Optimization of imidazole amide derivatives as cannabinoid-1 receptor antagonists for the treatment of obesity

Article abstract of DOI:10.1016/j.bmcl.2007.03.011

Several imidazole-based cyclohexyl amides were identified as potent CB-1 antagonists, but they exhibited poor oral exposure in rodents. Incorporation of a hydroxyl moiety on the cyclohexyl ring provided a dramatic improvement in oral exposure, together with a ca. 10-fold decrease in potency. Further optimization provided the imidazole 2-hydroxy-cyclohexyl amide 45, which exhibited hCB-1 K_i = 3.7 nM, and caused significant appetite suppression and robust, dose-dependent reduction of body weight gain in industry-standard rat models.

Full text of DOI:10.1016/j.bmcl.2007.03.011

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2706–2711
Optimization of imidazole amide derivatives as cannabinoid-1
receptor antagonists for the treatment of obesity
Roger A. Smith,a,* Zahra Fathi,b Furahi Achebe,a Christiana Akuche,a Su-Ellen Brown,b
Soongyu Choi,a Jianmei Fan,a Susan Jenkins,c Harold C. E. Kluender,a Anish Konkar,b
Rico Lavoie,a Ronald Mays,c Jennifer Natoli,b Stephen J. O’Connor,a Astrid A. Ortiz,b
Ning Su,a Christy Taing,b Susan Tomlinson,a Theresa Tritto,b Gan Wang,a
Stephan-Nicholas Wirtz,d Wai Wong,a Xiao-Fan Yang,a
Shihong Yinga and Zhonghua Zhanga
aDepartment of Chemistry Research, Bayer HealthCare, Pharmaceuticals Division, West Haven, CT 06516, USA
bDepartment of Metabolic Disorders Research, Bayer HealthCare, Pharmaceuticals Division, West Haven, CT 06516, USA
cDepartment of Research Technologies, Bayer HealthCare, Pharmaceuticals Division, West Haven, CT 06516, USA
dBayer HealthCare AG, Pharma Research, D-42096 Wuppertal, Germany
Received 12 January 2007; revised 1 March 2007; accepted 1 March 2007
Available online 12 March 2007
Abstract—Several imidazole-based cyclohexyl amides were identified as potent CB-1 antagonists, but they exhibited poor oral expo-
sure in rodents. Incorporation of a hydroxyl moiety on the cyclohexyl ring provided a dramatic improvement in oral exposure,
together with a ca. 10-fold decrease in potency. Further optimization provided the imidazole 2-hydroxy-cyclohexyl amide 45, which
exhibited hCB-1 Ki = 3.7 nM, and caused significant appetite suppression and robust, dose-dependent reduction of body weight gain
in industry-standard rat models.
Ó 2007 Elsevier Ltd. All rights reserved.
Obesity and excessive body weight are now recognized
as serious health concerns, as these conditions are
associated with decreased life span and several medical
complications such as diabetes, hyperlipidemia, coro-
nary artery disease, osteoarthritis, and some cancers.1
Furthermore, with the prevalence of obesity increasing
rapidly and current therapies being considered largely
inadequate,2 obesity has been declared one of the
most significant health problems faced by mankind.3
During the last decade, antagonism of the cannabi-
noid type 1 receptor (CB-1) has been pursued as
a highly promising strategy for the treatment of
obesity.4 To date, one CB-1 antagonist, the 1,5-diaryl-
pyrazole hydrazide rimonabant (1, SR-141716, Sanofi-
Aventis), has been approved in the European Union
for the treatment of obese or overweight patients with
associated risk factors, such as type 2 diabetes or
dyslipidemia.5
Of the various structure classes reported as CB-1 antago-
nists, the majority incorporate a central core fragment
substituted by two aromatic rings and a hydrogen bond
donor/acceptor functionality such as a carboxamide
group.4 Indeed, the pyrazole core exemplified in 1 has
been effectively replaced with other 5-membered hetero-
cycles such as dihydropyrazole,6 triazole,7–9 thiazole,8,10
pyrrole,11,12 and imidazole.7,8,13–15 For example, the 1,2-
diaryl-imidazole hydrazide 2 was reported by researchers
at Neurocrine Biosciences7 and Solvay8 to exhibit human
CB-1 Ki values of 85 and 23 nM, respectively. In our
assay, hCB-1 Ki = 7.8 nM was determined for 2.14 The
related cyclohexyl amide 3 was investigated as an isosteric
analog by Neurocrine Biosciences and was found to have
somewhat superior potency7 (hCB-1 Ki = 3.9 nM was
obtained for 3 in our assay). In this report, we describe
our investigation and optimization of 1,2-diaryl-imida-
zole amides related to 3 as effective CB-1 antagonists.
Keywords: Cannabinoid; Antagonist; Obesity; Appetite suppressant.
*
Corresponding author at present address: Angiotech Pharmaceuti-
cals, Inc., 1618 Station Street, Vancouver, BC, Canada V6A-1B6.
Tel.: +1 604 221 6927; e-mail: rosmith@angio.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.011

R. A. Smith et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2706–2711
2707
of PS-DIEA (polystyrene-supported tertiary amine),
followed by reaction in situ with an amine R4NH2 to
provide the desired product 7 generally in high purity.18
O
N
N
N
H
N
In our in vitro hCB-1 (human CB-1) binding assay,19
the imidazole amide 3 was determined to have hCB-1
Ki = 3.9 nM; for comparison, the binding affinity ob-
served in our assay for 1 was Ki = 1.1 nM. Amide 3
was also found to function as an antagonist,21 with
hCB-1 Kb = 18 nM, and to be selective for CB-1 over
CB-2 (hCB-2 Ki = 240 nM). A variety of additional
imidazole cyclohexyl amides were investigated, and
several were found to exhibit hCB-1 Ki val-
ues < 10 nM (Table 1). Some of the observed struc-
ture–activity relationships (SAR) were similar to
those reported for pyrazole hydrazides related to 1,23
imidazole hydrazides related to 2,7,8 and constrained
pyrrolopyridinone analogs.24 The chlorine atom of
the 4-chlorophenyl group in 3 (R1 in Table 1) could
be replaced with other small substituents without sub-
stantial loss in potency (8–10). In contrast, replace-
ment of the 2,4-dichlorophenyl substituent in 3 (R2
in Table 1) with 2,5-dichlorophenyl caused a signifi-
cant decrease in binding affinity (11), and replacement
with unsubstituted phenyl resulted in a further marked
decrease in potency (12). The 2,4-dichlorophenyl
group could be effectively replaced with a 2,4-dimeth-
ylphenyl group (13), but not with a 2,4-difluorophenyl
group (14), presumably due to insufficient steric bulk
at the 2-position. Similar to the SAR reported for
constrained pyrrolopyridinone analogs,24 removal of
the para-chloro substituent in 3 to provide 15 resulted
in comparable or slightly improved CB-1 binding
affinity. Replacement of the ortho-chloro in 15 with
other substituents caused a decrease in potency (19–
21). Finally, replacement of the phenyl ring at R1 in
16 with a cyclohexyl ring as in 22, substitution on
the amide nitrogen in 15 with a methyl group (23),
and replacement of the R1 substituent in 15 with a
3-pyridinyl group (24) all resulted in significant de-
creases in binding affinity (Table 1).
Cl
Cl
Cl
1 rimonabant
(SR-141716, AcompliaTM
hCB-1 Ki = 5.6 nM,5d 25 nM,6 12 nM7
)
O
N
X
N
H
N
Cl
Cl
Cl
2, X=N: hCB-1 Ki = 85 nM,7 23 nM,8
7.8 nM (this report)
3, X=CH: hCB-1 Ki = 48 nM,7
3.9 nM (this report)
The 1,2-diaryl-imidazole amides such as
3 were
efficiently synthesized by the route summarized in
Scheme 1, which is similar to the methods described
previously.7,8 In our protocol,14 an aniline was reacted
with an aromatic nitrile in the presence of ethylmagne-
sium bromide to give the amidine intermediate 4, that
was then condensed with a bromopyruvate.16 The
resulting imidazole ester 5 was hydrolyzed to the corre-
sponding carboxylic acid 6, which was then coupled
with an amine building block through the use of any of
a variety of coupling agents, such as EDCI (N-ethyl-N0-
dimethylaminoethyl-carbodiimide). For effective and
versatile parallel syntheses, the carboxylic acid 6 was
converted to the corresponding acyl fluoride by
treatment with TFFH (fluoro-N,N,N0,N0-tetramethyl-
formamidinium hexafluorophosphate)17 in the presence
The most potent imidazole cyclohexyl amide com-
pounds were evaluated by pharmacokinetics screening
in male Wistar rats,25 and unfortunately all of these
3
R
O
H
N
Br
OEt
CN
NH
EtMgBr
O
NH
2
+
2
1
1
R
R
R
THF
1. K CO , THF
2
2
3
R
o
2. AcOH, 110
C
4
O
O
O
4
3
3
3
R
R
R
R
OEt
OH
N
H
4
R NH , DCE
2
KOH
MeOH
N
N
N
N
N
N
EDCI
or
1
1
1
R
R
R
2
2
2
R
R
R
TFFH, PS-DIEA
5
6
7
Scheme 1. Synthesis of 1,2-diaryl-imidazole amides 7.

2708
R. A. Smith et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2706–2711
Table 1. Binding affinities of imidazole cyclohexyl amides to the human CB-1 and CB-2 receptors
O
N
R3
N
N
R1
R2
R2
Compound
R1
R3
hCB-1 Kia (nM)
3
8
4-Cl–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,5-Cl2–Ph
Ph
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
H
3.9b
1.9
4.9
8.2
25
4-CH3–Ph
4-CH3O–Ph
4-F–Ph
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
4-Cl–Ph
4-Cl–Ph
130
7.2
36
2.2c
4.6
9.0
20
21d
39
4-Cl–Ph
4-Cl–Ph
4-Cl–Ph
2,4-Me2–Ph
2,4-F2–Ph
2-Cl–Ph
4-CH3–Ph
4-CH3O–Ph
4-F–Ph
2-Cl–Ph
2-Cl–Ph
2-Cl–Ph
2-Me–Ph
2-Et–Ph
4-Cl–Ph
4-Cl–Ph
4-Cl–Ph
2-MeO–Ph
2-Cl–Ph
2-Cl–Ph
95
17
4-Me-cyHex
4-Cl–Ph
3-Pyridinyl
34
1100
2-Cl–Ph
a hCB-1 = 1.1 nM was determined for 1 in our assay.
b hCB-1 Kb = 18 nM; hCB-2 Ki = 240 nM.
c hCB-1 Kb = 11 nM; hCB-2 Ki = 2300 nM.
d hCB-1 Kb = 7.3 nM; hCB-2 Ki > 1000 nM.
Table 2. Binding affinities of imidazole amino- and hydroxyl-substituted cyclohexyl amides to the human CB-1 and CB-2 receptors
O
N
2
R
H
N
N
Cl
R2
1
R
Compound
R1
Cyclohexane isomer
hCB-1 Ki (nM)
25
26
27
28
29
30
31
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2,4-Cl2–Ph
2-Cl–Ph
NH2
NH2
S,S-trans
R,R-trans
S,S-trans
R,R-trans
trans–rac
S,S-trans
R,R-trans
270
940
620
920
22a
29b
97c
NHCH3
NHCH3
OH
OH
OH
2-Cl–Ph
a hCB-1 Kb = 4.4 nM; hCB-2 Ki = 1100 nM.
b hCB-1 Kb = 7.1 nM; hCB-2 Ki = 5300 nM.
c hCB-1 Kb = 22 nM.
compounds were found to provide quite poor plasma
exposure from oral dosing (10 mg/kg po, typically in
30% cyclodextrin suspension). For example, imidazole
amide 3 exhibited Cmax ꢀ 50 nM. Consistent with this
finding, 3 was also inactive in the fasted–refed rat
model for appetite suppression26 (10 mg/kg po, 30%
CD). In contrast, the corresponding imidazole hydra-
zide 2 exhibited plasma Cmax = 210 nM and caused
50 to 70% reduction in food intake at 0.5–4.0 h time
points in the fasted–refed rat model (10 mg/kg po,
30% CD).
cyclohexyl ring might improve the plasma exposure
of this series of compounds. However, incorporation
of either an amino or methylamino moiety was found
to give a substantial reduction in CB-1 binding affinity
(25–28, Table 2). Incorporation of a hydroxyl group
was less detrimental, causing approximately a 10-fold
decrease in potency (29 vs 3, 30 vs 15). The (S,S)-trans
isomers of these derivatives were found to be some-
what more potent than the (R,R)-trans isomers (Table
2). From Wistar rat pharmacokinetics screening, the
2-hydroxy-cyclohexyl amide 30 was found to provide
dramatically improved plasma exposure levels, with
We considered that the introduction of a basic nitro-
gen or polar moiety such as a hydroxyl group to the
Cmax = 1.64 lM
(10 mg/kg
po,
methanesulfonic acid (80:20) suspension (PEG/MSA)).
PEG400/25 mM

R. A. Smith et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2706–2711
2709
Brain exposure levels for 30 at 2 h were determined to
be 110 nM. Consistent with these results, compound 30
was found to have a significant effect in the fasted–re-
fed rat model, providing 34–62% reduction in food in-
take measured at 0.5–4.0 h time points (10 mg/kg po).
Likewise, 29 was active in this model, causing 21–
29% reduction in food intake at 0.5–4.0 h time points
(10 mg/kg po).
(26–56%) in food intake in the fasted–refed rat model.
Fusion of an aryl ring to the cyclopentyl group resulted
in significantly improved CB-1 binding affinity for ana-
log 35; however, this compound and other promising
aryl alkyl compounds were found to exhibit low plasma
exposure and/or insufficient activity in the fasted–refed
rat model to warrant further study.
Finally, alkyl substitutions were incorporated at the
imidazole 5-position. Introduction of methyl to n-pro-
pyl groups provided single-digit nanomolar hCB-1 Ki
values (39–41), while larger substituents were less effec-
tive (Table 4). Finally, replacement of the 4-chlorophenyl
group with a 4-bromophenyl group provided an appar-
ent further improvement in potency (45, hCB-1 Ki =
3.7 nM).
A diverse variety of additional hydroxyl-substituted
amides was therefore explored,14 with the goal of achiev-
ing greater CB-1 binding affinity while maintaining sig-
nificant plasma (and brain) exposure. Selected
examples of these derivatives are listed in Table 3. The
2-hydroxy-cyclopentyl amide 32 was found to be equi-
potent to the related cyclohexyl analog 30, it exhibited
similar exposure levels (plasma Cmax = 2.11 lM, brain
C(2h) = 330 nM; 10 mg/kg po in PEG/MSA) in PK
screening studies, and it caused a significant reduction
While compound 45 was found to be equipotent in the
hCB-1 binding assay to the cyclohexyl amide lead 3, it
exhibited considerably greater rat plasma exposure
(Cmax = 2.10 lM, 10 mg/kg po in PEG/MSA) and ro-
bust activity in the fasted–refed Wistar rat model. In this
study, the percent reduction in food intake caused by 45
(10 mg/kg po in PEG/MSA) at 0.5, 1.0, 1.5, 3.0, and
4.0 h time points was 63, 64, 62, 52, and 35%,
respectively.26
Table 3. Binding affinities of imidazole hydroxyl-substituted alkyl and
arylalkyl amides to the human CB-1 and CB-2 receptors
O
N
R
N
H
N
Cl
Cl
The genetically obese Zucker fa/fa rat has been used for
evaluating compound efficacy for the reduction of body
weight, including compounds such as rimonabant (1)
that have been shown to be effective in the management
of body weight in obese humans.27 Following the deter-
mination that 45 causes a significant suppression of
appetite, we also investigated its effect in this Zucker
rat model.28
Compound –NHR
Cyclopentane isomer hCB-1
Ki (nM)
32
33
S,S-trans
R,R-trans
29a
75
N
H
OH
OH
N
H
In the Zucker rat model, imidazole amide 45 was dosed
at 1, 5, and 10 mg/kg qd po, and was observed to cause a
significant and dose-dependent reduction in body weight
gain as compared to vehicle-treated rats (Fig. 1). The
effect on day 13 was ꢁ1.9% (not statistically significant),
ꢁ6.1%, and ꢁ10.6% at 1, 5, and 10 mg/kg, respectively.
34
35
36
trans–rac
S,R-cis
22
N
H
OH
OH
OH
Table 4. Binding affinities of imidazole 2-hydroxy-cyclohexyl amides
to the human CB-1 and CB-2 receptors
6.3b
N
H
O
R
N
H
OH
S,S
N
N
R,S-cis
18c
(
)
X
Cl
N
H
Compound
R
X
hCB-1 Ki (nM)
39
40
41
42
43
44
45
Me
Et
Cl
Cl
Cl
Cl
Cl
Cl
Br
6.9
5.0a
5.7b
10
37
38
trans–rac
cis–rac
41
17
N
H
OH
OH
nPr
nBu
iPr
Br
18
19
3.7c
N
H
Et
a hCB-1 Kb = 9.7 nM; hCB-2 Ki = 6300 nM.
b hCB-1 Kb = 2.2 nM; hCB-2 Ki = 1200 nM.
c hCB-1 Kb = 9.2 nM; hCB-2 Ki = 5100 nM.
a hCB-1 Kb = 0.9 nM; hCB-2 Ki = 2300 nM.
b hCB-1 Kb = 1.4 nM; hCB-2 Ki = 2700 nM.
c hCB-1 Kb = 13 nM; hCB-2 Ki = 2700 nM.

2710
R. A. Smith et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2706–2711
2. Stein, C. J.; Colditz, G. A. J. Clin. Endocrinol. Metab.
2004, 89, 2522.
Vehicle (PEG/MSA)
14
12
10
8
45, 1 mg/kg qd po
45, 5 mg/kg qd po
45, 10 mg/kg qd po
3. (a) Centers for Disease Control website, http:/
www.cdc.gov/nccdphp/dnpa/obesity; (b) Flier, J. S. Cell
2004, 116, 337.
4. For recent reviews, see (a) Muccioli, G. G.; Lambert, D.
M. Expert Opin. Ther. Pat. 2006, 16, 1405; (b) Antel, J.;
Gregory, P. C.; Nordheim, U. J. Med. Chem. 2006, 49,
4008; (c) Lange, J. H. M.; Kruse, C. G. Drug Discovery
Today 2005, 10, 693; (d) Smith, R. A.; Fathi, Z. IDrugs
2005, 8, 53; (e) Hertzog, D. L. Expert Opin. Ther. Pat.
2004, 14, 1435; (f) Lange, J. H. M.; Kruse, C. G. Curr.
Opin. Drug Discov. Dev. 2004, 7, 498.
6
4
2
0
5. (a) Pharma Marketletter, June 23, 2006; (b) Scrip, June 26,
2006; (c) Fernandez, J. R.; Allison, D. B. Curr. Opin.
Invest. Drugs 2004, 5, 430; (d) Rinaldi-Carmona, M.;
Barth, F.; Heaulme, M.; Shire, D.; Calandra, B.; Congy,
C.; Martinez, S.; Maruani, J.; Neliat, G.; Caput, D.;
Ferrara, P.; Soubrie, P.; Breliere, J. C.; Le Fur, G. FEBS
Lett. 1994, 350, 240.
-2
-4
-6
all points @ 5 & 10 mg/kg: p ≤ 0.05
0
2
4
6
8
10
12
Days of Treatment
Figure 1. Effect of the imidazole 2-hydroxycyclohexyl amide 45 on
body weight in genetically-obese Zucker fa/fa rats, upon dosing at 1, 5,
and 10 mg/kg po qd as a suspension in PEG/20 mM methanesulfonic
acid (80:20).
6. Lange, J. H.; Coolen, H. K.; van Stuivenberg, H. H.;
Dijksman, J. A.; Herremans, A. H.; Ronken, E.; Keizer,
H. G.; Tipker, K.; McCreary, A. C.; Veerman, W.; Wals,
H. C.; Stork, B.; Verveer, P. C.; den Hartog, A. P.; de
Jong, N. M. J.; Adolfs, T. J. P.; Hoogendoorn, J.; Kruse,
C. G. J. Med. Chem. 2004, 47, 627.
7. Dyck, B.; Goodfellow, V. S.; Phillips, T.; Grey, J.;
Haddach, M.; Rowbottom, M.; Naeve, G. S.; Brown,
B.; Saunders, J. Bioorg. Med. Chem. Lett. 2004, 14, 1151.
8. Lange, J. H.; van Stuivenberg, H. H.; Coolen, H. K.;
Adolfs, T. J.; McCreary, A. C.; Keizer, H. G.; Wals, H. C.;
Veerman, W.; Borst, A. J.; de Looff, W.; Vermeer, P. C.;
Kruse, C. G. J. Med. Chem. 2005, 48, 1823.
In a separate study, 45 and rimonabant (1) were both
dosed at 10 mg/kg qd po for eight days, and the com-
pounds were observed to cause 5.6% and 6.5% reduction
in body weight gain, respectively. The difference in the
effects observed for 45 and 1 in this study was not statis-
tically significant.
9. Jagerovic, N.; Hernandez-Folgado, L.; Alkorta, I.; Goya,
P.; Navarro, M.; Serrano, A.; de Fonseca, F. R.; Dannert,
M. T.; Alsasua, A.; Suardiaz, M.; Pascual, D.; Martin, M.
I. J. Med. Chem. 2004, 47, 2939.
10. Berggren, A. I. K.; Bostrom, S. J.; Elebring, S. T.;
Fallefors, L.; Wilstermann, J. M.; Greasley, P. PCT Int.
Appl. WO2004058255, 2004.
11. Smith, R. A.; Kluender, H. C.; Su, N.; Lavoie, R.; Fan,
J. PCT Int. Appl. WO2003027069, 2003.
12. Mayweg, A.; Marty, H. P.; Mueller, W.; Narquizian, R.;
Neidhart, W.; Pflieger, P.; Roever, S. PCT Int. Appl.
WO2004060870, 2004.
13. Plummer, C. W.; Finke, P. E.; Mills, S. G.; Wang, J.;
Tong, X.; Doss, G. A.; Fong, T. M.; Lao, J. Z.; Schaeffer,
M. T.; Chen, J.; Shen, C. P.; Stribling, D. S.; Shearman, L.
P.; Strack, A. M.; Van der Ploeg, L. H. T. Bioorg. Med.
Chem. Lett. 2005, 15, 1441.
14. Smith, R. A.; O’Connor, S. J.; Wirtz, S. N.; Wong, W. C.;
Choi, S.; Kluender, H. C.; Su, N.; Wang, G.; Achebe, F.;
Ying, S. PCT Int. Appl. WO2003040107, 2003.
In summary, imidazole amides were investigated and
optimized as CB-1 antagonists. SAR studies for a series
of cyclohexyl amides highlighted the importance of an
ortho substituent such as chloro or methyl on the phenyl
group at the imidazole 2-position. Although several very
potent examples of these analogs could be identified, they
lacked sufficient plasma exposure from oral dosing to
enable in vivo efficacy. Incorporation of a hydroxyl moi-
ety on the cyclohexyl ring provided a dramatic improve-
ment in oral exposure, together with a ca. 10-fold
decrease in in vitro potency. However, further optimiza-
tion of substituents provided the imidazole 2-hydroxy-
cyclohexyl amide 45, which exhibited hCB-1 Ki = 3.7 nM,
a significant anorexigenic effect in the fasted–refed Wistar
rat model, and robust, dose-dependent reduction in body
weight gain in the chronic Zucker rat model.
Acknowledgments
15. Hagmann, W. K.; Qi, H.; Shah, S. K. PCT Int. Appl.
WO2003063781, 2003.
16. Tsuji, K.; Nakamura, K.; Konishi, N.; Tojo, T.; Ochi, T.;
Senoh, H.; Matsuo, M. Chem. Pharm. Bull. 1997, 45, 987.
17. (a) Carpino, L. A.; El-Faham, A. J. Am. Chem. Soc. 1995,
117, 5401; (b) Carpino, L. A.; Ionescu, D.; El-Faham, A.
J. Org. Chem. 1996, 61, 2460.
18. General procedure for parallel syntheses: In a 20-mL
screw-cap vial, 0.5 mmol of 1,2-diaryl-imidazole-4-carbo-
xylic acid 6, 145 mg (0.55 mmol) TFFH (e.g., Advanced
Chemtech, Louisville, KY), and 5.0 equiv PS-DIEA
(loading level: 3.50 mmol/g, 716 mg, 2.5 mmol) (e.g.,
Argonaut Technologies Inc., San Carlos, CA) were heated
in 10 mL 1,2-dichloroethane at 35 °C overnight. The
formation of acyl fluoride was monitored by LC–MS.
To the mixture, 1.1 equiv (0.55 mmol) of amine building
We would like to thank Romulo Romero for assistance
with chromatographic separations, Anthony Paiva for
LC–MS analyses, and members of the Analytical Chem-
istry Group for NMR analyses. We are grateful to J.
Evenski, S. Hudgins, D. Kelly-Sullivan, L. Milardo,
and C. Ralphalides for assistance with in vivo pharma-
cology studies. We also thank R. Schoenleber, D. Camp-
bell, C. Mahle, and S. Gardell for helpful discussions.
References and notes
1. Bray, G. A. J. Clin. Endocrinol. Metab. 2004, 89, 2583.

R. A. Smith et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2706–2711
2711
block R4NH2 was added and the reaction was continued
at 35 °C overnight. The mixture was filtered through a
filter tube (polypropylene frit), and the filtrate was
evaporated under reduced pressure to provide the product,
typically in >70% purity. For compounds of particular
interest, the crude product was redissolved in 1 mL MeOH
and purified by preparative HPLC. The methodology was
also successfully applied to library synthesis (10 lmol
scale) by using 2-mL 96-well Robbins FlexChemTM reac-
tion blocks and Robbins rotator ovens.
final volume) for 20 min at 37 °C with various concentra-
tions of cannabinoid receptor agonist and forskolin
(3 lM, final concentration) in the presence or absence of
a fixed concentration of CB-1 receptor antagonist. The cell
suspension was centrifuged at 1000g for 5 min in a table
top centrifuge (4 °C). The supernatant was aspirated and
100 lL of cell lysis buffer added to each well. The cyclic
AMP content of the supernatant was measured by
scintillation proximity assay (cAMP SPA Biotrak Direct
Screening Assay System, Amersham Biosciences, UK)
according to the manufacturer’s instructions. Apparent
dissociation constants of antagonists (Kb values) were
calculated by using the formula pKb = log10(CR-
1) ꢁ log10B, where B is the concentration of antagonist
used and CR (concentration ratio) is the ratio of agonist
EC50 measured in the presence of antagonist over that
measured in the absence of antagonist.22
19. The hCB-1 and hCB-2 data reported in this article are the
average of two or more triplicate determinations for
purified and characterized (1H NMR, LC–MS) samples.
Cannabinoid receptor binding assays were performed with
cell membranes from human CB-1 or human CB-2
receptor expressing HEK 293 cells using [3H]CP 55940
as radioligand. Membrane pellets were suspended in ice-
cold binding buffer (50 mM Tris, pH 7.4, 2.5 mM EDTA,
5 mM MgCl2, 0.1% fatty acid free BSA) and immediately
used for determining protein content (Bio-Rad Assay) and
binding assays. Competition binding assays were per-
formed in triplicate by incubating cell membranes (corre-
sponding to 3.3 lg protein) with 0.3 nM [3H]CP 559440
and varying concentrations of competing compounds.
Reactions were carried out in a final volume of 200 lL
binding buffer in polypropylene plates with constant
shaking at 30 °C for 90 min. Non-specific binding was
determined in the presence of unlabeled 10 lM WIN
55212-2. The binding reaction was terminated by filtration
through pretreated (50 mM Tris, pH 7.4) Millipore GF/C
filter plates using a vacuum manifold. Filters were washed
seven times with ice cold 50 mM Tris (pH 7.4). Microscint
O (25 lL) was added to each well and radioactivity bound
to the filters was measured using a Wallac 1450 MicroBeta
Trilux liquid scintillation counter. All competition binding
and concentration–response curves were analyzed using
nonlinear regression with Prism software (GraphPad
Software, San Diego, CA). Ki values were calculated from
IC50 values according to the Cheng and Prusoff formula.20
20. Cheng, Y. C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
22, 3099.
21. Rimonabant has been demonstrated to behave as a CB-1
receptor inverse agonist, that is, it decreases the constitu-
tive level of CB-1 receptor activation in the absence of an
agonist (a) Bouaboula, M.; Perrachon, S.; Milligan, L.;
Canat, X.; Rinaldi-Carmona, M.; Portier, M.; Barth, F.;
Calandra, B.; Pecceu, F.; Lupker, J.; Maffrand, J.-P.; Le
Fur, G.; Casellas, P. J. Biol. Chem. 1997, 272, 22330; (b)
Recombinant cells expressing human CB-1 were used to
examine cell-based efficacy of compounds as antagonists,
that is, ability to block receptor agonist-stimulated inhi-
bition of forskolin-stimulated cAMP levels. Cannabinoid
receptor inhibition of adenylyl cyclase activity was deter-
mined in CHO K1 cells stably transfected with cDNA
encoding the human CB-1 receptor (Euroscreen). Cells
were maintained at 37 °C and 5% CO2 in Ham’s F12
medium supplemented with 20 mM HEPES (pH 7.4),
2 mM L-glutamine, 400 lg/mL G418, and 5% FBS. CB-1
receptor expressing CHO K1 cells grown to 90% conflu-
ence in 150 cm2 flasks were washed with 10 mL PBS.
Sigma cell dissociation solution (5 mL; Cat# C5914) was
added to flasks and the cells collected in Ham’s F12 media
and centrifuged for 5 min at 500g. The cell pellet was
resuspended in assay buffer (Dulbecco’s PBS containing
0.1 mM 3-isobutyl-1-methylxanthine and 0.1 mM Ro20-
17241). The cells (110,000/well) were incubated (250 lL,
22. Arunlakshana, O.; Schild, H. O. Br. J. Pharmacol. 1959,
14, 48.
23. Lan, R.; Liu, Q.; Fan, P.; Lin, S.; Fernando, S. R.;
McCallion, D.; Pertwee, R.; Makriyannis, A. J. Med.
Chem. 1999, 42, 769.
24. Smith, R. A.; Fathi, Z.; Brown, S.; Choi, S.; Fan, J.;
Jenkins, S.; Kluender, H. C. E.; Konkar, A.; Lavoie, R.;
Mays, R.; Natoli, J.; O’Connor, S. J.; Ortiz, A. A.;
Podlogar, B.; Taing, C.; Tomlinson, S.; Tritto, T.; Zhang,
Z. Bioorg. Med. Chem. Lett. 2007, 17, 673.
25. In pharmacokinetics screening studies, plasma exposure
levels were determined typically at 0.5, 1, and 2 h time
points following a 10 mg/kg oral dose to Wistar rats. As
the CB-1 receptor is a CNS target, brain exposure levels at
2 h after compound dosing were also determined for
selected compounds. However, insufficient data were
collected to permit an interpretation of the SAR for
brain/plasma exposure ratio.
26. Male Wistar rats were individually housed in suspended
cages with a mesh floor and were kept in standard animal
rooms under controlled temperature and humidity, and a
12-h/12-h light/dark cycle for a minimum of week before
the start of the experiment. Rats were fasted overnight
(18 h) and then dosed orally with vehicle or test compound
at 10 mg/kg in a volume of 2 mL/kg, 1 h before re-feeding.
Cumulative food intake was recorded 30, 60, 90, 180, and
240 min after the return of the pre-weighed food jar,
taking food spillage into account. Data were collected for
10 rats per treatment group.
27. Vickers, S. P.; Webster, L. J.; Wyatt, A.; Dourish, C. T.;
Kennett, G. A. Psychopharmacology 2003, 167, 103.
28. Genetically obese male Zucker fa/fa rats (average 550 g
body weight at start of study) were kept in standard
animal rooms under controlled temperature and humidity,
and a reversed 12-h/12-h light/dark cycle. Water and food
were continuously available. Rats were single-housed in
large rat shoeboxes containing grid floor. Animals were
adapted to the grid floors and sham dosed with vehicle
(PEG/25 mM methanesulfonic acid (80:20)) for at least
five days before the recording of two-days baseline
measurement of body weight and 24 h food and water
consumption. Using the baseline body weight data, rats
were weight-matched and assigned to the different treat-
ment groups. Rats received a daily oral dose of vehicle or
test compound in a volume of 2 mL/kg, before their
feeding phase (nocturnal cycle). Body weights were
recorded daily during the treatment period. Data shown
in Figure 1 represent means SEM for 9–10 rats per
treatment group.