J. Agric. Food Chem.

1998, 46, 401−406 401

Scavenging Effect of Water Soluble Proteins in Broad Beans on Free

Radicals and Active Oxygen Species
Yoshinori Okada*,† and Mizue Okada‡
Laboratory on Ageing, Chukyo Junior College, Mizunami City, Gifu 509-61, Japan, and
Department of Hygiene, Gifu University School of Medicine, Gifu City, Gifu 500, Japan

Water soluble proteins (WSP) in broad beans, Vicia faba, were purified, and their scavenging effects
on free radicals and active oxygen species were investigated. The purification steps included
ammonium sulfate precipitation followed by sequential chromatography on Sephadex G-75. The
final gel filtration step yielded two peaks of scavenging activity, each containing Mr values of 70
kDa (peak I) and 28 kDa (peak II). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of
peak II fraction gave a single band with Mr of 14 kDa, indicating that peak II protein is dimeric.
WSP exhibited a marked scavenging effect on superoxide, and also an effect on hydrogen peroxide,
but not so much on 1,1-diphenyl-2-picrylhydrazyl radical. WSP had only a small amount of sulfhydryl
groups. Thus, the sulfhydryl groups are not responsible for the scavenging activity of WSP.

Keywords: Broad beans (Vicia faba); antioxidant protein; free radicals; antioxidants

INTRODUCTION and glutathione peroxidase, have already been isolated

from many kinds of organisms. Some of them were
All aerobic organisms are at risk of being damaged tested also as antioxidants with the proper enzymic
by activated oxygen species such as superoxide, hydro- activity for skim milk (Taylor and Richardson, 1980b),
gen peroxide, hydroxyl radical, and singlet oxygen. soybean (Pratt, 1972), yeast (Kim et al., 1988), and
Because of the high reactivity, these agents cause human red blood cells (Lim et al., 1994). Antioxidant
various injuries such as carcinogenesis, inflammation, proteins (PRP) besides those enzymes have also been
and aging (Ames, 1993). The aerobes have natural found in these biological specimens. Among them PRP
antioxidant activities due to enzymes and/or some from yeast or human red blood cells has been well
metabolites. These agents are known to be important investigated. It is shown that they functioned also as
for the prevention of pollution damage (Bandiani et al., sulfur radical scavengers. Moreover, the yeast PRP
1993) in plants as well as various diseases in both plants (YPRP) did not show any known antioxidant enzyme
and animals. Antioxidants are found in many kinds of activity, which was the first one not to (Kim et al., 1988).
fruits and vegetables and include ascorbic acid, R-toco- Sulfhydryl groups on PRP contributed essentially to the
pherol, glutathione, β-carotene, chlorogenic acid, quer- thiol-specific antioxidant activities (Lim et al., 1993).
cetin, and other phenolic compounds (Jones et al., 1992; In the case of human blood cells, PRP (HPRP) had
Larson, 1988). More importantly, some of the proteins known enzymic activities such as SOD, catalase, and
and amino acids have known antioxidant activities. glutathione peroxidase (Lim et al., 1994). HPRP showed
Each of the amino acids of protein has been reported to also the antioxidant activity: it directly scavenged
be either anti-, pro-, or non-oxidant. Cysteine, methion- reactive oxygen species, presumably hydroxyl radical-
ine, histidine, tryptophan, and lysine are of antioxidant like species.
activity (Taylor and Richardson, 1980a). Among them,
Broad beans (Vicia faba) contain 5% protein, 4%
cysteine is the only one being active in a linoleate
carbohydrates, and 15-16% solid residue per 100 g of
emulsion. It is more potent at 10-4 M in a linoleate
fresh weight. Thus, the beans are a good source of
emulsion than tert-butylhydroxyanisole (BHA), tert-
protein and amino acids in the diet as commonly used.
butylated hydroxytoluene, or R-tocopherol and than
It is interesting and important to clarify the PRP
BHA at 10-3 M. The antioxidant activity of cysteine is
activity of broad beans. The aims of this study were
due to its sulfhydryl group, acting as the free radical
(1) to determine what type of oxidants WSP can
scavenger, and thus as the antioxidants, in biological
scavenge; (2) to test the ability of the WSP to protect
and other systems (Taylor and Richardson, 1980a). The
the tert-butylhydroperoxide (BHP)-induced peroxidation
net effect of thiols in many chemical systems as the
of bovine liver homogenate; (3) to evaluate the ability
antioxidants results from the ability of sulfhydryls to
of WSP to chelate metal ions; and (4) to identify proteins
donate the labile hydrogen to free radicals. Various
responsible for the antioxidant activities.
kinds of proteins are also found to be antioxidants or
free radical scavengers in many living organisms.
These enzymes, superoxide dismutase (SOD), catalase, MATERIALS AND METHODS

Chemicals. Catechin, BHA, and cytochrome c were ob-

* Author to whom correspondence should be addressed tained from Sigma Chemical Co., Ltd. (St. Louis, MO).
(telephone +81-572-68-4555; fax +81-572-68-4568). Hydrogen peroxide was purchased from E. Merck (Darmstadt,
† Chukyo Junior College. Germany). DPPH was obtained from Fluka Chemie AG
‡ Gifu University School of Medicine. (Buchs, Switzerland). Sephadex G-75 was the product of

S0021-8561(97)00470-6 CCC: $15.00 © 1998 American Chemical Society

Published on Web 01/30/1998
402 J. Agric. Food Chem., Vol. 46, No. 2, 1998 Okada and Okada

WSP, which has been used for the appraisal of its various
activities as an antioxidant.
Sodium Dodecyl Sulfate-Polyacrylamide Gel Elec-
trophoresis (SDS-PAGE). Peaks I and II were subjected
to SDS-PAGE. The resolving gel cotained 15% acrylamide
(2.6% Bis) and 0.1% SDS in 1.5 M Tris-HCl buffer at pH 8.8,
and the stacking gel contained 4% acrylamide and 0.1% SDS
in 0.5 M Tris-HCl buffer at pH 6.8. The sample proteins were
dispersed in 0.5 M Tris-HCl buffer at pH 6.8 containing 2%
SDS, 0.05% β-mercaptoethanol, and 0.01% bromophenol blue
and heated at 100 °C for 2 min. Electrophoresis was per-
formed usually at 40 mA for 4 h, at room temperature.
Figure 1. Sephadex G-75 column chromatography I of WSP Following electrophoresis, the gels were stained with 0.1%
from broad beans: b, absorbance at 280 nm; 2, superoxide- Coomassie Brilliant Blue (R) in 50% (v/v) methanol and 7%
scavenging activity. Absorbance at 280 nm and superoxide- acetic acid solution for 20 min. After staining, the gels were
scavenging activity were monitored in eluted 3-mL fractions. destained with five changes of a 7% acetic acid solution for a
total period of 24 h and photographed.
Determination of the Effects on Superoxide. This
activity of the WSP to inhibit the superoxide-dependent
reduction of cytochrome c was monitored by measuring the
rate of the increase in absorbance at 550 nm (Nihonbunko UV-
460). The superoxide was generated enzymatically. The
reaction mixture (3 mL) contained in 50 mM potassium
phosphate, pH 7.8, and 0.1 mM ethylenediaminetetraacetic
acid (EDTA): 10 µM ferricytochrome c, 50 µM xanthine, 0.01
µM xanthine oxidase, and an appropriate volume of the test
WSP-fraction (McCord and Fridovich, 1969). The reaction at
25 °C was initiated by the addition of xanthine oxidase to the
mixture. The superoxide- scavenging activity was expressed
Figure 2. Sephadex G-75 column chromatography II of WSP as units/ml sample or units/mg protein, where one unit of the
from broad beans: b, absorbance at 280 nm; 2, superoxide- activity was the amount of enzyme needed to cause half-
scavenging activity. Absorbance at 280 nm and superoxide- maximal inhibition of cytochrome c reduction. All the tests
scavenging activity were assayed in every fractionated eluate were replicated three times, and the results were averaged.
of 3 mL. Determination of the Effects on Hydrogen Peroxide.
The ability of WSP to scavenge hydrogen peroxide was
Pharmacia (Uppsala, Sweden). All other chemicals were measured with a spectrophotometer (Ruch et al., 1989). The
analytical grade products from Wako Pure Chemical Industries reaction at 20 °C was started by mixing an appropriate volume
Ltd. (Osaka, Japan). of the WSP fraction with 2 mL of 2 mM hydrogen peroxide
Protein Concentration. Protein was assayed spectro- solution to make the total volume 3 mL with phosphate-
photometrically either directly on protein solutions at A280 or buffered saline. After 10 min reaction, the change in A230 of
according to the method of Bradford (1976) using bovine serum the reaction mixture was measured against the reference
albumin as the standard. solution containing only the WSP. The molar absorption
Purification. Step 1: Extraction. Broad beans were coefficient of hydrogen peroxide was 81 M-1 cm-1 (Beers and
obtained from a market and immediately shelled by hand, Sizer, 1952). All the tests were replicated three times, and
being stored at -20 °C until used. Frozen specimens (150 g) the results were averaged.
were suspended in a buffer (50 mM potassium phosphate, pH Determination of the Effects on DPPH Radical. Two
7.8) and homogenized for 5 min at 10 000 rpm in a chilled (0 milliliters of WSP solution and 1 mL of DPPH radical solution
°C) Waring blender. The homogenates were centrifuged at 4 were mixed (the final concentration of DPPH was 2.0 × 10-4
°C for 10 min at 10000g to give the water soluble extracts. M) and shaken vigorously. The mixture was left to stand for
Step 2: Ammonium Sulfate Precipitation. The extracts 30 min, and the change in A517 was measured (Blois, 1958).
obtained in step 1 were brought to 40% as to ammonium The same assay was replicated three times, and the results
sulfate by slow addition of 100% ammonium sulfate solution. were averaged.
After being kept for 2 h at 4 °C (all of the following procedures Determination of the Chelating Activity on Metal
were carried out at 4 °C), the suspension was centrifuged for Ions. The chelating activity of WSP on Fe2+ and Cu2+ was
10 min at 10000g. The supernatant obtained was brought to measured according to the method of Shimada et al. (1992).
90% as to ammonium sulfate by the gradual addition of 100% The frozen WSP solution was thawed, and a buffer containing
ammonium sulfate solution. After the same centrifugation hexamine and KCl was added to make the final concentration
procedure, the resulting precipitates were suspended in 50 mM 10 mM and the pH 5.0. Two milliliters of the WSP solution
Tris-HCl buffer, pH 7.4 (gel filtration buffer). was added to 2 mL of 10 mM hexamine buffer, pH 5.0,
Step 3: Gel Filtration I. The suspension (∼5 mL) obtained containing 10 mM KCl and 3 mM FeSO4 or 3 mM CuSO4. After
in step 2 was applied at a flow rate of 0.5 mL/min on a the addition of 0.2 mL of 1 mM tetramethyl murexide to the
Sephadex G-75 column (2.5 × 30 cm), having previously been final solution, absorbance at 480 nm was measured with a
equilibrated with the gel filtration buffer. Three milliliters spectrophotometer. To prevent Fe2+ or Cu2+ oxidation, the
of each eluate was collected. Each of the fractions was measurement was carried out at 20 °C. All of the tests were
subjected to A280 measurement and to the superoxide-scaveng- replicated three times, and the results were averaged.
ing assay, as described later. The fractions including the peak Antioxidant Activity. Bovine liver was homogenized in
(fractions 14-16) were reserved (Figure 1). 19 volumes of 50 mM phosphate buffer, pH 7.4, and 0.5 mL of
Step 4: Gel Filtration II. The reserved fractions were homogenate was added to a mixture of 25 µL of 1.0% aqueous
subjected to the same chromatographic procedure for purifica- solution of BHP, 0.5 mL of a sample solution, and 2.225 mL
tion of WSP. Each of the fractions (fractions 24 and 25) gave of H2O. This reaction mixture was incubated at 37 °C for 15
the results that the activity peak coincided with that of the min, and the content of thiobarbituric acid reactive substances
protein (Figure 2). The fractions eluted at the retention (TBARS) in the mixture was determined according to the
volume of 48 mL (fraction 16: peak I) and 72-75 mL (fractions method of Yoshino et al. (1994) (TBARS 1). As the control,
24 and 25: peak II) from gel filtration II were pooled, divided the homogenate was peroxidized by BHP without the antioxi-
into aliquots, and stored at -20 °C. The protein in these is dants (TBARS 2). The reactions without BHP were carried
Free Radical Scavengers in Broad Bean Proteins J. Agric. Food Chem., Vol. 46, No. 2, 1998 403

Table 1. Purification of WSP from Broad Beansa

scavenging specific activity purification
fraction protein (mg) activity (units) (units/mg) (-fold) recovery (%)
homogenate 1160 23700 20.4 1.00 100
40-90% (w/v) ammonium sulfate fraction 228 5270 23.1 1.13 22.2
chromatography I
sum of fractions 14-16 0.801 51.0 63.7 3.12 0.215
chromatography II
sum of fractions 24-25 0.050 5.35 107 5.25 0.0226
a The starting material was 150 g of broad beans. The superoxide-scavenging activity of WSP was measured at 25 °C by inhibition of
the superoxide-dependent reduction of cytochrome c, measured at 550 nm. Each value is the mean ( standard deviation of three replicate
analyses.

Table 2. Effect of Dialysis on Superoxide-Scavenging

Activity of WSPa
dialysis activity (units/mL)
before 4.49 ( 0.10
after 0.506 ( 0.05
a WSP concentration was 16.7 µg/mL. The activities given were

measured as described under Materials and Methods. The dialysis

against 50 mM Tris-HCl buffer, pH 7.4, was carried out at 4 °C
for 24 h. Each value is the mean ( standard deviation of three
replicate analyses.

out for each of the test substance as the blank (TBARS 3 and
TBARS 4, respectively). The antioxidant potential of the
sample was calculated by using the following equation:

antioxidant activity (%) ) [1 - (TBARS 1 - TBARS 3)/ Figure 3. SDS-PAGE of WSP. The WSP eluted at the
(TBARS 2 - TBARS 4)] × 100 retention volume of 48 mL (lane I ) peak I) and 72-75 mL
(lane II ) peak II) were subjected to SDS-polyacrylamide gel
All of the tests were replicated three times, and the results (15%). Arrows indicate Mr standards (14 000; 24 000; 36 000;
were averaged. 48 000).
Sulfhydryl Assay. Sulfhydryl groups were measured with Table 3. Effect of Heat Treatment on Superoxide-
5,5′-dithiobis(2-nitrobenzoic) acid (DTNB) (Ellman, 1959). Scavenging Activity of WSPa
Typically, 0.1 mL of the sample solution and 3 mL of a buffer
(50 mM Tris-HCl, pH 8.0) were mixed, and the pH of the heat treatment activity (units/mL) inhibition (%)
mixture was readjusted to pH 8.0. To this was added 0.025 before 5.59 ( 0.10 0
mL of DTNB solution (10 mM, in 50 mM potassium phosphate after 5.02 ( 0.15 11.2
buffer, pH 7.0). The reaction mixture was incubated at 25 °C a The scavenging capacity of the WSP was investigated by
for 5 min, and A414 was measured against the blank solution
(water in place of sample). To correct the turbidity, the incubation for 2 min at 100 °C. The activities given were measured
as described under Materials and Methods. Each value is the mean
absorbance of samples without DTNB was measured against
( standard deviation of three replicate analyses.
water, and a standard curve was constructed. All of the tests
were replicated three times, and the results were averaged. Table 4. Superoxide-Scavenging Activity of Various
Substancesa
RESULTS
substance concentration activity (units/mL)
Purification of WSP. Procedures for the purifica- WSP 16.7 µg/mL 4.29 ( 0.15
tion of WSP from broad beans are summarized in Table 1.67 µg/mL 1.58 ( 0.08
1. Application of dialysis irreversibly reduced the 0.167 µg/mL 0.556 ( 0.05
superoxide-scavenging activity of WSP (Table 2). The BHA 75.0 µM 5.00 ( 0.10
catechin 2.50 µM 4.29 ( 0.08
last purification step (Figure 2) yielded two superoxide- L-cysteine 100 µM 2.73 ( 0.12
scavenging activity peaks I (fraction 16) and II (fractions 200 µM 3.00 ( 0.10
24 and 25), at retention volumes of 48 and 72-75 mL, a The activities given were measured as described under
respectively. The molecular weights of these proteins
Materials and Methods. Each value is the mean ( standard
determined from a plot of the logarithm of molecular deviation of three replicate analyses.
weight versus mobility (figure not shown) are ap-
proximately 70 000 and 28 000, respectively. SDS- for 2 min at 100 °C. The half-life of the protein at 100
PAGE of peak II showed a single band with a molecular °C was calculated to be 12.9 min, the decay being
weight of 14 000 (Figure 3, lane II). The peak II protein assumed to be exponential.
may appear to contain two small, identical subunits. Scavenging of Superoxide by WSP. The ability
Peak I yielded a band of 18 000 and several impurity of WSP to scavenge superoxide generated by the xan-
bands of higher molecular weight (Figure 3, lane I). thin/xanthine oxidase system (X-XOD) is shown in Table
Although peak I (WSP) was not completely homoge- 4. The addition of WSP (16.7 µg/mL) to the X-XOD
neous, it was possible to identify the 14-kDa band as system significantly (p < 0.05) diminished the produc-
the polypeptide of WSP in several trials of the same tion of superoxide. Catechin and BHA at concentrations
procedure. of 2.5 and 75 µM, respectively, also exhibited marked
The heat inactivation of the superoxide-scavenging superoxide-scavenging activities. Cysteine showed less
activity of the purified WSP is shown in Table 3. The scavenging activity than WSP. To rule out the inhibi-
activity of the WSP was little changed in the incubation tion of xanthine oxidase activity by the WSP solution,
404 J. Agric. Food Chem., Vol. 46, No. 2, 1998 Okada and Okada

Table 5. Scavenging Effects of Various Substances on Table 7. Sulfhydryl Content of Broad Bean Fractionsa
Hydrogen Peroxidea
fraction sulfhydryl content (µM)
substance concentration H2O2 (mM) inhibition (%)
homogenate 731 ( 10.0
WSP 0 µg/mL 2.00 0 40-90% (w/v) ammonium 446 ( 15.0
0.167 µg/mL 1.98 ( 0.05 1.00 sulfate fraction
1.67 µg/mL 1.86 ( 0.04 4.00 chromatography I
16.7 µg/mL 1.08 ( 0.06 46.0 sum of fractions 14-16 12.9 ( 2.00
BHA 43.0 µM 1.93 ( 0.02 3.50 chromatography II
catechin 14.0 µM 1.70 ( 0.06 15.0 peak I (fraction 16) 5.72 ( 1.00
L-cysteine 100 µM 1.99 ( 0.05 0.500 peak II (fractions 24 and 25) 12.6 ( 1.20
200 µM 1.99 ( 0.08 0.500 a The values given were measured as described under Materials
a The values given were measured as described under Materials and Methods. Each value is the mean ( standard deviation of
and Methods. Each value is the mean ( standard deviation of three replicate analyses.
three replicate analyses.
Table 8. Antioxidant Activity of Various Substancesa
Table 6. Scavenging Effects of Various Substances on
antioxidant
DPPH Radicala
substance concentration activity (%)
substance concentration A517 nm inhibition (%)
WSP 16.7 µg/mL 66.0 ( 1.60
WSP 0 µg/mL 0.300 ( 0.010 0 BHA 240 µM 91.3 ( 1.50
0.167 µg/mL 0.296 ( 0.005 1.33 catechin 8.00 µM 81.0 ( 2.50
1.67 µg/mL 0.280 ( 0.004 6.67 a The activities given were measured as described under
16.7 µg/mL 0.300 ( 0.006 0
BHA 240 µM 0.030 ( 0.005 90.0 Materials and Methods. Each value is the mean ( standard
catechin 8.00 µM 0.048 ( 0.006 84.0 deviation of three replicate analyses.
L-cysteine 100 µM 0.222 ( 0.005 26.0
Table 9. Chelating Effects of Various Substances on Fe2+
200 µM 0.211 ( 0.008 29.7
and Cu2+a
a The values given were measured as described under Materials
A480 nm
and Methods. Each value is the mean ( standard deviation of
three replicate analyses. substance concentration Fe2+ Cu2+
control 0 0.944 ( 0.002 1.211 ( 0.004
we carried out the nonenzymatic reduction of alloxan WSP 16.7 µg/mL 0.695 ( 0.004 1.111 ( 0.004
by reduced nicotinamide nucleotide (Miwa et al., 1982). 1.67 µg/mL 0.907 ( 0.003 1.189 ( 0.003
The results showed that WSP worked as the superoxide 0.167 µg/mL 0.935 ( 0.005 1.207 ( 0.006
EDTA 0.5 M 0.379 ( 0.004 0.388 ( 0.002
scavenger in this system as well (data not shown). citric acid 0.5 M 0.568 ( 0.003 0.611 ( 0.003
Scavenging of Hydrogen Peroxide by WSP. The
a The values given were measured as described under Materials
scavenging activity of WSP on hydrogen peroxide is
and Methods. Each value is the mean ( standard deviation of
shown in Table 5. Though WSP had scavenging activity
three replicate analyses.
on hydrogen peroxide, the effect was less sharp than
that on superoxide. The 46% inhibition (near D50) at
antioxidants. The three substances were effective, and
16.7 µg/mL of WSP gives a specific activity of 55.1 µmol
WSP was the lowest. BHA had the strongest protective
of hydrogen peroxide reduction/mg of WSP. Catechin
effect on the BHP-induced peroxidation of bovine liver
possessed a slight hydrogen peroxide scavenging activity
homogenates. The antioxidant activity was observed to
(15%). Cysteine had negligible scavenging activity.
increase in the following order: WSP < catechin < BHA.
Measurement of DPPH Radical-Scavenging Ac-
Measurements of Chelating Activity on Metal
tivity. The scavenging activity of WSP on DPPH
Ions. The ability of WSP to form complexes with metal
radical is shown in Table 6. It was strange that no
ions is shown in Table 9. Citric acid and EDTA showed
inhibition was observed at 16.7 µg/mL. This could be
chelating effects, but WSP showed no chelating activity
due to some of the contaminants protecting DPPH,
except the case of Fe2+ at the highest concentration.
because of the so much higher protein content of the
WSP specimen used. Catechin and BHA at the con-
DISCUSSION
centrations tested, respectively, exhibited marked scav-
enging activities (>80%). Cysteine also inhibited the The purification procedures for WSP from broad
DPPH radical. beans described in this paper involve ammonium sulfate
Properties of WSP. The ultraviolet-visible spec- fractionation and gel filtration chromatography, but the
trum of the purified scavenging protein is similar to that final enzyme preparation may still contain trace amounts
of most other proteins. At a concentration yielding A280 of other proteins as seen in the electrophoresis photo-
) 0.10, no detectable absorption was observed in the grams. The subunit structure of WSP, therefore, was
range of 320-600 nm, showing that this protein does not clear. In the present study, gel filtration chroma-
not contain prosthetic groups such as heme and flavin. tography yielded two superoxide-scavenging activity
An absorbance intensity of 0.001 could have been peaks I and II. SDS-PAGE of peak II showed a single
detected in the range, easily (data not shown). band with a molecular weight of 14 000 as far as WSP
The number of cysteine residues in WSP was mea- amount used in the experiments. Sephadex column
sured using Ellman’s reagent, DTNB (Table 7). WSP chromatography is a powerful tool for the isolation of
(16.7 µg/mL ≈ 1.2 µM) had a small amount of sulfhydryl WSP. Several investigators (Somers, 1966; Woof et al.,
groups (12 µM). Thus, 10 sulfhydryl groups per 14 kDa 1967) have reported successful separations of flavonoid
polypeptide were detected. compounds on Sephadex columns, even though Sepha-
Comparison of the Antioxidant Activities of dex tends to absorb aromatic compounds, particularly
WSP and Several Known Antioxidants. Table 8 phenols. Sephadex column chromatography was suc-
shows the antioxidant activity of WSP and two known cessfully used for separating WSP in the broad beans.
Free Radical Scavengers in Broad Bean Proteins J. Agric. Food Chem., Vol. 46, No. 2, 1998 405

A DEAE-cellulose column retained HPRP (Lim et al., apparently do not contribute to the scavenging activity
1994), while WSP of broad beans was not adsorbed onto of WSP. Our results suggest that sulfhydryl groups are
the column (data not shown). Thus, the properties of not the principal antioxidant agents in broad beans. The
WSP are somewhat different from those of HPRP. hydroxyl groups and/or carboxyl groups in amino acids
Application of dialysis against a buffer irreversibly may be responsible for the scavenging activity of WSP.
reduced the superoxide-scavenging activity of WSP We thus conclude that the WSP of broad beans has a
(Table 2). This fact may principally be due to the strong hydrogen-donating ability and is a good scaven-
removal of low molecular weight substance(s) necessary ger of the active oxygen species, including superoxide
for the WSP activity. The details, however, are not and hydrogen peroxide. This property seems to be
clear. For the purpose of separating of the scavenging important to explain how the antioxidant activity of
activity peak, a different condition may be more effec- WSP arises.
tive. However, when the fractions (fractions 14-16)
from gel filtration I were subjected to the dialysis ABBREVIATIONS USED
procedure, the WSP sample lost the superoxide-scav-
enging activity. Therefore, we did not use a different WSP, water soluble protein; DPPH, 1,1-diphenyl-2-
medium or even different conditions for the same picrylhydrazyl; DTNB, 5,5′-dithiobis(2-nitrobenzoic) acid;
medium. SDS-PAGE, sodium dodecyl sulfate-polyacrylamide
The molecular weight of WSP was 14 000, while those gel electrophoresis; BHA, tert-butylhydroxyanisole; BHP,
of PRP obtained from human red blood cells and from tert-butylhydroperoxide; EDTA, ethylenediaminetet-
yeast was 25 000 and 27 000, respectively, having been raacetic acid; PBS, phosphate-buffered saline; TBARS,
determined by SDS-PAGE. This indicates that the size thiobarbituric acid reactive substances; X-XOD, xanthin/
of WSP is almost half that of PRPs. xanthine oxidase; PRP, thiol-specific antioxidant pro-
tein; HPRP, thiol-specific antioxidant protein in human
Ilan et al. (1976) showed that superoxide acts as red blood cells; YPRP, thiol-specific antioxidant protein
either an oxidizing or a reducing agent, depending on in yeast; SOD, superoxide dismutase.
the substrate oxidation potential. Superoxide indirectly
initiates lipid oxidation as a result of superoxide and
ACKNOWLEDGMENT
the emerging hydrogen peroxide acting as precursors
of hydroxyl radicals (Kellogg and Fridovich, 1975; We thank Dr. Tatsuichi Iwamura for discussions and
Aurand et al., 1977). Superoxide also decomposes to a critical review of the manuscript. We also thank Ms.
form stronger oxidative species such as hydroxyl radical Wanda Miyata for a critical review of the manuscript.
and hydrogen peroxide, which initiate the peroxidation
of lipid (Dahl and Richardson, 1978). The antioxidant LITERATURE CITED
activity of WSP is considered to affect its scavenging of
superoxide. Hydrogen peroxide has only a weak activity Ames, B. N.; Shigenaga, M. K.; Hagen, T. M. Oxidants,
on the initiation of lipid peroxidation (Namiki, 1990). antioxidants, and the degenerative diseases of aging. Proc.
Natl. Acad. Sci. U.S.A. 1993, 90, 7915-7922.
However, its activity as an active oxygen species comes
Aurand, L. W.; Boonme, N. H.; Giddings, G. G. Superoxide
from its potential to produce a highly reactive hydroxyl and singlet oxygen in milk lipid peroxidation. J. Dairy Sci.
radical through the Fenton reaction (Cohen and Heik- 1977, 60, 363-369.
kila, 1974). Therefore, the ability of WSP to scavenge Badiani, M.; Schenone, G.; Paolacci, A. R.; Fumagalli, I. Daily
hydrogen peroxide may contribute to inhibition of the fluctuations of antioxidants in bean (Phaseolus vulgaris L.)
peroxidation of lipids. leaves as affected by the presence of ambient air pollutants.
By using the method of Shimada et al. (1992), we Plant Cell Physiol. 1993, 34, 271-279.
could not show that WSP was capable of chelating Fe2+ Beers, R. F.; Sizer, I. W. A spectrofluorometric method for
and Cu2+ (Table 9). This implies that transferrin and measuring the breakdown of hydrogenperoxide by catalase.
J. Biol. Chem. 1952, 195, 133-140.
lactoferrin can afford some degree of chelation, but WSP Blois, M. S. Antioxidant determination by the use of a stable
may not be one of those structures. This leads to the free radical. Nature 1958, 4617, 1198-1200.
conclusion that, in vivo, WSP would be unlikely to Bradford, M. M. A rapid and sensitive method for the quan-
chelate sufficiently the endogenous metal ions. titation of microgram quantities of protein utilizing the
The fact that YPRP and HPRP possessed neither principle of protein-dye binding. Anal. Biochem. 1976, 72,
superoxide dismutase nor catalase is based on the direct 248-254.
measurement of these enzyme activities by using the Cohen, G.; Heikkila, R. E. The generation of hydrogen perox-
protein preparations, while WSP from broad beans ide, superoxide radical and hydroxyl radical by P-hydroxy-
dopamine, dialuric acid, and related cytotoxic agents. J. Biol.
possessed antioxidant activity toward superoxide and
Chem. 1974, 249, 2447-2452.
hydrogen peroxide. These results clearly show the Dahl, M. K.; Richardson, T. Photogeneration of superoxide
structural differences between PRP and WSP. anion in serum of bovine milk and in model systems
We do not yet know exactly how WSP works as the containing riboflavin and amino acids. J. Dairy Sci. 1978,
antioxidant agent. A plausible mechanism may be due 61, 400-407.
to the existence of sulfhydryl groups in a small amount, Ellman, G. L. Tissue sulfhydryl groups. Arch. Biochem. Bio-
the content of which was calculated to be 10 sulfhydryl phys.1959, 82, 70-77.
groups per one 14-kDa WSP molecule. WSP showed Frei, B.; Stocker, R.; Ames, B. N. Antioxidant defenses and
significant activity to scavenge superoxide upon being lipid peroxidation in human blood plasma. Proc. Natl. Acad.
Sci. U.S.A. 1988, 85, 9748-9752.
compared with cysteine: ∼17 () 200 µM/12 µM) times Ilan, Y. A.; Gzapski, G.; Meisel, D. The one-electron-transfer
more active than cysteine, being compared at nearly the redox potential of free radical I. The oxygen/superoxide
same activity observed. The concentration dependence system. Biochim. Biophys. Acta 1976, 430, 209-224.
of the cysteine-scavenging activity was much different Jones, D. P.; Coates, R. J.; Flagg, E. W.; Eley, J. W.; Block,
from that of WSP. Cysteine does not scavenge hydrogen G.; Greenberg, R. S.; Gunter, E. W.; Jackson, B. Glutathione
peroxide, but WSP did. Therefore, sulfhydryl groups in foods listed in the National Cancer Institute’s health
406 J. Agric. Food Chem., Vol. 46, No. 2, 1998 Okada and Okada

habits and history food frequency questionnaire. Nutr. Pratt, D. E. Water soluble antioxidant activity in soybeans.
Cancer 1992, 17, 57-75. J. Food Sci. 1972, 37, 322-323.
Kellogg, E. W., III; Fridovich, I. Superoxide, hydrogen peroxide Ruch, K. J.; Cheng, S. J.; Klaunig, J. E. Prevention of
and singlet oxygen in lipid peroxidation by a xanthine cytotoxicity and inhibition of intercellular communication
oxidase system. J. Biol. Chem. 1975, 250, 8812-8817. by antioxidant catechin isolated from Chinese green tea.
Kim, K.; Kim, I. H.; Lee, K. Y.; Rhee, S. G.; Stadtman, E. R. Carcinogenesis 1989, 10, 1003-1008.
The isolation and purification of a specific “protector” protein Shimada, K.; Fujikawa, K.; Yahara, K.; Nakamura, T. Anti-
which inhibits enzyme inactivation by a thiol/Fe(III)/O2 oxidative properties of xanthan on the autoxidation of
mixed-function oxidation system. J. Biol. Chem. 1988, 263, soybean oil in cyclodextrin emulsion. J. Agric. Food Chem.
4704-4711. 1992, 40, 945-948.
Larson, R. A. The antioxidants of higher plants. Phytochem- Somers, T. C. Wine tannins isolation of condensed flavonoid
istry 1988, 27, 969-978. pigments by gel-filtration. Nature 1966, 209, 368-370.
Lim, Y. S.; Cha, M. K.; Kim, H. K.; Uhm, T. B.; Park, J. W.; Taylor, M. J.; Richardson, T. Antioxidant activity of cysteine
Kim, K.; Kim, I. H. Removals of hydrogen peroxide and and protein sulfhydryls in a linoleate emulsion oxidized by
hydroxyl radical by thiol-specific antioxidant protein as a hemoglobin. J. Food Sci. 1980a, 30, 1223-1227.
possible role in vivo. Biochem. Biophys. Res. Commun. 1993, Taylor, M. J.; Richardson, T. Antioxidant activity of skim
192, 273-280. milk: effect of heat and resultant sulfhydryl groups. J. Dairy
Lim, Y. H.; Cha, M. K.; Yun, C. H.; Kim, H. K.; Kim, K.; Kim, Sci. 1980b, 63, 1783-1795.
I. H. Purification and characterization of thiol-specific Woof, J. B.; Pierce, J. S. Separation of complex mixtures of
antioxidant protein from human red blood cell: a new type polyhydroxy phenols on columns of Sephadex. J. Chro-
of antioxidant protein. Biochem. Biophys. Res. Commun. matogr. 1967, 28, 94-103.
1994, 199, 199-206. Yoshino, K.; Hara, Y.; Sano, M.; Tomita, I. Antioxidative effects
McCord, J. M.; Fridovich, I. Superoxide dismutase. An enzymic of black tea theaflavins and thearubigin on lipid peroxida-
function for erythro cuprein (hemo cuprein). J. Biol. Chem. tion of rat liver homogenates induced by tert-butyl hydro-
1969, 244, 6049-6055. peroxide. Biol. Pharm. Bull. 1994, 17, 146-149.
Miwa, I.; Okuda, J. Nonenzymatic reduction of alloxan by
reduced nicotinamide nucleotide. Biochem. Pharmacol. 1982, Received for review June 4, 1997. Revised manuscript received
31, 921-925. November 17, 1997. Accepted November 20, 1997.
Namiki, M. Antioxidants/antimutagens in foods. Crit. Rev.
Food Sci. Nutr. 1990, 29, 273-300. JF970470L