Discovery of long-chain salicylketoxime derivatives as monoacylglycerol lipase (MAGL) inhibitors

Article abstract of DOI:10.1016/j.ejmech.2018.08.038

Monoacylglycerol lipase (MAGL) is the enzyme hydrolyzing the endocannabinoid 2-arachidonoylglycerol (2-AG) to free arachidonic acid and glycerol. Therefore, MAGL is implicated in many physiological processes involving the regulation of the endocannabinoid system and eicosanoid network. MAGL inhibition represents a potential therapeutic target for many diseases, including cancer. Nowadays, most MAGL inhibitors inhibit this enzyme by an irreversible mechanism of action, potentially leading to unwanted side effects from chronic treatment. Herein, we report the discovery of long-chain salicylketoxime derivatives as potent and reversible MAGL inhibitors. The compounds herein described are characterized by a good target selectivity for MAGL and by antiproliferative activities against a series of cancer cell lines. Finally, modeling studies suggest a reasonable hypothetical binding mode for this class of compounds.

Full text of DOI:10.1016/j.ejmech.2018.08.038

Accepted Manuscript
Discovery of long-chain salicylketoxime derivatives as monoacylglycerol lipase
(MAGL) inhibitors
Giulia Bononi, Carlotta Granchi, Margherita Lapillo, Massimiliano Giannotti, Daniela
Nieri, Serena Fortunato, Maguie el Boustani, Isabella Caligiuri, Giulio Poli, Kathryn E.
Carlson, Sung Hoon Kim, Marco Macchia, Adriano Martinelli, Flavio Rizzolio, Andrea
Chicca, John A. Katzenellenbogen, Filippo Minutolo, Tiziano Tuccinardi
PII:
S0223-5234(18)30708-6
10.1016/j.ejmech.2018.08.038
EJMECH 10649
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 6 May 2018
Revised Date: 19 July 2018
Accepted Date: 14 August 2018
Please cite this article as: G. Bononi, C. Granchi, M. Lapillo, M. Giannotti, D. Nieri, S. Fortunato, M.e.
Boustani, I. Caligiuri, G. Poli, K.E. Carlson, S.H. Kim, M. Macchia, A. Martinelli, F. Rizzolio, A. Chicca,
J.A. Katzenellenbogen, F. Minutolo, T. Tuccinardi, Discovery of long-chain salicylketoxime derivatives
as monoacylglycerol lipase (MAGL) inhibitors, European Journal of Medicinal Chemistry (2018), doi:
10.1016/j.ejmech.2018.08.038.
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

ACCEPTED MANUSCRIPT
Discovery of long-chain salicylketoxime derivatives as
monoacylglycerol lipase (MAGL) inhibitors
a
a,
a
a
Giulia Bononi , Carlotta Granchi *, Margherita Lapillo , Massimiliano Giannotti , Daniela
b
a
c, d
c
a
Nieri , Serena Fortunato , Maguie el Boustani , Isabella Caligiuri , Giulio Poli , Kathryn E.
e
e
a
a
c, f
Carlson , Sung Hoon Kim , Marco Macchia , Adriano Martinelli , Flavio Rizzolio , Andrea
b
e
a
a
Chicca , John A. Katzenellenbogen , Filippo Minutolo , Tiziano Tuccinardi
a
Department of Pharmacy, University of Pisa, Via Bonanno 6, 56126 Pisa, Italy.
b
Institute of Biochemistry and Molecular Medicine, NCCR TransCure, University of Bern, CH-
012 Bern, Switzerland.
3
c
Pathology Unit, Department of Molecular Biology and Translational Research, National Cancer
Institute and Center for Molecular Biomedicine, Aviano (PN), Italy.
d
Doctoral School in Molecular Biomedicine, University of Trieste, 34100 Trieste, Italy.
e
Department of Chemistry, University of Illinois at Urbana-Champaign, 600 S. Mathews Avenue,
Urbana, IL 61801, USA.
f
Department of Molecular Sciences and Nanosystems, Ca' Foscari University, Venezia 30123,
Italy.
*
Corresponding author.
E-mail address: carlotta.granchi@unipi.it (C. Granchi).
Keywords: Monoacylglycerol lipase inhibitors, MAGL, cancer, ketoxime.
Highlights
•
35 salicylketoxime derivatives were synthesized.
1

ACCEPTED MANUSCRIPT
•
•
•
Compounds 13a and 14a are potent reversible MAGL inhibitors.
Compounds 13a and 14a proved to be selective for MAGL.
Compounds 13a and 14a reduced the proliferation of a series of cancer cell lines.
ABSTRACT
Monoacylglycerol lipase (MAGL) is the enzyme hydrolyzing the endocannabinoid 2-
arachidonoylglycerol (2-AG) to free arachidonic acid and glycerol. Therefore, MAGL is implicated
in many physiological processes involving the regulation of the endocannabinoid system and
eicosanoid network. MAGL inhibition represents a potential therapeutic target for many diseases,
including cancer. Nowadays, most MAGL inhibitors inhibit this enzyme by an irreversible
mechanism of action, potentially leading to unwanted side effects from chronic treatment. Herein,
we report the discovery of long-chain salicylketoxime derivatives as potent and reversible MAGL
inhibitors. The compounds herein described are characterized by a good target selectivity for
MAGL and by antiproliferative activities against a series of cancer cell lines. Finally, modeling
studies suggest a reasonable hypothetical binding mode for this class of compounds.
1. Introduction
Monoacylglycerol lipase (MAGL) is a 33 kDa serine hydrolase that peripherally (i.e. adipose tissue,
liver, etc.) cleaves monoacyglycerols to fatty acids and glycerol. In particular, in the central nervous
system, MAGL is the main enzyme responsible of the degradation of 2-arachidonoylglycerol (2-
AG) to arachidonic acid and glycerol. 2-AG is one of the two major endocannabinoids, the other
being anandamide (AEA), which is instead hydrolyzed by the enzyme fatty acid amide hydrolase
(FAAH) [1]. 2-AG is produced on demand and represents one of the most important endogenous
ligands that activate the G protein-coupled cannabinoid receptors, CB1 and CB2. Therefore, 2-AG
2

ACCEPTED MANUSCRIPT
is involved in the modulation of many pathological and physiological processes, including pain,
inflammation, appetite, memory and emotion.
In 2011, a functional proteomic analysis of a panel of aggressive and non-aggressive human cancer
cell lines from multiple tumors was conducted in order to identify enzyme activities that contributed
to cancer pathogenesis, and it was found that MAGL levels were consistently more elevated in
aggressive cancer cells than in their non-aggressive counterparts. Moreover, it was demonstrated
that MAGL was responsible for high serine hydrolase activities found in aggressive cancer cells [2–
4]. It is intriguing that the mechanism of cancer aggressiveness provoked by elevated MAGL
activity involves the supply of fatty acids for the production of pro-tumorigenic signaling lipids:
MAGL controls the pools of free fatty acids that are the building blocks for these pro-tumorigenic
signaling lipids, such as prostaglandin E2 (PGE2) and lysophosphatidic acid (LPA) [5]. MAGL
blockade leads to an accumulation of the endocannabinoid 2-AG, thus suppressing arachidonic acid
formation and the consequent production of pro-inflammatory prostaglandins [6] At the same time
.
the inhibition of the enzymatic degradation of 2-AG leads to amplified effects on CB1, indirectly
enhancing endocannabinoid signaling without causing the problems that are typical of CB1 agonists
[
5,7,8]. Therefore, inhibition of MAGL was found to reduce pain, inflammation, anxiety, nausea
and depression. Considering these multiple beneficial effects, from a therapeutic point of view, it is
clear that MAGL is a promising target for cancer as well as for other pathologies, such as chronic
pain or inflammatory diseases.
Most of the MAGL inhibitors so far reported in the literature bind irreversibly to the enzyme.
However, chronic and irreversible MAGL inactivation is often associated with CB1 receptor
desensitization leading to unwanted effects, such as functional antagonism at CB1 receptors and
physical dependence [9]. Presently, only few of the currently discovered MAGL inhibitors show a
reversible mode of action, although this mode of action is considered to be a preferable and
potentially safer alternative to irreversible inhibition [10]. Two widely known examples of
irreversible and potent inhibitors are JZL184 (4-nitrophenyl-4-[bis(1,3-benzodioxol-5-
3

ACCEPTED MANUSCRIPT
yl)(hydroxy)methyl]piperidine-1-carboxylate 1, Figure 1)[11] and CAY10499 (benzyl(4-(5-
methoxy-2-oxo-1,3,4-oxadiazol-3(2H)-yl)-2-methylphenyl)carbamate 2, Figure 1) [12]. To the best
of our knowledge, the naturally occurring terpenoids Euphol (3, Figure 1) and Pristimerin (4, Figure
1) are the first reversible MAGL inhibitors ever reported. However, these two compounds act on a
large number of secondary targets; therefore, they are unsuitable for further development as MAGL
inhibitors [13,14]. Another MAGL inhibitor from natural sources is the triterpenoid β-amyrin,
structurally related to Pristimerin and Euphol, although it was found to be more active on other
hydrolases than on MAGL [7].
Hernández-Torres et al. reported a MAGL inhibitor c21 (benzo[d][1,3]dioxol-5-ylmethyl 6-([1,1′-
biphenyl]-4-yl)-hexanoate 5, Figure 1), which was demonstrated to improve the clinical outcome of
multiple sclerosis in vivo using the experimental allergic encephalomyelitis mouse model and for
the first time it was observed that the therapeutic effects of this reversible inhibitor were not
accompanied by catalepsy or other motor impairments that often occur after the administration of
irreversible inhibitors [15]. More recently, a growing number of reversible MAGL inhibitors have
been reported, such as (4-(4-chlorobenzoyl)piperidin-1-yl)(3-hydroxyphenyl)methanone 6 (Figure
1), a nanomolar inhibitor developed by our group, which originated from an optimization study of
an initial compound discovered by a virtual screening procedure [16,17] and (4-benzylpiperidin-1-
yl)(5-(4-hydroxyphenyl)-1-(3-methylbenzyl)-1H-pyrazol-3-yl)methanone 7 (Figure 1), which was
found to relieve the neuropathic hypersensitivity induced in vivo by oxaliplatin [18]. A loratidine
analogue was recently discovered as a potent MAGL inhibitor that also showed anti-histaminergic
activity, making it a potentially useful agent in inflammatory pathologies in which the simultaneous
blockade of these targets would be desirable [19].
4

ACCEPTED MANUSCRIPT
Figure 1. Structures of some of the most relevant MAGL inhibitors.
In the search for new MAGL inhibitors, we screened an in-house library of published compounds,
and we found that compounds 8 and 9 (Figure 2) showed a certain inhibition of MAGL activity,
with IC values of 34 and 11 µM, respectively. These two compounds, which were previously
5
0
developed as estrogen receptor ligands [20], share the same salicylketoxime scaffold, with a
peripheral phenolic ring substituted with a fluorine atom in a position ortho to the hydroxyl group,
and differ only in the alkyl group of the ketoxime moiety. In particular, we observed about a three-
fold increase of activity in passing from the methyl (compound 8) to the ethyl ketoxime substituent
(compound 9), thus suggesting a possible improvement in inhibitory activity by further increasing
5

ACCEPTED MANUSCRIPT
the length of this alkyl chain. This trend is reasonable considering that the MAGL active site
normally hosts the substrate 2-AG, which is a very long-chain ester of arachidonic acid and
glycerol. Therefore, the presence of long alkyl chains or bulky groups are likely to be well tolerated
by the enzyme, and the establishment of lipophilic interactions between molecules having long
alkyl chains and the enzyme active site could enhance the inhibitory potency of such compounds.
These data prompted us to further develop this scaffold and generate new derivatives, by
introducing the following modifications on the parent compounds 8 and 9: a) we progressively
increased the length of the linear saturated ketoximic alkyl chain in compounds 10-15. In fact,
compounds 10a-d have relatively small chains with 4 carbon atoms (n = 3, Figure 2), while
compounds 11a-d, 12a-d, 13a-d, 14a-d and 15a-d have gradually longer groups with 7, 9, 11, 13 or
15 carbon atoms, respectively (n = 6, 8, 10, 12, 14, Figure 2); b) simple aromatic groups in the
ketoxime moiety were also explored, in particular a phenyl ring (n = 0, compounds 16a-d, Figure 2)
or a benzyl group (n = 1, compounds 17a-d, Figure 2); c) in all the previous compounds mentioned
at points a) and b), we varied the substituents of the peripheral phenolic ring, in order to determine
their importance; therefore, we initially maintained the p-OH/m-F substitution pattern (R = OH, R1
2
=
=
F, compounds “a” of Figure 2), and we also introduced the relative methoxylated derivatives (R2
OCH , R = F, compounds “b” of Figure 2). Moreover, we removed the fluorine atom to assess its
3
1
role in the inhibitory activity of these compounds, thereby obtaining the p-hydroxyl derivatives (R2
OH, R = H, compounds “c” of Figure 2) and the related p-methoxy analogues (R = OCH , R =
=
1
2
3
1
H, compounds “d” of Figure 2). Finally, in order to acquire information about the binding mode of
this class of compounds, the last modifications were applied only to compound 14, which had
shown the best enzymatic inhibitory activity: we either removed all the substituents on the
peripheral phenyl ring, or introduced a fluorine atom or a hydroxyl group in the meta position
(compounds 14e-g, Figure 2).
6

ACCEPTED MANUSCRIPT
increasing MAGL inhibitory activity
OH
HO
OH
F
HO
N
F
N
HO
HO
8
9
R2
R1
R2
R1
n
n
HO
HO
N
N
HO
HO
10-15
16-17
1
1
1
1
1
1
0a-d: n = 3;
1a-d: n = 6;
2a-d: n = 8;
3a-d: n = 10;
4a-d: n = 12;
5a-d: n = 14.
a: R = F, R = OH;
a: R = F, R = OH;
1
2
1
2
b: R1 = F, R2 = OCH3; 16a-d: n = 0; b: R1 = F, R2 = OCH3;
17a-d: n = 1. c: R = H, R = OH;
c: R = H, R = OH;
1
2
1
2
d: R = H, R = OCH .
d: R = H, R = OCH .
1 2 3
1
2
3
12
HO
N
R1
HO
1
1
1
4e: R = H;
1
4f: R = F;
1
4g: R = OH.
1
Figure 2. Design of salicylketoximes with bulky chains (in red): starting compounds 8 and 9 and
newly designed derivatives, bearing alkyl (10-15) or aromatic chains (16-17).
2
2
. Results and discussion
.1. Chemistry
The synthesis of derivatives 10-15 started from the corresponding commercially available
carboxylic acid, valeric acid 18 (n = 3, Scheme 1), ottanoic acid 19 (n = 6, Scheme 1), decanoic
acid 20 (n = 8, Scheme 1), dodecanoic acid 21 (n = 10, Scheme 1), mirystic acid 22 (n = 12,
7

ACCEPTED MANUSCRIPT
Scheme 1) or palmitic acid 23 (n = 14, Scheme 1), which was converted to the acid chloride by
refluxing it with thionyl chloride for three hours. After evaporation of the excess of SOCl , the acid
2
chloride was subjected to a Friedel-Crafts reaction with 4-bromoanisole in the presence of
aluminum trichloride and 1,2-dichloroethane as the solvent: as expected, the acylation was directed
to the meta position with respect to the bromine atom, but at the same time we also observed the
desired demethylation of the methoxy group, thus giving derivatives 24-29 directly (Scheme 1). The
acylated intermediates were subjected to a Pd-catalyzed cross-coupling reaction under classical
Suzuki conditions [21], with the appropriate arylboronic acid: 3-fluoro-4-methoxyphenylboronic
acid for intermediates 30, 32, 34, 36, 38 and 40, 4-methoxyphenylboronic acid for compounds 31,
33, 35, 37, 39 and 41, phenylboronic acid for compound 42, 3-fluorophenylboronic acid for
compound 43 and 3-methoxyphenylboronic acid for compound 44 (Scheme 1). The biaryl
derivatives obtained were in part O-demethylated with BBr at 0 °C to afford free phenols 45-57.
3
All the resulting phenolic derivatives 45-57, their methoxylated precursors 30-41 and compounds
42 and 43, bearing no substituents or only a m-fluorine group, respectively, were directly
transformed into the corresponding salicylketoximes 10a-d, 11a-d, 12a-d, 13a-d, 14a-g and 15a-d
by a condensation reaction with hydroxylamine hydrochloride in ethanol (Scheme 1).
8

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of salicylketoximes 10-15. Reagents and conditions: a) i. SOCl , 95 °C, 3 h; ii.
2
4
-bromoanisole, AlCl , DCE, 65 °C, 2 h; b) ArB(OH) , Pd(OAc) , PPh , aqueous 2 M Na CO ,
3 2 2 3 2 3
toluene/EtOH (1:1), 100 °C, 24 h; c) BBr , CH Cl , - 10 °C to 0 °C, 1 h; d) NH OH·HCl, EtOH, 50
3
2
2
2
°
C, 12-48 h.
Salicylketoximes 16-17, bearing an aromatic substituent such as phenyl (n = 0, Scheme 2) or benzyl
(n = 1, Scheme 2) in the ketoximic portion, were prepared following a similar synthetic scheme,
9

ACCEPTED MANUSCRIPT
differing only for the initial step in which the starting acid chlorides were commercially available
(benzoyl chloride 58 or phenylacetyl chloride 59) and were directly used in the Friedel-Crafts
reaction (step a, Scheme 2). Both the methoxylated derivatives 62-65 as well as the phenolic
intermediates 66-69 were transformed in the corresponding oxime derivatives 16a-d and 17a-d by a
condensation reaction with hydroxylamine hydrochloride in ethanol under mild heating.
Scheme 2. Synthesis of salicylketoximes 16-17. Reagents and conditions: a) 4-bromoanisole, AlCl3,
DCE, 65 °C, 2 h; b) ArB(OH) , Pd(OAc) , PPh , aqueous 2 M Na CO , toluene/EtOH (1:1), 100
2
2
3
2
3
°
C, 24 h; c) BBr , CH Cl , - 10 °C to 0 °C, 1 h; d) NH OH·HCl, EtOH, 50 °C, 12-48 h.
3 2 2 2
Finally, we decided to insert an isopropyl group as the ketoximic chain, in order to compare the
biological results obtained from compounds 10a-d, 11a-d, 12a-d, 13a-d, 14a-g bearing long linear
1
0

ACCEPTED MANUSCRIPT
alkyl chains with those of compounds 16a-d and 17a-d possessing bulky aromatic groups. The
isopropyl-substituted compounds 76 and 77 (Scheme 3) may represent an intermediate situation
between very flexible chains and highly bulky groups. These two compounds differ only for the
presence of an additional fluorine atom in meta position of the peripheral phenolic ring of
compound 76, which is not present in compound 77, since in other analogues of this series these
two substitution combinations gave the best results in enzymatic assays (see section 2.2. MAGL
enzymatic assays). Compounds 76 and 77 were prepared following the same synthetic scheme
described above starting from isobutyryl chloride 70, which was subjected to a Friedel-Crafts
reaction to produce phenol 71. A cross-coupling reaction with the properly substituted
phenylboronic acid gave compounds 72 and 73. These biphenylic compounds were subjected to a
BBr -promoted demethylation of the methoxy group and the resulting compounds 74 and 75 were
3
then reacted with hydroxylamine hydrochloride to obtain the final ketoximic products.
Scheme 3. Synthesis of salicylketoximes 76-77. Reagents and conditions: a) 4-bromoanisole, AlCl3,
DCE, 65 °C, 2 h; b) ArB(OH) , Pd(OAc) , PPh , aqueous 2 M Na CO , toluene/EtOH (1:1), 100
2
2
3
2
3
°
C, 24 h; c) BBr , CH Cl , - 10 °C to 0 °C, 1 h; d) NH OH·HCl, EtOH, 50 °C, 24-72 h.
3 2 2 2
1
1

ACCEPTED MANUSCRIPT
2.2. MAGL enzymatic assays
The synthesized compounds were tested for their inhibitory activity on human MAGL (Table 1)
together with the known irreversible MAGL inhibitor CAY10499 (2) and the (4-(4-
chlorobenzoyl)piperidin-1-yl)(3-hydroxyphenyl)methanone 6, which were used as reference
compounds. Moreover, the new compounds were compared with the parent methylketoxime 8 and
ethylketoxime 9.
We observed a common trend in the p-OH, m-F-phenyl- and p-OH-phenyl-substituted series of
compounds bearing alkyl chains: the potency of inhibition increases in parallel with the number of
methylene units in the alkyl ketoximic chain; therefore, as long as the length of this chain increases,
the compounds become more potent, eventually reaching nanomolar inhibition values. In particular,
this trend was observed when n = 10, 12 or 14 in the p-OH m-F group (compounds 13a, 14a and
15a), and when n = 12 or 14 (compounds 14c and 15c) for the p-OH group. We were pleased to
observe these encouraging results from these newly developed compounds, because it confirmed
our initial hypothesis regarding the correlation between chain length and inhibitory activity. Indeed,
the parental short-chain ketoximic derivatives 8 and 9 were about 50 or 16-fold less potent,
respectively, than the best compound of this new series (14a).
Our data suggest that a further increase of the chain length over 12 carbons becomes detrimental for
the interaction with the enzyme. The palmitoyl derivative (n = 14) showed a slight decrease of
potency as compared to the myristoyl derivative (n = 12). In fact, compound 14a resulted to be the
most potent MAGL inhibitor of this series of compounds, showing an IC50 value of 0.68 µM.
Compound 14a is slightly more potent than the close analogue 13a which bears a shorter chain
(
IC = 0.71 µM), and more active than compound 15a (IC = 0.91 µM), which differs from 14a
50 50
only for an extra two methylene units. Similarly, the same behaviour was observed for the non-
fluorinated p-OH substituted derivatives, among which 14c was the most potent inhibitor (IC =
5
0
1
2

ACCEPTED MANUSCRIPT
0.77 µM). When the length of the alkyl chain was increased over 12 carbons, there was a loss of
potency (n = 14, compound 15c, IC = 0.96 µM).
5
0
From the comparison between the just-mentioned series of derivatives, it is reasonable to state that
the presence of the fluorine atom in combination with the hydroxyl group led to a slight
improvement of activity with respect to the analogue phenolic compound without the fluorine atom
(i.e., comparison between 14a and 14c). This fact could be explained by considering that the
electron-withdrawing fluorine atom increases the acidity of the adjacent OH group, increasing its
polarization, and thus enhancing its ability to establish hydrogen bonds (see Molecular modeling
section). Poorer activities were generally obtained with the compounds possessing aromatic
ketoximic chains: both the benzyl and phenyl-substituted compounds showed similar activities in
the low micromolar range, and also in this case the fluorine-bearing compounds 16a and 17a (IC50
=
1.7 µM for both of them) proved to inhibit MAGL with a higher potency compared to compounds
1
6c and 17c (IC value of 7.2 and 4.8 µM, respectively).
5
0
Overall these data suggest that the presence of a long aliphatic chain in the ketoxime moiety is well
tolerated by the enzyme, but bulkier and less flexible groups, such as a phenyl or a benzyl ring, lead
to a detrimental loss of activity. However, the length of the alkyl chain reaches a sort of “plateau” at
n = 12, after which a further increase results in a significant decrease of the biological activity. In
order to confirm the importance of the p-hydroxy group, the relative methoxylated derivatives were
tested (series “b” and “d”): all these compounds were less active than their hydroxylated
counterparts, with IC values in the range of 12-119 µM or greater than 200 µM.
5
0
Since compound 14a was the most promising compound of this class of derivatives, with an
inhibitory potency that was even higher than that of reference compound 6, we decided to further
explore substitutions on the peripheral phenyl ring. Removal of all the substituents (14e) or the
presence of a meta-fluoro or a meta-hydroxy group (14f and 14g, respectively) resulted in only
modest inhibition of enzyme activity for all the three compounds; the 3 to 37-fold decrease of
1
3

ACCEPTED MANUSCRIPT
activity for these derivatives supported the hypothesis that the p-hydroxy m-fluoro substitution
pattern is a key to improve inhibition of MAGL activity.
For what concerns i-propyl-substituted compounds 76 and 77, they showed IC values in the low
5
0
micromolar range (IC = 3.3 µM for 76 and 5.1 µM for 77). Compound 76 was slightly less active
5
0
than the other corresponding p-OH/m-F substituted compounds, such as the bulkier aryl derivatives
6a and 17a (IC = 1.7 µM), or the compounds bearing linear alkyl chains 10a-15a (IC = 0.68-2
1
5
0
50
µM). Compound 77 showed approximately the same inhibition potency as those displayed by p-OH
substituted compounds 16c and 17c (IC = 4.8-7.2 µM) and by compound 10c bearing the shortest
5
0
linear chain (IC = 5.2 µM), whereas its activity was lower than those of compounds 11c-15c with
5
0
longer alkyl chains (IC = 0.77-2.5 µM). These two i-propyl-substituted compounds maintain the
5
0
same trend that was observed for the previous series of compounds, which consisted in an inhibition
activity improvement given by the additional presence of the fluorine atom in ortho position to the
phenolic hydroxyl group.
Table 1. MAGL inhibitory activity of compounds 2, 6, 8, 9 10a-d, 11a-d, 12a-d, 13a-d, 14a-g,
15a-d, 16a-d, 17a-d, 76 and 77.
Compound
Structure
IC (mean ± SD, µM)
50
CAY10499
0
.14 ± 0.03
(2)
6
8
0.84 ± 0.04
34 ± 1
1
4

ACCEPTED MANUSCRIPT
9
11 ± 2
1
1
1
1
1
1
0a
1a
2a
3a
4a
5a
n = 3
2.0 ± 0.1
1.3 ± 0.2
n = 6
n = 8
1.0 ± 0.2
n = 10
n = 12
n = 14
0.71 ± 0.03
0.68 ± 0.01
0.91 ± 0.02
1
1
6a
7a
n = 0
n = 1
1.7 ± 0.9
1.7 ± 0.1
1
1
1
1
1
1
0b
1b
2b
3b
4b
5b
n = 3
36 ± 2
16 ± 1
19 ± 3
53 ± 3
> 200
>200
n = 6
n = 8
n = 10
n = 12
n = 14
1
1
6b
7b
n = 0
n = 1
26 ± 5
12 ± 1
1
1
1
1
1
1
0c
1c
2c
3c
4c
5c
n = 3
5.2 ± 0.2
2.5 ± 0.2
n = 6
n = 8
1.5 ± 0.3
n = 10
n = 12
n = 14
1.1 ± 0.1
0.77 ± 0.04
0.96 ± 0.14
1
1
6c
7c
n = 0
n = 1
7.2 ± 0.4
4.8 ± 0.2
1
1
0d
1d
n = 3
n = 6
119 ± 13
30 ± 2
1
5

ACCEPTED MANUSCRIPT
1
1
1
1
2d
3d
4d
5d
n = 8
36 ± 4
69 ± 6
28 ± 3
70 ± 5
n = 10
n = 12
n = 14
1
1
6d
7d
n = 0
n = 1
34 ± 3
20 ± 1
1
4e
4f
4g
R
R
R
1
1
1
= H
25 ± 5
18 ± 3
1
= F
1
= OH
2.4 ± 0.5
7
6
7
R
R
1
1
= F
3.3 ± 0.5
5.1 ± 0.4
7
= H
In order to verify whether the compounds could interact with the cysteines of the MAGL enzyme,
the activity of the most potent inhibitor 14a was also tested in the presence of the thiol-containing
agent 1,4-dithio-DL-threitol (DTT). It is known that MAGL is sensitive to inhibition by sulfhydryl-
specific agents, since some compounds reported in literature irreversibly inhibit MAGL, being
involved in a Michael addition reaction that leads to covalent modifications of some key cysteine
residues, such as Cys201, Cys208 or Cys242 [22–24]. DTT-based experiments were used as a first
demonstration that these compounds are reversible MAGL inhibitors, ruling out their covalent
interactions with cysteines. As shown in Figure 3A, the IC value of compound 14a was only very
5
0
slightly, but not significantly, influenced by the presence of DTT, shifting from 0.68 µM when
assayed in the absence of DTT to 0.70 µM when assayed in the presence of 10 µM DTT, thus
excluding any interaction of these compounds with the cysteine residues present in MAGL.
Furthermore, with the aim of establishing whether the mechanism of inhibition was reversible or
irreversible, the effects of dilution and preincubation on the inhibitory activity of compound 14a
were evaluated. In case of an irreversible inhibition, potency should not decrease after dilution,
1
6

ACCEPTED MANUSCRIPT
whereas in case of a reversible inhibition, the potency level should be substantially reduced after
dilution [24]. In our experiment, the inhibition produced by preincubation with a 20 µM
concentration of 14a was measured after a 40X dilution and compared to the potency observed by a
20 µM and a 0.5 µM of compound 14a. As shown in Figure 3B, the results were consistent with a
reversible inhibition mechanism, as the inhibition produced by 0.5 µM of the compound was similar
to that after a 40X dilution and was different to that produced by the compound at a concentration
of 20 µM. As a second test of mechanism, the biological activity of 14a was also tested at different
preincubation times of compound with MAGL. In this assay, the compound was preincubated with
the enzyme for 0, 30 and 60 minutes before adding the substrate to start the enzymatic reaction. An
irreversible inhibitor should show a higher potency after longer incubation times, whereas a
reversible inhibitor should show a constant inhibition potency that was independent of the
incubation time. As shown in Figure 3C, this assay confirmed the reversible property of 14a, as it
did not show any significant increase in inhibitory potency at longer incubation times.
Figure 3. Compound 14a-MAGL inhibition analysis. A) Effect of DTT on the MAGL inhibition
properties. B) Dilution assay: the first two columns indicate the inhibition percentage of compound
14a at a concentration of 20 µM and 0.5 µM. The third column indicates the inhibition percentage
1
7

ACCEPTED MANUSCRIPT
of compound 14a after dilution (final concentration = 0.5 µM). C) IC (µM) values of 14a at
5
0
different preincubation times with MAGL (0 min, 30 min and 60 min).
2.3. Selectivity towards other targets
With the aim of evaluating the selectivity of the most potent MAGL inhibitors towards the other
targets of the endocannabinoid system, compounds 13a and 14a were tested on FAAH, ABHD6,
ABHD12 and cannabinoid receptors CB1 (CB1R) and CB2 (CB2R) (Table 2). Both compounds
had low activity on ABHD6 and ABHD12 (IC values greater than 10 µM), and only a weak
5
0
activity on FAAH (IC = 4.1 and 8.7 µM for 13a and 14a, respectively). The binding data showed
5
0
that compounds 13a and 14a have a moderate affinity towards CB1R (K values of 2.6 and 4.1 µM
i
for 13a and 14a, respectively), while they do not bind to CB2R at the concentration of 10 µM (5%
of binding, data not shown). The long alkyl chain present in the structure of the salicylketoxime
derivatives 13a and 14a resembles the arachidonoyl tail of the endocannabinoids AEA and 2-AG,
and this may be responsible of the moderate effects observed on FAAH and CB1R. However, as
AEA and 2-AG have different affinities for CB1R and CB2R and the degrading enzymes, our
compounds 13a and 14a also showed a preferential binding towards MAGL as compared to FAAH
(selectivity factor = 6 and 13 for 13a and 14a, respectively), ABHDs (selectivity factor ≥ 14 for 13a
and 14a), CB1R (selectivity factor = 4 and 6 for 13a and 14a, respectively) and CB2R (selectivity
factor ≥ 15 for 13a and 14a).
Table 2. Biological activities (IC values for FAAH, ABHD6 and ABHD12 and K values for
5
0
i
CB1R and CB2R, means ± SD) of compounds 13a and 14a on the major components of the
endocannabinoid system.
IC (mean ± SD, µM)
K (mean ± SD, µM)
i
5
0
1
8

ACCEPTED MANUSCRIPT
Compound
FAAH
4.1 ± 1.3
8.7 ± 1.8
ABHD6
> 10
> 10
ABHD12
> 10
CB1R
2.6 ± 0.8
4.1 ± 1.1
CB2R
> 10
> 10
1
1
3a
4a
> 10
Considering that the scaffold of these molecules derives from a structural evolution that started
from compounds previously developed as estrogen receptor ligands [20], we decided to test whether
the newly synthesized compounds maintained a certain affinity for ERα and ERβ. The binding
affinity for ERα and ERβ of salicylketoximes 13a and 14a, showing the best inhibitory potency on
3
MAGL, was measured by a radiometric competitive binding assay with [ H]estradiol, using
previously reported methods [25,26]. The relative binding affinity (RBA) values for the newly
reported compounds are summarized in Table 3. RBA values are reported as percentage (%) of that
of estradiol, which is set at 100% and compared to those of the initial compound 9. Compound 9
showed very similar values to those previously published, which showed good affinity and
selectivity for ERβ, compared to estradiol. Differently, both the new compounds, 13a and 14a, had
a negligible affinity for ERα and a very low residual affinity for ERβ, which appears to decrease
with the increasing length of the ketoximic chain (Table 3). In fact, ethyl-substituted derivative 9
had a >5-fold greater ERβ affinity than that of compound 13a (n = 10, Figure 2), which showed a
RBA value of 6.8. The ERβ affinity of the longer-chain derivative 14a (n = 12, Figure 2) was found
to be further reduced (RBA = 2.6), being 2.6-fold lower than that of 13a.
Table 3. Relative Binding Affinities of compounds 9, 13a and 14a for the Estrogen Receptors α and
a
β (RBA, %)
Compound
ERα
ERβ
Estradiol
(100)
(100)
9
2.1
37
1
3a
0.31
6.8
1
9

ACCEPTED MANUSCRIPT
1
4a
0.17
2.6
a
This long-used binding assay has a coefficient of variation of 0.3.
2.4. Antiproliferative assays
Compounds 13a and 14a were also selected for further in vitro experiments to evaluate their
antiproliferative potency against cancer cells. Compound CAY10499 (2) was used as the reference
compound. Due to the key role that MAGL plays in the tumor progression of breast, colon and
ovarian cancers [4,27,28], five tumor cell lines were chosen: the human breast MDA-MB-231, the
colorectal HCT116 and the ovarian CAOV3, OVCAR3 and SKOV3 cancer cells (Table 4).
Salicylketoximes 13a and 14a produced an appreciable inhibition of cell viability in all the tested
cell lines, with IC values ranging from 7.6 to 73 µM. With respect to the covalent reference
5
0
inhibitor CAY10499, both compounds 13a and 14a showed a more potent cytotoxic activity on
CAOV3 and MDA-MB-231, with IC values of 11 and 7.6 µM for 13a, respectively, and of 14 and
5
0
1
1 µM for 14a, respectively (IC = 92 and 89 µM, respectively, for CAY10499). The proliferation
50
of the HCT116 tumor cells was similarly affected by 13a and CAY10499, whereas 14a was slightly
less potent (about 1.7-fold). Finally, compounds 13a and 14a exerted good antiproliferative potency
on OVCAR3 and SKOV3, always maintaining better activities than those of the reference inhibitor
(
IC ranging from 20 to 40 µM for ketoxime derivatives vs. 34 and 50 µM of CAY10499).
50
Table 4. Cell growth inhibitory activities (IC values) of compounds 2, 13a, 14a.
5
0
IC (mean ± SD, µM)
5
0
Compound
CAY10499 (2)
HCT116
42 ± 2
45 ± 5
MDA-MB-231
89 ± 4
CAOV3
92 ± 5
11 ± 2
14 ± 1
OVCAR3
50 ± 3
40 ± 6
SKOV3
34 ± 3
20 ± 2
26 ± 2
1
1
3a
4a
7.6 ± 1.2
11 ± 1
73 ± 6
28 ± 2
2.5. Molecular modeling
2
0

ACCEPTED MANUSCRIPT
To suggest a possible binding mode for this class of derivatives, the interaction of compound 14a
with MAGL was analysed by means of docking and molecular dynamic (MD) simulations. This
compound was thus docked by using Autodock 4.2, and the best docking pose was subjected to 100
ns of MD simulation with explicit water molecules, as described in the Experimental section. Figure
4
shows the main interactions of 14a with MAGL. The p-hydroxy-m-fluorophenyl fragment forms
two H-bonds with the backbone nitrogen of A51 and the hydroxyl group of S122 in the oxyanion
hole of the protein binding site. The biphenyl fragment shows lipophilic interactions with I179,
V183, L184 and L241, whereas the salicylic hydroxyl group establishes an H-bond with the
backbone nitrogen of D180 and forms a pseudocyclic system together with the oxime portion of the
molecule. With regards to the alkyl chain, it occupies a portion of the binding site characterized by
a high number of lipophilic residues and, in particular, it interacts with A151, A156, F159, L205,
L213 and L214.
Figure 4. Minimized average structure of compound 14a docked into MAGL.
2.6. LogP analysis
2
1

ACCEPTED MANUSCRIPT
In order to have information about the lipophilic properties of the herein reported compounds, an
evaluation of the logP has been carried out. Table 5 shows the consensus logP values obtained
through the Swiss ADME web tool [29], which combines five different logP calculation methods,
and the clogP values calculated by DataWarrior software [30]. As shown in Figure S75, there is a
good agreement between consensus and clogP, although this second method predicted logP values
with a scaling factor of 1.24 ± 0.05 with respect to the consensus logP. As expected, the higher the
number of carbon atoms in the compounds, the bigger the corresponding logP values calculated by
both methods, with the maximum value obtained for compound 15b. Nevertheless, no evident
correlation between logP and activity of the different compounds was observed; in fact, by plotting
2
the logP values against the measured pIC values of the ligands a correlation (R ) lower than 0.1
5
0
was obtained (see Figure S76). Finally, from this analysis we can highlight that, comparing the two
compounds 13a and 14a that showed the most promising activities, compound 13a should be
considered as the most promising candidate, since it shows a lower logP value than that of 14a. It is
worth mentioning that, although derivatives with moderate lipophilicity would be desirable to
assure a good bioavailability and toxicity profile, ligands with a predominantly lipophilic nature
could still be used for in vivo administration. As an example, compound 5 (Figure 1), for which a
consensus logP value of 5.5 was predicted (Table 5), showed a suitable pharmacokinetic profile and
demonstrated a pronounced activity in a mouse EAE in vivo model [15]. Nevertheless, structural
optimization studies aimed at developing further salicylketoxime derivatives with improved ADME
profiles and MAGL inhibition potencies will be performed in an attempt of developing better
candidates for in vivo studies.
Table 5. Calculated consensus logP and clogP values for the reported compounds.
Compound
Cons logP
2.80
clogP
3.22
3.68
8
9
2.95
2
2

ACCEPTED MANUSCRIPT
1
1
1
1
1
1
1
1
0a
1a
2a
3a
4a
5a
6a
7a
3.68
4.66
5.53
6.19
6.91
7.68
3.84
3.91
4.05
5.14
5.90
6.51
7.30
8.03
3.13
4.27
3.48
4.40
5.22
5.92
6.60
7.33
3.50
3.63
3.78
4.82
5.59
6.31
7.00
7.76
3.92
4.04
7.02
7.33
6.61
3.40
3.07
5.50
4.58
5.95
6.86
7.76
8.67
9.58
4.48
4.66
4.86
6.22
7.13
8.04
8.95
9.86
3.50
4.94
4.48
5.85
6.76
7.66
8.57
9.48
4.38
4.56
4.76
6.12
7.03
7.94
8.85
9.76
4.65
4.84
8.92
9.02
8.57
3.89
3.79
6.58
1
1
1
1
1
1
1
1
0b
1b
2b
3b
4b
5b
6b
7b
1
1
1
1
1
1
1
1
0c
1c
2c
3c
4c
5c
6c
7c
1
1
1
1
1
1
1
1
0d
1d
2d
3d
4d
5d
6d
7d
1
1
1
7
4e
4f
4g
6
7
7
5
3. Conclusions
In this work, we identified a series of long-chain salicylketoximes as MAGL inhibitors, starting
from two compounds that had been previously designed as estrogen receptor ligands, and also
showed an appreciable inhibition of MAGL activity. The newly developed compounds possess a
2
3

ACCEPTED MANUSCRIPT
long linear alkyl chain in the ketoxime moiety, thus gaining a low micromolar inhibition potency on
MAGL and strongly reducing their affinity to ERα and ERβ. The most potent MAGL inhibitors
herein reported bear 12 or 14 carbon atom chains (compounds 13a and 14a) and molecular
modeling studies confirmed that this chain is located in a lipophilic region of the enzyme active site.
Nevertheless, in spite of their abundant lipophilic portion, these compounds also appear to establish
hydrogen bonds with key residues of the protein. An enzymatic study of their binding mode
revealed that these compounds are not cysteine binders, and they inhibit the enzyme reversibly.
Moreover, the salicylketoximes had only moderate effects on the rest of the endocannabinoid
system, since compound 14a shows a MAGL-selectivity vs. FAAH, ABHDs and CBRs always
greater that six-fold. Finally, these compounds were evaluated for their ability to reduce the
proliferation of cancer cells, and they were particularly effective in both ovarian and breast cancer
cell lines, thus supporting their potential role as anticancer agents.
4
4
. Experimental section
.1. Chemistry
All solvents and chemicals were purchased from Sigma-Aldrich and Alfa Aesar and used without
further purification. Chromatographic separations were performed on silica gel columns by flash
chromatography (Kieselgel 40, 0.040−0.063 mm; Merck). Reactions were followed by thin layer
chromatography (TLC) on Merck aluminum silica gel (60 F254) sheets that were visualized under a
UV lamp. Evaporation was performed in vacuo (rotating evaporator). Sodium sulfate was always
1
13
used as the drying agent. Proton ( H) and carbon ( C) NMR spectra were obtained with a Bruker
1
13
Avance III 400 MHz spectrometer (400 MHz for H NMR and 100 MHz for C NMR) using the
indicated deuterated solvents. Chemical shifts are given in parts per million (ppm) (δ relative to
1
13
1
residual solvent peak for H and C). H NMR spectra are reported in this order: multiplicity and
number of protons. Standard abbreviation indicating the multiplicity were used as follows: s =
2
4

ACCEPTED MANUSCRIPT
singlet, d = doublet, dd = doublet of doublets, ddd = doublet of doublet of doublets, dddd = doublet
of doublet of doublet of doublets, t = triplet, tt = triplet of triplets, quint = quintet, sext = sextet, sept
=
septet, m = multiplet, and bs = broad signal. Compounds were named following IUPAC rules.
HPLC analysis: all target compounds (i.e., assessed in biological assays) were ≥ 95% pure by
HPLC, confirmed via UV detection (λ = 254 nm). Analytical reversed-phase HPLC was conducted
using a Kinetex EVO C18 column (5 µm, 150 mm × 4.6 mm, Phenomenex, Inc.). Compounds 10a-
d, 11a-d, 16a-d, 17a-d and 76-77 were analysed by using the following method “A”: eluent A,
water; eluent B, CH CN; after 5 min at 25% B, a gradient was formed from 25% to 75% of B in 5
3
min and held at 75% of B for 10 min; flow rate was 1 mL/min. Compounds 12a-d, 13a-d, 14a-g
and 15a-d (more lipophilic compounds, due to the long alkyl chain of the ketoxime moiety) were
analysed by using the following method “B”: eluent A, water; eluent B, CH CN; after 4 min at 30%
3
B, a gradient was formed from 30% to 75% of B in 3 min and held at 75% of B for 3 min; then a
gradient was formed from 75% to 95% of B in 3 min and held at 95% of B for 12 min; flow rate
was 1 mL/min. Yields refer to isolated and purified products derived from nonoptimized
procedures. CAY10499 (2) was purchased from Cayman Chemical and (4-(4-
chlorobenzoyl)piperidin-1-yl)(3-hydroxyphenyl)methanone 6 was synthesized as previously
reported.[17] Elemental analysis was used to further characterize the final compounds. Analytical
results are within ± 0.4 % of the theoretical values.
4
.1.1. General procedure for the synthesis of compounds 24-29, 60-61 and 71.
Thionyl chloride (10 eq) was added dropwise at 0 °C to the appropriate aliphatic carboxylic acid
8-23 (1.0 g, 1 eq). After addition was complete, the ice bath was removed, and the reaction
1
mixture was stirred at 95 °C for 3 h under argon atmosphere. The mixture was cooled to room
temperature and then evaporated to give the corresponding acid chloride that was immediately used
in the next step without further purification. Aromatic acid chlorides 58, 59 and 70 were
commercially available and were used as purchased without further purification. 4-Bromoanisole
(0.87 mL, 1 eq) was dissolved in anhydrous 1,2-dichloroethane (3.2 mL), cooled in an ice bath and
2
5

ACCEPTED MANUSCRIPT
then aluminium trichloride (1.4 eq) was added. Finally, a previously prepared solution of the acid
chloride (1 eq) in anhydrous 1,2-dichloroethane (5.3 mL) was added and the resulting mixture was
heated at 65 °C for 2 h. The reaction mixture was allowed to cool to room temperature, poured
carefully in ice and then repeatedly extracted with EtOAc. The combined organic phase was washed
with brine and then dried over Na SO , filtered and concentrated under reduced pressure to obtain a
2
4
residue that was purified by flash chromatography to afford the title compounds.
1-(5-Bromo-2-hydroxyphenyl)pentan-1-one (24). Yellow oil, yield: 42% from valeric acid 18
1
(
eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 0.97 (t, 3H, J = 7.3 Hz), 1.43 (sext, 2H, J =
3
7
1
1
.4 Hz), 1.73 (quint, 2H, J = 7.5 Hz), 2.97 (t, 2H, J = 7.4 Hz), 6.90 (d, 1H, J = 8.9 Hz), 7.53 (dd,
H, J = 8.9, 2.4 Hz), 7.86 (d, 1H, J = 2.4 Hz), 12.31 (exchangeable s, 1H).
-(5-Bromo-2-hydroxyphenyl)octan-1-one (25). Yellow solid, yield: 62% from ottanoic acid 19
1
(
eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 0.89 (t, 3H, J = 7.2 Hz), 1.20-1.48 (m, 8H),
3
1
2
1
.74 (quint, 2H, J = 7.3 Hz), 2.96 (t, 2H, J = 7.4 Hz), 6.89 (d, 1H, J = 8.9 Hz), 7.53 (dd, 1H, J = 8.9,
.4 Hz), 7.86 (d, 1H, J = 2.4 Hz), 12.31 (exchangeable s, 1H).
-(5-bromo-2-hydroxyphenyl)decan-1-one (26). White solid, yield: 45% from decanoic acid 20
1
(
eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 0.88 (t, 3H, J = 6.9 Hz), 1.20-1.44 (m, 12H),
3
1
2
1
2
1
.73 (quint, 2H, J = 7.4 Hz), 2.95 (t, 2H, J = 7.4 Hz), 6.89 (d, 1H, J = 8.9 Hz), 7.53 (dd, 1H, J = 8.9,
.5 Hz), 7.86 (d, 1H, J = 2.5 Hz), 12.31 (exchangeable s, 1H).
-(5-Bromo-2-hydroxyphenyl)dodecan-1-one (27). White solid, yield: 34% from dodecanoic acid
1
1 (eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 0.88 (t, 3H, J = 6.9 Hz), 1.19-1.45 (m,
3
6H), 1.73 (quint, 2H, J = 7.4 Hz), 2.95 (t, 2H, J = 7.4 Hz), 6.89 (d, 1H, J = 8.9 Hz), 7.53 (dd, 1H, J
=
8.9, 2.4 Hz), 7.86 (d, 1H, J = 2.4 Hz), 12.31 (exchangeable s, 1H).
1-(5-Bromo-2-hydroxyphenyl)tetradecan-1-one (28). White solid, yield: 52% from myristic acid 22
1
(
eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 0.88 (t, 3H, J = 6.8 Hz), 1.20-1.44 (m, 20H),
3
1
.73 (quint, 2H, J = 7.4 Hz), 2.95 (t, 2H, J = 7.4 Hz), 6.89 (d, 1H, J = 8.9 Hz), 7.53 (dd, 1H, J = 8.9,
.4 Hz), 7.86 (d, 1H, J = 2.4 Hz), 12.30 (exchangeable s, 1H).
2
2
6

ACCEPTED MANUSCRIPT
1-(5-Bromo-2-hydroxyphenyl)hexadecan-1-one (29). White solid, yield: 39% from palmitic acid 23
1
(
eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 0.88 (t, 3H, J = 6.7 Hz), 1.20-1.44 (m, 24H),
3
1
.73 (quint, 2H, J = 7.3 Hz), 2.95 (t, 2H, J = 7.4 Hz), 6.89 (d, 1H, J = 8.9 Hz), 7.53 (dd, 1H, J = 8.9,
.4 Hz), 7.86 (d, 1H, J = 2.5 Hz), 12.31 (exchangeable s, 1H).
2
(5-Bromo-2-hydroxyphenyl)(phenyl)methanone (60). White solid, yield: 60% from benzoyl
1
chloride 58 (eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 6.99 (d, 1H, J = 9.2 Hz), 7.49-
3
7
1
.72 (m, 7H), 11.92 (exchangeable s, 1H).
-(5-Bromo-2-hydroxyphenyl)-2-phenylethanone (61). White solid, yield: 71% from phenacetyl
1
chloride 59 (eluent: petroleum ether/EtOAc 99:1). H-NMR (CDCl ) δ (ppm): 4.28 (s, 2H), 6.90 (d,
3
1H, J = 8.9 Hz), 7.23-7.40 (m, 5H), 7.55 (dd, 1H, J = 8.9, 2.4 Hz), 7.97 (d, 1H, J = 2.4 Hz), 12.11
(exchangeable s, 1H).
1-(5-Bromo-2-hydroxyphenyl)-2-methylpropan-1-one (71). Light yellow oil, yield: 36% from
1
isobutyryl chloride 70 (eluent: petroleum ether). H-NMR (CDCl ) δ (ppm): 1.25 (d, 6H, J = 6.8
3
Hz), 3.54 (sept, 1H, J = 6.8 Hz), 6.91 (d, 1H, J = 8.9 Hz), 7.54 (dd, 1H, J = 8.9, 2.4 Hz), 7.88 (d,
1H, J = 2.5 Hz), 12.42 (exchangeable s, 1H).
4.1.2. General procedure for the synthesis of compounds 30-44, 62-65 and 72-73.
A solution of Pd(OAc) (0.03 eq) and triphenylphosphine (0.12 eq) in ethanol (0.9 mL/0.90 mmol
2
bromo-derivative) and toluene (0.9 mL/0.90 mmol bromo-derivative) was stirred at room
temperature under argon for 10 minutes. After that period, bromo-substituted derivative 24-29, 60-
6
1, 71 (320 mg, 1 eq), 2M aqueous Na CO (0.9 mL/0.90 mmol bromo-derivative) and the
2 3
appropriate substituted phenylboronic acid (1.3 eq) were sequentially added. The resulting mixture
was heated at 100 °C in a sealed vial under nitrogen for 24 h. After being cooled to room
temperature, the mixture was diluted with water and extracted with EtOAc. The combined organic
phase was dried and concentrated. The crude product was purified by flash chromatography using
the indicated eluent and pure fractions containing the desired compound were evaporated to dryness
affording the desired product.
2
7

ACCEPTED MANUSCRIPT
1-(3'-Fluoro-4-hydroxy-4'-methoxy-[1,1'-biphenyl]-3-yl)pentan-1-one (30). Yellow solid, yield:
1
6
2% from 24 and 3-fluoro-4-methoxyphenylboronic acid (eluent: n-hexane/EtOAc 98:2). H-NMR
(
CDCl ) δ (ppm): 0.98 (t, 3H, J = 7.3 Hz), 1.45 (sext, 2H, J = 7.5 Hz), 1.76 (quint, 2H, J = 7.5 Hz),
3
3
8
1
.06 (t, 2H, J = 7.4 Hz), 3.94 (s, 3H), 7.05 (d, 1H, J = 8.7 Hz), 7.21-7.29 (m, 3H), 7.63 (dd, 1H, J =
.6, 2.3 Hz), 7.87 (d, 1H, J = 2.3 Hz), 12.40 (exchangeable s, 1H).
-(4-Hydroxy-4'-methoxy-[1,1'-biphenyl]-3-yl)pentan-1-one (31). White solid, yield: 69% from 24
1
and 4-methoxyphenylboronic acid (eluent: petroleum ether/EtOAc 98:2). H-NMR (CDCl ) δ
3
(ppm): 0.98 (t, 3H, J = 7.4 Hz), 1.44 (sext, 2H, J = 7.5 Hz), 1.76 (quint, 2H, J = 7.5 Hz), 3.05 (t,
2
8
H, J = 7.4 Hz), 3.86 (s, 3H), 6.99 (AA’XX’, 2H, JAX = 8.8 Hz, JAA’/XX’ = 2.6 Hz), 7.04 (d, 1H, J =
.6 Hz), 7.46 (AA’XX’, 2H, JAX = 8.9 Hz, JAA’/XX’ = 2.7 Hz), 7.65 (dd, 1H, J = 8.6, 2.3 Hz), 7.89
(d, 1H, J = 2.3 Hz), 12.37 (exchangeable s, 1H).
1-(3'-Fluoro-4-hydroxy-4'-methoxy-[1,1'-biphenyl]-3-yl)octan-1-one (32). Light-yellow solid, yield:
1
6
1% from 25 and 3-fluoro-4-methoxyphenylboronic acid (eluent: n-hexane/EtOAc 98:2). H-NMR
(
CDCl ) δ (ppm): 0.89 (t, 3H, J = 6.9 Hz), 1.20-1.50 (m, 8H), 1.77 (quint, 2H, J = 7.4 Hz), 3.05 (t,
3
2
7
1
2
0
3
H, J = 7.5 Hz), 3.94 (s, 3H), 7.00-7.07 (m, 2H), 7.21-7.30 (m, 2H), 7.63 (dd, 1H, J = 8.6, 2.3 Hz),
.86 (d, 1H, J = 2.3 Hz), 12.40 (exchangeable s, 1H).
-(4-Hydroxy-4'-methoxy-[1,1'-biphenyl]-3-yl)octan-1-one (33). Off-white solid, yield: 46% from
1
5 and 4-methoxyphenylboronic acid (eluent: n-hexane/EtOAc 98:2). H-NMR (CDCl ) δ (ppm):
3
.89 (t, 3H, J = 6.9 Hz), 1.20-1.45 (m, 8H), 1.77 (quint, 2H, J = 7.5 Hz), 3.04 (t, 2H, J = 7.4 Hz),
.86 (s, 3H), 6.99 (AA’XX’, 2H, JAX = 8.8 Hz, JAA’/XX’ = 2.6 Hz), 7.04 (d, 1H, J = 8.6 Hz), 7.46
(AA’XX’, 2H, JAX = 8.8 Hz, JAA’/XX’ = 2.6 Hz), 7.65 (dd, 1H, J = 8.6, 2.3 Hz), 7.88 (d, 1H, J = 2.3
Hz), 12.37 (exchangeable s, 1H).
1-(3'-Fluoro-4-hydroxy-4'-methoxy-[1,1'-biphenyl]-3-yl)decan-1-one (34). Off-white solid, yield:
1
7
2% from 26 and 3-fluoro-4-methoxyphenylboronic acid (eluent: n-hexane/EtOAc 98:2). H-NMR
(
CDCl ) δ (ppm): 0.88 (t, 3H, J = 6.9 Hz), 1.19-1.46 (m, 12H), 1.77 (quint, 2H, J = 7.4 Hz), 3.05 (t,
3
2
8