Potassium Canrenoate, an Aldosterone Receptor Antagonist, Reduces Isoprenaline-Induced Cardiac Fibrosis in the Rat

The treatment of congestive heart failure has greatly improved with the introduction of inhibitors of angiotensin I-converting enzyme, β-adrenergic blockers, and blockade of aldosterone receptors by spironolactone, to judge not only from studies on experimental models of the disease but also from large-scale randomized clinical trials (Mason et al., 1979; DiBianco, 1990; RALES, 1996; Sleight, 2002). These studies suggest that the neurohormonal compensatory mechanisms that take place during the development of heart failure, in particular the activation of the renin-angiotensin-aldosterone system and the chronic increase in sympathetic tone (CIBIS-II, 1999), have long-term deleterious effects.

Activation of the sympathetic system results in greater release of norepinephrine, which in turn stimulates the β-adrenoceptors of the heart, especially the β1-adrenoceptors (Remme, 1986; van Zwieten and de Jonge, 1986). Such stimulation increases myocardial contractility and tends to restore “normal” hemodynamic conditions despite the deep alteration in cardiac function. This initial benefit is, however, counterbalanced by the induced increase in cardiac work and the potential chronic ischemia, which progressively leads to loss of contractile tissue, apparition of reparative fibrosis, and further alteration of left ventricular function.

The stimulation of β-adrenergic receptors on fibroblasts and the resulting increase in cAMP level can directly induce myocardial fibrosis (Kahn et al., 1969) in particular of the interstitial type. This fibrosis may also involve the action of the angiotensin II and aldosterone receptors. Indeed, both aldosterone and angiotensin II can induce cardiac fibrosis (Brilla and Weber, 1992; Brilla et al., 1994). Because β-adrenergic stimulation induces renin release and therefore increases the plasma levels of angiotensin II and aldosterone, part of the fibrotic process induced by β-adrenergic stimulation could be related to the overproduction of aldosterone and angiotensin II. Therefore, the aim of the present study was to determine whether aldosterone receptor blockade with potassium canrenoate (PC) reduces the cardiac remodeling and fibrosis induced by the acute administration of isoprenaline (ISO) in the rat.

Materials and Methods

Animals

Normotensive male Wistar rats (body weight 300-324 g, age 8 weeks) were purchased from CERJ (Saint Berthevin, France). All the procedures and protocols involving animals and their care were conducted in conformity with the institutional guidelines that are in compliance with national and international laws and policies (Council directive #87.848,October 19, 1987, Ministère de l'Agriculture et de la Forêt, Service vétérinaire de la Protection Animale, permission #0299 to M.H.) and were approved by the institutional animal care committee.

Characterization of the Experimental Model

Dose of Isoprenaline. The experimental model used was a modification of that proposed by Grimm et al. (1998), inducing cardiomyopathy with subcutaneous injection of high doses of isoprenaline in the female rat. In a preliminary study, the most effective dose inducing a significant myocardial fibrosis 2 months after the injection of isoprenaline was determined: the animals (male Wistar rats) received one, two, or three subcutaneous injections of isoprenaline with an interval of 1 week between the injections. Three doses of isoprenaline were tested: 150, 200, and 400 mg/kg. Extent of cardiac fibrosis (preferential location at the apex and at the endocardial level) was very variable for each dose level but clearly increased with isoprenaline dose. Therefore, to induce the largest extent of cardiac fibrosis in the surviving animals, 400 mg/kg was chosen. Thus, this dose was selected for the present study despite a high mortality rate, which was not considered as a limitation for this study.

Telemetry in Conscious Rats. The amplitude and duration of the effects of the selected subcutaneously administrated 400 mg/kg isoprenaline dose was evaluated by recording heart rate in three additional surviving conscious rats using telemetry captors (Physio Tel Telemetry System ETA-F20; Data Sciences International, St. Paul, MN) placed into the abdominal cavity and fixed by ligature. The administration of isoprenaline was performed 15 days after surgery. Recordings of the electrocardiogram were obtained for 3 min repeated each day three times (at 10, 13, and 16 h) with a specific acquisition board connected to Snap Master software. Recordings were interpreted with DADISP software (Data Science International telemetric system).

β-Adrenergic Responsiveness. To study the level of desensitization of cardiac β-adrenoceptors 2 months after the acute administration of high-dose isoprenaline, dose-response curves to isoprenaline on heart rate (0.0025-10 μg/kg) were obtained in control rats (n = 7) and in rats having received 400 mg/kg isoprenaline (n = 12). The animals were anesthetized with pentobarbital and the left femoral vein was cannulated for isoprenaline injections. Heart rate response was studied using ECG recording (Gould RS 3200). The EC₅₀ and the maximal effect of isoprenaline were determined using an E_max model (Sigma Plot software).

Choice of Potassium Canrenoate Dose. To study the effects of blockade of cardiac aldosterone receptors on the isoprenaline-induced fibrosis, potassium canrenoate was used as aldosterone antagonist (20 mg/kg/day in drinking water) (Araki et al., 1995). Such dose completely inhibits cardiovascular effects of aldosterone. However, because extracardiac effects of potassium canrenoate, in particular the diuretic effects, could participate indirectly to the cardiac remodeling, we wanted to exclude such diuretic effects and checked that at the chosen dose of potassium canrenoate, no diuretic effect was induced in normal rats. Thus, 12 rats were submitted to a 1-wk treatment with 20 mg/kg potassium canrenoate in drinking water. At the end of such treatment period, animals were placed in metabolic cages during 3 h for diuresis, natriuresis, and kaliuresis measurements compared with a control group (n = 12). No significant modification of diuresis, natriuresis, and kaliuresis was observed with this selected dose of potassium canrenoate in those normal rats (data not shown). Thus, with 20 mg/kg potassium canrenoate, we can assume that the observed cardiovascular do not depend on a diuretic action.

Study Design and Evaluation Parameters

Study Design. Animals were divided into four groups: two control groups with or without potassium canrenoate, and two groups of rats submitted to isoprenaline (one subcutaneous dose of 400 mg/kg) with or without potassium canrenoate, which was initiated 5 days before injection of isoprenaline. The rats were followed for 8 weeks and then noninvasive and invasive cardiovascular investigations were carried out before sacrifice of animals. The number of animals in each group was 12, 10, 25, and 25, respectively, for control, control + PC, Iso 400, and Iso 400 + PC group.

Noninvasive Measurements. The left ventricular dimensions were assessed in vivo under pentobarbital anesthesia (50 mg/kg intraperitoneally) by echocardiographic examination using a linear probe emitting ultrasounds at 7 to 10 MHz (ACUSON 128 XP; Acuson Corporation, Mountain View, CA). The two-dimensionally guided M-mode recording of the left ventricle provided the measurements of left ventricle cavity dimensions and wall thicknesses: internal end-diastolic diameters (EDD), end-systolic diameters (ESD) and fractional shortening (FS = ((EDD-ESD)^*100)/EDD), and diastolic and systolic thicknesses of septum and posterior wall. All measurements were the mean of at least three independent measures.

Invasive Measurements. Once echocardiographic measurements were performed, the animals still being under pentobarbital anesthesia, a micro-tip pressure transducer catheter (2-French; Millar Instruments Inc., Houston, TX) was inserted into the right carotid artery and connected to a Gould recorder (model SR 3200). The following pressures were obtained: aortic blood pressure, left ventricular pressure, and maximal positive and minimal negative left ventricle dP/dt. The heart rate was determined from the ventricular pressure tracing.

Morphometric Analysis. After completion of hemodynamic measurements, the heart was removed rapidly, weighed, and the heart-to-body weight ratio was calculated. All ventricles were immersed into 10% buffered formalin, dehydrated in 95°C ethanol, and then in acetone. They were impregnated with methyl salicylate and embedded in paraffin. Three sections of 6 μm were obtained: one at the basis, one at the middle of the ventricles, and one at the apex.

Quantification of Fibrosis (Collagen Extent). Sirius red staining was used to characterize collagen tissue (Junqueira et al., 1978). Sirius red-stained slides of left ventricle, were placed under a 3CCD color camera (KY-F55B; JVC, Tokyo, Japan) that was connected to a quantimeter Qwin (Leica, Rueil Malmaison, France) with the acquisition software Leica Win (version 2.2; Leica Microsystems, Rueil-Malmaison Cedex, France). The extent of fibrosis was determined by quantification of the red intensity and was expressed as a percentage of the entire area of the left ventricle.

Hormonal Dosages. At the time of sacrifice, blood samples were taken, and plasma was extracted after centrifugation (4°C during 10 min, 4000g) for determination of plasma levels of renin activity, angiotensin I, and aldosterone by radioimmunoassay using commercially available kits (Schaison et al., 1996).

Drugs

Isoprenaline (dl-isoproterenol hydrochloride) and potassium canrenoate were purchased from Sigma (Saint, Quentin Fallavier, France). Sodium pentobarbital (SANOFI Santé Animal, Montpellier, France) was used in injectable solution. Isoprenaline was prepared under sterile conditions with distilled water.

Statistical Analysis

The statistical analysis of each parameter was carried out with a two-factor variance analysis using as factors isoprenaline administration and potassium canrenoate administration. For the analysis of extent of fibrosis, a third factor was used: the level of the section (basis, medium, and apex). Between-group comparisons were performed using a Newman-Keuls test.

For comparison of mortality rates between the different groups, a chi square test was used. p values of less than 0.05 were considered to be statistically significant.

Results

Preliminary Experiments: Cardiac Effects of Isoprenaline

Telemetry, Holter Recording. The subcutaneous injection of isoprenaline (400 mg/kg) induced an important increase in heart rate (Fig. 1), from 358 ± 2 beats/min at baseline to 536 ± 14 beats/min 5 min after the first injection of isoprenaline. Heart rate returned to baseline values 3 days after injection (358 ± 14 beats/min).

Dose-Response Curve to Isoprenaline (Fig. 2). The maximal chronotropic effect of isoprenaline 2 months after the acute isoprenaline administration was reduced: for the control group, E_max (heart rate maximal increase) was 133 ± 27 beats/min and the EC₅₀ was 0.02 ± 0.02 μg/kg; and for the isoprenaline-treated group, E_max was 84 ± 18 beats/min and EC₅₀ was 0.01 ± 0.01 μg/kg. Difference between groups were significant for E_max (p < 0.001) but not for EC_50.

Mortality Rate

During the 2 months of study duration, no death was recorded in the two control groups but seven animals died in the isoprenaline-treated group (27% death rate) and 15 died in the isoprenaline + potassium canrenoate-treated group (52% death rate), with p = 0.07 between these last two groups (chi square test).

Some animals died immediately after onset of anesthesia or during hemodynamic investigation, explaining the different group numbers for analysis of noninvasive and invasive cardiovascular measurements.

Morphological Parameters (Table 1)

Heart and left ventricular weights were increased significantly by isoprenaline. Such increase was prevented partly by the administration of potassium canrenoate.

Echocardiographically measured at the level of end of mitral valve chordae (mid-ventricle) end diastolic diameter and end-systolic diameter were increased significantly in the isoprenaline-treated group and such enlargement was enhanced further by administration of potassium canrenoate. Between these last two groups (isoprenaline and isoprenaline + potassium canrenoate), the difference was however significant only for EDD (p < 0.003). The left ventricular shortening fraction was significantly reduced by the administration of isoprenaline compared with control groups.

Hemodynamic Parameters (Table 2)

No significant difference was found between groups for the systolic, diastolic, and mean arterial blood pressures; the left ventricular systolic and end-diastolic pressures; heart rate; or the dP/dt+. A significant reduction of dP/dt- was observed in the isoprenaline + potassium canrenoate group compared with the control group.

Extent of Fibrosis

The left ventricular area was significantly increased with potassium canrenoate and isoprenaline alone or in combination (Fig. 3). The extent of fibrosis was significantly increased in the isoprenaline-treated group but such increase was partly but significantly prevented by potassium canrenoate at the three levels studied (Fig. 4). Isoprenaline-induced fibrosis preferentially developed at the apex compared with the basal level (p < 0.001).

Plasma Neurohormone Concentrations (Table 3)

Plasma renin activity and plasma angiotensin I levels were increased significantly in rats submitted initially to the injection of isoprenaline. No statistical difference was recorded between the other groups.

Plasma levels of aldosterone were elevated significantly in the group of animals submitted to isoprenaline administration and treated with potassium canrenoate (p < 0.05 versus isoprenaline-treated group).

Discussion

The present study shows that antagonism of aldosterone receptors with potassium canrenoate interacts with cardiac remodeling after the acute administration of high doses of isoprenaline. The drug reduces the extent of fibrosis and such effect is associated with a greater left ventricular enlargement.

Isoprenaline is a nonselective agonist of β1- and β2-adrenoceptors and can exert toxic effects by several mechanisms, including intracellular cAMP increase, calcium overload (Dhalla et al., 1996), alteration of electrophysiological properties of cardiomyocytes (Hart, 1994), ischemia, and increased oxidative stress (Bindoli et al., 1992; Remiao et al., 2001). The latter two alterations can lead to the destruction of myocytes either by apoptosis or necrosis, with loss of contractile tissue, fibrosis (reparative fibrosis), and severe arrhythmias. Indeed, following the acute administration of such high doses of isoprenaline, a trend toward a mortality increase was observed, whether isoprenaline was injected alone or in association with potassium canrenoate. The acute administration of isoprenaline induces a cardiac remodeling process with a moderate degree of cardiac hypertrophy and dilatation, as described previously (Grimm et al., 1998), with an important interstitial fibrosis, especially at the subendocardial and apical cardiac levels, in confirmation of previous studies (VanVleet et al., 1983). Ng et al. (2002) have shown that the subendocardial layers of the left ventricle are more sensitive to low doses of isoprenaline than the epicardial layer. At the same dose (5 mg/kg) of isoprenaline, the necrosis was 10-fold higher at the endocardial level than that at epicardial level (Ng et al., 2002). The density of adrenoceptors is 7- to 8-fold higher in the heart than in skeletal muscle (Rothwell et al., 1987). A high density of the β-adrenoceptors has been observed in the endocardial layers of the heart (Murphree and Saffitz, 1989). The abundance of these receptors can explain the larger response of the heart to isoprenaline. One major mechanism underlying this cardiac toxicity is likely to be the myocardial ischemia secondary to the severe increase in myocardial oxygen consumption during β-adrenergic stimulation, which is maintained during 48 h according to the Holter recordings obtained in the present study. Indeed, both the chronotropic and the inotropic actions of β-adrenergic stimulation impact on cardiac oxygen consumption and have the potential to induce sustained ischemia, leading to necrosis of myocytes and reparative fibrosis (Benjamin et al., 1989). However, reparative fibrosis may not be the only mechanism for the observed fibrosis because a direct activation of β-adrenergic receptors of fibroblasts by isoprenaline may induce collagen synthesis. Thus, two types of fibrosis can occur: reactive fibrosis involving the intramyocardial, perivascular, or interstitial spaces, and reparative fibrosis, which is a scar tissue replacing necrotic cardiomyocytes (Weber et al., 1990). Because fibroblasts can be also activated by angiotensin II and aldosterone, and because β-adrenergic stimulation increases renin release by the juxtaglomerular apparatus (van Zwieten and de Jonge, 1986; Vatta et al., 1992), the objective of the present study was to test the interaction between β-adrenergic stimulation and aldosterone action.

The present experiments show that blockade of aldosterone receptors with potassium canrenoate partly reduces the isoprenaline-induced fibrosis. The present results confirm previous experiments by Gallego et al. (2001) who also found that spironolactone significantly reduce isoprenaline-induced fibrosis but not captopril. Their experimental conditions were however different and spironolactone was only active at very high doses (200 mg/kg subcutaneously). The effectiveness of potassium canrenoate at 20 mg/kg in drinking water in our experiments is in favor of a better bioavailability compared with that of spironolactone (used in Gallego experiments) and which is extremely difficult to solubilize at these concentrations. Grimm et al. (1998) reported only a transient increase in components of the renin-angiotensin system after isoprenaline and a partial inhibition of cardiac fibrosis by ramipril. Our results, along with those of Gallego et al. (2001) and Grimm et al. (1998), argue in favor of a major role for mineralocorticoïd receptor stimulation in the isoprenaline-induced cardiac fibrosis secondary to fibroblast stimulation in addition to the reparative fibrosis. They also suggest that part of the isoprenaline-induced fibrosis is mediated by circulating aldosterone, resulting from β-adrenergic induced renin release. Aldosterone may potentiate catecholamines by reducing their uptake from the extracellular space, which is an essential step for the disposition of adrenaline and noradrenaline (Yamamoto et al., 1976). Aldosterone stimulates fibroblasts and collagen synthesis by activating metalloproteinases (Funck et al., 1997). In particular, aldosterone induces an increase of messenger RNA for type I and III collagen in myocytes in both ventricles (Robert et al., 1994).

To judge from experiments in rats and in humans, tissues other than the adrenal gland can produce aldosterone, including the heart itself (Silvestre et al., 1998), vascular smooth muscle (Takeda et al., 1995), and the brain (Gomez-Sanchez et al., 1997). However, the relative importance of the extra-adrenal gland production of aldosterone remains unknown, and the adrenal gland remains the major source of aldosterone because adrenalectomy almost abolishes the plasma levels of the hormone (Rocha et al., 2000).

Potassium canrenoate is a competitive antagonist of aldosterone, used as a diuretic in the treatment of hypertension. Although spironolactone seems to have beneficial therapeutic effects in humans as demonstrated in the RALES study by inhibiting the effects of aldosterone on cardiac remodeling, potassium canrenoate, its main metabolite, may have direct toxic effects. Indeed, potassium canrenoate is genotoxic (Martelli et al., 2002) and induces DNA damage in different tissues, including the liver, thyroid, brain, and mammary gland (Cook et al., 1988). Studies in primary cultures of rat hepatocytes demonstrate that potassium canrenoate induces DNA fragmentation and DNA repair (Martelli et al., 1999). These potential toxic effects of potassium canrenoate may explain the observed trend of survival reduction in the rats treated with combination of potassium canrenoate and isoprenaline.

Together with the partial inhibition of fibrosis, the treatment with potassium canrenoate was associated with a greater enlargement of the left ventricle, as shown by echocardiographic measurements of left ventricular dimensions in vivo. These findings may have important implications in the treatment of heart failure because an accurate balance must be obtained between the prevention of fibrosis and prevention of left ventricular dilatation. Therefore, the combined actions of angiotensin II and aldosterone may represent a relevant mechanism for remodeling of the extracellular matrix in the myocardium.

In conclusion, the present study shows that the acute administration of high doses of isoprenaline induces a cardiac remodeling with dilatation and fibrosis of the left ventricle, which is partly reduced by an antagonist of aldosterone receptors. The potential role of the interaction between β-adrenergic stimulation and aldosterone remains to be established in human heart failure where both the sympathetic tone and the renin-angiotensin-aldosterone system are activated as compensatory mechanisms. The present results argue in favor of the synergic therapeutic action of blockade of both β-adrenergic and aldosterone receptors in heart failure. No randomized trial, however, has tested, in chronic heart failure, the benefit-risk ratio of a systematic combination of β-blockers and aldosterone antagonist.