Rapid Electrical Stimulation of Contraction Reduces the Density of β-Adrenergic Receptors and Responsiveness of Cultured Neonatal Rat Cardiomyocytes
Possible Involvement of Microtubule Disassembly Secondary to Mechanical Stress
Background—Although tachycardia is commonly present in patients with congestive heart failure, its role in the development of congestive heart failure remains unclear. We studied the effect of rapid electrical stimulation of contraction on β-adrenergic receptor (β-AR) signal pathway in cultured cardiomyocytes of neonatal rats.
Methods and Results—Contraction of cardiomyocytes was induced by electrical stimulation at 50 V with twice the threshold pulse width. β-ARs were identified by [3H]CGP-12177 and [3H]dihydroalprenolol. Electrical stimulation reduced cell-surface but not total β-AR density; the effect was dependent on pacing frequency (a reduction of 11%, 28%, and 18% in cells paced at 2.5, 3.0, and 3.3 Hz, respectively). This reduction was apparent at 3 hours, in contrast to reduced β-AR density after exposure to isoproterenol (ISP) for 1 hour. The fraction and inhibition constant of β-AR binding agonist with high affinity were not affected by rapid electrical stimulation. In cardiomyocytes paced at 3.0 Hz for 24 hours, the response to ISP decreased compared with unpaced cells, 142% versus 204% of baseline with 1 μmol/L ISP, whereas the responses to forskolin or acetylcholine were not different. Treatment of cardiomyocytes with 2,3-butanedione monoxime (10 mmol/L) or taxol (10 μmol/L) inhibited the rapid pacing–induced reduction in β-AR density.
Conclusions—Our results suggest that contractile activity is involved in regulation of cardiac function by modulating the β-AR system independently of hemodynamic and neurohormonal factors. This may help to elucidate the role of mechanical stress in the development of heart failure.
Tachycardia-induced cardiomyopathy (TCM) has been widely used as a model to study the mechanism responsible for the onset and progression of congestive heart failure (CHF), because alterations in left ventricular function and the neurohormonal system are similar to those seen in patients with developing heart failure.1 2 3 4 Furthermore, tachycardia is one of the most common clinical syndromes seen in CHF patients. However, this symptom has been viewed only as a marker of sympathetic nervous tone, whereas its role in the development of CHF has received little attention.
Recent studies have demonstrated that treatment with β-blockers fails to prevent the progression of cardiac dysfunction and neurohormonal activation in TCM.5 6 Furthermore, clinical CHF trials reporting on the effectiveness of β-blockers generally show that improvement of cardiac pump function parallels reductions in resting heart rate.7 8 The beneficial cardiac effects of amiodarone in patients with cardiac dysfunction are dependent on basal heart rate and its reduction during treatment.9 These findings suggest that alterations in β-adrenergic receptor (β-AR) signal pathway noted in CHF patients might be attributable primarily to excessive mechanical load associated with increased contraction rate rather than β-AR overstimulation by exposure to elevated catecholamines.
In the present study, we compared the effects of rapid electrical stimulation of contraction with those of isoproterenol (ISP) on the β-AR system in cultured neonatal rat cardiomyocytes. Our model allows a direct assessment of the mechanical stress in the absence of neurohormonal and hemodynamic influences. The possible roles of microtubules and active tension generation in mediating the effects of rapid contraction on the β-AR system were pharmacologically assessed by use of 2,3-butanedione monoxime (BDM) and taxol.
The experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of Oita Medical University. Cardiac myocytes were prepared from 3- to 5-day-old Wistar rats, as described previously.10 11 Briefly, hearts removed under ether anesthesia were minced into 1-mm3 pieces in PBS. After treatment with 0.02% EDTA, the pieces were incubated in 5 mL of HBSS containing 2 mg/mL collagenase (type IV; Cooper Biochemical) for 10 minutes at 37°C, and then in 5 mL of Ca2+- and Mg2+-free HBSS containing 1000 IU/mL dispase (Godo Shusei) for 20 minutes at 37°C. The isolated cells were stored in DMEM, seeded in plastic culture dishes, and incubated at 37°C under 5% CO2 in air for 90 minutes to allow the attachment of fibroblasts to the bottom of dishes. Floating cardiomyocytes were then collected by decanting.
Cardiomyocytes were plated in 4-well culture trays at a concentration of 2×106 cells/well containing DMEM supplemented with 5% FBS (Gibco), 10 mmol/L HEPES, and kanamycin (100 IU). After 48 hours of culture, >70% of the cells adhered to the culture dishes and began to oscillate (beat spontaneously). Thereafter, the culture medium was changed daily. On day 4, most cardiomyocytes beat synchronously and at a constant frequency.
On day 4 of culture, cardiomyocytes were stimulated at a high frequency by the method of Johnson et al,12 with some modifications. An electronic stimulator (SEN-7203 Nihon Kohden) was used to deliver square-wave electrical pulses at a constant voltage of 50 V through the culture medium via platinum wires submersed at opposite ends of each well of 4-well culture trays. The distance between electrodes was 6.5 cm; therefore, the strength of the electrical field was <8 V/cm, which is well below that reported to cause cell injury.13 To minimize electrolysis at electrodes and possible generation of oxidative species, the polarity of the electrodes was altered with each electrical pulse, and 250 μmol/L ascorbic acid was added to the culture medium.14 To set the threshold pulse width at 50 V, the lowest pulse width required for synchronous contraction in most cells was determined at applied frequencies of 2.0 to 3.3 Hz, and then cells were stimulated at twice this threshold. We discarded cardiomyocytes in which the threshold pulse width was >1.5 ms.
To assess any additive effects of ISP on the reduced β-AR density induced by rapid contraction, ISP was added to the culture medium during the final 3 hours of pacing.
Measurement of Cardiac β-AR Density
As previously described,10 cardiac cell-surface β-ARs were identified by binding of the hydrophilic radioligand [3H]CGP-12177 (specific activity, 1.55 TBq/mmol; Amersham Pharmacia Biotech). Stimulated or unstimulated cardiomyocytes cultured for 24 hours were washed 3 times with assay buffer (0.25 mol/L sucrose, 10 mmol/L MgCl2, and 50 mmol/L Tris-HCl [pH 7.4]) and incubated for 16 hours at 4°C with 2 mL of assay buffer containing [3H]CGP-12177 at concentrations of 0.25 to 15 nmol/L. After 2 washings with assay buffer to remove free [3H]CGP-12177, cells were lysed with 0.5 mol/L NaOH, and then 5 mL of Aquasol-2 (New England Nuclear) was added. The associated radioactivity was determined by scintillation spectroscopy. All measurements were performed in duplicate. Total β-AR density was measured with [3H]dihydroalprenolol (DHA) (specific activity, 3.52 TBq/mmol; Amersham Pharmacia Biotech) at concentrations of 0.30 to 20 nmol/L, as described above. The maximal number of binding sites (Bmax) and the dissociation constant (Kd) were calculated by Scatchard linear regression analysis, with r>0.90 as a criterion for acceptability of the data. Nonspecific binding was defined as binding in the presence of 10 μmol/L d,l-propranolol and was <20% of total binding at 5 nmol/L [3H]CGP-12177 and <30% at 10 nmol/L [3H]DHA. The protein content of the cells was measured by the biuret method.15
We also evaluated the frequency- and time-dependent effects of rapid pacing on β-AR density and inhibitory effects of BDM and taxol on rapid pacing–induced reduction in β-AR density in the presence of 10 nmol/L [3H]CGP-12177.
Agonist competition binding curves were constructed with 5 nmol/L [3H]CGP-12177 and ISP (10−10 to 5×10−4 mol/L) at 14 different concentrations. Binding data were analyzed by nonlinear regression with Graphpad PRISM (Graphpad Software, Inc), in which best fit, 1-site versus 2-site, was determined by the P value for the F test. When there was no significant difference in the fit to the 1-site or 2-site model, the data were adopted for the 2-site model.
Measurement of Response of Cultured Cardiomyocytes to ISP, Acetylcholine, and Forskolin
As previously described,10 spontaneous beating of cultured cardiomyocytes was measured. Paced and unpaced cultured cells were exposed to agents after 2 washings with the culture medium, and then spontaneous beating was monitored at 37°C for 15 minutes. Cardiomyocytes in each culture dish were exposed to a single concentration of agents and then discarded. For comparison between stimulated and unstimulated cardiomyocytes, responses to agents were assessed by changes in the spontaneous beating frequency.
Drugs and Solutions
BDM, taxol, forskolin, ISP, and acetylcholine were purchased from Sigma Chemical Co. Forskolin was dissolved in 50% (vol/vol) DMSO. BDM was dissolved in 50% (vol/vol) methanol. ISP, taxol, and acetylcholine were dissolved in distilled water. The final concentrations of DMSO (<0.1%) and methanol (<0.1%) had no effect on the spontaneous beating and β-AR density.
Data are expressed as mean±SEM and were analyzed by 1- or 2-way ANOVA and either Scheffé’s F test or Fisher’s protected least significant difference. A level of P<0.05 was considered statistically significant.
Response of Paced Cardiomyocytes to ISP, Forskolin, and Acetylcholine
We investigated the effect of ISP, forskolin, and acetylcholine on the spontaneous beating of cultured cardiomyocytes either unpaced or paced at 3.0 Hz for 24 hours. The spontaneous beating frequency at baseline was high in paced cells compared with unpaced cells (114±5 versus 96±6 bpm; n=4 to 8, P<0.05). Although the response to ISP increased dose-dependently in both paced and unpaced cells, the beating frequency was significantly lower in paced cells treated with 0.1 and 1 μmol/L ISP than in unpaced cells (Figure 1⇓). In contrast, subthreshold pacing did not affect the response of cultured cells to ISP (data not shown).
The response to forskolin was the same in paced and unpaced cells (164±36% versus 171±15% of baseline at 10 μmol/L, n=5 and 6; 125±4% versus 146±12% at 1 μmol/L, n=4 and 5; and 113±9% versus 124±14% at 0.1 μmol/L, n=6 and 5). Treatment with 50 nmol/L acetylcholine reduced the spontaneous beating frequency of paced and unpaced cardiomyocytes (a reduction of 14±4% versus 15±5%, n=5). In cardiomyocytes treated with 0.1 μmol/L ISP, acetylcholine further decreased spontaneous beating frequency of paced cells (n=4) and unpaced cells (n=8) (a reduction of 16±4% versus 22±6% at 50 nmol/L and 63±22% versus 61±12% at 1 μmol/L). However, the response to acetylcholine was not different in paced and unpaced cells in either basal or ISP-stimulated conditions.
Effects of Rapid Electrical Stimulation of Contraction on β-AR Density and Morphology of Cultured Cardiomyocytes
To explore the mechanism of blunted response to ISP in electrically stimulated cardiomyocytes, we measured cell-surface β-AR density in these cells. Representative saturation curves for [3H]CGP-12177 and Scatchard analysis of the data indicated that binding of [3H]CGP-12177 to cultured cardiomyocytes was saturated at a concentration of 10 nmol/L and that the ligand interacted with a single class of binding sites (Figure 2⇓). Electrical stimulation of cardiomyocytes at 3.0 Hz for 24 hours decreased Bmax (109±9 versus 74±8 fmol/mg protein, n=4, P<0.05) but had no effect on the affinity for the ligand (3.1±0.4 versus 2.6±0.5 nmol/L, n=4). Therefore, subsequent determinations of the cell-surface β-AR density were performed in the presence of 10 nmol/L [3H]CGP-12177. The effect of pacing on cell-surface β-AR density was frequency- and time-dependent, as shown in Figures 3⇓ and 4⇓, respectively. A reduction by 29% of cell-surface β-AR density was apparent at 3 hours and reached a plateau of 33% at 6 hours. Cell-surface β-AR density in cardiomyocytes paced at a subthreshold pulse width did not differ from that of unpaced cells (105±4 fmol/mg protein, n=4, versus 108±3 fmol/mg protein, n=10). Total β-AR density was not different between paced and unpaced cardiomyocytes, as identified with [3H]DHA (Bmax, 434±118 versus 410±137 fmol/mg protein; Kd, 11±3 versus 12±3 nmol/L, n=4).
ISP competed for binding sites with high-affinity and low-affinity constants that were not different between paced and unpaced cardiomyocytes (paced cells, 1.42±0.78 nmol/L, 1.3 μmol/L, n=5; unpaced cells, 1.92±0.58 nmol/L, 5.1±1.8 μmol/L, n=6). The proportion of β-ARs showing high-affinity binding for ISP was not different between paced and unpaced cells (23.4±11.2% versus 26.3±3.9%, respectively). A representative competitive inhibition agonist binding curve is shown in Figure 5⇓.
As shown in Figure 6⇓, there were no gross morphological differences between paced and unpaced cardiomyocytes for 24 hours at 3 Hz as assessed by phase-contrast microscopy.
Mechanism of Reduced β-AR Density Induced by Rapid Electrical Stimulation of Contraction
We compared the effects of ISP and rapid electrical stimulation of contraction on β-AR density. As shown in Figure 7⇓, exposure to 1 μmol/L ISP reduced β-AR density on unpaced cardiomyocytes significantly, by 32% at 1 hour and 23% at 3 hours. In contrast, the effect of pacing at 3.0 Hz on β-AR density appeared only at 3 hours but not at 1 hour. Exposure to 1 μmol/L ISP caused a further decrease in β-AR density on paced cells at 2.5 Hz but not at 3.0 Hz for 24 hours (Figure 8⇓). The duration of ISP-induced increase in beating frequency was <1 hour (90±6 and 102±9 bpm at baseline and 45 minutes after exposure to ISP, respectively, n=6), which was not long enough to increase β-AR density.
Next, we investigated the effects of BDM and taxol on rapid pacing–induced reduction in β-ARs. Treatment of cardiomyocytes with BDM at 10 mmol/L or taxol at 10 μmol/L abolished the effect of rapid pacing on β-AR density (unpaced, 99±5; paced, 82±3; paced+BDM, 106±4; paced+taxol, 100±1 fmol/mg protein), whereas they did not affect β-AR density on unpaced cardiomyocytes (Figure 9⇓). Furthermore, treatment with 10 μmol/L taxol had no effect on spontaneous beating and did not interfere with electrical stimuli. In the presence of 10 mmol/L BDM, the response to electrical stimuli was diminished in some cardiomyocytes (<20%) paced for 24 hours in 1 of the 4 cultures, whereas spontaneous beating was not affected in unpaced cells.
We demonstrated in the present study that rapid electrical stimulation of contraction reduced cell-surface β-AR density and the response of cultured neonatal rat cardiomyocytes to ISP. This is the first report that provides evidence for the direct involvement of contractile activity in the regulation of cardiac function by modulating the β-AR system through microtubules, independent of hemodynamics and neurohormonal factors.
Rapid Pacing Reduces Cell-Surface β-AR Density and Response to ISP
The potential role of contractile activity in the development of CHF has been underinvestigated, although tachycardia is one of the most common clinical syndromes seen in CHF patients. Recently, Spinale et al5 investigated the effects of β-blockers on cardiac function in TCM, which has been used as a heart failure model with an alteration in the β-AR signal pathway. They concluded that modulation of chronotropy contributes to the beneficial cardiac effects of β-blockers during the progression of TCM. Levett et al6 also showed that treatment with β-blockers did not prevent the progression of cardiac dysfunction and neurohormonal activation in TCM. These findings suggest that alterations of the β-AR system noted in TCM might be attributed primarily to increased contractile activity rather than β-AR overstimulation by exposure to elevated plasma catecholamines, which has been proposed as a mechanism for the blunted β-AR responsiveness.16 To address this issue, using cultured cardiomyocytes, we examined the effects of rapid stimulation of contraction on the β-AR system. This experimental setup allows the direct assessment of the role of contractile activity in the absence of confounding factors, such as neurohormonal and hemodynamic influences. In the present study, cell-surface but not total β-AR density and the response of cardiomyocytes to ISP were diminished by rapid electrical stimulation of contraction, as shown in Figures 1⇑, 3⇑, and 4⇑. In addition, although both rapid pacing– and ISP-induced β-AR stimulation reduced cell-surface β-AR density in cultured cardiomyocytes, the time courses of their respective effects on β-AR density obviously differed, and their effects were in part additive, as shown in Figures 7⇑ and 8⇑. These results show that increased contractile activity reduces cell-surface β-AR density and diminishes the response to ISP independently of neurohormonal and hemodynamic influences.
In TCM, abnormalities in components downstream of β-ARs, such as guanine nucleotide–binding regulatory (G) proteins and adenylyl cyclase activity, have been reported.1 In the present study, however, responses to forskolin and acetylcholine were not different between paced and unpaced cardiomyocytes when assessed by change of spontaneous beating frequency, a physiological end point. These results clearly indicated that G protein (Gs and Gi) levels and adenylyl cyclase activity were not markedly affected or involved in blunted response to ISP in cardiomyocytes paced at high frequency. Although competitive inhibition binding curves were performed here with [3H]CGP-12177, the values obtained, such as inhibition constants for ISP and the proportion of β-ARs displaying high-affinity binding, were not significantly different from those of previous studies using 125I-labeled cyanopindolol or [3H]DHA in dogs, pigs, or calves.1 17 18 The present data obtained from these curves were not different between paced and unpaced cardiomyocytes, suggesting no change in coupling to Gs of β-ARs reduced by rapid stimulation of contraction.
The present study was carefully designed to minimize the confounding effects of oxidative species and nonspecific toxic effects of the voltage on cultured cardiomyocytes, as described in detail in the Methods section. Phase-contrast microscopy demonstrated no significant differences in histological changes between paced and unpaced cardiomyocytes (Figure 6⇑). Thus, it appears that the effects of rapid pacing on the β-AR system are associated with an increase in contractile activity rather than cell damages by nonspecific effects exerted by electrical stimulation.
Possible Involvement of Microtubules and Active Tension Generation in Reduced β-AR Density Induced by Rapid Pacing
The role of the cytoskeleton, including microtubules, in contractile dysfunction has been shown in hypertrophied19 and failing myocardium in humans20 and in TCM.21 In cultured neonatal rat cardiomyocytes, there are ≥2 reported mechanisms of microtubular disassembly: one is Ca2+ overload,22 and the other is mechanical stress.23 In the present study, the effect of rapid pacing on β-AR density was abolished by treatment with 10 μmol/L taxol, which hyperpolymerizes and stabilizes microtubules.24 These results are consistent with those of Palmer et al,25 who demonstrated that the microtubule assembly and β-AR responsiveness of isolated cardiomyocytes were depressed in a rat model of cardiac hypertrophy 30 weeks after aortic contraction and that increased microtubule assembly by taxol partially recovered the response to ISP. In the present study, the effect of rapid pacing on β-AR density was also inhibited by treatment with 10 mmol/L BDM. This concentration of BDM is reported to effectively inhibit actin-myosin crossbridge formation and thereby uncouple the electrical and mechanical components of excitation-contraction coupling with relatively little effect on peak concentration of intracellular free Ca2+ of cardiomyocyte transient.26 27 Therefore, the data obtained with BDM suggest that the active tension generation during contraction contributes to the rapid pacing–induced reduction in β-AR density and is probably more relevant for this process than the increase of transient Ca2+ influxes. Of interest, this finding contrasted well with the primary importance of elevated cytosolic Ca2+ levels for microtubular disruption by β-AR activation.22 Furthermore, we showed previously that microtubules regulate cell-surface β-AR density in cultured cardiomyocytes through their involvement in the receptor recycling process,11 a finding supported by data from other laboratories.28 29 Taken together, it is possible that microtubule disassembly secondary to mechanical stress is involved in reduced β-AR density induced by rapid pacing via changes in the receptor recycling process. However, the intracellular Ca2+ level and the state of microtubule assembly remain to be investigated in future studies.
Our results suggest that tachycardia noted in CHF patients may be a relevant and independent component of the vicious circle of heart failure. Therefore, if CHF patients have an additional reversible component of cardiac dysfunction due to tachycardia, heart rate reduction itself should contribute to improved ventricular function or prevention of progression of ventricular dysfunction. This may also provide insight into the beneficial effects of β-blockers and amiodarone in CHF patients and the poor prognosis associated with the basal heart rate in patients with cardiovascular diseases.7 9 30 However, care must be applied in extrapolating this relatively short-term study in vitro to the pathogenesis of chronic heart failure in vivo.
In conclusion, the present results suggest that β-AR downregulation induced by contractile activity might be one of the mechanisms underlying the onset and/or progression of heart failure. Accordingly, reduction of heart rate might represent an important component of treatment of CHF. Our findings also help to elucidate the role of mechanical stress in the development of heart failure.
- Received September 7, 1999.
- Revision received December 3, 1999.
- Accepted December 22, 1999.
- Copyright © 2000 by American Heart Association
Kiuchi K, Shannon RP, Helmer GA, et al. Myocardial β-adrenergic receptor function during the development of pacing induced heart failure. J Clin Invest. 1993;91:907–914.
Cory CR, McCutcheon LJ, O’Grady M, et al. Compensatory downregulation of myocardial Ca channel in SR from dogs with heart failure. Am J Physiol. 1993;264:H926–H937.
Armstrong PW, Stopps TP, Ford SE, et al. Rapid ventricular pacing in the dog: pathophysiological studies of heart failure. Circulation. 1986;74:1075–1084.
Wilson JR, Douglas P, Hickey WF, et al. Experimental congestive heart failure produced by rapid ventricular pacing in the dog: cardiac effects. Circulation. 1987;75:857–867.
Levett JM, Marinelli CC, Lund DD, et al. Effects of β-blockade of neurohormonal responses and neurochemical markers in pacing-induced heart failure. Am J Physiol. 1994;266:H468–H475.
CIBIS Investigators and Committees. A randomized trial of β-blockade in heart failure: the Cardiac Insufficiency Bisoprolol Study (CIBIS). Circulation. 1994;90:1765–1773.
Just H. Pathophysiological targets for beta-blocker therapy in congestive heart failure. Eur Heart J. 1996;17(suppl B):2–7.
Yonemochi H, Yasunaga S, Teshima Y, et al. Mechanism of β-adrenergic receptor upregulation induced by ACE inhibition in cultured neonatal rat cardiac myocytes: roles of bradykinin and protein kinase C. Circulation. 1998;97:2268–2273.
Yonemochi H, Saikawa T, Takakura T, et al. Effects of calcium antagonists on β-receptors of cultured cardiac myocytes isolated from neonatal rat ventricle. Circulation. 1990;81:1401–1408.
Johnson TB, Kent RL, Bubolz BA, et al. Electrical stimulation of contractile activity accelerates growth of cultured neonatal cardiocytes. Circ Res. 1994;74:448–459.
Tovar O, Tung L. Electroporation and recovery of cardiac cell membrane with rectangular voltage pulses. Am J Physiol. 1992;32:H1128–H1136.
Gornall AG, Bardawill CJ, David MM. Determination of serum proteins by means of the biuret reaction. J Biol Chem. 1948;177:751–766.
Patel M, Stewart JM, Loud AV, et al. Altered function and structure of the heart in dogs with chronic elevation in plasma norepinephrine. Circulation. 1991;84:2091–2100.
Anzai T, Lai NC, Gao M, et al. Dissociation between regional dysfunction and β-adrenergic receptor signaling in heart failure. Am J Physiol. 1998;275:H1267–H1273.
Vatner DE, Young MA, Knight DR, et al. β-Receptors and adenylate cyclase: comparison of nonischemic, ischemic, and postmortem tissue. Am J Physiol. 1990;258:H140–H144.
Tsutsui H, Ishihara K, Cooper G. Cytoskeletal role in the contractile dysfunction of hypertrophied myocardium. Science. 1993;260:682–687.
Schaper J, Froede R, Hein St, et al. Impairment of myocardial ultrastructure and changes of cytoskeleton in dilated cardiomyopathy. Circulation. 1991;83:504–514.
Eble D, Spinale FG. Contractile and cytoskeletal content, structure, and mRNA levels with tachycardia-induced cardiomyopathy. Am J Physiol. 1995;268:H2426–H2439.
Hori M, Sato H, Kitakaze M, et al. β-Adrenergic stimulation disassembles microtubules in neonatal rat cultured cardiomyocytes through intracellular Ca2+ overload. Circ Res. 1994;75:324–334.
Watson PA, Hannan R, Carl LL, et al. Contractile activity and passive stretch regulate tubulin mRNA and protein content in cardiac myocytes. Am J Physiol. 1996;271:C684–C689.
Tsutsui H, Tagawa H, Kent RL, et al. Role of microtubules in contractile dysfunction of hypertrophied cardiocytes. Circulation. 1994;90:533–555.
Palmer B, Valent S, Holder EL, et al. Microtubules modulate cardiomyocyte β-adrenergic response in cardiac hypertrophy. Am J Physiol. 1998;275:H1707–H1716.
Marsh JD, Lachance D, Kim D. Mechanisms of beta-adrenergic receptor regulation in cultured chick heart cells. Circ Res. 1985;57:171–181.