Identification, biological characterization and pharmacophoric analysis of a new potent and selective NK1 receptor antagonist clinical candidate

Article abstract of DOI:10.1016/j.bmc.2013.09.001

The last two decades have provided a large weight of preclinical data implicating the neurokinin-1 receptor (NK₁) and its cognate ligand substance P (SP) in a broad range of both central and peripheral disease conditions. However, to date, only the NK₁ receptor antagonist aprepitant has been approved as a therapeutic and this is to prevent chemotherapy-induced nausea & vomiting (CINV). The belief remained that the full therapeutic potential of NK₁ receptor antagonists had yet to be realized; therefore clinical evidence that NK₁ receptor antagonists may be effective in major depression disorder, resulted in a significant further investment in discovering novel CNS penetrant druggable NK₁ receptor antagonists to address this condition. At GlaxoSmithKline after the discovery of casopitant, that went on to demonstrate efficacy as a novel antidepressant in the clinic, additional novel analogues of this NK₁ receptor antagonist were designed to further enhance its drug developability characteristics. Herein, we therefore describe the discovery process and the vivo pharmacological and pharmacokinetic profile of the new NK₁ receptor antagonist 3a (also called orvepitant), selected as clinical candidate and further progressed into clinical studies for major depressive disorder. Moreover, molecular modeling studies enabled us to improve the pharmacophore model of the NK₁ receptor antagonists with the identification of a region able to accommodate a variety of heterocycle moieties.

Full text of DOI:10.1016/j.bmc.2013.09.001

Bioorganic & Medicinal Chemistry 21 (2013) 6264–6273
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Identification, biological characterization and pharmacophoric
analysis of a new potent and selective NK1 receptor antagonist
clinical candidate
,
⇑
Romano Di Fabio , Giuseppe Alvaro à, Simone Braggio , Renzo Carletti , Philip A. Gerrard ,
Cristiana Griffante , Carla Marchioro , Alfonso Pozzan , Sergio Melotto §, Alessandro Poffe ,
Laura Piccoli , Emiliangelo Ratti k, Elvira Tranquillini –, Michael Trower k, Simone Spada , Mauro Corsi
Neurosciences Centre of Excellence for Drug Discovery, Chemical Development and Molecular Discovery Research, GlaxoSmithKline Medicines Research Centre, Via A. Fleming 4, 37135
Verona, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 June 2013
Revised 29 August 2013
Accepted 1 September 2013
Available online 11 September 2013
The last two decades have provided a large weight of preclinical data implicating the neurokinin-1 recep-
tor (NK1) and its cognate ligand substance P (SP) in a broad range of both central and peripheral disease
conditions. However, to date, only the NK1 receptor antagonist aprepitant has been approved as a ther-
apeutic and this is to prevent chemotherapy-induced nausea & vomiting (CINV). The belief remained that
the full therapeutic potential of NK1 receptor antagonists had yet to be realized; therefore clinical
evidence that NK1 receptor antagonists may be effective in major depression disorder, resulted in a sig-
nificant further investment in discovering novel CNS penetrant druggable NK1 receptor antagonists to
address this condition. At GlaxoSmithKline after the discovery of casopitant, that went on to demonstrate
efficacy as a novel antidepressant in the clinic, additional novel analogues of this NK1 receptor antagonist
were designed to further enhance its drug developability characteristics. Herein, we therefore describe
the discovery process and the vivo pharmacological and pharmacokinetic profile of the new NK1 receptor
antagonist 3a (also called orvepitant), selected as clinical candidate and further progressed into clinical
studies for major depressive disorder. Moreover, molecular modeling studies enabled us to improve
the pharmacophore model of the NK1 receptor antagonists with the identification of a region able to
accommodate a variety of heterocycle moieties.
Keywords:
Neurokinin
Substance P
G-protein coupled receptor
Central nervous system
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
The undecapeptide substance P (SP) is one of the most studied
neurotransmitters in the CNS. Its pharmacological action is medi-
ated by the preferential interaction with the 7-transmembrane
G-coupled protein (GPCR) neurokinin 1 (NK1) receptor. This recep-
tor belongs to the tachykinin superfamily that consists of three
receptor subtypes (NK1, NK2 and NK3) and is expressed widely in
the mammalian nervous system, spinal cord and peripheral
tissues.1–3 A large body of both pre-clinical and clinical evidence
has linked the NK1 receptor/SP system with a plethora of patholog-
ical conditions, ranging from pain, migraine, asthma, nausea,
inflammatory bowel syndrome, cancer and urinary incontinence
to anxiety and depression.4–13
Since the publication of the first non-peptidic NK1 receptor
antagonist in the early 1990s, there has been much effort to
discover drug-like NK1 receptor antagonists and bring them
through clinical trials to the market. To date aprepitant14 is the first
and only NK1 receptor antagonist to gain approval by the FDA; this
Abbreviations: NK, neurokinin; SP, substance P; GPCR,
G protein-coupled
receptor; CNS, central nervous system; SAD, social anxiety disorders; CINV,
chemotherapy-induced nausea and vomiting; PoC, prove of concept; SAR, struc-
ture–activity relationship; CDI, carbonyldiimidazole; THF, tetrahydrofuran; CHO,
Chinese hamster ovary; FLIPR, fluorescence imaging plate reader; GFT, gerbil foot
tapping; CRC, concentration–response curve; SSRIs, selective serotonin reuptake
inhibitors; HTT, human threat test; SI, social interaction; FDA, food and drug
administration; i.v., intravenous; p.o., per os; Clp, clearance plasmatic.
⇑
Corresponding author. Tel.: +39 0458218879.
E-mail address: romano.difabio@aptuit.com (R. Di Fabio).
Present address: Aptuit S.r.l., Via A. Fleming 4, 37135 Verona, Italy.
Present address: Autifony S.r.l., Via A. Fleming 4, 37135 Verona, Italy.
à
§
Present address: Ablicon S.r.l., Via L. Einaudi 29, 62012 Civitanova Marche (MC),
Italy.
–
Present address: Sandoz Industrial Products S.p.A., Corso Verona 165, 38068
Rovereto (TN), Italy.
Present address: NeRRe Therapeutics Ltd, Gunnels Wood Road, Stevenage SG1
2FX, UK.
k
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.09.001

R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
6265
occurred in 2003 to control chemotherapy-induced nausea and
vomiting (CINV) and, later in 2006, as a therapy for post-operative
nausea and vomiting (PONV). However in the last fifteen years on
the basis of initially encouraging preclinical data, several NK1
receptor antagonists have been progressed into the clinic as poten-
tial antidepressant drugs.15 The results from these studies have
been controversial, as although efficacy in depression studies was
shown initially with brain penetrant, high affinity, selective NK1
receptor antagonists including aprepitant in their proof of concept
(PoC) trials;14,16,17 the Phase III program with aprepitant failed.18
This led to the development of a working hypothesis that full
and long lasting saturation of the central NK1 receptor compart-
ment is a critical requirement for the antidepressant effects of
NK1 receptor antagonists to be realized.19 This theory was first
tested in a depression clinical study with, an orally active, highly
potent, selective and non-surmountable NK1 receptor antagonist
that we have previously described and termed casopitant20
(GW679769), designated as (2R,4S)-4-(4-acetylpiperazin-1-yl)-N-
[(1R)-1-[3,5-bis(trifluoromethyl)phenyl]ethyl]-2-(4-fluoro-2-meth-
ylphenyl)-N-methylpiperidine-1-carboxamide amine (see Figure 1).
In the Phase II study casopitant delivered both a statistically and
clinically significant antidepressant effect on the primary endpoint
at a dose that ensured full (>99%) and sustained (24 h coverage)
receptor occupancy.19 To further improve on casopitant’s profile
we aimed to identify related analogues that had an enhanced met-
abolic stability and more favorable pharmacokinetic profile in pre-
clinical animal species. Herein we describe the synthesis and the
pharmacological and pharmacokinetic profiles of novel 4-substi-
tuted-octahydro-1H-pyrrolo[3,2-c]pyridine derivatives, shown in
Figure 1. In addition, we report the results of some molecular mod-
eling studies we performed by comparing the broad set of NK1
receptor antagonist disclosed in literature shown in Figure 5 with
the new NK1 receptor antagonist described in this paper. This anal-
ysis confirmed that the NK1 receptor antagonists adopt a U-shaped
bioactive conformation and enabled us to identify the presence of a
bulk tolerant site, able to accommodate a variety of both hydro-
philic and hydrophobic moieties, as a region useful for designing
novel sub-classes of NK1 receptor antagonist analogues.
afford pure enantiomers 1a and 1b.23 The following reductive ami-
nation performed on the known ketone intermediate 2,20 with
amino bicycle derivatives 1a and 1b in the presence of NaBH(OAc)3
in dry CH3CN, gave a mixture of epimers 3a and 3c and 3b and 3d,
respectively, which were separated by flash chromatography to
give title compounds.
2.2. Molecular modeling studies
NK1 receptor antagonist structures shown in Figure 5 where
used to build a common hypothesis pharmacophore applying the
pharmacophore elucidate module available in the MOE software,53
as reported in the experimental section. Then, compound 3a was
fitted into this pharmacophoric model to identify both structural
similarities and differences with respect to the set of reference
molecules selected in literature.
3. Discussion
Previous studies had shown that casopitant after a single oral
administration in rats, was eliminated almost exclusively by loss
of the N-acetyl group and the oxidation of the piperazine ring,
leading to the cleavage of this heterocycle moiety ring.24–26 There-
fore, our design approach in preparing novel casopitant analogues
was to modify the N-acetyl piperazine moiety of the molecule by
synthesizing the corresponding more rigid bicycle c-lactam deriv-
ative, shown in Figure 1. The four diasteroisomers 3a, 3b, 3c and
3d, as depicted in Scheme 1, obtained as described above, were
characterized in the in vitro assays by increasing antagonist
concentrations both in terms of receptor binding profile by dis-
placement of [3H]SP on membranes derived from CHO cells
expressing the recombinant human NK1 (hNK1) receptor and
antagonism of SP-induced release of cytosolic Ca2+ in hNK1–CHO
cells by using fluorometric imaging plate reader (FLIPR)
technology.
As reported in Table 1, compounds 3a and 3c (1 pM–10 lM),
exhibited the highest in vitro potency in the receptor binding stud-
ies at the hNK1 receptor (pKi = 10.20 0.10 and 10.19 0.13,
respectively; n = 3 experiments). Based on these initial encourag-
ing results, both compounds were further characterized in terms
of the ability to functionally inhibit SP-induced release of cytosolic
Ca2+ in hNK1–CHO cells. Compounds 3a and 3c (0.3–10 nM),
pre-incubated for 1 h at 37 °C before adding the agonist SP, as
shown in Figure 2, each produced a non-surmountable antagonism
of agonist CRC. For 3a and 3c apparent pKB values of 10.30 0.30
and 10.50 0.20 (n = 4), respectively, were calculated by using
the dose ratio of the agonist concentration response obtained in
2. Materials and methods
2.1. Chemistry
The racemic bicycle c-lactam moiety 1, shown in Scheme 1, was
synthesized as previously reported.21,22 The resolution of the
racemate was performed by preparative chiral HPLC (Chiralpack
AD ꢀ 2 cm; n-hexane/EtOH 8:2; flux = 1 ml/min; k = 225 nm) to
cyclize
O
N
O
N
N
N
CF3
CF3
N
N
modify
N
N
N
N
CF3
CF3
O
O
F
F
A (casopitant)
3a-d
Figure 1. Modification of the piperazine moiety.

6266
R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
O
N
CF3
N
N
N
CF3
CF3
O
O
O
b,c
N
N
N
CF3
+
3a
CF3
F
O
O
N
H
N
N
1a
F
a
N
N
2
CF3
O
O
N
F
3c
N
H
1
O
CF3
N
N
CF3
a
O
O
N
N
b,c
CF3
N
N
N
O
CF3
+
O
N
H
3b
F
1b
F
2
O
CF3
N
N
N
N
CF3
O
F
3d
Scheme 1. (a) Chiralpack AD ꢀ 2 cm, n-hexane/EtOH 8:2, flux = 1 ml/min, k = 225 nm; (b) compound 2, dry CH3CN, NaBH(OAc)3; (c) HCl 1 M in Et2O, 30 min, 0 °C.
Table 1
1 mg/kg dose of 3a and 3c, respectively. Therefore both compounds
3a and 3c showed good pharmacokinetic properties when com-
pared with casopitant,20 with lower Clp, longer half-life, higher
oral bioavailability and lower variability. However, compound 3a
was selected for progression over 3c because of the more favorable
prediction of the human PK profile assessed in allometric scaling
studies. Compound 3a was characterized in vivo in the gerbil foot
tapping (GFT) pharmacodynamic model and later on, in the human
threat test (HTT), an anxiety model in marmoset. Prior to perform-
ing these studies, receptor binding experiments were performed
on homogenate derived membranes prepared from gerbil and mar-
moset brain cortices following the general membrane preparation
protocol used with hNK1–CHO cells, with [3H]-GR205171, a highly
selective radioligand for the NK1 receptor, instead of [3H]-SP, as
In vitro binding affinities (pKi) and functional potencies (pKB) at human-NK1 receptors
for NK1 receptor antagonists
Compound
pKia
n
pKB
3a
3b
3c
3d
10.20 0.10
9.70 0.04
10.19 0.13
9.86 0.05
3
2
4
2
10.30 0.30
NTb
4
4
10.50 0.20b
NTb
Data are the mean SE of n experiments.
a
pKi and pKB values have been determined as described in the experimental
section and are the mean SEM of n experiments performed in duplicate.
b
NT = not tested. n = number of experiments in duplicate.
the absence and presence of 1 nM con animal’s back is turned to
the observer with elevation of the tail to expose the centration of
antagonist.
previously described.27 Compound 3a (1 pM–1
lM), tested in
filtration binding experiments, completely displaced [3H]-
GR205171 specific binding with pKi values of 10.5 0.2 (n = 3
experiments) on gerbil and 10.4 0.1 (n = 4 experiments) on mar-
moset brain homogenates. In addition, 3a was evaluated in terms
of its pharmacokinetic profile in gerbils, showing a B/P ratio of
1.2 and a total brain concentration of 91 ng/g when administered
p.o. at 1 mg/kg. The gerbil foot tapping (GFT) model is one of the
preferred in vivo assay to screen NK1 receptor antagonists.28 This
model is characterized by a typical intense foot tapping behavior
induced by intracerebroventricular (icv) administration of selec-
tive NK1 receptor agonists (e.g. GR73632).29,30 This behavior is
antagonized by CNS penetrating NK1 receptor antagonists. In this
As reported in Table 5, compounds 3a and 3c exhibited good
pharmacokinetics in vivo, both in rat and dog, although 3a was
more bioavailable and metabolically stable in dog than 3c. Indeed,
3a showed an oral bioavailability (F) of 17% in rat and 55% in dog,
plasma clearance (Clp) of 29 ml/min/kg in rat and 6 ml/min/kg in
dog and a half-life of 2.3 h in rat and 6.1 h in dog. Compound 3c
showed instead an oral bioavailability of 19% in rat and 28% in
dog, a plasma clearance (Clp) of 25 ml/min/kg in rat and 11 ml/
min/kg in dog and a half-life of 2.8 h in rat and 5.1 h in dog. As
far as the brain penetration in rats is concerned, a B/P ratio of 1.2
and 0.9 was observed 5 min after the i.v. administration of a

R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
6267
150
100
50
150
100
50
0
0
-13 -12 -11 -10
-9
-8
-7
-6
-13 -12 -11 -10
-9
-8
-7
-6
SP (logM)
SP (logM)
Figure 2. Increase of cytosolic calcium on hNK1–CHO cells. Antagonism of 3a (left) and 3c (right) on SP-induced release of cytosolic Ca2+ in hNK1–CHO. Control (j), 3a and 3c
0.3 nM (N), 3a and 3c 1 nM (.), 3a and 3c 3 nM (ꢀ) and 3a and 3c 10 nM (d). Data are the mean SEM of four experiments performed in duplicate.
Table 2
Table 3
Effect of 3a on GFT test at different times (hours) after oral pre-treatment
Number of postures and their reduction (% of the reduction of the number of
postures) induced by 3a in the HTT in marmoset
Compound
ꢁ1 h
0.11
ꢁ4 h
0.09
ꢁ8 h
0.09
ꢁ24 h
3.36
Doses (po)
Number of
postures
Reduction of the
number of postures (%)
3a
(0.07–0.17)
(0.06–0.13)
(0.04–0.32)
(2.60–4.40)
Vehicle
0.3 mg/kg
Vehicle
1.0 mg/kg
Vehicle
3.0 mg/kg
Vehicle
12.8 2.2
11.3 1.3
12.3 1.5
8.0 0.6a
11.8 1.3
7.5 1.3a
11.2 1.3
6.0 1.0a
—
11.8
—
34.9
—
36.6
—
Data are presented as ID50 values (compound dose causing 50% inhibition of the
tapping behavior elicited by the icv injection of GR73632, 3 pmol/5
95% C.L.
lL) followed by
model, compounds 3a (Table 2) exhibited an excellent, dose
dependent inhibition of foot tapping in gerbils after oral adminis-
tration with an ID50 value of 0.11 mg/kg (and an ID50 = 0.07 mg/
kg after 1hr s.c. pre-treatment) versus GR73632-induced tapping
10 mg/kg
46.4
Data are the mean SEM.
a
P <0.05 compared to vehicle.
(3 pmol/5 lL). Notably, a long lasting effect was observed, the
compound still being active 8 and 24 h after its oral administration
(ID50 = 0.09 mg/kg and 3.36 mg/kg, respectively), providing evi-
dence of the presence of compound 3a in the brain several hours
after its oral administration. Preclinical studies have reported that
NK1 receptor antagonists exhibited anxiolytic-like effects in
preclinical in vivo models of anxiety.31 Clinical evidence suggest
that depression is co-morbid with anxiety in up to 60% of patients
suffering for major depressive disorders. Many symptoms overlap
and more than 50% of patients with generalized anxiety disorder
and more than 30% of those experiencing social anxiety disorder,
meet the criteria for major depressive disorders. Furthermore,
SSRIs are common drugs to treat both anxiety and major depres-
sion disorders, thus suggesting that similar neurobiological altera-
tions may occur in the two diseases. On this basis, we decided to
use preclinical anxiety models, namely: human threat test in the
marmoset (HTT) and social interaction (SI) in gerbils, as surrogate
models to predict for potential effects of compound 3a in major
depression disorders. As far as the HTT model in marmosets is con-
cerned, it has been suggested that the number of territorial pos-
tures elicited by marmosets, when confronted with a human
observer in close proximity to the home cage, may reflect the level
of stress induced in the animals.32,33 Compound 3a was tested at
the doses of 0.3, 1, 3 and 10 mg/kg p.o. (1 h pretreatment) and,
as reported in Table 3 and depicted in Figure 3, a dose dependant
reduction of the number of postures was observed at 1 mg/kg
(34.9% reduction vs vehicle treated animals; P <0.05), 3 mg/kg
(36.6% reduction vs vehicle treated animals; P <0.05) and 10 mg/
kg (46.4% reduction vs vehicle treated animals; P <0.05), suggesting
a potential anxiolytic-like effect of the compound. Then, to explore
whether the pharmacodynamic effect of NK1 receptor antagonists
was, as expected, dependent on their level of receptor occupancy
3a was tested in the social interaction34 (SI) model in gerbils and
the behavioral readouts were correlated with the level of receptor
engagement measured as compound receptor occupancy in gerbil
striatum by using an ex vivo autoradiography method with the
No. of Postures
16
14
12
10
8
*
*
*
6
4
2
0
Vehicle 0.3 Vehicle 1 Vehicle
3
Vehicle 10
3a mg/kg p.o.
Figure 3. Number of postures and their reduction (% of the reduction of the number
of postures) induced by 3a (1 h pretreatment time) in the HTT in marmoset. Data
are the mean SEM. ⁄P <0.05 compared to vehicle.
Table 4
Effect of oral 3a (1 h pretreatment time) in SI test in gerbils; data are presented as
means SEM of 10 animals per group
Doses (po)
GSI Time (s)
% NK1 receptor occupancy (optical density)b
Vehicle
21.2 2.9
22.1 1.8
27.383.7
0% (55.3 1.2)
0.1 mg/kg
0.3 mg/kg
1.0 mg/kg
47.8% (28.9 3.3)
74.6% (14 3.6)
95.2% (2.7 0.03)
33.1 3.7a
a
P <0.05 versus vehicle control. The right column reports the optical densities
values of NK1 binding as means SEM of 3 animals per group.
b
Values of NK1 receptor occupancy are expressed as percentage: see paragraph
5.2.16.
radioligand [3H]-GR205171. Although, other models of anxiety
are available in rat or mouse, their use in pre-clinical research for
NK1 receptor antagonists is complicated by species variation in

6268
R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
Seconds
compound (Table 4 and Figure 4). NK1 receptor occupancy in the
40
gerbil striatum at the effective dose of 3a (1 mg/kg) in the SI test
was 95.2% whereas the non-effective doses of 0.3 and 0.1 mg/kg
produced levels of NK1 receptor occupancy of 74.6% and 47.8%,
respectively (Table 4). These results confirm that high levels
(>90%) of NK1 receptor occupancy are necessary to elicit a signifi-
cant pharmacodynamic effect in the SI model in gerbils.
Finally the in vitro receptogram screen (by MDS Pharma
Service, Taiwan) confirmed that 3a is highly selective for the
hNK1 receptor since at 10 lM concentration it produced less than
35
30
25
20
15
10
5
0
Vehicle
0.1
0.3
1
50% displacement of binding in the full battery of 67 receptors,
transporters, ion channels and enzymes with exception of the
3a mg/kg p.o.
following targets: sigma
sodium channel (site 2) for which the following pKi values were
estimated: sigma 1 (Ki = 0.94 M), platelet activating factor
M), sigma (Ki = 3.26 M), sodium channel
M). In light of the subnanomolar affinity shown for 3a
at human NK1 receptor (pKi = 10.2; Ki = 0.06 nM), it is estimated
that the selectivity for hNK1 receptor over all the receptors
screened by MDS Pharma Services is at least 10,000-fold.
Molecular modeling studies were performed on compound 3a
which was fitted into a pharmacophoric model created from the
set of known NK1 receptor antagonists, showed in Figure 5, with
the aim to identify both structural similarities and/or differences
amongst compound 3a and the reference compounds. In a previous
pharmacophore model hypothesis for NK1 receptor antago-
nists,2,28,37–40 the presence of two distal aryl groups, undertaking
r1, platelet activating factor, sigma r2,
Figure 4. Effect of oral 3a (1 h pretreatment time) in SI test in gerbils; data are
presented as means SEM of 10 animals per group. ⁄P <0.05 versus vehicle control.
r
l
(Ki = 2.27
(Ki = 4.54
l
l
r
2
l
Table 5
Preclinical pharmacokinetic parameters of compound 3a and 3c versus casopitant in
rats and dogsa,b
Compound
Clp ((ml/min)/kg)
Vss (l/kg)
Half-life (h)
F (%)
A (casopitant)
(Rat)
(Dog)
37
10
3.2
1.9
1.8
4.5
8
34
3a
(Rat)
(Dog)
29
6
3.7
2.0
2.3
6.1
17
55
3c
(Rat)
(Dog)
25
11
3.4
2.0
2.8
5.1
19
28
a
P
-stacking interaction, is considered critical to enable NK1 recep-
tor antagonists to adopt a U-shaped bioactive conformation.41
Other authors reported the presence of an edge- interaction
a
PK parameters derived from plasma concentrations.
The PK profile was assessed at 0.5 mg/kg i.v. and 1 mg/kg p.o.
b
P
between similar aryl moieties.42 Then, NMR studies on tripepdides
NK1 receptor antagonists confirmed that two distal aromatic rings
are stacked each other.43 In Figure 6 is reported the superimposi-
tion of compound 3a and the set of NK1 receptor antagonists
shown in Figure 5. These compounds share at least four pharmaco-
phoric features, namely: two distal aryl rings, an H-bond acceptor
group, present in the right hand side chain, and a heterocycle core
bearing an amine or amide function. The presence of the two distal
aromatic rings at the appropriate distance and correctly oriented
within the chemical space, confirm the hypothesis of the NK1
NK1 receptor pharmacology.27,35 Ex vivo autoradiography experi-
ments were performed essentially as described previously.36 3a
was tested at the doses of 0.1, 0.3 and 1 mg/kg p.o. (1 h pre-treat-
ment) and a dose-dependent increase of the time spent in active
social interaction was observed. A significant increase in seconds
(s) was recorded at 1 mg/kg (21.2 2.9 s in vehicle to 33.1 3.7 s
after treatment, P <0.05). No effect was observed on animals’ loco-
motor activity, thus excluding any sedative effect induced by the
CF3
CF3
CH3O
CH3O
CF3
O
N
O
CF3
MeO
NH
NH
O
N
N
N
N
N
F
N
HN
NH
O
8, L754030
6, CP99994
7,CP211754
5, CP96345
4, L709210
IC50=0.09 nM
IC50=0.17 nM
IC50=0.61 nM
IC50=0.66 nM
IC50=0.2 nM
N
N
N
CF3
N
CF3
CF3
N
CF3
HN
N
N
CF3
N
N
CF3
CF3
N
O
NH OMe
CF3
O
O
O
N
H
O
N
H
N
H
F
F
13, Vestipitant
11,Merck 9c
12, L733060
10, Netupitant
9, GR203040
IC50=0.25 nM
IC50=0.09 nM
IC50=1 nM
IC50=0.95 nM
IC50=0.04 nM
Figure 5. Chemical structure and in vitro activity of the set of NK1 receptor antagonists used to build a common features pharmacophore model.

R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
6269
test in marmoset and in the social interaction model in gerbils. In
the latter test a full NK1 receptor occupancy was observed to be
needed to produce
Compared to casopitant, due to its purposefully engineered more
rigid bicycle -lactam structure, compound 3a showed an improved
a significant pharmacodynamic effect.
c
preclinical pharmacokinetic profile. Based upon its favorable phar-
macokinetics and developability profile 3a (termed orvepitant) was
selected as a drug candidate and has been progressed to clinical
studies. The outcome of these studies, including in patients suffer-
ing from major depressive disorder, is reported elsewhere.44
5. Experimental
5.1. Chemistry
NMR spectra were recorded in DMSO-d6 or CDCl3 at constant
temperature of 25 °C and complete assignment were made by
means of several 1D and 2D-techniques including 19F, gCOSY,
gHSQC, gHMQC NMR-experiments when needed. 1H and gHSQC
experiments were collected on a 400 Unity Varian or 500 Inova
Varian. Chemical shifts are reported in parts per million (ppm)
downfield from CHCl3 residual line (d = 7.27 ppm) or from DMSO
residual line (d = 2.49 ppm) and were assigned as singlets (s),
doublets (d), triplets (t), quartet (q), broad quartet (bq), broad sing-
lets (bs), doublets of doublets (dd), or multiplets (m). Coupling
constants (J) are given in Hz. IR spectra were recorded on a Nicolet
Magna 760 (Thermo Fisher Scientific) spectrometer. Mass spectra
analyses were performed on a VG Platform (Waters, Manchester,
UK), mass spectrometer operating in positive electrospray ion
mode. Analytical thin layer chromatography (t.l.c.) was performed
on glass plates (Merck Kieselgel 60 F254). Visualization was
accomplished by UV light (254 nm), I2. Column chromatography
was performed on silica gel (Merck Kieselgel 70–230 mesh). Chem-
ical purity of compounds has been assessed by HPLC (>95%). All
reactions were carried out under anhydrous N2 atmosphere using
standard Schlenk techniques. Most chemicals and solvents were
analytical grade and used without further purification.
Figure 6. Superimposition of the set of NK1 receptor antagonists 4–13 selected in
literature, shown in Figure 5, and compound 3a and identification of four distinct
pharmacophoric features, namely: two hydrophobic regions (green) one aromatic
point (orange) and one hydrogen bond acceptor (cyano). The two red arrows
indicate the hydrogen bonds vectors projected from the putative hydrogen bond
donor represented by the cyano sphere. The green contoured area represents a
region of the space in which different hydrophilic (heterocycle rings) and
hydrophobic (phenyl rings) moieties are tolerated.
5.1.1. General procedure for the preparation of compounds 3a,
3b, 3c, 3d
receptor antagonists adopt a U-shaped (stacked or slightly dis-
placed) bioactive conformation. In addition, despite the significant
structural difference existing between compound 3a and aprepit-
ant, both in terms of heterocycle core and scaffold decoration,
these NK1 receptor antagonists seem to share the same pharmaco-
phoric features. In particular, it is worth noting that the hydro-
A solution of intermediate 2 (2.4 g, 5 mmoles) in 80 ml of dry
CH3CN was added to a solution of intermediate 1a or 1b (5.7 g,
40.7 mmol) dissolved in 30 ml of dry CH3CN under nitrogen atmo-
sphere at room temperature. Sodium triacetoxyborohydride
(4.36 g, 20.6 mmol) was added portionwise at room temperature
over 1 h min. The reaction mixture was stirred at room temperature
for 22 h. Then 75 ml of H2O and 25 mL of a saturated solution of
NaHCO3 were added and the solution was extracted with AcOEt
(2 ꢀ 200 ml). The organic layers were collected and dried over dry
Na2SO4, filtered and evaporated in vacuo. The crude residue was
purified by flash chromatography (silica gel, AcOEt/CH3OH 9:1) to
give title diasteroisomers 3a and 3c from intermediate 1a, and 3b
and 3d from intermediate 1b. Then, the solution of the single dias-
teroisomers 3a, 3b, 3c and 3d in dry Et2O, was treated with
1.5 equiv of HCl (1 M in Et2O). The solution was stirred for 30 min
at 0 °C, evaporated in vacuo and the solid residue was washed with
n-pentane to afford title hydrochloride derivatives as white solids.
philic moieties, namely the bicycle
c-lactam and the
triazolidinone present in the north-western and south-western re-
gion of these compounds, respectively, occupy the same binding
site region, enabling the relative carbonyl groups to act as H-bond
acceptors. In summary, this analysis shown that compound 3a
shares the same U-shaped bioactive conformation hypothesized
in the past for other structural diverse series of NK1 receptor antag-
onist; in addition this study enabled to identify an accessory bulk
tolerant site, able to accommodate a variety of both hydrophilic
and hydrophobic moieties, as a useful region to design novel
sub-series of NK1 receptor antagonists and/or optimizing their
physicochemical and pharmacokinetic properties.
5.1.2. (2R,4S)-4-[(8aS)-6-Oxo-octahydropyrrolo[1,2-a]piperazin-
2-yl]-N-[(1R)-1-[3,5-bis(trifluoromethyl)phenyl]ethyl]-2-(4-
fluoro-2-methylphenyl)-N-methylpiperidine-1-carboxamide
hydrochloride (3a)
4. Conclusion
In summary, compound 3a is a new in vitro and in vivo potent
highly brain penetrating NK1 receptor antagonist, showing long
lasting pharmacological activity in the gerbil foot tapping test (up
to 24 h), and anxiolytic-like properties both in the human threat
1H NMR (DMSO-d6): d (ppm) 10.22 (br s, 1H); 7.99 (s, 1H); 7.67
(s, 2H); 7.22 (dd, 1H); 6.94 (dd, 1H); 6.81 (t, 1H); 5.31 (q, 1H); 4.20
(dd, 1H); 4.00–3.86 (br m, 2H); 3.60–3.40 (m, 2H); 3.1–2.7 (m, 4H);

6270
R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
2.73 (s, 3H); 2.40–2.00 (m, 5H); 2.35 (s, 3H); 1.94 (m, 1H); 1.74 (q,
1H); 1.57 (d, 3H); 1.46 (d, 3H).
4 °C. Cells were then resuspended in 10 volumes of HEPES
(50 mM) buffer (pH 7.4), containing leupeptin (0.1 mM), bacitra-
MS (ES/+) m/z = 629 [M+H]+.
cin (40
lg/ml), EDTA (1 mM), Pefabloc (1 mM) and pepstatin A
HPLC: supercolis ABZ plus column (25 cm ꢀ 4.6 mm ꢀ 5
l
);
(2 M) and homogenized using a Polytron. The suspension was
l
mobile phase NH4OAc 10 mM/CH3CN from 60:40 to 10:90 in
5 min, then NH4OAc 10 mM/CH3CN 10:90 for 10 min; flux:
0.8 ml/min; k = 220 nm; retention time = 9.27 min.
centrifuged at 48,000g for 20 min at 4 °C. The final pellet was
resuspended in 10 volumes of membrane preparation buffer and
re-homogenized. Suspensions of membrane were then frozen at
ꢁ80 °C until required. Protein concentration was determined by
the Bio-Rad Protein assay using bovine serum albumin (BSA) as
the standard.
5.1.3. (2R,4S)-4-[(8aR)-6-Oxo-octahydropyrrolo[1,2-
a]piperazin-2-yl]-N-[(1R)-1-[3,5-
bis(trifluoromethyl)phenyl]ethyl]-2-(4-fluoro-2-
methylphenyl)-N-methylpiperidine-1-carboxamide
hydrochloride (3b)
5.2.2. Preparation of brain homogenates from gerbil and
marmoset
1H NMR (DMSO-d6): d (ppm) 11.05 (br s, 1H); 7.95 (s, 1H); 7.64
(s, 2H); 7.19 (dt, 1H); 6.90 (dd, 1H); 6.78 (dt, 1H); 5.28 (q, 1H); 4.16
(dd, 1H); 3.53 (m, 2H); 3.41 (m, 2H); 3.17 (t, 1H); 2.94 (m, 2H);
2.96–2.80 (m, 2H); 2.75 (t, 1H); 2.69 (s, 3H); 2.30–2.00 (m, 1H);
1.90 (m, 1H); 1.75 (q, 1H); 1.50 (m, 1H); 1.43 (d, 3H).
MS (ES/+) m/z = 629 [M+H]+.
Mongolian gerbil (60 g, Charles River) and common marmoset
(300–400 g, GSK colony, Verona, Italy) homogenates obtained from
cerebral cortex and striatum (gerbil) and cerebral striatum (mar-
moset) were prepared as follows. Fresh tissues were weighed,
crumbled and homogenized in 10 volumes of membrane-prepara-
tion buffer (HEPES 50 mM pH 7.4, leupeptin 0.1 mM, bacitracin
HPLC: supercolis ABZ plus column (25 cm ꢀ 4.6 mm ꢀ 5
l
);
40 lg/ml, EDTA 1 mM, Pefabloc 0.2 mM, pepstatin 2 lM). The
mobile phase NH4OAc 10 mM/CH3CN from 60:40 to 10:90 in
5 min, then NH4OAc 10 mM/CH3CN 10:90 for 10 min; flux:
0.8 ml/min; k = 220 nm; retention time = 8.86 min.
homogenate was then centrifuged at 48,000g for 20 min, and the
pellet was washed once more by resuspension in 10 volumes of
membrane preparation buffer and centrifugation at 48,000g for
20 min. The final pellet was resuspended in 7–10 volumes of mem-
brane preparation buffer and subdivided in aliquots frozen at
ꢁ80 °C until use. Protein concentration was determined by the
Bio-Rad Protein assay using bovine serum albumin (BSA) as the
standard.
5.1.4. (2R,4R)-4-[(8aS)-6-Oxo-octahydropyrrolo[1,2-a]piperazin-
2-yl]-N-[(1R)-1-[3,5-bis(trifluoromethyl)phenyl]ethyl]-2-(4-
fluoro-2-methylphenyl)-N-methylpiperidine-1-carboxamide
hydrochloride (3c)
1H NMR (DMSO-d6): d (ppm) 10.65 (br m, 1H); 7.99 (s, 1H); 7.76
(s, 2H); 7.37 (dd, 1H); 7.01 (dd, 1H); 6.93 (dd, 1H); 5.24 (br m, 1H);
5.04 (q, 1H); 4.00–3.95 (br m, 2H); 3.68 (m, 1H); 3.58 (m, 2H); 3.51
(m, 1H); 3.24–3.15 (m, 2H); 2.96 (m, 1H); 2.85 (m, 1H); 2.54 (m,
3H); 2.36–2.13 (m, 6H); 2.21 (s, 3H); 1.72 (m, 1H); 1.59 (m, 1H);
1.57 (d, 3H).
5.2.3. [3H]-SP binding on membranes from hNK1–CHO cells
Binding assays were carried out in 96 well, flat bottom, plates
(Sarstedt). The assay volume of 200
or increasing concentrations of test compound dissolved in DMSO
(1 pM–1 M final concentration), 100
l of [3H]-SP, (0.5 nM final
concentration), and 100 l of membrane suspension (8 g of pro-
ll consisted of 2 ll of DMSO
l
l
MS (ES/+) m/z = 629 [M+H]+.
l
l
HPLC: supercolis ABZ plus column (25 cm ꢀ 4.6 mm ꢀ 5
l
);
tein per well) in incubation buffer (containing 50 mM HEPES, pH
mobile phase: NH4OAc 10 mM/CH3CN from 60:40 to 10:90 in
5 min, then NH4OAc 10 mM/CH3CN 10:90 for 10 min; flux:
0.8 ml/min; k = 220 nm; retention time = 8.84 min.
7.4, 3 mM MnCl2, and 0.02% BSA).
In saturation experiments, increasing concentrations of [3H]-SP
(1 pM–3 nM) were incubated as above. Non specific binding was
defined by the addition of 1
out at room temperature for 40 min. Non-specific binding was de-
fined by the addition of cold SP (1 M). The reaction was stopped
lM SP. The incubation was carried
5.1.5. (2R,4R)-4-[(8aR)-6-Oxo-octahydropyrrolo[1,2-
a]piperazin-2-yl]-N-[(1R)-1-[3,5-
l
bis(trifluoromethyl)phenyl]ethyl]-2-(4-fluoro-2-
methylphenyl)-N-methylpiperidine-1-carboxamide
hydrochloride (3d)
by rapid filtration through GF/B filterplates pre-soaked in 0.5%
polyethylenimine (PEI) using a Packard cell harvester. Filters were
washed 5 times with 200 ll of ice-cold 0.9% w/v NaCl, and radioac-
tivity was counted in a microplate scintillation counter (Top Count,
Packard). In each experiment, every concentration of displacer was
tested in duplicate.
1H NMR (DMSO-d6): d (ppm) 10.90–10.60 (br m, 1H); 7.99 (s,
1H); 7.76 (s, 2H); 7.36 (dt, 1H); 7.00 (dd, 1H); 6.92 (dt, 1H); 5.25
(br t, 1H); 5.05 (q, 1H); 3.98 (m, 2H); 3.67 (m, 2H); 3.58 (m, 1H);
3.44 (m, 1H); 3.20 (m, 2H); 2.90 (m, 2H); 2.53 (s, 3H); 2.22 (s,
3H); 2.40–2.10 (m, 6H); 1.73 (m, 1H); 1.56 (m, 1H); 1.57 (d, 3H).
MS (ES/+) m/z = 629 [M+H]+.
5.2.4. [3H]-GR205171 binding on cerebral tissues from gerbil
and marmoset
HPLC: supercolis ABZ plus column (25 cm ꢀ 4.6 mm ꢀ 5
l
);
Binding assays were carried out in 96 deep well polypropylene
plates (Whatman). In displacement experiments, the assay volume
mobile phase NH4OAc 10 mM/CH3CN from 60:40 to 10:90 in
5 min, then NH4OAc 10 mM/CH3CN 10:90 for 10 min; flux:
0.8 ml/min; k = 220 nm; retention time = 9.28 min.
of 400
50 mM HEPES, pH 7.4, 3 mM MnCl2, and 0.02% BSA), 4
or increasing concentrations of test compound dissolved in DMSO
(1 pM–1 M final concentration), 100
l of [3H]-GR205171 (0.2 nM
final concentration) in incubation buffer and 200 l of membrane
suspension containing 0.055 mg (gerbil) or 0.080 mg (marmoset)
protein dissolved in incubation buffer with the addition of 2 g/
M phosphoramidon.
l
l consisted of 100
ll of incubation buffer (containing
ll of DMSO
5.2. Biology
l
l
l
5.2.1. Membrane preparation from hNK1–CHO cells
CHO cells expressing hNK1 receptors45 were cultured in Ham’s
F-12 medium supplemented with 10% (v/v) heat-inactivated foetal
bovine serum, 2 mM glutamine and 0.2 mg/ml G 418.
l
ml leupeptin, 20 lg/ml Bacitracin and 0.5 l
In saturation experiments, increasing concentrations of [3H]-
GR205171 (1 pM–3 nM) were incubated as above. Non specific
Membrane preparation from hNK1–CHO cells was performed
essentially as described by Beattie et al. (1995).46 When conflu-
ent, hNK1–CHO cells were harvested in phosphate buffered saline
(PBS) containing 5 mM EDTA and centrifuged at 913g for 8 min at
binding was defined by the addition of 1
bation proceeded at room temperature for 60 min. Non-specific
binding was defined by the addition of cold GR205171 (1 M).
lM GR205171. The incu-
l

R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
6271
The reaction was stopped by rapid filtration through GF/C filter-
mats pre-soaked in 0.5% polyethyleneimine (PEI) using a Brandel
M/96R cell harvester. Filters were washed 3 times with 1 ml ice
cold wash buffer (containing 50 mM HEPES, pH 7.4, and 3 mM
MnCl2), and radioactivity was counted in a liquid scintillation
counter (Beta Counter, Packard). In each experiment, every concen-
tration of displacer was tested in duplicate.
groups of 5 under standard laboratory conditions (23 1 °C; light
on 0600–1800 h). Animals were moved from the housing room
to the experimental room one hour prior to the start of the exper-
iment. Gerbils were briefly anaesthetised by inhalation of an isoflu-
rane/oxygen mixture (3% v/v). The skull was exposed and 5
ll of
GR73632 (3 pmol/5 l) injected directly into the lateral ventricle
l
by the insertion of a cuffed 25 gauge needle to a depth of 4 mm be-
low bregma. Immediately after treatment the animals were placed
individually into clear Perspex observation boxes containing a
layer of sawdust. Following recovery, the duration of repetitive
hind foot tapping was recorded for 3 min, giving a maximum pos-
sible duration of 180 sec.
5.2.5. Measurement of [Ca2+]i using FLIPR in hNK1–CHO cells
hNK1–CHO cells were seeded into black walled clear-bottom 96-
well plates (Costar, UK) at a density of 60,000 cells per well and
cultured overnight. The cells were then incubated for the labelling
in the culture medium containing the fluorescent calcium indicator
Fluo-4 AM (2
lM), the organic anion transport blocker probenecid
5.2.9. Human threat test (HTT) in marmoset
(5 mM), and HEPES (20 mM) for 30 min in a humidified atmosphere
of 5% CO2. After washing with Hanks’ Balanced Salts Solution (HBSS)
containing 20 mM HEPES and 2.5 mM probenecid (wash buffer), the
cells were incubated for 60 min at 37 °C in wash buffer containing
0.02%BSA (assay buffer) either in the absence (control) or in the
presence of test compound (0.3–10nM). The plates were then
placed into a FLIPR (Molecular Devices, Sunnyvale, CA) to monitor
cell fluorescence (kex = 488 nm, kem = 510–570 nm) before and after
the addition of different concentrations of SP (2 pM–300 nM) in
assay buffer. FLIPR experiments were carried out by using a laser
setting of 1.0 W and a 0.4 sec CCD (charge coupled device) camera
shutter speed. In each experiment, the test compound was tested
in duplicate versus every concentration of SP.
The studies utilised laboratory-bred male (vasectomised) and
female common marmosets over 2 years of age, weighing 300–
500 g. The animals were caged in couples, in a housing room main-
tained at 25 1 °C, 60% humidity and a 12 h light/dark cycle (lights
on at 0600, with 30 min simulated dawn and twilight). Both ani-
mals in each pair were involved in the test, which was carried
out with the animals situated in the home cage. Because there
are great differences in behavior between subjects, the ‘responder’
animals included in an anxiety study were preselected and had to
meet the baseline criteria of at least 10 postures exhibited in the
2 min of test period.
The postures recorded in the test were those described by Cost-
all et al. (1988).32
5.2.6. Drugs and materials for in vitro studies
Each test compound was freshly prepared in each experiment as
100 lM solution in DMSO diluted from a 10 mM stock solution in
– Genital presenting (‘Tail Posture’): the animal’s back is turned to
the observer with elevation of the tail to expose the genital
region;
DMSO. All further dilutions were performed in either DMSO in the
binding experiments or in the assay buffer in the functional exper-
iments. [3H]-SP (1–2 TBq/mmol) was purchased from Amersham
Life Science. [3H]-GR205171, di-trifluoroacetic acid salt, synthes-
ised in GSK Chemical Department and labelled with [3H] by Amer-
sham, has been provided as 3.07 TBq/mmol, in ethanol solution. SP
was obtained from Bachem. Fluo-4 was purchased from Molecular
Probes. G-418 was purchased from Calbiochem. Probenecid was
purchased from Sigma–Aldrich. Leupeptin, bacitracin, Pefabloc,
pepstatin A, phosphoramidon were purchased from Sigma–Aldrich.
FLIPR was purchased from Molecular Devices (Sunnyvale, CA).
– Scent-marking: the animal scent-marks the cage surfaces using
circum-anal and circum-genital scent glands;
– Slit-stare: the animal stares at the observer with flattened ear
tufts and eyes reduced to ‘slits’
– Arch-piloerection: the animal moves around the cage with
arched back and full-body piloerection, failing to make eye con-
tact with the observer.24
The number of jumps from the back of the cage to the cage front
provided an index of locomotor activity.
5.2.10. GFT experimental design
5.2.7. Data analysis for receptor binding and functional studies
Data from competition binding experiments on membranes
from hNK1–CHO cells were analyzed by using the non-linear
curve-fitting program MicroSAT 5.2.1. For [3H]-SP displacement
binding experiments, IC50 values were obtained constraining the
slope factor to unity. IC50 values were converted to Ki using the
Cheng–Prusoff equation.47 Data from competition binding experi-
ments on brain homogenates were analyzed using the non-linear
curve-fitting program LIGAND.48 Previous determination of KD of
radioligands was assessed by elaborating saturation experiments
with LIGAND. In FLIPR experiments on hNK1–CHO cells, functional
responses were measured as fluorescence intensity (FI) produced
after receptor stimulation minus basal FI. Curve fitting was deter-
mined by non-linear regression analysis using GraphPad prism
Version 3.02. In functional experiments, the apparent pKB value
for 3a and 3c were calculated by using the Gaddum equation:
pKB = log(concd ratioꢁ1) ꢁ log(antag concd), where concd ratio
represents the ratio between the EC50 values of the agonist in the
presence and in absence of the antagonist.
1, 4, 8 or 24 h before the GR73632 administration, at least 7 ani-
mals received each treatment of vehicle or test compound (10 ml/
kg). All treatments were randomly distributed between the five
animals of each cage. On completion of experiments animals were
killed humanely.
5.2.11. HTT in marmoset experimental design
Drug treatments were assigned according to a blind random
crossover design. One hour before the test, four to seven animals
were dosed orally with vehicle or test compounds (1 ml/kg). After
a washout period of at least three days, treatments were reas-
signed and the study was complete when all animals had received
all treatments.
5.2.12. Drugs and materials for GFT and HTT test
GR73632 was dissolved in phosphate buffer solution (0.01 M)
and frozen at ꢁ20 °C in 0.5 ml aliquots at the required concentra-
tion until use. Compound 3a was prepared as a solution in metho-
cel (0.5% in distilled water) and DMSO (5% in gerbil and 1% in
marmoset). All further dilutions were performed maintaining the
same ratio between water and DMSO. This vehicle was used for
the treatment of the control group of animals.
5.2.8. Gerbil foot tapping (GFT)
The subjects were male Mongolian Gerbils (50–70 g) obtained
from the Charles River colony (Italy). Animals were housed in

6272
R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
5.2.13. Data analysis for GFT and HTT test
Diazepam (M/890001-339, F.I.S. S.p.A.) was used as standard
anxiolytic drugs and was prepared as a solution in distilled water.
[3H]-GR205171 (code TRQ10127, Amersham Biosciences). Specific
activity: 83 Ci/mmol. Data Analysis for SI test and receptor
autoradiography.
Behavioral data were subjected to one-way analysis of variance
(ANOVA) followed by Dunnett’s t-test, or a paired t-test comparing
each compound dose with the related vehicle treatment, using GB-
stat software (Version 6.5, Dynamic Microsystems Inc.).
The doses of 3a inhibiting foot tapping by 50% (ID50) were
determined by a linear regression analysis using RS-1 software
( C.I. 95%). Data of number of postures or jumps were subjected
to a paired t-test, comparing each compound dose with the related
vehicle treatment, using GB-stat software. Data are expressed as
mean standard deviation.
5.2.14. Social interaction (SI) test in gerbils
The subjects were male Mongolian Gerbils (50–70 g) obtained
from the Charles River colony (Germany). Animals were housed
in groups of 5 under standard laboratory conditions (23 1 °C;
lights on 0600–1800 h). Food and water were available ad libitum.
Testing took place in a square Perspex box (40 ꢀ 40 cm and 30 cm
high) the floor of which was divided into squares (16 ꢀ 16 cm) to
allow assessment of line crossings as an index of locomotor activ-
ity. Experiments were performed during the light phase between
the hours of 0900 and 1300 and behavior was recorded remotely
by videocamera. The day before testing, each gerbil was placed
singly into the test arena and given a 10 min familiarization trial.
On the day of the test, pairs of gerbils, matched for the treatment
condition, but unfamiliar to each other, were placed in the centre
of the test arena and their social interaction (sniffing, following,
grooming the partner, wrestling) was recorded for 5 min. After
each trial the floor was wiped with alcohol solution (20%) and al-
lowed to dry prior to further testing. An observer who was blind
to the drug treatment administered to the animals, analysed the
tape recordings to assess the time spent in active social interaction
and locomotor activity (number of line crossings).26 Each gerbil
was allocated to an unfamiliar partner that came from a different
cage. Pairs of gerbils received vehicle or test compound (i.p. 2 ml/
kg or p.o. 10 ml/kg) according to a randomized experimental de-
sign. Each animal was treated with compounds or vehicle one hour
prior to test. No animal was tested on more than one occasion. On
completion of experiments animals were humanely killed by cervi-
cal dislocation, and brains were immediately removed and frozen
in isopentane at ꢁ30 °C.
Ex vivo [3H]-GR205171 binding was expressed as the percent-
age of optical density versus saline treated animals. Non-specific
binding on adjacent sections, determined in the presence of
100 nM GR205171, was subtracted from each total binding section
to obtain the specific binding. Therefore, percentages of receptor
occupancy correspond to 100% minus the percentage ratio of spe-
cific binding in the drug treated over the vehicle treated animals.
6. Molecular modeling studies
Molecular modeling studies were performed using the Molecu-
lar Operating Environment (MOE).53 The set of NK1 receptor antag-
onists selected in literature antagonists standards and compound
3a were initially sketched in 2D and then fully minimized using
the MMFF94x force field. After visual inspection the molecules
were then submitted to the conformational search module, using
the LowModeMD method and defaults parameters. A conforma-
tional database of 513 conformations was then used as input for
the pharmacophoric elucidate module. Also in this case defaults
parameters where applied. The resulting best hypothesis based
on the maximum common conformation overlap index (6.98 in
our case) was then used to search a molecule’s 3a, 3b conforma-
tional database prepared using the same protocol described for
the NK1 receptor antagonists reference molecules.
Acknowledgments
This work was sponsored by G.S.K. Romano Di Fabio, Giuseppe
Alvaro, Simone Braggio, Renzo Carletti, Philip A. Gerrard, Cristiana
Griffante, Carla Marchioro, Alfonso Pozzan, Sergio Melotto, Ales-
sandro Poffe, Laura Piccoli, Emiliangelo Ratti, Elvira Tranquillini,
Michael Trower, Simone Spada, Mauro Corsi were employees of
GSK (Neurosciences Centre of Excellence for Drug Discovery,
Medicines Research Centre, Via A. Fleming 4, 37135 Verona, Italy)
at the time this work was performed. Emiliangelo Ratti and
Michael Trower are currently employees and shareholders in
NeRRe Therapeutics Ltd, which owns the intellectual property
and development rights to 3a (also called orvepitant).
5.2.15. Ex vivo receptor autoradiography
Coronal sections (12 lm thickness) of gerbil brains (3 animals/
group) were used for ex vivo autoradiography experiments with
the radiolabeled NK1 receptor antagonist [3H]-GR205171. To min-
imise the dissociation of the drug/receptor complex formed in vivo,
brain sections were not washed prior to incubation with the radi-
oligand. Total binding (0.3 nM [3H]-GR205171 in Tris–HCl pH
7.4 + 3 mM MnCl2, 0.02% Bovine Serum Albumin + protease inhibi-
tors) was determined at a restricted incubation time of 10 min to
minimise dissociation. The concentration of [3H]-GR205171 used
in this study approximates its affinity constant for the NK1 receptor
and in previous autoradiography experiments it was demonstrated
to show an optimal signal/noise ratio. Non-specific binding on
adjacent sections was determined in the presence of 100 nM
GR205171. Following 4 ꢀ 1 min washes in the same buffer, sec-
tions were exposed to Fuji imaging plates (Fuji Photo Film, Tokyo)
for 10 days. The film autoradiograms were scanned using a BAS-
5000 Bio-imaging Analyzer (Fuji Photo Film, Tokyo) and quantified
using an image analysis software system (AIS 4.0, Imaging
Research Inc., Canada). Ex vivo binding was measured in the stria-
tum, a brain region with high density of NK1 binding sites, both in
preclinical species49–51 and humans, often considered a reference
brain area in human PET studies.52
References and notes
1. Harrison, S.; Geppetti, P. Int. J. Biochem. Cell B 2001, 33, 555.
2. Zhongli, G.; Peet, N. P. Curr. Med. Chem. 1999, 6, 375.
3. Satake, H.; Kawada, T. Curr. Drug Targets 2006, 7, 963.
4. Chahl, L. A. Curr. Drug Targets 2006, 7, 993.
5. Ebner, K.; Singewald, N. Amino Acids 2006, 31, 251.
6. McLean, S. Curr. Pharm. Des. 2005, 11, 1529.
7. Nemeroff, C. B. Psychopharmacol. Bull. 2002, 36, 6.
8. Mantyh, P. W. J. Clin. Psychiatry 2002, 63, 6.
9. Wong, M. L.; Licinio, J. Nat. Rev. Drug Disc. 2004, 3, 136.
10. Holtzheimer, P. E.; Nemeroff, C. B. Novel Curr. Pshychiatry Rep. 2008, 10, 465.
11. Patacchini, R.; Maggi, C. A. Eur. J. Pharmacol. 2001, 429, 13.
12. Rupniak, N. M.; Kramer, M. S. TiPS 1999, 20, 485.
13. Muñoz, M.; Coveñas, R. Curr. Med. Chem. 2011, 18, 1820.
14. 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.
15. Czéh, B.; Fuchs, E.; Simon, M. Expert Opin. Investig. Drugs 2006, 15, 479.
5.2.16. Drugs and materials for SI and autoradiography study
Compound 3a was suspended in methocel (0.5% w/v in distilled
water) + DMSO (5%).

R. Di Fabio et al. / Bioorg. Med. Chem. 21 (2013) 6264–6273
6273
16. Kramer, M. S.; Winokur, A.; Kelsey, J.; Preskorn, S. H.; Rothschild, A. J.; Snavely,
D.; Ghosh, K.; Ball, W. A.; Reines, S. A.; Munjack, D.; Apter, J. T.; Cunningham, L.;
Kling, M.; Bari, M.; Getson, A.; Lee, Y. Neuropsychopharmacology 2004, 29, 385.
17. Chappell, P.B. 11th Congress of the Association of, European Psychiatry, 2002.
18. Keller, M.; Montgomery, S.; Ball, W.; Morrison, M.; Snavely, D.; Liu, G.;
Hargreaves, R.; Hietala, J.; Lines, C.; Beebe, K.; Reines, S. Biol. Psychia. 2006, 59,
216.
35. Beresford, I. J.; Birch, P. J.; Hagan, R. M.; Ireland, S. J. Br. J. Pharmacol. 1991, 104,
292.
36. Langlois, X.; Riele, P. T.; Wintmolders, C.; Leysen, J. E.; Jurzak, M. J. Pharmacol.
Exp. Ther. 2001, 299, 712.
37. Seabrook, G. R.; Shepheard, S. L.; Williamson, D. J.; Tyrer, P.; Rigby, M.; Cascieri,
M. A.; Harrison, T.; Hargreaves, R. J.; Hill, R. G. Eur. J. Pharm. 1996, 317,
129.
19. Ratti, E.; Bellew, K.; Bettica, P.; Bryson, H.; Zamuner, S.; Archer, G.; Squassante,
L.; Bye, A.; Trist, D. G.; Krishnan, K. R.; Fernandes, S. J. Clin. Psychopharmacol.
2011, 31, 727.
20. Di Fabio, R.; Alvaro, G.; Griffante, C.; Pizzi, D. A.; Donati, D.; Mattioli, M.;
Cimarosti, D.; Guercio, G.; Marchioro, C.; Provera, S.; Zonzini, L.; Montanari, D.;
Melotto, S.; Gerrard, P. A.; Trist, D. G.; Ratti, E.; Corsi, M. J. Med. Chem. 2011, 54,
1071.
21. Manetti, D.; Gherlandini, C.; Bartolini, A.; Bellucci, C.; Dei, S.; Galeotti, N.;
Gualtieri, F.; Romanelli, M. N.; Scapecchi, S.; Teodori, E. J. Med. Chem. 2000, 43,
1969.
38. Beattie, D. T.; Beresford, J. M.; Connor, H. E.; Marshall, F. H.; Hawcock, A. B.;
Hagan, R. M.; Bowers, J.; Birch, P. J.; Ward, T. P. Br. J. Pharmacol. 1995, 116, 3149.
39. Hoffmann, T.; Bös, M.; Stadler, H.; Schnider, P.; Hunkeler, W.; Godel, T.; Galley,
G.; Ballard, T. M.; Higgins, G. A.; Poli, S. M.; Sleight, A. J. Bioorg. Med. Chem. Lett.
2006, 16, 1362.
40. Elliott, J. M.; Carlson, E. J.; Chicchi, G. G.; Dirat, O.; Dominguez, M.; Gerhard, U.;
Jelley, R.; Jones, A. B.; Kurtz, M. M.; Lan Tsao, K.; Wheeldon, A. Bioorg. Med.
Chem. Lett. 2006, 16, 2929.
41. Jacoby, E.; Boudon, A.; Kucharczyk, N.; Michel, A.; Fauchere, J. L. J. Recept. Signal
Transduction Res. 1997, 17, 855.
22. Piwinski, J. J.; Wong, J. K.; Green, M. J.; Kaminski, J. J.; Colizzo, F.; Albanese, M.
M.; Ganguly, A. K.; Billah, M. M.; Anthes, J. C.; West, R. E., Jr. Bioorg. Med. Chem.
Lett. 1998, 8, 3469.
23. The absolute stereochemistry of intermediates 1a and 1b was assigned by X-
ray studies; see Alvaro, G.; Di Fabio, R.; Tranquillini, E.; Spada, S. Piperidine
Derivatives, WO2003066635.
42. Takeuchi, Y.; Shands, E. F.; Beusen, D. D.; Marshall, G. R. J. Med. Chem. 1998, 41,
3609.
43. Caliendo, G.; Grieco, P.; Perissutti, E.; Santagada, V.; Saviano, G.; Tancredi, T.;
Temussi, P. A. J. Med. Chem. 1997, 40, 594.
44. Ratti, E.; Bettica, P.; Alexander, R.; Archer, G.; Carpenter, D.; Evoniuk, G.;
Gomeni, R.; Lawson, E.; Lopez, M.; Millns, H.; Rabiner, E. A.; Trist, D.; Trower,
24. Pellegatti, M.; Bordini, E.; Fizzotti, P.; Roberts, A.; Johnson, B. M. Drug Met.
Dispos. 2009, 37, 1635.
M.; Zamuner, S.; Krishnan, R.; Fava, M.
424.
J Psychopharmacol. 2013, 27,
25. Ferrari, L.; Reami, D.; Michi, M. J. Chromatogr. B 2010, 878, 2974.
26. Miraglia, L.; Pasciarusco, S.; Bordini, E.; Martinucci, S.; Pellegatti, M. Drug Met.
Dispos. 2010, 38, 1876.
27. Griffante, C.; Carletti, R.; Andretta, F.; Corsi, M. Br. J. Pharmacol. 2006, 148, 39.
28. Di Fabio, R.; Alvaro, G.; Griffante, C.; Pentassuglia, G.; Pizzi, D. A.; Donati, D.;
Rossi, T.; Guercio, G.; Mattioli, M.; Cimarosti, Z.; Marchioro, C.; Provera, S.;
Zonzini, L.; Montanari, D.; Melotto, S.; Gerrard, P. A.; Trist, D. G.; Ratti, E.; Corsi,
M. J. Med. Chem. 2009, 52, 3238.
45. Lundstrom, K.; Hawcock, A. B.; Vargas, A.; Ward, P.; Thomas, P. Z.; Naylor, A.
Eur. J. Pharmacol. 1997, 337, 73.
46. Beattie, D. T.; Beresford, I. J.; Connor, H. E.; Marshall, F. H.; Hawcock, A. B.;
Hagan, R. M.; Bowers, J.; Birch, P. J.; Ward, P. Br. J. Pharmacol. 1995, 116, 3149.
47. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
48. Munson, P. J.; Rodbard, D. Anal. Biochem. 1980, 107, 220.
49. Saffroy, M.; Beaujouan, J. C.; Torrens, Y.; Besseyre, J.; Bergstrom, L.; Glowinski, J.
Peptides 1988, 9, 227.
29. Ballard, T. M.; Breitmaier, E.; Grottick, A. J.; Koch, S.; Higgins, G. A. Behav.
Pharmacol. 1999, 10, S5.
30. Rupniak, N. M. J.; Williams, A. R. Eur. J. Pharmacol. 1994, 265, 179.
31. Steckler, T. Curr. Top. Behav. Neurosci. 2010, 2, 415.
32. Costall, B.; Domeney, A. M.; Gerrard, P. A.; Naylor, R. J. Br. J. Pharmacol. 1988, 95,
475P.
50. Dam, T. V.; Quirion, R. Peptides 1986, 7, 855.
51. Caberlotto, L.; Hurd, Y. L.; Murdock, P.; Wahlin, J. P.; Melotto, S.; Corsi, M.;
Carletti, R. Eur. J. Neurosci. 2003, 17, 1736.
52. Hargreaves, R. J. Clin. Psychiatry 2002, 63, 18.
53. Molecular Operating Environment (MOE), 2012.10; Chemical Computing
Group Inc., 1010 Sheerbrooke St. West, Suite #910, Montral, QC, Canada, H3A
2R7, 2012.
33. Lipp, H. P. Brain Behav. Biol. 1978, 15, 241.
34. Cheeta, S.; Tucci, S.; Sandhu, J.; Williams, A. R.; Rupniak, N. M. J.; File, S. E. Brain
Res. 2001, 915, 170.