Quinazoline and benzimidazole MCH-1R antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.11.092

We have modified the previously reported 2-aminoquinoline 1 to provide two novel series of MCH-1R antagonists. Representative compounds from the quinazoline and benzimidazole series have been shown to be potent and selective, with promising in vitro eADME profiles.

Full text of DOI:10.1016/j.bmcl.2006.11.092

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1403–1407
Quinazoline and benzimidazole MCH-1R antagonists
Rosa Arienzo, Sue Cramp, Hazel J. Dyke,* Peter M. Lockey, Dennis Norman,
Alan G. Roach, Phil Smith, Melanie Wong and Stephen P. Wren
Argenta Discovery Limited, 8/9 Spire Green Centre, Flex Meadow, Harlow, Essex CM19 5TR, UK
Received 21 September 2006; revised 20 November 2006; accepted 30 November 2006
Available online 2 December 2006
Abstract—We have modified the previously reported 2-aminoquinoline 1 to provide two novel series of MCH-1R antagonists. Rep-
resentative compounds from the quinazoline and benzimidazole series have been shown to be potent and selective, with promising
in vitro eADME profiles.
Ó 2006 Elsevier Ltd. All rights reserved.
Melanin-concentrating hormone (MCH) is a cyclic 19-
amino acid peptide that is produced predominantly by
neurons in the lateral hypothalamus and zona incerta,
which project throughout the brain. The effects of
MCH are mediated through two distinct receptors,
MCH-1R and MCH-2R, both of which are members
of the G protein-coupled receptor (GPCR) super-family.
MCH-1R has been identified in mammals and rodents,
whereas functional MCH-2R has not been found in ro-
dents. The evidence for the role of MCH and the MCH-
1R in body weight and feeding regulation is abundant,1
and supports the hypothesis that MCH-1R antagonists
should provide a novel treatment for obesity. The large
number of patents that disclose novel MCH-1R antago-
nists provides an indication of the interest that this tar-
get has generated.2
ity to our compounds is evident. Related 2-aminoquino-
lines have been the subject of a recent publication by
Jiang et al. at Merck,7 with compound 4 selected for
in vivo efficacy studies. In order to obtain compounds
that did not lie within the generic scope of the Merck
patent, we decided to investigate the replacement of
the quinoline by alternative ring systems. The first
quinoline replacement investigated was the quinoxaline.
Subsequently, we investigated quinazolines and benzim-
idazoles in parallel, and the results obtained are dis-
closed herein.
Quinoxalines were prepared using the route depicted in
Scheme 1. Commercially available 2-hydroxyquinoxa-
line was nitrated using potassium nitrate in concentrated
sulfuric acid to give the 6-nitro derivative 5 selectively in
95% yield. Treatment of 5 with phosphorus oxychloride
and phosphorus pentachloride gave the chloro deriva-
tive 6 in 89% yield. Displacement of the chloro substitu-
ent with the appropriate amine followed by reduction of
the nitro group gave the amine 7, which was coupled
with the appropriate carboxylic acids under standard
conditions to provide an initial set of quinoxaline
amides.
We previously reported the discovery3 and optimisa-
tion4 of a series of quinolines as MCH-1R antagonists.
Compound 1 (Fig. 1) was one of the most potent and
selective analogues prepared during this optimisation.
Recently, others have reported the identification and
structure–activity relationships (SAR) of a related series
of quinolines, exemplified by compound 2.5
During the course of our work on the quinoline series, a
patent application disclosing related compounds was
published by Devita et al. at Merck.6 Example 1 in this
application is depicted as compound 3, and the similar-
The synthetic route used to prepare a series of quinazo-
lines is depicted in Scheme 2. Acylation of 2-aminoace-
tophenone with trichloroacetyl chloride followed by
cyclisation mediated by ammonium acetate in dimethyl-
sulfoxide provided dihydroquinazolinone 8 in overall
Keywords: Obesity; Melanin-concentrating hormone; MCH-1.
*
Corresponding author. Tel.: +44 (0) 1279 645631; fax: +44 (0) 1279
645657; e-mail: hazel.hunt@argentadiscovery.com
The structure depicted in reference 7 does not match the compound
name, and we believe that compound 4 is the correct structure.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.092

1404
R. Arienzo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1403–1407
CF3
OCF3
H
N
H
N
O
O
O
N
O
N
N
N
H
N
2
1
Cl
CF3
H
N
H
N
N
O
O
N
N
N
N
3
4
Figure 1. Compound structures.
Scheme 1. Synthesis of quinoxalines.
Scheme 2. Synthesis of quinazolines.
60% yield. Nitration of 8 gave the poorly soluble mono-
nitro derivative which was directly converted to the key
chloro intermediate 9 by treatment with phosphorus
oxychloride in low yield. Nucleophilic displacement of
the 2-chloro substituent by an amine followed by reduc-
tion of the nitro group using tin (II) chloride gave the
corresponding aniline 10. Amide formation under stan-
dard conditions provided a range of quinazolines.
87% yield of a 1:1 mixture of regioisomers, which was
subjected to reduction with tin (II) chloride in hydro-
chloric acid to give the mixture of regioisomers in 93%
yield. The regioisomers were separated at this stage to
provide 12 and 13. Aminobenzimidazole 12 was con-
verted into an amide using standard coupling condi-
tions, and the 2-chloro substituent was displaced with
an amine to provide the desired 2-aminobenzimidazoles.
The benzimidazoles were synthesized using the route
shown in Scheme 3. The key intermediate, 2-chloro-5-
nitrobenzimidazole 11, was prepared in two steps from
commercially available 4-nitro-1,2-phenylenediamine in
overall 45% yield. N-methylation of 11 provided an
Compounds were screened in a scintillation proximity as-
say in competition with 125I-[Phe13,Tyr19]-MCH binding
to MCH-1R membranes. IC50 values were determined in
duplicate wells using a 6-point dose–response curve. All
compounds were tested on two separate occasions and

R. Arienzo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1403–1407
1405
O
H
N
H2N
H2N
NO2
H
N
NO2
NO2
H2N
NH2
POCl3
O
Cl
N
H
84%
54%
N
11
NaH, DMF
MeI
87%
NH2
NO2
N
N
N
SnCl2, HCl
NH2
NO2
N
N
N
N
Cl
Cl
+
+
Cl
Cl
N
93%
13
12
Separation of isomers
RCO2H, DMF
HATU, DIPEA
94%
H
H
R2
R1R2NH, NMP
microwave
N
R
N
R
N
N
N
Cl
N
O
O
R1
N
Scheme 3. Synthesis of benzimidazoles.
compared with a standard MCH-1R inhibitor. IC50 val-
ues from the two determinations were within 2-fold of
each other. Functional activity was determined by mea-
suring the ability of selected compounds to inhibit
MCH-mediated calcium influx into cells expressing hu-
man MCH-1R using a Flexstation. Assays were run in
triplicate wells using a 7-point dose–response curve.
Compounds were only tested on a single occasion, since
the purpose of the functional assay was to ensure that
compounds acted as functional antagonists, not to delin-
eate specific SAR. Given the differences between the
binding and functional assays (including MCH concen-
tration, amount of serum present, timing of addition of
reagents) the correlation between the two assays was con-
sidered satisfactory. In general, the rank order of potency
was the same in both assays.
ies. We were disappointed to find that none of these
analogues possessed any activity in the MCH-1R bind-
ing assay, as shown in Table 1. Since we considered that
the loss of activity could be due to the lack of the 4-
methyl substituent on the biaryl ring, we prepared the
four analogous quinolines lacking this methyl substitu-
ent. Although there was some loss of activity compared
with the corresponding 4-methylquinolines, the quino-
lines were consistently significantly more potent than
the quinoxalines, as shown in Table 1.
Next we investigated replacement of the quinoline by a
quinazoline, and in this series we were able to maintain
the presence of the 4-methyl substituent. The first qui-
nazoline synthesized was ADS102891 (see Table 2), the
direct analogue of compound 1, and we were gratified
to find that this compound possessed useful activity.
This compound was approximately 10-fold less active
than the corresponding quinoline in both the binding as-
say and the functional assay (compound 1 IC50 15 and
10 nM; ADS102891 165 and 100 nM, respectively).
We prepared four quinoxalines, incorporating two of
the best amines (N-methylpiperazine and morpholine)
and two of the best aryl substituents (4-chloro and 4-tri-
fluoromethyl) previously identified in the quinoline ser-
Table 1. Comparison of quinoxalines and quinolines
R
R
H
H
N
O
N
N
N
O
R1
R1
O
O
N
N
R2
N
R2
NR1R2
R
Quinoxaline
Quinoline IC50 (nM)
Morpholine
Morpholine
CF3
Cl
IA at 1 lM
IA at 1 lM
IA at 1 lM
IA at 1 lM
172
961
N-methyl piperazine
N-methyl piperazine
CF3
Cl
122
38% at 1 lM
IA = <10% inhibition.

1406
R. Arienzo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1403–1407
Investigation of the substitution on the phenyl ring pro-
vided SAR consistent with those observed previously.4
Substitution at the 4-position is clearly favoured over
substitution at the 2- or 3-position; the results are shown
in Table 2. Replacement of the pyrrolidine by a substi-
tuted ethylenediamine (ADS103254) provided a com-
pound with comparable MCH-1R binding affinity
(IC50 260 nM compared with 165 nM). The 4-chloro
compound ADS103253 was selected for further profil-
ing, and the results obtained against a range of other hu-
man GPCRs are provided in Table 3. In addition,
ADS103253 was investigated for inhibition of five hu-
man cytochrome P450 enzymes (2D6, 3A4, 1A2, 2C9
and 2C19). No inhibition was observed when the com-
ent was carried out, using amines that had been identi-
fied in the quinoline series, see Table 4. Pyrrolidine and
isopropylamine provided the two most potent quino-
lines,4 and were also found to be advantageous in the
benzimidazole series. In contrast, although N-methyl-
piperazine provided a potent compound in the original
quinoline series, this substituent was clearly not benefi-
cial in the benzimidazole series. The N-methyl substitu-
ent on the benzimidazole ring was not required for
potency (compare ADS103168 with ADS103214) but
the regioisomeric N-methylbenzimidazole was signifi-
cantly less active, with an IC50 of 1.5 lM (data not
shown). ADS103274 was selected for further profiling,
and the results obtained against other human GPCRs
are shown in Table 3. ADS103274 did not inhibit hu-
man or rat cytochrome P450 enzymes at a concentra-
tion of 1 lM, was stable in human and rat
microsomes, achieved moderate permeability in a
Caco-2 assay and had excellent solubility (>1 mg/ml
at pH 7.4).
pound was tested at
a concentration of 1 lM.
ADS103253 was not cytotoxic in in vitro assays measur-
ing LDH release and intracellular ATP when tested at
concentrations up to 300 lM. In a hERG patch-clamp
assay, an IC50 of 10 lM was obtained.
The first benzimidazole synthesized was ADS103214,
and we were gratified to find that this compound was
almost as potent as the corresponding quinoline (IC50
40 nM vs 15 nM). Investigation of the amino substitu-
In summary, we have modified our previously identified
series of 2-aminoquinolines to provide two new series of
MCH-1R antagonists. Representative compounds from
Table 2. Investigation of the quinazoline series
H
R
N
O
N
R2
O
N
R1
N
Compound
R
NR1R2
IC50 (binding assay) nM
IC50 (functional assay) nM
ADS102891
ADS103253
ADS103293
ADS103317
ADS103316
ADS103254
4-CF3
4-Cl
2-Cl
Pyrrolidine
Pyrrolidine
Pyrrolidine
Pyrrolidine
Pyrrolidine
165
250
100
58
1,930
7% at 1 lM
390
Not tested
Not tested
41
3-Cl
4-OCH3
4-CF3
(CH3)2N(CH2)2N(CH3)
260
164
Table 3. Profile of ADS103253 and ADS103274
Compound
MCH-1R
MCH-2R
5-HT2A
5-HT2B
5-HT2C
D2L
a1
28%
27,000
ADS103253
ADS103274
250
25
2%
>30,000
0%
22%
1900
1500
39%
7%
>100 lM
19%
IC50 in nM, or % inhibition at 10 lM.
Table 4. Investigation of the amino substituent in the benzimidazole series
CF3
R
H
N
R1
N
N
O
O
R2
N
Compound
R1R2N
R
IC50 (binding assay) nM
IC50 (functional assay) nM
ADS103168
ADS103214
ADS103274
ADS103294
ADS103255
Pyrrolidine
Pyrrolidine
H
35
40
9
CH3
CH3
CH3
CH3
9
28
(CH3)2N(CH2)2N(CH3)
iPrNH
N-methylpiperazine
25
70
101
Not tested
37% at 1 lM

R. Arienzo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1403–1407
1407
2. (a) Browning, A. Expert Opin. Ther. Patents 2004, 14, 313;
(b) Dyke, H. J.; Ray, N. C. Expert Opin. Ther. Patents
2005, 15, 1303.
3. Clark, D. E.; Higgs, C.; Wren, S. P.; Dyke, H. J.; Wong,
M.; Norman, D.; Lockey, P. M. J. Med. Chem. 2004, 47,
3962.
the quinazoline and benzimidazole series have been
shown to be potent and selective, with a good in vitro
ADME profile.
Acknowledgments
4. Arienzo, R.; Clark, D. E.; Cramp, S.; Daly, S.; Dyke, H. J.;
Lockey, P.; Norman, D.; Roach, A. G.; Stuttle, K.;
Tomlinson, M.; Wong, M.; Wren, S. P. Bioorg. Med.
Chem. Lett. 2004, 14, 4099.
5. (a) Ulven, T.; Frimurer, T. M.; Receveur, J.-M.; Little, P.
B.; Rist, Ø.; Nørregaard, P. K.; Ho¨gberg, T. J. Med. Chem.
2005, 48, 5684; (b) Ulven, T.; Little, P. B.; Receveur, J.-M.;
Frimurer, T. M.; Rist, Ø.; Nørregaard, P. K.; Ho¨gberg, T.
Bioorg. Med. Chem. Lett. 2006, 16, 1070.
6. Devita, R. J.; Chang, L.; Chaung, D.; Hoang, M.; Jiang, J.;
Lin, P.; Sailer, A. W.; Young, J. R. WO03/045313.
7. Jiang, J.; Hoang, M.; Young, J. R.; Chaung, D.; Eid, R.;
Turner, C.; Lin, P.; Tong, X.; Wang, J.; Tan, C.; Feighner,
S.; Palha, O.; Hreniuk, D. L.; Pan, J.; Sailer, A. W.;
MacNeil, D. J.; Howard, A.; Shearman, L.; Stribling, S.;
Camacho, R.; Strack, A.; Van de Ploeg, L. H. T. Bioorg.
Med. Chem. Lett. 2006, 16, 5270.
The authors acknowledge Colin Bright and Anne White
for the provision of eADME data, and Russell Scamm-
ell and Mike Podmore for the provision of analytical
data.
References and notes
1. Reviews: (a) Carpenter, A. J.; Hertzog, D. L. Expert Opin.
Ther. Patents 2002, 12, 1639; (b) Collins, C. A.; Kym, P. R.
Curr. Opin. Investig. Drugs 2003, 4, 386; (c) Hervieu, G.
Expert Opin. Ther. Targets 2003, 7, 495; (d) Kowalski, T. J.;
McBriar, M. D. Expert Opin. Investig. Drugs 2004, 13,
1113.