Novel pyridine derivatives as potent and selective CB2 cannabinoid receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2009.08.063

Replacement of the phenyl ring in our previous (morpholinomethyl)aniline carboxamide cannabinoid receptor ligands with a pyridine ring led to the discovery of a novel chemical series of CB₂ ligands. Compound 3, that is, 2,2-dimethyl-N-(5-methyl-4-(morpholinomethyl)pyridin-2-yl)butanamide was identified as a potent and selective CB₂ agonist exhibiting in vivo efficacy after oral administration in a rat model of neuropathic pain.

Full text of DOI:10.1016/j.bmcl.2009.08.063

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5931–5935
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel pyridine derivatives as potent and selective CB2 cannabinoid
receptor agonists
a
a
a
a
Guo-Hua Chu a, , Christopher T. Saeui , Karin Worm , Damian G. Weaver , Allan J. Goodman ,
Robert L. Broadrup a, Joel A. Cassel b, Robert N. DeHaven b, Christopher J. LaBuda b, Michael Koblish b,
Bernice Brogdon c, Steve Smith c, Bertrand Le Bourdonnec a, Roland E. Dolle a
*
a Department of Chemistry, Adolor Corporation, 700 Pennsylvania Drive, Exton, PA 19341, USA
b Department of Pharmacology, Adolor Corporation, 700 Pennsylvania Drive, Exton, PA 19341, USA
c Department of DMPK, Adolor Corporation, 700 Pennsylvania Drive, Exton, PA 19341, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Replacement of the phenyl ring in our previous (morpholinomethyl)aniline carboxamide cannabinoid
receptor ligands with a pyridine ring led to the discovery of a novel chemical series of CB2 ligands. Com-
pound 3, that is, 2,2-dimethyl-N-(5-methyl-4-(morpholinomethyl)pyridin-2-yl)butanamide was identi-
fied as a potent and selective CB2 agonist exhibiting in vivo efficacy after oral administration in a rat
model of neuropathic pain.
Received 21 May 2009
Revised 13 August 2009
Accepted 13 August 2009
Available online 21 August 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Cannabinoid receptors
Selective CB2 agonists
Pain
Pyridine derivatives
The cannabinoid receptors CB1 and CB2 belong to the superfam-
ily of G-protein-coupled receptors (GPCRs).1 Cannabinoid receptors
are present at key sites involved in the relay and modulation of
nociceptive information.1 The analgesic effects of cannabinoids
have been extensively studied.2 However, administration of tetra-
hydrocannabinol (THC) or other non-selective synthetic cannabi-
noids is followed by severe psychotropic side effects mediated by
activation of CB1 receptors within the central nervous system
(CNS).2–4 The CB1 receptors are located centrally and peripherally,
whereas the CB2 receptors are expressed primarily in immune cells
and tissues.1–3 In preclinical studies, CB2 selective receptor ago-
nists inhibit signs of acute nociceptive, inflammatory and neuro-
pathic pain, and importantly they do not cause the CNS side
effects typically produced by cannabinoid ligands displaying ago-
nist activity at the CB1 receptor.2–7 Therefore, selective CB2 recep-
tor agonists are very promising candidates for the treatment of
pain.
1 suffered from poor in vitro metabolic stability as measured in rat
and human liver microsomes (abbreviated as RLM and HLM,
respectively), and the lead optimization campaign focused on
improving CB2 affinity as well as in vitro metabolic stability, while
retaining good selectivity for the CB2 receptor. As indicated in Fig-
ure 1, introduction of a methyl group at the 2-position of the ben-
zene core of 1 (compound 2) led to a clear improvement in the
metabolic stability in rat liver microsomes. However, this subtle
structural modification also led to a fivefold reduction in the CB2
binding affinity. This result suggested that the 2-position of the
benzene ring of 1 might represent one of the ‘soft’ spots of the mol-
ecule (position at which metabolism may occur). Based on this
hypothesis, we investigated whether replacing the phenyl moiety
of 1 with a pyridine core (compound 3) would also result in an
improvement in metabolic stability. As shown in Figure 1, the pyr-
idine derivative 3 displayed increased metabolic stability in rat li-
ver microsomes when compared to its benzene analog 1. In
addition, this modification led to a fivefold increase in the binding
affinity toward the CB2 receptor [1: Ki (CB2) = 120 nM; 3: Ki
(CB2) = 24 nM]. As a follow up to this result, we explored the struc-
ture–activity relationships (SAR) in this new series of pyridine-
based CB2 agonists (general formula I, Fig. 1).
As part of a research program aimed at identifying novel selec-
tive CB2 agonists, we identified the (morpholinomethyl)aniline car-
boxamide derivative 1 (Fig. 1) as the starting point for a lead
optimization campaign.8 Compound 1 bound with moderate affin-
ity to the CB2 receptor (Ki = 120 nM), while lacking significant
affinity at the CB1 receptor (4% inh. @ 10
l
M). However, compound
The syntheses of compound 3 and its regioisomeric analogs 4, 5
and 6 are summarized in Schemes 1 and 2. Treatment of commer-
cially available 2-amino-4-methyl-5-bromopyridine (17a) with
2,2-dimethylbutanoyl chloride gave the amide 18a (98%).
* Corresponding author. Tel.: +1 484 595 1083; fax: +1 484 595 1551.
E-mail address: ghchu@adolor.com (G.-H. Chu).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.08.063

5932
G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5931–5935
N
N
N
Z
Y
X
N
O
N
O
O
W
O
HN
O
HN
O
HN
O
HN
O
R
I
3
2
1
Ki (CB2) = 24 nM
Ki (CB1) = 3800 nM
RLM: 43% remaining
@ 30 min
HLM: 64% remaining
@ 30 min
Ki (CB2) = 530 nM
Ki (CB1): 18% inh. @ 10 µM
RLM: 68% remaining
@ 30 min
HLM: 61% remaining
@ 30 min
Ki (CB2) = 120 nM
Ki (CB1): 4% inh. @ 10 µM
RLM: 2.2% remaining
@ 30 min
HLM: 56% remaining
@ 30 min
CN
NH2
N
N
N
N
N
N
N
N
O
O
O
N
N
O
O
N
N
O
HN
O
HN
O
HN
O
HN
O
HN
O
O
HN
O
5
4
6
7
8
9
N
N
N
N
N
HN
O
N
O
N
O
N
O
HN
O
HN
O
HN
O
10
11
12
13
N
N
N
N
O
N
O
N
O
HN
O
HN
O
HN
O
14
15
Figure 1. Design of novel pyridine derivatives as CB2 cannabinoid receptor agonists.
16
Treatment of amide 18a with NBS in the presence of AIBN provided
the bromide 19a which reacted with morpholine to yield 20a (40%
for two steps). Compound 20a was subjected to a Negishi coupling
reaction by treatment with methylzinc chloride to afford the de-
sired target compound 3 (84%). Compounds 4 and 5 were synthe-
sized following the same route described for the preparation of 3
(Scheme 1).
Br
Y
Br
Br
Br
a
Y
b
O
Y
O
W
W
W
X
NH2
X
N
H
X
18a,b,c
a:
N
H
17a,b,c
19a,b,c
W = CH, X = N, Y = CH;
Compound 6 was prepared from commercially available 2-ami-
no-4,6-dimethylpyridine (17d) according to Scheme 2. Bromina-
b: W = N, X = CH, Y = CH;
c:
W = CH, X = CH, Y = N
tion of amide 18d with NBS yielded
a mixture of two
O
O
O
O
regioisomeric bromides 19d and 19e, which was reacted with mor-
pholine to furnish the desired compound 6 as the minor product
(ꢀ10%), and the regioisomeric analog of 6, that is, compound 21,
as the major product (45%).
Scheme 3 outlines the synthesis of analogs 7–9. Suzuki–Miya-
ura cross-coupling of the bromide 20a with 4,4,5,5-tetramethyl-
2-vinyl-1,3,2-dioxaborolane in the presence of Pd(PPh3)4 under
microwave conditions gave the vinyl compound 22 (60%), which
was hydrogenated to furnish the 5-ethyl analog 7 (70%). Substitu-
tion reaction of 20a with copper(I) cyanide in refluxing DMF affor-
ded the 5-cyano substituted analog 8 (21%). Palladium-catalyzed
carbon monoxide insertion reaction of the bromide 20a under
N
Y
N
c
Br
d
Y
W
W
X
X
N
N
H
H
20a: W = CH, X = N, Y = CH;
20b:
3: W = CH, X = N, Y = CH;
4:
W = N, X = CH, Y = CH;
20c: W = CH, X = CH, Y = N,
W = N, X = CH, Y = CH;
5: W = CH, X = CH, Y = N
Scheme 1. Synthesis of compound 3–5. Reagents: (a) 2,2-dimethylbutanoyl
chloride, TEA, DCM, 93–96%; (b) NBS, AIBN, CCl4; (c) morpholine, THF, NaHCO3,
10–65% (two steps); (d) MeZnCl, Pd(PPh3)4, THF, 76–88%.

G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5931–5935
5933
Br
a
O
b
O
O
+
N
N
H
N
NH2
N
N
H
N
N
H
Br
17d
18d
19d
19e
O
O
O
N
N
N
c
O
+
N
N
H
N
H
6
21
Scheme 2. Synthesis of compound 6. Reagents: (a) 2,2-dimethylbutanoyl chloride, TEA, DCM, 95%; (b) NBS, AIBN, CCl4; (c) morpholine, THF, NaHCO3, 10% for 6, 45% for 21
(two steps).
O
O
O
O
O
O
N
N
N
Br
a
d
b
N
N
H
N
22
N
H
N
7
N
H
20a
c
O
O
O
O
O
O
O
O
N
N
N
N
O
O
e
f
H2N
NC
HO
MeO
N
9
N
H
N
8
N
H
N
24
N
H
N
23
N
H
Scheme 3. Synthesis of compounds 7–9. Reagents: (a) 4,4,5,5-tertamethyl-2-vinyl-1,3,2-dioxaborlane, Pd(PPh3)4, K2CO3, dioxane–H2O, microwave, 120 °C, 60%; (b) H2, Pd/C,
MeOH, 70%; (c) CuCN, DMF, reflux, 21%; (d) MeOH, CO, dppp, Pd(OAc)2, TEA, DMF, 54%; (e) LiOH, THF–MeOH–H2O, 98%; (f) (i) DPPA, TEA, dioxane, t-BuOH; (ii) HCl, MeOH,
Et2O, 26%.
standard conditions gave the methyl ester 23 (54%) which was
hydrolyzed to the corresponding carboxylic acid 24 (98%) by treat-
ment with lithium hydroxide. Curtius rearrangement of 24 by reac-
tion with diphenylphosphoryl azide (DPPA) in the presence of t-
BuOH9 followed by treatment of the resulting tert-butyl carbamate
with HCl yielded the 5-amino substituted analog 9 (26%).
The synthesis of compounds 10–16 is described in Scheme 4.
Treatment of 3 with 6 N HCl at reflux provided the aminopyridine
25 (82%), a common intermediate used for the preparation of 10–
16. Compounds 10–16 were prepared in 7–55% yield by coupling
the aminopyridine derivative 25 with carboxylic acids under stan-
dard peptide coupling conditions using bis(2-oxo-3-oxazolidi-
nyl)phosphinic chloride (BOPCl) as coupling reagent or by
coupling 25 with the corresponding acyl chloride derivatives.
The novel pyridine derivatives 3–16 were evaluated for their
affinity toward the CB2 and the CB1 receptors as measured by their
ability to displace [3H]-CP-55960 from its specific binding sites in
membranes expressing CB1 and CB2 receptors.10 The agonist poten-
cies of selected compounds (CB2: Ki <100 nM) were assessed by
their abilities to stimulate guanosine 50-O-(3-[35S]thio)triphos-
O
O
O
O
O
N
N
N
a
b
N
3
N
H
N
10 16
-
N
H
R
N
25
NH2
phate ([35S]GTP
cS) binding to membranes containing CB2 cannab-
inoid receptors.10 Compounds displaying good affinity at CB2
receptors (Ki <50 nM) were also evaluated for in vitro metabolic
stability in rat and human liver microsomes.11 As indicated in Ta-
ble 1, compounds 4 and 5, the regioisomeric analogs of 3, displayed
weak affinity at the CB2 receptor. Changing the methyl moiety of 3
from the 5-position to the 6-position relative to the pyridine ring
(compound 6)12 also led to a threefold reduction in the CB2 binding
affinity, indicating that the substitution pattern of the pyridine ring
has a significant impact on the optimal CB2 binding affinity. The 5-
ethyl analog of 3 (compound 7) displayed potent affinity at the CB2
RCOOH:
O
O
O
O
HO
HO
HO
n = 1-4
HO
n
Scheme 4. Synthesis of compounds 10–16. Reagents: (a) 6 N HCl, reflux, 82%; (b) (i)
RCOOH, SOCl2; (ii) 25, TEA, DCM or RCOOH, bis(2-oxo-3-oxazolidinyl)phosphinic
chloride (BOPCl), DIEA, DCM, 7–55%.

5934
G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5931–5935
Table 1
577%
18
15
12
9
Cannabinoid receptor binding and in vitro metabolic stability of compounds 1–16
OF VEHICLE
a
a
a
Compd
Ki CB1
Ki CB2
Ratio
(CB1/CB2)
EC50CB2
(nM)
Microsomal
stabilityb
*
(nM or % inh @ 10
l
M)
RLM
HLM
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
4%
120
530
24
46%
1800
67
7.9
100
870
81
5.3
63
nd
nd
160
nd
nd
nd
550
nd
nd
nd
330
nd
nd
85
nd
ndc
nd
41
nd
nd
260
9.6
200
nd
70
4.1
62
2.2
68
43
nd
nd
nd
29
nd
nd
nd
nd
nd
48
nd
0
56
61
64
nd
nd
nd
34
nd
nd
nd
15
nd
51
12
0
271%
18%
3800
4%
28%
0%
4300
11%
10%
37%
1800
25%
13%
1100
36%
2500
OF VEHICLE
*
6
3
0
SHAM
VEHICLE
MORPHINE
COMPOUND 3
45
13
7.0
1.9
140
13
8.8
2.8
Figure 2. Antiallodynic activity of compound 3 (100 mg/kg PO) in the L5 SNL model
(9 days after surgery) (morphine: 3 mg/kg s.c.). p <0.05 compared to vehicle-
treated rats. n = 8 (SNL Groups), n = 4 (Sham-operated animals); Vehicle thresh-
*
1300
0
1
old = 2.4 grams.
a
b
c
For assay description see Ref. 10.
% remaining of parent compound after 30 min, see Ref. 11.
nd: not determined.
kg. After a 10 mg/kg PO dose, maximum plasma concentrations
were reached at around 2 h post-dose. The oral bioavailability of
3 in rat was 57%. Table 2 summarizes the pharmacokinetics of 3
in male Beagle dogs after single IV and PO doses of 1 and 3 mg/
kg, respectively. In contrast to the high clearance observed in rat,
the systemic plasma clearance of 3 in dog was low (0.9 L/h/ kg).
The half-life of 3 in dog was 1.4 h after a 1 mg/kg IV dose. After a
single 3 mg/kg oral dose of 3, maximum plasma concentrations
averaging 535 ng/mL were reached 30 min post-dose. The oral bio-
availability of 36 in dog was 29%. Additional studies demonstrated
that 3 exhibited antiallodynic activity in the L5 spinal nerve liga-
tion (L5 SNL) rat model13 of neuropathic pain at a dose of
100 mg/kg PO (Fig. 2). The weak in vivo activity of the compound
3 is not optimized and possibly caused by its rapid clearance. Fur-
ther work is underway to enhance in vivo activity of this class of
CB2 agonists.
In summary, replacement of the benzene template in our previ-
ous (morpholinomethyl)aniline carboxamide cannabinoid receptor
ligands with a pyridine core led to the discovery of a novel chem-
ical series of pyridine-based potent and selective CB2 agonists.
Compound 3, that is, 2,2-dimethyl-N-(5-methyl-4-(morpholinom-
ethyl)pyridin-2-yl)butanamide, the best compound in this series,
displayed good affinity at the CB2 receptor (Ki = 24 nM), 160-fold
selectivity versus CB1 and moderate metabolic stability in rat and
human liver microsomes. Importantly, compound 3 displayed anti-
allodynic activity after oral administration in a rat model of neuro-
pathic pain.
receptor. However, this modification also led to a decrease in the
in vitro metabolic stability (see Table 1). Replacement of the 5-
methyl group of 3 with a polar substituent such as CN (compound
8) or NH2 (compound 9) caused a significant decrease in CB2 bind-
ing affinity. We investigated further the SAR at the 2-position of 3,
concentrating on carboxamide derivatives. Substitution of the 2,2-
dimethylbutanoyl moiety of 3 with the less lipophilic pivaloyl
group (compound 10) led to a threefold reduction in CB2 binding
affinity. In contrast, extending the acyl substituent from dimeth-
ylbutanoyl to 2,2-dimethylpentanoyl, as in 11, was beneficial to
improve the binding affinity at the CB2 receptor. Unfortunately,
compound 11 displayed poor in vitro metabolic stability. Among
the constrained analogs of 3 that were prepared, the two cyclo-
hexyl derivatives 15 and 16 were identified as potent and selective
CB2 agonists. However, as was the case for 11, compounds 15 and
16 could not be studied further due to their poor in vitro metabolic
stability.
After exploring the SAR in this new chemical series of pyridine-
based CB2 agonists, 2,2-dimethyl-N-(5-methyl-4(morpholinom-
ethyl)pyridin-2-yl)butanamide (3) was the best compound identi-
fied in terms of CB2 affinity, selectivity versus CB1 and metabolic
stability. We then determined the pharmacokinetic profile of 3 in
rat and dog after IV and PO administration. As indicated in Table
2, 3 is a high clearance compound in rat. The elimination of 3
was high with a half-life of 0.5 h after a single IV dose of 1 mg/
References and notes
Table 2
1. Pan, H.-L.; Wu, Z.-Z.; Zhou, H.-Y.; Chen, S.-R.; Zhang, H.-M.; Li, D.-P. Pharmacol.
Ther. 2008, 117, 141.
2. Malan, T. P., Jr.; Ibrahim, M. M.; Lai, J.; Vanderah, T. W.; Makriyannis, A.;
Porreca, F. Curr. Opin. Pharmacol. 2003, 3, 62.
Pharmacokinetics of 3 in male Sprague-Dawley rats and male Beagle dogs after IV and
PO administration
Route
IV
PO
3. Ermann, M.; Riether, D.; Walker, E. R.; Mushi, I. F.; Jenkins, J. E.; Noya-Marino,
B.; Brewer, M. L.; Taylor, M. G.; Kahrs, A. F.; Ebneth, A.; Lobbe, S.; O’Shea, K.;
Shih, D.-T.; Thomson, D. Bioorg. Med. Chem. Lett. 2008, 18, 1725.
4. Giblin, G. M. P.; 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.
5. Hohmann, A. G.; Farthing, J. N.; Zvonok, A. M.; Makriyannis, A. J. Pharmacol. Exp.
Ther. 2004, 308, 446.
6. Ibrahim, M. M.; Porreca, F.; Lai, J.; Albrecht, P. J.; Rice, F. L.; Khodorova, A.;
Davar, G.; Makriyannis, A.; Vanderah, T. W.; Mata, H. P.; Malan, T. P., Jr. Proc.
Natl. Acad. Sci. U.S.A. 2005, 102, 3093.
Species
Rat
Dog
Rat
Dog
Dose (mg/kg)
CLs (L/h/kg)
Vdss (L/kg)
at1/2 (h)
1.0
1.0
10
—
—
3.0
—
—
5.8 0.8
3.5 0.8
0.5 0.2
158 23
—
0.9 0.0
0.6 0.1
1.4 0.0
1127 30
—
2.1 2.4
893 487
2.0 (0.25–2.0)
185 133
57 31
1.1 0.1
977 227
0.5 (0.25–2.0)
535 225
AUC0–1 (ng h/mL)
bTmax (h)
Cmax (ng/mL)
F (%)
—
—
—
—
29
7
Values represent the mean standard deviation of three animals.
7. Worm, K.; Zhou, Q. J.; Saeui, C. T.; Green, R. C.; Cassel, J. A.; Stabley, G. J.;
DeHaven, R. N.; Conway-James, N.; LaBuda, C. J.; Koblish, M.; Little, P. J.; Dolle,
R. E. Bioorg. Med. Chem. Lett. 2008, 18, 2830.
a
Expressed as harmonic mean.
Expressed as median and range.
b

G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5931–5935
5935
8. Worm, K.; Weaver, D. G.; Green, R. C.; Saeui, C. T.; Dulay, D.-M. S.; Barker, W.
M.; Cassel, J. A.; Stabley, G. J.; DeHaven, R. N.; LaBuda, C. J.; Koblish, M.;
Brogdan, B. L.; Smith, S. A.; Dolle, R. E. Bioorg. Med. Chem. Lett. 2009, 19, 5004.
9. Ninomiya, K.; Shioiri, T.; Yamada, S. Tetrahedron 1974, 30, 2151.
Stark, S.; Sim, L. J.; Childers, S. R. Life Sci. 1996, 59, 659–668. Additional details
are given in Ref. 7; (c) All the compounds listed in Table 1 are full CB2 receptor
agonists as determined in a [35 S]GTP
cS functional binding assay.
11. Obach, S. R. Drug Metab. Dispos. 1999, 27, 1350.
10. Binding assays were performed by modification of the method of Pinto, J. C.;
Potie, F.; Rice, K. C.; Boring, D.; Johnson, M. R.; Evans, D. M.; Wilken, G. H.;
Cantrell, C. H.; Howlett, A. Mol. Pharmacol. 1994, 46, 516–522: (b) The [35
12. Compound 21, the regioisomeric analog of 6, exhibited much weaker CB2
binding affinity as compared to 6 and 3 with Ki (CB2) = 470 nM.
13. LaBuda, C. J.; Little, P. J. J. Neurosci. Methods 2005, 144, 175.
S]GTPcS binding method is a major modification of the method by Selley, D. E.;