Synthesis and pharmacological evaluation of new 1,2-dithiolane based antioxidants

Article abstract of DOI:10.1016/S0223-5234(02)01424-1

Molecules containing a dithiolane moiety are widely investigated due to their antioxidant properties. The archetypal representative of this class of compounds is lipoic acid and indeed the lipoic acid-dihydrolipoic acid couple is part of the antioxidant defence system of the cell. In the course of a program aiming to find improved antioxidants effective in vivo, we designed, synthesised and pharmacologically investigated new lipoic acid analogs. The salient feature of these structures is the connection, via a thioamide or a thiocarbamate, of a 1,2-dithiolane moiety bearing a carbon chain and a N-alkyl-substituted morpholine ring. It was expected that the antioxidant and chelating properties of these functional groups combined with the basicity of the morpholine ring will impact on the antioxidant as well as on the partition and solubility characteristics of the compounds. Indeed in vitro and in vivo pharmacological investigation showed that these new molecules and especially those containing a thiocarbamate linker possess superior antioxidant properties compared with α-lipoic acid and to the amide or carbamate linker analogs. In particular, some of these compounds efficiently cross the blood brain barrier (BBB) thus providing efficient protection from lethality in a situation of induced oxidative stress. Moreover the absence of the 1,2-dithiolane moiety does not completely abolish antioxidant effects thus demonstrating that these compounds are distinct new chemical entities and not merely lipoic acid prodrugs. The chemical and pharmacological features of these new antioxidants are presented and discussed in the following paper.

Full text of DOI:10.1016/S0223-5234(02)01424-1

European Journal of Medicinal Chemistry 38 (2003) 1ꢀ11
/
www.elsevier.com/locate/ejmech
Original article
Synthesis and pharmacological evaluation of new 1,2-dithiolane
based antioxidants
Claude Guillonneau a, Yves Charton a, Yves-Michel Ginot b,
Marie-Victoire Fouquier-d’He´rouel b, Marc Bertrand b, Brian Lockhart c,
¨
Pierre Lestage c, Solo Goldstein a,*
a Chemistry Research Division A, institut de recherches Servier, 11, rue des Moulineaux, 92150 Suresnes, France
b Technologie Servier, 25-27, rue Euge`ne Vignat, 45007 Orle´ans, France
c Cerebral Pathology Division, institut de recherches Servier, 125, chemin de Ronde, 78290 Croissy-Sur-Seine, France
Received 9 April 2002; received in revised form 3 October 2002; accepted 3 October 2002
Abstract
Molecules containing a dithiolane moiety are widely investigated due to their antioxidant properties. The archetypal
representative of this class of compounds is lipoic acid and indeed the lipoic acidꢀdihydrolipoic acid couple is part of the
/
antioxidant defence system of the cell. In the course of a program aiming to find improved antioxidants effective in vivo, we
designed, synthesised and pharmacologically investigated new lipoic acid analogs. The salient feature of these structures is the
connection, via a thioamide or a thiocarbamate, of a 1,2-dithiolane moiety bearing a carbon chain and a N-alkyl-substituted
morpholine ring. It was expected that the antioxidant and chelating properties of these functional groups combined with the basicity
of the morpholine ring will impact on the antioxidant as well as on the partition and solubility characteristics of the compounds.
Indeed in vitro and in vivo pharmacological investigation showed that these new molecules and especially those containing a
thiocarbamate linker possess superior antioxidant properties compared with a-lipoic acid and to the amide or carbamate linker
analogs. In particular, some of these compounds efficiently cross the blood brain barrier (BBB) thus providing efficient protection
from lethality in a situation of induced oxidative stress. Moreover the absence of the 1,2-dithiolane moiety does not completely
abolish antioxidant effects thus demonstrating that these compounds are distinct new chemical entities and not merely lipoic acid
prodrugs. The chemical and pharmacological features of these new antioxidants are presented and discussed in the following paper.
´
# 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
Keywords: Antioxidant; Dithiolanes; Oxidative stress
1. Introduction
stress, as well as its role in the aggravation of a plethora
of pathological conditions, are widely documented in
the literature [1].
The deleterious effects of an imbalance between
reactive oxygen species (ROS) production and the
available antioxidant defence capacity, termed oxidative
A variable sensitivity to this excessive oxidative state
is observed between tissues and organs, the brain being
the most vulnerable due to its high oxygen consumption,
high content of unsaturated membrane fatty acids and
low level of antioxidant enzymes or vitamins [2].
In the last decade much effort has been directed
towards finding molecules capable of opposing the
oxidative stress challenge [3]. Thus one of the natural
molecules known to prevent or retard oxidation is a-
Abbreviations: BBB, blood brain barrier; DHLA, dihydrolipoic
acid; DCC, dicyclohexylcarbodiimide; GSH, reduced glutathione;
HOBT, 1-hydroxybenzotriazole; i.p., intraperitoneal; p.o., per os;
i.c.v., intracerebroventricular; i.v., intravenous; LA, a-lipoic acid;
NADH, reduced nicotinamide adenine dinucleotide; ROS, reactive
oxygen species; TBTU, O-(1H-benzotriazol-1-yl)-N,N,N?,N?-
tetramethyl
butylhydroperoxide.
* Correspondence and reprints.
uronium
tetrafluoroborate;
t-BuOOH,
tert-
lipoic acid and, therefore, the a-lipoic acid (LA)ꢀ
/
dihydrolipoic acid (DHLA) redox couple (Fig. 1) has
received considerable attention. Indeed, beneficial ef-
E-mail address: solo.goldstein@fr.netgrs.com (S. Goldstein).
´
0223-5234/02/$ - see front matter # 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S 0 2 2 3 - 5 2 3 4 ( 0 2 ) 0 1 4 2 4 - 1

2
C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ11
/
DHLA at a slower rate is also less efficiently incorpo-
rated into the enzyme complexes and might be fully
available for radical scavenging or other actions which
would explain the observed pharmacological results (i.e.
no difference between the enantiomers) [16ꢀ19].
/
Our research on novel antioxidants focused on
structures which, besides the 1,2-dithiolane moiety,
also contained a thioamide or a thiocarbamate func-
tional groups as well as a morpholine ring (Fig. 1) [20].
It was expected that the metal chelating and antiox-
idant properties of these two functional groups com-
bined with the basicity of the morpholine ring will
positively modulate the antioxidant as well as the
partition and solubility characteristics of the com-
pounds. Moreover based on calculated partition coeffi-
Fig. 1. Structures of dithiolane based antioxidants.
fects were demonstrated in a number of paradigms
implicating oxidative stress such as ischemia-reperfu-
sion, diabetes, cataract-formation, HIV activation, and
neurodegeneration [4].
cients (log P ꢀ
/
2ꢀ4) it was also expected that these
/
compounds will exhibit good BBB crossing capabilities.
The pharmacological profiles have been compared with
those obtained for lipoic acid, for amide and carbamate
linker analogs as well as for an analog lacking the 1,2-
dithiolane moiety (vide infra).
Growing evidence suggests that these compounds,
initially known as cofactors of a-keto-acid dehydrogen-
ase complexes, are part of the antioxidant defence
system of the cell [5]. It is well-known that LA is readily
absorbed from diet and also crosses the blood brain
barrier (BBB) [6,7]; the intracellular reduction by
lipoamide dehydrogenase (which is NADH-dependent)
affords DHLA and since mammalian cells are capable
of releasing it, the effects of LA and DHLA might be
present both intra and extracellularly even if LA alone is
administered extracellularly [8].
2. Chemistry
The synthetic methodologies employed for the pre-
paration of amide and thioamide containing structures
are summarised in Fig. 2. Commercially available a-
lipoic acid (racemic mixture) served as a convenient
starting material. The presence of the morpholino group
in all end-products allowed easy transformation into the
corresponding hydrochlorides, which could be crystal-
lised (see Section 6).
LA and DHLA are capable of quenching a number of
radicals both in a lipid and in an aqueous environment
[9ꢀ11]. However, it also appears that LA and DHLA act
/
as antioxidants not only by direct radical trapping and/
or metal chelation but also by recycling other antiox-
idants (ascorbate, vitamin E) and also by inducing
increased cellular levels of reduced glutathione (GSH).
Indeed DHLA prevents lipid peroxidation by reducing
glutathione, which in turn recycles vitamin E [12,13].
The pro-oxidant effects of DHLA obtained under
certain circumstances raised some concerns but the
results obtained are controversial [13,14].
There is strong evidence that the neuroprotective
effects of LA and DHLA are mediated by antioxidant
and free radical scavenging mechanisms. However, LA
exerts its neuroprotective activity only after a long
period of treatment suggesting that endogenous reduc-
tion to DHLA has to occur in order to diminish
neuronal damage [15]. LA contains an asymmetric
centre but no clear difference between (R) and (S)
enantiomers in terms of neuroprotective activity could
be observed. However, it is known that (R)-LA is
transformed by lipoamide dehydrogenase to DHLA
with a higher rate than (S)-LA. Also (R)-LA is more
effectively incorporated into mitochondrial enzyme
complexes. Therefore, (S)-LA which is converted into
Thus HOBTꢀTBTU mediated coupling [21] of the
/
above acid with the commercially available 2-morpho-
lin-4-yl-ethylamine proceeds smoothly in tetrahydro-
furan to afford the desired amide containing structure
while the corresponding hydrochloride 1 is obtained by
treatment with 2 N HCl.
The thioamide analog 2 was obtained by reaction of
1a (free base) with Lawesson’s reagent in toluene at
80 8C. (R,S) Lipolol, the alcohol obtained by catechol-
borane reduction of (R,S) LA [22], was employed as
starting material for the obtention of carbamate and
thiocarbamate containing members of the series (Fig. 3).
Thus triethylamine promoted reaction with N,N?-dis-
uccinimidyl carbonate in an acetonitrileꢀtoluene mix-
/
ture afforded derivative 3a that was used as such for the
next step. Carbamate 3 was obtained by reacting 3a with
2-morpholin-4-yl-ethylamine [23] in the presence of
triethylamine in dichloromethane followed by acid
treatment to yield the corresponding hydrochloride
(Fig. 3A).
The thiocarbamate analog 4 was obtained by reacting
racemic lipolol with a tin-oxide reagent followed by
further reaction of the crude 4a with an isothiocyanate

C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ
/
11
3
Fig. 2. Conditions: (i) HOBTꢀ
/
TBTUꢀ
/
ethyl diisopropylamineꢀ
/
, THF, 20 8C, 88%; (ii) 2 N HClꢀ
/
ethanol; (iii) Lawesson’s
reagent, toluene, 80 8C, 31%; (iv) 4 N HClꢀ
/
dioxane, ethanol.
derivative [24] and then hydrochloride formation as
usual (Fig. 3B).
their antioxidant protective effects both in vitro and in
vivo.
Enantiomers 5 and 6 (Table 1) of the racemic
thiocarbamate 4 were obtained by the same reaction
sequence but using optically pure (R) and (S) a-lipoic
acid that could be reduced to the corresponding
optically pure lipolol (see Section 6).
Thus the capacity to inhibit Fe2ꢁ ꢀ
H2O2 induced lipid
/
peroxidation was studied in fresh cortical membrane
preparations, by comparing percentage inhibition in the
presence or absence of antioxidant (see Section 6). Only
the global effect was considered and no attempt was
made to analyse the result in terms of metal chelating,
radical scavenging and/or other effects. However, we
also included in our study, compound 10 which does not
contain the dithiolane moiety.
Experimentation with antioxidants has shown that in
vitro effects as observed in lipid peroxidation or other
tests, do not necessarily reflect the in vivo profile [25].
Therefore, in order to pharmacologically characterise
our compounds in vivo, several pharmacological models
were employed. Thus intracerebroventricular (i.c.v.)
administration of tert-butylhydroperoxide to mice in-
duces a direct oxidative stress leading to an apoptotic
type neurodegeneresence and ultimately to animal
death. The protective effect of our compounds was
evaluated by the lethality observed in animals who
received the antioxidant agent comparative to those
receiving vehicle alone (see Section 6). It was expected
that the protective potency observed in this model will
Furthermore, compounds 7ꢀ10 (Table 1) were also
/
prepared following the same procedure as employed for
compound 4.
4-(2-Isothiocyanato-ethyl)-2,6-dimethyl-morpholine
11 and 5-[1,2] dithiolan-4-yl-pentan-1-ol (12), used for
preparation of compounds 9 and 8, respectively, were
prepared as indicated in Figs. 4 and 5, respectively. For
compound 11, the procedure given in the Section 6 is the
same as a typical procedure previously described [32].
For compound 12, the procedure is given in the Section
6.
3. Pharmacology
The compounds described above were examined in a
series of pharmacological models in order to evaluate
Fig. 3. (A) Conditions, i: N,N? disuccinimidyl carbonate, Et3N, CH3CN, toluene, 20 8C; ii:
HClꢀdioxane. (B) Conditions: i: (Bu3Sn)2O, toluene; ii: , toluene 60 8C; iii: 4 N HClꢀ
, Et3N, CH2Cl2, 20 8C, 78%; iii: 4 N
/
/
dioxane, ethanol, 41%.

4
C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ11
/

C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ
/
11
5
Fig. 4. Conditions: (i) N-(2-Bromoethyl) phtalimide, acetonitrile, 49%; (ii) Hydrazine hydrate, methanol, 79 8C; (iii) DCC, carbondisulphide,
diethylether, 45%.
be strongly influenced by the BBB permeability of the
investigated compound.
The in vivo antioxidant pattern as observed in the
above tests could be compared with the in vitro results,
concerning lipid peroxidation (vide infra). Moreover in
order to further characterise the in vivo behaviour of
our compounds, the whole body temperature of animals
was also measured.
Compounds were also evaluated in an additional
model whereby a peripheral oxidative stress created by
alloxan selectively damages b cells of the pancreas via a
redox cycling mechanism, thus inducing a type I
diabetes in mice. Indeed evidence of oxidative stress in
diabetic rat tissues has been reported in the literature
[26]. Moreover it is known that free radicals are
increasingly formed in diabetes mellitus by the autoox-
idation of glucose and glycosilated proteins [27]. The
potency of our compounds to inhibit alloxan-induced
hyperglycemia in mice was evaluated, in comparison
with lipoic acid.
4. Results and discussion
Investigation of our compounds, in the above men-
tioned models, resulted in the pharmacological profiles
summarised in Table 1. Also indicated are octanolꢀ
/
Fig. 5. Conditions: (i) 1,5-pentanediol, NaOH, H2O, 31%; (ii) P(C6H5)3, Br2, CH3CN, 71%; (iii) sodium ethoxyde, diethylmalonate, 72%; (iv)
LiA1H4, diethylether, 67%; (v) N(C2H5)3, CH2Cl2, methanesulfonylchloride, 91%; (vi) sulphur, Na2S, DMF, 73%; (vii) boron tribromideꢀmethyl
sulphide complex, 1,2-dichloethane, 73%.
/

6
C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ11
/
water partition coefficients measured at pH 7.4 (log D),
using a Sirius apparatus (see Section 6).
shows a very weak level of activity in the alloxan model
but moderate to good activities in the t-BuOOH and
lipid peroxidation tests, respectively. Also, the hypother-
mia effect induced by this compound is lower than the
one usually observed in this series and is similar to the
one observed for lipoic acid. These results tend to
demonstrate, at least partly, that the investigated
compounds are not merely lipoic acid prodrugs but
new chemical entities per se. Nevertheless and despite
the fact that specific enzymes for carbamate and
thiocarbamates are not known, this does not imply
complete resistance of these functions to esterases.
As already documented in the literature for the R and
S enantiomers of lipoic acid, in our case too, no
difference has been observed between racemic com-
pound 4 and its enantiomers 5 and 6. However,
lengthening the chain B by one carbon atom has a
negative impact on the protective effect (compare 4 and
7 in the t-BuOOH model).
It is worthwhile to stress again that not all compounds
that are effective in the lipid peroxidation model (in
vitro test) display activities in the in vivo tests, especially
in what the protection against t-BuOOH induced
lethality model is concerned. Indeed while compound 4
and lipoic acid behave similarly in the lipid peroxidation
test, only the former provides an excellent protective
action against t-BuOOH induced lethality. Also com-
pound 4 is more efficient than lipoic acid in what
hyperglycemia inhibition is concerned. Another striking
example is provided by compounds 2 and 7 both
inhibiting the lipid peroxidation but being completely
inactive in the t-BuOOH test. However, both these
compounds are active and better than lipoic acid in the
alloxan test.
Albeit a genuine correlation between measured parti-
tion data (log D) and protective effect in the t-BuOOH
test cannot be established for this series, it might be
observed that a good BBB crossing is probably linked to
a distribution coefficient higher than 3 as can be judged
from the protection against t-BuOOH induced lethality.
Compound 7 is an outlayer in this respect at least if a
passive mechanism for the BBB crossing is assumed.
This observation tends to indicate that transport me-
chanisms and/or partial metabolisation cannot be com-
pletely ruled out as factors influencing the
pharmacological results in this model.
The obtained results seem to indicate a tendency of
higher hypothermia with increasing log D.
An excellent level of activity in all examined models is
obtained for the 1,2-dithiolane regioisomer 8 and the
morpholine substituted analog 9.
For comparative purposes, commercially available
lipoic acid was also studied in the same models. Thus,
in our hands, this compound provides at a 5 mM
concentration, a high percentage of lipid peroxidation
inhibition (no inhibition was observed at 2 mM).
However, at a dose of 150 mg kgꢂ1 intraperitoneally
(i.p.), no evidence of protection against t-BuOOH
mediated lethality was observed, only toxic effects
(convulsion, morbidity) being apparent in this case.
No toxic effects but also no reduction in t-BuOOH
mediated lethality were observed at a lower dose (50 mg
kgꢂ1 i.p.). On the other hand, a moderate effect in
reducing alloxan induced hyperglycemia was evidenced
after oral administration (400 mg kgꢂ1).
As mentioned above, the chemical structure of the
new antioxidants is characterised by the presence of a
1,2-dithiolane moiety bearing a carbon chain and a N-
alkyl-substituted morpholine ring, both these units
being connected via one of amide, thioamide, carbamate
or thiocarbamate linker (Z). It should be pointed out
that even so compounds including similar connectivities
have been claimed in the recent patent literature, no
integrated in vivo tests have been employed in their
characterisation [28,29].
In our case it was found that the nature of the linker
(Z) as well as the length of the alkyl chains (A and B),
have an impact on the pharmacological results. Thus,
the mere change of the linker from amide to thioamide
increases the lipid peroxidation inhibition (compare 1
and 2) to a level comparative to the one observed for
lipoic acid. However, no real reduction in lethality is
observed for both 1 and 2 in the t-BuOOH model at 50
mg or 150 mg kgꢂ1 i.p. (toxic effects are observed for 1
at the higher dose). Interestingly, however, compound 2
totally inhibits alloxan induced hyperglycemia when
administered orally at a dose of 300 mg kgꢂ1 while
lipoic acid shows a lower effect at a higher dose.
Unfortunately 2 also displays at 150 mg kgꢂ1 i.p. a
more marked hypothermic effect higher than the one
observed for lipoic acid.
The change of the linker from carbamate to thiocar-
bamate also leads to an improvement in the pharmaco-
logical profile (compare 3 and 4). Indeed, while 3 shows
very weak activities or even toxic effects in the investi-
gated models, 4 shows excellent levels of activity both in
the in vitro (lipid peroxidation inhibition) and in vivo
tests. These results indicate an impact of the thiocarba-
mate function on the biological behaviour of our
compounds and are also in line with known literature
data indicating that thiocarbamates alone exhibit anti-
oxidant properties [30].
In summary, we presented in this study new 1,2-
dithiolane derivatives, which demonstrate in vivo phar-
macological profiles superior to those of lipoic acid.
Moreover some of our data indicate the possibility of
modulating peripheric (alloxan test) vs. central (t-
In order to further substantiate these facts, compound
10 lacking the 1,2-dithiolane ring was synthesised and
tested. As can be seen from Table 1, this compound

C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ
/
11
7
BuOOH test) effects. These compounds might be
valuable starting points for drugs acting in situations
of oxidative stress.
2-(4-Morpholino) ethyl isothiocyanate, 3-(4-morpho-
lino) propyl isothiocyanate and 2,6-dimethylmorpholin
were obtained from Lancaster.
Racemic a-lipoic acid (LA) was obtained from Sigma.
All compounds, prepared in this publication, containing
the [1,2-dithiolan-3-yl] or [1,2-dithiolan-4-yl] moiety
were dried at 30 8C maximal temperature due to their
sensitivity to heat induced polymerisation.
5. Conclusion
New antioxidants containing an alkylated 1,2-dithio-
lane unit connected via an amide, thioamide, carbamate
or thiocarbamate linker to a N-alkylated morpholine
unit, have been synthesised and tested. The thioamide
and thiocarbamate linker analogs demonstrate superior
in vivo pharmacological profiles in comparison to their
amide and carbamate counterparts and to the arche-
typal lipoic acid. The stereochemistry of the 1,2-dithio-
lane unit does not influence the obtained
pharmacological profile, both possible enantiomers
being equally active.
6.1.1. (R,S)-5-[1,2] dithiolan-3-yl-pentanoic acid (2-
morpholin-4-yl-ethyl)-amide hydrochloride (1)
To a stirred solution of racemic a-lipoic acid (6.2 g, 30
mmol) in THF (180 mL) were added HOBT (4.5 g, 33.3
mmol), TBTU (10.5 g, 32.7 mmol), ethyl-diisopropyl-
amine (7.8 g, 60 mmol) and 2-morpholin-4-yl-ethyla-
mine (3.9 g, 30 mmol). The mixture was stirred at 20 8C
for 20 h and concentrated under vacuum at a tempera-
ture below 30 8C. The residue was taken up in dichlor-
omethane followed by washing with a solution of
sodium hydrogencarbonate in water. The organic layer
was dried (Na2SO4) and concentrated to dryness.
Column chromatography of the residue (5% methanol
in dichloromethane) afforded 5-[1,2] dithiolan-3-yl pen-
tanoic acid (2-morpholin-4-yl-ethyl)-amide 1a which
was dissolved in ethanol. A small excess of a 2 N
solution of HCl in ethanol was then added and the
mixture was cooled to 5 8C. The precipitated solid was
collected by filtration, washed with ethanol and dried at
Ongoing studies aiming to further characterise this
class of compounds are in course.
6. Experimental
6.1. Chemistry
Compounds prepared in this publication were named
using AUTONOM 2 Software.
All reagents were fresh, commercial grade chemicals.
Melting points (m.p.) were determined on a Mel-
temps capillary apparatus and are uncorrected. 1H-
NMR spectra were recorded on a Bruker AC 200 or
300 spectrometer in DMSO-d6 solutions.
Chemical shifts are expressed in ppm downfield from
a tetramethylsilane internal standard. Significant 1H-
NMR data are reported in the following order: multi-
plicity (bs, broad signal; d, doublet; t, triplet; m,
multiplet; bd, broad triplet; dd, double doublet), num-
ber of protons. Elemental analyses were performed on a
Carlo Erba analyser 1108. Column chromatography was
30 8C under vacuum to afford compound 1 (8.4 g).
1
Yield (88%); m.p. 80ꢀ
/
82 8C. H-NMR (200 MHz): dꢃ
/
11.1 (bs, 1H, exchanged by D2O); 8.2 (t, 1H, exchanged
by D2O); 3.9 (m, 4H); 3.6 (m, 1H); 3.45 (m, 4H); 3.15
(m, 6H); 2.4 (m, 1H); 2.1 (m, 2H); 1.9 (m, 1H); 1.5 (m,
4H); 1.35 (m, 2H). Anal. C14H26N2O2S2, HCl (C, H, N,
S, Cl).
6.1.2. (R,S)-5-[1,2] Dithiolan-3-yl pentanethioic acid
(2-morpholin-4-yl-ethyl)-amide hydrochloride (2)
A mixture of 1a (8.3 g, 26 mmol), Lawesson’s reagent
(5.3 g, 13 mmol) and toluene (250 mL) was stirred at
80 8C during 4 h. After cooling, the mixture was
extracted with an aqueous solution of 1 N HCl. The
aqueous layer was brought to a basic pH with sodium
carbonate and then extracted with ethyl acetate. The
organic layer was dried (Na2SO4) and concentrated
under vacuum. The residue was taken up in ethanol, a
small excess of a 2 N HCl solution in ethanol was added
and the mixture was concentrated under vacuum. The
residue was stirred in ether and the solid collected by
filtration was washed with ether and dried at 30 8C
performed using Merck Silica gel (0.040ꢀ0.063 mm)
/
under a 1 bar nitrogen pressure (flash chromatography).
All reactions were carried out under a nitrogen atmo-
sphere.
Separation of a-lipoic acid enantiomers was achieved
by salt formation with a-methylbenzylamine followed
by crystallisation from toluene [31].
Elemental analyses indicated by the symbols of the
elements were within 90.4% of the theoretical values.
/
Dicyclohexylcarbodiimide (DCC), O-(1H-benzotria-
zol-1-yl)-N,N,N?,N? tetramethyl uronium tetrafluoro-
borate (TBTU), 1-hydroxybenzotriazole (HOBT),
Lawesson’s reagent, bis (tributyltin) oxide, N,N?-dis-
uccinimidyl carbonate, malonic acid diethyl ester, tri-
phenylphosphine, n-pentanol and N-(2-bromoethyl)
phtalimide were obtained from Aldrich.
under vacuum to afford compound 2 (3.0 g). Yield
1
(31%); m.p. 98ꢀ
/
100 8C. H-NMR (200 MHz): dꢃ
/11.4
(bs, 1H, exchanged by D2O); 10.4 (t, 1H, exchanged by
D2O); 3.95 (m, 6H); 3.6 (m, 1H); 3.45 (m, 4H); 3.15 (m,
4H); 2.6 (m, 2H); 2.4 (m, 1H); 1.9 (m, 1H); 1.65 (m, 4H);

8
C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ11
/
1.35 (m, 4H). Anal. C14H26N2O S3, HCl (C, H, N, S,
Cl).
product was dissolved in ethanol and treated with a
small excess of 4 N HCl in dioxane followed by
concentration of the solution under vacuum. The
residue was stirred in ether, the crystals were collected
by filtration, washed with ether and hexane and then
6.1.3. (R,S)-(2-Morpholin-4-yl-ethyl)-carbamic acid 5-
[1,2] dithiolan-3-yl pentyl ester hydrochloride (3)
N,N?-disuccinimidyl carbonate (11.5 g, 0.045 mol)
and triethylamine (9 g) were added to a stirred suspen-
sion of (R,S) lipolol (5.77 g, 0.03 mol) in toluene (50
mL) and acetonitrile (150 mL). After 3 h of stirring, a
further amount of N,N?-disuccinimidyl carbonate (2.5
g) and triethylamine (2 g) was added. The mixture was
stirred for 16 h at 20 8C, concentrated under vacuum at
a temperature below 30 8C, the residue was taken up in
an aqueous solution of sodium chloride, dried (Na2SO4)
and concentrated under vacuum at 30 8C. Column
chromatography of the residue using dichloromethane
as eluent afforded carbonic acid 2,5-dioxo-pyrrolidin-1-
yl ester 5-[1,2] dithiolan-3-yl pentyl ester 3a. The crude
residue was dissolved in dichloromethane (60 mL) and
added over 30 min to a stirred solution of 2-morpholin-
4-yl-ethylamine (4.68 g, 0.036 mol) and triethylamine (6
g) in dichloromethane (100 mL). The mixture was
stirred for 18 h at 20 8C, washed with an aqueous
solution of sodium chloride, dried (Na2SO4) and con-
centrated under vacuum at a temperature below 30 8C.
Chromatography of the residue using a mixture of 4%
methanol in dichloromethane as eluent gave (R,S)-(2-
morpholinyl-4-yl-ethyl)-carbamic acid 5-[1,2] dithiolan-
3-yl pentyl ester which was dissolved in ethanol; the
solution was treated with a small excess of 4 N HCl in
dioxane and concentrated under vacuum at a tempera-
ture below 30 8C. The residue was stirred in ether, the
crystals were collected by filtration and dried under
vacuum at 30 8C to afford compound 3 (8.2 g). Yield
dried under vacuum at 30 8C, affording compound 4
1
(8.2 g). Yield (41%); m.p. 105ꢀ
/
110 8C. H-NMR (300
MHz): dꢃ/10.75 (bs, 1H, exchanged by D2O); 9.25 (t,
1H, exchanged by D2O); 4.3 (t, 2H); 3.95 (m, 2H); 3.75
(m, 4H); 3.6 (m, 1H); 3.45 (m, 2H); 3.2 (m, 2H); 3.15 (m,
4H); 2.4 (m, 1H); 1.9 (m, 1H); 1.65 (m, 4H); 1.4 (m, 4H).
Anal. C15H28N2O2S3, HCl (C, H, N, S, Cl).
6.1.5. (R)-(2-Morpholin-4-yl-ethyl)-thiocarbamic acid
0-(5-[1,2] dithiolan-3-yl-pentyl) ester hydrochloride (5)
(R)-Lipolol (5.8 g, 0.03 mol) used as starting material
afforded compound 5 (4.3 g) by the same method as
described for 4. Yield (36%); m.p. 108ꢀ
/
110 8C. 1H-
NMR (300 MHz): dꢃ10.75 (bs, 1H, exchanged by
/
D2O); 9.25 (t, 1H, exchanged by D2O); 4.3 (t, 2H); 3.95
(m, 2H); 3.75 (m, 4H); 3.6 (m, 1H); 3.45 (m, 2H); 3.2 (m,
2H); 3.15 (m, 4H); 2.4 (m, 1H); 1.9 (m, 1H); 1.65 (m,
4H); 1.4 (m, 4H). Anal. C14H26N2O2S2, HCl (C, H, N,
S, Cl).
6.1.6. (S)-(2-Morpholin-4-yl-ethyl)-thiocarbamic acid
0-(5-[1,2] dithiolan-3-yl-pentyl) ester hydrochloride (6)
(S)-Lipolol (5.8 g, 0.03 mol) used as starting material
afforded compound 6 (4.7 g) by the same method as
described for 4. Yield (39%); m.p. 118ꢀ
/
120 8C.1H-NMR
(300 MHz): dꢃ10.75 (bs, 1H, exchanged by D2O); 9.25
/
(t, 1H, exchanged by D2O); 4.3 (t, 2H); 3.95 (m, 2H);
3.75 (m, 4H); 3.6 (m, 1H); 3.45 (m, 2H); 3.2 (m, 2H);
3.15 (m, 4H); 2.4 (m, 1H); 1.9 (m, 1H); 1.65 (m, 4H); 1.4
(m, 4H). Anal. C15H28N2O2S3, HCl (C, H, N, S, Cl).
(71%); m.p. 113ꢀ
/
115 8C. 1H-NMR (300 MHz); dꢃ
10.5
/
(bs, 1H, exchanged by D2O); 7.35 (t, 1H, exchanged by
D2O); 3.75 (m, 2H); 3.6 (m, 1H); 3.4 (m, 4H); 3.15 (m,
6H); 2.4 (m, 1H); 1.9 (m, 1H); 1.65 (m, 1H); 1.55 (m,
3H); 1.35 (m, 4H). Anal. C15H28N2O3S2, HCl (C, H, N,
S, Cl).
6.1.7. (R,S)-(3-Morpholin-4-yl-propyl)-thiocarbamic
acid 0-(5-[1,2] dithiolan-3-yl-pentyl) ester hydrochloride
(7)
3-(4-Morpholinyl)propylisothiocyanate (8.4 g, 0.045
mol) and racemic lipolol (5.8 g, 0.03 mol) used as
starting materials afforded compound 7 (1.55 g) by the
6.1.4. (R,S)-(2-Morpholin-4-yl-ethyl)-thiocarbamic
acid O-(5-[1,2] dithiolan-3-yl pentyl) ester hydrochloride
(4)
same method as described for 4. Yield (11%); m.p. 86ꢀ
/
88 8C. 1H-NMR (300 MHz): dꢃ
/10.65 (bs, 1H, ex-
Racemic lipolol (9.6 g, 0.05 mol) and bis-(tributyltin)-
oxide (30 g, 0.05 mol) in toluene (400 mL) were stirred at
a reflux temperature for 20 h. The reaction mixture was
cooled to 60 8C and 2-(4-morpholinyl) ethyl isothiocya-
nate (17.2 g, 0.1 mol) were added dropwise; the mixture
was maintained at 40 8C overnight. After cooling, the
mixture was washed with an aqueous solution of sodium
carbonate and concentrated to 100 mL under vacuum.
Column chromatography of the solution eluting with a
mixture 4% THF in dichloromethane afforded (R,S)-(2-
morpholin-4-yl-ethyl)-thiocarbamic acid O-(5-[1,2]
dithiolan-3-yl pentyl) ester as the free base. The above
changed by D2O); 9.15 (t, 1H, exchanged by D2O); 4.3
(t, 2H); 3.9 (m, 2H); 3.75 (m, 2H); 3.6 (m, 1H); 3.4 (m,
2H); 3.3 (m, 2H); 3.15 (m, 6H); 2.4 (m, 1H); 1.9 (m, 3H);
1.65 (m, 4H); 1.4 (m, 4H). Anal. C15H28N2O2S3, HCl (C,
H, N, S, Cl).
6.1.8. (2-Morpholin-4-yl-ethyl)-thiocarbamic acid 0-5-
[1,2] dithiolan-4-yl-pentyl) ester hydrochloride (8)
5-[1,2] dithiolan-4-yl-pentan-1-ol (12) (5.8 g, 0.03 mol)
used as starting material afforded compound 8 (3.0 g) by
the same method as described for 4. Yield (25%); m.p.
80ꢀ
/
85 8C. 1H-NMR (300 MHz): dꢃ
11.05 (bs, 1H,
/

C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ
/
11
9
exchanged by D2O); 9.25 (t, 1H, exchanged by D2O); 4.3
(t, 2H); 3.95 (m, 2H); 3.8 (m, 4H); 3.4 (m, 2H); 3.25 (m,
3H); 3.05 (m, 2H); 2.8 (m, 2H); 2.5 (m, 2H); 1.65 (m,
2H); 1.4 (m, 6H). Anal. C15H28N2O2S3, HCl (C, H, N,
S, Cl).
methane, the organic layer dried (Na2SO4) and concen-
trated to dryness. Distillation of the residue gave
compound 14 (29.8 g, 79%): b.p. 50 8C/0.1 mm.
6.1.13. 4-(2-Isothiocyanato-ethyl)-2,6-dimethyl-
morpholine (11)
6.1.9. (R,S)-[2-(2,6-Dimethyl-morpholin-4-y)l-ethyl]-
thiocarbamic acid O-(5-[1,2] dithiolan-3-yl-pentyl) ester
hydro-chloride (9)
4-(2-Isothiocyanato-ethyl)-2,6-dimethyl-morpholine
(11) (11.2 g, 0.06 mol) and racemic lipolol (5.8 g, 0.03
mol) used as starting materials afforded compound 9
2-(2,6-Dimethyl-morpholin-4-yl)-ethylamine (14) (20
g, 0.126 mol) and DCC (26 g, 0.126 mol) were dissolved
in diethyl ether (200 mL). Carbon disulphide (9.6 g,
0.126 mol) in diethyl ether (7.6 mL) was added over a 30
min period. The mixture was stirred for 16 h at 20 8C.
The precipitate was filtered and the filtrate concentrated
to dryness. Chromatography of the residue on silicagel
eluting with dichloromethane gave compound 11 (11.4
(2.9 g) by the same method as described for 4. Yield
1
(23%); m.p. 85ꢀ
/
88 8C. H-NMR (300 MHz): dꢃ
/
11.3
(bs, 1H, exchanged by D2O); 9.25 (t, 1H, exchanged by
D2O); 4.3 (t, 2H); 3.95 (m, 2H); 3.8 (m, 2H); 3.6 (m, 1H);
3.45 (m, 2H); 3.15 (m, 4H); 2.65 (m, 2H); 2.4 (m, 1H);
1.9 (m, 1H); 1.65 (m, 4H); 1.4 (m, 4H); 1.15 (d, 6H).
Anal. C17H32N2O2S3, HCl (C, H, N, S, Cl).
g, 45%). 1H-NMR (300 MHz): dꢃ
3.72 (m, 2H); 3.61 (t,
/
2H); 2.74 (m, 2H); 2.68 (t, 2H); 1.91 (t, 2H); 1.22 (d,
6H).
6.1.14. 5-Benzyloxy-pentan-1-ol (15)
Benzyl chloride (506 g, 4 mol) was added to a stirred
solution of 1,5-pentanediol (416 g, 4 mol) and a 12 N
sodium hydroxide aqueous solution (510 mL). The
mixture was stirred at 90 8C for 16 h. After cooling, it
was filtered, the organic layer obtained from the filtrate,
dried (Na2SO4) and concentrated to dryness. The
residue was taken up with toluene, concentrated and
distilled under vacuum to give compound 15 as an oil
6.1.10. (2-Morpholin-4-yl-ethyl)-thiocarbamic acid O-
pentyl ester (10)
n-Pentanol (4.3 g, 0.042 mol) used as starting material
afforded compound 10 (4.5 g) by the same method as
described for 4. Yield (31%); m.p. 108ꢀ
/
110 8C. 1H-
NMR (300 MHz): dꢃ11.0 (bs, 1H, exchanged by D2O);
/
9.25 (t, 1H, exchanged by D2O); 4.3 (t, 2H); 3.95 (m,
2H); 3.8 (m, 4H); 3.45 (m, 2H); 3.25 (m, 2H); 3.1 (m,
2H); 1.65 (m, 2H); 1.3 (m, 4H); 0.9 (m, 3H). Anal.
C12H24N2O2S, HCl (C, H, N, S, Cl).
(240 g, 31%): b.p. 124ꢀ130/0.1 mm.
/
6.1.15. (5-Bromo-pentyloxymethyl)-benzene (16)
Bromine (4.9 mL) and 4-benzyloxy-pentan-1-ol (15)
(19.4 g, 0.1 mol) in acetonitrile (150 mL) was added
successively to a stirred solution of triphenylphosphine
(25 g) in acetonitrile (200 mL) at a temperature below
10 8C. The mixture was stirred 2 h at 10 8C and 1 h at
20 8C. Triphenylphosphine oxide was filtered and the
filtrate concentrated under vacuum. Column chromato-
graphy of the residue eluting with a mixture of 15%
dichloromethane in cyclohexane gave compound 16
(18.3 g, 71%).
6.1.11. 2-[2-(2,6-Dimethyl-morpholin-4-yl)-ethyl]-
isoindole-1,3-dione (13)
N-(2-Bromoethyl)-phtalimide (25.4 g, 0.1 mol) and
2,6-dimethylmorpholine (23 g, 0.2 mol) in acetonitrile
(250 mL) are stirred at reflux temperature for 16 h. The
mixture was concentrated under vacuum to dryness. The
residue was dissolved in diethylether, washed succes-
sively with aqueous solutions of sodium carbonate and
sodium chloride, dried (Na2SO4) and concentrated to
dryness. Chromatography of the residue on silicagel
eluting with 4% tetrahydrofuran in dichloromethane
gave compound 13 (14 g, 49%).
6.1.16. 3-(5-Benzyloxy-pentyl)-pentanedioic acid diethyl
ester (17)
Sodium (11 g, 0.48 mol) was added over 20 min to
ethanol (400 mL) to give sodium ethoxide. Malonic acid
diethyl ester (80 g, 0.5 mol) was added and the mixture
stirred for 10 min and concentrated under vacuum to
dryness. The residue was taken up with toluene and the
suspension concentrated under vacuum to dryness. The
residue was dissolved in dimethylformamide (1 L), (5-
bromo-pentyloxymethyl)-benzene 16 (123 g, 0.48 mol)
was added, the mixture stirred for 4 days at 20 8C and
concentrated to dryness under vacuum. The residue was
dissolved in ether, washed successively with aqueous
solutions of sodium hydrogencarbonate and lithium
chloride, dried (Na2SO4), concentrated and distilled
6.1.12. 2-(2,6-Dimethyl-morpholin-4-yl)-ethylamine
(14)
2-[2-(2,6-Dimethyl-morpholin-4-yl)-ethyl]-isoindole-
1,3-dione (13) (69 g, 0.239 mol) was dissolved in
methanol (1 L). Hydrazine hydrate (13 mL) was added
and the mixture was stirred under reflux for 6 h and at
20 8C for 16 h. The mixture was cooled at 5 8C and an
aqueous solution of 4 N hydrochloric acid (200 mL) was
added. The solid was collected by filtration, washed with
water and the combined filtrates concentrated to dry-
ness. The residue was taken up with an aqueous solution
of 4 N sodium hydroxyde, extracted with dichloro-

10
C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ11
/
under reduced pressure to give compound 17 as an oil
(116 g, 72%): b.p. 175 8C/0.1 mm.
residue on silicagel eluting with dichloromethane gave
1
compound 12 (0.15 g, 73%). H-NMR (300 MHz): dꢃ
4.29 (bt, 1H, exchangeable by D2O); 3.33 (d, 2H); 3.23
(dd, 2H); 2.78 (dd, 2H); 2.45 (m, 1H); 1.48ꢀ1.2 (m, 8H).
/
6.1.17. 2-(5-Benzyloxy-pentyl)-propane-1,3-diol (18)
3-(5-Benzyloxy-pentyl)-pentanedioic acid diethyl ester
(17) (14.5 g, 0.043 mol) dissolved in diethyl ether (50
mL) was added dropwise to a lithium aluminium
hydride in diethyl ether (50 mL) suspension. The
mixture was stirred for 2 h at reflux temperature and
20 h at 20 8C. After cooling at 5 8C, water (2.2 mL), 2 N
aqueous solution of sodium hydroxyde (4.4 mL) and
water (2.2 mL) was successively added. The suspension
was filtered, the solid washed with diethylether and the
combined filtrates dried (Na2SO4) and concentrated
under vacuum. Chromatography of the residue on
silicagel (1 L) eluting with a mixture 20% tetrahydro-
furan in dichloromethane gave compound 18 (7.3 g,
67%).
/
6.2. Lipophilicity
Octanolꢀwater (0.15 M KCl) distribution coefficients
/
(log D) at pH 7.4 were measured by a potentiometric
method using a Sirius GlpKa apparatus [33,34]. The
experiments were carried out under strictly controlled
conditions of temperature (25 8C) and ionic strength
(0.15 M KCl). At least two different Voctanol/Vaqueous
phase ratios were examined. Log D values for investi-
gated compounds are indicated in Table 1.
6.3. Pharmacology
6.3.1. Peroxidation assay [35]
Fresh mice cortical membrane preparations (1 g per
6.1.18. Methanesulfonic acid 7-benzyloxy-2-
methanesulfonyloxymethyl-heptyl-ester (19)
20 mL, 20 mM) TrisꢀHCl, pH 7.4) were preincubated
/
Methanesulfonyl chloride (20.7 g, 0.18 mol) was
added at 5 8C to a stirred solution of 2-(5-benzyloxy-
pentyl)-propane-1,3-diol (18) (7 g, 0.028 mol) and
triethylamine (56 mL) in dichloromethane (115 mL).
The mixture was stirred at 5 8C for 5 h and at 20 8C for
20 h. Dichloromethane (200 mL) was added and the
mixture washed with an aqueous solution of sodium
hydrogencarbonate, dried (Na2SO4) and concentrated
to dryness under vacuum. Chromatography of the
residue on silicagel (1.2 L) eluting with 5% ethyl acetate
in dichloromethane gave compound 19 (10 g, 91%).
for 15 min at 27 8C with the tested antioxidant agent (5
mM). Membranes were then treated for 15 min with
FeSO4 (100 mM)ꢀ
The reaction was stopped by addition of trichloroa-
cetic acid (20% v/v) at ꢁ4 8C, followed by centrifuga-
tion for 5 min at ꢁ4 8C (1500 g). After addition of 1 mL
/
ascorbic acid (1 mM)ꢀ/H2O2 (1 mM).
/
/
thiobarbituric acid (0.67%), the supernatants were
incubated at 100 8C for 20 min. The reaction was
stopped in ice and tissue concentration of malonodial-
dehyde (expressed as TBARS formation) were measured
at 532 nm (Beckman DU 640 spectrophotometer; a
standard curve of malonodialdehyde, 0ꢀ40 mM, was
/
6.1.19. 4-(5-Benzyloxy-pentyl)-[1,2] dithiolane (20)
Methane sulfonic acid 7-benzyloxy-2-methanesulfo-
nyloxymethyl-heptyl ester (19) (36.5 g, 0.092 mol) was
dissolved in dimethylformamide (360 mL). Sulphur
(2.95 g) and sodium sulphide monohydrate (22 g) was
added and the mixture was stirred at 85 8C for 4 h.
Diethylether (1 L) was added and the mixture was
washed with an aqueous solution of sodium chloride,
dried (Na2SO4) and concentrated under vacuum to
dryness. Chromatography of the residue on silicagel
eluting with a mixture of 50% cyclohexane in dichlor-
omethane gave compound 20 (19 g, 73%).
established for each experiment). Percentage inhibition
of lipid peroxidation was expressed in drug treated
animals relative to controls (FeSO4ꢀ
/
ascorbic acidꢀ
/
H2O2 treated animals).
6.3.2. t-Butylhydroperoxide mediated lethality [36]
Based on the method previously described by Adams
et al., male NMRI (28ꢀ35 g) mice (nꢃ10 per group)
/
/
were injected i.p. with the compound under study (150
mg kgꢂ1 in 0.2 mL Tween-saline solution, gauge needle
4.48 mm) 30 min before an i.c.v. injection of a 70%
solution of t-BuOOH (22 mg kgꢂ1 per 1 mL saline).
Lethality was assessed 2 h after administration of t-
BuOOH and was expressed as the percent survival
relative to the lethality observed in t-BuOOH plus
Tween-saline vehicle-treated animals.
6.1.20. 5-[1,2] Dithiolan-4-yl-pentan-1-ol (12)
4-(5-Benzyloxy-pentyl)-[1,2] dithiolane (20) (0.3 g,
0.00106 mol) was dissolved in 1,2-dichloroethane (3
mL). A 1 M solution of boron tribromide-methylsul-
phide complex in dichloromethane (3 mL) was added
and the mixture stirred at 20 8C for 2 h, poured in an
aqueous solution of sodium hydrogencarbonate and
extracted with diethyl ether. The combined organic
layers was dried (Na2SO4) and concentrated to dryness
at a temperature below 30 8C. Chromatography of the
6.3.3. Alloxan-induced hyperglycemia [37]
Fasted (ꢁ
/
18 h) male NMRI (28ꢀ
/
35 g) mice (nꢃ10
/
per group) were administered by oral route with the
compound under study (400 mg kgꢂ1 in Tween-water;
20 mL kgꢂ1), 60 min before an intravenous (i.v.)
injection of alloxan monohydrate (40 mg kgꢂ1). Ani-

C. Guillonneau et al. / European Journal of Medicinal Chemistry 38 (2003) 1ꢀ
/
11
11
[11] V.E. Kagan, A. Shvedova, E. Serbinova, S. Khan, C. Swanson, R.
Powell, L. Packer, Biochem. Pharmacol. 44 (1992) 1637ꢀ1649.
[12] D.P. Xu, W.W. Wells, J. Bioenerg. Biomembr. 28 (1996) 78ꢀ85.
[13] A. Bast, G.R.M.M. Haenen, Biochim. Biophys. Acta 963 (1988)
558ꢀ562.
mals were sacrificed at 24 h, the plasma recovered and
-glucose levels were evaluated by assay in a Roche
CabasꢀFara Analyser. The protection against hypergly-
cemia was estimated as the difference between alloxan
drug-treated animals relative to alloxan-treated animals.
/
D
/
/
/
[14] B.C. Scott, O.I. Aruoma, P.J. Evans, C. O’Neill, A. van der Vliet,
C.E. Cross, H. Tritschler, B. Halliwell, Free Radic. Res. 20 (1994)
119ꢀ
[15] L. Packer, H.J. Tritschler, K. Wessel, Free Radic. Biol. Med. 22
(1997) 359ꢀ378.
[16] U. Muller, J. Krieglstein, Brain Res. 678 (1995) 265ꢀ
/
133.
6.3.4. Effect on whole body temperature
The whole body temperature of male NMRI mice
/
(nꢃ/10 per group) was measured with a rectal probe
/
268.
[17] P. Wolz, J. Krieglstein, Neuropharmacology 35 (1996) 369ꢀ
[18] S. Loffelhardt, C. Bonaventura, M. Locher, H.O. Borbe, H.
¨
(Physitemp, Bat-12) at tꢃ
/0 h.
/375.
Animals (NMRI) were then treated with either vehicle
(20 mL kgꢂ1) or the compound under study, by
intraperitoneal administration. Rectal temperature was
assessed every 30 min for up to 2 h post-injection.
Results are expressed as the mean maximal temperature
difference, at a given time, between control saline-
treated animals and drug-treated mice.
¨
Bisswanger, Biochem. Pharmacol. 50 (1995) 637ꢀ
[19] H. Schempp, H. Ulrich, E.F. Elstner, Z. Naturforsh 49 (1994)
691ꢀ692.
/
646.
/
[20] S. Goldstein, C. Guillonneau, Y. Charton, P. Lestage, B. Lock-
hart, 1,2 Dithiolane derivatives, their process of preparation and
pharmaceutical compositions containing them EP 1010698 ADIR
& Cie, 92415 Courbevoie, France, 2001.
[21] O. Frey, M. Hoffmann, H. Kessler, Angew. Chem. Int. Ed. Engl.
34 (1995) 2026ꢀ
[22] M. Saah, W.M. Wu, K. Eberst, E. Marvanyos, N. Bodor, J.
Pharm. Sci. 85 (1996) 496ꢀ504.
[23] A.K. Ghosh, T.T. Duong, S.P. McKee, W.J. Thompson, Tetra-
hedron Lett. 33 (1992) 2781ꢀ2784.
[24] K. Nishiyama, M. Oba, Tetrahedron Lett. 34 (1993) 3745ꢀ
[25] T.L. Yue, J.L. Gu, P.G. Lysko, H.Y. Cheng, F.C. Barone, G.
Feuerstein, Brain Res. 574 (1992) 193ꢀ197.
[26] B. Matkovics, M. Sasvari, M. Kotorman, I.S. Varga, D.Q. Hai,
C. Varga, Acta Physiol. Hung. 85 (1998) 183ꢀ192.
[27] R.V. Aruna, B. Ramesh, V.N.R. Kartha, Indian J. Exp. Biol. 37
(1999) 399ꢀ401.
/
2028.
Acknowledgements
/
We thank Dr. J.P. Volland and his staff for analytical
and NMR data, the staff of Cerebral Pathology Divi-
sion for the pharmacological testing and Mrs. F.
Chevereau for the typing of the manuscript.
/
/
3748.
/
/
References
/
[28] T. Fujita, T. Yokoyama, Dithiolan derivatives, their preparation
and their therapeutic effect EP0869126 Sankyo Company Ltd.,
Tokyo 103-8426, Japan, 1998.
[1] B. Halliwell, J.M.C. Gutteridge, Free Radicals in Biology and
Medicine, 3rd ed, Oxford, London, 1999, pp. 246ꢀ350.
/
[2] K. Hensley, Q.N. Pye, T. Tabatabaie, C.A. Stewart, R.A. Floyd,
in: P.L. woods (Ed.), Neuroinflammation: Mechanism and
[29] L. Packer, O. Tirosh, C.K. Sen, S. Roy, Lipoic acid analogs WO
99/45922 The Regents of the University of California, Oakland,
CA 94612-3550, USA, 2000.
Management, Humana Press, Totowa, NJ, 1998, pp. 265ꢀ276.
/
[3] H. Sies (Ed.), Antioxidants in Disease, Mechanisms and Therapy,
Academic Press, London, 1997.
[30] R. Schreck, R. Grassmann, B. Fleckenstein, P.A. Baeuerle, J.
Virol. 66 (1992) 6288ꢀ6293.
/
[4] J. Fuchs, L. Packer, G. Zimmer (Eds.), Lipoic Acid in Health and
Disease, Marcel Dekker, New York, 1997.
[5] S. Roy, C.K. Sen, H.J. Tritschler, L. Packer, Biochem. Pharma-
[31] H. Bethge, R. Moller, T. Deibvvenger, K. Huthmacker, G.
¨
Blaschke, Preparation and use of enantiomerically pure a-Lipoic
acid salts Deutsches Petentamt DE 4137773A1, 1994.
col. 53 (1997) 393ꢀ
[6] M. Podda, H.J. Tritschler, H. Ulrich, L. Packer, Biochem.
Biophys. Res. Commun. 204 (1994) 98ꢀ104.
[7] M. Panigrahi, Y. Sadguna, B.R. Shivakumar, S.V. Kolluri, S.
Roy, L. Packer, V. Ravindranath, Brain Res. 717 (1996) 184ꢀ188.
[8] G.J. Handelman, D. Han, H.J. Tritschler, L. Packer, Biochem.
Pharmacol. 47 (1994) 1725ꢀ1730.
[9] Y.J. Suzuki, M. Tsuchiya, L. Packer, Free Radic. Res. Commun.
15 (1991) 255ꢀ263.
[10] L.J. Yan, M.G. Traber, H. Kobuchi, S. Matsugo, H.J. Tritschler,
L. Packer, Arch. Biochem. Biophys. 327 (1996) 330ꢀ334.
/
399.
[32] J.P. He´nichart, J.L. Bernier, R. Houssin, Synthesis (1980), 311ꢀ
312.
[33] F.H. Clarke, N.M. Cahoon, J. Pharm. Sci. 76 (1987) 611ꢀ
[34] B. Slater, A. McCormack, A. Avdeef, J.E.A. Comer, J. Pharm.
Sci. 83 (1994) 1280ꢀ1283.
[35] M. Mihara, M. Uchiyama, Anal. Biochem. 86 (1978) 271ꢀ
/
/
/
619.
/
/
/
278.
/
[36] J.D. Adams, Jr., L.K. Klaidman, Y.M. Huang, J.J. Cheng, Z.J.
Wang, M. Nguyen, B. Knusel, A. Kuda, Mol. Chem. Neuro-
/
pathol. 22 (1994) 123ꢀ
/142.
[37] L.W. Oberley, Free Radic. Biol. Med. 5 (1988) 113ꢀ
/124.
/