Synthesis and structure-activity relationship of a new series of COX-2 selective inhibitors: 1,5-Diarylimidazoles

Article abstract of DOI:10.1021/jm030765s

The synthesis and the pharmacological activity of a series of 1,5-diarylimidazoles developed as potent and selective cyclooxygenase-2 (COX-2) inhibitors are described. The new compounds were evaluated both in vitro (COX-1 and COX-2 inhibition in human whole blood) and in vivo (carrageenan-induced paw edema, air-pouch, and hyperalgesia tests). Modification of all the positions of two regioisomeric imidazole cores led to the identification of 4-[4-chloro-5-(3-fluoro4-methoxyphenyl)imidazol-1-yl]benzenesulfonamide (UR-8880, 51f) as the best candidate, which is now undergoing Phase I clinical trials.

Full text of DOI:10.1021/jm030765s

J . Med. Chem. 2003, 46, 3463-3475
3463
Syn th esis a n d Str u ctu r e-Activity Rela tion sh ip of a New Ser ies of COX-2
Selective In h ibitor s: 1,5-Dia r ylim id a zoles
Carmen Almansa,* J ose´ Alfo´n, Alberto F. de Arriba, Fernando L. Cavalcanti, Ignasi Escamilla, Luis A. Go´mez,
Agust´ı Miralles, Robert Soliva, J avier Bartrol´ı, Elena Carceller, Manuel Merlos, and J ulia´n Garc´ıa-Rafanell
Research Center, Grupo Uriach. Av. Cam´ı Reial, 51-57, E-08184, Palau-Solita` i Plegamans, Spain
Received J anuary 13, 2003
The synthesis and the pharmacological activity of a series of 1,5-diarylimidazoles developed
as potent and selective cyclooxygenase-2 (COX-2) inhibitors are described. The new compounds
were evaluated both in vitro (COX-1 and COX-2 inhibition in human whole blood) and in vivo
(carrageenan-induced paw edema, air-pouch, and hyperalgesia tests). Modification of all the
positions of two regioisomeric imidazole cores led to the identification of 4-[4-chloro-5-(3-fluoro-
4-methoxyphenyl)imidazol-1-yl]benzenesulfonamide (UR-8880, 51f) as the best candidate, which
is now undergoing Phase I clinical trials.
Inhibition of cyclooxygenase (COX), one of the key
enzymes in the arachidonic acid (AA) cascade1 is the
main mechanism by which nonsteroidal antiinflamma-
tory drugs (NSAIDs) exert their antiinflammatory ac-
tion. This enzyme bis-oxygenates AA to PGG2, which is
subsequently degraded to vasoactive and inflammatory
mediators such as prostaglandins (PGs), prostacyclin
(PGI2), and thromboxane-A2. The therapeutic use of
NSAIDs, especially in chronic diseases, has revealed
their association with well-known side effects at the
gastrointestinal level (mucosal damage, bleeding)2 and,
less frequently, at the renal3 level.
After the discovery a decade ago of two COX iso-
forms,4 it was recognized5 that selective inhibitors of the
inducible form (COX-2, cytokine inducible, and ex-
pressed mainly in inflammatory cells) could provide
antiinflammatory agents devoid of the undesirable
effects associated with classical, nonselective NSAIDs.
The inhibition of COX-1, the form constitutively present
in many tissues such as stomach, kidney, and platelets,
by nonselective NSAIDS may be responsible for the
secondary effects associated with their use.
F igu r e 1.
gastrointestinal events, although it is becoming increas-
ingly apparent that they can cause nearly identical
renal effects to those observed with nonselective NSAIDs.
Cardiovascular concerns regarding the use of these
agents have recently emerged. Direct long-term studies
are needed before the relevance of this issue can be
determined.13
Although modifications of established nonselective
agents, such as lengthening the carboxyl side chain of
indomethacin6 (1), have been strategies for the design
of COX-2 selective inhibitors,7 the main effort has been
addressed to the diarylheterocycle class,8 based on early
known antiinflammatory drugs, such as flumizole. Two
compounds in this class, celecoxib9 (2) and rofecoxib10
(3), were the first COX-2 selective inhibitors to reach
the market for the oral treatment of acute pain, os-
teoarthritis, and rheumatoid arthritis, and two new
diarylheterocyclic derivatives valdecoxib (4)11a and etori-
coxib (5)12 have recently been introduced (Figure 1). A
water-soluble prodrug of valdecoxib, parecoxib,11b has
also been recently marketed for the parenteral treat-
ment of postoperative pain. In addition, some of them
are undergoing clinical trials for the treatment of certain
forms of cancer and Alzheimer’s disease.
To achieve good activity and selectivity, the diaryl-
heterocycle COX-2 selective inhibitors require the pres-
ence of a 4-methylsulfonylphenyl group attached to an
unsaturated (generally five-membered) ring in which an
additional vicinal lipophilic moiety is present. The
methylsulfonyl group can only be replaced by a SO2NH2
group, whereas the lipophilic pocket is usually occupied
by an optionally substituted phenyl ring or a bulky
alkoxy substituent. The nature of the central scaffold
is crucial for the activity, although in many instances
it does not establish well-defined electrostatic interac-
tions with any of the amino acid residues, as shown
either in the crystal structure of COX-2/inhibitor14
complexes or in different calculations, by ourselves15 and
others.16 Presumably, the highly lipophilic active site
Overall, these selective COX-2 inhibitors have fulfilled
the hope that they would exhibit a reduced risk in
* To whom correspondence should be addressed. Spain. Tel:
902471511; Fax: 938649692; e-mail: chem-almansa@uriach.com.
10.1021/jm030765s CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/25/2003

3464 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16
Almansa et al.
Sch em e 2a
a
(i) IR1, ClNEt3Bn, CH2Cl2, NaOH, 1-5 h, -20-0 °C; (ii)
TosMIC, K2CO3, DME, MeOH, reflux, 3 h.
good to high yield from imines 8, which were prepared
by reaction of 4-methylsulfonylaniline 6 with the cor-
responding aldehydes 7 under Dean-Stark conditions.
Amine 6 was obtained in 75% yield and high purity, by
reaction of 4-methylmercaptoaniline with H2O2 in the
presence of Na2WO4.21 The use of other oxidizing agents
afforded higher quantities of amino-oxidation byprod-
ucts (mainly the azoxy and nitro derivatives). Aldehydes
7 were commercially available or were synthesized from
the corresponding acids by successive conversion to the
ethyl ester, reduction (LiAlH4), and Swern oxidation.
Finally, chlorination of 9 was effected with N-chloro-
succinimide (NCS) in acetonitrile to give 4-chloroimi-
dazoles 10 (Tables 3 and 4) in 80-90% yield. This
reaction was highly regioselective, since only small
amounts of the 2-chloro (i.e., 14, around 2%) and
dichloro derivatives (i.e., 18, around 5%) were generally
observed. Both were separated from 10 either by
crystallization or flash chromatography.
4-Bromoimidazole 11 (Table 1) was obtained from 9a
in a similar way using N-bromosuccinimide (NBS) and
the dichloro derivative 18 was obtained from 9a using
2 equiv of NCS. The 4-alkylimidazoles 13a -e were
obtained by reaction of imine 8a with the corresponding
alkyl-substituted isocyanides 12. These were prepared,
as depicted in Scheme 2, by phase-transfer alkylation
of TosMIC using the conditions previously described.22
In our hands, the alkylation with the more reactive
benzyl and allyl bromides had to be performed at -20
°C in order to avoid formation of dialkylated products.
The 2-substituted imidazoles 15 and 16 (Table 1) were
regioselectively prepared on treatment of 9a with Ph-
COCl/NEt323 and aqueous formaldehyde24 respectively.
Chlorination of 16 with NCS gave 17 in 80% yield.
As shown in Scheme 3, the synthesis of the 4-chloro-
2-methyl derivative 23 required the preparation of the
sulfanyl imidazole 20. Formation of imine 19 was
effected as described above for 8, but in this case the
reaction with TosMIC provided a complex mixture of
products. However, when BetMIC20 was used, 20 was
isolated in 59% yield. Alkylation of 20 with BuLi and
IMe in THF25 afforded 21 in 36% yield, which was
finally transformed into 23 by oxidation with m-chlo-
roperbenzoic acid (MCPBA) and chlorination with NCS.
The 1-(4-fluorophenyl)imidazoles (II) were prepared,
as shown in Scheme 4, using similar methods as
described above for their regioisomers. The 2-substitut-
ed derivatives 32-34 were obtained from 4-methylsul-
fanylphenylimidazole 27, which was prepared upon
successive imine formation and cyclization with Bet-
MIC. Reaction of 27 with LDA and the corresponding
electrophile (NCS, NBS or MeI) gave compounds 29-
F igu r e 2.
Sch em e 1a
a
H2O2, Na2WO4, H2O, 65 °C, 1.5 h; (ii) toluene, Dean-Stark,
110 °C, 24-48 h; (iii) TosMIC, K2CO3, DME, MeOH, reflux, 3 h;
(iv) NCS, acetonitrile, 81 °C, 24 h.
requires a given (low) polarity of the central scaffold.
Furthermore inclusion of water molecules and the
rearrangement of some lateral chains inside the active
site makes general predictions using molecular model-
ing very difficult.15
Regioisomeric imidazoles have been explored previ-
ously as replacements of the central heterocyclic core.
The 4,5-diarylimidazoles III17 showed diminished po-
tency vs 2, while the 1,2-diarylimidazoles IV were
shown to be potent COX-2 selective inhibitors only when
a trifluoromethyl group was present in the 4 position.18
None of them, however, was pursued in further develop-
ment.
As a new approach to the diarylheterocycle class of
COX-2 inhibitors, we report here the structure-activity
relationships (SAR) of new 1,5-diarylimidazoles I and
II (Figure 2), in which a defined combination of sub-
stituents confers appropriate polarity and charge dis-
tribution for good activity.
Ch em istr y
The key step in the synthesis of imidazoles I and II
involves reaction of tosylmethyl isocyanide (TosMIC)
with an imine, in which the presence of the electron-
withdrawing 4-methylsulfonyl group allows the reaction
to take place in good yield. The reaction is performed
in the presence of K2CO3, which initiates a two-step
anionic 1,3-dipolar cycloaddition.19 In cases where the
reaction needs to be performed with 4-methylsulfanyl
imines, and thus no electron-withdrawing group is
present, the more reactive benzotriazol-1-yl-methyl iso-
cyanide (BetMIC)20 must be used.
As shown in Scheme 1, 1-(4-methylsulfonylphenyl)
imidazoles 9 were obtained using these conditions in

New COX-2 Selective Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16 3465
Sch em e 3a
Sch em e 5a
a
(i) Toluene, Dean-Stark, 110 °C, 24-48 h; (ii) BetMIC (23),
KOBut, DMSO, 75 °C, 1 h; (iii) LDA, IMe, THF, -20-20 °C, 1.5
h; (iv) MCPBA, CH2Cl2, 20 °C, 2 h; (v) NCS, acetonitrile, 81 °C,
24 h.
Sch em e 4a
a
(i) NH2But/DME, reflux, 4 h; (ii) KOH/MeOH/H2O, 100 °C, 4
h; (iii) 4-methoxy-3-fluorobenzaldehyde, toluene, Dean-Stark, 110
°C, 24 h; (iv) TosMIC, K2CO3, DME, MeOH, reflux, 3 h; (v) NCS,
acetonitrile, 81 °C, 24 h; (vi) 6 N HCl, reflux, 3 h.
F igu r e 3.
chlorination with NCS and deprotection of the sulfon-
amide group yielded 51f in 40% overall yield.
Some triazoles 52a -c and 53a ,b (Figure 3) were also
prepared following a route previously described for
similar compounds.28
a
(i) Toluene, Dean-Stark, 110 °C, 24-48 h; (ii) BetMIC (23),
KOBut, DMSO, 75 °C, 1 h or TosMIC (24), K2CO3, DME, MeOH,
reflux, 3 h; (iii) LDA, NCS (or NBS or IMe), THF, -20 to 20 °C,
1.5 h; (iv) MCPBA, CH2Cl2, 20 °C, 2 h; (v) MCPBA, CH2Cl2, 25
°C, 2 h.
Resu lts a n d Discu ssion
31, which were finally oxidized to afford 32-34, respec-
tively. The remaining compounds in Table 2 were
obtained from imidazole 28. This was converted to 35
and 40 as described above for 16 and 15, respectively.
Reaction of 35 with diethylaminosulfur trifluoride (DAST)
afforded compound 36 and oxidation of 35 with MnO2
gave a mixture of aldehyde 37 and methyl ester 38.
Aldehyde 37 was converted to nitrile 39 on treatment
with hydroxylamine-O-sulfonic acid in the presence of
pyridine.26 Compounds 41-44 were obtained following
similar methods to those described previously.
Sulfonamides 51 (Table 5) were initially prepared as
described in the experimental part for compound 51a .
The sulfonamide functionality was introduced using the
transformation of sulfinyl derivatives previously de-
scribed for related compounds.27
A more direct method was developed for the synthesis
of sulfonamide 51f, as described in Scheme 5. Treatment
of 4-(acetylamino)sulfonyl chloride 45 with tert-butyl-
amine yielded 46, which on deprotection with potassium
hydroxide gave 47. Reaction of 47 with 4-methoxy-3-
fluorobenzaldehyde gave imine 48, which was cyclized
with TosMIC to afford imidazole 49. Regioselective
The activity and selectivity of the new compounds was
first evaluated in vitro by obtaining their IC50 in human
cell lines expressing COX-1 and COX-2. The COX-2
activity was also determined in a human whole blood
(HWB) assay. Percent inhibition at 10 µM was initially
determined and active compounds were also tested at
1 µM. All compounds showing more than 60% inhibition
at 10 µM were tested orally in the rat carrageenan-
induced paw edema assay (CPE) at 10 mg/kg (Tables
1-5).29 The most potent derivatives were also tested in
the carrageenan-induced air pouch model30 to estimate
PG production at 1 mg/kg po and in the hyperalgesia
model31 at 3 mg/kg po (Table 6).
The 1-methylsulfonylimidazole I, in which a 4-fluo-
rophenyl group was present in position 5, was initially
selected to explore the SAR around the imidazole
nucleus (Table 1). Whereas the parent unsubstituted
compound (9a ) was completely inactive, a substantial
increase in activity was seen on introduction of small
substituents in position 4. 4-Halo derivatives 10a and
11 showed remarkable potency and selectivity in both
in vitro tests, similar to that of reference compounds 2
and 3. The chloro derivative 10a also showed good

3466 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16
Ta ble 1. 5-(4-Fluorophenyl)-1-(4-methylsulfonyl)imidazoles 9-23
Almansa et al.
c
h cell lines IC50
HWB COX-2,d % inhib
COX-1
(µM)
COX-2
(µM)
CPEe % inhib,
10 mg/kg
compd
R1
R2
mp,a °C
formulab
16H13FN2O2S
16H12ClFN2O2S
16H12BrFN2O2S
17H15FN2O2S‚0.25H2O
18H17FN2O2S‚0.5H2O
10 µM
1 µM
9a
H
Cl
Br
Me
Et
H
H
H
H
H
H
H
H
Cl
151-155
167
148
143
168-169
114-115 C19H17FN2O2S‚0.25H2O
173
103
C
C
C
C
C
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
56.9
100
100
86.5
3.8
-
NTf
41.2
23.0g
35.1g
NT
NT
NT
NT
NT
NT
NT
NT
NT
10a
11
13a
13b
13c
13d
13e
14
15
16
17
18
23
1
0.014
0.036
0.144
>10
89.0
85.2
34.5
-
-
-
-
-
-
-
-
-
-
Allyl
>10
3.4
CH2Ph
C23H19FN2O2S‚0.25H2O
21H13FN2O2S
16H12ClFN2O2S
23H17FN2O3S‚0.75H2O
>10
60.5
64.4
-
0
27.4
2.4
Pent
H
H
H
Cl
Cl
Cl
C
C
C
>10
162-166
225-227
>10
COPh
>10
CH2OH 208-211 C17H15FN2O3S‚0.75H2O
CH2OH 199-202 C17H14ClFN2O3S‚0.5H2O >10
>10
>10
Cl
194
188-199
C
C
16H11Cl2FN2O2S
17H14ClFN2O2S‚HCl
>10
>10
0.002
5.1
>10
>10
0
Me
>10
27.5
100
96.4
100
NT
indomethacin
celecoxib
rofecoxib
0.009
0.079
0.012
67.9
67.0
84.1
66.2
35.7
33.3
2
3
a
b
Melting points correspond generally to chromatographed compounds. Elemental analyses for C, H, N, and S were within 0.4% of
the theoretical values indicated. c IC50 for human COX activity in U-937 cells (COX-1) and 143982 cells (COX-2). Assays were performed
d
in duplicate. Percentage inhibition of COX-2 activity in human whole blood (at 10 and 1 µM). Assays were performed in duplicate.
e Carrageenan paw edema test (10 mg/kg, po). f NT: not tested. Percentage of inhibition at 30 mg/kg.
g
Ta ble 2. 1-(4-Fluorophenyl)-5-(4-methylsulfonyl)imidazoles 28-44
c
h cell lines IC50
HWB COX-2,d % inhib
COX-1
(µM)
COX-2
(µM)
CPEe % inhib,
10 mg/kg
compd
R1
R2
mp,a °C
formulab
16H13FN2O2S
16H12ClFN2O2S‚0.25H2O
16H12BrFN2O2S
17H15FN2O2S‚0.5H2O
C17H15FN2O3S
C17H14F2N2O2S‚0.75H2O
17H13FN2O3S‚0.5H2O
18H15FN2O4S‚1.25H2O
17H12FN3O2S‚0.25H2O
23H17FN2O3S‚0.25H2O
16H12ClFN2O2S
16H12BrFN2O2S
16H11Cl2FN2O2S
17H14ClFN2O2S‚0.25H2O
10 µM
1 µM
28
32
33
34
35
36
37
38
39
40
41
42
43
44
H
Cl
Br
H
H
H
H
H
H
H
H
H
H
Cl
Br
Cl
Me
133-134
218-220
207-208
160-162
211-212
152-154
198
C
C
C
C
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
63.4
100
92.8
68.0
8.1
16.8
70.9
39.1
100
95.0
15.3
36.0
51.6
42.6
-
NTf
20.4
40.4
21.7
NTf
NTf
27.5
NTf
28.5
9.8g
NTf
NTf
NTf
NTf
0.014
0.048
0.051
>10
85.6
56.2
-
-
-
-
-
74.0
59.1
-
-
-
-
CH3
CH2OH
CH2F
CHO
COOMe
CN
>10
C
C
C
C
C
C
C
C
> 1
192-194
192
>10
0.051
0.030
>10
COPh
H
225-226
133-134
187-189
211-212
201-205
H
Cl
Cl
>10
>10
>10
a-g
See footnotes a-g of Table 1.
activity in vivo, similar to 2 and 3 and superior to that
of 11. Among the 4-alkyl derivatives, only 13a showed
some activity, but inferior to that of 10a . Elongation of
the chain led to a complete loss of activity (13b-e).
Substitution in the 2-position of the imidazole ring was
clearly detrimental (14-16, 17, 18, 23).
The SAR around the imidazole nucleus of 5-methyl-
sulfonylimidazole II was also performed with a 4-fluo-
rophenyl group in position 1 (Table 2). Again, the parent
unsubstituted compound 28 was devoid of activity, but
on introduction of small substituents in position 2 (R to
the 4-fluorophenyl group) activity increased substan-
tially. The 2-chloro derivative, 32, was the most potent
in vitro, showing a similar trend to that observed in the
case of its regioisomer 10a . The bromo derivative 33 as
well as the methyl 34 and the 2-carbonitrile 39 also
exhibited high in vitro potency. In this case, the activity
in the CPE test was superior for the bromo- and
carbonitrile derivatives 33 and 39. The replacement of
the methyl group with hydroxymethyl (35), fluoromethyl

New COX-2 Selective Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16 3467
F igu r e 4. Proposed binding mode of compound 51f inside the active site of COX-2, resulting from a molecular dynamics simulation.
The most important amino acids are shown together with their respective numbers. The inhibitor only forms hydrogen bonds
with the enzyme through its sulfonamide moiety (to Gln 192, Arg 513, and Phe 518). Two structural water molecules hydrating
the imidazole ring and bridging Tyr 385 to Ser 530 are marked with arrows.
(36), and methyl ester (38) led to a complete loss of
activity, while aldehyde 37 retained some potency. As
in the case of isomers I, the introduction of substituents
in the position R to the 4-methylsulfonylphenyl group
led to a complete loss of activity (41-44).
An interesting result was obtained on introduction of
a benzoyl group in position 2 (40), which provided
excellent activity in vitro, although low potency in vivo.
This compound probably occupies the same space in the
active site of COX-2 as other related compounds,32 in
which a large lipophilic residue and a carbonyl group
are present in similar positions.
The introduction of a third nitrogen atom (Figure 3)
in the position occupied by the chlorine of 10a was
explored with triazole 53a , which was devoid of activity.
The same result was obtained with the regioisomeric
triazole 52a . Compounds 52b and 52c, which can be
regarded as aza analogues of 34 and 32, respectively,
were also completely inactive. Clearly the introduction
of a third nitrogen atom is detrimental, either because
of its electronegative nature or as a result of its
hydrogen bond acceptor properties.
Methyl triazole 53b was also inactive at 10 µM.
Although it is sterically similar to the potent COX-2
inhibitor valdecoxib (4), their different H-bonding ca-
pacities or electronic features may explain their differ-
ent activity. In the case of both regioisomeric imidazoles
23 and 44, the introduction of a methyl substituent
vicinal to the methylsulfonylphenyl group, provided also
completely inactive compounds. However, introduction
of methyl groups in similar positions of rofecoxib (i.e.,
in the lactone methylene) maintains potency.33 Clearly
this indicates the danger of drawing up general rules
to explain the main interactions and derive a generic
pharmacophore for COX-2 selective inhibitors. Thus,
although many compounds have similar or even identi-
cal topologies, their electronic distributions may play a
key role in determining their activities at two levels.15,16
First, the active site of COX-2 is very flexible, and some
key amino acids could shift the orientation of their
lateral chains depending on the inhibitor, leading to
changes in the interaction pattern. Second, the active
site is at the bottom of a long and buried channel,
sterically restricted and mainly lipophilic, and therefore
the free energy of desolvation of the inhibitors as well
as their van der Waals interaction with the amino acids
lining the channel wall must also have an impact on
activity.
This has recently been studied for both celecoxib
derivatives and 1,5-diarylimidazoles.15 In the case of
celecoxib, elimination of the trifluoromethyl group leads
to a dramatic decrease of activity mainly due to a change
in desolvation energy (the presence of CF3 reduces the
desolvation of the molecule upon binding) and to a
decrease of the drug-protein van der Waals interaction.
Also in the case of 1,5-diarylimidazoles, elimination of
the chlorine atom on the heterocycle again has a
negative impact on biological activity since the presence
of this halogen decreases the desolvation of the molecule
and increases the drug-protein van der Waals contacts.
The predicted binding mode of 51f (Figure 4) is
similar to that of other compounds within this series.15
It shows a very strong van der Waals contact with many
of the amino acids lining the bottom of the active site,
but the inhibitor only hydrogen bonds to the enzyme
through its phenylsulfonamide moiety (the oxygen
atoms with Phe518 and Arg513, the amino group with
Gln192).

3468 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16
Ta ble 3. Phenyl-Substituted Sulfones of Formula 10
Almansa et al.
c
h cell lines IC50
HWB COX-2,d % inhib
COX-1 COX-2
CPEe % inhib,
10 mg/kg
compd
X
yield % mp,a °C
formulab
16H12ClFN2O2S
16H12ClFN2O2S‚HCl
16H12ClFN2O2S‚HCl‚0.5H2O
16H13ClN2O2S‚0.25H2O
(µM)
>10
>10
100
>10
(µM)
10 µM
1 µM
10a
10b
10c
10d
10e
10f
10g
10h
10i
4-F
3-F
2-F
H
4-Cl
4-Me
80
62
65
51
55
76
63
65
60
80
73
76
65
15
34
80
85
65
70
45
61
74
73
48
73
71
76
75
70
76
84
60
75
60
80
167
C
C
C
C
C
C
C
C
0.014
0.065
0.028
0.123
0.018
0.016
0.011
0.004
10.0
100
100
100
96
89.0
51.9
66.2
82.6
88.4
90.5
100
100
0
41.2
22.8
21.5
29.2
9.1
21.6
33.4
16.3
NTf
NTf
NTf
0
17.0
21.7
12.3
NTf
15.3
NTf
24.2
8.6
177-179
177-178
145-146
222-223
182
16H12Cl2N2O2S‚HCl‚0.25H2O >10
100
100
100
100
NTf
NTf
NTf
91.9
100
88.5
97.3
NTf
71.6
NTf
100
100
100
100
100
100
100
100
100
98.9
100
100
95.2
92.6
99.8
0
17H15ClN2O2S‚0.25H2O
17H15ClN2O3S‚0.5H2O
18H17ClN2O3S‚HCl
>10
1.3
4-OMe
4-OEt
205
207
>10
>10
>10
>10
>10
>10
4-OPr
161-163 C19H19ClN2O3S‚HCl‚0.5H2O
153
136-138
151
10j
4-OPri
C
C
C
C
C
C
19H19ClN2O3S‚0.25H2O
17H12ClF3N2O3S‚HCl
19H19ClN2O2S
>10
10.0
0.100
NTf
NTf
60.0
70.0
49
10k
10l
4-OCF3
4-Pr
10m
10n
10o
10p
10q
10r
10s
10t
10u
10v
10w
10x
10y
10z
4-Pri
161
19H19ClN2O2S‚0.25H2O
17H15ClN2O2S2‚0.75H2O
18H17ClN2O2S2‚0.5H2O
0.039
0.011
0.053
4-SMe
4-SEt
4-SO2Et
4-NH2
4-AcNH
4-NEt2
2,4-di-F
4-OMe-2-F
3,4-di-Cl
4-OMe-3-F
4-Me-3-F
4-OMe-3-Me
4-Me-3-OMe
216-220
181-185
-
0.9
>10
>10
>10
>10
>10
23.8
0.11
9.0
>10
>10
>10
>10
4.0
1.9
>10
>10
>10
>10
>10
>10
>10
76.7
NTf
NTf
NTf
9.7
79.6
100
93.8
79
75.1
100
77.1
-
64.6
84.1
47.7
96.5
97.4
37.3
NTf
NTf
C18H17ClN2O4S2
>10
0.187
>10
170
C
C
C
C
C
C
C
C
C
C
C
16H14ClN3O2S‚H2O
18H16ClN3O3S‚0.5H2O
20H22ClN3O2S‚0.5H2O
16H11ClF2N2O2S
17H14ClFN2O3S‚HCl‚0.25H2O
16H11Cl3N2O2S
238-241
207-209
183-184
176-198
156-157
196
176
198
178
0.080
0.007
0.015
0.006
0.004
0.006
0.013
0.015
0.011
0.027
0.008
0.007
0.025
0.016
0.007
33.0
23.5
24.2
22.6
21.4
9.2
17H14ClFN2O3S
17H14ClFN2O2S‚HCl‚0.5H2O
18H17ClN2O3S‚HCl
18H17ClN2O3S‚HCl
17H14Cl2N2O2S‚HCl
C18H17Cl2N3O2S
17H14Cl2N2O3S
18H16Cl2N2O3S‚0.5H2O
18H16ClFN2O3S‚0.75H2O
17H14ClFN2O3S
10a a 4-Cl-3-Me
10a b 4-NMe2-3-Cl
10a c 4-OMe-3-Cl
10a d 4-OEt-3-Cl
10a e 4-OEt-3-F
181-182
169
NTf
33.3
31.4
2.0
14.1
25.0
16.5
NTf
NTf
191-192
218-219
199-203
178-179
230
C
C
C
C
C
C
C
10a f
4-F-3-OMe
10a g 4-OMe-3,5Cl
10a h 3,5-di-OEt
17H13Cl3N2O3S
20H21ClN2O4S‚HCl
16H11ClF2N2O2S
227-231
196
>10
10a i
3,5-diF
0.075
61.6
a-f See footnotes a-f of Table 1.
When the most potent compounds identified so far
(10a , 11, 32, 33, and 39), were submitted to other tests
in vivo (Table 6), compound 10a emerged as the most
potent derivative. For this reason the 4-chloro-1-meth-
ylsulfonylphenylimidazole nucleus was selected for study-
ing the SAR around the aryl group in position 5 (Tables
3 and 4), the only portion in the molecule yet to be
explored. Similar analogues of compounds 32 and 39
were also prepared, but none of them showed increased
potency in vivo (results not shown).
Changing the fluorine position of 10a decreased
potency both in vitro and in vivo (10b,c). The unsub-
stituted derivative 10d was less potent, as confirmed
in the air-pouch model (Table 6). The 4-chloro and
methyl derivatives 10e and 10f, although highly potent
in vitro, were less so in vivo. The 4-methoxy derivative
10g was potent in the CPE test, but was less selective
than 10a . However elongation of the alkoxy chain to
ethoxy (10h ) increased selectivity, possibly caused by
the occupation of an additional pocket in the COX-2
enzyme by the longer alkyl chain.15 This pocket may be
easier to achieve in the case of these 4-chloroimidazoles
than in other related structures such as the pyrazoles
represented by Celecoxib, where the ethoxy group only
produced a loss in potency.9
Further elongation of the alkoxy chain (10i-j), as well
as introduction of fluorine atoms (10k ), blocked all
activity. Maintaining the same chain length, the oxygen
atom was replaced by sulfur (10n ,o), sulfone (10p ), CH2
(10l), and CHMe (10m ). The effect of an electron-
withdrawing sulfone (10p ) was clearly negative, and
although the remaining compounds retained potency in
vitro, they were all less potent in vivo than their alkoxy
counterparts.
The poor in vitro potency of the 4-amino and acetyl-
amino derivatives 10q and 10r was improved upon
substitution (10s), although this compound failed to
show sufficient activity in vivo (Table 6).
Disubstitution of the phenyl ring was initially ex-
plored by introduction of fluorine in the 2-position of
10a and 10g. In the first case (10t), the in vivo activity
was lost and in 10u selectivity decreased. Among the
3,4-disubstituted derivatives (10v-a f) highly potent
compounds were identified, which were submitted to
additional tests in vivo (Table 6). The trisubstituted and
the 3,5-disubstituted derivatives were poorly active both
in vivo (10a g) and in vitro (10a h ,a i).
Replacing the phenyl ring in position 5 by other
heterocycles was also explored (Table 4). The naked
3-pyridyl and 4-pyridyl derivatives 10a j and 10a p

New COX-2 Selective Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16 3469
Ta ble 4. Aryl Sulfones of Formula 10
a-f See footnotes a-f of Table 1.
showed poor activity in the whole cell assay but a
surprisingly high activity in the HWB test. The 4-meth-
yl, 4-chloro and 4-ethoxy derivatives 10a k -a m showed
good activity in the CPE test, but not in the air-pouch
or hyperalgesia models. Compound 10a q was the most
potent and selective of all the heterocycle variations
studied.
We next replaced the SO2Me group by a SO2NH2,
which has been shown to improve the activity in vivo
by enhancing bioavailability.9 As shown in Table 5 the
selectivity of the sulfonamide derivatives was generally
lower than their sulfone counterparts. Three of them
(51a ,b,g) showed a diminished potency in the HWB test,
which may be explained by a high percentage of binding
to serum proteins. Although some compounds (51i-j)
failed to improve activity in vivo, others were highly
potent COX-2 inhibitors both in vitro and in vivo (Tables
5 and 6).
pounds.34 For this reason early pharmacokinetic evalu-
ation in rats was performed for the most active com-
pounds (Table 6). The 4-fluoro derivatives 10a and 51a
showed long half-lives in this species, which were hardly
diminished in the unsubstituted 10d and 51b. Introduc-
tion of metabolizable methyl groups decreased the half-
lives (10f and 51c), but in vivo activity was also reduced.
The pyridyl derivative 10a l also had a long half-life,
suggesting that N-oxidation was not an effective elimi-
nation route. It is known that substitution of aromatic
rings with electron-releasing groups such as alcoxy or
alkylamino increases metabolic elimination, either by
increasing the potential for hydroxylation of such rings
or by giving dealkylated products, which are more easily
eliminated. These assumptions were accomplished in
our series, where introduction of electron-releasing
groups gave compounds with shorter half-lives (10w ,
10a b, 10a c, 10a q, 51d , 51f). Compounds 10a b and
10a c were discarded due to their inferior in vivo activity
and the remaining four compounds were submitted to
in vitro evaluation of cytochrome inhibition.35 Com-
pound 10a q was discarded due to high CYP3A4 inhibi-
The highly lipophilic and insoluble character of the
diarylheterocycle class of COX-2 inhibitors together with
lack of hepatic elimination routes may explain the
prolonged half-lives described for some of these com-

3470 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16
Ta ble 5. Phenyl-Substituted Sulfonamides of Formula 51
Almansa et al.
c
h cell lines IC50
HWB COX-2d % inhib
COX-1
(µM)
COX-2
(µM)
CPEe % inhib,
10 mg/kg
compd
X
mp,a °C
formulab
10 µM
1 µM
51a
51b
51c
51d
51e
51f
51g
51h
51i
4-F
H
4-Me
4-OEt
3,4-Cl
4-OMe-3-F
4-F-3-OMe
4-OEt-3-F
4-OMe-3-Cl
4-OEt-3-Cl
4-Cl-3-Pyg
223
235
255
265
251
211
191
236
C
C
C
C
C
C
C
C
C
C
C
15H11ClFN3O2S‚0.25H2O
15H12ClN3O2S‚0.25H2O
16H14ClN3O2S‚0.25H2O
17H16ClN3O3S‚0.25H2O
15H10Cl3N3O2S‚0.5H2O
16H13ClFN3O3S
16H13ClFN3O3S
17H15ClFN3O3S
16H13Cl2N3O3S
17H15Cl2N3O3S
3.1
5.1
2.7
2.2
0.6
0.018
0.009
0.003
0.002
0.002
0.005
0.008
0.020
0.005
0.007
0.042
61.4
59.5
100
100
100
100
79.3
100
99.9
99.9
97.6
NTf
NTf
76.1
100
93.5
100
NTf
57.0
-
38.6
36.3
32.6
29.2
29.6
31.0
14.3
30.3
24.0
18.5
18.1
3.3
> 1
76.0
1.2
15.5
> 10
245
272
276-277
51j
51k
26.8
19.6
14H10Cl2N4O2S
g
a-f See footnotes a,b,c,d,e of Table 1. X-Ph ) 4-Chloro-3-pyridyl.
Ta ble 6. In Vivo Data of Selected Compounds
Ta ble 7. Pharmacological Data of Selected Compounds
air pouch,a
% inhib
(1 mg/kg, po)
hyperalgesiaa
% inhib
(3 mg/kg, po)
adjuvant
51Cr leakagec
100 mg/kg, bid,
5 days (%)
ratb half-life
(h)
COX-2 HWB, arthritis,
UR
compd
IC50 (nM)
% inhibb
2c
93.4
95.3
98.1
83.2
65.5
64.1
18.5
75.6
89.5
43.0
71.1
15.1
66.6
91.4
70.1
50.2
NTe
91.4
45.0
55.2
35.5
79.0
93.2
96.3
62.3
98.9
98.2
58.9
NTe
NTe
21
86.7
83.4
NTe
NTe
NTe
64.1
NTe
NTe
80.0
63.8
66.6
20.4
NTe
NTe
62.9
69.4
69.9
NTe
NTe
NTe
NTe
NTe
NTe
NTe
NTe
82.8
83.4
38.0
indomethacin (1)
celecoxib (2)
rofecoxib (3)
10w
51d
51f
640
600
280
156
27
94
89
87
86
97
93
7.5d
0.7
0.6
0.4
0.5
0.9
3d
10a
10d
10f
17
1.7
10s
NTe
NTe
NTe
2.2
66
10x
10y
10w
10a b
10a c
10a f
10a k
10a l
10a m
10a p
10a q
11
a
Inhibition of contralateral paw swelling after oral treatment
for 28 days (1 mg/kg od). Percentage of 51Cr secreted in feces.37
b
c Dose: 5 mg/kg, bid.
1.2
3.6
NTe
NTe
26
showed 90% binding to human serum proteins, good
bioavailability in rats (53%) and dogs (81%) and phar-
macokinetic behavior suggestive of one daily dosing in
man. Compound 51f (UR-8880)37 has successfully com-
pleted preclinical development as it is now finishing
phase I clinical trials for the treatment of osteoarthritis
and pain.
In summary, we have described the structure activity
relationships of two isomeric imidazole cores, which has
led to the identification of a highly selective COX-2
inhibitor, 51f. This compound showed high potency in
all inflammation tests assayed together with good
pharmacokinetics, which led to its selection as a clinical
candidate.
NTe
NTe
3.5
NTe
NTe
NTe
NTe
NTe
17
13a
32
33
39
51a
51b
51c
51d
51f
26
5.2
2.2
2.9
51g
3.8
a
b
See Experimental Section. Half-life in rat after iv adminis-
tration of compounds in DMSO solution at 1 mg/kg. c Celecoxib.
Exp er im en ta l Section
Rofecoxib. e Not tested.
Melting points were determined with a Mettler FP 80
central processor melting point apparatus and are uncorrected.
13C (75 MHz) NMR spectra and 1H (300 MHz) NMR spectra
were recorded on a Bru¨cker Avance DPX-300 spectrometer.
They are reported in ppm on the δ scale, from the reference
indicated. Combustion analyses were performed with a Carlo
Erba 1106 analyzer. Liquid chromatography was performed
with a forced flow (flash chromatography) of the indicated
solvent system on SDS silica gel Chromagel 60 ACC (230-
400 mesh). Analytical thin-layer chromatography (TLC) was
performed with Macherey-Nagel 0.25 mm silica gel SIL G-25
plates.
4-Meth ylsu lfon yla n ilin e (6). A mixture of Na2WO4 (0.067
g), 8 drops of acetic acid, and H2O (19 mL) was placed in a
flask and heated to 65 °C. 4-Methylthioaniline (19 mL, 153
mmol) was added followed by dropwise addition of H2O2 (34.5
mL, 337 mmol). The mixture was stirred at 65 °C for 1.5 h
and, after cooling, 800 mL of 1 N HCl and 500 mL of CHCl3
d
tion, and the remaining compounds, 10w , 51d , and 51f
were selected for further evaluation (Table 7).
Gastric toxicity was assessed using the 51Cr leakage
test.36 In this assay imidazoles 10w and 51d and 51f
did not induce gastric bleeding at 100 mg/kg bid. In
contrast, indomethacin produced a 7.5% leakage at a
much lower dose (5 mg/kg bid). In the adjuvant arthritis
model, compounds 51d and 51f were more potent than
10w and comparable to the reference compounds 2 and
3. Compound 51f showed a reduction in total radio-
graphic scores and secondary paw swelling (ID50: 0.18
mg/kg) in a dose-dependent manner.
Compound 51f was selected for further development
after pharmacokinetic evaluation in other species. It

New COX-2 Selective Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16 3471
were added. The layers were separated, and the aqueous phase
was washed with more CHCl3. The aqueous phase was basified
with 25% NaOH and extracted with CHCl3. The organic phase
was washed with brine and dried over MgSO4. The solvent
was removed to give 6 as a white solid (19.80 g, 75%): mp
was removed, and the residue was chromatographed on silica
gel (hexane-AcOEt mixtures) to afford 10r as a yellow solid
(0.028 g, 18%): mp 238-241 °C; 1H NMR (300 MHz, CDCl3 δ
TMS) 2.31 (s, 3 H), 3.11 (s, 3 H), 5.32 (s, 1 H), 7.14 (d, J ) 8.5
Hz, 2 H), 7.32 (d, J ) 8.5 Hz, 2 H), 7.37 (d, J ) 8.5 Hz, 2 H),
7.70 (s, 1 H), 8.02 (d, J ) 8.5 Hz, 2 H); Anal. (C18H16ClN3O3S‚
0.5H2O) C, H, N, S.
4-Ch lor o-5-(3-ch lor o-4-d im e t h yla m in op h e n yl)-1-(4-
m eth yl-su lfon ylp h en yl)im id a zole (10a b). Following a simi-
lar procedure to that described for the preparation of 10a , but
starting from 5-(4-dimethylaminophenyl)-1-(4-methyl-sulfo-
nylphenyl)imidazole and using 2 equiv of N-chlorosuccinimide,
compound 10a b was obtained as a yellow solid (45%): mp 169
°C; 1H NMR (300 MHz, CDCl3 δ TMS) 2.85 (s, 6 H), 3.09 (s, 3
H), 6.9 (m, 2 H), 7.2 (m, 1 H), 7.35 (d, J ) 8.6 Hz, 2 H), 7.64
(s, 1 H), 8.00 (d, J ) 8.6 Hz, 2 H). Anal. (C18H17Cl2N3O2S) C,
H, N, S.
5-(4-F lu or op h en yl)-4-m eth yl-1-(4-m eth ylsu lfon ylp h e-
n yl)im id a zole (13a ). Following a similar procedure to that
described for the preparation of 9a , but using R-tosylethyl
isocyanide22 (12a ) instead of tosylmethyl isocyanide, 13a was
obtained as a white solid (45%): mp 143-143 °C; 1H NMR
(300 MHz, CDCl3 δ TMS) 2.31 (s, 3 H), 3.08 (s, 3 H), 7.05 (m,
4 H), 7.27 (d, J ) 9 Hz, 2 H), 7.71 (s, 1 H), 7.93 (d, J ) 9 Hz,
2 H); Anal. (C17H15FN2O2S‚0.25H2O) C, H, N, S.
1
134 °C; H NMR (300 MHz, CDCl3 δ TMS) 2.97 (s, 3 H), 4.04
(s, 2 H), 6.66 (d, J ) 9 Hz, 2 H), 7.56 (d, J ) 9 Hz, 2 H).
N-(4-F lu or oben zylid en e)-4-m eth ylsu lfon yla n ilin e (8a ).
A mixture of 6 (19.60 g, 115 mmol), 4-fluorobenzaldehyde
(12.19 mL, 115 mmol), and toluene (590 mL) was refluxed in
a Dean-Stark for 2 days. The solvent was removed, and the
crude 8a thus obtained was directly used in the next reaction.
A sample was recrystallized from Et2O to give the analyti-
1
cally pure compound: mp 142 °C; H NMR (300 MHz, CDCl3
δ TMS) 3.08 (s, 3 H), 7.20 (m, 2 H), 7.30 (m, 2 H), 7.98 (m, 4
H), 8.38 (s, 1 H).
5-(4-F lu or op h en yl)-1-(4-m eth ylsu lfon ylp h en yl)im id a -
zole (9a ). A mixture of 8a (31.8 g, 115 mmol), tosylmethyl
isocyanide (33.4 g, 172 mmol), K2CO3 (31.7 g, 229 mmol),
MeOH (795 mL), and DME (340 mL) was refluxed for 2 h. The
solvent was removed, and the residue was redissolved in a CH2-
Cl2/brine mixture. The layers were separated, and the aqueous
phase was extracted with CH2Cl2. The combined organic
phases were dried over MgSO4 and concentrated. The crude
product was washed with Et2O and recrystallized from EtOAc/
hexane to afford 9a as a creamy solid (27.5 g, 75%): mp 151-
Compounds 13b-e were obtained in a similar way from the
corresponding tosyl isocyanides (12b-e), which were prepared
as described for 10.22
1
155 °C; H NMR (300 MHz, CDCl3 δ TMS) 3.10 (s, 3 H), 7.05
(m, 2 H), 7.13 (m, 2 H), 7.26 (s, 1 H), 7.36 (d, J ) 9 Hz, 2 H),
7.75 (s, 1 H), 7.99 (d, J ) 9 Hz, 2 H); Anal. (C16H13FN2O2S) C,
H, N, S.
N-(4-Meth ylsu lfa n ylben zylid en )-4-flu or oa n ilin e (23). A
mixture of 4-fluoroaniline (10.0 g, 90 mmol), 4-methylsulfa-
nylbenzaldehyde (16.5 g, 90 mmol), and benzene (500 mL) was
refluxed in a Dean-Stark for 2 days. The solvent was removed
and the crude product obtained was directly used in the
following reaction. A sample was recrystallized from Et2O to
give analytically pure 25: mp 93 °C; 1H NMR (300 MHz, CDCl3
δ TMS) 2.54 (s, 3 H), 7.07 (m, 2 H), 7.20 (m, 2 H), 7.31 (d, J )
9 Hz, 2 H), 7.79 (d, J ) 9 Hz, 2 H), 8.38 (s, 1 H).
5-(6-Eth oxy-3-p yr id yl)-1-(4-m eth ylsu lfon ylp h en yl)im i-
d a zole (9a m ). A mixture of 9a l (0.20 g, 0.6 mmol), 18-crown-6
(0.007 g), KOH (0.079 g, 1.2 mmol), EtOH (0.1 mL), and
toluene (10 mL) was refluxed in a Dean Stark for 12 h. The
mixture was poured on ice, and the layers were separated. The
aqueous phase was extracted with EtOAc, and the organic
phases were dried over MgSO4 and concentrated. Chromatog-
raphy on silica gel (hexane-AcOEt mixtures) afforded 9a m
as a yellow solid (0.20 g, 100%): mp 167-169 °C; 1H NMR
(300 MHz, CDCl3 δ TMS) 1.40 (t, J ) 7.5 Hz, 3 H), 3.10 (s, 3
H), 4.35 (q, J ) 7.5 Hz, 2 H), 6.65 (d, J ) 8.5 Hz, 1 H), 7.30
(m, 2 H), 7.38 (d, J ) 8.5 Hz, 2 H), 7.79 (m, 1 H), 8.02 (m, 3
H); Anal. (C17H17N3O3S‚0.5H2O) C, H, N, S.
4-Meth ylsu lfon ylben za ld eh yd e (24). To a solution of
4-methylthiobenzaldehyde (5 g, 33 mmol) in CH2Cl2 (132 mL)
was added, at 0 °C, m-chloroperbenzoic acid (20.61 g, 66 mmol).
The mixture was stirred for 3 h at room temperature and was
poured over CHCl3, washed with saturated NaHCO3 solution,
and dried over MgSO4. The solvent was removed, and the
residue was chromatographed on silica gel (hexane-AcOEt
mixtures) to afford 24 as a white solid (3.96 g, 65%): mp 157-
4-Ch lor o-5-(4-flu or oph en yl)-1-(4-m eth ylsu lfon ylph en yl)-
im id a zole (10a ). A mixture of 9a (27.2 g, 86 mmol), N-
chlorosuccinimide (12.05 g, 90 mmol), and CHCl3 (81 mL) was
refluxed for 18 h. The solvent was removed, and the residue
was redissolved in CH2Cl2 and washed with 1 N HCl, 1 N
NaOH, and brine. The organic phase was dried over MgSO4
and concentrated to a residue, which was chromatographed
on silica gel (hexane-AcOEt mixtures) to afford 10a as a white
1
159 °C; H NMR (300 MHz, CDCl3 δ TMS) 3.10 (s, 3 H), 8.09
(m, 4 H), 10.14 (s, 1 H).
In a similar manner, compound 19 was obtained from
4-methylsulfanylaniline and 4-fluorobenzaldehyde.
1-(4-F lu or op h en yl)-5-(4-m eth ylsu lfa n ylp h en yl)im id a -
zole (27). A mixture of 25 (6 g, 24.5 mmol), benzotriazolyl-
methylisocyanide (3.87 g, 24.5 mmol), potassium tert-butoxide
(5.49 g, 49 mmol), and DMSO (98 mL) was heated to 75 °C for
1 h. After cooling, Et2O was added, and it was washed with
H2O. The organic phase was dried over MgSO4, and the solvent
was removed. The residue thus obtained was chromatographed
on silica gel (hexane-AcOEt mixtures) to afford 27 as a white
solid (4.06 g, 58%): mp 96-99 °C; 1H NMR (300 MHz, CDCl3
δ TMS: 2.46 (s, 3 H), 7.0-7.3 (m, 9 H), 7.67 (s, 1 H); Anal.
(C16H13FN2S‚0.5H2O) C, H, N, S.
1
solid (24.0 g, 80%): mp 167 °C; H NMR (300 MHz, CDCl3 δ
TMS) 3.13 (s, 3 H), 7.12 (m, 2 H), 7.20 (m, 2 H), 7.32 (d, J )
9 Hz, 2 H), 7.71 (s, 1 H), 8.02 (d, J ) 9 Hz, 2 H); Anal. (C16H12
ClFN2O2S) C, H, N, S.
-
Following a similar procedure, but using NCS, NBS, or 2
equiv of NCS, and starting from the corresponding imidazoles,
compounds 11, 18, 23, 41, 42, 43, and 44 were obtained.
5-(4-Am in oph en yl)-4-ch lor o-1-(4-m eth ylsu lfon ylph en yl)-
im id a zole (10q). A mixture of 4-chloro-1-(4-methylsulfo-
nylphenyl)-5-(4-nitrophenyl)imidazole (1.14 g, 3 mmol), SnCl2
(2.88 g, 15 mmol), and EtOH (21 mL) was refluxed for 1.5 h.
The solvent was removed, and the residue was basified with
25% NaOH and extracted with CHCl3. The organic phase was
dried over MgSO4 and concentrated. Chromatography on silica
gel (hexane-AcOEt mixtures) afforded 10q as a yellow solid
In a similar manner, compound 20 was obtained from 19.
1-(4-F lu or op h en yl)-5-(4-m eth ylsu lfon ylp h en yl)im id a -
zole (28). Following a similar procedure to that described for
the preparation of 9a , but starting from 24 and 4-fluoroaniline,
1
28 was obtained as a white solid (70%): mp 133-134 °C; H
NMR (300 MHz, CDCl3 δ TMS) 3.05 (s, 3 H), 7.20 (m, 4 H),
7.31 (d, J ) 9 Hz, 2 H), 7.41 (s, 1 H), 7.73 (s, 1 H), 7.83 (d, J
) 9 Hz, 2 H); Anal. (C16H13FN2O2S) C, H, N, S.
1
(0.85 g, 81%): mp 170 °C; H NMR (300 MHz, CDCl3 + CD3-
OD δ TMS) 3.08 (s, 3 H), 4.0 (s, 2 H + H2O), 6.60 (d, J ) 8.5
Hz, 2 H), 6.90 (d, J ) 8.5 Hz, 2 H), 7.35 (d, J ) 8.5 Hz, 2 H),
7.66 (s, 1 H), 7.93 (d, J ) 8.5 Hz, 2 H); Anal. (C16H14ClN3O2S‚
H2O) C, H, N, S.
2-Ch lor o-1-(4-flu or oph en yl)-5-(4-m eth ylsu lfan ylph en yl)-
im id a zole (29). To a solution of diisopropylamine (0.35 mL,
2.5 mmol) in THF (8.5 mL) was added, at -20 °C, BuLi (1.6
M in hexane, 1.57 mL, 2.5 mmol), and the mixture was stirred
for 10 min. Compound 27 (0.56 g, 2 mmol) in THF (14 mL)
was added, and after stirring for 30 min, N-chlorosuccinimide
5-(4-Acet yla m in op h en yl)-4-ch lor o-1-(4-m et h ylsu lfo-
n ylp h en yl)im id a zole (10r ). A mixture of 10q (0.15 g, 0.4
mmol) and Ac2O (0.15 mL) was refluxed for 4 h. The solvent

3472 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16
Almansa et al.
(0.78 g, 5.8 mmol) in THF (8 mL) was added. The mixture was
stirred for 30 min at -20 °C and for 1.5 h at room temperature.
The solvent was removed, and the residue was dissolved in
an EtOAc-H2O mixture. The layers were separated, and the
aqueous phase was extracted with EtOAc. The organic phase
was dried over MgSO4, and the solvent was removed to give a
residue, which was chromatographed on silica gel (hexane-
AcOEt mixtures). Compound 29 was obtained as a white solid
(0.23 g, 37%): 1H NMR (300 MHz, CDCl3 δ TMS) 2.43 (s, 3
H), 6.9-7.2 (m, 8 H), 7.67 (s, 1 H).
Following a similar procedure but using NBS or methyl
iodide instead of NCS, compounds 30 and 31 were obtained.
Following a similar procedure, but starting from compound
20 and using methyl iodide instead of NCS, compound 21 was
obtained.
Compound 39 was obtained as a white solid (0.090 g, 37%):
mp 192 °C; 1H NMR (300 MHz, CDCl3 + CD3OD δ TMS) 2.99
(s, 3 H), 7.1 (m, 6 H), 7.39 (s, 1 H), 7.75 (d, J ) 8.5 Hz, 2 H);
Anal. (C17H12FN3O2S‚0.25H2O) C, H, N, S.
2-Ben zoyl-1-(4-flu or op h en yl)-5-(4-m eth ylsu lfon ylp h e-
n yl)im id a zole (40). To a solution of 28 (0.20 g, 0.6 mmol) in
acetonitrile (1 mL), NEt3 (0.08 mL, 0.6 mmol), and benzoyl
chloride (0.07 mL, 0.6 mmol) was added, and the mixture was
stirred at room-temperature overnight. The mixture was
poured over H2O and extracted with EtOAc. The organic phase
was washed with 1 N HCl and H2O, dried over Na2SO4, and
concentrated. The residue was chromatographed on silica gel
(hexane-AcOEt mixtures) to give 40 as a white solid (0.060
1
g, 22%): mp 225-226 °C; H NMR (300 MHz, CDCl3 δ TMS)
3.05 (s, 3 H), 7.0-7.5 (m, 10 H), 7.85 (d, J ) 8.5 Hz, 2 H), 8.25
(d, J ) 8.5 Hz, 2 H); Anal. (C23H17FN2O3S‚0.25H2O) C, H, N,
S.
2-Ch lor o-1-(4-flu or oph en yl)-5-(4-m eth ylsu lfon ylph en yl)-
im id a zole (32). Following a similar procedure to that de-
scribed for the preparation of 24, but starting from 29,
compound 32 was obtained as a white solid (80%): mp 218-
Following a similar procedure, but starting from compound
9a , compound 15 was obtained.
1
220 °C; H NMR (300 MHz, CDCl3 δ TMS) 3.04 (s, 3 H), 7.1
N-ter t-Bu tyl-4-a cetyla m in oben zen esu lfon a m id e (46).
To a suspension of 4-acetylaminobenzenesulfonyl chloride (10
g, 43 mmol) in DME (103 mL) was added, at 0 °C, tert-
butylamine (9 mL, 86 mmol) in DME (103 mL). The bath was
removed, and the mixture was stirred for 5 h at reflux. The
solvent was concentrated, and CHCl3 was added to the residue.
The suspension thus obtained was filtered, and the solid was
washed with CHCl3, H2O, and Et2O and vacuum-dried. Com-
pound 46 was obtained as a white solid (8.1 g, 70%): mp 200-
201 ° C; 1H NMR (CDCl3 + CD3OD δ TMS): 1.15 (s, 9 H),
2.12 (s, 3 H), 4.21 (s, 2 H + CD3OD), 7.66 (d, J ) 9 Hz, 2 H),
7.75 (d, J ) 9 Hz, 2 H); Anal. (C12H18N2O4S) C, H, N, S.
N-ter t-Bu tyl-4-a m in oben zen esu lfon a m id e (47). A solu-
tion of 46 (8.0 g, 30 mmol), KOH (8.3 g, 148 mmol), H2O (6
mL), and MeOH (24 mL) was heated at 100 ° C during 2 h.
Water (24 mL) was added and heating continued for 2 h more.
The mixture was allowed to cool, H2O was added, and the pH
was adjusted to 8 with 1 N HCl. The suspension thus obtained
was extracted with AcOEt and dried over Na2SO4. The solvent
was removed to afford 47 as a white solid (6.0 g, 89%): mp
127 ° C; 1H NMR (CDCl3 + CD3OD δ TMS): 1.19 (s, 9 H),
3.74 (s, CD3OD + 1 H), 6.67 (d, J ) 9 Hz, 2 H), 7.60 (d, J ) 9
Hz, 2 H); Anal. (C10H16N2O2S) C, H, N, S.
(m, 6 H), 7.32 (s, 1 H), 7.82 (d, J ) 8.5 Hz, 2 H); Anal. (C16H12
Cl2FN2O2S‚0.25H2O) C, H, N, S.
-
In a similar manner compounds 33, 34, and 22 were
obtained from 30, 31, and 21, respectively.
1-(4-F lu or op h en yl)-2-h yd r oxym eth yl-5-(4-m eth ylsu lfo-
n ylp h en yl)im id a zole (35). A mixture of 28 (2.0 g, 6.3 mmol)
and CH2O (40% in H2O, 10 mL) was heated at 130 °C for 72
h. The solvent was removed, and the residue was dissolved in
a EtOAc-H2O mixture. The layers were separated, and the
aqueous phase was extracted with EtOAc. The organic phase
was dried over MgSO4, and the solvent was removed to give a
residue which was chromatographed on silica gel (hexane-
AcOEt mixtures). Compound 35 was obtained as a white solid
1
(0.94 g, 43%): mp 211-212 °C; H NMR (300 MHz, CDCl3 +
CD3OD δ TMS) 3.07 (s, 3 H), 3.8 (s, 1 H + H2O), 4.45 (s, 2 H),
7.2 (m, 7 H), 7.80 (d, J ) 8.2 Hz, 2 H); Anal. (C17H15FN2O3S)
C, H, N, S.
2-F lu or om et h yl-1-(4-flu or op h en yl)-5-(4-m et h ylsu lfo-
n ylp h en yl)im id a zole (36). To a solution of 35 (0.2 g, 0.6
mmol) in CH2Cl2 (4 mL) was carefully added diethylamino-
sulfur trifluoride (0.14 g, 0.9 mmol), and the mixture was
stirred at room temperature for 2 h. Water was added, the
layers were separated, and the aqueous phase was extracted
with CH2Cl2. The organic phase was dried over MgSO4, and
the solvent was removed to give a residue, which was chro-
matographed on silica gel (hexane-AcOEt mixtures). Com-
pound 36 was obtained as a white solid (0.06 g, 28%): mp 152-
154 °C; Anal. (C17H14F2N2O2S‚0.75H2O) C, H, N, S.
1-(4-F lu or op h en yl)-5-(4-m eth ylsu lfon ylp h en yl)im id a -
zol-2-ca r boxa ld eh yd e (37) a n d Meth yl 1-(4-F lu or op h e-
n yl)-5-(4-m eth ylsu lfon ylp h en yl)im id a zol-2-ca r boxyla te
(38). A mixture of 35 (0.2 g, 0.6 mmol), MnO2 (1.26 g, 14.5
mmol), 3 Å molecular sieves (0.100 g), MeOH (6.5 mL), and
THF (4 mL) was stirred at room temperature for 24 h. The
resulting suspension was filtered through Celite and washed
with hot THF. The solvent was removed, and the crude product
was chromatographed on silica gel (hexane-AcOEt mixtures)
to give the following:
N-(3-F lu or o-4-m eth oxyben zylid en e)-4-(ter t-bu tyla m i-
n osu lfon yl)a n ilin e (48). Following a similar procedure to
that described for the preparation of 8a , but starting from 47,
compound 48 was obtained as a yellow solid (100%): mp 129-
1
131 ° C; H NMR (CDCl3 δ TMS): 1.23 (s, 9 H), 3.98 (s, 3 H),
4.65 (s, 1 H), 7.04 (t, J ) 8.1 Hz, 1 H), 7.21 (d, J ) 6.7 Hz, 2
H), 7.58 (m, 1 H), 7.73 (dd, J H-F ) 11.8 Hz, J ) 2 Hz, 1 H),
7.90 (d, J ) 6.7 Hz, 2 H), 8.33 (s. 1 H); Anal. (C18H21FN2O3S)
C, H, N, S.
1-(ter t-Bu tyla m in osu lfon ylp h en yl)-5-(3-flu or o-4-m eth -
oxyp h en yl)im id a zole (49). Following a similar procedure to
that described for the preparation of 9a , but starting from 48,
compound 49 was obtained as a white solid (87%): mp 229
1
°C; H NMR (CDCl3 δ TMS): 1.24 (s, 9 H), 3.89 (s, 3 H), 4.51
(s, 1 H), 6.90 (m, 3 H), 7.23 (s, 1 H), 7.29 (d, J ) 8.7 Hz, 2 H),
7.73 (s, 1 H), 7.94 (d, J ) 8.7 Hz, 2 H); Anal. (C20H22FN3O3S)
C, H, N, S.
1
37: (0.073 g, 36%); mp 198 °C; H NMR (300 MHz, CDCl3
δ TMS) 3.05 (s, 3 H), 7.1 (m, 6 H), 7.62 (s, 1 H), 7.85 (d, J )
8.5 Hz, 2 H), 9.84 (s, 1 H); Anal. (C17H13FN2O3S‚0.5H2O) C,
H, N, S.
38: (0.061 g, 28%); mp 192-194 °C; 1H NMR (300 MHz,
CDCl3 δ TMS) 3.03 (s, 3 H), 3.87 (s, 3 H), 7.1 (m, 6 H), 7.50 (s,
1 H), 7.87 (d, J ) 8.5 Hz, 2 H); Anal. (C18H15FN2O4S‚1.25H2O)
C, H, N, S.
1-(4-F lu or op h en yl)-5-(4-m eth ylsu lfon ylp h en yl)im id a -
zol-2-ca r bon itr ile (39). A mixture of 37 (0.24 g, 0.7 mmol),
hydroxylamine-O-sulfonic acid (0.155 g, 1.4 mmol), pyridine
(3 mL), and EtOH (30 mL) was stirred under reflux for 18 h.
The mixture was poured over CHCl3 and washed with satu-
rated NaHCO3 solution. It was dried over MgSO4, and the
solvent was removed, affording a crude product, which was
chromatographed on silica gel (hexane-AcOEt mixtures).
1-(ter t-Bu tylam in osu lfon ylph en yl)-4-ch lor o-5-(3-flu or o-
4-m eth oxyp h en yl)im id a zole (50). Following a similar pro-
cedure to that described for the preparation of 10a , but
starting from 49, compound 50 was obtained as a white solid
(82%): mp 209 °C; 1H NMR (CDCl3 δ TMS): 1.24 (s, 9 H),
3.89 (s, 3 H), 4.51 (s, 1 H), 6.90 (m, 3 H), 7.23 (d, J ) 8.7 Hz,
2 H), 7.63 (s, 1 H), 7.92 (d, J ) 8.7 Hz, 2 H); Anal. (C20H21
-
ClFN3O3S) C, H, N, S.
4-[4-Ch lor o-5-(3-flu or o-4-m et h oxyp h en yl)im id a zol-1-
yl]ben zen esu lfon a m id e (51f). A mixture of 50 (37.0 g, 85
mmol), concd HCl (200 mL), and H2O (200 mL) was heated
under reflux for 3 h. The mixture was allowed to cool and
basified with 6 N NaOH at pH ) 6. The precipitate thus
obtained was filtered and washed with H2O and CHCl3. The
solid thus obtained was crystallized from acetonitrile to give

New COX-2 Selective Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16 3473
51f as a white solid (27. 1 g, 84%): mp 211-212 °C; 1H NMR
(CDCl3 + CD3OD δ TMS) 3.90 (s, 3 H), 4.16 (s, CD3OD + 2
H), 6.93 (m, 3 H), 7.30 (d, J ) 8.6 Hz, 2 H), 7.73 (s, 1 H), 7.95
(d, J ) 8.6 Hz, 2 H); Anal. (C16H13ClFN3O3S) C, H, N, S.
4-[4-Ch lor o-5-(4-flu or op h en yl)im id a zol-1-yl]ben zen e-
su lfon a m id e (51a ).
pelleted by centrifugation at 1200g for 5 min at 4 °C, and PGE2
was determined in the supernatant by specific ELISA.
Hyp er a lgesia Test.31 The analgesic activity was measured
in the inflammatory hyperalgesia model. Oedema was induced
by injection of 1% carrageenan suspension (0.1 mL) in the right
hind foot pad of male Sprague-Dawley rats. Three hours later,
test compounds, suspended in 1 mL of tween-80 (1%), were
administered orally via a gavage needle. Thirty minutes after
dosing, hyperalgesic stimulus was induced by a heat source
under the inflamed paw. The withdrawal latency in seconds
was recorded for the control and treated groups. Percentage
inhibition of the stimulus-induced decrease in withdrawal
latency was determined and compared with measurement
taken in the left (untreated) paw.
Ad ju va n t-In d u ced Ar th r itis (AIA) in Ra t. AIA was
induced in the right hind foot pad of Lewis rats (140-170 g, 7
weeks old) by an intradermal injection of 0.5 mg of Mycobac-
terium butyricum (Difco Laboratory, Detroit, MI) in light
mineral oil. Ten rats were injected with incomplet adjuvant
and served as nonadjuvant controls. Animals were distributed
randomly into groups of 10 and dosed with 1 mg/kg po of tests
compounds daily for 28 days beginning the day of adjuvant
injection. Foot volume of left rat hind paws was measured
plethysmographically at the end of the experiment. The mean
increase in paw volume of treated animals was compared with
that of controls, and results were expressed as inhibition
percentage.
Ga str oin testin a l Lesion in Ra ts.36 Ulcerogenicity of test
compounds was evaluated by a modified version of the 51Cr
excretion method (2) in rats. a . Red Blood Cell La belin g.
Blood from a donor rat was collected 1:7 in ACD anticoagulant
(30 mM citric acid, 80 mM sodium citrate, 60 mM dextrose).
A red blood cell (RBC) pellet was obtained by centrifugation
and washed twice with ACD saline solution (5.4 mM citric acid,
12.2 mM sodium citrate, 9.5 mM dextrose, 0.9% sodium
chloride, pH 6.50). RBCs were resuspended up to the original
volume in saline solution and incubated with 14 Ci/mL sodium
51-chromate (CJ S11, Amersham PharmaciaBiotech). Unin-
corporated 51Cr was washed away, and radioactivity was
measured in aliquots of resuspended labeled-RBC in a gamma-
counter (Wallac). A volume of 0.5 mL of 51Cr-RBC (about 7
Ci) was subsequently iv injected into the animal. b. F eca l
Dr u g-In d u ced Bleed in g. Drugs (control, 5 mg/kg indometha-
cin, 10 mg/kg diclofenac, 100 mg/kg UR-8880, and 100 mg/kg
celecoxib, bid) were administered by gavage for 4 days. Feces
were collected daily, and radioactivity was quantified and
corrected for background and isotope decay. Data are ex-
pressed as percentage of injected radioactivity.
Step 1: 4-Meth ylsu lfin yla n ilin e. To a solution of 4-me-
thylsulfanylaniline (20 g, 144 mmol), CH2Cl2 (420 mL), and
MeOH (80 mL) at 0 °C was added magnesium monoperoxyph-
thalate hexahydrate (46.6 g, 75 mmol), and the mixture was
stirred at room temperature for 3 h. The suspension thus
obtained was filtered and concentrated. The residue was
dissolved in CH2Cl2 and washed with 1 N NaOH until basic.
The organic phase was dried over MgSO4 and concentrated.
A crude product was obtained, which was washed with Et2O
to afford 4-methylsulfinylaniline as a creamy solid (20.1 g,
90%): 1H NMR (300 MHz, CDCl3 δ TMS) 2.68 (s, 3 H), 4.02
(s, 2 H), 6.75 (d, J ) 8.7 Hz, 2 H), 7.45 (d, J ) 8.7 Hz, 2 H).
Step 2: 1-[4-(Acetoxym eth ylsu lfa n yl)p h en yl]-5-(4-flu o-
r op h en yl)im id a zole. A mixture of 1-[4-(methylsulfonyl)-
phenyl]-5-(4-fluorophenyl)imidazole (obtained from 4-methyl-
sulfonylaniline and 4-fluorobenzaldehyde after heating with
toluene and cyclization with TosMIC, as described for 9a , 1.6
g, 5.3 mmol), Ac2O (16 mL), and NaOAc (1.60 g, 20 mmol) was
stirred at reflux under a nitrogen for 8 h. The solvent was
removed, and the crude product was chromatographed on silica
gel (hexane-AcOEt mixtures) to give the desired product as
a foamy solid (1.6 g, 84%).
Step 3: 1-[4-(Acetoxym eth ylsu lfa n yl)p h en yl]-4-ch lor o-
5-(4-flu or op h en yl)im id a zole. Following a similar procedure
to that described for the preparation of 10a , but starting from
1-[4-(acetoxymethylsulfanyl)phenyl]-5-(4-fluorophenyl)imida-
zole the title compound was obtained (0.9 g, 51%).
Step 4: Sod iu m [4-Ch lor o-5-(4-flu or op h en yl)im id a zol-
1-yl]ben zen esu lfin a te. To a solution of the previous com-
pound (0.9 g), CH2Cl2 (8 mL), and MeOH (4 mL) at 0 °C was
added magnesium monoperoxyphthalate hexahydrate (1.5 g,
2.6 mmol), and the mixture was stirred at room-temperature
overnight. A 5% NaHCO3 solution (12 mL) was added, and
the mixture was extracted with CH2Cl2. The solvent was
removed, and the residue was dissolved in a mixture of THF
(8 mL) and MeOH (4 mL). After the mixture was cooled to 0
°C, a 1 N NaOH solution (2.6 mL) was added. The mixture
was stirred for 1 h at room temperature and was then
concentrated. Water was removed by azeotropic distillation
with EtOH/toluene mixtures. The residue was dried in vacuo,
to give the title compound (0.9 g, 100%).
Molecu la r Mod elin g. Compound 51f was modeled inside
the active site of COX-2 from the crystal structure of murine
COX-2 complex with the celecoxib analogue SC558. Atomic
Step 5: 4-[4-Ch lor o-5-(4-flu or op h en yl)im id a zol-1-yl]-
ben zen esu lfon a m id e (51a ). A mixture of sodium [4-chloro-
5-(4-fluorophenyl)imidazol-1-yl]benzenesulfinate (0.9 g), H2O
(13 mL), NaOAc (0.21 g, 2.7 mmol), and hydroxylamine-O-
sulfonic acid (0.3 g, 2.7 mmol) was stirred overnight at room
temperature. The resulting suspension was filtered, and the
solid was washed with EtOAc and H2O. The layers were
separated, and the aqueous phase was extracted with EtOAc.
The combined organic phases were concentrated, and the
residue was chromatographed on silica gel (hexane-AcOEt
mixtures) to give 51a as a white solid (0.42 g, 48%): mp 223
charges for this inhibitor were calculated in
a two-step
procedure using Spartan (Wave function Inc.): (i) geometry
optimization at the AM138 semiempirical level; (ii) ab initio
single point calculation at the HF/6-31G(d) level to obtain the
molecular electrostatic potential (MEP) from which ESP
charges were derived.
Only one monomer of the enzyme was modeled. The
protein-inhibitor system was hydrated with TIP3P water
molecules39 and equilibrated by a series of minimizations
interspersed by short 10 ps molecular dynamics simulations
using the Cornell et al. all atom force field40 implemented in
the program TINKER.41 Only atoms within 20 Å of the
inhibitor were allowed to move. The equilibrated system was
subject to a 1.0 ns molecular dynamics simulation at a constant
temperature of 300 K. The nonbonded pair-list was updated
every 20 steps (nonbonded cutoff was set to 14 Å). SHAKE42
was used together with a time step of 2 fs. Visualization was
performed with the Insight II package (Accelrys Inc.).
1
°C; H NMR (CDCl3 δ TMS) 4.84 (s, 2 H), 7.05 (m, 2 H), 7.18
(m, 2 H), 7.26 (d, J ) 8.6 Hz, 2 H), 7.65 (s, 1 H), 7.96 (d, J )
8.6 Hz, 2 H); Anal. (C15H11ClFN3O2S‚0.25H2O) C, H, N, S.
Air P ou ch Mod el of In fla m m a tion . Deter m in a tion of
P GE2 in Exu d a te.30 Male Lewis rats (175-200 g) were used.
Air cavities were produced by a subcutaneous injection of
sterile air (20 mL) into the intracapsular area. Every 2 days
air (10 mL) was injected into the cavity to keep the space open.
Seven days after the first injection, lambda-carrageenan
(Sigma) in saline (2 mL of a 1% solution) was injected into
the air pouch to produce an inflammatory reaction. The
compounds were administered by oral route as a suspension
in carboxymethyl cellulose and Tween-80 at 1% (10 mL/kg),
0.5 h before carrageenan injection. The animals were killed 6
h later, and the exudate volume was measured. Cells were
Ack n ow led gm en t . We thank M. Carmen Torres,
Concepcio´n Gonza´lez, Consol Ferreri, Guadalupe Mar-
t´ınez, Nuria Recasens, Teresa Gamero, J ose´ Antonio
Garc´ıa, Ana Ester Sanahuja, J ordi Vilar, and Assumpta

3474 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16
Almansa et al.
M.; Yuan, J .; Zhang, Y. Y.; Seibert, K. N-[[(5-Methyl-3-phenyl-
isoxazol-4-yl)-phenyl]sulfonyl]propanamide, Sodium Salt, Pare-
coxib Sodium: A Potent and Selective Inhibitor of COX-2 for
Parenteral Administration. J . Med. Chem. 2000, 43, 1661-1663.
(12) Riendeau, D.; Percival, M. D.; Brideau, C.; Charleson, S.; Dube´,
D.; Ethier, D.; Falgueyret, J . P.; Friesen, R. W.; Gordon, R.;
Greig, G.; Guay, J .; Mancini, J .; Oellet, 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.; Can,
C. C. Etoricoxib (MK-0663): Preclinical Profile and Comparison
with other Agents that Selectively Inhibit Cyclooxygenase-2. J .
Pharmacol. Exp. Ther. 2001, 296, 558-566.
(13) (a) J ackson, L. M.; Hawkey, C. J . Gastrointestinal effects of
COX-2 Inhibitors. Exp. Opin. Invest. Drugs, 1999, 8, 963-971;
(b) Breyer, M. D.; Hao, C.; Qi, Z. Cyclooxygenase-2 selective
inhibitors and the kidney. Curr. Opin. Crit. Care 2001, 7, 393-
400; (c) Emery, P. Cyclooxygenase-2: A Major Therapeutic
Advance? Am. J . Med. 2001, 110, 42s-45s.
(14) Kurumbail, R. G.; Stevens, A. M.; Gierse, J . K.; McDonald, J .
J .; Stegeman, R. A.; Pak, J . Y.; Gildehaus, D.; Miyashiro, J . M.;
Penning, T. D.; Seibert, K.; Isakson, P. C.; Stallings, W. C.
Structural Basis for Selective Inhibition of Cyclooxygenase-2 by
Anti-inflammatory Agents. Nature 1996, 384, 644-48.
(15) Soliva, R.; Almansa, C.; Kalko, S. G.; Luque, F. J .; Orozco, M.
Theoretical studies on the inhibition mechanism of COX-2. Is
there a unique recognition site? J . Med. Chem., in press.
(16) Plount Price, M. L.; J orgensen, W. L. Analysis of Binding
Affinities for Celecoxib Analogues with COX-1 and COX-2 from
Combined Docking and Monte Carlo Simulations and Insight
into the COX-2/COX-1 Selectivity. J . Am. Chem. Soc. 2000, 122,
9455-9466.
(17) Barta, T. E.; Stealy, M. A.; Collins, P. W.; Weier, R. M.
Antiinflammatory 4,5-Diarylimidazoles as Selective Cyclooxy-
genase Inhibitors. Bioorg. Med. Chem. Lett. 1998, 8, 3443-3448.
(18) Khanna, I. K.; Weier, R. M.; Yu, Y.; Xu, X. D.; Koszyk, F. J .;
Collins, P. W.; Koboldt, C. M.; Veenhuizen, A. W.; Perkins, W.
E.; Casler, J . J .; Masferrer, J . L.; Gregory, S. A.; Seibert, K.;
Isakson, P. C. 1,2-Diarylimidazoles as Potent, Cyclooxygenase-2
Selective and Orally Active Antiinflammatory Agents. J . Med.
Chem. 1997, 40, 1634-1647.
Oliveras for their excellent technical assistance and
J ordi Belloc for the GC, MS, and HPLC analysis. We
thank also the “Centre de Supercomputacio´ de Catalu-
nya (CESCA)” for computational facilities and Modesto
Orozco and Francisco J avier Luque (Molecular Modeling
and Bioinformatics Unit, Institut de Recerca Biome`dica,
Parc Cient´ıfic de Barcelona) for helpful discussions.
Su p p or tin g In for m a tion Ava ila ble: Description of the
synthesis and experimental data for the preparation of tria-
zoles 52a -c and 53a ,b. This material is available free of
charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) (a) Vane, J . R. Inhibition of Prostaglandin Synthesis as
a
Mechanism of Action of Aspirin-like Drugs. Nature 1971, 231,
232-5. (b) Vane, J . R.; Ferreira, S. H. Antiinflammatory Drugs;
Springer-Verlag: Berlin, 1979; pp 305-347.
(2) (a) Sontag, S. J . Prostaglandins in Peptic Ulcer Disease. An
Overview of Current Status and Future Directions. Drugs 1986,
32, 445-457. (b) Allison, M. C.; Howatson, A. G.; Torrance, C.
J .; Lee, F. D.; Russell, R. Y. G. Gastrointestinal Damage
Associated with the Use of Nonsteroidal Antiinflammatory
Drugs. N. Engl. J . Med. 1992, 327, 749-754.
(3) Clive, D. M.; Stoff, J . S. Renal Syndromes Associated with
Nonsteroidal Antiinflammatory Drugs. N. Engl. J . Med. 1984,
310, 563-572.
(4) (a) 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 Splic-
ing. Proc. Natl. Acad. Sci. U.S.A., 1991, 88, 2692-2696. (b)
J ujubu, D. A.; Fletcher, B. S.; Varnum, B. C.; Lim, R. W.;
Herschman, H. R. TIS10, a Phorbol Ester Tumor Promoter-
Inducible mRNA from Swiss 3T3 Cells, Encodes a Novel Pros-
taglandin Synthase/Cyclooxygenase Homologue. J . Biol. Chem.
1991, 266, 12866-72.
(19) van Leusen, A. M.; Wildeman, J .; Oldenziel O. H. Base_Induced
Cycloaddition of Sulfonylmethyl Isocyanides to C,N Double
Bonds. Synthesis of 1,5-Disubstituted and 1,4,5-Trisubstituted
Imidazoles form Aldimines and Imidoyl Chlorides. J . Org. Chem.
1977, 42, 1153-59.
(5) (a) Vane, J . R. Towards a Better Aspirin. Nature, 1994, 367,
215-16. (b) Masferrer, J . L.; Zweifel, B. S.; Manning, P. T.
Selective Inhibition of Inducible Cyclooxygenase-2 in vivo is
Antiinflammatory and Nonulcerogenic. Proc. Natl. Acad. Sci.
U.S.A. 1994, 91, 3228-32.
(6) Leblanc, Y.; Black, W. C.; Chan, C. C.; Charleson, S.; Delorme,
D.; Denis, D.; Gauthier, J . Y.; Grimm, E. L.; Gardon, R.; Guay,
D.; Hamel, P.; Kargman, S.; Lau, C. K.; Mancini, J .; Ouellet,
M.; Percival, D.; Roy, P.; Skorey, K.; Tagari, P.; Vickers, P.;
Wong, E.; Xu, L.; Prasit, P. Synthesis and Biological Evaluation
of Both Enantiomers of L-761,000 as Inhibitors of Cycloogyge-
nase 1 and 2. Biorg. Med. Chem. Lett. 1996, 6, 731-6.
(7) Kalgutkar, A. S. Selective Cyclooxygenase-2 Inhibitors as Non-
ulcerogenic Anti-inflammatory Agents. Exp. Opin. Ther. Pat.
1999, 9, 831-849.
(20) Katritzky, A. R.; Cheng, D.; Musgrave, R. P. Syntheses of
Imidazoles and Pyrroles: BetMIC and TosMIC as Complemen-
tary Reagents. Heterocycles 1997, 44, 67-70.
(21) Giam, C. S.; Kikukawa, K.; Trujillo, D. A.
A Convenient
Preparation of Alkyl and Arylpyridines. OPPI 1992, 137-140.
(22) van Leusen, A. M.; Bouma, R. J .; Possel, O. Phase-Transfer
Mono-alkylation of Tosylmethylisocyanide. Tetrahedron Lett.
1975, 40, 3487-8.
(23) Begg, G. C.; Grimmett, M. R.; Yu-Man, L. The Synthesis of
2-Alkyl and 2-Acyl-imidazoles by Substitution Methods. Aust.
J . Chem. 1973, 26, 415-420.
(24) Balaban, I. E.; Pyman, F. L. The Bromo-derivatives of 1-Meth-
ylglyoxaline, and the Constitution of Chloroxalmethylin. J .
Chem. Soc. 1924, 1564-1572.
(25) Kirk, K. L. Facile Synthesis of 2-Substituted Imidazoles. J . Org.
Chem. 1978, 22, 4381-2.
(8) (a) Reitz, D. B.; Isakson, P. C. Cyclooxygenase-2 Inhibitors.
Current Pharm. Design 1995, 1, 211-220. (b) Carter, J . S.
Recently Reported Inhibitors of Cyclooxygenase-2. Exp. Opin.
Ther. Pat. 1997, 8, 21-29.
(9) 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.; Cogburn, 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 Cyclooxygenase-2 Inhibitors:
Identification of 4[5-(4-Methylphenyl)-3-(trifluoromethyl)-1H-
pyrazol-1-yl]benzenesulfonamide (SC-58635, Celecoxib). J . Med.
Chem. 1997, 40, 1347-1365.
(26) Szczepankiewicz, B. G.; Heathcock, C. H. Total synthesis of
Diplamine, a Cytotoxic Pyridoacridine Alkaloid from a Pacific
Tunicate. J . Org. Chem. 1994, 59, 3512-3523.
(27) DeVleeschauwer, M.; Gauthier, J . Y. Remarkably Mild and
Simple Preparation of Sulfinates, Sulfonyl Chlorides and Sul-
fonamides from Thioanisoles. Synlett 1997, 375-6.
(28) Ito, S.; Tanaka, Y.; Kakehi, A. Base-induced Dehydrosulfinato-
cyclization of N-Alkyl-N-phenylsulfonyl-N’’-arylbenzamidrazones
to 3,4-Diaryl-4H-1,2,4-triazoles. Bull. Chem. Soc. J pn. 1984, 57,
544-547.
(10) Prasit, P.; Wang, Z.; Brideau, 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.; Le´ger, S.; Mancini, J .; O′Neill, G. P.;
Ouellet, M.; Percival, M. D.; Perrier, H.; Riendeau, D.; Rodger,
Y.; Tagari, P.; The´rien, M.; Vickers, P.; Wong, E.; Xu, L.-J .;
Young, R. N.; Zamboni, R. The Discovery of Rofecoxib, [MK 966,
Vioxx, 4-(4′-Methylsulfonylphenyl)-3-phenyl-2(5H)furanone], an
Orally Active Cyclooxygenase-2 Inhibitor. Bioorg. Med. Chem.
Lett. 1999, 9, 1773-1778.
(29) Experimental methods described in Almansa, C.; de Arriba, A.
F.; Cavalcanti, F. L.; Go´mez, L. A.; Miralles, A.; Merlos, M.;
Garc´ıa-Rafanell, J .; Forn, J . Synthesis and SAR of a New Series
of COX-2-Selective Inhibitors: Pyrazolo[1,5-a]pyrimidines. J .
Med. Chem. 2001, 44, 350-361.
(30) Seibert, K.; Zhang, Y.; Leahy, K.; Hauser, S.; Masferrer, J .;
Perkins, W.; Lee, L.; Isakson, P. Pharmacological and Biochemi-
cal Demonstration of the Role of Cyclooxygenase-2 in Inflam-
mation and Pain. Natl. Acad. Sci. U.S.A. 1994, 91, 12013-12017.
(31) Hargreaves, K.; Dubner, R.; Brown, F.; Flores, C.; J oris, J . A
New and Sensitive Method for Measuring Thermal Nociception
in Cutaneous Hyperalgesia. Pain 1988, 32, 77-88.
(32) Li, C. S.; Brideau, C.; Chan, C. C.; Savoie, C.; Claveau, D.;
Charleson, S.; Gordon, R.; Greig, G.; Gauthier, J . Y.; Lau, C. K.;
Riendeau, D.; Therien, M.; Wrong, E. Poster Communication,
Presented at the 216th American Chemical Society National
Meeting, Boston, 1998.
(11) (a) 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,
Valde-
coxib: A Potent and Selective Inhibitor of COX-2. J . Med. Chem.
2000, 43, 775-777; (b) Talley, J . J .; Bertershaw, S. R.; Brown,
D. L.; Carter, J . S.; Graneto, M. J .; Kellogg, M. S.; Koboldt, C.

New COX-2 Selective Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 16 3475
(33) Li, C. S.; Black, W. C.; Brideau, C.; Chan, C. C.; Charleson, S.;
Cromlish, W. A.; Claveau, D.; Gauthier, J . Y.; Gordon, R.; Greig,
G.; Grimm, E.; Guay, J .; Lau, C. K.; Therien, M.; Visco, D. M.;
Wong, E.; Xu, L.; Prasit, P. A New Structural Variation on the
Methanesulphonylphenyl Class of Selective Cyclooxygenase-2
Inhibitors. Bioorg. Med. Chem. Lett. 1999, 3181-3186.
(34) Pinto, D. J . P.; Copeland, R. A.; Covington, M. B.; Pitts, W. J .;
Batt, D. G.; Orwat, M. J .; Lam, G. N.; J oshi, A.; Chan, Y.-C.;
Wang, S.; Trzaskos, J . M.; Magolda, R. L.; Kornhauser, D. M.
Chemistry and Pharmacokinetics of Diarylthiophenes and Ter-
phenyls as Selective Cyclooxygenase-2 Inhibitors. Bioorg. Med.
Chem. Lett. 1996, 2907-2912.
(35) Menasse´-Gdynia R., Krupp P. Quantitative measurement of
gastrointestinal bleeding in rats: The effect of nonsteroidal anti-
inflammatory drugs. Toxicol. Appl. Pharmacol. 1974, 29, 389-
396.
(36) Standard in vitro CYP inhibition of isoforms CYP2C9, CYP2C19,
CYP2D6, and CYP3A4 was performed. For a reference see:
Miller V. P.; Stresser D. M.; Blanchard A. P.; Turner S.; Crespi
C. L.; Fluorometric high-throughput screening for inhibitors of
cytochrome P450. Ann. N. Y. Acad. Sci. 2000, 919, 26-32.
(37) (a) Almansa, C.; Alfo´n, J .; Cavalcanti, F. L.; Escamilla, I.; Go´mez,
L. A.; Miralles, A. Discovery of UR-8880: A New COX-2 Selective
Inhibitor. Poster Communication, 222nd ACS National Meeting,
Chicago, 2001. (b) Almansa, C.; Gonza´lez, C.: Torres, C. Patent
Appl. WO 0023426.
(38) Dewar, M. J . S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J . J . P.
AM1: A New General Purpose Quantum Mechanical Molecular
Model. J . Am. Chem. Soc. 1985, 107, 3902-3909.
(39) J orgensen, W. L.; Chandrasekhar, J .; Madura, J .; Impey, R. W.;
Klein, M. L. Comparison of Sample Potential Functions for
Simulating Liquid Water. J . Chem. Phys. 1983, 79, 926-933.
(40) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K.
M.; Ferguson MD.; Spellmeyer, D. C.; Fox, T.; Caldwell, J . W.;
Kollman, P. A Second Generation Force Field for the Simulation
of Proteins, Nucleic Acids and Organic Molecules. J . Am. Chem.
Soc. 1995, 117, 5179-5197.
(41) Ponder, J . W. TINKER, Software Tools for Molecular Design.
(42) Ryckaert, J . P.; Ciccotti, G.; Berendsen, H. J . C. Numerical
Integration of the Cartesian Equations of Motion of a System
with Constraints: Molecular Dynamics of n-Alkanes. J . Comput.
Phys. 1977, 23, 327-341.
J M030765S