Synthesis and biological activity of novel peptide mimetics as melanocortin receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2007.11.109

A series of novel peptidomimetic analogs was prepared containing cyclohexyl, phenyl, or heterocyclic groups to ostensibly orient the guanidine or mimic of an arginine in a putative melanocortin receptor ligand pharmacophore. Some binding affinity at the melanocortin receptors MC₃ and MC₄ was noted. In silico docking also indicated that the relative positions of the hydrogen-bonding sites and hydrophobic regions of the compounds are reasonably well matched to the receptor-binding site. This may present a lead entry into a selective series of MC₄R agonists.

Full text of DOI:10.1016/j.bmcl.2007.11.109

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1223–1228
Synthesis and biological activity of novel peptide mimetics
as melanocortin receptor agonists
Xue-Wei Liu,a,* Jimei Ma,a Anny-Odile Colson,b Doreen Cross Doersenb
and Frank H. Ebetinob
aDivision of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences,
Nanyang Technological University, Singapore 637371, Singapore
bProcter & Gamble Pharmaceuticals, Health Care Research Center, 8700 Mason Montgomery Road, Mason, OH 45040, USA
Received 16 October 2007; revised 25 November 2007; accepted 28 November 2007
Available online 4 December 2007
Abstract—A series of novel peptidomimetic analogs was prepared containing cyclohexyl, phenyl, or heterocyclic groups to osten-
sibly orient the guanidine or mimic of an arginine in a putative melanocortin receptor ligand pharmacophore. Some binding affinity
at the melanocortin receptors MC3 and MC4 was noted. In silico docking also indicated that the relative positions of the hydrogen-
bonding sites and hydrophobic regions of the compounds are reasonably well matched to the receptor-binding site. This may present
a lead entry into a selective series of MC4R agonists.
Ó 2007 Elsevier Ltd. All rights reserved.
Melanocortin receptors (MCR) are members of the G-
protein coupled receptor subfamily consisting of
ported to increase the affinities for the MC receptors.9,10
Also, replacement of the His residue was shown to in-
crease selectivity for the MC4 receptor.11,12 It is only re-
cently that subtype selective, orally active, small
(MW < 500) ligands were reported. Workers at Merck
synthesized a new class of cyclohexyl substituted piperi-
dines that presumably mimicked this HFRW sequence
(Fig. 1).13 Compound A is a potent (EC50 = 2.1 nM),
selective (1184-fold vs. MC3R, 350-fold vs. MC5R),
small-molecule full agonist of the human MC4 receptor.
In an attempt to better understand this finding and re-
lated theories on the minimum pharmacophore required
for MC4 activity, we designed a related series of cyclo-
hexyl-, phenyl-, and heterocyclic-containing linear com-
pounds. These compounds representing the piperidine
ring-opened analogs of compound A were thus designed
to test the criticality of the conformational restriction
aspects of this drug lead. In this letter, we report the syn-
thesis and biological activity of novel analogs based on
this concept.
1,2
MC1,2,3,4,5
.
The MC1 receptor has been found in mel-
anocytes where it controls skin pigmentation while re-
cent studies indicate that MC4R is involved in the
control of feeding behavior. As a result, the MC4 recep-
tor has received increased attention as a receptor target
from both academia and industry. Since this subtype is
localized in the central nervous system, transport of
exogenous ligands across the blood-brain barrier re-
mains one of the challenges in this field.3,4
Melanocyte stimulating hormone, a-MSH is a natural li-
gand for four of the melanocortin receptor subtypes,
namely, the MC1, MC3, MC4, and MC5 receptors. This
peptide hormone is 13 amino acids in length and binds
with high affinity to the MC1, MC3, MC4, and MC5
receptors. Other ligands designed earlier shared a com-
mon core sequence motif: His-Phe-Arg-Trp.5 A number
of linear and cyclic peptide and peptidomimetics con-
taining this sequence have now been reported. Most
show a similar selectivity profile to the natural MSH.6–8
Interestingly, replacement of L-Phe with D-Phe was re-
The novel cyclohexyl-directed triazole analog was syn-
thesized according to the route presented in Scheme 1.
The commercially available cyclohexyl methyl ketone 1
was brominated with benzyltrimethylammonium tribro-
mide yielding a-bromo ketone 2 in high yield (98%).
Overnight reaction of 2 with sodium triazole in DMF
at 50 °C yielded the a-triazole ketone 3 (91%) which,
Keywords: Melanocortin receptor; Peptidomimetics; MC4R agonist;
Homology model.
*
Corresponding author. Tel.: +65 6316 8901; fax: +65 6791 1961;
e-mail: xuewei@ntu.edu.sg
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.109

1224
X.-W. Liu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1223–1228
guanidine
mimic
N
N
HN
HN
N
O
RHN
O
N
N
N
N
N
O
cyclohexyl
phenyl
heterocyclic
X
halogen
R=H
B
Merck
A
Cl
Figure 1. Constrained (A) and linear (B) conformations of the ligand.
N
N
N
N
Br
N
N
a
c
b
O
O
O
N
1
2
3
4
N
N
N
N
N
O
N
d
e,f
H2N
N
H
H2N
5
6
F
Scheme 1. Reagents and conditions: (a) PhCH2N(CH3)3Br3, CH2Cl2, 5 °C–rt, 4 h, 98%; (b) Sodium triazole, DMF, 5 °C, 91%; (c) diethyl
phosphonoacetonitrile, LiCl, DBU, CH3CN, rt, 90%; (d) H2, Ni, NH4OH, MeOH, 40 psi, 4 h, 98%; (e) Boc-D-Phe(p-F)-OH, EDC, HOBt, NMM,
DMF, rt, overnight, 84%; (f) TFA, CH2Cl2, 97%.
via a facile Horner–Wadsworth–Emmons reaction with
diethyl phosphonoacetonitrile in the presence of LiCl
and DBU, provided a-, b-unsaturated nitrile 4 in 90%
yield. The primary amine 5 was obtained in almost
quantitative yield by hydrogenation of 4 on a Parr sha-
ker at 40 psi. This amine was coupled with Boc-D-Phe(p-
F)-OH in the presence of HOBt, EDC, and NMM. The
crude product after reaction workup was treated with
trifluoroacetic acid and purified by reverse-phase HPLC
to generate the final compound 6 as trifluoroacetate salt
in a yield of 81% for two steps.
Phenyl analog 16 of cyclohexyl compound 12 was pre-
pared as shown in Scheme 3 starting from a-, b-unsatu-
rated nitrile 8. Catalyzed by Raney nickel, it was
reduced to saturated amine 13 with the phenyl ring in-
tact. The final target 16 was obtained from primary
amine 13 via 14, and 15, following the procedures used
to prepare 12.
Schemes 4 and 5 outline the methodology for preparing
dipeptides D-Phe(p-F)-Arg with terminal modifications.
Boc-D-Phe(p-F)-OH 17 was coupled with the methyl ester
of Arg 18 under typical EDC/HOBt coupling conditions,
resulting in dipeptide 19. The terminal methyl ester group
was converted to the corresponding acid 20 with LiOH in
THF and water (3:1) followed by acidification. EDC/
HOBt coupling of the acid 20 and piperidine gave 21,
which was transformed to dipeptide 22 after removal of
the Boc group with TFA in CH2CH2 and hydrogenolytic
removal of the nitro group in structure 21.
Scheme 2 illustrates the synthesis of dipeptidomimetics
12 from the commercially available N-t-Boc-L-phenyl-
alanial 7. The aldehyde 7 was coupled to diethyl phos-
phonoacetonitrile via a Horner–Wadsworth–Emmons
reaction in 89% yield. The resulting a-, b-unsaturated ni-
trile 8 was reduced to the amine 9 (92%). Rh/alumina
was an optimal catalyst for this step to hydrogenate
the double bond nitrile and the phenyl ring in one step.
The free primary amine 9 was converted to the di-Cbz
protected guanidine derivative 10 by guanylation with
N,N-di-Cbz-S-methylpseudothiourea in the presence of
HgCl2 and Et3N. After deprotection with 50% TFA in
CH2Cl2, 10 was coupled to Boc-D-Phe(p-F)-OH to gen-
erate 11, followed by removal of Boc with 50% TFA in
CH2Cl2 and catalytic hydrogenation to remove the Cbz
group. The dipeptidomimetics 12 was obtained after
HPLC purification.
As shown in Scheme 5, methyl ester 19 was reduced to
alcohol 23 with LiBH4 in THF at room temperature in
a reasonable yield (77%). In the presence of DEAD
and Ph3P, this alcohol reacted with piperidine in a Mits-
unobu reaction in a moderate yield (41%), generating
dipeptidomimetics 24. Treatment of 24 with TFA in
CH2CH2 and hydrogenolysis of the nitro group, fol-
lowed by HPLC purification, provided the tertiary
amine analog 25.

X.-W. Liu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1223–1228
CN CH2NH2
1225
O
H
a
c
b
BocHN
BocHN
BocHN
7
8
9
CbzHN
NCbz
H2N
NH
NH
CbzHN
NCbz
NH
NH
f,g
d,e
O
O
R
BocHN
H2N
N
H
N
R
H
BocHN
F
F
10
11
12
Scheme 2. Reagents and conditions: (a) diethyl phosphonoacetonitrile, LiCl, DBU, CH3CN, rt, 89%; (b) H2, Rh/alumina, NH4OH, MeOH, 45 psi,
22 h, 92%; (c) 1,3-bis(benzyloxycarbonyl)-2-methyl-2-thiopseudourea, HgCl2, Et3N, DMF, 74%; (d) TFA, CH2Cl2, 98%; (e) Boc-D-Phe(p-F)-OH,
EDC, HOBt, NMM, DMF, rt, overnight, 80%; (f) TFA, CH2Cl2, 97%; (g) H2, Pd/C, MeOH; 99%.
CbzHN
NCbz
NH
CN
CH2NH2
a
b
BocHN
BocHN
BocHN
8
13
14
CbzHN
NCbz
NH
H2N
NH
NH
O
O
c,d
e,f
R
BocHN
H2N
N
H
N
R
H
F
15
16
F
Scheme 3. Reagents and conditions: (a) H2, Ni, NH4OH, MeOH, 40 psi, 4 h, 95%; (b) 1,3-bis(benzyloxycarbonyl)-2-methyl-2-thiopseudourea,
HgCl2, Et3N, DMF, 74%; (c) TFA, CH2Cl2, 98%; (d) Boc-D-Phe(p-F)-OH, EDC, HOBt, NMM, DMF, rt, overnight, 80%; (e) TFA, CH2Cl2, 97%; (f)
H2, Pd/C, MeOH; 99%.
The screening data obtained for the substances de-
scribed herein and shown in Table 1 contribute to the
understanding of the pharmacophoric requirements.
Compound 6 is a linear analog based on the Merck scaf-
fold A. Its weak binding affinity to MC4R suggests that
the conformational restriction of this drug lead is criti-
cal. Interestingly, when the triazole group is replaced
by a guanidine moiety, compounds 12 and 16 show in-
creased affinity to MC4R. This guanidine function might
therefore be forming stronger or more favorable interac-
tions with the receptor than the triazole moiety.
Replacement of the cyclohexyl group with a piperidine
functionality for compounds 22 and 25 led to poor affin-
ity analogs (Ki > 25 lM). In this case, a stereochemical
change (R ! S) was also made at the guanidinium side
chain. From this preliminary evidence, it appears that
the guanidine function with an R configuration and a
directing cyclohexyl or phenyl group are necessary in
addition to conformational restriction.
In an effort to understand the differences in the binding
affinity of various analogs, we constructed a homology
model of the MC4 receptor using the known human
MC4 receptor amino acid sequence15,16 and site-directed
mutagenesis studies.17,18 The computer model was cre-
ated by mapping the transmembrane sections of the hu-
man MC4 receptor from the primary amino acid
sequence onto the crystal structure of rhodopsin.19
The extracellular loops were constructed using a proto-
col described previously.20 The model was subsequently
optimized with molecular dynamics simulations and re-
fined using SAR data generated in our laboratory (data
not shown) as well as results from site-directed mutagen-
esis experiments. Individual ligands were then docked
manually and energy minimized within the putative
binding site of the receptor.21 Figure 2 shows compound
12 docked in the receptor and indicates specific interac-
tions between residues in the transmembrane (TM)
region of the receptor and the ligand. More specifically,

1226
X.-W. Liu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1223–1228
H2N
N
NO2
H2N
N
NO2
NH
O
NH
BocHN
OH
O
b
a
BocHN
OMe
+
N
OMe
H
H2N
O
F
F
O
17
18
19
H2N
N
H2N
N
H2N
NH
NO2
NO2
NH
NH
NH
O
O
O
R
c
d,e
BocHN
OH
BocHN
N
H2N
N
N
N
H
N
S
H
H
O
F
O
F
O
F
20
21
22
Scheme 4. Reagents and conditions: (a) EDC, HOBt, NMM, DMF, rt, overnight, 80%; (b) LiOH, THF, H2O; (c) EDC, HOBt, NMM, DMF, rt,
overnight, 71%; (d) H2, Pd/C, MeOH; 99%; (e) TFA, CH2Cl2, 97%.
Table 1. Binding affinity Ki (lM) and potency EC50 (lM) on MCR
H2N
N
H2N
N
subtypesa
NO2
NO2
NH
NH
Compound
MC1
EC50
MC3
MC4
Ki
Ki
EC50
Ki
EC50
O
O
a
BocHN
OMe
BocHN
OH
A
0.3
0.005
2.0
0.125
2.3
4.5
0.16
0.23
>50
0.003
0.023
>50
6.7
0.001
0.016
>25
N
N
H
H
B
>10
>10
—
O
F
6
>25
>25
—
12
16
22
25
9.1
3.5
>15
>50
F
19
23
>25
>100
>100
>100
>25
>100
4.2
>25
>100
>100
>50
>100
>50
>50
>100
>100
b
a The analogs were screened against the human MC1R, MC3R, and
MC4R.14 Data represent means of at least three experiments.
H2N
NH
NH
H2N
N
NO2
NH
O
O
compound 16 where the cyclohexyl ring is substituted
for a phenyl ring. The change in stereochemistry of the
guanidinium side chain for compounds 22 and 25 is det-
rimental to their binding which is most probably due to
the loss of interaction with Asp122 and Glu100. The
alternate location of the putative binding site proposed
by Wikberg et al.22,23 based on work performed in the
MC1 receptor may further explain the structure activity
results described here.
c,d
R
H2N
N
BocHN
N
N
H
N
H
S
F
F
25
24
Scheme 5. Reagents and conditions: (a) LiBH4, THF, rt, overnight,
77%; (b) piperidine, DEAD, Ph3P, THF, ꢀ78 °C, 41%; (c) H2, Pd/C,
MeOH; 99%; (e) TFA, CH2Cl2, 97%.
In summary, we have described the design, synthesis,
and modeling of a series of novel peptidomimetics with
low molecular weight (<500) via very practical synthetic
strategies. We utilized cyclohexyl, phenyl, and piperidine
rings as Trp mimics to direct the orientation of the gua-
nidine side chain or its replacements within the Arg
component of the HfRW pharmacophore. Docking
experiments within a receptor binding model offer a rea-
sonable understanding of the SAR for some members of
this series, although alternate binding modes should not
be excluded. It is likely that the increased conforma-
tional freedom of these compounds caused by the open-
ing of the piperidine ring is partly responsible for the
the model suggests that the guanidinium group of com-
pound 12 forms a salt bridge with the carboxylates of
Glu100 (TM2) and Asp122 (TM3), while the terminal
amine interacts with Asp126 (TM3). The hydrophobic
fluoro-phenyl is anchored deeper in the transmembrane
domain of the receptor and appears to reside in a hydro-
phobic pocket where it interacts with Trp258 (TM5),
Phe261 (TM6), and Phe262 (TM6). The cyclohexyl ring
occupies a hydrophobic region located at the putative
extracellular interface of TM7 and possibly interacts
with Met281 (TM7), Phe284 (TM7), and Leu288
(TM7). These interactions appear to be maintained in

X.-W. Liu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1223–1228
1227
Figure 2. Hydrophobic surface of the MC4R model with docked ligand 12. The hydrophobic regions are shown in brown/green, the hydrophilic
regions in blue/gray. The surface on portions of extracellular loop 1 has been removed for clarity.
12. Prusis, P.; Muceniece, R.; Mutule, I.; Mutulis, F.; Wik-
berg, J. E. S. Eur. J. Med. Chem. 2001, 36, 137.
lower binding affinity. Further optimization should be
pursued, perhaps including attempts to add new contact
residues within the MC4R pharmacophore.
13. Sebhat, I. K.; Martin, W. J.; Ye, Z. X.; Barakat, K.;
Mosley, R. T.; Johnston, D. B. R.; Bakshi, R.; Palucki, B.;
Weinberg, D. H.; MacNeil, T.; Kalyani, R. N.; Tang, R.;
Stearns, R. A.; Miller, R. R.; Tamvakopoulos, C.; Strack,
A. M.; McGowan, E.; Cashen, D. E.; Drisko, J. E.; Hom,
G. J.; Howard, A. D.; MacIntyre, D. E.; van der Ploeg, L.
H. T.; Patchett, A. A.; Nargund, R. P. J. Med. Chem.
2002, 45, 4589.
Acknowledgment
We thank Prof. Andrew S. Kende for his chemistry in-
put over the course of this work.
14. The agonist activity of the MCR ligands was evaluated at
three human MCR using a cell-based assay that is specific
for each subtype of MCR (MC1R, MC3R, MC4R). Each
class of the receptors was stably transfected into HEK293
cells. The MCR expressing cells were then stably trans-
fected with a reporter system consisting of a cyclic-AMP
responsive element (CRE) coupled to a luciferase reporter
gene. Agonist activity was determined by assaying cells in
a 96-well format for luciferase activity. Responses were
compared to the effect of NDP-MSH (MT-I) and
expressed as a percent of maximum activity of MT-1
(Emax). MT-I is considered to be a full agonist at each of
the three MCR subtypes. The binding activity of MCR
ligands was evaluated at three human MCR using a cell-
based assay that is specific for each subtype of MCR
(MC1R, MC3R, MC4R). Each class of the receptors was
stably transfected into the HEK293 cells. Binding affinity
(calculated as IC50 and Ki values) was determined in these
cell lines by measuring the displacement of a constant
concentration of Europium labeled NDP-a-MSH with
competing unlabeled ligands. Europium activity was
detected using time resolve fluorometry.
References and notes
1. Mountjoy, K. G.; Robbins, L. S.; Mortrud, M. T.; Cone,
R. D. Science 1992, 257, 1248.
2. Gantz, I.; Konda, Y.; Tashiro, T.; Shimoto, Y.; Miwa, H.;
Munzert, G.; Watson, S. J.; Delvalle, J.; Yamada, T.
J. Biol. Chem. 1993, 268, 8246.
3. Fan, W.; Boston, B. A.; Kesterson, R. A.; Hruby, V. J.;
Cone, R. D. Nature 1997, 385, 165–168.
4. Vergoni, A. V.; Bertolini, A. Eur. J. Pharmacol. 2000, 405,
25.
5. Hruby, V. J.; Wilkes, B. C.; Hadley, M. E.; Alobeidi, F.;
Sawyer, T. K.; Staples, D. J.; Devaux, A. E.; Dym, O.;
Castrucci, A. M. D.; Hintz, M. F.; Riehm, J. P.; Rao, K.
R. J. Med. Chem. 1987, 30, 2126.
6. HaskellLuevano, C.; Sawyer, T. K.; Hendrata, S.;
North, C.; Panahinia, L.; Stum, M.; Staples, D. J.;
Castrucci, A. M. D.; Hadley, M. E.; Hruby, V. J.
Peptides 1996, 17, 995.
7. Chen, M.; Georgeson, K. E.; Harmon, C. M.; Haskell-
Luevano, C.; Yang, Y. K. Peptides 2006, 27, 2836.
8. Xiang, Z. M.; Pogozheva, I. D.; Sorenson, N. B.;
Wilczynski, A. M.; Holder, J. R.; Litherland, S. A.;
Millard, W. J.; Mosberg, H. I.; Haskell-Luevano, C.
Biochemistry 2007, 46, 8273.
15. Gantz, I.; Miwa, H.; Konda, Y.; Shimoto, Y.; Tashiro, T.;
Watson, S. J.; Delvalle, J.; Yamada, T. J. Biol. Chem.
1993, 268, 15174.
16. Mountjoy, K. G.; Mortrud, M. T.; Low, M. J.; Simerly,
R. B.; Cone, R. D. Mol. Endocrinol. 1994, 8, 1298.
17. Haskell-Luevano, C.; Cone, R. D.; Monck, E. K.; Wan,
Y. P. Biochemistry 2001, 40, 6164.
9. Sugg, E. E.; Cody, W. L.; Abdelmalek, Z.; Hadley, M. E.;
Hruby, V. J. Biopolymers 1986, 25, 2029.
10. Haskellluevano, C.; Miwa, H.; Dickinson, C.; Hruby, V.
J.; Yamada, T.; Gantz, I. Biochem. Biophys. Res. Commun.
1994, 204, 1137.
18. Yang, Y.; Fong, T. M.; Dickinson, C. J.; Mao, C.; Li, J.
Y.; Tota, M. R.; Mosley, R.; van der Ploeg, L. H. T.;
Gantz, I. Biochemistry 2000, 39, 14900.
11. Holder, J. R.; Bauzo, R. M.; Xiang, Z. M.; Haskell-
Luevano, C. J. Med. Chem. 2002, 45, 2801.
19. Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.;
Motoshima, H.; Fox, B. A.; Le Trong, I.; Teller, D. C.;

1228
X.-W. Liu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1223–1228
Okada, T.; Stenkamp, R. E.; Yamamoto, M.; Miyano, M.
Science 2000, 289, 739.
20. Colson, A. O.; Perlman, J. H.; Smolyar, A.;
Gershengorn, M. C.; Osman, R. Biophys. J. 1998,
74, 1087.
22. Prusis, P.; Schioth, H. B.; Muceniece, R.; Herzyk, P.;
Afshar, M.; Hubbard, R. E.; Wikberg, J. E. S. J. Mol.
Graph. Model. 1997, 15, 307.
23. Prusis, P.; Muceniece, R.; Andersson, P.; Post, C.;
Lundstedt, T.; Wikberg, J. E. S. Biochim. Biophys. Acta
2001, 1544, 350.
21. SYBYL Version 6.7; Tripos, St. Louis, USA.