Discovery of Indazoles as Potent, Orally Active Dual Neurokinin 1 Receptor Antagonists and Serotonin Transporter Inhibitors for the Treatment of Depression

Article abstract of DOI:10.1021/acschemneuro.6b00337

Combination studies of neurokinin 1 (NK1) receptor antagonists and serotonin-selective reuptake inhibitors (SSRIs) have shown promise in preclinical models of depression. Such a combination may offer important advantages over the current standard of care. Herein we describe the discovery and optimization of an indazole-based chemotype to provide a series of potent dual NK1 receptor antagonists/serotonin transporter (SERT) inhibitors to overcome issues of ion channel blockade. This effort culminated in the identification of compound 9, an analogue that demonstrated favorable oral bioavailability, excellent brain uptake, and robust in vivo efficacy in a validated depression model. Over the course of this work, a novel heterocycle-directed asymmetric hydrogenation was developed to facilitate installation of the key stereogenic center.

Full text of DOI:10.1021/acschemneuro.6b00337

Letter
pubs.acs.org/chemneuro
Discovery of Indazoles as Potent, Orally Active Dual Neurokinin 1
Receptor Antagonists and Serotonin Transporter Inhibitors for the
Treatment of Depression
Andrew P. Degnan,*,† George O. Tora,† Hong Huang,† David A. Conlon,‡ Carl D. Davis,†
Umesh M. Hanumegowda,† Xiaoping Hou,§ Yi Hsaio,‡ Joanna Hu,† Rudolph Krause,† Yu-Wen Li,†
Amy E. Newton,† Rick L. Pieschl,† Joseph Raybon,† Thorsten Rosner,‡ Jung-Hui Sun,§
Matthew T. Taber,† Sarah J. Taylor,† Michael K. Wong,§ Huiping Zhang,§ Nicholas J. Lodge,†
Joanne J. Bronson,† John E. Macor,† and Kevin W. Gillman†
†Research and Development, Bristol-Myers Squibb Company, Wallingford, Connecticut 06492, United States
‡Chemical Development, Bristol-Myers Squibb Company, New Brunswick, New Jersey 08903, United States
§Department of Chemical Synthesis, Bristol-Myers Squibb Company, Princeton, New Jersey 08543, United States
S
* Supporting Information
ABSTRACT: Combination studies of neurokinin 1 (NK1)
receptor antagonists and serotonin-selective reuptake inhib-
itors (SSRIs) have shown promise in preclinical models of
depression. Such a combination may offer important
advantages over the current standard of care. Herein we
describe the discovery and optimization of an indazole-based
chemotype to provide a series of potent dual NK1 receptor
antagonists/serotonin transporter (SERT) inhibitors to over-
come issues of ion channel blockade. This effort culminated in
the identification of compound 9, an analogue that demonstrated favorable oral bioavailability, excellent brain uptake, and robust
in vivo efficacy in a validated depression model. Over the course of this work, a novel heterocycle-directed asymmetric
hydrogenation was developed to facilitate installation of the key stereogenic center.
KEYWORDS: Depression, neurokinin 1, serotonin transporter inhibitor, hERG, ion channels, asymmetric hydrogenation
n antiquity, depression was thought to be due to a buildup of
environmental,7 and neurobiological8 factors all play a role in
1
Ian excess of black bile, one of the four basic bodily humors. the development of depression and may lie upstream of the
serotonergic effects.
Indeed, the word “melancholy” is derived from ancient Greek
2
First approved in the mid-1980s for the treatment of major
depressive disorder (MDD), serotonin-selective reuptake
inhibitors (SSRIs) inhibit reuptake of serotonin, increasing
the synaptic concentrations of the neurotransmitter. Nearly 30
years since their introduction, SSRIs remain the bedrock of
depression treatment, accounting for more prescriptions than
all other antidepressant classes combined.9 It should be noted
that the shift from older MAOIs and tricyclic antidepressants to
SSRIs has been driven more by their improved safety profile,
rather than increased efficacy.10 Despite their widespread use,
only about 30% of patients achieve remission on SSRI-based
therapy.11 The utility of SSRIs is further limited by a delayed
onset of action, requiring approximately 4 weeks of treatment
prior to alleviation of symptoms.12 As such, there remains a
need to identify novel strategies to improve depression patient
outcomes.
́
́
for “black bile” (μελαινα χολη, melaina chole) where it was
linked with persistent fear and despondency. In modern times,
the symptoms of depression have been blamed not on an excess
of black bile, but on a deficiency of the neurotransmitter
serotonin.
First described in 1967, the original serotonin hypothesis
posited that reduced serotonin levels in the brain were the
direct cause of depression, and that pharmacological restoration
of serotonin levels would reduce symptomatology in patients.3
This theory was based on clinical trials carried out by Coppen
and co-workers, which showed that patients who received both
a monoamine oxidase inhibitor (MAOI) and tryptophan, a
brain-penetrant biogenic precursor to serotonin, showed more
robust reductions in their depression than those given an
MAOI alone.4 While it was known that MAOIs increased the
levels of monoamine neurotransmitters through the inhibition
of enzymes that degrade them in the brain, the central role of
serotonin was not elucidated until researchers combined these
two serotonin-enhancing mechanisms into a single study. Since
that time, it has become clear that genetic,5 epigenetic,6
Received: October 3, 2016
Accepted: October 16, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acschemneuro.6b00337
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Letter
Figure 1. Genesis of the indazole chemotype.
Table 1. In Vitro Profiling of Compounds 1−9
IC50, nM
% inh @1 μM
a
compd
hNK1
hSERT
SERT/NK1 ratio
MetStab h/r/m (% remain)
hERG
Na
hPPB (% bound)
cLogP
LipE (SERT)
1
0.29
0.28
3.6
9.5
5.8
33
99/79/87
97/84/97
91/1/3
80
75
6.3
6.4
5.4
4.3
3.7
3.8
4.2
4.2
4.6
5.2
1.8
1.8
3.9
4.9
5.4
5.2
3.7
3.2
3.7
3.1
2
21
95
99.2
98.0
96.5
3
0.50
0.60
0.83
1.0
0.1
0.1
0.2
0.2
58
4
5.4
100/100/90
65
5
3.4
6
4.4
7
0.21
200
0.11
0.46
12
ent-7
39
0.2
48
8
9
5.4
65/5/35
36
90
73
90
96.6
99.1
5.2
11
88/81/96
a
Percentage remaining after 10 min incubation with human (h), rat (r), or mouse (m) liver microsomes.
One such approach has been the combination of an SSRI
dual acting compound reduces patient pill burden without
undertaking the often difficult and costly process of
coformulating multiple active pharmaceutical ingredients. It
was in this context that we undertook the development of a
dual NK1 receptor antagonist/serotonin transporter (SERT)
inhibitor.
with a neurokinin 1 (NK1) receptor antagonist. Interest in this
combination was initiated by Froger and co-workers’ study of
SSRIs in NK1 knockout mice.13 In this study, they
demonstrated that wild-type and NK1 knockout mice show
indistinguishable extracellular levels of serotonin. Upon acute
administration of an SSRI, the levels of extracellular serotonin
increased. However, the maximum SSRI-induced increase in
NK1 knockouts was more than twice that observed in wild type
animals. The authors further showed that the NK1 knockout
mice demonstrated marked desensitization/downregulation of
the 5-HT1A autoreceptor. 5-HT1A autoreceptors mediate
negative feedback control on both the synthesis and release
of serotonin.14 Activation and desensitization of 5-HT1A
autoreceptors are known to be major mediators in the clinical
efficacy of serotonergic therapies (busparone, SSRIs, MAOIs,
serotonin precursor supplementation),15 and the timing of
desensitization mirrors the onset of clinical efficacy, suggesting
the possibility for more rapid onset of action via combination
therapy. Such a combination of an NK1 receptor antagonist
with an SSRI, both targeting 5-HT1A autoreceptor desensitiza-
tion, may be as seminal to understanding and treating
depression as Coppen’s combination trials which first linked
serotonin with depression and led to improved patient
outcomes.
Indeed, multiple groups have shown that combination of
subefficacious doses of NK1 antagonists with subefficacious
doses of SSRIs leads to robust in vivo efficacy in validated
rodent models of depression.16 While approved medicines in
both classes exist today, there are significant advantages to
targeting dual activity in a single molecule.17 First, this strategy
necessarily aligns the absorption, distribution, metabolism, and
excretion (ADME) properties of the two mechanisms, and thus
gives rise to predictable relative inhibition of both targets across
time and across bodily compartments in vivo. Additionally, a
In an earlier paper,18 we reported the discovery of 1 (Figure
1) as a potent dual NK1 receptor antagonist and SERT
inhibitor. This compound had excellent potency at both targets
and a ratio consistent with our desired profile (SERT IC50/
NK1 IC50 ratio >10). The compound was orally bioavailable
and highly brain penetrant (B:P = 2.5). Consistent with this
profile, in ex vivo gerbil occupancy studies, 1 demonstrated the
ability to maintain NK1 receptor saturation (>88% occupancy)
while titrating the desired level of SERT occupancy (11−84%)
via dose selection, suggesting the potential for antidepressant
efficacy with reduced potential for SSRI-related side effects.
Ultimately, on the basis of modest increases in SERT potency
and metabolic stability, compound 2 became the lead
compound from the biphenyl chemotype and was selected
for more extensive in vitro profiling. Electrophysiology studies
of 2 indicated that it was a potent inhibitor of both hERG and
sodium channels. Screening of other leads from this class of
compounds indicated that hERG inhibition was endemic to the
series. There are three primary strategies routinely employed in
attempts to reduce hERG binding: (1) formation of zwitterions
to limit access to the transmembrane binding site, (2) reduction
of the basicity of the amine moiety, and (3) reduction of
log P.19 As we were targeting effects in the CNS, we did not
pursue a zwitterion-based strategy due to anticipated issues
with brain uptake. We pursued pKa modulation of the basic
nitrogen, but, unfortunately, analogues which reduced the
basicity of the piperidine moiety showed unacceptable losses in
SERT potency (data not shown). Thus, we embarked upon a
log P-based hERG mitigation strategy. It was recognized that
B
DOI: 10.1021/acschemneuro.6b00337
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Letter
Figure 2. Evolution of indazole chemotype into lead 9.
leads from this chemotype were highly lipophilic, with N-
methylated derivatives (a requirement for favorable CNS
penetration18) such as 1 and 2 having a cLogP > 6. As such,
we sought to identify opportunities to introduce polar
functional groups into this series. We drew inspiration from
our biphenyl SAR where it was noted that that the methoxy
group of 3 was well-tolerated, giving a compound with potent
SERT and NK1 activity (albeit with reversed selectivity relative
to the desired target profile, see Table 1). This result suggested
the possibility of appending a heterocycle across C1 and C6 of
the tetra-substituted phenyl group. We sought a heterocycle
that would be electron donating at C1 (to mimic the C1-
methoxy group of 3) and electron-withdrawing at C6 (to mimic
the trifluoromethyl of 2), and felt that the indazole of 4 might
achieve this goal. Furthermore, in choosing an indazole as our
heterocyclic replacement, we anticipated a > 1 log unit drop in
lipophilicity (cLogP) relative to related biphenyls. We were
delighted to find that 4 had an essentially identical in vitro
target profile to that of 3 (Table 1). It was also gratifying to find
that the reduction in cLogP in silico correlated with reductions
in plasma protein binding in vitro relative to 2 and 3 (Table 1).
With this lead in hand, we began to prepare analogues to
develop SAR in this series with a particular interest in the
identification of compounds with improved NK1 potency. The
evolution of 4 to an optimized analogue is shown in Figure 2.
The strategic decision was made to focus our efforts on the
preparation of unsubstituted (NH) piperidines rather than the
corresponding N-methylated derivatives for reasons of synthetic
expediency. From earlier studies, we had found that, upon
methylation, NK1 was only minimally impacted while SERT
potency was generally within 3-fold of the unmethylated
derivative. Unsubstituted piperidines that approached the
desired in vitro profile (SERT/NK1 ratio >10, SERT < 10
nM) were subsequently converted to the corresponding N-
methylpiperidine and characterized independently.
The 3,5-bistrifluoromethylphenyl motif is a privileged
pharmacophore among NK1 receptor antagonists.20,21 As the
indazole of 4 had replaced the trifluoromethyl group of 2, we
replaced the cyanophenyl of 4 with this moiety. We were
pleased to find that 5 had slightly improved NK1 potency and
reduced lipophilicity. Over the course of the medicinal
chemistry effort, the fluorophenyl of 2 overtook the
unsubstituted phenyl of 1 as the preferred scaffold on the
basis of trends toward increased liver microsomal stability and
was thus installed to afford 6. The critical advancement in NK1
receptor potency was achieved by incorporation of a methyl
group in the benzylic position of the ether. Upon installation of
a methyl group in the R-configuration, a 20× increase in NK1
receptor affinity was realized, while the SERT potency was
reduced 10× to give a compound (7) which approached our
targeted in vitro profile. It is of interest to note that while the S-
methyl derivative (ent-7) is ∼1000× less potent at the NK1
receptor, SERT activity was reduced by only ∼3×. Previously,
we demonstrated that unsubstituted piperidines showed poor
brain uptake.18 Therefore, the piperidine was methylated to
afford 8, a compound with exquisite potency at both targets and
with SERT/NK1 ratio > 10.
In transforming from biphenyl 2 to indazole 8, we had
identified a molecule with a nearly identical in vitro target
profile, while substantially reducing molecular weight (483 →
435) and cLogP (6.4 → 4.6). We also monitored lipophilic
ligand efficiency22 (LipE) as means of relating our lipophilicity
to SERT potency. SERT-based LipE was chosen because, given
our target profile of a SERT/NK1 ratio of >10, it was SERT
potency that would ultimately impact dose selection. LipE also
indicated substantial improvement, increasing by nearly two
units (1.8 → 3.7). To measure the effect of reduced lipophility
on our ion channel inhibition, 8 was examined in sodium and
hERG patch clamp experiments. We were pleased to find that 8
had reduced inhibition at both channels. Unfortunately, liver
microsomal stability was only moderate in human and low in
rat and mouse. Suspecting that the most likely site of
metabolism was at C3 of the indazole, a small library of C3-
substituted indazoles was prepared. It was found that
substitution of C3 of the indazole gave analogues with greatly
improved microsomal stability, but with unacceptable losses in
SERT potency (data not shown). By incorporating substitution
at C3 and reoptimizing for SERT potency at C5, we identified
9, a compound with high microsomal stability in all species.
This compound also had an ideal target in vitro profile (NK1 =
0.45 nM, SERT = 5.2 nM; SERT/NK1 ratio = 11).
Unfortunately, incorporation of the dichlorindazole led to a
substantial increase in lipophilicity which was accompanied by a
corresponding increase in hERG and sodium channel
inhibition. A concentration response in patch clamp hERG
and sodium channel indicated an IC50 ∼ 0.3 μM at both targets.
However, given its high microsomal stability and exquisite
potency at both receptors, 9 was prioritized for advancement
into efficacy studies in order to relate the ion channel liability to
the exposures required to achieve in vivo efficacy.
The forced swim test (FST) is a validated model of
antidepressant efficacy23 which is based on the premise that
rodents forced to swim in an inescapable environment will
eventually become immobile (despair), restricting movements
to only those which prevent drowning. After administration of
the test compound, the animal is placed into an inescapable
transparent tank of water and scored for the characteristic
behavioral despair response (immobility). NK1 receptor
radioligand binding studies have shown significant species
differences in affinity for individual receptor antagonists. While
mouse and rat receptor binding have proven poor predictors for
human pharmacology, the gerbil has demonstrated good
correlation with human.24 Indeed, potency at gerbil NK1 and
SERT were found to be in good agreement with human (gerbil
IC50s = 0.30 nM and 8.0 nM, respectively). As such, we
employed gerbils in our forced swim test experiments. We were
delighted to find that compound 9 was active in this depression
model with a minimum effective dose (MED) of 3 mg/kg
C
DOI: 10.1021/acschemneuro.6b00337
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Letter
(Figure 3). In parallel with this experiment, exposure was
obtained using satellite animals which indicated excellent brain
performed with compound 9 across a range of doses (Figure 4).
3
125I-substance P and H-citalopram were used to assess NK1R
and SERT occupancy, respectively. Appreciable occupancy was
observed at both receptors at doses ≥1 mg/kg, and, as
expected, the relative in vitro potency at each target translated
to greater NK1 receptor occupancy at all doses. At the MED (3
mg/kg), NK1R and SERT occupancy were determined to be
88% and 20%, respectively (Figure 4).
The oral bioavailability of compound 9 was high (≥62%)
across preclinical species (mouse, rat, dog, cyno). A human
pharmacokinetic projection was prepared using simple
allometry, yielding a prediction of low clearance (6.1 mL/
min/kg) and moderate volume of distribution (4.9 L/kg). A 2
mg/kg dose (140 mg for a 70 kg human) anticipated 24 h
target coverage (>120 nM plasma concentration at trough;
MED in gFST) and a Cmax = 260 nM (2.3 nM free) at steady
state (attained ∼48 h after initiation). This corresponded to
>100× margins against both hERG and sodium ion channels.
Follow-up studies in rabbit demonstrated that there were no
effects on cardiac repolarization or cardiac conduction,
indicating no hERG or sodium channel inhibition-related
effects up to 10 mg/kg [Cmax = 3.5 μM (∼70 nM free)].25
Having demonstrated acceptable margins against ion channel
liabilities, a scale-up of compound 9 was initiated in support of
toxicology studies (Scheme 1).
Figure 3. Gerbil forced swim test. Efficacy of compound 9 versus
positive control (fluoxetine @ 10 mg/kg).
uptake (B:P ∼ 7). At the MED, the total plasma concentration
was 120 53 nM and the total brain concentration of 9 was
809 284 nM.
To assess the absolute and relative level of engagement at
both targets, ex vivo occupancy experiments in gerbils were
The synthesis of 9 began with diazotization of commercially
available aniline 10. Quenching of the diazonium with t-butyl
thiol followed by base-induced cyclization afforded indazole 11
in nearly quantitative yield. The two major fragments were
united via EDC-mediated esterification of carboxylic acid 11
with alcohol 12.25 Selective protection of N2 of the indazole
was achieved using a combination of SEMCl and dicyclohex-
ylmethylamine26 to afford 14 as a single regioisomer. The key
step of the synthesis was installation of the stereogenic methyl
group. We opted to pursue a Petasis methylenation27/
asymmetric hydrogenation strategy. The Petasis methylenation
proceeded uneventfully to afford enol ether 15. While we could
find no precedent for the use of a heterocycle to direct an
asymmetric hydrogenation, we reasoned that N1 of the
indazole could serve as a three atom donor which, in
conjunction with a chiral hydrogenation catalyst, might
Figure 4. Neurokinin 1 receptor and serotonin transporter occupancy
in gerbil.
a
Scheme 1. Optimized Route to Compound 9
a
Reagents and conditions: (a) (i) NaNO2, HCl; (ii) t-BuSH; (iii) KOt-Bu (97%); (b) EDC, DMAP (81%); (c) SEMCl, N,N-
dicyclohexylmethylamine (88%); (d) Petasis reagent, PhCH2C(CH3)2OAc; (e) Ru(OAc)2[(R)-p-TolBINAP], 100 psi H2 (85%, two steps); (f)
NCS then HCl (86%); (g) HCHO, NaHB(OAc)3, DIEA (99%).
D
DOI: 10.1021/acschemneuro.6b00337
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Letter
facilitate delivery of hydrogen to a single face of the substrate.
In collaboration with our process research group, a catalyst
screen was performed using enol ether 15. In all, 25 catalysts
were screened in each of two solvents (methanol and
dichloroethane). We were delighted to find that a Ru-based
BINAP catalyst in DCE gave excellent conversion, affording the
product in high optical purity (>99.8% ee). It is well-known
that this class of catalysts require a coordinating group in close
proximity to the site of hydrogenation in order to achieve high
conversion and high enantioselectivity,28 thus supporting our
original hypothesis that the heterocycle would coordinate to the
chiral catalyst to deliver hydrogen to a single face.
Having demonstrated proof of concept for the key step on a
small scale, we initiated the synthesis of a 70 g lot of 9 by this
route. While the Petasis methylenation had proceeded
smoothly using a standard protocol on a small scale, upon
scale-up, a substantial amount of an ethylidene byproduct25 was
observed. It was found that the byproduct could be suppressed
by addition of 2-methyl-1-phenylpropan-2-yl acetate (PhCH2C-
(CH3)2OAc) which served as a sacrificial ester, reacting only
after enol ether 15 had been consumed.29 After an extractive
workup and a solvent switch from toluene to dichloroethane,
the crude enol ether could be directly used in the asymmetric
hydrogenation. This two-step Petasis/hydrogenation sequence
was repeated successfully on batches that employed as much as
46 g of ester 14. Installation of the second chlorine with NCS
in acetic acid occurred with partial loss of the Boc group. A
two-step one pot procedure was employed, adding HCl in
isopropanol when chlorination was complete. This effected
removal of both protecting groups and had the added benefit of
allowing isolation of 17 as the crystalline HCl salt directly from
the reaction mixture. Reductive amination using formalin and
sodium triacetoxyborohydride completed the synthesis of 9.
This route was employed to successfully deliver 70 g of the
compound of interest.
acceptable safety margins with respect to hERG and sodium
channels, margins in a two-week rat toxicology study precluded
further advancement of this novel NK1/SERT dual inhibitor.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acschemneur-
o.6b00337.
Detailed experimental procedures for the preparation of
9, characterization data for compounds 1−8, supple-
mental pharmacokinetic and toxicology data, and
experimental details of biological assays (PDF)
AUTHOR INFORMATION
Corresponding Author
*Tel: 1-203-677-5750. Fax: 1-203-677-7884. E-mail: andrew.
degnan@bms.com.
■
Author Contributions
A.P.D., J.J.B., J.E.M., and K.W.G. oversaw analogue design.
A.P.D., G.O.T, and H.H. performed the synthesis of analogues.
A.P.D., D.A.C., X.H., Y.H., T.R., J.-H.S., M.K.W., and H.Z.
contributed the scale-up of compound 9 for toxicology studies.
C.D.D., J.H., J.R., and S.J.T. contributed to the design and
analysis of pharmacokinetic experiments. U.M.H. designed and
analyzed toxicology experiments. R.K. performed NK1 and
SERT binding assays. M.T.T. and N.J.L. oversaw the design and
execution of in vitro and in vivo biological experiments. A.E.N.
performed and analysed in vivo biology experiments. Y.-W.L.
and R.L.P. generated occupancy data. A.P.D. wrote the
manuscript with input from all authors.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank members of Discovery Analytical Sciences for
detailed characterizations and optical purity determinations.
In toxicology studies, a single dose of 9 was well tolerated in
mouse and rat up to 300 mg/kg (Cmax = 29 and 9.1 μM,
respectively).25 In telemeterized rats, a single dose of 100 mg/
kg did not alter blood pressure, heart rate, or locomotor activity
(Cmax = 7.3 μM). Unfortunately, upon repeat dosing for 2
weeks in rats, significant toxicity was observed. At 100 mg/kg,
significant increases in aspartate- and alanine-aminotransferase
(∼5× and ∼3× control, respectively) were observed.
Histopathology revealed extensive tissue vacuolation. The
frequency, severity, and the number of tissue types impacted
increased with dose and was observed in all dose groups (10,
30, 100, and 300 mg/kg). The vacuolation was consistent with
phospholipidosis which was subsequently confirmed via
transmission electron microscopy.25 At the lowest observed
effect level (LOEL, 10 mg/kg), exposure was only 5−8× the
projected human exposure and further progression of 9 was not
pursued due to an insufficient margin for safety.
In conclusion, an analysis of the structure activity relation-
ships of our dual NK1/SERT biphenyl chemotype suggested
that a heterocyclic tie-back might be tolerated. This strategy
was pursued to incorporate an indazole which reduced cLogP,
mitigated ion channel inhibition, and reduced protein binding.
Early leads from this series were optimized to afford 9, a
compound with favorable oral bioavailability, excellent brain
uptake, and robust in vivo efficacy. Subsequent scale-up of 9
required development of a novel heterocycle-directed asym-
metric hydrogenation to afford the key stereogenic center in
high yield and excellent optical purity. While 9 demonstrated
■
ABBREVIATIONS
■
CNS, central nervous system; FST, forced swim test; hERG,
human ether-a-go-go-related gene; LipE, lipophilic ligand
efficiency; MED, minimum efficacious dose; NCS, N-
chlorosuccinimide; NK1, neurokinin 1; PPB, plasma protein
binding; SAR, structure−activity relationship; SERT, serotonin
transporter; SSRI, serotonin-selective reuptake inhibitor
REFERENCES
■
(1) Hippocrates (400 B.C.E.) Aphorisms, Section 6.23.
(2) Bascome, E. (1852) On the nature and causes of fever, especially
that termed yellow fever, John Churchill, London.
(3) Coppen, A. (1967) The biochemistry of affective disorders. Br. J.
Psychiatry 113, 1237−1264.
(4) Coppen, A., Shaw, D. M., and Farrell, J. P. (1963) Potentiation of
the antidepressive effect of a monoamine-oxidase inhibitor by
tryptophan. Lancet 281, 79−81.
(5) Sullivan, F. P., Neale, M. C., and Kendler, K. S. (2000) Genetic
epidemiology of major depression: review and meta-analysis. Am. J.
Psychiatry 157, 1552−1562.
(6) (a) Tsankova, N., Renthal, W., Kumar, A., and Nestler, E. J.
(2007) Epigenetic regulation in psychiatric disorders. Nat. Rev.
Neurosci. 8, 355−367. (b) Mill, J., and Petronis, A. (2007) Molecular
studies of major depressive disorder: the epigenetic perspective. Mol.
Psychiatry 12, 799−814.
E
DOI: 10.1021/acschemneuro.6b00337
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Letter
(7) (a) Cadoret, R. J., O’Gorman, T. W., Heywood, E., and
Troughton, E. (1985) Genetic and environmental factors in major
depression. J. Affective Disord. 9, 155−164. (b) Kendler, K. S.,
Karkowski, L. M., and Prescott, C. A. (1999) Causal relationship
between stressful life events and the onset of major depression. Am. J.
Psychiatry 156, 837−841.
(8) (a) Nestler, E. J., Barrot, M., DiLeone, R. J., Eisch, A. J., Gold, S.
J., and Monteggia, L. M. (2002) Neurobiology of depression. Neuron
34, 13−25. (b) Krishnan, V., and Nestler, E. J. (2008) The molecular
neurobiology of depression. Nature 455, 894−902.
(9) (a) Libby, A. M., Orton, H. D., and Valuck, R. J. (2009)
Persisting decline in depression treatment after FDA warnings. Arch.
Gen. Psychiatry 66, 633−639. Supporting Information (b) Noh, E. H.
(2013) Use and Unintended Consequences of Antidepressant Medications
by Depressed Older Adults, Dissertation, University of Rhode Island,
Kingston, RI.
potent, tunable dual neurokinin 1 receptor antagonists and serotonin
transporter inhibitors. Bioorg. Med. Chem. Lett. 25, 3039−3043.
(19) Jamieson, C., Moir, E. M., Rankovic, Z., and Wishart, G. (2006)
Medicinal chemistry of hERG optimizations: Highlights and hang-ups.
J. Med. Chem. 49, 5029−5046 and references therein..
(20) Soloshonok, V. A., and Yasumoto, M. (2006) Simple and
convenient synthesis of 3,5-bis-(trifluoromethyl)benzylamine via
biomimetic 1,3-proton shift reaction. J. Fluorine Chem. 127, 889−893.
(21) All four approved NK1 receptor antagonists retain the bis-
trifluoromethylphenyl moiety.
(22) Leeson, P. D., and Springthorpe, B. (2007) The influence of
drug-like concepts on decision-making in medicinal chemistry. Nat.
Rev. Drug Discovery 6, 881−890.
(23) (a) Wallace-Boone, T. L., Newton, A. E., Wright, R. N., Lodge,
N. J., and McElroy, J. F. (2008) Behavioral and pharmacological
validation of the gerbil forced-swim test: effects of neurokinin-1
receptor antagonists. Neuropsychopharmacology 33, 1919−1928.
(b) Porsolt, R. D., Le Pichon, M., and Jalfre, M. (1977) Depression:
a new animal model sensitive to antidepressant treatments. Nature
266, 730−732.
(24) Gitter, B. D., Waters, D. C., Bruns, R. F., Mason, N. R., Nixon, J.
A., and Howbert, J. J. (1991) Species differences in affinities of non-
peptide antagonists for substance P receptors. Eur. J. Pharmacol. 197,
237−238.
(25) See the Supporting Information for additional detail.
(26) Luo, G., Chen, L., and Dubowchik, G. (2006) Regioselective
protection at N-2 and derivatization at C-3 of indazoles. J. Org. Chem.
71, 5392−5395.
(27) Petasis, N. A., and Bzowej, E. I. (1990) Titanium-mediated
carbonyl olefinations. 1. Methylenations of carbonyl compounds with
dimethyltitanocene. J. Am. Chem. Soc. 112, 6392−6394.
(28) Noyori, R., Kitamura, M., and Ohkuma, T. (2004) Toward
Efficient Asymmetric Hydrogenation: Architectural and Functional
Engineering of Chiral Molecular Catalysts. Proc. Natl. Acad. Sci. U. S. A.
101, 5356−5362.
(29) Payack, J. F., Huffman, M. A., Cai, D., Hughes, D. L., Collins, P.
C., Johnson, B. K., Cottrell, I. F., and Tuma, L. D. (2004)
Dimethyltitanocene: From Millimole to Kilomole. Org. Process Res.
Dev. 8, 256−259.
(10) (a) Kasper, S., Fuger, J., and Moller, H.-J. (1992) Comparative
̈
efficacy of antidepressants. Drugs 43, 11−23. (b) Georgotas, A.,
McCue, R. E., Hapworth, W., Friedman, E., Kim, O. M., Welkowitz, J.,
Chang, I., and Cooper, T. B. (1986) Comparative efficacy and safety of
MAOIs versus TCAs in treating depression in the elderly. Biol.
Psychiatry 21, 1155−1166.
(11) Trivedi, M. H., Rush, A. J., Wisniewski, S. R., Nierenberg, A. A.,
Warden, D., Ritz, L., Norquist, G., Howland, R. H., Lebowitz, B.,
McGrath, P. J., Shores-Wilson, K., Biggs, M. M., Balasubramani, G. K.,
and Fava, M. (2006) Evaluation of outcomes with citalopram for
depression using measurement-based care in STAR*D: implications
for clinical practice. Am. J. Psychiatry 163, 28−40.
(12) Thompson, C. (2002) Onset of action of antidepressants:
results of different analyses. Hum. Psychopharmacol. 17, S27−S32.
(13) Froger, N., Gardier, A. M., Moratalla, R., Alberti, I., Lena, I.,
Boni, C., De Felipe, C., Rupniak, N. M. J., Hunt, S. P., Jacquot, C.,
Hamon, M., and Lanfumey, L. (2001) 5-hydroxytryptamine (5-HT)1A
autoreceptor adaptive changes in substance P (neurokinin 1) receptor
knock-out mice mimic antidepressant-induced desensitization. J.
Neurosci. 21, 8188−8197.
(14) (a) Hamon, M. (1997) The main features of central 5-HT1A
receptors. In Serotoninergic neurons and 5-HT receptors in the CNS.
Handbook of experimental pharmacology (Baumgarten, H. G., and
Gothert, M., Ed.), pp 239−268, Springer, Berlin. (b) Guiard, B. P.,
̈
Froger, N., Hamon, M., Gardier, A. M., and Lanfumey, L. (2005)
Sustained pharmacological blockade of NK1 substance P receptors
causes functional desensitization of dorsal raphe 5-HT 1A
autoreceptors in mice. J. Neurochem. 95, 1713−1723.
(15) (a) Loane, C., and Politis, M. (2012) Buspirone: What is it all
about? Brain Res. 1461, 111−118. (b) Celada, P., Puig, M. V.,
́
Amargos-Bosch, M., Adell, A., and Artigas, F. (2004) The therapeutic
role of 5-HT1A and 5-HT2A receptors in depression. J. Psychiatr.
Neurosci. 29, 252−265. (c) Gardier, A. M., Malagie, I., Trillat, A. C.,
́
Jacquot, C., and Artigas, F. (1996) Role of 5-HT1A autoreceptors in
the mechanism of action of serotoninergic antidepressant drugs: recent
findings from in vivo microdialysis studies. Fundam. Clin. Pharmacol.
10, 16−27.
(16) (a) 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., and
Lodge, N. J. (2013) NK1 receptor antagonism lowers occupancy
requirement for antidepressant-like effects of SSRIs in the gerbil forced
swim test. Neuropharmacology 73, 232−240. (b) Chenu, F., Guiard, B.
P., Bourin, M., and Gardier, A. M. (2006) Antidepressant-like activity
of selective serotonin reuptake inhibitors combined with a NK1
receptor antagonist in the mouse forced swimming test. Behav. Brain
Res. 172, 256−263.
(17) Morphy, R., and Rankovic, Z. (2005) Designed multiple ligands.
An emerging drug discovery paradigm. J. Med. Chem. 48, 6523−6543.
(18) Degnan, A. P., Tora, G. O., Han, Y., Rajamani, R., Bertekap, R.,
Krause, R., Davis, C. D., Hu, J., Morgan, D., Taylor, S. J., Krause, K., Li,
Y.-W., Mattson, G., Cunningham, M. A., Taber, M. T., Lodge, N. J.,
Bronson, J. J., Gillman, K. W., and Macor, J. E. (2015) Biaryls as
F
DOI: 10.1021/acschemneuro.6b00337
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX