Synthesis and pharmacological profile of 6-methyl-3-isopropyl-2H-1,2-benzothiazin-4(3H)-one 1,1-dioxide derivatives: Non-steroidal anti-inflammatory agents with reduced ulcerogenic effects in the rat

Article abstract of DOI:10.1016/S0928-0987(02)00046-5

The new anti-inflammatory agents 6-methyl-3-isopropyl-2H-1,2-benzothiazin-4(3H)-one 1,1-dioxide 6a and its analogues 6b-f were synthesized from L-valine. All compounds were characterized by physical, chemical and spectral studies. Preliminary pharmacologi

Full text of DOI:10.1016/S0928-0987(02)00046-5

European Journal of Pharmaceutical Sciences 16 (2002) 221–228
www.elsevier.nl/locate/ejps
Synthesis and pharmacological profile of 6-methyl-3-isopropyl-2H-1,2-
benzothiazin-4(3H)-one 1,1-dioxide derivatives: non-steroidal
anti-inflammatory agents with reduced ulcerogenic effects in the rat
Yakdhane Kacema, Jamil Kraiema, Emna Kerkenib, Abderrahman Bouraouib , , Bechir Ben Hassine
a
*
´
a
`
´
`
´
Laboratoire de Synthese Organique Asymetrique et Catalyse Homogene, Faculte des Sciences, Avenue de l’Environnement, 5000 Monastir, Tunisia
b
´
Laboratoire de Pharmacologie, Faculte de Pharmacie, 5000 Monastir, Tunisia
Received 23 October 2001; received in revised form 2 April 2002; accepted 8 April 2002
Abstract
The new anti-inflammatory agents 6-methyl-3-isopropyl-2H-1,2-benzothiazin-4(3H)-one 1,1-dioxide 6a and its analogues 6b–f were
synthesized from L-valine. All compounds were characterized by physical, chemical and spectral studies. Preliminary pharmacological
evaluation of the resulting products showed that compounds 6a–f (5–20 mg/kg, i.p.) are active anti-inflammatory agents in
carrageenan-induced rat paw oedema assay in albino rats, and their effects are comparable to that of piroxicam (5 mg/kg, i.p.), used as a
reference drug. The nature of the substituents on the sulfonamide nitrogen and those on position three had a pronounced effect on the
anti-inflammatory activity. Studies of structure–activity relationships have led to selection of compound 2,6-dimethyl-3-isopropyl-1,2-
benzothiazin-3,4-diol 1,1-dioxide 6f which exhibited the most potent activity (61.7% inhibition at 5 mg/kg, i.p. and ED5054.5 mg/kg,
i.p.). Comparison of the gastrointestinal safety of compounds 6a–f with that of piroxicam showed a far better tolerability for our products.
This comparison was based on the ulcer index and the pH of gastric content.
2002 Elsevier Science B.V. All rights reserved.
Keywords: Non-steroidal anti-inflammatory; Piroxicam; Synthesis; Ulcerogenic; Rat
1. Introduction
currently available NSAIDs, including indomethacin and
piroxicam, are more potent inhibitors of COX-1 than of
COX-2 (Vane and Botting, 1995). This preferential inhibi-
tion of COX-1 may be responsible for many of the adverse
effects associated with NSAIDs. It has been postulated that
NSAIDs which preferentially inhibit COX-2, such as
meloxicam (Lipscomb et al., 1998), celecoxib (Simon et
al., 1998), rofecoxib (Scott and Lamb, 1999) and several
experimental drugs including NS 398, L-745,337 and DFP,
should produce the same or better anti-inflammatory
effects with less gastrointestinal, hematologic and renal
toxicities than classical NSAIDs.
On this basis, we directed our attention to the synthesis
of novel 6-methyl-3-isopropyl-2H-1,2-benzothiazin-4(3H)-
one 1,1-dioxide derivatives 6a–f (Fig. 1) related to ox-
icams with the aim of improving their anti-inflammatory
activity and reducing their ulcerogenic properties as it
appeared to be plausible that variation of the active
compound structures could exert a pronounced influence
on activity, as was the case with meloxicam 8 (Fig. 1)
Non-steroidal anti-inflammatory drugs (NSAIDs) are
very widely used to treat a variety of acute and chronic
inflammatory diseases. Today these drugs are being in-
creasingly used for the treatment of postoperative pain
(Moote, 1992), with or without supplemental opioid
agents. The pharmacological action of these agents was
assigned to inhibit two enzymes, known as
cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-
2) (Vane et al., 1998). The constitutive isoform COX-1 is
present in most tissues and is involved in the synthesis of
prostaglandins vital to normal cell function. In contrast, the
inducible isoform COX-2 appears to be produced primarily
in response to growth factors or inflammatory mediators,
such as cytokines (Vane and Botting, 1996). Many of the
*
Corresponding author. Tel.: 1216-73-461-000; fax: 1216-73-461-
830.
E-mail address: abder.bouraoui@fphm.rnu.tn (A. Bouraoui).
0928-0987/02/$ – see front matter
PII: S0928-0987(02)00046-5
2002 Elsevier Science B.V. All rights reserved.

222
Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
piroxicam (5 mg/kg) were dissolved in a vehicle solution
consisting of Tween 80/absolute ethanol/saline solution
(0.9%) in the ratio 1:1:18 and were administered intraperi-
toneally. Control rats received an equal volume of vehicle.
After 1 h, the rats were injected with 0.1 ml of 1%
carrageenan solution into the subplantar surface of the
right hind paw. The volume of the paw was measured
using a glass plethysmometer (cat. No. 7150) just before
carrageenan injection and hourly afterwards for up to 7 h.
Swelling was calculated as the difference in volume
between carrageenan and vehicle-treated paws. Anti-in-
flammatory activity was expressed as percentage inhibition
of oedema compared with control group, and calculated by
the formula I(%)51003(12dt/dc), where dt is the
difference in paw volume in the drug-treated group and dc
is the difference in paw volume in the control group.
Lower and/or higher doses were administered in order to
study dose-dependence of the anti-inflammatory activity
and to calculate ED50 values.
Fig. 1. Structures of compounds 6a–f, piroxicam 7 and meloxicam 8.
which has an improved gastrointestinal tolerance compared
with its analogue piroxicam 7 (Dequeker et al., 1998) (Fig.
1).
2. Materials and methods
2.1. Chemistry
¨
Melting points (mp) were determined on a Buchi
2.2.3. Ulcerogenic study
capillary apparatus and were uncorrected. pH of aqueous
solutions were measured with a pH meter (Hanna, model
HI 8820). Nuclear magnetic resonance (NMR) spectra
were recorded on a Bruker 300 spectrometer (1H at 300
MHz and 13C at 75 MHz) with deuterochloroform
(CDCl3) as solvent and tetramethylsilane (TMS) as inter-
nal standard reference. Infra-red (IR) spectra were re-
corded on a Bio-Rad FTS-6000 spectrometer. Elemental
analyses were carried out by the Microanalytical Labora-
tory, Claude Bernard University, Lyon I (France). For
column chromatography Merck silica gel (70–230 mesh)
was used. Solvents used in reactions were dried and
distilled before use. The purity of all synthesized com-
pounds was controlled by thin layer chromatography
(TLC; Merck silica gel plates 60F-254) and gas chroma-
tography (HP-5890) coupled to a mass spectrometer (HP-
5792).
The subchronic ulcerogenic or gastric mucosal eroding
effect of piroxicam and compounds 6a–f was determined
using the method reported by Kulkarni (1993). Rats were
chronically treated with piroxicam or compounds 6a–e at
one dose level (5 and 10 mg/kg, p.o., respectively) and
compound 6f at three dose levels (5, 10 and 20 mg/kg,
p.o.) for 7 days. On day 7, the rats were fasted for 24 h but
had free access to water. On day 8 the rats were anaesthet-
ized with ether and killed by decapitation. To determine
the mean ulcer score, the stomach was opened along the
greater curvature, washed under running tap water, and
examined with a dissecting microscope (310 magnifica-
tion) for the presence of ulcerations in mucosa. Ulcer index
(UI) was calculated by the formula UI5[(ADU)2(%
RU)]/100, where ADU is the average degree of single
ulceration of the group, and % RU is the percentage of rats
with ulcers (05normal coloured stomach; 0.55red colou-
ration; 15spot ulcers; 1.55haemorrhagic streaks; 25.3
but ,5 ulcers; 35$5 ulcers).
2.2. Biology
2.2.1. Animals
2.2.4. Analysis of gastric contents
All experiments were performed according to the Guide-
lines for Animal Experimentation of Monastir University.
Wistar rats (150–200 g) and albino mice (20–25 g) of
both sexes purchased from Pasteur Institute (Tunis,
Tunisia) were used. They were housed in groups of eight
or ten animals in plastic cages at 20–25 8C and maintained
on a standard pellet diet with free access to water. Animals
were fasted for 24 h before the experiments.
Each stomach was ligated at both ends and the gastric
content was collected in a graduated centrifuge tube and
centrifuged at 2000 rpm over a period of 10 min. The pH
of filtered gastric content was recorded. The free acidity
and the total acid production were determined as follows.
A 1-ml sample of filtered gastric content was poured into a
small beaker, then two to eight drops of Toepfer’s reagent
(indicator) were added. The solution was titrated to pH 3.5
with aqueous sodium hydroxide (0.01 M). The volume of
added alkali was recorded until the solution colour
changed from red to yellowish-orange. This volume mea-
sured the free acidity. After that, two to three drops of
phenolphthalein were added to the solution, and titration to
pH 8.0 was continued until a definite red tinge reappeared.
2.2.2. Anti-inflammatory activity
Anti-inflammatory activity was determined in vivo using
the carrageenan-induced rat paw oedema test (Winter et
al., 1962). Fasted albino rats of both sexes (150–200 g)
were used (n58). Compounds 6a–f (5–20 mg/kg) and

Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
223
This volume measured combined acidity. The sum of free
and combined acidity volumes measured the total acidity.
2.2.5. Toxicity study
All tested compounds were investigated for their acute
toxicity and approximate lethal dose (ALD50). Albino mice
weighing 20–25 g were randomised in ten groups of ten
animals. Tested substances were dissolved in a vehicle
solution consisting of Tween 80/absolute ethanol/saline
solution (0.9%) in the ratio 1:1:18 and administered (by
oral gavage and intraperitoneally) in graduated doses (25–
50–100–200–400–600–800 and 1000 mg/kg), one dose
being used per group. No mortality was observed in the
control group of animals treated only with the vehicle
solution. ALD50 values were determined by observing
mortality within 24 h after drug administration.
2.3. Statistics
Results are expressed as the mean6S.E.M. of eight or
ten animals per group. The data were analysed using
Student’s t-test, and P,0.05 was considered significant.
3. Results and discussion
3.1. Synthesis and spectral data of compounds 1–5 and
6a–f
Fig. 2. Synthetic procedure of compounds 6a–f.
The synthetic routes to target compounds 6a–f are
outlined in Fig. 2. The commercially available (S)-2-
amino-3-methylbutanoic acid (L-valine), used as a starting
material, was converted to (S)-2-N-tosylamino-3-
methylbutanoic acid 1 according to the procedure of Craig
furnished the desired 6-methyl-3-isopropyl-2H-1,2-ben-
zothiazin-4(3H)-one 1,1-dioxide 6a in 51% yield as a
yellow solid. The same procedure gave a crystalline 6-
methyl-3-isopropyl-N-methyl-1, 2-benzothiazin-4(3H) - one
1,1-dioxide 6b from 4 in 69% yield.
and
Berry
(1992).
(S)-3-Methyl-2-(N-methyl-N-
It is interesting to note that for efficient directed ortho
metalation reactions and to improve yields of products 6a
and 6b, an excess of LDA was required. However, this
excess involves racemisation of the final adducts, deduced
by the measure of their optical rotations. Analysis of the
nuclear magnetic resonance and infra-red spectra indicated
that these compounds exist in the keto (non-enolized)
form.
Hydroxylation of 6b was carried out using the Davis
procedure (Davis et al., 1991). The lithium enolate of 6b
was first prepared in situ upon treatment with LDA at
278 8C in THF. Then it was treated with (6)-camphorsul-
fonyloxaziridine 5 to afford the racemate 6c in 86% yield
as a white solid. Note that (6)-camphorsulfonyloxaziridine
tosylamino)butanoic acid 2 was prepared in two steps from
1 following a similar method reported by Freindinger et al.
(1983). The first step involves p-toluenesulfonic acid
(PTSA)-catalyzed condensation of 1 with paraformal-
dehyde (CH2O)n in refluxing benzene to form (S)-4-iso-
propyl-3-tosyl-1,3-oxazolidin-5-one. This intermediate was
then reduced with triethylsilane (Et3SiH) and trifluoro-
acetic acid (CF3COOH) to give the product 2 in an overall
yield of 92%. Treatment of 1 and 2 with thionyl chloride
(SOCl2) furnished the corresponding acid chloride inter-
mediates which subsequently were transformed to the
corresponding N,N-diisopropylamides 3 and 4 upon treat-
ment with diisopropylamine [(i-Pr)2NH] in the presence of
triethylamine (NEt3).
¨
was synthesized according to our procedure (Kraıem et al.,
To synthesize the heterocycles 6a and 6b, we followed
the procedure described by Bakker et al. (1997) with a
slight modification, involving a tosylsulfonyl ortho de-
protonation based on the well-established directed ortho
metalation reaction (Snieckus, 1990). Indeed, reaction of
compound 3 with four equivalents of lithium diiso-
propylamide (LDA) in tetrahydrofuran (THF) at 25 8C
2001). Attempts at asymmetric oxidation of the lithium
enolate of 6b with enantiomerically pure N-sulfonylox-
aziridines and camphorsulfonyloxaziridines (Davis and
Chen, 1992) were also investigated but since the maximum
optical purity obtained was still unsatisfactory (ee546%),
the pharmacological study at present will be in preliminary
evaluation restricted to the racemic derivatives.

224
Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
Finally the synthesis of the alcohol derivatives 6d, 6e
J58.0 Hz), 7.70 (d, 1H, J58.0 Hz); RMN 13C (CDCl3, d
ppm) 11.03, 12.24, 12.89, 13.52, 16.45, 19.11, 21.33,
31.42, 40.61, 41.29, 57.59, 127.39, 129.16, 136.12, 143.45,
169.17.
and 6f was completed by sodium borohydride (NaBH4)
reduction of the corresponding ketones 6a, 6b and 6c,
respectively.
3.1.4. (S)-N,N-Diisopropyl-3-methyl-2-(N-methyl-N-
tosylamino)butanamide (4)
3.1.1. (S)-2-(N-Tosylamino)butanoic acid (1)
(S)-2-(N-tosylamino)butanoic acid was prepared from
L-valine following the reported procedure (Craig and
Berry, 1992). Recrystallization from ethanol-diethyl ether
afforded pure 1 in 98% yield as white crystals, mp 146–
148 8C.
Compound 4 was prepared according to the procedure
used above. Yield 85%; mp 135–137 8C; RMN 1H
(CDCl3, d ppm) 0.81–1.02 (m, 6H), 1.09–1.22 (m, 12H),
1.63–1.74 (m, 1H), 2.39 (s, 3H), 2.91 (s, 3H), 3.05–3.25
(m, 2H), 3.98 (d, 1H, J510.5 Hz), 7.27 (d, 1H, J58.0
Hz), 7.73 (d, 1H, J58.0 Hz); RMN 13C (CDCl3, d ppm)
11.10, 12.23, 12.78, 13.49, 16.49, 19.25, 21.47, 29.56,
31.52, 40.51, 41.27, 57.31, 127.26, 128.94, 136.24, 143.36,
169.46.
3.1.2. (S)-3-Methyl-2-(N-methyl-N-tosylamino)butanoic
acid (2)
To a stirred solution of paraformaldehyde (2 g) and
p-toluenesulfonic acid (200 mg) in benzene (150 ml), were
added N-tosyl-L-valine (10 mmol). The mixture was
refluxed for 1 h with azeotropic water removal. The
solution was cooled, washed with 1 M aqueous NaHCO3
(2325 ml) and dried over magnesium sulfate (MgSO4).
Concentration in vacuo gave the intermediate (S)-4-iso-
propyl-3-tosyl-1,3-oxazolidin-5-one, which was dissolved
without further purification in chloroform (35 ml). Tri-
fluoroacetic acid (35 ml) and triethylsilane (30 mmol)
were then added. The solution was stirred at room tem-
perature for 20 h, followed by concentration in vacuo to an
oil which was dissolved in chloroform and reconcentrated
three times. Recrystallization from ethyl acetate–petroleum
ether afforded 9.2 mmol of 2 (92% yield) as colourless
3.1.5. (6)-Camphorsulfonyloxaziridine (5)
(6)-Camphorsulfonyloxaziridine was prepared using
trichloroacetonitrile (CCl3CN) instead of benzonitrile
(PhCN) as follows. To a stirred solution of trichloro-
acetonitrile (30 mmol), (6)-camphorsulfonylimine (15
mmol) and tetrabutylammonium chloride (0.2 mmol) in 50
ml of dichloromethane were added dropwise 5 ml of 30%
hydrogen peroxide H2O2 (45 mmol). The H2O2 solution
was adjusted to pH 7 by addition of 2.5 g of K2HPO4,
prior to addition. The biphasic mixture was stirred at 0 8C.
After 1 h the mixture was washed with water (250 ml) and
brine (10 ml), and dried over MgSO4. The solvent was
removed and the residue obtained was purified by column
chromatography (SiO2; ethyl acetate–hexane, 5:95) to give
compound 5 in 89% yield. mp 163–165 8C; RMN 1H
1
crystals. mp 155–157 8C; RMN H (CDCl3, d ppm) 0.93
(d, 3H, J56.4 Hz), 1.02 (d, 3H, J56.4 Hz), 2.38 (m, 1H),
2.12 (m, 1H), 2.43 (s, 3H), 2.88 (s, 1H), 4.13 (d, 1H,
J510.7 Hz), 7.29 (d, 1H, J58.2 Hz), 7.69 (d, 1H, J58.2
Hz); RMN 13C (CDCl3, d ppm) 19.07, 21.40, 27.73, 29.83,
51.13, 64.50, 127.25, 129.11, 135.84, 143.12, 180.53.
(CDCl3, d ppm) 1.08 (s, 3H), 1.21 (s, 3H), 1.46–2.20 (m,
6H), 2.63 (m, 1H), 3.25 (d, 2H, J514.1 Hz); RMN 13C
(CDCl3, d ppm) 19.51, 20.49, 26.35, 28.40, 33.71, 45.86,
48.33, 48.42, 55.02, 99.24; IR (n cm21) 1156, 1338 (SO2),
2963 (CH); EI-MS, m/z (rel. abund.) 229 (M1, 6.7), 122
(70.3), 93 (55.9), 67 (66.2), 41 (100).
3.1.3. (S)-N,N-Diisopropyl-3-methyl-2-(N-
tosylamino)butanamide (3)
3.2. General procedure for synthesis of ketones 6a and
6b
Thionyl chloride (14.3 mmol) was added dropwise at
25 8C to a solution of (S)-2-(N-tosylamino)butanoic acid 1
(13 mmol) in benzene (120 ml). The mixture was heated at
reflux for 3 h and concentrated in vacuo to give a yellow
oil, to which was added dichloromethane (100 ml),
diisopropylamine (13 mmol) and triethylamine (14.3
mmol). The resulting solution was stirred at 25 8C for 24 h,
the reaction mixture was mixed with distilled water (30
ml) and extracted with dichloromethane (3325 ml). The
organic extracts were combined, dried over MgSO4,
concentrated in vacuo up to a small volume and purified by
column chromatography (SiO2; ethyl acetate–hexane,
20:80). By means of this procedure compound 3 was
To a stirred solution of lithium diisopropylamide (LDA)
20 mmol, freshly prepared in anhydrous tetrahydrofuran
(THF) at 25 8C, a solution of the N,N-diisopropylamino
amide (5 mmol) was added. The reaction mixture was
allowed to warm to room temperature over 1 h, and
quenched with hydrochloric acid (1 M, 25 ml). The
aqueous layer was extracted with dichloromethane (3320
ml), and the organic extracts were dried over MgSO4 and
evaporated. The residue obtained was purified by column
chromatography (SiO2, ethyl acetate–hexane).
1
obtained in 76% yield. mp 131–132 8C; RMN H (CDCl3,
3.2.1. (6)-6-Methyl-3-isopropyl-2H-1,2-benzothiazin-
4(3H)-one 1,1-dioxide (6a)
d ppm) 0.83–0.99 (m, 6H), 1.02–1.18 (m, 12H), 1.79–1.88
(m, 1H), 2.42 (s, 3H), 3.02–3.21 (m, 2H), 3.83 (dd, 1H,
J54.0 and 9.2 Hz), 5.73 (d, 1H, J59.2 Hz), 7.28 (d, 1H,
1
Yield 51%; mp 75–77 8C; RMN H (CDCl3, d ppm)
0.94 (d, 3H, J57 Hz), 1.16 (d, 3H, J57 Hz), 2.38 (m,

Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
225
1H), 2.49 (s, 3H), 4.45 (dd, 1H, J510.6, 4 Hz), 5.17 (d,
1H, J510.6 Hz), 7.56 (d, 1H, J57.6 Hz), 7.74 (d, 1H,
J57.6 Hz), 7.88 (s, 1H); RMN 13C (CDCl3, d ppm) 16.49,
19.03, 21.51, 27.96, 68.07, 123.36, 128.80, 129.54, 134.99,
139.20, 143.78, 191.71; IR (n cm21) 1679 (C=O), 1128,
1323 (SO2N), 1552 (NH), 1044 (SO2); Elemental analysis
(C12H15NO3S): Calculated (%) C 56.91, H 5.92, N 5.53;
Found (%) C 56.51, H 6.02, N 5.33.
(30 ml) at 0 8C for 30 min. The mixture was stirred at the
same temperature for 2 h. After removal of the solvent, the
residue was diluted with water (20 ml), acidified with
diluted hydrochloric acid (0.5 M) to pH 6, and extracted
with dichloromethane (3330 ml). The combined extracts
were dried over MgSO4 and concentrated to give a residue
which was purified by flash chromatography (SiO2,
CH2Cl2 /MeOH).
3.2.2. (6)-2,6-Dimethyl-3-isopropyl-1,2-benzothiazin-
4(3H)-one 1,1-dioxide (6b)
3.4.1. (6)-6-Methyl-3-isopropyl-2H-1,2-benzothiazin-
4(3H)-ol 1,1-dioxide (6d)
1
Crystallisation from chloroform–hexane afforded 6b in
69% yield; mp 121–123 8C; RMN 1H (CDCl3, d ppm)
0.97 (d, 3H, J57 Hz), 1.19 (d, 3H, J57 Hz), 2.40 (m,
1H), 2.51 (s, 3H), 282 (s, 3H), 4.46 (d, 1H, J53.5 Hz),
7.31 (d, 1H, J57.8 Hz), 7.39 (s, 1H), 7.68 (d, 1H, J57.8
Hz); RMN 13C (CDCl3, d ppm) 17.01, 19.21, 21.51, 28.05,
39.02 67.25, 127.08, 127.23, 129.79, 129.91, 135.96,
143.62, 201.28; IR (n cm21) 1674 (C=O), 1139, 1339
(SO2N), 2821 (NCH3), 1040 (SO2); Elemental analysis
(C13H17NO3S): Calculated (%) C 56.91, H 5.92, N 5.53;
Found (%) C 56.51, H 6.02, N 5.33.
Yield 91%; mp 89–91 8C; RMN H (CDCl3, d ppm)
1.08 (d, 3H, J57 Hz), 1.23 (d, 3H, J57 Hz), 2.36 (m,
1H), 2.51 (s, 3H), 4.39 (dd, 1H, J510.6, 4 Hz), 5.28 (d,
1H, J510.6 Hz), 5.89 (br, 1H), 7.86 (d, 1H, J57.6 Hz),
7.91 (d, 1H, J57.6 Hz), 7.97 (s, 1H); RMN 13C (CDCl3, d
ppm) 17.45, 20.13, 22.58, 27.86, 65.27, 72.56, 122.26,
127.60, 128.59, 133.90, 138.29, 142.68; IR (n cm21) 1138,
1321 (SO2N), 1542 (NH), 1048 (SO2), 3614 (OH);
Elemental analysis (C12H16NO3S): Calculated (%) C
56.69, H 6.29, N 5.51; Found (%) C 56.81, H 6.12, N
5.58.
3.3. Synthesis of ketone 6c
3.4.2. (6)-2,6-Dimethyl-3-isopropyl-1,2-benzothiazin-
4(3H)-ol 1,1-dioxide (6e)
1
A solution of 6b (10 mmol) in 20 ml of THF was added
dropwise to a stirred solution of LDA (11 mmol) in 10 ml
of THF cooled at 278 8C. After the mixture was stirred at
the same temperature for 30 min, 16 mmol of (6)-
camphorsulfonyloxaziridine 5 in 10 ml of THF was added
dropwise. The reaction was monitored by thin layer
chromatography (TLC) and quenched by the addition of 15
ml of distilled water. The reaction mixture was extracted
with chloroform (3350 ml) and the combined organic
extracts were dried over MgSO4. The solvent was removed
to give a solid which was purified by flash chromatography
(SiO2; ethyl acetate–hexane, 2:8) giving 8.6 mmol of 6c
(86% yield).
Yield 96%; mp 97–99 8C; RMN H (CDCl3, d ppm)
0.99 (d, 3H, J57 Hz), 1.24 (d, 3H, J57 Hz), 2.44 (m,
1H), 2.63 (s, 3H), 280 (s, 3H), 4.51 (d, 1H, J53.5 Hz),
5.77 (br, 1H) 7.38 (d, 1H, J57.8 Hz), 7.42 (s, 1H), 7.71
(d, 1H, J57.8 Hz); RMN 13C (CDCl3, d ppm) 19.12,
21.08, 23.37, 29.15, 39.42, 66.26, 76.05, 128.18, 128.85,
130.29, 130.98, 134.86, 144.52; IR (n cm21) 1129, 1337
(SO2N), 2835 (NCH3), 1048 (SO2), 3589 (OH); Elemen-
tal analysis (C13H18NO3S): Calculated (%) C 58.20, H
6.71, N 5.22; Found (%) C 58.32, H 6.66, N 5.31.
3.4.3. (6)-2,6-Dimethyl-3-isopropyl-3-hydroxy-1,2-
benzothiazin-3,4-diol 1,1-dioxide (6f)
1
Yield 94%; mp 112–114 8C; RMN H (CDCl3, d ppm)
3.3.1. (6)-2,6-Dimethyl-3-isopropyl-3-hydroxy-1,2-
benzothiazin-4-one 1,1-dioxide (6c)
1.21 (d, 6H, J57 Hz), 2.39 (s, 3H), 2.73 (s, 3H), 3.26 (m,
1H), 5.49–5.59 (br, 1H), 6.58–6.62 (br, 1H), 7.35 (d, 1H,
J57.8 Hz), 7.43 (s, 1H), 7.87 (d, 1H, J57.8 Hz); RMN
13C (CDCl3, d ppm) 21.76, 20.67, 23.52, 28.89, 60.25,
72.49, 77.26, 123.75, 124.57, 129.48, 139.44, 142.71,
144.04; IR (n cm21) 1133, 1320 (SO2N), 2843 (NCH3),
1
Yield 86%; mp 134–136 8C; RMN H (CDCl3, d ppm)
1.28 (d, 6H, J57 Hz), 2.46 (s, 3H), 2.80 (s, 3H), 3.15 (m,
1H), 5.51–5.62 (br, 1H), 7.31 (d, 1H, J57.8 Hz), 7.39 (s,
1H), 7.68 (d, 1H, J57.8 Hz); RMN 13C (CDCl3, d ppm)
20.81, 21.87, 23.31, 28.67, 59.72, 73.21, 123.75, 124.57,
129.48, 139.44, 142.71, 144.04, 192.73; IR (n cm21) 1677
(C=O), 1132, 1315 (SO2N), 2833 (NCH3), 1052 (SO2),
3610 (OH); Elemental analysis (C13H17NO4S): Calculated
(%) C 55.12, H 6.00, N 4.94; Found (%) C 55.31, H 6.09,
N 4.86.
1055
(SO2),
3649
(OH);
Elemental
analysis
(C13H18NO4S): Calculated (%) C 54.92, H 6.33, N 4.92;
Found (%) C 54.86, H 6.42, N 4.89.
3.5. Anti-inflammatory and ulcerogenic effects of
compounds 6a–f
3.4. General procedure for reduction of ketones 6a–c
The
3-carboxamide-4-hydroxy-2-methyl-1,2-benzo-
thiazines 1,1-dioxide referred to as oxicams are a well-
known class of NSAIDs with several products on the
market (Wiseman et al., 1976; Binder et al., 1987;
Sodium borohydride (6 mmol) was added portion-wise
to a suspension of ketone 6a–c (5 mmol) in dry methanol

226
Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
Table 1
Anti-inflammatory effects of compounds 6a–c (the ketone series) against carrageenan-induced paw oedema in ratsa
c
Treatment
Dose
(mg/kg, i.p.)
Volume variation
(ml)
Inhibition (%)b
at 3 h
ED50 (mg/kg, i.p.)
(95% confidence limits)
Controld
Piroxicam
–
5
5
10
20
5
10
20
5
948.8635.8
310.6619.1
685.9630.5
615.6622.8
609.9620.4
503.5619.2
470.2623.5
454.1618.6
476.9619.5
427.5618.9
424.6621.3
–
–
67.3*
27.8*
35.2*
35.8*
47.0*
50.5*
52.2*
49.8*
55.0*
55.3*
4.4 (2.7–7.3)
6a
6a
6a
6b
6b
6b
6c
6c
6c
29.8 (14.9–63.5)
9.2 (4.4–21.2)
5.3 (2.7–11)
10
20
*P,0.05 (Student’s t-test).
a Each experimental group included eight rats (150–200 g).
b Percentage inhibition compared with control group and expressed by mean volume variation6S.E.M. 3 h after carrageenan injection.
c ED50 was calculated by the method of Litchfield and Wilcoxon (1949).
d Control group received equivalent volume of vehicle solution.
Engelhardt et al., 1995). The structure–activity relation-
ships (SAR) for these compounds have been extensively
explored for optimization of anti-inflammatory activity
during the three last decades since this class was intro-
duced (Lombardino et al., 1972; Farre et al., 1986; Berg et
al., 1999; Lee et al., 1999). In continual efforts to find
potentially safer and more efficacious parent agents
through further exploration of SAR of this class, we
decided to study the pharmacological profiles of com-
pounds 6a–f, analogues of the oxicam family. We ex-
amined the effect of modification of both the steric and
electronic nature of substituents on various portions of
enol-carboxamide-type NSAIDs. For this objective the
carboxamide moiety (position 3) is replaced by alkyl
and/or hydroxyl group, the enol form (position 4) is
replaced by a ketone or alcohol function, and the hydrogen
atom (position 6) is replaced by a methyl group.
and ulcerogenic effect (Table 3) of products 6a–f shows
that modifications of the structure of the oxicam basic
skeleton results on the one hand in a comparable anti-
inflammatory potential activity, and on the other hand in
significantly less irritation to the gastric mucosa compared
to piroxicam.
Table 1 reveals the following. (a) The ketones 6a, 6b
and 6c dose dependently exerted an anti-inflammatory
effect in carrageenan-induced paw oedema in rats. (b) The
influence of the substituent R9 (in the sulfonamide nitro-
gen) on activity is remarkable. Compound 6a is less active
than the N-methyl derivatives 6b and 6c, so a methyl
group linked to the sulfonamide nitrogen increases the
activity compared to the case of a hydrogen atom. At the
same dose, compound 6a has 27.8% anti-inflammatory
activity against 47.0 and 49.8% for 6b and 6c, respective-
ly. (c) Compound 6c was slightly more active than 6b
according to the ED50 values (5.3 vs. 9.2 mg/kg, i.p.).
Study of the anti-inflammatory activity (Tables 1 and 2)
Table 2
Anti-inflammatory effects of compounds 6d–f (the alcohol series) against carrageenan-induced paw oedema in ratsa
c
Treatment
Dose
(mg/kg, i.p.)
Volume variation
(ml)
Inhibition (%)b
at 3 h
ED50 (mg/kg, i.p.)
(95% confidence limits)
Controld
Piroxicam
–
5
5
10
20
5
10
20
5
989.8637.6
328.6619.7
669.2625.9
611.8625.1
588.0623.4
450.4621.6
398.9624.0
401.9619.2
379.1620.5
354.4621.4
347.4618.3
–
–
66.8*
32.4*
38.2*
40.6*
54.5*
59.7*
59.4*
61.7*
64.2*
64.9*
4.4 (2.7–7.3)
6d
6d
6d
6e
6e
6e
6f
23.4 (12.5–48.7)
4.7 (2.3–9.2)
4.5 (2.5–9.1)
6f
6f
10
20
*P,0.05 (Student’s t-test).
a Each experimental group included eight rats (150–200 g).
b Percentage inhibition compared with control group and expressed by mean volume variation6S.E.M. 3 h after carrageenan injection.
c ED50 was calculated by the method of Litchfield and Wilcoxon (1949).
d Control group received equivalent volume of vehicle solution.

Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
227
pH
Table 3
Ulcerogenic effects of compounds 6a–f and piroxicam in rats after subchronic treatment for 7 days
Treatment
Dose
Ulcer index
(UI)
Free acidity
(meq/l)
Total acidity
(meq/l)
(mg/kg, p.o.)
Control
Piroxicam
–
5
0
31.061.1
86.960.9
38.261.0
39.160.8
41.661.2
38.560.9
44.360.9
39.361.3
40.061.1
42.761.3
40.662.7
112.564.6
45.961.8
46.261.6
48.861.9
46.762.0
52.962.2
48.162.3
49.362.1
50.862.3
4.9
2.6
3.4
3.3
3.1
3.5
3.0
3.6
3.3
3.2
6.2*
1.4*
1.2*
1.4*
1.1*
1.3*
0.9*†
1.3*
1.3*
6a
6b
6c
6d
6e
6f
10
10
10
10
10
5
6f
6f
10
20
Values are mean6S.E.M., n510.
*P,0.05 compared with control.
†P,0.05 compared with the same dose of piroxicam.
This difference in potency results from the presence of the
polar hydroxyl functionality in position three of the ring.
This group increases the water solubility, and at the same
time, provides one more binding site in the molecule.
As Table 2 illustrates: (a) compound 6e is |1.5 times
more potent than 6d in the dose–response comparison in
the rat paw oedema test. The increase observed in the
anti-inflammatory activity may be attributed to the change
in the lipophilicity of these compounds, which in turn
improves the pharmacokinetic parameters of the molecule;
(b) compounds 6e and 6f displayed anti-inflammatory
activity almost equal to that of piroxicam (ED50 54.7 and
4.5 vs. 4.4 mg/kg, i.p., respectively).
The most active substance in this series is 6-methyl-3-
isopropyl-3-hydroxy-N-methyl-1,2-benzothiazin-3,4-diol
1,1-dioxide 6f with approximately the same activity as
piroxicam.
Mucosal erosion and ulceration are produced by most
NSAIDs with varying degrees. Inhibition of synthesis of
gastroprotective prostaglandins (PGE2 and PGI2) is clearly
involved (Nezamin and Philips, 1967; Main and Whittle,
1973; Cryer and Feldman, 1992). Thus deficiency of PGs
reduces mucous and hydrogenocarbonate (HCO23 ) secre-
tion, tends to enhance acid secretion and may promote
mucosal ischaemia.
In our study, the gastric tolerability based on ulcer
index, pH and acidity of gastric content of products 6a–f is
compared with piroxicam after subchronic treatment for 7
days. Piroxicam, which has a peculiar acidic character (pKa
values 1.86 and 5.46) (Von Bruchhausen et al., 1993)
exhibited high gastric content acidity which enhances
aggressive factors and curtails defensive factors in gastric
mucosa, and is therefore ulcerogenic (UI56.2 at 5 mg/kg,
p.o.). Compounds 6a–f at a daily dose of 10 mg/kg clearly
and significantly prevented gastric damage (UI range
between 0.9 and 1.4). Moreover, ulcer index and acidity
were significantly lower in the 6f treated group, compared
with the piroxicam treated group at the same dose level of
5 mg/kg, p.o. (UI50.9 vs. 6.2), whereas the anti-in-
flammatory efficacy of both agents was shown to be
equivalent. These data indicated that an isopropyl group
(bulkier and non-acidic substituent) at position three of the
1,2-benzothiazine framework, instead of an acidic 2-
pyridylcarboxamide moiety, preserved the anti-inflamma-
tory activity and protected the gastric mucosa from injury.
In this respect, our results shed new light on the chemical
requisite to design new agents with reduced ulcerogenic
effects in gastric mucosa.
In consideration of the efficacy of compounds 6a–f in
anti-inflammatory and ulcerogenic assays, those com-
pounds were further studied at graded doses for their acute
toxicity (ALD50) in albino mice. All them were essentially
non-toxic at the highest dose tested (ALD50 .800 mg/kg,
p.o. and ALD50 .700 mg/kg, i.p.). At these dose levels,
death generally occurred 12–24 h post drug treatment in
40–60% of the animals, whereas the surviving mice
appeared to be normal throughout a 7-day observation
period.
In conclusion, we have synthesized a new series of
6-methyl-3-isopropyl-2H-1,2-benzothiazin-4(3H)-one 1,1-
dioxide derivatives related to oxicam drugs. The anti-
inflammatory activity of tested compounds administered
intraperitoneally in their racemic forms seems to be
significantly influenced by the nature of substituents at the
two and the three positions of the 1,2-benzothiazine
framework. Indeed, according to the ED50 values, com-
pounds 6b and 6c compared to 6a (9.2 and 5.3 vs. 29.8
mg/kg, i.p.) prove that N-methylation and the introduction
of a hydroxyl group increase the anti-inflammatory activi-
ty, probably due to increased lipophilicity. The most active
compound in this series with an ED50 value of 4.6 mg/kg,
i.p., is 6f with N-methyl moiety and two hydroxyl groups.
The same result is found with the exchange of a carbonyl
(6a) by an hydroxyl group (6d) (ED50 529.8 vs. 23.4
mg/kg, i.p.). Our results on the ulcerogenic effects of
compounds 6a–f compared with piroxicam indicate that
replacement of a carboxamide residue by an isopropyl
group remarkably reduces the gastrointestinal adverse

228
Y. Kacem et al. / European Journal of Pharmaceutical Sciences 16 (2002) 221–228
Farre, A.J., Colombo, M., Fort, M., Gutierrez, B., Rodriguez, L., Roser,
R., 1986. Pharmacological properties of droxicam, a new non-steroidal
anti-inflammatory agent. Methods Find. Exp. Clin. Pharmacol. 8,
407–422.
Freindinger, R.M., Hinkle, J.S., Perlow, D.S., Arison, B.H., 1983.
Synthesis of 9-fluorenylmethyloxycarbonyl-protected N-alkyl amino
acids by reduction of oxazolidinones. J. Org. Chem. 48, 77–81.
effects. Further studies either in vivo or in vitro are under
investigation to elucidate the exact mechanism of action of
tested products and to evaluate the pharmacological profile
for each optical isomer of compounds 6a–f.
¨
Kraıem, J., Kacem, Y., Khiari, J., Ben Hassine, B., 2001. Stereospecific
Acknowledgements
synthesis of 2-alkyl-3-aryloxaziridines from prochiral and chiral N-
arylidenealkylamines by use of the benzonitrile–hydrogen peroxide
system. Synth. Commun. 31, 263–271.
´ ´
We are grateful to ‘la Direction Generale de la Re-
`
cherche Scientifique et Technique’ and ‘le Ministere de
Kulkarni, S.K., 1993. In: Handbook of Experimental Pharmacology.
Vallabh Prakashan, Delhi, pp. 69–71.
Lee, E.B., Kwon, S.K., Kim, S.G., 1999. Synthesis and analgesic and
anti-inflammatory activities of 1,2-benzothiazine derivatives. Arch.
Pharm. Res. 22, 44–47.
´
l’Enseignement Superieur Tunisien’ for their support of
our program.
Lipscomb, G.R., Wallis, N., Armstrong, G., Rees, W.D., 1998. Gastroin-
testinal tolerability of meloxicam and piroxicam: a double-blind
placebo-controlled study. Br. J. Clin. Pharmacol. 46, 133–137.
Litchfield, J.T., Wilcoxon, F.A., 1949. A simplified method of evaluating
dose–effect experiments. J. Pharmacol. Exp. Ther. 96, 99–113.
Lombardino, J.G., Wiseman, E.H., Chiaini, J., 1972. Potent anti-in-
flammatory N-heterocyclic 3-carboxamides of 4-hydroxy-2-methyl-
2H-1,2-benzothiazine 1,1-dioxide. J. Med. Chem. 16, 493–496.
Main, I.H.M., Whittle, B.J.R., 1973. The effect of E and A prostaglandin
on gastric mucosal blood flow and acid secretion in the rat. Br. J.
Pharmacol. 49, 428.
Moote, C., 1992. Efficacy of non-steroidal anti-inflammatory drugs in the
management of postoperative pain. Drugs 44 (Suppl. 5), 14–30.
Nezamin, R.J.E., Philips, J.P., 1967. Inhibition of gastric secretion by
prostaglandins. Am. J. Digest. Dis. 12, 1073–1076.
Scott, L.J., Lamb, H.M., 1999. Rofecoxib. Drugs 58, 499–505.
Simon, L.S. et al., 1998. Preliminary study of the safety and efficacy of
SC-58635, a novel cyclooxygenase-2 inhibitor. Arthritis Rheum. 41,
1591–1601.
Snieckus, V., 1990. Directed ortho metalation. Tertiary amide and O-
carbamate directors in synthetic strategies for polysubstituted aro-
matics. Chem. Rev. 90, 879–933.
Vane, J.R., Botting, R.M., 1995. New insights into the mode of action of
anti-inflammatory drugs. Inflamm. Res. 44, 1–10.
Vane, J.R., Botting, R.M., 1996. Mechanism of action of anti-inflamma-
tory drugs. Scand. J. Rheumatol. 25, 9–25.
Vane, J.R., Botting, R.M., Bakhle, Y.S., 1998. Cyclooxygenases 1 and 2.
Annu. Rev. Pharmacol. Toxicol. 38, 97–120.
References
Bakker, W.I.I., Familoni, O.B., Padfield, J., Snieckus, V., 1997. In-
tramolecular anionic Friedel-Crafts equivalents. An expeditious syn-
thesis of 4H-1,2-benzothiazin-4-one 1,1-dioxides from N-arylsul-
fonylated amino acids. Synlett. 1079–1080.
Berg, J., Fellier, H., Christoph, T., Grarup, J., Stimmeder, D., 1999. The
analgesic NSAID lornoxicam inhibits cyclooxygenase (COX)-1/-2,
inductible nitric oxide synthase (iNOS), and the formation of inter-
leukin (IL)-6 in vitro. Inflamm. Res. 48, 369–379.
Binder, D., Hromatka, O. et al., 1987. Analogues and derivatives of
tenoxicam. Synthesis and anti-inflammatory activity of analogues with
different residues on the ring nitrogen and the amide nitrogen. J. Med.
Chem. 30, 678–682.
Craig, D., Berry, M.B., 1992. A convenient method for the preparation of
enantiomerically pure 2-substituted N-tosylaziridines. Synlett. 41–44.
Cryer, B., Feldman, M., 1992. Effects of non-steroidal anti-inflammatory
drugs on endogenous GI prostaglandins and therapeutic strategies for
prevention and treatment of non-steroidal anti-inflammatory drug-
induced damage. Arch. Intern. Med. 152, 115.
Davis, F.A., Chen, B.-C., 1992. Asymmetric hydroxylation of enolates
with N-sulfonyloxaziridines. 92, 919–934.
Davis, F.A., Kumar, A., Chen, B.-C., 1991. Chemistry of oxaziridines 16.
A short, highly enantioselective synthesis of the AB-ring segments of
g-rhodomycinone and a-citromycinone using (1)-[(8,8-dimethox-
ycamphoryl)sulfonyl]oxaziridine. J. Org. Chem. 56, 1143–1145.
Dequeker, J., Hawkey, C. et al., 1998. Improvement in gastrointestinal
tolerability of the selective cyclooxygenase (COX)-2 inhibitor, melox-
icam, compared with piroxicam: results of the safety and efficacy
large-scale evaluation of COX-inhibiting therapies (SELECT) trial in
osteoarthritis. Br. J. Rheumatol. 37, 946–951.
Engelhardt, G., Homma, D., Schlegel, K., Utzmann, R., Schnitzler, C.,
1995. Anti-inflammatory, analgesic, antipyretic and related properties
of meloxicam, a new non-steroidal anti-inflammatory agent with
favorable gastrointestinal tolerance. Inflamm. Res. 44, 423–433.
Von Bruchhausen, F., Ebel, S., Frahm, A.W., 1993. Handbuch der
Pharmazeutischen Praxis, Vol. 7. Springer, Heidelberg.
Winter, C.A., Risley, G.A., Nuss, G.W., 1962. Carrageenan-induced
oedema in hind paw of the rat as an assay for anti-inflammatory drugs.
Proc. Soc. Exp. Biol. Med. 3, 544–547.
Wiseman, E.H., Chang, Y.H., Lombardino, J.G., 1976. Piroxicam, a novel
anti-inflammatory agent. Arzneim.-Forsch. 26, 1300–1303.