Antioxidative effects of flavonols and their glycosides against the free-radical-induced peroxidation of linoleic acid in solution and in micelles

Article abstract of DOI:10.1002/chem.200400391

The antioxidative effect of flavonols and their glycosides against the peroxidation of linoleic acid has been studied in homogeneous solution (tBuOH/H₂O, 3:2) and in sodium dodecyl sulfate and cetyl trimethylammonium bromide micelles. The peroxidation was initiated thermally by the water-soluble initiator 2,2′-azobis(2-methyl-propionamidine) dihydrochloride, and the reaction kinetics were studied by monitoring the formation of linoleic acid hydroperoxides. The synergistic antioxidant effect of the flavonols with α-tocopherol (vitamin E) was also studied by following the decay kinetics of α-tocopherol and the α-tocopheroxyl radical. Kinetic analysis of the antioxidative process demonstrates that the flavonols are effective antioxidants in solution and in micelles, either alone or in combination with α-tocopherol. The antioxidative action involves trapping the initiating radicals in solution or in the bulk-water phase of the micelles, trapping the propagating lipid peroxyl radicals on the surface of the micelles, and regenerating α-tocopherol by reducing the α-tocopheroxyl radical. It was found that the antioxidant activity of the flavonols and their glycosides depends significantly on the position and number of the hydroxy groups, the oxidation potential of the molecule, and the reaction medium. The flavonols bearing ortho-dihydroxy groups possess significantly higher antioxidative activity than those without such functionalities, and the glycosides are less active than their parent aglycones. The activity of the flavonols is higher in micelles than in solution, while the activity of α-tocopherol is lower in micelles than in solution. This is because the predominant factor for controlling the activity is the hydrogen-bonding interaction of the antioxidant with the micellar surface in the case of hydrophilic flavonols, while it is the inter- and intramicellar diffusion in the case of lipophilic α-tocopherol.

Full text of DOI:10.1002/chem.200400391

Antioxidative Effects of Flavonols and Their Glycosides against the Free-
Radical-Induced Peroxidation of Linoleic Acid in Solution and in Micelles
Bo Zhou, Qing Miao, Li Yang, and Zhong-Li Liu*[a]
Abstract: The antioxidative effect of
flavonols and their glycosides against
the peroxidation of linoleic acid has
been studied in homogeneous solution
(tBuOH/H2O, 3:2) and in sodium dode-
cyl sulfate and cetyl trimethylammoni-
um bromide micelles. The peroxidation
was initiated thermally by the water-
soluble initiator 2,2’-azobis(2-methyl-
propionamidine) dihydrochloride, and
the reaction kinetics were studied by
monitoring the formation of linoleic
acid hydroperoxides. The synergistic
antioxidant effect of the flavonols with
a-tocopherol (vitamin E) was also stud-
ied by following the decay kinetics of
a-tocopherol and the a-tocopheroxyl
radical. Kinetic analysis of the antioxi-
dative process demonstrates that the
flavonols are effective antioxidants in
solution and in micelles, either alone or
in combination with a-tocopherol. The
antioxidative action involves trapping
the initiating radicals in solution or in
the bulk-water phase of the micelles,
trapping the propagating lipid peroxyl
radicals on the surface of the micelles,
and regenerating a-tocopherol by re-
ducing the a-tocopheroxyl radical. It
was found that the antioxidant activity
of the flavonols and their glycosides
depends significantly on the position
and number of the hydroxy groups, the
oxidation potential of the molecule,
and the reaction medium. The flavo-
nols bearing ortho-dihydroxy groups
possess significantly higher antioxida-
tive activity than those without such
functionalities, and the glycosides are
less active than their parent aglycones.
The activity of the flavonols is higher
in micelles than in solution, while the
activity of a-tocopherol is lower in mi-
celles than in solution. This is because
the predominant factor for controlling
the activity is the hydrogen-bonding in-
teraction of the antioxidant with the
micellar surface in the case of hydro-
philic flavonols, while it is the inter-
and intramicellar diffusion in the case
of lipophilic a-tocopherol.
Keywords: antioxidation
·
flavo-
nols · peroxides · radical reactions ·
reaction mechanisms
Introduction
as flavones, flavanone, flavonols, flavanols, and isoflavones,
are naturally occurring polyphenolic compounds present in
vegetables, fruits, tea, and red wine and possess a wide
range of biological activities,[3] of which antioxidation has
been extensively explored.[4] One interesting example of fla-
vonoid activity is the so-called “French paradox”,[5] that is,
despite high fat intake, mortality from coronary heart dis-
ease is lower in some regions of France, a fact attributed to
the regular drinking of red wine which contains high levels
of flavonoids (approximately 200 mg per glass)[3a,6] and re-
sveratrol (0.1–15 mgLꢀ1).[7] These compounds have been
proved to be good antioxidants against low-density lipopro-
tein (LDL) peroxidation,[3b,4a] a process believed to be criti-
cal in the risk of human atherosclerosis,[1h,8] as well as to
possess cancer chemopreventive activity.[3b,9] We have re-
cently found that flavanols isolated from green-tea leaves
are good antioxidants against free-radical-initiated lipid per-
oxidation in solution,[10] in micelles,[11] in human red blood
cells,[12] in human low-density lipoprotein,[13] and in rat liver
microsomes,[14] and that the antioxidant activity of these fla-
Epidemiological, biological, and clinical studies have provid-
ed various lines of evidence in the past decade to indicate
that free-radical-induced oxidative damage of cell mem-
branes, DNA, and proteins might play a causative role in
aging and several degenerative diseases, such as cancer,
atherosclerosis, and cataract formation, and that antioxi-
dants, such as a-tocopherol (vitamin E), l-ascorbic acid (vi-
tamin C), and b-carotene, might have beneficial effects in
protecting against these diseases.[1] Therefore, inhibition of
free-radical-induced oxidative damage by supplementation
of antioxidants has become an attractive therapeutic strat-
egy to reduce the risk of these diseases.[2] Flavonoids, such
[a] Dr. B. Zhou, Q. Miao, Prof. L. Yang, Prof. Z.-L. Liu
National Laboratory of Applied Organic Chemistry
Lanzhou University, Lanzhou, Gansu 730000 (P.R. China)
Fax : (+86)931-862-5657
E-mail: liuzl@lzu.edu.cn
680
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200400391
Chem. Eur. J. 2005, 11, 680 – 691

FULL PAPER
vanols depends significantly on the structure of the mole-
cules and the initiation conditions.[10–14] It was also found
that these green-tea flavanols and resveratrol might interact
with a-tocopherol (vitamin E) synergistically to enhance the
antioxidant activity.[15,16] Therefore, it is of interest to extend
this research and study the structure–activity relationships
of other dietary flavonoids and their glycosides, since many
dietary flavonoids exist in the form of glycosides.[17] We
report herein a quantitative kinetic study of the antioxida-
tive behavior of a set of typical flavonols and their glyco-
sides against linoleic acid peroxidation in solution and in mi-
celles, with emphasis placed on the structure–activity rela-
tionships and the mechanistic details of the antioxidation,
including the synergistic interaction between these flavonols
and a-tocopherol (vitamin E). The flavonols studied were
myricetin (MY), quercetin (Q), quercetin galactopyranoside
(QG), quercetin rhamnopyranoside (QR), rutin (R), morin
(MO), kaempferol (K), and kaempferol glucoside (KG).
The peroxidation was thermally initiated at physiological
temperature by a water-souble azo initiator, 2,2’-azobis(me-
thylpropionamidine) dihydrochloride (AAPH), and con-
ducted either in tert-butyl alcohol/water (3:2) solution or
in sodium dodecyl sulfate (SDS) and cetyl trimethylammoni-
um bromide (CTAB) micelles to study the effect of micro-
environment on the reaction. The interaction of these com-
pounds with a-tocopherol (TOH, vitamin E) was also inves-
tigated.
Results
Inhibition of linoleic acid peroxidation by flavonols and
their glycosides in solution: Peroxidation of linoleic acid or
its esters gives different hydroperoxides depending on the
reaction conditions.[18] Hydroperoxide substitution at the C-
9 or C-13 positions produces either trans,trans- or cis,trans-
conjugated dienes, which are the major products in the ab-
sence of antioxidants or in the presence of small amount of
antioxidants, for example, millimolar concentrations of a-to-
copherol.[18a,b] It was found recently that these conjugated
dienes were formed from the rapid b scission of the primari-
ly formed bisallylic 11-peroxyl radical,[18c,d] and the kinetical-
ly controlled product, that is,
the nonconjugated 11-substitut-
ed
hydroperoxide,
might
become the major product in
the presence of high concentra-
tions of antioxidant, for exam-
ple, molar concentrations of a-
tocopherol.[18d] These experi-
mental observations have been
rationalized recently by theo-
retical calculations.[19] The pres-
ent experiment used very small
amounts of the antioxidants
(micromolar a-tocopherol and/
or flavonols and their glyco-
sides), hence the production of
the nonconjugated 11-hydroper-
oxide should be negligible, and
the conjugated hydroperoxides
were the predominant products.
The latter showed a characteris-
tic ultraviolet (UV) absorption
at 235 nm[20] that was used to
monitor the formation of the
total hydroperoxides formed
during the peroxidation after
separation of the reaction mix-
ture by high-performance liquid
chromatography (HPLC).
A set of representative kinet-
ic curves of the total hydroper-
oxide formation during the per-
oxidation of linoleic acid in
tBuOH/H2O (3:2) solution is
shown in Figures 1 and 2. It can
be seen from the figures that,
Chem. Eur. J. 2005, 11, 680 – 691
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
681

Z.-L. Liu et al.
the original rate of the propagation, a result indicating the
exhaustion of the antioxidant. The turning point from the in-
hibition period to the restoration of fast peroxidation relates
to the inhibition time, tinh, which is also an indication of the
efficacy of the antioxidants. Addition of only the flavonols
and their glycosides (FOHs) to the solution decreased the
rate of hydroperoxide formation from Rp to Rinh, but no in-
hibition period was observed (lines b–f in Figure 1). The an-
tioxidant activity can be expressed by the percentage inhibi-
tion of the peroxidation, Pinh =(RpꢀRinh)/Rp ꢁ100%, which
follows the sequence MY>Q>KꢁMO>QGꢁQR>Rꢁ
KG (see Table 1 below).
Figure 2 shows the antioxidative effect of representative
FOHs in the presence of a-tocopherol (TOH) in solution.
The addition of TOH produced a typical kinetic curve for a
chain-breaking antioxidative reaction, with a clear inhibition
period (line b in Figure 2) as reported previously.[10] Interest-
ingly, the addition of FOHs, which showed no inhibition
period in the absence of TOH, remarkably prolonged the in-
hibition period of TOH (lines c–g in Figure 2). This demon-
strates a synergistic antioxidation effect of FOHs and TOH
(see below).
Figure 1. Representative kinetic curves for the formation of total hydro-
peroxides (LOOH) during the peroxidation of LH in tert-butyl alcohol/
water (3:2) solution at 378C, initiated with AAPH and inhibited with fla-
vonols and their glycosides (FOHs). [LH]0 =0.1 molLꢀ1
, [AAPH]0 =
10 mmolLꢀ1, [FOHs]0 =20 mmolLꢀ1. a) Uninhibited reaction, b) reaction
inhibited with MY, c) reaction inhibited with Q, d) reaction inhibited
with QG, e) reaction inhibited with MO, f) reaction inhibited with K.
Curves of other FOHs are not shown for clarity.
Inhibition of linoleic acid peroxidation by flavonols and
their glycosides in micelles: It has been recognized that anti-
oxidant activity in homogenous solutions may not parallel
that in heterogeneous media, let alone the activity in vivo.[21]
To bridge the gap between chemical and biological activities,
it is essential to understand and evaluate the dependence of
the antioxidant activity upon the microenvironment of the
reaction media. A general methodology for this is to carry
out the reaction in membrane mimetic systems, such as mi-
celles and artificial bilayers.[22] Therefore, the antioxidative
effect of FOHs was investigated in sodium dodecyl sulfate
(SDS) and cetyl trimethylammonium bromide (CTAB) mi-
celles, as shown in Figures 3 and 4, respectively. It can be
Figure 2. Representative kinetic curves for the formation of total hydro-
peroxides (LOOH) during the peroxidation of LH in tert-butyl alcohol/
water (3:2) solution at 378C, initiated with AAPH and inhibited with
TOH and FOHs. [LH]0 =0.1 molLꢀ1, [AAPH]0 =10 mmolLꢀ1, [TOH]0 =
20 mmolLꢀ1, [FOHs]0 =20 mmolLꢀ1. a) Uninhibited reaction, b) reaction
inhibited with TOH, c) reaction inhibited with MY and TOH, d) reaction
inhibited with Q and TOH, e) reaction inhibited with QG and TOH,
f) reaction inhibited with MO and TOH, g) reaction inhibited with K and
TOH. Curves of other FOHs are not shown for clarity.
upon AAPH initiation, the concentration of linoleic acid hy-
droperoxides increased quickly and linearly with time in the
absence of antioxidants (line a in Figures 1 and 2), a result
indicating the fast peroxidation of the substrate. The slope
of the line corresponds to the rate of propagation, Rp, of the
peroxidation. The formation of the hydroperoxides was re-
markably inhibited by the addition of a-tocopherol (TOH)
during the so-called “inhibition period” or “induction
period” (line b in Figure 2). During the inhibition period,
the concentration of the hydroperoxides also increased ap-
proximately linearly with time, and the slope of this line was
designated as Rinh, the value of which reflects the antioxida-
tive potential of the antioxidant. After the inhibition period,
the rate of hydroperoxide formation increased to close to
Figure 3. Representative kinetic curves for the formation of total hydro-
peroxides (LOOH) during the peroxidation of linoleic acid (LH) in SDS
micelles (0.1 molLꢀ1) at pH 7.4 and 378C, initiated with AAPH and in-
hibited by flavonols and their glycosides (FOHs). [LH]0 =15.2 mmolLꢀ1
,
[AAPH]0 =6.3 mmolLꢀ1, [FOHs]0 =10 mmolLꢀ1. a) Uninhibited peroxi-
dation, b) reaction inhibited with MY, c) reaction inhibited with Q, d) re-
action inhibited with QG, e) reaction inhibited with MO, f) reaction in-
hibited with K. Curves of other FOHs are not shown for clarity.
682
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 680 – 691

Antioxidative Effects of Flavonols and Their Glycosides
FULL PAPER
seen from Figure 3 that the antioxidative behavior of the fla-
vonols in SDS micelles is distinctly different from that in the
homogeneous solution. All of the FOHs, which showed no
inhibition period in solution, exhibited clear inhibition peri-
ods in SDS micelles and behaved well as chain-breaking an-
tioxidants. Similar results were obtained in CTAB micelles
(Figure 4), but the kinetic parameters in the two micelles
were appreciably different. The details will be discussed in
the following sections.
Figure 6. Representative kinetic curves for the formation of total hydro-
peroxides (LOOH) during the peroxidation of linoleic acid (LH) in
CTAB micelles (15 mmolLꢀ1) at pH 7.4 and 378C, initiated with AAPH
and inhibited with TOH and FOHs. [LH]0 =15.2 mmolLꢀ1, [AAPH]0 =
6.3 mmolLꢀ1, [FOHs]0 =10 mmolLꢀ1, [TOH]0 =10 mmolLꢀ1. a) Uninhib-
ited peroxidation, b) reaction inhibited with TOH, c) reaction inhibited
with MY and TOH, d) reaction inhibited with Q and TOH e) reaction in-
hibited with QG and TOH, f) reaction inhibited with MO and TOH,
g) reaction inhibited with K and TOH. Curves of other FOHs are not
shown for clarity.
Figure 4. Representative kinetic curves for the formation of total hydro-
peroxides (LOOH) during the peroxidation of linoleic acid (LH) in
CTAB micelles (15 mmolLꢀ1) at pH 7.4 and 378C, initiated with AAPH
and inhibited by flavonols and their glycosides (FOHs). [LH]0 =
15.2 mmolLꢀ1, [AAPH]0 =6.3 mmolLꢀ1, [FOHs]0 =10 mmolLꢀ1. a) Unin-
hibited peroxidation, b) reaction inhibited with MY, c) reaction inhibited
with Q, d) reaction inhibited with QG, e) reaction inhibited with MO,
f) reaction inhibited with K. Curves of other FOHs are not shown for
clarity.
is much more pronounced in the micelles than in the homo-
geneous solution, especially in the case of Q and MY, which
almost completely inhibited the peroxidation for a very long
time. For example, the inhibition period produced by TOH
and Q in combination in SDS micelles was 293 min, which is
much longer than the sum of the inhibition periods pro-
duced by TOH (78 min) and Q (90 min) when they were
used individually under the same experimental conditions.
The mechanistic details of this antioxidant synergism will be
discussed in the following sections.
Figures 5 and 6 show the antioxidative effect of FOHs in
the presence of TOH in SDS and CTAB micelles, respec-
tively. A comparison of Figures 5 and 6 with Figure 2 clearly
indicates that the antioxidant synergism of FOHs with TOH
Decay kinetics of a-tocopherol and the a-tocopheroxyl radi-
cal: To rationalize the mechanism of the antioxidant syner-
gism of TOH and the FOHs, the decay kinetics of TOH and
the a-tocopheroxyl radical (TOC) in the absence and pres-
ence of quercetin (Q) were studied. The decay of TOH was
determined by HPLC separation of the reaction mixture,
followed by electrochemical determination of the amount of
remaining TOH. Representative results are illustrated in
Figures 7 and 8. It was found that the decay of TOH was ap-
proximately linear in the absence of Q in the solution and in
SDS and CTAB micelles (line a in Figure 7, lines a and c in
Figure 8), in accordance with the kinetic demand for chain-
breaking antioxidation reactions (see Equation (8) below).
The decay rates were 9.0ꢁ10ꢀ9, 1.6ꢁ10ꢀ9, and 4.1ꢁ
10ꢀ9 molLꢀ1 in solution and in SDS and CTAB micelles, re-
spectively, due to the different rates of initiation in these
media (see below). On the other hand, the decay of TOH in
the presence of Q was different in the solution and in the
micelles. In the solution the decay of TOH was only slightly
reduced by the coexistent Q (line b in Figure 7), while it was
significantly diminished in the micelles, especially in SDS
micelles (line b in Figure 8). The decay rates of TOH in the
presence of Q were determined to be 7.5ꢁ10ꢀ9, 0.5ꢁ10ꢀ9,
Figure 5. Representative kinetic curves for the formation of total hydro-
peroxides (LOOH) during the peroxidation of linoleic acid (LH) in SDS
micelles (0.1 molLꢀ1) at pH 7.4 and 378C, initiated with AAPH and in-
hibited with TOH and FOHs. [LH]0 =15.2 mmolLꢀ1
, [AAPH]0 =
6.3 mmolLꢀ1, [FOHs]0 =10 mmolLꢀ1, [TOH]0 =7.5 mmolLꢀ1. a) Uninhib-
ited peroxidation, b) reaction inhibited with TOH, c) reaction inhibited
with MY and TOH, d) reaction inhibited with Q and TOH, e) reaction
inhibited with QG and TOH, f) reaction inhibited with MO and TOH,
g) reaction inhibited with K and TOH. Curves of other FOHs are not
shown for clarity.
Chem. Eur. J. 2005, 11, 680 – 691
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
683

Z.-L. Liu et al.
Figure 7. Decay of a-tocopherol (TOH) during the inhibition of linoleic
acid peroxidation in tert-butyl alcohol/water (3:2) solution at 378C, initi-
ated with AAPH and inhibited with TOH in the absence (a) and pres-
Figure 9. Decay of a-tocopheroxyl radical (TOC) in SDS micelles
(0.2 molLꢀ1) at pH 7.4 and room temperature under air in the absence
(a) and presence (b) of quercetin (0.21 mmolLꢀ1). The inset shows the
EPR spectrum of TOC obtained under fast flow.
ence (b) of quercetin (Q). [LH]0 =0.1 molLꢀ1, [AAPH]0 =10 mmolLꢀ1
,
[TOH]0 =20 mmolLꢀ1, [Q]0 =20 mmolLꢀ1
.
Figure 8. Decay of a-tocopherol (TOH) during the inhibition of linoleic
acid peroxidation in micelles at 378C, initiated with AAPH and inhibited
with TOH and quercetin (Q). a) Decay of TOH in the absence of Q in
SDS micelles (0.1 molLꢀ1), b) decay of TOH in the presence of Q in
SDS micelles (0.1 molLꢀ1), c) decay of TOH in the absence of Q in
CTAB micelles (15 mmolLꢀ1), d) decay of TOH in the presence of Q in
Figure 10. Plot of the pseudo-first-order rate constants, kobs, of the decay
of TOC versus the initial concentration of quercetin in SDS micelles
(0.2 molLꢀ1).
pherol regeneration reaction (see Equation (12) below), was
CTAB micelles (15 mmolLꢀ1). [LH]0 =15.2 mmolLꢀ1
6.3 mmolLꢀ1 [TOH]0 =7.5 mmolLꢀ1 (SDS), [TOH]0 =10 mmolLꢀ1
(CTAB), [Q]0 =10 mmolLꢀ1
,
[AAPH]0 =
obtained as 64mꢀ1 sꢀ1.
,
.
Discussion
and 3.0ꢁ10ꢀ9 molLꢀ1 in solution and in SDS and CTAB mi-
celles, respectively.
Reaction kinetics of lipid peroxidation: It has been proven
that the reaction kinetics of lipid peroxidation in micelles
and biomembranes follow the same rate law as those in ho-
mogenous solutions;[22,24] therefore, the same rate law was
accepted in solution and micelles. The kinetics of linoleic
acid (LH) peroxidation initiated by azo compounds and its
inhibition by a chain-breaking antioxidant (AH) have been
discussed in detail previously.[10,11a,22,24] The rate of propaga-
tion (Rp) and the rate of peroxide formation in the inhibi-
tion period (Rinh) are given by Equations (1) and (2), respec-
tively.
The a-tocopheroxyl radical (TOC) was produced by oxidiz-
ing TOH with PbO2, and its decay kinetics were determined
by stopped-flow electron paramagnetic resonance (EPR)
spectroscopy as described previously.[15,16] TOC is much more
persistent in micelles than in homogenous solutions,[16,23]
a
fact that makes it easy to determine the reaction kinetics of
the radical by EPR spectroscopy at ambient temperatures in
micelles. As shown in Figure 9, TOC decayed fairly slowly in
SDS micelles with a rate constant of 75mꢀ1 sꢀ1. Addition of
Q remarkably increased the decay of TOC, the kinetics of
which were found to be pseudo first order in the presence of
a large excess of Q (line b in Figure 9). Plotting this first-
order rate constant, kobs, against the concentration of Q gave
a straight line (Figure 10), from which the bimolecular rate
constant between TOC and Q, that is, the rate for the a-toco-
d½LOOHꢂ=dt ¼ Rp ¼ ½kp=ð2ktÞ ꢂRi1=2½LHꢂ
Rinh ¼ kpRi½LHꢂ=ðn kinh½AHꢂÞ
1=2
ð1Þ
ð2Þ
684
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 680 – 691

Antioxidative Effects of Flavonols and Their Glycosides
FULL PAPER
Here kp, kt, and kinh are rate constants for the chain propa-
gation [Eq. (3)], chain termination [Eq. (4)] and chain inhib-
ition by the antioxidant [Eq. (5)], respectively, Ri is the ap-
parent rate of chain initiation [Eq. (6)], and n is the stoi-
chiometric factor (see below).
by Equations (9) and (10) for the uninhibited and inhibited
peroxidations, respectively. The kinetic parameters deduced
from Figures 1–6 are listed in Tables 1–3.
kclp ¼ Rp=Ri
ð9Þ
kp
LOOC þ LH
LOOH þ LC
ð3Þ
ð4Þ
ð5Þ
ð6Þ
kclinh ¼ Rinh=Ri
ð10Þ
ƒ!
2kt
2 LOOC
molecular products
ƒ!
Micellar effects on the initiation and antioxidation: It can
be seen from Figures 1–8 and Tables 1–3 that the reaction
medium exerts significant effects on the rate of initiation,
the rate of propagation, and the antioxidant activity of a-to-
copherol (TOH) and flavonols and their glycosides (FOHs).
LOOC þ AH kinh LOOH þ AC
ƒ!
Ri ¼ 2kge½RꢀN¼NꢀRꢂ
The Ri values calculated from the inhibition periods
[Eq. (7)] are 1.6ꢁ10ꢀ8, 3.1ꢁ10ꢀ9, and 8.3ꢁ10ꢀ9 molLꢀ1 in so-
lution and in SDS and CTAB micelles, respectively. These
values are in reasonable agreement with the values of 1.8ꢁ
10ꢀ8, 3.2ꢁ10ꢀ9, and 8.2ꢁ10ꢀ9 molLꢀ1, respectively, that were
obtained from the decay of TOH [Eq. (8)]. The values in so-
lution and in CTAB micelles are very close to the value of
(1.4ꢃ0.2)ꢁ10ꢀ6[AAPH] sꢀ1 reported previously for liposo-
mal dispersions,[25,26] with the concentrations of AAPH
taken as 10 and 6.3 mmolLꢀ1 in solution and in the micelles,
respectively, in the present experiments. However, the Ri
value of AAPH in SDS micelles is significantly smaller than
that in CTAB micelles. This can be understood because
AAPH is positively charged, hence it is prone to being ad-
sorbed onto the surface of the SDS micelles; this, in turn, re-
duces the effective initiation, due to the cage effect.
Although the radical generation rate, Ri, of AAPH is
known as (1.4ꢃ0.2)ꢁ10ꢀ6[AAPH] sꢀ1 at 378C for protein-
containing solutions and liposomal dispersions,[25,26] the cage
effect parameter, e, varies appreciably depending on the
medium and the concentrations of the antioxidant and the
initiator.[25] Therefore, the Ri value is generally determined
by the inhibition period and/or by the decay rate of TOH
[Eqs. (7) and (8), respectively].
Ri ¼ n½AHꢂ0=tinh
ð7Þ
ð8Þ
Ri ¼ ꢀn d½AHꢂ=dt
Here, n is the stoichiometric factor that designates the
number of peroxyl radicals trapped by each antioxidant mol-
ecule. Since the n value of a-tocopherol is generally as-
sumed to be 2,[22,24] the Ri value can be determined from the
inhibition period and/or the decay rate of a-tocopherol.
The kinetic chain length (kcl) defines the number of chain
propagations initiated by each initiating radical and is given
The inhibition rate constant, kinh, for the antioxidation re-
action [Eq. (5)] by TOH in tert-butyl alcohol/water solution
was calculated to be 4.9ꢁ105 mꢀ1 sꢀ1 by taking kp =100mꢀ1 sꢀ1
at 378C.[27] This value is close to the value obtained in tert-
Table 1. Inhibition of AAPH-initiated peroxidation of linoleic acid by flavonols and their glycosides (FOHs) and a-tocopherol (TOH) in solution.[a,b]
Antioxidant
Rp/10ꢀ8
Rinh/10ꢀ8
tinh/102
[s]
kinh[c]/105
kclp
kclinh
Pinh
[%]
SE%
[%]
[molLꢀ1 sꢀ1
]
[molLꢀ1 sꢀ1
]
[Lmolꢀ1 sꢀ1
]
none
14.3
9.0
MY
Q
R
QG
QR
MO
K
KG
3.8
8.2
12.4
11.4
11.7
9.2
9.1
12.6
1.7
1.2
1.3
1.8
3.2
2.4
5.1
7.7
7.1
7.3
5.8
5.7
7.9
1.1
0.8
0.8
1.1
2.0
1.2
1.2
1.9
1.4
73
43
13
20
18
36
36
12
TOH
14.8[d]
10.3[d]
9.0[d]
25.2
45.0
37.2
28.7
31.5
29.9
34.8
33.6
27.3
4.9
3.1
3.6
3.1
1.5
2.9
2.5
1.9
2.8
9.3
6.5
5.7
7.7
7.8
10.4
7.7
7.2
7.2
MY + TOH
Q + TOH
R + TOH
QG + TOH
QR + TOH
MO + TOH
K + TOH
KG + TOH
78
48
14
25
19
38
33
8
12.2[d]
12.4[d]
16.5[d]
12.3[d]
11.4[d]
11.4[d]
1.8
2.0
3.1
2.2
[a] In tBuOH/H2O (3:2) mixed solvent at 378C initiated with AAPH (10 mmolLꢀ1). The initial concentrations of linoleic acid, FOHs and TOH were
0.1 molLꢀ1, 20 mmolLꢀ1, and 20 mmolLꢀ1, respectively. Data are the averages of three determinations with a deviation of less than ꢃ10%. [b] The Ri
[27]
value was taken as 1.6ꢁ10ꢀ8 molLꢀ1 sꢀ1; see text for further details. [c] Calculated from Equation (2) by taking the kp value as 100 Lmolꢀ1 sꢀ1
.
[d] Rate
of propagation after exhaustion of the antioxidants.
Chem. Eur. J. 2005, 11, 680 – 691
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
685

Z.-L. Liu et al.
Table 2. Inhibition of AAPH-initiated peroxidation of linoleic acid by flavonols and their glycosides (FOHs) in micelles.[a,b]
Micelle
FOHs
Rp/10ꢀ8
Rinh/10ꢀ8
tinh/103
[s]
kinh[c]/104
n
kclp
kclinh
Pinh
[%]
[molLꢀ1 sꢀ1
]
[molLꢀ1 sꢀ1
]
[Lmolꢀ1 sꢀ1
]
SDS
SDS
SDS
SDS
SDS
SDS
SDS
SDS
none
MY
Q
7.1
22.9
9.0
2.8[d]
3.9[d]
3.3[d]
3.4[d]
3.7[d]
5.8[d]
6.3[d]
5.9[d]
18
0.7
1.1
2.2
2.2
2.4
1.6
1.9
2.9
5.5
5.4
2.9
2.4
2.4
4.5
4.7
1.6
2.5
1.7
1.4
1.6
1.5
1.6
1.1
1.8
1.7
1.7
0.9
0.7
0.7
1.4
1.5
0.5
2.3
3.5
7.1
7.1
7.7
5.1
6.1
9.4
75
72
33
35
35
72
70
51
12.6
10.6
11.0
11.9
18.7
20.3
19.0
21.7
10.0
6.9
5.8
6.1
7.6
16.4
19.4
19.8
R
QG
QR
MO
K
KG
none
MY
Q
SDS
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
8.2[d]
5.7[d]
4.8[d]
5.1[d]
6.3[d]
13.6[d]
16.1[d]
16.4[d]
1.2
1.5
4.1
3.3
3.1
1.5
2.6
7.9
6.3
4.7
2.8
3.5
3.2
2.6
3.6
1.4
1.4
1.4
0.8
0.8
0.8
1.8
1.1
0.8
5.2
3.9
2.3
2.9
2.7
2.2
3.0
1.2
1.4
1.8
4.9
4.0
3.7
1.8
3.1
9.5
85
74
15
35
51
89
84
52
R
QG
QR
MO
K
KG
[a] The reaction conditions and the initial concentration of substrates are the same as those described in the legends of Figures 3 and 4 for reactions con-
ducted in SDS and CTAB micelles, respectively. The calculations were based on the total reaction volume. Data are the averages of three reproducible
determinations with a deviation of less than ꢃ10%. [b] The Ri values were taken as 3.1 and 8.3 nmolLꢀ1 sꢀ1 in SDS and CTAB micelles, respectively;
[30]
see the text for further details. [c] Calculated from Equation (2) by taking the kp value as 37 Lmolꢀ1 sꢀ1
.
[d] Rate of propagation after exhaustion of
the antioxidants.
Table 3. Inhibition of AAPH-initiated peroxidation of linoleic acid by flavonols and their glycosides (FOHs) and a-tocopherol (TOH) in micelles.[a,b]
Micelle
Antioxidant
Rp/10ꢀ8
Rinh/10ꢀ8
tinh/103
[s]
kinh[c]/104
n’[d]
kclp
kclinh
SE%
[%]
[molLꢀ1 sꢀ1
]
[molLꢀ1 sꢀ1
]
[Lmolꢀ1 sꢀ1
]
SDS
SDS
SDS
SDS
SDS
SDS
SDS
SDS
TOH
7.8[e]
5.1[e]
4.3[e]
2.7[e]
2.8[e]
4.3[e]
5.7[e]
4.5[e]
7.0[e]
22[e]
0.6
0.2
0.2
0.3
0.3
0.4
0.3
0.2
0.4
1.8
0.3
0.4
0.9
1.3
1.1
0.4
0.4
0.4
4.7
19.5
17.6
10.1
6.9
3.6
3.1
3.7
2.9
4.9
4.2
2.7
3.9
4.3
2.0
3.0
2.7
2.2
1.6
1.9
5.0
3.1
6.8
2.0
3.0
2.7
1.8
1.1
1.1
2.0
2.0
1.0
2.0
4.6
3.5
1.9
2.2
2.1
2.2
3.2
1.3
24
1.9
0.6
0.6
1.0
1.0
1.3
1.0
0.6
1.3
2.1
0.4
0.5
1.1
1.6
1.3
0.5
0.5
0.5
MY + TOH
Q + TOH
R + TOH
QG + TOH
QG + TOH
MO + TOH
K + TOH
KG + TOH
TOH
MY + TOH
Q + TOH
R + TOH
QG + TOH
QR + TOH
MO + TOH
K + TOH
KG + TOH
16.5
13.9
8.7
91
75
33
0
9.0
6.8
13.9
18.4
14.5
22.6
27
14.8
8.2
5.7
8.3
7.8
17.7
20.2
18.1
0
12.8
12.6
6.2
2.3
11.0
8.5
4.6
5.2
5.1
5.2
40
34
0
SDS
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
CTAB
12.3[e]
6.8[e]
4.8[e]
6.9[e]
6.5[e]
14.7[e]
16.8[e]
15.0[e]
28
21
0
0
0
0
29
0
7.6
3.2
[a] The reaction conditions and the initial concentration of substrates are the same as those described in the legends of Figures 5 and 6 for reactions con-
ducted in SDS and CTAB micelles, respectively. The calculations were based on the total reaction volume. Data are the averages of three reproducible
determinations with a deviation of less than ꢃ10%. [b] The Ri values were taken as 3.1 and 8.3 nmolLꢀ1 sꢀ1 in SDS and CTAB micelles, respectively;
[30]
see the text for further details. [c] Calculated from Equation (2) by taking the kp value as 37 Lmolꢀ1 sꢀ1
.
[d] n’=Ritinh/([FOH]0 +[TOH]0). [e] Rate of
propagation after exhaustion of the antioxidants.
butyl alcohol (5.1ꢁ105 mꢀ1 sꢀ1) that was reported previous-
ly[26] but is remarkably smaller than that obtained in chloro-
benzene (3.2ꢁ106 mꢀ1 sꢀ1).[23] This decrease of the kinh value
in the polar solvent is expected since the strong hydrogen-
bonding ability of tert-butyl alcohol with TOH would make
the antioxidation reaction more difficult.[27–29] On the other
hand, the kinh value in micelles (3.6ꢁ104 and 2.0ꢁ104 mꢀ1 sꢀ1
in SDS and CTAB micelles, respectively, for kp =
37mꢀ1 sꢀ1[30]) was more than one order of magnitude smaller
than that in tert-butyl alcohol. Obviously, this lower reactivi-
ty of TOH in micelles cannot be explained by the hydrogen-
bonding interaction of water with TOH since water is not as
good a hydrogen-bond acceptor as tert-butyl alcohol. The
solute hydrogen-bond basicity parameters, bH2, of water and
686
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 680 – 691

Antioxidative Effects of Flavonols and Their Glycosides
FULL PAPER
tert-butyl alcohol were reported to be 0.38 and 0.49, respec-
tively.[31] Therefore, it must be a consequence of the physical
separation and the slow diffusion rate of the lipid peroxyl
radicals and TOH in the interior of the micelles.[32,33] Castle
and Perkins[32] and we[33] have proven previously that inter-
and intramicellar diffusions are the rate-limiting steps for
the antioxidation reaction conducted in micelles. Indeed, the
kinh value of TOH in SDS micelles (3.6ꢁ104 mꢀ1 sꢀ1) is about
two times larger than that in CTAB micelles (2.0ꢁ
104 mꢀ1 sꢀ1) because the microviscosity in the interior of
CTAB micelles is 2.6 times larger than that in SDS mi-
celles;[34] this makes the intramicellar diffusion slower in
CTAB micelles. In addition, since lipid peroxyl radicals are
polar (dipole moment of approximately 2.6 Debye) and
electrophilic,[24] they should move to the surface of micelles
and be subject to intermicellar diffusion[32,33] more quickly in
SDS than in CTAB micelles, so as to react with TOH whose
reactive phenoxyl functional group resides on the surface of
the micelle.[32]
celles),[16] and to those of green-tea polyphenols (0.3–3.7ꢁ
104 mꢀ1 sꢀ1 in micelles).[11a]
The antioxidant efficacy of these FOHs in micelles can be
assessed by comparing their inhibition times (tinh) and/or
kclinh values. This comparison gives the efficacy sequence
MY>Q>KꢁMO>QGꢁQRꢁR>KG, which is similar to
the sequence in the homogeneous solution.
Antioxidant synergism of flavonols and their glycosides with
a-tocopherol: It can be seen from Figures 1–6 and Tables 1
and 3 that addition of flavonols and their glycosides (FOHs)
together with a-tocopherol (TOH) significantly increased
the inhibition period of the latter, even if no inhibition
period could be observed when the FOHs was used alone.
This antioxidant synergism can be quantified by the percent-
age increment of the inhibition period when the two antioxi-
dants are used in combination with reference to the sum of
the inhibition periods when the two antioxidants are used
individually; this vale is termed the synergistic efficiency,
SE% [Eq. (11)].[35]The antioxidant synergism of a-tocopher-
Antioxidative activity of the flavonols and their glycosides:
It cab be seen from Figure 1 and Table 1 that the addition of
flavonols and their glycosides (FOHs) decreased the rate of
hydroperoxide formation and the kinetic chain length, but
no inhibition period was observed in the homogeneous so-
lution. This indicates that, in solution, FOHs can trap the in-
itiating radicals (ROOC) derived from the thermal decompo-
sition of AAPH but are unable to trap the propagating lino-
leic acid peroxyl radicals (LOOC), probably because the
latter reaction is too slow to compete with the former reac-
tion in solution. This behavior is similar to that reported
previously for green-tea polyphenols in homogeneous so-
lution.[10] The antioxidative efficacy can be assessed by the
percentage inhibition (Pinh) or the kinetic chain length in the
inhibition period (kclinh) and follows the sequence MY>
Q>KꢁMO>QGꢁQR>RꢁKG.
SE% ¼ ftinhðTOH þ FOHÞꢀ½tinhðTOHÞ
ð11Þ
þtinhðFOHÞꢂg=½tinhðTOHÞ þ tinhðFOHÞꢂ ꢄ 100
ol (TOH) with coexistent antioxidants, such as l-ascorbic
acid (vitamin C),[26] green-tea polyphenols,[11,15] and resvera-
trol analogues,[16] has been extensively studied and rational-
ized as due to the reduction of the a-tocopheroxyl radical
(TOC) by the coexistent antioxidant to regenerate TOH. The
prolonged decay of TOH in the presence of Q (Figures 7
and 8) and, especially, the stopped-flow EPR spectroscopy
experiments (Figures 9 and 10) demonstrate clearly that Q
can also reduce TOC to regenerate TOH [Eq. (12)], hence
providing a rationale for the antioxidant synergism.
TOC þ FOH kreg TOH þ FOC
ƒ!
ð12Þ
It is worth noting that the antioxidation behavior of
FOHs in micelles is distinctly different from that in homoge-
neous solution. In SDS and CTAB micelles all the flavonols
and their glycosides showed clear inhibition periods in
which the rate of propagation and the kinetic chain length
are remarkably reduced. This indicates that these flavonols
and their glycosides are able to trap the propagating linoleic
acid peroxyl radicals in micelles and behave well as chain-
breaking antioxidants. This might be due to the fact that
these flavonols and their glycosides can bind to the Stern
layer of the micelles by hydrogen-bonding interactions, con-
centrated on the micellar surface, hence facilitating their in-
teraction with the lipid peroxyl radical whose polar phenox-
yl group also resides in the surface of the micelle.[32] It can
be seen from Table 2 that the rate constant for trapping the
propagating peroxyl radicals, kinh, of these flavonols and
their glycosides is in the range of 0.8–2.5ꢁ104 mꢀ1 sꢀ1 in both
SDS and CTAB micelles. These values are comparable to
those of TOH (3.6ꢁ104 and 2.0ꢁ104 mꢀ1 sꢀ1 in SDS and
CTAB micelles, respectively; see Table 3), to those of re-
sveratrol and its analogues (0.5–3.1ꢁ104 mꢀ1 sꢀ1 in mi-
It should be pointed out in this context, however, that this
a-tocopherol regeneration reaction may not be adequate to
explain the antioxidant synergism. The steady-state concen-
tration of TOC during the antioxidation reaction in SDS mi-
celles can be estimated as 0.06 mmolLꢀ1 [36]
;
hence the rate of
TOH regeneration by
Q is calculated to be 0.04ꢁ
10ꢀ9 molLꢀ1 sꢀ1 if the kreg value [Eq. (12), the rate for the a-
tocopherol regeneration] of Q in SDS micelles is taken as
64 mꢀ1 sꢀ1 (see above) and the initial concentration of Q as
10 mmolLꢀ1. Obviously, this rate of regeneration can only
account for approximately 4% of the rate decrease of the
decay of TOH (from 1.6ꢁ10ꢀ9 molLꢀ1 sꢀ1 in the absence of
Q to 0.5ꢁ10ꢀ9 molLꢀ1 sꢀ1 in the presence of Q; Figure 8),
hence the increase of the inhibition period [Eqs. (7) and
(8)]. Therefore, there must be an additional mechanism re-
sponsible for the remarkable antioxidant synergism of Q
with TOH.
It can be seen from Figures 1 and 2 and Table 1 that al-
though the FOHs did not produce inhibition periods in so-
lution, they significantly prolonged the inhibition period of
Chem. Eur. J. 2005, 11, 680 – 691
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
687

Z.-L. Liu et al.
TOH when the FOHs were used together with the latter.
Comparison of the percentage inhibition, Pinh, with the syn-
ergistic efficiency, SE%, demonstrates that these two param-
eters are almost the same within the experimental deviation
for every FOH. Since the inhibition period is inversely pro-
portional to the initial concentration of AAPH,[22] that is,
the rate of initiation, the reduced rate of the initiation due
to the trapping of initiating radicals by the FOHs must be
the predominant factor for the antioxidant synergism in the
solution.
The same mechanism should also play an important role
in the antioxidant synergism of TOH with FOHs in micelles.
In the case of Q in SDS micelles, for example, the reduced
initiation contributes 72% of the antioxidant synergism,
while the TOH regeneration contributes only 4%. This is in
good agreement with the 75% synergistic efficiency of TOH
and Q calculated from the inhibition period (Tables 2 and
3). Other FOHs showed similar results, as listed in Tables 2
and 3. It should be pointed out, however, that if the FOH
itself can produce an inhibition period, the diminished rate
of propagation in the inhibition period, Rinh, would be
caused not only by the reduced initiation but also by the in-
hibited propagation [Eq. (5)]. In this case the calculated Pinh
value would no longer reflect the true value of the reduced
inhibition; hence, in some cases the value of Pinh is apprecia-
bly larger than the value of SE%.
to reduce the effective initiation. This, in turn, reduces the
rate of propagation and decreases the kinetic chain length.
The antioxidant synergism with TOH is caused by the re-
duction of the rate of initiation and/or by the regeneration
of TOH, depending on the reaction medium and the struc-
ture of the FOH, as discussed in the previous section.
Structure–activity relationship: It can be seen from the re-
sults listed in Tables 1–3 that the antioxidative activity of
MY and Q is appreciably higher than MO and K. That is,
the molecules bearing an ortho-diphenoxyl functionality
possess higher activity than those bearing no such function-
ality. The higher activity of flavonoids bearing ortho-diphe-
noxyl functionality on the B ring has been reported pre-
viously.[4b,i,m,n] This can be understood because the oxidation
intermediate of MY and Q, the ortho-hydroxy phenoxyl rad-
ical, is more stable due to the intramolecular hydrogen-
bonding interaction, as evidenced recently from both experi-
ments[4b,37] and theoretical calculations.[38] The theoretical
calculation showed that the hydrogen bond in the ortho-hy-
droxy phenoxyl radical is approximately 4 kcalmolꢀ1 stron-
ger than that in the parent catechol and that the bond disso-
ciation energy (BDE) of catechol is 9.1 kcalmolꢀ1 lower
than that of phenol and 8.8 kcalmolꢀ1 lower than that of re-
sorcinol.[38a] In addition, it should be easier to further oxi-
dize the ortho-hydroxy phenoxyl radical and/or ortho-semi-
quinone radical anion to form the ortho-quinone intermedi-
ate and/or product.[37,39] (Scheme 2). The fact that the stoi-
chiometric factor, n, of MY and Q in micellles is larger than
one (Table 2) suggests that the second peroxyl radical must
be involved in the antioxidation reaction that leads to the
formation of the corresponding ortho-quinones, as shown in
Scheme 2.
Antioxidation mechanism of flavonols and their glycosides:
Based on the above discussions the antioxidation mecha-
nism of flavonols and their glycosides (FOHs) in micelles
might involve trapping the initiating radical (ROOC) that re-
duces the effective initiation, trapping the propagating lino-
leic acid peroxyl radical (LOOC) that produces the inhibition
period, and reducing the a-tocopheroxyl radical (TOC) that
regenerates TOH, therby prolonging the inhibition period,
as depicted in Scheme 1. On the other hand, in homogene-
ous solutions the FOHs can only trap the initiating radicals
It can also be seen that the flavonol glycosides (R, QG,
QR, KG) possess appreciable lower antioxidant activity
than their aglycones (Q and K) in solution and in micelles.
It has been reported that the hydroxy group at position 3 is
required for the maximal radical scavenging activity of fla-
vonols.[40] Theoretical calculations also indicated that the O
H bond with the lowest BDE value in these flavonols (MY,
ꢀ
Q and K) is that of the 3-OH group on ring C.[8d] For exam-
ꢀ
ple, the BDE values of the O H bond at positions 3, 4’, 3’,
7, and 5 in myricetin (MY) were calculated to be 26.9, 29.1,
33.2, 42.4, and 44.9 kcalmolꢀ1, respectively; thus, the order
of preference of the position from which a hydrogen atom is
most likely to be abstracted in the MY molecule should be
3, 4’, 3’, 7, and 5.[8d] In addition, the 3-OH group of quercetin
is highly effective in stabilizing the reaction intermediate
radical anion and allows the formation of the p-quinonoid
structure[4l] as exemplified in Scheme 2. Obviously, blocking
position 3 by binding with a sugar residue will decrease the
activity of the molecule. Introduction of a sugar to the flavo-
nols would make them more hydrophilic,[41a] and hence in-
hibit them from reaching the micellar surface and reacting
with the lipophilic peroxyl radicals residing on the interior
of the micelle. It has been pointed out that the partition be-
tween the water and oil phases is an important factor for
Scheme 1. Antioxidative and TOH-regeneration reactions of FOH in mi-
celles.
688
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 680 – 691

Antioxidative Effects of Flavonols and Their Glycosides
FULL PAPER
Scheme 2. Antioxidation reactions of quercetin (Q) by hydrogen abstraction and electron transfer.
the antioxidant activity of flavonoids in membranes and in
lipid bilayers, and the flavonoid partitioned more in the
water phase possesses less activity.[41b–d] The lower activity of
QG than Q in phospholipid bilayers has been reported pre-
viously.[4j] Our recent results also show that flavonol glyco-
sides are less active than their aglycones in the inhibition of
lipid peroxidation of human low-density lipoprotein[42] and
erythrocyte ghosts.[43]
It is also noticeable that the antioxidative activity of
FOHs is correlated with the electrochemical behavior of the
molecule. The oxidation peak potential, Epa, was reported to
be 0.02, 0.11, 0.17, and 0.20 V (versus a saturated calomel
electrode (SCE)) for MY, Q, K, and MO, respectively,[4h]
and the introduction of a sugar to flavonoids increases their
oxidation potential. For example, the oxidation potentials at
pH 7, E7, of Q and R were reported to be 0.33[44] and
0.6 V[45] (versus a normal hydrogen electrode (NHE)), re-
spectively, and their half-peak oxidation potentials, Ep/2 are
0.03 and 0.18 V (versus an SCE), respectively.[46] Therefore,
the increase of the oxidation potential correlates well with
the decrease of the antioxidative activity in the sequence of
MY>Q>K>MO>R. This correlation suggests that elec-
tron-transfer antioxidation might take place simultaneously
with the direct hydrogen-abstraction reaction, as exempli-
fied in Scheme 2. It is well known that phenoxides undergo
electron-transfer oxidation more easily to produce relatively
stable phenoxide radical anions in alkaline media. The acid
dissociation constants, pK1, of Q, K and, R were reported to
be 6.7, 8.2, and 7.1, respectively.[45] This demonstrates that
these FOHs can partially dissociate under our experimental
conditions (pH 7.4) and the electron-transfer reaction is
therefore feasible. Cooperation between hydrogen-abstrac-
tion and electron-transfer processes in antioxidation reac-
tions by phenolic antioxidants has recently been discusse-
d.[38a,47] It has been pointed out that the relative contribution
of hydrogen abstraction and electron transfer in the antioxi-
dation of flavanols depends on the experimental conditions,
such as the character of the attacking radical, the pH value,
and the stability of the intermediate radical species.[47] In the
present case the hydrogen abstraction might take place first
ꢀ
from the 3-OH group because this O H bond is the weakest
[8d]
ꢀ
O H bond in the flavonols,
and the electron transfer
might take place first from the 4’-phenolate of the B ring be-
cause the 4’-OH group is the most acidic in the molecule.[47]
The proposed mechanism is depicted in Scheme 2.
Chem. Eur. J. 2005, 11, 680 – 691
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
689

Z.-L. Liu et al.
Conclusion
Acknowledgement
We thank the National Natural Science Foundation of China (Grant
nos.: 20172025, 20332020, and 20021001) for financial support.
Flavonols and their glycosides (FOHs), that is, MY, Q, MO,
K, R, QG, QR, and KG, are effective antioxidants against li-
noleic acid peroxidation in solution and micelles. The reac-
tion medium exerts significant influence on the antioxidant
activity of FOHs and the synergistic antioxidation mecha-
nism between TOH and the FOHs. The activity of flavonols
is higher in micelles than in solution, while the activity of a-
tocopherol is lower in micelles than in solution. This is be-
cause the predominant factor of controlling the activity is
the hydrogen-bonding interaction of the antioxidant with
the micellar surface in the case of hydrophilic flavonols,
while it is the inter- and intramicellar diffusion in the case
of lipophilic a-tocopherol. The observation that flavonols
and their glycosides bearing ortho-diphenoxyl and 3-hydroxy
functionalities possess remarkably higher antioxidant activi-
ty than those without such functionalities gives us useful in-
formation for antioxidant drug design.
[1] a) T. Finkel, N. J. Holbrook, Nature 2000, 408, 239–247; b) K. J.
Barnham, C. L. Masters, A. I. Bush, Nat. Rev. Drug Discovery 2004,
3, 205–214; c) S. P. Hussain, L. J. Hofseth, C. C. Harris, Nat. Rev.
Cancer 2003, 3, 276–285; d) M. S. Cooke, M. D. Evans, M. Dizdaro-
glu, J. Lunec, FASEB J. 2003, 17, 1195–1214; e) L. J. Marnett, Carci-
nogenesis 2000, 21, 361–370; f) A. R. Collins, BioEssays 1999, 21,
238–246; g) B. Halliweill, J. M. C. Gutteridge, Free Radicals in Biol-
ogy and Medicine, 3rd ed., Oxford University Press, Oxford, 1999;
h) H. Esterbauer, P. Ramos, Rev. Physiol. Biochem. Pharmacol.
1995, 127, 31–63.
[2] a) C. A. Rice-Evans, A. T. Diplock, Free Radical Biol. Med. 1993,
15, 77–96; b) D. E. Brash, P. A. Havre, Proc. Natl. Acad. Sci. USA
2002, 99, 13969–13971; c) Y.-J. Surh, Nat. Rev. Cancer 2003, 3, 768–
780; d) O. Blokhina, E. Virolainen, K. Fagerstedt, Ann. Bot. 2003,
91, 179–194.
[3] a) K. D. Croft, Ann. N. Y. Acad. Sci. 1998, 854, 435–442; b) J. B.
Harborne, C. A. Williams, Phytochemistry 2000, 55, 481–504; c) E.
Haslam, Practical Polyphenolics, Cambridge University Press, Cam-
bridge, 1998; d) B. A. Bohm, Introduction to Flavonoids, Harwood
Academic Publisher, Amsterdam, 1998.
[4] a) P.-G. Pietta, J. Nat. Prod. 2000, 63, 1035–1042; b) D. B. McPhail,
R. C. Hartley, P. T. Gardner, G. G. Duthie, J. Agric. Food Chem.
2003, 51, 1684–1690; c) M. J. T. J. Arts, G. R. M. M. Haenen, L. C.
Wilms, S. A. J. N. Beetstra, C. G. M. Heijnen, H.-P. Voss, A. Bast, J.
Agric. Food Chem. 2002, 50, 1184–1187; d) P. Goupy, C. Dufour, M.
Loons, O. Dangles, J. Agric. Food Chem. 2003, 51, 615–622; e) J. A.
Manthey, N. Guthrie, J. Agric. Food Chem. 2002, 50, 5837–5843;
f) V. Roginsky, Arch. Biochem. Biophys. 2003, 414, 261–270;
g) C. D. Sadik, H. Sies, T. Schewe, Biochem. Pharmacol. 2003, 65,
773–781; h) K. Furuno, T. Akasako, N. Sugihara, Biol. Pharm. Bull.
2002, 25, 19–23; i) W. Bors, C. Michel, Free Radical Biol. Med. 1999,
27, 1413–1425, and references therein; j) K. Ioku, T. Tsushida, Y.
Takei, N. Nakatani, J. Terao, Biochim. Biophys. Acta 1995, 1234, 99–
104; k) J. Terao, M. Piskula, Q. Yao, Arch. Biochem. Biophys. 1994,
308, 278–284; l) O. Dangles, G. Fargeix, C. Dufour, J. Chem. Soc.
Perkin Trans. 2 1999, 1387–1395; m) W. Bors, W. Heller, C. Michel,
M. Saran, Methods Enzymol. 1990, 186, 343–355; n) V. A. Belyakov,
V. A. Roginsky, W. Bors, J. Chem. Soc. Perkin Trans. 2 1995, 2319–
2326.
[5] a) G. J. Soleas, E. P. Diamandis, D. M. Goldberg, Clin. Biochem.
1997, 30, 91–113; b) R. Corder, J. A. Douthwaite, D. M. Lees, N. Q.
Khan, A. C. V. D. Santos, E. G. Wood, M. J. Carrier, Nature 2001,
414, 863–864.
[6] a) R. A. Larson, Phytochemistry 1988, 27, 969–978; b) E. N. Frankel,
J. Kanner, J. B. German, E. Parks, J. E. Kinsella, The Lancet 1993,
341, 454–456; c) E. N. Frankel, A. L. Waterhouse, J. E. Kinsella,
The Lancet 1993, 342, 1103–1104.
Experimental Section
Materials: QR, QG, and KG were isolated from apple peels and green-
tea leaves, respectively, by consecutive extraction with methanol, water,
and ethyl acetate and chromatographic separation on a Sephadex LH-20
column, with reference to procedures reported previously.[48] Their struc-
1
tures and purity were confirmed by H and 13C NMR spectra and HPLC.
MY and Q (from Sigma), R (from Aldrich), MO (from Tokyo Kaset
Kogyo), K (from Fluka), TOH (from Merck), and linoleic acid (from
Sigma) were purchased with the highest purity available and used as re-
ceived. 2,2’-Azobis(2-methylpropioamidine) dihydrochloride (AAPH;
from Aldrich) was used as received. The surfactants SDS and CTAB
were recrystallized from ethyl alcohol and acetone/water (9:1), respec-
tively.
Determination of linoleic acid hydroperoxide quantities: Aliquots of the
reaction mixture were taken out of an open vessel at appropriate time in-
tervals and subjected to HPLC analysis on a Gilson liquid chromato-
graph with
a ZORBAX ODS reversed-phase column (6ꢁ250 mm,
Du Pont Instruments), then eluted with methanol/water (9:1), for the ex-
periments conducted in homogeneous solution, or with methanol/water
(5:1) for the experiments conducted in micelles. The flow rate was set at
1.0 mLminꢀ1. A Gilson 116 UV detector was used to monitor the total li-
noleic acid hydroperoxides at 235 nm. Every determination was repeated
three times and the experimental deviations were within ꢃ10%.
Determination of a-tocopherol quantities: The procedure was the same
as that described above for the determination of linoleic acid hydroper-
[7] L. Fremont, Life Sci. 2000, 66, 663–671.
oxides, except that
a Gilson 142 electrochemical detector set at +
[8] a) A. Mertens, P. Holvoet, FASEB J. 2001, 15, 2073–2084; b) W. A.
Pryor, Free Radical Biol. Med. 2000, 28, 1681–1682; c) J. M. Upston,
L. Kritharides, R. Stocker, Prog. Lipid Res. 2003, 42, 405–422; d) J.
Vaya, S. Mahmood, A. Goldblum, M. Aviram, N. Volkova, A. Shaa-
lan, R. Musa, T. Snait, Phytochemistry 2003, 62, 89–99.
[9] a) M. V. Eberhardt, C. Y. Lee, R. H. Liu, Nature 2000, 405, 903–
904; b) B. N. Ames, L. S. Gold, W. C. Willett, Proc. Natl. Acad. Sci.
USA 1995, 92, 5258–5265; c) K. Nakachi, K. Suemasu, K. Suga, T.
Takeo, K. Imai, Y. Higashi, Jpn. J. Cancer Res. 1998, 89, 254–261;
d) B. Stavric, Clin. Biochem. 1994, 27, 319–322; e) M. G. L. Hertog,
P. C. H. Hollman, M. B. Katan, D. Kromhout, Nutr. Cancer 1993, 20,
21–29; f) B. E. Henderson, R. K. Ross, M. C. Pike, Science 1991,
254, 1131–1138.
700 mV (versus an SCE) was used for monitoring TOH. The column was
eluted with methanol/formic acid (99:1) containing sodium perchlorate
(50 mmolLꢀ1) as the supporting electrolyte for the experiment conducted
in homogeneous solution or with methanol/propan-2-ol/formic acid
(80:20:1) for the experiment conducted in micelles.
Determination of a-tocopheroxyl radical quantities: EPR spectroscopy
measurements were carried out on a Bruker ER200D spectrometer oper-
ated in the X-band with 100 kHz modulation, modulation amplitude of
0.25 mT, time constant of 0.2 s, and microwave power of 25 mW. A flat
quartz flow cell (0.4ꢁ5.5ꢁ60 mm) was used for the stopped-flow deter-
mination of the reaction kinetics as described previously.[33] The a-toco-
pheroxyl radical was generated by vigorously stirring a-tocopherol
(1 mmolLꢀ1) and excess lead oxide with a Vortex mixer for 3 min in SDS
(0.2 molLꢀ1) micelles at pH 7.4 and room temperature.
[10] Z. S. Jia, B. Zhou, L. Yang, L. M. Wu, Z.-L. Liu, J. Chem. Soc.
Perkin Trans. 2 1998, 911–915.
690
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 680 – 691

Antioxidative Effects of Flavonols and Their Glycosides
FULL PAPER
[11] a) B. Zhou, Z. S. Jia, Z. H. Chen, L. Yang, L. M. Wu, Z.-L. Liu, J.
Chem. Soc. Perkin Trans. 2 2000, 785–791; b) Z. H. Chen, B. Zhou,
L. Yang, L. M. Wu, Z.-L. Liu, J. Chem. Soc. Perkin Trans. 2 2001,
1835–1839; c) B. Zhou, Z. Chen, Y. Chen, Z. Jia, Y. Jia, L. Zeng,
L. M. Wu, L. Yang, Z.-L. Liu, Appl. Magn. Reson. 2000, 18, 397–
406.
[12] L. Ma, Z. Liu, B. Zhou, L. Yang, Z.-L. Liu, Chin. Sci. Bull. 2000, 45,
2052–2056.
[13] Z. Q. Liu, L. P. Ma, B. Zhou, L. Yang, Z.-L. Liu, Chem. Phys. Lipids
2000, 106, 53–63.
[30] W. A. Pryor, T. Strickland, D. F. Church, J. Am. Chem. Soc. 1988,
110, 2224–2229.
[31] M. H. Abraham, P. L. Grellier, D. V. Prior, J. J. Morris, P. J. Taylor, J.
Chem. Soc. Perkin Trans. 2 1990, 521–529.
[32] L. Castle, M. J. Perkins, J. Am. Chem. Soc. 1986, 108, 6381–6382.
[33] Z.-L. Liu, Z.-X. Han, K.-C. Yu, Y.-L. Zhang, Y.-C. Liu, J. Phys. Org.
Chem. 1992, 5, 33–38.
[34] F. M. Witte, J. B. F. N. Engberts, J. Org. Chem. 1988, 53, 3085–3091.
[35] O. S. Yi, D. Han, H. K. Shin, J. Am. Oil Chem. Soc. 1991, 68, 881–
883.
[36] B. Zhou, Z.-H. Chen, Z. S. Jia, L. Yang, Z. L. Liu, unpublished re-
sults.
[14] Y. J. Cai, J. G. Fang, L. P. Ma, L. Yang, Z.-L. Liu, Biochim. Biophys.
Acta 2003, 1637, 31–38.
[37] M. Foti, G. Ruberto, J. Agric. Food Chem. 2001, 49, 342–348.
[38] a) J. S. Wright, E. R. Johnson, G. A. DiLabio, J. Am. Chem. Soc.
2001, 123, 1173–1183; b) H. Y. Zhang, Y. M. Sun, X. L. Wang,
Chem. Eur. J. 2003, 9, 502–508.
[39] a) N. Kerry, C. Rice-Evans, FEBS Lett. 1998, 437, 167–171; b) H.
Tazaki, D. Taguchi, T. Hayashida, K. Nabeta, Biosci. Biotechnol.
Biochem. 2001, 65, 2613–2621.
[15] Z. Chen, B. Zhou, H. Zhu, L.-M. Wu, L. Yang, Z.-L. Liu in EPR in
the 21st Century (Eds.: A. Kawamori, J. Yamuchi, H. Ohta), Elsevi-
er, Amsterdam, 2002, pp. 421–428.
[16] J. G. Fang, M. Lu, Z. H. Chen, H. H. Zhu, Y. Li, L. Yang, L. M. Wu,
Z.-L. Liu, Chem. Eur. J. 2002, 8, 4191–4198.
[17] a) Methods in Polyphenol Analysis (Eds.: C. Santos-Buelga, G. Wil-
liamson), The Royal Society of Chemistry, Cambridge, 2003; b) K.
Herrmann, J. Food Technol. 1976, 11, 433–448.
[18] a) N. A. Poter, Acc. Chem. Res. 1986, 19, 262–268; b) N. A. Porter,
S. E. Caldwell, K. A. Mills, Lipids 1995, 30, 277–290; c) A. R.
Brash, Lipids 2000, 35, 947–952; d) K. A. Tallman, D. A. Pratt,
N. A. Porter, J. Am. Chem. Soc. 2001, 123, 11827–11828.
[19] D. A. Pratt, J. H. Mills, N. A. Porter, J. Am. Chem. Soc. 2003, 125,
5801–5810.
[20] H. Okawa, N. Ohishi, K. Yagi, J. Lipid Res. 1978, 19, 1053–1057.
[21] a) Z.-L. Liu in Bioradicals Detected by ESR Spectroscopy (Eds.:
H. O. Nishiguchi, L. Packer), Birkhauser, Basel, 1995, pp. 259–275;
b) W. A. Pryor, T. Strickland, D. F. Church, J. Am. Chem. Soc. 1988,
110, 2224–2229.
[22] L. R. C. Barclay, Can. J. Chem. 1993, 71, 1–16.
[23] G. W. Burton, T. Doba, E. J. Gabe, L. Hughes, F. L. Lee, L. Prasad,
K. U. Ingold, J. Am. Chem. Soc. 1985, 107, 7053–7065.
[24] L. R. C. Barclay, K. U. Ingold, J. Am. Chem. Soc. 1981, 103, 6478–
6485.
[25] V. W. Bowry, R. Stocker, J. Am. Chem. Soc. 1993, 115, 6029–6044.
[26] E. Niki, T. Saito, Y. Yoshikawa, Y. Yamamoto, Y. Kamiya, Bull.
Chem. Soc. Jpn. 1986, 59, 471–477.
[27] E. Niki, T. Saito, A. Kawakami, Y. Kamiya, J. Biol. Chem. 1984,
259, 4177–4182.
[40] W. Bors, C. Michel, M. Saran, Methods Enzymol. 1994, 234, 420–
429.
[41] a) S. Kitao, T. Ariga, T. Matsudo, H. Sekine, Biosci. Biotechnol. Bio-
chem. 1993, 57, 2010–2015; b) C. A. Rice-Evans, N. J. Miller, G. Pa-
ganga, Free Radical Biol. Med. 1996, 20, 933–956; c) K.-L. Liao, M.-
C. Yin, J. Agric. Food Chem. 2000, 48, 2266–2270; d) M. Shiral, J.-
H. Moon, T. Taushida, J. Terao, J. Agric. Food Chem. 2001, 49,
5602–5608.
[42] L. Hou, B. Zhou, L. Yang, Z.-L. Liu, Chem. Phys. Lipids 2004, 129,
209–219.
[43] L. Hou, B. Zhou, L. Yang, Z.-L. Liu, Org. Biomol. Chem. 2004, 2,
1419–1423.
[44] S. V. Jovanovic, S. Steenken, Y. Hara, M. G. Simic, J. Chem. Soc.
Perkin Trans. 2 1996, 2497–2504.
[45] S. V. Jovanovic, S. Steenken, M. Tosic, B. Marjanovic, M. G. Simic, J.
Am. Chem. Soc. 1994, 116, 4846–4851.
[46] S.-A. B. E. van Acker, D.-J. van den Berg, M-N. J. L. Tromp, D.-H.
Griffioen, W.-P. van Bennekom, W.-J. F. van der Vijgh, A. Bast, Free
Radical Biol. Med. 1996, 20, 331–342.
[47] C. Cren-Olivꢂ, C. Rolands in Methods in Polyphenol Analysis (Eds.:
C. Santos-Buelga, G. Williamson), The Royal Society of Chemistry,
Cambridge, 2003, pp. 157–176.
[48] G. Nonaka, O. Kawakami, I. Nishioka, Chem. Pharm. Bull. 1983, 31,
3906–3910.
[28] L. Valgimigli, J. T. Banks, J. Lusztyk, K. U. Ingold, J. Org. Chem.
1999, 64, 3381–3383.
Received: April 22, 2004
[29] V. W. Bowry, K. U. Ingold, Acc. Chem. Res. 1999, 32, 27–34.
Published online: December 2, 2004
Chem. Eur. J. 2005, 11, 680 – 691
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
691