Design and synthesis of substituted quinolines as novel and selective melanin concentrating hormone antagonists as anti-obesity agents

Article abstract of DOI:10.1016/j.bmcl.2006.07.006

A novel series of substituted quinoline analogs were designed and synthesized as potent and selective melanin concentrating hormone (MCH) antagonists. These analogs show potent (nM) activity (12a-k) with a moderate selectivity. Conversely, the conformatio

Full text of DOI:10.1016/j.bmcl.2006.07.006

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5207–5211
Design and synthesis of substituted quinolines as novel and selective
melanin concentrating hormone antagonists as anti-obesity agents
Namal C. Warshakoon,* Justin Sheville, Ritu Tiku Bhatt, Wei Ji, Jose L. Mendez-Andino,
Kenneth M. Meyers, Nick Kim, John A. Wos, Chrissy Mitchell, Jennifer L. Paris,
Beth B. Pinney, Ofer Reizes and X. Eric Hu
Drug Discovery Division, Procter and Gamble Pharmaceuticals Inc., 8700 Mason-Montgomery Road, Mason, OH 45040, USA
Received 24 May 2006; revised 26 June 2006; accepted 5 July 2006
Available online 25 July 2006
Abstract—A novel series of substituted quinoline analogs were designed and synthesized as potent and selective melanin concentrat-
ing hormone (MCH) antagonists. These analogs show potent (nM) activity (12a–k) with a moderate selectivity. Conversely, the con-
formationally constrained thienopyrimidinone analogs (18a–g) showed improved activity in MCH-1R and selectivity over 5HT2C.
Ó 2006 Elsevier Ltd. All rights reserved.
Obesity is a growing health issue in industrialized coun-
tries and developing nations.1 With the increased obese
and overweight population are increased risk of morbid-
ity and mortality as well as an increase in health care
costs due to the primary and secondary complications
of the epidemic.2 Thus, obesity and overweight present
a major burden on the economies of developed coun-
tries.2 Currently available therapeutic agents sibutr-
amine (Reductil or Meridia), an appetite suppressant,
and orlistat (Xenical), an inhibitor of fat absorption,
both suffer from limited efficacy and significant undesir-
able side effects.3 Therefore, there is great need for effec-
tive obesity pharmacotherapy. A major pharmaceutical
effort has focused on central nervous system (CNS) tar-
gets for treating obesity. These CNS targets include the
melanocortin, cannabionoid, and melanin concentrating
hormone receptors.
Furthermore, mice lacking MCH are lean and resistant
to diet induced obesity, a consequence of reduced feed-
ing and increased metabolic rate.6 In contrast, transgen-
ic mice that overexpress MCH are obese and insulin
resistant.7
Several small molecule MCH antagonists are reported in
the literature (Fig. 1),8–11 among them aminotetralin (1,
T-226296) was the first reported small molecule MCH
antagonist.8 Oral dosing of T-226296 (IC50 5.5 and
8.6 nM at human and rat MCH-1R, respectively)
blocked the hyperphagic effects of MCH injections into
the lateral ventricle of lean rats.8
Most of the recently disclosed MCH antagonists contain
a basic amine group and two aromatic groups joined by
an appropriate linker. Based on this design principle,
our proposed MCH antagonists were created via linking
a basic amino group and the aromatic portion to a quin-
oline platform (Fig. 2).12 The quinoline scaffold was
selected due to presumed desirable PK properties (e.g.,
solubility, metabolism, etc.)
MCH is a cyclic 19 amino acid peptide, which is synthe-
sized in the lateral hypothalamus and zona incerta in the
brain.4 MCH is highly conserved in vertebrate species
and appears to play an important role in regulating
energy feeding behavior and body weight. Fasting leads
to an increase in brain MCH expression in rats and CNS
administration of MCH stimulates food intake in rats.5,6
The requisite 3,8-substituted quinoline central core was
constructed via a Gould–Jacob condensation of substi-
tuted aniline 5 with ethoxymethylene malonate.13
The quinolone was then converted to the final com-
pounds via a series of synthetic transformations as
illustrated in Scheme 1. These compounds were then
evaluated for binding to the MCH-R1 using cells that
Keywords: Melanin concentrating hormone antagonists; Anti-obesity;
Quinolines.
*
Corresponding author. Tel.: +1 513 622 2284; fax: +1 513 622
0523; e-mail: Warshakoon.nc@pg.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.07.006

5208
N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5207–5211
F
F
O
N
O
O
N
H
O
O
N
N
N
H
H
N
N
O
H
F
O
2 SNAP-7941 (Synaptic)9
1 T-226296 (Takeda)
N
O
O
O
N
N
N
O
N
S
N
O
N
H
N
Cl
H
3 Abbott10
4 GSK11
Figure 1. Small molecule MCH antagonists.
N
O
N
O
N
N
H
basic amine
N
H
linker
F
linker
F
aromatic region
1 T-226296 (Takeda)
Figure 2. Design concept of new MCH antagonists.
O
Cl
CO2Et
CO2Et
a
c
b
NH2
N
N
H
CN
CN
CN
5
6
7
CO2Et
CO2Et
CO2Et
e
d
O
N
N
N
Ar
N
H2N
CN
H
8
9
10
R
OH
N
g
R
f
O
N
O
N
Ar
N
Ar
N
11
12
H
H
Scheme 1. Reagents and conditions: (a) Ph2O/250 °C/ethoxymethylenemalonate (60%); (b) SOCl2 (80%); (c) H2/Pd/EtOH/Et3N (65%); (d) RaNi/H2/
N2H4 (60%); (e) ArCO2H/HBTU/iPr2EtNH/DMF (80%); (f) LiB(C2H5)3H/THF/(70%); (g) (1) MsCl/Et3N/CH2Cl2; (2) NHR1R2/DMF (80%).
overexpressed the short form of the receptor (SLC-
1).14 Selectivity against the serotonin receptor
(5HT2C)15 was also evaluated because we and others
had determined this receptor showed similarity in the
pharmacophore binding pocket to the MCH-1R
(Table 1).

N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5207–5211
5209
Table 1. Activity and SAR relationship of substituted quinoline analogs against MCH and 5HT2C
X
O
N
Ar
N
R
12
Entry
Ar
R
H
X
MCH (nM)
102
5HT2C IC50 (nM)
12a
4.163
F
N
12b
12c
12d
12e
12f
12g
12h
12i
H
H
H
H
H
H
H
H
H
H
Me
1,408
195
7.467
O
F
F
F
N
N
2.807
OH
OH
142
3.595
N
N
310
>100,000
10,321
14,123
>100,000
14,641
16,515
20,230
4188
F
F
SO2
414
N
N
N
N
N
N
N
N
N
N
F
F
F
F
F
310
20,119
803
O
OH
OH
12j
1153
3544
31
N
N
12k
12l
N
F
N
In the aromatic region, the biaryl compounds (12a–e)
exhibited potent binding to the MCH receptor. Placing
a heteroatom in the inner benzene of the biaryl ring sys-
tem (12f–k) was detrimental to the activity. We were
able to obtain potent MCH activity as well as selectivity
by placing polar, hydrogen bond accepting groups on
the tertiary amine region (12d). The most active
(31 nM) as well as the most selective (ꢀ130-fold) com-
pound was 12l. In this instance, the hydrogen of the
amide nitrogen was replaced with a methyl group. Con-
formational restriction around the amide bond may be
directing these compounds to interact with the receptor
in a favorable fashion. We and others have hypothesized
that constraining this amide bond into a cyclic, rigid
ring system might enhance the activity further. Such a
ring system was described in two recent patents by scien-
tists at GlaxoSmithKline (Fig. 1, 4)11 and Neurocrine
Biosciences (Fig. 3, 13).16,17
O
O
N
N
S
N
N
N
Cl
13
Figure 3. An example from Neurocrine Biosciences.

5210
N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5207–5211
OH
N
16
H2N
S
CO2Et
a
b
Cl
S
CO2Et
NH2
Cl
N
N
15
14
R1
OH
N
R2
O
N
N
N
O
N
N
S
c
S
N
Cl
Cl
17
18
Scheme 2. Reagents and conditions: (a) DMF–DMA/EtOH/100 °C/MW (90%); (b) 16/DMF/100 °C/MW (85%); (c) (1) MsCl/Et3N/CH2Cl2; (2)
NHR1R2/DMF (60%).
Table 2. Activity and SAR relationship of substituted quinoline
derivatives against MCH and 5HT2C
The thienopyrimidinone core can be assembled rapidly
via a condensation between commercially available
thiophene carboxylate 14 and DMF–dimethylacetal
under microwave conditions. The resultant formami-
dine 15 was then condensed with amino alcohol 16
under yet another reaction using microwave condi-
tions. The alcohol was then transformed into a mesy-
late and subsequently displaced with appropriate
secondary amines to generate the desired compounds
(Scheme 2).
X
O
N
N
S
N
Cl
18
We were pleased to find that these compounds were not
only potent MCH ligands but also completely selective
with polar groups and hydrogen bond acceptors at the
amine terminus (18d–g) continuing to be selective li-
gands for MCH receptor over 5HT2C. In this series,
the tertiary amine derived from piperidine-4-ol (18d)
has the best activity (Ki 51 nM), while being completely
selective (Table 2).
Entry
X
MCH Ki (nM) 5HT2C IC50 (nM)
18a
197
126
91
1313
N
N
N
18b
18c
18d
18e
18f
18g
25,210
>100,000
>100,000
>100,000
>100,000
>100,000
In summary, we were able to design, synthesize, and
evaluate a novel series of substituted quinolines as
MCH antagonists. SAR investigation revealed that con-
formationally constrained analogs proved to be not only
more active in the MCH binding assay, but also more
selective against 5HT2C, warranting them for further
evaluations. The quinoline derived MCH antagonists
were potent binders to the hERG channel and this
problem was solved by further structural modifications.
Furthermore, selected compounds in this class show
efficacy in animal models. Those results will be reported
in due course.
51
N
N
OH
N S
773
136
110
O
O
O
N
N
N
N
Acknowledgments
NCW thanks Dr. Joseph Gardner for his continuous
support and advice. Ms. Anne Russell and Ms. Marica
Ketcha are gratefully acknowledged for the instrumen-
tation support.
References and notes
1. Flegal, K.; Carrol, M. S.; Ogden, C. L.; Johnson, C. L.
J. Am. Med. Assoc. 2002, 288, 1723.

N. C. Warshakoon et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5207–5211
5211
2. Kopelman, P. G. Nature 2000, 404, 635.
3. Gura, T. Science 2003, 299, 849.
4. Browning, A. Expert Opin. Ther. Patents 2004, 14, 313.
5. Qu, D.; Ludwig, D. S.; Gammeltoft, S. Nature 1996, 380,
243.
6. Shimada, M.; Tritos, N. A.; Lowell, B. B.; Flier, J. S.;
Maratos-Flier, E. Nature 1998, 396, 670.
7. Ludwig, D. S.; Tritos, N. A.; Mastaitis, J. W. J. Clin.
Invest. 2001, 107, 379.
8. Takehawa, S.; Asami, A.; Ishihara, Y. Eur. J. Pharmacol.
2002, 438, 1135.
9. Borowsky, B.; Durkin, M. M.; Ogozalek, K. Nat. Med.
2002, 8, 779.
10. Souers, A. J.; Gao, J.; Wodka, D. Bioorg. Med. Chem.
Lett. 2005, 15, 2752.
11. Carpenter, A. J.; Cooper, J. P.; Handlon, L.; Hertzog, D.
L.; Hyman, C. E.; Guo, Y. C.; Speake, J. D.; Witty, D. R.
WO Patent 2003033476 A1, 2003.
compound concentration varying from 100 lM to 11 fM.
Potency (Ki) and maximal efficacy were determined and
used to define structure–activity relationship.
15. 5HT2C assay was performed using a membrane prepara-
tion from cells that overexpress the receptor. Membranes
were purchased from Euroscreen (ES-318-M) and used
according to established protocols. Briefly, membranes
were incubated in a reaction mixture containing varying
concentrations of the compound and [3H] mesulergine
(final concentration 0.33 nM; Amersham), a compound
known to bind to the 5HT2C receptor. The binding buffer
contained 50 mM Tris HCl, 0.1% ascorbic acid, 5 mM
CaCl2, and 10 lg/ml Saponin. Nonspecific binding was
assessed by incubating the membranes with [3H] mesuler-
gine and excess 10 lM mianserin (ICN). After incubating
the mixture at room temperature for 60 min, bound
mesulergine was separated from unbound mesulergine by
filtration through glass fiber B filter (GF/B) plates (Mil-
lipore). Bound radioactive mesulergine was detected using
a scintillation counter.
12. Ishihara, Y.; Kamata, M.; Takekawa, S.; Suzuki, N.;
Kato, K. WO Patent 2003035624 A1, 2003.
13. Gould, R.; Jacob, W. J. Am. Chem. Soc. 1939, 61, 2890.
14. MCH binding using a Flashplate radioligand binding
assay was performed by Perkin–Elmer Biosignal (Toronto,
Canada). Full competition curves were generated with
16. Goodfellow, V.; Zhang, M.; Tamiya, J. WO Patent
2005047293 A1, 2005.
17. Barvian, K. K.; Handlon, A. L.; Hertzog, D.L.; Hyman,
C. E. WO Patent 2005042541 A1, 2005.