Pharmacodynamic Contribution to the Vasodilator Effect of Chronic AT1 Receptor Blockade in SHR
Abstract—The present study investigated the pharmacodynamic contribution of AT1 receptor blockade to the regional hemodynamic effects of long-term treatment with the AT1 receptor antagonist candesartan cilexetil in adult spontaneously hypertensive rats (SHR). Blood pressure and Doppler flowmetry measurements were made during and after withdrawal of candesartan cilexetil, representing times of maximal and negligible blockade of AT1 receptor–mediated vasoconstriction. There was marked renal, mesenteric, and hindquarter vasodilation in SHR treated for 4 weeks with candesartan cilexetil (2 mg/kg per day in drinking water, n=8) compared with vehicle (n=8). Blood pressure increased after withdrawal of candesartan cilexetil but was still reduced after 6 days, whereas regional flows and conductances did not reduce significantly compared with the last day of treatment. There was more prolonged inhibition of angiotensin (Ang) I–induced than Ang II–induced pressor responses after withdrawal of candesartan cilexetil, but these returned to control levels before blood pressure reached fully hypertensive levels. The renal and mesenteric vasoconstrictor effects of exogenously administered Ang I and Ang II returned to control levels just 2 days after withdrawal of candesartan cilexetil. Therefore, sustained inhibition of tonic Ang-mediated vasoconstriction caused by blockade of the AT1 receptor is not the only factor contributing to the hemodynamic profile after long-term administration of candesartan cilexetil. In addition, compared with the vehicle group, blood pressures at maximum vasoconstriction and maximum vasodilation (an indirect measure of vascular hypertrophy) were significantly reduced in candesartan cilexetil–treated SHR on the last day of treatment, as was mesenteric media wall-to-lumen ratio in a separate group of similarly treated SHR. Collectively, these findings indicate that Ang-mediated vasoconstriction rapidly normalizes on withdrawal of AT1 receptor blockade and that regression of vascular hypertrophy is important in determining blood pressure and hemodynamic status in candesartan cilexetil–treated SHR at this time.
AT1 receptor antagonists, like ACE inhibitors, decrease blood pressure (BP), and this effect is associated with regression of cardiac and vascular hypertrophy.1 Thus, the renin-angiotensin system (RAS) plays a key pathophysiological role in the development and maintenance of hypertension. Moreover, the beneficial antihypertensive and remodeling effects caused by RAS inhibition often persist to varying degrees after the withdrawal of treatment, depending on the duration of initial treatment (see Reference 11 for review). However, the relative contribution of pharmacodynamic as opposed to vascular structural components to the overall hemodynamic effect of RAS inhibition is not well characterized. In this context, in adult spontaneously hypertensive rats (SHR), we have observed that after withdrawal of long-term treatment (for 4 weeks) with the AT1 receptor antagonist candesartan cilexetil, there is an abrupt increase in BP in the immediate postwithdrawal period (≈1 week). Thereafter, BP returns to control levels after ≈10 days.2 However, it is unclear as to whether or not this change involves the removal of a vasodilator effect of the AT1 receptor antagonist alone and/or changes in vascular structure.
In addition, the antihypertensive effect of AT1 receptor antagonists, including candesartan cilexetil, is not well correlated with their antagonist profile against angiotensin (Ang) II itself.3 4 5 6 Indeed, on the basis of short-term studies in separate groups of anesthetized rats, it was suggested that the antihypertensive effect of AT1 receptor blockade was better correlated temporally with inhibition of Ang I–induced pressor responses rather than with Ang II.4 5 On the assumption that Ang I responses represent the local generation of Ang II in the vasculature, it was suggested that the BP-lowering effect of AT1 receptor blockade involved inhibition of locally generated as well as plasma-borne Ang II.5 Therefore, it is likely that prolonged inhibition of local RASs within various vascular beds occurs after long-term administration of AT1 receptor antagonists.
Therefore, the aims of the present study were 2-fold: (1) to determine the pharmacodynamic contribution of AT1 receptor blockade to the overall effect in SHRs by measuring the basal regional hemodynamics either at the end of 4 weeks of treatment with candesartan cilexetil or immediately after drug withdrawal (representing times of maximal and negligible AT1 receptor-mediated vasoconstriction) and (2) to determine if there was differential inhibition of the regional hemodynamic effects of Ang I and Ang II during the withdrawal phase. In addition, an indirect assessment of whole-body vascular hypertrophy was also made at the end of the treatment period.
Adult male SHR, 19 weeks of age and weighing 330 to 350 g, were obtained from the Biological Research Laboratories at the Austin Hospital, Heidelberg, Melbourne, Australia, and maintained on a 12-hour day/night cycle at 20° to 22°C with free access to food and water. All experiments were approved by the Monash University animal ethics committee and carried out in accordance with National Health and Medical Research Council of Australia guidelines for the use of animals in research.
Rats were anesthetized with methohexitone sodium (60 mg/kg IP, supplemented as required) and had pulsed Doppler flow probes (Crystal Biotec) implanted around the left renal and superior mesenteric arteries and the distal abdominal aorta for the recording of renal, mesenteric, and hindquarter Doppler shift, as previously described.7 At least 7 days later, rats were anesthetized as before for the implantation of catheters into the jugular vein for drug administration and into the carotid artery for the measurement of mean arterial pressure (MAP) and heart rate (HR). Experiments were carried out 24 to 48 hours after catheterization. All variables were displayed on a MacLab-8 system (ADInstruments Pty Ltd) interfaced with a Macintosh computer. Doppler shift is an index of blood flow; regional vascular conductances were calculated by dividing the appropriate mean Doppler shift signal by MAP.
After an initial baseline measurement of systolic blood pressure (SBP), SHR were randomly assigned to 2 groups to be treated with either candesartan cilexetil (2 mg/kg per day, n=8) or vehicle (n=8) in drinking water for 4 weeks. Candesartan cilexetil was a gift from Takeda Chemical Industries Ltd; it was dissolved in a mixture of ethanol, polyethylene glycol, sodium bicarbonate, and distilled water according to the method of Mackenzie et al.8 During the treatment period, SBP was measured once per week until the third week, when Doppler flow probes were implanted. SBP was measured noninvasively by tail-cuff plethysmography as previously described.9 After the initial 4-week treatment period and after surgery for the implantation of vascular catheters, rats were maintained on drug or vehicle for 1 additional day for the assessment of baseline hemodynamics during treatment (day 0). Treatment was then stopped, and MAP, HR, and regional flows were assessed for the following 6 days (days 1 to 6).
On days 0 to 4, cardiovascular responses evoked by Ang I and Ang II (1 to 40 ng IV) were assessed in all SHR. To test if the slightly impaired pressor response to Ang I (see Results section) reflected a potential ACE-inhibitory action of candesartan cilexetil, cardiovascular responses evoked by bradykinin (BK) (1 to 2 μg IV) were also assessed on the last day of treatment in several SHR. Doses of Ang I, Ang II, and BK were administered every 10 to 15 minutes as a bolus in a volume of 0.05 or 0.1 mL. In addition, an indirect assessment of whole-body vascular hypertrophy was carried out on the last day of drug treatment. Briefly, after ganglion blockade (10 mg/kg IV pentolinium) and β-adrenoceptor blockade (1 mg/kg IV propranolol), BP was measured at maximum vasodilation (BPmin) in response to sodium nitroprusside (0.5 to 128 μg/kg IV) and at maximum vasoconstriction (BPmax) in response to methoxamine (2 to 512 μg/kg IV).10
Changes in basal MAP and regional flows and conductances over time within each group were analyzed by 1-way ANOVA with repeated measures. Differences in body weight, SBP, MAP, HR, and hemodynamic variables over time between candesartan cilexetil–treated and vehicle-treated SHR and differences in hemodynamic responses evoked by Ang I, Ang II, and BK (maximum change from baseline) were analyzed by 2-way ANOVA with repeated measures. Changes in BPmax and BPmin parameters were analyzed by unpaired Student’s t test. All data are expressed as mean±SEM. Statistical significance was set at P<0.05.
Body weights in SHR to be treated with candesartan cilexetil and vehicle were 341±9 g and 341±12 g, respectively. Body weight was unaffected by treatment with candesartan cilexetil.
Baseline SBP was similar in both groups of SHR. SBP was markedly reduced by treatment with candesartan cilexetil compared with the vehicle group (ANOVA, P<0.01) (Figure 1⇓). MAP measured directly on the last day of treatment was significantly lower in candesartan cilexetil–treated SHR compared with the vehicle group (ANOVA, P<0.01) (Figure 2⇓). After withdrawal of candesartan cilexetil, MAP increased over the following days but was reduced compared with the vehicle group up to 6 days later (P<0.01) (Figure 2⇓). There were no differences in HR between candesartan cilexetil–treated or vehicle-treated SHR throughout the withdrawal period (Figure 2⇓).
Flow (Doppler shift) in renal and hindquarter vascular beds was increased in candesartan cilexetil–treated SHR compared with the vehicle group (day 0, P<0.05) (Figure 2⇑). Mesenteric vascular flow tended to be increased in treated SHR compared with the vehicle group, although this increase was not significant. Thus, given the reduced MAP, renal, mesenteric, and hindquarter conductances were significantly higher in candesartan cilexetil–treated SHR compared with the vehicle group (day 0, P<0.05) (Figure 3⇓). During the withdrawal phase (days 1 to 6), each vascular flow and conductance was unchanged compared with the last day of treatment (day 0) in candesartan cilexetil–treated SHR, and the vehicle group was similarly unchanged over this period. However, renal, mesenteric, and hindquarter flows and conductances in candesartan cilexetil–treated SHR were not significantly different from those in vehicle-treated rats on day 6 of withdrawal (2-way ANOVA) (Figures 2⇑ and 3⇓).
Ang I–induced and Ang II–induced pressor responses in vehicle-treated SHR (Figure 4⇓) were associated with marked reductions in renal and mesenteric blood flows and variable changes in hindquarter flow (data not shown). Thus, Ang peptides evoked marked reductions in renal and mesenteric vascular conductances (Figure 5⇓ and 6⇓), whereas there was no consistent response in the hindquarter vascular bed (data not shown). On day 0, pressor responses evoked by Ang I and Ang II were markedly attenuated in candesartan cilexetil–treated SHR compared with the vehicle group (ANOVA, P<0.01) (Figure 3⇑). The renal and mesenteric vasoconstrictor effects of Ang I and Ang II were also markedly attenuated by candesartan cilexetil on day 0 (ANOVA, P<0.01) (Figures 4⇓ and 5⇓).
Compared with the vehicle group, pressor responses evoked by Ang I were attenuated up to 3 days after withdrawal of candesartan cilexetil (days 1 to 3) (ANOVA, P<0.05 to 0.01) (Figure 4⇑), whereas the accompanying renal and mesenteric vasoconstriction was attenuated only on days 0 and 1 in candesartan cilexetil–treated SHR (ANOVA, P<0.01) (Figures 5⇑ and 6⇑). In contrast, pressor as well as renal and mesenteric vasoconstriction responses caused by Ang II were attenuated for just 1 day after withdrawal in candesartan cilexetil–treated SHR (ANOVA, P<0.05 to 0.01) (Figures 5⇑ and 6⇑).
Administration of BK at 2 doses (1 and 2 μg) in vehicle-treated SHRs (n=5) induced hemodynamic responses, of which the major components were a depressor response (−26±6 and −27±6 mm Hg) accompanied by an increase in hindquarter flow (%Δ Doppler shift=100±27 and 116±21) leading to marked vasodilation of the hindquarter vascular bed (%Δ conductance=141±36 and 161±28). However, there were no differences in BK-induced responses in SHR treated with candesartan cilexetil (depressor responses: −15±4 and −12±7 mm Hg; %Δ hindquarter Doppler shift=93±10 and 119±43; %Δ hindquarter conductance=116±6 and 140±46; all n=3).
An indirect assessment was made of whole-body vascular hypertrophy on the last day of the drug treatment period. After autonomic blockade, resting MAP was slightly lower in candesartan cilexetil–treated SHR (102±3 mm Hg) compared with the vehicle group (117±6 mm Hg) (P<0.05). In candesartan cilexetil–treated SHR, BPmax (183±2 mm Hg) (P<0.01) and BPmin (50±1 mm Hg) (P<0.05) were reduced compared with the vehicle group (217±4 mm Hg and 58±4 mm Hg, respectively).
In separate groups of SHR that had received identical treatments with either vehicle (n=9) or candesartan cilexetil (n=8), these indirect vascular morphology measurements were again obtained. In addition, the media wall-to-lumen ratios of third-order branches of mesenteric arteries from the same animals were measured with a small-vessel myograph. As expected, candesartan cilexetil caused similar reductions in BPmax and BPmin as well as a decrease in the media wall-to-lumen ratio. These changes resulted in there being a very good correlation between BPmax values and media wall-to-lumen ratio (r=0.69, n=17, P<0.01) in these animals, confirming that vascular remodeling had occurred in the candesartan cilexetil–treated SHR in which hemodynamic measurements were previously performed.
In the present study, the antihypertensive effect of long-term administration of candesartan cilexetil was associated with marked renal, mesenteric, and hindquarter vasodilation. In contrast, short-term administration of AT1 receptor antagonists caused renal vasodilation11 12 13 that we have subsequently shown to be relatively selective for this vascular bed in SHR, because vascular tone was largely unaffected by the active metabolite of candesartan cilexetil, candesartan, in the mesenteric and hindquarter circulations.3 Similarly, with the microsphere technique, 1-week treatment with candesartan cilexetil caused a selective increase in renal blood flow in SHRs.14 Collectively, these data suggest that longer-term (4 weeks) administration of an AT1 receptor antagonist causes substantial vasodilatation in all vascular beds. This effect is likely to be due to a general reduction in Ang II–induced tonic vasoconstriction and/or regression of vascular hypertrophy in various vascular beds that would not occur after single administration. Thus, hemodynamic measurements at times of maximal and negligible AT1 receptor–mediated vasoconstriction should address the role of the pharmacodynamic contribution to the overall response.
In the present study, BP was still reduced up to 6 days after withdrawal of candesartan cilexetil. This time course is consistent with similarly treated SHR in a previous study, although BP in those rats normalized after ≈10 days.2 Any hemodynamic changes that are manifested in the immediate postwithdrawal period (particularly days 1 to 3) are likely to represent the extent of the pharmacodynamic contribution of candesartan cilexetil to the overall antihypertensive effect. Surprisingly, after withdrawal of candesartan cilexetil, regional flows and conductances in SHRs did not significantly decrease compared with the final day of drug treatment (in line with corresponding increases in MAP), suggesting a minimal pharmacodynamic contribution at this time. However, there was a tendency after 5 to 6 days of withdrawal from candesartan cilexetil for regional flows and conductances to decrease toward levels observed in the vehicle-treated group, resulting in both groups being not significantly different from each other at this time.
The regional hemodynamic effects of Ang I and Ang II measured after withdrawal allowed the contribution of individual vascular beds to pressor responses to be determined. As expected, both Ang I and Ang II evoked marked renal and mesenteric vasoconstriction, as reported previously.3 6 7 There was more prolonged inhibition of Ang I–induced than Ang II–induced pressor responses after withdrawal of candesartan cilexetil (3 days versus 1 day). Moreover, the reversal of inhibition of Ang I–induced pressor responses was more closely paralleled with the increase in BP toward vehicle levels after withdrawal of treatment than was the time course of inhibition of pressor responses evoked by Ang II. Previous studies also found a similar dissociation between the antihypertensive effect of AT1 receptor antagonists and the inhibition of Ang II responsiveness.3 4 5 6 In particular, the antihypertensive effect of the AT1 receptor antagonist GR138950 was shown to be better temporally correlated with the inhibition of Ang I–induced than Ang II–induced pressor responses,4 5 as was demonstrated in the present study with candesartan cilexetil, although both Ang I–induced and Ang II–induced pressor responses had returned to control levels well before MAP reached fully hypertensive levels.
Unlike pressor responses, the renal and mesenteric vasoconstrictor effects of exogenously administered Ang I and Ang II were not differentially modulated after withdrawal of candesartan cilexetil. Instead, hemodynamic responses evoked by Ang peptides were restored to control levels just 2 days after withdrawal of candesartan cilexetil, when basal BP and basal regional conductances were still divergent from the vehicle group (day 2), as well as being largely unchanged from those values obtained during treatment. These findings suggest that after withdrawal of candesartan cilexetil, the pharmacological inhibition of AT1 receptors reverses rapidly at the level of the renal and mesenteric vasculature. The fact that there was more prolonged inhibition of pressor responses evoked by Ang I than Ang II after withdrawal may suggest that AT1 receptors in other intravascular compartments were still partially inhibited and/or changes in cardiac output were differentially affected. Alternatively, it was possible that the longer inhibition of Ang I–induced than Ang II–induced pressor responses involved inhibition of ACE activity. However, in the present study, there was no difference in BK-induced responses after treatment with candesartan cilexetil. Because ACE inhibition enhanced both the hypotensive and regional vasodilator effects of BK,15 it is unlikely that inhibition of ACE was the mechanism responsible for the different time course for Ang I–mediated and Ang II–mediated pressor responses in the present study. This finding lends support to a previous study in which the AT1 receptor antagonist losartan was shown to have no effect on tissue ACE levels, whereas plasma and tissue BK levels tended to be reduced.16
Collectively, it appears that sustained tonic inhibition of Ang-mediated vasoconstriction caused by AT1 receptor blockade is not the major factor contributing to the hemodynamic profile observed with long-term treatment with candesartan cilexetil. It is likely that after initial pharmacodynamic changes with candesartan cilexetil, more permanent alterations in vascular structure occurred. Such progressive changes can be inferred by the fact that basal BP and regional flows and conductances took longer to adjust after withdrawal of treatment with candesartan cilexetil than did restoration of Ang-mediated vasoconstriction. This point is reinforced by the lack of effect of Ang I and Ang II on hindquarter vascular conductance despite the observation of markedly increased basal flow and conductances in this vascular bed.
Finally, there are many studies that have reported reductions in vascular hypertrophy during AT1 receptor blockade and after withdrawal of treatment.1 17 18 19 Importantly, for the present study, Rizzoni et al18 showed reductions in media wall-to-lumen ratios in mesenteric vessels from SHRs taken 3 to 4 days after washout of candesartan cilexetil as well as 1 week after withdrawal of losartan.19 Thus, these structural changes were present at a time when we have shown remarkably little pharmacodynamic involvement of AT1 receptor blockade in the overall hemodynamic effects. Further support for the involvement of a major structural component in the persistent BP reduction after withdrawal of treatment comes from the finding that candesartan cilexetil–treated SHR had reduced BPmax and BPmin on the last day of the treatment period. These parameters have been used previously as indicators of whole-body vascular hypertrophy, whereby ACE inhibition produced an effect similar to that seen in the present study.10 Direct vascular morphology was also assessed in separate groups of rats in which the reduced BPmax was found to correlate very well with the reduced media wall-to-lumen ratio in candesartan cilexetil–treated SHR. Thus, it can be inferred from the current and previous studies1 17 18 19 that there was regression of vascular hypertrophy after 4 weeks of treatment with candesartan cilexetil, and this effect would have contributed substantially to the elevated regional vascular flows and conductances.
In summary, the antihypertensive effect of candesartan cilexetil, administered chronically to adult SHR, is associated with marked renal, mesenteric, and hindquarter vasodilation, which contrasts with acute hemodynamic effects of AT1 receptor blockade in this strain.3 Moreover, it appears that this effect may involve a greater structural than pharmacodynamic component, because there were persistent hemodynamic effects after withdrawal of treatment at a time when Ang-mediated vasoconstrictor responses had normalized. Thus, regression of vascular hypertrophy during long-term treatment with candesartan cilexetil in adult SHR may be important in determining BP and hemodynamic status after withdrawal of treatment.
These studies were supported in part by a grant from the National Health and Medical Research Council of Australia.
Paull JRA, Widdop RE. Cardiovascular effects of chronic treatment with the AT1 receptor antagonist, TCV-116, in adult spontaneously hypertensive rats. J Hypertens.. 1998;16:S123. Abstract.
Hilditch A, Hunt AAE, Travers A, Polley J, Drew GM, Middlemiss D, Judd DB, Ross BC, Robertson MJ. Pharmacological effects of GR138950, a novel angiotensin AT1 receptor antagonist. J Pharmacol Exp Ther. 1995;272:750–757.
Li XC, Widdop RE. Regional hemodynamic effects of the AT1 receptor antagonist CV-11974 in conscious renal hypertensive rats. Hypertension. 1995;26:989–997.
Mackenzie HS, Troy JL, Rennke HG, Brenner BM. TCV 116 prevents progressive renal injury in rats with extensive renal mass ablation. J Hypertens.. 1994;12:S11–S16.
Rizzoni D, Porteri E, Bettoni G, Piccoli A, Castellano M, Muiesan ML, Pasini G, Guelfi D, Rosei EA. Effects of candesartan cilexetil and enalapril on structural alterations and endothelial function in small resistance arteries of spontaneously hypertensive rats. J Cardiovasc Pharmacol. 1998;32:798–806.
Rizzoni D, Porteri E, Piccoli A, Castellano M, Bettoni G, Muiesan ML, Pasini G, Guelfi D, Mulvany MJ, Rosei EA. Effects of losartan and enalapril on small artery structure in hypertensive rats. Hypertension. 1998;32:305–310.