Chromenopyrazole, a Versatile Cannabinoid Scaffold with in Vivo Activity in a Model of Multiple Sclerosis

Article abstract of DOI:10.1021/acs.jmedchem.6b00397

A combination of molecular modeling and structure-activity relationship studies has been used to fine-tune CB₂ selectivity in the chromenopyrazole ring, a versatile CB₁/CB₂ cannabinoid scaffold. Thus, a series of 36 new derivatives covering a wide range of structural diversity has been synthesized, and docking studies have been performed for some of them. Biological evaluation of the new compounds includes, among others, cannabinoid binding assays, functional studies, and surface plasmon resonance measurements. The most promising compound [43 (PM226)], a selective and potent CB₂ agonist isoxazole derivative, was tested in the acute phase of Theiler's murine encephalomyelitis virus-induced demyelinating disease (TMEV-IDD), a well-established animal model of primary progressive multiple sclerosis. Compound 43 dampened neuroinflammation by reducing microglial activation in the TMEV.

Full text of DOI:10.1021/acs.jmedchem.6b00397

Subscriber access provided by UNIVERSITY OF KENTUCKY
Article
Chromenopyrazole, a Versatile Cannabinoid Scaffold
with in Vivo Activity in a Model of Multiple Sclerosis.
Paula Morales, Maria Gómez-Cañas, Gemma Navarro, Dow P. Hurst, Francisco
J. Carrillo-Salinas, Laura Lagartera, Ruth Pazos, Pilar Goya, Patricia H. Reggio,
Carmen Guaza, Rafael Franco, Javier Fernández-Ruiz, and Nadine Jagerovic
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00397 • Publication Date (Web): 16 Jun 2016
Downloaded from http://pubs.acs.org on June 17, 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 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chromenopyrazole, a Versatile Cannabinoid
Scaffold with in Vivo Activity in a Model of
Multiple Sclerosis.
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
#
§$‡
¶ ‡
£
Paula Morales, María GómezꢀCañas, Gemma Navarro, Dow P. Hurst, Francisco J.
Ψ
#
§$
#
£
CarrilloꢀSalinas, Laura Lagartera, Ruth Pazos, Pilar Goya, Patricia H. Reggio, Carmen
Ψ
¶
§$
#,
Guaza, Rafael Franco, Javier FernándezꢀRuiz, and Nadine Jagerovic *
#
Instituto de Química Médica, Consejo Superior de Investigaciones Científicas, Calle Juan de la
Cierva, 3, Eꢀ28006 Madrid, Spain
§
Instituto Universitario de Investigación en Neuroquímica, Departamento de Bioquímica y
Biología Molecular, Facultad de Medicina, Universidad Complutense de Madrid, Eꢀ28040
Madrid, Spain
$
Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas
(
CIBERNED), and Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), 28040 Madrid,
Spain
¶
Departamento de Bioquímica y Biología Molecular, Facultad de Biología, Universidad de
Barcelona, Eꢀ08028 Barcelona, Spain
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 70
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
£
Department of Chemistry and Biochemistry, University of North Carolina Greensboro,
Greensboro NC27402, USA
Ψ
Grupo de Neuroinmunología Neurobiología Funcional y de Sistemas, Instituto Cajal, Consejo
Superior de Investigaciones Científicas, Madrid, Spain.
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
KEYWORDS. Cannabinoid; CB1; CB2; docking; cAMP; binding; GTP; multiple sclerosis;
HSA; AGP; SPR; in silico ADME.
ABSTRACT. A combination of molecular modeling and structureꢀactivity relationship studies
have been used to fine tune CB selectivity in the chromenopyrazole ring, a versatile CB /CB
2
2
1
cannabinoid scaffold. Thus, a series of 36 new derivatives covering a wide range of structural
diversity have been synthesized and docking studies have been performed for some of them.
Biological evaluation of the new compounds includes, among others, cannabinoid binding
assays, functional studies and surface plasmon resonance measurements. The most promising
compound [43 (PM226)], a selective and potent CB agonist isoxazole derivative, was tested in
2
the acute phase of Theiler’s murine encephalomyelitis virusꢀinduced demyelinating disease
(
TMEVꢀIDD), a wellꢀestablished animal model of primary progressive multiple sclerosis.
Compound 43 dampened neuroinflammation by reducing microglial activation in the TMEV.
ACS Paragon Plus Environment

Page 3 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
The endocannabinoid system (ECS) composed of at least two cannabinoid Gꢀprotein coupled
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
1
,2
receptors (CB and CB receptors),
endogenous ligands such as anandamide and 2ꢀ
1
2
arachidonoylglycerol, and the enzymes for their biosynthesis and degradation, is involved in
3
–6
numerous physiological and pathological conditions. Therefore, for several years the ECS has
been considered a potential therapeutic target for the clinical management of various disorders
including inflammatory and neuropathic pain, neurological pathologies and cancer among
7
others. Some few diseases can be treated nowadays with cannabinoidꢀbased medicines, mainly,
9
plant derived compounds. A mixture of synthetic tetrahydrocannabinol (ꢁ ꢀTHC) [Figure 1] and
9
nabilone, a ꢁ ꢀTHC synthetic analogue, can be prescribed in several countries as antiemetic
8
,9
10
drugs for chemotherapyꢀinduced nausea and vomiting, and for anorexia treatment in patients
9
with AIDS. A combination of ꢁ ꢀTHC and cannabidiol is used for the symptomatic relief of
spasticity in adults suffering multiple sclerosis, and as an adjunctive analgesic treatment in adult
1
1
patients with neuropathic pain or cancer. Rimonabant, a CB receptor antagonist/inverse
1
1
2
agonist, commercialized in 2006 for the management of obesity was withdrawn few years later
because of increase of depression, anxiety, headache, and suicidal thoughts. Even though
CB /CB receptor agonists are currently in the forefront of clinical research for different
1
2
13
applications, there is an increasing interest in exploiting novel pharmacological strategies.
Whereas CB receptor is abundantly expressed in the central nervous system, CB receptor is
1
2
mainly associated with the peripheral immune system. The CB receptors are also expressed in
2
14–18
the central nervous system in microglia and neuronal cells.
Therefore, CB receptor
2
selective agonists exhibit a promising therapeutic potential for treating various pathologies while
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 70
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
avoiding the adverse psychotropic effects related to the modulation of the CB receptors in the
1
1
9
brain. Different therapeutic applications have been proposed for CB receptor agonists.
2
2
0
21
Treatment of neuropathic and osteoarthritis pain,
and diagnosis and treatment of
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
22
23
osteoporosis are currently in advanced development. Activation of CB receptors offers an
2
attractive opportunity for treating neuroinflammatory events in neurological disorders, such as
1
3,24–31
multiple sclerosis, cerebral ischemia, Alzheimer’s and Parkinson’s diseases.
Cannabinoidꢀ
based therapies are already approved for multiple sclerosisꢀassociated symptoms such as pain
and spasticity. The development of CB receptor ligands to regulate neural inflammation and
2
30,32
neurogenesis in multiple sclerosis is much more recent.
Different structures endowed with CB receptor affinity and partial or full selectivity were
2
identified mainly by pharmaceutical companies through high throughput screening as reviewed
33
recently by Han et al. In an attempt to target the CB type receptor, we decided to undergo
2
structureꢀactivity relationship studies around the chromenopyrazole scaffold. In previous studies,
we had identified chromenopyrazoles as a novel cannabinoid scaffold which leads to nonꢀ
34
psychoactive and selective CB agonists with peripheral antinociceptive properties (Figure 1).
1
On the basis of these previous findings, structural modifications around the chromenopyrazole
have been explored to achieve CB receptor selectivity. From combined pharmacological and
2
modeling studies, 36 new compounds covering structural diversity have been synthesized and
evaluated. Among them, the most promising, 43, a fully selective CB agonist, has shown
2
activity in the acute inflammatory phase of TMEVꢀIDD a wellꢀestablished animal model of
3
5
primary progressive multiple sclerosis .
ACS Paragon Plus Environment

Page 5 of 70
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
9
9
Figure 1. Structures of ∆ –tetrahydrocannabinoid (∆ ꢀTHC) and the previously identified CB
1
34
fully selective chromenopyrazoles.
Synthesis
Compounds 5−39 were synthesized as depicted in Scheme 1 from 7ꢀ(1,1ꢀdimethylheptyl)ꢀ5ꢀ
hydroxyꢀ3ꢀ(hydroxymethylen)ꢀ2,2ꢀdimethylchromanꢀ4ꢀone (4). At the outset, chromanone 4 was
prepared by demethylation of the commercially available 5ꢀ(1,1ꢀdimethylheptyl)ꢀ1,3ꢀ
3
4
dimethoxybenzene (1) followed by treatment with 3,3ꢀdimethylacrylic acid and α–formylation.
Then, condensation of the βꢀketoaldehyde 4 with the appropriate hydrazine gave the
corresponding chromeno[4,3ꢀc]pyrazolꢀ9ꢀoles 5−14 following a procedure previously published
3
4
by us for 5−9. Two regioisomers, N1ꢀ and N2ꢀsubstituted pyrazoles, can be form and their
1
13
structure could be attributed thanks to the combined H, C, HSQC and HMBCꢀNMR spectra.
From alkylhydrazines, the regioisomers 7a/7b and 10a/10b were isolated with relative ratios
varying from 1/5 to 2/1 (N1/N2). Condensation with hydroxymethylhydrazine only generated
N2ꢀpyrazole substitution (11) probably due to steric and electrostatic repulsions. In the case of
arylhydrazines and cyclohexylhydrazine, the N1ꢀregioisomers (8, 9, 12, and 13) were formed.
Preparation of compounds 15−39 was achieved by phenolic alkylation of the appropriate
chromenopyrazoles (5−9) with the corresponding alkyl halide.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 1. Synthesis of chromenopyrazoles 5−39.
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
Reagents and conditions: (i) BBr , CH Cl , overnight, from 0 ºC to rt, 92 %; (ii) 3,3ꢀdimethylacrylic acid,
methanesulfonic acid, P O , 8 h, 70 ºC, 81 %; (iii) a) NaH, THF, MW, 25 min, 45 ºC; b) ethyl formate, MW, 25
min, 45 ºC, 76 %; (iv) corresponding hydrazine, EtOH, 1ꢀ4 h, 40 ºC, 36ꢀ50 %; (v) a) NaH, THF, 10 min, b) 1ꢀbromo
or iodoalkane, 8ꢀ72 h, reflux, 36ꢀ50 %.
3
2
2
2
5
To explore the structureꢀactivity relationships (SAR) on the phenol substitution,
conformationally restricted analogues (40 and 41) adopting a semiꢀplanar geometry were
ACS Paragon Plus Environment

Page 7 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
synthesized as described in Scheme 2. Cyclodehydration of chromenopyrazoles 30 and 31 with
phosphorus pentoxide provided the desired condensed cyclic ethers 40 and 41.
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
a
Scheme 2. Synthesis of pyranoꢀchromenopyrazoles 40 and 41.
a
Reagents and conditions: (i) P O , toluene, 3 h, reflux, 36ꢀ66 %.
2
5
Further exploration of chromenopyrazoles as scaffold led us to consider the bioisosteric
replacement of the pyrazole into an isoxazole moiety. For that purpose, condensation of βꢀ
diketone 4 with isoxazole 42 was efficiently achieved upon reaction with hydroxylamine
hydrochloride as shown in Scheme 3. Phenolic alkylation of the chromenoisoxazole 42 yielded
the methoxy (43) and the hydroxypropoxy (44) chromenoisoxazole derivatives (Scheme 3).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 3. Synthesis of chromenoisoxazoles 42−44.
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
Reaction conditions: (i) hydroxylamine hydrochloride, acetic acid, 15 min, reflux, 91 %; (ii) a) NaH, THF, 10
min, 0 ºC; b) 1ꢀbromo or iodoalkane, 8ꢀ72 h, reflux, 36ꢀ50 %.
Results and Discussion
Receptor Affinity
Chromenopyrazoles and chromenoisoxazoles 5−44 were evaluated for their ability to compete
3
the binding of the radiolabeled nonꢀselective agonist [ H]CP55,940 to human CB (hCB R) and
1
1
CB (hCB ) cannabinoid receptors. As a source for these receptors, commercial membrane
2
2
preparations of HEK293 cells stably expressing the respective receptor type were used. Tables
1
−3 list the affinity constants values (K ) for hCB and hCB receptors obtained by fitting data
i
1
2
from competition curves.
34
In prior studies we have reported fully selective CB agonists 5−6 and 9. On the basis of
1
these previous findings, we have explored the SAR around the chromenopyrazole scaffold in
order to achieve affinity and selectivity for the other cannabinoid receptor type CB . In the
2
course of these studies, molecular modeling (discussed below) helped us to identify the structural
features necessary to fineꢀtune CB affinity and selectivity. The best results in terms of CB2
2
affinity were obtained for the chromenopyrazole 34 and the chromenoisoxazole 42 with K < 6
i
3
6
nM. In what concerns CB receptor affinity and selectivity, the chromenoisoxazole 43 (PM226)
2
showed CB selectivity with a K value of 12.8 ± 2.4 nM.
2
i
ACS Paragon Plus Environment

Page 9 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In order to highlight SAR, the results may be discussed taking into account structural
modifications on the pyrazole substituent (Table 1), on the phenol substituent (Table 2) and on
the chromenopyrazole scaffold (Table 3).
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
Table 1. Binding affinity of chromenopyrazoles 5−14 and reference cannabinoids for hCB and
1
hCB cannabinoid receptors.
2
1
CB
K (nM)
1
CB
K (nM)
2
CB
Select
1
CB
2
Select.
Compd
R
a
a
.b
c
i
i
3
4
5
H
28.5 ± 23.7
14.2 ± 2.9
4.5 ± 0.6
>40000
>40000
>1400
ꢀ
3
4
6
b
N2ꢀMe
>2500
ꢀ
ꢀ
3
4
7
7
a
N1ꢀEt
>40000
>8000
3
4
b
N2ꢀEt
18.6 ± 2.9
514 ± 205
5.2 ± 4.3
>40000
>2000
ꢀ
34
8
N1ꢀ3,4ꢀdiClPh
N1ꢀ2,4ꢀdiClPh
270
ꢀ
1.9
ꢀ
34
9
>40000
>7500
1
0a
N1ꢀ(CH
N2ꢀ(CH
2
2
)
)
2
2
OH
OH
19.1 ± 8.9
54.4 ± 8.1
218.1 ± 40.5
>40000
366 ± 169
39.6 ± 7.1
59.4 ± 31.2
>40000
19.1
ꢀ
10b
ꢀ
ꢀ
ꢀ
ꢀ
1.4
3.7
ꢀ
1
1
N2ꢀCH OH
2
12
N1ꢀ3,5ꢀdiFPh
N1ꢀ3ꢀOMePh
1
3
6440 ± 655
562.0 ± 13.2
11.5
21.5
14
N1ꢀCy
ꢀ
1140 ± 190
7.3 ± 0.9
53.7 ± 11.8
>40000
ꢀ
SR141716
>5400
WIN55,212ꢀ2
ꢀ
45.6 ± 8.6
3.7 ± 0.2
ꢀ
12.3
a
3
K : Affinity constants. Values obtained from competition curves using [ H]CP55940 as
i
radioligand for hCB and hCB cannabinoid receptors and are expressed as the mean ± SEM
1
2
b
c
of at least three experiments. K (CB )/K (CB ) selectivity ratio. K (CB )/K (CB ) selectivity
i
1
i
2
i
2
i
1
ratio.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 70
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
The nature of the pyrazole substituent clearly influenced the affinity for cannabinoid receptors
(Table 1). NꢀMethyl, Nꢀethyl or unsubstituted chromenopyrazoles (5−7) bearing a free phenol
group showed significant to high CB affinity and selectivity independently of the substitution
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
4
position on the pyrazole ring (N1 or N2). NꢀCyclohexyl substitution (14) drastically changed
the selectivity profile exhibiting preference for the CB receptor. The role played by the
2
cyclohexyl moiety for binding to the CB receptor is discussed later in the modeling section. The
2
presence of Nꢀhydroxyalkyl groups (10a, 10b and 11) resulted in moderate to high affinities for
both receptors. Aromatic Nꢀsubstituents (8, 9, 12, and 13) did not result beneficial for CB2
affinity.
Regarding the Oꢀalkylated chromenopyrazoles (Table 2), in general, the loss of the free
phenolic group led to compounds displaying high to moderate CB affinity and selectivity.
2
Modeling studies on the hydroxychromenopyrazole 7b and methoxychromenopyrazole 18
allowed determining critical interactions in the CB and CB receptor binding sites (discussed
1
2
below).
It is interesting to note that the condensation of a 2,4ꢀdihydropyran to the chromenopyrazoles
structure (40 and 41), which was performed for restricted conformational issues, elicited
excellent CB selectivity even though they showed moderate affinity. In the Oꢀalkylated series
2
(
Table 2), the pyrazole substitution contributes to determine affinity and selectivity issues.
Unsubstituted pyrazoles 15, 22, 28 and 34 revealed high affinity towards both receptors
regardless of the nature of the phenol substitution. NꢀAryl substitution (19, 20, 26, 27, 32, and
33) led to inactive derivatives while Nꢀmethyl or ethyl substituents (16, 18, 23, 25, 28, 31, 35,
and 39) resulted in high to moderate CB affinity and selectivity independently of the nature of
2
ACS Paragon Plus Environment

Page 11 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the phenol substituent. These results suggest that CB receptor binding site does not tolerate
2
bulky aromatic substituents on the pyrazole ring.
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
Table 2. Binding affinity of chromenopyrazoles 15−41 for hCB and hCB cannabinoid
1
2
receptors.
1
2
CB
K (nM)
1
CB
K (nM)
2
CB
Select.
1
CB
2
Select.
Compd
R
R
a
a
b
c
i
i
1
5
H
Me
272 ± 75
87.1 ± 10.6
ꢀ
3.1
1
1
1
1
2
6
7
8
9
0
N2ꢀMe
N1ꢀEt
Me
Me
Me
Me
Me
Et
4159 ± 542
5040 ± 670
2930 ± 470
2324 ± 327
1693 ± 239
>40000
66.4 ± 13.29
159.5 ± 31.2
92.6 ± 17.1
2256 ± 499
1493 ± 272
3545 ± 89
ꢀ
ꢀ
62.6
31.6
31.8
ꢀ
N2ꢀEt
ꢀ
N1ꢀ3,4ꢀdiClPh
N1ꢀ2,4ꢀdiClPh
N1ꢀ3,4ꢀdiClPh
H
ꢀ
ꢀ
ꢀ
21
ꢀ
>11.3
ꢀ
2
2
Bn
Bn
Bn
Bn
Bn
22.4 ± 4.1
93.3 ± 29.5
4.2
ꢀ
23
N2ꢀMe
702.4 ± 98.6 208.8 ± 39.5
613.7 ± 206.9 295.9 ± 53.6
671.3 ± 166.1 212.2 ± 49.4
3.3
2.0
3.1
>10.7
2
4
N1ꢀEt
ꢀ
25
N2ꢀEt
ꢀ
2
2
6
7
N1ꢀ3,4ꢀdiClPh
>40000
3740 ± 297
ꢀ
N1ꢀ2,4ꢀdiClPh
H
Bn
>40000
450.4 ± 9.9
>40000
>40000
26.0 ± 7.1
364.0 ± 68.9
440 ± 145
97.4 ± 9.9
>40000
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
17.3
>109.9
3.6
28
(CH
(CH
(CH
(CH
2
2
2
2
2
2
2
)
)
)
)
)
)
)
3
3
3
3
3
3
2
OH
OH
OH
OH
OH
OH
OH
2
9
N2ꢀMe
N1ꢀEt
30
1613 ± 284
>40000
3
3
3
3
1
2
3
4
N2ꢀEt
>410.7
ꢀ
N1ꢀ2,4ꢀdiClPh (CH
N1ꢀ3,4ꢀdiClPh (CH
>40000
>40000
>40000
ꢀ
H
(CH
64.8 ± 18.0
3.6 ± 0.7
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 70
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
3
3
3
5
6
7
N2ꢀMe
N2ꢀEt
N2ꢀMe
N2ꢀEt
(CH
(CH
2
2
)
)
2
2
OH
OH
1086 ± 198
6512 ± 714
1482 ± 221
657 ± 159
39.8 ± 24.9
210.6 ± 93.2
77.3 ± 0.87
87.1 ± 14.2
ꢀ
ꢀ
ꢀ
ꢀ
27.2
31.0
19.1
7.5
(CH
(CH
)
Br
Br
2
2
3
38
)
)
3
2
3
9
N1ꢀEt
(CH
2
Br
1331 ± 320
78.7 ± 11.3
ꢀ
16.9
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
2
3
4
5
6
7
8
9
0
4
4
0
1
N1ꢀEt
N2ꢀEt
ꢀ
ꢀ
>10000
>40000
563.8 ± 13.1
121.6 ± 43.5
ꢀ
ꢀ
>17.7
>330.5
a
3
K : Affinity constants. Values determined from competition curves using [ H]CP55940 as
i
radioligand for hCB and hCB cannabinoid receptors and are expressed as the mean ± SEM of
1
2
b
c
at least three experiments. K (CB )/K (CB ) selectivity ratio. K (CB )/K (CB ) selectivity ratio.
i
1
i
2
i
2
i
1
Table 3. Binding affinity of chromenoisoxazoles 42−44 for hCB and hCB cannabinoid
1
2
receptors.
2
CB
K (nM)
1
CB
2
CB
Select.
1
CB
2
Compd
R
a
a
b
c
K (nM)
Select.
i
i
4
2
H
15.4± 12.2
>40000
5.3 ± 0.8
ꢀ
2.9
43
Me
(CH
12.8 ± 2.4
ꢀ
ꢀ
>3100
5.1
4
4
2
)
3
OH 332.6 ± 143.9 65.5 ± 21.8
a
K : Affinity constants. Values obtained from competition curves using
H]CP55940 as radioligand for hCB and hCB cannabinoid receptors and are
expressed as the mean ± SEM of at least three experiments. K (CB )/K (CB )
selectivity ratio. K (CB )/K (CB ) selectivity ratio.
i
3
[
1 2
b
i
1
i
2
c
i
2
i
1
Bioisosteric replacement of the pyrazole by an isoxazole (Table 3) led to very potent
cannabinoid ligands. Chromenoisoxazole 42 exhibited high affinity for both CB and CB2
1
receptors with K values in the low nanomolar range (K CB = 15.4 nM; K CB = 5.3 nM). The
i
i
1
i
2
fact that this isoxazole derivative (42) but not its pyrazole analog (5) binds to the CB receptor
2
has been instrumental to propose a docking mode of these compounds to the active CB receptor
2
ACS Paragon Plus Environment

Page 13 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
model (CB R*) as discussed in the modeling section. Furthermore, these results and the data
2
obtained for Oꢀalkylated chromenopyrazoles prompted us to synthesize the
methoxychromenoisoxazole 43 which resulted to be the most potent and CB selective ligand in
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
this study, being over 3000ꢀfold selective for CB receptor with an affinity constant of 12.8 nM.
2
This CB affinity was similar to other selective synthetic CB receptor agonists, [(1R,2R,5R)ꢀ2ꢀ
2
2
[
2,6ꢀdimethoxyꢀ4ꢀ(2ꢀmethyloctanꢀ2ꢀyl)phenyl]ꢀ7,7ꢀdimethylꢀ4ꢀbicyclo[3.1.1]heptꢀ3ꢀenyl]
3
7
methanol (HUꢀ308) (K = 22.7 nM) or (6aR,10aR)ꢀ3ꢀ(1,1ꢀdimethylbutyl)ꢀ6a,7,10,10aꢀ
i
38
tetrahydro ꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyran (JWHꢀ133) (K = 3.4 nM). Moreover, the
i
selectivity of compound 43 (CB /CB > 3125) for CB receptor was higher than JWHꢀ133
1
2
2
(
CB /CB = 200) or HUꢀ308 (CB /CB = 450).
1 2 1 2
Potency in Functional Assays
Compounds showing high affinity for CB receptor (K values under 160 nM) (16−18, 22, 28,
2
i
31, 34, 35, 37, 39, 41, 42, 43, and 44) were selected for functional evaluation in cAMP
determination experiments. To further assess CB receptor activity through a different outcome,
2
GTPγS binding assays of 34, 42−44 were performed.
Functional properties of CB ligands (16−18, 22, 28, 31, 34, 35, 37, 39, and 41−44) were
2
appraised through cAMP assays using HEK293 cells stably expressing hCB R and treated with
2
forskolin to activate adenyl cyclase. Effects of tested compounds on forskolinꢀstimulated cAMP
levels were determined in a preliminary screening at two concentrations, at 200 nM and at 1 ꢂM
(Figure 2). The cannabinoid reference compound, 2ꢀ[(1R,2R,5R)ꢀ5ꢀhydroxyꢀ2ꢀ(3ꢀ
39
hydroxypropyl) cyclohexyl]ꢀ5ꢀ(2ꢀmethyloctanꢀ2ꢀyl)phenol (CP55,940) was also tested in this
assay as reference. Compounds 17, 18, 22, 28, 34, 37, 39, and 41−44 displayed low to high
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 70
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
inhibition of forskolinꢀstimulated cAMP accumulation at both concentrations. The tested
compounds were also screened in normal HEK293 cells at the same two concentrations to
confirm that these effects were CB receptorꢀmediated (Supporting Information).
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
Figure 2. cAMP screening in HEK293ꢀCB R cells. Results are expressed as percent of forskolin
2
(Fk)ꢀstimulated cAMP accumulation at a concentration of 200 nM and 1 ꢂM of 16−18, 22, 28,
31, 34, 35, 37, 39, or 41−44. All data result from at least three independent experiments,
performed in triplicates.
Full concentration−response curves were measured to determine IC and maximum inhibition
5
0
values for the most potent compounds (34 and 41−44); the other compounds (17, 18, 22, 28, 37,
and, 39) behave as partial or weak agonists. As shown in Table 4, 34 and 41−44 are potent CB2
agonists with IC values in the nanomolar range being chromenoisoxazole 43 the most potent
5
0
and efficient (IC = 4.2 nM).
5
0
ACS Paragon Plus Environment

Page 15 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. Functional potencies of 34 and 41−44 and reference cannabinoids at CB receptor
2
determined by measuring the decrease in forskolinꢀstimulated cAMP levels in HEK293ꢀCB2R
3
5
cells or by [ S]ꢀGTPγS binding to hCB receptor.
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
3
5
cAMP assay
[ S]ꢀGTPγS assays
(
CB R)
(CB R)
2
2
Compd
a
b
c
d
IC (nM)
Emax (%)
EC (nM)
Emax (%)
50
50
3
4
4
1
21.6 ± 1.5
25.5 ± 2.0
134.0 ± 2.3
4.2 ± 1.5
14.0 ± 1.9
8.3 ± 1.5
81.8 ± 1.7
n.d.
95 ± 6
98 ± 11
98 ± 9
101± 10
91± 10
100 ± 9
98 ± 11
n.d.
95.1 ± 7.2
n.d.
110 ± 17
n.d.
42
50.2 ± 24.9
38.6 ± 6.7
114 ± 37
98 ± 8
4
3
4
4
539.6 ± 208.1 96 ± 15
CP55,940
JWH133
HU308
n.d.
n.d.
n.d.
n.d.
64.5 ± 1.6
91 ± 7
a, c
IC50 nd EC50 values were calculated using nonlinear regression
analysis. Data are expressed as the mean ± SEM of at least three
independent experiments, each one run in triplicates. Forskolin
b
stimulated cAMP levels were normalized to 100%. Emax ids the
maximum inhibition of forskolin stimulated cAMP levels. Emax
:
maximal agonist effect, determined using nonlinear regression
analysis.
3
5
CB agonist properties of 34 and 42−44 were confirmed by [ S]ꢀGTPγS binding assays
2
performed in commercial CB receptorꢀcontaining membranes. The EC values obtained from
2
50
the respective concentrationꢀresponse curves are collected in Table 4. Agonist potencies of 34
35
and 42−44 vary from [ S]ꢀGTPγS (EC : 38.6−539.6 nM) to cAMP experiments (EC :
50
50
4.2−134.0 nM) experiments. However, in both assay types, chromenoisoxazole 43 standed out
the most potent selective CB receptor ligand of the series.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 70
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
CB receptor ligands (16, 31, and 35) that in the preliminary screening exhibited no effect on
2
forskolinꢀstimulated cAMP accumulation were tested for CB R antagonism in cAMP assays.
2
Compounds 16 and 35 were able to antagonize the effect of the cannabinoid agonist CP55,940
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
(
100nM and 1 µM) at a concentration of 200 nM, whereas 31 behaved as CP55,940 antagonist
only at the highest concentration (Figure 3).
(A)
(B)
(C)
(D)
ACS Paragon Plus Environment

Page 17 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. Effect of compounds 16, 31, and 35 on CP55,940ꢀinduced inhibition of cAMP
accumulation in HEK293ꢀCB2R cells. (A) 16 at 200 nM; (B) 34 at 200 nM; (C) 31 at 200 nM;
(D) 31 at 1 µM. All data result from at least three independent experiments, performed in
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
triplicates. Data were assessed by oneꢀway analysis of variance (F(5,43)=20.57, p<0.0001 for 16;
F(5,42)=17.92, p<0.0001 for 35; F(5,33)=27.16, p<0.0001 for 31 (C); F(5,33)=22.03, p<0.0001
for 31 (D); *p<0.05, **p<0.01,***p<0.005 compounds alone versus control (Fk cAMP
accumulation); &p<0.05, &&p<0.01, &&&p<0.005 CB agonist versus CB agonist + 16, 31 or
2
2
35).
Table 5. [35S]ꢀGTPγS binding of compounds 10b, 22 and 42 and the reference cannabinoid
WIN55,212ꢀ2 to hCB receptor.
1
35
[
S]ꢀGTPγS assays
(
CB R)
1
Compd
a
b
EC (nM)
Emax (%)
97 ± 52
50
1
0b
298.8 ± 54.2
22
42
191.0 ± 90.8 234 ± 79
15.8 ± 8.6
44 ± 30
196 ± 111
153 ± 70
WIN55,212
a
EC50 values were calculated using
nonlinear regression analysis. Data are
expressed as the mean ± SEM of at least
three independent experiments, each one
b
run in triplicates. Emax: maximal agonist
effect,
determined
using
nonlinear
regression analysis.
In what concerns the CB receptor activity, only the effect of the new compounds showing the
1
highest affinity for CB receptor (10b, 22, and 42) was determined. Potencies of 10b, 22, and 42
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 70
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
were evaluated through GTPγS binding assays performed in membranes extracted from HEK293
cells stably expressing the human CB receptor (Table 5). Tested compounds, 10b, 22 and 42,
1
35
produced an increase of basal [ S]ꢀGTPγS binding, being 42 at least 10 fold more potent than
0b and 22.
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
Molecular Modeling
During the course of our studies, cannabinoid selectivity and structural patterns led us to select
molecules 5, 7b, 18, and 42 for docking studies. Given the high CB selectivity of the
1
chromenopyrazole 5, it was surprising that the corresponding isoxazole derivative 42 showed
high affinity for both cannabinoid receptors. Therefore, we report here docking studies of both
compounds (5 and 42) in the refined CB and CB active state models. We also considered
1
2
interesting that the replacement of the hydroxyphenyl by a methoxyphenyl that led to full CB2
selectivity in the chromenopyrazole series. Varying degrees of CB selectivity and affinity were
2
3
7
8
reported in the literature for some classical cannabinoids such as the HUꢀ308 and methoxyꢀꢁ ꢀ
40
THC derivatives. However, these 0ꢀmethylations have not been fully explored in terms of
interactions with the binding site. Therefore, the CB R selective chromenopyrazole 7b was
1
studied in relation to its methoxy derivative 18 through docking studies.
Firstly, conformational analysis of 5, 7b, 18, and 42 was performed to determine the global
minimum energy conformers (Supporting Information). In what concerns the NꢀHꢀ
chromenopyrazole 5, this pyrazole can exist as a mixture of two tautomeric forms. However, we
only considered the N2ꢀHꢀtautomer based on our previous studies concerning annular
tautomerism (OH···N and/or NH···O) of hydroxychromenopyrazoles, in which the tautomer
41
OH···N was shown to be the predominant species in solution. We have then calculated the
ACS Paragon Plus Environment

Page 19 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
electrostatic potential maps of the minimum energy conformers of 5, 7b, 18, and 42 (Figure 4).
The phenolic hydroxyl group revealed to be the most negative electrostatic potential region (in
red) of 5 (CB ) and 7b (CB ). Chromenoisoxazole 42 (CB /CB ) showed two electron rich hot
1
1
1
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
spots (in red) originated by the phenolic hydroxyl group and by the isoxazole nitrogen, the latter
being the most electronegative region of the molecule. As expected, the
methoxychromenopyrazole 18 (CB ) displayed a weaker electronegative region due to the low
2
exposure of the free lone pair of electrons of the methylated phenolic oxygen. A positive
electrostatic potential at the corresponding region has also been detected for the highly CB2
selective ligand 43 (Supporting Information).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 70
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
Figure 4. The molecular electrostatic potential maps of the minimum energy conformations of 5,
7b, 18 and 42 are illustrated here. The electrostatic potential scale (in kJ/mol) is provided as a
color scale. This scale is from blue (most electropositive) to red (most electronegative).
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
The global energy minima of 5, 7b, 18 and 42 were docked using a model of the CB and of
1
42
the CB in their active state (CB R*; CB R*).
These models include the extracellular and
2
1
2
intracellular loops, the Nꢀterminus (truncated in CB R) and the Cꢀterminus, including the
1
intracellular helix portion of each receptor, termed Helix 8. Docking studies were performed in
4
3
44,45
the same binding site described for HU210 in the CB R* model and for AM841
in the
1
CB R* model.
2
Docking studies of 5 and 42 in CB R*. As illustrated in Figure 5, the energyꢀminimized
1
hydroxychromenopyrazole 5/CB R* complex shows two main binding site anchoring
1
interactions. The phenolic oxygen of 5 is engaged in hydrogen bond with K3.28(192) [hydrogen
bond (NꢀO) distance = 2.75 Å and (NꢀHꢀꢀO) angle = 151°]. The N1ꢀpyrazole nitrogen establishes
a hydrogen bond with serine S7.39(383) [hydrogen bond (NꢀO) distance = 3.02 Å and (OꢀHꢀꢀN)
angle = 140°].
ACS Paragon Plus Environment

Page 21 of 70
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. Binding site of 5 (panel A) and 42 (panel B) in the CB R* model. Side view of
1
ligand/receptor complexes as if looking through TMH1 (not displayed). Hydrophilic interacting
residues are represented with yellow carbons. Hydrogen bonds are shown with yellow dash lines.
Select hydrophobic interacting residues, F2.57 and L7.43, are displayed with green carbons.
Docking studies of 42 in the CB R*. Docking 42 in CB R* revealed a similar occupation of the
1
1
binding site with hydrogen bonds involving K3.28(192) [hydrogen bond (NꢀO) distance = 2.82 Å
and (NꢀHꢀꢀO) angle = 153°] and S7.39(383) [ hydrogen bond (NꢀO) distance = 2.77 Å and (OꢀHꢀ
ꢀN) angle = 130°] as key residues (Figure 5, right). It is interesting to note that an additional
hydrogen bond between the pyran oxygen and cysteine C7.42(386) was revealed in the
42/CB R* complex [hydrogen bond (SꢀO) distance = 3.19 Å and (SꢀHꢀO) angle = 167°].
1
Docking studies of 5 and 42 in CB R*. The CB R* model contains a salt bridge between the
2
2
aspartic acid D275 in the ECꢀ3 loop and lysine K3.28(109). Docking studies of 5 revealed a
steric clash between the pyrazole moiety of the structure and the lysine involved in the ionic lock
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 70
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
(
Figure 6, left) and in agreement with the experimental pharmacological data (K (CB R) > 40
i
2
µM), whereas the energy minimized 42/CB R* complex presents two main interactions (Figure
2
6
, right), a hydrogen bond between the isoxazole nitrogen and K3.28(109) [hydrogen bond (NꢀO)
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
distance = 2.86 Å and (NꢀHꢀꢀO) angle = 157º] and a hydrogen bond involving the phenolic
oxygen of 42 and S6.58(268) [hydrogen bond (OꢀO) distance = 2.63 Å and (OꢀHꢀꢀO) angle =
1
70º].
Figure 6. Binding site of 42 (panel B, in purple) in the CB R* model. Side view of
2
ligand/receptor complex as if looking through TMH1 (not displayed). Hydrophilic interacting
residues are represented with yellow carbons. Hydrogen bonds are shown with yellow dash lines.
A selected hydrophobic interacting residue, V3.32, is displayed with green carbons. Structure of
5
(panel A, in cyan) was superimposed on the 42ꢀCB R* complex. The magenta circle indicates
2
van der Waals steric overlap of 5 with K3.28, which forms a salt bridge with D(275) in the
CB R* binding site.
2
ACS Paragon Plus Environment

Page 23 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Docking studies of 7b and 18 in CB R*. The hydroxychromenopyrazole 7b and the
1
methoxychromenopyrazole 18 were compared at CB R* binding site (Supporting Information).
1
The 7b/CB R* complex presents two hydrogen bonds with K3.28(192) that involve the phenolic
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
oxygen and the pyrazole N1 nitrogen which also forms a hydrogen bond with S7.39(383).
Consistent with its poor CB affinity, 18 was unable to form any hydrogen bond with any residue
1
of the receptor. Indeed, the methoxy group of 18 shows steric overlap with the phenylalanine
F2.57(170) and the leucine L7.43(387) inducing low accessibility of the lone pairs of electrons of
the phenolic oxygen to K3.28(192) (showed in the Supporting Information).
Docking studies of 7b and 18 in CB R*. Figure 7 illustrates the CB R* docking studies of 7b
2
2
and 18. As previously shown for 5/CB R* complex, the hydroxychromenopyrazole 7b exhibits a
2
steric clash between the pyrazole moiety and the ionic lock formed by D275 and K3.28(109)
with an additional major steric overlap with F7.35(281). These findings likely explain the lack of
affinity of chromenopyrazole 7b for the CB receptor. The methoxy derivative 18 displays
2
similar occupation of the binding site, however, 18 adopts a different orientation than the free
phenolic hydroxyl ligands 5, 42, and 7b. This orientation enables S6.58(268) to form a hydrogen
bond with the pyran oxygen [hydrogen bond (OꢀO) distance = 3.21 Å and (OꢀHꢀꢀO) angle =
110°].
The results obtained with our docking studies in the CB R* model reveal the importance of
2
two residues, S6.58(268) and K3.28(109). The serine S6.58 had been previously mentioned in
48
45
the interactions of the classical cannabinoid AM841 with CB R*. However, the phenolic
2
hydroxyl of AM841 participated in this interaction. In what concerns the lysine K3.28, its
4
9
importance to classical cannabinoid binding to CB R* is well documented, whereas K3.28 was
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 70
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
45
considered not essential to bind to CB R*. As shown in our docking studies, K3.28 establishes
2
a hydrogen bond with the isoxazole nitrogen in the 42/CB R* complex.
2
Furthermore, the docking studies suggested that substitution of the hydroxyl phenol of the potent
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
CB /CB compound 42 would lead to a selective CB ligand, as was confirmed by the synthesis
1
2
2
of 43 which turned out to be the most interesting of the series.
Figure 7. Binding site of 18 (panel B, in orange) in the CB R* model. Side view of
2
ligand/receptor complex as if looking through TMH1 (not displayed). Hydrophilic interacting
residues are represented with yellow carbons. Hydrogen bonds are shown with yellow dash lines.
A selected hydrophobic interacting residue, V3.32, is displayed with green carbons. Structure of
7b (panel A, in pink) was superimposed on the 42ꢀCB R* complex. The magenta circle indicates
2
van der Waals steric overlap with K3.28, D(275) and F7.35. The residues K3.28 and D(275)
form a salt bridge within the CB R* binding site.
2
ACS Paragon Plus Environment

Page 25 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DrugꢀLike Properties in Silico
Multivariate statistical analysis of a relative large set of 34 physicochemical descriptors
calculated on the global minimum energy conformer of the chromenopyrazoles and
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
9
chromenoisoxazoles 5−44, cannabinol (CBN), CP55,940, and ∆ ꢀTHC has been realized which
data were compared to a range of 95% of drugs. The predicted data (Supporting Information) for
the chromenoheterocycle derivatives indicated that Lipinski’s and Jorgensen’s pharmacokinetics
rules are followed. One of the physicochemical properties that need to be optimized for
cannabinoids is the lipophilicity, as it is an important factor affecting its bioavailability.
Therefore, it is predicted that the chromenoisoxazoles 42−44 have a better bioavailability profile
9
compared to classical cannabinoids such as ∆ꢀTHC, or CBN calculated in the same predictive
model. In what concerns the bloodꢀbrain barrier, predictive data of the compounds suggest that
they can cross this barrier.
Assessment of Binding of Compounds to Plasma Proteins.
Surface plasmon resonance (SPR) experiments were performed with a Biacore X–100
apparatus to study the different binding levels of selected compounds (14, 16−18, 28, 31, 35−37,
3
9, and 41−43) to two plasma proteins, human serum albumin (HSA) and α ꢀacid glycoprotein
1
(AGP). As this experiment had not been previously reported in the literature for cannabinoids,
5
0
51
different reference cannabinoid ligands (rimonabant, SR144528, WIN55,212ꢀ2, HU308,
CBD, 2ꢀAG, AEA, and CP55,940) were also assessed for comparison. The results (Supporting
Information) indicated that 14−16, 39, 40 and 43 could exhibit medium HSA and AGP binding
levels suggesting appropriate free drug concentrations in plasma, whereas 18, 28, 36, 37, and 42
showed very high plasma protein binding that, if any, could be of interest for in vivo retard
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 70
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
effects. The tested reference cannabinoid ligands CBD, 2AG, and AEA showed very high levels
of binding to AGP, whereas CBD, 2ꢀAG, and CP55,940 showed high levels of binding to HSA.
It has to be kept in mind that cannabinoids are, in general, very lipophilic compounds.
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 vivo Efficacy of 43 in the Acute Inflammatory Phase of a Multiple Sclerosis Animal Model.
One of the most promising therapeutic applications of cannabinoids selectively activating the
5
2
CB receptor is the reduction of neuroinflammatory events. Thus, we wanted to investigate the
2
potential in vivo of 43 in experimental models reproducing neuroinflammatory conditions.
Multiple sclerosis is a complex inflammatory disease that affects the CNS white matter. The
Theiler’s murine encephalomyelitis virus
(TMEV) model is one of the best viralꢀbased models of multiple sclerosis. Intracerebral
infection of susceptible inbred mouse strains with TMEV leads to the induction of a lateꢀonset
demyelinating disease, termed TMEVꢀinduced demyelinating disease (TMEVꢀIDD) which is
pathologically similar to human multiple sclerosis. The acute inflammatory phase of TMEVꢀIDD
(day 7 post infection) has a strong neuroinflammatory response with the participation of
microglial cells as antigen presenting cells of viral antigens. To assess the efficacy of 43, mice
infected with TMEV for 7 consecutive days were injected intraperitoneally with vehicle (10%
DMSO in phosphateꢀbuffered saline, PBS) or 43 at a dose of 5 mg/kg. Brain sections of each
animal (described in Experimental Section) were analyzed by immunofluorescence staining.
35
Because microglial cell activation plays a pivotal role in TMEVꢀIDD, we analyzed the effect of
43 on the expression of Ibaꢀ1, a marker of microglia, in the brain of TMEVꢀinfected mice. As
illustrated in Figure 8, fluorescent staining revealed that TMEVꢀinfection increased the intensity
of fluorescence of Ibaꢀ1+ cells in the brain. Microglia activation was greatly prevented by
ACS Paragon Plus Environment

Page 27 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
administration of 43 leading to a significant reduction of the intensity of microglia activation to
levels close to those quantified in the control group (Sham, Figure 8). Therefore, we can
conclude that administration of 43 significantly reduced microglial activation in TMEVꢀinfected
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
35
mice that based on previous studies should necessarily reduce inflammatory events and
improve the neurological status of treated animals.
Figure 8. Compound 43 significantly reduces microglial activation on TMEVꢀinfected mice.
(
A) Brain sections were stained with an antiꢀIba1 Ab for microglia labeling and analyzed by
confocal microscopy. Representative images are shown. Fluorescence intensity was measured
with ImageJ software and plotted to show quantification of the analyzed images (B), mean ±
SEM. *p<0.05 vs Sham, #p<0.05 vs TMEV+veh, in a nonꢀparametric KruskalꢀWallis test. Scale
bar: 50 ꢂm.
Compound 43 has been also recently evaluated with positive results in inflammatory models of
Huntington’s disease that will be published in due time.
Conclusion
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 70
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
The pharmacological and docking studies carried out with the newly synthesized
chromenopyrazoles and ꢀisoxazoles along with our earlier findings allowed us to determine key
structural features for CB receptor binding. OꢀAlkylated chromenopyrazoles led to compounds
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
displaying high to moderate CB affinity and selectivity vs CB , whereas replacement of the
2
1
pyrazole core by an isoxazole led to very potent cannabinoid ligands. Even though different
degrees of CB selectivity and affinity were reported in the literature for classical cannabinoids,
2
4
5
their binding site interactions have not been fully explored. The results obtained with our
docking studies in the CB R* model reveal the importance of two residues, S6.58(268) and
2
K3.28(109). The serine S6.58 that had been previously described as interacting with the phenolic
4
5
hydroxyl of classical cannabinoid, established a hydrogen bond with the pyran oxygen and our
compounds. The residue K3.28 that was considered not essential for classical cannabinoids to
45
bind to CB R*, formed a hydrogen bond with the isoxazole nitrogen that conferred CB affinity
2
2
to the chromenoisoxazole compared to the corresponding chromenopyrazole. Finally, these
studies led to the synthesis of the chromenoisoxazole 43 that showed to be fully CB selective
2
with high affinity constant.
Multiple sclerosis is the major immuneꢀmediated, demyelinating and neurodegenerative disease
9
of the central nervous system (CNS). A mixture of ∆ ꢀTHC and cannabidiol oromucosal spray
has shown clinical benefit in reducing spasticity symptoms in multiple sclerosis and it is now
licensed for the treatment of multiple sclerosis symptoms. However, there is now abundant
experimental evidence that cannabinoids can act as inmunomodulators and neuroprotective
3
0
agents in both in vitro and in vivo models of neurodegeneration. In particular, the CB receptor
2
has been associated with the antiꢀinflammatory and immunomodulatory actions exerted by
5
3
54,55
cannabinoids and has been suggested to play a role in multiple sclerosis models.
The
ACS Paragon Plus Environment

Page 29 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
modulation of the innate immunity including microglia responses to TMEV infection by
cannabinoids treatment affected the development and progression of disabilities in the TMEVꢀ
3
5,56
IDD model.
In this context, compound 43 has been tested in the acute inflammatory phase of
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
the TMEV model. Administration of 43 significantly reduced microglial activation.
Experimental Section
Chemistry
General Methods and Materials. Reagents and solvents were purchased from SigmaꢀAldrich
Co., Fluorochem, Acros Organics, Manchester Organics or LabꢀScan and were used without
further purification or drying. Silica gel 60 F254 (0.2 mm) thin layer plates were purchased from
Merck GmbH. Microwave assisted organic synthesis was performed using the microwave reactor
Biotage Initiator. Products were purified using flash column chromatography (Merck Silica gel
60, 230ꢀ400 mesh) or medium pressure chromatography using Biotage Isolera One with preꢀ
packed silica gel columns (Biotage SNAP cartridges). The compounds were characterized by a
combination of NMR experiments, HPLCꢀMS, highꢀresolution mass spectrometry (HRMS) and
elemental analysis. HPLCꢀMS analysis was performed on a Waters 2695 HPLC system equipped
with a photodiode array 2996 coupled to Micromass ZQ 2000 mass spectrometer (ESIꢀMS),
using a reverseꢀphase column SunFireTM (Cꢀ18, 4.6 x 50 mm, 3.5 ꢂm) in a 10 min gradient A:
CH CN/0.1% formic acid, B: H O/0.1% formic acid visualizing at λ = 254 nm. Flow rate was 1
3
2
mL/min. Elemental analyses of the compounds were performed using a LECO CHNSꢀ932
1
apparatus. Deviations of the elemental analysis results from the calculated are within ± 0.4%. H,
1
3
C, HSQC and HMBCꢀNMR spectra were recorded on a Bruker 300 (300 and 75 MHz), or a
Varian 500 (500 and 126 MHz) at 25 °C. Samples were prepared as solutions in deuterated
ACS Paragon Plus Environment