Design and synthesis of novel celecoxib analogues as selective cyclooxygenase-2 (COX-2) inhibitors: Replacement of the sulfonamide pharmacophore by a sulfonylazide bioisostere

Article abstract of DOI:10.1016/j.bmc.2003.07.005

A group of celecoxib analogues in which the para-SO₂NH ₂ substituent on the N¹-phenyl ring was replaced by a para-sulfonylazido (SO₂N₃) 4, or a meta-SO ₂N₃ 8, substituent were designed for evaluation as selective cyclooxygenase-2 (COX-2) inhibitors. In vitro COX-1 and COX-2 inhibition studies showed that 4-[5-(4-methylphenyl)-3-trifluoromethyl-1H- pyrazol-1-yl]benzenesulfonyl azide (4) with a para-SO₂N₃ substituent was a selective COX-1 inhibitor. In contrast, 3-[5-(4-methylphenyl)- 3-trifluoromethylpyrazol-1-yl]benzenesulfonyl azide (8a) having a meta-SO ₂N₃ substituent (COX-1 IC₅₀ >100 μM; COX-2 IC₅₀=5.16 μM; COX-2 selectivity index >19.3) is a selective COX-2 inhibitor. A molecular modeling (docking) study showed that the SO₂N₃ group of 8a inserts deep inside the secondary pocket of the COX-2 binding site. The SO₂N₃ moiety of 8a can undergo a dual H-bonding interaction via one of its SO₂ oxygen-atoms, and an electrostatic (ion-ion) interaction via the terminal azido (N₃) nitrogen-atom, to the guanidino NH₂ of Arg ⁵¹³ in the secondary pocket of COX-2. These observations indicate that an appropriately positioned SO₂N₃ moiety is a novel alternative bioisostere to the traditional SO₂NH₂ and SO₂Me pharmacophores present in selective COX-2 inhibitors, that are only capable of H-bonding interactions with the COX-2 isozyme, for use in drug design.

Full text of DOI:10.1016/j.bmc.2003.07.005

Bioorganic & Medicinal Chemistry 11 (2003) 5273–5280
Design and Synthesis of Novel Celecoxib Analogues as Selective
Cyclooxygenase-2 (COX-2) Inhibitors:Replacement of the
Sulfonamide Pharmacophore by a Sulfonylazide Bioisostere
Md. Jashim Uddin, P. N. Praveen Rao and Edward E. Knaus*
Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2N8
Received 5 June 2003; accepted 28 July 2003
Abstract—A group of celecoxib analogues in which the para-SO2NH2 substituent on the N1-phenyl ring was replaced by a para-sul-
fonylazido (SO2N3) 4, or a meta-SO2N3 8, substituent were designed for evaluation as selective cyclooxygenase-2 (COX-2) inhibitors.
In vitro COX-1 and COX-2 inhibition studies showed that 4-[5-(4-methylphenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]benzenesulfonyl
azide (4) with a para-SO2N3 substituent was a selective COX-1 inhibitor. In contrast, 3-[5-(4-methylphenyl)-3-trifluoromethylpyrazol-
1-yl]benzenesulfonyl azide (8a) having a meta-SO2N3 substituent (COX-1 IC50 >100 mM; COX-2 IC50=5.16 mM; COX-2 selectivity
index >19.3) is a selective COX-2 inhibitor. A molecular modeling (docking) study showed that the SO2N3 group of 8a inserts deep
inside the secondary pocket of the COX-2 binding site. The SO2N3 moiety of 8a can undergo a dual H-bonding interaction via one
of its SO2 oxygen-atoms, and an electrostatic (ion–ion) interaction via the terminal azido (N3) nitrogen-atom, to the guanidino NH2
of Arg513 in the secondary pocket of COX-2. These observations indicate that an appropriately positioned SO2N3 moiety is a novel
alternative bioisostere to the traditional SO2NH2 and SO2Me pharmacophores present in selective COX-2 inhibitors, that are only
capable of H-bonding interactions with the COX-2 isozyme, for use in drug design.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
tyrosine (Tyr355),5 and interaction of Arg513 with the
bound drug is a requirementfor time-dependentinhibi-
tion of COX-2.6 The SO2NH2 moiety of SC-558, an
analogue of celecoxib (1) where Me is replaced by a Br
substituent, interacts with His90, Gln192 and Arg513. One
sulfonamide oxygen atom forms a H-bond to His90, t he
other oxygen atom is linked by a H-bond to Arg513, and
the nitrogen atom is H-bonded to the carbonyl oxygen
of Phe518.3 Recently we exploited, for the first time, the
amino acid Arg513 to design selective COX-2 inhibitors
having a dipolar azide (N3) pharmacophore that can
undergo an electrostatic (ion–ion) interaction with
Arg513 in the COX-2 2o-pocket. In this regard, a
molecular modeling study, where the azido compound
2, namely 1-(4-azidophenyl)-5-(4-methylphenyl)-3-tri-
fluoromethyl-1H-pyrazole was docked in the active site
of the COX-2 isozyme, showed that (i) the terminal
nitrogen atom of the azide substituent was inserted
deep into the 2ꢀ-COX pocket, and (ii) the ligand–
enzyme intermolecular and electrostatic binding ener-
gies for the azido compound 2 were significantly larger
than that for celecoxib (1) in which the corresponding
SO2NH2 pharmacophore is only capable of H-bonding
(Fig. 1).7
Development of the non-ulcerogenic selective COX-2
inhibitor celecoxib (Celebrex1, 1)1 provided a sig-
nificant advancement in the treatment of rheumatoid
arthritis and osteoarthritis.2 The SO2NH2 pharmaco-
phore presentin celecoxib is believed ot induce COX-2
selectivity by insertion into the secondary (2o) pocketof
the COX-2 binding site that is absent in COX-1. This
2o-pocket in COX-2 is formed due to a conformational
355
change atTyr
that is attributed to the presence of
Ile523 in COX-1 relative to Val523 having a smaller side
chain in COX-2.3 Accordingly, the combined volume of
the primary COX-2 binding site and its associated
2o-pocket(394 A ) is about 25% larger than the COX-1
3
binding site (316 A3).4 Ithas been reporetd ht atrepla-
513
cementof His
in COX-1 by Arg513 in COX-2 plays a
key role in the hydrogen-bond network of the COX-2
binding site. Access of ligands to the 2-pocket of COX-2
is controlled by histidine (His90), glutamine (Gln192) and
*Corresponding author. Tel.: +1-780-492-5993; fax: +1-780-492-
1217; e-mail: eknaus@pharmacy.ualberta.ca
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.07.005

5274
Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
Figure 1. Structures for celecoxib (1) and 1-(4-azidophenyl)-5-(4-
methylphenyl)-3-trifluoromethyl-1H-pyrazole (2).
As a partof our ongoing program to develop new drug
design concepts,8,9 we now reporta novel dipolar sulfo-
nylazide pharmacophore that has the potential to
undergo dual H-bonding (sulfonyl oxygens) and elec-
trostatic (ion–ion) interactions (azido) with amino acid
residues lining the 2ꢀ-pocketof the COX-2 binding site.
Accordingly, a group of celecoxib analogues having a
SO2N3 moiety at the ortho-, meta-, or para-position of
the N1-phenyl ring were synthesized, their COX-1/
COX-2 inhibition selectivity indices were determined,
and some molecular modeling studies are described
where target compounds are docked in the binding site
of the COX-2 isozyme.
Chemistry
The diazotization of the previously reported1,7 anilino
compound 3 gave rise to the corresponding diazonium
salt which on treatment with 37% w/v SO2 in HOAc in
the presence of CuCl2 using a method described by Gil-
bert10 yielded the intermediate sulfonyl chloride deriva-
tive. Subsequent azidation11 of this SO2Cl intermediate
productusing NaN 3 afforded the target product 4-[5-(4-
methylphenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]bene-
zenesulfonyl azide (4, 22% yield) where the SO2NH2
pharmacophore of celecoxib has been replaced by a
SO2N3 substituent (Scheme 1).
Scheme 2. Synthesis of 8a–f. Reagents and conditions: (a) 95% EtOH,
reflux, 24 h; (b) ClSO3H, 25 ꢀC, 1 h; (c) NaN3, acetone, H2O, 0 ꢀC, 3 h.
Condensation1,7 of a 1-(para-substituted-phenyl)-4,4,4-tri-
fluorobutane-1,3-dione 5a–e1 with either phenylhydrazine
hydrochloride (6a), or 4-sulfonamidophenylhydrazine
hydrochloride (6b),12 afforded the respective pyrazole
product 7a–f in 39–92% yield (see Scheme 2). Reactiꢀon
of the N1-phenylpyrazoles (7a–e) with ClSO3H at 25 C
afforded the respective meta-benzenesulfonyl chloride in
situ,13 which on reaction with NaN3, gave the respective
meta-benzenesulfonyl azide product 8a–e (46–70%
yield). In contrast, similar in situ chlorosulfonation13
and azidation of 7f having a N1-(4-sulfonamidophenyl)
substituent yielded the corresponding N1-(4-sulfon-
amido-2-sulfonylazidophenyl) product 8f (45%).
Electrophilic chlorosulfonation of compound 7a, and
the related compounds 7b–e, occurred atthe meta-posi-
tion of the N1-phenyl ring system that is attributed to
the electron-withdrawing effect of the protonated pyra-
zole ring. Chlorosulfonation at the meta-position of the
N1-phenyl ring is consistent with data obtained from a
geometry optimized PM3 calculation14 for 7a which
showed that the electron density was highest at the
meta-position (À0.097) relative to the ortho- (À0.0065)
and para- (À0.082) positons of the N1-phenyl ring. In
contrast, chlorosulfonation of compound 7f having a
sulfonamido substituent gave compound 8f in which the
chlorosulfonyl group was incorporated ortho to the
Scheme 1. Synthesis of 4. Reagents and conditions: (a) NaNO2, con-
centrated HCl, 0–5 ꢀC, 30 min (in dark); (b) 37% w/v SO2 in HOAc,
CuCl2, 40–50 ꢀC, 30 min (in dark); (c) NaN3, acetone, H2O, 0 ꢀC, 3 h.

Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
5275
N1-pyrazole nitrogen or meta to the sulfonamido sub-
stituent. This exclusive chlorosulfonation meta to the
sulfonamide substituent of 7f is due to the strong elec-
tron-withdrawing effect of the protonated sulonamido
group rather than protonation of the pyrazole ring. The
position at which this chlorosulfonation occurred is in
agreement with electron densities data acquired from a
PM3 calculation which showed the electron density was
low (0.009) atthe ortho-position, and high (À0.1000) at
the meta-position, to the sulfonamido substituent.
In a recent study we reported that the dipolar azido (N3)
substituent present in the azido compound 2 undergoes
an electrostatic (ion–ion) interaction with Arg513 in the
COX-2 2ꢀ-pocket. This observation, together with the
fact that the azido compound 2 is a selective COX-2
inhibitor, indicates that the dipolar azido substituent is
a pharmacophore of the H-bonding SO2NH2 moiety
presentin celecoxib ( 1).7 A comparison of substituent
volumes, calculated by PM3 geometry optimization using
the Alchemy 2000 program, showed the azido substituent
(25.5 A3) is smaller than the traditional SO2NH2
(44.8 A3) and SO2Me (49.7 A3) pharmacophores present
in selective COX-2 inhibitors, or a SO2N3 (58.6 A3)
substituent. A SO2N3 substituent has the potential to
undergo dual H-bonding (sulfonyl oxygens) and elec-
trostatic (ion-ion) interactions (dipolar N3) suggesting
that the SO2N3 moiety could serve as a new and alter-
native COX-2 pharmacophore to the SO2NH2 and
SO2Me pharmacophores.
Results and Discussion
The dipolar azido substituent has attracted our atten-
tion as a pharmacophore for use in drug design since it
has the potential to undergo electrostatic (ion–ion)
binding interactions with enzymes or pharmacological
receptors. In this regard, covalent azides can exist as
resonance hybrids between ionic species A, B and C (see
Fig. 2).15 Pauling eliminated C as a major contributor
based on the adjacent charge rule.16 Accordingly, the
remaining hybrids A and B predicta 2.5 bond order for
the N2–N3 bond and a 1.5 bond order for the N1–N2
bond (see D, Fig. 2). A structure determination for
methyl azide (D) is also in good agreement with this
prediction since the bond lengths from an electron dif-
fraction study17 are as follows: N2–N3=1.12 A,
N1–N2=1.24 A, C–N1=1.47 A, and the C–N1–N2 bond
angle was 120ꢀ. The azido group has a linear configur-
ation that is consistent with the sp3 hybridization indi-
cated by the lack of nonbonded electron pairs on N2.15
Replacementof the SO 2NH2 moiety of celecoxib (1) by
a SO2N3 substituent gave the target compound 4-[5-(4-
methylphenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]benze-
nesulfonylazide (4). In vitro COX inhibition studies
showed that 4 inhibited COX-1 selectively (COX-1
IC50=3.3 mM; COX-2 IC50 >100 mM) (see data in
Table 1). A molecular modeling study was therefore
performed where 4 was docked in the active site of the
murine COX-2 isozyme in order to explain its failure to
inhibitCOX-2. This docking experimentshowed 4 binds
in the center of the enzyme active site with the phe-
nylsulfonylazide moiety oriented towards the mouth of
the COX-2 channel such that the SO2N3 group is in
close proximity with Arg120 and Tyr355 (see Fig. 3). The
linear N3 group is involved in extensive electrostatic
interaction with the side chains of Arg120 and Tyr355
.
Llorens and coworkers have shown the importance of
the perturbation of the hydrogen bonding network
Figure 2. Resonance hybrid structures for the azido moiety.
Table 1. In vitro COX-1 and COX-2 inhibition data and molecular volumes of 4-[5-(4-methylphenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]benzene-
sulfonyl azide 4, 3-[5-(4-substituted-phenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]benzenesulfonyl azide 8a, 8b and 8d, and 5-sulfonamido-2-[5-(4-
methylphenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]benzenesulfonyl azide 8f
Compd
R1
R2
R3
R4
Vol (A3)a
IC50 (mM)b
COX-2
SIc
COX-1
4
Me
Me
OMe
F
Me
Me
SO2N3
H
H
H
SO2N3
SO2N3
SO2N3
H
H
H
H
H
SO2N3
H
312.6
312.6
321.3
300.4
357.9
279.4
3.3
>100
>500
38.1
>100
5.16
>100
>100
178.81
0.0507
<0.033
>19.36
>5.00
<0.38
0.0032
404
8a
8b
8d
8f
1
H
SO2NH2
SO2NH2
0.59
22.9
H
aThe volume of the molecule, after minimization using the MM3 force field, was calculated using the Alchemy 2000 program.
bThe in vitro test compound concentration required to produce 50% inhibition of COX-1 or COX-2. The result (IC50, mM) is the mean of two
determinations.
cSelectivity Index (SI)=COX-1 IC50/COX-2 IC50
.

5276
Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
selectivity index >19 (see data in Table 1). The selective
in vitro COX-2 inhibition exhibited by 8a is consistent
with observations from a molecular modeling experi-
mentwhere 8a was docked in the active site of the
COX-2 isozyme (see Fig. 4). Compound 8a binds in the
center of the active site with the phenylsulfonyl azide
moiety oriented towards the 2ꢀ-pocketregion. The
terminal N-atom of the m-SO2N3 is inserted about
4.25 A deep inside the entrance to the 2ꢀ-pocketof
COX-2 (Val523) and about5.20 A removed (within elec-
trostatic ion-ion interaction distance) from the NH2 of
Arg513. A SO2 oxygen atom of the SO2N3 group is
about2.74 A away (within H-bonding distance) from
the NH2 of Arg513. The C-3 CF3 substituent is posi-
tioned about 5.63 A from the NH2 of Arg120 and the
N2-nitrogen atom of the central pyrazole ring is located
about3.11 A from the OH of Tyr355. The center of the
C-5 phenyl ring is about5.87 A from the OH of Ser530
with the Me group at the para-position orienting in a
hydrophobic region surrounded by Leu384, Tyr385 and
Trp387. Molecular modeling studies of 1,5-diarylpyr-
azoles have shown that stereoelectronic effects of sub-
stituents at the para-position of the C-5 phenyl ring are
determinants of cyclooxygenase inhibition where
H-bonding, and electron-withdrawing, substituents
exhibitdiminished COX-1 and COX-2 binding affi-
nity.20 Accordingly, among this group of celecoxib ana-
logues 8, compound 8a with a SO2N3 moiety at the
meta-position of the N1-phenyl ring and a para-Me
substituent on the C-5 phenyl ring was the most potent
and selective COX-2 inhibitor. Introduction of a para-
OMe substituent on the C-5 phenyl ring abolished both
COX-1 and COX-2 inhibitory activity, and compound
8d having a 5-(4-fluorophenyl) substituent selectively
inhibited COX-1 (COX-1 IC50=38.1 mM; COX-2 IC50
Figure 3. Docking the pyrazole 4 (ball and stick) in the active site of
murine COX-2.
involving Arg120, Glu524, Tyr355 and His/Arg513 athte
mouth of the channel by different ligands and their
5
effecton COX inhibition. The SO2N3 terminal N-atom
is located about 4.93 A from the OH of Tyr355, a SO2
O-atom of the SO2N3 substituent is about 4.30 A away
from the OH of Tyr355, and the C-5 phenyl ring is
oriented toward Val523. The C-3 CF3 substituent is
oriented towards a hydrophobic pocket comprised of
Leu384, Tyr385 and Trp387, and itis posiitoned about
3.32 A from the OH of Tyr385 and about4.92 A from
the OH of Ser530. The N2-atom of the central pyrazole
ring is about5.88 A away from the OH of Ser530. Recent
studies have shown the importance of ionic interactions
involving the CO2H group of NSAIDs with Arg120 and
its critical role in COX-1 inhibition.18,19 The in vitro
COX inhibition data showing that 4 is a selective
COX-1 inhibitor are consistent with this docking
experiment which indicates that the dipolar N3 moiety
of the SO2N3 substituent is interacting with Arg120 at
the mouth of the primary binding site, like the CO2H
group of NSAIDs, rather than insertion into the
2ꢀ-pocketof hte COX-2 isozyme near Val
required for selective COX-2 inhibition.
523
that is
It was anticipated that relocation of the SO2N3 sub-
tituent from the para- to the meta-position of the
N1-phenyl ring of compound 4 may resultin a more
suitable orientation within the COX-2 binding side that
could provide selective COX-2 inhibition. In this regard,
in vitro COX inhibition studies showed that 3-[5-(4-
methylphenyl)-3-trifluoromethylpyrazol-1-yl)benzene-
sulfonylazide (8a) is a selective COX-2 inhibitor (COX-
2 IC50=5.16 mM; COX-1 IC50 >100 mM) with a COX-2
Figure 4. Docking the pyrazole 8a (ball and stick) in the active site of
murine COX-2.

Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
5277
>100 mM). Introduction of a SO2N3 substituent at the
ortho-position of the N1-phenyl ring of celecoxib pro-
duced 8f which was a more selective inhibitor of COX-1
(IC50=0.59 mM) than COX-2 (IC50=178.8 mM).
the solvent was removed in vacuo to yield the crude 4-
[5-(4-methylphenyl)-3-trifluoromethyl-1H-pyrazol-1-yl]-
benzenesulfonyl chloride as a brown syrup.10 Dissolu-
tion of this syrup in acetone (2 mL) and then drop wise
addition to
a stirred solution of NaN3 (0.04 g,
0.61 mmol) in water (2 mL) at À10 ꢀC to initiate the
azidation reaction.11 The reaction was allowed to pro-
ceed for 3 h at0 ꢀC with stirring. After warming to
25 ꢀC, the solvent was removed in vacuo, and the resi-
due was purified by silica gel column chromatography
using n-hexane–EtOAc (2:1, v/v) as eluent to afford the
benzenesulfonyl azide 4 (0.04 g, 22%) as a white solid;
mp 130–132 ꢀC; IR (film) n3100, 3020 (CHarom), 2138
Conclusions
It has been demonstrated that the meta-sulfonylazido
analogue 8a of celecoxib (1) is a selective COX-2 inhi-
bitor. Molecular modeling (docking) studies showed
that the sulfonylazido (SO2N3) group of 8a inserts deep
inside the 2ꢀ-pocketof the COX-2 isozyme where itcan
undergo both H-bonding via one of its SO2 oxygen-
atoms, and an electrostatic ion–ion interaction via its
terminal azide nitrogen atom, with the guanidino NH2
group of Arg513. These results indicate that a suitably
positioned sulfonylazido moiety is a potentially novel
dual H-bonding/electrostatic pharmacophore for the
design of potent COX-2 inhibitors with a high COX-2
selectivity index.
(N3), 1168, 1335 (SO2) cmÀ1 1H NMR (300 MHz;
;
CDCl3) d 7.34 (d, 2H, J=8.8 Hz, benzenesulfonyl
azido H-2, H-6), 7.26 (d, 2H, J=8.8 Hz, benzenesulfo-
nyl azido H-3, H-5), 7.16 (d, 2H, J=8.5 Hz, 4-methyl-
phenyl H-2, H-6), 7.10 (d, 2H, J=8.5 Hz, 4-
methylphenyl H-3, H-5), 6.72 (s, 1H, pyrazole H-4),
2.37 (s, 3H, CH3); 13C NMR (75 MHz; CDCl3) d
144.83 (pyrazole C-5), 144.34 (q, JCCF=38.4 Hz, pyra-
zole C-3), 139.32 (benzenesulfonyl azido C-4), 137.82
(4-methylphenyl C-4), 134.14 (benzenesulfonyl azido C-
1), 129.50, 129.21 and 128.65 (benzenesulfonyl azido C-
2, C-6; 4-methylphenyl C-3, C-5 and C-2, C-6), 126.56
(benzenesulfonyl azido C-3, C-5), 126.00 (4-methylphe-
nyl C-1), 121.19 (q, JCF=269.1 Hz, CF3), 105.59 (pyr-
azole C-4), 21.31 (CH3). Anal. calcd for
C17H12F3N5O2S: C, 50.12; H, 2.97; N, 17.19. Found: C,
50.21; H, 2.83; N, 17.28.
Experimental
Melting points were determined using a Buchi capillary
apparatus and are uncorrected. Infrared (IR) spectra
were recorded as films on NaCl plates with a Nicolet
550 Series II Magna FT-IR spectrometer. 1H NMR, 13
C
NMR and 19F NMR spectra were recorded on a Bruker
AM-300 spectrometer, where J (coupling constant)
values are estimated in Hz. 13C NMR spectra were
acquired using the J modulated spin echo technique
where methyl and methine carbons appear as positive
peaks and methylene and quaternary carbon resonances
appear as negative peaks. Elemental Analysis (EA) were
performed for C, H and N (Micro-analytical Service
Laboratory, Department of Chemistry, University of
Alberta, Canada). Silica gel column chromatography was
performed using Merck silica gel 60 ASTM (70–230
mesh). Compounds 3,1,7 5a–e,1 6b,12 and 7f1 were pre-
pared using literature methods. All other reagents were
purchased from the Aldrich Chemical Company (Mil-
waukee, WI, USA) and used without further purification.
General procedure for synthesis of 1-phenyl-5-(4-methyl-
phenyl)-3-trifluoromethyl-1H-pyrazole (7a)
Phenylhydrazine hydrochloride 6a (1.66 g, 11.48 mmol)
was added to a solution of the 4,4,4-trifluoro-1-(4-
methylphenyl)butane-1,3-dione 5a1 (2.60 g, 11.29 mmol)
in 95% ethanol (100 mL) with stirring and the reaction
was alꢀlowed to proceed at reflux for 24 h. After cooling
t o 25C, the solvent was removed in vacuo. Water
(40 mL) was added to the residue that was extracted
with EtOAc (3 Â 50 mL), the combined organic extracts
were dried (Na2SO4) and the solvent was removed in
vacuo to give a brown syrup. Purification of this pro-
ductby silica gel column chromaot graphy using n-hex-
ane–EtOAc (1:1, v/v) as eluent afforded 7a (1.92 g, 55%)
Synthesis of 4-[5-(4-methylphenyl)-3-trifluoromethyl-1H-
pyrazol-1-yl]benzenesulfonyl azide (4). Sodium nitrite
(0.42 g, 4.94 mmol) was added slowly to a solution of
the amine 37 (0.14 g, 0.44 mmol) in concentrated HCl
as a yellow syrup; IR (film) n 3137, 3055 (CHarom) cmÀ1
;
1H NMR (300 MHz; CDCl3) d 7.30–7.38 (m, 5H,
N1-phenyl hydrogens), 7.12 (d, 2H, J=8.5 Hz, 4-methyl-
phenyl H-2, H-6), 7.14 (d, 2H, J=8.5 Hz, 4-methylphe-
nyl H-3, H-5), 6.73(s, 1H, pyrazole H-4), 2.36 (s, 3H,
CH3). Anal. calcd for C17H13F3N2: C, 63.74; H, 4.68; N,
8.74. Found: C, 64.18; H, 4.41; N, 8.44.
ꢀ
(5 mL) at0 C with stirring. This mixture was stirred for
30 min at0–5 ꢀC in the dark and the 4-[5-(4-methylphe-
nyl)-3-trifluoromethyl-1H-pyrazol-1-yl]benzenediazo-
nium chloride produced in situ was added slowly with
stirring to a solution of 37% w/v SO2 in HOAc (3.7 mL,
21.34 mmol), CuCl2 (0.19 g, 2.11 mmol), KCl (0.51 g,
6.84 mmol), benzene (2 mL) and 1,4-dioxane (4 mL) at
5–10 ꢀC. The reaction mixture was stirred for 30 min at
40–50 ꢀC in the dark at which time nitrogen gas evolu-
tion was complete, the mixture was cooled to 25 ꢀC,
water (12 mL) was added, and the mixture was extracted
with benzene (3 Â 15 mL). The combined organic
extracts were washed with 10% w/v aqueous NaOH
(20 mL), the organic extract was dried (Na2SO4), and
Compounds 7b–e were prepared using a similar proce-
dure where 5b–e were used in place of 5a, and 7f was
obtained from the reaction of 5a with 6b according to a
literature method.1 Physical and spectral data for 7b–e
are listed below.
5-(4-Methoxyphenyl)-1-phenyl-3-trifluoromethyl-1H-pyr-
azole (7b). Brown syrup (92%); IR (film) n 3137, 3070
(CHarom
) ; d
cmÀ1 1H NMR (300 MHz; CDCl3)

5278
Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
7.31–7.38 (m, 5H, N1-phenyl hydrogens), 7.15 (d, 2H,
J=9.0 Hz, 4-methoxyphenyl H-2, H-6), 6.85 (d, 2H,
J=9.0 Hz, 4-methoxyphenyl H-3, H-5), 6.70 (s, 1H,
pyrazole H-4), 3.81 (s, 3H, OCH3); 13C NMR (75 MHz;
CDCl3) d 159.40 (4-methoxyphenyl C-4), 144.49 (pyra-
zole C-5), 142.77 (q, JCCF=38.4 Hz, pyrazole C-3),
132.95 (N1-phenyl C-1), 129.23 (4-methoxyphenyl C-1),
128.74, 128.68, 128.54 and 126.77 (4-methoxyphenyl
C-2, C-6, N1-phenyl C-2, C-6; C-3, C-5; C-4), 121.28 (q,
JCF=269.1 Hz, CF3), 114.17 (4-methoxyphenyl C-3,
C-5), 105.08 (pyrazole C-4), 55.48 (OCH3). Anal. calcd
for C17H13F3N2O: C, 64.15; H, 4.12; N, 8.80. Found: C,
64.39; H, 4.01; N, 8.45.
and removal of the solvent in vacuo gave a yellow syrup
which was purified by silica gel column chromatography
using n-hexane–EtOAc (2:1, v/v) as eluent to give 8a
(2.48 g, 61%) as pale yellow needles; mp 65–67 ꢀC; IR
(film) n 3133, 3083 (CHarom), 2125 (N3), 1183, 1356
(SO2) cmÀ1; 1H NMR (300 MHz; CDCl3) d 7.89 (d, 1H,
J=1.53 Hz, benzenesulfonyl azido H-2), 7.28–7.44 (m,
7H, benzenesulfonyl azido H-4, H-5, H-6, 4-methylphe-
nyl H-2, H-3, H-5, H-6), 6.86 (s, 1H, pyrazole H-4), 2.67
(s, 3H, CH3); 13C NMR (75 MHz; CDCl3) d 143.38 (q,
JCCF=38.4 Hz, pyrazole C-3), 142.12 (pyrazole C-5),
139.12, 138.63 (benzenesulfonyl azido C-1 and C-3),
137.55 (4-methylphenyl C-4), 134.11, 133.49 (benzene-
sulfonyl azido C-2 and C-6), 129.45, 129.15 (4-methyl-
phenyl C-2, C-6, C-3, C-5), 127.87 (4-methylphenyl C-
1), 125.68 (benzenesulfonyl azido C-4, C-5), 121.02 (q,
JCF=269.1 Hz, CF3), 105.91 (pyrazole C-4), 20.26
(CH3); dF (282 MHz; CDCl3) 99.40 (s, CF3). Anal. calcd
for C17H12F3N5O2S: C, 50.12; H, 2.97; N, 17.19.
Found: C, 50.34; H, 2.81; N, 16.93.
5-(4-Chlorophenyl)-1-phenyl-3-trifluoromethyl-1H-pyra-
zole (7c). Yellow syrup (66%); IR (film) n 3136, 3044
(CHarom
) ; d
cmÀ1 1H NMR (300 MHz; CDCl3)
7.32–7.40 (m, 3H, N1-phenyl H-3, H-4, H-5), 7.29–7.31
(m, 4H, N1-phenyl H-2, H-6, 4-chlorophenyl H-2, H-6),
7.15 (d, 2H, J=8.5 Hz, 4-chlorophenyl H-3, H-5), 6.75
(s, 1H, pyrazole H-4). Anal. calcd for C16H10ClF3N2
requires C, 58.72; H, 3.20; N, 8.55. Found: C, 58.98; H,
3.39; N, 8.57.
Compounds 8b–f were prepared using a similar proce-
dure where 7b–f were used in place of 7a. The physical
and spectral data for 8b–f are listed below.
5-(4-Fluorophenyl)-1-phenyl-3-trifluoromethyl-1H-pyra-
zole (7d). Brown syrup (58%); IR (film) n 3145, 3065
3-[5-(4-Methoxyphenyl)-3-trifluoromethylpyrazol-1-yl]-
benzenesulfonyl azide (8b). Yellow solid (70%); mp
121–123 ꢀC; IR (film) n 3141, 3067 (CHarom), 2142 (N3),
1
(CHarom) cmÀ1; H NMR (300 MHz; CDCl3) d 7.20–
7.42 (m, 7H, N1-phenyl hydrogens, 4-fluorophenyl H-2,
H-6), 7.07 (d, 2H, JHCCF=8.5 Hz of d, JHCCH=8.5 Hz,
4-fluorophenyl H-3, H-5), 6.76 (s, 1H, pyrazole H-4).
Anal. calcd for C16H10F4N2: C, 62.75; H, 3.29; N, 9.15.
Found: C, 62.72; H, 3.06; N, 9.09.
1
1165, 1355 (SO2) cmÀ1; H NMR (300 MHz; CDCl3) d
7.86 (d, 1H, J=2.1 Hz, benzenesulfonyl azido H-2),
7.29–7.43 (m, 6H, benzenesulfonyl azido H-5, H-6,
4-methoxyphenyl H-2, H-3, H-5, H-6), 7.03 (d, 1H,
J=8.8 Hz, benzenesulfonyl azido H-4), 6.79 (s, 1H,
pyrazole H-4), 4.05 (s, 3H, OCH3); 13C NMR
(75 MHz; CDCl3) d 157.04 (4-methoxyphenyl C-4),
143.19 (q, JCCF=38.5 Hz, pyrazole C-3), 142.00 (pyra-
zole C-5), 138.61 (benzenesulfonyl azido C-1), 136.08
(benzenesulfonyl azido C-2), 130.49 (benzenesulfonyl
azido C-6), 129.39 (4-methoxyphenyl C-2, C-6), 129.00
(benzenesulfonyl azido C-4), 128.03 (benzenesulfonyl
azido C-3), 125.61 (benzenesulfonyl azido C-5), 121.93
(4-methoxyphenyl C-1), 119.1 (q, JCF=269.0 Hz, CF3),
112.61 (4-methoxyphenyl C-3, C-5), 105.71 (pyrazole
C-4), 56.67 (OCH3). Anal. calcd for C17H12F3N5O3S: C,
48.02; H, 2.86; N, 16.46. Found: C, 48.34; H, 2.79; N,
16.06.
5-(4-Bromophenyl)-1-phenyl-3-trifluoromethyl-1H-pyra-
zole (7e). Brown solid (39%); mp 92–94 ꢀC; IR (film) n
3116, 3050 (CHarom) cmÀ1
;
1H NMR (300 MHz;
CDCl3) d 7.46 (d, 2H, J=8.8 Hz, 4-bromophenyl H-3,
H-5), 7.25–7.40 (m, 5H, N1-phenyl hydrogens), 7.09 (d,
2H, J=8.8 Hz, 4-bromophenyl H-2, H-6), 6.76 (s, 1H,
pyrazole H-4). Anal. calcd for C16H10BrF3N2: C,
52.34; H, 2.75; N, 7.63. Found: C, 52.30; H, 2.60; N,
7.22.
General procedure for synthesis of 3-[5-(4-methylphenyl)-
3-trifluoromethylpyrazol-1-yl]benzenesulfonyl azide (8a)
Chlorosulfonic acid (3.3 mL, 49.64 mmol) was added
drop wise to the ice-cold 1-phenyl-5-(4-methylphenyl)-3-
trifluoromethyl-1H-pyrazole 7a (3.0 g, 9.92 mmol) with
vigorous stirring. The cooling bath was removed, the
mixture was stirred for 1 h at 25 ꢀC, and then poured
into crushed ice (50 g) very slowly. Extraction with
EtOAc (3 Â 60 mL), washing the combined organic
extracts with water, and removal of the solvent in vacuo
afforded a brown syrup.12 This intermediate product, 3-
[5-(4-methylphenyl)-3-trifluoromethylpyrazol-1-yl]ben-
zenesulfonyl chloride, was dissolved in acetone (10 mL)
and added drop wise to a stirred solution of NaN3
(1.0 g, 15.38 mmol) in water (10 mL) at À10 ꢀC.11 The
reaction was allowed to proceed for 3 h at 0 ꢀC, the sol-
ventwas removed in vacuo, and waetr (30 mL) was
added to the residue. Extraction with EtOAc (3 Â
40 mL), drying the combined organic extracts (Na2SO4),
3-[5-(4-Chlorophenyl)-3-trifluoromethylpyrazol-1-yl]ben-
zenesulfonyl azide (8c). Yellow syrup (46%); IR (film) n
3110, 3084 (CHarom), 2131 (N3), 1150, 1355 (SO2) cmÀ1
;
1H NMR (300 MHz; CDCl3) d 7.98 (d, 1H, J=2.1 Hz,
benzenesulfonyl azido H-2), 7.16–7.70 (m, 7H, chloro-
phenyl hydrogens, benzenesulfonyl azido H-4, H-5, H-
6), 6.88 (s, 1H, pyrazole H-4); 13C NMR (75 MHz;
CDCl3) d 143.52 (q, JCCF=39.5 Hz, pyrazole C-3),
141.09 (pyrazole C-5), 138.24, 137.08 (benzenesulfonyl
azido C-1, C-3), 133.14 (4-chlorophenyl C-4), 128.73
(4-chlorophenyl C-1), 134.69, 132.55, 131.04, 129.57,
129.35, 125.58 (benzenesulfonyl azido and chlor-
ophenyl Carom–H), 120.84 (q, JCF=269.1 Hz, CF3),
106.26
(pyrazole
C-4).
Anal.
calcd
for
C16H9ClF3N5O2S: C, 44.92; H, 2.12; N, 16.37. Found:
C, 45.02; H, 2.15; N, 16.68.

Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
5279
3-[5-(4-Fluorophenyl)-3-trifluoromethylpyrazol-1-yl]ben-
zenesulfonyl azide (8d). Pale yellow solid (54%); mp
123–125 ꢀC; IR (film) n 3100, 3050 (CHarom), 2125 (N3),
for 1000 iterations reaching a convergence of 0.01 kcal/
mol A. The energy minimized ligands were super-
imposed on SC-558 in the PDB file 1cx2 after which
SC-558 was deleted. The resulting structure was sub-
jected to docking using the Affinity command in the
Docking module of InsightII by defining subsets of the
enzyme such that residues within 10 A of the ligand
were allowed to relax, while the rest of the enzyme resi-
dues were fixed. The ESFF (force field) was employed
for all docking purposes. The optimized ligand-enzyme
complex was further subjected to molecular dynamics
simulation at a constant temperature of 300 K for over
1000 fs with a time step of 1 fs.
1
1380, 1150 (SO2) cmÀ1; H NMR (300 MHz; CDCl3) d
7.91 (d, 1H, J=2.1 Hz, benzenesulfonyl azido H-2),
7.06–7.16 (m, 4H, aryl hydrogens), 7.18–7.32 (m, 3H,
aryl hydrogens), 6.85 (s, 1H, pyrazole H-4); 19F NMR
(282 MHz; CDCl3) (proton decoupled spectrum) d 104.0
(s, 3F, CF3), 50.2 (s, 1F, F-C6H4–). Anal. calcd for
C16H9F4N5O2S: C, 46.72; H, 2.21; N, 17.0. Found: C,
46.65; H, 2.23; N, 17.21.
3-[5-(4-Bromophenyl)-3-trifluoromethylpyrazol-1-yl]ben-
zenesulfonyl azide (8e). Yellow syrup (63%); IR (film) n
3165, 3100 (CHarom), 2140 (N3), 1380, 1180 (SO2) cmÀ1
;
Volume determination
1H NMR (300 MHz; CDCl3) d 7.97 (dt, 1H, J=7.9, 1.8,
1.8 Hz, benzenesulfonyl azido H-6), 7.84 (dd, 1H,
J=1.8, 1.8 Hz, benzenesulfonyl azido H-2), 7.62 (dd,
1H, J=7.9, 7.9 Hz, benzenesulfonyl azido H-5), 7.54
(dt, 1H, J=7.9, 1.8, 1.8 Hz, benzenesulfonyl azido H-4),
7.40 (d, 2H, J=8.5 Hz, 4-bromophenyl H-2, H-6), 7.26
(d, 2H, J=8.5 Hz, 4-bromophenyl H-3, H-5), 6.89 (s,
1H, pyrazole H-4); 13C NMR (75 MHz; CDCl3) d
143.63 (q, JCCF=38.4 Hz, pyrazole C-3), 141.17 (pyra-
zole C-5), 138.92 and 138.31 (benzenesulfonyl azido
C-1, C-3), 129.29 (4-bromophenyl C-1), 121.24 (4-bro-
mophenyl C-4), 136.16, 134.52, 131.39, 129.63, 129.39,
125.65 (Carom–H), 120.86 (q, JCF=268.0 Hz, CF3),
The Alchemy 2000 program14 was used to calculate the
molecular volume (A3) of compounds 4 and 8 after
minimization using PM3.
In vitro cyclooxygenase inhibition studies
The compounds listed in Table 1 were tested for their
ability to inhibit COX-1 and COX-2 using a COX-
(ovine) inhibitor screening kit (catalog no. 560101,
Cayman Chemical, Ann Arbor, MI, USA) using the
method previously reported.22
106.25
(pyrazole
C-4).
Anal.
calcd
for
C16H9BrF3N5O2S: C, 40.69; H, 1.92; N, 14.83. Found:
C, 40.88; H, 1.84; N, 14.64.
Acknowledgements
We are grateful to the Canadian Institutes of Health
Research (MOP-14712) for financial supportof htis
research, to the Alberta Heritage Foundation for Med-
ical Research for a postdoctoral fellowship to one of us
(MJU), and to Rx&D-HRF/CIHR for a graduate
scholarship to one of us (PR).
5-Sulfonamido-2-[5-(4-methylphenyl)-3-trifluoromethyl-
pyrazol-1-yl]benzenesulfonyl azide (8f). White solid
(45%); mp 176–178 ꢀC; IR (film) n 3400, 3300 (NH2),
3150, 3100 (CHarom), 2135 (N3), 1350, 1150 (SO2) cmÀ1
;
1H NMR (300 MHz; DMSO-d6) d 7.97 (1H, s, benze-
nesulfonyl azido H-6), 7.92 (2H, two overlapping d,
J=8.5 Hz, benzenesulfonyl azido H-3, H-4), 7.59–7.62
(4H, m, 4-methylphenyl hydrogens), 7.53 (2H, s, NH2),
7.46 (1H, s, pyrazole H-4), 2.61 (3H, s, CH3); 13C NMR
(75 MHz; DMSO-d6) d 144.25 (pyrazole C-5), 142.85
(benzenesulfonyl azido C-5), 142.15 (q, JCCF=37.3 Hz,
pyrazole C-3), 140.52 (benzenesulfonyl azido C-2),
138.83 (4-methylphenyl C-4), 136.37 (benzenesulfonyl
azido C-1), 135.20 (benzenesulfonyl azido C-6), 133.67
(benzenesulfonyl azido C-4), 129.11 (benzenesulfonyl
azido C-3), 126.92 (4-methylphenyl C-1), 126.82 and
126.09 (4-methylphenyl C-2, C-3, C-5, C-6), 121.10 (q,
JCF=268.0 Hz, CF3), 107.15 (pyrazole C-4), 19.53
(CH3). Anal. calcd for C17H13F3N6O4S2: C, 41.97; H,
2.69; N, 17.28. Found: C, 42.33; H, 2.65; N, 17.06.
References and Notes
1. Penning, T. D.; Talley, J. J.; Bertenshaw, S. R.; Carter, J. S.;
Collins, P. W.; Docter, S.; Graneto, M. J.; Lee, L. F.; Mal-
echa, J. W.; Miyashiro, J. M.; Rogers, R. S.; Rogers, D. J.;
Yu, S. S.; Anderson, G. D.; Burton, E. G.; Cogburn, J. N.;
Gregory, S. A.; Koboldt, C. M.; Perkins, W. E.; Seibert, K.;
Veenhuizen, A. W.; Zhang, Y. Y.; Isakson, P. C. J. Med.
Chem. 1997, 40, 1347.
2. Goldstein, J. L.; Silverstein, F. E.; Agrawal, N. M.; Hub-
bard, R. C.; Kaiser, J.; Maurath, C. I.; Verburg, K. M.; Geis,
G. S. Am. J. Gastroenterol. 2000, 95, 1681.
3. Kurumbail, R. G.; Stevens, A. M.; Gierse, J. K.; McDo-
nald, J. J.; Stegeman, R. A.; Pak, J. Y.; Gildehaus, D.; Miya-
shiro, J. M.; Penning, T. D.; Seibert, K. P.; Isakson, C.;
Stallings, W. C. Nature 1996, 384, 644.
4. Luong, C.; Miller, A.; Barnett, J.; Chow, J.; Ramesha, C.;
Browner, M. F. Nat. Struct. Biol. 1996, 3, 927.
5. Llorens, O.; Perez, J. L.; Palomer, A.; Mauleon, D. Bioorg.
Med. Chem. Lett. 1999, 9, 2779.
Molecular modeling (docking) studies
Docking studies were performed using Insight II soft-
ware (Version 97) running on a Silicon Graphics Indigo
2 workstation.21 The coordinates for the X-ray crystal
structure of the selective COX-2 inhibitor SC-558
bound to the murine COX-2 enzyme was obtained from
the RCSB Protein Data Bank (1cx2) and hydrogens
were added. The ligand molecules were constructed
using the Builder module and were energy minimized
6. Garavito, R. M.; DeWitt, D. L. Biochim. Biophys. Acta
1999, 1441, 278.
7. Habeeb, A. G.; Rao, P. N. P.; Knaus, E. E. J. Med. Chem.
2001, 44, 3039.
8. Habeeb, A. G.; Rao, P. N. P.; Knaus, E. E. J. Med. Chem.
2001, 44, 2921.

5280
Md. Jashim Uddin et al. / Bioorg. Med. Chem. 11 (2003) 5273–5280
9. Rahim, M. A.; Rao, P. N. P.; Knaus, E. E. Bioorg. Med.
Chem. Lett. 2002, 12, 2753.
10. Gilbert, E. E. Synthesis 1969, 3.
11. Leffler, J. E.; Tsuno, Y. J. Org. Chem. 1962, 28, 902.
12. Soliman, R. J. Med. Chem. 1979, 22, 321.
13. Hashimoto, H.; Imamura, K.; Haruta, J.-I.; Wakitani, K.
J. Med. Chem. 2002, 45, 1511.
14. Alchemy 2000 (program), Version 2.0; Tripos Inc.: St.
Louis, MO, USA.
15. Lieber, E.; Curtice, J.; Rao, C. Chem. Ind. 1966, 586.
16. Pauling, L. In The Nature of the Chemical Bond, 3rd ed.;
Cornell University Press: Ithaca, NY, 1960.
17. Livingston, R.; Rao, R. R. J. Phys. Chem. 1960, 64, 756.
18. Greig, G. M.; Francis, D. A.; Falgueyret, J. P.; Ouellet,
M.; Percival, M. D.; Roy, P.; Bayly, C.; Mancini, J. A.;
O’Neill, G. P. Mol. Pharmacol. 1997, 52, 829.
19. Selinsky, B. S.; Gupta, K.; Sharkey, C. T.; Loll, P. J. Bio-
chemistry 2001, 40, 5172.
20. Price, M. L. P.; Jorgensen, W. L. J. Am. Chem. Soc. 2000,
122, 9455.
21. Insight II, Version 97 and Discover, Version 2.98; Molec-
ular Simulations, Inc: San Diego, CA, USA, 1997.
22. Habeeb, A. G.; Rao, P. N. P.; Knaus, E. E. Drug Dev.
Res. 2000, 51, 273.