1,3-Biarylureas as selective non-peptide antagonists of the orexin-1 receptor

Article abstract of DOI:10.1016/S0960-894X(01)00343-2

This communication reports SARs for the first orexin-1 receptor antagonist series of 1-aryl-3-quinolin-4-yl and 1-aryl-3-naphthyridin-4-yl ureas. One of these compounds, 31 (SB-334867), has excellent selectivity for the orexin-1 receptor, blood-brain barrier permeability and shows in vivo activity following ip dosing.

Full text of DOI:10.1016/S0960-894X(01)00343-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1907–1910
1
,3-Biarylureas as Selective Non-peptide Antagonists of the
Orexin-1 Receptor
a,
a
a
a
Roderick A. Porter, * Wai N. Chan, Steven Coulton, Amanda Johns,
a
a
b
a
c
Michael S. Hadley, Katherine Widdowson, Jeffrey C. Jerman, Stephen J. Brough,
c
a
a
a
Martyn Coldwell, Darren Smart, Frances Jewitt, Phillip Jeffrey and Nigel Austin
aGlaxoSmithKline Pharmaceuticals, New Frontiers Science Park North, Third Avenue, Harlow, Essex CM19 5AW, UK
bGlaxoSmithKline Pharmaceuticals, King of Prussia, PA 19406, USA
cGlaxoSmithKline Pharmaceuticals, The Frythe, Welwyn, Herts AL6 9AR, UK
Received 6 February 2001; revised 30 April 2001; accepted 7 May 2001
Abstract—This communication reports SARs for the first orexin-1 receptor antagonist series of 1-aryl-3-quinolin-4-yl and 1-aryl-
-naphthyridin-4-yl ureas. One of these compounds, 31 (SB-334867), has excellent selectivity for the orexin-1 receptor, blood–brain
barrier permeability and shows in vivo activity following ip dosing. # 2001 Elsevier Science Ltd. All rights reserved.
3
Introduction
To explore the physiological role of the OX receptor a
1
programme of work was initiated to identify selective
OX1 receptor antagonists. This communication
describes 1,3-biaryl ureas as the first antagonists of the
Recently, two novel hypothalamic neuropeptides
orexin-A and orexin-B were reported. Both orexin-A, a
3
3 amino acid peptide, and orexin-B, a 28 amino acid
OX receptor and progress in optimising this structural
1
peptide, are derived from the same precursor, prepro-
orexin. These two peptides activate two closely related,
previously orphan, G-protein coupled receptors, orexin-
class, leading to a compound, 31, with excellent selec-
tivity for the OX receptor, demonstrable blood–brain
barrier permeability and in vivo activity.
1
1
1
(OX ) and orexin-2 (OX ). Orexin-A shows an
1
2
approximately 10-fold higher functional affinity for the
OX receptor than orexin-B, whilst at the OX receptor,
orexin-A and orexin-B showsimilar functional affinity.
High throughput functional screening of the in-house
collection against a CHO cell line expressing the human
OX receptor, using a FLIPR based calcium assay,
1
1
2
2
2
In the adult rat brain, prepro-orexin mRNA and
immunoreactive orexin-A are localised in neurons
within and around the lateral hypothalamus. These
areas are known to be involved in a range of physio-
logical actions, including regulation of feeding and
energy balance. Indeed, orexin-A has been shown to
stimulate feeding when injected into the lateral ven-
tricles of male Wistar rats and prepro-orexin mRNA is
elevated in rats after withdrawal of food, suggesting a
identified the 1-aryl-3-(quinolin-4-yl)urea 1.
3
physiological role of orexins in feeding. Recent data
also suggest roles for the orexins in regulation of blood
4
5
pressure, the neuroendocrine system and the sleep–
6
wake cycle.
This compound has good functional affinity for the OX
7
1
receptor and generally good selectivity over other
GPCRs, including the OX2 receptor (pKb 5.6).
However, high binding affinity for both the 5-HT2B and
*Corresponding author. Fax: +44-1279-622790; e-mail: rod_a_
porter@sbphrd.com
0
960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00343-2

1908
R. A. Porter et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1907–1910
5-HT2C receptors was seen. This poor selectivity limited
the potential of 1 as a tool compound.
The unsubstituted phenyl analogue 7 was inactive, no 2-
substituted analogue showed significant activity (data
not shown) while 3-substituted analogues, for example
8
–10, gave only modest OX receptor functional affinity.
1
However, the 4-N,N-dimethylamino substituted ana-
logue 11 showed excellent functional affinity, 50-fold
selectivity over OX (pK 6.1) and 8-fold selectivity over
the 5-HT2C receptor, consistent with the hypothesis.
The pyrrolidine 12 had 20-fold lower affinity suggesting
the binding site accommodating this substituent is
sterically constrained, although the N,N-diethyl analo-
gue 13 was only marginally less active. It is likely that a
lipophilic interaction is involved as demethylation (14)
Indole Ring SAR
To develop an understanding of the SAR of 1, attention
was initially directed to the role of the indole ring in
determining both functional affinity and selectivity
2
b
(
Table 1). Amongst indole modifications, functional
affinity comparable to 1 was seen with the 1, 2-dimethyl-
indole 2. A large N-substituent—as in the N-benzyl
analogue 3—was not tolerated while the des-methyl
analogue 4 was only marginally active. The indole
ring could be replaced by the indoline 5 retaining full
OX functional affinity but also high 5-HT affinity.
Previous studies on substituted 1-(3-pyridylcarbamoyl)-
indoline 5-HT2C receptor antagonists, for example 6,
had established that a lipophilic 5,6-disubstituted indo-
gave a >50-fold drop in OX receptor functional affi-
1
nity and the polar nitrile 15 was inactive. The lipophilic
methylthio analogue 16 retained excellent functional
affinity but unlike 11 lacked significant selectivity over
5-HT2B or 5-HT2C receptors.
1
2C
8
line ring gave high 5-HT2C receptor affinity. It was
Although 3-mono-substitution gave only modest affinity,
increased affinity could be achieved with 3,4-disub-
stitution. For example, increased affinity was seen with a
3-chloro-4-methoxy substituent 18 relative to the corre-
sponding modestly active mono-substituted analogues 8
and 17. This may be due to conformational constraint
of the 4-substituent optimising an interaction of the
methyl group with a small lipophilic pocket. Dis-
appointingly no such increase was seen for the 3-chloro-
therefore hypothesised that removal of the lipophilic
pyrrolo ring from indole 1 to give substituted phenyl
derivatives should reduce 5-HT2C receptor affinity.
4-dimethylamino analogue 19 relative to 11. 3,4-Dime-
thoxy substitution (20) resulted in loss of activity, but
activity was restored when constrained in benzodioxepin
21.
Table 1. 1-Quinolin-4-yl-3-(hetero)arylurea SAR
To further develop the hypothesis that reducing lipo-
philicity of the indole ring 3-position of 1 should
increase selectivity over the 5-HT2C receptor, a series of
benzoxazole derivatives was prepared. The un-sub-
stituted benzoxazole 22 retained moderate OX func-
1
tional affinity. Introduction of a 2-methyl substituent,
23, gave a 5-fold increase in functional affinity con-
Compound
Ar
OX
pK
1
5-HT2B
5-HT2C
pK
i
sistent with the presence of a small hydrophobic pocket
in the receptor. Furthermore, 23 showed 30-fold selec-
tivity over both 5-HT2B and 5-HT2C receptors. Com-
pared to 1, 23 shows a 160-fold reduction in binding
affinity at the 5-HT2B receptor and 25-fold reduction at
the 5-HT2C receptor. The regioisomeric benzoxazole 24
was inactive.
a
a
a
b
pK
i
1
2
3
4
5
7
8
9
1-Methylindol-5-yl
1,2-Dimethylindol-5-yl
1-Benzylindol-5-yl
Indol-5-yl
7.5
7.3
<6.1
6.2
7.6
<6.1
6.4
6.1
6.6
7.8
6.4
7.4
<6.1
<6.2
7.9
6.8
7.4
7.4
8.6
8.4
7.6
7.1
1-Methylindolin-5-yl
7.9
7.3
C
6
H
5
3-Cl-C
3-NMe -C
3-OMe-C
4-NMe -C
6
H
4
2
6
H
H
H
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
6
4
Quinoline Ring SAR
2
6
4
7.1
7.3
6.9
7.5
4-Pyrrolidin-1-yl-C
4-NEt -C
4-NH -C
6
H
4
Having demonstrated that good functional affinity and
significant selectivity could be achieved by modification
of the indole ring of 1, attention was directed to the
quinoline ring (Fig. 1). The lack of activity of the sub-
stituted pyridyl (25) and naphthyl (26) analogues high-
lighted the importance of both the quinoline benzo-ring
and the quinoline ring nitrogen for good functional
affinity.
2
6
H
H
4
2
6
4
4-CN-C
4-SMe-C
6
H
4
6
H
H
4
7.5
7.8
7.5
7.3
4-OMe-C
3-Cl-4-OMe-C
3-Cl-4-NMe -C
3,4-diOMe-C
3,4-O(CH O-C
Benzoxazol-6-yl
6
4
6
H
3
2
6
H
3
6
H
H
6 3
3
<6.3
7.4
7.0
7.7
<6.2
2
)
3
2-Methylbenzoxazol-6-yl
Benzoxazol-5-yl
6.2
6.2
Seeking to decrease lipophilicity and to increase solubi-
lity, aza-substitution was investigated (Table 2). The
inactivity of the quinazolin-4-yl analogue 27 may reflect
a
Values are means of three experiments standard deviation <0.1.

R. A. Porter et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1907–1910
1909
an intramolecular hydrogen bond mediated EZ con-
formation in solution. This contrasts with an extended
following ip dosing reflected its short half-life (t1=2=
0.4 h). It also showed low oral bioavailability (10%).
ZZ urea conformation in biaryl ureas lacking intramo-
Good func-
9À11
lecular hydrogen bonding potential.
tional affinity was seen with 1,5-naphthyridines 28–31.
Furthermore, the benzoxazole 31 (SB-334867) showed
Despite a modest pharmacokinetic profile, SB-334867
shows a range of in vivo activities mediated by orexin-A
following ip dosing. These activities include blockade of
1
3
5
greater selectivity over 5-HT , 5-HT and approxi-
0-fold selectivity over OX2 (pKb 5.7), 100-fold or
orexin-A induced grooming
1
and both orexin-A
4
induced and natural feeding.
2
B
2C
mately 50 other molecular targets, in particular other
GPCRs and ion channels. Full details of the in vitro phar-
macology of SB-334867, including binding data, are
described elsewhere. Broadly, aryl ring substituent SAR
in the 1,5-naphthyridines followed that seen in the corre-
sponding quinolines although functional affinities were
marginally reduced. 1,6- 32 and 1,8-naphthyridines 33
were substantially less active than the 1,5-naphthyridines.
Synthesis of SB-334867
12
SB-334867 was prepared according to Scheme 1. The
hydroxy compound 35 was prepared according to lit-
1
5
erature methods. Conversion to amine 37 was accom-
plished through a two-step process via triflate 36 and
subsequent amination with propylamine hydrochloride
1
6
in pyridine—a modification of a literature procedure.
This route gave moderate yields but products were
readily isolated. The final key urea formation involved
conversion of 2-methylbenzoxazole-6-carboxylic acid to
the acyl azide in situ using diphenyl phosphoryl azide at
room temperature, adding the weakly nucleophilic 4-
amino-1,5-naphthyridine and heating to reflux in tolu-
ene. Under these conditions, a good yield of SB-334867
was obtained with only small quantities of symmetrical
urea by-product.
SB-334867 was isolated as the HCl salt.17
Fig 1. Quinoline modifications.
Other compounds described were prepared by a similar
1
8,19
Table 2. Aza-substituted quinolines
procedure or by procedures described elsewhere.
Compound
Ar
N
OX
pK
1
5-HT2B 5-HT2C
a
a
a
b
pK
i
pK
i
27
28
29
30
31
32
33
1-Methylindol-5-yl
1-Methylindol-5-yl
-3- <6.3
-5-
-5-
-5-
7.3
7.4
7.5
7.4
5.9
6.5
2 6 4
4-NMe C H
4-SMeC
6
H
4
2-Methylbenzoxazol-6-yl -5-
2-Methylbenzoxazol-6-yl -6-
2-Methylbenzoxazol-6-yl -8-
5.4
<5.3
Scheme 1. Synthesis of SB-334867. Conditions: (i) Diethyl
ethoxymethylenemalonate, Dowtherm A; (ii) 4% NaOH; (iii) heat,
4
3% over three steps; (iv) (CF
2.
3 2
SO )O, 2,6-lutidine, 34%; (v)
a
Values are means of three experiments standard deviation <0.1.
nPrNH HCl 85%; (vi) 2-methylbenzoxazole-6-carboxylic acid,
PhO) P(O)N , 72% (as HCl salt).
(
2
3
CNS Penetration of OX Receptor Antagonists
1
Table 3. CNS penetration of selected compounds
cLogPb
Several compounds were assessed for CNS penetration
by intravenous infusion of drug to steady state (Table
Compound
CNS penetration
ratio brain/blood
Css (mM)
a
3
). Disappointingly, only lowlevel penetration wa s
1
1
1
3
0.06:1
0.16:1
0.1:1
4.2
0.7
1.0
0.49
4.0
5.1
4.4
2.8
achieved with quinoline analogues with the most pene-
trant being the highly lipophilic, non-selective 13. In
contrast, the more polar naphthyridine SB-334867
showed CNS penetration as determined by intravenous
infusion to steady state and also from a bolus 10 mg/kg
ip dose (brain level 2 h post dose 3.8 mM, 4 h <30 nM).
The relatively short duration of SB-334867 in the brain
16
31 (SB-334867)
0.4:1
aSteady state brain–blood ratio following intravenous infusion of 1.8
mg/kg/h free base drug for 12 h.
b
cLogP version 3 licensed by Daylight CIS Inc. of Mission Viejo, CA,
USA and BioByte Corp. of Claremont, CA, USA.

1910
R. A. Porter et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1907–1910
Summary
resuspended in, Tyrode’s medium containing 2.5 mM probenecid
ꢀ
and 1% gelatine, before being incubated for 30 min at 37 C
with either buffer alone (control) or buffer containing test
compound (10 pM–10 mM). The fluorescence (lex=488 nm,
lEM=540 nm) was then monitored using a fluorometric ima-
ging plate reader (FLIPR; Molecular Devices, UK) before and
after the addition of orexin-A in the continued presence of the
test compound. Data analysis: responses were measured as
peak fluorescence intensity (FI) minus basal FI, and curve-fit-
ting and parameter estimation were carried out according to a
four parameter logistic model, and results are expressed as a
In this communication, SAR of the high throughput
screening hit 1 at the OX receptor has been outlined.
1
The importance of the quinoline nitrogen and the quin-
oline benzo-ring has been demonstrated. The presence
of a small hydrophobic pocket associated with the aryl
ring substituent is postulated. The urea is proposed to
bind to the receptor in the extended lowenergy ZZ
conformation. Selectivity has been enhanced by mod-
ifying both the indole and quinoline motifs of 1. Finally
by replacement of the quinoline by a 1,5-naphthyridine,
CNS penetration has been increased. The naphthyridine
SB-334867 possesses CNS penetration, excellent selec-
tivity over a range of other receptors and ion channels
and in vivo activity following ip dosing.
pK
For further details see refs 2 and 12.
. Bromidge, S. M.; Dabbs, S.; Davies, D. T.; Duckworth,
b
. Results are the mean of three separate determinations.
8
D. M.; Forbes, I. T.; Ham, P.; Jones, G. E.; King, F. D.;
Saunders, D. V.; Starr, S.; Thewlis, K. M.; Wyman, P. A.;
Blaney, F. E.; Naylor, C. B.; Bailey, F.; Blackburn, T. P.;
Holland, V.; Kennett, G. A.; Riley, G. J.; Wood, M. D. J.
Med. Chem. 1998, 41, 1598.
9
1
. Ferraro, M. B. THEOCHEM 2000, 528, 199.
0. Singha, N. C.; Sathyanarayana, D. N. J. Chem. Soc.,
References and Notes
Perkin Trans. 2 1997, 157.
1. Yamaguchi, K.; Shudo, K. J. Agric. Food Chem. 1991, 39,
793.
1
1
. Sakurai, T.; Amemiya, A.; Ishii, M.; Matsuzaki, I.; Che-
melli, R. M.; Tanaka, H.; Williams, S. C.; Richardson, J. A.;
Kozlowski, G. P.; Wilson, S.; Arch, J. R. S.; Buckingham,
R. E.; Haynes, A. C.; Carr, S. A.; Annan, R. S.; McNulty,
D. E.; Liu, W.-S.; Terrett, J. A.; Elshourbagy, N. A.; Bergsma,
D. J.; Yanagisawa, M. Cell 1998, 92, 573.
2. Smart, D.; Jerman, J. C.; Brough, S. J.; Rushton, S. L.;
Murdock, P. R.; Jewitt, F.; Elshourbagy, N. A.; Ellis, C. E.;
Middlemiss, D. N.; Brown, F. Br. J. Pharmacol. 1999, 128, 1.
12. Smart, D.; Sabido-David, C.; Brough, S. J.; Jewitt, F.;
Johns, A.; Porter, R. A.; Jerman, J. C. Br. J. Pharmacol. 2001,
132, 1179.
13. Duxon, M. S.; Stretton, J.; Starr, K.; Jones, D. N. C.;
Holland, V.; Riley, G.; Jerman, J.; Brough, S. J.; Smart, D.;
Johns, A.; Chan, W.; Porter, R. A.; Upton, N. Psycho-
pharmacology 2001, 153, 203.
14. Haynes, A. C.; Jackson, B.; Chapman, H.; Tadayyon, M.;
Johns, A.; Porter, R. A.; Arch, J. R. S. Regul. Pep. 2000, 96,
45.
3. Haynes, A. C.; Jackson, B.; Overend, P.; Buckingham,
R. E.; Wilson, S.; Tadayyon, M.; Arch, J. R. S. Peptides 1999,
2
0, 1099.
. Chen, C.-T.; Hwang, L.-L.; Chang, J.-K.; Dun, N. J. Am.
15. Adams, J. T.; Bradsher, C. K.; Breslow, D. S.; Amore,
S. T.; Hauser, C. R. J. Am. Chem. Soc. 1946, 68, 1317.
16. Radinov, R.; Haimova, M.; Simova, E. Synthesis 1986,
886.
4
J. Physiol. Regul. Integr. Comp. Physiol. 2000, 278, R692.
5
1
6
. Pu, S.; Jain, M. R.; Kalra, P. S.; Kalra, S. Regul. Pept.
998, 78, 133.
. Piper, D. C.; Upton, N.; Smith, M. I.; Hunter, A. J. Eur. J.
17. Characterisation of SB-334867 hydrochloride 1H NMR
6
(250 MHz, DMSO-d ) d 2.61 (3H, s), 7.32 (1H, dd, J=2 and
Neurosci. 2000, 12, 726.
. Screening protocol in brief : CHO cells stably expressing
OX1 receptors (CHO-OX ), were seeded (20,000 cells/well)
into black walled clear-base 96-well plates (Costar, UK) in
MEM-Alpha medium and cultured overnight. Cells were
loaded with the cytoplasmic calcium indicator, Fluo-3AM
9 Hz), 7.65 (1H, d, J=9 Hz), 8.08 (1H, d, J=2 Hz), 8.17 (1H,
dd, J=4 and 9 Hz), 8.67 (1H, dd, J=1 and 10 Hz), 8.77 (1H,
d, J=7 Hz), 9.09 (1H, d, J=7 Hz), 9.22 (1H, dd, J=1 and
4 Hz), 10.80 (1H, s), 10.92 (1H, s). LC–MS (ES , TFA/
2 13 5 2
MeCN/H O variable gradient) calcd for C17H N O +H 320,
found 320 (100%).
7
1
+
(4 mM; Teflabs, Austin, TX, USA) in the presence of 2.5 mM
probenecid, and then washed four times with, and finally
18. Johns, A.; Porter R. A. Patent WO 9909024-A1, 1999.
19. Johns, A.; Porter R. A. Patent WO 9958533-A1, 1999.