Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Impressive evidence has been given for a beneficial effect on mortality mediated by -adrenoceptor antagonist therapy in chronic heart failure patients (CIBIS-II Investigators and Committees, 1999; MERIT-HF Study Group, 1999). In a recent study the nonselective -adrenoceptor antagonist carvedilol was associated with a decrease in mortality even in New York Heart Association (NYHA) class IV patients (Packer et al., 2001). However, beneficial effects could not be shown for all -adrenoceptor antagonists. Xamoterol, a ₁-selective -adrenoceptor antagonist with intrinsic sympathomimetic activity (ISA) (Schwinger et al., 1990; Böhm et al., 1990), increased mortality in heart failure patients (Xamoterol in Severe Heart Failure Study Group, 1990). It was assumed that the ISA of xamoterol was responsible for this negative outcome. Celiprolol, a -adrenoceptor antagonist with ₂-agonistic properties (Thulesius et al., 1982; Böhm et al., 1992), had no beneficial effect in chronic heart failure (Witchitz et al., 2000). These data suggest an unfavorable role of any agonistic property in heart failure. In contrast, metoprolol and carvedilol have been shown to exert no ISA in human myocardium (Brixius et al., 2001). However, in ₁-adrenoceptor transgenic mice, a slight, but not statistically significant stimulation of spontaneous beating frequency of isolated right atria was detected after carvedilol treatment (Engelhardt et al., 2001).

Recently, the Beta-Blocker Evaluation of Survival Trial (2001) using the nonselective -adrenoceptor antagonist bucindolol was terminated because no benefit in overall mortality reduction was observed. Subgroup analysis of this study suggests that the nonefficacy of bucindolol was mainly due to a rather enhanced mortality in the black population and in NYHA class IV patients, whereas a survival benefit in nonblack NYHA class III patients was observed. However, in contrast to bucindolol, carvedilol was also effective in the treatment of black patients suffering from chronic heart failure (Yancy et al., 2001). This also holds true for a relatively small trial (a cohort of 54 patients) of metoprolol (Freudenberger et al., 1997). Thus, the lack of overall survival benefit of bucindolol in heart failure patients still remains unclear.

There is ongoing discussion regarding whether bucindolol may have intrinsic sympathomimetic activity. In the rat, bucindolol has clearly shown ISA (Willette et al., 1999); in human myocardium, the situation is controversial (Hershberger et al., 1990; Maack et al., 2000; Sederberg et al., 2000).

Only recently, the presence of additional -adrenoceptors was described in human myocardium. The ₃-adrenoceptor is reported to mediate negative inotropic effects in human myocardium by activation of nitric oxide synthetase involving inhibitory G-proteins (Gauthier et al., 1998) and seems to be up-regulated in human heart failure (Gauthier et al., 1998; Moniotte et al., 2001). However, the physiological role of the ₃-adrenoceptor still remains unclear in the human heart.

The existence of a new "putative ₄-adrenoceptor" was postulated because several -adrenoceptor antagonists (so called nonconventional partial agonists) cause cardiostimulant effects at concentrations that exceed their affinity for ₁- and ₂-adrenoceptors. CGP 12177, a hydrophilic -adrenergic ligand (Staehelin et al., 1983), which was supposed to be a ₁- and ₂-adrenoceptor antagonist with ₃-partial agonistic properties (Mohell and Dicker, 1989) is the most frequently investigated nonconventional partial agonist. The cardiostimulant effects of these nonconventional partial agonists are resistant to blockage by propranolol.

This putative ₄-adrenoceptor has been functionally demonstrated in several cardiac tissues, e.g., atrial myocardium of ₃-knockout mice (Kaumann et al., 1998), rat myocytes (Malinowska and Schlicker, 1996; Sarsero et al., 1999), and human myocardium (Sarsero et al., 1996; Kaumann and Molenaar, 1997). However, recent studies have demonstrated, that the putative ₄-adrenoceptor does not exist as a distinct entity. It was reported, that the putative ₄-adrenoceptor response no longer existed in brown adipose tissue of ₁-knockout mice (Konkar et al., 2000). This holds true for cardiostimulant effects as in ₁/₂-double knockout mice, the cardiostimulant effects of CGP 12177 was absent, whereas in both wild-type and ₂-knockout mice the effect was present (Kaumann et al., 2001). In addition, it could be demonstrated in a rat model of cardiac failure that the pharmacology of the putative ₄-adrenoceptor parallels that of the ₁-adrenoceptor (Kompa and Summers, 1999). Thus, the putative ₄-adrenoceptor has to be defined as a propranolol-insensitive state of the ₁-adrenoceptor. However, it cannot be excluded that this rarely described state of the ₁-adrenoceptor has an impact on inotropic or chronotropic cardiac response due to -agonistic stimulation.

Previous studies have shown that increases in cAMP worsen the prognosis of heart failure patients (Xamoterol in Severe Heart Failure Study Group, 1990; Cruickshank, 1993). Thus, besides detrimental effects due to ISA mediated by typical ₁- and ₂-adrenoceptors, it may be possible that -adrenoceptor antagonists induce detrimental effects on heart failure patients by interaction with additional atypical ₁-adrenoceptors.

Therefore, besides determination of ISA on ₁- and ₂-adrenoceptors, the present study investigated the interaction of bucindolol with the atypical state of the ₁-adrenoceptor in human myocardium. For comparison, the "typical" nonconventional partial agonist CGP 12177 as well as the frequently administrated metoprolol and the newly developed highly ₁-selective adrenoceptor antagonist nebivolol were studied.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Preparation of Isolated Auricular Trabeculae

Right atrial tissue was taken from patients undergoing aortocoronary bypass operation (n = 10, seven males, three females; age 59 ± 2 years) who were without clinical signs of cardiac failure as measured by cardiac catheterization (normal ejection fraction, end diastolic volume, and stroke volume) and echocardiography. None of the patients had received Ca²⁺ channel antagonists or Ca²⁺ channel agonists within 7 days of surgery or -adrenoceptor agonists 48 h before surgery. All of the patients received -adrenoceptor antagonists. Drugs used for general anesthesia were propofol, flunitrazepam, fentanyl, and pancuronium bromide. The tissue was delivered within 5 min into the laboratory in ice-cold pre-aerated Bretschneider solution of the following composition: 15 mM NaCl, 10 mM KCl, 4mM MgCl₂, 180 mM histidine, 2 mM tryptophan, 30 mM mannitol, and 1 mM potassium dihydrogen oxoglutarate. From each native myocardial tissue sample, auricular trabeculae were selected of 0.4 to 0.6 mm thickness and 6 to 8 mm length under microscopic control (Axiovert 100; Carl Zeiss, Oberkochen, Germany).

Preparation of Left Ventricular Papillary Muscle Strips

Nonfailing human myocardium was obtained from five donors with brain death caused by traumatic injury (n = 5, four males, one female; age 52 ± 6 years). The nonfailing hearts could not be transplanted because of technical reasons. Failing myocardium was obtained during cardiac transplantation due to dilated cardiomyopathy (n = 3, two males, one female; age 61.3 ± 3.5 years; ejection fraction, 22.7 ± 1.8; cardiac index, 2.2 ± 0.1 l/m² × min). Patients suffered from heart failure clinically classified as NYHA class IV on the basis of clinical symptoms and signs as judged by the attending cardiologist shortly before operation. Medical therapy consisted of diuretics, nitrates, angiotensin-converting enzyme inhibitors, and cardiac glycosides. Patients receiving catecholamines, -adrenoceptor- or Ca²⁺-antagonists were withdrawn from the study.

Immediately after excision, the papillary muscles were placed in ice-cold pre-aerated Bretschneider solution and delivered to the laboratory within 5 min. Muscle strips 0.6 to 0.8 mm thick and 6 to 8 mm long with muscle fibers running approximately parallel to the length of the strips were carefully dissected under microscopic control in aerated bathing solution at room temperature. Connective tissue, if visibly present, was carefully trimmed away.

The experiments were performed on isolated electrically driven muscle preparations. The preparations were attached to a bipolar platinum stimulating electrode and suspended individually in 75-ml glass tissue chambers to record the isometric contractions. The bathing solution used was modified Tyrode's solution that contained 119.8 mM NaCl, 5.4 mM KCl, 1.05 mM MgCl₂, 1.8 mM CaCl₂, 22.6 mM NaHCO₃, 0.42 mM NaH₂PO₄, 0.05 mM Na₂EDTA, and 5.5 mM glucose. It was continuously gassed with 95% O₂ and 5% CO₂ and maintained at 37°C; the pH was 7.4. The isometric force of contraction was measured with an inductive force transducer (W. Fleck, Mainz, Germany) attached to a Hellige Helco Scriptor (Hellige, Freiburg, Germany) or Gould recorder (Cleveland, OH). The muscles were electrically stimulated at 1 Hz with rectangular pulses of 5-ms duration (Grass stimulator SD 9, Quincy, MA), and the voltage was 20% above threshold. All preparations were allowed to equilibrate at least 90 min in drug-free bathing solution until complete stabilization. After 45 min, the solution was changed. The duration of stimulation for a given concentration was constant until there was complete stabilization of force development. Control strips of native myocardium showed no change in baseline isometric tension during the time necessary to complete pharmacological testing. In all experiments, the final concentration of solvent (dimethyl sulfoxide) was 0.1%. Compound-dependent changes in force of contraction were determined. None of the substances changed pH or temperature.

Determination of ISA. To investigate ISA of the different -adrenoceptor antagonists, left ventricular muscle preparations from nonfailing myocardium were preincubated with forskolin (0.3 µM) (Jasper et al., 1988). Forskolin facilitates the coupling of the stimulatory G-protein with the catalytic subunit of the adenylate cyclase. Dose-response curves for bucindolol, metoprolol, and nebivolol were measured (0.1-10 µM). In addition, the influence of low and high concentrations of bucindolol and CGP 12177 on right auricular trabeculae under baseline conditions was measured to determine possible partial agonistic activity in right atrial myocardium.

Investigations on Propranolol-Insensitive State ₁-Adrenoceptors. Effects mediated by propranolol-insensitive state ₁-adrenoceptors were investigated after successive administration of 1) propranolol (1 µM) to block cardiac ₁- and ₂-adrenoceptors (Wellstein et al., 1986); 2) N-nitro-L-arginine (L-NMA; 100 µM), to inhibit the nitric-oxide synthase (Griffith and Kilbourn, 1996) and thus the suggested key enzyme of the ₃-adrenoceptor pathway (Gauthier et al., 1998; Moniotte et al., 2001); 3) forskolin (0.3 µM); and 4) CGP 12177, bucindolol, metoprolol, or nebivolol (all 10 µM). In each experiment CGP 12177 (10 µM) was added at the end of the experiment. Under these conditions, the inotropic effects of the -adrenoceptor antagonists should be mediated by stimulation of the atypical state of the ₁-adrenoceptor.

Materials

Bucindolol was generously provided by Knoll AG (Mannheim, Germany), metoprolol by Astra GmbH (Wedel, Germany), and nebivolol by Berlin Chemie (Berlin, Germany). CGP 12177A was obtained from BioTrend (Köln, Germany). Bupranolol was provided by Schwarz Pharma (Zwickau, Germany). All other chemicals were of analytical grade or the best grade commercially available.

Statistics

All values are means ± S.E.M. Statistical significance was analyzed with Student's t test for paired observations. Significance was imparted at a p value of <0.05.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Investigation of CGP 12177. The influence of the ₄-adrenoceptor agonist CGP 12177 on basal force of contraction was investigated using electrically driven (1 Hz, 37°C) right auricular trabeculae. Basal force of contraction of right auricular trabeculae was 10.3 ± 1.1 mN/mm². CGP 12177-treatment mediated a biphasic inotropic response. In a low concentration (10 nM), CGP 12177 decreased force of contraction (9.4 ± 2.9%), whereas an increase in force of contraction could be observed in a high concentration (10 µM) (+12.8 ± 2.5%). Figure 1 shows one representative trace recording (A) as well as the means ± S.E.M. of four experiments (B).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Influence of CGP 12177 on right auricular trabeculae. A, representative trace recordings demonstrating the influence of a low and high concentration of CGP 12177 on force of contraction (FOC) (10 nM and 10 µM, respectively); B, means ± S.E.M. of four individual experiments. In a low concentration, CGP 12177 decreased force of contraction, whereas in a high concentration, force of contraction was significantly increased. , p < 0.05 versus basal; #, p < 0.05 versus 10 nM.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Original force recordings of electrically driven isolated right auricular trabeculae illustrating the inotropic response of CGP 12177, bucindolol, nebivolol, and metoprolol to propranolol-insensitive state 1-adrenoceptors. 1- and 2-adrenoceptors were blocked by propranolol (1 µM). L-NMA (100 µM) inhibits the key enzyme of the 3-adrenoceptor pathway, whereas forskolin (0.3 µM) sensitizes adenylate cyclase.

View this table: [in this window] [in a new window]   TABLE 1 Data of functional studies on the propranolol-insensitive state of the 1-adrenoceptor in right auricular trabeculae Substances were cumulatively administered as presented. Force of contraction is given in millinewtons per millimeter squared.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Influence of CGP 12177 (CGP) on propranolol-insensitive state 1-adrenoceptor in right auricular trabeculae (n = 4) as well as in left ventricular papillary muscle strips from terminal failing (NYHA class IV) myocardium (n = 4). In both cardiac tissues, the atypical state 1-adrenoceptor mediated an increase in contractility. Note that the increase in force of contraction was more pronounced in right auricular trabeculae. Prop, propranolol (1 µM); Fors, forskolin (0.3 µM); and L-NMA (100 µM). , p < 0.05 versus basal; #, p < 0.05 versus +Prop, +L-NMA, and +Fors.

Investigation of Bucindolol. The influence of bucindolol on force of contraction was investigated under basal conditions in right auricular trabeculae and after forskolin stimulation in left ventricular papillary muscle strips. In contrast to CGP 12177, bucindolol concentration dependently decreased force of contraction in right auricular trabeculae. Basal force of contraction of right auricular trabeculae was 10.4 ± 0.8 mN/mm² (n = 4). Bucindolol (10 nM) significantly decreased force of contraction by 13.8 ± 5.9%, whereas 10 µM further decreased force of contraction by 40.1 ± 1.5% (Fig. 4A).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   A, influence of bucindolol on right auricular trabeculae (n = 4) under baseline condition (basal). B, influence of bucindolol on left ventricular papillary muscle strips (n = 5) after forskolin treatment (0.3 µM). Bucindolol decreased force of contraction concentration dependently in right atrial as well as in left ventricular myocardium. , p < 0.05 versus basal; #, p < 0.05 versus 10 nM; §, p < 0.05 versus + forskolin.

View larger version (6K): [in this window] [in a new window]   Fig. 5.   Influence of bupranolol (10 µM) on the agonistic activity of bucindolol on the propranolol-insensitive state of the 1-adrenoceptor. Note that in the presence of bupranolol, the positive inotropic effect of bucindolol was no longer evident.

Investigation of Nebivolol and Metoprolol. The influence of nebivolol and metoprolol on the propranolol-insensitive state of the ₁-adrenoceptor was investigated according to CGP 12177 and bucindolol. Neither nebivolol nor metoprolol mediated positive inotropic response to atypical state ₁-adrenoceptors as demonstrated in Fig. 2, C and D, respectively. Both lead to a decrease in force of contraction after pretreatment with propranolol, L-NMA, and forskolin (26.0 ± 0.1% and 18.0 ± 0.1%, respectively) (Fig. 6 and Table 1) and thus seem to be antagonists.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Bar graphs give the means of all experiments in percentage of basal force of contraction (FOC). The first bar represents basal force of contraction, the second bar force of contraction after cumulative addition of propranolol (prop, 1 µM), L-NMA (100 µM), and forskolin (Fors, 0.3 µM). Effects of CGP 12177 (CGP, panel A), bucindolol (Buc, panel B), nebivolol (Neb, panel C) and metoprolol (Met, panel D) (all 10 µM) on force of contraction are demonstrated in the third bar. The last bar represents further addition of CGP 12177 (CGP, 10 µM). Note that the additional increase in force of contraction by CGP 12177 was less pronounced when bucindolol was added before and more pronounced after nebivolol and metoprolol. #, p < 0.05 versus -adrenoceptor antagonist; , p < 0.05 versus +Prop,+L-NMA, +Fors.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Representative original tracings after stimulation of atypical state 1-adrenoceptors demonstrating the influence on the time course of the contractile twitch of right auricular trabeculae. The solid twitch was recorded after cumulative application of propranolol (1 µM), L-NMA (100 µM), and forskolin (0.3 µM); the dotted twitch after further addition of CGP 12177 (panel A), bucindolol (panel B), nebivolol (panel C), or metoprolol (panel D) (all 10 µM).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Recently, a putative ₄-adrenoceptor has been described that mediates positive inotropic effects in myocardial tissue by activation of adenylate cyclase. However, it has been demonstrated that this is not a distinct -adrenoceptor entity but an atypical, propranolol-insensitive state of the ₁-adrenoceptor (Kaumann et al., 2001). Agonistic activation on this state of the ₁-adrenoceptor may be one candidate leading to the neutral effects on mortality in the Beta-Blocker Evaluation of Survival Trial (2001) using bucindolol. Thus, we studied whether bucindolol exhibits agonistic activity to this atypical state of the ₁-adrenoceptor.

In heart failure ₁-adrenoceptors are down-regulated (Bristow et al., 1982; Brodde, 1991; Schwinger et al., 1991). Therefore, it might be possible that the functional importance of other pathways, such as ₃-adrenoceptors or atypical state ₁-adrenoceptors may increase. Since increase in cAMP worsens the prognosis of heart failure patients (Xamoterol in Severe Heart Failure Study Group, 1990; Cruickshank, 1993), it may be possible that the different effects of -adrenoceptor antagonists on mortality are due to their agonistic activity on the propranolol-insensitive state of the ₁-adrenoceptor. Furthermore, it has been demonstrated that stimulation of this -adrenoceptor mediates arrhythmic Ca²⁺-transients in mice ventricular myocytes (Freestone et al., 1999), which might also be of importance for treating heart failure patients.

In this study, bucindolol did not possess ISA in a multicellular system containing all receptor types that might impact cardiac contractility. Since net increase/decrease in force of contraction can still be negative although ISA is present, the experiments do not provide final evidence for a lack of ISA of bucindolol.

We investigated the influence of bucindolol, nebivolol, and metoprolol on the putative ₄-adrenoceptor in human myocardium on a functional level. Functional studies on this propranolol-insensitive state of ₁-adrenoceptors were performed in isolated muscle preparations after ₁- and ₂-adrenoceptor antagonism (propranolol,1 µM), inhibition of ₃-mediated inotropic effects (L-NMA, 100 µM), and forskolin treatment (0.3 µM). The present study indicates that bucindolol, like CGP 12177, exhibits positive inotropic effects mediated by the atypical state of the ₁-adrenoceptor (of the same magnitude as the forskolin-mediated increase in force of contraction), whereas nebivolol and metoprolol appear to be antagonists in human hearts.

Another explanation for the increase in force of contraction induced by bucindolol might be a reversal of the negative inotropic effect of propranolol due to the lower inverse agonistic activity of bucindolol compared with propranolol. However, one would expect a similar effect after nebivolol treatment because nebivolol has low inverse agonistic activity comparable with bucindolol (Brixius et al., 2001; Maack et al., 2001). In addition, the positive inotropic effect of bucindolol was absent in the presence of bupranolol.

There is an ongoing discussion whether bucindolol may have ISA in human myocardium. Despite investigations that did not reveal ISA (Hershberger et al., 1990; Sederberg et al., 2000), Maack et al. (2000) detected ISA in three of eight specimens from human left ventricular myocardium, suggesting that ISA of bucindolol may depend on the examined tissue. In a recent study we did not reveal any agonistic activity (Brixius et al., 2001). The present study provides new information about an "ISA-like" intrinsic activity that appears at high concentrations and is unmasked only if the classic ₁- and ₂-pathway is blocked.

It may be suggested that bucindolol has failed to decrease mortality in heart failure patients due to activation of the cAMP-dependent atypical ₁-adrenoceptor stimulation. However, this has to be further clarified in studies focusing on the molecular levels of action of the different -adrenoceptors. The presented agonistic activity on atypical ₁-adrenoceptor may influence regulation of intracellular ion homeostasis and/or cell regulatory processes. This has to be addressed in further studies as well.

In our studies, no change in time to half-peak relaxation could be demonstrated by activation of the propranolol-insensitive state of the ₁-adrenoceptor. An agonistic activation of the -adrenoceptor pathway results in a decrease in time to half-peak relaxation (Brixius et al., 1997). Thus, the typical feature for a cAMP-mediated mechanism is not shown. One speculative explanation for this is that cAMP formation by the atypical state of the ₁-adrenoceptor is preliminarily located in a subsarcolemmal microdomain and thus force development is increased due to agonistic activation without changes in relaxation parameters. This feature has also been described for the ₂-adrenoceptor (Xiao et al., 1999). Additionally, forskolin stimulation, which acts via activation of the catalyst itself, may lead to shortened relaxation, which may not be further enhanced by the agents studied.

The present study was performed under in vitro conditions, and experiments were either performed on isolated muscle preparations or crude membrane preparations of human hearts. All of the patients received -blocking agents. It cannot be excluded that in vivo effects may differ from those observed in vitro. Furthermore, the interaction on the propranolol-insensitive state of the ₁-adrenoceptors cannot be studied with a more direct approach. Additional studies focusing on the molecular level are necessary. In this study, the -adrenoceptor antagonists were investigated at concentrations that exceed their affinity for ₁-and ₂-adrenoceptors, respectively. In addition, the investigated concentrations exceed the plasma concentrations of the investigated -adrenoceptor antagonists. The positive inotropic effect of bucindolol has been observed in nonfailing myocardium from patients undergoing aortocoronary bypass operation. Thus, it cannot be excluded that the results may differ in heart failure patients.

Whether a positive inotropic effect of the atypical state of the ₁-adrenoceptors, as has been shown for bucindolol in the present study, may have consequences for its clinical use has to be further investigated and remains at the moment speculative. However, the ₁-selective -adrenoceptor antagonists metoprolol (studied by the MERIT-HF Study Group, 1999) and nebivolol (now investigated in the Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalisation in Seniors with Heart Failure (SENIOR) trial) are not capable of any agonistic activation.