Conformationally restricted homotryptamines. part 7: 3- cis -(3-aminocyclopentyl)indoles as potent selective serotonin reuptake inhibitors

Article abstract of DOI:10.1021/jm100515z

A series of conformationally restricted homotryptamines has been synthesized and shown to be potent inhibitors of hSERT. Conformational restriction of the homotryptamine side chain was attained by the insertion of a cyclopentyl ring, with the indole ring

Full text of DOI:10.1021/jm100515z

7564 J. Med. Chem. 2010, 53, 7564–7572
DOI: 10.1021/jm100515z
Conformationally Restricted Homotryptamines. Part 7: 3-cis-(3-Aminocyclopentyl)indoles
As Potent Selective Serotonin Reuptake Inhibitors
H. Dalton King,* Zhaoxing Meng, Jeffrey A. Deskus, Charles P. Sloan, Qi Gao, Brett R. Beno, Edward S. Kozlowski,
Melissa A. LaPaglia, Gail K. Mattson, Thaddeus F. Molski, Matthew T. Taber, Nicholas J. Lodge,
Ronald J. Mattson, and John E. Macor
Bristol Myers Squibb, 5 Research Parkway, Wallingford, Connecticut 06492, United States
Received April 27, 2010
A series of conformationally restricted homotryptamines has been synthesized and shown to be potent
inhibitors of hSERT. Conformational restriction of the homotryptamine side chain was attained by the
insertion of a cyclopentyl ring, with the indole ring and the terminal dialkylamino group occupying the
1- and 3-positions, respectively. Nitrile and fluoro substitutions at the indole 5-position gave highest hSERT
potency. Preferred cyclopentane ring stereochemistry in both series was cis (1S,3R for 5-CN compound 8a,
1R,3S for 5-F compound 9a). High hSERT binding affinity was observed for 8a and 9a (0.22 and 0.63 nM,
respectively). The corresponding trans isomers were 4-9 times less potent. 8a, dosed at 1 and 3 mg/kg po,
produced a robust, dose-dependent increase in extracellular serotonin in the frontal cortex of rats, similar to
that induced by paroxetine at 5 mg/kg, po. By contrast, 9a did not produce a significant increase in
extracellular serotonin in rat frontal cortex at 3 mg/kg po due to relatively low brain and plasma levels.
Introduction
terminal amino group, was accomplished via the incorporation of
a 1,2-substituted cyclopropyl6 or cyclopentyl group10 as repre-
sented by 2 and 3, respectively. In these examples, the conforma-
tionally restricted analogues were more potent in vitro than the
corresponding straight chain homotryptamine 1by 4-15-fold. In
vivo, both 2 and 3 produced dose-dependent increases in cortical
serotonin levels consistent with serotonin reuptake inhibition.11
In an effort to further refine the conformational and stereo-
chemical requirements of the most potent homotryptamine
SSRI’s, several alternate conformationally restricted homo-
typtamine side chains were considered. We now report the
1,3-cyclopentane series 4, in which the amino group is directly
attached to the ring, unlike 2 and 3, which utilized the
exocyclic aminomethyl group.
The serotonergic system constitutes an important target for
psychoactive drugs. The human serotonin transporter (hSERTa)
is responsible for the regulation of synaptic serotonin (5-hydro-
xytryptamine, 5-HT) levels. Selective serotonin reuptake inhibi-
tors (SSRIs) act by blocking the reuptake of serotonin into the
presynaptic neuron, thus increasing the effective concentration
of this neurotransmitter to enervate the postsynaptic 5-HT
receptors. SSRIs have found utility primarily in depression
disorders, although other indications including anxiety have
been developed as well.1
Although many clinically useful SSRIs have been discovered,2
the development of compounds with higher affinity and target
selectivity continues to be an area of interest.3 Our own efforts to
achieve this have concentrated on a number of homotryptamine
analogues of serotonin.4-10 Ranging from analogues with simple
flexible side chains4,5 to those with conformationally restricted
elements,6-10 homotryptamines have been shown to be potent
inhibitors of hSERT. In these series, conformational restriction
of the C3 spacer, which separates the indole nucleus and the
*To whom correspondence should be addressed. Phone: (203) 677-
6685. Fax: (203) 677-7702. E-mail: dalton.king@bms.com.
a Abbreviations used: hSERT, human serotonin transporter; 5-HT,
serotonin or 5-hydroxytryptamine; SSRI, selective serotonin reuptake
inhibitor; hDAT, human dopamine transporter; hNET, human norepi-
nephrine transporter.
Results
To enable a rapid assessment of R1 substituent effects on
hSERT affinity, we initially synthesized the racemic/dia-
stereomeric mixtures 6-17 (Table 1). Previous experience
r
pubs.acs.org/jmc
Published on Web 10/15/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7565
with other homotryptamine analogues suggested that this
approach would provide adequate guidance concerning
overall SAR trends. We chose to concentrate on com-
pounds with alkyl groups R2 and R3 = CH3, knowing that
these would likely yield compounds with relatively high
hSERT potency.5,6,10
potency. Increasing the size of the R2 and/or R3 substituents
beyond methyl in compounds 20-32 also decreased potency
relative to 8 (Table 2). Although we examined a more limited
set of R2 and R3 substituents in the similar 5-fluoro series
33-35 (Table 3), the same trend was observed relative to 9.
Therefore, in both series, dimethyl substitution (i.e., R2=
R3=Me) of the terminal side chain amine was optimal for
Michael addition of R1-substituted indoles to cyclopent-2-
enone in the presence of the Lewis acid Yb(OTf)3 yielded
racemic ketone 5 (Scheme 1).12 In the case of R1 = 5-CN,
longer reaction times were necessary because of the presence of
the strong electron withdrawing group. Reductive amination of
5 with dimethylamine then afforded 6-17 in quantitative yields.
Confirming a trend we have observed in other homotryp-
tamine analogues,5,6,10 5-CN substitution yielded 8, a sub-nM
inhibitor of hSERT and the most potent analogue examined
(Table 1). Halogen substitution at alternate indole positions
(6-7, 9-17) failed to improve on this result, although appre-
ciable potency was maintained. On the basis of these results,
we decided to explore analogues of both 8 and 9 in an effort to
further improve hSERT potency via optimization of the
dialkylamino side chain.
Table 2. hSERT Binding Affinities of Racemic 5-Cyano-3-(3-dia-
lkylaminocyclopentyl)-indole Diastereomers
A representative group of side chain alkyl group substitu-
tions (R2 and R3) in the 5-CN and 5-F series was synthesized
as shown in Scheme 1. Results for the racemic/diastereomeric
mixtures are listed in Tables 2 and 3, respectively. Relative to
8, removal of both methyl groups in 18 and a single methyl
group in 19 resulted in a significant decrease in hSERT
Table 1. hSERT Binding Affinities of Racemic 3-(3-Dimethylamino-
cyclopentyl)-indole Diastereomers
Table 3. hSERT Binding Affinities of Racemic 5-Fluoro-3-(3-dia-
lkylaminocyclopentyl)-indole Diastereomers
compd
R1
hSERT IC50 (nM ( SEM)
6
4-F
31 ( 4
7
4-Br
5-CN
5-F
5.3 ( 1.2
0.48 ( 0.18
1.7 ( 0.2
3.0 ( 0.4
4.3 ( 1.4
5.1 ( 1.1
6.5 ( 3.0
9.6 ( 4.4
3.3 ( 1.5
1.4 ( 0.3
1.6 ( 0.7
8
9
10
11
12
13
14
15
16
17
5-Cl
5-Br
6-F
compd
R2
R3
hSERT IC50 (nM ( SEM)
6-Cl
6-Br
7-F
9
Me
Me
Et
Me
Et
1.7 ( 0.2
3.7 ( 1.4
7.3 ( 1.2
2.9 ( 0.9
33
34
35
7-Cl
7-Br
Et
-((CH2)4)-
Scheme 1. Racemic Synthesis of Diastereomeric 3-(3-Dialkylaminocyclopentyl)-indoles

7566 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
King et al.
hSERT potency, as has been the case in the other homotryp-
tamine series we have reported.5,6,10
To determine the absolute stereochemistry of 8a and 9a and
to produce larger quantities for in vivo testing than available
via the chiral preparative HPLC method, it was necessary to
devise a stereospecific synthesis of each. We were unable to
find a general method applicable to both compounds, there-
fore, the individual approaches are outlined in the following
section.
Synthesis and Absolute Stereochemistry of 8a. Large-scale
synthesis of 8a was accomplished by the method outlined
in Scheme 2. Racemic ketone racemic-5a (compd 5, R1 =
5-CN) was resolved by partial enzymatic reduction using
ketoreductase KRED-1004 to yield multigram quantities of
the easily separable mixture of chiral ketone S-5a and
alcohol 36.13 The absolute configuration of the chiral ketone
was determined by X-ray crystallography of its N-3,4-di-
chlorobenzoyl derivative 37 (Figure 1). Chiral ketone S-5a
was then subjected to reductive amination with dimethyla-
mine and the mixture of diastereomers now efficiently sepa-
rated by chiral preparative HPLC. The cis isomer was
determined by NMR, and its correlation to the original
isolate 8a was confirmed by all analytical methods. Thus,
the configuration of the most potent isomer 8a was deter-
mined as 1S,3R. By analogy, the other isomers of 8 were
identified as shown in Table 4.
On the basis of our prior experience with other stereoiso-
meric homotryptamines,6,10 we believed that one of the
various stereoisomers of 8 and 9 would preferentially bind
to hSERT. To obtain all four possible stereoisomers of 8, we
initially obtained working quantities of the ketones S- and
R-5a (compd 5, R1 = 5-CN) by chiral preparative HPLC
resolution of racemic-5a (Scheme 1). After standard reductive
amination of S- and R-5a with dimethylamine, the two sets of
diastereomers were resolved by chiral preparative HPLC to
give 8a-d (Table 4). An identical procedure was followed
starting with racemic-5b (compd 5, R1 = 5-F) to give the
fluoro-series 9a-d. A slight potency advantage was observed
in the 5-CN series for isomer 8a, which inhibited hSERT with
an IC50 of 0.22 nM (Table 4). In the 5-fluoro series, 9a was the
most potent stereoisomer of the 5-fluoro series, although still
3-fold less potent than 8a in the 5-CN series.
Table 4. hSERT Binding Affinities of Stereoisomers of 8 and 9
An alternate synthetic approach which avoids preparative
chiral HPLC is outlined in Scheme 3. Starting from chiral
ketone S-5a, the mixture of diastereomers 38 and 39 obtained
by reductive alkylation with N-methyl-N-benzylamine fol-
lowed by N1-BOC protection was separated by silica gel
chromatography. 38 was then debenzylated and realkylated
with formaldehyde/sodium cyanoborohydride under reduc-
tive conditions and deprotected to yield 8a in good yield,
identical in all respects to that obtained by the other methods.
Synthesis and Absolute Stereochemistry of 9a. In the 5-F
series, we found it most convenient to obtain working
quantities of chiral ketone S-5b via the procedure outlined
in Scheme 4. Condensation of (1S,2S)-1,2-diphenylethane-
1,2-diol with racemic-5b (compd 5, R1= F) yielded a mixture
compd
cis/trans
configuration
hSERT IC50 (nM ( SEM)
8a
8b
8c
8d
9a
9b
9c
9d
cis
trans
cis
1S,3R
1S,3S
1R,3S
1R,3R
1R,3S
1S,3S
1S,3R
1R,3R
0.22 ( 0.10
0.97 ( 0.22
0.38 ( 0.07
1.3 ( 0.7
0.63 ( 0.21
5.9 ( 1.2
trans
cis
trans
cis
trans
2.4 ( 0.4
2.8 ( 0.5
Scheme 2. Stereospecific Synthesis of 8aa
a (a) Ketoreductase KRED-1004, NADPH, acetone; (b) 3,4-dichlorobenzoyl chloride, diisopropylethylamine, 76%; (c) dimethylamine, sodium
triacetoxyborohydride; (d) preparative chiral LC.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7567
Figure 1. ORTEP drawing of 37 and 40 with thermal ellipsoids at 30% (37) and 25% (40) probability for non-H atoms and open circles for
H-atoms.
Scheme 3. Alternate Stereospecific Synthesis of 8aa
a (a) N-Methyl-N-benzylamine, sodium cyanoborohydride; (b) di-t-butyldicarbonate, DMAP, triethylamine; (c) a-chloroethylchloroformate,
MeOH; (d) formaldehyde, sodium cyanoborohydride; (e) TFA
Scheme 4. Stereospecific Synthesis of 9aa
a (a) (1S,2S)-1,2-diphenylethane-1,2-diol, p-tosic acid, 17%; (b) HCl/MeOH, 63%; (c) dimethylamine, sodium triacetoxyborohydride; (d) preparative
chiral LC.
of two diastereomeric ketals. 40 was obtained by methanol
recrystallization in 86.8% de and 17% yield.14 The absolute
configuration of 40 was established by X-ray crystallography
as 2S,3S,7S (Figure 1). Acid hydrolysis then gave chiral
ketone S-5b, which was subjected to reductive amination and
preparative chiral HPLC. The cis isomer obtained by this
method, 9a, was established by NMR and correlated to the
original material by all analytical methods. The absolute
configuration of 9a was determined as 1R,3S. By analogy,
absolute configurations of the other three isomers 9b-d were
assigned (Table 4).
discovered thus far are the 5-CN analogues 2 and 3, with hSERT
binding affinity IC50 values of 0.48 and 0.13 nM, respectively.
In the current series, we maintained the C3 side chain common
to the homotryptamines but arranged the indole and the tertiary
amine in a 1,3 substitution pattern on a cyclopentane backbone.
Several members of this series exhibited high affinity binding to
hSERT, with the 5-CN and 5-F analogues again being
particularly active. Resolution of all enantiomers allowed
the identification of 8a and 9a, with hSERT binding affinity
IC50 values of 0.22 and 0.63 nM, respectively, as the most
potent of the series. These compounds shared the same cis-
stereochemistry around the cyclopentane ring, ie., S at the
indole and R at the dimethylamino group. They were 2-4
times more potent than the other respective cis enantio-
podes and 4-9 times more potent than either of the
respective trans enantiomers.
Discussion
In a series of previous communications, we have identified a
number of homotryptamines with high binding affinity for
hSERT.4-10 These studies have shown that 5-indole substitution
is generally optimal, with 5-CN and 5-F being the most active at
inhibiting hSERT. The most potent homotrytamine SSRI’s
To examine the relationship of putative binding conforma-
tions of 3 and 8a with established SSRIs such as sertraline and

7568 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
King et al.
Figure 2. (a) Model shows superposition of SSRIs sertraline ((1S,4S)-stereochemistry, yellow carbon atoms), S-citalopram (green carbon
atoms), compound 2 ((þ)-12a from reference 6 (orange carbon atoms)), and compounds 3 and 8a (gray carbon atoms). (b) Same model as
shown in (a) with sertraline and S-citalopram removed to highlight similarities between possible bioactive conformations of compounds 2, 3,
and 8a. Magenta spheres represent centroid atoms and putative hydrogen bond acceptor site points used in the rms fitting procedure.
Table 5. Binding Affinities of 8a and 9a versus Selected Off-Target
S-citalopram and the previously reported potent 1,2-substi-
tuted cyclopropyl homotryptamine 2,6 a molecular modeling
study was undertaken. This revealed that low-energy confor-
Receptors
IC50 (nM ( SEM)a
mers of compounds 3 and 8a can be superimposed reasonably
well with low-energy conformers of sertraline, S-citalopram,
and compound 2 (Figure 2). The conformers of sertraline,
S-citalopram, and 2 were identified and superimposed as
described previously.6 The conformers of compounds 3 and
8a were identified via similar methods described in ref 15. The
energy of the conformer of compound 3 shown in Figure 2 is
1.5 kcal/mol above that of the lowest energy conformer
identified, and the conformer of compound 8a shown is the
lowest energy conformer found for that compound. The
conformers of compounds 3 and 8a were each superimposed
with sertraline using three corresponding points for an rms-
fitting procedure: the aromatic ring centroids, the basic nitro-
gen atoms and putative hydrogen-bond acceptor site points
target
hSERT
hNET
8a
9a
0.22 ( 0.10
6900
1300
0.63 ( 0.21
2100
1100
hDAT
5HT1a
5HT1b
5HT2a
5HT2b
5HT2c
340b
500
840
28000
>10000
>10000
NTc
>10000
>10000
>10000
>30000
>1000
37b
5HT4
>30000
NT
NT
5HT5a
5HT6
5HT7
>1000
7200
>30000
NT
25000
>30000
dopamine D1
dopamine D2
˚
located 2.8 A from the nitrogen atoms in the directions of the
a All data n = 1, except where indicated. b n = 2. c Not tested.
N-H bonds. As shown in Figure 2a, superposition in this
manner results in reasonable overlap of the substituted aro-
matic rings and putative hydrogen-bond acceptor site points
for all five SSRI compounds. Note, however, that the sub-
stituted aromatic rings deviate significantly from coplanarity,
and the relative positions of the ring substituents differ as well.
These observations suggest that although compounds 3 and 8a
fit the basic SSRI pharmacophore as represented by sertraline
and S-citalopram, the SAR for substitution on the aromatic
rings may be somewhat different for the homotryptamines and
analogues of sertraline and S-citalopram. In Figure 2b, for
clarity, only compounds 2, 3, and 8a are shown. Despite the
variation in the linker moieties connecting the basic amines to
the indole rings, there is good correspondence between the
aromatic rings and putative hydrogen-bond acceptor site
points, indicating that the different homotryptamine com-
pounds interact with the serotonin transported in a similar
manner. Finally, it is interesting to note that the potency of
compound 8a is within experimental error of that determined
for its enantiomer 8c. Although not shown in Figure 2, 8c can
also adopt a conformation which fits the SSRI pharmacophore
and differs from the conformer of 8a shown primarily in the
region of the cyclopentyl linker.
Figure 3. Effects of po administration of 8a on extracellular sero-
tonin levels in the frontal cortex of rats measured by microdialysis.
10000 versus hSERT. Both compounds were also highly
selective against most of the other serotonin and dopamine
receptors listed in Table 5. In the single exception observed, 8a
was 200-fold less selective for 5HT6 than hSERT.
To confirm selectivity for the serotonin transporter, 8a and
9a were evaluated for binding affinity at the human dopamine
(hDAT) and norepinephrine (hNET) transporters (Table 5).
Neither of these compounds was a significant inhibitor of
hDAT or hNET, with selectivity ratios ranging from 1000 to
As a demonstration of the ability of these compounds to
produce functional changes in serotonin levels in vivo, micro-
dialysis experiments were carried out in rats.6 Oral dosing of 8a
produced a robust, dose-dependent increase in extracellular
serotonin in the frontal cortex (Figure 3). Doses of both 1 and 3

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7569
These compounds were found to be 2-4 times more potent
than their other respective cis enantiopodes, and 4-9 times
more potent than either of their respective trans enantiomers.
Both compounds were highly selective relative to other neuro-
transporters in vitro.
While the in vitro properties of8aand 9a weresimilar, the in
vivoeffectsdifferedbetween the twocompounds. 8aproduced
a robust, dose-dependent increase in extracellular serotonin in
the rat frontal cortex at 1 and 3 mg/kg po, similar to the SSRI
paroxetine. In contrast, no effect was observed for 9a at 3 mg/
kg po, although an effect similar to 8a was observed when 9a
was dosed at 1 mg/kg iv. This lack of oral activity is consistent
with the relatively poor pharmacokinetic profile of 9a, which
failed to attain suitable brain concentrations under oral
administration.
With the conclusion of the current series, it was informative
to compare the relative binding modes of other conforma-
tionally restricted homotryptamines with SSRIs. Molecular
models suggested that sertraline, S-citalopram, 2, 3, and 8a
may be able to interact with the serotonin transporter binding
site in a similar manner. Furthermore, there was good corre-
spondence between the aromatic rings and putative hydrogen-
bondacceptorsite pointsin2, 3, and 8a despite the variationin
the cyclic structures connecting the basic amine to the indole
ring. This proves that the homotryptamine side chain can be
locked into a common favorable hSERT binding confor-
mation, leading to higher in vitro and in vivo potency via a
number of different conformationally restricted orientations.
Figure 4. Effects of po and iv administration of 9a on extracellular
serotonin levels in the frontal cortex of rats measured by micro-
dialysis.
Table 6. Pharmacokinetic Parameters Measured in Rat Microdialysis
Experimentsa
brain conc,
nM
plasma conc,
nM
compd
Fpo, %
Tmax, h
T1/2, h
8a
9a
1000
53
820
37
51
nd
1.0
nd
0.55
nd
a Compounds 8a and 9a were administered at 3 mg/kg po. Brain and
plasma concentrations were measured at 2 h.
mg/kg po produced effects which were significantly greater
than vehicle control. An increase of 150% over vehicle was
observed at the maximum dose tested, 3 mg/kg. This effect was
similar to that seen for paroxetine dosed at 5 mg/kg po.
By contrast, 9a did not produce a significant increase in
extracellular serotonin in the frontal cortex when dosed in rats
at 3 mg/kg po (Figure 4). This result, in contrast to that of 8a,
can best be understood in terms of their relative pharmacoki-
netic properties. Compound 9a demonstrated low bioavail-
ability in rats when dosed orally at 3 mg/kg; brain and plasma
levels of drug measured at2 h were, respectively, 53 and 37 nM
(Table 6). By contrast, brain and plasma levels of 8a at 2 h
after 3 mg/kg oral dosing were 1000 and 820 nM, respectively
Thus, brain and plasma exposures were 20-fold higher for 8a
versus 9a. Oral bioavailability of 8a was 51%, with a Tmax of
1 h and a T1/2 of 0.55 h in the rat. Additionally, when dosed at
1.0 mg/kg iv, 9a didproduce a robust increase in frontal cortex
serotonin levels, similar to that observed with oral dosing of
8a. This result further confirmed the poor oral bioavailability
of 9a.
Experimental Section
Chemistry. NMRs were recorded on Bruker Avance spectrom-
eters. Elemental analyses were performed at Robertson Microlit
Laboratories, Madison, NJ. Exact mass determinations were made
using a Micromass LCT unit, and all samples were determined
using a TOF ESI (þ) source. ORTEP drawings of the X-ray crystal
structures for 37 and 40 are given in Figure 1, with thermal
ellipsoids at 30% (37) and 25% (40) probability for non-H atoms
and open circles for H-atoms. Full crystallographic data have been
deposited to the Cambridge Crystallographic Data Center (CCDC
793899 and 793900). Copies of the data can be obtained free of
charge via the Internet at http://www.ccdc.cam.ac.uk. Purity of all
compounds was determined to be >95% by either combustion
analysis or analytical HPLC using at least two perpendicular
methods as listed in the Supporting Information. Chiral purity
was determined by chiral HPLC using one of the methods listed in
the Supporting Information.
Specific details related to the synthesis of compounds 5-35
are included in the Supporting Information.
General Procedure for the Synthesis of 3-(1H-Indol-3-yl)cyclo-
pentanones 5. 2-Cyclopenten-1-one (4.1 g, 4.2 mL, 50 mmol) was
added to a stirred solution of the substituted indole (10 mmol)
and ytterbium triflate hexahydrate (124 mg, 0.2 mmol) in
acetonitrile (15 mL). After stirring at room temperature for
18 h to 7 days, the reaction was concentrated to an oil and
diluted with ether. The red oily mixture was filtered through
Celite, and the filtrate was concentrated in vacuo. The crude
product was purified by flash chromatography on silica gel
using a gradient of ethyl acetate in hexanes. Pure product
fractions were concentrated and dried under high vacuum to
give the 3-(1H-indol-3-yl)cyclopentanones.
General Procedure for the Synthesis of 6-17. The appropri-
ately substituted 3-(1H-indol-3-yl)cyclopentanone (0.5 mmol)
and dimethylamine (2.0 M solution in THF, 5.0 mmol) were
dissolved in ethanol to a final volume of 5 mL. After stirring for
15 min, sodium triacetoxyborohydride (430 mg, 2.0 mmol) was
added and the reaction stirred for 3 h to 3 days. The reaction was
then diluted with water (10 mL) and extracted three times with
Conclusion
In this report we presented a series of novel conformation-
ally restricted homotryptamines and demonstrated their
hSERT activity. Conformational restriction was imparted
by arranging the indole and alkylamino groups in a 1,3-
substitution pattern on a cyclopentane ring spacer. Nitrile
and fluoro substituents at the C5-position of the indole
provided compounds with highest hSERT affinities, although
compounds with halogen substituents at other positions on
the indole were active as well. Additionally, N,N-dimethyl
substitution of the cyclopentane amino group was found to be
the optimal alkylation pattern for the amine, consistent with
previous findings. Resolution of the most active enantiomers
in both series provided the highly potent hSERT inhibitors 8a
and 9a, which shared the same cyclopentane cis-stereochem-
istry, i.e., S at the indole and R at the dimethylamino group.

7570 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
King et al.
ethyl acetate (10 mL). The organic extracts were dried over
sodium sulfate and concentrated in vacuo. The residue was
purified by preparative reverse phase HPLC to give the product
as an oily trifluoroacetic acid salt of the 4 cis/trans diastereo-
mers. Where indicated, the free base was isolated by extraction
of the TFA salt from saturated sodium carbonate solution with
ethyl acetate.
3-[(1R,3S)-3-Dimethylaminocyclopentyl]-1H-indole-5-carbo-
nitrile 8c. H NMR δ (500 MHz, CD3OD) 8.03 (1 H, d, J =
0.92), 7.46 (1 H, d, J = 8.55), 7.36 (1 H, dd, J = 8.24, 1.53), 7.25
(1 H, s), 3.38 (1 H, m), 2.86 (1 H, m), 2.43 (1 H, m), 2.38 (6 H, s),
2.23 (1 H, m), 2.09 (1 H, m), 1.86 (1 H, m), 1.78 (1 H, m), 1.68
(1 H, m). [R]25 -8.12 (589 nm, c 1.71 mg/mL, EtOH). Analytical
chiral HPLC (method J) tR 9.7 min.
1
3-[(1R,3R)-3-Dimethylaminocyclopentyl]-1H-indole-5-carbo-
nitrile 8d. 1H NMR (500 MHz, CD3OD) δ 8.01 (1 H, s), 7.47 (1 H,
d, J = 8.24), 7.37 (1 H, dd, J = 8.24 1.52), 7.23 (1 H, s), 3.53 (1 H,
m), 3.10 (1 H, m), 2.47 (6 H, s), 2.29 (1 H, m), 2.19 (2 H, m), 2.07
(1 H, m), 1.84 (1 H, m), 1.73 (1 H, m). [R]25 þ13.99 (589 nm, c 1.5
mg/mL, EtOH). Analytical chiral HPLC (method J) tR 8.6 min.
Synthesis of 9a-d via Chiral HPLC Resolution of Stereo-
isomers. The S- and R-enantiomers of 3-(5-fluoro-1H-indol-
3-yl)-cyclopentanone 5b were resolved by chiral HPLC on
a Chiral Technologies Chiralcel OD column (20 μ, 50 mm ꢀ
500 mm) using a mobile phase of 15% isopropyl alcohol/
hexanes. Flow rate was 75 mL/min. The first isomer to elute
was S-3-(5-fluoro-1H-indol-3-yl)-cyclopentanone S-5b; analy-
tical chiral HPLC (method K) tR 21.0 min. The second isomer to
elute was R-3-(5-fluoro-1H-indol-3-yl)-cyclopentanone R-5b;
analytical chiral HPLC (method K) tR 26.7 min.
A solution of S-5b (290 mg, 1.34 mmol) and dimethylamine
(2.0 M solution in THF, 6.7 mL, 13.4 mmol) in ethanol (10 mL)
was stirred for 15 min. Sodium triacetoxyborohydride (1.1 g,
5.4 mmol) was added and the reaction stirred for 1 h. The reac-
tion was extracted three times with ethyl acetate/aqueous so-
dium bicarbonate solution. The ethyl acetate extracts were dried
over magnesium sulfate and concentrated in vacuo to give [(3S)-
3-(5-fluoro-1H-indol-3-yl)-cyclopentyl]-dimethylamine (400 mg,
100%) as a cis/trans diastereomeric mixture of 9a and 9b. The 1R
and 1S diastereomers were resolved by preparative chiral HPLC
on a Chiral Technologies Chiralpak AD column (20 μ, 50 mm ꢀ
500 mm) with a mobile phase of 10% ethanol in hexanes/0.1%
diethylamine at a flow rate of 60 mL/min.
General Procedure for the Synthesis of 18-35. The indicated
3-(1H-indol-3-yl)cyclopentanone (0.5 mmol) and amine (5.0
mmol) were dissolved in ethanol to a final volume of 5 mL.
After stirring for 15 min, sodium triacetoxyborohydride
(430 mg, 2.0 mmol) was added and the reaction stirred for 3 h
to 3 days. The reaction was then diluted with water (10 mL) and
extracted three times with ethyl acetate (10 mL). The organic
extracts were dried over sodium sulfate and concentrated in
vacuo. The residue was purified by preparative reverse phase
HPLC to give the product as an oily trifluoroacetic acid salt of
the 4 cis/trans diastereomers. Where indicated, the free base was
isolated by extraction of the TFA salt from saturated sodium
carbonate solution with ethyl acetate.
Synthesis of 8a-d via Chiral HPLC Resolution of Stereo-
isomers. The S- and R-enantiomers of 3-(3-oxocyclopentyl)-1H-
indole-5-carbonitrile 5a were resolved by chiral HPLC on
a Chiral Technologies Chiralcel OD column (20 μ, 50 mm ꢀ
500 mm) using a mobile phase gradient of ethanol/hexanes
(10-100% containing 0.01% diethylamine). Flow rate was
varied over the gradient from 60 to 50 mL/min. The first isomer
to elute was 3-(1S-3-oxocyclopentyl)-1H-indole-5-carbonitrile
S-5a ([R]25 -24.4 (589 nm, c 2.62 mg/mL, MeOH); analytical
chiral HPLC (method I) tR 10.8 min. The second isomer to elute
was 3-(1R-3-oxocyclopentyl)-1H-indole-5-carbonitrile R-5a
([R]25 þ10.5 (589 nm, c 2.64 mg/mL, EtOH); analytical chiral
HPLC (method I) tR 12.5 min.
S-5a (112 mg, 0.5 mmol) and dimethylamine (2.0 M solution
in THF, 2.5 mL, 5.0 mmol) were dissolved in ethanol (2 mL).
After stirring for 15 min, sodium triacetoxyborohydride
(424 mg, 2.0 mmol) was added and the reaction continued for
2 h. The reaction was then diluted with water (5 mL) and made
acidic (pH 3) with 6 M HCl. The reaction was then adjusted to
pH 10 with sodium carbonate. It was extracted two times with
ethyl acetate (50 mL), and the extracts were dried over sodium
sulfate, concentrated in vacuo, and dried under high vacuum to
give 3-[(1S)-3-dimethylaminocyclopentyl]-1H-indole-5-carbo-
nitrile (125 mg, 100%) as a cis/trans diastereomeric mixture of
8a and 8b. The 3R and 3S diastereomers were resolved by
preparative chiral HPLC on a Chiral Technologies Chiralpak
AD column (20 μ, 50 mm ꢀ 500 mm) with a mobile phase of
10% ethanol in hexanes/0.1% diethylamine at a flow rate of
75 mL/min.
[(1R,3S)-3-(5-Fluoro-1H-indol-3-yl)-cyclopentyl]-dimethylamine
9a. 1H NMR (500 MHz, CD3OD): δ 7.27 (dd, J = 9.0, 4.5 Hz,
1H), 7.24 (dd, J = 8.7, 4.2 Hz, 1H), 7.21 (s, 1H), 6.86 (dt, J = 9.3,
2.4 Hz, 1H), 3.35 (m, 1H), 3.18 (m, 1H), 2.58 (s, 6H), 2.47 (m, 1H),
2.18 (m, 2H), 1.86 (m, 2H), and 1.75 (q, J = 10.5 Hz, 1H). FIMS:
m/z 247.4 (M þ H)þ; m/z 245.4 (M - H)-. [R]25 þ2.54 (589 nm,
c 2.79 mg/mL, EtOH). Analytical HPLC (method H): tR, 2.38 m,
>99% purity; analytical chiral HPLC (method L) >98% ee,
tR 15.7 min.
[(1S,3S)-3-(5-Fluoro-1H-indol-3-yl)-cyclopentyl]-dimethylamine
9b. 1H NMR (500 MHz, CD3OD): δ 7.28 (dd, J = 8.7, 4.5 Hz,
1H), 7.22 (dd, J = 10.2, 2.4 Hz, 1H), 7.14 (s, 1H), 6.86 (dt, J = 2.4
Hz, 1H), 3.68 (t, 1H), 3.55 (m, 1H), 2.84 (s, 6H), 2.28 (m, 4H), and
1.88 (m, 2H). FIMS: m/z 247.4 (M þ H)þ; m/z 245.4 (M - H)-.
[R]25 -13.54 (589 nm, c 3.07 mg/mL, EtOH). Analytical HPLC
(method H): tR, 2.32 m, >97% purity; analytical chiral HPLC
(method L) >99% ee, tR 12.1 min.
R-5b was reacted with dimethylamine by the same procedure
on a 0.92 mmol scale to give 240 mg (100%) of [(3R)-3-(5-fluoro-
1H-indol-3-yl)-cyclopentyl]-dimethylamine as a cis/trans dia-
stereomeric mixture of 9c and 9d. The 1S and 1R diastereomers
were resolved by preparative chiral HPLC on a Chiral Techno-
logies Chiralpak AD column (20 μ, 50 mm ꢀ 500 mm) with a
mobile phase of 10% ethanol in hexanes/0.1% diethylamine at a
flow rate of 60 mL/min.
[(1S,3R)-3-(5-Fluoro-1H-indol-3-yl)-cyclopentyl]-dimethylamine
9c. 1H NMR (500 MHz, CD3OD): δ 7.26 (dd, J = 8.7, 4.5 Hz,
1H), 7.21 (dd, J = 9.9, 2.4 Hz, 1H), 7.11 (s, 1H), 6.83 (dt, J = 2.4
Hz, 1H), 3.31 (m, 1H), 2.90 (m, 1H), 2.41 (s, 6H), 2.39 (m, 1H),
2.20 (m, 1H), 2.10 (m, 1H), 1.80 (m, 2 H), and 1.68 (q, J = 10.5 Hz,
1H).FIMS:m/z248.3 (M þ H)þ;m/z245.4 (M - H)-.[R]25 -12.32
(589 nm, c 1.93 mg/mL, EtOH). Analytical HPLC (method H): tR,
2.34 m, >97% purity; analytical chiral HPLC (method L) >98%
ee, tR 14.7 min.
3-[(1S,3R)-3-Dimethylaminocyclopentyl]-1H-indole-5-carbo-
1
nitrile 8a. H NMR (500 MHz, CD3OD) δ 8.03 (1 H, d, J =
0.92), 7.46 (1 H, d, J = 8.55), 7.36 (1 H, dd, J = 8.24 1.53), 7.25
(1 H, s), 3.39 (1 H, m), 2.81 (1 H, m), 2.42 (1 H, m), 2.35 (6 H, s),
2.22 (1 H, m), 2.08 (1 H, m), 1.86 (1 H, m), 1.77 (1 H, m), 1.67
(1 H, m). [R]25 þ12.95 (589 nm, c 1.58 mg/mL, EtOH). Analy-
tical chiral HPLC (method J) tR 13 min.
3-[(1S,3S)-3-Dimethylaminocyclopentyl]-1H-indole-5-carbo-
1
nitrile 8b. H NMR (500 MHz, CD3OD) δ 7.99 (1 H, s), 7.45
(1 H, d, J = 8.54), 7.36 (1 H, dd, J = 8.24 1.52), 7.21 (1 H, s),
3.49 (1 H, m), 2.85 (1 H, m), 2.32 (6 H, s), 2.25 (1 H, m), 2.12
(2 H, m), 1.99 (1 H, m), 1.80 (1 H, m), 1.66 (1 H, m). [R]25 -26.50
(589 nm, c 1.58 mg/mL, EtOH). Analytical chiral HPLC
(method J) tR 8.4 min.
R-5a was reacted with dimethylamine by the same procedure
to give 3-[(1R)-3-dimethylaminocyclopentyl]-1H-indole-5-car-
bonitrile (125 mg, 100%) as a cis/trans diastereomeric mixture
of 8c and 8d. The 3S and 3R diastereomers were resolved by
chiral HPLC on a Chiral Technologies Chiralpak AD column
(20 μ, 50 mm ꢀ 500 mm) with a mobile phase of 10% ethanol in
hexanes/0.1% diethylamine at a flow rate of 75 mL/min.

Article
[(1R,3R)-3-(5-Fluoro-1H-indol-3-yl)-cyclopentyl]-dimethylamine
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7571
concentrated to dryness. The crude product was dissolved in 50
mL DCM, and to this was added di-t-butyldicarbonate (5.45 g,
25 mmol), DMAP (100 mg), and triethylamine (3.5 mL, 25
mmol). The reaction was stirred at room temperature for 18 h
and then washed three times with aqueous NaHCO3 solution
and twice with brine. The organic layer was dried over sodium
sulfate, concentrated to dryness, and purified by flash chro-
matography on 120 g silica gel with a gradient of 0-30%
EtOAc in hexanes (50 min). The faster eluting component was
isolated, concentrated, and dried to tert-butyl 3-((1S,3R)-
3-(benzyl(methyl)amino)cyclopentyl)-5-cyano-1H-indole-1-
carboxylate 38 (1.15 g, 54%). The slower eluting trans
component 39 was not characterized.
9d. 1H NMR (500 MHz, CD3OD): δ 7.27 (dd, J = 8.7, 4.5 Hz,
1H), 7.22 (dd, J = 9.9, 2.4 Hz, 1H), 7.11 (s, 1H), 6.84 (dt, J = 9.0,
2.4 Hz, 1H), 3.49 (t, 1H), 3.31 (m, 1H), 2.61 (s, 6H), 2.24 (m, 2H),
2.19 (q, J = 15.3, 6.9 Hz, 2 H), and 1.80 (m, 2H). FIMS: m/z 247.4
(M þ H)þ; m/z 245.4 (M - H)-. [R]25 þ14.03 (589 nm, c 1.71 mg/
mL, EtOH). Analytical HPLC (method H): tR, 2.26 m, >89%
purity; analytical chiral HPLC (method L) >99% ee, tR 13.0 min.
Enzymatic Resolution of 3-(3-Oxocyclopentyl)-1H-indole-5-
carbonitrile rac-5a and 3-(1R,3S-3-hydroxycyclopentyl)-1H-in-
dole-5-carbonitrile 36. (S)-3-(3-Oxocyclopentyl)-1H-indole-
5-carbonitrile S-5a was obtained by enzymatic resolution of
racemic 3-(3-oxocyclopentyl)-1H-indole-5-carbonitrile rac-5a
utilizing ketoreductase KRED-1004 (Biocatalytics, Inc., Pasa-
dena, CA) in the presence of isopropyl alcohol as cosubstrate
and NADPH as cofactor. The 1 L reaction mixture consisted of
10 mM potassium phosphate buffer (pH 6.0), 15% methanol,
2% isopropyl alcohol, 50 mg NADPH, 50 mg KRED-1004,
and 500 mg rac-5a in water. After incubating at 30 °C, 75 rpm
for 3 days, the reaction reached completion by HPLC analysis.
The reaction mixture was then extracted with 1 L of ethyl
acetate to afford a mixture of S-5a and 3-(1R,3S-3-hydro-
xycyclopentyl)-1H-indole-5-carbonitrile 36 (516 mg). The ke-
tone/alcohol mixture obtained from multiple runs (2.4 g) was
purified by flash chromatography on 110 g silica gel with a step
gradient of 0, 1, and 2% methanol in methylene chloride. The
two components were concentrated and dried under high
vacuum to yield S-5a (1.1 g, 46%) and 36 (0.94 g, 39%).
(S)-3-(3-Oxocyclopentyl)-1H-indole-5-carbonitrile S-5a. 1H
NMR (500 MHz, CDCl3) δ 8.38 (1 H, bs), 7.98 (1 H, s), 7.45
(2 H, m), 7.11 (1 H, dd, J = 2.44, 0.91 Hz), 3.71 (1 H, m), 2.77
(1 H, dd, J = 7.63, 18.31 Hz), 2.56 (1 H, m), 2.40 (3 H, m), 2.10
(1 H, m). MS m/e 223.2 (M - H)þ. [R]25 -22.3 (589 nm, c 1.54
mg/mL, MeOH). Analytical chiral HPLC (method I) >95% ee.
3-(1R,3S-3-Hydroxycyclopentyl)-1H-indole-5-carbonitrile 36.
1H NMR (500 MHz, CDCl3) δ 8.26 (1 H, bs), 8.04 (1 H, s), 7.40
(2 H, m), 7.15 (1 H, dd, J = 2.44, 0.91 Hz), 4.52 (1 H, m), 3.31 (1
H, p, J = 8.24), 2.55 (1 H, m), 2.15 (1 H, m), 1.98 (2 H, m), 1.83 (1
H, m), 1.76 (1 H, m). MS m/e 225.2 (M - H)þ. Anal. Calcd. for
1H NMR (400 MHz, CDCl3) δ ppm 8.21 (1 H, d, J = 8.31
Hz), 7.87-7.90 (1 H, m), 7.54 (1 H, dd, J = 8.56, 1.51 Hz), 7.47
(1 H, s), 7.28-7.35 (4 H, m), 7.20-7.27 (1 H, m), 3.55 (2 H, dd,
J = 37.02, 13.09 Hz), 3.16-3.30 (1 H, m), 2.94-3.04 (1 H, m),
2.39 (1 H, ddd, J = 12.02, 6.30, 6.11 Hz), 2.18-2.24 (1 H, m),
2.17 (3 H, s), 2.01-2.12 (1 H, m), 1.79-1.91 (2 H, m), 1.73 (1 H,
dd, J = 21.91, 11.58 Hz), 1.66 (9 H, s). 13C NMR (100 MHz,
CDCl3) δ ppm 28.19, 29.59, 30.96, 34.68, 37.74,40.07, 60.72,
66.06, 84.62, 105.68, 116.15, 119.95, 123.11, 124.43, 124.97,
126.95, 127.41, 128.27, 129.12, 130.49, 137.78, 139.24, 149.19.
LCMS (method D) tR 2.73 m, MHþ 430.18. Analytical HPLC
(method F) tR 21.57 m.
3-((1S,3R)-3-(Dimethylamino)cyclopentyl)-1H-indole-5-carbo-
nitrile 8a. R-Chloroethylchloroformate (2.6 mL, 23 mmol) was
added to a solution of 38 (1.0 g, 2.3 mmol) in dichloroethane
(100 mL) at room temperature. The reaction was heated to reflux
for 2 h. After concentration of the reaction to dryness, MeOH
(100 mL) was added and the reaction was heated at reflux for
30 m. The reaction was concentrated and purified by flash
chromatography on 120 g silica gel with a gradient of 0 to 30%
MeOH in DCM (50 min) to obtain the intermediate tert-butyl
5-cyano-3-((1S,3R)-3-(methylamino)cyclopentyl)-1H-indole-
1-carboxylate (686 mg, 88%).
1H NMR (400 MHz, CDCl3) δ ppm 8.39 (1 H, br. s.), 8.14
(1 H, d, J = 8.31 Hz), 7.94 (1 H, s), 7.52 (1 H, s), 7.47 (1 H, d, J =
8.56 Hz), 3.58 (1 H, qd, J = 7.60, 7.43 Hz), 3.12-3.24 (1 H, m),
2.67 (3 H, s), 2.60-2.65 (1 H, m), 1.98-2.33 (5 H, m), 1.62 (9 H, s).
13C NMR (100 MHz, CDCl3) δ ppm 28.15, 28.61, 30.35, 31.90,
35.35, 36.43, 59.81, 84.83, 105.74, 116.15, 119.78, 122.26, 123.94,
124.33, 127.47, 129.81, 137.64, 148.97. LCMS (method E) tR 2.82
m, MHþ 340.21. Analytical HPLC (method G) tR 7.71 m.
The intermediate from the previous experiment (100 mg,
0.29 mmol) was dissolved in MeOH (5 mL) and treated with
37% formaldehyde (0.5 mL) and NaBH3CN (100 mg, 1.6 mmol)
at room temperature for 20 m. The reaction was concentrated
and partitioned between EtOAc (20 mL) and water. The organic
layer was dried over sodium sulfate and concentrated to dryness
to yield tert-butyl 5-cyano-3-((1S,3R)-3-(dimethylamino)cyclo-
pentyl)-1H-indole-1-carboxylate (86 mg, 84%). LCMS (method E)
tR 2.80 m, MHþ 354.20.
The intermediate from the previous experiment (86 mg, 0.24
mmol) was dissolved in DCM (5 mL) and treated with TFA
(2 mL) for 2 h. The reaction was concentrated and dried under
high vacuum and then partitioned between EtOAc and satu-
rated NaHCO3. The organic layer was dried over sodium sulfate
and concentrated to dryness to yield 3-((1S,3R)-3-(dimethyl-
amino)cyclopentyl)-1H-indole-5-carbonitrile 8a (49 mg, 81%).
1H NMR (400 MHz, CD3OD) δ ppm 7.99 (1 H, d, J = 0.76 Hz),
7.43 (1 H, d, J = 8.56 Hz), 7.33 (1 H, dd, J = 8.31, 1.51 Hz), 7.21
(1 H, s), 3.25-3.37 (1 H, m), 2.68-2.79 (1 H, m), 2.32-2.40 (1 H,
m), 2.30 (6 H, s), 2.11-2.22 (1 H, m), 1.94-2.07 (1 H, m), 1.67-
1.86 (2 H, m), 1.63 (1 H, q, J = 11.50 Hz). LCMS (method E) tR
1.72 m, MHþ 254.13. Analytical HPLC (method F) tR 8.09 m.
Synthesis of 9a via Procedure of Scheme 4. 3S-3-(5-Fluoro-
1H-indol-3-yl)-cyclopentanone (S,S)-Hydro-benzoin Ketal 40. A
solution of rac-5b (5 g, 23 mmol), (S,S)-(-) hydrobenzoin (5 g,
23 mmol) and p-toluenesulfonic acid monohydrate (0.44 g, 2.3
C14H14N2O 0.65 H2O: C, 70.66; H, 6.48; N, 11.77. Found: C,
3
70.87; H, 6.80; N, 11.44. [R]25 -13.8 (589 nm, c 1.54 mg/mL,
MeOH). The configuration of the alcohol was established as cis
by a NMR-NOE method.
(S)-1-(3,4-Dichlorobenzoyl)-3-(3-oxocyclopentyl)-1H-indole-
5-carbonitrile 37. 3,4-Dichlorobenzoyl chloride (243 mg, 1.16
mmol) was added to a mixture of (S)-3-(3-oxocyclopentyl)-1H-
indole-5-carbonitrile S-5a (200 mg, 0.892 mmol) and diisopro-
pylethylamine (0.234 mL, 1.34 mmol) in DCM (1 mL) at 0 °C.
Stirring was continued at room temperature overnight. The
reaction was concentrated, and the residue was purified by silica
gel chromatography (EtOAc/hexanes, gradient: 0-40%). Re-
crystallization from EtOAc/EtOH yielded (S)-1-(3,4-dichloro-
benzoyl)-3-(3-oxocyclopentyl)-1H-indole-5-carbonitrile 37 (300
mg, 76%) as a white solid. 1H NMR (400 MHz, CDCl3) δ ppm
1.97-2.15 (m, 1 H) 2.27-2.46 (m, 3 H) 2.58 (m, J = 8.30, 4.20,
4.20, 2.10 Hz, 1 H) 2.78 (dd, J = 18.07, 7.53 Hz, 1 H) 3.58-3.74
(m, 1H) 7.16(d, J=1.25Hz, 1H) 7.56(dd,J= 8.28, 2.01 Hz, 1 H)
7.67 (d, J = 8.00 Hz, 1 H) 7.69 (dd, J = 8.70, 1.50 Hz, 1 H) 7.86 (d,
J = 2.01 Hz, 1 H) 7.96 (dd, J = 1.51, 0.50 Hz, 1 H) 8.45 (dd, J =
8.53, 0.50 Hz, 1 H). LCMS: MHþ = 397 and 399.
Synthesis of 3-[(1S,3R)-3-Dimethylaminocyclopentyl]-1H-indole-
5-carbonitrile 8a via Procedure of Scheme 3. tert-Butyl 3-((1S,3R)-
3-(Benzyl(methyl)amino)cyclopentyl)-5-cyano-1H-indole-1-carboxy-
late 38 and tert-Butyl 3-((1S,3S)-3-(benzyl(methyl)amino)cyclo-
pentyl)-5-cyano-1H-indole-1-carboxylate 39. NaBH3CN (1.26 g,
20 mmol) was added to a mixture of S-5a (1.12 g, 5.0 mmol) and
N-methylbenzylamine (6.45 mL, 50 mmol) in EtOH (75 mL). The
reaction was stirred for 3 h and then diluted into EtOAc and
extracted twice with aqueous NaHCO3 solution and twice with
brine. The organic layer was dried over sodium sulfate and

7572 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
King et al.
mmol) in benzene (150 mL) was heated to reflux under a
Dean-Stark trap for 40 min. The reaction mixture was con-
centrated and the residue was purified by flash chromatography
on silica gel using ethyl acetate/hexanes (0-20%) as the eluant.
The pure fractions were concentrated to give a mixture of two
diastereomers (5 g, 53%). The mixture was dissolved in ethyl
acetate (5 mL) and diluted with hexanes (30 mL). The resulting
solution was cooled in a refrigerator for 2 days to give the
crystalline single diastereomer, 3S-3-(5-fluoro-1H-indol-3-yl)-
cyclopentanone (S,S)-hydro-benzoin ketal 40 (1.6 g, 17% yield,
86.8% de by chiral HPLC method M). [R]25 -7.35 (589 nm, c
6.04 mg/mL, MeOH). 1H NMR (500 MHz, CDCl3) δ 1.98 (m,
1H), 2.36 (m, 4H), 2.67 (m, 1H), 3.50 (m, 1H), 4.75 (s, 2H), 6.94
(t, 1H), 7.09 (s, 1H), 7.30 (m, 11H), 7.93 (s, 1H). M - 1 = 412.
3S-3-(5-Fluoro-1H-indol-3-yl)-cyclopentanone S-5b. A solu-
tion of 40 (207 mg, 0.5 mM) in methanol (35 mL) and 3N HCl
(1 mL) was stirred for 18 h. The solution was concentrated, and
the residue was dissolved in ethyl acetate. The solution was washed
with aqueous sodium bicarbonate and brine and then dried over
magnesium sulfate. The solution was concentrated to give the crude
product, which was purified by flash chromatography on silica gel
using ethyl acetate/hexanes (0-50%) as the eluent to give 3S-3-
(5-fluoro-1H-indol-3-yl)-cyclopentanone S-5b (68 mg, 63% yield,
81% ee by chiral HPLC method K). [R]25 -10.67 (589 nm, c 12.36
mg/mL, MeOH). 1H NMR (500 MHz, CDCl3) δ2.09 (m, 1H), 2.50
(m, 4H), 2.80(m, 1H), 3.64 (m, 1H), 6.96(m, 1H), 7.02(d, 1H), 7.26
(m, 2H), 8.20 (s,1H). M þ 1 = 218.
R,β-Disubstituted R,β-Unsaturated Aldehyde. Org. Lett. 2005, 7,
3437–3440.
(8) King, H. D; Denhart, D. J.; Deskus, J. A.; Ditta, J. L.; Epperson,
J. R.; Higgins, M. A.; Kung, J. E.; Marcin, L. R.; Sloan, C. P.;
Mattson, G. K.; Molski, T. F.; Krause, R. G.; Bertekap, R. L.;
Lodge, N. J.; Mattson, R. J.; Macor, J. E. Conformationally
restricted homotryptamines. Part 4: Heterocyclic and naphthyl
analogs of a potent selective serotonin reuptake inhibitor. Bioorg.
Med. Chem. Lett. 2007, 17, 5647–5651.
(9) Deskus, J. A.; Epperson, J. R.; Sloan, C. P.; Cipollina, J. A.;
Dextraze, P.; Qian-Cutrone, J.; Gao, Q.; Ma, B.; Beno, B. R.;
Mattson, G. K.; Molski, T. F.; Krause, R. G.; Taber, M. T.; Lodge,
N. J.; Mattson, R. J. Conformationally restricted homotrypta-
mines 3. Indole tetrahydropyridines and cyclohexenylamines as
selective serotonin reuptake inhibitors. Bioorg. Med. Chem. Lett.
2007, 17, 3099–3104.
(10) Denhart, D. J.; Deskus, J. A.; Ditta, J. L.; Gao, Q.; King,
H. D.; Kozlowski, E. S.; Meng, Z.; LaPaglia, M. A.; Mattson,
G. K.; Molski, T. F.; Taber, M. T.; Lodge, N. J.; Mattson, R. J.;
Macor, J. E. Conformationally restricted homotryptamines.
Part 5: 3-(trans-2-aminomethylcyclopentyl)indoles as potent selec-
tive serotonin reuptake inhibitors. Bioorg. Med. Chem. Lett. 2009,
19, 4031–4033.
(11) Taber, M. T.; Wright, R. N.; Molski, T. F.; Clarke, W. J.; Brassil,
P. J.; Denhart, D. J.; Mattson, R. J.; Lodge, N. J. Neurochemical,
pharmacokinetic, and behavioral effects of the novel selective
serotonin reuptake inhibitor BMS-505130. Pharmacol., Biochem.
Behav. 2005, 80, 521–528.
(12) Harrington, P. E.; Kerr, M. A. Reaction of indoles with electron
deficient olefins catalyzed by Yb(OTf)3 3H2O. Synlett 1996, 1047–
3
1048.
(13) Kaluzna, I. A.; Rozzell, J. D.; Kambourakis, S. Ketoreductases:
stereoselective catalysts for the facile synthesis of chiral alcohols.
Tetrahedron: Asymmetry 2005, 16, 3682–3689.
(14) An analogous resolution of the CN-ketone 5a proved unsuccess-
ful. We were unable to crystallize either of the diastereomers
formed upon condensation of 5a with (1S,2S)-1,2-dipheny-
lethane-1,2-diol.
[(1R,3S)-3-(5-Fluoro-1H-indol-3-yl)-cyclopentyl]-dimethylamine
9a. Using S-5b obtained in the previous step, conversion to 9a was
carried out according to the earlier procedure wherein chiral HPLC
resolution of ketone 5b was utilized as the source of S-5b. All
analytical data were in agreement with the previous synthesis.
(15) The conformers of compounds 3 and 8a shown in Figure 2 were
identified as follows: Initially, 7500 step Monte Carlo conforma-
tional searches with 2500 steps of PRCG16 energy minimization per
conformer using the OPLS2005 force-field17 and GB/SA implicit
water solvation model18 were performed for compounds 3 and 8a in
their N-protonated cationic forms using MacroModel version 9.5;
Supporting Information Available: Synthesis details for com-
pounds 5-35. This material is available free of charge via the
Internet at http://pubs.acs.org.
€
References
Schrodinger, LLC: New York, NY, 2007. For each compound, the set
of conformations obtained was then ranked in ascending order of
OPLS2005 GB/SA energy and clustered using a non-hydrogen atom
RMSD cutoff of 0.25 Å. The lowest energy conformer from each cluster
was then optimized using density functional theory (DFT) calculations
with the B3LYP functional, 6-31þG* basis set and a self-consistent
reaction field implicit water solvation model.19,20 DFT calculations
(1) Horst, W. D. Biochemical and physiological processes in brain func-
tion and drug actions. Handb. Exp. Pharmacol. 2004, 157, 3–33.
(2) Preskorn, S. H.; Ross, R.; Stanga, C. Y. Selective serotonin
reuptake inhibitors. Handb. Exp. Pharmacol. 2004, 157, 241–
262.
(3) Spinks, D.; Spinks, G. Serotonin reuptake inhibition: an update
on current research strategies. Curr. Med. Chem. 2002, 9,
799–810.
(4) Denhart, D. J.; Mattson, R. J.; Ditta, J. L.; Macor, J. E. One-pot
synthesis of homotryptamines from indoles. Tetrahedron Lett.
2004, 45, 3803–3805.
(5) Schmitz, W. D.; Denhart, D. J.; Brenner, A. B.; Ditta, J. L.;
Mattson, R. J.; Mattson, G. K.; Molski, T. F.; Macor, J. E.
Homotryptamines as potent and selective serotonin reuptake
inhibitors (SSRIs). Bioorg. Med. Chem. Lett. 2005, 15, 1619–
1621.
(6) Mattson, R. J.; Catt, J. D.; Denhart, D. J.; Deskus, J. A.; Ditta,
J. L.; Higgins, M. A.; Marcin, L. R.; Sloan, C. P.; Beno, B. R.; Gao,
Q.; Cunningham, M. A.; Mattson, G. K.; Molski, T. F.; Taber,
M. T.; Lodge, N. J. Conformationally Restricted Homotrypta-
mines. Indole Cyclopropylmethylamines as Selective Serotonin
Reuptake Inhibitors. J. Med. Chem. 2005, 48, 6023–6034.
(7) King, H. D.; Meng, Z.; Denhart, D.; Mattson, R.; Kimura, R.; Wu,
D.; Gao, Q.; Macor, J. E. Enantioselective Synthesis of a Highly
Potent Selective Serotonin Reuptake Inhibitor. An Application
of Imidazolidinone Catalysis to the Alkylation of Indoles with an
€
were performed with Jaguar version 7.0; Schrodinger, LLC: New York,
NY, 2007. For compounds 3 and 8a, the lowest-energy DFT-optimized
conformer that yielded a reasonable three-point overlay (see text) with
the conformer of sertraline shown in Figure 5a of reference 6 was
selected.
ꢀ
(16) Polak, E.; Ribiere, G. Note sur la convergence de methodes de
directions conjuguees. Rev. Fr. Autom. Inf. Rech. Oper., Ser. Rouge
ꢀ
1969, 16, 35–43.
(17) Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. Development
and Testing of the OPLS All-Atom Force Field on Conformational
Energetics and Properties of Organic Liquids. J. Am. Chem. Soc.
1996, 118, 11225–11236.
(18) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T. A.
Semianalytical treatment of solvation for molecular mechanics and
dynamics. J. Am. Chem. Soc. 1990, 112, 6127–6129.
(19) For details of the B3LYP DFT functional and implicit solvation
method used see: Jaguar 7.0 User Manual; Schrodinger, LLC: New
York, NY, 2007, and references therein.
(20) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; John Wiley & Sons: New York, 1986;
Chapter 4.