Synthesis, biological evaluation and molecular modeling study of pyrazole and pyrazoline derivatives as selective COX-2 inhibitors and anti-inflammatory agents. Part 2

Article abstract of DOI:10.1016/j.bmc.2012.03.044

New pyrazole and pyrazoline derivatives have been synthesized and their ability to inhibit ovine COX-1/COX-2 isozymes was evaluated using in vitro cyclooxygenase (COX) inhibition assay. Among the tested compounds, N-((5-(4-chlorophenyl)-1-phenyl-3-(triflu

Full text of DOI:10.1016/j.bmc.2012.03.044

Bioorganic & Medicinal Chemistry 20 (2012) 3306–3316
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, biological evaluation and molecular modeling study of pyrazole
and pyrazoline derivatives as selective COX-2 inhibitors and anti-inflammatory
agents. Part 2
Magda A.-A. El-Sayed a, Naglaa I. Abdel-Aziz b, , Alaa A.-M. Abdel-Aziz b,c, Adel S. El-Azab c,d
,
⇑
Kamal E.H. ElTahir e
a Department of Pharmaceutical Organic Chemistry, Faculty of Pharmacy, University of Mansoura, Mansoura 35516, Egypt
b Department of Medicinal Chemistry, Faculty of Pharmacy, University of Mansoura, Mansoura 35516, Egypt
c Department of Pharmaceutical Chemistry, College of Pharmacy, King Saud University, Riyadh 11451, Saudi Arabia
d Department of Organic Chemistry, Faculty of Pharmacy, Al-Azhar University, Cairo 11884, Egypt
e Department of Pharmacology, College of Pharmacy, King Saud University, Riyadh 11451, Saudi Arabia
a r t i c l e i n f o
a b s t r a c t
Article history:
New pyrazole and pyrazoline derivatives have been synthesized and their ability to inhibit ovine COX-1/
COX-2 isozymes was evaluated using in vitro cyclooxygenase (COX) inhibition assay. Among the tested
compounds, N-((5-(4-chlorophenyl)-1-phenyl-3-(trifluoromethyl)-1H-pyrazol-4-yl)methylene)-3,5-
bis(trifluoromethyl)aniline 8d exhibit optimal COX-2 inhibitory potency (IC50 = 0.26 lM) and selectivity
(SI) = >192.3] comparable with reference drug celecoxib (IC50 value of 0.28 lM and selectivity index of
178.57). Moreover, the anti-inflammatory activity of selected compounds, which are the most selective
COX-2 inhibitors in the COX inhibition assay, was investigated in vivo using carrageenan-induced rat
paw edema model. Molecular modeling was conducted to study the ability of the active compounds to
bind into the active site of COX-2 which revealed a similar binding mode to SC-558, a selective COX-2
inhibitor.
Received 9 January 2012
Revised 15 March 2012
Accepted 21 March 2012
Available online 29 March 2012
Keywords:
Pyrazole
Pyrazoline
Synthesis
COX-2 inhibitors
Anti-inflammatory
Molecular modeling
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
safety profile and therapeutic potency of NSAIDs. The improved
safety profile of COX-2 inhibitors may allow the use of these new
Inflammation is a normal response to any noxious stimulus that
threatens the host and may vary from a localized response to a
generalized one.1 Non-steroidal anti-inflammatory drugs (NSAIDs)
are widely used for the treatment of rheumatism diseases, as
rheumatoid arthritis, and pain.2,3 NSAIDs achieve their anti-inflam-
matory action through an inhibitory effect on cyclooxygenase
enzyme (COX), a protein essential for prostaglandin biosynthesis
from arachidonic acid. COX exists under two isoforms, namely
COX-1 and COX-2. In general terms, COX-1 is responsible for pro-
tecting the gastric mucosa and maintaining homeostasis, whereas
COX-2 is induced by proinflammatory stimuli at the inflammatory
sites.4–6 Gastrointestinal (GI) irritation, bleeding and ulceration
side effects are associated with long term use of the classical
NSAIDs due to high COX-1 versus COX-2 selectivity.2,7,8 Thus, it is
reasonable to speculate increasing the specificity for COX-2 over
COX-1 is one of the strategies that could be employed to improve
agents for long-term prophylactic use in certain chronic dis-
eases.9–11 This has led intense efforts in search for potent and
selective COX-2 inhibitors which could provide anti-inflammatory
drugs with fewer risks.12–16 Research attempts in the discovery of
selective COX-2 inhibitors have produced many diarylheterocycles
and central ring pharmacophore templates (Fig. 1).12–14 However,
the recent success of pyrazole-based COX-2 inhibitors has high-
lighted the importance of these heterocycles in medicinal chemis-
try.12 Several classes of compounds having selective COX-2
inhibitory activity have been reported in the literature such as
SC-55817 and celecoxib18 (Fig. 1, A and B, respectively).
Thus our main objective is to design novel cyclic imides as
specific inhibitors of COX-2 in the hope that these molecules
may be further explored as powerful and novel non-ulcerogenic
anti-inflammatory lead-candidates. Our strategy is intended to ob-
tain potent anti-inflammatory activity with selective inhibition of
COX-2 using traditional medicinal chemistry techniques motivated
by the comparative modeling of a COX-1 and -2 complexed with A
and B together with the available pharmacophore. We have re-
cently studied a series of arylhydrazone and 1,5-diphenylpyrazole
⇑
Corresponding author. Tel./fax: +20 502247496.
E-mail address: naglaabdalaziz2005@yahoo.com (N.I. Abdel-Aziz).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.03.044

M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
3307
H2N
C: IC50= 0.45 M
S
O
O
N
N
N
N
CF3
CF3
O
Cl
CH
Br
SC-558 ( )
A
N
N
Fragment insersion
CF3
NH2
S
O
Cl
New small
CHO
N
N
molecular COX-2 inhibitors
O
CF3
O
E: IC50= 0.55 M
N
Cl
N
CH
CF3
N
Lead optimization
H3C
Et
D: IC50= 0.45 M
B
Celecoxib ( ): IC50= 0.30 M
Cl
H3CO
H3CO
CF3
CF3
N
H
C
F3C
N
N
N
N
F3C
8d
Cl
3d
Figure 1. Representative examples of selective COX-2 inhibitors (A–E) the designed pyrazoles (3d and 8d).
derivatives which were evaluated for their anti-inflammatory and
COX-2 inhibitory activities (Fig. 1).19 Continuing our studies on
pyrazole derivatives as attractive candidates as anti-inflammatory
agents and COX-2 inhibitors we have designed a number of new
pyrazole and pyrazoline derivatives and biologically evaluated
their in vitro COX-1/COX-2 and in vivo assessment as anti-inflam-
matory activities ( Fig. 1). Drug-likeness and molecular docking
methodology were used to identify the structural features required
for the COX-2 inhibition properties of these new series. The results
of this molecular docking could support the postulation that our
active compounds may act on the same enzyme target where
COX-2 inhibitors act confirming the molecular design of the
reported class of COX-2 inhibitors.
active methylene compounds is a very important transformation
in organic synthesis. Anhydrous ferric chloride has been emerged
as a powerful lewis acid catalyst for C–C bond formation between
active methylene compound and alcohol since direct substitution
of the hydroxyl group is difficult due to its poor leaving ability.20
Hence, reaction of b-diketone 1 with benzyl alcohol was performed
in dichloromethane in the presence of catalytic amount of anhy-
drous ferric chloride to afford compound 4. Cyclocondensation of
the later compound with hydrazine hydrate in refluxing ethanol
afforded the corresponding pyrazole derivative 5.
The structures of the prepared compounds were substantiated
by elemental and spectral analyses. For example, the structure of
compound 5 was confirmed by IR data which exhibited absorption
bands at 3200, 3096 cmꢀ1 (NH) and 1610, 1590 cmꢀ1 (C@C, C@N)
and no absorption bands for (C@O). The 1H-NMR spectra of com-
pound 5 showed the methylene of the benzyl group and NH at
4.2 and 14.10 ppm, respectively.
2. Results and discussion
2.1. Chemistry
2.1.2. Synthesis of compounds 6–9 (Scheme 2)
The synthetic route used to synthesize the target compounds is
outlined in Schemes 1–3.
Cyclocondensation of phenylhydrazine with trifluoromethyl-
1,3-diketone 1 afforded the corresponding 1,5-diphenyl pyrazole
derivative 6 as the major product. The NH2 in phenylhydrazine is
more nucleophilic than NH and would react preferentially with
the more reactive carbonyl group (COCF3) leading to the produc-
tion of 1,5-diphenyl pyrazole regioisomer as a major product spe-
cially if the reaction is carried out in acidic medium.18 Vilsmeier
formylation of 1,5-diphenyl pyrazole derivative 6 provided the cor-
responding aldehyde 7 which reacted with different substituted
aromatic amines affording the corresponding Schiff bases 8a–d in
a good yield. In addition, Knoevenagel condensation of compound
7 with malononitril in DMF and sodium acetate gave compound 9.
The structures of the isolated products 8a–d and 9 were estab-
2.1.1. Synthesis of compounds 1–5 (Scheme 1)
Claisen condensation of p-chloroacetophenone with ethyl tri-
fluoroacetate provided trifluoromethyl-1,3-dicarbonyl adduct 1 in
a good yield. Interaction between compound 1 and different alde-
hydes in ethanolic sodium hydroxide afforded the corresponding
chalcone derivatives 2a–f. The b-diketone moieties in the later
compounds were reacted with hydrazine hydrate in refluxing eth-
anol affording the corresponding pyrazole derivatives 3a–f. The
spectral and microanalytical data for compounds 2,3a–f were con-
sistent with their chemical structures. Moreover, alkylation of

3308
M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
F
F
F
F
F
F
O
O
ArCHO
O
O
NaOH/EtOH
Cl
Ar
Cl
1
2a-f
OH
FeCl3/CH2Cl2
NH2NH2
EtOH
F
F
F
F
F
O
F
Ar
O
N
N
Cl
4
Cl
3a-f
NH2NH2
EtOH
2a
Ar= 2-hydroxyphenyl ( and
3a
)
)
2b
3b
Ar= 4-hydroxyphenyl ( and
2c
3c
)
F
Ar= 4-methoxyphenyl ( and
2d
F
F
3d
Ar= 3,4-dimethoxyphenyl ( and
2e 3e
)
Ar= 3-nitrophenyl ( and
2f 3f
)
NH
Ar= 2-f uryl ( and
)
N
Cl
5
Scheme 1. Synthesis of diarylpyrazole derivatives (3 and 5).
F
F
F
O
NHNH2
C2H5OH/H2SO4
N
N
Cl
O
F
F
Cl
F
1
6
/DMF
POCl3
NH2
N
N
N
CH
(CN
)
2
Cl
N
2
N
R
N
Cl
F
Cl
F
F
F
EtOH
F
DMF/N
aOAc
HC
F
OHC
N
F
F
F
CN
7
NC
R
8a:R= 4-Br
8a-d
9
8b:R = 2-Cl
8c:R= 4-NO2
8d:R= 3,5-diCF3
Scheme 2. Synthesis of triarylpyrazole derivatives (8 and 9).

M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
3309
R
COCH3
N
O
HN
F
1- NH(CH3)2
2- CH COOH/HCl
/
3
F
O
+
O
3- NH2NH2.H2O/C2H5OH
N
N
H
H
R
10
11a:
11a,b
R= Br
11b: R= CH3
Scheme 3. Synthesis of spiropyrazoline derivatives (11).
Table 1
lished on the basis of their elemental and spectral analyses. IR data
for compound 9 exhibited characteristic absorption bands at 2211,
2205 (CN).
Data of the in vitro COX-1/COX-2 enzyme inhibition assay of the designed compounds
SIc
a
Compd No.
IC50
(
lM)
COX-1b
COX-2
2.1.3. Synthesis of compounds 11a,b (Scheme 3)
2a
2b
2c
2d
2e
2f
3a
3b
3c
3d
3e
3f
5
8a
8b
8c
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
0.22
>50
46.0
1.2
3.3
>1.09
Crossed aldol condensation between isatin and acetophenone
derivatives was reported to be achieved in the presence of dimeth-
ylamine as a basic catalyst followed by dehydration to give the cor-
responding chalcones.21 Nucleophilic Michael addition of
hydrazine hydrate to the later compounds gave spiropyrazo-
lines.22–25 Moreover, spiropyrazolines were prepared in a good
yield via a one-pot synthesis without the isolation of the interme-
diate chalcones.25,26 Hence, in this work, fluoroisatin reacted with
acetophenone derivatives in the presence of dimethylamine for
30 min at room temperature and the formed solid was treated with
glacial acetic acid and hydrochloric acid for another 30 min at
80 °C. Hydrazine hydrate was then added to the previous acidic
solution to give the corresponding spiropyrazoline derivatives
11a,b. The structures of the final products were established by
physical and spectral methods.
>41.67
>15.15
>100.00
>75.76
>1.24
0.50
0.66
40.2
44.4
0.48
2.1
0.29
0.43
48.0
>50
3.2
49.0
0.41
0.26
>50
>50
>50
3.1
>1.13
>104.17
>23.81
>172.41
>116.28
>1.04
>1.00
>15.63
>1.02
>121.95
>192.31
>1.0
>1.0
>1.0
8d
9
11a
11b
Diclofenac
Celecoxib
0.07
>178.57
2.2. Biological evaluation
0.28
a
IC50 value is the compound concentration required to produce 50% inhibition of
2.2.1. In vitro cyclooxygenase (COX) inhibition assay
COX-1 or COX-2 for means of two determinations and deviation from the mean is
It was found that the structurally simple compounds, pyrazoles
(Fig. 1, A–D), were obtained as a potent COX-2-selective inhibitor.
According to the previous rational, the compounds synthesized in
this work were evaluated for their ability to inhibit COX-1 and
COX-2 using an ovine COX-1/COX-2 assay kit (Catalog No.
560101, Cayman Chemicals Inc., Ann Arbor, MI, USA) according to
<10% of the mean value.
b
No inhibition of COX-1 up to 50
Selectivity index (COX-1 IC50/COX-2 IC50).
lM.
c
selectivity compared with compound 2f. It was also noticed that
replacement of the phenyl ring in compounds 2b–e and
3b–e with a furyl one as in compounds 2f and 3f led to dramatic de-
crease in the COX-2 activity. In addition, replacement of the benzyl-
idene fragment with a benzyl moiety abolishes the COX-2 activity
as in compound 5. Moreover, no activity was observed for the spi-
ropyrazoline compounds containing isatin moiety (11a and 11b).
the protocol recommended by the supplier. IC50 (lM) are deter-
mined as the means of two determinations acquired and the devi-
ation from the mean is <10% of the mean value. The selectivity
index (SI values) was defined as IC50(COX-1)/IC50(COX-2). In the as-
say system, the IC50 values of celecoxib on COX-1 and COX-2 were
determined to be >50 and 0.28
celecoxib is a selective COX-2 inhibitor. None of the tested com-
pounds showed inhibitory activity on COX-1 down to 50 M. Some
lM, respectively indicating that
l
2.2.2. In vivo anti-inflammatory studies
of the tested compounds (2d, 3b, 3d, 3e, 8c and 8d) were found to
be potent and selective similar to celecoxib against COX-2. IC50 val-
ues in lM (Table 1) acquired by determination of the in vitro ability
of the tested compounds to inhibit COX-2 showed that compound
Anti-inflammatory activity of eleven compounds, which
showed the highest specificity for COX-2 enzyme in the in vitro as-
say including compounds 2b–e, 3b–e, 8a, 8c and 8d was evaluated
by employing the well-known rat carrageenan-induced foot paw
edema model (Table 2). The percentages of edema reduction given
by the tested compounds, diclofenac sodium and celecoxib, as a
reference drugs, are depicted in Table 2. Moreover, ED50 was mea-
sured after 2 h of treatment with carrageenan at which maximum
percentage of inhibition of carrageenan-induced edema was
reached. All of the ED50 values were determined using three doses
8d with 3,5-ditrifluoromethyl substituents on the C4-phenyl ring
was more potent and selective COX-2 inhibitor (IC50 = 0.26
SI = >192.3) compared with the reference drug celecoxib
(IC50 = 0.28 M, SI = >178.57). Despite, COX-2 potency and selectiv-
lM,
l
ity were increased in the cyclized derivatives 3 relative to the cor-
responding chalcones 2, compound 3f had less potency and

3310
M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
Table 2
2.3.2. 2D-QSAR study
Data of the in vivo anti-inflammatory activity of the designed compounds against
carrageenan-induced rat paw edema (IP)
The COX-2 inhibition of the twenty tested compounds, pre-
sented in ꢀlog 1/IC50 (Table 3), were used in the 2D-QSAR studies
using VLife Molecular Design Suite.30 Compounds were divided in
training and test set. This was achieved by setting aside four com-
pounds as test set which have regularly distributed activity. Selec-
tion of molecules in the training set and test is a key and important
feature of any QSAR model. Therefore the care was taken in such a
way that biological activities of all compounds in test set lie within
the maximum and minimum value range of biological activities of
training set of compounds. A Uni-Column statistics for training set
and test set were generated to check correctness of selection
criteria for trainings and test set molecules. The maximum and
minimum value in training and set were compared in a way that:
% Inhibition of edemaa
ED50
c
Compd No.
100 mg/kg
200 mg/kg
mg/kg
mmol/kg
2b
2c
2d
2e
3b
3c
3d
3e
39.2
26.3
38.7
37.9
35.3
33.4
42.0
37.6
25.2
43.7
52.0
48.7
57.0
78.9
52.4
80.2
80.0
81.4
69.1
88.7
84.5
61.1
86.7
89.3
87.3
83.0
130 7.5
195 11.3
128 6.3
132 9.1
130 7.9
148 8.1
118 4.9
121 6.7
170 79.9
115 6.1
96 4.3
0.366
0.528
0.321
0.343
0.370
0.405
0.298
0.318
0.335
0.243
0.170
0.198
0.185
8a
8c
8d
Diclofenacb
Celecoxibc
63 7.5
70.4 0.8
1. The maximum value of ꢀlog 1/IC50 of test set should be less
than or equal to maximum value of ꢀlog 1/IC50 of training set.
2. The minimum value of ꢀlog 1/IC50 of test set should be higher
than or equal to minimum value of ꢀlog 1/IC50 of training set.
a
The results are expressed as means SEM (n = 4–6) after 2 h following IP of the
test compounds.
b
Values are determined at 50 and 100 mg/kg IP.
ED50 was the effective dose calculated after 2 h.
c
This observation showed that test set was interpolative and
derived within the minimum–maximum range of training set.
The mean and standard deviation of ꢀlog 1/IC50 values of sets
of training and test provide insights to relative difference of mean
and point density distribution of two sets. 2D-QSAR models were
generated for training set of 16 compounds using multiple linear
regression (MLR) method. The best QSAR model was selected on
the basis of value of statistical parameters like r2 (square of
correlation coefficient for training set of compounds), q2 (cross
validated r2), and pred_r2 (predictive r2 for the test set of
compounds). All QSAR model was validated and tested for its pre-
dictability using an external test set of 4 compounds. Statistical
results generated by 2D-QSAR analysis showed that QSAR models
have good internal as well as external predictability. The results
obtained for actual and predicted activity are presented in Table
3 and the residuals were found to be minimal (Table 3, Fig. 2).
The regression equation obtained for the different series of
compounds is given below.
of 50, 100 and 200 mg/kg in the test compounds and 25, 50 and
100 mg/kg in the reference drugs diclofenac and celecoxib. Out of
11 compounds, 10 compounds (2b, 2d, 2e, 3b–e, 8a, 8c and 8d)
were found to possess potent anti-inflammatory activity (61–89%
reduction in inflammation). Compound 8d showed the highest
anti-inflammatory activity among the tested compounds with
ED50 of 0.170 mmol/kg and the rest of compounds showed ED50
in the rang of 0.243–0.528 mmol/kg, as compared to diclofenac
and celecoxib (ED50 = 0.198 and 0.185 mmol/kg, respectively).
2.3. Molecular calculation and results
Modeling studies are required in order to construct molecular
models that incorporate all experimental evidence reported.27,28
These models are necessary to obtain a consistent and more pre-
cise picture of the biological active molecules at the atomic level
and furthermore, provide new insights that can be used to design
novel therapeutic agents.15,19
2D-QSAR model for COX-2 inhibition in ꢀlog 1/IC50
ꢀLog1=IC50 ¼ ꢀ6:1502ðꢂ0:9175ÞSssNHcount
þ 0:7191ðꢂ0:0576ÞT N N 4
2.3.1. ADME profiling
The bioavailability of the most active compounds 2b, 2d, 2e, 3b,
3d, 3e, 8c and 8d and the reference drug celecoxib was assessed
using ADME (absorption, distribution, metabolism and elimina-
tion; http://www.molinspiration.com/cgi-bin/properties) predic-
tion methods. In particular, we calculated the compliance of
compounds to the Lipinski’s rule of five.29 This approach has been
widely used as a filter for substances that would likely be further
developed in drug design programs. Briefly, this simple rule is
based on the observation that most orally administered drugs have
a molecular weight ꢁ500, a logPꢁ5, hydrogen bond donor sites ꢁ5
and hydrogen bond acceptor sites (N and O atoms) ꢁ10. In addi-
tion, we calculated the total polar surface area (TPSA) since it is an-
other key property that has been linked to drug bioavailability.
Thus, passively absorbed molecules with a TPSA >140 are thought
to have low oral bioavailability. Molecules violating more than one
of these rules may have problems with bioavailability. Predictions
of ADME properties for studied compounds (Supplementary data)
showed that compounds (2b, 2d, 2e, 3b, 3d, 3e and 8c) fulfilled this
rule, similarly to clinically used drugs (i.e., celecoxib). Compound
8d was violated from the rule of five by two represented by molec-
ular weight and logP values. Theoretically, these compounds
should present good passive oral absorption and differences in
their bioactivity can not be attributed to this property.
þ 0:2439ðꢂ0:0421ÞT O F 7 þ 20:4400n
¼ 16 r2 ¼ 0:8301 q2 ¼ 0:7416 F ꢀ test
¼ 27:9176 r2se ¼ 0:3315 q2se ¼ 0:4322 pred r2
¼ 0:8633 pred r2se ¼ 0:3346
A brief idea of statistical result of 2D-QSAR models, the require-
ment of different physicochemical parameters and their contribu-
tions (positive or negative influence on biological activity),
required for potential antibacterial activity was obtained. The
regression equation so obtained will be useful for the prediction
of biological activities of the designed series of compounds, in
future. The QSAR model generated is a tri-parametric model and
the statistical best model for COX-2 inhibition (ꢀlog 1/IC50) with
a coefficient of determination (r2) = 0.8301 was considered, as the
model showed an internal predictive power (q2 = 0.7416) of 74%
and a predictivity for the external test set (pred_r2 = 0.8633) of
about 86%. The low standard error of r2_se = 0.3315 demonstrates
accuracy of the model. The F-test value of 27.9176 shows the over-
all statistical significance level to be 99.99% of the model, which
means that the probability of failure for model is 1 in 10,000.

M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
3311
detail. As it is clear from X-ray crystal structure data,31,32 the
COX-2 binding site possesses an additional 2°-pocket that is ab-
sent in COX-1, which is highly relevant to the design of selective
COX-2 inhibitors. This COX-2 2°-pocket arises due to a confor-
mational change at Tyr355 that is attributed to the presence of
Ile523 in COX-1 relative to Val523 having a smaller side chain in
COX-2.31 It has also been reported that replacement of His513
in COX-1 by Arg513 in COX-2 plays a key role with respect to
the H-bond network in the COX-2 binding site. Access of ligands
to the 2°-pocket of COX-2 is controlled by histidine (His90), glu-
tamine (Gln192), and tyrosine (Tyr355).33 Interaction of Arg513
with the bound drug is a requirement for time dependent inhi-
bition of COX-2.34
Table 3
Observed, predicted COX-2 inhibition activity (ꢀlog 1/IC50) and residual values of the
designed compounds
Compd No.
ꢀlog 1/IC50
Observed
Predicted
Residual
2a
2b
2c
2d
2e
2f
3a
3b
3c
3d
3e
3f
1.66
0.08
0.52
ꢀ0.30
ꢀ0.18
1.60
1.65
0.42
ꢀ0.20
ꢀ0.42
0.01
0.01
ꢀ0.34
0.77
0.12
ꢀ0.19
1.38
1.66
0.22
1.65
ꢀ0.01
ꢀ0.13
ꢀ0.11
ꢀ0.11
ꢀ0.38
0.27
ꢀ0.07
ꢀ0.27
0.47
ꢀ0.21
ꢀ0.35
0.11
0.72
0.10
ꢀ0.32
ꢀ0.19
0.32
0.43
ꢀ0.54
ꢀ0.37
1.68
1.70
0.51
ꢀ0.43
0.01
1.41
1.77
0.78
All the docking calculations were performed using MOE 2008.10
software installed on 2.0G Core 2 Duo.15,19,35,36 The crystal struc-
tures of COX-2 enzymes complexed with SC-558 [6COX] were used
for the docking.17 The active site of the enzyme was defined to in-
clude residues within a 10.0 Å radius to any of the inhibitor atoms.
The automated docking program of MOE 2008.10 was used to dock
compounds 3d, 11a and 8d on the active sites of COX-2 enzymes (
Fig. 2). The most stable docking model was selected according to
the best scored conformation predicted by the MOE scoring func-
tion. The complexes were energy-minimized with a MMFF94 force
field37 till the gradient convergence 0.05 kcal/mol was reached.
The docking energies of ꢀ40.36, ꢀ5.01 and ꢀ43.68 kcal/mol were
obtained for 3d, 11a and 8d, respectively (Fig. 3). The compounds
3d and 8d could dock into the active site of COX-2 successfully (
Fig. 3). Compounds 3d and 8d produces a deep moving into the
hydrophilic pocket of COX-2. 3,5-di-triflouromethyl fragment of
compound 8d was able to reach the hydrophilic pocket and is in-
volved in classical hydrogen bonding with Gln192 (2.61, 2.93 Å)
and Arg513 (1.85, 2.14 Å). Such interactions are almost essential
for COX-2 inhibitory activity, as exemplified by the binding inter-
action of SC-558, an analogue of celecoxib cocrystallized in the
COX-2 active site.17 Moreover trifluoropyrazole core formed addi-
tional three hydrogen bonding with Tyr355 (2.09 Å) and Arg513
(2.98, 3.29 Å) while 4-chlorophenyl moiety showed weak interac-
tion with amino acid residue Trp387 (2.92 Å). In addition; such
interaction forces the triarylpyrazole core to adopt a specific orien-
tation at the top of the channel. This moiety is involved in hydro-
5
8a
8b
8c
8d
9
1.69
1.22
ꢀ0.39
ꢀ0.59
1.70
1.70
1.70
ꢀ0.18
ꢀ0.24
1.59
0.98
1.60
11a
11b
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
phobic interaction with Trp387, Tyr385, Val523, Ala527 and Phe518
.
Observed -log 1/IC50
The lateral pocket of COX-2 would therefore be responsible for
the COX-2 selectivity of 8d and contributed to stabilize the li-
gand–enzyme complexes ( Fig. 3). Similarly compound 3d was
modeled in the active site of COX-2 enzyme (Fig. 3). The dimethoxy
fragment of 3d is able to reach the hydrophilic pocket and is in-
volved in hydrogen bonding forming a three hydrogen bonding
interaction with His90 (3.07, 2.56 Å) and Gln192 (2.97 Å). Moreover
trifluoropyrazole core of compound 3d formed additional three
hydrogen bonding with Tyr355 (2.56, 3.20 Å) and Arg120 (2.76,
3.34 Å). In a similar manner of compound 8d, 4-chlorophenyl moi-
Figure 2. Plot between observed and predicted COX-2 inhibition of the compounds
in the training set and test set for the linear regression developed model.
The model is validated by
_ran_ pred_r2 = 0.01. The randomization test suggests that the
a a
_ran_r2 = 0.001, _ran_q2 = 0.01,
a
developed model have a probability of less than 1% that the model
is generated by chance.
The positive coefficients (36.00%) of T_N_N_4 (count of number
of Nitrogen atoms separated from any other Nitrogen atom by four
bonds) and (15.5%) of T_O_F_7 (count of number of Oxygen atoms
separated from any other fluorine atom by seven bonds) showed
that increase in the values of this descriptor is beneficial for the
COX-2 inhibition activity (like in compounds 2d, 2e, 3d, 3e and
8c, 8d). The next descriptor SssNHcount is a descriptor defines
the total number of –NH group connected with two single bonds,
indicates a negative contribution to the biologic activity (like in
compounds 5, 11a and 11b).
ety of 3d is involved in hydrophobic interaction with Trp387, Tyr385
,
Val523, Ala527 and Phe518. The complex generated by docking stud-
ies of 8d with COX-2 and superimposition with the structure of the
selective inhibitor, SC-558, co-crystallized with COX-2, illustrated
in Figure 3, shows that compound 8d can bind in the active site
of this enzyme in approximately similar fashion as the pyrazolic
prototype (SC-558) with additional hydrogen bonds. Additionally,
3,5-di-triflourophenyl system are positioned in the same region
as p-sulfonamido-phenyl ring of SC-558, while, the 4-chlorophenyl
fragment of compound 8d is close to the p-Br-phenyl ring of SC-
558, in the aromatic region of the active site lined by aromatic ami-
no acid residues such as Phe518, Tyr385 and Trp387, among others. In
short, the described interactions are typical of selective inhibitors
of COX-2, confirming the molecular design of the reported class
of anti-inflammatory pyrazole derivatives.38,39
2.3.3. Docking studies
The level of COX-2 inhibition and anti-inflammatory activities
of compounds 3d, 11a and 8d prompted us to perform molecular
docking studies to understand the ligand–protein interactions in

3312
M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
Figure 3. Orientation of 8d (color red) in COX-2 active pocket (Upper left panel; SC-558 is shown as cyan). Docking of compounds 8d (Upper right panel), 3d (Lower left
panel) and 11a (Lower right panel) into the active site of COX-2. Hydrogen bonds are shown in green.
Figure 4. Docking of compounds 3d (left panel) and 8d (right panel) into the active site of COX-1. Hydrogen bonds are shown in green.
The lower interaction energy observed for 11a rationalizes
the insufficient binding of spiropyrazoline core into the COX-2
active site than that of the pyrazole core of 8d (Fig. 3). The insuf-
ficient binding can be explained in terms of the occurrence of
only one hydrogen bond between the carbonyl group and
Arg120 (2.86 Å) and the absence of hydrogen bonding with
Gln192, His90 and Arg513 which is essential for COX-2 inhibitory
activity.

M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
3313
On the other hand, compounds 3d and 8d as representative
examples were docked into the active site of COX-1 enzyme to
understand the inactivity of such compounds toward COX-1 inhi-
bition (Fig. 4). The crystal structures of COX-1 enzymes com-
plexed with flurbiprofen40 [1EQH] were used for the docking
using the same protocol described above. The two compounds
could dock into the active site of COX-1 successfully. The binding
energies of ꢀ4.22 and ꢀ7.67 kcal/mol were obtained for 3d and
8d, respectively (Fig. 3). The insufficient binding of both com-
pounds can be explained in terms of the occurrence of very weak
hydrogen bonds in which the 3,4-dimethoxyphenyl group of com-
pound 3d formed such weak bonds with Tyr355 (3.48 Å) while
3,5-diCF3phenyl group of compound 8d formed similar weak
bonds with Tyr355 (3.54 Å). The hydrophobic phenyl rings of 3d
and 8d were surrounded by active site amino acid residues
were recorded on FT-NMR spectrometer (200 MHz) Gemini Varian
and 13C NMR on Ultra shield Bruker (125 MHz) using DMSO-d6
and/or CDCl3 and using TMS as internal standard (chemical shifts
in d ppm). Mass spectra were measured on JEOL JMS-600H spec-
trometer. Elemental analysis was carried out for C, H and N at
the Microanalytical Centre of Cairo University. All reagents were
purchased from the Aldrich Chemical Company. The well-known
compounds,
(1),18 5-(4-chlorophenyl)-1-phenyl-3-(trifluoromethyl)-1H-pyra-
zole and 5-(4-chlorophenyl)-1-phenyl-3-(trifluoro-
(6)41,42
methyl)-1H-pyrazole-4-carbaldehyde were prepared
(7)19
following the procedures reported in the literature.
1-(4-chlorophenyl)-4,4,4-trifuorobutane-1,3-dione
4.2. General method for synthesis of 2-(2-arylidene)-1-(4-chloro
phenyl)-4,4,4-trifuorobutane-1,3-dione (2a–f) (Scheme 1)
Tyr385, Val349, Trp387 and Ser530
.
A mixture of compound 1 (2.50 g, 10 mmol) and the appropriate
aldehyde (10 mmol) in ethanolic sodium hydroxide solution (5%,
40 mL) was refluxed for 2 h. The reaction mixture was poured onto
ice/cold water and the formed precipitate was filtered off and
recrystallized from ethanol to give compounds 2a–f.
3. Conclusion
Pyrazole and pyrazoline derivatives were synthesized and
screened for COX-1/COX-2 inhibition and anti-inflammatory
activity. Compounds which showed significant COX-2 inhibition
were subjected to anti-inflammatory studies. Compound 8d exhi-
4.2.1. 1-(4-Chlorophenyl)-2-(2-hydroxybenzylidene)-4,4,4-triflu-
orobutane-1,3-dione (2a)
bit optimal COX-2 inhibitory potency (IC50 = 0.26 lM) and selec-
Yield, 70%; mp 110–111 °C; IR (KBr)
m
max/cmꢀ1 3328 (OH),
tivity (SI) = >192.3] comparable with celecoxib, so it appears
promising. Various physicochemical indices are helpful for the
understanding of COX-2 inhibition results, as shown by our 2D-
QSAR study. The 2D-QSAR results could not be addressed to a
concrete drug–receptor interaction, but they can reveal trends
in the relationship between ligand structures and their activities
for our set of COX-2 inhibitors. Analysis of 2D-QSAR models pro-
vides details on the fine relationship linking structure and activity
and offers clues for structural modifications that can improve the
activity. These trends should prove to be an essential guide for
the future work. Molecular docking studies further help in under-
standing the various interactions between the ligands and en-
zyme active sites in detail and thereby help to design novel
potent inhibitors. It is clear that the trifluoromethyl moieties of
8d inserts deep inside the COX-2 2°-pocket and forming strong
hydrogen bond with Gln192 (2.61, 2.93 Å) and Arg513 (1.85,
2.14 Å). This result was corroborated by molecular docking stud-
ies with COX-2 inhibition, which showed that this compound pre-
sents the pharmacophoric requisites for COX-2 inhibition. Indeed
molecular docking studies further supported the strong inhibitory
activity of 8d and further help understanding the various interac-
tions between the ligands and enzyme active sites in detail and
thereby help to design novel potent inhibitors. The structure–
activity relationships acquired show that appropriately substi-
tuted triarylpyrazole have the necessary geometry to provide
potent and selective inhibition of the COX-2 isozyme, and to
exhibit excellent anti-inflammatory activities. Furthermore analy-
sis of the predicted ADME properties for newly prepared com-
pounds opens the possibility for further optimization of studied
compounds. Compound 8d is found to be too lipophilic due to
the larger number of trifluoromethyl fragment; therefore, it could
serve as lead molecule for further optimization of both specific
biological activity and ADME properties.
1650, 1589 (C@O). 1H NMR (DMSO-d6); d 9.71 (br s, 1H, OH),
8.08–7.20 (m, 8H, Ar-H), 7.11 (s, 1H, @CH). Anal. Calcd for
C17H10ClF3O3: C, 57.56; H, 2.84. Found: C, 57.43; H, 2.91.
4.2.2. 1-(4-Chlorophenyl)-2-(4-hydroxybenzylidene)-4,4,4-triflu-
orobutane-1,3-dione (2b)
Yield, 60%; mp 120–121 °C; 1H NMR (DMSO-d6); d 9.61 (br s,
1H, OH), 8.08–7.10 (m, 9H, Ar-H, @CH). Anal. Calcd for
C17H10ClF3O3: C, 57.56; H, 2.84. Found: C, 57.30; H, 2.60.
4.2.3. 1-(4-Chlorophenyl)-2-(4-methoxybenzylidene)-4,4,4-
trifluorobutane-1,3-dione (2c)
Yield, 55%; mp 90–92 °C; 1H NMR (DMSO-d6); d 8.18–7.20 (m,
8H, Ar-H), 7.08 (s, 1H, @CH), 3.81 (s, 3H, OCH3). 13C NMR (CDCl3):
d 191.18, 159.33, 138.98, 136.34, 134.06, 131.85, 129.71, 129.10,
128.58, 127.17, 118.75, 118.59, 115.25, 55.67. Anal. Calcd for
C18H12ClF3O3: C, 58.63; H, 3.28. Found: C, 58.35; H, 3.10.
4.2.4. 1-(4-Chlorophenyl)-2-(3,4-dimethoxybenzylidene)-4,4,4-
trifluorobutane-1,3-dione (2d)
Yield, 60%; mp 105–107 °C; IR (KBr)
m
max/cmꢀ1 1645, 1595
(C@O). 1H NMR (DMSO-d6); d 8.20–7.00 (m, 8H, Ar-H, @CH), 3.90
(s, 3H, OCH3), 3.80 (s, 3H, OCH3). 13C NMR (DMSO): d 187.89,
151.41, 148.98, 145.06, 137.83, 136.47, 130.33, 128.80, 128.47,
127.36, 124.17, 119.18, 111.51, 55.71, 55.57. MS m/z (%); 399.00
(1.72, M++1), 398.00 (1.77, M+), 302.00 (100.00), 160.05 (0.83),
158.00 (2.10), 109.05 (1.26), 107.05 (3.36). Anal. Calcd for
C19H14ClF3O4: C, 57.23; H, 3.54. Found: C, 57.00; H, 3.20.
4.2.5. 1-(4-Chlorophenyl)-2-(3-nitrobenzylidene)-4,4,4-triflu-
orobutane-1,3-dione (2e)
Yield, 60%; mp 125–127 °C; 1H NMR (DMSO-d6); d 8.18–7.10
(m, 9H, @CH, Ar-H). Anal. Calcd for C17H9ClF3NO4: C, 53.21; H,
2.36; N, 3.65. Found: C, 53.00; H, 2.10; N, 3.30.
4. Experimental
4.1. General
4.2.6. 1-(4-Chlorophenyl)-2-((furan-2-yl)methylene)-4,4,4-triflu-
orobutane-1,3-dione (2f)
Yield, 55%; mp 130–131 °C; 1H NMR (CDCl3); d 8.40–6.80 (m,
8H, Ar-H, furyl-H, @CH). MS m/z (%); 328.00 (37.78, M+), 327.00
(32.94), 264.00 (32.25), 246.00 (100.00), 168.00 (30.25) 82.00
Melting points was determined using Fisher-Johns melting
point apparatus and are uncorrected. Infrared (IR) spectra were
recorded on Mattson 5000 FT-IR spectrometer. 1H NMR spectra

3314
M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
(44.45). Anal. Calcd for C15H8ClF3O3: C, 54.81; H, 2.45. Found: C,
54.51; H, 2.15.
pressure and the residue was crystallized from isopropanol to give
compound 4. Yield, 40%; mp 90–91 °C; 1H NMR (CDCl3); d 7.81–
7.40 (m, 9H, Ar-H), 3.8 (t, 1H, CH), d 3.2 (d, 2H, CH2, J = 7.5 Hz).
Anal. Calcd for C17H12ClF3O2: C, 59.93; H, 3.55. Found: C, 60.10;
H, 3.83.
4.3. General method for synthesis of 3,4,5-trisubstituted pyrazole
derivatives (3a–f) (Scheme 1)
A mixture of chalcone derivatives 2a–f (3 mmol) and hydrazine
hydrate (0.32 g, 3 mmol) in absolute ethanol (30 mL) was refluxed
for 3–6 h. Upon cooling, the obtained product was filtered off and
recrystallized from ethanol to give compounds 3a–f.
4.5. General method for synthesis of 4-benzyl-3-(4-
chlorophenyl)-5-(trifluoromethyl)-1H-pyrazole (5) (Scheme 1)
A mixture compound 4 (1.02 g, 3 mmol) and hydrazine hydrate
(0.32 g, 3 mmol) in absolute ethanol (30 mL) was refluxed for 6 h.
The reaction mixture was evaporated under reduced pressure
and the residue was crystallized from chloroform to give com-
4.3.1. 2-((3-(4-Chlorophenyl)-5-(trifluoromethyl)-4H-pyrazol-4-
ylidene)methyl)phenol (3a)
Yield, 60%; mp 109–110 °C; IR (KBr) m
max/cmꢀ1 3340 (OH), 1600,
pound 5. Yield, 50%; mp 148–150 °C; IR (KBr)
m
max/cmꢀ1 3200,
1589 (C@N). 1H NMR (DMSO-d6); d 9.71 (br s, 1H, OH exchangable
with D2O), 8.18–7.20 (m, 8H, Ar-H), 6.15 (s, 1H, @CH). Anal. Calcd
for C17H10ClF3N2O: C, 58.22; H, 2.87; N, 7.99. Found: C, 58.10; H,
2.57; N, 8.11.
3096 (NH), 1610, 1590 (C@C, C@N) 1H NMR (CDCl3); d 14.10 (s,
1H, NH), 7.86–7.27 (m, 9H, Ar-H), 4.2 (s, 2H, CH2). Anal. Calcd for
C17H12ClF3N2: C, 60.64; H, 3.59; N, 8.32. Found: C, 60.35; H, 3.81;
N, 8.62.
4.3.2. 4-((3-(4-Chlorophenyl)-5-(trifluoromethyl)-4H-pyrazol-4-
ylidene)methyl)phenol (3b)
4.6. General method for synthesis of N-((5-(4-chlorophenyl)-1-
phenyl-3-(trifluoromethyl)-1H-pyrazol-4-yl)methylene)aniline
derivatives (8a–d) (Scheme 2)
Yield, 65%; mp 123–125 °C; 1H NMR (DMSO-d6); d 9.80 (br s,
1H, OH exchangable with D2O), 8.18–7.10 (m, 9H, Ar-H @CH). 13C
NMR (CDCl3): d 164.12, 144.87, 141.23, 140.99, 136.77, 130.00,
127.56, 126.05, 124.69, 122.07, 119.33, 117.01. Anal. Calcd for
To a mixture of compound 7 (17.53 g, 0.05 mol) and the appro-
priate aromatic aldehyde (0.05 mol) in absolute ethanol (20 mL),
glacial acetic acid (0.5 mL) was added and the reaction mixture
was refluxed for 5 h. The resulting clear solution was evaporated
under reduced pressure and the residue was crystallized from
aqueous ethanol to give compounds 8a–d.
C17H10ClF3N2O: C, 58.22; H, 2.87; N, 7.99. Found: C, 58.00; H,
2.99; N, 8.13.
4.3.3. 3-(4-Chlorophenyl)-4-(4-methoxybenzylidene)-5-(trifluoro-
methyl)-4H-pyrazole (3c)
Yield, 45%; mp 110–111 °C; 1H NMR (CDCl3); d 8.22–7.30 (m,
8H, Ar-H), 7.18 (s, 1H, @CH), 3.75 (s, 3H, OCH3). 13C NMR (CDCl3):
d 162.00, 143.29, 142.70, 142.40, 134.51, 128.53, 125.85, 125.38,
123.08, 120.94, 118.80, 116.66, 62.40. Anal. Calcd for
4.6.1. 4-Bromo-N-((5-(4-chlorophenyl)-1-phenyl-3-
(trifluoromethyl)-1H-pyrazol-4-yl)methylene)aniline (8a)
Yield, 80%; mp 125–126 °C; 1H NMR (CDCl3); d 7.95–7.40 (m,
14H, CH@N, Ar-H). 13C NMR (CDCl3): d 184.45, 153.21, 145.03,
141.11, 138.91, 135.45, 132.00, 131.12, 130.20, 129.86, 129.75,
128.95, 128.13, 126.32, 122.53, 120.57, 119.75. MS m/z (%);
505.00 (0.71, M+), 504.00 (0.57), 360.40 (0.78), 358.20 (0.80),
150.00(100.00), 144.05 (3.50), 142.00 (3.60). Anal. Calcd for
C18H12ClF3N2O: C, 59.27; H, 3.32; N, 7.68. Found: C, 59.53; H,
3.12; N, 7.93.
4.3.4. 3-(4-Chlorophenyl)-4-(3,4-dimethoxybenzylidene)-5-
(trifluoromethyl)-4H-pyrazole (3d)
Yield, 60%; mp 105–107 °C; 1H NMR (DMSO-d6); d 7.70–7.03
(m, 7H, Ar-H), 6.93 (s, 1H, @CH), 3.75 (s, 3H, OCH3), 3.62 (s, 3H,
OCH3). Anal. Calcd for C19H14ClF3N2O2: C, 57.81; H, 3.57; N, 7.10.
Found: C,57.62; H, 3.32; N, 7.31.
C23H14BrClF3N3: C, 54.73; H, 2.80; N, 8.33. Found: C, 55.00; H,
2.61; N, 8.60.
4.6.2. 2-Chloro-N-((5-(4-chlorophenyl)-1-phenyl-3-
(trifluoromethyl)-1H-pyrazol-4-yl)methylene)aniline (8b)
Yield, 85%; mp 139–140 °C; 1H NMR (CDCl3); d 7.83–7.52 (m,
14H, CH@N, Ar-H). 13C NMR (CDCl3): d 184.45, 153.21, 145.00,
141.00, 138.91, 135.44, 132.07, 130.20, 129.86, 129.74, 128.93,
128.11, 126.30, 122.52, 120.31, 119.73. Anal. Calcd for
4.3.5. 3-(4-Chlorophenyl)-4-(3-nitrobenzylidene)-5-(trifluoro-
methyl)-4H-pyrazole (3e)
Yield, 55%; mp 98–99 °C; 1H NMR (DMSO-d6); d 8.18–7.10 (m,
9H, @CH, Ar-H). Anal. Calcd for C17H9ClF3N3O2: C, 53.77; H, 2.39;
N, 11.07. Found: C, 53.52; H, 2.15; N, 11.31.
C23H14Cl2F3N3: C, 60.02; H, 3.07; N, 9.13. Found: C, 60.33; H,
3.30; N, 9.35.
4.3.6. 3-(4-Chlorophenyl)-4-((furan-2-yl)methylene)-5-
(trifluoromethyl)-4H-pyrazole (3f)
4.6.3. N-((5-(4-chlorophenyl)-1-phenyl-3-(trifluoromethyl)-1H-
pyrazol-4-yl)methylene)-4-nitroaniline (8c)
Yield, 55%; mp 96–98 °C; 1H NMR (CDCl3); d 8.50-6.70 (m, 8H,
Ar-H, furyl-H, @CH). MS m/z (%); 324.60 (10.14, M+), 262.15
(9.90), 245.95 (100.00), 136.70 (15.46), 82.45 (10.85) 68.20
(15.59). Anal. Calcd for C15H8ClF3N2O: C, 55.49; H, 2.48; N, 8.63.
Found: C, 55.21; H, 2.22; N, 8.32.
Yield, 75%; mp 235–237 °C; 1H NMR (CDCl3); d 7.81–7.40 (m,
14H, CH@N, Ar-H). Anal. Calcd for C23H14ClF3N4O2: C, 58.67; H,
3.00; N, 11.90. Found: C, 58.89; H, 3.30; N, 12.10.
4.6.4. 3,5-bi(Trifluoromethyl)-N-((5-(4-chlorophenyl)-1-phenyl-
3-(trifluoromethyl)-1H-pyrazol-4-yl)methylene)aniline (8d)
Yield, 80%; mp 235–237 °C; 1H NMR (CDCl3); d 7.80–7.45 (m,
13H, CH@N, Ar-H). 13C NMR (CDCl3): d 184.45, 153.20, 145.11,
141.21, 138.91, 135.45, 132.01, 131.11, 130.20, 129.85, 129.75,
128.94, 128.13, 126.81, 122.53, 120.44, 119.75, 118.51. Anal. Calcd
for C25H13ClF9N3: C, 53.44; H, 2.33; N, 7.48. Found: C, 53.75; H,
2.00; N, 7.55.
4.4. General method for synthesis of 2-benzyl-1-(4-chlorophenyl)-
4,4,4-trifluorobutane-1,3-dione (4) (Scheme 1)
To a stirred solution of compound 1 (0.25 g, 1 mmol) and benzyl
alcohol (0.10 g, 1 mmol) in dry dichloromethane (4 mL) was added
anhydrous ferric chloride (0.16 g, 0.1 mmol). The mixture was re-
fluxed for 3 h. The reaction mixture was evaporated under reduced

M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
3315
4.7. General method for synthesis of 2-(5-(4-chlorophenyl)-1-
5.2. Anti-inflammatory experiment
phenyl-3-(trifluoromethyl)-1H-pyrazol-4-yl)methylene)-
malononitrile (9) (Scheme 2)
The anti-inflammatory activity was evaluated using in vivo rat
carrageenin-induced foot paw edema model reported
previously.15,19
A mixture of compound 7 (1.75 g, 0.005 mol), malononitril
(0.30 g, 0.005 mol) and sodium acetate (3.28 g, 0.04 mol) in abso-
lute ethanol (30 mL) was refluxed for 10 h. The solvent was evap-
orated under reduced pressure and the residue obtained was
triturated with water, filtered and dried to give compound 9 which
crystallized DMF/water.
6. Molecular modeling methods
6.1. 2D-descriptor calculations
Yield, 40%; mp 105–110 °C; IR (KBr) m
max/cmꢀ1 2211,2205 (CN),
6.1.1. Data set
1579 (C@N). 1H NMR (CDCl3); d 8.00–7.10 (m, 10H, CH@C, Ar-H).
13C NMR (CDCl3): d 163.12, 138.21, 137.35, 136.00, 132.50,
129.89, 129.50, 129.07, 128.90, 128.51, 126.11, 116.00, 113.11,
100.00, 90.21. MS m/z (%); 399.00 (1.72, M++1), 398.00 (1.77,
M+), 302.95(35.98), 302.00 (100.00), 271.00 (42.35), 139.00
(29.39), 111.00 (22.13). Anal. Calcd for C20H10ClF3N4: C, 60.24; H,
2.53; N, 14.05. Found: C, 60.00; H, 2.70; N, 14.20.
A data set of 20 compounds was used for the present 2D QSAR
study. There is high structural diversity and a sufficient range of
the biological activity in the selected series of these derivatives.
The biological activity values (IC50) were converted to negative log-
arithmic scale (ꢀlog 1/IC50) and subsequently used as the depen-
dent variable for the QSAR analysis. The energy-minimized
geometry of the compounds was used for the calculation of the
various 2D descriptors (Individual, Chi, ChiV, Path count, ChiChain,
ChiVChain, Chainpathcount, Cluster, Pathcluster, Kapa, Element
Count, Estate number, Estate contribution, Hydophillic–hydopho-
bic and Polar surface area). The various alignment-independent
(AI) descriptors were also calculated. For calculation of alignment,
the independent descriptor was assigned the utmost three attri-
butes. The first attribute was T to characterize the topology of
the molecule. The second attribute was the atom type, and the
third attribute was assigned to atoms taking part in the double
or triple bond. The pre-processing of the independent variables
(i.e., 2D descriptors) was done by removing invariable (constant
column).
4.8. General method for synthesis of 50-(4-substituted phenyl)-
5-fluoro-20,40-dihydrospiro(indol-3,30-pyrazol)-2(1H)-one 11a,b
(Scheme 3)
To a mixture of fluoroisatin (0.82 g, 5 mmol) and acetophenone
derivatives (5 mmol), dimethylamine (5 drops) was added and the
mixture was stirred for 30 min at room temperature. To the formed
solid, glacial acetic acid (10 mL) and concentrated hydrochloric
acid (3 drops) were added and the reaction mixture was heated
at 80 °C for 30 min. Hydrazine hydrate (6 mmol) was added to
the previous reaction mixture and heating was continued at
80 °C for 1 h to give spiro compounds 11a,b which crystallized
from DMF.
6.1.2. Statistical analysis
The descriptors were taken as independent variables and COX-2
inhibition as dependent variable. Multiple linear regression (MLR)
method of analysis was used to derive the 2D QSAR equations. The
developed QSAR models are evaluated using the following statisti-
cal measures: r2, (the squared correlation coefficient); r2se, (stan-
dard error of squared correlation coefficient); F test, (Fischer’s
value) for statistical significance; q2, (cross-validated correlation
coefficient); q2_se, (standard error of cross-validated square corre-
lation co-efficient); pred_r2, (r2 for external test set); pred_r2se,
(standard error of predicted squared regression); Z score, (Z score
calculated by the randomization test); best_ran_q2, (highest q2 va-
lue in the randomization test). The regression coefficient r2 is a rel-
ative measure of fit by the regression equation. It represents the
part of the variation in the observed data that is explained by the
regression. However, a QSAR model is considered to be predictive,
if the following conditions are satisfied: r2 >0.6, q2 >0.6 and pred_r2
>0.5.43,44 The F-test reflects the ratio of the variance explained by
the model and the variance due to the error in the regression. High
values of the F-test indicate that the model is statistically signifi-
cant. The low standard error of r2 (r2_se), q2 (q2_se) and pred_r2
(Pred_r2se) shows absolute quality of fitness of the model.
4.8.1. 50-(4-Bromophenyl)-5-fluoro-20,40-dihydrospiro(indol-
3,30-pyrazol)-2(1H)-one (11a)
Yield, 45%; mp 120–122 °C; IR (KBr)
m
max/cmꢀ1 3210, 3087
(NH), 1706 (C@O), 1587 (C@N). 1H NMR (DMSO-d6); d 10.2 (s,
1H, NH), 7.85–7.01 (m, 7H, Ar-H), 6.6 (s, 1H, NH), 3.5 (d, 2H, CH2,
J = 15.0 Hz). 13C NMR (DMSO): d 167.04, 156.97,139.92, 137.93,
137.21, 135.72, 131.43, 129.98, 127.72, 125.71, 123.80, 120.52,
62.92, 50.00. MS m/z (%); 359.75 (5.62, M+), 358.00 (48.59),
356.05 (33.66), 350.55 (43.64), 348.65 (46.40), 236.10 (100.00),
92.55 (11.10), 90.30 (13.21). Anal. Calcd for C16H11BrFN3O: C,
53.35; H, 3.08; N, 11.67. Found: C, 53.80; H, 3.20; N, 11.92.
4.8.2. 50-(4-Methylphenyl)-5-fluoro-20,40-dihydrospiro(indol-
3,30-pyrazol)-2(1H)-one (11b)
Yield, 50%; mp 142–143 °C; IR (KBr)
m
max/cmꢀ1 3417, 3031
(NH), 1600 (C@O), 1562 (C@N). 1H NMR (DMSO-d6); d 11.0 (s,
1H, NH), 7.75–7.00 (m, 7H, Ar-H), 6.8 (s, 1H, NH), 3.5 (d, 2H, CH2,
J = 15.1 Hz), 2.5(s, 3H, CH3). 13C NMR (DMSO):
d 172.00,
157.41,139.36, 137.24, 135.24, 130.00, 129.63, 129.45, 128.96,
126.55, 126.89, 124.00, 63.00, 50.00, 20.87. MS m/z (%); 294.00
(50.62, M+-H), 278.10 (100.00), 277.05 (52.29), 132.25 (57.03),
129.45 (63.46), 111.75(62.19). Anal. Calcd for C17H14FN3O: C,
69.14; H, 4.78; N, 14.23. Found: C, 69.40; H, 4.95; N, 14.61.
6.2. Docking methodology
Docking studies have been performed using MOE 2008.10.35
With this purpose, crystal structure of both COX-2/SC-558 (a selec-
tive inhibitor) and COX-1/flurbiprofen (a non-selective-inhibitor)
complexes (PDB codes: 6COX and 1EQH, respectively) was ob-
tained from the Protein Data Bank in order to prepare both proteins
for docking studies. Docking procedure was followed using the
standard protocol implemented in MOE 2008.10 and the geometry
of resulting complexes was studied using the MOE’s Pose Viewer
utility.
5. Biological evaluation
5.1. In vitro cyclooxygenase (COX) inhibition assay
The ability of the test compounds to inhibit ovine COX-1 and
COX-2 was determined using an enzyme immuno assay (EIA) (Cat-
alog No. 560101, Cayman chemicals, Ann Arbor, MI) according to
our previous reports.15,19

3316
M. A.-A. El-Sayed et al. / Bioorg. Med. Chem. 20 (2012) 3306–3316
17. Wang, J. L.; Limburg, D.; Graneto, M. J.; Springer, J.; Hamper, J. R. B.; Liao, S.;
Acknowledgments
Pawlitz, J. L.; Kurumbail, R. G.; Maziasz, T.; Talley, J. J.; Kiefer, J. R.; Carter, J.
Bioorg. Med. Chem. Lett. 2010, 20, 7159.
The authors extend their appreciation to the Deanship of Scien-
tific Research at King Saud University for funding the work through
the research group Project No. RGP-VPP-163. The authors wish to
express gratitude to V-life Science Technologies Pvt. Ltd for provid-
ing the trial full version software. Our sincere acknowledgments to
Chemical Computing Group Inc., 1010 Sherbrooke Street West,
Suite 910, Montreal, H3A 2R7, Canada, for its valuable agreement
to use the package of MOE 2008.10 software.
18. Penning, T. D.; Talley, J. J.; Bertenshaw, S. R.; Carter, J. S.; Collins, P. W.; Docter,
S.; Graneto, M. J.; Lee, L. F.; Malecha, J. W.; Miyashiro, J. M.; Rogers, R. S.; Rogier,
D. J.; Yu, S. S.; Anderson, G. D.; Burton, E. G.; Cogburn, J. N.; Gregory, S. A.;
Koboldt, C. M.; Perkins, W. E.; Seibert, K.; Veenhuizen, A. W.; Zhang, Y. Y.;
Isakson, P. C. J. Med. Chem. 1997, 40, 1347.
19. El-Sayed, M. A.-A.; Abdel-Aziz, N. I.; Abdel-Aziz, A. A.-M.; El-Azab, A. S.; Asiri, Y.
A.; ElTahir, K. E. H. Bioorg. Med. Chem. 2011, 19, 3416.
20. Jana, U.; Biswas, S.; Maiti, S. Tetrahedron Lett. 2007, 48, 4065.
21. Das, N.B.; Mittra, A.S. Ind. J. Chem. 1978, 16b, 638.
22. Azarifar, D.; Shaebanzadeh, M. Molecules 2002, 7, 885.
23. Le‘vai, A. J. Heterocycl. Chem. 2002, 39, 1.
24. Levai, A. ARKIVOC 2005, iv, 344.
25. Azizian, J.; Shaebanzadeh, M.; Hatamjafari, F.; Mohammadizadeh, M. R.
ARKIVOC 2006, xi, 47.
Supplementary data
26. McGettigan, P.; Henry, D. JAMA 2006, 296, 1633.
27. El-Ayaan, U.; Abdel-Aziz, A. A.-M.; Al-Shihry, S. Eur. J. Med. Chem. 2007, 42,
1325.
28. Goda, F. E.; Abdel-Aziz, A. A.-M.; Ghoneim, H. A. Bioorg. Med. Chem. 2005, 13,
3175.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.03.044.
29. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Deliv. Rev.
2001, 46, 3.
References and notes
30. V-Life Molecular Design Suite 3.5, VLife Sciences Technologies Pvt. Ltd;
www.Vlifesciences.com.
31. Meade, E. A.; Smith, W. L.; DeWitt, D. L. J. Biol. Chem. 1993, 268, 6610.
32. Kurumbail, R. G.; Stevens, A. M.; Gierse, J. K.; McDonald, J. J.; Stegemen, R. A.;
Pak, J. Y.; Gildehaus, D.; Miyashiro, J. M.; Penning, T. D.; Seibert, K.; Isakson, P.
C.; Stallings, W. C. Nature 1996, 384, 644.
1. Williams, D. A.; Lemke, T. L. Non-steroidal anti-inflammatory drugs. In Foye’s
Principles of Medicinal Chemistry; Lippincott: Williams & Wilkins, 2002; pp 751–
793. 5th ed.
2. Dannhardt, G.; Kiefer, W.; Krämer, G.; Maehrlein, S.; Nowe, U.; Fiebich, B. Eur. J.
Med. Chem. 2000, 35, 499.
3. Dubost, J. J.; Soubrier, M.; Sauvezie, B. Rev. Med. Intern. 1999, 20, 171.
4. Frank, M. M.; Fries, L. F. Immunol. Today 1991, 322, 12.
5. Collier, H. O. J. Nature 1971, 17, 232.
6. Cryer, B.; Feldman, M. Arch. Intern. Med. 1992, 152, 1145.
7. Dannhardt, G.; Kiefer, W. Eur. J. Med. Chem. 2001, 36, 109.
8. Bayly, C. I.; Black, C.; Leger, S.; Ouimet, N.; Ouellet, M.; Percival, M. D. Bioorg.
Med. Chem. Lett. 1999, 9, 307.
33. Llorens, O.; Perez, J. L.; Palomer, A.; Mauleon, D. Bioorg. Med. Chem. Lett. 1999, 9,
2779.
34. Garavito, R. M.; DeWitt, D. L. Biochem. Biophys. Acta 1999, 1441, 278.
35. MOE 2008.10 of Chemical Computing Group. Inc.
36. El-Azab, A. S.; Al-Omar, M. A.; Abdel-Aziz, A. A.-M.; Abdel-Aziz, N. I.; El-Sayed,
M. A.-A.; Aleisa, A. M.; Sayed-Ahmed, M. M.; Abdel-Hamide, S. G. Eur. J. Med.
Chem. 2010, 45, 4188.
37. Halgren, T. A. J. Comp. Chem. 1996, 17, 490.
38. Sondhi, S. M.; Rani, R.; Roy, P.; Agrawal, S. K.; Saxena, A. K. Bioorg. Med. Chem
Lett. 2009, 19, 1534.
39. Lima, L. M.; Castro, P.; Machado, A. L.; Fraga, C. A. M.; Lugnier, C.; Concalves de
Moraes, V. L.; Barreiro, E. J. Bioorg. Med. Chem. 2002, 10, 3067.
40. Selinsky, B. S.; Gupta, K.; Sharkey, C. T.; Loll, P. J. Biochemistry 2001, 40, 5172.
40. Fustero, S.; Román, R.; Sanz-Cervera, J. F.; SimÓn-Fuentes, A.; Cuñat, A. C.;
Villanova, S.; Murgu, M. J. Org. Chem. 2008, 73, 3523.
42. Singh, S. P.; Kumar, V.; Aggarwal, R. J. Heterocycl. Chem. 2006, 43, 1003.
43. Abdel-Aziz, A. A.-M.; Asiri, Y. A.; Al-Agamy, M. H. M. Eur. J. Med. Chem. 2011, 46,
5487.
9. Vane, J. R. Nat. New Biol. 1971, 231, 232.
10. Buttar, N. S.; Wang, K. K. Mayo Clin. Proc. 2000, 75, 1027.
11. Michaux, C.; Charlier, C. Mini Rev. Med. Chem. 2004, 4, 603.
12. Patel, C. K.; Rami, C. S.; Panigrahi, B.; Patel, C. N. J. Chem. Pharm. Res. 2010, 2, 73.
13. Sakya, S. M.; Lundy DeMello, K. M.; Minich, M. L.; Rast, B.; Shavnya, A.; Rafka, R.
J.; Koss, D. A.; Cheng, H.; Li, J.; Jaynes, B. H.; Ziegler, C. B.; Mann, D. W.; Petras, C.
F.; Seibel, S. B.; Silvia, A. M.; George, D. M.; Lund, L. A.; Denis, S. St.; Hickman,
A.; Haven, M. L.; Lynch, M. P. Bioorg. Med. Chem. Lett. 2006, 16, 288.
14. Lehman, F. S.; Beglinger, C. Curr. Top. Med. Chem. 2005, 5, 449.
15. Abdel-Aziz, A. A.-M.; ElTahir, K. E. H.; Asiri, Y. A. Eur. J. Med. Chem. 2011, 46,
1648.
16. Chakraborti, A. K.; Garg, S. K.; Kumar, R.; Motiwala, H. F.; Jadhavar, P. S. Curr.
Med. Chem. 2010, 17, 1563.
44. Golbraikh, A.; Tropsha, A. J. Mol. Graph. Model. 2002, 20, 269.