Stimulation of angiotensin AT₂ receptors by the non-peptide agonist, Compound 21, evokes vasodepressor effects in conscious spontaneously hypertensive rats

Abstract

Background and purpose:

Angiotensin type 2 receptor (AT₂ receptor) stimulation evokes vasodilator effects in vitro and in vivo that oppose the vasoconstrictor effects of angiotensin type 1 receptors (AT₁ receptors). Recently, a novel non-peptide AT₂ receptor agonist, Compound 21, was described, which exhibited high AT₂ receptor selectivity.

Experimental approach:

Functional cardiovascular effects of the drug candidate Compound 21 were assessed, using mouse isolated aorta and rat mesenteric arteries in vitro and in conscious spontaneously hypertensive rats (SHR).

Key results:

Compound 21 evoked dose-dependent vasorelaxations in aortic and mesenteric vessels, abolished by the AT₂ receptor antagonist, PD123319. In vivo, Compound 21 administered alone, at doses ranging from 50 to 1000 ng·kg⁻¹·min⁻¹ over 4 h did not decrease blood pressure in conscious normotensive Wistar-Kyoto rats or SHR. However, when given in combination with the AT₁ receptor antagonist, candesartan, Compound 21 (300 ng·kg⁻¹·min⁻¹) lowered blood pressure in SHR only. Further analysis in separate groups of conscious SHR revealed that, at a sixfold lower dose, Compound 21 (50 ng·kg⁻¹·min⁻¹) still evoked a significant depressor response in adult SHR (∼30 mmHg) when combined with different doses of candesartan (0.01 or 0.1 mg·kg⁻¹). Moreover, the Compound 21-evoked depressor effect was abolished when co-infused (50 µg·kg⁻¹·min⁻¹ for 2 h) with the AT₂ receptor antagonist PD123319.

Conclusion and implications:

Collectively, our results indicate that acute administration of Compound 21 evoked blood pressure reductions via AT₂ receptor stimulation. Thus Compound 21 can be considered an excellent drug candidate for further study of AT₂ receptor function in cardiovascular disease.

Introduction

The octapeptide angiotensin II (Ang II) is the main biologically active mediator of the renin-angiotensin system and plays an important role in cardiovascular function by influencing vascular tone, structure, fluid and electrolyte balance via direct effects on endothelial and smooth muscle cells (Widdop et al., 2003; Jones et al., 2008). Two main receptor subtypes have been identified as binding sites for Ang II: angiotensin type 1 receptor (AT₁ recepter) and type 2 receptor (AT₂ receptor) (de Gasparo et al., 2000); nomenclature follows Alexander et al., 2008). Ang II has similar affinity for both AT₁ receptors and AT₂ receptors, whereas CGP42112 and PD123319 are the prototypical examples of an agonist and antagonist, respectively, at the AT₂ receptor subtype. On the other hand, compounds such as candesartan and losartan are selective AT₁ receptor antagonists that are used clinically for the treatment of hypertension. It is well established that most of the cardiovascular effects induced by Ang II, such as vasoconstriction, water and salt retention, are mediated via AT₁ receptor (de Gasparo et al., 2000; Carey, 2005). In contrast, it has been suggested that the function of the AT₂ receptor is to counter-regulate AT₁ receptor-mediated actions, mainly based on experiments in which AT₂ receptor function was deduced from the effects of AT₂ receptor blockade, or altered responses in genetically modified animal models of AT₂ receptor overexpression or deletion. However, demonstration of AT₂ receptor-mediated effects, particularly in an in vivo setting, has been hampered by a lack of non-peptide AT₂ receptor selective agonists and antagonists that exhibit oral bioavailability.

In this context, Wan et al. (2004b) have recently described the first non-peptide, selective AT₂ receptor agonist, Compound 21 (N-butyloxycarbonyl-3-(4—imidazol-1-ylmethylphenyl)-5-isobutylthiophene-2-sulphonamide). Compound 21 was derived from a medicinal chemistry programme aimed at transforming the drug-like but non-selective AT₁ and AT₂ receptor agonist L-162313 (Wan et al., 2004a) into a selective AT₂ receptor agonist. Compound 21 exhibits a K_i value of 0.4 nM for the AT₂ receptor and a K_i > 10 µM for the AT₁ receptor. In addition, due to the non-peptide nature of the drug, Compound 21 has an estimated oral bioavailability of 20–30% and 4 h half-life in rats (Wan et al., 2004b). Compound 21 has been shown to induce neurite outgrowth in cell culture and to increase duodenal mucosal alkalinization in the rat via stimulation of MAPK and NO/cGMP signalling pathways (Wan et al., 2004b). Furthermore, Compound 21 decreased mean arterial blood pressure (MAP) in anaesthetized spontaneously hypertensive rats (SHR), although detailed and systematic evaluation of haemodynamic responses to Compound 21 were not performed in this earlier study.

AT₂ receptor-mediated relaxation is a well-established effect in isolated resistance vessels (Matrougui et al., 1999; Dimitropoulou et al., 2001; Widdop et al., 2002); conversely, there is less consensus regarding the influence of AT₂ receptors on blood pressure regulation in vivo. Studies using AT₂ receptor knockout mice support a role for AT₂ receptors in haemodynamic control, as these animals exhibit elevated basal blood pressure and enhanced sensitivity to the vasopressor effects of Ang II (Hein et al., 1995; Ichiki et al., 1995). Conversely, overexpression of AT₂ receptors in vasculature did not alter basal blood pressure, but markedly impaired Ang II-induced pressor activity (Tsutsumi et al., 1999). In conscious SHR, Ang II-mediated vasodilatation during AT₁ receptor blockade was not observed (Gohlke et al., 1998), presumably because the hypotensive effect of AT₂ receptor stimulation was masked by the concomitant, dominant AT₁ receptor-mediated pressor action during Ang II infusion. In order to avoid such confounding influences of AT₁ receptor stimulation on potential AT₂ receptor vasodilator function, we and others have also assessed the effect of selective AT₂ receptor agonists and antagonists during AT₁ receptor blockade. Using this approach, selective stimulation of AT₂ receptors by CGP42112 lowered blood pressure, provided that there was a background of AT₁ receptor blockade in conscious SHR (Wistar-Kyoto rat, WKY) (Barber et al., 1999), and Sprague-Dawley rats (Carey et al., 2001) in a PD123319-reversible manner. Furthermore, this blood pressure-lowering response to AT₂ receptor stimulation was shown to be associated with increased blood flow in renal, mesenteric and hindquarter circulations in conscious SHR suggesting widespread vasodilatation (Li and Widdop, 2004).

Therefore, in the current study we determined the effects of Compound 21 on blood pressure in conscious SHR and WKY rats, as well as in isolated vasculature. In addition, AT₂ receptor selectivity of Compound 21 was determined by simultaneous administration of the selective AT₂ receptor antagonist, PD123319, to determine whether or not these effects were AT₂ receptor-mediated.

Methods

Animals

All animal care and experimental procedures were approved by the Monash University Animal Ethics Committee and performed according to the guidelines of the National Health and Medical Research Council of Australia for animal experimentation.

Male 16- to 18-week-old SHR and WKY rats, weighing approximately 300 to 350 g and male 16-week-old FVB/N mice, weighing approximately 25–30 g were obtained from the Animal Resource Centre (Perth, WA, USA). Animals were maintained on a 12 h day/night cycle with standard laboratory rat or mice chow and water available ad libitum.

In vitro reactivity

Mice were killed by isoflurane inhalation followed by decapitation. The thoracic aorta was removed and cut transversely into ring segments for organ bath studies. Two stainless steel wires were threaded through the lumen of each aortic ring and the rings were then mounted and suspended in vertical 10 mL organ baths containing Krebs bicarbonate solution [composition (mM): NaCl 118, KCl 4.7, KH₂PO₄ 1.2, MgSO₄.7H₂O 1.2, CaCl₂ 2.5, NaHCO₃ 25 and glucose 11.7; pH 7.4], which was maintained at 37°C and continuously bubbled with carbogen (95% oxygen and 5% CO₂). Isometric tension was continuously measured via a force transducer (Grass FT03) interfaced to a MacLab data acquisition device (ADInstruments, Sydney, Australia) displayed on a Macintosh computer. Aortic rings were set to 0.5 g resting tension and allowed to equilibrate for 90 min, during which time Krebs bicarbonate solution was changed every 15 min. After equilibration, 0.3 µM of the thromboxane A₂ receptor agonist, U46619, was used to obtain the maximum contractile response. Once a maximum response was determined, tissues were then washed with Krebs' solution and allowed to equilibrate for 30 min or until baseline had been reached.

Tissues were pre-contracted with U46619 to attain 30–40% of the maximum contractile response. In the first series of experiments, all vessels were pretreated with the AT₁ receptor antagonist, losartan (0.1 µM). Cumulative dose–response curves at log intervals to Compound 21 (1 pM to 1 µM) were performed in absence or presence of the AT₂ receptor antagonist PD123319 (0.1 µM). In a further series, AT₁ receptor blockade was omitted and cumulative dose-response curves to Compound 21 were performed in the absence and presence of either the NOS inhibitor, L-NAME (10 µM) or PD123319. A parallel tissue served as a time control in which only U46619 was given. At the end of the experiment, 10 µM of the endothelium-independent vasodilator, sodium nitroprusside was added to the organ bath to test the integrity of the vascular smooth muscle cells.

In analogous studies, thoracic aortic rings from male SHR were set up at 2 g resting tension and a maximum response was obtained to 124 mM K+. After washing, tissues were pre-contracted with the α₁-adrenoceptor agonist phenylephrine to 30–40% of maximum K+ response and cumulative dose–response curves to either Ang II (in the presence of 0.1 µM candesartan) or Compound 21 in the presence or absence of candesartan were obtained.

Mesenteric artery

Male WKY rats, approximately 16 weeks of age, were killed by isoflurane inhalation followed by decapitation and the gut was removed in order to dissect 3 to 5 mm long sections of the third order branch from mesenteric artery. The arterial sections were cannulated at both ends and mounted in a video-monitored perfusion system (Living Systems Instrumentation, Burlington, VT, USA), as previously described (Matrougui et al., 1999; Loufrani et al., 2001; Widdop et al., 2002). Mesenteric artery sections were bathed in 20 mL organ baths that contained Krebs solution to which the AT₁ receptor antagonist, candesartan (1 µM) was added. The solution was bubbled with carbogen (95% O₂ and 5% CO₂), with temperature maintained at 37°C and the pH at 7.4. The arterial sections were superfused at a rate of 4 mL·min⁻¹ and perfused at a rate of 100 µL·min⁻¹. The intraluminal pressure was set at 75 mmHg. The diameter of the arterial sections was constantly measured and recorded with a video-monitoring system. Following an equilibration period of approximately 30 min, phenylephrine was added to achieve 20–30% of the maximum contractile response. Once the plateau was reached, a concentration response curve to Compound 21 (0.1 nM to 1 µM) was constructed. Analogous experiments were performed in which the AT₂ receptor antagonist, PD123319 (1 µM) was added 30 min before Compound 21. An additional tissue was set up that served as a time control and was only pre-contracted with phenylephrine.

In vivo procedures

Rats were anaesthetized (ketamine and xylazine; 75 mg·kg⁻¹ and 10 mg·kg⁻¹, i.p., respectively; supplemented as required). Two catheters were inserted into the right jugular for i.v. drug administration. A catheter was inserted into the right carotid artery for direct blood pressure measurement as described previously (Barber et al., 1999; Li and Widdop, 2004; Walters et al., 2005). Rats were housed in individual cages and allowed free access to food and water while maintained on 12 h day/night cycle. The arterial catheter was infused overnight with heparinized saline using an infusion pump.

Twenty-four hours after the surgery, the arterial catheter was attached to a pressure transducer (Gould Inc., Eichstetten, Germany), connected to a MacLab-8 data acquisition system (ADInstruments) and interfaced to a Macintosh computer. Mean arterial pressure (MAP) and heart rate were computed from the phasic blood pressure signal.

Experimental protocol

Rats received drug combinations in a randomized fashion over a 4 or 5 day protocol, as described previously (Barber et al., 1999; Walters et al., 2005). Doses of candesartan and PD123319 were chosen on the basis of previous studies (Barber et al., 1999; Walters et al., 2005). Animals in group 1 (WKY; dose-ranging) and group 2 (SHR; dose-ranging) were randomized to receive following treatments: (i) a 4 h Compound 21 infusion (50, 100 or 300 ng·kg⁻¹·min⁻¹ in WKY rats or 100, 300 and 1000 ng·kg⁻¹·min⁻¹ in SHR); and (ii) a 4 h Compound 21 infusion (50 and 300 ng·kg⁻¹·min⁻¹ in WKY rats or 300 and 1000 ng·kg⁻¹·min⁻¹ in SHR) given simultaneously with candesartan (0.1 mg·kg⁻¹ i.v.).

Based on the dose-ranging results, additional SHR (group 3) received the following treatments in randomized fashion: (i) candesartan (0.1 mg·kg⁻¹ i.v.); (ii) Compound 21 infusion (50 ng·kg⁻¹·min⁻¹ for 4 h); (iii) a 4 h Compound 21 infusion together with candesartan; and (iv) a 4 h Compound 21 infusion in the presence of candesartan and PD123319 infusion (50 µg·kg⁻¹·min⁻¹ for 2 h). The dose of PD123319 was based on our previous experience using this compound in similar in vivo experiments (Barber et al., 1999; Li and Widdop, 2004). In analogous experiments in separate SHR (group 4), an identical protocol was repeated to that of group 3 but a 10-fold lower dose of candesartan (0.01 mg·kg⁻¹ i.v.) was used. We have previously shown that basal BP recordings are stable over these time periods. Nevertheless, in this latter group, SHR also received a 4 h infusion (0.1 mL·kg⁻¹·h⁻¹ i.v.) of saline (0.9% NaCl) to confirm a lack of effect on MAP.

Statistical analysis

All data are presented as mean responses ± standard error of the mean (SEM). Differences in vasorelaxation or MAP between treatments were analysed using a two-way repeated measure, anova. Statistical analysis was performed using GraphPad Prism (Version 5.0). P-values <0.05 were considered statistically significant.

Materials

Compound 21 was provided by A Hallberg, Department of Medicinal Chemistry, Uppsala University; PD123319 and candesartan were kind gifts from Pfizer and AstraZeneca respectively. All other chemicals were purchased from commercial sources: L-NAME (Sigma), sodium nitroprusside (Sigma, Sydney, Australia), ketamine (Troy Laboratories, Sydney, Australia), xylazine (Troy Laboratories), phenylephrine (Sigma), isoflurane (Baxter, Deerfield, IL, USA) and U46619 (Saphire Bioscience, Sydney, Australia).

Results

In vitro relaxation evoked by Compound 21

Compound 21 caused a dose-dependent relaxation of mouse aorta in the presence of AT₁ receptor blockade, which was markedly inhibited by PD123319 (Figure 1A). Furthermore, Compound 21-evoked relaxation was also evident in the absence of AT₁ receptor blockade and was abolished by L-NAME (Figure 1B). In addition, Compound 21 caused vasodilatation in third order perfused rat mesenteric artery, which was also blocked by PD123319 (Figure 1C).

Figure 1.

Dose–response curves to Compound 21 (C21) performed in (A) mouse aorta and (C) third order perfused rat mesenteric artery in the absence (n= 5 and 9 for aorta and mesenteric artery respectively) and presence (n= 5 and 3 for aorta and mesenteric artery respectively) of PD123319 (PD); all experiments in Figure 1A and C performed in the presence of angiotensin type 1 receptor (AT₁ receptor) blockade, as described in the Methods. (B) Effect of Compound 21 in mouse aorta in the absence of AT₁ receptor blockade and in the presence of PD123319 or L-NAME (n= 10 for each). Values represent mean ± SEM. **P < 0.01 for treatment effect between Compound 21 and Compound 21 + PD123319 or Compound 21 + L-NAME (two-way repeated measures anova).

Experiments were also performed using aortic rings obtained from naive SHR, a preparation that is reported to be unresponsive to the AT₂ receptor-mediated vasorelaxant effects of Ang II (Cosentino et al., 2005). Indeed, Ang II did not cause relaxation; on the other hand, Compound 21 evoked dose-dependent relaxation that was similar in the presence or absence of AT₁ receptor blockade (Figure 2).

Figure 2.

Dose–response curves to either angiotensin II (Ang II) or Compound 21 (C21) performed in aorta obtained from naive spontaneously hypertensive rat (SHR) in the absence (Compound 21) and presence (Ang II & Compound 21) of candesartan (0.1 µM) (n= 5 for each). Values represent mean ± SEM. **P < 0.01 for treatment effect between Compound 21 and time control (two-way repeated measures anova).

In vivo effect of Compound 21 in conscious rats

Basal MAP of SHR over the four or five experimental days for each group are listed in the Table 1. There was no significant difference between resting MAP over the experimental period for any of the treatment groups, suggesting that none of the acute treatments had effects that lasted more than 24 h, and therefore did not influence baseline MAP on subsequent days.

Table 1.

Resting mean arterial pressure (MAP) of spontaneously hypertensive rat recorded on separate days before drug treatments, as indicated

Treatment	MAP (mmHg)
Group 2 (n= 5)
Compound 21 (100 ng·kg−1·min−1)	174 ± 8
Compound 21 (300 ng·kg−1·min−1)	177 ± 8
Compound 21 (1000 ng·kg−1·min−1)	165 ± 5
Compound 21 (300 ng·kg−1·min−1) & candesartan (0.1 mg·kg−1)	164 ± 8
Compound 21 (1000 ng·kg−1·min−1) & candesartan (0.1 mg·kg−1)	163 ± 5
Group 3 (n= 7)
Compound 21 (50 ng·kg−1·min−1)	178 ± 6
Candesartan (0.1 mg·kg−1)	191 ± 6
Compound 21 & candesartan	190 ± 7
Compound 21, candesartan & PD123319 (50 µg·kg−1·min−1)	177 ± 7
Group 4 (n= 7)
Saline	175 ± 6
Compound 21 (50 ng·kg−1·min−1)	179 ± 7
Candesartan (0.01 mg·kg−1)	184 ± 8
Compound 21 & candesartan	190 ± 7
Compound 21, candesartan & PD123319 (50 µg·kg−1·min−1)	170 ± 10

The values shown in the Table are means ± SEM. n= 5–7 per group.

In both WKY (Figure 3) and SHR (Figure 4), infusion of Compound 21 alone, at doses ranging from 50 to 300 ng·kg⁻¹·min⁻¹, had no significant effect on MAP. Combined administration of Compound 21 (100 and 300 ng·kg⁻¹·min⁻¹) and candesartan (0.1 mg·kg⁻¹) also had no effect on MAP in WKY rats (Figure 3); however, the combination of Compound 21 (300 ng·kg⁻¹·min⁻¹) and candesartan, significantly decreased MAP in SHR compared with Compound 21 alone (P < 0.001) (Figure 4). Interestingly, at the highest dose tested (1000 ng·kg⁻¹·min⁻¹), Compound 21 alone caused an increase in MAP in SHR (P < 0.05), suggesting a lack of selectivity of Compound 21 at this dose. This Compound 21-mediated pressor effect was attenuated by simultaneous AT₁ receptor blockade (Figure 4).

Figure 3.

Effects of Compound 21 (50, 100 and 300 ng·kg⁻¹·min⁻¹), administered as a 4 h infusion (shown by horizontal line), on mean arterial pressure (MAP) in Wistar-Kyoto rats (n= 5). Compound 21 was given in the presence or absence of the angiotensin type 1 receptor antagonist, candesartan (0.1 mg·kg⁻¹ i.v. bolus; shown by an arrow). Values represent mean ± SEM.

Figure 4.

Effects of Compound 21 (100, 300 and 1000 ng·kg⁻¹·min⁻¹), administered as a 4 h infusion (shown by horizontal line), on mean arterial pressure (MAP) in spontaneously hypertensive rats (n= 5). Compound 21 was given in the presence or absence of the angiotensin type 1 receptor agonist, candesartan (0.1 mg·kg⁻¹ i.v. bolus, shown by an arrow). Values represent mean ± SEM. #P < 0.001 for treatment effect between Compound 21 (300 ng·kg⁻¹·min⁻¹) + candesartan and Compound 21(300 ng·kg⁻¹·min⁻¹) (two-way repeated measures anova); *P < 0.05 for treatment effect between Compound 21(1000 ng·kg⁻¹·min⁻¹) + candesartan and Compound 21(1000 ng·kg⁻¹·min⁻¹) (two-way repeated measures anova).

Given the lack of response to Compound 21 in WKY rats, further examination of the effect of Compound 21 on MAP was performed in separate groups of SHR, to determine the effects of lower dose of Compound 21 in combination with AT₁ receptor block. As had been previously determined, Compound 21 infusion alone (50 ng·kg⁻¹·min⁻¹) had no effect on MAP in SHR (Figure 5). However, when combined with either high-dose (0.1 mg·kg⁻¹; Figure 5) or low-dose (0.01 mg·kg⁻¹; Figure 6) candesartan, Compound 21 caused a significant reduction in MAP in SHR (P < 0.001). Importantly, when the AT₂ receptor antagonist, PD123319 (50 µg·kg⁻¹·min⁻¹), was co-infused for 2 h with Compound 21 and candesartan, this blood pressure-lowering effect was abolished, indicating AT₂ receptor selectivity of Compound 21 (P < 0.001). Furthermore, infusion of saline had no effect on blood pressure (Figure 6).

Figure 5.

Effect of Compound 21 (50 ng·kg⁻¹·min⁻¹; 4 h infusion shown by full line), high-dose candesartan (0.1 mg·kg⁻¹ bolus i.v.; shown by an arrow), Compound 21 + candesartan and Compound 21 + candesartan + PD123319 (50 µg·kg⁻¹·min⁻¹ for 2 h; (dashed line), on mean arterial pressure (MAP) in spontaneously hypertensive rat (n= 7). Values represent mean ± SEM. #P < 0.001 for overall effect of treatment versus Compound 21 (two-way repeated measures anova); †P < 0.01 for treatment effect between Compound 21 + candesartan and candesartan (two-way repeated measures anova).

Figure 6.

Effect of Compound 21 (50 ng·kg⁻¹·min⁻¹; 4 h infusion shown by full line), low-dose candesartan (0.01 mg·kg⁻¹ bolus i.v.; shown by an arrow), Compound 21 + candesartan and Compound 21 + candesartan + PD123319 (50 µg·kg⁻¹·min⁻¹ for 2 h; shown by dashed line), on mean arterial pressure (MAP) in spontaneously hypertensive rat (n= 7). Values represent mean ± SEM. *P < 0.05 for treatment effect between Compound 21 and candesartan; #P < 0.001 for overall effect of individual treatment versus Compound 21 (two-way repeated measures anova); †P < 0.01 for treatment effect between Compound 21 + candesartan and candesartan alone (two-way repeated measures anova).

Discussion

The main finding of the current study was that the novel non-peptide AT₂ receptor agonist, Compound 21, evoked vasorelaxation in vitro, which translated into vasodepressor responses in conscious SHR against a background of AT₁ receptor blockade. To our knowledge, this study represents the first systematic study of the vascular effects of Compound 21, particularly in hypertension.

Compound 21 exhibits a similar binding profile at AT₂ receptors to that of Ang II and CGP42112 but with little affinity for AT₁ receptors. In mouse aortic rings, Compound 21, in the presence of AT₁ receptor antagonists, caused concentration-dependent vasorelaxation that was inhibited by the AT₂ receptor antagonist, PD123319. This in vitro effect was not dependent on background AT₁ receptor blockade, as it was in vivo, and was also inhibited by L-NAME. Thus, Compound 21 elicited classical AT₂ receptor-mediated NO signalling, a hallmark of AT₂ receptors in vascular tissue (Widdop et al., 2003; Jones et al., 2008). This compound was also tested using rat isolated mesenteric arteries, a preparation well recognized as exhibiting AT₂ receptor-mediated vasorelaxation (Matrougui et al., 1999; Henrion et al., 2001; Widdop et al., 2002). Indeed in these resistance-like vessels, Compound 21 caused vasorelaxation that was inhibited by PD123319, confirming selective AT₂ receptor-mediated actions of the compound.

In the present study, dose–response analysis was performed in conscious SHR and WKY rats in which multiple doses were tested in the same animals on different days, allowing within-animal analysis. When tested over a wide dose range, Compound 21 alone did not reduce MAP in SHR or WKY rats except during AT₁ receptor blockade in SHR, as seen previously with CGP42112 (Barber et al., 1999; Li and Widdop, 2004). It is likely that these results can be explained by the fact that circulating endogenous Ang II itself exerts tonic AT₁ receptor-mediated vasoconstriction that, once removed, allows AT₂ receptor-mediated vasodilatation to be manifest. These data are consistent with previous studies in conscious SHR in which Ang II did not cause vasodilatation during AT₁ receptor blockade (Gohlke et al., 1998), presumably because the hypotensive effect of AT₂ receptor stimulation was masked by the concomitant, dominant AT₁ receptor-mediated pressor action during Ang II infusion. A differential effect on vascular tone to AT₂ receptor stimulation has also been noted by others using SHR and WKY rats (Savoia et al., 2005), and is consistent with a lack of effect of Compound 21 on blood pressure in anaesthetized normotensive rats (Wan et al., 2004b). This lack of effect of Compound 21 on BP in normotensive animals may relate to the fact that subtle AT₂ receptor-mediated depressor responses are more easily observed from a higher basal blood pressure. Alternatively, it may represent strain-dependent differences in sensitivity to AT₂ receptor stimulation or drug-induced changes in vascular AT₂ receptor expression. In this context, aortic AT₂ receptor expression is higher in adult SHR compared with age-matched WKY, whereas mesenteric AT₂ receptor expression is increased in young SHR but decreased in adult SHR (see Widdop et al., 2008).

Limited in vivo studies in anaesthetized SHR implied a role of this compound on vascular tone as, when given as bolus i.v. injections to anaesthetized SHR, Compound 21 lowered BP (Wan et al., 2004b). The discrepancy between the two studies most likely reflects the different experimental designs between the current and previous studies (Wan et al., 2004b). The highest effective dose (0.05 mg·kg⁻¹) of Compound 21 previously tested (Wan et al., 2004b) was probably higher than our maximally effective depressor dose achieved during a 4 h infusion (i.e. 300 ng·kg⁻¹·min⁻¹∼ 0.072 mg·kg⁻¹ total; during AT₁ receptor blockade), once half-life and pharmacokinetic considerations are taken into account, although it should be noted that a higher dose of Compound 21 (1000 ng·kg⁻¹·min⁻¹) alone also did not decrease blood pressure. Another important difference is that the previous study (Wan et al., 2004b) tested barbiturate-anaesthetized SHR, which are less physiological than consciously instrumented SHR that would be more able to buffer potential blood pressure reductions by homeostatic reflex mechanisms. Moreover, we were keen to be able to make direct comparisons with previous studies that also reported AT₂ receptor-mediated depressor effects of the selective agonist CGP42112 during AT₁ receptor blockade (Barber et al., 1999; Li and Widdop, 2004).

Curiously, AT₂ receptor stimulation does not generally cause vasorelaxation in vessels isolated from SHR strains (Matrougui et al., 2000; Cosentino et al., 2005; Savoia et al., 2005; You et al., 2005). Chronic treatment with AT₁ receptor antagonists is associated with increased AT₂ receptor expression in aortae from SHR (Cosentino et al., 2005; Savoia et al., 2005), and human subcutaneous gluteal arteries (Savoia et al., 2007), as well as in mesenteric arteries from SHR, which normally exhibited decreased AT₂ receptor expression under basal conditions (You et al., 2005). Indeed, sartan-induced up-regulation of AT₂ receptors unmasked ex vivo AT₂ receptor-mediated vasorelaxation in otherwise unresponsive vessels (Yayama et al., 2004; Cosentino et al., 2005; Savoia et al., 2005; Savoia et al., 2007). Therefore, it is conceivable that enhanced in vivo sensitivity of untreated SHR, compared with WKY rats, to Compound 21 was due to higher AT₂ receptor expression, although elevated MAP per se could still contribute to AT₂ receptor-mediated depressor activity in vivo. In any case, whether or not ex vivo AT₂ receptor-mediated relaxation evoked by Compound 21 is more manifest following chronic treatment with an AT₁ receptor antagonist awaits further investigation. However, we did test the in vitro effects of Compound 21 acutely in naive SHR. As expected, Ang II did not evoke vasorelaxation whereas, strikingly, Compound 21 relaxed aortae. Thus, in aortic tissue known to be refractory to acute AT₂ receptor-mediated effects of Ang II (Cosentino et al., 2005; Savoia et al., 2005), Compound 21 caused vasorelaxation, in SHR, both in vitro and in vivo, at least in the presence of AT₁ receptor blockade. Thus, the current study highlights the importance of using subtype selective compounds.

At the highest dose tested (1000 ng·kg⁻¹·min⁻¹), Compound 21 alone actually increased MAP in SHR, most likely representing a lack of AT₂ receptor selectivity at this concentration. Although the sensitivity of this pressor effect of Compound 21 to blockade by PD123319 was not tested in the current study, simultaneous AT₁ receptor inhibition restored MAP responses to baseline. This finding could indicate that Compound 21 caused AT₁ receptor stimulation at higher doses, which may relate to the higher AT₁/AT₂ receptor ratio in vasculature, or that the hypotensive effect of candesartan offset Compound 21-mediated vasoconstriction via other mechanisms. The fact that a lower dose of Compound 21 (300 ng·kg⁻¹·min⁻¹), in combination with candesartan, significantly lowered MAP in these same SHR makes it likely that a more selective AT₂ receptor vasodilator effect was manifest at doses <1000 ng·kg⁻¹·min⁻¹. Similarly, pressor doses of Ang II infused in the presence of AT₁ receptor blockade do not always reduce blood pressure (Gohlke et al., 1998), most likely for the same reasons of opposing vascular effects of AT₁ and AT₂ receptor stimulation (Barber et al., 1999; Li and Widdop, 2004).

When infused at a sixfold lower dose, the depressor effect of Compound 21 (50 ng·kg⁻¹·min⁻¹) was similar in groups concomitantly administered either high- or low-dose candesartan, suggesting that the maximum achievable fall in MAP had been reached. By contrast, using CGP42112, we did not find additional depressor effects when combined with high-dose candesartan (Barber et al., 1999), which may reflect a difference in metabolic fate of these two compounds as Compound 21 is a non-peptide compound. The additive effect of Compound 21 and candesartan may also reflect the fact that sartan-induced elevation in Ang II levels was probably not maximal. In any case, we have found that AT₂ receptors do not functionally desensitize even in the face of raised Ang II levels (Widdop et al., 2002), therefore the non-peptide AT₂ receptor agonist may exert a prolonged effect. Importantly, in these same animals, we also tested PD123319, which completely abolished the depressor effect of combined Compound 21 and candesartan; consistent with the AT₂ receptor selectivity demonstrated by both the current in vitro data and radioligand binding assays performed with this compound (Wan et al., 2004b). Interestingly, although tested in separate animal groups, it also appeared that there was no difference in the maximal Compound 21-mediated depressor effect using either 50 or 300 ng·kg⁻¹·min⁻¹. In this context, a bell-shaped dose–response relationship for the effect of Compound 21 on BP was also reported by Wan et al. (2004b), who found that the depressor effect of bolus Compound 21 administration in anaesthetized SHR was present at lower doses, but lost at higher doses (>0.05 mg·kg⁻¹).

Collectively, these data implicate the AT₂ receptor as a potential target for the treatment of hypertension, although until now, there have been no drug-like candidates available to directly test this premise. In this context, the effects of Compound 21 have recently been reported in the setting of myocardial infarction. In that study, Compound 21, given for 7 days after myocardial infarction, improved systolic and diastolic function and reduced infarct size (Kaschina et al., 2008), thus illustrating the potential use of this non-peptide compound in a number of cardiovascular settings. Indeed, our current findings suggest additive effects of AT₁ receptor blockade and AT₂ receptor stimulation would be beneficial for BP reduction, and fit with clinical findings of increased vascular AT₂ receptor expression after long-term sartan treatment (Savoia et al., 2007), highlighting the need for future determination of the chronic effects of Compound 21 in hypertensive settings.

In conclusion, we have established that Compound 21 evoked vasorelaxation in mouse and SHR aortae or rat mesenteric arteries, and vasodepressor responses in conscious SHR, via AT₂ receptor stimulation. The BP-lowering effect of Compound 21 was additive to candesartan when the latter compound was given at a dose that itself lowered BP. Further studies are warranted on the chronic effects of Compound 21, alone and in combination with AT₁ receptor antagonists, in hypertensive-related diseases. These studies implicate the AT₂ receptor as a potential therapeutic target in the setting of hypertension and related cardiovascular diseases.

Glossary

Abbreviations:

Ang II

angiotensin II

AT₁ receptor

angiotensin type 1 receptor

AT₂ receptor

angiotensin type 2 receptor

MAP

mean arterial pressure

SHR

spontaneously hypertensive rat

WKY rat

Wistar-Kyoto rat

Conflicts of interest

None.