Scand J Clin Lab Invest 1986; 46: 121-130

Relative potencies of various P-adrenoceptor antagonists (BAA) at the level of the human myocardial (S-adrenoceptor-adenylate cyclase (AC) complex. Is intrinsic sympathomimetic activity (ISA) due to AC activation?

S V E I N G O L F * & V I D A R H A N S S O N t

*Medical Department B, Rikshospitalet, Oslo, and ths t i tu te of Pathology and Institute of Medical Biochemistry, University of Oslo, Norway

Golf S, Hansson V. Relative potencies of various p-adrenoceptor antagonists (BAA) at the level of the human myocardial fi-adrenoceptor-adenylate cyclase (AC) complex. Is intrinsic sympathomimetic activity (ISA) due to A C activation? Scan J Clin Lab Invest 1986; 46: 121-130.

Nine different p-adrenoceptor antagonists (BAA), five with intrinsic sympatho- mimetic activity (ISA), were examined for their ability to inhibit isoproterenol- stimulated adenylate cyclase (AC) activity and spe~ific''~I-cyanopindolol (CUP) binding in crude membrane particles from human myocardium. The BAA'S were: propranolol, pindolol, timolol, alprenolol. metoprolol. atenolol, prenal- terol. ICI 141.292 'Visacor', and ICI 118.587 'Corwin'. Whether BAAS with strong ISA were able to stimulate A C activity by themselves were examined in separate experiments and compared to the A C stimulation by full agonists. All the BAAS caused a concentration dependent, and at high doses apparently complete, inhibition of both isoproterenol-stimulated A C activity and i2sI-CYP binding. Both assays made possible a 'potency-ranking' of the different BAAS (pindolol Ppropranolol and timolol > ICI 142.292 and alprenolol > ICI 118.587, prenalterol and metoprolol > atenolol). Corrected IC,,,-values, derived from inhibition curves with both techniques, show that receptor binding and inhibition of receptor function follow each other closely. Prenalterol caused a very weak AC activation (5.4% of maximum), whereas the 'ISA-blockers', pindolol, ICI 141.292, and ICI 118.587 were unable to stimulate A C activity at concentrations which completely displaced i2s1-CYP binding. In comparison, norepinephrine stimulated A C activity to the same level as isoproterenol (three to four times basal activity) and the (&selective agonist terbutaline caused some 5O'X of maximal A C stimulation. This raises the question whether ISA is due to A C activation. The effect upon A C activation and "'I-CYP binding of drugs with fi-selectivity shows that both 0,- and P?-receptors are coupled to the AC.

Key words: adenylate cyclase; p-adrenoceptors; human myocardium; intrinsic sympathomimetic activity (ISA)

Svrin Golf', M D , Medicul Departmetif B , Section of Cardiology, Rikshospifuler, 0027 Oslo 1. Norwuy

121

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

122 S. Golf & V. Hansson

The potency of a P-adrenoceptor antagonist (BAA) on the target cell membrane is defined by its receptor affinity. Receptor affinity is probably best evaluated in radioligand assays, however, this technique does not distinguish between agonistic and antagonistic activities. Measurements of adenylate cyclase (AC) ac- tivity is possibly a more functional test of the interaction between drug and receptor, and the AC assay should potentially distinguish be- tween drugs which are agonists, partial agonists/antagonists, and pure antagonists.

The intrinsic sympathomimetic activity (ISA) of some BAAs has been well documented in haemodynamic and electrophysiological studies [l-31, and in isolated organs from laboratory animals [4-51. The ISA has also been described in several clinical studies [6-91.

The ISA of some BAAS is suggested as being caused by a partial agonist-like effect of the antagonist upon the same receptor which is blocked [10,11].

The main objective of the present study was t o explore the relative potency of different BAAs on human myocardial P-adrenoceptors by measuring the ability of each B A A to inhibit isoproterenol-stimulated (lo-' mol/l) A C ac- tivity and specific '251-CYP binding. Furthermore, we wanted to examine to what extent BAAS with documented ISA revealed any partial agonist-like activity in the A C assay, and then compare the magnitude of this stimulation with that of full PI a n d o r P2-agonists like isoprotere- nol, norepinephrine and terbutaline. The AC assay should be useful for this purpose, since only the receptor, the coupling proteins, and the catalytic subunit of the adenylate cyclase are involved. Carried out in vitro on membrane preparations, autonomic responses are ex- cluded, as well as interactions between pre- synaptic and postsynaptic receptors responsible for stimulatiodinhibition of catecholamine re- lease from nerve terminals [12].

M A T E R I A L S A N D M E T H O D S

Collection of material. Human myocardial biop- sies were obtained during open heart surgery, either from the right atrial auricle at the introduction of the drainage tubes of the headlung machine (five patients operated for coronary disease), from the tip of the papillary

muscle when preparing the mitral orifice for prosthetic valve implantation (three patients operated for mitral valve disease), or from excision of infundibular muscular hypertrophy in one patient with valvular and subvalvular pul- monary stenosis. None of the patients had severe heart failure. The biopsies were collected in the operation theatre and immediately sub- merged in liquid nitrogen and stored in the same medium until preparation for the assay.

Informed consent was obtained from the patients, and the procedure for obtaining myo- cardial material was approved by the ethical committee of The Norwegian Research Council for Science and the Humanities.

Preparation of crude membrane particles. Visible fat and fibrous tissue were carefully removed and the biopsies were minced with a sharp scissor in ice-cold assay buffer. Following homogenization (Ultraturrax, Janke Kunkel, FRG) for 2 x 3 seconds, the homogenates were filtered through a nylon mesh and centrifuged at 27,000 g for 30 minutes at &2 "C. The pellet was resuspended in 2@30 volumes (wthol) of assay buffer containing 0.1 % bovine serum albumin (BSA), and the final protein concen- tration in the suspension, measured according to Lowry et al. [13], was approximately 1-2 mg proteidml.

Adenylate cyclase assay. The A C assay was performed as described previously [ 141. Inhi- bition curves for the different BAAS were established by the ability of different concen- trations of the compound at test to inhibit isoproterenol- stimulated ( lo-' mol/l) A C ac- tivity. The ability of 'ISA-blockers' to stimulate AC activity was examined in separate experi- ments. in which basal A C activity was estab- lished by 12 separate determinations, and the A C activity at each BAA concentration by eight separate determinations. The presence of A C activation by any concentrations of the BAAS with ISA was explored by performing comparisons to basal A C activities using the Wilcoxon rank sum test, and if present, cal- culated in percentage of maximal A C activation by isoproterenol (lo-' mol/l). Finally, complete dose/response curves of A C activation by iso- proterenol, norepinephrine and terbutaline were established. All experiments using the A C assay were carried out two to four times with essentially identical results.

Radioligand assay. The ligand binding assay

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

/3-blocking agents and human myocardial P-adrenoceptors 123

was performed in principle as described pre- viously [15] in a total volume of 150 pl containing 100 pl of membrane particle prep- aration, 25 pl radioactive ligand '2sI-cyanopin- dolol, (CYP, New England Nuclear, Boston, USA, final concentration 400 pmol/l), and 25 pl of either buffer (for zero inhibition), o r B A A to be tested. fPropranolo1 in excess (lo-' moV1) was used for determining non-specific binding. The p2-selective agonist terbutaline was also examined in a binding.inhibition experiment in order to establish the concentration ranges for its binding to p2- and pl- receptors, respectively. The samples were incubated in the same buffer as used in the A C assay (10 mol/l Tris-HCI, 1.0 mmoVl EDTA, p H 7.4, 5 mmoVl Mg", and 0.04 mmol/l guanosine triphosphate (GTP)). The Mg2+ and GTP were added to avoid any affinity changes due to allosteric regu- lation of the receptors [16]. All ligand binding experiments were carried out at least twice, with essentially identical results.

The equilibrium constant of dissociation (Kd) of both (+/-) and (-) '"1 -CYP binding has been established in previous experiments [ 15, 171 by employing different concentrations of the ligand ('hot' dose/response curves) and

Scatchard analysis of the binding data [18]. These Kd values were 230 pmoVl and 120 pmoVl. respectively, which are close to the values obtained by others 1191.

Drug and chemicals. The BAAS examined were non-selective without ISA (propranolol, timolol), non-selective with weak or moderate ISA (alprenolol, pindolol), P1-selective without ISA (atenolol, metoprolol) or p,-selective with strong ISA (ICI 118 587 'Corwin', ICI 141 292 'Visacor'). Whether prenalterol is pl-selective or not is uncertain, but this drug possesses strong ISA.

Propranolol, atenolol, ICI 118 587 ('Cor- win'), and ICI 141 292 ('Visacor') were all generously provided by ICI, Pharmaceutical Division, UK. Metoprolol, alprenolol and pre- nalterol were gifts from Hassle AB, Sweden. Pindolol was a gift from Sandoz AG, Switzer- land; timolol was a gift from MSD, The Netherlands. Most of these drugs and all other compounds used in the assays are commercially available.

R E S U L T S

Figure 1 shows the inhibition of isoproterenol- stimulated (lo-' moVI) A C activity by the

I I I

I o - ~ 10-8 lo-' 10-6 I O - ~ Concentration /3-receptor ontogontst (mol / l )

Frc. 1. Concentration dependent inhibition of isoproterenol-stimulated adenylate cyclase (AC) activity by nine different P-adrenoceptor antagonists (BAA). (+/-) Isoproterenol- stimulated (lo-" moVI) AC activity (1X-27 pmoVmg protein . minute)=100%. Basal activity (6-10 pmoVmg protein . minute)=0%. All BAAS apparently inhibited isoproterenol- stimulated AC activity to basal level. Pindolol (w). tirnolol (A-A), propanolol (A-A), ICI 142.292 'Viscacor' (U-W), alpernolol (v-v), ICI 118.587 'Corwin' (M), Prenalterol (V-V), metoprolol (0-0) and atenolol (w).

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

124 S. Golf & V. Hansson

I I I 1 I 1 I

I o - ~ 10-8 10'' 10-6 10-5 I O - ~ Concentrat ion p-receptor ontogonis t (mol/l1

FIG. 2. Concentration dependent inhibition of i"I-cyanopindolol (CUP) binding by the same BAAs as in Fig. 1. Specific binding of 400 pmol/l "'I-CYP=lOO"/o. The "'I-CYP binding in presence of k propranolol 1K5 moYI=O%. All BAAs apparently inhibited specific "'I-CYP binding to basal level. See legend to Fig. 1 for symbols.

various BAAs. The AC activity in the presence of isoproterenol moI/I) was set at 1OO%, while 0% represented basal A C activity. The inhibition curves for the various BBAs were nearly parallel. The concentrations causing half maximal inhibition of A C activity varied by three orders of magnitude, and all the BAAs apparently inhibited A C activity completely to basal level.

Figure 2 describes the inhibition of specific "'1 -CYP binding by the same antagonists. The BAAs with PI selective properties exhibit biphasic curves with a 'foot' in the lower right part, compatible with binding to myocardial fi2-receptors at high concentrations [ lS , 191. No differences were seen in the inhibitory activities of the BAAs on artial and ventricular prepara- tions.

The ICs0 values derived from the curves in Figs 1 and 2 were corrected according to Cheng & Prusoff [20] using an equilibrium constant of dissociation (K, value) for isoproterenol A C activation of 2 x 1 W 7 mol/l [14] and Kd values for CYP binding as mentioned above. The corrected ICs(,-values are shown in Table I and the correlation between these two sets of data in Fig. 3.

Figure 4 exhibits inhibition of specific 1251- CYP binding by different concentrations of

terbutaline. A biphasic curve is revealed with a plateau at 20-2S% inhibition of "'I-CYP bind- ing. At this concentration range terbutaline obviously occupied the myocardial P2- receptors. However, the figure shows that this drug also binds to the BI-receptors when em-

lo-'' 10-9 10-8 10-7 10-6 1 0 - 5

Correlation IC,, inhibition ACoctlvlty FIG. 3 . Correlation between the corrected IC,, values derived from Figs 1 and 2. Each BAA is represented by the same symbol as in Figs 1 and 2. The identity line is indicated.

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

P-blocking agents and human myocardial b adrenoceptors 125

I00

0.

U c : 75 a

CT > I4

Y) ‘ 50 u a ’ 25

L - V

0

-

-

-

1 2 3 4 5 -\

Terbutolin ( rno l / l )

FIG. 4. Inhibition of “‘I-CYP binding by different concentrations of terbutaline. The ordinate is represented by a percentage scale in analogy with Fig. 2. Each point was determined by triplicate measurements, the bars indicate the standard deviations (SD) of each deter- mination. The diphasic form of the curve is compatible with occupancy of P2- receptors (2C-25% of total receptor density) at low concentrations and of PI-receptors at high concen- trations of terbutaline. The inserted figure exhibits a modified Hofstee plot [21] of the binding data. A two-component configuration of the curve is obvious. The low-affinity component (flat part of the curve) represents binding of terbutaline to the P,-receptors. Percentage inhibition of ”‘1-CYP binding=B; concentrations of terbutaline in wmoVI=C.

ployed in higher concentrations. A modified Hofstee plot [21] of the binding data is shown by the inserted figure. The two-component configuration of the curve demonstrates pre- sence of both PI and P2-receptors. The high affinity component (Pz) constitutes some 20- 25% of the total receptor number.

Figure 5 shows the A C doselresponse experi- ments with the ‘ISA-blockers’ prenalterol, ICI 115.587 ‘Corwin’, ICI 141.292 ‘Visacor’, and pindolol. In the figure, isoproterenol- stimulated moUl A C activity is set as 100% and basal activity is 0%. The standard deviations (SD) of basal activity, based on 12 separate determinations, are indicated by the shaded areas. The SD of measured AC activity at each concentration of the BAAs are also in- dicated, as well as the statistical significance in deviation from basal activity. Only prenalterol was able to perform a very weak (max 5.4%) stimulation of A C activity. This was found in three different experiments, while the other ‘ISA-blockers’ showed no such stimulation in two repeated experiments.

The A C dose/response experiments with the full agonists isoproterenol and norepinephrine

and the @+elective agonist terbutaline are shown in Fig. 6. Norepinephrine reached the same level of A C activation as isoproterenol when used in proper concentrations, whereas terbutaline at concentrations compatible with binding to P2-receptors caused approximately 50% of maximal A C activation. At higher concentrations of terbutaline, no further AC activation was observed. Actually, at the highest concentrations of terbutaline, A C acti- vation was completely inhibited.

D I S C U S S I O N

The present study shows that the affinity of various BAAs for the @-adrenoceptors in hu- man heart, can be determined in vitro both by the ability to inhibit isoproterenol-stimulated A C activity and the ability to inhibit 1251-CYP binding. A ranking of BAAs according to receptor affinity is thus possible by direct methods (Table I . ) . The correlation between the corrected ICso-values for inhibition of isoproterenol-stimulated A C activity and for inhibition of 1251-CYP binding (Fig. 3), indi- cates that receptor binding and receptor func-

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

126 S. Golf & V. Hansson 50 -

25 -

* * * * * * * * O J

0 c I O - ~ 1 0 . ~ +

5 0 4 Pindolol Imol/l)

- 0

li 10-8 lo-' z IC I 141.292 ' Visacor' (mol/ I)

* * * * * * * * * * * * * *

1 0 ' ~ 10.' 10 Prenalterol (mol 'I )

FIG. 5. Stimulation of A C activity by the 'ISA-blockers' prenalterol. ICI 118.587 'Corwin', ICI 141.292 'Visacor'. and pindolol. The ordinates are represented by a percentage scale where basal AC activity, as determined by 12 separate measurements. is 0%. and iso- proterenol-stimulated (lo-' moVl) A C activity is 100%. In the different experiments maximal isoproterenol-stimulated AC activity was three to four times basal. (Basal A C activity %11 pmoUmg protein . minute. and stimulated AC activity 28-45 p m o l h g protein . minute.) The A C activity at each concentration of BAA was deter- mined by eight separate measurements. Standard deviations (SD) of basal activities are indicated by the shaded areas. and SD of AC activity at each concentration of BAA by bars. The significance in deviation from basal activity. calculated by Wilcoxon rank sum test, are also indicated. *Not significant, **p<O.01. ***pt0 .05 .

tion of BAAs are interrelated so that a certain ined for their ability to inhibit isoproterenol level of receptor occupancy of a BAA corres- stimulated AC activity in human heart. The ponds with a similar level of effect upon ICs0 (K,) values of inhibition of AC activity in receptor function. our study are comparable to those reported [22].

Three out of the nine BAAs under investiga- In human myocardium there are both P I - and tion in the present study (propranolol, atenolol Pz-receptors, however, their relative number and metoprolol) have previously been exam- is controversial (19,23.24]. In our laboratory. we

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

&blocking agents and human myocardial /3-adrenoceptors 127

10-8 10-7 10-6 I 0-5 I O - ~ 10-3 10-2

FIG. 6. Dose/response curves of AC activation by isoproterenol, norepinephrine and terbuta- line. The ordinate is represented by a percentage scale in analogy with the other figures. Each point is the mean of duplicate measurements. The SD are represented by bars. Norepinephrine, which possess weak B,-selective properties stimulated AC activity to the same level as iso- proternol. Terbutaline, on the other hand, provoked 5G60% of maximal AC activation at concentrations compatible with only P,-receptor occupancy (according to Fig. 4). No further AC activation could be achieved, although Fig. 4 also indicates that binding to P,-receptors occurred. Hence, terbutaline is a &agonist and probably a P,-antagonist. The fall in stimulated AC activity at high concentrations of terbutaline may be a non-specific effect. Isoproternol (B-B), norepinephrine (M) and terbutaline (U).

rnol / I

have found approximately 2&25% of the P-adrenoceptors to be of the Pz-subtype both in the right atrial, the right ventricular and the left ventricular human myocardium [15, 171. The

TABLE I. Corrected IC,,,-values of inhibition of isoproterenol-stimulated AC activity and "51-CYP binding by the different BAAS

Isoproterenol- '"I-CYP binding stimulated AC

(mow activity (moV1)

Atenolol Metoprolol Prenalterol ICI 118.587

'Convin' Alprenolol ICI 141.292

'Visacor' Propranolol Timolol PicJolol

4.2X 7 . 7 ~ 1 0 - ~ 7.7X lo-'

5 . 8 ~ 1 0 - ~ 5.6X lo-'

3.5x 10 -* 7.4x10-' 5.6X10-" 4.7x lo-'

2 . 7 x W 6

4.2x10-'

1 . 7 ~ 1 0 - ~

1 . 7 ~ lo-' 3.3X lo-' 4.0X lo-' 3.7x lo-"

5 . 0 ~

8 . 3 ~

The values are based on 'crude' IC,,,-values derived from Figs 1 and 2 and corrected according to Cheng & Prusoff (20).

binding experiments shown in the present study, and especially that of terbutaline (Fig. 4). emphasize the validity of our previous estimations of P-adrenoceptor subtypes. The inserted Hofstee plot (Fig. 4) gives an un- equivocal demonstration of receptor subtypes in the same way as previously shown with the P-antagonist ICI 118.551 [15, 171.

The coexistence of both P-adrenoceptor sub- types in human heart enables a calculation of the affinity constants of inhibition (K,) of both receptor subtypes for those BAAS with pl-selective properties to be made. Such calcu- lations were, however, beyond the scope of the present study, hence the use of the less precise term 'corrected IC5,,-values' (which is equal to K1 values for the drugs with non-selective properties). The presented ICso-values of the p,-selective BAAS constitute, however, a good approximation of the K1 values of the pl-re- ceptor subtype, since the P*-receptors are sup- posed to be responsible for the greater part of the slopes of the curves in both Figs 1 and 2. In Fig. 2 this is exhibited by the two-component configuration of the curves of the pl-selective

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

128 S. Golf & V . Hansson

BAAs, in which the (J,-component constitutes the small terminal 'foot'. Besides, similar ex- periments with the very strongly p2-selective B A A ICI 118.551 clearly illustrates the domi- nant role of B,-receptors in such curves [15, 171.

Based on the ICs,,-values for inhibition of isoproterenol-stimulated A C activity and '"I- CYP binding, the various BAAS can grossly be separated into two groups (Table I . Figs 1-3). All the drugs in the low affinity group are Pi-selective (atenolol. metoprolol, ICI 118.587) [ l , 25, 261. There are conflicting reports of the selectivity of prenalterol [27-301. In the high affinity group. the BAAS are nonselective P-blocking agents; except for ICI 141.292, which has been reported to be pi-selective

Five out of the nine BAAS examined in the present study (alprenolol, pindolol, ICI 141.292, ICI 118.587, and prenalterol) have well documented ISA both in experimental and clinical studies [ l , 2. 11, 311. The ISA of alprenolol is relatively weak. The other four drugs have all substantial ISA. In the case of prenalterol, the ISA is so strong that it may be considered an agonist rather than an antagonist [27, 281, and prenalterol has been used as antidote to metoprolol intoxication in clinical medicine [32].

None of the BAAS with ISA have previously been examined for receptor affinity or ability to inhibit or stimulate A C activity in membrane particle preparation from human heart. Results from the present study raises the question whether ISA of these drugs is mediated through stimulation of AC activity. This is based on the fact that all the BAAs inhibited iso- proterenol-stimulated A C activity apparently completely to basal level. Neither pindolol, ICI 141.292 'Visacor' nor ICI 118.587 'Corwin' were able to stimulate AC activity in concen- trations compatible with full receptor occup- ancy (Fig. 5) . The marginal AC activation caused by prenalterol was seen only at very high concentrations (Fig. S ) , and appears to be out of proportion with respect to its sympath- omimetic effect seen in clinical studies. The very strong A C activation by isoproterenol in the same preparations (three to four times basal A C activity) demonstrates that the AC-system was intact. Thus, neither the membrane prep- aration procedure nor any other causes of reduced A C activity (e.g. heart failure) are

P I .

responsible for the lack of A C activation by the 'ISA-blockers'.

On kitten heart preparations, Hedberg et al. [30, 331 observed positive chronotropic and inotropic effects of prenalterol, although no increase in myocardial cAMP content could be detected, nor any stimulatory effect on A C activity. Furthermore, prenalterol, in contrast to isoproterenol, was not able to increase cAMP content in rat heart [33]. It has, how- ever, been demonstrated that isoproterenol and other full P-agonists provoke an excessive stimulation of AC in relation to its functional cardiac stimulation [34, 351, and it has been hypothesized that antagonists with ISA may exert their agonistic activity by stimulating A C by a fraction (corresponding to their ISA) of what is necessary for maximal functional stimu- lation by a full agonist (< lo% of maximal AC-stimulation). Thus, it is possible that a drug such as pindolol (ISA some 30% of isopro- terenol) needs to activate A C by only approxi- mately 3%. A sizeable 'spare' capacity [34, 3.51 should thus be involved (i.e. a very high number of 'spare' receptors), making the sensi- tivity of our assay too low for detecting any ISA.

Some observations have, however, chal- lenged the actuality of the spare receptor concept in human heart. Heart failure in humans is associated with loss of both adrenergic receptors and reduced response of the A C to P-adrenergic agonists [3&38]. This has been considered to be the cause of the decreased 0-agonist-stimulated muscle contrac- tion of heart failure, and interpreted as if little or no spare receptors are present in human heart [39]. Supporting evidence for such a causal relationship has emerged from the observation that failing human hearts, with a reduced number of P-receptors, preserve their inotropic response to Ca2+, when the inotropic response to catecholamines is impaired [40]. Whether or not there are spare receptors in the human myocardium, cannot be solved by the present data. The observations presented in Fig. 5 are. however, dependent upon the presence of a very high 'spare capacity' between A C activation and the physiological response if A C activation is to be responsible for the ISA.

Hedberg et al. [33] have suggested that other mechanisms than cAMP formation may be involved in the complete physiological response

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

/3-blocking agents and human myocardial /3-adrenoceptors 129

to catecholamines, that is, a direct effect upon Ca2+ influx. Several years ago Kaumann [34] raised the possibility that some physiological effects of P-adrenergic agonists are mediated through different P-receptors than those mediating stimulation of AC. Both these and other possibilities deserve further investiga- tions. A third mechanism, by which ISA of BAAS could be explained, is direct interaction with presynaptic receptors responsible for catecholamine release [12]. However, the ex- periments showing ISA in animals depleted of catecholamines with syrosingopine and reser- pine [ l , 31 do not support this hypothesis.

It has recently been postulated that only the receptors of the Pz-subtype are coupled to the catalytic moiety of the human A C [22]. The partial A C activation by terbutaline observed in this study, and the inhibition of A C activity by pl-selective BAAS at concentrations compat- ible with only P1-receptor occupancy, strongly opposes this hypothesis.

Terbutaline in high concentrations does not, however, lead to further A C stimulation (via the P,-receptors) but rather to an inhibition of A C activity (Fig. 6). Thus, although terbutaline seems to compete with "'1-CYP for pl-re- ceptor occupancy (Fig. 4). its receptor effect at these concentrations is uncertain and requires further exploration

In conclusion, both the AC assay and the radioligand assay are capable of determining receptor affinities and perform a ranking of P-blocking agents at the receptor level in human myocardium. Different BAAS exhibit a high correlation between receptor binding and inhibition of A C activation by isoproternol. Both PI- and P2-receptors are coupled to the AC-enzyme. Furthermore, our data raise some doubt as to whether ISA of the BAAS is mediated via stimulation of AC. Therefore, more investigations aimed at eventually demon- strating alternative mechanisms for the ISA are warranted.

R E F E R E N C E S 1 Nutall A , Snow HM. The cardiovascular effects of

ICI 118.587 a PI-adrenoceptor partial agonist. Br J Pharmacol 1982; 77: 381.

2 Smith HJ. Halliday SE. Earl DCN. Stribling D . Effects of selective (beta-1 and beta-2) and nonselective beta adrenoceptor antagonists on the cardiovascular and metabolic responses to isopro-

terenol: comparison with ICI 141.292. J Pharma- col Exp Ther 1983; 226: 211-6.

3 Raeder E A , Verrier RL. Lown B. Intrinsic sympathomimetic activity and the effects of beta- adrenergic blocking drugs on vulnerability to ventricular fibrillation. J Am Coll Cardiol 1983;

4 Kaumann A. Blinks JR. P-adrenoceptor blocking agents as partial agonists in isolated heart muscle: dissociation of stimulated and blockade. Naunyn- Schmiedebergs Arch Pathol 1980; 311: 237-48.

5 Clark BJ, Menninger K. Bertholet A . Pindnlol- the pharmacology of a partial agonist. Br J Clin Pharmacol 1982; 13 (Suppl. 2): 149-58.

6 Heikkila J. Nieminen MS. Cardiac safety of acute beta blockade: intrinsic sympathomimetic activity is superior to beta-I selectivity. Am Heart J 1982; 104: 464-72.

7 Hombach V, Braun V, Hopp H-W, Gil-Sanches D, Behrenbeck DW. Tauchert M. Hilger HH. Electrophysiological effects of cardioselective and non-cardioselective P-adrenoceptor blockers with and without ISA at rest and during exercise. Br J Clin Pharmacol 1982; 13 (Suppl. 2): 285-93.

8 Hansson L, Svensson A , Gudbrandsson T. Sivert- s o n R. Treatment of hypertension with 0- blockers with and without intrinsic sympatho- mimetic activity. J Cardiovasc Pharmacol 1983; 5 (Suppl. 1): 26-9.

9 Svendson TL. Central hemodynamics of P- adrenoceptor blocking drugs: P,-selectivity versus intrinsic sympathomimetic activity. J Cardiovasc Pharmacol 1983; 5 (Suppl. 1): 21-5.

10 Ariens EJ. Intrinsic activity: partial agonists and partial antagonists. J Cardiovasc Pharmacol 1983; 5 (Suppl. 1): 8-15.

11 Clark BJ. Beta-adrenoceptor-blocking agents: are pharmacologic differences relevant? Am Heart J 1982; 104: 334-46.

12 Langer SZ. Presynaptic regulation of the release of catecholamines. Pharmacol Rev 1981; 32:

13 Lowry O H , Rosebrough NJ, Farr AL, Randall RJ . Protein measurement with the fohn phenol reagent. J Biol Chem 1951; 193: 265-75.

14 Golf S. Jahnsen T. Attramadal H , FrBysaker T. Hansson V. Catecholamine responsive adenylate cyclase in human myocardial preparations. Scand J Clin Lab Invest 1984; 44: 317-27.

15 Golf S. L ~ v s t a d R , Hansson V. 6-adrenoceptor density and relative number of P-adrenoceptor subtypes in biopsies from human right atrial, left ventricular and right ventricular myocardium. Cardiovasc Res 1985; 19: 636-41.

16 Abramowitz J. Iyengar R , Birnbaumer L. Guanine nucleotide and magnesium ion regulation of the interaction of gonadotropic and 0- adrenergic receptors with their hormones: a comparative study using a single membrane system. Endocrinology 1982; 110: 33646.

17 Golf S. Myhre E. Abdelnoor M, Andersen D , Hansson V. Hypertrophic cardiomyopathy (HOCM) characterized by p-adrenoceptnr den- sity, relative amount of 0-adrenoceptor subtypes

1 (6): 1442-6.

337-62.

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.

130 S. Golf & V. Hansson

and adenylate cyclase activity. Cardiovasc Res

18 Schatchard G. The attraction of proteins for small molecules and ions. Ann NY Acad Sci 1949; 51: 66(&72.

19 Stiles GL. Taylor S, Lefowitz RJ. Human cardiac beta-adrenergic receptors; subtype heterogenity delineated by direct radioligand binding. Life Sciences 1983; 33: 467-73.

20 Cheng Y-C, Prusoff WH. Relationship between the inhibition constant (K,) and the concentration of inhibitor which causes 50 per cent inhibition (Is,)) of an enzymatic reaction. Biochem Pharma-

21 Minneman KP. Hedberg H , Molinoff PB. Com- parison of beta adrenergic receptor subtypes in mammalian tissues. J Pharmacol Exp Ther 1979; 211: 502-8.

22 Waelbroeck M, Taton G. Delhaye M, Chatelain P. Camus JC, Pochet R , Leclerc JL, D e Smet JM, Robberecht P. Christophe J . The human heart beta-adrenergic receptors. 11. Coupling of beta- adrenergic receptors with the adenylate cyclase system. Mol Pharmacol 1983; 24: 174-86.

23 Brodde 0-E, Karad K, Zerkowski H-R, Rohm N, Reiderneister JC. Coexistence of PI- and Pz- adrenoceptors in human right atrium. Direct identification by (5 ) - [ l’sl]-iodocyano-pindolol binding. Circ Res 1983; 53: 752-8.

24 Robberecht P. Delhaye M. Taton G . D e Neff P, Waelbroeck M, De Smet JM, Leclerc JL, Chate- lain P, Christophe J . The human heart beta- adrenergic receptors. I. Heterogeneity of the binding sites: presence of 50% beta,- and 50% beta,-adrenergic receptors. Mol Pharmacol 1983;

25 Ablad B, Borg KO, Carlsson E, Ek L, John- son G, Malmfors T, Regirdh C-G. A survey of the pharmacological properties of metoprolol in animals and man. Acta Pharmacol Toxicol 1975;

26 Lundgren B, Carlsson E , Herrman I. p-ad- renoceptor blockade by atenolol, metoprolol and propranolol in the anaesthetized cat. Europ J Pharmacol 1979; 55: 263-8.

27 Ronn 0, Graffner C, Johnson G, Jordo L, Lundborg P, Wikstrand J . Haemodynamic effects and pharmacokinetics of a new selective beta,- adrenoceptor agonist, prenalterol. and its inter- action with metoprolol in man. Europ J Clin Pharmacol 1979; 15: 9-13.

28 Jennings G, Bobik A, Oddie C , Hargreaves M, Korner P. Cardioselectivity of prenalterol and isoproterenol. Clin Pharmacol Ther 1983; 34: 74947.

1985; 19: 693-9.

COI 1973; 22: 3099-108.

24: 169-73.

36 (SUPPI. V): 7-23.

29 Rohn N , Wagner J , Schumann HJ. The lack of a pronounced preference of prenalterol for the beta-I adrenoceptor subtype. Naunyn- Schmiedebergs Arch Pharmacol 1980; 315: 85-8.

30 Hedherg A , Mattsson H. Beta adrenoceptor interaction of full and partial agonists in the cat heart and soleus muscle. J Pharmacol Exp Therap 1981; 219: 798-808.

31 Ahlquist RP. Adrenergic beta-blocking agents. Prog Drug Res 1976; 20: 27-42.

32 Wallin C-J, Hulting J . Massive metoprolol poison- ing treated with prenalterol. Acta Med Scand

33 Hedberg A . Carlsson E, Fellenius E , Lundgren B. Cardiostimulatory effects of prenalterol, a beta-I adrenoceptor partial agonist, in vivo and in vitro. Naunyn-Schiedebergs Arch Pharmacol 1982; 318:

34 Kaumann AJ . On spare P-adrenoceptors for inotropic effects of catecholamines in kitten ven- tricle. Naunyn-Schmiedebergs Arch Pharmacol

35 Venter JC. High efficiency coupling between beta-adrenergic receptors and cardiac contratility. Direct evidence for ‘spare’ beta-adrenergic recep- tors. Mol Pharmacol 1979; 16: 429-40.

36 Bristow MR, Ginsburg R , Minobe W, Cubicotti RS, Scott Sageman W, Lurie K, Billingham ME, Harrison DC. Stinson EB. Decreased catechola- mine sensitivity and beta-adrenergic-receptor density in failing human hearts. N Engl J Med

37 Bristow MR, Ginsburg R , Minobe WA, Harrison DC, Reitz BA, Stinson EB. P-adrenergic receptor measurements in normal and failing human right and left ventricle. Circulation 1982; 66 (Suppl. 11): 207.

38 Ginsburg R , Bristow MR, Billingham ME, Stin- son EB, Schroeder JS, Harrison DC. Study of the normal and failing isolated human heart: de- creased response of failing heart to isoproterenol. Am Heart J 1983; 106: 534-40.

39 Bristow MR. Myocardial P-adrenergic receptor downregulation in heart failure. Intern J Cardiol 1984; 5: 648-52.

40 Fowler MB, Bristow MR, Hopkins D G , Laser JA, Ginsburg R , Alderman EL, Schroeder JS. Impaired beta-adrenergic inotropic response in severe heart failure. Circulation 1984; 70 (Suppl. 11): 191.

1983; 214: 253-5.

185-91.

1978; 305: 97-102.

1982; 307: 205-11.

Received 28 June 1985 Accepted 13 September 1985

Scan

d J

Clin

Lab

Inv

est D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

11/2

0/14

For

pers

onal

use

onl

y.