Nitric Oxide Synthesis Inhibition Blocks Reversal of Two-Kidney, One Clip Renovascular Hypertension After Unclipping
Abstract It is well established that two-kidney, one clip renovascular hypertension can be rapidly reversed by unclipping. We hypothesized that rapid renal reperfusion and the subsequent fall in blood pressure are mediated in part by nitric oxide, the endothelium-derived relaxing factor. We tested whether the hypotensive response to unclipping could be blocked by nitric oxide synthesis inhibition using a bolus of 10 mg/kg body wt Nω-nitro-l-arginine methyl ester. Rats were made hypertensive by placing a silver clip on the left renal artery. After 4 weeks, they were anesthetized and either not treated (controls) or had nitric oxide synthesis blockade. After 10 minutes, the clip was removed and blood pressure monitored over 60 minutes. Initial pressure in controls was 157±8 mm Hg, and heart rate was 310±21 beats per minute. Unclipping resulted in pressure falling to 125±6 mm Hg within 45 minutes (P<.005). Heart rate was unchanged (312±9 beats per minute). In contrast, nitric oxide synthesis inhibition increased blood pressure from 149±6 to 174±9 mm Hg (P<.001). Unclipping did not change blood pressure, which was 167±8 mm Hg after 60 minutes (P<.005 versus controls), and heart rate remained unchanged (282±13 versus 276±16 beats per minute). We determined the blood flow to the clipped kidneys using radioactive microspheres. Unclipping untreated hypertensive rats resulted in a 10-fold increase in renal blood flow (P<.001), concomitant with a decrease in blood pressure. In rats with nitric oxide synthesis inhibition, unclipping resulted in an increase in renal blood flow that was only a third of that seen in untreated rats, with no change in blood pressure. Our results show that nitric oxide synthesis inhibition eliminates the acute reversal of renovascular hypertension caused by unclipping. This suggests that endothelium-derived nitric oxide may be an important component in the reversal of two-kidney, one clip renovascular hypertension, either by facilitating renal reperfusion or by mediating the systemic response secondary to renal reperfusion.
- endothelium-derived relaxing factor
- nitric oxide
- hypertension, renovascular
- renal circulation
Two-kidney, one clip (2K1C) Goldblatt hypertension is an animal model of human renovascular hypertension in which renal artery stenosis is induced by placing a silver clip on the renal artery to partially obstruct perfusion.1 Renal artery stenosis results in a fall in renal blood flow (RBF) and glomerular filtration rate in the clipped kidney.2 3 This model induces a largely angiotensin-dependent hypertension within 3 to 4 weeks.2 Blockade of the renin-angiotensin system partially reduces the hypertension, but unclipping reduces blood pressure (BP) further.4 5 Despite reducing BP, blocking the renin-angiotensin system has been reported to depress perfusion and function of the clipped kidney.4
In 1949, Byrom and Dodson6 observed that removing the clip in the Goldblatt model resulted in a rapid reversal of hypertension. Subsequently, many studies have tried to explain the mechanism involved; for instance, it has been shown that ablation of the stenotic kidney results in a more gradual normalization of BP than unclipping,5 suggesting that renal reperfusion is an important component of the process. Denervation of the clipped kidney has been reported to partially reduce 2K1C hypertension, while unclipping further reduces BP to normal.7 In contrast, inhibition of prostaglandin synthesis shows very little effect on the reversal of hypertension by unclipping.8
Volume replacement during the marked diuresis that accompanies unclipping does not retard the reversal of high BP in 2K1C hypertension,9 10 although it does attenuate the antihypertensive effect of unclipping in the 1K1C model.11 These observations suggest that volume is the critical component in the 1K1C but not in the 2K1C model. Thus, several factors, including the renin-angiotensin system, volume and sodium balances, and increased renal nerve activity, contribute to the hypertension but not necessarily to the phenomenon of reversal by unclipping.
We have previously described an important relation between the vasoconstrictor renin-angiotensin system and the endothelium-derived vasodilator nitric oxide (NO) in controlling renal perfusion.12 13 In normotensive rats, inhibition of NO synthesis results in acute hypertension and renal vasoconstriction.14 15 This reduction in RBF is largely due to the influence of angiotensin II (Ang II) in the absence of NO, which is particularly apparent if basal plasma renin activity (PRA) is increased.16 17 Because 2K1C renovascular hypertension is characterized by a high circulating level of Ang II, with particularly elevated concentrations in the clipped or stenotic kidney,3 we hypothesized that reversal of 2K1C hypertension by unclipping is primarily due to sudden renal reperfusion. This reperfusion serves to facilitate local vasodilation by increasing vascular shear stress and thereby stimulating dormant NO synthesis.18 This should counteract the constrictor effect of locally elevated Ang II. Thus, we hypothesized that NO-mediated renal reperfusion is the initial step in the reversal of systemic hypertension, and therefore by inhibiting NO synthesis, we should retard the reversal of hypertension by unclipping.
We used a rat model of 2K1C renovascular hypertension as previously described.19 Briefly, male Sprague-Dawley rats weighing 180 to 200 g were anesthetized with 50 mg/kg body wt sodium pentobarbital (Abbott Laboratories). With the use of an antiseptic technique, the left renal artery was exposed through a retroperitoneal flank incision and carefully dissected free of the renal vein. A sterile silver clip with an internal diameter of 0.23 mm was placed around the renal artery to cause partial occlusion. The wound was then closed and the rat allowed to recover for 4 weeks before the study. Only rats that developed a systolic BP of greater than 160 mm Hg by 4 weeks after clipping were used.
All experiments on 2K1C hypertensive rats were run 4 weeks after clipping. At this time, rats were deprived of food but allowed free access to water the night before the study. After injection of 125 mg/kg body wt IP thiobutabarbital (Andrew Lockwood), the rats were placed on a heating pad to maintain normal body temperature. A tracheostomy was then performed using PE-260 tubing for spontaneous breathing of room air. The femoral artery was catheterized with PE-50 tubing to facilitate direct measurement of BP with a Statham pressure transducer (Viggo-Spectramed) and a chart recorder (Gould Instruments). Heart rate (HR) was derived electronically from the arterial pulse wave. The femoral vein was catheterized with PE-50 tubing for supplemental administration of 1 mL plasma obtained from a donor rat nephrectomized 24 hours earlier and for a constant infusion of saline at 40 μL/min as well as drug administration. The left (clipped) kidney was exposed through a midventral incision, and the renal artery and clip were carefully exposed to facilitate unclipping. Once prepared, rats were allowed to stabilize for 60 minutes. Removing the clip from the renal artery was done quickly and carefully, and any bleeding at the site was grounds for exclusion. The progression of BP after unclipping, sham unclipping, or tying off the kidney was monitored for 60 to 90 minutes.
Duration of NO Synthesis Inhibition
NO synthesis was blocked with an intravenous bolus of 10 mg/kg body wt Nω-nitro-l-arginine methyl ester (L-NAME, Sigma Chemical Co). We have previously shown that this dose produces complete and sustained inhibition of systemic and renal endothelium-dependent vasodilation.14 However, since our present experiments were designed to be extended over longer periods than we have previously reported, we measured BP, HR, and RBF in six normotensive rats over 90 minutes after L-NAME administration to determine how persistent the inhibition was. RBF was measured by fitting the left renal artery with a noncannulating electromagnetic flow probe (2.0 mm internal diameter) connected to a flowmeter (both Carolina Medical Electronics). The flow probe was calibrated in situ at the conclusion of each experiment. The distal end of the renal artery was cannulated (with the flow probe still in place), and timed blood collections were taken.
Response to Unclipping in 2K1C Rats
We studied three groups of rats. The first group of eight 2K1C rats was unclipped and BP monitored continuously over 60 minutes. The second group of eight 2K1C rats was given a bolus of 10 mg/kg body wt L-NAME; 10 minutes later, the clip was removed from the stenotic kidney and BP monitored over the following 60 minutes. The third group of six 2K1C rats was given 10 mg/kg body wt L-NAME, but the renal artery clip was not removed (sham unclipping). BP was monitored over the following 60 minutes.
In the first two groups (unclipping and unclipping after L-NAME), we took 0.5-mL blood samples for determination of PRA. The first blood sample was taken during the stabilization period, at least 30 minutes before experimental manipulations, and the sample volume was replaced with an equal volume of blood from a 24-hour nephrectomized donor rat. The second blood sample was taken at the conclusion of the experiment.
Response to Eliminating the Clipped Kidney After L-NAME
To determine whether eliminating rather than unclipping the clipped kidney would lower BP in L-NAME–treated rats, six 2K1C rats were given 10 mg/kg body wt L-NAME. After stabilization, the clipped kidney was functionally eliminated by completely tying off all renal circulation. A loop of 3-0 silk was placed around the entire kidney and advanced beyond the hilus to occlude the renal artery, vein, ureter, and anything else entering or leaving the kidney, after which BP was monitored over the following 60 minutes.
Determination of RBF to the Clipped and Contralateral Kidneys
In two groups of six 2K1C rats, RBF to both kidneys was determined using radioactive microspheres. In the first group, this was done before and 20 minutes after unclipping. In the second group, rats were first treated with 10 mg/kg body wt L-NAME, and RBF was measured 10 minutes later. The rats were then unclipped and measured again 20 minutes later.
RBF was determined with radioactive microspheres 15 mm in diameter tagged with either 141Ce or 85Sr (Du Pont–New England Nuclear) as described previously.19 Before unclipping, a PE-10 catheter was passed through the right common carotid artery into the left ventricle, and the position of the catheter tip was adjusted until the left ventricular pulse pressure could be read without artifacts. In addition, the right femoral vein and artery were catheterized with PE-50 tubing as before. The venous catheter was used for drug infusion and blood replacement, and the arterial catheter was used for monitoring of BP and HR. The rats were allowed to stabilize for 60 minutes. Microspheres were suspended in 2.5 mol/L glucose at a concentration of 400 000/mL and mechanically agitated for approximately 15 minutes. A volume of 200 mL of the suspension, corresponding to approximately 80 000 microspheres, was then drawn up into a syringe together with 200 mL saline and infused into the left ventricle over 20 seconds. While the microsphere infusion was begun, an arterial reference blood sample was withdrawn at a rate of 0.48 mL/min over 75 seconds and replaced with blood obtained from a donor rat nephrectomized 24 hours earlier. After the second set of microspheres was injected, the rats were killed with an overdose of sodium pentobarbital, and the right and left kidneys were removed, decapsulated, weighed, and then placed in empty scintillation vials and counted in a gamma counter using dual window settings of 10 to 250 and 400 to 700 meV. RBF (milliliters per minute per gram kidney weight) was measured by comparing counts per minute in the kidney to the controlled blood sample using the formula (cpm kidney×pump speed)/(cpm blood×g kidney weight).
Influence of Phenylephrine on Unclipping 2K1C Rats
Because L-NAME produces an increase in total peripheral resistance that could alone account for the sustained hypertension irrespective of the alteration of RBF, we ran control experiments in which we produced a similar sustained increase in BP in 2K1C rats using an infusion of 4 mg/min phenylephrine (Sigma). In preliminary studies in normotensive rats (not presented here), we determined that this dose produced a sustained increase in BP of 20 to 25 mm Hg, similar to the pressor response to 10 mg/kg L-NAME. We studied two groups of rats. The first group of six 2K1C rats was started on the phenylephrine infusion. After 10 minutes, the rats were unclipped, and BP was monitored continuously over 90 minutes. A second group of three 2K1C rats was given the phenylephrine infusion, but after 10 minutes the clip was not removed (sham unclipping). BP was monitored over the following 90 minutes as phenylephrine time controls.
Response to Unclipping After a Kinin Antagonist (Hoe 140)
To determine whether the hypotensive response to unclipping could be mediated in part by kinins, we treated four 2K1C rats with a continuous infusion of 1 mg/min of the kinin analogue antagonist Hoe 140 (Hoechst), which we have found to be 10 times more potent an inhibitor of kinin-induced hemodynamic responses than antagonists previously reported.20 After 10 minutes of infusion, the clip was removed and BP monitored for 90 minutes.
Response to Unclipping After Inhibition of Platelet-Activating Factor
To determine whether the hypotensive response to unclipping was mediated in part by platelet-activating factor (PAF), we treated four rats with an oral dose of 10 mg/kg body wt of the PAF inhibitor U46,195 (Upjohn Co) suspended in approximately 1 mL of 0.2% ethanol in saline.21 An initial dose of 0.5 mL was given by gavage and the animal allowed to rest in its cage for 4.5 hours. Then the rat was anesthetized and the surgical preparation carried out. When the abdominal cavity was first opened, the remaining drug was injected directly into the stomach. The experiment was then carried out as described above.
PRA was determined by radioimmunoassay of Ang I generation using a modification of the technique of Haber et al22 as described previously.2 Results are expressed as nanograms Ang I per milliliter per hour. Acute changes were analyzed using a paired Student’s t test, whereas changes over time were analyzed by ANOVA with probability (P) modified according to the Bonferroni adjustment. Differences between acute changes were compared using an unpaired t test. A value (or adjusted value) of P<.05 was considered statistically significant.
Duration of NO Synthesis Inhibition
As shown in Fig 1⇓, intravenous bolus administration of 10 mg/kg body wt L-NAME in six normotensive rats increased BP by 22±2 to 128±6 mm Hg (P<.001); HR fell from 296±8 to 267±3 beats per minute (bpm) (P<.05); and RBF fell 25% from 7.65±0.73 to 6.10±0.61 mL/min per gram kidney weight (P<.025). The increase in BP remained constant over 90 minutes, ending unchanged at 131±9 mm Hg. HR returned to pre–L-NAME levels (310±11 bpm) by 90 minutes. RBF remained depressed throughout the full 90 minutes (5.81±0.70 mL/min per gram kidney weight). Despite the reflex correction of HR, these data suggest that our blocking dose of L-NAME was sustained throughout the 90-minute study period.
Similarly, in six 2K1C hypertensive rats (sham unclipping group), L-NAME resulted in a rapid increase in BP of 41±4 mm Hg (P<.001) (from 154±7 to 194±9 mm Hg). HR dropped from 321±22 to 271±17 bpm. RBF was not determined in these rats. After 60 minutes, BP remained elevated (200±5 mm Hg), and HR had only partially returned to pre–L-NAME levels (287±13 bpm). Thus, as in the normotensive rats, the results after 60 minutes gave little indication that the systemic inhibition of NO synthesis by this dose of L-NAME was reversed.
Response to Unclipping in 2K1C Rats
Basal BP of eight 2K1C rats was 157±8 mm Hg, and HR was 310±21 bpm. As shown in Fig 2⇓, unclipping resulted in a drop in BP over 45 minutes to 125±6 mm Hg, after which no further decreases were apparent. At 60 minutes, BP was 128±6 mm Hg, and HR was unchanged at 312±9 bpm. In contrast, rats given L-NAME to block NO synthesis had no hypotensive response to unclipping (Fig 2⇓). Basal BP in the second set of eight 2K1C rats was 149±6 mm Hg, and HR was 282±7 bpm. When these rats were given 10 mg/kg body wt L-NAME, BP increased 25±3 mm Hg up to 174±9 mm Hg (P<.001), and HR dropped to 233±5 bpm (P<.05). Sixty minutes after unclipping, BP was 167±8 mm Hg, and HR was 276±16 bpm. Thus, BP was the same as before unclipping but significantly different from BP in untreated 2K1C rats 60 minutes after unclipping (P<.005).
Basal BP in the third group of six 2K1C rats was 154±7 mm Hg. In response to L-NAME, BP rose to 190±9 mm Hg (P<.001). In these rats, the renal artery clip was not removed, and BP remained elevated and unchanged over the following 60 minutes (196±5 mm Hg at 60 minutes).
PRA of the anesthetized 2K1C rats during the stabilization period was 51±6 ng Ang I/mL per hour. There were no differences between the rats in either group before treatment. In the untreated rats, 60 minutes after unclipping, PRA had decreased to 16±3 ng Ang I/mL per hour (P<.005). In rats treated with L-NAME, after 60 minutes PRA was decreased to 13±4 ng Ang I/mL per hour (P<.001), a level not different from that in untreated rats.
Response to Eliminating the Clipped Kidney After L-NAME
In six 2K1C rats, L-NAME increased BP from 153±6 to 201±5 mm Hg (P<.001). Once the increased BP had stabilized, the entire clipped kidney was tied off from the circulation. Over 60 minutes, after perfusion of the clipped kidney was eliminated, BP did not change further (195±4 mm Hg at 60 minutes).
Determination of RBF to the Clipped and Contralateral Kidneys
RBF to the clipped kidney was determined by radiolabeled microspheres before and 20 minutes after unclipping in five untreated and six L-NAME–treated rats. Basal BP in untreated rats was 159±9 mm Hg. In the clipped kidney within 20 minutes after unclipping, blood flow increased more than 10 times from 0.63±0.31 to 6.46±0.23 mL/min per gram kidney weight (P<.001) (Fig 3⇓). BP dropped to 141±11 mm Hg (P<.01). This contrasted with L-NAME–treated rats, in which unclipping increased RBF only a fraction of that seen in untreated rats (from 0.34±0.13 to 1.90±0.36 mL/min per gram kidney weight, P<.001). BP remained unchanged.
In the contralateral kidney of untreated rats, RBF was not significantly changed after unclipping (from 4.92±0.48 to 7.34±0.58 mL/min per gram kidney weight). With L-NAME treatment, RBF in the contralateral kidney was 1.90±0.41 mL/min per gram kidney weight and was slightly less after unclipping (1.72±0.47 mL/min per gram kidney weight).
Influence of Phenylephrine on Unclipped 2K1C Rats
The six rats that were unclipped had a basal BP of 161±11 mm Hg, and BP in the time controls was 149±16 mm Hg. As shown in Fig 4⇓, infusion of 4 mg/min phenylephrine produced a similar pressor response in both rat groups (to 199±12 and 187±16 mm Hg, respectively). Unclipping of phenylephrine-treated rats resulted in a drop in BP of 37±15 mm Hg (P<.03), similar to the drop of 32 mm Hg seen in untreated 2K1C rats after unclipping (above). In the 2K1C phenylephrine time controls (sham unclipping), the pressor response to phenylephrine was sustained throughout the 90 minutes (186±3 mm Hg after 90 minutes).
Response to Unclipping After a Kinin Antagonist (Hoe 140)
As shown in Fig 5⇓, pretreatment of four 2K1C rats with the kinin antagonist Hoe 140 did not eliminate the decrease in BP after unclipping. Basal BP was 152±15 mm Hg. There was an apparent trend to accelerate the decrease in BP as it dropped to 124±8 mm Hg within the first 10 minutes after unclipping. By 90 minutes BP was 137±9 mm Hg.
Response to Unclipping After Inhibition of PAF
Also shown in Fig 5⇑, pretreatment of four 2K1C rats with the PAF inhibitor U46,195 did not eliminate the decrease in BP after unclipping. However, the response was attenuated such that the rate of decrease in BP was only two thirds as rapid as seen after unclipping of untreated rat kidneys.
Our results show that the reversal of 2K1C renovascular hypertension by unclipping could be completely eliminated by blocking NO synthesis with L-NAME. This suggests that reversal of 2K1C hypertension is largely mediated by restoration of blood flow to the kidney, as previously proposed by others4 8 23 ; however, we further suggest that restoration of renal perfusion is largely mediated by stimulation of renal endothelium-derived NO synthesis in the stenotic kidney.
Goldblatt or 2K1C hypertension is an animal model of human renovascular hypertension characterized by depressed renal perfusion and glomerular filtration rate in the clipped kidney.2 3 This model results in angiotensin-dependent hypertension that is fully developed within 3 to 4 weeks after clipping.2 It has long been known that surgical repair of a discrete renal artery stenosis rapidly resolves the accompanying hypertension.6 Simply excising rather than repairing the stenotic kidney results in a slower return to normal BP than unclipping,5 suggesting that renal reperfusion is an important component of the process of reversing the hypertension.4 5 8 23 It has been reported that denervation of the clipped kidney only partially reverses 2K1C hypertension, as subsequent unclipping further reduces BP to normotensive levels.7 Inhibition of prostaglandin synthesis seems to have very little effect on the reversal of hypertension by unclipping.8 Although 2K1C hypertension is highly angiotensin dependent, acutely blocking Ang II does not completely reverse the hypertension, and Ang II blockade without unclipping may further impair function of the stenotic kidney.4 After the renin-angiotensin system is blocked, unclipping provides an additional hypotensive influence in this model. Replacement of volume lost during the diuresis after unclipping has been reported not to alter the hypotensive response to unclipping in 2K1C hypertension9 10 and is apparently not a major factor in this model. Overall, although various factors such as renal nerve activity, the renin-angiotensin system, and plasma volume may influence BP in 2K1C hypertension, they do not explain the acute reversal of hypertension by unclipping.
Our previous work has addressed the interaction between the vasoconstrictor Ang II and the dilator NO in controlling renal perfusion12 13 16 in normotensive rats. NO synthesis inhibition causes acute hypertension and Ang II–mediated renal vasoconstriction.13 The Ang II dependence of decreased renal perfusion is amplified if basal PRA is increased with either anesthesia12 or dietary salt restriction.17
If the interaction between Ang II and NO is a critical regulator of renal perfusion in the normal kidney as Ang II increases, we hypothesized that in an Ang II–dependent model of hypertension, this interaction might be even more important. We have previously reported that in 2K1C renovascular hypertension, inhibition of NO synthesis results in an exaggerated acute systemic pressor response, suggesting that NO synthesis is increased to buffer the elevated BP.19 We also found that L-NAME decreased RBF and increased renal vascular resistance in the nonclipped (contralateral) kidney but did not change RBF in the stenotic or clipped kidney.19 From this we inferred that local NO synthesis is related to the degree of renal perfusion. Shear stress is the primary endogenous signal for endothelial NO production18 ; thus, impaired renal perfusion should result in diminished shear stress and retard the stimulus for NO, resulting in an imbalance between NO and high Ang II in the clipped kidney.3 We hypothesized that the reperfusion resulting from unclipping would reverse this imbalance by increasing renal vascular NO synthesis (assuming the endothelium was still functional). Therefore, NO-mediated renal reperfusion would act as the initiating step in the reversal of systemic hypertension.
We found that NO synthesis inhibition produced a sustained and exaggerated hypertension that was similar with or without unclipping. This contrasted to the rapid reversal of hypertension (within 45 minutes) by unclipping seen in untreated 2K1C rats. In the untreated rats, the clipped kidney was characterized by greatly compromised blood flow that was completely restored to normal within 20 minutes after unclipping, suggesting that resumption of renal perfusion precedes the hypotension. In L-NAME–treated rats, unclipping resulted in only a slight increase in RBF, despite the fact that renal perfusion pressure remained elevated. Interestingly, PRA was decreased similarly after unclipping in either untreated or L-NAME–treated rats, suggesting that the presence of NO more than a decrease in PRA is the important variable in reperfusion and decreased BP in these protocols. However, we did not sample PRA after L-NAME but before unclipping. We have previously found in normotensive rats that the pressor response to L-NAME reduces PRA,24 and presumably at least part of the decrease in PRA in L-NAME–treated rats may precede unclipping. Overall, these observations support our hypothesis that NO-mediated restitution of renal perfusion is a key element in the acute reversal of hypertension in this model.
Previously it has been reported that ablation of the clipped kidney does not induce the rapid acute hypotensive response seen with unclipping,5 supporting the concept that renal reperfusion is the initiating step in this response.4 8 23 Likewise, completely tying off the clipped kidney did not alter the sustained hypertension seen in L-NAME–treated rats.
Since L-NAME blocks systemic NO synthesis and raises BP, it could be argued that despite our results in the renal circulation, the maintenance of hypertension after unclipping seen in L-NAME–treated rats could be due to the influence of increased total peripheral resistance and not to the response in the unclipped kidney. To test this, we produced a sustained pressor response in 2K1C hypertensive rats, similar to that found with L-NAME, using a constant infusion of phenylephrine. Despite the increased hypertension in 2K1C rats given phenylephrine, unclipping resulted in a BP decrease similar in magnitude to that seen in untreated rats after unclipping. Thus, an increase in total peripheral resistance alone cannot account for the lack of a hypotensive response to unclipping observed when NO synthesis is inhibited. As an interesting caveat, in preliminary studies (not reported) in normotensive rats, phenylephrine infusion that produced a pressor response of 20 to 25 mm Hg (similar to that seen after L-NAME) resulted in only a 2% decrease in RBF in contrast to the 35% decrease in RBF seen in normotensive rats after NO synthesis inhibition.12 14 This supports our contention that in the kidney the influence of NO is of particular importance in regulating renal perfusion.
The pressor response to L-NAME in 2K1C hypertensive rats was generally almost double that found in normotensive rats. If the systemic response to L-NAME is interpreted as a bioassay for the role of NO in maintaining total peripheral resistance, then these data suggest overall that there is no deficiency in the endogenous synthesis of NO in this model of hypertension, despite the apparent diminished NO synthesis in the stenotic kidney. This exaggerated systemic pressor response is consistent with our previous findings in 2K1C hypertensive rats.19
Although NO is a potent vasodilator, it is rapidly inactivated in the blood and therefore cannot serve as a systemic endocrine vasoactive factor.18 Thus, if NO-mediated renal vasodilation and reperfusion is the initial step in the rapid reversal of 2K1C hypertension by unclipping, some signal secondary to the renal reperfusion must influence the systemic resistance. We tested the possible role of two candidate vasoactive systems: the kallikrein-kinin system and PAF. Since kinins are endothelium-dependent vasodilators that can directly stimulate endothelial NO synthesis,18 25 we tested whether the systemic response to unclipping would be impaired after blocking kinins with Hoe 140. However, the antagonist not only failed to block the depressor response to unclipping but actually seemed to potentiate it. Likewise, PAF has been shown to stimulate renal NO synthesis25 and is reportedly released from the renal medulla, which might make it a good candidate for a systemic factor that is stimulated by a rapid increase in renal perfusion after unclipping. However, after PAF was blocked, BP still dropped after unclipping. There was a tendency for BP to drop more slowly than after unclipping in untreated rats. Overall, since neither manipulation mimicked the effect of L-NAME to block the reversal of hypertension, neither of these systems seems to be a likely systemic vasodilator signal resulting from renal reperfusion. The identity of such a signal remains unresolved.
In conclusion, we found that acute reversal of 2K1C renovascular hypertension by unclipping could be completely eliminated by blocking NO synthesis with L-NAME. This was not the case after a similar increase in BP seen using phenylephrine. Our data suggest that reversal of hypertension by unclipping is largely influenced by NO-mediated restoration of blood flow to the kidney, as the increase in RBF seen after unclipping was greatly attenuated by L-NAME. However, NO-mediated restoration of renal perfusion can serve only as an initiating step, as renal-derived NO cannot act as a systemic endocrine vasodilator. Presumably, it is the initiating event in some yet undetermined secondary signal, stimulated by restoration of renal perfusion, which results in systemic vasodilation and ultimately normotension.
This work was supported by grant HL-46683-A02 from the National Institutes of Health, Bethesda, Md. The authors wish to thank the Upjohn Co for the gracious gift of the PAF inhibitor U46,195 and Hoechst Pharmaceutical for the gift of the kinin antagonist Hoe 140.
Part of this work previously appeared in abstract form (Hypertension. 1992;20:428).
- Received March 17, 1994.
- Revision received May 4, 1994.
- Accepted September 20, 1994.
Goldblatt H, Lynch J, Hanazal RF. Studies on experimental hypertension, I: the production of persistent elevation of systolic blood pressure by means of renal ischemia. J Exp Med. 1937;59:347-378.
Martinez-Maldonado M. Pathophysiology of renovascular hypertension. Hypertension. 1991;17:707-719.
Russell GI, Bing RF, Thurston H, Swales JD. Surgical reversal of two-kidney one clip hypertension during inhibition of the renin-angiotensin system. Hypertension. 1982;4:69-76.
Byrom FB, Dodson LF. The mechanism of the viscous circle in chronic hypertension. Clin Sci. 1949;8:1-14.
Katholi RE, Whitlow PL, Winetnitz SR, Oparil S. Importance of renal nerves in established two-kidney, one clip Goldblatt hypertension. Hypertension. 1982;4(suppl II):II-166-II-174.
Romero JC, Ott CE, Aguilo JJ, Torres VE, Strong CG. Role of prostaglandins in the reversal of one-kidney hypertension in the rabbit. Circ Res. 1975;37:683-689.
Neubig RR, Hoobler SW. Reversal of chronic renal hypertension: role of salt and water excretion. Proc Soc Exp Biol Med. 1975;150: 254-256.
Otsuka Y, Carretero OA, Albertini R, Binia A. Angiotensin and sodium balance: their role in chronic two-kidney Goldblatt hypertension. Hypertension. 1979;1:389-396.
Akahoshi M, Carretero OA. Body fluid volume and angiotensin II in maintenance of one-kidney, one clip hypertension. Hypertension. 1989;14:269-273.
Sigmon DH, Beierwaltes WH. Angiotensin II: nitric oxide interaction and the distribution of blood flow. Am J Physiol. 1993;265: R1276-R1283.
Sigmon DH, Carretero OA, Beierwaltes WH. Angiotensin dependence of endothelium-mediated renal hemodynamics. Hypertension. 1992;20:643-650.
Beierwaltes WH, Sigmon DH, Carretero OA. Endothelium modulates renal blood flow but not autoregulation. Am J Physiol. 1992;262:F943-F949.
Baylis C, Harton P, Engels K. Endothelium derived relaxing factor controls renal hemodynamics in normal rat kidney. J Am Soc Nephrol. 1990;1:875-881.
Sigmon DH, Carretero OA, Beierwaltes WH. Plasma renin activity and the renal response to nitric oxide synthesis inhibition. J Am Soc Nephrol. 1992;3:1288-1294.
Sigmon DH, Beierwaltes WH. Angiotensin II: nitric oxide interaction in conscious rats. J Am Soc Nephrol. 1994;4:1675-1682.
Sigmon DH, Beierwaltes WH. Renal nitric oxide and angiotensin II interaction in renovascular hypertension. Hypertension. 1993;22: 237-242.
Beierwaltes WH, Carretero OA, Scicli AG, Vavrek RJ, Stewart JM. Competitive analog antagonists of bradykinin in the canine hindlimb. Proc Soc Exp Biol Med. 1987;186:79-83.
Thurston T, Bing RF, Swales JD. Reversal of two-kidney one clip renovascular hypertension in the rat. Hypertension. 1980;2:256-265.
Sigmon DH, Carretero OA, Beierwaltes WH. Endothelium-derived relaxing factor regulates renin release in vivo. Am J Physiol. 1992;263:F256-F261.
Marsden PA, Brock TA, Ballerman BJ. Glomerular endothelial cells respond to calcium-mobilizing agonists with release of EDRF. Am J Physiol. 1990;258:F1295-F1303.