Discovery of 1-amino-4-phenylcyclohexane-1-carboxylic acid and its influence on agonist selectivity between human melanocortin-4 and -1 receptors in linear pentapeptides

Article abstract of DOI:10.1016/j.bmcl.2005.08.012

Linear pentapeptides (Penta-cis-Apc-DPhe-Arg-Trp-Gly-NH₂) containing 1-amino-4-phenylcyclohexane-1-carboxylic acid (cis-Apc) and substituted Apc are potent hMC4R agonists and they are inactive or weakly active in hMC1R, hMC3R, and hMC5R agonist assays. This study, together with our earlier report on 5-BrAtc, demonstrated the importance of replacing His ⁶ with phenyl-containing rigid templates in achieving good hMC4R agonist potency and selectivity against hMC1R in linear pentapeptides.

Full text of DOI:10.1016/j.bmcl.2005.08.012

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4910–4914
Discovery of 1-amino-4-phenylcyclohexane-1-carboxylic acid
and its influence on agonist selectivity between human
melanocortin-4 and -1 receptors in linear pentapeptides
Xin-Jie Chu,* David Bartkovitz, Waleed Danho, Joseph Swistok,
Adrian Wai-Hing Cheung, Grazyna Kurylko, Karen Rowan, Mitch Yeon,
Lucia Franco, Lida Qi, Li Chen and Keith Yagaloff
Roche Research Center, Hoffmann-La Roche Inc., 340 Kingsland Street, Nutley, NJ 07110, USA
Received 6 June 2005; revised 9 August 2005; accepted 9 August 2005
Available online 16 September 2005
Abstract—Linear pentapeptides (Penta-cis-Apc-DPhe-Arg-Trp-Gly-NH2) containing 1-amino-4-phenylcyclohexane-1-carboxylic
acid (cis-Apc) and substituted Apc are potent hMC4R agonists and they are inactive or weakly active in hMC1R, hMC3R, and
hMC5R agonist assays. This study, together with our earlier report on 5-BrAtc, demonstrated the importance of replacing His6 with
phenyl-containing rigid templates in achieving good hMC4R agonist potency and selectivity against hMC1R in linear
pentapeptides.
Ó 2005 Elsevier Ltd. All rights reserved.
In the last decade, five human melanocortin receptor
subtypes (hMC1R–hMC5R) have been cloned and
characterized.1 These melanocortin receptors are sev-
en-transmembrane G-protein coupled receptors
(GPCRs) which mediate a wide range of physiological
functions including pigmentation regulation (MC1R),
glucocorticoid production (MC2R), food intake and
energy expenditure (MC3R and MC4R) as well as exo-
crine gland function (MC5R).1 MC4R, in particular,
has become a current drug target for the treatment of
obesity due to its involvement in feeding behavior.2,3
It was recently suggested that MC4R also plays a role
in sexual function.4 Our laboratories are interested in
the design of potent and selective human melanocor-
tin-4 receptor (hMC4R) agonists for the treatment of
obesity.5
against hMC1R (about 15-fold).6 N-cap modification
of peptide 2 from n-butanoyl (Bu-) to n-pentanoyl
(Penta-) gave peptide 3 which showed slight improve-
ment in agonist potency at both hMC1R and hMC4R,
compared with peptide 2.6 To improve hMC4R poten-
cy and selectivity of Penta-Atc-DPhe-Arg-Trp-Gly-
NH2 (3), one strategy involves systematic substitution
of the Atc phenyl ring which resulted in 5-BrAtc.6
An alternative approach involves the use of phenyl-
containing rigid templates7 which led to the discovery
of
1-amino-4-phenylcyclohexane-1-carboxylic
acid
(Apc) (Fig. 1). As described in this report, linear pen-
tapeptides containing cis-Apc and substituted Apc
showed excellent hMC4R agonist potency and are
inactive or weakly active in hMC1R, hMC3R, and
hMC5R agonist assays.
We previously reported that using linear pentapeptide
Bu-His6-DPhe7-Arg8-Trp9-Gly10-NH2 (1, a-MSH num-
bering) as the template, replacement of His with race-
mic 2-aminotetraline-2-carboxylic acid (Atc) gave two
separable diastereomers, one of which (2) showed mod-
est hMC4R potency (EC50 = 290 nM) and selectivity
3'
6'
3'
6'
5
8
4'
5'
2'
4'
5'
2'
3
3
6
7
4
4
NH2
2
2
1
1
NH2
CO2H
cis-Apc
NH2
CO2H
CO2H
rac-Atc
trans-Apc
*
Figure 1. 2-Aminotetraline-2-carboxylic acid (Atc) and 1-amino-4-
phenylcyclohexane-1-carboxylic acid (Apc).
Corresponding author. Fax: +1 973 235 7239; e-mail: xin-jie.chu@
roche.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.08.012

X.-J. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4910–4914
4911
Synthesis and stereochemical determination of Fmoc-pro-
tected 1-amino-4-phenylcyclohexane-1-carboxylic acid.
Commercially available 4-phenyl cyclohexanone (4)
was converted via Bucherer–Bergs reaction8 into hydan-
toin 5 as a mixture of two isomers (ꢀ6:1 by NMR anal-
ysis). Direct hydrolysis of hydantoin 5 into Apc amino
acid 7 involved drastic reaction conditions (prolonged
reflux with concentrated aqueous NaOH solution) and
lengthy work-up procedures. An improved protocol
was adopted in which the hydantoin N–HÕs of 5 were
first protected with tert-butoxylcarbonyl (Boc) groups9
(giving compound 6) which allowed the subsequent
hydrolysis of the hydantoin group to proceed under
mild conditions (1 N NaOH, room temperature, over-
night). Without work-up, the pH of the above basic
reaction mixture containing Apc amino acid 7 was
adjusted to ꢀ11 and Fmoc-protection gave a readily
separable mixture of cis-Apc 8 and its minor trans-iso-
mer 9 (Scheme 1).
a
b
NHFmoc
OH
NHFmoc
OMe
99%
98%
O
O
8
10
O
O
ClO2S
c, 91%
O
S
NH2
O
NH
OMe
O
11
OMe
O
12
Scheme 2. Reagents and conditions: (a) CH2N2, Et2O, 0 °C, 30 min;
(b) piperidine, DMF, rt, 20 min; (c) (iPr)2EtN, DMAP, CH2Cl2, rt,
18 h.
The cis-configuration between the phenyl and amino
groups of the major isomer 8 was unambiguously estab-
lished by X-ray crystallographic study of the corre-
sponding camphorsulfonamide 1210 prepared via a
three-step sequence (Scheme 2).
X
O
X
OH
b,c
a
+
O
80-92%
89-98%
O
Br,I
X
O
O
14
13
15
Synthesis of Fmoc-protected substituted Apc. Addition of
various aryl lithium reagents (generated from the corre-
sponding aryl bromides/iodides 14 and n-BuLi) to 1,4-
cyclohexanedione mono-ethylene ketal (13) gave alcohol
15. Dehydration of the tertiary alcohol of 15, hydroge-
nation of the resulting olefin, followed by hydrolysis of
the cyclic ketal gave substituted phenyl cyclohexanone
17. Using a four-step sequence previously used in the
synthesis of cis-Apc 8, ketone 17 was converted into
the corresponding Fmoc-protected substituted Apc 20.
As in the case of the unsubstituted Apc isomers 8 and
9, the major cis-isomer 20 was readily separated from
its minor trans-isomer by column chromatography
(Scheme 3).
X
X
e
d
O
>95%
>98%
O
NH
HN
O
O
O
16
18
17
X
X
O
N
O
g,h
80-92%
f
>96%
CO2H
Boc
N
Boc
19
NHFmoc
20
All new peptides11 were synthesized on solid phase from
Fmoc-protected Apc (or Fmoc-protected substituted
Apc) and other suitably protected amino acids using
Scheme 3. Reagents and conditions: (a) n-BuLi, ꢁ78 °C, THF, 2 h;
(b) cat. TsOH, reflux, benzene, 3–4 h; (c) H2, 5 wt% Pd/C, EtOAc, 3 h;
(d) acetone, TsOH, 60 °C, 4–6 h; (e) KCN, (NH4)2CO3, EtOH/H2O,
overnight; (f) (Boc)2O, Et3N, DMAP, rt, THF, overnight; (g) 1 N
NaOH, DME/H2O, rt, overnight; (h) Fmoc-Cl, pH 11, dioxane/H2O,
rt, overnight.
O
O
a
b
95%
91%
Boc
N
NH
HN
N
standard Fmoc methodology. The crude peptides were
purified to homogeneity using reversed-phase HPLC
and characterized by fast atom bombardment mass
spectroscopy.
O
Boc
O
O
5
6
4
c
d
Agonist assays were performed using HEK293 cells
transfected with hMC1R–hMC5R as reported in detail
elsewhere.11,12 The EC50 values reported in Tables 1
and 2 are the average of at least two separate experi-
ments. Binding assays were performed using radiola-
beled NDP-MSH as reported in detail elsewhere.12
The IC50 values reported in Table 3 are the average of
at least two separate experiments.
+
CO2H
CO2H
CO2H
NHFmoc
85%
from 6
NHFmoc
9 (minor)
NH2
6:1
(major)
Scheme 1. Reagents and conditions: (a) KCN, (NH4)2CO3, EtOH/
H2O, 90 °C, overnight; (b) (Boc)2O, Et3N, DMAP, THF, rt, overnight;
(c) 1 N NaOH, DME, H2O, rt, overnight; (d) Fmoc-Cl, pH 11,
dioxane/H2O, rt.

4912
X.-J. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4910–4914
Table 1. Agonist activity of His6 modified pentapeptides at the human melanocortin receptors
a
EC50 (nM)
Peptide
Amino acid sequence
hMC4R
hMC1R
1
2
Bu-His-(D)Phe-Arg-Trp-Gly-NH2
20
290
10
4400
Bu-Atc-(D)Phe-Arg-Trp-Gly-NH2 (2nd isomer)b
Penta-Atc-(D)Phe-Arg-Trp-Gly-NH2 (2nd isomer)b
Penta-cis-Apc-(D)Phe-Arg-Trp-Gly-NH2
3
45
830
25% at 50 lMc
1300
55% at 50 lMc
45% at 50 lMc
0% at 50 lMc
0% at 50 lMc
60% at 50 lMc
1300
21
22
23
24
25
26
27
28
29
30
31
32
2
88
Penta-trans-Apc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-40-MeApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-40-ClApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-40-MeOApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-40-EtOApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-40-iPrOApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-40-HOApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-30-MeOApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-30,50-diiPrApc-(D)Phe-Arg-Trp-Gly-NH2
Penta-cis-Apc-(D)Phe-Arg-Trp-NH2
3
4
9
13
6
5
2
690
210
60% at 50 lMc
130
550
1300
60% at 50 lMc
trans-Pcc-(D)Phe-Arg-Trp-Gly-NH2
Bu stands for CH3CH2CH2C(@O) and Penta stands for CH3CH2CH2CH2C(@O).
a Concentration of peptide at 50% maximum cAMP accumulation or the % of cAMP accumulation (relative to NDP-MSH) observed at the highest
peptide concentration tested.
b 1st isomer and 2nd isomer refer to the order in which the two diastereomers eluted under our HPLC conditions.13
c Not tested for antagonist activities.
Table 2. Agonist activity of pentapeptides containing various His
surrogates at the human melanocortin receptors
9) is about 40-fold less potent as a hMC4R agonist
and is significantly less hMC4R selective, compared with
peptide 21. The different agonist profiles of peptides 21
and 22 clearly showed that the relative orientation of
the phenyl group in Apc has a significant impact on
the hMC4R agonist potency and selectivity of the result-
ing peptide. Due to the highly flexible nature of linear
peptides, we were unable to determine the solution
structures of peptides 1 and 21 using NMR spectrosco-
py. On the other hand, incorporation of cis-Apc into
conformationally constrained (MT-II-like) cyclic pep-
tide templates gave highly rigid solution structures
which led to a predictive pharmacophore model of
hMC4R.14
a
EC50 (nM)
Peptide
hMC1R
10
hMC3R
hMC4R hMC5R
1
21
26
0% at 50 lMb 20
0% at 50 lMb
36% at 50 lMb
46% at 50 lMb
25% at 50 lMb 57% at 50 lMb
2
0% at 50 lMb 28% at 50 lMb 13
a Concentration of peptide at 50% maximum cAMP accumulation or
the % of cAMP accumulation (relative to NDP-MSH) observed at
the highest peptide concentration tested.
b Not tested for antagonist activities.
Table 3. Binding affinity of pentapeptides containing various His
surrogates at the human melanocortin receptors
a
IC50 (nM)
In an effort to optimize cis-Apc, a series of substituents
(methyl-, chloro-, methoxy-, ethoxy-, isopropoxy-, hy-
droxy-, and isopropyl-) were introduced into cis-Apc
(general structure 20) and the agonist activities of the
resulting peptides (peptides 23–30) are shown in Table
1. Peptides 23–28, bearing 40-substituents with different
electronic and steric properties, all showed hMC4R
EC50 values within a 4-fold range (3–13 nM). Since pep-
tide 27, containing bulky 40-isopropoxy group, retains
excellent hMC4R agonist activity, it should be interest-
ing to use Apc with bulkier 40-substitution (e.g., t-but-
oxy-, phenoxy- or phenyl) to establish the size limit of
the hMC4R His binding pocket.
Peptide
hMC1R hMC3R
hMC4R hMC5R
1
21
26
580
7100
4700
4000
150
13000
Not determined
Not determined
Not determined 13
Not determined 43
a Concentration of peptide at 50% radiolabeled NDP-MSH
displacement.
As shown in Table 1, the lead pentapeptide Bu-His-
DPhe-Arg-Trp-Gly-NH2 (1) is a potent hMC4R agonist
(EC50 = 20 nM) but is not selective against hMC1R
(EC50 = 10 nM). Using Atc as a His replacement, mod-
estly hMC4R potent and selective peptides 2 and 3 were
obtained.6 Apc, in which the phenyl group is attached to
C-4 of the cyclohexane ring in an exocyclic fashion, is
structurally quite different from Atc (Fig. 1). We were
delighted to find that peptide 21 (made from cis-Apc
8) is over 22-fold (the standard error in our assays is
about 2-fold) more potent as a hMC4R agonist com-
pared with peptide 3 and at the same time possessed
minimal agonist activity at hMC1R (25% activation at
50 lM). In contrast, peptide 22 (made from trans-Apc
Peptide 29, which contains 30-MeOApc, possessed the
same hMC4R potency (EC50 = 2 nM) as peptide 21.
Peptide 30, with isopropyl groups at both 30 and 50 posi-
tions of Apc, is weakly active as a hMC4R agonist, per-
haps due to an unfavorable steric interaction with the
hMC4R binding pocket. Peptide 30 was not tested for
its hMC4R binding and antagonist activity. Interesting-
ly, both peptides 29 and 30 with 30-substitution showed
higher hMC1R agonist activities compared with peptide
21. More diverse substitutions need to be explored at the

X.-J. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4910–4914
4913
20- and 30-positions of cis-Apc to fully establish the
structure–activity relationship in this region.
rigid templates in achieving good hMC4R agonist
potency and selectivity against hMC1R in linear penta-
peptides.18 Further modification of Apc amino acid,
incorporation of Apc and related analogs into cyclic
peptides, and in vivo studies using Apc-containing line-
ar/cyclic peptides would be reported in due course.
In an attempt to reduce the molecular weight of peptide
21, truncated peptides such as 31 and 32 were prepared.
Tetrapeptide 31, with the Gly residue removed, is signif-
icantly less potent and selective as a hMC4R agonist,
compared to pentapeptide 21. Peptide 32, containing
trans-4-phenylcyclohexane-1-carboxylic acid (trans-Pcc)
(Fig. 2), which could be considered as Apc with its
Acknowledgments
N-cap truncated, is only moderately potent (EC50
=
The authors are grateful to Dr. Christopher S. Framp-
ton (Roche Welwyn, UK) and Dr. Bradford Graves
for X-ray crystallographic study of compound 12 and
to the Physical Chemistry Department for spectroscopic
measurements and interpretations. The authors would
also like to thank Drs. Ramakanth Sarabu and Jefferson
Tilley for their critical reading of the manuscript.
550 nM) as a hMC4R agonist. The reduced hMC4R
agonist activities of peptides 31 and 32 clearly demon-
strate the important roles played by the Gly residue
and N-cap of peptide 21.
The most hMC4R selective peptides described above
were tested in hMC3R and hMC5R agonist assays
and the results are shown in Table 2. The lead pentapep-
tide 1 (Bu-His-DPhe-Arg-Trp-Gly-NH2) was inactive in
both hMC3R and hMC5R agonist assays. The low ago-
nist activities in hMC3R and hMC5R are maintained
when His of peptide 1 was replaced by cis-Apc (peptide
21) and cis-40-EtOApc (peptide 26). The above peptides
were also, in certain cases, tested in hMC1R–hMC5R
binding assays (Table 3). The binding affinities of the
peptides 1, 21, and 26 toward hMC4R, track with their
agonist activities in the order of 21 > 26 > 1. Peptides 21
and 26 bind to hMC1R with micromolar affinity, but
showed no significant agonist activity.
References and notes
1. The Melanocortin Receptors; Cone, R. D., Ed.; Humana:
Totowa, 2000.
2. Fan, W.; Boston, B. A.; Kesterson, R. A.; Hruby, V. J.;
Cone, R. D. Nature 1997, 385, 165.
3. Huszar, D.; Lynch, C. A.; Fairchild-Huntress, V.; Dun-
more, J. H.; Fang, Q.; Berkemeier, L. R.; Gu, W.;
Kesterson, R. A.; Boston, B. A.; Cone, R. D.; Smith, F. J.;
Campfield, L. A.; Burn, P.; Lee, F. Cell 1997, 8, 131.
4. (a) Wessells, H.; Gralnek, D.; Dorr, R.; Hruby, V. J.;
Hadley, M. E.; Levine, N. Urology 2000, 56, 641; (b)
Wessells, H.; Fuciarelli, K.; Hansen, J.; Hadley, M. E.;
Hruby, V. J.; Dorr, R.; Levine, N. J. Urol. 1998, 160, 389;
(c) Martin, W. J.; MacIntyre, D. E. Eur. Urol. 2004, 45,
706.
5. Boyce, R. S.; Duhl, D. M. Curr. Opin. Invest. Drugs 2004,
5, 1063.
6. Cheung, A. W.-H.; Danho, W.; Swistok, J.; Qi, L.;
Kurylko, G.; Rowan, K.; Yeon, M.; Franco, L.; Chu,
X.-J.; Chen, L.; Yagaloff, K. Bioorg. Med. Chem. Lett.
2003, 13, 133.
Hruby et al. proposed that Ôthe position 6 of the synthet-
ic melanocortin ligands is important for enhancing
potency and selectivity at MC3 and MC4 melanocortin
receptorsÕ.15,16 HrubyÕs study was carried out using a
cyclic peptide antagonist as the template (SHU9119)
and the resulting compounds are all hMC4R antago-
nists with no hMC1R biochemical data.15,16 In a study
by Haskell-Luevano et al., 17 histidine surrogates were
introduced into a Ac-His6-DPhe7-Arg8-Trp9-NH2 tetra-
peptide template17 and the tetrapeptide containing ami-
no-2-naphthylcarboxylic acid (Anc) showed good mouse
MC4R agonist potency (EC50 = 21 nM) and good selec-
tivity against mouse MC1R (EC50 = 7900 nM). The
biological activities of these tetrapeptides at human mel-
anocortin receptors were not reported.17 Our work de-
scribed herein showed that linear pentapeptides
7. Phenyl-containing rigid templates investigated include:
O
O
O
N
N
NH
N
N
(Penta-cis-Apc-DPhe-Arg-Trp-Gly-NH2)
containing
N
cis-Apc and substituted Apc are potent hMC4R ago-
nists and they are inactive or weakly active in hMC1R,
hMC3R, and hMC5R agonist assays. This report on
Apc and substituted Apc, together with our earlier re-
port on 5-BrAtc and related Atc,6 demonstrated the
importance of replacing His6 with phenyl-containing
H
O
O
O
rac-
rac-
None of the pentapeptides derived from these amino acids
is as potent or hMC4R selective as peptide 3 (data not
shown).
8. Obrecht, D.; Spiegler, C.; Schonholzer, P.; Muller, K.;
Heimgartner, H.; Stierli, F. Helv. Chim. Acta 1992, 75,
1666.
9. Wysong, C. L.; Yokum, T. S.; Morales, G. A.; Gundry, R.
L.; McLaughlin, M. L.; Hammer, R. P. J. Org. Chem.
1996, 61, 7650.
10. Crystallographic data for compound 12 have been depos-
ited with the Cambridge Crystallographic Data Center as
supplementary publication number CCDC 283241.
11. (a) Chen, L.; Cheung, A. W-H.; Chu, X-J.; Danho, W.;
Swistok, J.; Yagaloff, K. A. WO 0174844, 2001; CAN
CO2H
trans-Pcc
Figure 2. Structure of trans-4-phenylcyclohexane-1-carboxylic acid
(trans-Pcc).

4914
X.-J. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4910–4914
135:304143.; (b) Chen, L.; Cheung, A. W-H.; Chu, X-J.;
Danho, W.; Swistok, J.; Wang, Y.; Yagaloff, K. A. WO
0218437, 2002; CAN 136:217052.
15. Hruby, V. J.; Grieco, P.; Balse, P.; Han, G.; Weinberg, D.;
McNeil, T. Abstract of Papers, 217th National Meeting of
the American Chemical Society, Anaheim, CA; American
Chemical Society: Washington, DC, 1999; Abstract
MEDI-231.
16. Grieco, P.; Han, G.; Weinberg, D.; MacNeil, T.; Van der
Ploeg, L. H. T.; Hruby, V. J. Biochem. Biophys. Res.
Commun. 2002, 292, 1075.
12. Benoit, S. C.; Schwartz, M. W.; Lachey, J. L.; Hagan, M.
M.; Rushing, P. A.; Blake, K. A.; Yagaloff, K. A.;
Kurylko, G.; Franco, L.; Danho, W.; Seeley, R. J. J.
Neurosci. 2000, 20, 3442.
13. The diastereomeric mixture of pentapeptides containing
racemic Atc are separable using reversed-phase high
performance liquid chromatography (HPLC) on a Vydac
17. Holder, J. R.; Bauzo, R. M.; Xiang, Z.; Haskell-Luevano,
C. J. Med. Chem. 2002, 45, 2801.
C
18 column. Gradient elution (10% buffer B to 60% buffer
18. A portion of this work was previously presented. Danho,
W.; Swistok, J.; Cheung, A.; Chu, X-J.; Wang, Y.; Chen,
L.; Bartkovitz, D.; Gore, V.; Qi, L.; Fry, D.; Greeley, D.;
Sun, H.; Guenot, J.; Franco, L.; Kurylko, G.; Rumennik,
L.; Yagaloff, K. Abstracts of Papers, 2nd International/
17th American Peptide Symposium; Kluwer Academic:
The Netherlands, 2001; 701.
B) was carried out over 90 min at a flow rate of 8 ml/min
using 0.1% TFA/H2O (buffer A) and 0.1% TFA/CH3CN
(buffer B) with UV detection at 280 nm.
14. Sun, H.; Greeley, D. N.; Chu, X.-J.; Cheung, A.; Danho,
W.; Swistok, J.; Wang, Y.; Zhao, C.; Chen, L.; Fry, D. C.
Bioorg. Med. Chem. 2004, 12, 2671.