INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE PROGNOSIS of patients with heart failure remains poor, although chronic treatment with angiotensin-converting enzyme (ACE) inhibitors (8, 37, 42) and, more recently, -receptor blockers (4, 6, 44) may reduce mortality. The mechanisms of the protective effects of ACE inhibitors and -receptor blockers are only partially understood and are, almost certainly, complex and multifactorial. One attractive hypothesis is that the failing heart is energy depleted (16, 19, 20, 30), and that the beneficial functional effects of ACE inhibitors and -receptor blockers are accompanied and possibly accounted for, by a preservation of energy metabolism. Although a large number of experimental studies have demonstrated the protective functional effects of ACE inhibitors and, to a lesser extent, of -receptor blockers in heart failure models, only sporadic evidence exists on the energetic effects of these compounds in heart failure (24, 36, 45). The poor prognosis of patients with heart failure necessitates the search for novel therapeutic approaches, and the demonstration of a protective effect of established heart failure drugs such as ACE inhibitors or -receptor blockers on energy metabolism may fuel the search for therapies targeted more specifically to cellular systems involved in the regulation of energy metabolism.

The purpose of this study was thus to define the geometric, functional, and energetic cardiac effects of chronic treatment with the -receptor blocker bisoprolol and the ACE inhibitor captopril in a clinically highly relevant model of cardiac dysfunction, occurring postmyocardial infarction (MI) in the rat. Using a combination of nuclear magnetic resonance (NMR) spectroscopy, high-performance liquid chromatography (HPLC), and enzyme analysis, we define steady-state levels, turnover rates, and free energy change of high-energy phosphates as well as tissue activities (V_max) of creatine kinase (CK) isoenzymes.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals and experimental MI. Infarcts or sham operations were performed in 12-wk-old Wistar rats kept in a 12:12-h light-dark cycle. Left coronary artery ligation (MI) was induced by a previously described technique (11, 35). A left thoracotomy was performed under ether anesthesia and positive pressure ventilation. The heart was rapidly exteriorized by applying gentle pressure on both sides of the thorax. The left coronary artery was ligated between the pulmonary outflow tract and the left atrium. The heart was then replaced into the thorax, lungs were inflated by increasing positive end-expiratory pressure, and the wound was closed immediately. Sham operation was performed using an identical procedure except that the suture was passed under the coronary artery without ligation. Mortality rate of infarcted rats for the first 24 h after the operation was 40-50%. Surviving rats were kept on commercial rat chow and water ad libitum. The investigation conformed with the "Guiding Principles for Research Involving Animals and Human Beings."

Bisoprolol and captopril treatment and experimental groups. Rats were randomly assigned to one of six groups: untreated sham (sham, n = 9), untreated MI (MI, n = 8), bisoprolol-treated sham (sham-Biso, n = 13), bisoprolol-treated MI (MI-Biso, n = 6), captopril-treated sham (sham-Capto, n = 11), and captopril-treated MI (MI-Capto, n = 6). After surgery, bisoprolol-treated groups received 60 mg · kg¹ · day¹ bisoprolol with the drinking water. Captopril was added to the drinking water at 2 g/l. These concentrations were chosen because they were shown to exert a small but significant hemodynamic effect [10% reduction of blood pressure (captopril) or heart rate (bisoprolol) (3, 24)]. Because bisoprolol has a slightly bitter taste, 5% glucose was added to the drinking water of all groups. Drinking water was freshly prepared every other day. Therapy was continued for 8 wk and was stopped 1 day before the isolated heart experiment.

Isolated rat heart preparation. Eight weeks after the left coronary artery ligation or sham operation, rats were anesthetized with the injection of 50 mg pentobarbital sodium intraperitoneally. A transverse laparotomy and left and right anterolateral thoracotomy were performed, and the heart was rapidly excised and immersed in ice-cold buffer. The aorta was dissected free and mounted onto a cannula attached to a perfusion apparatus, as previously described (2). Retrograde perfusion of the heart was begun in the Langendorff mode at a constant temperature of 37°C and at a constant coronary perfusion pressure of 100 mmHg. A small vent made out of polyethylene tubing was pierced through the apex of the left ventricle to allow drainage of flow from Thebesian veins. For control perfusion, phosphate-free Krebs-Henseleit buffer was used containing (mM) 118 NaCl, 4.7 KCl, 1.75 CaCl₂, 1.2 MgSO₄, 0.5 ethylenediaminetetraacetate tetrasodium, 25.0 NaHCO₃, and 11.0 glucose. Equilibrating the buffer with 95% O₂-5% CO₂ yielded a pH of 7.4. Coronary flow was continuously measured by an ultrasonic flow probe (Transonic Systems, Ithaca, NY) built into the perfusate inflow tubing. As previously shown (2), the perfusion system allowed maintenance of hearts in a steady state for at least 90 min with changes of <5% for all mechanical and metabolic parameters.

Cardiac performance measurements. A water-filled latex balloon was inserted into the left ventricle through an incision in the left atrial appendage, via the mitral valve, and secured by a ligature. The balloon was connected to a Statham P23 Db pressure transducer (Gould Instruments, Glen Burnie, MD) via a small-bore polyethylene tubing for continuous measurement of left ventricular pressure and heart rate on a four-channel recorder (Graphtec, Tokyo, Japan). Performance was estimated as the product of heart rate and left ventricular developed pressure (LVDP, mmHg/min). Frank-Starling curves were obtained by increasing the volume of the intraventricular balloon by 0.05-ml increments until LVDP reached a maximum.

³¹P NMR spectroscopy. The perfused hearts were placed into a 20-mm NMR sample tube and inserted into a probe seated in the bore of a superconducting superwide bore (150 mm), 7.05-Tesla magnet (Bruker, Rheinstetten, Germany) as previously described (30). Hearts were bathed in their own perfusate, which was pumped from the NMR tube at a level immediately above the heart. An Aspect 3000 computer (Bruker) was used in the pulsed Fourier transform mode to generate ³¹P NMR spectra at 121.50 MHz. A 14-channel Shim Unit served to homogenize the magnetic field. Single ("one pulse") spectra were accumulated over 5-min periods, and averaging data from 152 free induction decays were obtained using a pulse time of 37.6 µs, a pulse angle of 45°, and an interpulse delay of 1.93 s. The resonance areas corresponding to ATP, phosphocreatine (PCr), and P_i, which are proportional to the number of phosphorus atoms of the respective compound, were measured by integration using the NMR1 software (TRIPOS, Munich, Germany). Relative saturation factors for each resonance were determined by comparing spectra to fully relaxed spectra obtained using a pulse angle of 45° and an interpulse delay of 15 s; correction factors were 1.12 ± 0.03 for PCr and 1.08 ± 0.05 for P_i (n = 15); spectra were corrected accordingly. Myocardial ATP content was found to be 28.3 ± 5.6 nmol/mg protein in sham and 28.0 ± 3.0 nmol/mg protein in MI hearts by HPLC (30); the myocardial contents of the other metabolites were calculated by multiplying the ratio of resonance area of metabolite to the [-P]ATP area by 28.3 nmol/mg protein for sham and by 28.0 nmol/mg protein for MI hearts. This assumes that neither bisoprolol nor captopril affected normal ATP content, demonstrated before for sham-operated and infarcted hearts (30), which is most likely the case. ATP could not be determined by HPLC in this study, because left ventricles were fixed in Formalin for infarct size measurements, and right ventricles were cut off and frozen allowing determination of total creatine and CK activity. ATP levels were not determined in the right ventricle because we cut off the right ventricle from the beating heart and froze it after several seconds, because freeze-clamping the beating heart was impossible when we wanted to determine infarct size. In our experience, ATP levels cannot reliably be determined when tissue is not directly freeze-clamped. Intracellular pH (pH_i) was measured by comparing the chemical shift between P_i and creatine phosphate with values obtained from a standard curve. The free cytosolic ADP concentration was calculated by using [ATP], [PCr], and [H⁺] measured in the intact beating heart by ³¹P-NMR spectroscopy, and total creatine was measured chemically in the right ventricular homogenates by assuming that CK is in equilibrium and by using a CK equilibrium constant of 1.66 × 10⁹ M¹ (25, 43)
The free energy change of ATP hydrolysis (G) was calculated as
where G° (30.5 kJ/mol) is the value of G under standard conditions of molarity, temperature, pH, and [Mg²⁺] (14); R is the gas constant (8.3 J/mol K); and T is the temperature in Kelvin (K).

³¹P-NMR magnetization transfer measurements of CK kinetics. For magnetization transfer experiments, each broad-band pulse was preceded by a low-power, narrow-band pulse at the resonance frequency of [-P]ATP for 0, 0.3, 0.6, 1.2, 2.4, or 3.6 s as previously described (29, 30). Separate studies showed that the narrow-band pulse directly attenuated the PCr magnetization by <5% when the carrier frequency was placed 2.5 ppm downfield from the resonance of PCr. For each of the six saturation transfer spectra, 64 scans were accumulated by repetitively cycling through the six different times of presaturation. Thus any metabolic deterioration occurring during the saturation transfer measurement was equally distributed among the spectra. A complete saturation transfer experiment was acquired in 32 min. Stability of the preparation was assessed by comparing one-pulse spectra obtained before and after each magnetization transfer experiment. Magnetization transfer measurements of the forward CK reaction phosphocreatine  [-P]ATP were analyzed according to the two-site chemical exchange model of Forsen and Hoffman (9), providing estimates of the pseudo first-order rate constant (k_for) and the intrinsic longitudinal relaxation time for PCr (T₁). Briefly, as the time of saturation at [-P]ATP, t, is increased from 0 to 3.6 s, the integrated signal intensity of the PCr resonance peak (M_t) decays from M₀ to M (defined as magnetization at zero and infinite saturation times, respectively) with a time constant ₁ as
where
Nonlinear regression of PCr magnetization at different saturation times of [-P]ATP was used to determine M₀, M, and ₁. T₁ and k_for were then calculated by solving the equations
and
Multiplying the rate constant by substrate concentration yielded reaction velocity (2).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Heart weight, body weight, and infarct size. Table 1 shows infarct size, body weight, heart weight, heart-to-body weight ratios, and left and right ventricular weight of the six groups studied. In the three infarcted groups, infarct size was ~40% of the left ventricular circumference and was not significantly different. There was a tendency for body weight reduction in captopril-treated, sham-operated rats and in the bisoprolol-treated MI group. With bisoprolol the increase in heart weight and left ventricular weight, which occurs in infarcted hearts, was no longer significant. Heart weight was substantially reduced in both sham-operated and infarcted captopril-treated rats. Infarction led to an increase in right ventricular weights, which was prevented by captopril (P < 0.007) but not significantly by bisoprolol.

Cardiac performance and pressure-volume relations. Table 2 shows heart rate, coronary flow, and LVDP of the six experimental groups, all recorded at an end-diastolic pressure of 10 mmHg. On average, heart rate was 273 ± 3 beats/min and was not significantly different among groups. In contrast, LVDP was significantly reduced in chronically infarcted hearts (85 ± 11 vs. 135 ± 9 mmHg in sham, P < 0.007). Although treatment did not affect LVDP in sham-operated hearts, bisoprolol completely and captopril treatment partially prevented the decrease of LVDP in infarcted hearts. Coronary flow was 23 ± 1 ml/min on average and was not significantly different among groups.

Figure 1 shows pressure-volume relations for LVDP (Fig. 1A) and left ventricular end-diastolic pressure (Fig. 1B) in the six groups. The LVDP curve for the infarcted untreated group was shifted to the right with a significantly reduced maximum developed pressure (Table 2) compared with LVDP curve for the sham-operated untreated group. In infarcted hearts, the shift of developed pressure-volume curves was partially prevented by bisoprolol and captopril treatment. Similarly, the reduction of maximum LVDP was completely prevented by bisoprolol, whereas captopril showed only a trend to increase maximum LVDP (Table 2). Bisoprolol shifted the developed pressure-volume curves of sham-operated hearts somewhat upward. The end-diastolic pressure-volume curves for the infarcted groups were also shifted rightward and downward and were not significantly affected by treatment. However, the shift between captopril-treated, sham-operated and infarcted hearts did not reach significance, an indication that structural dilation was less than in untreated or bisoprolol-treated infarcted hearts.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   A: pressure-volume curves for left ventricular developed pressure (LVDP) of all experimental groups. B: pressure-volume curves for left ventricular end-diastolic pressure (EDP) of all experimental groups. MI, myocardial infarction; Capto, captopril; Biso, bisoprolol. * P < 0.007 sham vs. MI,  P < 0.007 Biso- or Capto-treated vs. untreated.

High-energy phosphate metabolism. Representative ³¹P NMR spectra from the various groups of hearts are shown in Fig. 2. Mean values for high- and low-energy phosphates and pH_i are given in Table 3. P_i resonances were comparable in sham and chronically infarcted hearts, but there was a significant reduction of the PCr resonance in untreated chronically infarcted hearts (10.6 ± 0.6 vs. 13.5 ± 0.7 mM in untreated sham-operated hearts, P < 0.007). With bisoprolol treatment, PCr remained almost unchanged and also captopril could limit the extent of the PCr reduction. There was no effect of treatment on ³¹P metabolites in the sham-operated groups. The total creatine pool tended to be reduced by chronic infarction (19.7 ± 1.2 vs. 23.0 ± 1.0 mM in sham-operated rats). Treatment with bisoprolol prevented this reduction, whereas captopril did not. Free creatine was, on average, 9.4 ± 0.5 mM and tended to reduce in infarcted hearts but not in treated hearts. Intracellular pH was 7.15 ± 0.00, and the values were comparable among sham and infarcted, treated and untreated, hearts. Free cytosolic ADP concentrations were, on average, 89 ± 5 µM, and G values were 59.5 ± 0.4 kJ/mol. TANs measured in the right ventricle were 32.1 ± 1.0 nmol/mg protein on average. There were no significant differences among groups (Table 3).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Representative 31P nuclear magnetic spectroscopy (NMR) spectra of all experimental groups. PCr, phosphocreatine.

CK reaction velocity. Figure 3 depicts stacked plots of ³¹P NMR spectra obtained from saturation transfer experiments for sham-operated untreated, infarcted untreated, infarcted bisoprolol-treated, and infarcted captopril-treated hearts. The spectra show that the extent of saturation transfer from the PCr to [-P]ATP resonance is reduced in infarcted hearts and that this reduction is partially prevented by treatment either with bisoprolol or captopril.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Stacked plots of 31P-NMR spectra from saturation transfer experiments of an untreated sham-operated (A) and infarcted heart (B), and a Biso (C)- and Capto (D)-treated infarcted heart. For reasons of clarity, only PCr and -ATP resonances are shown.

Mean data from saturation transfer experiments are shown in Table 3 and in Fig. 4. On average, the T₁ of PCr was 3.49 ± 0.14 s and was not different among groups (Table 3). There was a trend for a decrease of the CK rate constant in all infarcted groups. CK reaction velocity was reduced in all infarcted hearts but significantly less so in hearts treated with either bisoprolol or captopril (Fig. 4). However, the reduction post-MI did not reach significance in the bisoprolol-treated group. Thus energy reserve via CK remained at higher levels when infarcted hearts were treated with either the -blocker bisoprolol or the ACE inhibitor captopril.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Mean creatine kinase reaction velocity (CK flux) of all experimental groups. * P < 0.007 sham vs. MI;  P < 0.007 Biso- or Capto-treated vs. untreated.

Biochemical analysis. Table 4 summarizes the results of enzyme analysis of right ventricular homogenates. Total CK activity showed a trend for reduction after MI from 8.1 ± 0.5 to 6.0 ± 0.7 IU/mg protein in untreated but not in hearts treated with bisoprolol or captopril. CK isoenzyme distribution showed a trend for the changes characteristic of failing myocardium: a decrease of the absolute MM-CK and mitochondrial CK activities and an increase of the relative activities of fetal -containing isoenzymes. Treatment with bisoprolol or captopril completely prevented all changes in total CK and isoenzyme activities after MI both in absolute and relative terms. Activities of the mitochondrial enzyme citrate synthase and of the glycolytic enzyme lactate dehydrogenase did not change in any experimental group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Definition of model. The present study defines cardiac function and energy metabolism under various forms of treatment in a clinically highly relevant model of heart failure, occurring post-MI. For untreated groups, changes of cardiac geometry and function were as previously reported (23, 30). End-diastolic and systolic pressure-volume relations were shifted to the right in infarcted hearts, indicating structural dilatation (40). At the same time, maximum LVDP was substantially reduced, attesting to contractile dysfunction. This model is well suited to study beneficial or adverse effects of various forms of pharmacological treatment (10, 17).

Similarly, changes in cardiac energy metabolism in untreated infarcted hearts were as previously reported: steady-state TAN levels are unaltered, whereas PCr is reduced by 22% and total creatine content by 14% (20, 21, 30). On the basis of CK equilibrium assumptions, we calculated unchanged free ADP and G levels at least for the baseline performance conditions studied here. CK reaction velocity, a measure of ATP transfer from mitochondrium to myofibrils (21), was reduced by 41%. Also, changes in total CK and CK isoenzyme distribution, measured in the right ventricle, were as previously described (23, 30), indicating that the chronically failing heart is characterized by reduced MM-CK and mitochondrial CK activity. The alterations of energy metabolism found here are characteristic of many models of cardiac hypertrophy and failure, i.e., they occur independent of the etiology of heart failure (5, 20, 21, 28).

Effects of -receptor blockade and ACE inhibition on cardiac geometry and function. In untreated infarcted hearts, substantial left and right ventricular hypertrophy occurred, as indicated by increased left and right ventricular weights occurring despite the loss of ~40% of left ventricular tissue due to infarction. The increase in heart weight was reduced by both captopril and bisoprolol treatment, most effectively in the right ventricle, indicating that both forms of treatment blunted the hypertrophic response post-MI. However, even under captopril treatment, there was still significant hypertrophy after MI. This is in agreement with previous findings on -blockers (13, 24) and ACE inhibitors (17, 46). Aside from a small increase in maximum LVDP by bisoprolol, both forms of treatment did not affect function and geometry in sham-operated hearts, as assessed by pressure-volume curves. In contrast, in infarcted groups, ACE inhibition with captopril partially prevented the rightward and downward shift of the systolic and diastolic pressure-volume curves, a well-characterized effect (12, 33, 46). Data are controversial on the functional effects of chronic -receptor blockade in the post-MI rat model. Several studies suggested that -blockers promote left ventricular dilation (13, 18). In contrast, our present work showed unchanged end-diastolic pressure to end-diastolic volume relations with bisoprolol treatment. In addition, bisoprolol improved LVDP (end-diastolic pressure = 10 mmHg) and maximum LVDP in infarcted hearts studied under the same loading conditions as sham-operated hearts. The present study clearly demonstrates that chronic bisoprolol treatment partially prevented the decrease of LVDP at various loading conditions but did not change end-diastolic pressure-volume relations. For the dosages used, the extent of the beneficial effect was similar to that exerted by captopril.

In the isolated isovolumic heart preparations used in this study, global coronary flow was not significantly different among the six groups of hearts. Therefore, the beneficial effects of chronic -blocker or ACE inhibitor treatment on function and energy metabolism may be due at most in part to improved global perfusion and microcirculation after MI. Because isolated hearts were not perfused with bisoprolol or captopril, these changes reflect chronic alterations of myocardial perfusion rather than acute coronary vascular effects of these drugs.

Effects of -receptor blockade and ACE inhibition on cardiac energy metabolism. In clinical studies of heart failure, both -receptor blockers (7, 44) and ACE inhibitors (34, 37) have been shown to chronically reduce mortality and improve left ventricular function. The exact mechanisms of these beneficial effects remain to be determined. Energy metabolism is compromised in heart failure (19), and these compounds may, at least in part, act by maintaining normal energy metabolism. Previous studies on this subject are limited: Sanbe et al. (36) showed in a post-MI rat model that improvement of cardiac index and high-energy phosphate levels by various ACE inhibitors after MI was associated with an increased mitochondrial oxidative function. A more recent study by Nascimben et al. (27) using Syrian myopathic hamsters showed maintenance of CK reaction velocity by treatment with enalapril. Waagstein et al. (45) showed that myocardial lactate production turns to lactate extraction in dilated cardiomyopathy patients treated chronically with metoprolol. Previous work by Laser et al. (24) showed that the changes in CK and lactate dehydrogenase isoenzyme composition could be prevented by bisoprolol treatment in the post-MI rat model. Also, in six patients with dilated cardiomyopathy, we showed that the myocardial PCr-to-ATP ratio, measured noninvasively with ³¹P NMR spectroscopy, increased during chronic drug therapy, including (in 4 of 6 cases) metoprolol (31). In the present work, we systematically analyze the effects of -receptor blockers and ACE inhibitors on the various components of cardiac energy metabolism in the post-MI rat model. Unequivocally, we demonstrate that the beneficial functional effects of both bisoprolol and captopril treatment are accompanied by beneficial effects on cardiac energy metabolism: increased PCr content and CK reaction velocity; increased MM and mitochondrial CK activities; and prevention of the fetal reprogramming with relative increase of the -containing CK isoenzymes. For the dosages used, both compounds were similarly effective, one exception being that bisoprolol also prevented the loss of total creatine whereas captopril did not. The reason for this discrepancy warrants further study. It is likely that -receptor blockers and ACE inhibitors exert their beneficial effects mainly by chronically reducing the energetic needs of the heart, via reduction of heart rate and pre- and afterload, respectively. In addition, it is possible that these agents interfere more directly with energy metabolism, one potential site of action being the sarcolemmal creatine transporter (15). This remains to be further evaluated.

Are the observed beneficial effects on cardiac energy metabolism causally related to the improvement of left ventricular function and geometry, or are they merely epiphenomena of the treatment with these compounds? Our data do not provide a final answer but allow us to speculate: In principle, energy metabolism could limit performance in heart failure by three distinct mechanisms: 1) a reduction of ATP content, 2) decrease of ATP transfer and thus ATP availability at the myofibrils, and 3) a reduction of the free energy change of ATP hydrolysis. Because steady-state ATP levels are unchanged in our model of heart failure, simply a preservation of ATP levels can be ruled out as a mechanism. In contrast, ATP transfer (CK flux) is reduced substantially in post-MI rat hearts, and this reduction is almost completely prevented by bisoprolol and, to a lesser extent, by captopril. Thus maintenance of ATP transfer, i.e., energy reserve via CK, may be involved in the protective effect of -receptor blockers and ACE inhibitors. Finally, G values remained unchanged for the baseline performance conditions studied here. However, Tian et al. (40) and Tian and Ingwall (41) have demonstrated that hearts with a compromised CK system show reduced contractile reserve. It is thus conceivable that the effects of reduced CK flux and the inability to maintain high G combine to limit the contractile reserve of the failing heart during inotropic stimulation, and -receptor blockers and ACE inhibitors may be able to maintain energy metabolism under these conditions. This remains to be studied. Therefore, our results do not prove a causal relation between the observed beneficial functional and energetic effects, but our findings are consistent with the view that preservation of energy metabolism explains, at least in part, the favorable effects of -receptor blockers and ACE inhibitors in chronic heart failure.

Study limitations. Enzyme and HPLC analyses were performed on intact residual right ventricular tissue, because the left ventricle was Formalin pretreated for histological determination of infarct size. However, Laser et al. (23) previously showed that changes in energy metabolism are very similar for the left and right ventricle. The present study involves 53 successful long-term experiments, yet only a single dose for each compound, one that showed a mild hemodynamic effect, was tested. Therefore, it is open whether the agents tested here exert dose-dependent effects on energy metabolism. Our study, however, shows that alterations in energy metabolism can be largely prevented when, post-MI, rats are chronically treated with -receptor blockers or ACE inhibitors.