Structure-activity relationships of thiazole and benzothiazole derivatives as selective cannabinoid CB2 agonists with in vivo anti-inflammatory properties

Article abstract of DOI:10.1016/j.ejmech.2019.07.002

The strong therapeutic potential of CB2 receptor agonists for use as anti-inflammatory agents that lack psychiatric side effects has attracted substantial interest. We herein describe the rational design and synthesis of novel thiazole and benzothiazole derivatives and the evaluation of their binding affinity and functional activity on CB1 and CB2 receptors. The series with the general formula N-(3-pentylbenzo [d]thiazol-2(3H)-ylidene) carboxamide (compounds 6a-6d) exhibited the highest affinity and selectivity towards CB2 receptors with K_is in the picomolar or low nanomolar range, and selectivity indices (K_i hCB1/K_i hCB2) reaching up to 429 fold. Notably, these compounds also demonstrated an agonistic functional activity in cellular assays with EC₅₀s in the low nanomolar range. More interestingly, compound 6d, the 3-(trifluoromethyl)benzamide derivative, exhibited remarkable protection against DSS-induced acute colitis in mice model.

Full text of DOI:10.1016/j.ejmech.2019.07.002

Accepted Manuscript
Structure-activity relationships of thiazole and benzothiazole derivatives as selective
cannabinoid CB2 agonists with in vivo anti-inflammatory properties
Aya E. Ghonim, Alessia Ligresti, Alessandro Rabbito, Ali Mokhtar Mahmoud,
Vincenzo Di Marzo, Noha A. Osman, Ashraf H. Abadi
PII:
S0223-5234(19)30623-3
DOI:
https://doi.org/10.1016/j.ejmech.2019.07.002
EJMECH 11499
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 19 May 2019
Revised Date: 1 July 2019
Accepted Date: 1 July 2019
Please cite this article as: A.E. Ghonim, A. Ligresti, A. Rabbito, A.M. Mahmoud, V. Di Marzo, N.A.
Osman, A.H. Abadi, Structure-activity relationships of thiazole and benzothiazole derivatives as selective
cannabinoid CB2 agonists with in vivo anti-inflammatory properties, European Journal of Medicinal
Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.07.002.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
TABLE OF CONTENTS GRAPHIC

ACCEPTED MANUSCRIPT
Structure-Activity Relationships of Thiazole and
Benzothiazole Derivatives as Selective
Cannabinoid CB2 Agonists with In Vivo Anti-
Inflammatory Properties
1
2
2
2
Aya E. Ghonim , Alessia Ligresti , Alessandro Rabbito , Ali Mokhtar Mahmoud ,
2
1*
1
Vincenzo Di Marzo , Noha A. Osman , Ashraf H. Abadi .
1
Department of Pharmaceutical Chemistry, Faculty of Pharmacy and Biotechnology, German
University in Cairo, Cairo 11835, Egypt.
2
Endocannabinoid Research Group, Institute of Biomolecular Chemistry, Consiglio
Nazionale delle Ricerche, Pozzuoli, Italy.
*
To whom correspondences should be addressed:
Noha A. Osman; email: noha.osman@guc.edu.eg, nohaabdelazizosman@gmail.com, Phone:
201005444305.
+

ACCEPTED MANUSCRIPT
ABSTRACT
The strong therapeutic potential of CB2 receptor agonists for use as anti-inflammatory agents
that lack psychiatric side effects has attracted substantial interest. We herein describe the
rational design and synthesis of novel thiazole and benzothiazole derivatives and the
evaluation of their binding affinity and functional activity on CB1 and CB2 receptors. The
series with the general formula N-(3-pentylbenzo[d]thiazol-2(3H)-ylidene) carboxamide
(compounds 6a-6d) exhibited the highest affinity and selectivity towards CB2 receptors with
K s in the picomolar or low nanomolar range, and selectivity indices (K hCB1/K hCB2)
i
i
i
reaching up to 429 fold. Notably, these compounds also demonstrated an agonistic functional
activity in cellular assays with EC s in the low nanomolar range. More interestingly,
5
0
compound 6d, the 3-(trifluoromethyl)benzamide derivative, exhibited remarkable protection
against DSS-induced acute colitis in mice model.
KEYWORDS: thiazole; benzothiazole; CB2 ligands; CB2 agonists; structure-activity
relationship; DSS-induced colitis.
ABBREVIATIONS
cAMP, cyclic adenosine monophosphate; CB1, cannabinoid receptor 1; CB2, cannabinoid
receptor 2; CDCl , deuterated chloroform; Cpd, compound; DCM, dichloromethane; DMF,
3
dimethylforamide; DMSO, dimethyl sulfoxide; DSS, dextran sulfate sodium; EtOAc, ethyl
acetate; EtOH, ethanol; h, hour; HEK-293, human embryonic kidney 293;
i.p,
intraperitoneal; IC , 50% inhibitory concentration; IL-1β, interleukin 1beta; K , inhibition
5
0
i
constant; min, minutes; MW, microwave; RLU, relative luminescence units; SAR, structure-

ACCEPTED MANUSCRIPT
activity relationship; SEM, standard error of the mean; TFA, trifluoroacetic acid; TNF-α,
tumor necrosis factor alpha; W, watt.
INTRODUCTION
Interest in the medicinal uses of phytocannabinoids sparked-off ongoing research in
pursuance of a better understanding of the endocannabinoid system, the endogenous
1
responder of such interesting molecules. Characterised by a number of endogenous
2
cannabinoids (endocannabinoids) that bind to two G-protein coupled receptors; CB1 and
3
4
CB2, in addition to biosynthetic and inactivation enzymes and transport proteins, the
endocannabinoid system is capable of maintaining homeostasis and balance in the body
5
6
through controlling key regulatory functions including inflammation, immune function,
7
8
9
cancer, appetite regulation, metabolism, and nociception, in a rather complex neuroactive
signalling fashion. Despite how all the components of this complex system present excellent
therapeutic targets for a range of pathological conditions, targeting cannabinoid receptor CB2
for an anti-inflammatory psychotropic-free effect have recently gained significant
momentum. By virtue of being predominantly expressed by the immune cells in addition to
its inducible nature; being dramatically upregulated during inflammatory conditions in
1
0-12
peripheral and central tissues,
CB2 receptor is thought to re-achieve a homeostatic
balance following tissue injury. The latter has been reported to take place through induction
of apoptosis in immune cells, inhibition of T cell proliferation, inhibition of cytokine and
chemokine production by T cells and impairment of macrophage proliferation, function and
1
3-15
release of chemokines.
This anti-inflammatory role is complemented by several recent
9
experimental findings suggesting CB2 involvement in ∆ -tetrahydrocannabinol (Chart 1) and
1
6-17
cannabidiol anti-inflammatory properties.
This is in addition to the distinguishable
number of synthetic selective CB2 agonists that have also shown promising central and
18
peripheral anti-inflammatory activities in animal models of multiple sclerosis, Parkinson's

ACCEPTED MANUSCRIPT
1
9
20
21
22
23
disease, Alzehimers, uveitis, autoimmune encephalitis and ulcerative colitis. In the
gastrointestinal tract, the components of the endocannabinoid system have been detected
2
4
under normal and inflamed conditions, during which an upregulation of CB2 receptors, an
elevation in endocannabinoids levels and a reduction in the corresponding metabolic enzymes
2
5-26
have been witnessed, perfectly fitting a protective mechanism theory.
These events
facilitate an on-target mission for selective CB2 agonists to depress the inflammation process
in ulcerative colitis, thereupon, their use as a novel therapeutic tool for treating various GI
inflammatory conditions has emerged. In proof of this concept, CB2 agonist JWH133 (Chart
1
) was found to attenuate TNBS-induced colitis in animal models, an effect that was
completely abolished by co-injecting CB2 antagonist AM630 while extensively aggravated
2
5
by single administration of the latter. More recently, CB2 agonist HU308 (Chart 1)
mediated protective effect in colitis animal model was accredited to suppression of the
NLRP3 inflammasome responsible for inflammatory process initiation, activation and
2
7
amplification.
Chart 1: Chemical structure of some cannabinoid receptor agonists.
Despite the favourable preclinical outcomes of a big number of highly affine ligands
specifically developed for targeting CB2 receptors, only few compounds were considered for
clinical trials and none of them has currently been approved for human use, principally due to
2
8-29
poor efficacy.
Provoked by the preclinical/clinical mismatching outcomes that are

ACCEPTED MANUSCRIPT
standing in the way of harnessing the promising therapeutic utilities of CB2 selective
agonists, there is an increasing interest in exploiting novel chemical scaffolds for potent and
selective CB2 agonists that are clinically effective for the treatment of GI inflammatory
disorders. A diverse number of chemical scaffolds have been reported to act as selective CB2
ligands. A trend of mono- or bicyclic core bearing one or more heteroatoms such as oxygen
or nitrogen, a carbonyl containing linker bound to bulky aliphatic or aromatic groups is
3
1
generally seen with reported selective CB2 chemotypes. In this context, an interactive SAR
study, composed of three phases, was carried out, thoroughly describing structural and core
enhancement strategies needed to construct a novel promising series of potent CB2 agonists
(Chart 2).
Chart 2: Phases of structural modifications and core enhancements employed in the interactive SAR study.
Aliphatic, aromatic and
halogenated substituents
Bulky aliphatic/ aromatic/
halogenated substituents
Amide/reversed amide
Simple/ branched aliphatic
and aromatic substituents
Lipophilic extension
Core enlargement
PHASE I
PHASE II
PHASE III
We started off by the synthesis of monocyclic 4-methyl-2-substituted thiazole-5-
32
carboxamides (Phase I, Chart 2) in alignment with previously reported isoxazoles,
33
34
imidazoles, pyrazoles. Next, we explored the effect of core enlargement through
condensing a benzene ring to form benzothiazole-2-carboxamides (Phase II, Chart 2)
bearing different bulky alicyclic, aromatic and halogenated amide substituents. In order to
achieve chemical diversity, the 2-carboxamide linkage was reversed to form the N-

ACCEPTED MANUSCRIPT
(benzothiazol-2-yl) carboxamide counterparts (Phase III, Chart 2). Driven by the favourable
biological data, the latter chemotype was modified by n-pentyl substitution on basis of other
3
5-36
cannabinoid pharmacophores.
The newly synthesised compounds were tested in a
competitive radio-ligand binding assay towards both hCB1 and hCB2 receptors
overexpressed on membranes of HEK-293 cells. Affinity data (K values) were used to
i
calculate the selectivity indices of these compounds. These ligands were also examined in the
3
7
cAMP Hunter™ assay and the GPCR β-arrestin recruitment Tango assay, with the aim of
evaluation of their functional activity. To assess the in vivo efficacy of the newly developed
chemotypes, a representative from the most potent series was selected to be tested for its anti-
inflammatory action in mice model of acute DSS-induced colitis.
RESULTS AND DISCUSSION
Chemistry. A general route to the synthesis of the targeted thiazole-2-carboxamides is shown
in Scheme 1. The synthesis is mainly comprised of 4 steps unless the starting thioamide was
commercially available (2b, 3b). In the first step, 2,2-dimethylpropyl thioamide (1b) was
prepared in good yields from the corresponding carboxamide through thionation of the
carbonyl using Lawesson’s reagent in THF. The prepared (1b) or commercially available
thioamides (2b, 3b) underwent Hantzsch thiazole synthesis by reacting with ethyl 2-
chloroacetoacetate in refluxing ethanol for 24 h to give the corresponding 2-substituted-4-
methyl 5-thiazole ethyl ester (1c, 2c, 3c). The resulting esters (1c, 2c, 3c) were then subjected
to base-catalyzed ester hydrolysis using 10% NaOH solution in refluxing ethanol. Subsequent
direct coupling of the obtained acids (1d, 2d, 3d) to different amines was done using HBTU
in the presence of tributylamine as a base giving rise to the final amide products (1m, 2e-2q,
3
l-3n).

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of 4-methyl-2-substituted thiazole-5-carboxamides; 1m, 2e-2q, 3l-
a
3
n .
NH2
N
NH2
O
OH
N
O
ii
i
iii
R1
S
R
1
S
R1
R1
S
O
O
1a
1b, 2b, 3b
1c, 2c, 3c
1d, 2d, 3d
1
1
2
3
: R = t-butyl
iv
1
: R = phenyl
1
: R = methyl
N
NHR2
2
1
2
2
2
2
2
2
2
2
2
2
2
2
m, 2m, 3m, R = adamantan-1-yl
2
R1
S
e, R = butyl
O
2
f, R = hexyl
2
1m, 2e-2q, 3l-3n
g, R = cyclopropyl
h, R = cyclopentyl
2
2
i, R = cyclohexyl
2
j, R = morpholin-4-yl
2
k, R = cyclohexylmethyl
2
l, 3l, R = tetralin-1-yl
2
n, 3n, R = benzyl
2
o, R = 2-fluorophenylmethyl
2
p, R = 4-fluorophenylmethyl
2
q, R = 2-chlorophenylmethyl
a
Reagents and conditions: (i) 1 equiv of the appropriate amide, 1 equiv 2,4-bis(4-methoxyphenyl)-1,3-dithia-2,4-
diphosphetane-2,4-disulfide (Lawesson's reagent), THF, reflux, 4 h. (ii) 1 equiv of the thioamide, 1 equiv ethyl
2
-chloroacetoacetate, EtOH, room temperature, overnight. (iii) 10 equiv 10% aqueous NaOH, ethanol, reflux, 4
h. (iv) 1 equiv of the appropriate carboxylic acid derivative, anhydrous DMF, 6 equiv tributylamine, 1.2 equiv
HBTU, room temperature 30 min, 1.2 equiv of the appropriate amine, room temperature, 4-5 h.
Synthesis of benzothiazole-2-carboxamides was achieved in three steps (Scheme 2). The first
0
step comprises refluxing 2-aminothiophenol together with diethyl oxalate at 170 C for 2
0
hours then dropping the temperature to 90 C for another 2 hours to obtain benzothiazole-2-
carboxylic acid ethyl ester 4a in average yields. Treating the resulting beznothiazole ester 4a

ACCEPTED MANUSCRIPT
with refluxing 10% NaOH and a mixture of H O, THF and methanol led to the formation of
2
the acid derivative 4b that was then coupled with different amines to provide the expected
final amides of this series (4c-4f). In an attempt to enhance the greenness of the synthetic
scheme, an alternative route was adopted to directly convert the ester 4a to different amides
0
in excellent yields through a direct fusion reaction at 100 C with aliphatic amines in the
presence of catalytic amounts of NH Cl under neat conditions, giving rise to final amide
4
products requiring no chromatographic purification (4g-4i).
a
Scheme 2. Synthesis of Benzothiazole-2-carboxamides; 4c-4i
a
Reagents and conditions: (i) 1 equiv 2-aminothiophenol, 2 equiv diethyl oxalate, reflux, 4 h. (iia) 10 equiv 10%
aqueous NaOH methanol/THF, reflux 4 h. (iib) 1 equiv ethyl benzothiazole-2-carboxylate, 4 equiv appropriate
aliphatic amine, catalytic amount of NH Cl, 120 ºC, 2-3 h. (iii) 1 equiv of the carboxylic acid derivative,
4
anhydrous DMF, 6 equiv tributylamine, 1.2 equiv HBTU, room temperature 30 min, 1.2 equiv of the
appropriate amine, room temperature, 4-5 h.
As reported in Scheme 3, N-(benzothiazol-2-yl) carboxamides were synthesized through the
reaction between the commercially available benzothiazole-2-amine and different acyl
chlorides in the presence of triethylamine under inert conditions. Allowing the mixture to stir
overnight gave the amides (5a-5d) in average yields. These, in turn, were subjected to N-
alkylation with 1-bromopentane in the presence of NaH providing N-(3-pentylbenzothiazol-
2
(3H)-ylidene) carboxamides (6a-6d).

ACCEPTED MANUSCRIPT
Scheme 3. Synthesis of N-(benzothiazol-2-yl) and N-(3-pentylbenzothiazol-2(3H) - ylidene) carboxamides;
a
5
a-5d, 6a-6d
a
Reagents and conditions: (i) 1 equiv 2-aminobenzothiazole in dry DCM, 1.5 equiv triethylamine, 1.5 equiv of
the appropriate acyl chloride added at 0 ºC to room temperature, 18 h. (ii) 1.3 equiv of the appropriate N-
(
benzo[d]thiazol-2-yl) carboxamide in DMF, 1.9 equiv NaH, 1,4 equiv 1-bromopentane, room temperature (1 h)
0
to 80 C overnight.
Structure-Affinity/Activity Relationships
The newly synthesized thiazoles and benzothiazole carboxamides were evaluated in a
3
competitive radio-ligand binding assay for their ability to displace [ H]-CP-55,940 (a high
affinity radio-ligand) from HEK-293 cells over-expressing human recombinant CB2 and CB1
receptors. Compounds showing greater than 50% displacement at 10 µM screening doses
were tested in a dose-response curve to determine their IC values and calculate the latter
5
0
into K values. The obtained data are shown in Table 1. In vitro binding data for the initial
i
thiazole-5-carboxamide chemotype revealed how aliphatic/alicyclic amide substituents
seem to possess a size dependant effect on the binding affinity on CB2 receptors as it was
generally noticed that increasing the size of the amide substituent positively influenced the
binding affinity of the compounds. This is exemplified by the proportional increase in CB2
binding affinity through replacing the cyclohexyl amide substituent in the inactive 2i (Ki

ACCEPTED MANUSCRIPT
>
10,000 nM) with the bulkier tetraline in 2l (K = 2300 nM) or adamantyl in 2m (K = 450
i
i
nM), the latter, representing the most active compound in this series.
Table 1: Affinities (K values) on CB1 and CB2 of synthesized compounds towards hCB2 and hCB1
i
cannabinoid receptors, selectivity index (K hCB1/K hCB2), and functional activities (EC values) on CB1 and
i
i
50
CB2 receptors. Data represent mean values for at least three separate experiments performed in duplicate and
are expressed as K ± SD (nM).
i
Binding affinity K
i
CB2 EC50 (nM)
Tango assay
CB1 EC50 (nM)
Tango assay
Cpd
Structure
(nM)
SI
hCB2
hCB1
Agonist
Antagonist
Agonist
Antagonist
1
m
395 ± 113
>10,000 >25
>10,000
3981.0
>30,000
>30,000
>30,000
>30,000
2
e
>10,000
>10,000
>30,000
>30,000
>30,000
>30,000
>30,000
2
f
>10,000
>30,000
>30,000
2
g
h
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
O
2
S
N
N
H
2
2
i
j
2860 ± 890 >10,000
>3
2
k
>10,000
>10,000
>30,000
>30,000
>30,000
>30,000
2
300 ± 190 >10,000
>4
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
2
l
4
50 ± 120
>10,000 >22
2
m
>
10,000
>10,000
>10,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
2
n
o
2
>10,000

ACCEPTED MANUSCRIPT
>10,000
>10,000
2
p
>30,000
>30,000
>30,000
>30,000
>
10,000
10,000
>10,000
>10,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
2
3
q
n
>
3
l
NT*
NT*
>30,000
>30,000
436.5
>30,000
>30,000
1174
3
m
1530 ± 288 >10,000
1324 ± 193 >10,000
>6
>7
>30,000
>30,000
4
c
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
4
g
>
10,000
>10,000
4
h
e
1084 ± 345 >10,000
>9
3311
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
4
>10,000
>10,000
>10,000
>10,000
>30,000
>30,000
4
f
4
i
288 ± 27
1060
3
>30,000
>30,000
>30,000
>30,000
4
d
a
82 ± 7
>10,000 >121 >30,000
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
5
272 ± 77
>10,000 >36
>30,000
>30,000
5
b
c
1288 ± 58
>10,000
>10,000
>10,000
>7
>30,000
>30,000
>30,000
>30,000
>30,000
>30,000
5
>30,000
>30,000
5
d
>10,000
>10,000
>30,000
>30,000
>30,000

ACCEPTED MANUSCRIPT
0
0
.10 ±
.002
6
6
a
12 ± 3
2 ± 0.5
120
26.30
12.88
>30,000
>30,000
1621.8
125.89
>30,000
>30,000
0
0
.11 ±
.002
b
22.2
6
6
c
25 ± 7
>10,000
393
2511.8
>30,000
>30,000
>30,000
d
1 ± 0.4
460 ± 73 429
117 ± 22 64
128.82
1360
>30,000
41
>30,000
-
>30,000
-
SR144528
2 ± 0.5
*NT = not tested
The introduction of heteroatoms to the cyclohexyl amide 2i (K >10,000 nM) to afford the
i
corresponding morpholine amide 2j (K = 2860 nM) was found to enhance the CB2 binding
i
affinity. The size of the 2-substituent on the thiazole ring also appeared to deeply influence
the CB2 binding affinity. A significant decrease in affinity was observed upon replacing the
phenyl ring in 2m (K = 450 nM) with a small methyl group in 3m (K = 1530 nM). On the
i
i
other hand, an aliphatic branched tert-butyl slightly enhanced the binding affinity 1m (Ki=
94 nM). Compounds of this initial chemotype showed no binding affinity to CB1 receptors
3
granting them superior CB2 selectivity despite of their modest affinity. Thus, a general
preference for bulky lipophilic substituents either at position 2 or at the amide moiety was
noticed for the initial set of the synthesised compounds. Subsequently, the adopted strategy
during the second round of structural modifications was to increase the size of the core
thiazole ring by synthesizing a set of benzothiazole 2-carboxamides. At this point of the
SAR study, the influence of different amide substituents and of reversing the amide
functionality on CB2 binding affinity were envisaged. As successfully anticipated, increasing
nd
the size of the core in the 2 chemotype; benzothiazole 2-carboxamides, was generally
accompanied with a marked enhancement in the CB2 binding affinity while retaining the

ACCEPTED MANUSCRIPT
relatively higher CB2 over CB1 selectivity. This was demonstrated in the benzothiazole
adamantly amide 4d (K = 82 nM) which displayed a 5-fold increase in CB2 affinity
i
compared to its thiazole counterpart 2m (K = 450 nM). In a similar fashion, the benzothiazole
i
tetralin amide derivative 4i (K = 288.5 nM) exhibited an 8-fold rise in affinity versus its
i
thiazole analogue 2l (K = 2300 nM). As seen with the initial chemotype, a progressive
i
enhancement in the CB2 binding affinity was observed upon increasing the bulkiness of the
amide moiety from cyclohexyl methyl to tetralin then to 1-adamantyl, compounds 4c (K =
i
1
324 nM), 4i (K = 288.5 nM) and 4d (K = 82 nM), respectively. It was also clear that bulky
i i
aliphatic amide substituents were superior over aromatic ones where naphthyl amido
derivatives 4f and 4e appeared to be disfavoured in terms of CB2 binding affinity. In
addition, the attachment position of the naphthyl moiety to the carboxamide seemed to have
no influence on the activity as both regioisomers; 1 and 2-naphthyl amides(4e and 4f,
respectively) were found to be inactive (K >10,000 nM). It has been noted that the
i
chlorinated benzyl amide analog, 4h, displayed a slightly higher, yet a relatively modest
binding behaviour as that seen with its cyclohexyl methyl analog 4c (K = 1084 nM, 1324
i
nM, respectively). As previously noticed, optimal CB2 affinity within this series of
compounds was obtained with the aliphatic adamantly amide derivative 4d (K = 82 nM).
i
In attempts to achieve chemical diversity other forms of the amide functionality were also
assessed through reversing the amide linkage to afford the N-(benzothiazol-2-yl)
carboxamides. Pharmacomodulation of the acyl substituent, through usage of different bulky
lipophilic groups, was also carried out. Comparing the adamantly amide derivatives 4d and
5
a, we can find that K values have raised from 82.3 nM to 217.8 nM when the carboxamide
i
functionality in 4d was reversed to give the respective retroamide 5a. However, in a
contradicting manner, applying the same modification to the naphthylamide derivative 4e (Ki
>
10,000 nM) afforded compound 5b (K = 1288 nM) with a greatly enhanced CB2 affinity. To
i

ACCEPTED MANUSCRIPT
complete the SAR, we investigated the impact of an n-alkyl substituent by introducing a
pentyl chain at the N-3 position to yield the corresponding ylidenes; N-(3-
pentylbenzothiazol-2(3H)-ylidene) carboxamides. As obvious from the binding results, this
modification induced a dramatic increase in CB2 binding affinity. This pattern was consistent
with all derivatives synthesized. N-pentyl substitution of the p- and m- trifluoromethyl phenyl
amide derivatives (5c and 5d, respectively) granted these inactive derivatives CB2 binding
affinities in the low nanomolar range (6c K = 25 nM, 6d K = 1 nM, respectively), indicating
i
i
that substitution at this position is a crucial modulator of affinity at the CB2 receptor.
Analogously, K values of the adamantyl and 1-naphthyl amides (5a and 5b) were powerfully
i
lowered from the nanomolar to the picomolar range (6a K = 0.10 nM, 6b K = 0.11 nM,
i
i
respectively). This substantial enhancement in affinity implies a potential hydrophobic
interaction with the binding site of the receptor at this part of the molecule. Through
exploring amide substituents of different natures, elements controlling potency and
CB2/CB1 subtype selectivity were deduced. The highest CB2 binding affinities were
achieved by the aliphatic adamantyl 6a and aromatic naphthyl 6b amide derivatives,
however, with a concomitant enhancement in binding affinity to the CB1 receptor (SI: 6a=
1
20, 6b= 22). In an attempt to resolve the poor selectivity issue, halogenated aromatic amide
substituents were tested. Interestingly, both 3- and 4-trifluoromethylphenyl amide derivatives
6d and 6c, respectively) displayed a favourable selectivity profile and a regio-dependent
(
affinity outcome. The 4-trifluoromethyl regio-isomer 6c, successfully eliminated the activity
on CB1 receptors (K > 10,000 nM) at the expense of a relatively lower CB2 binding affinity
i
(
K = 25 nM). However, more conveniently, shifting the trifluoromethyl group to the meta
i
position in 6d contributed to a 25-fold enhancement in CB2 binding affinity (K = 1 nM)
i
while retaining an appreciable selectivity index (SI, 6d = 429).

ACCEPTED MANUSCRIPT
In vitro functional characterization.
The most potent compounds, 6a, 6b and 6d, were selected for additional in vitro
pharmacological evaluation to assess their ability to functionally activate CB2. Two
orthogonal assays were employed, one evaluated their ability to inhibit cAMP accumulation,
one of the primary second messengers in the G-protein-coupled effector system, and the other
one tested their ability to recruit β-arrestin, an important response regulating signal
transduction beyond G protein activation. To measure intracellular cAMP levels, we
performed the enzyme fragment complementation (EFC) technology-based cAMP Hunter™
assay in CHO cells optimized to overexpress CB2. Compounds were tested in presence of
2
5µM NKH-477 solution (a water-soluble analogue of Forskolin) to stimulate adenylate
cyclase and enhance basal cAMP levels. As shown in Figure 1 (left side), compounds 6a, 6b,
6
d were able to inhibit NKH-477-stimulated cAMP accumulation in cells, with EC s 57nM,
50
6
8nM, 306 nM, respectively, suggesting their action as agonists at CB2. To further confirm
the efficacy of compounds 6a, 6b, 6d as CB2 agonists, the GPCR β-arrestin recruitment
Tango assay was performed Figure 1 (right side). It relies on the finding that β-arrestins are
37
essential to stimulate agonist-induced receptor internalization. Again, they showed agonistic
pattern with EC s of 26.3, 12.88 and 128.82 nM, respectively. More interestingly, in both
50
assays their observed efficacy as agonists demonstrated a similar trend to their binding data,
confirming a strong correlation between both. To characterize the functionality of all other
3
7
synthesised compounds, the GPCR β-arrestin recruitment Tango assay was performed. For
measurement of agonistic activity, the synthesized compounds were subjected to full
concentration-response studies against different concentrations of the non-selective
cannabinoid receptor reference agonist CP-55,940, with a maximum concentration of 30 µM.

ACCEPTED MANUSCRIPT
While for measurement of the antagonistic activity, 0.3 µM (EC80 concentration) of the
reference agonist (CP55,940) was initially added to elicit the activation of the receptors above
the basal level. Subsequently, synthesized compounds were subjected to full concentration-
response studies against a series of dilutions of reference antagonist SR144528; (a selective
CB antagonist) and/or Rimonabant ((SR141716); a selective CB antagonist) with the
2
1
maximum concentration used was 30 µM in order to determine the EC50 values of our
compounds. All functional results were analyzed using Graph Pad Prism v7.0 using its built-
in functions.
125
100
6d
75
50
25
0
NKH-477
JWH-133 (5µM, EC100
)
-9
-8
-7
-6
-5
-4
log [M]
Figure 1: Left side; concentration-response curves of compounds 6a, 6b and 6d (red) in cAMP Hunter™ assay,
Effect of EC100 JWH-133 as reference agonist is also shown (violet). Right side; concentration-response curves

ACCEPTED MANUSCRIPT
of compounds 6a, 6b and 6d (red) in GPCR β-arrestin recruitment functional assay. Effect of CP-55,940 as a
reference agonist is also shown (violet).
In Vivo Efficacy: Mouse Model of DSS-Induced Colitis
Driven by a favourable selectivity and affinity profile (hCB2 K = 1 nM, hCB1 K
i
/hCB2 K
i
SI
i
>
429) and its efficacy, the CB2 agonist 6d has been chosen for assessing the in vivo anti-
inflammatory potential, of this novel class of compounds, in mice model of acute DSS-
induced colitis. Swiss male albino mice (20-35 g) were provided with 3% DSS in distilled
drinking water ad libitum for 7 days during which treated mice (n=7) recieved daily
intraperitoneal injections of 10 mg/kg of compound 6d in 4% DMSO, 2% tween in saline.
Positive controls recieved vehicle and DSS water while negative controls recieved vehicle
only. At day 7, the clinical severity of the colitis was assesed in the form of a disease activity
index (DAI) based on weight loss, stool consistency, and the presence of blood in stool.
Significantly, treatment with 10 mg/kg of compound 6d lowered the DAI (3.286 ± 0.8081, P
<
0.05) compared to the untreated control group (7.143 ± 1.29, P < 0.05), Figure 2A. After
sacrificing the test subjects, colon length -being a parameter for assessing severity of colitis-
was evaluated and a significant shortening of the colon was observed with the untreated
control mice (6.729 ± 0.1911, P= 0.0220) relative to treated ones (7.314 ± 0.1143, P=
0
.0220), Figure 2B. Histological data obtained from the distal part of the colon showed
ulceration and desquamation of the lining of the mucosal epithelium and displayed oedema,
haemorrhage and lymphoid hyperplasia in the sub-musoca of untreated control mice, being
absent in 6d treated mice. However, treated mice showed comparable degree of inflammatory
cells infiltration in the submucosa to untreated controls (Figure 3).

ACCEPTED MANUSCRIPT
A
B
C
D
Figure 2: Effect of daily 10 mg/kg i.p treatment with compound 6d on disease activity index DAI score (A) and
colon length (B) in DSS-induced colitis mice model. Quantitative determination of inflammatory cytokines IL-
1
β and TNFα using ELISA in the colons of 6d-treated mice, positive and negative controls (C, D) (values are
the mean ± SEM of 8 mice per group; *p < 0.05; ****p < 0.0001 vs positive control).
To further confirm an anti-inflammatory role of 6d, protein levels of major inflammatory
cytokines such as TNF-α and IL-1β were quantified in colon tissues using ELISA. Treatment
with 6d brought about a powerful anti-inflammatory action apparent with remarkably lower
levels of both TNF-α (58.71 ± 4.434, P<0.0001) and IL-1β (65.81 ± 3.694, P<0.0001)
compared to untreated controls; TNF-α (102.4 ± 5.043, P<0.0001) and IL-1β (140.9 ± 8.551,
P<0.0001), Figure 2C, 2D. Taken together, these data strongly indicate that compound 6d,
being a representative of our most potent series of CB2 agonists N-(3-pentylbenzothiazol-

ACCEPTED MANUSCRIPT
2
(3H)-ylidene) carboxamides, exhibits a protective effect against DSS-induced acute colitis
in mice model.
A
B
C
D
Figure 3: Representative micrographs of H&E (hematoxylin & eosin)-stained sections of distal colon from
experimental groups of DSS-induced colitis model. In positive controls; massive inflammatory infiltration and
aggregation in lamina propria with mucosal destruction and ulceration, as well as diffuse haemorrhage with
lymphoid follicles hyperplasia was recorded (A), and (B), which is largely ameliorated in compound 6d-treated
mice that showed no histopathological alterations except for inflammatory cells infiltration in lamina propria.
(
C) and (D). (Magnification: A, B, C: x16, D: x40).
Molecular Docking
To explain the impact of different substituents on the CB2 receptors binding, compound 6d
was docked to the CB2 receptor homology model derived from the CB1 co-crystal structure
PDB 5XRA. The binding site is identified from the CB1 template and from the site finder of
MOE software and both identified the same binding pocket. The protein is prepared by

ACCEPTED MANUSCRIPT
adding hydrogens and partial charges. Compound 6d is subjected to conformational search
and docked to the site using default conditions. The best scoring pose and close poses showed
mainly hydrophobic interactions between the benzothiazole and lipophilic pocket lined with
lipophilic residues like Phe106, Ile110, Val113, Phe117, the other aryl part is also embedded
in a lipophilic pocket outlined with Trp194, Trp258, Val261, Met285. Moreover, the pentyl
substituent is involved in hydrophobic interactions with Phe91, Phe281, this may explain the
extra-potency of the disubstituted benzothiazoles (6a-6d) relative to their monosubstituted
congeners (5a-5d). In addition, there is always a hydrogen bonding between the amide
oxygen and Ser285, Figure 4.

ACCEPTED MANUSCRIPT
4
A

ACCEPTED MANUSCRIPT
4
B

ACCEPTED MANUSCRIPT
4
C
Figure 4: 2D (4A) and 3D (4B) predicted interaction of compound 6d with the homology model of human CB2
receptor. The interactions are mainly hydrophobic and with on H-bonding with Ser 285. Figure 4C, shows the
distances between key residues and corresponding groups of 6d; Hydrogen bonds shown as red lines,
hydrophobic interaction and π-π shown as blue lines
EXPERIMENTAL SECTION

ACCEPTED MANUSCRIPT
Chemistry. Solvents and reagents were obtained from commercial suppliers and were used
without further purification. Anhydrous reactions were carried out under argon. Reaction
progress was monitored by TLC using fluorescent pre-coated silica gel plates and detection
of the components was made by short UV light (λ= 254 nm). Column chromatography was
1
13
carried out using silica-gel 40-60 µm mesh. H-NMR, C-NMR spectra were recorded at
1
Varian 400 MHz spectrometer or at Varian 300 MHz spectrometer. H shifts are referenced
1
3
to the residual protonated solvent signal (δ 2.50 for DMSO-d and δ 7.26 for CDCl ) and C
6
3
shifts are referenced to the deuterated solvent signal (δ 39.5 for DMSO-d and δ 77.0 for
6
CDCl ). Chemical shifts are given in parts per million (ppm), and all coupling constants (J)
3
are given in Hz. The purities of the tested compounds were determined by HPLC coupled
with mass spectrometry and were higher than 95% purity for all compounds. Mass
spectrometric analysis (UPLC-ESI-MS) was performed using Waters ACQUITY Xevo TQD
TM
system, which consisted of an ACQUITY UPLC H-Class system and Xevo TQD triple-
quadrupole tandem mass spectrometer with an electrospray ionization (ESI) interface (Waters
Corp., Milford, MA, USA). Acquity BEH C 100 mm × 2.1 mm column (particle size, 1.7
1
8
µm) was used to separate analytes (Waters, Ireland). The solvent system consisted of water
containing 0.1% TFA (A) and 0.1% TFA in acetonitrile (B). HPLC-method: flow rate 200
µL/min. The percentage of B started at an initial of 5% and maintained for 1 minute, then
increased up to 100% during 10 min, kept at 100% for 2 min, and flushed back to 5% in 3
min. The MS scan was carried out at the following conditions: capillary voltage 3.5 kV, cone
voltage 20V, radio frequency (RF) lens voltage 2.5V, source temperature 150 ◦C, and
desolvation gas temperature 500 ◦C. Nitrogen was used as the desolvation and cone gas at a
flow rate of 1000 and 20 L/h, respectively. System operation and data acquisition were
controlled using Mass Lynx 4.1 software (Waters). Melting points were determined on
BuchiB-540 Melting Point apparatus and are uncorrected.

ACCEPTED MANUSCRIPT
General synthetic methods and experimental details for some key compounds.
Procedure for the synthesis of 2,2-dimethyl propanthioamide (1b).
A solution of 2,2-dimethylpropanamide 1a (1 g, 9.8 mmol) and 2,4-bis(4-methoxyphenyl)-
1
,3-dithia-2,4-diphosphetane-2,4-disulfide (Lawesson's reagent) (1.98 g, 4.9 mmol) in THF
(15 mL) was heated to reflux for 4 h. The reaction mixture was then cooled to room
temperature and poured into a saturated aqueous solution of NaHCO3 (50 mL). The mixture
was extracted with ether (35 mL) and the organic layer was dried over anhydrous MgSO4
then evaporated under reduced pressure giving the desired white powder 1b that was used
o
39
without further purification in a yield of 78.9%; mp: 117-119 C.
General procedure for the synthesis of ethyl 2-substituted-4-methyl 5-thiazole
carboxylates (1c, 2c, 3c).
To a suspension of different thioamides 1b, 2b, 3b (7.28 mmol) in ethyl alcohol (15 mL) was
added at room temperature ethyl 2-chloroacetoacetate (1 equiv.). The solution was refluxed
for 24 h, and then the solvent was removed under reduced pressure. The solid material was
washed with cooled hexane (2 x 10 mL) to give the corresponding pure ethyl carboxylate
derivatives 1c, 2c, 3c in different yields.
Ethyl 2-(tert-butyl)-4-methylthiazole-5-carboxylate (1c). Prepared from 2,2-dimethyl
o
40
propanthioamide (1b). Yellow solid 75.3%; mp: 43-47 C.
Ethyl 4-methyl-2-phenylthiazole-5-carboxylate (2c). Prepared from commercially
o
1
available thiobenzamide 2b. White solid 78%; mp: 42-43 C; H-NMR (δ): (300 MHz,
DMSO) δ 8.01 – 7.92 (m, 2H), 7.59 – 7.45 (m, 3H), 4.28 (q, J = 7.1 Hz, 2H), 2.67 (s, 3H),

ACCEPTED MANUSCRIPT
1
3
1
1
.29 (t, J = 7.1 Hz, 3H); C-NMR (δ): (75 MHz, DMSO) δ 169.28, 161.68, 160.57, 132.39,
4
1
31.74, 129.66, 126.84, 121.58, 61.51, 17.54, 14.44.
Ethyl 2,4-dimethylthiazole-5-carboxylate (3c). Prepared from commercially available
o
+ 42
methylthioamide 2c. Red solid 91.3%; mp: 48-50 C; MS (ESI): m/z 186.08 [M + H] .
Procedure for the synthesis of 4-methyl 2-substituted thiazole-5-carboxylic acid (1d, 3d).
The appropriate ethyl ester (1c, 3c) (5.8 mmol) was refluxed for 4 h in a mixture of equal
volumes of aqueous 10% NaOH (20 mL) and Ethanol (20 mL). After completion of the
reaction, ethanol was evaporated under reduced pressure and the remaining solution was then
extracted with ethyl acetate. The organic layer was discarded while the aqueous layer was
acidified to pH 3 with aqueous 10% hydrochloric acid. The mixture was then evaporated
under reduced pressure and the solid obtained was then dissolved in methanol while the
remaining insoluble NaCl was discarded. Evaporating the methanol afforded the desired pure
acids 1d and 3d.
2
-(tert-butyl)-4-methylthiazole-5-carboxylic acid (1d). Prepared from ethyl 2-(tert-butyl)-
o
43
4
-methylthiazole-5-carboxylate 1c. White solid 95%; mp: 154-157 C.
2
,4-Dimethylthiazole-5-carboxylic acid (3d). Prepared from ethyl 2,4-dimethylthiazole-5-
o
44
carboxylate 3c. Yellow solid 90%; mp: 233-235 C.
Procedure for the synthesis of 4-methyl-2-phenylthiazole-5-carboxylic acid (2d).
Ethyl 4-methyl-2-phenylthiazole-5-carboxylate 2c (5.8 mmol) was refluxed for 4 h in a
mixture of equal volumes of aqueous 10% NaOH (20 mL) and ethanol (20 mL). After
cooling in ice, the solution was acidified to pH 3 with 10% aqueous hydrochloric acid. The
resulting precipitate 2d was collected by filtration, washed with H O, dried under vacuum
2
o
1
and used without further purification. White solid 97%; mp: 210-212 C; H-NMR (δ): (300

ACCEPTED MANUSCRIPT
1
3
MHz, DMSO) δ 7.96 (dd, J = 5.7, 1.7 Hz, 2H), 7.51 (d, J = 5.6 Hz, 3H), 2.66 (s, 3H); C-
NMR (δ): (75 MHz, DMSO) δ 168.72, 163.21, 159.89, 132.59, 131.51, 129.60, 126.75,
4
2
1
23.14, 17.40.
Procedure for the synthesis of 4-methyl 2-substituted thiazole-5-carboxamides (1m, 3l-
n):
3
To a magnetically stirred suspension of the carboxylic acid derivative 1d or 3d (1.6 mmol) in
anhydrous DMF were successively added tributylamine (Hunig’s base) (6 equiv.) and
N,N,N′,N′-Tetramethyl-O-(1H-benzotriazol-1-yl) uronium hexafluorophosphate (HBTU) (1.2
equiv.). The mixture was allowed to stir at room temperature for 30 minutes. The appropriate
amine (1.2 equiv.) was then added, and the solution was stirred at room temperature for 4-5 h.
The mixture was diluted with EtOAc, washed with aqueous NaHCO3 solution, 1N HCl and
brine, dried over anhydrous MgSO4 and concentrated to give a crude resin. The crude
product was further purified using column chromatography.
N-(adamantan-1-yl)-2-(tert-butyl)-4-methylthiazole-5-carboxamide (1m). Prepared from
1
1
1
-adamantyl-amine. Column chromatography on silica gel (Hexane:Acetone, 9:1) afforded
1
m as a white solid (58.2%); mp: 138-139 ºC; H-NMR (δ): (400 MHz, CDCl ) δ 5.40 (s,
3
1
3
H), 2.61 (s, 3H), 2.10 (s, 3H), 2.08 (s, 6H), 1.70 (s, 6H), 1.40 (s, 9H); C-NMR (δ): (101
MHz, CDCl ) δ 180.26, 161.45, 154.11, 125.83, 53.01, 41.92, 37.83, 36.45, 30.80, 29.60,
3
+
1
7.35; MS (ESI): m/z 333.24 [M + H] .
N-benzyl-2,4-dimethylthiazole-5-carboxamide (3n). Prepared from benzylamine. Column
chromatography on silica gel (DCM:Methanol, 98.5:1.5) afforded 3n as a yellow solid
1
(
75.6%); mp: 90-91 ºC; H-NMR (δ): (300 MHz, CDCl ) δ 7.40 – 7.26 (m, 5H), 6.03 (s, 1H),
3
1
3
4
1
.57 (d, J = 5.7 Hz, 2H), 2.64 (s, 6H); C-NMR (δ): (75 MHz, CDCl ) δ 166.01, 161.87,
3
55.36, 137.95, 128.96, 127.93, 127.85, 125.32, 44.25, 19.44, 17.34; MS (ESI): m/z 247.073
+
[
M + H] .

ACCEPTED MANUSCRIPT
2
,4-Dimethyl-N-(1,2,3,4-tetrahydronaphthalen-1-yl)thiazole-5-carboxamide
(3l).
Prepared from 1,2,3,4-tetrahydronaphth-1-yl amine. Column chromatography on silica gel
1
(
Hexane:Acetone, 7.5:2.5) afforded 3l as a yellow solid (71.4%); mp: 123-125 ºC; H-NMR
(
δ): (400 MHz, CDCl ) δ 7.35 – 7.29 (m, 1H), 7.23 – 7.15 (m, 2H), 7.15 – 7.10 (m, 1H), 5.92
3
(d, J = 7.8 Hz, 1H), 5.37 – 5.28 (m, 1H), 2.90 – 2.74 (m, 2H), 2.65 (s, 6H), 2.19 – 2.08 (m,
1
3
1
1
1
H), 1.94 – 1.83 (m, 3H); C-NMR (δ): (101 MHz, CDCl ) δ 165.95, 161.28, 155.03,
3
37.83, 136.38, 129.44, 128.72, 127.63, 126.55, 125.73, 48.32, 30.37, 29.35, 20.19, 19.44,
+
7.37; MS (ESI): m/z 287.22 [M + H] .
N-(adamantan-1-yl)-2,4-dimethylthiazole-5-carboxamide (3m). Prepared from 1-
adamantyl-amine. Column chromatography on silica gel (Heptane:Acetone, 7.5:2.5) afforded
1
3
m as a white solid (75.2%); mp: 76-79 ºC; H-NMR (δ): (400 MHz, CDCl ) δ 5.38 (s, 1H),
3
1
3
2
.63 (s, 3H), 2.60 (s, 3H), 2.10 (s, 3H), 2.08 (s, 6H), 1.70 (s, 6H); C-NMR (δ): (101 MHz,
CDCl ) δ 165.38, 161.07, 153.86, 127.21, 53.11, 41.90, 36.43, 29.59, 19.37, 17.20; MS
3
+
(
ESI): m/z 291.06 [M + H] .
Procedure for the synthesis of 4-methyl 2-substituted thiazole-5-carboxamides (2e-2q):
To a magnetically stirred suspension of the carboxylic acid derivative 2d (1.6 mmol) in
anhydrous DCM were successively added tributylamine (Hunig’s base) (6 equiv.) and
N,N,N′,N′-Tetramethyl-O-(1H-benzotriazol-1-yl) uronium hexafluorophosphate (HBTU) (1.2
equiv.) and the mixture was allowed to stir at room temperature for 30 minutes. The
appropriate amine (1.2 equiv.) was then added, and the solution was stirred at room
temperature for 4-5 h. The solvent was then evaporated under reduced pressure and the
resulting crude product was dissolved in EtOAc, washed with aqueous NaHCO solution, 1N
3
HCl and brine, dried over anhydrous MgSO and concentrated to give a crude resin. The
4
crude product was further purified using column chromatography.