Benzimidazole CB2 agonists: Design, synthesis and SAR

Article abstract of DOI:10.1016/j.bmcl.2013.12.068

A new series of CB2-selective agonists containing a benzimidazole core is reported. Design, synthesis, SAR and pharmacokinetic data for selected compounds are described.

Full text of DOI:10.1016/j.bmcl.2013.12.068

Bioorganic & Medicinal Chemistry Letters 24 (2014) 1218–1221
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Benzimidazole CB2 agonists: Design, synthesis and SAR
b
b
b
b
Kausik K. Nanda a, , Darrell A. Henze , Kimberly Della Penna , Reshma Desai , Michael Leitl ,
Wei Lemaire b, Rebecca B. White c, Suzie Yeh c, Janine N. Brouillette d, George D. Hartman a,
Mark T. Bilodeau a, B. Wesley Trotter e
⇑
a Department of Medicinal Chemistry, Merck Research Laboratories, West Point, PA, United States
b Department of Pain Research, Merck Research Laboratories, West Point, PA, United States
c Department of Drug Metabolism & Pharmacokinetics, Merck Research Laboratories, West Point, PA, United States
d Department of Structure Elucidation, Merck Research Laboratories, West Point, PA, United States
e Department of Medicinal Chemistry, Merck Research Laboratories, Boston, MA, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A new series of CB2-selective agonists containing a benzimidazole core is reported. Design, synthesis, SAR
and pharmacokinetic data for selected compounds are described.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 31 October 2013
Revised 17 December 2013
Accepted 17 December 2013
Available online 25 December 2013
Keywords:
CB2 agonist
Cannabinoid 2 receptor
Benzimidazole
Cannabis, originating from central Asia, is probably the oldest
psychotropic drug known to humanity and has been used for rec-
reational as well as medicinal (analgesic, anticonvulsant, antie-
metic, appetite stimulant, etc.) purposes.1 Cannabinoids are
pharmacologically active components of cannabis (Cannabis sativa)
and partially act on two known subtypes of cannabinoid receptors,
CB12 and CB2,3 which are members of the G-protein coupled
receptor (GPCR) family. The CB1 receptor is abundantly expressed
in the central nervous system (CNS)4 as well as in the periphery5,6
and is believed to be responsible for the psychotropic effects of
cannabinoids.7 Conversely, CB2 receptors are expressed predomi-
nantly, but not exclusively, outside of the CNS, where they are
found mainly in the cells of the immune system, though they
may be up-regulated in the CNS under pathological conditions
(inflammation, pain).8,9
Activation of CB2 receptors triggers the signal transduction
pathway through Gi proteins resulting in inhibition of adenylate
cyclase activity.10 In contrast to CB1 receptors, CB2 receptors do
not couple to calcium-Q or inward-rectifying potassium chan-
nels,11 whereas agonism of CB1 receptors suppresses calcium and
activates inward-rectifying potassium conductance, associated
with depression and neuronal excitability. Additionally, the first
moderately selective CB2-agonist, HU-308 showed antiinflamma-
tory and antihyperalgesic properties in mice which were reversed
by a CB2 antagonist, but not by a CB1 antagonist.12 Moreover, HU-
308 showed no activity in mice in the tetrad13 of behaviorial tests.
It has thus been hypothesized that CB2-selective agonists could be
therapeutically useful to treat pain without the undesirable psy-
chotropic side effects. Based on this hypothesis, significant efforts
have been dedicated to the pursuit of CB2-selective agonists, both
in academia and in industry.14,15
Large numbers of publications have detailed CB2-agonists with
varying degrees of selectivity over CB1. In our own efforts to find
CB2-selective agonists, we have previously reported imidazopyri-
dines16 and decahydroquinolines17 as potent and selective CB2
agonists. In these studies, we observed16 that highly selective
CB2 agonists (e.g., 2, Fig. 1) did not show efficacy18 in a rat CFA
hyperalgesia model19 despite high exposure in vivo, both peripher-
ally and centrally, while moderately selective CB2/CB1 agonists
(e.g., 1, Fig. 1) show a significant analgesic effect.18 In light of the
above observations, the question remains—how does residual
CB1 agonism affect the outcome of agonist dosing in the in vivo
pain models? Continuing to identify novel structural classes of
CB2-selective agonists may enable additional experiments towards
answering this question.
In the course of our imidazopyridine series efforts, we identified
hetero-aryl alcohol substituents as viable analogs of the morpholi-
nomethylene substituent of imidazopyridine 1 (imidazopyridine
3,20 Fig. 2). Examination of 3 suggested that a similar display of
⇑
Corresponding author. Tel.: +1 2156520371.
E-mail address: kausik_nanda@merck.com (K.K. Nanda).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.12.068

K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1218–1221
1219
Cl
Cl
Cl
Cl
N
Cl
NH
Cl
Cl
O
NH2
Cl
N
O
HN
H2N
Cl
Cl
N
hCB2 cAMPIC50 = 5 nM (94% activity)
hCB1 cAMPIC50 = 2565 nM (66% activity)
AcOH
(72%)
AgNO3
EtOH
4
5
N
75 o
C
O
1
O
OH
O
N
N
O
R
O
HN
NH
HN
Cl
Cl
HN
+
N
N
N
Ar
hCB2 cAMP IC50 =33 nM (99% activity)
hCB1 cAMP IC50 >17000 nM
R
N
7
6
N
(50% by LC/MS)
(50% by LC/MS)
O
2
Scheme 1.
Figure 1. Imidazopyridine CB2 agonists.
B(OH)2
NO2
NO2
NH2
Pd[P(t-Bu)3]2
KF
H2N
Cl
H2N
SnCl2
H2N
+
R
NH
THF, 100°C
(90%)
O
EtOH, 75°C
(94%, crude)
O
8
9
10
11
NH
N
N
N
N
O
Cl
Cl
NH
Cl
O
Cl
Cl
N
EtI
K2CO3
AgNO3
EtOH
O
N
N
Cl
HO
3 (racemic)
HN
HN
AcOH
(86%)
DMF
75°C
(96%)
hCB2 cAMP IC50 = 4.5 nM
(88% activity)
hCB1 cAMP IC50 = 3881 nM
(82% activity)
12
13
(71%)
O
N
N
O
Figure 2. Design of benzimidazole core from imidazopyridine scaffold.
OH
16
O
RNH2
(80% by LC/MS)
N
N
EDC, HOAT DIEA
aq NaOH
(80%)
N
N
+
substituents could be achieved using a completely different benz-
imidazole scaffold (Figs. 2 and 3).
DMF, 80°C
R
NH
14
15
O
To test this hypothesis, synthesis of benzimidazole containing
compounds was investigated. The initial route employed 3-
chloro-1,2-diaminobenzene 4 as starting material (Scheme 1).
Reaction21 of 4 with methyl-2,2,2-trichloroacetimidate in acetic
acid provided 2-trichloromethyl benzimidazole 5. Unfortunately,
ethanolysis of 5 to give 6 also provided 50% decarboxylated
byproduct 7.
Although the reason behind the formation of decarbonylated
byproduct 7 during ethanolysis of 5 was not well understood, an
alternate route was devised where the 7-chloro group of 5 was
substituted with a phenyl group prior to ethanolysis of the 2-tri-
chloromethyl group. In this route (Scheme 2), phenyl boronic acid
was coupled to 3-chloro-2-nitroaniline to give 10. The nitro group
of 10 was then reduced with tin(II)chloride to give 11, which in
turn was transformed to 2-trichloromethyl benzimidazole 12 via
N
N
R
NH
O
17
RNH2
PyBOP, DIEA
N
N
(20% by LC/MS)
DMF, rt
(~80%)
17
Scheme 2.
treatment with methyl-2,2,2-trichloroacetimidate. Ethanolysis of
12 cleanly gave benzimidazole-2-ethyl ester in 96% yield. Alkyl-
ation of 13 produced exclusively one product whose regiochemist-
ry was determined by NMR spectrometry on the basis of ROESY
correlations.22 The alkylated product, 14 was then hydrolyzed to
15. When standard peptide coupling conditions (EDC, HOAT, DIEA
in DMF) were employed to couple amines with 15, reactions did
not proceed at rt. At higher temperature (80 °C), predominantly
decarboxylated byproduct 16 was observed. This problem was
alleviated by using a more active coupling agent, PyBOP, which
yielded the desired amides in good yields.
This synthetic route allowed us to make benzimidazole analogs
for which the amide group as well as the N-alkyl group could be
varied. Evaluation of the initial compounds revealed that this
chemotype did afford potent CB2 agonists with a high degree of
CB2/CB1 selectivity.23 Table 1 lists selected compounds containing
arylmethyl amides at the benzimidazole 2-position. As evident
from Table 1, N-ethyl benzimidazoles show more potent CB2 agon-
ism (lower EC50 value and higher Emax) when compared to N-
unsubstituted benzimidazoles. The most potent compound in this
series is 2-chlorobenzyl amide 7, which shows 92-fold selectivity
over CB1. 3-Chlorobenzyl amide 8, although it exhibits 4-fold
weaker potency vs CB2 (compared with 7), shows >425-fold selec-
Figure 3. Overlay of gas phase conformations (within 3 kcal/mol of global
minimum) of imidazopyridine 3 and benzimidazole 12.

1220
K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1218–1221
Table 1
SAR: arylmethyl amides
Table 2
SAR: aliphatic amides
R2
NH
R2
NH
O
O
N
R1
N
N
R1
N
Entry R1 R2
hCB2 cAmp EC50 nMa
hCB1 cAmp EC50 nMa
(Emax)
Entry R1
R2
hCB2 cAmp EC50
hCB1 cAmp EC50
b
b
(Emax
)
nMa (Emax
1.9 (99%)
6.3 (75%)
)
nMa (Emax
)
b
b
Cl
1
2
3
H
H
H
165 (67%)
256 (49%)
256 (69%)
>17,000
>17,000
>17,000
12
H
123 (100%)
F
HO
13
14
H
H
>17,000
>17,000
OH
F3C
14.7 (95%)
24.6 (93%)
OH
OH
4
5
H
H
443 (64%)
991 (64%)
>17,000
>17,000
Cl
15
16
H
H
>17,000
>17,000
N
43 (92%)
6
H
1437 (39%)
>17,000
F
HO
17
18
19
H
236 (82%)
490 (67%)
1255 (68%)
>17,000
913 (87%)
>17,000
7
Et
Et
Et
Et
Et
9.9 (101%)
40 (87%)
45 (98%)
84 (99%)
1104 (89%)
913 (87%)
>17,000
>17,000
>17,000
>17,000
Cl
Cl
8
Et
Et
OH
OH
9
F
10
11
NC
N
20
21
22
CH2CF3
CH2CF3
CH2CF3
1973 (45%)
>17,000
>17,000
>17,000
>17,000
N
a
OH
See Ref. 23.
See Ref. 24.
b
>17,000
a
See Ref. 23.
See Ref. 24.
b
tivity over CB1. SAR also reveals that heteroatoms in the aromatic
ring are not favored (6 and 11).
Table 2 lists benzimidazoles where the 2-position is substituted
with aliphatic amides and the 1-position is either unsubstituted or
substituted with alkyl groups. A bulky, lipophilic adamantyl amide
gives a very potent CB2 agonist (12), but its selectivity over CB1 is
poor (65-fold). When the adamantyl group is replaced with a t-
butylmethyl group (14), a potent CB2 agonist is obtained which
shows >1156-fold selectivity over CB1. In an effort to reduce lipo-
philicity, aliphatic amides bearing hydroxy groups were prepared.
As evident from Table 2, these molecules were potent and selective
CB2 agonists. The most potent and selective compound from this
category is 13. As noted for the arylmethyl amides (Table 1), ali-
phatic amides bearing N-ethyl substitution of the benzimidazole
were more potent than their unsubstituted counterparts (compare
14 and 18; 15 and 19).
F
HO
O
NH
NH
NH
N
O
O
N
N
N
N
N
9
14
15
CLa = 21
t1/2b = 0.8
Vdc = 1.1
CLa = 39
t1/2b = 0.5
Vdc = 1.3
CLa = 37
t1/2b = 0.5
Vdc = 1.7
Figure 4. PK (dog cassette) data. amL/min/kg; bh; cL/kg.
As evident from Table 2 (20, 21, 22), introduction of an electron
withdrawing trifluoroethyl group makes these compounds much
less active as CB2 agonists.
In summary, a new structural class of CB2 agonists has been
Evaluation of analgesic effects in animal models of pain requires
pharmacokinetic profiles that provide adequate exposures for CB2
agonism. Following our PK screening paradigm for this series, rep-
resentative compounds were selected for initial pharmacokinetic
profiling in dog i.v. cassette PK experiments. Figure 4 shows data
for compounds 9, 14 and 15. 2-Fluorobenzyl amide 9 shows mod-
erate clearance (21 mL/min/kg) whereas two aliphatic amide con-
taining compounds (14 and 15) show moderately higher clearance.
This data suggests that, although further optimization is needed,
compounds with suitable pharmacokinetic profile for in vivo dos-
ing could be obtained from this chemical series.
identified. Exploratory efforts have provided CB2-selective com-
pounds with moderate pharmacokinetic profiles. Further optimiza-
tion could provide new tools for understanding the viability of
selective CB2 agonists for the treatment of pain.
Acknowledgment
We thank Deping Wang for generation of the overlay of gas
phase conformations of imidazopyridine and benzimidazole con-
taining structures.

K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1218–1221
1221
18. Compound 1: rat CB2 cAMP EC50 = 11 nM (87% activity); rat CB1 cAMP
EC50 = 1068 nM (104% activity); rat plasma protein binding 76%. Compound 1
exhibited naproxen-like reversal in paw withdrawal threshold in rat CFA
model at 100 mpk (PO dosing) at 60 min post dose. Plasma, brain and CSF
levels at 100 mpk dose (60 min timepoint) were 4500, 5400 and 110 nM,
respectively.
References and notes
1. Ben Amar, M. J. Eyhnopharm. 2006, 105, 1.
2. Matsuda, L. A.; Lolait, S. J.; Brownstein, M. J.; Young, A. C.; Bonner, T. I. Nature
1990, 346, 561.
3. Munro, S.; Thomas, K. L.; Abu-Shaar, M. Nature 1993, 365, 61.
4. Herkenham, M.; Lynn, A. B.; Little, M. D.; Johnson, M. R.; Melvin, L. S.; de Costa,
B. R.; Rice, K. C. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 1932.
5. Straiker, A.; Stella, N.; Piomelli, D.; Mackie, K.; Karten, H. J. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 14565.
6. Galiegue, S.; Mary, S.; Marchand, J.; Dussossoy, D.; Carriere, D.; Carayon, P.;
Bouaboula, M.; Shire, D.; LeFur, G.; Eur Casellas, P. J. Biochem. 1995, 232, 54.
7. Adams, I. B.; Martin, B. R. Addiction 1996, 91, 1585.
8. Patel, K. D.; Davison, J. S.; Pitman, Q. J.; Sharkey, K. A. Curr. Med. Chem. 2010, 17,
1393.
9. Benito, C.; Tolon, R. M.; Pazos, M. R.; Nunez, E.; Castillo, A. I.; Romero, J. Br. J.
Pharmacol. 2008, 153, 277.
Compound 2: rat CB2 cAMP EC50 = 58 nM (87% activity); rat CB1 cAMP
EC50 > 17,000 nM; rat plasma protein binding 84%. Compound 1 exhibited no
change in paw withdrawal threshold in rat CFA model at 100 mpk (PO dosing)
at 120 min post dose. Plasma, brain and CSF levels at 100 mpk dose (120 min
timepoint) were 4300, 1300 and 300 nM, respectively.
19. Nagakura, Y.; Okada, M.; Kohara, A.; Kiso, T.; Toya, T.; Iwai, A.; Wanibuchi, F.;
Yamaguchi, T. J. Pharmacol. Exp. Ther. 2003, 306, 490.
20. Bilodeau, M. T.; Burgey, C. S.; Deng, Z. J.; Hartnett, J. C.; Kett, N. R.; Melamed, J.;
Munson, P. M.; Nanda, K. K.; Thompson, W.; Trotter, B. W. PCT Int. Appl., 2008,
WO2008085302.
21. Louvet, P.; Lallement, G.; Pernot-Marino, I.; Luu-Duc, C.; Blanchet, G. Eur. J. Med.
Chem. 1993, 28, 71.
22. Diagnostic NOE correlations (indicated by the arrows) to assign regiochemistry
of the alkylated product:
10. Howlett, A. C.; Barth, F. I.; Cabral, G.; Casellas, P.; Devane, W. A.; Felder, C. C.;
Herkenham, M.; Mackie, K.; Martin, B. R. Pharmacol. Rev. 2002, 54, 161.
11. Felder, C. C.; Joyce, K. E.; Briley, E. M.; Mansouri, J.; Mackie, K.; Blond, O.; Lai, Y.;
Ma, A. L.; Mitchell, R. L. Mol. Pharmacol. 1995, 48, 443.
12. Hanus, L.; Breuer, A.; Tchilibon, S.; Shiloah, S.; Goldenberg, D.; Horowitz, M.;
Pertwee, R. G.; Ross, R. A.; Fride, R.; Fride, E. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
14228.
13. Martin, B. R.; Compton, D. R.; Thomas, B. E.; Prescott, W. R.; Little, P. J.; Razdan,
R. J.; Johnson, M. R.; Melvin, L. S.; Mechoulam, R.; Ward, S. J. Pharmacol.
Biochem. Behav. 1991, 40, 471.
O
N
N
HN
F
23. Assay results represent the average of N = 1 to 3 results. Inter-assay variability
was 20%.
14. Thakur, G. A.; Tichkule, R.; Bajaj, S.; Makriyannis, A. Expert Opin. Ther. Pat. 2009,
19, 1647.
24. Compounds were added to a Greiner black 384 well low volume assay plate.
1000 CHO-K1 cells (expressing human or rat CB1 or CB2) were added to each
well of the assay plate containing compound, then incubated at room
temperature for 15 min. Then, forskolin at EC70 was added and incubated at
room temperature for an additional 30 min. Detection of cAMP was performed
using Cisbio’s HRTF cAMP dynamic 2 kit following the manufacturer’s protocol
for cAMP detection. After adding d2-cAMP and anti-cAMP-cryptate to all the
wells, there was a final incubation at room temperature for 1 hour. Plates were
then read on an EnVision plate reader (Perkin Elmer). CV values calculated for a
positive control compound (>250 replicates) were 59% and 70% respectively,
for CB1 and CB2 IP values.
15. Hollinshead, S. P.; Tidwell, M. W.; Palmer, J.; Guidetti, R.; Sanderson, A.;
Johnson, M. P.; Chambers, M. G.; Oskins, J.; Stratford, R.; Astles, P. J. Med. Chem.
2013, 56, 5722.
16. Trotter, B. W.; Nanda, K. K.; Burgey, C. S.; Potteiger, C. M.; Deng, J.; Green, A. I.;
Hartnet, J. C.; Kett, N. R.; Wu, Z.; Henze, D. A.; Della Penna, K.; Desai, R.; Leitl, M.
D.; Lemaire, W.; White, R. B.; Yeh, S.; Urban, M. O.; Kane, S. A.; Hartman, G. D.;
Bilodeau, M. T. Bioorg. Med. Chem. Lett. 2011, 21, 2354.
17. Manley, P. J.; Zartman, A.; Paone, D. V.; Burgey, C. S.; Henze, D. A.; Della Penna,
K.; Desai, R.; Leitl, M. D.; Lemaire, W.; White, R. B.; Urban, M. O.; Kane, S. A.;
Hartman, G. D.; Bilodeau, M. T.; Trotter, B. W. Bioorg. Med. Chem. Lett. 2011, 21,
2359.