3-(1-Piperazinyl)-4,5-dihydro-1H-benzo[g]indazoles: High affinity ligands for the human dopamine D4 receptor with improved selectivity over ion channels

Article abstract of DOI:10.1016/S0968-0896(98)00028-5

3-(4-Piperidinyl)-5-arylpyrazoles, such as 1, were selective for the cloned human dopamine D₄ receptor (hD₄), but also showed affinity at voltage sensitive calcium, sodium and potassium ion channels. A combination of substituent changes to reduce the basicity of the piperidine nitrogen and conformational restriction to give 4,5-dihydro-1H-benzo[g]indazoles reduced this ion channel affinity at the expense of selectivity for hD₄ over other dopamine receptors. Incorporation of piperazine into the 4,5-dihydro-1H-benzo[g]indazoles in place of piperidine gave a novel series of high affinity, selective, orally bioavailable hD₄ ligands, such as 16, with improved selectivity over ion channels. Copyright (C) 1998 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(98)00028-5

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 743±753
3-(1-Piperazinyl)-4,5-dihydro-1H-benzo[g]indazoles: High
Anity Ligands for the Human Dopamine D4 Receptor with
Improved Selectivity over Ion Channels
Ian Collins,a,* Michael Rowley,a William B. Davey,a Frances Emms,a
Rosemarie Marwood,a Shil Patel,a Smita Patel,a Alan Fletcher,a Ian C. Ragan,a
Paul D. Leeson,a Ann L. Scottb and Theodore Brotenb
aMerck Sharp & Dohme Research Laboratories, Neuroscience Research Centre, Terlings Park, Eastwick Road,
Harlow, Essex, CM20 2QR, UK
bMerck Research Laboratories, West Point, PA., USA
Received 18 October 1997; accepted 5 December 1997
AbstractÐ3-(4-Piperidinyl)-5-arylpyrazoles, such as 1, were selective for the cloned human dopamine D4 receptor
(hD4), but also showed anity at voltage sensitive calcium, sodium and potassium ion channels. A combination of
substituent changes to reduce the basicity of the piperidine nitrogen and conformational restriction to give 4,5-dihydro-
1H-benzo[g]indazoles reduced this ion channel anity at the expense of selectivity for hD4 over other dopamine
receptors. Incorporation of piperazine into the 4,5-dihydro-1H-benzo[g]indazoles in place of piperidine gave a novel
series of high anity, selective, orally bioavailable hD4 ligands, such as 16, with improved selectivity over ion channels.
# 1998 Elsevier Science Ltd. All rights reserved.
Introduction
pharmacologically into two classes: D1-like (D1 and D5)
and D2-like (D2, D3 and D4).5±9 The atypical neurolep-
tic clozapine,10 which does not provoke the movement
disorders, shows, amongst many other actions, some
selectivity for the human D4 (hD4) subtype.8,11 Addi-
tionally, there have been reports that hD4 receptors are
located preferentially in the areas of the brain involved
with antipsychotic activity,12 and that hD4 receptor
density may be elevated in post-mortem schizophrenic
brain,13±15 although this latter ®nding is disputed.16,17
Schizophrenia is a serious mental illness characterised
by episodes of disordered thought, hallucinations, delu-
sions, social withdrawal and other estranging beha-
viours.1 Classical neuroleptic drugs for the treatment of
the disorder are presumed to act by the unselective
blockade of D2-like dopamine receptors.2 Although
these are e€ective for the treatment of positive symp-
toms, there is a great need for more selective treatments
that reduce or eliminate the often severe movement dis-
orders3 and hyperprolactinaemia4 associated with these
drugs, and that also show greater ecacy in the treat-
ment of the negative symptoms of the illness. Recent
advances in molecular biology have identi®ed ®ve
cloned human subtypes of dopamine receptor, divided
For these reasons, we18±20 and others21±25 have descri-
bed the development of novel, selective hD4 receptor
antagonists with the aim of determining the role of
the hD4 receptor in psychosis. In particular, we
recently reported the discovery of the pyrazolopiper-
idine (1) (Table 1) and its elaboration to the very
highly selective antagonist 2.18,19 Further characterisa-
tion of these ligands has revealed that, although they
are selective with respect to other central nervous system
G-protein coupled receptors, they exhibit moderately
high anities at voltage sensitive sodium, calcium and
Key words: 4,5-1H-Benzo[g]indazoles; selective dopamine
receptor ligands.
*Corresponding author: Tel: 01279440419; Fax: 01279440187;
E-mail: ian_collins@merck.com
0968-0896/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00028-5

744
I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
Table 1. Bindinga and physicochemical properties of the lead compounds
No.
1
2
16
Structure
hD2 Ki (nM)
hD3 Ki (nM)
hD4 Ki (nM)
644
>4000
>1700
770
>1500
>4100
5.8
3.5
0.82
4.3
Ca channel Ki (mM)
Na channel Ki (mM)
Ikr channel EC25 (mM)
pKa
1.4
0.56
0.40
>7.5c
4.2
58%@10 mM
33%@10 mM
>30
1.9
n.d.b
>7.4c
4.5
6.55
>4
log P (pH 7.4)
aSee footnotes to Table 2 for de®nition of binding parameters.
bn.d., not determined.
cPoor solubility prevented accurate measurement.
potassium ion channels (Table 1). This represents a
potential bar to the eventual use of these compounds as
investigational tools in the clinic. In this paper we
describe the evolution of ligands related to 1 to produce
the title compounds: high anity, selective hD4 ligands
with reduced ion channel activities and good oral bio-
availability.
alkylation of 3b gave the compounds 6 and 7. The
ketone (7) was reduced with sodium borohydride to give
the alcohol (8). 3-(4-Piperidinyl)-4,5-dihydro-1H-
benzo[g]indazoles (10 and 11; Scheme 2) were synthe-
sised by an extension of the previously described route.
Thus, the condensation of two equivalents of the lithium
enolate of a-tetralone with protected isonipecotic acid,
activated as the imidazolide, gave the diketopiperidine
(9) in good yield after N-deprotection. Alkylation of 9
and subsequent condensation with hydrazine a€orded
the target tricyclic compounds (10 and 11).
Chemistry
The pyrazolopiperidines (3a, 3b, and 4; Scheme 1) were
prepared by the general procedure described pre-
viously.19 Alkylation of 3a gave the analogue (5) and
The piperazine (12; Table 2) was prepared as described
previously.19 The analogous 3-(1-piperazinyl)-4,5-dihy-
dro-1H-benzo[g]indazoles (14, 15, and 16; Scheme 3)
were synthesised from a-tetralone, starting with a con-
densation with carbon disul®de under basic conditions,
followed by double S-methylation with methyl iodide.
The resulting dithioketeneacetal was re¯uxed with N-
Scheme 2. Reagents: (i) N,N-carbonyldiimidazole, solvent, rt;
(ii) a-tetralone, LDA, THF, 78 ꢀC; (iii) HCl, EtOAc, rt, 48%
(three steps); (iv) RBr, Pr2NEt, DMF, rt 70 ꢀC, 47±49%; (v)
i
Scheme 1. Reagents: (i) RBr, DMF, Pr2NEt, 60 ꢀC, 25±46%;
i
NH2NH2-H2O, MeOH, rt, 55±81%.
(ii) NaBH4, EtOH, rt, 43%.

I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
745
BOC-piperazine in acetonitrile to yield the mono-sub-
stitution product (13). The regiochemistry of 13 was
demonstrated by the observation of an NOE between
the S-methyl group and the proximal bridging methy-
lene. Further substitution of the remaining thiomethyl
group of 13 with hydrazine was accompanied by ring
closure to introduce the pyrazole ring. The syntheses of
14, 15, and 16 were completed by N-deprotection and
alkylation in poor to moderate yield. For the analogous
7-membered ring compound (18; Scheme 4) a variation
of this route was used, starting from 1-benzosuberone,
in which the N-substituted piperazine was introduced
complete in a single step to give 17, albeit in low yield.
Elaboration of 17 as before gave the target (18).
The N-methylated analogues (21 and 22; Scheme 5)
were prepared from 13 by condensation with N-methyl-
hydrazine to give a mixture of 19 and 20 (1.7:1), which
were separated by ¯ash column chromatography. The
regiochemistry of 19 was demonstrated by the observa-
tion of a reciprocal NOE between the pyrazole methyl
group and the neighbouring aromatic proton. The two
compounds were taken on separately as before to give
21 and 22.
Table 2. Variation of the substituent on the basic nitrogen
Kia (nM)
Inhibition (at 10 mM) ion channels
No.
Structure
hD2
125
hD3
hD4
Cab
Nac
4
160 9.1
80%
n.d.d
5
400
140
450
1500 40
1800 1.2
970 38
52%
77%
33%
77%
4%
100%
61%
n.d.
6e
7
8
660 >4400 4.4
10
11
41
14
908 0.40
71%
62%
n.d.
n.d.
55 4.5
aAnities at cloned human dopamine receptors stably expressed in cell lines. Results are the mean of two or more determinations.
bInhibition of speci®c binding of [3H] diltiazem to rabbit skeletal muscle at 10 mM concentration of test compound. Results are the
mean of two or more determinations.
cInhibition of speci®c binding of [3H] batrachotoxin to rat cortex at 10 mM concentration of test compound. Results are the mean of
two or more determinations.
dn.d., not determined.
eAnities are the result of a single determination only.

746
I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
were expressed as the percentage inhibition of radiolabel
binding at a ®xed test compound concentration of
10 mM and were the mean of at least two determina-
tions. Binding to voltage sensitive potassium channels
(mainly IKR) was estimated in vitro by the measurement
of prolongation of e€ective refractory period (ERP) in
ferret papilliary muscle.30
E€orts to introduce a larger window between the hD4
anity and ion channel activities began by varying the
substitution, and consequently the basicity, of the
piperidine nitrogen of 1 (Table 2). The introduction of
N-benzyl in 4 as a replacement for the N-phenethyl
group of 1, previously found to be optimal for selectiv-
ity at hD4 receptors over hD2 and hD3,19 produced the
expected decrease in selectivity for hD4 over the other
dopamine subtypes whilst little change in calcium chan-
nel activity was observed. Substitution by the more
electron withdrawing 3-nitrobenzyl group did produce a
compound (5) with signi®cantly decreased activity at
both calcium and sodium ion channels. However, this
change also reduced anity at the hD4 receptor by
sevenfold, suggesting that modulation of the basicity of
the nitrogen atom alone would not be sucient to
achieve the desired selectivity over the ion channels.
Scheme 3. Reagents: (i) NaH, CS2, MeI, THF, rt, 59%; (ii) N-
tert-butoxycarbonylpiperazine, MeCN, re¯ux, 49%; (iii)
NH2NH2±H2O, EtOH, rt, 80%; (iv) CF3CO2H, rt; (v) RBr,
iPr2NEt, DMF, rt±60 ꢀC, 7±51% (two steps).
Earlier studies had shown that a 4-methyl substituent on
the central pyrazole ring of ligands related to 1 was
bene®cial for hD4 anity,19 possibly as a result of
introducing a conformational bias in the relative orien-
tation of the aromatic heterocycle and the two sub-
stituent rings. Also, removal of the 4-chloro substituent
from the pendant 5-phenyl ring did not reduce anity
for the hD4 receptor.19 Thus the 4-methylpyrazole (6)
had ®vefold improved anity at hD4 and the same
selectivity as 1 with regard to hD2. Again, gradual
reduction of the basicity of the piperidine nitrogen, this
time by substitution on the ethylene linker to give the
alcohol (8) and ketone (7), produced a progressive
decline in hD4 anities. The ketone (7) had a similar
pro®le to the nitrobenzyl analogue (5) in terms of
dopamine receptor anities and calcium ion channel
activity. However, the di€erence in sodium channel
activities between 5 and 7 indicated that the ion chan-
nels had divergent structure±activity relationships
(SAR) in this series of compounds and, furthermore,
that these SAR were not wholly dominated by the pKa
of the piperidine.
Scheme 4. Reagents: (i) NaH, CS2, MeI, THF, rt, 65%; (ii)
phenethylpiperazine tri¯uoroacetate, iPr2NEt, MeCN, re¯ux,
13%; (iii) NH2NH2±H2O, EtOH, rt, 71%.
Results and Discussion
The compounds prepared were tested in vitro for their
ability to displace [3H]-spiperone from cloned human
dopamine receptors (hD2 stably expressed in CHO
cells26 and hD3 and hD4 in HEK293 cells27). Each inhi-
bition constant (Ki) was the mean of at least two
experimental determinations: errors in the mean were
within twofold of the mean. Binding to voltage sensitive
ion channels was evaluated in vitro in rat cerebral cortex
for the sodium (Na) channel28 and in rabbit skeletal
muscle for the calcium (Ca) channel.29 Values obtained
The 4-methyl substituent on the pyrazole (6) was exten-
ded to create an ethylene bridge to the 5-aryl ring, pro-
viding compound (10), with the intention of restricting
the rotation of the aromatic ring whilst providing a
similar conformational bias with regard to the piper-
idine as the original 4-methyl group. A signi®cant
increase in anity at all dopamine subtypes was

I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
747
Scheme 5. Reagents: (i) MeNHNH2, EtOH, temp, 30%, (19)/(20) 1.7/1; (ii) CF3CO2H, rt; (iii) PhCH2CH2Br, iPr2NEt, DMF, 60 ꢀC,
37±41% (two steps).
observed on making these changes, leading to sub-
nanomolar anity at hD4 for 10 with reduced selectivity
over hD2. There was also a marginal reduction in cal-
cium ion channel activity. The increase in anity asso-
ciated with the planar tricyclic 4,5-dihydro-1H-
benzo[g]indazole system meant the analogue (11)
retained good hD4 activity when N-benzyl was sub-
stituted for N-phenethyl. Although this combination of
changes also reduced calcium channel activity, this was
at the expense of the selectivity for hD4 over the other
dopamine receptors.
also
inactive
at
potassium
channels
(IKR
EC25>30 mM). Regrettably, despite an otherwise pro-
mising pro®le, compound 14 remained active at the
sodium channel.
Replacement of the N-phenethyl group of 14 by N-ben-
zyl to yield 15 was not tolerated either in terms of
anity or selectivity at dopamine receptors. Several
analogues of 14 bearing simple substituents on the aro-
matic ring of the N-phenethyl group were preparedÐfor
example; 4-Cl, 2-NO2 and 2-, 3-, and 4-OMeÐbut were
generally signi®cantly less active than the parent (14) at
dopamine receptors (hD4 Ki>20 nM). An exception was
the 2-chloro substituted derivative (16) which showed
slightly improved hD4 binding. More importantly, 16
had a much improved sodium ion channel pro®le (33%
inhibition at 10 mM) whilst remaining inactive at potas-
sium channels (EC25>30 mM). Although the activity of
16 at calcium channels was more than 14, this still
represented a satisfactory improvement on the lead
piperidines 1 and 2.
An alternative approach to ®nding more selective com-
pounds was based on piperazine analogues of the
piperidine lead 1 (Table 2). The piperazine (12; pKa 7.2),
which was somewhat less basic than 1 (pKa>7.5), had
lower hD4 anity than 1, but represented a better lead
pro®le than the modi®ed piperidines such as 5 and 7. In
particular, good selectivity over hD2 and hD3 subtypes
was maintained and a signi®cant reduction in calcium
and sodium ion channel activity was associated with
only a modest drop in hD4 anity. The piperazine (12)
also had improved selectivity over the potassium chan-
nels (IKR EC25 5.8 mM) compared to the piperidine (1;
Table 1).
Comparison of 6, 10, 14, and 16 suggested that although
a large component of the binding at ion channels was
due to the basicity of the heterocycle in these series,
there were more subtle e€ects in the structure±activity
relationships arising from aromatic substitution on the
N-substituent in the case of the sodium channel, and
also more generally from the local conformation around
the heteroaromatic ring. The strong in¯uence of the
latter property was demonstrated by the ring expanded
analogue (18), where the longer propylene tether was
expected to twist the pyrazole and 5-phenyl substituent
out of coplanarity. This led to a large increase in the
calcium channel activity compared to the more planar
14. Similarly, methylation of the pyrazole nitrogen to
give 21 also enhanced calcium channel activity. In this
case, the buttressing e€ect of the methyl group and the
Replacement of the phenylpyrazole moiety of 12 with
the 4,5-dihydro-1H-benzo[g]indazole structure, which
had led to increased hD4 anity in the piperidine series,
now gave the high anity compound (14) which
retained the high selectivity over hD2 and hD3 expected
from the N-phenethyl substitution. The small decrease
in calcium channel activity associated with the change
from phenylpyrazole to 4,5-dihydro-1H-benzo[g]ind-
azole, previously observed in the comparison of 6 with
10, was apparently additive to that associated with the
change from piperidine to piperazine. Thus 14 had neg-
ligible activity at the calcium ion channel, and was now

748
I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
5-phenyl substituent may disrupt the planarity of the
tricycle relative to the parent (14). This hypothesis was
supported by the regioisomeric N-methyl pyrazole (22),
where the conformation of the tricycle would again be
more planar as the peri interaction is removed, which
showed an ion channel pro®le more like that of 14. The
high anity of both the N-methyl pyrazoles (21 and 22)
suggested that neither the pyrazole N-H nor the lone
pair on the pyrazole nitrogen were engaged in a direct
interaction with the receptor. This was consistent with
earlier observations that the central pyrazole ring in
ligands related to 1 could be replaced successfully by a
variety of heterocycles and acyclic linkers.19 In view of
the promising dopamine receptor and ion channel pro-
®le of 16 (Table 1), the pharmacokinetic properties of
this compound were investigated in rats. The compound
(16) was orally bioavailable (31%) with a plasma half-
life of 1 h.
Table 3. Replacing the piperidine ring with piperazine
Kia (nM)
hD3
Inhibition (at 10 mM) ion channels
No.
Structure
hD2
hD4
Cab
Nac
12
>1900
1200 23
32%
43%
14
15
16
18
21
22
>1800
2000 5.3
17%
14%
58%
72%
78%
44%
100%
n.d.d
33%
n.d.
270
810 22
>1500 >4100 4.3
>1800
700 3.9
400 7.9
1200 1.9
480
n.d.
250
76%
aAnities at cloned human dopamine receptors stably expressed in cell lines. Results are the mean of two or more determinations.
bInhibition of speci®c binding of [3H] diltiazem to rabbit skeletal muscle at 10 mM concentration of test compound. Results are the
mean of two or more determinations.
cInhibition of speci®c binding of [3H] batrachotoxin to rat cortex at at 10 mM concentration of test compound. Results are the mean of
two or more determinations.
dn.d., not determined.

I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
749
Conclusions
was evaluated by displacement of [3H]-batrachotoxin
(30±60 Ci mmol 1, NEN, USA) binding to rat cerebral
cortex. Binding to the voltage sensitive potassium
channels (particularly IKR channels) was estimated by
measurement of the prolongation of e€ective refractory
period (ERP) in ferret papilliary muscle.
A signi®cant improvement in the selectivity for hD4
receptors over ion channels in the series of hetero-
cyclylpyrazoles derived from 1 was achieved by a com-
bination of three approaches. A general reduction in the
basicity of the heterocycle on changing from piperidine
to piperazine reduced anity at the ion channels, but
also at dopamine receptors. The hD4 anity was
restored by conformational restriction of the pendant 5-
phenyl group to produce the more planar, tricyclic 4,5-
dihydro-1H-benzo[g]indazoles. Generally lower ion
channel activities were also associated with this more
planar structure, although the three ion channel types
studied did not have identical structure±activity rela-
tionships. Appropriate aromatic substitution in the N-
phenethyl group attached to the piperazine further
improved selectivity over ion channels and allowed the
identi®cation of the high anity, selective, orally bio-
available hD4 ligand (16).
Chemistry
Melting points were determined on a Reichert Thermo-
var apparatus and are uncorrected. Proton NMR were
measured on Bruker AM 360 or AC 250 spectrometers
and chemical shifts are reported in parts per million (d)
down®eld from tetramethylsilane as internal standard,
coupling constants are in Hertz. Chemical ionisation
mass spectra were recorded on a VG70/250 spectro-
meter. Merck Kieselgel (230±400) mesh was used for
column chromatography. For reactions, anhydrous sol-
vents were used as purchased from Aldrich or Fluka.
Organic solutions were dried with anhydrous magne-
sium sulfate or anhydrous sodium sulfate. Elemental
analyses were performed by Butterworth Laboratories
Ltd, Teddington, Middlesex, UK.
Experimental
Biochemical methods
Compounds 5, 6, and 7 were prepared from the inter-
mediates (3a) and (3b) according to the published pro-
cedure19 using the appropriate alkyl halide:
[3H]-Spiperone binding studies.26,27 Clonal cell lines
expressing the human dopamine D2, D3 and D4 receptor
subtypes were harvested in PBS (phosphate bu€ered
saline) and then lysed in 10 mM Tris±HCl pH 7.4 bu€er
containing 5 mM MgSO4 for 20 min on ice. Membranes
were centrifuged at 50,000 g for 15 min at 4 ꢀC and the
resulting pellets resuspended in assay bu€er (50 mM
Tris±HCl) pH 7.4 containing 5 mM EDTA, 1.5 mM
CaCl2, 5 mM MgCl2, 5 mM KCl, 120 mM NaCl and
0.1% ascorbic acid) at 20 mg wet weight/mL (human D4
HEK cells), 10 mg wet weight/mL human D2 CHO cells
and D3 HEK cells). Incubations were carried out for
120 min at ambient temperature (22 ꢀC) in the presence
of 0.2 nM [3H]-spiperone for displacement studies and
were initiated by the addition of 20±100 mg protein in a
®nal assay volume of 0.5 mL. The incubation was ter-
minated by rapid ®ltration over GF/B ®lters presoaked
in 0.3% PEI (polyethylenimine) and washed with ice-
cold 50 mM Tris±HCl, pH 7.4. Speci®c binding was
determined by 10 mM apomorphine and radioactivity
determined by counting in a LKB beta counter. Binding
parameters were determined by nonlinear least squares
regression analysis, from which the inhibition constant
Ki could be calculated for each test compound.
3-(1-(3-Nitrobenzyl)-4-piperidinyl)-5-(4-chlorophenyl)pyr-
azole (5). Pale-yellow needles, mp 174±176 ꢀC (from
EtOH); 1H NMR (360 MHz, DMSO-d6) d 1.69 (2 H, dq,
J=2.8 and 12), 1.90 (2 H, br d, J=12), 2.13 (2 H, dt,
J=2.1 and 10), 2.60±2.80 (1 H, m), 2.88 (2 H, br d,
J=10), 3.64 (2 H, s), 6.53 (1 H, s), 7.43 (2 H, d, J=8),
7.64 (1 H, t, J=8), 7.77±7.80 (3 H, m), 8.13 (1 H, d,
J=8), 8.18 (1 H, s), 12.69 (1 H, s); MS m/z 397
(M++H). Found: C, 63.85; H, 5.31; N, 14.10.
C21H21N4O2Cl requires C, 63.55; H, 5.33; N, 14.12.
3-(1-(2-Phenylethyl)-4-piperidinyl)-4-methyl-5-phenylpyr-
azole (6). White crystals, mp 154±157 ꢀC (from EtOAc);
1H NMR (360 MHz, DMSO-d6) d 1.76±1.82 (4 H, m),
2.00±2.15 (5 H, m), 2.53±2.57 (2 H, m), 2.60±2.70 (1 H,
m), 2.77 (2 H, t, J=7), 3.05 (2 H, d, J=11), 7.18±7.57
(10 H, m), 12.47 (1 H, br s); MS m/z 346 (M++H).
Found: C, 79.97; H, 7.78; N, 12.40. C23H27N3 requires
C, 79.96; H, 7.88; N, 12.16.
3-(1-(2-Oxo-2-phenylethyl)-4-piperidinyl)-4-methyl-5-
phenylpyrazole (7). Pale-yellow crystals, mp 176±178 ꢀC
1
Ion channel activities.28±30 Activity at the voltage sensi-
tive calcium channel (diltiazem allosteric site) was eval-
uated by displacement of [3H]-diltiazem (60±87
Ci mmol 1, NEN, USA) binding to rabbit sketetal
muscle. Binding to the voltage sensitive sodium channel
(from EtOAc); H NMR (360 MHz, DMSO-d6) d 1.72±
1.84 (4 H, m), 2.10 (3 H, s), 2.26 (2 H, t, J=11), 2.58±
2.68 (1 H, m), 3.01 (2 H, d, J=11), 3.86 (2 H, s), 7.32
(1 H, t, J=7), 7.42 (2 H, t, J=7), 7.51±7.58 (4 H, m),
7.64 (1 H, t, J=7), 8.02 (2 H, d, J=8), 12.48 (1 H, s);

750
I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
MS m/z 360 (M++H). Found: C, 76.85; H, 6.91; N,
11.76. C23H25N3O requires C, 76.85; H, 7.01; N, 11.69.
a gradient of 1% to 5% v/v methanol in dichloro-
methane containing 1% v/v triethylamine to give 2-(4-
(1-(2-phenylethyl)) piperidinocarbonyl)-3,4-dihydro-1-
hydroxynaphthalene (0.64 g, 47%) as o€-white needles,
mp 85±86 ꢀC (from EtOH); 1H NMR (360 MHz,
DMSO-d6) d 1.60±1.80 (4 H, m), 2.04 (2 H, t, J=11),
2.50±2.60 (2 H, m), 2.62 (2 H, t, J=7), 2.75 (2 H, t,
J=7), 2.85 (2 H, t, J=7), 2.90±3.10 (3 H, m), 7.10±7.50
(8 H, m), 7.81 (1 H, d, J=7), 15.2 (1 H, br s); MS m/z
362 (M++H). Found: C, 79.93; H, 7.47; N, 4.04.
C24H27NO2 requires C, 79.74; H, 7.53; N, 3.87.
3-(1-(2-Hydroxy-2-phenylethyl)-4-piperidinyl)-4-methyl-
5-phenylpyrazole (8). A mixture of the ketone (7)
(0.081 g, 0.225 mmol) and sodium borohydride (0.017 g,
0.50 mmol) in ethanol (2 mL) was stirred at rt under
nitrogen for 4 h. The solids were ®ltered o€, washed
with warm ethanol, and the ®ltrate was concentrated.
The resulting white powder was dissolved in ethyl acet-
ate and washed with water. The organic layer was dried
and concentrated, and the residual white powder was
recrystallised from ethanol to give 8 (0.035 g, 43%), mp
2-(4-(1-(2-Phenylethyl))piperidinocarbonyl)-3,4-dihydro-
1-hydroxynaphthalene (0.158 g, 0.44 mmol) was stirred
with hydrazine hydrate (0.050 g, 1.6 mmol) in methanol
(3 mL) under nitrogen for 17 h. The mixture was con-
centrated and the residue was dissolved in ethyl acetate,
washed with water and brine, then dried, concentrated
and recrystallised from ethanol to give 10 (0.086 g, 55%)
as white needles, mp 151±152 ꢀC (from EtOH); 1H
NMR (360 MHz, DMSO-d6) d 1.70±1.80 (4 H, m), 2.00±
2.10 (2 H, m), 2.50±2.60 (2 H, m), 2.60±2.70 (3 H, m),
2.75 (2 H, t, J=7), 2.85 (2 H, t, J=7), 3.02 (2 H, d,
J=11), 7.10±7.30 (8 H, m), 7.60 (1 H, br s), 12.30 (1 H,
br s); MS m/z 358 (M++H). Found: C, 78.77; H, 7.53;
173±175 ꢀC (from EtOH); H NMR (360 MHz, DMSO-
1
d6) d 1.74±1.83 (4 H, m), 2.10±2.22 (5 H, m), 2.40±2.64
(3 H, m), 3.04 (2 H, t, J=11), 4.71±4.75 (1 H, m), 4.95
(1 H, s), 7.23 (1 H, t, J=7), 7.29±7.57 (9 H, m), 12.45
(1 H, s); MS m/z 362 (M++H). Found: 76.07; H, 7.43;
N, 11.31. C23H27N3O requires C, 76.42; H, 7.53; N, 11.62.
3-(1-(2-Phenylethyl)-4-piperidinyl)-4,5-dihydro-1H-benzo-
[g]indazole (10). a-Tetralone (6.4 g, 44 mmol) in THF
(15 mL) was added over 5 min to a solution of lithium
diisopropylamide (45 mmol) in THF (150 mL) under
nitrogen at 78 ꢀC, then stirred for 45 min to give an
orange solution. Meanwhile, carbonyldiimidazole
(3.89 g, 24 mmol) was added to a solution of N-tert-
butyloxycarbonyl isonipecotic acid (5 g, 22 mmol) in
THF (100 mL) at rt under nitrogen. The mixture was
stirred for 45 min, then cannulated into the above
orange solution at 78 ꢀC. After a further 40 min the
mixture was warmed to rt, diluted with ethyl acetate,
and washed sequentially with saturated sodium hydro-
gencarbonate solution, water and brine. The organic
layer was dried and concentrated to give a yellow oil
(13.1 g). The oil was dissolved in ethyl acetate (15 mL),
then a saturated solution of HCl in ethyl acetate (40 mL)
was added. When e€ervescence had stopped, the mix-
ture was heated to boiling then cooled. The solid was
collected, washed with ethyl acetate and dried to give 2-
(4-piperidinocarbonyl)-3,4-dihydro-1-hydroxynaphtha-
lene hydrochloride (9) (3.12 g, 48%) as an o€-white
.
N, 11.39. C24H27N3 0.5(H2O) requires C, 78.65; H, 7.70;
N, 11.46.
3-(1-Benzyl-4-piperidinyl)-4,5-dihydro-1H-benzo[g]indazole
(11). This was prepared in the same way as 10 from the
intermediate (9) using benzyl bromide as the alkylating
agent (49%). The ®nal product was puri®ed by pre-
parative thin layer chromatography, eluting with 3%
methanol and 1% v/v triethylamine in dichloromethane,
to give 11 as a white foam (0.081 g, 16%); 1H NMR
(360 MHz, DMSO-d6) d 1.70±1.80 (4 H, m), 2.00±2.10 (2
H, m), 2.70±2.80 (3 H, m), 2.90±3.00 (4 H, m), 3.49 (2
H, s), 7.00±7.70 (9 H, m), 12.40 (1 H, br s); MS m/z 344
(M++H). Found: C, 77.97; H, 7.33; N, 11.78.
.
C23H25N3 0.6(H2O) requires C, 77.98; H, 7.45; N, 11.86.
3-(4-(2-Phenylethyl)piperazin-1-yl)-4,5-dihydro-1H-benzo-
[g]indazole (14). Sodium hydride (60% in oil, 35 g,
880 mmol) was added with care to a solution of a-tetra-
lone (54 g, 370 mmol), carbon disul®de (27 mL,
443 mmol) and methyl iodide (52 mL, 810 mmol) in
THF (400 mL) at 0 ꢀC. The mixture was stirred at rt
overnight, giving a yellow solution with a white precipi-
tate. Saturated aqueous ammonium chloride solution
and ethyl acetate were added and the mixture was sepa-
rated. The organic layer was washed with water and
brine, dried and concentrated. The resulting solid was
recrystallised from ethyl acetate-hexanes to give 2-(bis-
methylthiomethylene)-4,5-dihydro-2H-naphthalen-1-one
(67 g, 59%) as yellow cubes, mp 54±56 ꢀC; 1H NMR
(360 MHz, CDCl3) d 2.43 (6 H, br s), 2.98 (2 H, t, J=7),
amorphous solid, mp 257±259 ꢀC. H NMR (360 MHz,
1
DMSO-d6) d 1.80±2.00 (4 H, m), 2.67 (2 H, t, J=8), 2.86
(2 H, t, J=8), 2.90±3.00 (2 H, m), 3.10±3.20 (1 H, m),
3.31 (2 H, d, J=12), 7.30±7.40 (2 H, m), 7.49 (1 H, t,
J=8), 7.81 (1 H, d, J=8).
A suspension of 9 (1.12 g, 3.8 mmol), diisopropylethyl-
amine (1.49 mL; 8.6 mmol) and 2-phenylethyl bromide
(0.63 mL, 4.6 mmol) in DMF (10 mL) was stirred at rt
for 3 days, then at 70 ꢀC for 17 h. Water (50 mL) was
added and the mixture was extracted with ethyl acetate.
The combined organic extracts were washed with water
and brine, then dried and concentrated. The crude pro-
duct was puri®ed by ¯ash chromatography, eluting with

I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
751
3.26 (2 H, t, J=7), 7.22 (1 H, d, J=8), 7.32 (1 H, t,
J=8), 7.43 (1 H, dt, J=1 and 8), 8.10 (1 H, dd, J=1
and 8).
(0.098 mg, 30% over two steps) as white needles, mp
216±219 ꢀC (from DMF-EtOH); 1H NMR (360 MHz,
DMSO-d6) d 2.66 (2 H, t, J=8), 2.87 (2 H, t, J=8),
2.88 (2 H, t, J=7), 3.10±3.20 (6 H, m), 3.2±3.40 (4 H,
m), 7.10±7.40 (8 H, m), 7.55 (1 H, d, J=7); MS
m/z 359 (M++H). Found: C, 66.55; H, 6.32; N,
2-(Bis-methylthiomethylene)-4,5-dihydro-2H-naphthalen-
1-one (11.56 g, 46 mmol) and 1-tert-butyloxycarbonyl-
piperazine (10.3 g, 55 mmol) were re¯uxed in acetonitrile
(300 mL) for 24 h. The mixture was cooled, water
(500 mL) was added, and the mixture was extracted with
ethyl acetate. The organic layer was washed with water
and brine, dried and concentrated. The resulting oil was
puri®ed by ¯ash chromatography, eluting with
dichloromethane then dichloromethane:methanol (97:3
v/v) to give 2-(methylthio[4-(tert-butyloxycarbonyl)-1-
piperazinyl]methylene)-4,5-dihydro-2H-naphthalen-1-
.
12.30. C23H24N4 C2H2O4 requires C, 66.94; H, 6.29; N,
12.49.
The following compounds were prepared in the same
way as 14 from the intermediate (13) using the appro-
priate alkyl halide:
3-(4-Benzyl-1-piperazinyl)-4,5-dihydro-1H-benzo[g]indazole
(15). Characterised as the oxalate salt, white micro-
crystalline solid, mp 254±256 ꢀC (from EtOH); 1H NMR
(360 MHz, DMSO-d6) d 2.63 (2 H, t, J=8), 2.86 (2 H, t,
J=8), 2.96 (4 H, br s), 3.26 (4 H, br s), 4.04 (2 H, s),
7.10±7.30 (3 H, m), 7.40±7.50 (5 H, m), 7.54 (1 H, d,
J=7); MS m/z 345 (M++H). Found: C, 65.71; H, 6.04;
1
one (13) (8.75 g, 49%) as a foam; H NMR (360 MHz,
CDCl3) d 1.49 (9 H, s), 3.31 (3 H, s), 2.86±2.94 (4 H, m),
3.3±3.5 (8 H, m), 7.18 (1 H, d, J=8), 7.30 (1 H, t, J=8),
7.37 (1 H, dt, J=1 and 8), 8.10 (1 H, dd, J=1 and 8).
Irradiation of the singlet at 3.31 ppm gave a positive
NOE to the signal at 2.86±2.94 ppm.
.
.
N, 12.40. C22H24N4 C2H2O4 0.25(H2O) requires C,
65.66; H, 6.08; N, 12.76.
The intermediate (13) (5.3 g, 14 mmol) and hydrazine
hydrate (3.4 g, 69 mmol) were stirred in ethanol
(100 mL) at rt for 16 h. The mixture was concentrated
and the resulting oil was puri®ed by ¯ash chromato-
graphy, eluting with dichloromethane:methanol (95:5 v/
v) to give 3-(4-(tert-butyloxycarbonyl)-1-piperazinyl)-
4,5-dihydro-1H-benzo[g]indazole (3.9 g, 80 %) as a yel-
low oil; H NMR (360 MHz, CDCl3) d 1.48 (9 H), 2.72
(2 H, t, J=8), 2.95 (2 H, t, J=8), 3.13 (4 H, t, J=5),
3.53 (4 H, t, J=5), 7.20±7.40 (4 H, m).
3-(4-(2-(2-Chlorophenyl)ethyl)-1-piperazinyl)-4,5-dihydro-
1H-benzo[g]indazole (16). Characterised as the oxalate
salt, pale-yellow plates, mp 134±136 ꢀC (from EtOH);
1H NMR (360 MHz, DMSO-d6) d 2.66 (2 H, t, J=8),
2.88 (2 H, t, J=8), 3.03±3.07 (8 H, m), 3.20±3.36 (4 H,
m), 7.19±7.35 (5 H, m), 7.41±7.47 (2 H, m) and 7.55 (1
H, d, J=7); MS m/z 393 (M++H). Found: C, 61.58; H,
1
.
.
5.61; N, 11.33. C23H25N4Cl (COOH)2 0.2(H2O) requires
C, 61.71; H, 5.68; N, 11.51.
3-(4-(tert-Butyloxycarbonylpiperazin-1-yl)-4,5-dihydro-
1H-benzo[g]indazole (1.65 g, 4.8 mmol) was dissolved in
tri¯uoroacetic acid (10 mL) at rt. After 30 min the sol-
vent was evaporated in vacuo to give 3-(1-piperazinyl)-
4,5-dihydro-1H-benzo[g]indazole bis-tri¯uoroacetate
(2.64 g), which still contained an excess amount of tri-
¯uoroacetic acid, as a light brown solid; (360 MHz,
DMSO-d6) d 2.66 (2 H, t, J=7), 2.95 (2 H, t, J=7), 3.22
(4 H, br s), 3.30 (4 H, br s), 7.20±7.30 (3 H, m), 7.56 (1
H, d, J=7), 8.8 (2 H, br s).
3-(4-(2-Phenylethyl)-1-piperazinyl)-1,4,5,6-tetrahydro-1,2-
diazabenzo[e]azulene (18). 1-Benzosuberone (10 g,
63 mmol) was treated with carbon disul®de (5.7 g,
75 mmol), methyl iodide (21.3 g, 150 mmol) and sodium
hydride (60% in oil, 6 g, 150 mmol) according to the
method described in the synthesis of 13 to give 6-(bis-
methylthiomethylene)-6,7,8,9-tetrahydrobenzocyclohep-
ten-5-one (10.7 g, 65%) as yellow cubes, mp 63±64 ꢀC
1
(from EtOAc-hexanes); H NMR (250 MHz, CDCl3) d
2.00 (2 H, quintet, J=7), 2.40 (6 H, s), 2.70±2.80 (4 H),
7.17 (1 H, d, J=8), 7.32 (1 H, t, J=8), 7.44 (1 H, t,
J=8), 7.88 (1 H, d, J=8).
This crude material (0.416 g, 1.1 mmol), diisopropyl-
ethylamine (0.59 mL, 3.3 mmol) and 2-phenethyl bro-
mide (0.168 mL, 1.23 mmol) were heated in DMF
(3 mL) at 60 ꢀC for 4 h. The mixture was cooled, diluted
with water (20 mL) and extracted with ethyl acetate. The
organic layer was washed with brine, dried and con-
centrated. The resulting light brown oil was dissolved in
ethanol (2 mL), heated to boiling, and oxalic acid
(1.3 mL of a 1M solution in ethanol) was added. After
cooling to room temperature the resulting solid was
collected, washed with ethanol, and recrystallised from
DMF:ethanol (1:9 v/v) to give the oxalate salt of 14
6-(Bis-methylthiomethylene)-6,7,8,9-tetrahydrobenzo-
cyclohepten-5-one (1.06 g, 4 mmol), phenethylpiperazine
tri¯uoroacetate (2.2 g, 5.3 mmol) and diisopropylethyl-
amine (2 mL, 11 mmol) were re¯uxed in acetonitrile
(10 mL) for 24 h. The mixture was cooled, diluted with
water, and extracted with ethyl acetate. The organic
layer was washed with water and brine, dried and con-
centrated. The crude product was puri®ed by ¯ash
chromatography, eluting with dichloromethane:metha-
nol (92:8 v/v) to give 6-(methylthio-(4-phenethylpiper-

752
I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
azin-1-yl)methylene)-6,7,8,9-tetrahydrohydrobenzocycl-
ohepten-5-one (17) (0.208 g, 13%) as a yellow oil; 1H
NMR (360 MHz, CDCl3) d 1.95 (2 H, quintet, J=7),
2.25 (3 H, s), 2.50±2.90 (16 H, m), 7.10 (1 H, d, J=8),
7.10±7.40 (7 H, m), 7.73 (1 H, d, J=8).
Compound 22: (41% over 2 steps from 20) characterised
as the oxalate salt, mp 244±245 ꢀC (from EtOH) 1H
NMR (360 MHz, DMSO-d6) d 2.80±2.90 (4 H, m), 2.90±
2.95 (2 H, m), 3.10±3.20 (6 H, m), 3.20±3.25 (4 H, m),
3.69 (3 H, s), 7.10±7.40 (8 H, m), 7.61 (1 H, d, J=8); MS
m/z 373 (M++H). Found: C, 66.94; H, 6.42; N, 11.71.
.
.
A solution of 17 (0.208 g, 0.52 mmol) and hydrazine
hydrate (0.5 mL, 10 mmol) in ethanol (3 mL) was stood
for 16 h at rt. Water was added and the mixture was
extracted with ethyl acetate. The organic layer was
washed with water and brine, dried, concentrated and
recrystallised to give 18 (0.132 g, 71%) as white needles,
mp 147±148 ꢀC (from EtOAc-hexanes); 1H NMR
(360 MHz, DMSO-d6) d 1.89 (2 H, quintet, J=5), 2.55±
2.7 (10 H, m), 2.76±2.82 (2 H, m), 3.00±3.05 (2 H, m),
3.25±3.35 (2 H, m), 7.10±7.25 (8 H, m), 7.69 (1 H, d,
J=7), 12.1 (1 H); MS m/z 373 (M++H). Found: C,
C24H28N4 C2H2O4 0.2(H2O) requires C, 66.99; H, 6.57;
N, 12.02.
Acknowledgements
We would like to thank S. Thomas for NMR experi-
ments and K. Locker for the measurement of
pharmacokinetic data on 16.
References
.
76.56; H, 7.62; N, 14.80. C24H28N4 0.2(H2O) requires C,
76.64; H, 7.61; N, 14.90.
1. Diagnostic and Statistical Manual of Mental Disorders IVth
Edition; First, M. B., Ed.; American Psychiatric Association:
Washington DC, 1994; pp 273±290.
3-(4-(2-Phenylethyl)-1-piperazinyl)-4,5-dihydro-1-methyl-
1H-benzo[g]indazole (21) and 3-(4-(2-phenylethyl)-1-piper-
azinyl)-4,5-dihydro-2-methyl-2H-benzo[g]indazole (22).
A solution of 13 (3.2 g, 8.2 mmol) and methylhydrazine
(5 mL, 94 mmol) in ethanol (30 ml) was stood at rt for 4
days. Water (150 mL) was added and the mixture was
extracted with ethyl acetate. The organic layer was
washed with brine, dried and concentrated. Puri®cation
by ¯ash column chromatography, eluting with hex-
ane:ethyl acetate (4:1 v/v) gave 3-(4-(tert-butyloxy-
carbonyl)piperazin-1-yl)-4,5-dihydro-1-methyl-1H-benzo-
[g]indazole (19) (0.342 g, 11%) as a colourless oil; 1H
NMR (360 MHz, CDCl3) d 1.52 (9 H, s), 2.67 (2 H, t,
J=7.3), 2.94 (2 H, t, J=7.3), 3.15 (4 H, t, J=5), 3.61 (4
H, t, J=5), 4.15 (3 H, s), 7.22±7.34 (3 H, m), 7.55 (1 H,
d, J=7.7). Irradiation at d 4.15 gave a positive NOE to
the doublet at d 7.55, and the reverse. Further elution
gave 3-(4-(tert-butyloxycarbonyl)piperazin-1-yl)-4,5-di-
2. Snyder, S. H. Am. J. Psych. 1981, 138, 461.
3. Baldessarini, R. J.; Tarsy, D. Annu. Rev. Neurosci. 1980, 3, 23.
4. Ben-Jonathon, N. Endocr. Rev. 1985, 6, 564.
5. Dearry, J. R.; Gingrich, J. A.; Falardeau, R. T.; Fremeau,
R. T.; Bates, M. D.; Caron, M. Nature (London) 1990, 347, 72.
6. Grandy, D. K.; Marchionni, M. A.; Makam, H.; Stofko, R.
E.; Alfano, M.; Frothingham, L.; Fischer, J. B.; Burk-Howie,
K. J.; Bunzow, J. R.; Sewer, A. C.; Civelli, O. Proc. Natl. Acad.
Sci. U. S. A. 1989, 86, 9762.
7. Sokolo€, P.; Giros, B.; Martres, M.-P.; Bouthenet, M.-L.;
Schwartz, J.-C. Nature (London) 1990, 347, 146.
8. Van Tol, H. H. M.; Bunzow, J. R.; Guan, H.-C.; Sunahara,
R. K.; Seeman, P.; Niznik, H. B.; Civelli, O. Nature (London)
1991, 350, 610.
9. Sunahara, R. K.; Laurier, L. G.; Ng, G.; George, S. R.;
Torchia, J.; Van Tol, H. H. M.; Niznik, H. B. Nature (London)
1991, 350, 614.
10. Fitton, A.; Heel, R. Drugs 1990, 40, 722.
hydro-2-methyl-2H-benzo[g]indazole
(20)
(0.561 g,
11. Seeman, P.; Corbett, R.; Van Tol, H. H. M. Neuropsycho-
pharmacology 1997, 16, 93.
19%) as a white solid; 1H NMR (360 MHz, CDCl3) d
1.54 (9 H, s), 2.86 (2 H, t, J=8), 2.95 (2 H, t, J=8), 3.08
(4 H, t, J=5), 3.60 (4 H, t, J=5), 7.20±7.30 (3 H, m),
7.83 (1 H, d, J=8).
12. Matsumoto, M.; Hidaka, K.; Tada, S.; Tasaki, Y.; Yama-
guchi, T. Mol. Brain. Res. 1995, 29, 157.
13. Seeman, P.; Guan, H.-C.; Van Tol, H. H. M. Nature
(London) 1993, 365, 441.
Compounds 19 and 20 were taken on separately
according to the procedure described in the synthesis of
14 to give:
14. Murray, A. M.; Hyde, T. M.; Knable, M. B.; Herman, M.
M.; Bigelow, L. B.; Carter, J. M.; Weinberger, D. R.; Klein-
man, J. E. J. Neurosci. 1995, 15, 2186.
15. Sumiyoshi, T.; Stockmeier, C. A.; Overholser, J. C.;
Thompson, P. A.; Meltzer, H. Y. Brain Res. 1995, 681, 109.
Compound 21: (37% over 2 steps from 19) characterised
as the oxalate salt, mp 225±226 ꢀC (from EtOH) 1H
NMR (360 MHz, DMSO-d6) d 2.56 (2 H, t, J=8), 2.83
(2 H, t, J=8), 2.95 (2 H, t, J=8), 3.10±3.20 (6 H, m),
3.25±3.35 (4 H, m), 3.95 (3 H, s), 7.20±7.40 (8 H, m),
7.63 (1 H, d, J=7); MS m/z 373 (M++H). Found: C,
16. Reynolds, G. P.; Mason, S. L. J. Neurochem. 1994, 63,
1576.
17. Mulcrone, J.; Kerwin, R. W. Neurosci. Lett. 1996, 219,
163.
18. Rowley, M.; Broughton, H. B.; Collins, I.; Baker, R.;
Emms, F.; Marwood, R.; Patel, S.; Patel, S.; Ragan, C. I.;
Freedman, S.; Leeson, P. D. J. Med. Chem. 1996, 39, 1943.
.
.
67.04; H, 6.66; N, 11.92. C24H28N4 C2H2O4 0.2(H2O)
requires C, 66.99; H, 6.57; N, 12.01.

I. Collins et al./Bioorg. Med. Chem. 6 (1998) 743±753
753
19. Rowley, M.; Collins, I.; Broughton, H. B.; Davey, W. D.;
Baker, R.; Emms, F.; Marwood, R.; Patel, S.; Patel, S.; Ragan,
C. I.; Freedman, S. B; Leeson, P. D. J. Med. Chem. 1997, 40,
2374.
24. Kula, N. S.; Baldessarini, R. J.; Kebabian, J. W.; Baktha-
vachalam, V.; Xu, L. Eur. J. Pharmacol. 1997, 331, 333.
25. Wright, J. L.; Gregory, T. F.; He€ner, T. G.; MacKenzie,
R. G.; Pugsley, T. A.; Vander Meulen, S.; Wise, L. D. Bioorg.
Med. Chem. Lett., 1997, 7, 1377.
20. Kulagowski J. K.; Broughton, H. B.; Curtis, N. R.; Mawer,
I. M.; Ridgill, M. P.; Baker, R.; Emms, F.; Freedman, S. B.;
Marwood, R.; Patel, S.; Patel, S.; Ragan, C. I.; Leeson, P. D. J.
Med. Chem. 1996, 39, 1941.
26. Freedman, S. B.; Patel, S.; Marwood, R.; Emms, F.; Seab-
rook, G. R.; Knowles, M. R.; McAllister, G. J. Pharmacol.
Exp. Ther. 1994, 268, 417.
27. McAllister, G.; Knowles, M. R.; Wrad-Booth, S. M.; Sin-
clair, H. A.; Patel, S. Marwood, R.; Emms, F.; Patel, S.; Smith,
G. R.; Seabrook, G. R.; Freedman, S. B. J. Recept. Sig. Trans.
Res. 1995, 15, 267.
21. Boy®eld, I.; Brown, T. H.; Coldwell, M. C.; Cooper, D. G.;
Hadley, M. S.; Hagan, J. J.; Healy, M. A.; Johns, A.; King, R.
J.; Middlemiss, D. N.; Nash, D. J.; Riley, G. J.; Scott, E. E.;
Smith, S. A.; Stemp, G. J. Med. Chem. 1996, 39, 1946±48.
28. Catterall, W. A.; Morrow, C. S.; Daly, J. W.; Brown, G. B.
J. Biol. Chem. 1981, 256, 8922.
22. Boy®eld, I.; Coldwell, M. C.; Hadley, M. S.; Healy, M. A.;
Johns, A.; Nash, D. J.; Riley, G. J.; Scott, E. E.; Smith, S. A.;
Stemp, G.; Wilson, K. Bioorg. Med. Chem. Lett. 1996, 6, 1227.
29. Reynolds, I.; Snowman, A. M.; Snyder, S. H. J. Pharma-
col. Exp. Ther. 1986, 237, 731.
23. TenBrink, R. E.; Bergh, C. L.; Duncan, J. N.; Harris, D.
W.; Hu€, R. M.; Lahti, R. A; Lawson, C. F.; Lutzke, B. S.;
Martin, I. J.; Rees, S. A.; Schleachter, S. K.; Sih, J. C.; Smith,
M. W. J. Med. Chem. 1996, 39, 2435.
30. Baskin, E. P.; Serik, C. M.; Wallace, A. A.; Brookes, L.
M.; Selnik, H. G.; Claremon, D. A.; Lynch, J. J. J. Cardiovasc.
Pharmacol. 1991, 18, 406.