Daidzein as an antioxidant of lipid: Effects of the microenvironment in relation to chemical structure

Article abstract of DOI:10.1021/jf801907m

Isoflavone daidzein (D, pK_a1 = 7.47 ± 0.02 and pK _a2 = 9.65 ± 0.07) was, through a study of the parent compound and its three methyl anisol derivatives 7-methyldaidzein (7-Me-D, pK_a = 9.89 ± 0.05), 4'-methyldaidzein (4'-Me-D, pK_a = 7.43 ± 0.03), and 7,4'-dimethyldaidzein (7,4'-diMe-D), found to retard lipid oxidation in liposomal membranes through two mechanisms: (i) radical scavenging for which the 4'-OH was more effective than the 7-OH group in agreement with the oxidation potentials: 0.69 V for 4'-OH and 0.92 V for 7-OH versus Ag/AgCl in acidic solution and 0.44 V for 4'-O- and 0.49 V for 7-O- in alkaline solution and (ii) change in membrane fluidity through incorporation of the isoflavones, in effect hampering radical mobility. The radical scavenging efficiency measured by the rate of the reaction with the ABTS?⁺ in aqueous solution followed the order D > 7-Me-D > 4'-Me-D > 7,4'-diMe-D, as also found for antioxidant efficiency in liposomes when oxidation was initiated with the water-soluble AAPH radical and monitored as the formation of conjugate dienes. For oxidation initiated by the lipid-soluble AMVN radical, the antioxidant efficiency was ranked as 4'-Me-D > D > 7,4'-diMe-D > 7-Me-D, and change in fluorescence anisotropy of fluorescent probes bound to the membrane surface or inside the lipid bilayer confirmed the effects of isoflavones on the membrane fluidity, especially for 7,4'-diMe-D.

Full text of DOI:10.1021/jf801907m

10376 J. Agric. Food Chem. 2008, 56, 10376–10383
Daidzein as an Antioxidant of Lipid: Effects of the
Microenvironment in Relation to Chemical Structure
†
,‡
‡
‡
‡
‡
JUN LIANG,
YU-XI TIAN, LI-MIN FU, TIAN-HE WANG, HAI-JUN LI,
‡
‡
,‡
§
PENG WANG, RUI-MIN HAN, JIAN-PING ZHANG,* AND LEIF H. SKIBSTED
State Key Laboratory for Structural Chemistry of Unstable and Stable Species, Institute of Chemistry,
Chinese Academy of Sciences, Beijing 100080, China, Department of Chemistry, Renmin University
of China, Beijing 100872, China, and Food Chemistry, Department of Food Science, Faculty of Life
Sciences, University of Copenhagen, Rolighedsvej 30, DK-1058 Frederiksberg C, Denmark
Isoflavone daidzein (D, pKa1 ) 7.47 ( 0.02 and pKa2 ) 9.65 ( 0.07) was, through a study of the
parent compound and its three methyl anisol derivatives 7-methyldaidzein (7-Me-D, pK
a
) 9.89 (
0
.05), 4′-methyldaidzein (4′-Me-D, pK ) 7.43 ( 0.03), and 7,4′-dimethyldaidzein (7,4′-diMe-D), found
a
to retard lipid oxidation in liposomal membranes through two mechanisms: (i) radical scavenging for
which the 4′-OH was more effective than the 7-OH group in agreement with the oxidation potentials:
-
0
0
.69 V for 4′-OH and 0.92 V for 7-OH versus Ag/AgCl in acidic solution and 0.44 V for 4′-O and
-
.49 V for 7-O in alkaline solution and (ii) change in membrane fluidity through incorporation of the
isoflavones, in effect hampering radical mobility. The radical scavenging efficiency measured by the
•
+
rate of the reaction with the ABTS in aqueous solution followed the order D > 7-Me-D > 4′-Me-D
7,4′-diMe-D, as also found for antioxidant efficiency in liposomes when oxidation was initiated with
>
the water-soluble AAPH radical and monitored as the formation of conjugate dienes. For oxidation
initiated by the lipid-soluble AMVN radical, the antioxidant efficiency was ranked as 4′-Me-D > D >
7,4′-diMe-D > 7-Me-D, and change in fluorescence anisotropy of fluorescent probes bound to the
membrane surface or inside the lipid bilayer confirmed the effects of isoflavones on the membrane
fluidity, especially for 7,4′-diMe-D.
KEYWORDS: Daidzein derivative; liposome; radical scavenge; antioxidation
INTRODUCTION
recent studies have also found other biochemical activities of
isoflavones, which may play important physiological roles, such
as the possible influence on intracellular redox status (11), the
interactions with specific proteins central to intracellular-
signaling cascades (12), the promotion of the cellular level of
endogenous antioxidant GSH (reduced glutathione) (13), and
the specific interactions with nuclear acids (14). Most likely,
the beneficial health effects of isoflavones derive from different
mechanisms, i.e., the antioxidation capacities and specific
biochemical interactions, for which the concentration levels
relative to other antioxidants have to be taken into account. The
present work attempts to investigate the antioxidation properties
of daidzein and its derivatives in Vitro, with an emphasis on
the influence from the environmental factors, e.g., the pH, the
disposition in lipid bilayer, and the fluidity of the membrane in
correlation to the molecular structures of the isoflavones.
Extensive studies on free-radical biology suggest that reactive
•
- 1
•
oxygen species (ROS), such as O2 , O2, and OH, are involved
in the course of aging, development of cancers, and cardiovas-
cular diseases (1). ROS may also be responsible for the toxicity
of environmental xenobiotics, some antineoplastic agents, as
well as radiation damages and may even be involved in the
mechanism behind antibiotics (2-7). The use of antioxidants
for the prevention of such effects induced by ROS has great
input on advices for human nutrition and health (8). Daidzein
(
D), an isoflavone abundant in soybean and accordingly with a
high intake in Asian populations, has been demonstrated to be
a potent antioxidant and has been receiving much attention in
relation to human health (9).
While the antioxidant properties of isoflavones, such as the
reducing capacities and the known ability to chelate transition-
metal ions (10), etc., have been relevant to the health benefits,
The antioxidation effect of isoflavones had been demonstrated
in many biologically relevant systems, such as homogeneous
solution (15), phospholipid bilayers (16, 17), low-density
lipoproteins (18, 19), and whole cells (20, 21). Studies of the
structure-activity relationship have shown that the B-ring 4′-
hydroxyl (4′-OH) is essential in scavenging free radicals,
whereas the effect of the A-ring 7-hydroxyl (7-OH) appears to
*
To whom correspondence should be addressed. Telephone:
86-10-62516604. Fax: +86-10-62516444. E-mail: jpzhang@
chem.ruc.edu.cn.
+
†
Chinese Academy of Sciences.
Renmin University of China.
University of Copenhagen.
‡
§
1
0.1021/jf801907m CCC: $40.75  2008 American Chemical Society
Published on Web 10/09/2008

Daidzein as an Antioxidant
J. Agric. Food Chem., Vol. 56, No. 21, 2008 10377
Scheme 1. Synthetic Routes of D Derivatives
be less significant (22). In homogeneous solution, the oxidation
potentials of (iso)flavones and the lifetimes of their radicals
formed by oxidation are important for the antioxidation activi-
ties. In cell membranes, liposome, and other structured media,
however, additional factors such as the distribution and orienta-
tion of (iso)flavone molecules at the surface or inside the
membranes together with any changes in the physical properties
of the membrane, owing to the incorporation of (iso)flavonoids,
also have effects on the antioxidative activity (23-25). In
addition, it had been shown that the effective concentration of
an antioxidant may vary over a 1000-fold depending upon the
type of target in the oxidation process (26). The reactivity of
an antioxidant may also vary dramatically depending upon
whether the oxidation reaction is induced by a hydrophilic or
lipophilic free-radical initiator (27). Accordingly, it is important
to consider both chemical and physical aspects in any evaluation
of antioxidant performance.
(Beijing, China). 16-(9-Anthroyloxy) palmitic acid (16-AP) was
synthesized from 16-hydroxyhexadecanoic acid and 9-anthracenecar-
boxylic acid (28).
Modification of Daidzein. Modification of D was achieved via the
Williamson reaction (29, 30). Scheme 1 shows the synthetic routes. D
(0.5 M) mixed with methyl iodide (0.5 or 1.0 M) and sodium hydroxide
(
0.5 or 1.0-1.25 M) in methanol was stirred for 24 h at 78 °C. The
reaction was traced by thin-layer chromatography (TLC). The D
derivatives, 7-methyldaidzein (7-Me-D), 4′-methyldaidzein (4′-Me-D),
and 7,4′-dimethyldaidzein (7,4′-diMe-D), were purified with TLC using
a mixture of dichloromethane and methanol (40:1, v/v) as the eluent
and were characterized by nuclear magnetic resonance (NMR) and mass
spectroscopies.
Determination of pK . The aqueous solutions of D and derivatives
a
-
5
were prepared with Britton-Robinson buffers (1.0 × 10 M) in the
pH range of 4.0-12.0, for which the ionic strength was controlled at
0
.1 using NaCl. The pH was measured with a Mettler-Toledo pH meter
(
Process Analytics, Urdorf, Switzerland), and the UV-vis absorption
spectra were recorded on a Cary 50 spectrophotometer (Varian, Walnut
Creek, CA). The solution temperature was controlled at 25 °C using a
RTE-110 thermostat (Neslab Instruments, Inc., Newington, NH) during
the measurements of pH and optical absorption. The pH was standard-
ized using international activity pH standards, and the so-called “mixed”
We have carried out, in this work, systematic studies of the
rather lipophilic D and its methyl anisol derivatives with even
more lipophilic properties for their thermodynamic properties
(
pKa and oxidation potential), radical scavenging capabilities
in homogeneous aqueous solution of varying pH, as well as
antioxidation activities in liposome. In addition, the distribution
and orientation of D and its derivatives in liposome and their
interaction with the local membranal environments were ex-
amined. When these properties and the antioxidation perfor-
manceofDandthederivativesarecompared,thestructure-activity
relationship of these antioxidants with regarding to the interac-
tions with the microenvironment is discussed. It is concluded
that both A- and B-ring hydroxyls of D play crucial roles in
radical scavenging and antioxidation activities; however, the
underlying mechanisms are different depending upon the
microenvironments.
constant was obtained from the relations, K
a
) aH+(base)/(acid) and
pK ) -log(K ), where a was measured electrochemically and the
a
a
H+
concentrations of the corresponding acids and bases were measured
spectrophotometrically.
For 7-Me-D and 4′-Me-D, the pK
experimental data by a linear regression with the relation of ApH
(Abase - ApH)/aH+ + Aacid. For D, the pKa1 and pKa2 values were
a
values were calculated from the
)
K
a
calculated on the basis of the absorbance at 336 nm using the method
described in ref 31.
Determination of Oxidation Potentials. Cyclic voltammetry was
performed on a CHI 660A electrochemical analyzer (CH Instruments,
Inc., Austin, TX) with a three-electrode configuration (32), for which
-4
D, 7-Me-D, and 4′-Me-D solutions at a concentration of 1.0 × 10
M (ethanol/water ) 1:10; 0.5 M Tris buffer with varying pH) were
used. The working electrode was a glassy carbon piece (diameter of 4
mm); the reference electrode was of the Ag/AgCl type (KCl saturated);
and the counter electrode was a platinum wire. To expel oxygen, the
MATERIALS AND METHODS
Chemicals. Daidzein (>98%) was purchased from Huike Plant
Exploitation, Inc. (Shanxi, China). 8-Anilino-1-naphthalenesulfonic acid
(
ANS), soybean L-R-phosphatidyl choline (PC), 2,2′-azino-bis(3-
solutions were bubbled with N
Radical Scavenging Assay. The radical scavenging assay was based
on the stoichiometry of the reaction of (iso)flavones with ABTS (33).
Potassium persulphate was added to a solution of 7 mM ABTS (final
concentration of 2.45 mM), which was kept at a constant temperature
2
for 10 min before measurement.
ethylbenzothiazoline-6-sulfonic) acid (ABTS), and 2,2′-azobis(2-me-
thylpropionamidine) dihydrochloride (AAPH) were from Sigma-Aldrich
Chemical Co. (St. Louis, MO). 2,2′-Azobis(2,4-dimethylvaleronitrile,
AMVN) was purchased from Huichang Petrochemical Auxiliary Co.
Ltd. (Shandong, China). Methyl iodide (>98%), methanol (analytical
grade), potassium persulfate (>99%), ethanol (HPLC grade), and
chloroform (>99%) were purchased from Beijing Chemical Plant
•
+
•
+
• +
for 12 h in the dark to generate ABTS (34). The ABTS was stable
•
+
for two days in the dark. The final concentrations of ABTS
and
-
5
-5
isoflavone were 2.5 × 10 and 1.25 × 10 M, respectively. For

10378 J. Agric. Food Chem., Vol. 56, No. 21, 2008
Liang et al.
Scheme 2. Thermolysis Generation of AMVN and AAPH Radicals
kinetics measurements, ABTS + solution and antioxidant were filled
in each syringe of a stopped-flow spectrometer (Biologic, Claix, France),
and the reaction was monitored at the characteristic absorption
•
factor, which was determined to be ∼1.
Calculation of the Dipole Moments. The geometry optimization
of the radicals from AMVN or AAPH (Scheme 2) was performed using
a density functional theory with Becher’s three-parameter hybrid
exchange functional (UB3LYP functional) at the 6-31g (d, p) level,
and the quantum chemical calculations for the dipole moments of the
radicals were performed with the Gaussian 03 program suite (37).
•
+
wavelength of ABTS
different pH were calculated following the rate equations given in ref
2.
Evaluation of Antioxidation in Liposome. Liposome was prepared
(734 nm). The initial reaction rates under
2
by using an extrusion method (35). Briefly, PC (0.68 mg) was dissolved
in absolute chloroform (3 mL), and then D or derivatives dissolved in
absolute ethanol (45 µL; the concentration of each isoflavone stock
solution was 500 µM) were added to reach an isoflavone/PC molar
ratio of 3% (the molecular mass of soybean PC was taken as 900). In
experiments with the initiation of lipid oxidation in the lipid phase,
the lipid-soluble radical generator AMVN in absolute ethanol (25 µL,
RESULTS AND DISCUSSION
pKa Values and the Oxidation Potentials of Daidzein and
Derivatives. The derivatives of D were synthesized by selective
-
-
functionalization of the 7-O or 4′-O groups (Scheme 1). The
difference between the pKa values of 7-OH (7.47) and 4′-OH
6
4.64 mM) was added. The solvent was then removed under reduced
pressure by a rotary evaporator run at a water-bath temperature of 30
C. Nitrogen gas was introduced to re-establish atmospheric pressure,
(9.65) of D is large enough to allow selective ionization of these
two groups, on the basis of which the regioselective, methyl-
substituted D derivatives were synthesized (38, 39). When 1
equiv NaOH was added, only the 7-OH of D was deprotonated
and the reaction of 7-O- with methyl iodide led to 7-Me-D,
whereas when 2 equiv NaOH was added, both hydroxyls of D
were deprotonated and 7,4′-diMe-D was obtained as the minor
product via methyl substitution, together with 4′-Me-D as the
major one.
°
and the flask was covered with aluminum foil. Then, an oil-free vacuum
pump was used to maintain the flask vacuum at <0.5 mmHg for >1 h.
The lipid residue was rehydrated with phosphate buffer (5 mL, 10 mM,
pH 7.4). The flask was then shaken while being sonicated for 1 min,
producing a homogeneous white suspension of multilamellar liposomes.
Large unilamellar liposomes were obtained by pushing the multilamellar
liposome solution through the polycarbonate membrane with 100 nm
sieve pores (Whatman, Maidstone, U.K.) for 5 repeats, and then the
water-soluble radical generator AAPH (25 µL, 150 mM) in sodium
phosphate buffer (pH 7.4) was added for the initiation of lipid oxidation
from the aqueous phase.
Lipid peroxidation was followed by monitoring the formation of
conjugated dienes using the absorbance change at 234 nm (A234). The
unilamellar liposome suspension (3.5 mL) was pipetted into a quartz
cuvette, thermostatted at 43 °C with a RTE-110 thermostat (Neslab
Instruments, Inc., Newington, NH), and the change of A234 was
monitored on a Cary 50 spectrophotometer (Varian, Walnut Creek, CA).
The lag phase was determined as the evolution time to the point where
a tangent to the propagation phase intercepted that of the initial phase
with little or no oxidation (35).
The pH dependence of the UV-vis spectra of D shown in
Figure 1A can be accounted for by the two acid/base equilibria
of the 7- and 4′-OH groups. The pKa values determined from
Figure 1B for 7-Me-D (9.89 ( 0.05) and from Figure 1C for
4
′-Me-D (7.43 ( 0.03) facilitate the assignment of the two pKa
values of D, i.e., pKa1 ) 7.47 ( 0.02 to the A-ring 7-OH and
pKa2 ) 9.65 ( 0.07 to the B-ring 4′-OH. Notably, the pKa1 of
D is almost identical to the pKa of 4′-Me-D, indicating that the
transformation of the B-ring to an anisole does not affect the
deprotonation of the A-ring phenol. On the other hand, the pKa2
of D is slightly lower than the pKa of 7-Me-D, suggesting that
the negative charge left by the dissociation of the first proton
enhances the ionization of the B-ring phenol probably through
an extension of conjugation in the fully deprotonated form of
D.
Evaluation of Membrane Fluidity. The effects of D and derivatives
on membrane fluidity were evaluated using the fluorescent probes 16-
AP (23) and ANS (36). To evaluate the fluidity of membrane surfaces,
an isoflavone (4.5 µM) was added to the liposome suspension (150
µM), and ANS in the stock phosphate buffer solution (10 mM, pH
The oxidation potentials of D and derivatives in buffer with
varying pH were determined as the spiking potentials from their
cyclic voltamograms, and the results as are listed in Table 1.
The oxidation potentials of D at pH 5.0 and 11.5 can be attribute
7.4) was added to yield a final concentration of 25 µM. The preparation
was incubated at 43 °C for 5 min. Fluorescence spectra (420-600 nm)
were measured on a LS-55 luminescence spectrophotometer (Perkin-
Elmer, Beaconsfield, U.K.), for which ANS was excited at 400 nm to
minimize the isoflavone absorption. To evaluate the influence of D and
derivatives on the fluidity of the interior of lipid bilayer, 16-AP was
added to the PC solution during the preparation of the liposome (final
concentration of 0.5 µM; 16-AP/PC molar ratio of 1:300) and the
preparations were then obtained by adding 4.5 µM isoflavone and
incubating at 43 °C for 5 min. Fluorescence polarization (P) was
determined at 415 nm (excited at 380 nm) according to the relation
-
to 4′-OH (0.69 V) and 4′-O (0.44 V), respectively, on the basis
of a comparison to the oxidation potentials of 7-Me-D having
only 4′-OH oxidizable. At pH 8.5, approximately 87% of the
7
-OH of D are deprotonated (pKa1 ) 7.47; see the inset of
Figure 1A), and the oxidation potential of 0.49 V for D at this
pH can be attribute to 7-O by comparison to the value of 0.50
-
V for 4′-Me-D at pH 11.5.
The above results prove that, for D at pH 7.4, the 7- and 4′-OH
groups are both important as radical scavengers, because about a
(
27)
P ) (I - G · I )/(I + G · I )
(1)
-
|
⊥
|
⊥
half of the 7-OH are in anionic form (7-O ) and the reducing power
-
of 7-O (0.49 V) is high enough to compete with that of the 4′-
where I and I , respectively, represent the fluorescence intensities
| ⊥
measured with the emission polarization parallel and vertical to the
excitation polarization and G stands for the instrumental polarization
OH (pKa ) 9.65; 99% in neutral form with a potential of 0.69 V).
Interestingly, for D in acidic form for which both 7-OH and 4′-

Daidzein as an Antioxidant
J. Agric. Food Chem., Vol. 56, No. 21, 2008 10379
Figure 1. Room-temperature UV-vis spectra of (A) D, (B) 7-Me-D, and
• +
Figure 2. (A-C) Room-temperature stopped-flow kinetics of ABTS
(C) 4′-Me-D in aqueous solution with varying pH under an ionic strength
scavenged by D, 4′-Me-D, 7-Me-D, and 7,4′-diMe-D in aqueous solutions
of indicated pH values.
of 0.1. Arrows indicate the increase of pH. Acid-base fractional distribution
diagrams are shown in the insets.
Radical Scavenging and Antioxidant Activities of Daidzein
and Derivatives. Scavenging of the metastable ABTS by D
Table 1. Oxidation Potentials (versus Ag/AgCl; in Aqueous Phase,
0
• +
Ag/AgCl E ) 0.2223 V/SHE) for D and Derivatives in Aqueous Solution
-
4
a
and derivatives in aqueous solution were evaluated using the
stopped-flow technique, and the kinetic traces are shown in
Figure 2. A comparison of parts A-C of Figure 2 reveals the
(
Water/Methanol ) 10:1; 1 × 10 M) of Varying pH at 25 °C
oxidation potential (V)
•
+
effect of pH, i.e., the decay of ABTS
in the presence of
pH
D
7-Me-D
4′-Me-D
isoflavones became faster with increasing pH, which may be
explained by the decreased oxidation potentials of D upon
deprotonization (Table 1). A similar tendency in the decay
kinetics is seen in the absence of isoflavones (blank), which
5
.0
0.69 (4′-OH)
0.49 (7-O )
0.44 (4′-O )
0.71 (4′-OH)
0.92 (7-OH)
-
8.5
-
-
0.50 (7-O-)
11.5
0.46 (4′-O )
a
• +
-
Assignments of oxidation potentials to specific hydroxyls are given in
may be due to the reduction of ABTS
by OH in the
parentheses.
phosphate buffer at a high pH. The following two reactions,
respectively, have been assigned as rate-determining for the
radical scavenging by phenols under acidic and basic conditions
OH should be oxidizable, only oxidation of one phenol, i.e., 4′-
OH with the lower oxidation potential, could be seen in the cyclic
voltammagram. For D at pH 8.5, only oxidation of 7-O- was
(
40):
-
2-
-
ABTS•+ + D-OH f ABTS + D-O• + H
ABTS•+ + D-O f ABTS + D-O•
+
detected, while for D at pH 11.5, only oxidation of 4′-O was
detected. These observations suggest that charge distribution in the
D radical formed by oxidation of one phenol strongly affect the
oxidation property of the other. Thus, it may be concluded that
the D radical following one-electron transfer reaction becomes
strongly stabilized by resonance, and as the result, further oxidation
is hampered.
(2a)
(2b)
-
At pH 5.0 when D or derivatives stay mostly in their neutral
forms, the initial rates for the scavenging reactions in Table 2
show that 7-Me-D (B-ring phenol) has a much higher rate of
scavenging than 4′-Me-D (A-ring phenol). At higher pH 7.4

10380 J. Agric. Food Chem., Vol. 56, No. 21, 2008
Liang et al.
Table 3. LP Determined by Spectrophotometry of the Conjugated Dienes
(cf. Figure 3) as a Result of the Oxidation of Soybean Phosphatidylcholine
Induced at 43 °C by Radical Initiators in Aqueous Phase (AAPH) or Lipid
Phase (AMVN)
lag phase (min)
sample
blank
AAPH (water-soluble)
AMVN (lipid-soluble)
13.6
20.1
18.9
17.1
15.6
146.8
192.6
158.4
251.7
173.3
D
7
4
7
-Me-D
′-Me-D
,4′-diMe-D
Scheme 3. Molecular Structures of Fluorescent Probes
7.4), i.e., D > 7-Me-D > 4′-Me-D > 7,4′-diMe-D. This is rather
reasonable in view of the oxidation potentials and the number
of hydroxyl groups of the isoflavones. We had confirmed that
the antioxidation activities of D and derivatives with the same
AAPH concentration but slightly lower isoflavone concentration
(3.0 µM), and the same rank held though the differences among
the LPs of different isoflavones were smaller. To our surprise,
as seen from Figure 3B and Table 3 for the oxidation initiated
by AMVN in the lipid phase, 4′-Me-D is the most efficient
antioxidant and the rank becomes 4′-Me-D > D > 7,4′-diMe-D
Figure 3. Change in absorbance at 234 nm of liposomes containing 4.5
µM antioxidants with (A) water-soluble AAPH (750 µM) or (B) lipid-soluble
AMVN (320 µM) as the initiators of lipid oxidation. The inset in A
schematically shows the method of lag-phase (LP) determination.
>
7-Me-D. Interestingly, 7,4′-diMe-D without any phenolic
groups seems a better antioxidant than 7-Me-D, an observation
to be explained in terms of the spatial disposition of radicals
and antioxidants in the PC bilayer as well as the heterogeneity
of the membrane fluidity induced by D and derivatives (Vide
infra).
•
+
Table 2. Initial Rates of Scavenging ABTS by D, 7-Me-D, 4′-Me-D, and
,4′-diMe-D in Aqueous Solution of Varying pH as Derived from the
Stopped-Flow Kinetics
7
rate (M s-
1
)
Interaction between Daidzein or Derivatives and Mem-
brane. To understand the antioxidation behavior of D and
derivatives in liposomes, the interactions between antioxidants
and membrane must be taken into consideration, because the
distribution and orientation of antioxidants in liposome and the
modification that they brought to the membrane fluidity would
significantly influence their antioxidation reactivities (23-25).
To investigate these effects, we made use of the fluorescent
probes ANS and 16-AP (Scheme 3), which distribute in different
regions of the membrane; i.e., ANS stays preferentially at the
surface, whereas the fluorophore of 16-AP locates mainly in
the interior of the bilayer.
The membrane surface with higher rigidity incorporates more
ANS molecules and, therefore, results in more intense fluores-
cence from ANS (41). As seen from Figure 4A, the addition
of D and 4′-Me-D to PC liposome enhanced the fluorescence
of ANS, whereas little fluorescence change by 7,4′-diMe-D and
7-Me-D was observed. Because ANS was excited at 400 nm,
where D or its derivatives have little absorption (Figure 1), the
ANS fluorescence enhancement must be induced by the increase
in ANS molecules incorporated by the membrane. These results
suggest that the concentration of D or derivatives in the
proximity of the membrane surface follows an order of D >
4′-Me-D > 7-Me-D > 7,4′-diMe-D, a sequence which agrees
with that of the increase in the lipophilicity of these isoflavones.
The fluorescence polarization of 16-AP as defined in eq 1 is
known to increase with the decrease in membrane fluidity (27).
The fluorophore of 16-AP is at the 16 position of the fatty acyl
chain (Scheme 3) and will accordingly locate at the central
sample
pH 5.0
pH 7.4
pH 8.5
-
-
-
-
-
11
-8
-6
-6
-7
-8
-6
blank
D
-Me-D
′-Me-D
,4′-diMe-D
6.0 × 10
3.5 × 10
1.9 × 10
1.5 × 10
5.1 × 10
8.0 × 10
1.3 × 10
1.4 × 10
1.1 × 10
4.3 × 10
1.5 × 10
7
7
8
9
-5
-5
-6
-6
1.1 × 10
1.0 × 10
1.4 × 10
1.1 × 10
7
4
7
(
(
8.5) when the deprotonation fraction of the 7-OH group is 50%
87%), the rate of scavenging of 7-Me-D is still higher than
-
that of 4′-Me-D, although the oxidation potential of 7-O is
lower than the significantly less deprotonated 4′-OH (Table 1).
These results lead to the conclusion that, for radical scavenging
by D in aqueous solution, the B-ring phenol is more efficient
than the A-ring one. It is intriguing to see from Figure 2 and
Table 2 that the scavenging efficiency of D, despite two
hydroxyls, is comparable to that of 7-Me-D, which can be
explained by the aforementioned one-electron oxidation mech-
anism in combination with resonance stabilization of the resulted
D radical.
Lipid peroxidation of the PC liposomes at pH 7.4 was induced
by the radicals derived from thermolysis of the lipid-soluble
AMVN or the water-soluble AAPH (Scheme 2). In either case,
the lag phase (LP; see the inset of Figure 3A) was used to
evaluate the antioxidation performance of the isoflavones.
It is seen from Figure 3A and Table 3 for oxidation triggered
by AAPH in the aqueous phase that the rank of antioxidation
efficiency of D and derivatives is consistent with that of these
•
+
compounds in scavenging of ABTS in aqueous solution (pH

Daidzein as an Antioxidant
J. Agric. Food Chem., Vol. 56, No. 21, 2008 10381
from AMVN or AAPH have the dipole moments of 3.71 or
1
.91 D, respectively, as obtained from the quantum chemical
calculations based on the chemical structures shown in Scheme
. The radical from AMVN with higher polarity is prone to be
2
adsorbed on the liposome surface, whereas that from AAPH
with considerably lower polarity may diffuse more freely within
the PC bilayer (Scheme 4), as also supported by the nearly 10-
fold reduced LP compared to the case of AMVN (Table 3).
With these in mind, we would like to discuss below the
mechanisms of retardation of AMVN- and AAPH-induced
peroxidation by the isoflavones.
Daidzein, 7-Me-D, and 4′-Me-D have been shown to scavenge
•
+
ABTS with the rates mainly associated with the oxidation
potentials and further dependent upon pH. In comparison to the
homogeneous solution, the reactivities of D and derivatives in
liposome rely on, besides their oxidation potentials, the spatial
distribution in the membrane and the membrane fluidity. For
AMVN-triggered lipid oxidation, the radical will approach
closely the surface of liposomes, where D and 4′-Me-D are
concentrated. The 4′-Me-D penetrating relatively deeper into
the membrane induces a decrease in the fluidity of the
hydrophobic domain of PC and, therefore, restricts the diffusion
of AMVN free radical, in effect decreasing lipid oxidation. This
mechanism explains why 4′-Me-D acts as the most efficient
antioxidant compared to the others under the actual conditions.
7
,4′-DiMe-D decreases the fluidity of the central region more
than 7-Me-D and, accordingly, behaves as a more efficient
antioxidant. The AAPH radical is more uniformly distributed
across the membrane, owing to its low polarity; therefore, for
lipid oxidation initiated by AAPH, the relative antioxidation
efficiency of the flavones in liposome resemble the rank of
Figure 4. (A) Fluorescence spectra of liposomes suspension containing
ANS fluorophore (25 µM) and D or derivatives (4.5 µM). (B) Fluorescence
polarization increase of 16-AP (0.5 µM; cf. eq 1) as a function of the
addition of D or derivatives (4.5 µM). Liposome preparations containing
the fluorescent probes 16-AP were incubated at 43 °C.
•
+
scavenging ABTS in homogeneous solution (D > 7-Me-D
>
4′-Me-D > 7,4′-diMe-D).
region of the PC bilayer. Because D or derivatives incorporated
in the hydrophobic region of the PC bilayer would increase the
stiffness there, the amplitude of fluorescence polarization of 16-
AP thus reflects the concentration of these compounds in its
proximity (42). As shown in Figure 4B, the order of polarization
increase induced via isoflavone incorporation follows 7,4′-
diMe-D > 7-Me-D > 4′-Me-D > D, reflecting the concentration
of these isoflavones at the central domain of the bilayer. These
results prove that 7,4′-diMe-D is predominantly in the central
region of the bilayer, whereas 7-Me-D locates mainly in the
hydrophobic region of PC and less in the membrane surface
Guo and co-authors reported, in an electron spin resonance
(ESR) study, that D in homogeneous solution with a concentra-
tion up to 1 mM exhibited no significant scavenging effects on
hydroxyl, superoxide, nitric oxide, galvinoxyl, and lipid-derived
radicals initiated with AMVN (13). The present work, however,
shows that D and derivatives at a rather low concentration (12.5
•
+
µM) can efficiently scavenge ABTS in aqueous solution and
that the protection against AMVN-initiated lipid oxidation
functions well at an even lower concentration of D (4.5 µM).
Thus, apparent discrepancies exist between our results and Guo
and co-authors’ with regard to the radical scavenging and
antioxidation activities of D. We suggest that, besides the
detection sensitivity of different spectroscopic methods, the
significant difference in the physicochemical environments
between homogeneous solution and the liposome phase may
be one of the origins.
(
Scheme 4). This tendency agrees with the aforementioned
concentration distribution reported by the ANS probe.
In liposome solution, the pKa values of the 7-OH of D or
4
′-Me-D are both close to 7.4; thus, at this value of pH, about
a half amount of 7-OH groups become ionized. As illustrated
in Scheme 4, the affinity between 7-O and the hydrophilic
-
-
-
head group of PC makes D-O and 4′-Me-D-O preferentially
adsorb to the liposome surface, creating an environment rich
in D and 4′-Me-D. Such an enrichment could limit the access
of oxidants to the bilayer and control the rate of propagation of
free-radical chain reactions occurring in the center of mem-
branes. D-O- has the additional polar group 4′-OH and will
interact more strongly with the membrane surface than a
derivative. Because 4′-Me-D is substituted with a hydrophobic
methyl group, it will insert relatively deeper toward the
hydrophobic region of the PC bilayer. On the other hand,
Preferential accumulation of D to the interface of aqueous
and lipid phases results in local enrichment of D; this may offer
effective protection against AMVN-induced lipid peroxidation
under a rather low concentration of D (4.5 µM). The same
mechanism may also be in operation in biological systems to
preserve the structure and function of biological membranes by
offering appreciable level of protection against lipid peroxidation
induced by external or internal aggressors (oxidants). It was
reported that flavonoids could protect neurons against oxidative
stress more effectively than ascorbate, even when the latter was
kept at a 10-fold higher concentration, and that the antioxidation
activity of flavonoids was relevant to their specific structural
characteristics, particularly to those defining their reduction
potentials and partition coefficients and those affecting their
interactions with the microenvironments (44). In a dietary
7
-Me-D and 7,4′-diMe-D are present in the central region, with
7
-Me-D interacting relatively stronger with the membrane
surface because of its polar 4′-OH group.
AMVN and AAPH on thermolysis decompose into free
radicals inducing lipid oxidation (43). The radicals generated

10382 J. Agric. Food Chem., Vol. 56, No. 21, 2008
Liang et al.
Scheme 4. Schematic Illustration of the Distribution of D and Derivatives and the Radicals Derived from AMVN (Tilted 4) or AAPH (0) in the PC Bilayer
of the Liposome
intervention study subjected to model animals, consumption of
soya protein products containing D led to its mean plasma level
of 0.32 µM. In addition, the bioaccumulation of D into low-
density lipoprotein (LDL) particles and into lipophilic atheromal
plaques associated with the arterial wall was suggested (9).
Therefore, the antioxidation activity of (iso)flavone, influenced
by its redox property and its interaction with the microenvi-
ronment, may be of significant physiological importance in ViVo.
LITERATURE CITED
(
1) Thomas, M. J. The role of free radicals and antioxidants. Nutrition
000, 16, 716–718.
2
(
2) Cervellati, R.; Renzulli, C.; Guerra, M. C.; Speroni, E. Evaluation
of antioxidant activity of some natural polyphenolic compounds
using the Briggs-Rauscher reaction method. J. Agric. Food Chem.
2
002, 50, 7504–7509.
(
3) Block, G.; Patterson, B.; Subar, A. Fruit, vegetables and cancer
prevention: A review of epidemiological evidence. Nutr. Cancer
1
992, 18, 1–29.
CONCLUSIONS
(
4) Burr, M. L. Antioxidants and cancer. J. Hum. Nutr. Diet 1994, 7,
The present work has demonstrated that the microenviron-
ments of D or derivatives influence the efficiency of these
isoflavones as antioxidants. In homogeneous solution, the
fractional acid/base forms and the scavenging rate of radicals,
409–416.
(5) Maxwell, S. R. J.; Thomason, H.; Sandler, D.; Leguen, C.; Baxter,
M. A.; Thorpe, G. H. G.; Jones, A. F.; Barnett, A. H. Antioxidant
status in patients with uncomplicated insulin-dependent and non-
insulin dependent diabetes mellitus. Eur. J. Clin. InVest. 1997,
27, 484–490.
(6) Acheson, R. M.; Williams, D. R. R. Does consumption of fruit
and vegetables protect against stroke? Lancet 1983, 1, 1191–1193.
7) Kohanski, M. A.; Dwyer, D. J.; Hayete, B.; Lawrence, C. A.;
Collins, J. J. A common mechanism of cellular death induced by
bactericidal antibiotics. Cell 2007, 130, 797–810.
8) Cross, C. E.; Halliwell, B.; Borish, E. T.; Pryor, W. A.; Ames,
B. N.; Saul, R. L.; McCord, J. M.; Harman, D. Oxygen radicals
and human disease. Ann. Int. Med. 1987, 107, 526–545.
9) Vedavanam, K.; Srijayanta, S.; O’Reilly, J.; Raman, A.; Wiseman,
H. Antioxidant action and potential antidiabetic properties of an
isoflavone-containing soyabean phytochemical extract (SPE).
Phytother. Res. 1999, 13, 601–608.
10) Fraga, C. G.; Oteiza, P. I. Iron toxicity and antioxidant nutrients.
Toxicology 2002, 180, 23–32.
11) Williams, R. J.; Spencer, J. P. E.; Rice-Evans, C. A. Flavonoids:
Antioxidant or signaling molecules. Free Radical Biol. Med. 2004,
•
+
such as ABTS , depend upon pH. The 4′-OH associated with
the B-ring of D was found to be the key functional group for
radical scavenging, in agreement with its low oxidation potential.
In liposome, significant differences are found in the spatial
distribution of D and derivatives, which together with the
modification of membrane fluidity by the incorporation of these
compounds considerably affect their antioxidation efficiency.
Deprotonation of 7-OH enables D to interact with the polar head
of PC and, therefore, to approach the membrane surface, while
the insert of the isoflavone B-ring in the hydrophobic domain
of the bilayer restricts the diffusion of free radicals therein
because of radical scavenging. In conclusion, the antioxidative
effect of D or derivatives as examples of isoflavones compre-
hensively depends upon the rate with which they may scavenge
the free radicals involved in lipid oxidation, their spatial
distribution in the lipid bilayer, and the local modification of
the membrane fluidity that they induced.
(
(
(
(
(
3
6, 838–849.
(
12) Jiang, B.; Liu, J. H.; Bao, Y. M.; An, L. J. Hydrogen peroxide-
induced apoptosis in pc12 cells and the protective effect of
puerarin. Cell Biol. Int. 2003, 27, 1025–1031.
ABBREVIATIONS USED
AAPH, 2,2′-azobis(2-methylpropionamidine) dihydrochloride;
•
+
ABTS , radical cation of 2,2′-azino-bis(3-ethylbenzothiazoline-
(13) Guo, Q.; Rimbach, G.; Moini, H.; Weber, S.; Packer, L. ESR
and cell culture studies on free radical-scavenging and antioxidant
activities of isoflavones. Toxicology 2002, 179, 171–180.
6
-sulfonic); AMVN, 2,2′-azobis(2,4-dimethylvaleronitrile); ANS,
8-anilino-1-naphthalenesulfonic acid; 16-AP, 16-(9-anthroyloxy)
palmitic acid; D, daidzein; GSH, reduced glutathione; 7-Me-
D, 7-methyl daidzein; 4′-Me-D, 4′-methyl daidzein; 7,4′-diMe-
D, 7,4′-dimethyl daidzein; P, polarization; PC, soybean L-R-
phosphatidyl choline; ROS, reactive oxygen species.
(14) Lin, C. C.; Tsai, Y. L.; Ho, C. T.; Teng, S. C. Determination of
the differential estrogenicity of isoflavones by E -ER-ERE-
2
dependent gene expression in recombinant yeast and MCF-7
human breast cancer cells. Food Chem. 2007, 108, 719–726.
15) Torres, J. L.; Carreras, A.; Jim e´ nez, A.; Brillas, E.; Torrelles, X.;
Rius, J.; Juli, L. Reducing power of simple polyphenols by
electron-transfer reaction using a new stable radical of the PTM
series, Tris (2,3,5,6-tetrachloro-4-nitrophenyl) methyl radical. J.
Org. Chem. 2007, 72, 3750–3756.
(
ACKNOWLEDGMENT
We thank Prof. S.-X. Dou at the Institute of Physics (CAS)
for his assistance in stopped-flow measurements.

Daidzein as an Antioxidant
J. Agric. Food Chem., Vol. 56, No. 21, 2008 10383
(
16) Arora, A.; Nair, M. G.; Strasburg, G. M. Antioxidant activities
of isoflavones and their biological metabolites in a liposomal
system. Arch. Biochem. Biophys. 1998, 356, 133–141.
17) Decharneux, T.; Dubois, F.; Beauloye, C.; Wattiaux-De, C. S.;
Wattiaux, R. Effect of various flavonoids on liposomes subjected
to an oxidative or an osmotic stress. Biochem. Pharmacol. 1992,
(32) Luo, H. X.; Shi, Z. J.; Li, N. Q.; Gu, Z. N.; Zhuang, Q. K.
Investigation of the electrochemical and electrocatalytic behavior
of single-wall carbon nanotube film on a glassy carbon electrode.
Anal. Chem. 2001, 73, 915–920.
(33) Sekher Pannala, A.; Chan, S. T.; O’Brien, P. J.; Rice-Evans, C. A.
Flavonoid B-ring chemistry and antioxidant activity: Fast reaction
kinetics. Biochem. Biophys. Res. Commun. 2001, 282, 1161–1168.
(34) Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M;
Rice-Evans, C. A. Antioxidant activity applying an improved
ABTS radical cation decolonization assay. Free Radical Biol. Med.
(
4
4, 1243–1248.
(
18) Hwang, J.; Hodis, H. N.; Sevanian, A. Soy and alfalfa phytoestro-
gen extracts become potent low-density lipoprotein antioxidants
in the presence of acerola cherry extract. J. Agric. Food Chem.
1
999, 26, 1231–1237.
2
001, 49, 308–314.
(
35) Roberts, W. G.; Gordon, M. H. Determination of the total
antioxidant activity of fruits and vegetables by a liposome assay.
J. Agric. Food Chem. 2003, 51, 1486–1493.
(
(
19) Hwang, J.; Sevanian, A.; Hodis, H. N.; Ursini, F. Synergistic
inhibition of LDL oxidation by phytoestrogens and ascorbic acid.
Free Radical Biol. Med. 2000, 29, 79–89.
20) Wei, H.; Zhang, X.; Wang, Y.; Lebwohl, M. Inhibition of
ultraviolet light-induced oxidative events in the skin and internal
organs of hairless mice by isoflavone genistein. Cancer Lett. 2002,
(
36) Jagannadham, M. V.; Rao, V. J.; Shivaji, S. The major carotenoid
pigment of a psychrotrophic Micrococcus roseus strain: Purifica-
tion, structure, and interaction with synthetic membranes. J.
Bacteriol. 1991, 173, 7911–7917.
1
85, 21–29.
(37) Frisch, M. J.; et al. Gaussian 03W, Revision B.01; Gaussian, Inc.,
(
21) Jiang, B.; Liu, J. H.; Bao, Y. M.; An, L. J. Hydrogen peroxide-
induced apoptosis in pc12 cells and the protective effect of
puerarin. Cell Biol. Int. 2003, 27, 1025–1031.
Pittsburgh, PA, 2003.
(
38) W a¨ h a¨ l a¨ , K.; Valo, T.; Brunow, G.; Hase, T. Monoalkylation of
-
D (7,4′-dihydroxyisoflavone): Synthesis of 7-O (carboxybutyl)
equal. Finn. Chem. Lett. 1989, 16, 79–83.
(
22) Han, R. M.; Tian, Y. X.; Becher, E. M.; Andersen, M. L.; Zhang,
J. P.; Skibsted, L. H. Puerarin and conjugate bases as radical
scavengers and antioxidants: Molecular mechanism and synergism
with ꢀ-carotene. J. Agric. Food Chem. 2007, 55, 2384–2391.
23) Arora, A.; Byrem, T. M.; Nair, M. G.; Strasburg, G. M.
Modulation of liposomal membrane fluidity by flavonoids and
isoflavones. Arch. Biochem. Biophys. 2000, 373, 102–109.
24) Oteiza, P. I.; Erlejman, A. G.; Verstraeten, S. V.; Keen, C. L.;
Fraga, C. G. Flavonoid-membrane interactions: A protective role
of flavonoids at the membrane surface. Clin. DeV. Immunol. 2005,
(
39) Gao, G. Y.; Li, D. J.; Keung, W. M. Synthesis of potential anti-
dipsotropic isoflavones: Inhibitors of the mitochondrial monoam-
ine oxidase-aldehyde dehydrogenase pathway. J. Med. Chem.
(
(
2
001, 44, 3321–3328.
(40) Miller, N. J.; Rice-Evans, C. A. Spectrophotometric determination
of antioxidant activity. Redox Rep. 1996, 2, 161–171.
(
41) Lenaz, G.; Bertoli, E.; Curatola, G.; Mazzanti, L.; Biqi, A. Lipid
protein interaction in mitochondria. Spin and fluorescence probe
studies on the effect of n-alkanols on phospholipid vesicles and
mitochondrial membranes. Arch. Biochem. Biophys. 1976, 172,
1
2, 19–25.
(
25) Maggio, B.; Lucy, J. A. Interactions of water-soluble fusogens
2
78–288.
with phospholipids in monolayers. FEBS Lett. 1978, 94, 301–
(
(
(
42) Gasymov, O. K.; Abduragimov, A. R.; Prasher, P.; Yusifov, T. N.;
Glasgow, B. J. Tear lipocalin: Evidence for a scavenging function
to remove lipids from the human corneal surface. InVest. Oph-
thalmol. Vis. Sci. 2005, 46, 3589–3596.
43) Barclay, L. R. C.; Antunes, F.; Egawa, Y.; McAllister, K. L.;
Mukai, K.; Nishi, T.; Vinqvist, M. R. The efficiency of antioxi-
dants delivered by liposomal transfer. Biochim. Biophys. Acta
3
04.
(
(
26) Liu, Z. L. Microenvironmental effects and synergistic effects of
bio-antioxidants. Chin. J. Org. Chem. 2001, 11, 884–889.
27) Mojumdar, S. C.; Becker, D. A.; Dilabio, G. A.; Ley, J. J; Barclay,
L. R. C.; Ingold, K. U. Kinetic studies on stilbazulenyl-bis-nitrone
(
STAZN), a nonphenolic chain-breaking antioxidant in solution,
micelles, and lipid membranes. J. Org. Chem. 2004, 69, 2929–
936.
1
997, 1328, 1–12.
2
44) Schroeter, H.; Williams, R. J.; Matin, R.; Iversen, L.; Rice-Evans,
C. A. Phenolic antioxidants attenuate neuronal cell death following
uptake of oxidized low-density lipoprotein. Free Radical Biol.
Med. 2000, 29, 1222–1233.
(
28) Tilley, L.; Thulborn, K. R.; Sawyer, W. H. An assessment of the
fluidity gradient of the lipid bilayer as determined by a set of
n-(9-anthroyloxy) fatty acids (n ) 2, 6, 9, 12, 16). J. Biol. Chem.
1
979, 8, 2592–2594.
(
(
(
29) Keung, W. M. Anti-dipsotropic isoflavones: The potential thera-
Received for review June 23, 2008. Revised manuscript received August
0, 2008. Accepted September 4, 2008. This work has been supported
by the grants-in-aid from the Natural Science Foundation of China
20673144 and 20703067) and from the Ministry of Science and
Technology of China (2009CB220008 and 2006BAI08B04-06). Support
by LMC, Centre for Advanced Food Studies to the Food Chemistry
group at the University of Copenhagen is also acknowledged.
peutic agents for alcohol dependence. Med. Res. ReV. 2003, 23,
3
6
69–696.
30) Han, R. M.; Tian, Y. X.; Wang, P.; Xiang, J. F.; Ai, X. C.; Zhang,
J. P. Synthesis and characterization of puerarin and the mechanism
of derivation reaction. Chem. J. Chin. UniV. 2006, 9, 1716–1720.
31) Albert, A.; Serjeant, E. P. Ionization Constants of Acids and Bases:
A Laboratory Manual; Methuen: London, U.K. and Wiley: New
York, 1962; pp 24-45.
(
JF801907M