Synthesis and pharmacological evaluation of new biphenylic derivatives as CB2 receptor ligands

Article abstract of DOI:10.1016/j.ejmech.2016.03.072

Targeting type-2 cannabinoid receptor (CB₂) is considered a feasible strategy to develop new drugs for the treatment of diseases like neuropathic pain, chronic inflammation, neurodegenerative disorders and cancer. Such drugs are devoid of the u

Full text of DOI:10.1016/j.ejmech.2016.03.072

Accepted Manuscript
Synthesis and pharmacological evaluation of new biphenylic derivatives as CB
receptor ligands
2
Simone Bertini, Andrea Chicca, Chiara Arena, Stefano Chicca, Giuseppe
Saccomanni, Jürg Gertsch, Clementina Manera, Marco Macchia
PII:
S0223-5234(16)30254-9
10.1016/j.ejmech.2016.03.072
EJMECH 8497
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 December 2015
Revised Date: 24 March 2016
Accepted Date: 25 March 2016
Please cite this article as: S. Bertini, A. Chicca, C. Arena, S. Chicca, G. Saccomanni, J. Gertsch, C.
Manera, M. Macchia, Synthesis and pharmacological evaluation of new biphenylic derivatives as CB
2
receptor ligands, European Journal of Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.03.072.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
Synthesis and pharmacological evaluation of new biphenylic derivatives as
CB receptor ligands
2
a
b
a,b
b
Simone Bertini, * Andrea Chicca, * Chiara Arena, Stefano Chicca,
a
b
a
a
Giuseppe Saccomanni, Jürg Gertsch, Clementina Manera, Marco Macchia
Dipartimento di Farmacia, Università di Pisa, Via Bonanno 6, 56126 Pisa, Italy
Institute of Biochemistry and Molecular Medicine, National Center of Competence in
Research (NCCR) TransCure, University of Bern, Bühlstrasse, 28 CH-3012 Bern,
Switzerland
*
SB and AC equally contributed to this work.
To whom correspondence should be addressed. Simone Bertini: phone, +39(0)502219579;
fax, +39(0)502219605; e-mail: simone.bertini@unipi.it
RECEIVED DATE (to be automatically inserted after your manuscript is accepted if
required according to the journal that you are submitting your paper to)
a
Dipartimento di Farmacia, Università di Pisa, Via Bonanno 6, 56126 Pisa, Italy
b
Institute of Biochemistry and Molecular Medicine, National Center of Competence in
Research (NCCR) TransCure, University of Bern, Bühlstrasse, 28 CH-3012 Bern,
Switzerland

ACCEPTED MANUSCRIPT
Highlights
•
•
•
We synthesized 18 biphenylic carboxamides as new CB -selective ligands.
2
The functional activity is influenced by the substituent at position 4′ and 5.
The methoxyl group at position 4′ is responsible for neutral antagonist behaviour.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
Abstract
Targeting type-2 cannabinoid receptor (CB ) is considered a feasible strategy to develop new
2
drugs for the treatment of diseases like neuropathic pain, chronic inflammation,
neurodegenerative disorders and cancer. Such drugs are devoid of the undesired central side
effects that are typically mediated by the CB receptor. In this work we synthesized 18
1
biphenylic carboxamides as new CB -selective ligands and evaluated their pharmacological
2
profiles.The functional activity of these compounds is strongly influenced by the nature of the
substituent at position 4′ and 5 of the biphenyl scaffold. Position 5 seems to be responsible
for the agonist or inverse agonist behaviour independently of the substituent in position 4′,
with the exception of the methoxyl group which transforms both full agonists and inverse
agonists into neutral antagonists. This study provides a novel complete toolbox of CB2
functional modulators that derive from the same chemical scaffold. Such probes may be
useful to investigate the biological role of CB receptors in cellular assays.
2
Keywords: CB , cannabinoid receptor, biphenyl-carboxamides, CB agonist, CB antagonist,
2
2
2
CB inverse agonist, methylhonokiol, endocannabinoid system
2

ACCEPTED MANUSCRIPT
1
. Introduction
The endocannabinoid system (ECS) comprises the two main cannabinoid receptors (CB and
1
CB ), their endogenous ligands (endocannabinoids) and the enzymes responsible for the
2
biosynthesis and metabolism of endocannabinoids [1, 2].
Over the past two decades, great efforts have been made in order to fully understand the
biological role and regulatory functions of CB receptors in pathophysiological conditions.
CB receptors are mainly expressed in the central nervous system, but also in peripheral
1
districts including spleen, heart, reproductive organs, lungs and adipose tissue [1]. CB2
receptors are widely localized in cells and tissues of the immune system, cardiovascular
system, bone, liver, kidney and the gut [3-5]. In healthy brain, CB receptors are almost
2
absent, although their presence in some restricted neuronal populations has been reported [6].
However, CB receptors become significantly expressed in activated microglial cells and
2
astrocytes upon specific conditions such as neuroinflammation [3-5]. In the perpiphery, CB2
receptors are expressed in different immune cells where they show dynamic trafficking
between cytoplasm and cell membrane where they may even form heterodimers with CB1
receptors [7].
CB and CB receptors have been proposed as potential therapeutic targets for the treatment
1
2
of several diseases including actute and chronic inflammation, neurodegenerative and eating
disorders, neuropathic pain, cancer and osteoporosis [4, 8, 9]. Non-selective CB /CB2
1
®
receptor agonists are the active principles of some approved medicines (i. e. Marinol ,
®
®
Cesamet , Sativex ). However, non-selective agonists produce adverse effects (i. e.
psyichotropic) almost exclusively due to the activation of central CB receptors. The
1
occurrence of these side effects represents the major limitation in the therapeutic use of these
compounds, especially for chronic treatment. One of the proposed strategies to avoid the

ACCEPTED MANUSCRIPT
unwanted consequences of the central CB receptor activation is to selective target CB2
1
receptors.
In some therapeutic field such as pain, inflammation and osteoporosis, the potential
usefulness of both CB -agonists and CB -inverse agonists has been reported [10-16], while
2
2
very few compounds acting as neutral CB -antagonists have been described [17, 18].
2
These considerations strongly support the importance of the synthesis and pharmacological
characterization of new CB -selective ligands bearing such functional profiles (agonists or
2
inverse agonists or neutral antagonists) in order to delineate a precise structure-activity
relationship for this kind of molecules and to deepen the role of the CB receptors in
2
pathophysiological conditions.
We recently described the synthesis, binding properties, functional activity and molecular
modelling of a series of biphenylic carboxamides (general structure A, Figure 1) as selective
CB receptor ligands [19]. Some of these compounds showed CB -affinity levels in the nM
2
2
range. The most potent and selective derivative of that series showed an interesting
pharmacological profile as a selective neutral antagonist for CB receptors, thus being a silent
2
ligand when tested alone and dose-dependently reverting the response of a CB agonist (HU-
2
2
10) and a CB inverse agonist (SR144528) [19].
2
5
'
R3 4'
6
1
'
'
O
2
3
'
1
R2
3
2
'
NH
OMe
4
6
5
R1
A)
(
Figure 1. General structure of biphenylic carboxamides [19].
In an effort to identify additional and more potent or functionally distinct biphenyl
derivatives as selective CB receptor ligands, and to develop structure-activity relationships
2
(SAR) for these type of compounds, we engaged on the synthesis and characterization of the

ACCEPTED MANUSCRIPT
pharmacological properties of a number of biphenyl derivatives (compounds 1a-r, Table I).
To that aim, other combinations of substituents from the ones already mentioned in the
previous report were explored and new substituents were also introduced.
In particular, position 5 (R ) was functionalized with n-butyl, benzyl or p-fluorobenzyl; in
1
position 4′ (R ) there was a hydrogen, a fluorine or a methoxy group; finally the substituent
3
on carboxamide moiety (R ) was a cycloheptyl or a 4-methylcyclohexyl (cis or trans isomer).
2
The new compounds were tested on cannabinoid receptors in order to evaluate the binding
affinity and selectivity. The displacement assays were performed using 0.5 nM of
3
[
H]CP55,940 and 15 µg of clean membrane preparations obtained from CHO-K1 cells stably
transfected with hCB and hCB receptors, respectively. The most affine ligands were also
1
2
3
5
assessed in functional tests using the [ S]GTPγS assay as previously described [20].
2
. Results and discussion
2
.1 Chemistry
The synthesis of compounds 1a-f is shown in Scheme 1. The atom numbering of carbon
skeleton (and of substituents) of compounds 1a and 1b (as examples for all the final
products) is also included in Scheme 1. Compound 1a-f were synthesized starting from
methyl 5-bromo-3-butyl-2-methoxybenzoate (1), which was prepared as previously reported
by our research group [19] (Scheme 1). Compound 1 was submitted to a Suzuki reaction with
the proper arylboronic acid (phenyl, or 4-fluorophenyl boronic acid), aqueous sodium
carbonate, palladium acetate and triphenylphosphine in methanol/toluene [21], affording the
intermediates 2 and 3 which were then hydrolysed in presence of potassium hydroxide in
methanol [22] leading to the corresponding acid derivatives 4 and 5. Intermediates 4 and 5
were treated with thionyl chloride and subsequently with cycloheptylamine in anhydrous

ACCEPTED MANUSCRIPT
dichloromethane [23], obtaining the desired final products 1a and 1d. Treatment of 4 and 5
with thionyl chloride and subsequently with 4-methyl-cyclohexylamine (cis/trans mixture) in
anhydrous dichloromethane [23] followed by chromatographic separation of the cis- and
trans-isomers, allowed to obtain the desired final products 1b-c and 1e-f. The two isomers
were identified on the basis of the chemical shift of the proton bound to the C1′′′ position (see
the atom numbering of compound 1b in Scheme 1). In the cis isomer this proton resonates at
higher ppm (δ from 4.20 to 4.39 ppm) compared to the same proton in the trans isomer (δ
from 3.87 to 4.11 ppm), as previously reported [19].
F
O
O
O
Br
OMe
OMe
OMe
OMe
a
[86%]
OMe
OMe
a
[85%]
F
B(OH)
2
B(OH)
2
2
1
3
b
[98%]
b
[69%]
F
O
O
5
'
6'''
5'''
3'''
4
3
'
'
6'
7'''
'''
F
O
4'''
O
OH
2
1
OH
1
3
'
1
3
2
'
NH
OMe
2'''
c
[63%]
NH
OMe
OMe
OMe
c
[45%]
6
4
5
''
2''
1''
H
2
N
4
H
2
N
5
4
''
1a
1d
c
H
2
N
c
H2
N
cis/trans
cis/trans
mixture
mixture
[3%]
[5%]
[13%]
[24%]
5
'''
7'''
5
'
4
'
6'
6
'''
4'''
3'''
O
F
F
O
O
O
2
1'''
3
'
1
3
NH
OMe
1
'
2
'
NH
OMe
2'''
NH
OMe
NH
OMe
4
6
5
1''
3
''
2''
4
''
1c
1
b
1e
1f
Scheme 1. Reagents and conditions: a) Suitable arylboronic acid, Pd(PPh ) , aq. Na CO ,
3
4
2
3
toluene/MeOH, microwave (150 °C, 5 bar, 200 W), 10 min.; b) KOH, MeOH, reflux, 19 h; c)
) SOCl , reflux, 30 min.; 2) Suitable cycloalkylamine, CH Cl , RT, overnight.
1
2
2
2
Compounds 12a-d, direct precursors of final products 1g-r, were synthesized as shown in
Scheme 2, starting from methyl 3-bromo-2-methoxybenzoate (6), which was prepared as
previously reported by our research group [19]. Compound 6 was converted in the
corresponding arylboronic acid 7 through a two steps procedure. 6 was first submitted to a

ACCEPTED MANUSCRIPT
Miyaura
borylation
in
presence
of
potassium
acetate
and
bis(diphenylphosphinoferrocene)palladium dichloride ((dppf)PdCl ) in anhydrous 1,4-
2
dioxane [24]. After that, the obtained pinacol ester was hydrolyzed in presence of ammonium
acetate and sodium periodate in a mixed solution of acetone and water (1:1) to afford the
desired arylboronic acid 7, which was subjected to a cross-coupling reaction with benzyl
bromide
or
4-fluorobenzyl
bromide
in
the
presence
of
tetrakis(triphenylphosphine)palladium(0) as catalyst and sodium carbonate as base, in a
mixture of anhydrous 1,2-dimethoxyethane and water (2:1) [25] to give derivatives 8a-b.
Subsequently, these compounds were brominated using bromine in chloroform [26],
affording the desired compounds 9a-b and a significant amount of the corresponding
demethylated products 10a-b, which were converted again in the derivatives 9a-b upon
treatment with dimethyl sulphate and sodium hydroxide in the presence of
tetrabutylammonium bromide in water/dichloromethane [27]. The subsequent Suzuki
reaction with the suitable arylboronic acid (phenyl, 4-fluorophenyl and 4-methoxyphenyl
boronic acid), aqueous sodium carbonate, palladium acetate and triphenylphosphine in
methanol/toluene [21], afforded the intermediates 11a-d which were hydrolysed in the
presence of potassium hydroxide in methanol [22] leading to the corresponding acid
derivatives 12a-d.

ACCEPTED MANUSCRIPT
O
OMe
OMe
Br
6
a
[73%]
O
O
O
OMe
OMe
b
[63%]
b
[66%]
OMe
Br
OMe
B(OH)2
Br
OMe
OMe
F
7
8a
8b
F
c
c
[
42%]
[24%]
Br
[
32%]
[35%]
O
O
O
O
OMe
Br
Br
Br
OMe
OH
OMe
OMe
OMe
d
OH
d
OMe
[88%]
F
[42%]
9a
10a
9b
10b
F
(HO)2B
B(OH)2
[85%]
e
B(OH)2
F
e
e
58%]
B(OH)2
e
86%]
[
[59%]
[
OMe
MeO
F
O
O
O
O
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
11b
11c
11d
11a
F
F
F
f
[72%]
f
[89%]
f
[78%]
f
[88%]
F
MeO
O
O
O
O
OH
OMe
OH
OMe
OH
OMe
OH
OMe
12a
12b
12c
12d
F
F
F
Scheme 2. Reagents and conditions: a) 1) Bis(pinacolate)diboron, KOAc, Pd(dppf)Cl2,
anhydrous 1,4-dioxane, 110 °C, 2 h (or microwave: 130 °C, 5 bar, 200 W, 30 min.); 2)
NH OAc, NaIO , acetone/H O, RT.; b) Suitable aryl bromide, Na CO , Pd(PPh ) ,
4
4
2
2
3
3 4
DME/H O, 100 °C, 4 h (or microwave: 140 °C, 5 bar, 200 W, 15 min.); c) Br , CHCl , RT,
2
2
3
overnight; d) TBAB, (CH ) SO , aq. NaOH, CH Cl , RT, overnight; e) Suitable arylboronic
3
2
4
2
2
acid, Pd(PPh ) , aq. Na CO , toluene/MeOH, microwave (150 °C, 5 bar, 200 W), 10 min.; f)
3
4
2
3
KOH, MeOH, reflux, 19 h.

ACCEPTED MANUSCRIPT
Final products 1g-r were obtained from intermediates 12a-d as shown in Scheme 3. 12a-d
were treated with thionyl chloride and subsequently with cycloheptylamine in anhydrous
dichloromethane [23], affording the desired final products 1g, 1j, 1m and 1p. Treatment of
1
2a with thionyl chloride and subsequently with 4-methyl-cyclohexylamine (cis/trans
mixture) in anhydrous dichloromethane [23] allowed to obtain the desired final products 1h
trans-isomer) and 1i (cis-isomer) after chromatographic separation. Trans-isomer 1h was
(
also obtained by treatment of 12a with thionyl chloride and subsequently with trans-4-
methyl-cyclohexylamine in anhydrous dichloromethane [23]. Treatment of 12d with thionyl
chloride and subsequently with 4-methyl-cyclohexylamine (cis/trans mixture) in anhydrous
dichloromethane [23] allowed to obtain the desired final products 1q (trans-isomer) and 1r
(cis-isomer) after chromatographic separation. Trans-isomer 1q was also obtained by
treatment of 12d with thionyl chloride and subsequently with trans-4-methyl-
cyclohexylamine in anhydrous dichloromethane [23]. Finally, treatment of 12b and 12c with
thionyl chloride and subsequently with 4-methyl-cyclohexylamine (cis/trans mixture) in
anhydrous dichloromethane [23], followed by chromatographic separation of the cis- and
trans-isomers, allowed to obtain the desired final products 1k, 1l, 1n and 1o.

ACCEPTED MANUSCRIPT
O
[7%]
NH
OMe
O
O
OH
OMe
1i
NH
OMe
a [62%]
a
F
H N
2
H N
2
O
cis/trans
mixture
1g
12a
F
NH
OMe
F
[11%]
a [32%]
O
H N
2
1h
H N
NH
OMe
2
F
a
[32%]
[2%]
1q
O
O
a
OH
H N
NH
OMe
O
2
H N
2
OMe
a [62%]
NH
OMe
cis/trans
mixture
1p
[6%]
12d
1r
MeO
O
NH
OMe
[4%]
MeO
MeO
O
O
1l
OH
OMe
NH
OMe
F
a [23%]
a
H N
MeO
2
H N
O
2
12b
cis/trans
mixture
1j
NH
OMe
F
F
F
[4%]
1k
O
F
NH
OMe
[23%]
F
F
O
O
1n
OH
NH
OMe
F
a
H N
2
OMe
a [53%]
H N
F
2
O
cis/trans
mixture
12c
1m
NH
OMe
[32%]
F
F
1o
F
Scheme 3. Reagents and conditions: a) 1) SOCl , reflux, 30 min.; 2) Suitable
2
cycloalkylamine, CH Cl , RT, overnight.
2
2

ACCEPTED MANUSCRIPT
2
.2 Binding to CB and CB receptors
1 2
We synthesized 18 biphenyl carboxamides to investigate the impact of different substituents
on the binding and activation of CB receptors. First, we assessed the impact of the
2
carboxamide moiety keeping constant the substituents in R and R (Fig. 1, Table I).
1
3
a
Table I. Radioligand binding data of biphenylic derivatives.
R3
O
R2
NH
OMe
R1
K (nM)
i
b
Compound R1
R2
R3
CB1
CB2
352
S.I.
5.3
0.7
2.3
1a
1b
1c
1d
1e
1f
1
1
1
1
1
1
n-butyl
n-butyl
n-butyl
n-butyl
n-butyl
n-butyl
cycloheptyl
H
1859
3608
549
trans-4-methylcyclohexyl
cis-4-methylcyclohexyl
cycloheptyl
H
2390
1262
166
H
F
6111
36.8
≥5.8
1.8
trans-4-methylcylclohexyl
cis-4-methylcyclohexyl
F
≥10000 1763
F
2523
904.5
3086
826
1413
84.1
2617
1106
811
g
h
i
4-fluorobenzyl cycloheptyl
H
10.7
1.2
4-fluorobenzyl trans-4-methylcyclohexyl
4-fluorobenzyl cis-4-methylcyclohexyl
4-fluorobenzyl cycloheptyl
H
H
1.3
k
j
OMe
1173
1.4
4-fluorobenzyl trans-4-methylcyclohexyl OMe
4-fluorobenzyl cis-4-methylcyclohexyl OMe
≥10000 ≥10000 n.d.
380 196 2.0
l

ACCEPTED MANUSCRIPT
1
1
1
1
1
1
1
1
m
n
o
4-fluorobenzyl cycloheptyl
F
2831
1852
1070
1557
2502
n.d.
265
10.7
1.7
3.2
2.6
1.9
n.d.
4.5
6.1
n.d.
4-fluorobenzyl trans-4-methylcyclohexyl
4-fluorobenzyl cis-4-methylcyclohexyl
F
1086
337
F
p
q
r
benzyl
benzyl
benzyl
methyl
methyl
n-butyl
cycloheptyl
H
594
trans-4-methylcyclohexyl
cis-4-methylcyclohexyl
cycloheptyl
H
4809
2667
148
H
c
s
H
656
c
t
cycloheptyl
OMe
OMe
16542
n.d.
2742
249
c
1
u
cycloheptyl
a
Binding experiments were carried out using purified membranes generated in-house from
3
CHO-hCB and CHO-hCB stably transfected cell lines. [ H]CP55940 concentration was 0.5
1
2
nM and its K value for CB and CB receptor is 0.5 nM and 0.69 nM, respectively. Values
d
1
2
b
represent means and 95% confidence intervals. n.d.: not determined S.I.: selectivity index
c
for CB receptor calculated as K (CB )/K (CB ) ratio. See Ref. [19].
2
i
1
i
2

ACCEPTED MANUSCRIPT
R
1
R
1
R
1
OCH
3
OCH
3
OCH
3
H
N
H
N
H
N
O
O
R
3
O
R
3
R
3
A
R : n-butyl
B
R : n-butyl
C
R : benzyl
1
1
1
R : H
R : F
R : H
3
3
3
1
20
00
120
100
80
60
40
20
0
120
100
80
60
40
20
0
1
8
6
4
2
0
0
0
0
0
A
1 dC 22
1
c
p
omp.5
1
1
A
A
a
c
C40
C42c
C42a
A
1r
GR28b
1
AC23a
e
GR29a
1q
1b
A
-
10 -9 -8 -7 -6 -5 -4
-10 -9 -8 -7 -6 -5 -4
-10 -9 -8 -7 -6 -5 -4
Log [compound] (M)
Log [compound] (M)
Log [compound] (M)
D
E
F
R : p-Fluoro-benzyl
R : p-Fluoro-benzyl
R : p-Fluoro-benzyl
1
1
1
R : H
R : F
R : Methoxyl
3
3
3
1
1
20
00
1
20
1
1
20
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
8
6
4
0
0
0
1
1
j
AC17
1
AC35
C64b
C64a
m
1
A
g
C66
omp.7
omp.6
20
0
l
AC33b
1
c
i
1
A
o
1
c
h
1
A
n
1 kA C33a
-
10 -9 -8 -7 -6 -5 -4
-10 -9 -8 -7 -6 -5 -4
-10 -9 -8 -7 -6 -5 -4
Log [compound] (M)
Log [compound] (M)
Log [compound] (M)
Figure 2. Binding curves to CB receptors. The full concentration-dependent binding
2
curves to CB receptor are reported for different series of analogs. In each panel, the impact
2
of the type of substituent on the carboxamide moiety (R ) is reported for the cycloheptyl
2
(red), cis-4-methylcyclohexyl (green) and trans-4-methylcyclohexyl (blue) group. The role of
the substituent in R and R was evaluated by using groups with different size and polarity, as
1
3
reported in the graphs.

ACCEPTED MANUSCRIPT
The results show that cycloheptyl carboxamides have higher affinity to CB receptors
2
compared to 4-methylcyclohexyl-bearing molecules without major differences between the
cis and trans isomer (Fig. 2). This is in agreement with the previous molecular modelling
study performed on similar biphenyl compounds [19]. Only compound 1l bearing a cis-4-
methylcyclohexyl carboxamide showed a similar K value to the cycloheptyl derivative (1j)
i
and significantly higher potency compared to the trans isomer (1k) (Table I and Fig. 2F).
Compound 1j-l differ from the other set of compounds (Fig. 2A-E) for the substituent in R3,
which is neither a hydrogen (compounds 1a-c) nor a fluorine (compounds 1d-f) but a
methoxyl group (Fig. 2F). This might suggest that a bigger substituent in position 4′ could
abolish the privileged interaction with the CB receptor binding pocket provided by the
2
cycloheptyl carboxamide compared to the 4-methylcyclohexyl moiety. Then, we investigated
the impact of different substituents in R and R on CB and CB receptors binding. In
1
3
1
2
agreement with the previous docking study, the results here reported confirmed that the
presence of a n-butyl chain in position 5 accounts for a higher selectivity to CB over CB1
2
receptors. Indeed, compound 1d showed the highest selectivity for CB receptors (36-folds).
2
Furthermore, our data also indicate that by replacing the linear alkyl chain with an aromatic
group the selectivity towards CB receptors is not only retained (e.g. 1d vs 1m) but rather
2
increased (e.g. by a factor 2 for 1g vs 1a).
3
.3 Assessment of functional activity at CB receptors
2
We assessed the functional activity of the biphenyl derivatives at CB receptors using the
2
35
[
S]GTPγS binding assay (Table II). Based on the binding data, we decided to investigate
only the compounds bearing a cycloheptyl carboxamide moiety in combination with different
substituents in position 5 and 4′, with the exception of 1l, which showed a higher affinity
compared to corresponding cycloheptyl derivative (1j), as described above. Initially, we

ACCEPTED MANUSCRIPT
investigated the impact of three different groups in R (methyl, n-butyl, 4-fluorobenzyl)
1
keeping the position 4′ unsubstituted. The results show that the presence of a methyl group
leads to full agonism at CB receptors, while bigger substituents dramatically shift the
2
functional activity towards inverse agonism (1a and 1g, Fig. 3A).
Table II. Effects of different substituents in R and R to the biphenylic scaffold on the
1
3
functional activation of CB receptors.
2
R3
O
NH
OMe
R1
Compound
R1
R3
H
CB effects
2
1
a
n-butyl
n-butyl
n-butyl
Inverse agonist
Inverse agonist
1
d
u
F
1
Methoxyl
Antagonist
Inverse agonist
Inverse agonist
Antagonist
1
g
4-Fluorobenzyl
4-Fluorobenzyl
4-Fluorobenzyl
Benzyl
H
1
m
F
1
j
Methoxyl
1
p
H
Inverse Agonist
Agonist
*
1
s
Methyl
H
F
*
1
v
Methyl
Agonist
*
1
t
Methyl
Methoxyl
Antagonist
*
See Ref. [19].

ACCEPTED MANUSCRIPT
F
R
1
OCH
3
OCH
3
OCH
3
OCH
3
H
N
H
N
H
N
H
N
A
B
C
D
O
O
O
O
H
R
3
R
3
R
3
2
1
1
00
50
00
200
150
150
150
100
50
1
00
100
50
0
5
0
0
5
0
0
0
-
10 -9 -8 -7 -6 -5
-10 -9 -8 -7 -6 -5
-10 -9 -8 -7 -6 -5
-10 -9 -8 -7 -6 -5
Log [compound] (M)
Log [compound] (M)
Log [compound] (M)
Log [compound] (M)
Figure 3. Functional activity at CB receptors. Concentration-dependent curves of CB2
2
35
receptor-mediated G-protein activation ([ S]GTPγS binding). A) The effect of methyl
(
triangles solid), n-butyl (open triangels) and benzyl group (solid diamond) in R was
1
evaluated in molecules bearing a hydrogen in R . In panels B-D the impact of different
3
substituents in R is reported. Hydrogen (solid circles), fluorine (open circles) and methoxyl
3
(solid square) groups were tested in molecules bearing a methyl (B), n-butyl (C) and 4-
fluorobenzyl (D) group in R1.
Then, we carried out further experiments to assess the impact of different substituents in R3
35
keeping fixed R . In Fig. 3B-D the [ S]GTPγS binding curves obtained with molecules
1
bearing a methyl (Fig. 3B), n-butyl (Fig. 3C) and 4-fluorobenzyl (Fig. 3D) group in R are
1
showed. The combination of a methyl group in R and a hydrogen, (compound 1s [19]) or a
1
fluorine (compound 1v [19]) in R generates full agonism at CB receptors, while the
3
2
presence of a methoxyl group (compound 1t [19]) turns the molecule into a neutral antagonist
which does not shift the receptor population neither towards the active (i.e. agonist) nor the
inactive (i.e. inverse agonist) conformation (Fig. 3B). When the methyl group in R is
1
replaced by a bigger group such as a n-butyl or 4-fluorobenzyl group, the molecules bearing a
small substituent in R behave as inverse agonists at CB receptors, but again turning into
3
2

ACCEPTED MANUSCRIPT
neutral antagonists when a methoxyl group is present (Fig. 3C-D). As summarized in Table
II, our results indicate that the presence of a methoxyl group in R generates neutral
3
antagonists at CB receptors independently of the type of substituent present in R .
2
1
Nonetheless, when a hydrogen or fluorine is present in R , it is the type of the substituent
3
present in R to dictate the functional activity of the molecule at CB receptors. In particular,
1
2
the presence of a methyl group generates full agonism, while a linear alkyl chain or an
aromatic ring is responsible for inverse agonism. These results indicate that by introducing
little modifications of the size and position of the substituents on the biphenyl scaffold, we
could generate molecules that behave as full agonists, inverse agonists and neutral
antagonists at CB receptor. As hypothesized in the previous docking study, the methoxyl
2
group in R could perfectly fill the TM3-TM5 cleft establishing a T-shape interaction with
3
5.43
W
[19]. Our current results also suggest that molecules bearing the methoxyl group do not
perturb the endogenous balance among the different conformational states of CB receptors,
2
thus behaving as neutral antagonists. On the other side, when a smaller substituent (i.e.
hydrogen and fluorine) is present in R , the molecule preferentially stabilizes the active (e.g.
3
1
s) or the inactive (e.g. 1a, 1g) conformation, thus behaving as an agonist or an inverse
agonist, respectively. In line with these results, we recently reported that the 1,2-dihydro-2-
oxopiridine-3-carboxamides bearing a phenyl or a p-methoxyphenyl group in position 5
behave as inverse agonists or a neutral antagonists, respectively, while the unsubstituted
compound behaves as a full agonist [28]. In our first investigation of the biphenyl
carboxamide scaffold, we showed that the 5-n-butyl-4,4′-dimethoxy-N-cycloheptylbiphenyl-
3
-carboxamide (compound 1u [19]) possesses a strong affinity to CB receptors (K = 11.5
2 i
nM) and a high selectivity over CB (130 fold) behaving as a neutral antagonist in functional
1
assays [19]. In competition experiments, the antagonistic effects of 5-n-butyl-4,4′-dimethoxy-
N-cycloheptylbiphenyl-3-carboxamide (compound 1u [19]) against 1 nM of HU-210 became

ACCEPTED MANUSCRIPT
evident at concentrations that are 1000 to 10000 times higher than the full agonist (i.e. 1-10
µM). Increasing concentrations of the antagonist induced a right-shift of the agonist curve
without apparently changing the maximal response. Nonetheless, the derived Schild plot
indicates a non-competitive antagonism (slope = 0.65) which might likely derive from a
slightly different accommodation of the agonist and antagonist on the binding pocket rather
than an irreversible interaction of the antagonist with CB receptors. In addition, the
2
antagonistic effect of 1u started to occur at 1-10 µM which is 100-1000 higher than the
measured binding affinity. In our current study we included 1u as a reference compound and
we could confirm its neutral antagonist behaviour up to 10 µM. On the other hand, we
obtained a lower binding affinity to CB receptors (K = 249 nM, Table I) compared to the
2
i
previously reported K value (K = 11.5 nM) [19]. Our current binding data are more in line
i
i
35
with the antagonistic effects previously showed by 1u in [ S]GTPγS assays.
In our study, we further evaluated the effect of the new neutral antagonist 1l against the
structurally-related full agonist 1s. As shown in Fig. 4A, increasing concentrations of the
3
5
antagonist determined a right shift of the [ S]GTPγS binding curve of 1s without affecting
the maximal response. The slope obtained from the linear regression analysis reported in the
Schild plot (Fig. 4B, slope = 1.039 ± 0.025) do not differ from the unity, thus confirming a
competitive antagonism. Altogether, our current study provides new insights into the
functional SAR of the biphenyl carboxamides as CB ligands, indicating that the type of
2
substituent in position 5 and 4′ are crucially involved in the functional effect of the molecule
at CB receptors. Our data also suggest the presence of a subordinate relationship between
2
these two positions, with R being responsible for the agonist or inverse agonist behaviour
1
independently of the substituent in R with the exception of the methoxyl group which
3
transforms both full agonists and inverse agonists into neutral antagonists. Overall, our data
provide further information about the SAR of biphenyl carboxamides as CB receptor ligands
2

ACCEPTED MANUSCRIPT
and indicate important features which are responsible for the functional switch between
agonist, inverse agonist and neutral antagonist at CB receptor (see Table II for summary).
2
A
B
2
2
1
1
50
00
50
00
2.0
slope= 1.039 ± 0.025
1
1
.5
.0
plus vehicle
0.5
0.0
plus 12 0.3 µM
plus 11 2l 1 µM
1
l
5
0
0
plus 11 2l 3 µM
plus 11 2l 10 µM
-
10 -9 -8 -7 -6 -5 -4
-7
-6
Log (12)
[1l]
-5
Log[ [1S sS] 0( 5M ])(M)
Figure 4. Compound 1l comepetitively antagonized the 1s-induced CB receptor
2
activation. A) Compound 1l dose-dependently antagonizes the 1s-induced CB receptor
2
activation without attenuating the maximal agonist response. B) The slope calculated from
the Schild-plot (1.039 ± 0.025) is not statistically different from the unity (unpaired t-test)
indicating that the 1l and 1s competitively interact with CB receptors. The results represent
2
the average and S.D. calculated from least three independent experiments performed in
duplicates
We recently reported another class of natural and natural-derived biphenyl compounds as
CB ligands which showed different functional behaviours [12, 29]. Interestingly, some of the
2
compounds bear one or two alkyl ethers on the two phenyl rings similarly to the biphenyl
carboxamides here described. In order to further evaluate the role of the methoxyl group in
determining the antagonist behaviour we evaluated few more compounds of this class of
molecules (Table III).

ACCEPTED MANUSCRIPT
Table III. Cannabinoid receptor pharmacology of additional natural and natural-derived
biphenyl compounds.
OR2
OR1
R3
Compound
R1
H
R2
H
R3
H
K CB
K CB
(nM)
CB effects
i
1
i
2
2
(nM)
Antagonist/
‡
‡
6
3
.46 ±
.54
5.61 ±
2.02
Honokiol
inverse
agonist
*
4
-O′-
2400 ±
600
Agonist
*
H
Methyl
H
H
H
H
H
H
188 ± 116
Methylhonokiol
*
*
Agonist
Agonist
2
a
Methyl
790 ± 96
114 ± 96
1
375 ±
19
2
b
Methyl Methyl
n.d.
3
Antagonist
2
c
H
n-butyl
615 ± 186
103 ± 71
*
1
120 ±
06
Protean
agonist
*
2
d
n-butyl
H
325 ± 115
1
1
783 ±
01
Antagonist
2
e
n-butyl n-butyl
H
n.d.
4
-
-
-
2
f
H
H
H
Methyl
Methyl
Methyl
CH
3
CONH
n.d.
n.d.
n.d.
>10000
>10000
>10000
2
g
C
C
11
17
H
23CONH
35CONH
2
h
H
‡
*
See Ref. [25].
See Ref. [24].

ACCEPTED MANUSCRIPT
The natural compound 4′-O-methylhonokiol (MH) and its structural isomer 2a bear a
methoxyl and a hydroxyl group on the two phenyl rings and both behave as selective and
moderately potent agonists at CB receptors (Table III). Compound 2b, which has two
2
methoxyl groups, retains the agonist activity despite significantly losing the binding affinity
(by a factor 10). A longer alkyl chain (n-butyl) does not affect the binding affinity but
meaningfully modifies the functional activity at CB receptors. Compound 2c and 2d which
2
bear a n-butyl group instead of a methoxyl group compared to MH and 2a, behave as protean
agonist and neutral antagonist, respectively. The double n-butyl ether (2e) showed a 6-10 fold
lower affinity to CB receptors compared to the mono-ether derivatives, similarly to 2b, but
2
unlike the latter compound, 2e behaves as a neutral antagonist (Table III). Honokiol, which
bears two hydroxyl groups is a weak ligand at CB receptors (K value = 2-5 µM) [12, 30].
2
i
The presence of a carboxamide in R completely abolishes the binding to CB receptors
3
2
(compound 2f-h, Table III). Although the small number of compounds, the combination of
binding and functional data suggest that the presence of one alkyl ether is necessary for
generating moderately potent and selective CB ligands, while the presence of two ethers has
2
detrimental effects leading to a significant loss in binding affinity to CB receptors. The
2
length of the alkyl chain, rather than its position on the biphenyl scaffold appears to be
responsible for the functional effects indicating different tridimensional interactions for these
molecules within the binding pocket of CB2 receptors compared to the biphenyl
carboxamides. Assuming a certain similarity in the interaction with CB receptors between
2
the two classes of biphenyl compounds and based on the docking study on biphenyl
carboxamides, we might speculate that the methoxyl group of MH cannot reach the cleft
between the TM3-TM5 as reported for 1u thus leading the molecule to behave as a CB2
agonist. On the other hand, the longer alkyl chain of 2b could protrude into the cleft and
acting as a neutral antagonist, similarly to the proposed docking for 1u.

ACCEPTED MANUSCRIPT
Recently, Smoum et al., [31] reported two enantiomer agonists of the CB cannabinoid
2
receptor, HU-308 and HU-433, with paradoxical pharmacological properties. HU-308
exhibited stronger receptor binding but weaker potency while HU-433 showed opposite
features (i.e. lower binding affinity and higher potency) in several in vitro assays for
osteoblast proliferation and osteoclast differentiation and in mouse models for rescuing the
ovariectomy-induced bone loss and ear inflammation. A molecular-modelling analysis
suggested some small differences in possible binding conformations of the two enantiomers
within the CB receptor binding pocket which seem to be responsible for the striking
2
differences in binding affinity and potency [31]. This indicates that very similar chemical
structures may have significant different orientations relative to the same binding site, leading
to different biological properties, as we described here for some biphenyl carboxamides.
3
. Conclusion
In conclusion, our study provides new insights into the functional SAR of the biphenyl
carboxamides as CB ligands indicating that the type of substituent in position 5 and 4′ are
2
crucially involved in the functional activity of these molecules at CB receptors. Our data
2
also suggest the presence of a subordinate relationship between these two positions, with
position 5 being responsible for the agonist or inverse agonist behaviour, independently of the
substituent in position 4′ with the exception of the methoxyl group which transforms both full
agonists and inverse agonists into neutral antagonists. Our new compounds provide a
complete toolbox of CB functional modulators (i.e. full agonists, inverse agonists and
2
neutral antagonists) derived from the same chemical scaffold and apparently acting on the
same or an overlapping binding site. One of the major confounding factors in studying CB2
pharmacology is related to the constitutive activity of these receptors. This includes the
widespread use of antagonists at concentrations which exhibit inverse agonist activity (i.e.

ACCEPTED MANUSCRIPT
AM630 and SR144528), thus biasing the experimental read-outs. The biphenyl carboxamides
shown here would enable researchers with a valuable set of tool compounds which could be
applied to unambiguously investigate the biological roles of CB receptors in cellular systems
2
and potentially in pathophysiological conditions ex-vivo and in vivo.
4
. Experimental
4
.1 Chemistry
Commercially available reagents were purchased from Sigma Aldrich or Alfa Aesar, and
1
used without purification. H NMR spectra were recorded on a Varian Gemini 200
spectrometer (operating at 200 MHz) or on a Bruker AVANCE III™ 400 spectrometer
1
3
(
operating at 400 MHz). C NMR spectra were recorded on a Bruker AVANCE III™ 400
spectrometer. Chemical shift (δ) are reported in parts per million related to the residual
solvent signal, while coupling constants (J) are expressed in Hertz (Hz). Microwave-assisted
®
reactions were run in a Biotage microwave synthesizer. All of the final products undergoing
biological testing were >96% pure as demonstrated by analysis carried out with a Varian
Prostar HPLC system equipped with an PDA Detector at 260 nm (column Luna C18 (2) 5µ
(150 mm × 4.6 mm)), gradient A/B 70/30 to 90/10 in 20 min, A consisting of methanol, B
consisting of buffer ammonium acetate (pH = 4, 10 mM), flow rate of 0.6 mL/min, room
temperature). High-resolution mass spectra (HRMS) were recorded on a LTQ Orbitrap XL
(Thermo Fisher Scientific) with nanoESI in the positive ion mode (700 - 800 V on the
emitter). Evaporation was carried out under vacuum using a rotating evaporator. Silica gel
flash chromatography was performed using silica gel 60 Å (0.040-0.063 mm; MERK).
Reactions was monitored by TLC on Merck aluminium silica gel (60 F254) plates that were
visualized under a UV lamp (λ = 254 nm).

ACCEPTED MANUSCRIPT
4
.1.1. General procedure for the synthesis of derivatives 2-3 and 11a-d.
A microwave sealed-tube was charged, under nitrogen flux, with a solution of the
corresponding aryl-halide 1 or 9a-b (0.57 mmol) in anhydrous toluene (1.3 mL),
tetrakis(triphenylphosphine)palladium(0) (0.02 mmol), sodium carbonate (1.14 mmol) in
water (1.2 mL) and the proper boronic acid (0.86 mmol) in methanol (1.2 mL). The system
was sealed and heated at 150 °C in a microwave reactor for 10 min (5 bar, 200 W). After
cooling, the reaction mixture was diluted with water and extracted with ethyl acetate. The
organic phases were combined, dried over anhydrous Na SO , filtered, concentrated and
2
4
purified.
4
.1.1.1 Methyl 5-butyl-4-methoxy-[1,1′-biphenyl]-3-carboxylate (2)
A microwave sealed-tube was charged, under nitrogen flux, with a solution of methyl 5-
bromo-3-butyl-2-methoxybenzoate 1 (171.7 mg, 0.57 mmol) in anhydrous toluene (1.3 mL),
tetrakis(triphenylphosphine)palladium(0) (23.1 mg, 0.02 mmol), a solution of sodium
carbonate (120.8 mg, 1.14 mmol) in water (1.2 mL) and a solution of phenylboronic acid
(104.9 mg, 0.86 mmol) in methanol (1.2 mL). The system was sealed and heated at 150 °C in
a microwave reactor for 10 min (5 bar, 200 W). After cooling, the reaction mixture was
diluted with water and extracted with ethyl acetate. The organic phases were combined, dried
over anhydrous Na SO , filtered, concentrated in vacuo. The crude thus obtained was purified
2
4
by flash column chromatography (n-hexane/EtOAc 9:1), affording of pure intermediate 2
1
(
146.2 mg, 0.49 mmol). Yield: 86%. H NMR (200 MHz, CDCl ) δ (ppm): 7.88 (d, 1H, J =
3
2
.4 Hz, H2), 7.60-7.54 (m, 3H, H6, H2′, H6′), 7.48-7.30 (m, 3H, H3′, H4′, H5′), 3.95 (s, 3H,
OCH ), 3.87 (s, 3H, COOCH ), 2.72 (t, 2H, J = 7.8 Hz, CH CH CH CH ), 1.72-1.56 (m, 2H,
3
3
2
2
2
3
CH CH CH CH ), 1.51-1.32 (m, 2H, CH CH CH CH ), 0.95 (t, 3H, J = 7.1 Hz,
2
2
2
3
2
2
2
3
CH CH CH CH ).
2
2
2
3

ACCEPTED MANUSCRIPT
4
.1.1.2 Methyl 5-butyl-4′-fluoro-4-methoxy-[1,1′-biphenyl]-3-carboxylate (3)
Prepared from aryl-halide 1 (171.7 mg, 0.57 mmol) using 4-fluorophenylboronic acid.
Purification by flash column chromatography (n-hexane/EtOAc 9:1). 3 (152.9 mg, 0.48
1
mmol). Yield: 85%. H NMR (200 MHz, CDCl ) δ (ppm): 7.82 (d, 1H, J = 2.6 Hz, H2), 7.57-
3
7
.47 (m, 3H, H6, H2′, H6′), 7.18-7.05 (m, 2H, H3′, H5′), 3.94 (s, 3H, OCH ), 3.86 (s, 3H,
3
COOCH ), 2.71 (t, 2H, J = 7.7 Hz, CH CH CH CH ), 1.71-1.55 (m, 2H, CH CH CH CH ),
3
2
2
2
3
2
2
2
3
1
.50-1.31 (m, 2H, CH CH CH CH ), 0.95 (t, 3H, J = 7.2 Hz, CH CH CH CH ).
2 2 2 3 2 2 2 3
4
.1.1.3 Methyl 5-(4-fluorobenzyl)-4-methoxy-[1,1′-biphenyl]-3-carboxylate (11a)
Prepared from aryl-halide 9a (201.3 mg, 0.57 mmol) using phenylboronic acid. Purification
by flash column chromatography (n-hexane/EtOAc 9:1). 11a (169.7 mg, 0.48 mmol). Yield:
1
8
5%. H NMR (200 MHz, CDCl ) δ (ppm): 7.92 (d, 1H, J = 2.4 Hz, H2), 7.56-7.33 (m, 6H,
3
H6, H2′, H3′, H4′, H5′, H6′), 7.23-7.15 (m, 2H, H2′′, H6′′), 7.02-6.90 (m, 2H, H3′′, H5′′),
4
.07 (s, 2H, benzylic CH ), 3.95 (s, 3H, OCH ), 3.77 (s, 3H, COOCH ).
2 3 3
4
.1.1.4 Methyl 5-(4-fluorobenzyl)-4,4′-dimethoxy-[1,1′-biphenyl]-3-carboxylate (11b)
Prepared from aryl-halide 9a (201.3 mg, 0.57 mmol) using 4-methoxyphenylboronic acid.
Purification by flash column chromatography (n-hexane/EtOAc 9:1). 11b (125.5 mg, 0.33
1
mmol). Yield: 58%. H NMR (200 MHz, CDCl ) δ (ppm): 7.88 (d, 1H, J = 2.5 Hz, H2), 7.49-
3
7
.41 (m, 3H, H6, H2′, H6′), 7.21-7.14 (m, 2H, H2′′, H6′′), 7.01-6.91 (m, 4H, H3′, H5′, H3′′,
H5′′), 4.05 (s, 2H, benzylic CH ), 3.94 (s, 3H, OCH ), 3.84 (s, 3H, OCH ), 3.75 (s, 3H,
2
3
3
COOCH3).
4
.1.1.5 Methyl 4′-fluoro-5-(4-fluorobenzyl)-4-methoxy-[1,1′-biphenyl]-3-carboxylate (11c)

ACCEPTED MANUSCRIPT
Prepared from aryl-halide 9a (201.3 mg, 0.57 mmol) using 4-fluorophenylboronic acid.
Purification by flash column chromatography (n-hexane/EtOAc 9:1). 11c (180.5 mg, 0.49
1
mmol). Yield: 86%. H NMR (200 MHz, CDCl ) δ (ppm): 7.87 (d, 1H, J = 2.6 Hz, H2), 7.51-
3
7
.43 (m, 3H, H6, H2′, H6′), 7.22-7.04 (m, 4H, H3′, H5′, H2′′, H6′′), 7.03-6.93 (m, 2H, H3′′,
H5′′), 4.06 (s, 2H, benzylic CH ), 3.95 (s, 3H, OCH ), 3.76 (s, 3H, COOCH ).
2
3
3
4
.1.1.6 Methyl 5-benzyl-4-methoxy-[1,1′-biphenyl]-3-carboxylate (11d)
Prepared from aryl-halide 9b (191.0 mg, 0.57 mmol) using phenylboronic acid. Purification
by flash column chromatography (n-hexane/EtOAc 9:1). 11d (111.7 mg, 0.336 mmol). Yield:
1
5
9%. H NMR (400 MHz, CDCl ) δ (ppm): 7.92 (d, 1H, J = 2.4 Hz, H2), 7.53-7.50 (m, 3H,
3
H6, H2′, H6′), 7.45-7.38 (m, 2H, H3′, H5′), 7.36-7.17 (m, 6H, H4′, H2′′, H3′′, H4′′, H5′′,
H6′′), 4.11 (s, 2H, benzylic CH ), 3.95 (s, 3H, OCH ), 3.77 (s, 3H, COOCH ).
2
3
3
4
.1.2. General procedure for the synthesis of derivatives 4-5 and 12a-d.
The suitable ester 2-3 or 11a-d (1.96 mmol) was dissolved in methanol (90 mL) followed by
addition of solid potassium hydroxide (19.6 mmol). The resulting suspension was stirred until
completed dissolution of the solids. Then, the mixture was heated at reflux for 19 h. The
reaction was allowed to cool to room temperature, and methanol was removed under vacuum
to afford a yellow oil that was partitioned between water and ethyl acetate. After separation
of the two phases, the aqueous layer was acidified to pH = 2 with 0.6 N hydrochloric acid
solution, to obtain a white precipitate. The precipitate was repeatedly extracted with ethyl
acetate, and the combined organic fractions were dried over anhydrous Na SO , filtered and
2
4
evaporated under vacuum to afford the desired acid derivative, that was used in the next step
without further purification.