1-[2-[(Heteroarylmethoxy)aryl]carbamoyl]indolines are selective and orally active 5-HT(2C) receptor inverse agonists

Article abstract of DOI:10.1016/S0960-894X(00)00365-6

Bisarylmethoxyethers have been identified with nanomolar 5-HT(2C) affinity and selectivity over both 5-HT(2A) 5-HT(2A) and 5-HT(2B) receptors. Compounds such as 1, 2, 8, 12, 14 and 18 have potent oral activity in a centrally mediated pharmacodynamic model

Full text of DOI:10.1016/S0960-894X(00)00365-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1867±1870
1-[2-[(Heteroarylmethoxy)aryl]carbamoyl]indolines are Selective
and Orally Active 5-HT2C Receptor Inverse Agonists
Steven M. Bromidge,* Susannah Davies, D. Malcolm Duckworth, Ian T. Forbes,
Graham E. Jones, Jerome Jones, Frank D. King, Thomas P. Blackburn, Vicky Holland,
Guy A. Kennett, Sean Lightowler, Derek N. Middlemiss, Graham J. Riley,
Brenda Trail and Martyn D. Wood
SmithKline Beecham Pharmaceuticals, Discovery Research, New Frontiers Science Park, Third Avenue,
Harlow, Essex CM19 5AW, UK
Received 7 April 2000; revised 5 June 2000; accepted 21 June 2000
AbstractÐBisarylmethoxyethers have been identi®ed with nanomolar 5-HT2C anity and selectivity over both 5-HT2A and 5-HT2B
receptors. Compounds such as 1, 2, 8, 12, 14 and 18 have potent oral activity in a centrally mediated pharmacodynamic model of 5-
HT2C function and their therapeutic potential is currently under further investigation. # 2000 Elsevier Science Ltd. All rights
reserved.
In our previous letter we described bisaryl ethers which
are high anity and selective 5-HT2C receptor inverse
agonists with excellent in vivo activity.1 Such com-
pounds are of considerable potential for the treatment
of a range of CNS disorders, in particular anxiety and
depression.2 We now describe the discovery and SAR of
a novel series of 1-[2-[(2-heteroarylylmethoxy)aryl]-
carbamoyl]indolines such as 1 (SB-247853) and 2 which
are also high anity and selective 5-HT2C receptor inverse
agonists with potent oral activity in animal models.
5-nitropyridines, or 4-¯uoronitrobenzene in the case of
14, to a€ord the nitrobisarylmethoxyethers in generally
excellent yield (Scheme 1). Reduction to the corresponding
amines and coupling with the appropriate substituted
indolines,3 via the phenyl carbamate, a€orded the ®nal
compounds.4 The hydroxymethylheteroaromatics and 2-
chloro-5-nitropyridines were commercially available or
known in the literature. Compound 17 was similarly pre-
pared from 5-nitro-2-[2-(2-pyridinyl)ethoxy]pyridine,5
itself prepared from 2-hydroxy-5-nitropyridine and 2-(2-
hydroxyethyl)pyridine under Mitsunobu conditions.
The pyridazine 13 was prepared by reacting 6-chloro-3-
pyridazinecarboxylic acid ethyl ester with 2-(hydroxy-
methyl)pyridine under basic conditions to give the
bisarylmethoxyether. The ester was then converted to
the azide which was heated to generate the isocyanate
and reacted in situ with 5-methyl-6-tri¯uoromethyl-
indoline to a€ord 13 (Scheme 2). In an analogous pro-
cedure (Scheme 3), 16 was prepared by heating 2-hydro-
xypyridine and 6-(chloromethyl)-3-pyridinecarboxylic
acid ethyl ester with silver carbonate in toluene in the dark
to a€ord the bisarylmethoxyether which was then con-
verted to 16 as in Scheme 2. The ethyl linked compound 15
was prepared by condensing 2-pyridinecarboxaldehyde
with ethyl 2-methyl-5-pyridinecarboxylate in acetic anhy-
dride followed by hydrogenation of the resulting ethylene
linker (Scheme 4). The resulting ethyl linked bispyridyl
ester was then converted to 15 as described in Scheme 2.
Chemistry
The ®nal compounds were prepared according to Schemes
1±4. 1±12, 14 and 18±21 were obtained by treating the
anion of the appropriately substituted hydroxymethyl-
heteroaromatics with appropriately substituted 2-chloro-
*Corresponding author. Tel.: +44-1279-627684; fax: +44-1279-
627685; e-mail: steve_bromidge-1@sbphrd.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00365-6

1868
S. M. Bromidge et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1867±1870
Scheme 1. Reagents and conditions: (i) NaH, DMF, 20 ꢀC±rt, 18 h (31±91%); (ii) SnCl2, EtOH/conc. HCl, 50 ꢀC, 1 h (49±100%); (iii) PhOCOCl/
NEt3, CH2Cl2, 20 ꢀC, 1 h; (iv) NEt3/DMF, 100 ꢀC, 1 h (35±85%).
ꢀ
Scheme 2. Reagents and conditions: (i) POCl3, 100 ꢀC, 2 h (74%); (ii) 2-pyridylmethanol/KOtBu, THF, 70 C±rt, 18 h (22%); (iii) NH2NH2 H2O,
.
MeOH, re¯ux, 4 h (81%); (iv) NaNO2, aqueous HCl, 0 ꢀC, 0.5 h (100%); (v) 5-methyl-6-tri¯uoromethylindoline, toluene, re¯ux, 1 h (45%).
Scheme 3. Reagents and conditions: (i) SOCl2/pyridine, CH2Cl2, 0 ꢀC,
4 h (30%); (ii) 2-hydroxypyridine/Ag2CO3, toluene, re¯ux (dark), 18 h
Scheme 4. Reagents and conditions: (i) 2-pyridinecarboxaldehyde,
Ac2O, re¯ux, 18 h (12%); (ii) H2/ Pd/C, EtOH, rt, 4 h (66%); (iii)
.
.
(22%); (iii) NH2NH2 H2O, MeOH, re¯ux, 4 h (63%); (iv) NaNO2,
NH2NH2 H2O, MeOH, re¯ux, 4 h (83%); (iv) NaNO2, aqueous HCl,
aqueous HCl, 0 ꢀC, 0.5 h (53%); (v) 5-methyl-6-tri¯uoromethylindo-
line, toluene, re¯ux, 1 h (54%).
0 ꢀC, 0.5 h (79%); (v) 5-methyl-6-tri¯uoromethylindoline, toluene,
re¯ux, 1 h (59%).
model of 5-HT2C function, the ability of compounds to
block the hypolocomotion in rats produced by a stan-
dard dose of the moderately selective 5-HT2C agonist m-
chlorophenylpiperazine (mCPP) was measured.6 The
corresponding 5-tri¯uoro-6-methylindoline 1 showed an
even better pro®le with sub-nanomolar 5-HT2C anity,
further improved selectivity (1300-fold and 60-fold over 5-
HT2A and 5-HT2B, respectively) and potent in vivo activ-
ity (ID50 1 mg/kg p.o.). Interestingly, introduction of a 6-
methyl group 4 or a 3-methyl group 5 into the terminal
pyridine ring led to dramatic loss of selectivity and a 10-
fold drop in 5-HT2C anity, suggesting that the unsub-
stituted 2-pyridyl side-chain optimally ®lls the lipophilic
binding pocket. Introduction of an additional nitrogen
gave the pyrazine 8 with a similar binding pro®le to 1,
although selectivity over 5-HT2B and in vivo activity was
somewhat reduced. A number of heteroaromatic ®ve-
membered rings were also investigated such as the imi-
dazoles 9 and 10 and the thiazole 11. Although these
compounds had good 5-HT2C anity and reasonable
selectivity over 5-HT2A, selectivity over 5-HT2B was low
and 11 was inactive in vivo.
Results and Discussion
Our original series of bispyridyl ethers was developed
utilizing ligand docking studies into a model of the 5-
HT2C receptor.2 In the proposed binding mode the bis-
pyridyl ether moiety occupies a lipophilic pocket de®ned
by side-chain aromatic residues on transmembrane helices
5 and 6. In order to further explore the size and shape of
this binding pocket we prepared a range of analogues
incorporating the longer methoxy linker group between the
aromatic rings, initially in combination with the previously
optimized 5-chloro- or 5-tri¯uoro- 6-methylindoline.
Heteroarylmethoxyethers (Table 1)
Preparation of all three possible terminal pyridyl isomers
revealed that the position of the nitrogen atom was of
key importance. The 3-pyridylmethoxy analogue 6 had
relatively modest 5-HT2C receptor anity (reduced 25-
fold relative to the corresponding bispyridyl ether SB-
242084),1 moderate selectivity over 5-HT2A and no
selectivity over 5-HT2B receptors. The corresponding 4-
pyridylmethoxy isomer 7 showed 10-fold increased 5-
HT2C anity and somewhat improved, although still
unimpressive, selectivity. However, the 2-pyridylmethoxy
isomer 3 had similar target anity to 7 combined with
excellent selectivity and in addition this compound showed
signi®cant in vivo activity in the rat hypolocomotion
model. In this centrally mediated pharmacodynamic
Variation of central ring and linker (Table 2)
Having established that the 2-pyridyl isomer 1 was
optimal, a series of analogues was prepared investigating
modi®cations to the central ring and the linker group.
Considering the central ring, incorporation of a 5-methyl

S. M. Bromidge et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1867±1870
1869
Table 1. The 5-HT2A/B/C receptor binding anities,a selectivities over 5-HT2A and 5-HT2B and in vivo activitye of 1-[2-[2-(heteroarylmethoxy)-5-
pyridyl]carbamoyl]indolines 1 and 3±11
e
Compound
Ar
R
pKi
pKi (selectivity)
ID50
(mg/kg p.o.)
b
c
d
5-HT2C
5-HT2A
5-HT2B
3
1
4
5
6
7
8
9
2-Pyridyl
2-Pyridyl
6-Methyl-2-pyridyl
3-Methyl-2-pyridyl
3-Pyridyl
4-Pyridyl
2-Pyrazinyl
1-Methyl-2-imidazolyl
1-Methyl-5-imidazolyl
2-Thiazolyl
Cl
CF3
CF3
CF3
Cl
Cl
CF3
Cl
8.6
9.3
8.2
8.4
7.6
8.9
9.3
8.3
8.4
8.3
<6.1 (>320)
6.2 (1300)
6.0 (160)
6.8 (40)
<6.1 (>30)
7.1 (60)
6.5 (630)
6.6 (50)
7.0 (25)
7.0 (40)
7.5 (60)
7.6 (4)
7.8 (4)
7.6 (1)
8.1 (6)
8.0 (20)
7.9 (2.5)
8.2 (2)
41%
1.0
Ð
Ð
Ð
Ð
58%
Ð
10
11
CF3
Cl
Ð
2%
<6.1 (>160)
7.0 (20)
aAll values represent the mean of at least two determinations, with each determination lying within 0.2 log unit of the mean.
bBinding anity (human cloned receptors; HEK 293 cells; [3H]-mesulergine).2
cBinding anity (human cloned receptors; HEK 293 cells; [3H]-ketanserin).2
dBinding anity (human cloned receptors; HEK 293 cells; [3H]-5-HT).2
eDose of compound required to reverse mCPP (7 mg/kg ip administered 30 min pretest) induced hypolocomotion by 50% or percentage reversal at 5
mg/kg.2
Table 2. The 5-HT2A/B/C receptor binding anities,a selectivities over 5-HT2A and 5-HT2B and in vivo activitye of 1-[[(2-pyridyl)linked aryl]car-
bamoyl]indolines 1 and 12±17
e
pKi
pKi (selectivity)
ID50 (mg/kg p.o.)
Compound
A
b
c
d
5-HT2C
5-HT2A
5-HT2B
1
9.3
9.4
8.3
8.6
8.5
8.3
8.0
6.2 (1300)
7.4 (100)
6.6 (50)
7.5 (60)
8.1 (20)
7.1 (15)
7.6 (10)
7.7 (6)
1.0
0.3
Ð
12
13
14
15
16
6.4 (160)
6.8 (50)
84%
Ð
7.0 (20)
6.7 (40)
7.7 (2)
Ð
17
5.7 (200)
Ð
a
eSee corresponding footnotes in Table 1.
group (compound 12) retained anity and gave the
most potent in vivo activity yet achieved (ID50 0.3 mg/
kg) although selectivity was reduced relative to 1.
Replacement of the pyridine by pyridazine (compound
13) resulted in a 10-fold drop in anity and loss of
selectivity.
The corresponding phenyl derivative 14 also showed
reduced 5-HT2C receptor anity and selectivity, although
over 100-fold selectivity over 5-HT2A receptors and
reasonable oral activity was maintained. The analogues
incorporating an ethyl linker 15 and a reversed methoxy
linker group 16 had disappointing in vitro pro®les. The

1870
S. M. Bromidge et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1867±1870
Table 3. The 5-HT2A/B/C receptor binding anities,a selectivities over 5-HT2A and 5-HT2B and in vivo activitye of substituted 1-[[2-(2-pyr-
idylmethoxy)-5-pyridyl]carbamoyl]indolines 1±3 and 18±21
R4
R5
R6
pKi
5-HT2C
pKi (selectivity)
ID50
(mg/kg p.o.)
e
Compound
b
c
d
5-HT2A
5-HT2B
3
1
H
H
H
H
H
Cl
H
Me
Me
Br
CF3
Cl
Cl
CF3
H
H
H
8.6
9.3
7.8
8.2
7.5
7.1
8.1
<6.1 (>320)
6.2 (1300)
5.6 (160)
7.0 (40)
7.5 (60)
6.3 (30)
6.4 (60)
6.0 (30)
6.0 (13)
6.6 (30)
41%
1.0
2.4
1.6
Ð
18
2
19
20
21
5.3 (800)
<5.3 (>160)
5.3 (60)
6.6 (30)
H
H
H
CF3
Ð
Ð
a
eSee corresponding footnotes in Table 1.
three atom linked compound 17 showed a 10-fold
reduction in anity and selectivity relative to 1 suggest-
ing that a two atom linker is optimal.
reported elsewhere. Compound 1 (SB-247853) is currently
in pre-clinical development.
References and Notes
Optimization of indoline substitution (Table 3)
A limited programme of work to further investigate
indoline substitution in the 2-pyridylmethoxy series was
carried out. The 5-CF3 indoline 2 and the 5-bromo-
indoline 18 had reduced but still reasonable 5-HT2C
receptor anity and selectivity relative to 1 combined with
potent oral activity. In contrast, the 5-chloro 19, 4-chloro
20 and 6-tri¯uoromethyl 21 indolines had relatively mod-
est anity and selectivity.
1. Bromidge, S. M.; Dabbs, S.; Davies, S.; Duckworth, D. M.;
Forbes, I. T.; Jones, G. E.; Jones, J.; King, F. D.; Saunders, D.
V.; Blackburn, T. P.; Holland, V.; Kennett, G. A.; Lightowler,
S.; Middlemiss, D. N.; Riley, G. R.; Trail, B.; Wood, M. D.
Bioorg. Med. Chem. Lett. 2000, 10, 1863.
2. Bromidge, S. M.; Dabbs, S.; Davies, D. T.; Davies, S.; Duck-
worth, D. M.; Forbes, I. T.; Gaster, L. M.; Ham, P.; Jones, G. E.;
King, F. D.; Mulholland, K. R.; Saunders, D. V.; Wyman, P. A.;
Blaney, F. E.; Clarke, S. E.; Blackburn, T. P.; Holland, V.;
Kennett, G. A.; Lightowler, S.; Middlemiss, D. N.; Trail, B.;
Riley, G. J.; Wood, M. D. J. Med. Chem. 2000, 43, 1123.
3. Bromidge, S. M.; Dabbs, S.; Davies, D. T.; Duckworth, D.
M.; Forbes, I. T.; Ham, P.; Jones, G. E.; King, F. D.; Saun-
ders, D. V.; Starr, S.; Thewlis, K. M.; Wyman, P. A.; Blaney,
F. E.; Naylor, C. B.; Bailey, F.; Blackburn, T. P.; Holland, V.;
Kennett, G. A.; Riley, G. J.; Wood, M. D. J. Med. Chem.
1998, 41, 1598.
Summary
A number of bisarylmethoxyethers have been identi®ed
with nanomolar 5-HT2C anity and selectivity over both
5-HT2A and 5-HT2B receptors. On further cross-screen-
ing these compounds were also found to be selective over
a wide range of other monoamine receptors, including
serotinergic and dopaminergic subtypes. In an in vivo
human 5-HT2C receptor functional assay they were
found to be inverse agonists completely abolishing basal
activity.2 In addition 1, 2, 8, 12, 14 and 18 demonstrated
potent oral activity in the rat hypolocomotion assay.
Their activity in other animal models of anxiety will be
4. Bromidge, S. M. WO Patent 9748700, 1997; Chem. Abstr.
1998, 128, 75312.
5. Sohda, T.; Mizuno, K.; Imamiya, E.; Sugiyama, Y.; Fujita,
T.; Kawamatsu, Y. Chem. Pharm. Bull. 1982, 30, 3580.
6. Kennett, G. A.; Wood, M. D.; Bright, F.; Trail, B.; Riley,
G.; Holland, V.; Avenell, K. Y.; Stean, T.; Upton, N.; Bro-
midge, S. M.; Forbes, I. T.; Brown, A. M.; Middlemiss, D. N.;
Blackburn, T. P. Neuropharmacology 1997, 36, 609.