Sulfonamido, azidosulfonyl and N-acetylsulfonamido analogues of rofecoxib: 4-[4-(N-acetylsulfonamido)phenyl]-3-(4-methanesulfonylphenyl)-2(5H)furanone is a potent and selective cyclooxygenase-2 inhibitor

Article abstract of DOI:10.1016/j.bmcl.2004.01.076

4-[4-(N-Acetylsulfonamido)phenyl]-3-(4-methanesulfonylphenyl)-2(5H) furanone, possessing a N-acetylsulfonamido pharmacophore, has been identified as a potent and selective COX-2 inhibitor that has the potential to acetylate the COX-2 isozyme.

Full text of DOI:10.1016/j.bmcl.2004.01.076

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1957–1960
Sulfonamido, azidosulfonyl and N-acetylsulfonamido analogues of
rofecoxib: 4-[4-(N-acetylsulfonamido)phenyl]-3-(4-
methanesulfonylphenyl)-2(5H)furanone is a potent and selective
cyclooxygenase-2inhibitor
Afshin Zarghi, P. N. Praveen Rao and Edward E. Knaus*
Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2N8
Received 18 August 2003;revised 25 January 2004;accepted 26 January 2004
Abstract—4-[4-(N-Acetylsulfonamido)phenyl]-3-(4-methanesulfonylphenyl)-2(5H)furanone, possessing
pharmacophore, has been identified as a potent and selective COX-2 inhibitor that has the potential to acetylate the COX-2
a
N-acetylsulfonamido
isozyme.
# 2004 Elsevier Ltd. All rights reserved.
Cyclooxygenase (COX) inhibitors such as rofecoxib (1)1
and celecoxib (4)2 which selectively inhibit the inducible
COX-2 isozyme that causes inflammation, rather than
the constitutive COX-1 isozyme that provides gastro-
protection and maintains vascular homeostasis, are
clinically effective nonulcerogenic antiinflammatory
drugs. However, a precautionary concern regarding the
use of COX-2 inhibitors in patients at risk for an
adverse cardiovascular event such as myocardial infarc-
tion has been raised. One plausible explanation for this
increased incidence of a prothrombotic episode is
attributed to a lower level of the vasodilator and platelet
aggregation inhibitor prostacyclin (PGI2) in conjunction
with a higher level of the potent platelet activator and
aggregator thromboxane A2 (TxA2).3 Accordingly,
there is still a need for the design of COX-2 inhibitors
with a greater safety profile for the treatment of arthri-
tis. In this regard, a novel class of isomeric acetoxy
analogues of rofecoxib (2) were recently designed which
are potent and selective COX-2 inhibitors that, like
acetylsalicyclic acid (aspirin), have the potential to
acetylate the COX-2 isozyme.4 Other studies indicated
that the aspirin analogue O-(acetoxyphenyl)hept-2-ynyl
sulfide selectively acetylated and irreversibly inhibited
COX-2,5 and a diastereomeric acyl-Co-A-ketoprofen
conjugate may act as a reversible inhibitor of COX-1
and an irreversible inhibitor of COX-2.6 Extensive
structural–activity studies for the diphenylheterocycle
class have shown that a SO2NH2 or SO2Me substituent
at the para-position of one phenyl ring often provides
optimum COX-2 selectivity and potency.7 The SO2Me
and SO2NH2 H-bonding pharmacophores are believed
to induce COX-2 selectivity by insertion into the sec-
ondary (2ꢀ) pocket of COX-2 that is absent in COX-1.
The 2ꢀ-pocket in COX-2 has been attributed to the pre-
sence of isoleucine (Ile523) in COX-1 relative to the
smaller valine (Val523) in COX-2.8 In an earlier study we
showed that the dipolar azido substituent of 3 inserts
deep into the COX-2 2ꢀ-pocket binding site where it
undergoes a hitherto unreported electrostatic (ion–ion)
interaction with the charged guanidino moiety of
Arg513.9 It is also of interest to investigate the N-acetyl-
sufonamido (SO2NHCOMe) moiety as an pharmaco-
phore that is capable of acetylating the Ser530 hydroxyl
moiety in the primary binding site of COX-2.10 An azi-
dosulfonyl (SO2N3) substituent warrants investigation
as a COX-2 pharmacophore that is capable of under-
going dual H-bonding (SO2) and electrostatic (N3)
interactions with amino acid residues lining the COX-2
2ꢀ-pocket. As part of our ongoing program to design
COX-2 inhibitors, we now describe a novel group of
regioisomeric 3,4-diphenyl-2(5H)furanones (8–10) that
possess a SO2Me, SO2NH2, SO2NHCOMe or SO2N3
substituent at the para-position of either phenyl ring.
Keywords: N-Acetylsulfonamido;COX-2 inhibition.
* Corresponding author. Tel.: +1-780-492-5993;fax: +1-780-492-
1217;e-mail: eknaus@pharmacy.ualberta.ca
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.01.076

1958
A. Zarghi et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1957–1960
substituent are inactive inhibitors of COX-1 and COX-2
(IC50 values >100 mM). In contrast, incorporation of a
SO2N3 substituent at the para-position of the C-3
phenyl ring of rofecoxib (10a) conferred modest inhibi-
tory potency against COX-1 (IC50=31.5 mM) and
COX-2 (IC50=11 mM) with a moderate COX-2 selec-
tivity index (S.I.) of about 3. On the other hand, the
corresponding regioisomer 10b was a selective (COX-2
S.I.>31), but not particularly potent (COX-2
IC50=3.15 mM), inhibitor of COX-2. The N-acetyl-
sulfonamide regioisomers 9a and 9b, in view of their
potential ability to acetylate the COX-2 isozyme, could
provide a lead-compound for the development of a
novel type of acetylating COX-2 inhibitor. In vitro
COX-1/COX-2 enzyme inhibition studies showed that
incorporation of an additional para-N-acetylsulfonamido
substituent on the C-3 phenyl ring of rofecoxib provided
9a that was an approximately equipotent inhibitor of
both COX-1 (IC50=3.1 mM) and COX-2 (IC50=4.6 mM).
In contrast, the corresponding 4-[4-(N-acetylsulfon-
amido)phenyl]-3-(4-methanesulfonylphenyl)-2(5H)fur-
anone (9b) was a highly potent (COX-2 IC50=0.05
mM), and selective (COX-2 S.I. >2000) COX-2 inhi-
bitor relative to the reference drug rofecoxib (see data in
Table 1). These data suggest that the novel N-acetyl-
sulfonamdo compound 9b should inhibit the synthesis
of inflammatory prostaglandins via the COX pathway
at sites of inflammation, and be devoid of ulcerogenicity
due to the fact that it does not inhibit COX-1
(IC50>100 mM).
The target 3,4-diaryl-2(5H)furanone derivatives (8–10)
were synthesized using the reaction sequence illustrated
in Scheme 1. Reaction of the bromoketone (5a11 or b)
with the phenylacetic acid (6a or b12) proceeds via a 2-
step condensation–cyclization reaction performed as a
one-pot procedure.13 Thus, treatment of a mixture of
either 5a and 6a, or 5b and 6b, in acetonitrile with tri-
ethylamine at 25 ꢀC yields an ester intermediate. Sub-
sequent cooling to 0 ꢀC and then addition of DBU
effected the cyclization to afford the respective 4-(4-
methylsulfonylphenyl)-3-phenyl-2(5H)furanone or 3-(4-
methylsulfonyl)-4-phenyl-2(5H)furanone regioisomer (7a
or b, 53–57% yield). Chlorosulfonation of the furanones
7a or b with chlorosulfonic acid at 25 ꢀC, and then
reaction of the sulfonyl chloride intermediate with either
gaseous ammonia in THF, or NaN3 in aqueous acetone,
afforded the respective sulfonamide regioisomer (8a or
b, 100%), or sulfonylazide regioisomer (10a or b, 41–
44% yield). Acetylation of 8a and b using acetyl
chloride in acetic acid afforded the respective N-acetyl-
sulfonamide product (9a or b, 85–90%).
The orientation of the highly potent and selective
COX-2 inhibitor 4-[4-(N-acetylsulfonamido)phenyl]-3-
(4-methanesulfonylphenyl)-2(5H)furanone (9b) in the
COX-2 binding site was examined by a docking experi-
ment (Fig. 1).14 This molecular modeling shows that 9b
binds in the primary binding site such that the C-3 para-
SO2Me substituent inserts into the 2ꢀ-pocket present in
COX-2. One of the O-atoms of the SO2Me moiety is H-
bonding to the amide hydrogen (NH) of Phe518 (dis-
tance=3.6 A) whereas, the other O-atom is close to the
NH2 of Arg120 (distance=2.8 A). The C¼O oxygen
atom of the central furanone ring forms a hydrogen
bond (distance=3.3 A) with the OH of Tyr355 that
forms part of the entrance to the COX-2 2ꢀ-pocket. The
C-4 phenyl ring with a para-SO2NHCOMe substituent
A group of rofecoxib derivatives having an additional
SO2NH2, SO2NHCOMe or SO2N3 substituent at
the para-position of the C-3 phenyl ring (8a–10a), and the
corresponding rofecoxib regioisomers (8b–10b), were
prepared to investigate the effect of these substituents
on COX-2 selectivity and potency. In vitro COX-1/
COX-2 inhibition studies showed that the rofecoxib
analogues 8a and b possessing an additional SO2NH2
Table 1. In vitro inhibition of COX-1 and COX-2 by 3,4-diphenyl-
2(5H)furanone derivatives of rofecoxib (8–10)
IC50 (mM)a
COX-2
S.I.b
Compd
COX-1
COX-2
8a
8b
9a
9b
>100
>100
3.2
>100
31.5
>100
>100
4.6
—
—
<0.8
>2000
>2.8
>32
0.05
11
3.1
10a
10b
Rofecoxib
>100
>500
0.43
>1162
a Values are means of two determinations acquired using an ovine
COX-1/COX-2 assay kit (Catalog No. 560101, Cayman Chemicals,
Inc., Ann Arbor, MI), and the deviation from the mean is <10% of
the mean value.
Scheme 1. Reagents and conditions: (a) Et3N, MeCN, 25 ꢀC, 30 min,
and then DBU, 0 ꢀC, 30 min;(b) ClSO 3H, 25 ꢀC, 3 h;(c) NH gas,
3
THF, 5 min;(d) AcCl, AcOH, reflux, 30 min;(e) NaN 3, aqueous
acetone, 0 ꢀC, 2 h.
b In vitro COX-2 selectivity index (COX-1 IC50/COX-2 IC50).

A. Zarghi et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1957–1960
1959
Figure 1. Docking of 9b (ball and stick) in the active site of murine COX-2 (line and stick). Hydrogen atoms of the amino acid residues are not
shown to increase clarity.
is oriented towards a hydrophobic pocket comprised of
Trp387, Tyr385 and Tyr348 at the top of the COX-2 pri-
mary binding site. As per our hypothesis, the SO2NH-
COMe is located in the vicinity of Ser530 which is the
acetylation site for aspirin. In this regard, the C-atom of
the SO2NHCOMe group is positioned about 4.0 A
away from the OH of Ser530, and the SO2NHCOMe NH
and Ser530 OH are separated by almost 3.3 A. The SO2
moiety of the SO2NHCOMe substituent undergoes a
hydrophobic interaction with the backbone residues
Gly526, Ala527 and Met522. It is interesting to note that,
the O-atom of the C¼O (SO2NHCOMe) forms a
hydrogen bond with the OH of Tyr385 (distance=3.2 A)
which could potentially activate the C-atom of the
SO2NHCOMe moiety with respect to nucleophilic
attack by the OH of Ser530 that may lead to a covalent
acetylation of the COX-2 isozyme Ser530. This observa-
tion is consistent with the critical role of Tyr385 in the
acetylation of Ser530 by aspirin in the COX binding
site,15 and the in vitro COX inhibition data provide a
good explanation for the potent and selective inhibitory
activity exhibited by 9b.
compound 9b could serve as a useful probe to study the
function and catalytic activity of the COX-2 isozyme.
Acknowledgements
We are grateful to the Canadian Institutes of Health
Research (MOP-14712) for financial support of this
research, to Shaheed Beheshti University of Medical
Sciences for a visiting professorship (to A.Z.), and to
the Alberta Heritage Foundation for Medical Research
for a graduate scholarship (to P.R.).
References and notes
1. Hawkey, C.;Laine, L.;Simon, T.;Beaulieu, A.;
Maldonado-Cocco, J.;Acevedo, E.;Shahane, A.;Quan,
H.;Bolognese, J.;Mortensen, E. Arthritis Rheum. 2000,
43, 370.
2. Goldstein, J. L.;Silverstein, F. E.;Agrawal, N. M.;
Hubbard, R. C.;Kaiser, J.;Maurath, C. I.;Verburg,
K. M.;Geis, G. S. Am. J. Gastroenterol. 2000, 95, 1681.
3. Mukherjee, D.;Nissen, S. E.;Topol, E. J. JAMA 2001,
286, 954 and references cited therein.
The results of this investigation show (i) incorporation
of a para-N-acetylsulfonamido substituent on the C-3
phenyl ring of the rofecoxib regioisomer (9b) provides a
highly potent, and selective, COX-2 inhibitor, (ii) a
molecular modeling study indicates the para-N-acetyl-
sulfonamido substituent of 9b is suitably positioned to
acetylate the serine hydroxyl group in the COX-2
pimary binding site, and (iii) the N-acetylsulfonamido
4. Rahim, M. A.;Rao, P. N. P.;Knaus, E. E. Bioorg. Med.
Chem. Lett. 2002, 12, 2753.
5. (a) Kalgutkar, A. S.;Crews, B. C.;Rowlinson, S. W.;
Garner, C.;Seibert, K.;Marnett, L. J. Science 1998, 280,
1268. (b) Kalgutkar, A. S.;Kozak, K. R.;Crews, B. C.;
Hochgesang, G. P., Jr.;Marnett, J. R. J. Med. Chem.
1998, 41, 4800.

1960
A. Zarghi et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1957–1960
6. Levoin, N.;Chretien, F.;Lapicque, F.;Chapleur, Y.
Bioorg. Med. Chem. 2002, 10, 753.
7. Talley, J. J. Prog. Med. Chem. 1999, 36, 201.
8. Luong, C.;Miller, A.;Barnett, J.;Chow, J.;Ramesha, C.;
Browner, M. F. Nat. Struct. Biol. 1996, 3, 927.
structed using the Builder module and were energy mini-
mized for 1000 iterations reaching a convergence of 0.01
kcal/mol A. The energy minimized ligands were super-
imposed on SC-558 in the PDB file 1cx2 after which SC-
558 was deleted. The resulting ligand–enzyme complex
was subjected to docking using the Affinity command in
the Docking module of Insight II after defining subsets of the
enzyme such that residues within 10 A of the ligand were
allowed to relax, while the remainder of the enzyme residues
were fixed. The consistent valence force field (CVFF) was
employed for all docking purposes. The ligand–enzyme
assembly obtained after docking was then subjected to a
molecular dynamics (MD) simulation using the Discover
module Version 2.98 at a constant temperature of 300 K
with a 100 step equilibration for over 1000 iterations and
a time step of 1 fs using a distance dependent dielectric
constant 4r. The optimal binding orientation of the
ligand–enzyme assembly obtained was further minimized
for 1000 iterations using the conjugate gradient method
until a convergence of 0.001 kcal/mol A was reached.
15. Hochgesang, G. P.;Rowlinson, S. W.;Marnett, L. J.
J. Am. Chem. Soc. 2000, 122, 6514.
9. Habeeb, A. G.;Rao, P. N. P.;Knaus, E. E.
Chem. 2001, 44, 3039.
J. Med.
10. Sykes, N. O.;Macdonald, S. J. F.;Page, M. I. J. Med.
Chem. 2002, 45, 2850.
11. Culter, R. A.;Stenger, R. J.;Suter, C. M. J. Am. Chem.
Soc. 1952, 74, 5475.
12. Giroux, A.;Nadeau, C.;Han, Y. Tetrahedron Lett. 2000,
41, 7601.
13. Therien, M.;Gauthier, J. Y.;Leblanc, Y.;Leger, S.;
Perrier, H.;Prasit, P.;Wang, Z. Synthesis 2001, 12, 1778.
14. Docking studies were performed using Insight II software
Version 2000.1 (Accelrys Inc.) running on a Silicon Gra-
phics Octane 2 R14000A workstation. The coordinates of
the X-ray crystal structure of the selective COX-2 inhi-
bitor SC-558 bound to the murine COX-2 enzyme was
obtained from the RCSB Protein Data Bank (1cx2) and
hydrogens were added. The ligand molecules were con-