In the past few years, data have accumulated demonstrating the spontaneous activity of unliganded G protein-coupled receptors (GPCRs). Simultaneous with this discovery was the finding that some compounds previously classified as antagonists can behave as inverse agonists in some systems and produce a decrease in the amount of spontaneously active receptors (3). However, the relatively low amount of spontaneously active receptors (R*) in most native systems led to questions concerning the (patho)physiologic relevance of treatment with inverse agonists versus neutral antagonists. Neutral antagonists are defined here as compounds that do not alter the R-R* equilibrium but can antagonize the effects of both agonists and inverse agonists. It was thought that neutral antagonists and inverse agonists would produce similar effects in systems with little or no spontaneous receptor activity. Data have illustrated, however, that at relatively low levels of receptor expression, or in native systems, inverse agonists can produce differential effects on the amount of signaling proteins involved in receptor-mediated signaling (4). For example, in Chinese hamster ovary cells transfected with relatively low levels (250 fmol/mg of protein) of the serotonin type 2C receptor (5-HT2C), the inverse agonist, SB 206553, produced homologous and heterologous sensitization of receptors mediating their signal via activation of phospholipase C (5). This sensitizing effect of SB 206553 was observed in the absence of a detectable change in baseline inositol phosphate or arachidonic acid accumulation (second messenger assays involved in 5-HT2C receptor signaling) (5). Similarly, in cardiac tissue from control mice, infusion for 90 h of inverse agonists for the β2-adrenoceptor (β2-AR), but not neutral antagonists, resulted in an upregulation of G protein receptor kinase-2 (GRK2) (1). Again, this effect was observed in the absence of any detectable change in baseline atrial contractility (1).
To date, agonists and inverse agonists are reported to reciprocally modulate cellular activity. For example, if agonists increase baseline parameters, inverse agonists decrease baseline; if agonists promote receptor downregulation, inverse agonists produce upregulation; and if receptor alkylation treatment results in rightward shifts of agonist concentration-response curves, inverse agonist curves are shifted leftward (2,3,6). Chronic agonist exposure can result in homologous and heterologous receptor desensitization, whereas incubation with 5-HT2C receptor inverse agonists, but not antagonists, has been shown to produce homologous and heterologous sensitization (5). In the current study we investigated whether this reciprocal modulation can also be observed in wild-type and transgenic (TG35) mice cardiac-specifically overexpressing the human β2-AR. In β1-/β2-AR double-knockout mice, cardiac adenoviral infection of the β2-AR, but not the β1-AR, resulted in spontaneous (ligand-independent) activity of the receptor (7). Thus, spontaneous β2-AR signaling may contribute to heterologous desensitization. We have chosen to investigate three β-AR ligands, ICI-118,551, carvedilol, and alprenolol. Previous studies in transgenic (TG4) mice with marked cardiac-specific overexpression of the human β2-AR revealed that ICI-118,551, carvedilol, and alprenolol differed in their negative intrinsic activity at the human β2-AR (1,2). The rank order of negative intrinsic activity (inhibition of left atrial tension) was ICI-118,551 (∼80%), carvedilol (∼40%), and alprenolol (<20%).
Our results suggest that inverse agonists and a neutral antagonist are equally effective at restoring the elevated protein kinase A (PKA) activity and the left atrial inotropic response to histamine in TG35 mice overexpressing the human β2-AR. These results suggest that, as is the case with agonism, inverse agonism is system dependent. In some systems, treatment with antagonists versus inverse agonists results in differential cellular responses (1,3,5), whereas in other systems, such as the current study, there appears to be no difference in the response to treatment with inverse agonists or a neutral antagonist.
METHODS
Animals
TG35 mice (2-4 months old) and litter mate nontransgenic mice were group housed (maximum of 10 mice per cage) under standard laboratory conditions with ad libitum access to food and water (12-h light-dark cycle). The TG35 mice have cardiac-specific overexpression of the human β2-AR at levels of approximately 50-fold greater than the β-AR density of wild-type mice (8). The total β-adrenoceptor Bmax is approximately 9.2 ± 2 pmol, compared with a total β-adrenoceptor Bmax in wild-type mice of 0.18 ± 0.04 pmol/mg (8). The TG35 mice do not have significantly different baseline atrial contractility from wild-type mice but do show other signs of possible chronic β-AR activation such as increased Gαi and GRK2 protein levels (1).
Drug treatments in both wild-type and TG35 mice were performed using subcutaneous osmotic minipumps for chronic (14-day) infusion. The doses were chosen to provide >90% receptor occupancy based on prior experiments and were as follows: alprenolol, 1.2 mg/kg/h; carvedilol, 0.4 mg/kg/h; and ICI-118,551, 0.7 mg/kg/h. These three compounds were chosen because they vary in the degree of inverse agonism they exhibit. The mice were briefly anesthetized via inhalation with fluothane and a small incision was made on their back for insertion of the minipumps. The animals were killed during the last hour of pump infusion, and the left atria were isolated for isometric tension recordings. The right atria and ventricles were immediately frozen in liquid nitrogen until assay for determination of PKA activity and GRK2 protein levels. All protocols were approved by the Institutional Animal Care and Use Committee at the University of Houston.
Myocardial cytosolic fraction preparation
The right atria and ventricles were thawed on ice and minced in 1 ml of ice-cold homogenization buffer (50 m M 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid pH 7.2, 150 m M NaCl, 5 m M ethylenediamine tetra-acetic acid). The minced tissues were homogenized with a polytron (Brinkmann Instruments) and the volume was brought to 10 ml with the same buffer. The homogenate was centrifuged at 1,000 rpm for 10 min at 4°C. The supernatant was passed through two layers of cheesecloth and then spun at 18,000 rpm for 10 min at 4°C in a Sorvall RC-5B/rotorSS34 refrigerated superspeed centrifuge. The supernatant was stored as the cytosolic fraction at -20°C until assay.
Total protein estimation
Total amount of protein in the prepared cytosolic fraction was quantified using BCA assay reagent kit (Pierce, Inc., Rockford, IL, U.S.A.) according to manufacturer's instructions. Briefly, a sample of 10 μl was incubated with the reagent mixture. After incubation the samples were read at 560 nm using a spectrophotometer. A standard curve was obtained by using a range of concentrations of bovine serum albumin standard protein treated similarly to test samples. The regression equation obtained from the standard curve was used to estimate the quantity of total protein in the cytosolic fraction samples.
Protein kinase A activity assay
Cytosolic samples were thawed on ice and PKA activity was determined using a colorimetric assay (Pierce) according to manufacturer's instructions. This assay, for the rapid quantitative analysis of PKA, was used to isolate labeled, phosphorylated peptide from nonphosphorylated peptide based on the specific affinity of the incorporated phosphate groups for ligands immobilized on the separation unit membranes. The detection system used a fluorescent label attached to the substrate molecules, which was measured by absorbance properties. Briefly, the samples were incubated with a fluorescent dye-labeled peptide substrate. After incubation the reaction mixture was transferred to a separation unit containing an affinity membrane that specifically binds the phosphorylated substrate. The phosphorylated fraction was collected, and 200 μl of the sample was transferred to a flat-bottom 96-well plate and quantitated by measuring absorbance at 570 nm using a microplate reader. A standard curve was obtained using a cAMP-dependent protein kinase as control and run in parallel with each set of experiments.
G protein receptor kinase-2 level determinations
Western immunoblotting was performed to detect GRK2 protein levels using cytosolic samples obtained from treated and untreated wild-type and TG35 mice. After separation on 7.5% sodium dodecyl sulfate polyacrylamide minigels (20 μg protein/lane), samples were transferred to activated polyvinylidene difluoride (0.45 μm; Millipore Corp., Bedford, MA, U.S.A.) membranes. The membranes were incubated overnight at 4°C or 1 h at room temperature in phosphate-buffered saline containing 0.01% polysorbate-20 (PBS-T) and 5% nonfat dry milk (Bio-Rad, Hercules, CA, U.S.A.). Membranes were incubated 1 h at room temperature with GRK2 protein antibody diluted 1:7,500 in PBS-T containing 2.5% nonfat dry milk. After incubation with the primary antibody, membranes were washed three times for 30 min with PBS-T. Horseradish peroxidase conjugated anti-mouse secondary antibody (1:5,000 dilution in PBS-T containing 2.5% nonfat dry milk) was added and the membranes were incubated for 1 h at room temperature. After several washes with PBS-T, blots were made visible by chemiluminescence using the enhanced chemiluminescence Western blotting detection kit (Amersham Life Science, Arlington Heights, IL, U.S.A.) according to the manufacturer's instructions. Bands were quantified using densitometer reading. The GRK2 protein level in untreated wild-type mice was considered as 100%; thus the GRK2 protein amounts in the other groups were compared with that of untreated wild-type mice and expressed as a percentage of the untreated wild type. Purified GRK2 protein was used as the positive control reagent and quantified using densitometer reading.
Isolated left atrial isometric tension measurements
Left atria from wild-type and TG35 mice were used in this study. The left atria were isolated and paced (0.5 ms, voltage at threshold +20%, and optimal frequency of 3.2 Hz) as previously described and isometric tension was recorded (8). Concentration-response curves to histamine were performed. Each response to a concentration of histamine was allowed to plateau prior to addition of the next concentration. The histamine response was then expressed as a percentage of baseline response for each mouse.
Drugs and chemicals
Isoproterenol, alprenolol, and histamine were obtained from Sigma Chemical Company (St. Louis, MO, U.S.A.). ICI-118,551-(±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol-was purchased from Tocris Cookson (Ballwin, MO, U.S.A.). Carvedilol was a gift from SmithKline Beecham. Micro-osmotic minipumps (model 2002) were obtained from Alza Corp. (Palo Alto, CA, U.S.A.). Enhanced chemiluminescence reagent was purchased from Amersham Life Science. Electrophoresis materials were purchased from Bio-Rad. The colorimetric PKA and the BCA assay reagent kits were obtained from Pierce Inc. Anti-GRK2/3 antibody, GRK2 purified protein, and anti-mouse linked antibody as well as the founder mice for the TG35 line were gifts from Dr. Robert J. Lefkowitz, Duke University.
Statistical analysis
Data are shown as mean ± SEM. Analysis of variance was applied to compare more than two groups and the Games-Howell test was used as the post hoc test. Values of p < 0.05 were considered significant.
RESULTS
Effect of ligand infusion on protein kinase A activity in TG35 and wild-type mice
We measured PKA activity in wild-type and TG35 mice with and without chronic infusions of the β-AR ligands, alprenolol, carvedilol, and ICI-118,551. Untreated TG35 mice exhibited significantly enhanced PKA activity compared with untreated wild-type mice (Fig. 1A and B). In TG35 mice, 14-day infusion of all three ligands restored the PKA activity to the levels of untreated wild-type mice (Fig. 1A). In wild-type mice, ICI-118,551 significantly increased PKA activity, whereas carvedilol and alprenolol had no effect (Fig. 1B).
Effect of ligand infusion on G protein receptor kinase-2 protein levels in TG35 and wild-type mice
We performed Western immunoblotting for GRK2 protein levels in the cytosolic samples of wild-type and TG35 mice. The GRK2 protein level in wild-type mice was considered as 100%; thus GRK2 protein amounts in the other groups were compared with that of wild-type mice and indicated as the relative percent changes. As with PKA activity, TG35 mice had significantly higher amounts of GRK2 protein than wild-type mice (Fig. 2A and B). In TG35 mice, 14-day infusions of alprenolol and carvedilol produced significantly lower GRK2 protein levels than untreated TG35, whereas ICI-118,551 treatment had no effect on GRK2 protein levels (Fig. 2A). In wild-type mice, carvedilol significantly increased GRK2 protein levels, whereas alprenolol and ICI-118,551 had no effect (Fig. 2B).
Histamine responses in TG35 and wild-type mice
We performed histamine concentration-response curves on the left atria of wild-type and TG35 mice. The baseline left atrial tension levels were 241 ± 21 mg (n = 10) and 207 ± 41 mg (n = 5) for untreated wild-type and TG35, respectively (Table 1). Responses to histamine were expressed as a percentage increase of baseline tension. For wild-type mice, the maximal histamine response was a 45 ± 8% increase (Fig. 3). However, histamine produced only slight increases in isometric tension in TG35 mice, with the maximal response being 9 ± 2% increase of baseline tension (Fig. 3).
We also examined the effects of three β-AR ligands, alprenolol, carvedilol, and ICI-118,551, on the response to histamine in wild-type and TG35 mice. These three compounds were chosen because they vary in the degree of inverse agonism they exhibit. Left atrial isometric tension was measured and compared with untreated wild-type and TG35 mice. We also compared the maximal response to histamine of each treatment group to one another. Drug treatment had no significant effect on baseline atrial tension in any of the treatment groups (Table 1).
In TG35 mice, alprenolol, carvedilol, or ICI-118,551 treatment of 14 days each augmented the left atrial response to histamine. After treatment, the maximal histamine responses were not different from those in untreated wild-type mice (Fig. 3A). The maximum responses to histamine for alprenolol, carvedilol, and ICI-118,551 groups were 26 ± 4%, 24 ± 4%, and 28 ± 8%, respectively. There was no significant difference among these three treatment groups (Fig. 3A). However, 90 h of treatment with ICI-118,551 or carvedilol did not produce a change in the histamine response compared with untreated TG35 mice (data not shown).
The maximal histamine responses in wild-type mice tended to be lower with the drug infusions. The alprenolol infusion did produce a significant decrease in the maximal histamine response, reducing it from 45 ± 8% in untreated wild-type to 17 ± 2% (Fig. 3B); however, treatments with carvedilol (25 ± 7%) or ICI-118,551 (31 ± 5%) did not significantly reduce the maximal histamine response (Fig. 3B).
DISCUSSION
In the current study, we examined the effects that chronic treatment with three β-AR ligands, alprenolol, carvedilol, and ICI-118,551, have on PKA activity and GRK2 protein amounts and on another receptor system, i.e., the cardiac histamine receptor. We used both transgenic (TG35) mice with cardiac-specific overexpression of the human β2-AR and wild-type mice. Previous studies in transgenic (TG4) mice with marked cardiac-specific overexpression of the human β2-AR revealed that ICI-118,551, carvedilol, and alprenolol differed in their negative intrinsic activity at the human β2-AR (1,2). The rank order of negative intrinsic activity (inhibition of left atrial tension) was ICI-118,551 (∼80%), carvedilol (∼40%), and alprenolol (<20%).
Serine/threonine kinases have been implicated in mediating desensitization of GPCRs. Two of these are PKA, a cAMP-dependent protein kinase, and GPCR kinases (GRKs) or more specifically in the case of cardiac β-ARs, GRK2 (9). GRKs are generally associated with homologous desensitization, whereas PKA is generally associated with heterologous desensitization.
Cardiac tissue from TG35 mice exhibited higher amounts of PKA activity and GRK2 protein than wild-type mice (Fig. 1A and 2A). All of the ligands tested reduced PKA activity in TG35 mice to levels not different from those observed in wild-type mice (Fig. 1A). Because PKA is a second messenger-dependent kinase (cAMP), a likely interpretation of the results is that both a neutral antagonist and inverse agonists are effective at lowering cAMP levels in our system.
Both alprenolol and carvedilol produced significant decreases in GRK2 protein levels in TG35 mice (Fig. 2A). However, ICI-118,551 did not significantly alter GRK2 protein amounts in these mice. Based on the rank order of negative intrinsic activity of the three compounds (ICI-118,551 > carvedilol > alprenolol), it appears the effect on GRK2 protein amount was unrelated to the inverse agonist property of the compounds. Because our experimental design was to learn about effects of the drug during administration, the mice were killed while the infusion of the neutral antagonist or inverse agonists was ongoing. Thus, cardiac β2-ARs would have still been occupied by circulating drug, and we could not test the functional significance of possible modification of homologous desensitization through altered GRK2 amounts.
Fig. 3A shows that left atrial inotropic responses to histamine were impaired in TG35 mice compared with wild-type mice. Presumably, this impairment was due to the increased number of activated β2-ARs producing heterologous desensitization of other receptor systems also coupled to GS proteins and adenylate cyclase. Such cardiac heterologous desensitization also has been demonstrated in other systems of chronic β-AR activation, such as heart failure (10). Additional evidence in support of this hypothesis is that 14-day infusion of a β2-AR neutral antagonist or inverse agonists was able to partially restore the impaired histamine response (Fig. 3A). Thus, decreasing β2-AR activity and subsequent decrease in PKA activity resulted in increased inotropic responses to histamine in TG35 mice.
Activation of β2-ARs in the TG35 mice could occur via two mechanisms, agonist-dependent (AR*) and agonist-independent active states of the receptor (R*). Both of these states promote receptor coupling to GS proteins and the subsequent physiologic responses. If agonist-dependent active states of the β2-AR were the primary form producing the elevated PKA activity, then neutral antagonists and inverse agonists would be predicted to be equally effective at restoring the histamine responses. If, however, the elevated PKA activity were being primarily produced by spontaneous (agonist-independent) receptor activity resulting from the overexpression of the β2-AR (2), then inverse agonists would be predicted to be better at restoring the impaired histamine response. Because the neutral antagonist, alprenolol, was as effective as the inverse agonists, ICI-118,551 and carvedilol, our results suggest that it is AR* that is producing the elevated PKA activity. Both TG35 and wild-type mice would presumably have AR* complexes in vivo (in the presence of circulating norepinephrine and epinephrine). However, the TG35 mice would be predicted to have a greater number of AR* at the same concentration of agonist because total AR* is correlated to both the amount of agonist and the amount of receptor.
In this study, no changes were observed in baseline atrial tensions with any of the treatment groups (Table 1). The result is different from a previous study in which we demonstrated that 90-h infusions of inverse agonists, but not neutral antagonists, resulted in an increase in baseline left atrial tension (1). We can speculate on at least two reasons for the difference. First, there were temporal differences in the infusion times (90 h in the previous study and 14 days in the current study). Second, in the previous study, tissues were taken 6 h after the drug infusions had ceased to allow for metabolism and elimination of the drug, whereas in the current study the tissues were taken while the drug was still being infused.
Our results in the TG35 mice suggest that diminishing β-AR activity can lead to decreases in PKA activity and the restoration of responses through other GPCRs. In congestive heart failure (CHF), some β-AR antagonists/inverse agonists (e.g., carvedilol) have been shown to increase left ventricular ejection fraction and to decrease mortality. Yet even in CHF, the β-AR system, though downregulated and uncoupled, is still a positive inotropic mechanism capable of increasing contractility (11). Therefore, the paradox remains, why does blocking a positive inotropic mechanism in the heart result in an increase in contractility? Perhaps the resultant increase in contractility is due to the ability of some compounds to restore inotropic responses through other receptor systems coupling to GS proteins (i.e., histamine H2 and 5-HT4 receptors) that have undergone heterologous desensitization following chronic β-AR activation. Brodde et al. (10) have reported impaired responses to GPCRs (including histamine H2 and 5-HT4 receptors) in cardiac tissue from patients with CHF. In separate findings, Sanders et al. (12) and Kaumann and Sanders (13) have reported sensitization of histamine H2 and 5-HT4 receptors in non-CHF patients chronically treated with β-AR antagonists. Therefore, when the primary inotropic mechanism supporting the heart, β-ARs, is severely impaired as in CHF, blocking this mechanism may restore responses to other receptor systems coupling to GS proteins and thereby increase contractility.
This reversing of desensitized receptors to functional receptors may be another chronic effect of β-blockers, such as the reversal of remodeling changes, which produces beneficial effects in the treatment of heart failure.
Alprenolol has been shown to possess intrinsic sympathomimetic activity in β1-AR systems (14,15) and in our results decreased the maximal histamine response in atria of wild-type mice. ICI-118,551 and carvedilol, which have no positive intrinsic sympathomimetic activity on β-AR subtypes, had no effect on maximal histamine responses in wild-type mice. This suggests that ligands with intrinsic sympathomimetic activity for β1-ARs may not be beneficial in treating CHF.
In this study, the restoration of the histamine response required 14 days of treatment. Treatment with 90-h infusions of ICI-118,551 and carvedilol did not restore the histamine response (data not shown). This also correlates well temporally with the use of β-blockers in CHF, as improvement usually does not occur until weeks after treatment has started (16).
The effects of ligand treatment on PKA activity and GRK2 protein levels in wild-type mice are not readily explainable. PKA activity was increased by ICI-118,551 in wild-type mice, yet alprenolol and carvedilol had no effect. We can speculate that for PKA activity, ICI-118,551 differed because of its high preference for β2-ARs versus β1-ARs. Carvedilol and alprenolol have similar affinities for β2-ARs and β1-ARs. Carvedilol and alprenolol are likely to have more pronounced negative inotropic effects than ICI because they also will occupy β1-adrenoceptors (the predominant β-subtype in wild-type hearts). Furthermore, carvedilol can also block contractile α1-adrenoceptors on the vasculature, which may negate the vascular effects of β2-adrenoceptor blockade. Thus, perhaps the selectivity for β2-ARs makes the peripheral vascular effects of increased resistance dominant, and this upregulates PKA activity. Carvedilol increased GRK2 protein levels in wild-type mice, and ICI-118,551 and alprenolol had no effect. For this result in wild-type mice we can again speculate based on pharmacologic differences between the three ligands. Carvedilol, as stated, is also an α1-AR antagonist, a property not shared by either alprenolol or ICI-181,551. Perhaps blockade of peripheral vascular α1-ARs produces a reflexogenic increase in sympathetic tone, which leads to increased cardiac GRK2 protein amount.
In summary, 14-day infusions of ligands with varying degrees of negative efficacy decrease PKA activity in cardiac tissue from TG35 mice. All three ligands also restore the impaired histamine response in atria of TG35 mice. This restoration of contractile responses through other GPCRs may partly explain why some β-blockers improve left ventricular ejection fraction in patients with CHF.
Acknowledgments:
The authors thank Mr. Dennis Doan for his excellent technical assistance. The authors affirm that the original studies have been carried out in accordance with the Declaration of Helsinki or with the Guide for the care and use of laboratory animals as adopted and promulgated by the U.S. National Institutes of Health.
REFERENCES
1. Nagaraja S, Iyer S, Liu X, et al. Treatment with inverse agonists enhances baseline atrial contractility in transgenic mice with chronic beta2-adrenoceptor activation. Br J Pharmacol 1999; 127:1099-104.
2. Bond RA, Leff P, Johnson TD, et al. Physiological effects of inverse agonists in transgenic mice with myocardial overexpression of the β2-adrenoceptor. Nature 1995; 374:272-6.
3. Milligan G, Bond RA, Lee M. Inverse agonism: pharmacological curiosity or potential therapeutic strategy? Trends Pharmacol Sci 1995; 16:10-3.
4. de Ligt RA, Kourounakis AP, Ijzerman AP. Inverse agonism at G protein-coupled receptors: (patho)physiological relevance and implications for drug discovery. Br J Pharmacol 2000; 130:1-12.
5. Berg KA, Stout BD, Cropper JD, et al. Novel actions of inverse agonists on 5-HT2C receptor systems. Mol Pharmacol 1999; 55:863-72.
6. Milligan G, Bond RA. Inverse agonism and the regulation of receptor number. Trends Pharmacol Sci 1997; 18:468-74.
7. Devic E, Xiang Y, Gould D, et al. Beta-adrenergic receptor subtype-specific signaling in cardiac myocytes from beta(1) and beta(2) adrenoceptor knockout mice. Mol Pharmacol 2001; 60:577-83.
8. Milano CA, Allen LF, Rockman HA, et al. Enhanced myocardial function in transgenic mice overexpressing the β2-adrenergic receptor. Science 1994; 264:582-6.
9. Freedman NJ, Lefkowitz RJ. Desensitization of G protein-coupled receptors. Recent Prog Horm Res 1996;51:319-51; discussion 352-3.
10. Brodde OE, Vogelsang M, Broede A, et al. Diminished responsiveness of Gs-coupled receptors in severely failing human hearts: no difference in dilated versus ischemic cardiomyopathy. J Cardiovasc Pharmacol 1998; 31:585-94.
11. Brodde OE, Hillemann S, Kunde K, et al. Receptor systems affecting force of contraction in the human heart and their alterations in chronic heart failure. J Heart Lung Transplant 1992; 11(4 part 2):S164-74.
12. Sanders L, Lynham JA, Kaumann AJ. Chronic β1-adrenoceptor blockade sensitises the H1 and H2 receptor systems in human atrium: role of cyclic nucleotides. Naunyn Schmiedebergs Arch Pharmacol 1996; 353:661-70.
13. Kaumann AJ, Sanders L. 5-Hydroxytryptamine causes rate-dependent arrhythmias through 5-HT4 receptors in human atrium: facilitation by chronic β-adrenoceptor blockade. Naunyn Schmiedebergs Arch Pharmacol 1994; 349:331-7.
14. Nyberg G, Wilhelmsson C, Vedin A. Intrinsic sympathomimetic activity of penbutolol. Eur J Clin Pharmacol 1979; 16:381-6.
15. Jasper JR, Michel MC, Insel PA. Amplification of cyclic AMP generation reveals agonistic effects of certain β-adrenergic antagonists. Mol Pharmacol 1990; 37:44-9.
16. Hjalmarson A, Waagstein F. The role of beta-blockers in the treatment of cardiomyopathy and ischemic heart failure. Drugs 1994;47(suppl 4):31-9; discussion 39-40.
© 2002 Lippincott Williams & Wilkins, Inc.