Enalapril Treatment Increases Cardiac Performance and Energy Reserve Via the Creatine Kinase Reaction in Myocardium of Syrian Myopathic Hamsters With Advanced Heart Failure
Background Converting enzyme inhibitor treatment of congestive heart failure slows progression to failure and reduces mortality rate. It is not known whether these benefits are due solely to improved hemodynamics or also to improved myocyte energetics. This study examines the effect of enalapril treatment on both isovolumic contractile performance and its biochemical correlate, flux through the creatine kinase (CK) system, in an animal model of severely failing myocardium.
Methods and Results Seven-month-old Syrian cardiomyopathic (TO-2 strain) and normal golden Syrian (FIB strain) hamsters were each randomly assigned to one of three groups supplied daily with either no, low (25 mg/kg body wt), or high (100 mg/kg body wt) doses of enalapril for 12 to 14 weeks. At 10 months of age, all substrates and products and flux through the CK reaction were measured in isolated perfused hearts by 31P magnetization transfer and chemical assay. Compared with normal hamsters, the myopathic hamsters exhibited significantly lower body weights and higher biventricular heart weights, which were partially reversed by drug treatment. The Langendorff-perfused hearts showed decreased isovolumic contractile performance with identical load conditions. This was partially reversed by drug treatment. In the failing hearts, the following substrate and product concentrations and enzyme activities were decreased compared with nonfailing hearts but were unchanged by drug treatment: ATP (−28%), phosphocreatine (−48%), free creatine (−64%), ADP (−51%), and CK (−34%, primarily MM isoenzyme). Flux through the CK reaction for the untreated cardiomyopathic hamster hearts was decreased by 67%, and this decrease was almost completely reversed by enalapril treatment. The increased CK flux is due to an increase in the rate constant for the reaction, since substrate concentrations are unchanged, and is not predicted by the rate equation. In enalapril-treated failing hearts, phosphoryl transfer via the CK reaction increased with contractile performance. This was not observed in the nonfailing hearts, in which energy reserve is adequate to support changes in contractile performance.
Conclusions Decreased flux through CK reaction leads to decreased capacity for ATP synthesis and may contribute to decreased contractile performance in cardiomyopathic hamster hearts. Enalapril treatment results in increased phosphoryl transfer through the CK reaction in failing myocardium, and this increase is coupled to improved cardiac performance. Decreased CK flux in failing hearts is due to a combination of decreased Vmax and lower guanidino pool; this mechanism fails to explain changes in CK flux in enalapril-treated failing hearts.
Taken together, results from the clinical trials CONSENSUS I,1 SOLVD,2 3 VHeFT II,4 and Hy-C5 show that treating patients with congestive heart failure (CHF) with converting enzyme inhibitors (CEIs) ameliorates patients’ symptoms, slows the progression to heart failure, and reduces mortality. Hemodynamic improvements produced by supplying CHF patients with CEIs include decreased afterload, increased coronary perfusion, and reduced left ventricular (LV) dilatation.6 7 Treating patients with mild to moderate CHF (but with documented low ejection fraction) with the CEI enalapril reduced end-diastolic and end-systolic volumes, increased ejection fraction, and in the long term ameliorated symptoms and mortality rates.3 6 The CONSENSUS I study concluded that treating severely symptomatic (New York Heart Association class IV) patients with enalapril prolonged life; however, analysis of patients with the same functional class in the VHeFT II and in the symptomatic arm of the SOLVD studies did not show the same survival benefit.7 The Hy-C study of captopril-treated CHF patients with class IV symptoms attributed the observed decrease in mortality to a decrease in sudden death, not to changes in the progression of heart failure.5 Thus, our understanding of the responsiveness of severely failing myocardium to CEI treatment and its mechanisms of action is incomplete.
In this report, we use an animal model of heart failure to address these issues. Previous experimental studies using animal models to test the ability of drugs to slow the natural progression of CHF and to study their mechanisms of action have focused primarily on the early stages of CHF.8 9 In the present study, we take a different approach. Here, we focus on the late stages of heart failure and ask whether supplying the CEI enalapril improves contractile function in the severely failing mammalian myocardium. For these studies, we supplied either enalapril or no drug to severely failing (7-month-old) Syrian cardiomyopathic hamsters (TO-2 strain) for 3 months. We then compared isovolumic contractile performance in hearts isolated from these 10-month-old animals, perfused under the same conditions of preload and afterload and coronary flow. In addition to comparing drug-treated with untreated failing hearts, we also compared isovolumic contractile performance of failing hearts to performance of hearts isolated from age-matched nonfailing Syrian hamsters (FIB strain) treated or untreated with enalapril.
This study design comparing age-matched failing and nonfailing hearts from animals with and without drug treatment also allows us to examine a biochemical parameter that closely matches contractile performance, namely, the ability of the myocardium to synthesize ATP needed for excitation and contraction. In normal myocardium, ATP concentration is maintained constant despite variations in cardiac performance through balanced changes in the rates of ATP synthesis and utilization. ATP is synthesized by the rephosphorylation of ADP by oxidative phosphorylation in the mitochondria and by glycolysis in the cytosol. It is also resynthesized by the action of the primary energy reserve system in heart, the creatine kinase (CK) system. CK, present in high activity in heart, catalyzes the transfer of a phosphoryl group between creatine and ADP: phosphocreatine (PCr)+MgADP⇋MgATP+creatine. In vivo turnover of the phosphoryl group by CK, measured directly by 31P magnetization transfer, is an order of magnitude faster than net ATP synthesis estimated from oxygen consumption.10
Recent evidence from studies using myocardium obtained from CHF patients suggests that the capacity for ATP resynthesis via the CK system is compromised in failing myocardium.11 This conclusion is suggested by decreases both in the tissue activity of CK (Vmax) and in tissue content of the guanidino substrate for the reaction, ie, the sum of free creatine and PCr. 31P nuclear magnetic resonance (NMR) studies of the failing human myocardium also show that tissue content of the primary high-energy phosphate reserve compound in heart, PCr, is decreased compared with nonfailing myocardium.12 13 The observations that both CK activity and the PCr pool are decreased in failing myocardium are important because the velocity of the CK reaction, ie, the in vivo rate of phosphoryl turnover, is proportional to their product.14 For example, if CK activity decreases by 25% and PCr content decreases by 50% as described for class IV CHF patients,11 almost two thirds of the capacity for rapid ATP resynthesis (energy reserve) via this enzyme system is lost. Decreased energy reserve may also contribute to the decrease in myocardial ATP content observed in some models of CHF.15
In the present study, we determined, first, whether the capacity for ATP synthesis via the CK reaction is altered in the failing hamster heart and, second, whether supplying enalapril modifies this capacity. Substrates of the CK reaction were measured by 31P NMR spectroscopy and chemical assay. Tissue activity and isoenzyme distribution were measured by chemical assay, and in vivo reaction velocity was measured in the intact heart by 31P magnetization transfer. In this way, we determined not only the capacity of phosphoryl transfer of the failing hearts with and without enalapril treatment but also the actual turnover rate of the phosphoryl group in intact failing hearts.
Male Syrian cardiomyopathic hamsters of the TO-2 strain (failing) and normal golden Syrian hamsters of the FIB strain (nonfailing) were obtained from Bio Breeders, Inc. Contractile performance of hearts isolated from 7-month-old failing hamsters (methodology described below) was 34% lower than for aged-matched nonfailing hearts (R. Tian and J.S. Ingwall, unpublished data). Accordingly, we tested the effects of supplying enalapril to failing hamster hearts for the 3-month period beginning at 7 months of age.
Forty-eight 7-month-old animals of each strain were randomly assigned to one of three groups (n=16 animals per group) supplied with water containing either no, low (25 mg/kg body wt), or high (100 mg/kg body wt) concentrations of enalapril (Merck, Sharp & Dohme) for 12 to 14 weeks. During this time, these six groups of animals were maintained under identical conditions. Animals were housed four per cage and were allowed free access to laboratory chow (Purina) and water.
Drug doses of 25 and 100 mg/kg body weight were chosen on the basis of previous long-term studies using quinapril in cardiomyopathic hamsters9 and enalapril in spontaneously hypertensive rats.16 Although these doses are 200-fold higher than used clinically, the quinapril study showed that doses of 10 mg/kg were ineffective in cardiomyopathic hamsters. It appears, therefore, that hamsters are more resistant to the drug effect than humans. Average daily drug consumption was estimated from average daily water consumption (range, 13 to 16 mL per animal) determined weekly. Water intake was higher for all failing groups than for nonfailing groups (0.12 versus 0.08 mL · g body wt−1 · d−1), with no differences due to the addition of drug to the water. On the basis of these values, we conclude that the animals were given the planned doses. All animals were cared for according to the guidelines of the American Physiological Society.
Isolated Perfused Hearts
The hamsters were heparinized (1000 U IP) 15 minutes before being anesthesized with pentobarbital sodium (50 mg/kg body wt IP). The heart was quickly removed through a median sternotomy and rinsed in ice-cold buffer while pericardium and lung tissues were gently trimmed away. The aorta was cut just under the arch before being securely tied to a polyethylene cannula attached to an isolated heart perfusion apparatus. Retrograde aortic perfusion was carried out in the Langendorff mode at constant temperature of 37°C and constant pressure of 100 mm Hg with modified Krebs-Henseleit buffer.10 Choice of this perfusion pressure was based on results by Haleen et al9 showing that this pressure maximizes ventricular function and coronary flow in aged hamsters. This was confirmed in preliminary experiments using a subset of the cohort of hamsters entered into this study. The perfusate buffer was saturated with 95% O2/5% CO2 and contained (in mmol/L) NaCl 118, KCl 4.7, CaCl2 1.75, MgSO4 1.2, Na4EDTA 0.5, NaHCO3 25, and glucose 11.0. An incision in the pulmonary artery and a short polyethylene vent pierced through the apex of the left ventricle allowed drainage from the coronary sinus and thebesian veins. Coronary flow was measured periodically by collecting timed coronary effluent samples in a calibrated cylinder.
An oval-shaped polyethylene balloon was inserted into the LV through the mitral valve after the left auricle was cut. This was used in place of a latex balloon typically used in Langendorff-perfused rodent heart preparations to record systolic and diastolic pressures because its dimensions more closely matched the shape and volume of the LV of the hamster heart. The balloon was filled with water so that LV end-diastolic pressure was <8 mm Hg. The water-filled balloon was connected to a Statham P23Db pressure transducer (Gould Instruments) through a small-bore polyethylene tube to allow continuous monitoring of heart rate and LV developed pressure (difference between systolic and diastolic pressures) on a Gould 2200S recorder. Isovolumic contractile performance was estimated as the product of heart rate and developed pressure, the rate-pressure product (RPP, mm Hg/min). Nonfailing and failing hearts perfused in this way maintained a steady state, in which RPP did not vary by >10%, for 3 and 1.5 hours, respectively.
31P NMR Spectroscopy
The perfused heart was placed in a custom-built 16-mm horizontal probe and inserted into the bore of an 8.4-T magnet (Oxford Instruments). The magnet was connected to a NT360 spectrometer interfaced with a 1280 computer (Nicolet Instruments) operating in pulsed Fourier transform mode. An 18-channel Oxford Instruments Supply was used to homogenize the magnetic field by minimizing the width of the 1H signal. 31P NMR spectra were obtained at 145.75 MHz. Magnetization transfer was performed by applying a low-power (B1 field of ≈30 Hz) radiofrequency pulse centered at the [γ-P]ATP resonance for progressively longer time periods from 0 to 4.8 seconds. The low-power pulse was calibrated by irradiating the [γ-P]ATP peak to ensure complete saturation, followed by irradiation downfield at a frequency equidistant from the PCr resonance to ensure that there was negligible off-resonance saturation of the PCr resonance. Magnetization transfer spectra were obtained by signal averaging 64 scans of high-power broadband 53° (20-μs) read pulses following the low-power narrowband saturation pulse interleaved in groups of eight and separated by a constant delay of 7 seconds including the saturation pulse time. This increased the probability that effects due to any changes in contractile function or rhythm during the signal averaging were equally distributed among the spectra. A complete saturation transfer experiment was acquired in 55 minutes. Control spectra without presaturation, obtained over 4 minutes by signal averaging 104 scans of 53° read pulses separated by an interscan delay of 2.14 seconds, were acquired immediately before and after the magnetization transfer experiment to assess the stability of the preparation over the course of the experiment. A spectral width of 6000 Hz was used. Individual free induction decays were zero filled from 1024 to 2048 data points and weighted with a 20-Hz line, broadening decaying exponential before Fourier transformation.
Hearts were freeze-clamped immediately after acquisition of the NMR data for subsequent biochemical analysis. Five to 15 mg of ventricular tissue was homogenized for 10 seconds at 4°C in potassium phosphate buffer containing 1 mmol/L EDTA and 1 mmol/L β-mercaptoethanol, pH 7.4 (final concentration of 5 mg/mL). Aliquots were removed for measurement of protein according to the method of Lowry et al17 with bovine serum albumin as the standard and measurement of total creatine content according to the fluorometric assay of Kammermeier.18 Then Triton X-100 was added to the homogenate at a final concentration of 0.1% for analysis of CK activity.
Total CK activity was measured with the coupled enzyme scheme of Rosalki19 at 30°C by use of the Calbiochem-Behring CK-NAC SVR kit (Behring Inc). The ratio of activities at 37°C versus 30°C was empirically determined to be 1.8. Distribution of the CK isoenzymes was determined by cellulose-acetate strip electrophoresis coupled with scanning fluorometry.20 The same amount of enzyme activity (7 to 10 mIU) was applied to each strip. Since the specific activities of the isoenzymes are similar,21 multiplying total CK activity by percentage isoenzyme distribution provides an estimate of the tissue activity of each isoenzyme. Results are expressed both as percentage of total CK activity and as specific activities of each isoenzyme. Enzyme activities are expressed as international units (IU=μmol/min) per milligram of cardiac protein.
In a separate portion of the freeze-clamped heart, ATP content was determined by high-performance liquid chromatography.22 Results are expressed as nanomoles per milligram Lowry protein.
Integrated signal intensities in 31P NMR spectra corresponding to amounts of ATP and PCr in the whole heart were measured with the nmr1 curve-fitting routine (New Methods Research Inc). The integrated signal intensities of the PCr resonance were corrected for differential incomplete relaxation based on measured relaxation times for ATP and PCr.23 Absolute values of ATP were determined as described.22 Briefly, absolute ATP contents were determined from a standard curve obtained from a separate group of hamster hearts by relating [β-P]ATP peak areas and myocardial ATP content measured by high-performance liquid chromatography normalized in the same way. The conversion factor obtained was 0.0277 nmol ATP per area unit. Intracellular pH was measured by comparing the chemical shift between the inorganic phosphate and PCr resonances with values obtained from a standard curve.
To calculate intracellular concentrations of the substrates of the CK reaction using either 31P NMR spectra or chemical analysis, it is necessary to estimate the fraction of the heart that is intracellular water. For nonfailing hearts, the literature value of 0.5 mL/g wet wt obtained for hearts of small animals24 was used. We make the assumption that this value also applies to the myocytes of failing hearts and to drug-treated hearts. Because 10-month-old failing hearts contain regions of focal fibrosis and may be edematous, myocyte mass (and hence intracellular volume) per unit wet weight differs in failing and nonfailing myocardium. Estimates of the concentrations of metabolites (and enzymes) were adjusted to take these differences into account. The parameter that best accounts for these differences is Lowry protein. Lowry protein minimizes the contribution from proteins found in the extracellular space because these proteins contain relatively low amounts of aromatic amino acids, the target for this assay.17 It also eliminates any confounding effects of intracellular or extracellular edema. Thus, Lowry protein closely approximates the myocyte protein content. Values obtained here, 0.117±0.008 and 0.141±0.005 mg protein/mg wet wt for failing and nonfailing hearts, respectively, are consistent with the presence of regions of fibrosis in failing hearts.25 26 There were no differences in Lowry protein content due to administration of enalapril in either failing or nonfailing hamster hearts. In this report, we calculate metabolite (and enzyme) concentrations for failing and nonfailing hearts both as moles (or activity units) per milligram Lowry protein and, by scaling wet weight by the nonfailing Lowry protein content and using the value for intracellular water of 0.5 mL/g wet wt, as cytosolic concentrations (mmoles per liter). It should be emphasized that, for the failing myocytes, both the intracellular volume assumption and the scaling procedure minimize the differences observed in tissue metabolite and enzyme contents and provide an estimate of their maximum intracellular concentrations.
Cytosolic free [ADP] was estimated from the CK equilibrium expression and the ratio of ATP to PCr contents and pH measured using 31P NMR spectroscopy, total creatine measured chemically in heart homogenates, and an equilibrium constant of (1.66×109)×[H+]27 :
Measured CK Reaction Velocity
For the reaction
the velocity of the forward CK reaction, Vfor=kfor′[ADP][PCr]=kfor[PCr], where kfor′ is the second-order rate constant and kfor is the pseudo–first-order rate constant equal to the product of the second-order rate constant and [ADP]. The magnetization transfer measurements of the forward CK reaction (PCr→[γ-P]ATP) were analyzed according to the two-site chemical exchange model of Forsen and Hoffman,28 providing a measurement of the pseudo–first-order rate constant and the intrinsic longitudinal relaxation time constant for PCr, T1.10 23 As the time of saturation at [γ-P]ATP, t, is increased from 0 to 4.8 seconds, the integrated signal intensity of PCr magnetization, Mt, decays from M0 to M∞ (defined as magnetization at time zero and infinite time, respectively) with the time constant τ as follows:
where 1/τ=1/T1+kfor. Nonlinear regression of PCr magnetization at different saturation times of [γ-P]ATP was used to determine M0, M∞, and τ. kfor and T1 were then calculated by solving two simultaneous equations: kfor=(M0−M∞)/(M0τ) and T1=(M0/M∞)τ. Multiplying the pseudo–first-order rate constant by [PCr] yields the measured reaction velocity, Vfor=kfor[PCr].
Calculated CK Reaction Velocity
The velocity for CK in solution can be calculated from the rate equation derived from the work of Morrison and Cleland29 :
Measured input values are Vmax (total CK activity under saturating conditions) and the cytosolic concentrations of PCr, creatine, ATP, and ADP determined as described above. The estimates of kinetic constants for heart determined at pH 7.0 used in this analysis were taken from References 29 and 30.
To test for statistical significance among the different groups, two-factor (hamster strain [failing and nonfailing] and enalapril treatment [no-dose, low-dose, and high-dose]) ANOVA was used. In cases in which the two-factor ANOVA gave either a significant (P<.05) drug effect or factor interaction, the data were further analyzed as follows: (1) One-factor ANOVA was used to test for statistical significance of enalapril treatment on each strain individually. (2) A two-tailed unpaired Student’s t test comparing untreated failing with nonfailing hamsters was used to test for statistical significance between the two strains. In these cases, the nonfailing heart data were pooled because one-factor ANOVA showed no significant drug treatment effects for any parameter; this was not the case for the failing hamster hearts. It should be noted that pooling the data for nonfailing hearts did not produce any statistically significant strain effects that were not already identified by t tests of the unpooled data for untreated failing versus untreated nonfailing hearts. Linear regression was used to correlate CK reaction velocity to isovolumic contractile performance. All calculations were aided by statview 512+ (BrainPower Inc) and nmr1. All data are presented as mean±SEM.
Characteristics of Failing and Nonfailing Animals
All nonfailing hamsters survived to 10 months of age, independent of enalapril supply. In contrast, only 44% (7 of 16 animals) of untreated failing hamsters survived to 10 months of age. Drug treatment improved this percentage: 8 of 16 and 11 of 16 animals in low- and high-dose failing groups, respectively, survived to 10 months of age, for a total of 19 of 32, or 59%. Enalapril-treated failing animals displayed greater physical activity than untreated failing animals.
For each of the six groups studied, the body weights, biventricular weights, and ratio of biventricular weight to body weight are shown in Table 1⇓. The untreated failing hamsters had higher heart weights and lower body weights than nonfailing hamsters. Consequently, the ratio of biventricular heart weight to body weight was 59% higher in failing hearts than in nonfailing hearts. Drug treatment reduced both the biventricular weight and body weight of failing animals without changing their ratio. It also reduced the number of thrombi found in atria of failing hearts (4 of 7 hearts of untreated versus 2 of 19 hearts of drug-treated failing animals contained thrombi).
Isovolumic Contractile Performance of Failing and Nonfailing Isolated Perfused Hearts
Hearts of treated and untreated failing hamsters showed all the characteristics of dilated cardiomyopathy. The ventricles were irregular in shape, enlarged, and grossly dilated and contained many fibrotic foci. The atria were also very large. In contrast, the gross appearance of all nonfailing hearts was normal. Perfusing isolated hearts at the same constant pressure at the same fixed end-diastolic pressure (approximating constant preload and afterload) resulted in comparable values for coronary flow (milliliter per gram wet weight), allowing comparison of contractile performance of failing and nonfailing myocardium. Isovolumic contractile performance and coronary flow data obtained for the six groups of isolated perfused hearts are summarized in Table 2⇓.
Untreated failing hearts had similar heart rates but lower developed pressures than untreated nonfailing hearts. Consequently, the rate-pressure product for failing hearts was 44% of the value for nonfailing hearts. To take into account differences in heart size and the presence of fibrosis and increased extracellular volume in failing hearts, we also calculated rate-pressure product per gram heart weight and per gram Lowry protein. The rate-pressure product per gram biventricular heart weight for failing hearts was only 34% of the value for nonfailing hearts [13 000 versus 38 300 mm Hg · min−1 · (g wet wt)−1]. The rate-pressure product normalized to Lowry protein (which more closely approximates myocyte mass) for failing hearts was 40% of the value for nonfailing heart. Thus, whether expressed per heart weight, per biventricular weight, or per myocyte protein content, isovolumic contractile function of failing hearts was less than half that for aged-matched nonfailing hearts perfused under comparable conditions.
Supplying enalapril to nonfailing animals for 3 months had no effect on contractile performance measured in isolated hearts. In contrast, compared with untreated failing hearts, developed pressure in failing hearts isolated from animals supplied with enalapril at the high dose was 39% higher, rate-pressure product per heart 52% higher, rate-pressure product per gram biventricular weight 85% higher, and rate-pressure product per gram protein 84% higher. Parameters characterizing contractile performance for these enalapril-treated failing hearts were close to those obtained for enalapril-treated nonfailing hearts.
To determine whether the changes in contractile performance in failing and nonfailing hearts with and without enalapril are matched by differences in turnover rate of the phosphoryl group, we measured the concentrations and turnover rates of ATP and the energy reserve compound PCr.
The CK System in Failing and Nonfailing Hearts With and Without Enalapril Treatment
Values for total CK activity, CK isoenzyme distribution, and the cytosolic concentrations of the four substrates of the CK reaction are summarized for failing and nonfailing hearts in Table 3⇓. Since enalapril treatment did not significantly change any of these values, we present only the mean values for all failing versus all nonfailing hearts.
CK activity in failing myocardium was 36% lower than in age-matched nonfailing hearts. The isoenzyme distribution expressed as a percentage of the total activity differed in failing and nonfailing myocardium: compared with nonfailing myocardium, the failing myocardium has similar MB, less MM, and more mitochondrial CK activity. BB CK activity was <1%. When tissue activity of each isoenzyme is estimated by multiplying the percentage distribution of each isoenzyme by the total CK activity, a different profile is obtained. The tissue activity of MB-CK is unchanged, while both the MM and mitochondrial CK isoenzymes are lower (by 42% and 18%, respectively) in failing hearts.
[ATP] measured by chemical assay using homogenates of portions of freeze-clamped failing hearts was 28% lower than for nonfailing hearts. To test whether the decrease in ATP content in failing myocardium determined in this way was also observed for intact hearts, we compared the [β-P]ATP resonance areas in 31P NMR spectra obtained from isolated beating failing and nonfailing hearts; representative examples are shown in Fig 1⇓. Spectra such as these were used to determine the relative resonance areas of the phosphorus-containing metabolites in the six groups of hearts. To determine the relative concentrations, the areas were normalized for heart weight (Table 1⇑) and protein content (see “Methods”) as follows. The [β-P]ATP areas, heart weights, and protein contents of nonfailing hearts were unchanged by drug treatment, and thus [ATP] was unchanged. In contrast, the mean values determined for [β-P]ATP resonance areas of treated and untreated failing hearts were lower than for nonfailing hearts (71.5, 54.9 [−23%], 45.7 [−36%], and 46.0 [−36%]×103 area units for nonfailing, untreated failing, low-dose failing, and high-dose failing hearts, respectively). After taking into account individual differences in heart weight (see Table 1⇑ for the means), we obtain 0.157, 0.092 (−41%), 0.092 (−41%), and 0.096 (−39%)×103 area units/mg tissue wet wt, respectively. Finally, including differences in protein content (0.141 mg/mg wet wt for nonfailing and 0.117 mg/mg wet wt for failing hearts), we obtain values of 1.11, 0.79 (−29%), 0.79 (−29%), and 0.82 (−26%)×103 area units/mg Lowry protein, respectively. These 31P NMR results show that failing hearts contain, on average, 28% less ATP than nonfailing hearts and that the ATP content is unchanged with drug treatment. Absolute [ATP] was determined by use of an external standard (see “Methods”) and is shown in Table 3⇑.
Thus, both 31P NMR spectroscopy of the entire beating heart and chemical assay of heart homogenates of ventricular biopsy specimens show that [ATP] is 28% lower in failing myocardium. Calculated values for cytosolic [ADP] in failing myocardium were ≈50% lower than for nonfailing myocardium. Intracellular pH was the same in all groups and averaged 7.14.
Changes in the guanidino pool were also assessed in two ways: from the PCr content measured by 31P NMR spectroscopy of intact hearts and from the total creatine pool measured in heart homogenates. The examples illustrated in Fig 1⇑ show that the ratio of PCr to ATP is lower in failing than in nonfailing hearts. Combining results from 31P NMR spectroscopy and chemical assay shows that total creatine, PCr, and hence free creatine pools are >50% lower in failing than in nonfailing hearts.
Estimating CK Reaction Velocity Using the Rate Equation
With the kinetic constants given in References 29 and 30 and the CK reaction parameters given in Table 3⇑, the rate equation predicts that the CK reaction velocity in failing untreated hearts is 2.7 times lower than in nonfailing hearts. Thus, both the measured (Table 4⇓) and calculated reaction velocities for CK show that phosphoryl transfer between the energy reserve molecule PCr and ATP in failing myocardium is decreased by about the same factor compared with nonfailing myocardium. Neither Vmax nor the concentrations of substrates changed with enalapril treatment. Therefore, the rate equation fails to predict the increase in CK reaction velocity measured in failing hearts isolated from enalapril-treated animals.
Relations Between CK Reaction Velocity and Cardiac Performance
Fig 2⇓ plots measured CK reaction velocity (Table 4⇑) against isovolumic contractile performance estimated as the rate-pressure product for all treated and untreated failing (Fig 2A⇓) and nonfailing (Fig 2B⇓) hearts. The relation (r=.61) for treated and untreated failing hearts has a positive slope (P=.02) and an intercept indistinguishable from zero (P=.84). In contrast, there is no relation between measured reaction velocity and contractile performance in treated and untreated nonfailing hearts. If the rate-pressure product is normalized for myocyte mass (data not shown), one obtains similar results with a better correlation for the failing hearts (r=.74, P=.002). Note that the data points for no treatment and for high-dose enalapril treatment do not overlap for the failing hearts, whereas there is substantial overlap for the nonfailing hearts, consistent with the poor correlation for nonfailing hearts. This analysis shows that as contractile performance increases in enalapril-treated failing hearts, phosphoryl transfer via the CK reaction also increases.
Isovolumic Contractile Performance and Phosphoryl Turnover in Isolated Failing Hearts
Consistent with depressed cardiac performance in CHF in humans and in other animal models of CHF, isovolumic contractile performance measured as the rate-pressure product in hearts isolated from 10-month-old failing hamsters was much lower than for hearts isolated from age-matched nonfailing hamsters perfused under the same conditions of load and coronary flow. In concert with the decrease in contractile performance, we show here that (1) the tissue contents of ATP and PCr, the primary energy reserve compound in heart, (2) the capacity (Vmax) for ATP synthesis via the CK reaction measured as tissue enzyme activity, and (3) the rate of phosphoryl transfer (Vfor) measured as CK reaction velocity in vivo are all lower in the severely failing mammalian myocardium.
Results presented in this study, from both the noninvasive tool of 31P NMR spectroscopy and chemical assay of heart homogenates, show that failing myocardium of the 10-month-old hamster contains less ATP, ADP, PCr, and free creatine than nonfailing myocardium, even when normalized to account for any differences in extracellular protein and edema. The decrease in the content of the energy reserve compound PCr is greater than that for ATP.
Our results also show that, compared with age-matched nonfailing hearts, MM CK activity is substantially lower (42%) in failing hamster hearts. The decrease in MM isoenzyme activity is much greater than the changes in the other CK isoenzymes. There was no change in MB CK calculated as tissue activity and an 18% decrease in mitochondrial CK. A decrease in mitochondrial CK activity has also been observed in the spontaneously hypertensive rat heart in the transition from compensated hypertrophy to failure.31 It has also been described for the CHF 14.6 strain of cardiomyopathic hamster by Khuchua et al.32 They also reported a large increase in BB CK without an increase in the hybrid MB isoenzyme, which we did not observe. Since the M and B polypeptide chains associate randomly, we would expect that the percent of MB would always be greater than the percent of BB (not less than BB, as reported by Khuchua et al) if both gene products are present in the same cell type.
A major finding described in this report is that there are relatively large decreases both in total CK activities and in the guanidino pool in the dilated cardiomyopathic hamster heart in severe failure. Several recent meeting reports11 14 have described directionally similar changes in the CK system in failing myocardium of humans and in other animal models of CHF, suggesting that these defects in the CK system occur independently of species or pathogenesis and that this pattern is characteristic of this pathology.
Total CK activity (Vmax) and the size of the guanidino pool combine to set CK reaction velocity. On the basis of lower CK activity (primarily MM CK) and lower concentrations of guanidino (and purine) substrates in the failing heart, the CK rate equation predicts that the velocity through the CK reaction is 2.7-fold lower than in nonfailing myocardium. The measured in vivo CK reaction velocities also show that phosphoryl transfer between the energy reserve molecule PCr and ATP is 2.6-fold lower in failing than nonfailing hamster myocardium. Thus, phosphoryl transfer via CK decreases from being 10 times greater than the rate of ATP synthesis via oxidative phosphorylation (for normal heart) to only ≈3 times greater (assuming no change in the rate of oxidative phosphorylation) in the failing heart. The significance of this observation may be illustrated by considering the hypothetical case in which ATP synthesis occurs only via the CK reaction. In that case, it would take twice as long to replenish the ATP pool (even though the ATP pool is 30% smaller) from phosphoryl transfer via CK in failing than nonfailing myocardium. Moreover, because the PCr concentration is lower than that for ATP, the ATP pool would not even be fully repleted.
The CK system acts as an energy reserve system, especially during high-workload conditions. The link between energy reserve and contractile reserve is clearly shown during the ATP supply-demand mismatch that occurs in hypoxia and ischemia, when phosphoryl transfer from PCr to ADP slows the rate of net ATP depletion. The link has also been shown in three experiments designed to selectively perturb the CK reaction in intact striated muscle. First, in rats in which the myocardial PCr pool was chronically replaced with a poorly hydrolyzable guanidino analogue (β-guanidinopropionic acid), PCr content, CK reaction velocity, and heart function all decreased under high-workload conditions.33 Second, acute selective chemical inhibition of CK activity in intact rodent hearts with iodoacetamide resulted in decreased contractile reserve when the heart was inotropically stressed with either high extracellular Ca2+ or norepinephrine.34 Third, depleting MM CK by transgenic technology reduced the ability of skeletal muscle to sustain acute or burst work.35 Thus, by analogy, decreased energy reserve via the CK system described here for the severely failing hamster heart is likely to reduce the contractile reserve of the heart. It may also contribute to decreased baseline contractile performance.
Effects of Enalapril on Isovolumic Contractile Performance and Phosphoryl Turnover in Severely Failing Hearts
Because our study design required that the animals be killed at a specific time, we cannot quantitatively assess differences in survival rates due to CEI treatment. However, several characteristics of the treated failing animals provide evidence that supplying enalapril to severely failing animals for 3 months improved survival and hemodynamic performance: (1) increased physical activity, (2) a decrease (albeit small) in body and ventricular weights, and (3) fewer atrial thrombi. The apparent increase in survival rate and improved physiology in enalapril-treated failing hamsters are consistent with previous experiments using the CEI quinapril.9 Increased isovolumic contractile performance and phosphoryl transfer via the CK reaction measured in hearts isolated from enalapril-treated failing animals demonstrates that supplying enalapril to severely failing animals for 3 months was efficacious.
The major physiological consequences of administration of CEIs include decreased afterload and increased coronary flow. If the isolated isovolumic heart preparation is used, the confounding effects of variable preload and afterload that occur in vivo are either not present or minimized during the measurement of contractile performance. In the isolated isovolumic heart preparations used in this study, preload was held constant and coronary flow was not significantly changed in hearts isolated from enalapril-treated animals. Nevertheless, hearts isolated from enalapril-treated failing animals showed significantly greater isovolumic contractile performance compared with untreated failing hearts. Since isolated hearts were not perfused with enalapril, these changes reflect altered intrinsic capacity of the myocardium resulting from chronic enalapril supply.
We found no drug-related changes in gene expression (assessed here as enzyme activity) of any of the CK isoenzymes or in several mitochondrial and glycolytic enzymes (data not shown). Instead, we found that supplying enalapril to severely failing myocardium allows existing myocytes to perform more isovolumic work without changing the biochemical composition of the major pathways for ATP supply. Our results showing improved cardiac performance in hearts isolated from enalapril-treated animals can best be explained by the ability of the drug to chronically reduce load, thereby creating a better match between ATP supply and demand. This conclusion is supported by the positive relation described here for phosphoryl turnover via the CK system and isovolumic contractile performance.
Regulation of CK Reaction Velocity in Failing Hearts With and Without Enalapril
Because we measured CK enzyme activity under saturating conditions (Vmax) and concentrations of all of its substrates, we can draw some conclusions regarding the mechanism underlying the observed changes in CK reaction velocity. One way this can be done is by comparing measured (by 31P magnetization transfer) and predicted (from the rate equation) CK reaction velocities. Agreement between the relative changes in the measured and predicted velocities means that the reaction is regulated by the number of active enzyme molecules and the concentrations of substrates. The decrease in CK reaction velocity in failing versus nonfailing hearts observed here can be explained by the combined decreases in Vmax and the guanidino and purine pool sizes. Similarly, the lack of an effect of enalapril on CK reaction velocity in nonfailing hearts is consistent with no observed changes in Vmax or in substrate concentrations. Since drug treatment changed neither CK activity (Vmax) nor its substrate concentrations in enalapril-treated failing hearts, the rate equation predicts that the measured CK reaction velocity should also be unchanged. This was not observed; instead, the rate equation fails to predict the dose-dependent increase in phosphoryl turnover in enalapril-treated failing hearts.
Failure of the rate equation to predict measured reaction velocity suggests that the reaction mechanism developed by Morrison and Cleland29 for the enzyme in dilute solution does not apply to enalapril-treated failing hearts. This point is further illustrated by calculating the apparent second-order rate constant (kfor′=kfor/[ADP]) for failing and nonfailing hearts with and without enalapril treatment. As shown in Table 4⇑, the second-order rate constant is, as it should be, a constant for all three nonfailing groups, varying only from 12 to 16 (mmol/L)−1 · s−1, values well within experimental error. A similar value is obtained for the untreated failing hearts, 18 (mmol/L)−1 · s−1. In contrast, the second-order rate constant is at least two times higher for both groups of enalapril-treated failing hearts. Both the increase in the second-order rate constant and the failure of the rate equation to predict the measured change in reaction velocity in drug-treated failing hearts show that the conventional reaction scheme for the CK reaction developed for enzyme in dilute solution is inadequate to explain in vivo reaction kinetics in enalapril-treated failing hamster hearts.
Previous studies using myocardium of small animal hearts have demonstrated that flux through the CK reaction is usually under substrate control and is proportional to the concentration of enzyme (Vmax).36 Examples include the beating rat heart both in situ37 and as the isolated heart under different workloads.10 There are, however, three conditions in which the rate equation fails to predict measured changes in CK reaction velocity in heart: during KCl arrest in all small animal hearts studied to date,10 30 during mild38 and severe (A.C. Cave, C.S. Apstein, J.S. Ingwall, unpublished data) ischemia in the rat heart, and during hypoxia in the open-chest rat heart.37 In each of these cases, reaction velocity was lower than predicted. The underlying mechanism(s) for these discrepancies is not established, but several possibilities exist. One possibility is that the bulk cytosolic substrate levels do not regulate the reaction flux because the velocity of the CK reaction is determined by metabolite levels in microcompartments formed by localization of CK isoenzymes. Another is that the reaction scheme appropriate for MM CK in dilute solution changes in vivo.30 Another possible explanation is suggested by the observation that the ratio of Vfor to Vmax is <<1. This shows that a significant fraction of CK in vivo is not active; likely candidates for in vivo inhibitors are anions such as bicarbonate (L. Bolinger, J.S. Leigh, J.S. Ingwall, unpublished data) and a phosphorylation/dephosphorylation cycle.
In each of the examples of downregulation of CK activity in vivo described above, cardiac performance was also decreased. As in these examples, we observe a positive relation between measured CK reaction velocity and contractile performance in enalapril-supplied failing hearts. The major observation made in this report is that CK reaction velocity changes with pharmacological treatment as well as physiological maneuvers. Enalapril treatment resulted in increases in both cardiac performance and phosphoryl turnover in failing hearts to a value close to those observed for nonfailing hearts. It had no effect on nonfailing hearts, in which energy reserve is adequate to support changes in contractile performance. The results are consistent with an increase in apparent Vmax, a recruitment of inactive CK enzymes, in enalapril-supplied failing hearts. Whatever the mechanism for the increase in apparent Vmax, however, the results described here support the hypothesis that a decrease in energy reserve via the CK system contributes to decreased contractile performance in heart failure and that increasing CK reaction velocity in failing myocardium is coupled to improved cardiac performance.
First, we used a genetic model of cardiomyopathy that has many unknown physiological and biochemical alterations. This limitation applies to other heart failure models as well. Of central importance to this study, however, is that the biochemical changes in the CK system observed here are common to all heart failure models studied to date and to failing myocardium in humans.
Second, although we can explain why CK reaction velocity decreased in failing hearts, we are unable to establish the mechanism for the apparent increase in Vmax for CK in enalapril-treated hearts. Because there was no change in apparent Vmax in enalapril-treated nonfailing hearts, we can rule out direct action of the drug on the enzyme, but identifying the mechanism in vivo must await other strategies.
Third, chamber dilatation and afterload reduction are compensatory effects in the maintenance of stroke volume. However, elevation of wall stress and diastolic pressure is detrimental for myocardial viability and with time leads to a decrease in cardiac function proportional to the amount of viable myocardium lost. In this experimental setting, we did not measure ventricular geometry and thus cannot rule out improved performance of enalapril-treated hearts secondary to decreased dilatation. However, because the protein content remained unchanged and enalapril-treated hearts remained significantly dilated, it is unlikely that the changes in contractile performance and CK flux observed in enalapril-treated hearts are due primarily to decreased dilatation.
Alone and in combination with digoxin, nitrates, and diuretics, CEIs are widely used in the treatment of CHF. The efficacy of the CEI enalapril in slowing the progression to failure has been convincingly shown in clinical trials studying functional class I through class III CHF patients, but its beneficial effect is less clear for class IV patients.7 This is the case even though enalapril as well as other CEIs is widely used in the treatment of severe CHF. The results presented here showing improved myocardial energetics and performance at the terminal stages of heart failure in an animal model suggest that enalapril treatment would also be beneficial in class IV CHF patients. Our results also support the hypothesis that the physiological benefits at this stage of the disease are related to improved energetics.
This research was supported in part by National Institutes of Health grants HL-43170 and HL-52320. Dr Nascimben was supported by a Merck Sharpe & Dohme Italia Spa fellowship award. Dr Friedrich was supported by Harvard-MIT Division of Health Sciences and Technology Johnson & Johnson and Research Fund Fellowships. The authors thank Jonathan Rose, BS, and Ilana Reis, MSc, for technical assistance; Dr Judith K. Gwathmey for helpful discussions; and Dr Case Van Dongen of Bio Breeders, Inc, for his support and assistance during this project.
- Received July 11, 1994.
- Revision received October 10, 1994.
- Accepted October 30, 1994.
- Copyright © 1995 by American Heart Association
Konstam MA, Kronenberg MW, Rousseau MF, Udelson JE, Melin J, Stewart D, Dolan N, Edens TR, Ahn S, Kinan D, Howe DM, Kilcoyne L, Metherall J, Benedict C, Yusuf S, Pouleur H. Effects of the angiotensin converting enzyme inhibitor enalapril on the long-term progression of left ventricular dilatation in patients with asymptomatic systolic dysfunction. Circulation. 1993;88(pt 1): 2277-2283.
Cohn JN, Johnson G, Ziesche S, Cobb F, Francis G, Tristani F, Smith R, Dunkman WB, Loeb H, Wong M, Bhat G, Goldman S, Fletcher RD, Doherty J, Hughes CV, Carson P, Cintron G, Shabetai R, Haakenson C. A comparison of enalapril with hydralazine-isosorbide dinitrate in the treatment of chronic congestive heart failure. N Engl J Med. 1991;325:303-310.
Fonarow GC, Chelimsky-Fallick C, Stevenson LW, Luu M, Hamilton MA, Moriguchi JD, Tillisch JH, Walden JA, Albanese E. Effect of direct vasodilation with hydralazine versus angiotensin-converting enzyme inhibition with captopril on mortality in advanced heart failure: the Hy-C trial. J Am Coll Cardiol. 1992;19:842-850.
Fonarow GC, Stevenson LW. Angiotensin-converting enzyme inhibition in heart failure. In: Singh BN, ed. Cardiovascular Pharmacology and Therapeutics. New York, NY: Churchill Livingstone; 1994:797-814.
Camacho SA, Wikman-Coffelt J, Wu ST, Watters TA, Botvinick EH, Sievers R, James TL, Jasmin G, Parmley WW. Improvement in myocardial performance without a decrease in high-energy phosphate metabolites after isoproterenol in Syrian cardiomyopathic hamsters. Circulation. 1988;77:712-719.
Haleen SJ, Weishaar RE, Overhiser RW, Bousley RF, Keiser JA, Rapundalo SR, Taylor DG. Effects of quinapril, a new angiotensin converting enzyme inhibitor, on left ventricular failure and survival in the cardiomyopathy hamster. Circ Res. 1991;68:1302-1312.
Bittl JA, Ingwall JS. Reaction rates of CK and ATP synthesis in the isolated rat heart: a 31P NMR magnetization transfer study. J Biol Chem. 1985;260:3512-3517.
Nascimben L, Pauletto P, Pessina AC, Reis I, Ingwall JS. Decreased energy reserve may cause pump failure in human dilated cardiomyopathy. Circulation. 1991;84(suppl II):II-563. Abstract.
Ingwall JS. Is cardiac failure a consequence of decreased energy reserve? Circulation. 1993;87(suppl VII):VII-58-VII-62.
Scheuer J. Metabolic factors in myocardial failure. Circulation. 1993;87(suppl VII):VII-54-VII-57.
Childs TJ, Adams MA, Mak AS. Regression of cardiac hypertrophy in spontaneously hypertensive rats by enalapril and the expression of contractile proteins. Hypertension. 1990;16:662-668.
Lowry OH, Rosebrough MJ, Farr AL, Randall RJ. Protein measurement with the Folin reagent. J Biol Chem. 1951;193:265-275.
Bak MI, Ingwall JS. NMR-invisible ATP in heart: fact or fiction? Am J Physiol. 1992;262(Endocrinol Metab 25):E943-E947.
Polimeni PI. Measurement of myocardial electrolyte distributions. Tech Life Sci Cardiovasc Physiol. 1984;317:1-34.
Veech RL, Lawson JWR, Cornell NW, Krebs HA. Cytosolic phosphorylation potential. J Biol Chem. 1979;254:6538-6547.
Morrison JF, Cleland WW. Isotope exchange studies of the mechanism of the reaction catalyzed by adenosine triphosphate: creatine phosphotransferase. J Biol Chem. 1966;241:673-683.
Bittl JA, Ingwall JS. Intracellular high-energy phosphate turnover in normal and hypertrophied myocardium. Circulation. 1987;75(suppl I):I-96-I-101.
Zweier JL, Jacobus WE, Korecky B, Brandeis-Barry Y. Bioenergetic consequences of cardiac phosphocreatine depletion induced by creatine analogue feeding. J Biol Chem. 1991;266: 20296-20304.
Hamman BL, Bittl JA, Jacobus WE, Ingwall JS. Inhibition of creatine kinase decreases the contractile reserve of the isolated rat heart. Am J Physiol. In press.
Bittl JA, Balschi JA, Ingwall JS. Contractile failure and high-energy phosphate turnover during hypoxia: 31P-NMR surface coil studies in living rat. Circ Res. 1987;60:871-878.
Ingwall JS, Kobayashi K, Bittl JA. In vivo enzymology of the creatine kinase reaction in the isolated rat heart: 31P magnetization transfer studies. In: Smirnoff VN, Katz A, eds. Proc 6th Joint USA-USSR Symposium on Myocardial Metabolism. New York, NY: Gordon and Breach; 1987:3-48.