Racemic and chiral sulfoxides as potential prodrugs of 4-pyrone COX-2 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.03.101

The preparation of the sulfoxide analogues 7, 8, and 9 and their enantiomerically pure forms is discussed as well as their ability to act as prodrugs of the potent and selective sulfone-containing COX-2 inhibitors 1, 2, and 3. Sulfoxide derivatives 7 and

Full text of DOI:10.1016/j.bmcl.2006.03.101

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3605–3608
Racemic and chiral sulfoxides as potential prodrugs
of 4-pyrone COX-2 inhibitors
b
b
b
Francisco Caturla,a, Merce Amat, Raquel F. Reinoso, Elena Calaf and
*
`
Graham Warrellowa
aDepartment of Medicinal Chemistry, Almirall Prodesfarma S.A., Research Center, Cardener 68-74, 08024 Barcelona, Spain
bDepartment of Biology, Almirall Prodesfarma S.A., Research Center, Cardener 68-74, 08024 Barcelona, Spain
Received 30 January 2006; revised 15 March 2006; accepted 16 March 2006
Available online 2 May 2006
Abstract—The preparation of the sulfoxide analogues 7, 8, and 9 and their enantiomerically pure forms is discussed as well as their
ability to act as prodrugs of the potent and selective sulfone-containing COX-2 inhibitors 1, 2, and 3. Sulfoxide derivatives 7 and 9
were shown to be rapidly transformed in vivo into the corresponding sulfone derivatives 1 and 3, after oral administration to rats.
Ó 2006 Elsevier Ltd. All rights reserved.
F
Cl
O
Br
O
In spite of recent reports of side effects, the commer-
cial utilization of selective COX-2 inhibitors in the
treatment of pain and inflammatory disorders is well
established and such compounds may also have
potential in other therapeutic areas.1 We have been
interested in the design of novel COX-2 inhibitors in
our laboratories for some time.2 During the course
of these investigations we have discovered the poten-
tial of certain arylsulfoxides to act as both COX-2
inhibitors in their own right but also potential
prodrugs to the prototypical sulfones found in many
COX-2 inhibitors described in the literature. In
addition, we have found that such sulfoxides often
offer distinct advantages in terms of solubility and
pharmacokinetic profiles over the parent sulfones.
F
O
O
F
O
O
F
O
O
O
O
S
O
S
O
S
O
O
O
3
2
1
The medicinal use of sulfoxide-containing active agents
is fairly common. For example, such a moiety is found
in the H+/K+ ATPase inhibitor omeprazole (LosecÒ)
or its single enantiomer form esomeprazole (NexiumÒ),
in certain p38 kinase inhibitors,3 or in investigational
drugs such as RP52891 (aprikalim). However, the ability
of sulfoxides to act as prodrugs of sulfones in vivo is not
well documented and we are aware of only one other
recent report, also in the area of COX-2 inhibitors where
this concept has been described.4 In this paper, we
report our investigations into the biological activities
and ADME profiles of sulfoxides as potential prodrugs
of sulfones.
Given the interesting biological profiles presented by
the racemic sulfoxides synthesized, we decided to
undertake the enantioselective synthesis and biological
profiling of several chiral sulfoxides as potential
prodrugs of COX-2 inhibitors. We selected three
preclinical candidates (1–3) originating from our own
laboratories, shown below, in order to further explore
these findings.
The synthetic strategy employed to prepare the sulfoxide
derivatives was based on the racemic or enantioselective
oxidation of the corresponding thioethers.5 In the case
of pyran-4-one derivatives, the thioether intermediates
Keywords: COX-2; Sulfoxides; Prodrugs.
*
Corresponding author at present address: Treball 2-4, 08960 St. Just
Desvern. Tel.: +34 93 291 3583; fax: +34 93 312 8635; e-mail:
jfcaturl@almirall.es
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.03.101

3606
F. Caturla et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3605–3608
Table 1. Enantiomeric excesses of pyrones 7–9 obtained via assymetric
oxidation
were synthesized following
a protocol previously
described by our laboratories2c in which a series of
2-(4-methanesulfanylphenyl)pyran-4-one
derivatives
Substrate
R
Product (R)-Enantiomer (S)-Enantiomer
ee (%)
ee (%)
4–6 were prepared from the corresponding phenacyl
derivatives 10–12 by heating in a mixture of PPA
in Ac2O at 100 °C. The racemic sulfoxides 7–9 were
subsequently obtained by oxidation of the corresponding
thioethers with NaIO4 in MeOH–water (ratio 2:1) in
moderate to good yields (43–77%) (Scheme 1).
4
5
6
F
7
8
9
100
100
98.8
88.4
98.4
98.2
Cl
Br
Several asymmetric oxidation methods were attempted
to gain access to the enantiomerically enriched sulfox-
ides (R)- or (S)-7–9. Utilizing methodologies described
by Kagan6 or Bolm7 provided compounds with only
slight enantiomeric excesses. However, excellent results
were obtained utilizing the Modena catalytic system8
using Ti(OiPr)4 and either (R,R)- or (S,S)-DET (1:4)
and t-BuOOH at À20 °C in 1,2-dichloroethane, as
depicted in Scheme 2.
The enantiomeric excesses were determined using
capillary electrophoresis (Table 1) and were in
accordance with those determined by H NMR using
1
Pirkle’s alcohol9 as a shift reagent.
The absolute configuration at the sulfur center was as-
signed in each case following precedents described in
the literature.10 Thus, the (R)-sulfoxides were obtained
using (R,R)-DET as a chiral source, while the use of
(S,S)-DET provided the (S)-sulfoxides. An X-ray crystal
structure of compound (S)-7 (CCDC deposition num-
ber: 601850) confirmed the stereochemical assignment
(Fig. 1). It was subsequently assumed that this rule
would hold for the other sulfoxides synthesized.
Figure 1. X-ray structure of sulfoxide (S)-7.
R
O
R
O
F
O
F
O
O
F
O
a
b
O
R
17-21 %
43-77 %
MeS
O
Me
Me
Me
S+
O
MeS
10 (R=F)
11 (R=Cl)
12 (R=Br)
4 (R=F)
5 (R=Cl)
6 (R=Br)
7 (R=F)
8 (R=Cl)
9 (R=Br)
Scheme 1. Reagents and conditions: (a) PPA, Ac2O, 100 °C; (b) NaIO4, MeOH–H2O (2:1).
R
R
O
R
O
O
F
O
O
O
F
F
a or b
O
O
or
17-57%
Me
Me
O
Me
Me
Me
S+
O
S+
O
MeS
(R)-7 (R=F)
(R)-8 (R=Cl)
(R)-9 (R=Br)
(S)-7 (R=F)
(S)-8 (R=Cl)
(S)-9 (R=Br)
4 (R=F)
5 (R=Cl)
6 (R=Br)
Scheme 2. Reagents and conditions: (a) Ti(OiPr)4/(R,R)-DET (1:4), tert-BuOOH, 1,2-DCE, À20 °C; (b) Ti(OiPr)4/(S,S)-DET (1:4), t-BuOOH, 1,2-
DCE, À20 °C.

F. Caturla et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3605–3608
3607
All compounds described herein were tested for their
ability to inhibit human COX-1 and COX-2 using the
whole blood assay described by Patrignani et al.11 The
sulfoxides were found to be highly potent versus
COX-2. The COX-1/COX-2 selectivity of the sulfoxides
was similar to the corresponding sulfones of the 4-pyro-
nes and comparable to that of an included standard,
Etoricoxib (Merck). For the enantiomerically pure
sulfoxides, the S-forms were more potent for COX-2
and showed higher COX-1/COX-2 selectivity than the
corresponding R-forms (Table 2).
microsomes showed that the corresponding sulfone was
the main metabolite formed in all cases (data not shown).
In addition, all sulfoxides were assayed in human whole
blood and were found to be stable (data not shown).
Table 4. Therapeutic activity
Compound
Adjuvant arthritis EC50 or
% inhibition (dose) (mg/kg)
Pyresis ED50
(mg/kg)
1
68% (0.1)
63% (0.1)
60% (0.01)
59% (0.1)
56% (0.1)
58% (0.01)
0.440
0.7
0.8
3.2
1.3
2.3
2.8
1.0
0.9
2
3
In vitro studies using pooled microsomes from rat and
human livers showed that all the sulfoxides assayed were
significantly metabolized. In rat microsomes, the extent of
metabolism of the sulfoxides was greater than in human
microsomes (Table 3). Parallel studies carried out after
incubation of the sulfoxides with human hepatic
7
8
9
Etoricoxib
Rofecoxib
0.300
Data are indicated as ED50 (mg/kg) using four doses (six to eight
animals per group) or percentage of inhibition with dose (mg/kg) in
parentheses. Since SEM values never exceeded 15% of the media, they
have been omitted.
Table 2. In vitro inhibitory activities against COX-2 and selectivity
COX-1/COX-2 in human whole blood
Compound
COX-2 [IC50
(lM)
]
Selectivity ratio
COX-1/COX-2
rat, 1 mg/kg p.o.
4000
1
0.08
0.20
0.15
0.92
2.10
0.27
0.67
2.47
0.25
0.57
1.58
0.20
0.76
0.81
279
94
7
2
1 (7)
3
91
78
1
3000
7
(R)-7
(S)-7
8
24
320
121
30
2000
1000
0
(R)-8
(S)-8
9
400
135
20
(R)-9
(S)-9
Rofecoxib
Etoricoxib
430
15
0
1
2
3
4
5
6
122
Time (h)
Figure 2. Plasma concentrations of sulfoxide 7 (prodrug) and sulfone 1
(drug) after oral administration of prodrug or drug at 1 mg/kg to male
Wistar rats as a suspension of 0.5% methylcellulose and 0.1% Tween
80. Each data point represents the means SD of three values.
Table 3. In vitro metabolism and thermodynamic solubility of
sulfoxide prodrugs (7–9) and parent drugs (1–3)
Compound
Microsome
metabolisma
(%)
Thermodynamic solubilityb
(lg/mL)
rat, 1 mg/kg p.o.
Human Rat SGFc (pH 1.8) PBSd (pH 7.2)
300
1
2
6
26
27
34
54
55
38
86
87
67
77
93
80
9
18
4
7
5
9
2
3 (9)
3
12
24
20
17
57
48
41
64
62
42
1
3
7
527
—
—
70
—
—
31
—
—
150
—
—
37
—
—
15
—
—
200
(R)-7
(S)-7
8
(R)-8
(S)-8
9
100
0
(R)-9
(S)-9
0
1
2
3
4
5
6
a Assay conditions: metabolism (1 mg/mL microsomal protein, 5 lM
compound, 30 min and 37 °C incubation).
Time (h)
b Thermodynamic solubility for crystalline compound (Target con-
centration: 1 mg/mL, 24 h and 37 °C incubation).
c SGF, simulated gastric fluid.
Figure 3. Plasma concentrations of sulfoxide 9 (prodrug) and sulfone 3
(drug) after oral administration of prodrug or drug at 1 mg/kg to male
Wistar rats as a suspension of 0.5% methylcellulose and 0.1% Tween
80. Each data point represents the means SD of three values.
d PBS, phosphate buffer saline.

3608
F. Caturla et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3605–3608
Table 5. Pharmacokinetic parameters following sulfoxide (7 and 9) and sulfone (1 and 3) administration to male Wistar rats
Parameter
7
1
9
3
7
1
9
3
Cmax (ng/mL)
tmax (h)
AUC (ng.h/mL)
1309 (195)
0.3 (0.1)
1314 (174)
2959 (171)
2.3 (1.2)
18646 (5255)
1507 (245)
1.0 (0.0)
9216 (3423)
37 (23)
0.3 (0.0)
43 (19)
243 (26)
1.5 (1.3)
1125 (280)
177 (31)
2.2 (0.2)
1268 (445)
Results expressed as means (n = 3) and SD (in parentheses). Assay conditions: pharmacokinetic parameters of prodrug or parent drugs following
single oral administration of 1 mg/kg to male Wistar rats as a suspension of 0.5% methylcellulose and 0.1% Tween 80.
In general, the metabolism tended to be slightly lower
for the S-forms compared to the R-forms both in rat
and human microsomes (Table 3). These data suggest
that the R-form of sulfoxide 9 could be the best prodrug
in humans, based on its metabolism in human micro-
somes (Tables 2 and 3).
no significant improvement in its pharmacodynamics
was found.
In conclusion, the present study shows that the sulfox-
ides of 4-pyrones have COX-2 activity and good
COX-1/COX-2 selectivity. After oral administration of
the sulfoxides and the corresponding sulfones in the
rat, higher levels of the sulfone were achieved from sulf-
oxide prodrug than the sulfone itself. These results show
that such sulfoxides may be of use as prodrugs of the
corresponding 4-pyrone sulfones. Studies on potential
sulfoxide prodrugs of the sulfone-containing products
Etoricoxib and Rofecoxib will be reported in due course.
Regardless of the pH, the solubility of a given sulfoxide
was always higher than that of the corresponding sul-
fone. The greatest difference was found for the sulfoxide
7 at the acidic pH, with a solubility almost 60 times
higher than that of the sulfone 1 (Table 3).
Therapeutic activities were assessed for all racemic sulf-
oxides and the corresponding sulfones in the yeast-in-
duced pyresis model and the adjuvant-induced arthritis
model. The yeast-induced pyresis in a therapeutic proto-
col is an acute model. In this model, temperature mea-
surements were taken at 1 h intervals from 1 to 5 h
after single oral administration of the compounds. As
a chronic model of inflammation, the compounds were
tested in the adjuvant-induced arthritis in male Wistar
rats in a therapeutic protocol. The test compounds were
administered orally once daily for 7 days starting from
day 14 after arthritis induction and paw edema was mea-
sured 24 h after the last administration. The results
showed that the sulfoxides demonstrated good oral effi-
cacy at comparable doses to the sulfones, and higher
efficacy than that of both Etoricoxib and Rofecoxib in
the adjuvant arthritis model (Table 4).
References and notes
1. FitzGerald, G. A. Nat. Rev. Drug Discov. 2003, 2, 879.
2. (a) Puig, C.; Crespo, M. I.; Godessart, N.; Feixas, J.;
´
Ibarzo, J.; Jimenez, J. M.; Soca, L.; Cardelus, I.; Heredia,
A.; Miralpeix, M.; Puig, J.; Beleta, J.; Huerta, J. M.;
´
Lopez, M.; Segarra, V.; Ryder, H.; Palacios, J. M. J. Med.
Chem. 2000, 43, 214; (b) Feixas, J.; Jimenez, J. M.;
´
´
Godessart, N.; Puig, C.; Soca, L.; Crespo, M. I. Bioorg.
´
Med. Chem. Lett. 2001, 11, 2687; (c) Caturla, F.; Jimenez,
J. M.; Godessart, N.; Amat, M.; Cardenas, A.; Soca, L.;
´
Beleta, J.; Ryder, H.; Crespo, M. I. J. Med. Chem. 2004,
47, 3874.
3. de Laslo, S. E.; Visco, D.; Agarwal, L.; Chang, L.; Chin,
J.; Croft, G.; Forsyth, A.; Fletcher, D.; Frantz, B.;
Hacker, C.; Hanlon, W.; Harper, C.; Kostura, M.; Li,
B.; Luell, S.; MacCoss, M.; Mantlo, N.; O’Neill, E. A.;
Orevillo, C.; Pang, M.; Parsons, J.; Ronaldo, A.; Sahly,
Y.; Sidler, K.; Widmer, W. R.; O’Keefe, S. J. Bioorg. Med.
Chem. Lett. 1998, 8, 2689.
4. Moh, J. H.; Choi, Y. H.; Lim, K. M.; Lee, K.-W.; Shin, S.
S.; Choi, J. K.; Koh, H. J.; Chung, S. Bioorg. Med. Chem.
Lett. 2004, 14, 1757.
5. Caturla Javaloyes, F.; Warrellow, G. WO Patent
04072058, 2004.
6. Brunel, J.-M.; Diter, P.; Duetsch, M.; Kagan, H. B.
J. Org. Chem. 1995, 60, 8086.
7. Bolm, C.; Bienewald, F. Angew. Chem., Int. Ed. Engl.
1995, 34, 2640.
8. Di Furia, F.; Modena, G.; Seraglia, R. Synthesis 1984,
325.
The pharmacokinetic profiles were evaluated for the
sulfoxides 7 and 9, and compared to those of the cor-
responding sulfones 1 and 3. Oral administration of 7
and 9 to male Wistar rats (1 mg/kg) resulted in rapid
absorption followed by quick in vivo conversion of
the sulfoxides into the corresponding sulfones 1 and
3 (Figs. 2 and 3). The sulfone-to-sulfoxide AUC ratios
were 14 and 28 for 1 and 3, respectively. In each case,
plasma concentrations of the sulfone, derived from
prodrug administration of the sulfoxide, at the earliest
measured time point (0.25 h) were approximately two-
fold higher than those obtained after direct adminis-
tration of the sulfone itself. Similar increases were
noted in the Cmax and AUC parameters for sulfone
1. For sulfone 3, a tendency to higher Cmax and sim-
ilar AUC was observed (Table 5).
9. (R) or (S)-1-Anthracen-9-yl-2,2,2-trifluoro-ethanol.
10. Kagan, H. B. In Catalytic Asymmetric Synthesis; Ojima,
I., Ed.; VCH: New York, 1993; pp 203–226.
11. Patrignani, P.; Panara, MR.; Greco, A.; Fusco, O.; Natoli,
`
C.; Iacobelli, S.; Chipollone, F.; Ganci, A.; Creminon, C.;
However, although the pharmacokinetic properties
of sulfone 1 improved after prodrug administration,
Maclouf, J.; Patrono, C. J. Pharmacol. Exp. Ther. 1994,
271, 1705.