Stereoselective synthesis and biodistribution of potent [11C]-labeled antagonists for positron emission tomography imaging of muscarinic receptors in the airways

Article abstract of DOI:10.1021/jm960374w

Quantitation of muscarinic receptors in the lungs in vivo with positron emission tomography (PET) is of clinical interest. For that purpose we decided to develop [¹¹C]-labeled ligands with a high affinity (K(D) < 0.1 nM). Three quaternary muscarinic antagonists, racemic N-methylpiperidin-4-yl 2-cyclohexyl-2-hydroxy-2-phenylacetate methiodide 1a (pK(B) = 10.39), its (R)-isomer 1b (pK(B) = 11.08), and (R,R)-quinuclidin-3-yl 2-cyclohexyl-2- hydroxy-2-phenylacetate methiodide 2 (pK(B) = 11.28), were labeled by reacting [¹¹C]CH₃I with their tertiary amine precursors. The enantiomerically pure tertiary amine precursors were prepared by stereoselective synthesis starting from (R)-(-)-mandelic acid. In vitro binding assay of 1b and 2 demonstrated that both ligands bind with very high affinity to the muscarinic receptor subtypes M₁, M₂, and M₃. They are more potent than the muscarinic antagonist (R)-N-methylquinuclidinyl benzilate ((R)-MQNB). Distribution studies with 1a, 1b, and 2 in control and atropine- treated male Wistar rats demonstrated significant specific binding (90-99% of total issue uptake) in tissues containing cholinoceptors (heart, intestine, lung, pancreas, spleen, stomach, submandibular gland). Because the tissue/plasma concentration ratios of lb are most favorable, this ligand was used for further evaluation. Analysis of plasma samples showed a very rapid clearance (t(1/2) = 0.3 min) of the radioligand 1b and a relatively slow appearance of a hydrophilic metabolite. At 15 min postinjection of 1b, analysis of heart, lungs, and liver showed that respectively 99%, 88%, and 8% of the tissue radioactivity corresponded with the parent compound. Ligand 1b appears to be an excellent candidate for PET studies of mAChR receptors in heart and lungs.

Full text of DOI:10.1021/jm960374w

J . Med. Chem. 1997, 40, 117-124
117
Ster eoselective Syn th esis a n d Biod istr ibu tion of P oten t [11C]-La beled
An ta gon ists for P ositr on Em ission Tom ogr a p h y Im a gin g of Mu sca r in ic
Recep tor s in th e Air w a ys
Ton J . Visser,*,† Aren van Waarde,† Twan J . H. J ansen,† Gerben M. Visser,† Thom W. van der Mark,‡
J an Kraan,‡ Kees Ensing,§ and Willem Vaalburg†
Positron Emission Tomography (PET) Center and Department of Pulmonary Diseases, Groningen University Hospital, P.O.
Box 30001, 9700 RB Groningen, The Netherlands, and Department of Analytical Chemistry & Toxicology, University Center for
Pharmacy, Antonius Deusinglaan 2, 9713 AW Groningen, The Netherlands
Received May 23, 1996X
Quantitation of muscarinic receptors in the lungs in vivo with positron emission tomography
(PET) is of clinical interest. For that purpose we decided to develop [11C]-labeled ligands with
a high affinity (KD < 0.1 nM). Three quaternary muscarinic antagonists, racemic N-meth-
ylpiperidin-4-yl 2-cyclohexyl-2-hydroxy-2-phenylacetate methiodide 1a (pKB ) 10.39), its (R)-
isomer 1b (pKB ) 11.08), and (R,R)-quinuclidin-3-yl 2-cyclohexyl-2-hydroxy-2-phenylacetate
methiodide 2 (pKB ) 11.28), were labeled by reacting [11C]CH3I with their tertiary amine
precursors. The enantiomerically pure tertiary amine precursors were prepared by stereose-
lective synthesis starting from (R)-(-)-mandelic acid. In vitro binding assay of 1b and 2
demonstrated that both ligands bind with very high affinity to the muscarinic receptor subtypes
M1, M2, and M3. They are more potent than the muscarinic antagonist (R)-N-methylquinu-
clidinyl benzilate ((R)-MQNB). Distribution studies with 1a , 1b, and 2 in control and atropine-
treated male Wistar rats demonstrated significant specific binding (90-99% of total issue
uptake) in tissues containing cholinoceptors (heart, intestine, lung, pancreas, spleen, stomach,
submandibular gland). Because the tissue/plasma concentration ratios of 1b are most favorable,
this ligand was used for further evaluation. Analysis of plasma samples showed a very rapid
clearance (t1/2 ) 0.3 min) of the radioligand 1b and a relatively slow appearance of a hydrophilic
metabolite. At 15 min postinjection of 1b, analysis of heart, lungs, and liver showed that
respectively 99%, 88%, and 8% of the tissue radioactivity corresponded with the parent
compound. Ligand 1b appears to be an excellent candidate for PET studies of mAChR receptors
in heart and lungs.
In tr od u ction
pulmonary diseases. The efficacy of â-agonists (sym-
pathicomimetics) and muscarinic antagonists (parasym-
pathicolytics) in such patients changes with the dura-
tion of treatment and with age. The suppression of
bronchial hyperreactivity by sympathicomimetics is
much better after a single administration than when
the drugs are administered chronically. The protection
and airway relaxation of juvenile asthma patients is
much better after treatment with sympathicomimetics
than after treatment with parasympathicolytics. How-
ever patients over 50 years of age respond better to
parasympathicolytic drugs. These observations indicate
an altered drug/receptor interaction after prolonged
treatment and with increasing age. This could be due
to a change of pulmonary receptor density6,7 or to an
altered coupling of the receptors to the contraction
mechanism of the smooth muscle cells.
Muscarinic cholinergic receptors (mAChR) are impor-
tant in many physiological processes, such as smooth
muscle contraction and exocrine and endocrine secre-
tion.1 Cloning studies have confirmed the heterogeneity
of muscarinic receptors, and to date five different
subtype genes (m1, m2, m3, m4, and m5) have been
identified.2 Subtypes m1, m2, m3, and m4 correspond to
the pharmacologically defined M1, M2, M3, and M4
receptors, respectively, and the m5 subtype awaits
pharmacological characterization.3,4 Human lungs con-
tain M1, M2, and M3 receptors.5 M1 receptors facilitate
neurotransmission through parasympathetic ganglia
and enhance cholinergic reflexes. M2 receptors act as
autoreceptors on postganglionic cholinergic nerves and
inhibit acetylcholine release. The M3 subtype mediates
smooth muscle contraction via phosphoinositide hy-
drolysis. Pulmonary mucus secretion is stimulated via
the M1 and M3 subtypes. M4 and M5 receptors have not
been identified in human airways.5 Quantification of
mAChRs in the lungs in vivo with positron emission
tomography (PET) is of clinical interest, especially in
patients suffering from asthma and chronic obstructive
Although a few reports on labeled muscarinic ligands
for quantification of the mAChRs in heart8,9 and brain10,11
with PET have appeared, little attention has been paid
to the airways. A major obstacle in the quantitation of
muscarinic receptors in the airways is the relatively low
receptor density in this tissue. The concentrations in
human peripheral lung range from 25 to 100 fmol/mg
protein.12-14 Because of this low density, high-affinity
ligands (KD < 10-10 M) must be employed to obtain
reasonable (>10) signal to noise ratios in PET.15 Mus-
carinic receptor subtype selective ligands like piren-
zepine (M1),16 methoctramine (M2),16 AFDX 116 (M2),17
hexahydrosiladifenidol (M3),16,18 and 4-DAMP (M3)16,19,21
* Author to whom correspondence should be addressed.
† Positron Emission Tomography (PET) Center, Groningen Univer-
sity Hospital.
‡ Department of Pulmonary Diseases, Groningen University Hos-
pital.
§ University Center for Pharmacy.
X Abstract published in Advance ACS Abstracts, December 15, 1996.
S0022-2623(96)00374-3 CCC: $14.00 © 1997 American Chemical Society

118 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Visser et al.
Sch em e 1
activity studies. The route is depicted in Scheme 2. (R)-
(-)-Mandelic acid 3b was condensed with pivalaldehyde
4 under azeotropic removal of the water formed to give
a mixture of the trans- and cis-dioxolanones. 1H NMR
(200 MHz) showed a cis:trans ratio of 7.5:1. The cis-
dioxolanone 5 was obtained in diastereomerically pure
form after crystallization from pentane/ether in 71%
yield. Deprotonation of the cis-dioxolanone at -80 °C
with LDA did not result in formation of the lithium
enolate, but by carrying out the reaction at higher
temperatures (-40 to -10 °C), deprotonation went very
smoothly.
are not potent enough to be used for PET imaging of
cholinoceptors in the airways. Only a limited number
of drugs have sufficiently high receptor affinity. (R)-
N-Methylquinuclidinyl benzilate ((R)-MQNB) is a potent
muscarinic antagonist,21 but according to Delforge it is
not extracted by the lungs.22 Some esters of glycolic acid
are among the most potent anti-acetylcholine drugs
known. In 1973 Inch23 reported the synthesis of various
glycolic esters which were very potent antagonists in
functional experiments. For instance compounds 1 and
2 (Scheme 1) antagonized the carbachol-induced con-
tractions of guinea pig ileum in picomolar concentra-
tions.24 The (R)-isomer of 1 (pKB ) 11.08) and the (R,R)-
isomer of 2 (pKB ) 11.28) were most potent in functional
experiments.24 Inch synthesized these potent antago-
nists by the degradation of carbohydrate derivatives. We
explored a shorter route for the stereoselective synthesis
of these potent cholinoceptor ligands.
To assess whether these antagonists are useful can-
didates for the quantitation of pulmonary muscarinic
receptors, we reacted the tertiary amine precursors with
CH3I. The present paper describes the stereoselective
synthesis and the binding of the ligands to the M1, M2,
and M3 muscarinic receptor subtypes in vitro as well
as the tissue distribution, metabolism, and clearance
from plasma of the [11C]-labeled quaternized products
1a , 1b, and 2 in male Wistar rats.
Reaction of the lithium enolate with bromocyclohex-
ane did not afford any 5,5-disubstituted dioxolanone 7.
In this reaction step only starting material was isolated
in quantitative yield even after performing the reaction
at room temperature overnight. The lack of reactivity
of the secondary alkyl halides is probably due to steric
repulsion.32 By using 3-bromocyclohexene which is an
allyl halide and probably reacting via a SN2′ type of
mechanism, the 5,5-disubstituted dioxolanone 6 could
be isolated in good yields with a cis:trans ratio of 11:1.
In this reaction step an additional stereogenic center is
introduced (allylic carbon) with a selectivity ratio of 1:2.
1
After crystallization from hexane at -18 °C, H NMR
showed that the trans isomers were no longer present.
Subsequently the additional stereogenic center was
destroyed again by reduction of the double bond with a
catalytic amount of Pd/C and H2, and compound 7 was
obtained in 98% yield with a diastereomerical excess of
>99%. It is also possible to reduce the double bond with
Pd/C and H2 first and to separate the cis and trans
isomers afterward by crystallization from hexane at -18
°C.
Esterification of the 5,5-disubstituted dioxolanone
with N-methyl-4-hydroxypiperidine under the influence
of base (Na or NaH) as well as acid (p-TsOH) under
various reaction conditions gave yields below 10%.
However, deprotection with KOH at room temperature
afforded R-cyclohexyl-R-hydroxy-R-phenylacetic acid 8
in 85% yield. The enantiomeric excess of 8 was deter-
mined by means of a chiral â-cyclodextrin column and
appeared to exceed 99%.33 Reaction of 8 with diazo-
methane afforded the methyl ester 9 in 95% yield.
Transesterification with N-methyl-4-hydroxypiperidine
10 or (R)-3-quinuclidinol34 11 under the influence of
NaH yielded the esters 12b (R) and 13 (R,R) in 40%
and 18% yield, respectively.
Resu lts a n d Discu ssion
Ch em istr y. Some methods have appeared in litera-
ture for the preparation of enantiomerically pure esters
of R-hydroxy acids. These methods include the resolu-
tion of 2-cyclohexyl-2-hydroxy-2-phenylethanoic acid
with quinine25 or ephedrine26 as the resolving agent,
diastereoselective synthesis with 8-phenylmenthol as
the chiral auxiliary,27 and an asymmetric synthesis with
(S)-(anilinomethyl)-2-pyrrolidine.28 In 1984 the group
of Seebach showed that protected R-hydroxy acids could
be R-alkylated with various electrophiles (primary alkyl
halides, allyl halides, and ketones) to produce 5-disub-
stituted dioxolanones in high yield and high diastereo-
selectivities. This process was called self-reproduction
of chirality.29-31 Although Inch synthesized the antago-
nists 1 and 2 by the degradation of carbohydrate
derivatives, the route which we selected for the stereo-
selective synthesis of these potent cholinoceptor ligands
is based on the method used by Seebach. This meth-
odology is very attractive because all the isomers of
compounds 1 and 2 are easily prepared in enantiomeri-
cally pure form, in a few, good- to high-yield steps, from
readily available starting materials. Furthermore the
substituents on the R-position of the R-hydroxy esters
can be varied very easily, making this route particularly
attractive for the synthesis of a variety of structurally
related compounds. Therefore this method is highly
interesting for pharmacological studies, e.g. structure-
The racemic piperidyl ester 12a was prepared, start-
ing from racemic mandelic acid 3a . Reaction of 3a with
2,2-dimethoxypropane (Scheme 3) afforded the ace-
tonide 14, which after deprotonation with LDA was
reacted with 3-bromocyclohexene to give the 5,5-disub-
stituted acetonide 15. Again, reaction of the lithium
enolate with bromocyclohexane did not give any result.
Reduction of the double bond with Pd/C and H2 yielded
compound 16, which was subsequently transformed in
a few steps to the racemic piperidyl ester 12a , analogous
to the route as depicted in Scheme 2. Quaternization
of the tertiary amines 12a , 12b, and 13 with [11C]CH3I
in a closed vessel at 110 °C afforded the labeled
muscarinic antagonists 1a , 1b, and 2 in 40-60% radio-
chemical yield and with specific activities of 11.1-44.4
TBq/mmol (300-1200 Ci/mmol), 40 min EOB (Scheme
4).
Dyn a m ic P ET Stu d ies in Ma le Wista r Ra ts. To

PET Imaging of Muscarinic Receptors in the Airways
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 119
Sch em e 2. Stereoselective Synthesis of Muscarinic Antagonists 1b and 2
Sch em e 3. Synthesis of Racemic 1a
Sch em e 4. Quaternization with [11C]Methyl Iodidea
F igu r e 1. Dynamic PET study in male Wistar rats after
injection of racemic 1a , with and without pretreatment of
atropine. The myocardial time activity curves are shown.
Ta ble 1. Biodistribution of Muscarinic Antagonists 1a , 1b,
and 2 in Male Wistar Rats, 15 min Postinjectiona
racemate
1a (N ) 6)
(R)-isomer
1b (N ) 6)
(R,R)-isomer
2 (N ) 6)
controls
bladder
cerebellum
cortex
a
The same conditions were used for the labeling of compound
13.
4.71 ( 5.10 4.68 ( 1.53
5.11 ( 1.61
BLD
BLD
BLD
BLD
BLD
BLD
obtain insight in the pharmacokinetics of the tracers,
the time course of radioactivity in an organ with very
high receptor density, i.e. the heart of male Wistar rats,
was monitored over a 50 min period, using a Siemens
ECAT 951/31 PET camera. In this preliminary study,
male Wistar rats (240 g) were injected intravenously
in the tail vein with the racemic piperidyl ester 1a .
Heart was visible in two planes, and the time-activity
curve (Figure 1) showed an initial rapid, partial wash-
out, followed by a slow decay. In the same figure, the
myocardial time-activity curve of rats pretreated with
atropine is shown. This curve demonstrates that the
uptake of 1a is nearly completely blocked by atropine.
These results are a very good indication that the uptake
of 1a is receptor mediated and that there is low
nonspecific binding. The curves also show that a
plateau value is already reached within 10 min. There-
fore, a distribution period of 15 min was used in the
tissue distribution studies.
fat
heart
intestine
kidney
liver
0.50 ( 0.37 0.10 ( 0.08
0.44 ( 0.16
8.45 ( 1.47 12.78 ( 1.17b 12.18 ( 1.31
3.39 ( 1.02 6.08 ( 2.23b 5.40 ( 2.40
2.48 ( 0.42 2.75 ( 0.57
6.27 ( 1.44 5.42 ( 1.00
3.10 ( 0.47
5.73 ( 1.48
lung
0.67 ( 0.12 1.03 ( 0.22b 1.02 ( 0.10
muscle
pancreas
plasma
red blood cells
spleen
0.08 ( 0.04 0.07 ( 0.04
0.31 ( 0.14
1.97 ( 0.39 2.85 ( 0.50b 2.56 ( 0.51
0.14 ( 0.02 0.10 ( 0.01b 0.08 ( 0.01
BLD
BLD
BLD
1.41 ( 0.30
0.97 ( 0.27 1.23 ( 0.15
stomach
1.62 ( 0.40 2.76 ( 0.25b 1.90 ( 0.35
submandibular gland 1.68 ( 0.28 2.64 ( 0.39b 2.27 ( 0.34
testes
trachea
0.07 ( 0.02 0.12 ( 0.05
0.11 ( 0.04
0.54 ( 0.04 0.89 ( 0.15b 0.90 ( 0.22
a
Data are listed as mean ( SD and expressed as standardized
(body weight-corrected) uptake values. BLD ) below limit of
b
detection. Significant difference (one-way ANOVA p < 0.05)
between racemate 1a and (R)-isomer 1b. No significant difference
between (R)-isomer 1a and (R,R)-isomer 2 was observed.
noceptors are presented in Table 3. These ratios are
important because they give a good indication of the
signal to noise ratios which will be obtained in the PET
images. Tissue uptake was determined by ex vivo
counting, 15 min postinjection, and expressed as a
Tissu e Distr ibu tion Stu d ies in Ma le Wista r Ra ts.
The distribution of radioligands 1a , 1b, and 2 in various
tissues of male Wistar rats is shown in Table 1. The
tissue/plasma ratios of some organs containing choli-

120 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Visser et al.
Ta ble 2. Biodistribution of Muscarinic Antagonists 1a , 1b,
and 2 in Male Wistar Rats, 15 min Postinjection with
male Wistar rats. Its clearance from plasma was
determined and shown to be very rapid (Figure 2). The
clearance demonstrates a biphasic pattern with a half-
life of 0.3 min for the initial rapid distribution phase
and a half-life of 6.2 min for the slow phase. We have
also studied the metabolism of radioligand 1b in rat
tissues and plasma in vivo. For quantitation of the
receptor density it is important to know whether there
are labeled metabolites in the target tissues and in
blood. Therefore we analyzed extracts of heart, lung,
and liver 15 min after intravenous administration of the
antagonist 1b. HPLC analyses of the tissue extracts
indicated that respectively 99%, 88%, and 8% of radio-
activity in heart, lungs, and liver was the parent
compound. The parent compound eluted with a reten-
tion time of 7 min; one hydrophilic metabolite eluted
with a retention time of 4.1 min. We also analyzed the
time course of metabolism of ligand 1b in plasma over
a period of 40 min. The percentage of total plasma
radioactivity which coeluted with the parent compound
decreased from 99.5% at 0 min to 46% at 40 min
postinjection.
Pretreatment of the Muscarinic Antagonist Atropine 2.5 mg/kga
racemate
1a (N ) 5)
(R)-isomer
1b (N ) 5)
(R,R)-isomer
2 (N ) 6)
blocked
bladder
cerebellum
cortex
9.72 ( 9.77 5.92 ( 4.27
4.50 ( 2.82
BLD
BLD
BLD
BLD
BLD
BLD
fat
heart
intestine
kidney
liver
0.23 ( 0.20 0.37 ( 0.24
0.22 ( 0.15
0.16 ( 0.02b 0.12 ( 0.01b,c 0.10 ( 0.02b
1.10 ( 0.52b 0.97 ( 0.75b 4.68 ( 2.32d
1.49 ( 0.06b 4.32 ( 0.67b,c 2.86 ( 0.72
8.10 ( 0.78b 9.41 ( 1.18b 7.41 ( 1.83
0.14 ( 0.02b 0.11 ( 0.02b 0.12 ( 0.04b,d
lung
muscle
pancreas
plasma
red blood cells
spleen
0.07 ( 0.05 0.07 ( 0.05
0.14 ( 0.13
0.34 ( 0.12b 0.35 ( 0.11b 0.38 ( 0.16b
0.13 ( 0.02 0.19 ( 0.03b 0.12 ( 0.02b
B.L.D.
B.L.D.
B.L.D.
0.18 ( 0.03b 0.12 ( 0.02b,c 0.20 ( 0.08b,d
0.45 ( 0.28b 0.22 ( 0.07b 0.23 ( 0.07b
stomach
submandibular gland 0.20 ( 0.06b 0.23 ( 0.12b 0.25 ( 0.06b,d
testes
trachea
0.11 ( 0.06 0.16 ( 0.09
0.14 ( 0.05
0.24 ( 0.04b 0.26 ( 0.02b 0.31 ( 0.05b,d
a
Data are listed as mean ( SD and expressed as standardized
(body weight-corrected) uptake values. BLD ) below limit of
b
detection. Significant difference (Student’s t-test, p < 0.05)
Ra d ioliga n d Bin d in g Assa ys. Radioligand binding
affinities and reference muscarinic agents are docu-
mented in Table 4. The binding studies were conducted
in rat tissue and calf brain homogenates. The nonselec-
tive radioligand [3H]-N-methylscopolamine was used to
label muscarinic receptors in preparations of rat cortex
(M1), calf brain (M1), and rat heart (M2). Binding to the
M3 muscarinic subtype in guinea pig ileum was deter-
mined by Nova Screen (Hanover, MD) with [3H]-N-
methylscopolamine as the radioligand. In their assay,
4-DAMP was used as a reference compound. The
results show that ligands 1b and 2 have a very high
affinity for all the muscarinic receptor subtypes. The
antagonists are even more potent than (R)-MQNB.
Compound 1b shows the highest affinity for the M1 and
M3 receptors. In rats 91% of the pulmonary cholinocep-
tors is of the M2 subtype.35 Cholinergic receptor densi-
ties are in the order of 16-35 fmol/mg of protein,36,37
which is less than in human lungs. The receptor density
in human lungs varies between 25 and 100 fmol/mg of
protein.12-14 In human lungs the dominant muscarinic
receptors are of the M1 (60%) and the remaining
receptors are of the M3 subtype. There is only a very
small population of the M2 subtype present.13 So it is
clear that the rat studies in comparison to human
studies can be seen as a worst case scenario. Therefore
lung/blood ratios in humans will probably be higher
than those observed in rats, and we intend to evaluate
compound 1b in humans.
between controls and atropine treated animals. c Significant dif-
ference (one-way ANOVA, p < 0.05) between racemate 1a and (R)-
d
isomer 1b. Significant difference (one-way ANOVA, p < 0.05)
between (R)-isomer 1b and (R,R)-isomer 2.
differential absorption ratio (DAR) which is defined as
(cpm recovered/g of tissue)/(cpm injected/g of body
weight). These data demonstrate that the tissues rich
in mAChR (heart, intestine, pancreas, submandibular
gland) show the highest accumulation of radioactivity.
Although there are many muscarinic receptors located
in the brain, none of the ligands accumulated in this
tissue. Probably because of the low lipophilicity, the
quaternary antagonists cannot cross the blood/brain
barrier. Uptake in erythrocytes was below the limit of
detection. Injection of the more potent (R)-isomer (1b)
instead of its racemate (1a ) caused a significant increase
in uptake in nearly all the organs containing choli-
noceptors, and the plasma radioactivity was signifi-
cantly lowered. This resulted in a doubling of the tissue
to plasma ratios in these organs. Ligand 1b demon-
strated moderate lung/plasma ratios (10 ( 3) and lung/
blood ratios (20 ( 3) at 15 min postinjection. Injection
of (R,R)-quinuclidinyl ester (2) instead of the (R)-
piperidyl ester (1b) did not improve uptake values and
tissue/plasma ratios.
To prove that the uptake of the radioligands repre-
sents binding to cholinoceptors, blocking studies were
performed in anaesthesized male Wistar rats. In these
studies a solution of atropine (2.5 mg/kg) in saline was
injected intravenously 1 min before administration of
the radioligand. In Table 2 the results of the blocking
experiments are shown. Radioactivity in the target
organs (heart, intestine, kidney, liver, lung, pancreas,
spleen, stomach, and trachea) was significantly reduced
when the rats were pretreated with atropine. In heart
and lungs, respectively 99% and 84% of the uptake was
blocked by atropine, suggesting that the uptake in these
organs is indeed receptor-mediated. Injection of the
(R,R)-ligand 2 instead of the (R)-compound 1b slightly
increased nonspecific binding in intestine, lung, spleen,
submandibular gland, and trachea.
Con clu sion
We have synthesized racemic N-methylpiperidin-4-
yl 2-cyclohexyl-2-hydroxy-2-phenylacetate methiodide
(1a ), its more potent (R)-isomer 1b, and (R,R)-quinu-
clidin-3-yl 2-cyclohexyl-2-hydroxy-2-phenylacetate me-
thiodide (2) (most potent of the four possible isomers)
by means of a novel stereoselective synthesis starting
from (R)-(-)-mandelic acid. The binding affinity of the
ligands 1b and 2 to the M1, M2, and M3 muscarinic
receptor subtypes was determined in vitro. The com-
pounds appeared to be more potent than (R)-MQNB.
The tertiary amine precursors were labeled with [11C]-
CH3I and tested in rats. The time course of myocardial
radioactivity in male Wistar rats was monitored with a
PET camera, and the time-activity curve showed an
Clea r a n ce fr om P la sm a a n d Meta bolism . Be-
cause radioligand 1b was most promising in the biodis-
tribution studies, this ligand was further evaluated in

PET Imaging of Muscarinic Receptors in the Airways
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 121
Ta ble 3. Tissue/Plasma Ratios of the Muscarinic Antagonists 1a , 1b, and 2 in Male Wistar Rats
ligands
organ
heart
intestine
lung
pancreas
spleen
stomach
sub. gland
trachea
1a control
atropine
1b control
atropine
2 control
atropine
61.1 ( 8.5
33.0 ( 20.2
4.8 ( 0.7
14.2 ( 1.5
6.3 ( 1.7
11.6 ( 1.8
12.1 ( 1.0
4.0 ( 0.7
1.3 ( 0.3
8.7 ( 4.5
1.1 ( 0.1
2.3 ( 0.7
1.8 ( 1.1
3.0 ( 2.2
1.6 ( 0.5
2.1 ( 0.6
118.2 ( 24.4*
47.2 ( 24.4
10.1 ( 2.9*
28.8 ( 5.6*
11.0 ( 2.0*
24.7 ( 3.6*
26.3 ( 7.5*
8.0 ( 2.8*
0.6 ( 0.1*
5.3 ( 4.3
0.6 ( 0.1*
1.9 ( 0.8
0.6 ( 0.2*
1.2 ( 0.4
1.3 ( 0.8
1.6 ( 0.3
140.3 ( 9.9
54.7 ( 19.0
11.8 ( 1.1
30.3 ( 5.3
15.6 ( 4.3
22.6 ( 3.4
27.2 ( 4.9
10.8 ( 3.1
1.0 ( 0.4
33.5 ( 22.5#
1.0 ( 0.3#
3.4 ( 1.7
1.8 ( 0.8#
4.1 ( 3.2
2.7 ( 1.1#
3.0 ( 0.8#
* Significant difference (p < 0.05) between 1b and 1a . # Significant difference (p < 0.05) between 1b and 2.
converted to the TMS scale using δ (CDCl3) ) 7.26 ppm. 13C
NMR spectra (APT) were recorded on a Varian Gemini 200
(at 50.32 MHz). Chemical shifts are denoted in δ units (ppm)
relative to the solvent and converted to the TMS scale using
δ (CDCl3) ) 76.91 ppm. The splitting patterns are designated
as follows: s (singlet), 2s (2× singlet), d (doublet), dd (doublet
doublet), t (triplet), q (quartet), m (multiplet), and br (broad).
Optical rotations were measured at the NaD-line on a Perkin-
Elmer 241 MC polarimeter at room temperature. Elemental
analyses were performed in the Microanalytical Department,
University of Groningen. Mass spectra were recorded on an
AEI-MS-902 mass spectrometer (EI). Chemicals were pur-
chased from Merck, J anssen, or Aldrich. Ether, CH2Cl2,
hexane, and pentane were distilled over P2O5 except when the
solvent was used for extractions. Toluene and tetrahydrofuran
(THF) were distilled over Na (benzophenone). Benzene was
used without any further purification. Thin layer chromatog-
raphy was performed using silica gel 60 F254 (Merck, art. no.
7734 or 9385). 3-Bromocyclohexene was synthesized according
to a procedure described in literature.38
F igu r e 2. Clearance of radioactivity from plasma after
administration of (R)-isomer 1b. The solid line is a biexpo-
nential curve fit.
Ta ble 4. Binding Affinities of 1b, 2, and Other Muscarinic
(2R,5R)-2-ter t-Bu tyl-5-p h en yl-1,3-d ioxola n -4-on e (5). A
mixture of (R)-mandelic acid 3b (13.24 g, 87 mmol), pivalal-
dehyde 4 (15 g, 174 mmol), p-toluenesulfonic acid (0.25 g), and
concentrated H2SO4 (two drops) in pentane (70 mL) was
refluxed with azeotropic removal of the water formed. After
the mixture was cooled, a white solid precipitated. The
precipitate was dissolved in a large amount of ether (600 mL)
and washed with water (2 × 100 mL). The organic layers were
dried (MgSO4) and evaporated. The crude material (cis:trans
) 7.5:1) was recrystallized from hot ether/pentane, yielding 5
as colorless needles (13.59 g, 61.8 mmol, 71%, ds >99%): mp
142 °C (lit.29 mp 140 °C); [R]D ) -88.7° (c ) 1.2; CHCl3) [lit.29
[R]D ) +88.7° (c ) 1.2; CHCl3, (2S,5S)-isomer)]; 1H NMR
(CDCl3, 200 MHz) δ 7.51-7.39 (m, 5 H), 5.34 (d, 1 H, J ) 1.3),
5.25 (d, 1 H, J ) 1.3), 1.10 (s, 9 H); 13C NMR (CDCl3) δ 172.0
(s), 133.51 (s), 129.17 (d), 128.72 (d), 127.05 (d), 109.32 (d),
77.05 (d), 34.46 (s), 23.62 (q).
(2R,5R)-2-ter t-Bu tyl-5-cycloh ex-2-en yl-5-p h en yl-1,3-d i-
oxola n -4-on e (6). A solution of diisopropylamine (4.48 mL,
32 mmol) in THF (200 mL) was cooled to -40 °C, and n-BuLi
(20 mL, 32 mmol, 1.6 M solution in hexane) was added. The
reaction mixture was warmed up to -10 °C during 15 min
and then cooled to -80 °C. Subsequently a solution of 5 (7.0
g, 31.8 mmol) in THF (75 mL) was slowly added. The reaction
mixture was allowed to warm up to -15 °C over a period of 1
h and then cooled to -80 °C again. Then 3-bromocyclohexene
(3.7 mL, 31,9 mmol) was slowly added. The reaction mixture
was allowed to warm up slowly to -20 °C during 3 h and
subsequently poured into a half-saturated aqueous solution
of ammonium chloride (100 mL) and extracted with ether (2
× 100 mL). The organic layers were dried (MgSO4) and
evaporated. The crude product (cis:trans ) 12:1) was crystal-
lized from hexane at -18 °C, yielding 6 (7.35 g, 24.50 mmol,
77%, de > 99%): mp 109.7-111.3 °C; [R]D ) -12.9° (c ) 1,
CHCl3); 1H NMR (CDCl3) δ 7.72-7.31 (m, 5 H), 5.94-5.37 (m,
2 H), 5.45 and 5.40 (2s, two diastereoisomers from additional
stereogenic center, 1 H), 2.90-2.83 (m, 1 H), 1.99-1.41 (m, 6
H), 0.96 and 0.93 (2s, two diastereoisomers, 9 H); 13C NMR
(CDCl3) δ 173.0 (s), 137.6 (2s), 132.30 (d), 130.96 (d), 128.71
(d), 127.93 (2xd), 125.57 (2xd), 110.53 (2xd), 84.25 (2s), 46.02
(2d), 35.06 (2s), 24.90 (t), 24.47 (t), 23.54 (q), 22.81 (t), 21.64
(t), 21.11 (t). (Most of the peaks are doubled due to the
presence of an additional stereogenic center.)
Antagonists
Ki (nM)
M1 rat
cortex
M2 rat
heart
M3 guinea
pig ileum
M1 calf
brain
compound
1b
2
0.01
0.02
0.53
0.21
0.10
4.57
0.09b
0.12b
0.02
0.01
0.41
scopolamine
4-DAMP (MeI)
(R)-MQNBa
66.9b
0.65d
0.13c
0.43
a
b
Reference 21. Determined by NovaScreen, mean of two
determinations. c In NB-OK-1-cells. In rat pancreas.
d
initial rapid, partial washout followed by a slow decay.
Biodistribution studies in untreated and atropine-
treated rats demonstrated significant specific binding
in tissues containing muscarinic receptors. Use of the
active (R)-isomer (1b) instead of its racemate (1a )
resulted in a doubling of the tissue to plasma concentra-
tion ratios in most organs. However use of the slightly
more potent (R,R)-compound 2 instead of the (R)-
compound (1b ) did not improve tissue/plasma ratios,
and it slightly increased nonspecific binding. Ligand
1b demonstrated low but useful lung/plasma ratios (10
( 3) and lung/blood ratios (20 ( 3) at 15 min postin-
jection. Compound 1b was slowly metabolized in male
Wistar rats. Tissue uptake of labeled metabolites was
negligible in heart and 12% in lungs, 15 min postinjec-
tion. Ligand 1b appeared to be an excellent candidate
for PET studies of mACh receptors in heart and lungs.
Exp er im en ta l Section
Gen er a l Rem a r k s. IR spectra were recorded with a
Perkin-Elmer 841 infrared spectrometer. Melting points were
determined with a Mettler FP-2 melting point apparatus,
equipped with a Mettler FP-2 microscope (uncorrected). 1H
NMR spectra were recorded on a Varian Gemini 200 (200
MHz) spectrometer. Chemical shifts are denoted in δ units
(ppm) and are determined relative to the solvent (CDCl3) and

122 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Visser et al.
(2R,5R)-2-ter t-Bu tyl-5-cycloh exyl-5-ph en yl-1,3-dioxolan -
4-on e (7). To a solution of 6 (15.8 g, 52.7 mmol) in THF (175
mL) was added Pd/C (10%, 10 mol %). The suspension was
hydrogenated in a Parr apparatus at 5.0 bar of H2 pressure
until no hydrogen was adsorbed. After filtration, to remove
the catalyst, the solvent was evaporated to afford 7 as a white
crystalline compound (15.6 g, 51.7 mmol, 98%). Although the
compound is pure enough for the next step, it can be recrystal-
lized from hexane at -18 °C: mp 118.5-121 °C; [R]D ) -19.0°
(c ) 1, CHCl3); 1H NMR (CDCl3) δ 7.68-7.29 (m, 5 H), 5.41 (s,
1 H), 1.97-1.07 (m, 11 H), 0.94 (s, 9 H); 13C NMR (CDCl3)
173.90 (s), 137.82 (s), 127.83 (d), 127.51 (d), 125.35 (d), 110.63
(d), 85.14 (s), 48.40 (d), 35.54 (s), 28.02 (t), 26.09 (t), 25.93 (t),
23.52 (q); HRMS calcd 302.188, found 302.188. Anal. Calcd
for C19H26O3: C, 75.46; H, 8.67 Found: C, 75.57; H, 8.62.
2,2-Dim eth yl-5-p h en yl-1,3-d ioxola n -4-on e (14). A mix-
ture of racemic mandelic acid 3a (15.2 g, 0.10 mol), 2,2-
dimethoxypropane (12.5 g, 0.12 mol), and benzene (100 mL)
was refluxed for 2 h with azeotropic removal of methanol. After
the mixture was concentrated at reduced pressure, 14 was
obtained as a white solid (19.02 g, 99.1 mmol, 99%). The
product was pure enough to be used in the next step: mp 40-
42.5 °C (lit.31 mp 42.5-43.5 °C); 1H NMR (CDCl3) δ 7.51-7.35
(m, 5 H), 5.41 (s, 1 H), 1.74 (s, 3 H), 1.68 (s, 3 H); 13C NMR
(CDCl3) δ 171.48 (s), 134.48 (s), 128.94 (d), 128.71 (d), 126.43
(d), 110.94 (s), 75.90 (d), 27.20 (q), 26.13 (q).
2,2-Dim eth yl-5-cycloh ex-2-en yl-5-p h en yl-1,3-d ioxola n -
4-on e (15). The same synthetic procedure was used as
described for 6 except that compound 14 was used instead of
5. The crude product was obtained as an oil (1.35 g, 4.98
mmol, 99%) and was used in the next step without further
purification: 1H NMR (CDCl3) δ 7.65-7.22 (m, 5 H), 5.92-
5.23 (m, 2 H), 2.78 (m, 1 H), 2.20-1.20 (m, 6 H), 1.65 (2s, 3
H), 1.39 (2s, 3 H); 13C NMR (CDCl3) δ 171.00 (s), 138.36 (s),
130.90 (d), 128.17 (d), 128.10 (d), 125.60 (d), 125.35 (d), 110.26
(s), 85.30 (s). 45.44 (d), 27.94 (q), 27.23 (q), 24.90 (t), 24.46
(t), 21.45 (t). (Most of the peaks were doubled due to the
presence of an additional stereogenic center.)
2,2-Dim e t h yl-5-cycloh e xyl-5-p h e n yl-1,3-d ioxola n -4-
on e (16). The same synthetic procedure was used as described
for 7 except that compound 15 was used instead of 6. Bulb to
bulb distillation (0.02 mmHg, 125 °C) yielded the product as
a clear viscous oil (7.86 g, 28.77 mmol, 77.5%): 1H NMR
(CDCl3) δ 7.63-7.27 (m, 5 H), 1.89-0.89 (m, 11 H), 1.69 (s, 3
H), 1.41 (s, 3 H); 13C NMR (CDCl3) δ 173.20 (s), 138.84 (s),
128.02 (d), 127.68 (d), 125.45 (d), 110.08 (s), 86.13 (s), 47.29
(d), 28.02 (t), 27.95 (q), 27.29 (q), 26.27 (t), 26.01 (t); HRMS
calcd 274.157, found 274.157. Anal. Calcd for C17H22O3: C,
74.41; H, 8.09. Found: C, 74.30; H: 8.09.
(R)-2-Cycloh exyl-2-h yd r oxy-2-p h en yla cetic Acid (8).
To a solution of 7 (8.0 g, 26.49 mmol) in a mixture of methanol
(100 mL) and ether (100 mL) was added a solution of KOH
(8.18 g, 145.8 mmol) in water (25 mL). The solution was
stirred overnight, and the solvents were removed under
vacuum. The residue was dissolved in water (20 mL) and
acidified with HCl (6 M). The precipitate was taken up in
ether (100 mL) and separated. The water layer was extracted
with ether (2 × 30 mL). After the solvent was dried (MgSO4)
and evaporated, 8 was obtained as a white crystalline com-
methane was led into the solution with acid by means of a
stream of N2. After the diazald solution became colorless, the
reaction mixture was stirred for half an hour and then the
ether was evaporated in vacuum. Bulb to bulb distillation (110
°C, 0.01 mmHg) gave the product 9 (350 mg, 1.41 mmol, 98%)
as a clear oil which became solid on standing: mp 53.2-55.2
°C; [R]D ) -30° (c ) 2, CHCl3) [lit.23 mp 53.5 °C; [R]D ) -30°
1
(c ) 2, CHCl3)]; H NMR (CDCl3) δ 7.68-7.23 (m, 5 H), 3.79
(br, 3 H), 3.69 (s, 1 H), 2.24 (m, 1 H), 1.84-1.02 (m, 10 H); 13
C
NMR (CDCl3) δ 176.09 (s), 140.71 (s), 128.02 (d), 127.35 (d),
125.96 (d), 81.02 (s), 53.24 (q), 45.71 (d), 27.36 (t), 26.31 (t),
26.15 (t), 25.43 (t); HRMS calcd 248.141, found 248.141. Anal.
Calcd for C15H20O3: C, 72.55; H, 8.12 Found: C, 72.17; H, 8.02.
(R)-N-Met h ylp ip er id in -4-yl 2-Cycloh exyl-2-h yd r oxy-
p h en yla ceta te (12b). N-Methyl-4-hydroxypiperidine 10 (165
mg, 1.43 mmol) was dissolved in heptane (20 mL) and refluxed
for 30 min using a Dean-Stark reflux apparatus to remove
any traces of moisture. After the mixture was cooled, NaH
(10 mg, 60% dispersion in mineral oil) was added. After the
mixture was stirred at room temperature for 45 min, com-
pound 9 (350 mg, 1.43 mmol) was added. The reaction mixture
was refluxed for 24 h, and subsequently the heptane was
drained and discarded to a small volume ((10 mL). After the
mixture was cooled in ice, ether (20 mL) was added. The
organic layer was repeatedly washed with water until the
washings were approximately pH 7. After the solvents were
dried (Na2SO4) and evaporated, the crude product was ob-
tained as an oil. Column chromatography (silica gel, CHCl3:
MeOH:NH4OH (20%) ) 90:9:1) afforded 12b (187 mg, 0.56
mmol, 40%) as a clear oil: 1H NMR (CDCl3) δ 7.67-7.62 (m,
2H), 7.36-7.20 (m, 3H), 4.80 (m, 1H), 3.80 (br, 1H), 2.26 (s,
3H), 2.65-1.0 (m, 19H); 13C NMR (CDCl3) δ 175.01 (s), 140.87
(s), 127.92 (d), 127.26 (d), 126.03 (d), 80.70 (s), 71.66 (d), 52.30
(br, t), 46.18 (d), 45.69 (q), 30.56 (t), 30.21 (t), 27.27 (t), 26.33
(t), 26.16 (t), 25.48 (t).
(R)-N-Met h ylp ip er id in -4-yl 2-Cycloh exyl-2-h yd r oxy-
p h en yla ceta te Meth iod id e (1b). To a solution of 12b (48
mg, 0.14 mmol) in ether (2 mL) was added MeI (0.2 mL, 3.17
mmol), and the mixture was stirred overnight. After evapora-
tion of the excess methyl iodide and solvent, the product 1b
was crystallized from 2-butanone (2 mL) until one single peak
was observed on HPLC (CN radialpak 10 µm, mobile phase:
H2O with 1% Et3N (acidified to pH ) 4 with HOAc):CH3CN )
72:928, v/v) running at 4 mL/min. White crystals (50 mg, 0.11
mmol, 75%) were obtained: mp 169-172 °C (lit. racemate: mp
163-165 °C39 and 165-170 °C40); 1H NMR (CDCl3) δ 7.63-
7.24 (m, 5 H), 5.1 (m, 1 H), 3.80 (br, 1 H), 3.43 (s, 3 H), 3.51 (s,
3 H), 3.09-2.95 (dt, 1 H), 2.50-1.0 (m, 19 H); 13C NMR (CDCl3)
δ 174.21 (s), 140.41 (s), 128.19 (d), 127.65 (d), 125.94 (d), 81.00
(s), 65.52 (d), 58.73 (t), 58.54 (t), 54.51 (q), 49.84 (q), 44.80 (d),
27.31 (t), 26.20 (t), 25.98 (t), 25.37 (t), 24.96 (t), 24.39 (t); HRMS
calcd for C21H32NO3I 473.143, the parent peak was not
observed. Immediate fragmentation of methyl iodide occurs.
This results in the masses 141.96 and 331.11. MS (rel int)
331 (20), 189 (99), 142 (100), 127 (27), 98 (86), 96 (38), 55 (38),
28 (75).
(R,R)-Qu in u clid in -3-yl 2-Cycloh exyl-2-h yd r oxyp h en -
yla ceta te (13). The same synthetic procedure was used as
described for 12b except that (R)-quinuclidinol 11 was used
instead of N-methyl-4-piperidinol 10. Column chromatogra-
phy (silica gel, CHCl3:MeOH:NH4OH (20%) ) 90:9:1). The
product was obtained as a clear, viscous oil: yield 18%; 1H
NMR (CDCl3) δ 7.68-7.63 (m, 2 H), 7.40-7.26 (m, 3 H), 4.87
(m, 1 H), 3.6 (br, 1 H), 3.2 (m, 1 H), 2.90-1.10 (m, 21 H); 13C
NMR (CDCl3) δ 175.5 (s), 140.68 (s), 128.02 (d), 127.38 (d),
125.97 (d), 80.80 (s), 73.48 (d), 54.89 (t), 47.25 (t), 46.28 (t),
45.51 (d), 27.39 (t), 26.33 (t), 26.15 (t), 25.52 (t), 25.30 (d), 24.37
(t), 19.67 (t).
pound (5.28 g, 22.52 mmol, 85%): mp 139.5-141.0 °C; [R]D
)
-26.5° (c ) 1, EtOH) [lit.33 mp 139.5-140.5 °C; [R]D ) -26.3°
(c ) 1, EtOH)]. The enantiomeric excess (>99%) was deter-
mined by means of a chiral â-cyclodextrin column (cyclobond
I, 250 × 4.6, 5 µm, Astec) with the eluant K2HPO4‚3H2O (0.1
M, pH ) 4); CH3CN ) 40:60, flow rate 1 mL/min; UV detection
205 nm, injected was 20 µL from a solution of 1 mg of 8 in 2
mL of eluant. The product was pure enough to be used in the
next step: 1H NMR (CDCl3) δ 7.73-7.26 (m, 5H), 2.33-2.22
(m, 1 H), 2.02-1.00 (m, 10 H); 13C NMR (CDCl3) δ 180.68 (s),
139.86 (s), 128.18 (d), 127.70 (d), 125.95 (d), 80.99 (s), 45.67
(d), 27.33 (t), 26.23 (t), 26.09 (t), 25.39 (t).
(R,R)-Qu in u clid in -3-yl 2-Cycloh exyl-2-h yd r oxyp h en -
yla ceta te Meth iod id e (2). The same synthetic procedure
was used as described for 1b except that 13 was used instead
of 12b. The product 2 was crystallized from acetone until one
peak was observed on HPLC (conditions as described for 1b):
yield 85%; mp 243.4-244.1 °C; 1H NMR (CDCl3) δ 7.62-7.59
(m, 2 H), 7.40-7.22 (m, 3 H), 5.16 (m, 1 H), 4.35 (m, 1 H),
3.95 (s, 1 H), 3.8 (m, 2 H), 3.45 (m, 2 H), 3.25 (s, 3 H), 1.10 (m,
17H); 13C NMR (CDCl3) δ 174.00 (s), 140.42 (s), 128.43 (d),
(R)-Meth yl 2-Cycloh exyl-2-h ydr oxy-2-ph en yleth an oate
(9). To a solution of 8 (335 mg, 1.43 mmol) in ether (25 mL)
was added diazomethane (3.3 mmol). The diazomethane was
generated by adding a solution of NaOH (10%) to a yellow
solution of diazald (1.0 g) in ethanol (30 mL). The NaOH
solution was added until colorless. The generated diazo-

PET Imaging of Muscarinic Receptors in the Airways
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 123
127.76 (d), 125.73 (d), 81.59 (s), 68.47 (d), 62.13 (t), 57.11 (t),
56.02 (t), 51.88 (q), 44.96 (t), 27.76 (t), 26.19 (t), 25.54 (t), 23.40
(d), 21.04 (t), 18.06 (t); HRMS calculated for C22H32O3IN
485.142, the parent peak was not observed. Immediate
fragmentation of methyl iodide occurs. This results in the
fragments 141.95 and 343.64. MS (rel intens) 343 (1), 189 (26),
142 (80), 126 (24), 105 (76), 77 (22), 28 (100); HRMS (CI, NH3)
m/ e 486 (M + 1).
artery of anaesthetized rats was cannulated, and arterial blood
samples (0.2-0.3 mL) were drawn at 10, 30, 42, 55, 65, 80,
93, 108, 123, 300, 600, 1200, and 2400 s after injection of 5.55
MBq of labeled 1b, with a specific activity of 20.35 TBq/mmol.
Plasma and red blood cells were separated by centrifugation
(2 min, 3000g). Plasma (100 µL) was then counted in a
calibrated γ-counter. The clearance curve was analyzed by
means of a commercial nonlinear regression program EnzFit-
ter (Elsevier Biosoft, Cambridge, U.K.). For analysis of
metabolites, pentobarbital-anaesthetized rats (280 g) were
injected with radioligand 1b (5.55 MBq in 0.4 mL of saline,
specific activity 3.7 TBq/mmol) via a tail vein. After 15 min
the rats were killed by extirpation of the heart. Lung, heart,
and liver were quickly removed and homogenized in a mixture
of saline (1 volume) and acetonitrile (2 volumes), using a
Heidolph diax 600 apparatus (Salm & Kipp b.v., Breukelen,
The Netherlands). Subsequently the extracts were centrifuged
(5 min, 3000g), and the supernatant was injected onto an
HPLC column (C-18 radialpak 5 µm, mobile phase NaH2PO4
(50 mM):CH3CN:Et3N ) 50:50:2 (v/v); pH ) 6.0; flow rate 1.5
mL/min). Fractions (0.75 mL) of the eluate were collected
during 15 min and counted in a calibrated γ-counter. The
recovery from the column exceeded 99%.
Ra d iosyn th esis of [11C]-La beled Mu sca r in ic An ta go-
n ists 1a , 1b, a n d 2. [11C]CH3I was prepared from [11C]CO2,
which was generated by the 14N(p,R)11C nuclear reaction with
17 MeV protons. The [11C]CO2 was trapped in the reaction
vessel which was previously filled with a solution of LiAlH4
in THF (0.3 mL, a saturated solution of LiAlH4 in THF (0.15
mL) was diluted with THF to a final volume of 1 mL) using
an Anatech robotic system. After evaporation of the solvent
at 130 °C, HI (57%, 0.8 mL) was added with the robot and the
[11C]CH3I was formed. In this way 30 GBq [11C]-labeled
methyl iodide with a specific activity of 150 TBq/mmol was
obtained 11 min EOB. The radiochemical yield was 70-90%
(corrected for decay). The [11C]CH3I was trapped in a solution
of the desmethyl precursor (1.4 mg) in CH3CN (0.4 mL) in a
stervial which was cooled to 0 °C with ice. After the [11C]CH3I
was trapped, the stervial was capped and placed into an oil
bath at 110 °C for 12 min. Then the stervial was cooled in an
ice bath for 1 min and the solvents evaporated by rotary
evaporation. Subsequently the radioligand was taken up in
HPLC eluant (1 mL) and injected onto an HPLC column (CN-
10 µm radialpak, mobile phase: H2O with 1% Et3N (acidified
to pH ) 4 with HOAc):CH3CN ) 72:928, v/v), running at 4
mL/min. The radiolabeled product was completely separated
from the precursor: retention time precursor, 5 min; retention
time product, 9 min. The radiochemical yield was 40-60%
(corrected for decay), and the specific activity was 11.1-44.4
TBq/mmol, 40 min EOB. After evaporation of the solvent, the
radioligand was dissolved in saline (0.9% NaCl) and injected
in the tail vein of male Wistar rats.
Dyn a m ic P E T St u d ies in Ma le Wist a r R a t s. Animal
handling was in accordance with the Law on Animal Experi-
ments of The Netherlands. PET studies were carried out with
a Siemens ECAT 951/31 positron camera. Data acquisition
was performed using a dynamic protocol and in stationary
mode. In this mode the in-plane spatial resolution amounts
to 6 mm fwhm. During the reconstruction a zoom factor of
1.5 was applied and the matrix size was 128 × 128. The rats
were anaesthetized with pentobarbital. The rats were posi-
tioned in the PET camera parallel to the transaxial plane of
the tomograph, so sagittal sections were obtained. In the
control rats, a saline solution of ligand 1b (2.78 MBq, specific
activity 25.9 TBq/mmol) was administered by injection in the
tail vein. In the blocked group the rats were first pretreated
with atropine followed by injection of a saline solution of the
radioligand (2.78 MBq, specific activity 25.9 TBq/mmol). The
following frames were defined: 10 × 30, 3 × 300, and 3 × 600
s. Data analysis was performed using the Siemens ECAT
software (V6.5) on a Sun workstation.
Tissu e Distr ibu tion Stu d ies in Ma le Wista r Ra ts. The
radioligands were intravenously injected in the tail of pento-
barbital-anaesthetized (60 mg/kg body weight) male Wistar
rats (240 g) as described previously by Van Waarde et al.41
One minute before injection of the radioligand (1.85-2.78
MBq) which was dissolved in saline (0.4 mL), the rats were
pretreated with saline (0,5 mL, controls) or saline containing
atropine (0.5 mg/0.5 mL, blocking experiments). After 15 min
the animals were killed by extirpation of the heart, and the
following tissues were removed: bladder, cerebellum, cortex,
fat, heart, intestine, kidney, liver, lung, muscle, pancreas,
spleen, stomach, submandibular gland, testes, and trachea.
Plasma was obtained from the blood by centrifugation (5 min,
2000g). Next the tissue samples were weighed, and the
radioactivity was determined with a calibrated γ-counter
(LKB-Wallac Compugamma 1282 CS, Turku, Finland). Dif-
ferences between the various treatments were analyzed using
Student’s t-test or one-way analysis of variance (ANOVA); in
both cases a probability of p < 0.05 was considered statistically
significant.
Ra d ioliga n d Bin d in g Assa ys. The binding of the radio-
ligands 1b and 2 to muscarinic M1, M2, and M3 receptor
subtypes was determined as follows: A male Wistar rat was
anaesthetized with pentobarbital and sacrificed by extirpation
of the heart. The heart and cortex were removed and frozen
in liquid nitrogen and stored in the refrigerator at -20 °C.
The following buffers were used: PB7, 50 mM phosphate
buffer, pH 7.4, was prepared by dissolving 7.12 g of Na2-
HPO4‚2H2O and 1.3 g of NaH2PO4‚2H2O in 1 L of distilled
water; PBS, the PB7 buffer was supplemented with sucrose
to give a final concentration of the latter of 0.32 M. Subse-
quently the rat cortex was homogenized in 10 volumes (w/v)
of ice-cold PBS using a glass-Teflon Potter-Elvejehem ho-
mogenizer (R. W. 18, J anke & Kunkel, Staufen i. Breisgau,
FRG) at 1200 rpm, 10 up and down strokes. The homogenate
was centrifuged for 10 min at 1000g, and the resulting
supernatant was centrifuged for 60 min at 200000g. The pellet
was washed with PB7 and centrifuged for 30 min at 100000g
twice, and the final pellet was stored at -20 °C. For the
determination of the affinity of the muscarinic antagonists 1b
and 2 to rat cortex, 2.72 mg of receptor material in 400 µL of
buffer PB7 was used per sample. The preparation of the
muscarinic receptor from calf brain was similar, though this
preparation was lyophilized after reconstitution of the pellet
in 5 volumes of PB7. In the receptor assay 0.5 mg of this
preparation in 400 µL of PB7 was used per sample. For rat
heart the procedure was as follows: the tissue was suspended
in 25 mL of PB7 buffer and minced with an Ultraturrax
apparatus. Subsequently the material was homogenized using
a Potter-Elvejehem homogenizer at 1200 rpm, 10 up and down
strokes. For the assay, 400 µL of receptor material was used
per sample. The receptor suspensions were incubated with
50 µL (950 Bq) of labeled [3H]NMS with a specific activity of
2960 Bq/pmol and 50 µL of the enantiomers 1b, 2, or scopol-
amine in concentrations ranging from 1 × 10-12 to 1 × 10-7
M. The mixtures were incubated at 37 °C for 45 min. All the
incubations were terminated by the addition of 4 mL of ice-
cold PB7 buffer and filtered through Whatman GF/B filters
at a pressure of 500 mbar. The tubes were rinsed twice with
4 mL of ice-cold buffer, which was also filtered, and the filters
were dispersed in 3.5 mL of Rialuma by shaking for 2 h. The
vials were counted for 5 min in a liquid scintillation counter
(Packard TriCarb 4000, Downers Grove, IL). The binding
parameters were calculated using the RADLIG program
(version 4.0), purchased from Elsevier/Biosoft. The dissocia-
tion constant of [3H]NMS used was 0.1 nM for calf brain42 and
rat cortex21,42 and 0.4 nM for rat heart.21 Receptor concentra-
tion, N1 (aspecific binding of labeled ligand), and dissociation
constants were estimated by the computer program.
Ack n ow led gm en t. The authors gratefully thank
the Netherlands Asthma Foundation for their financial
support (Grant AF 92.20).
Clea r a n ce fr om P la sm a a n d Meta bolism . The carotid

124 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Visser et al.
(23) Inch, T.; Ley, R. V.; Rich, P. Asymmetric Synthesis. Part III.
Stereospecific Synthesis of (R)-2-Hydroxy-2-phenylpropionic Acid,
and (R)- and (S)-2-cyclohexyl-2-hydroxy-2-phenylacetic Acid.
Configurational relationship between (R) (-)-2-Hydroxy-2-
phenylpropionic Acid and (S) (+)-2-Phenylpropionic Acid. J .
Chem. Soc. C 1968, 1693-1699.
(24) Inch, T. D.; Green, D. M.; Thompson, P. B. J . The central and
peripheral activities of anti-acetylcholine drugs. Some concepts
of practical relevance. J . Pharm. Pharmacol. 1973, 25, 359-
370.
(25) Schjelderup, L.; Harbitz, O.; Groth, P.; Aasen, A. J . The absolute
configuration of oxyphencyclimine, a parasympatholytic drug.
Synthesis of both enantiomers. Acta Chem. Scand. 1986, B40,
601-603.
(26) Barlow, R. B.; Franks, F. M.; Pearson, J . D. M. Studies on the
stereospecificity of closely related compounds which block post-
ganglionic acetylcholine receptors in the guinea pig ileum. J .
Med. Chem. 1973, 16, 439-446.
Refer en ces
(1) Goyal, R. Muscarinic receptor subtypes. Physiologic and clinical
implications. N. Engl. J . Med. 1989, 321, 1022-1028.
(2) Buckley, N. J .; Bonner, T. I.; Buckley, C. M.; Brann, M. R.
Antagonist binding properties of five cloned muscarinic receptors
expressed in CHO-K1 cells. Mol. Pharmacol. 1989, 35, 469-
476.
(3) Caulfield, M. P. Muscarinic receptors, characterization, coupling
and function. Pharmacol. Ther. 1993, 58, 319-380.
(4) Lazareno, S.; Buckley, N. J .; Roberts, F. F. Characterization of
muscarinic M4 binding sites in rabbit lung, chicken heart and
NG108-15 cells. Mol. Pharmacol. 1990, 38, 805-815.
(5) Barnes, P. J . Muscarinic receptor subtypes in airways. Life Sci.
1993, 52, 521-527.
(6) Barnes, P. J .; Dollery, C. T.; MacDermot, J . Increased pulmonary
R-adrenergic and reduced â-andrenergic receptors in experimen-
tal asthma. Nature 1980, 285, 569-571.
(7) Raaijmakers, J . A. M.; Terpstra, G. K.; Van Rozen, A. J .; Witter,
A.; Kreukniet, J . Muscarinic cholinergic receptors in pheripheral
lung tissue of normal subjects and of patients with chronic
obstructive lung disease. Clin. Sci. 1983, 66, 585-590.
(8) Delforge, J .; Le Guludec, D.; Syrota, A.; Bendriem, B.; Crouzel,
C.; Slama, M.; Merlet, P. Quantification of myocardial musca-
rinic receptors with PET in humans. J . Nucl. Med. 1993, 34,
981-991.
(9) Mulholland, G. K.; Chin, F. I. N.; Hutchins, G. D.; Fain, R.; Zipes,
D.; New peripheral muscarinic receptor ligands N-[F-18] fluo-
roethyl-QNB and -TRB for parasympathetic imaging. J . Nucl.
Med. 1995, 36, 137P-138P (abstract).
(10) Frey, K. A.; Koeppe, R. A.; Mulholland, G. K.; J ewett, D. M.;
Hichwa, R. D.; Ehrenkaufer, R. L. E.; Carey, J . E.; Wieland, D.
M.; Kuhl, D. E. In vivo muscarinic cholinergic receptor imaging
in human brain with [C-11] scopolamine and positron emission
tomography. J . Cereb. Blood Flow Metab. 1992, 12, 147-154.
(11) Frey, K. A.; Koeppe, R. A.; Mulholland, G. K.; Kuhl, D. E.
Quantification of regional cerebral muscarinic receptors in brain
with the use of C11 tropanyl benzilate and positron emission
tomography. J . Nucl. Med. 1990, 31, 779.
(27) Kiesewetter, D. O. Asymmetric synthesis of benzylic acid
analogues using 8-phenylmenthol as a chiral auxiliary. Tetra-
hedron: Asymmetry 1993, 4 (10), 2183-2198.
(28) Tambute´, A.; Collet, A. De´doublement et de´termination de la
configuration absolute des e´nantiome`res de l’acide (thie´nyl-3)-
2-cyclohexyl-2-hydroxy-2-ace´tique. Application a` la synthe`se
d’esters anticholinergiques du quinuclidinol-3. (Resolution and
determination of the absolute configuration of the enantiomers
of 2-(3-thienyl)-2-cyclohexyl-2-hydroxy acetic acid. Application
to the synthesis of anticholinergic esters of 3-quinuclidinol.) Bull.
Soc. Chim. Fr. 1984, 11, 77-82.
(29) Seebach, D.; Naef, R.; Calderari, G. R-Alkylation of R-hetero-
substituted carboxylic acids without racemization; EPC-synthe-
ses of tertiary alcohols and thiols. Tetrahedron 1984, 40, 1313-
1324.
(30) Frater, G.; Mu¨ller, U.; Gu¨nther, W. Synthesis of enantiomerically
enriched atrolactic acid and other R-hydroxy acids Tetrahedron
Lett. 1981, 22, 4221-4224.
(31) Moorlag, H; Kellogg, R. M.; Kloosterman, M.; Kaptein, B.;
Kamphuis, J .; Schoemaker, H. E. Pig liver esterase catalyzed
hydrolysis of racemic R-substituted R-hydroxy esters. J . Org.
Chem. 1990, 55, 5878.
(32) Hof, R. P. Enantioselective synthesis and (bio)catalysis. Routes
to enantiomerically pure alcohols and thiols. Ph.D. Thesis,
University of Groningen, The Netherlands, 1995.
(33) Feitsma, K. Enantiomers of oxyphenonium bromide, analytical
and pharmacological aspects. Ph.D. Thesis, University of Gronin-
gen, The Netherlands, 1987.
(34) Ringdahl, B.; Resul, B.; Dahlbom, R. Facile preparation of the
enantiomers of 3-acetoxyquinuclidine and 3-quinuclidinol. Acta
Pharm. Suec. 1979, 16, 281-283.
(35) Wall, S. J .; Yasuda, R. P.; Li, M.; Wolfe, B. B. Development of
an antiserum against m3 muscarinic receptors: distribution of
m3 receptors in rat tissues and clonal cell lines. Mol. Pharmacol.
1991, 40, 783-789.
(36) Poaty, V.; Tavakoli, R.; Lockhart, A.; Frossard, N. Muscarinic
receptors after syngeneic unilateral lung transplantation. Life
Sci. 1993, 52, 613-620.
(37) Whitsett, J . A.; Hollinger, B. Muscarinic cholinergic receptors
in developing rat lung. Pediatr. Res. 1984, 18, 1136-1140.
(38) Vogel, A. Textbook of practical organic chemistry; 4th ed.;
Longman: London, p 401.
(39) Coan, S. B.; J affe, B.; Papa, D. Parasympathetic blocking agents.
III. Phenylglycolic acid esters of N-alkyl-4-piperidinol. J . Am.
Chem. Soc. 1956, 78, 3701-3703.
(40) Abramson, F. B.; Barlow, R. B.; Franks, F. M.; Pearson, J , D.
M. Relationships between chemical structure and affinity for
postganglionic acetylcholine receptors of the guinea pig ileum.
Br. J . Pharmacol. 1974, 51, 81-93.
(41) Van Waarde, A.; Meeder, J . G.; Blanksma, P. K.; Bouwer, J .;
Visser, G. M.; Elsinga, P. H.; Paans, A. M. J .; Vaalburg, W.; Lie,
K. I. Suitability of CGP-12177 and CGP-26505 for quantitative
imaging of â-adrenoceptors. Nucl. Med. Biol. 1992, 19, 711-
718.
(42) Smisterova, J . Purification and solubilization of muscarinic
receptors for the purpose of quantitative receptor assays of
anticholinergics. Ph.D. Thesis, University of Groningen, The
Netherlands, 1995.
(12) J oad, P. J .; Casale, T. B. [3H]-Quinuclidinyl benzilate binding
to the human lung muscarinic receptor. Biochem. Pharmacol.
1988, 37, 973-976.
(13) Mak, J . W. C.; Barnes, P. J . Muscarinic subtypes in human and
guinea pig lung. Eur. J . Pharmacol. 1989, 164, 223-230.
(14) Casale, T. B.; Ecklund, P. Characterization of muscarinic
receptor subtypes on human peripheral lung. J . Appl. Physiol.
1988, 65 (2), 594-600.
(15) Eckelman, W. C.; Reba, R. C.; Gibson, R. E.; Rzeszotarski, W.
J .; Vieras, F.; Mazaitis, J . K.; Francis, B. Receptor-binding radio-
tracers: A class of potential radiopharmaceuticals. J . Nucl. Med.
1979, 20, 350-357.
(16) Eglen, R. M.; Reddy, H.; Watson, N.; Challis, R. A. J . Muscarinic
acetylcholine receptor subtypes in smooth muscle. Trends
Pharmacol. Sci. 1994, 15, 114-119.
(17) Eberlein, W. G.; Engel, W.; Mihm, G.; Rudolf, K.; Wetzel, B.;
Entzeroth, M.; Mayer, N.; Doods, H. N. Structure-activity
relationships and pharmacological profile of selective tricyclic
antimuscarinics. Trends Pharmacol. Sci. 1989, Suppl. 50-54.
(18) Waelbroeck, M.; Camus, J .; Tastenoy, M.; Mutschler, E.; Stro-
hmann, C.; Tacke, R.; Lambrecht, G.; Christophe, J . Binding
affinities of hexahydrodifenidol and hexahydrosiladifenidol ana-
logues at four muscarinic receptor subtypes: constitutional and
stereochemical aspects. Eur. J . Pharmacol. 1991, 206, 95-103.
(19) Van Waarde, A.; Visser, G. M.; Visser, T. J .; Brouwer, J .; Paans,
M. J .; Vaalburg, W. Rodent biodistribution and metabolism of
tritiated 4-DAMP, a M3 subtype-selective cholinoceptor ligand.
Nucl. Med. Biol. 1994, 21, 41-47.
(20) Waelbroeck, M.; Camus, J .; Tastenoy, M.; Mutschler, E.; Chris-
tophe, J . Binding properties of nine 4-diphenyl-acetoxy-N-
methyl-piperidine (4-DAMP) analogues to M1, M2, M3 and
putative M4 muscarinic receptor subtypes. Br. J . Pharmacol.
1992, 105, 97-102.
(21) Waelbroeck, M.; Tastenoy, M.; Camus, J .; Christophe, J . Binding
kinetics of quinuclidinyl benzilate and methylquinuclidinyl
benzilate enantiomers at neuronal (M1), cardiac (M2), and
pancreatic (M3) muscarinic receptors. Mol. Pharmacol. 1991,
40, 413-420.
(22) Delforge, J .; J anier, M.; Syrota, A.; Crouzel, C.; Vallois, J . M.;
Cayla, J .; Lancon, J . P.; Mazoyer, B. M. Noninvasive quantifica-
tion of muscarinic receptors in vivo with positron emission
tomography in the dog heart. Circulation 1990, 82, 1494-1504.
J M960374W