Development and Pharmacological Characterization of Selective Blockers of 2-Arachidonoyl Glycerol Degradation with Efficacy in Rodent Models of Multiple Sclerosis and Pain

Article abstract of DOI:10.1021/acs.jmedchem.5b01812

We report the discovery of compound 4a, a potent β-lactam-based monoacylglycerol lipase (MGL) inhibitor characterized by an irreversible and stereoselective mechanism of action, high membrane permeability, high brain penetration evaluated using a human in vitro blood-brain barrier model, high selectivity in binding and affinity-based proteomic profiling assays, and low in vitro toxicity. Mode-of-action studies demonstrate that 4a, by blocking MGL, increases 2-arachidonoylglycerol and behaves as a cannabinoid (CB1/CB2) receptor indirect agonist. Administration of 4a in mice suffering from experimental autoimmune encephalitis ameliorates the severity of the clinical symptoms in a CB1/CB2-dependent manner. Moreover, 4a produced analgesic effects in a rodent model of acute inflammatory pain, which was antagonized by CB1 and CB2 receptor antagonists/inverse agonists. 4a also relieves the neuropathic hypersensitivity induced by oxaliplatin. Given these evidence, 4a, as MGL selective inhibitor, could represent a valuable lead for the future development of therapeutic options for multiple sclerosis and chronic pain.

Full text of DOI:10.1021/acs.jmedchem.5b01812

Subscriber access provided by HKU Libraries
Article
Development and Pharmacological Characterization of
Selective Blockers of 2-Arachidonoyl Glycerol Degradation
with Efficacy in Rodent Models of Multiple Sclerosis and Pain
Margherita Brindisi, Samuele Maramai, Sandra Gemma, Simone Brogi, Alessandro Grillo, Lorenzo Di
Cesare Mannelli, Emanuele Gabellieri, Stefania Lamponi, Simona Saponara, Beatrice Gorelli, Daniele
Tedesco, Tommaso Bonfiglio, Christophe Landry, Kwang-Mook Jung, Andrea Armirotti, Livio Luongo,
Alessia Ligresti, Fabiana Piscitelli, Carlo Bertucci, Marie-Pierre Dehouck, Giuseppe Campiani, Sabatino
Maione, Carla Ghelardini, Anna Pittaluga, Daniele Piomelli, Vincenzo Di Marzo, and Stefania Butini
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01812 • Publication Date (Web): 18 Feb 2016
Downloaded from http://pubs.acs.org on February 25, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Development and Pharmacological Characterization of Selective
Blockers of 2-Arachidonoyl Glycerol Degradation with Efficacy in
Rodent Models of Multiple Sclerosis and Pain
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
¥
¥
¥
¥
¥
Margherita Brindisi, Samuele Maramai, Sandra Gemma, Simone Brogi, Alessandro Grillo,
‡
¥
¥
●
Lorenzo Di Cesare Mannelli, Emanuele Gabellieri, Stefania Lamponi, Simona Saponara,
●
§
&
▀
Beatrice Gorelli, Daniele Tedesco, Tommaso Bonfiglio, Christophe Landry, KwangꢀMook
▼
◊
#
†
†
§
Jung, Andrea Armirotti, Livio Luongo, Alessia Ligresti, Fabiana Piscitelli, Carlo Bertucci,
▀
¥,
#
‡
MarieꢀPierre Dehouck, Giuseppe Campiani,
* Sabatino Maione, Carla Ghelardini, Anna
&,£
▼
†
¥
Pittaluga, Daniele Piomelli, Vincenzo Di Marzo, and Stefania Butini
¥
European Research Centre for Drug Discovery and Development (NatSynDrugs), University of
Siena, via Aldo Moro 2, Iꢀ53100 Siena, Italy
●
Department of Life Sciences, University of Siena, via Aldo Moro 2, Iꢀ53100 Siena, Italy
§
Department of Pharmacy and Biotechnology Alma Mater Studiorum, University of Bologna, Via
Belmeloro 6, 40126 – Bologna, Italy
‡
Department of Neuroscience, Psychology, Drug Research and Child Health, Section of
Pharmacology and Toxicology, University of Firenze, V.le G. Pieraccini 6, 50139 Firenze, Italy
&
Department of Pharmacy, Section of Pharmacology and Toxicology, University of Genova, Viale
Cembrano 4, Genova, 16148, Italy
▀
Laboratoire de la Barrière HématoꢀEncéphalique (LBHE), University of Artois, EA 2465, F62300
LensꢀFrance
▼
Department of Anatomy and Neurobiology, University of California, 92617 Irvine, California,
USA
◊
Drug Discovery and Development, Istituto Italiano di Tecnologia, Genova, 16163, Italy
ACS Paragon Plus Environment

Page 3 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
#
Endocannabinoid Research Group, Department of Experimental Medicine – Division of
Pharmacology “L. Donatelli”; Second University of Napoli, Napoli, Italy
†£
Endocannabinoid Research Group, Institute of Biomolecular Chemistry, C.N.R., Pozzuoli
(Napoli), Italy
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
£
Center of Excellence for Biomedical Research, University of Genova, Viale Benedetto XV, Genova,
1
6132, Italy
RECEIVED DATE (to be automatically inserted after your manuscript is accepted if
required according to the journal that you are submitting your paper to)
Abstract
We report the discovery of compound 4a, a potent βꢀlactamꢀbased monoacylglycerol lipase (MGL)
inhibitor characterized by an irreversible and stereoselective mechanism of action, high membrane
permeability, high brain penetration evaluated using a human in vitro blood brain barrier model,
high selectivity in binding and affinityꢀbased proteomic profiling assays, and low in vitro toxicity.
Modeꢀofꢀaction studies demonstrate that 4a, by blocking MGL, increases 2ꢀarachidonoylglycerol,
and behaves as cannabinoid (CB1/CB2) receptor indirect agonist. Administration of 4a in mice
suffering from experimental autoimmune encephalitis ameliorates the severity of the clinical
symptoms in a CB1/CB2ꢀdependent manner. Moreover, 4a produced analgesic effects in a rodent
model of acute inflammatory pain, which was antagonized by CB1 and CB2 receptor
antagonists/inverse agonists. 4a also relieves the neuropathic hypersensitivity induced by
oxaliplatin. Given these evidences, 4a, as MGL selective inhibitor, could represent a valuable lead
for the future development of therapeutic options for multiple sclerosis and chronic pain.
Keywords: Multiple sclerosis, MGL inhibitors, neuropathic pain, βꢀlactams.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
Multiple sclerosis (MS) is a neuroinflammatory autoimmune disease characterized by
1
demyelination and axonal damage leading to loss of cerebral functions. The treatment landscape of
MS includes several antiꢀinflammatory agents and monoclonal antibodies for immunotherapy.
Besides demyelination and inflammation, thought to play the major role in MS progression, central
neuronal defects also occur and have been associated to the onset of symptoms such as spasticity
and pain.
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 the last few years an increasing body of evidence has suggested that cannabinoids may
2
have beneficial effects on the symptoms of MS. More recently, attention has focused on small
molecules modulating the endocannabinoid system, although more investigation is needed to prove
whether endocannabinoid system modulation may represent a valuable therapeutic option to
3
successfully reverse the clinical progression of MS. In fact, therapeutic inhibition of
endocannabinoid degrading enzymes increases the endocannabinoids (ECs) tissue levels and is less
likely to cause psychoactive effects than direct agonism of CB1 receptors, whilst maintaining other
4
,5,6
beneficial effects of CB1/2 receptors activation.
ECs such as anandamide and 2ꢀarachidonoylꢀsnꢀglycerol (2ꢀAG) are signaling lipids that exert their
7
biological actions by interacting with cannabinoid receptors (CBRs); they modulate multiple
8
physiological processes including pain, inflammation, appetite, memory and emotion. In the brain,
2ꢀAG is produced by diacylglycerol lipases and it is deactivated through monoacylglycerol lipase
(MGL)ꢀmediated hydrolysis. Due to its prominent role in 2ꢀAG catabolism, selective inactivation of
MGL represents an attractive approach for eliciting the desirable effects of indirect CBR activation.
9
MGL is a serine hydrolase with a catalytic triad represented by Ser122, His269 and Asp239.
Inhibitors of 2ꢀAG degradation have been developed such as the carbamate biphenylꢀ3ꢀylcarbamic
10,11,12
acid cyclohexyl ester (URB602),
the tetrahydrolipstatin (S)ꢀ1ꢀ((2S,3S)ꢀ3ꢀhexylꢀ4ꢀoxooxetanꢀ
13
2
ꢀyl)tridecanꢀ2ꢀyl 2ꢀformamidoacetate (OMDM169), the 4ꢀ(bisꢀbenzo[1,3]dioxolꢀ5ꢀylꢀhydroxyꢀ
15
methyl)ꢀpiperidineꢀ1ꢀcarboxylic acid 4ꢀnitroꢀphenyl ester 1 (JZL184), the ureidic 4ꢀ(bis(4ꢀ
ACS Paragon Plus Environment

Page 5 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
fluorophenyl)methyl)piperazinꢀ1ꢀyl)(1Hꢀ1,2,4ꢀtriazolꢀ1ꢀyl)methanone 2 (SAR629),
and their
analogue (4ꢀ(bis(benzo[d][1,3]dioxolꢀ5ꢀyl)methyl)piperidinꢀ1ꢀyl)(1Hꢀ1,2,4ꢀtriazolꢀ1ꢀyl)methanone
16
3
(JJKKꢀ048).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Despite the variety of MGL inhibitors available, the structural diversity of chemical
templates explored to date remains poor. Compound 1, which shares with most MGL inhibitors the
diphenylmethane core, is frequently used as experimental tool but it exhibits only submicromolar
activity against the human MGL enzyme.
We describe herein the development of a set of βꢀlactamꢀbased inhibitors (4a-h, Chart 1) of human
MGL (hMGL) which led to the identification of 4a as the prototypic compound of this class of
analogues. 4a was characterized as a novel, ultraꢀpotent hMGL inhibitor, with high selectivity
towards the anandamideꢀinactivating enzyme, fatty acid amide hydrolase (FAAH), other serine
hydrolases from the rat brain proteome, and CBRs. The unique profile of this prototypic inhibitor
combines selectivity, potency, and favorable pharmacokinetic properties, highlighting its potential
as a novel MGL inhibitor for further development (Table 1). As expected for a potent and selective
MGL inhibitor, in vivo administration of 4a reduced the clinical severity of the disease in mice
suffering from Experimental Autoimmune Encephalomyelitis (EAE), a rodent demyelinating
17
disease model. In vivo tests confirmed the analgesic profile of 4a in mouse models and evidenced
a clearꢀcut dependence of its pharmacological efficacy upon the increased 2ꢀAG levels and
subsequent indirect CBRs modulation. Remarkably, the efficacy of 4a further supports the
2
hypothesis of a tight intersection between the endocannabinoid system and MS.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
F
F
O
O
O
O
O
O
O
O
OH
N
N
N
N
N
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
N
O
N
O
O
NO2
N
N
1
, JZL184
2, SAR629
3, JJKK048
R2
R
1
N
O
N
R3
O
4
a-h, as defined in Scheme 1 and Table 1
Chart 1. Representative diphenylmethane inhibitors (compounds 1ꢀ3) and general structure of title
compounds 4a-h.
Chart 2. Molecular deconstruction units of compound 4a.
ACS Paragon Plus Environment

Page 7 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemistry
Synthesis of compounds (±)ꢀ, (+)ꢀ and (ꢀ)ꢀ4a, 4b-h, (±)ꢀ5a-e, (±)-6a,e, (±)-7, (±)-8 and 9 is reported
in Schemes 1ꢀ3.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Aldehydes 10a-c were reacted with 4ꢀaminoꢀ1ꢀbenzylpiperidine to form the intermediate imines
1
8
1
1a-c. These latter were subjected to the Staudinger reaction using the suitable phenylacetic acid,
in the presence of triphosgene, to obtain the βꢀlactam systems (±)-5a-e in the trans configuration.
The benzyl groups of (±)-5a-e were then cleaved by means of a palladium catalyzed hydrogenation
leading to free amines (±)-6a-e. Amines (±)-6a,e were reacted with 1Hꢀ1,2,4ꢀtriazole in the
presence of phosgene and DMAP, providing final compound (±)-4a,e-h. Enantiomerically pure
amines (3R,4S)ꢀ(+)ꢀ6a and (3S,4R)ꢀ(ꢀ)ꢀ6a, separated by HPLC as reported in the Supporting
Information, were also submitted to final reaction with 1Hꢀ1,2,4ꢀtriazole in the presence of
phosgene, leading to enantiomerically pure final compounds (3R,4S)ꢀ(+)ꢀ4a and (3S,4R)ꢀ(ꢀ)ꢀ4a
respectively. Compounds 4b, 4c, and 4d were prepared by reaction of amine (±)ꢀ6a with 4ꢀ
nitrophenyl chloroformate, 1,1ꢀcarbonyldiimidazole, 1,1,1,3,3,3ꢀhexafluoroisopropanol and
triphosgene respectively (Scheme 1).
Scheme 1. Synthesis of Title Compounds 4a-h.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Reagents and conditions: a) 4ꢀaminoꢀ1ꢀbenzylpiperidine, EtOH, 80 °C, 12 h; b)
triphosgene, dry DCM, 50 °C, 0.5 h, TEA, 50 °C, 12 h; c) H , Pd/C 10%, MeOH, 25 °C, 2 h; d) 1Hꢀ
2
1,2,4ꢀtriazole, phosgene (20% in toluene), DMAP, dry DCM, 25 °C, 12 h; e) for 4b: 4ꢀnitrophenyl
chloroformate, TEA, dry DCM, 25 °C, 12 h; for 4c: 1,1ꢀcarbonyldiimidazole, dry DCM, 25 °C, 12
ACS Paragon Plus Environment

Page 9 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
h; for 4d: 1,1,1,3,3,3ꢀhexafluoroisopropanol, triphosgene, DIPEA, dry DCM, 25 °C, 12 h; for 4a,e-
h: 1Hꢀ1,2,4ꢀtriazole, phosgene (20% in toluene), DMAP, dry DCM, 25 °C, 12 h.
Compound (±)-12 was prepared by alane reduction of compound (±)ꢀ5a. Subsequent hydrogenation
and treatment of the resulting amine (±)-13 with 1Hꢀ1,2,4ꢀtriazole, phosgene and DMAP led to
derivative (±)-7 (Scheme 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Synthesis of Analogue (±)-7
Scheme 2. Synthesis of compounds (±)ꢀ7. Reagents and conditions: a) LiAlH
4
, AlCl
, 10% Pd/C, MeOH, 25 °C, 2 h; c) 1Hꢀ
,2,4ꢀtriazole, phosgene (20% in toluene), DMAP, dry DCM, 25 °C, 12 h.
3
, dry THF,
0
1
°C, 45 min then 10% aqueous NaOH, 0 °C, 30 min; b) H
2
Catalyst 16, used for the diastereoselective synthesis of the transꢀ3,4ꢀdiarylsubstituted βꢀlactam
system, was prepared by refluxing 14 with 3ꢀaminoꢀ1ꢀpropanol in chlorobenzene and successive
treatment of the intermediate 15 with mesyl chloride, in the presence of N,Nꢀdiisopropylethylamine
19
(
DIPEA). Meanwhile, imine 17, obtained by heating at reflux 10a with 4ꢀtoluenesulfonamide in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
0
absolute ethanol, was subjected to the cyclization reaction using 4ꢀfluorophenylacetic acid, in the
presence of tosyl chloride and catalyst 16, obtaining the Nꢀtosylated transꢀβꢀlactam (±)-18. This
2
latter was treated with a solution of SmI in tetrahydrofuran to give the final compound (±)-8.
Compound 9 was prepared by reaction of piperidine (19) with 1Hꢀ1,2,4ꢀtriazole and phosgene in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
presence of DMAP (Scheme 3).
Scheme 3. Synthesis of βꢀlactam (±)ꢀ8, and of compound 9
Scheme 3. Synthesis of compounds (±)ꢀ8. Reagents and conditions: a) 3ꢀaminoꢀ1ꢀpropanol,
DIPEA, chlorobenzene, 130 °C, 4 h; b) TEA, MsCl, dry DCM, 0 to 25 °C, 1.5 h then MeOH, TEA,
5
3 2
0 °C, 1.5 h; c) 4ꢀtoluensulfonamide, BF ·Et O, dry toluene, 130 °C, 12 h; d) 4ꢀfluorophenylacetic
acid, TsCl, DIPEA, dry DCM, 0 °C, 30 min then 16, DIPEA, 25 °C, 5 h; e) SmI 0.1 N solution in
2
ACS Paragon Plus Environment

Page 11 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
THF, dry THF, 25 °C, 15 min; f) 1Hꢀ1,2,4ꢀtriazole, phosgene (20% in toluene), DMAP, dry DCM,
2
5 °C, 12 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To evaluate the stereoselectivity of the interaction of (±)-4a with hMGL (see Table 1 in next
paragraph), we resolved the racemic mixture of amine (±)-6a by using chiral HPLC (see Supporting
Information). The two amines (+)ꢀ6a and (ꢀ)ꢀ6a, showing an optical purity > 98% (Figure S2),
underwent the final reaction to yield the two enantiomers of 4a. Enantiomers (+)ꢀ6a and (ꢀ)ꢀ6a
yielded (+)ꢀ4a and (ꢀ)ꢀ4a, respectively. The absolute configuration of (+)ꢀtransꢀ6a was established
by means of electronic circular dichroism (ECD) spectroscopy combined with timeꢀdependent
density functional theory (TDꢀDFT) calculations of the theoretical chiroptical properties of (3R,4S)ꢀ
2
1
6a. As shown in Figure 1, the theoretical UV and ECD spectra of (3R,4S)ꢀ6a are in agreement
with the experimental spectra of (+)ꢀtransꢀ6, allowing us to assign a (3R,4S) configuration to (+)ꢀ6a
and, consequently, to (+)ꢀ4a.
Figure 1. Absolute configuration determination of compound 6a. Experimental UV and ECD
spectra of (+)ꢀtransꢀ6a (solid line) compared to the PBE0/TZ2P//B97D/TZ2P calculated spectra of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(3R,4S)ꢀ6a (dashed line). Computational details are given in Experimental Section and Supporting
Information.
Results and Discussion
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
Rational Design, Structure-Activity Relationships, and Molecular Modelling Studies
22
The MGLꢀ2 coꢀcrystal structure defined the structural basis for MGL inhibition. This complex
also shed light on the size of the lipophilic portion of the MGL binding pocket, which is of
particular interest for designing novel and selective enzyme ligands. An inꢀdepth examination of
this Xꢀray structure, combined with the analysis of the structural elements of other diphenylmethane
inhibitors, allowed us to develop a novel 3D pharmacophoric model and to rationally design
innovative MGL inhibitors based on a βꢀlactam skeleton (Chart 1, and Figure S1 of the Supporting
Information). In particular, in order to improve the binding properties over current inhibitors such as
1, we chose the transꢀ3,4ꢀdiarylsubstituted βꢀlactam structural motif to obtain the correct distance
and reciprocal orientation of the aromatic rings (see Supporting Information for details). In this
frame, the combination of the 4ꢀfluorophenyl and the methyleneꢀ3,4ꢀdioxyphenyl moieties provides
the key hydrophobic interactions with the enzyme (Figure S1) and guarantees an improved drugꢀ
like profile. The diaryl βꢀlactam system of compounds 4a-h represents a novelty in the field of
MGL inhibition since the potent inhibitors so far described (compounds 1ꢀ3) all share a
diphenylmethaneꢀbased structure.
The experimental data shown in Table 1 confirmed that the βꢀlactam is a key substructure for potent
and selective MGL inhibition. When compound 4a was tested in vitro for its ability to inhibit
human and rat MGL, it resulted to be a very potent inhibitor of hMGL (IC50 = 7.4 nM, Table 1) and
a picomolar inhibitor of rat MGL (rMGL, IC50 = 250 pM). Selectivity towards human recombinant
FAAH was measured and 4a proved to be more than 380ꢀfold selective over human FAAH
(hFAAH) (Table 1). Notably, the two enantiomers showed a stereoselective interaction with hMGL
ACS Paragon Plus Environment

Page 13 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(Table 1), the eutomer (3R,4S)ꢀ(+)ꢀ4a being nearly 7 times more potent than the distomer (3S,4R)ꢀ(ꢀ
)ꢀ4a. The selectivity versus hFAAH was maintained, with both enantiomers still being highly
selective for hMGL (Table 1).
Further in vitro tests revealed that 4a does not significantly interact (IC > 10 ꢀM) with either CB1
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
or CB2 receptors in displacement assays carried out with a high affinity radiolabeled ligand of such
receptors. A timeꢀcourse experiment showed that 4a inhibits hMGL with an IC = 195.8 nM (Table
50
2
) when the assay was conducted without the standard 10 min preꢀincubation, thus suggesting an
irreversible mechanism of action.
Table 1. Inhibition activity of compounds 4a-h, 5a, 6a and 7-9 on human and rat MGL enzymes
a
(IC50, nM), and on human or rat FAAH enzymes (IC50, nM)
IC50 FAAH/
IC50 hMGL
IC50 FAAH/
IC50 hMGL
Cpd
IC50 hMGL
IC50 FAAH
Cpd
IC50 hMGL
IC50 FAAH
O
O
OH
O
O
e
7
25.3 ±
78
N
O
13
13
285.2 ± 30
>10,000
>35
61.43 ± 11
12
O
NO2
4
f
1
c
7
.4 ± 0.7
2880 ±
2
51.28 ±
23
389
e
(
0.25 on
>10,000
>40
287
rMGL)
4
g
4a
c
4050 ±
e
2
3.31 ±
7.2
3678 ±
146
4
.6 ± 0.3
880
158
320
(3R,4S)ꢀ(+)-4a
4h
c
7
740 ±
31.2 ± 1.5
248
>
10,000
ꢀ
ꢀ
650
(3S,4R)ꢀ(–)ꢀ4a
5a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
,410 ±
21
c
>10,000
>7
>10,000
ꢀ
ꢀ
2
6
a
4
b
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
>
10,000
ꢀ
ꢀ
376 ± 154
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
4
c
7
d
97.5 ±
e
>
50,000
>513
10,000
20.7
8
4d
e
1
3.22 ±
.6
2,590 ±
327
196
>10,000
3
9
4
e
a
Tests were performed as described in the experimental section; values are means of three
experiments and all SD are within 10%; hFAAH IC50; COS cells MGL IC50; rFAAH IC50
c
d
e
.
Table 2. Inhibition activity of reference compound 1 and compounds 4a, (3R,4S)ꢀ(+)ꢀ4a and
a
(
3S,4R)ꢀ(–)ꢀ4a on human MGL (IC50, nM).
IC50
h
MGL
(nM)
1
4440 ± 268
195.8 ± 9
4
a
(
3R,4S)ꢀ(+)ꢀ4a
104.4 ± 12
728.5 ± 21
(
3S,4R)ꢀ(–)ꢀ4a
a
The effect of the compounds on MGL activity was measured after a 20 min of incubation with the
enzyme in presence of the specific substrate, as described in the experimental section; values are
means of three experiments and all SD are within 10%
ACS Paragon Plus Environment

Page 15 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We developed and tested a small set of analogues of 4a (4b-h, Table 1) in order to investigate: i) the
role played by the leaving group linked to the piperidine ring (4b-d, Table 1), and ii) the effect of a
small number of focused substitutions aimed at modifying steric hindrance and electronic properties
of the aromatic systems placed on the
βꢀlactam skeleton (4e-h, Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
When studying the mechanism of action of irreversible MGL inhibitors, we applied the concept
16
advanced by Aaltonen and coꢀworkers according to whom the effects of the leaving group are
strictly dependent upon its physicoꢀchemical properties, although steric hindrance may also play a
role. In fact, we found that the effect on potency seems to be related to the pKa of the leaving
group’s conjugated acid (ideal pKa ~ 10). Accordingly, the best profile among compounds 4a-d
was achieved with the triazole (pKa = 10), while the imidazole (pKa = 14) provided a 1500ꢀfold
reduction in potency (4b vs 4a, Table 1), the pꢀnitrophenoxy group (1) (pKa = 7) displayed a 200ꢀ
fold drop of potency (4c vs 4a) and the hexafluoroisopropyl alcohol (pKa = 9) gave an analogue
(
4d) 17 times less potent than 4a.
The modification of steric hindrance and electronic properties of the aromatic systems placed on the
ꢀlactam skeleton was also crucial for potency and selectivity of our MGL inhibitors. The
β
replacement of the 4ꢀfluoro of 4a with a methoxy group led to inhibitor 4e with a slight loss of
inhibitory potency (IC50 = 13.2 nM, Table 1). Compound 4f, in which the methyleneꢀ3,4ꢀ
dioxyphenyl moiety was replaced by a dimethoxyphenyl system, resulted in a 10 times less potent
analogue (IC50 = 61.4 nM, Table 1) and displayed the lowest degree of selectivity toward FAAH
(
IC rFAAH/IC hMGL = 12). Substitution (with methoxy groups) at para and meta positions on
50 50
both the aromatic rings (4g) was detrimental and led to a sensible decrease of MGL inhibition
potency (IC50 = 251.3 nM). By contrast, a decreased steric hindrance, by substitution (with methoxy
groups) at the para positions only (4h), triggered a partial recovery of inhibitory potency (IC50
3.31 nM).
=
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Taken together, these studies point out to the 4ꢀFꢀphenyl system at C3 of the βꢀlactam skeleton and
a methyleneꢀ3,4ꢀdioxyphenyl moiety at C4, combined to the triazole system, (4a) as the optimal
structural arrangement for both potency and selectivity.
Further studies were dedicated to gain insight into the mechanism of action of 4a. We applied a
molecular deconstruction procedure aimed at defining the essential structural determinants
contributing to the high potency of 4a (Chart 2). To this aim, synthetic intermediates 5a and 6a
were also tested in order to assess the importance of the leaving triazole group. In line with the
hypothesized irreversible mechanism of action, both amines 5a and 6a lacked inhibitory activity
towards MGL (Table 1). We also explored the role of the azetidinone system by synthesizing the
azetidine 7. This compound displaying a twoꢀfold drop in inhibitory potency (IC50 = 375.9 nM),
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
underlined the pivotal role played by the
,4ꢀtransꢀdiarylꢀ ꢀlactam system (8) and the piperidine triazole urea disconnection unit 9 were
found to be extremely weak MGL inhibitors (IC50 >10 M). Overall, molecular deconstruction
βꢀlactam amide carbonyl (Table 1). Finally, the isolated
3
β
ꢀ
confirmed the original design hypothesis leading to 4a as a synergistic combination of key
structural elements for the fine tuning of potency.
The interaction of 4a with rMGL was also studied by combining topꢀdown and bottomꢀup
proteomics experiments (Figure 2) (see Supporting Information for details). These studies, which
were consistent with a covalent interaction of 4a with MGL, also indicated that the covalent MGLꢀ
4
a adduct was achieved with 1:1 stoichiometry, under the assay conditions. Furthermore, the
outcome of the bottomꢀup proteomic analyses, performed to map the binding site of 4a in MGL,
indicated that the carbamoyl adduct of the tryptic peptide with 4a (Figure 2C) contains six serine
residues, including the catalytic Ser122. Tandem mass analysis allowed us to unequivocally assign
Ser122 as the site of modification induced by the nucleophilic attack on the urea moiety of 4a
(Figures S2ꢀS5) (see Supporting Information for details).
ACS Paragon Plus Environment

Page 17 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Investigation of the mechanism of action of compound 4a by high resolution mass
spectrometry experiments. Multiply charged mass spectra of intact rMGL naïve (A) or incubated
with 4a (B) and nucleophilic attack of Ser122 on the urea moiety of 4a, with formation of a
carbamate (C).
The stereoselective interaction of 4a with hMGL was investigated by docking studies (Glide,
version 5.7, Schrödinger, LLC, New York, NY, 2011). The eutomer (3R,4S)ꢀ4a shows a different
pattern of interactions with respect to the distomer (3S,4R)ꢀ4a (Figures 3A,B). (3R,4S)ꢀ4a
establishes a series of Hꢀbonds with His269, Ala51, Met123, and with the catalytic Ser122, while
(3S,4R)ꢀ4a only interacts with Ser122 and Ala51. These differences of interactions are translated
into slightly different Glide XP score values ((3R,4S)ꢀ4a ꢀ11.181 kcal/mol, and (3S,4R)ꢀ4a ꢀ10.608
kcal/mol) and in different estimated freeꢀbinding energy values calculated by using Prime
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
MM/GBSA (ꢁGbind(3R,4S)ꢀ4a = ꢀ119.05 kcal/mol, and ꢁGbind(3S,4R)ꢀ4a = ꢀ117.39 kcal/mol).
Notably, the key interactions with His269, Ala51, Met123, the two latter residues being constituents
9
of the oxyanion hole, are relevant for MGL inhibition. Our computational analysis also highlights
the peculiar interactions of both enantiomers into the hMGL binding site, with the catalytic residues
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
(Ser122 and His269 for (3R,4S)ꢀ4a and only Ser122 for (3S,4R)ꢀ4a), and with the oxyanion hole
residues (Ala51 and Met123 for (3R,4S)ꢀ4a and only Ala51 for (3S,4R)ꢀ4a).
Proteomic analysis proved the formation of a carbamoylated adduct of the enzyme with 4a. This
data suggests the formation of an initial tetrahedral intermediate which was also probed, in silico,
23
by a covalent docking approach. A tetrahedral complex generated by using GOLD software for
(3R,4S)ꢀ4a and (3S,4R)ꢀ4a in complex with hMGL (Figure 3C) highlighted that the binding mode
found from classical docking studies (Figures 3AꢀC) is maintained, and (3R,4S)ꢀ4a and (3S,4R)ꢀ4a
similarly establish a Hꢀbond with Gly50 (oxyanion hole residue), deeply positioned into the binding
site. Consequently, we can assume that stereoselectivity of the interaction of 4a enantiomers with
hMGL can rely on the different enantiomer/enzyme reciprocal recognition at the early stage of
interactions, as explained by the docking studies (Figures 3A,B).
hFAAH/hMGL selectivity ratio (Table 1) for (3R,4S)ꢀ and (3S,4R)ꢀ4a was explained by docking
studies on the hFAAH enzyme. (3R,4S)ꢀ4a and (3S,4R)ꢀ4a are not able to completely reach the
active site of the enzyme (Figures 4A,B), as they do not lie in a favorable position into hFAAH
active site, but are accommodated in a pocket located in the upper part of the catalytic site. The XPꢀ
scoring function explains the micromolar affinity for hFAAH (4a Glide score values: ꢀ6.804
kcal/mol and ꢀ6.452 kcal/mol for (3R,4S)ꢀ and (3S,4R)ꢀ4a respectively).
ACS Paragon Plus Environment

Page 19 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Molecular modeling studies on 4a. Docked pose of (3R,4S)-4a (cyan sticks) (A) and of
(3S,4R)-4a (green sticks) (B) in the hMGL binding site (salmon cartoon). The catalytic triad is
represented by orange sticks, while oxyanion hole residues are represented by salmon sticks. Nonꢀ
polar hydrogen atoms have been omitted; (C) Tetrahedral complex of hMGL (salmon cartoon) and
(3R,4S)-4a (cyan sticks) and (3S,4R)-4a (green sticks). Hꢀbonds are indicated by dotted lines in the
same colors of the ligands.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
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. XPꢀGlide predicted binding mode into hFAAH (pink cartoon) binding site of (3R,4S)ꢀ4a
(cyan sticks) (A) and of (3S,4R)ꢀ4a (green sticks) (B). Catalytic triad (Ser241ꢀSer217ꢀLys142) is
indicated by sticks and the Hꢀbonds are represented by black dotted line. Moreover, the crystallized
ligand was superposed to the docking complex, and it is represented by orange lines, in order to
highlight the impossibility of both enantiomers of 4a to reach the catalytic Ser241 of hFAAH. In
fact, (3R,4S)ꢀ4a (A) is able to interact with one of the residues belonging to the catalytic triad
(Ser217) but no interactions with the catalytic Ser241 are observed. For (3S,4R)ꢀ4a (B) no contacts
with the catalytic triad are observed.
Affinity-Based Proteomic Profiling (ABPP) Assay.
Compound 4a was found to potently inhibit MGL with an IC50 of ~10 nM in a competitive ABPP
assay carried out in the rat brain proteome, and did not exhibit crossꢀreactivity with other rat brain
serine hydrolase offꢀtargets up to 500 nM, including FAAH, ABHD6 and ABHD12 (Figure 5A).
Only at concentrations >500 nM some inhibition of lysophopholipase A1/2 was observed (Figure
5
B).
ACS Paragon Plus Environment

Page 21 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. In vitro competitive ABPP of 4a using the serine hydrolaseꢀdirected probe FPꢀRh in the
membrane fraction of the rat brain proteome. (A) 4a potently and selectively inhibits FPꢀRh
labeling of MGL, with lysophopholipase A1/2 (LYPLA1/2) being the only detectable offꢀtarget (B)
(from 500 nM to 1 ꢂM of 4a).
Preliminary in vitro ADME+T Profiling.
Besides its in vitro profile, additional features that contribute to designate 4a as an excellent MGL
inhibitor are the lack of mutagenic effect in the S. typhimurium strains TA98 and TA100 at any of
the concentrations tested (following the assay conditions of the Ames’ test, Figure 6) along with its
experimentally determined physicoꢀchemical features. 4a showed a favorable solubility and
chemicalꢀstability profile at both neutral and acidic pH (Figure S7A). Metabolic stability to
hCYP3A4 and parallel artificial membrane permeability assay confirmed the stability and high
permeability through artificial membranes of 4a (Figures S7B and C).
Brain penetration, evaluated by using a human in vitro blood brain barrier model, is high (Papp =
ꢀ6
4
7 x 10 cm/s ± 6.81) showing that 4a may easily cross the human blood brain barrier. Notably, the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
measured Papp for Lucifer yellow, the non permeant molecule used as marker of endothelium
integrity and coꢀincubated with 100 µM 4a (see Experimental Section for details), is low (Papp =
ꢀ6
9
.48 x 10 cm/s ± 1.68) and is in the same range found when Lucifer yellow is administered alone.
This data indicates that transport of 4a across the blood brain barrier is not due to any toxic effect
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
on the endothelium.
Figure 6. Ames test performed on S. typhimurium TA98 and TA100 strains for compound 4a.
Cardiovascular adverse effects heavily contribute to drug withdrawals from the market and
represent one of the major hurdles in the development of new drugs. To evaluate the potential
cardiovascular toxicity of compound 4a, its effect on cardiac mechanical function and
24
electrocardiogram (ECG) in Langendorffꢀisolated rat hearts was assessed, as previously described.
Under control conditions, left ventricle pressure (LVP) and coronary perfusion pressure (CPP)
values of 60.07 ± 8.96 and 57.45 ± 2.58 mmHg (n = 4) respectively, were obtained. At the
maximum concentration tested (1 ꢂM), 4a significantly increased CPP to 75.19 ± 7.80 mmHg (n =
4
, ** P < 0.01, repeated measures ANOVA and Dunnett’s post test). Moreover, 4a, up to 1 µM, did
not affect surface ECG (Table 3).
ACS Paragon Plus Environment

Page 23 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
At the maximum concentration tested, 4a increases coronary perfusion pressure leaving unaltered
heart contractility and ECG, thus allowing us to exclude cardiac toxicity for this compound.
Table 3. Effects of 4a on HR, RR, PQ, QRS, QT, QTc in Langendorff Perfused Rat Hearts
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
a
M)
HR (BPM)
RR (ms)
PQ (ms)
QRS (ms) QT (ms)
QTc (ms)
(
ꢀ
none 252.78±9.28 238.38±9.23 45.62±4.49 14.26±0.71 75.65±0.87 155.26±2.96
0
0
0
1
.001 254.38±9.83 236.93±9.51 46.03±4.81 14.02±0.71 76.38±1.14 157.13±2.45
.01 254.18±8.35 236.80±7.85 46.46±5.25 13.50±0.29 76.17±0.48 156.30±3.27
.1
252.65±7.66 238.15±7.23 46.18±5.84 13.75±0.48 75.36±1.85 154.60±5.02
255.50±7.36 235.48±6.90 46.67±5.23 13.75±0.47 76.02±1.84 156.78±4.38
Each value represents mean ± SEM (n = 4). HR, frequency; RR, cycle length; PQ, atrioventricular
conduction time; QRS, intraventricular conduction time; QT, duration of ventricular depolarization
and repolarization, i.e., the action potential duration; QTc, corrected QT.
In vivo efficacy of compound 4a
Efficacy of 4a on the Experimental Autoimmune Encephalomyelitis (EAE) Model of MS in
Rodents.
Although no single animal model exists recapitulating all the features of MS, EAE is considered an
25
animal model suitable for studying it, since it shares some important features with human MS,
such as the chronic course and histoꢀpathological evidence of demyelination and
26
neuroinflammation. We therefore evaluated 4a in the myelin glycoprotein 35ꢀ55 (MOG35ꢀ55)ꢀ
induced EAE rodent model (the disease was induced as detailed in the Experimental Section). Drug
prophylactic treatment (4a, 3 mg/kg, i.p.) started at the asymptomatic stage of disease at day 6 postꢀ
immunization (d.p.i.), up to the acute stage of the disease (21 ±.1 d.p.i., when the maximal score
gravity is observed). Administration of 4a clearly influenced the disease progression, as assessed by
the significant lower daily clinical score observed in 4aꢀadministered EAE mice when compared to
vehicleꢀtreated EAE mice starting from 8 d.p.i., as well as by the dramatic change in the cumulative
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
score level (Figure 7). Interestingly, we found that the reduction of the daily clinical score induced
by prophylactic 4a could involve the modulation of the endocannabinoid system. Actually,
administration (starting from 6 d.p.i. for 14 days) of the selective CB1 antagonist/inverse agonist Nꢀ
(piperidinꢀ1ꢀyl)ꢀ5ꢀ(4ꢀiodophenyl)ꢀ1ꢀ(2,4ꢀdichlorophenyl)ꢀ4ꢀmethylꢀ1Hꢀpyrazoleꢀ3ꢀcarboxamide 20
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
7
(
AM251) and the CB2 antagonist/inverse agonist 6ꢀiodoꢀ2ꢀmethylꢀ1ꢀ[2ꢀ(4ꢀmorpholinyl)ethyl]ꢀ
2
8
1Hꢀindolꢀ3ꢀyl](4ꢀmethoxyphenyl)methanone 21 (AM630) , each used at 1 mg/kg, i.p., both
prevented the protecting effect of 4a starting at 12 d.p.i.. Although activation of both CB1 and CB2
2
receptors by endogenous 2ꢀAG could be beneficial for the course of the disease, we found that the
blockade of each one of the two receptors significantly reduced the positive effects exerted 4a,
leading us to conclude that in 4aꢀtreated EAE mice both the two receptors did not compensate, but
synergize one each other at mediating the 4aꢀinduced amelioration of the clinical score. The
respective contribution of CB1 and CB2 in ameliorating the course of the demyelinating disorder in
EAE mice chronically administered 4a deserves further investigation to be clarified.
Changes in spontaneous locomotor activity were not observed in 4aꢀadministered control (nonꢀ
immunized) mice when compared to control (not shown).
ACS Paragon Plus Environment

Page 25 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. In vivo effects of compound 4a on EAE mice. Female C57BL/6J mice were immunized
with the MOG35ꢀ55 peptide and were randomly assigned to the following group: untreated EAE
mice, 4a (3 mg/kg i.p.)ꢀtreated EAE mice; 4a (3 mg/kg i.p.)/20 (1 mg/kg)ꢀtreated EAE mice; 4a (3
mg/kg i.p.)/21 (1 mg/kg)ꢀtreated EAE mice. Mice were administered drugs daily. 20 and 21 were
administered 10 min before 4a. Clinical score is expressed as average (media ± S.E.M.) daily score
(panels A,B,C) or as cumulative score (D). Data are means ± SEM from untreated EAE mice (panel
A, white square, 7 animals), from 4a (3 mg/kg i.p.)ꢀtreated EAE mice (panels A,B,C white circle, 9
animals); from 4a (3 mg/kg i.p.)/20 (1 mg/kg)ꢀtreated EAE mice (panel B, black triangle, 4
animals); from 4a (3 mg/kg i.p.)/21 (1 mg/kg)ꢀtreated EAE mice (panel C, black square, 6 animals).
*P<0.05 vs EAE + vehicle 1 d.p.i.; ^P<0.05 vs EAE + vehicle in the same day; ^^P<0.01 vs EAE +
vehicle in the same day; ^^^P<0.001 vs EAE + vehicle.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Following these encouraging results, we performed histological evaluation of myelin using the
Luxol Fast Blue staining (Figure 8). Pallor of Luxol Fast Blue stain is associated with a reduction in
the density of staining for myelin. Representative images of spinal cord of each experimental group
are shown in Figure 8. In the spinal cord white matter of the control group a normal structural
organization of myelin is evident. By contrast, in EAE mice demyelinated areas are present
(arrows). The EAE mice treated with 4a exhibited myelinꢀdensity staining that was comparable to
that of control animals. Treatment with 4a plus 20 or 21 revealed the presence of some
demyelinated areas in comparison to control animals and to 4aꢀtreated groups, but these areas were
less extended than in EAE group (Figure 8).
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 8. Myelin in the white matter of mice spinal cord. Luxol Fast Blue staining shows areas of
intact myelin (blue) and demyelination (pink). Arrows indicate the demyelinated areas in EAE,
EAE + 4a + 20 and EAE + 4a + 21 groups. Scale bar: 100 µm.
In Figure 9A a panel of representative microglia staining in the spinal cord is shown. Cells were
stained with Iba1 antibody. Immunofluorescence staining for microglia revealed an increase of
ACS Paragon Plus Environment

Page 27 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Iba1ꢀpositive cells/optic field in the white matter anterior fasciculus of EAE mice (247.66 ± 11.50).
The results of the quantitative measurements of microglia density are shown in Figure 9B. 4a
administration in EAE mice caused a reduction in the number of Iba1 positive cells in the area
evaluated that was, at least in part, reverted by coꢀadministration of 20 or 21 (Figure 9A and 9B).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. Microglia positive cells in the white matter of mice spinal cord. A) Representative images
of microglia cells stained with IBA1 (red) and cell nuclei stained with DAPI (blue) in mice white
matter of each experimental group are shown. Scale Bar: 100 ꢂm. B) Quantitative measurements of
the number of Iba1 positive cells in the white matter spinal cord (anterior fasciculus). Statistical
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 68
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
analysis. OneWay ANOVA followed by Bonferroni comparison test: *** P<0.001 vs Control and
EAE + 4a; ^^ P<0.01 vs Control, EAE + vehicle and EAE + 4a.
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
Antinociceptive Effects of 4a in Animal Models of Inflammatory and Neuropathic Pain.
The in vivo antinociceptive profile of 4a was also evaluated in two different animal models of pain.
The compound inhibited both the first and second phase of the nocifensive response to formalin in
mice (model of acute inflammatory pain). The maximum effect was reached at the dose of 0.3
mg/kg (i.p.) (Figure 10A). MGL inhibition may also produce therapeutic actions by impeding the
conversion of 2ꢀAG into arachidonic acid and hence prostaglandins, thus acting in an
2
9
endocannabinoidꢀ and CBRꢀindependent manner. However, we found that the effect of 4a in the
formalin test was mediated by both CB1 and CB2 receptors, since both the selective CB1
antagonist/inverse agonist 20 and the CB2 antagonist/inverse agonist 21 (each used at 1 mg/kg,
i.p.), administered 10 min prior to 4a, antagonized the antinociceptive actions of the compound
(Figure 10B). Importantly, 5 min after the administration of 4a (0.3 mg/kg, i.p.), the levels of 2ꢀAG
in the spinal cord and paw skin of mice coꢀadministered with formalin, measured by LCꢀMS, were
found to be elevated (Table 3, see Experimental Procedures for details). These results suggest that
compound 4a alleviates the pain behavior in the formalin test by elevating endogenous 2ꢀAG levels
(Table 3) thus indirectly activating both CB1 and CB2 receptors, as previously reported for other
10
MGL inhibitors.
Table 3. 2ꢀAG Tissue Levels in Formalin Treated Mice administered (i.p.) with 4a.
a
2ꢀAG (pmol/g)
vehicle/saline 4a/saline vehicle/formalin 4a/formalin
b
Spinal cord
Paw skin
6.4 ± 0.4
6.1 ± 0.3
9.7 ± 1.3
3.8±0.4
6.6 ± 0.7
b
11.0 ± 2.8
10.8 ± 1.5
14.1 ± 1.6
ACS Paragon Plus Environment

Page 29 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
b
means ± SEM, n = 4, analyzed by ANOVA + Bonferroni’s postꢀhoc test;
P<0.05 vs. vehicle/formalin.
Next we evaluated the effect of 4a in a mouse model of nociceptive behavior caused by the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
0
chemotherapeutic agent, oxaliplatin (OXP, Figures 7C,D) (model of neuropathic pain which is
also a common symptom in MS patients). After oral administration, the inhibitor doseꢀdependently
reversed the lowering of the threshold to cold stimuli (cold plate test) induced by OXP, after acute
administration on day 14 of OXP treatment, when changes in pain thresholds were fully established
(
Figure 10C). As shown in Figure 10D, 4a maintained its efficacy after repeated treatments
following a preventive protocol (daily oral administration with OXP). Moreover, when subꢀ
chronically administered, 4a was efficacious 24 h after the last administration. These
antinociceptive properties of 4a are related to conditions of hypersensitivity, such as neuropathic
pain, since 4a did not modify normal pain thresholds evaluated in the hotꢀplate test (Table S1).
ACS Paragon Plus Environment