Direct labelling of myocardial ? 1-adrenoceptors

Naunyn-Schmiedeberg's Arch Pharmacol (1985) 331 : 27- 39 Naunyn-Schmiedeberg's

Archives of Pharmacology �9 Springer-Verlag 1985

Direct labelling of myocardial fll-adrenoceptors Comparison of binding affinity of 3H-(--)-bisoprolol with its blocking potency

Alberto J. Kaumann and Horst Lemoine Lehrstuhl ffir Klinische Physiologic, Physiologisches Institut der Universit/it Diisseldorf, Universit/itsstraBe 1, D-4000 Dfisseldorf, Federal Republic of Germany

Summary. A radioligand that selectively labels fl~-adrenoceptors, 3H-(-)-bisoprolol (3H-BIS), is introduced. The binding properties of 3H-BIS to membrane particles of kitten heart are compared with the blocking properties of (-)-bisoprolol against stimulant effects of (-)-adrenaline and (-)-noradrenaline in heart preparations of kitten and guinea pig.

1. On kitten heart tissues (-)-bisoprolol antagonized the positive chronotropic and inotropic effects of catecholamines competitively. The effects of (-)-adrenaline were antagonized considerably less by (-)-bisoprolol than the corresponding effects of (-)-noradrenaline on sinoatrial pacemakers. The antagonism was nearly the same against both (-)-adrenaline and (-)-noradrenaline in left atria and papillary muscles. The data were analyzed with a model for 2-receptor subtypes by non-linear regression. Equilibrium dissociation constants K~ (mol/1; - l o g KB =pKB) for a high-affinity fia-adrenoceptor of 8.8 and for a low-affinity flz-adrenoceptor of 7.0 were estimated in the three classes of tissues. In kitten sinoatrial pacemaker fl~-adrenoceptors contribute 76% to the stimulus induced by (-)-adrenaline and 97% to the stimulus induced by (-)-noradrenaline. In ventricle and left atrium fll-adrenoceptors contribute 9 7 - 99% and 100% to the stimulus caused by (-)-adrenaline and (-)-noradrenaline, respectively.

2. Both 3H-BIS and unlabelled (-)-bisoprolol caused competitive blockade of the positive chronotropic effects of (-)-noradrenaline in guinea-pig right atria, pKB-values of 8.7 were estimated for both unlabelled and tritiated ( - ) - bisoprolol. The positive chronotropic effects of (-)- adrenaline were antagonized considerably less by (-)-bisoprolol than those of (-)-noradrenaline in guinea-pig atria.

In the presence of low concentrations of flz-selective ICI 118,551, which did not antagonize fl~adrenoceptor mediated effects, (-)bisoprolol antagonized positive chronotropic effects of (-)-adrenaline to the same extent as those of (-)-noradrenaline. The results are consistent with the con- cept of a significant role of sinoatrial fl2-adrenoceptors of guinea pig for the effects of (-)-adrenaline but not for those of (-)-noradrenaline.

3. 3H-BIS associated and dissociated quickly with and from ventricular fi~-adrenoceptors. A koff of 1.0 min- 1 was estimated. An equilibrium dissociation constant pKL* of 8.2

fo r 3H-BIS was estimated from saturation binding. The binding competition of the optical isomers of bupranolol

Send offprint requests to A. J. Kaumann ici Pharmaceuticals Division, Macclesfield, Cheshire, SK10 4TF, UK

with aH-BIS revealed a degree of stereo-selectivity expected for fll-adrenoceptors.

4. (-)-Bisoprolol competed with high affinity and low affinity, respectively, for fia- and fl2-adrenoceptors of both ventricular and atrial membranes labelled with 3H-(-)- bupranolol or 3H-(-)-propranolol. The difference in affinity of (-)-bisoprolol of 1.7 + 0.3 log units between ill- and flz-adrenoceptors agrees with the selectivity for fi~- adrenoceptors found in intact heart tissues.

5. (-)-Noradrenaline and (-)-adrenaline competed with low affinity with 3H-BIS for ventricular fit-adrenoceptors. An approximately 100-fold difference between high Ke-values of these catecholamines and low inotropic ECso- values is consistent with a large fll-adrenoceptor reserve.

6. The binding affinity for fil-adrenoceptors of both 3H- BIS and (-)-bisoprolol is 3 - 8 times lower on membrane particles than on intact tissues, as also confirmed with the blockade of catecholamine-induced stimulation of the adenylate cyclase. By contrast, the affinities of (-)-propranolol and (-)-bupranolol are similar in intact tissues and on membrane particles. The low affinity of (-)-bisoprolol in membranes may be caused by a lesion of accessory recogni- tion sites of the fi~-adrenoceptors due to the procedure of membrane preparation. However, the lesion of these hypothetical accessory binding sites would not lower the affinity for (-)-propranolol and (-)-bupranolol.

Key words: Heart - 3H-(-)-Bisoprolol - Catechol- amines - fit-Adrenoceptors - Binding and blockade

Introduction

fi~- and flz-Adrenoceptors (Lands et al. 1967) were first suggested by Carlsson et al. (1972) to coexist in the same tissue. Following the visualization of 2 fl-adrenoceptor subtypes (usually considered fil and f12) in the radiolabelled fl- adrenoceptor population of the same tissue (Rugg et al. 1978) several methods have been developed to estimate the affinity of competing ligands for each subtype (Minneman et al. 1979, 1981). In such binding studies it is assumed that the radioligand has essentially the same affinity for the 2 subtypes. However, evidence is accumulating that the true binding affinity of radioligands - considered usually nonselective - is actually higher for one receptor subtype than for the other. For example, the affinity of 3H-(-)- dihydroalprenolol is at least 4 times higher for mammalian lung (usually mostly f12) (Rugg et al. 1978; Dickinson et al. 1980) than for mammalian heart (usually mostly ill) (Ch6nieux-Guicheney et al. 1978; Hancock et al. 1978;

28

Moustafa et al. 1978). Similarly, the binding affinity of 3H- (-)-propranolol is 3 times higher for lung than for heart fl-

adrenoceptors (Morris and Kaumann 1984). From theo- retical considerations and computer simulations De Lean et al. (1982) concluded that these small differences in binding affinity of the radioligand for the 2 subtypes can not be quantitated in the same tissue. Nevertheless, both the true affinity of a competing ligand for both subtypes and the true size of each subtype population are estimated wrongly when the affinity of the radioligand differs slightly for the 2 subtypes (De Lean et al. 1982). Furthermore, serious distortions in the estimates of affinity and of the size of subtype population occur with indirect fits (Hofstee plots - for example Minneman et al. 1979) and even with direct non-linear fits. To overcome these difficulties we introduce in the present paper the ill-selective ligand (-)-bisoprolol [ ( - ) -EMD 33512 - Manalan et al. 1981 ; Kaumann and Lemoine 1983] and in the accompanying paper (Lemoine et al. 1985) the flz-selective ligand ICI 118,551 (Bilski et al. 1983; Kaumann and Lemoine 1983) as radioligands. The advantages of using radioligands selective for fl-adrenoceptor subtypes are:

1. Direct estimate of the binding affinity for one subtype without interference from the other. A clean estimate of the affinity of an agonist for one subtype will improve the understanding of the physiological meaning of subtypes.

2. Direct estimate of the total amount of receptors of one subtype without interference from the other.

We verified the total fl-adrenoceptor population (containing both subtypes) by labelling heart receptors with 3H- (-)-propranolol and 3H-(-)-bupranolol (this paper) and lung receptors with 3H-(--)-bupranolol (accompanying paper of Lemoine et al. 1985). In order to check the physiological relevance of the binding affinities we used (-)-bisoprolol and ICI 118,551 as antagonists of the effects of (-)-noradrenaline in heart tissues (present paper) and tracheae (accompanying paper of Lemoine et al. 1985).

The stimulant potencies of the physiological catecholamines and the blocking potencies of the subtype- selective ligands were compared with the corresponding binding affinities of the agonists and antagonists. In the present paper we also compared the blocking potency of (-)-bisoprolol as antagonist of adenylate cyclase stimulation by catecholamines with the binding affinity of 3H-(-)- bisoprolol.

Methods

Experiments were carried out at 32.5 ~ C. Kitten of either sex (body weight 400-1000 g) were pretreated with 4 mg/kg reserpine (s.c.) for 24h. After the animals had been anaesthetized with chloroform, the chest was opened, the heart was removed and washed free of blood with oxygenated modified Krebs solution at room temperature. Atria and thin right ventricular papillary muscles (width 0.3-0.8 mm) were rapidly dissected. The remainder of the ventricles was immediately placed into icecold modified Krebs solution (see below) and used for the preparation of membrane particles. Experiments were also performed on right atria of reserpine-pretreated (J0mg/kg i.p., 24 h) guinea pigs (male, 250-280 g weight). The guinea pigs were sacrificed by a blow on the head.

Isolated tissues. Left atrial strips, right ventricular papillary muscles and spontaneously beating right atria of

kitten were dissected and mounted in pairs in an apparatus with a 50 ml bath described by Blinks (1965). The left atrial strips and papillary muscles were driven with square-wave pulses of 5 ms duration by a voltage which was barely above threshold at 2 s and 5 s intervals, respectively, delivered by a punctate electrode in contact with the tissues. Isometric contractions were measured with Statham (G 10B and G 7 B) and Swema (SG 4-45) strain gauge force transducers via preamplifiers on Watanabe (WTR 281) polygraphs. The resting length of spontaneously beating right atria was set to develop just enough force necessary to count the contractions on the polygraph. Length-force curves were determined on the driven preparations as soon as they had become reasonably stable. Papillary muscles were left at a length associated with maximal developed force of contraction. The final length of left atria was set at a resting force corresponding to approximately 1/2 of the resting force associated with maximum developed force of contraction.

The modified Krebs solution contained (mmol/1): Na + 140, K + 5, Ca 2+ 2.25, Mg z+ 0.5, C1- 98.5, SOl- 0.5, HCOj 34, HPO2 2 1, fumarate 0.5, pyruvate 5, L-glutamate 5, glucose 10, disodium edetate (EDTA) 0.04, equilibrated with 95% O2 and 5% COz (water was deionized and redistilled). Kitten tissues were pretreated once with 5 ~tmol/1 phenoxybenzamine for 2 h. This treatment results in irreversible blockade of tissue uptake of catecholamines (Kaumann 1972) and of heart e-adrenoceptors (Kaumann 1970). Spontaneously beating fight atria from guinea pigs were exposed to 6 gmol/1 cocaine throughout the experiment to prevent neuronal uptake of noradrenaline (Trendelenburg

1968). In order to detect possible stimulant and depressant

effects of (-)-bisoprolol cumulative concentration-effect curves were determined on right atria, left atria and papillary muscles of kitten. The time of incubation was 45 min for concentrations up to 20 nmol/1 and 30 min for higher concentrations.

In order to estimate the blocking potency of (-)-bisoprolol 3 or 4 cumulative successive concentration-effect curves for (-)-noradrenaline and (-)-adrenaline were performed, the first in the absence, the following curves in the presence of increasing concentrations of (-)-bisoprolol or 3H-(-)-bisoprolol. In control experiments carried out in the absence of (-)-bisoprolol successive curves were shifted slightly to the right without reduction of absolute maximum. Desensitization factors were estimated as ratios of equieffective concentrations (ECso'S) of agonists. The desensitization factors were used for correction of concentration ratios of agonists induced by (-)-bisoprolol. Desen- sitization factors were taken from Lemoine and Kaumann (1982, (-)-noradrenaline, kitten tissues) and from Lemoine and Kaumann (1983, guinea-pig right atria) or determined in this paper for (-)-adrenaline on the 3 classes of kitten tissue (n = 6 each). In order to determine the blocking potency, (-)-bisoprolol was incubated at least 1 h before the next concentration-effect curve for an agonist was determined.

Membrane particles were prepared as published (Kaumann and Birnbaumer 1974; Kaumann 1978) from right and left ventricles and left atria of kitten heart and stored at - 80 ~ C until used.

Binding experiments. The membrane suspension was incubated at 32.5~ with the indicated concentration of 3H-

29

(-)-bisoprolol, 3H-(-)-propranolol or 3H-(-)-bupranolol in the incubation buffer I containing (retool/l): 25 Tris HC1, pH 7.4, 2 MgC12, I EGTA, 0.1 ascorbic acid and 0.2 mmol/1 GTP. In order to estimate the rate of binding of 3H-(-)- bisoprolol, 200 gl samples were removed from the prewarmed membrane suspension at the times indicated. For the equilibrium binding experiments with 3H-(-)-bisoprolol the binding reaction was conducted for 10 rain in a volume of 200 ~tl. The binding reaction with 3H-(-)- bisoprolol was stopped by dilution with 2 ml ice-cold incubation buffer I. The membrane particles were collected by vacuum on Whatman GF/A glass-fiber filters and washed 5 times with 2 ml ice-cold washing solution (10 mmol/1 Tris HC1, pH 7.6, 5 retool/1 MgC12). 8 ml of Aquassure (NEN) was added to the filters and radioactivity was measured by ligand scintillation counting. Counting efficiency was 43% for 3H-(-)-bisoprolol. (-)-Bisoprolol was also used to inhibit the binding of 3H-(-)-propranolol or 3H-(-)- bupranolol. The binding reaction of 3H-(-)-propranolol was allowed to proceed for 10 rain. 10 min was shown to be sufficient for equilibrium of 3H-(-)-propranolol with //- adrenoceptors of membrane particles (Kaumann 1978). The binding reaction with 3H-(-)-bupranolol was allowed to proceed for 30min, a time reported sufficiefat for equilibrium (Morris et al. 1981). In order to detect possible influences of cations on (-)-bisoprolol binding, some assays with 3H-(-)-propranolol (2.8-6.2 nmol/1) were carried out in incubation buffer II containing (mmol/1): 122.6 Na +, 5.1 K § 1.125 Ca 2+, 0.5 Mg 2+, 98.4 CI-, 29.6 HCO3, 1 HPO]-, 0.5 SO~-, 0.2 GTP, 0.04 EDTA.

For 3H-(-)-propranolol and 3H-(-)-bupranolol the filtering procedure was as with 3H-(--)-bisoprolol but the filters were treated with 0.5 ml Protosol (NEN) for 30 rain at 60 ~ C, then cooled on ice. 8 ml of Econofluor (NEN) (acidified with 50 gl glacial acetic acid) were then added and the radioactivity was measured by liquid scintillation counting. (Econofluor could not be used for counting of 3H- (-)-bisoprolol because the activity decayed within hours). Counting efficiency was 46% for both 3H-(--)-propranolol and 3H-(-)-bupranolol.

For equilibrium experiments with 3H-(-)-bisoprolol non-specific binding was defined as the binding observed in the presence of concentrations equivalent to. 100 equilibrium dissociation constants KL of the agonists (-)-adrenaline, (-)-noradrenaline, and the antagonists (-)-bupranolol (relatively lipophilic) and (-)-pindolol (relatively hy- drophilic). Under these conditions, non-specific binding was independent of the choice of unlabelled ligands. For equilibrium experiments with aH-(-)-propranolol and 3H- (-)-bupranolol, non-specific binding was defined as binding observed in the presence of 200 gmol/1 (-)-isoprenaline as well as 0.6 gmol/1 (-)-propranolol. Non-specific binding did not differ with (-)-isoprenaline and (-)-propranolol. Protein content was determined by the procedure of Lowry et al. (1951) with bovine serum albumin as standard. All binding is expressed in fmol/mg protein after correction for counting efficiency and recovery (65%) (Kau- mann 1982).

In order to test whether significant radioligand was trapped by the membranes, particles were incubated at 32.5~ for 10 rain with 2.9 nmol/1 3H-(-)-bisoprolol or for 30 rain with 1.3 nmol/1 aH-(-)-bupranolol in 1 ml of the described incubation medium. After a 10 rain centrifugation (10,000 x g) at 32.5~ we compared the radioactivity of the

supernatant of the tubes containing membranes with the radioactivity of tubes without membranes. We did not detect significant depletion (Morris and Kaumann 1979; Ehle et al. 1985) of radioactivity of either radioligand by the membranes using 0.25, 0.5 or 1 mg/ml protein.

Adenylate cyclase assays were performed with slight modifications as by Kaumann and Birnbaumer (1974). In- cubations were made in a final volume of 50 gl containing: 25 mmol/1 Tris HC1, pH 7.6; 20 mmol/1 creatine phosphate; 1 mmol/l EGTA; 0.1 mmol/1 ascorbate; I mmol/1 3H-cyclic AMP (10,000 cpm/assay tube); 0.1 mmol/1 e-32p-ATP ( 100 - 300 cpm/pmol); 0.01 mmol/1 GTP, plus the enzymes: 15 IU/ml creatine phosphokinase and 9.8 IU/ml myokinase, and the indicated concentrations of ligands. The reaction was begun with the addition of the membrane suspension and continued for 10 rain at 32.5~ The reaction was terminated by the addition of a solution containing 1% sodium dodecylsulfate, 40 mmol/1 ATP, and 10 mmol/1 cyclic AMP. The cyclic AMP was isolated by double chromatography (Salomon et al. 1974) and quantified by liquid scintillation counting.

Equations and statistics

Kitten heart tissues. The antagonism of the positive inotropic and chronotropic effects of catecholamines by (-)-bisoprolol was analyzed with a model for 2 classes of receptor subtypes (Kaumann and Marano 1982; Lemoine and Kaumann 1983). Data were calculated with help of Eq. (11) of Lemoine and Kaumann (1983). For the estimation of fractional stimuli 6~ and 6z, and equilibrium dissociation constants Km and KB2 for /~1- and/~2-adrenoceptors, log (CR-1) data were corrected for desensitization and analyzed with a computer program for non-linear regression as described in Lemoine and Kaumann (1983) and Walter et al. (1984). The concentration-ratio (CR) of catecholamine was taken between ECs0-values in the presence and absence of (-)-bisoprolol.

61- and 62-values of an agonist are interpreted as follows: Agonist-concentrations [A~] and [A2], mediating their effects exclusively via/31- and exclusively via/~2-adrenoceptors, respectively, are equieffective, if [A1]/[A2] = o'2/61.

Guinea pig right atria. Data were analyzed with the equation of Arunlakshana and Schild (1959).

Binding. A single class of non-interacting sites was assumed throughout for 3H-(-)-bisoprolol binding. Non-specific binding was measured and was found to be invariant with time. Dissociation rate constants, equilibrium dissociation constants and fractions of receptor subtypes were estimated by a non-linear regression analysis (Biomedical Biostatistic Program 3R, Dixon and Brown 1979) according to conventional equations as published by Morris et al. (1981).

Two classes of non-interacting binding sites were assumed for the binding of 3H-(-)-bupranolol and 3H- (-)-propranolol. The estimation of parameters was performed as detailed in Ehle et al. (1985).

Because ofheteroscedasticity all binding data were trans- formed according to Eq. (7) of Ehle et al. (1985) before estimating parameters.

For graphical representation (Fig. 5) the standard error of the specific binding (SEMs) is calculated as

SEMs = VSEMtot 2 + SEMns z , (1)

30

where SEMto t and SEMn~ are the SEM of total binding and non-specific binding, respectively. Only for graphical representation (Fig. 5) data for specific binding are linearized according to Scatchard (1949). For the standard error of the ratio of bound (B) over free (F) (i.e. SEMwF ) we find:

F + B SEMB/F = ~ - SEMB, (2)

where SEMB is the standard error of bound radioactivity.

Adenylate cyclase. The data of enzyme stimulation by catecholamines were fitted with a model of 2 components as described by the Eq. (4) of Gille et al. (1985).

Materials and drugs. (-)-Bisoprolol hemifumarate [ ( - ) - EMD 33512, see Fig. 1 for chemistry], 3H-(-)-bisoprolol (specific activity 1447 GBq/mmol) and (-)-noradrenaline L-tartrate were from Merck (Darmstadt, FRG); materials for the 3H-(-)-propranolol binding assay were those of Kaumann (1978); 3H-(-)-propranolol hydrochloride (specific activity 585 GBq/mmol) was from NEN; 3H-( - ) - bupranolol hydrochloride (specific activity 577 GBq/mmol) and bupranolol enantiomers were from Sanol (Monheim, FRG); ICI 118,551 hydrochloride and ( - ) -propranolol hydrochloride were from ICI (Macclesfield, UK); ( - ) - adrenaline bitartrate was from Serva (Heidelberg, FRG); (-)-isoprenaline bitartrate was from Sterling-Winthrop (Rensselaer, NY, USA); reserpine phosphate was from Ciba-Geigy (Basel, Switzerland); phenoxybenzamine HC1 was from Smith, Kline and French (Philadelphia, PA, USA). Other chemicals were obtained from local commercial sources. The materials for the binding and adenylate cyclase assays were as in Kaumann (1982).

For experiments on intact tissues stock solutions were made with deionized, redistilled water; stock solutions of catecholamines contained 0.04 retool/1 EDTA and were adjusted to pH 4 with HC1. Stocks for biochemical studies were prepared in a solution containing 2.5 mmol/1 EGTA and 0.25 retool/1 ascorbic acid.

Results

Tissue experiments

Depressant effects of ( -- )-bisoprolol. Figure I illustrates that (-)-bisoprolol does not exert stimulant effects. With concentrations up to 2 gmol/1 (-)-bisoprolol force and rate remain unaffected by (-)-bisoprolol. Concentrations greater than 2 gmol/1 depress force in left atria and papillary muscles. Concentrations greater than 20 gmol/1 depress sinoatrial beating rate. We estimated ICs0-values ( - l o g , tool/l) for the depressant effects of (-)-bisoprolol of 4.5 for the left atria, 4.3 for papillary muscles and 3.3 for right atria.

Agonist-dependent antagonism by ( - )-bisoprolol in kitten heart muscle

The effects of (-)-noradrenaline were antagonized by ( - ) - bisoprolol to similar extent in sinoatrial pacemakers, left atria and papillary muscles (Fig. 2 A). By contrast, the effects of (-)-adrenaline were antagonized more extensively in lch: atria and papillary muscles than in right atria (Fig. 2B). (-)-Bisoprolol antagonized the effects of (-)-adrenaline to

ua 80

OH CH tl" *~ 6 0 O_CH2_~H_CH2NH_CH / 3 t- ~+'~---- CH3 o_c , , , ~ 4 0

O,, ~ 20 . . . . ~1 - ' '~-

0 _ I 10 9 8 7 6 5 4 3 2

- LOG. [(-)-BISOPROLOL] mol/I

Fig. 1. Depressant effects of(-)-bisoprolol on isolated myocardium of kitten: spontaneously beating right atria (�9 n = 4), paced left atria (O, n = 6) and paced papillary muscles (&, n = 6). The inset shows the structural formula of bisoprolol

a similar extent in left atria and papillary muscles. These results hint at the existence of fl-adrenoceptor subtypes, at least in kitten sinoatrial pacemakers. For both catecholamines the blockade produced by (-)-bisoprolol was surmountable, and the curves in the presence of agonists were nearly parallel to the corresponding control curve. These characteristics are necessary prerequisites for the application of the model for 2 receptor-subtypes (Kaumann and Marano 1982; Lemoine and Kaumann 1983). Log (CR - 1 ) - d a t a are shown in Fig. 2C, estimated parameters are listed in Table 1.

In kitten sinoatrial node the affinity of (-)-bisoprolol was 6 0 - 8 0 times higher for/71- than for/32-adrenoceptors (Table 1). One sixth to one quarter of the chronotropic stimulus of (-)-adrenaline is caused by/3z-adrenoceptors; the remainder stimulus of (-)-adrenaline and most of the stimulus of (-)-noradrenaline are caused by sinoatrial/~l- adrenoceptors.

In left atria and papillary muscles a technique of paired tissues was used: left atria were divided into 2 halves and at least 2 papillary muscles were dissected out of each right ventricle. The agonist-dependence of the antagonism of positive inotropic effects by (-)-bisoprolol was considerably smaller than that of the chronotropic effects of the catecholamines, suggesting only a small participation of/32- adrenoceptors in the mediation of positive inotropic effects. Thus, estimates of the affinity of (-)-bisoprolol (pKB2) and of fractional stimuli (o-2) for /~2-adrenoceptors remain uncertain (Table IA) in left atria and papillary muscle. Contrarily, in the sinoatrial node, the estimates ofpKB2 and o-2 appear to be statistically reliable because of the marked agonist-dependence of blockade. Assuming that the affinity of (-)-bisoprolol does not differ for the/~2-adrenoceptors of the different heart regions we analyzed the data from blockade with a common non-linear regression (Table 1 B). The analysis revealed a 60-fold selectivity of (-)-bisoprolol for /~l-adrenoceptory. The inotropic stimulus of ( - ) - noradrenaline is essentially due to /~l-adrenoceptors while that of (-)-adrenaline mainly to/31-adrenoeeptors and to a small extent to/?2-adrenoceptors.

Agonist-dependent blockade by (-)-bisoprolol in the sino-atrial node of guinea pig

The positive chronotropic effects of (--)-noradrenaline were antagonized in a surmountable and simple competitive

31

= 180

/ / / / I / 1 ~C Dr Jl 13

)- 130 I U i-- 100 z= i j1M/j/--2 oo

~e I ~ J I ' ' 4 ~ l l ~ w (-)-Bisoprolol

,= 8 0 L ' I v - - ' l ' r - ~ l l I I I I

,90[ 1, , oo,,

I.~,,"JPJ/'J ,ooo ,"-?~176176 I i

.. ~176 / ~ ~ 50 50

< 10 10 I

>- 5 0 }/ t ~ 1 ~ [ ~ 2 0 6 0 n m ~ 5 0 -- ~ ' / / I / -- /---600 < ~ / / / / / - 2 0 0 0

r / z , l , i , <-~-Bisopro,ol lo 10 I I I i I , I

10 9 8 7 6 5 4 3.5 10

-LOG, [(-)-NORADRENALINE] mol/I

////Xo s o p r .

I I I I I I I

/ o/ootooto, , / / / , , , : - , . , ,sop, . . , , , ,/':-'i',., 9 8 7 6 5 4 3.5

-LOG. [(-)-ADRENALINE] mol/I

0

T i 0 '~ -1 L.

L..

-o.~ / 2 "

"i[ 0 I

-o.5 10 - L O G . [(o}-Bisoprolol] mol/ I

Fig. 2. Concentration-dependence of the antagonism by (-)-bisoprolol of the effects of (-)-noradrenaline (A) and (-)-adrenaline (B). Shown are results from spontaneously beating right atria, paced left atria and paced papillary muscles of 16 kittens (2 litters of each 4 in A and B). Open symbols of A and B represent concentration-effect curves for a catecholamine in the absence of (-)-bisoprolol (n = 8, except for the lower panel of B were n = 4); closed symbols are n = 4. To avoid overcrowding an additional set of 4 experiments with (-)-adrenaline is not shown on papillary muscles. The antagonism by (-)-bisoprolol in left atria and papillary muscles was always completely surmountable by the catecholamines. Panel C represents double log plots obtained with concentration-ratios (n = 4 for each symbol) of (-)-adrenaline ( 0 ) and (-)-noradrenaline (�9 calculated from the experiments of A and B. Curved heavy lines and thin asymptotes represent functions calculated with the estimated parameters of Table 1 B. Bars through symbols in A, B and C represent SEM

Table 1. Equilibrium dissociation constants (pKB = --log Km mol/1) of (-)-bisoprolol and agonist fractional stimuli a for/3-adrenoceptor subtypes ~1 and/?z from 3 different heart regions

Tissue Agonist (A) Individual fits (B) Common fits

pKBI__+ASD pKBz• a1 = (1-o-2) p K m _ A S D pK~2___ASD 0-1=(1-0-z) -F ASD + ASD

Right atrium (--)-noradrenaline 8.65 __ 0.06 6.74 _+ 0.18 0.99 -t- 0.01 0.97 • 0.01 (-)-adrenaline 0.85 +_ 0.04 0.76 _ 0.04

Left atrium (-)-noradrenaline 8.84 __ 0.07 7.55 _+ 0.32 0.99 _+ 0.02 8.82 ___ 0.03 7.05 _+ 0.13 1.00 _+_ 0.01 (-)-adrenaline 0.93 __. 0.05 0.97 _-t- 0.01

Papillary ( )-noradrenaline 8.87 _+ 0.11 7.33 +_ 1.03 1.00 -t- 0.01 1.00 muscle (-)-adrenaline 0.98 • 0.04 0.99 Jr 0.01

Data were calculated from the experiments of Fig. 2 ASD are asymptotic standard deviations (A): Data fitted for each heart region (B): Data fitted under the assumption that the affinity of (-)-bisoprolol for the 2 receptor subtypes is the same regardless of heart region The goodness of fits of (A) and (B) did not differ significantly as examined with a likelihood-ratio test

fashion by ( - ) - b i s o p r o l o l (Fig. 3, left middle panel). How- ever, the effects o f ( - ) - a d r e n a l i n e were antagonized less by ( - ) - b i s o p r o l o l than the effects o f ( - ) - n o r a d r e n a l i n e . Fur thermore , unlike the parallel shifts of concentrat ion- effect curves of ( - ) - n o r a d r e n a l i n e the curves for ( - ) - adrenaline became flatter with increasing concentrat ions o f ( - ) - b i s o p r o l o l (Fig. 3, r ight upper panel). These results suggest that par t o f the effect induced by ( - ) - a d r e n a l i n e is resistant to b lockade by ( - ) - b i s o p r o l o l , and presumably

media ted through /?2-adrenoceptors. This hypothesis received suppor t by experiments, in which/?z-adrenoceptors were occupied by ICI 118,551 to the same degree as /?l- adrenoceptors were occupied by ( - ) - b i s o p r o l o l . Only concentrat ions of ICI 118,551 were used which were lower than the equil ibr ium dissociation constant Km for /~l- adrenoceptors . These concentrat ions do not cause antagonism o f / ~ - a d r e n o c e p t o r mediated effects by ( - ) - a d r e n a l i n e (Lemoine et al. 1985). In the presence of ICI 118,551

3 2

' - r l=5 E 26~ r

o~ 180 r o1/I

140 a'- i i ~ , i i i N , ~ O ~ S 3H'(-)'BISOPROLOL

D 0 LU

'~ 210 4

17o I ~j .=, ~4oo

r t I I - I I ' 10 9 8 7 6 5

-LOG. [(-)-NORADRENALINE]

/ e __. slope 1.06 --I M . . . . . 31"I:BIS

L ~ " �9 �9 , , I O L I o 9 8 7 6 -LOG [(-)-BISOPROLOL] mol/I

I 1

4 3 mol/I

_ ,oo r o~ n=12 :4 4 :4

~) BOO

Ok_,~r ,,r~ .I i I , I nl ee

[ o 4 & 0.4 d ! i

- l ~ /e ff . .~" C--)-Bisoprolol ~: I . . . / . . * C , / . , , / . . ,6 . ~, 19 , ,8..ss~

O ~ , b ' ~ = I ~ I i I

9 8 7 6 5 4 3 -LOG. [(-)-ADRENALINE] mol/I

. ~,o.. too /

/ ...I /

~ - I I I I I 10 9 8 7 6 -LOG [(-)-BISOPROLQLIc~ 118,5.51] mol/I

Fig. 3. Competitive antagonism of positive chronotropic effects of (-)-noradrenaline and (-)-adrenaline by (-)-bisoprolol (BIS) on spontaneously beating right atria of guinea pig. Three or four successive concentration-effect curves for (-)-noradrenaline or (-)-adrenaline were determined, the first in the absence, the following in the presence of the indicated concentrations of 3H-(-)-bisoprolol (3H-BIS) or (-)-bisoprolol. The concentrations of 3H-BIS were verified by scintillation counting. On one group of atria successive concentration-effect curves of (-)-adrenaline were determined in the absence or presence of the indicated combinations of (-)-bisoprolol and ICI 118,551 (right middle panel). The data with (-)-adrenaline are normalized because three groups (right upper panel) and two groups (right middle panel) of four atria were pooled; absolute maximum effects in the presence of antagonists were not significantly different to those in their absence. The lowest panels show the related Schild-plots; CR are concentration-ratios. The left-hand Schild-plot also shows data from additional experiments with 3H-BIS (open symbols) and BIS (closed symbols). The right-hand Schild-plot was derived from the experiment with ( - ) - bisoprolol + IC[ 118,551 (middle right panel)

concentration-effect curves of (-)-adrenaline were shifted in parallel fashion by (-)-bisoprolol and the blocking potency of( - ) -bisoprolol (pKBt = 8.67 + 0.06) enhanced to a similar degree as against the effects of (-)-noradrenaline (pKm = 8.73 + 0.04) in the absence ofICI 118,551 (compare the similarity of Schild-plots, bottom panels of Fig. 3).

We also used guinea-pig right atria in order to compare the blocking potencies of tritiated and unlabeled ( - ) -b i - soprolol against/~-selective (-)-noradrenaline (Fig. 3, left hand side). In this system virtually all positive chronotropic effects of (-)-noradrenaline appear to be mediated via ]~:- adrenoceptors (Lemoine et al. 1985). Neither 3H-(-)-bisoprolol nor unlabeled 3H-(-)-bisoprolol caused depres- sion of the sinoatrial rate in guinea pigs. Both ligands antagonized equally well the positive chronotropic effects of (-)-noradrenaline. Estimated pKm-values were 8.69 +_ 0.04 for 3H-(-)-bisoprolol.

Membrane experiments

Binding of 3H-(--)-bisoprolol. Binding data are summarized in Table 2A. Binding of 3H-(-)-bisoprolol to /~l-adrenoceptors was extremely fast (Fig. 4 top), being essentially complete by I rain and precluding further analysis. Dissocia- tion of 3H-(-)-bisoprolol from ~l-adrenoceptors was complete in 5 rnin (Fig. 4 bottom).

Binding of 3H-(-)-bisoprolol to ~l-adrenoceptors was saturable (Fig. 5) and stereoselective (Fig. 6) as shown with the optical isomers of bupranolol. ( -)-Noradrenaline and

(-)-adrenaline competed with relatively low affinity with 3H-(-)-bisoprolol for ]~-adrenoceptors (Fig. 6). All equilibrium binding experiments suggest the existence of a single class of non-interacting binding sites (i.e./3t-adrenoceptors). As expected, the affinity of 3H-(-)-bisoprolol estimated from saturation binding matched the affinity of unlabeled (-)-bisoprolol estimated from binding competition (Fig. 6, Table 2A).

Affinities of ( - )-bisoprolol and ( - )-propranolol for ~t-adrenoceptor subtypes labelled with 3 H- (-)-propranolol

In order to estimate the affinity of (-)-bisoprolol for/~2- adrenoceptors, as well as for/3~-adrenoceptors, binding inhibition experiments were performed with 3H-(-)-propranolol. Representative experiments are shown in Fig. 7. (-)-Bisoprolol competed with 3H-(-)-propranolol for 2 ventricular and atrial binding sites with 2 orders of magnitude difference in affinity (Table 2C). Neither the absence of GTP in the incubation medium I nor different ions (incubation medium II) modified the affinity of ( - ) - bisoprolol for the 2 sites or the fractions of high- and low- affinity binding-sites. The affinity of (-)-bisoprolol for the 2 sites was also not altered by increasing the time of incubation to 18 rain (data not shown). Two binding sites for ( -)-bisoprolol with similar characteristics were found in left ventricle, right ventricle and left atrium (Table 2). One sixth to 1/4 of the receptors were/3z, the remainder/~1. Unlabeled (- ) -propranolol competed with 3H-(-)-propranolol for

Table 2. Binding constants of ligands for fl-adrenoceptors of kitten myocardium

Competing Experimental n pKLi + ASD f l + ASD pKe2 4- ASD Incubation Membrane ligand condition buffer preparation

33

(A) Equilibrium dissociation constants pKL (--log, tool/l)

- - saturation binding 1 x 6 (--)-Bisoprolol binding inhibition 2 x 6 (-)-Bupranolol binding inhibition 1 x 6 (+)-Bupranolol binding inhibition l x 6 (-)-Noradrenaline binding inhibition 2 x 6 (-)-Adrenaline binding inhibition 3 x 6

for fl~-adrenoceptors radiolabelled by 3H-(-)-bisoprolol:

8.15 + 0.06 - - I left ventricle 8.27 _+ 0.10 1.00 - I left ventricle 9.15 4- 0.08 1.00 - I left ventricle 7.58 4- 0.07 1.00 - I left ventricle 6.32 4- 0.07 1.00 - I left ventricle 6.05 4- 0.05 1.00 - I left ventricle

(B) Equilibrium dissociation constants pKL (--log, tool/l) of (-)-bisoprolol for fit- and /32-adrenoceptors radiolabelled by 3H-(-)- bupranolot:

(-)-Bisoprolol binding inhibition 1 x 5 8.20 4- 0.1 i 0.67 4- 0.04 6,62 4- 0.19 I left ventricle { binding inhibition }

(-)-Bisoprolol + 66 nmol/1 1 x 5 8.16 4- 0.06 - - I left ventricle ICI 118.551

(C) Equilibrium dissociation constants pKL (--tog, tool/l) for fl~- and fl2-adrenoceptors radiolabelled by 3H-(-)-propranolol:

(-)-Bisoprolol binding inhibition t x 4 7.96 ~0.1t 0.84 _+ 0.04 6.12 4- 0.33 I without GTP right ventricle (-)-Bisoprolol binding inhibition 1 x 4 8.20 4- 0.11 0.82 4- 0.03 6.25 4- 0.24 I right ventricle (-)-Bisoprolol binding inhibition 1 x 4 8.00 _+ 0.08 0.86 4- 0.04 6.61 _+ 0.30 II left ventricle (-)-Bisoprolol binding inhibition 1 x 4 8.13 4- 0.13 0.75 4- 0.05 5.91 4- 0.29 I left ventricle (-)-Bisoprolol binding inhibition 2 x 4 7.95 4- 0.10 0.79 4- 0.05 6.45 4- 0.27 I left atrium (-)-Propranolol binding inhibition t x 4 8.72 4- 0.05 1.00 - I left ventricle

(B) pKL:- and pKL2-values for (-)-bisoprolol were calculated with pKL,-values for aH-(--)-bupranolol taken from saturation binding experiments on kitten heart (PKL*I --- 9.23 + 0.06; Morris et al. 1981) and on guinea-pig lung (pKL*2 = 9.61 + 0.02; Lemoine et al. 1985)

(C) pKLI- and pKL2-values for unlabelled ligands were calculated with pKL,-values for 3H-(-)-propranolol taken from saturation binding experiments on kitten heart (PKL*~ = 8.70 4- 0.04) and on calf lung (pKe*2 = 9.12 _-!- 0.04)

n number of experiments times sixtuple, quintuple or quadruple determinations Incubation buffers are defined under Methods f : is the fraction of/3:-adrenoceptors

ASD is the asymptotic standard deviation of non-linear regression analysis

ventricular /%adrenoceptors with a nearly 4-fold higher affinity than ( - ) -b i sop ro lo l for flx-adrenoceptors (Fig. 7, lower panel).

An alternative procedure of selective fll-adrenoceptor labelling

Conceivably, it should also be possible to radiolabel selectively one subtype of a heterogeneous fl-adrenoceptor popu- Iation using a non-selective radioligand in the presence of a saturating concentrat ion of an unlabelled subtype-selective Iigand. In Fig. 8 and Table 2B we compare inhibition by ( - ) -b i sop ro lo l o f binding o f little selective a H - ( - ) - b u p r a - nolol in the absence and presence of ICI 118,551. Because of the 300-fold fi2-selectivity of ICI 118,551 we calculated that the used 66nmol/1 will saturate 90% of the fiz-adrenoceptors while competing with less than 9% of the fi :-adrenoceptors. In the absence of ICI 118,551 ( - ) - b i s o p r o l o l competed for a high-affinity site and for a low-affinity site with pKL-values of 8.2 and 6.6, respectively. In the presence of ICI 118,551 only the high-affinity site (fl:) for ( - ) - b i s o p - rolol remained with a pKL-value of 8.2. As expected, the fraction of f l , -adrenoceptors was identical in the absence and presence of ICI I 18,551.

( - )-B&oprolol as antagonist of catecholamine-indueed stimulation of the adenylate cyclase

fi-Adrenoceptor coupled adenylate cyclase was examined to obtain an independent estimate of the affinity of ( - ) - bisoprolol for fit- and flz-adrenoceptors of ventricular

membranes. The stimulation of adenylate cyclase by ( - ) - adrenaline and ( - ) -no rad rena l ine was antagonized by 0.1, I and 10 gmol/1 ( - ) -b i sop ro lo l (Fig. 9). The concentration- effect curve o f ( - ) - ad rena l ine and ( - ) -no rad rena l ine in the absence of antagonist was monophasic and biphasic, respectively. A small fraction of the ( - ) -adrenal ine- induced stimulation of adenylate cyclase, observed in the lower par t of the concentration-effect curve, was antagonized with low potency by 10 gmol/1 ( - ) -b i sopro lo l . This effect is expected when the agonist is nonselective and the antagonist is selective for fl :-adrenoceptors. For ( - ) -noradrena l ine , the small low potency fraction, occurring in the absence of ( - ) - b i sop ro lo l in the upper par t o f the concentration-effect curve, disappeared in the presence of ( - ) - b i - soprolol. This effect is expected when the fl~-selectivity of the agonist and the antagonist is of similar magnitude. These experimental findings were accounted for sufficiently by a model of 2 receptors subtypes interacting with agonists and antagonists, The ECso-values and pKB-values were calculated by a common non-linear regression analysis of the concentration-effect curves for both agonists in the presence and absence of the antagonists, as detailed in Eq. (4) of the accompanying paper of Gille et al. (1985). For fl~- and flz-adrenoceptors the estimated ECso'S ( - l o g , tool/l) were, respectively, 6.66 + 0.03 and 6.47 _+ 0.08 for ( - ) - ad rena l ine and 6.94 4- 0.03 and 4.75 +_ 0.14 fo r ( - ) -noradrena l ine . The estimated pKB-values for ( - ) - b i sop ro lo l were 7.97 + 0.03 and 5.94 __+ 0.14 for/?a- and fl2-adrenoceptors, respectively. ( - ) -Adrena l ine and ( - ) -no rad rena l ine mediated 78 __+ 1%

34

4C B[fmol]

L--~TJ 3O

20

10

0 0

32 I 30 B fmol

++l+l,

I I 5 10

I I 20 30

[min] T I M E Fig. 4, Association and dissociation of 3H-(-)-bisoprolol with and from left ventricular/~l-adrenoceptors of kitten. The rate of association (top panel) was assessed by administrating 5.5 nmol/1 3H- (-)-bisoprolol to 560 mg/l protein of membranes prewarmed for 10 min at 32.5~ At the indicated times 200 gl aliquots were pipetted directly onto the Whatman GF/A filters under vacuum and immediately washed with ice-cold washing buffer as described under Methods. The rate of dissociation (bottom panel) was estimated by adding (-)-isoprenaline (final concentration 0.2 retool/l) to a membrane suspension which had previously been incubated for 11 rain (equals 0 rain of dissociation) with 3H(-)-bisoprolol. Specific binding (B) was 33 _+ 2 fiuol/mg and non-specific binding 41 _+ 1 fmol/mg protein 07+ASD). Each points represents g _+ SEM of quadruplicates. Curves represent functions calculated with the estimated parameters

and 86 _+ 1% of their effects, respectively, through fll- adrenoceptors.

30 and 300nmol/1 of each 3H-(=)-bisoprolol and unlabelled (-)-bisoprolol antagonized to the same extent the (-)-noradrenaline-induced stimulation of the adenylate cyclase (not shown).

Discussion Selectivity of ( - )-bisoprolol for myocardial fll-adrenoceptors The results from both antagonism of catecholamine-induced effects in intact tissues and binding on membrane particles are consistent with the existence of a high-affinity and low- affinity fi-adrenoceptor for (-)-bisoprolol. The difference in affinity between the 2/~-adrenoceptors is 1.7 + 0.3 log units, regardless of tissue, radioligand or technique. 3H- (-)-bisoprolol quickly associated to and dissociated from myocardial/3-adrenoceptors. As expected, unlabeUed ( - ) - bisoprolol competed with 3H-(-)-bisoprolol for the high- affinity/~-adrenoceptor with an affinity nearly identical to that of 3H-(-)-bisoprolol. Furthermore, (-)-bisoprolol also competed for binding with 3H-(-)-propranolol and 3H-( - )-bupranolol for a high-affinity fl-adrenoceptor with

frnol

90

6c ~0 3C

I l I I 0 10 2 0 3 0 33

6 0

4 0 lo

, ~ " FB-- [ ; ~ pKL*. 8"15

0 30 [ B ~fm=~ 61 ol i i [ . ]

0 10 20 3 0 33 [3H- ( - ) -B ISOPROLOL] nrnol/I

Fig. 5. Saturation binding of aH-(-)-bisoprolol to left ventricular fll-adrenoceptors of kitten. Upper panel: The binding of the indicated free concentrations (F) of 3H-( -)-bisoprolol was evaluated in the absence (total binding, 0) and presence of 50 nmol/l ( - ) - bupranolol (non-specific binding, rq). Lower panel: The experimental errors (SEM) of the specific binding (B) and of B/F ratios were calculated according to Eqs. (1) and (2), respectively. Vertical bars specific binding correspond to horizontal bars of the Scatchard-plot. Total, specific and non-specific binding and the Scatchard-plot (inset) are represented by functions calculated with the estimated parameters. Each point represents 2 + SEM of sextuplicate determinations

nearly the same affinity as against 3H-(-)-bisoprolol, regardless of tissue.

The high-affinity/%adrenoceptor for 3H-(-)-bisoprolol was stereoselective as demonstrated with the stereoisomers of bupranolol. The difference in affinity of 1.6 log units between (-)-bupranolol and (+)-bupranolol is similar to differences in affinity reported for these stereoisomers in a variety of myocardial systems with fil-adrenoceptors (Kaumann and Birnbaumer 1974; W/ichter et al. 1980; Morris et al. 1981; Kaumann et al. 1982). This agreement with published data supports the assumption that the/~l-adrenoceptors labelled with high affinity by 3H-( - ) - bisoprolol are the same as those labelled by the well characterized /3-blockers 3H-(-)-propranolol and aH-(-)- bupranolol.

Recent evidence supports the view that the steric properties of/~1- and ~2-adrenoceptors are different (Lemoine and Kaumann 1983; Morris and Kaumann 1984). Thus, while the affinity-ratio of stereoisomers of bupranolol is of 1.6-- 1.8 log units for /~1, it is only of 1 .2-1 .4 log units for /~z- The found affinity-ratio of 1.6 log units between ( - ) - bupranolot and (+)-bupranolol for fl-adrenoceptors

35

~ 0 ~ ~ , ~ RIGHT VENTRICLE

= / (-,-gupranolol..~, X"~.. .~j (+,-, u p r a n o Io I i30 ~ (-)-Bisoprolo, "6 50 n=4

i / ' ~ ~ I L ~ ~ rO.O. (3._ 0 100 ' ' ' 10 9 8 7 6 5 4 "T" IooL I I I I' T~'dl'-.a-~"'~J..l~'~--I ~- 0 " ~ 9 ~ L

11 10 9 8 ' 7 ~" '6 5 4 ~.e EFT ATRIUM

0 c ~ ~ (-,-Bisoprolol " n : 8

~ / (-)-Noradrenaline k l~(-)-Adrenaline "~ Q..

" 10 9 8 7 6 5 4

",~ 0 " ~ L E F T VENTRICLE �9 P,

1no/~ --,_ I I I I I I r ~ : : ~ - + 11 10 9 8 7 6 5' ~'4 "~ \ 't~ (-)- Bisoprolol

- L O G . [ D R U G ] mol / I ~ 5(2 (-)-Propranolo~ ~ n:4

, + , . +,,itio, of s , , + ,-,,isop o,ol b,nd,,l to left ve,,- tricular membranes of kitten. Binding of 8-11 nmol/1 3H-(-)- 10s I bisoprolol was evaluated in the absence and presence of the indicated concentrations of competing ligands. Non:specific binding (48 - 56%) was independent of the used ligands. Each point represents ff + SEM of sextuple determinations. Curves represent functions, for a single class of drug-receptor complex, calculated with the estimated parameters

labelled with aH-(-)-bisoprolol further supports its ill- nature.

From the agonist-dependent antagonism of the positive chronotropic and inotropic effects of catecholamines we uncovered a high-affinity and a low-affinity fl-adrenoceptor for (-)-bisoprolol. The difference in affinity between the 2 receptors was 1.8 log units. The difference of affinity for 2 subtypes estimated from blockade agrees with the average difference of affinity of 1.7 log units for 2 sites estimated from binding. We therefore postulate that the 2 sites are/Yl- and fl/-adrenoceptors. The low-affinity binding site for (-)-bisoprolol is not labelled by 3H-(-)-bupranolol when a f12 -saturating concentration of ICI 118,551 is present. The low-affinity fl-adrenoceptor for (-)-bisoprolol appears therefore to be f12. In all 3 heart regions (i.e., kitten right and left ventricle and left atrium) the fl2-adrenoceptor comprises 1/6 to I/3 of the total fl-adrenoceptor population, the remainder being//1 in character. Inspection of the curves of Figs. 7 and 8 would suggest that the fractions of fla- adrenoceptors are larger than actually calculated. However, this impression is due to the slight fl2-selectivity of the radioligands 3H-(-)-propranolol and 3H-(-)-bupranolol.

Loss o f accessory binding sites on the f l l -adrenoceptor f o r ( - )-bisoprolol but no t f o r ( - ) -propranolol and ( - ) -bupranolol in membrane part icles

Both the data from isolated tissues and that from membrane particles indicate the existence of 2 fl-adrenoceptor subtypes recognized with differential affinity by (-)-bisoprolol. We detected, however, a systematic, quantitative discrepancy: the affinities of (-)-bisoprolol are 3 to 8 times greater for

10 9 8 7 6 5 4 - L o g . [ L i g a n d ] m o l / I

Fig. 7. Inhibition by (-)-bisoprolol of specific 3H-(-)-propranolol binding to membranes derived from right ventricle, left atrium and left ventricle of kitten. Comparison of binding inhibition by ( - ) - bisoprolol (0) and (-)-propranolol (�9 on left ventricular membranes. Concentrations of 3H-(-)-propranolol were 3.2 nmol/1 in right ventricle, 4.3 nmol/1 in left atrium, and 4.1 (�9 or 4.5 (0) nmol/1 in left ventricle. Initial binding obtained with these concentrations of aH-(-)-propranolol was 95 + 4 fmol/mg in right ventricle, 55 + 2 fmol/1 in left atrium, and 61 + 4 (�9 or 63 _-t- 3 (O) fmol/mg in left ventricle. Non-specific binding observed in the presence of 0.2 mmol/1 (-)-isoprenaline was 28 __+ 3% in right ventricle, 44 + 2% in left atrium, and 30 +__ 3% (�9 or 29 _+_ 3% (0) in left ventricle. Binding inhibition curves of (-)-bisoprolol and (-)-propranolol were fitted for a model with two or one receptor subtypes, respectively, Curves represent functions calculated with the estimated parameters

the fll-adrenoceptor in intact tissues than in membrane particles. The discrepancy can not be accounted for by the choice of radioligand, because it was observed with 3H- ( - ) -propranolol and 3H-(-)-bupranolol as well as with 3H- (-)-bisoprolol. By contrast, we do not observe discrepancies between affinity in intact tissues and membranes for ( - ) - propranolol (Kaumann and Birnbaumer 1974; Kaumann et al. 1980) and (-)-bupranolol (Morris et al. 1981), two ligands with only a slightly higher affinity for f12- than for fll- adrenoceptors (Morris and Kaumann 1984; Lemoine and Kaumann 1983; Lemoine et al. 1985).

The discrepancy is unrelated to a possible radioisotope effect: 3H-(-)-bisoprolol and unlabelled (-)-bisoprolol have identical affinities as antagonists of (-)-noradrenaline- induced effects in guinea-pig sinoatrial pacemaker and in kitten ventricular adenylate cyclase.

(-)-Bisoprolol could perhaps bind to accessory non- specific sites with a high capacity (resulting in an effective reduction of free ligand concentration) and with an extremely fast dissociation kinetic. As a result (-)-bisoprolol

36

6C

fmol

4C

20 c 1 1 8 5

I I I I ~ ' ~ - ' ~ , . . - _ _ I 10 9 8 7 6 5 ,v 4

- L O G . [ ( - ) -B isopro lo l ] mol/ I

Fig. 8. Inhibition of specific 3H-(-)-bupranolol binding in the absence and presence of a concentration ofIC1118,551 which saturates the /32-adrenoceptors. Left ventricular membranes of kitten. The binding of 2 nmol/13H-(-)-bupranolol to/~1- and/~2-adrenoceptors (without ICI 118,551; 68 +_ 2 fmol/mg) and to /~l-adrenoceptors (with ICI 1t8,551; 54 + 1 fmol/mg) was determined in the absence and presence of the indicated concentrations of (-)-bisoprolol. Non-specific binding (24_+1 fmol/mg) was estimated with 0.2 retool/1 (-)-isoprenaline and with 0.6 gmol/1 (-)-propranolol; no significant difference was found between the two ligands. Each point represents )~ + SEM of quintuplicate determinations. Binding inhibition curves in the absence and presence of ICI 118,551 were fitted for a model with two or one receptor-subtypes, respectively. Curves represent functions calculated with the estimated parameters

'0 f 0.5 �9

9 8 7 6 5 4 3 2 -LOG. [(-)-ADRENALINE] mol l l

f i "5 0 5

8

0 I J L I 9 - - g - - 7 " 6 5 4 3 2

-LOG. [(-)-NORADRENALINE] mol//I

Fig. 9. Antagonism by (--)-bisoprolol of adenylate cyclase stimulation with (-)-adrenaline and (-)-noradrenaline. Membranes were derived from left ventricular myocardium of kittens. Each point represents a single determination with 65.9 -+ 0.8 gg protein per tube. Results are expressed as a fraction of maximum stimulation with 2mmol/1 (-)-isoprenaline. In the experiment with ( - ) - adrenaline, basal and maximum activities of triplicates (pmoles min -1 rag-l) were 41.5 + 0.3 and 136.4 __ 0.8 (�9 35.0 _+ 0.3 and 132.1 _+ 0.9 (O), 34.8 _ 0.3 and 135.0 +_ 0.9 ( I ) , 34.6 +- 0.3 and 131.2 _ 1.0 (A). In the experiment with (--)-noradrenaline, basal and maximum activities were 45.2_+0.5 and 139.4+_0.8 (�9 37.4 +_ 0.3 and 136.7 +_ 0.8 (O), 36.2 + 0.1 and 133.8 + 1.0 (m), 35.7-t-_ 0.1 and 133.5 +- 1.0 (A). Curves were calculated from the estimated parameters obtained with a common fit for a model of two receptor-subtypes

would dissociate completely during the separation (filtering procedure) of free and bound radioligand ( 5 - 1 0 s). This hypothetical process would result in an underestimation of the affinity of 3H-(-)-bisoprolol for /?-adrenoceptors in saturation binding experiments. However, we reject this mechanism because the separation of free fi'om bound radioligand by centrifugation revealed that the used protein concentrations (0.25, 0.5 and 1.0 mg/ml) did not reduce measurably the concentration of free radioligand.

Other factors that could explain the discrepancy were also ruled.out: neither GTP nor the presence of physiological cations and anions changed significantly the binding characteristics.

Conceivably the discrepancy could be due to lack of equilibrium of ( - ) -bisoprolol with/~-adrenoceptors. How- ever, the binding kinetics of 3H-(-)-bisoprolol are fast and the affinity of ( - ) -bisoprolol as inhibitor of 3H-( - ) -p ro- pranolol binding was the same with 10 rain as with 18 min incubation. Thus, the binding affinity to membrane particles does not appear underestimated because of lack of equilibrium. Alternatively, the affinity from blockade of catecholamine-induced effects may be overestimated in intact tissues because at the time the catecholamines surmount blockade, ( - ) -b isoprolol may continue to dissociate significantly from the/3-adrenoceptors. However, this latter mechanism can also be ruled out because the catecholamine effects in the presence of ( - ) -bisoprolol were essentially at equilibrium within 2 - 5 rain and did not increase thereafter over an additional 12 min observation period.

A possibility that had to be disposed of was that the discrepancy might be due to different distributions within membrane lipids of 3H-(-)-bisoprolol for membrane particles and membranes in intact tissues. Access of the ligand to the receptors may depend on its lipophilicity. Evidence based on partition coefficients in octanolol/water shows that the rank order of lipophilicity is propranolol > bupranolol > bisoprolol > pindolol (Hellenbrecht et al. 1973; Cruickshank 1980; Nowack, personal communication). Be- cause only ( - ) -bisoprolol but not the other 3 ligands (this paper; Morris et al. 1981; Walter et al. 1984) show the discrepancy between blocking potency and binding affinity, membrane partition can not account for the discrepancy.

By exclusion we conclude that some structure of the/3- adrenoceptor is lesioned during the procedure of membrane preparation, as has been previously suggested (Kaumann et al. 1980). Interestingly, this lesion seems specific for the binding of subtype-selective ligands, i.e. ( - ) -bisoprolol to /~l-adrenoceptors, because binding to membrane particles of ligands with small subtype-selectivity such as ( - ) - p r o - pranolol or ( - ) -bupranolo l does not show a decrease in affinity (vide supra). It is remarkable that the high affinity of ( - ) -propranolo l and ( - ) -bupranolo l observed in intact tissues is preserved in membrane particles even when ( - ) - bupranolol competes with 3H-(-)-bisoprolol for binding. This suggests that the accessory binding site for ( - ) - b i - soprolol, which was probably lost during the procedure of membrane preparation, is not necessary to contribute to the binding of ( - ) -bupranolol .

We conclude that the interpretation of affinity estimates for/~l-selective ligands on membrane particles requires cau- tion. We recommend that an independent estimate of affinity for the/?~-selective ligands be obtained from intact tissues. Thus, on intact tissues the blocking potencies of ( - ) - p r o - pranolol and ( - ) -b isoprolol are the same but on membrane

37

particles the affinity of (-)-bisoprolol is on average 4 times lower than the affinity of (-)-propranolol.

Low potency of ( - )-bisoprolol as antagonist of catecholamine-induced stimulation of adenylate cyclase

If a decrease in affinity to (-)-bisoprolol is due to the loss of some fi~-adrenoceptor component during the preparation of membrane particles, the blocking potency of (-)-bisoprolol against catecholamine-induced cyclase stimulation in membranes should equally be decreased in these membranes. As expected, the blocking potency of (-)-bisoprolol was lower in membrane particles than in intact tissues. (-)-Bisoprolol antagonizes the catecholamine-induced stimulation of the cyclase predominantly through fla- adrenoceptors but also to some extent through fl2- adrenoceptors. The affinity of (-)-bisoprolol estimated for fl~- and flz-adrenoceptors coupled to the adenylate cyclase agreed with the corresponding affinity estimated from membrane binding.

With one exception, good agreement between affinity estimates was found for 12 ligands (regardless of the degree of lipophilicity) in intact tissues and membrane particles (adenylate cyclase) in kitten myocardium (Birnbaumer et al. 1977). The exception is another ill-selective ligand, practolol, which exhibit a 4-fold lower affinity for/?-adrenoceptors in membrane particles than for those of intact tissues (see also Kaumann et al. 1980). The result with practolol is expected if, as for (-)-bisoprolol an accessory binding site for/~-selective ligands were lost during preparation of the membrane particles.

Positive inotropic effects of ( - )-noradrenaline and ( - )-adrenaline appear mediated mostly through spare fil-adrenoeeptors

We detected low affinities of both (-)-noradrenaline and (-)-adrenaline for ventricular fil-adrenoceptors from inhibition of ZH-(-)-bisoprolol binding. This is in contrast to the high inotropic potencies of the catecholamines and supports the assumption of a large fla-adrenoceptor reserve.

The inotropic potency of (-)-noradrenaline is 7-times greater than the potency of (-)-adrenaline on both papillary muscles and left atria (Fig. 2). The 2-fold higher binding affinity of (-)-noradrenaline over the affinity of ( - ) - adrenaline (Table 2) for ventricular fll-adrenoceptors only partially accounts for the 7-fold difference in inotropic potency. The resultant 3.5-fold difference is conceivably due to a weaker conformational change of the fix-adrenoceptor with (-)-adrenaline than with (-)-noradrenaline.

Inotropic concentration-effect curves are steeper in papillary muscle than in left atria regardless of the used catecholamines. The independence of the slope of the curves from the used cateeholamine makes it unlikely that steep and flat curves are attributable to different affinity characteristics or conformational changes of fl-adrenoceptors in atria and papillary muscles. A likely explanation of the difference in slopes is to postulate differences between atrium and ventricle in the physiological function between fl-adrenoceptor occupancy and inotropic effects. Further- more, the concentration-effect curves of both ( - ) - adrenaline and (-)-noradrenaline for the stimulation of the ventricular adenylate eyclase are also flatter than the corresponding curves for the inotropic effects in papillary

muscle. Thus, in papillary muscle, we conclude by exclusion that some sort of cooperativity takes place in processes which are induced by the activated receptor-cyclase complex.

In order to provide fractional stimuli for/~1- and /~2- adrenoceptors we must assume the existence of spare receptors for both subtypes (Kaumann and Marano 1982; Lemoine and Kaumann 1983). The 50- to 100-fold differences between high inotropic potencies (EC5o'S) and binding affinities (estimated in the presence of 0.2 retool/1 GTP) of the catecholamines supports the assumption of spare receptors.

It is well known for a variety of receptor systems in- cluding/~-adrenoceptors (Lefkowitz et al. 1983) that GTP changes a high-affinity form into a low-affinity form of the receptor. Under our conditions omission of GTP causes flattening of the binding inhibition curves with a 2- to 4-fold decrease of the overall ICs0 (unpublished experiments). Thus, large differences between inotropic potency and binding affinity for catecholamines remain, even in the absence of exogeneous GTP. Thus, the spare receptor hypothesis appears not invalidated by the lack of exogeneous GTP.

The estimation of a large/?-adrenoceptor reserve for a catecholamine from its potency in an intact tissue and from binding affinity on its cell free membranes may be artifactual. To overcome a possible artifact both binding affinity and physiological potency should be estimated in intact cells. However, even in intact beating cardiocytes a large chronotropic/%adrenoceptor reserve has been found (Kaumann 1982) and the binding affinity for a catecholamine is nearly independent on binding to intact cardiocytes (Porzig et al. 1982) or to a homogenate of such cells (Kaumann 1982). Thus, it is likely that the large inotropic /~l-adrenoceptor reserve found by us is also to be expected for intact cells.

Although we found that 16-18 % of the fight ventrieular //-adrenoceptor population is /~2, we estimated that they contribute less than 2% and 1% to the inotropic stimulus induced by (-)-adrenaline and (-)-noradrenaline, respectively. Similarly, in left atria 21% of the/~-adrenoceptors are /~2, yet this subtype only contributes 8% and 2% to the inotropic stimulus induced by (-)-adrenaline and ( - ) - noradrenaline, respectively. Clearly, there is a discrepancy between the relative amount of/~2-adrenoceptors detected and their relative contribution to the positive inotropic effects elicited by (-)-noradrenaline and (-)-adrenaline. The ultimate reason of this discrepancy is not yet known.

One possibility is that cAMP produced by/~l-stimulation is utilized more efficiently for inotropic effects than cAMP produced by/~2-stimulation. Evidence by Kaumann et al. (1983), using the flz-selective agonist procaterol, supports this hypothesis. Procaterol stimulates the ventricular adenylate cyclase mainly through /~2-adrenoceptors but causes positive inotropic effects mainly through//a-adrenoceptors in kitten papillary muscle.

Positive chronotropic effects of ( - )-adrenaline occur through both fla- and ~2-adrenoceptors, those of ( - )-noradrenaline mostly through fll-adrenoceptors

Because the sinoatrial node is not suitable for direct binding studies (Kaumann 1982), information on the physiological role of fl-adrenoceptor subtypes can only be obtained from observations in isolated tissues. We estimated in kitten that 97% and 76% of the ehronotropic stimuli induced by ( - ) -

38

noradrenaline and ( - ) -adrenal ine , respectively, are caused by/~l-adrenoceptors. The remaining chronotropic stimuli o f 3% for ( - ) -noradrena l ine and 24% for ( - ) -adrena l ine are caused by sinoatrial ]~2-adrenoceptors. The chronotropic potency of ( - ) -adrena l ine is 1/2 of the potency of ( - ) - noradrenaline (Fig. 2) in kitten sinoatrial node. Using classical criteria (Lands et al. 1967), this similarity of chronotropie potencies would suggest that the effects of both ( - ) - noradrenaline and ( - ) -adrena l ine are mediated by /~l- adrenoceptors. Our finding of the agonist-dependent antagonism by ( - ) -b i sopro lo l of the effects of the two physiological catecholamines and their different fractional stimuli for sinoatrial/~1- and/~2-adrenoceptors clearly shows that the classical criterion of potency ratio is insufficient and can even be misleading for the classification o f receptor subtypes in one tissue. We conclude that the similarity ofchronot ropic potencies of ( - ) -noradrena l ine and ( - ) -adrena l ine is fortuitous because, while both /~l- and /~2-adrenoceptors contribute to the effect of ( - ) -adrenal ine , nearly only/~l- adrenoceptors cause the effects o f ( - ) -noradrenal ine .

In guinea pig the antagonism of the sinoatrial effects o f ( - ) -noradrena l ine by ( - ) -b i sopro lo l is consistent with an almost exclusive involvement of /~radrenoceptors . On the other hand, ( - ) -adrena l ine causes its positive chronotropic effects predominantly t h r o u g h / ~ - and to a smaller extent through/~2-adrenoceptors. Only after selective blockade of /~2-adrenoceptors by ICI 118,551 most effects of ( - ) - adrenaline occur through/~i-adrenoeeptors. For further evidence of the role af b o t h / ~ - and/~2-adrenoceptors in the sinoatrial node o f guinea pig see Lemoine et al. (1985):

Acknowledgements. We are grateful to Prof. H~iusler, Dr. Harting, Dr. Enenkel, Dr. Nowak and Dr. Faro of Merck (Darmstadt, FRG) for the synthesis of 3H-(-)-bisoprolol, for a gift of (-)-bisoprolol and for determining the partition coefficients of t-blockers. Beate Hillebrand provided technical assistance. Sibille Holzmann carefully prepared the figures. We thank Bernhard Ehle for performing the computer calculations. This work was supported by the Deutsche Forschungsgemeinschaft (grant SFB 30, Kardiologie, 04).

References

Arunlakshana O, Schild HO (1959) Some quantitative uses of drug antagonists. Br J Pharmacol 14:48-58

Bilski A, Halliday SE, Fitzgerald JD, Wale JL (1983) The pharmacology of a fiz-selective adrenoceptor antagonist (ICI 118,551). J Cardiovasc Pharmacol 5: 430- 437

Birnbaumer L, Duran JM, Nakahara T, Kaumann AJ (1977) Adenylyl cyclases: stimulation by hormones and regulation by nucleotides. In: Jaimeson GA, Robinson DM (eds) Mammalian cell membranes, vol 5. Responses of plasma membranes. Butterworths, London Boston pp 105-150

Blinks JR (1965) Convenient apparatus for recording contractions of isolated muscle. J Appl Physiol 20: 755 - 757

Carlsson E, /kblad B, Brandstrom A, Carlsson B (1972) Dif- ferentiated blockade of the chronotropic effects of various adrenergic stimuli in the cat heart. Life Sci 11:953-958

Chtnieux-Guicheney P, Dausse JP, Meyer P, Schmitt H (1978) Inhibition of (3H)-dihydroalprenolol binding to rat cardiac membranes by various t-blocking agents. Br J Pharmacol 63:177-- 182

Cruickshank JM (1980) The clinical importance of cardioselectivity and lipophilicity in beta blockers. Am Heart J 100:160-- 178

De Lean A, Hancock AA, Lefkowitz RJ (1982) Validation and statistical analysis of a computer modeling method for

quantitative analysis of radioligand binding data for mixtures of pharmacological receptor subtypes. Mol Pharmacol 21 : 5 - 16

Dickinson K, Richardson A, Nahorski SR (1980) Homogeneity of beta2-adrenoceptors on rat erythrocytes and reticulocytes. A comparison with heterogeneous rat lung beta-adrenoceptors. Molee Pharmaeol 19:194- 204

Dixon W J, Brown MB (eds) (1979) Biomedical computer programs, P series. University of California Press, Berkeley Los Angeles London

Ehle B, Lemoine H, Kaumann AJ (1985) Improved evaluation of binding of ligands to membranes containing several receptor subtypes. Naunyn-Schmiedeberg's Arch Pharmacol 331:52- 59

Gille E, Lemoine H, Ehle B, Kaumann AJ (1985) The affinity of (-)-propranolol for ill- and/~2-adrenoceptors of human heart. Differential antagonism of the positive inotropic effects and adenylate cyclase stimulation by (-)-noradrenaline and ( - ) - adrenaline. Naunyn-Schmiedeberg's Arch Pharmacol 331: 60-70

Hancock AA, De Lean AL, Lefkowitz RJ (1979) Quantitative reso- lution of beta-adrenergic receptor subtypes by selective ligand binding: application of a computerized model fitting technique. Mol Pharmacol 16 :1 -9

Hellenbrecht D, Lemmer B, Wiethold G, Grobecker H (1973) Mea- surements of hydrophobicity, surface activity, local anaesthesia, and myocardial conduction velocity as quantitative parameters of non-specific membrane affinity of nine fl-adrenergic blocking agents. Naunyn-Schmiedeberg's Arch Pharmacol 277:211- 226

Kaumann AJ (1970) Adrenergic receptors in heart muscle: Re- lations among factors influencing the sensitivity of the cat papillary muscle to catecholamines. J Pharmacol Exp Ther 173: 383- 398

Kaumann AJ (1972) Potentiation of the effects of isoprenaline and noradrenaline by hydrocortisone in cat heart muscle. Naunyn- Schmiedeberg's Arch Pharmacol 273 : 134-153

Kaumann AJ (1978) On spare fl-adrenoceptors for inotropic effects of catecholamines in kitten ventricle. Naunyn-Schmiedeberg's Arch Pharmacol 305 : 97-102

Kaumann AJ (1982) Cultured heart cells as a model for fl- adrenoceptors in a heart pacemaker. Chronotropic spare fl- adrenoceptors and spare adenylyl cyclase for (-)-isoprenaline but not for (-)-diehloroisoprenaline in rat cardiocytes. Naunyn-Schmiedeberg's Arch Pharmacol 320:119-129

Kaumann AJ, Birnbaumer L (1974) Characteristics of the adrenergic receptor coupled to myocardial adenylyl cyclase. Stereo- specifity and determination of apparent affinity constants for beta blockers. J Biol Chem 249:7874-7885

Kaumann A J, Lemoine H (1983) Large discrepancies between membrane binding and blocking affinities in tissues for antagonists that select fl-adrenoceptor subtypes but not for nonselective antagonists. Naunyn-Schmiedeberg's Arch Pharmacol 324: R66

Kaumann AJ, Marano M (1982) On equilibrium constants for complexes of drug-receptor subtypes. Selective and nonselective interactions of partial agonists with two plausible/?- adrenoceptor subtypes mediating positive chronotropic effects of (-)-isoprenaline in kitten atria. Naunyn-Schmiedeberg's Arch Pharmacol 318 : 192- 201

Kaumann AJ, Mclnerny TK, Gilmour DP, Blinks JR (1980) Comparative assessment of/3-adrenoceptor blocking agents as simple competitive antagonists in isolated heart muscle: Similarity of inotropic and chronotropic blocking potencies against isoproterenol. Naunyn-Schmiedeberg's Arch Phar- macol 311:219-236

Kaumann AJ, Lemoine H, Morris TH, Schwederski U (1982) An initial characterization of human heart /~-adrenoceptors and their mediation of the positive inotropic effects of catecholamines. Naunyn-Schmiedeberg's Arch Pharmacol 319: 216--221

39

Kaumann A J, Morris TH, Bojar H (1983) On the functional role of heterogeneous binding sites. Heart ]~-adrenoceptors. J Receptor Res 3:61 - 70

Lands AM, Arnold A, Mc AuliffJP, Luduena FP, Brown TG (1967) Differentiation of receptor systems activated by sym- pathomimetic amines. Nature 214:597-598

Lefkowitz RJ, Stadel JM, Caron MG (1983) Adenylate cyclase- coupled beta-adrenergic receptors: Structure and mechanisms of activation and desensitization. Ann Rev Biochem 52, 159- 186

Lemoine H, Kaumann AJ (1982) A novel analysis of concentration-dependence of partial agonism. Ring-demethylation of bupranolol results in a high affinity partial agonist (K 105) for myocardial and tracheal/~-adrenoceptors. Naunyn-Schmiede- berg's Arch Pharmacol 320:130-144

Lemoine H, Kaumann AJ (1983) A model for the interaction of competitive antagonists with two receptor-subtypes characterized by a Schild-plot with apparent slope unity. Agonist- dependent enantiomeric affinity ratios for bupranolol in tracheae but not in right atria of guinea pigs. Naunyn- Schmiedeberg's Arch Pharmacol 322:111 - 120

Lemoine H, Ehle B, Kaumann AJ (1985) Direct labelling of/~2- adrenoceptors. Comparison of binding potency of 3H-ICI 118,551 and blocking potency of ICI 118,551. Naunyn- Schmiedeberg's Arch Pharmacol 331:40-51

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurements with the Folin phenol reagent. J Biol Chem 193:265 -275

Manalan AS, Bosch HR, Watanabe AM (1981) Characterization of (3H)-(+_)-carazolol binding to /~-adrenergic receptors. Application to study of/?-adrenergic receptor subtypes in canine ventricular myocardium and lung. Circulation Res 49: 326- 336

Minneman KP, Hegstrand LR, Molinoff PB (1979) Simultaneous determination of beta-I and beta-2 adrenergic receptors in tissues containing both receptor subtypes. Mol Pharmacol 16:34-46

Minneman KP, Pittman RN, Molinoff PB (1981) /?-Adrenergic receptor subtypes: Properties, distribution, and regulation. Ann Rev Neurosci 4:419-461

Morris TH, Kaumann AJ (1979) Characterization of 3H-(q-)- carazolol binding to myocardial //-adrenoceptors. Naunyn- Schmiedeberg's Arch Pharmacol 308:R 33

Morris TH, Kaumann AJ (1984) Different steric characteristics of /~1- and/~2-adrenoceptors. Naunyn Schmiedeberg's Arch Phar- macol 327:176-179

Morris TH, Sandrock K, Kaumann AJ (1981) 3H-(-)-Bupranolol, a new//-adrenoceptor radioligand: Characterization of its binding to kitten heart //-adrenoceptors. Naunyn-Schmiedeberg's Arch Pharmacol 317:19-25

Moustafa E, Giachetti A, Downey HF, Bashour FA (1978) Binding of (3H)-dihydroalprenolol to beta-adrenoceptors of cells isolated from adult rat heart. Naunyn-Schmiedeberg's Arch Pharmacol 303 : 107-109

Porzig H, Becker C, Reuter H (1982) Competitive and non-competitive interactions between specific ligands and beta-adrenoceptors in living cardiac cells. Naunyn-Schmiedeberg's Arch Pharmacol 321 : 8 9 - 99

Rugg EL, Barnett DB, Nahorski SR (1978) Coexistence of beta1 and betaz adrenoceptors in mammalian lung: evidence from direct binding studies. Mol Pharmacol 14: 996-1005

Salomon Y, Londos C, Rodbell M (1974) A highly sensitive adenylate cyclase assay. Anal Biochem 58: 541 - 548

Scatchard G (1949) The attractions of proteins for small molecules and ions. Ann NY Acad Sci 51:660-672

Trendelenburg U (1968) The effect of cocaine on the pacemaker of isolated guinea-pig atria. J Pharmacol Exp Ther 161:222- 231

Wfichter W, Miinch U, Lemoine H, Kaumann AJ (1980) Evidence that (+)-bupranolol interacts directIy with myocardial//-adrenoceptors: Control of optical purity with differential thermal analysis. Naunyn-Schmiedeberg's Arch Pharmacol 313 : 1 - 8

Walter M, Lemoine H, Kaumann AJ (1984) Stimulant and blocking effects of optical isomers of pindolol on the sinoatrial node and trachea of guinea pig. Role of ]~-adrenoceptor subtypes in the dissociation between blockade and stimulation. Naunyn- Schmiedeberg's Arch Pharmacol 327:159 - 175

Received June 1, 1984/Accepted July 5, 1985

Direct labelling of myocardial ? 1-adrenoceptors

Documents

Transcript of Direct labelling of myocardial ? 1-adrenoceptors