Synthesis and antioxidant properties of a new lipophilic ascorbic acid analogue

Article abstract of DOI:10.1016/S0968-0896(02)00446-7

4-(4-Hydroxyphenyl)-5-(4-hydroxyphenylmethyl)-2-hydroxyfurane-2-one 1 was prepared by an acidic dimerisation of 4-hydroxyphenylpyruvic acid and some of its antioxidant and spectroscopic properties have been measured and compared to that of ascorbic acid. 1 is as good an antioxidant as ascorbic acid in the DPPH (2,2-diphenyl-1-picryl hydrazyl radical) test and the inhibition of hydroxyl radical and a powerful inhibitor of the Cu²⁺ or AAPH (2,2′-azobis-(2-amidinopropane) dihydrochloride) induced oxidation of human LDL. 1 gives a stable radical characterised by its ESR spectrum similarly to ascorbic acid but in lower concentration and with a different reactivity towards nitroxides. Theoretical calculations allow us to propose the structure for the radical formed from 1, to explain its lower stability than ascorbyl radical and to evaluate the lipophilicity of 1.

Full text of DOI:10.1016/S0968-0896(02)00446-7

Bioorganic & Medicinal Chemistry 11 (2003) 1087–1093
Synthesis and Antioxidant Properties of a New Lipophilic
Ascorbic Acid Analogue
Philippe Cotelle, * Nicole Cotelle, Elisabeth Teissier and Herve
´
Vezina
a,
a
b
aLaboratoire de Chimie Organique et Macromol e´ culaire, UPRESA CNRS 8009, Universite´ des Sciences et Technologies de Lille,
5
9655 Villeneuve d’Ascq, France
b
D e´ partement de recherches sur les lipoprote´ines et l’ath e´ roscl e´ rose, INSERM U325, Institut Pasteur et Faculte´ de Pharmacie,
Universit e´ de Lille 2, 59000 Lille, France
Received 14 June 2002; accepted 6 September 2002
Abstract—4-(4-Hydroxyphenyl)-5-(4-hydroxyphenylmethyl)-2-hydroxyfurane-2-one 1 was prepared by an acidic dimerisation of 4-
hydroxyphenylpyruvic acid and some of its antioxidant and spectroscopic properties have been measured and compared to that of
ascorbic acid. 1 is as good an antioxidant as ascorbic acid in the DPPH (2,2-diphenyl-1-picryl hydrazyl radical) test and the inhi-
2
+
0
bition of hydroxyl radical and a powerful inhibitor of the Cu
or AAPH (2,2 -azobis-(2-amidinopropane) dihydrochloride)
induced oxidation of human LDL. 1 gives a stable radical characterised by its ESR spectrum similarly to ascorbic acid but in lower
concentration and with a different reactivity towards nitroxides. Theoretical calculations allow us to propose the structure for the
radical formed from 1, to explain its lower stability than ascorbyl radical and to evaluate the lipophilicity of 1.
#
2002 Elsevier Science Ltd. All rights reserved.
Introduction
on the role of ascorbic acid in the prevention of cardio-
vascular disease is inconclusive, with some studies
showing a very strong correlation between increased
ascorbic acid intake and incidence of CVD events and
other studies showing no correlation at all. Numerous
in vitro studies have demonstrated that ascorbic acid
inhibits LDL oxidation by a variety of mechanisms.
Oxidative stress is thought to play an important con-
tributory role in the pathogenesis of numerous degen-
erative or chronic diseases, such as atherosclerosis and
cancer. Considerable experimental evidence links the
production of reactive oxygen species to biological
damage that can potentially provide a mechanistic basis
for their initiation and/or progression. The human
antioxidant protections include enzymes such as cata-
lase, superoxide dismutase and glutathione peroxidase
as well as water- and lipid-soluble antioxidants such as
ascorbic acid (vitamin C) and a-tocopherol (vitamin
7
Retsky et al. have suggested two mechanisms by which
ascorbic acid may protect LDL against oxidation.
Ascorbic acid may scavenge free radicals in the aqueous
phase or dehydroascorbic acid (the two-electron oxida-
tion form of ascorbic acid), or its decomposition pro-
ducts may modify LDL leading to decreased copper
binding to the LDL particle and increased resistance to
copper-dependent oxidation. The recent efforts devoted
to the development of new ascorbic acid analogues have
1ꢀ4
E).
Ascorbic acid can be considered as one of the most
potent naturally occurring antioxidants because it
works by reacting with aqueous peroxyl radicals and by
restoring the antioxidant properties of vitamin E.
However, its high hydrophilicity limits this latter prop-
erty. The antioxidant and pro-oxidant activities of
8
ꢀ10
11
led to the obtention of antioxidant,
1
antitumoural
2
and potentially anti-inflammatory agents.
5
In connection with our interest in the synthesis of nat-
1
3,14
ural polyphenols and our knowledge on the acid-
catalysed dimerisation of arylethanoid and arylpropa-
6
ascorbic acid remain controversal. Similarly to animal
data, the current evidence from epidemiological studies
1
5ꢀ17
we felt that arylpyruvic acids may
noid derivatives,
lead to lipophilic ascorbic acid analogues by acidic
dimerisation. Indeed, such dimerisation has been
evoked in the biosynthesis of butyrolactone isolated
*
0
Corresponding author. Tel.: +333-20-33-72-31; fax: +333-20-33-63-
9; e-mail: philippe.cotelle@univ-lille1.fr
1
8
from Aspergillus terreus
and numerous phenolic
0
968-0896/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00446-7

1088
P. Cotelle et al. / Bioorg. Med. Chem. 11 (2003) 1087–1093
1
9
compounds from the mushroom Suillus spp. In order
condensation followed by a lactonisation and finally a
decarboxylation (Scheme 1). Treating 2-(4-acetoxy-
phenylmethylene)-5-methyloxazol-3-one, a precursor
of 4-hydroxyphenylpyruvic acid, with 3 N HCl during
24 h can increase the yield in 1 up to 50% with the for-
mation of 4-hydroxybenzaldehyde as an impurity.
2
0
to obtain true ascorbic acid analogues, we studied the
dimerisation of 4-hydroxyphenylpyruvic acid. To con-
firm the ascorbic acid character of the synthesised
molecule, we evaluated its antioxidant activity by the
2
4
2
1,22
DPPH test and the hydroxyl radical scavenging
and
evidenced the generation of a stable radical under
slightly alkaline conditions by ESR spectroscopy. The
The antioxidant properties of 1 and ascorbic acid were
expressed by the traditional EC50 parameter describing
the global reactivity of the tested compound towards
DPPH and the kinetic parameter log Z influenced only
by the first step of the reaction, that is, the hydrogen
abstraction (from the donor to give the hydrazine and a
radical) (see Experimental section for the calculation of
log Z). 1 And ascorbic acid did not show any differ-
ence (see Table 1) in this test, suggesting that the two
molecules interact in the same manner with DPPH. The
total stoichiometry ([DPPH]/2EC50) of 1 and ascorbic
acid is about 2and corresponds to the total number of
labile hydrogen atoms transferred to DPPH.
2
+
inhibitory effect of 1 on the Cu
or AAPH induced
oxidation of human LDL was also evaluated. Addi-
tionally, theoretical calculations were performed to
determine the structure of the radical formed and its
stability has been compared with that of ascorbyl radi-
cal. The lipophilic character of the new molecule was
also evaluated by the calculation of log P.
2
5
Results and Discussion
Our first attempts were directed towards the concomitant
demethylative dimerisation of 4-methoxyphenylpyruvic
acid using boron tribromide. Unfortunately, the de-
methylation occurred but the expected dimerisation
ꢁ
The inhibition of hydroxyl radical (OH ) using DMPO
(dimethylpyrroline-N-oxide) as spin trap was carried
out only for 1 since ascorbic acid presents several pro-
blems in this test: (i) the second-order rate constants for
2
3
failed. So, we turned to the treatment of 4-hydroxy-
phenylpyruvic acid with concentrated hydrochloric acid.
We serendipitously found that zinc amalgam in boiling
hydrochloric acid gives the desired dimerisation product
ꢁ
the reaction of OH with ascorbic acid and DMPO are
9
9
ꢀ1 ꢀ1
7.2ꢂ10 and 3.4ꢂ10 M
s respectively. Therefore
1
phenyllactic acid
in 20% yield beside the expected 4-hydroxy-
2
the ESR signal of ascorbyl radical may superimpose on
that of DMPO–OH and perturb the measurement of
DMPO–OH signal; (ii) ascorbic acid can reduce
4
in 50% yield. The supposed
mechanism of formation of 1 involves an aldol
Scheme 1.
Table 1. Antioxidant properties, ESR data and lipophilicity of 1 and ascorbic acid
ꢁ
DPPH inhibition
OH inhibition
LDL peroxidation induced by
ESR data
g-Factor
Lipophilicity
Log Z
EC50
EC50
Cu2+
AAPH
a
H5
Log P
1
Ascorbic acid
3.9
3.9
16 mM
13 mM
2 mM10
30 mM
0.1 mM
2.5 mM
0.18 mM
—
2.00529
2.00518
1.32 G
1.72 G
+1.89
ꢀ1.85
26

P. Cotelle et al. / Bioorg. Med. Chem. 11 (2003) 1087–1093
1089
paramagnetic nitroxides such as DMPO–OH spin
adduct; (iii) ascorbic acid can reduce Fe(III) to Fe(II)
and in the presence of H O can stimulate OH forma-
increase in absorbance at 234 nm caused by the forma-
2
6
tion of conjugated dienes. Its activity was expressed as
percentages of increase of their lag phase compared to
the one of the control. As its antioxidant capability can
be characterised by the evolution of the LDL oxidation
curves (Fig. 1) with its concentration, 1 has been tested
at concentrations of 0.1–100 mM. We considered that
the drug concentration had 100% activity (efficiency
dose 100%: ED100) when it doubled the control lag
phase duration and we defined the efficiency dose 50%
ꢁ
2
2
tion by the Fenton reaction. Whereas an IC50 of 30 mM
1
0
has recently been reported for ascorbic acid, we were
unable to reproduce this result. The ESR signal of
DMPO–OH is superimposed with that of ascorbyl
radical. In the process of time, the ESR signal of
DMPO–OH decreased with the concomitant increase of
the ESR signal of ascorbyl radical.
(ED ) as the concentration of the drug that increased
this control lag phase by 1.5 times. Increased con-
centrations of 1 have been used with a logarithmic scale
50
The inhibitory effect of 1 on the Cu2+ or AAPH
induced oxidation of human LDL was measured by the
Figure 1. Effect of 1 on the lag phase of LDL oxidation by copper (II) and AAPH by increase of the diene absorption. The LDL solution (125 mg
26
protein/mL) was supplemented with 0, 0.2, 0.5, 1, 2 mM of 1 in 0.01 M PBS. The reference cuvette contained LDL and 1 without copper or AAPH.

1090
P. Cotelle et al. / Bioorg. Med. Chem. 11 (2003) 1087–1093
to calculate the ED50 of antioxidant activity. The ED50
values of 1 in the various peroxidation experiments are
reported in Table 1 and compared to that of ascorbic
The lipophilicities of 1 and ascorbic acid were evaluated
by the calculation of the log P values using the Villar
method implemented in Spartan packages. Whereas the
calculated (and the experimental) value for ascorbic acid
is ꢀ1.88 (ꢀ1.85) the calculated one for 1 is 1.89. A good
correlation between the calculated and the experimental
values of log P of ascorbic acid validates the use of the
Villar method. The calculated positive sign for the log P
value indicates that 1 is much more lipophilic than
ascorbic acid but its intrinsic moderate lipophilicity
suggests that it may be localised at the interface of the
bilayers (water and lipid phases).
2
7
acid. The ED of 1 on the induction of LDL oxida-
5
0
2
+
tion by Cu and AAPH are 0.10 and 0.18 mM, respec-
tively, whereas Esterbauer et al. reported for ascorbic
2
7
acid an ED50 of about 2.5 mM for the oxidation of LDL
2
+
induced by Cu . The similar ED50 of 1 whatever the
peroxidation inducer shows that the activity of 1 is
expected to be linked to a radical scavenging property.
Nevertheless other mechanisms, that is, copper com-
plexation or reduction, may not be rejected. Therefore
2
+
the interaction of 1 with Cu
spectroscopy. The UV spectrum of 1 (Fig. 2) was not
was measured by UV
The ESR spectra displayed in Figure 3 show the ascor-
byl radical and the radical of 1 obtained under mild
alkaline solution. On both spectra, a doublet can be
observed arising from the hyperfine coupling with the
H5 proton. The hyperfine splitting constant aH5 and the
g-value for both molecules are given in Table 1 and are
very similar for both molecules with a small decrease of
2
+
modified by adding Cu as well as after the addition of
EDTA, denoting an absence of interaction or reaction of 1
with Cu . As a control, the UV spectrum of 1 in alkaline
solution was also recorded (Fig. 2). The powerful activity
of 1 compared to ascorbic acid may be due to a higher
lipophilicity associated with its antioxidant properties.
2+
0.4 G for 1. No additional proton feature was detected
for 1.
It is well known that ascorbic acid can reduce para-
magnetic nitroxides into diamagnetic hydroxylamines:
the ESR signal of TEMPO totally disappears after the
addition of ascorbic acid whereas when 1 was added to
a 1 mM solution of TEMPO, we have not observed any
modification of ESR signal even for a high concentra-
tion (i.e., 80 mM). On the contrary, DPPH was reduced
either by ascorbic acid or by 1. These results suggest
that the reduction potential of 1 is expected to be lower
0
28
than that of TEMPO (E =880 mV versus NHE) and
therefore 1 is unable to reduce it and to be higher than
0
29
that of DPPH (E =560 mV versus NHE). The oxida-
tion of ascorbic acid at pH 5 has been reported to occur
at 902mV versus NHE with a modified carbon elec-
3
0
trode.
Theoretical calculations were employed first to under-
stand the most favourable structure of radical formed
under auto-oxidation conditions. Effectively the three
Figure 2. UV spectra of 1 in aqueous solution, after addition of Cu2+
or after addition of NaOH M/2.
ꢁ
structures that can be considered are the radical 1 , the
radical anion 1
ꢀ
ꢁ
2ꢀ
ꢁ
and the radical dianion 1 .
ꢀ
ꢁ
Figure 3. ESR spectra of ascorbyl radical (left) and 1 (right).

P. Cotelle et al. / Bioorg. Med. Chem. 11 (2003) 1087–1093
1091
ꢀ
ꢁ
Modelling these structures using semiempirical UHF/
PM3 formalism has shown that the most favourable
structure was the radical anion form with a heat of for-
mation of ꢀ184 kcal/mol whereas the two other ones
have respective heats of formation of ꢀ134 kcal/mol
and ꢀ167 kcal/mol for radical and radical dianion
forms. This first approach does not take account of the
solvent and can be used only to estimate electrostatic
potentials and to provide a reasonable spatial repre-
the radical anion 1 . Moreover E of ascorbyl radical
hf
(ꢀ278 kcal/mol) is much lower than those calculated for
ꢀ
ꢁ
1
of about 90 kcal/mol that can explain the lowest
stability of the radical formed. Measuring the steady
state concentrations of the two radicals under the same
experimental conditions may easily check this assertion.
As expected the double integration of this signal shows
that the concentration of ascorbyl radical is 10-fold
ꢀ
ꢁ
higher than in radical anion 1 .
ꢀ
ꢁ
sentation of the species. The structure of 1 is pre-
sented in Fig. 4. The electrostatic potential contours
show the most favourable hydrogen bond donation sites
for the radical anion. As expected the densest regions of
Conclusion
In this paper, we have prepared a true ascorbic acid
analogue that presents antioxidant activities very close
to that of ascorbic acid. A radical anion that gives very
similar data has also been generated under the same
conditions than for ascorbic acid but its stability is
lower and its redox potential lower than those of ascor-
byl radical. Due to its higher lipophilicity, 1 has proved
to be a good candidate in the prevention of lipid per-
oxidation and is expected to be a good relay between
ascorbic acid (in the water phase) and a-tocopherol (in
the lipid phase). We are currently attempting to prepare
the isomer of 1 where the 4-hydroxyphenyl moiety
replaces the 2-hydroxy group of the molecule in order to
evaluate the relative importance of the hydroxy groups
0
spin are located at the O3 atom and the O4 atom (see
Scheme 1 for numbering of 1) that conform to the con-
cept of ascorbic acid analogues. Deprotonation of the
phenolic group that is conjugated with the heterocycle
probably stabilises the radical because of an extended
electron delocalisation. Then, deprotonation of the
remaining and less acidic phenolic group does not fur-
ther enhance radical stability since it is not conjugated
with the heterocycle. The apparent in vacuo stabilisa-
tion probably reflects the repulsion between the two
negative charges, a phenomenon greatly overestimated
in the absence of solvent. In order to overcome this
problem, the solvent effect (water or methanol) was
3
1,32
33
computed using PCM models
levels using 6-31G** basis set. Calculated ꢀꢀG between
at the DFT B3LYP
in ascorbic acid.
ꢁ
ꢀ
ꢁ
2ꢀ
ꢁ
the radical 1 and the radical anion 1 and dianion 1
are respectively 48 and 624 kcal/mol higher whereas the
Experimental
Chemistry
ꢁ
solvation energies are ꢀ25 kcal/mol for the radical 1 ,
ꢀ
ꢁ
ꢀ
56.7 kcal/mol for the radical anion 1 and ꢀ147 kcal/
ꢀ
ꢁ
2
mol for the radical dianion 1 . These results indicate
that the phenolic group conjugated to the heterocycle
can be easily deprotonated and therefore the order of
All reagents and solvents were purchased from Aldrich-
Chimie (Saint-Quentin-Fallavier, France) of ACS
reagent grade and were used as provided. TLC analyses
were performed on a 3ꢂ10 cm plastic sheet precoated
with silica gel 60F254 (Merck) (Solvent system: ethyl
ꢁ
ꢀ
ꢁ
2ꢀ
ꢁ
stability is 1 ꢃ1 >1 . Additional isotropic Fermi
contact coupling term was computed and yielded the
hyperfine splitting constants a =10.46 G (1 ), 2.13 G
ꢁ
H5
ꢀ
ꢁ
2ꢀ
ꢁ
(1 ) and 2.45 G (1 ). From the calculated and
experimental hyperfine constants, we can conclude that
the radical observed during the auto-oxidation of 1 is
acetate/hexane 1:4). SiO , 200–400 mesh (Merck), was
2
used for column chromatography. Melting points were
obtained on a Reichert Thermopan melting point
apparatus equipped with a microscope and are uncor-
rected. NMR spectra were obtained on an AC 200
Bruker spectrometer in the appropriate solvent with
TMS as internal reference.
Synthesis of 1
A mixture of 2-(4-acetoxyphenylmethylene)-5-methyl-
oxazol-3-one (10 mmol) in 3 N HCl (60 mL) was
refluxed during 24 h. The resulting mixture was filtered
while it was still hot. The filtrate was allowed to cool to
room temperature and was extracted with EtOAc
(
(
5ꢂ20 mL). The combined filtrate was then dried
Na SO ) and was evaporated. The residue was purified
2
4
by column chromatography on silica gel (cyclohexane/
EtOAc 3:7) (R =0.71), yield=50%; mp=248–249 C
ꢄ
lit. 246–249 C); H NMR (200 MHz, Acetone-6d): 2.84
f
ꢄ
1H, dd, J=14.5 Hz; J=6.0 Hz, H6), 3.27 (1H, dd,
1
(
(
2
3
2
3
3
J=14.5 Hz; J=3.3 Hz, H6), 5.59 (1H, dd, J=3.3 Hz;
00
3
3
J=6.0 Hz, H5), 6.66 (2H, d, J=8.5 Hz, H3 ), 6.87
3
0
0
3
0
(2H, d, J=8.5 Hz, H2 ), 6.98 (2H, d, J=8.8 Hz, H2 ),
7.61 (2H, d, J=8.8 Hz, H3 ), 8.60 (1H, br s); C NMR
Figure 4. Structure of 1 ꢁ.
ꢀ
3
0
13

1092
P. Cotelle et al. / Bioorg. Med. Chem. 11 (2003) 1087–1093
(50 MHz, Acetone-6d): 39.4 (t), 79.3 (d), 115.6 (d), 116.5
(d), 127.0 (s), 129.3 (s), 130.1 (d), 131.6 (d), 137.4 (s),
Theoretical calculations
1
57.0 (s), 158.7 (s), 169.7 (s), 172.1 (s).
Geometry optimisation calculations were performed
with Spartan Pro 1.0 package software for semi-
empirical at the UHF/PM3 level for heat of formation
and log P calculations. Solvation calculation was per-
ESR spectroscopy
3
2
ESR measurements: ESR experiments were performed
on a Bruker Elexys 580 spectrometer operating at
formed using Gaussian 98.
9
0
.7 GHz with a 100 kHz modulation frequency and
.8 G of amplitude modulation. Microwave power
LDL preparation
detection was set to 5 mW. Samples were systematically
prepared as follows: 10 mL of sodium hydroxide solu-
tion (0.1 M) was added to 500 mL of a solution of the
compound in methanol (0.01 M).
Human LDL was isolated from freshly drawn blood
from healthy, normolipidaemic, and fasting volunteers.
Blood was collected into EDTA and the plasma was
separated by low speed centrifugation. LDL was iso-
lated by sequential density gradient ultracentrifugation
in sodium bromide density solutions in the density
range 1.019–1.063 g/mL as previously reported.2 Then,
LDL was dialysed against 0.01 M PBS (phosphate buf-
fered saline: 0.15 M NaCl, 0.01 M Na-phosphate, pH
7.4) containing 0.01% EDTA and sterilised by filtration
Hydroxyl radical inhibition
6
Produced by the Fenton reaction, the hydroxyl radical
was detected by the ESR spin-trapping method. The
hydroxyl radical reacts with 5,5-dimethyl-1-pyrroline-1-
oxide (DMPO) to give DMPO–OH, which is ESR
detectable. Reaction mixtures contained ferrous sulfate
ꢄ
through a 0.22 mm pore-size filter and stored at 4 C
before use. The protein concentration was determined
3
4
(
0.5 mM), H O (0.5 mM), DMPO (10 mL, 0.4 M) and
2
by Peterson’s method using bovine serum albumin as
standard. Total and free cholesterol, triglycerides
(Boehringer kit) and total phospholipids (bioMerieux
kit) were analysed by enzymatic methods.
2
various amounts of the tested molecule in 0.21 mL of
.1 M sodium phosphate buffer, pH 7.4. The reaction
was initiated by the addition of H O . ESR spectrum
0
2
2
was taken 2min after addition of H O .
2
2
Induction of LDL oxidation by copper or AAPH
DPPH inhibition
Ethanolic solutions of derivatives were prepared at
ꢀ
2
Spectrophotometric measurements were performed with
a Uvikon 932spectrophotometer. Calibration curves
for DPPH were performed in triplicate and the initial
DPPH concentration was then calculated from eq (1).
10 M in ethanol and diluted in 0.01 M PBS to give
final concentrations of 0.1–100 mM in a total ethanol
concentration of 1% (v/v). The same amounts of pure
ethanol were added in blanks.
Prior to oxidation, EDTA was removed by extensive
dialysis of LDL solution against 0.01 M PBS under N ,
5
A517 nm ¼ 0:109 Â CDPPH Â 10 þ 0:0013
ð1Þ
2
ꢄ
then oxidation was induced at 30 C by adding 100 mL
0
Standard solutions of the antioxidants were prepared in
ethanol and rapidly mixed (volumes from 0.02to
.4 mL) with an ethanol solution of DPPH taken
between 50 and 63 mM. The decrease in absorbance was
recorded every 6 s for 5 min. Seven measurements per
potential antioxidant were recorded with [antioxidant]/
of 16.6 mM CuSO or of 2mM 2,2 -azobis(2-amidino-
4
propane)dihydrochloride (AAPH) to 800 mL of LDL
(125 mg protein/mL) and 100 mL drugs in 0.01 M PBS.
Since the tested compounds absorb at 234 nm, the diene
formation has been studied with a control containing
LDL and the compound to be tested without copper or
AAPH. During copper- or AAPH-induced LDL oxida-
tion the diene conjugate formation was followed by
measuring the optical density (OD) at 234 nm every
10 min for 8 h with a thermostatted Kontron Uvikon
930 spectrophotometer equipped with a 10 positions
sample changer. Analyses were performed in triplicate.
0
[
DPPH] ratios varying from 0.19 to 1.7. In parallel a
blank solution of DPPH was screened to estimate
DPPH decomposition during the time of measurement.
The decrease in absorbance was then plotted against
time, after correction for DPPH decomposition. The
effective concentration ratio (EC ) is the concentration
5
0
of antioxidant producing a 50% decrease in DPPH at
steady state. The initial reaction followed second-order
kinetics; the rate constants k (obtained from plots of 1/
References and Notes
[
oxidant]/[DPPH]. Linear regression (r >0.9) gave the
C] versus time) were plotted against the ratio [anti-
2
ꢀ1
ꢀ1
1. Halliwell, B.; Chirico, S. Am. J. Clin. Nutr. 1993, 57, 715 S.
parameter Z (slope of the line in L mol
s ) which
2
3
4
1
5
. Loft, S.; Poulsen, H. E. J. Mol. Med. 1996, 74, 297.
. Stahl, W.; Sies, H. Diabetes 1997, 46, S14.
. Bowen, P. E.; Mobarhan, S. Am. J. Clin. Nutr. 1995, 62,
403S.
was examined as a quantification of radical scavenging
activity.
. Bendich, A. Handbook of Free Radicals and Antioxidants in
UV spectroscopy
Biomedicine, vol II; CRC Press Inc.: Boca Raton, FL, 1989; pp
153–160.
6. Stait, S. E.; Leake, D. S. Biochem. J. 1996, 320, 373.
UV spectra were recorded on a thermostated Kontron
Uvikon 932spectrometer.

P. Cotelle et al. / Bioorg. Med. Chem. 11 (2003) 1087–1093
1093
7
1
8
. Restsky, K. L.; Freeman, M. W.; Frei, B. J. Biol. Chem.
993, 268, 1304.
. (a) Weber, V.; Coudert, P.; Rubat, C.; Duroux, E.; Leal, F.;
23. Dupont, R.; Cotelle, P. Tetrahedron Lett. 2001, 42, 597.
24. Wong, H. N. C.; Xu, Z. L.; Chang, H. M.; Lee, C. M.
Synthesis 1992, 793.
25. Dupont, R.; Goossens, J. F.; Cotelle, N.; Vrielynck, L.;
´
Vezin, H.; Henichart, J. P.; Cotelle, P. Bioorg. Med. Chem.
2001, 9, 229.
Couquelet, J. J. Pharm. Pharmacol. 2000, 52, 523. (b) Weber,
V.; Coudert, P.; Rubat, C.; Duroux, E.; Vallee-Goyet, D.;
Gardette, D.; Bria, M.; Albuisson, E.; Leal, F.; Gramain, J. C.;
´
Couquelet, J.; Madesclaire, M. Bioorg. Med. Chem. 2002, 10,
26. Lebeau, J.; Furman, C.; Bernier, J. L.; Duriez, P.; Teissier,
E.; Cotelle, N. Free Radic. Biol. Med. 2000, 29, 900.
27. Esterbauer, H.; Striegl, G.; Puhl, H.; Rotheneder, M. Free
Radic. Res. Commun. 1989, 6, 67.
28. Rychnovsky, S. D.; Vaidyanathan, R.; Beauchamp, T.;
Lin, R.; Farmer, P. J. J. Org. Chem. 1999, 64, 6745.
29. Zhuang, Q. K.; Scholz, F.; Pragst, F. Electrochem. Com-
mun. 1999, 1, 406.
30. Raoof, J. B.; Ojani, R.; Kiani, A. J. Electroanal. Chem.
2001, 515, 45.
31. Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem.
Phys. Lett. 1996, 255, 327.
32. Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G.
E., Robb, M. A., Cheeseman, J. R., Zakrzewski, V. G., Mon-
tgomery, J. A., Jr., Stratmann, R. E., Burant, J. C., Dapprich,
S., Millam, J. M., Daniels, A. D., Kudin, K. N., Strain, M. C.,
Farkas, O., Tomasi, J., Barone, V., Cossi, M., Cammi, R.,
Mennucci, B., Pomelli, C., Adamo, C., Clifford, S., Ochterski,
J., Petersson, G. A., Ayala, P. Y., Cui, Q., Morokuma, K.,
Malick, D. K., Rabuck, A. D., Raghavachari, K., Foresman,
J. B., Cioslowski, J., Ortiz, J. V., Baboul, A. G., Stefanov, B.
B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Gom-
perts, R., Martin, R. L., Fox, D. J., Keith, T., Al-Laham, M.
A., Peng, C. Y., Nanayakkara, A., Challacombe, M., Gill, P.
M. W., Johnson, B., Chen, W., Wong, M. W., Andres, J. L.,
Gonzalez, C., Head-Gordon, M., Replogle, E. S., and Pople,
J. A. Gaussian 98, Revision A.9. Gaussian, Inc., Pittsburgh,
PA, 1998.
1
9
647.
. Manfredini, S.; Vertuani, S.; Manfredi, B.; Rossini, G.;
Calviello, G.; Palozza, P. Bioorg. Med. Chem. 2000, 8, 2791.
0. Mashino, T.; Takigawa, Y.; Saito, N.; Wong, L. Q.;
Mochizuki, M. Bioorg. Med. Chem. Lett. 2000, 10, 2783.
1. Raic-Malic, S.; Svedruzic, D.; Gazivoda, T.; Marunovic,
1
1
A.; Hergold-Brundic, A.; Nagl, A.; Balzarini, J.; DeClercq, E.;
Mintas, M. J. Med. Chem. 2000, 43, 4806.
1
3
1
1
1
2
1
1
1
2. Hopper, A. T.; Witiak, D. T. J. Org. Chem. 1995, 60,
334.
3. Dalla, V.; Cotelle, P. Tetrahedron Lett. 1998, 39, 8285.
4. Dalla, V.; Cotelle, P. Tetrahedron 1999, 55, 6923.
5. Cotelle, P.; Catteau, J. P. Tetrahedron Lett. 1997, 38,
969.
6. Dupont, R.; Cotelle, P. Synthesis 1999, 1651.
7. Dupont, R.; Cotelle, P. Tetrahedron Lett. 1998, 39, 8457.
8. (a) Kiriyama, N.; Nitta, K.; Sakagushi, Y.; Taguchi, Y.;
Yamamoto, Y. Chem. Pharm. Bull. 1977, 25, 2593. (b) Nitta,
K.; Fujita, N.; Yoshimura, T.; Arai, K.; Yamamoto, Y. Chem.
Pharm. Bull. 1983, 31, 1528.
1
9. Yun, B. S.; Kang, H. C.; Koshino, H.; Yu, S. H.; Yoo,
I. D. J. Nat. Prod. 2001, 64, 1230, and references cited therein.
0. We consider that a true ascorbic acid analogue must pos-
2
sess a structural feature able to stabilise a radical anion by
mesomery.
2
1. Cotelle, N.; Bernier, J. L.; Catteau, J. P.; Pommery, J.;
Wallet, J. C.; Gaydou, E. Free Radic. Biol. Med. 1996, 20, 35.
2. Capelle, S.; Planckaert, B.; Cotelle, P.; Catteau, J. P. J.
Chim. Phys. 1992, 89, 561.
2
33. O’Malley, P. J. J. Phys. Chem. B 2001, 105, 11290.
34. Peterson, G. A. Anal. Biochem. 1977, 83, 346.