Discovery of 1,3-disubstituted-1H-pyrrole derivatives as potent Melanin-Concentrating Hormone Receptor 1 (MCH-R1) antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.07.079

A series of 1,3-disubstituted-1H-pyrrole-based antagonists of the human Melanin-Concentrating Hormone Receptor 1 (h-MCH-R1) are reported. High-throughput screening of the AstraZeneca compound collection yielded 1, a hit with moderate affinity towards MCH-

Full text of DOI:10.1016/j.bmcl.2008.07.079

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4859–4863
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of 1,3-disubstituted-1H-pyrrole derivatives as potent
Melanin-Concentrating Hormone Receptor 1 (MCH-R1) antagonists
Susanne Berglund a, Bryan J. Egner a, Henrik Gradén a, Joakim Gradén a, David G. A. Morgan b,
c,
Tord Inghardt a, , Fabrizio Giordanetto
*
*
a Medicinal Chemistry, AstraZeneca R&D Mölndal, Pepparedsleden 1, SE-431 83, Mölndal, Sweden
b CVGI Bioscience, AstraZeneca R&D Alderley Park, Macclesfield, Cheshire SK 104TG, UK
c Lead Generation, AstraZeneca R&D Mölndal, Pepparedsleden 1, SE-431 83, Mölndal, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 1,3-disubstituted-1H-pyrrole-based antagonists of the human Melanin-Concentrating Hor-
mone Receptor 1 (h-MCH-R1) are reported. High-throughput screening of the AstraZeneca compound
collection yielded 1, a hit with moderate affinity towards MCH-R1. Subsequent structural manipulations
and SAR analysis served to rationalize potency requirements, and 12 was identified as a novel, functional
MCH-R1 antagonist with favorable pharmacokinetic properties.
Received 16 June 2008
Revised 17 July 2008
Accepted 19 July 2008
Available online 24 July 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Melanin-Concentrating Hormone
MCH
MCH-R1 antagonists
Obesity
Worldwide obesity has been steadily increasing for the past 20
years.1 In the United States, obesity is now the most common
chronic disease, affecting one in three of all Americans, including
children.1 Morbidities associated with obesity and being over-
weight are contributing to more than 300,000 deaths per year
among the US population, whereas the related economic expendi-
ture has exceeded US $100 billions.1
The Melanin-Concentrating Hormone (MCH) system has at-
tracted the interest of pharmaceutical companies due to its role
in regulating food intake and energy expenditure in animal models.
Specifically, it was shown that intracerebroventricular (icv) injec-
tion of MCH, a cyclic 19 amino acid peptide, promoted food intake
in rats.2 Additionally, mice overexpressing the MCH gene demon-
strated hyperphagia and obesity traits,3 while transgenic MCH
Receptor 1 (MCH-R1) knockout mice displayed hyperactivity and
a lean phenotype and were resistant to diet-induced obesity.4 It
was therefore postulated that MCH-R1 antagonism could offer a
viable treatment for obesity and several MCH-R1 antagonists have
been reported.5–9 The discovery of novel, potent MCH-R1 antago-
nists with favorable in vivo pharmacokinetic properties is herein
presented.
A high-throughput screening (HTS) campaign against MCH-R1
was initiated. It identified 1 (Fig. 1) as a hit with good receptor
affinity (IC50 = 92 nM) but moderate functional antagonism
(MCH-R1 GTPcS IC50 = 804 nM). Docking of 1 to a MCH-R1 homol-
ogy model10 postulated that the tertiary amine group and the car-
bonyl function engaged in polar interactions with the receptor, as
shown in Figure 2. According to the docking, the 3,4-dichloro-ben-
zyl group partially filled a narrow and extended hydrophobic pock-
et formed by the transmembrane helices 3, 5, and 6. Interestingly,
these findings seem to be in agreement with a recently published
structural model for MCH-R1 antagonism.11 The nitro-thiophene
fragment of 1 was considered problematic due to a number of con-
cerns about metabolic stability and reactive metabolites formation.
Chemical scan of an acid chlorides lead generation set was per-
formed to identify appropriate structural replacements. Here, re-
agents were selected to ensure adequate coverage of chemical
N
O
O
Cl
N+
N
H
Cl
-O
S
1
MCH-R1 IC50 = 0.092 μM
MCH-R1 GTPγS IC50 = 0.804 μM
* Corresponding authors. Tel.: +46 31 7761954; fax: +46 31 7763724 (T.
Inghardt), tel.: +46 31 7065723; fax: +46 31 7763710 (F. Giordanetto).
E-mail addresses: tord.inghardt@astrazeneca.com (T. Inghardt), fabrizio.
giordanetto@astrazeneca.com (F. Giordanetto).
Figure 1. Hit originated from in-house high-throughput screening campaign.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.079

4860
S. Berglund et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4859–4863
space and, as a result, the 3-chloro-phenoxymethyl moiety was
identified as a more suitable drug-like segment with similar
MCH-R1 binding properties (MCH-R1 GTP S IC50 = 1 M) to the ni-
tro-thiophene (MCH-R1 GTP S IC50 = 0.8 M), and it was used in
c
l
c
l
the subsequent optimization efforts. Here, a number of elongated
fragments were designed to validate the proposed structural
framework and to investigate their effects on MCH-R1 potency.
The results for derivatives 2–21 are presented in Table 1. The syn-
thesis of 10 is reported in Scheme 1.12
Replacement of the phenyl ring in 1 with a lipophilic naphtha-
lene (2) improved MCH-R1 potency and functional antagonism to
41 and 104 nM, respectively. On the contrary, the subsequent isos-
teric substitution of naphthalene to more polarized moieties,
including 2,3-dihydro-benzo[1,4]dioxine (3), benzo[1,3]dioxole
(4) and indole (5), was detrimental to affinity, as shown in Table
1. Inclusion of more extended biaryl fragments (6–7) failed to im-
prove potency over 1 (Table 1). However, the introduction of 1-(4-
Figure 2. Predicted MCH-R1-docked conformation of 1. Observed electrostatic
interactions with D123 and Y272 side chains are displayed as dashed green lines.
Table 1
In vitro MCH-R1 binding and functional data for 2–21
R
R
O
N
N
O
O
Cl
Cl
O
N
H
N
H
a
b
a
b
Compound
R
MCH-R1 IC50 (GTP
c
S IC50
)
(l
M) Compound
R
*
*
MCH-R1 IC50 (GTP
c
S IC50
)
(l
M)
F
*
F
2
0.041 (0.104)
12
0.031 (0.015)
F
N
N
O
O
*
3
4
0.903
13
14
15
16
17
18
19
20
21
>33
O
O
*
*
F
F
0.082 (0.505)
2.34
*
*
0.043 (0.117)
9.65
N
N
O
F
Cl
5
N
H
N
N
*
*
O
6
0.120
1.5
N
N
N
*
F
F
N
7
0.297
0.048 (0.092)
0.078 (0.103)
0.927
*
*
F
F
F
F
O
O
8
0.710
*
*
N
N
O
F
*
9
0.100 (0.090)
0.052 (0.145)
0.029 (0.032)
O
F
N
F
N
S
*
*
*
*
10
0.347
Cl
N
N
N
F
F
F
F
11
2.14
N
F
N
F
a
Values are mean of at least two experiments. Compounds competed with 125I-MCH for binding at the human MCH1 receptor (h-MCH-R1) expressed in the HEK293 cell
line.
b
Values are mean of at least two experiments. Functional assay measuring [35S]GTP
cS accumulation.

S. Berglund et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4859–4863
4861
O
O
N
Cl
O
N
O
N
Cl
Cl
OH
N
H
Cl
N
t-BuOK, THF, rt, 2h
DCM, rt, 1h
H2N
H
86%
87%
Cl
O
1.
DCE, reflux, 2h
Cl
O
Cl
2. MeOH, reflux, 18h
O
NH
CN
O
N
N
O
1. DCM, rt, 10m
O
O
N
N
+
CN
H
H
H
N
2. MP-BH(OAc)3, rt, 3h
Cl
10 65%
72%
73%
Cl
O
O
AcOH, reflux, 18h
+
H2N
CN
O
O
H
Scheme 1.
Table 2
In vitro MCH-R1 binding and functional data for 22–34
F
F
N
N
F
CORE
CORE
O
O
F
Y
Y
F
F
R
R
a
b
a
b
Compound
R
Core
Y
C
MCH-R1 IC50 (GTP
c
S IC50
)
(
l
M) Compound
R
Core
Y
MCH-R1 IC50 (GTP
cS IC50) (lM)
O
O
*
11
3-Cl
0.029 (0.044)
28
3,4-diF
3,4-diF
C
C
2.26
*
*
*
N
H
N
H
N
H
N
N
*
*
O
O
H
N
22
23
24
3-Cl
3-Cl
3-Cl
C
C
N
0.109
0.636
3.05
29
30
31
0.104
1.25
*
N
H
N
*
*
O
*
N
H
N
3-Cl
3-Cl
N
C
N
N
*
*
H
O
*
O
*
N
*
0.169
N
N
N
H
N
H
*
*
O
O
N
*
*
25
26
3,4-diF
3-Cl
C
C
0.029 (0.048)
0.042 (0.073)
32
33
3-Cl
3-Cl
N
N
0.313
0.303
N
H
*
N
H
*
O
O
H
N
N
*
*
N
N
H
O
*
*
*
O
H
*
27
3-Cl
C
0.111
34
3,4-diF
C
0.269
N
N
*
N
H
N
H
O
H
a
Values are mean of at least two experiments. Compounds competed with 125I-MCH for binding at the human MCH1 receptor (h-MCH-R1) expressed in the HEK293 cell line.
b
Values are mean of at least two experiments. Functional assay measuring [35S]GTP
cS accumulation.

4862
S. Berglund et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4859–4863
substituted-phenyl)-1H-pyrroles, 9–12, afforded potent MCH-R1
antagonists. Here, electron-withdrawing groups seemed to be pre-
ferred over electron-donating substituents, probably reflecting the
presence of electron rich tyrosines in the binding pocket. Specifi-
cally, the trifluoromethyl and trifluoromethyloxy groups provided
the most potent, functional MCH-R1 antagonists (11: IC50 = 29 nM,
GTPcS IC50 = 32 nM and 12, IC50 = 31 nM, GTPcS IC50 = 15 nM,
respectively). Branching on the pseudobenzylic position was toler-
ated, and afforded a potent MCH-R1 antagonist (14, IC50 = 43 nM)
but no improvement in functional MCH-R1 antagonism (14, GTPcS
IC50 = 117 nM). Additionally, increasing the steric bulk on the in-
dole ring with a 1,5-dimethyl substitution resulted in a significant
loss of MCH-R1 affinity (15, IC50 = 9.65 lM), thereby highlighting
specific spatial constraints. Modification of the electronic proper-
ties of the distal ring had a significant effect on potency, as shown
by the isoxazole (16, IC50 = 1.5 lM) and pyridine (17, IC50 = 48 nM)
derivatives. Replacement of pyrrole with the 1,4-disubstituted fur-
ans, 18–19, always translated in a potency loss. Similarly, the thi-
azole (20) and imidazole (21) counterparts failed to improve
MCH-R1 potency, possibly validating the predicted lipophilic char-
acter of the MCH-R1 binding pocket.
The central part of the molecule was then subjected to exten-
sive structural modifications in an effort to optimize MCH-R1
antagonism and to further explore the SAR around the present ser-
ies. The results of such campaign are outlined in Table 2. Com-
pounds 22–34 were prepared in a similar manner to Scheme 1,
according to synthetic procedures previously disclosed.12
Direct manipulation of the amide function resulted in a dimin-
ished MCH-R1 affinity: the removal of the carbonyl group (23,
IC50 = 636 nM) had a more evident effect than N-methylation (22,
IC50 = 109 nM). This underlines the hypothesis of the involvement
of the amide function in polar interactions with MCH-R1 (Fig. 2).
Scaffold hopping to pyrrolidine (25) and azetidine (26) yielded po-
tent MCH-R1 antagonists but no improvement in GTPcS IC50, when
compared to 11 or 12. Fascinatingly, despite the obvious 2D and 3D
similarities, introduction of 3-Aza-bicyclo[3.1.0]hex-6-ylamide
(27) displayed a marked reduction in activity, which cannot simply
be explained on the basis of ring flexibility alone. Modification of
the interatomic distance between the amide and amine functions
with 1,3-disubstituted cyclopentane (28) and cyclohexane (29) or
when the 4-substituted-piperidine was flipped (30) did not im-
prove MCH-R1 potency, as shown in Table 2. This is supported
by the results obtained via one carbon chain homologation (31–
34), affording IC50 values in the 169–313 nM range. Variation of
the central scaffold indicated that there seem to be specific geo-
metric constraints around the amide and amine groups of the pres-
ent lead series. This might be due to the occurrence of electrostatic
interactions with MCH-R1, which could be impaired by even the
subtlest structural modifications, as exemplified by the data pre-
sented in Table 1. According to the emerging SAR, we concluded
that a hydrogen bond acceptor (HBA) was beneficial to MCH-R1
potency, and we set to investigate alternative HBA functionalities,
in the form of heterocycle fragments. The purpose was to retain the
key HBA pharmacophore while reducing molecular flexibility, in
order to lock the bioactive conformation of the compounds and im-
prove their functional MCH-R1 antagonism. Compounds 35–40
were synthesized starting from commercially available building
blocks, according to Scheme 2. Their MCH-R1 binding properties
are outlined in Table 3.
Introduction of 5-substituted-imidazole, as a simple amide
group bioisoster, afforded a compound (35) with good MCH-R1 po-
tency (IC50 = 20 nM) but average functional activity (GTPcS IC50:
122 nM). This indicates that while incorporation of a HBA feature
via ring closure is tolerated in the binding pocket, there is no direct
gain in efficacy. Interestingly, the 2-substituted-imidazole isomer
36 resulted in approximately sixfold potency loss, suggesting that
N
O
O
O
HN
1. CDI in THF, rt, 10m
NH2NH2 in EtOH, rt, 18h
N
N
Boc
Boc
HO
O
2. LDA
in THF
N
N
Boc
F
Boc
F
35%
64%
N
-78 °C to rt 2h
F
1. NH2OH in DCM, rt
O
N
O
2. TEA, MsCl in DCM
N
Boc
0 °C, 1h
+
F
F
30%
30%
NH
NH2
O
O
N
1. Oxalyl Chloride, DMF
DCM, rt, 2.5h
Cl
N
Cbz
F
N
H
HO
2. TMSCH2N2, Et2O, 0 °C
DIEA in THF/DMF
MW, 130 ˚C, 12m
N
N
Cbz
Cbz
F
58%
66%
O
O
1.
2.
DCM, 40 °C, 18h
Br
NCO
DCM, rt, 1h
CF3
N
N
CF3
N
N
N
3. NaOMe/MeOH, 1h
4. TFA, 5m
H2N
N
H
O
40, 25%
Scheme 2.

S. Berglund et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4859–4863
4863
Table 3
In vitro MCH-R1 binding and functional data for 35–40
F
R
R
F
N
N
F
F
N
N
F
F
a
b
a
b
Compound
R
MCH-R1 IC50 (GTP
0.020 (0.122)
c
S IC50
)
(l
M)
Compound
R
MCH-R1 IC50 (GTP
c
S IC50
)
(l
M)
O
N
N
N
N
O
F
F
35
38
0.025 (0.067)
0.066
N
H
*
*
*
*
N
F
F
F
36
0.113
39
40
N
H
H
H
O
N
N
37
0.037 (0.058)
0.026 (0.071)
N
*
*
a
Values are mean of at least two experiments. Compounds competed with 125I-MCH for binding at the human MCH1 receptor (h-MCH-R1) expressed in the HEK293 cell
line.
b
Values are mean of at least two experiments. Functional assay measuring [35S]GTP
c
S accumulation.
tent MCH-R1 antagonists. Modification of the pyrrole and phenyl
rings, as well as the central amine scaffold, indicated stringent
electronic constraints in the MCH-R1 pocket. As a result of our
structural exploration, 11 and 12 were shown to be functional
MCH-R1 antagonists with favorable pharmacokinetic properties
but intolerable cardiovascular safety margins. The subsequent
hERG optimization efforts will be the subject of future publications
from our group.
Table 4
Summary in vitro and in vivo data for 11 and 12
11
12
a
MCH-R1 GTP
hERG IC50
Fu (% free)
c
lM)
S IC50
(
lM)
0.032
0.56
1.2
38
41
6
24
11
10
100
0.015
0.49
3.2
21
34
8.3
20
1.2
14
100
b
(
HLM Clint
(
l
L/min/mg)
Mouse Clint (iv)c (mL/min/kg)
Mouse T1/2 (iv)c (h)
Mouse Vss (iv)c (L/kg)
c
Mouse AUC (po)
(
lM h)
Acknowledgments
c
Mouse T1/2 (po) (h)
Mouse bioavailabilityc (%)
The authors thank their AstraZeneca R&D Mölndal colleagues
Mia Boethius-Litzén for providing MCH-R1 binding and func-
tional data as well as Lars-Olof Larsson and his team for provid-
ing the DMPK data. Clas Landersjö is greatly acknowledged for
providing the structure elucidation by NMR of the isomeric
isoxazoles.
a
Functional assay measuring [35S]GTP
Values are mean of at least two experiments. Patch clamp assay using ION-
c
S accumulation.
b
WORKSTM technology in hERG-expressing CHO cells.
c
Determined following oral administration of a 20
in C57BL female mice.
lM/kg nanosuspension dose
References and notes
the hydrogen bond interaction is indeed very susceptible to dis-
tance and directionality. This is supported by the data for the pyr-
azole (37) and isoxazole (38) derivatives, where the HBA atom is
located in exactly the same position as in 35. As a result, they dis-
played similar potency (IC50 = 37 and 25 nM, respectively) when
compared to 35, as shown in Table 3. Finally, cyclization of the
amide functionality to provide the imidazol-2-one core resulted
in a potent MCH-R1 antagonist (40, IC50 = 26 nM) but without
1. Melnikova, I.; Wages, D. Nat. Drug Discov. 2006, 5, 369.
2. Gomori, A.; Ishiara, A.; Ito, M.; Mashiko, S.; Matsuhita, H.; Yumoto, M.; Ito, M.;
Tanaka, T.; Tokita, S.; Moriya, M.; Iwaasa, H.; Kanatani, A. Am. J. Physiol.
Endocrinol. Metab. 2003, 284, E583.
3. Ludwig, D. S.; Tritos, N. A.; Mastaitis, J. W.; Kulkarni, R.; Kokkotou, E.; Elmiqst,
J.; Lowell, B.; Flier, J. S.; Maratos-Flier, E. J. Clin. Invest. 2001, 107, 379.
4. Shimada, M.; Tritos, N. A.; Lowell, B. A.; Flier, J. S.; Maratos-Flier, E. Nature 1998,
396, 670.
5. Rivera, G.; Bocanegra-Garcia, V.; Galiano, S.; Cirauqui, N.; Ceras, J.; Perez, S.;
Aldana, I.; Monge, A. Curr. Med. Chem. 2008, 15, 1025.
the desired improvement in functional activity (GTP
cS
IC50 = 71 nM).
6. Kowalski, T. J.; Sasikumar, T. Biodrugs 2007, 21, 311.
7. Rokosz, L. L. Exp. Opin. Drug Discov. 2007, 2, 1301.
11 and 12 emerged as the most efficacious MCH-R1 antagonists
in this study, displaying GTP S IC50 values of 32 and 15 nM, respec-
8. Souers, A. J.; Gao, J.; Brune, M.; Bush, E.; Wodka, D.; Vasudevan, A.; Judd, A. S.;
Mulhern, M.; Brodjian, S.; Dayton, B.; Shapiro, R.; Hernandez, L. E.; Marsh, K. C.;
Sham, H. L.; Collins, C. A.; Kym, P. R. J. Med. Chem. 2005, 48, 1318.
9. Takekawa, S.; Asami, A.; Ishihara, Y.; Terauchi, J.; Kato, K.; Shimomura, Y.; Mori,
M.; Murakoshi, H.; Kato, K.; Suzuki, N.; Nishimura, O.; Fujino, M. Eur. J.
Pharmacol. 2002, 438, 129.
10. Giordanetto, F.; Lindberg, J.; Karlsson, O.; Inghardt, T. Abstract of Papers, 232nd
National Meeting of the American Chemical Society, San Francisco, CA, 2006,
Abstract 0410.
11. Cavasotto, C. N.; Orry, A. J. W.; Murgolo, N. J.; Czarniecki, M. F.; Kocsi, S. A.;
Hawes, B. E.; O’Neill, K. A.; Hine, H.; Burton, M. S.; Voigt, J. H.; Abagyan, R. A.;
Bayne, M. L.; Monsma, M. J., Jr. J. Med. Chem. 2008, 51, 581.
12. The preparation of key compounds is described in: (a) Brickmann, K.; Egner, B.
J.; Giordanetto F.; Inghardt, T.; Linusson, A.; Ponten, F. WO2005/090330
(Compounds 6–12, 17–18, 22, 25–26) and in: (b) Egner, B. J.; Giordanetto, F.;
Inghardt, T. WO2006/0685594 (Compounds 35–39).
c
tively, as shown in Table 4. Moreover, the two compounds demon-
strated acceptable free fractions in human plasma and appeared to
be metabolically stable when tested in vitro in Human Liver Micro-
somes (HLM) preparations, as summarized in Table 4. Both 11 and
12 achieved high bioavailability following intravenous (iv) and oral
(po) administration of a 20 lM/kg dose in mice (Table 4). Unfortu-
nately, both derivatives were also found to be potent inhibitors of
the hERG potassium channel, as outlined in Table 4, and this has
represented a major hurdle for their subsequent development.
In summary, the optimization of the HTS hit 1 resulted in the
discovery of 1-phenyl-1H-pyrrol-3-yl-derivatives as novel and po-