Radical-induced oxidation of trans-resveratrol

Article abstract of DOI:10.1016/j.biochi.2011.11.005

trans-Resveratrol (RVT) (3,5,4′-trihydroxystilbene), a polyphenolic constituent of red wine, is thought to be beneficial in reducing the incidence of cardiovascular diseases, partly via its antioxidant properties. However, the mechanism of action by which trans-resveratrol displays its antioxidant effect has not been totally unravelled. This study aimed at establishing a comprehensive scheme of the reaction mechanisms of the direct scavenging of HO and O₂^- radicals generated by water gamma radiolysis. Aerated aqueous solutions of trans-RVT (from 10 to 100 μmol L^-1) were irradiated with increasing radiation doses (from 25 to 400 Gy) and further analyzed by UV-visible absorption spectrophotometry for detection of trans-RVT oxidation products. Separation and quantification of RVT and its four oxidation products previously identified by mass spectrometry, i.e., piceatannol (PCT), 3,5-dihydroxybenzoic acid (3,5-DHBA), 3,5-dihydroxybenzaldehyde (3,5-DHB) and para-hydroxybenzaldehyde (PHB), were performed by HPLC/UV-visible spectrophotometry. Determination of the radiolytic yields of trans-RVT consumption and oxidation product formation has allowed us to establish balance between trans-RVT disappearance and the sum of oxidation products formation. Under our conditions, O₂^- radicals seemed to poorly initiate oxidation of trans-RVT, whereas the latter, whatever its initial concentration, quantitatively reacted with HO radicals, via a dismutation mechanism. Two reaction pathways involving HO-induced trans-RVT primary radicals have been proposed to explain the formation of the oxidation end-products of trans-RVT.

Full text of DOI:10.1016/j.biochi.2011.11.005

Biochimie 94 (2012) 741e747
Contents lists available at SciVerse ScienceDirect
Biochimie
journal homepage: www.elsevier.com/locate/biochi
Research paper
Radical-induced oxidation of trans-resveratrol
Laurent Camont , Fabrice Collin , Martine Couturier , Patrice Thérond e, Daniel Jore c,
a
b
,
c
d
a,
c
a,f,
*
Monique Gardès-Albert , Dominique Bonnefont-Rousselot
EA 4466, Département de Biologie Expérimentale, Métabolique et Clinique, Faculté des Sciences Pharmaceutiques et Biologiques, Université Paris Descartes, Sorbonne Paris Cité,
avenue de l’Observatoire, 75006 Paris, France
a
4
b
IRD UMR 152, Université Paul Sabatier, 35 chemin des Maraîchers, 31400 Toulouse, France
c
UFR Biomédicale des Saints-Pères, Université Paris Descartes, Sorbonne Paris Cité, 45 rue des Saints-Pères, 75006 Paris, France
Inserm UMR-S939, Hôpital de la Pitié, Pavillon Benjamin Delessert, 83 boulevard de l’Hôpital, 75013 Paris, France
Service de Biochimie, Hôpital de Bicêtre, 74 rue du Général Leclerc, 94275 Le Kremlin-Bicêtre, France
d
e
f
Service de Biochimie Métabolique, Groupe Hospitalier Pitié-Salpêtrière-Charles Foix (AP-HP), 47-83 boulevard de l’Hôpital, 75013 Paris, France
a r t i c l e i n f o
a b s t r a c t
0
Article history:
trans-Resveratrol (RVT) (3,5,4 -trihydroxystilbene), a polyphenolic constituent of red wine, is thought to
be beneficial in reducing the incidence of cardiovascular diseases, partly via its antioxidant properties.
However, the mechanism of action by which trans-resveratrol displays its antioxidant effect has not been
totally unravelled. This study aimed at establishing a comprehensive scheme of the reaction mechanisms
Received 30 September 2011
Accepted 10 November 2011
Available online 20 November 2011
ꢀ
ꢀ
ꢁ
of the direct scavenging of HO and O
solutions of trans-RVT (from 10 to 100
2
radicals generated by water gamma radiolysis. Aerated aqueous
mol L ) were irradiated with increasing radiation doses (from 25
Keywords:
ꢁ1
m
Reactive oxygen species
Hydroxyl radical
Oxidation mechanism
Gamma radiolysis
to 400 Gy) and further analyzed by UVevisible absorption spectrophotometry for detection of trans-RVT
oxidation products. Separation and quantification of RVT and its four oxidation products previously
identified by mass spectrometry, i.e., piceatannol (PCT), 3,5-dihydroxybenzoic acid (3,5-DHBA), 3,5-
dihydroxybenzaldehyde (3,5-DHB) and para-hydroxybenzaldehyde (PHB), were performed by HPLC/UV
evisible spectrophotometry. Determination of the radiolytic yields of trans-RVT consumption and
oxidation product formation has allowed us to establish balance between trans-RVT disappearance and
the sum of oxidation products formation. Under our conditions, O
oxidation of trans-RVT, whereas the latter, whatever its initial concentration, quantitatively reacted
with HO radicals, via a dismutation mechanism. Two reaction pathways involving HO -induced trans-
RVT primary radicals have been proposed to explain the formation of the oxidation end-products of
trans-RVT.
ꢀ
2
ꢁ
radicals seemed to poorly initiate
ꢀ
ꢀ
Ó 2011 Elsevier Masson SAS. All rights reserved.
1
. Introduction
have atheroprotective properties [5,6]. It has been suggested that
the antioxidant properties of RVT are responsible for the protective
0
trans-Resveratrol (3,5,4 -trihydroxystilbene) (RVT, Fig. 1) is
a polyphenolic phytoalexin which is present in biological materials
such as grapes, grape juice, red wine, plums and peanuts [1,2]. This
compound has been received much interest for a long time because
of its extensive physiological properties. Indeed, several studies
have shown its beneficial effects on neurological, hepatic and
cardiovascular systems [3,4]. Particularly, RVT has been reported to
effect against cardiovascular diseases of consuming moderate
amounts of red wine. RVT seems especially able to scavenge
hydroxyl radicals produced by a Fenton reaction [7] but as for other
polyphenols one has to distinguish between a direct scavenging of
free radicals and a chelating effect towards transition metals
involved in Fenton reaction [8]. It has also been reported to be
a good antioxidant against lipid peroxidation in human blood cells
treated by platinum compounds [9], or peroxynitrite [10], in cell
membranes [11,12] and low density lipoproteins (LDL) [13e18].
In order to improve the knowledge of RVT antioxidant mode
of action, this work focused on its direct antioxidant properties
in vitro against oxygen-derived free radical species generated in
aqueous solution by gamma radiolysis. Gamma radiolysis of water
is a well-known method that has many advantages, such as the
*
Corresponding author. EA 4466, Département de Biologie Expérimentale,
Métabolique et Clinique, Faculté des Sciences Pharmaceutiques et Biologiques,
Université Paris Descartes, Sorbonne Paris Cité, 4 avenue de l’Observatoire, 75006
Paris, France. Tel.: þ33 1 42 16 20 58.
E-mail
address:
dominique.bonnefont-rousselot@parisdescartes.fr
(
D. Bonnefont-Rousselot).
0300-9084/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.biochi.2011.11.005

742
L. Camont et al. / Biochimie 94 (2012) 741e747
purchased from ABCys (Paris, France), and 3,5-DHBA, 3,5-DHB and
PHB from SigmaeAldrich (St-Louis, MO, USA).
2.2. Gamma radiolysis
Radiolysis corresponds to the chemical transformations of
a solvent due to the absorption of ionizing radiations, which allows,
within a few nanoseconds, the production of a homogeneous
solution of reactive oxygen species. In addition, this method allows
selective generation of specific radicals from the solvent, so that it is
possible to study their action towards the dissolved compounds.
Radiolytically generated radicals are independent of the nature and
of the concentration of the dissolved compound as long as its
ꢁ
2
ꢁ1
concentration remains lower than or equal to 10 mol L [21].
Gamma radiolysis was carried out by using an IBL 637 irradiator
(
1
37
CIS Biointernational, Gif-sur-Yvette, France) of Cs source, whose
activity was z 222 TBq (6000 Ci). In our experiments the dose rate
ꢁ1
was 10 Gy min . The dosimetry was determined by the Fricke
ꢁ
1
method [22], namely radio-oxidation of a 1 mmol L of iron(II)
ꢁ
1
3þ
sulphate solution in 0.4 mol L sulphuric acid, taking
l
max (Fe ) ¼
ꢁ
1
ꢁ1
3
04 nm,
3
(304 nm) ¼ 2204 L mol cm
and a radiolytic yield of
3
þ
ꢁ7
ꢁ1
ꢃ
G(Fe ) ¼ 16.2 ꢂ10 mol J
ranging from 25 to 400 Gy, were delivered to 1.5 mL of solution
depending on the time of the exposure to the -ray source: the
at 25 C. Different radiation doses,
g
longer the time, the higher the radiation dose. For each experi-
mental set, 1.5 mL of non-irradiated solution was taken as a control.
Prior to each set of experiments, glassware was carefully washed
Fig. 1. Chemical structure of RVT (a) and of its oxidation products generated by free
radicals attacks [19], i.e., 3,5-dihydroxybenzoic acid (b), p-hydroxybenzaldehyde (c),
with TFD4 soap (Franklab, France), rinsed with ultra-pure water
ꢃ
(
resistivity 18.2 M
any pollution by organic compounds.
Water radiolysis by -rays generates radical and molecular
U
) and finally heated at 400 C for 4 h to avoid
3,5-dihydroxybenzaldehyde (d) and piceatannol (e).
g
ꢁ
ꢀ
ꢀ
homogeneous production of known quantities of free radicals (such
species (i.e., e aq, HO , H , H
e dissolved molecular oxygen concentration around 10 mol L
e hydroxyl and superoxide radicals (this latter resulting from the
2 2 2
and H O ). Under aerated conditions
ꢀ
ꢁ
ꢀ
ꢁ4
ꢁ1
as superoxide anion O
2
or hydroxyl radical HO ), as well as the
possibility to selectively produce one specific radical to be studied
at a time. Oxygen-derived radicals thus generated have been used
to initiate one-electron oxidation of RVT dissolved in water. In
a previous work, we have identified by mass spectrometry the
oxidation end-products of RVT resulting from the direct attack of
ꢁ
ꢀ
scavenging of e aq and H species by O
2
) are simultaneously
produced with respective radiolytic yields (G-values expressed in
ꢁ
1
ꢁ7
ꢁ7
ꢁ1
mol J ) of 2.8 ꢂ 10 and 3.4 ꢂ10 mol J [21]. To selectively
ꢁ1
obtain superoxide anion, 0.1 mol L
sodium formate was
ꢀ
ꢀ
ꢁ
HO /O
2
,
i.e., 3,5-dihydroxybenzoic acid (3,5-DHBA), 3,5-
added into aqueous solutions in order to convert all radical
ꢁ
ꢀ
ꢀ
ꢀ
ꢁ
dihydroxybenzaldehyde (3,5-DHB), para-hydroxybenzaldehyde
species (e aq, HO , H ) into O
2
radicals with a final G-value of
0
0
ꢁ7
ꢁ1
(
PHB) and piceatannol (trans-3,5,3 ,4 -tetrahydroxystilbene, PCT)
6.2 ꢂ 10 mol J [21].
(
Fig. 1) [19]. However, only one radiation dose (400 Gy) and only
ꢁ1
one concentration of RVT (100
specify the HO /O
mmol L ) were studied. In order to
2.3. Analysis
ꢀ
ꢀ
ꢁ
2
-induced oxidation mechanism of RVT, the
present paper focuses on the effect of increasing radiation doses
Detection of the formation of RVT oxidation products was ach-
ieved by spectrophotometric measurements with an UVevisible
spectrophotometer (Beckman DU 800, Villepinte, France) at room
temperature, with a 0.2-cm or 1-cm pathlength quartz cell.
Samples were scanned from 190 to 500 nm. RVT in phosphate-
(
from 25 to 400 Gy) on aqueous RVT solutions at five concentra-
ꢁ1
tions (from 10 to 100 mmol L ), in order to establish a compre-
hensive reaction mechanism of RVT oxidation by HO /O
species.
ꢀ ꢀ
ꢁ
2
radical
ꢁ1
buffered (10 mmol L , pH 7) aqueous solutions had a maximum
ꢁ
1
ꢁ1
[20]) and
absorption at 304 nm
(
3
304 ¼ 30 135 L mol cm
2
. Materials and methods
BeereLambert law was applicable within the studied range of
ꢁ1
concentration (10e100 mmol L ).
2
.1. Chemicals and reagents
Separation of RVT from its oxidation end-products was per-
formed by high pressure liquid chromatography (HPLC) (P1000 XR,
Thermo Separation Products, Les Ulis, France) coupled to a UVevi-
sible spectrophotometer (UV 3000, Thermo Separation Products,
Les Ulis, France). Chromatographic separation was performed using
trans-Resveratrol (RVT, M ¼ 228 g molꢁ1) was purchased from
Cayman Chemical Company (Spi-Bio, Montigny-le-Bretonneux,
France). RVT solutions (10, 20, 30, 50, 70 and 100 mmol L ) were
prepared by sonication for 2 h in 10 mmol L phosphate buffer at
pH 7 in ultra-pure water (Maxima Ultra-Pure Water, ELGA, resis-
tivity 18.2 M
isomerization into cis-RVT. Before irradiation, they were system-
atically checked by UVevisible spectrophotometry (according to
a previously developed method [20]) to make sure that no isom-
erization into cis-RVT has occurred. Authentic piceatannol was
ꢁ1
ꢁ
1
a reverse-phase C18 analytical column (250 mm ꢂ 4.6 mm, 5
Kromasil, A.I.T, Houilles, France). Gradient elution was carried out
with 0.5% (v/v) acetic acid in H O (mobile phase A) and 0.5% (v/v)
acetic acid in acetonitrile (mobile phase B) at a flow rate of
mm,
U
). Aqueous solutions of RVT were kept in dark to avoid
2
ꢁ
1
1.0 mL min . The mobile phase was degassed ultrasonically before
use. The gradient was programmed with the following time course:
20% mobile phase B at 0 min, linear increase to 80% mobile phase B

L. Camont et al. / Biochimie 94 (2012) 741e747
743
at 30 min, 20% mobile phase B at 31 min and held 5 min. A volume
of 100
m
L of solution was injected into the column, maintained at
ꢃ
3
0 C. Concentrations of RVT and its oxidation end-products (PCT,
3
,5-DHBA, 3,5-DHB and PHB) were determined from the HPLC peak
area. Calibration curves were built with standard solutions of
ꢁ1
authentic PCT, 3,5-DHBA, 3,5-DHB and PHB (0e100 mmol L ,
ꢁ1
phosphate buffer 10 mmol L , pH 7) by detection at 304 nm. Linear
regression was obtained with coefficient of determination
2
2
r > 0.99: y ¼ 153 793 ꢂ x with r ¼ 0.996 for RVT, y ¼ 92 800 ꢂ x
2
2
with r ¼ 0.996 for PCT, y ¼ 12 414 ꢂ x with r ¼ 0.992 for 3,5-DHBA,
2
y ¼ 11 400 ꢂ x with r ¼ 0.993 for 3,5-DHB and y ¼ 19 592 ꢂ x with
2
r ¼ 0.993 for PHB. All experiments were performed in triplicate,
and experimental data are expressed as means ꢄ standard devia-
tion (SD).
3
. Results
ꢀ
ꢀ
ꢁ
towards RVT, the
Before studying the direct attack of HO /O
scavenging capacities of phosphate buffer (PB) and radiolytically
2
ꢁ7
ꢁ1
generated H
assessed. At pH 7, H
2
O
2
(G ¼ 0.7 ꢂ 10 mol J ) towards free radicals were
ꢁ
2ꢁ
2
PO
4
and HPO
s
4
might react with the hydroxyl
6
ꢁ1 ꢁ1
radicals (k z 10 L mol
2 2
[23]), and H O might also be oxidized
ꢀ
7
ꢁ1 ꢁ1
[24]) thus resulting in misin-
by HO radicals (k z 10 L mol
s
ꢁ
1
terpretation of the experimental results. Aqueous 10
mmol L
solutions of RVT in the presence of absence of either 0.1, 1 and
ꢁ
1
ꢁ1
1
0 mmol L phosphate buffer, or of 0.05, 0.5 and 5 mmol L
were irradiated and analyzed by absorption spectrophotometry at
04 nm. The oxidation of RVT was not found to be influenced by the
presence of phosphate buffer or H (Supplementary material,
Fig. s1), even at the highest concentration of the latter compounds.
2 2
H O ,
3
2 2
O
ꢁ
1
Phosphate buffered (10 mmol L , pH 7) aqueous solutions of
ꢁ
1
RVT 100
mmol L
were irradiated at increasing radiation doses
ꢀ
ꢀ
2
ꢁ
Fig. 2. Spectral characterization of RVT oxidation by HO /O
. Absorption spectra of
from 25 to 400 Gy (þcontrol, 0 Gy) and absorbance was measured
ꢁ
1
ꢁ1
phosphate buffered (10 mmol L , pH 7) aerated aqueous solution of 100
mmol L RVT
between 190 and 490 nm. As RVT was oxidized, the absorption at
irradiated at 25, 50, 75, 100, 150, 200, 300 and 400 Gy. (a) Absolute absorption spectra
ꢁ
1
3
04 nm decreased (Fig. 2a), thus illustrating the oxidation of RVT. At
(reference: phosphate buffer, 10 mmol L , pH 7), (b) differential absorption spectra
(reference: non-irradiated RVT solutions). Optical pathlength: l ¼ 1 cm, dose rate:
I ¼ 10 Gy minꢁ . The arrows indicate the evolution of the absorption bands.
the same time, two new spectral bands at 252 and 360 nm
increased with the radiation dose, and can be attributed to the
oxidation products. Among them, the former one was previously
attributed to the isomerization of trans-RVT into cis-RVT
1
,
,
ꢁ7
ꢁ1
because G(ꢁRVT)¼½GHO ꢂ (GHO ¼ 2.8 ꢂ 10 mol J [21]), radio-
induced oxidation of RVT appears as proceeding via a dismutation
reaction (equation (1)).
ꢁ
1
ꢁ1
(
3
cis(252 nm) ¼ 6557 L mol cm [20]). These results are confirmed
by the differential absorption spectra (Fig. 2b).
The quantification of the disappearance of RVT and the forma-
tion of the four oxidation products was conducted by HPLC analysis.
$
HO
ꢁ
1
2RVT / primary radicals//RVT þ oxidation products
(1)
A chromatogram of RVT 50
m
mol L before and after irradiation
400 Gy) is presented in Fig. 3. Only RVT was detected in the control
¼ 13.9 min), whereas at 400 Gy, 3,5-DHBA, 3,5-DHB, PHB and
¼ 3.9 min, ¼ 6.6 min,
(
(
t
r
PCT were respectively detected at
t
r
t
r
t
r
¼ 7.9 min and 12.1 min by comparison with previous mass
spectrometric analyses and with retention times of standard solu-
tions [19]. The concentration of RVT and its oxidation end-products
was measured and reported as a function of the radiation dose for
ꢁ
1
an initial concentration of 50 mmol L RVT (Fig. 4). The concen-
tration of RVT decreased, thus confirming the oxidation of RVT as
a function of the radiation dose (Fig. 4a), whereas the concentration
of 3,5-DHBA, 3,5-DHB, PHB and PCT increased with the radiation
dose, indicating the simultaneous formation of these four oxidation
products (Fig. 4b). The same result was obtained whatever the
ꢁ1
initial concentration of RVT (10, 20, 30, 70 and 100
mmol L
)
(Supplementary material, Fig. s2). The radiolytic yields (G) of RVT
consumption and oxidation product formation e determined as the
slope of the initial tangent of each curve e are reported in Table 1.
Whatever its initial concentration, the radiolytic yield of RVT
consumption remains constant, with an average value of
Fig. 3. HPLC chromatograms of aerated aqueous solution of resveratrol (a) non-
ꢀ
ꢀ
ꢁ
irradiated and (b) irradiated at 400 Gy (action of HO /O
2
radicals), with UV detec-
ꢁ
1
tion at 304 nm; 50
m
mol L
initial RVT concentration, phosphate buffered
ꢁ1
ꢁ1
(10 mmol L
)
aqueous aerated solution at pH
7
(dose rate I ¼ 10 Gy min ).
ꢁ7
ꢁ1
ꢀ
G
(ꢁRVT) ¼ (1.43 ꢄ 0.08) ꢂ 10 mol J . All hydroxyl radicals (HO )
RVT ¼ trans-resveratrol; 3,5-DHBA ¼ 3,5-dihydroxybenzoic acid; 3,5-DHB ¼ 3,5-
generated during gamma radiolysis are scavenged by RVT, and
dihydroxybenzaldehyde; PHB ¼ para-hydroxybenzaldehyde; PCT ¼ piceatannol.

744
L. Camont et al. / Biochimie 94 (2012) 741e747
Fig. 5. Remaining non-oxidized RVT as a function of the radiation dose (dose rate
ꢁ
1
ꢁ1
I ¼ 10 Gy min ). Aqueous aerated solutions (phosphate buffered, 10 mmol L , pH 7)
ꢁ
1
ꢀ
ꢀ
2
ꢁ
of RVT 50
sodium formate 0.1 mol L (specific production of O
5, 100, 150, 200, 300 and 400 Gy (þcontrol, 0 Gy). Results expressed as mean ꢄ SD
n ¼ 3).
m
mol L without (simultaneous production of HO and O
radicals) or with
ꢁ
2
radicals), irradiated at 25, 50,
ꢁ
1
ꢀ
7
(
(
0.1 mol Lꢁ1). Under these experimental conditions, Oꢀ
the only radical species produced by water radiolysis with
2
ꢁ
radical is
ꢁ7
ꢁ1
a formation yield of 6.2 ꢂ 10 mol J (see Materials and methods).
The concentration of remaining non-oxidized RVT has been re-
ported as a function of the radiation dose for an aqueous solution of
ꢁ
1
ꢀ
ꢀ
ꢁ ꢁ
2 2
mmol L , oxidized either by HO /O or by O radicals alone
ꢀ
ꢀ
RVT 50
Fig. 5). The consumption of RVT is much less important when O
is the only species able to initiate RVT oxidation. At 300 Gy, 72%
ꢁ
2
(
ꢀ
ꢀ
ꢁ
ꢀ
ꢁ
of RVT was consumed by HO /O
2
vs. a few 10% by O alone, what
would represent about 5% in the conditions were HO and O
2
ꢀ
ꢀ
ꢁ
2
are
Fig. 4. Consumption ofꢀ resveratrol and production of radical-induced oxidation
ꢀꢁ
ꢁ7
ꢁ7
ꢁ1
produced together ðGðO
Þ
ꢀ
ꢀꢁ ¼ 3:4 ꢂ 10 mol J
and
ꢀ
2
ꢁ
2
HO =O
2
products (action of HO /O
radicals). Remaining non-oxidized resveratrol (a) and
ꢀꢁ
ꢁ1
GðO
Þ
ꢀꢁ
¼ 6:2 ꢂ 10 mol J , thus 10% ꢂ (3.4/6.2) ¼ 5.5%).
2
O
alone
generated 3,5-DHBA, PCT, PHB and 3,5-DHB (b) as a function of the radiation dose
2
ꢁ
1
ꢁ1
However, the radiolytic yield of RVT consumption e the slope of the
(
dose rate I ¼ 10 Gy min ); 50
m
mol L
aqueous aerated solutions of resveratrol,
ꢀ
ꢁ
2
is
ꢁ
1
phosphate buffered (10 mmol L ), pH 7, irradiated at 25, 50, 75, 100, 150, 200, 300 and
initial tangent to the curve e remains close to zero when O
specifically generated. Under our experimental conditions, the
400 Gy (þcontrol, 0 Gy). Results are expressed as mean ꢄ SD (n ¼ 3).
ꢀ
ꢁ
effect of O
initiation of RVT oxidation, thus implying that superoxide radicals
would mainly dismutate in such conditions
at pH 7, [28]). This
2
radicals can be considered as negligible regarding the
Among the oxidation products generated, 3,5-DHBA exhibits the
ꢀ
ꢁ
ꢀ
2
5
ꢁ1 ꢁ1
s
highest radiolytic yield of formation and has been previously
identified as an in vivo metabolite of RVT resulting from a biocon-
version by cytochrome P450 [25,26]. This result is supported by the
study of Makris and Rossiter [27], showing that the major products
from quercetin radical-induced oxidation was protocatechuic acid
(O
2
þ HO
/ H O
2 2
þ O
2
, k ¼ 6 ꢂ 10 L mol
result is in good accordance with Bader et al. [29] that showed that
only hydroxyl radicals were responsible for RVT oxidation in similar
experimental conditions.
(
3,4-dihydroxybenzoic acid) resulting from an oxidative fragmen-
4. Discussion
tation of quercetin. Two other minor products, 3,5-DHB and PHB,
were detected and accounted for 8% and 7% respectively (Table 1).
According to our experimental results, neither phosphate buffer
anions nor hydrogen peroxide compete with RVT towards the
ꢀ
ꢁ
Equation (1) implies that the effect of O
2
radicals on the initi-
ꢀ
ꢀ
ꢁ
ation of RVT oxidation is supposed to be very weak. To check for
this, RVT aqueous solutions at concentration ranging from 10 to
action of HO /O
2
radicals. Moreover, it seems that superoxide
radicals are poorly able to initiate RVT oxidation, which is in
agreement with a study of Jia et al. showing by EPR that the direct
ꢁ
1
1
00 mmol L were irradiated in the presence of sodium formate
Table 1
Radiolytic yields G (10 mol J , mean (standard deviation)) of disappearance of RVT submitted to the action HO /O
ꢁ7
ꢁ1
ꢀ
ꢀ
ꢁ
1
2
ꢁ
radicals, and radiolytic yields of production of the
oxidation products (along with their sum, SPox). Initial concentrations of RVT were 10, 20, 30, 50, 70 and 100
m
mol L . Radiolytic yields are given by the slopes of initial
P
tangents of the curves in Fig. 4 and Fig. s2. Average (av.) values (standard deviation) of radiolytic yields of RVT consumption and oxidation product formation (
indicated. RVT ¼ trans-resveratrol; 3,5-DHBA ¼ 3,5-dihydroxybenzoic acid; 3,5-DHB ¼ 3,5-dihydroxybenzaldehyde; PHB ¼ para-hydroxybenzaldehyde; PCT ¼ piceatannol.
Pox) are
Initial RVT concentration
Av.
10
20
30
50
70
100
RVT
1.39 (0.07)
0.75 (0.08)
0.10 (0.01)
0.09 (0.01)
0.26 (0.03)
1.2 (0.13)
1.32 (0.02)
0.92 (0.09)
0.12 (0.01)
0.09 (0.01)
0.30 (0.03)
1.43 (0.14)
1.55 (0.03)
0.86 (0.09)
0.11 (0.01)
0.09 (0.01)
0.29 (0.03)
1.35 (0.14)
1.42 (0.02)
0.93 (0.09)
0.13 (0.01)
0.12 (0.01)
0.30 (0.01)
1.48 (0.12)
1.50 (0.02)
1.00 (0.09)
0.13 (0.01)
0.10 (0.01)
0.31 (0.03)
1.54 (0.14)
1.41 (0.02)
0.96 (0.09)
0.13 (0.01)
0.11 (0.01)
0.32 (0.03)
1.52 (0.14)
1.43 (0.08)
0.90 (0.09)
0.12 (0.01)
0.10 (0.01)
0.30 (0.02)
1.42 (0.13)
3
3
,5-DHBA
,5-DHB
PHB
PCT
P
P
ox

L. Camont et al. / Biochimie 94 (2012) 741e747
745
Scheme 1. Proposed mechanism of primary RVT radical generation.
attack of superoxide radicals does not induce RVT oxidation for
two main pathways. First, because of the strong stabilization of
the radical, RVT is assumed not to react with molecular oxygen
ꢁ
1
ꢀ
a concentration range from 10 to 100
mmol L
[30]. Thus, the
hydroxyl radical is the only species initiating RVT oxidation.
A reaction mechanism can be proposed for the generation of the
primary radical, resulting from a single attack of one hydroxyl
radical towards one RVT molecule (Scheme 1). The first radical e
(addition) but with other radical transient species (biradicalꢀ
reactions). Among the two other primary radicals, RVT(HO)
would undergo the rapid addition of molecular oxygen (Scheme
2), to form the peroxyl radical (A). The condensation of two of
them leads to the tetroxide dimer (B) that could lose one
molecular oxygen to form the alkoxyl radical (C). A biradical
ꢀ
noted RVT e is generated by the abstraction of hydrogen located on
0
the hydroxyl group in position 4 of the phenol moiety of RVT. Bond
ꢀ
dissociation energy (BDE) is known to be small for phenol and, in
addition, the radical thus formed is close to a semiquinone reso-
nance structure with the most extensive resonance system
reaction between this latter and RVT , proceeding by the hydrogen
abstraction on the aliphatic hydroxyl group, finally leads to 3,5-
dihydroxybenzaldehyde
(3,5-DHB)
and
para-hydrox-
(
[
comparing to 3- and 5-radical, on the resorcinol moiety of RVT)
31]. The formation of this radical has been previously observed by
ybenzaldehyde (PHB), along with the regeneration of one mole-
cule of RVT. The simultaneous formation of 3,5-DHB and PHB by
a single reaction pathway is consistent with our experimental
results because the formation of 3,5-DHB and PHB represents 7-
8% of all the oxidation products, for each (Table 1). Thus,
Stojanovic et al. [32] by pulse radiolysis experiments of a system
involving RVT and trichloromethylperoxyl radicals. It has also been
proposed as one of the first step of primary radical formation
during reaction between RVT and galvinoxyl radicals [33]. The
second main pathway for primary radical generation corresponds
ꢀ
implicitly, the formation of RVT(HO) would appear as a non-
preferential way in the radical-induced oxidation of RVT.
ꢀ
ꢀ
to the addition of HO , which could occur itself via two ways: first,
HO radical could add to the phenol moiety of RVT, in position 3
The second way of the primary radical evolution starts with PCT
ꢀ
0
ꢀ
(Scheme 3). A biradical reaction with RVT leads to the formation of
(
ortho) because of the
p
-donor effect of the hydroxyl group in
piceatannol by re-aromatization and regenerates one molecule of
resveratrol. This first reaction competes with the rapid addition of
molecular oxygen, to form the peroxyl radical (D), followed by
a cyclization and a second addition of molecular oxygen to finally
lead to the endoperoxoperoxyl radical (E). Re-aromatization could
0
ꢀ
position 4 [34], leading to a piceatannol-like radical PCT ; second,
hydroxyl radical could add to the aliphatic double bond of RVT,
leading to RVT(HO) for which the carbon-centred radical is stabi-
ꢀ
0
lized by the
p-donor effect of the hydroxyl group in position 4 ,
0
extending the resonance system to the whole phenol moiety. The
latter radical has been previously proposed in similar experimental
conditions [29].
Starting with these three primary radicals, the formation of
the oxidation end-products of RVT would proceed according to
then proceed by hydrogen abstraction in position 3 and rear-
rangement, producing the cyclic non-radical di-endoperoxide (F).
The latter could chemically evolve to finally lead to 3,5-
dihydroxybenzoic acid (3,5-DHBA) and 3,4-dihydroxybenzoic acid
(3,4-DHBA, or protocatechuic acid). The two latter isomers were
ꢀ
Scheme 2. Proposed mechanism of RVT(HO) radical secondary reaction, leading to oxidation end-products.

746
L. Camont et al. / Biochimie 94 (2012) 741e747
ꢀ
Scheme 3. Proposed mechanism of PCT radical secondary reaction, leading to oxidation end-products.
probably co-eluted in our experimental conditions e i.e. protocol
based on reversed-phase chromatographic separation e that is
the reason why only one peak was detected (Fig. 3). The way of PCT
evolution would appear as the main reaction way of RVT oxidation
since the three products formed e i.e., PCT, 3,4-DHBA and 3,5-DHBA
References
ꢀ
[1] B. Fauconneau, P. Waffo-Teguo, F. Huguet, L. Barrier, A. Decendit, J.M. Merillon,
Comparative study of radical scavenger and antioxidant properties of phenolic
compounds from Vitis vinifera cell cultures using in vitro tests, Life Sci. 61
(1997) 2103e2110.
[
[
[
[
[
2] J.A. Baur, D.A. Sinclair, Therapeutic potential of resveratrol: the in vivo
evidence, Nat. Rev. Drug Discov. 5 (2006) 493e506.
3] S. Renaud, M. de Lorgeril, Wine, alcohol, platelets, and the French paradox for
coronary heart disease, Lancet 339 (1992) 1523e1526.
e account for 85% of the total amount of the oxidation products
ꢀ
(
Table 1). Moreover, the addition of molecular oxygen on PCT is
ꢀ
ꢀ
likely faster than the biradical reaction between PCT and RVT since
4] S. Pervaiz, Resveratrol: from grapevines to mammalian biology, FASEB J. 17
3
,4- and 3,5-DHBA are preferentially formed.
(2003) 1975e1985.
5] S. Das, D.K. Das, Resveratrol:
a
therapeutic promise for cardiovascular
diseases, Recent Pat. Cardiovasc. Drug Discov. 2 (2007) 133e138.
5
. Conclusion
6] G.D. Norata, P. Marchesi, S. Passamonti, A. Pirillo, F. Violi, A.L. Catapano, Anti-
inflammatory and anti-atherogenic effects of cathechin, caffeic acid and trans-
resveratrol in apolipoprotein E deficient mice, Atherosclerosis 191 (2007)
The aim of this study was to propose a comprehensive scheme
ꢀ
ꢀ
ꢁ
265e271.
of the reaction mechanisms of the direct scavenging of HO and O
2
[
7] H.J. Kim, E.J. Chang, S.H. Cho, S.K. Chung, H.D. Park, S.W. Choi, Antioxidative
activity of resveratrol and its derivatives isolated from seeds of Paeonia lac-
tiflora, Biosci. Biotechnol. Biochem. 66 (2002) 1990e1993.
radicals generated by water gamma radiolysis, thanks to a quanti-
tative analysis (determination of the radiolytic yields) of the
disappearance of trans-RVT and of the concomitant formation of
trans-RVT-derived oxidation products previously identified by mass
[
8] S. Ozgova, J. Hermanek, I. Gut, Different antioxidant effects of polyphenols on
lipid peroxidation and hydroxyl radicals in the NADPH-, Fe-ascorbate- and Fe-
microsomal systems, Biochem. Pharmacol. 66 (2003) 1127e1137.
ꢀ
ꢁ
spectrometry. Whereas O
2
radicals seemed to poorly initiate
[9] B. Olas, B. Wachowicz, I. Majsterek, J. Blasiak, A. Stochmal, W. Oleszek, Anti-
ꢀ
0
0
0
oxidation of trans-RVT, HO radicals quantitatively reacted with
trans-RVT, via a dismutation mechanism. Two reaction pathways
involving HO -induced trans-RVT primary radicals have been
oxidant properties of trans-3,3 ,5,5 -tetrahydroxy-4 -methoxystilbene against
modification of variety of biomolecules in human blood cells treated with
platinum compounds, Nutrition 22 (2006) 1202e1209.
ꢀ
[
10] B. Olas, P. Nowak, J. Kolodziejczyk, M. Ponczek, B. Wachowicz, Protective
effects of resveratrol against oxidative/nitrative modifications of plasma
proteins and lipids exposed to peroxynitrite, J. Nutr. Biochem. 17 (2006)
proposed to explain the formation of the oxidation end-products of
trans-RVT. We have shown that piceatannol (PCT) and 3,5-
dihydroxybenzoic acid (3,5-DHBA) accounted for w85% of the
oxidized trans-RVT, whereas 3,5-dihydroxybenzaldehyde (3,5-
DHB) and para-hydroxybenzaldehyde (PHB) accounted for w15%.
These results help understand the direct radical scavenging of
trans-RVT, property involved in the beneficial antioxidant effect of
this natural polyphenol.
96e102.
[
[
11] B. Tadolini, C. Juliano, L. Piu, F. Franconi, L. Cabrini, Resveratrol inhibition of
lipid peroxidation, Free Radic. Res. 33 (2000) 105e114.
12] Y.J. Cai, J.G. Fang, L.P. Ma, L. Yang, Z.L. Liu, Inhibition of free radical-induced
peroxidation of rat liver microsomes by resveratrol and its analogues, Bio-
chim. Biophys. Acta 1637 (2003) 31e38.
[13] L. Belguendouz, L. Fremont, A. Linard, Resveratrol inhibits metal ion-
dependent and independent peroxidation of porcine low-density lipopro-
teins, Biochem. Pharmacol. 53 (1997) 1347e1355.
[
14] L. Belguendouz, L. Fremont, M.T. Gozzelino, Interaction of transresveratrol
with plasma lipoproteins, Biochem. Pharmacol. 55 (1998) 811e816.
15] L. Fremont, L. Belguendouz, S. Delpal, Antioxidant activity of resveratrol and
alcohol-free wine polyphenols related to LDL oxidation and polyunsaturated
fatty acids, Life Sci. 64 (1999) 2511e2521.
Acknowledgements
[
The authors are indebted to Dr. Averbeck for the use of the
gamma irradiator facility (Institut Curie, Paris, France). Laurent
Camont was the recipient of a fellowship from the Ministère de
l’Enseignement Supérieur et de la Recherche.
[
16] P. Brito, L.M. Almeida, T.C. Dinis, The interaction of resveratrol with ferryl-
myoglobin and peroxynitrite; protection against LDL oxidation, Free Radic.
Res. 36 (2002) 621e631.
[17] D. Pietraforte, L. Turco, E. Azzini, M. Minetti, On-line EPR study of free radicals
induced by peroxidase/H(2)O(2) in human low-density lipoprotein, Biochim.
Biophys. Acta 1583 (2002) 176e184.
Appendix. Supplementary data
[18] H. Berrougui, G. Grenier, S. Loued, G. Drouin, A. Khalil, A new insight into
resveratrol as an atheroprotective compound: inhibition of lipid peroxidation
and enhancement of cholesterol efflux, Atherosclerosis 207 (2009) 420e427.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.biochi.2011.11.005
[19] L. Camont, F. Collin, C. Marchetti, D. Jore, M. Gardes-Albert, D. Bonnefont-
Rousselot, Liquid chromatographic/electrospray ionization mass spectrometric

L. Camont et al. / Biochimie 94 (2012) 741e747
747
identification of the oxidation end-products of trans-resveratrol in aqueous
solutions, Rapid Commun. Mass Spectrom. 24 (2010) 634e642.
resveratrol in human liver microsomes, Biochem. Pharmacol. 68 (2004)
773e782.
[
20] L. Camont, C.H. Cottart, Y. Rhayem, V. Nivet-Antoine, R. Djelidi, F. Collin,
J. Beaudeux, D. Bonnefont-Rousselot, Simple spectrophotometric assessment
of the trans-/cis-resveratrol ratio in aqueous solutions, Anal. Chim. Acta 634
[27] D.P. Makris, J.T. Rossiter, Heat-induced, metal-catalyzed oxidative degradation
of quercetin and rutin (quercetin 3-O-rhamnosylglucoside) in aqueous model
systems, J. Agric. Food Chem. 48 (2000) 3830e3838.
[28] B.H.J. Bielski, D.E. Cabelli, R.L. Arudi, A.B. Ross, Reactivity of HO
ꢁ
(
2009) 121e128.
2
/O
2
radicals in
[
[
21] J.W.T. Spinks, R.J. Woods, Water and inorganic aqueous systems, in: Intro-
duction to Radiation Chemistry. John Wiley & Sons, New York, 1990, pp.
aqueous solution, J. Phys. Chem. Ref Data 14 (1985) 1041e1100.
[29] Y. Bader, R.M. Quint, N. Getoff, Resveratrol products resulting by free radical
attack, Radiat. Phys. Chem. 77 (2008) 708e712.
[30] Z. Jia, H. Zhu, B.R. Misra, J.E. Mahaney, Y. Li, H.P. Misra, EPR studies on the
superoxide-scavenging capacity of the nutraceutical resveratrol, Mol. Cell.
Biochem. 313 (2008) 187e194.
[31] S. Xu, G. Wang, H.-M. Liu, L.-J. Wang, H.-F. Wang, A DMol3 study on the
reaction between trans-resveratrol and hydroperoxyl radical: dissimilarity of
antioxidant activity among OeH groups of trans-resveratrol, J. Mol. Struct.:
Theochem 809 (2007) 79e85.
[32] S. Stojanovic, H. Sprinz, O. Brede, Efficiency and mechanism of the antioxidant
action of trans-resveratrol and its analogues in the radical liposome oxidation,
Arch. Biochem. Biophys. 391 (2001) 79e89.
[33] Y.J. Shang, Y.P. Qian, X.D. Liu, F. Dai, X.L. Shang, W.Q. Jia, Q. Liu, J.G. Fang,
B. Zhou, Radical-scavenging activity and mechanism of resveratrol-oriented
analogues: influence of the solvent, radical, and substitution, J. Org. Chem.
74 (2009) 5025e5031.
243e313.
22] H. Fricke, S. Morse, The chemical action of Roentgen rays on dilute ferrosul-
fonate solutions as measure of dose, Am. J. Roentgenol. Radium Ther. 18
(
1927) 430e432.
23] P. Wardman, L.P. Candeias, Fenton chemistry: an introduction, Radiat. Res.
45 (1996) 523e531.
[
[
1
24] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical-review of rate
constants for reactions of hydrated electrons, hydrogen-atoms and hydroxyl
ꢀ
ꢀ
ꢁ
radicals ( OH/ O ) in aqueous-solution, J. Phys. Chem. Ref. Data 17 (1988)
13e886.
5
[
25] G.A. Potter, L.H. Patterson, E. Wanogho, P.J. Perry, P.C. Butler, T. Ijaz,
K.C. Ruparelia, J.H. Lamb, P.B. Farmer, L.A. Stanley, M.D. Burke, The cancer
preventative agent resveratrol is converted to the anticancer agent picea-
tannol by the cytochrome P450 enzyme CYP1B1, Br. J. Cancer 86 (2002)
774e778.
[26] B. Piver, M. Fer, X. Vitrac, J.M. Merillon, Y. Dreano, F. Berthou, D. Lucas,
Involvement of cytochrome P450 1A2 in the biotransformation of trans-
[34] J. Fossey, D. Lefort, J. Sorba, Free Radicals in Organic Chemistry. Masson, Paris,
1995.