Thiazole derivatives as inhibitors of cyclooxygenases in vitro and in vivo

Article abstract of DOI:10.1016/j.ejphar.2015.01.008

Cyclooxygenases (COXs) are important membrane-bound heme containing enzymes important in platelet activation and inflammation. COX-1 is constitutively expressed in most cells whereas COX-2 is an inducible isoform highly expressed in inflammatory condition

Full text of DOI:10.1016/j.ejphar.2015.01.008

European Journal of Pharmacology 750 (2015) 66–73
Contents lists available at ScienceDirect
European Journal of Pharmacology
journal homepage: www.elsevier.com/locate/ejphar
Immunopharmacology and inflammation
Thiazole derivatives as inhibitors of cyclooxygenases in vitro and in vivo
Ghewa A. El-Achkar a,b, Mariam Jouni c, May F. Mrad a, Taghreed Hirz a, Nehme El Hachem d,
Ali Khalaf e, Soukaina Hammoud e, Hussein Fayyad-Kazan f, Assaad A. Eid g,
a,i,
Bassam Badran h, Raghida Abou Merhi c, Ali Hachem e, Eva Hamade c, , Aïda Habib
n
nn
a Department of Biochemistry and Molecular Genetics, Faculty of Medicine, AUB, Beirut, PO Box 11-236, Lebanon
b INSERM U955, Equipe 12, Faculty of Medicine, University Paris-Est, Creteil, France
c Genomic and Health Laboratory ER 031/PRASE-EDST, Faculty of Sciences, Lebanese University, Beirut, Lebanon
d Bioinformatics and Computational Genomics Laboratory, Institut de Recherches Cliniques de Montréal, University of Montreal, Montreal, Quebec, Canada
e Département de Chimie et de Biochimie, Laboratoire de Chimie Médicinale et des Produits Naturels & PRASE, EDST Lebanese University, Hadath, Lebanon
f Laboratory of Experimental Hematology, Institut Jules Bordet, Université Libre de Bruxelles, 121, Boulevard de Waterloo, 1000 Bruxelles, Belgium
g Department of Anatomy, Cell Biology and Physiological Sciences, American University of Beirut, PO Box 11-236, Lebanon
h Laboratory of Immunology/EDST-PRASE, Lebanese University, Faculty of Sciences, Hadath, Beirut, Lebanon
i Centre de recherche sur l’inflammation, INSERM UMR 1149-Université Paris Diderot, Faculté de Médecine Xavier Bichat, 16 rue Henri Huchard,
F-75018 Paris, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Cyclooxygenases (COXs) are important membrane-bound heme containing enzymes important in
platelet activation and inflammation. COX-1 is constitutively expressed in most cells whereas COX-2 is
an inducible isoform highly expressed in inflammatory conditions. Studies have been carried out to
evaluate thiazole derivatives as anti-inflammatory molecules. In this study, we investigated the in vitro
and in vivo effects of two novel thiazole derivatives compound 1 (N-[4-(4-hydroxy-3-methoxyphenyl)-
Received 25 August 2014
Received in revised form
16 December 2014
Accepted 4 January 2015
Available online 21 January 2015
1,3-thiazol-2-yl] acetamide) and compound
2 (4-(2-amino-1,3-thiazol-4-yl)-2-methoxyphenol) on
Keywords:
Prostaglandin
NSAIDs
Cyclooxygenase
Thiazole derivatives
Air pouch model of inflammation
Platelet aggregation
prostaglandin E2 (PGE2) production and COX activity in inflammatory settings. Our results reveal a
potent inhibition of both compound 1 (IC50 9.0170.01 mM) and 2 (IC50 11.6576.20 mM) (Mean7S.E.
M.) on COX-2-dependent PGE2 production. We also determined whether COX-1 activity was inhibited.
Using cells stably over-expressing COX-1 and human blood platelets, we showed that compound 1 is a
specific inhibitor of COX-1 with IC50 (5.56 ꢀ 10ꢁ872.26 ꢀ 10ꢁ8 mM), whereas compound 2 did not affect
COX-1. Both compounds exhibit anti-inflammatory effect in the dorsal air pouch model of inflammation
as shows by inhibition of PGE2 secretion. Modeling analysis of docking in the catalytic site of COX-1 or
COX-2 further confirmed the difference in the effect of these two compounds. In conclusion, this study
contributes to the design of new anti-inflammatory agents and to the understanding of cyclooxygenase
inhibition by thiazole.
& 2015 Elsevier B.V. All rights reserved.
1. Introduction
the onset and progression of arthritic diseases (Crofford et al.,
1994; Martel-Pelletier et al., 1999). Despite the beneficial effect of
Cyclooxygenases (COXs) are enzymes responsible for the synth-
esis of prostanoids that have inflammatory and thrombotic effects
(Julemont et al., 2004; Patrono et al., 1985; Schror, 1997). COX-1 is
the constitutive isoform important for maintaining homeostasis
whereas COX-2 is the inducible isoform, highly upregulated by
pro-inflammatory mediators. Accumulation of COX-2 protein and
increase in prostaglandin (PG) E2 production are associated with
COX-2 inhibition, many undesirable side effects persist including
gastric irritation result from the inhibition of COX-1 isoform
(Abouzid and Bekhit, 2008; Masferrer et al., 1994).
Selective inhibition of COX-2 has potential clinical benefits in
cancer and neurological disorders. Novel drug design is carried out
with the aim to unveil the therapeutic value of thiazole derivatives
which are heterocycles compounds having frequent biological
interest.
Thiazoles, or 1,3-thiazoles are five member ring heterocyclic
organic compounds with three carbon, one sulfur, and one nitro-
gen atoms. They are important non-carcinogenic molecules that
can be easily metabolized by routine biochemical reactions and
exhibit a variety of biological activities manifested by their
n
Corresponding author.
Corresponding author at: Department of Biochemistry and Molecular Genetics,
nn
Faculty of Medicine, AUB, Beirut, PO Box 11-236, Lebanon.
E-mail addresses: eva.hamade@ul.edu.lb (E. Hamade),
aida.habib@inserm.fr (A. Habib).
http://dx.doi.org/10.1016/j.ejphar.2015.01.008
0014-2999/& 2015 Elsevier B.V. All rights reserved.

G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
67
presence in many potent biologically active compounds
(Apostolidis et al., 2013; Basile et al., 2012). Woods et al. (2001)
demonstrated a potent and selective inhibition of human recom-
binant COX-2 by a series of thiazole analogs of indomethacin
prepared by chemical substitution of carboxyl group with various
thiazoles substitutes. A more potent inhibition was observed when
the carboxyl group were replaced by an aromatic substitution on
the 4-position of the thiazole. These observations, in addition to
other reports, strongly suggest that thiazole derivatives when
manipulated produce potent anti-inflammatory drugs (Li et al.,
2005; Woods et al., 2001).
Scheme 1. Preparation of amino-thiazoles from their corresponding ketones.
Recently, we showed that the two synthesized thiazole deriva-
tives (N-[4-(4-hydroxy-3-methoxyphenyl)-1,3-thiazol-2-yl] aceta-
mide) and (4-(2-amino-1,3-thiazol-4-yl)-2-methoxyphenol) sign-
ificantly inhibited prostaglandin production in lipopolysaccharide
(LPS)-treated RAW 264.7 cells (Hamade et al., 2012). In the present
study we modified the (N-[4-(4-hydroxy-3-methoxyphenyl)-1,3-
thiazol-2-yl] acetamide) molecule by substituting with 2 different
side chains R¼C4H9 for the compound 1 and R¼CH2Ph for the
compound 2 and assessed its inhibitory activity on both COX-1 and
COX-2 on cells in culture and in mice model of inflammation.
Modeling analysis was also performed. We found that compound 1
blocked both COX-2 and COX-1 whereas compound 2 was more
selective for COX-2.
2.4. Cyclooxygenase activity
COX activity was tested in RAW cells as described previously
(Hamade et al., 2012). Briefly, cells were treated with LPS 100 ng/
ml for 24 h washed and incubated with compound 1 or 2 at
different concentrations (0.1, 5, 10 and 25
μ
M) or 10 M of NS 398
μ
for 30 min in Hanks buffer. 10 mM of arachidonic acid was then
added for an additional 30 min. Supernatants were collected and
PGE2 was measured by enzymeimmunoassay (EIA) (Pradelles
et al., 1985). Cells were lysed with RIPA buffer (50 mM Tris–HCl
(pH 7.5), 150 mM NaCl, 1% Nonidet P40, 0.5% sodium deoxycholate,
4% protease inhibitors and 1% phosphatase inhibitors) and COX-2
protein levels were detected by western blot as previously desc-
ribed (Habib et al., 1993).
2. Material and methods
2.5. Blood platelet aggregation
2.1. Reagents
Venous blood was obtained from healthy volunteers who have
not ingested any drugs for the last 14 days and after informed
consent in accordance with the Institutional Review Board (IRB) of
the American University of Beirut (Approval # BioCh.AH.03). The
effect of compounds 1 and 2 were tested on the platelet aggrega-
tion of washed platelets as described previously (Hamade et al.,
2012).
Cell viability assay kit (WST-1) was from Roche Diagnostic
(Mannheim, Germany). Ibuprofen was purchased from Calbio-
chem. NS 398, LPS, anti- -actin, zymosan A, and arachidonic acid
β
were purchased from Sigma Aldrich (St. Louis, MO, USA). Heat
inactivated fetal bovine serum (FBS), penicillin-streptomycin,
DMEM, G418 and Dulbecco's phosphate buffered Saline (PBS) were
obtained from Invitrogen. Anti-mouse IgG-HRP antibodies were
obtained from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA).
Bradford protein assay reagent and bovine serum albumin (BSA)
were from Bio-Rad (Hercules, California, USA). Enhanced chemi-
luminescence (Amersham, Buckinghamshire, UK). Monoclonal anti
COX-1 (COX 111) and anti-COX-2 (COX 214) were raised as
previously described (Habib et al., 1993).
2.6. Over-expression of human recombinant COX-1 in HEK-293 cells
and assessment of COX-1 activity
To assess the effect of compounds 1 and 2, HEK-293 cells stably
over expressing human COX-1 were obtained after transfecting
HEK-293 with 10 mg of human COX-1 cDNA (kind gift of Colin
Funk, Queen's University, Ontario, Canada) (Funk et al., 1991). Cells
were maintained for 3 weeks in culture media containing 1 mg/ml
G418 and further selected and characterized for enzymatic activity
and protein expression of COX-1. Cells were pre-treated in the
absence or presence of different concentrations of compounds 1, 2
2.2. Cell culture
HEK-293 and RAW 264.7 cells (American Type Culture Collec-
tion, Rockville, MD, USA) were cultured in DMEM supplemented
with 10% FBS, 1% penicillin and 1% streptomycin in a 5% CO2
humidified incubator at 37 1C. Cells were plated in 6 or 12 well
plates for 24 h and pre-treated with different concentrations of
inhibitors and/or LPS in serum-free condition. DMSO was used as
vehicle and did not exceed 0.1% (v:v).
or ibuprofen 25
BSA for 30 min prior to the addition of 25
After 30 min incubation, PGE2 was measured in the supernatants.
COX-1 protein was determined by western blot in protein extracts
(Habib et al., 1993).
μ
M in Hanks buffer (pH 7.4), containing 1 mg/ml
μM arachidonic acid.
2.7. Mice sterile-air pouch model of inflammation
Female C57BL/6 mice (8 weeks) were obtained from the animal
care facility at the American University of Beirut. All animal pro-
cedures were performed following the recommendations of the
IACUC (Institutional Animal Care and Use Committee approval #
12-10-238).
The sterile air pouch model was adapted from Paya et al. (1996)
with modifications. The mice were anesthetized with isoflurane
and 5 ml volume of sterile air was injected in the subcutaneous
tissue of the dorsal midline, below the scapula (Fig. 1). The pouch
2.3. Cell viability
Compounds 1 and 2, which are thiazole derivatives, are shown
in Scheme 1. Their preparation is described in Supplementary
materials. Cell viability was assessed on cells according to the
manufacturer's instructions and expressed as percentage of
untreated cells. In brief, 5 ꢀ 104 cells were plated in 96-well plates
and treated with different concentrations of inhibitors for 24 h.

68
G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
analysis was performed using t-test in Sigma Plot (Systat Software
Inc., San Jose, CA), P-valueso0.05 were considered significant.
Autoradiograms obtained from western blot analyses were scanned
using Epson 1680 pro scanner and the band intensities were
quantified by Image J software (NIH, USA).
3. Results
Fig. 1. Outline of the air pouch experiments. On day 1, 5 ml of sterile air was
injected in the subcutaneous tissue of the dorsal midline, below the scapula. The
pouch was re-inflated with an additional 2.5 ml of sterile air on day 4, to maintain
the cavity. On the sixth day after the initial air injection, 0.5 ml of 1% w/v zymosan
was injected into the pouch cavity. 24 h after zymosan administration, the mice
were sacrificed and the exudates from each mouse were collected by washing the
pouch with 1 ml of Hanks buffer containing 0.315% trisodium citrate. The collected
exudates (1 ml) contain the recruited inflammatory cells and resident cells. The
total number of cells was counted. The exudates were centrifuged at 300g for 5 min
at þ4 1C. The supernatants were collected for PGE2 measurement. The cell pellets
were lysed for protein extraction.
3.1. Compounds 1 and 2 inhibit PGE2 release in RAW 264.7 cells
Treatment of the Raw 264.7 cells with compound 1 at different
concentrations (0.1, 5, 10, 25 mM) inhibited significantly LPS-
dependent release of PGE2 compared to LPS-treated cells. There
was a concentration dependent effect starting at 5 mM with a max-
imal inhibitory effect at 25 mM (Fig. 2A). Similarly, treatment of cells
with compound 2 showed an inhibitory effect starting at 10 mM with
was re-inflated with an additional 2.5 ml of sterile air on day 4, to
maintain the cavity. On day 6 after the initial air injection, the mice
were injected with 0.5 ml of 1% zymosan A directly into the pouch
in order to induce local inflammation. 0.5 ml of compound 1 or 2
(0.7 mg/kg, eq. to 100 M in an air pouch injected with 0.5 ml total
μ
volume) were injected twice in the pouch: two hours prior to
zymosan and together with zymosan injection. 24 h after zymosan
administration, the mice were sacrificed and the cellular exudate
from each mouse was collected by washing the air pouch with
1 ml of Hanks buffer containing 0.315% trisodium citrate. Cells
were counted in the collected exudates (1 ml) and further cen-
trifuged at 300g for 5 min at 4 1C. PGE2 was measured in the sup-
ernatant as described above and cell pellet were lysed for protein
COX analysis.
2.8. Modeling analysis
The 3D structure of COX-2 complexed with celecoxib (PDBI-
D:1EQG) and the ovine COX-1 complexed with ibuprofen (PDBI-
D:1EQG) were selected for docking simulations (Hamade et al.,
2012). Water and other heteratoms were removed from the str-
ucture. Chain A was retained including ibuprofen and heme group.
Hydrogen atoms were added, atom typing and partial charges
were assigned using AMBER forcefield (Feng et al., 2009). The
coordinates of the binding site were extracted using the co-crys-
tallized ligand, ibuprofen. Docking and scoring: low energy con-
formations of the chemical compounds were generated using
catalyst (Accelrys, Inc.)
The (R)-groups of compounds 1 and 2 were selected; docking
simulations were carried out using Autodock 4.2. Each docking
simulation was achieved with 10 docking runs with 150 indivi-
duals using the Lamarckian genetic algorithm implemented in
Autodock and 250,000 energy evaluations. The binding energies
were estimated from a new free-energy scoring function based
on the AMBER force field, an updated charge-based desolvation
term and improved models of the unbound state. The best poses
were analyzed and visualized with Discovery Studio visualizer
(Accelrys, Inc.).
2.9. Data analysis
Fig. 2. Dose dependant inhibition of prostaglandin E2 production by compounds 1
and 2. (A) and (B) RAW 264.7 cells stimulated with 100 ng/ml of LPS for 24 h were
incubated with Hanks buffer and treated for 30 min with compounds 1 and 2 at
increasing concentration (0.1, 5, 10 and 25 mM) or with 10 μM of NS 398 and 20 mM
of arachidonic acid for additional 30 min. Results are expressed as percentage of
secreted PGE2 in ng/ml and correspond to mean7S.E.M. of 3 different experiments.
(C) Curve fitting for the PGE2 values obtained from (A) was done using Grafit
software. (n¼3; *Po0.05, ***Po0.001 vs LPS-treated cells).
Aggregation data was expressed after defining the slope for each
aggregation curve, which reflects the rate of platelet reaction, using
the Born's method (Li et al., 2005). Curve fitting and calculation of
the IC50 values were done using Grafit 7 software (Erithacus
software, Staines, UK). PGE2 concentration in ng/ml was expressed
as mean7S.E.M. for at least 3 different experiments and statistical

G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
69
a maximal inhibitory effect obtained at 25 mM (Fig. 2B). NS 398 a
selective COX-2 inhibitor, showed a strong inhibition of PGE2 release.
Both compounds 1 and 2 showed a similar IC50 (9.0170.01 mM and
11.65 76.20 mM for compounds 1 and 2, respectively) (Fig. 2C).
3.2. Compounds 1 and 2 do not modulate LPS-induced COX-2
expression in RAW 264.7 macrophages
In parallel, we checked whether compound 1 or 2 modified
COX-2 expression. Different concentrations of compounds 1 and 2
(5, 10 and 25 mM) were tested on LPS-dependent induction of COX-
2. Western blot results revealed no modulation of COX-2 expres-
sion by either compound 1 or 2 in RAW 264.7 cells (Fig. 3). Using
WST-1 assay, we further show that compound 1 and compound 2
did not affect the viability of the cells (Fig. 4).
Fig. 4. Cellular viability of RAW 264.7 cells following treatment with compounds 1
and 2. RAW 264.7 cells were incubated with 25 mM of compound 1 or 2 for 24 h.
Cell viability was assessed by WST-1 in comparison to control untreated cells.
Results are representative of 3 different experiments performed in triplicates.
3.3. Compound 1 but not compound 2 inhibits blood platelet
aggregation
We next evaluated the capacity of these compounds to block
platelet aggregation. Platelets were pre-treated for 1 min at 37 1C
prior to the addition of 25
was followed for 3 min. We first tested high concentrations (6.25,
12.5, and 25 M). The compound 2 did not show any inhibition of
μM of arachidonic acid and aggregation
μ
arachidonic acid-dependent platelet aggregation. Compound 1
(3, 10, and 100 nM) with pentyl side chain showed a potent inh-
ibition. Ibuprofen, a well described COX-1/COX-2 non-selective
Fig. 5. Aggregation curves of compounds
1 and 2. Platelet aggregation was
performed on human blood washed platelets. Platelets (0.4 ꢀ 109plt/ml) were
treated with increasing concentrations of compound 1 or 2 for 1 min at 37 1C prior
to the addition of 25 μM of arachidonic acid for 3 min under stirring conditions.
Results represent aggregation curves of compound 1 and compound 2 and are
representative of 4 different experiments.
inhibitor, which strongly block platelet aggregation was used as a
positive control (Fig. 5) (Hirz et al., 2012).
3.4. Compound 1 but not compound 2 inhibits PGE2 levels in HEK-
293 cells stably-overexpressing human COX-1
To further confirm that compound 1 was targeting COX-1 and
not other intermediate in targets, we tested the compounds on
recombinant COX-1. HEK-293 cells over-expressing human COX-1
were obtained. In these cells, arachidonic acid is metabolized into
PGH2, which is sequentially hydrolyzed into PGE2. Measurement of
PGE2 under these conditions reflects COX-1 activity. Results sho-
wed that compound 1 exerted a strong and significant inhibition
of PGE2 synthesis in HEK-293 over-expressing human COX-1 (IC50
was 5.56 ꢀ 10ꢁ872.26 ꢀ 10ꢁ8 mM, n¼3) (Fig. 6A and B), while
compound 2 did not show a significant inhibition (data not
shown). Ibuprofen strongly blocked COX-1 activity (Fig. 6A). By
assessing both platelet aggregation and enzyme activity of
Fig. 3. Compounds 1 and 2 do not modulate COX-2 protein level induced by LPS.
RAW 264.7 cells were stimulated with 100 ng/ml LPS and compounds 1(A) or 2(B)
at 5, 10 and 25 μM, respectively for 24 h. Cells were lysed and analyzed by western
blotting using anti-COX-2 and β-actin antibody. Results are representative of
3
different experiments. (C) Protein bands were scanned and its intensity
quantified using ImageJ software.

70
G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
Fig. 6. Inhibition of COX-1 activity. (A) HEK-293 cells stably overexpressing COX-1 were treated for 30 min in the absence or presence of different concentrations of
compound 1 and 2 or ibuprofen 25 μM prior to the addition of 25 μM arachidonic acid as described in Section 2. Results are expressed as percentage of the secreted PGE2 and
correspond to mean7S.E.M. of 3 different experiments (***Po0.001 vs LPS-treated cells). (B) HEK-293 cells stably overexpressing COX-1 were treated with different
concentrations of compound 1 (0.1, 1, 3, and 10 μM) for 24 h. Cell viability was assessed in comparison to untreated cells as described in the legend for Fig. 4. Results are
representative of 3 different experiments. (C) Western blot analysis of COX-1 using selective monoclonal COX-1 in untransfected and pcDNA3-COX-1transfected cells. Results
are representative of 3 different experiments. (D) Curve fitting for the PGE2 values obtained from (A) was done using Grafit software.
recombinant COX-1, we further confirm that compound 1 and not
compound 2 blocks COX-1.
acids of the active site of the enzymes involved in the interaction
with these compounds. Ibuprofen docked into COX-1 and as shown
previously (Hirz et al., 2012), the carboxyl group of ibuprofen
showed three hydrogen bonds, two with Arginine 120 (guanidine
–NH2 group) and one with Tyrosine 355 (p-OH group).
Compounds 1 and 2 were docked nearby Tyrosine 355 and
Arginine 120 of the COX-1 catalytic site where compound 1 was
found to exhibit three hydrogen bonds one between the methoxy
group and OH group of the side chain of Tyrosine 355 (active site
of COX-1), two between NH2 group and Serine 530, and Tyrosine
385 amino acids (Fig. 8A). On the other hand, compound 2 was
found to exhibit two hydrogen bonds, one between the methoxy
group and Arginine 120 (side chain) of active site of COX-1 and the
other between amine group and Leucine 352 of COX-1 (amide
bond) (Fig. 8B).
The binding scores of compounds 1 and 2 were ꢁ6.2 kcal/mol
and ꢁ6.9 kcal/mol respectively. These scores were comparable to
ibuprofen (ꢁ8 kcal/mol), which served as a positive control for
COX inhibition. Despite the comparable scores of compounds 1
and 2 to ibuprofen, compound 2 did not affect platelet aggregation
due to the absence of hydrogen bonding with Tyrosine 355 (Hirz
et al., 2012) that indispensible for platelet aggregation/COX-1
activity. A possibility for the absence of this hydrogen bonding
can be due to the shielding of the benzyl group.
Compounds 1 and 2 were docked nearby the catalytic site of COX-2
where compound 1 was found to exhibit one hydrogen bond between
the amine group and oxygen of peptide bond of Leucine 338 (active
site of COX-2) (Fig. 9A), while compound 2 was found to exhibit two
hydrogen bonds one between the amine group and oxygen of peptide
bond of Methionine 508 (active site of COX-2), and the other between
the hydroxyl group and Serine 339 of COX2 (oxygen of amide bond)
Viability of HEK-293 overexpressing cells was not affected dur-
ing the treatment as shown by WST-1 assay even when compound
1 was incubated for 24 h at high concentrations (Fig. 6C). More-
over, COX-1 protein expression in HEK-293 cells was not modified
after treatment with compounds 1 and 2 compared to untreated
cells (Fig. 6D).
3.5. In vivo effect of compounds 1 and 2 on PGE2 formation and
COX-2 expression
We next investigated the anti-inflammatory effects of com-
pounds 1 and 2 in vivo in a mouse model of inflammation. When
saline was injected in the air pouch, there was a low accumulation
of cells in the air pouch exudates. Injection of zymosan increased
the total cell number (Fig. 7A). Compounds 1 and 2 did not change
cell infiltration in the pouch exudates (Fig. 7A). The injection of
zymosan for 24 h increased PGE2 levels in exudates from zym-
osan-injected mice compared to saline alone. The intrapouch inje-
ction of 100 mM of compound 1 or 2 decreased significantly PGE2
levels by 51% and 45% (Pr0.01, unpaired t-test) respectively
(Fig. 7B). No effect on COX-2 protein expression was observed
(Fig. 7C). Ibuprofen, the COX-1/COX-2 non-selective inhibitor, also
decreased PGE2 formation and was used as a positive control.
3.6. Modeling analysis
We finally carried out modeling analysis of compounds 1 and 2
with ovine COX-1 and COX-2 to examine how these compounds
dock to the active site of the enzymes, and to determine the amino

G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
71
Fig. 7. Effect of compounds 1 and 2 on PGE2 levels and COX-2 expression in zymosan-injected air pouch model. Drugs were administered two hours before and with
zymosan. 24 h after zymosan the total number of cells was counted (A) and PGE2 levels were measured in exudates using enzymeimmuno assay (B). Cell pellet was lysed
using Laemmli buffer solution. Western blotting was performed using COX-2 and β-actin antibodies. Protein band signals were developed and scanned using Epson 1680 pro
scanner (C). Data represent means7S.E.M. (n¼4–6 animals); *Pr0.05 **Pr0.01 ***Pr0.001. Zym¼zymosan/Ibu¼ibuprofen.
(Fig. 9B). The docking of celecoxib, known to be a positive inhibitor of
COX-2, showed many hydrogen bondings among which are those
between the amine group, Leucine 338 (amide bond), Serine 339 side
chain, and Glutamine 178 (side chain), in addition to a hydrogen
bonding between oxygen of the celecoxib SO2 and the Arginine 499
(side chain) of the COX-2 active site (Fig. 9C).
Based on those results, we can assume that both compounds
exhibit an inhibitory COX-2 activity due to the similarity in the docking
features of these compounds compared to celecoxib. The poses for
compounds 1 and 2 showed a binding affinity of ꢁ7.7 and ꢁ8.1 kcal/
mol which are relatively similar to celecoxib binding profile as a
reference with a binding energy of ꢁ12.5 kcal/mol for the best pose.
4. Discussion
COX-2 inhibition is crucial for the anti-inflammatory properties
of NSAIDs. Since selective inhibitors of COX-2 spare COX-1 enzymatic
activity, COX-2-selective drugs can produce analgesic effects without
causing gastrointestinal ulcers despite the increased risk of asso-
ciated cardiovascular side effects reported (Warner et al., 1999).
We performed this study to analyze the effect of the two novel
thiazole derivatives on COX-1 and COX-2 enzymatic activity. We
demonstrated that compound 1 was a non-selective COX-1/COX-2
inhibitor, whereas compound 2 was a selective COX-2 inhibitor.
The zymosan subcutaneous air pouch model was performed to

72
G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
Fig. 9. Docking compounds and celecoxib into ovine COX-2. The amino acids of
active site of COX-2 were rendered in violet and green. H-bonds are indicated as
blue dashed lines for compounds 1(A) and 2(B) while blue and green for celecoxib
(C). (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
Fig. 8. Docking compounds into ovine COX-1. (A) The amino acids of active site of
COX-1 were rendered in violet and green. H-bonds are indicated as blue dashed
lines. Results for compound 1 (A), compound 2 (B) are illustrated. (For interpreta-
tion of the references to color in this figure legend, the reader is referred to the web
version of this article.)
When the two compounds were docked in COX-2, we observed
that they both interacted with amino acids previously described
for selective inhibitors of COX-2 although the 2 compounds inter-
acted with different groups within the active site. Compound 1
exhibit one hydrogen bond between the NH2 group and oxygen of
peptide bond of Leucine 338 (active site of COX-2) while com-
pound 2 was found to exhibit two hydrogen bonds one between
the NH2 group and oxygen of peptide bond of Methionine 508
(active site of COX-2), and the other between the hydroxyl group
and Serine 339 of COX-2 (oxygen of amide bond). Both compounds
were able to block COX-2 activity in both in vitro assay on Raw
264.7 cells and in air pouch model.
The docking of celecoxib, known to be a positive inhibitor of
COX-2, showed many hydrogen bindings among which are those
between the amine group, Leucine 338 (amide bond), Serine 339
side chain found for each compound (1 or 2), and Glutamine 178
(side chain), in addition to a hydrogen bonding between oxygen of
the celecoxib SO2 and the Arginine 499 (side chain) of the COX-2
active site.
investigate the anti-inflammatory effect of the two compounds
in vivo. The zymosan-dependent migrating cells express COX-2
which is along with mPGES-1, the main enzymes responsible for
PGE2 production in this model (Paya et al., 1997). The groups of
mice treated with compound 1 or compound 2 showed a sig-
nificant decrease in PGE2 levels without any effect on COX-2
protein expression supporting our in vitro findings with a direct
effect on PGE2 synthesis.
Modeling analysis results showed that both compounds are
docked nearby Tyrosine 355 and Arginine 120 of the COX-1 cat-
alytic site. Analysis revealed that compound 1 exhibit three
hydrogen bonds between methoxy and Tyrosine 355, between
NH2 and Serine 530, and between NH2 and Tyrosine 385 while
compound 2 exhibited only two hydrogen bonds between phenol
and Arginine 120 and between NH2 and the amide bond of
Leucine 352. These observations suggest that the absence of the
hydrogen bond between the methoxy group in the compound 2
and Tyr 355 of the active site of COX-1 is due to the steric effect of
the CH2Ph group, as we described previously for derivatives of 13-
hydroxyocatdecadienoic acid and 12-hydroxyeicosatetraenoic acid
(Hirz et al., 2012). Consistent with these results, compound 2 did
not block the enzymatic activity of COX-1 on platelet aggregation.
Many thiazole-containing molecules have been described tar-
geting multi-signaling pathways, which may increase the potential
side-effect. We have described these molecules as specific anti-
cyclooxygenase activities with no cytotoxic effects on cultured
cells. Additional investigation is required to characterize their side
effects, if any, in vivo (Leoni et al., 2014).

G.A. El-Achkar et al. / European Journal of Pharmacology 750 (2015) 66–73
73
In conclusion, our results further confirm a selective role of
thiazoles as COX inhibitors. Our study enabled a more advanced
characterization of structure-function relationship. Further inves-
tigations will be performed in the near future to assess the
absence of adverse effect of this thiazole compound 2 on gastro
intestinal tract lesions.
Effects of interleukin-1 beta, phorbol ester, and corticosteroids. J. Clin. Investig.
93, 1095–1101.
Feng, Z., Chu, F., Guo, Z., Sun, P., 2009. Synthesis and anti-inflammatory activity of
the major metabolites of imrecoxib. Bioorg. Med. Chem. Lett. 19, 2270–2272.
Funk, C.D., Funk, L.B., Kennedy, M.E., Pong, A.S., Fitzgerald, G.A., 1991. Human
platelet/erythroleukemia cell prostaglandin G/H synthase: cDNA cloning,
expression, and gene chromosomal assignment. FASEB J. 5, 2304–2312.
Habib, A., Creminon, C., Frobert, Y., Grassi, J., Pradelles, P., Maclouf, J., 1993.
Demonstration of an inducible cyclooxygenase in human endothelial cells
using antibodies raised against the carboxyl-terminal region of the
cyclooxygenase-2. J. Biol. Chem. 268, 23448–23454.
Author participation
Hamade, E., Habib, A., Hachem, A., Hussein, A.H., Abbas, M., Hirz, T., Al Masri, M.,
Faour, W.H., 2012. Biological and anti-inflammatory evaluation of two thiazole
compounds in RAW cell line: potential cyclooxygenase-2 specific inhibitors.
Med. Chem. 8, 401–408.
Hirz, T., Khalaf, A., El-Hachem, N., Mrad, M.F., Abdallah, H., Creminon, C., Gree, R.,
Merhi, R.A., Habib, A., Hachem, A., Hamade, E., 2012. New analogues of 13-
hydroxyocatdecadienoic acid and 12-hydroxyeicosatetraenoic acid block
human blood platelet aggregation and cyclooxygenase-1 activity. Chem. Cent.
J. 6, 152.
Julemont, F., de Leval, X., Michaux, C., Renard, J.F., Winum, J.Y., Montero, J.L., Damas,
J., Dogne, J.M., Pirotte, B., 2004. Design, synthesis, and pharmacological
evaluation of pyridinic analogues of nimesulide as cyclooxygenase-2 selective
inhibitors. J. Med. Chem. 47, 6749–6759.
AK, SH, and A. Hachem designed the molecules and carried out
the synthesis. GAE, A. Habib and EH designed the biological studies.
GAE, MJ, MFM, TH carried out the biological tests. NH performed
docking analysis. GAE, A. Hachem, A. Habib, and EH conceived the
study. A. Hachem, A. Habib, HFK, EH, AAE, BB, and RAM participated
in problem solving. GAE, AK, A. Hachem, A. Habib and EH analyzed
the data and wrote the paper. All authors read and approved the
final version of the manuscript.
Leoni, A., Locatelli, A., Morigi, R., Rambaldi, M., 2014. Novel thiazole derivatives: a
patent review (2008–2012. Part 2). Expert Opin. Ther. Pat. 24, 759–777.
Li, M.H., Yin, L.L., Cai, M.J., Zhang, W.Y., Huang, Y., Wang, X., Zhu, X.Z., Shen, J.K.,
2005. Design, synthesis, and anti-inflammatory evaluation of a series of novel
amino acid-binding 1,5-diarylpyrazole derivatives. Acta Pharmacol. Sin. 26,
865–872.
Martel-Pelletier, J., Mineau, F., Jovanovic, D., Di Battista, J.A., Pelletier, J.P., 1999.
Mitogen-activated protein kinase and nuclear factor kappaB together regulate
interleukin-17-induced nitric oxide production in human osteoarthritic chon-
drocytes: possible role of transactivating factor mitogen-activated protein
kinase-activated proten kinase (MAPKAPK). Arthritis Rheumatol. 42,
2399–2409.
Acknowledgment
This work was supported by grants from the American
University of Beirut, Lebanon (MPP and URB) and the Lebanese
National Council for Scientific Research (LNCSR grant number 02-
09-12). The funders had no role in study design, data collection
and analysis, decision to publish, or preparation of the manuscript.
We are grateful to the American University of Beirut Medical
Center for the aggregometer. The authors thank Oula Dagher (AUB)
for her help in performing IC50 determination and Dr. Mirvat El
Masri for her help in the synthesis of the molecules.
Masferrer, J.L., Zweifel, B.S., Manning, P.T., Hauser, S.D., Leahy, K.M., Smith, W.G.,
Isakson, P.C., Seibert, K., 1994. Selective inhibition of inducible cyclooxygenase
2 in vivo is antiinflammatory and nonulcerogenic. Proc. Natl. Acad. Sci. USA 91,
3228–3232.
Patrono, C., Ciabattoni, G., Patrignani, P., Pugliese, F., Filabozzi, P., Catella, F., Davi, G.,
Forni, L., 1985. Clinical pharmacology of platelet cyclooxygenase inhibition.
Circulation 72, 1177–1184.
Appendix A. Supplementary information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.ejphar.2015.01.008.
Paya, M., Garcia Pastor, P., Coloma, J., Alcaraz, M.J., 1997. Nitric oxide synthase and
cyclo-oxygenase pathways in the inflammatory response induced by zymosan
in the rat air pouch. Br. J. Pharmacol. 120, 1445–1452.
Paya, M., Terencio, M.C., Ferrandiz, M.L., Alcaraz, M.J., 1996. Involvement of
secretory phospholipase A2 activity in the zymosan rat air pouch model of
inflammation. Br. J. Pharmacol. 117, 1773–1779.
References
Pradelles, P., Grassi, J., Maclouf, J., 1985. Enzyme immunoassays of eicosanoids using
acetylcholine esterase as label: an alternative to radioimmunoassay. Anal.
Chem. 57, 1170–1173.
Schror, K., 1997. Aspirin and platelets: the antiplatelet action of aspirin and its role
in thrombosis treatment and prophylaxis. Semin. Thromb. Hemost. 23,
349–356.
Abouzid, K., Bekhit, S.A., 2008. Novel anti-inflammatory agents based on pyridazi-
none scaffold; design, synthesis and in vivo activity. Bioorg. Med. Chem. 16,
5547–5556.
Apostolidis, I., Liaras, K., Geronikaki, A., Hadjipavlou-Litina, D., Gavalas, A., Sokovic,
M., Glamoclija, J., Ciric, A., 2013. Synthesis and biological evaluation of some
5-arylidene-2-(1,3-thiazol-2-ylimino)-1,3-thiazolidin-4-ones as dual anti-
inflammatory/antimicrobial agents. Bioorg. Med. Chem. 21, 532–539.
Basile, L., Alvarez, S., Blanco, A., Santagati, A., Granata, G., Di Pietro, P., Guccione, S.,
Munoz-Fernandez, M.A., 2012. Sulfonilamidothiopyrimidone and thiopyrimi-
done derivatives as selective COX-2 inhibitors: synthesis, biological evaluation,
and docking studies. Eur. J. Med. Chem. 57, 149–161.
Warner, T.D., Giuliano, F., Vojnovic, I., Bukasa, A., Mitchell, J.A., Vane, J.R., 1999.
Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-
oxygenase-2 are associated with human gastrointestinal toxicity:
a full
in vitro analysis. Proc. Natl. Acad. Sci. USA 96, 7563–7568.
Woods, K.W., McCroskey, R.W., Michaelides, M.R., Wada, C.K., Hulkower, K.I., Bell, R.
L., 2001. Thiazole analogues of the NSAID indomethacin as selective COX-2
inhibitors. Bioorg. Med. Chem. Lett. 11, 1325–1328.
Crofford, L.J., Wilder, R.L., Ristimaki, A.P., Sano, H., Remmers, E.F., Epps, H.R., Hla, T.,
1994. Cyclooxygenase-1 and -2 expression in rheumatoid synovial tissues.