Synthesis and biological evaluation of 3,4-diphenyl-1,2,5-oxadiazole-2- oxides and 3,4-diphenyl-1,2,5-oxadiazoles as potential hybrid COX-2 inhibitor/nitric oxide donor agents

Article abstract of DOI:10.1016/j.bmc.2005.02.034

A group of 3,4-diphenyl-1,2,5-oxadiazole-2-oxides (3,4-diphenylfuroxans) and the corresponding N-desoxy 3,4-diphenyl-1,2,5-oxadiazoles (3,4-diphenylfurazans) analogs, were synthesized for in vitro evaluation as hybrid cyclooxygenase (COX) inhibitor/nitric

Full text of DOI:10.1016/j.bmc.2005.02.034

Bioorganic & Medicinal Chemistry 13 (2005) 2749–2757
Synthesis and biological evaluation of 3,4-diphenyl-
1,2,5-oxadiazole-2-oxides and 3,4-diphenyl-1,2,5-oxadiazoles
as potential hybrid COX-2 inhibitor/nitric oxide donor agents
a
Carlos Velazquez, P. N. Praveen Rao, Robert McDonald and Edward E. Knaus
a
b
a,
*
´
aFaculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2N8
bDepartment of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2
Received 25 January 2005; revised 16 February 2005; accepted 16 February 2005
Abstract—A group of 3,4-diphenyl-1,2,5-oxadiazole-2-oxides (3,4-diphenylfuroxans) and the corresponding N-desoxy 3,4-diphenyl-
1,2,5-oxadiazoles (3,4-diphenylfurazans) analogs, were synthesized for in vitro evaluation as hybrid cyclooxygenase (COX) inhibi-
tor/nitric oxide donor agents. Reaction of 1-[4-(methylsulfonyl)phenyl]-2-phenylethene with an aqueous sodium nitrite solution in
acetic acid afforded a mixture (3:1 ratio) of the inseparable 4-[4-(methylsulfonyl)phenyl]-3-phenyl-1,2,5-oxadiazole-2-oxide (13a) and
3-[4-(methylsulfonyl)phenyl]-4-phenyl-1,2,5-oxadiazole-2-oxide (13b) regioisomers. A group of related regioisomers possessing
either a p-aminosulfonylphenyl (16) or a p-azidosulfonylphenyl (17), moiety were obtained by chlorosulfonation of the unsubstituted
3,4-diphenylfuroxan (10) and subsequent reaction with either ammonium hydroxide or sodium azide, respectively. The metha-
nesulfonyl regioisomers 13a,b [COX-1 IC50 = 11.6 lM; COX-2 IC50 = 0.12 lM; COX-2 selectivity index (SI) = 97] and aminosulfo-
nyl regioisomers 16 (COX-1 IC50 = 9.8 lM; COX-2 IC50 = 0.78 lM; COX-2 SI = 12), like the reference drug celecoxib (COX-1
IC50 = 33.1 lM; COX-2 IC50 = 0.07 lM; COX-2 SI = 472), were potent in vitro COX-2 inhibitors with a good COX-2 selectivity
index. Release of nitric oxide (NO) from the 3,4-diphenylfuroxan compounds (10, 13a,b, 16, 17) was thiol-dependent since the %
NO released was higher upon incubation in the presence of L-cysteine (0.57–3.18%) compared to that in phosphate buffer solution
at pH 7.4 (0.06–0.15%). Molecular modeling (docking) studies show that the methanesulfonyl (MeSO2) COX-2 pharmacophore
present in regioisomers 13a,b is positioned in the vicinity of the COX-2 secondary pocket. The in vitro NO release data, COX-1/
COX-2 inhibition and COX-2 SI structure–activity relationships acquired, and molecular modeling docking studies suggest that
the 1,2,5-oxadiazole-2-oxide (furoxan) ring possesses beneficial features that should be present in a suitable central ring template
(bioisostere) pertinent to the design novel hybrid COX-2 inhibitor/nitric oxide donor agents with a low ulcerogenicity profile that
may be free from adverse cardiovascular effects.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
a lower incidence of associated gastrointestinal (GI) side
effects than NSAIDs.4 This hypothesis served as the ba-
sis for design of the highly selective tricyclic COX-2
inhibitors celecoxib (1) and rofecoxib (2), which have a
central heterocyclic ring scaffold (see Fig. 1). A diverse
group of structurally related tricyclic compounds that
possess vicinal diaryl substituents on a central five- or
six-membered ring template such as pyrazole (1),5 2-
(5H)furanone (2),6 cyclopentene (3),7 isoxazole (4 and
5),8,9 pyridine (6),10 and pyranone (7)11 have been inves-
tigated as selective COX-2 inhibitors.
The nonselective inhibition of both cyclooxygenase iso-
zymes (COX-1 and COX-2) by traditional ulcerogenic
nonsteroidal antiinflammatory drugs (NSAIDs) clearly
illustrated the clinical need for a new generation of non-
ulcerogenic selective COX-2 inhibitors.1–3 In this regard,
the differential tissue distribution of COX-1 and COX-2
provided a rationale for the design of selective COX-2
inhibitors as antiinflammatory and analgesic agents with
Despite the relatively safe pharmacological profile of
selective COX-2 inhibitors, there is now increasing con-
cern regarding their use in patients at risk for an adverse
cardiovascular event such as myocardial infarction. This
increased risk is thought to be triggered by a reduction
Keywords: 3,4-Diphenyl-1,2,5-oxadiazole-2-oxides; 3,4-Diphenyl-1,2,5-
oxadiazoles; Methanesulfonyl; Sulfonamide; Sulfonyl azide; Nitric
oxide donors; COX-2 inhibitors.
*
Corresponding author. Tel.: +1 780 492 5993; fax: +1 780 492
1217; e-mail: eknaus@pharmacy.ualberta.ca
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.02.034

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2750
associated with the use of selective COX-2 inhibitors,
SO2NH2
SO2CH3
N
F3C
and which elicit a decreased ulcerogencity relative to
that frequently observed on long-term use of traditional
NSAIDs.14 It has been reported that an increased gener-
ation of endothelial NO, or release of NO from a nitric
oxide donor drug, is expected to produce beneficial ef-
fects such as a reduction in blood pressure and preven-
tion of atherosclerosis.15 At nanomolar concentrations,
NO reversibly activates soluble guanylate cyclase by
400-fold, catalyzing the conversion of guanosine tri-
phosphate to cyclic guanosine monophosphate
(cGMP).16 Elevation of cGMP relaxes smooth muscle
in blood vessels, inhibits platelet aggregation and adhe-
sion, and blocks the adhesion of white cells to blood ves-
sel walls.17 In addition to these cardiovascular effects,
NO is now recognized as a critical mediator of gastroin-
testinal mucosal defense, exerting many of the same ac-
tions as prostaglandins in the gastrointestinal
tract.14,18,19 Drugs classified as NO–NSAIDs suppress
COX-2 derived prostaglandin synthesis as effectively as
the parent drugs, and they have been found to exert
comparable antiinflammatory and antipyretic activity
to the parent NSAID drug.20,21
N
O
O
CH3
Rofecoxib (2)
Celocoxib (1)
CH3
SO2CH3
SO2NH2
O
N
F
Valdecoxib (4)
(3)
SO2CH3
CH3
-
SO2NCO2Et
Cl
O
Na+
N
N
N
CH3
1,2,5-Oxadiazole-2-oxides (furoxans) represent one class
of heterocyclic compounds that are thiol-dependent NO
donor agents. In this context, furoxans are considered to
be NO releasing prodrugs whose biological activity is
produced by action on the sGC–cGMP pathway. Com-
pared to other NO donor agents, furoxans possess
rather favorable pharmacological properties, since they
frequently release NO slowly resulting in a longer dura-
tion of action. The absence of tolerance is an important
distinctive feature of the furoxan moiety. A mechanism
has been proposed by Granik and Grigor for the thiol-
induced release of NO from furoxans that involves at-
tack by a thiolate anion at C-3 and/or C-4 of the furo-
xan ring followed by ring opening and the subsequent
release of NO.22 As part of our ongoing research pro-
gram, we now report the synthesis, in vitro COX-1/
COX-2 inhibitory activities and nitric oxide release data,
for a group of 3,4-diphenyl-1,2,5-oxadiazole-2-oxides
(10, 13, 16, 17, 3,4-diphenylfuroxans) and the corre-
sponding desoxy 3,4-diphenyl-1,2,5-oxadiazole (11, 14,
19, 20, 3,4-diphenylfurazans) analogs, possessing a C-4
H, SO2CH3, SO2NH2, or SO2N3 moiety on one of the
phenyl groups, as potential hybrid COX-2 inhibitor/
NO donor drugs.
Parecoxib (5)
Etoricoxib (6)
SO2CH3
O
O
S
(7)
Figure 1. Some representative examples of selective tricyclic COX-2
inhibitors possessing a central five- or six-membered ring scaffold.
in the level of the desirable platelet aggregation inhibitor
and vasodilatory prostacyclin (PGI2) in conjunction
with an increased level of the undesirable potent platelet
activator and aggregator thromboxane A2 (TxA2).12
This biochemical explanation constitutes the rationale
for the recent voluntary worldwide withdrawal of
Vioxxꢁ (rofecoxib). This decision was based on data,
from a three-year clinical trial evaluating the ability of
rofecoxib to reduce colon polyp recurrence, that re-
vealed a Ôdiscernible and confirmedÕ higher risk of car-
diovascular events such as stroke and heart attack in
patients taking the drug for a period longer than 18
months.13 This latter clinical evidence presents a new
challenge in the design of selective COX-2 inhibitors
since it is now necessary to also consider the physiolog-
ical consequences associated with alterations in the ara-
chidonic acid cascade due to unbalanced inhibition of
the COX-1 and COX-2 isozymes.
2. Chemistry
Reaction of (E)-1,2-diphenylethene (trans-stilbene, 9)
with a saturated aqueous solution of sodium nitrite in
a mixture of acetic acid and 1,4-dioxane, a modification
of the procedure reported by Gasco and co-workers,23,24
afforded 3,4-diphenyl-1,2,5-oxadiazole-2-oxide (10, 26%
yield) as illustrated in Scheme 1. A similar reaction
using (E)-1-[4-(methylsulfonyl)phenyl]-2-phenylethene (12)
yielded a mixture of the two furoxan regioisomers 13a
and 13b. This product (13) was originally believed to
be a single regioisomer since a single spot was observed
on micro TLC plates irrespective of the polarity of the
Hybrid molecules, comprised of NSAID and nitric oxide
(NO) donor moieties, constitute one of the more prom-
ising approaches for the design of drugs, which are de-
void of the potential adverse cardiovascular effects

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2751
ii
i
H
H
N
N
N
N
O
O
O
11
10
9
H3CO2S
H3CO2S
H3CO2S
i
H
+
H
N
N
N
N
O
O
O
O
12
13a
13b
ii
H3CO2S
N
O
N
14
Scheme 1. Reagents and conditions: (i) NaNO2, CH3CO2H, 1,4-dioxane, 50–60 ꢂC, 6–24 h; (ii) (EtO)3P, reflux, 19–24 h.
development solvent. Subsequent purification by silica
gel column chromatography and then recrystallization
from hexanes provided a product that showed a single
15 with NH4OH afforded the sulfonamide product (16,
presumed to be a mixture of two regioisomers, 69%).
Alternatively, reaction of 15 with the nucleophile
NaN3 yielded the respective sulfonylazide product (17,
76%) that is assumed to exist as a mixture of two regio-
isomers. Similar reactions of the sulfonyl chloride 18
with NH4OH furnished 3-[4-(aminosulfonyl)phenyl]-4-
phenyl-1,2,5-oxadiazole (19, 82%), or with NaN3 yielded
3-[4-(azidosulfonyl)phenyl]-4-phenyl-1,2,5-oxadiazole
(20, 86%).
1
set of resonances in both the H NMR and 13C NMR
spectra. In order to determine which regioisomer of 13
was isolated, a X-ray crystallographic structure was
determined, which indicated that this product was a
mixture of the two regioisomers 4-[4-(methylsulfonyl)-
phenyl]-3-phenyl-1,2,5-oxadiazole-2-oxide (13a) and
3-[4-(methylsulfonyl)phenyl]-4-phenyl-1,2,5-oxadiazole-2-
oxide (13b) that were present in a 3:1 ratio (see Fig. 2).
Reaction of furoxan 10, or the regioisomers 13a,b, with
triethylphosphite at reflux for 19–24 h afforded the
respective deoxygenated product 3,4-diphenyl-1,2,5-
oxadiazole (11, 70%), or 3-[4-(methylsulfonyl)phenyl]-
4-phenyl-1,2,5-oxadiazole (14, 84%).
3. Results and discussion
In vitro COX-1 and COX-2 enzyme inhibition data (Ta-
ble 1) showed that replacement of the 2-(5H)furanone
central ring present in rofecoxib (1) by a 1,2,5-oxadiaz-
ole-2-oxide ring maintains COX-2 inhibitory activity.
Chlorosulfonation of the unsymmetrical furoxan 10
with chlorosulfonic acid at 25 ꢂC for 17 h afforded the
sulfonyl chloride 15 that was purified by silica gel
column chromatography to provide a 30% isolated
Compounds 10 (COX-1 IC50 = 12.0 lM; COX-2 IC50
>
100 lM) and 11 (COX-1 = 0.15 lM; COX-2 IC50 > 100
lM), having unsubstituted C-3 and C-4 phenyl substi-
tuents, are selective COX-1 inhibitors. However, incor-
poration of a para-SO2Me phenyl substituent (COX-2
pharmacophore) provided a mixture of the two regio-
isomers 13a and 13b [COX-1 IC50 = 11.6 lM; COX-2
IC50 = 0.12 lM; COX-2 selectivity index (SI) = 97],
and 14 (COX-1 IC50 = 1.6 lM; COX-2 IC50 = 0.74 lM;
COX-2 SI = 2), that possess a selectivity for the COX-2
isozyme. Compounds 16 and 19, possessing a para-
SO2NH2 phenyl substituent (COX-2 pharmacophore),
like 13 and 14, were also selective COX-2 inhibitors
where the respective COX-2 selectivity indexes were 12
and 5, respectively. In an earlier investigation,25 we
1
yield. Although the H NMR spectrum for the sulfonyl
chloride product 15 showed a single set of resonances
similar to 13a,b, it may also exist as a mixture of
the regioisomers 4-[4-(chlorosulfonyl)phenyl]-3-phenyl-
1,2,5-oxadiazole-2-oxide and 3-[4-(chlorosulfonyl)phen-
yl]-4-phenyl-1,2,5-oxadiazole-2-oxide as shown in
Scheme 2. A similar chlorosulfonation of the symmetri-
cal furazan 11 furnished 3-[4-(chlorosulfonyl)phenyl]-4-
phenyl-1,2,5-oxadiazole (18), which unlike 15 cannot
exist as a mixture of two regioisomers, in 16% isolated
yield after purification by silica gel column chromato-
graphy. The subsequent reaction of the sulfonyl chloride

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2752
H2NO2S
ii
ClO2S
16
N
N
O
O
i
N3O2S
N
N
N
N
O
O
O
iii
O
15
10
17
N
N
O
O
H2NO2S
ii
SO2Cl
19
i
N
N
O
N3O2S
N
N
N
N
O
O
iii
11
18
N
N
20
O
Scheme 2. Reagents and conditions: (i) ClSO3H, 25 ꢂC, 17 h; (ii) NH4OH, THF, 25 ꢂC, 1.5 h; (iii) NaN3, H2O, THF, 25 ꢂC, 20 h.
not possess a N-oxido substituent, shows that N-deoxy-
genation provided a small increase in COX-1 potency,
and a modest decrease in COX-2 potency, which re-
sulted in a lower COX-2 inhibitory selectivity index.
Table 1. In vitro COX inhibition data for 3,4-diphenylfuroxans (10,
13a,b, 16, and 17) and 3,4-diphenylfurazans (11, 14, 19, and 20)
a
a
Compound
COX-1 IC50
(lM)
COX-2 IC50
(lM)
COX-2 SIb
10
11
12.0
0.15
11.6
1.6
>100
>100
0.12
0.74
0.78
>100
0.91
>100
0.50
0.07
—
—
97
2
The percent nitric oxide (NO) released from the 3,4-di-
phenyl-1,2,5-oxadiazole-2-oxides (furoxans 10, 13a–b,
16, and 17) upon in vitro incubation with L-cysteine,
for 1.5 h at 37 ꢂC, was determined and the results are
summarized in Table 2. It has been reported that a re-
duced thiol such as L-cysteine, L-cysteamine, or glutathi-
one is required for the release of NO from certain NO
donor agents such as those containing a furoxan
(1,2,5-oxadiazole-2-oxide) moiety.26 The % NO release
data acquired in this study (see Table 2) is consistent
with this literature precedent since the % NO released
from the 3,4-diphenylfuroxan compounds (10, 13a,b,
16, and 17) was higher upon incubation in the presence
of L-cysteine (0.57–3.18%) compared to that determined
in phosphate buffer solution at pH 7.4 (0.06–0.15%). In
comparison, the reference drug glycerine trinitrate re-
leased 5.86% NO upon incubation in the presence of
L-cysteine, and 0.72% NO in phosphate buffer at
pH 7.4, per nitrooxy group. Antiinflammatory agents
releasing NO that reversibly activates soluble guanylate
cyclase which in turn catalyzes the conversion of guano-
sine triphosphate to cyclic guanosine monophosphate
(cGMP)16 would be expected to relax smooth muscle
in blood vessels, inhibit platelet aggregation and adhe-
sion, and block the adhesion of white cells to blood ves-
sel walls.17 Accordingly, hybrid COX-2 inhibitor/NO
donor agents should not cause adverse cardiovascular
13a,b
14
16
17
9.8
>100
4.6
12
—
5
19
20
>100
>100
33.1
—
>200
472
Rofecoxib
Celecoxib
a Values are means of two determinations acquired using an ovine
COX-1/COX-2 assay kit (catalog no. 560101, Cayman Chemicals
Inc., Ann Arbor, MI, USA) and the deviation from the mean is
<10% of the mean value.
b In vitro COX-2 selectivity index (COX-1/COX-2 IC50).
showed that 2-(5H)furanone derivatives possessing a
para-SO2N3 phenyl substituent exhibited significant
COX-2 inhibitory activity. In contrast, in this study
replacement of the SO2Me substituent present in com-
pounds 13 and 14, or the SO2NH2 substituent present
in compounds 16 and 19, by a SO2N3 phenyl substituent
provided compounds 17 and 20, which were completely
devoid of both COX-1 and COX-2 inhibitory activity
(COX-1 and COX-2 IC50 > 100 lM). A comparison of
the furoxan regioisomers (13a,b) having a N-oxido sub-
stituent, with the corresponding furazan (14) that does

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2753
Table 2. Nitric oxide release from 3,4-diphenylfuroxans possessing a
C-4 H (10), SO2Me (13a,b), SO2NH2 (16), or SO2N3 (17) substituent
Compound
% NO releasea
L-Cysteinec
PBSb
10
13a,b
0.15
0.06
0.10
0.08
0.72
0.57
1.48
0.84
3.18
5.86
16
17
Glycerine trinitrated
a Percent of nitric oxide released (mean value, n = 3) relative to a
theoretical maximum release of 1 mol of NO/mol of test compound
was determined using the Griess reaction. Variation from the mean %
value was 60.02%.
b Incubated in phosphate buffer solution (PBS, pH 7.4) at 37 ꢂC for
1.5 h.
c Incubated in the presence of 5 mM L-cysteine in phosphate buffer
solution (pH 7.4) at 37 ꢂC for 1.5 h.
d The percent nitric oxide released was estimated as the % NO pro-
duced/nitrooxy (ONO2) group present in the reference drug glycerine
trinitrate [O2NCH2CH(ONO2)CH2ONO2].
events such as an increased incidence of heart attacks
and strokes.
The binding interactions of the 3,4-diphenyl-1,2,5-
oxadiazole-2-oxide (3,4-diphenylfuroxan) regioisomers
(13a,b, COX-2 IC50 = 0.12 lM; COX-2 SI = 96) within
the COX-2 binding site were examined by molecular
modeling (docking) experiments (Figs. 3 and 4). The
most stable conformer of the regioisomer 13a orients
within the COX-2 primary binding site such that the C-4
para-methanesulfonylphenyl substituent is positioned
in the vicinity of the COX-2 secondary pocket where it
Figure 3. Docking of 4-[4-(methylsulfonyl)phenyl]-3-phenyl-1,2,5-oxa-
diazole-2-oxide (13a) (ball-and-stick) in the active site of murine COX-
2. Hydrogen atoms of the amino acid residues have been removed to
improve clarity.
is surrounded by amino acid residues Phe518, Arg513
,
Gln192, Val523, and Leu352 (Fig. 3). One of the O-atoms
of the SO2Me substituent forms a favorable hydrogen
toward the top of the COX-2 binding site closer to
Trp387 and Tyr385. The N5-atom of the central furoxan
ring participates in a hydrogen bonding interaction with
bond with the backbone NH of Ile517 (distance = 2.22 A)
˚
and a weak hydrogen bond with the NH2 of Gln192 (dis-
the OH of Tyr355 (distance = 3.16 A) and it is positioned
˚
˚
tance = 3.62 A). The distance between the second O-
120
˚
about 4.25 A away from the NH2 of Arg
that is lo-
atom of SO2Me and the NH2 of Arg513 was about
cated near the mouth of the COX-2 binding site. Accord-
ingly, the O-atom of the central furoxan ring undergoes a
weak hydrogen bonding interaction with the OH
˚
7.58 A. The unsubstituted C-3 phenyl ring was oriented
of Tyr355 (distance = 3.63 A) and it is located about
˚
5.4 A from the NH2 of Arg120. It is interesting to note
˚
that the N-oxide moiety of the central furoxan ring is ori-
ented close to Ser530, the acetylation site of aspirin. The
distance between negatively charged N2-oxido O-atom
and the OH of Ser530 was about 2.5 A. This observation is
˚
consistent with previous studies where the C@O of the
central furanone ring of rofecoxib undergoes a favorable
hydrogen bonding interaction with the OH of Ser530
.
27
A similar molecular modeling experiment where the reg-
ioisomer 13b was docked in the COX-2 binding site
shows, like that observed for 13a, that the C-3 phenyl
ring possessing the p-SO2Me COX-2 pharmacophore is
also oriented in the vicinity of the COX-2 secondary
pocket (Phe518, Arg513, Gln192, Val523, Ser353, and
Leu352) as shown in Figure 4. One of the O-atoms of
the –SO2Me substituent is hydrogen bonding with the
Figure 2. X-ray crystal structure of a mixture (3:1 ratio) of the 4-[4-
(methylsulfonyl)phenyl]-3-phenyl-1,2,5-oxadiazole-2-oxide (13a), and
3-[4-(methylsulfonyl)phenyl]-3-phenyl-1,2,5-oxadiazole-2-oxide (13b),
regioisomers. Atom labels ending in ÔAÕ (e.g., O1A, N1A, C1A) belong
to 13a, while those ending in ÔBÕ (O1B, N1B, C1B) belong to 13b (the
MeSO2 group is common to both).
backbone NH of Ile517 (distance = 2.28 A) and the
˚
NH2 of Gln192 (distance = 2.08 A). The C-4 unsubsti-
˚
tuted phenyl ring present in 13b, similar to that observed

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2754
1,2,5-oxadiazole-2-oxides (3,4-diphenylfuroxans) and
3,4-diphenyl-1,2,5-oxadiazoles (3,4-diphenylfurazans)
investigated indicate that a methylsulfonyl (SO2Me),
or aminosulfonyl (SO2NH2), COX-2 pharmacophore
located at the para-position of either a C-3 or C-4 phen-
yl ring is essential for COX-2 inhibitory activity. In con-
trast, compounds having an unsubstituted C-3 or C-4
phenyl substituent (10 and 11) were selective COX-1
inhibitors. The sulfonylazido (SO2N3) substituent is
not a suitable COX pharmacophore since compounds
possessing this substituent were devoid of both COX-1
and COX-2 inhibitory activity. A central ring N-oxido
substituent is not a major determinant of COX inhibi-
tory potency and/or selectivity. In this regard, the pres-
ence of a N-oxido moiety provided small increases in
COX-2 potency and selectivity. Alternatively, removal
of the N-oxido oxygen atom, which gives rise to the
N-deoxy derivatives, results in a small increase in
COX-1 potency and a small decrease in COX-2 potency.
The thiol-dependent release of NO, in conjunction with
relatively potent and selective COX-2 inhibitory activity,
suggest that the 1,2,5-oxadiazole-2-oxide (furoxan) ring
system possesses beneficial features that would be desir-
able for the design of hybrid COX-2 inhibitor/NO donor
antiarthritic agents with a low ulcerogencity profile and
minimal potential to induce adverse cardiovascular
events such as heart attacks and strokes.13
Figure 4. Docking of 3-[4-(methylsulfonyl)phenyl]-4-phenyl-1,2,5-oxa-
diazole-2-oxide (13b) (ball-and-stick) in the active site of murine COX-
2. Hydrogen atoms of the amino acid residues have been removed to
improve clarity.
5. Experimental
Melting points were determined using a Thomas–Hoo-
ver capillary apparatus and are uncorrected. Infrared
(IR) spectra were recorded using a Nicolet 550 Series
II Magna FT-IR spectrometer. Nuclear magnetic reso-
nance spectra (1H) were recorded on a Bruker AM-300
spectrophotometer. The assignment of exchangeable
protons (NH) was confirmed by the addition of D2O.
Ultraviolet (UV) spectra and quantitative analyses were
measured using a Philips PU 8740 UV/vis scanning spec-
trophotometer. Silica gel column chromatography was
performed using Silicycleꢁ (silica gel 70–230 mesh). X-
ray crystallographic (crystallography laboratory) and
elemental analyses (microanalytical service laboratory),
were performed in the Department of Chemistry, Uni-
versity of Alberta. Molecular modeling experiments
were performed on a Sillicon Graphics Workstation
(Octane 2) equipped with a R14000A microprocessor
using the Insight II software version 2000.1 (Accelerys
Inc.). 1-[4-(Methylsulfonyl)phenyl]-2-phenylethene (12)
was prepared according to a reported procedure.28 All
other reagents were purchased from Aldrich Chemical
(Milwaukee, WI) and used without further purification.
for regioisomer 13a, is also oriented toward a hydropho-
bic area comprised of Trp387 and Tyr385 at the top of the
COX-2 binding site. The major difference between the
binding modes observed for the two regioisomers 13a
and 13b within the COX-2 binding site is the orientation
of the N-oxido moiety of the central furoxan ring. In the
case of regioisomer 13b, the N-oxido moiety is oriented
toward the mouth of the COX-2 binding site close to
Tyr355 and Arg120. In contrast, the N-oxido moiety
present in 13a was oriented in a direction close to
Ser530. The distance between the negatively charged
O-atom of the N-oxido moiety in 13b and the OH of
Tyr355 is about 3.3 A, whereas the distance between
˚
the charged guanidino side chain of Arg120 and
˚
the O-atom of the N-oxido moiety is about 6.0 A.
Molecular dynamics (MD) simulations on the stabilities
of the enzyme–ligand complexes revealed that 13a
(Eintermolecular = À52.30 kcal/mol) has a slightly higher
binding affinity for the COX-2 isozyme as compared
to 13b (Eintermolecular = À52.10 kcal/mol).
5.1. 3,4-Diphenyl-1,2,5-oxadiazole-2-oxide (10)
4. Conclusions
trans-Stilbene (9, 2 g, 11.1 mmol) was dissolved in a
warm mixture (50–55 ꢂC) of glacial acetic acid (6 mL)
and 1,4-dioxane (10 mL). Aqueous sodium nitrite
(5.34 g, 77.6 mmol in 10 mL of water) was added drop-
wise during a period of 1 h, the reaction was allowed to
proceed with stirring at 50–60 ꢂC for 6.5 h, and the reac-
tion was quenched by addition of ice-water (100 mL of a
Replacement of the 2-(5H)furanone central ring present
in rofecoxib by an isosteric 1,2,5-oxadiazole-2-oxide ring
maintains COX-2 inhibitory activity, although there is a
decrease in COX-2 selectivity, relative to the refer-
ence compounds celecoxib (1) and rofecoxib (2).
Structure–activity data acquired for the 3,4-diphenyl-

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2755
50:50 v/v mixture) at 25 ꢂC. This mixture was extracted
with EtOAc (3 · 30 mL), the organic phase was dried
(Na2SO4), and the solvent was removed in vacuo. The
residue obtained was purified by silica gel column chro-
matography (hexanes–ether, 3:1 v/v) prior to recrystalli-
zation from hexanes to yield 10 (0.63 g, 26%) as pale
yellow crystals; mp 104–105 ꢂC; IR (KBr): 3059 (CH),
(10 mL) was heated at reflux for 24 h with stirring.
The reaction mixture was cooled to 25 ꢂC, the reaction
was allowed to proceed at 25 ꢂC for an additional 1 h
with stirring, and the reaction was quenched by addition
of H2SO4 (10 mL of 2 N) with subsequent stirring for
20 min. This mixture was extracted with CH2Cl2
(3 · 30 mL), the combined CH2Cl2 extracts were washed
with water (6 · 30 mL), the organic fraction was dried
(Na2SO4), and the solvent was removed in vacuo. The
residue obtained was purified by silica gel column chro-
matography (hexanes–ether, 3:1 v/v) to yield 14 (0.51 g,
84%) as a white powder; mp 135–137 ꢂC; IR (KBr): 3012
1588 (NO), 1420 (NO) cmÀ1 1H NMR (CDCl3): d
;
7.43–7.55 (m, 10H, phenyl hydrogens). Anal. Calcd for
C14H10N2O2: C, 70.58; H, 4.23; N, 11.76. Found: C,
70.33; H, 4.19; N, 11.61.
5.2. 4-[4-(Methylsulfonyl)phenyl]-3-phenyl-1,2,5-oxadiaz-
ole-2-oxide (13a) and 3-[4-(methylsulfonyl)phenyl]-4-phen-
yl-1,2,5-oxadiazole-2-oxide (13b) regioisomers (ratio 3:1)
(CHarom), 2925 (CHaliph), 1595 (NO), 1313 (SO) cmÀ1
;
1H NMR (CDCl3): d 8.03 (d, Jortho = 8.4 Hz, 2H, 4-
methylsulfonylphenyl H-3, H-5), 7.78 (d, Jortho = 8.4 Hz,
2H, 4-methylsulfonylphenyl H-2, H-6), 7.56–7.48 (m,
5H, phenyl hydrogens), 3.12 (s, 3H, CH3). Anal. Calcd
for C15H12N2O3S: C, 59.99; H, 4.03; N, 9.33. Found:
C, 59.71; H, 4.01; N, 9.09.
1-[4-(Methylsulfonyl)phenyl]-2-phenylethene (12, 1.12 g,
4.3 mmol) was dissolved in a warm mixture (55–60 ꢂC)
of glacial acetic acid (5 mL) and 1,4-dioxane (50 mL).
Aqueous sodium nitrite (2.2 g, 31.9 mmol in 4 mL of
water) was added dropwise during a period of 1 h, the
reaction was allowed to proceed with stirring at 60 ꢂC
for 17 h, and the reaction was quenched by addition of
ice-water (100 mL of a 50:50 v/v mixture) at 25 ꢂC. This
mixture was extracted with CH2Cl2 (3 · 40 mL), the or-
ganic extract was dried (Na2SO4), and the solvent was
removed in vacuo. The residue obtained was purified
by silica gel column chromatography (hexanes–
CHCl3–EtOAc, 50:35:15 v/v/v) to yield a mixture of
the two regioisomers 13a and 13b that were subse-
quently shown by X-ray crystallography to be present
in a ratio of 3:1 (0.65 g, 47%) as a white powder; mp
121–123 ꢂC; IR (KBr): 3059 (CHarom), 2918 (CHaliph),
5.5. 4-[4-(Chlorosulfonyl)phenyl]-3-phenyl-1,2,5-oxadiaz-
ole-2-oxide and 3-[4-(chlorosulfonyl)phenyl]-4-phenyl-
1,2,5-oxadiazole-2-oxide (15) regioisomers
3,4-Diphenyl-1,2,5-oxadiazole-2-oxide (10, 0.5 g, 2.1 mmol)
was added slowly to a solution of chlorosulfonic acid
(5 mL) at ice-bath temperature, the reaction mixture
was stirred at this temperature for 10 min, the ice bath
was removed, and the reaction was allowed to proceed
at 25 ꢂC for 17 h. The brown reaction mixture was
poured dropwise with caution onto crushed ice (100 g).
Extraction with EtOAc (3 · 30 mL), repeated washing
of the EtOAc extract with water until the water wash
achieved a neutral pH of 7, drying the EtOAc fraction
(Na2SO4), and the solvent was removed in vacuo. The
residue obtained was purified by silica gel column chro-
matography (hexanes–ether, 1:1 v/v) to afford 15 (0.21 g,
1
1595 (NO), 1313 (SO) cmÀ1; H NMR (CDCl3): d 7.96
(d, Jortho = 8.4 Hz, 2H, 4-methylsulfonylphenyl H-3, H-
5), 7.68 (d, Jortho = 8.4 Hz, 2H, 4-methylsulfonylphenyl
H-2, H-6), 7.43–7.40 (m, 5H, phenyl hydrogens), 3.12
(s, 3H, CH3). Anal. Calcd for C15H12N2O4S: C, 56.95;
H, 3.82; N, 8.86. Found: C, 56.96; H, 3.76; N, 8.89.
1
30%) as a pale brown liquid; H NMR (CDCl3): d 8.10
(d, Jortho = 8.4 Hz, 2H, 4-chlorosulfonylphenyl H-3, H-
5), 7.83 (d, Jortho = 8.4 Hz, 2H, 4-chlorosulfonylphenyl
H-2, H-6), 7.50–7.53 (m, 5H, phenyl hydrogens). The
sulfonyl chloride product 15 was used immediately after
purification for the synthesis of compounds 16 and 17.
5.3. 3,4-Diphenyl-1,2,5-oxadiazole (11)
A solution of 10 (0.3 g, 1.26 mmol) in triethylphosphite
(10 mL) was heated at reflux for 24 h with stirring, the
solution was cooled to 25 ꢂC, and the reaction mixture
was stirred for 25 ꢂC for 2 h. The reaction was quenched
by addition of H2SO4 (100 mL of 2 N), the mixture was
stirred for 20 min, and the product was extracted with
CH2Cl2 (2 · 50 mL). The combined CH2Cl2 extracts
were washed with water (6 · 50 mL), the organic phase
was dried (Na2SO4), and the solvent was removed in va-
cuo. The residue obtained was purified by silica gel col-
umn chromatography (hexanes–ether, 4:1 v/v) to yield
5.6. 3-[4-(Chlorosulfonyl)phenyl]-4-phenyl-1,2,5-oxadiaz-
ole (18)
3,4-Diphenyl-1,2,5-oxadiazole (11, 0.86 g, 3.87 mmol)
was added to chlorosulfonic acid (5 mL) at ice-bath tem-
perature with stirring, after 10 min the ice bath was re-
moved, the reaction mixture was allowed to warm to
25 ꢂC, and the reaction was allowed to proceed at
25 ꢂC for 17 h. The brown reaction mixture was poured
dropwise with caution onto crushed ice (100 g). The
product was extracted with EtOAc (3 · 30 mL), and
the organic phase was washed repeatedly with water un-
til the pH was neutral. The EtOAc fraction was dried
(Na2SO4), the solvent was removed in vacuo, and the
residue obtained was purified by silica gel column chro-
matography (hexanes–EtOAc, 96:4 v/v) to furnish the
sulfonyl chloride product 18 (0.20 g, 16%) as a pale
1
11 (0.19 g, 70%) as a white powder; mp 67–70 ꢂC; H
NMR (CDCl3): d 7.41–7.56 (m, 10H, phenyl hydro-
gens). Anal. Calcd for C14H10N2O: C, 75.66; H, 4.54;
N, 12.60. Found: C, 75.39; H, 4.80; N, 12.78.
5.4. 3-[4-(Methylsulfonyl)phenyl]-4-phenyl-1,2,5-oxadiaz-
ole (14)
A mixture (ratio of 3:1) of the two regioisomers 13a and
13b (0.65 g, 2.05 mmol) dissolved in triethylphosphite
brown liquid; 1H NMR (CDCl3):
Jortho = 8.1 Hz, 2H, 4-chlorosulfonylphenyl H-3, H-5),
d
8.11 (d,

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2756
1
7.84 (d, Jortho = 8.1 Hz, 2H, 4-chlorosulfonylphenyl H-2,
H-6), 7.50–7.52 (m, 5H, phenyl hydrogens). The sulfonyl
chloride 18 was used immediately after purification for
the synthesis of compounds 19 and 20.
(NH), 1333 (SO) cmÀ1; H NMR (CDCl3): d 7.74 (d,
Jortho = 8.4 Hz, 2H, 4-aminosulfonylphenyl H-3, H-5),
7.39 (d, Jortho = 8.4 Hz, 2H, 4-aminosulfonylphenyl
H-2, H-6), 7.23–7.27 (m, 5H, phenyl hydrogens), 6.76
(s, 2H, NH2). Anal. Calcd for C14H11N3O3S: C, 55.80;
H, 3.68; N, 13.95. Found: C, 55.99; H, 3.59; N, 13.75.
5.7. 4-[4-(Aminosulfonyl)phenyl]-3-phenyl-1,2,5-oxadiaz-
ole-2-oxide and 3-[4-(aminosulfonyl)phenyl]-4-phenyl-
1,2,5-oxadiazole-2-oxide regioisomers (16)
5.10. 3-[4-(Azidosulfonyl)phenyl]-4-phenyl-1,2,5-oxadiaz-
ole (20)
An excess of ammonium hydroxide (2 mL of 30% w/v)
was added dropwise to a mixture of the sulfonyl chloride
regioisomers 15 (0.24 g, 0.71 mmol) in THF (10 mL) at
25 ꢂC with stirring. The reaction was allowed to proceed
for 1.5 h at 25 ꢂC, water (100 mL) was added to quench
the reaction, and this mixture was extracted with EtOAc
(3 · 20 mL). The combined EtOAc extracts were washed
with water (3 · 25 mL) and the EtOAc fraction was
dried (Na2SO4). Removal of the solvent in vacuo affor-
ded 16 (0.15 g, 69%) as a white powder; mp 245–247 ꢂC;
IR (KBr): 3321 (NH), 3261 (NH), 1595 (NO), 1326
(SO) cmÀ1; 1H NMR (CDCl3): d 7.82 (d, Jortho = 8.4 Hz,
2H, 4-aminosulfonylphenyl H-3, H-5), 7.47 (d,
Jortho = 8.4 Hz, 2H, 4-aminosulfonylphenyl H-2, H-6),
7.31–7.37 (m, 5H, phenyl hydrogens), 6.70 (s, 2H,
NH2). Anal. Calcd for C14H11N3O4S: C, 52.99; H,
3.49; N, 13.24. Found: C, 53.31; H, 3.34; N, 13.15.
Sodium azide (72 mg, 1.11 mmol) was added to the sul-
fonyl chloride 18 (0.11 g, 0.37 mmol) in THF (10 mL)
and water (2 mL) at 25 ꢂC with stirring. The reaction
was allowed to proceed for 12 h with stirring at 25 ꢂC,
EtOAc (40 mL) was added to quench the reaction, the
solids removed by filtration were discarded, the organic
phase was washed with water (2 · 15 mL), the organic
phase was dried (Na2SO4), and the solvent was removed
in vacuo. The residue obtained was purified by silica gel
column chromatography (hexanes–EtOAc, 90:10 v/v) to
afford 20 (0.10 g, 86%) as a white semi-solid; IR (KBr):
1
3073 (CH), 2139 (N3), 1595 (NO), 1434 (SO); H NMR
(CDCl3): d 8.07 (d, Jortho = 8.4 Hz, 2H, 4-azidosulfonyl-
phenyl H-3, H-5), 7.71 (d, Jortho = 8.4 Hz, 2H, 4-azi-
dosulfonylphenyl H-2, H-6), 7.46–7.51 (m, 5H, phenyl
hydrogens). Anal. Calcd for C14H9N5O3S: C, 51.37; H,
2.77; N, 21.40. Found: C, 51.53; H, 2.45; N, 21.27.
5.8. 4-[4-(Azidosulfonyl)phenyl]-3-phenyl-1,2,5-oxadiaz-
ole-2-oxide and 3-[4-(azidosulfonyl)phenyl]-4-phenyl-
1,2,5-oxadiazole-2-oxide regioisomers (17)
5.11. In vitro cyclooxygenase (COX) inhibition assays
The ability of the test compounds listed in Table 1 to in-
hibit ovine COX-1 and COX-2 (IC50 value, lM) was
determined using an enzyme immuno assay (EIA) kit
(catalog no. 560101, Cayman Chemical, Ann Arbor,
MI, USA) according to our previously reported
method.4
Sodium azide (87 mg, 1.26 mmol) was added to a mix-
ture of regioisomers 15 (0.21 g, 0.63 mmol) in THF
(10 mL) and water (2 mL) at 25 ꢂC. The reaction was al-
lowed to proceed for 12 h with stirring at 25 ꢂC, EtOAc
(40 mL) was added to quench the reaction, the insoluble
solid was removed by filtration and discarded, the or-
ganic phase was washed with water (2 · 15 mL), dried
(Na2SO4), and the solvent was removed in vacuo. The
residue obtained was purified by silica gel column chro-
matography (hexanes–ether, 3:1 v/v) to furnish 17
(0.16 g, 76%) as white crystals; mp 85–90 ꢂC; IR
(KBr): 3059 (CH), 2146 (N3), 1595 (NO), 1380
(SO) cmÀ1; 1H NMR (CDCl3): d 8.04 (d, Jortho = 8.4 Hz,
2H, 4-azidosulfonylphenyl H-3, H-5), 7.82 (d,
Jortho = 8.4 Hz, 2H, 4-azidosulfonylphenyl H-2, H-6),
7.56–7.59 (m, 5H, phenyl hydrogens). Anal. Calcd for
C14H9N5O4S: C, 48.98; H, 2.64; N, 20.40. Found: C,
49.16; H, 2.40; N, 20.43.
5.12. In vitro nitric oxide release assay
In vitro nitric oxide release in phosphate buffer, in the
presence or absence of L-cysteine, was determined by
quantification of nitrite produced by the reaction of ni-
tric oxide with oxygen and water using the Griess reac-
tion. Nitric oxide release data acquired for the test
compounds (10, 13a,b, 16, and 17) using the reported
procedures are listed in Table 2.29
5.13. Molecular modeling (docking) studies
Docking experiments were performed using INSIGHT II
software Version 2000.1 (Accelrys Inc.) running on a Sil-
icon Graphics Octane 2 R14000A workstation accord-
ing to a previously reported method.4
5.9. 3-[4-(Aminosulfonyl)phenyl]-4-phenyl-1,2,5-oxadiaz-
ole (19)
Excess ammonium hydroxide (1 mL of 30% w/v) was
added dropwise to a solution of the sulfonyl chloride
18 (81 mg, 0.25 mmol) in THF (10 mL) at 25 ꢂC. The
reaction was allowed to proceed for 1.5 h at 25 ꢂC with
stirring and water (100 mL) was added to quench the
reaction. Extraction with EtOAc (3 · 20 mL), washing
the combined EtOAc extracts with water (3 · 25 mL),
drying the EtOAc fraction (Na2SO4), and removal of
the solvent in vacuo yielded 19 (62 mg, 82%) as a white
powder; mp 227–229 ꢂC; IR (KBr): 3314 (NH), 3224
5.14. Crystal structure data for the 4-[4-(methylsulfon-
yl)phenyl]-3-phenyl-1,2,5-oxadiazole-2-oxide (13a) and 3-
[4-(methylsulfonyl)phenyl]-4-phenyl-1,2,5-oxadiazole-2-
oxide (13b) mixture or regioisomers (3:1 ratio)
Molecular formula: C15H12N2O4S, formula weight:
316.33, crystal system: monoclinic, space group: P21/c
˚
(no. 14) with unit cell dimensions a = 18.1969 (18) A,
˚
˚
b = 7.1642 (7) A, c = 11.6272 (11) A, b = 106.4720

´
C. Velazquez et al. / Bioorg. Med. Chem. 13 (2005) 2749–2757
2757
3
(17)ꢂ, V = 1453.6 (2) A , Z = 4, qcalcd = 1.445 g cmÀ3
,
˚
Perkins, W.; Seiber, K.; Veenhuizen, A.; Isakson, P. J.
Med. Chem. 1997, 40, 1347.
l = 0.242 mmÀ1. A crystal fragment of approximate
dimensions (mm3) 0.49 · 0.25 · 0.04 was mounted in a
nonspecific orientation on a Bruker PLATFORM/
SMART 1000 CCD diffractometer. All intensity mea-
surements were performed using Mo Ka radiation
6. Chan, C.; Boyce, S.; Brideau, C.; Charleson, S.; Cromlish,
W.; Ethier, D.; Evans, J.; Ford-Hutchinson, A.; Forrest,
M.; Gauthier, J.; Gordon, R.; Gresser, M.; Guay, J.;
Kargman, S.; Kenedy, B.; Leblanc, Y.; Leger, S.; Mancini,
J.; OÕneil, G.; Ouellet, M.; Patrick, D.; Percival, M.;
Perrier, H.; Prasit, P.; Rodger, I. J. Pharmacol. Exp. Ther.
1999, 290, 551.
˚
(k = 0.71073 A) with a graphite crystal incident beam
monochromator. The intensity data were collected at
À80ꢂ using x scans (0.2ꢂ scans, 45 s exposures). A total
of 2970 independent reflections were collected to a max-
imum 2h limit at 52.86ꢂ. The structure was solved by di-
rect methods (SHELXS-86). Refinement of atomic
parameters was carried out by using full-matrix least-
squares on F2 (SHELXL-93). Although the initial model
refined was for 13a, elongation of the thermal parame-
ters for the phenyl and oxadiazole rings, plus the occur-
rence of a large residual electron-density peak near the
5-nitrogen of the 1,2,5-oxadiazole ring, suggested that
the isomer 13b was also present. The two isomers share
the MeSO2 group, but separate sets of positions for the
phenyl and 1,2,5-oxadiazole 2-oxide groups were refined
for both 13a and 13b, in 75:25 ratio. The final model
gave agreement factors (R indices) of R1(F) = 0.0685
(for 2432 data with I P 2r(I)) and wR2(F2) = 0.1982
(for all 2970 unique data). Crystallographic data
(excluding structure factors) have been deposited at
the Cambridge Crystallographic Data Centre as supple-
mentary publication number CCDC 252139. Copies of
the data can be obtained free of charge by application
to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
(fax: 44 01223 336033 or e-mail: deposit@ccdc.cam.
ac.uk or http://www.ccdc.cam.uk).
7. Li, J.; Anderson, G. D.; Burton, E. G.; Cogburn, N.;
Collins, J. T.; Garland, D. J.; Gregory, S. A.; Huang, H.;
Isakson, P. C. J. Med. Chem. 1995, 38, 4570.
8. Talley, J.; Brown, D.; Carter, J.; Graneto, M.; Koboldt,
C.; Masferrer, J.; Perkins, W.; Rogers, R.; Shaffer, A.;
Zhang, Y.; Zweifel, B.; Seibert, K. J. Med. Chem. 2000, 43,
775.
9. Padi, S. S.; Jain, N. K.; Singh, S.; Kulkarni, S. K. Eur. J.
Pharmacol. 2004, 491, 69.
10. Riendeau, D.; Percival, M. D.; Brideau, C.; Charleson, S.;
DubÕe, D.; Ethier, D.; Falgueyret, J.-P.; Friesen, R. W.;
Gordon, R.; Grieg, G.; Guay, J.; Mancini, J.; Ouellet, M.;
Wong, E.; Xu, L.; Boyce, S.; Visco, D.; Girard, Y.; Prasit,
P.; Zamboni, R.; Rodger, I. W.; Gresser, M.; Ford-
Hutchinson, A. W.; Young, R. N.; Chan, C.-C. J. Phar-
macol. Exp. Ther. 2001, 296, 558.
11. Rao, P. N. P.; Amini, M.; Li, H.; Habeeb, A.; Knaus, E.
Bioorg. Med. Chem. Lett. 2003, 13, 2205.
12. Mukherjee, D.; Nissen, S. E.; Topol, E. J. JAMA 2001,
286, 954.
13. (a) Marx, V. Chem. Eng. News 2004, 82, 8; (b) Voluntary
withdrawal notification published by Merck on September
30, 2004, available at http://www.vioxx.com.
14. Wallace, J.; Del-Soldato, P. Fund. Clin. Pharmacol. 2003,
17, 11.
15. Vallance, P. Fund. Clin. Pharmacol. 2003, 17, 1.
16. Ignarro, L. J. Kidney Int. 1996, 55, S2.
17. Russwurm, M.; Koesling, D. Mol. Cell. Biochem. 2002,
230, 159.
Acknowledgements
18. Kubes, P.; Wallace, J. Med. Inflamm. 1995, 4, 397.
19. Whittle, B. Br. J. Pharmacol. 1993, 110, 3.
20. Wallace, J.; Reuter, B.; Cicala, C.; McKnight, W.;
Grisham, M.; Cirino, G. Eur. J. Pharmacol. 1994, 257,
249.
We are grateful to the Canadian Institutes of Health Re-
search (MOP-14712) for financial support of this re-
search, and the National Council of Science and
´
Technology (CONACyT, Mexico) for a scholarship to
C.V.
21. Wallace, J.; Reuter, B.; Cicala, C.; McKnight, W.;
Grisham, M.; Cirino, G. Gastroenterology 1994, 107, 173.
22. Granik, V. G.; Grigor, N. B. Russ. Chem. Bull., Int. Ed.
2002, 51, 1375.
23. Gasco, A. M.; Fruttero, R.; Sorba, G.; Gasco, A. Liebigs
Ann. Chem. 1991, 1211.
References and notes
24. Frutero, R.; Ferraroti, B.; Serafino, A.; DiStilo, A.;
Gasco, A. J. Heterocycl. Chem. 1989, 26, 1345.
25. Zarghi, A.; Rao, P. N. P.; Knaus, E. E. Bioorg. Med.
Chem. Lett. 2004, 14, 1957.
26. Civelli, M.; Caruso, P.; Giossi, M.; Bergamaschi, M.;
Razzetti, R.; Bongrani, S.; Gasco, A. Br. J. Pharmacol.
1996, 118, 923.
1. Mitchell, J.; Warner, T. Br. J. Pharmacol. 1999, 128, 1121.
2. Mitchell, J.; Akarasereenont, P.; Thiemerman, C.; Flower,
R.; Vane, J. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 11693.
3. Warner, T.; Giuliano, F.; Vojnovic, I.; Bukasa, A.;
Mitchell, J.; Vane, J. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 7563.
4. Rao, P. N. P.; Amini, M.; Li, H.; Habeeb, A.; Knaus,
E. E. J. Med. Chem. 2003, 46, 4872.
27. Soliva, R.; Almansa, C.; Kalko, S. G.; Luque, F. J.;
Orozco, M. J. Med. Chem. 2003, 46, 1372.
5. Penning, T.; Talley, J.; Bertenshaw, S.; Carter, J.; Collins,
P.; Docter, S.; Graneto, M.; Lee, L.; Malecha, J.;
Miyashiro, J.; Rogers, R.; Rogier, D.; Yu, S.; Anderson,
G.; Burton, E.; Cogburn, J.; Gregory, S.; Koboldt, C.;
28. Rao, P. N. P.; Habeeb, A.; Knaus, E. E. Drug Dev. Res.
2002, 55, 79.
´
29. Velazquez, C.; Vo, D.; Knaus, E. E. Drug Dev. Res. 2003,
60, 204.