Synthesis, biological activity, and molecular modeling studies of selective 5-HT(2C/2B) receptor antagonists

Article abstract of DOI:10.1021/jm960571v

The synthesis and biological activity are reported for a series of analogues of the previously published indole urea 2 (SB-206553), designed to probe the 5-HT(2C) receptor binding site. Small molecule modeling studies have been used to define a region in space which is allowed at the 5-HT(2C) receptor but disallowed at the 5-HT(2A) receptor. In a complementary approach, docking of 2 into our model of the 5-HT(2C) receptor has allowed us to propose a novel primary binding interaction for this series of diaryl ureas, involving a potential double hydrogen-bonding interaction between the urea carbonyl oxygen of the ligand and two serine residues in the receptor. The difference of two valine residues in the 5-HT(2C) receptor for leucine residues in the 5-HT(2A) receptor is believed to account for the observed 5- HT(2C)/5-HT(2A) selectivity with 2.

Full text of DOI:10.1021/jm960571v

4966
J . Med. Chem. 1996, 39, 4966-4977
Syn th esis, Biologica l Activity, a n d Molecu la r Mod elin g Stu d ies of Selective
5-HT2C/2B Recep tor An ta gon ists
Ian T. Forbes,*,† Steven Dabbs,† D. Malcolm Duckworth,† Peter Ham,† Graham E. J ones,† Frank D. King,†
Damian V. Saunders,† Frank E. Blaney,† Christopher B. Naylor,† Gordon S. Baxter,‡ Thomas P. Blackburn,‡
Guy A. Kennett,‡ and Martyn D. Wood‡
SmithKline Beecham Pharmaceuticals, Discovery Research, New Frontiers Science Park, Third Avenue,
Harlow, Essex CM19 5AW, England
Received August 5, 1996X
The synthesis and biological activity are reported for a series of analogues of the previously
published indole urea 2 (SB-206553), designed to probe the 5-HT2C receptor binding site. Small
molecule modeling studies have been used to define a region in space which is allowed at the
5-HT2C receptor but disallowed at the 5-HT2A receptor. In a complementary approach, docking
of 2 into our model of the 5-HT2C receptor has allowed us to propose a novel primary binding
interaction for this series of diaryl ureas, involving a potential double hydrogen-bonding
interaction between the urea carbonyl oxygen of the ligand and two serine residues in the
receptor. The difference of two valine residues in the 5-HT2C receptor for leucine residues in
the 5-HT2A receptor is believed to account for the observed 5-HT2C/5-HT2A selectivity with 2.
5-Hydroxytryptamine (5-HT, serotonin) is an impor-
tant neurotransmitter in the central nervous system
(CNS).1 Dysfunction of the 5-HT system has been
implicated in various psychiatric and other CNS disor-
ders such as anxiety, depression, panic attacks, and
migraine, and these conditions can be treated by
pharmacological intervention with modulators of 5-HT
receptor function. The study of 5-HT receptors has been
revolutionized in recent years by molecular biological
techniques, with seven classes of receptors now identi-
fied (5-HT1-5-HT7) and numerous subclasses.2 The
5-HT2 family of receptors consists of three subtypes,
namely, 5-HT2A, 5-HT2B, and 5-HT2C, which have been
grouped together on the basis of molecular sequence,
secondary-messenger system, and operational profile.3
Sequence analysis indicates approximately 80% amino
acid identity in the transmembrane regions of these
three subtypes, and it is not surprising that many
compounds once thought to be selective for the 5-HT2A
(classical 5-HT2) receptor also bind with high affinity
to the 5-HT2B and 5-HT2C sites.
receptor antagonists might be useful anxiolytic agents.8
Indeed, compound 2 exhibits significant anxiolytic activ-
ity in three animal models of anxiety, which strengthens
our hypothesis that blockade of 5-HT2C/2B receptors
causes anxiolysis.9 Our previous work4 on modifications
to structure 1 established that the optimum position of
attachment of the urea group was at the 5-position of
the indole ring and at the 3-position of the pyridine. We
now report the synthesis and structure-activity rela-
tionships (SAR) of a number of analogues of 2 designed
to probe the size and shape of the receptor binding
pocket, along with our receptor modeling studies.
Ch em istr y
The indolyl ureas 2 and 3 were prepared as previously
reported.5 Compounds 4 and 5 were prepared as shown
in Scheme 1. 1-Acetyl-6-aminoindoline (6) was alky-
lated with bromoacetaldehyde diethyl acetal to give 7
which was then cyclized in refluxing TFAA/TFA5 to
afford a 4:1 mixture of 8 and 9. These compounds were
separated by chromatography, and each was subjected
to the sequence of hydrolysis, N-methylation, and hy-
drolysis, as detailed in Scheme 1, to give the tricyclic
intermediates 10 and 11. Conversion to the ureas 4 and
5 was achieved by treatment with 3-pyridyl isocyanate.
The indolyl urea 12 was prepared as shown in Scheme
2. 1-Methyl-6-nitroindole (13) was reacted with (4-
chlorophenoxy)acetonitrile10 to afford regioselectively
the 7-cyanomethyl derivative 14 which was converted
to the corresponding methyl ester 15. Treatment of 15
with LiAlH4 afforded the alcohol 16 which was con-
verted to its mesylate 17. Hydrogenation of 17 effected
reduction of the nitro group and concomitant cyclization
to the tricyclic 18. Treatment of 18 with 3-pyridyl
isocyanate provided the urea 12.
We have recently reported the syntheses of pyridyl-
urea 1 (SB-200646A)4 and the more potent conforma-
tionally restricted analogue 2 (SB-206553)5 as the first
5-HT2C/2B receptor antagonists with selectivity over the
closely related 5-HT2A site. Our interest in the 5-HT2C
receptor developed from literature reports that the
selective 5-HT2C/2B receptor agonist (m-chlorophenyl)-
piperazine causes anxiety both in animal models6 and
in humans.7 This has been attributed to stimulation
of 5-HT2C/2B receptors and implies that selective 5-HT2C/2B
N-Alkylated analogues of 2 were prepared by either
of two routes. In the first approach (Scheme 3), alky-
lation of the tricyclic intermediate 195 afforded the
N-alkyl derivatives 20 which were deprotected to give
the indolines 21. Reaction with 3-pyridyl isocyanate
afforded the desired ureas 22-24. Alternatively, the
N-acetyl intermediate 19 was hydrolyzed to the unsub-
stituted tricycle 25 (Scheme 4) which was reacted with
* To whom correspondence should be addressed.
† Department of Medicinal Chemistry.
‡ Department of Neurosciences.
X Abstract published in Advance ACS Abstracts, November 1, 1996.
S0022-2623(96)00571-7 CCC: $12.00 © 1996 American Chemical Society

Selective 5-HT2C/ 2B Receptor Antagonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4967
Sch em e 3. Synthesis of N-Alkylated Indoles 22-24a
Sch em e 1. Synthesis of N-Methylindoles 4 and 5a
a
Reagents: (a) NaH, RBr or RI, DMF, 1-18 h (79-99%); (b)
10% aqueous NaOH, EtOH, reflux, 8 h (68-98%); (c) 3-pyridyl
isocyanate, CH2Cl2-PhMe, 24 h (70-83%).
Sch em e 4. Synthesis of Indoles 26 and 27a
a
Reagents: (a) BrCH2CH(OEt)2, K2CO3, DMF, 100 °C, 18 h
(40%); (b) TFAA, TFA, reflux, 66 h; (c) (i) K2CO3, MeOH, 1 h, (ii)
NaH, MeI, DMF, 2 h, (iii) 10% aqueous NaOH, EtOH, reflux, 2 h;
(d) 3-pyridyl isocyanate, CH2Cl2-PhMe, 1 h.
Sch em e 2. Synthesis of N-Methylindole 12a
a
Reagents: (a) 10% aqueous NaOH, EtOH, reflux, 8 h (89%);
(b) 3-pyridyl isocyanate, CH2Cl2-PhMe, 24 h (80%); (c) NaH, iPrI,
DMF, 24 h (35%).
Sch em e 5. Synthesis of Indole 33a
a
Reagents: (a) 4-ClPhOCH2CN, KOtBu, DMF, 0 °C, 15 min
(78%); (b) (i) SOCl2, MeOH, reflux, 100 h, (ii) 10% aqueous NaOH,
EtOH, 3 h, (iii) SOCl2, MeOH, reflux, 2 h (42% overall); (c) LiAlH4,
THF, 10 min (47%); (d) MeSO2Cl, Et3N, CH2Cl2, 30 min (64%); (e)
EtOH, H2, 5% Pd-C, 2 h (85%); (f) 3-pyridyl isocyanate, CH2Cl2-
PhMe, 1 h (57%).
a
Reagents: (a) ClCH2C(Cl)dCH2, K2CO3, DMF, 70 °C, 16 h
(92%); (b) CH2O, NaBH4, THF, <20 °C, 15 min (100%); (c)
polyphosphoric acid, 140 °C, 24 h (12%); (d) 10% aqueous NaOH,
EtOH, reflux, 8 h (60%); (e) 3-pyridyl isocyanate, CH2Cl2-PhMe,
24 h (39%).
3-pyridyl isocyanate to give the urea 26. Regioselec-
tive alkylation at the 5-position was achieved by treat-
ment with excess sodium hydride and 2-iodopropane to
afford 27.
followed by intramolecular electrophilic cyclization af-
forded the linear tricyclic ring system 31, albeit in low
yield. Hydrolysis of the N-acetyl group in 31 to 32
followed by treatment with 3-pyridyl isocyanate afforded
the desired product 33.
Synthesis of the 5,6-dimethyl analogue 33 was
achieved as shown in Scheme 5. 1-Acetyl-5-aminoin-
doline (28) was alkylated with 2,3-dichloroprop-1-ene
to give 29, which was then reductively alkylated with
formaldehyde to 30. Thermal rearrangement11 of 30

4968 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Sch em e 6. Synthesis of Indole 39a
Forbes et al.
Sch em e 7. Synthesis of Benzothiophene 50a
a
Reagents: (a) TFAA, CH2Cl2, NEt3, 1 h (95%); (b) polyphos-
phoric acid, 140 °C, 1.5 h (7%); (c) K2CO3, MeOH, 30 min (91%);
(d) NaH, MeI, DMF, 24 h (79%); (e) 10% aqueous NaOH, EtOH,
reflux, 8 h (60%); (f) 3-pyridyl isocyanate, CH2Cl2-PhMe, 24 h
(42%).
a
Reagents: (a) concentrated H2SO4, concentrated HNO3, <15
°C (63%); (b) HSCH2CO2Et, NaOEt, EtOH, reflux, 3 h (96%); (c)
NaOH, EtOH, reflux, 3 h (96%); (d) quinoline, Cu, 190 °C, 3 h
(89%); (e) (i) DMFDMA, DMF, pyrrolidine, (ii) 5 M HCl, reflux,
30 min (78%); (f) NaBH4, EtOH, 1 h (98%); (g) MeSO2Cl, Et3N,
CH2Cl2, 1 h (88%); (h) EtOAc, H2, 10% Pd-C, Et3N, 1.5 h (100%);
(i) 3-pyridyl isocyanate, CH2Cl2-PhMe, 24 h (83%).
Synthesis of the 5,7-dimethyl derivative 39 (Scheme
6) started with the intermediate 29. Conversion of 29
to the trifluoroacetamide 34 followed by electrophilic
cyclization12 gave the tricyclic system of 35. Hydrolysis
of 35 to 36 and subsequent N-methylation provided the
intermediate 37. Deacetylation of 37 gave the indoline
38 which was reacted with 3-pyridyl isocyanate to give
the urea 39.
Sch em e 8. Synthesis of Benzofuran 58a
Compounds in Table 3 were prepared by several
different methods. Reduction of 2 with sodium cy-
anoborohydride afforded the reduced indoline 40. The
preparation of the benzothiophene 50 is summarized in
Scheme 7. Nitration of the aldehyde 41 gave a mixture
of nitrated products from which 42 was isolated by
recrystallization. Reaction of 42 with ethyl thioglycolate
under basic conditions13 afforded the benzothiophene 43,
which was converted to 45 via decarboxylation of the
corresponding carboxylic acid 44. Enamine formation
from 45 using Leimgruber's conditions14 followed by acid
hydrolysis provided the aldehyde 46, which was reduced
to the alcohol 47 and converted to the mesylate 48.
Reduction of the nitro group in 48 with concomitant
cyclization afforded the desired tricyclic intermediate
49 which was converted to the urea 50 by treatment
with 3-pyridyl isocyanate.
a
Reagents: (a) TFAA, CH2Cl2, NEt3, 1 h (93%); (b) EtOH, H2,
5% Pd-C, 4 h (99%); (c) concentrated H2SO4, NaNO2, aqueous
CuSO4, reflux, 5 min (67%); (d) K2CO3, ClCH2COCH3, DMF, 64 h
(97%); (e) concentrated H2SO4, 0 °C, 15 min (21%); (f) NaOH,
EtOH, 15 min (96%); (g) 3-pyridyl isocyanate, CH2Cl2-PhMe, 24
h (59%).
The synthesis of the furan 58 is shown in Scheme 8.
6-Nitroindoline (51) was converted to the trifluoroac-
etamide 52, which was then reduced to the aniline 53.
Diazotization of 53 followed by treatment with aqueous
acid afforded the phenol 54. Alkylation of 54 with
chloroacetone provided the ether 55 which was cyclized
to the tricyclic intermediate 56 by treatment with
concentrated sulfuric acid.15 Hydrolysis of 56 to 57
followed by reaction with 3-pyridyl isocyanate afforded
the desired product 58.
contractions of the rat stomach fundus, and for the
5-HT2C receptor using binding to rat cloned 5-HT2C
receptors expressed in HEK 293 cells and [3H]me-
sulergine as the radioligand.4 It should be noted that
the 5-HT2B receptor affinities are based on functional
data, and caution should be exercised when making
comparisons with data from the receptor binding assays
(5-HT2A and 5-HT2C).
We have previously reported compound 2 (SB-206553)
as a selective 5-HT2C/2B receptor antagonist with g100-
fold selectivity over the closely related 5-HT2A receptor.5
The angular indole isomer 3 was also found to possess
reasonable affinity at the 5-HT2C and 5-HT2B receptors.
Encouraged by these results, we have extended our
initial study with the synthesis of a range of pyrrole ring
Resu lts a n d Discu ssion
The affinity of compounds for the 5-HT2A receptor was
determined in rat frontal cortex preparations using [3H]-
ketanserin as radioligand in binding studies, for the
5-HT2B receptor from the inhibition of 5-HT-induced

Selective 5-HT2C/ 2B Receptor Antagonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4969
Ta ble 1. 5-HT2A, 5-HT2B, and 5-HT2C Receptor Binding Affinities of Indole Analoguesa
a
b
All values represent the mean of at least two determinations, with each determination lying within 0.2 log unit of the mean. Binding
affinity (rat frontal cortex, [3H]ketanserin). c Binding affinity (rat stomach fundus, inhibition of 5-HT). Binding affinity (rat clones
d
expressed in HEK 293 cells, [3H]mesulergine).
Ta ble 2. 5-HT2A, 5-HT2B, and 5-HT2C Receptor Binding Affinities of Indole Analoguesa
b
c
d
compd
R1
Me
H
Et
R2
R3
pKi, 5-HT2A
pA2, 5-HT2B
pKi, 5-HT2C
sel 5-HT2C/2A
2
26
22
23
27
24
33
39
H
H
H
H
H
H
Me
H
H
H
H
H
H
H
H
Me
6.3
5.8
5.9
5.5
5.4
<5.2
5.5
5.5
8.5
7.5
8.0
8.1
8.2
6.7
7.7
8.2
8.3
6.9
8.1
7.9
8.0
7.3
7.4
6.9
100
13
160
250
400
>130
80
Pr
iPr
CH2Ph
Me
Me
25
a
b
All values represent the mean of at least two determinations, with each determination lying within 0.2 log unit of the mean. Binding
affinity (rat frontal cortex, [3H]ketanserin). c Binding affinity (rat stomach fundus, inhibition of 5-HT). Binding affinity (rat clones
d
expressed in HEK 293 cells, [3H]mesulergine).
isomers (Table 1) designed to investigate the tolerance
of steric bulk at various positions around the rigid
aromatic ring system and also the effects of altering the
electrostatic potential orientation around the indole
ring. From Table 1, it can be seen that the [2,3-f] isomer
2 is the most potent in this series and has 100-fold
5-HT2C/5-HT2A selectivity. The [3,2-e], [2,3-g], and [2,3-
e] isomers 3, 5, and 12 all possess weaker affinity at
the 5-HT2C receptor having pKi’s between 0.7 and 1.6
log units less than 2. In contrast, the linear [3,2-f]
isomer 4 is over 100-fold less potent at the 5-HT2C
receptor compared with 2.
Since the linear [2,3-f] isomer 2 is the most potent
and most selective analogue, we decided to further
explore the structure-affinity relationships in this
specific isomeric series, with respect to the substitution
on the pyrrole ring. Removal of the N-methyl group in
2 to give 26 (Table 2) led to significant loss in 5-HT2C
and 5-HT2B receptor affinity, suggesting the presence
of a lipophilic pocket in these receptors accommodating
the N-methyl substituent. Alternatively, the presence
of the polar indole NH may be detrimental to binding.
The N-ethyl, N-propyl, and N-isopropyl analogues 22,
23, and 27 all retained a similar 5-HT2C receptor affinity
to that for 2, with a trend toward increasing 5-HT2C/5-
HT2A selectivity as size increases. The N-benzyl ana-
logue 24 however showed a reduction in affinity at the
5-HT2C and 5-HT2A receptors and more noticeably at the
5-HT2B receptor indicating a limit to the size of this
binding pocket in these receptors. We have further
probed the size of this lipophilic pocket with the
synthesis of the 5,6- and 5,7-dimethyl analogues 33 and
39. In both cases, a drop in 5-HT2C receptor affinity
and selectivity was observed which was more apparent

4970 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Forbes et al.
Ta ble 3. 5-HT2A, 5-HT2B, and 5-HT2C Receptor Binding Affinities of Indole Isosteresa
a
b
All values represent the mean of at least two determinations, with each determination lying within 0.2 log unit of the mean. Binding
affinity (rat frontal cortex, [3H]ketanserin). c Binding affinity (rat stomach fundus, inhibition of 5-HT). Binding affinity (rat clones
d
expressed in HEK 293 cells, [3H]mesulergine).
with 39. However it is interesting to note that the
5-HT2B receptor affinity of 39 is relatively unaffected
by this variation. Clearly, these results indicate subtle
differences between the 5-HT2C and 5-HT2B receptors.
Compound 2 has been shown to have anxiolytic
activity in three animal models of anxiety.9 However,
2 undergoes metabolic N-demethylation in the rat to
afford 26 as the principal metabolite. Unfortunately 26
is less potent and less selective than 2, and there was
evidence that 26 was present in significant quantities
in the brain.16 In order to obviate this problem, we
undertook the synthesis of several indole isosteres
lacking a pyrrole N-methyl group (Table 3) as poten-
tially metabolically stable analogues of 2.
The reduced indole 40 was found to be less potent at
the 5-HT2C receptor than 2 by a factor of 10, perhaps
due to the presence of the more polar indoline nitrogen.
Of more interest was the benzothiophene 50 which
retained high affinity at the 5-HT2C and 5-HT2B recep-
tors but unfortunately had increased affinity at the
5-HT2A receptor resulting in only 30-fold 5-HT2C/5-HT2A
selectivity. This confirms the earlier results, with
compound 26 and others, that steric bulk around the
F igu r e 1. Pharmacophore for 5-HT2C/2B receptor affinity.
an ID50 of 0.1 mg/kg, which compares favorably with
an ID50 of 0.3 mg/kg iv for compound 2, suggesting that
the poor oral bioavailability of 58 may be due to lack of
efficient gut absorption or extensive first-pass metabo-
lism.
Molecu la r Mod elin g Stu d ies. Our aim in the
ligand modeling studies was to use the “active analogue
approach” developed by Marshall17 and to overlap
common structural features of both active and inactive
analogues to gain information about the steric require-
ments at the 5-HT2C receptor. Compounds were over-
lapped by superimposing the urea functional group in
all compounds and using low-energy conformations
where appropriate. Our working hypothesis is based
on the assumption that weakly active compounds have
substituents in sterically disallowed areas. Data from
Table 1 indicate that with respect to 5-HT2C receptor
affinity, pyrrole ring fusion is tolerated across the e bond
of the indoline ring system in both orientations and
across the f bond in only one of the two possible
orientations (Figure 1). From the compounds reported
in Tables 1-3, along with other published derivatives,18
we were able to define an allowed volume at the 5-HT2C
receptor (Figure 2; shown in green) by overlapping
structures with a pKi g 7.5. The pyridine ring has been
excluded from this calculation because of a large number
of low-energy conformations. This allowed volume
should be similar to the size and shape of the binding
pocket. By also modeling compounds with a low 5-HT2A
receptor affinity (pKi < 5.5), we were able to define on
the surface of the 5-HT2C allowed volume a disallowed
area at the 5-HT2A receptor (Figure 2, shown in pink).
The use of this model in the design of potent and
selective 5-HT2C receptor antagonists will be the subject
of a future publication. In a complementary approach,
we sought to confirm these results by considering in
more detail how our series of diaryl ureas may be
interacting at the molecular level with the 5-HT2C and
5-HT2A receptors.
indole nitrogen in 2 is important for enhanced 5-HT2C
/
5-HT2A selectivity (see modeling section later). Accord-
ingly, we synthesized the inverted furan 58 which
possesses a potentially metabolically more stable C-
methyl substituent in the same area as the N-methyl
substituent in 2. As predicted, 58 demonstrated both
high 5-HT2C and 5-HT2B receptor affinity, along with
160-fold selectivity over the 5-HT2A receptor.
In Vivo Eva lu a tion . We have previously reported
the ID50 of 2 (5.5 mg/kg po) in an in vivo, centrally
mediated assay of 5-HT2C receptor function, the reversal
of m-CPP-induced hypolocomotion in the rat.5 Com-
pounds 22 and 27, which retained good in vitro activity,
were slightly less potent with ID50’s of 15 and 11 mg/kg
po, respectively. However, metabolic dealkylation of 22
and 27 remained a problem16 despite the increased
steric hindrance in 27. The benzothiophene 50 had an
ID50 of 1.5 mg/kg po representing the most potent
compound in vivo in this series. Rather disappointingly,
it was found that the more selective benzofuran 58
possessed an ID50 of 17 mg/kg po. However, after iv
administration, 58 was significantly more potent with

Selective 5-HT2C/ 2B Receptor Antagonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4971
F igu r e 2. 5-HT2C allowed (green) and 5-HT2A disallowed (pink) volumes viewed around the structure of compound 2.
Con str u ction of th e 5-HT2C Recep tor Mod el. The
Compound 2 however does not have a strongly basic
nitrogen (pKa ) 4.9), and an alternative binding mode
was therefore sought. Although ab initio quantum
calculations have shown that some rotation is allowed
around the urea bonds, the molecule is essentially rigid,
and its dimensions are such that it cannot fit in the
“agonist” cavity between TM3 and TM5. Other modes
of binding were examined, and on steric grounds, it
seemed likely that 2 is aligned along the axes of the
TM helices. SAR studies have shown that one essential
feature of this series of compounds is the carbonyl group
of the urea, and the receptor model was therefore
examined for possible hydrogen bond donors. TM3
contains two serine residues on its interior face, one and
two helical turns, respectively, below the conserved
aspartate. Within 5-HT receptors, the former of these
is only found in the 5-HT2 subtypes and is one of the
features that distinguishes the 5-HT2 family from other
5-HT sequences.27 As no binding has been observed of
2 to 5-HT1 cloned receptors, it seemed likely that this
serine (Ser-31228) might be a likely candidate as a
hydrogen bond donor. Positioning the urea carbonyl in
this pocket, it is possible to form two H-bonds with the
serines, Ser-312 and Ser-315 (see Figure 3). The two
aromatic regions of the molecule can now reside in well-
defined hydrophobic pockets. The first of these is
formed by a number of highly conserved aromatic
residues on helices 5 and 6, in the general vicinity of
Asp-308. These are Phe-508, Phe-616, Phe-617, and
Trp-613. The second hydrophobic pocket is a long cavity
bounded by His-318, Phe-609, Val-212, and Val-608. It
was not possible at first to decide which aromatic system
of 2 would dock into which pocket in the receptor. The
ligand was therefore initially positioned in both orienta-
tions and energy minimized using the CHARMm pro-
serotonin 5-HT2C receptor sequence19 was obtained from
the Swissprot database using the GCG software.
A
combination of hydropathic analysis using Kyte-Do-
little parameters and sequence alignments with other
related 5-HT receptors was used to identify the putative
transmembrane regions. The receptor model was gen-
erated with the transmembrane regions only, using the
GPCR_Builder program20 with its built-in standard
cationic neurotransmitter template. This template is
based very loosely on the conformation of the helix
bundle of bacteriorhodopsin. The inward faces of the
helices were determined by a Fourier transform of the
periodicity of conservation taken from a multiple-
sequence alignment (Clustal W21) of all mammalian
5-HT receptors. This procedure has been described in
detail by Donnelly et al.22 The ends of each helix were
capped with N-methyl amides at the C-termini and
acetamido groups at the N-termini. The backbone
dihedrals of this model were constrained to the confor-
mation of an “idealized” R-helix23 in a hydrophobic
environment but using weaker (10%) constraints on the
proline φ angles. Minimization was performed with
5000 steps of conjugate gradient and 5000 ABNR using
the CHARMm24 program. The dihedral constraints
were gradually weakened with continuous minimiza-
tion, and eventually these were completely removed.
The resulting model showed a completely acceptable
backbone φ-ψ plot, and as expected, “kinks” were
observed in the regions of the prolines.
Dock in g of Com p ou n d 2 (SB-206553) in to th e
5-HT2C Recep tor . It is generally accepted25,26 that the
standard agonists and antagonists at 5-HT receptors
bind their protonated amino sites to the highly con-
served aspartate on transmembrane helix 3 (TM3).

4972 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Forbes et al.
F igu r e 3. Structure 2 docked into the model of the 5-HT2C receptor.
gram. From the resulting interaction energies, it was
evident that 2 preferred to orient itself with the pyridine
ring in the upper (first) hydrophobic pocket. This places
the indole between the two valines on helices 2 and 6
(Val-212 and Val-608, see Figure 3). A study of the
sequences of the mammalian 5-HT2 receptors shows
that these two residues are among the principal differ-
ences between 5-HT2A and 5-HT2C. Both residues are
leucines in the 5-HT2A sequence, thus constricting the
pocket in this receptor. Binding is therefore less favor-
able in the 5-HT2A receptor compared to the 5-HT2C
receptor, and these valine-leucine differences are there-
fore postulated as being the major reason for the
observed selectivity. This result closely parallels the
result from our small molecule modeling studies (above).
of this model to design more potent and more selective
antagonists for the 5-HT2C receptor will be the subject
of a future publication.
Exp er im en ta l Section
NMR spectra were determined using Bruker AC-200 or AC-
250 spectrometers. Electron impact mass spectra were de-
termined using a Fisons VG 302 single quadrupole mass
spectrometer. Melting points were determined on a Buchi 510
melting point apparatus and are uncorrected. Tetrahydrofu-
ran (THF) was freshly distilled from sodium benzophenone
ketyl under an argon atmosphere before use. N,N-Dimethyl-
formamide (DMF) and dimethyl sulfoxide (DMSO) were of
commercial grade and dried over 4 Å molecular sieves before
use. Other solvents and reagents were of commercial grade
and used without purification. Organic extracts were dried
over anhydrous sodium sulfate before evaporation at reduced
pressure. Chromatography was performed on Merck Art. 7734
silica gel or Fluka silica gel 60 (60739).
Su m m a r y
In conclusion, we report the synthesis and biological
activity of a series of analogues of the previous lead
indole urea 2, designed to probe the 5-HT2C receptor
binding site. We have used small molecule modeling
studies to define a region in space which is allowed at
the 5-HT2C receptor but disallowed at the 5-HT2A
receptor. Docking of 2 into our model of the 5-HT2C
receptor has allowed us to propose a novel primary
binding interaction for this series of diaryl ureas,
involving a double hydrogen-bonding interaction be-
tween the urea carbonyl oxygen of the ligand and two
serine residues in the receptor. The difference of two
valine residues in the 5-HT2C receptor for leucine
residues in the 5-HT2A receptor is believed to account
for the observed 5-HT2C/5-HT2A selectivity with 2. Use
N-(1-Acetyl-6-in d olin yl)-2,2-d ieth oxyeth yla m in e (7). A
mixture of 1-acetyl-6-aminoindoline (6) (9.5 g, 54 mmol),
bromoacetaldehyde diethyl acetal (8.9 mL, 59 mmol), and
potassium carbonate (11.1 g) in DMF (100 mL) was heated at
100 °C for 18 h. The mixture was allowed to cool and
partitioned between ethyl acetate and water. The organic
layer was washed with brine, dried, and evaporated to give a
gray solid. Chromatography on silica gel using 0-100% ethyl
acetate/chloroform followed by 4% methanol/ethyl acetate
afforded 7 (6.1 g, 39%) as a light yellow solid: 1H NMR (CDCl3)
δ 1.26 (6H, t, J ) 9 Hz), 2.21 (3H, s), 3.09 (2H, t, J ) 9 Hz),
3.26 (2H, d, J ) 8 Hz), 3.5-3.8 (4H, m), 4.03 (2H, d, J ) 9
Hz), 4.68 (1H, t, J ) 8 Hz), 6.32 (1H, m), 6.97 (1H, d, J ) 9
Hz), 7.68 (1H, m).
1-Acetyl-7-(tr iflu or oa cetyl)-1,2,3,7-tetr a h yd r op yr r olo-
[3,2-f]in d ole (8) a n d 1-Acetyl-6-(tr iflu or oa cetyl)-1,2,3,6-
tetr a h yd r op yr r olo[2,3-g]in d ole (9). The acetal 7 (7.1 g, 24

Selective 5-HT2C/ 2B Receptor Antagonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4973
mmol) was added to a mixture of trifluoroacetic anhydride (35
mL) and trifluoroacetic acid (85 mL) and heated under reflux
for 66 h. After cooling, the solution was evaporated to dryness
and chromatographed on silica gel eluting with 0-60% ethyl
acetate/chloroform to give as a faster running band the angular
isomer 9 (1.3 g, 18%) as an impure product (MS m/ e 296 (M+)),
which was used without further purification, and as a slower
running band the linear isomer 8 (5.1 g, 71%): 1H NMR
(DMSO-d6) δ 2.32 (3H, s), 3.28 (2H, t, J ) 9 Hz), 4.20 (2H, t,
J ) 9 Hz), 7.30 (1H, d, J ) 9 Hz), 7.47 (1H, m), 7.76 (1H, m),
8.21 (1H, d, J ) 9 Hz); MS m/ e 296 (M+).
7-Meth yl-1,2,3,7-tetr ah ydr opyr r olo[3,2-f]in dole (10). The
linear isomer 8 (5.1 g, 17 mmol) was stirred in methanol (50
mL) containing potassium carbonate (3.6 g) for 1 h. The
mixture was diluted with water (500 mL) and the crude
product filtered off. The solid was resuspended in ethanol (20
mL) and stirred as 10% aqueous sodium hydroxide (3 mL) was
added. After 1 h, the mixture was diluted with water (200
mL), and the solid was filtered and dried to afford 1-acetyl-
1,2,3,7-tetrahydropyrrolo[3,2-f]indole (1.6 g, 45%) as a white
solid: 1H NMR (DMSO-d6) δ 2.15 (3H, s), 3.18 (2H, t, J ) 9
Hz), 4.10 (2H, t, J ) 9 Hz), 6.28 (1H, m), 7.20 (1H, m), 7.31
(1H, s), 8.18 (1H, s), 10.92 (1H, s).
This material was added to a stirred suspension of 80%
sodium hydride (0.31 g, 10.3 mmol) in DMF (20 mL) under
argon. After stirring for 30 min, methyl iodide (0.73 mL, 11.7
mmol) was added dropwise. The mixture was stirred for 1 h
and then diluted with water (200 mL). The resultant solid
was filtered off and dried to afford 1-acetyl-7-methyl-1,2,3,7-
tetrahydropyrrolo[3,2-f]indole (1.5 g, 91%): 1H NMR (CDCl3)
δ 2.28 (3H, s), 3.27 (2H, t, J ) 9 Hz), 3.78 (3H, s), 4.11 (2H, t,
J ) 9 Hz), 6.37 (1H, d, J ) 3 Hz), 6.98 (1H, d, J ) 3 Hz), 7.35
(1H, s), 8.29 (1H, s); MS m/ e 214 (M+).
δ 2.35 (3H, s), 3.22 (2H, t, J ) 9 Hz), 3.78 (3H, s), 4.16 (2H, t,
J ) 9 Hz), 6.95-7.10 (4H, m); MS m/ e 214 (M+).
This material was added to a mixture of ethanol (5 mL) and
10% aqueous sodium hydroxide (20 mL) and the solution
heated under reflux for 2 h. After cooling, the solution was
diluted with water (50 mL) and extracted with ethyl acetate.
The organic layer was dried and evaporated to afford 11 (0.31
g, 91%) as a brown oily solid: 1H NMR (CDCl3) δ 3.13 (2H, t,
J ) 9 Hz), 3.69 (2H, t, J ) 9 Hz), 3.74 (3H, s), 6.29 (1H, d, J
) 3 Hz), 6.74 (1H, d, J ) 9 Hz), 6.93 (1H, d, J ) 3 Hz), 7.07
(1H, d, J ) 9 Hz).
6-Met h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,6-t et r a h yd r o-
p yr r olo[2,3-g]in d ole (5). The title compound was prepared
in 62% yield from nicotinoyl azide (0.29 g, 1.9 mmol) and the
amine 11 (0.31 g, 1.8 mmol) using a procedure similar to that
1
for the preparation of 4: mp 197-199 °C; H NMR (DMSO-
d6) δ 3.20 (2H, t, J ) 9 Hz), 3.75 (3H, s), 4.20 (2H, t, J ) 9
Hz), 6.58 (1H, d, J ) 3 Hz), 7.05 (2H, m), 7.18 (1H, d, J ) 2
Hz), 7.32 (1H, m), 8.00 (1H, m), 8.21 (1H, m), 8.73 (1H, d, J )
2 Hz), 8.98 (1H, s); MS m/ e 292 (M+). Anal. (C17H16N4O) C,
H, N.
7-(Cya n om eth yl)-1-m eth yl-6-n itr oin d ole (14). 1-Meth-
yl-6-nitroindole (13) (1.1 g, 6.0 mmol) and (4-chlorophenoxy)-
acetonitrile (1.2 g, 7.2 mmol) were stirred in DMF (20 mL) at
0 °C, and potassium tert-butoxide (2.0 g, 17.9 mmol) was added
portionwise. After 15 min, the solution was diluted with water
(100 mL) and acidified with 5 M hydrochloric acid. The
resultant solid was filtered off and chromatographed on silica
gel using 0-20% ethyl acetate/dichloromethane as eluant to
afford 14 (1.0 g, 78%) as a yellow solid: 1H NMR (CDCl3/
DMSO-d6) δ 4.06 (3H, s), 4.25 (2H, s), 6.40 (1H, d, J ) 3 Hz),
7.14 (1H, d, J ) 3 Hz), 7.47 (1H, d, J ) 9 Hz), 7.51 (1H, d, J
) 9 Hz); MS m/ e 215 (M+).
This material (0.82 g, 3.8 mmol) was added to a mixture of
ethanol (5 mL) and 10% aqueous sodium hydroxide (20 mL),
and the solution was heated under reflux for 32 h. After
cooling, the solution was diluted with water (50 mL) and
extracted with ethyl acetate. The organic layer was dried and
evaporated to afford 10 (0.6 g, 90%) as a black oil: 1H NMR
(CDCl3) δ 3.10 (2H, t, J ) 9 Hz), 3.59 (2H, t, J ) 9 Hz), 3.67
(3H, s), 6.31 (1H, d, J ) 3 Hz), 6.57 (1H, s), 6.82 (1H, d, J )
3 Hz), 7.31 (1H, s).
Met h yl (1-Met h yl-6-n it r o-7-in d olyl)a cet a t e (15). The
nitrile 14 (0.88 g, 4.1 mmol) was stirred in methanol (100 mL)
and thionyl chloride (10 mL) added dropwise. The mixture
was then heated under reflux for 110 h. After cooling, the
mixture was evaporated to dryness and partitioned between
water and chloroform. The organic layer was dried and
evaporated to give a brown solid (1.1 g). This solid was
suspended in ethanol (10 mL) containing 10% aqueous sodium
hydroxide (1.5 mL) and stirred for 3 h. The mixture was then
diluted with water (100 mL) and washed with chloroform. The
aqueous layer was acidified and extracted with ethyl acetate
to afford a brown solid (0.4 g) which was dissolved in methanol
(25 mL) and stirred as thionyl chloride (2 mL) was added. The
mixture was heated under reflux for 2 h and then evaporated
to dryness to afford the methyl ester 15 (0.43 g, 41%) as a
brown solid:
7-Met h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,7-t et r a h yd r o-
pyr r olo[3,2-f]in dole (4). Nicotinoyl azide4 (CAUTION! Heat-
ing this material in the absence of solvent can lead to explosive
decomposition. Larger-scale (ca. 20 g or above) preparations
following this procedure are noticeably exothermic on reaching
70-80 °C, and copious quantities of nitrogen are rapidly
evolved. Appropriate precautions for the storage and utiliza-
tion of this reagent are strongly advised.) (0.57 g, 3.9 mmol)
in dry toluene (10 mL) was heated under reflux for 1 h and
allowed to cool to room temperature. This solution was added
to a stirred solution of the amine 10 (0.60 g, 3.5 mmol) in
dichloromethane (10 mL). After stirring for 1 h, the precipitate
was filtered off and dried to afford 4 (0.40 g, 43%) as a light
1H NMR (CDCl3) δ 3.79 (3H, s), 4.07 (3H, s), 4.38
(2H, s), 6.54 (1H, d, J ) 3 Hz), 7.20 (1H, d, J ) 3 Hz), 7.57
(1H, d, J ) 9 Hz), 7.78 (1H, d, J ) 9 Hz).
2-(1-Meth yl-6-n itr o-7-in d olyl)eth a n ol (16). The ester 15
(0.43 g, 1.7 mmol) was dissolved in THF (20 mL) under argon
and lithium aluminum hydride (0.16 g, 4.2 mmol) added
portionwise. After 30 min, the reaction was carefully quenched
with water; then the mixture was diluted with water (100 mL),
acidified with 5 M hydrochloric acid, and extracted with ethyl
acetate. The organic layer was washed with brine, dried, and
evaporated to afford the alcohol 16 (0.18 g, 47%) as a dark
gum: 1H NMR (MeOH-d4) δ 3.40 (2H, t, J ) 10 Hz), 3.90 (2H,
t, J ) 10 Hz), 4.12 (3H, s), 4.93 (1H, s), 6.46 (1H, d, J ) 3 Hz),
7.30 (1H, d, J ) 3 Hz), 7.44 (2H, s).
2-(1-Meth yl-6-n itr o-7-in d olyl)eth yl Meth a n esu lfon a te
(17). The alcohol 16 (0.18 g, 0.82 mmol), triethylamine (0.14
mL), and methanesulfonyl chloride (0.08 mL, 1.0 mmol) were
added to dichloromethane (10 mL) with stirring. After 30 min,
water was added and the mixture acidified with 5 M hydro-
chloric acid. The aqueous layer was extracted with chloroform,
and the combined organic layers were dried and evaporated.
Chromatography on silica gel using dichloromethane as eluant
afforded 17 (0.16 g, 64%) as a yellow gum: 1H NMR (CDCl3)
δ 2.91 (3H, s), 3.76 (2H, t, J ) 10 Hz), 4.16 (3H, s), 4.63 (2H,
t, J ) 10 Hz), 6.52 (1H, d, J ) 3 Hz), 7.21 (1H, d, J ) 3 Hz),
7.53 (1H, d, J ) 9 Hz), 7.67 (1H, d, J ) 9 Hz).
1
pink solid: mp 207-211 °C; H NMR (DMSO-d6) δ 3.31 (2H,
t, J ) 9 Hz), 3.77 (3H, s), 4.25 (2H, t, J ) 9 Hz), 6.33 (1H, d,
J ) 3 Hz), 7.22 (1H, d, J ) 3 Hz), 7.40 (2H, m), 8.00 (1H, s),
8.10 (1H, m), 8.29 (1H, m), 8.75 (1H, s), 8.84 (1H, d, J ) 2
Hz); MS m/ e 292 (M+). Anal. (C17H16N4O) C, H, N.
6-Meth yl-1,2,3,6-tetr a h yd r op yr r olo[2,3-g]in d ole (11).
The crude linear isomer 9 (1.3 g, 4.4 mmol) was stirred in
methanol (20 mL) containing potassium carbonate (0.92 g) for
1 h. The mixture was diluted with water (200 mL) and the
crude product filtered off to afford 1-acetyl-1,2,3,6-tetrahydro-
pyrrolo[2,3-g]indole (0.5 g, 60%) as a yellow solid: 1H NMR
(DMSO-d6) δ 2.21 (3H, s), 3.12 (2H, t, J ) 9 Hz), 4.11 (2H, t,
J ) 9 Hz), 6.84 (1H, m), 6.95 (1H, d, J ) 9 Hz), 7.09 (1H, d, J
) 9 Hz), 7.18 (1H, m); MS m/ e 200 (M+).
This material was added to a stirred suspension of 80%
sodium hydride (0.11 g, 3.7 mmol) in DMF (10 mL) under
argon. After stirring for 30 min, methyl iodide (0.25 mL, 4.0
mmol) was added dropwise. The mixture was stirred for 1 h
and then diluted with water (50 mL). The resultant solid was
filtered off and dried to afford 1-acetyl-6-methyl-1,2,3,6-
tetrahydropyrrolo[2,3-g]indole (0.41 g, 71%): 1H NMR (CDCl3)
4-Meth yl-1,2,3,4-tetr a h yd r op yr r olo[2,3-e]in d ole (18).
The mesylate 17 (0.16 g, 0.54 mmol) was dissolved in ethanol

4974 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Forbes et al.
(25 mL) and hydrogenated over 5% palladium on charcoal (0.23
g) at 50 psi for 2 h. The catalyst was removed by filtration,
and the filtrate was evaporated to dryness. The residue was
partitioned between chloroform and aqueous sodium bicarbon-
ate solution. The organic layer was dried and evaporated to
afford 18 (0.08 g, 85%) as a brown oil: 1H NMR (CDCl3) δ 3.41
(2H, t, J ) 9 Hz), 3.61 (2H, t, J ) 9 Hz), 3.86 (3H, s), 6.36 (1H,
d, J ) 3 Hz), 6.60 (1H, d, J ) 9 Hz), 6.79 (1H, d, J ) 3 Hz),
7.29 (1H, d, J ) 3 Hz).
4-Met h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,4-t et r a h yd r o-
p yr r olo[2,3-e]in d ole (12). The title compound was prepared
in 57% yield from nicotinoyl azide (0.08 g, 0.6 mmol) and the
amine 18 (0.08 g, 0.47 mmol) using a procedure similar to that
for the preparation of 4: mp 225 °C dec; 1H NMR (DMSO-d6)
δ 3.69 (2H, d, J ) 9 Hz), 3.91 (3H, s), 4.25 (2H, d, J ) 9 Hz),
6.33 (1H, d, J ) 3 Hz), 7.13 (1H, d, J ) 3 Hz), 7.32 (2H, m),
7.78 (1H, d, J ) 8 Hz), 7.99 (1H, m), 8.21 (1H, m), 8.64 (1H,
s), 8.75 (1H, d, J ) 2 Hz); MS m/ e 292 (M+). Anal. (C17H16N4O)
C, H, N.
1-Ace t yl-5-e t h yl-1,2,3,5-t e t r a h yd r op yr r olo[2,3-f]in -
d ole (20, R ) Et). To a stirred suspension of 80% sodium
hydride (0.42 g, 15 mmol) in DMF (20 mL) under argon was
added dropwise a solution of 1-acetyl-1,2,3,5-tetrahydropyr-
rolo[2,3-f]indole (19)5 (2.9 g, 15 mmol in DMF (10 mL)). After
stirring for 30 min, iodoethane (1.2 mL, 14.5 mmol) was added
and stirring continued for 18 h. The suspension was poured
onto water and extracted with ethyl acetate. The organic layer
was dried and evaporated to dryness to afford 20 (R ) Et) as
a yellow oil (3.1 g, 95%): 1H NMR (CDCl3) δ 1.44 (3H, t, J )
8 Hz), 2.23 (3H, s), 3.29 (2H, t, J ) 10 Hz), 4.0-4.25 (4H, m),
6.43 (1H, d, J ) 3 Hz), 7.03 (1H, d, J ) 3 Hz), 7.10 (1H, s),
8.48 (1H, s); MS m/ e 228 (M+).
1-Acetyl-5-p r op yl-1,2,3,5-tetr a h yd r op yr r olo[2,3-f]in d -
ole (20, R ) P r ). The title compound was prepared in 98%
yield from 1-acetyl-1,2,3,5-tetrahydropyrrolo[2,3-f]indole (19),5
sodium hydride, and iodopropane using a similar procedure
to that for 20 (R ) Et): 1H NMR (CDCl3) δ 0.92 (3H, t, J ) 8
Hz), 1.86 (2H, m), 2.29 (3H, s), 3.30 (2H, t, J ) 8 Hz), 4.0-
4.15 (4H, m), 6.47 (1H, d, J ) 3 Hz), 7.03 (1H, d, J ) 3 Hz),
7.12 (1H, s), 8.48 (1H, s); MS m/ e 242 (M+).
in 83% yield from nicotinoyl azide (2.2 g, 12 mmol) and the
amine 21 (R ) Et) (2.2 g, 12 mmol) using a procedure similar
to that for the preparation of 4. Chromatography of the
product on silica gel using 2% methanol/dichloromethane as
eluant afforded 22 (3.0 g, 83%) as a white solid: mp 202-203
1
°C; H NMR (DMSO-d6) δ 1.33 (3H, t, J ) 8 Hz), 3.28 (2H, t,
J ) 10 Hz), 4.16 (4H, m), 6.31 (1H, d, J ) 3 Hz), 7.24 (1H, d,
J ) 3 Hz), 7.30 (1H, s), 7.32 (1H, m), 8.00 (1H, m), 8.03 (1H,
s), 8.22 (1H, m), 8.65 (1H, s), 8.77 (1H, s); MS m/ e 306 (M+).
Anal. (C18H18N4O) C, H, N.
5-P r op yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,5-t et r a h yd r o-
p yr r olo[2,3-f]in d ole (23). The title compound was prepared
in 70% yield from nicotinoyl azide and the amine 21 (R ) Pr)
using a procedure similar to that for the preparation of 4: mp
1
175-176 °C; H NMR (CDCl3) δ 0.96 (3H, t, J ) 8 Hz), 1.88
(2H, m), 3.32 (2H, t, J ) 8 Hz), 4.04 (2H, t, J ) 8 Hz), 4.18
(2H, t, J ) 8 Hz), 6.46 (1H, d, J ) 3 Hz), 6.91 (1H, s), 7.07
(1H, d, J ) 3 Hz), 7.30 (2H, m), 7.94 (1H, s), 8.15 (1H, m),
8.34 (1H, m), 8.52 (1H, s); MS m/ e 320 (M+). Anal. (C19H20N4O)
C, H, N.
5-(P h en ylm et h yl)-1-(3-p yr id ylca r b a m oyl)-1,2,3,5-t et -
r a h yd r op yr r olo[2,3-f]in d ole (24). The title compound was
prepared in 80% yield from nicotinoyl azide and the amine 21
(R ) CH2Ph) using a procedure similar to that for the
preparation of 4: mp 199-200 °C; 1H NMR (DMSO-d6) δ 3.21
(2H, t, J ) 8 Hz), 4.13 (2H, t, J ) 8 Hz), 5.38 (2H, s), 6.40 (1H,
d, J ) 3 Hz), 7.1-7.4 (8H, m), 8.00 (1H, m), 8.05 (1H, m), 8.21
(1H, d, J ) 6 Hz), 8.62 (1H, s), 8.77 (1H, d, J ) 3 Hz); MS m/ e
368 (M+). Anal. (C23H20N4O) C, H, N.
1,2,3,5-Tetr a h yd r op yr r olo[2,3-f]in d ole (25). The title
compound was prepared in 89% yield from 1-acetyl-1,2,3,5-
tetrahydropyrrolo[2,3-f]indole (19)5 and aqueous sodium hy-
droxide using a procedure similar to that for the preparation
of 21 (R ) Et): 1H NMR (DMSO-d6) δ 2.89 (2H, t, J ) 10 Hz),
3.35 (2H, t, J ) 10 Hz), 6.08 (1H, m), 6.52 (1H, s), 7.00 (2H,
m); MS m/ e 158 (M+).
1-(3-P yr id ylca r ba m oyl)-1,2,3,5-tetr a h yd r op yr r olo[2,3-
f]in d ole (26). The title compound was prepared in 80% yield
from nicotinoyl azide and the amine 25 using a procedure
1
similar to that for the preparation of 4: mp 205-206 °C; H
1-Acetyl-5-(p h en ylm eth yl)-1,2,3,5-tetr a h yd r op yr r olo-
[2,3-f]in d ole (20, R ) CH2P h ). The title compound was
prepared in 79% yield from 1-acetyl-1,2,3,5-tetrahydropyrrolo-
[2,3-f]indole (19),5 sodium hydride, and benzyl bromide using
a similar procedure to that for 20 (R ) Et): 1H NMR (CDCl3)
δ 2.27 (3H, s), 3.24 (2H, t, J ) 8 Hz), 4.08 (2H, t, J ) 8 Hz),
5.29 (2H, s), 6.51 (1H, d, J ) 3 Hz), 7.00 (1H, s), 7.06 (3H, m),
7.28 (3H, m), 8.50 (1H, s); MS m/ e 290 (M+).
5-Eth yl-1,2,3,5-tetr a h yd r op yr r olo[2,3-f]in d ole (21, R )
Et). To a solution of the N-acetyl derivative 20 (R ) Et) (3.1
g, 14 mmol) in ethanol (60 mL) was added 10% aqueous sodium
hydroxide (60 mL) along with sodium hydroxide pellets (6 g)
under argon. The mixture was heated under reflux for 8 h,
allowed to cool, and poured onto water. The aqueous solution
was extracted with dichloromethane, and the organic layer was
dried and evaporated to give 21 (R ) Et) as a dark solid (2.5
g, 98%): 1H NMR (CDCl3) δ 1.41 (3H, t, J ) 8 Hz), 3.12 (2H,
t, J ) 10 Hz), 3.58 (2H, t, J ) 10 Hz), 4.08 (2H, q, J ) 8 Hz),
6.26 (1H, d, J ) 3 Hz), 6.84 (1H, s), 6.97 (1H, d, J ) 3 Hz),
7.12 (1H, s); MS m/ e 186 (M+).
5-P r op yl-1,2,3,5-tetr a h yd r op yr r olo[2,3-f]in d ole (21, R
) P r ). The title compound was prepared in 60% yield from
20 (R ) Pr) and aqueous sodium hydroxide using a procedure
similar to that for 21 (R ) Et): 1H NMR (CDCl3) δ 0.94 (3H,
t, J ) 8 Hz), 1.87 (2H, m), 3.12 (2H, t, J ) 8 Hz), 3.59 (2H, t,
J ) 8 Hz), 4.03 (2H, t, J ) 8 Hz), 6.29 (1H, d, J ) 3 Hz), 6.87
(1H, s), 6.98 (1H, d, J ) 3 Hz), 7.28 (1H, s); MS m/ e 200 (M+).
5-(P h en ylm et h yl)-1,2,3,5-t et r a h yd r op yr r olo[2,3-f]in -
d ole (21, R ) CH2P h ). The title compound was prepared in
96% yield from 20 (R ) CH2Ph) and aqueous sodium hydroxide
using a procedure similar to that for 21 (R ) Et): 1H NMR
(CDCl3) δ 3.08 (2H, t, J ) 8 Hz), 3.57 (2H, t, J ) 8 Hz), 5.28
(2H, s), 6.35 (1H, d, J ) 3 Hz), 6.88 (1H, s), 7.02 (2H, m), 7.10
(2H, m), 7.29 (3H, m); MS m/ e 248 (M+).
NMR (DMSO-d6) δ 3.21 (2H, t, J ) 10 Hz), 4.15 (2H, t, J ) 10
Hz), 6.31 (1H, m), 7.20 (1H, m), 7.31 (1H, m), 8.03 (2H, m),
8.21 (1H, m), 8.64 (1H, bs), 8.79 (1H, m), 10.88 (1H, bs); MS
m/ e 278 (M+). Anal. (C16H14N4O) C, H, N.
5-Isop r op yl-1-(3-p yr id ylca r ba m oyl)-1,2,3,5-tetr a h yd r o-
p yr r olo[2,3-f]in d ole (27). The indolyl urea 26 (0.38 g, 1.4
mmol) was dissolved in DMF (10 mL) at 0 °C and treated with
80% sodium hydride (0.09 g, 2.9 mmol). After 15 min,
2-iodopropane (0.14 mL, 1.5 mmol) was added, and the mixture
was allowed to warm to room temperature. After 4 h,
additional sodium hydride (0.05 g) and 2-iodopropane (0.07
mL) were added, and the mixture was stirred for 18 h. The
mixture was evaporated to dryness and extracted with ethyl
acetate. The ethyl acetate layer was dried and evaporated.
Chromatography of the residue on silica gel using 0-5%
methanol/dichloromethane afforded 27 (0.15 g, 35%): mp 182-
183 °C; 1H NMR (CDCl3) δ 1.52 (6H, d, J ) 7 Hz), 3.29 (2H, t,
J ) 8 Hz), 4.17 (2H, t, J ) 8 Hz), 4.60 (1H, m, J ) 7 Hz), 6.47
(1H, d, J ) 3 Hz), 6.90 (1H, s), 7.2-7.3 (3H, m), 7.95 (1H, s),
8.11 (1H, m), 8.30 (1H, d, J ) 6 Hz), 8.53 (1H, d, J ) 3 Hz);
MS m/ e 320 (M+). Anal. (C19H20N4O) C, H, N.
N-(1-Acet yl-5-in d olin yl)-2-ch lor oa llyla m in e
(29).
1-Acetyl-5-aminoindoline (28) (4.36 g, 24.8 mmol), potassium
carbonate (5.1 g, 37 mmol), and 2,3-dichloro-1-propene (4.5 mL,
49 mmol) were stirred in DMF (50 mL) at 70 °C for 16 h. The
mixture was then diluted with water (500 mL) and stirred for
10 min. The precipitate was filtered off and dried to afford
29 (5.7 g, 92%) as a dark olive solid: 1H NMR (CDCl3) δ 2.19
(3H, s), 3.13 (2H, t, J ) 8 Hz), 3.9-4.2 (5H, m), 5.32 (1H, m),
5.41 (1H, m), 6.4-6.6 (2H, m), 8.05 (1H, d, J ) 9 Hz).
N -(1-Ac e t y l-5-i n d o li n y l)-2-c h lo r o -N -m e t h y la lly l-
a m in e (30). To a mixture of 40% aqueous formaldehyde (2.8
mL, 36 mmol) and 3 M sulfuric acid (5 mL) was added
portionwise, with cooling, a suspension of sodium borohydride
(1.7 g, 44 mmol) and the amine 29 (3.1 g, 12 mmol) in
tetrahydrofuran (60 mL). The mixture was stirred at room
5-E t h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,5-t e t r a h yd r o-
p yr r olo[2,3-f]in d ole (22). The title compound was prepared

Selective 5-HT2C/ 2B Receptor Antagonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4975
temperature for 15 min and then basified with excess solid
sodium hydroxide. The mixture was extracted with ethyl
acetate and the combined organic solution dried and evapo-
rated to afford 30 (3.8 g) as a brown solid which was used
without further purification: 1H NMR (CDCl3) δ 2.20 (3H, s),
3.00 (3H, s), 3.16 (2H, t, J ) 7 Hz), 3.95-4.20 (4H, m), 5.22
(1H, m), 5.30 (1H, m), 6.56 (2H, m), 8.08 (1H, d, J ) 8 Hz).
1-Acetyl-5,6-d im eth yl-1,2,3,5-tetr a h yd r op yr r olo[2,3-f]-
in d ole (31). The allylamine 30 (2.1 g, 7.9 mmol) was stirred
in polyphosphoric acid (44 g) at 140 °C for 24 h, cooled, diluted
with water, and extracted with ethyl acetate. The organic
layer was washed with brine, dried, and evaporated to give a
pink solid. Chromatography on silica gel eluting with 0-20%
ethyl acetate/dichloromethane afforded, inter alia, the linear
tricyclic 31 (0.21 g, 12%) as a white solid: 1H NMR (CDCl3) δ
(major rotamer) 2.25 (3H, s), 2.39 (3H, s), 3.29 (2H, t, J ) 7
Hz), 3.62 (3H, s), 4.10 (2H, t, J ) 7 Hz), 6.22 (1H, s), 7.03 (1H,
s), 8.38 (1H, s).
using a procedure similar to that for the preparation of 4: mp
224-227 °C; 1H NMR (DMSO-d6) δ 2.19 (3H, s), 3.27 (2H, t, J
) 8 Hz), 3.67 (3H, s), 4.18 (2H, t, J ) 8 Hz), 6.96 (1H, s), 7.21
(1H, s), 7.32 (1H, dd, J ) 7, 4 Hz), 8.00 (1H, s), 8.02 (1H, dd,
J ) 7, 2 Hz), 8.22 (1H, dd, J ) 4, 2Hz), 8.62 (1H, s), 8.77 (1H,
s); MS m/ e 306 (M+). Anal. (C18H18N4O) C, H, N.
5-Met h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,5,6,7-h exa h y-
d r op yr r olo[2,3-f]in d ole (40). The indole 25 (0.80 g, 2.7
mmol) was treated with sodium cyanoborohydride (0.90 g, 14
mmol) in acetic acid at room temperature for 4 h. Water was
added, and the mixture was basified with 10% aqueous sodium
hydroxide and extracted with dichloromethane. The organic
layer was washed with brine, dried, and evaporated to give a
solid which was recrystallized from methanol/60-80 °C pe-
troleum ether ether to give 40 (0.40 g, 53%) as a white solid:
mp 153-155 °C; 1H NMR (DMSO-d6) δ 2.62 (3H, s), 2.80 (2H,
t, J ) 8 Hz), 3.05 (2H, t, J ) 8 Hz), 3.17 (2H, t, J ) 8 Hz), 4.10
(2H, t, J ) 8 Hz), 6.40 (1H, s), 7.30 (1H, q, J ) 4 Hz), 7.65
(1H, s), 7.91-7.98 (1H, m), 8.19 (1H, d, J ) 4 Hz), 8.55 (1H,
s), 8.72 (1H, d, J ) 4 Hz); MS m/ e 294 (M+). Anal. (C17H18N4O)
C, H, N.
2-Br om o-4-m eth yl-5-n itr oben za ld eh yd e (42). 2-Bromo-
4-methylbenzaldehyde (41) (57.0 g, 286 mmol) was added
dropwise as a melt to concentrated sulfuric acid (300 mL) with
stirring and ice cooling. Concentrated nitric acid (36 mL, 570
mmol) was then added dropwise keeping the temperature
below 15 °C. After the addition was completed, stirring was
continued for 20 min; then the mixture was poured onto ice/
water (2.5 L) with vigorous stirring. The resultant precipitate
was filtered off, extracted with 5% methanol/chloroform,
washed with saturated sodium hydrogen carbonate solution
followed by brine, and then evaporated to dryness. Recrys-
tallization from ethyl acetate/60-80 °C petroleum ether gave
42 (44.2 g, 63%) as a yellow solid: 1H NMR (CDCl3) δ 2.69
(3H, s), 7.70 (1H, s), 8.50 (1H, s), 9.80 (1H, s); MS m/ e 243,
245 (M+).
Eth yl 6-Meth yl-5-n itr oben zoth iop h en e-2-ca r boxyla te
(43). Sodium (4.3 g, 180 mmol) was added portionwise to dry
ethanol (250 mL) over 30 min. The solution was then cooled
in an ice bath, and ethyl 2-mercaptoacetate (20 mL, 180 mmol)
was added. The mixture was stirred at 5 °C for 20 min and
then diluted with ethanol (300 mL). The aldehyde 42 (44.2 g,
180 mmol) was then added portionwise over 10 min. Ad-
ditional ethanol (200 mL) was added and the mixture heated
under reflux for 3 h. The mixture was then cooled and the
ethanol evaporated. The residue was partitioned between
water and dichloromethane, and the organic layer was washed
with brine, dried, and evaporated to give 43 (46 g, 96%): 1H
NMR (CDCl3) δ 1.42 (3H, t, J ) 8 Hz), 2.75 (3H, s), 4.43 (2H,
q, J ) 8 Hz), 7.70 (1H, s), 8.10 (1H, s), 8.53 (1H, s); MS m/ e
265 (M+).
6-Meth yl-5-n itr oben zoth ioph en e-2-car boxylic Acid (44).
The ester 43 (46 g, 170 mmol) was heated under reflux in
ethanol (500 mL), water (300 mL), and sodium hydroxide (27
g, 680 mmol) for 3 h. The mixture was allowed to cool and
most of the ethanol removed under reduced pressure. The
residue was poured into a mixture of 5 M HCl (250 mL) and
ice/water (1 L) with vigorous stirring. The precipitate was
filtered off and dried at 80 °C to give 44 (41 g, 100%) as a
brown solid: 1H NMR (DMSO-d6) δ 2.62 (3H, s), 8.20 (1H, s),
8.26 (1H, s), 8.75 (1H, s), 13.7 (1H, bs); MS m/ e 237 (M+).
6-Meth yl-5-n itr oben zoth iop h en e (45). The acid 44 (41
g, 170 mmol), in quinoline (300 mL) containing copper powder
(25 g), was heated to 190 °C for 3 h. The mixture was cooled
and diluted with diethyl ether (1 L). The mixture was filtered
through Kieselguhr, then washed with 5 M HCl (3 × 500 mL),
water, and brine, dried, and evaporated to give 45 (29.6 g,
89%): 1H NMR (CDCl3) δ 2.72 (3H, s), 7.40 (1H, d, J ) 6 Hz),
7.55 (1H, d, J ) 6 Hz), 7.80 (1H, s), 8.51 (1H, s); MS m/ e 193
(M+).
5,6-Dim eth yl-1,2,3,5-tetr ah ydr opyr r olo[2,3-f]in dole (32).
The title compound was prepared in 60% yield from the
acetamide 31 and aqueous sodium hydroxide using a procedure
similar to that for 21 (R ) Et): 1H NMR (CDCl3) δ 2.36 (3H,
s), 3.12 (2H, t, J ) 7 Hz), 3.56 (2H, t, J ) 7 Hz), 3.58 (3H, s),
6.06 (1H, s), 6.78 (1H, s), 7.03 (1H, s).
5,6-Dim et h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,5-t et r a h y-
d r op yr r olo[2,3-f]in d ole (33). The title compound was pre-
pared in 39% yield from nicotinoyl azide and the amine 32
using a procedure similar to that for the preparation of 4: mp
1
225 °C dec; H NMR (DMSO-d6) δ 2.36 (3H, s), 3.26 (2H, t, J
) 8 Hz), 3.60 (3H, s), 4.17 (2H, t, J ) 8 Hz), 6.11 (1H, s), 7.21
(1H, s), 7.32 (1H, dd, J ) 7, 4 Hz), 7.93 (1H, s), 8.00 (1H, d, J
) 7 Hz), 8.21 (1H, d, J ) 4 Hz), 8.63 (1H, s), 8.75 (1H, d, J )
2 Hz); MS m/ e 306 (M+). Anal. (C18H18N4O) C, H, N.
N-(1-Acet yl-5-in d olin yl)-2-ch lor o-N-(t r iflu or oa cet yl)-
a llyla m in e (34). The amine 29 (5.7 g, 25 mmol) and trieth-
ylamine (3.3 mL, 27 mmol) in chloroform (100 mL) were stirred
during the addition of trifluoroacetic anhydride (3.8 mL, 27
mmol) dropwise. The mixture was stirred for 1 h and then
diluted with water (100 mL). The organic layer was dried and
evaporated to give 34 (7.5 g, 95%) as a dark oil: 1H NMR
(CDCl3) δ 2.25 (3H, s), 3.24 (2H, t, J ) 8 Hz), 4.16 (2H, t, J )
8 Hz), 4.52 (2H, s), 5.23 (1H, s), 5.36 (1H, s), 7.1 (2H, m), 8.23
(1H, d, J ) 8 Hz).
1-Acet yl-7-m et h yl-5-(t r iflu or oa cet yl)-1,2,3,5-t et r a h y-
d r op yr r olo[2,3-f]in d ole (35). The trifluoroacetamide 34 (7.6
g, 22 mmol) was stirred in polyphosphoric acid (38 g) at 140
°C for 1.5 h. The mixture was cooled, diluted with water (200
mL), and then extracted with ethyl acetate. The organic layer
was filtered through Kieselguhr, dried, and evaporated to give
a dark oil. Chromatography on silica gel eluting with 0-20%
ethyl acetate/chloroform gave 35 (0.5 g, 7%) as a yellow solid:
1H NMR (CDCl3) δ 2.28 (3H, s), 2.33 (3H, s), 3.36 (2H, t, J )
8 Hz), 4.18 (2H, t, J ) 8 Hz), 7.19 (1H, s), 8.24 (1H, s), 8.36
(1H, s).
1-Acet yl-7-m et h yl-1,2,3,5-t et r a h yd r op yr r olo[2,3-f]in -
d ole (36). The title compound was prepared in 91% yield from
the trifluoroacetamide 35 using potassium carbonate in metha-
nol as in the preparation of 11: 1H NMR (DMSO-d6) δ 2.15
(3H, s), 2.18 (3H, s), 3.17 (2H, t, J ) 8 Hz), 4.09 (2H, t, J ) 8
Hz), 7.00 (1H, s), 7.14 (1H, s), 8.16 (1H, s), 10.55 (1H, bs).
1-Acetyl-5,7-d im eth yl-1,2,3,5-tetr a h yd r op yr r olo[2,3-f]-
in d ole (37). The title compound was prepared in 79% yield
from the indole 36, sodium hydride, and methyl iodide using
a procedure similar to that for 20 (R ) Et): 1H NMR (CDCl3)
δ (major rotamer) 2.27 (3H, s), 2.30 (3H, s), 3.30 (2H, t, J ) 8
Hz), 3.68 (3H, s), 4.10 (2H, t, J ) 8 Hz), 6.76 (1H, s), 7.05 (1H,
s), 8.42 (1H, s).
5,7-Dim eth yl-1,2,3,5-tetr ah ydr opyr r olo[2,3-f]in dole (38).
The title compound was prepared in ∼60% yield from the
acetamide 37 and aqueous sodium hydroxide using a procedure
similar to that for 21 (R ) Et). Some starting material
remained in the crude product which was used in the next
step without further purification.
5,7-Dim et h yl-1-(3-p yr id ylca r b a m oyl)-1,2,3,5-t et r a h y-
d r op yr r olo[2,3-f]in d ole (39). The title compound was pre-
pared in 42% yield from nicotinoyl azide and the amine 38
5-Nitr oben zoth iop h en e-6-a ceta ld eh yd e (46). 6-Methyl-
5-nitrobenzothiophene (45) (29.6 g, 153 mmol) in DMF (300
mL) was treated with DMF dimethyl acetal (61 mL, 460 mmol)
and pyrrolidine (25.7 mL, 300 mmol) at 150 °C for 3 h under
argon. The mixture was cooled and concentrated under
reduced pressure. Toluene (600 mL), 5 M HCl (300 mL), and

4976 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Forbes et al.
water (800 mL) were added to the residue, which was heated
under reflux for 30 min. The mixture was cooled, extracted
with ethyl acetate, washed with water, dried, and evaporated.
Chromatography on silica gel eluting with dichloromethane
afforded 46 (26.4 g, 78%): 1H NMR (CDCl3) δ 4.22 (2H, s),
7.48 (1H, d, J ) 6 Hz), 7.64 (1H, d, J ) 6 Hz), 7.81 (1H, s),
8.67 (1H, s), 9.91 (1H, s); MS m/ e 221 (M+).
2-(5-Nitr o-6-ben zoth ien yl)eth a n ol (47). The aldehyde 46
(25.3 g, 114 mmol) in ethanol (300 mL) was treated portionwise
with sodium borohydride (8.7 g, 130 mmol), with stirring at
room temperature for 1 h. The solution was then concentrated
under reduced pressure, and water (500 mL) was added. The
mixture was carefully acidified with 5 M HCl, and extracted
with ethyl acetate. The organic layer was washed with water
and brine, dried, and evaporated to give 47 (25.0 g, 98%): 1H
NMR (CDCl3) δ 1.78 (1H, bs), 3.29 (2H, t, J ) 8 Hz), 4.12 (2H,
t, J ) 8 Hz), 7.41 (1H, d, J ) 6 Hz), 7.59 (1H, d, J ) 6 Hz),
7.89 (1H, s), 8.45 (1H, s); MS m/ e 223 (M+).
2-(5-Nitr o-6-ben zoth ien yl)eth yl Meth an esu lfon ate (48).
The alcohol 47 (25.0 g, 112 mmol) was dissolved in dry
dichloromethane (500 mL) and treated with triethylamine
(16.4 mL, 118 mmol), and methanesulfonyl chloride (9.2 mL,
120 mmol) was added dropwise over 15 min. The mixture was
stirred at room temperature for 1 h, then washed with
saturated aqueous sodium hydrogen carbonate, and brine,
dried, and evaporated to afford 48 (29.6 g, 88%): 1H NMR
(CDCl3) δ 2.96 (3H, s), 3.48 (2H, t, J ) 8 Hz), 4.60 (2H, t, J )
8 Hz), 7.46 (1H, d, J ) 7 Hz), 7.55 (1H, d, J ) 7 Hz), 7.90 (1H,
s), 8.56 (1H, s); MS m/ e 301 (M+).
6,7-Dih yd r o-5H-th ien o[2,3-f]in d ole (49). The mesylate
48 (10.3 g, 34.0 mmol) in ethyl acetate (500 mL) containing
triethylamine (10 mL) and 10% palladium on carbon (1 g) was
hydrogenated at atmospheric pressure at 50 °C for 1.5 h. The
reaction mixture was then filtered through Kieselguhr, washed
with water, and evaporated to dryness to afford 49 (6.0 g,
100%): 1H NMR (CDCl3) δ 3.13 (2H, t, J ) 8 Hz), 3.54 (1H,
bs), 3.65 (2H, t, J ) 8 Hz), 7.02 (1H, s), 7.15 (1H, d, J ) 7 Hz),
7.30 (1H, d, J ) 7 Hz), 7.57 (1H, s); MS m/ e 175 (M+).
6,7-Dih yd r o-5-(3-p yr id ylca r b a m oyl)-5H -t h ien o[2,3-f]-
in d ole (50). The title compound was prepared in 83% yield
from the amine 49 (17.7 g, 101 mmol) and nicotinoyl azide
(16.0 g, 108 mmol) using the same procedure as for 4: mp 189-
190 °C; 1H NMR (DMSO-d6) δ 3.35 (2H, t, J ) 8 Hz), 4.19 (2H,
t, J ) 8 Hz), 6.67 (1H, s), 7.2-7.4 (3H, m), 7.63 (1H, m), 8.12
(1H, m), 8.32 (2H, m), 8.53 (1H, m); MS m/ e 295 (M+). Anal.
(C16H13N3OS) C, H, N.
6-Nitr o-1-(tr iflu or oa cetyl)in d olin e (52). 6-Nitroindoline
(51) (6.5 g, 40 mmol) and triethylamine (6.6 mL, 47 mmol)
were stirred in dichloromethane (65 mL) as trifluoroacetic
anhydride (6.6 mL, 47 mmol) was added dropwise. This
mixture was stirred for 45 min; then water (100 mL) was
added. After stirring for 10 min, the mixture was acidified
with 5 M HCl. The organic layer was washed with brine,
dried, and evaporated to give 52 (9.6 g, 93%) as a yellow-brown
solid: 1H NMR (CDCl3) δ 3.40 (2H, t, J ) 8 Hz), 4.40 (2H, t,
J ) 8 Hz), 7.40 (1H, d, J ) 8 Hz), 8.10 (1H, dd, J ) 8, 2 Hz),
9.05 (1H, d, J ) 2 Hz).
6-Am in o-1-(tr iflu or oa cetyl)in d olin e (53). The nitroin-
doline 52 (4.1 g, 16 mmol) in ethanol (200 mL) was hydroge-
nated over 5% palladium on carbon (60% aqueous paste, 1 g)
for 4 h. The catalyst was removed by filtration and the filtrate
evaporated to afford 53 (3.6 g, 99%) as a light brown solid:
1H NMR (CDCl3) δ 3.15 (2H, t, J ) 8 Hz), 3.35 (2H, bs), 4.25
(2H, t, J ) 8 Hz), 6.50 (1H, dd, J ) 8 Hz, 2 Hz), 7.00 (1H, d,
J ) 8 Hz), 7.65 (1H, d, J ) 2 Hz).
NMR (DMSO-d6) δ 3.10 (2H, t, J ) 8 Hz), 4.25 (2H, t, J ) 8
Hz), 6.60 (1H, dd, J ) 8, 2 Hz), 7.10 (1H, d, J ) 8 Hz), 7.60
(1H, d, J ) 2 Hz), 9.05 (1H, bs).
6-(2-Oxop r op oxy)-1-(t r iflu or oa cet yl)in d olin e (55).
A
mixture of the phenol 54 (2.0 g, 8.7 mmol), potassium carbon-
ate (1.8 g, 13.0 mmol), and chloroacetone (0.8 mL, 10 mmol)
were stirred in dry DMF (20 mL) for 64 h. The mixture was
diluted with ethyl acetate (200 mL), filtered, washed with
water, dried, and evaporated to give 55 (2.4 g, 97%) as a brown
oil: 1H NMR (CDCl3) δ 2.30 (3H, s), 3.20 (2H, t, J ) 8 Hz),
4.30 (2H, t, J ) 8 Hz), 4.60 (2H, s), 6.75 (1H, dd, J ) 8, 2 Hz),
7.15 (1H, d, J ) 8 Hz), 7.85 (1H, d, J ) 2 Hz).
5,6-Dih yd r o-3-m eth yl-7-(tr iflu or oa cetyl)fu r o[3,2-f]in -
d ole (56). Concentrated sulfuric acid (25 mL) was added at
0 °C to the ketone 55 (2.4 g, 8.4 mmol). The dark mixture
was stirred at room temperature for 15 min and poured onto
ice. The crude product was extracted into ethyl acetate,
washed with water and brine, dried, and evaporated to
dryness. Chromatography on silica gel eluting with chloroform
gave 56 (0.5 g, 21%) as a yellow solid: 1H NMR (CDCl3) δ 2.25
(3H, s), 3.35 (2H, t, J ) 8 Hz), 4.35 (2H, t, J ) 8 Hz), 7.35 (1H,
s), 7.45 (1H, s), 8.35 (1H, s); MS m/ e 263 (M+).
5,6-Dih yd r o-3-m et h ylfu r o[3,2-f]in d ole (57). The tri-
fluoroacetamide 56 (0.50 g, 1.8 mmol) was stirred in ethanol
(10 mL) as 2.5 M aqueous sodium hydroxide (1 mL) was added.
The mixture was stirred for 15 min, diluted with water, and
extracted with ethyl acetate. The organic layer was washed
with brine, dried, and evaporated to give 57 (0.30 g, 96%) as
a brown oil: 1H NMR (CDCl3) δ 2.20 (3H, s), 3.10 (2H, t, J )
8 Hz), 3.30 (1H, bs), 3.60 (2H, t, J ) 8 Hz), 6.70 (1H, s), 7.20
(2H, m); MS m/ e 173 (M+).
5,6-Dih yd r o-3-m eth yl-7-(3-p yr id ylca r ba m oyl)fu r o[3,2-
f]in d ole (58). The title compound was prepared in 59% yield
from the amine 57 (0.15 g, 0.87 mmol) and nicotinoyl azide
(0.14 g, 0.94 mmol) using the same procedure as for 4: mp
227-228 °C; 1H NMR (DMSO-d6) δ 2.17 (3H, s), 3.27 (2H, t, J
) 8 Hz), 4.24 (2H, t, J ) 8 Hz), 7.35 (2H, m), 7.62 (1H, d, J )
2 Hz), 8.01 (2H, m), 8.24 (1H, m), 8.76 (2H, s); MS m/ e
293.1162 (M+). C17H15N3O2 requires 293.1164.
Com p u ta tion a l Stu d ies a n d Molecu la r Mod elin g. Mo-
lecular mechanics and dynamics calculations on the protein
structures with and without bound ligands were performed
using the CHARMm program. Geometries for the ligands
were generated by optimization with the COSMIC molecular
mechanics force field or, where parameters were unavailable,
with the semiempirical molecular orbital program VAMP,
using the AM1 Hamiltonian. Natural atomic orbital charges
were used for the docking calculations, and these were
generated from single-point Hartree-Fock calculations using
the Spartan suite of programs (Wavefunction Inc.) with a
3-21G basis set. The results for the CHARMm calculations
were visualized using the program QUANTA (Molecular
Simulations, Inc.). The allowed-disallowed volume and over-
lap studies on the ligands were carried out using the SYBYL
program (Tripos Assoc.). All calculations were performed on
either a Silicon Graphics 4D-380 Powerserver or a DEC alpha
2100 workstation and the results visualized on a Silicon
Graphics Indigo-2 Extreme workstation.
Ack n ow led gm en t. We are grateful to Steve Bro-
midge and David T. Davies for helpful discussions.
Refer en ces
6-Hyd r oxy-1-(tr iflu or oa cetyl)in d olin e (54). The aniline
53 (3.0 g, 13 mmol) was stirred in water (30 mL) as concen-
trated sulfuric acid (3 mL) was added dropwise. The solution
was cooled to 0 °C, and a solution of sodium nitrite (0.98 g, 14
mmol) in water (10 mL) was added dropwise, maintaining the
temperature e 0 °C. The mixture was stirred for 5 min and
then transferred to a boiling solution of copper sulfate (13.0
g, 52 mmol) in water (50 mL). The mixture was boiled for 5
min and cooled, and the black solid was filtered off and dried.
Chromatography on silica gel eluting with 0-5% methanol in
chloroform gave 54 (1.0 g, 67%) as a dark brown solid: 1H
(1) Blackburn, T. P. Pharmacology and Function of 5-HT Receptors.
In Advances in Neuropharmacology; Rose, F. C., Ed.; Smith-
Gordon and Nishimura: New York, 1993; 51-66.
(2) Hoyer, D.; Clarke, D. E.; Fozard, J . R.; Hartig, P. R.; Martin, G.
R.; Mylecharane, E. J .; Saxena, P. R.; Humphrey, P. P. A.
International Union of Pharmacology Classification of Receptors
for 5-Hydroxytryptamine. Pharmacol. Rev. 1994, 46, 157-203.
(3) (a) Humphrey, P. P. A.; Hartig, P.; Hoyer, D. A Proposed New
Nomenclature of 5-HT Receptors. Trends Pharmacol. Sci. 1993,
14, 233-236. (b) Nelson, D. L. The 5-HT2 Subfamily of Recep-
tors: Pharmacological Challenges. Med. Chem. Res. 1993, 3,
306-316.

Selective 5-HT2C/ 2B Receptor Antagonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4977
(4) Forbes, I. T.; Kennett, G. A.; Gadre, A.; Ham, P.; Hayward, C.
J .; Martin, R. T.; Thompson, M.; Wood, M. D.; Baxter, G. S.;
Glen, A.; Murphy, O. E.; Stewart, B. R.; Blackburn, T. P. N-(1-
Methyl-5-indolyl)-N'-(3-pyridyl)urea Hydrochloride: The First
Selective 5-HT1C Receptor Antagonist. J . Med. Chem. 1993, 36,
1104-1107.
(5) Forbes, I. T.; Ham, P.; Booth, D. H.; Martin, R. T.; Thompson,
M.; Baxter, G. S.; Blackburn, T. P.; Glen, A.; Kennett, G. A.;
Wood, M. D. 5-Methyl-1-(3-pyridylcarbamoyl)-1,2,3,5-tetrahydro-
pyrrolo[2,3-f]indole: A Novel 5-HT2C/5-HT2B Receptor Antagonist
with Improved Affinity, Selectivity and Oral Activity. J . Med.
Chem. 1995, 38, 2524-2530.
(6) Kennett, G. A.; Whitton, P.; Shah, K.; Curzon, G. Anxiogenic-
like Effects of mCPP and TFMPP in Animal Models are Opposed
by 5-HT1C Receptor Antagonists. Eur. J . Pharmacol. 1989, 164,
445-454.
(7) Kahn, R. S.; Wetzler, S. mCPP as a Probe of Serotonin Function.
Biol. Psychiat. 1991, 30, 1139-1166.
(8) (a) Kennett, G. A. 5-HT1C Receptor Antagonists have Anxiolytic-
like Actions in the Rat Social Interaction Model. Psychophar-
macology 1992, 107, 379. (b) Kennett, G. A.; Pittaway, K.;
Blackburn, T. P. Evidence that 5-HT1C Receptor Antagonists are
Anxiolytic in the Geller-Seifter Model of Anxiety. Psychophar-
macology 1994, 114, 90-96.
(9) Kennett, G. A.; Wood, M. D.; Bright, F.; Cilia, J .; Piper, D. C.;
Gager, T.; Thomas, D.; Baxter, G. S.; Forbes, I. T.; Ham, P.;
Blackburn, T. P. In Vitro and In Vivo Profile of SB-206553, A
Potent 5-HT2C/5-HT2B Receptor Antagonist with Anxiolytic-like
Properties. Br. J . Pharmacol. 1996, 117, 427-434.
(10) Wojciechowski, K.; Makosza, M.; Vicarious Nucleophilic Substi-
tution in 5- and 6-Nitroindole Derivatives. Synthesis 1989, 106-
109.
(11) McDonald, B. G.; Proctor, G. R. Conversion of 2-Chloroally-
lamines into Heterocyclic Compounds. Part 1. J . Chem. Soc.,
Perkin Trans. 1 1975, 1446-1450.
(12) Nordlander, J . E.; Catalane, D. B.; Kotian, K. D.; Stevens, R.
M.; Haky, J . E. Synthesis of Indoles from N-(Trifluoroacetyl)-
2-anilino Acetals. J . Org. Chem. 1981, 46, 778-782.
(13) Rahman, L. K. A.; Scrowston, R. M. 7-Substituted Benzo[b]-
thiophenes and 1,2-Benzisothiazoles. Part 2. J . Chem. Soc.,
Perkin Trans. 1 1984, 385-390.
(16) Unpublished results from the Department of Drug Metabolism
and Pharmacokinetics, SmithKline Beecham.
(17) Marshall, G. R.; Barry, C. D.; Bosshard, H. E.; Dammkoehler,
R. A.; Dunn, D. A. The Conformational Parameter in Drug
Design: The Active Analog Approach. In Computer-Assisted
Drug Design; Olson, E. C., Christoffersen, R. E., Eds.; American
Chemical Society: Washington, DC, 1979; pp 205-226.
(18) Forbes, I. T.; Ham, P.; J ones, G. E.; WO Patent 95/01976, J an.
19, 1995.
(19) J ulius, D.; MacDermott, A. B.; Axel, R.; J essell, T. M. Molecular
Characterization of a Functional cDNA Encoding the Serotonin
1c Receptor. Science 1988, 241, 558-564.
(20) Blaney, F. E.; Tennant, M. Computational Tools and Results in
the Construction of G Protein-coupled Receptor Models. In
Membrane Protein Models; Findlay, J ., Ed.; Bios Scientific Pub-
lishing: Oxford, 1996.
(21) Thompson, J . D.; Higgins, D. G.; Gibson, T. J . CLUSTAL W:
Improving the Sensitivity of Progressive Multiple Sequence
Alignment through Sequence Weighting, Positions-specific Gap
Penalties and Weight Matrix Choice. Nucleic Acids Res. 1994,
22, 4673-4680.
(22) Donnelly, D.; J ohnson, M. S.; Blundell, T. L.; Saunders, J . An
Analysis of the Periodicity of Conserved Residues in Sequence
Alignments of G Protein-coupled Receptors - Implications for the
3 Dimensional Structure. FEBS Lett. 1989, 251, 109-116.
(23) Blundell, T.; Barlow, D.; Borkakoti, N.; Thornton, J . Solvent-
induced Distortions and the Curvature of R-Helices. Nature
1983, 306, 281-283.
(24) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J .;
Swaminathan, S.; Karplus, M. CHARMM. A Program for Mac-
romolecular Energy, Minimisation and Dynamics Calculations.
J . Comput. Chem. 1983, 4, 187-217.
(25) Trumpp-Kallmeyer, S.; Hoklack, J .; Bruinvels, A.; Hibert, M.
Three-dimensional Models of Neurotransmitter G-Binding Pro-
tein-coupled Receptors. Mol. Pharmacol. 1991, 40, 8-15.
(26) Kristiansen, K.; Dahl, S. G. Molecular Modeling of Serotonin,
Ketanserin, Ritanserin and their 5-HT2C Receptor Interactions.
Eur. J . Pharmacol. 1996, 306, 195-210.
(27) Baxter, G.; Kennett, G.; Blaney, F.; Blackburn, T. 5-HT2 Receptor
Subtypes: A Family Re-united? Trends Pharmacol. Sci. 1995,
16, 105-110.
(14) Batcho, A. D.; Leimgruber, W. Indoles from 2-Methylnitroben-
zenes by Condensation with Formamide Acetals Followed by
Reduction. Organic Syntheses; Wiley: New York, 1990; Collect.
Vol. VII, pp 34-41.
(15) Kawase, Y.; Takata, S.; Hikishima, E. Synthesis of Nitro, Amino
and Acetamido Derivatives of 2,3-Dimethylbenzofuran. Bull.
Chem. Soc. J pn. 1971, 44, 749-753.
(28) The naming convention used here is that the first number refers
to the helix and the next two numbers refer to the position in
that helix. Thus Ser-312 is the twelfth residue in helix 3.
J M960571V