Radiolysis of quercetin in methanol solution: Observation of depside formation

Article abstract of DOI:10.1021/jf020165m

Radiolysis of the flavonol quercetin, a natural antioxidant, was performed in methanol. The degradation process was followed by HPLC analyses. The major product was identified as a depside (Q1) by NMR and LC-MS. The G(Q1) radiolytic factor was plotted versus the initial concentration of quercetin. This radiolytic process was attributed to the CH₃O^· radicals presented in the irradiated medium. The proposed mechanism invoked a stereospecific oxidation of the 3-hydroxyl group of quercetin which led to C-ring opening and to the formation of the depside Q1. In presence of water, Q1 was transformed into another depside, Q2, by an inverse esterification reaction. A chemical equilibrium was observed between Q1 and Q2. The comprehension of the radiolytic process of quercetin in methanol solution is of importance. Indeed, the same type of oxidative reactions could occur on flavonoids during preservation of food by ionizing radiation.

Full text of DOI:10.1021/jf020165m

J. Agric. Food Chem. 2002, 50, 4827−4833 4827
Radiolysis of Quercetin in Methanol Solution: Observation of
Depside Formation
ABDELGHAFOUR MARFAK,† PATRICK TROUILLAS,† DAOVY-PAULETTE ALLAIS,‡
YVES CHAMPAVIER,§ CLAUDE-ALAIN CALLISTE,† AND JEAN-LUC DUROUX*,†
UPRES EA 1085, Biomole´cules et Cibles Cellulaires Tumorales, Faculte´ de Pharmacie,
University of Limoges, 2 rue du Dr. Marcland, 87025 Limoges Cedex, France
Radiolysis of the flavonol quercetin, a natural antioxidant, was performed in methanol. The degradation
process was followed by HPLC analyses. The major product was identified as a depside (Q1) by
NMR and LC-MS. The G(Q1) radiolytic factor was plotted versus the initial concentration of quercetin.
This radiolytic process was attributed to the CH3O• radicals presented in the irradiated medium. The
proposed mechanism invoked a stereospecific oxidation of the 3-hydroxyl group of quercetin which
led to C-ring opening and to the formation of the depside Q1. In presence of water, Q1 was transformed
into another depside, Q2, by an inverse esterification reaction. A chemical equilibrium was observed
between Q1 and Q2. The comprehension of the radiolytic process of quercetin in methanol solution
is of importance. Indeed, the same type of oxidative reactions could occur on flavonoids during
preservation of food by ionizing radiation.
KEYWORDS: Quercetin; gamma irradiation; depsides; radiolytic mechanism
INTRODUCTION
quercetin was followed by HPLC analyses of the irradiated
quercetin solutions.
γ-Irradiation process is used for food preservation. Recently
this method has been investigated for fruits, but little is known
about the modification of constitutive compounds. Because of
the importance of flavonoids in human diet, it seems to be
important to develop knowledge of γ-irradiation-flavonoid
interaction.
MATERIALS AND METHODS
Chemicals and Reagents. Methanol (HPLC grade, 99.8%) was
purchased from SdS (Peypin, France), acetic acid was purchased from
Merck (Darmstadt, Germany), and quercetin was from Sigma (St. Louis,
MO).
The flavonoids are widely distributed in plants, especially in
the skins and seeds of fruits and vegetables. Because of their
presence in human food (1, 2), a large number of experimental
studies on the potential health effect of diets rich in flavonoids
has been carried out during the last 10 years (3). The flavonol
quercetin has been the most studied flavonoid. One of its major
interesting properties is antioxidant activity via free radical
scavenging or iron chelation (4-7). It has also been shown that
quercetin has multiple biological and pharmacological activities,
including antiviral (8), antiinflammatory (9-11), anti-allergy,
and anticancer properties (12-14).
None of the studies presented so far in the literature have
discussed the effect of γ-irradiation on the flavonol quercetin.
The degradation process of molecules in solution could be due
to indirect oxidation-reduction reactions generated by the
reactive species formed during the radiolysis of the solvent. In
the present study, we investigated the effect of γ-irradiation on
quercetin in methanol solutions. The degradation process of
Irradiation Treatment. Different solutions of quercetin dissolved
in methanol were prepared, and were irradiated in aliquots of 1 mL
with different doses (1-20 kGy) at a dose rate of 800 Gy/h in the
60Co source carrier type Oris experimental irradiator.
HPLC Analyses. For each analysis of pure quercetin (5 × 10-5 M,
10-4 M, 5 × 10-4 M, and 5 × 10-3 M), as control, and irradiated
quercetin, 20 µL was injected into the analytical HPLC system, a Waters
model equipped with a 600 model pump, a variable wavelength
photodiode array detector (PDA 996), and a 600 model controller. The
column used was a 250 × 4.6 mm i.d.,10 µm, µBondapak C18 cartridge
(Waters). The mobile phase consisted of 100% methanol (A) and 1%
aqueous acetic acid (B). Analyses were performed using a linear
gradient from 20% A to 80% A during 40 min at 1 mL/min.
Isolation Procedure of Products Q1 and Q2 Using Preparative
HPLC. A 5 × 10-3 M solution of quercetin (48 mg) in methanol was
irradiated at the dose of 14 kGy. The purification was performed by
semipreparative HPLC using a 100 × 25 mm i.d., 10 µm, µBondapak
C18 cartridge (Waters) and the same solvent system as above. The
gradient was 20% A to 80% A at 5 mL/min during 60 min. The eluant
containing Q1 was collected from 47 to 49 min and was evaporated at
45 °C under vacuum to give 9.4 mg. HPLC analysis was then carried
out, indicating the presence of Q1 and a new compound named Q2. A
second purification was done, giving 6.4 mg of product Q2.
Identification of Products Q1 and Q2. The identification was
* To whom correspondence should be addressed. Phone: 33 5 55 43 58
45. Fax: 33 5 55 43 58 44. E-mail: duroux@pharma.unilim.fr.
† Laboratoire de Biophysique.
‡ Laboratoire de Pharmacognosie.
1
§ Service commun de RMN.
carried out by the NMR and LC-MS techniques. H NMR and 13C
10.1021/jf020165m CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/17/2002

4828 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Marfak et al.
Figure 1. HPLC chromatograms, recorded at 280 nm, of quercetin (a) and irradiated quercetin (5 × 10-3 M) at different doses: 4 kGy (b); 8 kGy (c);
and 14 kGy (d).
1
NMR spectra were measured at 300 K in CD3OD at 400 MHz for H
NMR (dual probe) and 100 MHz for 13C NMR (dual probe) on a Bruker
DPX Avance Spectrometer using tetramethylsilane as internal standard.
The complete proton and carbon assignments were based on 1D (1H
standard, 13C Jmod) and 2D (1H-1H correlation spectroscopy (COSY),
1H-13C heteronuclear multiple quantum coherence (HMQC), and 1H-
13C heteronuclear multiple bond correlation (HMBC)) NMR experi-
ments. Mass spectroscopy was performed on a Waters Alliance system
equipped with a Waters electrospray interface. The source was operated
in the negative and positive ions modes (ES- and ES+) with a 40 V
cone voltage.
RESULTS AND DISCUSSION
γ-Irradiation of Quercetin. γ-Irradiation of a methanol
solution of quercetin was performed with different doses. The
HPLC chromatograms of quercetin at a 5 × 10-3 M concentra-
tion, as control, and of irradiated quercetin at different radiation
Figure 2. Change in concentration of Q (O) and Q1 (4) froma5× 10-3
M quercetin methanol solution irradiated at different doses of γ-ray at a
dose rate of 800 Gy/h from 60Co source.
doses are shown in Figure 1. The quercetin peak (Rt ) 30 min)
decreased when irradiation dose increased. For the high
concentration (5 × 10-3 M), the quercetin quantity decreased
significantly between 2 kGy and 14 kGy, and beyond this
irradiation dose, no significant differences were observed. We
concluded that γ irradiation was able to degrade quercetin almost
completely within 14 kGy at this concentration. In contrast, for
low concentrations (<10-3 M), short times of irradiation
treatment (dose <3 kGy) were enough to degrade all molecules
of the quercetin methanol solution.
Figure 1b, c, and d show the increase in new peaks, which
indicate the formation of several compounds. The area of each
peak increased while the quercetin one decreased. We deter-
mined the concentration of the major radiolytic compound eluted
at Rt ) 27 min and which we called Q1. The concentrations
were calculated by the use of calibration curves made with
purified materials.
Figure 2 shows the change in the concentrations of Q1 and
the consumption of quercetin versus the γ-irradiation doses,
obtained from radiolysis of methanol quercetin solution (5 ×
10-3 M). A good linear relationship was observed (R2 > 0.99).
The G-values of Q1 were determined from the slope of linear
concentration-dose equations and with the formula (1) (15)
9.65 × 106 ∆M
G )
(1)
0.79
D

Depside Formation in Irradiated Quercetin
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4829
where D is the dose (Gy), ∆M is the concentration (mol/L) of
the molecules formed, and G is the radiation chemical yield,
which corresponds to the number of molecules liberated per
100 eV of absorbed energy. G is the parameter which character-
izes the radiolysis efficiency of a product (15). The linear
relationship, observed in Figure 2, indicated that G(Q1) was
independent of the irradiation dose. This point confirmed that
the experimental conditions chosen conformed with a “simple”
model of radiolysis. All theoretical studies of solution radiolysis
indicate that γ-rays interact with the molecules of the medium
and form excited species and secondary electrons which are
confined into small zones around the primary interaction, called
“spurs” because of their 3D-geometrical configuration (15).
Hence, heterogeneous zones appear along the track of the
particles. G is constant when no overlapping of heterogeneous
zones occurs, which is the case for low linear energy transfer
(LET) radiation such as 60Co γ rays, used at low doses and low
solute concentrations. These conditions are required in order to
not complicate the radiolysis process and its interpretation.
Nevertheless, we observed, for high doses, saturation effects
of G (data not shown) which could be attributed to saturation
of the radical species formed in solvent or to overlapping
phenomena along the track.
When we determined the G values versus the concentrations,
we observed an increase of G when concentration increased. G
seemed to saturate for concentrations higher than 10-3 M. This
evolution conformed with a “classic” radiolytic process (16),
where the solute reacts with the radiolytic species from the
solvent, until all substrate was consumed.
To understand the problem of the radiolysis of quercetin in
solution, we had to consider different types of effects. First,
direct effects could occur on solute molecules, leading to
excitation and ionization of the molecules. Direct effects could
occur with solutions at high concentration or with large
molecules (polymers or DNA). In our case (∼10-3 M concen-
tration of quercetin), direct effects could be considered only as
a minor contribution. Second, indirect effects can be responsible
for chemical degradation of the molecules. In this scheme, the
solvent radiolysis products lead to chemical cascade events
which yield some very reactive species. The radiolysis of
methanol and some others alcohols has been established for
many years (17). It was shown that 50 ns after the beginning
irradiation, the irradiated solution becomes a homogeneous
solution of reactive species, and if the concentration of the solute
is small enough (<5 × 10-3 M), its degradation is activated
only by these radical species. In the case of methanol radiolysis,
the reactive homogeneous solution is composed by solvated
electron (eS-), H2, H•, CH3O•, •CH2OH, (CH2OH)2, H2CO, and
H+ such as CH3OH2+. The strong potential for electron transfer
by the cited species indicates that the degradation of the solute
is a consequence of oxidation-reduction reactions, such as
stereospecific oxidation, hydroxylation, and scission of mol-
ecules.
Figure 3. HPLC chromatogram obtained after evaporation of isolated
product Q1 from 5 × 10-3 M quercetin methanol solution irradiated at 14
kGy (a), and the UV spectra of products Q1, Q2, and quercetin (Q) (b).
UV/Vis Spectra of Q1 and Q2. The UV spectra of the
products (Figure 3b) exhibited two major absorption bands in
the 222-227 nm range and in the 270-298 nm range.
Compared to the UV spectrum of quercetin, we concluded that
both the characteristic absorption bands at 255 nm and at 369
nm (attributed to the A-C ring conjugation and to the B-C
ring conjugation, respectively) had disappeared. This indicated
that in the Q1 and the Q2 structures, the C ring was destroyed.
NMR and Mass Spectroscopic Identification of Q2 and
1
Q1. These two products were identified by H, 13C, COSY,
ROESY, NOESY, HMQC, and HMBC NMR experiments and
by the negative and positive ES mass data.
The 1H NMR spectrum of product Q2 showed signals
corresponding to five aromatic protons (Table 1). Two proton
signals at 6.24 and 6.17 ppm possessed the same coupling
constant (J ) 2.2 Hz) ascribable to the A-ring of quercetin,
and a series of three proton signals at 7.58 ppm (d, J ) 2.1
Hz), 7.56 ppm (dd, J ) 2.1, 8.2 Hz), and 6.86 ppm (d, J ) 8.2
Hz) were assigned to the B-ring of quercetin. The positions of
All of the reactions in solute are generally very fast and occur
during the first second of the process. In the present study, it
was interesting to notice that the chromatogram reached a
definitive pattern for high doses, which indicated that the
radiolysis products were stable against irradiation. However,
after irradiation, we observed during the isolation of Q1, the
formation of a second compound Q2 eluting at Rt ) 12 min
(Figure 3a). It was clear that this chemical transformation
occurred out of the irradiated medium and was not due to the
irradiation process.
1
all protons were confirmed by H-1H COSY spectra.
The 13C Jmod spectrum of Q2 showed three quaternary
carbonyl groups, the first ascribable to a ketone function (188.6
ppm, C-7), the second ascribable to an acid group (165.8 ppm,
C-8), and the third ascribable to an ester group (165.9 ppm,
C-7′). The identification of these carbonyl groups was demon-
strated by HMBC experiments (Table 1). Signals corresponding

4830 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Marfak et al.
Table 1. 1H (400 MHz) and 13C (100 MHz) NMR Data and
Correlations Observed in COSY, HMQC, and HMBC Spectra of
Products Q1 and Q2 in CD OD/TMS (δ ppm; J Hz)
3
13C J mod/
HMQC
position
1H
COSY
HMBC
product Q1
1
2
3
4
5
6
7
8
105.6
168.2
101.7
167.7
105.0
155.4
188.6
165.5a
53.2
6.27 d (2.2)
6.19 d (2.2)
H-5
C-1; C-2; C-4; C-5
C-1; C-3; C-4; C-6
H-5
8-OCH
3.44 s
C-8
3
1′
2′
3′
4′
5′
6′
7′
120.7
118.0
146.6
153.1
116.2
124.9
165.9a
7.48 d (2.1)
H-6′
H-6′
C-1′; C-3′; C-4′; C-6′; C-7′
6.89 d (8.2)
7.50 dd (8.2; 2.1) H-2′; H-5′
C-1′; C-3′; C-4′; C-6′
C-1′; C-2′; C-4′; C-5′; C-7′
product Q2
1
2
3
4
5
6
7
8
1′
2′
3′
4′
5′
6′
7′
105.5
168.1
101.5
167.2
104.7
155.5
188.6
165.8b
121.3
118.4
146.3
152.7
116.1
125.0
165.9b
6.24 d (2.2)
6.17 d (2.2)
H-5
H-3
C-1; C-2; C-4; C-5
C-1; C-3; C-4; C-6
Figure 4. Selected 1H−13C long-range correlations (3J) (T) observed in
HMBC spectra of products Q1 and Q2.
C7-C8 linkage. Finally, the product Q2 was identified as {2-
[(3′,4′-dihydroxybenzoil)oxy]-4,6-dihydroxyphenyl}(oxo) acetic
acid (Figure 4).
7.58 d (2.1)
6.86 d (8.2)
H-6′
H-6′
C-1′; C-3′; C-4′; C-6′; C-7′
The negative and positive ES-MS spectra of product Q2
showed molecular ion peaks at m/z 333 [M - H]- and at m/z
357 [M + Na]+, respectively. The molecular weight of Q2 was
determined to be 334, suggesting the formula C15H10O9. The
positive FAB-LSIMS mass spectroscopy, performed at Centre
Re´gional de Mesures Physiques de l’Ouest (Rennes, France),
using metanitrobenzylic alcohol (mNBA) matrix afforded
pseudo molecular ion peak at m/z 357 [M + Na]+. The high-
resolution mode of the FABMS revealed that the molecular
formula of Q2 was C15H10O9 (calculated for C15H10O9Na,
357.0223; found, 357.0234) corresponding to the structure
obtained from the NMR analysis. In addition, the positive ES-
MS spectrum displayed other ion peaks at m/z 137 and at m/z
199 (Figure 5a). The presence of the m/z 199 fragment confirms
the C7-C8 linkage. The proposed pathways for these fragmen-
tations are illustrated in Figure 6.
C-1′; C-3′; C-4′; C-6′
C-1′; C-2′; C-4′; C-5′; C-7′
7.56 dd (8.2; 2.1) H-2′; H-5′
a,b Assignments may be interchangeable.
to twelve aromatic carbons were also detected, five of which
were tertiary (101.5 ppm, C-3; 104.7 ppm, C-5; 118.4 ppm,
C-2′; 116.1 ppm, C-5′; 125.0 ppm, C-6′) and seven of which
were quaternary. Five of these latter were O-linked (168.1 ppm,
C-2; 167.2 ppm, C-4; 155.5 ppm, C-6; 146.3 ppm, C-3′; 152.7
ppm, C-4′), and two were C-linked (105.5 ppm, C-1; 121.3 ppm,
C-1′). The positions of the five tertiary carbons were assigned
from the H-13C HMQC experiments (Table 1). These data
1
confirmed the structure of the A-ring. In addition, the positions
of all O-linked and C-linked aromatic carbons were deduced
with the carbon-proton long-range couplings obtained from the
HMBC spectrum (Table 1).
The 1H NMR and 13C spectra (Table 1) of product Q1 were
found to be very similar to those of product Q2, and it appeared
that they were of the same analogous series differing only by
the presence of a methoxyl group, represented by a carbon
resonance at 53.2 ppm, and by three protons (3.44 ppm, s). The
chemical structure of Q1 was determined by procedures similar
to those used for Q2 identification. The position of the methoxyl
The ROESY and the NOESY experiments of Q2 were
performed. No NOE correlations were observed between H3
and H6′ or H2′. This result is not an indication of no connection
between the two rings, but indicating that the stable conforma-
tion is obtained for torsion angles between the two rings which
implicate a “high” distance (g4-5 Å) between the cited protons.
However, the connection is elucidated as the following. First,
the chemical shift of C1′ (121.3 ppm), which corresponds to a
C-linked carbon, and the 3J correlation between C7′ (165.9 ppm)
and the protons H2′(7.58 ppm), H6′(7.56 ppm), led us to
establish the C1′-C7′ linkage. Second, C7′ and C2 (168.1 ppm)
were identified as an ester group and an O-linked carbon
respectively, which allowed the C7′-O-C2 linkage.
3
was determined by the observation of J correlation between
the methoxyl protons (3.44 ppm) and the C-8 carbon (165.5
ppm). Consequently, the product Q1 was identified as {2-[(3′,4′-
dihydroxybenzoil)oxy]-4,6-dihydroxyphenyl}(oxo) methyl ace-
tate (Figure 4).
The observation of the molecular ions at m/z 347 [M - H]-,
at m/z 349 [M + H]+, and at m/z 371 [M + Na]+, respectively
in the negative and positive ES-MS spectra, suggested the
molecular formula C16H12O9 for Q1. The positive FAB-LSIMS
mass spectroscopy, using metanitrobenzylic alcohol (mNBA)
matrix, afforded pseudo molecular ion peaks at m/z 349 [M +
H]+ and at m/z 371 [M + Na]+. The high-resolution mode of
the FABMS revealed that the molecular formula of Q1 was
The absence of long-range couplings in the HMBC spectrum
between any protons and the C-7 (188.6 ppm) and C-8 (165.8
ppm) carbons, together with the presence of the quaternary
C-linked carbon (105.5 ppm, C-1), led us to establish the C1-

Depside Formation in Irradiated Quercetin
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4831
Figure 5. Positive ES−MS spectra of products Q2 (a) and Q1 (b).
C16H12O9 (calculated for C16H13O9 and C16H12O9Na, 349.0560
and 371.0379; found, 349.0560 and 371.0370, respectively)
confirming the structure obtained from the NMR analysis.
Other fragment ions, in the positive spectrum, were detected
at m/z 137 and at m/z 213 (Figure 5b). The proposed pathways
of these fragments are illustrated in Figure 6. The fragment at
m/z 199 showed in the positive spectrum of the product Q2,
was detected at m/z 213 for the product Q1. This difference
confirmed the -OCH3 group in Q1. In addition, the spectra of
both Q1 and Q2 possessed the same fragment at m/z 137, due
to the same formation pathway of this fragment from Q1 and
Q2.
Equilibrium between Q1 and Q2. In conclusion, the major
radiolytic compound of Q is a depside. Formation of depsides
during oxidation of flavonols has been proposed for many years
(18) and was recently confirmed by oxidation catalyzed by
flavonolato copper (II) phthalazine (19). However, Q1 seemed
to be thermodynamically unstable under mild conditions,
because it could be easily transformed in Q2 in the presence of
water. Indeed, Q1 eluted in HPLC with approximately 70%
MeOH/30% water. The water seemed to hydrolyze the methoxyl
group and form Q2. When Q2 was dissolved in methanol, the
esterification reaction was observed and the chemical equilib-
rium between Q1 and Q2 was established.
Proposed Mechanism for Quercetin Radiolysis in Metha-
nol Solution. Quercetin is a flavonol and its antioxidant capacity
is attributed to the presence of hydroxyl substituents in positions
3, 5, and 7 which permitted redox transfers, and to the π electron
delocalization, between the B ring, the 2,3-double bond, and
the keto group which was responsible for the stabilization of
the radical produced during antioxidant reactions. Quercetin
therefore exhibits a facility for oxidation-reduction reactions
which occur during the radiolysis process. Aerial oxidation of
flavonols has been established for many years. The primary
process involved in quercetin oxidation is the opening of the
2,3-double bond with formation of the depside Q2 which is then
hydrolyzed to the aromatic acids (18). In the case of quercetin
radiolysis in methanol solution, we obtained the depside Q1
because, after opening the C ring, methoxylation was more easy
Figure 6. Proposed pathways for formation of the fragments at m/z 213
and 137, detected in the positive ES−MS spectrum of the product Q1,
and the fragments at m/z 119 and 137, detected in that of compound
Q2.

4832 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Marfak et al.
Figure 7. Proposed transformation mechanism for the formation of depside Q1 by γ radiolysis of quercetin.
than hydroxylation in radiolyzed methanol solution due to the
presence of •OCH3 radicals. We localized the primary reactive
site of oxidation at position 3, because no irradiation product
was formed after irradiation of rutin in the same experimental
conditions used for quercetin. Rutin is quercetin glycosylated
at position 3. We performed irradiation with doses up to 30
kGy and no effect was observed and concluded that sugar
stabilized quercetin against oxidative reactions due to γ irradia-
tion.
formation of the major degradation product. However, it is
possible that these groups reacted with the radicals of the
irradiated medium to form some minor radiolytic compounds.
The irradiation process has been used to preserve the quality
of food by killing bacteria, and so extend the shelf life (20,
21). γ-Irradiation has been reported as a potential treatment for
extending the post-harvest life of fresh fruits and vegetables
(22-24). It was also established that γ-irradiation of Citrus
clementina Hort. ex. Tanaka increases the synthesis of total
phenolic compounds including phenolic acids and flavonoids
(25). However, little is known about the effect of γ-irradiation
on the constitutive molecules of fruits and the vegetables,
especially for flavonoids. γ-Irradiation of quercetin methanol
solution does not constitute an exact model of flavonoid
irradiation in fruits, however, it could contribute to the
understanding of the reaction mechanism of quercetin with the
free radicals produced in solution by irradiation. The free
radicals produced in methanol and in acid water solution, which
constitutes a better model of fruits, are not the same. Probably,
the major oxidative species, during γ-irradiation of fruits, are
produced from the water radiolysis. Pulsed radiolysis studies
of flavonoids in water solution (26) reported stereospecific
oxidation of the hydroxyl groups in B-ring. This process leads
to the formation of the corresponding quinone. However, van
Acker discussed the influence of pH on the oxidative sites in
flavonoids (27). More generally, the structure-activity relation-
ship studies showed that, according to the experimental condi-
The transformation mechanism of quercetin to depside Q1
•
is shown in Figure 7. We propose that the OCH3 radicals,
initiated by irradiation, react with the OH group in position 3
•
to form the flavonoxy radical Q11. Then OCH3 radical could
add to position 2 and form an intermediate product Q12. The
CH3 of the methoxy group could then be eliminated from Q12
and the formation of a keto group in position 2 lead to scission
of the molecule to compound Q13. Two mechanisms are
proposed for CH3 elimination. First, a H• radical could react
with the CH3 group to form CH4. The second mechanism
•
implicates a OCH3 radical which yields the ether CH3OCH3.
Such a mechanism seems to be more possible, because •OCH3
was implicated in all the quercetin transformation steps we
described in Figure 7. Probably because of their lifetime, the
•OCH3 radical must be the most likely candidate for these
•
transformation reactions. Finally, a OCH3 radical reacts with
the radical Q13 to form the depside Q1. The role of the OH
groups in positions 3′ and 4′ seemed to be a minor factor in

Depside Formation in Irradiated Quercetin
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4833
(13) Huang, M. T.; Wood, A. W.; Newmark, H. L.; Sayer, J. M.;
Yagi, H.; Jerina, D. M.; Conney, A. H. Inhibition of the
mutagenicity bay-region diol-epoxides of polycyclic aromatic
hydrocarbons by phenolic plant flavonoids. Carcinogenesis 1983,
4, 1631-1637.
(14) Sousa, R. L.; Marletta, M. A. Inhibition of cytochrom P-540
activity in rat liver microcosms by the naturally occurring
flavonoid quercetin. Arch. Biochem. Biophys. 1985, 240, 345-
357.
(15) Fricke, H.; Hart, E. J. Chemical dosimetry. In Radiation
Dosimetry, Vol. II, Instrumentation; Attix, F. H., Roesch, W.
C., Tochilin, E., Eds.; Academic Press: New York, 1966; pp
167-237.
(16) Ferrandini, C.; Pucheault, J. Biologie de l’action des rayonne-
ments; Masson: Paris, 1983.
(17) Freeman, G R., Ed. The radiolysis of alcohols. In Kinetics of
Nonhomogeneous Processes: A Practical Introduction for
Chemists, Biologists, Physicists, and Materials Scientists; Wiley-
Interscience: New York, 1987; pp 73-101.
(18) Gottlieb, O. R. Flavonols. In The FlaVonoids; Harborne, J. B.,
Mabry, T. J., Eds.; Chapman and Hall; London, 1975; pp 296-
375.
tions, the oxidative site could be the 3-OH group, the 4-keto
group, or the OH groups of B-ring (4, 5).
•
The chemical action of OCH3 observed probably does not
mimic all the physical conditions which occur during fruit
irradiation, but it constitutes the first step for investigating the
radiolysis process, especially the C-ring opening. The flavonol
scission was also obtained under favorable thermodynamic
conditions such as in aerial oxidized solution (18) and catalytic
reactions (28).
Because of the low water solubility of quercetin, and with
the aim of reproducing the irradiation conditions of fruits, we
actually try to irradiate quercetin methanol/water solutions.
ACKNOWLEDGMENT
We are grateful to H. Lotfi (Faculte´ de Pharmacie de Limoges,
France) for his help in LC-MS measurements throughout this
work.
LITERATURE CITED
(19) Balogh-Hergovich, E.; Kaiser, J.; Speier, G. Synthesis and
characterization of copper (I) and copper (II) flavonolate
complexes with phthalazine, and their oxygenation and relevance
to quercetinase. Inorg. Chim. Acta 1997, 256, 9-14.
(20) Josephson, E. S.; Peterson, M. S. PreserVation of Food by
Ionisation Radiation. CRC Press: Boca Raton, FL, 1982; Vol.
I-III.
(21) Zareena, A. V.; Variyar, P. S.; Gholap, A. S.; Bongirwar, D. R.
Chemical investigation of gamma-irradiated saffron (Crocus
satiVus L.). J. Agric. Food Chem. 2001, 49, 687-691.
(22) Kader, A. A. Potential applications of ionizing radiation in
postharvest handling of fresh fruits and vegetables. Food Technol.
1986, 40, 117-121.
(23) Urbain, W. M. Food irradiation. Food Science Technology; a
series of monographs; Academic Press: Orlando, FL, 1986.
(24) Thomas, P. Radiation preservation of foods of plant origins Part
VI. Mushrooms, tomatoes, minor fruits and vegetables, dried
fruits and nuts. CRC Crit. ReV. Food Sci. Nutr. 1988, 26, 313-
358.
(25) Oufedjikh, H.; Lacroix, M.; Mahrouz, M.; Amiot, M. J. Effect
of γ irradiation on phenolic compounds and phenylalanine
ammonia-lyase activity during storage in relation to peel injury
from peel of Citrus clementina Hortic. ex. Tanaka. J. Agric. Food
Chem. 2000, 48, 559-565.
(1) Herrmann, K. Flavonols and flavones in food plants: A review.
J. Food Technol. 1976, 11, 433-448.
(2) Justesen, U.; Knuthsen, P.; Teth, T. Quantitative analysis of
flavonols, flavones, and flavanones in fruits, vegetables and
beverages by high-performance liquid chromatography with
photodiode array and mass spectrometric detection. J. Chro-
matogr. A 1998, 799, 101-110.
(3) Hertog, M. G. L.; Feskens, E. J. M.; Hollman, P. C. H.; Katan,
M. B.; Kromhout, D. Dietary antioxidant flavonoids and risk of
coronary heart disease: The Zutphen elderly study. Lancet 1993,
342, 1007-1011.
(4) Pietta, P. G. Flavonoids as antioxidants. J. Nat. Prod. 2000, 63,
1035-1042.
(5) Rice-Evans, C. A.; Miller, N.; Paganga, G. Structure-antioxidant
activity relationships of flavonoids and phenolic acids. Free
Radical Biol. Med. 1996, 20, 933-956.
(6) Cotelle, N.; Bernier, J. L.; Catteau, J. P.; Pommery, J.; Wallet,
J. C.; Gaydou, E. M. Antioxidant properties of hydroxy-flavones.
Free Radical Biol. Med. 1996, 20, 35-43.
(7) Guohua, C.; Emin, S.; Ronald, L. Antioxidant and prooxidant
behaviour of flavonoids: Structure-activity relationships. Free
Radical Biol. Med. 1997, 22, 749-760.
(8) Van Hoof, L.; Van den Berghe, D. A.; Hatfield, G. M.; Vlietinck,
A. J. Plant antiviral agents: V 3-Methoxyflavones as potent
inhibitors of viral-induced block of cell synthesis. Planta Med.
1984, 50, 513-517.
(9) Brasseur, T. Anti-inflammatory properties in flavonoids. J.
Pharm. Belg. 1989, 44, 235-241.
(10) Wei, B. L.; Lu, C. M.; Tsao, L. T.; Wang, J. P.; Lin, C. N. In
vitro anti-inflammatory effects of quercetin 3-O-methyl ether and
other constituents from Rhamnus species. Planta Med. 2001, 67,
745-747.
(11) Fischer, M.; Mills, G. D.; Slaga, T. J. Inhibition of mouse skin
tumour promotion by several inhibitors of the arachidonic acid
metabolism. Carcinogenesis 1982, 3, 1243-1245.
(12) Huang, M.; Ferraro, T. Phenolic compounds in foods and cancer
prevention. In Phenolic Compounds in Food and their Effect on
Health II, Huang, M. T., Ho, C. T., Lee, C. Y., Eds.; ACS
Symposium Series; American Chemical Society: Washington,
DC, 1992; Chapter 2, p 507.
(26) Jovanovic, S. V.; Steenken, S.; Tosic, M.; Marjanovic, B.; Simic,
M. G. Flavonoids as antioxidants. J. Am. Chem. Soc. 1994, 116,
4846-4851.
(27) Van Acker, S. A. B. E.; de Groot, M. J.; van den Berg, D. J.;
Tromp, M. N. J. L.; den Kelder, G. D. O.; van der Vijgh, W. J.
F.; Bast, A. A quantum chemical explanation of the antioxidant
activity of flavonoids. Chem. Res. Toxicol. 1996, 9, 1305-1312.
(28) Kaiser, J.; Speier, G. Radical-initiated oxygenation of flavonols
by dioxygen. J. Mol. Catal. A 2001, 171, 33-36.
Received for review February 7, 2002. Revised manuscript received
May 28, 2002. Accepted May 30, 2002. This work was supported in
part by the Ministe`re de l’Education Nationale, de la Recherche et de
la Technologie and by the Conseil Re´gional du Limousin (France).
JF020165M