Response of Failing Canine and Human Heart Cells to β2-Adrenergic Stimulation
Background Failing human hearts lose β1- but not β2-adrenergic receptors. In canine hearts with tachypacing failure, the ratio of β2- to β1-adrenergic receptors is increased. The present study was designed to determine whether heart failure increases sensitivity to β2-adrenergic stimulation in isolated canine ventricular cardiomyocytes and to verify that myocytes from failing human ventricles contain functional β2-adrenergic receptors.
Methods and Results Myocytes from healthy dogs, dogs with tachypacing failure, and human transplant recipients were loaded with fura 2-AM and subjected to electric field stimulation in the presence of zinterol, a highly selective β2-adrenergic agonist. Zinterol significantly increased [Ca2+]i transient amplitudes in all three groups. The failing canine myocytes were significantly more responsive than normal to β2-adrenergic stimulation. We also measured isotonic twitches, indo-1 fluorescence transients, and L-type Ca2+ currents in healthy canine myocytes. Zinterol (10−5 mol/L) elicited large increases in the amplitudes of simultaneously recorded twitches and [Ca2+]i transients. Zinterol also increased L-type Ca2+ currents in the normal canine myocytes; this augmentation was abolished by 10−7 mol/L ICI 118,551. cAMP production by suspensions of healthy and failing canine myocytes was not increased by zinterol (10−9 to 10−5 mol/L), nor did 10−5 mol/L zinterol elicit phospholamban phosphorylation.
Conclusions Failing human ventricular cardiomyocytes contain functional β2-adrenergic receptors. Canine myocytes also contain functional β2-adrenergic receptors. The canine ventricular response to β2-agonists is increased in tachypacing failure. Positive inotropic responses to β2-stimulation are not mediated by increases in cAMP or cAMP-dependent phosphorylation of phospholamban.
- calcium channels
- heart failure
- myocardial contraction
- receptors, adrenergic, beta
- sarcoplasmic reticulum
β-Adrenergic stimulation increases the rate and force of myocardial contraction and enables the heart to respond appropriately to increased peripheral demands. Although there are several different types of β-adrenergic receptors (βAR), it had been thought that those in the myocardium are primarily the β1 subtype (β1AR). However, strong evidence is now available that ventricular muscle cells (myocytes) also contain functional β2-adrenergic receptors (β2AR).1 2 3 4 5 Because the reduced contractile function characteristic of chronic heart failure in humans is accompanied by a substantial loss of β1AR with little or no loss of β2AR,1 6 7 the potential role of β2AR for improving cardiac performance has received considerable attention.8
To better characterize myocardial β2AR receptor responses, we examined the effects of zinterol, a highly selective β2AR partial agonist, on healthy and failing canine cells. We chose to investigate the canine model of heart failure because of the many similarities between canine and human myocytes, both of which are from large, slow-heart-rate mammals. We also examined the effects of zinterol on failing human myocytes.
Canine Heart Failure
Rapid pacing heart failure was induced in six mongrel dogs. Under general anesthesia with ketamine-diazepam-halothane, a bipolar pacing catheter was advanced percutaneously through the left jugular vein so that the electrode tip could be screwed into the endocardium of the right ventricle near its apex. The catheter was attached to a Medtronic 8320 pacemaker carried by the dog in a protective cervical collar. The ventricle was paced with a pulse of 0.5 ms at 2.5 V and 240 beats per minute. Each week, the pacemaker was turned off briefly, and ventricular function during sinus rhythm was assessed by left ventricular shortening fraction and the presence of an S3 gallop. The degree of cardiomegaly and pulmonary congestion and/or edema was assessed radiographically. All dogs had slightly reduced ventricular function by week 1, and all dogs were in moderately severe to severe heart failure by week 5. This heart failure was manifested by tachycardia, an S3 gallop, ascites, exercise intolerance, a shortening fraction of approximately 10% (normal, >25%), and severe cardiomegaly and pulmonary congestion, usually with pulmonary edema. After 5 weeks of rapid pacing, dogs were killed with an overdose of sodium pentobarbital, and their hearts were removed for isolation of ventricular cardiomyocytes. Seven healthy dogs served as controls. Myocytes were prepared from four additional healthy dogs for studies not involving a comparison between healthy and failing cells. All procedures conformed to the guiding principles of the American Physiological Society and were approved by the Institutional Laboratory Animal Care and Use Committee.
Explanted Human Ventricles
Five failing human hearts were obtained at the time of cardiac transplant. Two hearts were from patients with dilated cardiomyopathy and three were from patients with coronary artery disease. The Table⇓ describes each individual patient.
Canine and human myocytes were isolated in Columbus, Ohio, by use of segmental collagenase perfusion as described previously.9 10 Healthy canine cells were sent by overnight mail to Baltimore, Md, where measurements of L-type Ca2+ currents, indo-1 fluorescence transients, and unloaded shortening were made.
[Ca2+]i-Dependent Fura 2 Fluorescence Ratio Transients
Canine and human cells in Columbus were loaded with 5 μmol/L fura 2-AM for 5 minutes, postincubated for 1 hour at room temperature, superfused with bicarbonate-buffered Krebs-Henseleit, pH 7.4, at 37°C, and field stimulated at 0.2 Hz with parallel platinum electrodes.11 Fluorescence measurements, with excitation alternating between 340 and 380 nm, were obtained with a PTI filterscan. Because cells loaded with fura 2-AM accumulate Ca2+-sensitive dye in the mitochondria,12 accurate calibration of the cytosolic dye signal is problematic. Data are therefore presented as the ratios of fluorescence intensity at 340- and 380-nm excitation. The amplitude and configuration of the fura 2 ratio transients are thought to reflect accurately the magnitude and time course of changes in cytosolic free [Ca2+].
Unloaded Shortening and Indo-1 Transients
Myocytes were loaded with the fluorescent Ca2+ probe indo-1 as described by Xiao and Lakatta.13 Cells were placed on the stage of a modified inverted Zeiss microscope equipped for simultaneous recording of indo-1 fluorescence and cell length and were superfused with HEPES-buffered Krebs-Henseleit at 23°C.
L-Type Ca2+ Currents
Canine myocytes were placed on the stage of an inverted microscope and superfused with HEPES buffer consisting of (mmol/L) CaCl2 1.0, NaCl 137, CsCl 5, dextrose 15, MgCl2 1.3, and HEPES 20; pH was adjusted to 7.4 with NaOH. Low-resistance (1.8 to 2 MΩ) patch pipette electrodes were filled with a solution containing (mmol/L) CsCl 120, HEPES 20, MgCl2 5, NaCl 10, EGTA 5, and MgATP 3; pH was adjusted to 7.2 with CsOH. All experiments were performed at room temperature. Membrane current was recorded in the whole-cell configuration with a discontinuous switch clamp on an Axoclamp amplifier (Axon Instruments) and controlled by a Vax 11/730 computer with an LPA-1 Lab Interface. The computer was also used to acquire (2 kHz), store, and analyze the cell length and the membrane current. The membrane current was measured as the difference between the peak of inward current and the current at the end of the 200-ms pulse. To inactivate sodium current, cells were voltage clamped at −40 mV, and 200-ms depolarizing test pulses to +20 mV were applied at 0.5 Hz.
Measurement of cAMP in Suspensions of Intact Canine Myocytes
Suspensions of canine myocytes (≈0.4 mg protein/mL) were incubated for 5 minutes at 37°C with the indicated concentration of zinterol or isoproterenol. Cells were then separated from their suspending medium and simultaneously extracted by rapid centrifugation through a layer of bromododecane into 2N perchloric acid. The acid extracts were neutralized with Freon-trioctylamine, and cAMP content was analyzed by radioimmunoassay as described by Hohl and Li.9
The degree of phospholamban phosphorylation was detected by its characteristic mobility shift on SDS-PAGE.14 Cells were incubated 5 minutes with the indicated drug, dissolved in SDS (10% final concentration), and shipped on dry ice to Indianapolis, Ind. Samples (80 μg myocyte protein per lane) were electrophoresed on a 7% to 18% polyacrylamide gradient and transferred to nitrocellulose in 50 mmol/L phosphate, pH 7.4, at 3 A for 90 minutes. The nitrocellulose sheet was blotted with BSA and then incubated with phospholamban monoclonal antibody 2D12 at a 1:500 dilution. Antibody binding was detected colorimetrically with alkaline phosphatase–coupled protein A (Sigma Chemical Co) and bromochloroindolyl–nitro blue tetrazolium (Promega).
All dose response curves were fitted by use of graphpad inplot. Tests for statistically significant differences between groups used Microsoft excel and a two-tailed Student’s t test. A value of P<.05 was considered statistically significant.
Zinterol Effects on Normal Canine Cells
In healthy canine myocytes superfused at 23°C, 10−5 mol/L zinterol increased the amplitude of the isotonic twitch and ([Ca2+]i)-dependent indo-1 ratio transient (Fig 1⇓). These effects were due, at least in part, to activation of L-type Ca2+ currents. Zinterol produced a substantial rise in nifedipine-sensitive Ca2+ currents in patch-clamped myocytes (Fig 2⇓) that was fully reversed by the highly selective β2-antagonist ICI 118,551.
Failing Canine Myocytes
To ascertain the effects of heart failure on canine myocytes, we compared fura 2 [Ca2+]i transients in 15 healthy cells and 17 cells from dogs with tachypacing heart failure. The mean baseline fluorescence ratio for the healthy cells was 1.68±0.09 compared with 1.45±0.09 for those from failing hearts. This difference was not statistically significant. The mean increase in the fluorescence ratio during electric field stimulation without added drugs was 0.54±0.02 for cells from healthy hearts and 0.56±0.03 for cells from dogs with rapid pacing failure. The [Ca2+]i transient duration tended to be longer in the failing myocytes (compare Figs 3⇓ and 4⇓), but the difference was not statistically significant when the data from all cells were analyzed.
β2AR activation with zinterol caused a dose-dependent abbreviation of the [Ca2+]i transients and an augmentation of transient amplitude in both healthy (Fig 3⇑) and failing myocytes (Fig 4⇑). The failing cells were significantly more responsive to zinterol, especially at low concentrations of the drug (Fig 5⇓).
Because zinterol is a partial β2AR agonist, we investigated the effects of isoproterenol, a full but nonselective βAR agonist, on [Ca2+]i transients in 7 myocytes from two healthy dogs. Both ICI 118,551, a selective β2AR antagonist, and CGP-20712A, a selective β1AR antagonist, attenuated the effects of 100 nmol/L isoproterenol (Fig 6⇓).
To characterize further the effects of β2AR stimulation, we examined the effects of zinterol on cAMP production by suspensions of healthy and failing canine myocytes. Intracellular cAMP did not increase significantly with any concentration of zinterol tested up to 10−5 mol/L, despite a pronounced rise in cAMP when portions of the same cell suspensions were titrated with isoproterenol (Fig 7⇓). In addition, 100 nmol/L ICI 118,551 did not significantly affect cAMP production in cells incubated with 100 nmol/L isoproterenol (Fig 8⇓), whereas 300 nmol/L CGP-20712A reduced cAMP to a value only 25% greater than the control value.
Phosphorylation of phospholamban decreases electrophoretic mobility, giving rise to multiple bands on SDS-PAGE.14 Our previous study of rat myocytes demonstrated that isoproterenol and norepinephrine but not zinterol produce multiple immunoreactive phospholamban bands.15 Similar results were obtained in the present study with canine myocytes: zinterol had no discernible effect on phospholamban phosphorylation, whereas isoproterenol-treated myocytes exhibited multiple phospholamban bands (Fig 9⇓).
Despite the absence of increased cAMP or altered phospholamban mobility in the zinterol-treated canine myocytes, it is clear from Figs 3⇑ and 4⇑ that zinterol abbreviated the [Ca2+]i-dependent fura 2 fluorescence transients. Such abbreviation of [Ca2+]i transients is commonly associated with cAMP-dependent phosphorylation of phospholamban. However, as Fig 10⇓ shows, [Ca2+]i transients in healthy canine myocytes also were abbreviated by increases in extracellular [Ca2+].
Although the canine myocyte is a useful model, direct studies in human myocytes are necessary to determine whether failing human ventricular muscle cells contain functional β2AR, which might prove clinically relevant. Accordingly, we measured [Ca2+]i transients in myocytes isolated from the excised ventricles of five human transplant recipients. In right ventricular myocytes from patient 1, [Ca2+]i transient amplitude was increased 50±3% with 10−9 mol/L zinterol (n=4, mean±SEM) and 77±14% with 10−8 mol/L zinterol (Fig 11⇓). Note the biphasic [Ca2+]i transient for the cell treated with 10−8 mol/L zinterol (Fig 11C⇓). This waveform is very common when failing human myocytes are stimulated at low frequency in the presence of catecholamines. Zinterol effects were completely abolished by 10−7 mol/L ICI 118,551 (Fig 11D⇓), confirming that the effects were mediated by the β2AR. In other cells from the same heart (n=3), where cumulative dose-response curves were obtained, the responses to 10−9, 10−8, and 10−7 mol/L zinterol were nearly identical (ie, 148±8%, 145±8%, and 144±4% of control values, respectively). It was therefore impractical to obtain full dose-response curves such as those described for the canine cells.
In 7 myocytes from two other explanted hearts, we tested the effects of both ICI 118,551 and CGP-20712A on the response to zinterol. First, the cells were challenged with zinterol followed by washout of the drug. Next, the cells were exposed to zinterol plus either ICI 118,551 or CGP-20712A. This was again followed by drug washout and a subsequent superfusion with zinterol plus the second antagonist. We alternated the order in which cells were exposed to ICI 118,551 or CGP-20712A. Fig 12⇓ shows results for a typical cell from patient 4. CGP-20712A had no effect on [Ca2+]i transient amplitude, whereas ICI 118,551 again abolished the stimulatory effect of 10 nmol/L zinterol.
The Table⇑ summarizes the effects of 10−8 mol/L zinterol on [Ca2+]i transient amplitudes for all cells studied from each explanted heart. The number of cells examined from each heart is given in parentheses.
The recent overexpression of the human β2AR gene in transgenic mouse hearts has been hailed as the first step toward a gene therapy approach to heart failure,8 but difficulties could arise from our incomplete understanding of ventricular β2AR signal transduction, especially in the hearts of larger mammals. There are substantial differences between large and small mammals with respect to the ability of β2AR to regulate myocardial contractility. Possible disease-related alterations in this pathway must also be explored.
Early studies with rat ventricular cardiomyocytes suggested that β2AR arising from nonmuscle cells accounted for nearly all β2AR in rat heart homogenates.16 More recently, however, Xiao and Lakatta13 demonstrated that isolated rat ventricular cardiomyocytes are responsive to β2AR stimulation, albeit in a manner different from that commonly associated with β1AR stimulation.15 The threshold concentration for β2AR stimulation of myocyte contractility in rat ventricular myocytes is relatively high (ie, >10−7 mol/L zinterol).
By contrast, in canine myocytes, the threshold zinterol concentration for stimulation of [Ca2+]i transient amplitudes was <10−9 mol/L (Figs 3⇑ and 4⇑). Another novel and important finding of the present study was that failing canine myocytes were significantly more responsive than healthy cells to the stimulatory effects of zinterol. This enhanced sensitivity of failing versus healthy myocytes to β2AR stimulation agrees with the observation of Kiuchi et al17 that β2AR-β1AR ratios are increased in ventricular membrane preparations from dogs with rapid pacing heart failure. It remains to be established whether β2AR sensitivity is enhanced in other forms of heart failure.
β2AR Effects in Human Myocytes
Human myocytes, like canine cells, appeared to be quite responsive to β2AR stimulation. A number of investigators demonstrated positive inotropic effects of β2AR stimulation at relatively low agonist concentrations in multicellular1 2 3 and single-cell preparations18 from explanted human hearts. The results of the present study confirm and extend those observations by showing that the highly selective β2AR agonist zinterol can elicit significant increases in [Ca2+]i transient amplitude in isolated failing human ventricular cardiomyocytes. In isolated cells, potential complications attributable to catecholamine release from endogenous stores are eliminated, whereas the use of a selective β2AR agonist should eliminate any residual concerns relative to the specificity of selective antagonists. The inability of the highly selective β1AR antagonist CGP-20712A to inhibit the effects of zinterol lends further support to the argument that failing human ventricular myocytes contain functional β2ARs.
Signal transduction for both β1AR and β2AR is generally thought to occur primarily through cAMP. But although β2AR stimulation clearly activates adenylyl cyclase in some systems, Brodde et al19 noted that zinterol does not stimulate cAMP production by membrane fragments from human cardiac muscle. This observation initially raised the possibility that inotropic effects of β2AR might be cAMP-independent.
Effects of βAR Stimulation on [Ca2+]i Transient Duration
In rat myocytes, β2AR stimulation increases the [Ca2+]i transient amplitude but, unlike isoproterenol, does not abbreviate its time course13 or increase phospholamban phosphorylation.15 While zinterol causes some accumulation of cAMP in rat myocytes, the maximum elevation of the membrane-bound pool of cAMP by β2AR is only half that induced by β1AR stimulation. Therefore, in rat myocytes, there is good evidence for a dissociation between activation of adenylyl cyclase and the positive inotropic response to β2AR stimulation.
The failure of zinterol to increase cAMP or phospholamban phosphorylation in canine myocytes further supports the hypothesis that positive inotropic effects of β2AR agonists in adult ventricular myocytes are largely cAMP-independent.13 Data showing direct G-protein effects on cardiac ion channels have accumulated recently20 21 22 and could account for the positive inotropic effects of β2AR stimulation in ventricular cardiomyocytes.
Unlike in rat myocytes, however, zinterol significantly abbreviated the [Ca2+]i transients measured in both normal and failing canine cells (Figs 3⇑ and 4⇑). In rat myocytes, the cAMP-dependent phosphorylation of phospholamban accelerates Ca2+ accumulation by the sarcoplasmic reticulum and is largely responsible for shortening the [Ca2+]i transient duration. Accordingly, in rat myocytes, when [Ca2+]i transient amplitudes are increased by increased [Ca2+]o from 1 to 3.5 mmol, the time course is unchanged. In canine cells, on the other hand, increasing [Ca2+]o caused significant abbreviation of the [Ca2+]i transient (Fig 10⇑). Thus, cAMP-dependent phosphorylation of phospholamban clearly is not responsible for shortening the [Ca2+]i transient duration in zinterol-treated canine myocytes. Alternative possibilities would be phospholamban phosphorylation in beating cells by a calcium-calmodulin–dependent protein kinase23 or direct phosphorylation of the sarcoplasmic reticular Ca2+ATPase by a calcium-calmodulin–dependent protein kinase.24
Isoproterenol Effects on cAMP Production in Normal and Failing Canine Myocytes
While the failing canine myocytes exhibited an increased [Ca2+]i transient response to zinterol, these same cell preparations produced significantly less cAMP than normal when challenged with isoproterenol. This depressed cAMP response to the nonselective β-agonist is consistent with both downregulation of the β1AR and depressed adenylyl cyclase activity in failing myocardium.25
The major limitation of the present study was the small number of observations in human myocytes and the lack of healthy human control cells. Another difficulty was the lack of human myocyte preparations with the yields needed for careful titrations of drug effects on cAMP production in suspensions of intact, viable cells. Thus, it is impossible to state whether failing human ventricular cardiomyocytes, like failing canine cells, exhibit a heightened response to β2AR agonists and whether such responses are cAMP-independent.
Preparations of healthy and failing human ventricular cardiomyocytes with high yields and excellent viability have been obtained in our laboratory,10 but preparations with 0% to 10% intact rod-shaped cells are more common. Therefore, in our hands, data from human ventricular cardiomyocytes tend to be anecdotal. Nevertheless, such data are useful for evaluating the clinical relevance of myocyte data from experimental animal models. That failing canine myocytes exhibit responses to β2AR stimulation similar to those observed in five consecutive preparations of failing human myocytes supports the use of canine models for studies of βAR alterations in heart failure.
With respect to the studies of canine myocytes, although cAMP did not increase in zinterol-treated cells, there could have been localized increases in cAMP that were not detected by measurement of whole-cell values. Additional studies are required to address this issue.
The robust response of failing myocytes from human and canine hearts to β2AR stimulation indicates that these receptors may take on added importance in at least some forms of heart failure. The question is whether enhanced contractile responses to β2AR stimulation are beneficial or detrimental over the long term. Positive inotropes that inhibit cAMP phosphodiesterase (ie, milrinone) shorten survival in patients with heart failure.26 In this regard, β1AR but not β2AR stimulation evokes spontaneous [Ca2+]i oscillations, increases the diastolic indo-1 fluorescence ratio, and causes a decline in resting cell length in rat myocytes.13 In sheep Purkinje fibers treated with isoproterenol, triggered activity resulting from oscillatory afterpotentials appears to result primarily from the stimulation of β1AR, with little or no contribution from the β2ARs.27 Therefore, selective activation of β2AR or its overexpression through gene therapy might provide safe inotropic support without increasing the likelihood of arrhythmias.
These studies were supported by Grant-in-Aid 91013490 (Dr Altschuld) from the American Heart Association, funds contributed by the Ohio Affiliate of the American Heart Association and Parke Davis, and NIH grants HL-36240 and HL-48835 (Dr Altschuld), PO1-HL-06308 and HL-49428 (Dr Jones), and HL-45120 (Dr Robitaille). Zinterol was a generous gift from Bristol-Myers, Evansville, Ind. We wish to thank Dr Pal Vaghy and Ron Phillips for their invaluable assistance.
- Received January 9, 1995.
- Revision received March 3, 1995.
- Accepted March 10, 1995.
- Copyright © 1995 by American Heart Association
Bristow MR, Ginsburg R, Umans V, Fowler M, Minobe W, Rasmussen R, Zera P, Menlove R, Shah P, Jamieson S, Stinson EB. β1- and β2-adrenergic–receptor subpopulations in nonfailing and failing human ventricular myocardium: coupling of both receptor subtypes to muscle contraction and selective β1-receptor downregulation in heart failure. Circ Res. 1986;59:297-309.
Hall JA, Kaumann AJ, Brown MJ. Selective β1-adrenoceptor blockade enhances positive inotropic responses to endogenous catecholamines mediated through β2-adrenoceptors in human atrial myocardium. Circ Res. 1990;66:1610-1623.
Fowler MB, Laser JA, Hopkins GL, Minobe W, Bristow MR. Assessment of the beta-adrenergic receptor pathway in the intact failing human heart: progressive receptor downregulation and subsensitivity to agonist response. Circulation. 1986;74:1290-1302.
Milano CA, Allen LF, Rockman HA, Dolber PC, McMinn TR, Chien KR, Johnson TD, Bond RA, Lefkowitz RJ. Enhanced myocardial function in transgenic mice overexpressing the β2-adrenergic receptor. Science. 1994;264:582-586.
Hohl CM, Li Q. Compartmentation of cAMP in adult canine ventricular myocytes: relation to single-cell free Ca2+ transients. Circ Res. 1991;69:1369-1379.
Jacobson SL, Altschuld RA, Hohl CM. Muscle cell cultures from human heart. In: Piper HM, ed. Cell Culture Techniques in Heart and Vessel Research. Berlin, Germany: Springer-Verlag; 1990:74-98.
Li Q, Hohl CM, Altschuld RA, Stokes BT. Energy depletion-repletion and calcium transients in single cardiomyocytes. Am J Physiol. 1989;257:C427-C434.
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.
Wegener AD, Simmerman HK, Lindemann JP, Jones LR. Phospholamban phosphorylation in intact ventricles: phosphorylation of serine 16 and threonine 17 in response to β-adrenergic stimulation. J Biol Chem. 1989;264:11468-11474.
Xiao RP, Hohl C, Altschuld R, Jones L, Livingston B, Ziman B, Tantini B, Lakatta EG. 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.
Buxton IL, Brunton LL. Direct analysis of beta-adrenergic receptor subtypes on intact adult ventricular myocytes of the rat. Circ Res. 1985;56:126-132.
Kiuchi K, Shannon RP, Komamura K, Cohen DJ, Bianchi C, Homcy CJ, Vatner SF, Vatner DE. Myocardial β-adrenergic receptor function during the development of pacing-induced heart failure. J Clin Invest. 1993;91:907-914.
Del Monte F, Kaumann AJ, Poole-Wilson PA, Wynne DG, Pepper J, Harding SE. Coexistence of functioning β1- and β2-adrenoceptors in single myocytes from human ventricle. Circulation. 1993;88:854-863.
Yatani A, Imoto Y, Codina J, Hamilton SL, Brown AM, Birnbaumer L. The stimulatory G protein of adenylyl cyclase, Gs, also stimulates dihydropyridine-sensitive Ca2+ channels: evidence for direct regulation independent of phosphorylation by cAMP-dependent protein kinase or stimulation by a dihydropyridine agonist. J Biol Chem. 1988;263:9887-9895.
Yatani A, Codina J, Imoto Y, Reeves JP, Birnbaumer L, Brown AM. A G protein directly regulates mammalian cardiac calcium channels. Science. 1987;238:1288-1292.
Xu A, Hawkins C, Narayanan N. Phosphorylation and activation of the Ca2+-pumping ATPase of cardiac sarcoplasmic reticulum by Ca2+/calmodulin-dependent protein kinase. J Biol Chem. 1993;268:8394-8397.
Chen L, Vatner DE, Vatner SF, Hittinger L, Homcy CJ. Decreased GsÂ mRNA levels accompany the fall in Gs and adenylyl cyclase activities in compensated left ventricular hypertrophy: in heart failure, only the impairment in adenylyl cyclase activation progresses. J Clin Invest. 1991;87:293-298.