Differences in the Bioenergetic Response of the Isolated Perfused Rat Heart to Selective β1- and β2-Adrenergic Receptor Stimulation
Background— In the heart, striking functional differences exist after stimulation of the β1- and β2-adrenergic receptor (AR) subtypes. These may be linked to differences in metabolic response during β1- and β2-AR stimulation.
Methods and Results— The relation between work and metabolism was examined during selective β1- and β2-AR stimulation (β1 and β2 groups, respectively) in the isolated perfused rat heart. Measurements were made of rate-pressure product (RPP=LV developed pressure × heart rate), phosphorus-containing metabolites, and pH by 31P nuclear magnetic resonance spectroscopy and of O2 consumption by fiber-optic oximetry. Experiments were performed under high constant flow (HCF) and under flow-limiting conditions (constant pressure, CP). Despite substantially greater RPP increases relative to baseline during β1-AR (HCF, 475%; CP, 150%) than β2-AR (HCF, 90%; CP, 72%) stimulation, the relative decrease in the intracellular energy charge relative to baseline was similar for the β1 (HCF, 49%; CP, 64%) and β2 (HCF, 59%; CP, 55%) groups. For each group, an increase in oxygen consumption (MV̇o2) occurred commensurate with workload during HCF (β1, 141%; β2, 30%). During CP, however, the MV̇o2 increase was similar (β1, 39%; β2, 34%), despite the large RPP difference between the groups. During both protocols, there was greater acidosis during β1-AR than during β2-AR stimulation. Thus, at a given workload, intracellular energy charge decreased, and MV̇o2 (CP) increased to a greater extent during β2 than β1-AR stimulation.
Conclusions— The bioenergetic differences are consistent with access to an additional substrate pool during β1-AR stimulation. This may occur via increased glycogenolysis during β1-AR stimulation, facilitating increased energy production by oxidative phosphorylation, and under flow-limiting conditions, anaerobic glycolysis.
Received December 9, 2002; accepted January 14, 2003.
Beta-adrenergic receptors (β-ARs) modulate the contractile response of the heart to stress. Although the importance of both the β1- and β2-AR subtypes has been recognized, there are substantial differences between β1- and β2-AR–mediated functional responses.1–4
A recent study using canines found that cAMP signaling was coupled to PKA activation during β1- but not during β2-AR stimulation.4 For β1-AR–stimulated hearts, this led to PKA-dependent phosphorylation of the key regulatory proteins phospholamban, troponin I, and C protein and phosphorylation of the inactive b form of glycogen phosphorylase to the active a form. This suggested that the metabolic responses to β1- and β2-AR stimulation may differ. In particular, greater glycogenolysis may occur during β1-AR than β2-AR stimulation.4
We hypothesized that as a result of these differences in cAMP signaling, the relationship between cardiac contractile function and bioenergetics differs between β1- and β2-AR stimulation. Studies in the intact heart permit an evaluation of the working organ with accurate regulation of substrate and oxygen delivery without the complexity of other neurohumoral factors present in the whole organism.
31P NMR spectroscopy provides real-time measurements of the intracellular concentrations of phosphocreatine (PCr), ATP, and inorganic phosphate (Pi), as well as pH. Previous applications of this technique have been made to β-adrenergic stimulation in isolated5–8 and in situ9,10 rat hearts, isolated guinea pig hearts,11 in situ rabbit hearts12 and human hearts,13,14 each demonstrating decreased PCr concentration, but without specifically addressing differences between β1- and β2-AR–mediated effects.
In the present study, we used 31P NMR spectroscopy, oxygen consumption measurements, and measurements of contractile function to establish relationships between bioenergetics and work in the isolated perfused rat heart during β1- and β2-AR stimulation, under both limiting and nonlimiting flow conditions.
Isolated Heart Preparation
Male Wistar rats (Charles River, Wilmington, Mass), 3 to 4 months of age, were injected with 20 mg/kg 6-hydroxydopamine 24 hours before experimentation to attenuate the effect of endogenous catecholamines.15 Hearts were rapidly excised under anesthesia; perfused through the aorta with a filtered (0.45 μm), warmed (37°C), and gassed (95% O2/5% CO2) buffer in a nonrecirculating system; and allowed to beat spontaneously. The buffer consisted of (in mmol/L) 118 NaCl, 5 KCl, 0.5 Na2EDTA, 1.2 MgSO4, 25 NaHCO3, 1.8 CaCl2, 11 glucose, and 1100 U/L heparin.
A water-filled polyethylene balloon connected to a pressure transducer was inserted into the left ventricle (LV) and used to measure the LV pressure. Balloon inflation was adjusted to achieve a preload of 10 to 15 mm Hg. RPP was used as an index of workload. To test the stability of the preparation, a control group of hearts (n=5) was perfused under constant perfusion pressure (CP) for 2 hours. During this time, LV developed pressure (LVDP), heart rate (HR), and PCr/Pi ratio remained within 11.5±4% of baseline.
An O2-sensitive fiber-optic fluorometer (Presens GmbH) was used to measure the O2 saturation in the arterial line and in the venous effluent via a cannula sutured into the coronary sinus. MV̇o2 was derived from the arteriovenous O2 difference per unit coronary flow.
31P NMR Spectroscopy
Perfused hearts were placed in a glass tube, then inserted into a 9.4-T Magnex magnet (Magnex Scientific) interfaced to a Bruker DMX spectrometer (Bruker Analytik GmbH). Shimming was performed to achieve a water proton line width of ≤40 Hz. After a stabilization period, 31P spectra were acquired continuously over a period of 3 minutes using a pulse flip angle of 45°, repetition time of 1.5 seconds, spectral width of 12 kHz, 50-Hz line broadening, and zero filling to 8K points. A small tube containing 400 mmol/L methylenediphosphonate (MDP) was placed outside the heart and used as a chemical shift and signal intensity standard. Quantification was performed with Lorentzian deconvolution. The ratio of the PCr and Pi resonances was used as an index of the intracellular energy charge (IEC).16 Intracellular pH was calculated from the chemical shift difference between the PCr and Pi resonances.17
Experiments were performed both under conditions of high constant flow (HCF) and CP. In the HCF experiments, the fixed flow rate (28.4±0.1 mL/min) was equal to the highest flow rates observed under CP (at the start of the perfusion) and is at the upper end of the range typically used in isolated rat heart perfusion experiments.
Under the CP protocol (CP=120 mm Hg), the coronary flow rate decreased over the duration of the experiment for both groups (β1, 24%; β2, 28%), with the respective coronary flow time courses not significantly different (P=0.3). Therefore, the CP protocol provided a convenient method for examining flow-limiting and hence oxygen-limiting conditions.
Under each protocol, hearts underwent either selective β1-AR stimulation (β1 group) or selective β2-AR stimulation (β2 group). The doses used were based on previous studies2,3 and are listed in the Table. Three 12-minute doses of increasing concentrations of the β1 agonist norepinephrine or the β2 agonist zinterol (Bristol-Myers Squibb) were administered to the β1 and β2 groups, respectively. To increase selectivity of the receptor stimulation, an α-AR antagonist, prazosin, or a β1-AR antagonist, bisoprolol (Merck), was used for β1 and β2-stimulated hearts, respectively, in combination with the agonists.2,4 Figure 1 shows typical responses of RPP and MV̇o2 for a β1- and a β2-AR–stimulated heart during HCF at baseline and during each of the 3 doses. Typical 31P NMR spectra obtained during baseline and β-AR stimulation are shown in Figure 2.
LVDP, HR, RPP, and MV̇o2 reached their maxima approximately 3 to 4 minutes after the beginning of each dose. The data from the last 8 minutes of each period were averaged to measure dose responses. The NMR-measured metabolite concentrations were assessed from an average over the last 3 of the 4 spectra acquired during each dose period. During the normal buffer and baseline antagonist periods, all parameters showed very little change, and averages were therefore taken over the entire duration of these periods.
Data are reported as mean±SEM. The dose responses for coronary flow, LVDP, HR, PCr/Pi, MV̇o2, and pH were analyzed for main and interaction effects using repeated-measures ANOVA. Statistical comparisons of the bioenergetic-workload correlations between the β1 and β2 groups (for each protocol) and between the CP and HCF protocols (for each group) were made by use of ANCOVA and analyzed for main and interaction effects. A value of P<0.05 was considered to be statistically significant.
Preliminary experiments using both perfusion protocols showed that the maximum agonist doses chosen elicited a near-maximal response in RPP. In particular, the larger range of RPPs achieved in response to β1-AR stimulation could not be achieved during β2-AR stimulation. The mean baseline values of all measured parameters were generally consistent between the β1 and β2 groups.
LV Developed Pressure
LVDP increased significantly in a dose-dependent manner in both groups (Figure 3A). Under HCF, the maximum response for the β1 group (415% of baseline) was significantly greater than that of the β2 group (158% of baseline). Under the flow-limiting CP protocol, the maximum β1-AR–mediated LVDP response was only half that under HCF (187% of baseline), whereas the response of the β2-AR–stimulated groups was similar to that under HCF (132% of baseline).
Dose-dependent increases in HR were observed for each group (Figure 3B). The range of this response was somewhat greater during β1-AR (141% of baseline) than during β2-AR (124% of baseline) stimulation. No such difference was found under the CP protocol (134% and 132% of baseline for the β1 and β2 groups, respectively).
The dose dependence of cardiac workload as measured by the RPP is shown in Figure 3C. For the β1-AR–stimulated hearts, the workload was influenced predominantly by the LVDP response, leading to much greater RPPs at maximum dose (HCF, 575%; CP, 250%), compared with the β2 group (HCF, 190%; CP, 172%).
MV̇o2 increased relative to baseline at all doses (Figure 3D) in both groups. During the HCF protocol, the β1 group showed substantially greater maximum MV̇o2 response (241% of baseline) than the β2 group (130% of baseline). Under the flow-limiting CP protocol, significantly lower maximal MV̇o2 was observed in the β1 group only (β1, 139%; β2, 134%).
Intracellular Energy Charge
In the HCF protocol, the IEC (Figure 3E) decreased in a dose-dependent manner in both groups to similar final values (β1, 36%; β2, 45%). In the CP protocol, the dose dependence of this decrease was not significantly different for either the β1-AR (P=0.2) or β2-AR (P=0.8) group (IEC=51% and 41% at maximum dose, respectively).
During HCF, significant acidosis occurred in the β1-AR group only (Figure 3F). During CP, acidosis developed in both groups as a function of dose, although the pH reduction at maximum dose was greater in the β1-AR group (HCF, 0.06; CP, 0.09).
Comparison of Bioenergetic Responses to β1- and β2-AR Stimulation
Substantially greater RPP dose responses were achieved during β1-AR than during β2-AR stimulation (HCF and CP), despite similar decreases in IEC (HCF and CP) and MV̇o2 (CP). We sought to determine the metabolic cost associated with a given workload during β1- and β2-AR stimulation by plotting IEC and MV̇o2 as functions of RPP. A significant difference in the IEC versus RPP relationship was observed under both HCF and CP (Figure 4), with β2-AR stimulation resulting in a greater decrease in IEC as workload increased (CP, P=0.001; HCF, P=0.05).
The slopes of the MV̇o2-RPP relationship (Figure 5) are indicative of the oxygen cost associated with a given RPP increase. During HCF, there was no significant difference in this relationship between the β1 and β2 groups (P=0.3). For the CP protocol, however, a significantly greater slope was observed in the β2 group than in the β1 group (P=0.05). Thus, under CP, although the maximal workload achieved under β1 stimulation was much greater than under β2 stimulation, maximal MV̇o2 was not significantly different between the 2 groups. However, the pH measurements (Figure 3F) indicate that these higher workloads were associated with significantly greater acidosis during β1- than during β2-AR stimulation.
The major aim of this study was to compare the correlations between bioenergetics and LV workload during stimulation of β1- and β2-AR subtypes. Figure 6 is a schematic of the major factors that are expected to influence or regulate energy supply and demand and their interactions. The IEC can be viewed as an index of the supply:demand ratio, but alternatively as a regulator of the supply and demand, as depicted by the double-headed arrows (8 and 9) in Figure 6.
Greater Energy Availability and Contractile Response During β1-AR Stimulation
During the CP protocol, we observed a far greater RPP response during β1- than during β2-AR stimulation. Despite this difference, the decrease in IEC was similar in the β1- and β2-AR–stimulated hearts (Figure 4B). This indicates a similar energy supply:demand ratio and therefore an increase in energy production commensurate with workload during β1-AR stimulation. For β1-AR stimulation, the increased flow (and therefore oxygen delivery) under the HCF protocol resulted in an almost 2-fold greater contractile response compared with the response under the flow-limiting CP protocol. For the β2-AR–stimulated hearts, however, the responses under the 2 flow protocols were similar. This suggests that under the CP protocol, the maximal β1-AR–mediated increases in workload were flow limited. Despite the very large difference in the maximum workload achieved during β1- and β2-AR stimulation with HCF, the net decrease in IEC was again similar (Figure 4A), implying that energy production was augmented during β1-AR stimulation to meet the greater demand.
Increased Anaerobic Metabolism During β1-AR Stimulation Under Oxygen-Limited Conditions
Under CP, the greater workload demand and energy production during β1- compared with β2-AR stimulation occurred without a large increase in oxygen requirements (Figure 5B). This suggests that during β1-AR stimulation, anaerobic metabolism was used for this additional energy production. This is supported by the intracellular pH measurements (Figure 3F), which demonstrated significantly greater acidosis during the maximum response to β1-AR stimulation, consistent with increased lactate production from glycolysis. In contrast, during HCF, the increased work achieved by β1-AR stimulation was associated with a commensurate increase in MV̇o2 (Figure 5A) and a smaller degree of acidosis.
β1-AR Stimulation Shows Increased IEC at a Given Workload
Our measurements allow us to examine a range of workloads common to both β1- and β2-AR–stimulated hearts and specifically the metabolic responses at a given workload during β1- or β2-AR stimulation. Under both flow protocols, at a fixed RPP, the IEC was preserved to a greater extent during β1-AR stimulation (Figure 4), indicating that during β1-AR stimulation there was greater energy reserve, consistent with increased energy production. Under CP, at a given RPP, a clear difference in the MV̇o2 was found between the β1- and β2-AR–stimulated hearts (Figure 5), a difference not present in the HCF protocol. This suggests that during β1-AR stimulation under O2-limiting conditions, the increase in workload was facilitated by energy production through anaerobic pathways.
Regulation of Energy Charge and the Workload Response
Intracellular calcium ([Ca2+]i) levels are known to be greater during β1-AR stimulation1–4 and may be a more significant regulator of oxidative phosphorylation (Figure 6, arrow 5) during β1-AR stimulation than during β2-AR stimulation. This regulation would be expected to be more effective when O2 is not limiting and is consistent with the greater augmentation of energy supply and workload demand during β1-AR stimulation under HCF compared with that under CP. This could lead to more effective augmentation of the other regulators of oxidative phosphorylation, such as phosphorylation potential and mitochondrial NADH/NAD ratio (Figure 6, arrow 7), and ultimately to increased feedback between these regulators and cardiac workload. During β2-AR stimulation, energy supply and demand regulation by [Ca2+], phosphorylation potential, and mitochondrial NADH/NAD ratio may be reduced, with a similar effect occurring under reduced flow during β1-AR stimulation.
The MV̇o2 data and pH data under CP suggest that when oxygen limits the aerobic energy supply, anaerobic metabolism can be used during β1-AR stimulation in conjunction with additional substrate delivery. When oxygen delivery is increased (HCF), [Ca2+]i regulation of oxidative phosphorylation could act in concert with the additional substrate to preserve the IEC, leading to the even larger workload responses.
At the maximum workloads achieved, there was no statistical difference in the IEC between the β1 and β2 groups during CP or HCF. This suggests that the energy supply:demand ratio was similar under these conditions of near maximal stimulation. This could indicate a lower limit for IEC, which might then limit the workload response. Alternatively, however, IEC can be thought of as an energy “reserve” that can be rapidly utilized after sudden increases in demand. In this context, this reserve may be “depleted” by workload to some limiting extent during maximal stimulation. The maintenance of this increased workload would then be regulated by energy supply through the usual aerobic and anaerobic pathways, which would in turn depend on both O2 and substrate supply.
Mechanisms for Increased Energy Charge
Although we cannot draw absolute conclusions about the origin of the additional energy available under β1-AR stimulation, an attractive possibility is increased substrate supply from the breakdown of endogenous glycogen, followed by anaerobic and/or aerobic glucose metabolism. This is consistent with previous results demonstrating that glycogen phosphorylase kinase is activated during β1- but not during β2-AR stimulation.4 Glycogen availability under β1-AR stimulation may be a result of specific signaling that leads to conversion of glycogen phosphorylase from the inactive b form to the active a form.
Significant reliance on glycogenolysis and subsequent glucose metabolism during epinephrine-induced mixed β-adrenergic stimulation in the isolated heart has been observed with radiolabeling.18,19 However, epinephrine is a potent agonist for both the β1- and β2-AR subtypes. Our results suggest, for the first time, that reliance on an alternative energy source to meet requirements under β-adrenergic stimulation occurs specifically under β1-AR stimulation.
Although these results are consistent with glycogen being the origin of additional substrate available during β1-AR stimulation, we cannot distinguish between this and the possibility of enhanced delivery of exogenous glucose, for example, via an unspecified β1-AR–mediated activation of glucose transporters.
The MV̇o2 and pH differences between the β1- and β2-AR–stimulated hearts suggest that during the flow-limiting conditions of the CP experiments, a significant proportion of any potential additional substrate is metabolized anaerobically. Under HCF, the IEC data are still consistent with increased use of glycogen in the β1 hearts, but the MV̇o2 and pH data suggest that given sufficient O2, this substrate is metabolized primarily aerobically.
Furthermore, it should be noted that the β1-/β2-AR differences in MV̇o2 and IEC are also consistent with increased oxidative efficiency during β1-AR stimulation. This may result from decreased utilization of O2 for noncontractile cellular processes or futile cycles. This could result in greater preservation of IEC because of an effectively greater energy supply for contraction. However, other evidence suggests that the opposite may be true, namely, that β1-AR stimulation is less energy efficient than β2-AR stimulation. One possible mechanism that has been proposed for this is spontaneous [Ca2+]i oscillations during β1-AR stimulation.3
Effect of Competing Substrates
In the intact organism, free fatty acids (FFAs) are the preferred substrate of the heart in the fasted state. However, it has been shown that under β-AR–induced stress in the perfused rat heart, carbohydrates become the preferred substrate, with glucose and glycogen showing large increases in oxidation compared with relatively minor increases in oxidation of exogenous FFAs and endogenous triglycerides.20,21 Furthermore, the endogenous lipolytic capacity of the heart is considerably smaller than the capacity for glycogen mobilization and subsequent carbohydrate oxidation, which is directly regulated by mitochondrial feedback. Therefore, we expect that metabolism of exogenous FFAs and endogenous triglycerides would make only a minor contribution to the observed differences between β1- and β2-AR stimulation.
We have demonstrated for the first time, in a working heart preparation, a difference in metabolic response to β1- and β2-AR stimulation. These differences are consistent with differences in energy supply, possibly mediated by β-AR subtype–dependent endogenous substrate utilization. These metabolic differences may underlie the large differences in β1- and β2-AR–mediated functional responses.
Xiao RP, Cheng H, Zhou YY, et al. Recent advances in cardiac β2-adrenergic signal transduction. Circ Res. 1999; 85: 1092–1100.
Xiao RP, Hohl C, Altschuld R, et al. β2-Adrenergic receptor-stimulated increase in cAMP in rat heart cells is not coupled to changes in Ca2+ dynamics, contractility, or phospholamban phosphorylation. J Biol Chem. 1994; 269: 19151–19156.
Xiao RP, Lakatta EG. β1-Adrenoceptor stimulation and β2-adrenoceptor stimulation differ in their effects on contraction, cytosolic Ca2+, and Ca2+ current in single rat ventricular cells. Circ Res. 1993; 73: 286–300.
Kuschel M, Zhou YY, Spurgeon HA, et al. β2-Adrenergic cAMP signaling is uncoupled from phosphorylation of cytoplasmic proteins in canine heart. Circulation. 1999; 99: 2458–2465.
Lamb HJ, Beyerbacht HP, Ouwerkerk R, et al. Metabolic response of normal human myocardium to high-dose atropine-dobutamine stress studied by 31P-MRS. Circulation. 1997; 96: 2969–2977.
Pluim BM, Lamb HJ, Kayser HW, et al. Functional and metabolic evaluation of the athlete’s heart by magnetic resonance imaging and dobutamine stress magnetic resonance spectroscopy. Circulation. 1998; 97: 666–672.
Goodwin GW, Taylor CS, Taegtmeyer H. Regulation of energy metabolism of the heart during acute increase in heart work. J Biol Chem. 1998; 273: 29530–29539.