Identification and characterization of pyrrolidine diastereoisomers as potent functional agonists and antagonists of the human melanocortin-4 receptor

Article abstract of DOI:10.1016/j.bmcl.2007.10.115

A series of trans-4-phenylpyrrolidine-3-carboxamides were synthesized and characterized as potent ligands of the human melanocortin-4 receptor. Interestingly, a pair of diastereoisomers 13b displayed potent functional agonist and antagonist activity, resp

Full text of DOI:10.1016/j.bmcl.2007.10.115

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 129–136
Identification and characterization of pyrrolidine
diastereoisomers as potent functional agonists and antagonists
of the human melanocortin-4 receptor
Chen Chen,a,* Wanlong Jiang,a Joe A. Tran,a Fabio C. Tucci,a, Beth A. Fleck,b
Stacy Markison,c Jenny Wen,d Ajay Madan,d Sam R. Hoare,b Alan C. Foster,c
Dragan Marinkovic,a Caroline W. Chen,a Melissa Arellanoa and John Saundersa,à
aDepartment of Medicinal Chemistry, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
bDepartment of Pharmacology, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
cDepartment of Neuroscience, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
dDepartment of Preclinical Development, Neurocrine Biosciences, Inc., 12790 El Camino Real, San Diego, CA 92130, USA
Received 25 September 2007; revised 30 October 2007; accepted 31 October 2007
Available online 4 November 2007
Abstract—A series of trans-4-phenylpyrrolidine-3-carboxamides were synthesized and characterized as potent ligands of the human
melanocortin-4 receptor. Interestingly, a pair of diastereoisomers 13b displayed potent functional agonist and antagonist activity,
respectively. Thus, the 3S,4R-pyrrolidine 13b-1 possessed a Ki of 1.0 nM and an EC50 of 3.8 nM, while its 3R,4S-isomer 13b-2 exhib-
ited a Ki of 4.7 and an IC50 of 64 nM. Both compounds were highly selective over other melanocortin receptor subtypes. The MC4R
agonist 13b-1 also demonstrated efficacy in a diet-induced obesity model in rats.
Ó 2007 Elsevier Ltd. All rights reserved.
The melanocortin receptor family belongs to the G-pro-
tein-coupled receptor (GPCR) superfamily and consists
of five receptor subtypes that are highly expressed in the
brain and other tissues and are regulated by the endog-
enous peptide agonists, melanocyte-stimulating hor-
mones (a-, b-, and c-MSH) and adrenocorticotropic
hormone (ACTH), and by the endogenous antagonists,
agouti-protein and agouti-related protein.1 The melano-
cortin-4 receptor (MC4R) plays an important role in the
regulation of feeding behavior,2 and thus, a potent and
selective MC4R agonist with brain penetration could
perhaps be useful in the treatment of obesity.3
Several small molecule MC4R agonists from different
chemical classes have been discovered.4 For example,
pyrrolidine derivatives exemplified by 1 (Fig. 1) have
been reported by Ujjainwalla to be potent and selective
at MC4R.5 We have previously reported that, by com-
bining the trans-4-arylpyrrolidine-3-carbonyl moiety in
1 with the piperazinebenzylamine portion in 2, presented
in a series of 3-aryl-2-methylpropionamide MC4R
antagonists,6 potent MC4R agonists such as 3 have been
identified.7 Here we report the optimization of this series
of compounds and the in vitro characterization of the
diastereoisomers 13b-1 and 13b-2 as a potent MC4R
agonist and antagonist, respectively. In vivo studies of
the agonist 13b-1 in feeding and obesity animal models
are also discussed.
Keywords: Melanocortin-4 receptor; Agonist; Antagonist; Potency;
Stereochemistry; Pyrrolidine; Obesity; Animal model.
The trans-N-isopropyl-4-arylpyrrolidin-3-ylcarboxylic
esters 6 were synthesized from the corresponding methyl
cinnamates and N-(trimethylsilylmethyl)-N-(methoxy-
methyl)isopropylamine 5, which was prepared by the
reaction of isopropylamine with trimethylsilylmethyl
chloride 4, followed by formaldehyde and methanol,
using a procedure similar to that described in the litera-
ture.8 Hydrolysis of 6 under basic conditions gave the
*
Corresponding author. Tel.: +1 858 617 7634; fax: +1 858 617
7602; e-mail: cchen@neurocrine.com
Present address: Department of Medicinal Chemistry, Tanabe
Research Laboratories, USA, Inc., 4540 Towne Centre Ct., San
Diego, CA 92121, USA.
à
Present address: Vertex Pharmaceuticals, Inc., 11010 Torreyana
Road, San Diego, CA 92121, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.10.115

130
C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
F
F
N1
2
N
N
N
5
3
N
N
N
4
O
NH
N
NH
F
O
O
O
O
O
O
NH2
NH2
Cl
Cl
F
1
2
3
Figure 1. Chemical structures of compounds 1–3.
acids 7a–n in good yields. The cis-isomer cis-7a was syn-
thesized using the same procedure but from methyl cis-
4-chlorocinnamate.9 The b-alanine derivative 9a was
prepared from the benzylamine 8a.10 Thus, de-sulfin-
ation of 8a with HCl in methanol afforded an amine
intermediate which was coupled with N,N-dimethyl-b-
alanine under standard peptide coupling conditions to
give the amine 9a after Boc-deprotection with trifluoro-
acetic acid. Coupling reactions of the trans-pyrrolidine-
carboxylic acids 7a–n with the amine 9a provided the
amides 10a–n in various yields (Scheme 1).
Alternatively, a coupling reaction of 11 with the amine
generated from 8b with TFA afforded the secondary
amine 15 after Boc-deprotection. Reductive alkylation
of 15 with tetrahydropyran-4-one provided the tetrahy-
dropyran-4-yl (4-THP) amine 16 after HCl/MeOH treat-
ment. The single isomers 16-1 and 16-2 were obtained,
respectively, using the same protocol from the single dia-
stereoisomers 11-1 (3S,4R) and 11-2 (3R,4S) that were
obtained from the separation of trans-N-benzylpyrroli-
dinecarboxylic acid using (4R)- or (4S)-benzyloxazoline
as previously described.11 Dimethylation of 16 using
formaldehyde under reductive conditions gave the ter-
tiary benzylamine 17 in 51% yield. A coupling reaction
of 16 with N,N-dimethylaminopropionic acid provided
the amides 13b. Two single isomers 13b-1 and 13b-2
were obtained from 16-1 and 16-2, respectively, using
the same method. Finally, coupling reactions of 16 with
various N-Boc amino acids afforded the amides 18 after
Boc-deprotection.
Compound 13a was synthesized from the piperazineben-
zylamine 8b as shown in Scheme 2. Treatment of 8b with
HCl in methanol afforded the corresponding amine
which was coupled to N,N-dimethylaminopropionic
acid to provide the amide intermediate 9b after TFA-
treatment. A coupling reaction of 9b with trans-N-Boc-
4-(4-chlorophenyl)pyrrolidine-3-carboxylic acid 11 gave
the intermediate 12 after Boc-deprotection. Reductive
alkylation of 12 with acetone afforded the isopropyl-
amine 13a. cis-13 was synthesized by directly coupling
9b with cis-7a using a procedure similar to that for
10a. The amides 14a–b were obtained by the reaction
of 12 with acetyl or propionyl chloride in the presence
of triethylamine.
The final compounds were tested in a binding assay in
HEK293 cells expressing human melanocortin-4 recep-
tors using [125I]-NDP-MSH as the radiolabeled ligand.
Compounds with high affinity were also tested in a
cAMP functional assay as previously described.12 The
results are summarized in Tables 1–3.
N
N
a
b
c
Si
Cl
Si
N
O
MeO
HO
Ar
Ar
O
(trans)
O
5
4
6
7a-n (trans)
7a: Ar = 4-ClPh
F
F
F
e
d
N
N
N
N
N
O
N
NH
NH
NH
NH
Ar
O
S
O
O
O
O
N
N
8a
10a-n
9a
Scheme 1. Reagents and conditions: (a) i—i-PrNH2/60 °C, 16 h, 73%, 37% CH2O/0 °C to rt, 0.5 h, then MeOH, 62%; (b) trans-ArCH@CHCOOMe/
CH2Cl2/0 °C to rt, 4 h, various yields; (c) NaOH/MeOH/H2O/rt, 8 h, >90%; (d) i—HCl/MeOH/rt, 1 h; ii—Me2NCH2CH2COOH/EDC/HOBt/Et3N/
CH2Cl2/rt, 8 h, TFA/CH2Cl2/rt, 1 h, 65%; (e) 7/EDC/HOBt/Et3/N/CH2Cl2/rt, 16 h, 5–80%.

C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
131
O
O
R1
N
NH
N
N
N
e or f
N
N
NH
NH
HO
O
O
O
O
O
13a: R1 = iPr
N
N
Cl
11 (trans
)
Cl
12
Cl
O
13b: R1 = THP
14a: R1 = MeCO
14b: R1 = EtCO
a,b,c,d,c
b
NH
N
N
c,d,c
g,a
N
O
N
N
N
NH
NH
S
N
S
O
O
NH2
O
O
O
8b
15
16
Cl
N
h
i,c
Cl
O
O
N
N
N
NH
N
N
N
O
N
N
O
NH
R2
N
O
O
N
O
Cl
cis-13a
17
Cl
Cl
18a-e
Scheme 2. Reagents and conditions: (a) HCl/MeOH/rt, 1 h; (b) Me2NCH2CH2COOH/EDC/HOBt/Et3N/DMF/rt, 16 h, ꢀ60%; (c) TFA/CH2Cl2/rt,
1 h, >90%; (d) 11/EDC/HOBt/Et3N/DMF/CH2Cl2/rt, 16 h, 60–88%; (e) acetone/NaBH(OAc)3/CH2Cl2/rt, 1 h, 64%; (f) MeCOCl or EtCOCl/Et3N/
CH2Cl2/rt, 1 h, 90%; (g) tetrahydropyran-4-one/NaBH(OAc)3/AcOH/CH2Cl2/rt, 8 h, 36%; (h) CH2O/NaBH(OAc)3/CH2Cl2/rt, 2 h, 51%; (i)
R2COOH/EDC/HOBt/Et3N/CH2Cl2/rt, 1–8 h, 20–60%.
For the 4-arylpyrrolidines 10a–n, the 4-chlorophenyl
compound 10a possessed good binding affinity
(Ki = 5.3 nM) and high agonist potency (EC50 = 20 nM)
with full intrinsic activity (IA = 103%) compared to the
internal standard a-MSH. The 3-chlorophenyl analog
10b (Ki = 370 nM) was 70-fold less potent in binding
affinity compared to 10a, and the 3-methoxyphenyl
10c was 10-fold less potent than 10b. The 2,4-dichloro-
phenyl 10f had a comparable Ki value to 10a, while
the 2-fluoro-4-trifluoromethylphenyl 10g was about 10-
fold less potent. However, 10f displayed very low effi-
cacy (IA = 24%), even at a 10 lM concentration. In a
functional antagonist assay, 10f dose-dependently inhib-
ited a-MSH-stimulated cAMP production with an IC50
of 560 nM. The regio-isomer of 10f (10k, Ki = 51 nM,
EC50 = 340 nM, IA = 47%) had moderate binding affin-
ity but was a partial agonist, while the 3-chloro-4-tri-
and agonist potency very similar to the 6-fluoro analog
10a. Its des-isopropyl analog (12, Ki = 7.2 nM,
EC50 = 79 nM, IA = 72%) possessed a slightly reduced
binding affinity, agonist potency, and efficacy compared
to 13a, indicative of a role of this small but lipophilic N-
isopropyl group. cis-13a (Ki = 90 nM) possessed much
lower binding affinity than its trans-isomer 13a, suggest-
ing a stereo-preference of the 4-chlorophenyl group. The
N-tetrahydropyran 13b (Ki = 1.4 nM, EC50 = 10 nM,
IA = 121%) exhibited high binding affinity as well as
functional activity. The N-acyl pyrrolidines 14a and
14b also showed high binding affinity and good agonist
potency, indicative of a minimal role of the basic nitro-
gen in 13 (Table 2).
We have previously shown that the diastereoisomers 16-
1 and 16-2 were a functional agonist and antagonist,
respectively (Table 3).11 Similar to this pair, the individ-
ual stereoisomers 13b-1 and 13b-2 displayed similar
binding affinities but different functional activities.
Thus, 13b-1 (Ki = 1.0 nM) was a potent functional ago-
nist with an EC50 of 3.8 nM and intrinsic activity of
122%. In contrast, 13b-2 (Ki = 4.7 nM) was only able
to stimulate cAMP accumulation to about 10% of the
fluoromethylphenyl
analog
10j
(Ki = 260 nM,
EC50 = 410 nM, IA = 78%) was a moderately potent
agonist. All other analogs 10 showed poor binding affin-
ity, including the 3-furyl compound 10n (Table 1).
The 4-methylphenylpiperazine 13a (Ki = 4.7 nM,
EC50 = 16 nM, IA = 102%) displayed binding affinity

132
C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
Table 1. Binding affinity of various trans-4-arylpyrrolidine-3-carbox-
amides 10a–na
While the binding affinity of the agonist 13b-1 was
about 5-fold better than that of the antagonist 13b-
2, 16-1 and 16-2 exhibited similar affinity. Unlike
13b (IA = 121%), the mixture of the trans-isomers 16
only displayed partial intrinsic activity (IA = 36%).
Its N,N-dimethyl analog 17 (Ki = 5.3 nM, IA = 35%)
also showed partial agonism. Other derivatives of 16
with an additional amino acid side-chain possessed
high binding affinity and high intrinsic activity (com-
pounds 18a–c), but the two alanine derivatives 18d
and 18e were partial agonists with high potency (Ta-
ble 3). The factor contributing to full or partial recep-
tor activation for these trans-isomers 18a–e is unclear.
The single isomer agonist 13b-1 was 11-fold more po-
tent than the agonist 16-1, suggesting that the amino-
propionyl side-chain strengthens the interaction of the
agonist 13b-1 with the receptor.
F
N
N
N
NH
Ar
O
O
N
10a-n
b
EC50 (nM)
Compound
Ar
Ki (nM)
10a
10b
10c
10d
10e
10f
10g
10h
10i
4-ClPhC6H4
3-ClC6H4
5.3
20 (103%)
370
3700
470
560
61
3-MeOC6H4
3-MeC6H4
3-FPhC6H4
In competition binding assays, both 13b-1 and 13b-2
were highly selective for hMC4R over other melanocor-
tin receptor subtypes (Table 4). These two compounds
were also characterized for their pharmacokinetic prop-
erties in rats. After an intravenous injection of 2.5 mg/
kg, 13b-1 displayed a high plasma clearance of
129 mL/min kg and a very high volume of distribution
(Vd = 80 L/kg), resulting in a long half-life (t1/2 = 7.1 h)
in this species (Table 4).
2-F,4-CF3C6H3
2,4-Cl2C6H3
2,5-F2C6H3
3,4-F2C6H3
3-Cl,4-FC6H3
3-F,4-CF3C6H3
3,5-F2C6H3
3,4,5-F3C6H2
3-Furyl
7.5
(24%)c
2050
290
260
51
10j
410 (78%)
340 (47%)
10k
10l
7800
250
850
10m
10n
a Data are average of two or more independent measurements.
b Intrinsic activity relative to a-MSH reported in parentheses.
c Dose-dependent inhibition of a-MSH-stimulated cAMP release with
IC50 = 560 nM.
After an oral gavage at a 10 mg/kg dose, the maximal
plasma concentration (Cmax) of 14 ng/mL appeared at
2.7 h, and the AUC value was 114 ng/mL h. The oral
bioavailability was 10%. Despite its high tissue distribu-
tion, which might be related to its dibasic structure, 13b-
1 had delayed brain penetration. The whole brain con-
centration was below the limit of detection at the 1-h
time point post po dosing and was only 19 ng/g at the
4-h time point. This resulted in a brain/plasma ratio of
3.0 due to a very low plasma concentration. The PK
parameters of 13b-2 were very similar to those of its ste-
reoisomer 13b-1 (Table 5).
Table 2. SAR of the N-group of pyrrolidine derivatives 13–14 at
hMC4Ra
R1
N
N
N
NH
O
In an in vitro Caco-2 assay, 13b-1 had a Papp of 1.6 nm/s
from the apical (a) to basolateral (b) direction and a (b
to a)/(a to b) ratio of 23, demonstrating high efflux
which might be associated with P-glycoprotein trans-
porter activity. Compound 13b-1 had a measured logD
value of 2.6 using a shake-flask method. Its high molec-
ular weight of 638 and an efflux mechanism are probable
causes for its limited brain penetration.
O
N
12-14
Cl
b
EC50 (nM)
Compound
R1
Ki (nM)
12
13a
H
7.2
4.7
90
79 (72%)
16 (102%)
i-Pr
i-Pr
c
cis-13a
13b
14a
4-THP
MeCO
EtCO
1.4
7.6
4.6
10 (121%)
44 (100%)
27 (129%)
The potent and selective agonist 13b-1 was also evalu-
ated in vivo. The effect of this compound on food intake
over 24 h was first tested in DIO rats that were food de-
prived overnight and on the test day. One hour after
administration of 13b-1 at 3, 10 or 30 mg/kg via intra-
peritoneal (ip) injection with vehicle (sterile water), food
was returned and intake measured at defined intervals
over a 24-h period. Food intake was significantly sup-
pressed compared to vehicle-treated rats at all time
points and all doses (Fig. 3, p < 0.05). In this study, fen-
fluramine was used as a positive control and it signifi-
cantly suppressed feeding at the measured time points
as expected.
14b
a Data are average of two or more independent measurements.
b Intrinsic activity relative to a-MSH in cAMP stimulation.
c Not determined.
a-MSH level at 10 lM concentration. In the functional
antagonist assay, 13b-2 dose-dependently inhibited a-
MSH-stimulated cAMP accumulation with an IC50 of
64 nM (Fig. 2).13

C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
Table 3. SAR of the N-group at the benzylamine derivatives 13 and 17–18a
133
O
N
N
N
N
R"
R'
O
Cl
13, 16-18
b
EC50 (nM)
Compound
R0
H
R00
Ki (nM)
16
16-1 (3S,4R)
H
12
11
(36%)
24 (104%)
(18%)c
(35%)
16-2 (3R,4S)
17
8.6
5.3
1.4
1.0
4.7
2.2
2.0
2.1
7.3
1.0
Me
COCH2CH2NMe2
Me
H
13b
13b-1 (3S,4R)
13b-2 (3R,4S)
10 (121%)
3.8 (122%)
(10%)d
18a
18b
18c
18d
18e
COCH2CH2NH2
COCH2NH2
H
H
H
H
H
6.0 (107%)
9.1 (114%)
19 (84%)
17 (56%)
5.2 (38%)
COCH2NHMe
R-COCH(Me)NH2
S-COCH(Me)NH2
a Data are average of two or more independent measurements.
b Intrinsic activity relative to a-MSH is indicated in parentheses.
c Dose-dependent inhibition of a-MSH-stimulated cAMP release with IC50 of 65 nM.
d Dose-dependent inhibition of a-MSH-stimulated cAMP release with an IC50 of 64 nM.
O
O
N
N
N
NH
N
S
S
N
N
S
R
NH
R
S
O
O
O
O
N
N
Cl
Cl
13b-1 (S,S,R-isomer):
Ki = 1.0 nM, EC50 = 3.8 nM, IA = 122%
13b-2 (S,R,S-isomer):
Ki = 4.7 nM, IA = 10%, IC50 = 64 nM
α-MSH
110
100
90
80
70
60
50
40
30
20
10
0
100
13b-
1
2
90
80
70
60
50
40
30
20
10
0
13b-
-10
-9
-8
-7
-6
-5
-10
-9
-8
-7
-6
-5
log [ ligand] (M)
log [13b-2] (M)
Figure 2. Chemical structures of 13b-1 and 13b-2 and their functional activity in cAMP assays.
Based on the robust effects observed with ip administra-
tion, the effects of oral (po) administration of 13b-1 were
also examined in the same model. Food intake following
oral administration of 13b-1 was suppressed by only

134
C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
Table 4. Binding affinity (Ki, nM) of 13b-1 and 13b-2 at the
melanocortin receptorsa
30 mg/kg, the highest dose tested (and fenfluramine) at
all the time points tested (p < 0.05).
Compound
MC1R
MC3R
MC4R
MC5R
Given the positive effects on acute feeding behavior after
a single oral dose of 13b-1, we next evaluated how
chronic administration of this selective MC4 agonist
would alter body weight and other related parameters
in the DIO rat.14 Figure 4 depicts the effects of daily
13b-1 treatment on food intake and body weight. Both
fenfluramine and 30 mg/kg of 13b-1 significantly sup-
pressed food intake on all test days. The 10 mg/kg dose
of 13b-1 inhibited feeding on day 1, 2, and 6 while the
3 mg/kg dose inhibited feeding on day 6. Compared to
vehicle-treated controls, body weight was also signifi-
cantly reduced with fenfluramine treatment on all days
and with 30 mg/kg of 13b-1 on days 2–7.
13b-1
13b-2
(38%)b
120
260
1300
1.0
4.7
800
530
a Data are average of two or more independent measurements.
b Percentage inhibition at 10 lM concentration.
Table 5. Pharmacokinetic parameters of 13b-1 and 13b-2 in ratsa
Compound
13b-1
13b-2
iv dose (mg/kg)
CL (mL/min kg)
Vd (L/kg)
2.5
129
80
2.5
82
75
11
t1/2 (h)
AUC (ng/mL h)
7.1
307
455
After 7 days of 13b-1 administration, body composition
(Rat EchoMRI, EchoMRI, TX) and fasting glucose
(OneTouch Ultra, LifeScan, CA) were measured. Rela-
tive to baseline measurements, compound 13b-1 pro-
duced a dose-dependent increase in fat loss (Fig. 5)
but did not significantly affect lean mass. These body
composition results demonstrate that body weight loss
due to 13b-1 treatment occurred solely as a result of
fat loss, while lean mass was spared. Furthermore,
po dose (mg/kg)
Cmax (ng/mL)
Tmax (h)
10
14
10
17
2.7
2.0
AUC (ng/mL h)
F (%)
114
10
161
14
Cbrain (ng/g) at 1, 4 h
Cbrain/Cplasma
0, 19
0, 3.0
0, 18
0, 1.0
a Data are average of three animals.
Vehicle
35
10
9
8
7
6
5
4
3
2
3 mg/kg
10 mg/kg
30
25
20
15
10
5
30 mg/kg
*
Fenfluramine
*
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
1
24
1
2
4
6
Time (hours)
Figure 3. Effect of 13b-1 (ip) on cumulative food intake in 24-h food deprived male DIO rats.
qd PO
2
0
25
20
15
10
5
qd PO
-2
-4
Vehicle
3 mg/kg
-6
10 mg/kg
30 mg/kg
Fenfluramine
-8
-10
0
-4
-2
0
2
4
6
8
-4
-2
0
2
4
6
8
Experimental Day
Experimental Day
Figure 4. Effect of 13b-1 (po) administered once a day for 7 days on food intake and body weight change in male DIO rats.

C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
135
0
-5
10
5
-10
-15
-20
-25
-30
-35
*
*
0
*
-5
-10
*
*
vehicle
3
10
30
Fen
vehicle
3
10
13b-1 (mg/kg, PO)
Figure 5. Effect of 7 days of oral administration of 13b-1 on fat and lean mass change in male DIO rats (*p < 0.05).
30
Fen
13b-1 (mg/kg, PO)
13b-1 administration seemed to positively affect glucose
regulation. Fasting glucose was slightly but significantly
decreased in the DIO rats given the 30 mg/kg dose of
13b-1 (Fig. 6). It is unlikely that this effect is simply a
correlate of weight loss considering the fenfluramine-
treated animals lost more weight but did not show de-
creased blood glucose levels.
In the chronic study, the plasma concentrations of the
compound were also monitored in a satellite group of
DIO rats, and the results are shown in Figure 7. On
day 1, plasma concentrations were relatively propor-
tional to the dose (3, 10, and 30 mg/kg). On day 7, the
plasma exposure was increased by about 5-fold com-
pared to day 1, suggesting accumulation of this com-
pound. The brain/plasma ratio also increased from
about 4 to over 10. This phenomenon was probably
associated with its high volume of distribution in rats.
120
100
*
80
60
40
20
0
In conclusion, we have studied a series of trans-4-aryl-
pyrrolidine-3-carboxamides as potent MC4R agonists.
Several potent compounds were identified. It was ob-
served that the 3S,4R-isomer of 13b-1 was able to acti-
vate the MC4 receptor with high potency. In contrast,
its 3R,4S-isomer 13b-2 possessed poor intrinsic activity
in the cAMP-stimulation assay and functioned as an
antagonist in the cAMP-inhibition assay. The potent
and selective agonist 13b-1 was characterized in vivo
and demonstrated efficacy in a dose-dependent manner
in a chronic DIO study.
vehicle
3
10
30
Fen
13b-1 (mg/kg, PO)
Figure 6. Effect of 7 days of oral administration of 13b-1 (po) on
fasting blood glucose in male DIO rats (*p < 0.05).
3 mg/kg (day 1)
10 mg/kg (day 1)
30 mg/kg (day 1)
3 mg/kg (day 7)
10 mg/kg (day 7)
30 mg/kg (day 7)
References and notes
100
1. Goodfellow, V. S.; Saunders, J. Curr. Top. Med. Chem.
2003, 3, 855.
2. Cone, R. D. Nat. Neurosci. 2005, 8, 571.
3. Yang, Y. K. Obes. Rev. 2003, 4, 239.
4. For recent reviews, see (a) Nargund, R. P.; Strack, A. M.;
Fong, T. M. J. Med. Chem. 2006, 49, 4035; (b) Chen, C.
Prog. Med. Chem. 2007, 45, 111.
10
1
5. Ujjainwalla, F. In 230th ACS National Meeting, Wash-
ington, DC, U.S.A. 2005; MEDI 275.
0.0
0.5
1.0
2.0
4.0
24.0
6. Tucci, F. C. In 16th International Conference on Organic
Synthesis, June 11–15, Merida, Yucatan, Mexico, 2006.
7. Tran, J. A.; Chen, C. W.; Jiang, W.; Tucci, F. C.; Fleck, B.
A.; Marinkovic, D.; Arellano, M.; Chen, C. Bioorg. Med.
Chem. Lett. 2007, 17, 5165.
Time (hours)
Figure 7. Time–concentration curves for compound 13b-1 after oral
administration on day 1 and day 7 in the chronic feeding study in DIO
rats.

136
C. Chen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 129–136
8. (a) Bucsh, R. A.; Domagala, J. M.; Laborde, E.; Sesnie, J.
C. J. Med. Chem. 1993, 36, 4139; (b) Achini, R. Helv.
Chim. Acta 1981, 64, 2203.
9. Huck, B. R.; Llamas, L.; Robarge, M. J.; Dent, T. C.;
Song, J.; Hodnick, W. F.; Crumrine, C.; Stricker-Kron-
grad, A.; Harrington, J.; Brunden, K. R.; Bennani, Y. L.
Bioorg. Med. Chem. Lett. 2006, 16, 4130.
10. Jiang, W.; Chen, C.; Marinkovic, D.; Tran, J. A.; Chen, C.
W.; Arellano, L. M.; White, N. S.; Tucci, F. C. J. Org.
Chem. 2005, 70, 8924.
11. Jiang, W.; Tran, J. A.; Tucci, F. C.; Hoare, S. R.; Markison,
S.; Wen, J.; Chen, C. W.; Marinkovic, D.; Arellano, M.;
Fleck, B. A.; Madan, A.; Foster, A. C.; Saunders, J.; Chen,
C. Bioorg. Med. Chem. Lett. 2007, 17, 6546.
13. Functional efficacy of compound 13b-1 (agonist) and
compound 13b-2 (antagonist): accumulation of cAMP
production in CHO cells transfected with hMC4R was
measured and analyzed using an ELISA. Data are
from representative experiments performed 2–8 times
with similar results. (left) Compound 13b-1, compound
13b-2, and a-MSH stimulation of cAMP production,
where 100% cAMP is defined as the Emax of a-MSH-
stimulated cAMP production. (right) Compound 13b-2
antagonism of a-MSH-stimulated cAMP production,
where 100% cAMP is defined as stimulation of cAMP
production by 10 nM a-MSH (ꢀEC50) in the absence
of antagonist.
14. Separate groups of DIO rats were orally dosed once a day
(1 h prior to lights out) with either vehicle, fenfluramine, 3,
10, or 30 mg/kg of 13b-1. Food intake (corrected for
spillage) and body weight were measured daily.
12. Nickolls, S. A.; Cismowski, M. I.; Wang, X.; Wolff, M.;
Conlon, P. J.; Maki, R. A. J. Pharmacol. Exp. Ther. 2003,
304, 1217.