Biaryls as potent, tunable dual neurokinin 1 receptor antagonists and serotonin transporter inhibitors

Article abstract of DOI:10.1016/j.bmcl.2015.04.098

Depression is a serious illness that affects millions of patients. Current treatments are associated with a number of undesirable side effects. Neurokinin 1 receptor (NK₁R) antagonists have recently been shown to potentiate the antidepressant e

Full text of DOI:10.1016/j.bmcl.2015.04.098

Bioorganic & Medicinal Chemistry Letters 25 (2015) 3039–3043
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Biaryls as potent, tunable dual neurokinin 1 receptor antagonists
and serotonin transporter inhibitors
⇑
Andrew P. Degnan , George O. Tora, Ying Han, Ramkumar Rajamani, Robert Bertekap, Rudolph Krause,
Carl D. Davis, Joanna Hu, Daniel Morgan, Sarah J. Taylor, Kelly Krause, Yu-Wen Li, Gail Mattson,
Melissa A. Cunningham, Matthew T. Taber, Nicholas J. Lodge, Joanne J. Bronson, Kevin W. Gillman,
John E. Macor
Bristol-Myers Squibb Co., Research and Development, 5 Research Parkway, Wallingford, CT 06492, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Depression is a serious illness that affects millions of patients. Current treatments are associated with a
number of undesirable side effects. Neurokinin 1 receptor (NK1R) antagonists have recently been shown
to potentiate the antidepressant effects of serotonin-selective reuptake inhibitors (SSRIs) in a number of
animal models. Herein we describe the optimization of a biaryl chemotype to provide a series of potent
dual NK1R antagonists/serotonin transporter (SERT) inhibitors. Through the choice of appropriate sub-
stituents, the SERT/NK1R ratio could be tuned to afford a range of target selectivity profiles. This effort
culminated in the identification of an analog that demonstrated oral bioavailability, favorable brain
uptake, and efficacy in the gerbil foot tap model. Ex vivo occupancy studies with compound 58 demon-
strated the ability to maintain NK1 receptor saturation (>88% occupancy) while titrating the desired level
of SERT occupancy (11–84%) via dose selection.
Received 9 February 2015
Accepted 30 April 2015
Available online 2 June 2015
Keywords:
Neurokin 1
NK1
SERT
SSRI
Serotonin transport inhibitors
Neuroscience discovery
Ó 2015 Elsevier Ltd. All rights reserved.
Major depressive disorder is a serious and debilitating illness
that has a lifetime prevalence in the United States of >16%.1 It is
a leading cause of disability as measured by years lost to disability
and is the fourth leading contributor to the global burden of dis-
ease as measured by disability-adjusted life years.2 The treatment
of depression was revolutionized in 1988 with the approval of flu-
oxetine, the first serotonin-selective reuptake inhibitor (SSRI), in
the United States.3 Despite the improvements in safety and tolera-
bility made by the advent of this and other SSRIs, these agents have
been associated with a number of negative side effects, including
sexual dysfunction, weight gain, nausea, insomnia, and somno-
lence.3 As such, there remains a serious unmet medical need that
awaits the identification of new pharmacological approaches for
the treatment of depression.
Neurokinin 1 receptor (NK1R) antagonists were first reported to
have anti-depressant activity in clinical trials in 1998 by Kramer
and co-workers.4 In these trials, aprepitant (1) showed improve-
ment in symptoms of anxiety and depression that was comparable
to the well-established SSRI, paroxetine. Aprepitant was found to
be well-tolerated at all doses, with adverse events comparable to
those of placebo. For example, sexual side effects occurred in 26%
of study participants given paroxetine versus 3% given aprepitant.
Despite this early promising result, a subsequent dose-finding
study with aprepitant failed to show superior efficacy to placebo
and development for depression was discontinued.5
More recently, it has been suggested that NK1R antagonists
potentiate the antidepressant effects of SSRIs.6 For instance, it
has been shown that combination of sub-active doses of citalo-
pram or paroxetine with an inactive dose of NK1R antagonist
GR205171 reduces immobility of Swiss mice in the forced swim
test.7 A similar result was observed in our own labs using parox-
etine and aprepitant in the gerbil forced swim test.8 As a part of
our studies, ex vivo occupancy of both the NK1 receptor and the
serotonin transporter (SERT) was measured. It was found that by
maintaining high NK1 receptor occupancy, the SERT occupancy
could be reduced while retaining activity in the gerbil forced swim
test. This result offered the hope of delivering an antidepressant
with fewer side effects than with an SSRI alone.
It was in this context that we undertook the development of a
dual NK1R/SERT antagonist.9 From the occupancy experiments
described above, it appeared that high NK1 receptor occupancy
would allow us to reduce SERT occupancy and, theoretically, reduce
the potential for SSRI-related side effects. Our studies suggested a
SERT inhibition/NK1R antagonism IC50 ratio of approximately 10
⇑
Corresponding author. Tel.: +1 203 677 5750; fax: +1 203 677 7702.
E-mail address: andrew.degnan@bms.com (A.P. Degnan).
http://dx.doi.org/10.1016/j.bmcl.2015.04.098
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

3040
A. P. Degnan et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3039–3043
Table 1
or greater would allow us to achieve the desired receptor/trans-
porter occupancy profile.
SAR studies: in vitro potency of tetrazoles 6–11a,b
In an earlier paper,9b we reported that 2 was a potent dual NK1R
antagonist and SERT inhibitor. We sought to improve the NK1R
potency of this series (and consequently increase the SERT/NK1R
ratio) and focused our initial work around modification of the
bis-trifluoromethylphenyl ring. We were inspired in this
endeavor by the work of Ward and co-workers,10 who showed that
the 3,5-bistrifluoromethylphenyl of NK1R antagonists such as
O
N
H
aprepitant (1) could be replaced with
a 2-methoxy-5-tetra-
Compound
R
NK1R
IC50 (nM) (nM)
SERT IC50 SERT/
NK1R
ratio
zolylphenyl to afford compounds such as 3 (Fig. 1). Importantly,
this substitution gave dramatic improvements in both metabolic
stability and oral bioavailability. In a subsequent disclosure,11 it
was reported that substitution of C-5 of the tetrazole with small
alkyl groups was also well-tolerated to afford compounds 4 and
5. We sought to apply this modification to our series.
N
N
N
6
7
87
37
70
0.4
0.4
N
MeO
The first compound prepared in this endeavor was 6 (Table 1).
Although this analog retained nanomolar potency against both tar-
gets, we were disappointed to find that it was significantly less
potent than the parent bis-trifluoromethylphenyl derivative, 2.
Additionally, this analog was somewhat more potent against
SERT than NK1R, contrary to our desired profile (vide supra).
Given this result, we investigated the possibility of replacement
of the 2-methoxy substituent with the 3-trifluoromethyl group of
our earlier lead series. Unfortunately, compound 7 was less potent
than our earlier lead (2) and had an unfavorable SERT/NK1R ratio.
In earlier work,11 substitution of the tetrazole with a trifluo-
romethyl group gave maximum NK1R potency, however on our
substrate, just the opposite trend was observed with compound
8 having a still lower SERT/NK1R ratio. Our most favorable balance
of potency and ratio was realized with the C-linked tetrazole 10,
which had an NK1R antagonism and SERT inhibition ratio of ꢀ1.
Interestingly, the desmethyl tetrazole 9 and the isomeric N-
methyltetrazole 11 achieved a desirable potency for SERT inhibi-
tion but did not have sufficient potency for NK1R antagonism to
achieve the targeted potency profile.
Encouraged by the result with 10, we sought to more exten-
sively explore SAR around biaryl ring systems, focusing most of
our work on C-linked aryl and heteroaryl substituents. The results
of these studies are summarized in Table 2. In general, 6-mem-
bered aromatic systems gave the best combination of potency
and reasonable SERT/NK1R ratios. We were especially encouraged
by phenyl-containing 12, having single digit nanomolar potency
at both targets and 4-pyridyl 13, having single digit nanomolar
potency against NK1R and an ideal SERT/NK1R ratio (10). Smaller
heterocycles such as furan (15), thiazole (16), and those in
N
N
N
N
N
N
180
F3C
F3C
F3C
F3C
F3C
N
N
8
700
200
65
57
0.08
0.04
0.8
F3C
N
N
N
9
8.6
50
N
H
N
N
N
N
10
11
N
N
N
N
36
9.6
0.3
a
Binding assays performed by methods described in Ref. 8.
All binding values represent a minimum of 2 replicates.
b
Encouraged by these results, we expanded our understanding
Table
1 tended to have lower NK1R potency and lower
around the SAR of 6-membered aryls through the preparation of
an extensive biaryl library using a combination of Suzuki and
Stille coupling approaches. The requisite precursors were prepared
SERT/NK1R ratios. Larger bicyclic aromatics (compounds 17 and
18) tended toward lower affinities at both targets and were not
pursued further.
CF3
MeO
R
CF3
CF3
O
N
O
N
N
N
N
O
O
CF3
N
H
N
F
HN
O
N
H
2
NH
Aprepitant, 1
NK1R IC50 = 0.09 nM
3: R=H, NK1R Ki = 0.05 nM
4: R=Me, NK1R Ki = 0.05 nM
5: R=CF3, NK1R Ki = 0.03 nM
NK1R IC50 = 1.0 nM
SERT IC50 = 1.0 nM
Figure 1. Structures of literature NK1R antagonists.

A. P. Degnan et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3039–3043
3041
Table 2
OH
SAR studies: in vitro potency of diverse biaryls 12–18
CO2H
a, b
c
CF3
O
N
H
.
N
TsOH
23
N
24
Boc
Aryl
CF3
CF3
N
H
O
O
d, e
Compound Aryl
NK1R IC50
(nM)
SERT IC50
(nM)
SERT/NK1R
ratio
Br
Ar
or
f, g, e
N
H
Boc
12
13
6.3
1.6
3.7
16
0.6
10
25
26
Scheme 2. Reagents and conditions: (a) Boc2O, Et3N, THF (98%); (b) BH3–THF, THF
(97%); (c) NaH, 22, DMF (96%); (d) ArB(OH)2, Pd(PPh3)4, 1 N KOH; (e) 33% TFA in
DCM; (f) n-BuLi, THF then Bu3SnCl (79%); (g) ArBr, PdCl2(PPh3)2, MeCN.
N
N
14
15
41
47
7.4
18
0.2
0.4
the tert-butyl carbamate by treatment with TFA/DCM completed
the synthesis. In all, approximately 150 compounds were prepared
by these methods.
O
N
Following up on lead 13, a variety of substituted 4-pyridines
were prepared. While several retained useful levels of potency
against both targets (data not shown), none gave a better combina-
tion of potency and SERT/NK1R ratio than that of the unsubstituted
4-pyridyl (13). On the other hand, a number of phenyl substituents
were found to have beneficial effects on both potency and
SERT/NK1R ratio in the biphenyl-containing series (Table 3). In gen-
eral, substitution of the 4-position was preferred relative to substi-
tution of other positions. Substitution of the 4-position with either
cyano or alkoxy groups gave compounds with potent NK1R
16
17
70
4.1
0.06
0.9
S
120
110
18
540
120
0.2
as outlined in Schemes 1 and 2. To begin, nitro acid 19 was reduced
in two steps to hydroxy aniline 20. A modified Sandmeyer reac-
tion12 converted the aniline to the corresponding aryl bromide
21. Conversion of the hydroxyl group to the benzyl bromide pro-
vided intermediate 22 (Scheme 1) for coupling to the piperidine.
Preparation of the piperidine began with the commercially avail-
able piperidine acid 23 (Scheme 2). Protection of the nitrogen as
the tert-butyl carbamate and reduction of the acid by the action
of borane-THF yielded alcohol 24. Benzyl bromide 22 and alcohol
24 were efficiently coupled by the action of sodium hydride in
DMF to afford the corresponding ether in 96% yield (Scheme 2).
Aryl bromide 25 proved itself to be a competent coupling partner
in Suzuki reactions with arylboronic acids. Alternately, the bro-
mide could be converted to the corresponding stannane for use
in Stille couplings. This greatly increased the number of commer-
cially available coupling partners by expanding the potential cou-
pling partners from arylboronic acids to arylhalides. Removal of
Table 3
SAR studies: effect of biphenyl substitution on potency
CF3
O
N
H
R
Compound
R
NK1R IC50
(nM)
SERT IC50
(nM)
SERT/NK1R
ratio
27
28
29
30
31
32
33
34
35
36
37
38
2-Cl
2-F
68
23
130
6.5
54
24
49
5.0
25
5.2
13
6.4
7.6
13
3.5
4.2
14
10
6.3
30
0.1
0.3
0.1
0.5
0.1
0.6
0.2
1
2-Me
2-NO2
2-OMe
3-CH3
3-CN
3-F
3-NH2
3-NO2
3-OH
3,4-
O
1
0.8
2
4.2
27
45
CF3
CF3
HO
HO
a, b
c
7.1
6
OCH2O
4-CF3
4-Cl
4-CO2Me
4-F
4-Me
4-NMe2
4-NO2
4-OEt
4-OH
4-OMe
4-CN
19
20
Br
39
40
41
42
43
44
45
46
47
48
49
30
2.8
17
36
1.8
25
1.5
6.4
3.5
7.3
14
1
0.6
1
0.7
0.8
0.2
2
4
8
16
9
NO2
NH2
2.2
8.4
18
3.8
3.4
3.6
0.93
0.45
CF3
CF3
d
HO
21
22
Br
Br
27
15
4.0
Scheme 1. Reagents and conditions: (a) BH3-THF, THF (85%); (b) Pd/C, H2, MeOH
(85%); (c) CuBr2, t-butylnitrite, MeCN (69%); (d) NBS, PPh3, THF (76%).

3042
A. P. Degnan et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3039–3043
antagonism and favorable SERT/NK1R ratios. The most attractive
compound identified was the 4-cyanophenyl analog 49, which
had potent NK1R antagonism and SERT inhibition (NK1R
IC50 = 0.45 nM; SERT IC50 = 4.0 nM) and a nearly ideal ratio (9).
Additional analogs were prepared around the 4-alkoxy and 4-
cyano 6-membered aryls to further refine our leads. No analogs
were identified within the 4-alkoxy series that showed any
improvements in potency over those outlined in Table 3 (data
not shown). A number of compounds in the 4-cyano series were
found to have attractive potencies and ratios and led to the identi-
fication of a number of related compounds with varying SERT/NK1R
profiles (Table 4). Incorporation of fluorine onto C-2 (50), gave a
compound with a profile very similar to that of 4-cyanophenyl lead
49. Replacement of both the C-2 and C-6 protons with fluorine
(51), boosted the SERT/NK1R ratio to 22 while retaining single digit
nanomolar potency against SERT. Alternately, substitution with
2,5-difluoro (52), 2-chloro (54), 2-methyl (55), or 2-aza (57) gave
compounds with high potency (<10 nM) against both NK1R and
SERT, but lower levels of NK1R selectivity (SERT/NK1R IC50’s = 2–
5). In this series, only compound 56 was found to be more selective
for SERT. This compound had low nanomolar potency against both
targets with a SERT/NK1R ratio of 0.5.
Given its excellent potency and its near ideal SERT/NK1R ratio,
49 was prioritized for advancement into behavioral studies. NK1R
radioligand binding studies have shown significant species differ-
ences. Whilst mouse and rat receptor binding have proven poor
predictors for human pharmacology, the gerbil and guinea pig have
demonstrated good receptor homology with human.13 As such, our
screening tier utilized the gerbil foot tap model14 as an in vivo tar-
get engagement model. In this assay, the ability of an NK1R antag-
onist, dosed orally, to block foot tapping induced by i.c.v.
administration of an NK1R agonist (GR-73632) was measured.
We were disappointed to find that 49, despite its excellent
in vitro potency, failed to give a significant reduction in foot tap-
ping. Measurement of drug concentration in the plasma and brain
indicated poor brain uptake (Table 6), and suggested that poor
brain penetrance was to blame for the lack of behavioral efficacy.
It was hypothesized that brain penetration could be improved
by removal of a hydrogen bond donor via substitution of the nitro-
gen of the piperidine. To this end, a small library of N-alkylated
piperidines was prepared (Table 5). Although substitution had lit-
tle effect on NK1R potency, it was found that SERT potency was
exquisitely sensitive to the steric bulk of the substituent. Only sub-
stitution with a methyl group (58) was tolerated, affording only a
small (ca. 2 fold) loss in potency at SERT. Substitution with larger
groups (compounds 59–62) was accompanied by dramatic reduc-
tions in SERT potency (>100 fold).
Table 4
SAR studies: impact of cyanobiphenyl functionalization on SERT/NK1R ratio
Compound 58 was advanced into the gerbil foot tap model. We
were gratified to find that 58, differing by only a single methyl
group, gave a robust reduction in foot tapping (Table 6). Again,
plasma and brain concentrations were measured. They indicated
CF3
O
greatly enhanced brain uptake ([brain] = 1.3
lM; B:P = 2.5) relative
Aryl
to unmethylated derivative 49 ([brain] = 0.19
l
M; B:P = 0.52).
N
H
Activity in the gerbil foot tap model, suggested significant occu-
pancy of the NK1 receptor, but said nothing about the occupancy
of serotonin transporters. To assess the absolute and relative level
of occupancy at both targets, ex vivo occupancy experiments were
performed with compound 58 across a range of doses in gerbil
(Fig. 2). 125I-substance P and 3H-citalopram were used to assess
NK1R and SERT occupancy, respectively. To our delight, NK1 recep-
tor occupancy was high (88–102% across all doses) while serotonin
transporter occupancy increased in a dose-dependent manner
from 11% (@3 mg/kg) to 84% (@30 mg/kg). This target profile sug-
gested the ability to maintain NK1 receptor saturation while titrat-
ing (via dose selection) SERT occupancy up to a level which
achieves clinical efficacy without the mechanism-based side
effects that can occur at high SERT occupancies.
Compound Aryl
NK1R IC50
(nM)
SERT IC50
(nM)
SERT/NK1R
ratio
F
F
50
1.1
8.9
7.0
8
CN
CN
51
0.32
22
F
F
CN
52
53
0.91
3.7
47
4
F
F
Table 5
F
F
SAR studies: impact of piperidine alkylation on potency
CF3
CN 4.8
10
O
F
CN
54
55
1.1
3.1
2.7
3
2
N
R
Cl
CN
CN
1.2
Compound
R
NK1R IC50 (nM)
SERT IC50 (nM)
SERT/NK1R ratio
49
58
59
60
61
62
H
Me
Et
i-Pr
c-Pr
Bn
0.45
0.29
0.25
0.39
1.8
4.0
9.5
420
450
2100
800
9
33
1700
1200
1200
400
N
56
57
8.1
4.0
10
0.5
5
CN
N
1.9
CN
2.0

A. P. Degnan et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3039–3043
3043
Table 6
piperidine afforded a compound that enhanced oral bioavailability
and greatly improved brain penetration. Ex vivo occupancy studies
with compound 58 demonstrated the ability to maintain NK1
receptor saturation (>88% occupancy) while titrating the desired
level of SERT occupancy (11–84%) via dose selection. Additional
characterization of compound 58 will be disclosed in due course.
Impact of piperidine alkylation on activity in gerbil foot tap model and brain uptake
Compound
49
58
hNK1R IC50 (nM)
0.59
0.20
73%
526
1.3
Foot Tappinga (% inh @ 10 mpk, po)
Plasma Conc. (nM)
29% (ns)b
365
Brain Conc. (
lM)
0.19
Brain:Plasma
0.52
2.5
References and notes
a
Gerbils treated with test compound (10 mg/kg p.o.). After 4 h, animals anes-
1. Kessler, R. C.; Berglund, P.; Demler, O.; Jin, R.; Koretz, D.; Merikangas, K. R.;
thetized with isoflurane and treated with GR-73632 (3 pmol, i.c.v.). Foot tapping
recorded for 5 min after animal regains righting reflex (2–3 min post-injection).
Rush, A. J.; Walters, E. E.; Wang, P. S. J. Am. Med. Assoc. 2003, 289, 3095.
b
2. World
Health
Organization
website:
http://www.who.int/mental_
ns = not significant.
health/management/depression/definition/en/. Downloaded 2/22/2010.
3. Ferguson, J. M. J. Clin. Psychiatry 2001, 3, 22.
4. Kramer, M. S.; Cutler, N.; Feighner, J.; Shrivastava, R.; Carman, J.; Sramek, J. J.;
Reines, S. A.; Liu, G.; Snavely, D.; Wyatt-Knowles, E.; Hale, J. J.; Mills, S. G.;
MacCoss, M.; Swain, C. J.; Harrison, T.; Hill, R. G.; Hefti, F.; Scolnick, E. M.;
Cascieri, M. A.; Chicchi, G. G.; Sadowski, S.; Williams, A. R.; Hewson, L.; Smith,
D.; Carlson, E. J.; Hargreaves, R. J.; Rupniak, N. M. J. Science 1998, 281, 1640.
5. (a) Rupniak, N. M. J.; Kramer, M. S. Trends Pharmacol. Sci. 1999, 20, 485; (b)
Herpfer, I.; Lieb, K. CNS Drugs 2005, 19, 275.
6. Gobert, A.; Brocco, M.; Dekeyne, A.; Di Cara, B.; Bouchez, G.; Lejeune, F.;
Gannon, R. L.; Millan, M. J. Neuropsychopharmacology 2009, 34, 1039.
7. Chenu, F.; Guiard, B. P.; Bourin, M.; Gardier, A. M. Behav. Brain Res. 2006, 172,
256.
8. Lelas, S.; Li, Y.-W.; Wallace-Boone, T. L.; Taber, M. T.; Newton, A. E.; Pieschl, R.
L.; Davis, C. D.; Molski, T. F.; Newberry, K. S.; Parker, M. F.; Gillman, K. W.;
Bronson, J. J.; Macor, J. E.; Lodge, N. J. Neuropharmacology 2013, 73, 232.
9. For other efforts to identify dual-acting NK1R/SERT antagonists, see: (a) Wu, Y.-
J.; He, H.; Gao, Q.; Wu, D.; Bertekap, R.; Westphal, R. S.; Lelas, S.; Newton, A.;
Wallace, T.; Taber, M.; Davis, C.; Macor, J. E.; Bronson, J. Bioorg. Med. Chem. Lett.
2014, 24, 1611; (b) Wu, Y.-J.; He, H.; Bertekap, R.; Westphal, R.; Lelas, S.;
Newton, A.; Wallace, T.; Taber, M.; Davis, C.; Macor, J. E.; Bronson, J. Bioorg.
Med. Chem. 2013, 21, 2217; (c) Gillman, K. W.; Parker, M. F.; Silva, M.; Degnan,
A. P.; Tora, G. O.; Lodge, N. J.; Li, Y.-W.; Lelas, S.; Taber, M.; Krause, R. G.;
Bertekap, R. L.; Newton, A. E.; Pieschl, R. L.; Lengyel, K. D.; Johnson, K. A.; Taylor,
S. J.; Bronson, J. J.; Macor, J. E. Bioorg. Med. Chem. Lett. 2013, 23, 407; (d) Millan,
M. J.; Dekeyne, A.; Gobert, A.; Mannoury la Cour, C.; Brocco, M.; Rivet, J.-M.;
DiCara, B.; Lejeune, F.; Cremers, T. I.; Flik, G.; de Jong, T. R.; Olivier, B.; de
Nanteuil, G. Eur. Neuropsychopharmacol. 2010, 20, 599.
Figure 2. Ex vivo NK1R and SERT occupancy in gerbils: Dose response of compound
58. Gerbils treated with test compound (po). Ex vivo occupancy assessed at 4 h.
10. Ward, P.; Armour, D. R.; Bays, D. E.; Evans, B.; Giblin, G. M. P.; Heron, N.;
Hubbard, T.; Liang, K.; Middlemiss, D.; Mordaunt, J.; Naylor, A.; Pegg, N. A.;
Vinader, M. V.; Watson, S. P.; Bountra, C.; Evans, D. C. J. Med. Chem. 1995, 38,
4985.
11. Armour, D. R.; Chung, K. M. L.; Congreve, M.; Evans, B.; Guntrip, S.; Hubbard, T.;
Kay, C.; Middlemiss, D.; Mordaunt, J. E.; Pegg, N. A.; Vinader, M. V.; Ward, P.;
Watson, S. P. Bioorg. Med. Chem. Lett. 1996, 6, 1015.
12. Doyle, M. P.; Siegfried, B.; Dellaria, J. F., Jr. J. Org. Chem. 1977, 42, 2426.
13. (a) Beresford, I. J. M.; Birch, P. J.; Hagan, R. M.; Ireland, S. J. Br. J. Pharmacol.
1991, 104, 292; (b) Gitter, B. D.; Waters, D. C.; Bruns, R. F.; Mason, N. R.; Nixon,
J. A.; Howbert, J. J. Eur. J. Pharmacol. 1991, 197, 237.
In conclusion, it was found that biaryls are suitable replace-
ments for the bis-trifluoromethyl group of 1. Replacement of one
of the trifluoromethyl groups with a 4-cyanophenyl gives a com-
pound with significantly improved potency against NK1R to afford
a compound with an improved SERT/NK1R ratio. By selection of
appropriate substitutions around the 4-cyanophenyl ring, a num-
ber of highly potent compounds were identified which were differ-
entiated primarily by their SERT/NK1R ratio. Methylation of the
14. Bristow, L. J.; Young, L. Eur. J. Pharm. 1994, 253, 245.