Discovery of agonists of cannabinoid receptor 1 with restricted central nervous system penetration aimed for treatment of gastroesophageal reflux disease

Article abstract of DOI:10.1021/jm301511h

Agonists of the cannabinoid receptor 1 (CB1) have been suggested as possible treatments for a range of medical disorders including gastroesophageal reflux disease (GERD). While centrally acting cannabinoid agonists are known to produce psychotropic effects, it has been suggested that the CB1 receptors in the periphery could play a significant role in reducing reflux. A moderately potent and highly lipophilic series of 2-aminobenzamides was identified through focused screening of GPCR libraries. Development of this series focused on improving potency and efficacy at the CB1 receptor, reducing lipophilicity and limiting the central nervous system (CNS) exposure while maintaining good oral absorption. Improvement of the series led to compounds having excellent potency at the CB1 receptor and high levels of agonism, good physical and pharmacokinetic properties, and low penetration into the CNS. A range of compounds demonstrated a dose-dependent inhibition of transient lower esophageal sphincter relaxations in a dog model.

Full text of DOI:10.1021/jm301511h

Subscriber access provided by University of Sussex Library
Article
Discovery of agonists of cannabinoid receptor 1 with restricted CNS
penetration aimed for treatment of gastroesophageal reflux disease
Alleyn T. Plowright, Karolina Nilsson, Madeleine Antonsson, Kosrat Amin, Johan Broddefalk,
Jörgen Jensen, Anders Lehmann, Shujuan Jin, Stephane St-Onge, Miros#aw J Tomaszewski,
Maxime Tremblay, Christopher S.J. Walpole, Zhongyong Wei, Hua Yang, and Johan Ulander
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm301511h • Publication Date (Web): 10 Dec 2012
Downloaded from http://pubs.acs.org on December 14, 2012
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of agonists of cannabinoid receptor 1 with
restricted CNS penetration aimed for treatment of
gastroesophageal reflux disease
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
Alleyn T. Plowright*, Karolina Nilsson, Madeleine Antonsson, Kosrat Amin, Johan Broddefalk, Jörgen
†
†
†
†
†
†
‡
‡
‡
Jensen, Anders Lehmann, Shujuan Jin, Stephane St-Onge, Mirosław J. Tomaszewski, Maxime Tremblay,
‡
‡
Christopher Walpole, Zhongyong Wei, Hua Yang, and Johan Ulander
‡
‡
†
†
‡
AstraZeneca Research and Development, Pepparedsleden 1, Mölndal, 43183, Sweden.
AstraZeneca Research and Development, 7171 Frederick-Banting, Saint-Laurent, Quebec, H4S 1Z9,
Canada
Corresponding Author: alleyn.plowright@astrazeneca.com
*
ABSTRACT. Agonists of the cannabinoid receptor 1 (CB1) have been suggested as possible treatments for
a range of medical disorders including gastroesophageal reflux disease (GERD). Whilst centrally acting
cannabinoid agonists are known to produce psychotropic effects it has been suggested that the CB1
receptors in the periphery could play a significant role in reducing reflux. A moderately potent and highly
lipophilic series of 2-aminobenzamides was identified through focused screening of GPCR libraries.
Development of this series focused on improving potency and efficacy at the CB1 receptor, reducing
lipophilicity and limiting the CNS exposure whilst maintaining good oral absorption. Improvement of the
series led to compounds having excellent potency at the CB1 receptor and high levels of agonism, good
physical and pharmacokinetic properties and low penetration into the CNS. A range of compounds
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 58
demonstrated a dose-dependent inhibition of transient lower esophageal sphincter relaxations in a dog
model.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
KEYWORDS CB1 agonists, CNS exposure, peripheral, drug design, gastroesophageal reflux disease,
aminobenzamides, ligand lipophilicity efficiency
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Introduction
1
The cannabinoid receptor 1 (CB1) is a GPCR that is located both in the CNS and on peripheral neurons.
Activation of CB1 receptors reduces calcium flux through voltage dependent calcium channels, resulting in
reduced neurotransmitter release. The endogenous ligand for the CB1 receptor, anandamide (1), has been
2
shown to block the N-type calcium channel (Figure 1). As a consequence of these effects agonists of the
CB1 receptor have been suggested as possible medical treatments for a range of disorders including pain,
inflammation, spasticity, neurodegenerative disorders, glaucoma and gastroesophageal reflux disease
3
-9
(
GERD). In healthy adults, the large majority of all reflux episodes results from transient lower
esophageal sphincter relaxation (TLESR) and is hence the single most important motility event as a cause of
10
reflux in all GERD patients. TLESRs are equally common in both healthy individuals and GERD patients.
However, TLESRs in GERD patients are more frequently associated with acid reflux than those in healthy
9
people. The centrally-acting mixed CB1/cannabinoid receptor 2 (CB2) agonists, WIN 55,212-2 (2) and ꢀ -
8
,9
tetrahydrocannabinol (3) (Figure 1), have been shown to dose dependently inhibit TLESRs in dogs.
Marinol, of which the active ingredient is 3, has been shown to reduce TLESRs in healthy human
9
volunteers. The effect of 2 on TLESRs in dog could be reversed by the selective CB1 antagonist,
11
rimonabant, but not by the selective CB2 antagonist, 5-(4-chloro-3-methylphenyl)-1-(4-methylbenzyl)-N-
12
[(1S,2S,4R)-1,3,3-trimethylbicyclo[2.2.1]hept-2-yl]-1H-pyrazole-3-carboxamide (SR144528), indicating
8
that CB1 alone mediates the action of 2 and thereby CB1 agonists rather than CB2 agonists were desired.
Although the exact location of CB1 receptors involved in the inhibition of TLESRs is still uncertain it has
13,14
been suggested that the CB1 receptors in the periphery could play a significant role.
Centrally acting
CB1 receptor agonists are known to produce psychotropic events which are not a viable therapeutic option
1
5
for GERD patients and therefore targeting CB1 receptors in the periphery would be required. Whilst a
number of cannabinoid ligands, particularly CB1 receptor antagonists, targeting peripheral restriction have
ACS Paragon Plus Environment

Page 3 of 58
Journal of Medicinal Chemistry
16-19
recently been reported,
potent CB1 or CB1/CB2 agonists which combine low CNS penetration together
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5,20
with good physicochemical, pharmacokinetic and in vivo efficacy characteristics are more scarce.
Herein, a novel CB1 receptor agonist series is reported with high in vitro binding affinities and full
2
1
agonism, good ligand lipophilicity efficiency (LLE) and physicochemical properties, and low ratios of
unbound concentration of compound in brain versus plasma (Cu,br/Cu,pl). Compounds in this series inhibit
TLESRs in a dose dependent manner in an in vivo dog model.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
Figure 1. Structures of CB1 receptor agonists: anandamide (1), WIN 55,212-2 (2) and ꢀ -
tetrahydrocannabinol (3).
Results and discussion
Design and Identification of the Hit Series
The driver for biological response is the concentration of the drug available to interact with the target
22,23
receptor in the required effect compartments.
In general, the amount of nonspecific binding differs
between brain and plasma and the pharmacologically relevant ratio is the unbound brain-to-plasma
24,25
concentration ratio, Cu,br/Cu,pl
.
The brain is separated from the systemic circulation by the BBB and the
blood cerebrospinal fluid barrier. The CNS endothelial cells differ from their peripheral counterparts by
2
6
having tighter junctions, limited pinocytic transport and no fenestrations. The net permeation across the
BBB is a combination between permeability and efflux properties and any substance with a finite
permeability can maintain asymmetry of unbound concentration only by active-transport. One of the more
2
7
prominent efflux transporters is P-gp. P-gp transporters are expressed both in the BBB and in the gut and it
26
has been reported that most marketed CNS-drugs are not PGP substrates. The x-ray structure of apo P-gp
2
8
has shed some light on putative mechanisms for P-gp mediated transport. The crystal structure reveals a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 58
3
large flexible cavity of 6000 Å making structure-based attempts to model P-gp specificity highly
challenging. In addition, P-gp is just one of several potential transporters. The transporter, breast cancer
resistance protein (BCRP) has been shown to have, along with P-gp, the highest expression in human brain
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
29
microvessels. Therefore, while one strategy in the present molecular design was to achieve a high level of
efflux from the CNS, it was not confined only to P-gp mediation. The extent of efflux (efflux ratio) was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
0
31
measured using Caco-2 cells. P-gp and BCRP are both expressed in the Caco-2 cell line. Brain exposure
was described as the steady-state unbound brain-to-plasma concentration ratio, Cu,br/Cu,pl measured in rats
where the compounds were administered as 4 h constant-rate intravenous infusions (both unbound and total
3
2
ratios are reported on key compounds for comparison). Figure 2 shows the C /C and Cu,br/Cu,pl ratios for
br pl
258 compounds from a variety of AstraZeneca projects where the colouring is based on measured Caco-2
33
efflux ratios. The relevance of the efflux ratio is demonstrated by a clear shift in the plot where compounds
with higher efflux generally have lower Cu,br/Cu,pl ratios.
Color by binned
HuCaco-2 Efflux Ratio
x < 0.4
0
0
1
.4 < x < 0.7
.7 < x < 1.3
.3 < x < 10
10 < x
ND
Log Cu,br/Cu,pl
Figure 2. Experimental data showing total (Cbr/Cpl) versus unbound (Cu,br/Cu,pl) ratios for 258 compounds
3
3
from a variety of AstraZeneca projects and marketed drugs. The coloring is based on experimental Caco-2
efflux data. The relevance of the efflux ratio is demonstrated by a clear shift in the plot where compounds
with higher efflux generally have lower Cu,br/Cu,pl ratios.
ACS Paragon Plus Environment

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
Much attention has recently been given to lipophilicity and in particular the consequences of making very
34
lipophilic compounds in the search for highly potent compounds. Lipophilicity influences permeability and
is an important parameter in the design of peripheral compounds. However, lipophilicity alone is not
expected to influence the asymmetry of unbound concentration directly, and the lipophilicity of non-CNS
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
35
drugs are not expected to differ from CNS drugs by default. LLE (pIC50-logD or pIC50-clogP) is one
2
1
simple measure of the degree to which the potency is driven by forces other than hydrophobicity. As the
lead compounds in this study were highly lipophilic, LLE was a key parameter driving the design. Two
human CB1 (hCB1) assays were used for estimation of CB1 potency, one competitive receptor binding
36,37
35
35
assay (IC50),
and one assay for stimulation of [ S]-guanosine 5’-O-(3-thiotriphosphate) (GTPγ[ S])
36,38
binding assay (EC50).
A corresponding competitive receptor binding assay was also utilized for the
36
estimation of human CB2 (hCB2) potency.
Screening a set of targeted GPCR libraries (75,000 compounds) led to the identification of the amino-
benzamide derivative 4 (Figure 3). Compound 4 displayed reasonable in vitro potency as measured in the
39
competitive receptor binding assay, but was highly lipophilic, and displayed partial agonism (Table 1).
Figure 3. Initial CB1 receptor agonist lead (4) identified from screening a set of targeted GPCR libraries
(
75,000 compounds).
Chemical Synthesis
The synthesis of the pyridyl and pyrazinyl derivatives are described in general in Schemes 1-3. Introduction
of the amide functional groups into the 2- and 3-position of the heterocyclic core structure was performed by
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 58
two methods. Coupling 3-aminopyridine-2-carboxylic acid 5a, 3-amino-6-methoxypyridine-2-carboxylic
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
40
acid 5b (synthesized from 3-(acetylamino)-6-methoxypyridine-2-carboxylic acid ) or 3-amino-pyrazine-2-
carboxylic acid 5c to the required acid chloride followed by heating in the presence of N,N,N′,N′-
tetramethyl-O-(7-azabenzotriazol-1-yl)uronium hexafluorophosphate (HATU) gave the oxazin-4-one
derivatives 6a-e (Scheme 1). Alternatively, the required amine was first coupled to 5a-c using HATU
mediated coupling conditions to give amides 7a-b. The oxazin-4-one derivatives 6a-e then underwent ring
opening upon treatment with the required amine, or the amides 7a-b underwent a second acid coupling with
the required substituted carboxylic acid to give the di-carboxamide derivatives 8a-o. Some compounds were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
derivatised further. For example, where R was a 4-substituted-1-naphthoic acid group, the acid was reduced
via the mixed anhydride to the alcohol, which was subsequently converted to a mesylate. The mesylate was
then displaced with a variety of nucleophiles to provide derivatives 9a-f. A methoxy group in the 6-position
1
of the pyridyl ring also provided a handle for further derivatisation. Treatment of 8n, where R was a
methoxy group, with pyridine hydrochloride at 160 ºC gave the pyridone which was subsequently O-
alkylated using either a variety of electrophiles in the presence of silver (I) carbonate at 130 ºC or using
41
Mitsunobu reaction conditions with the required alcohol to give the products 10a-b.
a
Scheme 1. Synthesis of substituted pyridine and pyrazine derivatives
a
3
Reagents and Conditions: (a) i. R C(O)Cl, N,N-diisopropylethylamine (DIPEA), DMF, CH
2
Cl
2
, 0 ºC ; ii. DIPEA, HATU, 50ºC;
2
2
3
(
b) R NH , DIPEA, DMF; (c) R NH , HATU, DIPEA, DMF; (d) R COOH, HATU, DIPEA, DMF; (e) i. oxalyl chloride, DCE, 70
2
2
ACS Paragon Plus Environment

Page 7 of 58
ºC; ii. NaBH
hydrochloride, 160 ºC; (i) Ag
Journal of Medicinal Chemistry
4
4
, I
2
; (f) CH
3
SO
2
Cl, Et
3
N, CH
2
Cl
2
, 0 ºC; (g) nucleophile (R H) (e.g. pyrrole for 9c), KI, DMF, 80 ºC; (h) pyridine
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
5
2
CO
3
, R Cl, DMF, 130ºC or R OH, diisopropyl azodicarboxylate, Ph
3
P, THF, 50 ºC. The structures
1
5
of R -R and the molecule cores are presented in Tables 1-3.
Further manipulation of the naphthalene ring led to the introduction of triazole and sulfur containing
functionality into the 4-position. The triazole derivatives were synthesized by first coupling 5a-c with 4-
methyl-1-naphthalenecarbonyl chloride and then trapping the resultant acid with methyl iodide to give the
esters 11a-c (Scheme 2). The methyl substituted naphthalene was then brominated under radical conditions
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
to give the bromomethyl intermediate, which then underwent a S 2 reaction upon treatment with 1,2,3-
triazole to give 12a-c. Finally the amide was introduced at the 2-position of the heterocyclic core by treating
1
1
2a-c with the required amine to give the triazoles 13a-e. Treatment of 13c, where R was a methoxy group,
with pyridine hydrochloride at 160 ºC gave the pyridone which was subsequently O-alkylated using silver
(
I) carbonate and 3,3,3-trifluoropropylsulphonyl chloride to give 14.
a
Scheme 2. Synthesis of triazole derivatives
a
Reagents and Conditions: (a) i. 4-methyl-1-naphthalenecarbonyl chloride, DIPEA, DMF, 50 ºC; ii. MeI, K
2
CO
3
; (b) i. NBS,
2
5
benzoyl peroxide, CCl ; ii. 1,2,3-triazole, DMF; (c) R NH , DMF, 80 ºC; (d) pyridine hydrochloride, 160 ºC; (e) Ag CO , R Cl,
4
2
2
3
1
2
5
DMF, 130ºC. The structures of R , R and R are presented in Tables 3-4.
The sulfur containing derivatives were synthesised by two methods (Scheme 3). Treatment of 5b-c with
HCl in methanol gave the methyl esters 15a-b, which were then reacted with 4-[(methylthio)methyl]-1-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 58
4
2
naphthoyl chloride (synthesized from 4-(bromomethyl)-1-naphthoic acid ) to give the amides 16a-b.
Treatment of 16a-b with the required amines gave the thiomethylnaphthalene intermediates 17b and 17d-e.
Alternatively, treatment of intermediates 7a-b with 4-[(methylthio)methyl]-1-naphthoyl chloride gave 17a
and 17c. Finally, oxidation of 17a-e using m-CPBA gave the sulfoxides 18a-g and the sulfone derivative 19.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
The R group was varied by treating 17e, where R was a methoxy group, with pyridine hydrochloride to
give the pyridone. The pyridone was then oxidised using m-CPBA to give the sulfoxide which was
subsequently O-alkylated with silver (I) carbonate and 2-(2-chloroethoxy)ethanol to give 20.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 3. Synthesis of sulfoxide derivatives
a
3
Reagents and Conditions: (a) HCl, MeOH; (b) 4-[(methylthio)methyl]-1-naphthoyl chloride, pyridine, DMAP, CHCl , 50 ºC (c)
2
5
2 2 2 2 3
R NH , DMF, 80 ºC; (d) m-CPBA, CH Cl , 0 ºC; (e) pyridine hydrochloride, 160 ºC; (f) Ag CO , R Cl, DMF, 130 ºC.
1
2
5
The structures of R , R and R are presented in Tables 3-4.
Biological Evaluation and Structure Activity Relationship Study
Explorative chemistry aimed at improving the overall properties showed that the phenyl core in the lead
compound 4 could be replaced with the pyridyl (8a) or pyrazinyl (8b) ring systems and lead to improved
ACS Paragon Plus Environment

Page 9 of 58
Journal of Medicinal Chemistry
potency (Table 1). In the case of the pyridyl-2-carboxamide derivative 8a, affinity to the hCB1 receptor was
increased greater than 10-fold and full efficacy achieved. The increase in binding affinity for 8a and 8b is
believed to be attributed to the intramolecular hydrogen bond that exists between either the pyridyl or
pyrazinyl nitrogen atom and the NH moiety of the 2-carboxamide group which conformationally restricts the
structure to provide a planar conformation. In contrast, moving the pyridine nitrogen atom to give the
corresponding 3-carboxamide (8c), 4-carboxamide (8d) or 2-amino-3-carboxamide (8e) derivatives gave a
significant decrease in potency compared to 4.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 1. In vitro binding, efficacy and lipophilicity data for the core variations
d
e
Compound
Core
hCB1 IC50
hCB2 IC50
(µM)
hCB1 EC50 hCB1 Emax logD
LLE
2.0
a
a
b
c
(
µM)
(µM)
(%)
2
0.059
0.16
0.0034
1.8
0.11
100
77
5.2
7.8
8.1
4
0.26
-1.0
-0.11
ND
8a
0.010
0.091
0.11
1.4
0.024
0.48
110
120
f
8b
ND
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8c
1.4
3.6
9.0
>8.9
>10
ND
ND
ND
ND
ND
ND
6.1
5.5
4.7
-0.24
-0.05
<0.30
8d
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8e
>10
a
IC50 values against hCB1 and hCB2 were determined using membranes from HEK293S cells expressing the cloned hCB1
receptor or Sf9 cells expressing the cloned hCB2 receptor and H-CP55,940 as the radioligand.
calculations were performed using Xlfit4 (IDBS, Inc). IC50 values were obtained from three independent experiments and are
3
36,37
Curve fitting and IC50
b
35
expressed as the mean. EC50 values from GTPγ[ S] binding assay were measured on cloned hCB1 receptors in membranes of
3
6,38
HEK 293S cells.
Curve fitting and EC50 calculations were performed using Xlfit4 (IDBS, Inc). EC50 values were obtained
c
from three independent experiments and are expressed as the mean. Emax is the maximal effect of the test compounds and
expressed as a percentage of that value obtained with 2. Chromatographic logD measurements performed in a similar fashion to
that described previously. LogD calculated from retention time versus standard curve. LLE defined as pIC50 (hCB1 IC50) –
d
3
9
e
21 f
logD. ND means not determined.
With the knowledge that good affinity and full agonism could be achieved the focus shifted towards
improving ADMET properties to meet the required peripheral candidate drug profile. The different positions
on the core structure were modified to build up knowledge of how different substitution patterns and groups
would affect these properties. The 2-carboxamide side chain of the pyridine ring was varied showing that
both lipophilicity and molecular weight could be reduced without significant loss of potency (Table 2). For
example, replacement of the cyclohexylmethyl moiety with tetrahydropyranylmethyl (8f) gave a 3-fold
reduction in binding affinity with a significant reduction in lipophilicity whilst replacement with
cyclobutylmethyl (8g) gave similar potency to 8a. Compound 8f showed significant CNS exposure (Cbr/Cpl
=
3), which was comparable to 2 (C /C = 2.5), highlighting that significant work was required to achieve
br pl
peripheral compounds in this series.
A large contributing factor to the overall lipophilicity of these compounds was the naphthalene in the 3-
position of the pyridine ring. Replacement of the naphthalene ring with alternative bicyclic systems, such as
benzodioxane (8h), benzothiazole (8i) or quinoxaline (8j) led to a drop in potency compared with the
naphthalene derivative 8f. Introduction of a nitrogen atom to give regioisomeric quinoline or isoquinoline
derivatives was also explored. The 5-substituted isoquinoline (8k) and 8-substituted quinoline (8l) both
showed reduced binding affinity compared with 8g. However, the 4-substituted quinoline (8m) displayed
ACS Paragon Plus Environment

Page 11 of 58
Journal of Medicinal Chemistry
greater potency than the other regioisomers and also maintained full efficacy and had no selectivity versus
hCB2. Further substitution in the 6-position of the pyridine ring of 8m with a methoxy group gave 8n which
was also well tolerated (Table 3).
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
As a nitrogen atom could be incorporated into the 4-position of the naphthalene ring, this led to an
investigation into alternative substitution at this position in an attempt to further enhance potency whilst
reducing lipophilicity. This was also an opportunity to introduce additional polarity, including hydrogen
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
Table 2. In vitro binding, efficacy, lipophilicity and LLE data for R and R variations
2
3
d
e
Compound
R
R
hCB1 IC50 hCB2 IC50 hCB1 EC50 hCB1 Emax logD
LLE
a
a
b
c
(
µM)
(µM)
(µM)
(%)
8
a
0.010
0.031
0.0061
1.2
0.11
0.024
0.081
0.023
ND
110
110
110
ND
ND
8.1
5.1
-0.11
2.4
8f
0.26
f
8g
0.026
5.7
ND
3.8
3.5
ND
2.1
8h
8
i
>6.7
>10
ND
<1.7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8j
>9.9
0.22
>10
0.038
0.36
ND
ND
ND
ND
ND
94
2.5
5.1
5.5
ND
<2.5
8k
1.5
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8l
0.32
ND
8
m
0.028
0.082
0.079
ND
a
IC50 values against hCB1 and hCB2 were determined using membranes from HEK293S cells expressing the cloned hCB1
receptor or Sf9 cells expressing the cloned hCB2 receptor and H-CP55,940 as the radioligand.
calculations were performed using Xlfit4 (IDBS, Inc). IC50 values were obtained from three independent experiments and are
3
36,37
Curve fitting and IC50
b
35
expressed as the mean. EC50 values from GTPγ[ S] binding assay were measured on cloned human CB1 receptors in membranes
3
6,38
of HEK 293S cells.
Curve fitting and EC50 calculations were performed using Xlfit4 (IDBS, Inc). EC50 values were obtained
c
from three independent experiments and are expressed as the mean. Emax is the maximal effect of the test compounds and
d
expressed as a percentage of that value obtained with 2. Chromatographic logD measurements performed in a similar fashion to
39
e
that described previously. LogD calculated from retention time versus standard curve. LLE defined as pIC50 (hCB1 IC50) –
2
1 f
logD. ND means not determined.
bond donors and acceptors, in an attempt to pick up recognition elements with CNS efflux transporters such
17,43
as P-gp and also modify molecular properties to restrict brain penetration.
Whilst acidic side chains such
as in 8o were not tolerated, it was found that a wide variety of polar substituents could successfully be
incorporated into this position to achieve high levels of hCB1 inhibition, including both neutral and basic
side chains (Table 3). This included substituents such as methoxymethyl (9a) and dimethylaminomethyl
(
9b). The pyrrole derivative (9c) lost potency compared to the unsubstituted naphthalene (8g), the imidazole
9d) and pyrazole (9e) derivatives had comparable potency whilst the triazole (9f) had excellent sub-
(
nanomolar binding affinity to hCB1 and full agonism. As expected, 9f had reduced lipophilicity relative to
a and hence achieved an improved LLE and also good selectivity over hCB2. This is most likely due to the
8
triazole ring picking up an additional binding interaction with the hCB1 receptor. Alternatively the
methylsulfoxide (18a) and methylsulfone (19) were also tolerated with the sulfoxide displaying significantly
greater binding affinity and LLE compared with the sulfone 19. As a methoxy group was well tolerated in
the 6-position of the pyridine ring as shown with 8n, additional modifications were made in this position.
ACS Paragon Plus Environment

Page 13 of 58
Journal of Medicinal Chemistry
This is exemplified with the hydroxyethers (10a and 10b) (Table 3). The derivative containing the
hydroxyethoxyethoxy side chain (10b) gave the highest affinity for hCB1 and highest LLE. Substitution at
the 6-position also had a large effect on the selectivity over hCB2, where increasing the size and length of
the substituent from a hydrogen atom to the hydroxyethoxyethoxy side chain in 10b increased the selectivity
significantly.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
Table 3. In vitro binding, efficacy, selectivity and LLE data for R and R variations.
1
3
d
e
Compound
R
R
hCB1 IC50 hCB2 IC50 hCB1 EC50 hCB1 Emax logD
LLE
a
a
b
c
(
nM)
(nM)
(nM)
(%)
8n
OMe
H
15
100
16
51
5.9
ND
ND
ND
ND
ND
1.9
f
g
8o
>10000
2.0
>10000
ND
ND
110
100
82
ND
ND
ND
ND
ND
9a
H
19
5.9
180
170
40
9b
H
41
190
9c
H
89
340
9d
H
16
330
100
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
9e
H
H
13
0.71
73
69
65
23
1.4
120
14
93
110
51
6.2
5.3
4
1.7
3.8
3.1
3.9
3.6
1.4
9f
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0a
0b
2500
1200
ND
ND
1
10
130
110
77
4.1
4.6
5.2
18a
19
H
H
7.0
240
5.9
260
a
IC50 values against hCB1 and hCB2 were determined using membranes from HEK293S cells expressing the cloned hCB1
receptor or Sf9 cells expressing the cloned hCB2 receptor and H-CP55,940 as the radioligand.
calculations were performed using Xlfit4 (IDBS, Inc). IC50 values were obtained from three independent experiments and are
3
36,37
Curve fitting and IC50
b
35
expressed as the mean. EC50 values from GTPγ[ S] binding assay were measured on cloned human CB1 receptors in membranes
3
6,38
of HEK 293S cells.
Curve fitting and EC50 calculations were performed using Xlfit4 (IDBS, Inc). EC50 values were obtained
c
from three independent experiments and are expressed as the mean. Emax is the maximal effect of the test compounds and
expressed as a percentage of that value obtained with 2. Chromatographic logD measurements performed in a similar fashion to
that described previously. LogD calculated from retention time versus standard curve. LLE defined as pIC50 (hCB1 IC50) –
logD. Compound 8o contains tetrahydropyranylmethyl instead of cyclobutylmethyl in the R position. ND means not
d
3
9
e
2
1
f
2
g
determined.
With the SAR of the three side chains established the best substituents were combined to address the
overall design criteria including achieving compounds with a very low Cu,br/Cu,pl ratio (Table 4). The series
of 1,4-naphthylmethyltriazoles (13a-d) achieved excellent hCB1 potency and showed a reduction in logD of
3
log units across the series whereas the drop in potency is less than 1 log unit. Thus, an increase in LLE is
achieved where the LLE for 13d is the greatest. Other properties including permeability, as measured in
Caco-2 cells, and solubility are also improved as logD decreases across this series. The Cu,br/Cu,pl ratio also
decreases from 13a to 13d. The polar surface area (PSA) is similar between 13a-d (PSA = 112 for 13a and
ACS Paragon Plus Environment

Page 15 of 58
Journal of Medicinal Chemistry
1
11 for 13d) and interestingly the efflux ratio decreases as the Cu,br/Cu,pl decreases. This may be caused by
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
several reasons, including that other transporter mechanisms are involved. The triazole derivatives 13a, 13c
and 13d maintained good selectivity over hCB2 whilst 13d also displayed full efficacy. At the beginning of
this investigation it was shown that a pyrazine ring could replace the pyridine core structure. The preferred
side chains from the SAR exploration with the pyridine derivatives were applied to the pyrazine core which
gave the triazole derivative 13e, a potent hCB1 receptor agonist. Further exploration of the 6-position of the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
44
pyridine ring led to the discovery of the 1,1,1-trifluoropropylsulfonate containing derivative 14, which was
a highly potent and full agonist of hCB1 with excellent selectivity over hCB2. Compound 14 had an even
lower Cu,br/Cu,pl ratio compared to 13d, most likely reflecting the increased PSA (158) as well as being an
efflux substrate as shown in the Caco-2 assay. A set of 1,4-naphthylmethylsulfoxides with different
substituents in the 2- and 6-position of the pyridyl ring were also prepared. This set of compounds, including
18b-c, again showed that good binding affinity against hCB1 can be achieved with compounds with
significantly reduced lipophilicity and hence improved LLEs. Compounds 18b-c have good permeability
and are efflux substrates in Caco-2 cells, which leads to low Cu,br/Cu,pl ratios. The enantiomers of the
sulfoxide moiety show a significant difference in binding affinity to hCB1 as exemplified by the
enantiomers of 18c, compounds 18d and 18e. The more potent enantiomer 18d has a high efflux ratio and is
peripherally restricted with a low Cu,br/Cu,pl ratio. Replacement of the pyridine core in 18a with a pyrazine
ring provided the sulfoxide derivative 18f, which had good potency against the hCB1 receptor. The
enantiomers of 18f were separated and, as with the pyridine core, one sulfoxide enantiomer (18g) showed
much greater potency versus hCB1. Analysis of the matched pairs of the pyridine and pyrazine containing
compounds showed that the pyrazine compounds were, in general, 5-10 fold less potent versus hCB1 with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
1
2
4
Table 4. In vitro binding, lipophilicity and Caco-2 data and Cu,br/Cu,pl data for the best R , R and R combinations.
1
2
4
d
e
h
i
Compound
R
R
R
hCB1
hCB2
hCB1
hCB1
Emax
logD
LLE
Hu Caco-2 Hu Caco-2 Cu,br/Cu,pl
Cbr/Cpl
IC50a
IC50a
EC50b
c
P
app A to B
-6
f
Efflux
Ratio
g
(
nM)
(nM)
(nM)
(%)
(1E cm/s)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
j
1
3a
OMe
OMe
OMe
H
0.85
0.97
2.4
5.0
4.3
14
110
ND
1.7
1.3
2.9
8.0
8.1
11
57
54
6.7
5.7
4.2
3.6
3.4
4.8
3.4
2.8
2.4
3.3
4.5
4.7
5.0
3.0
4.4
4.5
0.19
ND
0.18
0.18
0.31
13b
0.96
47
2.6
1.5
0.55
0.9
3.3
3.2
2.4
0.093
0.044
0.019
0.054
0.015
0.026
0.010
1
3c
470
60
0.10
1
3d
1200
ND
110
110
110
79
43
0.093
0.078
0.016
0.031
0.018
k
13e
H
50
1
4
>10000
>690
2500
21
1
8b
OMe
H
15
13
35
1
8c
49
87
110
23
ACS Paragon Plus Environment

Page 17 of 58
Journal of Medicinal Chemistry
1
2
3
l
1
8d
H
H
H
H
21
600
47
ND
ND
300
ND
500
31
ND
120
32
100
ND
120
110
120
2.9
2.9
2.7
2.7
3.2
4.8
3.3
4.6
4.7
4.9
26
0.61
20
17
ND
6.3
5.4
11
0.030
ND
0.031
ND
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
l
1
8e
8f
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
k
1
0.030
0.011
0.023
0.022
0.020
k,m
18g
37
9
20
7.2
5.1
26
0.020
a
3
IC50 values against hCB1 and hCB2 were determined using membranes from HEK293S cells expressing the cloned hCB1 receptor or Sf9 cells expressing the cloned hCB2 receptor and H-
CP55,940 as the radioligand.
as the mean. EC50 values from GTPγ[ S] binding assay were measured on cloned human CB1 receptors in membranes of HEK 293S cells.
performed using Xlfit4 (IDBS, Inc). EC50 values were obtained from three independent experiments and are expressed as the mean.
expressed as a percentage of that value obtained with 2. Chromatographic logD measurements performed in a similar fashion to that described previously. LogD calculated from retention
time versus standard curve. LLE defined as pIC (hCB1 IC ) – logD. Apparent permeability measurements carried out in Caco-2 cells at pH 7.4 for apical (A) to basolateral (B) transport.
Efflux ratio defined as Papp B to A/Papp A to B where both Papp A to B and Papp B to A are measured in Caco-2 cells. Calculated by combining the total brain to plasma ratio determined from
in vivo samples from Sprague-Dawley rats with estimates of the unbound fraction in the brain from rat brain slices, and unbound fraction in plasma by equilibrium dialysis. m The total
brain to plasma ratio determined from in vivo samples from Sprague-Dawley rats. ND means not determined. compounds contain pyrazine core. The two enantiomers of 18c. Most active
enantiomer of 18f.
36,37
Curve fitting and IC calculations were performed using Xlfit4 (IDBS, Inc). IC50 values were obtained from three independent experiments and are expressed
50
35 36,38
b
Curve fitting and EC50 calculations were
c
E
max
is the maximal effect of the test compounds and
d
39
e
21 f
50
50
g
h
4
5
32,46 i
j
k
l
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
7
similar efficacy. The pyrazines were also less lipophilic, had higher predicted PSA, similar selectivity
versus hCB2 and generally improved solubility (3-100 fold). Compounds 18f and 18g were peripherally
restricted with low C /C ratios. Finally, increasing the size of the 6-substituent on the pyridine core,
u,br u,pl
exemplified with compound 20, maintained excellent potency and led to excellent selectivity against hCB2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Now that compounds displaying good potency and efficacy versus hCB1 and low C /C ratios had
u,br u,pl
been identified, the concept of inhibition of TLESRs through CB1 receptor agonism was tested in an in vivo
9,48
dog model. Prior to testing in vivo, it was confirmed that compounds in this series, for example, 13c, 18b
and 20, showed excellent potency at the dog CB1 receptor with EC values that were similar to that shown
50
at the hCB1 receptor (Table 5). The compounds tested in vivo demonstrate a dose-dependent inhibition of
TLESRs in the dog, delivering greater than 50% inhibition for most of the compounds tested (Table 5).
Figure 4 shows the dose response plots for 13c and 14. The level of inhibition achieved is comparable in
magnitude to what has been shown previously for the more centrally acting CB1 agonists, 2 and 3, in the
8
,9
same animal model. In dogs, centrally acting cannabinoids induce a behavioral pattern known as static
49
ataxia, including sedation, prancing and motor incoordination. Single behavioral experiments with the
CB1 receptor agonists with restricted CNS penetration demonstrated an increased margin (e.g.
approximately 5 and 10 times for 13c and 14 respectively) between plasma concentrations producing 50%
inhibition of TLESRs and the plasma concentrations which induce these CNS side effects in comparison
with 2. Compound 2 showed no margin between the plasma concentrations required to reach 50% inhibition
of TLESRs and CNS side effects. The present data demonstrates the potential for achieving a robust
inhibition of TLESRs with CB1 agonists with restricted CNS penetration, reducing the risk for CNS
mediated side effects.
ACS Paragon Plus Environment

Page 19 of 58
Journal of Medicinal Chemistry
35
Table 5. In vitro dog CB1 EC values (GTPγ[ S] binding assay) and mean inhibition of transient lower
5
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
esophageal sphincter relaxations (TLESRs) in the dog in vivo model at different doses.
d
Compound Dog CB1
EC (nM)
Mean
+SEM
Dose
(µmol/kg)
n
a
c
inhibition ofb
50
TLESRs (%)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14.1
58
12.7
10.4
1.2
16.2
-
0.006
0.019
0.06
0.25
0.5
0.05
0.1
0.2
0.2
0.4
1.5
3
6
2
1
1
47
6
4
4
1
4
4
2
2
4
4
4
4
4
2
2
2
2
2
8
1.7
3.2
4
3a
3c
1.5
4.1
81.5
8.9
57.2
0.4
13.5
2
6.9
6.8
9.6
9.6
10.7
7.5
16.3
4.7
12.8
13
9
e
13d
ND
5
0
2
3
1.5
8.1
13e
31
28.6
33.5
0.1
0.2
0.4
0.05
0.1
1
14
19
58.4
38.6
62.2
2.6
23.4
2
18b
16
5.9
1.7
20
3.0
3
16.8
2
a
35
38
EC50 values from GTPγ[ S] binding assay were measured on cloned dog CB1 receptors in membranes of HEK 293S cells.
Curve fitting and EC50 calculations were performed using Xlfit4 (IDBS, Inc). EC values were obtained from three independent
experiments and are expressed as the mean. Inhibition of TLESRs through CB1 receptor agonism tested in an in vivo dog
The compounds tested demonstrated a dose-dependent inhibition of TLESRs. Administered intragastrically except
,14 and 18b, which were administered by an i.v. bolus. n = number of experiments for each dose. ND means not determined.
50
b
9
,48
c
model.
2
d
e
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Compounds 13c and 14 produced dose-dependent inhibition of transient lower esophageal
9,48
sphincter relaxations (TLESRs) in the dog in vivo.
The Figure shows mean percentage inhibition and
S.E.M. (n = number of experiments for each dose).
Further profiling of this series revealed that a number of compounds were strong inhibitors of the
cytochrome P450 (CYP) isoforms 3A4 and 2C9, in some cases with IC values around 100 nM (Table 6).
5
0
50
These interactions are highly undesirable due to potential drug-drug interactions in the clinic. The potency
1
versus CYP 2C9 was particularly sensitive to changes in R . Changing a H atom to a methoxy group
increased potency at CYP 2C9, for example comparing 13d with 13c or 18c with 18b. This effect of the
methoxy group on the interactions with CYP 2C9 could be decreased by increasing the size and flexibility
1
of the R substituent, as shown by comparing 18b with 20. The potency at CYP 3A4 was impacted to some
1
extent by the R substituent. However, it was particularly sensitive to variations on the naphthalene ring
4
(
R ). For example, whilst the methyltriazole substituted naphthalene in compound 13c significantly
increased the hCB1 potency and LLE, it also led to strong inhibition of CYP 3A4. It has been reported that
nitrogen atoms in heterocyclic ring systems can bind strongly to the haem moiety within the active site of
51
the CYP enzymes. Hence, attempts were made to substitute the triazole ring to increase the steric bulk
around the triazole nitrogen atoms to reduce the affinity. However, in this case this approach had minimal
impact on reducing the interactions. Alternatively, replacement of the triazole ring with a sulfoxide group
ACS Paragon Plus Environment

Page 21 of 58
Journal of Medicinal Chemistry
led to significantly reduced potency at the CYP enzymes, for example, comparing 13c and 18b. Applying
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
4
the learning from R and R in combination led to significantly reduced CYP interactions and provided
compounds with reduced inhibition of both CYP 2C9 and 3A4 isoforms (e.g. 18c). Applying the same
learning with the pyrazine core also provided compounds, such as 18f, with significantly reduced inhibition
of the CYP enzymes.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 6. CYP 3A4 and 2C9 inhibition data.
Compound
CYP 3A4
IC (µM)
CYP 2C9
a
a
IC (µM)
5
0
50
1
3c
0.36
0.11
0.66
1.6
1
3d
8b
0.61
8.8
1
20
5.5
6.7
18c
>20
>20
19
1
8f
>20
a
Inhibition values show the inhibition of metabolic degradation of the corresponding substrate by human recombinant CYP 3A4
and 2C9 at 37 °C (the percent of inhibition is determined at five different concentrations and is reported as IC ). Data were
50
analyzed using nonlinear regression analysis with ActivityBase software.
The most promising sulfoxide derivatives 18b, 18c and 18f were profiled further (Table 7). As well as
showing good permeability in Caco-2 cells and having low C /C ratios, these compounds display good
u,br u,pl
52
solubility and low inhibition of the hERG encoded potassium channel. These compounds were also tested
in in vivo rat PK studies and show moderate clearance and moderate to good oral bioavailability. For
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 58
example, despite efflux properties contributing to very low concentrations in the CNS, 18b showed
acceptable oral bioavailability and clearance whilst the pyrazine derivative 18f had similar clearance but
good oral bioavailability.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 7. C /C , rat PK, hERG and solubility data for leading derivatives.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
u,br u,pl
a
b
c
c
Compound C /C
C /C
br pl
Rat F
%)
Rat CL
hERG IC
(µM)
Solubility
(µM)
u,br u,pl
50
d
e
(
(mL/min/kg)
1
3c
0.044
0.10
0.031
0.018
0.022
5.8
21
31
52
22
17
14
26
11
>32
>33
17
3.7
93
79
87
1
8b
0.026
0.010
0.030
1
8c
18f
a
Calculated by combining the total brain to plasma ratio determined from in vivo samples from Sprague-Dawley rats with
4
5
32,46
estimates of the unbound fraction in the brain from rat brain slices, and unbound fraction in plasma by equilibrium dialysis.
b
c
The total brain to plasma ratio determined from in vivo samples from Sprague-Dawley rats. Compound 18b, 18c and 18f were
dosed at 2 ꢁmol/Kg (i.v.) and 10 ꢁmol/Kg (p.o.) and 13c was dosed at 1 ꢁmol/Kg (i.v.) and 15 ꢁmol/Kg (p.o.) to Sprague-
d
53 e
Dawley rats. Patch clamp assay using IonWorks technology in hERG-expressing CHO cells. Solubility determined by adding
Hanks balanced saline solution (pH 7.4) to a solution of the compound in DMSO and shaking for 24 h. The mixtures were then
filtered and analyzed to give an estimate of the solubility.
Conclusion
The design of efficient orally bioavailable yet peripherally restricted compounds involves optimizing
properties that are often in conflict with each other. Sufficient bioavailability is required, thus permeability
across the intestinal membrane is needed, whilst permeation across the BBB needs to be minimized. This
can be achieved as the concentration of the compound is much higher in the intestine, potentially leading to
54
saturation of efflux transporters. A moderately potent and highly lipophilic series of 2-aminobenzamides
was identified through focused screening of GPCR libraries. Opportunities were found to vary the core
ACS Paragon Plus Environment

Page 23 of 58
Journal of Medicinal Chemistry
structure to pyridine or pyrazine to increase potency at the hCB1 receptor by restricting the conformation
through an intramolecular hydrogen bond. The series was developed further focusing on improving potency
and efficacy at the hCB1 receptor and reducing lipophilicity to improve compound properties. Design
strategies were applied to limit CNS exposure, including increasing PSA and introducing polar functionality
in an attempt to increase recognition by CNS efflux transporters, whilst simultaneously maintaining
properties for good oral absorption. A rewarding strategy to achieve the required criteria was to introduce
polar functionality, particularly triazole or sulfoxide groups, into the 4-position of the naphthalene ring
located in the 3-position of the heterocyclic scaffold. This led to compounds with excellent potency at the
hCB1 receptor, lipophilicity values in the preferred range for oral drug-like compounds and hence,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
significantly improved LLEs. These modifications also gave compounds with low C /C ratios. Changes
u,br u,pl
were tolerated in the 6-position of the core heterocyclic ring structure allowing further modification of
compound properties as well as increasing selectivity versus hCB2. In general, the sulfoxide derivatives,
such as 18b and 18f gave higher efflux ratios than the triazoles and also better properties including less
inhibition of the CYP enzymes and good in vivo rat PK profiles.
The ultimate goal for this series was to demonstrate that CB1 agonists with restricted CNS penetration are
efficacious in inhibition of TLESRs. The in vivo data presented demonstrate that compounds from this
series, such as 18b, dose dependently inhibit TLESRs in dogs in a comparable way to what has previously
been reported for the centrally acting compounds, 2 and 3. This suggests that the peripheral concept of CB1
receptor agonism can deliver high efficacy whilst minimizing CNS-related side effects and have a potential
benefit for the treatment of GERD.
Experimental Section
Chemistry. All solvents and reagents were obtained from commercially available sources and used
without further purification. Reactions were carried out under nitrogen atmosphere unless otherwise stated.
Reactions carried out using a microwave reactor were performed using a Biotage Initiator or Personal
Chemistry [Biotage] Emrys Optimizer. Flash chromatography was carried out on prepacked silica gel
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 58
columns supplied by Biotage and using Horizon/Biotage systems. Analytical HPLC/MS was conducted on a
Waters Zevo QTof or Waters LCT Premiere mass spectrometer using an Acquity PDA (Waters) UV
detector monitoring either at (a) 210 nm with an Acquity BEH C18 column (2.1x100 mm, 1.7 µm, 0.7
mL/min flow rate), using a gradient of 2% v/v CH CN in H O (ammonium carbonate buffer pH10) to 98%
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
2
v/v CH CN in H O or (b) 230 nm with an Acquity HSS C18 column (2.1x100 mm, 1.8 µm, 0.7 mL/min
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
2
flow rate), using a gradient of 2% v/v CH CN in H O (ammonium formate buffer pH3) to 98% v/v CH CN
3
2
3
in H O. Preparative HPLC was conducted using a Waters Fraction Lynx Purification System using either (i)
2
Xbridge Prep C18 5 µm OBD 19 mm x 150 mm columns. The mobile phase used was varying gradients of
CH CN and 0.2% NH buffer; flow rate 30 mL/min, or (ii) Kromasil, C8, 10µm, 50.8 mm x 300 mm
3
3
columns. The mobile phase used was varying gradients of CH CN and 0.1M NH OAc buffer; flow rate 30
3
4
mL/min or (iii) Kromasil, C8, 10µm, 20 mm x 50 mm columns. The mobile phase used was varying
gradients of CH CN and 0.1 M HCO H buffer; flow rate 30 mL/min. MS-triggered fraction collection was
3
2
used for (a) – (c). All tested compounds were determined to be > 95% pure using the analytical methods (a)
1
or (b) described above based on the peak area percentage. H NMR spectra were generated on a Varian 300
MHz, Varian 400 MHz, Varian 500 MHz or Varian 600 MHz instrument as indicated. Chemical shifts (δ)
are given in parts per million (ppm), with the residual solvent signal used as a reference. Coupling constants
(J) are reported as Hz. NMR abbreviations are used as follows: br = broad, s = singlet, d = doublet, t =
triplet, q = quartet, m = multiplet.
N-[4-Chloro-2-[[(cyclohexylmethyl)amino]carbonyl]phenyl]-1-naphthalene-carboxamide (4). 1-Cyclo-
hexylmethanamine (6.8 g, 60 mmol) was added to a solution of 6-chloro-2H-3,1-benzoxazine-2,4(1H)-
dione (6.0 g, 30 mmol) and N,N-diisopropylethylamine (DIPEA, 3.8 g, 30 mmol) in DMF (50 mL) at rt.
After 2 h, the reaction mixture was quenched with H O (100 mL) and Et O (50 mL). The precipitate was
2
2
collected to provide 2- amino-5-chloro-N-(cyclohexylmethyl)benzamide (7.0 g, 87%).
A solution of 1-naphthoyl chloride (296 mg, l.5 mmol) in CH Cl (1 mL) was added to a mixture of DIPEA
2
2
(190 mg, 1.5 mmol), 2-amino-5-chloro-N-(cyclohexylmethyl)benzamide (270 mg, 1.0 mmol) in DMF (10
ACS Paragon Plus Environment

Page 25 of 58
Journal of Medicinal Chemistry
mL) at 0 °C. The reaction mixture was then stirred for 2 h at rt, and then quenched with H O (20 mL). The
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
precipitate was collected to provide the title compound (178 mg, 43%). H NMR (400 MHz, CDCl ) δ 0.99
3
(m, 2H), 1.22 (m, 3H), 1.56 (m, 1H), 1.75 (m, 5H), 3.23 (d, J = 6.4, 2H), 6.21 (brs, 1H), 7.46 (m, 1H), 7.53
(m, 4H), 7.84 (dd, J = 7.2, 1.2, 1H), 7.89 (m, 1H), 7.97 (d, J = 8.0, 1H), 8.52 (m, 1H), 8.88 (d, J = 9.2, 1H),
+
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
11.53 (brs, 1H); HRMS (ESI) m/z calcd for C H ClN O [M + H] 421.1683; found 421.168.
25
25
2
2
3
-Amino-6-methoxypyridine-2-carboxylic acid (5b). A mixture of 3-(acetylamino)-6-methoxypyridine-2-
40
carboxylic acid, (8.0 g, 37.9 mmol) and a 2.5 M aqueous solution of NaOH (80 mL) were refluxed for 80
min. The solution was cooled to 0 °C and 4 M aqueous HCl was added until the solution was pH 4. The
1
precipitate was collected and washed with cold H O to give the title compound as a solid (5.7 g, 89%). H
2
NMR (300 MHz, CD OD) δ 3.87 (s, 3H), 6.86 (d, J = 9.0, 1H), 7.25 (d, J = 9.0 Hz, 1H).
3
3-Amino-N-(tetrahydro-2H-pyran-4-ylmethyl)pyridine-2-carboxamide (7a). HATU (2.63 g, 6.93
mmol) and 1-(tetrahydro-2H-pyran-4-yl)methanamine (0.80 g, 6.94 mmol) were added to a solution of 3-
amino-2-pyridine carboxylic acid (5a, 0.91 g, 6.60 mmol) and DIPEA (1.26 mL, 7.26 mmol) in DMF (120
mL) at 0°C. The reaction mixture was warmed to rt and heated to 50 °C for 3 h. The solvent was
concentrated in vacuo and the residue was partitioned between EtOAc and water. The organic phase was
washed with saturated aqueous NaHCO solution, saturated aqueous NaCl solution and dried (Na SO ) and
3
2
4
concentrated in vacuo. The residue was purified by flash column chromatography (Et N 0.1%, MeOH 3%
3
1
and acetone 5% in CH Cl ) to give the title compound as a white solid (1.40 g, 90 %). H NMR (500 MHz,
2
2
CDCl ) δ 1.34-1.48 (m, 2H), 1.66-1.77 (m, 2H), 1.82-1.96 (m, 1H), 3.34 (t, J = 6.6, 2H), 3.37-3.45 (m, 2H),
3
4
.01 (dd, J = 11.2, 4.0, 2H), 5.96 (s, 2H), 7.03 (d, J = 11.2, 1H), 7.17 (dd, J = 8.3, 4.3, 1H), 7.81-7.91 (m,
1
H), 8.25 (s, 1H).
3-Amino-N-(cyclobutylmethyl)pyridine-2-carboxamide (7b). O-(Benzotriazol-1-yl)-N,N,N′,N′-tetra-
methyluronium tetrafluoroborate (2.32 g, 7.24 mmol) and DIPEA (2.02 mL, 11.6 mmol) were added to a
solution of 5a (1.00 g, 7.24 mmol) in DMF (5 mL) and after 15 min 1-cyclobutylmethanamine (0.62 g, 7.24
mmol) was added. The reaction mixture was stirred at rt for 1 h and was then partitioned between EtOAc
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 58
and H O. The organic phase was washed with H O, saturated aqueous solution of NaHCO and saturated
2
2
3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
solution of NaCl. The combined aqueous phases were extracted with EtOAc. The combined organic phases
was dried (Na SO ) and concentrated in vacuo. The residue was purified by flash column chromatography
2
4
1
(
heptane/EtOAc 1:1) to give the title compound as a clear oil (1.05 g, 71%). H NMR (400 MHz, CDCl ) δ
3
1
2
.68-1.81 (m, 2H), 1.83-1.97 (m, 2H), 2.01-2.15 (m, 2H), 2.46 – 2.65 (m, 1H), 3.34-3.5 (m, 2H), 5.93 (s,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H), 6.96 (dd, J = 8.3, 1.3, 1H), 7.12 (dd, J = 8.4, 4.3, 1H), 7.82 (dd, J = 4.3, 1.3, 1H), 8.07 (s, 1H).
N-(Cyclohexylmethyl)-3-[(1-naphthalenylcarbonyl)amino]-2-pyridinecarboxamide (8a). 1-Naphtha-
lenecarbonyl chloride (400 mg, 2.1 mmol) in CH Cl (2 mL) was added to a solution of 3-aminopyridine-2-
2
2
o
carboxylic acid (5a, 280 mg, 2.0 mmol) and DIPEA (280 mg, 2.2 mmol) in DMF (10 mL) at 0 C. The
reaction mixture was stirred for 16 h at rt, and then DIPEA (280 mg, 2.2 mmol) and HATU (0.84 g, 2.2
o
mmol) were added. After stirring for 1 h at rt, the reaction mixture was heated at 50 C to provide 2-(1-
naphthalenyl)-H-pyrido-[3,2-d][1,3]oxazin-4-one (6a) which was used in the next step with no further
purification.
A solution of 1-cyclohexylmethanamine (0.261 g, 2.3 mmol) in MeOH (0.5 mL) was added to a solution of
o
6
a (130 mg, 0.47 mmol) in DMF (2 mL) at 0 C. The reaction mixture was stirred for 18 h at rt and then
concentrated in vacuo to leave a residue, which was purified by flash chromatography (hexane/EtOAc 9:1)
1
to provide the title compound as a solid (172 mg, 95%). H NMR (400 MHz, CD OD) δ 0.90-1.00 (m, 2H),
3
1
.13-1.28 (m, 3H), 1.52-1.75 (m, 6H), 3.16 (d, J = 6.83, 2H), 7.55-7.61 (m, 4H), 7.88-7.90 (m, 1H), 7.94-
7
.96 (m, 1H), 8.05-8.07 (m, 1H), 8.36 (dd, J = 4.5, 1.6, 1H), 8.41-8.43 (m, 1H), 9.29 (dd, J = 8.6, 1.4, 1H);
+
HRMS (ESI) m/z calcd for C H N O [M + H] 388.2025; found 388.2014.
2
4
25
3
2
N-(Cyclohexylmethyl)-3-(1-naphthoylamino)pyrazine-2-carboxamide (8b). Following the procedure for
a, using 3-amino-pyrazine-2-carboxylic acid (5c, 139 mg, 1.0 mmol) in place of 5a gave 2-(naphthalen-1-
8
yl)-4H-pyrazino[2,3-d][1,3]oxazin-4-one (6b). Treatment of 6b (69 mg, 0.25 mmol) with 1-
cyclohexylmethanamine (113 mg, 1.0 mmol) provided the title compound as its TFA salt after purification
1
by reversed-phase HPLC (6 mg, 5%). H NMR (400 MHz, CD OD) δ 0.96 (m, 2H), 1.22 (m, 3H), 1.72 (m,
3
ACS Paragon Plus Environment

Page 27 of 58
Journal of Medicinal Chemistry
6
H), 3.19 (m, 2H), 7.55 (m, 3H), 7.95 (m, 2H), 8.06 (m, 1H), 8.48 (m, 3H); MS (ESI) m/z C H N O [M +
23 24 4 2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
H] found 389.0.
N-(Cyclohexylmethyl)-4-[(naphthalen-1-ylcarbonyl)amino]pyridine-3-carboxamide (8c). Following the
procedure for 8a, using 4-aminopyridine-3-carboxylic acid (138 mg, 1.0 mmol) in place of 5a gave 2-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(naphthalen-1-yl)-4H-pyrido[4,3-d][1,3]oxazin-4-one (6c). Treatment of 6c (0.137 mg, 0.5 mmol) with 1-
cyclohexylmethanamine (226 mg, 2.0 mmol) provided the title compound as its TFA salt after purification
1
by reversed-phase HPLC (39 mg, 16%). H NMR (400 MHz, CDCl ) δ 0.99 (m, 2H), 1.23 (m, 3H), 1.63
3
(m, 1H), 1.76 (m, 5H), 3.22 (d, J = 6.8, 2H), 7.61 (m, 3H), 7.98 (m, 2H), 8.14 (d, J = 8.4, 1H), 8.53 (m, 1H),
+
8
.72 (m, 1H), 9.05 (s, 1H), 9.22 (d, J = 6.8, 1H); HRMS (ESI) m/z calcd for C H N O [M + H]
24 25 3 2
3
88.2025; found 388.2064.
N-(Cyclohexylmethyl)-3-[(naphthalen-1-ylcarbonyl)amino]pyridine-4-carboxamide (8d). Following
the procedure for 8a, using 3-aminopyridine-4-carboxylic acid (138 mg, 1.0 mmol) in place of 5a gave 2-
(naphthalen-1-yl)-4H-pyrido[3,4-d][1,3]oxazin-4-one (6d). Treatment of 6d (137 mg, 0.5 mmol) with 1-
cyclohexylmethanamine (226 mg, 2.0 mmol) provided the title compound as its TFA salt after purification
1
by reversed-phase HPLC (55 mg, 22%). H NMR (400 MHz, CD OD) δ 0.99 (m, 2H), 1.22 (m, 3H), 1.70
3
(m, 6H), 3.22 (d, J = 7.2, 2H), 7.59 (m, 3H), 7.90 (dd, J = 7.2, 1.2, 1H), 7.96 (m, 1H), 7.99 (brs, 1H), 8.08
(d, J = 8.4, 1H), 8.47 (m, 1H), 8.64 (brs, 1H), 10.1 (brs, 1H); HRMS (ESI) m/z calcd for C H N O [M +
24
25
3
2
+
H] 388.2025; found 388.2064.
N-(Cyclohexylmethyl)-2-[(naphthalen-1-ylcarbonyl)amino]pyridine-3-carboxamide (8e). Following the
procedure for 8a, using 2-aminopyridine-3-carboxylic acid in place of 5a gave 2-(naphthalen-1-yl)-4H-
pyrido[2,3-d][1,3]oxazin-4-one (6e). Treatment of 6e (100 mg, 0.36 mmol) with 1-cyclohexylmethanamine
(226 mg, 2.0 mmol) provided the title compound as its TFA salt after purification by reversed-phase HPLC
1
(89 mg, 62%). H NMR (400 MHz, DMSO-d ) δ 0.76-0.93 (m, 2H), 0.99-1.14 (m, 3H), 1.39-1.79 (m, 6H),
6
3
.02 (t, J = 6.3, 2H), 7.30 (dd, J = 7.6, 4.9, 1H), 7.49-7.64 (m, 3H), 7.80 (dd, J = 7.2, 1.0, 1H), 7.94-8.10 (m,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 58
3
H), 8.33-8.39 (m, 1H), 8.48 (dd, J = 4.9, 1.8, 1H), 8.60 (brs, 1H), 11.24 (brs, 1H); HRMS (ESI) m/z calcd
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
for C H N O [M + H] 388.2025; found 388.2051.
24
25
3
2
3-[(1-Naphthalenylcarbonyl)amino]-N-[(tetrahydro-2H-pyran-4-yl)methyl]-2-pyridinecarboxamide
8f). Following the procedure for 8a, treating 6a (122 mg, 0.446 mmol) with 1-(tetrahydro-2H-pyran-4-
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
yl)methanamine (62 mg, 0.535 mmol) in place of 1-cyclohexylmethanamine provided the title compound as
1
a solid (139 mg, 90%). H NMR (400 MHz, CDCl ) δ 0.98 (m, 2H), 1.23 (m, 3H), 1.56 (m, 1H), 1.76 (m,
3
5
H), 3.25 (t, J = 6.4, 2H), 7.54 (m, 4H), 7.90 (m, 2H), 7.98 (d, J = 8.0, 1H), 8.28 (dd, J = 8.4, 1.6, 1H), 8.53
+
(m, 2H), 9.41 (dd, J = 8.4, 0.8, 1H), 12.87 (s, 1H); HRMS (ESI) m/z calcd for C H N O [M + H]
23 23 3 3
3
90.1818; found 390.1812.
N-(Cyclobutylmethyl)-3-[(1-naphthalenylcarbonyl)amino]-2-pyridinecarboxamide (8g). Following the
procedure for 8a, treating 6a (100 mg, 0.37 mmol) with 1-cyclobutylmethanamine (45 mg, 0.53 mmol) in
1
place of 1-cyclohexylmethanamine provided the title compound as a solid (160 mg, 83%). H NMR (400
MHz, CDCl ) δ 1.69-1.78 (m, 2H), 1.81-1.91 (m, 2H), 1.99-2.07 (m, 2H), 2.51-2.62 (m, 1H), 3.34 (d, J =
3
7
8
.0, 2H), 7.52-7.59 (m, 4H), 7.87-7.89 (m, 1H), 7.92-7.96 (m, 1H), 8.03-8.05 (m, 1H), 8.30-8.35 (m, 1H),
+
.42-8.45 (m, 1H), 9.27 (dd, J = 8.6, 1.2, 1H). MS (ESI) m/z C H N O [M + H] found 360.0.
22
21
3
2
3
-[(2,3-Dihydro-1,4-benzodioxin-5-ylcarbonyl)amino]-N-(tetrahydro-2H-pyran-4-ylmethyl)pyridine-
-carboxamide (8h). A mixture of 7a (125 mg, 0.53 mmol), crude 1,4-benzodioxan-5-carboxylic acid
2
chloride (0.88 mmol, prepared from 1,4-benzodioxan-5-carboxylic acid and oxalyl chloride) and DIPEA
0.15 mL, 0.88 mmol) in CH Cl (10 mL) were stirred for 16 h. A saturated aqueous solution of NaHCO
3
(
2
2
was added and the phases were separated. The organic phase was dried (phase separator) and concentrated
in vacuo. The residue was purified by preparative reversed phase HPLC to give the title compound as a
1
solid (150 mg, 71%). H NMR (400 MHz, CD OD) δ 1.34-1.47 (m, 2H), 1.68-1.81 (m, 2H), 1.89-2.01 (m,
3
1H), 3.31-3.40 (m, 2H), 3.41-3.51 (m, 2H), 3.95-4.04 (m, 2H), 4.37-4.44 (m, 2H), 4.52-4.57 (m, 2H), 6.98
(t, J = 7.9, 1H), 7.09 (dd, J = 1.6, 8.0, 1H), 7.5-7.62 (m, 2H), 8.36 (dd, J = 1.3, 4.4, 1H), 9.26 (dd, J = 1.3,
+
8
.6, 1H); HRMS (ESI) m/z calcd for C H N O [M + H] 398.1716; found 398.1725.
2
1
24
3
5
ACS Paragon Plus Environment

Page 29 of 58
Journal of Medicinal Chemistry
N-{2-[(Tetrahydro-2H-pyran-4-ylmethyl)carbamoyl]pyridin-3-yl}-1,3-benzothiazole-6-carboxamide
8i). Following the procedure for 8h, treating 7a (71 mg, 0.30 mmol) with 1,3-benzothiazole-6-carboxylic
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
acid chloride (1.0 mmol, prepared from 1,3-benzothiazole-6-carboxylic acid and oxalyl chloride) in place of
1
1
,4-benzodioxan-5-carboxylic acid chloride provided the title compound as a solid (54 mg, 45%). H NMR
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(400 MHz, CDCl ) δ 1.35-1.49 (m, 2H), 1.71 (brd, 2H), 1.84-1.98 (m, 1H), 3.34-3.44 (m, 4H), 3.99 (dd,
3
2
H), 7.50 (dd, 1H), 8.17-8.28 (m, 3H), 8.67 (brt, 1H), 8.71 (s, 1H), 9.14 (s, 1H), 9.34 (d, 1H), 13.30 (s, 1H);
+
HRMS (ESI) m/z calcd for C H N O S [M + H] 397.1334; found 397.1313.
2
0
20
4
3
N-{2-[(Tetrahydro-2H-pyran-4-ylmethyl)carbamoyl]pyridin-3-yl}quinoxaline-5-carboxamide
(8j).
Following the procedure for 8h, treating 7a (125 mg, 0.53 mmol) with quinoxaline-5-carboxylic acid
chloride (0.883 mmol, prepared from quinoxaline-5-carboxylic acid and oxalyl chloride) in place of 4-
1
benzodioxan-5-carboxylic acid chloride provided the title compound as a solid (63 mg, 30%). H NMR (400
MHz, CD OD) δ 1.33-1.48 (m, 2H), 1.68-1.80 (m, 2H), 1.87-2.01 (m, 1H), 3.33-3.38 (m, 2H), 3.39-3.49
3
(m, 2H), 3.93-4.04 (m, 2H), 7.56-7.68 (m, 1H), 7.98-8.09 (m, 1H), 8.3-8.46 (m, 2H), 8.65 (dd, J = 1.3, 7.4,
1H), 9.06 (d, J = 1.6, 1H), 9.13 (d, J = 1.6, 1H), 9.29 (dd, J = 1.3, 8.6, 1H); HRMS (ESI) m/z calcd for
+
C H N O [M + H] 392.1722; found 392.1708.
21
21
5
3
N-[2-[[(Cyclobutylmethyl)amino]carbonyl]-3-pyridinyl]-5-isoquinolinecarboxamide (8k). DIPEA (0.17
mL, 0.97 mmol) was added to a solution of 7b (100 mg, 0.49 mmol) and isoquinoline-5-carboxylic acid
(168 mg, 0.97 mmol) in DMF (20 mL) at 0 °C. The mixture was stirred for 20 min and then HATU (369
mg, 0.97 mmol) was added. The reaction mixture was stirred for 24 h at rt, and was then quenched with
water and extracted with EtOAc. The combined organic phases were combined and concentrated in vacuo to
leave a residue which was purified by reversed-phase HPLC to provide the title compound as its TFA salt
1
(97 mg, 42%). H NMR (400 MHz, CD OD) δ 1.47-1.96 (m, 4H), 2.02-2.10 (m, 2H), 2.58-2.65 (m, 1H),
3
3
8
.39 (d, J = 7.2, 2H), 7.62 (dd, J = 8.6, 4.6, 1H), 8.10 (dd, J = 8.3, 7.3, 1H), 8.39 (dd, J = 4.6, 1.4, 1H), 8.59-
.64 (m, 3H), 8.98-8.90 (m, 1H), 9.26 (dd, J = 8.6, 1.4, 1H), 9.73-9.80 (brs, 1H); HRMS (ESI) m/z calcd for
+
C H N O [M + H] 361.1664; found 361.1643.
21
20
4
2
ACS Paragon Plus Environment