J Nutr Sci Vitaminol, 52, 437–444, 2006

Protective Effects of Broccoli (Brassica oleracea) against Oxidative

Damage in Vitro and in Vivo

Eun Ju CHO1, Young A LEE2, Hye Hyun YOO2 and Takako YOKOZAWA2,*
1
Department of Food Science and Nutrition, and Research Institute of Ecology for the Elderly, Pusan National
University, 30 Jangjeon-dong, Geumjeong-gu, Busan 609–735, Korea
2
Institute of Natural Medicine, University of Toyama, 2630 Sugitani, Toyama 930–0194, Japan
(Received May 15, 2006)

Summary The antioxidative effect and protective potential against diabetes of the broc-
coli flower were investigated both in vitro and in a diabetic rat model. Among fractions of
MeOH, CH2Cl2, BuOH, and H2O, the BuOH fraction exerted the strongest inhibitory activi-
ties on 1,1-diphenyl-2-picrylhydrazyl radical, radical-induced protein oxidation, and nitric
oxide generation by sodium nitroprusside. The in vitro results suggest that the BuOH frac-
tion from the broccoli flower has a protective potential against oxidative stress. The rat
model with diabetes induced by streptozotocin was employed to evaluate the protective
effect of the BuOH fraction in vivo. Diabetic rats showed reduced body weight gain and
heavier kidney and liver weights than normal rats, while oral administration of the BuOH
fraction at an oral dose of 100 or 200 mg/kg body weight/d for 20 d attenuated the physio-
logical changes induced by diabetes. In addition, oral administration of the BuOH fraction to
diabetic rats led to significant decreases in serum glucose and glycosylated protein, while it
resulted in the increase of serum albumin, implying that the BuOH fraction improves the
abnormal metabolism of glucose and protein that leads to oxidative stress. Moreover, it sig-
nificantly reduced thiobarbituric acid-reactive substance levels in serum, hepatic and renal
mitochondria. This suggests that the BuOH fraction would alleviate the oxidative stress
associated with diabetes through the inhibition of lipid peroxidation. The present study dem-
onstrates that the BuOH fraction has an antioxidative effect in vitro and it protects against
oxidative stress induced by diabetes in an in vivo model.
Key Words broccoli, antioxidative activity, oxidative stress, diabetes, streptozotocin

Oxidative stress that results from excessive free radi-
cal formation and limited antioxidant defense leads to
changes in proteins, lipids, polysaccharides, and DNA in
biological systems (1). Eventually, it results in patho-
physiological conditions such as aging, cardiovascular
disease, Alzheimer’s disease, and diabetes (2–5). In bio-
logical systems, inactivation and removal of free radi-
cals depend on reactions involving the antioxidant
defense system. Therefore, much attention has been
focused on the search for antioxidants with potentials
to scavenge free radicals and enhance the antioxidative
defense system. In particular, antioxidants from natural
food sources that could have a synergistic effect with
other components have attracted much attention.
Diabetes mellitus is characterized by abnormal glu-
cose homeostasis. Under diabetes, the failure of insulin-
stimulated glucose uptake by fat and muscle tissues
causes glucose concentrations in blood to remain high.
This increased glucose both enhances oxidant produc-
tion and impairs antioxidant defenses (6–8). Thus, pos-
sible causes of oxidative stress in diabetes include an
excessive production of radical oxygen species, espe-
* To whom correspondence should be addressed.
E-mail: yokozawa@inm.u-toyama.ac.jp
cially from glycation or lipoxidation processes, and a
decrease in antioxidant defense systems. Therefore,
scavenging free radicals and/or elevating defenses via
the regulation of blood glucose are therapeutic ap-
proaches for diabetes.
Vegetables that contain high amounts of polyphe-
nols, vitamin C, vitamin E,
-carotene, and lycopene
have protective effects against the risk of chronic degen-
erative disease (9–12). In particular, cruciferous vegeta-
bles, including broccoli, cabbage, kale, brussels sprouts,
and cauliflower, were found to have glucosinolate at a
high concentration, and it has been well established
that their breakdown products induce endogenous anti-
oxidant defences such as quinine reductase and glu-
tathione S-transferase in vitro and in vivo (13). Among
the cruciferous vegetables, broccoli was established to
have an anticancer effect mainly attributable to the
active component of sulforaphane. However, other ben-
eficial effects on health and disease prevention of broc-
coli have rarely been investigated. Therefore, we investi-
gated the antioxidant activities of fractions (MeOH,
CH2Cl2, BuOH, and H2O) from the broccoli flower under
an in vitro system. Furthermore, the effects of the active
fractions against diabetes under an in vivo system using
rats with streptozotocin (STZ)-induced diabetes were

437
438
Fig. 1. Extraction and fractions from broccoli flower.
also investigated.
MATERIALS AND METHODS
Extract preparation and fractions from broccoli. Broc-
coli flowers (570 g) were freeze-dried and then refluxed
with MeOH for 3 h, and the MeOH extract (20 g) was
suspended in distilled H2O and partitioned with CH2Cl2,
n-BuOH, and H2O in sequence, thus yielding 4.5, 5.5,
and 8.4 g of the respective fractions (Fig. 1).
Bioactivity-guided fractionation and comparison of radi-
cal scavenging activity. From the BuOH fraction, activ-
ity-guided isolation by silica gel chromatography eluted
with chloromethane-MeOH (gradient elution, 10:1Æ
1:1) gave 7 fractions, and 2,2-diphenyl-1-picrylhydra-
zyl (DPPH) radical scavenging activity of the 7 fractions
was measured and compared.
Isolation and identification of active components. The
most active fraction among the 7, fraction 5, was fur-
ther purified by silica gel chromatography using an
EtOAc-MeOH (10:1) system. The two major compo-
nents were isolated and their chemical structures were
characterized by spectroscopic techniques of mass and
NMR. NMR spectra were measured on a Bruker Avance
500 spectrometer (Bruker, Ettlingen, Germany, 500
MHz for 1H). The Fab-MS was recorded on a Jeol JMS-
700 mass spectrometer (Jeol, Tokyo, Japan). Compound
I (5 mg) and II (6 mg) were identified as 1,2-disinapoyl-
gentiobiose and 1-sinapoyl-2-feruloylgentiobiose,
respectively, and their chemical structures are illus-
trated in Fig. 2.
(1) 1,2-Disinapoylgentiobiose: Yellow crystal (MeOH),
C34H42O19, FAB-MS m/z: 777 [MNa], 1H-NMR
(500 MHz, DMSO-d6): 7.63 (1H, d, J15.8 Hz, H-7¢),
7.62 (1H, d, J15.8, H-7), 6.88 (2H, s, H-2¢, 6¢), 6.85
(2H, s, H-2, 6), 6.40 (1H, d, J15.8 Hz, H-8¢), 6.32
(1H, d, J15.8 Hz, H-8), 3.85 (3H, s, OCH3¢), 3.83 (3H,
s, OCH3¢), 5.77 (1H, d, J8.83 Hz, glu-1), 5.08 (1H, dd,
J8.8, 8.8 Hz, glu-2), 4.35 (1H, d, J7.9 Hz, glu-1¢),
4.11 (1H, d, J10.2 Hz, glu-6b), 3.20 (1H, dd, J7.9,
CHO EJ et al.
Fig. 2. Chemical structures of active components iso-
lated from broccoli flower.
8.7 Hz, glc-3¢), 3.12 (1H, dd, J7.9, 8.7 Hz, glc-2¢), 13C-
NMR (125 MHz, DMSO-d6): 165.0 (C-9¢), 164.2 (C-9),
147.5 (C-3, 3¢, 5, 5¢), 146.8 (C-7¢), 145.5 (C-7), 138.3
(C-4¢), 137.9 (C-4), 123.6 (C-1), 123.5 (C-1¢), 114.1
(C-8¢), 113.1 (C-8), 106.1 (C-2¢, 6¢), 105.5 (C-2, 6),
102.6 (glc-1¢), 91.5 (glc-1), 76.5 (glc-5¢), 76.2 (glc-5),
75.9 (glc-3¢), 73.2 (glc-3), 72.9 (glc-2), 72.1 (glc-2¢),
69.4 (glc-4¢), 69.0 (glc-4), 67.2 (glc-6), 60.5 (glc-6¢),
55.6 (OCH3¢), 55.5 (OCH3).
(2) 1-Sinapoyl-2-feruloylgentiobiose: Yellow crystal
(MeOH), C33H40O18, FAB-MS m/z: 747 [MNa], 1H-
NMR (500 MHz, DMSO-d6): 7.62 (1H, d, J15.8 Hz, H-
7¢¢¢), 7.60 (1H, d, J15.8 Hz H-7), 7.30 (1H, d,
J2.0 Hz, H-2¢¢¢), 7.09 (1H, dd, J2.0, 7.9 Hz, H-6¢¢¢),
7.02 (2H, s, H-2, 6), 6.79 (1H, d, J7.9 Hz, H-5¢¢¢),
6.48 (1H, d, J15.8 Hz, H-8), 6.48 (1H, d, J15.8 Hz,
H-8¢¢¢), 3.84 (3H, s, OCH3¢), 3.83 (3H, s, OCH3), 5.76
(1H, d, J8.8 Hz, glu-1), 4.92 (1H, dd, J8.8, 8.8 Hz,
glu-2), 4.35 (1H, d, J7.9 Hz, glu-1¢), 4.07 (1H, d,
J11.2 Hz, glu-6b), 3.15 (1H, dd, J7.9, 8.7 Hz, glc-
3¢), 3.02 (1H, dd, J7.9, 8.7 Hz, glc-2¢), 13C-NMR
(125 MHz, DMSO-d6) d: 165.2 (C-9), 164.4 (C-9¢),
148.8 (C-4¢), 147.5 (C-3, 5), 147.4 (C-3¢), 146.8 (C-7),
145.0 (C-7¢), 138.3 (C-4), 124. 9 (C-1¢), 123.5 (C-1),
122.8 (C-6¢), 114.9 (C-5¢), 113.6 (C-8¢), 113.1 (C-8),
110.6 (C-2¢), 106.1 (C-6), 102.5 (glc-1¢), 91.5 (glc-1),
76.5 (glc-5¢), 76.2 (glc-5), 75.9 (glc-3¢), 73.2 (glc-3),
72.8 (glc-2), 72.1 (glc-2¢), 69.4 (glc-4¢), 69.0 (glc-4),
67.2 (glc-6), 60.5 (glc-6¢), 55.6 (OCH3), 55.0 (OCH3¢).
Reagents. DPPH and 2,2¢-azobis (2-amidinopro-
 Broccoli Ameliorates Oxidative Damage
pane) dihydrochloride (AAPH) were purchased from
Wako Pure Chemical Industries, Ltd. (Osaka, Japan)
and allophycocyanin was obtained from Funakoshi Co.,
Ltd. (Tokyo, Japan). STZ and sodium nitroprusside
(SNP) were purchased from Sigma Chemical Co. (St.
Louis, MO, USA).
In vitro experiments.
(1) DPPH radical scavenging activity: One hundred
microliters of 50% EtOH solution of the sample (control:
100 L of 50% EtOH) was added to microwells followed
by 100 L of 60 M DPPH in EtOH, according to the
method of Hatano et al. (14). After being mixed gently
and left to stand for 30 min at room temperature, the
DPPH radical level was measured with a microplate
reader, Model 3550-UV (Bio-Rad, Tokyo, Japan). The
antioxidant activity of each sample was expressed as
the IC50 value (concentration in g/mL required to
inhibit DPPH radical formation by 50%) calculated
from the log-dose inhibition curve. L-Ascorbic acid was
used as a positive control.
(2) AAPH-induced protein oxidation: According to
the methods of Courderot-Masuyer et al. (15), a reac-
tion mixture containing 37.5 nM allophycocyanin,
3 mM AAPH, and an aqueous solution of test sample in
75 mM phosphate buffer (pH 7.0) was incubated at
37˚C. The fluorescence obtained just before the addition
of the radical generator AAPH was used as the 100%
value for that sample. Loss of fluorescence was mea-
sured every 5, 10, 20, or 30 min at an emission wave-
length of 651 nm and an excitation wavelength of
598 nm using a fluorescence spectrophotometer (model
RF-5300 PC, Shimadzu, Kyoto, Japan).
(3) Nitric oxide (NO) scavenging effect: NO produc-
tion from SNP was measured according to the method
of Sreejayan (16). Briefly, SNP (5 mM) in phosphate
buffered saline (pH 7.4) was mixed with different con-
centrations of test samples and incubated at 25˚C for
150 min. The amount of NO produced by SNP was
assayed by measuring nitrite accumulation using a
microplate assay method based on the Griess reaction
(17).
Animal experiments.
(1) Animal preparation: The ‘Guidelines for Animal
Experimentation’ approved by the University of Toyama
were followed during these experiments. Male Wistar
rats (120–130 g) from Japan SLC Inc. (Hamamatsu,
Japan) were used. They were kept in a wire-bottomed
cage and exposed to a 12 h light/dark cycle. The room
temperature (about 25˚C) and humidity (about 60%)
were controlled automatically. Laboratory pellet chow
(CE-2, comprising 24.0% protein, 3.5% lipid, and
60.5% carbohydrate; CLEA Japan Inc., Tokyo, Japan)
and water were given ad libitum. After several days of
adaptation, STZ (50 mg/kg of body weight) dissolved in
citrate buffer (pH 4.5) was injected intraperitoneally fol-
lowing overnight fasting. One week after injection, the
glucose level of blood taken from the tail vein of each rat
was determined and then the animals were divided into
three groups, avoiding any inter group differences in
the blood glucose level. For the oral administration, the
439
BuOH fraction was dissolved with distilled water, and it
was administered at the doses of 100 or 200 mg/kg
body weight/d with total volume of 1 mL. On the other
hand, the control group was given the same amount of
plain drinking water orally. After administration for 20
consecutive days, the rats were sacrificed by decapita-
tion, a blood sample was collected, and serum was sep-
arated immediately by centrifugation. The livers and
kidneys were removed, rinsed with cold physiological
saline, and frozen at 80˚C until analysis.
(2) Assays of serum sample: The glucose levels were
measured using a commercial kit (Glucose CII-Test
Wako; Wako Pure Chemical Industries, Ltd.). Glycosy-
lated protein levels were determined by a modified thio-
barbituric acid (TBA) assay of Fluckiger and Winterhal-
ter (18), in which nonenzymatically bound glucose is
released as 5-hydroxymethylfurfural and is quantified
colorimetrically. Briefly, serum (100 L) was diluted to
1.0 mL, mixed with 0.5 mL oxalic acid (0.1 M), hydro-
lyzed for 4.5 h at 100˚C, and then the glycosylated pro-
tein level was quantified by measuring the absorbance
at 443 nm after reaction with TBA. The albumin con-
tents were quantified using a commercial protein assay
kit (A/G B-Test Wako; Wako Pure Chemical Industries,
Ltd.). TBA-reactive substance levels were measured
using the method of Naito and Yamanaka (19).
(3) Preparation of mitochondria and measurement
of TBA-reactive substance levels: According to the
methods of Johnson and Jung with a slight modification
(20, 21), the homogenates of liver and kidney were cen-
trifuged for 10 min at 800 g in a refrigerated centri-
fuge (4˚C), and the resulting supernatant was centri-
fuged for 15 min at 12,000 g. Each pellet was
resuspended in preparation medium, and the TBA-reac-
tive substance level was determined according to the
method of Mihara and Uchiyama (22). Protein levels
were determined by the method of Itzhaki and Gill (23)
with bovine serum albumin as the standard.
Statistical analysis. The in vitro results are expressed
as meanSD values. Data were analyzed by the one
way ANOVA among each groups by SAS (SAS Institute
Inc., Cary, NC, USA). Significant differences by Dun-
can’s multiple range test were determined among
groups at p0.05. The in vivo results are presented as
meanSE values. Data were analyzed by Dunnett’s test
and significant differences were determined among
groups at p0.05.
RESULTS
In vitro experiments
Table 1 shows the DPPH radical scavenging activities
of MeOH, CH2Cl2, BuOH, and H2O fractions from the
broccoli flowers. Among them, the BuOH fraction
exerted the strongest inhibition of DPPH formation by
50% at a concentration of 74.9 g/mL, whereas the
other fractions showed relatively weak DPPH radical
scavenging activities.
Figure 3 represents the protective effect of fractions
against protein oxidation induced by AAPH. The frac-
tions from the broccoli flowers inhibited the protein oxi-
440 CHO EJ et al.

dation by AAPH in a dose-dependent manner. In partic- after exposure to AAPH for 30 min was 28.8, 43.4, and
ular, compared with other fractions at the same 69.8% compared with the control value of 2.6%. On
concentration, the BuOH fraction exerted the strongest the other hand, MeOH, CH2Cl2, and H2O fractions
protective activity against protein damage. At 5, 10, showed relatively low inhibitory activities compared
and 50 g/mL of the BuOH fraction, the fluorescence with the BuOH fraction.
As shown in Table 2 concerning the NO scavenging
Table 1. DPPH radical scavenging activity of fractions activities of the fractions from the broccoli flowers, the
from broccoli flower. BuOH fraction exhibited the greatest inhibitory effect
on NO generation by SNP, which reacts with oxygen to
Fraction IC50 (g/mL) form nitrite. In particular, the BuOH fraction showed
high NO scavenging activities of 13.2, 31.9, and 47.3%
MeOH 199.16.9b at concentrations of 12.5, 25, and 50 g/mL, respec-
CH2Cl2 633.69.2d tively, being significantly higher than those of the other
BuOH 74.92.6a fractions.
H2O 466.018.8c
In vivo experiments
As shown in Table 3, the intakes of food and water in
Ascorbic acid 1.10.0
rats with diabetes increased significantly compared
Each value is expressed as the meanSD. Means with with normal rats. The administration of BuOH fraction
different letters are significantly different (p0.05) from the broccoli flower led to the significant decrease
according to Duncan’s multiple range test. in water intake. In addition, the oral dose of 100 mg
decreased the food intake, whereas that of 200 mg did
Table 2. NO scavenging effect of fractions from broccoli not change it significantly. Table 4 shows the effects on
flower. changes in body and tissue weights of rats with STZ-
induced diabetes after oral administration of the BuOH
Concentration NO scavenging effect fraction from the broccoli flowers. The body weight gain
Fraction
(g/mL) (%) of rats with STZ-induced diabetes was significantly
lower than that of normal rats. After oral administra-
12.5 MeOH 12.43.7a
CH2Cl2 —
BuOH 13.20.3a Table 3. The intakes of food and water in normal rats
H2O 7.41.1a and rats with diabetes.

25 MeOH 17.93.2b Dose Food intake Water intake

Group
CH2Cl2 — (mg/kg B.W./d) (mg/d) (mL/d)
BuOH 31.91.7a
H2O 11.40.7b Normal rats 18.70.5 41.42.8
Diabetes rats
50 MeOH 33.61.8a Control — 27.40.9* 126.66.6*
CH2Cl2 — BuOH fraction 100 25.12.0*,# 108.98.3*,##
BuOH 47.31.2b BuOH fraction 200 28.60.3* 110.55.6*,##
H2O 13.51.3c
Each value is expressed as the meanSE of 6 rats per
Each value is expressed as the meanSD. Means with group. Means with different symbols are significantly dif-
different letters are significantly different (p0.05) ferent according to Dunnett’s test. * p0.001 vs. normal
according to Duncan’s multiple range test. values; # p0.05, ## p0.01 vs. diabetic control values.

Fig. 3. Effects of MeOH (
), CH2Cl2 (), BuOH (), and H2O () fractions from broccoli flower on free radical-mediated pro-
tein damage. Each value is expressed as the meanSD. Means with different letters are significantly different (p0.05)
according to Duncan's multiple range test.
 Broccoli Ameliorates Oxidative Damage 441

Table 4. Effects of the BuOH fraction on the body and tissue weights of rats with STZ-induced diabetes.

Body weight
Dose Liver weight Kidney weight
Group
(mg/kg B.W./d) (g/100 g B.W.) (g/100 g B.W.)
Initial (g) Final (g) Gain (g/20 d)

Normal rats — 2236 2819 58.04.1 3.200.11 0.710.01

Diabetes rats
Control — 1955* 2019* 6.66.6* 4.190.10* 1.080.02*
BuOH fraction 100 1987* 21312* 14.15.0* 4.030.10* 0.930.02*,###
BuOH fraction 200 1996* 22012*,# 20.96.8*,## 3.960.11*,# 0.970.02*,###

Each value is expressed as the meanSE of 6 rats per group. Means with different symbols are significantly different accord-
ing to Dunnett’s test. * p0.001 vs. normal values; # p0.05, ## p0.01, ### p0.001 vs. diabetic control values.

Table 5. Effects of the BuOH fraction on glucose, glycosylated protein, and albumin levels in the serum of rats with STZ-
induced diabetes.

Dose Glucose Glycosylated protein Albumin

Group
(mg/kg B.W./d) (mg/dL) (nmol/mg protein) (g/dL)

Normal rats — 131.61.1 13.41.6 3.310.05

Diabetes rats
Control — 431.917.1** 21.91.4** 3.090.07*
BuOH fraction 100 405.515.1**,# 19.21.7**,# 3.090.10*
BuOH fraction 200 402.85.1**,## 17.91.3**,## 3.160.08

Each value is expressed as the meanSE of 6 rats per group. Means with different symbols are significantly different accord-
ing to Dunnett’s test. * p0.01, ** p0.001 vs. normal values; # p0.05, ## p0.01 vs. diabetic control values.

Fig. 4. Effects of the BuOH fraction on malondialdehyde levels of serum, hepatic mitochondria, and renal mitochondria of
rats with STZ-induced diabetes. N, normal; C, STZ-induced control; B1 and B2, administrations of BuOH fraction at 100
and 200 mg/kg body weight/d, respectively. Each value is expressed as the meanSE of 6 rats per group. Means with dif-
ferent symbols are significantly different according to Dunnett’s test. * p0.05, ** p0.01, *** p0.001 vs. normal values;
#
p0.05, ## p0.01, ### p0.001 vs. diabetic control values.

tion of the BuOH fraction at a dose of 100 or 200 mg/kg
body weight/d for 20 d, body weight gain increased sig-
nificantly compared to the control. In addition, the liver
and kidney weights of rats with STZ-induced diabetes
showed significant increases compared to those of nor-
mal rats, whereas oral administration of the BuOH frac-
tion significantly inhibited the weight increases of the
liver and kidney.
As shown in Table 5, serum levels of glucose and gly-
cosylated protein in rats with STZ-induced diabetes
were markedly higher than those of normal rats. How-
ever, the elevated levels showed a dose-dependent
decrease after oral administration of the BuOH fraction.
After oral administration of the BuOH fraction at a dose
of 200 mg for 20 d, the serum level of glucose decreased
from 431.9 to 402.8 mg/dL and that of glycosylated
protein from 21.9 to 17.9 nmol/mg protein. On the
other hand, the serum albumin level of rats with STZ-
induced diabetes was lower than that of normal rats,
whereas it was increased in rats orally administered the
BuOH fraction at a dose of 200 mg compared to the
control.
Figure 4 represents the effect of the BuOH fraction on
lipid peroxidation under STZ-induced diabetes. The con-
trol rats with STZ-induced diabetes showed a higher
TBA-reactive substance level in serum, and increased
hepatic and renal mitochondria compared to normal
rats; however, these levels in rats administered the
442 CHO EJ et al.
BuOH fraction orally for 20 d were decreased. The
serum TBA-reactive substance level was lowered from
6.30 to 5.93 and 5.64 nmol/mL after oral administra-
tion of the BuOH fraction at 100 and 200 mg/kg body
weight/d, respectively. The hepatic TBA-reactive sub-
stance level of diabetic control rats was slightly higher
than that of the normal rats without significance, while
that of rats administered the BuOH fraction at 200 mg
was decreased significantly from 0.59 to 0.43 nmol/mg
protein. In addition, the markedly elevated renal TBA-
reactive substance level of diabetic control rats signifi-
cantly declined after administration of the BuOH frac-
tion at 100 and 200 mg for 20 d from 2.45 to 1.61 and
1.49 nmol/mg protein, respectively.
DISCUSSION
Natural antioxidants from vegetables and fruits that
can reduce the risk of chronic diseases such as diabetes
have been the focus of much investigation (9–12). In
particular, broccoli contains many antioxidants, includ-
ing flavonol glucosides, hydroxycinnamic acids, and
sulfur-containing compounds such as glucosinolates;
thus, its protective potential against oxidative stress
could be expected. Therefore, we investigated the anti-
oxidative effect and protective potential against oxida-
tive damage of fractions from broccoli flower under an
in vitro system and verified the ameliorative effects of
the active fractions against diabetic oxidative stress
induced by STZ.
The broccoli flower was reported to contain high phe-
nol content among vegetables and show strong antioxi-
dative activity (24–26). We also previously reported its
antioxidative effect and two active components from
the broccoli flower, 1,2-disinapolygentiobiose and 1-
sinapoly-2-feruloylgentiobiose, as polyphenolic compo-
nents (27). However, studies on the biological activity of
broccoli have mainly focused on the anticancer effect of
its active components, especially sulforaphane, a well-
known compound with an anticancer effect. On the
other hand, broccoli is expected to have other active
components and show beneficial effects against various
pathological conditions.
In the present study, the BuOH fraction showed the
strongest DPPH radical scavenging effect among frac-
tions from the broccoli flower, suggesting that the anti-
oxidative potential of broccoli is mainly attributed to
components of the BuOH fraction, such as flavonol gly-
cosides and hydroxycinnamic acid. The broccoli flower
also exerted NO scavenging activity. NO, a short-lived
free radical, has been demonstrated to play crucial roles
in the homeostatic regulation of cardiovascular, neu-
ronal, and immune systems. Despite these important
physiological functions, the oxidized metabolite of nitro-
gen is also a well-known toxic agent. Furthermore, NO
formed endogenously is thought to promote a number
of chronic inflammatory bowel diseases, septic and
hemorrhagic shock, and certain autoimmune disor-
ders. It is known, for example, that in the presence of
molecular oxygen, NO can form peroxynitrite that can
damage DNA, inhibit a variety of enzymes, and initiate
lipid peroxidation (28, 29). Therefore, we investigated
the scavenging activities of extracts on excess NO using
SNP as an NO radical generator, and verified that the
BuOH fraction scavenged NO most effectively compared
to other fractions, suggesting that the BuOH fraction
exhibits antioxidative potential by scavenging NO as
well as the DPPH radical.
To investigate the protective effect against protein
oxidation induced by free radicals, allophycocyanin is
used as one of the phycobiliproteins of phycobilisomes,
the accessory light-harvesting complexes in cyanobac-
teria and red algae, as a protein donor (30). Modified
proteins and amino acids including protein carbonyls,
carboxymethyl, and carboxyethyl lysine can also be
formed by the direct attack of reactive oxygen species on
protein, which contributes to pathologies such as diabe-
tes, atherosclerosis, and neurodegenerative disease
(31). To investigate the protective potential of the broc-
coli flower against protein oxidation induced by AAPH,
we also used allophycocyanin, characterized by its abil-
ity to rapidly lose fluorescence when exposed to a
source of free radicals (15). As shown in the results, the
BuOH fraction showed the strongest protective effect
against protein damage induced by radicals, implying
its protective potential against oxidation by free radicals
as well as a direct free radical scavenging effect.
From the in vitro results on active fractions from the
broccoli flower, we can also expect the BuOH fraction to
have an ameliorative effect against diabetic oxidative
stress. Diabetes mellitus is a disease characterized by
hyperglycemia caused by increased oxidative stress in
various tissues, with evidence of increased levels of oxi-
dized DNA, proteins, and lipids. In addition, oxidative
stress also triggers a series of cellular responses, includ-
ing the activation of protein kinase C and the transcrip-
tion factor nuclear factor- B, and so forth. Inappropri-
ate activations of these important regulatory molecules
would have deleterious effects on cellular functions,
leading to the pathogenesis of various diabetic compli-
cations including retinopathy, nephropathy, and neur-
opathy (4, 32, 33). Since oxidative stress also contrib-
utes to several deleterious changes caused by STZ,
antioxidants or NO scavengers substantially attenuate
STZ toxicity (34, 35).
Hyperglycemia causes toxic effects on the structure
and function of organs, including the pancreatic islets,
by multiple biochemical pathways and mechanisms of
action such as glucose autoxidation, protein kinase C
activation, methylglyoxal formation, glycation, hex-
osamine metabolism, sorbitol formation, and oxidative
phosphorylation. As hyperglycemia worsens,
cells
steadily undergo deterioration, secrete less and less
insulin, and become part of a downward spiral of func-
tion loss (36). Furthermore, glucose possesses a reactive
aldehyde moiety that reacts non enzymatically with the
amino groups of proteins, forming slowly reversible
amadori products. Accordingly, the elevation of glycosy-
lated protein products has been implicated in diabetic
complications. Therefore, hyperglycemia has an impor-
tant role in the pathogenesis of diabetes and its compli-
 Broccoli Ameliorates Oxidative Damage
cations by increasing protein glycation (37). The pro-
tective effect against diabetes through the measurement
of the glucose level and glycosylated protein levels was
also evaluated. As shown in the results, oral adminis-
tration of the BuOH fraction to diabetic rats dose-depen-
dently reduced serum levels of glucose and glycosylated
protein, indicating an improvement in abnormal glu-
cose metabolism and the reduced toxicity of excessive
glucose. Moreover, under physiological conditions, the
body weight gain of the diabetic rats given the BuOH
fraction was elevated and the increase in the weights of
the liver and kidney under diabetes was inhibited, sug-
gesting that the BuOH fraction from the broccoli flower
ameliorated physiological changes induced by abnor-
mal glucose metabolism.
Serum albumin has inverse relations with coronary
heart disease, stroke, cardiovascular mortality, cancer,
and death from all causes (38). In particular, the condi-
tions under liver and kidney diseases may have a pro-
found influence on albumin plasma elimination. Fur-
thermore, excessive urinary excretion of albumin in
nephritic states is thought to arise from structural
changes in capillary wall permeability (39). Accelerated
protein breakdown is a hallmark of metabolic alter-
ations; thus, protein catabolism has been attributed to
the acute disease process, uremic toxins, inflammatory
mediators, metabolic acidosis, hormonal imbalance,
and insulin resistance (39, 40). The present results
demonstrated that the serum level of albumin in dia-
betic control rats decreased compared to normal rats,
while oral administration of the BuOH fraction to dia-
betic rats slightly increased the serum level of albumin,
indicating that albumin loss induced by urinary elimi-
nation under diabetic conditions was attenuated by oral
administration of the BuOH fraction. Therefore, we con-
sider the increase in the serum albumin level by broc-
coli flower extract administration as a sign of the allevi-
ation of diabetic pathological conditions.
Diabetes leads to the disturbance of lipid profiles,
especially an increased susceptibility to lipid peroxida-
tion (41). In particular, lipid peroxidation products such
as malondialdehyde are formed when polyunsaturated
fatty acyl chains are attacked by hydroxyl radicals, so
they are frequently used to determine the oxidant and
antioxidant balance. The present results indicated the
elevation of lipid peroxidation under diabetes. Other
studies also reported that the formation of lipid peroxi-
dation products in serum and tissues under diabetes is
elevated (42, 43), and leads to diabetic complications
including retinopathy and neuropathy (44, 45). The
decrease in lipid peroxidation by the BuOH fraction
from the broccoli flower implies a protective role against
diabetic oxidative stress induced by STZ.
The present study demonstrates that the BuOH frac-
tion has the strongest antioxidative effect among the
fractions by scavenging radicals and inhibiting protein
oxidation induced by radicals. Furthermore, the effect
of the broccoli flower against oxidative stress induced by
diabetes was also confirmed under an in vivo system.
On the basis of the present study, the related protective
443
mechanisms of the broccoli flower have to be elucidated
and a further study on the active compounds has to be
carried out.
REFERENCES
1) Turrens JF. 2003. Mitochondrial formation of reactive
oxygen species. J Physiol 552: 335–344.
2) Cavalca V, Cighetti G, Bamonti F, Loaldi A, Bortone L,
Novembrino C, De Franceschi M, Belardinelli R, Guazzi
MD. 2001. Oxidative stress and homocysteine in coro-
nary artery disease. Clin Chem 47: 887–892.
3) Bokov A, Chaudhuri A, Richardson A. 2004. The role of
oxidative damage and stress in aging. Mech Ageing Dev
125: 811–826.
4) Scott JA, King GL. 2004. Oxidative stress and antioxi-
dant treatment in diabetes. Ann NY Acad Sci 1031:
204–213.
5) Gibson GE, Huang HM. 2005. Oxidative stress in Alzhe-
imer’s disease. Neurobiol Aging 26: 575–578.
6) Wolff SP. 1993. Diabetes mellitus and free radicals. Br
Med Bull 49: 642–652.
7) King GL, Loeken MR. 2004. Hyperglycemia-induced
oxidative stress in diabetic complications. Histochem Cell
Biol 122: 333–338.
8) West IC. 2000. Radicals and oxidative stress in diabetes.
Diabet Med 17: 171–180.
9) Gillman MW, Cupples LA, Gagnon D, Posner BM, Ellison
RC, Castelli WP, Wolf PA. 1995. Protective effect of fruits
and vegetables on development of stroke in men. JAMA
273: 1113–1117.
10) Steinmetz KA, Potter JD. 1996. Vegetables, fruit, and
cancer prevention: a review. J Am Diet Assoc 96: 1027–
1039.
11) Joshipura KJ, Hu FB, Manson JE, Stampfer MJ, Rimm EB,
Speizer FE, Colditz G, Ascherio A, Rosner B, Spiegelman
D, Willett WC. 2001. The effect of fruit and vegetable
intake on risk for coronary heart disease. Ann Intern
Med 134: 1106–1114.
12) Liu S, Serdula M, Janket SJ, Cook NR, Sesso HD, Willett
WC, Manson JE, Buring JE. 2004. A prospective study of
fruit and vegetable intake and the risk of type 2 diabetes
in women. Diabetes Care 27: 2993–2996.
13) Plumb GW, Lambert N, Chambers SJ, Wanigatunga S,
Heaney RK, Plumb JA, Aruoma OI, Halliwell B, Miller
NJ, Williamson G. 1996. Are whole extracts and puri-
fied glucosinolates from cruciferous vegetables antioxi-
dants? Free Radic Res 25: 75–86.
14) Hatano T, Edamatsu R, Hiramatsu M, Mori A, Fujita Y,
Yasuhara T, Yoshida T, Okuda T. 1989. Effects of the
interaction of tannins with co-existing substances. VI.
Effects of tannins and related polyphenols on superoxide
anion radical, and on 1,1-diphenyl-2-picrylhydrazyl
radical. Chem Pharm Bull 37: 2016–2021.
15) Courderot-Masuyer C, Dalloz F, Maupoil V, Rochette L.
1999. Antioxidant properties of aminoguanidine. Fun-
dam Clin Pharmacol 13: 535–540.
16) Sreejayan, Rao MN. 1997. Nitric oxide scavenging by
curcuminoids. J Pharm Pharmacol 49: 105–107.
17) Green LC, Wagner DA, Glogowski J, Skipper PL, Wish-
nok JS, Tannenbaum SR. 1982. Analysis of nitrate,
nitrite, and [15N]nitrate in biological fluids. Anal Bio-
chem 126: 131–138.
18) Fluckiger R, Winterhalter KH. 1976. In vitro synthesis
of hemoglobin AIC. FEBS Lett 71: 356–360.
19) Naito C, Yamanaka T. 1978. Atherosclerotic disorder
444 CHO EJ et al.
and lipid per-oxidation. Jpn J Geriatr 15: 187–191.
20) Johnson D, Lardy H. 1967. Isolation of liver or kidney
mitochondria. Meth Enzymol 10: 94–96.
21) Jung K, Pergande M. 1985. Influence of cyclosporin A
on the respiration of isolated rat kidney mitochondria.
FEBS Lett 183: 167–169.
22) Mihara M, Uchiyama M. 1978. Determination of mal-
onaldehyde precursor in tissues by thiobarbituric acid
test. Anal Biochem 86: 271–278.
23) Itzhaki RF, Gill DM. 1964. A micro-biuret method for
estimating proteins. Anal Biochem 9: 401–410.
24) Cao G, Sofic E, Prior RL. 1996. Antioxidant capacity of
tea and common vegetables. J Agric Food Chem 44:
3426–3431.
25) Vinson JA, Hao Y, Su X, Zubik L. 1998. Phenol antioxi-
dant quantity and quality in foods: vegetables. J Agric
Food Chem 46: 3630–3634.
26) Chu YF, Sun J, Wu X, Liu RH. 2002. Antioxidant and
antiproliferative activities of common vegetables. J Agric
Food Chem 50: 6910–6916.
27) Piao XL, Kim HY, Yokozawa T, Lee YA, Piao XS, Cho EJ.
2005. Protective effects of broccoli (Brassica oleracea)
and its active components against radical-induced oxi-
dative damage. J Nutr Sci Vitaminol 51: 142–147.
28) Grisham MB, Jourd’Heuil D, Wink DA. 1999. Nitric
oxide. I. Physiological chemistry of nitric oxide and its
metabolites: implications in inflammation. Am J Physiol
276: G315–G321.
29) Li J, Billiar TR. 1999. Nitric oxide. VI. Determinants of
nitric oxide protection and toxicity in liver. Am J Physiol
276: G1069–G1073.
30) Foguel D, Weber G. 1995. Pressure-induced dissociation
and denaturation of allophycocyanin at subzero temper-
atures. J Biol Chem 270: 28759–28766.
31) Dean RT, Fu S, Stocker R, Davies MJ. 1997. Biochemis-
try and pathology of radical-mediated protein oxidation.
Biochem J 324: 1–18.
32) Baynes JW, Thorpe SR. 1999. Role of oxidative stress in
diabetic complications: a new perspective on an old par-
adigm. Diabetes 48: 1–9.
33) Chung SS, Ho EC, Lam KS, Chung SK. 2003. Contribu-
tion of polyol pathway to diabetes-induced oxidative
stress. J Am Soc Nephrol 14: S233–S236.
34) Szkudelski T. 2001. The mechanism of alloxan and
streptozotocin action in
cell of the rat pancreas. Phys-
iol Res 50: 537–546.
35) Bolzan AD, Bianchi MS. 2002. Genotoxicity of strepto-
zotocin. Mutat Res 512: 121–134.
36) Robertson RP. 2004. Chronic oxidative stress as a cen-
tral mechanism for glucose toxicity in pancreatic islet
beta cells in diabetes. J Biol Chem 279: 42351–42354.
37) Ahmed N. 2005. Advanced glycation endproducts-role
in pathology of diabetic complications. Diabetes Res Clin
Pract 67: 3–21.
38) Shaper AG, Wannamethee SG, Whincup PH. 2004.
Serum albumin and risk of stroke, coronary heart dis-
ease, and mortality: the role of cigarette smoking. J Clin
Epidemiol 57: 195–202.
39) Koltun M, Nikolovski J, Strong K, Nikolic-Paterson D,
Comper WD. 2005. Mechanism of hypoalbuminemia in
rodents. Am J Physiol Heart Circ Physiol 288: H1604–
H1610.
40) Druml W. 1998. Protein metabolism in acute renal fail-
ure. Miner Electrolyte Metab 24: 47–54.
41) Giugliano D, Ceriello A, Paolisso G. 1996. Oxidative
stress and diabetic vascular complications. Diabetes Care
19: 257–267.
42) Feillet-Coudray C, Rock E, Coudary C, Grzelkowska K,
Azais-Braesco V, Dardevet D, Mazur A. 1999. Lipid per-
oxidation and antioxidant status in experimental diabe-
tes. Clin Chim Acta 284: 31–43.
43) Pasaoglu H, Sancak B, Bukan N. 2004. Lipid peroxida-
tion and resistance to oxidation in patients with type 2
diabetes mellitus. Tohoku J Exp Med 203: 211–218.
44) Low PA, Nickander KK, Tritschler HJ. 1997. The roles of
oxidative stress and antioxidant treatment in experi-
mental diabetic neuropathy. Diabetes 46: S38–S42.
45) van Reyk DM, Gillies MC, Davies MJ. 2003. The retina:
oxidative stress and diabetes. Redox Rep 8: 187–192.