Bis(aminopyrrolidine)-derived ureas (APUs) as potent MCH1 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2004.12.036

Ureas derived from two substituted 3-aminopyrrolidine subunits were prepared as constrained analogs of a linear lead compound and tested as antagonists of the MCH₁ receptor. The series was optimized for substitution and stereochemistry to generate a functional antagonist with a K_i of 3.3 nM and IC₅₀ of 12 nM (GTPγS).

Full text of DOI:10.1016/j.bmcl.2004.12.036

Bioorganic & Medicinal Chemistry Letters 15 (2005) 999–1004
Bis(aminopyrrolidine)-derived ureas (APUs) as potent
MCH1 receptor antagonists
Jonathan Grey, Brian Dyck,* Martin W. Rowbottom, Junko Tamiya,
Troy D. Vickers, Mingzhu Zhang, Liren Zhao, Christopher E. Heise,
David Schwarz, John Saunders and Val S. Goodfellow
Departments of Medicinal Chemistry, Pharmacology and Molecular Biology, Neurocrine Biosciences Inc.,
12790El Camino Real, San Diego, CA 92130, USA
Received 12 November 2004; revised 10 December 2004; accepted 14 December 2004
Available online 13 January 2005
Abstract—Ureas derived from two substituted 3-aminopyrrolidine subunits were prepared as constrained analogs of a linear lead
compound and tested as antagonists of the MCH1 receptor. The series was optimized for substitution and stereochemistry to gen-
erate a functional antagonist with a Ki of 3.3 nM and IC50 of 12 nM (GTPcS).
Ó 2004 Elsevier Ltd. All rights reserved.
In mammals, melanin-concentrating hormone (MCH) is
believed to be a key neuropeptide responsible for bal-
ancing energy intake and expenditure.1–4 This equili-
brium is modulated by interaction of the hormone with
one or more G-protein coupled receptors. In rodents, a
single receptor (MCH1) has been identified, and homo-
zygous knockouts of this receptor have been gener-
ated.5,6 These animals demonstrate a lean phenotype
characterized by hyperphagia and increased energy
expenditure. Acute and chronic infusion of MCH into
the brains of wild type rats leads to increased feeding7
and body weight,8 respectively. Non-peptide MCH1
receptor antagonists have demonstrated efficacy both
in acute9 and chronic feeding models.10 In humans, at
least two MCH receptors are present. Of the two,
MCH1 shares greater homology with the rodent recep-
tors, and is more widely distributed in the human
CNS.11 This evidence is suggestive of a role for a selec-
tive MCH1 antagonist in the treatment of human obes-
ity, an epidemic of the developed world. There is also an
CH3
N
O
CH3
N
H
T-226296 (1)
F
O
CH3
O
H3C
N
N
R
Ar
N
N
N
CH3
CH3
N
H3C
N
O
O
Ph
3
2
Keywords: Melanin-concentrating hormone; Obesity; Anxiety; Depression.
*
Corresponding author. Tel.: +1 858 617 7778; fax: +1 858 617 7619; e-mail: bdyck@neurocrine.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.036

1000
J. Grey et al. / Bioorg. Med. Chem. Lett. 15 (2005) 999–1004
O
H3C
H3C
H3C
HN
a
d
N
Bn
H3C
Bn
N
R
N
Bn
N
N
N
N
N
N
Boc
H3C
O
e
O
b
9
6
4
H3C
NH
O
N
H3C
N
Boc
H3C
R'
N
5
R
N
N
H3C
N
N
N
N
Boc
R''
O
H3C
O
S
10
O
R'
7
O
c
H3C
*3'
O
H3C
R'
N
R2
N
N
R1
N
R
N
N
*3
N
H3C
R''
O
11
O
8
Scheme 1. Reagents and conditions: (a) (i) 4-nitrophenyl chlorocarbonate, THF; (ii) 5, DMF, 100 °C (54–74%, two steps); (b) (i) ammonium
formate, 10% Pd on carbon, ethanol, reflux; (ii) aldehyde or ketone, NaBH3CN, methanol (21–86%, two steps); (c) (i) TFA, DCM; (ii) RCOCl, TEA,
DCM or RCO2H, HOBt, EDC, DCM (44–90%, two steps); (d) (i) TFA, DCM; (ii) RCOCl, TEA, DCM or RCO2H, HOBt, EDC, DCM (90–100%,
two steps); (e) (i) ammonium formate, 10% Pd on carbon, ethanol, reflux; (ii) vinyl sulfone or sulfonamide, ACN, 60 °C (12–27%, two steps).
increasing body of evidence suggesting that an MCH1
antagonist may have application in the treatment of
anxiety and/or depression.4,10
transfer hydrogenation and reductive alkylation with a
ketone or an aldehyde provided 7. Removal of the Boc
protecting group was effected by treatment with trifluo-
roacetic acid, and acylation of the resulting exocyclic
amine group afforded the desired APU compounds 8.
Alternatively, for some examples it was preferable to re-
move the Boc protecting group and acylate to give com-
pound 9. At this stage, debenzylation and alkylation
with various electrophiles, including a,b-unsaturated
sulfones and sulfonamides, afforded compounds 10.
Tetrahydronaphthalene-derived T-226296 (1) is a potent
MCH1 receptor antagonist, which, when given orally,
blocks the induced hyperphagia brought on by MCH
injections into the lateral cerebral ventricle of rats.9 Dur-
ing our investigation of the pharmacophore of MCH1
antagonists related to 1, we discovered a related series
of compounds based on a c-aminobutyramide (GABA)
core. When the GABA scaffold is substituted with 3-
aminopyrrolidine and biphenyl carboxamide groups,
compounds with sub-micromolar binding affinities re-
sult (i.e., 2).12 Further optimization of the substitution
on the pyrrolidine nitrogen atom produces compounds
with binding constants below 10 nM. However, the flex-
ible nature of the GABA molecules was a concern, both
because the multitude of conformations available would
likely cause selectivity issues, and also due to the nega-
tive impact of a large number of rotable bonds on phar-
macokinetic parameters.13 Therefore, in an attempt to
develop a potent and selective MCH1 antagonist, we
sought to constrain the GABA-based lead compound
2 into a urea-linked compound containing two amino-
pyrrolidine units (i.e., 3). Introduction of a second
aminopyrrolidine unit offered additional opportunity
to adjust the selectivity of the compounds, whilst relying
on the same source of optically pure starting materials.
Preliminary experiments with the APU core indicated
that the (3S,30R)-stereochemistry was likely to produce
the most potent examples (cf. annotation in 11). When
the (3S,30R)-APU nucleus was substituted similarly to
lead compound 2, there was a modest increase in po-
tency (12). This is consistent with previous observations
that a restriction in the number of possible conforma-
tions of a ligand results in a concomitant entropic en-
ergy gain leading to an increase in binding potency.14
In the GABA series, replacement of the N-methyl substi-
tuent of the pyrrolidine ring with more lipophilic groups
enabled a 20-fold improvement in binding potency.12
Hence, an attempt was made to impart a similar
improvement in the affinity of the APU compounds by
the analogous modifications. Conversion of the N-
methyl group to hydrogen (13) or larger lipophilic (14)
or hydrophilic (15) groups had little effect on binding
potency. An exception was the introduction of lipophiles
attached via a three-atom spacer, which inevitably led to
potent compounds (16 and 17), one exhibiting a Ki of
7.3 nM. Attempts to modify the lipophile in order to
lower the logD and attenuate the basicity of the pyrrol-
idine nitrogen atom, resulted in compounds that were
2- to 10-fold less potent than 16 (18–20). Dimethyl-
cyclohexyl derivative 21 possesses lower affinity than
16, but the compound was of interest due to the in-
The synthesis of the bis(aminopyrrolidine)ureas (APUs)
was performed as shown in Scheme 1. Benzyl-protected
aminopyrrolidine 4 was converted to the p-nitrophenyl
carbamate by treatment with the corresponding chloro-
formate. The carbamate was then heated with 3-(N-Boc-
N-methylamino)pyrrolidine (5) to produce orthogonally
protected 6. Removal of the benzyl protecting group by

J. Grey et al. / Bioorg. Med. Chem. Lett. 15 (2005) 999–1004
Table 1. Binding affinities of 1, 2, and APU compounds 11 at the MCH1 receptor15
1001
Compda
R1b
R2b
Ki SD (nM)c
1d
2e
N.A.
N.A.
N.A.
N.A.
5.5
142 54
O
CF3
CF3
12
13
Me
H
57 14
N A
H3C
O
64 62
N A
H3C
O
CF3
CF3
14
15
n-Butyl
64 38
N A
H3C
O
OH
96
6
2
A
A
N A
H3C
O
CF3
CF3
CF3
CF3
CF3
16
17
18
19
20
7.3
N A
Cl
H3C
H3C
H3C
H3C
H3C
O
A
A
20 10
N A
O
O
18
14
82
9
6
6
N A
Cl
Cl
O
O
O
S
S
A
N A
O
O
O
A
A
N
N A
O
CH3
CH3
CF3
21
31 15
N A
H3C
O
CF3
CF3
22
23
75 23
A
O
N
A
H3C
O
83
9
A
A
N CH3
N A
H3C
O
O
Cl
24
25
48 11
N A
H3C
O
A
H3C
H3C
H3C
67 67
N
A
O
Cl
H3C
(continued on next page)

1002
J. Grey et al. / Bioorg. Med. Chem. Lett. 15 (2005) 999–1004
Table 1 (continued)
Compda
R1b
R2b
Ki SD (nM)c
O
A
F
A
26
5.3
3
N
Cl
Cl
CH3
O
O
CH3
A
N
27
28
22 20
N
H
A
O
CH3
CH3
CH3O
CF3
A
A
1043 56
N A
H3C
N N
O
CH3
CH3
29
30
3.9
1
N A
N
H3C
O
N
N
CH3
CH3
A
A
1096 584
N
N
H
A
CF3
CF3
H3C
O
CH3
CH3
31
456 288
S
N
A
H3C
H3C
O
CH3
CH3
S
A
A
32
1.8 0.2
N
A
CF3
CF3
O
33f
1019 211
N
A
a (3S,30R)-configuration unless otherwise specified.
b A: site of attachment to APU core.
c Values are averaged from at least two experiments.
d Ref. 9.
e Racemic.
f (3S,30S)-configuration.
creased rigidity. More hydrophilic analogs of 21 showed
a similar drop in potency (22 and 23), as was observed
when similar changes were made to 16 (Table 1).
O
O
CH3
O
a
OH
CH3
Br
Br
Considerable effort was also invested in modifying the
biphenyl carboxamide group of the APU compounds.
The replacement of the internal phenyl ring with other
linkers led to less potent analogs (24 and 25). However,
an ether linkage between the phenyl rings produced 26,
which was comparably potent to 16. Introduction of a
urea functional group to replace the carboxamide group
led to a threefold drop in potency (27). Attempts to
introduce nitrogen-containing heterocycles to replace
either of the phenyl rings of the biphenyl moiety typi-
cally resulted in a substantial drop in potency (28, 30,
and 31), although the pyridyl derivative 29 exhibited a
Ki of 3.9 nM. The introduction of a thiophene replace-
ment of the internal phenyl ring produced 32, which is
the most potent compound of the series with a Ki of
1.8 nM. Thiophene 32 was also shown to be a functional
antagonist with an IC50 of 12 8 nM (n = 11) in an
MCH1-derived GTPcS binding inhibition assay.16 Preli-
35
Br
34
H3C
b
R2
N
N
N
N
H2N
O
36
H3C
O
N
R2
N
N
O
37: R = Br
38: R = 4-CF3C6H4
c
R
Scheme 2. Reagents and conditions: (a) (i) trimethylsilyldiazomethane,
DCM (100%); (ii) NBS, benzoyl peroxide, carbon tetrachloride, reflux
(30%); (b) DIEA, methanol (45%); (c) 4-trifluoromethylbenzene-
boronic acid, tetrakis(triphenylphosphine)palladium, 2 M sodium
carbonate, toluene, ethanol, 100 °C, sealed tube (20%).

J. Grey et al. / Bioorg. Med. Chem. Lett. 15 (2005) 999–1004
Table 2. Effect of stereochemistry of APU core on MCH1 receptor binding15
1003
Compd
R1a
R2a
Stereochemistry
Ki SD (nM)b
O
CH3
CH3
S
S
32
A
(3S,30R)
1.8 0.2
N
A
A
CF3
H3C
H3C
O
CH3
CH3
(3R,30R)
(3S,30S)
(3R,30S)
338 15
A
A
A
39
40
41
N
CF3
CF3
CF3
O
CH3
CH3
S
S
27 22
N
A
H3C
O
CH3
CH3
1125 784
N
A
H3C
a A: site of attachment to APU core.
b Values are averaged from at least two experiments.
minary screening showed this compound to have a Ki
value for the human MCH2 receptor greater than
10 lM.
Studies to determine the pharmacokinetics and activity
of 32 and related compounds in animal models of obes-
ity, anxiety, and depression will be reported in due
course.
Finally, in an attempt to remove the metabolically labile
amide N-methyl group, a cyclized analog was prepared
as shown in Scheme 2. Compound 36 was prepared in
an analogous fashion to its methyl counterpart 7 of
Scheme 1. This material was cyclized with the benzylic
bromide 35 to afford 37, which was subjected to Suzuki
coupling conditions with 4-(trifluoromethyl)benzenebo-
ronic acid to give 38. Unfortunately, compounds of this
class (i.e., 33) were typically more than 100-fold less
potent than their acyclic analogs.
References and notes
1. Forray, C. Curr. Opin. Pharmacol. 2003, 3, 85.
2. Pissios, P.; Maratos-Flier, E. Trends Endocrinol. Metabol.
2003, 14, 243.
3. Collins, C. A.; Kym, P. R. Curr. Opin. Invest. Drugs 2003,
4, 386.
4. Hervieu, G. Expert Opin. Ther. Targets 2003, 7, 495.
5. Marsh, D. J.; Weingarth, D. T.; Novi, D. E.; Chen, H. Y.;
Trumbauer, M. E.; Chen, A. S.; Guan, X.-M.; Jiang, M.
M.; Feng, Y.; Camacho, R. E.; Shen, Z.; Frazier, E. G.;
Yu, H.; Metzger, J. M.; Kuca, S. J.; Shearman, L. P.;
Gopal-Truter, S.; MacNeil, D. J.; Strack, A. M.; MacIn-
tyre, D. E.; Van der Ploeg, L. H. T.; Qian, S. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99, 3240.
6. Chen, Y.; Hu, C.; Hsu, C.-K.; Zhang, Q.; Bi, C.; Asnicar,
M.; Hsiung, H. M.; Fox, N.; Slieker, L. J.; Yang, D. D.;
Heinman, M. L.; Shi, Y. Endocrinology 2002, 143, 2469.
7. Qu, D.; Ludwig, D. S.; Gammeltoft, S.; Piper, M.;
Pelleymounter, M. A.; Cullen, M. J.; Mathes, W. F.;
Przypek, J.; Kanarek, R.; Maratos-Flier, E. Nature 1996,
380, 243.
8. Della-Zuana, O.; Presse, F.; Ortola, C.; Duhault, J.;
Nahon, J. L.; Levens, N. Int. J. Obesity 2002, 26, 1289.
9. Takekawa, S.; Asami, A.; Ishihara, Y.; Terauchi, J.; Kato,
K.; Shimomura, Y.; Mori, M.; Murakoshi, H.; Kato, K.;
Susuki, N.; Nishimura, O.; Fujino, M. Eur. J. Pharmacol.
2002, 438, 129.
10. Borowsky, B.; Durkin, M. M.; Ogozalek, K.; Marzabadi,
M. R.; DeLeon, J.; Heurich, R.; Lichtblau, H.; Shaposh-
nik, Z.; Daniewska, I.; Blackburn, T. P.; Branchek, T. A.;
Gerald, C.; Vaysse, P. J.; Forray, C. Nat. Med. 2002, 8, 825.
11. Schlumberger, S. E.; Talke-Messerer, C.; Zumsteg, U.;
Eberle, A. N. J. Recept. Signal Transduc. Res. 2002, 22,
509.
In order to verify that the (3S,30R)-stereochemistry was
optimal for the current series of APU compounds, we
synthesized the three stereoisomers of 32 using modifica-
tions of the chemistry described in Scheme 1. The bind-
ing affinities of these compounds are shown in Table 2.
The (3R)-epimer 39 was 100-fold less potent than 32,
and the (30S)-epimer 40 was 15-fold less potent. This
confirms that the stereochemistry of the pyrrolidine ring
residing closer to the carboxamide group is more critical
for high affinity binding than is the analogous center on
the right hand side of the molecule. Compound 41, the
enantiomer of 32, is more than 600-fold less potent. This
suggests that the decreased potency resulting from
inverting the stereocenters of 32 is additive.
In this study, we discovered a new class of MCH1 recep-
tor ligands containing a urea moiety derived from two
substituted 3-aminopyrrolidine subunits. With this new
class of ligands, the MCH1 receptor shows a strong ste-
reochemical preference and an affinity for lipophilic sub-
stituents. Our attempts to circumvent the latter met with
limited success; however, a potent, functional antagonist
was discovered, which prompted its evaluation in vivo.

1004
J. Grey et al. / Bioorg. Med. Chem. Lett. 15 (2005) 999–1004
12. Details of the SAR of the GABA-based MCH1 receptor
antagonists will be reported elsewhere.
13. Veber, D. F.; Johnson, S. R.; Cheng, H.-Y.; Smith, B. R.;
Ward, K. W.; Kopple, K. D. J. Med. Chem. 2002, 45, 2615.
14. Bohm, H.-J.; Klebe, G. Angew. Chem., Int. Ed. Engl. 1996,
35, 2588.
15. Measuring displacement of [
125I]MCH from human
MCH1 receptor modified for optimal expression in
HEK293 cells.
16. Measuring inhibition of MCH-stimulated G-protein/
[35S]GTPcS binding in CHO cells expressing human
MCH1 receptor.