Biosci. Biotechnol. Biochem.

, 74 (2), 397–401, 2010

Note
Protective Effects of Kaempferol (3,40 ,5,7-tetrahydroxyflavone)
against Amyloid Beta Peptide (A)-Induced Neurotoxicity in ICR Mice
Jae Kyeom K IM,1;5 Soo Jung C HOI,2 Hong Yon C HO,1 Han-Joon HWANG,1 Young Jun K IM,1
Seung Taik L IM,2 Chang-Ju K IM,3 Hye Kyung K IM,4 Sabrina P ETERSON,5 and Dong-Hoon S HIN1; y
1
Department of Food and Biotechnology, Korea University, Seoul 136-701, Republic of Korea
2
Graduate School of Life Sciences and Biotechnology, Korea University, Seoul 136-701, Republic of Korea
3
Department of Physiology, Kyung Hee University, Seoul 130-701, Republic of Korea
4
Department of Food and Biotechnology, Hanseo University, Seosan 356-706, Republic of Korea
5
Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN, 55108, USA

Received August 6, 2009; Accepted November 20, 2009; Online Publication, February 7, 2010
[doi:10.1271/bbb.90585]

To determine the effects of kaempferol, rat pheo- prominent feature observed in AD patients. Although
chromocytoma cells (PC12) and Institute of Cancer the cause of the vulnerability of basal forebrain
Research (ICR) mice were utilized as neuronal models. cholinergic neurons and its relationship to A
remain
Using in vitro assays, kaempferol was shown to have unclear, several studies have reported that acetylcholi-
protective effects against oxidative stress-induced cyto- nesterase promotes the assembly of A
into fibrils.7,8)
toxicity in PC12 cells. Administration of kaempferol Flavonoids are polyphenolic compounds, ubiquitous
also significantly reversed amyloid beta peptide (A)- in fruits, that can be classified into six categories:
induced impaired performance in a Y-maze test. Taken flavones, flavonols, flavanones, isoflavones, flavanols,
altogether, the results reported here suggest that further and anthocyanins.9) Epidemiological and pharmaco-
investigation is warranted of the influence of kaempferol logical studies have indicated that consumption of
on pathways related to Alzheimer’s disease. flavonoids is associated with many beneficial effects,
including antioxidative, antiviral, anticancer, and anti-
Key words: kaempferol (3,40 ,5,7-tetrahydroxyflavone); inflammatory effects, as well as the prevention of
Alzheimer’s disease; amyloid beta peptide; cardiovascular diseases.10–14) Because some of the
oxidative stress beneficial effects of flavonoids are believed to result
mainly from their antioxidative properties, these anti-
Alzheimer’s disease (AD) is a neurodegenerative oxidative activities have gained increasing interest.
disorder of the central nervous system. It is correlated The objective of this study was to examine the
with cognitive malfunction, loss of memory, unusual protective activities of kaempferol against A
, a major
behavior, and declining perception ability. AD is representative flavonol that has been isolated from tea,
accompanied by the following pathological changes in broccoli, grapefruit, and other plant sources, including
the brain: diffuse loss of neurons, abnormal deposits of medicinal herbs.15) To investigate the protective effects
intracellular proteins, and extracellular proteins such as of kaempferol against AD, rat pheochromocytoma
senile plaques. The most abundant constituent of senile (PC12) cells and A
-injected ICR (Institute of Cancer
plaques in AD is a 40–42 amino acid peptide named Research) mice were utilized as neuronal models.
amyloid-beta peptide (A
), which is formed by cleavage PC12 cells were cultured and maintained as previ-
of the amyloid precursor protein by
- and -secre- ously described.16) Although PC12 cell line was origi-
tase.1,2) The mechanism and cellular pathway of neuro- nally isolated from adrenal medulla tumors, the adrenal
nal degeneration induced by this peptide are not yet fully medulla originates in the neural crest, and it displays
understood, but A
-induced cytotoxicity has been found properties associated with sympathicoblasts as well.
to be caused by the intracellular accumulation of Hence, it is widely used in research related to AD.1,2,17)
reactive oxygen species (ROS), eventually resulting in Briefly, RPMI 1640 supplemented with 10% heat
peroxidation of cell membranes, modification of pro- inactivated horse serum, 5% fetal bovine serum, and
teins, damage to DNA/RNA, and cell death.3,4) The fact 1% antibiotic-antimycotic was provided as the growth
that antioxidants, vitamin E and C, protect neuronal cells medium in a humidified incubator at 37
C under 5%
in vitro against A
-induced cytotoxicity confirms CO2 . Cultured cells were dislodged from culture dishes,
this.5,6) Besides the progressive deposition of A
, a and subcultured when each dish was 80–90% confluent.
deficiency of acetylcholine (Ach), a neurotransmitter The split ratio for subculture was 1:4. The medium was
found in the synapses of the cerebral cortex, is another changed at least 3 times a week.

y
To whom correspondence should be addressed. Fax: +82-2-3290-3429; E-mail: dhshin@korea.ac.kr
Abbreviations: AAPH, 2,20 -azobis(2-amidinopropane)dihydrochloride; ABTS, 2,20 -azinobis(3-ethylbenzothiazoline-6-sulfonic acid); A
, amyloid
beta peptide; Ach, acetylcholine; AChE, acetylcholinesterase; AD, Alzheimer’s disease; APP, amyloid precursor protein; DCF-DA, 20 ,70 -
dichlorofluorescin diacetate; ICR, Institute of Cancer Research; ICV, intracerebroventricular; MDA, malondialdehyde; MTT, 3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide; PBS, phosphate buffered saline; PC12, rat pheochromocytoma cells; PUFA, polyunsaturated fatty acid; ROS,
reactive oxygen species; TBARS, thiobarbituric acid reactive substances
398 J. K. KIM et al.

Intracellular oxidative stress levels were evaluated by memory impairment. The control group was injected
the 20 ,70 -dichlorofluorescin diacetate (DCF-DA) assay to with non-toxic reverse A
42-1 . A
was dissolved in
detect ROS by a method previously described.18) Briefly, 0.85% (v/v) sodium chloride solution and 5 ml (410
seeded PC12 cells in 96-well plates were pretreated for pmol/mouse) was injected into each mouse using a
48 h at 25 mM, 50 mM, 75 mM, and 100 mM (final concen- Hamilton micro-syringe inserted to a depth of 2.5 mm.
tration) of kaempferol. The cells were treated with and All experimental procedures were approved by the
without hydrogen peroxide for 2 h, and then 50 mM of guidelines established by the Animal Care and Use
DCF-DA was added. After a 50 min incubation period, Committee of Korea University. Spontaneous alterna-
DCF was quantified with a fluorometer using a 485 nm tion behavior in a Y-maze test was recorded to assess
excitation filter, and a 535 nm emission filter (GENois immediate working memory performance. The Y-maze
TECAN, Mannedorf, Switzerland). test was performed as described previously.16) Sponta-
Blue-green 2,20 -azinobis(3-ethylbenzothiazoline-6- neous alternation of mice was calculated as the ratio of
sulfonic acid) (ABTS) radicals were used to measure actual alternations to possible alternations (defined as
the radical scavenging activity of kaempferol. The ABTS the total number of arm entries minus two), multiplied
radical is of blue-green color when it is in its odd electron by 100.
state but it loses its color when the electron from Each result was expressed as the mean  SD. The
kaempferol or vitamin C pairs with its unpaired elec- statistical significance of differences among groups was
tron.19) One mM 2,20 -azobis(2-amidinopropane) dihydro- calculated by one-way analysis of variance (ANOVA).
chloride, a radical initiator, was added to 2.5 mM ABTS Data were analyzed by Duncan’s multiple-range test
in phosphate buffered saline (PBS; pH 7.4). The mixed using the Statistical Analysis System (SAS) software
solution was heated in a water bath at 68
C. The ABTS package (SAS Institute Inc., Cary, NC). p < 0:05 was
radical solution was adjusted to an absorbance of considered statistically significant.
0:650  0:020 at 734 nm with additional PBS. Twenty Since ROS (e.g., hydrogen peroxide and hydroxyl
ml of the sample was added to 980 ml of the ABTS radical) have been found to mediate cell injury, a DCF-
solution. The mixture was incubated in a 37
C water DA assay was performed to evaluate the inhibitory
bath without any light. The decrease in absorbance at effects of kaempferol against intracellular ROS forma-
734 nm was measured after 10 min of incubation. tion.22) In this assay, accumulation of ROS in PC12 cells
Cell viability was evaluated by 3-(4,5-dimethylthia- was found to increase by 153:2  2:0% after treatment
zol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) re- with 100 mM of hydrogen peroxide over 2 h as compared
duction assay, as described previously.2) PC12 cells with the control group (Fig. 1A). Forty-two h of
were plated at a density of 104 -105 cells/well on 96-well pretreatment of PC12 cells with vitamin C and kaemp-
plate and incubated with the concentrations of kaemp- ferol resulted in up to an 83:7  2:8% and a 120:3 
ferol described above for 48 h. After incubation with 1:5% decrease in ROS levels respectively. In addition,
hydrogen peroxide or A
25-35 , cell viability was deter- the antioxidant activity of kaempferol was examined by
mined by conventional MTT reduction.20) The amount measuring ABTS radical scavenging activity; kaemp-
of MTT formazan crystal that formed in the sample was ferol quenched radicals in a dose-dependent manner
quantified by measuring the absorbance using a micro- (Fig. 1B; r 2 ¼ 0:993), and was more effective than
plate reader (GENois TECAN, Mannedorf, Switzerland) vitamin C. ROS readily damage biological molecules,
at a test wavelength of 570 nm and a reference wave- including lipids, carbohydrates, proteins, and DNA,
length of 630 nm. and this can ultimately lead to apoptotic or necrotic
The level of lipid peroxidation was determined by cell death. Increased oxidation of protein and DNA,
measuring the amount of thiobarbituric acid reactive increased lipid peroxidation, and lowered levels of
substances (TBARS) by a modification of the method polyunsaturated fatty acids (PUFAs) have been reported
of Mihara and Uchiyama.21) Briefly, PC12 cells were in the AD brain.23) A
has been reported to be a possible
pretreated with various concentrations of kaempferol source of oxidative stress in the brain of AD patients
and vitamin C for 48 h. The cells were homogenized because it can acquire a free radical state and con-
with PBS (pH 7.4) after removal of the growth medium, sequently induce neurotoxicity.6) Conversely, it was also
and then mixed with H3 PO4 solution (1% v/v) and suggested that oxidative stress activates
-secretase and
160 ml of thiobarbituric acid solution (0.67% w/v). The increases A
secretion in late-onset AD. These cleaved
mixture was heated at 95
C for 45 min. After cooling, A
s can enter the mitochondria and induce free radicals,
n-butanol was added to the supernatant. The absorbance which eventually lead to neuronal degeneration in
of the extracted pigment was measured at 532 nm (UV- AD.24) Thus removal of excess ROS or inhibition of
1601 Shimadzu, Kyoto, Japan). The protein content was their generation by antioxidants might be an effective
determined by the Bradford method using a protein therapeutic strategy in preventing neurodegenerative
assay kit (Bio-Rad Laboratories, Hercules, CA). disorders. Recently, work has been done to determine
ICR mice (male, 4 weeks old; Daehan Biolink, the antioxidative effects of kaempferol.25) In agreement
Chungnam, Korea) were housed nine per cage in a with these studies, hydrogen peroxide-induced deposits
room maintained under a 12 h light-dark cycle, 55% of ROS were found to be significantly reduced by the
humidity at 23–25
C. Kaempferol was mixed with a presence of kaempferol in PC12 cells; in contrast, the
commercial diet at concentrations of 5, 10, 20, and same concentration of vitamin C was less effective than
40 mg/kg of body weight (0.0003, 0.0006, 0.0012, and kaempferol (p < 0:01). In accordance with the results of
0.0024% respectively). The mice had free access to feed the DCF-DA assays, kaempferol demonstrated the same
and water for 4 weeks, and then A
1-42 was administered pattern of radical quenching activity, which was again
by intracerebroventricular (ICV) injection to induce higher than that of vitamin C. These results indicate that
 Protective Effects of Kaempferol against A
399

A A

B
B

C
Fig. 1. Effects of Kaempferol on Oxidative Stress in PC12 Cells and
Scavenging of Radicals.
A, The control group was not treated. The hydrogen peroxide
group was treated only with 100 mM hydrogen peroxide for 2 h.
Sample groups were preincubated with 25 mM, 50 mM, 75 mM, and
100 mM of vitamin C or kaempferol for 48 h before treatment with
100 mM hydrogen peroxide. Levels of oxidative stress in PC12 cells
were measured using DCF (20 ,70 -dichlorofluorescein), as described
in ‘‘Materials and Methods.’’ Data represent the mean value
(n ¼ 4) SD. Duncan’s multiple-range test of SAS indicated a
significant difference ( p < 0:01 vs. the control group, # p, ## p,
###
p < 0:01 vs. the hydrogen peroxide group). B, The relationships
between vitamin C and kaempferol and absorbance reduction of the Fig. 2. Protective Effects of Kaempferol on Cell Viability and
free blue-green ABTS radical at 734 nm. Data represent mean Oxidative Stress on PC 12 Cells.
absorbance (n ¼ 4Þ  SD. A, The control group was not treated. The hydrogen peroxide
group was treated with 100 mM hydrogen peroxide for 2 h. Sample
groups were preincubated with 25 mM, 50 mM, 75 mM, and 100 mM of
kaempferol can exhibit a neuroprotective effect by kaempferol before treatment with 100 mM hydrogen peroxide. The
scavenging radicals directly in PC12 cells, which can be same concentrations of vitamin C were also evaluated as a positive
control. Data represent the mean (n ¼ 3)  SD. Duncan’s multiple-
produced from A
in AD. range test of SAS indicated a significant difference ( p < 0:01 vs.
Since yellow tetrazolium salt is metabolically reduced the control group, # p, ## p, ### p < 0:01 vs. the hydrogen peroxide
in live cells to form insoluble formazan crystals, the group). B, The control group was not treated. The amyloid beta
amount of these crystals is directly proportional to the group was treated with 100 mM A
25-35 for 24 h. Sample groups were
preincubated with the concentrations described above. The same
number of viable cells. The cellular reactions in this
concentrations of vitamin C were also tested as a positive control.
reduction are not fully understood, but several studies Data represent the mean (n ¼ 4)  SD. Duncan’s multiple-range
have suggested that the mitochondrial succinate dehy- test of SAS indicated a significant difference ( p < 0:01 vs. the
drogenase system of active mitochondria is involved.2,19) control group, # p, ## p < 0:01 vs. the amyloid beta group). C, The
The protective effects of kaempferol and vitamin C on level of lipid peroxidation was measured after treatment with 100 mM
hydrogen peroxide for 2 h. The sample groups and positive control
the loss of PC12 cell viability induced by hydrogen were treated under the conditions described above. Data represent
peroxide or A
25-35 were assessed with this assay the mean (n ¼ 4)  SD. Duncan’s multiple-range test of SAS
(Fig. 2A and B). Treatment with 100 mM of hydrogen indicated a significant difference ( p < 0:01 vs. the control group,
#
peroxide and with A
25-35 decreased the viability of p, ## p < 0:05 vs. the hydrogen peroxide group).
neuronal cells, by 43:3  2:0% and 40:3  10:2%
respectively, as compared with the control group Malondialdehyde (MDA), a major product of lipid
(p < 0:01). For the groups pretreated with kaempferol peroxidation, spectrophotometrically measured by de-
and vitamin C, cell viability increased in a dose- tecting TBARS.20) The level of MDA was higher in cells
dependent manner. However, cell viability for the cells treated with hydrogen peroxide than with the control
pretreated with vitamin C was less than when the same (126:8  3:91% and 100:0  0:47% respectively, p <
concentrations of kaempferol were used (p < 0:01 for 0:01). Pretreatment of cells with all but the lowest dose
all concentrations treated). These results suggest that of vitamin C and kaempferol resulted in statistically
kaempferol protects PC12 cells has been possibly by significant reductions in MDA as compared with cells
protecting the mitochondria. treated only with hydrogen peroxide (Fig. 2C). PUFAs
400 J. K. KIM et al.
Table 1. Protective Effects of Kaempferol on Y-Maze Alternation in blueberry, and spinach) improved learning and memory
the A
-Treated Mice retention in aged rats.29) In addition, it has also been
Alternation Numbers of reported that antioxidant-enriched food reversed age-
Experimental related changes in antioxidant defenses in neuronal
behavior arm entries
group
(%) (%) tissue.30) However, it is still unclear whether the
Controla 66:14  5:93 100:00  20:07 protective effects of dietary flavonoids are associated
A
b 51:41  2:69 103:97  25:88 with their own reducing properties or with other
mechanisms, because they are substrates for several
A
+ Kaempferol
enzymes that are located in the small intestine, colon,
þ5 mg/kg body weight 58:18  3:59 117:01  16:58 and liver.31) Possible modifications of those enzymes
þ10 mg/kg body weight 66:43  4:03# 115:53  14:33
include methylation, glucuronidation, and sulfation, and
þ20 mg/kg body weight 66:32  6:19# 116:92  13:32
þ40 mg/kg body weight 77:26  6:82# 103:65  11:31
these modifications can alter the antioxidative properties
or the lipophilicity of flavonoids, which is positively
Data represent the mean (n ¼ 9)  SD. Controla , non-toxic reverse A
42-1 related to the permeability of flavonoids through the
injected (410 pmol/mouse). A
b , A
1-42 injected (410 pmol/mouse). The
mice were fed different concentrations of kaempferol (5, 10, 20, and
blood-brain barrier.29) Therefore, it is not clear whether
40 mg/kg, body weight). Duncan’s multiple-range test of SAS indicated a kaempferol diminished the neurotoxicity of A
directly
significant difference ( p < 0:01 vs. control group, # p < 0:01 vs. A
group). or indirectly in the brains of the mice. However, the
There were no significant differences in the number of arm entries for any
present data suggest that the administration of kaemp-
group.
ferol effectively reversed A
-induced damage in the
behavior test; thus, it might have contributed to
are abundantly present in brain phospholipids, which are ameliorating or improving the spatial working memory
the major constituents of the cellular membrane. They of the mice. To our knowledge, this is the first report that
are not only one of the most vulnerable targets for free- kaempferol has protective activity against A
-induced
radical damage but also the substrate of lipid perox- memory impairment in vivo.
idation. Lipid peroxidation produces secondary bioac- The present work clearly indicates that kaempferol
tive aldehydes, including 4-hydroxy-2-nonenal and exerted protective activity against oxidative stress,
acrolein, which can adduct to DNA and proteins. which plays a crucial role in the onset of AD. Taken
Increased levels of both aldehydes have been observed together, these results indicate that kaempferol, which
in brain regions of AD patients.26,27) Moreover, perox- has an antioxidative capacity, might be protective
idation of membrane lipids and proteins (carbonyl group against oxidative stress-induced neuronal damage, pos-
formation) can disrupt membrane fluidity, reduce mem- sibly by ameliorating disruption of neuronal cell
brane potential, elevate membrane permeability to ions membranes and mitochondria. Moreover, kaempferol
by inactivating membrane-bound enzymes, and even- also improved A
-induced memory impairment in
tually lead to cell death.28) Therefore, the results of this mice. These actions of kaempferol can account for
study imply that kaempferol suppresses lipid peroxida- its therapeutic profile. Based on the results of these
tion in PC12 cells and thus might be beneficial in in vitro and in vivo studies, further investigation is
protecting cells against oxidative damage. warranted to determine the effects of kaempferol at the
To confirm the anti-neurotoxic activity of kaempferol, molecular level and to examine its potential for
an animal model of A
-induced memory deficit, which therapeutic use.
involves administering an ICV injection of A
to ICR
mice, was used. Since A
1-42 is reported to cause Acknowledgments
memory impairment in mice, this method of A
1-42
exposure is widely used as an in vivo model of A
- This work was supported by a Korea University Grant
induced memory deficit.2) In the mouse model, all the (2009).
mice treated with the kaempferol mixed diet had normal
increases in body weight, and no acute toxicity or liver References
toxicity was observed (data not shown). Spontaneous
1) Heo HJ, Cho HY, Hong BS, Kim HK, Kim EK, Kim BG, and
alternation behavior, which is used as a measure of Shin DH, Amyloid, 8, 194–201 (2001).
spatial working memory, was recorded using the Y- 2) Choi SJ, Kim MJ, Heo HJ, Kim HK, Hong BS, Kim CJ, Kim
maze test. Administration of A
1-42 effectively impaired BG, and Shin DH, Amyloid, 13, 6–12 (2006).
spatial working memory as compared with the control 3) Markesbery WR and Carney JM, Brain Pathol., 9, 133–146
group (p < 0:01, Table 1). The groups pretreated with (1999).
kaempferol had an ameliorative A
1-42 -induced decrease 4) Behl C, Davis J, Cole GM, and Schubert D, Biochem. Biophys.
Res. Commun., 186, 944–950 (1992).
in alternation behavior (50 mM: 66:43  4:03%; 75 mM:
5) Shin EJ, Suh SK, Lim YK, Jhoo WK, Hjelle OP, Ottersen OP,
66:32  6:19%; 100 mM: 77:26  6:82%, p < 0:01 for Shin CY, Ko KH, Kim WK, Kim DS, Chun W, Ali S, and Kim
all concentrations). There was no significant difference HC, Neurosci., 133, 715–727 (2005).
in the numbers of arm entries for any group. These 6) Behl C, Int. J. Vitam. Nutr. Res., 69, 213–219 (1999).
results indicate that the administration of kaempferol 7) Ulrich J, Meier-Ruge W, Probst A, Meier E, and Ipsen S, Acta
attenuated impaired spatial working memory in the Neuropathol., 80, 624–628 (1990).
Y-maze test but did not affect the general locomotive 8) Morán MA, Mufson EJ, and Gómez-Ramos P, Acta Neuro-
pathol., 85, 362–369 (1993).
activity of the mice. A number of studies have shown 9) Hollman PC and Katan MB, Food Chem. Toxicol., 37, 937–942
that dietary intervention with antioxidants has beneficial (1999).
effects on cognitive functions. Supplementation with 10) Burda S and Oleszek W, J. Agric. Food Chem., 49, 2774–2779
vitamins C and E and plant extracts (strawberry, (2001).
 Protective Effects of Kaempferol against A
401
11) Guo Q, Zhao L, You Q, Yang Y, Gu H, Song G, Lu N, and Xin 20) Berridge MV, Herst PM, and Tan AS, Biotechnol. Annu. Rev.,
J, Antiviral Res., 74, 16–24 (2007). 11, 127–152 (2005).
12) Cárdenas M, Marder M, Blank VC, and Roguin LP, Bioorg. 21) Mihara M and Uchiyama M, Anal. Biochem., 86, 271–278 (1978).
Med. Chem., 14, 2966–2971 (2006). 22) Buttke TM and Sandstrom PA, Immunol. Today, 15, 7–10
13) Li N, Liu JH, Zhang J, and Yu BY, J. Agric. Food Chem., 56, (1994).
3876–3883 (2008). 23) Heo HJ, Choi SJ, Choi SG, Shin DH, Lee JM, and Lee CY,
14) Kris-Etherton PM, Hecker KD, Bonanome A, Coval SM, J. Food Sci., 73, 28–32 (2008).
Binkoski AE, Hilpert KF, Griel AE, and Etherton TD, Am. J. 24) Reddy PH, J. Neurochem., 96, 1–13 (2006).
Med., 113, 71S–88S (2002). 25) Dai F, Miao Q, Zhou B, Yang L, and Liu ZL, Life Sci., 78,
15) Park JS, Rho HS, Kim DH, and Chang IS, J. Agric. Food Chem., 2488–2493 (2006).
54, 2951–2956 (2006). 26) Williams TI, Lynn BC, Markesbery WR, and Lovell MA,
16) Kim JK, Bae H, Kim MJ, Choi SJ, Cho HY, Hwang HJ, Kim Neurobiol. Aging, 27, 1094–1099 (2006).
YJ, Lim ST, Kim EK, Kim HK, Kim BY, and Shin DH, Biosci. 27) Lovell MA, Xie C, and Markesbery WR, Neurobiol. Aging, 22,
Biotechnol. Biochem., 73, 1105–1112 (2009). 187–194 (2001).
17) Fujita K, Lazarovici P, and Guroff G, Environ. Health 28) Yu BP and Yang R, Ann. NY Acad. Sci., 768, 1–11 (1996).
Prospect., 80, 127–142 (1989). 29) Youdim KA, Shukitt-Hale B, and Joseph JA, Free Radic. Biol.
18) Wolfe KL and Liu RH, J. Agric. Food Chem., 55, 8896–8907 Med., 11, 1683–1693 (2004).
(2007). 30) Lynch MA, Nutr. Neurosci., 4, 419–438 (2001).
19) Kim DO, Lee KW, Lee HJ, and Lee CY, J. Agric. Food Chem., 31) Rice-Evans C, Free Radic. Biol. Med., 7, 827–828 (2004).
50, 3713–3717 (2002).