Optimization of a Benzoylpiperidine Class Identifies a Highly Potent and Selective Reversible Monoacylglycerol Lipase (MAGL) Inhibitor

Article abstract of DOI:10.1021/acs.jmedchem.8b01483

Monoacylglycerol lipase (MAGL) is the enzyme degrading the endocannabinoid 2-arachidonoylglycerol, and it is involved in several physiological and pathological processes. The therapeutic potential of MAGL is linked to several diseases, including cancer. The development of MAGL inhibitors has been greatly limited by the side effects associated with the prolonged MAGL inactivation. Importantly, it could be preferable to use reversible MAGL inhibitors in vivo, but nowadays only few reversible compounds have been developed. In the present study, structural optimization of a previously developed class of MAGL inhibitors led to the identification of compound 23, which proved to be a very potent reversible MAGL inhibitor (IC₅₀ = 80 nM), selective for MAGL over the other main components of the endocannabinoid system, endowed of a promising antiproliferative activity in a series of cancer cell lines and able to block MAGL both in cell-based as well as in vivo assays.

Full text of DOI:10.1021/acs.jmedchem.8b01483

Subscriber access provided by Iowa State University | Library
Article
Optimization of a benzoylpiperidine class identifies a highly potent
and selective reversible monoacylglycerol lipase (MAGL) inhibitor
Carlotta Granchi, Margherita Lapillo, Sandra Glasmacher, Giulia Bononi, Cristina
Licari, Giulio Poli, Maguie el Boustani, Isabella Caligiuri, Flavio Rizzolio, Jürg
Gertsch, Marco Macchia, Filippo Minutolo, Tiziano Tuccinardi, and Andrea Chicca
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01483 • Publication Date (Web): 04 Feb 2019
Downloaded from http://pubs.acs.org on February 5, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Optimization of a benzoylpiperidine class identifies a
highly potent and selective reversible
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
monoacylglycerol lipase (MAGL) inhibitor
a
a
b
a
a
Carlotta Granchi , Margherita Lapillo , Sandra Glasmacher , Giulia Bononi , Cristina Licari ,
a
c,d
c
c,e
b
Giulio Poli , Maguie el Boustani , Isabella Caligiuri , Flavio Rizzolio , Jürg Gertsch , Marco
a
a
a,*
b
Macchia , Filippo Minutolo , Tiziano Tuccinardi , Andrea Chicca
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, 33081 Aviano (PN), Italy.
d
Doctoral School in Molecular Biomedicine, University of Trieste, 34100 Trieste, Italy.
e
Department of Molecular Sciences and Nanosystems, Ca' Foscari University, 30123 Venezia,
Italy.
*
Corresponding author.
E-mail address: tiziano.tuccinardi@unipi.it
ABSTRACT
Monoacylglycerol lipase (MAGL) is the enzyme degrading the endocannabinoid 2-
arachidonoylglycerol and it is involved in several physiological and pathological processes. The
therapeutic potential of MAGL is linked to several diseases, including cancer. The development of
MAGL inhibitors has been greatly limited by the side effects associated with the prolonged MAGL
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inactivation. Importantly, it could be preferable to use reversible MAGL inhibitors in vivo, but
nowadays only few reversible compounds have been developed. In the present study, structural
optimization of a previously developed class of MAGL inhibitors led to the identification of
compound 23, which proved to be a very potent reversible MAGL inhibitor (IC = 80 nM), selective
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
for MAGL over the other main components of the endocannabinoid system, endowed of a promising
antiproliferative activity in a series of cancer cell lines and able to block MAGL both in cell-based as
well as in vivo assays.
INTRODUCTION
The endocannabinoids system (ECS) is an endogenous system involved in many physiological and
pathological processes. ECS is composed of two seven-transmembrane G protein-coupled receptors
named cannabinoid receptors type 1 and type 2 (CB1, CB2),1,2 a class of lipid signaling molecules
called endocannabinoids (eCBs) and several biosynthetic and degrading enzymes that are involved
in the production and transformation of these molecules. CB1 receptors are mainly expressed in the
brain where they regulate the release of neurotransmitters from pre-synaptic neurons, whereas CB2
receptors are principally expressed in immune cells. Anandamide (AEA) and 2-arachidonoylglycerol
3
(2-AG) are the two most abundant and well-studied eCBs, which are biosynthesized on-demand
from phospholipid precursors in the plasma membrane and released into the extracellular milieu.4
After activating cannabinoid receptors, eCBs are transported into the cytoplasm via facilitated
5
,6
diffusion mediated by a putative endocannabinoid membrane transporter and degraded by specific
enzymes. AEA is hydrolyzed by fatty acid amide hydrolase (FAAH) in arachidonic acid and
ethanolamine, whereas 2-AG is hydrolyzed into arachidonic acid and glycerol by monoacylglycerol
lipase (MAGL) and α/β hydrolase-6 and -12 (ABHD6 and ABHD12). MAGL is the main degrading
enzyme for 2-AG and it is responsible for approximately 85% of 2-AG hydrolysis in the brain, with
7
,8
a minor contribution of ABHD6 and ABHD12. Several studies indicate the therapeutic potential of
2
ACS Paragon Plus Environment

Page 3 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
selective FAAH and MAGL inhibitors in several disease models of inflammation, pain, anxiety and
other neuroinflammatory diseases.9–16 The inhibition of eCBs degradation represents a promising
pharmacological strategy to activate the ECS limiting the potential side effects associated with direct
receptor agonists.7,17,18
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9
20
Many MAGL inhibitors were published and patented and some representative compounds are
reported in Figure 1; however, most MAGL inhibitors reported to date in the literature are
characterized by an irreversible binding mode. The irreversible mechanism of action shown by these
inhibitors hampered their subsequent clinical development, since chronic administration of
irreversible inhibitors, as well as genetic deletion of MAGL, provokes many negative effects in vivo.
However, there are some recent noteworthy examples of irreversible inhibitors, endowed with a good
pharmacokinetic profile and highly effective in reducing inflammation in vivo.21,22 In some
experimental studies reported in literature, sustained inactivation of MAGL by irreversible inhibition
led to a loss of the therapeutic effects and a cross-tolerance to CB1 agonists in mice, accompanied by
physical dependence. Moreover, prolonged MAGL inhibition provokes marked changes in some
brain regions, such as CB1 receptor downregulation and desensitization, together with a reduction of
the endocannabinoid-mediated synaptic plasticity. These effects are due to chronic MAGL
inactivation in the central nervous system, provoking the maintenance of elevated 2-AG levels, which
continuously activate and thus desensitize CB1 receptors in the brain, since this enzyme has a great
2
3
impact on the brain ECS. These just mentioned effects were observed in vivo by administration of
irreversible inhibitors such as compound 1 (JZL184, 4-nitrophenyl-4-[bis(1,3-benzodioxol-5-
yl)(hydroxy)methyl]piperidine-1-carboxylate, Figure 1), which is a very potent carbamate-based
MAGL inhibitor.24 Moreover, compound 2 (CAY10499, benzyl(4-(5-methoxy-2-oxo-1,3,4-
oxadiazol-3(2H)-yl)-2-methylphenyl)carbamate, Figure 1) is a widely known irreversible MAGL
2
5
inhibitor, used in many experimental studies concerning MAGL inhibition. Despite the variety of
irreversible MAGL inhibitors, at present their use is mainly limited as experimental tools in many
studies. To the best of our knowledge, only a few reversible MAGL inhibitors are known and some
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
examples are reported in Figure 1. For example, natural terpenoids Euphol (compound 3, Figure 1)
and Pristimerin (compound 4, Figure 1) were found to be potent MAGL inhibitors in vitro (IC values
5
0
in the nanomolar range), although they were not selective, acting also on other targets different from
MAGL.26–28 Natural triterpenoid β-amyrin 5 (Figure 1) is a MAGL inhibitor structurally related to,
although less potent than, Pristimerin and Euphol, but it also inhibits other enzymes that hydrolyse 2-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
9
AG, such as ABHDs. In 2014, benzo[d][1,3]dioxol-5-ylmethyl 6-phenylhexanoate 6 (Figure 1) was
reported by Hernández-Torres et al. as a potent, reversible and selective MAGL inhibitor. This
compound was able to alleviate the clinical progression and the symptoms of a multiple sclerosis
mouse model, without inducing catalepsy or other motor impairments, which were instead previously
observed after the administration of irreversible MAGL inhibitors.30 More recent examples of
reversible MAGL inhibitors include derivative 7 (JZP-361, Figure 1), a nanomolar MAGL inhibitor
that was characterized by a dual action, since it showed an anti-histaminergic activity, due to its high
3
1
structural similarity to Loratadine, a histamine H receptor antagonist. Very recently, reversible
1
MAGL inhibitor 1,5-diphenylpyrazole-3-carboxamide 8 (Figure 1) was reported to exert
antiproliferative effects in cancer cell lines and to relieve the neuropathic hypersensitivity induced in
3
2
vivo by oxaliplatin. Moreover, compound 9 (Figure 2) is a benzoylpiperidine-based reversible
MAGL inhibitor developed by our group in 2016 and it has been used as the starting point for the
chemical optimization of the class of compounds reported in this paper.
4
ACS Paragon Plus Environment

Page 5 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Irreversible inhibitors
O
O
N
O N
2
H
O
N
OH
O
N
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
O
O
O
O
O
2
O
H
O
(
CAY10499)
1
(JZL184)
Reversible inhibitors
Natural
Synthetic
O
O
O
5
O
6
O
HO
H
Cl
3
O
(Euphol)
N
N
H
O
N
HO
O
N
4
(
Pristimerin)
N
7
O
N
N
H
N
H
HO
HO
H
5
(
-amyrin)
8
Figure 1. Structures of some of the most important known MAGL inhibitors.
RESULTS AND DISCUSSION
Design. In 2016, our group published the discovery of a chemical class of reversible MAGL
inhibitors, based on a benzoylpiperidine scaffold, and within this class compound 9 (Figure 2) was
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
identified as the most potent inhibitor, showing an IC value of 0.84 µM, thus representing one of
5
0
3
3
the first example of potent and selective reversible MAGL inhibitor. The significant improvement
of the inhibitory potency of this compound with respect to the other synthesized compounds is
supposed to be due to a strategic hydrogen bond network established between the meta-hydroxyl
group on the amidic phenyl ring of compound 9, which replaces a structural water molecule observed
in many MAGL X-ray structures, and a couple of active site residues (E53 and H272). Starting from
this consideration, we were intrigued by the possibility to increase the acidity of the phenolic hydroxyl
group of compound 9, in order to strengthen the established hydrogen bonds and thus potentially lead
to an increase in the inhibitory activity. At the same time, we took into account that this part of the
molecule is located into a small pocket of the protein, where some polar residues are present, as
suggested by modeling studies. Therefore, the insertion of bulky groups in the amidic phenyl ring is
discouraged in order to avoid steric clashes and maintain the same binding disposition observed for
compound 9. On these bases, we started to modify compound 9 by adding halogen atoms (iodine,
bromine, chlorine and fluorine atoms) in all the possible substitution positions of the amidic phenyl
ring, due to their electron withdrawing properties and at the same time their different dimensions
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 2, panel A). This modification gave rise to compounds 10a-d, 11a-d, 12a-d and 13a-d (Figure
2, panel B) where iodine, bromine, chlorine and fluorine were inserted in the two possible ortho
positions, in the meta or in the para positions with respect the phenolic hydroxyl group, still
maintaining the 4-chlorobenzoylpiperidine portion on the other side of the molecule fixed. Regarding
this last portion of compound 9, molecular modeling studies suggested that the benzoyl moiety is
directed toward the open cavity of the protein, pointing at the lipophilic channel of the enzyme (Figure
2
, panel A). This consideration was confirmed by the good inhibition activity shown by a previously
obtained analogue of 9, where the p-Cl atom in the benzoyl ring was replaced by a second phenyl
3
3
ring, which showed a 2-fold increase of inhibitory activity. As an additional structural investigation,
in the present work we have explored the possibility to fill the wide open enzymatic cavity, where the
benzoyl part of compound 9 is located, with suitable groups. This purpose was achieved by removing
6
ACS Paragon Plus Environment

Page 7 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the chlorine atom from the benzoyl moiety (compound 14, Figure 2, panel B) or replacing it with
alkyl groups of different dimensions, such as methyl, ethyl, n-propyl, i-proyl or n-butyl (compounds
15-19, Figure 2, panel B). Then, in order to verify whether this region preferentially hosts lipophilic
groups, the p-Cl-phenyl was also replaced by more polar fragments, such as a phenolic ring, a
benzodioxane and a 4-phenyl-morpholine moiety (compounds 20-22, Figure 2, panel B).
Finally, we investigated the combinations of chemical modifications on both sides of the scaffold,
such as: 1) the presence of a fluorine atom in para position to the phenolic hydroxyl group or to the
amide carbonyl moiety, as in compounds 11d and 13d, respectively; 2) the presence of a i-propyl or
n-butyl chain on the benzoyl ring, as in compounds 18 and 19, respectively. The resulting compounds
23-26 (Figure 3) maintained the same central benzoylpiperidine scaffold and derived from the
combination of different substituents in the two terminal portions of the parent compound 9 and they
were synthesized accordingly.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
small polar pocket
open hydrophobic cavity
O
Cl
N
X
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OH
X = halogens
O
9
OH
R
R = H, alkyl chains,
polar groups
A
X
X
O
Cl
N
X
X
X
OH
OH
OH
12a: X = I;
OH
O
OH
10a: X = I;
11a: X = I;
13a: X = I;
1
1
1
0b: X = Br; 11b: X = Br; 12b: X = Br; 13b: X = Br;
0c: X = Cl; 11c: X = Cl; 12c: X = Cl; 13c: X = Cl;
0d: X = F. 11d: X = F. 12d: X = F.
10a-d, 11a-d, 12a-d, 13a-d
13d: X = F.
1
4:
19:
3
15:
20:
21:
HO
O
N
O
O
Ar
16:
17:
O
OH
14-22
2
22:
O
N
18:
B
Figure 2. Design of new benzoylpiperidine derivatives. Upper panel (A): previously published
starting compound 9 and its schematic location in the MAGL binding site: planned modifications of
8
ACS Paragon Plus Environment

Page 9 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the benzoyl part (in red) and the amidic moiety (in blue). Lower panel (B): newly synthesized
compounds, modified in the amidic (10a-d, 11a-d, 12a-d, 13a-d) and in the benzoyl (14-22) portions.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
R
N
X
O
OH
O
O
F
O
O
N
N
N
F
F
O
O
OH
F
O
O
OH
OH
2
3
4
25
3
3
N
OH
2
26
Figure 3. Design of new benzoylpiperidine derivatives: compounds 23-26 resulting from the
combination of the best structural motifs present in compounds 10-22.
Chemistry. For the synthesis of halogenated compounds, in most cases the starting halogenated
carboxylic acids were commercially available and, therefore, they were directly submitted to the
amidic condensation with the appropriate amine, as shown in Scheme 1. Basically, halogenated 3-
methoxy-benzoic acids 27a, 28 and 29a-f were reacted with commercially available 4-(4-
chlorobenzoyl)piperidine 30 in the presence of 1-[bis(dimethylamino)methylene]-1H-1,2,3-
triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HATU) as the condensing agent, DIPEA as
the base and dry N,N-dimethylformamide as the solvent, as previously reported,33,34 to obtain the
corresponding amidic derivatives 31, 32 and 33a-f (Scheme 1). Finally, the methoxy groups of these
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
intermediates were deprotected with boron tribromide to yield the final hydroxy-substituted
compounds 10c, 11b, 12b-d and 13b-d. Just one case differs from the commonly adopted synthetic
strategy, since we directly used the halogenated benzoic acid bearing the free phenolic group, in order
to skip the last deprotection step. 2-Fluoro-3-hydroxybenzoic acid 27b and piperidine derivative 30
straightforwardly furnished the final product 10d (Scheme 1). However, the purification of this
compound from the impurities formed in the reaction was not straightforward, finally resulting in a
low reaction yield (23%). We believe that this problem was mainly due to the presence of a free
phenolic OH group in precursor 27b. Therefore, we slightly modified the synthetic strategy for the
preparation of the other compounds of this series, by generally using the appropriate methoxylated
benzoic acid in the amidic condensation reaction.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
HOOC
Cl
Cl
NH
N
a
R1
R1
OR2
O
OR2
O
O
Cl
27a-b
30
31, 10d
N
b
1
2
1
2
a: R = Cl, R = CH ;
31: R = Cl, R = CH ;
3
3
1
2
1
2
Cl
b: R = F, R = H.
10d: R = F, R = H.
O
OH
10c
Br
O
Br
O
Br
HOOC
Cl
Cl
Cl
Cl
Cl
Cl
NH
NH
N
N
N
a
a
b
b
OMe
28
O
O
O
OMe
O
OH
30
32
11b
O
O
HOOC
R1
R2
R1
R2
R1
R2
N
OMe
9a-f
O
OMe
O
OH
2
30
33a-f
12b-d, 13b-d
1
2
1
2
1
2
a: R = Br, R = H;
a: R = Br, R = H;
12b: R = Br, R = H;
1
2
1
2
1
2
b: R = Cl, R = H;
b: R = Cl, R = H;
12c: R = Cl, R = H;
1
2
1
2
1
2
c: R = F, R = H;
c: R = F, R = H;
12d: R = F, R = H;
1
2
1
2
1
2
d: R = H, R = Br;
d: R = H, R = Br;
13b: R = H, R = Br;
1
2
1
2
1
2
e: R = H, R = Cl;
e: R = H, R = Cl;
13c: R = H, R = Cl;
1
2
1
2
1
2
f: R = H, R = F.
f: R = H, R = F.
13d: R = H, R = F.
Scheme 1. Synthesis of halogenated derivatives 10c-d, 11b, 12b-d, 13b-d.
1
0
ACS Paragon Plus Environment

Page 11 of 83
Journal of Medicinal Chemistry
1
2
3
Reagents and conditions: (a) HATU, DIPEA, dry DMF, RT, 2-6.5 h [23-77%]; (b) 1M BBr , dry
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CH Cl , −10 to 0 °C, then RT, 1.5-4 h [40-81%].
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In details, considering the just mentioned problems encountered with the amide formation starting
from a hydroxy-substituted benzoic acid, and that in some cases benzoic acids or methyl benzoates
bearing free phenolic hydroxyl groups were the only commercially available starting materials, we
methylated the phenolic derivatives 34a-c with methyl iodide by using potassium carbonate in DMF
(Scheme 2). Under these conditions, both the carboxylic acid (if present, such as in compounds 34a,b)
and the hydroxyl groups were methylated and consequently, methoxylated methyl benzoates 35a-c
were hydrolyzed under aqueous basic conditions in order to obtain the free acids 36a-c. At this point,
the previously adopted reaction conditions were followed, which consisted in the amide condensation
with compound 30 and the subsequent BBr -promoted deprotection of intermediates 37a-c to obtain
3
final compounds 11c-d and 13a (Scheme 2).
Cl
NH
R2
R2
R2
R2
O
30
c
O
COOR1
COOMe
COOH
Cl
N
a
b
R3
R3
R3
R3
OH
OMe
OMe
O
OMe
34a-c
35a-c
36a-c
37a-c
1
2
3
2
3
2
3
2
3
a: R = H, R = Cl, R = H;
a: R = Cl, R = H;
a: R = Cl, R = H;
a: R = Cl, R = H;
1
2
3
2
3
2
3
2
3
b: R = H, R = F, R = H;
b: R = F, R = H;
b: R = F, R = H;
b: R = F, R = H;
1
2
3
2
3
2
3
2
3
c: R = CH , R = H, R = I.
c: R = H, R = I.
c: R = H, R = I.
c: R = H, R = I.
3
d
O
R2
Cl
N
R3
O
OH
11c-d, 13a
2
3
11c: R = Cl, R = H;
2
3
11d: R = F, R = H;
2
3
13a: R = H, R = I.
Scheme 2. Synthesis of halogenated derivatives 11c-d, 13a.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Reagents and conditions: (a) MeI, K CO , DMF, RT, 24 h [79-99%]; (b) aq. 2N LiOH, THF/MeOH
2
3
1
:1 v/v, RT, overnight [95-99%]; (c) HATU, DIPEA, dry DMF, RT, 3-5 h [39-75%]; (d) 1M BBr3,
dry CH Cl , −10 to 0 °C, then RT, 1.5-7 h [33-88%].
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
An appropriate aniline derivative was used as the starting material instead of the halogenated
precursor, when the latter was not commercially available. Hence, amino-substituted benzoic acids
38a-b (Scheme 3) were converted to the corresponding methyl esters 39a-b by refluxing them with
thionyl chloride in methanol. Compounds 39a-b and commercially available methyl-2-amino-5-
methoxybenzoate 40 were reacted in a Sandmeyer reaction with the appropriate source of nucleophile
(copper (I) bromide for compound 41b or potassium iodide for compounds 41a, c, d) to obtain
halogenated compounds 41a-d. Methyl esters 41a-d were hydrolyzed to benzoic acids 42a-d, which
were then condensed with 4-(4-chlorobenzoyl)piperidine 30 to get amides 43a-d. The final O-
demethylation reaction furnished the last groups of halogenated benzoylpiperidine derivatives 10a-
b, 11a and 12a (Scheme 3).
1
2
ACS Paragon Plus Environment

Page 13 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NH2
COOMe
OMe
40
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
R3
R3
R2
COOH
R1
R2
COOMe
R1
R2
COOMe
R1
R2
COOH
R1
a
b
c
OMe
8a-b
OMe
OMe
OMe
3
39a-b
41a-d
42a-d
1
2
1
2
1
2
3
1
2
3
a: R = NH , R = H;
a: R = NH , R = H;
a: R = I, R = H, R = H;
a: R = I, R = H, R = H;
2
2
1
2
1
2
1
2
3
1
2
3
b: R = H, R = NH .
b: R = H, R = NH .
b: R = Br, R = H, R = H;
b: R = Br, R = H, R = H;
2
2
1
2
3
1
2
3
c: R = H, R = H, R = I;
c: R = H, R = H, R = I;
Cl
1
2
3
1
2
3
d: R = H, R = I, R = H.
d: R = H, R = I, R = H.
NH
O
O
R3
O
R3
30
Cl
R2
Cl
R2
N
N
d
e
R1
R1
O
OMe
O
OH
4
3a-d
2
10a-b, 11a, 12a
1
3
1
2
3
a: R = I, R = H, R = H;
10a: R = I, R = H, R = H;
1
2
3
1
2
3
b: R = Br, R = H, R = H;
10b: R = Br, R = H, R = H;
1
2
3
1
2
3
c: R = H, R = H, R = I;
11a: R = H, R = H, R = I;
1
2
3
1
2
3
d: R = H, R = I, R = H.
12a: R = H, R = I, R = H.
Scheme 3. Synthesis of halogenated derivatives 10a-b, 11a, 12a.
Reagents and conditions: (a) SOCl , MeOH, reflux, 3-24 h [39a: 69%, 39b: 97%]; (b) if R = I (41a,
2
c, d): KI, NaNO , H SO , H O, - 5 °C to RT, 1-3 h; if R = Br (41b): CuBr, HBr 48%, NaNO , H O,
2
2
4
2
2
2
dioxane, - 5 °C to RT, then 110 °C, 5 h [25-78%]; (c) aq. 2N LiOH, THF/MeOH 1:1 v/v, RT,
overnight [64-99%]; (d) HATU, DIPEA, dry DMF, RT, 3-5 h [43-60%]; (e) 1M BBr , dry CH Cl ,
3
2
2
−10 to 0 °C, then RT, 1.5-4.5 h [24-94%].
Compounds 14-21 bearing different aromatic groups in the place of the para-chlorobenzoyl ring of
the previous series of derivatives were obtained according to the synthetic pathway outlined in
Scheme 4. Commercially available isonipecotic acid 44 was acetylated on the piperidine nitrogen
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
atom by using acetic anhydride in pyridine to obtain compound 45, which was then converted to the
corresponding acyl chloride by refluxing it with thionyl chloride in dichloroethane, and then reacted
in a Friedel-Crafts reaction with the appropriate aromatic moiety, in the presence of aluminum
chloride in dichloroethane. The aromatic systems utilized in this step were either a simple benzene
ring (46a) or mono-substituted aromatic rings with alkyl chains of increasing length, such as toluene
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(46b), ethylbenzene (46c), n-propylbenzene (46d), cumene (46e) or n-butylbenzene (46f). More polar
groups were also inserted, such as an anisole (46g) and a benzo-1,4-dioxane moiety (46h). After the
different functionalization of the piperidine portion, hydrolysis under aqueous basic conditions and
heating removed the acetyl group, thus making the piperidine nitrogen atom free to react with 3-
methoxybenzoic acid 48 to get the corresponding amides 49a-h. The final demethylation step
afforded compounds 14-21; in particular, it should be noticed that the two antipodal methoxy groups
of compound 20, one in meta position of the amidic phenyl ring and the other in para position of the
benzoyl ring, were both converted to free hydroxyls.
1
4
ACS Paragon Plus Environment

Page 15 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
O
NH
N
NH
a
b
c
N
Ar
Ar
HOOC
HOOC
HOOC
O
O
4
4
45
46a-h
47a-h
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
O
OMe
Ar
4
d
8
N
9a-h
3
N
e
Ar
O
OMe
O
OH
4
14-21
Ar
f:
a:
b:
g:
RO
c:
4
6g, 47g, 49g: R = CH3
2
0: R = H
d:
e:
O
O
2
h:
Scheme 4. Synthesis of aryl-substituted derivatives 14-21.
Reagents and conditions: (a) Ac O, pyridine, 140 °C, 2 h [85%]; (b) i. SOCl , dry 1,2-DCE, 60 °C,
2
2
4 h; ii. appropriate aromatic system, AlCl , dry 1,2-DCE, 90 °C, overnight [43-76%]; (c) NaOH 1N,
3
EtOH, 90 °C, overnight [66-95%]; (d) HATU, DIPEA, dry DMF, RT, 2-3 h [33-97%]; (e) 1M BBr3,
dry CH Cl , −10 to 0 °C, then RT, 1-2.5 h [43-76%].
2
2
Slight synthetic modifications were used for the preparation of morpholine-substituted derivative 22
Scheme 5), when compared to the procedure used for the synthesis of compounds 14-21. The
(
piperidine nitrogen atom of isonipecotic acid 44 was protected by benzoylation and then its acyl
chloride was reacted with bromobenzene in a Friedel-Crafts reaction to obtain compound 53. This
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bromo-derivative was subjected to a C-N coupling reaction with morpholine in the presence of
tris(dibenzylideneacetone)dipalladium as the catalyst and XPhos as the ligand, by using potassium
phosphate as the base under heating conditions, in order to replace the bromine atom with the
morpholine ring. The protection of the piperidine nitrogen with a benzoyl group instead of the
previous acetyl moiety was due to the harsh conditions in which this intermediate was subjected
during the coupling reaction: our intent was to avoid the unwanted hydrolysis of the protecting amide
group by using a quite resistant protecting group such as the benzoyl instead of the acetyl. Once the
benzamide group was removed by basic hydrolysis for prolonged times, piperidine 55 was condensed
with 3-methoxybenzoic acid 48 and finally the methoxy group was converted to the free phenolic OH
to give final compound 22 (Scheme 5).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
O
O
O
Br
N
NH
N
Ph
N
Ph
d
e
N
Ph
ROOC
ROOC
O
O
4
4 R = H
51 R = Et
52 R = H
53
54
a
b
c
50 R = Et
O
O
O
HOOC
OMe
N
N
NH
48
N
f
g
O
O
56
OMe
55
h
O
O
N
N
O
OH
22
Scheme 5. Synthesis of derivative morpholine-substituted derivative 22.
Reagents and conditions: (a) SOCl , EtOH, reflux, 3 h [74%]; (b) benzoyl chloride, Et N, dry DCM,
2
3
RT, overnight [93%]; (c) NaOH, EtOH/H O, RT, overnight [99%]; (d) i. SOCl , dry 1,2-DCE, 60 °C,
2
2
4
h; ii. PhBr, AlCl , dry 1,2-DCE, 90 °C, overnight [43%]; (e) morpholine, Pd (dba) , XPhos, K PO ,
3 2 3 3 4
1
6
ACS Paragon Plus Environment

Page 17 of 83
Journal of Medicinal Chemistry
1
2
3
toluene, 100 °C, 20 h [63%]; (f) NaOH 1N, EtOH, 95 °C, overnight [72%]; (g) HATU, DIPEA, dry
DMF, RT, 3 h [99%]; (h) 1M BBr , dry CH Cl , −10 to 0 °C, then RT, 1.5 h [27%].
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Finally, we combined the structural portions that gave the best results in enzymatic assays (see
below): the i-propyl and n-butyl chains were favored as substituents in the benzoyl ring and o-/p-F-
phenolic portions were chosen in the other terminal benzamide portion. Therefore, amines 47e and
47f, which were prepared as displayed in Scheme 4, were reacted either with benzoic acids 36b,
synthesized as outlined in Scheme 2, or with commercially available compound 29f (Scheme 1), since
it was observed that a fluorine atom in para to the phenolic hydroxyl group on the amidic phenyl ring
or to the amide carbonyl group was beneficial for MAGL inhibition potency. The corresponding
amides (57, 58 and 59, 60) were converted to the final compounds by BBr -promoted demethylation
3
of the methoxy groups, to give compounds 23-26 (Scheme 6).
F
COOH
R
HOOC
F
O
F
OMe
36b
a
OMe
29f
a
O
R
R
N
NH
N
F
O
OMe
O
47e, f
e: R = i-Pr;
f: R = n-Bu.
O
OMe
5
7, 58
59, 60
59: R = i-Pr;
60: R = n-Bu.
5
5
7: R = i-Pr;
8: R = n-Bu.
b
b
O
N
F
O
N
F
F
O
O
OH
F
O
O
OH
OH
2
3
4
25
O
O
3
N
3
N
OH
2
26
Scheme 6. Synthesis of compounds 23-26.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Reagents and conditions: (a) HATU, DIPEA, dry DMF, RT, 3 h [86-93%]; (b) 1M BBr , dry CH Cl ,
3
2
2
−
10 to 0 °C, then RT, 1-1.5 h [17-87%].
1
13
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Analysis of H and C-NMR spectra of some of these benzoylpiperidine derivatives confirmed the
presence of two rotational conformers generated by the hindered rotation around the C-C bond
between the amidic phenyl ring and the carbonyl group nearby, as previously observed by us for
3
3
1
13
similar compounds. The splitting of H and C-NMR signals was observed for compounds bearing
bulkier halogen atoms in the ortho position to the amidic carbonyl group, such as I (10a and 11a), Br
(10b and 11b) and Cl (10c and 11c). Conversely, the presence of a smaller fluorine atom in the same
position, as in compounds 10d and 11d, led to the presence of a unique species detected by NMR,
and this fact may be explained by considering the smaller size of the fluorine atom, which allows a
free rotation around the above-mentioned C-C bond.
Enzymatic assays and molecular modeling. The newly synthesized compounds were evaluated for
their inhibition potency in human MAGL by using an enzymatic assay in which 4-nitrophenylacetate
was used as the substrate and the absorbance values relative to the amount of the formed product 4-
nitrophenol were measured at 405 nm. The IC values obtained for the new series of compounds
5
0
were compared with those of two reference inhibitors: the previously published (4-(4-
chlorobenzoyl)piperidin-1-yl)(3-hydroxyphenyl)methanone (9), which represents the parent
compound of this new class of derivatives, and the irreversible MAGL inhibitor CAY10499 (2),
which is one of the most potent MAGL inhibitors present in literature. The MAGL inhibition activity
data for the first series of compounds, halogenated derivatives 10a-d, 11a-d, 12a-d and 13a-d, are
reported in Table 1. It is evident that the presence of the halogen atom in the meta or ortho position
to both the amidic carbonyl moiety and the phenolic hydroxyl group (12a-d and 10a-d, respectively)
was not satisfactory for the MAGL inhibitory activity, because the IC values of these compounds
50
1
8
ACS Paragon Plus Environment

Page 19 of 83
Journal of Medicinal Chemistry
1
2
3
were only in the micromolar range. In particular, they were greater than 1 µM for compounds 12a-d
and even greater than 5 µM for compounds 10a-d, thus resulting to be less active than original
compound 9. On the contrary, the other two substitution positions led to very positive effects in some
cases. In fact, the shift of the halogen atom to the para position to either the amidic carbonyl moiety
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(13a-d) or the phenolic hydroxyl group (11a-d) generally produced more potent inhibitors and
highlighted a possible trend: the inhibitory activity increased in parallel with the electron withdrawing
property of the halogen present in the molecule, since it gradually increased going from iodine- (11a,
1
3a) to bromine- (11b, 13b), chlorine- (11c, 13c) and fluorine- (11d, 13d) substituted compounds. In
particular, in the series of compounds 11a-d, iodine and bromine led to IC values in the range of
5
0
2.4-2.5 µM, but when the halogen atom was a chlorine the activity of the resulting compound 11c
increased (IC value of 960 nM). This effect was even more pronounced in the presence of a fluorine
5
0
atom, such as in compound 11d, whose IC value was of 240 nM, thus reaching a higher inhibition
5
0
potency than that of parent compound 9. Similarly, in the series 13a-d we observed an evident
improvement by passing from iodine (IC = 1.3 µM) to bromine and chlorine (IC = 457 and 424
5
0
50
nM, respectively) and finally reaching the best inhibition activity in this group of halogenated
benzoylpiperidines with the insertion of a fluorine atom in para to the amidic carbonyl group in
compound 13d, which showed an IC value of 389 nM (Table 1). The above-mentioned MAGL
5
0
inhibition improvement given by fluorine atom in two specific positions of the amidic phenyl ring
may rely on the small size and the powerful electron withdrawing property of fluorine. This atom is
an isoster of hydrogen, but additionally it increases the acidity of the neighboring phenolic OH,
stabilizing the anionic form, the phenoxide, thus leading to a highly polarized hydroxyl and,
consequently, to the formation of stronger hydrogen bonds involving this group. The negative charge
on the phenoxide can be delocalized on the carbons in para or ortho positions, that correspond to the
same carbon atoms where the fluorine atom is bound (compounds 11d and 13d), thus contributing to
a further stabilization of the charge. On the contrary, considering the two possible ortho positions to
the phenolic OH, the one in which F is in the middle between the carbonyl and the phenolic OH leads
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to negative results for the inhibition activity (compound 10d). This effect could be rationalized by
considering that the steric bulk in compound 10d induces a plausible block of rotation around the
adjacent C-C bond, thus preventing an ideal approach of the hydroxyl group to the protein residues
involved in the hydrogen bond network.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Table 1. In vitro inhibitory activity on MAGL (IC , nM) of halogenated derivatives 10a-d, 11a-d,
5
0
1
2a-d, 13a-d.
O
Cl
N
X
O
OH
Compounds
X
IC (nM)
50
a
b
c
d
a
I
5300 ± 200
8200 ± 300
8700 ± 500
12300 ± 600
2500 ± 400
2400 ± 300
960 ± 50
Br
Cl
F
1
0
X
OH
X
I
b
c
Br
Cl
11
d
a
b
c
d
a
b
c
F
I
Br
Cl
F
240 ± 26
1400 ± 200
1300 ± 200
1100 ± 100
1300 ± 200
1300 ± 100
457 ± 6
OH
X
X
1
1
2
3
OH
OH
I
Br
Cl
F
424 ± 46
389 ± 15
d
O
Cl
N
9
840 ± 40
O
OH
CAY10499
JZL184
KML29
134 ± 15
49.3 ± 3.9
3.0 ± 0.1
a
Enzymatic values are the mean of three or more independent experiments, performed in duplicate.
2
0
ACS Paragon Plus Environment

Page 21 of 83
Journal of Medicinal Chemistry
1
2
3
With the aim of investigating the effect of the introduction of halogen atoms in different positions of
the amidic phenyl ring of the benzoylpiperidine derivatives, the compounds showing the most
interesting structure-activity relationship (SAR) have been analyzed through molecular modeling
studies, thus evaluating their binding mode within MAGL and the impact of their different
substitution pattern on the ligand-protein interactions. In particular, compounds 11d and 10d,
presenting the highest and lowest MAGL inhibitory activity, respectively, among the halogenated
derivatives, were subjected to a robust docking procedure (see the Experimental section for details),
which further confirmed the reliability of the binding mode already predicted for the parent inhibitor
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
. In fact, these two ligands shared a very similar disposition within the protein, with the 4-
chlorobenzoyl moiety directed toward the entrance of the catalytic site and showing lipophilic
33
interactions with L148, L205, L213 and L241 already observed for compound 9. Moreover, the
carbonyl oxygen of 10d and 11d forms H-bonds with the backbone nitrogen of A51 and M123. The
fluorophenol ring of the ligands is placed within the small pocket of the enzyme binding site,
sandwiched between Y194 and V270 that form lipophilic interactions with the inhibitors (Figure 4).
Interestingly, the docking calculations suggested that only compound 11d was able to form the H-
bond network with E53 and H272 through its hydroxyl group, as predicted for the parent compound
9. In fact, the fluorophenol ring of compound 10d was predicted to be about 180° rotated with respect
to that of 11d, with the hydroxyl group pointing toward the inner side of the binding site cavity, far
from both E53 and H272. In this orientation, the hydroxyl group can only form an H-bond with the
side chain of R57 (Figure 4B). This difference in the binding mode predicted for the two derivatives
might be ascribed to the steric hindrance that would occur between the fluorine atom and the central
piperidine core of 10d if its fluorophenol ring assumed the opposite orientation in order to place the
OH group in close proximity to the E53/H272 pair. Moreover, a repulsive electrostatic interaction
between the highly electronegative fluorine atom and the close carboxylic group of E53 would need
to be overcome for allowing the formation of the H-bond between E53 and the ligand hydroxyl group.
Such steric and electrostatic factors could make this binding mode energetically unfavorable for 10d,
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compared to that displayed in Figure 4B. Therefore, the consequential loss of the H-bond network
with E53 and H272 could explain the drop of activity observed for compound 10d with respect to
11d.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Docking results of compounds 11d (A) and 10d (B) into MAGL (PDB code 3PE6).
In order to analyze the binding modes proposed for the two compounds from an energetic point of
view, the predicted 11d-MAGL and 10d-MAGL complexes were further studied through a 101 ns
molecular dynamics (MD) simulation protocol followed by ligand-protein binding energy evaluations
performed using the Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) method (see
the Experimental section for details). The MD studies confirmed the reliability of the predicted
binding modes, as both compounds maintained their disposition within MAGL catalytic site during
the simulations. The last 80 ns of MD simulation were employed to estimate the binding free energies
2
2
ACS Paragon Plus Environment

Page 23 of 83
Journal of Medicinal Chemistry
1
2
3
associated with the two ligand-protein complexes. The results of the energetic evaluations were found
to be consistent with the experimental activities of the compounds and further confirmed the
reliability of the predicted ligand binding modes. In fact, the ligand-protein binding energy calculated
for compound 11d (-61.3 kcal/mol) exceeded of about 11 kcal/mol the binding energy predicted for
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0d (-50.4 kcal/mol), in agreement with the remarkably higher MAGL inhibition activity of 11d
compared to 10d (Table S1). The whole computational protocol including docking, MD simulations
and binding free energy evaluations was then employed to evaluate and analyze the binding mode of
compound 13d, which showed an IC50 value for MAGL inhibition considerably close to that
determined for compound 11d. As expected, compound 13d showed a binding disposition very
similar to that predicted for 11d, consistently with the very similar activity of the two compounds. In
particular, compound 13d was able to form the H-bond network with E53 and H272 through its
hydroxyl group (Figure S1). In agreement with the experimental data, free energy evaluations
performed on MAGL in complex with 13d predicted a ligand-protein binding energy of -61.0
kcal/mol, comparable to that calculated for the 11d-MAGL complex, thus supporting the reliability
of the binding mode predicted for the p-fluoro derivative 13d.
The inhibition activity results obtained for compounds 14-22, which are variously substituted on the
ketone part of the scaffold, are reported in Table 2. The removal of the chlorine atom of compound 9
led to compound 14, possessing an unsubstituted benzoyl ring, but this modification was not
beneficial for the inhibition activity on MAGL, resulting in an IC value of about 2.2 µM. On the
5
0
other hand, the insertion of alkyl groups of increasing dimensions – going from a methyl group
(compound 15) to a n-butyl chain (compound 19) – generated a progressive improvement of the
inhibitory activity. However, this behavior was subjected to a sort of “plateau”: after an initial
decrease of the IC values from methyl, to ethyl and n-propyl chain of compounds 15, 16 and 17
5
0
(
IC = 368, 194 and 162 nM, respectively), the best results were reached by i-propyl (18) and n-butyl
50
(
19)-substituted compounds that showed nearly the same IC values (138 and 136 nM, respectively),
5
0
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
thus indicating a limit in the increase of the activity induced by the increase of the dimension of the
alkyl chain. Moreover, it is worth noting that compounds 18 and 19 were not only more potent than
parent compound 9, but their IC values were comparable to that of irreversible inhibitor CAY10499
5
0
3
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and were only about 3-fold less active than JZL184. With regards to compound KML29, it was
about 46-fold more active than compounds 18 and 19. The presence of polar groups in the place of
the aliphatic chains, such as a phenol (20), a benzodioxane (21) or a morpholine ring (22), was
detrimental for the enzyme inhibition activity of the resulted compound, which raised to values in the
micromolar range (in the range of 1-10 µM), thus suggesting that the insertion of polar moieties in
this position is strongly disadvantageous for MAGL inhibition.
a
Table 2. In vitro inhibitory activity on MAGL (IC , nM) of compounds 14-22.
5
0
O
N
Ar
O
OH
Compounds
IC (nM)
50
1
4
2200 ± 200
368 ± 25
1
5
16
194 ± 19
162 ± 5
2
1
1
1
7
8
9
138 ± 9
136 ± 9
3
2
4
ACS Paragon Plus Environment

Page 25 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HO
2
2
0
1
10200 ± 1100
1000 ± 100
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
22
5000 ± 900
O
Cl
N
9
840 ± 40
O
OH
a
Enzymatic values are the mean of three or more independent experiments, performed in duplicate.
Considering the inhibitory activities of the compounds belonging to the two previous series of MAGL
inhibitors (Table 1 and 2), we selected the more potent inhibitors with the aim of combining the best
structural modifications in new molecules. Therefore, the presence of a fluorine atom in para position
to the amidic carbonyl group (compound 13d) or to the phenolic hydroxyl group (compound 11d), as
well as the insertion of a n-butyl (compound 19) or i-propyl (compound 18) chain on the ketone-type
benzoyl ring were the most promising modifications, which were therefore combined to obtain
“
hybrid” compounds 23-26 (Table 3). These new four compounds showed IC50 values in the
nanomolar range, which were comparable to, or even better than, those of the original compounds
which inspired their design. In fact, compounds 25 and 26 (IC = 110 and 262 nM, respectively),
50
bearing a common p-F,m-OH-benzamido motif, were more active than their simpler analogue 13d
(
IC = 389 nM). However, 26 was less potent than its n-butyl-substituted non-fluorinated analogue
50
1
9 (IC values of 262 and 136 nM, respectively), whereas the inhibition potency of 25 was quite
50
similar to that of its i-propyl-substituted non-fluorinated analogue 18 (IC values of 110 and 138 nM,
5
0
respectively). The strongest synergistic effect was observed in compounds 23 and 24, which both
shared a common o-F,m-OH-benzamido portion. In fact, 23 and 24 gave IC values of 80 and 74
5
0
nM, respectively, thus being remarkably more active than their close analogues 11d, 18 and 19. These
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
two new compounds were also found to be more potent than both reference compounds 9 and 2.
Therefore, 23 and 24 currently represent the most active benzoylpiperidine-based MAGL inhibitors.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Table 3. In vitro inhibitory activity on MAGL (IC , nM) of compounds 23-26, 57.
5
0
Compounds
IC (nM)
50
O
O
F
N
N
2
2
3
4
80 ± 12
74 ± 1
O
O
OH
F
3
OH
O
N
N
2
2
5
6
110 ± 8
F
O
OH
OH
O
3
262 ± 26
F
O
O
F
N
57
2600 ± 400
O
O
OMe
O
Cl
N
9
840 ± 40
OH
a
Enzymatic values are the mean of three or more independent experiments, performed in duplicate.
Finally, in order to confirm the importance of the phenolic hydroxyl group, which was maintained
fixed in all synthesized derivatives, we tested the methoxylated precursor of compound 23, compound
5
7 (Scheme 6, Table 3). The decrease of activity of compound 57 was evident, since it displayed an
IC value of 2.6 µM, compared to its analogue 23 with the free OH (IC = 80 nM), thus confirming
5
0
50
that the hydroxyl group is essential for the interaction with MAGL, as previously suggested by
3
3
modeling studies. In order to have information about the lipophilic properties of the herein reported
2
6
ACS Paragon Plus Environment

Page 27 of 83
Journal of Medicinal Chemistry
1
2
3
compounds, an evaluation of the logP has been carried out. Table S2 shows the consensus logP values
obtained through the Chemicalize tool. None of the analyzed compounds showed a logP value higher
than five and, furthermore, no evident correlation between logP and activity of the different
compounds was observed; in fact, by plotting the logP values against the measured pIC values of
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
2
the ligands a correlation (R ) lower than 0.2 was obtained. In order to confirm the hypothesized
reversible mechanism of inhibition, the effects of dilution and preincubation on the inhibitory activity
of compound 23 were evaluated. In the presence of an irreversible mechanism of inhibition, the
potency should not decrease after dilution, whereas for a reversible inhibition, the potency level
3
6
should be strongly reduced after dilution. Therefore, the inhibition produced by incubation with a
4000 nM concentration of 23 was measured after a 40X dilution and compared to the potency
observed by a 4000 nM and a 100 nM of compound 23. The results showed in Figure 5A supported
a reversible mechanism of inhibition, since the inhibition produced by 100 nM of this compound was
similar to that obtained after a 40X dilution and was considerably lower than that produced by the
same compound at a concentration of 4000 nM. As a second assay, the inhibition activity of 23 was
measured at different preincubation times with MAGL. Compound 23 was preincubated with the
enzyme for 0, 30 and 60 min before adding the substrate to start the enzymatic reaction. An
irreversible inhibition should produce a higher potency after longer incubation times, whereas a
reversible inhibitor should produce a constant inhibition potency over all the different incubation
times. As shown in Figure 5B, this test agreed with the reversible property of 23, as it showed a very
similar activity for the three different incubation times.
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Analysis of the mechanism of MAGL inhibition by compound 23-MAGL inhibition
analysis. A) Dilution assay: the first two columns indicate the inhibition percentage of compound 23
at a concentration of 4000 nM and 100 nM. The third column indicates the inhibition percentage of
compound 23 after dilution (final concentration = 100 nM). B) IC (nM) values of 23 at different
5
0
preincubation times with MAGL (0 min, 30 min and 60 min).
Once we verified the reversible mechanism of inhibition of compound 23, its mode of inhibition was
then evaluated by measuring Michaelis–Menten kinetics at various inhibitor concentrations. The
dataset was plotted as substrate concentration versus enzyme activity and analyzed by applying the
mixed-model inhibition fit of GraphPad Prism 5.0, which includes competitive, uncompetitive, and
noncompetitive inhibition terms. This model evaluates the Vmax, K and the α parameter, which is
m
indicative of the inhibition mechanism. When α corresponds to one, the inhibitor does not alter the
binding of the substrate to the enzyme, and the mixed-model can be considered as a noncompetitive
2
8
ACS Paragon Plus Environment

Page 29 of 83
Journal of Medicinal Chemistry
1
2
3
inhibition. When α is a very large value, the inhibitor interaction prevents the substrate binding and
the mixed-model corresponds to a competitive inhibition. Finally, when α is a very small value, the
binding of the inhibitor increases the binding of the substrate and the mixed model corresponds to an
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
uncompetitive model. Kinetic studies indicate for 23 a K value of 39 ± 4 nM and an α value greater
i
than 10000, thus supporting a competitive behavior for this compound (Figure 6).
Figure 6. Inhibition of the activity of MAGL and competitive nature of compound 23 (K = 39 ± 4
i
nM, α >10000). 4-NPA = 4-nitrophenylacetate.
The binding mode of the most promising derivative of the series compound 23, which showed the
best activity in the cell-based assay and selectivity profile (see also following sections), was studied
through the same robust docking procedure applied on 10d, 11d and 13d. As expected, the predicted
disposition of compound 23 within MAGL catalytic site, as well as its H-bond interactions with the
binding site residues (Figure 7), were highly similar to those predicted for compound 11d. The
predicted MAGL-23 complex was further studied through the same MD simulation protocol applied
on the other derivatives, in order to evaluate the reliability of the predicted ligand-protein interactions
(see the Experimental section for details). The ligand was found to maintain its binding disposition
with remarkable stability during the whole simulation, showing an average root-mean-square
2
9
ACS Paragon Plus Environment