0022-3565/06/3182-649–656$20.00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2006 by The American Society for Pharmacology and Experimental Therapeutics
JPET 318:649–656, 2006
Vol. 318, No. 2
103093/3126006
Printed in U.S.A.
Differential Coupling of Muscarinic M1, M2, and M3 Receptors
to Phosphoinositide Hydrolysis in Urinary Bladder and
Longitudinal Muscle of the Ileum of the Mouse
John A. Tran, Minoru Matsui, and Frederick J. Ehlert
Department of Pharmacology, College of Medicine, University of California, Irvine, California (J.A.T., F.J.E.); and Division of
Neuronal Network, Department of Basic Medical Sciences, the Institute of Medical Science, the University of Tokyo, Minato-ku,
Tokyo, Japan (M.M.)
Received February 16, 2006; accepted May 3, 2006
Muscarinic receptors (M1–M5) have many vital functions
throughout the central and peripheral nervous system (Caulfield, 1993). With regard to smooth muscle, a heterogeneous
mixture of M2 and M3 receptors is expressed, with the M2
outnumbering the M3 by at least 4-fold (Ehlert et al., 1997).
The M3 receptor is known to mediate direct contraction of the
ileum, urinary bladder, and other smooth muscle-containing
tissues (Sawyer and Ehlert, 1998; Choppin and Eglen, 2001).
In contrast, the M2 receptor mediates contraction through
indirect mechanisms, including an inhibition of relaxation
(Thomas et al., 1993; Matsui et al., 2003) and an enhance-
This work was supported by Pfizer (to F.J.E.), National Institutes of Health
Grant NS 30882 (to F.J.E.), a grant from Pharmacia (to M.M.) and the Detrol
LA Research Grant Program from Pfizer (to M.M.), the Industrial Technology
Research Grant 02A09001a from The New Energy and Industrial Technology
Development Organization (NEDO) of Japan (to M.M.), and Grant-in-aid for
Scientific Research on Priority Areas 16067101 from The Ministry of Education, Culture, Sports, Science and Technology (MEXT) (to M.M.).
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.106.103093.
0.37 and 0.52 ␮M and Emax values of 35.8 and 34.7%, respectively. Oxotremorine-M also elicited phosphoinositide hydrolysis in ilea from M3 and M2/M3 knockout mice, although these
responses were less sensitive (EC50 values of 1.6 and 1.4 ␮M;
Emax values of 31.2 and 20.8%, respectively). The response in
ileum from the M2/M3 knockout was significantly smaller than
that from the M3 knockout. The muscarinic phosphoinositide
response in ilea from M2/M3 knockout mice originated in the
smooth muscle and exhibited a profile for competitive antagonism consistent with an M1 mechanism. These data suggest a
major role for the M3 receptor in eliciting phosphoinositide
hydrolysis in the ileum and urinary bladder and minor roles for
the M1 and M2 in ileum.
ment in M3 receptor-mediated contraction (Sawyer and Ehlert, 1998; Ehlert et al., 2005).
Muscarinic receptors belong to the G protein-coupled receptor family, which elicit cellular responses by interacting
with heterotrimeric G proteins (Stryer and Bourne, 1986).
When transfected into cell lines, the M1, M3, and M5 receptors couple to Gq to mediate phosphoinositide hydrolysis,
whereas the M2 and M4 receptors couple to Gi/o to mediate an
inhibition of adenylyl cyclase (Bonner et al., 1988; Peralta et
al., 1988). A similar picture emerges from pharmacological
studies on smooth muscle. The muscarinic phosphoinositide
response is antagonized in a manner consistent with an M3
receptor mechanism (Noronha-Blob et al., 1989; Candell et
al., 1990), whereas muscarinic inhibition of adenylyl cyclase
exhibits an M2 profile in competitive antagonism studies
(Noronha-Blob et al., 1989; Candell et al., 1990). The coupling of the M3 receptor to phosphoinositide hydrolysis in
smooth muscle is consistent with its direct role in mediating
contraction, presumably through mobilization of Ca2⫹. Likewise, the coupling of M2 receptors to inhibition of adenylyl
cyclase is consistent with its role in mediating an inhibition
ABBREVIATIONS: M, muscarinic receptor; KO, knockout; KRB, Krebs’ Ringer Bicarbonate; AF-DX 116, [[2-[(diethylamino) methyl]-1-piperidinyl]acetyl]-5,11-dihydro-6H-pyrido[2,3b][1,4]-benzodiazepine-6-one; [3H]NMS, [3H]N-methylscopolamine; PLC-␤, phospholipase-C␤.
649
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
ABSTRACT
We investigated the coupling of muscarinic receptor (M) subtypes to phosphoinositide hydrolysis in ileum and urinary
bladder using muscarinic receptor knockout mice. In urinary
bladder from wild-type mice, the muscarinic agonist oxotremorine-M, elicited a robust phosphoinositide response characterized by an EC50 value of 0.22 ␮M and a maximal response
(Emax) of 32.8% conversion of [3H]inositol-labeled phosphoinositides into [3H]inositol phosphates. A similar response was
observed in urinary bladder from M2 knockout mice, whereas
no measurable response was observed in urinary bladder from
M3 and M2/M3 knockout mice. In ilea from wild-type and M2
knockout mice, substantial phosphoinositide responses to oxotremorine-M were measured, characterized by EC50 values of
650
Tran et al.
Materials and Methods
Mice. M2 receptor knockout (M2–/–; M2 KO), M3 receptor knockout
(M3–/–; M3 KO), and M2/M3 receptor double knockout (M2–/– and
M3–/–; M2/M3 KO) mice were generated as described in Matsui et al.
(2000, 2002). These hybrid lines were backcrossed with C57BL/6
mice to yield an N4 generation of M2 KO mice, an N8 generation of
M3 KO mice, and an N2 generation of M2/M3 KO mice. Only male
knockout mice were used as well as male wild-type (M2⫹/⫹, M3⫹/⫹)
C57BL/6 mice.
Phosphoinositide Hydrolysis. Our method is based on the
[3H]inositol-labeling method of Berridge et al. (1982) and the tissue
extraction method of Kendall and Hill (1990). Segments of ileum (10
cm) from one to two mice were removed and washed with Krebs’
Ringer bicarbonate buffer (KRB; 124 mM NaCl, 5 mM KCl, 1.3 mM
MgCl2, 26 mM NaHCO3, 1.2 mM KH2PO4, 1.8 mM CaCl2, and 10 mM
glucose). The longitudinal muscle was isolated by rubbing off the
outer layer of the ileum with a cotton swab. The muscle was cut into
small strips (0.5 cm). The urinary bladders from two mice were
removed, washed with KRB buffer, and trimmed of excess fat. Each
bladder was cut into nine to 10 strips. The pooled strips of tissue
from ileum and urinary bladder were incubated separately in 5 ml of
KRB buffer containing myo-[3H]inositol (100 ␮Ci) in a 50-ml Erlenmeyer flask. The flask was gassed with O2/CO2 (19:1), capped with a
rubber stopper, and incubated at 37°C for 90 min. The strips were
washed with fresh KRB buffer and put into conical tubes containing
KRB buffer, 10 mM LiCl, and 0.1 ␮M tetrodotoxin in a final volume
of 0.3 ml. One strip of tissue was added to each tube, except in initial
experiments on the ileum where two strips were used. The reaction
was started with the addition of oxotremorine-M, and the mixture
was gassed with O2/CO2, capped with a rubber stopper, and incubated at 37°C for 30 min. The reaction was stopped with 5% perchloric acid (200 ␮l), and the tubes were placed on ice. [3H]Inositol
phosphates were isolated as described previously (Ehlert et al.,
2001).
When the competitive effects of pirenzepine (0.2 ␮M) and AF-DX
116 (2 ␮M; [[2-[(diethylamino) methyl]-1-piperidinyl]acetyl]-5,11-dihydro-6H-pyrido[2,3b][1,4]-benzodiazepine-6-one) were investigated, these antagonists were preincubated with the oxotremorine-M in the absence of LiCl for 30 min at 37°C to allow
equilibration. LiCl (final concentration of 10 mM) was added, and the
incubation was allowed to proceed for 30 min at 37°C. The reaction
was then stopped with 5% perchloric acid (200 ␮l) as described above.
Radioligand Binding. The binding of the muscarinic antagonist
[3H]N-methylscopolamine ([3H]NMS) was measured as described in
Thomas and Ehlert (1994). The urinary bladder and longitudinal
muscle of the ileum were isolated as described under Phosphoinositide Hydrolysis. The tissue was minced with scissors and homogenized with a Polytron homogenizer at setting 5 in 20 ml of a binding
buffer containing 100 mM NaCl, 30 mM NaHEPES, pH 7.4, 0.5 mM
EGTA, and 0.5 mM dithiothreitol The homogenates were centrifuged
30,000g for 10 min. The pellet was resuspended to a concentration of
5 mg (original wet weight/milliliter in fresh binding buffer). An
aliquot (0.7 ml) of the tissue homogenate was added to glass tubes
containing the binding buffer, 0.5 nM [3H]NMS, and other drugs in
a final volume of 1.0 ml. [3H]NMS bound to the membrane of the
homogenates was trapped on glass-fiber filters (Whatman Inc.,
Clifton, NJ) by vacuum filtration on a cell harvester (Brandel, Gaithersburg, MD). The filters were washed four times with aliquots (4 ml)
of ice-cold saline. Nonspecific binding was defined as binding in the
presence of atropine (10 ␮M). When hypotonic buffer was used, it
contained 1.5 mM MgSO4, 20 mM HEPES/Tris, pH 7.5, and 0.5 mM
dithiothreitol.
In Vivo Pertussis Toxin Treatment. C57BL/6 male mice
(20 – 40 g) were injected i.p. with a dose of pertussis toxin (isletactivating protein; 70 –100 ␮g/kg body weight) 3 days prior to experimentation.
Dissociated Smooth Muscle. Ileal longitudinal muscle was dissociated by a method similar to Yang et al. (1991), with adaptations.
The longitudinal muscle was obtained as described above. The strips
were incubated in Medium 199 (5 ml; Invitrogen, Carlsbad, CA)
containing 0.2% collagenase II and 0.01% DNase I in a 50-ml centrifuge tube for 90 min at 37°C and gassed with O2/CO2 (19:1). The
tissue was also incubated with myo-[3H]inositol (100 ␮Ci) for phosphoinositide hydrolysis assays. The tissue was triturated with a
disposable Pasteur pipette to release smooth muscle cells (⬃5 min).
The mixture was filtered through a 500-␮m Nitex mesh to remove
connective tissue and undissociated tissue. The dissociated muscle
cells were washed twice by centrifugation at 300g for 5 min. The cells
were incubated in enzyme-free KRB for 1 h at 37°C between the first
and second centrifugation. Afterward, the cells were suspended in a
volume of 2 ml, and aliquots (0.3 ml) of the cell suspension were
added to assay tubes. An aliquot (3 ␮l) of LiCl (1 M) was added
followed by an aliquot (3 ␮l) of oxotremorine-M (10 mM), and phosphoinositide hydrolysis was measured as described earlier.
Analysis of the Loss of Function in Knockout Mice. We
developed a simple method to estimate the contribution of muscarinic receptor subtypes to the phosphoinositide hydrolysis in smooth
muscle. This analysis is based on the assumption that there is no
compensatory change in responses elicited by residual muscarinic
receptors in tissue from knockout mice. We observed that the phosphoinositide response to oxotremorine-M exhibited a logistic concentration-response curve with a Hill slope similar to one in ileum and
urinary bladder. Consequently, we used the operational model to
describe the stimulus-response function (Black and Leff, 1983):
Response ⫽
SMsys
S ⫹ KE
(1)
in which S denotes the total stimulus, Msys denotes the maximal
response of the system, and KE denotes a constant representing the
sensitivity of the stimulus-response function. We assume that the
degree of receptor activation is proportional to the stimulus as defined by Furchgott (1966). The total stimulus is equivalent to the
summation of the stimuli generated by each receptor contributing to
the response and is given by,
冘
n
S⫽
i⫽1
X␧ iRi
X ⫹ Ki
(2)
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
of the relaxant effect of agents that increase cAMP levels
(Thomas et al., 1993; Sawyer and Ehlert, 1998). Activation of
M2 receptor has also been shown to mediate an increase in a
nonselective cationic conductance (Bolton, 1979) and an inhibition of Ca2⫹-activated K⫹ channels (Cole et al., 1989),
which might explain how M2 receptors mediate an enhancement in contractions elicited by the M3 receptor.
In most prior studies on muscarinic signaling mechanism
in smooth muscle, a pharmacological approach has been used
to identify the coupling of receptor subtypes to specific coupling mechanisms. However, the pharmacological approach
has limitations in situations where more than one receptor
contributes to the response. Consequently, we used muscarinic receptor knockout mice to investigate the coupling of
muscarinic receptor subtypes to phosphoinositide hydrolysis
in the smooth muscle of mouse ileum and urinary bladder.
Our findings demonstrate that, in the urinary bladder, only
the M3 receptor couples to phosphoinositide hydrolysis,
whereas in the longitudinal muscle of the mouse ileum, significant contributions of the M1 and M2 receptor are apparent in addition to a major role for the M3 receptor.
Phosphoinositide Hydrolysis in Urinary Bladder and Ileum
in which ␧i and Ki denote the intrinsic efficacy and dissociation
constant of oxotremorine-M at receptor subtype Ri, respectively, and
n denotes the number of receptor subtypes contributing to the response. When used in this context, ␧i denotes the ability of oxotremorine-M to activate the receptor as well as the ability of the receptor
to trigger phosphoinositide hydrolysis in smooth muscle. We assume
that each receptor acts independently to generate a stimulus but
that the independent stimuli converge on the same effector system.
As described under Results, our data suggest that the M1, M2, and
M3 receptor subtypes contribute to the phosphoinositide response in
ileal smooth muscle. It has been shown that the dissociation constants of oxotremorine-M for M1 (2.5 ␮M; Mei et al., 1991), M2 (3.9
␮M; Ehlert, 1985), and M3 (2.9 ␮M; Ringdahl, 1985) receptors are
approximately the same. Consequently, we assumed that the dissociation constant of oxotremorine-M for these subtypes was the same,
which leads to the following simplification of eq. 2,
冘
n
S⫽
(3)
in which, KX denotes the dissociation constant of oxotremorine-M for
muscarinic receptor subtypes eliciting phosphoinositide hydrolysis.
Substitution of eq. 3 into 1 yields
冘
冘
n
Response ⫽
i⫽1
n
i⫽1
X␧iRi
M
X ⫹ KX sys
(4)
X␧iRi
⫹ KE
X ⫹ KX
XEmax
X ⫹ EC50
(5)
in which
E max ⫽
Msys␶
␶⫹1
(6)
EC50 ⫽
KX
␶⫹1
(7)
冘
␧ iR i
KE
(8)
n
␶⫽
i⫽1
In the analysis of our data, we assume that the product ␧iRi is a
measure of the contribution of receptor Ri to the total phosphoinositide response. This product is essentially equal to the maximal
stimulus elicited by the receptor. In our study, there are three
discernable populations of muscarinic receptors (n ⫽ 3): M2 receptors
(R2), M3 receptors (R3), and the residual receptors (R1). It is possible
to obtain an estimate of the parameter ␧iRi for a given receptor
relative to the total contribution of all receptors using the following
equation, which specifically applies to the M3 receptor,
E max⫺KO3EC50 ⫺ WT
RC 3 ⫽ 1 ⫺
Emax⫺WTEC50 ⫺ KO3
(9)
in which RC3 denotes the relative contribution of the M3 receptor to
the phosphoinositide response, Emax–KO3 and Emax-WT denote the
maximal phosphoinositide responses, and EC50-KO3 and EC50-WT denote the EC50 values for oxotremorine-M in M3 KO and wild-type
mice, respectively. Substitution of eqs. 6 and 7 for the EC50 and Emax
values in eq. 9 followed by simplification yields
(10)
Using an analogous approach, the relative contribution of the M2
receptor to the phosphoinositide response (RC2) can be estimated as
RC 2 ⫽ 1 ⫺
E max⫺KO2EC50 ⫺ WT
Emax⫺WTEC50 ⫺ KO2
(11)
in which Emax-KO2 and EC50-KO2 denote the Emax and EC50 values of
oxotremorine-M in M2 KO mice. Equation 11 simplifies to
RC 2 ⫽
␧ 2 R2
␧1R1 ⫹ ␧2R2 ⫹ ␧3R3
(12)
Likewise, the contribution of the residual receptors (CE1) can be
estimated from the following equation:
RC 1 ⫽
E max⫺KO23EC50 ⫺ WT
Emax⫺WTEC50 ⫺ KO23
(13)
in which Emax-KO23 and EC50-KO23 denote the Emax and EC50 values
of oxotremorine-M in M2/M3 KO mice. Equation 13 simplifies to
RC 1 ⫽
␧ 1 R1
␧1R1 ⫹ ␧2R2 ⫹ ␧3R3
(14)
Analysis of Competitive Antagonism. The dissociation constants of antagonists (KB), based on competitive antagonism of the
phosphoinositide response, were estimated using the following equation:
K B ⫽ 关B兴/共CR ⫺ 1兲
This equation simplifies to
Response ⫽
␧ 3 R3
␧1R1 ⫹ ␧2R2 ⫹ ␧3R3
(15)
in which [B] denotes the concentration of the antagonist and CR
denotes the ratio of the EC50 value of the agonist measured in the
presence of the antagonist divided by that measured in the absence
of the antagonist.
Statistical Analysis of Concentration-Response Curves.
Global nonlinear regression analysis was used to determine the
significance of difference between the EC50 and Emax values determined in experiments on phosphoinositide hydrolysis in ilea from
wild-type and muscarinic receptor knockout mice. Using this approach, the data of the concentration-response curves are fitted
simultaneously, allowing the parameter of interest to be shared
among the curves for the different mouse strains or estimated independently for each strain. The significance of the reduction in the
residual sum of squares that occurs when the parameter is fitted
independently was determined using an F test.
Materials. The reagents used in this study were obtained from
the following sources. myo-[3H]Inositol (PerkinElmer Life and Analytical Sciences (Boston, MA), tetrodotoxin and DNase I (SigmaAldrich, St. Louis, MO), AF-DX 116 (Boehringer Ingelheim Pharmaceuticals, Ridgefield, CT), pirenzepine (Research Biochemicals,
Natick, MA), pertussis toxin (List Biological Laboratories, Campbell,
CA), and collagenase type II (Worthington Biochemical, Lakewood,
NJ).
Results
Phosphoinositide Hydrolysis in the Urinary Bladder
and Ileum. Oxotremorine-M elicited a robust muscarinic
phosphoinositide response in urinary bladder from wild-type
and M2 KO mice (Fig. 1a). In contrast, little or no significant
muscarinic response was detected in urinary bladder from
M3 KO and M2/M3 KO mice. These results indicate that the
muscarinic phosphoinositide response in mouse urinary
bladder is mediated by the M3 receptor. Table 1 summarizes
the EC50 and Emax values of oxotremorine-M for eliciting
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
i⫽1
X␧ iRi
X ⫹ KX
RC 3 ⫽
651
652
Tran et al.
TABLE 1
Oxotremorine-M-induced phosphoinositide hydrolysis in mouse urinary
bladder and ileuma
Mouse Strain
Bladder
Wild type (n ⫽ 3)
M2 KO (n ⫽ 5)
Ileum
Wild-type (n ⫽ 8)
M2 KO (n ⫽ 9)
M3 KO (n ⫽ 10)
M2/M3 KO (n ⫽ 4)
pEC50
Maximal % Conversion
6.66 ⫾ 0.16
6.55 ⫾ 0.14
32.8 ⫾ 1.81
31.8 ⫾ 1.57
6.44 ⫾ 0.16
6.28 ⫾ 0.10
5.79 ⫾ 0.12b
5.83 ⫾ 0.18
35.8 ⫾ 1.84
34.7 ⫾ 1.23
31.2 ⫾ 1.35c
20.8 ⫾ 1.41d,e
a
Data were calculated from experiments shown in Fig. 1. Mean values ⫾ S.E.M.
are shown. The number of experiments is shown in parentheses.
b
Significantly different from wild type (F1,86 ⫽ 10.1; P ⫽ 0.0021).
c
Significantly different from wild type (F1,86 ⫽ 4.13; P ⫽ 0.045).
d
Significantly different from wild type (F1,56 ⫽ 20.3; P ⫽ 3.4 ⫻ 10⫺5).
e
Significantly different from M3 KO (F1,66 ⫽ 17.3; P ⫽ 9.6 ⫻ 10⫺5).
phosphoinositide hydrolysis in urinary bladder from wildtype and M2 KO mice.
We also investigated the effects of oxotremorine-M on
phosphoinositide hydrolysis in the longitudinal muscle of the
mouse ileum (Fig. 1b). Oxotremorine-M mediated an accumulation of inositol phosphates in all of the knockout mouse
strains studied, with the more potent responses measured in
wild-type and M2 KO mice (pEC50 ⫽ 6.44 ⫾ 0.16 and 6.28 ⫾
0.10, respectively) and less potent responses in M3 KO and
M2/M3 KO mice (pEC50 ⫽ 5.79 ⫾ 0.12 and 5.83 ⫾ 0.18,
respectively). The responses observed in M3 KO and M2/M3
KO mice also exhibited a diminished maximal response relative to wild type. A summary of these results is shown in
Table 1. A substantial loss of function associated with the
lack of the M3 receptor is apparent when the data from the
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
Fig. 1. a and b, oxotremorine-M-induced phosphoinositide hydrolysis in
wild-type, M2 KO, M3 KO, and M2/M3 KO mouse urinary bladder strips
(a) and longitudinal muscle strips of the ileum (b). Each point represents
the mean values ⫾ S.E.M. from four to 10 separate experiments.
M3 KO mouse are compared with those of the wild type as
well as when the data from the M2/M3 KO mouse are compared with those of the M2 KO mouse. A small role for the M2
receptor in phosphoinositide hydrolysis is suggested by the
nonsignificant loss of function in the M2 KO relative to wild
type, which was confirmed by the significant loss of function
in the M2/M3 KO relative to M3 KO. The contribution of
muscarinic receptor subtypes to the phosphoinositide response was estimated using the approach described under
Materials and Methods. Using eqs. 9 and 13, we estimated
that the M3 receptor contributes to 80% of the response,
whereas the residual muscarinic receptors in the M2/M3 KO
mouse contribute to 15% of the response. If the contributions
of the M2, M3, and residual receptors equal to 100%, then the
contribution of the M2 receptor is approximately 5%. It is
difficult to estimate the contribution of the M2 receptor by
comparison of muscarinic phosphoinositide response in wildtype and M2 KO mice, because a contribution of 5% would be
manifest as only a 1.054-fold increase in the EC50 value of
oxotremorine-M. The observed increase (1.42-fold) was not
significantly different from this value. Consequently, it is
more reliable to estimate the contribution of the M2 receptor
by the loss of function in the M2/M3 KO mouse relative to the
M3 KO mouse.
If the dissociation constants of oxotremorine-M for M1, M2,
and M3 receptor subtypes are approximately the same, then
the concentration-response curves in wild-type and muscarinic receptor knockout mice can be compared with those
obtained when the method of partial receptor inactivation
(Furchgott, 1966) is used to estimate the dissociation constant of an agonist. To test this postulate, we used Furchgott’s method to estimate the dissociation constant of oxotremorine-M by comparing equiactive concentrations of
oxotremorine-M in wild-type, M3 KO, and M2/M3 KO mice.
The overall estimate of the pKD value of oxotremorine-M
(5.23 ⫾ 0.17) did not differ significantly when estimated by
comparing the wild type data with those obtained in M3 KO
or M2/M3 KO mice. Consequently, these results are consistent with the postulate that the dissociation constants of
oxotremorine-M are similar for M1, M2, and M3 receptors.
This result is consistent with the assumption made in eq. 3 of
the mathematical derivation of our method for estimation of
the RC value.
Competitive Antagonism of the Phosphoinositide
Response in M2/M3 KO Ileum. We measured the competitive antagonism of the phosphoinositide response to oxotremorine-M in the ileum of the M2/M3 KO mouse to characterize the nature of the muscarinic receptor mediating this
response (see Fig. 2a). At a concentration of 2 ␮M, the M2selective muscarinic antagonist AF-DX 116 caused a 6.3-fold
dextral shift in the concentration-response curve to oxotremorine-M, which yields a pKB estimate of 6.42 ⫾ 0.42
(see eq. 15). When present at a concentration of 0.2 ␮M, the
M1-selective antagonist pirenzepine shifted the concentration-response curve to oxotremorine-M 22.9-fold, yielding a
pKB estimate of 8.04 ⫾ 0.47. The pKB values for AF-DX 116
and pirenzepine are in close agreement with their corresponding binding affinities for the human M1 receptor (6.2
and 7.8, respectively), but not the M2 (7.3 and 6.0), the M3
(6.1 and 6.6), the M4 (6.0 and 7.2), or the M5 (5.3 and 6.6)
(Esqueda et al., 1996). These results suggest that the M1
Phosphoinositide Hydrolysis in Urinary Bladder and Ileum
653
Fig. 2. Oxotremorine-M-mediated phosphoinositide in ileum from M2/M3 KO mice. The competitive antagonism of the response by AF-DX 116 and
pirenzepine is shown in a, and the effect of oxotremorine-M (0.1 mM) on the response in dissociated smooth muscle cells is shown in b. Data are
expressed as the percentage conversion of [3H]inositol-labeled phospholipid into [3H]inositol phosphates. The mean values ⫾ S.E.M. of three
experiments are shown.
investigated whether our method for treating smooth muscle
with the toxin was effective in ADP-ribosylating Gi/o. We
have previously shown that injection of the toxin intraperitoneally into guinea pigs three days before is effective in
ADP-ribosylating Gi/o as determined by SDS-polyacrylamide
gel electrophoresis, cAMP assays, and radioligand binding
assays (Thomas and Ehlert, 1994). Consequently, we used
the latter method in the present study, which is based on the
assumption that the agonist-receptor-G protein complex exhibits high affinity in the absence of GTP. Figure 4a shows
the results of oxotremorine-M/[3H]NMS competitive binding
experiments done on homogenates of the longitudinal muscle
of the ileum from control wild-type mice and those that had
been injected three days before with pertussis toxin (70 –100
␮g/kg). In control tissue, oxotremorine-M exhibited a relatively potent IC50 value of ⬃0.23 ␮M (pIC50 ⫽ 6.23 ⫾ 0.12)
and a low Hill slope (nH) of 0.50. GTP (1 mM) caused a
steepening of the curve (nH ⫽ 0.74) and a reduction in the
potency of oxotremorine-M, as shown by the 6-fold dextral
shift in the binding curve. In tissue from animals that had
been treated with pertussis toxin, the potency of oxotremorine-M was only one-fourth that of control as shown by a
significantly smaller pIC50 value of 5.67 ⫾ 0.06. In addition,
the Hill slope was greater (nH ⫽ 0.69). GTP (1 mM) caused
less of a shift in the binding curve of oxotremorine-M (2-fold)
in pertussis toxin-treated tissue. Collectively, these results
are consistent with the idea that M2 receptors in the ileum
have much less ability to generate a high affinity, GTPsensitive Gi/o complex following pertussis toxin treatment.
Because of the small size of the mouse urinary bladder,
there was insufficient tissue in a single mouse to measure
complete oxotremorine-M/[3H]NMS competitive binding
curves to monitor the effectiveness of pertussis toxin treatment. Consequently, we developed a different assay based on
the observation that M2 receptors in the myocardium spontaneously form G protein complexes in low ionic strength
buffer (Hulme et al., 1981). Under these conditions, the affinity of the antagonist [3H]NMS is reduced and guanine
nucleotides cause an increase in [3H]NMS binding. The
mechanism can be explained by 1) the precoupling of the M2
receptor to Gi/o in low-ionic strength buffer, 2) the dissociation of the M2 receptor-G protein complex by guanine nucleotides, and 3) the preferential binding of [3H]NMS to the free
receptor complex with higher affinity than that of the receptor-G protein complex. Thus, we used the GTP-induced in-
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
receptor is primarily responsible for mediating the muscarinic phosphoinositide response in M2/M3 KO mouse ileum.
Dissociated Smooth Muscle. To determine whether M1
receptor-mediated phosphoinositide hydrolysis in the M2/M3
KO mouse originated from smooth muscle and not from other
cell types, such as myenteric neurons, dissociated smooth
muscle cells of the ileum were prepared. Because the protocol
yielded a small amount of cells, the phosphoinositide response to only a single concentration of oxotremorine-M (0.1
mM) was measured. In the wild-type mouse, the phosphoinositide response to oxotremorine-M was 11.0 ⫾ 0.98%
when expressed as conversion of [3H]phosphoinositides into
[3H]inositol phosphates (Fig. 2b). Oxotremorine-M also
caused a significant phosphoinositide response in the M2/M3
KO mouse ileum of 6.1 ⫾ 0.34%. The size of the response in
smooth muscle cells from the M2/M3 KO mouse ileum relative
to that measured in wild-type cells is similar to the relative
sizes observed in intact tissue. We conclude that most, if not
all, of the response in intact smooth muscle is contributed by
smooth muscle cells.
Effects of Pertussis Toxin. Pertussis toxin catalyzes the
ADP-ribosylation of Gi/o proteins, which prevents the coupling of receptors to these G proteins (Kurose et al., 1983). M2
receptors are known to signal through Gi/o (Ashkenazi et al.,
1987); therefore, pertussis toxin should be a useful probe for
determining the extent to which muscarinic receptor signaling through Gi/o is involved in the phosphoinositide response
in smooth muscle. In wild-type urinary bladder, pertussis
toxin treatment had little significant effect on the phosphoinositide response to oxotremorine-M (Fig. 3a). Likewise, in
ilea from wild-type (Fig. 3b) and M3 KO mice (Fig. 3c), pertussis toxin had little or no inhibitory effect on oxotremorineM-stimulated phosphoinositide hydrolysis. We also observed
no significant effect of pertussis toxin on the pEC50 and Emax
values of oxotremorine-M in ilea from M2 KO (6.25 ⫾ 0.11
and 33.6 ⫾ 1.2%, respectively) and M2/M3 KO mice (5.91 ⫾
0.17 and 21.3 ⫾ 1.4%, respectively). It can be seen that the
latter estimates are approximately the same as the corresponding values for the control data listed in Table 1. Collectively, these results provide no evidence for the role of Gi/o in
mediating phosphoinositide hydrolysis in mouse ileum and
urinary bladder.
Effects of Pertussis Toxin on Muscarinic Receptor
Binding. To establish the significance of our [3H]inositol
phosphate measurements in pertussis toxin-treated mice, we
654
Tran et al.
pertussis toxin-treated bladder, the increase was insignificant (31 ⫾ 19%). The increase in control urinary bladder was
significantly greater than that observed in pertussis toxintreated bladder (P ⫽ 0.03). These results indicate that, in
mouse urinary bladder, pertussis toxin was able to catalyze
the ADP-ribosylation of Gi/o and thereby reduce GTP-induced
increase in [3H]NMS binding.
Discussion
crease in [3H]NMS binding in hypotonic buffer as a measure
of M2 receptor-G protein interactions in urinary bladder
treated with pertussis toxin (see Fig. 4b). GTP (1 mM) significantly increased the binding of [3H]NMS at a low concentration (0.25 nM) in untreated wild-type urinary bladder by
greater than 2-fold (118 ⫾ 18% over control binding). In
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
Fig. 3. a– c, effects of pertussis toxin on muscarinic receptor-mediated
phosphoinositide hydrolysis in wild-type mouse urinary bladder (a), wildtype ileum (b), and M3 KO ileum (c). Pertussis toxin (70 –100 ␮g/kg body
weight) was injected intraperitoneally 3 days before experiments. Mean
values ⫾ S.E.M. from three to 10 experiments are shown. The control
concentration-response curves for the ileum shown in b and c are the
same as those shown in Fig. 1b.
In our experiments, we focused on the contribution of M2
and M3 receptors to the phosphoinositide response because
these are the two main muscarinic receptor subtypes expressed in smooth muscle. Our results show that the M3
receptor mediates the muscarinic phosphoinositide response
in urinary bladder because the response is completely lost in
mice lacking M3 receptors (Fig. 1a). These results agree with
previous pharmacological studies demonstrating an M3-mediated mechanism for the muscarinic phosphoinositide response in guinea pig urinary bladder (Noronha-Blob et al.,
1989).
Our results on the ileum show a complex picture, suggesting that more than one muscarinic subtype mediates phosphoinositide hydrolysis. We observed a small loss of function
in the M2 KO, a large loss of function in the M3 KO, and a
substantial response that still persisted in the M2/M3 KO
mouse (Fig. 1b) that exhibited an M1 profile in competitive
antagonism experiments (Fig. 2a). In situations where more
than one receptor mediates a response and there is a receptor
reserve, it is not a trivial matter to estimate the contribution
of each receptor to the response. Nevertheless, we developed
a parameter called RC that accurately estimates the contribution of each receptor to the response assuming that the
observed dissociation constant of oxotremorine-M for each
receptor is the same (see Materials and Methods). Accordingly, we estimated that the M1, M2, and M3 receptor contributed 15, 5, and 80% respectively, to the phosphoinositide
response in mouse ileum. These results are generally consistent with a prior pharmacological study demonstrating that
the M3 receptor mediates phosphoinositide hydrolysis in the
longitudinal muscle of the rat ileum (Candell et al., 1990).
The M1 and M2 effects on phosphoinositide response were
most likely too low to be detected in prior pharmacological
studies on rat ileum. Moreover, the M1 effect would have
been masked because the only so-called M3 selective antagonists employed also exhibited high affinity for M1 receptors.
In a situation where a response exhibits a receptor reserve
and is mediated by both M1 and M3 receptors, an M1-selective antagonist would block the response with low M3 potency, and an M3 selective antagonist would block with low
M1 potency. An antagonist with high affinity for both M1 and
M3 receptors would block the response with high potency.
This phenomenon occurs because the M1 receptor can rescue
the M3 receptor and vice versa.
Although the contribution of the M2 receptor to phosphoinositide hydrolysis is small in the ileum, it was detected as
a substantial loss of function in the M2/M3 KO relative to the
M3 KO (Fig. 1b). Evidence for recombinant M2 receptors
coupling to phosphoinositide hydrolysis through a pertussis
toxin-sensitive mechanism has been noted in murine fibroblasts (Lai et al., 1991) and CHO cells (Ashkenazi et al.,
1989). In rabbit intestinal smooth muscle, M2 receptors have
Phosphoinositide Hydrolysis in Urinary Bladder and Ileum
655
Fig. 4. a and b, the effects of pertussis toxin treatment on [3H]NMS binding in the longitudinal muscle of the ileum (a) and urinary bladder of the
wild-type mouse (b). a, the competitive inhibition of [3H]NMS binding by oxotremorine-M was measured in the presence and absence of GTP. The
concentration of [3H]NMS was 0.5 nM. b, the percentage enhancement of [3H]NMS binding by GTP was measured. In these experiments, the binding
of [3H]NMS was measured at a single concentration of 0.25 nM. Mean values ⫾ S.E.M. are shown in a and b.
Blob et al., 1989; Candell et al., 1990) and is consistent with
the direct role that the M3 receptor has in contraction of
smooth muscle. Many receptors that couple to Gq elicit contraction of smooth muscle. However, the details of the mechanism are unclear because the source of Ca2⫹ for contraction
is mainly extracellular (Bolger et al., 1983) and not through
inositol-1,4,5-trisphosphate-mediated release of intracellular
Ca2⫹. M2 receptors have been shown to trigger signaling
mechanisms that are contingent upon M3 receptor activation,
including activation of a nonselective cation conductance
(Bolton, 1979) and inhibition of Ca2⫹-activated K⫹ channels
(Cole et al., 1989). Activation of M2 receptors has been shown
to enhance M3 receptor-mediated contractions and inhibit
the relaxant effects of isoproterenol and forskolin on contractions mediated by Gq-linked receptors, including the M3
(Thomas et al., 1993; Sawyer and Ehlert, 1998). All of these
M2 responses can be attributed to Gi-mediated effects. The
relationship between the M2 phosphoinositide response observed in ileum here and the contractile response is unclear.
M2 receptors have been shown to mediate a weak, direct
contractile response in both the ileum and urinary bladder of
M3 KO mouse, and this weak response is comparatively
greater in the ileum. One unresolved question is that the
ileum from M2/M3 KO mice does not exhibit a muscarinic
agonist-induced contractile response (Matsui et al., 2002),
yet as described herein, it does exhibit a significant phosphoinositide response elicited through activation of M1 receptors.
One possible explanation is that there is a differential cellular compartmentalization of M1 and M3 receptors in smooth
muscle of the ileum. It has been shown that Gq, downstream
effectors, and receptors that couple to Gq exhibit a differential distribution in lipid rafts (see review by Pike, 2003).
Perhaps M3 receptors, Gq, PLC-␤, and a downstream effector
protein that initiates contraction are colocalized in lipid
rafts, whereas M1 receptors are localized elsewhere with Gq
and PLC-␤ but not with the relevant proteins involved in
initiating contraction. Our results suggest that M1 receptors
may have a role in smooth muscle distinct from contraction
that is mediated through phosphoinositide hydrolysis.
References
Ashkenazi A, Peralta EG, Winslow JW, Ramachandran J, and Capon DJ (1989)
Functionally distinct G proteins selectively couple different receptors to PI hydrolysis in the same cell. Cell 56:487– 493.
Ashkenazi A, Winslow JW, Peralta EG, Peterson GL, Schimerlik MI, Capon DJ, and
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
been shown to mediate phosphoinositide hydrolysis through
a pertussis toxin-sensitive mechanism (Murthy et al., 2003).
The authors suggest that the mechanism involves activation
of PLC-␤3 by the ␤␥i3 subunits of Gi (Murthy et al., 2003). In
several prior studies on the longitudinal muscle of the guinea
pig ileum, we have consistently found that pertussis toxin
treatment causes a small increase in muscarinic agonistmediated and histamine-mediated phosphoinositide hydrolysis (Thomas and Ehlert, 1994; Ehlert et al., 2001; Shehnaz et
al., 2001). This result implies that pertussis toxin treatment
(intraperitoneal injection in vivo, 3 days before) causes a
small increase in agonist-mediated phosphoinositide hydrolysis in the ileum. Perhaps this enhancing effect prevented us
from detecting a small loss of M2 receptor GI-mediated phosphoinositide hydrolysis in mouse strains expressing M2 receptors (Fig. 3a,b). Consequently, we cannot discern whether
the M2 response is pertussis toxin-sensitive.
Our results on dissociated smooth muscle cells indicate
that the M1 response is in the muscle itself (Fig. 2b). If the M1
response resided in a different cell type (e.g., neurons), the
concentration-response curve for oxotremorine-M in the wildtype mouse should be equivalent to the summation of that for
the M2–M3 response in smooth muscle plus the M1 response
in neurons. Because the maximal response in the M2/M3 KO
was ⬃57% that of wild type, this neuronal M1 hypothesis
would imply that the maximal response in smooth muscle
only accounted for 43% of the total response. This situation
seems unlikely. It is also possible that the M1 response in
smooth muscle is compensatory, caused by the lack of expression of M2 and M3 receptors. This hypothesis is difficult to
test. A small amount of M1 receptors (Wall et al., 1991) and
mRNA (Maeda et al., 1988) has been detected in wild-type
intestinal smooth muscle from rats, which could account for
the muscarinic phosphoinositide response observed here. No
compensatory increases in muscarinic receptor expression
have been detected in muscarinic receptor knockout mice
(Gautam et al., 2005), and evidence for a decrease in expression has actually been observed in brain (Oki et al., 2005). It
would be difficult to ascribe potential changes in other signaling proteins in M2/M3 knockout mice to the small M1
response that we observed.
The prominent coupling of M3 receptors to phosphoinositide hydrolysis in mouse smooth muscle is consistent with
prior pharmacological studies on other animals (Noronha-
656
Tran et al.
relaxant action of forskolin and isoproterenol against muscarinic agonist-induced
contractions in smooth muscle from M2 receptor knockout mice. J Pharmacol Exp
Ther 305:106 –113.
Matsui M, Motomura D, Fujikawa T, Jiang J, Takahashi S, Manabe T, and Taketo
MM (2002) Mice lacking M2 and M3 muscarinic acetylcholine receptors are devoid
of cholinergic smooth muscle contractions but still viable. J Neurosci 22:10627–
10632.
Matsui M, Motomura D, Karasawa H, Fujikawa T, Jiang J, Komiya Y, Takahashi S,
and Taketo MM (2000) Multiple functional defects in peripheral autonomic organs
in mice lacking muscarinic acetylcholine receptor gene for the M3 subtype. Proc
Natl Acad Sci USA 97:9579 –9584.
Mei L, Lai J, Yamamura H, and Roeske W (1991) Pharmacologic comparison of
selected agonists for the M1 muscarinic receptor in transfected murine fibroblast
cells (B82). J Pharmacol Exp Ther 256:689 – 694.
Murthy KS, Zhou H, Grider JR, Brautigan DL, Eto M, and Makhlouf GM (2003)
Differential signalling by muscarinic receptors in smooth muscle: m2-mediated
inactivation of myosin light chain kinase via Gi3, Cdc42/Rac1 and p21-activated
kinase 1 pathway and m3-mediated MLC20 (20-kDa regulatory light chain of
myosin II) phosphorylation via Rho-associated kinase/myosin phosphatase targeting subunit 1 and protein kinase C/CPI-17 pathway. Biochem J 374:145–155.
Noronha-Blob L, Lowe V, Patton A, Canning B, Costello D, and Kinnier WJ (1989)
Muscarinic receptors: relationships among phosphoinositide breakdown, adenylate cyclase inhibition, in vitro detrusor muscle contractions and in vivo cystometrogram studies in guinea pig bladder. J Pharmacol Exp Ther 249:843– 851.
Oki T, Takagi Y, Inagaki S, Taketo MM, Manabe T, Matsui M, and Yamada S (2005)
Quantitative analysis of binding parameters of [3H]N-methylscopolamine in central nervous system of muscarinic acetylcholine receptor knockout mice. Brain Res
Mol Brain Res 133:6 –11.
Peralta EG, Ashkenazi A, Winslow JW, Ramachandran J, and Capon DJ (1988)
Differential regulation of PI hydrolysis and adenylyl cyclase by muscarinic receptor subtypes. Nature (Lond) 334:434 – 437.
Pike LJ (2003) Lipid rafts: bringing order to chaos. J Lipid Res 44:655– 667.
Ringdahl B (1985) Structural requirements for muscarinic receptor occupation and
receptor activation by oxotremorine analogs in the guinea-pig ileum. J Pharmacol
Exp Ther 232:67–73.
Sawyer GW and Ehlert FJ (1998) Contractile roles of the M2 and M3 muscarinic
receptors in the guinea pig colon. J Pharmacol Exp Ther 284:269 –277.
Shehnaz D, Ansari KZ, and Ehlert FJ (2001) Acetylcholine-induced desensitization
of the contractile response to histamine in guinea pig ileum is prevented by either
pertussis toxin-treatment or by selective inactivation of muscarinic M3 receptors.
J Pharmacol Exp Ther 297:1152–1159.
Stryer L and Bourne HR (1986) G proteins: A family of signal transducers. Annu Rev
Cell Biol 2:391– 419.
Thomas EA, Baker SA, and Ehlert FJ (1993) Functional role for the M2 muscarinic
receptor in smooth muscle of guinea pig ileum. Mol Pharmacol 44:102–110.
Thomas EA and Ehlert FJ (1994) Pertussis toxin blocks M2 muscarinic receptormediated effects on contraction and cyclic AMP in the guinea pig ileum, but not
M3-mediated contractions and phosphoinositide hydrolysis. J Pharmacol Exp Ther
271:1042–1050.
Wall SJ, Yasuda RP, Li M, and Wolfe BB (1991) Development of an antiserum
against m3 muscarinic receptors: distribution of m3 receptors in rat tissues and
clonal cell lines. Mol Pharmacol 40:783–789.
Yang CM, Chou SP, and Sung TC (1991) Muscarinic receptor subtypes coupled to
generation of different second messengers in isolated tracheal smooth muscle cells.
Br J Pharmacol 104:613– 618.
Address correspondence to: Dr. Frederick J. Ehlert, Department of Pharmacology, University of California, Irvine, CA 92697-4625. E-mail: fjehlert@
uci.edu
Downloaded from jpet.aspetjournals.org at ASPET Journals on June 15, 2017
Ramachandran J (1987) An M2 muscarinic receptor subtype coupled to both
adenylyl cyclase and phosphoinositide turnover. Science (Wash DC) 238:672– 675.
Berridge MJ, Downes CP, and Hanley MR (1982) Lithium amplifies agonistdependent phosphatidylinositol responses in brain and salivary glands. Biochem J
206:587–595.
Black JW and Leff P (1983) Operational models of pharmacological agonism. Proc R
Soc Lond Ser B Biol Sci 220:141–162.
Bolger GT, Gengo P, Klockowski R, Luchowski E, Siegel H, Janis RA, Triggle AM,
and Triggle DJ (1983) Characterization of binding of the Ca2⫹ channel antagonist,
[3H]nitrendipine, to guinea pig ileal smooth muscle. J Pharmacol Exp Ther 225:
291–309.
Bolton TB (1979) Mechanisms of action of transmitters and other substances on
smooth muscle. Physiol Rev 59:606 –718.
Bonner TI, Young AC, Brann MR, and Buckley NJ (1988) Cloning and expression of
the human and rat m5 muscarinic acetylcholine receptor genes. Neuron 1:403–
410.
Candell LM, Yun SH, Tran LL, and Ehlert FJ (1990) Differential coupling of
subtypes of the muscarinic receptor to adenylate cyclase and phosphoinositide
hydrolysis in the longitudinal muscle of the rat ileum. Mol Pharmacol 38:689 – 697.
Caulfield MP (1993) Muscarinic receptors– characterization, coupling, and function.
Pharmacol Ther 58:319 –379.
Choppin A and Eglen RM (2001) Pharmacological characterization of muscarinic
receptors in mouse isolated urinary bladder smooth muscle. Br J Pharmacol
133:1035–1040.
Cole WC, Carl A, and Sanders KM (1989) Muscarinic suppression of Ca2⫹-dependent
K current in colonic smooth muscle. Am J Physiol 257:C481–C487.
Ehlert FJ (1985) The relationship between muscarinic receptor occupancy and adenylate cyclase inhibition in the rabbit myocardium. Mol Pharmacol 28:410 – 421.
Ehlert FJ, Ansari KZ, Shehnaz D, Sawyer GW, and Griffin MT (2001) Acetylcholineinduced desensitization of muscarinic contractile response in guinea pig ileum is
inhibited by pertussis toxin treatment. J Pharmacol Exp Ther 299:1126 –1132.
Ehlert FJ, Griffin MT, Abe DM, Vo TH, Taketo MM, Manabe T, and Matsui M (2005)
The M2 muscarinic receptor mediates contraction through indirect mechanisms in
mouse urinary bladder. J Pharmacol Exp Ther 313:368 –378.
Ehlert FJ, Ostrom RS, and Sawyer GW (1997) Subtypes of the muscarinic receptor
in smooth muscle. Life Sci 61:1729 –1740.
Esqueda EE, Gerstin EH Jr, Griffin MT, and Ehlert FJ (1996) Stimulation of cyclic
AMP accumulation and phosphoinositide hydrolysis by M3 muscarinic receptors in
the rat peripheral lung. Biochem Pharmacol 52:643– 658.
Furchgott RF (1966) The use of ␤-haloalkylamines in the differentiation of receptors
and in the determination of dissociation constants of receptor-agonist complexes.
Adv Drug Res 3:21–55.
Gautam D, Han SJ, Heard TS, Cui Y, Miller G, Bloodworth L, and Wess J (2005)
Cholinergic stimulation of amylase secretion from pancreatic acinar cells studied
with muscarinic acetylcholine receptor mutant mice. J Pharmacol Exp Ther 313:
995–1002.
Hulme EC, Berrie CP, Birdsall NJ, and Burgen AS (1981) Two populations of binding
sites for muscarinic antagonists in the rat heart. Eur J Pharmacol 73:137–142.
Kendall DA and Hill SJ (1990) Measurement of [3H]inositol phospholipid turnover,
in Methods in Neurotransmitter Analysis (Yamamura HI, Enna SJ, and Kuhar MJ
eds) pp 69 – 87, Raven Press, New York.
Kurose H, Katada T, Amano T, and Ui M (1983) Specific uncoupling by isletactivating protein, pertussis toxin, of negative signal transduction via ␣-adrenergic, cholinergic and opiate receptors in neuroblastoma x glioma hybrid cells. J Biol
Chem 258:4870 – 4875.
Lai J, Waite SL, Bloom JW, Yamamura HI, and Roeske WR (1991) The m2 muscarinic acetylcholine receptors are coupled to multiple signaling pathways via pertussis toxin-sensitive guanine nucleotide regulatory proteins. J Pharmacol Exp
Ther 258:938 –944.
Maeda A, Kubo T, Mishina M, and Numa S (1988) Tissue distribution of mRNAs
encoding muscarinic acetylcholine receptor subtypes. FEBS Lett 239:339 –342.
Matsui M, Griffin MT, Shehnaz D, Taketo MM, and Ehlert FJ (2003) Increased