Effects of three different L-type Ca2+ entry blockers on airway constriction induced by muscarinic receptor stimulation

K. Hirota, E. Hashiba, H. Yoshioka, S. Kabara and A. Matsuki

Department of Anesthesiology, University of Hirosaki, School of Medicine, Hirosaki 036-8562, Japan

Corresponding author. E-mail: masuika@cc.hirosaki-u.ac.jp

Accepted for publication: January 20, 2003


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Background. The crucial role of L-type Ca2+ channels in airway smooth muscle contraction suggests that these channels could be an important therapeutic target. There are three separate drug binding sites on this channel: those for dihydropyridines, benzothiazepines and phenyl alkylamines. In this study, we examined the effects of the dihydropyridines nifedipine and nicardipine, the benzothiazepine diltiazem, and the phenylalkylamine verapamil on airway constriction.

Methods. Tension of guinea-pig tracheal strips was measured isometrically in vitro with a force displacement transducer. Strips were precontracted with carbachol 10–7 M with or without 4-aminopyridine 10–3 M, a voltage-sensitive K+ channel blocker. Then, nifedipine 10–8–10–4 M, diltiazem 10–8–3x10–4 M or verapamil 10–8–3x10–4 M was added cumulatively to the organ bath (n=6 each). The bronchial cross-sectional area of pentobarbital-anaesthetized dogs was assessed using a bronchoscopy method. Bronchoconstriction was elicited with methacholine 0.5 µg kg–1 plus 5 µg kg–1 min–1, and then nicardipine 0–1000 µg kg–1, diltiazem 0–3000 µg kg–1 or verapamil 0–3000 µg kg–1 were given i.v. (n=7 each).

Results. In the in vitro experiments, nifedipine and diltiazem fully reversed carbachol-mediated tracheal contraction with logIC50 values of 4.76 (SEM 0.22) (mean 17.5 µM) and 4.60 (0.33) (mean 24.8 µM), respectively. Although verapamil 10–6–10–4 M reversed the contraction by 87.2%, strip tension re-increased by 18.1% following maximal relaxation with verapamil 3x10–4 M. This re-increase was almost fully abolished by pretreatment with 4-aminopyridine. In the in vivo experiments, nicardipine and diltiazem dose-dependently reversed methacholine-induced bronchoconstriction, with logID50 values of 3.22 (0.05) (mean 0.60 mg kg–1) and 1.85 (0.32) (mean 14.0 mg kg–1), respectively. Verapamil worsened methacholine-induced bronchoconstriction.

Conclusions. Although supraclinical doses of dihydropyridines and benzothiazepines can produce airway relaxant effects, these agents are unlikely to be used in the treatment of bronchoconstriction. In addition, verapamil may aggravate airway constriction.

Br J Anaesth 2003; 90: 671–5

Keywords: ions, ion channels; receptors, muscarinic; muscle, smooth


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Calcium (Ca) plays an important role in airway hyperreactivity because synthesis of inflammatory mediators and release from mast cells, airway smooth muscle contraction and neuronal conduction are dependent on cytosolic calcium ions (Ca2+), which are elevated via either Ca2+ channel opening or release from intracellular stores.1 Voltage-sensitive Ca2+ channels on airway smooth muscle are predominantly of the L-type.2 Thus, blockade of L-type Ca2+ channels may attenuate airway smooth muscle contraction.

There are three different binding sites on L-type Ca2+ channels: for dihydropyridine, benzothiazepine and phenyl alkylamine. These distinct high-affinity binding sites are located on the {alpha}1 subunit of the L-channel. However, the binding domains for dihydropyridines3 and benzothiazepines4 have been reported to be located on the extracellular side of the membrane in myocytes, while the phenylalkyl amine binding site is located on the intracellular surface.5 These antagonists may therefore produce differential effects on airway smooth muscle contractility.

As described above, the important role of intracellular Ca2+ led to speculation that Ca2+ entry blockers may be beneficial in the treatment of asthma. However, most clinical investigations suggest that Ca2+ entry blockers may produce only modest and highly variable effects on airway tone. It has been reported that an acute asthmatic attack increases sympathetic tone as plasma norepinephrine levels are significantly elevated.6 Moreover, theophylline, and ß-adrenoceptor agonists such as epinephrine administered as bronchodilators also produce positive inotropic effects. Thus, Ca2+ entry blockers are often used in the treatment of tachycardia and hypertension in asthmatic patients even if these agents are not used as bronchodilators. Verapamil has been reported to attenuate catecholamine-induced acceleration of AV junctional automaticity in patients7 and to inhibit sympathetically induced tachycardia and pressor responses in anaesthetized cats.8

Ca2+ mobilization by muscarinic stimulation consists of two components: an initial transient increase resulting from inositol-1,4,5-triphosphate-induced Ca2+ release and a sustained rise in intracellular Ca2+ concentration, resulting from Ca2+ influx via L-type Ca2+ channels.10 Thus, in the present study, we used a muscarinic-receptor-mediated airway constriction model. Using this model, we have evaluated the relaxant effect of three different types of Ca2+ entry blockers: dihydropyridines (nicardipine and nifedipine), benzothiazepines (diltiazem) and phenylalkylamines (verapamil).


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The study protocol was approved by our university animal care committee.

In vitro study
Female guinea pigs (350–400 g) were killed with an overdose of pentobarbital 75 mg kg–1 i.p. and the aorta was sectioned. The trachea was removed, carefully dissected from surrounding connective tissues, and then cut spirally into two tracheal strips (3x15 mm). Each strip was mounted in a 10 ml organ bath filled with Krebs-bicarbonate buffer (NaCl 124 mM, NaHCO3 25 mM, CaCl2 2.5 mM, MgSO4 1.3 mM, KCl 5 mM, NaH2PO4 0.6 mM, glucose 10 mM) oxygenated with oxygen/carbon dioxide 95%/5% at 37°C. Strip tension was measured isometrically with a force displacement transducer (Isometric transducer TB651T, Nihon Kohden, Tokyo, Japan). Each strip was subjected to a load of 2 g for at least 2 h, with frequent changes of the bath fluid until a stable baseline tension was obtained. This baseline was set as 0. The strips were then contracted with carbachol 10–7 M (EC50 for contraction). Nifedipine 10–8–10–4 M (n=7), diltiazem 10–8–3x10–4 M (n=7) or verapamil 10–8–3x10–4 M (n=7) were added cumulatively to the organ bath. To confirm that carbachol-induced tension did not change during the experiment, one of four organ baths received saline only. Each strip received only one Ca2+ entry blocker or saline. Relaxant effects of verapamil 3x10–4 M (n=7) in the presence of 4-aminopyridine 10–3 M, a voltage-sensitive K+ channel blocker, was also evaluated.

In vivo study
Twenty-one mongrel dogs (8–10 kg) were anaesthetized with pentobarbital 30 mg kg–1 i.v. plus 10 mg kg–1 h–1 and paralysed with pancuronium 0.2 mg kg–1 h–1 infusion. Dogs were randomly assigned to three groups (n=7 each): nicardipine, diltiazem or verapamil. The trachea was intubated with a special tracheal tube (ID 7 mm) with a second lumen for insertion of a superfine fibreoptic bronchoscope (OD 2.2 mm) and the lungs were ventilated using a volume-controlled respirator (Servo 900C, Siemens-Elema AB, Sonnla, Sweden) with oxygen 100%. End-tidal CO2 was maintained at 4.0–4.5%. A femoral artery was cannulated to monitor arterial pressure and to obtain arterial blood samples. A femoral vein was also cannulated with a double-lumen catheter to infuse methacholine and fluid and to administer Ca2+ entry blockers and propranolol. Airway tone was assessed as change in the bronchial cross-sectional area, as described previously.11 Briefly, images of the bronchial cross-sectional area were printed using a videoprinter (Videoprinter VY-170, Hitachi, Tokyo, Japan) during the end-expiratory pause, and then measured using image analysis software (MacSCOPE 2.56, Mitani Co., Fukui, Japan). The measured bronchial cross-sectional area was expressed as a percentage of the basal (=100%). The coefficient of variation for this measurement was 2.36%. Bronchoconstriction was elicited with methacholine 0.5 µg kg–1 i.v. plus continuous i.v. infusion at 5 µg kg–1 min–1 until the end of the experiment. Thirty min later, when stable bronchoconstriction was achieved, Ca2+ entry blockers (nicardipine 0 [saline], 1, 10, 100, 1000 µg kg–1, diltiazem 0, 3, 30, 300, 3000 µg kg–1 or verapamil 0, 3, 30, 300, 3000 µg kg–1) were administered cumulatively i.v. Propranolol 0.4 mg kg–1 was then administered. At least 15 min elapsed between each dose of Ca2+ entry blocker. Propranolol was given immediately after printing out the bronchial cross-sectional area at the highest dose of Ca2+ entry blockers. Bronchial cross-sectional area was assessed before (basal) and 30 min after methacholine infusion started and 5 min after each dose of Ca2+ entry blocker and propranolol. Arterial blood (6 ml) was collected simultaneously and centrifuged at 3000 rpm for 10 min at –10°C to separate the plasma, which was kept frozen at –70 °C until assay. Plasma catecholamine concentrations were determined by high performance liquid chromatography with electrochemical detection. The intra-assay coefficients of variation for epinephrine and norepinephrine were 3.31% and 2.93%, respectively. The lower limits of detection for epinephrine and norepinephrine were 9 pg ml–1 and 12.5 pg ml–1, respectively.

Data analysis
All data are expressed as mean (SEM). Bronchial cross-sectional area is presented as percentage of the basal value. Data were analysed using one-way or repeated measures analysis of variance followed by Fisher’s protected least significant difference test, as appropriate, using Stat View II on a Macintosh computer. P<0.05 was considered significant.

Dose–response curves for Ca2+ entry blocker induced relaxation were expressed as percentage relaxation where peak constriction by carbachol or methacholine was taken as 0% and fully relaxed (baseline) as 100%. Sigmoid dose–response curves were fitted using GraphPad Prism 1.03, from which –log[concentration or dose producing 50% relaxation] (pC50 or pD50) for each Ca2+ entry blocker were estimated.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
In vitro study
Carbachol 10–7 M increased tracheal strip tension to a maximum of 1.4 (SEM 0.1) g which, in the saline group, was sustained until the end of experiment. Nifedipine and diltiazem, fully and in a concentration-dependent manner, reversed carbachol-induced contraction, with pIC50 values of 4.76 (SEM 0.22) (mean IC50 17.5 µM) and 4.60 (0.33) (mean IC50 24.8 µM), respectively (Fig. 1). Verapamil also concentration-dependently relaxed tracheal strips, with pIC50 of 4.32 (0.08) (mean IC50 47.4 µM), but in contrast to nifedipine and diltiazem, relaxation reached only 87.2 (3.8)% then decreased to 69.1 (2.7)% (Fig. 1).



View larger version (15K):
[in this window]
[in a new window]
 
Fig 1 Relaxant effects of nifedipine, diltiazem and verapamil on tracheal smooth muscle contraction elicited by carbachol. Nifedipine (closed circle) and diltiazem (open circle) inhibited contraction in a concentration-dependent manner. Whilst verapamil also concentration-dependently inhibited contraction (closed square), the tension re-increased after maximal relaxation (represented by an open triangle). Values are mean (SEM).

 
4-Aminopyridine 10–3 M increased tension to 0.9 (0.1) g, which was further increased by carbachol 10–7 M to 1.6 (0.2) g. In the presence of 4-aminopyridine, the decrease in relaxation following maximal relaxation by verapamil 3x10–4 M was essentially abolished – from 82.0 (6.3)% to 78.0 (8.9)%.

In vivo study
In agreement with our in vitro data, nicardipine and diltiazem dose-dependently reversed methacholine-induced bronchoconstriction, with pID50 values of 3.22 (SEM 0.05) (mean ID50 0.60) mg kg–1 and 1.85 (0.32) (mean ID50 14.0) mg kg–1, respectively. Verapamil worsened the constriction (Fig. 2). Propranolol did not change the airway tone (Fig. 2). Plasma epinephrine and norepinephrine concentrations were increased by all Ca2+ entry blockers (Table 1).



View larger version (19K):
[in this window]
[in a new window]
 
Fig 2 (A) Effects of nicardipine, diltiazem and verapamil on methacholine (Mch)-induced bronchoconstriction assessed by measuring changes in bronchial cross-sectional area. Pre=before Mch infusion; Mch30=30 min after Mch infusion started; PPL=propranolol. **P<0.01 compared with Mch30. (B) Dose–response curves of nicardipine, diltiazem and verapamil for Mch-induced bronchoconstriction. 100%=bronchial cross-sectional area at Pre (full relaxation); 0%=bronchial cross-sectional area at Mch30 (peak Mch-bronchoconstriction). Values are mean (SEM).

 

View this table:
[in this window]
[in a new window]
 
Table 1 Effects of Ca2+ entry blockers on plasma catecholamine concentrations. Data are mean (SEM), n=7. *P<0.05; **P<0.01 vs Mch30
 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The relaxant effect of nifedipine and diltiazem on carbachol-induced contraction of tracheal smooth muscle strips was more potent than that produced by verapamil. In addition, our in vivo data indicate that nicardipine and diltiazem attenuate methacholine-induced airway constriction whereas verapamil may worsen constriction. Collectively, these data suggest that blockade of L-type Ca2+ channels at the extracellular (dihydropyridine and benzothiazepine) rather than intracellular (phenylalkylamine) surface of the smooth muscle cell membrane may attenuate bronchoconstriction. Similarly, Chapman and colleagues12 reported that extracellular-site Ca2+ entry blockers inhibited histamine- and antigen-induced bronchoconstriction whereas the intracellular-site Ca2+ entry blockers did not.

In the present study, nicardipine and diltiazem concentrations above 10 and 300 µg kg–1, respectively, produced significant spasmolytic effects on methacholine-induced bronchoconstriction in vivo. Verapamil 300 µg kg–1 significantly worsened bronchoconstriction. Are these doses in dogs clinically relevant? The steady-state distribution volumes of nicardipine, diltiazem and verapamil, respectively, are 2- to 3-fold, 1- to 2-fold and 1- to 3-fold higher in dogs1315 than humans.1619 Thus, nicardipine 100 µg kg–1 and diltiazem or verapamil 300 µg kg–1 would be clinically relevant in the present study. Collectively, the present data suggest that clinically relevant doses of nicardipine and diltiazem may produce significant broncho dilation. However, as the ID50 values of nicardipine (0.60 mg kg–1) and diltiazem (14.0 mg kg–1) were much higher than clinically relevant doses, bronchodilation may not be large. Hence, nicardipine and diltiazem may not be particularly useful therapeutic agents for treatment of bronchoconstriction.

Several reports suggest that verapamil may have bronchodilating effects. However, the present in vivo data show that verapamil may worsen bronchoconstriction. In addition, as described above, clinically relevant doses may also do this. Similarly, inhaled verapamil has been reported to produce bronchoconstriction2022 or reduce atropine-induced relaxant effects,23 although several reports demonstrated a small bronchodilating action.24 25 In the present in vitro study, we observed that verapamil 3x10–4 M produced a maximal decrease in tension of tracheal smooth muscle followed by a re-increase of 18%. Verapamil has been reported to inhibit voltage-sensitive K+ channels26 and it is known that blockade of these channels increases airway tone.27 In the present study, the re-increase in tension was almost fully abolished in the presence of 4-aminopyridine, a voltage-sensitive K+ channel blocker. Thus, the inhibitory effects of verapamil on voltage-sensitive K+ channels might be involved in the mechanism of deterioration of methacholine-induced bronchoconstriction in vivo. In contrast to our in vitro data showing relaxant effects of verapamil, bronchodilating effects of verapamil were not observed in vivo. Further studies will be necessary to explain the discrepancies between these in vitro and in vivo data.

Plasma levels of the catecholamines increased significantly following i.v. administration of Ca2+ entry blockers in the present study. This may be because Ca2+-entry-blocker-induced vasodilatation promoted baroreceptor reflexes to increase sympathetic tone.28 Direct neural supply of the sympathetic system in the lung is limited. Thus, as circulating catecholamines mainly influence airway tone,29 Ca2+-entry-blocker-induced catecholamine release may attenuate bronchoconstriction. However, as propranolol 0.4 mg kg–1 did not antagonize the relaxant effects of Ca2+ entry blockers, spasmolytic effects of Ca2+ entry blockers may not be caused by catecholamine release.

In the present study, pancuronium was infused i.v. to prevent spontaneous respiration and bucking. As pancuronium has been reported to antagonize M2 rather than M1 or M3 muscarinic receptors,30 this relaxant may antagonize M2 receptors activated by methacholine. As M2 activation reduces acetylcholine release and counteracts ß-agonist-induced relaxation, pancuronium might potentiate broncho dilation caused by endogenous epinephrine (a ß-agonist) release. However, methacholine-induced bronchoconstriction is mainly mediated via M3 receptors. In addition, we previously observed that aminophylline-induced relaxant effects on methacholine-induced bronchoconstriction with pancuronium infusion were antagonized by the same dose of propranolol as used in the present dose. Thus, in the present study, the absence of antagonistic effects of propranolol could not be the result of M2 antagonism by pancuronium. Therefore, the dose of pancuronium used in this study is unlikely to have affected our results.

In conclusion, the present data suggest that dihydropyridines and benzothiazepines may not be useful therapeutic agents for the treatment of bronchoconstriction, although supraclinical doses of these agents may produce airway relaxation. Moreover, verapamil may aggravate airway constriction.


    Acknowledgement
 
We thank Dr T. Kudo for his help in the assay of plasma catecholamines.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Ahmed T, D’Brot J, Abraham W. The role of calcium antagonists in bronchial reactivity. J Allergy Clin Immunol 1988; 81: 133–44[ISI][Medline]

2 Yamakage M, Cheng X, Tsujiguchi N, Kamada Y, Namiki A. Different inhibitory effects of volatile anesthetics on T- and L-type voltage-dependent Ca2+ channels in porcine tracheal and bronchial smooth muscles. Anesthesiology 2001; 94: 683–93[CrossRef][ISI][Medline]

3 Striessnig J, Murphy BJ, Catterall WA. Dihydropyridine receptor of L-type Ca2+ channels: identification of binding domains for [3H](+)-PN200-110 and [3H]azidopine within the {alpha}1 subunit. Proc Natl Acad Sci USA 1991; 88: 10 769–73[Abstract]

4 Kurokawa J, Adachi-Akahane S, Nagao T. 1,5-Benzothiazepine binding domains is located on the extracellular side of the cardiac L-type Ca2+ channel. Mol Pharmacol 1997; 51: 262–8[Abstract/Free Full Text]

5 Striessnig J, Glossmann H, Catterall WA. Identification of a phenylalkylamine binding region within the {alpha}1 subunit of skeletal muscle Ca2+ channels. Proc Natl Acad Sci USA 1991; 87: 9108–12[ISI]

6 Pichurko BM, Sullivan B, Porcelli RJ, McFadden ER Jr. Endogenous adrenergic modification of exercise-induced asthma. J Allergy Clin Immunol 1986; 77: 796–801[ISI][Medline]

7 Lee KL, Lauer MR, Young C, et al. Spectrum of electrophysiologic and electropharmacologic characteristics of verapamil-sensitive ventricular tachycardia in patients without structural heart disease. Am J Cardiol 1996; 77: 967–73[CrossRef][ISI][Medline]

8 Lee KL, Chun HM, Liem LB, Sung RJ. Effect of adenosine and verapamil in catecholamine-induced accelerated atrioventricular junctional rhythm: insights into the underlying mechanism. Pacing Clin Electrophysiol 1999; 22: 866–70[ISI][Medline]

9 Malo JL, Pineau L, Cartier A, Martin RR. Reference values of the provocative concentrations of methacholine that cause 6% and 20% changes in forced expiratory volume in one second in a normal population. Am Rev Respir Dis 1983; 128: 8–11[ISI][Medline]

10 Kajita J, Yamaguchi H. Calcium mobilization by muscarinic cholinergic stimulation in bovine single airway smooth muscle. Am J Physiol 1993; 264: L496–503[ISI][Medline]

11 Otomo N, Hirota K, Hashimoto Y, et al. Measurement of bronchodilatation using a superfine fibreoptic bronchoscope. Br J Anaesth 1997; 78: 583–5[Abstract/Free Full Text]

12 Chapman RW, Danko G, Siegel MI. Effect of extra- and intracellular calcium blockers on histamine and antigen-induced bronchospasms in guinea pigs and rats. Pharmacology 1984; 29: 282–91[ISI][Medline]

13 Chelly JE, Hysing ES, Abernethy DR, et al. Role of isoflurane on hemodynamic properties and disposition of nicardipine. J Pharmacol Exp Ther 1987; 241: 899–906[Abstract]

14 duSouich P, Hartmann D, Saunier C. Effect of acute and chronic moderate hypoxia on diltiazem kinetics and metabolism in the dog. Pharmacology 1993; 47: 378–85[ISI][Medline]

15 McAllister RG Jr, Bourne DW, Dittert LW. The pharmacology of verapamil. I. Elimination kinetics in dogs and correlation of plasma levels with effect on the electrocardiogram. J Pharmacol Exp Ther 1977; 202: 38–44[Abstract]

16 Gurret M, Cheymol G, Julien-Larose C, Lavene D. Simultaneous study of the pharmacokinetics of intravenous and oral nicardipine using a stable isotope. Eur J Clin Pharmacol 1989; 37: 381–5[CrossRef][ISI][Medline]

17 Hermann P, Rodger SD, Remones G, Thenot JP, London DR, Morselli PL. Pharmacokinetics of diltiazem after intravenous and oral administration. Eur J Clin Pharmacol 1983; 24: 349–52[CrossRef][ISI][Medline]

18 Koike Y, Shimamura K, Shudo I, Saito H. Pharmacokinetics of verapamil in man. Res Commun Chem Pathol Pharmacol 1979; 24: 37–47[ISI][Medline]

19 McAllister RG Jr, Kirsten EB. The pharmacology of verapamil. IV. Kinetic and dynamic effects after single intravenous and oral doses. Clin Pharmacol Ther 1982; 31: 418–26[ISI][Medline]

20 Fish JE, Norman PS. Effects of the calcium channel blocker, verapamil, on asthmatic airway responses to muscarinic, histaminergic, and allergenic stimuli. Am Rev Respir Dis 1986; 133: 730–4[ISI][Medline]

21 Kivity S, Yarchovsky J, Greif J, Topilsky M. Combined effect of inhaled verapamil and sodium cromoglycate on the airway response to hypertonic saline. Ann Allergy 1990; 64: 163–5[ISI][Medline]

22 Harman E, Hill M, Pieper JA, Hendeles L. Inhaled verapamil-induced bronchoconstriction in mild asthma. Chest 1991; 100: 17–22[Abstract]

23 Krivoy N, Brandt HD, Bunn AE. Interaction of verapamil hydrochloride and atropine sulphate on histamine-induced bronchoconstriction. Respiration 1985; 47: 185–9[ISI][Medline]

24 Patel KR. Calcium antagonists in exercise-induced asthma. Br Med J 1981; 282: 932–3[ISI][Medline]

25 McIntyre E, Fitzgibbon B, Otto H, Minson R, Alpers J, Ruffin R. Inhaled verapamil in histamine-induced bronchoconstriction. J Allergy Clin Immunol 1983; 71: 375–81[CrossRef][ISI][Medline]

26 Rauer H, Grissmer S. Evidence for an internal phenylalkylamine action on the voltage-gated potassium channel Kv1.3. Mol Pharmacol 1996; 50: 1625–34[Abstract]

27 Adda S, Fleischmann BK, Freedman BD, Yu M, Hay DW, Kotlikoff MI. Expression and function of voltage-dependent potassium channel genes in human airway smooth muscle. J Biol Chem 1996; 271: 13 239–43[Abstract/Free Full Text]

28 Blank TJJ, Lee DL. Cardiac physiology. In: RD Miller, ed. Anesthesia 4th Edn. New York: Churchill-Livingstone, 1994; 247–89

29 Gal TJ. Bronchial hyperresponsiveness and anesthesia: physiologic and therapeutic perspectives. Anesth Analg 1994; 78: 559–73[ISI][Medline]

30 Cembala TM, Sherwin JD, Tidmarsh MD, Appadu BL, Lambert DG. Interaction of neuromuscular blocking drugs with recombinant human m1–m5 muscarinic receptors expressed in Chinese hamster ovary cells. Br J Pharmacol 1998; 125: 1088–94[Abstract]