Synthesis and biological characterization of 1-methyl-1,2,5,6-tetrahydropyridyl-1,2,5-thiadiazole derivatives as muscarinic agonists for the treatment of neurological disorders

Article abstract of DOI:10.1021/jm0301235

Muscarinic agonists might be useful in the treatment of neurological disorders, including Alzheimer's disease, schizophrenia, chronic pain, and drug abuse. Previous studies identified a series of bis-1,2,5-thiadiazole derivatives of 1,2,5,6-tetrahydropyridine with high activity and selectivity for muscarinic receptors. To develop compounds with improved central nervous system penetration, several new derivatives were synthesized and characterized for muscarinic receptor binding and activity. One ligand (11) exhibited agonist activity at M₁, M₂, and M₄ receptors, a selectivity profile suggesting potential utility in the treatment of schizophrenia.

Full text of DOI:10.1021/jm0301235

J . Med. Chem. 2003, 46, 4273-4286
4273
Syn th esis a n d Biologica l Ch a r a cter iza tion of
1-Meth yl-1,2,5,6-tetr a h yd r op yr id yl-1,2,5-th ia d ia zole Der iva tives a s Mu sca r in ic
Agon ists for th e Tr ea tm en t of Neu r ologica l Disor d er s
Yang Cao,† Minjia Zhang,† Cindy Wu,† Selina Lee,† Mary Elizabeth Wroblewski,† Trisha Whipple,†
Peter I. Nagy,† Krisztina Taka´cs-Nova´k,‡ Attila Bala´zs,§ Szila´rd T_oro¨s,§ and William S. Messer, J r.*,†
Departments of Medicinal & Biological Chemistry and Pharmacology, College of Pharmacy, The University of Toledo,
2801 West Bancroft Street, Toledo, Ohio 43606, Institute of Pharmaceutical Chemistry, Semmelweis University,
Ho¨gyes E. u. 9, H-1092 Budapest, Hungary, and Department of Organic Chemistry, University of Veszprem,
Wartha V. u. 1., H-8200 Veszprem, Hungary
Received March 14, 2003
Muscarinic agonists might be useful in the treatment of neurological disorders, including
Alzheimer’s disease, schizophrenia, chronic pain, and drug abuse. Previous studies identified
a series of bis-1,2,5-thiadiazole derivatives of 1,2,5,6-tetrahydropyridine with high activity and
selectivity for muscarinic receptors. To develop compounds with improved central nervous
system penetration, several new derivatives were synthesized and characterized for muscarinic
receptor binding and activity. One ligand (11) exhibited agonist activity at M1, M2, and M4
receptors, a selectivity profile suggesting potential utility in the treatment of schizophrenia.
In tr od u ction
Muscarinic receptors belong to the super family of G
protein-coupled receptors, and mediate many of the
actions of acetylcholine throughout the brain and pe-
ripheral tissues. Five muscarinic receptor subtypes have
been identified and characterized using a variety of
pharmacological and molecular biological techniques.1
Several lines of evidence have suggested that selective
muscarinic agonists could be useful in the treatment of
schizophrenia. Anticholinergic drugs, including the gly-
colate esters, produce psychotomimetic effects in hu-
mans.2 High doses of muscarinic antagonists used in
the treatment of Parkinson’s disease (e.g., benztropine,
trihexyphenidyl) produce hallucinations and confusion.
In contrast, Alzheimer’s disease patients treated with
cholinesterase inhibitors, which elevate levels of ace-
tylcholine, exhibit improvements in neuropsychiatric
symptoms such as agitation, hallucinations, and psy-
chosis.3-5 In addition, the selective M1 muscarinic
agonist xanomeline (1) significantly improved psychi-
atric symptoms such as hallucinations in phase II
clinical trials in Alzheimer’s patients.6 Unfortunately,
xanomeline produced unwanted side effects associated
with activation of M3 receptors, including salivation,
diarrhea, and profuse sweating, which limited patient
compliance.7
produce very few of the adverse side effects (e.g.,
catalepsy) associated with classical antipsychotics such
as haloperidol, suggesting that selective muscarinic
agonists might provide a useful alternative therapeutic
approach to treating the symptoms of schizophrenia.
Moreover, muscarinic agonists might be particularly
useful in improving cognitive function (including memory
function, language use, and constructional praxis) in
schizophrenic patients.9 A key factor in developing
muscarinic agonists as antipsychotics will be limiting
the side effects associated with M3 receptor activation.
Behavioral studies of muscarinic receptor knockout
mice highlight the potential utility of selective M1 and
M4 agonists in the treatment of psychosis.10 For ex-
ample, M4 receptors modulate locomotor activity pro-
duced by the stimulation of D1 dopamine receptors. M4
knockout mice also show enhanced sensitivity to the
effects of phencyclidine on the prepulse inhibition model
of psychosis. M1 agonists might be particularly useful
Follow-up preclinical studies with structurally related
compounds identified strong antipsychotic activity in
(5R,6R)-6-(3-butylthio-1,2,5-thiadiazol-4-yl)-1-azabicyclo-
[3.2.1]octane (2), which displays partial agonist activity
at M2 and M4 receptors.8 Compounds 1 and 2 may exert
an antipsychotic action by regulating the release of
dopamine in the frontal cortex. Compounds 1 and 2
* To whom correspondence should be addressed. Tel: 419-530-1958.
Fax: 419-530-1909. E-mail: wmesser@utnet.utoledo.edu.
† The University of Toledo.
‡ Semmelweis University.
§ University of Veszprem.
10.1021/jm0301235 CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/30/2003

4274 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
Sch em e 1a
a
(a) KCN/H2O/HOAc, NH4Cl/NH3(aq), S2Cl2/DMF. (b) HO(CH2)12OH (6), NaH, THF, reflux. (c) Compound 12, NaH, THF, reflux. (d)
CH3I, acetone, room temperature. (e) NaBH4, CH3OH/CHCl3, 0-5 °C. (f) HCl(g), CH3OH/CHCl3, 0-5 °C. (g) CH3I, acetone/CHCl3, room
temperature. (h) Ac2O, ZnCl2, 85 °C. (i) 1 M HCl/ether, acetone/ether, 0-5 °C.
Ch a r t 1
assessed due to the lack of compounds exhibiting an
appropriate combination of agonist activity and selectiv-
ity for M1/M4 receptors. Recently, a novel series of
bivalent ligands was developed with strong agonist
activity at M1 and M4 receptors and very low activity
at M3 and M5 receptors.11 Among the compounds, 3
displayed high potency and agonist activity at M1
receptors. However, the inherent weaknesses associated
with the structural features of bivalent ligands (rela-
tively large size, hydrophobicity of alkyl linking groups,
and the presence of two basic amines in one molecule)
likely limit membrane permeability and bioavailability.
The subtype selectivity of bivalent ligands may derive
from interactions with nonconserved residues within the
extracellular domains of muscarinic receptors. The
hypothesized mechanism of agonist activation calls for
the tetrahydropyridyl group penetrating into the bind-
ing pocket formed by Asp105, Thr192, and Asn382
residues, while the polar group at the other end of the
bivalent ligand interacts with the extracellular loops.
To explore whether the bivalent ligand is flexible enough
to take the required conformation from the one adopted
in the biological fluid within the synaptic cleft, molec-
ular dynamics (MD) calculations were performed in the
present studies.
Ch a r t 2
in treating memory and cognitive deficits associated
with schizophrenia.9 Taken together, the data suggest
a role for M1/M4 agonists in the treatment of schizo-
phrenia.
The clinical utility of muscarinic agonists for the
treatment of schizophrenia has not been adequately
To develop selective muscarinic agonists with thera-
peutic potential, two new series of tetrahydropyridyl

Muscarinic Agonists for Neurological Disorder Treatment
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20 4275
Sch em e 2a
a
(a) NaH, THF, reflux. (b) CH3I, acetone/CHCl3, room temperature. (c) NaBH4, CH3OH/CHCl3, 0-5 °C. (d) HCl(g), CH3OH/CHCl3,
0-5 °C. (e) Ac2O, ZnCl2, 85 °C. (f) 1 M HCl/ether, 0-5 °C. (g) CH3I, acetone, room temperature.
thiadiazole derivatives were designed. Three compounds
were synthesized in the series of tetrahydropyridyl
thiadiazole derivatives with an alkyl linker (see Chart
1), while five compounds were synthesized with a poly-
(ethylene glycol) linker (see Chart 2). MD simulations
also examined the structural consequences of replacing
a cationic tetrahydropyridyl-thiadiazole moiety with
different neutral groups. One compound (11) displayed
robust agonist activity at M1, M2, and M4 receptors
suggesting potential utility in the treatment of psychi-
atric disorders.
95.3% yield. The ethers were treated with excess CH3I
to yield the quaternary ammonium iodides (compounds
15 and 17a -c) in 91.5-99.7% yield. The quaternary
salts then were treated with 4 equivalents of NaBH4 in
CH3OH and CHCl3 to give the free bases (compounds
19 and 21a -c). Then, dry HCl gas was used to convert
the purified free bases into their corresponding hydro-
chloride salts (23 and 24a -c) in 46.2-76.9% yield.
For 10 and 11, prior to quaternization, the resulting
12-[3-(pyrid-3-yl)-1,2,5-thiadiazol-4-yloxy]-1-dodecanol
(8) and tetra(ethylene glycol) mono-[3-(pyrid-3-yl)-1,2,5-
thiadiazol-4-yl] ether (9b) were reacted further with 12
to give compounds 13 and 14 in 26.6 and 41.9% yield,
respectively. The corresponding quaternary ammonium
iodides 16 and 18 were obtained in 81.8 and 98.4% yield,
which were converted into their hydrochloride salts 10
and 11 in 65.3 and 16.1% yield, respectively, following
NaBH4 reduction and treatment with hydrogen chloride
gas.
Resu lts a n d Discu ssion
Ch em istr y. As shown in Schemes 1 and 2, the
starting material 3-(3-chloro-1,2,5-thiadiazol-4-yl)pyri-
dine (5) was synthesized from 3-pyridinecarboxaldehyde
(4) following published procedures.12 Tetra(ethylene
glycol) monoethyl ether (7c) was prepared from the
reaction of tetra(ethylene glycol) (7b) with NaH and
BrCH2CH3 in tetrahydrofuran (THF) in a yield of 25.5%.
4-Chloro-3-methoxy-1,2,5-thiadiazole (12) was synthe-
sized in 10.1% yield by reacting 3,4-dichloro-1,2,5-
thiadiazole (29) with NaOCH3. Compound 5 was reacted
with 1,12-dodecanediol (6), tri(ethylene glycol) monoet-
hyl ether (7a ), 7b, or 7c in the presence of NaH in
refluxing THF to give mono-[3-(pyrid-3-yl)-1,2,5-thia-
diazol-4-yl] ethers (compounds 8 and 9a -c) in 41.7-
For 25 and 26, the free bases (19 and 21b) were
transformed to the corresponding acetyl esters (com-
pounds 27 and 28) by reacting with excess acetic
anhydride in the presence of catalytic anhydrous ZnCl2
before conversion into hydrochloride salts. Ethereal HCl
then was added into the methylene chloride solution of
compounds 27 or 28. After crystallization from acetone/

4276 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
Ta ble 1. Pharmacological Properties of Novel Compounds at
Ta ble 4. Pharmacological Properties of Novel Compounds at
M1 Receptors Expressed in A9 L Cellsa
M4 Receptors Expressed in A9 L and CHO Cellsa
M1 receptors
M4 receptors
pKi
pEC50
Smax
pKi
pIC50
Imax
carbachol
5.5 ( 0.13
6.8 ( 0.01
10 ( 0.23
5.7 ( 0.10
7.6 ( 0.26
7.3 ( 0.56
6.2 ( 0.13
7.1 ( 0.24
6.4 ( 0.25
8.4 ( 0.64
6.7 ( 0.26
5.5 ( 0.17
8.0 ( 0.04
8.1 ( 0.05
6.4 ( 0.15
7.2 ( 0.13
5.5 ( 0.03
nd
6.2 ( 0.10
5.2 ( 0.77
5.6 ( 0.19
nd
1500 ( 300
770 ( 120
960 ( 81
500 ( 13
600 ( 56
250 ( 29
nd
240 ( 64
420 ( 220
280 ( 17
nd
carbachol
1
3
10
11
23
25
6.4 ( 0.26
7.4 ( 0.05
8.7 ( 0.05
7.7 ( 0.10
8.6 ( 0.12
7.5 ( 0.12
7.1 ( 0.08
6.0 ( 0.38
50 ( 10
39 ( 5.9
nd
nd
20 ( 5.4
nd
1
7.6 ( 0.18
3
nd
nd
10
11
23
24a
24b
24c
25
26
7.8 ( 0.50
nd
nd
nd
a
Data include the inhibition of [3H]-(R)-QNB binding (from A9
L cells) and inhibition of cAMP formation (from CHO cells). Data
represent the mean ( SEM from at least two assays, each
performed in triplicate; nd, not determined.
a
Data include the inhibition of [3H]-(R)-QNB binding and
stimulation of phosphoinositide metabolism. Data represent the
mean ( SEM from at least two assays, each performed in
triplicate; nd, not determined.
ligands, compound 11 displayed the highest affinity for
M2, M3, and M4 receptors with pKi values of 8.1, 7.6,
and 8.6, respectively (see Figure 1). Compound 23
displayed the highest affinity for M5 receptors with a
pKi of 6.9. None of the compounds exhibited marked
selectivity for a single muscarinic receptor subtype in
terms of binding affinity, although all compounds
displayed relatively low affinity for M5 receptors.
Agonist activity was assessed by measuring phos-
phoinositide metabolism in A9 L cells expressing M1,
M3, and M5 receptors and by assaying cAMP levels in
A9 L and Chinese hamster ovary (CHO) cells expressing
M2 and M4 receptors, respectively (see Tables 1-5).
Compound 11 displayed the highest activity at M1
receptors, with a pEC50 of 5.7 and a maximal response
of 40% of that of carbachol (see Figure 2). Compound
11 was essentially inactive at M3 receptors and dis-
played very low potency (pEC50 of 3.7) and activity (14%
of the carbachol response) at M5 receptors. Of the
compounds tested, 25 displayed the highest activity at
M3 receptors (57% of the carbachol response), while 26
displayed the highest activity at M5 receptors (28% of
the carbachol response). On the basis of the phosphoi-
nositide metabolism data, compound 11 would not be
expected to produce peripheral side effects (e.g., saliva-
tion, lacrimation, and diarrhea) associated with activa-
tion of M3 receptors, which are present in smooth muscle
and exocrine glands.
Agonist activity also was assessed by measuring the
ability of selected compounds to inhibit the formation
of cAMP stimulated by forskolin in A9 L cells expressing
M2 receptors and in CHO cells expressing M4 receptors.
Compound 11 displayed strong M2 agonist activity (76%
inhibition with a pIC50 of 9.3) as shown in Figure 3A
and Table 2. Carbachol was slightly less efficacious and
less potent (65% inhibition with a pIC50 of 7.9). All
responses were blocked by addition of the muscarinic
antagonist l-hyoscyamine (data not shown). Compound
11 also displayed partial agonist activity at M4 receptors
expressed in CHO cells (see Figure 3B and Table 4). On
the basis of the adenylyl cyclase activity data, compound
11 would be expected to produce hypothermia through
activation of central M2 receptors.
Ta ble 2. Pharmacological Properties of Novel Compounds at
M2 Receptors Expressed in A9 L Cellsa
M2 receptors
pKi
pIC50
Imax (%)
carbachol
1
3
10
11
23
25
7.0 ( 0.21
7.3 ( 0.15
9.9 ( 0.25
7.8 ( 0.26
8.1 ( 0.02
7.8 ( 0.21
7.8 ( 0.25
7.9 ( 0.23
nc
9.2 ( 0.29
nd
9.3 ( 0.30
nd
nd
65 ( 4.3
na
92 ( 18
nd
76 ( 4.8
nd
nd
a
Data include the inhibition of [3H]-(R)-QNB binding and
inhibition of cAMP formation. Data represent the mean ( SEM
from at least three assays, each performed in triplicate; na, no
significant change from basal levels; nc, not calculable due to
inconsistent dose response curves from assay to assay; nd, not
determined.
Ta ble 3. Pharmacological Properties of Novel Compounds at
M3 Receptors Expressed in A9 L Cellsa
M3 receptors
pKi
pEC50
Smax
carbachol
1
3
10
11
23
25
5.2 ( 0.24
7.1 ( 0.07
8.3 ( 0.28
7.1 ( 0.17
7.6 ( 0.06
7.1 ( 0.10
6.9 ( 0.14
4.7 ( 0.23
4.6 ( 0.34
nc
3.4 ( 0.39
nc
5.0 ( 0.39
4.4 ( 0.16
350 ( 48
280 ( 43
na
50 ( 5.6
na
73 ( 12
200 ( 28
a
Data include the inhibition of [3H]-(R)-QNB binding and
stimulation of phosphoinositide metabolism. Data represent the
mean ( SEM from at least two assays, each performed in
triplicate; na, no significant change from basal levels; nc, not
calculable due to inconsistent dose response curves from assay to
assay.
ether, the final compounds 25 and 26 were obtained in
36.3-48.0% yield.
P h a r m a cology. Compounds were initially examined
for receptor binding properties and abilities to stimulate
biochemical responses in cell lines expressing muscar-
inic receptor subtypes. Ligand binding affinities (pKi
values) were determined by measuring the inhibition
of specific [3H]-(R)-QNB (quinuclidinyl benzilate) bind-
ing to human muscarinic receptor subtypes expressed
in A9 L cells and compared with carbachol, 1, and 3 (see
Tables 1-5). In general, the tetrahydropyridyl thiadia-
zole derivatives inhibited the binding of [3H]-(R)-QNB
in a dose-dependent manner with pKi values ranging
from 5.7 to 8.6. Compound 25 displayed the highest
affinity for M1 receptors with a pKi of 8.4. Among the
In pharmacological screens in mice, M3 receptor-
mediated side effects (i.e., diarrhea, salivation, and
lacrimation) were minimal for compound 11 at doses up
to 10 mg/kg i.p (data not shown). The in vivo data for
compound 11 are consistent with a lack of efficacy at
M3 receptors expressed in A9 L cells. In contrast,
xanomeline produces diarrhea, salivation, and lacrima-

Muscarinic Agonists for Neurological Disorder Treatment
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20 4277
F igu r e 1. Inhibition of [3H]-(R)-QNB binding to muscarinic receptor subtypes expressed in A9 L cells by carbachol, 1, and 11.
Data represent the mean ( SEM from a minimum of three assays, each performed in triplicate.
tion at 1.0 and 10 mg/kg i.p.13 Compound 11 did produce
slight tremors and induced hypothermia (see Figure 4),
consistent with activity at central M2 receptors. Taken
together, the in vivo data suggest that compound 11 is
able to penetrate into the brain and act as an agonist
at M2 receptors involved in regulating movement and
body temperature. Substantial M2 effects were only seen
at the highest dose.
Ex vivo binding studies were conducted to examine
the ability of 11 to cross the blood-brain barrier. Rats
were injected with either saline, 11, or oxotremorine
(both at doses of 0.1 or 1.0 mg/kg, i.p.) and sacrificed at
either 30 or 60 min following injections. The brain was
removed rapidly, and membranes were prepared for
receptor binding assays. Displacement of [3H]QNB
binding was assessed in treated animals as compared
to saline-injected controls, and the results were ex-
pressed as the percentage inhibition of specific [3H]QNB
binding. Both 11 and oxotremorine inhibited [3H]QNB
binding in a dose-dependent manner, as shown in
Figure 5. Oxotremorine inhibited [3H]QNB binding at
both 30 and 60 min, while 11 inhibited [3H]QNB binding
only at 30 min. Thus, compound 11 appears to penetrate
readily into the brain at doses that produce minimal
peripheral side effects.
The pharmacological profile of compound 11 suggests
potential utility in the treatment of schizophrenia.
Compound 11 acted as an agonist at both M2 and M4
receptors, which may play important roles in regulating
dopamine release in the central nervous system. In
addition, 11 activated M1 receptors, which play a role
in memory and cognitive function, suggesting potential

4278 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
F igu r e 4. Hypothermia produced by 11 following i.p. admin-
istration in mice. Data represent the mean core body temper-
ature (in °C) of mice.
F igu r e 2. PI hydrolysis mediated by carbachol, 1, and 11
through human M1 receptors expressed in A9 L cells. Data
represent the mean ( SEM from a minimum of three assays,
each performed in triplicate.
F igu r e 5. Inhibition of [3H]QNB binding to rat brain by
oxotremorine (A) and 11 (B) following i.p. injections. Data
represent the mean ( SEM from a minimum of three animals
for each dose and time point (30 and 60 min).
for 11 in the treatment of schizophrenia or even drug
abuse. Additional studies are necessary to assess the
activity of 11 in peripheral tissues to verify the selectiv-
ity profile found in cell lines.
F igu r e 3. Inhibition of forskolin-stimulated cAMP formation
mediated by carbachol and 11. (A) Inhibition of cAMP forma-
tion through human M2 receptors expressed in A9 L cells. (B)
Inhibition of cAMP formation through human M4 receptors
expressed in CHO-K1 cells. Data represent the mean ( SEM
from a minimum of three assays, each performed in triplicate.
Com p u ta tion a l Ch em istr y. Molecular modeling of
molecules can help reveal important structural features
for consideration in studying the possible binding and
activation mechanisms of muscarinic agonists. Acetyl-
choline, the natural agonist for muscarinic receptors,
is subject only to limited conformational flexibility.
Acetylcholine is a small molecule, and its conformation
can be primarily characterized with gauche or trans
N-C-C-O arrangements. As a neurotransmitter, ace-
tylcholine diffuses from presynaptic neurons through an
extracellular solution into the binding cavity of receptors
on either pre- or postsynaptic neurons.
utility in treating cognitive deficits associated with
schizophrenia. The lack of activity at M3 receptors
suggests that compound 11 should have a low side effect
profile, particularly with respect to salivation, lacrima-
tion, and diarrhea. In addition, 11 did not activate M5
receptors yet bound with reasonable affinity. Because
M5 receptors promote dopamine release, the ability to
block M5 receptors may provide an additional advantage

Muscarinic Agonists for Neurological Disorder Treatment
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20 4279
F igu r e 6. Conformations of novel compounds after 130 ps gas phase and 500 ps in solution MD calculations.
According to a recent tandem binding model,14 an
acetylcholine molecule binds to the Asp99 carboxylate
group as an initial binding site. The Asp99 residue is
located at the junction of the third transmembrane
domain and the first extracellular loop but can become
exposed for binding of the ligand. Acetylcholine subse-
quently relocates to a more stable secondary binding
site, which is Asp105 in M1 receptors. The relocation
requires a further opening of the pore among trans-
membrane domains 3, 5, and 6, to accommodate ace-
tylcholine in the hypothesized binding cavity. In the
J akubik model, a second acetylcholine molecule binds
to Asp99 thereby covering the opened pore.
a large ligand. The binding process starts in the
conformation that the ligand adopts as it approaches
the extracellular loops. In a previous study,11 the
flexibility of the spacer was analyzed based on MD
simulations for bivalent dications including compound
3. The in solution stable, horseshoe-shaped structure
was docked to the first extracellular loop of the M1
receptor model. It was found, however, that the binding
energy is remarkably more negative when 3 is docked
in an extended conformation into the binding pocket of
the receptor. Thus, it is important to consider the
possible shapes that the bivalent molecules can adopt.
Their flexibility is crucial for reaching the binding
pocket in an extended conformation, which may not be
the lowest energy structure of the ligand. The previous
Such a concerted mechanism is easily imaginable for
a small agonist but becomes much more complicated for

4280 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
Ta ble 5. Pharmacological Properties of Novel Compounds at
an opened V structure, as for the other three com-
pounds. The average S‚‚‚S distance was 9.40 ( 2.45 Å,
allowing the location of the Cl- between the rings for
large and favorable van der Waals interactions. The net
result of such interactions may help keep the anion close
to the cationic headgroup.
M5 Receptors Expressed in A9 L Cellsa
M5 receptors
pKi
pEC50
Smax (%)
carbachol
1
3
10
11
23
25
26
4.7 ( 0.28
5.8 ( 0.08
7.2 ( 0.03
6.8 ( 0.29
6.4 ( 0.31
6.9 ( 0.08
6.2 ( 0.49
nd
5.2 ( 0.27
5.6 ( 0.16
nc
nc
3.7 ( 2.1
nc
200 ( 23
92 ( 15
na
23 ( 14
27 ( 15
32 ( 20
34 ( 26
55 ( 22
Counterions and ions of NaCl molecules added to the
modeling system had large effects on both the confor-
mation and the flexibility of molecules with a 13 atom
ethylene glycol spacer. In a separate MD run (using the
same simulation parameters), three NaCl molecules
were added to the system (for compound 11) in order to
mimic the isotonic saline concentration of 0.153 M NaCl
solution. A remarkable decrease was found for the
average S‚‚‚S distance (4.30 ( 0.96 Å). Thus, the
presence of a small excess of ions in the model leads to
remarkable stabilization of some ligand conformations.
When NaCl was not present in the solution, five simple
conformational conversions were identified between
local gauche and trans arrangements, and one trans-
gauche-trans conversion was identified along the O1-
C‚‚‚C-O5 path in the last 250 ps of the MD trajectory.
Not a single conversion was found, however, in the same
time interval when three NaCl molecules were added
to the system.
nc
5.9 ( 2.2
a
Data include the inhibition of [3H]-(R)-QNB binding and
stimulation of phosphoinositide metabolism. Data represent the
mean ( SEM from at least two assays, each performed in
triplicate; na, no significant change from basal levels; nc, not
calculable due to inconsistent dose response curves from assay to
assay.
Ta ble 6. Calculated N‚‚‚O5 Distances (in Å) from MD
Simulations at T ) 310
compd
type
dmin
dmax
dave
3
dication
7.1
8.4
5.4
14.3
5.7
10.7
13.0
14.6
17.8
18.4
17.2
16.5
10.6 ( 1.2
10.9 ( 1.2
11.2 ( 2.8
16.8 ( 0.7
11.0 ( 2.9
13.9 ( 1.3
24b
24c
26
monocation
monocation
monocation
monocation
11
A possible explanation for such differences in confor-
mational flexibility is that the NaCl ions create strongly
bound and relatively rigid hydration spheres. Many
water molecules belong to the spheres, and their trans-
lational and rotational freedom is reduced as compared
to water molecules in solutions without excess NaCl.
In the somewhat “frozen” ionic environment, conforma-
tional changes along the spacer become less feasible as
well.
studies were extended for 24b,c, 26, and 11 to include
an analysis of the consequences of replacing a cationic
end group with neutral groups of different polarities.
The structures after 130 ps gas phase and 500 ps in
solution MD calculations differed essentially from the
starting structures. All four molecules could be charac-
terized as conformers with opened V or U shapes (see
Figure 6). Table 6 compares the N‚‚‚O5 distances,
including those for the dicationic compound 3. Average
distances (dave) were close for 3, 24b, and 26 and differed
significantly from the value for 24c. Such differences
may not be very surprising, considering that 24c has
the only entirely nonpolar end group in the series
studied. Thus, the smaller dave values are associated
with more polar end groups.
The results from MD calculations are counterintuitive
to the concept that dicationic headgroups tend to
separate the most for compound 3. In aqueous solution,
however, there are counterions that can shield the
cationic headgroups thereby balancing the dicationic
repulsions with attractive anion-cation interactions. In
fact, previous modeling studies for 3 indicated Cl-
counterions located in close proximity to the protons of
the tertiary amines. As a result, there are altogether
two repulsive interactions (one dicationic, one dianionic),
as compared to four favorable (cation-anion) electro-
static interactions. Furthermore, the bulky end groups
exert large and favorable van der Waals interactions
with each other.
During the development of the M1 receptor model, the
receptor was placed in a water box. The Sybyl software
removed water molecules that formed unfavorable
contacts with receptor atoms but left the cavities and
pores with the transmembrane helices filled with water.
Even within a short MD run, however, most of the water
molecules rapidly left the cavities and returned to the
bulk water. Accordingly, neither a profound aqueous
environment nor Na+ and Cl- ions are expected to
surround the cationic headgroup of agonist molecules
located within the transmembrane domain cavities.
Consequently, molecules with long spacers (such as
compound 11) become more flexible when leaving the
saline solution. At the same time, the receptor has to
account for the electrostatic repulsion for the two
protonated amines in 3 and the electrostatic stabiliza-
tion of the protonated tetrahydropyridyl ring in each
molecule.
For 3, the average N‚‚‚N distance is 11.8 ( 1.4 Å in
aqueous solution,11 a distance favorable for tandem
binding to the Asp99 and Asp105 carboxylate groups
by a single bivalent molecule. Such a binding mode
would require, however, an unlikely wide opening of the
pore formed by transmembrane helices 3, 5, and 6 in
order to accommodate a horseshoe-shaped structure for
3 or a thermodynamically unfavorable, wormlike motion
of the ligand, to avoid repulsion by the helical elements
crossing this theoretical path. No such problem is
present with compounds 24b,c, 26, and 11, with only
one protonated ring and without the possibility for a
In the new series of monocationic ligands, there is no
electrostatic repulsion to be balanced by counterions.
Accordingly, the Cl- counterion was found at a fairly
large average distance of 14-18 Å from the proton
connecting to the N-atom within the tetrahydropyridyl
ring in 24b,c and 26. In contrast, the average Cl-‚‚‚H+
distance was found to be 5.3 Å for compound 11. For
this system, however, the two thiadiazole rings stayed
relatively close to each other forming a U rather than

Muscarinic Agonists for Neurological Disorder Treatment
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20 4281
single molecule, tandem protonated nitrogen‚‚‚carboxylate
interaction pair. For such molecules, however, the
flexibility of the spacer is important. A flexible spacer
may help compounds reach conformations allowing
favorable interactions between the variable end groups
and the receptor residues either in the extracellular loop
or within the transmembrane helices. Such interactions
may play a decisive role in receptor subtype selectivity.
ing pKa value for monocationic 3. The difference,
however, is significant considering the very small
standard deviations (no more than 0.004) for the two
molecules. In principle, the alcoholic end group in 24b
may get close to the tertiary amine and, through
formation of an intramolecular hydrogen bond, could
reduce the protonation capacity of the nitrogen. None
of the snapshots throughout the MD simulations showed,
however, such a structure either in the gas phase or in
the aqueous solution. Such a conformation would impose
a strong structural restraint on the molecule and could
be conceivable only in the gas phase. Hydrogen bond
formation is much easier with a water molecule in
aqueous solution. Thus, in the absence of a returning
side chain that reduces the accessibility of the tertiary
nitrogen by water molecules, the pKa of 8.27 vs 8.44
appears to reflect small changes in the electron struc-
ture of the two molecules. The results also suggest
that only small deviations from the pKa of 8.27 are
expected for monoamines in this series with different
end groups.
Differences in the protonation constants (pKa) may
be related to variations in the spatial orientation of the
molecules. Molecular flexibility appears to be a function
of the chemical environment. Different pKa values may
reflect that the tertiary amine groups are more or less
open to hydration and protonation, depending on the
horseshoe, V, or U shape of the spacer. The pKa values
for 3 were measured as 8.44 ( 0.004 and 7.77 ( 0.004.
For a representative of the new mono-tertiary-amine
molecules, 24b was selected, and a pKa of 8.27 ( 0.003
was found. The results should be analyzed with respect
to the length of the spacer and the possible interaction
of the polar end groups.
Su m m a r y. Compound 11 displays an interesting
activity profile with agonist activity at M1, M2, and M4
receptors. The pharmacological profile suggests poten-
tial utility in the treatment of psychiatric disorders
including schizophrenia. Compound 11 displays a very
low side effect profile in preliminary pharmacological
studies in mice, consistent with its lack of activity at
M3 receptors in biochemical assays. Compound 11 does
exhibit robust activity at M2 receptors, which may result
in cardiovascular side effects in clinical studies. It
remains unclear, however, whether M2 receptor activity
is desirable for antipsychotic activity. Further studies
are necessary to optimize the pharmacological proper-
ties of the novel series of compounds for the treatment
of schizophrenia. Molecular modeling studies revealed
that monocationic bivalent ligands with neutral and
polar end groups take similar effective lengths as
characterized by the N‚‚‚O5 distance as compared with
the dicationic (R,ω-di-tetrahydropyridyl-thiadiazole)
ligand. The flexibility of the spacer of the present
monocationic ligands increases when the molecules
leave the fluid in the synaptic cleft and penetrate into
the receptor binding pocket. Upon replacement of the
(neutral) polar end group by an apolar moiety (ethyl),
the effective length is considerably increased. This
structural feature can be usefully considered in design-
ing new derivatives with optimal effective length and
flexibility. Ultimately, derivatives of compound 11 may
be useful in treating psychosis and cognitive deficits
associated with schizophrenia.
The basicity of monoamines is practically independent
of the length of the chain. Ethylamine and octadecy-
lamine (C18H37NH2) have pKa values of 10.65 and 10.60
at 25 °C.15 Thus, for primary amines, even if the alkyl
chain folds back, basicity is not affected greatly. How-
ever, pKa values for R-ω diamines are sensitive to the
length of the spacer. The pKa values for the first and
second protonations of 1,2-diamino-ethane are 9.92 and
6.86, respectively. It is normal that the pKa for the
dication is smaller than for the monocation, since it
reflects that a more acidic environment is required for
the protonation of a cation. The corresponding values
for the longest R-ω diamine, 1,8-diamino-octane (H2N-
(CH2)8-NH2), found in ref 15, are 11.0 and 10.1. These
values reveal interesting peculiarities for diamines: the
basicity increases with a longer alkyl chain (in contrast
to the nearly constant value for monoamines), and the
difference of the pKa values decreases to about 0.9 units.
This finding can be explained theoretically, assuming
that the two amine groups interact in 1,2-diamino-
ethane. In a gauche conformation, a five-membered ring
with an H2N‚‚‚+HNH2 intramolecular hydrogen bond
can be formed in the monocation, making the protona-
tion of the second amino group less feasible.16 As a
result, the pKa for the dication is relatively low for 1,2-
diamino-ethane. For 1,6-diamino-hexane, the pKa values
are 11.86 and 10.76 at 0 °C. Thus, the increase in
basicity is not monotonic, but the trend is clear for the
difference of the pKa values. The decreasing difference
indicates that the two amine groups become indepen-
dent with respect to protonation, and implicitly suggests
that the shape of the spacer may not be decisive.
Exp er im en ta l Section
Ch em istr y. Reactions were carried out under nitrogen.
Melting points were determined on a Fisher-J ohns melting
point apparatus and are presented uncorrected. 1H and 13C
NMR spectra were obtained with a Bruker ACF 300 MHz
spectrometer. Elemental analyses (C, H, N) were performed
by Atlantic Microlab, Inc., GA; analytical results were within
(0.4% of the theoretical values for the formula given unless
otherwise indicated. Precoated silica gel GHLF UNIPLATES
(250 µm) purchased from Analtech, Inc., DE, were used for
thin-layer chromatography, and spots were examined with UV
light at 254 nm or with iodine vapor. Column chromatography
purification was performed on Davisil silica gel (200-425
mesh) obtained from Fisher Scientific. THF was dried over
Although a -(CH2)8- spacer may behave differently
than the long and chemically different spacer in 3, the
obtained results indicate that the protonation of 3 can
be understood on the basis of the above qualitative
argument. The O-CH2-(CH2-O-CH2)3-CH2-O spacer
is long enough to separate the two protonation sites in
any conformation, and the pKa for the formation of the
dication is smaller by about 0.7 units than that for the
monocation.
It is important to notice that the (single) pKa for 24b
is 8.27, which differs only slightly from the correspond-

4282 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
sodium-benzophenone ketyl and distilled. All other com-
mercially available solvents and reagents were used without
further purification unless otherwise specified.
61.9, 69.4, 70.5, 70.6, 70.8, 70.9, 71.0, 72.7, 123.7, 127.8, 135.2,
145.3, 148.8, 150.3, 162.7.
Tetr a (eth ylen e glycol) Eth yl[3-(p yr id -3-yl)-1,2,5-th ia -
d ia zol-4-yl]eth er (9c). Compound 9c was prepared from 7c
(336 mg, 1.5 mmol) and 5 (100 mg, 0.5 mmol) using a similar
procedure described earlier to afford the title compound 9c
(185 mg, as a yellow brown oil, 95.3% yield); Rf ) 0.63 (5%
Com p ou n d 7c. A suspension of 60% NaH in mineral oil (1
g, 25.23 mmol) was washed with anhydrous hexane (10 mL)
and suspended in freshly distilled THF (30 mL). Then, 7b (4
mL, 22.94 mmol) was added to the above suspension and the
reaction mixture was refluxed for 1 h. After the mixture was
cooled to 0 °C using an ice bath, bromoethane (0.88 mL, 11.47
mmol) was added slowly, and the reaction mixture was kept
at 0 °C for 1 h. The solvent was evaporated under a vacuum,
ice water was added, and then, the aqueous solution was
extracted twice with CHCl3. The combined extract was washed
with brine, dried over anhydrous Na2SO4, and concentrated
to yield the crude compound. After silica gel chromatography
(4% CH3OH in CHCl3 as eluting solvent), 7c was obtained (650
mg, as a colorless oil, 25.5% yield); Rf ) 0.50 (2% CH3OH in
1
CH3OH in CHCl3). H NMR (CDCl3): δ 1.58 (t, 3H), 3.50 (q,
2H), 3.53-3.70 (m, 12H), 3.90 (t, 2H), 4.65 (t, 2H), 7.36 (m,
1H), 8.41 (d, J ) 8.0 Hz, 1H), 8.61 (dd, J ) 1.6 and 4.8 Hz,
1H), 9.36 (s, 1H). 13C NMR (CDCl3): δ 15.3, 66.7, 69.3, 69.9,
70.7, 70.8, 70.9, 123.5, 127.6, 134.8, 145.2, 148.9, 150.3, 162.6.
1-(3-Met h oxy-1,2,5-t h ia d ia zol-4-yloxy)-12-[3-(p yr id -3-
yl)-1,2,5-th ia d ia zol-4-yloxy]d od eca n e (13). A suspension of
60% NaH in mineral oil (80 mg, 2 mmol) was washed with
anhydrous hexane (10 mL) and suspended in freshly distilled
THF (20 mL). Then, 8 (182 mg, 0.5 mmol) in 20 mL of THF
was added to the above suspension. The reaction mixture was
refluxed for 30 min. Compound 12 (61 mg, 0.4 mmol) in THF
(20 mL) was added, and the reaction mixture was refluxed for
24 h. The mixture was evaporated under a vacuum and
quenched with ice water (20 mL), and then, the aqueous
solution was extracted twice with CH2Cl2. The combined
extract was washed with water, dried over anhydrous
Na2SO4, and concentrated to yield the crude compound. After
silica gel chromatography (25% EtOAc in hexane as eluting
solvent), 13 was obtained (80 mg, as a pale yellow oil, 41.9%
1
3
CHCl ). H NMR (CDCl3): δ 1.17 (t, 3H), 2.75 (t, 1H, -OH),
3.50 (q, 2H), 3.54-3.71 (m, 16H). 13C NMR (CDCl3): δ 15.1,
61.5, 66.5, 69.7, 70.3, 70.5, 70.6, 72.6.
Com p ou n d 12. A suspension of 60% NaH in mineral oil
(800 mg, 20 mmol) was washed with anhydrous hexane (10
mL), and then, anhydrous methanol (20 mL) was added under
an ice bath. The resulting suspension was refluxed for 1 h
before being transferred to an addition funnel. Under an ice
bath, the above sodium methoxide suspension was added
slowly to a solution of 29 (3.196 g, 20 mmol) in THF (20 mL),
and the reaction mixture was maintained at room temperature
for 24 h. The mixture was evaporated under a vacuum and
quenched with ice water (20 mL), and then, the aqueous
solution was extracted with hexane three times. The combined
extracts were washed with water, dried over anhydrous Na2-
SO4, and concentrated to yield the crude compound. After silica
gel chromatography (hexane, then 5% CH3OH/CHCl3 as elut-
ing solvents), 12 was obtained (305 mg, as a white powder,
10.1% yield); mp 52.5-53.5 °C (sublimated); Rf ) 0.40 (hex-
1
yield); Rf ) 0.44 (50% EtOAc in hexane). H NMR (CDCl3): δ
1.27-1.50 (m, 16H), 1.76-1.92 (m, 4H), 4.08 (s, 3H, -OCH3),
4.37 (t, 2H), 4.50 (t, 2H), 7.38 (m. 1H), 8.41 (d, J ) 8.0 Hz,
1H), 8.64 (s, 1H), 9.40 (s, 1H). 13C NMR (CDCl3): δ 25.9, 26.1,
29.0, 29.1, 29.4, 29.6, 57.6, 71.0, 71.6, 123.6, 127.8, 134.8, 145.1,
148.8, 150.3, 152.0, 152.5, 163.0.
Tetr a (eth ylen e glycol) (3-Meth oxy-1,2,5-th ia d ia zol-4-
yl)[3-(p yr id -3-yl)-1,2,5-t h ia d ia zol-4-yl]et h er (14). Com-
pound 14 was prepared from 9b (355.4 mg, 1.0 mmol) and 12
(180.7 mg, 1.2 mmol) using a similar procedure described in
the synthesis of 13 to afford the title compound 14 (125 mg,
as a yellow oil, 26.6% yield); Rf ) 0.60 (5% CH3OH in CH2-
Cl2). 1H NMR (CDCl3): δ 3.65-3.72 (m, 8H), 3.84 (t, 2H), 3.93
(t, 2H), 4.06 (s, 3H, -OCH3), 4.53 (t, 2H), 4.67 (t, 2H), 7.39
(m. 1H), 8.43 (d, J ) 8.1 Hz, 1H), 8.63 (dd, J ) 1.6 and 4.8 Hz,
1H), 9.38 (s, 1H). 13C NMR (CDCl3): δ 57.6, 69.2, 69.3, 69.9,
70.4, 70.9, 123.6, 127.7, 134.9, 145.3, 148.9, 150.3, 151.7, 152.5,
162.7.
P r ep a r a t ion of Mon o[3-(1-m et h ylp yr id in iu m -3-yl)-
1,2,5-th ia d ia zol-4-yl]eth er Iod id es. Gen er a l P r oced u r e.
12-[3-(1-Meth ylp yr id in iu m -3-yl)-1,2,5-th ia d ia zol-4-yloxy]-
1-d od eca n ol Iod id e (15). To a solution of 8 (100 mg, 0.275
mmol) in acetone (5 mL) and CHCl3 (5 mL) was added CH3I
(3 mL, 48 mmol), and the solution was stirred under nitrogen
for 36-48 h at room temperature. The residual quaternary
iodide 15 was obtained by removing the solvents under reduced
pressure and then recrystallized in acetone/ether (136 mg, as
a pale yellow-white powder, 97.8% yield); mp 107-109 °C. 1H
NMR (CDCl3): δ 1.30-1.58 (m, 18H), 1.95 (m, 2H), 3.64 (m,
2H, -CH2OH), 4.62 (t, 2H, -OCH2-), 4.81 (s, 3H, -NCH3),
8.25 (m, 1H), 9.16 (d, J ) 8.4 Hz, 1H), 9.39 (s, 1H), 9.71 (d, J
) 5.9 Hz, 1H).
1
ane). H NMR (CDCl3): δ 4.15 (s, -OCH3). 13C NMR (CDCl3):
δ 58.4, 133.3, 160.4.
P r ep a r a tion of Mon o[3-(p yr id -3-yl)-1,2,5-th ia d ia zol-4-
yl]eth er Der iva tives. Gen er a l P r oced u r e. Com p ou n d 8.
A suspension of 60% NaH in mineral oil (560 mg, 14 mmol)
was washed with anhydrous hexane (10 mL) and suspended
in freshly distilled THF (20 mL). Then, 6 (1.012 g, 5 mmol)
was added to the above suspension and the reaction mixture
was refluxed for 1 h. Compound 5 (400 mg, 2 mmol) in THF
(20 mL) was added, and the reaction mixture was refluxed for
24 h. The mixture was evaporated under a vacuum and
quenched with ice water (20 mL). After adjusting the pH < 7,
the aqueous solution was extracted twice with CHCl3. The
combined extract was washed with water, dried over anhy-
drous Na2SO4, and concentrated to yield the crude compound.
After silica gel chromatography (25-50% EtOAc/hexane as
eluting solvent), 8 was obtained (450 mg, as a white powder,
61.9% yield); Rf ) 0.63 (50% EtOAc in hexane). 1H NMR
(CDCl3): δ 1.25-1.56 (m, 18H), 1.86 (m, 2H), 2.30 (m, 1H,
-OH), 3.61 (t, 2H, -CH2OH), 4.49 (t, 2H, -OCH2-), 7.38 (m,
1H), 8.41 (d, 1H, J ) 6.5 Hz), 8.62 (d, 1H, J ) 4.6 Hz), 9.37 (s,
1H). 13C NMR (CDCl3): δ 25.9, 26.1, 29.0, 29.4, 29.5, 29.6, 33.0,
63.0, 71.6, 123.6, 127.9, 134.9, 145.1, 148.7, 150.2, 163.0.
Tr i(eth ylen e glycol) Eth yl[3-(p yr id -3-yl)-1,2,5-th ia d ia -
zol-4-yl]eth er (9a ). Compound 9a was prepared from 7a (750
mg, 4 mmol) and 5 (200 mg, 1 mmol) using a similar procedure
described earlier to afford the title compound 9a (300 mg, as
a brown oil, 88.4% yield); Rf ) 0.56 (2.5% CH3OH in CHCl3).
1H NMR (CDCl3): δ 1.14 (t, 3H), 3.50 (q, 2H), 3.53-3.70 (m,
8H), 3.90 (t, 2H), 4.64 (t, 2H), 7.35 (m, 1H), 8.40 (d, J ) 8.0
Hz, 1H), 8.59 (dd, J ) 1.6 and 4.8 HZ, 1H), 9.35 (s, 1H).
Com p ou n d 9b. Compound 9b was prepared from 7b (10.1
mmol) and 5 (500 mg, 2.53 mmol) using a similar procedure
described earlier to afford the title compound 9b (375 mg, as
a pale yellow oil, 41.7% yield); Rf ) 0.60 (10% CH3OH in
CHCl3). 1H NMR (CDCl3): δ 3.57-3.75 (m, 12H), 3.95 (t, 2H),
4.70 (t, 2H), 7.41 (m. 1H), 8.44 (d, J ) 8.1 Hz, 1H), 8.65 (dd,
J ) 1.6 and 4.8 Hz, 1H), 9.40 (s, 1H). 13C NMR (CDCl3): δ
1-(3-Met h oxy-1,2,5-t h ia d ia zol-4-yloxy)-12-[3-(1-m et h -
ylp yr id in iu m -3-yl)-1,2,5-th ia d ia zol-4-yloxy]d od eca n e Io-
d id e (16). Compound 16 was prepared from 13 (80 mg, 0.17
mmol) and CH3I (1 mL, 16 mmol) using a similar procedure
described earlier to afford the quaternary iodide 16 (85 mg,
as a yellow powder, 81.8% yield); mp 112-113 °C. 1H NMR
(CDCl3): δ 1.30-1.45 (m, 16H), 1.83 (m, 2H), 1.95 (m, 2H),
4.11 (s, 3H, -OCH3), 4.40 (t, 2H), 4.61 (t, 2H), 4.82 (s, 3H,
-NCH3), 8.26 (m, 1H), 9.15 (d, J ) 8.4 Hz, 1H), 9.41 (s, 1H),
9.70 (d, J ) 6.0 Hz, 1H). 13C NMR (CDCl3): δ 25.9, 26.1, 29.0,
29.4, 29.6, 29.7, 50.9, 57.7, 71.1, 72.7, 128.8, 131.7, 139.6, 142.1,
143.0, 146.0, 152.0, 152.6, 163.4.
Tr i(et h ylen e glycol) E t h yl[3-(1-m et h ylp yr id in iu m -3-
yl)-1,2,5-th ia d ia zol-4-yl]eth er Iod id e (17a ). Compound 17a
was prepared from 9a (300 mg, 0.88 mmol) and CH3I (3 mL,

Muscarinic Agonists for Neurological Disorder Treatment
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20 4283
48 mmol) using a similar procedure described earlier to afford
the quaternary iodide 17a (425 mg, as a yellow oil, 99.7%
yield). 1H NMR (DMSO-d6): δ 1.05 (t, 3H), 3.48-3.3.54 (q, 2H),
3.55-3.3.75 (m, 8H), 3.90 (t, 2H), 4.45 (s, 3H, -NCH3), 4.67
(t, 2H), 8.29(m, 1H), 9.07 (m, 2H), 9.51(s, 1H). 13C NMR
(DMSO-d6): δ 15.0, 48.5, 65.5, 68.1, 69.1, 69.8, 70.8, 128.0,
130.2, 141.2, 142.6, 143.3, 146.0, 162.5.
1H, vinylic). 13C NMR (CDCl3): δ 25.9, 26.1, 26.7, 29.0, 29.3,
29.6, 29.7, 33.0, 46.1, 51.4, 55.2, 63.2, 71.1, 128.5, 129.5, 162.8.
The above free base (19) was redissolved in acetic anhydride
(6 mL, redistilled), and ZnCl2 powder (10 mg, freshly dried in
a vacuum oven for 4 days) was added. Then, the reaction
mixture was maintained in an oil bath at 85 °C for 3 h. Ice
was added to quench the reaction. The aqueous solution was
extracted twice with CHCl3. The combined organic extract was
washed with 10% K2CO3 (50 mL) and then water and dried
over anhydrous Na2SO4. The solvent was removed under
reduced pressure, and the residue was chromatographed on a
column of silica gel using CHCl3-CH3OH (97.5:2.5) as eluent.
The purified free base of the ester (27) was redissolved in
anhydrous CH2Cl2 (5 mL) and was cooled to 0 °C. Then, 1 M
HCl/ether (1 mL) was added, and the flask was kept at 0 °C
for 5 min. The title compound 25 was recrystallized from
acetone/ether (53.4 mg, as a hygroscopic yellow-white powder,
48.0% yield); mp 85-86 °C; free base Rf ) 0.45 [CHCl3/CH3-
Tetr a (eth ylen e glycol) Mon o[3-(1-m eth ylp yr id in iu m -
3-yl)-1,2,5-th ia d ia zol-4-yl]eth er Iod id e (17b). Compound
17b was prepared from 9b (100 mg, 0.28 mmol) and CH3I (1.5
mL, 24 mmol) using a similar procedure described earlier to
afford the quaternary iodide 17b (128 mg, as a yellow oil,
1
91.5% yield). H NMR (CDCl3): δ 3.10 (m, 1H, -OH), 3.53-
3.79 (m, 2H), 3.99 (t, 2H), 4.74 (t, 2H), 4.81 (s, 3H, -NCH3),
8.32 (m, 1H), 9.16 (d, J ) 8.3 Hz, 1H), 9.51(s, 1H), 9.70 (d, J
) 6.0 Hz, 1H). 13C NMR (CDCl3): δ 50.6, 61.4, 68.9, 70.2, 70.6,
70.7, 70.8, 70.9, 128.9, 131.4, 140.2, 142.7, 143.2, 146.1, 162.8.
Tetr a (eth ylen e glycol) Eth yl[3-(1-m eth ylp yr id in iu m -
3-yl)-1,2,5-t h ia d ia zol-4-yl]et h er Iod id e (17c). Compound
17c was prepared from 9c (185 mg, 0.48 mmol) and CH3I (2
mL, 32 mmol) using a similar procedure described earlier to
afford the quaternary iodide 17c (243 mg, as a yellow oil, 95.9%
1
OH (95:5)]. H NMR (CDCl3): δ 1.28-1.43 (m, 18H), 1.62 (m,
2H), 1.72 (m, 2H), 2.05 (s, 3H, acetyl), 2.59 (m, 1H), 2.97 (s,
3H, -NCH3), 3.25 (m, 1H), 3.57 (m, 1H), 3.78 (m, 1H), 4.05 (t,
2H), 4.46 (t, 2H), 7.27 (s, 1H, vinylic). 13C NMR (CDCl3): δ
21.2, 23.0, 26.1, 28.8, 29.0, 29.4, 29.7, 43.5, 52.2, 64.8, 71.7,
124.0, 126.9, 144.1, 162.7, 171.4. Anal. (C22H37N3O3S1‚HCl‚
0.5H2O) C, H, N.
1-(3-Meth yoxy-1,2,5-th ia d ia zol-4-yloxy)-12-[3-(1-m eth -
yl-1,2,5,6-tetr a h yd r o-p yr id -3-yl)-1,2,5-th ia d ia zol-4-yloxy]-
d od eca n e Hyd r och lor id e (10) (CDD-0308-A). Compound
10 was prepared from 16 (85 mg, 0.237 mmol) using a similar
procedure described earlier to afford the title compound 10
(50 mg, as a pale yellow powder, 65.3% yield); mp 108-109
°C; free base Rf ) 0.33 [EtOAc/hexane (50:50)]. 1H NMR
(CDCl3): δ 1.19-1.29 (m, 16H), 1.42 (m, 4H), 2.58 (m, 1H),
2.96 (s, 3H, -NCH3), 2.96 (m, 1H), 3.27 (m, 1H), 3.50 (m, 1H),
3.77 (m, 1H), 4.11 (s, 3H, -OCH3), 4.37-4.53 (m, 5H), 7.27 (s,
1H, vinylic), 13.1 (s, 1H). 13C NMR (CDCl3): δ 22.8, 26.0, 26.1,
29.0, 29.5, 29.7, 43.3, 50.0, 52.0, 57.7, 71.1, 71.7, 124.0, 126.9,
144.1, 152.0, 152.6, 162.8. Anal. (C23H37N5O3S2‚HCl‚1.5H2O)
C, H, N.
1
yield). H NMR (CDCl3): δ 1.17 (t, 3H), 3.48-3.3.54 (q, 2H),
3.55-3.3.75 (m, 12H), 4.00 (t, 2H), 4.73 (t, 2H), 4.79 (s, 3H,
-NCH3), 8.28(m, 1H), 9.14 (d, J ) 8.3 Hz, 1H), 9.55(s, 1H),
9.57 (d, J ) 6.0 Hz, 1H). 13C NMR (CDCl3): δ 15.3, 50.7, 66.9,
69.9, 70.7, 70.8, 71.0, 128.8, 131.5, 140.0, 142.6, 143.3, 146.0,
162.9.
Tetr a (eth ylen e glycol) (3-Meth oxy-1,2,5-th ia d ia zol-4-
yl)[3-(1-m et h yl-p yr id in iu m -3-yl)-1,2,5-t h ia d ia zol-4-yl]-
eth er Iod id e (18). Compound 18 was prepared from 14 (125
mg, 0.27 mmol) and CH3I (1 mL, 16 mmol) using a similar
procedure described earlier to afford the quaternary iodide 18
(160 mg, as a yellow oil, 98.4% yield). 1H NMR (CDCl3): δ
3.58-3.3.66 (m, 6H), 3.69 (m, 2H), 3.76 (m, 2H), 3.93 (m, 2H),
4.00 (s, 3H, -OCH3), 4.43 (t, 2H), 4.67 (t, 2H), 4.70 (s, 3H,
-NCH3), 8.23 (m, 1H), 9.09 (d, J ) 8.3 Hz, 1H), 9.40 (d, J )
6.0 Hz, 1H), 9.52 (s, 1H). 13C NMR (CDCl3): δ 50.5, 57.6, 68.8,
69.0, 69.7, 70.4, 70.6, 70.8, 128.7, 131.3, 139.9, 142.5, 143.2,
145.6, 151.4, 152.2, 162.7.
P r ep a r a tion of Mon o[3-(1-m eth yl-1,2,5,6-tetr a h yd r o-
p yr id -3-yl)-1,2,5-t h ia d ia zol-4-yl]et h er H yd r och lor id es.
Gen er a l P r od ed u r e. 12-[3-(1-Meth yl-1,2,5,6-tetr a h yd r o-
p yr id -3-yl)-1,2,5-th ia d ia zol-4-yloxy]-1-d od eca n ol Hyd r o-
ch lor id e (23) (CDD-0297-A). The pyridinium iodide 15 (100
mg, 0.198 mmol) was dissolved in a mixture of CH3OH (10
mL) and CHCl3 (10 mL). The solution was cooled to 0-5 °C,
and NaBH4 (30 mg, 0.792 mmol) was added. After the mixture
was stirred at 0-5 °C for 2 h, another 30 mg of NaBH4 was
added. The reaction continued for 2 h. Ice water was added to
the reaction mixture, which was then extracted twice with
CHCl3. The combined organic extract was washed with water
and dried over anhydrous Na2SO4. The solvent was removed
under reduced pressure, and the residue was chromatographed
on a column of silica gel using CHCl3-CH3OH (95:5) as eluent.
The purified free base 19 was redissolved in a mixture of CH3-
OH/CHCl3 (10 mL, 5:5) and was cooled to 0 °C. Then, dry HCl
gas was bubbled through the solution for 3 min, and the title
compound 23 was recrystallized from methanol/ether (61.5 mg,
as a white powder, 72.7% yield); mp 97-98 °C; free base Rf )
0.69 [CHCl3/CH3OH (90:10)]. 1H NMR (D2O): δ 1.01-1.23 (m,
18H), 1.63 (m, 2H), 2.53 (m, 2H), 2.84 (s, 3H, -NCH3), 3.27
(m, 4H), 4.03 (m, 2H), 4.22 (m, 2H), 7.03 (s, 1H, vinylic). 13C
NMR (CDCl3): δ 22.6, 25.7, 25.9, 28.8, 29.1, 29.4, 32.8, 43.1,
49.7, 51.7, 63.0, 71.4, 123.7, 126.6, 143.9, 162.5. Anal.
(C20H35N3O2S1‚HCl‚0.5H2O) C, H, N.
Tr i(eth ylen e glycol) Eth yl[3-(1-m eth yl-1,2,5,6-tetr a h y-
d r op yr id -3-yl)-1,2,5-th ia d ia zol-4-yl]eth er Hyd r och lor id e
(24a ) (CDD-0282-A). Reduction of the pyridinium iodide 17a
(420 mg, 0.87 mmol) yielded the hydrochloride 24a (180 mg,
as a yellow hygroscopic solid, 51.3% yield); mp: 75-76 °C; free
1
base Rf ) 0.53 [CHCl3/CH3OH (95:5)]. H NMR (D2O): δ 0.95
(t, 3H), 2.53 (m, 2H), 2.85 (s, 3H, -NCH3), 3.10 (m, 1H), 3.35
(m, 1H), 3.44-3.57 (m, 10H), 3.80 (m, 2H), 3.87 (m, 1H), 4.25
(m, 1H), 4.46 (m, 2H), 7.06 (s, 1H, vinylic). Anal. (C16H27N3O4S1‚
HCl‚0.5H2O) C, H, N.
Tetr a (eth ylen e glycol) Mon o[3-(1-m eth yl-1,2,5,6-tet-
r a h yd r op yr id -3-yl)-1,2,5-t h ia d ia zol-4-yl]et h er H yd r o-
ch lor id e (24b) (CDD-0300-A). Reduction of the pyridinium
iodide 17b (120 mg, 0.24 mmol) yielded the hydrochloride 24b
(75.7 mg, as a pale yellow oil, 76.9% yield); free base Rf ) 0.38
[CHCl3/CH3OH (90:10)]. 1H NMR (CDCl3): δ 2.56-2.62 (m,
2H), 2.95 (s, 3H, -NCH3), 3.03 (m, 1H), 3.55-3.82 (m, 14H),
3.90 (m, 2H), 4.50 (m, 1H), 4.63 (m, 2H), 7.30 (s, 1H, vinylic).
13C NMR (CDCl3): δ 22.8, 43.3, 50.0, 51.9, 61.9, 69.3, 70.3,
70.5, 70.9, 72.7, 123.7, 127.5, 144.3, 162.4. Anal. (C16H27N3O5S1‚
HCl‚0.5H2O) C, H, N.
Tetr a (eth ylen e glycol) Eth yl[3-(1-m eth yl-1,2,5,6-tet-
r a h yd r op yr id -3-yl)-1,2,5-t h ia d ia zol-4-yl]et h er H yd r o-
ch lor id e (24c) (CDD-0301-A). Reduction of the pyridinium
iodide 17c (243 mg, 0.46 mmol) yielded the hydrochloride 24c
(93 mg, as a pale white powder, 46.2% yield); mp 64.5-66 °C;
1
free base Rf ) 0.49 [CHCl3/CH3OH (95:5)]. H NMR (CDCl3):
δ 1.20 (t, 3H), 2.58 (m, 1H), 2.95 (s, 3H, -NCH3), 3.03 (m, 1H),
3.23 (m, 1H), 3.47-3.88 (m, 16H), 3.90 (m, 2H), 4.48 (m, 1H),
4.62 (m, 2H), 7.29 (s, 1H, vinylic), 13.1 (s, 1H). 13C NMR
(CDCl3): δ 15.4, 22.7, 43.2, 49.9, 51.9, 66.8, 69.3, 70.0, 70.3,
70.8, 123.7, 127.4, 144.3, 162.4. Anal. (C18H31N3O5S1‚HCl‚
0.5H2O) C, H, N.
12-[3-(1-Meth yl-1,2,5,6-tetr a h yd r op yr id -3-yl)-1,2,5-th ia -
d ia zol-4-yloxy]-1-d od ecyl Aceta te Hyd r och lor id e (25)
(CDD-0299-A). Compound 25 was prepared from 15 (120 mg,
0.237 mmol) using a similar procedure described earlier to
afford the free base 19 (60 mg yellow oil, 66.3% yield); Rf )
Tetr a (eth ylen e glycol) Mon o[3-(1-m eth yl-1,2,5,6-tet-
r ah ydr opyr id-3-yl)-1,2,5-th iadiazol-4-yl]eth er Acetate Hy-
d r och lor id e (26) (CDD-0303-A). Reduction of 17b (100 mg,
0.20 mmol) yielded the free base 21b (40 mg, as a yellow oil,
1
0.69 [CHCl3/CH3OH (90:10)]. H NMR (CDCl3): δ 1.27-1.57
(m, 18H), 1.87 (m, 2H), 1.95 (m, 2H), 2.45 (s, 3H, -NCH3),
2.57 (t, 2H), 3.45 (m, 2H), 3.60 (t, 2H), 4.44 (t, 2H), 7.07 (s,

4284 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
53.6% yield); Rf ) 0.31 [CHCl3/CH3OH (90:10)]. The above free
base was treated with acetic anhydride (4 mL) and ZnCl2
powder (10 mg) following the similar procedure as described
in the synthesis of compound 25 to afford the title compound
26 (35.5 mg, as a yellow-white powder, 36.3% yield); mp 55-
55.5 °C (hygroscopic); free base Rf ) 0.48 [CHCl3/CH3OH (90:
10)]. 1H NMR (CDCl3): δ 2.07 (s, 3H, acetyl), 2.58 (m, 1H),
2.95 (s, 3H, -NCH3), 2.96 (m, 1H), 3.20 (m, 1H), 3.56 (m, 1H),
3.66-3.71 (m, 12H), 3.80 (m, 1H), 3.90 (t, 2H), 4.48 (m, 1H),
4.62 (t, 2H), 7.29 (s, 1H, vinylic), 13.1 (s, 1H). 13C NMR
(CDCl3): δ 21.2, 22.9, 43.4, 50.0, 52.0, 63.7, 69.3, 70.3, 70.8,
70.9, 123.7, 127.4, 144.3, 162.4, 171.2. Anal. (C18H29N3O6S1‚
HCl‚0.5H2O) C, H, N.
Tetr a (eth ylen e glycol) (3-Meth oxy-1,2,5-th ia d ia zol-4-
yl)[3-(1-m et h yl-1,2,5,6-t et r a h yd r op yr id -3-yl)-1,2,5-t h ia -
d ia zol-4-yl]eth er Hyd r och lor id e (11) (CDD-0304-A). Re-
duction of the pyridinium iodide 18 (160 mg, 0.26 mmol)
yielded the hydrochlodride 11 (23 mg, as a hygroscopic pale
yellow powder, 16.1% yield); mp 35-37 °C; free base Rf ) 0.70
[CH2Cl2/CH3OH (95:5)]. 1H NMR (CDCl3): δ 2.62 (m, 2H), 2.95
(m, 5H), 3.23 (m, 2H), 3.55 (m, 2H), 3.69 (m, 6H), 3.89 (m,
4H), 4.10 (s, 3H), 4.56 (t, 2H), 4.63 (m, 2H), 7.30 (s, 1H). 13C
NMR (CDCl3): δ 22.9, 43.5, 50.1, 52.0, 57.7, 69.3, 70.0, 70.4,
70.9, 71.0, 123.7, 127.4, 144.3, 151.7, 152.6, 162.4. Anal.
(C19H29N5O6S2‚HCl‚H2O) C, H, N.
(Packard, Meriden, CT). Consistent results were obtained from
both systems for saturation binding and inhibition binding
assays.
The M1, M3, and M5 subtypes of muscarinic receptor couple
efficiently to phospholipase Câ via the Gq/11 family of G
proteins. Compounds were examined for their abilities to
activate G proteins by measuring accumulation of [3H]inositol
phosphates in stably transfected cells (A9 L cells and/or NIH
3T3 cells) upon agonist stimulation as described previously.11
Cells were transfected with plasmids containing the genes for
human M1, M3, or M5 receptors (provided by Dr. Tom Bonner
of the National Institutes of Health).22,23 Activity is presented
as the percentage activation above basal levels. Carbachol was
utilized in each assay as a positive control for muscarinic
receptor activation.
In contrast to the M1, M3, and M5 muscarinic receptor
subtypes, the M2 and M4 receptors signal preferentially
through inhibition of the cAMP pathway. Hence, ligands were
investigated for their abilities to inhibit forskolin-stimulated
cAMP accumulation in cells expressing either human M2 or
M4 receptors. The Assay Designs Correlate-EIA Direct cyclic
AMP kit was used to determine cAMP levels. The day before
the experiment, cells were treated with trypsin and seeded
into 24 well tissue culture plates with 500 µL of DMEM
(without serum) per well and incubated at 37 °C and 5% CO2
for 16-20 h to about 90% confluence. The media was removed,
and cells were treated with 500 µL of DMEM (without serum)
containing 300 µM IBMX, 20 µM forskolin, and appropriate
ligand concentrations (with DMEM serving as the control). The
reaction was terminated after 15 min incubation at 37 °C by
aspiration of media and addition of 200 µL 0.1 M HCl. After
10 min, the cells were inspected visually to verify cell lysis
and then transferred to Eppendorf tubes and centrifuged at
10 000 rpm at room temperature, and then, the supernatant
was used directly in the assay. A polyclonal antibody from the
Correlate-EIA Direct cyclic AMP kit was used to bind cAMP.
After a simultaneous incubation (2 h) at room temperature,
the excess reagents were washed away and substrate was
added. After a short incubation time (1 h), the enzyme reaction
was stopped and the yellow color generated was read on a
microplate reader at 405 nm. The measured optical density
was used to calculate the concentration of cAMP.
P h a r m a cologica l Scr een s. Observations of specific be-
haviors helped determine the incidence of side effects associ-
ated with injections of compound 11 and oxotremorine (for
comparison). Mice were injected (i.p.) with saline or 0.1 or 1.0
mg/kg of drug dissolved in saline and observed for 2 h
according to methods described previously.13,24 Three animals
received each dose. (A single animal was tested with 10 mg/
kg of compound 11.) Animals were observed over a 10 s
interval three times during each of three 30 min periods.
Relevant behaviors included salivation, lacrimation, tremors,
micturition, diarrhea, grooming, rearing and sniffing, respira-
tory rate (elevated or depressed), and locomotor behavior. Body
temperature also was recorded using a rectal probe. Particular
attention was paid toward behaviors associated with musca-
rinic receptor activity, including hypothermia, salivation,
lacrimation, tremors, and diarrhea.
Deter m in a tion of P r oton a tion Con sta n ts. The standard
potentiometric method (temperature, 25.0 ( 0.1 °C; ionic
strength, 0.15 M KCl; N2 atmosphere) and the GLpKa auto-
mated instrument (Sirius Anal. Ltd., U.K.) were used for the
determination of protonation constants of the samples. The
details of the experiments are the same as published previ-
ously.17 Five separate measurements were carried out, and the
average pKa values along with the standard deviations were
calculated.
P h a r m a cology. Mouse fibroblast (A9 L) cells and CHO
cells were purchased from American Type Culture Collection
(ATCC) (Rockville, MD). Dulbecco’s modified Eagle media
(DMEM), l-gGlutamine solution, penicillin/streptomycin solu-
tion, trypsin, geneticin (G-418 Sulfate Solid), 3-isobutyl-1-
metbylxanthine (IBMX), and all other inorganic chemicals
were purchased from Fisher Scientific Company L. L. C.
(Hanover Park, IL). Fetal bovine serum was purchased from
EQUITECH-BIO, Inc. (Kerrville, TX). Bovine serum albumin,
atropine-sulfate, carbamylcholine chloride (carbachol), trichlo-
roacetic acid, and Triton X-100 were purchased from Sigma-
Aldrich Co. (St. Louis, MO). Opti-MEM I reduced-serum media
and MEM nonessential amino acids solution were purchased
from Life Technologies, Inc. (Gaithersburg, MD). Forskolin and
UniverSol ES scintillation cocktail were bought from ICN
Biomedicals, Inc. (Irvine, CA). MicroScint 20 was purchased
from PerkinElmer Life Science (Boston, MA). Both myo-[2-3H]-
inositol (10-25Ci/mmol) and [3H]-(R)-QNB (30-60Ci/mmol)
were purchased from NEN Life Science Products, Inc. (Boston,
MA). The QIAGEN endofree plasmid Maxi kit was bought from
QIAGEN Inc. (Valencia, CA). LipofectAMINE 2000 reagent
was purchased from Invitrogen (Carlsbad, CA). The direct
Cyclic AMP Enzyme Immunoassay Kit was purchased from
Assay Designs, Inc. (Ann Arbor, MI).
If a behavior was observed during that 10 s period, a score
of 1 was given, while 0 was recorded if the behavior did not
occur. Scores were averaged for each half hour period and
again averaged for the group of three animals at each dose.
Thus, a score of 1.0 indicates that all three mice exhibited a
particular behavior each time it was observed during a 30 min
period. The studies provided an indication of the in vivo
selectivity and the time course of action for compound 11 as
compared with the classical muscarinic agonist oxotremorine.
Ex Vivo Bin d in g Assa ys. To assess the ability of ligands
to cross the blood-brain barrier following peripheral injections,
young, male Long-Evans rats were injected with either saline
or 0.1 or 1.0 mg/kg of drug in saline intraperitoneally (i.p.)
and sacrificed by cervical dislocation at 0.5 and 1.0 h after
injection. Brains were removed, and a crude homogenate was
prepared from whole brain to be used in ex vivo binding
Recep tor Assa ys. Plasmids for muscarinic receptor sub-
types were obtained as a gift from Dr. Tom I. Bonner of the
National Institutes of Health. Membrane homogenates were
prepared from transfected cells using established pro-
cedures.18-21 All binding assays were conducted in a 1 mL
mixture of binding buffer [25 mM sodium phosphate (pH 7.4)
containing 5 mM magnesium chloride]. In saturation binding
assays, eight concentrations of [3H]-(R)-QNB were used rang-
ing from 5 to 300 pM. In ligand inhibition binding assays, 0.1
nM [3H]-(R)-QNB and 14 points for the test ligand (ranging
from 0.01 nM to 3 mM) were used. The mixture was incubated
for 2 h at room temperature. Total binding and nonspecific
binding were determined in the absence and presence of 1000-
fold excess of unlabeled (R)-QNB. Radioactivity was counted
using either a 6895 BetaTrac liquid scintillation counter (TM
Analytic, Elk Grove Village, IL) or a TopCount NXT system

Muscarinic Agonists for Neurological Disorder Treatment
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20 4285
studies. The binding of 1 nM [3H]QNB was assayed using
methods outlined previously,13,25 including 1000-fold excess of
unlabeled atropine to measure nonspecific binding. The ability
for each ligand to penetrate the brain was assessed by
comparing the level of [3H]QNB binding between saline- and
drug-injected animals at each time point. Results were ex-
pressed as a percentage of specific [3H]QNB binding to brains
of control animals.
Da ta An a lysis. Nonlinear least squares curve fitting was
performed using DeltaGraph Pro (DeltaPoint Inc., version
4.0.1) for the Macintosh. Dose-response data were fit to a one
site stimulation model to obtain Smax (maximal levels of
stimulation), Imax (maximal levels of inhibition), and pEC50 or
pIC50 values. Saturation binding data were fit to a one site
binding model for Bmax and Kd values. Ligand inhibition
binding data were fit to one site, two site, and three site
binding models. Statistical comparisons between one site and
multiple site models were carried out using an F test with and
R set at the 0.05 level. Then, pKi values were converted from
pIC50 values according to the Cheng-Prusoff formula.26 Analy-
sis of variance with posthoc Tukey-Kramer tests were utilized
for comparing differences in receptor binding affinity and
receptor activation.
(4) Levy, M. L.; Cummings, J . L.; Kahn-Rose, R. Neuropsychiatric
symptoms and cholinergic therapy for Alzheimer’s disease.
Gerontology 1999, 45 (Suppl. 1), 15-22.
(5) Cummings, J . L. The role of cholinergic agents in the manage-
ment of behavioural disturbances in Alzheimer’s disease. Int.
J . Neuropsychopharmacol. 2000, 3, 21-29.
(6) Bodick, N. C.; Offen, W. W.; Shannon, H. E.; Satterwhite, J .;
Lucas, R.; van Lier, R.; Paul, S. M. The selective muscarinic
agonist xanomeline improves both the cognitive deficits and
behavioral symptoms of Alzheimer disease. Alzheimer Dis. Assoc.
Disord. 1997, 11, S16-S22.
(7) Bodick, N. C.; Offen, W. W.; Levey, A. I.; Cutler, N. R.; Gauthier,
S. G.; Satlin, A.; Shannon, M. E.; Tollefson, G. D.; Rasmussen,
K.; Bymaster, F. P.; Hurley, D. J .; Potter, W. Z.; Paul, S. M.
Effects of xanomeline, a selective muscarinic receptor agonist,
on cognitive function and behavioral symptoms in Alzheimer
disease. Arch. Neurol. 1997, 54, 465-473.
(8) Rasmussen, T.; Fink-J ensen, A.; Sauerberg, P.; Swedberg, M.
D.; Thomsen, C.; Sheardown, M. J .; J eppesen, L.; Calligaro, D.
O.; DeLapp, N. W.; Whitesitt, C.; Ward, J . S.; Shannon, H. E.;
Bymaster, F. P. The muscarinic receptor agonist BuTAC, a novel
potential antipsychotic, does not impair learning and memory
in mouse passive avoidance. Schizophr. Res. 2001, 49, 193-201.
(9) Friedman, J . I.; Temporini, H.; Davis, K. L. Pharmacologic
strategies for augmenting cognitive performance in schizophre-
nia. Biol. Psychiatry 1999, 45, 1-16.
Com p u ta tion a l Ch em istr y. Molecular modeling was per-
formed utilizing the Sybyl 6.6 software of Tripos.27 Extended
conformations were generated for 24b,c, 26, and 11, and
geometries were optimized using the AM1 semiempirical
quantum chemical method.28 Atomic charges were obtained
from the calculations and implemented in the Tripos force field
with a distance-dependent dielectric constant and a 20 Å
nonbonded cutoff. MD simulations (130 ps) were performed
for all four molecules in the gas phase. The time step was set
to 1 fs, and no SHAKE restriction was applied. Each molecule
started from an all-trans conformation along the O-CH2-‚‚
‚-CH2-O spacer with the torsional angle set to 90° for the
central CH2-CH2 bond. The molecules rapidly left this con-
formation at T ) 310 K and formed generally some sort of a V
shape structure, which fluctuated, however, between an almost
U and a largely opened V conformation. The last conformation
taken at 130 ps in the gas phase was accepted as the starting
structure for NpT MD simulations in aqueous solution.
Isothermal-isobaric simulations at T ) 310 K and p ) 1
atm mimic natural conditions: the temperature corresponds
to that of the human body, and the constant pressure applied
leads to reasonable density (near to 1 g/cm3) for the equili-
brated solution. A TIP3P water box was used with periodic
boundary conditions and with a nonbonded cutoff of 8 Å. A
chloride counterion was added to the system and set to 6-7 Å
apart from the proton of the tetrahydropyridyl ring in the
starting solution structure. Utilization of the SHAKE option
for single bonds allowed a time step of 2 fs. A 250 ps
equilibration phase was followed by another 250 ps simulation
period used for structure analysis. In calculations with three
NaCl molecules in the system, the isotonic saline concentration
(0.153 M) was mimicked with about 1300 TIP3P water
molecules in the water box. The simulations, using the same
parameters as described previously, were performed for 11
with a 250 ps simulation time for each of the equilibration
and averaging phases.
(10) Felder, C. C.; Porter, A. C.; Skillman, T. L.; Zhang, L.; Bymaster,
F. P.; Nathanson, N. M.; Hamilton, S. E.; Gomeza, J .; Wess, J .;
McKinzie, D. I. Elucidating the role of muscarinic receptors in
psychosis. Life Sci. 2001, 68, 2605-2613.
(11) Rajeswaran, W. G.; Cao, Y.; Huang, X. P.; Wroblewski, M. E.;
Colclough, T.; Lee, S.; Liu, F.; Nagy, P. I.; Ellis, J .; Levine, B.
A.; Nocka, K. H.; Messer, W. S., J r. Design, Synthesis, and
Biological Characterization of Bivalent 1-Methyl-1,2,5,6-tetrahy-
dropyridyl-1,2,5-thiadiazole Derivatives as Selective Muscarinic
Agonists. J . Med. Chem. 2001, 44, 4563-4576.
(12) Sauerberg, P.; Olesen, P. H.; Nielsen, S.; Treppendahl, S.;
Sheardown, M. J .; Honore, T.; Mitch, C. H.; Ward, J . S.; Pike,
A. J .; Bymaster, F. P.; et al. Novel functional M1 selective
muscarinic agonists. Synthesis and structure-activity relation-
ships of 3-(1,2,5-thiadiazolyl)-1,2,5,6-tetrahydro-1-methylpy-
ridines. J . Med. Chem. 1992, 35, 2274-2283.
(13) Messer, J . W. S.; Abuh, Y. F.; Ryan, K.; Shepherd, M. A.;
Schroeder, M.; Abunada, S.; Sehgal, R.; El-Assadi, A. A.; Sza´sz,
G.; Brann, M. R. Tetrahydropyrimidine derivatives display
functional selectivity for M1 muscarinic receptors in brain. Drug
Dev. Res. 1997, 40, 171-184.
(14) J akubik, J .; El-Fakahany, E. E.; Tucek, S. Evidence for a tandem
two-site model of ligand binding to muscarinic acetylcholine
receptors. J . Biol. Chem. 2000, 275, 18836-18844.
(15) CRC Handbook of Chemistry and Physics, 83rd ed.; CRC Press
LLC: Boca Raton, FL, 2002.
(16) Boudon, S.; Wipff, G. Conformational analysis of protonated
ethylenediamine in the gas phase and in water. J . Mol. Struct.
(THEOCHEM) 1991, 228, 61-70.
(17) Taka´cs-Nova´k, K.; Nosza´l, B.; Hermecz, I.; Keresztu´ri, G.;
Poda´nyi, B.; Abunada, S.; Sehgal, R.; El-Assadi, A. A.; Sza´sz,
G.; Brann, M. R. Protonation equilibria of quinolone antibacteri-
als. J . Pharm. Sci. 1990, 79, 1023-1028.
(18) Dorje, F.; Wess, J .; Lambrecht, G.; Tacke, R.; Mutschler, E.;
Abunada, S.; Sehgal, R.; El-Assadi, A. A.; Sza´sz, G.; Brann, M.
R. Antagonist binding profiles of five cloned human muscarinic
receptor subtypes. J . Pharmacol. Exp. Ther. 1991, 256, 727-
733.
(19) Huang, X. P.; Williams, F. E.; Peseckis, S. M.; Messer, W. S.,
J r. Pharmacological characterization of human m1 muscarinic
acetylcholine receptors with double mutations at the junction
of TM VI and the third extracellular domain. J . Pharmacol. Exp.
Ther. 1998, 286, 1129-1139.
(20) Huang, X. P.; Nagy, P. I.; Williams, F. E.; Peseckis, S. M.;
Messer, W. S., J r. Roles of threonine 192 and asparagine 382 in
agonist and antagonist interactions with M1 muscarinic recep-
tors. Br. J . Pharmacol. 1999, 126, 735-745.
Ack n ow led gm en t. The work was supported by
research grants NS31173 and NS35127 from the Na-
tional Institutes of Health. We thank Ms. Karen Pa-
padakis for her secretarial assistance in the preparation
of the manuscript.
(21) Huang, X. P.; Williams, F. E.; Peseckis, S. M.; Messer, W. S.,
J r. Differential modulation of agonist potency and receptor
coupling by mutations of Ser388Tyr and Thr389Pro at the
junction of transmembrane domain VI and the third extracel-
lular loop of human M1 muscarinic acetylcholine receptors. Mol.
Pharmacol. 1999, 56, 775-783.
(22) Bonner, T. I.; Buckley, N. J .; Young, A. C.; Brann, M. R.
Identification of a family of muscarinic acetylcholine receptor
genes. Science 1987, 237, 527-532.
Refer en ces
(1) Caulfield, M. P.; Birdsall, N. J . International Union of Pharma-
cology. XVII. Classification of muscarinic acetylcholine receptors.
Pharmacol. Rev. 1998, 50, 279-290.
(2) Abood, L. G.; Biel, J . H. Anticholinergic psychotomimetic agents.
Int. Rev. Neurobiol. 1962, 4, 217-273.
(3) Cummings, J . L.; Back, C. The cholinergic hypothesis of neu-
ropsychiatric symptoms in Alzheimer’s disease. Am. J . Geriatr.
Psychiatry 1998, 6, S64-S78.

4286 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 20
Cao et al.
(23) Bonner, T. I.; Young, A. C.; Brann, M. R.; Buckley, N. J . Cloning
and expression of the human and rat m5 muscarinic acetylcho-
line receptor genes. Neuron 1988, 1, 403-410.
(24) Messer, W. S., J r.; Bachmann, K. A.; Dockery, C.; El-Assadi, A.
A.; Hassoun, E.; Haupt, N.; Tang, B.; Li, X. Development of CDD-
0102 as a selective M1 agonist for the treatment of Alzheimer’s
disease. Drug Dev. Res. 2002, 57, 200-213.
(25) Bymaster, F. P.; Heath, I.; Hendrix, J . C.; Shannon, H. E.
Comparative behavioral and neurochemical activities of cholin-
ergic antagonists in rats. J . Pharmacol. Exp. Ther. 1993, 267,
16-24.
(26) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes
50% inhibition (I50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(27) Sybyl, 6.6 ed.; Tripos, Inc.: St. Louis, MO.
(28) Dewar, M. J . S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J . J . P.
AM1: A new general purpose quantum mechanical molecular
model. J . Am. Chem. Soc. 1985, 107, 3902-3909.
J M0301235