Fast regeneration of carotenoids from radical cations by isoflavonoid dianions: Importance of the carotenoid keto group for electron transfer

Article abstract of DOI:10.1021/jp907349x

Electron transfer to radical cations of β-carotene, zeaxanthin, canthaxanthin, and astaxanthin from each of the three acid/base forms of the diphenolic isoflavonoid daidzein and its C-glycoside puerarin, as studied by laser flash photolysis in homogeneous methanol/chloroform (1/9) solution, was found to depend on carotenoid structures and more significantly on the deprotonation degree of the isoflavonoids. None of the carotenoid radical cations reacted with the neutral forms of the isoflavonoids while the monoanionic and dianionic forms of the isoflavonoids regenerated the oxidized carotenoid. Electron transfer to the β-carotene radical cation from the puerarin dianion followed second order kinetics with the rate constant at 25 °C k₂ = 5.5 × 10⁹ M^-1 s^-1, zeaxanthin 8.5 × 10⁹ M^-1 s^-1, canthaxanthin 6.5 × 10¹⁰ M^-1 s^-1, and astaxanthin 11.1 × 10¹⁰ M^-1 s^-1 approaching the diffusion limit and establishing a linear free energy relationship between rate constants and driving force. Comparable results found for the daidzein dianion indicate that the steric hindrance from the glucoside is not important suggesting the more reducing but less acidic 4′-OH/4′-O^- as electron donors. On the basis of the rate constants obtained from kinetic analyses, the keto group of carotenoids is concluded to facilitate electron transfer. The driving force was estimated from oxidation potentials, as determined by cyclic-voltametry for puerarin and daidzein in aqueous solutions at varying pH conditions, which led to the standard reduction potentials E° = 1.13 and 1.10 V versus NHE corresponding to the uncharged puerarin and daidzein. For pH > pk_a2, the apparent potentials of both puerarin and daidzein became constants and were E° = 0.69 and 0.65 V, respectively. Electron transfer from isoflavonoids to the carotenoid radical cation, as formed during oxidative stress, is faster for astaxanthin than for the other carotenoids, which may relate to astaxanthins more effective antioxidative properties and in agreement with the highest electron accepting index of astaxanthin.

Full text of DOI:10.1021/jp907349x

1
26
J. Phys. Chem. A 2010, 114, 126–132
Fast Regeneration of Carotenoids from Radical Cations by Isoflavonoid Dianions:
Importance of the Carotenoid Keto Group for Electron Transfer
†
†
†
†
,‡
Rui-Min Han, Chang-Hui Chen, Yu-Xi Tian, Jian-Ping Zhang, and Leif H. Skibsted*
Department of Chemistry, Renmin UniVersity of China, Beijing 100872, People’s Republic of China and Food
Chemistry, Department of Food Science, Faculty of Life Sciences, UniVersity of Copenhagen, RolighedsVej 30,
DK-1058 Frederiksberg C, Denmark
ReceiVed: July 31, 2009; ReVised Manuscript ReceiVed: September 30, 2009
Electron transfer to radical cations of ꢀ-carotene, zeaxanthin, canthaxanthin, and astaxanthin from each of
the three acid/base forms of the diphenolic isoflavonoid daidzein and its C-glycoside puerarin, as studied by
laser flash photolysis in homogeneous methanol/chloroform (1/9) solution, was found to depend on carotenoid
structures and more significantly on the deprotonation degree of the isoflavonoids. None of the carotenoid
radical cations reacted with the neutral forms of the isoflavonoids while the monoanionic and dianionic forms
of the isoflavonoids regenerated the oxidized carotenoid. Electron transfer to the ꢀ-carotene radical cation
9
-1
from the puerarin dianion followed second order kinetics with the rate constant at 25 °C k
2
) 5.5 × 10 M
-
1
9
-1 -1
10
-1 -1
10
-1 -1
s , zeaxanthin 8.5 × 10 M s , canthaxanthin 6.5 × 10
M
s , and astaxanthin 11.1 × 10
M
s
approaching the diffusion limit and establishing a linear free energy relationship between rate constants and
driving force. Comparable results found for the daidzein dianion indicate that the steric hindrance from the
glucoside is not important suggesting the more reducing but less acidic 4′-OH/4′-O as electron donors. On
-
the basis of the rate constants obtained from kinetic analyses, the keto group of carotenoids is concluded to
facilitate electron transfer. The driving force was estimated from oxidation potentials, as determined by cyclic-
voltametry for puerarin and daidzein in aqueous solutions at varying pH conditions, which led to the standard
reduction potentials E° ) 1.13 and 1.10 V versus NHE corresponding to the uncharged puerarin and daidzein.
For pH > pKa2, the apparent potentials of both puerarin and daidzein became constants and were E° ) 0.69
and 0.65 V, respectively. Electron transfer from isoflavonoids to the carotenoid radical cation, as formed
during oxidative stress, is faster for astaxanthin than for the other carotenoids, which may relate to astaxanthins
more effective antioxidative properties and in agreement with the highest electron accepting index of
astaxanthin.
Introduction
lipid interfaces.11,12 Results of theoretical calculations recently
became available that predict the relative ease of one-electron
oxidation of carotenoids to form the carotenoid radical cation
and which accordingly also predict the relative efficiency of
carotenoids as antioxidants.13 In order to study the mechanism
of electron transfer to carotenoids, we have selected four well-
characterized and structurally related carotenoids including
The positive effects on human health of a high intake of
carotenoids through a diet rich in fruits and vegetables are often
assigned to the functions of carotenoids as antioxidants.1
ꢀ
-Carotene and certain other carotenoids have provitamin A
activity, and lutein and zeaxanthin have been shown to protect
the retina of the eye against light-induced oxidative stress, but
otherwise antioxidative functions of carotenoids are still
uncertain.2
ꢀ
-carotene as the parent compound together with zeaxanthin,
canthaxanthin, and astaxanthin for a systematic study of the
effects of hydroxyl- and keto group substitution in ꢀ-carotene
on interaction with polyphenolic antioxidants, cf. Scheme 1.
Daidzein and its C-glycoside puerarin with only two phenolic
groups were selected as relative simple plant phenols for the
kinetic studies of reduction of carotenoid radical cations, since
the radical dynamics of the isoflavonoids skeleton recently has
-6
In biological systems, many factors affect the efficiency of
antioxidants, including the radical scavenging capacity of the
individual antioxidant, the actual distribution of antioxidants in
heterogeneous environment, the influence of antioxidants on
membrane structure, and the interaction with other antioxidants
present.7 It has long been known that carotenoids may interact
with other antioxidants in membranes and protect against
-9
14
been characterized.
10
development of cardio-vascular diseases and cancer. However,
the molecular basis of carotenoid interaction and possible
synergism with other types of antioxidants like polyphenols have
not been studied in any detail, although there are some
indications of electron transfer between ꢀ-carotene radical
cations formed from oxidative stress and polyphenols at water/
Materials and Methods
Chemicals. ꢀ-Carotene, zeaxanthin, and astaxanthin were
15
from the same source as previously reported. Canthaxanthin
sealed in ampules under argon was supplied by Roche A/S
16
(
Hvidovre, Denmark) and used as-received. Puerarin and
daidzein (>98%) were purchased from Huike Plant Exploitation
Inc., (Shanxi, China) and used as received. Water was supplied
by a Milli-Q apparatus (Millipore Corp., Billerica, MA).
Methanol of high-performance liquid chromatography (HPLC)
*
To whom correspondence should be addressed: Tel: +45 3528 3221.
Fax: +45 3528 3344. E-mail: ls@life.ku.dk.
†
Renmin University of China.
University of Copenhagen.
‡
1
0.1021/jp907349x  2010 American Chemical Society
Published on Web 12/03/2009

Fast Regeneration of Carotenoids from Radical Cations
J. Phys. Chem. A, Vol. 114, No. 1, 2010 127
SCHEME 1: Molecular Structures of (a) ꢀ-Carotene, (b) Zeaxanthin, (c) Canthaxanthin, (d) Astaxanthin, (e) Puerarin,
and (f) Daidzein
TABLE 1: Standard Reduction Potential of Carotenoid Radical Cations versus NHE at 25°C, Absorption Maximum (λmax) of
Carotenoid Radical Cation in near Infrared Spectral Region, The Potential Difference between Carotenoid and Puerarin/
-
1
-1
Daidzein Dianion ∆E (V), Gibbs Free Energy ∆G (kJ·mol ) and ∆∆G (kJ ·mol , relative to ∆G for of ꢀ-carotene) of the
Electron Transfer Reaction between Carotenoid Radical Cations and Puerarin or Daidzein Dianions
•
+
2-
•+
2-
E°/V(ref)
aqueous33
CH
car /puerarin
car /daidzein
15,18
max/nm20,27
e
∆G°/kJ·mol-1
e
∆G°/kJ·mol-1
∆∆G°/kJ·mol-1
carotenoid
-carotene
zeaxanthin
canthaxanthin
astaxanthin
2
Cl
2
λ
∆E /V
∆E /V
a
a
a
a
b
b
ꢀ
1.06
1.03
1.04
1.03
0.84
0.85
940
940
860
0.15
0.16
0.26
0.28
-14.48
-15.44
-25.09
-27.02
0.19
0.20
0.30
0.32
-18.34
-19.30
-28.95
-30.88
0
-0.96
-10.61
-12.54
0.95c
d
b
0.97
850
a
In Triton X-100 micelles. b Corrected from value relative to Ag/AgCl electrode by addition of 0.22 V. Corrected from value relative to
c
d
e
2-
2-
saturated calomel electrode by addition of 0.24 V. In ditert-butyl peroxide/benzene (7/3 v/v). E′(puerarin ) ) 0.69 V, E′(daidzein ) ) 0.65 V.
grade (Caledon Laboratories, Georgetown, Ontario, Canada) was
used as received. Chloroform (>99.0%, Beijing Chemical Plant)
was purified before use by passing through an alumina column
was achieved by bubbling the solution with high-purity argon
for ∼30 min. Near-infrared (NIR) kinetics at individual
wavelengths was recorded with a xenon lamp probe (CW, 350
W) and a photodiode detector (S8890-02, Hamamatsu,
Hamamatsu City, Japan).
(
Wusi Chemical Reagent Ltd., Shanghai, China). The solution
-
2-
of puerarin and puerarin were prepared by addition 1 or 2
equiv. tetramethylammonium hydroxide (CH ) N OH (97%,
3 4
+
-
Results
Sigma, Louis, USA) to puerarin neutral solutions and similar
procedures were used for daidzein. Other chemicals were all of
analytical grade.
The rate of electron transfer from puerarin or daidzein to each
of the four carotenoids ꢀ-carotene, zeaxanthin, canthaxanthin,
and astaxanthin were determined in methanol/chloroform (1/9,
v/v). The rate of electron transfer between ꢀ-carotene and
puerarin has previously been found to depend strongly on the
charge of puerarin corresponding to the three isoflavonoid acid/
Determination of Oxidation Potentials. Cyclic voltammetry
(
(
CV) was performed on a CHI 660B electrochemical analyzer
CH Instruments Inc., Austin, TX, USA) with a three-electrode
17
configuration, for which the solutions of puerarin at a
concentration of 2.0 × 10 M in Britton-Robinson(B-R) buffer
were used (pH ) 2-12). For the less water-soluble daidzein
-5
11
base forms. The change in oxidation potentials for puerarin
and daidzein corresponding to the standard reduction potentials
of their radicals upon deprotonation is accordingly important
for calculating and comparing the differences in the driving force
between carotenoid radical cations and various acid/base forms
of the isoflavonoids. The oxidation potentials of the carotenoids
corresponding to the standard reduction potentials of their radical
cations are available in literature as shown in Table 1. The
-5
(
2.0 × 10 M), the B-R buffer containing 5% dimethylsulfoxide
was used to increase the solubility. The working electrode was
a glassy carbon piece (diameter, 4 mm), the reference electrode
was of the Ag/AgCl type (KCl-saturated), and the counter
electrode was a platinum wire. The ionic strength was adjusted
to 0.1 M with NaCl.
Laser Flash Photolysis. Submicrosecond time-resolved
absorption spectra were obtained at room temperature (25 ( 1
2 2
reported potentials in CH Cl are used for the calculations for
the kinetic study of the electron transfer rather than the values
reported for aqueous solution due to the consistency between
the absorption maximum of the carotenoid radical cations in
°
C). The excitation laser pulses at 500 nm (2 mJ/pulse, 7 ns,
0 Hz) was obtained by an optical parametric oscillator (OPO,
Quanta-Ray MOPO-SL, Spectra Physics, Mountain View, CA,
1
2 2
the NIR spectral region, and the potentials determined in CH Cl
3+
15,18-21
USA) driven by an Nd :YAG pumped laser (Quanta-Ray PRO-
30; Spectra Physics). The optical path length of the flow cuvette
used for laser flash photolysis was 1 cm. The anaerobic condition
rather than those determined in micelles in aqueous solution.
2
The oxidation potential of puerarin and daidzein in aqueous
solution were determined by cyclic voltammetry for 2 < pH <

1
28 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Han et al.
(6)
12 and the standard reduction potential for one-electron oxidized
E′ ) E° - 0.060·pH
forms of the isoflavonoids were estimated by extrapolation to
pH ) 0.
both for conditions where the fully protonated form of puerarin
or the monoanion of puerarin (or daidzein) are oxidized. The pH
dependence of the oxidation potential E′ obtained from the CV
diagrams of Figure 1 is shown in Figure 2. For 2 < pH < 10, E′
is seen to depend linearly on pH, while for higher values of pH,
the value of E′ becomes independent of pH and nearly constant
Oxidation Potentials of Puerarin and Daidzein. CV scan
of puerarin and daidzein at varying pH as shown in Figure 1
indicates no reversible reduction peak after the initial oxidation
peak. The clear oxidation peaks seen are ascribed to the
oxidation of the B-ring phenolic 4′-OH group, which is more
reducing than the A-ring phenolic 7-OH group according to our
23
in agreement with what is seen for simple phenols and the
11,22
previous experimental and theoretical results.
The electrode
24
3
-hydroxyl group of catechin. For puerarin (and daidzein), this
oxidation processes during CV, the neutral and monoanionic
forms of puerarin (and daidzein), are suggested to be the
following:
corresponds to an electrode process,
-
-
-
•
-
O(7)-P-(4′)O f O(7)-P-(4′)O + e
(7)
•
+
-
HO(7)-P-(4′)OH f HO(7)-P-(4′)O + H + e
for which the following equation is valid.
(1)
[-
[-
O(7)-P-(4′)O ]
-
-
-
•
+
-
O(7)-P-(4′)OH f O(7)-P-(4′)O + H + e
E ) E° - 0.060·log
(8)
•
O(7)-P-(4′)O ]
(2)
From the linear relationship predicted by eq 6 applied to the
results of puerarin and daidzein (Figure 2)
in which P represents the isoflavonoid skeleton, and for which
the Nernst equation at 25 °C yields,
E′ ) E° - R·pH
(9)
[
HO(7)-P-(4′)OH]
E ) E° - 0.060·log
(3)
(4)
•
+
[
HO(7)-P-(4′)O ][H ]
the standard reduction potential of the fully protonated puerarin
is estimated to be E° ) 1.13 V versus NHE (0.91 ( 0.01 V
relative to Ag/AgCl) together with the slope R ) 0.046 ( 0.002,
while for neutral daidzein E° ) 1.10 V versus NHE (0.88 (
[-
O(7)-P-(4′)OH]
E ) E° - 0.060·log
[-
O(7)-P-(4′)O ][H ]
•
+
0
.01 V relative to Ag/AgCl) and R ) 0.049 ( 0.002 are
obtained. For pH > 10, the apparent potential is E′ ) 0.47 V
for puerarin and 0.43 V for daidzein as seen from in Figure 2,
parts a and b, which are ascribed to the oxidation potential of
puerarin and daidzein dianions and are corrected to 0.69 and
Rearranging further yields,
E′ ) E° - 0.060·pH - log Q
(5)
0
.65 V versus NHE, respectively.
The 7-phenolate and the 4′-phenol coexist at an intermediate
in which Q for reaction 1 is [HO(7)-P-(4′)OH]/[HO(7)-P-
pH interval between pKa1 ) 7.20 (7-OH) and pKa2 ) 9.84 (4′-
OH) for puerarin and between pKa1 ) 7.47 and pKa2 ) 9.65
for daidzein, and it is not possible using the present experimental
methods to differentiate between the initial oxidation of the
7-phenolate (i) as shown in eq 10 from the initial oxidation of
•
-
-
(
4′)O ] and for reaction 2 is [ O(7)-P-(4′)OH]/[ O(7)-P-
•
(
4′)O ], and from which it is further seen that the value of the
apparent potential measured at varying pH, E′, is related to the
standard reduction potential as follows:
-
5
Figure 1. Cylic voltammogram of 2.0 × 10 M (a) puerarin and (b) daidzein at varying pH in aqueous solution of ionic strength 0.1 adjusted with
-
1
NaCl at a scan rate of 100 mV ·s at room temperature (∼25°). For daidzein, 5% dimethylsulfoxide was used to increase the solubility.

Fast Regeneration of Carotenoids from Radical Cations
J. Phys. Chem. A, Vol. 114, No. 1, 2010 129
SCHEME 2: Acid/Base Equilbria of Puerarin in
Aqueous Solutiona
Figure 2. Oxidation potential of 4′-hydroxyl (O) and glucoside (0)
-5
-5
for (a) puerarin (2 × 10 M) and (b) daidzein (2 × 10 M) in aqueous
solution at varying pH as determined by cyclic voltametry at room
temperature (∼25°).
the 4′-phenol at the same pH or in oxidation parallel processes.
However, oxidation of the 7-phenolate (i) leads to a fast
5
-1
intramolecular electron transfer with an rate constant ∼10 s
from the 7-phenoxyl radical (ii) to the 4′-phenoxyl radical (iii)
as also formed by direct oxidation of the 4′-phenol in eq 7.
Both the two reaction paths correspond accordingly to the same
overall oxidation process.
a
K ) 8.3 × 10-22, is calculated from the difference of the
deprotonation energy (DE) for 7-OH and 4′-OH from ref 11.
versus Ag/AgCl is estimated with R ) 0.047 ( 0.002 V by
extrapolating to pH ) 0 and corrected to yield the value of E°
)
1.47 V versus NHE.
Reduction of Carotenoid Radical Cations by Isoflavonoids.
Upon direct photoexcitation to the optically allowed S excited
2
state with 500 nm laser pulse, carotenoids will form radical
cations in an electron withdrawing solvent such as chloro-
form.1
5,26-28
A mixture of chloroform and methanol was used
-
6
in which both carotenoids (5 × 10 M) and isoflavonoids
-
5
(
∼10 M) are sufficiently soluble. The kinetics of the decay
of the carotenoid radical cations initially formed by nanosecond
laser flash photolysis was recorded in real time by NIR detection
as previously described.11 It was found that the kinetics of the
four carotenoids was hardly affected by the presence of neutral
puerarin or daidzein as shown for astaxanthin in Figure 3. The
decay of carotenoid radical cations were in contrast significantly
accelerated by the presence of monoanions and especially
dianions of puerarin and daidzein as seen in Figure 3 for
astaxanthin.
The deviation from the Nernst equation observed for both
puerarin and daidzein with the slope of ∼0.05 rather than the
theoretical value of 0.06 as found for catechol-containing
flavonoids such as catechin and quercetin,24,25 may be due
to the acid/base equilibria shown in Scheme 2. The most
reducing form of puerarin (and daidzein) monoanion is the
HO(7)-P-(4′)O- form, which is present in extremely low
concentration, and kinetic aspects related to the acid/base
equilibria may become important. From the bond dissociation
energy (BDE) reported in ref 22 the equilibrium constant of
the following equation:
In Figure 4, the kinetics for decay of the four carotenoid
radical cations in the presence of excess puerarin and daidzein
dianions are presented. The relatively slow decay of the
carotenoid radical cation in the absence of isoflavonoids (Figure
3
) has been found for ꢀ-carotene to follow second-order
•
•
HO(7) - P - (4′)O h O(7) - P - (4′)OH (11)
11
kinetics. For the dianion form of puerarin and daidzein,
carotenoid radical cations decay monoexponentially, and the
pseudofirst order rate constant for decay depends on the
concentration of (excess) puerarin or daidzein, as shown in
Figure 5 for ꢀ-carotene radical cation and puerarin anion. The
initial carotenoid radical cation concentration is low compared
to the puerarin dianion concentration given the ∼10% quantum
is calculated to be K7,4′RAD ) 1.4 × 10 , and if the deprotonated
-6
7
-OH is oxidized it will rearrange also complicating the
electrode kinetics. In Scheme 3, the acid/base equilibria are
combined with the one-electron oxidation reactions showing the
most important reaction paths.
For puerarin, a new peak at higher potential appears for 8.5
2
yield for its formation from the S excited state ensuring the
pseudofirst order reaction conditions.15 The pseudofirst order
rate constant depends not strictly linearly on the concentration
of puerarin or daidzein dianions, as shown in Figure 5 for
ꢀ-carotene radical cation and puerarin dianion. This kind of
nonlinear deviation for the higher puerarin dianion concentration
may be ascribed to association between the carotenoid radical
<
pH < 12, which shows the same pH-dependence for E′ as the
initial oxidation at lower potential shown in Figure 2a. Notably,
this new peak at higher potential is not observed for daidzein
(vide infra) and could accordingly be ascribed to the involvement
of glucoside moiety of puerarin. From the measured E′ at 8 <
pH < 12 shown in Figure 2, a value of E° ) 1.25 ( 0.02 V

1
30 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Han et al.
SCHEME 3: One-Electron Oxidation/Reduction of Acid/Base Forms of Puerarin in Aqueous Solution
cation and the puerarin dianion prior to or in parallel with
electron transfer as presented in the following equation,29
daidzein dianions collected in Table 2 were based on experi-
ments with low dianion concentration. A similar kinetics analysis
was not attempted for the puerarin/daidzein monoanions due
to the complicated acid/base equilibria as shown in Scheme 2
with simultaneous presence of neutral, mono- and dianionic
forms during reaction with the carotenoid radical cation.
Car•+ + Pue h {Car , Pue } f Car + Pue
2-
•+
2-
•-
(12)
In order to simplify the kinetic analysis, the pseudofirst order
rate constants were, however, converted to second-order rate
constants from the initial slope of the pseudofirst order rate
constants as a function of puerarin dianion concentration as
shown in Figure 5. The second-order rate constants for reaction
between the three other carotenoid radical cations and puerarin/
Discussion
Structural aspects of interactions between different antioxi-
dants are important for understanding antioxidant synergism.
Carotenoids are known to be regenerated from their radical
cations formed during oxidative stress by tocopherols and
tocotrienols, depending on their structures.16 More recently,
flavonoids also have been found to regenerate ꢀ-carotene from
its radical cations as anionic forms of the isoflavonoid puerarin
and of the flavonoid baicalin, found to reduce the radical
cation.1
1,12
For regeneration of carotenoids from their radical
cations by other carotenoids, an antioxidant hierarchy has been
established in which astaxanthin ranks as the least efficient,
while lycopene, ꢀ-carotene, and zeaxanthin are among the most
efficient and comparable with the tocopherols.1 The low rank
in this hierarchy of especially astaxanthin is remarkable, since
astaxanthin has been found to interact with R-tocopherol in
6,30
31
salmon, which is being considered as a protector in carcino-
gensis5 and is demonstrated in a model system to be an
efficient antioxidant.3
Figure 3. Time trace of astaxanthin radical cation at 900 nm measured
,32
by transient absorption spectroscopy following a laser flash photolysis
3,34
-
6
of astaxanthin (5 × 10 M) in methanol/chloroform (1/9) at 25° in
-5
presence of 1.2 × 10 M neutral, mono- and dianionic forms of (a)
puerarin and (b) daidzein.
Figure 4. Time trace of carotenoid radical cations measured by
transient absorption spectroscopy following laser flash photolysis of
Figure 5. Observed pseudofirst order rate constant for electron transfer
from puerarin dianion to ꢀ-carotene radical cation. From the initial
-
6
solutions of carotenoid (5 × 10 M) in methanol/chloroform (1/9) at
-
5
9
-1 -1
2
5° in presence of 1.2 × 10 M dianion of (a) puerarin and (b)
slope, a second order rate constant k
2
) (5.5 ( 0.3) × 10 l·mol ·s
daidzein.
is calculated.

Fast Regeneration of Carotenoids from Radical Cations
J. Phys. Chem. A, Vol. 114, No. 1, 2010 131
-5
TABLE 2: Second Order Rate Constants for Electron Transfer from Puerarin Dianion (1.2 × 10 M) and Daidzein Dianion
-5
#
-1
(
1.2 × 10 M) to Carotenoid Radical Cation in Methanol/Chloroform (1/9) and Activation Free Energy ∆G (kJ·mol ) at 25°
puerarin2
-
daidzein2-
k (l·mol- ·s-1
1
)
∆G /kJ·mol
#
-1
k (l·mol-1 ·s-1
)
∆G /kJ·mol
#
-1
carotenoid
9
9
ꢀ
-carotenea
zeaxanthin
(5.5 ( 0.1) × 10
(8.5 ( 0.1) × 10
(6.5 ( 0.1) × 10
(11.1 ( 0.1) × 10
17.4
16.3
11.3
10.0
(5.8 ( 0.3) × 10
(8.3 ( 0.2) × 10
(5.7 ( 0.1) × 10
(9.2 ( 0.1) × 10
17.3
16.4
11.6
10.4
a
9
9
b
10
10
10
10
canthaxanthin
astaxanthin
a
a
Reaction followed at 950 nm. b Reaction followed at 900 nm.
the more important group for electron transfer than the hydroxyl.
The presence of both hydroxyl and keto groups as in astaxanthin
therefore yields the highest rate of reaction. As for the
isoflavonoids, the rate of puerarin dianion reacting with caro-
tenoid radical cations is quite similar to that of daidzein dianion,
and the presence of glucoside in puerarin is not resulting in
any steric hindrance. The 4′-phenolate group, which notably is
the more reducing in both isoflavonoids, may be identified as
electron donor. The π-bond of the keto group of the carotenoid
δ-
δ+
is polarized according to O ) C accelerating the interaction
with the phenolate groups of the isoflavonoids.
#
Figure 6. Activation free energy ∆G as a function of ∆∆G°, reaction
free energy relative to reaction with ꢀ-carotene radical cation, for
electron transfer to carotenoid radical cations from puerarin or daidzein
Astaxanthin is found to be an efficient antioxidant despite
its low rank in the regeneration hierarchy among the carotenoids
as shown both experimentally and based on theoretical
#
dianion. From linear regression, the relation ∆G ) 0.55 · ∆∆G° +
1
7.0 is obtained.
calculations.2
1,37,38
On the basis of the high electron accepting
index, astaxanthin was recently classified as the best antiradical
substance among the carotenoids, while canthaxanthin, zeax-
anthin and especially ꢀ-carotene were concluded to be less
effective. This classification relates, however, to carotenoids
as “antireductants” rather than as an antioxidants, and under
conditions of oxidative stress, the reverse ordering applies with
Astaxanthin contains two hydroxyl and two keto groups which
make this carotenoid more hydrophilic than the parent ꢀ-car-
otene, a feature which has been associated with the antioxidative
properties of astaxanthin. In order to explore the roles of these
two functional groups, we have accordingly selected the four
carotenoids seen in Scheme 1, in which hydroxyl and keto
groups vary systematically facilitating structural comparison.
The reduction of the radical cations of these carotenoids by
the isoflavonoid dianions followed second-order kinetics and
the second-order rate constants are collected in Table 2. The
second-order rate constants are approaching the diffusion limit,
which may indicate that some preassociation is involved prior
to electron transfer as shown in eq 12. The reaction is an electron
transfer and the driving force for the reaction may be calculated
from the standard reduction potential for the carotenoid radical
cations and for the isoflavonoid radicals shown in Table 1. Since
the potentials for the carotenoid in dichloromethane need a
correction not available for specific solvation effects to be
comparable with the potential determined for puerarin and
daidzein dianions in aqueous solution, the ∆G° calculated for
the electron transfer reactions are in Table 1, expressed relative
38
ꢀ
-carotene as the best antiradical substance due to its highest
38
relative electron donating index among these carotenoids. Plant
phenols, like isoflavonoids, are good antioxidants and, as such,
also good antiradical substance. The major finding of the
39
present study is that when carotenoids under oxidative stress
act as electron donors and antioxidants rather than electron
acceptors and “antireductants”, the carotenoid radical cation
formed is efficiently reduced by the isoflavonoids, acting as good
electron donors. Moreover, the rate for this electron transfer
depends on the driving force as calculated from the standard
reduction potential, and the rate is approaching the diffusion
limit for the radical cation of astaxanthin, the carotenoid with
the highest electron accepting index.
Acknowledgment. This work has been supported by grants-
in-aid from Natural Science Foundation of China (Nos. 20673144
and 20803091) and from the Ministry of Science and Technol-
ogy of China (Nos. 2009CB220008 and 2006BAI08B04-06).
Continuing support from LMC (U-10 grant), Centre for
Advanced Food Studies to the Food Chemistry group at
University of Copenhagen, is gratefully acknowledged.
to the value for ꢀ-carotene as ∆∆G°. The activation free energy,
#
∆
G , is calculated using transition state theory from the second-
3
5
order rate constants and shown in Table 2. These thermody-
namic data together establish a free energy linear relationship
#
between ∆G and ∆∆G° with the slope 0.55 ( 0.04 V as seen
in Figure 6. The linearity indicates that a common mechanism
is operative for the different carotenoids, and the intermediate
value of the slope indicate that both the reactants and the
products have important contributions to the structure of the
transition state for electron transfer. A similar analysis for 13
vitamin E analogues and the stable 2,6-ditert-butyl-4-(4-meth-
References and Notes
(1) Key, T. J.; Thorogood, M.; Appleby, P. N.; Burr, M. L. Br. Med. J.
1996, 313, 775.
(
2) El-Agamey, A.; Cantrell, A.; Land, E. J.; McGarvey, D. J.; Truscott,
T. G. Photochem. Photobiol. Sci. 2004, 3, 802.
3) Gerster, H. J. Am. Coll. Nutr. 1997, 16, 109.
36
oxylhexyl)phenoxyl radical led to a slope of 0.32 ( 0.04.
(
From a comparison of the four carotenoids, it is seen that
the astaxanthin radical cation as the most oxidizing radical is
reacting with the highest rate and faster than the radicals of
both canthaxanthin and zeaxanthin. Furthermore, canthaxanthin
reacts faster than zeaxanthin, suggesting the keto-group to be
(4) McNulty, H. P.; Byun, J.; Lockwood, S. F.; Jacob, R. F.; Mason,
R. P. Biochim. Biophys. Acta 2007, 1768, 167.
(
5) Hosokawa, M.; Okada, T.; Mikami, N.; Konishi, I.; Miyashita, K.
Food Sci. Biotechnol. 2009, 18, 1.
6) Edge, R.; McGarvey, D. J.; Truscott, T. G. J. Photochem. Photobiol.
1997, 41, 189.
(

1
32 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Han et al.
(
7) Oteiza, P. I.; Erlejman, A. G.; Verstraeten, S. V.; Keen, C. L.; Fraga,
C. G. Clin. DeV. Immunol. 2005, 12, 19.
8) Arora, A.; Byrem, T. M.; Nair, M. G.; Strasburg, G. M. Arch.
Biochem. Biophys. 2000, 373, 102.
9) Uri, N. Mechanism of Antioxidation. In Autoxidation and Antioxi-
dants; Lundberg, W. O., Ed.; Wiley: New York, 1961; 133.
(25) Hendrickson, H. P.; Kaufman, A. D.; Lunte, C. E. J. Pharm.
Biomed. Anal. 1994, 12, 325.
(26) Konovalova, T. A.; Kispert, L. D.; Konovalov, V. V. J. Phys. Chem.
(
B 1997, 101, 7858.
(
(
27) Hill, T. J.; Land, E. J.; McGarvey, D. J.; Schalch, W.; Tinkler,
J. H.; Truscott, T. G. J. Am. Chem. Soc. 1995, 117, 8322.
28) Tian, Y.-X.; Han, R.-M.; Zhang, J.-P.; Skibsted, L.-H. Free Rad.
Res. 2008, 42, 281.
29) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms;
McGraw-Hill, Inc.: New York, 1981.
30) Edge, R.; Land, E. J.; McGarvey, D.; Mulroy, L.; Truscott, T. G.
J. Am. Chem. Soc. 1998, 120, 4087.
31) Jensen, C.; Birk, E.; Jokumsen, A.; Skibsted, L. H.; Bertelgen, G.
Z. Lebensm.-Unters. Forsch. 1998, 207, 189.
32) Prabhu, P. N. N.; Ashokkumar, P.; Sudhandiran, G. Fund. Clin.
(
(
10) Palozza, P.; Krinsky, N. I. Acta Med. Rom. 1993, 31, 131.
11) Han, R.-M.; Tian, Y.-X.; Becher, E. M.; Andersen, M. L.; Zhang,
J.-P.; Skibsted, L. H. J. Argric. Food Chem. 2007, 55, 2384.
12) Liang, R.; Han, R.-M.; Fu, L.-M.; Ai, X.-C.; Zhang, J.-P.; Skibsted,
(
(
(
L. H. J. Agric. Food Chem. 2009, 57, 7118–7124.
(
(
(
13) Galano, A. J. Phys. Chem. B 2007, 111, 12898.
14) Tian, Y.-X.; Han, R.-M.; Fu, L.-M.; Zhang, J.-P.; Skibsted, L. H.
(
J. Phys. Chem. B 2008, 112, 2273.
15) Han, R.-M.; Tian, Y.-X.; Wu, Y.-S.; Wang, P.; Ai, X.-C.; Zhang,
J.-P.; Skibsted, L.-H. Photochem. Photobiol. 2006, 82, 538.
(
(
Pharmacol. 2009, 23, 225.
(33) Matsushita, Y.; Suzuki, R.; Nara, E.; Yokoyama, A.; Miyashita,
K. Fish. Sci. 2000, 66, 980.
(34) Burke, M.; Edge, R.; Land, E. J.; McGarvey, D. J.; Truscott, T. G.
FEBS Lett. 2001, 500, 132.
(
(
16) Mortensen, A.; Skibsted, L. H. FEBS Lett. 1997, 417, 261.
17) Luo, H. X.; Shi, Z. J.; Li, N. Q.; Gu, Z. N.; Zhuang, Q. K. Anal.
Chem. 2001, 73, 915.
(
18) Liu, D.; Gao, Y.; Kispert, L. D. J. Electroanal. Chem. 2000, 488,
1
40.
(
19) Gao, G.; Wei, C. C.; Jeevarajan, A. S.; Kispert, L. D. J. Phys. Chem.
(35) Santosh, K. U. Chemical Kinetics and Reaction Dynamics; jointly
published with, Anamaya Publishers: New Delhi, India, 2006; p 199.
(36) van Acker, S. A. B. E.; Roymani, L. M. H.; Bast, A. Free Radical
Biol. Med. 1993, 15, 311.
1
996, 100, 5362.
(20) Liang, J.; Tian, Y.-X.; Yang, F.; Zhang, J.-P.; Skibsted, L. H. Food
Chem. 2009, 115, 1437.
21) Mortensen, A.; Skibsted, L. H. J. Agric. Food Chem. 1997, 45,
970.
22) Han, R.-M.; Tian, Y.-X.; Liu, Y.; Chen, C.-H.; Ai, X.-C.; Zhang,
J.-P.; Skibsted, L. H. J. Agric. Food Chem. 2009, 57, 3780.
(
(37) Naguib, Y. M. A. J. Agric. Food Chem. 2000, 48, 1150.
2
(
38) Mart ´ı nez, A.; Rodr ´ı guez-Giron e´ s, M. A.; Barbosa, A.; Costas, M.
J. Phys. Chem. A 2008, 112, 9037.
39) Mart ´ı nez, A. J. Phys. Chem. B 2009, 113, 4915.
(
(
(
(
23) Li, C.; Hoffman, M. Z. J. Phys. Chem. B 1999, 103, 6653.
24) Janeiro, P.; Oliveira Brett, A. M. Anal. Chim. Acta 2004, 518, 109.
JP907349X