Antioxidative activities of oxindole-3-acetic acid derivatives from supersweet corn powder

Article abstract of DOI:10.1271/bbb.100124

The components contributing to the antioxidative activity of supersweet corn powder (SSCP), which is commonly used in corn soup and snacks in Japan, were clarified and the effects investigated. 7-(0-/3-Glucosy-loxy)oxindole-3- acetic acid (GOA) was found

Full text of DOI:10.1271/bbb.100124

Biosci. Biotechnol. Biochem., 74 (9), 1794–1801, 2010
Antioxidative Activities of Oxindole-3-acetic Acid Derivatives
from Supersweet Corn Powder
y
Naoki MIDOH,1; Akihito TANAKA,1 Makiko NAGAYASU,1 Chie FURUTA,2 Katsuya SUZUKI,2
2
Takeshi ICHIKAWA,1 Takashi ISOMURA,1 and Kenzo NOMURA
1Knorr Foods Co., Ltd., 2-12-1 Simonoge, Takatsu-ku, Kawasaki 213-8505, Japan
2Ajinomoto Co., Inc., 1-1 Suzuki-cho, Kawasaki-ku, Kawasaki 210-8681, Japan
Received February 22, 2010; Accepted June 12, 2010; Online Publication, September 7, 2010
[doi:10.1271/bbb.100124]
The components contributing to the antioxidative
activity of supersweet corn powder (SSCP), which is
commonly used in corn soup and snacks in Japan, were
clarified and the effects investigated. 7-(O-ꢀ-Glucosy-
loxy)oxindole-3-acetic acid (GOA) was found to be the
component most strongly contributing to the 1,1-
diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging
activity of the 80% ethanol extract of SSCP, and the
presence of its aglycone, 7-hydroxy-oxindole-3-acetic
acid (HOA) was confirmed. GOA and HOA respectively
contributed 35.1% and 10.5% to the DPPH radical-
scavenging activity of the 80% ethanol extract of SSCP.
Mice orally administered with HOA at doses of both 500
and 1500 mg/kg showed a significantly lower (p < 0:05)
level of thiobarbituric acid reactive substances (TBARS)
in the plasma than the vehicle-treated control. These
results suggest that GOA and HOA were at least partly
involved in the antioxidative activity of SSCP in vitro
and that HOA might have possessed antioxidative
activity in vivo.
corn consumed as a vegetable. Immature kernels are
harvested in which the water content is high.9) It has been
shown in supersweet corn that starch synthesis is
inhibited via the mutation of a gene involved in the
synthesis of starch from glucose (gene mutation sh2:
ADP-glucose pyrophosphorylase), increasing the sugar
content and contributing to the sweet taste.10) The sugar
content reaches approximately 30% of the dry weight,
markedly differing from other corn types. This is the
reason why supersweet corn is utilized as a vegetable,
and it is also used as a component of processed foods. In
particular, supersweet corn powder (SSCP), which is
manufactured by heating, drying and crushing super-
sweet corn, is commonly employed in Japan as a material
to add flavor/color to powdered corn soup and snacks.11)
The antioxidative activities of heated/processed veg-
etables have been considered to be less potent than those
of fresh vegetables; however, studies using corn and
tomato have shown that heating decreased the level of
L-ascorbic acid, whereas it increased their antioxidative
activities.12,13) We therefore became interested in the
antioxidative activity of SSCP manufactured via a
heating process in comparison with that of commercially
available supersweet corn.
Key words: supersweet corn; antioxidative activity;
oxindole-3-acetic acid
Oxidative stress plays an important role in the devel-
opment of cancer, diabetes, and cardiovascular disease.
A large number of food-derived, natural, antioxidants
have been identified, and those in foods consumed in
daily life may inhibit development of the clinical
conditions of such diseases.1,2) The levels of antioxidants
in grains, vegetables, and fruit are high, and the
consumption of these foods may be important to prevent
oxidative stress-related disorders.3–6) Grains are the main
supply source of antioxidants in daily life. In particular,
the antioxidative activity of corn is relatively high,7) and
its consumption methods vary; corn may therefore be a
possible source of antioxidative substances.
Dent corn, which is consumed as a grain, comprises
the greatest proportion of corn consumption. This is
harvested as dry kernels and is employed as a food
component, including corn flour for corn bread and
cereals, oil, high-fructose corn syrup, and alcohol, as
well as feed for domestic animals.8) On the other hand,
supersweet corn comprises the greatest proportion of
Supersweet corn has been reported to contain such
antioxidants as lutein, zeaxanthin, tocopherols, ferulic
acids, and L-ascorbic acid.14–16) These substances are
thought to contribute to the antioxidative activity of
supersweet corn and also of SSCP; however, our
preliminary analysis showed that these substances
contributed to only part of the 1,1-diphenyl-2-picrylhy-
drazyl (DPPH) radical-scavenging activity of the SSCP
extract; other antioxidants in the SSCP extract were
therefore predicted to contribute to its DPPH radical-
scavenging activity.
To clarify the components contributing to the anti-
oxidative activity of SSCP, we evaluated its DPPH
radical-scavenging activity and identified two oxindole-
3-acetic acid derivatives, 7-(O-ꢀ-glucosyloxy)oxindole-
3-acetic acid (GOA, Fig. 1) and its aglycone, 7-hydroxy-
oxindole-3-acetic acid (HOA) which has been reported
to be isolated as an antioxidant from corn gluten meal.17)
We measured their contents and antioxidative activities
in this study.
y
To whom correspondence should be addressed. Tel: +81-44-811-3117; Fax: +81-44-812-7511; E-mail: naoki mido@knorr.co.jp
Abbreviations: SSCP, supersweet corn powder; DM, dry matter; DPPH, 1,1-diphenyl-2-picrylhydrazyl; GOA, 7-(O-ꢀ-glucosyloxy) oxindole-3-
acetic acid; HOA, 7-hydroxy-oxindole-3-acetic acid; TBARS, thiobarbituric acid reactive substances; MDA, malondialdehyde

Antioxidants of Supersweet Corn Powder
1795
the same condition was continued until 100 min. Detection was at an
absorbance of 250 nm. The eluate was collected at 8-min intervals by
dividing it into 11 fractions. Each fraction was dried, dissolved in
3.0 ml of 80% ethanol, and used as a sample to measure the DPPH
radical-scavenging activity.
H
H
COOH
COOH
O
O
N
H
N
H
R
R
HPLC-MS analysis. Ten microliters of the SSCP extract was eluted
at 30 ꢁC, using HPLC with a COSMOSIL 5C18-MS-II column
(ꢁ4:6 ꢂ 250 mm, Nacalai Tesque). Elution was performed employing
a linear gradient, with a flow velocity of 0.3 ml/min. We employed
solutions A (0.1% formic acid) and B (70% methanol and 0.1% formic
acid), with the initial concentration of solution A set as 100%. Gradient
elution was started after 15 min, using respective concentrations of
solution A 30, 50, 65, 100, 135, and 160 min after the start of elution of
90, 80, 70, 60, 40, and 0%. This 0% condition was continued until
175 min. Detection was performed with a Micromass Quattro micro
API mass spectrometer (Waters Corp., Milford, MA, USA) in the
positive-ion electrospray mode. A drying nitrogen gas temperature of
300 ꢁC, drying gas flow rate of 10 l/min, fragmentation voltage of
20 V, and capillary voltage of +4,600 V were used.
GOA: R = O-β-D-glucose
HOA: R = OH
Fig. 1. Structures of GOA and HOA.
The two-headed arrow indicates the equilibrium of two stereo-
isomers.
Materials and Methods
Chemicals. 1,1-Diphenyl-2-picrylhydrazyl (DPPH), L-ascorbic acid,
Trolox, ferulic acid, and indole-3-acetic acid were purchased from
Sigma-Aldrich Co. (St. Louis, MO, USA). ꢀ-Glucosidase was
purchased from Oriental Yeast Co. (Tokyo, Japan), and thiobarbituric
acid was purchased from Wako Pure Chemical Industries (Osaka,
Japan).
Purification of the antioxidative substances. SSCP (36 g) was
extracted twice with 90 ml of 80% ethanol, and the liquid extract was
dried by using a concentrating centrifuge. The dried extract was
dissolved in methanol, and the target fraction was collected by
conducting HPLC several times on 30 ml of the extract eluted at room
temperature by HPLC with a CAPSELPAK C18 UG80 column
(ꢁ20 ꢂ 250 mm, Shiseido Co., Tokyo, Japan). Elution was performed
at a flow velocity of 6 ml/min for 30 min, employing 25% methanol
and 0.1% formic acid as vehicles. The next sample was then injected.
The target fraction was collected, dried by using the concentrating
centrifuge, and dissolved in a small volume of methanol. The dried
sample was subsequently purified as a single substance by conducting
HPLC several times on 40 ml of the fraction eluted at 30 ꢁC, using
HPLC with a COSMOSIL 5C18-MS-II column (ꢁ4:6 ꢂ 250 mm,
Nacalai Tesque). Elution was performed by employing a linear
gradient with a flow velocity of 1 ml/min. We employed solutions A
(15% methanol and 0.1% formic acid) and B (99.9% methanol and
0.1% formic acid) with the initial concentration of solution A set at
100%. Gradient elution was started after 20 min, the concentration of
Materials. Corn flour (dent corn), popcorn, supersweet corn,
tomato, spinach, pumpkin, onion, potato, barley, wheat, rice, and
soybean were purchased from a market for use as materials. The fresh
materials, supersweet corn, tomato, spinach, pumpkin, onion and
potato, were freeze-dried and crushed. The dry products, corn flour and
popcorn, were crushed to prepare dry powder. SSCP was purchased
from Knorr Trading Co. (Kawasaki, Japan); SSCP was manufactured
from the kernels of supersweet corn which were harvested, crushed,
heated, dried, and processed to a powder.11) This powder was stored at
ꢀ20 ꢁC until needed.
DPPH radical-scavenging activity. Each agricultural product
sample was prepared by mixing 1 g of dry powder with 10 ml of
80% ethanol, agitating at 280 rpm for 20 min, and passing through a
0.45-mm filter for use as the test sample.
Stepwise-diluted 37.5 ml of each sample was mixed with 75 ml of a
0.1 M MES (pH 6.0)/10% ethanol solution and 37.5 ml of 400 mM
DPPH in ethanol, and the solution was kept at room temperature for
20 min. A DPPH-free reactive solution was simultaneously prepared as
a blank, and the absorbance of each solution at 520 nm was measured.
The DPPH radical-scavenging activity of each solution was calculated
from assay lines of Trolox (0, 12.5, 25, 50, 100, 200, and 400 mM) and
is expressed as the Trolox equivalent (Trolox eq.). The activity of each
product was determined by using four samples differing in origin. The
activity of each fraction was measured by using three samples
employing high-performance liquid chromatography (HPLC). The
activity of each substance was determined three times.
solution
A 22 min after the start of elution being 30%. This
concentration was held until 32 min, before a linear return to the
initial condition (100% of solution A) up to 35 min after the start of
elution, and the same concentration held until 50 min. The next sample
was then injected. The target fraction was dried in the concentrating
centrifuge and then overnight in a desiccator, substituting the
atmosphere with nitrogen gas and then storing at ꢀ80 ꢁC. We obtained
approximately 10 mg of the purified substance.
Structure of the purified substance. A 1-mg sample of the purified
substance was dissolved in 1 ml of water, and the solution mixed with
18 units of ꢀ-glucosidase. This solution was reacted at 30 ꢁC for 2 h,
and then mixed with 5 ml of ethanol to stop the reaction. After the
precipitated protein had been removed by centrifugation, the super-
natant was dried in the concentrating centrifuge. A mixture of a formic
acid solution (pH 3) and 1-butanol (1:1) was added to this sample and
distributed to the respective layers. The formic acid solution layer
containing constitutive sugars was concentrated and dried, and the
material compared with an authentic substance by using 1H-NMR and
HSQC with an AV-600 spectrometer (Bruker Co., Rheinstetten,
Germany). The 1-butanol layer containing the aglycone was dried,
purified by HPLC as just described, and analyzed by 1H-NMR, HMBC,
NOESY, and H-H COSY with the Bruker AV-600 spectrometer, and
by 13C-NMR, TOCSY, and HSQC with a DMX-600 spectrometer
(Bruker Co., Rheinstetten, Germany).
Phenolic compound content. The polyphenolic content of an extract
prepared in the presence of 80% ethanol was measured by the Folin &
Ciocalteu method with gallic acid as a standard sample, and is
expressed as the gallic acid equivalent (gallic acid eq.). We added 100
ml of the sample, 100 ml of the Folin & Ciocalteu phenol reagent (Wako
Pure Chemical Industries, Osaka, Japan), and 200 ml of a saturated
sodium carbonate solution to 1.6 ml of purified water, let the solution
stand for 30 min, and measured the absorbance at 760 nm. The activity
of each product was determined for four samples differing in origin.
Fractionation by HPLC. Five milliliters of an SSCP extract was
eluted at 35 ꢁC by HPLC with a COSMOSIL 5C18-MS-II column
(ꢁ20 ꢂ 250 mm, Nacalai Tesque, Kyoto, Japan) to fractionate the
eluate. Elution was performed with a linear gradient, using a flow
velocity of 10 ml/min. We employed solutions A (0.1% formic acid)
and B (99.9% methanol and 0.1% formic acid) with the initial
concentration of solution A set at 98%. Its subsequent concentrations
10, 30, 35, 60, and 65 min after the start of elution were 79, 77, 59, 58,
and 0%, respectively. This 0% concentration of solution A was
continued until 80 min. A linear return to the initial condition (98% of
solution A) was performed until 85 min after the start of elution, and
Purified substance (GOA). LC/MS (ESI-positive): m=z 370.1
½M þ Hꢃþ; UV (H2O): ꢂmax 248, 283 nm.
Aglycone (HOA). LC/MS (ESI-positive): m=z 207.9 ½M þ Hꢃþ;
UV (H2O): ꢂmax 246, 290 nm; 1H-NMR (DMSO-d6) ꢃ: 2.63 (1H, dd,
J ¼ 7:2, 16.7 Hz), 2.86 (1H, dd, J ¼ 4:3, 16.8 Hz), 3.67 (1H, m), 6.86
(1H, m), 6.94 (1H, dd, J ¼ 4:5, 6.8 Hz), 7.10 (1H, dd, J ¼ 3:3, 8.2 Hz),
10.19 (1H, s); 13C-NMR (DMSO-d6) ꢃ: 34.1 (d), 42.5 (s), 115.7 (d),
117.8 (d), 121.9 (s), 130.5 (s), 132.2 (d), 141.5 (d), 172.2 (s), 178.1 (d).

1796
N. MIDOH et al.
COOEt
COOH
OK
NO2
O
NO2
NO2
NO2
O
OH
O
O
1
2
3
4
O
HO
COOH
O
COOH
NO2
O
O
N
H
O
N
H
N
O
O
H
OH
O
5
6
7
HOA
Fig. 2. Chemical Synthesis of HOA.
Quantification of GOA and HOA. One hundred microliters of
500 mM flavone was added to 150 mg of a sample as an internal
standard, and the mixture extracted three times with 1 ml of 80%
ethanol. After centrifugation, the supernatant was dried in the
concentrating centrifuge, dissolved in 500 ml of methanol, and passed
through a 0.25-mm filter for use as the sample for analysis. A 10-ml
amount of the sample was eluted at 35 ꢁC by using HPLC with a
COSMOSIL 5C18-MS-II column (ꢁ4:6 ꢂ 250 mm, Nacalai Tesque).
Elution was performed with a linear gradient at a flow velocity of
1.0 ml/min. We employed solutions A (0.1% formic acid) and B
(99.9% methanol and 0.1% formic acid) with the initial concentration
of solution A set at 98%, before its respective concentrations 15, 30,
and 35 min after the start of elution being 80, 70, and 0%. This
condition was continued up to 45 min. A linear return to the initial
condition (98% Solution A) was subsequently made up to 50 min after
the start of elution, and the same condition was continued until 60 min.
The next sample was then injected, detecting at an absorbance of
250 nm. We calculated the level of each substance from the assay line
of the standard solution that was simultaneously analyzed with each
sample. The detection limit for GOA was 0.04 mmol/g of dry matter
(DM), and that for HOA was 0.03 mmol/g DM when employing this
method. Four samples differing in origin were employed in the analysis
of corn, while three samples differing in origin were employed when
analyzing GOA and HOA in the other products.
methyl ether layer, and dehydrated with anhydrous magnesium sulfate.
The vehicle was removed to obtain 174.7 g of compound 4.
Compound 5 was synthesized from compound 4, as described by
King et al.20) In a nitrogen atmosphere, a mixture of compound 4
(174.7 g) and a 0.05 N sodium hydroxide solution (2,183 ml) was
cooled to 0 ꢁC, and a 6% hydrogen peroxide solution was dripped into
the mixture. This was then mixed with 1,310 ml of dioxane, placed at
room temperature for 1 h, and cooled. The pH value was subsequently
adjusted to 3 by adding 1 N hydrochloric acid. The formed crystals
were filtered out, washed, and dried under reduced pressure to obtain
138.0 g of compound 5.
HOA was synthesized from compound 5, as described by Lewer
et al.21) Under an atmosphere of nitrogen, a mixture of compound 5
(97.4 g), acetic acid (1,770 ml), and zinc (177 g) was agitated at 110 ꢁC
for 4 h. After this solution had been filtered, the collected compound
was washed in heated acetic acid and then concentrated under reduced
pressure. This concentrate was purified by using silica gel column
chromatography (hexane/ethyl acetate/chloroform = 3:1:4) to obtain
29.2 g of compound 6. In a nitrogen atmosphere, a tetrahydrofurane
solution (27.0 g/221 ml) containing compound 6 was dripped into a
tetrahydrofurane suspension (7.39 g/123 ml) containing sodium hy-
droxide at room temperature, and the mixture agitated for 2 h. Dibenzyl
oxalate (33.6 g) was then dripped in, and the mixture agitated
overnight. The solution was mixed with 1 N hydrochloric acid to
prepare a neutral mixture which was concentrated under reduced
pressure and further mixed with 1 N hydrochloric acid until the pH
value reached 2. The formed crystals were filtered out and washed to
obtain 41.6 g of compound 7. In a nitrogen atmosphere, 16.2 g of 10%
palladium-carbon was added to the mixture of compound 7 (41.5 g),
high-concentration sulfuric acid (2.0 ml), and acetic acid (808 ml). The
solution was agitated at room temperature for 26 h, and then mixed
with 1.9 ml of high-concentration sulfuric acid, 758 ml of acetic acid,
and 15.1 g of 10% palladium-carbon. The mixture was agitated at room
temperature for 4 h, mixed with 31.3 g of sodium acetate, and filtered.
The residue was washed in ethyl acetate, concentrated under reduced
pressure, mixed with 1 N hydrochloric acid, and then extracted with
ethyl acetate. The organic vehicle layer was subsequently dehydrated
with anhydrous sodium sulfate, the vehicle was removed, and the
residue was washed and dried under reduced pressure to obtain 17.4 g
of HOA. This product was dissolved in isopropyl alcohol to remove the
trace vehicle, and the solution dried under reduced pressure to obtain
16.2 g of HOA.
Chemical synthesis of HOA (Fig. 2). Compounds 2 and 3 were
synthesized from compound 1, as described by Cogan et al.18) Under
an atmosphere of nitrogen, 145.6 g of benzyl bromide was dripped
into a mixture of 1-benzyloxy-3-methyl-2-nitrobenzene (130.4 g),
potassium carbonate (147.1 g), and acetonitrile (2,077 ml) at 35–
40 ꢁC. The solution was agitated at 75 ꢁC for 2.5 h after dripping,
cooled to room temperature, mixed with water, and extracted with
ethyl acetate. The organic vehicle layer was then washed and
dehydrated with anhydrous magnesium sulfate, and the vehicle was
removed to obtain 203.5 g of compound 2. In a nitrogen atmosphere,
131.4 g of diethyl oxalate was added to a mixture of potassium
t-butoxide (100.9 g), tetrahydrofurane (899 ml), and t-butyl methyl
ether (2,936 ml). Compound 2 (a t-butyl methyl ether solution, 219 g/
367 ml) was dripped into this mixed solution on ice, heated and
agitated at perfusion temperature for 24 h, and then cooled to 5 ꢁC. The
formed crystals were filtered out and then washed and dried under
reduced pressure to obtain 258.9 g of compound 3.
Compound 4 was synthesized from compound 3, as described by
Beer et al.19) In a nitrogen atmosphere, 787 ml of a 5% sodium
hydroxide solution was added to a mixture of compound 3 (109.7 g),
ethanol (1,055 ml), and water (714 ml). The mixed solution was
agitated at room temperature for 2.5 h and then dried under reduced
pressure. This procedure was repeated for one more batch. These dried
samples were mixed and extracted with t-butyl methyl ether. The
aqueous layer was mixed with 1 N hydrochloric acid to adjust the pH
value to 3. The solution was subsequently extracted with a mixed
chloroform/methanol (3:1) vehicle, added to the foregoing t-butyl
Ex vivo plasma oxidation. Fifty or 150 mg/ml of the HOA solution
in 50% polyethylene glycol 400, or a control vehicle (50% poly-
ethylene glycol 400) was administered orally to 9-week-old male
C57BL/6J mice (n ¼ 6 per group) at a dosage of 10 ml/kg. Blood was
collected 120 min after this administration, mixed with heparin, and
centrifuged to isolate the plasma. Since a preliminary time-course
experiment had given the highest activity at 120 min, the time for
blood collection was determined as 120 min. Plasma samples were
diluted with physiological saline at a ratio of 10, and mixed with

Antioxidants of Supersweet Corn Powder
1797
CuSO4 to prepare a final concentration of 100 mM. Each of these
samples was reacted at 37 ꢁC for 240 min. Thiobarbituric acid reactive
substances (TBARS) in the reactive solution were quantified, as
described by Ohkawa et al.,22) TBARS values being expressed as the
malondialdehyde equivalent (MDA eq.). This experiment was ap-
proved by the Animal Experiment Ethics Committee of Ajinomoto
Co., Ltd.
scavenging activity and the phenolic compound con-
tent, with a Pearson correlation coefficient of 0.956
(p ¼ 7:89 ꢂ 10ꢀ9).
Identification of the SSCP components with DPPH
radical-scavenging activity
We measured the activity of each fraction by using
HPLC to identify the components with DPPH radical-
scavenging activity contained in SSCP (Fig. 3). Fraction
4 showed the strongest activity (Table 2) and was
analyzed by using HPLC-MS. Two peaks were noted,
with an ESI-MS value of 370.1 m=z ½M þ Hꢃþ,
suggesting that these two peaks were stereoisomers of
7-(O-ꢀ-glucosyloxy)oxindole-3-acetic acid (GOA,
Fig. 1);24,25) however, there was no sample of this, so
Hydrolysis of GOA with the brush border membrane fraction. A rat
jejunum-derived brush border membrane fraction was extracted by
modifying the method described by Kessler et al.23) Briefly, a
laparotomy was performed under ether anesthesia on 6- to 8-week-
old SD rats to extirpate the jejunum. The inner area of the jejunum was
washed in physiological saline. The mucosa was exfoliated by using a
slide glass, mixed with a 2 mM Tris–HCl solution (pH 7.0) containing
50 mM D-mannitol, and homogenized. Calcium chloride was added to
the homogenate to achieve a final concentration of 10 mM. The solution
was allowed to stand at 4 ꢁC for 15 min and then centrifuged (750 ꢂ g
at 4 ꢁC for 30 min). The supernatant was then also centrifuged
(9;000 ꢂ g at 4 ꢁC for 30 min). The precipitate was mixed with 10 mM
phosphate-buffered saline (pH 7.0) to prepare a final concentration of
100 mg/ml, and then stored at ꢀ80 ꢁC as the brush border membrane
fraction. This experiment was approved by the Animal Experiment
Ethics Committee of Ajinomoto Co., Ltd.
Table 1. DPPH Radical-Scavenging Activities and Phenolic Con-
tents of SSCP and the Other Corn Types
DPPH activityA
(mmol Trolox
eq./g DM)
Phenolic contentA
(mmol gallic acid
eq./g DM)
Sample
A 1-ml sample of the brush border membrane fraction was added to
100 ml of 0.2 mM GOA and a 50 mM maleic acid buffer (pH 6.0), and
the mixture reacted at 37 ꢁC for 24 h. A 5-ml aliquot of the solution was
used as a sample for HPLC according to the ‘‘Quantification of GOA
and HOA.’’
Corn flour (dent corn)
Popcorn
2:2 ꢄ 0:9a
3:2 ꢄ 1:2a
6:7 ꢄ 1:8b
7:8 ꢄ 0:7b
4:8 ꢄ 1:5a
5:0 ꢄ 2:4a
14:4 ꢄ 1:1b
16:2 ꢄ 0:9b
Supersweet corn
SSCP
AData in the same column with different letters are significantly different at
p < 0:05.
Statistical analysis. The data are expressed as the mean ꢄ standard
deviation (SD). We used SPSS ver. 13.0J software (SPSS Japan,
Tokyo, Japan) for the statistical analysis. Significance was tested by
using ANOVA, and we employed Bonferroni’s multiple comparison to
compare individual values.
Table 2. DPPH Radical-Scavenging Activity of Each Fraction from
the SSCP Extract after HPLC
DPPH activityA
(mmol Trolox eq./g DMB)
Proportion
(%)C
Fraction
Results
1
2
0:52 ꢄ 0:06a
0:64 ꢄ 0:02ab
0:55 ꢄ 0:01a
3:94 ꢄ 0:05e
0:72 ꢄ 0:09bc
1:11 ꢄ 0:04d
0:73 ꢄ 0:04bc
0:57 ꢄ 0:03a
1:01 ꢄ 0:03d
1:14 ꢄ 0:03d
0:84 ꢄ 0:07c
11.77
4.4
5.4
DPPH radical-scavenging activity and phenolic com-
pound content
3
4.7
The DPPH radical-scavenging activity of the 80%
ethanol extract of SSCP was compared with corn and
found to be higher than that of corn flour and popcorn.
There was no significant difference between this extract
and commercially available supersweet corn (Table 1).
The phenolic compound content of the 80% ethanol
extract of SSCP was also compared with corn and found
to be higher than that of corn flour and popcorn. There
was no significant difference between this extract and
supersweet corn (Table 1).
4
33.5
6.1
5
6
9.5
6.2
7
8
4.8
8.6
9
10
11
Total
9.6
7.1
100.0
AData in the same column with different letters are significantly different at
p < 0:05;
Bper
SSCP
dry
weight
of
pre-fractionation;
Cactivity ꢂ 100=total activity
The data on individual samples of corn (n ¼ 16)
showed a high correlation between the DPPH radical-
8
7
1
2
3
4
5
6
8
9
10
11
4
0
GOA HOA
8
16
24
32
40
48
56
64
72
80
88
Retention time (min)
Fig. 3. HPLC Trace of the SSCP Extract.
The 80% ethanol extract of SSCP was eluted by reversed-phase HPLC according to the conditions described in the Materials and Methods
section. The eluate was collected at 8-min intervals and divided into 11 fractions. The peaks with retention times of 28.5–29.5 min and 34.0 min
are GOA and HOA, respectively.

1798
N. MIDOH et al.
COOH
HO
O
O
HO
O
OH
N
H
HO
HO
OH
L
Indole-3-acetic acid
Ferulic acid
-Ascorbic acid
Fig. 4. Structures of the Antioxidants Assayed in This Study.
Table 3. Contents of GOA and HOA in Corn
Content (mmol/g DM)
Table 5. Contribution Rates of GOA and HOA to the DPPH Radical-
Scavenging Activities of Supersweet Corn and SSCP
Sample
Contribution rate (%)
Substance
Supersweet corn
GOA
HOAA
SSCP
Corn flour (dent corn)
Popcorn
ND
ND
0:12 ꢄ 0:02a
0:08 ꢄ 0:02a
0:21 ꢄ 0:14a
0:43 ꢄ 0:04b
GOA
HOA
Total
39.1
5.9
35.1
10.5
45.6
Supersweet corn
SSCP
3:03 ꢄ 1:20
45.0
3:20 ꢄ 0:30
AData in the same column with different letters are significantly different at
p < 0:05.
Table 6. Effect of Orally Administered HOA on the Susceptibility to
Plasma Oxidation ex Vivo
Table 4. DPPH Radical-Scavenging Activities of GOA and HOA
Dose (mg/kg)
TBARSA (mM MDA eq.)
DPPH activity
Substance
Vehicle
HOA
—
500
275 ꢄ 29a
198 ꢄ 19b
176 ꢄ 25b
(mol Trolox eq./mol)A
1,500
GOA
HOA
0:86 ꢄ 0:04b
1:91 ꢄ 0:12d
0:35 ꢄ 0:03a
1:11 ꢄ 0:04c
1:20 ꢄ 0:01c
AData in the same column with different letters are significantly different at
Indole-3-acetic acid
Ferulic acid
L-Ascorbic acid
p < 0:05.
The contribution rates of GOA and HOA to the DPPH
radical-scavenging activity of the 80% ethanol extract of
SSCP and supersweet corn were calculated from their
contents (Table 3) and activities (Table 4). Table 5
shows that the respective contribution rates of GOA to
the DPPH radical-scavenging activity of the 80%
ethanol extract of SSCP and supersweet corn were
35.1 and 39.1%, both figures being higher than those of
HOA.
AData in the same column with different letters are significantly different at
p < 0:05.
its presence was confirmed by purification. The NMR
spectrum of the purified substance was complicated, so
it was cleaved by ꢀ-glucosidase to a sugar moiety and an
aglycone. We confirmed that the sugar moiety of the
purified substance was glucose by comparing with the
1H-NMR and HSQC data for the authentic substance,
and that the aglycone was 7-hydroxy-oxindole-3-acetic
acid (HOA) by 1H-NMR, HMBC, NOESY, H-H COSY,
13C-NMR, TOCSY, and HSQC data. The purified
substance was finally confirmed to be GOA (Fig. 1). It
was also confirmed that its aglycone, HOA, was
contained in fraction 5 by using HPLC-MS/MS (Fig. 3).
Chemical synthesis of HOA
Although it was extremely difficult to obtain the
required quantity of GOA for animal experiments by
both purification and chemical synthesis, it was possible
to chemically synthesize HOA as shown in Fig. 2 to
investigate its effect by animal experiments. The
structure of chemically synthesized HOA was confirmed
1
Quantification of GOA and HOA
from its H-NMR data, and its purity was estimated to
be 95.9% by using HPLC.
The GOA and HOA contents of corn (Table 3) and
other agricultural products were measured by using
purified GOA and HOA as standard samples. GOA was
detected in SSCP and supersweet corn at similar levels
among the corn types, while HOA was detected in all
corn types investigated and was highest in SSCP. On the
other hand, the GOA and HOA levels in such other
agricultural products as tomato, spinach, pumpkin,
onion, potato, barley, wheat, rice, and soybean were
below the limits for detection.
Ex vivo plasma oxidation
We investigated whether chemically synthesized
HOA absorbed via an oral route had any effects. A
dose of 500 or 1,500 mg/kg of HOA orally administered
to mice resulted in the plasma TBARS level being
significantly lower than the control value in vehicle-
treated mice (Table 6).
Hydrolysis of GOA with the brush border membrane
fraction
DPPH radical-scavenging activities of GOA and HOA
We compared the DPPH radical-scavenging activities
of GOA and HOA with those of other antioxidants
(Fig. 4, Table 4). HOA showed the highest DPPH
radical-scavenging activity, followed by L-ascorbic acid,
ferulic acid, GOA, and indole-3-acetic acid.
We investigated the possibility that GOA might be
hydrolyzed into HOA in the animal intestine. GOA
incubated with the rat brush border membrane fraction
in vitro resulted in GOA being converted to HOA
(Fig. 5).

Antioxidants of Supersweet Corn Powder
1799
crushing, heating with a steamer at approximately 80 ꢁC
for 30 min or more, and drying in a drum drier,11) the
antioxidative activity may be increased during the
manufacturing process. Measurement of the antioxida-
tive activities and quantification of GOA, HOA and the
phenolic compounds should be performed in the future
before and after SSCP manufacture.
We measured the DPPH radical-scavenging activities
of HPLC fractions of the 80% ethanol extract of SSCP,
and identified GOA in a fraction showing the highest
activity. GOA was present as two peaks on the HPLC
trace; however, even when one was isolated and
employed for HPLC, two similar peaks were detected
(data not shown), suggesting the presence of stereo-
isomers in GOA, showing equilibrium in a solution from
the formation of a double bond between the 2nd and 3rd
carbon by keto-enol tautomerization. In respect of the
structure of GOA, two diastereomers for the asymmetric
carbon at the 3rd position exist due to the presence of an
asymmetric carbon at the 1st position of ꢀ-D-glucose
connected to the 7th position (Fig. 1).
HOA
120
80
120
80
Before
After
GOA
40
0
40
0
16
18
20
22
24
26
16
18
20
22
26
24
Retention time (min)
Retention time (min)
Fig. 5. Conversion of GOA to HOA by Rat Brush Border Mem-
branes.
A 100-ml sample of the 0.2 mM GOA solution was incubated with
1 ml of the rat brush border membrane fraction at 37 ꢁC for 24 h.
Ten-microliter aliquots before and after incubation were eluted by
reversed-phase HPLC according to the conditions described in the
Materials and Methods section.
Discussion
To our knowledge, neither GOA nor HOA has been
found in any vegetable/grain other than corn, and they
may be specific to corn. Furthermore, GOA was only
found in SSCP and supersweet corn, and may be specific
to supersweet corn cultivars (Table 3) which have
abnormal accumulations of glucose and sucrose com-
pared with other corn types;10) it is thus possible that
supersweet corn cultivars have greater ability to add
glucose to HOA. However, a survey should be con-
ducted to clarify species/cultivar-related differences in
the levels of these substances, considering the tissue
distribution of these substances and the degree of kernel
maturation, in addition to quantifying these substances
in the part that is eaten.
It has been reported that GOA was synthesized from
indole-3-acetic acid, which is known to be a plant
hormone and also an antioxidant,26) through HOA in the
kernels of supersweet corn after seed germination. GOA
and HOA have been suggested to be inactive metabo-
lites of indole-3-acetic acid;25) however, the antioxida-
tive effects of these substances were demonstrated here
and may protect seeds against oxidative stress, rather
than simply being inactive metabolites of indole-3-
acetic acid.
Nonhebel et al.24) have reported that the GOA level in
the kernels of supersweet corn after seed germination
was in the range of 10.6–14.3 nmol/g of fresh matter.
The GOA content was calculated as 42.4–57.2 nmol/g
of DM when estimating the water content of kernels
after seed germination as 75%. The GOA level in SSCP
and supersweet corn in our study, which were harvested
as immature kernels, may have been 50–80 times higher
than that in germinated kernels of supersweet corn
(Table 3). Since immature kernels are in an anabolic
state and germinated kernels are in a catabolic state, the
difference in GOA content between the two types may
reflect differences in the physiological state.
The contribution rate of GOA to the DPPH radical-
scavenging activity of the 80% ethanol extract of SSCP
was 35.1% (Table 5). This is consistent with the activity
of fraction 4 by HPLC (33.5%, Table 2). GOA may be
the major component of the 80% ethanol extract of
SSCP that is involved in its DPPH radical-scavenging
We identified in this study GOA and HOA, an
aglycone of GOA, as antioxidative components of
SSCP. The GOA content of SSCP and of commercially
available supersweet corn was high among the corn
types, and neither GOA nor HOA was contained in the
nine other agricultural products measured in this study.
HOA decreased the blood TBARS level when orally
administered to mice, indicating its in vivo effect. The
presence of GOA and HOA in corn and the antioxidative
effect of HOA in vitro have been reported;17,24,25)
however, no study has examined the antioxidative effect
of GOA, a quantitative determination of GOA and HOA
in agricultural products other than corn, or the ex vivo
effect of HOA. This study is the first to report these
aspects.
The DPPH radical-scavenging activity of an SSCP
extract prepared in the presence of 80% ethanol was
higher than that of corn flour or popcorn, and was
similar to that of supersweet corn (Table 1). Further-
more, there was a high correlation between the DPPH
radical-scavenging activity and the phenolic compound
content, suggesting that SSCP had similar or more
potent antioxidative activity to that of other types of
corn, despite the heating process, and that phenolic
compounds may have contributed to this activity.
Processing is generally thought to reduce the antiox-
idative activity of foods. However, Dewanto et al.13)
have indicated that the antioxidative activity of sweet
corn treated at 100 ꢁC for 25 min was increased by 36%,
and that the activity of free ferulic acid was elevated by
approximately 125%. They speculated that these in-
creases in antioxidative activity might have been
associated with the heating-related release of antioxida-
tive phenolic compounds from the tissues. Heating may
therefore not reduce the antioxidative activity of sweet
corn, but rather may increase it. The HOA content of
SSCP was higher than that of supersweet corn (Table 3),
which might have been due to the heating-related release
of HOA from the tissues. We did not investigate changes
in the antioxidative activity, HOA content and other
phenolic compounds before and after SSCP manufacture
in this study. As the SSCP production process consists of

1800
N. MIDOH et al.
activity. Similarly, the contribution rate of GOA to the
DPPH radical-scavenging activity of supersweet corn
was 39.1%, suggesting GOA also to be the major
component of the 80% ethanol extract of supersweet
corn in its DPPH radical-scavenging activity. Although
the DPPH radical-scavenging activities of most HPLC
fractions were lower than the fraction of GOA (Table 2),
the individual components cumulatively contributing to
the activity have not been elucidated. Further compo-
nents must be specified to clarify the components
contributing to the DPPH radical-scavenging activity
of the SSCP and supersweet corn extracts.
related conditions. Furthermore, GOA was converted to
HOA in the presence of a rat brush border membrane-
derived extract (Fig. 5); GOA may therefore be ab-
sorbed in vivo by its conversion to HOA with higher
activity. GOA may also exhibit an antioxidative effect
when orally administered to mice. Further investigation
is required to clarify GOA/HOA absorption in animals
and the in vivo activity of GOA.
We identified in this study GOA and HOA as
antioxidants of SSCP and supersweet corn. We also
showed that HOA exhibited an antioxidative effect when
orally administered. The effects of GOA and HOA on
in vivo oxidation-related conditions should be inves-
tigated in the future.
The antioxidative activity of GOA was identified for
the first time in this study; however, that of HOA has
previously been reported by Niwa et al.17) They isolated
HOA from corn gluten meal, a by-product obtained from
dent corn in the industrial starch manufacturing process,
and indicated the hydroxyl radical-scavenging activity
of HOA. Furthermore, they reported the order of the
hydroxyl radical-scavenging activity to be indole-3-
acetic acid ¼ HOA > oxindole-3-acetic acid (a sub-
stance without a hydroxyl group compared with
HOA). We found in our study that the order of the
DPPH radical-scavenging activity was HOA > GOA >
indole-3-acetic acid (Table 4). DPPH radical-scaveng-
ing activity is mainly based on the ability of an
antioxidant to transfer one electron to reduce the DPPH
radical, whereas hydroxyl radical-scavenging activity is
based on the ability of an antioxidant to donate a
hydrogen atom to quench the hydroxyl radical.27) It was
therefore suggested that the difference in activity
between these two parameters might be due to a
difference in the assay mechanism. The high activity
of GOA and HOA in DPPH radical-scavenging activity
compared with indole-3-acetic acid might be attributable
to the easier transfer of an electron from GOA or HOA
to the DPPH radical than from indole-3-acetic acid due
to the loss in aromatic stability of the pyrrole moiety by
the addition of the carbonyl group at the 2nd position to
the indole ring. Since phenolic compounds are known to
have antioxidative activity by their hydroxyl group,28)
the hydroxyl group at the 7th position of HOA is thought
to be related to its activity; therefore, the lower activity
of GOA in DPPH radical scavenging than that of HOA
might be attributable to masking of the hydroxyl group
at the 7th position by the addition of glucose. On the
other hand, a radical produced by the abstraction of a
hydrogen atom from the NH group of indole-3-acetic
acid is highly resonance stable, since the unpaired
electron may be present not only on the nitrogen but can
be delocalized across the indole moiety.29) Thus indole-
3-acetic acid can readily form a resonance-stable radical
which might account for its higher antioxidative activity
in hydroxyl radical scavenging than that of oxindole-3-
acetic acid. The higher activity of HOA in hydroxyl
radical scavenging than that of oxindole-3-acetic acid
might be attributable to the presence of a hydroxyl group
at the 7th position, as with the DPPH radical-scavenging
activity; however, details about this remain to be
clarified.
Acknowledgments
We thank Prof. Kazuki Kanazawa of Kobe University
for his advice regarding the identification of antioxida-
tive substances, Mr. Naoto Koyama and Ms. Kanna
Kuribayashi of Ajinomoto Co., Ltd., for their advice
regarding the measurement of antioxidative activity, Dr.
Nobuhisa Shimba and Ms. Mai Shinagawa of Ajinomoto
Co., Ltd., for their support in the structural analysis
of GOA and HOA, Dr. Keiji Iwasaki and Dr. Yusuke
Amino of Ajinomoto Co., Ltd., for their advice
regarding the chemical synthesis of HOA, and Ms.
Motoko Ogura of Knorr Trading Co., Ltd., for useful
discussions.
References
1) Hamilton CA, Miller WH, Al-Benna S, Brosnan MJ,
Drummond RD, McBride MW, and Dominiczak AF, Clin.
Sci. (Lond.), 106, 219–234 (2004).
2) Willcox JK, Ash SL, and Catignani GL, Crit. Rev. Food Sci.
Nutr., 44, 275–295 (2004).
3) Liu S, Stampfer MJ, Hu FB, Giovannucci E, Rimm E, Manson
JE, Hennekens CH, and Willett WC, Am. J. Clin. Nutr., 70,
412–419 (1999).
4) Dauchet L, Amouyel P, Hercberg S, and Dallongeville J,
J. Nutr., 136, 2588–2593 (2006).
5) Vainio H and Weiderpass E, Nutr. Cancer, 54, 111–142
(2006).
6) Jacobs DRJr, Andersen LF, and Blomhoff R, Am. J. Clin. Nutr.,
85, 1606–1614 (2007).
7) Adom KK and Liu RH, J. Agric. Food Chem., 50, 6182–6187
(2002).
8) Leath MN, ‘‘Corn: Chemistry and Technology’’ 2nd ed., eds.
White PJ and Johnson LA, American Association of Cereal
Chemists, Inc., Minnesota, pp. 241–288 (2003).
9) Laughnan JR, Genetics, 38, 485–499 (1953).
10) Tracy WF, Plant Breed. Rev., 14, 189–236 (1997).
11) Ogura M and Nakamura K, Japan Kokai Tokkyo Koho,
2000342202 (Dec. 12, 2000).
12) Dewanto V, Wu X, Adom KK, and Liu RH, J. Agric. Food
Chem., 50, 3010–3014 (2002).
13) Dewanto V, Wu X, and Liu RH, J. Agric. Food Chem., 50,
4959–4964 (2002).
14) Kurilich AC and Juvik JA, J. Agric. Food Chem., 47, 1948–
1955 (1999).
15) Sakakibara H, Honda Y, Nakagawa S, Ashida H, and Kanazawa
K, J. Agric. Food Chem., 51, 571–581 (2003).
16) Ibrahim KE and Juvik JA, J. Agric. Food Chem., 57, 4636–4644
(2009).
This study has shown that HOA orally administered to
mice exhibited an antioxidative effect in blood in
addition to its antioxidative activity in vitro (Table 6);
HOA may therefore be effective for oxidative stress-
17) Niwa T, Ishii S, Hiramatsu A, and Osawa T, Biosci. Biotechnol.
Biochem., 67, 1870–1874 (2003).
18) Cogan D, Goldberg DR, Hao M-H, Xiong Z, Aungst RA, and
Davis AL, WIPO Patent, 2005108387 (Apr. 29, 2005).

Antioxidants of Supersweet Corn Powder
1801
19) Beer RJS, Clarke K, Khorana HG, and Robertson A, Journal of
the Chemical Society (Resumed), 1605–1609 (1948).
20) King LJ, Parke DV, and Williams RT, Biochem. J., 98, 266–277
(1966).
25) Nonhebel HM, Kruse LI, and Bandurski RS, J. Biol. Chem.,
260, 12685–12689 (1985).
26) Cano A, Alcaraz O, and Arnao MB, Anal. Bioanal. Chem., 376,
33–37 (2003).
21) Lewer P, J. Chem. Soc., Perkin Trans. I, 1, 753–757 (1987).
22) Ohkawa H, Ohishi N, and Yagi K, Anal. Biochem., 95, 351–358
(1979).
27) Turjanski AG, Rosenstein RE, and Estrin DA, J. Med. Chem.,
41, 3684–3689 (1998).
28) Wright JS, Johnson ER, and DiLabio GA, J. Am. Chem. Soc.,
123, 1173–1183 (2001).
23) Kessler M, Acuto O, Storelli C, Murer H, Muller M, and
¨
Semenza G, Biochim. Biophys. Acta, 506, 136–154 (1978).
24) Nonhebel HM and Bandurski RS, Plant Physiol., 76, 979–983
(1984).
29) Encinas MV, Lissi EA, Majmud C, and Olea AF, Int. J. Chem.
Kinet., 23, 761–766 (1991).