Center for Drug Design and Development, Department of Medicinal & Biological Chemistry, College of Pharmacy, The University of Toledo,
Toledo, Ohio
A mutant human m5 receptor containing the mutations of Ser465 to
Tyr and Thr466 to Pro showed constitutive activity. By replacing the
equivalent Ser388 with Tyr and Thr389 with Pro, we created a mutant
human m1 (Hm1) receptor with comparable double mutations. The mutant
receptor, Hm1(Ser388Tyr, Thr389Pro), was stably expressed in A9 L cells
and displayed enhanced responses to classical muscarinic agonists with
significantly increased potencies. Choline, a normal component of
growth media, showed an efficacy comparable to acetylcholine and
carbachol at Hm1(Ser388Tyr, Thr389Pro) receptors. Methylcarbachol, a
selective nicotinic agonist, exhibited partial agonist activity at
human m1 wild-type receptors and full agonist activity at
Hm1(Ser388Tyr, Thr389Pro) receptors. l-Hyoscyamine inhibited
the activities of choline and methylcarbachol. Muscarinic antagonists
displayed small reductions in binding affinities, although muscarinic
agonists showed greatly increased binding affinities for Hm1(Ser388Tyr, Thr389Pro) receptors. All agonists, including choline and
methylcarbachol, showed multiple affinity states at Hm1(Ser388Tyr,
Thr389Pro) receptors in the absence of GppNHp. The high affinity
binding sites for acetylcholine, arecoline and choline were shifted in
the presence of GppNHp. These results suggest that Hm1(Ser388Tyr,
Thr389Pro) is conformationally favorable for agonist binding and
receptor activation.
 |
Introduction |
Muscarinic
acetylcholine receptors are members of the large family of G
protein-coupled receptors mediating signal transduction and represent
important targets for drug design and development. Five subtypes of
muscarinic receptors, named m1 to m5, have been cloned. Functionally,
they can be classified into two groups, the m1, m3 and m5 subtypes are
preferentially coupled to the activation of phospholipase
C
through the pertussis toxin-insensitive Gq/11 family of G proteins; while the m2 and m4 subtypes
are mainly coupled to the inhibition of adenylyl cyclase through the
pertussis toxin-sensitive Gi family of G proteins. Previous
molecular modelling studies (Ward et al., 1992
; Nordvall and
Hacksell 1993, 1995) and site-directed mutagenesis and pharmacological
studies have identified several conserved amino acid residues that are
believed to be involved in ligand binding and/or receptor activation
processes. Important residues include Asp105 in TM III of m1 receptors
(Fraser et al., 1989), the corresponding Asp103 of m2
receptors (Schwarz et al., 1995), Thr231 and Thr234 in TM V
of rat m3 receptors (Wess et al., 1991; 1992) and Tyr506 and
Asn507 in TM VI of rat m3 receptors (Wess et al., 1992;
Blüml et al., 1994). The residues corresponding to
Thr234 and Tyr506 of rat m3 receptors are Thr192 and Tyr381 in human m1
receptors, and Tyr403 in porcine m2 receptor. These residues were
recently demonstrated to participate in agonist binding and/or receptor
activation (Allman et al., 1997; Vogel et al.,
1997; Ward and Hulme, 1997a, b). These amino acids are highly conserved
within the muscarinic acetylcholine receptor subfamily, and Asp105 is
conserved in all G protein-coupled receptors that bind biogenic amines.
However, the molecular mechanisms by which muscarinic receptors are
activated upon agonist binding are still not clear (for recent reviews,
see Wess 1996 and 1997).
A mutant Hm5 receptor with Ser465 and Thr466 at the junction of TM VI
and the N-terminal of the third extracellular domain (Ne3) mutated to
Tyr and Pro, respectively, showed constitutive activity (Spalding
et al., 1995). The constitutively active m5 receptor
displayed increased agonist potencies and agonist binding affinities,
but no change in antagonist binding. In the same study, no significant
effects of GTP analogues were observed on agonist binding and the
constitutive activity was reported to be independent of an increased
sensitivity of the mutant receptor to potential traces of
acetylcholine, or its breakdown product choline, in the growth media.
One drawback for this study was that the receptor functional activities
and ligand binding properties were characterized using different cell
lines. Agonist-dependent and -independent functional properties were
indirectly characterized in transfected NIH 3T3 cells with unknown
receptor expression levels, although ligand binding properties were
analyzed using membrane homogenates from transiently transfected COS-7
cells with expression levels of 1.9 to 5.2 pmol/mg (Spalding et
al., 1995).
The development of selective m1 agonists may be beneficial for the
treatment of Alzheimer's disease (Dunbar et al., 1993, 1994; Messer and Dunbar, 1996; Messer et al., 1997).
Understanding the molecular mechanisms for agonist activation and
receptor subtype selectivity could aid in the design of new lead
compounds. Therefore we created comparable double mutations at the
junction portion of TM VI and Ne3 of the human m1 receptor subtype with
Ser388 and Thr389 residues replaced by Tyr and Pro residues,
respectively (fig. 1). A9 L cells are
particularly suitable for evaluating muscarinic receptor agonist
activity and selectivity, and regulation of binding by GTP analogues
(Brann et al., 1987; Messer et al., 1997). We
therefore transfected the A9 L cells and created stable A9 L cell lines
expressing Hm1(WT) or the mutant receptors, Hm1(Ser388Tyr, Thr389Pro).
The receptor functional activities and ligand binding properties were
characterized in the stably transfected A9 L cells.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1.
Partial sequence alignment of transmembrane (TM) VI,
the third extracellular loop (e3), and TM VII of human muscarinic
acetylcholine receptors. Ser388Thr389 residues in Hm1 and Ser465Thr466
residues in Hm5 are boxed. Amino acid sequences were from Bonner
et al., 1988.
|
|
| |
Materials and Methods |
Materials.
Plasmids Hm1pCD1 (Bonner et al., 1988)
encoding the human m1 muscarinic receptor and pCDneo were kindly
provided by Dr. Tom I. Bonner (NIH) and Dr. Jürgen Wess (NIDDK),
respectively. DMEM was purchased from GIBCO BRL (Grand Island, NY). FBS
was ordered from HyClone (Logan, UT). L-Glutamine solution
and penicillin/streptomycin solution were obtained from GIBCO BRL or
Fisher (Pittsburgh, PA). Geneticin (G418) was ordered from Sigma
Chemical Co. (St. Louis, MO) or Fisher.
Acetylcholine, methylcarbachol, oxotremorine, oxotremorine-M, APE,
trihexyphenidyl hydrochloride and pirenzepine were purchased from
Research Biochemicals Int. (RBI, Natick, MA). Carbachol, l-hyoscyamine, NMS, lithium chloride and GppNHp were ordered
from Sigma. Choline and poly(ethylenimine) were from Aldrich
(Milwaukee, WI). Both myo-[3H]-inositol and
[3H]-(R)-QNB were purchased from Du Pont-New England
Nuclear (Boston, MA). All other inorganic chemicals were from Fisher.
The QuickChange site-directed mutagenesis kit was purchased from
Stratagene (La Jolla, CA). The restriction enzyme Bst Z17 I
was purchased from New England BioLabs (Beverly, MA). The Super Separator-24 and T7 Sequenase PCR product sequencing kit were obtained
from Amersham Corp. (Arlington Heights, IL). The QIAprep Spin Plasmid
Mini Kit and the QIAGEN Plasmid Maxi Kit were bought from QIAGEN
(Chatsworth, CA). The LIPOFECTIN Reagent was obtained from GIBCO BRL.
The SEP-PAK anion exchange cartridges were purchased from Waters
(Franklin, MA). UniverSol ES scintillation cocktail was ordered from
ICN Biomedicals (Irvine, CA) and FP-100 Whatman GF/B filters were
obtained from Brandel (Gaithersburg, MD).
Mutation strategy.
The double mutations of Ser388 to Tyr and
Thr389 to Pro were carried out using the QuickChange kit. The sense
primer was
5'-CCGTACAACATCATGGTGCTGGTATACCCCTTCTGCAAGGACTGTGTTCCCGAG-3' with mutated bases in bold, and the antisense primer was
5'-CTCGGGAACACAGTCCTTGCAGAAGGGGTATACCAGCACCATGATGTTGTACGG-3'. A
silent mutation (Val387) as underlined was generated in order to
incorporate a unique restriction site for Bst Z17 I. The
mutations were confirmed by Bst Z17 I digestion and dideoxy
nucleotide sequencing.
Creation of stable A9 L cell lines.
A9 L cells were
cotransfected with Hm1pCD1 (or mutant Hm1pCD1) and pCDneo in a ratio of
at least 10:1 using LIPOFECTIN Reagent. The transfected A9 L cells were
selected in DMEM (supplemented with 10% FBS, 4 mM
L-glutamine, 50 U/ml penicillin and 50 µg/ml streptomycin) containing 800 µg/ml G418. The surviving Hm1(WT) transfected A9 L cells were subcultured for testing. The surviving Hm1(Ser388Tyr, Thr389Pro) transfected A9 L cell colonies were subcultured into 6-cm dishes for screening as described below. To
obtain high expression levels of Hm1(WT) receptors, A9 L cells were
cotransfected with Hm1pCD1 and pCDneo in a ratio of 50:1 using calcium
phosphate precipitation method (Chen and Okayama, 1987).
Cells were subcultured into 24-well tissue culture plates and grown for
2 days to about 100% confluence. Receptor expression levels then were
assessed using a whole cell binding assay in the plates according to
previously reported methods (Bulseco and Schimerlik, 1996) with
modification. Briefly, cells were washed twice in ice-cold binding
buffer [25 mM sodium phosphate (pH 7.4) containing 5 mM magnesium
chloride]. The specific binding was determined using 3 nM
[3H]-(R)-QNB in the absence and presence of 6 µM cold
QNB in a total volume of 0.4 ml. The plate was incubated at 4°C for
about 2 hr. After three washings with ice-cold binding buffer, the
cells were solubilized using 0.4 ml of binding buffer containing 1%
Triton X-100. The cell lysate and 0.5 ml of rinse water were combined and counted for radioactivity. Cells expressing mutant receptors were
used for PI hydrolysis assays.
PI hydrolysis assays.
A9 L cells expressing Hm1(WT) or
Hm1(Ser388Tyr, Thr389Pro) were trypsinized and seeded into 24-well
tissue culture plates with 1 µCi of [3H]-inositol per
well and cultured for 48 hr to about 80 to 90% confluence. The cells
were washed twice in 0.5 ml of warmed KH buffer (118 mM NaCl, 4.7 mM
KCl, 1.3 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, 25 mM NaHCO3, 11.7 mM glucose, pH
7.4) and incubated for 30 min in 0.45 ml of KH buffer containing 10 mM
LiCl. Test ligands were added for an additional 30 min of incubation.
The reaction was stopped by removal of the incubation mixture and addition of 0.75 ml of 5% (g/100 ml) ice-cold TCA. The plate was incubated at 4°C for 30 min. The levels of IPs accumulated upon ligand stimulation were determined using SEP-PAK cartridges according to Wreggett and Irvine (1987) with minor modifications (Hoss et al., 1990). Briefly, the neutralized 5% TCA extracts were
combined with 0.5 ml of wash water and transferred to balanced SEP-PAK anion exchange cartridges (formate form). The cartridges then were
washed with 6 ml distilled water and 10 ml of 5 mM disodium tetraborate. The radioactive [3H]-IPs were eluted from
the cartridges by 1 ml of 0.6 M ammonium formate/0.06 M formic acid/5
mM disodium tetraborate (pH 4.75). Then 0.5 ml of the 1 ml eluate was
counted in 6 ml of UniverSol ES scintillation cocktail in a 6895 BetaTrac Liquid Scintillation counter. KH buffer containing 10 mM LiCl
served for determination of basal levels, and activities were presented
as percentage stimulation above basal levels. Background cpm counts
from 6 ml cocktail were subtracted from both basal counts and sample
counts. For antagonist activity measurements, 0.1 mM
l-hyoscyamine was added 5 min before agonist incubation.
Membrane preparation and binding assays.
Membrane
homogenates were prepared from cells according to procedures described
previously (Dörje et al., 1991) with modifications. A9
L cells were collected using a cell scraper and washed twice with
ice-cold binding buffer. The pellets were resuspended in a small volume
of binding buffer and homogenized by a Brinkmann Polytron. Large
cellular debris and nuclei were removed by centrifugation (2500 × g) for 10 min at 4°C. Membrane proteins were pelleted by
high speed centrifugation (50,000 × g) for 30 min at
4°C and suspended in binding buffer. Protein concentrations were
determined by a modified Lowry method (Lowry et al., 1951,
Markwell et al., 1981). The membrane homogenates were stored
at 
70°C for future use.
[3H]-(R)-QNB saturation binding assays and ligand
inhibition binding assays were carried out in 12 × 75 mm
borosilicate glass tubes in 1 ml total volume in triplicate. Eight
different concentrations ranging from 5 to 300 pM were used in
[3H]-(R)-QNB saturation binding experiments. In ligand
inhibition binding experiments, 14 different concentrations of test
ligand were used. The total binding and nonspecific binding were
determined in the absence and presence of 1000-fold excess of cold QNB,
respectively. Reactions began with the addition of membrane proteins to
mixtures and were incubated at room temperature for 2 hr. The
incubations were terminated by addition of 5 ml ice-cold binding buffer
and rapid transfer to Whatman GF/B filters that were soaked with cold binding buffer containing 0.3% poly(ethylenimine) immediately before
filtration.
In stability tests of Hm1(WT) receptors and Hm1(Ser388Tyr, Thr389Pro)
receptors, membrane homogenates were proportioned into 15 ml of binding
buffer. For heat stability tests, tubes were incubated in a 37°C
water bath for 1 to 3 hr. Tubes were kept on ice to serve as a control.
For the freezing/thawing stability test, tubes were immersed in a dry
ice/ethanol mixture for 60 min followed by a 37°C incubation for 10 min. This cycle was repeated one, two or three times. The total binding
activity and nonspecific binding activity were determined by a
one-point binding assay using a saturating concentration of 0.5 nM
[3H]-(R)-QNB in the absence and presence of 0.5 µM cold
QNB, respectively. Total specific binding activity was expressed as the
percentage of control sample that was kept on ice. To test protection
by the antagonist l-hyoscyamine and the agonist
acetylcholine, designated concentrations of antagonist or agonist were
added to tubes before the freezing/thawing cycle. Tubes containing
ligands, but kept on ice, served as secondary controls.
Data analysis.
All data were analyzed using nonlinear least
squares curve-fitting as implemented in DeltaGraph for Macintosh (1993, DeltaPoint Inc.). PI hydrolysis data were fit to a one-site stimulation
model. Binding data were fit to one-site, two-site and/or three-site binding models. Statistical comparisons were applied using an F test
with 
set at the 0.05 level. More complex models were chosen only if
they provided a significantly better fit to the data from each
experiment. Ki values were calculated from
IC50 values according to the formula, Ki = IC50/(1 + L/Kd),
described by Cheng and Prusoff (1973).
| |
Results |
Receptor expression and antagonist binding properties.
A total
of 18 cell lines were chosen after G418 selection for screening using a
whole cell [3H]-(R)-QNB binding assay, and were found to
have specific binding with varying expression levels (data not shown).
Table 1 lists Hm1(WT) and Hm1(Ser388Tyr,
Thr389Pro) expression levels and binding affinities for
[3H]-(R)-QNB. Three cell lines expressing the mutant
receptor are included for comparison. All three cell lines expressed
Hm1(Ser388Tyr, Thr389Pro) receptors at significantly higher levels than
the wild-type receptor, and exhibited slightly lower affinities for QNB
than Hm1(WT) receptors.
To determine the effect of the double mutations on the binding of other
muscarinic antagonists, we tested four ligands belonging to different
structural classes (see table 2; fig.
2). All tested antagonists exhibited
slightly reduced (2.4- to 4.2-fold) affinities for the mutant receptor
as compared to the wild-type receptor.
Receptor functional properties.
The double mutations were
originally designed to determine if constitutive activity could be
observed for the m1 receptor subtype as found previously for the mutant
m5 receptor subtype (Spalding et al., 1995). Initially, it
was found that the mutant receptor was extremely sensitive to both
agonists and antagonists, and all tested antagonists appeared to be
inverse agonists with significant inhibition (up to 50%) of basal IP
levels, when the PI assays were conducted in DMEM with serum
supplements. Acetylcholine was found to have variable responses only at
high concentrations, presumably due to the presence of cholinesterases
within the serum as previously shown (Spalding et al., 1995;
Burstein et al., 1995). PI hydrolysis assays conducted in
PBS exhibited increased maximal agonist responses and decreased basal
activity levels and antagonist inhibition (data not shown). These
preliminary results suggested that potentially active compound(s) exist
in DMEM and/or serum. To test this possibility, we compared KH buffer,
and DMEM with and without serum supplements as incubation media to
determine their effects on basal IP levels and the activity of 100 µM
NMS. As indicated in table 3, basal IP
levels were lowest in KH buffer and highest in DMEM with serum
supplements. In addition, 100 µM NMS exhibited significant inhibition
of basal activity in DMEM with or without serum supplements, but much
lower inhibition was observed in KH buffer. Choline is a normal
component in DMEM at 28.6 µM yet was reportedly inactive at Hm5 wild
type and Hm5 with Ser465Tyr and Thr466Pro mutations (Spalding et
al., 1995). In follow-up studies, we found that choline
surprisingly showed strong agonist activity at Hm1(Ser388Tyr,
Thr389Pro) receptors at 1 mM (see table
4). The average stimulation produced by
30 µM choline is 150 ± 50% (n = 3) above basal
levels for Hm1(Ser388Tyr, Thr389Pro) receptors when tested in KH
buffer. In DMEM with or without serum supplement, most of the
inhibition produced by antagonists (inverse agonists) was due to the
elevated basal levels associated with choline in DMEM and trace amount
of endogenous agonists in serum. Thus the widely used KH buffer, but
not DMEM with serum supplement, was used in subsequent PI assays.
View this table:
[in this window]
[in a new window]
|
TABLE 4
Pharmacology of muscarinic ligands on Hm1(WT) and Hm1(Ser388Tyr,
Thr389Pro) receptors stably expressed in A9 L cells
|
|
Table 4 summarizes the levels of stimulation for nine commercially
available agonists (including choline and methylcarbachol) and four
antagonists at the 1 mM concentration at both Hm1(WT) and
Hm1(Ser388Tyr, Thr389Pro) receptors. A comparison of their potencies is
listed in table 5. The endogenous
muscarinic agonist acetylcholine displayed full agonist activity (table
4; fig. 3), as did carbachol and
oxotremorine-M at Hm1(WT) receptors. Methylcarbachol, a selective
nicotinic agonist, exhibited partial agonist activity at Hm1(WT)
expressed in A9 L cells (fig. 3). Methylcarbachol-mediated PI
hydrolysis was completely inhibited by 0.1 mM l-hyoscyamine
(the active enantiomer of atropine).
View this table:
[in this window]
[in a new window]
|
TABLE 5
Comparison of potencies of muscarinic agonists at Hm1(WT) and
Hm1(Ser388Tyr, Thr389Pro) receptors stably expressed in A9 L cells
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 3.
PI hydrolysis mediated by classical muscarinic
agonists (acetylcholine and carbachol) as well as methylcarbachol and
choline. Hm1(WT) and Hm1(Ser388Tyr, Thr389Pro) receptors expression
levels were listed in table 4. Data represent the mean ± S.E.M.
from a minimum of three assays and are fit to a one-site stimulation
model.
|
|
At Hm1(Ser388Tyr, Thr389Pro) receptors, all agonists tested, including
choline and methylcarbachol, exhibited full agonist activities with
about 500% stimulation above basal levels. Furthermore, all ligands
displayed higher potencies at the mutant receptor than at wild type
receptors, especially acetylcholine and carbachol (see table 5; fig. 3)
with over 300- and 200-fold increases in potencies, respectively. The
full agonist activities of choline and methylcarbachol were inhibited
by 0.1 mM l-hyoscyamine.
The m1 selective antagonist, trihexyphenidyl, was surprisingly found to
stimulate PI accumulation in A9 L cells expressing Hm1(WT) or
Hm1(Ser388Tyr, Thr389Pro) receptors, however, the effects were not
blocked by 0.1 mM l-hyoscyamine, which completely blocked carbachol activity (table 4). This result suggested that
trihexyphenidyl stimulates PI hydrolysis through receptor(s) other than
human m1 receptors. To test this possibility, we measured
trihexyphenidyl-stimulated PI hydrolysis in untransfected A9 L cells.
At 1 mM, trihexyphenidyl produced a maximal stimulation of 130 ± 15% (n = 4) above basal levels. This activity was
comparable to that found in transfected A9 L cells expressing Hm1(WT)
or Hm1(Ser388Tyr, Thr389Pro) receptors (table 4), although 1 mM
acetylcholine (2.9 ± 9.9% above basal, n = 3)
and carbachol (0.15 ± 6.6% above basal, n = 3)
did not exhibit activities in untransfected A9 L cells. Three other
tested antagonists did not inhibit basal activity at wild-type
receptors, yet caused small but significant inhibition of basal
activity at mutant receptors, indicating that the mutant receptor may
be constitutively activated, but to a limited degree. We were unable to
obtain a dose-response inhibition curves for l-hyoscyamine in individual assays. Curve-fitting with data averaged from three experiments indicated a maximal inhibition of 13% with an
IC50 value of 1.9 nM (fig. 3).
Hm1(Ser388Tyr, Thr389Pro) receptors were expressed at relatively higher
levels than Hm1(WT) receptors, although the expression levels varied
from passage to passage. To determine if the activity of choline and
enhanced agonist activities at the mutant receptor were due to
significantly higher expression levels than Hm1(WT) receptors, we
created A9 L cell lines which overexpressed Hm1(WT) receptors. As
indicated in table 6 and figure
4A, with overexpression of Hm1(WT) in A9
L cells, 30 µM choline (the concentration found in DMEM without serum
supplements) showed negligible activity, although 1 mM choline
displayed strong ability (up to 370 ± 110% above basal) to
stimulate PI metabolism. To determine EC50 values for
choline at overexpressed Hm1(WT), up to 10 mM choline was used in PI
assays. There was no significant change in EC50 values for
choline at different levels of Hm1(WT) receptor expression (table 6).
With overexpression of Hm1(WT) receptors in A9 L cells, acetylcholine
exhibited significantly enhanced efficacies and potencies (fig. 4A;
table 6). The partial agonist arecoline (fig. 4B) functioned as a full
agonist in stimulating PI hydrolysis to 1100 ± 160%
(n = 2) above basal with an EC50 of
0.87 ± 0.055 µM (n = 2). Muscarinic
antagonists, l-hyoscyamine (7.9 ± 3.1%, n = 5), NMS (3.9 ± 5.1%, n = 5),
and pirenzepine (8.7 ± 11%, n = 5), did not show
significant inhibition of basal activity; although DMEM with serum
supplements exhibited a stimulation of 49 ± 12% (n = 6) above basal levels at the overexpressed Hm1(WT)
receptors. Generally, both acetylcholine and choline activities were
proportional to the expression levels of Hm1(WT), however, choline
efficacy and potency were much lower than acetylcholine at
overexpressed Hm1(WT) receptors.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of expression levels of Hm1(WT) in A9 L cells
on agonist dose response profiles. Receptor expression levels were
listed in tables 4 and 6. Data represent the mean ± S.E.M. from
two to five assays and are fit to a one-site stimulation model. A,
Acetylcholine and choline dose responses. For convenient comparison, an
acetylcholine dose response curve from figure 3 is included. B,
Arecoline dose responses.
|
|
To ensure that the enhanced agonist activity at the mutant receptors or
overexpressed Hm1(WT) receptors was not due to enriched G-proteins
and/or phospholipase C in transfected cells, PI hydrolysis activity by
20 mM NaF was measured in untransfected and transfected A9 L cells
expressing high levels of Hm1(WT) and mutant receptors (data not shown
here). There was no significant difference in the effects of NaF
between untransfected and transfected A9 L cells overexpressing Hm1(WT)
receptors; however A9 L cells expressing Hm1(Ser388Tyr, Thr389Pro)
receptors displayed significantly reduced (about 25%) PI hydrolysis
with 20 mM NaF.
Agonist inhibition binding properties.
As seen with the
constitutively-active m5 receptor (Spalding et al., 1995),
antagonists did not show significantly different binding affinities to
Hm1(Ser388Tyr, Thr389Pro) receptors. Agonists however, bound with much
higher affinities to the mutant receptor than to Hm1(WT) receptors (see
table 7; fig. 5-7). Full agonists at
Hm1(WT) receptors, such as acetylcholine and carbachol, had a much
greater shift in potency than partial agonists, such as arecoline and
APE. Choline showed the smallest shift in affinity at the mutant m1
receptor. All agonists, including choline and methylcarbachol,
displayed multiple affinity states at Hm1(Ser388Tyr, Thr389Pro)
receptors, including those classical partial agonists which did not
exhibit multiple binding affinity states at Hm1(WT), such as arecoline,
APE, oxotremorine and pilocarpine. Generally, the multiple binding
affinity states were easily observed for full agonists in a
GTP-sensitive manner, as for acetylcholine and carbachol at the Hm1(WT)
receptors (table 7; figs. 5 and 6), presumably due to their strong
abilities to induce receptor conformational changes recognizable by
G-proteins. Acetylcholine exhibited three binding sites at Hm1(WT)
receptors, and the high affinity binding sites disappeared in the
presence of 100 µM GppNHp.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 5.
Acetylcholine inhibition binding properties of
Hm1(WT) and Hm1(Ser388Tyr, Thr389Pro) receptors in the presence of 100 pM or 200 pM [3H]-(R)-QNB. Assays were performed in
triplicate. Data represent the mean (± S.E.M.) from a minimum of three
experiments (table 7).
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6.
Carbachol inhibition binding properties of Hm1(WT)
and Hm1(Ser388Tyr, Thr389Pro) receptors in the presence of 100 pM or
200 pM [3H]-(R)-QNB. Assays were performed in triplicate.
Data represent the mean (± S.E.M.) from a minimum of three experiments
(table 7).
|
|
To determine if the coupling interaction between the Hm1(Ser388Tyr,
Thr389Pro) mutant receptors and G-proteins is GTP-sensitive, acetylcholine, arecoline and choline inhibition binding properties were
examined in the presence of GppNHp (see table 7; figs. 5, 7 and 8).
Acetylcholine and arecoline displayed one-site binding profiles with
Ki values of 37 ± 5.7 nM
(n = 2) and 0.57 µM (n = 1),
respectively, in the presence of 100 µM GppNHp. Choline retained a
two-site binding profile (n = 3) yet with a
significantly reduced percentage of high affinity sites in the presence
of 100 µM GppNHp, indicating the possibility that the interaction
between the mutant receptor and G-protein was relatively strong. To
test this possibility, we measured choline inhibition binding in the presence of 200 µM of GppNHp. With the higher concentration of GppNHp, choline exhibited only a one-site binding profile as shown in
figure 8 (n = 2).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 7.
Arecoline inhibition binding properties of Hm1(WT)
and Hm1(Ser388Tyr, Thr389Pro) receptors in the presence of 100 pM or
200 pM [3H]-(R)-QNB. Assays were performed in triplicate.
Data refer to table 7.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 8.
Choline inhibition binding properties of Hm1(WT) and
Hm1(Ser388Tyr, Thr389Pro) receptors in the presence of 100 pM or 200 pM
[3H]-(R)-QNB. Assays were performed in triplicate. Data
represent the mean (± S.E.M.) from three experiments (table 7).
|
|
Receptor stability.
As indicated in table 1, frozen mutant
receptors lost many more binding sites than frozen wild-type receptors
(30-50 vs. 10%) as compared with fresh membrane
preparations, respectively. Both frozen Hm1(WT) and Hm1(Ser388Tyr,
Thr389Pro) receptors exhibited homogeneous one-site binding with
similar Kd values for [3H]-(R)-QNB
as observed for fresh membrane preparations, respectively. This
suggested that the mutant receptor may not be as stable as the
wild-type receptor. To determine if the mutant receptor was physically
unstable to heat treatment as found previously for a mutant
beta-2 adrenergic receptor (Gether et al., 1997),
membrane preparations were incubated at 37°C for 0 to 3 hr. The total
remaining binding activities were measured by a one-point
[3H]-(R)-QNB binding assay using a saturating
concentration (0.5 nM). Unexpectedly, wild-type and mutant receptors
displayed similar stability to 37°C incubation for up to 3 hr (data
not shown here). When Hm1(WT) and Hm1(Ser388Tyr, Thr389Pro) receptors
were exposed to freezing/thawing treatments (fig.
9), more than 70% of total Hm1(WT)
binding activity was recovered after three cycles, as compared to less
than 50% of total Hm1(Ser388Tyr, Thr389Pro) binding activity after the
same treatment.
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 9.
Freezing/thawing stability tests of Hm1(WT) and
Hm1(Ser388Tyr, Thr389Pro) receptors. Membrane homogenates were
frozen/thawed as described in "Materials and Methods." The
remaining total binding and nonspecific binding were determined at the
saturating concentration of 0.5 nM of [3H]-(R)-QNB in the
absence and presence of 0.5 µM (R)-QNB. Specific binding was
presented as the mean (± S.E.M., n = 3) percentage of
the control sample, which was kept on ice.
|
|
To test if receptors could be protected from this loss by agonists or
antagonists, different concentrations of the antagonist l-hyoscyamine or agonist acetylcholine were added while the
membranes were treated by freezing/thawing. Surprisingly, the
percentage loss of specific binding sites was decreased from 58%
(n = 4) to 40% (n = 2) in the presence
of 10 µM acetylcholine but not with 3 to 10 nM
l-hyoscyamine.
| |
Discussion |
A Hm5 receptor with Ser465Tyr and Thr466Pro mutations at the
junction of TM VI and Ne3 was reported to have constitutive activity as
well as increased potencies and binding affinities for agonists but not
for antagonists (Spalding et al., 1995). In this study, we
made equivalent Ser388Tyr and Thr389Pro mutations at the junction of TM
VI and Ne3 of the Hm1 receptor. The mutant receptor was potently
activated by muscarinic agonists and exhibited much more significant
inhibition of basal activity by antagonists when PI assays were
conducted in DMEM with or without serum supplements than in KH buffer.
This led us to examine the activity of choline, a component of DMEM,
which is reportedly inactive at Hm5(Ser465Tyr, Thr466Pro) receptors. At
30 µM (the concentration found in DMEM), choline stimulated PI
hydrolysis by 150 ± 50% above basal at the mutant receptors. KH
buffer was used for further characterization of Hm1(Ser388Tyr,
Thr389Pro) to exclude the effects of choline and potential traces of
the endogenous muscarinic agonist acetylcholine in serum. Choline
showed no activity at A9 L cells expressing low levels of Hm1(WT)
receptors, but displayed partial agonist activity at overexpressed
Hm1(WT) receptors with low potency. At Hm1(Ser388Tyr, Thr389Pro)
receptors, choline exhibited full agonist activity with significantly
increased potency.
Overexpression of Hm1(WT) resulted in small increases in efficacy
(2.3-fold) and potency (12-fold) for acetylcholine. However, the double
mutations increased acetylcholine potency by over 300-fold and high
binding affinity by 190-fold. Arecoline showed full agonist activity
comparable to acetylcholine at the overexpressed Hm1(WT) and gained a
4.1-fold increase in potency, although the double mutations increased
arecoline potency by 60-fold. Both acetylcholine and arecoline
exhibited similar binding affinities at low and high levels of Hm1(WT)
receptor expression (data not shown), suggesting that high receptor
expression levels can not account for the increased agonist activities.
Compared with untransfected or transfected A9 L cells overexpressing
Hm1(WT) receptors, the slightly reduced PI hydrolysis activity by 20 mM
NaF at A9 L cells overexpressing the mutant receptors indicates that
possible changes of important signalling proteins could not contribute
to significantly enhanced agonist activity. These data indicate that
overexpression of Hm1(WT) receptors results in enhanced agonist
efficacies and potencies. It is possible that the enhanced agonist
activities at the mutant receptor could be due partially to high
expression levels. However, the double mutation produced much
greater increases in agonist potencies than can be attributed to
elevated receptor expression. Overexpression of Hm1(Ser388Tyr,
Thr389Pro) may contribute to choline activity as in overexpressed
Hm1(WT), but the double mutation rendered choline full agonist
activity. Further studies are necessary to examine the effects of
receptor expression levels on agonist activity. Creation of stable cell
lines expressing different levels of mutant receptors could assess the
possibility that enhanced agonist activity or constitutive activity is
dependent on the cell line examined as previously observed in mutant
human m1 receptors (Högger et al., 1995; Shockley
et al., 1997).
As shown in this study, enhanced agonist activities can be due to
elevated receptor expression and/or mutations of Hm1(WT) receptors.
Serum showed significant activity at overexpressed Hm1(WT) and mutant
receptors, suggesting that serum contains unknown compound(s) with
muscarinic agonist activities, perhaps acetylcholine and/or choline.
Also, other constitutively activated mutant receptors exhibited higher
binding affinities and potencies for agonists as compared to wild-type
receptors (Kjelsberg et al., 1992; Samama et al.,
1993; Högger et al., 1995; Perez et al.,
1996). Therefore, it is important to exclude endogenous agonist
activities in growth media in characterizing overexpressed wild-type
and mutant receptors. A defined buffer system rather than growth media
is the ideal choice to test potentially active compounds at wild-type
and mutant receptors.
Methylcarbachol was found to have weak muscarinic agonist activity in
transfected cell lines but not in mouse and rat brain tissue (Wang
et al., 1994). Here, we found that methylcarbachol exhibited
partial agonist activity at Hm1(WT) and full agonist activity at
Hm1(Ser388Tyr, Thr389Pro) receptors expressed in A9 L cells. The
activities were blocked by l-hyoscyamine in A9 L cells
expressing either Hm1(WT) or mutant receptors. l-Hyoscyamine also blocked choline-dependent PI hydrolysis in A9 L cells expressing Hm1(Ser388Tyr, Thr389Pro) receptors. Surprisingly, the clinically useful muscarinic antagonist trihexyphenidyl showed comparable PI
hydrolysis activities at untransfected and transfected A9 L cells
expressing either Hm1(WT) or mutant receptors. The weak stimulation was
not inhibited by l-hyoscyamine. These data suggest that
choline and methylcarbachol both act as agonists at m1 receptors while
trihexyphenidyl may elevate PI hydrolysis through a mechanism other
than activation of m1 receptors.
In the Hm5(Ser465Tyr, Thr466Pro) receptors, no significant GppNHp
effect was observed for agonist binding (Spalding et al., 1995). In the Hm1(Ser388Tyr, Thr389Pro) receptor, all tested agonists showed multiple binding affinity states. The GTP shift in agonist binding affinities for Hm1(Ser388Tyr, Thr389Pro) receptors was associated with a disappearance of the high affinity binding sites. In
the presence of 100 µM GppNHp, the high affinity binding of acetylcholine and arecoline disappeared, although high affinity binding
of choline to the mutant receptor was reduced from 19 to 9%. The
choline high affinity binding disappeared in the presence of 200 µM
GppNHp. It is reasonable to believe that all high-affinity agonist
binding, including methylcarbachol, would be GppNHp sensitive. These
data suggest that the interaction between G-proteins and the
choline/Hm1(Ser388Tyr, Thr389Pro) receptor complex might be stronger
than that between G-proteins and the acetylcholine/Hm1(Ser388Tyr, Thr389Pro) complex. Thus different ligands may induce conformational changes to varying degrees, resulting in different susceptibility to
GTP modulation. The enhanced interaction may contribute to the full
agonist activity (yet low potency) of choline, despite the fact that it
has almost the same low binding affinity for both Hm1(WT) and
Hm1(Ser388Tyr, Thr389Pro) receptors. The enhanced interaction could
also contribute to elevated activities of classical partial muscarinic
agonists (e.g., arecoline) at Hm1(Ser388Tyr, Thr389Pro)
receptors.
A constitutively activated alpha-1B adrenergic receptor was
unstable to heat treatment and showed exaggerated conformational changes upon ligand binding (Gether et al., 1997). In this
study, we applied a simple stability test to obtain some information about the structural and conformational stability of the mutant m1
receptor. The data indicate that the mutant receptor is as stable as
the wild-type receptor at 37°C for incubations up to 3 hr, but is
much more vulnerable to freezing/thawing treatments. The presence of
acetylcholine protected the mutant receptor, thereby maintaining
[3H]-(R)-QNB binding activity. It is possible that the
mutant receptor is in structurally and conformationally flexible
states, and agonist, but not antagonist, binding could induce
conformational changes to stabilize the receptor. As mentioned above,
conformational changes were directly reflected by the existence of
multiple affinity states for all tested agonists. To our knowledge,
this is the first experimental evidence that mutant receptors may not
be as stable as wild-type receptors to freezing/thawing conditions.
KSHV-GPCR showed constitutive activity in stimulating PI hydrolysis and
cell proliferation in COS-1 cells (Arvanitakis et al., 1997)
and oncogenic angiogenesis in NIH-3T3 cells (Bais et al.,
1998). It is not clear which G protein(s) is involved in this
constitutively activated pathway in NIH-3T3 cells, but the constitutive
activity of KSHV-GPCR in COS-1 cells was pertussis-toxin insensitive,
probably coordinated through Gq/11 proteins (Arvanitakis et al., 1997). It was reported recently that Hm1 and
G
12 (also 
subunits) and the small GTP-binding
proteins RhoA are involved in a novel signal transduction pathway that
promotes NIH 3T3 cell transformation (Fromm et al., 1997).
It is interesting to notice that the constitutively activated
Hm5(Ser465Tyr, Thr466Pro) receptors were characterized in transfected
NIH-3T3 cells using the R-SAT which is based on the cell transformation
mediated by G-proteins and G-protein coupled receptors (Spalding
et al., 1995). It is unclear if m1 muscarinic receptors and
the KSHV-GPCR act through similar mechanisms, however, it is clear that
NIH-3T3 cells have signal transduction pathways capable of linking
G-proteins and PI metabolism to cell proliferation. It is possible that
the mutant m5 receptor might adopt conformations that constitutively activate the signaling pathway that is used for KSHV-GPCR-induced oncogenic angiogenesis. In this study, we characterized Hm1(WT) and
mutant receptors stably expressed in A9 L cells by monitoring ligand-dependent PI hydrolysis, presumably mediated by
Gq/11 family of G-proteins. We did not observe constitutive
activity for Hm1(WT) receptors, yet recorded a limited degree of
constitutive activity for Hm1(Ser388Tyr, Thr389Pro) receptors. It is
unclear if A9 L cells have comparable machinery for cell proliferation
mediated by G-protein coupled receptors as in NIH-3T3 cells.
Taken together, we have observed that a mutant Hm1 receptor,
Hm1(Ser388Tyr, Thr389Pro), does not exhibit significant constitutive activity but does show enhanced sensitivity to muscarinic agonists, with dramatically increased potencies and binding affinities for agonists but not for antagonists. Moreover, choline, which is a normal
component of growth media and inactive at Hm1(WT) receptors expressed
at low levels, showed partial agonist activity at highly expressed
Hm1(WT) receptors and full agonist activity at Hm1(Ser388Tyr, Thr389Pro) receptors. These data suggest that studies characterizing highly expressed receptors and/or constitutively activated mutant receptors should exclude the presence of potentially active compounds in growth media. Further biochemical and site-directed mutagenesis studies of Hm1(Ser388Tyr, Thr389Pro) receptors should help delineate the amino acid residues involved in muscarinic agonist binding and
receptor activation.
The authors thank Dr. Tom I. Bonner for kindly providing Hm1pcD1
and Dr. Jürgen Wess for pcDneo. We also thank Afif El-Assadi for
his technical assistance.
ACh, acetylcholine;
APE, arecaidine propargyl
ester hydrobromide, DMEM, Dulbecco's Modified Eagle Media;
FBS, fetal
bovine serum;
GppNHp, guanylylimidodiphosphate;
Hm1, human muscarinic
acetylcholine receptor subtype 1;
Hm1(Ser388Tyr, Thr389Pro), Hm1 mutant
receptor with the mutations of Ser388 to Tyr and Thr389 to Pro;
Hm5, human muscarinic acetylcholine receptor subtype 5;
IP, inositol
phosphate;
KH buffer, Krebs-Henseleit buffer, mAChR, muscarinic
acetylcholine receptor;
(R)-QNB, (R)-3-quinuclidinyl benzilate;
Ne3, N-terminal region of the third extracellular domain;
TCA, trichloroacetic acid, TM, transmembrane domain;
WT, wild-type;
KSHV-GPCR, G-protein-coupled receptor of Kaposi's sarcoma-associated
herpesvirus.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
molecular interactions with agonists and antagonists, in
Molecular Mechanisms of Muscarinic Acetylcholine Receptor Function (Wess J ed) pp 19-32,
R. G. Landes Company, Austin, TX.
