Investigations of new lead structures for the design of novel cyclooxygenase-2 inhibitors

Article abstract of DOI:10.1016/S0223-5234(02)01373-9

On the basis of molecular modelling studies, five new compounds were synthesised and studied in an attempt to design new lead structures as selective COX-2 inhibitors.

Full text of DOI:10.1016/S0223-5234(02)01373-9

Eur. J. Med. Chem. 37 (2002) 461–468
www.elsevier.com/locate/ejmech
Original article
Investigations of new lead structures for the design of novel
cyclooxygenase-2 inhibitors
Chang Ha Park a, Xavier Siomboing b, Sa¨ıd Yous a, Bernard Gressier b,
Michel Luyckx b, Philippe Chavatte a,*
a Yang Ji Chemical Co., Ltd. 638-6, Sungkok-Dong, Ansan-City, Kyunggi-Do, South Korea
b Laboratoire de Pharmacologie, Pharmacocine´tique et Pharmacie Clinique, Faculte´ des Sciences Pharmaceutiques et Biologiques, BP 83,
59006 Lille Cedex, France
Received 20 August 2001; received in revised form 30 November 2001; accepted 6 December 2001
Abstract
On the basis of molecular modelling studies, five new compounds were synthesised and studied in an attempt to design new lead
´
structures as selective COX-2 inhibitors. © 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
Keywords: NSAIDs; Cyclooxygenase (COX-1, COX-2); Selective COX-2 inhibitors; Thromboxane B2
or treat a number of cancers [5,6] and to delay or slow
the clinical expression of Alzheimer’s disease [7,8].
Therefore, considerable efforts have been made to dis-
cover selective COX-2 inhibitors, which are structurally
related to two general classes, the diarylheterocycles
and the methanesulphonanilides. Two compounds in
the first class, celecoxib [9] and rofecoxib [10] and one
in the second, nimesulid, have recently been marketed
for the treatment of acute pain, osteoarthritis and
rheumatoid arthritis (Fig. 1). Their clinical efficacy was
shown similar to that of classical NSAIDs and was
associated with a greater gastrointestinal safety [11].
More recently, the structures of some non selective
NSAIDs such as indomethacin, zomepirac, aspirin and
flurbiprofen have been successfully converted into
COX-2 selective inhibitors [12]. In order to facilitate the
design and to predict the activity of new selective
COX-2 inhibitors, we have previously performed a
three-dimensional quantitative structure-activity rela-
tionship study (3D-QSAR) [13] using comparative
molecular field analysis (CoMFA) [14]. We obtained a
predictive model from an extensive series of various
derivatives known as selective COX-2 inhibitors. This
robust model has been used to design novel selective
COX-2 inhibitors and to predict their biological activi-
ties. Five compounds were then selected according to
their predicted COX-2 inhibitory potency and their
1. Introduction
Non-steroidal anti-inflammatory drugs (NSAIDs)
such as aspirin, ibuprofen or indomethacin are widely
used in the treatment of pain and inflammatory diseases
[1]. Their clinical efficacy is closely related to their
ability to inhibit both COX-1 and COX-2 isoforms of
the enzyme cyclooxygenase (COX) also referred to as
prostaglandin H2 synthase since it catalyses the conver-
sion of arachidonic acid to prostaglandin H2 (PGH2)
[2]. The constitutive COX-1 isoform is mainly responsi-
ble for the synthesis of prostaglandins which exert
cytoprotective effect in the gastrointestinal tract and
control renal function in the kidneys, whereas, the
inducible COX-2 is selectively activated by pro-inflam-
matory stimuli cytokines (IL-1) and growth factors
(FGF, PDGF) and facilitates the release of
prostaglandins involved in the inflammatory process [3].
These observations suggest that selective COX-2 in-
hibitors could provide antiinflammatory, analgesic and
antipyretic drugs devoid of the unwanted side effects
such as ulcers and renal failure associated with the
classical nonselective NSAIDs [4]. Selective COX-2 in-
hibition could also be an important strategy to prevent
* Correspondence and reprints
E-mail address: pchavatt@phare.univ-lille2.fr (P. Chavatte).
´
0223-5234/02/$ - see front matter © 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S0223-5234(02)01373-9

462
C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
chemical accessibility (Table 1). We report here, the
synthesis and the pharmacological evaluation of these
derivatives, which can be considered as new leads for
chemical optimisation.
2. Chemistry
Compounds listed in Table 1 were synthesised ac-
cording to Figs. 2–5.
Fig. 2. (a) PPA. (b) Br2, MeCO2H, chloroform. (c) N(Et)3, acetone.
(d) MeSO2Cl, pyridine.
Fig. 1. Structures of selective COX-2 inhibitors.
Fig. 3. (a) NaOH, MeOH, H2O. (b) i, NaOEt; ii, MeI, DMF. (c)
MeSO2Cl, pyridine. (d) CuBr2, AcOEt, chloroform. (e) N(C2H5)3,
acetone.
Table 1
Predicted and experimental IC50 values of studied compounds
Fig. 4. (a) Na2SO3, NaHCO3, ClCH2CO2H; (b) ClCH2C6H3Cl2(3,4),
K2CO3, acetone; (c) NaOH, MeOH, H2O; (d) (Ac)2O, pyridine,
methylene chloride.
The synthesis of indenone 5 is shown in Fig. 2.
Intramolecular cyclisation of chalcone 1 [15] was
achieved by treatment with polyphosphoric acid [16] at
120 °C affording indanone 2 in 30% yield. The reaction
of 2 with bromine in acetic acid–chloroform gave
compound 3. Dehydrobromination of 3 with triethy-
lamine in acetone proceeded at reflux, giving 4. Treat-
ment of 4 with methanesulphonyl chloride in pyridine
gave 5 in 76% yield.

C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
463
3. Pharmacology
Fig. 3 illustrates the synthesis of compound 10. The
oxazolinone ring of 2 was opened using sodium hydro-
xide in methanol–water medium. Selective alkylation of
the phenolic group of 6 by methyl iodide in the pre-
sence of sodium ethoxide in DMF gave 7. N-sulphona-
tion of 7 using methanesulphonyl chloride in pyridine
afforded 8. The indenone 10 was obtained in two-step
process. First, generation of the brominated compound
9 with copper (II) bromide in ethyl acetate–chloroform,
and second, dehydrobromination with triethylamine in
acetone.
Access to compound 15 is presented in Fig. 4. From
compound 11 [17], the sulphone 12 was obtained by
action of sodium sulphite and chloroacetic acid in
water [18]. N-alkylation with 3,4-dichlorobenzyl chlo-
ride in the presence of potassium carbonate in DMF
gave 13. Ring opening of the oxazolinonic ring afforded
compound 14, which by treatment with acetic anhy-
dride in methylene chloride in the presence of pyridine
led to the aminoester 15.
The ability of compounds 5, 10, 15, 19 and 20 to
inhibit both COX-1 and COX-2 isoforms was deter-
mined in a cell assay using mononuclear cells (mono-
cytes, lymphocytes and platelets) isolated from human
whole blood by density gradient centrifugation [20].
Their activities against the two isoforms were evaluated
measuring the production of thromboxane B2 (TxB2)
[21] by a specific radioimmunoassay. Western blot
analyses were previously carried out to detect the pre-
sence of both isoforms. To assess the accuracy of the
model, different inhibitors with distinct selectivity for
COX-1 versus COX-2 were used as reference subs-
tances: aspirin [22,23], ibuprofen [24,25] and nimesulid
[26,27]. The cytotoxicity of our compounds was also
estimated measuring lactate dehydrogenase activity
[28].
4. Results and discussion
The synthesis of compounds 19–20 is outlined in Fig.
5, starting from 1-ethyl-3-(4-methylsulphanylphenyl)-
urea (16), which was prepared as previously described
[19]. According to standard methods, compound 16 was
N-acylated with the appropriate benzoyl chloride in
methylene chloride to give derivatives 17–18. Finally,
compounds 19–20 were obtained by oxidation of the
appropriate compounds 17–18 with oxone in acetone.
The five compounds which were synthesised on ac-
count of their activity values predicted in our previous
3D-QSAR model [13] are reported in Table 1. They
present some common structural features of most
COX-2 inhibitors, namely a phenylmethylsulphonyl or
a methanesulphonamide group in the vicinity of an-
other phenyl ring generally substituted by halogens or
lower alkyl groups [29,30]. Compounds 5 and 10 can be
considered as structural analogues of nimesulid and
flosulid whereas 15, 19 and 20 are related to opened
diarylheterocycles. In our 3D-QSAR model these com-
pounds exhibit predicted IC50 values lower than 1 mM
whereas the predicted IC50 values of celecoxib is 0.03
mM. Compound 15 was, therefore, predicted as active
as celecoxib and compound 5 ten fold more potent.
The design of these compounds was facilitated by the
availability of the X-ray crystal structure of murine
COX-2 bound with SC-558, a selective COX-2 inhibitor
[31]. The SC-558 conformation extracted from the X-
ray crystallographic inhibitor-enzyme complex was used
to direct the design of compounds 15, 19 and 20,
assuming that this conformation represents the most
probable bioactive conformation of the diarylheterocy-
cle derivatives at the enzyme level site. The flexible
docking of 5, the most interesting derivative according
to its predicted activity value, was performed into the
enzyme active site using the FlexiDock module of
SYBYL [32]. This study suggests that hydrogen bonding
occurs between 5 and three residues of the enzyme
active site, His90, Arg120 and Tyr355 (Fig. 6). Com-
pound 5 binds to the positive-charged Arg120 in the
COX-2 nonselective binding site. Its methylsulphonyl
group occupies the COX-2 side-pocket without inter-
acting with the hydrophilic residues His90 or Arg513
Fig. 5. (a) EtNCO, THF; (b) ClCOC6H4F(4) or ClCOC6H3F2(3,5),
pyridine, CH2Cl2; (c) oxone, acetone.
Fig. 6. Compound 5 docked into the COX-2 active site. H-bonds are
shown (dashed lines).

464
C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
[29]. However, we notice that sp2 and sp3 oxygen atoms
of the benzoxazolinone ring accept H-bonds from
His90 and Tyr355, respectively. This study has also
shown that 5 could be correctly fitted on the bioactive
conformation of SC-558 when positioned in the protein
cavity.
COX-2 ratio). Furthermore, we think that this could
also explain the great difference between its COX-2
predicted and experimental inhibitory values. Neverthe-
less, the high COX-2 predicted value for this compound
needs and justify further chemical and pharmacological
stu-dies, which are developed after present time.
The results of the experimental IC50 values obtained
with both COX-1 and COX-2 are summarised in Table
1. The reference compound nimesulid is around 200
times more potent against COX-2 than COX-1,
whereas aspirin and ibuprofen are ca. equipotent. These
results are in agreement with previous published data
[24–27]. As for references, our compounds were able to
inhibit the TxB2 production by stimulated human
mononuclear cells in a dose-dependent manner. The
experimental values are in good agreement with the
predicted ones except for compound 5, which theoreti-
cally was the best inhibitor. A possible explanation for
this exception could be related with the test used in this
study and the physicochemical properties of this com-
pound. As a matter of fact: (1) the COX-1 IC50 value
of 5 could not be accurately evaluated because of its
lack of solubility. (2) In our 3D-QSAR model, the
inhibitory values of the selected compounds had been
obtained from the same biological method using the
recombinant enzyme [33]. (3) In the current study we
preferred a whole cell test since it takes into account the
penetration of the compounds through the membranes.
Compounds 10, 15 and 19 were found to be almost
three to four times more potent than nimesulid in terms
of COX-2 inhibition but their COX-1/COX-2 ratios
were lower than this reference. Compound 20 emerges
as the most promising one since it exhibits a COX-1/
COX-2 ratio value of 298 and a COX-2 inhibitory
activity of 0.57 mM, higher than that of the reference
inhibitor. For all compounds tested, we have also
checked the lack of cytotoxicity (B5%) dosing lactate
dehydrogenase activity.
6. Experimental
6.1. Chemistry
Melting points (m.p.) were determined on a Bu¨chi
SMP-20 capillary apparatus and are uncorrected. IR
spectra were recorded on a Vector 22 Brucker spec-
1
trophotometer. H-NMR spectra were recorded on a
AC 300 Brucker spectrometer. Chemical shifts are re-
ported in l units (parts per million) relative to (Me)4Si.
Elemental analyses for final substances were performed
by CNRS Laboratories (Vernaison, France). Obtained
results were within 0.4% of the theoretical values.
6.1.1. 5-(4-Chlorophenyl)-3H-indano[5,6-d]oxazol-2,7-
dione (2)
Compound 1 (39 g, 130 mmol) and PPA (400 g) were
heated at 120 °C during 1 h. The reaction mixture was
quenched with ice water and the precipitate was
filtered, washed with water and dried. The obtained
solid was refluxed in chloroform (200 mL) for 2 h.
After filtration and evaporation of the organic phase,
the residue was recrystallised from acetonitrile (12 g,
30%): m.p. 250–251 °C; IR w NH 3200 cm−1, CO
1775, 1670 cm−1; H-NMR (300 MHz, DMSO-d6) l
1
2.55 (dd, 1H, J=3.79, 18.90 Hz), 3.22 (dd, 1H, J=
7.56, 18.90 Hz), 4.65 (dd, 1H, J=3.79, 7.56 Hz), 6.79
(s, 1H), 7.22 (d, 1H, J=8.64 Hz), 7.38 (d, 1H, J=8.64
Hz), 7.50 (s, 1H), 12.00 (br s, 1H, exchangeable with
D2O).
5. Conclusions
6.1.2. 6-Bromo-5-(4-chlorophenyl)-3H-indano[5,6-d]-
oxazol-2,7-dione (3)
In conclusion, we have described the synthesis of five
novel selective COX-2 inhibitors that we designed on
the basis of a 3D-QSAR study. Their biological evalua-
tion using a whole cell assay has confirmed the predic-
tive power of our 3D-QSAR model since they were
found at least as potent as nimesulid against COX-2.
Nevertheless only compound 20 has shown an interes-
ting selectivity with a COX-1/COX-2 ratio reaching
298. These results allow us to select it as a lead com-
pound in view to develop a new series of potent and
selective COX-2 inhibitors. Moreover, compound 5 that
was the best predicted one cannot be dissolved to a
concentration higher 5 mM. Its lack of solubility did not
allow us to accurately determine its selectivity (COX-1/
To a cooled suspension of compound 2 (3.0 g, 10
mmol) in 60 mL of a mixture of chloroform–acetic acid
(1/1) was added dropwise under stirring a solution of
bromine (0.52 mL, 10 mmol) in chloroform (20 mL).
The reaction mixture was stirred at r.t. for 12 h.
Solvents were evaporated under reduced pressure and
the solid was triturated with petroleum ether. The
precipitate was dried and recrystallised from acetoni-
trile (1.63 g, 43%): m.p. 235–236 °C; IR w NH 3140
cm−1, CO 1770, 1670 cm−1 1H-NMR (300 MHz,
;
DMSO-d6) l 4.81 (d, 1H, J=4.29 Hz), 5.12 (d, 1H,
J=4.29 Hz), 6.54 (s, 1H), 7.21 (d, 2H, J=8.37 Hz),
7.46 (d, 2H, J=8.37 Hz), 7.68 (s, 1H), 12.25 (br s, 1H,
exchangeable with D2O).

C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
465
6.1.3. 5-(4-Chlorophenyl)-3H-indeno[5,6-d]oxazol-
2,7-dione (4)
pressure. The residue was dissolved in dimethylfor-
mamide (30 mL) and methyl iodide (1.18 mL, 18.9
mmol) was added slowly. After stirring for 1 h at r.t.,
the reaction mixture was poured into water (200 mL)
and acidified with an aqueous solution of 6 M HCl.
The precipitate was filtered, washed with water, dried
and recrystallised from EtOAc (3.21 g, 63%): m.p.
100–110 °C; IR w NH+3 3000–2600 cm−1, CO 1700
To a solution of compound 3 (1.0 g, 2.64 mmol) in
acetone (30 mL) was added triethylamine (1.47 mL, 11
mmol). The reaction mixture was refluxed for 12 h and
acetone was evaporated under reduced pressure. The
solid was triturated with water (100 mL) and acidified
with an aqueous solution of 1 M HCl. The precipitate
was filtered, washed with water, dried and recrystallised
from dioxane (0.48 g, 61%): m.p.\270 °C; IR w NH
3209 cm−1, CO 1815, 1690 cm−1; 1H-NMR (300 MHz,
DMSO-d6) l 6.19 (s, 1H), 7.01 (s, 1H), 7.38 (s, 1H),
7.61 (d, 2H, J=8.57 Hz), 7.74 (d, 2H, J=8.57 Hz),
12.04 (br s, 1H, exchangeable with D2O).
1
cm−1; H-NMR (300 MHz, DMSO-d6) l 2.55 (dd, 1H,
J=3.20, 18.87 Hz), 3.16 (dd, 1H, J=7.60, 18.87 Hz),
3.86 (s, 3H), 4.43 (dd, 1H, J=3.20, 7.60 Hz), 6.37 (s,
1H), 6.93–7.01 (m, 4H, NH+3 exchangeable with D2O),
7.14–7.28 (m, 4H).
6.1.7. N-[3-(4-Chlorophenyl)-6-methoxy-1-oxo-
6.1.4. 5-(4-Chlorophenyl)-3-methanesulphonyl-3H-
indeno[5,6-d]oxazol-2,7-dione (5)
indan-5-yl]methanesulphonamide (8)
To a suspension of compound 7 (0.3 g, 0.93 mmol) in
pyridine (3 mL) was added slowly methanesulphonyl
chloride (0.1 mL, 1.2 mmol). The reaction mixture was
stirred for 12 h at r.t., poured into water and acidified
with an aqueous solution of 3 M HCl. The precipitate
was filtered, washed with water, dried and recrystallised
from EtOAc (0.20 g, 60%): m.p. 169–170 °C; IR w NH
To a suspension of compound 4 (2.0 g, 6.7 mmol) in
pyridine (10 mL) was added slowly methanesulphonyl
chloride (0.70 mL, 8.7 mmol). The reaction mixture was
stirred for 12 h at room temperature (r.t.), poured into
water and acidified with an aqueous solution of 3 M
HCl. The precipitate was filtered, washed with water,
dried and recrystallised from acetonitrile (1.91 g, 76%):
m.p. 222–223 °C; IR w CO 1804, 1708 cm−1; 1H-NMR
(300 MHz, DMSO-d6) l 3.73 (s, 3H), 6.34 (s, 1H), 7.54
(s, 1H), 7.68 (s, 1H), 7.69 (d, 2H, J=8.72 Hz), 7.78 (d,
2H, J=8.72 Hz). Anal. Calc. for C17H10ClNO5S: C,
54.34; H, 2.68; N, 3.73. Found: C, 54.16; H, 2.66; N,
3.85%.
3340 cm−1, CO 1700 cm−1 1H-NMR (300 MHz,
;
DMSO-d6) l 2.60 (dd, 1H, J=3.46, 18.61 Hz), 3.20
(dd, 1H, J=7.79, 18.61 Hz), 3.51 (s, 3H), 3.99 (s, 3H),
4.53 (dd, 1H, H3, J=3.46, 7.79 Hz), 7.21 (s, 1H), 7.24
(s, 1H), 7.26 (d, 2H, J=8.66 Hz), 7.52 (d, 2H, J=8.66
Hz), 10.95 (br s, 1H, exchangeable with D2O).
6.1.8. N-[2-Bromo-3-(4-Chlorophenyl)-6-methoxy-
1-oxo-indan-5-yl]methanesulphonamide (9)
6.1.5. 5-Amino-3-(4-chlorophenyl)-6-hydroxyindan-
1-one (6)
Compound 8 (4.3 g, 11.8 mmol) and copper(II) bro-
mide (3.2 g, 14.1 mmol) in 120 mL of a mixture of
chloroform–EtOAc(1/1) were refluxed for 12 h. The
reaction mixture was filtered and concentrated under
reduced pressure. The precipitate was filtered, dried and
recrystallised from ethanol (2.6 g, 45%): m.p. 182–
To a solution of compound 2 (10.0 g, 34 mmol) in
methanol (120 mL) was added a solution of sodium
hydroxide (20 g, 500 mmol) in water (120 mL). The
reaction mixture was refluxed for 24 h. After cooling,
the solution was acidified with an aqueous solution of
6M HCl, filtered and basified with a saturated solution
of NaHCO3. The precipitate was filtered, washed with
water, dried and recrystallised from EtOAc(4.65 g,
50%): m.p. 260–261 °C; IR w OH 3360 cm−1, NH2
184 °C; IR w NH 3262 cm−1, CO 1720 cm−1 1H-
;
NMR (300 MHz, DMSO-d6) l 3.09 (s, 3H), 3.94 (s,
3H), 4.75 (d, 1H, J=4.42 Hz), 5.02 (d, 1H, J=4.42
Hz), 6.54 (s, 1H), 7.21 (d, 2H, J=8.37 Hz), 7.46 (d,
2H, J=8.37 Hz), 7.68 (s, 1H), 12.25 (br s, 1H, ex-
changeable with D2O).
3260 cm−1, CO 1650 cm−1 1H-NMR (300 MHz,
;
DMSO-d6) l 2.29 (dd, 1H, J=2.99, 18.55 Hz), 2.95
(dd, 1H, J=7.77, 18.55 Hz), 4.36 (dd, 1H, J=2.99,
7.77 Hz), 5.73 (br s, 2H, exchangeable with D2O), 6.25
(s, 1H), 6.85 (s, 1H), 7.14 (d, 2H, J=8.38 Hz), 7.34 (d,
2H, J=8.38 Hz), 9.72 (br s, 1H, exchangeable with
D2O).
6.1.9. N-[3-(4-Chlorophenyl)-6-methoxy-1-oxo-1H-
inden-5-yl]methanesulphonamide (10)
To a solution of compound 9 (2.6 g, 5.85 mmol) in
acetone (80 mL) was added triethylamine (3.3 mL, 23.4
mmol). The reaction mixture was stirred at r.t. for 72 h
and then evaporated under reduced pressure. An
aqueous (aq.) 1 M HCl solution was added and the
precipitate was filtered, washed with water, dried and
recrystallised from acetonitrile (0.40 g, 19%): m.p. 195–
6.1.6. 5-Amino-3-(4-chlorophenyl)-6-methoxyindan-
1-one, hydrochloride (7)
To a freshly prepared solution of sodium (0.44 g,
0.0189 at. gr) in absolute ethanol (120 mL) was added
6 (4.3 g, 15.7 mmol). The reaction mixture was stirred
at r.t. during 1 h and then evaporated under reduced
197 °C; IR w NH 3247 cm−1, CO 1699 cm−1
;
1H-
NMR (300 MHz, DMSO-d6) d 3.04 (s, 3H), 3.92 (s,

466
C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
3H), 6.20 (s, 1H), 7.28 (s, 1H), 7.39 (s, 1H), 7.38 (s,
1H), 7.65 (d, 2H, J=8.53 Hz), 7.77 (d, 2H, J=8.53
Hz), 9.29 (br s, 1H, exchangeable with D2O). Anal.
C17H14ClNO4S: C, 56.12; H, 3.87; N, 3.85. Found: C,
56.35; H, 3.86; N, 3.72%.
quenched with water and extracted with methylene
chloride. The organic layer was separated, washed with
a 2 M HCl solution, dried over MgSO4 and evaporated
under reduced pressure. The solid was triturated with
diethylether, filtered and recrystallised from toluene
(0.34 g, 30%): m.p. 127–128 °C; IR w NH 3393 cm−1
,
6.1.10. 6-Methanesulphonyl-2(3H)-benzoxazolone (12)
To a solution of Na2SO3 (0.63 g, 5.0 mmol) and
NaHCO3 (1.26 g, 15.0 mmol) in water (60 mL) was
added at 75 °C portionwise and over 10 min compound
11 (1.0 g, 5.0 mmol). Heating was continued for 1 h and
chloroacetic acid (0.7 g, 7.5 mmol) was added. The
resulting solution was refluxed for 48 h. After cooling
the precipitate was collected by filtration, dried and
recrystallised from acetonitrile (0.34 g, 32%): m.p.\
CO 1764 cm−1; 1H-NMR (300 MHz, DMSO-d6) l 2.36
(s, 3H), 3.00 (s, 3H), 4.40 (d, 2H, J=5.60 Hz), 6.66 (d,
1H, J=8.38 Hz), 7.11 (br s, 1H, 1H exchangeable with
D2O), 7.28 (dd, 1H, J=1.20, 8.38 Hz), 7.48–7.50 (m,
2H), 7.58–7.64 (m, 2H). Anal. C16H15Cl2NO4S: C,
49.48; H, 3.89; N, 3.60. Found: C, 49.62; H, 3.76; N,
3.56%.
6.1.14. 1-(4-Fluorobenzoyl-1-(4-methylsul-
phanylphenyl)-3-ethylurea (17)
270 °C; IR w NH 3271 cm−1, CO 1788 cm−1
;
1H-
NMR (300 MHz, DMSO-d6) l 3.35 (s, 3H), 7.30 (d,
1H, J=8.27 Hz), 7.72 (d, 1H, J=8.27 Hz), 7.85 (s,
1H), 12.23 (br s, 1H, exchangeable with D2O).
To a cooled solution of compound 16 (0.8 g, 3.8
mmol) in dichloromethane (60 mL) were added pyri-
dine (1.3 mL, 15.3 mmol) and slowly 4-fluorobenzoyl
chloride (0.55 mL, 4.6 mmol). The reaction mixture was
refluxed for 12 h, poured into water and acidified with
an aq. 1 M HCl solution. The organic layer was washed
with water, dried over MgSO4 and evaporated under
reduced pressure. The residue was recrystallised from
cyclohexane (0.34 g, 27%): m.p. 143–145 °C; IR w NH
3310 cm−1, CO 1683, 1632 cm−1; 1H-NMR (300 MHz,
DMSO-d6) l 1.28 (t, 3H, J=7.44 Hz), 2.42 (s, 3H),
3.44 (m, 1H), 6.86–7.31 (m, 4H), 8.13–8.19 (m, 4H),
8.95 (m, 1H, exchangeable with D2O).
6.1.11. 3-(3,4-Dichlorobenzyl)-6-methanesul-
phonyl-2(3H)-benzoxazolone (13)
A mixture of compound 12 (1.6 g, 7.5 mmol), K2CO3
(3.1 g, 22.5 mmol) in dimethylformamide (10 mL) was
heated at 70 °C during 1 h. 3, 4-Dichlorobenzyl chlo-
ride chloride (1.25 mL, 9 mmol) was added and heating
was continued for an additional 2 h. After cooling, the
reaction mixture was poured into water and acidified
with an aqueous solution of 6 M HCl. The precipitate
was filtered, washed with water, dried and recrystallised
from toluene from acetonitrile (2.74 g, 98%): m.p.
6.1.15. 1-(3,5-Difluorobenzoyl)-1-(4-methylsulphanyl-
phenyl)-3-ethylurea (18)
211–212 °C; IR w CO 1760 cm−1 1H-NMR (300
;
MHz, DMSO-d6) d 3.22 (s, 3H), 5.12 (s, 2H), 7.38 (d,
1H, J=8.25 Hz), 7.47 (d, 1H, J=8.26 Hz), 7.62 (d,
1H, J=8.26 Hz), 7.76–7.80 (m, 2H), 7.93 (s, 1H).
This compound was prepared following the proce-
dure described for compound 17. From compound 16
(1.0 g, 4.8 mmol), pyridine (1.2 mL, 14.4 mmol) and
3,5-difluorobenzoyl chloride (0.8 mL, 5.76 mmol). Re-
crystallisation from cyclohexane (0.8 g, 48%): m.p.
119–121 °C; IR w NH 3313 cm−1, CO 1712, 1660
6.1.12. 2-(3,4-Dichlorobenzylamino)-5-methanesul-
phonyl-phenol (14)
1
This compound was prepared following the proce-
dure described for compound 6. From compound 13
(3.1 g, 8.3 mmol), sodium hydroxide (1.15 g, 29 mmol),
methanol (150 mL) and water (50 mL). Recrystallisa-
tion from acetonitrile (2.0 g, 71%): m.p. 173 °C; IR w
OH 3410 cm−1; NH 3400 1H-NMR (300 MHz,
DMSO-d6) l 3.00 (s, 3H), 4.45 (d, 2H, J=5.70 Hz),
6.42 (d, 1H, J=8.80 Hz), 6.55 (br s, 1H, exchangeable
with D2O), 7.10–7.15 (m, 2H), 7.32 (dd, 1H, J=2.07,
8.28 Hz), 7.55 (d, 1H, J=8.28 Hz), 7.60 (s, 1H), 10.20
(br s, 1H, exchangeable with D2O).
cm−1; H-NMR (300 MHz, DMSO-d6) l 1.27 (t, 3H,
J=7.37 Hz), 2.44 (s, 3H), 3.40 (m, 1H), 6.69–6.82 (m,
3H), 7.02 (d, 2H, J=8.42 Hz), 7.12 (d, 2H, J=8.42
Hz), 8.72 (m, 1H, exchangeable with D2O).
6.1.16. 1-(4-Fluorobenzoyl)-1-(4-methanesulphony-
lphenyl)-3-ethylurea (19)
To a solution of compound 17 (0.17 g, 0.50 mmol) in
acetone (5 mL) was added a solution of oxone (1.2 g,
1.8 mmol) in 10 mL water. After stirring for 1 h at r.t.,
the reaction mixture was diluted with water and ex-
tracted with EtOAc. The organic phase was dried over
MgSO4 and evaporated under reduced pressure. The
residue was recrystallised from EtOAc (0.11 g, 60%):
m.p. 130–131 °C; IR w NH 3312 cm−1, CO 1711, 1664
6.1.13. 2-(3,4-Dichlorobenzylamino)-5-methanesul-
phonylphenyl acetate (15)
To a solution of compound 14 (1.0 g, 2.9 mmol) in
methylene chloride (40 mL) were added pyridine (0.73
mL, 8.7 mmol) and acetic anhydride (0.4 mL, 4.3
mmol). The reaction mixture was stirred at r.t. for 12 h,
cm−1 1H-NMR (300 MHz, CDCl3) l 1.26 (m, 3H,
;
J=7.11 Hz), 3.02 (s, 3H), 3.40 (m, 2H), 6.87–6.92 (m,
2H), 7.24–7.29 (m, 2H) 7.35 (d, 2H, J=8.23 Hz), 7.84

C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
467
(d, 2H, J=8.23 Hz), 8.88 (m, 1H, exchangeable with
D2O). Anal. C17H17FN2O4S: C, 56.04; H, 4.70; N,
7.68. Found: C, 56.12; H, 4.75; N, 7.75%.
(400×g for 10 min at 4 °C), and the supernatant
was also separated. The enzymatic activity of lactate
deshydrogenase was measured at 30 °C by enzymatic
kinetic read at 340 nm after adding 0.2 mM NADH
and 1.6 mM pyruvate, as described in manufacturer’s
protocol [28]. The percentage of lactate deshydroge-
nase released was estimated in comparison with the
maximal activity of this enzyme.
6.1.17. 1-(3,5-Difluorobenzoyl)-1-(4-methanesulphonyl-
phenyl)-3-ethylurea (20)
This compound was prepared following the proce-
dure described for compound 19. From 18 (0.5 g,
1.43 mmol), acetone (20 mL) and oxone (3.2 g, 5.2
mmol) in 25 mL water. Purification by column chro-
matography (EtOAc–cyclohexane: 6/4) (0.48 g, 88%):
m.p. 118–120 °C; IR w NH 3312 cm−1, CO 1711,
6.2.4. In 6itro COX-1 assay
The lymphocytes and the monocytes were preincu-
bated with the inhibition, if necessary, then they were
incubated with calcium ionophore 50 mM A 23187
and 50 mM arachidonic acid for 1 h at 37 °C. At the
end of the incubation, the tubes were centrifuged
(400×g for 10 min at 4 °C). The TxB2 was extracted
and measured from the supernatant as described
later.
1
1664 cm−1; H-NMR (300 MHz, CDCl3) l 1.26 (m,
3H, J=7.28 Hz), 3.02 (s, 3H), 3.42 (m, 2H), 6.74–
6.79 (m, 3H), 7.36 (d, 2H, J=8.33 Hz), 7.87 (d, 2H,
J=8.33 Hz), 8.73 (m, 1H, exchangeable with D2O).
Anal. C17H16F2N2O4S: C, 53.40; H, 4.21; N, 7.32.
Found: C, 53.42; H, 4.32; N, 7.43%.
6.2. Pharmacology
6.2.5. In 6itro COX-2 assay
When the lymphocytes and the monocytes had
been incubated with 500 mM acetyl salicylic acid for
30 min to block the catalytic site of COX-1, the com-
pound was eliminated with several washings. The cells
were placed in contact with the inhibitors, if neces-
sary, and incubated with 8 mg mL−1 lipopolysaccha-
ride and 50 mM arachidonic acid for 3 h at 37 °C.
Cell suspensions were centrifuged (400×g for 10 min
at 4 °C) and TxB2 was extracted and measured from
the supernatant as described later.
6.2.1. Chemical products
Buffer compounds and chemical products were pur-
chased from Sigma Chemical (St Louis, MO, USA).
Primary and secondary antibodies were purchased
from Tebu (Le Perray en Yvelines, France). ECL
Western blotting detection reagents and Hyperfilm
ECL were purchased from Amersham International
plc (Buckinghamshire, UK).
6.2.2. Cell separation technique
Mononuclear cells were isolated as described in the
first step of PMN preparation [20]. Once 15 mL fresh
by drawn human heparinised blood, obtained from
healthy donors, has been diluted 1:2 with a 0.1 M
phosphate buffer saline, pH 7.4, 10 mL of Histo-
paque®-1077 was placed at the bottom of a conical
tube. After centrifugation (400×g for 30 min at
20 °C), the upper layer, composed of monocytes,
lymphocytes and platelets, was recovered. The suspen-
sion was centrifuged (400×g for 10 min at 4 °C),
and the pellet was resuspended in a Hanks’HEPES
buffer pH 7.4. Cell viability, determined by blue Try-
pan exclusion, was over 95%. The cells were kept at
4 °C until use.
6.2.6. Extraction of thromboxane B2 [34]
Two milliliter of acetone were added to 1 mL of
supernatant, and the tubes were shaken for 2 min.
They were then centrifuged (400×g for 10 min at
4 °C). The supernatants were transferred to a sepa-
rate tube and 2 mL of hexane were added. The tubes
were shaken for 2 min and then centrifuged (400×g
for 5 min at 4 °C). The upper layers were kept and
the pH of each lower layer was adjusted to 3.0–4.0
with 1 M citric acid solution. Following this, 2 mL of
chloroform was added, the tubes were shaken for 2
min, and centrifuged (400×g for 5 min at 4 °C). The
lower chloroform layers containing TxB2, were sepa-
rated, and the top layer was processed, for another
extraction, with 2 mL of chloroform. The two chloro-
form extracts were combined. The tubes were dried
under nitrogen, and TxB2 was assayed as described
by the manufacturer’s protocol.
6.2.3. Assessment of the cytotoxicity of the compounds
The cytotoxicity of the different inhibitors was esti-
mated by dosing lactate deshydrogenase activity in
the cellular supernatant 5×106 cells were incubated
in 1 mL of Hank’s Hepes buffer at 37 °C with sev-
eral concentrations of the compound. The suspension
was centrifuged (400×g for 10 min at 4 °C), all su-
pernatants were separated, the pellet of the reference
tube was resuspended with 1 mL of water and shaken
for 5 min to lyse the cells. This tube was centrifuged
6.2.7. Statistical studies
The data obtained were subjected to statistical ana-
lysis using a non-parametric test (Student’s), where
P-values B0.05 were considered significant versus
reference.

468
C.H. Park et al. / European Journal of Medicinal Chemistry 37 (2002) 461–468
Miyashiro, R.S. Rogers, D.J. Rogier, S.S. Yu, G.D. Anderson,
6.3. Molecular modelling
E.G. Burton, J.N. Cogburn, S.A. Gregory, C.M. Koboldt, W.E.
Perkins, K. Seibert, A.W. Veenhuizen, Y.Y. Zhang, P.C. Isak-
son, J. Med. Chem. 40 (1997) 1347–1365.
Molecular modelling studies were performed using
SYBYL software version 6.4 [32] running on Silicon
Graphics workstations. The geometry of all compounds
was subsequently optimised using the Tripos force field
[35] including the electrostatic term calculated from
Gasteiger and Hu¨ckel atomic charges. The method of
Powell available in Maximin2 procedure was used for
energy minimisation until the gradient value was
[10] P. Prasit, Z. Wang, C. Brideau, C.C. Chan, S. Charleson, W.
Cromlish, D. Ethier, J.F. Evans, A.W. Ford-Hutchinson, J.Y.
Gauthier, R. Gordon, J. Guay, M. Gresser, S. Kargman, B.
Kennedy, Y. Leblanc, S. Leger, J. Mancini, G.P. O’Neill, M.
Ouellet, M.D. Percival, H. Perrier, D. Riendeau, I. Rodger, R.
Zamboni, S. Boyce, N. Rupniak, M. Forrest, D. Visco, D.
Patrick, Bioorg. Med. Chem. Lett. 9 (1999) 1773–1778.
[11] G.W. Cannon, F.C. Breedveld, Am. J. Med. 110 (2001) 6–12.
[12] A.S. Kalgutkar, A.B. Marnett, B.C. Crews, R.P. Remmel, L.J.
Marnett, J. Med. Chem. 43 (2000) 2860–2870.
smaller than 0.001 kcal mol−1
−1. The structure of
,
A
the COX-2 enzyme was obtained from its complexed
X-ray crystal structure with SC-558 available in the
RCSB Protein Data Bank (1CX2) [31]. Flexible doc-
king of compounds into the enzyme active site was
performed using the FlexiDock module of SYBYL. The
complexes were energy-minimised using the Powell
method available in Maximin2 procedure with the
MMFF94 force field [36] and a dielectric constant of
4.0 until the gradient value reached 0.05 kcal mol−1
[13] P. Chavatte, S. Yous, C. Marot, N. Baurin, D. Lesieur, J. Med.
Chem. 44 (2001) 3223–3230.
[14] R.D. Cramer III, D.E. Patterson, J.D. Bunce, J. Am. Chem. Soc.
110 (1988) 5959–5967.
[15] J.P. Bonte, D. Lesieur, C. Lespagnol, J.C. Cazin, Eur. J. Med.
Chem. 9 (1974) 497–500.
[16] B. Merdji, D. Lesieur, C. Lespagnol, D. Barbry, D. Couturier, J.
Heterocycl. Chem. 18 (1981) 1223–1227.
[17] B. Scudi, J. Am. Chem. Soc. 63 (1941) 879–885.
[18] R.W. Brown, J. Org. Chem. 56 (1991) 4974–4976.
[19] J.W. Chern, Y.L. Leu, S.S. Wang, J. Med. Chem. 40 (1997)
2276–2286.
[20] A. Cabanis, B. Gressier, S. Lebegue, C. Brunet, T. Dine, M.
Luyckx, M. Cazin, J.C. Cazin, APMIS 102 (1994) 119–121.
[21] F.A. Kuehl, R.W. Egan, Science 210 (1980) 978–983.
[22] E.A. Meade, W.L. Smith, D.L. De Witt, J. Biol. Chem. 268
(1993) 6610–6614.
,
A
−1. The anneal function was used defining around the
,
inhibitor a hot region (10 A) and an interesting region
,
(15 A).
Acknowledgements
[23] J.M. Young, S. Panah, C. Satchawatcharaphong, P.S. Cheung,
Inflamm. Res. 45 (1996) 246–253.
The authors are grateful to Yanikem laboratory for
financial support. They also thank Dr. C. Bochu, Labo-
ratoire d’Application de Re´sonance Magne´tique Nucle´-
aire de Lille 2 for his aid with the interpretation of the
1H-NMR spectra.
[24] G.P. O’Neill, J.A. Mancini, S. Kargman, J. Yergey, M.Y. Kwan,
J.P. Falgueyret, Mol. Pharmacol. 45 (1994) 245–254.
[25] D. Riendeau, M.D. Percival, S. Boyce, C. Brideau, S. Charleson,
W. Cromlish, Br. J. Pharmacol. 121 (1997) 105–117.
[26] I.A. Tavares, P.M. Bishai, A. Bennett, Arzneimittelforschung 45
(1995) 1093–1095.
[27] J.P. Famaey, Inflamm. Res. 46 (1997) 437–446.
[28] H. Levert, B. Gressier, I. Moutard, C. Brunet, T. Dine, M.
Luyckx, M. Cazin, J.C. Cazin, Inflammation 22 (1998) 191–201.
[29] C.J. Hawkey, Lancet 353 (1999) 307–314.
References
[30] I.K. Khanna, Y. Yu, R.M. Huff, R.M. Weir, X. Xu, F.J.
Koszyk, P.W. Collins, J.N. Cogburn, P.C. Isakson, C.M.
Koboldt, J.M. Masferrer, W.E. Perkins, K. Seibert, A.W. Veen-
huizen, J. Yuan, D.C. Yang, Y.Y. Zhang, J. Med. Chem. 43
(2000) 3168–3185.
[31] R.G. Kurumbail, A.M. Stevens, J.K. Gierse, J.J. McDonald,
R.A. Stegeman, J.Y. Pak, D. Gildehaus, J.M. Miyashiro, T.D.
Penning, K. Seibert, P.C. Isakson, W.C. Stallings, Nature 384
(1996) 644–648.
[1] D.B. Reitz, P.C. Isakson, Curr. Pharm. Des. 1 (1995) 211–220.
[2] J.Y. Fu, J.L. Masferrer, K. Seibert, A. Raz, P. Needleman, J.
Biol. Chem. 265 (1990) 16737–16740.
[3] J.L. Masferrer, B.S. Zweifel, P.T. Manning, S.D. Hauser, K.M.
Leahy, W.G. Smith, P.C. Isakson, K. Seibert, Proc. Natl. Acad.
Sci. USA 95 (1994) 3228–3232.
[4] A. Ballinger, G. Smith, Expert Opin. Pharmacother. 2 (2001)
31–40.
[5] K. Subbaramaiah, D. Zakim, B.B. Weksler, A.J. Dannenberg,
Proc. Soc. Exp. Biol. Med. 216 (1997) 201–210.
[6] A.L. Hsu, T.T. Ching, D.S. Wang, X. Song, V.M. Rangnekar,
C.S. Chen, J. Biol. Chem. 275 (2000) 11397–11403.
[7] G.M. Pasinetti, J. Neurosci. Res. 54 (1998) 1–6.
[8] M. Hull, K. Lieb, B.L. Fiebich, Expert. Opin. Investig. Drugs 9
(2000) 671–683.
[32] SYBYL 6.4, Tripos Associates, Inc., 1699 South Hanley Road, St
Louis, MO 63144.
[33] J.K. Gierse, S.D. Hauser, D.P. Creely, C. Koboldt, S.H. Rang-
wala, P.C. Isakson, K. Seibert, Biochem. J. 305 (1995) 479–484.
[34] J.A. Salmon, R.J. Flower, Meth. Enz. 86 (1986) 477–493.
[35] M. Clark, R.D. Cramer III, N. Van Opdenbosch, J. Comput.
Chem. 10 (1989) 982–1012.
[9] T.D. Penning, J.J. Talley, S.R. Bertenshaw, J.S. Carter, P.W.
Collins, S. Docter, M.J. Graneto, L.F. Lee, J.W. Malecha, J.M.
[36] T. Halgren, J. Am. Chem. Soc. 112 (1990) 4710–4723.