Structure-based design of cyclooxygenase-2 selectivity into ketoprofen.

Article abstract of DOI:10.1016/S0960-894X(01)00800-9

We have recently described how to achieve COX-2 selectivity from the non-selective inhibitor indomethacin (1) using a combination of a pharmacophore and computer 3-D models based on the known X-ray crystal structures of cyclooxygenases. In the present study we have focused on the design of COX-2 selective analogues of the NSAID ketoprofen (2). The design is similarly based on the combined use of the previous pharmacophore together with traditional medicinal chemistry techniques motivated by the comparative modeling of the 3-D structures of 2 docked into the COX active sites. The analysis includes use of the program GRID to detect isoenzyme differences near the active site region and is aimed at suggesting modifications of the basic benzophenone frame of the lead compound 2. The resulting series of compounds bearing this central framework is exemplified by the potent and selective COX-2 inhibitor 17 (LM-1669).

Full text of DOI:10.1016/S0960-894X(01)00800-9

Bioorganic & Medicinal Chemistry Letters 12 (2002) 533–537
Structure-Based Design of Cyclooxygenase-2 Selectivity
into Ketoprofen
Albert Palomer,* Jaume Pascual, Marta Cabre, Liset Borras, Gracia Gonzalez,
Monica Aparici, Assumpta Carabaza, Francesc Cabre,y M. Luisa Garcıa and
David Mauleon
R&D Department, Laboratorios Menarini S.A., Alfonso XII 587, 08918 Badalona, Spain
Received 26 June 2001; revised 26 September 2001; accepted 22 November 2001
Abstract—We have recently described how to achieve COX-2 selectivity from the non-selective inhibitor indomethacin (1) using a
combination of a pharmacophore and computer 3-Dmodels based on the known X-ray crystal structures of cyclooxygenases. In
the present study we have focused on the design of COX-2 selective analogues of the NSAIDketoprofen ( 2). The design is similarly
based on the combined use of the previous pharmacophore together with traditional medicinal chemistry techniques motivated by
the comparative modeling of the 3-Dstructures of 2 docked into the COX active sites. The analysis includes use of the program
GRIDto detect isoenzyme differences near the active site region and is aimed at suggesting modifications of the basic benzo-
phenone frame of the lead compound 2. The resulting series of compounds bearing this central framework is exemplified by the
potent and selective COX-2 inhibitor 17 (LM-1669). # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
Rofecoxib (3) and Celecoxib (4), (see Scheme 1), indu-
cing a great interest in obtaining isozyme-specific drugs.2
Non-steroidal antiinflammatory drugs (NSAIDs) are
useful tools for the treatment of inflammation, pain and
fever although they show an undesirable profile of gas-
tric side effects. Because NSAIDs directly target
cyclooxygenases (COXs), the discovery of the COX-2
isoform has opened the possibility of developing COX-2
selective inhibitors to act as an effective NSAIDwith
reduced gastric side effects.1 At present, two COX-2
selective inhibitors have successfully reached the market,
To our knowledge, several attempts to derive COX
selective inhibitors from the non-selective NSAIDs
indomethacin (1),3 ketoprofen (2),4 zomepirac5 or flur-
biprofen6 have been published. These strategies intro-
duced the desired selectivity by systematic structural
modification of the lead NSAIDs. Alternatively, selec-
tivity may be introduced by using the available infor-
mation on the tricyclic COX-2 selective inhibitors
Scheme 1.
*Corresponding author at present address: Centro de Investigacion, Ferrer Internacional S.A., Juan de Sada 32, 08028 Barcelona, Spain. Tel.: +34-
93-509-3266; fax: +34-93-411-2764; e-mail: apalomer-research@ferrergrupo.com
yPresent address: Preclinical Pharmacology, Vita-Invest S.A., Av. Barcelona 69, 08970 Sant Joan Despı, Barcelona, Spain. Tel.: +34-93-602-2425;
fax: +34-93-373-8751; e-mail: fcabre@vita-invest.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00800-9

534
A. Palomer et al. / Bioorg. Med. Chem. Lett. 12 (2002) 533–537
structurally related to 3 and 4. We have recently descri-
bed the derivation of COX-2 selectivity from the non-
selective inhibitor 17 based on a pharmacophore
accounting for the activity of the tricyclic COX-2 inhi-
bitors. This methodology allowed us to transfer struc-
tural information onto 1 and to identify a small set of
novel COX-2 selective inhibitors having the basic N-
benzyl-indole core.7
Results and Discussion
The modeled 3-Dstructures of the COX isoenzymes
complexed with ketoprofen (2) were used as the starting
point for this study.4 Comparison of the resulting keto-
profen COX-1 and -2 complexes were performed with
the program GRIDfor the presence of isoenzyme-spe-
cific binding sites.11 Calculations were carried out with
the oxygen-containing GRIDprobes including OH,
OC2, O, O:: and O1. The specific water (OH2) and sul-
fonyl oxygen probes (O¼for sulfonamide and OS for
sulfone) rendered most significant differences. Particular
attention was focused on the cavity near the active site
of COX-2 where the selective inhibitors of the tricyclic
series position the sulfonyl group.10 In this region,
GRIDwas able to assess the presence of energetically
favourable sulfonyl positions in COX-2, meanwhile,
could not predict equivalent sites of interaction in COX-
1. The results are shown in Figure 1 with the 5.5 A
spacing between the GRIDsulfonyl oxygen spot and the
preferred ketoprofen meta substitution. The disposition
allows sulfonyl-containing substituents extending 4–6 A.
Thus, a number of spacing groups accessible for synth-
esis were modeled, for example, vinyl, phenyl and so on.
Computer limitations excluded systematic modeling of
all possible substituents by performing enzyme–ligand
The structure of the COX-1 and COX-2 isoenzymes,
obtained from X-ray crystallography8 or by homology
modeling,9 are known. Kurumbail et al.10 have pub-
lished crystallographic structures of a COX-2 in free
form as well as complexed with the COX-2 selective
inhibitor SC-558 (5), the non-selective inhibitors indo-
methacin (1), and flurbiprofen. In this paper we detail
our strategy to obtain selective COX-2 inhibitors based
on the modification of the structure of the known
potent but non-selective COX inhibitor ketoprofen (2).
The strategy is intended to obtain selectivity using tradi-
tional medicinal chemistry techniques motivated by the
comparative modeling of a COX-1 and -2 complexed
with 2 together with the available pharmacophore. The
latter model contains structural information on the tri-
cyclic COX-2 selective inhibitors having the sulfonyl
group believed to play a crucial role on selectivity.
Figure 1. Computer modeled complex of ketoprofen (2) docked into the COX-2 active site: Interaction energy maps contoured at ꢀ5 kcal/mol
(yellow surface) obtained with the GRIDstandard sulfonamide probe (O ¼). The ligand and the additional water molecule considered in the com-
plex4 are shown together with a selected subset of the residues near the GRIDspots.

A. Palomer et al. / Bioorg. Med. Chem. Lett. 12 (2002) 533–537
535
complexes. Moreover, the success on the application of
the pharmacophore to the design of COX-2 selective
indomethacin analogues prompted us to use this
rational approach to the design of new compounds.7
The model bearing structural information on the tri-
cyclic COX-2 selective inhibitors, for example, having
the characteristic sulfonyl group believed to play a cru-
cial role on selectivity, is described in detail in literature7
and is shown in Figure 2. Moreover, the alignment of 5
and 17 illustrates that the sulfonylvinyl benzophenone
skeleton present in 17 may be suitable to obtain selective
COX-2 inhibition.
against COX-2. Because none of these compounds
showed activity, we assumed that, in addition to the
pharmacophore minimal requirements, an extra sub-
stituent on the benzophenone phenyl ring is needed.
Accordingly, compounds with an additional polar
hydroxymethyl (10 and 19), acetyl (20), cyano (21),
amide (22), amine (23) or 2-propanoate (12) were syn-
thesized together with the compounds 8, 9, 11 and 15–18,
having an extra alkyl substituent. Among these com-
pounds only the alkyl-substituted 3-sulfonylvinyl benzo-
phenones 16–18 were able to significantly inhibit the
COX-2 and only 17 inhibited this enzyme selectively (16
and 18 also presented a significant ability to inhibit
COX-1). These results are consistent with the absence of
activity described for the compounds of the sulfonyl tri-
cyclic series bearing, simultaneously, a carboxylate.14 All
in all, we hypothesized that the extra substituent should
be neither polar nor acidic and, most importantly, that it
is an orienting group crucial to retain the pharmacophore
geometry. In order to assess the importance of the sulfo-
nyl group for selectivity, the sulfonamide group in 17 was
modified for sulfone (24), ester (25), carboxylate (26) and
carboxamide (27) but none of the variations improved
The modeling results prompted us to synthesize com-
pounds 6–27 having the essential sulfonylvinyl benzo-
phenone frame. Our primary in vitro screening scheme
included the inhibition of the PGE2 generation by LPS-
stimulated monocytes isolated from human blood
(COX-2)12 and of the TxB2 generation in the presence of
1 mM arachidonic acid by platelets isolated from human
blood (COX-1).13 In vitro COX inhibition results are
summarized in Table 1. In this set of in vitro tests,
compound 17 demonstrated to inhibit the COX-2
activity (IC50=0.25 mM) with moderate effect on COX-1
(IC50=18 mM; see Table 1).
the COX-2 inhibition. Finally, in the human whole blood
COX-2
assay15 17 inhibited the COX activity (IC50
=
12.0ꢁ0.7 mM and IC50COX-1=100ꢁ3.3 mM) with
comparable selectivity to the reference compound 3
(0.83ꢁ0.13 and 18.3ꢁ5.3 mM respectively).
Benzophenones 6, 7, 13 and 14, having the additional
sulfonylvinyl substituent, were synthesized and tested
Figure 2. Pharmacophore with 5 (SC-558, green) and 17 (red) fitted. The pharmacophore contains a sulfonyl S-atom, an aromatic ring plane A with
a fixed position of the normal to the plane, an additional aromatic ring plane B and an excluded volume. The chemical features are drawn as brown
globes except for the additional globe accounting for the position orthogonal to the aromatic ring plane A and the excluded volume (black globe).

536
A. Palomer et al. / Bioorg. Med. Chem. Lett. 12 (2002) 533–537
Scheme 2.
Synthesis
groups of the formed 5-bromo-isophthalic acid dimethyl
ester was then methylated [CH3MgCl, LiN(SiCH3)] to
produce the acetyl that was reduced with NH2NH2.H2O/
KOH to form the 3-bromo-5-ethylbenzoic acid. Friedel–
Craft reaction of the corresponding acyl chloride (PCl5/
CCl4, 60ꢂC) with benzene (AlCl3, 80ꢂC) produced the 5-
amino-isophthalic acid dimethyl ester that was reacted
with ethenesulfonic acid amide to give the desired pro-
duct 17 in 20% yield over the 5 steps (see Scheme 2).
The compounds in Table 1, exemplified by the synthesis
of 17 shown in Scheme 2, were prepared as previously
described in the literature.16,17 In summary, 17 was
obtained as follows: the amino group in 5-amino-iso-
phthalic acid dimethyl ester was converted into the cor-
responding bromo derivative by treatment with NaNO2/
HBr followed by HBr/CuBr. One of the carboxylate
Table 1. In vitro COX inhibition results in human blood monocytes (COX-2) and platelets (COX-1)
Compd
Subst.a
R1
R2
COX-2,a,b IC50 (mM)
COX-1,b,c IC50 (mM)
6
7
8
9
10
11
12
13
14
15
16
A
B
B
B
B
B
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
SO2NH2
SO2CH3
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2CH3
SO2CH3
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2CH3
CO2CH2CH3
CO2H
H
H
3-CH3
3-CH2CH3
3-CH2OH
2-CH3
43% @ 20 mM
6% @ 20 mM
14% @ 20 mM
5.6
8.4
41% @ 20 mM
9.7
Inact. @ 500 mM
37% @ 20 mM
19
39% @ 10 mM
14% @ 10 mM
0% @ 10 mM
6.3
25% @ 10 mM
0% @ 10 mM
n.t.
n.t.
0% @ 10 mM
49% @ 10 mM
0.20
3-CH(CH3)COOH
H
H
2-CH3
2-CH2CH3
3-CH2CH3
3-CH=C(CH3)2
3-CH2OH
3-COCH3
3-CN
3-CON(CH3)2
3-CH2NH(CH2CH3)2
3-CH2CH3
3-CH2CH3
3-CH2CH3
3-CH2CH3
0.77
0.20
0.98
9.7
20
5.6
17 (LM-1669)
18
19
20
21
22
23
24
18
3.6
5.4
6.4
1.7
n.t.
n.t.
31% @ 20 mM
7.1
40% @ 20 mM
5.0
16% @ 10 mM
n.t.
n.t.
25
26
27
2.6
1.8
0.026
0.28
0.11
CONH2
0.074
2 (ketoprofen)d
3 (rofecoxib)
4 (celecoxib)
0.0020
0% @ 10 mM
1.88
aPGE2 generation by LPS-stimulated monocytes isolated from human blood.
bDose–response curves were analysed by non-linear regression using the Hill equation as implemented in the software Prism.18 The values shown
correspond either to the IC50 or to the % inhibition produced by tested compounds at the selected doses. Values are the mean of at least two
independent determinations.
cTxB2 generation in the presence of 1 mM arachidonic acid by platelets isolated from human blood.
dActivity was measured with the S-(+)-enantiomer of ketoprofen.

A. Palomer et al. / Bioorg. Med. Chem. Lett. 12 (2002) 533–537
537
Conclusions
6. Bayly, C. I.; Black, C.; Leger, S.; Ouimet, N.; Ouellet, M.;
Percival, M. D. Bioorg. Med. Chem. Lett. 1999, 9, 307.
7. Palomer, A.; Cabre, F.; Pascual, J.; Campos, J.; Trujillo,
M. A.; Entrena, A.; Gallo, M. A.; Garcıa, M. L.; Mauleon,
D.; Espinosa, A. J. Med. Chem. 2002, in press.
8. Picot, D.; Loll, P. J.; Garavito, R. M. Nature 1994, 367,
243. McKeever, B. M.; Pandya, S.; Percival, M. D.; Ouellet,
M.; Bayly, C.; O’Neill, G. P.; Bastien, L.; Kennedy, B. P.;
Adam, M.; Cromlish, W.; Roy, P.; Black, W. C.; Guay, D.;
This work describes the development of novel COX-2
selective inhibitors starting from a non-selective inhi-
bitor ketoprofen (2) to obtain the potent and selective
inhibitor 17. Traditional medicinal chemistry techniques
have been employed driven by the combined use of a
pharmacophore together with the comparative model-
ing of COX-1 and 2 using GRIDto detect isoenzyme
differences. The optimal activity has been obtained with
the 3-sulfonylvinyl benzophenone frame with an addi-
tional non-polar substituent required for selectivity, for
example meta-ethyl. 17 (LM-1669) has shown to selec-
tively inhibit COX-2 in human isolated cell and whole
blood assays.
LeBlanc, Y. 10th Conf. Prost. RelatedCpsd.
(abstr. 63).
1996, 55, 20
9. Filizola, M.; Perez, J. J.; Palomer, A.; Mauleon, D. J. Mol.
Graph. 1997, 15, 290.
10. 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. Nature 1996, 384, 644.
11. GRIDversion 17.0. MDL, Oxford, 1998. Goodford,
P. J. A. J. Med. Chem. 1985, 28, 849. Boobbyer, D. N. A.;
Goodford, P. J.; McWhinnie, P. M.; Wade, C. R. J. Med.
Chem. 1993, 36, 148.
References and Notes
1. Smith, W. L.; Garavito, R. M.; DeWitt, D. L. J. Biol.
Chem. 1996, 271, 33157.
12. Carabaza, A.; Cabre, F.; Rotllan, E.; Gomez, M.; Gutier-
rez, M.; Garcıa, M. L.; Mauleon, D. J. Clin. Pharmacol. 1996,
36, 505.
13. Riendeau, D.; Charleson, S.; Cromlish, W.; Mancini, J. A.;
Wong, E.; Guay, J. Can. J. Phys. Pharmacol. 1997, 75, 1088.
14. Reitz, D. B.; Isakson, P. C. Curr. Pharm. Des 1995, 1, 211.
15. Brideau, C.; Kargman, S.; Liu, S.; Dallob, A. L.; Ehrich,
E. W.; Rodger, I. W.; Chan, C. C. Inflammation. Res. 1996,
45, 68.
2. Ehrich, E.; Dallob, A.; van Hecken, A.; Depre, M.; DeLe-
peleire, I.; Brideau, C.; Hilliard, D.; Tanaka, W.; Mukho-
padhyay, S.; Seibold, J.; DeSchepper, P.; Gertz, B. Arthr.
Rheum. 1996, 39, s81 (abstr. 328). 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.; Bur-
ton, E. G.; Cogburn, J. N.; Gregory, S. A.; Koboldt, C. M.;
Perkins, W. E.; Seibert, K.; Veenhuizen, A. W.; Zhang, Y. Y.;
Isakson, P. C. J. Med. Chem. 1997, 40, 1347.
3. Black, W. C.; Bayly, C.; Belley, M.; Chan, C. C.; Charleson,
S.; Denis, D.; Gauthier, J. Y.; Gordon, R.; Guay, D.; Kargman,
S.; Lau, C. K.; Leblanc, Y.; Mancini, J.; Ouellet, M.; Percival,
D.; Roy, P.; Skorey, K.; Tagari, P.; Vickers, P.; Wong, E.; Xu,
L.; Prasit, P. Bioorg. Med. Chem. Lett. 1996, 6, 725.
4. Palomer, A.; Perez, J. J.; Navea, S.; Llorens, O.; Pascual,
J.; Garcıa, M. L.; Mauleon, D. M. J. Med. Chem. 2000, 43,
2280.
16. Mauleon, D.; Garcia, M. L.; Palomer, A.; Pascual, J.
Spanish Patent Application 9,902,287, 1999.
17. Analytical data for the selected active compound: 2-(3-
Benzoyl-5-ethyl-phenyl)-ethenesulfonic acid amide (17, LM-
1669). Mp 123–126 ꢂC; 1H NMR (300 MHz, CDCl3): d 1.24
(3H, t, CH3), 2.20 (2H, broad, NH), 2.59 (2H, d, CH2), 7.29
(1H, d, ethyl–H), 7.35 (1H, d, Ar–H), 7.36 (2H, dd, Ar–H), 7.44
(1H, s, Ar–H), 7.45 (1H, dd, Ar–H), 7.55 (1H, ddd, ethyl–H),
7.56 (1H, d, Ar–H), 7.70 (2H, dd, Ar–H); 13C NMR (100 MHz,
CDCl3): d 187.0, 139.8, 137.8, 137.6, 136.0, 134.5, 132.2, 130.1,
130.6, 129.3, 128.7, 128.4, 128.2, 125.0, 112.0, 29.0, 16.1.
18. GraphPad PRISM Version 2.0. GraphPad Software: San
Diego, CA, 1998.
5. Luong, C.; Miller, A.; Barnett, J.; Chow, J.; Ramesha, C.;
Browner, M. F. Nat. Str. Biol. 1996, 3, 927.