Novel benzimidazole derivatives as selective CB2 agonists

Article abstract of DOI:10.1016/j.bmcl.2008.05.073

The preparation and evaluation of a novel class of CB2 agonists based on a benzimidazole moiety are reported. They showed binding affinities up to 1 nM towards the CB2 receptor with partial to full agonist potencies. They also demonstrated good to excellent selectivity (>1000-fold) over the CB1 receptor.

Full text of DOI:10.1016/j.bmcl.2008.05.073

Bioorganic & Medicinal Chemistry Letters 18 (2008) 3695–3700
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel benzimidazole derivatives as selective CB2 agonists
a
a
a
a
a
Daniel Pagé a, , Elise Balaux , Luc Boisvert , Ziping Liu , Claire Milburn , Maxime Tremblay ,
Zhongyong Wei a, Simon Woo a, Xuehong Luo a, Yun-Xing Cheng a, Hua Yang a, Sanjay Srivastava a, Fei Zhou a,
William Brown a, Miroslaw Tomaszewski a, Christopher Walpole a, Leila Hodzic b, Stéphane St-Onge b,
Claude Godbout b, Dominic Salois b, Keymal Payza b
*
a Department of Medicinal Chemistry, AstraZeneca R&D Montréal, 7171 Frederick-Banting, St-Laurent, Que., Canada H4S 1Z9
b Department of Molecular Pharmacology, AstraZeneca R&D Montréal, 7171 Frederick-Banting, St-Laurent, Que., Canada H4S 1Z9
a r t i c l e i n f o
a b s t r a c t
Article history:
The preparation and evaluation of a novel class of CB2 agonists based on a benzimidazole moiety are
reported. They showed binding affinities up to 1 nM towards the CB2 receptor with partial to full agonist
potencies. They also demonstrated good to excellent selectivity (>1000-fold) over the CB1 receptor.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 4 March 2008
Revised 16 May 2008
Accepted 16 May 2008
Available online 22 May 2008
Keywords:
Cannabinoids
Benzimidazoles
CB2 receptor
Agonists
The endocannabinoid system has been extensively studied be-
cause of its implication in many biological processes such as met-
abolic regulation,1,2 pain control,3–5 cellular proliferation6 and
many other CNS functions.7 These effects are mediated mainly
via two subtypes of cannabinoids receptors, located either in the
central nervous system (CB1) or in the peripheral tissues (CB2).
The extensive pharmacology of cannabinoid receptors have been
reviewed in the literature over the past few years.8–12 Huge
emphasis has been placed on the CB1 receptor due to its implica-
tion in many therapeutic applications. Selective CB1 antagonists
are currently in clinical studies for the treatment of obesity and
metabolic syndrome1,2,13–15 or different types of addictions,2,14,16
while CB1 agonists are being used for the treatment of multiple
sclerosis, cancer, neuropathic and inflammatory pain.17–19
Despite the great homology between the CB1 and CB2 recep-
tors,20 the exact physiological roles of the CB2 receptor still remain
to be fully defined. Recent studies have demonstrated that CB2
selective compounds were active in different neuropathic and
inflammatory pain models.21–25 Some neuroprotective roles have
also been associated with CB2 agents,26–29 that could lead to the
prevention of some neurodegenerative diseases. Other studies
have also highlighted potential roles for CB2 in cancer,30,31 multi-
ple sclerosis32 and bone regeneration.33,34 Furthermore, since the
majority of CB2 receptors are distributed in peripheral tissues, cen-
trally mediated side-effects would be greatly diminished with CB2
selective agents.
Several classes of natural or synthetic CB2 ligands have been re-
ported in the literature.35–42 In a high throughput CB1/CB2 screen-
ing campaign originated a few years ago at our company, several
hit compounds were obtained from a variety of chemical series.
This program enabled us to develop different selective CB2 ligands
such as 1,2,3,4-tetrahydropyrroloindole agonists43 or benzimid-
azole inverse agonists.44 One of our first hits obtained was shown
to be compound (1) (see Fig. 1). This benzimidazole derivative
demonstrated very good binding affinity towards the CB2 receptor
with decent selectivity over CB1 while showing partial agonist po-
tency. This molecule represented a good starting template for SAR
studies in order to increase the CB2 potency and CB1/CB2 selectiv-
ity of these new ligands. We report herein the synthesis and phar-
macological evaluation of this novel class of benzimidazole ligands.
O
N
N
Ki hCB2: 36nM
EC50 hCB2: 38nM (58%)
N
Ki hCB1/ Ki hCB2: >130
1
O
N
* Corresponding author. Tel.: +1 514 832 3200; fax: +1 514 832 3232.
E-mail address: daniel.page@astrazeneca.com (D. Pagé).
Figure 1. Initial benzimidazole hit ligand.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.05.073

3696
D. Pagé et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3695–3700
The synthetic strategy used was designed to maintain the benz-
tions of the p-ethoxy group into other ether functionalities were
investigated (Scheme 2). The compounds were prepared following
a similar procedure as described in Scheme 1. When the desired
para-substituted phenylacetic acids were not commercially avail-
able, they could be prepared by the alkylation of methyl (4-
hydroxyphenyl)acetate (12) with corresponding alkyl bromides,
followed by hydrolysis of the methyl ester. Amide coupling of
the acids (13a–j) with diamine intermediate (4j), followed by cycli-
zation under acidic conditions afforded compounds 14a–j.
In order to maximize the diversity on the left-hand side of the
molecule, two different templates were used. Modifications of
the diethyl amide could be done starting from 3-fluoro-4-nitro-
benzonitrile (15) (Scheme 3) following a similar synthetic route
as the one described above. p-Ethoxy pyridine derivatives (right-
hand side) were also prepared the same way in order to look at
the potential effect on the CB2 binding. Hydrolysis of the nitriles
(18a–c) under basic conditions into the corresponding acids
(19a–c), followed by HATU (N,N,N0,N0-tetramethyl-O-(7-azabenzo-
triazole-1-yl) uranium hexafluorophosphate) coupling using
different amines afforded final compounds (20a–j) in good yields
(70–90%).
imidazole core intact while modifying the lower side-chain, the
diethyl amide (left-hand side) and the p-ethoxyphenyl part
(right-hand side) of the molecule. The modifications were initially
performed on the lower side-chain of the molecule by introducing
various alkyl or aromatic substituents. Two different synthetic
routes were developed in order to produce the desired molecules
(Scheme 1). The first approach used a SnAr reaction involving the
fluoro-nitro derivative (2) with different amines in refluxing etha-
nol. Reduction of compounds (3a–j) by catalytic hydrogenation
afforded anilines (4a–j), which were then readily coupled with
(4-ethoxyphenyl)acetic acid using a standard amide coupling strat-
egy. The final cyclization under acidic conditions afforded the
desired benzimidazole products (9) in relatively good yields
(65–90%). Initial studies done at our site demonstrated that
a
-branched amines gave lower cyclization yields and the subse-
quent products gave poor binding results. In light of this, the focus
was placed on amines free of any -substituents. A more conver-
a
gent approach could also be used by introducing initially the
p-ethoxy moiety in the synthesis. N,N-Diethyl-3-fluoro-4-nitrob-
enzamide (5) was reacted with ammonium hydroxide to give the
corresponding anilino product (6). The coupling of this deactivated
aniline with (4-ethoxyphenyl)acetyl chloride could be performed
in the presence of zinc dust,45 which afforded the desired amide
(7) in a good yield (82%). Reduction of the nitro group gave inter-
mediate (8) that was alkylated with different aldehydes, followed
by the acidic cyclization to give the analogous products (9a–s).
The benzylic position could also be oxidized using MnO2 to the cor-
responding ketone (10) in 71% yield. Subsequent reduction of the
ketone afforded the corresponding alcohol (11).
It became evident throughout the initial screening campaign
that the presence of the p-ethoxy group on the phenyl ring, on
the right-hand side of the molecule, greatly enhanced the CB2
binding affinity for this class of compounds when compared to
any other tested para-, ortho- or meta-substituents. Transforma-
Diversity could also be created on the left-hand side starting
from N-(4-fluoro-3-nitro-phenyl)-acetamide (21) in order to intro-
duce different reversed-amide or urea functionalities (Scheme 4).
Following a similar synthetic procedure, the benzimidazole core
could be easily prepared. The N-acetyl amide benzimidazole (24)
could be alkylated under phase transfer catalysis conditions. Re-
moval of the acetate group followed by reaction with different acid
chlorides of isocyanates afforded the final compounds (26a–g) in
75–90% yields. All final products were purified by reversed-phase
chromatography with a water/acetonitrile gradient containing
0.05% TFA v/v and isolated as their corresponding TFA salts.46
The CB1/CB2 binding results are summarized in Tables 1–4.47
Only compounds that exhibited Ki hCB2 <100 nM were tested in
the hCB2 GTPc[
35S] assay.48 The mixed CB1/CB2 agonists
O
O
NO2
F
F
N
N
NO2
2
5
iv
i
O
O
NH2
NO2
X
N
N
R1
N
6
H
3a-j X = NO2
4a-j X = NH2
v
ii
O
H
N
iii
N
N
N
O
N
O
O
X
Y
7 X = NO2
8 X = NH2
ii
R1
vi
O
9a-s Y = CH2
10R1 = -CH2-CH(CH3)2, Y = C(O)
11R1 = -CH2-CH(CH3)2, Y = CH-OH
vii
viii
Scheme 1. Reagents and conditions: (i) R1CH2NH2, Et3N, EtOH, 75 °C, o/n; (ii) H2, 10% Pd/C, EtOAc, rt, o/n; (iii) a—(4-ethoxyphenyl)acetic acid, HATU, DIPEA, DMF, rt, 3 h;
b—DCE/HCl or glacial AcOH, 80 °C, 2 h; (iv) NH4OH, EtOH, 65 °C, o/n; (v) (4-ethoxyphenyl)acetyl chloride, zinc dust, toluene, rt, o/n; (vi) a—aldehyde, BH3-pyridine, DCE,
AcOH, rt, 1 h; b—DCE/HCl or glacial AcOH, 80 °C, 2 h; (vii) MnO2, dioxane, 65 °C, 24–48 h; (viii) NaBH4, EtOH, rt, 1 h.

D. Pagé et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3695–3700
3697
O
NH2
N
NH
O
4j
R2
O
O
O
N
N
O
i
N
iii, iv
OH
O
R1
12
13a-j R2 = Me
13a-j R2 = H
14a-j
ii
R1
O
Scheme 2. Reagents and conditions: (i) R1-Br, K2CO3, DMF, 100 °C, 8 h; (ii) LiOH, THF/H2O, rt, 3 h; (iii) 4j, HATU, DIPEA, DMF, rt, 3 h; (iv) DCE/HCl or glacial AcOH, 80 °C, 2 h.
NC
NO2
NH
NC
NC
NH2
NH
N
N
NO2
F
NC
i
ii
iii, iv
X
Y
16
17
18a-c
15
O
X, Y = C or N
v
O
O
R1
N
N
N
N
N
HO
vi
R2
X
X
Y
Y
19a-c
20a-j
O
O
Scheme 3. Reagents and conditions: (i) cyclopropylmethylamine, Et3N, EtOH, 75 °C, o/n; (ii) H2, 10% Pd/C, EtOAc, rt, o/n; (iii) acid, HATU, DIPEA, DMF, rt, 3 h; (iv) DCE/HCl or
glacial AcOH, 80 °C, 2 h; (v) 20% KOH, EtOH, reflux, o/n; (vi) amine, HATU, DIPEA, DMF, rt, 2 h.
H
H
H
NO2
NH
N
N
NH2
NH
NO2
F
N
ii
i
O
O
O
22
23
21
iii, iv
R1
N
R1
N
N
N
R2
N
vi, vii
O
N
26a-g
24R1 = H
25R1 = CH3
v
O
O
Scheme 4. Reagents and conditions: (i) cyclopropylmethylamine, Et3N, EtOH, 75 °C, o/n; (ii) H2, 10% Pd/C, EtOAc, rt, o/n; (iii) (4-ethoxyphenyl)acetic acid, HATU, DIPEA, DMF,
rt, 3 h; (iv) glacial AcOH, 80 °C, 2 h; (v) Bu4NBr, MeI, 50%KOH/ DCM, rt, o/n; (vi) 1 M HCl, EtOH, 120 °C, waves, 0.5 h; (vii) RC(O)Cl/ RNCO, Et3N, DCM, rt, 1 h.
l
WIN55212-249 and
D
9-THC were tested as standards for compari-
(9o) or N-oxides (9j) at the position 4 of the substituent rings were
not tolerated. Furthermore, the introduction of either a carbonyl
(10) or alcohol (11) group at the benzylic position was also detri-
mental for the CB2 binding affinity. Even though the nature of
the group seemed to exert an influence on the potency of the com-
pounds, with Emax values ranging from 55% to 93%, no specific
trend could be observed.
The results of the p-ethoxy group substitutions are reported in
Table 2. It seems that only the isopropoxy moiety (14c) gave
similar CB2 binding affinity (4.5 nM) than compound 9a. Smaller
(14a–b) and/or bigger substituents (14d–14i) had a detrimental ef-
fect on the CB2 binding, with the exception of the phenoxy
substituent (14j) which showed a comparable Ki (4.9 nM) to
son purposes (Table 1). The results from Table 1 suggest that a
wide variety of groups can be accommodated at the lower part of
the molecule, with a preference for alkyl substituents (9a–9f).
Increasing the bulk of the substituents also had the effect of
increasing the CB1 binding affinity. Noticeable discrepancies be-
tween the CB1 and CB2 binding affinities were observed when po-
lar atoms were introduced in the lower region of the molecule.
Hydrophilic substituents such as pyridines (9h–9i), furan (9k), tet-
rahydropyran (9l), tetrahydrofurans (9m–9n) and 2-substituted
piperidines (9r–9s) all showed good CB2 binding affinities while
being essentially devoid of any CB1 binding affinity. However,
charged groups at physiological pH, such as secondary amines

3698
D. Pagé et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3695–3700
Table 1
Binding results of benzimidazole derivatives with lower part modifications
O
N
N
N
Y
R1
O
*
R1
Y
hCB2 Ki (nM)
hCB1 Ki (nM)
hCB2 EC50 (nM)
D
9-THC
—
—
—
—
41
20
2
3
2.9 0.3
140 42
>5000
1.5 0.1 (100%)
14 3 (64%)
2.9 0.2 (63%)
2.1 0.7 (68%)
1.3 0.1 (64%)
0.7 0.2 (79%)
0.52 0.04 (79%)
1.3 0.2 (81%)
1.1 0.2 (88%)
25 5 (77%)
5.0 0.7 (78%)
nd
3.3 0.9 (78%)
1.0 0.3 (89%)
4.8 1.1 (76%)
8.3 1.9 (62%)
nd
WIN5512-2
9a
9b
9c
9d
9e
9f
9g
9h
9i
(CH3)2CH–CH2–
Cyclopropyl
Cyclobutyl
Cyclopentyl
Cyclohexyl
1-Adamantane
Phenyl
2-Pyridine
4-Pyridine
4-Pyridine-N-oxide
2-Furan
4-Tetrahydropyran
2-Tetrahydrofuran(S)
2-Tetrahydrofuran(R)
4-Piperidine
2-Piperidine(S)
2-Piperidine(R)
2-N-Me-piperidine(R)
2-N-Me-pyrrolidine(R)
(CH3)2CH–CH2–
(CH3)2CH–CH2–
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
C@O
4.5 0.9
4.1 0.6
1.6 0.1
1.0 0.1
3.7 1.3
2.8 0.2
12
39
16
588 191
11
3.9 0.9
15
23 4.5
2920 78
75
35
8.9 1.1
12
35
57
>5000
3115 149
491 27
110 1.9
406 44
4766 348
>5000
>5000
>5000
>5000
1209 53
>5000
>5000
>5000
>5000
>5000
4315 270
>5000
>5000
3
3
2
9j
9k
9l
3
9m
9n
9o
9p
9q
9r
9s
10
11
3
9
6
25 6 (77%)
13 2 (72%)
2.3 0.3 (81%)
3.4 0.7 (93%)
25 6 (55%)
69 18 (70%)
2
3
4
CH–OH
>5000
nd, not determined.
*
Emax are reported in brackets.
Table 2
Table 3
Binding results of benzimidazole derivatives with p-ethoxy group replacements
Binding results of benzimidazole derivatives with diethyl amide group replacements
O
O
R1
N
N
N
N
R2
N
N
X
Y
O
R1
O
*
R1
R2
X
Y
hCB2 Ki
(nM)
hCB1 Ki
(nM)
hCB2 EC50
(nM)
*
R1
hCB2 Ki (nM)
hCB1 Ki (nM)
hCB2 EC50 (nM)
9b
Ethyl
Ethyl
H
H
Methyl
Propyl
—
CH CH 4.1 0.6 >5000
2.1 0.7 (68%)
52 18 (53%)
nd
9a
Ethyl
H
Methyl
4.5 0.9
799 169
>5000
>5000
>5000
4679 190
>5000
>5000
>5000
>5000
>5000
2.9 0.2 (63%)
nd
20a Ethyl
20b
20c Propyl
20d Propyl
20e –(CH2)4–
CH CH 63
CH CH >5000
6
>5000
>5000
14a
14b
14c
14d
14e
14f
14g
14h
14i
14j
H
17
4
7.3 1.8 (43%)
2.4 0.6 (79%)
24 5 (73%)
18 1 (83%)
nd
11 2 (29%)
65 18 (68%)
43 8 (75%)
3.2 0.1 (66%)
CH CH 7.6 2.1 >5000
CH CH 2.8 0.6 2608 245 2.0 0.1 (67%)
3.2 0.4 (54%)
Isopropyl
Cyclobutyl
Cyclopentyl
Cyclohexyl
CF3
–CH2–CF3
–CH2-cyclopropyl
Phenyl
4.5 0.3
44 15
CH CH 14
CH CH 6.9 1.7 >5000
CH CH 11 >5000
2
>5000
6.2 1.2 (51%)
3.0 0.4 (45%)
4.2 0.4 (60%)
43
526 125
25
57 15
53
4.9 0.2
8
20f
–(CH2)5–
—
20g Cyclohexyl Methyl
20h CF3CH2–
20i
20j
5
7
CF3CH2– CH CH 5.6 1.0 4534 134 2.0 0.4 (51%)
Ethyl
Ethyl
Ethyl
Ethyl
N
CH 27
4
>5000
22 5 (63%)
nd
8
4190 149
2396 124
CH
N 133 25 >5000
nd, not determined.
nd, not determined.
*
Emax are reported in brackets.
*
Emax are reported in brackets.
were all well tolerated demonstrating probably the presence of a
large binding pocket at the receptor site. Reducing the size of the
group to a mono ethyl (20a) showed about a 10-fold decrease in
CB2 binding affinity, whereas having only a primary amide on
the left-hand side (20b) resulted in a total loss of activity. The
introduction of a nitrogen atom in the benzyl ring on the right-
compound 9a, probably coming through an additional p–p interac-
tion with the receptor binding pocket.
Most of the diethyl amide modifications reported in Table 3
showed CB2 Ki values in the same range as compound 9b. Linear
(20c–d), cyclic (20e–g) or fluorinated (20h) bis-alkyl substituents

D. Pagé et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3695–3700
3699
Table 4
20. Davis, M.; Maida, V.; Daeninck, P.; Pergolizzi, J. Support Care Cancer 2007, 15,
63.
Binding results of benzimidazole derivatives with left-hand side modifications
21. Whiteside, G. T.; Lee, G. P.; Valenzano, K. J. Curr. Med. Chem. 2007, 14, 917.
22. Ashton, J. C. Curr. Opin. Investig. Drugs 2007, 8, 373.
R1
23. Giblin, G. M.; O’Shaughnessy, C. T.; Naylor, A.; Mitchell, W. L.; Eatherton, A. J.;
Slingsby, B. P.; Rawlings, D. A.; Goldsmith, P.; Brown, A. J.; Haslam, C. P.;
Clayton, N. M.; Wilson, A. W.; Chessell, I. P.; Wittington, A. R.; Green, R. J. Med.
Chem. 2007, 50, 2597.
24. Guindon, J.; Hohmann, A. G. Br. J. Pharmacol. 2008, 153, 319.
25. Khanolkar, A. D.; Lu, D.; Ibrahim, M.; Duclos, R. I.; Thakur, G. A.; Malan, T. P.;
Porreca, F.; Veerappan, V.; Tian, X.; George, C.; Parrish, D. A.; Papahatjis, D. P.;
Makriyannis, A. J. Med. Chem. 2007, 50, 6493.
26. Fernandez-Ruiz, J.; Romero, J.; Velasco, G.; Tolon, R. M.; Ramos, J. A.; Guzman,
M. Trends Pharmacol. Sci. 2007, 28, 39.
27. Maccarrone, M.; Battista, N.; Centonze, D. Prog. Neurobiol. 2007, 81, 349.
28. Centoze, D.; Finazzi-Agro, A.; Bernardi, G.; Maccarrone, M. Trends Pharmacol.
Sci. 2007, 28, 180.
N
N
N
R2
O
R1
R2
H
hCB2 Ki
hCB1 Ki
(nM)
hCB2 EC50* (nM)
(nM)
24
25
26a
Acetyl
Acetyl
H
>5000
>5000
>5000
>5000
>5000
2483 271
>5000
>5000
>5000
nd
nd
nd
29. Micale, V.; Mazzola, C.; Drago, F. Pharmacol. Res. 2007, 56, 382.
30. McKallip, R. J.; Lombard, C.; Fisher, M.; Martin, B. R.; Ryu, S.; Grant, S.;
Nagarkatti, P. S.; Nagarkatti, M. Blood 2002, 100, 627.
31. Velasco, G.; Galve-Roperh, I.; Sánchez, C.; Blázquez, C.; Guzmán, M.
Neuropharmacology 2004, 47, 315.
Me 1349 359
Me 4342 117
Me 4.0 0.7
Me 5.0 1.3
Me 41
Me 5.9 0.7
Me 5.7 1.2
26b –C(O)CH2CH(CH3)2
26c –C(O)-2-thiophene
26d –C(O)N(CH3)2
2.7 0.4 (54%)
2.6 0.3 (51%)
36 5 (55%)
2.0 0.1 (48%)
6.4 1.5 (51%)
3
32. Pertwee, R. G. Pharmocol. Therapeut. 2002, 95, 165.
26e –C(O)N(CH2CH3)2
33. Ofek, O.; Karsak, M.; Leclerc, N.; Fogel, M.; Frenkel, B.; Wright, K.; Tam, J.; Attar-
Namdar, M.; Kram, V.; Shohami, E.; Mechoulam, R.; Zimmer, A.; Bab, I. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 696.
34. Idris, A. I.; van’t Hof, R. J.; Greig, I. R.; Ridge, S. A.; Baker, D.; Ross, R. A.; Ralston,
S. H. Nat. Med. 2005, 11, 774.
26f
–C(O)NHCH(CH3)2
26g
Me 4.0 0.6
3475 206
3.3 0.3 (60%)
–C(O)
N
35. Gertsch, J.; Raduner, S.; Altmann, K. H. J. Recept. Signal Transduct. 2006, 26, 709.
36. Muccioli, G. G.; Lambert, D. M. Expert Opin. Ther. Patents 2006, 16, 1405.
37. Pavlopoulos, S.; Thakur, G. A.; Nikas, S. P.; Makriyannis, A. Curr. Pharm. Design
2006, 12, 1751.
nd, not determined. Abbreviations: Me, methyl.
*
Emax are reported in brackets.
38. Manera, C.; Cascio, M. G.; Benetti, V.; Allara, M.; Tuccinardi, T.; Martinelli, A.;
Saccomanni, G.; Vivoli, E.; Ghelardini, C.; Di Marzo, V.; Ferrarini, P. L. Bioorg.
Med. Chem. Lett. 2007, 17, 6505.
39. Kai, H.; Morioka, Y.; Murashi, T.; Morita, K.; Shinonome, S.; Nakazato, H.;
Kawamoto, K.; Hanasaki, K.; Takahashi, F.; Mihara, S.; Arai, T.; Abe, K.; Okabe,
H.; Baba, T.; Yoshikawa, T.; Takenaka, H. Bioorg. Med. Chem. Lett. 2007, 17,
4030.
40. Stern, E.; Muccioli, G. G.; Millet, R.; Goossens, J.-F.; Farce, A.; Chavatte, P.;
Poupaert, J. H.; Lambert, D.; Depreux, P.; Hénichart, J.-P. J. Med. Chem. 2006, 49,
70.
41. Adam, J.; Cowley, P. M.; Kiyoi, T.; Morrison, A. J.; Mort, C. J. W. Prog. Med. Chem.
2006, 44, 207.
42. Verbist, B. M. P.; De Cleyn, M. A. J.; Surkyn, M.; Fraiponts, E.; Aerssens, J.; Nijsen,
M. J. M. A.; Gijsen, H. J. M. Bioorg. Med. Chem. Lett. 2008, 18, 2574.
43. Pagé, D.; Yang, H.; Brown, W.; Walpole, C.; Fleurent, M.; Fyfe, M.; Gaudreault,
F.; St-Onge, S. Bioorg. Med. Chem. Lett. 2007, 17, 6183.
44. Pagé, D.; Brochu, M.-C.; Yang, H.; Brown, W.; St-Onge, S.; Martin, E.; Salois, D.
Lett. Drug Des. Discov. 2006, 3, 298.
hand side of the molecule did not have any positive influence on
the CB2 binding, as compounds 20i and 20j showed a 6.5- to 33-
fold decrease in the binding affinity, respectively.
Similar observations can be drawn from the results of Table 4
where compounds bearing smaller groups (24–26a) showed much
lower CB2 binding affinities than those bearing either bulkier
amide (20b–c) or urea (20d–g) moieties. The binding interactions
at this specific site seem to be mainly hydrophobic and non-spe-
cific since it can accommodate different groups and/or functional-
ities. The left-hand part of the molecule also seem to have less
influence on the nature of the ligand since all the reported modifi-
cations in Tables 3 and 4 only resulted in compounds showing par-
tial agonism (Emax 45–67%).
In conclusion, these molecules, based on a benzimidazole core
represent new scaffolds in the development of cannabinoid ago-
nists. These ligands demonstrated good binding affinities with de-
cent potencies towards the CB2 receptor, along with excellent
selectivity over the CB1 receptor. Further investigations of this
new class of ligands are currently underway in our laboratories
in order to look at their potential biological application.
45. Yadav, J. S.; Reddy, G. S.; Reddy, M. M.; Meshram, H. M. Tetrahedron Lett. 1998,
39, 3259.
46. All products gave satisfactory analytical characterization showing purity >95%
as determined by HPLC using a Zorbax C-18 column (k = 215, 254 and 280 nm).
1H NMR spectra were obtained from
a 400 MHz Varian Unity Plus
spectrometer. Mass spectra were obtained on a Micromass Quattro micro API
or an Agilent 1100 Series LC/MSD instrument using loop injection. Selected
analytical characterizations: Compound 9a: 1H NMR (DMSO-d6) d 7.80 (d, 1H),
7.67 (s, 1H), 7.42 (d, 1H), 7.25 (d, 2H), 6.91(m, 2H), 4.46 (s, 2H), 4.31 (t, 2H),
3.97 (q, 2H), 3.16 (b, 4H), 1.61 (m, 1H), 1.35 (m, 2H), 1.28 (t, 3H), 1.07 (b, 6H),
0.85 (m, 6H); 422.29 (MH+ monoisot.); Compound 9b: 1H NMR (DMSO-d6) d
7.59 (d, 1H), 7.30 (s, 1H), 7.08 (d, 1H), 6.91 (d, 2H), 6.53 (d, 2H), 4.14 (s, 2H),
3.96 (d, 2H), 3.58 (q, 2H), 3.04 (b, 2H), 2.80 (b, 2H), 0.90 (t, 3H), 0.78 (b, 7H),
0.10–0.03 (m, 4H); 406.17 (MH+ monoisot.); Compound 14j: 1H NMR (CD3OD)
d 7.84 (d, 1H), 7.72 (s, 1H), 7.54 (dd, 1H), 7.33 (m, 4H), 7.10 (t, 1H), 6.98 (m, 4H),
4.57 (s, 2H), 4.41 (dd, 2H), 3.55 (br s, 2H), 3.27 (br s, 2H), 1.69 (m, 1H), 1.51 (m,
2H), 1.24 (br s, 3H), 1.11 (br s, 3H), 0.96 (s, 6H); 470.32 (MH+ monoisot.);
Compound 20h: 1H NMR (CD3OD) d 8.00 (d, 1H), 7.79 (s, 1H), 7.59 (d, 1H), 7.27
(s, 2H), 6.95 (d, 2H), 4.56 (s, 2H), 4.36 (m, 6H), 4.02 (q, 2H), 1.37 (t, 3H), 1.24 (m,
1H),0.61 (m, 2H), 0.48 (m, 2H); 514.22 (MH+ monoisot.); Compound 20i: 1H
NMR (CD3OD) d 8.13 (d, 1H), 7.94 (d, 1H), 7.69 (s, 1H), 7.53 (dd, 1H), 7.45 (d,
1H), 7.40 (dd, 1H), 4.80 (s, 2H), 4.37 (d, 2H), 4.04 (q, 2H), 3.58–3.46 (m, 2H),
3.26–3.19(m, 2H), 1.32 (t, 3H), 1.25–1.15 (m, 4H), 1.13–1.00 (m, 3H), 0.59–0.52
(m, 2H), 0.45–0.38 (m, 2H); 407.27 (MH+ monoisot.); Compound 26g: 1H NMR
(CD3OD) d 0.51 (m, 2H), 0.66 (m, 2H), 1.28 (m, 1H), 1.41 (m, 7H), 1.55 (m, 2H),
3.28 (s, 3H), 3.34 (m, 4H), 4.07 (q, 2H), 4.39 (d, 2H), 4.57 (s, 2H), 7.01 (m, 2H),
7.30 (m, 2H), 7.39 (dd, 1H), 7.44 (d, 1H), 7.90 (d, 1H); 447.36 (MH+ monoisot.).
47. General procedure for the CB1/CB2 binding assay: cannabinoids membranes
are thawed at 37 °C, passed 3 times through a 25-gauge blunt-end needle,
diluted in the cannabinoid binding buffer (50 mM Tris, 2.5 mM EDTA, 5 mM
References and notes
1. Cervino, C.; Pasquali, R.; Pagotto, U. Mini-Rev. Med. Chem. 2007, 7, 21.
2. Tucci, S. A.; Halford, J. C. G.; Harrold, J. A.; Kirkham, T. C. Curr. Med. Chem. 2006,
13, 2669.
3. Hohmann, A. G.; Suplita, R. L. AAPS J. 2006, 8, E693.
4. Goya, P.; Jagerovic, N.; Hernandez-Folgado, L.; Martin, M. I. Mini-Rev. Med.
Chem. 2003, 3, 765.
5. Richardson, J. D. J. Pain 2000, 1, 2.
6. Lopez-Rodriguez, M.; Viso, A.; Ortega-Gutiérrez, S.; Diaz-Laviada, I. Mini-Rev.
Med. Chem. 2005, 5, 97.
7. Drysdale, A. J.; Platt, B. Curr. Med. Chem. 2003, 10, 2719.
8. Pertwee, R. G. Br. J. Pharmacol. 2006, 147, S163.
9. Jonsson, K. O.; Holt, S.; Fowler, C. Basic Clin. Pharmacol. Toxicol. 2006, 98, 124.
10. Mackie, K. Annu. Rev. Pharmacol. Toxicol. 2006, 46, 101.
11. Pacher, P.; Batkai, S.; Kunos, G. Pharmacol. Rev. 2006, 58, 389.
12. Howlett, A. C.; Breivogel, C. S.; Childers, S. R.; Deadwyler, S. A.; Hampson, R. E.;
Porrino, L. J. Neuropharmacology 2004, 47, 345.
13. Muccioli, G. G.; Lambert, D. M. Curr. Med. Chem. 2005, 12, 1361.
14. Muccioli, G. G. Chem. Biodivers. 2007, 4, 1805.
MgCl2, and 0.5 mg/mL BSA fatty acid free, pH 7.4) and 80
the appropriate amount of protein are distributed in 96-well plates. The IC50 of
compounds (150 L) are evaluated from 10-point dose–response curves done
with 70
L of 3H-CP55, 940 at 20,000 to 25,000 dpm per well (0.17–0.21 nM) in
a final volume of 300 L. The total and non-specific binding are determined in
the absence and presence of 0.2 M of HU210 (150 L). The plates are vortexed
lL aliquots containing
15. Das, S. K.; Chakrabarti, R. Curr. Med. Chem. 2006, 13, 1429.
16. Basavarajappa, B. S. Mini-Rev. Med. Chem. 2007, 7, 769.
17. Ashton, J. C.; Milligan, E. D. Curr. Opin. Investig. Drugs 2008, 9, 65.
18. Russo, E.; Guy, G. W. Med. Hypotheses 2006, 66, 234.
19. Montero, C.; Campillo, N. E.; Goya, P.; Paez, J. A. Eur. J. Med. Chem. 2005, 40,
75.
l
l
l
l
l

3700
D. Pagé et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3695–3700
56 M of GDP (15 M final). Finally, 70 lL of the tracer GTPc[
35S] (100,000 to
and incubated for 60 min at room temperature, filtered through Unifilters GF/B
l
l
(presoaked in 0.1% polyethyleneimine) with the Tomtec or Packard harvester
using 3 mL of wash buffer (50 mM Tris, 5 mM MgCl2, 0.05% BSA, pH 7.0). The
filters are dried for 1 h at 55 °C. The specific binding (SB) is calculated as TB-NS,
and the SB in the presence of various ligands is expressed as percentage of
control SB. Values of IC50 and Hill coefficient (nH) for ligands in displacing
specifically bound radioligand are calculated in Activity base with ExcelFit. The
concentration of compounds to use and dilutions are also calculated with
Activity base. The radioactivity (cpm) is counted in a TopCount (Packard) after
130,000 dpm/well) is added to start the reaction. Eight wells were used to
define basal (negative control) binding and 8 for positive control (maximal
binding) using 1 lM WIN55,212-2. Plates were then mixed by hand on an
orbital mixer and incubated for 1 h at room temperature. Filter plates were
presoaked in deionized water. Filtration was performed with a Packard cell
harvester using 3ꢀ 1 mL of wash buffer. Filter plates were then dried at 55 °C
for 1.5 h before adding 50 lL of Microscint 20 (Packard Biosciences)
scintillation fluid. Filter plates were counted in a Packard Top Count. The
cpm values of GTP
35S] binding in the 8 wells containing GTP 35S] and
membranes were averaged to define basal binding (BB) and the values of the 8
wells containing WIN55,212-2 were averaged to define GTP
35S]
maximal binding (MB). The stimulation of GTP
35S] binding observed for
each concentration of compound (cpm ECC) was expressed as a percentage of
maximal effect elicited by 1 M WIN55,212-2. GTP
35S] specific binding is
calculated by subtracting the BB.
adding 65
48. GTP
35S] binding on cloned human CB2 receptors in Sf9 cell membranes:
human CB2 assay was performed in a buffer consisting of 50 mM Hepes,
20 mM NaOH, 100 mM NaCl, 1 mM EDTA, 5 mM MgCl2, 0.1% BSA, and 15
l
L/well of MS-20 scintillation liquid.
c[
c[
c
[
1
l
M
c[
l
M
c[
GDP. The pH was set at 7.4 at room temperature. The washing buffer consisted
of 50 mM Tris, 5 mM MgCl2, 50 mM NaCl, pH 7.4, at 4 °C. Compounds were
tested in 10-point dose–response curves. The assay, performed in 96-well
l
c[
plates, consisted of 300
varying concentrations, 80
l
L, containing 150
l
L of buffer alone or compound at
g of protein/well) mixed with
49. Compton, D. R.; Gold, L. H.; Ward, S. J.; Balster, R. L.; Martin, B. R. J. Pharmacol.
Exp. Ther. 1992, 263, 1118.
l
L of membranes (5
l