Cardioprotective Effects of Ramipril and Losartan in Right Ventricular Pressure Overload in the Rabbit
Importance of Kinins and Influence on Angiotensin II Type 1 Receptor Signaling Pathway
Background— The role of kinins in the cardioprotective effects of ACE inhibitors remains controversial.
Methods and Results— Right ventricular pressure overload in rabbits was produced by pulmonary artery banding for 21 days. Rabbits were untreated, or they received the ACE inhibitor ramipril with or without bradykinin B1 and B2 receptor blockers or the angiotensin (Ang) II type I (AT1) receptor blocker losartan. Pulmonary artery banding caused right ventricular hypertrophy, depressed papillary muscle contractility, and loss of Ang II contractile effects because of a signaling defect downstream of AT1 receptors. Paradoxically, AT1 receptor density and G protein α subunits αq and αi1/2 increased. Inotropic responsiveness to the α-receptor agonist phenylephrine was normal. Ramipril preserved cardiac contractility, but this effect was attenuated by simultaneous use of kinin receptor blockers. Ramipril also maintained responsiveness to Ang II and prevented AT1 receptor and G protein upregulation. The simultaneous use of a kinin receptor blocker attenuated but did not prevent upregulation in the AT1 receptor and G protein. Losartan had no effect on baseline contractility, but it maintained cardiac inotropic responsiveness to Ang II, prevented upregulation of AT1 receptors, but did not modify G protein upregulation.
Conclusions— Pressure overload of the right ventricle decreases contractility, uncouples AT1 receptors to downstream signaling pathways, and changes the expression of components of the AT1 receptor signaling pathway. Ramipril attenuates these effects via kinins. Interventions that prevent local increases in Ang II or block AT1 receptors also prevent decreased responsiveness of the AT1 receptor in this model.
Received January 25, 2001; revision received April 5, 2001; accepted April 24, 2001.
The cardioprotective effect of ACE inhibitors was originally thought to be nearly exclusively the result of inhibition of the conversion of angiotensin (Ang) I to Ang II. Other enzymes, however, are also capable of converting Ang I to Ang II, the most important of these being chymase.1,2 Indeed, in the heart and vasculature, chymase is an important mechanism for the conversion of Ang I to Ang II.1,2 More recently, inhibition of kinin metabolism by ACE inhibitors has also been shown to have important cardioprotective effects.3 Increasing kinin levels results in an increase in nitric oxide and vasodilatory prostaglandins, both of which have numerous actions that counter the effects of vasoconstrictor–growth-promoting substances and stimuli.3 These effects include vasodilatation, attenuation of cardiac and vascular growth, antiatherogenic and antithrombotic effects, and improved myocardial metabolic efficiency.3 Their cardioprotective effects appear to be both direct and indirect. The importance of the cardioprotective effects of bradykinin were most elegantly demonstrated in bradykinin B2 receptor–knockout mice, which were found to have hypertension and both left ventricular dilatation and dysfunction.4
A new class of medication, the Ang II type I (AT1) receptor blocker, has been developed to better block the vasoconstrictor–growth-promoting effects of Ang II directly at the receptor level.5 The AT1 receptor is responsible for most of the known biological effects of Ang II.5 The aortic banding pressure overload model of left ventricular hypertrophy in rats by Weinberg et al6,7 suggests that in this setting, ACE inhibitors may have some advantages over AT1 receptor blockers. The only clinical study comparing the 2 classes of drugs is the ELITE 2 study,8 which found a statistically insignificant trend in favor of the ACE inhibitor (12% reduction in mortality, P=0.16). Thus, the superiority of one class over the other still remains controversial and may depend on the setting in which it is studied. More studies comparing the cardioprotective effects of these 2 classes of medications would be helpful, particularly after the HOPE study,9 which demonstrated that the ACE inhibitor ramipril significantly reduced the mortality of patients with atherosclerosis.
In this study, we created right ventricular (RV) pressure overload by banding the pulmonary artery of rabbits and studied them 21 days later. Rabbits were untreated for control, or they received the ACE inhibitor ramipril with or without bradykinin (B1 and B2) receptor blockers or the Ang II AT1 receptor blocker losartan. B1 and B2 receptor blockers were given because with inflammation, such as may occur in this model, B1 receptors may be expressed and exert effects similar to those of B2 receptors. Cardiac hypertrophy and myocardial contractility were assessed. We then evaluated the changes in Ang AT1 and AT2 receptors and in G protein α subunit αq and αi1/2 expression and the effects of these changes on the integrity of the cardiac angiotensin II signaling pathway by using an isolated papillary muscle bioassay.
New Zealand White male rabbits (Charles River, St-Constant, Québec) weighing ≈1.95 kg were used. The protocols used in this study were approved by the Institutional Animal Research Ethics Committee, and all animals were treated and prepared in accordance with the guidelines of the “Canadian Council on Animal Care Regulations.”
Creation of the Model and Therapeutic Interventions
RV heart failure was induced in 90 rabbits by pulmonary artery banding according to methods previously described by Alpert and Mulieri.10 Once the rabbits were sedated, a spiral model metal coil (inside diameter 2.8 to 3.2 mm; length 6.6 mm; 2.5 turns; 0.7-mm-diameter wire) was applied to the outside of the central end of the pulmonary artery.
The rabbits were separated into 5 groups: (1) sham-operated controls; (2) banded, no therapy; (3) banded, ramipril (Avantis Pharmaceuticals) 37.5 μg/kg IP 1 hour after surgery, then 1 mg · kg−1 · d−1 in the drinking water11; (4) banded, with the same ramipril regimen, except that the kinin B1 and B2 receptor blockers Lys-[Leu 8]desArgBK at 1 mg/kg and HOE 140 at 1 mg/kg (Avantis Pharmaceutical), respectively, were also given by osmotic pump placed subcutaneously at the time of surgery3; and (5) banded, losartan 0.25 mg/kg IP 1 hour after surgery, then 50 mg · kg−1 · d−1 in the drinking water.11
Experiments were conducted 21 days after surgery, after ramipril was stopped for 72 hours, the kinin receptor antagonists for 24 hours, and losartan for 48 hours to permit adequate washout of the medications before further studies.
Hemodynamic and Morphological Studies
Animals were anesthetized with halothane 0.5% to 1% and RV pressure measured by cannulating the jugular vein and advancing a 5F Millar catheter (Millar Instruments Inc) connected to a Gould 2400s recorder (Gould Inc Instruments Division) to the RV. The animals were then euthanized, and 1 or 2 RV papillary muscles were removed and mounted in an isolated bath. The heart was then separated into the left ventricular free wall, the septum, the RV free wall, and the atria before being weighed and rapidly frozen in liquid nitrogen for measurements of Ang II AT1 and AT2 receptor density and G protein α subunit expression.
Papillary Muscle Studies
Papillary muscles were mounted in an isolated bath with Krebs-Henseleit solution at 29°C (in mmol/L: NaCl 118, KCl 4.7, MgSO4 · 0.7H2O 1.2, CaCl2 · H2O 1.25, KH2PO4 0.18, NaHCO3 24, glucose 5, and albumin 6.01×10−4) and bubbled with 95% O2/5% CO2. The base of the muscle was held by a Lucite clamp, and the other end was tied to a lever with an electromagnetic feedback system previously described12 to allow control of force, length, and velocity. Only muscles with a cross section <1.2 mm2 were accepted.
The muscles were stimulated at 6 stimuli/min and stabilized at Lmax for 15 minutes. Isometric, isotonic, and unloaded (Vmax) contractions were recorded as previously described.12
A cardiac inotropic concentration-response relationship was performed with Ang II (Sigma Chemical Co) by increasing doses from 1×10−10 to 1×10−5 mol/L. Then, to assess maximum contractile capacity, extracellular calcium concentration was increased to 10 mmol/L. At each concentration, isometric, isotonic, and unloaded muscle contractions were recorded.
An additional 6 papillary muscles from rabbits with pulmonary artery banding but no treatment had a phenylephrine concentration-response relationship (10−10 to 10−5 mol/L) done, in the presence of propranolol 10 μmol/L, to assess responsiveness of phospholipase C to stimulation by another receptor system.
Ang II AT1 and AT2 Receptor Binding Assays
[Sar1, Ile8]Ang II (sarile) was iodinated by a solid-phase method, and the monoiodinated peptide was purified by high-performance liquid chromatography.13 Radioligand binding assays were conducted as previously described.13 Equilibrium binding constants are reported as dissociation constant (Kd), and receptor concentrations are expressed as fmol/mg protein.
Immunoblot Analysis of G Protein α Subunits
Cardiac membranes were isolated as described.13 Equal amounts of membrane proteins (50 μg) were resolved on 10% acrylamide gels and analyzed by immunoblotting using the following polyclonal antibodies: anti-αq W082-8 (1:1000) and anti-αi1/2 SG2.5 (1:5000). Protein bands were quantified by densitometric analysis, and the area of each band was evaluated with NIH Image software.
Values are expressed as mean±SEM. Quantitative data were compared between groups by 1-way ANOVA and adjusted for inequalities of variances with a Brown-Forsythe’s test. Student’s t test or, when appropriate, a Tukey or Bonferroni t test was used to compare individual differences between groups. Statistical significance was determined by a value of P<0.05.
Pulmonary Artery Banding and Therapeutic Interventions
The survivals of the various pulmonary artery banded groups beyond the first 24 hours after operation were similar: untreated 94%, ramipril 83%, losartan 82%, and ramipril+kinin B1 and B2 receptor blockers 90%.
Hemodynamics and Measurements of Cardiac Hypertrophy
Heart rate was similar in all groups whether pulmonary artery banding had been performed or not (Table 1). RV systolic pressure and maximum rate of pressure rise (+dP/dt) and decline (−dP/dt) increased similarly in the pulmonary artery banded groups. RV end-diastolic pressure was unchanged.
The ratios of RV to body weight and heart to body weight were significantly increased in all banded groups (Table 2), although there was a tendency for this increase to be smaller in the treated groups.
Isolated Papillary Muscle Studies
Papillary muscles from untreated banded rabbits demonstrated marked depression of all tension-generating and shortening indices and lost responsiveness to Ang II (Tables 3 and 4⇓). Muscles responded normally to addition of calcium or phenylephrine 10−5 mol/L, however, indicating that the signaling pathway from phospholipase C to the myofibril was still functional. With the addition of phenylephrine, tension increased from 17 to 51 mN/mm2 and +dT/dt from 39 to 206 mN · mm−2 · s−1, both P<0.001.
Ramipril best preserved cardiac contractility, particularly in isometric contractions in which contractility and responsiveness to Ang II were normal. The beneficial effects of ramipril during isotonic contraction were less marked, but inotropic responsiveness to Ang II was maintained. Blockade of the kinin B1 and B2 receptors significantly attenuated the cardioprotective effects of ramipril. This was most obvious during isometric contractions in which values were essentially identical to those of the untreated banded group. Interestingly, the use of the kinin receptor blockers with ramipril also resulted in significant prolongation of twitch duration. Treatment with the kinin receptor blockers did not modify the normal response to Ang II and extracellular calcium.
The Ang II AT1 receptor blocker losartan had no obvious beneficial effect on contractile indices during baseline isometric or isotonic contractions. It did, however, preserve contractile responsiveness to Ang II. Interestingly, treatment with losartan resulted in the greatest prolongation of twitch duration (time to half tension decline from initiation of twitch of 671 ms).
Ang II AT1 and AT2 Receptor Binding Properties
AT1 receptor density was significantly upregulated in the banded group compared with control animals (Table 5). Treatment with ramipril completely prevented this upregulation of the AT1 receptor, whereas the combined administration of kinin receptor antagonists and ramipril only partially prevented the increased expression of the receptor (not significantly different from the banded group). Losartan also completely blocked the upregulation of AT1 receptors observed after pulmonary artery banding. No significant change in the affinity of losartan for the AT1 receptor was observed between the various experimental groups (Table 5). We also measured the expression of the AT2 receptor subtype in these membrane preparations. The AT2 receptor density was very low and represented <5% of the total 125I-labeled sarile binding activity. No detectable change in the abundance of AT2 receptors was observed between control rabbits or the various pulmonary artery banded rabbit groups (data not shown).
Analysis of G Protein α Subunit Expression
Pulmonary artery banding significantly increased the expression of αq and αi1/2 subunits (Figure). Treatment with ramipril completely reversed the increased expression of Gαi1/2 proteins but had no significant effect on the expression of Gαq. Interestingly, coadministration of kinin B1/B2 receptor antagonists with ramipril restored the high-level expression of Gαi1/2 subunits. In contrast to ramipril, treatment with losartan failed to prevent the upregulation of Gαi1/2 expression in the RV of banded rabbits.
This study demonstrates that pressure overload of the RV in the rabbit results in marked RV hypertrophy and depression of myocardial contractility. It also results in the nearly total loss of responsiveness to the inotropic effects of Ang II by a mechanism occurring downstream of the AT1 receptor, despite an increase of AT1 receptor density and G protein αq and αi1/2 subunit expression. Of the 2 therapeutic interventions evaluated, the one that best preserved cardiac contractility and best normalized AT1 receptor density and G protein expression was the ACE inhibitor ramipril. This difference compared with the AT1 receptor blocker losartan appeared to be largely due to its effect on kinins, because many of the beneficial effects of ramipril were eliminated by the simultaneous use of kinin receptor blockers. All therapeutic interventions preserved the contractile responsiveness to Ang II, suggesting that reduction in the production of Ang II or blockade of the AT1 receptor prevents the functional uncoupling of the receptor to downstream signaling events.
The importance of specific enzyme systems in the conversion of Ang I to Ang II varies according to the species and the tissue being studied.1,2 For this reason, in pilot studies (data not shown), we verified that in the rabbit, the conversion of Ang I to Ang II occurred at the cardiac level. We also determined that the major enzyme involved in that conversion was ACE, chymase being a less important enzyme for this conversion. The combined use of an ACE inhibitor and a chymase inhibitor resulted in much greater inhibition of the contractile effects of Ang I than either alone, however, indicating that when one pathway is inhibited, the other compensates by increasing the conversion rate of Ang I to Ang II.
Cardioprotective Effect of the ACE Inhibitor Ramipril Compared With the Ang II AT1 Receptor Blocker Losartan: Importance of Kinins
RV pressure overload led to significant RV hypertrophy and papillary muscle dysfunction. Despite this marked reduction in contractility, no in vivo reduction in RV function was documented, perhaps because of the contribution of compensatory hypertrophy and in vivo activation of compensatory mechanisms such as neurohumoral activation.
The ACE inhibitor ramipril had the most important cardioprotective effect of the 3 therapeutic interventions evaluated. Although ramipril tended to reduce RV hypertrophy, this did not reach statistical significance, perhaps because of the fixed afterload of pulmonary artery banding as reflected by similar RV systolic pressures in all groups, or the relatively short follow-up period. The lack of effect of losartan on RV hypertrophy in this model was reported by Koide et al.14 Ramipril nearly completely prevented the decrease of contractility, however, this effect being more marked for isometric contractions than isotonic contractions. The more marked effect on isometric contractions may result from changes in expression of myosin isozymes to a slower, more energy-efficient one in this model.10 As a result, contraction would be slowed and prolonged, such that tension-generating abnormalities would be minimized while abnormalities in velocity of shortening would be maximized.
The protective effects of ramipril on cardiac contractility were significantly reduced by the simultaneous infusion of kinin B1 and B2 receptor blockers. The mechanisms remain speculative and probably multifactorial; however, this may result in part from nitric oxide–mediated improvement in cardiac metabolism.15 With left ventricular hypertrophy, subendocardial blood flow reserve is reduced16 such that improved cardiac metabolism and coronary flow would be expected to improve cardiac energetic balance and be cardioprotective. The effects of ramipril without its kinin component were comparable to those of losartan, suggesting that the major cardioprotective effect of ramipril in this model is due to its effect on kinins.
Functional Uncoupling of the Ang II AT1 Receptor: Effects of ACE Inhibition and AT1 Receptor Blockade
There was complete functional uncoupling of the AT1 receptor from its contractile effects despite an increase in AT1 receptor density. In previous studies,17,18 AT1 receptor density and affinity were evaluated in various models of ventricular dysfunction and overload, but little information exists as to the functional ability of these receptors. One study in the pacing-overdrive model of heart failure demonstrated a loss of the positive inotropic effects of Ang II, but they did not measure the integrity of the Ang II signaling pathway.19 In the present study, we found significant upregulation of AT1 receptor expression despite functional uncoupling of the AT1 receptor, indicating a functional defect beyond the receptor. Although the pattern of uncoupling of Ang II to its effector organ response differs somewhat from that of the β-adrenergic system,20 the end result is protection against chronic overactivation of an endogenous neurohumoral system. Whether protection of the integrity of the AT1 signaling pathway with the various interventions that we tested is beneficial or potentially detrimental if the drug is discontinued, as occurs with β-blockers, remains to be determined, but a risk of rebound overactivation clearly exists.
We did not find any modification in AT2 receptor expression. This finding is consistent with previous studies in heart failure17 and cardiac hypertrophy,18 in which AT2 receptor density has been found to be unchanged. The expression of the G protein subunit αq, which is primarily responsible for coupling the AT1 receptor to phospholipase C-β and Ca2+ mobilization in cardiomyocytes,21 increased by 65% in banded rabbits, excluding the possibility that the loss of Ang II responsiveness was due to a decreased expression of Gαq. Also, the contractile effects of phenylephrine, another potent agonist of the Gαq–phospholipase C-β22 signaling pathway, were preserved, if not exaggerated, indicating that the signaling pathway from phospholipase C to the myofibril was intact.
Interestingly, all pharmacological interventions tested reestablished the cardiac responsiveness to Ang II, suggesting that the functional uncoupling of the AT1 receptor in the untreated banded group results from the local increase in Ang II levels and the chronic stimulation of the receptor. The most plausible explanation to reconcile all these observations is that the increased level of agonist in cardiac tissue uncouples the AT1 receptor from the Gq protein, resulting in functional desensitization of the receptor.
The expression of G protein αi subunits increases in heart failure and in various models of cardiac hypertrophy.23–25 We found that expression of the Gαi1/2 proteins was upregulated and that treatment with ramipril, but not with losartan or the combination of ramipril with kinin receptor blockers, completely prevented this increase. These results suggest that the increase in G protein αi1/2 subunits may be causally linked to the contractile dysfunction observed in these animals but not related to the desensitization of the AT1 receptor.
This work was supported in part by grants from the Medical Research Council of Canada (MT-8566 and MT-14168). S. Pelletier has a studentship from the Heart and Stroke Foundation of Canada. Dr Meloche is a Scientist of the Medical Research Council of Canada.
Akasu M, Urata H, Kinoshita A, et al. Differences in tissue angiotensin II–forming pathways by species and organs in vitro. Hypertension. 1998; 32: 514–520.
Urata H, Healy B, Stewart RW, et al. Angiotensin II–forming pathways in normal and failing human hearts. Circ Res. 1990; 66: 883–890.
Emanueli C, Maestri R, Corradi D, et al. Dilated and failing cardiomyopathy in bradykinin B2 receptor knockout mice. Circulation. 1999; 100: 2359–2365.
Weinberg EO, Schoen FJ, George D, et al. Angiotensin-converting enzyme inhibition prolongs survival and modifies the transition to heart failure in rats with pressure overload hypertrophy due to ascending aortic stenosis. Circulation. 1994; 90: 1410–1422.
Weinberg EO, Lee MA, Weigner M, et al. Angiotensin AT1 receptor inhibition: effects on hypertrophic remodeling and ACE expression in rats with pressure-overload hypertrophy due to ascending aortic stenosis. Circulation. 1997; 95: 1592–1600.
Alpert NR, Mulieri LA. Increased myothermal economy of isometric force generation in compensated cardiac hypertrophy induced by pulmonary artery constriction in the rabbit: a characterization of heart liberation in normal and hypertrophied right ventricular papillary muscles. Circ Res. 1982; 50: 491–500.
Gohlke P, Kuwer I, Schnell A, et al. Blockade of bradykinin B2 receptors prevents the increase in capillary density induced by chronic angiotensin-converting enzyme inhibitor treatment in stroke-prone spontaneously hypertensive rats. Hypertension. 1997; 29: 478–482.
Legault F, Rouleau JL, Juneau C, et al. Functional and morphological characteristics of compensated and decompensated cardiac hypertrophy in dogs with chronic infrarenal aorto-caval fistulas. Circ Res. 1990; 66: 846–859.
Lambert C, Massillon Y, Meloche S. Upregulation of cardiac angiotensin II AT1 receptors in congenital cardiomyopathic hamsters. Circ Res. 1995; 77: 1001–1007.
Koide M, Carabello BA, Conrad CC, et al. Hypertrophic response to hemodynamic overload: role of load vs. renin-angiotensin system activation. Am J Physiol. 1999; 276: H350–H358.
Zhang X, Xie YW, Nasjletti A, et al. ACE inhibitors promote nitric oxide accumulation to modulate myocardial consumption. Circulation. 1997; 96: 176–182.
Hettinger L, Shen YT, Patrick TA, et al. Mechanisms of subendocardial dysfunction in response to exercise in dogs with left ventricular hypertrophy. Circ Res. 1992; 71: 423–434.
Asano K, Dutcher DL, Port JD, et al. Selective downregulation of the angiotensin II AT1-receptor subtype in failing human ventricular myocardium. Circulation. 1997; 95: 1193–1200.
Lopez JJ, Lorell BH, Ingelfinger JR, et al. Distribution and function of cardiac angiotensin AT1- and AT2-receptor subtypes in hypertrophied rat hearts. Am J Physiol. 1994; 36: H844–H852.
Cheng CP, Suzuki M, Ohte N, et al. Altered ventricular and myocyte response to angiotensin II in pacing-induced heart failure. Circ Res. 1996; 78: 880–892.
Graham RM, Perez DM, Hwa J, et al. α1-Adrenergic receptor subtypes: molecular structure, function, and signaling. Circ Res. 1996; 78: 737–749.
Bohm M, Lippoldt A, Wienen W, et al. Reduction of cardiac hypertrophy in TGR(mREN2)27 by angiotensin II receptor blockage. Mol Cell Biochem. 1996; 163–164: 217–221.