contractions of the mouse prostate elicited by...

Contractions of the Mouse Prostate Elicited by AcetylcholineAre Mediated by M3 Muscarinic Receptors

Carl W. White, Jennifer L. Short, John M. Haynes, Minoru Matsui, and Sabatino VenturaMedicinal Chemistry and Drug Action, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, Victoria,Australia (C.W.W., J.L.S., J.M.H., S.V.); and Department of Clinical Research and General Medicine, Tokyo-Nishi TokushukaiHospital, Tokyo, Japan (M.M.)

Received August 10, 2011; accepted September 1, 2011

ABSTRACTIncreased smooth muscle tone in the human prostate contrib-utes to the symptoms associated with benign prostatic hyper-plasia. In the mouse prostate gland, cholinergic innervation isresponsible for a component of the nerve-mediated contractileresponse. This study investigates the muscarinic receptor sub-type responsible for the cholinergic contractile response in themouse prostate gland. To characterize the muscarinic receptorsubtype, mouse prostates taken from wild-type or M3 musca-rinic receptor knockout mice were mounted in organ baths. Theisometric force that tissues developed in response to electrical-field stimulation or exogenously applied cholinergic agonists inthe presence or absence of a range of muscarinic receptorantagonists was evaluated. Carbachol elicited reproducibleand concentration-dependent contractions of the isolatedmouse prostate, which were antagonized by the presence ofmuscarinic receptor antagonists. Calculation of antagonist af-

finities (pA2 values) indicated a rank order of antagonist poten-cies in the mouse prostate of: darifenacin (9.08) � atropine(9.07) � 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide(9.02) � cyclohexyl-hydroxy-phenyl-(3-piperidin-1-ylpropyl)si-lane (7.85) � cyclohexyl-(4-fluorophenyl)-hydroxy-(3-piperidin-1-ylpropyl)silane (7.39) � himbacine (7.19) � pirenzipine(6.88) � methoctramine (6.20). Furthermore, genetic deletion ofthe M3 muscarinic receptor inhibited prostatic contractions toelectrical-field stimulation or exogenous administration of ace-tylcholine. In this study we identified that the cholinergic com-ponent of contraction in the mouse prostate is mediated by theM3 muscarinic receptor subtype. Pharmacological antagonismof the M3 muscarinic receptor may be a beneficial additionaltarget for the treatment of benign prostatic hyperplasia in thehuman prostate gland.

IntroductionMuscarinic receptors are G protein-coupled receptors con-

sisting of five subtypes denoted M1–5, which are activated bythe endogenous agonist acetylcholine. Muscarinic receptorscan be divided into two groups depending on their G protein-coupling properties. M1, M3, and M5 receptor subtypes coupleto Gq/11 proteins and signal through the inositol phosphatepathway, whereas the M2 and M4 receptor subtypes couple toGi/o proteins and inhibit adenylate cyclase activity. In the

prostate gland, cholinergic innervation acting at muscarinicreceptors in the glandular epithelium is thought to be respon-sible primarily for the secretion of prostatic fluids (Venturaet al., 2002). However, in the mouse prostate gland, in addi-tion to noradrenaline acting at �1L-adrenoceptors (Gray andVentura, 2006), we have shown a significant cholinergic com-ponent of the nerve-mediated contractile response wherebyacetylcholine acts at muscarinic receptors to mediate con-traction (White et al., 2010). Currently, the muscarinic re-ceptor subtype mediating the contractile response in themouse prostate gland is unknown.

In the canine prostate, M2 muscarinic receptors are re-sponsible for mediating contraction (Fernandez et al., 1998).In contrast, M1 muscarinic receptors mediate the facilitationof contraction in the guinea pig prostate (Lau et al., 2000). Inthe rat prostate, the M3 muscarinic receptor subtype hasbeen reported to mediate contractile responses to carbachol(Lau and Pennefather, 1998). Moreover, immunohistochem-ical studies of the rat prostate have shown colocalization ofthe M3 muscarinic receptor subtype with the prostatic

This work was supported by the ANZ Trustees [Grant 09/3164] (to S.V.) andthe National Health and Medical Research Council of Australia [Grant334136] (to S.V.).

Part of this work has been presented previously in abstract form: White C,Short J, and Ventura S (2010) Characterization of the muscarinic receptorsubtype mediating contraction in the mouse prostate gland, at the 16th WorldCongress on Basic and Clinical Pharmacology; 2010 July 17–23; Copenhagen,Denmark. International Union of Basic and Clinical Pharmacology, KansasCity, KS.

Article, publication date, and citation information can be found athttp://jpet.aspetjournals.org.

doi:10.1124/jpet.111.186841.

ABBREVIATIONS: M3R, M3 muscarinic receptor; 4-DAMP, 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide; ANOVA, analysis of variance;HHSiD, cyclohexyl-hydroxy-phenyl-(3-piperidin-1-ylpropyl)silane; p-F-HHSiD, cyclohexyl-(4-fluorophenyl)-hydroxy-(3-piperidin-1-ylpropyl)silane.

0022-3565/11/3393-870–877$25.00THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 339, No. 3Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics 186841/3729516JPET 339:870–877, 2011 Printed in U.S.A.

870

at ASPE

T Journals on June 26, 2018

jpet.aspetjournals.orgD

ownloaded from

http://jpet.aspetjournals.org/

smooth muscle (Nadelhaft, 2003). However, a number ofstudies have shown that subtype-specific muscarinic receptorantibodies are largely nonspecific (Jositsch et al., 2009; Mi-chel et al., 2009; Pradidarcheep et al., 2009). Previous immu-nohistochemical studies of the prostate using muscarinicsubtype-specific antibodies should be viewed cautiously with-out additional functional evidence for the role of a particularmuscarinic receptor subtype in the prostate.

Both the M1 and M2 muscarinic receptor subtypes havebeen observed in binding studies of whole human prostatehomogenates (Anisuzzaman et al., 2008). However, Ruggieriet al. (1995) observed only the M1 muscarinic subtype, whichwas found to be localized immunohistochemically in the ep-ithelium. In cultures of human prostatic smooth muscle cellsonly M2, M3, and M4 muscarinic receptor subtype mRNAexpression is observed (Obara et al., 2000), with the M2

muscarinic receptor being the most abundant. In a separatestudy of human prostatic smooth muscle cell cultures, M2

muscarinic receptors were observed by radioligand bindingexperiments and mediated a decrease in cAMP accumulation(Yazawa et al., 1994).

In the mouse prostate both the clinically selective M3 mus-carinic receptor antagonists darifenacin and solifenacin in-hibited [3H]N-methylscopolamine binding after oral admin-istration, indicating a population of muscarinic receptors(Oki et al., 2005; Yamada et al., 2006). With the availabilityof muscarinic receptor subtype knockout mice the M1 and M3,but not the M2, M4, or M5, muscarinic receptor subtypes wereshown to be present in mouse prostate by [3H]N-methylsco-polamine binding studies (Ito et al., 2009). However, thelocation and function of either the M1 or M3 muscarinicreceptor subtypes present in the mouse prostate were notelucidated. The muscarinic receptor subtype mediating con-traction in the mouse prostate is of interest because an in-crease in prostatic smooth muscle tone contributes to thelower urinary tract symptoms associated with benign pros-tatic hyperplasia in humans. The aim of this study was topharmacologically characterize the muscarinic receptor sub-type mediating the cholinergic contractile response in themouse prostate gland and to confirm the muscarinic receptorsubtype by using muscarinic receptor knockout mice.

Materials and MethodsAnimals. Heterozygous [M3R(�/�)] breeding pairs of adult M3

muscarinic receptor knockout [M3R(�/�)] mice were purchased fromthe Centre for Animal Resources and Development (Kumamoto Uni-versity, Kumamoto, Japan), and a colony was established at theMonash Animal Services facility (Clayton, Australia). Age-matchedlittermates were used at 10 weeks for experimentation and obtainedby matings of M3R(�/�) breeding pairs maintained on a B6.129background. The resulting litters of M3R(�/�), M3R(�/�), and wild-type [M3R(�/�)] mice were routinely genotyped by polymerase chainreaction using genomic DNA from tail samples obtained at weaning(21 days) as described previously (Matsui et al., 2000). For pharma-cological characterization experiments wild-type adult male mice (�8 weeks) from a C57BL/6 background were used. All mice were bredand housed at the Monash Animal Services facility, were exposed toa 12-h light/dark photoperiod, and had free access to food and water.Mice were weighed before being sacrificed by cervical dislocation.Prior approval for animal experimentation was granted by the Mo-nash University Standing Committee on Animal Ethics, ethics num-bers VCPA 2009/14 and VCPA 2009/15 for the use of geneticallymodified and wild-type mice, respectively. All studies conformed to

the National Health and Medical Research Council, Australian codeof practice for the care and use of animals for scientific purposes.

Isolated Organ Bath Studies. The whole mouse ventral pros-tate was obtained by an incision along the midline of the lowerabdomen that exposed the urogenital tract. To reveal the prostate,the penile muscles, excess fat, and connective tissue were cut away.The prostate was then carefully dissected out and placed in a spec-imen jar containing Krebs-Henseleit solution (118.1 mM NaCl, 25.0mM NaHCO3, 11.7 mM glucose, 4.69 mM KCl, 1.2 mM KH2PO4, 0.5mM MgSO4, and 2.5 mM CaCl2, pH 7.4). The prostate was mountedin a 10-ml water-jacketed organ bath containing Krebs-Henseleitsolution maintained at 37°C and bubbled with 95% O2/5% CO2. Oneend of the tissue was attached to a perspex tissue holder and theother to a Grass FT03 force displacement transducer (Grass Instru-ments, Quincy, MA) for the measurement of isometric contractions.The force of contraction was recorded using a PowerLab 4/SP dataacquisition system (ADInstruments Pty Ltd., Castle Hill, Australia)and LabChart software version 5 (ADInstruments Pty Ltd.). Tissueswere maintained under a resting force of approximately 0.7g. Beforeexperimentation tissue preparations were equilibrated for 1 h, dur-ing which time prostates were stimulated with electrical pulses of0.5-ms duration, 60 V at 0.01 Hz. Electrical stimulation occurred viatwo parallel platinum electrodes incorporated into the tissue holder,connected to a Grass S88 stimulator (Grass Instruments). Afterexperimentation, prostates were removed from the organ baths, blot-ted dry, and weighed.

Muscarinic Receptor Characterization. After the equilibra-tion period the prostates from wild-type C57BL/6 mice were exposedto a priming dose of 1 mM carbamylcholine chloride (carbachol) toensure reproducible responses. Once a maximum contractile re-sponse was observed the tissue was washed four times with thevolume of the organ bath. After a 30-min recovery period an initialdiscrete half-log concentration response curve to carbachol (10nM-30 �M) was then constructed with 15-min intervals betweendrug additions, whereby for each concentration the response wasallowed to reach a maximum and plateau before the tissue waswashed out with four times the volume of the organ bath. To assessconfounding effects on potency by �1-adrenoceptors or cholinesteraseactivity, an appropriate time control curve to carbachol was con-structed in the presence or absence of prazosin (0.3 �M) or phys-ostigmine (10 �M). In subsequent experiments to prevent any con-founding effects by �1-adrenoceptors on characterization of themuscarinic receptor subtype, prazosin (0.3 �M) was added to theKrebs-Henseleit solution. To characterize the muscarinic subtypemediating contraction, after the initial concentration response curve,isolated prostates were exposed to a muscarinic receptor antagonistfor 1 h before a second concentration response curve to carbachol (10nM-1 mM) was constructed in the same manner. The muscarinicreceptor antagonist was replaced after each wash.

Effect of M3 Muscarinic Receptor Deletion on Nerve-Medi-ated Contraction. Frequency response curves were constructed inprostates taken from M3R(�/�), M3R(�/�), and M3R(�/�) miceafter the equilibration period. Prostates were exposed to electrical-field stimulation at frequencies of 0.1, 0.2, 0.5, 1, 2, 5, 10, and 20 Hz(0.5-ms duration, 60 V) delivered at 10-min intervals in trains ofpulses lasting 10 pulses or 10 s. An initial control frequency responsecurve was constructed to determine the contractile response of thetissue at each frequency. A second frequency response curve wasthen constructed after the prostate had been washed three timeswith the volume of the organ bath and exposed to atropine (1 �M) for1 h. When a third frequency response curve was constructed, theprostate was washed as described previously and exposed to atropine(1 �M) plus prazosin (0.3 �M). After the final frequency responsecurve the prostate was washed again and allowed to rest for 1 hbefore the Krebs-Henseleit solution was replaced with a high K� (80mM) Krebs-Henseleit solution to measure the nonspecific contractilepotential of the preparation.

M3 Muscarinic Receptors in the Mouse Prostate 871

at ASPE



ownloaded from


Effect of M3 Muscarinic Receptor Deletion on Acetylcho-line-Mediated Contraction. After the 1-h equilibration period, aninitial discrete concentration response curve to acetylcholine (10nM-1 mM), as described previously for carbachol, was constructed inprostates taken from M3R(�/�), M3R(�/�), and M3R(�/�) mice.After the initial concentration response curve, a second concentra-tion response curve was constructed in the same manner; however,the tissue was exposed to atropine (1 �M) for 1 h before the firstaddition of acetylcholine and was replaced after each wash. In aseparate experiment, using prostates taken from M3R(�/�) mice,the effect of mecamylamine (30 �M) on acetylcholine concentrationresponse curves (control, 100 nM-1 mM; in the presence of mecam-ylamine, 100 nM-10 mM) was observed.

Data Analysis. The peak force (g) elicited by electrical-field stim-ulation or the exogenous application of muscarinic receptor agonistswas measured at each frequency or concentration, in the presenceand absence of antagonists. Mean frequency and concentration re-sponse curves were constructed using Prism version 5.00 for Win-dows (GraphPad Software, Inc., San Diego, CA). The mean curveswere formed from the average of data from n experiments, where nis equal to the number of mice used. Results are expressed as mean �S.E.M.

For characterization studies, the contractile responses to carba-chol were expressed as a percentage of the maximum response of theinitial concentration response curve. The nonlinear regression func-tion of Prism version 5.00 was then used to fit an unconstrainedsigmoidal concentration response curve to the data. For each antag-onist, concentration ratios were determined by dividing the EC50

value of carbachol in the presence of the antagonist by the EC50

value in the absence of the antagonist. After which, Arunlakshana-Schild plots were constructed whereby log (concentration ratio-1)was plotted versus log (antagonist concentration) (Arunlakshanaand Schild, 1959). When the slope of the linear regression did notdiffer from unity, the slope was constrained to 1, and the pA2 valuewas calculated. Correlation and linear regression analysis was usedto determine the muscarinic receptor subtype mediating contractionof the mouse prostate. pA2 values of the antagonists at the musca-rinic receptors in the mouse prostate were compared against themean published antagonist pKD or pKi values at each of the fivemuscarinic receptor subtypes (Table 1). Ionic strength of assay bufferis known to influence the binding affinity of muscarinic receptorantagonists (Loury et al., 1999). As such, pKD or pKi values wereselected from Loury et al. (1999) and the International Union ofBasic and Clinical Pharmacology G protein-coupled receptor data-base only from studies using buffers of comparable ionic strength tothe Krebs-Henseleit solution used in this study. Where differentantagonist affinities were given for the same subtype, the meanvalue was used for the correlation analysis. Pearson correlationcoefficients (r) and associated p values, as well as the slope of linearregression, were calculated using Prism version 5.00.

In mean frequency and concentration response curves, differences

between the initial and subsequent drug-exposed curve were ana-lyzed by Prism version 5.00 using a two-way, repeated-measureanalysis of variance (ANOVA), followed by a Bonferroni post-testwhere required. The p values stated were used to evaluate thestatistical significance of any difference between the two curves andrepresent the probability of the drug treatment causing a change.p � 0.05 was considered significant. All receptor nomenclature con-forms to the International Union of Basic and Clinical Pharmacologynomenclature guidelines.

Reagents Used. (�)-Himbacine, acetylcholine chloride, atropinesulfate, carbamoylcholine chloride (carbachol), eserine hemisulfate salt(physostigmine), mecamylamine hydrochloride, methoctramine hemi-hydrate, and prazosin hydrochloride were obtained from Sigma-Aldrich(St. Louis, MO). 1,1-Dimethyl-4-diphenylacetoxypiperidinium iodide (4-DAMP), cyclohexyl-hydroxy-phenyl-(3-piperidin-1-ylpropyl)silane(HHSiD), and cyclohexyl-(4-fluorophenyl)-hydroxy-(3-piperidin-1-ylpropyl)silane (p-F-HHSiD) were obtained from Sigma/RBI (Natick,MA). Darifenacin hydrobromide was obtained from Santa Cruz Biotech-nology, Inc. (Santa Cruz, CA), and pirenzipine dihydrochloride wasfrom Boehringer Ingelheim (Artamon, Australia). All drugs were dis-solved and diluted in distilled water except HHSiD, which was dis-solved in 10% dimethyl sulfoxide; all further dilutions were made indistilled water.

ResultsPharmacological Characterization. In the isolated

wild-type mouse, prostate administration of exogenously ap-plied carbachol (10 nM-1 mM) produced reproducible concen-tration-dependant contractile responses with a �log (EC50)of 5.99 � 0.05 (n � 18; p � 0.95). The contractile response tocarbachol was unaffected by physostigmine (10 �M; Fig. 1A;p � 0.20); however, prazosin (0.3 �M; Fig. 1B; p � 0.05)caused a 3-fold rightward shift in the concentration responsecurve to carbachol.

The muscarinic receptor antagonists used were: 4-DAMP(3, 10, and 30 nM), atropine (6, 20, and 60 nM), p-F-HHSiD(0.3, 1, and 3 �M), HHSiD (0.1, 0.3, and 1 �M), darifenacin(Fig. 2; 10, 30, and 300 nM), himbacine (0.1, 0.3, and 1 �M),methoctramine (1, 3, and 10 �M), and pirenzipine (0.6, 1, and3 �M). All caused concentration-dependent parallel right-ward shifts in the concentration response curve to carbachol.The slopes and pA2 values calculated by Arunlakshana-Schild analysis are shown in Table 2. None of the antagonistsproduced a slope that was significantly different from unity(Fig. 3; p � 0.05). The rank order of antagonist affinities (pA2

values) at the muscarinic receptor mediating contraction inthe isolated mouse prostate (Table 2) was: darifenacin

TABLE 1pKD or pKi values reported for muscarinic receptor antagonists at the five muscarinic receptor subtypesValues shown are selected pKD (indicated by *) or pKi values. Where a range is given the mean value in parentheses was used for the correlation analysis. Only estimatesobtained from studies that used a physiological buffer are included here.

Antagonist M1 M2 M3 M4 M5

Atropinea 9.4, r 9.0, r 9.3, r 9.7, r 9.4, rDarifenacinb 7.8, h 7.0, h 8.8, h 7.7, h 8.0, h4-DAMPa 8.9, r 8.2, r 9.2, r 9.1, r 8.9, rHimbacinea 6.7, h 7.9, h 6.9, h 7.8, h 6.1, hMethoctraminea 6.7–7.0*, h (6.85) 7.3*, h 6.1–6.3*, h (6.2) 7.0*, h 6.3*, hPirenzepinea 7.8, h 6.0–6.4, h (6.2) 6.6, h 7.3, h 6.6, hp-F-HHSiDa 7.5, h 6.4, h 7.6, h 7.3, h 6.3, hHHSiDa 7.7, h 6.7–6.8, h (6.75) 7.7, h 7.7, h 6.8, h

r and h indicate studies using tissues containing muscarinic receptors taken from rats or humans, respectively.a Values from the International Union of Basic and Clinical Pharmacology database, http://www.iuphar-db.org/DATABASE/FamilyMenuForward?familyId�2, for

acetylcholine receptors (muscarinic).b Values from Loury et al. (1999).

872 White et al.

at ASPE



ownloaded from


(9.08) � atropine (9.07) � 4-DAMP (9.02) � HHSiD (7.85) �p-F-HHSiD (7.39) � himbacine (7.19) � pirenzipine (6.88) �methoctramine (6.20).

Correlation and linear regression analysis revealed signif-icant correlation and linear regression for the M3 (Fig. 4C;r � 0.98; p � 0.001; slope � 1.05 � 0.08) and M5 (Fig. 4E; r �0.88; p � 0.01; slope � 1.02 � 0.22) muscarinic receptorsubtypes; whereas weaker linear regressions and correla-

tions were observed for the M1 (Fig. 4A; r � 0.77; p � 0.05;slope � 0.64 � 0.21), M2 (Fig. 4B; r � 0.51; slope � 0.44 �0.30), and M4 (Fig. 4D; r � 0.78; p � 0.05; slope � 0.66 �0.22) subtypes.

Genetic Confirmation. Deletion of the M3 muscarinicreceptor resulted in a reduction in mouse weight (Fig. 5A; p �0.001); however, no effect on prostate weight was observed(Fig. 5B). Furthermore, the contractile response to high K�

(80 mM) Krebs-Henseleit was reduced in prostates takenfrom both M3R(�/�) and M3R(�/�) mice (Fig. 5C; p � 0.01).

In prostates taken from M3R(�/�), M3R(�/�), andM3R(�/�) mice, electrical-field stimulation (0.1–20 Hz) elic-ited frequency-dependent contractions. Contractile responseswere reduced in prostates taken from M3R(�/�) mice but notM3R(�/�) mice compared with prostates taken fromM3R(�/�) mice (Fig. 6). A maximum inhibition of contrac-tion of 76% for M3R(�/�) mice compared with M3R(�/�)mice was observed at 2 Hz. Furthermore, atropine (1 �M)

Fig. 1. Left, mean log concentration response curves to carbachol (Cch) inthe isolated mouse prostate expressed as a percentage of the maximumresponse in the initial curve before and after the administration of phys-ostigmine (10 �M) (A) or prazosin (0.3 �M) (B). Points represent meanforce � S.E.M. (n � 6). p values determined by a two-way, repeated-measures of ANOVA represent the probability of a significant interactionbetween concentration and treatment. Right, histograms represent themean � S.E.M. (n � 6 maximum contractile response in the absence(filled bars) or presence (open bars) of physostigmine (10 �M) (A) orprazosin (0.3 �M) (B).

Fig. 2. Left, representative graph showing the contractileresponse of the isolated mouse prostate to carbachol (Cch)in the absence and presence of increasing concentrations ofthe muscarinic receptor antagonist darifenacin (10, 30, and100 nM). Responses are expressed as a percentage of themean � S.E.M. (n � 6) maximum contractile response ofcarbachol in the absence of antagonist. Right, histogramrepresents the mean � S.E.M. (n � 6) maximum contrac-tile response in the absence of darifenacin (open bar) andthe presence of 10 nM darifenacin (gray bar), 30 nM dar-ifenacin (black bar), and 100 nM darifenacin (crosshatchedgray bar).

TABLE 2pA2 values and slope factors (slope � S.E.M.) calculated fromcharacterization studies in the mouse prostate gland by the range ofmuscarinic antagonists used to carbachol

Antagonist Slope � S.E.M. pA2� S.E.M.

Atropine 0.92 � 0.04 9.07 � 0.02Darifenacin 1.02 � 0.24 9.08 � 0.074-DAMP 1.01 � 0.35 9.02 � 0.11Himbacine 0.98 � 0.06 7.19 � 0.02Methoctramine 0.95 � 0.18 6.20 � 0.06Pirenzepine 1.25 � 0.38 6.88 � 0.09p-F-HHSiD 1.16 � 0.05 7.39 � 0.05HHSiD 0.91 � 0.21 7.85 � 0.07

Fig. 3. Schild analysis of the muscarinic receptor antagonists atropine,pirenzepine, 4-DAMP, himbacine, methoctramine, HHSiD, p-f-HHSiD,and darifenacin. Each point represents the mean � S.E.M. (n � 6). Slopesof the linear regression and pA2 values are shown in Table 2.


at ASPE



ownloaded from


attenuated electrical-field stimulation-mediated contractionin prostates taken from both M3R(�/�) (Fig. 7A; p � 0.001)and M3R(�/�) mice (Fig. 7B; p � 0.001) with a maximuminhibition of contraction observed at 5 Hz of 76 and 70%,respectively. However, atropine (Fig. 7C; p � 0.54; 1 �M) hadno effect on the magnitude of contraction in prostates takenfrom M3R(�/�) mice, whereas prazosin (Fig. 7C; p � 0.001;0.3 �M) abolished the residual noncholinergic contraction.

The administration of the endogenous agonist acetylcholine(10 nM-100 �M) elicited concentration-dependent contractileresponses (Fig. 8A) that were inhibited by atropine (1 �M) inprostates taken from M3R(�/�) and M3R(�/�) mice. A slightlydecreased potency for acetylcholine was observed in prostatestaken from M3R(�/�) mice [Fig. 8A; �log(EC50) � 6.06; 95%confidence intervals 5.80–6.37] compared with M3R(�/�) mice[Fig. 8A; �log(EC50) � 6.46; 95% confidence intervals 6.31–6.68]. The maximum contractile response was unchanged (Fig.8A; p � 0.83). In prostates taken from M3R(�/�) miceacetylcholine elicited a small contractile response at con-centrations higher than 10 �M (Fig. 8A). This responsewas unaffected by the addition of atropine (Fig. 8B; p �0.47; 1 �M), but was abolished by mecamylamine (Fig. 8C;p � 0.001; 30 �M).

DiscussionWe have previously identified that in the mouse prostate

gland acetylcholine acting at muscarinic receptors mediatesa nonadrenergic contractile response; however, the musca-rinic receptor subtype mediating this response was not elu-cidated. Furthermore, it has been reported that both the M1

and M3 muscarinic receptor subtypes are present in themouse prostate (Ito et al., 2009); however, their locations andfunctions are unknown. This study used pharmacological andgenetic techniques to examine the muscarinic receptor sub-type responsible for the functional contractile response.

The contractile response to carbachol was unaffected bycholinesterase activity and was therefore seen as more suit-able than acetylcholine for use in the pharmacological char-acterization study because it eliminated this confoundingfactor. However, the response to carbachol was inhibited bythe �1-adrenoceptor antagonist prazosin, suggesting thatcarbachol facilitates the adrenergic response. This is poten-tially caused by the release of noradrenaline mediated byprejunctional muscarinic receptors on adrenergic nerves ornicotinic receptor activation of postganglionic adrenergicnerves. In the guinea pig prostate, cholinergic agonists facil-

Fig. 4. Correlation of the pA2 values observed for the seven muscarinic receptor antagonists (x-axis) with the published (Table 1, y-axis) muscarinicreceptor antagonist pKi values at the M1 (A), M2 (B), M3 (C), M4 (D), and M5 (E) muscarinic receptor subtypes. Solid lines represent the linearregression analyses, and broken lines represent a line with a slope of 1. Also indicated are the slope � S.E.M of linear regression and the Pearsonscoefficient (r), calculated by correlation analysis.

Fig. 5. Comparison of mean mouseweights (wt) (A), isolated prostateweights (B), and contractile responsesto 80 mM K� Krebs-Henseleit solu-tion (C) from wild-type [M3R(�/�),open bars], M3 muscarinic receptorheterozygous mice [M3R(�/�), graybars], and M3 muscarinic receptorknockout mice [M3R(�/�), blackbars]. Bars represent mean weight/contractile force � S.E.M. (n � 6). ��,p � 0.01; ��, p � 0.001 versus controlcalculated by an unpaired t test.

874 White et al.

at ASPE



ownloaded from


itate nerve-mediated contraction hypothesized to be medi-ated by noradrenaline release by prejunctional muscarinicreceptors (Lau et al., 2000). In contrast, in both the human(Hedlund et al., 1985) and canine (Arver and Sjostrand, 1982)prostate, muscarinic receptors produce a slight reduction innoradrenaline release. Likewise, sympathetic neurotrans-mitter release is inhibited in the mouse atria, urinary blad-der, and vas deferens by muscarinic receptors (Trendelen-burg et al., 2005). In the present study, atropine had nodetectable effect on the contractile response to high concen-trations of acetylcholine in prostates taken from M3R(�/�)mice, whereas the noncompetitive nicotinic antagonistmecamylamine abolished the contractile response. This sug-gests that postganglionic nicotinic receptors rather thanprejunctional muscarinic receptors mediate the release ofneuronal noradrenaline.

Correlation plots show that the antagonist affinities ob-tained in the pharmacological characterization studies corre-late most noticeably with the binding affinities of the antag-onists at the M3 muscarinic receptor subtype. Progressivelypoorer correlations were observed for the M5, M4, M1, and M2

muscarinic receptor subtypes. The slope of the regressionanalysis was also used to determine the muscarinic receptorsubtype, with a slope of the line closest to unity (x � y)considered to be the best correlation. The best regressionanalysis was observed for the M3 and M5 muscarinic receptorsubtypes. Taken together, these results are indicative of theM3 muscarinic receptor subtype mediating contraction in themouse prostate gland. These results are consistent with find-ings in the rat prostate, where the M3 muscarinic receptorsubtype mediates contraction (Lau and Pennefather, 1998)and is localized in the stromal tissue (Nadelhaft, 2003). How-ever, this is in contrast to findings in the human (Yazawa etal., 1994) and canine (Fernandez et al., 1998) prostates whereM2 muscarinic receptors are thought to mediate contractionand in the guinea pig prostate where M1 muscarinic recep-tors facilitate contraction (Lau et al., 2000).

It should be noted that a good correlation, with poor linearregression, was also observed for the M1 subtype. Given that a

population of M1 receptors has previously been reported in themouse prostate (Ito et al., 2009) this is intriguing. At the M1

muscarinic receptor the binding affinities for all of the antago-

Fig. 6. Comparison of mean contractile responses to electrical-field stim-ulation (0.5 ms, 60 V, 0.1–20 Hz, 10-s pulses) in prostates taken fromwild-type mice [M3R(�/�), open bars], M3 muscarinic receptor heterozy-gous mice [M3R(�/�), gray bars], and M3 muscarinic receptor knockoutmice [M3R(�/�), black bars]. Bars represent mean force � S.E.M. (n �6). p values determined by a two-way, repeated-measures of ANOVArepresent the probability of the M3 muscarinic receptor deletion causinga significant change in response.

Fig. 7. Mean contractile responses to electrical-field stimulation (0.5 ms, 60 V,1–20 Hz, 10-s pulses) in prostates taken from wild-type mice [M3R(�/�)](A), M3 muscarinic receptor heterozygous mice [M3R(�/�)] (B), and M3 mus-carinic receptor knockout mice [M3R(�/�)] (C) in the absence (open bars) orpresence of atropine (1 �M) (closed bars) (A–C) as well as atropine (1 �M) plusprazosin (0.3 �M) (gray bars) (C). Bars represent mean force � S.E.M. (n � 6).p values determined by a two-way, repeated-measures of ANOVA represent theprobability of the drug treatment causing a significant change.


at ASPE



ownloaded from


nists used, except darifenacin and pirenzipine, are close to thebinding affinities for the M3 muscarinic receptor subtype(http://www.iuphar-db.org/DATABASE/FamilyMenuForward?familyId�2). Therefore darifenacin and pirenzipine impart thegreatest strength when differentiating between M1 and M3

muscarinic receptors. The binding affinity for darifenacin(pA2 � 9.08) exclusively suggests the M3 subtype over the M1.For pirenzipine, however, the binding affinity (pA2 � 6.88) fallswithin the range for the M3 subtype and somewhat within therange of the M1 subtype. The Schild slope (Table 2) for piren-zipine is steeper than what is expected of a truly competitiveantagonist, albeit not significantly. The lack of statistical sig-nificance in this difference may be the result of a broad S.E.M.(� 0.38) confounding the analysis and can be explained by theconcentration range of antagonist used (0.6–3 �M) being insuf-ficiently large (Kenakin, 2009). When the binding affinity iscalculated as a pKB value, instead of a pA2 value using theequation described by Furchgott (1972), the binding affinityobtained is 5.89, which is well within the range of the M3 butnot the M1 muscarinic receptor subtype.

Likewise, good linear regression analysis was observed forboth the M3 and M5 subtypes. This is unsurprising given thedearth of muscarinic receptor antagonists capable of distin-guishing between M3 and M5 muscarinic receptors. Giventhat Ito et al. (2009) failed to find a significant population ofM5 receptors in the mouse prostate gland, in addition to theresults presented here using M3 muscarinic receptor knock-out mice, it seems unlikely that M5 muscarinic receptorsmediate contraction.

M3R(�/�) mice were used to confirm the muscarinic receptorsubtype mediating the cholinergic contractile response in themouse prostate gland. However, a significant difference in theresponse to 80 mM K� Krebs-Henseleit solution between geno-types was also observed. This result is interesting and unex-plained; however, it may suggest that endogenously releasedacetylcholine acting at M3 muscarinic receptors contributes tothe contractile response mediated by 80 mM K� Krebs-Henseleit solution. Equally, it may suggest that the deletion ofthe M3 muscarinic receptor affects the contractile machinery inthe mouse prostate gland. Nevertheless, to allow fair compari-

son between genotypes, results to electrical-field stimulationwere plotted as a percentage of the contractile response to 80mM K� Krebs-Henseleit solution.

In prostates taken from M3R(�/�) mice, we observed thatthe nerve-mediated contractile response was inhibited. Fur-thermore, atropine was without effect in M3R(�/�) mice.These observations suggest that the M3 muscarinic receptorsubtype is solely responsible for the cholinergic component ofnerve-mediated contraction and that compensatory changesby up-regulation of other muscarinic receptor subtypes do notoccur in prostates taken from M3R(�/�) mice. This does notexclude the possibility of up-regulation of noncholinergicmechanisms such as purinergic mechanisms, as observed inthe bladders of M3R(�/�) mice (Igawa et al., 2004). However,this seems unlikely in the prostate because the residualcontractile response to electrical-field stimulation in pros-tates taken from M3R(�/�) mice is abolished by the �1-adrenoceptor antagonist prazosin. Furthermore this con-firms that the nerve-mediated contractile response in themouse prostate is mediated by both muscarinic receptors and�1L-adrenoceptors (White et al., 2010).

The decrease in the potency of acetylcholine observed inprostates taken from M3 muscarinic receptor heterozygousmice may be a product of fewer M3 muscarinic receptorsbeing available, whereas the response to acetylcholine inprostates taken from M3R(�/�) mice was comparable withour previous results (White et al., 2010). The absence of acontractile response to acetylcholine, except at high concen-trations (as discussed previously) in M3R(�/�) mice, furtherconfirms the M3 muscarinic receptor as the subtype mediat-ing contraction in the mouse prostate gland.

The weight of the prostate was unaffected by the deletionof the M3 muscarinic receptor, indicating the M3 muscarinicreceptor and/or gene does not play a role in growth of themouse prostate gland. In primary human cell cultures andprostate cancer cell lines (Witte et al., 2008), as well as in therat prostate (McVary et al., 1998), growth can be regulated bycholinergic agonists or parasympathetic innervation, respec-tively. Although the results of the present study exclude arole for M3 muscarinic receptors in regulating growth, they

Fig. 8. A, comparison of meancontractile response to ace-tylcholine (Ach) in prostatestaken from wild-type mice[M3R(�/�), F], M3 musca-rinic receptor heterozygousmice [M3R(�/�), f], and M3muscarinic receptor knockoutmice [M3R(�/�), Œ]. B andC, mean log concentration re-sponse curves to acetylcholinein prostates taken from M3muscarinic receptor knockoutmice [M3R(�/�)] before andafter the administration of at-ropine (1 �M) (B) or mecam-ylamine (30 �M) (C). Pointsrepresent mean force � S.E.M.(n � 4–6). p values determinedby a two-way, repeated-mea-sures of ANOVA represent theprobability of the M3 muscarinicreceptor deletion, or test drug,causing a significant change inresponse.

876 White et al.

at ASPE



ownloaded from


can not exclude the possibility of additional muscarinic re-ceptor subtypes regulating mouse prostate growth. It shouldbe noted that the reduction in mouse weight at this age inM3R(�/�) mice, caused by an impairment of saliva produc-tion (Matsui et al., 2000) or appetite (Yamada et al., 2001),may have influenced this result.

Unlike in the mouse prostate, the cholinergic component ofnerve-mediated contraction in the human prostate is smallcompared with the adrenergic response (Caine et al., 1975;Hedlund et al., 1985; Gup et al., 1989; Kester et al., 2003).However, another study has observed an adreno-muscarinicsynergy in contraction of the human prostate (Roosen et al.,2009). Although the M3 muscarinic receptor subtype has notbeen demonstrated in the human prostate, recent reviews ofclinical trials using muscarinic receptor antagonists (whichincluded the M3 muscarinic receptor antagonists solifenacinand darifenacin) for the treatment of lower urinary tractsymptoms, usually in combination with an adrenoceptor an-tagonist, show varying improvement in storage symptomswithout worsening of voiding symptoms or significant devel-opment of acute urinary retention (Athanasopoulos, 2010;Chapple, 2010). Although these drugs may act by inhibitionof detrusor instability in the bladder, prostatic smooth mus-cle tone can not be ruled out as an additional site of action forthe treatment of lower urinary tract symptoms by muscarinicreceptor antagonists.

In conclusion, the results of the present study have shownthat the M3 muscarinic receptor subtype is solely responsiblefor the nerve-mediated cholinergic component of contractionin the mouse prostate gland.

Authorship Contributions

Participated in research design: White, Short, and Ventura.Conducted experiments: White.Contributed new reagents or analytic tools: Matsui.Performed data analysis: White and Ventura.Wrote or contributed to the writing of the manuscript: White,

Short, Haynes, and Ventura.

ReferencesAnisuzzaman AS, Morishima S, Suzuki F, Tanaka T, Yoshiki H, Sathi ZS, Akino H,

Yokoyama O, and Muramatsu I (2008) Assessment of muscarinic receptor sub-types in human and rat lower urinary tract by tissue segment binding assay.J Pharmacol Sci 106:271–279.

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

Arver S and Sjostrand NO (1982) Functions of adrenergic and cholinergic nerves incanine effectors of seminal emission. Acta Physiol Scand 115:67–77.

Athanasopoulos A (2010) Antimuscarinics and bladder outlet obstruction: From acontraindication to an indication? Neurourol Urodyn 29(Suppl 1):S46–S50.

Caine M, Raz S, and Zeigler M (1975) Adrenergic and cholinergic receptors in thehuman prostate, prostatic capsule and bladder neck. Br J Urol 47:193–202.

Chapple C (2010) Antimuscarinics in men with lower urinary tract symptoms sug-gestive of bladder outlet obstruction due to benign prostatic hyperplasia. CurrOpin Urol 20:43–48.

Fernandez JL, Rivera L, Lopez PG, Recio P, Vela-Navarrete R, and García-SacristanA (1998) Characterization of the muscarinic receptor mediating contraction of thedog prostate. J Auton Pharmacol 18:205–211.

Furchgott RF (1972) The classification of adrenoceptors (adrenergic receptors). Anevaluation from the standpoint of receptor theory, in Catecholamines, Handbook ofExperimental Pharmacology (Blaschko H and Muscholl E eds) Vol 33, pp 283–335,Springer, Berlin.

Gray KT and Ventura S (2006) �1L-adrenoceptors mediate contractions of theisolated mouse prostate. Eur J Pharmacol 540:155–161.

Gup DI, Shapiro E, Baumann M, and Lepor H (1989) Contractile properties ofhuman prostate adenomas and the development of infravesical obstruction. Pros-tate 15:105–114.

Hedlund H, Andersson KE, and Larsson B (1985) �-Adrenoceptors and muscarinicreceptors in the isolated human prostate. J Urol 134:1291–1298.

Igawa Y, Zhang X, Nishizawa O, Umeda M, Iwata A, Taketo MM, Manabe T, MatsuiM, and Andersson KE (2004) Cystometric findings in mice lacking muscarinic M2or M3 receptors. J Urol 172:2460–2464.

Ito Y, Oyunzul L, Seki M, Fujino Oki T, Matsui M, and Yamada S (2009) Quantita-tive analysis of the loss of muscarinic receptors in various peripheral tissues inM1–M5 receptor single knockout mice. Br J Pharmacol 156:1147–1153.

Jositsch G, Papadakis T, Haberberger RV, Wolff M, Wess J, and Kummer W (2009)Suitability of muscarinic acetylcholine receptor antibodies for immunohistochem-istry evaluated on tissue sections of receptor gene-deficient mice. Naunyn Schmie-debergs Arch Pharmacol 379:389–395.

Kenakin TP (2009) Orthosteric drug antagonism, in A Pharmacology Primer, 3rd ed,pp 101–127, Academic Press, New York.

Kester RR, Mooppan UM, Gousse AE, Alver JE, Gintautas J, Gulmi FA, Abadir AR,and Kim H (2003) Pharmacological characterization of isolated human prostate.J Urol 170:1032–1038.

Lau WA and Pennefather JN (1998) Muscarinic receptor subtypes in the rat prostategland. Eur J Pharmacol 343:151–156.

Lau WA, Pennefather JN, and Mitchelson FJ (2000) Cholinergic facilitation ofneurotransmission to the smooth muscle of the guinea-pig prostate gland. Br JPharmacol 130:1013–1020.

Loury DN, Hegde SS, Bonhaus DW, and Eglen RM (1999) Ionic strength of assaybuffers influences antagonist binding affinity estimates at muscarinic M1–M5cholinoceptors. Life Sci 64:557.

Matsui M, Motomura D, Karasawa H, Fujikawa T, Jiang J, Komiya Y, Takahashi S,and Taketo MM (2000) Multiple functional defects in peripheral autonomic organsin mice lacking muscarinic acetylcholine receptor gene for the M3 subtype. ProcNatl Acad Sci U S A 97:9579–9584.

McVary KT, McKenna KE, and Lee C (1998) Prostate innervation. Prostate Suppl8:2–13.

Michel MC, Wieland T, and Tsujimoto G (2009) How reliable are G-protein-coupledreceptor antibodies? Naunyn Schmiedebergs Arch Pharmacol 379:385–388.

Nadelhaft I (2003) Cholinergic axons in the rat prostate and neurons in the pelvicganglion. Brain Res 989:52–57.

Obara K, Arai K, Miyajima N, Hatano A, Tomita Y, and Takahashi K (2000)Expression of m2 muscarinic acetylcholine receptor mRNA in primary culture ofhuman prostate stromal cells. Urol Res 28:196–200.

Oki T, Sato S, Miyata K, and Yamada S (2005) Muscarinic receptor binding, plasmaconcentration and inhibition of salivation after oral administration of a novelantimuscarinic agent, solifenacin succinate in mice. Br J Pharmacol 145:219–227.

Pradidarcheep W, Stallen J, Labruyere WT, Dabhoiwala NF, Michel MC, and Lam-ers WH (2009) Lack of specificity of commercially available antisera againstmuscarinergic and adrenergic receptors. Naunyn Schmiedebergs Arch Pharmacol379:397–402.

Roosen A, Blake-James BT, Wood D, and Fry CH (2009) Clinical and experimentalaspects of Adreno-muscarinic synergy in the bladder base and prostate. NeurourolUrodyn 28:938–943.

Ruggieri MR, Colton MD, Wang P, Wang J, Smyth RJ, Pontari MA, and Luthin GR(1995) Human prostate muscarinic receptor subtypes. J Pharmacol Exp Ther274:976–982.

Trendelenburg AU, Meyer A, Wess J, and Starke K (2005) Distinct mixtures ofmuscarinic receptor subtypes mediate inhibition of noradrenaline release in dif-ferent mouse peripheral tissues, as studied with receptor knockout mice. Br JPharmacol 145:1153–1159.

Ventura S, Pennefather J, and Mitchelson F (2002) Cholinergic innervation andfunction in the prostate gland. Pharmacol Ther 94:93–112.

White CW, Short JL, Haynes JM, Evans RJ, and Ventura S (2010) The residualnonadrenergic contractile response to nerve stimulation of the mouse prostate ismediated by acetylcholine but not ATP in a comparison with the mouse vasdeferens. J Pharmacol Exp Ther 335:489–496.

Witte LP, Chapple CR, de la Rosette JJ, and Michel MC (2008) Cholinergic inner-vation and muscarinic receptors in the human prostate. Eur Urol 54:326–334.

Yamada M, Miyakawa T, Duttaroy A, Yamanaka A, Moriguchi T, Makita R, OgawaM, Chou CJ, Xia B, Crawley JN, et al. (2001) Mice lacking the M3 muscarinicacetylcholine receptor are hypophagic and lean. Nature 410:207–212.

Yamada S, Maruyama S, Takagi Y, Uchida S, and Oki T (2006) In vivo demonstra-tion of M3 muscarinic receptor subtype selectivity of darifenacin in mice. Life Sci80:127–132.

Yazawa H, Saita Y, Iida E, Honma Y, Morita T, and Honda K (1994) Characteriza-tion of muscarinic cholinoceptor in primary culture of smooth muscle cells fromhuman prostate. J Urol 152:2173–2177.

Address correspondence to: Dr. Sabatino Ventura, Medicinal Chemistryand Drug Action, Monash Institute of Pharmaceutical Sciences, Monash Uni-versity, 381 Royal Parade, Parkville, Victoria 3052, Australia. E-mail:[email protected]


at ASPE



ownloaded from


contractions of the mouse prostate elicited by...

Documents