Synthesis and Pharmacological Evaluation of 1-Phenyl-3-Thiophenylurea Derivatives as Cannabinoid Type-1 Receptor Allosteric Modulators

Article abstract of DOI:10.1021/acs.jmedchem.9b01161

We previously reported diarylurea derivatives as cannabinoid type-1 receptor (CB₁) allosteric modulators, which were effective in attenuating cocaine-seeking behavior. Herein, we extended the structure-activity relationships of PSNCBAM-1 (2) at the central phenyl ring directly connected to the urea moiety. Replacement with a thiophene ring led to 11 with improved or comparable potencies in calcium mobilization, [³⁵S]GTPγS binding, and cAMP assays, whereas substitution with nonaromatic rings led to significant attenuation of the modulatory activity. These compounds had no inverse agonism in [³⁵S]GTPγS binding, a characteristic that is often thought to contribute to adverse psychiatric effects. While 11 had good metabolic stability in rat liver microsomes, it showed modest solubility and blood-brain barrier permeability. Compound 11 showed an insignificant attenuation of cocaine seeking behavior in rats, most likely due to its limited CNS penetration, suggesting that pharmacokinetics and distribution play a role in translating the in vitro efficacy to in vivo behavior.

Full text of DOI:10.1021/acs.jmedchem.9b01161

Subscriber access provided by Stockholm University Library
Article
Synthesis and Pharmacological Evaluation of 1#Phenyl-3#Thiophenylurea
Derivatives as Cannabinoid Type-1 Receptor Allosteric Modulators
Thuy Nguyen, Thomas Gamage, Ann M. Decker, Daniel Barrus,
Tiffany L. Langston, Jun-Xu Li, Brian F Thomas, and Yanan Zhang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01161 • Publication Date (Web): 09 Oct 2019
Downloaded from pubs.acs.org on October 12, 2019
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 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 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.
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 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis and Pharmacological Evaluation of
1
‐Phenyl-3‐Thiophenylurea Derivatives as
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
Cannabinoid Type-1 Receptor Allosteric Modulators
a
a
a
a
a
Thuy Nguyen, Thomas F. Gamage, Ann M. Decker, Daniel Barrus, Tiffany L. Langston,
b
a
a,*
Jun-Xu Li, Brian F. Thomas, and Yanan Zhang
a
Research Triangle Institute, Research Triangle Park, North Carolina 27709, United States
b
Department of Pharmacology and Toxicology, University of Buffalo, the State University of
New York, Buffalo, New York 14214, United States
Abstract
We previously reported diarylurea derivatives as cannabinoid type-1 receptor (CB ) allosteric
1
modulators which were effective in attenuating cocaine seeking behavior. Herein, we extended the
structure-activity relationships of PSNCBAM-1 (2) at the central phenyl ring directly connected
to the urea moiety. Replacement with a thiophene ring led to 11 with improved or comparable
3
5
potencies in calcium mobilization, [ S]GTPγS binding, and cAMP assays, whereas substitution
with non-aromatic rings led to significant attenuation of the modulatory activity. These compounds
3
5
had no inverse agonism in [ S]GTPγS binding, a characteristic that is often thought to contribute
to adverse psychiatric effects. While 11 had good metabolic stability in rat liver microsomes, it
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 68
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
showed modest solubility and blood-brain barrier permeability. 11 showed an insignificant
attenuation of cocaine seeking behavior in rats, most likely due to its limited CNS penetration,
suggesting that pharmacokinetics and distribution play a role in translating the in vitro efficacy to
in vivo behavior.
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
ACS Paragon Plus Environment

Page 3 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The cannabinoid type-1 and type-2 receptors (CB and CB ) belong to the class A Rhodopsin-like
1
2
superfamily of G protein-coupled receptors (GPCRs), with CB being one of the most abundantly
1
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
expressed receptors in the brain. CB plays an important role in many physiological processes,
1
such as addiction, pain, learning and memory, appetite and feeding behaviors, anxiety and
depression.2-4 CB agonists such as (-)-trans-Δ -tetrahydrocannabinol (THC), the major
9
1
phytocannabinoid found in marijuana, have been known to induce appetite and weight gain as well
as euphoria, tolerance, dependence, and withdrawal, whereas CB inhibition has been widely
1
investigated to develop therapeutic interventions for obesity, metabolic disorders, and substance
abuse.5-11 Unfortunately, rimonabant, the first CB inverse agonist/antagonist that received
1
approval in Europe for the treatment of obesity in 2006 was subsequently withdrawn due to adverse
effects including depression and suicidal ideation.
Despite this setback, the research community continues to target this important receptor system
and explore different strategies to overcome the adverse psychiatric effects of CB1
antagonism/inverse agonism while preserving the beneficial therapeutic effects. These strategies
include neutral antagonists, peripherally restricted antagonists, and allosteric modulators.12-17 Like
many GPCRs, CB displays a high level of constitutive activity in the absence of exogenous
1
ligands in both neurons18, 19 and non-neuronal cells. Rimonabant’s inhibition of CB constitutive
20
1
activity in the central nervous system (CNS) may underly its psychiatric side effects;21-24 therefore,
neutral antagonists which maintain basal CB signaling may not have these side effects. In
1
addition, peripherally restricted antagonists which do not cross the blood-brain barrier (BBB) have
shown promising therapeutic efficacy in the treatment of obesity, diabetes, and nonalcoholic
2
5
steatohepatitis (NASH) without the liability of the CNS side effects, although these drugs are
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 68
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
unviable for the treatment of substance abuse for obvious reasons. Finally, CB allosteric
1
antagonists, such as Org27569 (1) and PSNCBAM-1 (2) (Fig. 1),26-28 bind to a location that is
topographically distinct from the orthosteric site and can reduce signaling in a probe dependent
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
manner.29
Allosteric modulators offer several advantages compared to orthosteric ligands, including lower
risk of overdosing due to the “ceiling” effect and preservation of the temporal characteristics of
1
5
receptor signaling. In particular with CB , the allosteric antagonists of the CB reported so far
1
1
are positive allosteric modulator (PAM)-antagonists,29 since they increase the affinity of
orthosteric agonists for the receptor but decrease their signaling capability. This unique mechanism
may provide great advantages for therapeutic blockade of pathological signaling by receptors such
as CB . Compared to competitive CB antagonists/inverse agonists whose potency diminishes with
1
1
increasing agonist concentration, the potency of PAM-antagonists is enhanced as agonist
concentration increases due to the bidirectionality of the cooperativity vector.29
With our long-standing interest in developing CB allosteric modulators for the treatment of drug
1
addiction, we have conducted structure-activity relationship (SAR) studies on both 1 and 2, two
30, 31
of the most characterized CB allosteric antagonists (Fig. 1).
Our initial efforts in optimization
1
3
0
of 2 led to compound RTICBM-28 (3) (Fig 1), which has comparable in vitro potencies. Further
SAR efforts demonstrated that the pyrrolidinyl ring is not required for the CB modulatory activity
1
and the pyridinyl ring could be replaced with other substituted phenyl rings or five-membered
heterocycles, leading to the discovery of RTICBM-74 (4) and RTICBM-229 (5), respectively (Fig
32
1
).32, 33 We found that 4 attenuated prime-induced restatement of cocaine seeking and 3 reduced
3
4
THC’s potency in drug discrimination, demonstrating the therapeutic potential of these CB
1
allosteric modulators for the treatment of the relapse of cocaine addiction and THC dependence.
ACS Paragon Plus Environment

Page 5 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Cl
O
N
N
N
Cl
HN
O
N
N
H
H
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
N
H
PSNCBAM-1 (2)
hCB1 Ca2+ assay IC50 = 33 nM
Org27569 (1)
2
+
hCB Ca assay IC50 = 853 nM
[3
mCB1 GTPS assay IC50 = 115 nM
1
3
H]CP55,940 binding K = 217 nM,  = 6.9
[ H]CP55,940 binding KB = 54 nM,  = 7.3
B
Reduced food intake
Rat liver microsomes T1/2 = 13 min
Reduced food intake and body weight
Attenuated the reinstatement of
cocaine re-seeeking behavior
Reduced CP55,940-induced hypothermia in rats
Attenuated cue and drug-induced reinstatement
of cocaine- and methamphetamine-seeking behavior
NC
O
N
N
N
N
H
H
RTICBM-28 (3)
2+
hCB Ca assay IC50 = 33 nM
1
Reduced THC's potency in drug discrimination test
Cl
O
Cl
O
N
N
H
H
N
N
H
H
F
HN
RTICBM-74 (4)
RTICBM-229 (5)
2
+
hCB Ca assay IC50 = 23 nM
1
2
+
hCB Ca assay IC50 = 169 nM
1
mCB GTPS assay IC = 151 nM
1
50
mCB GTPS assay IC = 40 nM
H]CP55,940 binding K = 1.4 nM
d
1
50
Rat liver microsomes T1/2 > 300 min
Attenuated the reinstatement of
cocaine re-seeeking behavior
3
[
Figure 1. Structures of representative indole-based and diarylurea-based CB allosteric
1
modulators with in vitro and in vivo potency.27, 30, 32-39
Herein, we report our optimization effort to extend the SAR understanding of this diarylurea
scaffold by structural optimization at the central phenyl ring (Fig. 2), a position that has not yet
been investigated. Specifically, we replaced this phenyl ring with a variety of heterocylic rings
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 68
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
including pyridine, thiophene and thiazole, as well as non-aromatic rings such as cyclopropyl or
piperidinyl rings. Newly synthesized compounds were evaluated in human calcium mobilization
3
5
and cAMP assays as well as human and mouse [ S]GTPγS binding assays for their potencies in
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
modulating the activity of the orthosteric agonist CP55,940. The compounds were additionally
3
5
assessed alone for their effects on the CB basal tone in the [ S]GTPγS binding assay. Select
1
compounds were also examined in preliminary ADME assays, including metabolic stability
against rat liver microsomes, blood-brain barrier permeability and aqueous solubility. Finally, the
most potent compound from this series, 11, was evaluated in the reinstatement of cocaine seeking
behavior in rats.
adding substituent or
replacing with other
aromatic or aliphatic linker
retained or removed
Cl
O
N
N
N
N
H
H
replaced with phenyl
or substituted phenyl
PSNCBAM-1 (2)
Figure 2. Strategies to investigate SAR of 2.
RESULTS AND DISCUSSION
Chemistry
3
0
To prepare 6, 2‐bromo‐6‐(pyrrolidin‐1‐yl)pyridine underwent Suzuki coupling with 4-methoxy-
3-nitro-phenylboronic acid to give 36 which was subsequently reduced by Raney-nickel and
hydrazine to 37. Coupling of 37 with 4-chlorophenyl isocyanate yielded the desired product 6
(Scheme 1).
ACS Paragon Plus Environment

Page 7 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1
MeO
MeO
b
MeO
O
Cl
Br
N
N
a
N
N
N
N
c
O N
2
H N
2
N
N
N
N
H
H
3
6
37
6
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
Reagents and conditions: a) 4-methoxy-3-nitro-phenylboronic acid, Pd(PPh ) , aq. NaHCO ,
3 4
3
o
DME, reflux, 16 h, 30%; b) N H .H O, Raney-Ni, EtOH, 60 C, 1 h, 93%; c) 4-ClPhNCO, CHCl ,
2
4
2
3
o
6
0 C, 16 h, 64%.
To prepare the pyridinyl analogs, the corresponding bromopyridinyl amines underwent Suzuki
coupling with phenylboronic acid to give intermediates 38 – 41, which were subsequently coupled
with 4-chlorophenyl isocyanate to afford compounds 7 - 10 (Scheme 2).
Scheme 2
Z
Cl
Z
X
Z
X
Y
W
O
Y
W
b
Y
W
a
H N
X
2
N
N
H N
2
Br
H
H
38 - 41
7
- 10
38, 7: Y=N; X,Z,W=CH
9, 8: X=N; Y,Z,W=CH
0, 9: Z=N; X,Y,W=CH
1, 10: W=N; X,Y,Z=CH
3
4
4
Reagents and conditions: a) phenylboronic acid, Pd(PPh ) , aq. NaHCO , DME, reflux, 16 h, 44-
3
4
3
o
89%; b) 4-ClPhNCO, CHCl , 60 C, 16 h, 35-60%.
3
Commercially available bromothiophenecarboxylic acids underwent Suzuki coupling with
phenylboronic acid to afford the intermediates 42 and 43. The urea derivatives 12 and 13 were
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 68
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
obtained via a microwave-assisted coupling of these carboxylic acids via a Curtius rearrangement40
with 4-chlorophenylamine in the presence of diphenylphosphoryl azide. Similarly, 15 and 16 were
afforded from the Curtius rearrangement reaction of cis- or trans-2-phenylcyclopropane-1-
carboxylic acid and 4-chlorophenylamine (Scheme 3).
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
Scheme 3
Cl
X Y
X Y
X
Y
O
b
a
Br
HOOC
HOOC
N
N
H
H
42 - 43
12 - 13
42, 12: X=S; Y=CH
43, 13: X=CH; Y=S
Cl
O
b
HOOC
N
H
N
H
cis or trans mixture
15 (trans mixture)
6 (cis mixture)
1
Reagents and conditions: a) phenylboronic acid, Pd(PPh ) , aq. NaHCO , DME, reflux, 16 h, 62-
3 4
3
o
99%; b) 4-ClPhNH , DPPA, Et N, toluene, mw, 100 C, 5 min, 65-83%.
2
3
As the Suzuki coupling between 2-bromo-5-nitrothiazole and phenylboronic acid failed to give the
desired product 14, a different route was sought to form the thiazole ring. Phenyl acetaldehyde was
treated with bromine to give 2‐bromo‐2‐phenylacetaldehyde, which underwent cyclization with
4
1
thiourea to give the intermediate 44. Coupling of 44 with 4-chlorophenyl isocyanate yielded 14
Scheme 4).
Scheme 4
(
ACS Paragon Plus Environment

Page 9 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
.
Cl
O
a, b
c
O
N
H N
2
S
N
N
H
S
HBr
H
4
4
14
o
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
Reagents and conditions: a) bromine, DCM, -10 C, 30 min then reflux, 16 h b) thiourea, EtOH,
o
reflux, 8 h, 2-step yield 77%; c) 4-ClPhNCO, CHCl , 60 C, 16 h, 34%.
3
Copper-catalyzed coupling between (R)-3-(Boc-amino)piperidine and phenylboronic acid yielded
the intermediate 45. Removal of the Boc protecting group gave 46, which underwent coupling
with 4-chlorophenyl isocyanate to afford 17 (Scheme 5).
Scheme 5
Cl
O
a
N
b
N
c
BocHN
H N
2
N
NH
N
N
BocHN
.
HCl
H
H
4
5
46
17
o
Reagents and conditions: a) PhB(OH) , Cu(OAc) , Et N, DCM, 60 C, 72 h, 32%; b) 4N HCl/1,4-
2
2
3
o
dioxane, rt, 1 h, quant.; c) 4-ClPhNCO, CHCl , 60 C, 16 h, 32%.
3
Standard procedures were followed to prepare the 5-phenyl-thiophen-2-yl analogues, starting with
Suzuki coupling between 2-bromo-5-nitrothiophene with corresponding substituted phenylboronic
acid to give intermediates 47 - 65, which were reduced by Raney-Ni and hydrazine to amines 66 -
84. Coupling of these amines with 4-chlorophenyl isocyanate afforded compounds 11 and 18 - 35
(
scheme 6).
Scheme 6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 68
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
a
Cl
R
b
c
R
R
O
X
Y
Br
O N
S
S
2
H N
2
O N
2
S
N
N
S
H
H
47 - 65
66 - 84
1
1, 18 - 35
X, Y = C or N
R = H, F, diF, Cl, diCl, Ac, COOMe,
SO Me, OMe, Me, NMe
2
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
Reagents and conditions: a) corresponding phenylboronic acid, Pd(PPh ) , aq. NaHCO , DME,
3
4
3
o
reflux, 16 h, 8-85% b) N H .H O, Raney-Ni, EtOH, 60 C, 1 h, 12%-quant. Yield; c) 4-ClPhNCO,
2
4
2
o
CHCl , 60 C, 16 h, 7-88%.
3
Biological Evaluations
We assessed the ability of our new compounds to antagonize the CB -mediated responses to
1
3
5
30, 32, 33
CP55,940 in calcium mobilization and [ S]GTPγS binding assays as previously described.
The calcium mobilization assay was conducted with CHO-RD-HGA16 cells overexpressing the
human CB and the promiscuous Gα protein, which enables the CB to signal through
1
16
1
35
mobilization of internal calcium. The [ S]GTPγS binding assays were conducted in HEK293 cells
overexpressing the human CB1.32, 33 Select compounds were also assessed in the [ S]GTPγS
35
3
3
binding assay in mouse cerebellum which has a high expression of CB . Table 1 shows the IC
1
50
values against the EC concentration of CP55,940 (100 nM) in all three assays.
8
0
SAR of the central phenyl ring of 2 (Compounds 6 – 17)
Table 1. Allosteric Modulating Activities of Compounds 2 and 6 - 17 in the CB Calcium
1
3
5
Mobilization and [ S]GTPγS Binding Assays.
ACS Paragon Plus Environment

Page 11 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Cl
O
Ar -Ar
1
2
N
N
H
H
hCB1
hCB1
mCB1
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
3
5
35
Calcium [ S]GTPγS [ S]GTPγS
Compound
Ar1
Ar2
assay
IC50
binding
assay
binding
assay
nM)a
IC (nM)
b
IC (nM)
c
(
50
50
4
55 (307 – 288 (152 –
N
N
2
6
33 ± 8
6
73)
589)
MeO
1
270 ±
20
N
N
>10,000
N.D.
4
1
56 ±
N
7
8
Ph
>10,000
>10,000
>10,000
N.D.
71d
3530 ±
40
Ph
Ph
Ph
2
N
N
9
292 ± 55
233 ± 34
>10,000
N.D.
N.D.
2
770 (1830
4190)
N
1
0
–
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 68
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
24 (283 –
969)
63 (42 –
92)
1
1
1
2
Ph
Ph
Ph
Ph
6.8 ± 1
S
667 (417 – 776 (546 –
S
37 ± 2
60 ± 8
79 ± 10
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
060)
1103)
8
6
88 (621 –
1270)
S
13
N.D.
04 (348 –
1050)
N
1
1
4
5
N.D.
S
1
220 ±
90
Ph
Ph
Ph
>10,000
>10,000
>10,000
>10,000
1
trans mixture
1
1
6
7
>10,000e
N.D.
N.D.
cis mixture
N
3
880 ±
140
a
Against 100 nM CP55,940 in stable human CB -CHO-RD-HGA16 cells. Values are the mean ±
1
b
SEM of at least three independent experiments in duplicate. Against 100 nM CP55,940 in
HEK293 cells stably expressing CB . Values are expressed as mean (95% confidence interval)
1
c
from at least three independent experiments in duplicate. Against 100 nM CP55,940 in mouse
cerebellar membranes. Values are expressed as mean (95% confidence interval) from at least three
d
e
independent experiments in duplicate. Partial inhibition at 33%. Value is the result of two
independent experiments in duplicate. N.D. Not determined
ACS Paragon Plus Environment

Page 13 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
First, we introduced a methoxy group at the 2-position of the central phenyl ring (6) to probe the
electronic properties of this ring. As the diarylurea possesses a flat structure, resulting in tight
packing and limited aqueous solubility, this 2-methoxy group also introduces steric hindrance, thus
disrupting the planar structure and preventing the tight packing. Unfortunately, the allosteric
modulating activity was significantly weakened (6, IC = 1270 nM) compared to 2 (IC = 33
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
5
0
50
nM).
3
2
As previously demonstrated, the pyrrolidinylpyridine ring can be replaced with a phenyl ring
without compromising the activity. Therefore, to simplify the synthesis effort, the SAR studies at
the central phenyl ring were conducted without the pyrrolidinyl ring. The central phenyl ring was
replaced with various aromatic heterocycles such as pyridine, thiophene, and thiazole. All three
pyridinyl analogues (7, 9, 10) exhibited a modest drop in activity and were 4.7-fold, 8.8-fold, and
7
.1-fold, respectively, less potent than 2. The activity of the 2,6-pyridinyl analogue (8) was
significantly diminished (IC = 3530 nM). Interestingly, the five membered ring analogues,
5
0
thiophenes and thiazole, all displayed activities better than their six-membered pyridinyl
counterparts. In particular, the 2,5-thiophenyl analogue (11, IC = 6.8 nM) had significantly
5
0
improved activity compared to 2 (Fig 3). Other thiophene and thiazole (12 - 14) analogues had
comparable potencies to 2, with 12 being the most active of the three compounds (IC = 37 nM).
50
When the central phenyl ring was replaced with non-aromatic rings such as cyclopropyl (15 and
1
6) or piperidinyl (17), the allosteric modulating activity was greatly diminished and resulted in
IC values of 1220 nM (15), >10,000 (16), and 3880 nM (17).
50
3
5
The results from the [ S]GTPγS binding assay against human CB were generally in agreement
1
with the calcium mobilization assays with similar potency ranking orders, albeit the potencies
appeared to be weaker. In general, compounds with weak activities (6, 8, 15 - 17) in the calcium
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 68
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
35
assay were mostly inactive in the hCB [ S]GTPγS binding assay. The pyridinyl analogues (7, 9,
1
and 10) with moderate activities in the calcium assay demonstrated no or weak activities in the
3
5
[ S]GTPγS binding assay. In contrast, the five membered ring thiophene and thiazole analogues
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
(
11 - 14) still maintained good activities in the [ S]GTPγS binding assay comparable to that of 2.
35
Evaluation in the [ S]GTPγS binding assay using mouse cerebellum, which natively expresses
CB , revealed some differences between assays and species/source of receptors. In particular, 7
1
3
5
had no activities in both [ S]GTPγS binding assays but displayed a moderate activity in the
3
5
calcium assay. 11 demonstrated better potencies in the calcium and [ S]GTPγS binding assay
3
5
against mouse CB than 2 but appeared to be equipotent in the [ S]GTPγS binding assay against
1
3
5
human CB . On the other hand, 12 had comparable potencies in the calcium and [ S]GTPγS
1
3
5
binding assay against human CB but weaker potency in the [ S]GTPγS binding assay against
1
mouse CB . Lastly, 15 had weak or no activities across three assays. These discrepancies may be
1
attributed to the different G protein coupling (Gα vs. Gα ) or species (human vs. mouse) between
1
6
i/o
these assays (see further discussions below).
Finally, consistent with the positive allosteric modulation, 2 increased the specific binding of
[3H]CP55,940, an effect that has been well-established.27, 30, 32 Interestingly, 11 increased the %
specific binding of the radioligand to a greater degree (343.1±63.94), than did 2 (190.0±29.65)
following Student’s t-test (t(4) = 3.761, p<0.05; Fig. 3D), suggesting greater cooperativity.
ACS Paragon Plus Environment

Page 15 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CP55,940
B
A
11 + CP55,940 100 nM
CP55,940 100 nM
Vehicle
2
1
1
125
1
00
100
7
5
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
75
50
50
2
5
25
0
0
-
11 -10
-9
-8
-7
-6
-5
-4
-12
-10
-8
-6
-4
log [compound] M
log [compound] M
D
350
C
2
1
1
100
2
300
11
7
5
0
2
2
1
1
50
00
50
00
5
25
0
5
0
-10
-9
-8
-7
-6
-5
-4
-10
-9
-8
-7
-6
-5
-4
log [compound] M
log [compound] M
Figure 3. Activities of 2 and 11 against 100 nM CP55,940 in A) calcium mobilization assay in
35
human CB -CHO-RD-HGA16 cells stably expressing the human CB . B) [ S]GTPγS binding
1
1
3
5
assay in HEK293 cells stably expressing the human CB . C) [ S]GTPγS binding assay in cerebella
1
of male ICR mice. D) Effects of 2 and 11 on the equilibrium binding levels of 1 nM of
3
[ H]CP55,940 to membranes of HEK293 cells stably expressing the human CB .
1
Substituted Phenylthiophenyl analogues and their in vitro pharmacological activities
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 68
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 2. Allosteric Modulating Activities of Compounds 2, 11, and 18 - 35 in the hCB Calcium
1
3
5
Mobilization and hCB and mCB [ S]GTPγS Binding, and cAMP BRET Assays.
1
1
Cl
O
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
Ar
N
N
S
H
H
hCB1
mCB1
[35S]GTPγS
binding
hCB1
[35
S]GTPγS
hCB1
Calcium
assay
Compound
Ar
binding
assay
cAMP IC50
(nM)d
assay
a
IC (nM)
5
0
b
c
IC (nM)
IC (nM)
5
0
50
455 (307 –
73)
288 (152 –
589)
2620 (1690-
4070)
2
-
33 ± 8
6.8 ± 1
40 ± 2
39 ± 6
6
5
24 (283 –
969)
1760 (1260-
2440)
1
1
1
8
63 (42 – 92)
174 (78-399)
N.D.
5
37 (319-
904)
N.D.
F
F
425 (281-
44)
19
N.D.
6
F
2290 (1460
2
2
0
1
22 ± 4
21 ± 2
84 (45-157) 138 (39-468)
–
3590)
F
Cl
2
72 (181-
10)
363 (159-
813)
1190 (675 –
2100)
4
ACS Paragon Plus Environment

Page 17 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
53 (326-
37)
Cl
Cl
2
2
2
2
3
4
40 ± 7
43 ± 7
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
9
774 (506-
1180)
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
720 (941-
3130)
Cl
Cl
111 ± 14
Cl
6
3
330 (4350- 3311 (1950-
2
5
92 ± 12
>10,000
N.D.
9200)
5630)
Cl
O
430 (2150-
5490)
26
108 ± 17
N.D.
2
3
350 (1600-
3460)
COOMe
2
2
7
8
345 ± 48
501 ± 58
N.D.
N.D.
N.D.
N.D.
230 (2370-
SO Me
2
4
400)
020 (1410-
890)
3280 (2390- 3020 (1320- 6350 (4290-
OMe
2
2
3
9
0
58 + 5
N.D.
N.D.
2
OMe
61 + 2
31 ± 2
48 ± 8
4
520)
7080)
9410)
1850 (1150-
2950)
4
49 (278-
725)
31
110 (48-252)
OMe
Me
1
180 (849-
650)
3
2
N.D.
N.D.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 68
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
020 (2800- 3467 (1450-
NMe2
3
3
3
4
89 ± 14
107 ± 6
79 ± 13
>10,000
N.D.
9
010)
298 (184-
84)
8320)
N
N.D.
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
4
2
63 (133-
522)
1800 (1160-
2770)
35
204 (67-617)
N
a
Against 100 nM CP55,940 in stable human CB -CHO-RD-HGA16 cells. Values are the mean ±
1
SEM of at least three independent experiments in duplicate. b Against 100 nM CP55,940 in
HEK293 cells stably expressing CB . Values are expressed as mean (95% confidence interval)
1
c
from at least three independent experiments in duplicate. Against 100 nM CP55,940 in mouse
cerebellar membranes. Values are expressed as mean (95% confidence interval) from at least three
independent experiments in duplicate. d Against 100 nM CP55,940 in HEK293 cells stably
expressing the human CB . Replicates of at least 3 independent experiments performed in
1
duplicate were normalized to forskolin (%) and used to calculate IC best-fit parameters following
5
0
logistic non-linear regression with the top and bottom of curves globally shared. N.D. Not
determined.
With the encouraging results of the thiophene 11 which exhibited better or equipotent activities in
all three assays described above, we proceeded to prepare a focused series to investigate the
substituent effect on the phenyl ring of the thiophene analogs and screened these compounds in all
three in vitro assays. As shown in table 2, the presence of one or two fluoro substituent(s) or one
chloro group (18 - 23) resulted in a comparable potency to 2 in the calcium assay. Particularly, 20
3
5
(
IC = 84 nM) was more potent than 2 (IC = 455 nM) in the hCB [ S]GTPγS binding assay
5
0
50
1
ACS Paragon Plus Environment

Page 19 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
35
with comparable potencies in the hCB calcium and mCB [ S]GTPγS binding assays. However,
1
1
inclusion of two chloro groups, as in 3,4-dichloro and 3,5-dichloro analogues (24 and 25),
weakened the activities. Addition of electron-withdrawing groups such as acetyl (26),
methoxycarbonyl (27) or methylsulfonyl (28) at the 3-position also dampened the activities.
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
Among three positional isomeric methoxy analogues, 4-methoxy analogue (31) was the most
potent CB allosteric modulator with comparable potencies to 2 in all three assays. The other two
1
3
5
methoxy analogues (29 and 30) exhibited slightly lower potencies in the [ S]GTPγS binding
assays. The 3-methyl analogue (32) displayed good activity in the calcium mobilization assay (IC50
3
5
=
48 nM) but had a weaker potency in the hCB [ S]GTPγS binding assay (IC = 1180 nM). The
1 50
3
-N,N-dimethylamino analogue (33) demonstrated diminished activities in all three assays.
Although two pyridinyl analogues (34 and 35) had slightly weaker potencies in the calcium
35
mobilization assay, they were equipotent to 2 in two [ S]GTPγS binding assays. Overall, these
results indicate that small electron-withdrawing substituents such as fluoro or chloro are tolerated
on the phenyl ring. Electron-donating groups such as methoxy or methyl are preferably located at
the 4-position of the phenyl ring. The inclusion of a heteroatom such as in the pyridine ring is also
tolerated.
3
5
In general, the results from the two [ S]GTPγS binding assays were relatively consistent with the
calcium assay. Compounds with equipotent activities to 2 in the calcium assay also displayed
3
5
comparable activities in the [ S]GTPγS binding assay (18, 19, 21 - 23, and 31). Compounds with
35
weaker activities in the calcium assay also showed lower potencies in the [ S]GTPγS binding
3
5
assay (24 - 29, and 31 - 33). Several compounds demonstrated better activities in the [ S]GTPγS
binding assay than in the calcium assay. For example, the two pyridinyl analogues (34 and 35)
35
showed comparable potencies to 2 in the [ S]GTPγS binding assay while they exhibited weaker
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 68
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
activities in the calcium assay. Fascinatingly, the 2,4-difluoro analogue demonstrated significantly
3
5
better potencies in the [ S]GTPγS binding assay compared to 2 (20, IC = 84 nM vs 2 IC = 455
5
0
50
nM).
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
Interestingly, the allosteric modulatory activities of some compounds appeared to be more potent
35
at the higher CP55,940 concentration (1 µM) in the [ S]GTPγS binding assays using mouse
cerebellar membranes (Table 3) following two-way ANOVA [main effect of CP55,940
concentration: F(1,58) = 18.3, p<0.001; main effect of compound: F(8,58) = 75.6, p<0.001; no
significant interaction, F(8,58) = 1.15, p=0.345]. A significant shift in potency was observed for
2
0 with a 3-fold reduction in its calculated IC value following Bonferroni post hoc analysis
50
(
p<0.05). 35 also exhibited a significant change in IC values, from 287 nM to 129 nM. This shift
50
in inhibitory potencies reflects positive cooperativity characteristics of these PAM-antagonists
consistent with our previous study.34
3
5
Table 3. CB allosteric modulating activities in the [ S]GTPγS binding assays using mouse
1
cerebellum membrane at 1 µM and 100 nM CP55,940.
pIC ±
S.E.M.
6.91 ± 0.055 6.73 – 7.08
6.54 ± 0.081 6.23 – 6.82
6.84 ± 0.041 6.72 – 6.95
6.76 ± 0.13
7.32 ± 0.1 * 6.99 – 7.65
6.86 ± 0.17
6.69 ± 0.11
6.44 ± 0.11
pIC 95%
IC50
(nM)
123
288
145
174
48
138
204
363
IC (nM)
IC Fold
Change
5
0
50
50
50
Compound [CP55,940]
C.I.
95% C.I.
84 – 187
152 – 589
113 – 191
78 – 399
23 – 103
39 – 468
96 – 447
159 – 813
1 µM
100 nM
2
2.3
1
1
1
100 nM
1 µM
µM
00 nM
µM
18
1.2
2.9
1.8
6.4 – 7.11
2
2
0
1
6.33 – 7.41
6.35 – 7.02
6.09 – 6.8
1
00 nM
1
5
1
5630
080 –
380
950 –
1 µM
5.62 ± 0.11
5.27 – 5.97
2399
3311
2
5
1.4
1
00 nM
5.48 ± 0.074 5.25 – 5.71
ACS Paragon Plus Environment

Page 21 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
260 –
1 µM
5.79 ± 0.035
5.52 ± 0.11
5.67 – 5.9
1622
2
140
3
3
3
0
1
3
1.9
0.9
1.3
320 –
1
00 nM
5.15 – 5.88
3020
7080
76 – 195
48 – 252
1
4
1 µM
6.92 ± 0.074 6.71 – 7.12
6.96 ± 0.13 6.6 – 7.32
120
110
1
00 nM
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
630 –
080
450 –
1 µM
5.59 ± 0.064 5.39 – 5.79
5.46 ± 0.12 5.08 – 5.84
6.92 ± 0.063 6.72 – 7.12
6.69 ± 0.15 6.21 – 7.18
2570
3467
120
1
8320
1
00 nM
µM
100 nM
7
6
6 – 191
1
35
1.7
7 – 617
204
Mean IC values ± SEM of N=4-5 experiments performed in duplicate. Data were normalized (0-
50
1
00%) and fit to logistic non-linear regression with top constrained to 100 and bottom constrained
to 0. * p < 0.05 compared to respective 100 nM concentration of CP55,940 following two-way
ANOVA and Bonferroni post hoc.
All compounds were screened in the calcium mobilization assay for agonist activity at the CB ; no
1
significant agonist effects (<30% of CP55,940 Emax) were observed for any of the compounds. All
these compounds were also screened for agonist and antagonist activity at the CB to determine
2
3
0, 32
receptor subtype selectivity using our previously established CB calcium mobilization assay.
2
None of the compounds had significant CB agonist activity (<10% of CP55,940 Emax) or CB2
2
antagonist activity (<50% inhibition of CP55,940 EC concentration at 10 μM or IC values >10
8
0
50
μM).
Real-time kinetic BRET CAMYEL cAMP Studies
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 68
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
We then assessed several representative compounds in the real-time kinetic BRET CAMYEL
4
2
cAMP assay. In HEK-hCB cells, forskolin (10 µM) induced significant cAMP production that
1
plateaued after 5 min which was inhibited by the agonist CP55,940 (10 nM), consistent with
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
coupling to G proteins. A one-phase exponential association model was preferred for the majority
i/o
of the kinetic curves following an extra-sum-of-squares F-test, except for some of the curves for
the highest concentrations of modulator for 11 (3.16 and 10; p < 0.01), 20 (31.6 μM; p < 0.05),
and 31 (31.6 μM; p < 0.01), which could reflect inverse agonism as the accumulation of cAMP
appears greater than forskolin before the curve begins to plateau. Potencies did not appear to
change over the course of the assay when examined in 5 min bins (data not shown) in contrast to
4
2
1 which exhibited increases in potency over the course of the assay. This can, at least in part, be
explained by the fact that the allosteric modulators were pre-equilibrated in our assay whereas 1
was co-applied in the previous study; while this delay has been observed with other indole-2-
4
2
43
carboxamides, previously 2 did not exhibit a delay when it was co-applied, suggesting there
may be kinetic differences between these scaffolds.
Compounds 11, 20, 21, 31, and 35 were equipotent to 2 while 25, 30 and 33 were either inactive
or exerted weak activities. As seen in Fig 4, these compounds exhibited immediate dose-dependent
CB antagonism effects which reached equilibrium at around 5-6 min. In the cAMP BRET assay,
1
3
5
compounds appeared to be less potent than in the calcium and [ S]GTPγS binding assay (Table
2).
ACS Paragon Plus Environment

Page 23 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 2
Basal
1
.0
0.9
.8
Forskolin (10 M)
FSK + CP55,940 (10 nM)
FSK + CP + compound (10 M)
FSK + CP + compound (3.16 M)
0
FSK + CP + compound (1 M)
FSK + CP + compound (316 nM)
FSK + CP + compound (100 nM)
FSK + CP + compound (31.6 nM)
FSK + CP + compound (10 nM)
FSK + CP + compound (3.16 nM)
0.7
0.6
0.5
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
0
3
6
9
12
15
18
21
Time (min)
Compound 11
Compound 30
Compound 31
1
.0
.9
.8
1.0
1.0
0.9
0.8
0.7
0.6
0.5
0
.9
.8
.7
.6
0
0
0
0
0
0.7
0
.6
.5
0.5
0
0
3
6
9
12
15
18
21
0
3
6
9
12
15
18
21
0
3
6
9
12
15
18
21
Time (min)
Time (min)
Time (min)
Figure 4. Temporal cAMP BRET signaling raw data of select diarylurea-based CB allosteric
1
modulators.
Potency differences between various pharmacological assays
35
35
Overall, potencies were consistent across the human [ S]GTPγS, mouse [ S]GTPγS, and cAMP
assays as supported by significant positive correlations between IC values for human and mouse
5
0
[r(10)=0.959 (0.830 – 0.990), p < 0.0001], human and cAMP [r(7)=0.958 (0.736 – 0.994), p <
.001], and mouse and cAMP [r(7)=0.957 (0.729 – 0.994), p < 0.0001]. However, a significant
0
3
5
correlation was not observed for IC50 values between calcium and human [ S]GTPγS
r(20)=0.366 (-0.0917 – 0.696), p=0.113] or calcium and cAMP [r(7)=0.433 (-0.474 – 0.894),
[
p=0.332]. There are several distinctions between assays that may have contributed to a lack of
congruence in potency for these compounds.
While the calcium mobilization assay was performed in CHO cells with CB coupling to Gα , the
1
16
3
5
44
[ S]GTPγS binding assay predominantly reflects coupling to the Gα family of G proteins. G-
i/o
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 68
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
proteins increase agonist affinity for CB and agonists can exhibit biased signaling for individual
1
G-protein subtypes.45, 46 For example, when overexpressed, Gα shifts the conformation of the
1
6
kappa opioid receptor and increases the affinity of salvinorin A by 18-fold while not affecting the
affinity of U69,593 or dynorphin A. 47 Thus, higher potencies observed in the calcium assay could
reflect the contribution of this distinct G-protein subtype having a greater impact on the affinity of
these PAM-antagonists.
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
5
Differences between the [ S]GTPγS and cAMP assays could be due to receptor reserve. Sufficient
receptor reserve exists for cAMP signaling in our HEK293 cells such that agonists of high and low
3
5
efficacy in [ S]GTPγS binding typically exhibit equi-efficacy in cAMP signaling, as this pathway
is easily saturated due the artificially high level of CB . Only a fraction of receptors must be
1
occupied to saturate the pathway and reach Emax in cAMP, which requires greater receptor
coverage by the modulator to antagonize this response, i.e. a higher IC .
5
0
Other factors that may also have impacted relative potencies include differences in equilibration
time and receptor:G-protein ratios. Because the calcium assay measures the response for 90 s
4
8
immediately after agonist CP55,940 is added, it is conducted at hemi-equilibrium. Therefore,
differences in cooperativity between these allosteric modulators will impact the binding kinetics
of CP55,940,49, 50 which could result in apparent differences in relative potencies compared to
3
5
assays in which equilibrium was reached or approached (e.g. cAMP is 20 min, [ S]GTPγS is 60
min plus 30 min pre-equilibration with agonist and modulator). In addition, cerebellar membranes
contain high levels of CB with receptor/G-protein stoichiometry that is more physiologically
1
relevant, whereas the HEK293 cells have overexpressed human CB which results in a higher
1
receptor to G-protein ratio, which means a large proportion of CB will be uncoupled and exhibit
1
5
1
low affinity for agonist. Interestingly, IC values in the human CB HEK293 membranes were
5
0
1
ACS Paragon Plus Environment

Page 25 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
generally higher than in the mouse CB which would be consistent with this, though differences
1
in the human versus mouse form of the receptor cannot be discounted. While the CB Gα CHO
1
16
cells have overexpressed CB , they also overexpress Gα so the ratio may be favoring
1
16
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
coupled/high affinity receptors resulting in a concomitant increase in affinity of the modulator due
to cooperativity.
Intrinsic activities of CB allosteric modulators
1
CB has been demonstrated to possess constitutive activities which are critical to maintain normal
1
physiological functions. SR141716 acts as an CB inverse agonist, reducing the CB signaling on
1
1
its own. It has been suggested that inhibition of this basal activity results in the adverse effects of
5
2
SR141716. Hence, we were interested in investigating the intrinsic activities of these allosteric
modulators in the absence of the CB agonist CP55,940. As shown in Fig 5, SR141716 imparted
1
significant inverse agonism with an IC of 2.8 nM. 2 reached the same level of inverse agonism
5
0
produced by SR141716, but only at the highest concentration of 10 µM (IC = 1.47 µM). At 10
5
0
µM, the allosteric modulators in this new series only exhibited a modest degree of inverse agonism.
Indeed, the thiophene analogue 11 produced very little inverse agonism up to 10 µM. These results
imply that these CB allosteric modulators may have less propensity to impart adverse effects than
1
SR141716/rimonabant.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 68
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
10
0
SR141716
2
9
-
-
10
20
11
14
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
-30
40
-
10
-8
-6
-4
-
log [compound] M
Figure 5. Intrinsic activities of the CB allosteric modulators and SR141716 in the absence of the
1
CB agonist CP55,940.
1
ADME Studies on compounds 2 and 11
As pharmacokinetics and biodistribution plays an important role determining whether compounds
reach the site of action at sufficient levels to impart in vivo activities, we examined preliminary
ADME properties for the most potent compound in vitro, 11, along with 2 (Table 4). Compound
1
1 (T1/2 = 65 min) showed significantly improved metabolic stability than 2 (T1/2 = 13 min). As
expected with the diarylurea scaffold, 2 displayed poor kinetic solubilities (<1 µM) whereas 11
has slightly better solubility (1.5 µM). The slightly enhanced solubility could be attributed to the
less tight packing of the five membered ring thiophene compared to the phenyl ring. Finally, we
employed the bidirectional MDCK-MDR1 assay to evaluate the blood-brain barrier permeability
and the probability to be a P-glycoprotein substrate. Both 2 and 11 exhibited low permeabilities in
-6
this assay, with Papp values of 2.6 and 1.6 x 10 cm/s (A to B), significantly lower than the
-6
53
commonly desired Papp value of 15 x 10 cm/s for CNS-penetrant compounds. However, neither
is a P-glycoprotein substrate (efflux ratio < 2.5).53
ACS Paragon Plus Environment

Page 27 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. ADME properties of 2 and 11.
Compound
2
11
Half-life (min) a
Clearance (µL/min/mg) a
Solubility (µM) b
13.4 ± 4.1
113.7 ± 34.4
< 0.5
65.0 ± 19.1
22.2 ± 6.5
1.5 ± 0.1
1.6 ± 0.1
1.1 ± 0.4
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
-
6
a
a
MDCK-MDR1 Papp A to B (x10 cm/s)
2.6 ± 0
2.2 ± 0.1
0.8
-6
MDCK-MDR1 Papp B to A (x10 cm/s)
MDCK-MDR1 efflux ratio BA/AB
0.7
a
b
Values are expressed as mean ± SD from two independent experiments. Values are expressed
as mean ± SD from three independent experiments.
Compound 11 on reinstatement of cocaine-seeking behaviors
Blockade of the CB in vivo by antagonists/inverse agonists has been demonstrated to reduce food
1
intake and self-administration of several drugs of abuse.54-56 We previously demonstrated that rats
pretreated with 1, 2, and 4 were less likely to seek drugs of abuse, such as cocaine or
methamphetamine, after a period of extinction.32, 38 Herein, we examined one of the most potent
compounds, 11, in this new series in the reinstatement of cocaine seeking behavior in rats.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 68
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
Drug-induced reinstatement of cocaine-seeking
Active responses
Inactive responses
8
6
4
0
0
0
20
15
10
5
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
20
0
0
Vehicle
11
Vehicle
11
Figure 6. Behavioral effects of 11 (10 mg/kg, i.p.) in the drug-induced reinstatement of
cocaine-seeking models in rats (n = 10 male Sprague-Dawley rats/group, p > 0.05 by Student’s t-
test).
As 11 had equipotent or better in vitro potencies compared to 4, we tested the effect of the same
3
2
dose of 10 mg/kg (i.p.) pretreatment on drug-induced cocaine-seeking behavior. As shown in
Fig. 6, this dose of 11 failed to significantly reduce drug-induced cocaine-seeking behavior (p >
0
.05). Both active and inactive responses were not significantly altered by 10 mg/kg of 11. This
lack of efficacy is likely due to the poor brain penetration of 11, as demonstrated in the MDCK-
MDR1 studies. Future studies administering the drug directly into the brain may better assess the
pharmacological effects of 11 on cocaine-seeking behavior.
Conclusions
In this study, we investigated a series of diarylurea-based CB allosteric modulators centering
1
around modifications on the central phenyl ring of 2. SAR revealed that replacement with non-
aromatic rings resulted in significant reduction of CB modulating activity. Interestingly, the
1
ACS Paragon Plus Environment

Page 29 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
presence of a five-membered ring such as thiophene led to the identification of 11 with better or
3
5
comparable potencies than 2 in calcium, cAMP, and [ S]GTPγS binding assays. This series of
3
5
compounds do not exhibit inverse agonism in [ S]GTPγS, indicating that they may not have
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
liability of psychiatric side effects associated with the CB antagonist/inverse agonist SR141716.
1
Although 11 had good metabolic stability, its solubility and blood-brain barrier permeability need
further improvement. In contrast to 4, 11 showed little effect in attenuating cocaine seeking
behavior in rats and seems to be comparable to 2 in this in vivo assay. The unsuccessful
extrapolation of in vitro pharmacological potencies to in vivo efficacy in the cocaine relapse model
could be due to the limited brain penetration of this compound. If follow-up pharmacokinetic
studies demonstrate that 11 is indeed restricted to the periphery, it could stand as a promising
candidate for peripherally CB -mediated conditions. More importantly, our SAR studies reveal
1
several modifiable regions to optimize pharmacokinetic properties to achieve in vivo efficacy.
Experimental Section
Chemistry. All solvents and chemicals were reagent grade. Unless otherwise mentioned, all
reagents and solvents were purchased from commercial vendors and used as received. Flash
column chromatography was carried out on a Teledyne ISCO CombiFlash Rf system using
prepacked columns. Solvents used include hexanes, ethyl acetate (EtOAc), dichloromethane, and
methanol. Purity and characterization of compounds were established by a combination of high
pressure liquid chromatography (HPLC), thin layer chromatography (TLC), mass spectrometry
(MS), and nuclear magnetic resonance (NMR) analyses. Melting point was recorded by the Mel-
1
13
Temp II instrument (laboratory Devices Inc., US). H and C NMR spectra were recorded on a
Bruker Avance DPX-300 (300 MHz) spectrometer and were determined in CDCl , DMSO-d , or
3
6
ACS Paragon Plus Environment