In Vitro Structure-Activity Relationship and in Vivo Studies for a Novel Class of Cyclooxygenase-2 Inhibitors: 5-Aryl-2,2-dialkyl-4-phenyl-3(2H)furanone Derivatives

Article abstract of DOI:10.1021/jm020545z

5-Aryl-2,2-dialkyl-4-phenyl-3(2H)furanone derivatives were studied as a novel class of selective cyclooxygenase-2 inhibitors with regard to synthesis, in vitro SAR, antiinflammatory activities, pharmacokinetic considerations, and gastric safety. If, a representative compound for methyl sulfone derivatives, showed a COX-2 IC₅₀ comparable to that of rofecoxib. In case of 20b, a representative compound for sulfonamide derivatives, a potent antiinflammatory ED₅₀ of 0.1 mg kg^-1 day^-1 was observed against adjuvant-induced arthritis by a preventive model, positioning 20b as one of the most potent COX-2 inhibitors ever reported. Furthermore, 20b showed strong analgesic activity as indicated by its ED₅₀ of 0.25 mg/kg against carrageenan-induced thermal hyperalgesia in the Sprague-Dawley rat. 3(2H)Furanone derivatives showed due gastric safety profiles as selective COX-2 inhibitors upon 7-day repeat dosing. A highly potent COX-2 inhibitor of the 3(2H)furanone scaffold could be considered suitable for a future generation COX-2 selective arthritis medication with improved safety profiles.

Full text of DOI:10.1021/jm020545z

792
J . Med. Chem. 2004, 47, 792-804
Articles
In Vitr o Str u ctu r e-Activity Rela tion sh ip a n d in Vivo Stu d ies for a Novel Cla ss
of Cyclooxygen a se-2 In h ibitor s: 5-Ar yl-2,2-d ia lk yl-4-p h en yl-3(2H)fu r a n on e
Der iva tives
Song Seok Shin,* Youngjoo Byun,† Kyung Min Lim, J in Kyu Choi, Ki-Wha Lee, J oo Hyun Moh, J in Kwan Kim,
Yeon Su J eong, J i Young Kim, Young Hoon Choi, Hyun-J u Koh, Young-Ho Park, Young Im Oh,
Min-Soo Noh,‡ and Shin Chung*
Drug Discovery, AmorePacific R&D Center, 314-1 Bora-ri, Kiheung-eup, Yongin-si, Kyounggi-do 449-729, South Korea
Received December 3, 2002
5-Aryl-2,2-dialkyl-4-phenyl-3(2H)furanone derivatives were studied as a novel class of selective
cyclooxygenase-2 inhibitors with regard to synthesis, in vitro SAR, antiinflammatory activities,
pharmacokinetic considerations, and gastric safety. 1f, a representative compound for methyl
sulfone derivatives, showed a COX-2 IC50 comparable to that of rofecoxib. In case of 20b, a
representative compound for sulfonamide derivatives, a potent antiinflammatory ED50 of 0.1
mg kg-1 day-1 was observed against adjuvant-induced arthritis by a preventive model,
positioning 20b as one of the most potent COX-2 inhibitors ever reported. Furthermore, 20b
showed strong analgesic activity as indicated by its ED50 of 0.25 mg/kg against carrageenan-
induced thermal hyperalgesia in the Sprague-Dawley rat. 3(2H)Furanone derivatives showed
due gastric safety profiles as selective COX-2 inhibitors upon 7-day repeat dosing. A highly
potent COX-2 inhibitor of the 3(2H)furanone scaffold could be considered suitable for a future
generation COX-2 selective arthritis medication with improved safety profiles.
In tr od u ction
responsible for bodily homeostasis such as the gas-
trointestinal integrity, vascular dilatation, renal func-
tions, and so on. Overt inhibition of COX-1 can, there-
fore, elicit undesirable side effects such as gastric PUB
and blood thinning.4 In the meantime, COX-2 is induced
upon inflammatory stimuli and is known to be respon-
sible for progression of inflammation. Traditional
NSAIDs, such as aspirin, naproxen, piroxicam, ibupro-
fen, diclofenac, etc., inhibit both COX-1 and COX-2,
which accounts for NSAIDs’ antiinflammatory effects
as well as their notorious side effects of GI toxicity and
blood thinning. Thus, selective inhibition of COX-2 over
COX-1 should be useful for treatment of inflammation
without incurring the side effects associated with inhi-
bition of COX-1.
Nonsteroidal antiinflammatory drugs (NSAIDs) have
been widely used over 100 years to alleviate symptoms
of arthritis, arthritis-associated disorders, fever, post-
operative pain, migraine, and so on.1 Despite their
widespread use and desirable therapeutic efficacy for
the treatment of inflammation and inflammation-as-
sociated disorders, NSAIDs are generally regarded to
have life-threatening toxicity in the gastrointestinal (GI)
tract. Severity of the GI toxicity is well illustrated by a
report that 16 500 patients on NSAIDs therapy died due
to the GI toxicity in the year of 1994 alone in the US.2
Frequently, the gastric toxicity of perforation, ulcer-
ation, and bleeding (PUB) is not noticed by patients
before hospitalization, leading to such a high mortality
rate upon chronic use of NSAIDs.
Despite the huge amount of efforts directed to reduce
the GI toxicity of NSAIDs, it was only about a decade
ago that the origin of the GI toxicity began to be
understood through the discovery of an inducible iso-
form of cyclooxygenases.3 There are at least two kinds
of cyclooxygenases, cyclooxygenase-1 (COX-1) and cy-
clooxygenase-2 (COX-2). COX-1 is constitutively ex-
pressed in various tissues including the GI tract, the
kidneys, and the platelets. COX-1 is known to be
A study with COX-2 knockout mice suggests that
complete inhibition of COX-2 could lead to peritonitis
secondary to intestinal toxicity.5 Animal safety data for
COX-2 inhibitors indicated that the intestinal toxicity
was the dose-limiting toxicity in the dog and the rat.6
However, primates seem to possess robust intestinal
tolerance to selective inhibition of COX-2.6c Indeed,
COX-2 inhibitors are regarded to have better gas-
trointestinal safety profiles than traditional NSAIDs.7
Long term use of traditional NSAIDs has been known
to elicit cardiorenal toxicity such as edema and worsen-
ing blood pressure. There have been some attempts to
assess cardiorenal safety of COX-2 inhibitors;8 however,
more clinical data are needed to estimate the cardiore-
nal safety of COX-2 inhibitors. Considering that COX-2
inhibitors are supposed to be chronically taken mostly
* To whom correspondence should be addressed. S.S.S.: Tel: +82-
31-2805915. E-mail: ssshin@amorepacific.com. S.C.: Tel: +82-31-
2805910. Fax: +82-31-2818391. E-mail: schung@amorepacific.com.
† Current address: Department of Medicinal Chemistry, Ohio State
University, OH 43210, USA.
‡ Current address: Department of Bioengineering, Massachusetts
Institutes of Technology, MA 02139, USA.
10.1021/jm020545z CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/14/2004

Novel Class of Cyclooxygenase-2 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 793
Ta ble 1. In Vitro COX-1/COX-2 Inhibitory Activities by the
Mouse Peritoneal Macrophage Method for 2,2-Dialkyl-
4-aryl-5-{4-(methylsulfonyl)phenyl}-3(2H)-furanone Derivatives
1∼ 3
by elderly arthritis patients, the importance of the long-
term cardiorenal safety can never be overemphasized.
COX-2 is constitutively expressed in the glomerular
region and the small blood vessels of the kidneys in
primates including the human,8d suggesting that the
smaller inhibition of renal COX-2 could lead to smaller
renal and consequently cardiovascular adverse effects.
Given that only protein-unbound drug molecules are
subject to glomerular filtration, a drug with higher
plasma protein binding is expected to exert a smaller
renal effect for a given lipophilicity or hydrophilicity of
drug.
IC50, µg/mL
COX-2
entry
AR
COX-2 COX-1 selectivity
There are already several COX-2 inhibitors being
prescribed for chronic indications, and they mostly
maintain a tricyclic structure as in rofecoxib,9a celecoxib,9b
valdecoxib,9c and etoricoxib.9d,9e As more clinical data
accumulate, it may become possible to differentiate
profiles of COX-2 inhibitors based on efficacy and safety.
There have been some, but not conclusive, attempts to
differentiate profiles of COX-2 inhibitors, though.10,11
1a
1b
1c
1d
1e
phenyl
0.05
0.02
0.01
0.03
0.3
3
5
2
0.03
3
60
3-fluorophenyl
3-chlorophenyl
4-bromophenyl
3-chloro-4-
fluorophenyl
3,5-difluorophenyl
3,5-dichlorophenyl
250
200
1
10
1f
1g
1h
0.05
0.03
20
3
3
400
100
60
3-(trifluoromethyl)- 0.05
phenyl
Previously, we reported 2,2-dimethyl-4,5-diaryl-3(2H)-
furanone (1) derivatives as orally active COX-2 inhibi-
tors. In case of 1f (see Table 1 for the chemical
structure), the observed adjuvant-induced arthritis ED50
was 0.03 mpk/day by the therapeutic model.12 Such a
pronounced oral activity in the adjuvant arthritis is one
of the most potent among the reported COX-2 inhibitors.
The 3(2H)furanone derivatives are structurally similar
to celecoxib and rofecoxib, in that they have a scaffold
or pharmacophore of cis-1,2-diaryl-alkene type.13 The
scaffold of COX-2 inhibitors not only is important for
COX-2 selectivity but also can be critical to their in vivo
profiles. For example, the major primary metabolic
pathway for celecoxib in humans involves cytochrome
P450 isozymes.14a On the other hand, rofecoxib is
metabolized mainly by unidentified cytosolic re-
ductases.14b Some COX-2 inhibitors with a scaffold of
cis-1,2-diaryl-alkene type failed to show pharmacological
profiles acceptable for clinical development. For ex-
ample, a metabolite of Du P 697 from Dupont Merck
was reported to accumulate in the body.15 SC-236 from
G. D. Searle has been reported to show an unacceptably
long plasma half-life.9b In this article will be presented
the synthesis, structure-activity relationship, and phar-
macological properties of 4,5-diaryl-3(2H)furanone de-
rivatives as a novel class of highly potent selective
COX-2 inhibitors.
1i
1j
1k
1l
1m
1n
1o
1p
1q
1r
3-isopropylphenyl
4-isopropylphenyl
4-n-butylphenyl
3-acetylphenyl
4-acetylphenyl
3-methoxyphenyl
2-furanyl
3-pyridyl
3-thienyl
4-(1-N-methyl-
pyrazolyl)
phenyl
2-fluorophenyl
3-chlorophenyl
3,5-dichlorophenyl
3-(trifluoromethyl)- 0.05
phenyl
3-methoxyphenyl
phenyl
4-(trifluoromethyl)- 0.03
phenyl
0.03
0.03
0.002
0.05
0.3
0.1
0.2
0.5
0.2
30
3
5
50
50
30
30
10
30
30
1000
100
2500
1000
167
300
150
20
150
60
0.5
2a
2b
2c
2d
2e
0.2
0.3
0.03
0.03
20
30
5
20
10
100
100
167
666
400
2f
3a
3b
0.3
5
30
30
3
100
6
100
3c
3d
3-methylphenyl
4-methylphenyl
0.5
10
3
1.86
>100
20
100
93
0.03
0.02
0.06
celecoxib
r ofecoxib
>1667
Syn th esis
2,2-Dialkyl-4,5-diaryl-3(2H)furanone derivatives (1 ∼
5) with methylsulfonyl group were prepared as outlined
in Scheme 1. We previously reported synthesis of 2,2-
dimethyl-4,5-diphenyl-3(2H)furanone derivatives by the
same reaction sequence as Scheme 1.12 4-(Methylthio)-
benzaldehyde was coupled with the in situ-generated
lithium acetylenide of 6 to afford diol 7. The benzylic
hydroxyl group of 7 was then oxidized to the corre-
sponding ketone 8 by reaction either with MnO2 or with
pyridinium dichromate (PDC). Often MnO2 allowed
smoother conversion of 7 to 8 than PDC. The acyclic
ketone 8 was then cyclized to give 2,2-dialkyl-5-(4-
methylthiophenyl)-3(2H)furanone 9 in the presence of
diethylamine as described in the literature.16 The me-
thylthio group of furanone 9 was smoothly converted
to the methylsulfonyl group of 3(2H)furanone 10 by
reaction with OXONE. 3(2H)furanone 10 was then
subjected to bromination with bromine in acetic acid to
afford the corresponding bromide compound 11. Suzuki
coupling17 of 11 with an appropriate arylboronic acid

794 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Shin et al.
Sch em e 1a
a
Reagents and conditions: (a) n-BuLi, 4-(methylthio)benzaldehyde, THF, -78 °C; (b) CrO3/acetone, PDC/CH2Cl2, or MnO2/CH2Cl2; (c)
(Et)2NH/CH3OH; (d) Br2/Ac2O/CCl4; (e) oxone/THF/CH3OH/H2O; (f) arylboronic acid, Pd(PPh3)4, aq NaHCO3/toluene/EtOH, ∆.
Sch em e 2a
sequence of Scheme 1 was modified to obtain desired
target compounds. For example, the sequence of the
OXONE and the bromination reactions were reverted
to yield target molecules 1 ∼ 5.
Alternatively, 2,2-dimethyl-4,5-diphenyl-3(2H)fura-
none derivatives (1 and 12) with methylsulfonyl group
were prepared according to Scheme 2. The enolate of
1-(4-methylthiophenyl)-2-phenyl-ethan-1-one (13) was
reacted with 2-bromo-isobutyryl-cyanide (14: BBC)18 to
effect the desired C-acylation against the O-acylation.
BBC worked very effectively as a soft electrophile for
this specific case to resolve the classical issue of “C- vs
O-acylation”. The acyclic C-acylation product 15 could
be isolated; however, use of excess base allowed 15 to
be cyclized to 3(2H)furanone derivatives 16 in one pot.
The effectiveness of the one-pot cyclization reaction was
discussed recently in detail with regard to its potential
application to a GMP-compatible scale-up process of 2,2-
a
Reagents and conditions: (a) NaHMDS, THF; (b) NaH, THF;
(c) OXONE, THF/CH3OH/H2O.
was employed to generate structural diversity for 3(2H)-
furanone derivatives 1 ∼ 5. In some cases, the reaction
Sch em e 3a
a
Reagents and conditions: (a) m-CPBA/THF, 0 °C; (b) Br2/Ac2O/CH2Cl2 or cat. BTI/I2/CCl4; (c) (i) TFAA, 0 °C; (ii) TEA/CH3OH; (iii)
Cl2/AcOH; and (iv) aq NH4OH/THF; (d) arylboronic acid, Pd(PPh3)4, aq NaHCO3/tolene/EtOH, ∆.

Novel Class of Cyclooxygenase-2 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 795
Sch em e 4a
a
Reagents and conditions: (a) m-CPBA, THF, 0 °C; (b) (i) TFAA, 0 °C; (ii) TEA, CH3OH; (iii) Cl2, AcOH; (iv) aq NH4OH, THF.
dimethyl-4,5-diaryl-3(2H)furanone derivatives.19 The
methyl sulfide group of 16 was then converted to the
corresponding methyl sulfone group of 1 and 12 by
reacting with OXONE.
ages, celecoxib and rofecoxib showed about 100-fold and
over 1500-fold COX-2 selectivity over COX-1, respec-
tively. It is interesting that the two blockbusters showed
comparable COX-2 IC50’s by this cell-based assay,
though. The 3(2H)furanone derivatives in Table 1
showed a wide range of COX-2 selectivity. For example,
compounds 1d and 1e showed COX-2 selectivity too
marginal to be classified as a selective COX-2 inhibitor.
Compounds such as 1b, 1c, 1f, and 1h showed in vitro
profiles comparable to celecoxib in terms of COX-2
selectivity and COX-2 IC50. A very high COX-2 selectiv-
ity was achieved with 1k , reaching the level of the
rofecoxib’s COX-2 selectivity. It is often the case that
relatively better COX-2 selectivity was observed by
having a substituent at the 3-position of the 4-phenyl
group of 1. Structural flexibility of the n-butyl moiety
of 1k seems to allow itself to maneuver into the ideal
space for COX-2 selectivity near the 3-position of the
4-phenyl group.
3(2H)Furanone derivatives 20 with a sulfonamide
moiety were prepared as described in Scheme 3. First,
the methylthio group of 2,2-dialkyl-5-(4-methylthiophe-
nyl)-3(2H)furanone 9 was subjected to partial oxidation
with m-CPBA to give the corresponding sulfoxide 17.
Then, the sulfoxide 17 was halogenated either with
bromine in acetic acid, or with iodine in the presence of
a catalytic amount of BTI [{bis(trifluoroacetoxy)iodo}-
benzene] to yield the corresponding halide 18. Sulfoxide
18 was converted to the corresponding sulfonamide 19
by a series of reactions with (1) trifluoroacetic anhy-
dride, (2) triethylamine in methanol, (3) chlorine in
acetic acid, and then (4) ammonia water according to
the literature.20 Finally, Suzuki coupling of 19 with
appropriate arylboronic acids afforded the desired 3(2H)-
furanone derivative 20 for evaluation of COX-1/COX-2
inhibition.
Sulfonamide-containing 3(2H)furanone derivatives
could also be prepared by modifying Scheme 3. In
Scheme 4, 2,2-dimethyl-5-(4-methylthiophenyl)-4-phen-
yl-3(2H)furanone derivatives were used as starting
materials to prepare sulfonamide-containing COX-2
inhibitors 20 or 22. Step 1 is basically partial oxidation
of methylthio group to the sulfoxide moiety similar to
the step 1 of Scheme 3. The step 2 of Scheme 2 refers to
a series of reactions for the step 3 of Scheme 3. Schemes
3 and 4 are essentially the same. However, the former
scheme may be better suited to generate structural
diversity, and the latter could be more appropriate for
a scale-up synthesis due to facile availability of starting
material 16 in large quantity.
Heterocycle-containing compounds 1o ∼ 1r showed
relatively poor in vitro COX-2 activities when compared
with 4-phenyl-3(2H)furanone derivatives as well as the
blockbuster drugs. Their COX-2 selectivity was not far
off from that of celecoxib, though.
In case of 2-ethyl-2-methyl-5-{4-(methylsulfonyl)phe-
nyl}-4-phenyl-3(2H)-furanone derivatives (2) and 2,2-
diethyl-5-{4-(methylsulfonyl)phenyl}-4-phenyl-3(2H)-
furanone derivatives (3), there was not much noticeable
improvement of in vitro COX-2 inhibitory potency and
COX-2 selectivity when compared with 2,2-dimethyl-4-
aryl-5-{4-(methylsulfonyl)phenyl}-3(2H)furanone de-
rivatives (1). There was some improvement in COX-2
selectivity for 2d and 2e when compared to the corre-
sponding gem-dimethyl furanone analogues 1g and 1h ,
respectively. However, there are cases of decreased
COX-2 selectivity such as with 2f, when compared with
the corresponding gem-dimethyl analogue 1n .
Figure 1 shows an energy-minimized structure of 1k
bound to the active site of COX-2, where the 3-dimen-
sional structure for the COX-2 active site was con-
structed from the reported crystallographic structure.24
According to Figure 1, the dimethyl moiety on the 3(2H)-
furanone scaffold occupies the vacancy near the en-
trance of the active site of COX-2. The hydroxyl group
of the Tyr355 residue of COX-2 is close to the enol ether
moiety enough to potentially form a hydrogen bond with
the enolic oxygen of the 3(2H)furanone scaffold. Previ-
ously, the role of Tyr355 was implicated as a building
block for a hydrophobic pocket in the reported complex
structure for SC-558.24 Whatever the underlying mi-
croscopic binding interactions, it is interesting that 1k
showed a COX-2 IC50 reaching down to 2 ng/mL.
In the case of spirocyclopentyl moiety containing
derivatives 4, COX-2 activities are better than the
Detailed analytical and physical data of 3(2H)fura-
none COX-2 inhibitors are summarized in Experimental
Section along with actual synthetic examples for rep-
resentative compounds according to the aforementioned
synthetic schemes.
Resu lts a n d Discu ssion
In Vitr o COX-1 a n d COX-2 Activities. Compounds
in this article were evaluated against COX-1 and COX-2
mostly by using freshly harvested mouse peritoneal
macrophages.21 Some compounds were evaluated by the
human whole blood method as reported in the litera-
ture,22 to assess their inhibitory activities against
human COX-1 and COX-2.23
In Table 1, in vitro COX-1/COX-2 inhibitory activities
are summarized for 2,2-dimethyl-4-aryl-5-{4-(methyl-
sulfonyl)phenyl}-3(2H)furanone derivatives (1) along
with COX-2 inhibitors celecoxib and rofecoxib. By the
assay method employing mouse peritoneal macroph-

796 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Shin et al.
Ta ble 3. In Vitro COX-1/COX-2 Inhibitory Activities against
the Mouse Peritoneal Macrophage Method for 2,2-Dimethyl-
5-{4-(aminosulfonyl)phenyl}-4-aryl-3(2H)-furanone Derivatives
20
IC50, µg/mL
COX-2
entry
AR
COX-2 COX-1 selectivity
20a
20b
20c
20d
20e
phenyl
0.008
0.003
0.014
0.007
0.011
0.061
0.3
3.8
1.9
0.82
77
100
270
290
74
3-fluorophenyl
3-chlorophenyl
3,4-difluorophenyl
3-(trifluoromethyl)-
phenyl
20f
20g
20h
20i
4-acetylphenyl
2-furanyl
3-thienyl
0.013
0.057
0.12
0.26
>10
4.8
20
>175
93
4-(1-N-methylpyrazolyl)
0.03
100
3333
F igu r e 1. A docking structure for 3(2H)furanone derivative
1k (orange balls and sticks) in the active site of human COX-2
(green blue and gray capped sticks). In this docking structure,
the cavity near the entrance of the active site is visually
present around Arg120. The enolic oxygen of the 3(2H)-
furanone scaffold is close to the hydroxyl group of Tyr355 with
a distance of 2.66 Å, and the potential hydrogen bonding
between the enolic oxygen and the phenolic hydrogen of Tyr355
is highlighted by a red dashed line.
3,3-spiro group on the cyclopentene moiety on the in
vitro COX-2 inhibitory activity and COX-2 selectivity
for 1,2-diaryl cyclopentene derivatives.25 According to
their reports, the spirocyclopropyl moiety as represented
by SC-58451 appeared to be optimal for the in vitro
COX-2 inhibitory activity for their COX-2 inhibitors of
the 1,2-diarylcyclopentane scaffold. Thus, the SAR with
respect to the aliphatic substitution at the 2-position of
the 3(2H)furanone scaffold seems to correlate well with
the prior observations made with the cyclopentene
scaffold. For the sake of completeness in the SAR, it
would have been nicer to study 3(2H)furanone deriva-
tives without aliphatic substituents at the 2-position of
the 3(2H)furanone scaffold. Unfortunately, synthetic
attempts were hampered by unexpected problems relat-
ing to the stability and tautomerization of the key
intermediate 2,2-dihydro-5-{4-(methylsulfonyl)phenyl}-
3(2H)furanone leading to 3(2H)furanone derivatives
without aliphatic substituents at the 2-position of the
3(2H)furanone scaffold.26
Ta ble 2. In Vitro COX-1/COX-2 Inhibitory Activities against
the Mouse Peritoneal Macrophage Method for Spirocyclic
Compounds 4 and 5
IC50, µg/mL
entry
AR
COX-2 COX-1 COX-2 selectivity
4a
4b
4c
4d
5a
5b
phenyl
0.03
3
0.03
0.05
2
5
3
3
10
10
3
167
1
100
200
5
3-fluorophenyl
3-methylphenyl
4-isopropylphenyl
phenyl
3-methylphenyl
0.5
6
corresponding 2,2-diethyl-3(2H)furanone derivatives 3
(see Table 2): compare the COX-2 inhibitory potencies
of 3a and 3c with those of 4a and 4c, respectively. The
spirocyclopentyl moiety of 4 occupies a smaller space
than the diethyl moiety of 3, which may have to do with
the improved the COX-2 inhibitory potency of 4 when
compared to that of 3. Upon the basis of the reported
crystal structure of COX-2,24 there should be a cavity
of limited size available to the 2,2-dialkyl moiety of
3(2H)furanone COX-2 inhibitors. Spirocyclohexyl moiety
containing derivatives 5 showed quite poor in vitro
COX-2 inhibitory activities, again supporting the ratio-
nale for “the cavity of limited size near the entrance of
the COX-2 active site”. For now, the gem-dimethyl
moiety appears to be the optimal size for the cavity near
the entrance of the COX-2 active site. Previously,
scientists of G. D. Searle investigated the effect of the
The sulfonamide moiety has been reported to be a
reasonable replacement for the methyl sulfone moiety
for selective binding to COX-2.13 When the methyl
sulfone group of 3(2H)furanone derivatives 1 was re-
placed with the sulfonamide group, two major changes
were observed as previously reported by scientists of G.
D. Searle:25a COX-2 selectivity decreased; however, in
vitro COX-2 inhibitory potency improved considerably
(Table 3). For example, the COX-2 IC50 of 20b is better
than the corresponding methylsulfonyl compound 1b by
∼7-fold. Many sulfonamide analogues of the 3(2H)-
furanone scaffold showed COX-2 IC50’s reaching a few
ng/mL. However, many of the sulfonamide derivatives
20 showed reduced COX-2 selectivity, when compared

Novel Class of Cyclooxygenase-2 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 797
Ta ble 4. In Vitro COX-1/COX-2 Inhibitory Activities against
the Mouse Peritoneal Macrophage Method for 2,2-Dimethyl-5-
{Fluoro-4-(methylsulfonyl)phenyl}-4-phenyl-3(2H)furanone
Derivatives 12 and 2,2-Dimethyl-5-{Fluoro-4-(aminosulfonyl)-
phenyl}-4-phenyl-3(2H)furanone Derivatives 22
Ta ble 5. Oral Antiinflammatory Activity of COX-2 Inhibitors
and Traditional NSAID Indomethacin against CFE in the
Sprague-Dawley Rata
% inhibition
entry
( SD
entry
% inhibition ( SD
1b
1e
1h
1l
2c
12g
20c
20i
22e
celecoxib
36 ( 22
31 ( 7
31 ( 12
22 ( 12
23 ( 7
28 ( 5
39 ( 13
29 ( 12
27 ( 5
34 ( 4
1c
1f
1k
1r
16 ( 26
34 ( 8
8 ( 12
36 ( 11
26 ( 15
47 ( 9
26 ( 17
32 ( 11
31 ( 11
45 ( 4
4c
20b
20f
22c
22f
IC50, µg/mL
fluoride
position
COX-2
selectivity
entry
R
COX-2
COX-1
in d om eth a cin
a
12a
12b
12c
12d
12e
12f
12g
12h
22a
22b
22c
22d
22e
22f
H
3-F
3-Cl
3-F, 5-F
H
0.03
0.03
0.02
0.03
0.03
0.03
0.05
0.03
0.03
0.003
0.003
0.03
0.03
0.02
20
10
3
667
333
150
667
667
6667
600
6667
667
3333
1000
667
Antiinflammatory activities were measured using at least 5
animals/group at 3 mg/kg of body weight.
2
3
20
20
200
30
200
20
10
3
oral route, in that it takes only several hours to evaluate
a drug by CFE. Oral antiinflammatory potencies of
COX-2 inhibitors of the 3(2H)furanone scaffold against
CFE are summarized in Table 5. Celecoxib showed 34%
inhibition when orally administered at 3 mg/kg, and
most of the COX-2 inhibitors of Table 5 showed 30 ∼
35% CFE inhibition at 3 mg/kg. Traditional NSAID
indomethacin was the most potent among the com-
pounds of Table 5 at the 3 mg/kg dose level. Considering
that COX-2 induction is not fully responsible for this
acute inflammatory animal model, most of the COX-2
inhibitors in Table 5 could be regarded to possess due
and reasonable oral antiinflammatory potency as COX-2
inhibitors. Thus, the 3(2H)furanone scaffold turned out
to be a good choice for oral bioavailability.
Su bch r on ic in Vivo An tiin fla m m a tor y Activity.
Adjuvant-induced arthritis (AA) is a reasonable animal
model to simulate the chronic inflammatory situation
of arthritis, in that what is monitored for an antiin-
flammatory drug is suppression of the secondary hy-
perimmunological response developing over 10 ∼ 22
days or so after induction of arthritis by inoculation with
Mycobacterium butyricum in the Freund’s adjuvant.
There have been two methods to evaluate the antiin-
flammatory effect of a drug by AA, preventive and
therapeutic model. In the preventive model, drug treat-
ment is commenced 1 day before the induction of
arthritis, and is continued usually up to day 21 or so
on a daily basis.29b In the therapeutic model, however,
drug is usually administered on a daily basis from day
14 to day 21 or so after induction of arthritis.25b In the
preventive model, the uninflammed foot is taken as the
reference foot for evaluation of AA activity. In the
meantime, the inflamed foot is used as the reference in
the therapeutic model. Usually, an antiinflammatory
drug tends to show nominally stronger potency by the
therapeutic model than by the preventive model. In-
house experiences indicated that a 5-fold difference in
ED50 between the two models was not uncommon when
male SD rats were used for the therapeutic model.30a
Antiinflammatory effects of COX-2 inhibitors against
AA are summarized along with that of positive com-
parator indomethacin in Table 6.
3-F
3-Cl
3-F, 5-F
H
3-F
3-F, 5-F
H
2
3
20
5
100
3-F
3-F, 5-F
167
5000
to the corresponding methyl sulfone derivatives 1.
Improved COX-2 activity was traded off with reduced
COX-2 selectivity by having sulfonamide in place of
methyl sulfone. It should be nice if a COX-2 inhibitor
retained the potency level of sulfonamide compounds
20 while maintaining the selectivity level of the corre-
sponding methyl sulfone compounds 1.
Previously, scientists of J apan Tobacco Company
devised an innovative approach to improve COX-2
selectivity of their COX-2 inhibitors with 2-methylox-
azole scaffold. Substitution with a halide next to the
aminosulfonyl group improved COX-2 selectivity from
the corresponding compound without the halide sub-
stituent, as exemplified by J TE-522.27 To improve the
COX-2 selectivity of sulfonamide derivatives 20, a halide
moiety was introduced to the phenyl ring substituted
with sulfonamide group. According to Table 4, the trend
observed by the J apan Tobacco scientists appears to be
maintained to a large extent in COX-2 inhibitors of the
3(2H)furanone scaffold. Many of sulfonamide deriva-
tives 20 have only modest COX-2 selectivity. The
sulfonamide compounds 22 in Table 4 showed COX-2
selectivity well over 100-fold and frequently over several
1000-fold. When the same approach was applied to
methyl sulfone derivatives as in 12 (Table 4), the same
trend was observed: COX-2 selectivity improved while
maintaining in vitro COX-2 inhibitory potency. Thus,
fluoride substitution at the 5-phenyl ring should be a
valid option to improve COX-2 selectivity while main-
taining in vitro COX-2 inhibitory potency for COX-2
inhibitors of the 3(2H)furanone scaffold.28
Acu te in Vivo An tiin fla m m a tor y Activity. Car-
rageenan-induced rat foot edema (CFE)29a can be re-
garded to be an animal model for acute inflammation,
and the data obtained therefrom would not be appropri-
ate to be extrapolated to chronic inflammatory situa-
tions such as oesteoarthritis and rheumatoid arthritis.
However, CFE could still serve as an efficient animal
model to test the antiinflammtory efficacy of a drug by
According to Table 6, COX-2 inhibitors of the 3(2H)-
furanone scaffold are highly potent against AA. Some
of them are significantly more potent than currently
marketed COX-2 inhibitors rofecoxib and etoricoxib.30
Especially the antiinflammatory potency of compound
20b against AA is impressively high. Its ED50 of 0.1

798 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Shin et al.
F igu r e 3. Pharmacokinetic diagrams for intravenous and oral
administration of 1b at 10 mg/kg in male SD rats. For the
intravenous route, AUC0-24h was (7.75 ( 1.86) µg/mL × h. with
an apparent plasma half-life of 3.78 ( 1.01 h. The oral route
yielded Cmax ) 0.57 µg/mL, Tmax ) 3.17 h, and AUC0-24h ) (5.01
( 2.70) µg/mL × h.
F igu r e 2. Inhibition of contralateral foot swelling by 20b and
indomethacin as positive control against adjuvant-induced
arthritis in male Lewis rats. The mode of administration was
by oral gavage, BID, and 10 animals per group. The observed
% inhibitions on day 17 were (40 ( 10), (51 ( 9), (63 ( 9), and
(77 ( 4)%, for 0.03, 0.1, 0.3, and 1.0 mg kg-1 day-1 20b,
respectively. The positive comparator indomethacin showed
(51 ( 4)% inhibition at 0.3 mg kg-1 day-1. The errors are by
standard error.
Ta ble 7. Gastric Ulcerogenicity Data for 3(2H)furanone COX-2
Inhibitors and Indomethacin (male SD rats)a,b
ulcerogenicity
compound
placebo
oral daily dose
score ( SD
Ta ble 6. Oral Antiinflammatory Potency of COX-2 Inhibitors
and Indomethacin against Adjuvant-Induced Arthritis
QD
1.8 ( 1.0
1.8 ( 0.7
1.8 ( 1.0
1.5 ( 1.0
3.5 ( 0.7
1b
30 mg/kg, QD
30 mg/kg, QD
5 mg/kg, BID
5 mg/kg, QD
AA ED50, mg kg-1 day-1 (QD or BID)a
1f
compound
preventiveb
therapeuticc
20bc
in d om eth a cin d
1b
1f
1h
0.2 ∼ 0.3, BID
0.07, QD
0.03, QD
0.06, QD
a
Ulcerogenicity was scored according to the following scales.
score 1: apparently normal stomach; score 2: ulcers of pinpoint
size only; score 3: one or two erosion(s); score 4: erosions more
than two; and score 5: erosions with hemorrhage, in which erosion
is defined as ulcerative lesion of diameter larger than 1 mm.
1m
1r
0.1 ∼ 0.2, BID
∼0.3, QD
12h
20b
20c
celecoxib
>0.1, QD
b
0.1, BID
0.02, QD; 0.01, BID
0.03, QD
Eight animals were used for each group and the treatment
schedule throughout one week was QD, i.e., daily once, except for
0.37, BID in Lewis rats;d
0.2, QD in SD rats
the case of 20b. c Ten animals. 6/8 animals died from peritonitis
d
during the one-week treatment. The ulcerogenicity was scored only
for the two animals that survived.
r ofecoxibd
0.74, BID
0.28, QD
0.7, BID
m eloxica m d
etor icoxibd
va ld ecoxibd
in d om eth a cin
0.032, BID in Lewis rats
compound 1b was appropriately cleared from the plasma
with an apparent plasma half-life of 3.8 h after oral
administration at 10 mg/kg in male SD rats.31 Figure 3
shows pharmacokinetic data for compound 1b at 10 mg/
kg in male SD rats, where the intravenous decay
diagram overlays well with the oral counterpart.
0.2 ∼ 0.3, BID
a
b
QD ) once daily, BID ) twice daily (dosing regimen). Per-
formed using Lewis rats, 10 animals per group. c Performed using
SD rats unless noted otherwise, 5 ∼ 8 animals per group.
d
Literature data. See ref 30 for the relevant literature.
Ga str ic Sa fety. Selective COX-2 inhibitors are ex-
pected to show reduced gastric toxicity of PUB, when
compared with traditional NSAIDs. To substantiate that
the COX-2 inhibitors of the 3(2H)furanone scaffold meet
the expectation of selective COX-2 inhibition in vivo,
some of the 3(2H)furanone COX-2 inhibitors were
evaluated for gastric safety according to the literature.32
Gastric safety data for representative compounds are
provided in Table 7. In the case of indomethacin, repeat
dosing for 7 days at 5 mg kg-1 day-1 resulted in
substantial toxicity. Six out of eight SD rats died from
the GI toxicity and peritonitis during the one-week
treatment, which could be explained by overt inhibition
of COX-1. When gastric ulcerogenicity of indomethacin
was scored for the two surviving animals, a significantly
high level of gastric ulcerogenicity was observed. Con-
sidering that indomethacin is a nonselective COX
inhibitor, such a high level of gastric toxicity is not
surprising. In the case of 3(2H)furanone COX-2 inhibi-
tors 1b and 1f, their ulcerogenicity level was essentially
the same as the placebo level at 30 mg kg-1 day-1 for 7
mpk/day by the preventive model could be regarded as
one of the most potent among the COX-2 inhibitors ever
reported. It is notable that 20b is more potent than
indomethacin. Figure 2 illustrates a dose-response
diagram for AA by the preventive model, from which
the AA ED50 of 0.1 mg kg-1 day-1 was read for 20b. A
similar dose response curve by the therapeutic model
was reported for compound 1f previously.12
P h a r m a cok in etics. Sometimes AA ED50 values can
be misleading, if a drug molecule is not cleared from
the body fast enough. If a drug molecule accumulates
upon repeated dosing due to slow elimination of drug,
its apparent antiinflammatory potency is likely to be
overestimated. Such a situation is more pronounced
especially when multiple and long-term dosing is re-
quired as in AA. Pharmacokinetic studies have been
conducted for many COX-2 inhibitors of the 3(2H)-
furanone scaffold, which indicated that the antiinflam-
matory potencies presented in Table 6 should not be an
artifact from slow drug elimination. For example,

Novel Class of Cyclooxygenase-2 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 799
days. Considering their strong antiinflammatory poten-
cies in AA by the therapeutic model, the dose level of
30 mg kg-1 day-1 is considered huge for treatment of
human arthritis with 1b and 1f. Thus, the in vitro
COX-2 selectivity of the 3(2H)furanone derivatives was
further substantiated by the in vivo animal model for
gastric safety. Reflecting its moderate COX-2 selectivity,
the gastric ulcerogenicity of 20b was evaluated at a
therapeutic dose would facilitate development of an
injectable formulation without resorting to a prodrug
approach or an invasive vehicle for injection.
Exp er im en ta l Section
Biologica l a n d P h a r m a cologica l Stu d ies. Biological and
pharmacological profiles of compounds in this article were
evaluated according to the methods described as follows. The
reagents used for the following biological and pharmacological
studies were purchased mostly from Sigma. The ELISA kits
used for the in vitro COX-1/COX-2 assays were purchased from
Cayman Chemical.
relatively small daily dose, that is, 5 mg kg-1 day-1
BID.
,
An a lgesic Activity. Traditionally, the analgesic
activity of NSAIDs has been assessed by several animal
models, including carrageenan-induced hyperalgesia33
and the mouse writhing test.34 However, it may not be
appropriate to evaluate the analgesic activity of COX-2
inhibitors by the mouse writhing test. It is unlikely that
a significant level of COX-2 is induced during such a
short time frame of the analgesia induction by a noxious
chemical stimulus in the mouse writhing test. In the
meantime, COX-2 inhibitors have been reported respon-
sive to the carrageenan-induced hyperalgesia.30b When
20b was evaluated against the carrageenan-induced
thermal hyperalgesia in the SD rat, an ED50 of 0.25 mg/
kg was observed. Under the same measurement condi-
tions, indomethacin inhibited the thermal hyperalgesia
by 47% at 1 mg/kg.35a Rofecoxib has been reported to
show an ED50 of 1.03 mg/kg by the Randall-Selitto
method, a variation of the carrageenan-induced hype-
ralgesia, where the positive comparator indomethacin’s
ED50 has been reported to be 1.5 mg/kg.30b It is difficult
to make a direct comparison of analgesic potency
between 20b and rofecoxib, due to differences in the
experimental conditions employed for the two COX-2
inhibitors. Considering that similar analgesic activities
were observed with indomethacin in the two separate
experiments, however, the analgesic potency of 3(2H)-
furanone derivative 20b may be regarded comparable
to that of rofecoxib.35b Thus, 20b should be regarded as
one of the most potent COX-2 inhibitors in terms of
analgesic as well as chronic antiinflammatory activity.
In Vitr o Assa y for COX-1 In h ibition (Hu m a n Wh ole
Blood Meth od ). Venous blood was sampled from healthy
young adults using acid citrate dextrose (170 mM trisodium
citrate, 142 nM citric acid, and 222 nM dextrose) as an
anticoagulant. A total of 140 µL of blood was added to each
well of a 96-well plate, which was previously treated with 1.5
µL of an inhibitor dissolved in DMSO. After a 10-min incuba-
tion at 37 °C, 10 µL of PBS solution of 150 µM calcium
ionophore A23187 was added to each well and the plate was
incubated for another 10 min. The COX-1 reaction was
terminated by adding to each well 50 µL of 50 µg/mL in-
domethacin in 0.1% BSA buffer. Then the 96-well plate was
spun for 10 min at 1000 rpm, and the resulting TXB2 level in
the supernatant was measured by ELISA according to the
manufacturer’s instructions (Cayman cat. no. 419030).
In Vitr o Assa y for COX-2 In h ibition (Hu m a n Wh ole
Blood Meth od ). To 140 µL of human whole blood in each well
of a 96-well plate which was pretreated with 1.5 µL of DMSO
stock solution of an inhibitor was added 10 µL of 1.5 mg/mL
LPS in PBS. To the reaction mixture was added 50 µL of 50
µg/mL indomethacin in 0.1% BSA buffer after 18 h incubation
at 37 °C to terminate the reaction. Then the 96-well plate was
spun for 10 min at 1000 rpm and the resulting PGE2 level in
the supernatant was measured by ELISA according to the
manufacturer’s instructions (Cayman cat. no. 414010).
In Vitr o Assa y for COX-1/COX-2 In h ibition (Mou se
P er iton ea l Ma cr op h a ge Meth od ). The mouse peritoneal
macrophage method employed for this article is basically the
same as the above-described human whole blood method,
except for the source of the enzymes. In this murine method
mouse peritoneal macrophages which were freshly harvested
according to the literature21 were used as the enzyme source.
For the COX-2 assay, however, the COX-1 activity of the
macrophages was knocked out prior to the induction of COX-2
with LPS stimulation by incubation with 50 µM aspirin for 2
h, which was followed by washing two times with PBS to
remove the residual aspirin.
Con clu sion
The COX-2 inhibitors of the 3(2H)furanone scaffold
manifested strong in vitro COX-2 inhibitory activity
with moderate to high COX-2 selectivity, strong oral
antiinflammatory, and analgesic activity. Also COX-2
inhibitors of this class were observed to meet the due
gastric safety profile as COX-2 inhibitors. Some of the
COX-2 inhibitors, such as 1f and 20b, showed highly
potent oral antiinflammatory activities by adjuvant-
induced arthritis, positioning those COX-2 inhibitors as
the most potent COX-2 inhibitors. Obviously, the smaller
therapeutic dose is expected to elicit the smaller renal
and hepatic burden upon long-term use, at least,
through non-COX-2 dependent pathways. Also a small
therapeutic dose may be useful in reducing the intes-
tinal burden during gastrointestinal drug influx after
oral intake, through use of a fast absorption formula-
tion.36 The shorter absorptive influx through the intes-
tinal mucosa may lead to a smaller chance of COX-
associated toxicity at a given COX-2 selectivity in the
intestinal tract. Therefore, the importance of a small
therapeutic dose needs to be addressed with regard to
clinical safety of COX-2 inhibitors. Finally, a small
Ca r r a geen a n -In d u ced Ra t P a w Ed em a . Acute inflam-
mation was induced by subplantar injection of 0.05 mL of 1%
carrageenan saline solution to the right hindpaw of male SD
rats (160 ∼ 190 g) fasted overnight. A designated amount of
an inhibitor was suspended in 1% methylcellulose and admin-
istered by oral gavage to the rats 1 h prior to the injection of
carrageenan. The antiinflammatory effect by the inhibitor was
determined by % inhibition of the paw swelling measured 4 h
after the administration by using a plethysmometer (Ugobasil
7240, Comerio, Italy) according to the literature.33
Ad ju va n t-In d u ced Ar th r itis (P r even tive Mod el). Ar-
thritis was induced on day 0 by subplantar injection of 0.1 mL
of a 1% suspension of heat-killed M. butyricum mixed in the
incomplete Freund’s adjuvant into right hindpaw of male
Lewis rats (200 ∼ 210 g). Designated daily doses of an inhibitor
were administered to the randomized rats with arthritis by
oral gavage from day -1 to day 16 after the injection of the
adjuvant. The left hindpaw volumes of the control and the
treatment groups on day 17 were measured using a plethys-
mometer and averaged to determine % inhibitions of adjuvant
arthritis at the designated doses of the inhibitor according to
the literature.29b
Ad ju va n t-In d u ced Ar th r itis (Th er a p eu tic Mod el). Ar-
thritis was induced by subplantar injection of 0.1 mL of 1%
suspension of heat-killed M. butyricum mixed in the incom-

800 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Shin et al.
plete Freund’s adjuvant into the right hindpaw of male SD
rats (160 ∼ 190 g) on day 0. On day 14, arthritis-induced rats
with foot volume increase (swelling) of the left hind paw over
0.37 mL were selected and randomized for grouping. The
arthritis-induced rats were subjected to daily drug treatment
from day 14 to day 22. The foot volumes on day 22 were used
for assessment of % inhibitions of adjuvant arthritis at
designated daily dose of the inhibitor.
Ca r r a geen a n -In d u ced Th er m a l Hyp er a lgesia . Male SD
rats (140 ∼ 160 g) were subjected to overnight fasting and were
adapted to polypropylene test cage (210 × 210 × 210 mm) for
10 min before experiment. A total of 100 µL of 1% carrageenan
saline solution was intradermally injected to the left footpad
except for the vehicle control group, where saline was injected
instead. After 2 h, designated doses of an inhibitor were
administered to the rats by oral gavage. At 3 h after the
carrageenan injection, focused heat radiation (Model 336
Analgesia Meter, IITC INC/Life Science Instruments) was
applied to the left footpad and subsequent withdrawal latency
was measured. Withdrawal latency for the control and the
treatment groups was averaged to determine % inhibitions of
carrageenan-induced thermal hyperalgesia by the inhibitor
according to the literature.30b
Ga str ic Ulcer ogen icity. Male SD rats (130 ∼ 140 g) were
administered by oral gavage for 7 days at a designated daily
dose of an inhibitor suspended in 1% methylcellulose. The
animals were fasted overnight before necropsy examination.
Four hours after the last administration, the rats were
sacrificed and the stomachs were scored for ulceration. Ul-
cerogenicity of the inhibitor was scored according to the scales
as defined in the footnote of Table 7.
P h a r m a cok in etics. Male SD rats (250 ∼ 280 g) were
fasted overnight prior to dosing and until approximately 4 h
after drug administration either by oral gavage or intrave-
nously through the tail vein. Blood samples were collected from
the retro-orbital sinus at various time points over the following
24 h. Immediately after each collection, the plasma was
separated from the blood by centrifugation and stored at 4 °C
until analysis. The plasma samples were analyzed using a
reverse phase HPLC method with an appropriate internal
standard.
Molecu la r Mod elin g. The basic modeling methodology
leading to the energy-minimized complex was performed using
the Tripos force field implemented in Sybyl 6.8 Dock module
(Tripos Associates, Inc.) with a cutoff to compute the non-
bonded interactions. A human COX-2 structure complexed
with SC-558, determined by X-ray diffraction (Brookhaven
PDB entries 1CX2), was utilized.24 The conformations of the
ligand were optimized using the Tripos force field including
the electrostatic term calculated from Gasteiger and Huckel
atomic charges. The method of conjugated gradients was used
for energy minimization until the gradient value was smaller
than 0.05 kcal/mol. The lowest energy conformation obtained
in this way was superimposed onto the original position of SC-
558 in the active site channel using the Sybyl 6.8 Dock module,
after which SC-558 was deleted. Energy minimization of the
inhibitor-enzyme complex was performed with 8 Å cutoff to
compute the nonbonded interactions by the conjugate gradi-
ents method until the gradient value was smaller than 0.05
kcal/mol.
National Center for Inter-University Research Facilities,
Seoul. Detailed experimental procedures for synthesis of
compounds in this article are provided as follows.
2,2-Dim et h yl-5-{4-(m et h ylsu lfon yl)p h en yl}-4-p h en yl-
3(2H)-fu r a n on e (1a ). 1a was prepared according to the
following six-step procedure (also see Scheme 1).
Step 1: P r ep a r a tion of 1-{4-(Meth ylth io)p h en yl}-4-
h yd r oxy-4-m eth yl-2-p en tyn -1-ol (7a ). To a stirred solution
of 2-methyl-3-butyn-2-ol (416 mg) and dry THF (30 mL) under
argon and at -78 °C was added 1.6 M butyllithium in hexane
(5 mL) dropwise over 10 min. A total of 20 min later, 0.5 mL
of p-methylthiobenzaldehyde was added dropwise to the reac-
tion solution. Then, the reaction solution was allowed to warm
to the ambient temperature by removing the cold bath. After
the reaction mixture was stirred for another 2 h, the reaction
solvent was removed in vacuo, which was followed by neu-
tralization with dilute aqueous HCl. The reaction mixture was
then extracted with dichloromethane (50 mL × 3), and the
organic layer was washed with water (50 mL × 1). The organic
layer was concentrated in vacuo and the resulting residue was
subjected to column chromatographic separation (hexane/ethyl
acetate ) 1:1) to yield 724 mg of 1-{4-(methylthio)phenyl}-4-
hydroxy-4-methyl-2-pentyn-1-ol. NMR: δ 1.54 (s, 6H), 2.36 (s,
1H), 2.48 (s, 3H), 2.62 (d, 1H), 5.43 (d, J ) 5.4 Hz, 1H), 7.25
(d, J ) 6.9 Hz, 2H), 7.43 (d, J ) 6.9 Hz, 2H).
Step 2: P r ep a r a tion of 1-{(4-Meth ylth io)p h en yl}-4-
h yd r oxy-4-m eth yl-2-p en tyn -1-on e (8a ). To a stirred solu-
tion of 724 mg of 7a in 30 mL of acetone was added dropwise
451 mg of chromium trioxide dissolved in 10 mL of water and
0.25 mL of concentrated sulfuric acid. After being overnight
at the ambient temperature, the reaction solution was con-
centrated in vacuo. The resulting residue was extracted with
50 mL of water and dichloromethane (50 mL × 3). The organic
layer was concentrated in vacuo and subjected to column
chromatographic separation (hexane/ethyl acetate ) 1:1) to
yield 200 mg of 8a as a solid. mp: 102-103 °C. NMR: δ 1.67
(s, 6H), 2.41 (s, 1H), 2.54 (s, 3H), 7.28 (d, J ) 8.7 Hz, 2H),
8.02 (d, J ) 8.7 Hz, 2H). IR (cm-1): 3404, 1613, 1176, 747.
Alternatively, 8a was prepared by using PDC (pyridinium
dichromate) in place of chromium trioxide as follows. To a
stirred suspension of 7a (20 g) and Celite (30 g) in 500 mL of
dichloromethane was added PDC (40 g) portion-wise over 10
min. The reaction mixture was stirred for 12 h at room
temperature. The suspension was then filtered through a
Florisil pad (150 g) and the Florisil pad was washed with 500
mL of methylene chloride. The filtrate was washed with dilute
aqueous HCl (200 mL × 1) and the organic layer was
concentrated in vacuo. The resulting residue was chromato-
graphed as above to afford 12.3 g of 8a . Another variation of
oxidation reaction to prepare 8a from 7a was performed as
follows: A suspension of 7a (150 g) and activated manganese
dioxide (200 g) in methylene chloride (2 L) was stirred using
an overhead stirrer for 20 h at room temperature. Then the
suspension was filtered through Celite (300 g) and the filtrate
was concentrated under reduced pressure. The resulting crude
solid was recrystallized from ethyl acetate/hexane to afford 120
g of 8a .
Step 3: P r ep a r a tion of 2,2-Dim eth yl-5-{4-(m eth ylth io)-
p h en yl}-3(2H)-fu r a n on e (9a ). To a stirred solution of 8a (120
mg) in 20 mL of ethanol was added dropwise diethylamine
(0.08 mL) diluted in 7 mL of ethanol over 5 min at room tem-
perature. The reaction solution was stirred for another 1 h and
then the solvent was removed in vacuo. The resulting residue
was diluted with 50 mL of water and then extracted with
dicloromethane (30 mL × 3). The organic layer was concen-
trated in vacuo and subjected to column chromatographic
separation (hexane/ethyl acetate ) 4:1) to give 90 mg of 9a as
a solid. mp: 107-109 °C. NMR: δ 1.48 (s, 6H), 2.54 (s, 3H),
5.91 (s, 1H), 7.30 (d, J ) 8.4 Hz, 2H), 7.72 (d, J ) 8.4 Hz, 2H).
IR (cm-1): 1676, 1579, 1485, 1376, 1174, 1095, 1050, 809.
Ma ter ia ls a n d Meth od s for Ch em ica l Syn th esis. Re-
agents and solvents were obtained from commercial suppliers
and used as received, unless noted otherwise. Dry solvents
used for this study were purchased from Aldrich (Aldrich Sure/
Seal) and used without further drying manipulation. Column
chromatography was performed on silica gel 60 (Merck Catalog
1
no. 1.09385). H NMR spectra were recorded on a Varian 300
MHz NMR spectrometer (Varian Gemini2000) with TMS as
an internal standard. Melting points were measured on a
capillary melting point apparatus (Thomas Scientific) and were
reported as uncorrected. Mass spectral analyses were carried
out either at National Center for Inter-University Research
Facilities, Seoul, or at Korea Basic Science Support Center,
Taejon, South Korea. Elemental analyses were performed at
Step 4: P r ep a r a tion of 4-Br om o-2,2-d im eth yl-5-{4-
(m eth ylth io)p h en yl}-3(2H)-fu r a n on e (10a ). To a stirred
solution of 9a (58 mg) in 20 mL of carbon tetrachloride were
added 0.5 mL of acetic acid and 0.1 mL of bromine. The

Novel Class of Cyclooxygenase-2 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 801
reaction solution was stirred at room temperature for 1 h. Then
the reaction was quenched with 20 mL of saturated aq sodium
thiosulfate. Removal of the carbon tetrachloride in vacuo was
followed by extraction with dichloromethane (50 mL × 3).
Then, the organic layer was washed with water (50 mL × 1),
concentrated in vacuo, and then subjected to chromatographic
separation (hexane/ethyl acetate ) 2:1) to yield 69 mg of 10a
as a solid. NMR: δ 1.52 (s, 6H), 2.55 (s, 3H), 7.33 (d, J ) 9.3
Hz, 2H), 8.15 (d, J ) 9.0 Hz, 2H); IR (cm-1): 1704, 1594, 1574,
1486, 1348, 1184, 1069.
Step 5: P r ep a r a tion of 4-Br om o-2,2-d im eth yl-5-{(4-
m eth ylsu lfon yl)-p h en yl}-3(2H)fu r a n on e (11a ). A total of
42 mg of 10a was dissolved in 15 mL of THF and 15 mL of
ethanol, to which was added 178 mg of OXONE. The mixture
was stirred overnight at room temperature. Then the solvent
was removed in vacuo. The resulting residue was extracted
with 50 mL of water and methylene chloride (50 mL × 3). The
organic layer was concentrated under reduced pressure and
subjected to column chromatography (hexane/ethyl acetate )
2:1) to afford 45 mg of desired 11a as a solid. mp: 196-196.5
°C. NMR: δ 1.57 (s, 6H), 3.11 (s, 3H), 8.11 (d, J ) 8.7 Hz,
2H), 8.40 (d, J ) 8.7 Hz, 2H). IR (cm-1): 2928, 1703, 1559,
1270, 1148, 1076, 847. MS (EI): 346 (m).
Step 6: P r ep a r a tion of 2,2-Dim eth yl-5-{4-(m eth ylsu l-
fon yl)p h en yl}-4-(p h en yl)-3(2H)fu r a n on e (1a ). To a stirred
solution of 11a (380 mg) and tetrakis(triphenylphosphine)-
palladium(0) (40 mg) in 50 mL of toluene/15 mL of ethanol
were added 2 M aq sodium carbonate (15 mL) and benzenebo-
ronic acid (200 mg). The reaction solution was stirred at 90
°C for 12 h. Then the solvent was evaporated off under reduced
pressure. The resulting residue was extracted with 50 mL of
water and dichloromethane (30 mL × 3). Then the organic
layer was concentrated in vacuo and separated by column
chromatography (hexane/ethyl acetate ) 2:1) to give 200 mg
of 1a as a solid. mp: 192-193 °C. NMR: δ 1.58 (s, 6H), 3.07
(s, 3H), 7.27 (m, 2H), 7.36 (m, 3H), 7.84 (d, J ) 8.7 Hz, 2H),
7.93 (d, J ) 8.4 Hz, 2H). IR (cm-1): 1697, 1503, 1404, 1245,
1148, 959, 770. HRMS (EI) calcd for C19H18O4S 342.0926 (M+),
found 342.0936.
solvent was evaporated off in vacuo, and the resulting residue
was extracted with 50 mL of water and dichloromethane (30
mL × 3). The organic layer was concentrated in vacuo and
purified by column chromatography (hexane/ethyl acetate) to
give 130 mg of 3a as a solid. mp: 139-140 °C. NMR: δ 0.93
(t, J ) 7.5 Hz, 6H), 1.99 (q, J ) 7.5 Hz, 4H), 3.07 (s, 3H), 7.24
(m, 2H), 7.37 (m, 3H), 7.86 (d, J ) 8.7 Hz, 2H), 7.93 (d, J )
8.7 Hz, 2H). IR (cm-1): 2972, 1695, 1621, 1403, 1318, 1149.
HRMS (EI) calcd. for C21H22O4S 370.1239 (M+), found 370.1243.
Compounds 3b ∼ 3d were prepared by reacting bromide 11c
with appropriate arylboronic acids by Suzuki coupling as
similarly as described for compound 3a . The analytical and
physical data for compounds 3b ∼ 3d may be found in
Supporting Information.
2-{4-(Meth ylsu lfon yl)p h en yl}-3-p h en yl-1-oxa -sp ir o[4,4]-
n on -2-en -4-on e (4a ). To a stirred solution of 3-bromo-2-(4-
(methylsulfonyl)phenyl)-1-oxa-spiro[4,4]non-2-en-4-one (11d ,
77 mg; see Supporting Information for details) in 5 mL of
toluene and 15 mL of ethanol were added 15 mg of tetrakis-
(triphenylphosphine)palladium(0), 5 mL of 2 M aqueous
sodium carbonate, and 27 mg of benzeneboronic acid. Then
the reaction solution was stirred at 90 °C for 12 h. The reaction
solvent was evaporated off under reduced pressure and the
resulting residue was extracted with 20 mL of water and
dichloromethane (30 mL × 1). The organic layer was concen-
trated in vacuo and purified by column chromatography
(hexane/ethyl acetate) to obtain 57 mg of 4a as a solid. mp:
184-185 °C. NMR: δ 2.09 (m, 6H), 2.17 (m, 2H), 3.06 (s, 3H),
7.26 (m, 2H), 7.35 (m, 3H), 7.81 (d, J ) 8.1 Hz, 2H), 7.92 (d, J
) 8.1 Hz, 2H). IR (cm-1): 2925, 1695, 1591, 1403, 1150, 771.
HRMS (EI) calcd for C21H20O4S 368.1082 (M+), found 368.1092.
Compounds 4b ∼ 4d were prepared by reacting bromide 11d
with appropriate arylboronic acids by Suzuki coupling as
similarly as described for compound 4a . The analytical and
physical data for compounds 4b ∼ 4d may be found in
Supporting Information.
2-{4-(Meth ylsu lfon yl)p h en yl}-3-p h en yl-1-oxa -sp ir o[4,5]-
d ec-2-en -4-on e (5a ). To 102 mg of 3-bromo-2-{4-(methylsul-
fonyl)phenyl}-1-oxa-spiro[4,5]dec-2-en-4-one (11e; see Sup-
porting Information for details) dissolved in 3 mL of toluene
and 15 mL of ethanol were added 16 mg of tetrakis-(triph-
enylphosphin)palladium(0), 3 mL of 2 M aqueous sodium
carbonate, and 35 mg of benzeneboronic acid. The reaction
mixture was stirred at 90 °C for 12 h. The reaction mixture
was then purified by a procedure similar to that for 4a to afford
50 mg of 5a as a solid. mp: 126-127 °C. NMR: δ 1.77-1.85
(m, 10H), 3.06 (s, 3H), 7.27 (m, 2H), 7.35 (m, 3H), 7.83 (d, J )
8.1 Hz, 2H), 7.94 (d, J ) 8.1 Hz, 2H). IR (cm-1): 2936, 1693,
1621, 1404, 1318, 1147, 1129, 730. HRMS (EI) calcd for
Compounds 1b ∼ 1r were prepared similarly as the case of
compound 1a by reacting bromide compound 11a with appro-
priate arylboronic acids. The analytical and physical data for
compounds 1b ∼ 1r may be found in Supporting Information.
2-Eth yl-2-m eth yl-5-{4-(m eth ylsu lfon yl)ph en yl}-4-ph en -
yl-3(2H)fu r a n on e (2a ). To a stirred solution of 4-bromo-2-
ethyl-2-methyl-5-{4-(methylsulfonyl)phenyl}-3(2H)furanone
(11b, 200 mg; see Supporting Information for details) in 15
mL of toluene were added 40 mg of tetrakis-(triphenylphos-
phine)palladium(0), 5 mL of 2 M aq sodium carbonate, and
100 mg of benzeneboronic acid. The reaction solution was
stirred at reflux for 12 h. Then the reaction solvent was
evaporated under reduced pressure. The resulting residue was
extracted with water and dichloromethane (30 mL × 3). The
organic layer was concentrated in vacuo and purified by
column chromatography (hexane/ethyl acetate ) 2:1) to give
60 mg of 2a as a solid. mp: 115-117 °C. NMR: δ 0.96 (t, J )
7.5 Hz, 3H), 1.55 (s, 3H), 1.97 (q, J ) 7.5 Hz, 2H), 3.07 (s,
3H), 7.37 (m, 5H), 7.85 (d, J ) 8.7 Hz, 2H), 7.93 (d, J ) 8.7
Hz, 2H). IR (cm-1): 2928, 1697, 1620, 1403, 1318, 1149, 959,
769, 552. HRMS (EI) calcd for C20H20O4S 356.1082 (M+), found
356.1068.
C
22H22O4S 382.1239 (M+), found 382.1229.
5b was prepared by reacting bromide 11e with 3-meth-
ylphenylboronic acid by Suzuki coupling as similarly as
described for compound 5a . Analytical and physical data for
5b may be found in Supporting Information.
2,2-Dim eth yl-5-{2-flu or o-4-(m eth ylth io)ph en yl}-4-ph en -
yl-3(2H)fu r a n on e (16a ). 1-{2-Fluoro-4-(methylthio)phenyl}-
2-phenyl-ethan-1-one (675 mg) was dissolved in 50 mL of dry
THF at 0 °C, which was followed by portionwise addition of
sodium hydride (120 mg). The reaction mixture was stirred
for an hour at 0 °C, which was followed by dropwise addition
of 0.35 mL of R-bromoisobutyryl cyanide diluted in 20 mL of
dry THF. The reaction mixture was allowed to warm to room
temperature and stirred overnight. The reaction solvent was
removed in vacuo and the resulting residue was extracted with
50 mL of water and dichloromethane (30 mL × 3). The organic
layer was concentrated under reduced pressure and purified
by column chromatography (hexane/ethyl acetate ) 6:1) to
yield 353 mg of 16a . NMR: δ 1.54 (s, 6H), 2.49 (s, 3H), 6.90
(dd, J ) 10.8, 1.8 Hz, 1H), 7.00 (dd, J ) 8.4, 1.8 Hz, 1H), 7.22-
7.29 (m, 5H), 7.40 (dd, J ) 8.4, 7.2 Hz, 1H). IR (cm-1): 1699,
1610, 1388, 1175, 1049.
Compounds 2b ∼ 2f were prepared by reacting bromide 11b
with appropriate arylboronic acids by Suzuki coupling as
similarly as described for compound 2a . The analytical and
physical data for compounds 2b ∼ 2f may be found in
Supporting Information.
2,2-Die t h yl-5-{4-(m e t h ylsu lfon yl)p h e n yl}-4-p h e n yl-
3(2H)fu r a n on e (3a ). To a stirred solution of 4-bromo-2,2-
diethyl-2-5-{4-(methylsulfonyl)phenyl}-3(2H)furanone (11c,
300 mg; see Supporting Information for details) in 25 mL of
toluene and 10 mL of ethanol were added 40 mg of tetrakis-
(triphenylphosphine)palladium(0), 10 mL of aqueous 2 M
sodium carbonate, and 140 mg of benzeneboronic acid. Then
the reaction solution was stirred at 95 °C for 12 h. The reaction
2,2-Dim eth yl-5-{2-flu or o-4-(m eth ylsu lfon yl)p h en yl}-4-
p h en yl-3(2H)fu r a n on e (12a ). A total of 305 mg of 16a was
dissolved in 30 mL of methanol, 20 mL of THF, and 20 mL of

802 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Shin et al.
water, to which was added 1.4 g of OXONE. The reaction
mixture was stirred overnight at room temperature. Then the
solvent was removed in vacuo and the resulting residue was
extracted with 50 mL of water and dichloromethane (30 mL
× 3). The organic layer was concentrated in vacuo and purified
by column chromatography (hexane/ethyl acetate ) 3:2) to
obtain 70 mg of 12a as a solid. mp: 175-176 °C. NMR: δ 1.57
(s, 6H), 3.10 (s, 3H), 7.22 (m, 2H), 7.28 (m, 3H), 7.75 (dd, J )
8.7, 1.8 Hz, 1H), 7.77 (m, 2H). IR (cm-1): 1702, 1408, 1321,
1147. MS (FAB): 361 (MH+). HRMS (EI) calcd for C19H17FO4S
360.0832 (M+), found 360.0833.
solution of 150 mg of 19 dissolved in 15 mL of toluene and 5
mL of ethanol were added 25 mg of tetrakis(triphenylphos-
phine)palladium(0), 5 mL of 2 M aq sodium carbonate, and 70
mg of benzeneboronic acid. The reaction solution was stirred
at 95 °C for 24 h, followed by removal of the solvent in vacuo.
Then the residue was extracted with 30 mL of water and
dichloromethane (30 mL × 3). The organic layer was concen-
trated in vacuo and purified by column chromatography
(hexane/ethyl acetate ) 1:1) to give 20 mg of sulfonamide 20a
as a solid. mp: 180-182 °C. NMR: δ 1.58 (s, 6H), 4.91 (br s,
2H), 7.30 (m, 5H), 7.78 (d, J ) 8.7 Hz, 2H), 7.89 (d, J ) 8.4
Hz, 2H). IR (cm-1): 3375, 3245, 1697, 1617, 1559, 1395, 1241,
1161, 899, 752. HRMS (EI) calcd for C18H17NO4S 343.0878
(M+), found 343.0891.
Methyl sulfone compounds 12b ∼ 12h were prepared from
a suitable 1-{4-(methylthio)phenyl}-2-phenyl-ethan-1-one de-
rivative (13), as similarly as the procedure employed for
preparation of 12a . For the analytical and physical data of
compounds 12b ∼ 12h , see Supporting Information.
Sunfonamide derivatives 20b ∼ 20i were prepared by
Suzuki coupling of iodide 19 with appropriate arylboronic
acids, as similarly as employed for sulfonamide 20a . The
physical and analytical data sulfonamides 20b ∼ 20i may be
found in Supporting Information.
5-{4-(Am in osu lfon yl)p h en yl}-2,2-d im et h yl-4-p h en yl-
3(2H)fu r a n on e (20a ). 20a was prepared according to the fol-
lowing four-step synthetic procedure (also refer to Scheme 3).
5-{4-(Am in osu lfon yl)-2-flu or op h en yl}-2,2-d im eth yl-4-
p h en yl-3(2H)fu r a n on e (22a ). A total of 154 mg of 2,2-
dimethyl-5-{2-fluoro-4-(methylsulfinyl)phenyl}-4-phenyl-3(2H)-
furanone (21a ; see Supporting Information for details) was
stirred in 30 mL of trifluoroacetic anhydride at 0 °C for 2 h.
Then the volatile material was removed in vacuo, to which
was added 50 mL of 1:1 methanol and triethylamine. The
solution was stirred for 20 min at 0 °C and the solvent was
removed in vacuo. The resulting residue was dissolved in 40
mL of dichloromethane, which was followed by dropwise
addition of 15 mL of acetic acid saturated with chlorine. The
reaction solution was stirred for 20 min at 0 °C. Then the
solvent and unreacted chlorine were removed in vacuo. The
resulting residue was dissolved in 30 mL of toluene and the
toluene was evaporated off under reduced pressure. The
resulting residue was stirred overnight in 40 mL of THF and
reacted with 5 mL of ammonia water. The solvent was removed
in vacuo and the resulting residue was extracted with 30 mL
of water and dichloromethane (30 mL × 3). The organic layer
was concentrated in vacuo and then purified by column
chromatography (hexane/ethyl acetate ) 3:2) to give 42 mg of
sulfonamide 22a as a solid. mp: 79-81 °C. NMR: δ 1.58 (s,
6H), 5.01 (br s, 2), 7.20-7.31 (m, 5H), 7.69-7.76 (m, 3H). IR
(cm-1): 3340, 3274, 1591, 1526, 1328. MS (FAB): 362 (MH+).
HRMS (EI) calcd for C18H16FNO4S 361.0784 (M+), found
361.0781.
Step 1: P r ep a r a tion of 2,2-Dim eth yl-5-{4-(m eth ylsu lfi-
n yl)p h en yl}-3(2H)fu r a n on e (17a ). To a stirred solution of
2,2-dimethyl-5-{4-(methylthio)phenyl}-3(2H)furanone (9a : 10
g) in 200 mL of dichloromethane at 0 °C was added dropwise
4.8 g of m-chloroperoxybenzoic acid dissolved in 50 mL of
dichloromethane. The reaction mixture was stirred at 0 °C for
another 2 h and then concentrated in vacuo. The resulting
residue was extracted with water and dichloromethane (50 mL
× 3), followed by washing of the organic layer with aqueous
sodium carbonate. The organic layer was then subjected to
column chromatography (hexane/ethyl acetate) to give 7.5 g
of sulfoxide 17a as a solid. mp: 108-109 °C. NMR: δ 1.51 (s,
6H), 2.78 (s, 3H), 6.06 (s, 1H), 7.78 (d, J ) 8.4 Hz, 2H), 8.02
(d, J ) 8.4 Hz, 2H). IR (cm-1): 2925, 1697, 1603, 1558, 1173,
1087, 1049.
Step 2: P r ep a r a tion of 2,2-Dim eth yl-4-iod o-5-{4-(m e-
th ylsu lfin yl)p h en yl}-3(2H)-fu r a n on e (18). To a stirred
solution of 6 g of 17a in 200 mL of carbon tetrachloride and
100 mL of chloroform were added at room temperature 5.15 g
of [bis(trifluoroacetoxy)iodo]benzene and 6.5 g of iodine. The
reaction was quenched after 4 h by adding saturated aqueous
sodium thiosulfate until the characteristic color of iodine
disappeared. The quenched solution was extracted with water
and dichloromethane (100 mL × 3). The organic layer was
concentrated in vacuo and the resulting crude product was
recrystallized from hexane/ethyl acetate to yield 4.5 g of 18.
NMR: δ 1.53 (s, 6H), 2.79 (s, 3H), 7.81 (d, J ) 8.1 Hz, 2H),
8.38 (d, J ) 8.1 Hz, 2H). IR (cm-1): 2975, 2929, 1699, 1595,
1404, 1319, 1150, 969, 766, 552.
Sulfonamides 22b ∼ 22f were prepared from appropriate
2,2-dimethyl-5-{4-(methylsulfinyl)-fluoro-phenyl}-4-phenyl-
3(2H)furanone derivatives (21), by a procedure similar to that
employed for preparation of 22a .
Step 3: P r ep a r a tion of 5-{4-(Am in osu lfon yl)p h en yl}-
2,2-d im eth yl-4-iod o-3(2H)fu r a n on e (19). A total of 1.11 g
of 18 was reacted with 30 mL of trifluoroacetic anhydride for
2 h at 0 °C. After removal of the volatile solvent in vacuo, the
resulting residue was dissolved in 50 mL of 1:1 methanol/
triethylamine. The solvent was removed again in vacuo. Then
the resulting residue was dissolved in 30 mL of carbon
tetrachloride, to which was added slowly 40 mL of acetic acid
saturated with chlorine at 0 °C. After the reaction solution
was stirred at 0 °C for 20 min, the reaction solvent and the
unreacted chlorine were removed under reduced pressure. The
resulting residue was dissolved in 30 mL of toluene and the
toluene was removed again under reduced pressure. Then the
residue was the reacted with 3 mL of ammonia water in 40
mL of THF at 0 °C for 30 min. The reaction solution was
concentrated in vacuo and the resulting residue was diluted
with 30 mL of water. The aqueous solution was then extracted
with dichloromethane (30 mL × 3). The organic layer was
concentrated in vacuo and was purified by column chroma-
tography (hexane/ethyl acetate ) 3:2) to obtain 450 mg of 19
as a solid. mp: 179-180 °C. NMR: δ 1.50 (s, 6H), 5.63 (br. s,
2H), 8.05 (dd, J ) 9.0, 1.5 Hz, 2H), 8.29 (dd, J ) 9.0, 5.7 Hz,
2H). IR (cm-1): 3367, 3261, 2985, 1684, 1582, 1405, 1188, 913.
MS (FAB): 393 (MH+).
Ack n ow led gm en t. First of all, we would like to
thank Dr. Woo Young Lee of Pacific Pharmaceuticals
for in-depth discussions and understanding with regard
to the COX-2 inhibitors program in general. Also
technical support by The National Center for Inter-
University Research Facilities located in Seoul National
University is acknowledged for mass spectral analysis.
Su ppor tin g In for m ation Available: Analytical and physi-
cal data for compounds 1b ∼ 1r , 2b ∼ 2f, 3b ∼ 3d , 4b ∼ 4d ,
5b, 12b ∼ 12h , 20b ∼ 20i, and 22b ∼ 22f. This material is
available free of charge via the Internet at http://pubs.acs.org.
Note Ad d ed a fter ASAP P ostin g
Incorrect units for IC50 in Tables 1-3 were included
in the manuscript posted J anuary 14, 2004. The correct
units are µg/mL. The manuscript was reposted J anuary
20, 2004.
Refer en ces
(1) (a) Roth, S. H. Nonsteroidal Anti-inflammatory Drugs: Gastr-
opathy, Deaths, and Medical Practice. Ann. Intern. Med. 1988,
109, 353-354. (b) J ones, R. Nonsteroidal Anti-inflammatory
Drug Prescribing: Past, Present, and Future. Am. J . Med. 2001,
110, 4S-7S.
Step 4: P r ep a r a tion of 5-{4-(Am in osu lfon yl)p h en yl}-
2,2-d im eth yl-4-p h en yl-3(2H)fu r a n on e (20a ). To stirred

Novel Class of Cyclooxygenase-2 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 803
(2) (a) Singh, G. Recent Considerations in Nonsteroidal Anti-
inflammatory Drug Gastropathy. Am. J . Med. 1988, 105, 31S-
38S. (b) Wolfe, M. M.; Lichtenstein, D. R.; Singh, G. Gastrointes-
tinal Toxicity of Nonsteroidal Anti-inflammatory Drugs. N. Engl.
J . Med. 1999, 340, 1888-1899.
(3) (a) Fu, J .-Y.; Masferrer, J . L.; Seibert, K.; Raz, A.; Needleman,
P. The Induction and Suppression of Prostaglandin H2 Synthase
(Cyclooxygenase) in Human Monocytes. J . Biol. Chem. 1990,
265, 16737-16740. (b) Xie, W.; Chipman, J . G.; Robertson, D.
L.; Erikson, R. L.; Simmons, D. L. Expression of a Mitogen-
responsive Gene Encoding Prostaglandin Synthase is Regulated
by mRNA Splicing. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 2692-
2696.
J ournet, M.; Cai, D.-W.; Paluki, M.; Wu, J .; Larsen, R. D.;
Rossen, K.; Pye, P. J .; DiMichele, L.; Dormer, P.; Reider, P. J . A
Practical Synthesis of a COX-2 Specific Inhibitor. J . Org. Chem.
2000, 65, 8415-8420.
(10) Geba, G. P.; Weaver, A. L.; Polis, A. B.; Dixon, M. E.; Schnitzer,
T. J . Efficacy of Rofecoxib, Celecoxib, and Acetaminophen in
Osteoarthritis of the Knee: a Randomized Trial. J AMA 2002,
287, 64-71.
(11) Whelton, A.; Fort, J . G.; Puma, J . A.; Normandin, D.; Bello, A.
E.; Verburg, K. M.Cyclooxygenase-2-Specific Inhibitors and
Cardiorenal Function: A Randomized, Controlled Trial of Cele-
coxib and Rofecoxib in Older Hypertensive Osteoarthritis Pa-
tients. Am. J . Ther. 2001, 8, 85-95.
(4) (a) Smith, W. L.; DeWitt, D. L. Prostaglandin Endoperoxidase
H Synthase-1 and -2. Adv. Immunol. 1996, 62, 167-215. (b)
Pariet, M.; Churchill, L.; Engelhardt, G. In New Targets in
Inflammation: Inhibitors of COX-2 or Adhesion Molecules;
Bazan, N.; Botting, J .; Vane, J ., Eds.; Kluwer Academic: London,
1996; pp 23-38.
(5) Morham, S. G.; Langenbach, R.; Loftin, C. D.; Tiano, H. F.;
Vouloumanos, N.; J ennette, J . C.; Mahler, J . F.; Kluckman, K.
D.; Ledford, A.; Lee, C. A.; Smithies, D. Prostaglandin Syn-
thase-2 Gene Disruption Causes Severe Renal Pathology in the
Mouse. Cell 1995, 83, 473-482.
(6) For animal safety data of COX-2 inhibitors, see the FDA NDA
reports as follows: (a) Celecoxib: Pharmacology Review, Ap-
plication Number 20-998. (b) Rofecoxib: Pharmacology Review,
Application Number 21,042. (c) Valdecoxib: Pharmacology
Review, Application Number 21-341.
(12) Shin, S. S.; Noh, M.-S.; Byun, Y. J .; Choi, J . K.; Kim, J . Y.; Lim,
K. M.; Ha, J .-Y.; Kim, J . K.; Lee, C. H.; Chung, S. 2,2-Dimethyl-
4,5-diaryl-3(2H)furanone Derivatives as Selective Cyclooxyge-
nase-2 Inhibitors. Bioorg. Med. Chem. Lett. 2001, 11, 165-168.
(13) (a) Prasit, P.; Riendeau, D. Selective Cyclooxygenase-2 Inhibi-
tors. Ann. Rep. Med. Chem. 1997, 32, 211-220. (b) Leval, X. D.;
J ule´mont, F.; Delarge, J .; Sanna, V.; Pirotte, B.; Dogne´, J .-M.
Advances in the Field of COX-2 Inhibition. Expert Opin. Ther.
Patents 2002, 12, 969-989.
(14) (a) Tang, C.; Shou, M.; Rushmore, T. H.; Mei, Q.; Sandhu, P.;
Woolf, E. J .; Rose, M. J .; Gelmann, A.; Greenberg, H. E.; De
Lepeleire, I.; Hecken, A. V.; De Schepper, P. J .; Ebel, D. L.;
Schwartz, J . I.; Rodrigues, A. D. In-vitro Metabolism of Cele-
coxib, a Cyclooxygenase-2 Inhibitor, by Allelic Variant Forms
of Human Liver Microsomal Cytochrome P450 2C9: Correlation
with CYP2C9 Genotype and in-vivo Pharmacokinetics. Phar-
macogenetics 2001, 11, 223-235. (b) Nicoll-Griffith, D. A.;
Yergey, J . A.; Trimble, L. A.; Silva, J . M.; Li, C.; Chauret, N.;
Gauthier, J . Y.; Grimm, E.; Le´ger, S.; Roy, P.; The´rien, M.; Wang,
Z.; Prasit, P.; Zamboni, R.; Young, R. N.; Brideau, C.; Chan, C.-
C.; Mancini, J .; Riendeau, D. Synthesis, Characterization, and
Activity of Metabolites Derived from the Cyclooxygenase-2
Inhibitor Rofecoxib (MK-0966, Vioxx). Bioorg. Med. Chem. Lett.
2000, 10, 2683-2686.
(15) J oshi, A. S.; Raghavan, N.; Williams, R. M.; Takahashi, K.;
Shingu, H.; King, S.-Y. P. Simultaneous Quantification of an
Antiinflammatory Compound (Du P 697) and a Potential Me-
tabolite (X6882) in Human Plasma and Urine by High-
Performance Liquid Chromatography. J . Chromatogr. B: Biomed.
Sci. Appl. 1994, 660, 143-150.
(16) Parker, W.; Raphael, R. A.; Wilkinson, D. I. The Structure and
Synthesis of Bullatenone. J . Chem. Soc. 1958, 3871-3875.
(17) (a) Miyaura, N.; Yanagi, T.; Suzuki, A. The Palladium-catalyzed
Cross-coupling Reaction of Phenylboronic Acid with Haloarenes
in the Presence of Bases. Synth. Commun. 1981, 11, 513-517.
(b) Hoshino, Y.; Miyaura, N.; Suzuki. A. Novel Synthesis of
Isoflavones by the Palladium-catalyzed Cross-coupling Reaction
of 3-Bromochromones with Arylboronic Acids of Its Esters. Bull.
Chem. Soc. J pn. 1988, 61, 3008-3010.
(18) Herrmann, K.; Simchen, G. Synthese von Acylcyaniden (R-
Oxonitrilen) mit Cyanotrimethylsilan. Synthesis 1979, 204-205.
(19) Lee, K.-W.; Choi, Y. H.; J oo, Y. H.; Kim, J . K.; Shin, S. S.; Byun,
Y. J .; Kim, Y.; Chung, S. A Facile One-Pot Synthesis of
4,5-Diaryl-2,2-dimethyl-3(2H)-furanones. Bioorg. Med. Chem.
2002, 10, 1137-1142.
(20) (a) Kharasch, N.; Bruice, T. C. Derivatives of Sulfenic Acids. V.
1-Fluorenone Sulfur Compounds. J . Am. Chem. Soc. 1951, 73,
3240-3244. (b) Vleeschauwer, M. D.; Gauthier, J . Y. Remarkably
Mild and Simple Preparations of Sulfinates, Sulfonyl Chlorides
and Sulfonamides from Thioanisoles. Synlett 1997, 375-377.
(21) Mitchell, J . A.; Akarasereenont, P.; Thiemermann, C.; Flower,
R. J .; Vane, J . R. Selectivity of Nonsteroidal Antiinflammatory
Drugs as Inhibitors of Constitutive and Inducible Cyclooxyge-
nase. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 11693-11697.
(22) Brideau, C.; Kargman, S.; Liu, S.; Dallob, A. L.; Ehrich, E. W.;
Rodger, I. W.; Chan, C.-C. A Human Whole Blood Assay for
Clinical Evaluation of Biochemical Efficacy of Cyclooxygenase
Inhibitors. Inflammation Res. 1996, 45, 68-74.
(23) The following COX-1/COX-2 IC50’s were observed by the human
whole blood method: (a) for 1b, COX-1 IC50 ) 0.26 µg/mL and
COX-2 IC50 ) 0.005 µg/mL; (b) for 1f, COX-1 IC50 ) 0.72 µg/mL
and COX-2 IC50 ) 0.005 µg/mL; (c) for 22f, COX-1 IC50 ) 12
µg/mL and COX-2 IC50 ) 1.5 µg/mL.
(24) Kurumbail, R. G.; Stevens, A. M.; Gierse, J . K.; Mcdonald, J . J .;
Stegeman, R. A.; Pak, J . Y.; Gildhaus, D.; Miyashiro, J . M.;
Penning, T. D.; Seibert, K. Isakson, P. C.; Stallings, W. C.
Structual Basis for Selective Inhibition of Cyclooxygenase-2 by
Anti-inflammatory Agent. Nature 1996, 384, 644-648
(25) Li, J . J .; Anderson, G. D.; Burton, E. G.; Cogburn, J . N.; Collins,
J . T.; Garland, D. J .; Gregory, S. A.; Huang, H.-C.; Isakson, P.
C.; Koboldt, C. M.; Logusch, E. W.; Norton, M. B.; Perkins, W.
E.; Reinhard, E. J .; Seibert, K.; Veenhuizen, A. W.; Zhang, Y.;
Reitz, D. B. 1,2-Diarylcyclopentenes as Selective Cyclooxyge-
nase-2 Inhibitors and Orally Active Anti-inflammatory Agents.
J . Med. Chem. 1995, 38, 4570-4578. (b) Huang, H.-C.; Li, J . J .;
(7) (a) Bombardier, C.; Laine, L.; Reicin, A.; Shapiro, D.; Burgos-
Vargas, R.; Davis, B.; Day, R.; Ferraz, M. B.; Hawkey, C. J .;
Hochberg, M. C.; Kevin, T. K.; Schnitzer, T. J . The VIGOR Study
Group. Comparison of Upper Gastrointestinal Toxicity of Rofe-
coxib and Naproxen in Patients with Reumatoid Arthritis. N.
Engl. J . Med. 2000, 343, 1520-1528. (b) Silverstein, F. E.; Faich,
G.; Goldstein, J . L.; Simon, L. S.; Pincus, T.; Whelton, A.;
Makuch, R.; Eisen, G.; Agrawal, N. M.; Stenson, W. F.; Burr, A.
M.; Zhao, W. W.; Kent, J . D.; Lefkowith, J . B.; Verburg, K. M.;
Geis, G. S. Gastrointestinal Toxicity with Celecoxib vs Nonste-
roidal Anti-inflammatory Drugs for Osteoarthritis and Rheu-
matoid Arthritis: The CLASS Study: A Randomized Controlled
Trial. J AMA 2000, 284, 1247-1255. (c) Distel, M.; Mueller, C.;
Bluhmki, E.; Fries, J . Safety of Meloxicam: A Global Analysis
of Clinical Trials. Br. J . Rheumatol. 1996, 35 (Suppl. 1), 68-77.
(8) (a) Mukherjee, D.; Nissen, S. E.; Topol, E. J . Risk of Cardiovas-
cular Events Associated with Selective COX-2 Inhibitors. J AMA
2001, 286, 954-959. (b) Barkin, R. L.; Sable, K. S. Caution
Recommended for Prescribing and Administering COX1/COX2
and COX2 Specific NSAIDs. P&T 2000, 25, 196-202. (c) Cheng,
Y.; Austin, S. C.; Rocca, B.; Koller. B. H.; Coffman, T. M.;
Grosser, T.; Lawson, J . A.; FitzGerald, G. A. Role of Prostacyclin
in the Cardiovascular Response to Thromboxane A2. Science
2002, 296, 539-541. (d) Khan, K. N. M.; Paulson, S. K.; Verburg,
K. M.; Lefkowith, J . B.; Maziasz, T. J . Pharmacology of Cy-
clooxygenase-2 Inhibition in the Kidney. Kidney Int. 2002, 61,
1210-1219.
(9) (a) Prasit, P.; Wang, Z.; Briedeau, C.; Chan, C.-C.; Charleson,
S.; Cromlish, W.; Ethier, D.; Evans, J . F.; Ford-Hutchinson, A.
W.; Gauthier, J . Y.; Gordon, R.; Guay, J .; Gresser, M.; Kargman,
S.; Kennedy, B.; Leblanc, Y.; Legar, S.; Mancini, J .; O’Neill, G.
P.; Ouellet, M.; Percival, M. D.; Perrier, H.; Riendeau, D.; Rodger,
I.; Tagari, P.; Therien, M.; Vikers, P.; Wong, E.; Xu, L.-J .; Young,
R. N.; Zamboni, R.; Boyce, S.; Rupiniak, N.; Forrest, M.; Visco,
D.; Patrick, D. The Discovery of Rofecoxib, [MK966, VIOXX,
4-(4′-Methylsulfonylphenyl)-3-phenyl-2(5H)-furanone], an Orally
Active Cyclooxygenase-2 Inhibitor. Bioorg. Med. Chem. Lett.
1999, 9, 1773-1778. (b) Penning, T. D.; Talley, J . J .; Bertenshaw,
S. R.; Carter, J . S.; Collins, P. W.; Docter, S.; Graneto, M. J .;
Lee, L. F.; Malecha, J . W.; Miyashiro, J . M.; Rogers, R. S.; Rogier,
D. J .; Yu, S. S.; Anderson, G. D.; Burton, E. G.; Gogburn, J . N.;
Gregory, S. A.; Koboldt, C. M.; Perkins, W. E.; Seibert, K.
Veenhuizen, A. W.; Zhang, Y. Y.; Isakson, P. C. Synthesis and
Biological Evaluation of the 1,5-Diarylpyrazole Class of Cy-
clooxygenase-2 Inhibitors: Identification of 4-[5-(4-Methylphe-
nyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benzene Sulfonamide (SC-
58635, Celecoxib). J . Med. Chem. 1997, 40, 1347-1365. (c)
Talley, J . J .; Brown, D. L.; Carter, J . S.; Graneto, M. J .; Koboldt,
C. M.; Masferrer, J . L.; Perkins, W. E.; Rogers, R. S.; Shaffer,
A. F.; Zhang, Y. Y.; Zweifel, B. S.; Seibert, K. 4-[5-Methyl-3-
phenylisoxazol-4-yl]-benzenesulfonamide, Valdecoxib: A Potent
and Selective Inhibitor of COX-2. J . Med. Chem. 2000, 43, 775-
777. (d) Friesen, R. W.; Brideau, C.; Chan, C. C.; Charleson, S.;
Descheˆnes, D.; Dube´, D.; Ethier, D.; Fortin, R.; Gauthier, J . Y.;
Girard, Y. Gordon, R.; Greig, G. M.; Riendeau, D.; Savoie, C.;
Wang, Z.; Wong, E.; Visco, D.; Xu, L. J .; Young, R. N. 2-Pyridnyl-
3-(4-methylsulfonyl)phenylpyridines: Selective and Orally Ac-
tive Cyclooxygenase-2 Inhibitors. Bioorg. Med. Chem. Lett. 1998,
8, 2777-2782. (e) Davies, I. W.; Marcoux, J .-F.; Corley, E. G.;

804 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Shin et al.
Garland, D. J .; Chamberlain, T. S.; Reinhard, E. J .; Manning,
R. E.; Seibert, K.; Koboldt, C. M.; Gregory, S. A.; Anderson, G.
D.; Veenhuizen, A. W.; Zhang, Y.; Perkins, W. E.; Burton, E.
G.; Cogburn, J . N.; Isakson, P. C.; Reitz, D. B. Diaryl-sprio[2,4]-
heptenes as Orally Active, Highly Selective Cyclooxygenase-2
Inhibitors: Synthesis and Struture-activity Relationships. J .
Med. Chem. 1996, 39, 253-266. (c) Reitz, D. B.; Huang, H.-C.;
Li, J . J .; Garland, D. J .; Manning, R. E.; Anderson, G. D.;
Gregory, S. A.; Koboldt, C. M.; Perkins, W. E.; Seibert, K.;
Isakson, P. C. Selective Cyclooxygeanse-2 Inhibitors: Novel
4-Spiro 1,2-Diarylcyclopentenes Are Potent and Orally Active
COX-2 Inhibitors. Bioorg. Med. Chem. Lett. 1995, 5, 867-872.
(26) {4-(Methylsulfonyl)phenyl}-3(2H)furanone was actually pre-
pared and identified by 1H NMR after chromatographic separa-
tion. Unfortunately, the key intermediate spontaneously decom-
posed upon standing for a few days at ambient temperature.
Furthermore, the 3(2H)furanone intermediate appeared to be
in equilibrium with its tautomer 3-hydroxy-5-{4-(methylsulfo-
nyl)-phenyl}-furane in a 1:4 ratio in the NMR solvent (CDCl3)
by careful analysis of the initial 1H NMR spectrum. The furan
form was the major tautomer.
(27) (a) Matsushita, M.; Masaki, M.; Yagi, Y.; Tanaka, T.; Wakitani,
K. Pharmacological Profile of J TE-522, a Novel Prostaglandin
H Synthase-2 Inhibitor, in Rats. Inflammation Res. 1997, 46,
461-466. (b) J TE-522. Drugs of the Future 1998, 23, 598-601.
(c) Hashimoto, H.; Imamura, K.; Haruta, J .; Wakitani, K. 4-(4-
Cycloalkyl/aryl-oxazol-5-yl)-benzenesulfonamides as Selective
Cyclooxygenase-2 Inhibitors: Enhancement of the Selectivity by
Introduction of a Fluorine Atom and Identification of a Potent,
Highly Selective, and Orally Active COX-2 Inhibitor J TE-522.
J . Med. Chem. 2002, 45, 1511-1517.
(28) Even though only derivatives substituted with fluorine were
described in Table 4 due to space limitation, derivatives substi-
tuted with a halogen other than fluorine may be found in the
following patent. The trend observed in Table 4 was grossly
maintained also with the derivatives substituted with chlorine
and bromine. Shin, S. S.; Noh, M.-S.; Byun, Y. J .; Choi, J . K.;
Kim, J . K.; Lim, K. M.; Kim, J . Y.; Choi, Y. H.; Ha, J .-Y.; Lee,
K.-W.; Moh, J . H.; J eong, Y. S.; Chung, S.; J oo, Y. H.; Lee, C.
H.; Kang, S. H.; Park, Y.-H.; Yi, J . B. 4,5-Diaryl-3(2H)furanone
Derivatives as Cyclooxygenase-2 Inhibitors. U.S. Patent 6,492,-
416, 2002.
(29) (a) Lombardino, J . G. Nonsteroidal Antiinflammatory Drugs. In
Chemistry and Pharmacology of Drugs; Lednicer, D. Ed.; J ohn
Wiley & Sons: New York, 1985; pp 116-129. (b) Lombardino,
J . G. Nonsteroidal Antiinflammatory Drugs. In Chemistry and
Pharmacology of Drugs; Lednicer, D. Ed.; J ohn Wiley & Sons:
New York, 1985; pp 150-154.
Methylsulfonylphenyl)-3-phenyl-2-(5H)-furanone]: A Potent and
Orally Active Cyclooxygenase-2 Inhibitor. Pharmacological and
Biochemical Profiles. J . Pharmacol. Exp. Ther. 1999, 290, 551-
560. (c) For the AA data for celecoxib, see ref 9b. (d) For the AA
data for meloxicam, refer to Pairet, M.; Ryn, J . v.; Schierok, H.;
Mauz, A.; Trummlitz, G.; Engelhardt, G. Differential Inhibition
of Cyclooxygenases-1 and -2 by Meloxicam and Its 4′-Isomer.
Inflammation Res. 1998, 47, 270-276. (e) For the AA data for
etoricoxib, see Drugs of the Future 2001, 26, 346-353. (f) For
the AA data for valdecoxib, refer to Drugs of the Future 2001,
26, 133-140.
(31) Toxicokinetic studies also suggested no indication of drug
accumulation upon repeat dose in male SD rats for compounds
1b, 1f, and 20b.
(32) Lombardino, J . G. Nonsteroidal Antiinflammatory Drugs. In
Chemistry and Pharmacology of Drugs; Lednicer, D., Ed.; J ohn
Wiley & Sons: New York, 1985; pp 219-230.
(33) Riendeau, D.; Percival, M. D.; Boyce, S.; Brideau, C.; Charleson,
S.; Cromlish, W.; Ethier, D.; Evans, E. J .; Falgueyret, J .-P.; Ford-
Hutchinson, A. W.; Gordon, R.; Greig, G.; Gresser, M.; Guay, J .;
Kargman, S.; Le´ger, S.; Mancini, J . A.; O’Neil, G.; Ouellet, M.;
Rodger, I. W.; The´rien, M.; Wang, Z.; Webb, J . K.; Wong, E.;
Xu, L.; Young, R. N.; Zamboni, R.; Prasit, P.; Chan, C.-C.
Biochemical and Pharmacological Profile of a Tetrasubstituted
Furanone as
a Highly Selective COX-2 Inhibitor. Brit. J .
Pharmacol. 1997, 121, 105-117.
(34) Futaki, N.; Yoshikawa, K.; Hamasaka, Y.; Arai, I.; Higuchi, S.;
Iizuka, H.; Otomo, S. NS-398, a Novel Non-steroidal Anti-
inflammatory Drug with Potent Analgesic and Antipyretic
Effects, which Causes Minimal Stomach Lesions. Gen. Pharma-
col. 1993, 24, 105-110.
(35) (a) The observed % inhibitions against carrageenan-induced
thermal hyperalgesia were (26 ( 10), (59 ( 21), and (118 ( 30)
% for 0.1, 0.3, and 1.0 mg/kg 20b in the male SD rat by oral
gavage, respectively. In case of indomethacin as the positive
control, the observed % inhibitions were (18 ( 13) and (47 (
30)% for 0.3 and 1.0 mg/kg, respectively. N ) 8 per group. (b)
The oral analgesic activity of 20b was not directly compared with
that of rofecoxib. Considering that rofecoxib has comparable
lipophilicity to 20b as judged by calculated logP values (1.83
for rofecoxib and 2.38 for 20b by the Crippen’s method on CS
ChemDraw Ultra version 5.0), rofecoxib is a reasonable COX-2
inhibitor for comparison of analgesic activity with 20b, though.
(36) Chung, S.; et al. Bioorg. Med. Chem. Lett., submitted. When
sulfoxide analogue 21 (X ) H, R ) 3,5-difluoro; see Scheme 4
for structure) was orally administered to rats, the sulfoxide
analogue was absorbed fast and subsequently in vivo trans-
formed into methyl sulfone compound 1f according to pharma-
cokinetic studies. Repeat dose oral safety studies for both 21 and
1f indicated that rats were significantly more tolerant to 21 than
to 1f at comparable systemic exposure level of 1f, suggesting
that there should be a notable level of local gastrointestinal
toxicity occurring during the drug absorption in the gastrointes-
tinal tract. Thus, minimization of the time needed for GI drug
influx could be important for further improvement of GI safety
of COX-2 inhibitor of a given COX-2 selectivity.
(30) (a) There are not much data available to directly compare AA
ED50’s assessed by the therapeutic model using Lewis rats with
those by the preventive model. (b) For the AA data for rofecoxib,
refer to Chan, C. C.; Boyce, S.; Brideau, C.; Charleson, S.;
Cromlish, W.; Ethier, D.; Evans, J .; Ford-Hutchinson, A. W.;
Forrest, M. J .; Gauthier, J . Y.; Gordon, R.; Gresser, M.; Guay,
J .; Kargman, S.; Kennedy, B.; Leblanc, Y.; Leger, S.; Mancini,
J .; O’Neill, G. P.; Ouellet, M.; Patrick, D.; Percival, M. D.; Perrier,
H.; Prasit, P.; Rodger, I.; Tagari, P.; Therien, M.; Vickers, P.;
Visco, D.; Wang, Z.; Webb, J .; Wong, E.; Xu, L.-J .; Young, R. N.;
Zamboni, R. Riendeau, D. Rofecoxib [Vioxx, MK-0966; 4-(4′-
J M020545Z