Autoimmunity Against the Second Extracellular Loop of β1-Adrenergic Receptors Induces β-Adrenergic Receptor Desensitization and Myocardial Hypertrophy In Vivo
Abstract—Although immunoapheresis removing autoantibodies against the second extracellular domain of β1-adrenergic receptors (ARs) improves cardiac function in patients with cardiomyopathy, the underlying mechanisms have not been defined. We examined the role of autoimmunity against the domain in the development of cardiac dysfunction in vivo. Japanese white rabbits were immunized with a synthetic peptide corresponding to the second extracellular loop of β1-AR once a month with (β+biso rabbits, n=10) or without (β rabbits, n=13) bisoprolol treatment (2 mg/kg per day). Control rabbits received vehicle without bisoprolol treatment (n=13). Autoantibodies of IgG isotype against the domain were persistently detected in β and β+biso rabbits. Purified IgG from sera of β and β+biso rabbits increased cAMP production in a rabbit cardiac membrane preparation, which was blocked by bisoprolol. At 3 months, β-AR uncoupling with increased G protein–coupled receptor kinase 5 (GRK5) expression was found in β rabbits. At 6 months, left ventricular hypertrophy was noted with hemodynamic derangements in β rabbits. This was accompanied by decreased β1-AR density and increased inhibitory G protein and GRK5 expression, which were related to marked decrease in membrane cAMP production. These changes in β rabbits at 6 months were prevented in β+biso rabbits. There was no difference in the plasma norepinephrine concentration in the 3 groups over the observation period. Thus, autoimmunity against the second extracellular loop of β1-ARs induced profound β-AR desensitization and myocardial hypertrophy in vivo, associated with cardiac dysfunction. Sustained sympathomimetic-like actions of autoantibodies against the domain may be partly responsible for these changes.
Certain forms of acquired cardiomyopathy are associated with autoimmunity as well as virus infection.1 Autoantibodies against the second extracellular domain of β1-adrenergic receptors (ARs) are prevalent in patients with cardiomyopathy.2 3 4 Immunoapheresis removing these autoantibodies ameliorates their symptoms and hemodynamic abnormalities.5 This suggests that autoimmunity against the domain plays a role in the development of cardiac dysfunction in cardiomyopathy. However, the underlying mechanisms have not been defined.
Previous studies have shown that autoantibodies against the second extracellular domain of β1-ARs exert sustained intrinsic sympathomimetic-like actions on cardiomyocytes in vitro.2 Because chronic stimulation of β-ARs gives rise to alterations in cardiac β-adrenergic signaling,6 7 8 myocardial hypertrophy,7 9 or damage,10 we speculated that autoimmunity might be responsible for such changes, leading to the development of cardiomyopathy and cardiac dysfunction. Actually, autoantibodies against the domain are found in patients with hypertrophic cardiomyopathy as well as dilated cardiomyopathy.3 4 Although it is clear that β-adrenergic signaling is altered in cardiomyopathy,11 12 13 relations to autoimmunity are not completely understood in vivo. Especially, G protein–coupled receptor kinases (GRKs) are extremely important in modulating myocardial adrenergic signaling and cardiac function.8 13 14 Nevertheless, relationships between autoimmunity and GRKs have not been previously reported.
Beneficial effects of β blockers on cardiac function have been confirmed in patients with cardiomyopathy.15 16 17 β blockers might improve cardiac performance in patients with autoantibodies against the domain by preventing autoimmune-mediated myocardial damage. We sought to clarify the role played by autoimmunity against the second extracellular domain of β1-ARs in the development of cardiac dysfunction in vivo, by immunizing rabbits with a peptide corresponding to the domain, together with treatment by bisoprolol, a selective β1 antagonist.
Materials and Methods
Experiments were performed in 36 male Japanese white rabbits, which were 10 weeks old (1.80 to 2.20 kg). The experimental protocol was approved by our Institutional Review Board. A synthetic peptide corresponding to the second extracellular loop of rabbit β1-AR (residues 197 to 222; HWWRAESDEARRCYNDPKCCDFVTNR) was produced by Peptide Institute, Inc. Twenty-six rabbits were randomly divided into 2 groups, and they were given distilled water throughout the observation period. Thirteen rabbits were immunized by subcutaneous injection of the peptide (1 mg) dissolved in 1 mL of saline conjugated with 0.5 mL of complete and incomplete Freund’s adjuvant once a month (β rabbits). Thirteen control rabbits received saline containing adjuvant in the same manner. Ten additional rabbits that were immunized with peptide in the same manner were given oral bisoprolol (2 mg/kg per day) dissolved in distilled water (β+biso rabbits). Three months after starting immunization, 5 rabbits in each group were euthanized (3M rabbits). The remaining rabbits were immunized over 6 months (6M rabbits). The mean plasma concentration of bisoprolol at 6 hours after oral administration in β+biso rabbits was 9.1 ng/mL (2.4×10−8 mol/L), which corresponds approximately to the therapeutic plasma level.18
Peptide (50 μL; 50 μg/mL in 0.1 mol/L Na2CO3) was used to coat individual wells of a 96-well microtiter plate. The wells were then saturated with PBS supplemented with 3% skim milk, 0.1% Tween-20, and 0.01% merthiolate. Rabbit sera (50 μL) diluted 1:200 were added to the coated plates and incubated overnight at 4°C. After 3 washes with PBS, an affinity-purified biotinylated goat anti-rabbit IgG antibody solution, diluted 1:1000 in merthiolate, was allowed to react for 1 hour at room temperature. After 3 washes, the bound biotinylated antibodies were detected using streptavidin-peroxidase (1 μL/mL), H2O2 (2.5 mmol/L) and 2,2′-azino-di-(ethylbenzthiazoline) sulfonic acid (2 mmol/L). After 30 minutes, the optical densities at 492 nm were determined using an ELISA reader (Sanko Junyaku Co).
Echocardiography was performed with the rabbit under anesthesia using 3 mg/kg thiopental sodium (Tanabe Seiyaku Co). The parasternal long-axis views were obtained using a 7.5-MHz transducer connected to a Hewlett-Packard Sonos 500 ultrasonic echocardiographic system (model 77010CF), with the rabbit placed in the lateral decubitus position.
At the end of each observation period, hemodynamic measurements were performed in the open-chest condition under anesthesia with chloral hydrate as described previously,19 and finally, the hearts were rapidly removed.
Excised hearts were fixed in 10% formalin and embedded in paraffin, and both transverse and cross sections (3 μm thickness) were obtained. Sections were stained with hematoxylin and eosin and were subjected to light microscopic examinations. Planimetry was performed macroscopically on cross sections of the hearts to evaluate left ventricular (LV) muscle mass and cavity area and microscopically on 100 cross-sectioned myocytes to determine cross-sectional area per myocyte, using NIH Image.
β-AR Binding Studies
Membrane fractions were prepared using the LV myocardium as described previously.19 The densities of myocardial β-ARs were determined by a ligand binding assay with [125I]iodocyanopindolol (ICYP) (Amersham Biotech) using membrane samples as previously described.19 Competition binding experiments were performed in duplicate by incubating 50 pmol/L ICYP with (−)isoproterenol using 16 different concentrations ranging from 10−10 to 10−3 mol/L. Nonspecific binding was determined using 1 μmol/L (−)propranolol. The ratio of high- to low-affinity binding sites was determined using nonlinear regression analysis with Prism (GraphPad Software, Inc). Receptor–G protein coupling was assessed using 30 μmol/L 5′-guanylylimidodiphosphate (Gpp[NH]p). The β1/β2 subpopulations were determined by competitive binding experiments, which were performed in duplicate by incubating 50 pmol/L ICYP with CGP27012A (Novartis Pharma), a selective β1 antagonist using 16 different concentrations ranging from 10−10 to 10−3 mol/L. The percentage of β1-ARs was calculated from the high-affinity binding subpopulation.
cAMP production was determined using cardiac membrane preparations under basal conditions and after addition of 10−5 mol/L isoproterenol+10−5 mol/L GTP or of 10−5 mol/L Gpp(NH)p, as previously described.19 Concentration-response relationships for isoproterenol were examined at concentrations ranging from 10−7 to 10−4 mol/L.
Immunodetection of stimulatory (Gs) and inhibitory (Gi) G-protein levels in the membrane fraction, as well as of GRK2 and GRK5 protein levels in both the membrane and cytosolic fractions, was performed using standard SDS-PAGE and immunoblotting techniques, as previously described.20 Protein from each fraction (50 μg) was electrophoresed and then immunoblotted. Fixed samples were included on each gel as standard for quantification of the densities of each blot. Antisera against Gsα (Upstate Biotechnology, Inc) and Giα, GRK2, and GRK5 (Santa Cruz Biotechnology) were used as primary antibodies, and horseradish peroxidase–linked anti-rabbit IgG (Boehringer Mannheim) was used as a secondary antibody to detect individual protein levels. The densities of each blot were quantified by densitometric scanning and standardized by defining the mean density of membrane fraction from control rabbits at 6 months as 1.0 densitometric unit. The specificities of antibody binding for each antigen were confirmed by neutralization assays using blocking peptides.
Functional Assay With Purified IgG From Sera of Rabbits
After hemodynamic measurements, IgG fractions were isolated from sera of rabbits with an Affi-Gel Protein A Monoclonal Antibody Purification System II Kit (Bio-Rad Laboratories) according to the protocol and desalted over a Bio-Gel P-6DG desalting gel (Bio-Rad Laboratories) to be eluted with PBS (pH 7.4). A cardiac membrane preparation from a control rabbit was incubated in the presence or absence of bisoprolol with purified IgG or with PBS (30°C, 30 minutes), followed by incubation in the presence of 10−5 mol/L GTP with isoproterenol or with buffer alone (30°C, 10 minutes) to determine cAMP production.
Data are expressed as mean±SEM. Comparisons between 3 groups were performed by 1-way ANOVA accompanied by a Bonferroni post hoc test when appropriate. Changes in optical densities on ELISA and isoproterenol concentration–dependent data on cAMP production in cardiac membrane preparations from 3 groups were analyzed using a repeated-measures ANOVA, whereas differences in mean values between the groups at a specific dose were determined by 1-way ANOVA. Statistical significance was defined as P<0.05.
IgG isotype autoantibodies against the second extracellular loop of β1-ARs were persistently detected by ELISA, as early as 1 month after starting immunization, in both β and β+biso rabbits, but not in the control rabbits (Figure 1A⇓). Purified IgG (100 μg/mL) from β and β+biso rabbits increased cAMP production in cardiac membrane preparations in the absence of either bisoprolol (10−7 mol/L) or isoproterenol (10−6 mol/L), whereas IgG from control rabbits did not have a significant effect. In contrast, isoproterenol-stimulated cAMP production was partly inhibited by the IgG from β or β+biso rabbits. In the presence of bisoprolol, the stimulatory effects of the IgG from β and β+biso rabbits, as well as those of isoproterenol, were abolished (Figure 1B⇓). Moreover, we examined the concentration-response relationships on cAMP production for the purified IgG and isoproterenol and the effects of bisoprolol on these parameters. IgG from β and β+biso rabbits increased cAMP production in a concentration-dependent manner, whereas the maximum responses were lower than those of isoproterenol (Figure 1C⇓-1). On the other hand, IgG from β and β+biso rabbits inhibited isoproterenol-stimulated cAMP production in a concentration-dependent manner (Figure 1C⇓-2). These results demonstrate that autoantibodies from β and β+biso rabbits act as partial agonists for β1-ARs, competing with the full agonist isoproterenol. In the presence of bisoprolol, basal cAMP production in the absence of either the IgG or isoproterenol was lower than that in the absence of bisoprolol (Figures 1C⇓-1 and 1C-2), showing that bisoprolol acts as an inverse agonist.21 Concentration-response curves in the presence of bisoprolol revealed that the stimulatory effects of IgG from β and β+biso rabbits were reversed (Figure 1C⇓-1), whereas the effects of isoproterenol were partially blocked by bisoprolol (Figure 1C⇓-2).
Data on cardiac function are summarized in Table 1⇓. At 3 months, there was no difference in any of the parameters measured among the 3 groups. At 6 months, thickening of both the interventricular septum and the posterior wall was noted with a decrease in the LV end-diastolic dimension in β rabbits compared with control rabbits on echocardiography (Table 1⇓ and Figure 2A⇓). Hemodynamic data demonstrated an elevation in LV end-diastolic pressure and a decrease in peak negative dp/dt and cardiac output in β rabbits. There was an increase in LV weight in β rabbits at 6 months (Table 1⇓). Macroscopic examination of the hearts revealed LV hypertrophy, as evidenced by an increase in LV muscle mass, and a decrease in the cavity area (Figure 2B⇓). LV muscle mass corrected by the total cross-sectional area was increased in β rabbits. These macroscopic findings were endorsed by microscopic planimetry showing an increase in cross-sectional area per myocyte in β rabbits (Table 1⇓).
Histological findings at 3 months included mononuclear cell infiltrates in 3 rabbits each from the β and β+biso groups, but not in rabbits from the control group (Figure 3A⇓). At 6 months, although cellular infiltrates in the hearts subsided, myocardial hypertrophy with large nuclei, severe disorganizations of the myofibers, and interstitial fibrosis were present in the LV myocardium from β rabbits. β+biso rabbits demonstrated this histopathology to a lesser extent (Figure 3B⇓).
There was no difference in plasma and LV norepinephrine concentrations among the 3 groups at either 3 or 6 months (Table 2⇓). At 3 months, the maximal binding sites of total β-ARs were similar in the 3 groups, whereas the percentages of high-affinity binding sites were remarkably decreased in β rabbits. At 6 months, total β-ARs tended to decrease, primarily as a function of diminished β1 receptor density in β rabbits. The percentage of high-affinity binding sites was also decreased (Table 2⇓). Isoproterenol-competition curves revealed a loss of guanine nucleotide modulatable binding in β rabbits (Figure 4A⇓). With regard to cAMP production, at 3 months, although there was no difference in the basal and Gpp(NH)p-stimulated production in the 3 groups, the isoproterenol-stimulated production was decreased in β rabbits. At 6 months, the basal and Gpp(NH)p-stimulated, as well as isoproterenol-stimulated cAMP production, was decreased in β rabbits (Table 2⇓). Moreover, the concentration-response curves of isoproterenol revealed that the maximal cAMP production was decreased in β rabbits at both 3 and 6 months (Figure 4B⇓). These β-AR signaling abnormalities in β rabbits were prevented in β+biso rabbits.
As assessed by immunoblots, at 3 months, there was no difference in protein levels of Gsα, Giα, or GRK2 in the 3 groups, whereas GRK5 protein levels were increased in the membrane fraction from β rabbits. At 6 months, Giα protein levels as well as GRK5 protein levels were increased in the membrane fraction from β rabbits. GRK2 tended to increase in the membrane fraction from β rabbits, although there was no statistical significance. The above findings at 3 and 6 months in β rabbits were prevented in β+biso rabbits (Figures 5⇓ and 6⇓).
The present study provides 3 new major findings. First, autoimmunity against the second extracellular loop of β1-ARs induced early myocardial β-AR uncoupling with increased GRK5 expression. Profound desensitization with increased Gi protein expression ensued. Second, the autoimmune process induced cardiac hypertrophy. Third, these changes were prevented by a β1-selective antagonist bisoprolol, which has an inverse agonist activity.
Cardiomyopathic Changes Induced by Autoimmunity
By immunizing rabbits with a synthetic peptide corresponding to the second extracellular loop of β1-AR, we observed sequential changes in the hearts. As early as 1 month after starting immunization, autoantibodies against the second extracellular loop of β1-ARs were persistently detected in both β and β+biso rabbits. These autoantibodies exert agonist-like actions on β1-ARs. At 3 months, mononuclear cell infiltrates were noted in both β and β+biso rabbits, whereas β-AR uncoupling with increased GRK5 expression was observed only in the β rabbits. These findings suggest that cellular inflammation with infiltrates does not account for β-AR uncoupling and may be a nonspecific immune reaction. Whereas cellular infiltration in the hearts subsided at 6 months, LV hypertrophy with altered cardiac function occurred in the β rabbits. Profound β-AR desensitization was noted in the β rabbits and was accompanied by decreased β1-AR density and increased Gi protein as well as GRK5 expression, although there was no difference in the plasma norepinephrine concentration in the 3 groups. These results suggest that sustained sympathomimetic actions of the autoantibodies may at least partly account for β-AR signaling abnormalities and myocardial hypertrophy, because isoproterenol increases GRK8 and Giα6 expression, and induces cardiomyocyte hypertrophy.7 9
Matsui et al22 have demonstrated autoimmune-mediated cardiomyopathy by immunizing rabbits for up to 12 months with the peptide. Biventricular dilation occurred with upregulation of β-AR in immunized rabbits, which differs from the findings of the present study. Differences in the immunization period as well as in adjuvant composition or species differences may account for the inconsistent results. Preliminary results from our laboratory have revealed that cavity dilation is not observed in rabbits immunized for up to 12 months. In addition, at 12 months, LV hypertrophy is no longer present and total β-AR density was significantly decreased in β rabbits in association with increased plasma norepinephrine concentrations. It is possible that such phenomena may represent an early stage of dilated cardiomyopathy. The different results in the β-AR assay could be attributable to differential enrichment of cell membranes due to the filtration methods used in preparation.
Alterations in β-Adrenergic Signaling Associated With Autoimmunity
Previous experiments in vitro have shown that incubation of neonatal rat ventricular myocytes with autoantibodies against β1-ARs decreases β1-AR protein and mRNA expression.23 This is consistent with the present findings at 6 months, although this finding was preceded by an increase in the GRK5 protein. GRKs have a key role in modulating myocardial adrenergic signaling and cardiac function.8 13 14 GRK5 and GRK2 are enzymes normally expressed in the heart that are rapidly activated after agonist occupancy of receptors. These kinases phosphorylate the receptors, resulting in receptor–G protein uncoupling. Iaccarino et al8 showed that β-AR stimulation with isoproterenol increases expression of GRK2 but not GRK5 in mice. One previous study also showed increased GRK2 but not GRK5 expression in failing rabbit heart after myocardial infarction,24 whereas another study demonstrated increased GRK5 but not GRK2 expression in a porcine pacing-induced model of heart failure, suggesting that increased GRK5 expression may be one of the earliest changes in heart failure.25 Ping et al26 reported that although chronic inhibition of β1-AR activation by bisoprolol resulted in downregulation of GRK activity in porcine heart, no change in immunodetectable GRK2 was found, suggesting that reduced GRK5 expression might be responsible for the decreased GRK activity. Thus, it seems possible that selective stimulation of β1-ARs by the autoantibodies is partly responsible for the predominant expression of GRK5 in the present study. According to the previous study,8 GRK2 expression may also increase as the plasma norepinephrine concentration increases in the present model. Actually, GRK2 tended to increase in the membrane fraction from β rabbits at 6 months. Choi et al14 reported that although pressure-overload cardiac hypertrophy is associated with increased GRK activity, sympathoadrenal activation but not cellular hypertrophy is responsible for the increased GRK activity, which is consistent with our findings at 3 months.
Although relationships between β-AR stimulation6 or heart failure11 23 27 and Giα have been reported, those between autoimmunity and Gi expression have not previously. In the present study, increased Gi protein, coupled with decreased β1-AR density or increased GRK5, was responsible for the profound decrease in cAMP production related to altered cardiac function at 6 months. According to a previous report,28 increased Gi protein, unlike GRK5, may be related to myocardial hypertrophy, although this issue was not clarified in the present study.
Study Limitation and Clinical Implications
We checked cross-reaction of autoantibodies with other cardiac proteins in Western blots using rabbit cardiac ventricular homogenates to confirm the specificity of autoimmunity against β1-ARs. We could not detect β-ARs clearly, possibly because of their low expression. Therefore, we immunized rabbits together with bisoprolol treatment to confirm that the partial agonist autoantibodies were effecting cardiac toxicity via β1-ARs. Cardiomyopathic changes induced by autoimmunity in the present study may involve various factors, including humoral and cellular immunity, and neurohumoral factors such as catecholamines or cytokines. Because there are complex interrelationships among them, it is difficult to specify the exact cause of an individual change. We observed 2 time points for sequential changes in the hearts in addition to bisoprolol treatment and attempted to identify the mechanisms of cardiomyopathic changes seen in the present study.
Bisoprolol prevented β-AR desensitization and myocardial hypertrophy induced by autoimmunity in the present study. Bisoprolol appears to eliminate toxicity from the partial agonist autoantibodies, rather than toxicity from elevated plasma levels of norepinephrine, because there was no difference in plasma norepinephrine concentration in the 3 groups over the observation period. Based on the fact that the second extracellular domain of β-ARs, although not the normal ligand binding site, includes disulfide-bonded cysteine residues and plays an role in ligand affinity regulation and receptor activation,29 it is speculated that autoantibody binding to the domain may induce the conformational change of the receptor to an active state.2 On the other hand, although the binding site of ligands such as β-adrenergic antagonists resides within the hydrophobic transmembrane domains of the receptor,29 functional assays demonstrated that bisoprolol has an inverse agonist activity, which is the agonist-independent modulation of receptor activity inducing an inactive conformation with loss of the active state.21 This suggests that the 2-state model of G protein–coupled receptor activation21 could explain our results and that inverse agonism rather than direct interaction with autoantibodies may be responsible for the protective effects of bisoprolol in the present study. General hemodynamic effects of β blockers appear insufficient to account for the protective effects of bisoprolol in the present study because the effects on β-adrenergic signaling have been apparent at 3 months when there is no difference in any of the hemodynamic parameters among the 3 groups. The elimination of agonist-like actions of the autoantibodies by bisoprolol may also prevent myocardial expression of proinflammatory cytokines induced by chronic β-adrenergic stimulation, which are believed to have key roles in the pathophysiology of congestive heart failure,30 although there is no evidence to support this hypothesis in the present study.
The present results suggest a possible mechanism by which the removal of the autoantibodies through immunoglobulin adsorption improves cardiac performance in patients with cardiomyopathy.5 In this regard, β blockers, especially with inverse agonism, as well as immunoglobulin adsorption, could be rational therapeutic tools for patients with cardiomyopathy who have autoantibodies directed against the second extracellular domain of β1-ARs.
This study was supported in part by Grants 09670748 (to T.Y.) and 12770356 (to M.I.) from the Ministry of Education, Science and Culture of Japan, and by funds from the Idiopathic Cardiomyopathy Research Group of the Ministry of Health and Welfare of Japan. We thank Yukio Shimoguchi for technical assistance in histological analysis.
Original received September 29, 2000; revision received February 1, 2001; accepted February 1, 2001.
- © 2001 American Heart Association, Inc.
Richardson P, McKenna W, Bristow M, Maisch B, Mautner B, O’Connell J, Olsen E, Thiene G, Goodwin J, Gyarfas I, Martin I, Nordet P. Report of the 1995 World Health Organization/International Society and Federation of Cardiology Task Force on the Definition and Classification of Cardiomyopathies. Circulation. 1996;93:841–842.
Magnusson Y, Wallukat G, Waagstein F, Hjalmarson Å, Hoebeke J. Autoimmunity in idiopathic dilated cardiomyopathy. Circulation. 1994;89:2760–2767.
Peukert S, Fu ML, Eftekhari P, Poepping I, Voss A, Thalhammer C, Hempel A, Menz M, Dietz R, Osterziel KJ. The frequency of occurrence of anti-cardiac receptor autoantibodies and their correlation with clinical manifestation in patients with hypertrophic cardiomyopathy. Autoimmunity. 1999;29:291–297.
Müller J, Wallukat G, Dandel M, Bieda H, Brandes K, Spiegelsberger S, Nissen E, Kunze R, Hetzer R. Immunoglobulin adsorption in patients with idiopathic dilated cardiomyopathy. Circulation. 2000;101:385–391.
Müller FU, Boheler KR, Eschenhagen T, Schmitz W, Scholz H. Isoprenaline stimulates gene transcription of the inhibitory G protein α-subunit Giα-2 in the rat heart. Circ Res. 1993;72:696–700.
Iaccarino G, Tomhave ED, Lefkowitz RJ, Koch WJ. Reciprocal in vivo regulation of myocardial G protein-coupled receptor kinase expression by β-adrenergic receptor stimulation and blockade. Circulation. 1998;98:1783–1789.
Yamazaki T, Komuro I, Zou Y, Kudoh S, Shiojima I, Hiroi Y, Mizuno T, Aikawa R, Takano H, Yazaki Y. Norepinephrine induces the raf-1 kinase/mitogen-activated protein kinase cascade through both α1- and β-adrenoceptors. Circulation. 1997;95:1260–1268.
Bristow MR. β-Adrenergic receptor blockade in chronic heart failure. Circulation. 2000;101:558–569.
Böhm M, Gierschik P, Jakobs KH, Pieske B, Schnabel P, Ungerer M, Erdmann E. Increase of Giα in human hearts with dilated but not ischemic cardiomyopathy. Circulation. 1990;82:1249–1265.
Choudhury L, Rosen SD, Lefroy DC, Nihoyannopoulos P, Okaley CM, Camici PG. Myocardial β adrenoceptor density in primary and secondary left ventricular hypertrophy. Eur Heart J. 1996;17:1703–1709.
Cho MC, Rapacciuolo A, Koch WJ, Kobayashi Y, Joens LR, Rockman HA. Defective β-adrenergic receptor signaling precedes the development of dilated cardiomyopathy in transgenic mice with calsequestrin overexpression. J Biol Chem. 1999;274:22251–22256.
Choi DJ, Koch WJ, Hunter JJ, Rockman HA. Mechanism of β-adrenergic receptor desensitization in cardiac hypertrophy is increased β-adrenergic receptor kinase. J Biol Chem. 1997;272:17223–17229.
Kim MH, Devlin WH, Das SK, Petrusha J, Montgomery D, Starling MR. Effects of β-adrenergic blocking therapy on left ventricular diastolic relaxation properties in patients with dilated cardiomyopathy. Circulation. 1999;100:729–735.
Anzai T, Lai CL, Gao M, Hammond HK. Dissociation between regional dysfunction and β-adrenergic receptor signaling in heart failure. Am J Physiol. 1998;275:H1267–H1273.
Podlowski S, Luther HP, Morwinski R, Müller J, Wallukat G. Agonistic anti-β1-adrenergic receptor autoantibodies from cardiomyopathy patients reduce the β1-adrenergic receptor expression in neonatal rat cardiomyocytes. Circulation. 1998;98:2470–2476.
Maurice JP, Shah AS, Kypson AP, Hata JA, White DC, Glower DD, Koch WJ. Molecular β-adrenergic signaling abnormalities in failing rabbit hearts after infarction. Am J Physiol. 1999;276:H1853–H1860.
Ping P, Anzai T, Gao M, Hammond HK. Adenylyl cyclase and G protein receptor kinase expression during development of heart failure. Am J Physiol. 1997;273:H707–H717.
Ping P, Gelzer-Bell R, Roth DA, Kiel D, Insel PA, Hammond HK. Reduced β-adrenergic receptor activation decreases G-protein expression and β-adrenergic receptor kinase activity in porcine heart. J Clin Invest. 1995;95:1271–1280.
Feldman AM, Cates AE, Veazey WB, Hershberger RE, Bristow MR, Baughman KL, Baumgartner WA, Van Dop C. Increase of the 40,000-mol wt pertussis toxin substrate (G protein) in the failing human heart. J Clin Invest. 1988;82:189–197.
Murray DR, Prabhu SD, Chandrasekar B. Chronic β-adrenergic stimulation induces myocardial proinflammatory cytokine expression. Circulation. 2000;101:2338–2341.