Novel pyrazolo-tetrahydropyridines as potent orexin receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.01.070

A novel series of dual orexin receptor antagonists was prepared by heteroaromatic five-membered ring system replacement of the dimethoxyphenyl moiety contained in the tetrahydroisoquinoline core skeleton of almorexant. Thus, replacement of the dimethoxyph

Full text of DOI:10.1016/j.bmcl.2010.01.070

Bioorganic & Medicinal Chemistry Letters 20 (2010) 1539–1542
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel pyrazolo-tetrahydropyridines as potent orexin receptor antagonists
*
*
Thierry Sifferlen , Christoph Boss , Emmanuelle Cottreel, Ralf Koberstein, Markus Gude, Hamed Aissaoui,
Thomas Weller, John Gatfield, Catherine Brisbare-Roch, Francois Jenck
Actelion Pharmaceuticals Ltd, Drug Discovery and Preclinical Research & Development, Gewerbestrasse 16, CH-4123 Allschwil, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of dual orexin receptor antagonists was prepared by heteroaromatic five-membered ring
system replacement of the dimethoxyphenyl moiety contained in the tetrahydroisoquinoline core skel-
eton of almorexant. Thus, replacement of the dimethoxyphenyl by a substituted pyrazole and additional
optimization of the substitution pattern of the phenethyl motif allowed the identification of potent
antagonists with low nanomolar affinity for hOX1R and hOX2R. The synthesis and structure–activity rela-
tionship of these novel antagonists will be discussed in this communication. These investigations fur-
nished several suitable candidates for further evaluation in in vivo studies in rats.
Received 17 December 2009
Accepted 14 January 2010
Available online 22 January 2010
Keywords:
Orexin receptors
Neuropeptides
G protein-coupled receptors
Sleep
Ó 2010 Elsevier Ltd. All rights reserved.
In 1998, two independent research groups discovered two neu-
ropeptides, orexin A and orexin B (or hypocretin-1 and hypocretin-
2), which are produced exclusively by neurons in the lateral hypo-
thalamus.1,2 Two orphan G protein-coupled receptors (GPCR) for
these endogenous ligands were as well identified and are known
as orexin 1 (OX1R) and orexin 2 (OX2R) receptors. Orexin A shows
potent agonistic activity toward both subtypes of orexin recep-
tors,2 whereas, orexin B exhibited greater affinity for OX2R as com-
pared to the OX1R.2 Several studies highlighted the potential role of
the orexin neuropeptides in the regulation of a variety of biological
functions including feeding,2 and the sleep/wake cycle.3,4 Several
groups have developed orexin antagonists to explore the physio-
logical role of the orexin receptors.5 We have recently reported
that a dual orexin receptor antagonist, almorexant (Fig. 1) enabled
somnolence without cataplexy in rats, dogs, and humans.6 In the
course of our efforts toward the identification of small molecule
orexin antagonists, we were interested in the heterocyclic replace-
ment of the dimethoxyphenyl moiety contained in the tetrahydro-
isoquinoline core skeleton.7 Here, we describe the efforts in
optimizing a novel series of nonpeptidic, dual orexin antagonists
structurally related to almorexant where the dimethoxyphenyl
part was replaced by a substituted pyrazole (Fig. 1).
ylamine regioselectively led to the formation of the pyrazolo-deriv-
ative 4.9 Hydrolysis under acidic conditions (p-TsOH in methanol)
resulted in the formation of methyl ester 5 which was quantita-
tively converted to the corresponding primary amide 6 by treat-
ment with a solution of ammonia in methanol. Dehydration of
amide 6 (trifluoroacetic anhydride, pyridine) smoothly delivered
the expected nitrile 7.10 Reduction of 7 (LiAlH4, THF) followed by
protection of the resulting primary amine with a benzyloxycar-
bonyl moiety gave after purification the protected compound 8. A
final deprotection of the amino function by hydrogenolysis using
palladium over charcoal delivered the expected primary amine 9.
For the preparation of the homologous primary amine 13, devi-
ation from the first synthetic route began with methyl ester 5
(Scheme 2). Reduction of methyl ester 5 (LiAlH4) quantitatively
led to the corresponding primary alcohol 10. Activation of the alco-
hol 10 as the mesylate and subsequent nucleophilic displacement
mediated by tetrabutylammonium cyanide gave the expected
nitrile 11. Borane reduction of 11 followed by protection of the
resulting primary amine with a benzyloxycarbonyl moiety afforded
O
O
N
The preparation of the planned pyrazolo-tetrahydropyridine
derivatives began with the synthesis of the trisubstituted pyrazole
9 (Scheme 1). Acylation of Meldrum’s acid 1 with diketene 2
quantitatively afforded intermediate 3.8 Subsequent reaction of a
methanolic solution of 3 with ethylhydrazine in presence of trieth-
N
O
N
H
O
N
N
N
H
Y
X
Ri
CF3
Almorexant
* Corresponding authors. Tel.: +41 615656561; fax: +41 615656500.
E-mail addresses: thierry.sifferlen@actelion.com (T. Sifferlen), christoph.bos-
s@actelion.com (C. Boss).
(ACT-078573)
Figure 1. Almorexant and the corresponding pyrazolo-tetrahydropyridines.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.070

1540
T. Sifferlen et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1539–1542
O
O OH O
N
a
h
N
O
O
O
c
b
O
N
N
O
O
O
+
O
O
O
O
O
O
O
5
1
2
3
4
d
f, g
e
N
N
N
N
N
N
N
O
N
CN
NH2
NHCO2Bn
NH2
9
8
7
6
Scheme 1. Reagents and conditions: (a) Et3N, CH2Cl2, room temperature (100%); (b) EtNHNH2, Et3N, MeOH, 60 °C; (c) p-TsOH, MeOH, 60 °C, (70% for two steps); (d) 7 M NH3
in MeOH, room temperature (100%); (e) trifluoroacetic anhydride, pyridine, dioxane, room temperature (88%); (f) LiAlH4, THF, 0 °C; (g) ClCO2Bn, NaHCO3, THF, H2O, room
temperature (29% for two steps); (h) H2, 10% Pd/C, MeOH, room temperature (94%).
f
a
b, c
OH
d, e
N
N
N
N
N
N
N
N
N
O
N
NHCO2Bn
NH2
O
CN
13
12
5
10
11
Scheme 2. Reagents and conditions: (a) LiAlH4, THF, 0 °C (100%); (b) MeSO2Cl, Et3N, THF, 0 °C; (c) n-Bu4NCN, THF, 75 °C (86% for two steps); (d) 1 N BH3ÁTHF, THF, 75 °C; (e)
ClCO2Bn, K2CO3, THF, H2O, room temperature (74% for two steps); (f) H2, 10% Pd/C, MeOH, room temperature (100%).
12. Finally, deprotection of the amino function by hydrogenolysis
using palladium over charcoal delivered the expected primary
amine 13.
A cell-based FLIPR assay (fluorometric imaging plate reader)
measuring Ca2+ flux as a functional determinant of orexin binding
allowed to evaluate the antagonistic activity of the synthesized pyr-
azolo-tetrahydropyridines with both orexin receptors.11 Prelimin-
ary structure–activity relationship (SAR) studies explored the
potency against both receptors of derivatives containing a mono-
substituted phenyl ring. A variety of substituents representing dif-
ferent electronic properties were investigated. Moving an electron-
donating methyl substituent from the ortho to the meta-position al-
lowed a sixfold increase in hOX1R potency as illustrated with the
two mixtures of diastereoisomers 20 and 21 (Table 1). A para-
methyl group allowed an additional gain of potency toward both
orexin receptors (diastereoisomers 22). Compared to methyl, a
para-methoxy moiety was detrimental and resulted in a complete
loss of potency toward hOX1R (23). Previous investigations in the
tetrahydroisoquinoline series (X–Y = CH2CH2) indicated that the
(S;R)-stereoisomer was the most active one (Fig. 1). The influence
of the stereochemistry was investigated with pyrazolo-tetrahydro-
pyridine derivatives containing an electron-donating para-ethyl or
para-isopropyl substituent. Evaluation of the antagonistic activity
of the separated diastereoisomers highlighted that the (R;R) iso-
mers were significantly less potent against both receptors (stereo-
isomers 24–27). Derivatives with a para-ethyl or a para-methyl
group are almost equipotent (22 and 24). A larger para-isopropyl
residue induced a threefold decrease in hOX1R potency (26). Similar
effects on potency could be observed with electron-withdrawing
substituents. Thus, moving a fluorine atom from the ortho to the
meta-position allowed an enhancement in potency toward both
orexin receptors (diastereoisomeric mixtures 28 and 29). A larger
meta-trifluoromethyl moiety afforded an almost equipotent com-
pound (30), and a para-CF3 was even more beneficial (31). Substitu-
tion of the para-position with a chlorine atom resulted in decreased
potency toward hOX1R as shown with diastereoisomers 32. The
importance of the linker between the pyrazolo-tetrahydropyridine
and the phenyl ring was as well investigated. Replacement of the
CH2CH2 linker with the CH2O linker gave a complete loss of affinity
for hOX1R as illustrated with the selective orexin 2 receptor antag-
onist 33, a direct analogue of 22. Shortening of the C2-linker as in 31
resulted in reduced potency as exemplified with 34.
The target pyrazolo-tetrahydropyridine derivatives could be
prepared according to two convenient synthetic routes (Scheme 3).
Primary amine 9 was allowed to react with diversely substituted
aldehydes 14 in a microwave-assisted Pictet–Spengler reaction
(AcOH, EtOH, microwave) affording with good yields a racemic
mixture of pyrazolo-tetrahydropyridine derivatives 15. Finally, N-
alkylation with the chiral (S)-tosylate 16 yielded, after purification,
the target pyrazolo-tetrahydropyridine derivatives 17. According
to an alternative synthetic route, primary amines 9 and 13 were
acylated with diversely substituted carboxylic acid derivatives 18
(PyBOP, DIPEA, DMF) resulting in the amides 19 which were trans-
formed into the pyrazolo-tetrahydropyridine derivatives 15 via
Bischler–Napieralski reaction (POCl3, MeCN) and subsequent imine
reduction (NaBH4, MeOH). Again, a final alkylation of the resulting
racemic mixture of secondary amines 15 with the chiral (S)-tosyl-
ate 16 gave after purification the target pyrazolo-tetrahydropyri-
dine derivatives 17.
14
X
N
NH2
n
N
n
Y CHO
N
N
Ri
NH
9, n = 1
Y
a
X
15
13,
n = 2
Ri
O
CO2H
Y
d, e
TsO
N
H
X
c
b
Ri
16
18
N
O
N
n
N
N
H
H
Ri
Y
N
n
Y
N
X
X
N
O
17
19
Ri
After this first exploration of derivatives containing a mono-
substituted phenyl ring, it was apparent that a para-methyl or a
para-trifluoromethyl moiety afforded the most potent compounds.
Scheme 3. Reagents and conditions: (a) AcOH, EtOH, microwave (100 W; 130 °C;
14 bar; 6 min); (b) DIPEA, 3-methyl-2-butanone, 90 °C; (c) PyBOP, DIPEA, DMF,
room temperature; (d) POCl3, MeCN, 90 °C; (e) NaBH4, MeOH, 0 °C.

T. Sifferlen et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1539–1542
1541
Table 1
SAR studies of the phenyl-X–Y moiety
N
N
O
O
N
N
(R)
(R)
N
N
*
*
N
H
N
H
Y
Ph
Y
Ph
X
X
R2
R3
R2
R3
R4
R4
X-Y = CH2CH2 or CH2: (S;R)-stereoisomer
X-Y = OCH2: (R;R)-stereoisomer
X-Y = CH2CH2 or CH2: (R;R)-stereoisomer
X-Y = OCH2: (S;R)-stereoisomer
Compound
X–Y
Stereochemistry
R2
R3
R4
hOX1Ra
hOX2Ra
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
OCH2
(S;R) + (R;R)
(S;R) + (R;R)
(S;R) + (R;R)
(S;R) + (R;R)
(S;R)
(R;R)
(S;R)
(R;R)
(S;R) + (R;R)
(S;R) + (R;R)
(S;R) + (R;R)
(S;R)
(S;R) + (R;R)
(R;R)
Me
H
H
H
H
H
H
H
F
H
H
H
H
H
H
H
Me
H
H
H
H
H
H
H
F
CF3
H
H
H
H
H
Me
OMe
Et
1031
180
108
10,000
63
1424
161
1919
669
402
358
128
21
19
6
219
3
174
9
179
22
10
7
Et
i-Pr
i-Pr
H
H
H
CF3
Cl
Me
CF3
3
993
10,000
769
12
19
25
CH2
(S;R) + (R;R)
H
a
IC50 values in nM (FLIPR assay).
In an attempt to further improve potency in this series of orexin
receptor antagonists, additional substituents were added on the
phenyl moiety (Table 2).12 Having a para-methyl substituent, the
addition of an ortho-fluorine atom was unfavorable mainly for
hOX1R potency (35) whereas a meta-fluorine was well tolerated
as illustrated with compound 36. Difluorination of the para-tolyl
motif was also nicely tolerated as shown with derivative 37, and
an almost threefold increase in hOX1R potency could be even
achieved with the isomeric and potent dual orexin receptor antag-
onist 38. The addition of a second methyl substituent in the para-
tolyl moiety was then investigated and like with fluorine, an addi-
tional ortho-methyl group was unfavorable mainly for hOX1R po-
tency (39). The addition of a meta-methyl residue was beneficial
due to an additive effect on orexin activity which allowed for a
substantial eightfold increase of the affinity for the hOX1R resulting
in compound 40, the most potent dual orexin receptor antagonist
identified so far in this series. In a next step, the addition of fluorine
substituents was explored with the derivative containing a para-
trifluoromethyl group on the phenyl ring. An additional ortho-fluo-
rine atom induced a twofold increase in hOX1R potency (41). Diflu-
orination of the para-trifluoromethylphenyl allowed for
a
significant gain in potency as shown with 42, and the affinity for
the hOX1R could be even increased with the isomeric derivative
43 which is one of the most potent dual orexin receptor antagonist
identified in this series. Finally, investigations of seven-membered
analogues (44 and 45) showed that the corresponding six-mem-
bered derivatives (40 and 43) were more potent toward both orex-
in receptors. In vivo effects of compounds 40 and 43 on sleep and
wake cycles were evaluated in adult male wistar rats implanted
with radiotelemetric probes.13 Compounds were orally adminis-
trated at 100 mg/kg at the beginning of the dark active phase
and electroencephalographic (EEG) and electromyographic (EMG)
signals, home cage activity and body temperature recorded. Both
compounds showed the same pattern of activity on sleep and wake
cycles as almorexant (Fig. 1)6 but to a lesser extent. Over the 12 h
dark period following administration, the compounds decreased
the home cage activity (p <0.05 for compound 40 and p <0.001
for compound 43, paired t-test, Fig. 2A) and the time spent in active
wake (p <0.01 for compound 40 and p <0.05 for compound 43,
paired t-test, Fig. 2B). Time spent in non-REM (rapid eye move-
ment) sleep was increased significantly (p <0.05 for both com-
pound, paired t-test). We also observed a non-significant increase
of the time spent in REM sleep (15% and 8% increase for com-
pounds 40 and 43, respectively). Like almorexant, the two dual
orexin receptor antagonists showed a very particular profile, that
is, they increased both non-REM and REM sleep in physiological
proportions.
Table 2
SAR studies of derivatives containing a poly-substituted phenyl ring
n
N
O
N
(S) (R)
N
N
H
Ph
R2
R6
R5
R3
R4
Compound
n
R2
R3
R4
R5
R6
hOX1Ra
hOX2Ra
35
36
37
38
39
40
41
42
43
44
45
1
1
1
1
1
1
1
1
1
2
2
F
H
F
H
Me
H
F
H
F
F
F
H
Me
H
F
F
Me
F
Me
Me
Me
Me
Me
Me
CF3
CF3
CF3
Me
CF3
H
H
H
F
H
H
H
H
F
H
H
H
H
H
H
H
H
H
H
H
147
49
57
16
131
7
52
23
5
8
2
5
5
7
3
8
8
4
5
12
F
H
H
H
H
F
36
22
a
IC50 values in nM (FLIPR assay).

1542
T. Sifferlen et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1539–1542
Home cage activity
Active Wake
B
A
C
50
40
30
20
10
0
50
40
30
20
10
0
*
**
*
***
Veh
Veh
Compound 43
Veh
Compound 40
Veh
Compound 40
Compound 43
Non-REM sleep
REM sleep
D
50
40
30
20
10
0
10
8
*
*
6
4
2
Veh
Veh
Compound 43
Veh
Veh
0
Compound 40
Compound 40
Compound 43
Figure 2. Effects of compound 40 and compound 43 on home cage activity (A), time spent in AW (AW, B), time spent in non-REM sleep (NREM, C) and time spent in REM sleep
*
**
***
(D). Effects measured and integrated over 12-h night period following administration. Data are presented as means SEM. p <0.05, p <0.01, p <0.001 (n = 7 for compound
40 and n = 8 for compound 43).
4. Chemelli, R. M.; Willie, J. T.; Sinton, C. M.; Elmquist, J. K.; Scammell, T.; Lee, C.;
In conclusion, we have described a novel series of dual orexin
receptor antagonists based on the heterocyclic replacement of
Richardson, J. A.; Williams, S. C.; Xiong, Y.; Kisanuki, Y.; Fitch, T. E.; Nakazato,
M.; Hammer, R. E.; Saper, C. B.; Yanagisawa, M. Cell 1999, 98, 437.
the dimethoxyphenyl moiety present in the tetrahydroisoquino-
line series around almorexant. Additional structure–activity rela-
tionship (SAR) studies of the phenethyl motif allowed the
identification of potent dual receptor antagonists with low nano-
molar potency for hOX1R and hOX2R. This series demonstrated
excellent in vitro cell-based activity and exhibited activity in the
in vivo sleep-model.
5. For recent reviews on the medicinal chemistry of orexin antagonists, see: (a)
Roecker, A. J.; Coleman, P. J. Curr. Topics Med. Chem. 2008, 8, 977; (b) Boss, C.;
Brisbare-Roch, C.; Jenck, F. J. Med. Chem. 2009, 52, 891; (c) Boss, C.; Brisbare-
Roch, C.; Jenck, F.; Aissaoui, H.; Koberstein, R.; Sifferlen, T.; Weller, T. Chimia
2008, 62, 974.
6. Brisbare-Roch, C.; Dingemanse, J.; Koberstein, R.; Hoever, P.; Aissaoui, H.;
Flores, S.; Mueller, C.; Nayler, O.; van Gerven, J.; de Haas, S. L.; Hess, P.; Qiu, C.;
Buchmann, S.; Schertz, M.; Weller, T.; Fischli, W.; Clozel, M.; Jenck, F. Nat. Med.
2007, 13, 150.
7. Koberstein, R.; Aissaoui, H.; Bur, D.; Clozel, M.; Fischli, W.; Jenck, F.; Mueller, C.;
Nayler, O.; Sifferlen, T.; Treiber, A.; Weller, T. Chimia 2003, 57, 270.
8. Kang, J.; Kim, Y. H.; Park, M.; Lee, C. H.; Kim, W. Synth. Commun. 1984, 14, 265.
9. Vicentini, C. B.; Manfrini, M.; Mazzanti, M.; Manferdini, M.; Morelli, C. F.;
Veronese, A. C. Heterocycles 2000, 53, 1285.
10. Campagna, F.; Carotti, A.; Casini, G. Tetrahedron Lett. 1977, 21, 1813.
11. FLIPR assay: Chinese hamster ovary (CHO) cells expressing the human orexin
receptors (hOX1R or hOX2R) were seeded into 96-well plates and incubated at
37 °C in 5% CO2 with the cytoplasmic fluorescent calcium indicator fluo-3 AM
(Molecular Probes). After washing the cells, intercellular Ca2+ mobilization was
monitored as a change in cell fluorescence intensity by FLIPR (Molecular
Devices). Differing concentrations of orexin antagonists were added to the
plates prior to addition of orexin A. For each antagonist, IC50 (the concentration
of compound needed to inhibit 50% of the agonistic response) is calculated.
12. For the purposes of this communication, a dual antagonist is defined as having
less than 20-fold selectivity for either OX1R or OX2R.
13. For pharmacodynamic sleep studies, adult male wistar rats were implanted
with radiotelemetric probes (Data Sciences International) under general
anesthesia. Those implants allow the recording of the electroencephalogram
(EEG), the electromyogram (EMG), home cage activity and body temperature in
freely moving animals. Compounds were administrated orally at the beginning
of the dark active phase and formulated in 100% PEG-400. We used groups of 7
or 8 animals, in a cross over design, with at least 4-day washout periods
separating consecutive administration.
Acknowledgments
The authors would like to thank Katalin Menyhart, Celia Müller,
Viktor Ribic, Pascal Rebmann, Ursula Fusco-Hug and Daniel Trach-
sel for expert technical support and Henri Ramuz for continuous
stimulating discussions.
References and notes
1. de Lecea, L.; Kilduff, T. S.; Peyron, C.; Gao, X.; Foye, P. E.; Danielson, P. E.;
Fukuhara, C.; Battenberg, E. L.; Gautvik, V. T.; Bartlett, F. S., 2nd; Frankel, W. N.;
van den Pol, A. N.; Bloom, F. E.; Gautvik, K. M.; Sutcliffe, J. G. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 322.
2. Sakurai, T.; Amemiya, A.; Ishii, M.; Matsuzaki, I.; Chemelli, R. M.; Tanaka, H.;
Williams, S. C.; Richardson, J. A.; Kozlowski, G. P.; Wilson, S.; Arch, J.;
Buckingham, R. E.; Haynes, A. C.; Carr, S. A.; Annan, R. S.; McNulty, D. E.; Liu,
W.; Terrett, J. A.; Elshourbagy, N. A.; Bergsma, D. J.; Yanagisawa, M. Cell 1998,
92, 573.
3. Lin, L.; Faraco, J.; Li, R.; Kadotani, H.; Rogers, W.; Lin, X.; Qiu, X.; de Jong, P. J.;
Nishino, S.; Mignot, E. Cell 1999, 98, 365.