Muscarinic Agonists with Antipsychotic Activity
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 22 4383
and 4b had intrinsic activity at M2 and M4 receptor
2.00 (1H, m), 1.84-1.74 (3H, m); mp 144-146 °C. Anal.
2
0
(C H N O S ) C, H, N.
subtypes. Further, the compounds had equally high
affinity for all five muscarinic receptor subtypes using
12 17
3
4
2
exo-6-(3-E t h ylt h io-1,2,5-t h ia d ia zol-4-yl)-1-a za b icyclo-
1
3
H]NMS binding.20
[3.2.1]octa n e (8) oxa la te: H NMR (DMSO) δ 10-8.0 (2H,
[
br s), 3.99-3.77 (2H, m), 3.67-3.59 (1H, m), 3.85-3.14 (6H,
m), 2.63-2.57 (1H, m), 2.20-1.99 (1H, m), 1.90-1.75 (3H, m),
Con clu sion s
1
.91 (3H, t); mp 159-160 °C. Anal. (C13
19 3 4 2
H N O S ) C, H, N.
We have shown that standard muscarinic agonists
and muscarinic agonists of the alkylthio-1,2,5-thiadi-
azole azacyclic type both are active in behavioral models
predictive of antipsychotic activity. SAR investigations
revealed the propyl/butylthio side chain to be the
optimal size for functional antipsychotic-like activity in
the 1,2,5-thiadiazole azacyclic series. The azabicyclo-
exo-6-(3-P en tylth io-1,2,5-th ia d ia zol-4-yl)-1-a za bicyclo-
1
[3.2.1]octa n e (10) oxa la te: H NMR (DMSO) δ 7.5-6.0 (2H,
br s), 3.95-3.78 (2H, m), 3.68-3.49 (1H, m), 3.34-3.13 (6H,
m), 2.63-2.56 (1H, m), 2.19-1.98 (1H, m), 1.90-1.67 (5H, m),
.45-1.25 (4H, m), 0.88 (3H, t); mp 117-118 °C. Anal.
16 25 3 4 2
C H N O S ) C, H, N.
1
(
exo-6-(3-Hexylth io-1,2,5-th ia d ia zol-4-yl)-1-a za bicyclo-
1
[
3.2.1]octa n e (11) oxa la te: H NMR (DMSO) δ 7.7-6.3 (2H,
[3.2.1]octane ring was identified to give compounds with
br s), 3.95-3.86 (2H, m), 3.68-3.49 (1H, m), 3.36-3.13 (6H,
m), 2.65-2.56 (1H, m), 2.20-1.98 (1H, m), 1.90-1.65 (5H, m),
the most potent antipsychotic-like activity without
having the undesired cholinergic side effects of saliva-
tion and tremor. These compounds, 4a ,b, and 9a ,b, were
1
.50-1.20 (6H, m), 0.85 (3H, t); mp 118-119 °C. Anal.
) C, H, N.
Recep tor Bin d in g Stu d ies. Binding to rat brain homo-
17 27 3 4 2
(C H N O S
1
0 times more potent than the typical antipsychotic
agent haloperidol. The lack of enantiomeric selectivity
of the exo-6-(3-propyl/butylthio-1,2,5-thiadiazol-4-yl)-1-
azabicyclo[3.2.1]octane enantiomers suggested that the
pharmacophores lie in the mirror plane of the com-
pounds.
The described muscarinic agonists lacking the tradi-
tional cholinergic side effects have the potential of
becoming a novel treatment for schizophrenia.
3
genates using [ H]Oxo-M as ligand was performed using
standard conditions as described previously.13
Con d ition ed Avoid a n ce Resp on d in g. Male rats (Fisher-
derived F344, Harlan Sprague-Dawley, Indianapolis, IN)
weighing 250-300 g were trained to avoid or escape foot shock
in an operant conditioning chamber. Briefly, at the start of
each trial, the houselight and a tone were presented. A
response within 10 s immediately terminated the trial (avoid-
ance response). If the rat did not respond within 10 s, foot
shock (2 mA) was presented. A reponse during the shock
immediately terminated the trial (escape response). If the rat
did not respond within 10 s after the onset of shock, the trial
terminated automatically (response failure). Compounds were
initially screened at a dose of 3.0 mg/kg sc, and the percent
avoidance response (%AR) and the percent escape failures
(%AF) were recorded. Active compounds were tested sc for full
dose response. Sessions typically ended after 50 trials. Each
point represents the mean ( SEM of percent of responses for
six rats. The ED50 values were calculated as the dose producing
50% of the maximal effect produced by the drug.
Exp er im en ta l Section
Melting points were determined on a Buchi capillary melt-
1
ing point apparatus and are uncorrected. H NMR spectra were
recorded at 200 MHz on a Bruker AC-200 MHz FT-NMR
instrument, and mass spectra were recorded on a Finnigan
5
100 mass spectrometer. Column chromatography was per-
formed on silica gel 60 (70-230 mesh, ASTM; Merck). Ele-
mental analyses were performed by Novo Nordisk Microana-
lytical Laboratory, Denmark, and were within (0.4% of the
calculated values.
Experimental data on preparation and physical character-
ization have previously been published on the following
Ap om or p h in e Clim bin g. Groups of n ) 10 male Bom:
NMRI mice were injected with saline or a dose of drug sc, 50
min prior to the trial. Apomorphine (2 mg/kg, sc) was injected
1
3
14
21
15
compounds: 1; 2, 5, 6; 4, 4a ; 9a ,b, 12a ,b.
The com-
pounds below were made in the same manner:
2
0 min prior to the trial. Mice were placed individually in
en d o-6-(3-Bu tylth io-1,2,5-th ia d ia zol-4-yl)-1-a za bicyclo-
cylindrical wire cages, and the time each mouse climbed on
the wire cage was automatically recorded. Data are expressed
as the mean ( SEM of number of seconds the mice were
climbing during the 60-s trial. ED50 values and 95% confidence
limits were calculated using logistic nonlinear regression
equations (Graph Pad Prism, Graph Pad Software Inc., San
Diego, CA). Compounds were screened at 3-4 doses giving
relatively large 95% confidence limits. Side effects were scored
1
[
3.2.1]octa n e (3) oxa la te: H NMR (DMSO) δ 12.0-9.0 (2H,
br s), 4.1-3.85 (3H, m), 3.56-3.46 (1H, m), 3.43-3.24 (5H,
m), 3.12-3.05 (1H, s), 2.04-1.84 (1H, m), 1.77-1.63 (3H, m),
1
.60-1.37 (3H, m), 1.05-0.88 (4H, m); mp 123-124 °C. Anal.
(C
15
H
23
N
3
O
4
S
2
) C, H, N.
(
5S,6R)-6-(3-Bu tylth io-1,2,5-th iadiazol-4-yl)-1-azabicyclo-
1
[
3.2.1]octa n e (3a ) oxa la te: H NMR (DMSO) δ 10.5-8.5 (2H,
br s), 4.08-3.84 (3H, m), 3.56-3.47 (1H, m), 3.43-3.23 (5H,
m), 3.12-3.04 (1H, br s), 2.04-1.83 (1H, m), 1.77-1.63 (3H,
m), 1.58-1.35 (3H, m), 1.05-0.86 (4H, m); mp 198-199 °C.
3
0 min after injection of saline or a dose of drug, using three
grades, and the data are expressed as the percent of the
maximum possible score.14 ED50 values were calculated by
graphic interpolation of the dose-response curve.
Anal. (C15
H
23
N
3
O
4
S
2
‚0.4H
2
O) C, H, N.
(
5R,6S)-6-(3-Bu tylth io-1,2,5-th iadiazol-4-yl)-1-azabicyclo-
1
Molecu la r Mod elin g. The geometry of all compounds were
optimized using the AM1 Hamiltonian in SPARTAN (Wave-
function Inc., Irvine, CA 92715) and subsequently analyzed
using SYBYL 6.4 (R4000, TRIPOS Inc., St. Louis, MO 63144).
Torsional scans were performed from 0 to 360 in 10 intervals
using the MMFF force field in MacroModel (W. Clark Still,
Department of Chemistry, Columbia University, New York,
NY 10027). At each point in the scan, all coordinates were
refined while the dihedral angle being investigated was
constrained. An additional harmonic potential with a force
[
(
(
3.2.1]octa n e (3b) h yd r och lor id e: H NMR (DMSO) δ 10.94
1H, br s), 4.10-3.82 (3H, m), 3.55-3.46 (1H, m), 3.43-3.23
5H, m), 3.11-3.05 (1H, m), 2.02-1.83 (3H, m), 1.75-1.60 (3H,
m), 1.57-1.36 (3H, m), 1.04-0.88 (4H, m); mp 198-199 °C.
Anal. (C13 ) C, H, N.
5R,6R)-6-(3-Bu t ylt h io-1,2,5-t h ia d ia zol-4-yl)-1-a za b i-
H
3 2
22ClN S
(
1
cyclo[3.2.1]octa n e (4b) ta r tr a te: H NMR (DMSO) δ 7.5-
6
.0 (4H, br s), 3.84-3.66 (2H, m), 3.63-3.56 (1H, m), 3.30 (2H,
t), 3.22-3.05 (4H, m), 2.57-2.51 (1H, m), 2.13-1.94 (1H, m),
1
1
.85-1.65 (5H, m), 1.50-1.36 (2H, m), 0.90 (3H, t); mp 140-
42 °C. Anal. (C17 ) C, H, N.
exo-6-(3-Meth ylth io-1,2,5-th ia d ia zol-4-yl)-1-a za bicyclo-
2
H
27
N
3
O
6
S
2
constant of 1000 kJ /mol × (deg) was added to maintain the
dihedral angle during the force field minimization. The SCF
energy convergence criterion for the Spartan calculations was
set to 1.0D-6. Default parameters were used for geometry
optimization.
1
[
3.2.1]octa n e (7) oxa la te: H NMR (DMSO) δ 5.0-3.0 (2H,
br s), 3.96-3.80 (2H, m), 3.67-3.61 (1H, m), 3.31-3.22 (3H,
m), 3.21-3.13 (1H, m), 2.74 (3H, s), 2.65-2.59 (1H, m), 2.18-