Synthesis, anti-inflammatory activity, and in vitro antitumor effect of a novel class of cyclooxygenase inhibitors: 4-(Aryloyl)phenyl methyl sulfones

Article abstract of DOI:10.1021/jm100398z

Following our previous research on anti-inflammatory drugs (NSAIDs), we report on the design and synthesis of 4-(aryloyl)phenyl methyl sulfones. These substances were characterized for their capacity to inhibit cyclooxygenase (COX-1 and COX-2) isoenzymes. Molecular modeling studies showed that the methylsulfone group of these compounds was inserted deep in the pocket of the human COX-2 binding site, in an orientation that precludes hydrogen bonding with Arg120, Ser353, and Tyr355 through their oxygen atoms. The N-arylindole 33 was the most potent inhibitor of COX-2 and also the most selective (COX-1/COX-2 IC₅₀ ratio was 262). The indole derivative 33 was further tested in vivo for its anti-inflammatory activity in rats. This compound showed greater inhibitory activity than ibuprofen. Other compounds (20, 26, 9, and 30) showed strong activity against carrageenan-induced inflammation. The latter compounds showed a weak capacity to inhibit the proliferation of human cell lines K562, NCI-H460, and HT-29 in vitro.

Full text of DOI:10.1021/jm100398z

6560 J. Med. Chem. 2010, 53, 6560–6571
DOI: 10.1021/jm100398z
Synthesis, Anti-Inflammatory Activity, and in Vitro Antitumor
Effect of a Novel Class of Cyclooxygenase Inhibitors: 4-(Aryloyl)phenyl Methyl Sulfones
†
§
†
†
§
§
ꢀ
Youssef Harrak, Giovanni Casula, Joan Basset, Gloria Rosell, Salvatore Plescia, Demetrio Raffa,
Maria Grazia Cusimano,§ Ramon Pouplana,*,‡ and Maria Dolors Pujol*,†
†
‡
Laboratori de Quımica Farmaceutica (Unitat Associada al CSIC), and Laboratori de Fısico-Quımica, Facultat de Farmacia,
´
ꢀ
´
´
ꢀ
Universitat de Barcelona, Av. Diagonal 643, E-08028 Barcelona, Spain, and §Dipartamento di Chimica e Tecnologie Farmaceutiche,
Facoltaꢀ di Farmacia, Universitaꢀ degli Studi di Palermo, Via Archirafi, 32-90123 Palermo, Italy
Received March 30, 2010
Following our previous research on anti-inflammatory drugs (NSAIDs), we report on the design and
synthesis of 4-(aryloyl)phenyl methyl sulfones. These substances were characterized for their capacity to
inhibit cyclooxygenase (COX-1 and COX-2) isoenzymes. Molecular modeling studies showed that the
methylsulfone group of these compounds was inserted deep in the pocket of the human COX-2 binding
site, in an orientation that precludes hydrogen bonding with Arg120, Ser353, and Tyr355 through their
oxygen atoms. The N-arylindole 33 was the most potent inhibitor of COX-2 and also the most selec-
tive (COX-1/COX-2 IC50 ratio was 262). The indole derivative 33 was further tested in vivo for its
anti-inflammatory activity in rats. This compound showed greater inhibitory activity than ibuprofen.
Other compounds (20, 26, 9, and 30) showed strong activity against carrageenan-induced inflammation.
The latter compounds showed a weak capacity to inhibit the proliferation of human cell lines K562,
NCI-H460, and HT-29 in vitro.
Introduction
liver metastasis. While regular treatment with NSAIDs corre-
lates with a reduced risk of lymphoma and leukemia devel-
opment, long-term use of acetaminophen enhances the devel-
opment of leukemia.10 Of 20 published epidemiologic studies
focusing on the association between NSAIDs and the risk
of cancer in humans, 13 reported a significant reduction of
the disease.10 Although epidemiological studies have demon-
strated that NSAID compounds are effective in chemopre-
vention and may also provide a treatment for several cancers,
more detailed studies are required before their medicinal
application.11 Thus, ibuprofen inhibits cell proliferation in
vitro in human cell lines (HT-29) and also potentiates the
antitumor activity of agents such as 5-fluorouracile.12 Sulindac
and celecoxib inhibit the growth of adenomatous polyps of
patients with familial adenomatous polyposis.13 The devel-
opment of safe and effective pharmaceutical compounds is
complicated. However, recent studies suggest that the long-
term use of selective COX-2 inhibitors is limited because of
cardiovascular thrombotic events related to the aggregatory
properties of these drugs.14 Moreover, COX-2 makes a signif-
icant contribution to the production of inflammatory PGs,
and the inhibition of COX-2 attenuates the expression of inflam-
matory mediators such as TNF-R, iNOS, and IL-1β.15
Nonsteroidal anti-inflammatory drugs (NSAIDsa) are widely
used to alleviate inflammation and pains associated with
many pathological conditions and are often the initial therapy
for common inflammation. NSAIDs inhibit the biosynthesis
of prostaglandins (PGs). In general, biosynthesis involves the
conversion of arachidonic acid to prostaglandin G2 (PGH2), a
reaction catalyzed by the sequential action of cyclooxygenase
(COX). Most NSAIDs inhibit COX-1 and COX-2 isoforms.1-3
COX-1 is responsible for the synthesis of cytoprotective
prostaglandins in the gastrointestinal tract and for the proag-
gregatory thromboxane in blood platelets.4 COX inhibitors
have recently been reported to have a protective effect against
colon cancer and Alzheimer’s disease. Moreover, these com-
pounds continue to be the most commonly used remedies
for rheumatic diseases.5 Several studies have indicated that
the COX inhibitors may prevent colon cancer.6 The molecular
mechanism responsible for the chemopreventive action of
NSAIDs is not understood. However, recent literature has
revealed several mechanisms involved in the inhibition of prolif-
eration or invasive growth, such as apoptosis,7 angiogenesis,8
and activation of protein kinase G.9 NSAIDs are potent
antioxidantsthatexertbothanti-inflammatory andantitumor
activity. In this regard, ibuprofen inhibits tumor growth and
In an attempt to rationalize the selective activity of COX-1/
COX-2 and the cytotoxicity of NSAIDs, here we designed a
series of new compounds from classic arylpropionic acids,
such as ketoprofen and suprofen (Figure 1), which are con-
sidered nonselective for COX-1 and COX-2. The diarylketone
subunit was slightly modified, and the propionic acid group
was changed to methylsulfone, a more lipophilic group with
acidic properties. The modifications were chosen to eval-
uate whether the haptophore moiety of arylpropionic acids
*To whom correspondence should be addressed. For M.D.P.: phone,
þ34-93-4024534; fax, þ34-93-4035941; e-mail, mdpujol@ub.edu. For
R.P.: phone, þ34-93-4024557; fax, þ34-93-40325987; e-mail, rpouplana@
ub.edu.
a Abbreviations: COX-1, cyclooxigenase-1; COX-2; cyclooxigenase-
2; NSAIDs, nonsteroidal anti-inflammatory drugs; PG, prostaglandin;
PGH2, prostaglandin G2; MTT, 3(4,5-dimethylthiazol-2-yl)-2,5-diphe-
nyltetrazolium bromide; SAR, structure-activity relationship; QSAR,
quantitative structure-activity relationship.
r
pubs.acs.org/jmc
Published on Web 08/30/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18 6561
influencedthe selectivity ofCOX isoforms for methylsulfones.
Here we report on the synthesis of a new series of sulfones as a
novel class of NSAIDs and also their anti-inflammatory and
antitumor activity and structure-activity relationships.
We developed a synthetic approach for N-arylamines
30-33 using Pd catalysts for the amination of aryl halides
(Scheme 5). We previously reported the synthesis of the di-
arylamine 30, the N-arylpyrrole 32, and the N-arylindole 33.21
We included these compounds in this study in order to facili-
tate structure-activity relation studies.
Chemical Section
The methods used for the synthesis of arylsulfones are con-
venient because the reagents are readily available and the
protocols straightforward.
Biological Results. Discussion
We evaluated the capacity of these compounds to inhibit
COX-1 and COX-2 in vitro. The potency and selectivity with
respect to two isoforms was determined from IC50values,
expressed as the concentration of compound required to
inhibit 50% of the initial rate of the amount of PGs by enzyme
immunoassay.
Two compounds (9 and 13) were nonselective COX-2 inhib-
itors (selectivity index SI = 1), whereas seven (11, 31, 29, 21,
20, 32, and 12) were moderately selective against this isoform
(SI=2.1, 4.2, 6.7, 9.4, 10, 10.1, 4.2, respectively). In contrast
four compounds (10, 22, 30, and 33) showed very high COX-2
selectivity indexes (36.6, 150, 70.3, and 262, respectively).
Compounds 14 and 23 showed moderate COX-2 selectivity
indexes. The indole derivative 11 had an inhibition profile
similar to 13, whereas the diarylketone 23 showed less inhibi-
tion capacity than the bromo analogue 22.
The general synthetic routes are illustrated in Schemes 1-5.
These methylsulfones were prepared by oxidation of the 2- or
3-(p-(methylthio)benzoyl) heterocycle to the corresponding
sulfone using m-CPBA (Scheme 1). The intermediate diarylk-
etone (3-8 and 16-19) was obtained in a two steps from the
corresponding arylaldehyde and p-bromothioanisole using
butyllithium for the Br/Li interchange,16 followed by oxida-
tion of the carbinol obtained (2a-f) (Scheme 1) or directly by
acylation of the thioanisole 15 withthe corresponding arylacyl
chloride under classical conditions (Scheme 2).17
The arylketones (16-19) were transformed to the corre-
sponding methylsulfones 20-23 by oxidation with m-CPBA
(Scheme 2).18
The 2-arylpyridine 26 was prepared by a coupling reaction
of 2-bromopyridine (24) and the p-methylthiobromobenzene
(1) under classical Ullmann conditions19 followed by oxida-
tion of the methylthio group using m-CPBA (Scheme 3). The
method using m-CPBA was highly effective and simple for the
oxidation to methylsulfones.
The N-(2-fluorophenyl)pyrrole 27 was obtained from
2-fluoroaniline following the Paal-Knorr condensation.20
The racemic methylsulfone 29 was prepared by intramolecu-
lar cyclization of the racemic carbinol 2d in basic media
followed by oxidation of the methylthio intermediate 28 as a
racemate with m-CPBA in the optimized conditions described
above (Scheme 4). Enantiomers of 29 were not separated, and
the biological tests were performed with the racemic mixture.
The compound most selective for COX-2, 33, was a weak
inhibitor of COX-1 (IC50=78.7 μM) and a potent inhibitor of
COX-2 (IC50= 0.3 μM). As a positive control, we used NS398
Scheme 2a
a Reaction conditions: (i) ArCOCl/CH2Cl2; (ii) m-CPBA, 0 °C.
Scheme 3a
Figure 1. Structure of NSAIDs.
a Reaction conditions: (i) Cu/K2CO3; (ii) m-CPBA, 0 °C.
Scheme 1a
a Reaction conditions: (i) BuLi/ArCHO; (ii) MnO2/CH2Cl2, (iii) m-CPBA, 0 °C.

6562 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18
Harrak et al.
Scheme 4a
a Reaction conditions: (i) 1, BuLi, THF, -78 °C; (ii) NaH/DMF; (iii) m-CPBA, 0 °C.
Scheme 5a
Table 1. In Vitro Inhibition of COX-1 and COX-2 by Compounds
9-14, 20-23, 26, 29, and 30-33a
% inhibition
IC50 [μM]
compd
COX-1
COX-2
selectivity index
SI = COX-1/COX-2
9
15.1 ( 0.94
150 ( 11.30
76.4 ( 8.50
183 ( 15.3
34.6 ( 1.65
26.7 ( 2.74
231.2 ( 3.10 23.1 ( 3.00
143.7 ( 4.20 15.2 ( 2.10
150.0 ( 4.20 1 ( 0.15
106.5 ( 4.32 7.7 ( 0.28
89.6 ( 3.22
177.3 ( 4.70 26.5 ( 1.20
140.7 ( 1.9
80 ( 7.50
86 ( 3.50
78.7 ( 3.60
3.2
19.2 ( 1.80
4.1 ( 1.00
36.8 ( 6.60
43.3 ( 5.70
19.7 ( 2.70
9.4 ( 0.36
0.79
36.6
2.1
10
11
12
13
14
20
21
22
23
26
29
30
31
32
33
4.20
1.76
2.8
a Reaction conditions: (i) m-CPBA, 0 °C; (ii) Pd[P(o-tolyl)3]2Cl2,
BINAP, Cs2CO3, NHR1R2.
10
9.40
150
which showed an inhibition of 73.2% for COX-2 and 14.1%
for COX-1 when used at a concentration of 0.2 μM (Table 1).
All the compounds tested inhibited COX. Comparison
results with standard drugs ibuprofen, NS398, celecoxib,
and rofecoxib are listed in Table 1. The N-arylindole 33 elic-
ited maximum inhibition of COX-2, whereas ibuprofen showed
efficacy in inhibiting both COX-1 and COX-2. Also, the
diarylketone 22 exhibited greater activity (1 ( 0.15 μM) than
23 (7.7 ( 0.28 μM), in which the substituent of the aryl was a
fluorine instead of a bromine atom. Thus, 33 and 22 showed
262 and 150 times more inhibitory activity against COX-2
than COX-1, respectively. Moreover, the indole derivative 33
presents inhibition of COX-2 equivalent to known AINEs
such as celecoxib and rofecoxib while its COX-2/COX-1 selec-
tivity was 4 and 8 times greater, respectively.
13.8
68.9
6.70
70.3
4.20
10.10
262
1.3 ( 0.03
2 ( 0.3
18.8 ( 2.80
8.5 ( 1.00
0.3 ( 0.04
1.5
(S)-ibuprofen
2.10
0.19
69
NS398 [0.2 μM] 14.1
celecoxib
rofecoxib
73.2
0.10
6.86
23.5
0.78
30.12
a Results are expressed as the mean (n = 3) of the % inhibition of
PGH2 production by test compounds with respect to control samples. In vitro
COX-2 selectivity index is IC50(COX-1)/IC50(COX-2). ND: not determined.
In order to test the structural requirements, we studied the
isomeric compounds. The inhibitory activity of 3-substituted
furan (10) against COX-2 was greater than the observed for
the 2-substituted furan (9). Also, the anti-inflammatory activ-
ity of the bioisoster 21, which had a 3-substituted thiophene,
was found to be greater than that of 2-substituted thiophene
(20). In the pyrrole ring the 3-substituted compound showed
2-fold more COX-2 inhibition than the corresponding 2-
substituted isomer (comparison of 13 with 12). The piperazine 31
caused 9-fold reduction in COX-2 inhibition compared with
the diarylamine 30. In comparison with the indole (33), the
pyrrole derivative (32) reduced 30 times the COX-2 inhibition.
The methylation of N-indole (compound 14) caused a de-
crease in COX-2 inhibition that was 4-fold that produced
by compound 11. These compounds were then assessed for
anti-inflammatory activity by using the carrageenan rat paw
edema assay.
The characteristic feature of the title compounds is the pres-
ence of a sulfone group at the phenyl ring. Compounds were
tested two times at 70 mg/kg po, and the results were compa-
red withthoseregistered for standard drugssuchasibuprofen.
The thiophene derivative 20 elicited maximum inhibition of
edema (59.7% and 56% at 3 and 4 h postadministration,
respectively) (Table 2). This compound was the most potent
of this series, showing twice more activity than the parent
compound ibuprofen. Moreover, 20 exhibited extended ac-
tion, and 5 h after oral administration (4 h postcarrageenan)
this compound was 2.40 times more active than ibuprofen.
Compounds 20, 26, 33, 9, and 30 were the most potent in
inibiting the in vivo inflammation induced by carrageenan,
compared tothe action of ibuprofen. Incontrast10, 22, and 12
showed less anti-inflammatory activity than ibuprofen in the
rat paw edema assay.
A comparison of data revealed that 20, 26, 33, 9, and 30
showed significantly longer anti-inflammatory activity than
ibuprofen. Ibuprofenshoweda considerable reduction in anti-
inflammatory activity 4 h after carrageenan administration,
while the activity of 20, 26, and 33 practically did not change
between 4 and 5 h after the administration of product. There-
fore, 20, 26, and 33 could be considered for the potential
development of new AINEs with anti-inflammatory activity.
Regarding structure-activity relationships, 20 showed the
strongest anti-inflammatory activity. Replacement of the thio-
phene group by furan, such as in compound 9, decreased the
anti-inflammatory activity of the arylmethylsulfone moiety.
The presence of substituents on the aryl group (12, 13, 22, and
29) decreased the anti-inflammatory activity of 20 compared
to compounds containing a heterocyclic ring (20, 26, and 33).
The thiophene nucleus in 20 showed greater anti-inflamma-
tory activity in vivo than the pyridine moiety in 26.

Article
Our results on biological activity indicate that the intro-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18 6563
Effect of Compounds on Cell Viability. The antiprolifera-
tive activities of compounds 9-13, 20-22, and 29-33 were
evaluated against the following cell lines: K562 (leukemia-
lymphoma), NCI-H460 (human nonsmall lung carcinoma)
and HT29 (human colon cancer). We used the trypan blue
dye exclusion assay and MTT for this purpose. Antitumor
activities were compared with celecoxib (anti-inflammatory)
and purvalanol A (specific inhibitor of cyclin-dependent
kinase activities).
duction of 2-thiophene to the arylsulfones is relevant, as 20
showed more activity than the analogue 9, which has a
furan ring. The nature of the heterocyclic aromatic on the
p-position of the methyl sulfone phenyl had a strong effect
on anti-inflammatory activity. The methylsulfone 20, a
highly potent anti-inflammatory with a diarylketone scaf-
fold, could be considered suitable for a future generation of
COX inhibitors. This compound presented reasonable oral
bioavailability compared with the other compounds of the
same series.
Treatment of K562 cells with 9-13, 20-22, and 29-33 at
10 μM did not affect cell viability, and the effects were
relatively poor (Table 3).
Treatment of NCI-H460 cells with 10 μM pyrrole deriva-
tive 13 resulted in a decrease in cell viability of 54.8 ( 0.70%.
In addition, treatment with 10 μM other compounds had no
cytotoxic effect, while 100 μM 12, 22, 29, 30, 31, and 33 showed
poor cytotoxicity with a mean of growth inhibition of 22.0%,
20.3%, 42.8%, 43.2%, 27.7%, and 27.3%, respectively.
Finally, treatment of HT-29 with 10 μM compounds
9-13, 20-22, and 29-33 did not have a significant effect
on cell viability. The IC50 ranged from 34 (compound 12) to
383 (compound 9) μM (Table 3).
To examine the effects of compounds 9-13, 20-22, and
29-33 on primary culture, HuDe cells were treated with
100 μM test compounds. At this tested concentration none
of the compounds showed a significant effect on cell viability
except 12 (Table 3).
SAR of Arylsulfones. A comparison of these structures
indicated that the presence of the N-(2-fluorophenyl)pyrrole
subunit resulted in greater cytotoxicity than that of other
aromatic rings. The presence of an aroyl group at the C-2
position of the pyrrole (12) produced greater activity against
HT-29 and HuDe cells than the corresponding isomer of
position (13). Compound 13, with the aroyl group at the C-3
position, showed considerable capacity to inhibit H-460 cell
growth. The carbonyl linkage that acts as spacer between the
two aryl rings offers the possibility of adopting a conforma-
tion that is suitable for interaction with the site active. The
3-substituted furan (10) and the 3-substituted thiophene (21)
exhibited slightly more inhibitory activity in HT-cell than
Table 2. Anti-Inflammatory Activity of the Compounds in Vivo (Edema
Induced by Carrageenan)a
3 h
4 h
compd
20
swelling
% inhibition
59.7
swelling
% inhibition
56.0
(a) 17.5 ( 5.6***
(b) 43.4 ( 6.9
(a) 20.1 ( 4.9***
(b) 43.4 ( 6.9
(a) 22.8 ( 4.5***
(b) 43.4 ( 6.9
(a) 28.3 ( 5.6***
(b) 51.9 ( 6.8
(a) 31.7 ( 5.6***
(b) 54.5 ( 6.8
(a) 37.9 ( 4.8**
(b) 51.9 ( 6.8
(a) 40.5 ( 5.6**
(b) 51.9 ( 6.8
(a) 32.2 ( 5.1*
(b) 37.3 ( 3.1
NA
(a) 19.1 ( 5.3***
(b) 43.4 ( 8.4
(a) 21.1 ( 5.5***
(b) 43.4 ( 8.4
(a) 24.1 ( 5.1***
(b) 43.4 ( 8.4
(a) 30.0 ( 4.4**
(b) 48.2 ( 7.7
(a) 25.2 ( 3.6**
(b) 38.9 ( 2.3
(a) 37.5 ( 5.7*
(b) 48.2 ( 7.7
(a) 44.3 ( 4.6
(b) 48.2 ( 7.7
(a) 34.9 ( 5.0
(b) 36.5 ( 4.3
NA
26
33
9
53.7
48.7
45.5
41.9
27.0
22.0
13.7
51.4
46.0
37.8
32.6
22.2
8.1
30
10
22
12
4.4
13
29
NA
NA
(a) 34.4 ( 6.9**
(b) 45.1 ( 5.0
ibuprofen (a) 27.1 ( 3.6***
(b) 38.9 ( 6.0
30.3
23.9
a (a) Swelling for a dose of 70 mg/kg. (b) Swelling control. NA no
activity attributable to low solubility in water. Values significantly dif-
fer from controls as indicated: /, P < 0.05; //, P < 0.01; ///, P < 0.001;
#, does not differ significantly according to unpaired one-tailed
Student’s t-test. The data represent the mean ( SEM of six animals
(p < 0.05). The rat paw edema occurred at 3 and 4 h after carrageenan
administration.
Table 3. Inhibition of K562, NCI-H460, and HT-29 Cancer Cells and HuDe Cell Proliferationa
% inhibition [10 μM]
% inhibition [100 μM]
HuDe cell viability
IC50 (μM)
HT-29 cell viability
compd
K562 cell viability
NCI-H460 cell viability
9
8.3 ( 0.50
8.3 ( 1.00
2.1 ( 0.20
2.1 ( 0.02
12.5 ( 1.00
1.0 ( 0.10
14.6 ( 2.00
6.2 ( 0.10
13.5 ( 3.90
13.5 ( 0.60
17.7 ( 2.00
1.0 ( 0.05
16.6 ( 0.90
38.8 ( 2.00
75.1 ( 2.30
3.0 ( 0.70
6.0 ( 1.00
7.0 ( 0.01
13.3 ( 0.50
54.8 ( 0.70
3.0 ( 0.40
2.0 ( 0.09
12.7 ( 1.90
6.6 ( 0.08
1.8 ( 0.01
3.1 ( 0.20
3.0 ( 0.30
12.2 ( 1.00
9.75 ( 1.10
3.98 ( 0.10
30.7 ( 6.40
36.2 ( 2.00
30.8 ( 1.70
45.0 ( 2.10
13.4 ( 0.90
25.7 ( 2.80
31.2 ( 0.40
29.4 ( 0.05
12.3 ( 1.00
37.4 ( 1.90
36.1 ( 0.80
28.9 ( 2.00
34.3 ( 0.70
92.0 ( 1.60
88.6 ( 1.90
383.0 ( 6.40
104.0 ( 1.60
70.4 ( 0.50
34.1 ( 2.10
96.1 ( 1.00
122.0 ( 2.10
75.7 ( 9.00
100.0 ( 8.40
100.0 ( 1.20
130.0 ( 5.20
65.8 ( 4.70
146.4 ( 7.90
85.0 ( 5.70
45.5 ( 0.90
.100
10
11
12
13
20
21
22
29
30
31
32
33
celecoxib
purvalanol A
a Cancer cell viability on K-562 and H-460 after treatment with 10 μM 9-13, 20-22, and 29-33. Cell viability is on human derm fibroblasts (HuDe)
after treatment with 100 μM. Values represent [control (DMSO) - treated] /control (DMSO) and are expressed as average values of two separate
experiments repeated in duplicate. HT-29 cell viability is for after treatment with a range of concentrations of compounds 9-13, 20-22, and 29-33. IC50
values are expressed as average values of two separate experiments repeated four times. Purvalanol: Sigma-Aldrich, ref P4484. Celecoxib: LC Labo-
ratories, ref 1502.

6564 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18
Harrak et al.
Table 4. MD (debye), Experimental Activity (ΔGbind), Free Energy of Solvation (ΔGsolv), and Calculated Binding Affinity (ΔGbind(MM-GBSA)) of
Compounds 9-14, 20-23, 26, and 29-33
MD (D)
ΔGbind (kcal/mol)
ΔGsolv (kcal/mol)
IC50(COX-2) in vitro ( μM)
compd
ΔGbind(MM-GBSA) (kcal/mol)
8.333
7.226
6.348
8.028
8.571
6.084
8.398
7.480
6.121
8.344
7.933
7.979
7.350
6.234
5.657
5.084
-6.41
-7.32
-6.02
-5.93
-6.39
-6.83
-6.30
-6.55
-8.15
-6.95
-8.00
-6.22
-7.74
-6.42
-6.89
-8.86
-10.27
-10.71
-12.42
-10.32
-10.73
-15.96
-10.48
-10,76
-9.57
-9.48
-9.45
-10.99
-11.55
-10.38
-9.47
19.2
4.1
9
-26.1 ( 2.0
-24.87 ( 2.0
-30.55 ( 3.5
-38.96 ( 3.1
-44.86 ( 2.5
-35.13 ( 2.4
-26.36 ( 2.7
-27.95 ( 2.3
-38.25 ( 1.5
-34.35 ( 2.1
-32.94 ( 1.8
-28.12 ( 1.7
-34.98 ( 1.6
-31.86 ( 1.9
-30.74 ( 1.9
-37.65 ( 1.4
10
11
12
13
14
20
21
22
23
26
29
30
31
32
33
36.8
43.3
19.7
9.4
23.1
15.2
1.0
7.7
1.3
26.5
2.0
18.8
8.5
-9.39
0.3
that shown by corresponding 2-substituted isomers. The
indole derivative 33 showed greater activity than the pyrrole
analogue (32). The piperazine-arylsulfone (31) exhibited
modest growth inhibition of HT-29 cells but was superior
to the effect of the diarylamine (30). In contrast 22 and 29
which had distinct structural characteristics showed similar
and modest cytotoxicity. The cytotoxic activity of the com-
pounds series at the concentrations tested was not sufficient
to inhibit cancer cell proliferation; however, in a time-
dependent manner, these compounds may concentrate in
cancer tissue and exert cumulative effects. Clinical studies
have indicated that a dose of 400 mg b.i.d. of celecoxib
reduces polyp growth while this effect was insignificant at
100 mg b.i.d. and rofecoxib does not show antitumor activity
even at relatively high doses. Molecular modeling studies
may explain these effects and also contribute to the design of
novel compounds that provide either a more specific COX
inhibitor or potent inhibitor of cancer cell proliferation.
QSAR Analysis. The optimized geometries for the 16 methyl
sulfone derivatives were obtained at the B3LYP/6-31G(d)
level, and the relative free energies of hydration were calcu-
lated using the MST B3LYP/6-31G(d) version of the PCM
continuum model.
Information), with the exception of 32 and 10. However, the
HOMO was delocalized over aromatic rings on the less active
compounds (12, 29). The NHOMO was located at the thio-
phene or pyrrole ring, and the NLUMO was delocalized over
the whole structure on the compounds that showed the most
effective inhibition (12, 21) (Figure 2C in Supporting Infor-
mation) of human cancer (HT-29) cell proliferation in vitro.
Molecular Modeling Studies. Visual inspection of the com-
plex of methyl sulfone derivatives and COX-2 (Figure 3)
revealed the same binding orientation as that observed for
the carboxylic group of flurbiprofen in COX-2 (3PGH). This
orientation involved hydrogen bonding interactions of the
oxygen atoms of the sulfone group at the mouth of the bind-
ing channel with Arg120 and Tyr355. Furthermore, Val349,
Phe518, and Leu352 were within van der Waals contact.
Additional hydrogen bonds were observed between the
oxygen atom of the carbonyl group and the hydroxyl groups
of Ser530.
These studies demonstrate that the sulfone group of these
nonsteroidal anti-inflammatory drugs bind to COX enzyme
in an orientation that precludes the formation of hydrogen
bonding with Arg120, Ser353, and Tyr355 through their
oxygen atoms. Further relevant hydrophobic interactions
were observed in a top cavity formed by Phe518, Leu384,
Tyr385, and Trp387. Hydrophobic residues Leu352, Ala527,
and Val523 surrounded the phenyl ring.
Physicochemical parameters were examined in order to
find a possible explanation for the observations described
above and to provide a better understanding of the relationship
between structure and anti-inflammatory activity. The dipole
moment and the energies of the NHOMO and NLUMO orbitals
in aqueous solution for the 16 compounds were studied.
Also the CH3 of the methylsulfone group was bound to an
adjacent pocket formed by Met113, Val116, Val349, and
Leu531. Finally, the oxygen atom of the carbonyl group
formed hydrogen bonds with Ser530.
The dipole moment plays an interesting role in aligning
optimally with the receptor. The orientation and magnitude
of the dipole moment are crucial for the COX-2 inhibition
activity22 (9 and 10) (Figure 2A in Supporting Information).
A statistically significant correlation (n=11, r2=0.808) was
observed between COX-2 inhibition (ΔGbin =-RT ln(1/IC50)
and dipole moment. However, 5 compounds (11, 14, 26, 31,
and 32, outlying points) showed very large free energy of
solvation or the HOMO delocalized over aromatic rings. For
these compounds, the magnitude of the dipole moment
established a worse relation with capacity to inhibit COX-2
than the same potent compounds of this series (Table 4).
The electron density maps of the HOMO and NHOMO
orbitals appear to be centered in the phenyl, thiophene, or
N-indole group, exhibiting a π-character, on the most active
compounds (33, 22, 30, and 26) (Figure 2B in Supporting
Compounds 33, 32, and 26, which showed potent inhibi-
tory activity of COX-2 in vitro and anti-inflammatory activ-
ity in vivo, interacted with COX-2 via one of the oxygen atoms
of the sulfone group with Arg120 and Tyr355 (Figure 4).
An additional hydrogen bonding interaction was observed
between one of the oxygen atoms of its sulfone group and
Ser353. Further multiple hydrophobic contacts involving the
Leu352, Ala527, and Val523 were observed. Also, the indole
group was in close van der Waals contacts with Leu384 and
Trp387 only for 33. However, notable differences were
detected in the relative arrangement of the indole ring
compared to the position occupied by the corresponding
moiety of pyrrole or pyridine ring. This arrangement thus led
to a more efficient π-π stacking interaction between the
indole ring of 33 and the phenyl ring of Phe518.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18 6565
Figure 5. Superposition of the triphenyl methylsulfone derivatives
(13 (green) and 31 (light green) compounds) in the binding site of
mouse COX-2, highlighting selected residues that form the main
interactions of the inhibitors. Most hydrogen atoms are omitted for
the sake of clarity.
Figure 3. Superposition representation of the thiophene and furane
methylsulfone derivatives (9 (light blue), 10 (green), 20 (blue), and
21 (brown)) in the binding site of mouse COX-2, highlighting
selected residues that form the main interactions of the inhibitors.
Most hydrogen atoms are omitted for the sake of clarity.
The orientation and magnitude of the dipole moment on
these compounds are very crucial for their best docked posi-
tion in the COX-2 binding pocket in such a way that
the π-aromatic contacts involving Phe518 and Trp387 are
revealed. Analysis of the MM-GBSA computations of the
binding free energy and dipole moments showed a close
relation (n =13, r2 =0,746).
Compounds 13 and 31, which are active in the antitumor
bioassay, and shown the same anti-inflammatory activity,
interacted with COX-2 in a distorted mode compared to the
other compounds. This interaction occurred via the binding
of one hydrogen atom with Arg120 through its sulfone group
(Figure 5). Additional hydrophobic contacts with the methyl-
sulfone phenyl ring involving Ile345, Val359, Ala527, and
Leu531 were detected.
The main hydrophobic interactions were found in a top
cavity formed by Phe518, Leu384, Tyr385, and Trp387.
Analysis of the complexes of compounds 12 and 29 (less
active) in their docked position with COX-2 (Figure 6 in
Supporting Information) revealed hydrogen bonds of the
former involving the oxygen atoms of the sulfone group with
Arg120 and with Ser353.
Also, the benzoxazine or pyrrole groups were in close
van der Waals contact with Leu352, Phe518, and Ala527
while the fluorophenyl ring showed an inversed binding
orientation compared to 29 and displacement of Tyr385.
Compounds 22, 23, and 30, which showed the same anti-
inflammatory activity, interacted with COX-2 via one of the
oxygen atoms of the sulfone group with Ser353 and Tyr355
at the mouth of the channel, and the oxygen atom of the
carbonyl/amine moiety formed hydrogen bonds with Ser530
(Figure 7 in Supporting Information). Additional hydro-
phobic contacts involving Ile345, Val359, Ala527, and Leu531
were observed with the methylsulfonephenyl ring. However,
in 22 and 23 the methylsulfone group was slightly rotated.
The main hydrophobic interactions were observed in a top
cavity formed by Leu384, Tyr385, and Trp387. Finally, the
Figure 4. Superposition of the N-arylpyrrole and N-arylindole
methylsulfone derivatives (32 (violet) and 33 (green)) in the binding
site of mouse COX-2, highlighting selected residues that form the
main interactions of the inhibitors. Most hydrogen atoms are
omitted for the sake of clarity.
Thiophene and furan derivates (9, 10, 20, and 21) formed
hydrogen bonds with COX-2 via one of the oxygen atoms of
the sulfone group with Arg120 and Ser353 (except 21, which
formed a hydrogen bond with Tyr355 (Figure 3). A few
hydrophobic contacts were observed close to Ile345, Val349,
Leu352, and Val523 for these compounds.

6566 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18
Harrak et al.
Figure 9. (A) Correlation (n = 13, r2 = 0.69) between COX-2 experimental inhibition activity (ΔGbind) and calculated ΔGbind(GBSA) and
(B) correlation (n = 10, r2 = 0.808) between experimental inhibitory activity of COX-2 (ΔGbind) and dipole moment (DM).
fluoro-/bromophenyl rings were roughly stacked onto the
aromatic ring of Phe518.
However, the best docked position of 14 and 11 with COX-2
(Figure 8 in Supporting Information) showed the oxygen atoms
of the sulfone moiety hydrogen-bonded to Tyr355, Arg120,
and Ser353 at the mouth of the channel, and the oxygen atom
of the carbonyl formed hydrogen bonds with Ser530. Also,
the indole group was in close van der Waals contact with
Leu384, Trp387, and Phe518; however, 11 showed notable
differences in the relative arrangement of the indole ring
through the methyl group compared to 14.
The calculation of binding free energy, ΔGbind(MM-
GBSA), between protein and ligands was evaluated using GBSA.
Statistically significant correlations (n = 13, r2 = 0.69) were
observed between experimental ΔGbin and calculated
ΔGbind(GBSA) (Figure 9A).
The orientation and magnitude of the dipole moment on
these compounds are highly relevant for their the best docked
position in the COX-2 binding pocket, as it reveals π-aromatic
contacts involving Phe518 and Trp387. Statistically significant
correlations (n=10, r2=0.808) were observed between COX-2
Figure 10. Compound 33 found in its respective complex with
COX-2. Relevant active site residues for one representative protein
inhibition activity and dipole moment (Figure 9B). The main
are shown. A solid solvent-accessible surface envelops the side
hydrophobic interactions observed in a top cavity formed by
Leu384, Tyr385, Trp38, and Phe518 could explain the notable
chains of the labeled residues (except Phe518, Trp387, and Ser530)
to delineate the active site cavity. Water molecules and the hydrogen
differences in the relative anti-inflammatory activity of these
atom have been omitted for clarity.
compounds (Figure 10).23
way for further research involving structures with a low dipolar
moment and minimum free energy of solvation. Our results
indicate that a simple methylsulfone with an aryl group in-
volved in the affinity and in the activity showed interesting anti-
inflammatory activity by inhibiting COX and weak cytotoxic.
This study may suggest that these compounds possess the
correct electronic density and the dipole moment directionality
to an important COX-2 receptor affinity.
In summary, here we have described a straightforward and
efficient synthesis of new anti-inflammatory compounds
containing an arylcarbonyl or an aryl group at the para
position of the pharmacophoric methylsulfonephenyl group.
The compounds synthesized showed anti-inflammatory and
cytotoxic activities of diverse intensity. Compound 20 had
the strongest in vivo anti-inflammatory activity which
exceeds (2- and 2.4-fold at 3 and 4 h after the administration
of carrageenan, respectively) that of the parent reference,
ibuprofen, followed by 26, 33, 9, and 30 which showed more
activity than ibuprofen 3 and 4 h after administration. Our
findings may contribute to the future construction of a
pharmacophoric model of COX-2 inhibitors and pave the
Experimental Section
(A) Chemical. General. Melting points were obtained on an
MFB-595010M Gallenkamp apparatus in open capillary tubes
and are uncorrected. IR spectra were obtained using a FTIR

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18 6567
Perkin-Elmer 1600 infrared spectrophotometer. Only noteworthy
IR absorptions are listed (cm-1). 1H and 13C NMR spec-
tra were recorded on a Varian Gemini-200 (200 and 50.3 MHz,
respectively) or Varian Gemini-300 (300 and 75.5 MHz) instru-
ment using CDCl3 as solvent with tetramethylsilane as internal
standard or (CD3)2CO. Other 1H NMR spectra and heterocor-
relation 1H-13C (HMQC and HMBC) experiments were re-
corded on a Varian VXR-500 (500 MHz). Mass spectra were
recorded on a Helwett-Packard 5988-A. Column chromatog-
raphy was performed with silica gel (E. Merck, 70-230 mesh).
Reactions were monitored by TLC using 0.25 mm silica gel
F-254 (E. Merck). Microanalysis was determined on a Carlo
Erba 1106 analyzer. All reagents were of commercial quality or
were purified before use. Organic solvents were of analytical
grade or were purified by standard procedures. Commercial
products were obtained from Sigma-Aldrich. Elemental anal-
ysis was used to ascertain purity of >95% for all compounds of
this work for which biological activities were determined.
General Method for the Preparation of the Diarylalcohols and
Oxidation to Ketones. (A) To a solution of the bromothioanisole
(450 mg, 2.22 mmol) in THF (4 mL) was added a solution of
n-BuLi in hexane (1.4 mL, 2.22 mmol) at -78 °C under an argon
atmosphere. The mixture was stirred at the same temperature
for 1 h. Then a solution of the corresponding aldehyde or ketone
(2.50 mmol) in freshly distilled THF (2 mL) was added, and the
mixture was allowed to warm slowly to 20 °C. At this time a
solution of saturated NH4Cl was added (3 mL), and the mixture
was stirred for 20 min. Finally it was extracted with ether (3 ꢀ
20 mL) and the aqueous phase was acidified with 2 N HCl and
extracted with CH2Cl2 (3 ꢀ 20 mL). The organic layers were
dried (Na2SO4), filtered, and concentrated under vacuum, and
the crude product was purified by silica gel column chromatog-
raphy. Using a mixture of hexane/ethyl acetate (5:95) as eluent,
the desired alcohol was obtained.
obtained as a white solid (406 mg, 1.87 mmol) in a 91% yield.
Mp: 81-83 °C (hexane/ethyl acetate). IR (KBr), ν (cm-1): 1633
(CO); 1296 (Ar-O); 1091 (C-S). 1H NMR (CDCl3, 200 MHz) δ
(ppm): 2.54 (s, 3H, CH3-); 6.59 (m, 1H, C-4H), 7.23 (m, 1H,
C-3H); 7.30 (d, J=8.4 Hz, 2H, C-30H and C-50H); 7.69 (m, 1H,
C-50H); 7.94 (d, J = 8.4 Hz, 2H, C-20H and C-60H). 13C NMR
(CDCl3, 50.3 MHz) δ (ppm): 14.8 (CH3, CH3-); 112.1 (CH, C-4);
119.9 (CH, C-3); 124.8 (CH, C-30 and C-50); 129.8 (CH, C-20and
C-60); 133.2 (C, C-10); 145.4 (C, C-2); 146.8 (CH, C-5); 152.3 (C,
C-40); 182.4 (C, CO).
General Method for the Preparation of Diarylketones. A
stirred solution of thioanisole (127 mg, 1.02 mmol) in dry
dichoromethane (15 mL) was cooled to 0 °C under argon
atmosphere. The solution was stirred vigorously, and a solution
of titanium tetrachloride (0.15 mL, 1.36 mmol) was added
slowly. A solution of the corresponding carboxylic chloride
(0.80 mmol) in dichloromethane (2 mL) was added to the
mixture. The reaction mixture was stirred for 10 h at room
temperature. The crude of reaction was poured into 40 mL of
ice-water, and the aqueous layer was separated. The organic
layer was washed with saturated NaHCO3, dried over Na2SO4,
and evaporated to afford a residue which was purified by
column chromatography using 8:2 hexane/ethyl acetate.
2-(p-Methylthiobenzoyl)thiophene (16). The title compound
was obtained as a white solid from 2-thiophene carboxylic chlo-
ride and thioanisole following the above-described conditions in
50% yield after purification by column chromatography. Mp:
64-66 °C (hexane/ethyl acetate). IR (KBr), ν (cm-1): 1622
(CdO); 1088 (S-CH3). 1H NMR (CDCl3, 200 MHz) δ (ppm):
2.54 (s, 3H, CH3-S); 7.16 (dd, J1=5.2, J2=3.6 Hz, 1H, C-4H);
7.31 (d, J=8.8 Hz, 2H, C-30H y C-50H); 7.64 (dd, J1=3.8, J2=
1 Hz, 1H, C-3H); 7.71 (dd, J1=5.2, J2 =1 Hz, 1H, C-5H); 7.82
(d, J=8,8, 2H, C-20H y C-60H). 13C NMR (CDCl3, 50.3 MHz)
δ (ppm): 14.8 (CH3, CH3-); 124.8 (CH, C-30 y C-50); 127.7
(CH, C-4); 129.6 (CH, C-20 and C-60); 133.7 (CH, C-3); 133.9
(C, C-10); 134.1 (CH, C-5); 143.4 (C, C-2); 144.9 (C, C-40); 186.9
(C, CO).
General Procedure for the Oxidation of Methylthio Derivatives
to Sulfones. To a solution of the methylthio derivative (100 mg,
0.34 mmol) in 25 mL of dry dichloromethane cooled at 0 °C,
m-CPBA (129 mg, 0.75 mmol) was slowly added. The reaction
mixture was stirred at room temperature for 4 h. After the
mixture was treated with a solution of 2 N NaOH (3 ꢀ 25 mL),
the organic layer was separated and dried over anhydrous sodium
sulfate, filtered off, and the solvent was removed by evaporation.
The residue was then purified by silica gel column chromatog-
raphy, eluting with a mixture 7:3 hexane/ethyl acetate.
2-(p-Methylsulfonylbenzoyl)furan (9). From the correspond-
ing methylthio derivative (218.27 mg, 1 mmol) and following the
general procedure described above, the title compound was
obtained (207.1 mg, 0.95 mmol) as a white powder in 95% yield.
Mp: 96-98 °C (hexane/ethyl acetate). IR (KBr), ν (cm-1): 1642
(CO); 1434 (SO2); 1278 (Ar-S); 1099 (C-O). 1H NMR (CDCl3,
200 MHz) δ (ppm): 3.11 (s, 3H, CH3-S); 6.62 (m, 1H, C-4H);
7.31 (m, 1H, C-3H); 7.79 (s,1H, C-5H); 8.13 (d, J=5.6 Hz, 4H,
C-20H, C-30H, C-50H and C-60H). 13C NMR (CDCl3, 50.3 MHz)
δ (ppm): 44.4 (CH3, CH3-); 112.7 (CH, C-4); 121.4 (CH, C-3);
127.4 (CH, C-30, and C-50); 130.0 (CH, C-20, and C-60); 141.5 (C,
C-10); 143.4 (C, C-2); 147.8 (CH, C-5); 151.7 (C, C-40); 180.6 (C,
CO). Anal Calcd for C12H10NO4S: C, 57.59%; H, 4.03%.
Found: C, 57.21%; H, 4.42%.
(B) The intermediate alcohol (1 mmol) was rapidly treated
with MnO2 (3 mmol) in dichloromethane (15 mL). The reaction
mixture was stirred for 12 h at room temperature. The crude of
reaction was filtered, washed with dichloromethane, and the
solvent was removed in rotator evaporator. The obtained
residue was purified by silica gel column chromatography using
8:2 hexane/ethyl acetate mixture as eluent.
2-( p-Methylthiophenylhydroxymethyl)furan (2a). By use of
the above-described procedure, the furan-2-carbaldehyde was
converted to the title alcohol as oil in 92% yield. The crude of
reaction obtained was rapidly oxidized with MnO2 because of
high instability.
N-(2-Fluorophenyl)-2-(p-methylthiophenylhydroxymethyl)pyrrole
(2d). Starting from p-bromothioanisole (450 mg, 2.22 mmol)
and N-(2-fluorophenyl)pyrrole-2-carbaldehyde (23) (350 mg,
1.85 mmol), following the procedure described above, and using
1.6 M BuLi in THF (1.38 mL, 2.22 mmol) at -78 °C, the corre-
sponding diarylmethanol was obtained. The carbynol was ex-
tracted with ether (3 ꢀ 25 mL), dried over Na2SO4, filtered off,
and the solvent was removed under vacuum. The intermediate
carbynol was obtained with sufficient purity to go to the new
step of oxidation to the ketone using MnO2. The title compound
was obtained as a colorless oil after purification by column
chromatography, eluting with a mixture of hexane/ethyl acetate
80:20 (520 mg, 1.66 mmol) in a 90% yield. IR (KBr), δ (cm-1):
1
3220 (OH); 1180 (C-O); 1091 (C-S). H NMR (CDCl3, 200
MHz) δ (ppm): 2.47 (s, 3H, CH3-); 5.60 (d, J=4 Hz, 1H, CH-O),
6.08 (m, 1H, C-4H); 6.28 (m, 1H, C-3H); 6.78 (m, 1H, C-5H);
7.22 (m, 8H, Ar). The crude of reaction was rapidly oxidized
with MnO2 because of high instability.
2-(p-Methylsulfonylbenzoyl)thiophene (20). From the corre-
sponding methylthio derivative (120 mg, 0.51 mmol) and follow-
ing the general procedure described above, the title compound
was obtained (100 mg, 0.37 mmol) as a white powder in 73%
yield. Mp: 136-138 °C (hexane/ethyl acetate). IR (KBr),
ν (cm-1): 1636 (CO); 1413 (SO2); 1286 (Ar-S); 1154 (SO2). 1H
NMR (CDCl3, 200 MHz) δ (ppm): 3.13 (s, 3H, CH3-S); 7.20
(dd, J1 = 4.8, J2 = 3.6 Hz, 1H, C-4H); 7.62 (dd, J1 = 3.6, J2 =
1 Hz, 1H, C-3H); 7.81 (dd, J1=4.8, J2 =1 Hz, 1H, C-5H); 8.05
2-(p-Methylthiobenzoyl)furan (3). Starting from the p-bromo-
thioanisole (500 mg, 2.46 mmol) and the furfural (196.8 mg, 2.05
mmol), following the procedure described before, and using
BuLi in THF at -78 °C, the corresponding diarylmethanol was
obtained, which was of sufficient purity to go to the new step of
oxidation to the ketone using MnO2. The title compound was

6568 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18
Harrak et al.
(dd, J1=8.4, J2=1 Hz, 4H, C-20H, C-30H, C-50H y C-60H). 13C
NMR (CDCl3, 50.3 MHz) δ (ppm): 44.3 (CH3, CH3-); 127.5
(CH, C-30 and C-50); 128.3 (CH, C-4); 129.7 (CH, C-20 and C-60);
135.4 and 135.5 (CH, C-3, and C-5); 142.5 and 142.6 (C, C-2 and
C-10); 143.2 (C, C-40); 186.5 (C, CO). Anal Calcd for C12H10-
O3S2: C, 54.11%; H, 3.78%. Found: C, 53.86%; H, 3.64%.
General Procedure for Obtaining Diaryl Compounds. A sus-
pension of bromoaryl (2.46 mmol), bromopyridine (387 mg,
2.46 mmol), Cu (134 mg, 3.69 mmol), and anhydrous K2CO3
(509 mg, 3.69 mmol) without solvent in a Schlenk tube was
heated at 300 °C for 8 h. At this time, the reaction mixture was
poured into 25 mL of water and was extracted with ether (3 ꢀ
30 mL). The organic phase was dried and evaporated at vacuum
to yield colorless oil. The residue was purified by silica gel
column chromatography, eluting with 8:2 hexane/ethyl acetate.
2-(p-Methylthiophenyl)pyridine (25). Starting from (500 mg,
2.46 mmol) of 4-bromothianisole and following the general
procedure described before, the title compound was obtained
as a white solid (50 mg, 0.25 mmol) in 10% yield. Mp: 55-57 °C
(hexane/ethyl acetate). IR (KBr), ν (cm-1): 1584 (CdC); 1463
(CdC), 843 C-S). 1H NMR (CDCl3, 200 MHz) δ (ppm): 2.53 (s,
3H, CH3-); 7.20 (m, 1H, H-Ar); 7.40 (d, J=8.8 Hz, 2H, C-20H
and C-60H); 7.68 (m, 2H, H-pyridine); 7.91 (d, J = 8.8 Hz, 2H,
C-30H and C-50H); 8.65 (m, 1H, C-6H pyridine).
2-(p-Methylsulfonylphenyl)pyridine (26). Starting from (50 mg,
0.25 mmol) of the corresponding methylthio compound and
following the general procedure described before, the title com-
pound was obtained as a white solid (35 mg, 0.15 mmol) in 60%
yield. Mp: 86-88 °C (hexane/ethyl acetate). IR (KBr), ν (cm-1):
1300 (SO2); 1152 (S-O). 1H NMR (CDCl3, 200 MHz) δ (ppm):
3.10 (s, 3H, CH3-); 7.28 (m, 1H, C-5H), 7.81 (m, 2H, C-3H, and
C-4H); 8.02 (d, J=8.8, 2H, C-20H, and C-60H); 8.07 (d, J=8.8,
2H, C-30H, and C-50H); 8.73 (d, J=5.5, C-6H). 13CNMR(CDCl3,
50.3 MHz) δ (ppm): 44.6 (CH3, CH3-); 121.1 (CH, C-3); 123.4
(CH, C-5); 127.7 (CH, C-20 and C-60); 127.8 (CH, C-30and C-50);
137.1 (CH, C-4); 140.4 (C, C-10); 144.4 (C, C-40); 149.9 (CH, C-6);
155.1 (C, C-2). Anal Calcd for C12H11NO2S: C, 61.78%; H,
4.75%; N, 6.00%. Found: C, 61.63%; H, 4.65%; N, 6.17%.
Preparation of 1,4-Benzoxazines 4-(p-Methylthiophenyl)-4H-
pyrrolo[2,1-c][1,4]benzoxazine (28). To a solution of N-(2-fluoro-
phenyl)-2-((p-methylthiophenyl)hydroxymethyl)pyrrole (2d)
(500 mg, 1.59 mmol) in dry DMF (5 mL) was added (127 mg,
3.18 mmol) sodium hydride (a 60% oil dispersion). The reaction
mixture was stirred for 24 h at 130 °C. Then the mixture was
allowed to stir to room temperature for 30 min, and 20 mL of ether
was added. The organic phase was separated and washed with
water three times, dried, filtered, and the solvent was removed by
rotary evaporation. The residue was then purified by silica gel
column chromatography by elution with 8:2 hexane/ethyl acetate
mixtures to yield the title compound 28 (233 mg, 0.795 mmol) in
50% yield as a white solid. Mp: 122-123 °C (hexane/ethyl acetate).
IR (KBr), ν (cm-1): 1504 (CdC); 1216 (Ar-O); 1088 (C-O). 1H
NMR (CDCl3, 200 MHz) δ (ppm): 2.5 (s, 3H, CH3-S); 5.73 (m,
1H, C-3H); 6.06 (s, 1H, C-4H); 6.31 (t, 1H, J=3.8 Hz, C-2H); 7.06
(m, 3H, Ar); 7.21 (dd, J1=3 Hz, J2=1.5 Hz, 1H, C-1H); 7.27 (d,
J = 8.4 Hz, 2H, Ar); 7.40 (m, 3H, Ar). 13C NMR (CDCl3, 50.3
MHz) δ (ppm) (/ indicates interchangeable): 15.7 (CH3, CH3-S);
76.0 (CH, C-4); 106.3 (CH, C-3); 110.5 (CH, C-2); 114.7 (CH, C-1);
114.9 (CH, C-9); 118.3 (CH, C-6); 122.3* (CH, C-7); 124.9* (CH,
C-8); 126.2 (CH, C-20, and C-60); 126.6 (C, C-3a); 127.3 (C, C-10);
128.5 (CH, C-30, and C-50); 134.5 (C, C-9a); 139.4 (C, C-40); 145.9
(C, C-5a).
1H, C-2H); 7.10 (m, 3H, Ar); 7.23 (dd, J1=3 Hz, J2=1 Hz, 1H,
C-1H); 7.40 (m, 1H, C-9H); 7.70 (d, J=8.2 Hz, 2H, C-20H, and
C-60H); 7.98 (d, J=8,2 Hz, 2H, C-30H, and C-50H). 13C NMR
(CDCl3, 50.3 MHz) δ (ppm) (/ indicates interchangeable): 44.5
(CH3, CH3-); 75.2 (CH, C-4); 106.8 (CH, C-3); 110.7 (CH, C-2);
114.8 (CH, C-1); 115.4 (CH, C-6); 118.3 (CH, C-9); 122.8* (CH,
C-7); 125.2* (CH, C-8); 125.8 (C, C-3a); 126.5 (C, C-10); 127.6
(CH, C-20, and C-60); 128.5 (CH, C-30, and C-50); 140.6 (C,
C-9a); 146.1 (C, C-40); 145.2 (C, C-5a). Anal Calcd for C18H15-
NO3S: C, 66.44%; H, 4.65%; N, 4.30%. Found: C, 66.59%; H,
4.77%; N, 4.56%.
Preparation of N-Arylanilines 30-33. The preparation of
arylanilines 30-33 was described previously by us.21
(B) Anti-Inflammatory Activity. In Vitro Assay. The ability of
the test compound to inhibit the conversion of arachidonic acid
to prostaglandin H2 (PGH2) was determined using a COX-1/
COX-2 inhibitor screening assay kit (catalog no. 560101; Cayman
Chemical, Ann Arbor, MI).
Briefly, cyclooxygenase catalyzes the first step in the bio-
synthesis of arachidonic acid to PGH2. The COX (ovine) inhi-
bitor screening assay directly measures PGF2R produced by
SnCl2 reduction, by enzyme immunoassay (ACE competitive
EIA). This assay is based on the competition between PGs and
PG-acetylcholinesterase (AChE) conjugate (PG tracer) for a
limited amount of PG monoclonal antibody. Because the con-
centration of PGs varies, the amount of PG tracer that is able to
bind to the PGs monoclonal antibody will be inversely propor-
tional to the concentration of PG in the well. This antibody-PG
complex binds to goat polyclonal anti-mouse IgG that has been
previously attached to the well. The plate was washed to remove
any unbound reagents, and Ellman’s reagent (which contains
the substrate to acetylcholinesterase) is added to the well. The
product of this enzymatic reaction absorbs at 405 nm. The inten-
sity of this color, determined spectrophotometrically, is pro-
portional to the amount of PG trace bound to the well, which
is inversely proportional to the amount of free PG present in the
well during the incubation: absorbance µ [bound PG tracer] µ
1/[PG].
In Vivo Test. The carrageenin-induced rat paw edema assay
was carried out using a modified Winter’s method as a prelimi-
nary screening test. The rats (in groups of six animals weighing
160-200 g, young adult male Sprague-Dawley) were starved
for 24 h before the test compound (70 mg/kg po) was adminis-
tered. The drugs were given orally 1 h before carrageenan. Rat
paw edema was induced by subplantar injection of 0.1 mL of
a 1% solution of carrageenan in sterile pyrogen-free 0.9%
NaCl solution, and the volume of paw was measured by water
displacement in a plethysmometer S-5128, Ugo Basile. Three
and four hours after the injection of carrageenan, the volume of
the paw was again measured. All statistical analyses of data were
processed by computer. Any treatment mean value significantly
less than the control mean was indicative of significant anti-
inflammatory activity. Rat paw edema volume of treated ani-
mals was compared to that animals receiving ibuprofen for
comparing the relative potency. No toxic symptoms were ob-
served after oral administration of 70 mg/kg in the animal test.
Antitumor Activity. Cell Lines and Cell Cultures.24-27 A human
chronic myelougenous leukemia cell line, K562, was maintained as
suspension cultures in RPMI1640 medium (SIGMA-Aldrich,
catalog no. R8758) supplemented with 10% fetal bovine serum
(FBS; Sigma-Aldrich, catalog no. F6178), 100 U/mL penicillin/100
μg/mL streptomycin (Sigma-Aldrich, catalog no. P4333), and
0.25 μg/mL amphothericin B (Sigma-Aldrich, catalog no. A2411)
at 37 °C in 5% CO2 (complete medium).
4-(p-Methylsulfonylphenyl)-4H-pyrrolo[2,1-c][1,4]benzoxazine
(29). 29 was prepared from the corresponding methylthio
following the procedure described above. The title compound
was obtained as a white solid in 45% yield. Mp: 109-111 °C
(hexane/ethyl acetate). IR (KBr), ν (cm-1): 1519 (CdC); 1413
(SO2); 1224 ((Ar-O); 956 (C-O); 1043 (SdO). 1H NMR
(CDCl3, 200 MHz) δ (ppm): 3.10 (s, 3H, CH3-S); 5.75 (dt, J1 =
3 Hz, J2=1 Hz, 1H, C-3H); 6.21 (s, 1H, C-4H); 6.33 (t, J=3 Hz,
A human nonsmall lung cancer cell line, NCI-H460, a human
colon cancer cell line, HT-29, which constitutively expresses
COX-2, and a human fibroblastic cell line, HuDe, were cultured
in MEM medium (Sigma-Aldrich, catalog no. M4655) supple-
mented with 10% FBS, 100 U/mL penicillin, 100 μg/mL strep-
tomycin, and 0.25 μg/mL amphothericin B at 37 °C in 5% CO2.

Article
Proliferation Assay. K562 cell line was cultured in a 24-well
microplate by adding 1 mL of complete medium containing
2.0 ꢀ 105 cells in the absence or presence of test compounds
[10 μM].
The effects of test compounds on cell viability were assayed
on a portion of the cell suspension. The cell number was deter-
mined with a hemocytometer, and viability was estimated by
trypan blue dye exclusion.
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18 6569
and SC-558 (6COX).32 The X-ray crystal structure of COX-2
with an inhibitor SC-558 was treated as a template molecule to
construct the complexes of COX-2 with NSAIDS.
The AutoDock 3.0 program33 was used to explore the dock-
ing for different conformations of the 15 NSAIDs in the active
site of the enzyme. The AutoDock exploration was carried out
within a 30 A cube by using 0.25 A grid spacing. Seven affinity
grids were calculated (C, N, O, S, H, Cl, F).
Instead, the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-
zolium bromide (MTT) assay was used to determine cell viabil-
ity of NCI-H460, HT-29, and HuDe cell lines.
MTT is a yellow tetrazolium salt that is taken up and cleaved
only by metabolically active cells, which reduce it to a colored,
water-insoluble formazan salt. The solubilized formazan prod-
uct can be quantified via absorbance at 570 nm (690 nm for
blank), which is measured using a 96-well-format spectrophotom-
eter (ELx-800-BioTek instruments). The absorbance correlates
directly with the cell number.
The simulated annealing protocol consisted of 100 runs of
50 cycles, each cycle including 25 000 accepted or 25 000 rejected
relative positions. A distance dependent dielectric constant
equal to 4r was used to simulate a partially solvated state. The
annealing temperature was set to 298 K during the first cycle and
then linearly reduced at the end of each cycle.
Ligands were considered conformationally flexible by defin-
ing the torsion angles about which rotation was allowed. Auto-
Dock 3.0 was used to generate conformer binding mode with the
lowest docked energy. From the 100 simulations with each com-
pound, the COX-2-ligand complex structures in the top ranked
energy cluster were selected. To take into account protein flexi-
bility, the stability and behavior of all complexes were studied in
a dynamic context and the van der Waals and electrostatic
components of the interaction energy monitored. Further anal-
ysis of the binding orientation of the aryl methyl sulfones
indicated that this substituent adopts the arylpropionic position
of flurbiprofen in the crystal structure of COX-2 (3PGH).30
The lowest docked energy resulting complexes were energy-
minimized and equilibrated using the AMBER 10 program.34
The standard ionization state at neutral pH was considered for
the ionizable residues. The system was hydrated by an octahe-
dral box of TIP3P water molecules extending 11 A outside the
protein at all sides, resulting in an average of 13 850 waters. The
parm-99 file of the AMBER force field was used to describe the
enzyme. Restricted electrostatic potential fitted charges deter-
mined at the HF/6-31G(d) level, using the RESP procedure and
van der Waals parameters taken for related atom types in the
AMBER-99 force field, were used for each inhibitor. Torsional
parameters for the methylsulfone group were taken from those
reported in our previous studies.35-38 SHAKE was used to
maintain all the bonds at their equilibrium distances, and a
nonbonded 11 A cutoff and a distance-dependent dielectric
constant were used throughout. In each case, 100 steps of
steepest descent were followed by conjugate gradient until the
root-mean-square value of the potential energy gradient was
Cells were plated at 2.0 ꢀ 104 cell/well in 100 μL volume in
96-well plates and grown for 72 h in MEM complete medium.
Different concentrations of test drugs or 0.1% DMSO were
added to the wells. Then cells were incubated with 10 μL of MTT
(5 mg/mL) at 37 °C for 3 h. The tetrazolium crystals were solu-
bilized by the addition of 4.5 mL of isopropyl alcohol-0.5 mL of
Triton X-100 in 150 μL of 37% HCl.
Conformational Analysis. Molecular modeling of NSAIDs
was carried out on an O2 Silicon Graphics computer using the
X-ray crystallography data in the Cambridge Structural Data-
base and conformational analysis The conformational prefer-
ences in aqueous solution were determined computations using
the geometries optimized at the B3LYP/6-31G(d) level using the
program Gaussian 0328 and the relative free energies of hydra-
tion from the MST B3LYP/6-31G(d) version of the PCM
continuum model.29 A systematic exploration at the B3LYP/
6-31G(d) level was performed by changing the torsional angle
around the bonds CO-Ar and C-N-R1-R2 by increments of
60° and excluding those conformers where a steric clash was
observed. The geometry of the local minimum energy structures
was then fully optimized. The minimum energy nature of all the
stationary points located from B3LYP/6-31G(d) geometry opti-
mizations was verified from the analysis of the vibrational
frequencies, which were positive in all cases.
For all compounds the MO (molecular orbitals) wave func-
tions were determined and the following properties were exam-
ined: energies of the HOMO, NHOMO, LUMO, and NLUMO
frontier orbitals, net atomic charges, dipole moment (MD), and
the free energy of solvatation (ΔGsolv).
The energy of the HOMO is related to the ionization potential
and characterizes the molecular reactivity in front of the attack
by electrophiles, while the energy of the LUMO is related to the
electron affinity and describes the reactivity toward nucleo-
philes. In processes involving radicalary species, both HOMO
and LUMO have a large contribution. The HOMO-LUMO
gap, i.e., the difference in energy between the HOMO and the
LUMO, is an important stability index. A large HOMO-
LUMO gap implies high stability for the molecule because of
its lower reactivity in chemical reactions. The HOMO-LUMO
gap has thus been used as an approximation to the excitation
energy of the molecule.
below 0.01 kcal mol-1 A-1
.
The system was equilibrated by running five 60 ps molecular
dynamics (MD) simulations to increase the temperature to
298 K. Subsequently a 10 ns MD simulation was carried out
with an integration time step of 2 fs.
In all cases the positional root-mean square deviations (rmsd)
determined for the backbone and heavy atoms in the system
with regard to the corresponding X-ray crystallographic struc-
ture was in the range 0.9-1.8 A. The characterization of the
structural features that mediate the binding to the enzyme was
determined by averaging the geometrical parameters for the
snapshots (saved every ps) sampled along the last 2 ns of the MD
simulation. The ptraj module of AMBER was used to perform
most numerical analyses of the MD trajectories.
To refine the structures to be used for free energy analysis,
energy minimizations of the entire protein-ligand complexes
were performed in implicit water solvent models.
The solvated complexes were minimized with 2000 steps of
conjugate-gradient minimization without restraints, employing
a residue-based cutoff of 11 A. After each energy minimization,
visual inspection of the complexes was performed to make sure
that the protein and the ligand remained close to conformation
observed in the crystal structures.
Overall, there are two major conformational families that
differ in the relative orientation of the aromatic rings and several
subfamilies that arise from rotation about the R1SO2 bond. The
energy differences among them were sufficiently small to con-
sider them all as candidates for the bound conformation; other
conformers were located, but they were not considered because
of the large destabilization (more than 4 kcal/mol) relative to the
most stable conformer.
Molecular Modeling Methods. NSAID docking was done in
mouse COX-2 enzyme taken from X-ray crystallographic struc-
tures, complexed to flubiprofen (3PGH),30 diclofenac (1PXX),31
MM-GBSA Computations. The calculation and decomposi-
tion of binding free energy, ΔGbind(MM-GBSA), between protein
and ligands were evaluated using the MM-GBSA (molecular

6570 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18
Harrak et al.
mechanics generalized Born/surface area) method as implemen-
ted in AMBER 9.37
(3) Gund, P.; Shen, T. Y. A model for the prostaglandin synthetase
cyclooxygenase site and its inhibition by antiinflammatory aryla-
cetic acids. J. Med. Chem. 1977, 20, 1146–1152.
MM-GBSA has consistently been shown to be a good method
for comparing binding energies of complexes similar to those in
this case.38,39 MM-GBSA computes the binding free energy
by using a thermodynamic path that combines the molecular
mechanical energies with the continuum solvent approaches.
In the MM-GBSA method, the total binding free energy in
(4) Jouzeau, J. Y.; Terlain, B.; Abid, A.; Nedelec, E.; Netter, P. Cyclo-
oxygenase isoenzymes. How recent findings affect thinking about
nonsteroidal anti-inflammatory drugs. Drugs 1997, 53, 563–582.
(5) (a) Mc Gettigan, P.; Henry, D. Current problems with non-specific
COX inhibitors. Curr. Pharm. Des. 2000, 6, 1693–1724. (b) Barbaric,
M.; Kralj, M.; Marjanovic, M.; Husnjak, I.; Pavelic, K.; Filipavic-Grcic,
J.; Zorc, D.; Zorc, B. Synthesis and in vitro antitumor effect of
diclofenac and fenoprofen thiolated and non-thiolated polyaspartamide-
drug conjugates. Eur. J. Med. Chem. 2007, 42, 20–29.
(6) Moore, B. C.; Simmons, D. L. COX-2 inhibition. Apoptosis and
chemoprevention by nonsteroidal anti-inflammatory drugs. Curr.
Med. Chem. 2000, 7, 1131–1144.
(7) Chen, L.; He, Y.; Huang, H.; Liao, H.; Wei, W. Selective COX-2
inhibitor celecoxib combined with EGFR-TKI ZD 1839 on non-
small cell lung cancer cell lines: in vitro toxicity and mechanism
study. Med. Oncol. 2008, 25, 161–171.
(8) Thun, M. J.; Henley, S. J.; Pairono, C. Nonsteroidal anti-inflammatory
drugs as anticancer agents: mechanistic, pharmacologic, and clini-
cal issues. J. Natl. Cancer Inst. 2002, 94, 252–266.
(9) Soh, J.-W.; Kazi, J. V.; Li, H.; Thompson, W. J.; Weinstein, B.
Celcoxib-induced growth inhibition in SW480 colon cancer cells is
associated with activation of protein kinase G. Mol. Carcinog.
2008, 47, 519–525.
(10) Harris, R. E.; Beebe-Donk, J.; Doss, H.; Burr-Doss, D. Aspirin,
ibuprofen, and other non-steroidal anti-inflammatory drugs in
cancer prevention: a critical review of non-selective COX-2 block-
ade (review). Oncol. Rep. 2005, 13, 559–583.
(11) Arber, N.; Levin, B. Chemoprevention of colorectal neoplasia: the
potential for personalized medicine. Gastroenterology 2008, 134,
1224–1237.
(12) Williams, J. L.; Borgo, S.; Hasan, I.; Castillo, E.; Traganos, F.;
Rigas, B. Nitric oxide-releasing nonsteroidal anti-inflammatory
drugs (NSAIDs) alter the kinetics of human colon cancer cell lines
more effectively than traditional NSAIDs: implications for colon
cancer chemoprevention. Cancer Res. 2001, 61, 3285–3289.
(13) Robak, P.; Smolewski, P.; Robak, T. The role of non-steroidal anti-
inflammatory drugs in the risk of development and treatment of
hematologic malignances. Leuk. Lymphoma 2008, 49, 1452–1462.
(14) Chen, L. C.; Ashcroft, D. M. Risk of myocardial infarction
associated with selective COX-2 inhibitors: meta-analysis of ran-
domised controlled trials. Pharmacoepidemiol. Drug Saf. 2007, 16,
762–772.
water is approximated by ΔGbind =ΔEGAS þ ΔGGB þ ΔGSUR
.
The ΔEGAS is the energy difference of the solutes in the two
(bound and unbound) states. The ΔGGB is the polar part of the
solvation free energy represented by the generalized Born
approach. The ΔGSUR is the apolar surface part of the free
energy solvation of a cavity inside the solvent). In this formula,
the conformational entropy of the solute is not explicitly con-
sidered, although the solvent entropy is implicitly considered in
the ΔGGB and ΔEGAS
.
The electrostatic term is computed by adding the solvent
screened electrostatic interaction between ligand and enzyme,
and the corresponding change in the desolvation free energy of
the ligand upon binding. According to this procedure, the
desolvation cost of the enzyme is assumed to be common for
all complexes between COX-2 and inhibitors, which is justified
by the fact that all of them share a common chemical skeleton
and occupy the same position in the binding site while providing
a substantial saving in computer time.
MM-GBSA computations were performed for a set of 100
structures of the enzyme-inhibitor complex taken from the last
2 ns of the MD simulation and were obtained using the Amber
package. Prior to the calculations, the complexes were energy-
minimized to eliminate bad contacts; at the end all water mole-
cules were removed. The relative binding affinity was deter-
mined from the most favorable binding free energies determined
for each inhibitor.
Acknowledgment. The authors express their sincere grati-
tude to the Ministerio de Ciencia y Tecnologı
CTQ2007-60614/BQU) and the Departament d’Universitats,
´
a (Grant
ꢁ
Recerca i Societat de la Informacio de la Generalitat de
(15) Peri, K. G.; Hardy, P.; Li, D. Y.; Varma, D. R.; Chemtob, S.
Prostaglandin G/H synthase-2 is a major contributor of brain
prostaglandins in the newborn. J. Biol. Chem. 1995, 270, 24615–
24620.
Catalunya, Spain (Grant 2005-SGR-00180), for the financial
support. J.B. acknowledgesthe Generalitat deCatalunya for a
predoctoral fellowship.
(16) Capilla, A. S.; Pujol, M. D. A convenient method for the prepara-
tion of substituted naphtho[2,3-b]-1,4-dioxin by the Diels-Alder
reaction. Synth. Commun. 1996, 26, 1729–1738.
Supporting Information Available: Dipole moments (Figure
2A-C), molecular modeling studies (Figure 6-8), and the
preparation of organic compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
ꢁ
(17) Vazquez, M. T.; Rosell, G.; Pujol, M. D. Synthesis and anti-
inflammatory activity of rac-2-(2,3-dihydro-1,4-benzodioxin)pro-
pionic acid and its R and S enantiomers. Eur. J. Med. Chem. 1997,
32, 529–534 and references cited herein.
(18) Hamdouchi, C.; de Blas, J.; Ezquerra, J. A novel application of the
Ullmann coupling reaction for the alkylsulfenylation of 2-amino-
imidazo[1,2-a]pyridine. Tetrahedron 1999, 55, 541–552.
(19) Lindley, J. Copper-assisted nucleophilic substitution of aryl halo-
gen. Tetrahedron 1984, 40, 1433–1438.
References
(1) (a) Bertolini, A.; Ottani, A.; Sandrini, M. Selective COX-2 inhibi-
tors and dual anti-inflammatory drugs: critical remarks. Curr.
Med. Chem. 2002, 9, 1033–1043. (b) Howardell, M. J., Ed. COX-2
Inhibitor Research; Nova Science: New York, 2006. (c) Yu, G.; Praveen
Rao, P. N.; Chowdhury, M. A.; Abdellatif, K. R. A.; Dong, Y.; Das, D.;
ꢁ
(20) Sanchez, I.; Pujol, M. D. A convenient synthesis of pyrrolo[2,1-
c][1,4]benzodioxines. Tetrahedron 1999, 55, 5593–5598.
(21) Romero, M.; Harrak, Y.; Basset, J.; Ginet, L.; Constans, P.; Pujol,
M. D. Preparation of N-arylpiperazines and other N-aryl com-
pounds from aryl bromides as scaffolds of bioactive compounds.
Tetrahedron 2006, 62, 9010–9016.
ꢁ
Velazquez, C. A.; Suresh, M. R.; Knaus, E. E. Synthesis and biological
evaluation of N-difluoromethyl-1,2-dihydropyrid-2-one acetic acid
regioisomers: dual inhibitors of cyclooxygenases and 5-lipoxygenase.
Bioorg. Med. Chem. Lett. 2010, 20, 2168–2173. (d) Chowdhury,
M. A.; Abdellatif, K. R. A.; Dong, Y.; Das, D.; Suresh, M. R.; Knaus,
E. E. Synthesis of celecoxib analogues possessing a N-difluoromethyl-
1,2-dihydropyrid-2-one 5-lipoxygenase pharmacophore: biological
evaluation as dual inhibitors of cyclooxygenases and 5-lipoxygenase
with anti-inflammatory activity. J. Med. Chem. 2009, 52, 1525–1529.
(2) (a) Dannhardt, G.; Kiefer, W. Cyclooxygenase inhibitors;current
status and future prospects. Eur. J. Med. Chem. 2001, 36, 109–126.
ꢁ
ꢁ
(22) Lozano, J. J.; Lopez, M.; Ruiz, J.; Vazquez, I. J.; Pouplana, R.
QSAR in the Nonsteroidal Antiinflammatory Agents: The Fena-
mic Acids. Trends in QSAR and Molecular Modelling 92; Wermuth,
C. G., Ed.; ESCOM: Leiden, The Netherlands, 1993; pp 560-61.
(23) Limongelli, V.; Bonomi, M.; Marinelli, L.; Gervasio, F. L.; Cavalli,
A.; Novellino, E.; Parrinello, M. Molecular basis of cyclooxy-
genase enzymes (COXs) selective inhibition. Proc. Natl. Acad. Sci.
U.S.A. 2010, 107, 5411–5416.
ꢁ
(b) Puig, C.; Crespo, M. I.; Godessart, N.; Feixas, J.; Ibarzo, J.; Jimenez,
(24) Waskewich, C.; Blumenthal, R. D.; Li, H.; Stein, R.; Goldenberg,
D. M.; Burton, J. Celecoxib exhibits the greatest potency amongst
cyclooxigenase (COX) inhibitors for growth inhibition of COX-2
negative hemapoietic and epithelial cell lines. Cancer Res. 2002, 62,
2029–2033.
J. M.; Soca, L.; Cardelus, I.; Heredia, A.; Miralpeix, M.; Puig, J.; Beleta,
J.; M. Huerta, J. M.; Lopez, M.; Segarra, V.; Ryder, H.; Palacios, J. M.
Synthesis and biological evaluation of 3,4-diaryloxazolones: a new
class of orally active cyclooxygenase-2 inhibitors. J. Med. Chem.
2000, 43, 214–223. (c) Kiefer, W.; Dannhardt, G. Novel insights and
therapeutical applications in the field of inhibitors of COX-2. Curr.
Med. Chem. 2004, 11, 3147–3161.
ꢁ
(25) Giles, J. F.; Kantarjian, H. M.; Bekele, B. N.; Cortes, J. E.; Faderl,
S.; Thomas, D. T.; Manshouri, T.; Rogers, A.; Keating, M. J.;

Article
Talpaz, M.; O’Brien, S.; Albitar, M. Bone marrow cyclooxygenase-2
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 18 6571
(31) Rowlinson, S. W.; Kiefer, J. R.; Prusakiewicz, J. J.; Pawliotz, J. L.;
Kozak, K. R.; Kalgutkar, A. S.; Stallings, W. C.; Kurumbail,
R. G.; Marnett, L. J. A novel mechanism of cyclooxygenase-2
inhibition involving interactions with Ser-530 and Tyr-385. J. Biol.
Chem. 2003, 278, 45763.
(32) Kurumbail, R. G.; Stevens, A. M; Gierse, J. K.; McDonald, J. J.;
Stegeman, R. A.; Pak, J. Y.; Gildehaus, D.; Miyashiro, J. M.;
Penning, T. D.; Seibert, K; Isakson, P. C.; Stallings, W. C. Struc-
tural basis for selective inhibition of cyclooxygenase-2 by anti-
inflammatory agents. Nature 1997, 385, 555.
levels are elevated in chronic phase chronic myeloid leukaemia
and are associated with reduced survival. Br. J. Haematol. 2002,
119, 38–45.
(26) Cianchi, F.; Cortesini, C.; Fantappie, O.; Messerini, L.; Sardi, I.;
Lasagna, N.; Perna, F.; Fabbroni, V.; Di Felice, A.; Perigli, G.;
Mazzanti, R.; Masini, E. Cyclooxygenase-2 activation mediates the
proangiogenic effect of nitric oxide in colorectal cancer. Clin.
Cancer Res. 2004, 10, 2694–2704.
(27) Shin, Y. K.; Park, J. S.; Kim, H. S.; Jun, H. J.; Kim, G. E.; Suh,
C. O.; Yun, Y. S.; Pyo, H. Radiosensivity enhancement by cel-
ecoxib, a cyclooxygenase (COX)-2 selective inhibitor, via COX-2-
dependent cell cycle regulation on human cancer cells expressing
differential COX-2 levels. Cancer. Res. 2005, 65, 9501–9509.
(28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi,
T.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts,
R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth,
G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. B.;
Cui, Q.; Baboul, A. J.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Gonzalez, C.; Pople, J. A.. Gaussian 03, revision
D.02; Gaussian, Inc.: Wallingford, CT, 2004.
(33) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,
W. E.; Belew, R. K.; Olson, A. J. Automated docking using a
Lamarckian genetic algorithm and empirical binding free energy
function. J. Comput. Chem. 1998, 19, 1639–1662.
(34) Case, D. A.; Darden, T. A.; Cheatham, T. E.; Simmerling, C. L.;
Wang, J.; Duke, R. E.; Luo, R.; Crowley, M.; Walker, R. C.;
Zhang, W.; Merz, K. M.; Wang, B.; Hayik, S.; Roitberg, A.;
ꢁ
Seabra, G.; Kolossvary, I.; Wong, K. F.; Paesani, F.; Vanicek, J.;
Wu, X.; Brozell, S. R.; Steinbrecher, T.; Gohlke, H.; Yang, L.; Tan,
C.; Mongan, J.; Hornak, V.; Cui, G.; Mathews, D. H.; Seetin,
M. G.; Sagui, C.; Babin, V. Kollman, P. A. AMBER 10; University
of California: San Francisco, CA, 2008.
ꢁ
(35) Pouplana, R.; Lozano, J. J.; Perez, C; Ruiz, J. Structure-based
QSAR study on differential inhibition of human prostaglandin endo-
peroxide H synthase-2 (COX-2) by nonsteroidal anti-inflammatory
drugs. J. Comput.-Aided Mol. Des. 2002, 16, 683–710.
(36) Pouplana, R.; Lozano, J. J.; Ruiz, J. Molecular modeling of the dif-
ferential interaction between several non-steroidal anti-inflammatory
drugs and human prostaglandin endoperoxide H synthase-2
(H-PGHS-2). J. Mol. Graphics Modell. 2002, 20, 329–343.
(37) Kollman, P. A.; Massova, I; Reyes, C. Calculating structures
and free energies of complex molecules: combining molecular
mechanics and continuum models. Acc. Chem. Res. 2000, 33, 889–
897.
(29) Soteras, I.; Curuchet, A.; Bidon-Chanal, A.; Orozco, M.; Luque,
ꢁ
F. J. Extension of the MST model to the IEF formalism: HF and
(38) Srinivasan, J.; Cheatham, T. E.; Cieplak, P.; Kollman, P. A.; Case,
D. A. Continuum solvent studies of the stability of DNA, RNA,
and phosphoramidate-DNA helices. J. Am. Chem. Soc. 1998, 120,
9401–9409.
B3LYP parametrizations. J. Mol. Struct.: THEOCHEM 2005,
727, 29–40.
(30) Kurumbail, R. G; Stevens, A. M; Gierse, J. K.; McDonald, J. J.;
Stegeman, R. A.; Pak, J. Y.; Gildehaus, D.; Miyashiro, J. M.;
Penning, T. D.; Seibert, K; Isakson, P. C.; Stallings, W. C. Struc-
tural basis for selective inhibition of cyclooxygenase-2 by anti-
inflammatory agents. Nature 1996, 384, 644.
ꢁ
ꢁ
ꢁ
(39) Perez, C.; Sanchez, J.; Marmol, F.; Puig-Parellada, P.; Pouplana,
R. Reactivity of biologically important NSAID compounds with
superoxide (O2 -) and nitric oxide ( NO) and cyclooxygenase
3
3
inhibition. QSAR Comb. Sci. 2007, 26, 368–377.