β-Alanine dipeptides as MC4R agonists

Article abstract of DOI:10.1016/j.bmcl.2003.09.036

β-Alanine derivative 2 (IC₅₀=16 nM) and related compounds were found to be potent MC4R agonists.

Full text of DOI:10.1016/j.bmcl.2003.09.036

Bioorganic & Medicinal Chemistry Letters 13 (2003) 4341–4344
ꢀ-Alanine Dipeptides as MC4R Agonists
Rejean Ruel,a,* Timothy F. Herpin,c Lawrence Iben,b Guanglin Luo,b
Alain Martel,a Helen Mason,c Gail Mattson,b Brigitte Poirier,a
Edward H. Ruediger,a Dan Shi,c Carl Thibault,a Guixue Yu,c Ildiko Antal Zimanyi,c
Graham S. Poindexterb and John E. Macorb
aBristol-Myers Squibb Pharmaceutical Research Institute, 100 boul. de l’Industrie, Candiac, Quebec, Canada J5R 1J1
bDepartment of Neuroscience, 5 Research Parkway, Wallingford, CT 06492, USA
cDepartment of Metabolic Research, POB 5400 Princeton, NJ08543-5400, USA
Received 14 July 2003; revised 9 September 2003; accepted 17 September 2003
Abstract—b-Alanine derivative 2 (IC50=16 nM) and related compounds were found to be potent MC4R agonists.
# 2003 Elsevier Ltd. All rights reserved.
The melanocortin (MC) receptorsbelong to the G-pro-
tein coupled receptor superfamily and are involved in a
number of physiological functions including feeding
behavior, sexual behavior, immune regulation and skin
pigmentation.1 There are five MC receptors(MC1R–
MC5R) which, except for MC2R, all recognize the
melanocyte-stimulating hormones (a-, b-, and g-MSH)
and their activation leadsto a downstream elevation of
cAMP.1h,2
alanine unit and a 4,4-disubstituted piperidine moiety.
Using 1 asa guide, we initiated a program to identify
related bis-amide MC4R agonists with improved
physicochemical properties (Mr, lipophilicity, solubi-
lity). Herein we report our effortsat the identification
and optimization of novel N-terminal analoguesof 1,
which have enhanced physicochemical properties.
We first examined simple tetrahydroisoquinoline repla-
cementswith lower molecular weights, lower lipophili-
Based on experiments in various rodent feeding models,
the MC4R of the hypothalamusand brain tsem may
play an important role in food intake and energy
homeostasis. For example, it has been shown that icv
administration (<10 mg/animal) of the agonist peptide
a-MSH resulted in long lasting (6–9 h) inhibition of
spontaneous feeding in rats.3 The non-selective MC4R
agonist MTII has also been shown to reduce food con-
sumption in rats.4 Furthermore, MC4R knockout mice
were shown to develop morbid obesity,5 and thisphe-
notype was insensitive to administration of the cyclic a-
MSH analogue MTII.6
citiesand enhanced oslubilitiescompared to
1.8 The
simple b-alanine derivative 29,13 wasfound to retain
potent binding affinity for MC4R (IC50=16 nM)10
compared to 1 (IC50=7 nM, Table 1). The calculated
lipophilicity of 2 [cLog D (pH 7.4)=2.94] wasreduced
by at least three log units compared to 1 [cLog D (pH
7.4)=5.92].8 Furthermore, b-alanine 2 wasfound to be
a full agonist in a functional cAMP assay11 (EC50=170
nM and Intrinsic Activity (IA)=98%). These results
prompted usto expand the SAR around thislead. The
structure/activity around substitution of the basic
nitrogen atom of the b-alanine issummarized in Table
1. The N-methyl and N,N-dimethyl derivatives( 3a and
3b, respectively) were equipotent with the primary
amine 2. Additionally, there appeared to be some steric
tolerance at thispoistion isnce the mono-benzylated
analogue 3c wasalso equipotent with the primary amine
2. However, additional large substitution of the nitro-
gen atom led to a significant diminution of binding affi-
nity asdemonstrated by the N,N-dibenzyl analogue 3d.
Researchers from Merck recently reported a number of
MC4R agonists exemplified by structure 1 (Fig. 1).7
These lipophilic, tetrahydroisoquinoline (TIC)-based
dipeptideshsare a common para-substituted d-phenyl-
*Corresponding author. Tel.:+1-450-444-6129; fax:+1-450-444-4166;
e-mail: rejean.ruel@bms.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.09.036

4342
R. Ruel et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4341–4344
Table 2. MC4R binding of substituted and cyclic b-alanine deriva-
tives 4
Compd
R
MC4R (IC50, nM)a
30b
4a
4b
4c
4d
4e
4f
18Æ10
16b
Figure 1.
Table 1. MC4R binding of 1 and b-alanine derivatives 2 and 3
Compd
R1
R2
MC4R (IC50, nM)a
9b
1
2
3a
3b
3c
3d
—
H
Me
Me
H
—
H
H
Me
Bn
Bn
9Æ3
16Æ8
10Æ8
10Æ8
14Æ4
110b
12Æ4
22b
Bn
aValues are means of three or more experiments unless otherwise
noted.
bValuesare meansof two experiments.
aValuesare meansof three or more experimentsunlessotherwies
noted.
Substitution on the sp3 carbon atomsof the b-alanine
pharmacophore was also investigated (Table 2). The
gem-dimethyl analogues 4a and 4b and the cyclic ana-
logues 4c–4f were similarly potent to b-alanine 2. Also,
the configuration of the chiral center in the cyclic dia-
stereomers 4d and 4e had no influence on MC4R bind-
ing affinity. Thus, this region of the agonists was
reasonably promiscuous as a variety of basic amine
moietiesled to potent MC4R agonists( Table 2).
bValuesare meansof two experiments.
Table 3. MC4R binding of b-alanine amide derivatives 5
In order to examine the effect of basicity of 2 on MC4R
affinity, we prepared a series of related amide deriva-
tives 5, and the results are summarized in Table 3. Based
on the hypothesis that the minimal core sequence of
His-Phe-Arg-Trp was necessary for MC4R recogni-
tion,12 we anticipated that the corresponding amides 5
would show reduced potency at MC4Rs compared with
their amine counterparts. Surprisingly, acetamide 5a
wasfound to be equipotent with the primary amine 2.
Compd
R
MC4R (IC50, nM)a
5a
5b
5c
5d
5e
Me
MeOCH2
Ph
CF3
Me2NCH2CH2
12b
21Æ19
90b
180b
11Æ6
12b
The methoxymethyl ether 5b wasalso equipotent with
2
and 5a, but the bulkier benzamide 5c had reduced
MC4R affinity compared to the simple acetamide 5a.
The more lipophilic trifluoroacetamide 5d also lost sig-
nificant binding affinity. Interestingly, re-introduction of
a basic nitrogen in the b-alanine amide side chain
afforded compounds( 5e–5h) with potent MC4R affi-
nitiescomparable to that of the original b-alanine 2
(Table 3). The fact that smaller and hydrophilic sub-
stituents (i.e., 4a – 4f, 5a, 5g and 5h) were tolerated at
this position, but that larger hydrophobic substituents
were not (i.e., 3d, 5c, 5d), coupled with the lack of ste-
reogenic differentiation at thisposition (i.e., 4d and 4e,
5f
5g
5h
25b
22Æ14
aValuesare meansof three or more experimentsunlessotherwies
noted.
bValuesare meansof two experiments.

R. Ruel et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4341–4344
4343
fluoroacetic acid to give b-alanine 2.13 Thiscompound
served as the common intermediate for the synthesis of
compounds 3–5 (Scheme 1). N-Methyl b-alanine 3a was
prepared using an EDCI-induced coupling of amine 6
and N-Boc-N-methyl-b-alanine 7, followed by treatment
with trifluoroacetic acid. N,N-dimethyl-b-alanine 3b was
prepared via a reductive amination of b-alanine 2 using
excess formaldehyde. Reductive amination of 2 with
stoichiometric and excess benzaldehye afforded 3c and
3d, respectively. Standard amide coupling of 2 with the
appropriate activated acids, followed by deprotection
(where needed), afforded compounds 4a–f and com-
pounds 5a–h.
Table 4. Functional activity of selected compounds
Compd
MC4R
MC4R
IA (%)b
(IC50, nM)a
(EC50, nM)a
1
2
3a
3b
7
16
10
10
3.6
170
70
99Æ3
92Æ5
98Æ10
98Æ7
40
aValuesare meansof three or more experiments.
bValuesare meansof four or more experiments.
5g and 5h) could suggest that this portion of the mol-
ecule wassolvent-exposed in the receptor. It would then
follow that potency wasdriven by that portion of the
molecule’shydrophilicity more than anything else. Our
present SAR studies are being directed to examine this
hypothesis.
Conclusion
Replacement of the tetrahydroisoquinoline moiety in 1
by a simple b-alanine moiety provided compoundswith
potent affinity for the MC4 receptor and enhanced
physicochemical properties. Optimization of this series
provided several compounds with low nanomolar bind-
ing affinity for MC4R and full agonist activity.14 Our
SAR study described herein indicated that a wide range
of substitution are tolerated on the N-terminus of these
b-alanine dipeptides. Small hydrophilic substituents
seemed to be tolerated the best, possibly suggesting that
this portion of the agonist was solvent-exposed. Our
future effortswill be directed at further tetsing and
evaluating this hypothesis.
The functional propertiesof b-alanines 2, 3a and 3b are
provided in Table 4. All compoundswere found to be
full agonists with EC50 valuesranging from 40 to 170
nM11 and with intrinsic activities of >90%. Intrinsic
activity was expressed as % cAMP stimulation com-
pared to that of 100 nM NDP-a-MSH, which causes
maximal stimulation in this system. For each compound
the assays were performed twice or three times in
duplicate. NDP-a-MSH and the natural ligand, a-
MSH, stimulated cAMP to similar levels, with an EC50
of 0.3 and 5.5 nM, respectively. NDP-a-MSH wasused
as a reference compound in these studies due to its
higher affinity and greater stability. Intrinsic activity of
>90% wasgeneral for all compoundsreported in this
manuscript (data not shown).
Acknowledgements
We are grateful to Dr. William R. Ewing for helpful
suggestions during the preparation of the manuscript.
Synthesis
References and Notes
Scheme 1 describes the synthesis of the b-alanines 2–5.
Amine 67 wascoupled with N-Boc-protected b-alanine,
and the resulting carbamate was treated with tri-
1. For some recent reviews on melanocortin receptors, see: (a)
Antal Zimanyi, I.; Pelleymounter, M. A. Curr. Pharma. Des.
2003, 9, 627. (b) Wikberg, J. E. S. Expert Opin. Ther. Patents
2001, 11, 61. (c) Cone, R. D. The Melanocortin Receptors;
Humana: Totwa, NJ, 2000. (d) Van der Ploeg, L. H. T. Curr.
Opin. Chem. Biol. 2000, 4, 452. (e) Wikberg, J. E. S.; Muce-
niece, R.; Mandrika, I.; Prusis, P.; Lindblom, J.; Post, C.;
Skottner, A. Pharmacol. Res. 2000, 42, 393. (f) Schioth, H. B.
Vitam. Horm. 2001, 63, 195. (g) Vergoni, A. V.; Schioth, H. B.;
Bertolini, A. Biomed. & Pharmacother. 2000, 54, 129. (h)
Wikberg, J. E. S. Eur. J. Pharmacol. 1999, 375, 295. (i) Speake,
J. D.; Bishop, M. J. Expert Opin. Ther. Pat. 2002, 12, 1631. (j)
Pan, K.; Scott, M. K.; Lee, D. H. S.; Fitzpatrick, L. J.;
Crooke, J. J.; Rivero, R. A.; Rosenthal, D. I.; Vaidya, A. H.;
Zhao, B.; Reitz, A. B. Bioorg. Med. Chem. 2003, 11, 185.
2. Schioth, H. B.; Chhajlani, V.; Muceniece, R.; Klusa, V.;
Wikberg, J. E. S. Life Sci. 1996, 59, 797.
3. Poggioli, R.; Vergoni, A. V.; Bertolini, A. Peptides 1986, 7,
843.
4. (a) Fan, W.; Boston, B. A.; Kesterson, R. A.; Hruby, V. J.;
Cone, R. D. Nature 1997, 385, 165. (b) Murphy, B.; Nunes,
C. N.; Ronan, J. J.; Hanaway, M.; Fairhurst, A. M.; Mellin,
T. N. J. Appl. Physiol. 2000, 89, 273.
5. Huszar, D.; Lynch, C. A.; Fairchild-Huntress, V.; Dun-
more, J. H.; Fang, Q.; Berkemeir, L. R.; Gu, W.; Kesterson,
Scheme 1. Synthesis of the b-alanine derivatives: (a) (1) N-Boc-b-
ALA, EDCI, HOBt, CH2Cl2; (2) TFA, CH2Cl2; (b) RCHO, NaBH
(OAc)3, DMF; (c) (1) N-Boc-b-aminoacid, EDCl, HOBt, CH2Cl2; (2)
TFA, CH2Cl2; (d) RCOOH, EDCl, HOBt, CH2Cl2; (e) (1) 7, EDCl,
HOBt, CH2Cl2; (2) TFA, CH2Cl2.

4344
R. Ruel et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4341–4344
R. A.; Boston, B. A.; Cone, R. D.; Smith, F. J.; Campfield,
L. A.; Burn, P.; Lee, F. Cell 1997, 88, 131.
12. Hruby, V. J.; Wilkes, B. C.; Hadley, M. E.; Al-Obeidi, F.;
Sawyer, T. K.; Staples, D. J.; Devaux, A.; Dym, O.; Castrucci,
A. M.; Hintz, M. F.; Riehm, J. P.; Rao, K. R. J. Med. Chem.
1987, 30, 2126.
13. Synthesis of 2: To a solution of a-amino amide 6 (1.1 g,
2.56 mmol) and N-Boc-b-alanine (531 mg, 2.81 mmol) in di-
6. Chen, A. S.; Metzger, J. M.; Trumbauer, M. E.; Myrna, E.;
Guan, X.-M.; Yu, H.; Frazier, E. G.; Marsh, D. J.; Forrest,
M. J.; Gopal-Truter, S.; Fisher, J.; Camacho, R. E.; Strack,
A. M.; Mellin, T. N.; MacIntyre, D. E.; Chen, H. Y.; Van der
Ploeg, L. H. T. Transgenic Res. 2000, 9, 145.
chloromethane
(12
mL)
wasadded
1-ethyl-3-(3-
7. (a) Sebhat, I. K.; Martin, W. J.; Ye, Z.; Barakat, K.; Mos-
ley, R. T.; Johnston, D. B. R.; Bakshi, R.; Palucki, B.; Wein-
berg, 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.
(b) Bakshi, R. K.; Barakat, K. J.; Nargund, R. P.; Palucki, B. L.;
Patchett, A. A.; Sebhat, I.; Ye, Z.; Van der Ploeg, L. H. T.
WO 0074679, and referencestherein.
dimethylaminopropyl) carbodiimide hydrochloride (EDCI)
(736 mg, 3.8 mmol) and 1-hydroxybenzotriazole hydrate
(HOBt) (518 mg, 3.8 mmol) at room temperature. The mixture
was stirred at room temperature overnight and a saturated
solution of ammonium chloride (15 mL) was added. The
separated aqueous layer was extracted with dichloromethane
(3Â25 mL) and the combined organic layerswere dried
(MgSO4), filtered, and evaporated to afford the tert-butylcar-
bamate which was used in the next step without further pur-
ification. To a solution of this carbamate (1.0 g, 1.7 mmol) in
dichloromethane (10 mL) wasadded a 20% (v/v) solution of
trifluoroacetic acid in dichloromethane (1.6 mL) at room
temperature. The mixture wasstirred at room temperature for
8 h and then evaporated under reduced pressure. The residue
waspurified using preparative (Column: YMC-PACK S5-C18
30Â100 mm) HPLC (acetonitrile–0.1% TFA/water: 7 min
gradient from 10% AcCN to 90% AcCN at 220 nm. Flow
rate: 30 mL/min) and, after evaporation, the residue was lyo-
philized to afford 0.9 g (47% yield) of the trifluoroacetate salt.
HPLC/MS (Column: Premisphere C18 4.6Â30 mm; Flow rate:
4 mL/min, Solvent system: 0–100% B in 2 min. Solvent A:
10% CH3CN– 90%H2O – 5 mM NH4OAc; Solvent B: 90%
CH3CN– 10%H2O – 5 mM NH4OAc; UV: 220 nm; Micro-
mass ZMD 2000, ESI): retention time 1.50 min, MS pos. m/z
8. 1: Mr=589, clogD (pH 7.4)=5.92, polar surface area
2
˚
(PSA)=85.1 A . 2: Mr=501, clogD (pH 7.4)=2.94,
2
˚
PSA=97.6 A . WindowsXP ACD Software verison 5.0 was
used to perform these calculations.
9. Related glycine and 4-aminobutyryl amide derivativeswere
also studied. This work will be reported in due course.
10. Binding activity of compounds was measured using mem-
branes of Hi5 cells expressing the human MC4R receptors.
Homogenateswere incubated with 100 pM [ 125I]-NDP-a-
MSH and increasing concentrations of unlabeled test com-
pound (at final DMSO concentration of 1%) for 90 min at
37 ꢀC in buffer consisting of 25 mM HEPES (pH 7.4), 140 mM
NaCl, 2.5 mM CaCl2, 1.2 mM MgCl2, and 0.1% BSA. Assays
were stopped by addition of ice cold wash buffer (20 mM
HEPES, 5 mM MgCl2) and filtrated over GF/B glass fiber fil-
ters that were previously soaked in 1% PEI. Non-specific
binding wasdefined in the preesnce of 1 mM NDP-a-MSH.
Filters were measured by liquid scintillation counting, and the
data was analyzed by non-linear regression using the XLfit
function of EXCEL.
11. Stimulation of adenylate cyclase in HEK293 cells expres-
sing the human MC4R receptor was determined by measuring
cAMP accumulation using the RPA559 SPA assay kit from
Amersham. Recombinant HEK293 cells were cultured in
MEM with 400 mg/mL G418, 100 mM sodium pyruvate and
10% heat inactivated FBS. Cellswere seeded on 96-well plates
at 5Â104 cells/well and incubated overnight. The culture med-
ium was aspirated and changed to the assay medium contain-
ing 0.1 mM IBMX, and increasing concentrations of the test
compound at final DMSO concentration of 1%. The cellswere
incubated for 10 min at 37 ꢀC, then the medium wasaspirated
and replaced with the lysis medium, and cAMP concentration
wasmeausred according to the manufacturer’sdirection.
Intrinsic activity was expressed as percent of maximal activity
achieved by NDP-a-MSH. Data wasanalyzed by non-linear
regression using the PRIZM program.
501 (M+H)+; IR (nmax, KBr) 3600–2880, 1695, 1620 cmÀ1
;
1H NMR (400 MHz, MeOH-d4) d ppm (two rotamers; 1:2
ratio) 8.43 (1H, s, minor rotamer), 8.42 (1H, s, major rota-
mer), 7.96 (1H, s, minor rotamer), 7.92 (1H, s, major rotamer),
7.26 (2H, d, J=8.3 Hz, major rotamer), 7.23 (2H, d, J=8.4
Hz, minor rotamer), 7.18 (2H, d, J=8.3 Hz, major rotamer),
7.15 (2H, d, J=8.6 Hz, minor rotamer), 4.98 (1H, t, J=7.8
Hz), 4.21 (2H, s, major rotamer), 4.18 (2H, s, minor rotamer),
3.60 (1H, m), 3.31 (3H, m), 3.08 (2H, m), 2.87 (2H, m), 2.54
(2H, t, J=6.5 Hz), 1.95–0.82 (15H, m). Anal. calcd for
.
.
C26H37ClN6O2 3HCl H2O: C, 49.69; H, 6.74; N, 13.37.
Found: C, 49.96; H, 6.75; N, 12.88.
15
14. Selectivity towardsthe MC1 receptor
of selected com-
pounds was also measured. The IC50 values(nM) for MC1R
of selected compounds are provided along with the MC1R/
MC4R ratio given in parentheses: 2: 690 (43), 3a: 500 (55), 3b:
110 (12) and 5a: 4400 (367).
15. For a recent disclosure of related compounds with high
affinity on MC1 receptor, see: Herpin, T. F.; Yu, G.; Carlson,
K. E.; Morton, G. C.; Wu, X.; Kang, L.; Tuerdi, H.; Khanna,
H.; Tokarski, J. S.; Lawrence, R. M.; Macor, J. E. J. Med.
Chem. 2003, 46, 1123.