Evaluation and comparison of the antioxidative potency of various carbohydrates using different methods

Article abstract of DOI:10.1021/jf804020u

A detailed analysis of the antioxidative activity of 12 carbohydrates including chondroitin sulfate, fucoidan, agaro-oligosaccharide, 2-deoxy-scyllo-inosose (DOI), Gal β1-4DOI, D-glucuronic acid, chitobiose, D-mannosamine, D-galactosamine, D-glucosamine,

Full text of DOI:10.1021/jf804020u

3102 J. Agric. Food Chem. 2009, 57, 3102–3107
Evaluation and Comparison of the Antioxidative
Potency of Various Carbohydrates Using Different
Methods
KATSUMI AJISAKA,* SAYURI AGAWA, SATOMI NAGUMO, KEISUKE KURATO,
TATSUYA YOKOYAMA, KUMIKO ARAI, AND TATSUO MIYAZAKI
Department of Food Science, Niigata University of Pharmacy and Applied Life Sciences,
265-1, Higashijima, Akiha-ku, Niigata 956-8603, Japan
A detailed analysis of the antioxidative activity of 12 carbohydrates including chondroitin sulfate,
fucoidan, agaro-oligosaccharide, 2-deoxy-scyllo-inosose (DOI), Galꢀ1-4DOI, D-glucuronic acid,
chitobiose, D-mannosamine, D-galactosamine, D-glucosamine, heparin, and colominic acid was
performed using four established methods: 2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging
assay, ferric reducing antioxidant power (FRAP) assay, superoxide dismutase (SOD) activity assay,
and the deoxyribose method. Ascorbic acid and/or catechin were used as positive standards. In the
DPPH radical scavenging activity measurements, fucoidan, DOI, and Galꢀ1-4DOI showed remarkable
levels of activity, although at lower levels than ascorbic acid. The SOD assay revealed that DOI,
Galꢀ1-4DOI, and agaro-oligosaccharide had high antioxidant activity, with DOI and Galꢀ1-4DOI notably
showing almost half of the antioxidative potency of ascorbic acid. Using the deoxyribose method,
chitobiose and heparin showed the highest hydroxyl radical scavenging activity, followed by chondroitin
sulfate, colominic acid, Galꢀ1-4DOI, and D-glucosamine. Given that 11 of the carbohydrates analyzed
share a common structure, agaro-oligosaccharide being the exception, we propose from our current
results that at least one amino, carboxyl, carbonyl, or sulfonyl group is required, but is not in itself
sufficient, for carbohydrates to function as antioxidants.
KEYWORDS: Antioxidative activity; radical scavenging; DPPH; FRAP; SOD; deoxyribose assay; carbo-
hydrate; fucoidan; chondroitin sulfate
INTRODUCTION
Recently, several carbohydrates including fucoidan (3),
agaro-oligosaccharides (4, 5), chondroitin sulfate (6), chitobiose
(7), and D-glucosamine (8-10), have been reported to possess
antioxidative properties. Chitobiose is often used to denote
N-acetylglucosaminyl-ꢀ-(1,4)-N-acetylglucosamine, but we use
this term herein to define D-glucosaminyl-ꢀ-(1,4)-D-glucosamine.
Carbohydrate antioxidants are predicted to have significant
utility as they are generally water soluble and have low toxicity.
Moreover, most of these molecules have been safely consumed
as constituents of food products for many years. In this context,
much attention has recently been paid to the development of
novel carbohydrate additives with potent antioxidative activity.
However, most recent reports lack any data that compare the
antioxidative potency of these compounds with that of estab-
lished antioxidants such as ascorbic acid, catechin, or trolox. It
is therefore unclear whether any of the carbohydrates so far
tested are equivalent or even superior to known antioxidants.
The purpose of our present study was to examine this issue by
comparing the relative antioxidative potencies of various
carbohydrates with well-known antioxidants.
Antioxidative materials have recently attracted much attention
because of their potential as scavengers of the reactive oxygen
species which are known to cause damage to nucleic acids and
membrane lipids (1). These damaging events can trigger a
variety of diseases, most notably cancer when DNA is affected
(2). In addition, the cumulative damage to membrane lipids by
reactive oxygen species can cause accelerated aging. Many
antioxidative materials have thus been identified for use as food
additives or medical supplements as scavengers of reactive
oxygen species that will thus negate their damaging effects.
However, most of the reported antioxidant materials are water
insoluble with the exception of ascorbic acid and several
polyphenolic compounds. Water soluble antioxidants such as
ascorbic acid are excreted in the urine after the trapping of
oxygen radicals from water insoluble antioxidative materials
such as vitamin E. Water soluble antioxidants thus play an
important role in assisting their water insoluble counterparts in
the removal of reactive oxygen species from the body.
Four generally established methods were employed for this
purpose: the measurement of 2-diphenyl-1-picrylhydrazyl
* Corresponding author. Tel: +81-250-25-5168. Fax: +81-250-25-
5021. E-mail: ajisaka@nupals.ac.jp.
10.1021/jf804020u CCC: $40.75  2009 American Chemical Society
Published on Web 03/23/2009

Antioxidative Potency of Various Carbohydrates
J. Agric. Food Chem., Vol. 57, No. 8, 2009 3103
Table 1. 13C NMR Chemical Shifts Measured for the Gal-DOI Conjugatea
C-1
C-2
C-3
C-4
C-5
C-6
Gal
DOI
103.3
71.2
72.5
68.6
75.2
61.0
hydrate form
keto form
92.9
206.4
39.9
43.2
67.0
66.7
86.0
84.4
73.0
73.4
75.6
77.3
a NMR analysis was performed for a D2O solution of Gal-DOI conjugate using
a Bruker AVANCE DPX-250 spectrometer.
Figure 1. Enzymatic synthesis of Galꢀ1-4DOI by transglycosylation using
ꢀ-galactosidase from Aspergillus oryzae.
Measurement of DPPH Radical Scavenging Activity. A previ-
ously described standard procedure was used for the measurement
of DPPH radical scavenging activity (17). Briefly, 1 mL of DPPH
(100 µM) in ethanol and 1 mL of each carbohydrate sample (2 mg/
mL) in 100 mM Tris-HCl buffer (pH 7.4) were mixed to adjust the
final sample concentration to 1 mg/mL. This reaction mixture was
then incubated for 20 min in the dark at room temperature. The
absorbance at 515 nm was measured against a blank control (100
mM Tris-HCl buffer instead of carbohydrate solution). Measurements
were performed in triplicate over a 60 s period for each sample.
The measurements of all the samples were not performed within
one day, but it took several days. The DPPH radical scavenging
activity, namely, the inhibitory ratio, was calculated using the
(DPPH) radical scavenging activity, which reflects the ability
of a compound to scavenge general radical species (11); the
ferric reducing antioxidant power (FRAP) assay, which
indicates the potency of reduction of the Fe3+/tripyridyl-s-
triazine (TPTZ) complex to Fe2+ (12); the superoxide
dismutase (SOD) assay, which measures the capacity to
decrease the levels of superoxide radicals (13); and the
deoxyribose assay, which is used to estimate hydroxyl radical
scavenging activity (14). Another crucial aim of our present
study was to assess the structure/activity relationship of the
carbohydrates under analysis.
following equation: scavenging activity (%) ) (1 - Asample/Ablank
× 100, where Ablank is the absorbance of the blank.
)
Measurement of the Concentration-Dependent Antioxidative
Activity of Carbohydrates. (a) DPPH Radical ScaVenging ActiVity.
The radical scavenging activities of 12 carbohydrates at different
concentrations (0.5-20 mM) were measured in triplicate using the
DPPH procedure as described above. Ascorbic acid and catechin were
used as the positive standard.
MATERIALS AND METHODS
Materials. Chitobiose was purchased from Seikagaku Biobusiness
Co. (Tokyo, Japan). Colominic acid was obtained from Funakoshi Co.
(Tokyo, Japan), and fucoidan derived from kelp was purchased from
Sigma-Aldrich Japan (Tokyo, Japan). Agarobiose and agaro-oligosac-
charide were prepared as described previously (15). Most disaccharides
used in this study were synthesized via a condensation (reverse
hydrolysis) reaction using the corresponding glycosidase or through a
glycosidase assisted regioselective transglycosylation reaction using the
corresponding p-nitrophenyl glycoside as a donor. DOI and galactose
was also combined by a transglycosylation reaction using ꢀ-galactosi-
dase from Aspergillus oryzae or Bacillus circulans. (Figure 1) The
structures of the compounds synthesized for this study were con-
firmed by NMR spectroscopy after HPLC purification. The more
common carbohydrates such as D-glucose or N-acetyl-D-glucosamine
that could be readily sourced commercially were used without further
purification.
Measurement of NMR Spectroscopy. NMR spectra were recorded
at 27 °C with a Bruker AVANCE DPX-250 spectrometer in D2O.
Enzymatic Synthesis of Galꢀ1-4DOI. 2-Deoxy-scyllo-inosose
(DOI) was prepared from D-glucose as the starting material from
cultures of metabolically engineered recombinant Escherichia coli (16).
DOI (168 mg, 1 mmol) and p-nitrophenyl ꢀ-D-galactopyranoside (303
mg, 1 mmol) were dissolved in 0.1 M phosphate buffer (pH 6.0)
containing 20% (v/v) dimethylformamide (DMF). ꢀ-Galactosidase from
Aspergillus oryzae (3 U) was then added to the solution. The solution
was then incubated at 37 °C, and the time course of the reaction was
monitored by HPLC using an Asahipak NH2P-50 column (4.5 mm
i.d. × 25 cm, Shimadzu GLC Ltd., Tokyo, Japan) and RI monitor at
25 °C. Elution was with 80% (v/v) acetonitrile at 1 mL/min, and the
reaction was stopped after 180 min by heating in a boiling water bath
for 3 min. The reaction mixture was then applied to an activated carbon
column (2.5 cm i.d. × 48 cm) and eluted with a gradient of water
(1 L) and 10% (v/v) ethanol (1 L). The eluate was collected in 10 mL
fractions, each of which was examined by HPLC. Fractions containing
Galꢀ1-4DOI were pooled and concentrated, yielding 19.7 mg of this
product.
(b) FRAP Assay. The FRAP assay was performed as described by
Benzie et al. (12). FRAP reagent (40 mM TPTZ and 20 mM ferric
chloride in 0.3 M acetate buffer; pH 3.6) was freshly prepared on each
occasion. Carbohydrates at concentrations of 1-20 mM (in a 30 µL
volume), 900 µL of FRAP reagents, and 70 µL of water were mixed
and incubated at 37 °C for 4 min. Catechin was used as a positive
standard. Complex formation was measured by the decrease in the
absorption at 593 nm at room temperature.
(c) SOD ActiVity. SOD activity was measured using the SOD assay
kit-WST (Dojin Chemicals Co., Kumamoto, Japan) according to the
manufacturer’s instructions. The method involved is based on previous
studies of Ukeda et al. (18, 19). Briefly, the mixed solution of
carbohydrate sample of various concentrations in water (20 µL), WST
working solution (200 µL) containing 2-(4-iodophenyl)-3-(4-nitrophe-
nyl)-5-(2,4-disulfophenyl)-2-H-tetrazolium in 50 mM carbonate buffer
(pH 10.2), and enzyme working solution (20 µL) containing xanthine
oxidase in the same buffer was incubated at 37 °C for 20 min, and the
absorbance at 450 nm was measured. The SOD activity (percentage
inhibition rate) was calculated as follows: inhibition rate (%) ) (1 -
Asample/Ablank) × 100, where Ablank is the absorbance of the blank (dilution
buffer instead of carbohydrate solution).
(d) Deoxyribose Method. The deoxyribose assay for hydroxyl radical
scavenging activity was performed using catechin as a positive standard.
To a mixture of 2-deoxyribose (10 mM, 0.36 mL), Fe3+ chloride (10
mM, 0.01 mL), EDTA (1 mM, 0.1 mL), and H2O2 (10 mM, 0.1 mL)
were added 1 mL of carbohydrate sample (1-20 mM) or 1 mL of
mannitol (1.0 µM), 0.33 mL of 50 mM potassium phosphate buffer
(pH 7.4), and 1 mL of ascorbic acid (1 mM). The mixture was then
incubated for 1 h at 37 °C. A solution of thiobarbituric acid (TBA) in
50 mM NaOH (1 mL, 1% w/v) and trichloroacetic acid (TCA) in
aqueous solution (1 mL, 2.8% w/v) was next added, and the mixture
was heated for 20 min in a boiling water bath. The amount of
chromogen produced was measured at 532 nm. The hydroxyl radical
scavenging activity (percentage inhibition rate) was calculated as
follows: scavenging activity (%) ) (1 - Asample/Ablank) × 100, where
Ablank is the absorbance of the blank (10 mM potassium phosphate buffer
instead of carbohydrate solution).
13C NMR data for the transglycosylation product are listed in Table
1. As DOI exists in equilibrium between its keto and hydrate forms in
aqueous solution, each carbon in a DOI residue exhibits two signals
with different chemical shifts. Both of the C-4 carbon signals (74.7
ppm and 74.4 ppm) for DOI shifted to a lower field (84.4 ppm and
86.0 ppm) following conjugation with D-galactose. Hence, the structure
of the product was concluded to be Galꢀ1-4DOI.
In all measurements obtained using methods (a)-(d) above, the
experiments were performed in triplicate, and the averages of the
obtained values were plotted against the carbohydrate concentration.

3104 J. Agric. Food Chem., Vol. 57, No. 8, 2009
Ajisaka et al.
Table 2. DPPH Radical Scavenging Activity of Various Carbohydratesa
6Man (21), FucR1-6GlcNAc (22), Galꢀ1-6Galꢀ-OMe, maltitol,
isomaltitol, di-N-acetylchitobiose, chitotriose, NeuAcR2-3Galꢀ1-
4GlcNAc, NeuAcR2-6Galꢀ1-4GlcNAc, Galꢀ1-4GlcNAcꢀ1-
2Man (23), agaro-oligosaccharide (a mixture of disaccharide-
hexasaccharide) (15), Galꢀ1-6GlcR1-2Frc.
functional
group
activity
(%)
carbohydrate
L-ascorbic acid
catechin
positive control
monosaccharide
96.4 ( 0.2
85.0 ( 0.2
In these measurements, the final carbohydrate concentration
in each case was adjusted to 1 mg/mL. Most carbohydrates in
Table 2 showed 2-7% scavenging activity. D-Glucosamine,
which showed 8.6 ( 0.6% scavenging activity in our measure-
ment, was reported to be active (8). For the next stage of our
evaluation, a DPPH radical scavenging activity of 10% was
tentatively used as a benchmark for selection. According to this
criterion, chondroitin sulfate, fucoidan, agarobiose, 2-deoxy-
scyllo-inosose (DOI), and Galꢀ1-4DOI were judged to have high
antioxidative potency, since they showed nearly 30% scavenging
activity. In contrast, D-galactosamine, D-mannosamine, D-
glucosamine, chitobiose, and D-glucuronic acid were deemed
to have only mild antioxidative potency as they showed about
10% activity. Chondroitin sulfate purchased from Extrasynthese
Co. (Genay, France) showed high antioxidative activity (chon-
droitin sulfateb in Table 2), but the origin was not disclosed
from the manufacturer. In contrast, chondroitin sulfate from
shark cartilage (Sigma-Aldrich Japan, Tokyo, Japan, chondroitin
sulfatec in Table 2) showed much lower antioxidative activity.
The results observed here are quite surprising. We purified the
chondroitin sulfate from Extrasynthese Co., by HPLC connected
with anion exchange column (Mono Q) followed by gel
chromatography column (Superdex Peptide). The 1H NMR
spectrum of the purified chondroitin sulfate from Extrasynthese
Co. was quite analogous to that of chondroitin sulfate from shark
cartilage (data not shown). We are now analyzing the structure
and origin of the chondroitin sulfate from Extrasynthese Co.
Therefore the purified chondroitin sulfate from Extrasynthese
Co. was used hereafter, though the origin was not clear.
Although agarobiose showed almost equal activity with fucoidan
or chondroitin sulfate, it was not easy to prepare large quantities
of pure agarobiose, therefore agaro-oligosaccharide was used
in our subsequent experiments.
L-arabinose
D-fructose
D-galactose
D-mannose
D-glucosamine
D-galactosamine
D-mannosamine
N-acetyl-D-glucosamine
methyl-R-D-glucopyranoside
methyl-2-acetamide-2-deoxy-
ꢀ-D-glucopyranoside
1.6 ( 0.4
5.7 ( 0.1
6.5 ( 0.9
5.1 ( 0.2
8.6 ( 0.6
9.4 ( 1.1
10.8 ( 0.3
6.3 ( 0.5
5.9 ( 0.3
6.2 ( 0.6
-NH2
-NH2
-NH2
-NHAc
-NHAc
D-glucuronic acid
-COOH
-COOH
-SO3H
-SO3H
-COOH
11.0 ( 0.3
5.2 ( 0.1
5.9 ( 0.7
5.7 ( 0.5
5.3 ( 0.2
6.2 ( 0.4
2.0 ( 0.2
D-galacturonic acid
D-glucose-3-sulfate
D-glucose-6-sulfate
potassium D-gluconate
D-glucono-1,5-lactone
N-acetylneuraminic acid
-COOH
-NHAc
disaccharide
Galꢀ1-3GlcNAc
GalR1-6Glc
Glcꢀ1-6Glc
Galꢀ1-6Gal
FucR1-3GlcNAc
ManR1-6Man
ManR1-2Man
Galꢀ1-3Galꢀ-OMe
Galꢀ1-4Galꢀ-OMe
agarobiose
5.4 ( 0.3
4.4 ( 0.4
3.3 ( 0.2
3.4 ( 0.2
6.4 ( 0.4
2.1 ( 0.2
3.4 ( 0.5
2.6 ( 0.3
2.2 ( 0.1
29.7 ( 0.8
9.6 ( 0.8
-NHAc
chitobiose
-NH2
oligosaccharide
polysaccharide
Galꢀ1-6GlcR1-2Frc
tri-N-acetyl chitotriose
agaro-oligosaccharide
2.6 ( 0.2
3.7 ( 0.1
8.7 ( 0.3
-NHAc
chondroitin sulfateb
chondroitin sulfatec
fucoidan
heparin sodium salt
colominic acid sodium salt
-SO3H
-SO3H
-SO3H
-SO3H
-COOH
35.6 ( 1.6
4.3 ( 0.7
30.1 ( 0.9
3.7 ( 1.1
2.1 ( 0.7
miscellaneous
2-deoxy-scyllo-inosose (DOI)
Galꢀ1-4DOI
-CdO
-CdO
24.8 ( 1.1
28.4 ( 3.2
Heparin and colominic acid were included in our panel as
with the exception of agarobiose, all of the sugars showing
antioxidative activity contain at least one amino, carboxyl,
carbonyl, or sulfonyl group. From this point of view, we
speculated that, based upon their structure, heparin and colo-
minic acid would likely show potent antioxidative activity.
Although D-glucose-3-sulfate and D-glucose-6-sulfate may fit
the above criteria, these monosaccharides were not included,
since oligo- or polysaccharide seems to have higher antioxidative
activity than monosaccharide due to the cluster effect. Hence,
a total of 12 carbohydrates were selected for further detailed
analysis of their antioxidative properties using a DPPH radical
scavenging activity assay, FRAP assay, superoxide radical
scavenging activity assay, and hydroxyl radical scavenging
activity assay (deoxyribose method). For each of these assays,
ascorbic acid and/or catechin was used as the positive
standard.
DPPH Radical Scavenging Activity. The DPPH radical
scavenging activity of each compound in our carbohydrate panel
at various concentrations was measured. The most striking
antioxidant activities measured using this method are shown in
Figure 2. In Table 2, the concentration of all samples was
normalized to 1 mg/mL. However, for the comparison of the
structure-activity relationship including polysaccharides, we
assessed the antioxidative activity levels based on the concentra-
tion of mono- or disaccharide repeats. Therefore, the concentra-
a Activity was measured in triplicate at a final concentration of 1 mg/mL.
Experiments were performed in triplicate, and values are means and SE.
b Chondroitin sulfate purchased from Extrasynthese Co. (Genay, France), but the
origin was not disclosed from the manufacturer. c Chondroitin sulfate from shark
cartilage purchased from Sigma-Aldrich Japan (Tokyo, Japan).
RESULTS AND DISCUSSION
Evaluation of the Antioxidative Activity of Selected
Carbohydrates Including Mono-, Di-, Oligo-, and Polysac-
charides. To our knowledge, this is the first report that compares
the antioxidative activities of a wide range of carbohydrates.
As a first screening, DPPH radical scavenging activity was
measured for 67 different carbohydrate molecules including
mono-, di-, oligo-, and polysaccharides of various structures,
most of which were synthesized in our laboratory. In Table 2,
only the representative results are presented. In addition to the
carbohydrates listed in Table 2, the following carbohydrates
were also examined: N-acetyl-D-galactosamine, N-acetyl-D-
mannosamine, methyl-ꢀ-D-glucopyranoside, methyl-R- and -ꢀ-
D-galactopyranoside, methyl-R- and -ꢀ-D-mannopyranoside,
methyl-R- and -ꢀ-D-xylopyranoside, methyl-R- and -ꢀ-D-rham-
nopyranoside, GalR- and Galꢀ-1,3Glc, Galꢀ1-6Glc, GalR1-3Gal
(20), GalR1-6Gal (20), Galꢀ1-4GlcNAc, Galꢀ1-6GlcNAc,
Galꢀ1-2Man (21), Galꢀ1-3Man (21), Galꢀ1-4Man (21), Galꢀ1-

Antioxidative Potency of Various Carbohydrates
J. Agric. Food Chem., Vol. 57, No. 8, 2009 3105
Figure 2. DPPH radical scavenging activity of various carbohydrates. Experiments were performed in triplicate, and values shown are the means ( SE.
tion (mM) of each polysaccharide solution was roughly calcu-
lated using the molecular weights of the repeat mono- or
disaccharide units. Similar to the findings of our preliminary
analysis, chondroitin sulfate and Galꢀ1-4DOI showed high
activity. The activity of DOI was almost equivalent to that of
fucoidan, and Galꢀ1-4DOI showed higher activity than DOI,
possibly due to the lower stability of DOI during the assay. A
1 mg/mL concentration of fucoidan and chondroitin sulfate
corresponds to approximately 4.1 mM and 2.2 mM, respectively.
As shown in Figure 2, the DPPH radical scavenging activity
of fucoidan and chondroitin sulfate of the above concentration
was measured as 24% and 28%, respectively. These values are
just a little lower than the values in Table 2 (30.1% and 35.6%,
respectively). These differences can be estimated within ex-
perimental error.
Figure 3. SOD activities of various carbohydrates. Experiments were
performed in triplicate, and values shown are the means ( SE.
The DPPH radical scavenging activity of each of the
carbohydrates we tested was lower than that of either ascorbic
acid or catechin. Agaro-oligosaccharide, chondroitin sulfate, and
chitobiose showed only mild activity in the DPPH radical
scavenging assay. The activities of the remaining carbohydrates
in this assay are not shown as they were very marginal in each
case.
at 0.2 mM. The SOD scavenging potencies of DOI and Galꢀ1-
4DOI were supposed as about half that of ascorbic acid.
As can be seen from the results shown in Figure 3, agaro-
oligosaccharide also showed remarkably high activity in the
SOD assay, although this was lower than either DOI or Galꢀ1-
4DOI.
FRAP Assay. DOI and Galꢀ1-4DOI showed relatively high
reducing power compared with other carbohydrates. However,
none of the other carbohydrates showed appreciable antioxida-
tive activity by FRAP assay when compared with catechin (data
not shown). Ruperez et al. reported that a sulfated polysaccharide
fraction extracted from the edible marine brown seaweed Fucus
Vesiculosus showed high potential to be antioxidant by the FRAP
assay (24). The fraction they used for the FRAP assay consisted
of fucoidan with an average molecular weight of 1.6 × 106 Da.
Our result on fucoidan was not so high as catechin. We used a
commercial fucoidan of the same species (Fucus Vesiculosus)
without further purification; therefore the molecular weight and
composition were unknown. The difference between both
fucoidan samples is not clear, but the number or position of
sulfate group may play an important role in antioxidant activity
measured by the FRAP method.
SOD Assay. Many of the carbohydrates examined showed
high antioxidative activity when analyzed by the SOD assay,
but the activities of agaro-oligosaccharide, DOI, and Galꢀ1-
4DOI were the most notable in this regard. Only the results for
these three carbohydrates are therefore presented in Figure 3.
As shown from these measurements, the SOD activities of DOI
and Galꢀ1-4DOI were found to be high at 20% and 22%,
respectively, of the superoxide scavenging activity at 0.1 mM.
Ascorbic acid shows 39% activity at the same concentration.
Moreover, DOI and Galꢀ1-4DOI showed about 90% scavenging
activity at 0.5 mM, while ascorbic acid showed the same activity
In the previous report of Xing et al. (8), D-glucosamine
exhibited 80% scavenging activity at a concentration of 0.8 mg/
mL, which corresponds to 4.5 mM under our experimental
conditions. However, D-glucosamine at a 5 mM concentration
showed only about 10% superoxide scavenging activity. Chen
et al. (7) have also reported previously that D-glucosamine does
not show scavenging activity toward superoxide radicals, though
chitobiose displayed such activity. These discrepancies are
difficult to explain at present as Xing et al. did not include a
positive control in their analysis.
Deoxyribose Method. The hydroxyl radical scavenging
activity results for our carbohydrate panel using the deoxyribose
method are shown in Figure 4. Chitobiose and Galꢀ1-4DOI at
1 mM concentration showed high hydroxyl radical scavenging
activity whereas chondroitin sulfate, D-glucosamine, and colo-
minic acid showed medium levels of activity. Although heparin
and colominic acid did not show high DPPH scavenging activity
in our preliminary analysis (Table 1), these two polysaccharides
showed this potency when assayed using the deoxyribose
method. Both methods measure hydroxyl radical scavenging
activity. A plausible explanation for this finding is that these
carbohydrates efficiently target different radicals, i.e. the DPPH
method measures the ability to scavenge a phenolic radical
whereas the deoxyribose method measures the scavenging
potency toward aliphatic hydroxyl radicals. The mechanisms
by which phenolic and hydroxyl radicals are neutralized may
also be different. Our contention that polysaccharides containing

3106 J. Agric. Food Chem., Vol. 57, No. 8, 2009
Ajisaka et al.
Figure 4. Scavenging activity of various carbohydrates toward hydroxy radicals measured by the deoxyribose method. Experiments were performed in
triplicate, and values shown are the means ( SE.
carboxyl or sulfonyl group may have superoxide and hydroxyl
radical scavenging activities was thus supported by these data.
Although the experimental system used herein is different
from that employed in the previous study of Chen et al. (7),
our findings for chitobiose are in accordance with the data from
this previous study in that this compound showed higher
hydroxyl radical scavenging activity than D-glucosamine. In the
report of Xing et al. (8), the hydroxyl radical scavenging activity
of a 3.2 mg/mL D -glucosamine solution, which corresponds to
an 18 mM concentration, was 55%. The hydroxyl radical
scavenging activity of chitobiose at an 18 mM concentration
was approximately 43% in our analysis (Figure 4). The
difference in this case may fall within the experimental error.
The antioxidative activity of DOI could not be measured using
the deoxyribose method because the color of the solution became
dark brown soon after the start of the assay. The procedure for
the deoxyribose method includes the heating of the sample
solution at 95 °C for 20 min, and DOI is likely to be heat labile.
In contrast, Galꢀ1-4DOI showed higher activity in this assay,
possibly due to its greater stability at high temperature, but these
levels were still below those of catechin.
Comparison of the Antioxidant Activities of Selected
Carbohydrates. The results shown in Table 2 confirmed that
the activity of fucoidan was greater than that of the other
carbohydrates based on the measurements from the concentra-
tion dependent DPPH method (Figure 2). However the absolute
antioxidant activity of fucoidan was found to be much lower
than that of ascorbic acid or catechin. The DPPH radical
scavenging activities of DOI and Galꢀ1-4DOI were almost
equivalent to that of fucoidan as demonstrated in Figure 2,
indicating that the absolute antioxidant activities of these
carbohydrates were also far lower than that of ascorbic acid. In
contrast, agaro-oligosaccharide as well as DOI and Galꢀ1-4DOI
showed high SOD activity, which was at almost half the level
of ascorbic acid. Ascorbic acid is known to be heat labile, and
this has hindered its wide use as a food additive. To overcome
this problem, derivatives of this compound have been developed
to increase its stability at higher temperatures in vitro (25). In
contrast to ascorbic acid, Galꢀ1-4DOI and agaro-oligosaccharide
are heat stable. Agaro-oligosaccharide in particular is quite safe
for human consumption as it is derived from agarose, a
component of seaweed. Hence, agaro-oligosaccharide may be
a viable alternative to ascorbic acid as an additive in food
products that require heating during their manufacture. Another
possible candidate is Galꢀ1-4DOI since it is also heat stable
and possesses both high SOD and hydroxyl radical scavenging
activity. High hydroxyl radical scavenging activity was also
observed for chitobiose and heparin sodium salt, both of which
could be potentially used as food additives since these sugars
have prebiotic characteristics and can increase the levels of
intestinal microflora.
In summary, each of the aforementioned carbohydrates has
potential for use as a food additive to serve a different purpose.
Although none of the carbohydrates examined in the present
study exceeded the antioxidative activity of catechin and
ascorbic acid, carbohydrates are of great potential as antioxidant
constituents of food since they are safe for use as an additive
and stable during heating. For example, to control for superoxide
anions, agaro-oligosaccharide or Galꢀ1-4DOI could be used,
whereas to diminish hydroxyl radicals, heparin sodium salt or
chitobiose could be employed.
Relationships between the Structure and Antioxidative
Activities of Carbohydrates. This is the first report to deal
with the relationship between carbohydrate structure and anti-
oxidative activity. It was of considerable interest to evaluate
the functional groups in carbohydrates that showed antioxidative
activity. Normal hexoses and their glycosides do not show any
scavenging activity, though hexoses have an aldehyde group.
We reveal from our present experiments that, with the exception
of agaro-oligosaccharide, all of the tested carbohydrates that
showed antioxidative activity possessed either an amino, car-
boxyl, carbonyl, or sulfonyl group.
As shown in Table 2, chondroitin sulfates of different origins
(chondroitin sulfateb and chondroitin sulfatec) showed large
difference in antioxidative activity. To make the matter worse,
the origin of the sample from Extrasynthese Co. was not clear.
1
The H NMR spectra of both samples were very close to each
other, meaning that a small difference in structure must have
caused the difference of the antioxidative activity.
The fact that D-glucosamine, D-galactosamine, and D-man-
nosamine all showed activity suggests that the orientation of
the hydroxyl groups is not an important determinant of these
properties, but the amino group adjacent to the aldehyde may
be an important common partial structure in the case of these
three sugars. In contrast, D-glucuronic acid and D-galacturonic
acid showed quite different antioxidant properties, though the
structural difference between these two sugars is only in terms
of the orientation of the hydroxyl group at the 4-position.
Potassium D-gluconate, N-acetylneuraminic acid, D-glucose-3-
sulfate, and D-glucose-6-sulfate did not show antioxidant
activity, although they possessed carboxyl or sulfonyl groups.
This indicated that possession of these functional groups is
required but not sufficient to confer antioxidative activity, and
that the position or number of those groups may thus be the
determining factor. It is noteworthy also in this regard that agaro-
oligosaccharide has no such functional groups. It may be the
case therefore that the unstable five membered bicyclic structure

Antioxidative Potency of Various Carbohydrates
J. Agric. Food Chem., Vol. 57, No. 8, 2009 3107
(12) Benzie, I. F. F.; Strain, J. J. The ferric reducing ability of plasma
(FRAP) as a measure of “antioxidant power”: The FRAP assay.
Anal. Biochem. 1996, 239, 70–76.
(13) Robak, J.; Gryglewski, R. J. Flavonoids are scavenging or
scavengers of superoxide anions. Biochem. Pharmacol. 1988, 37,
837–841.
(14) Halliwell, B.; Gutteridge, J. M. C.; Aruoma, O. I. The deoxyribose
method: a simple “test-tube” assay for determination of rate
constants for reactions of hydroxyl radicals. Anal. Biochem. 1987,
165, 215–219.
of agaro-oligosaccharide may be an origin of the radical
scavenging activity.
The purpose of our present study was to elucidate the
functional groups necessary on carbohydrates to confer anti-
oxidative properties. We were not able to precisely determine
this in our present analyses, but our data do show that at least
one amino, carboxyl, carbonyl, or sulfonyl group is required,
but is not in itself sufficient, for carbohydrates to function as
antioxidants.
(15) Chen, H. M.; Zheng, L.; Lin, W.; Yan, X. J. Product monitoring
and quantitation of oligosaccharides composition in agar hydroly-
sate by precolumn labeling HPLC. Talanta 2004, 64, 773–777.
(16) Kogure, T.; Wakisaka, N.; Takaku, H.; Takagi, M. Efficient
production of 2-deoxy-scyllo-inosose from D-glucose by metaboli-
cally engineered recombinant Escherichia coli. J. Biotechnol.
2007, 129, 502–509.
ABBREVIATIONS USED
ASA, ascorbic acid; DMF, dimethylformamide; DOI, 2-deoxy-
scyllo-inosose; DPPH, 2-diphenyl-1-picrylhydrazyl; FRAP, fer-
ric reduction antioxidant power; SOD, superoxide dismutase;
TBA, thiobarbituric acid; TCA, trichloroacetic acid; TPTZ,
tripyridyl-s-triazine.
(17) Ordoudi, A. A.; Tsimidou, M. Z.; Vafiadis, A. P.; Bakalbassis,
E. G. Structure-DPPH radical scavenging activity relationships:
parallel study of catechol and guaiacol acid derivatives. J. Agric.
Food Chem. 2006, 54, 5763–5768.
(18) Ukeda, H.; Sarker, A. K.; Kawata, D.; Sawamura, M. Flow-
injection assay of superoxide dismutase based on the reduction
of highly water-soluble tetrazolium. Anal. Sci. 1999, 15, 353–
357.
LITERATURE CITED
(1) Dimitrios, B. Sources of natural phenolic antioxidants. Trends
Food Sci. Technol. 2006, 17, 505–512.
(2) Barnett, R. N.; Bongiorno, A.; Cleveland, C. L.; Joy, A.; Landman,
U.; Schster, G. B. Oxidative damage to DNA: counterion-assisted
addition of water to ionized DNA. J. Am. Chem. Soc. 2006, 128,
10795–10800.
(19) Ukeda, H.; Kawata, D.; Maeda, S.; Sawamura, M. Spectropho-
tometric assay for superoxide dismutase based on the reduction
of highly water-soluble tetrazolium salts by xanthine-xanthine
oxidase. Biosci., Biotechnol., Biochem. 1999, 63, 485–488.
(20) Matsuo, I.; Fujimoto, H.; Isomura, M.; Ajisaka, K. Chemoenzy-
matic synthesis of GalR1-3Gal, GalR1-3Galꢀ1-4GlcNAc, and their
PEG-conjugates. Bioorg. Med. Chem. Lett. 1997, 7, 255–258.
(21) Miyasato, M.; Ajisaka, K. Regioselectivity in ꢀ-galactosidase-
catalyzed transglycosylation for the enzymatic assembly of
D-galactosyl-D-mannose. Biosci., Biotechnol., Biochem. 2004, 68,
2086–2090.
(22) Ajisaka, K.; Fujimoto, H.; Miyasato, M. An R-fucosidase from
Penicillium multicolor as a candidate enzyme for the synthesis
of R-(1-3)-linked fucosyl oligosaccharides by transglycosylation.
Carbohydr. Res. 1998, 309, 125–129.
(23) Matsuo, I.; Isomura, M.; Miyazaki, T.; Sakakibara, T.; Ajisaka,
K. Chemoenzymatic synthesis of the branched oligosaccharides
which correspond to the core structure of N-linked sugar chains.
Carbohydr. Res. 1998, 305, 401–413.
(3) Xue, C.; Chen, L.; Li, Z.; Cai, Y.; Li, H.; Fang, Y. Antioxidative
activities of low molecular fucoidans from kelp Laminaria
japonica. In More Efficient Utilization of Fish and Fisheries
Products, Sakaguchi, M., Ed.; Elsevier Ltd: 2004; pp 139-145.
(4) Chen, H. M.; Yan, X. J. Antioxidant activities of agaro-
oligosaccharides with different degrees of polymerization in cell-
based system. Biochem. Biophys. Acta 2005, 1722, 103–111.
(5) Chen, H.; Zheng, L.; Yan, X. J. The preparation and bioactivity
research of agaro-oligosaccharides. Food Technol. Biotechnol.
2005, 43, 29–36.
(6) Campo, G. M.; Avenoso, A.; Campo, S.; Ascola, A. D’.; Ferlazzo,
A. M.; Calatrni, A. The antioxidant and antifibrogenic effects of
the glycosaminoglycans hyaluronic acid and chondroitin-4-sulfate
in a subchronic rat model of carbon tetrachloride-induced liver
fibrogenesis. Chem. Biol. Interact. 2004, 148, 125–138.
(7) Chen, A.-S.; Taguchi, T.; Sakai, K.; Kikuchi, K.; Wang, M.-W.;
Miwa, I. Antioxidant activities of N,N′-diacetylchitobiose and
chitotriose. Biol. Pharm. Bull. 2003, 26, 1326–1330.
(8) Xing, R.; Liu, S.; Guo, Z.; Yu, H.; Li, C.; Ji, X.; Feng, J.; Li, P.
The antioxidant activity of glucosamine hydrochloride in Vitro.
Bioorg. Med. Chem. 2006, 14, 1706–1709.
(9) Yan, Y.; Wan-Shun, L.; Bao-Qin, H.; Chang-Hong, W.; Chen-Wei,
F.; Bing, L.; Lie-Huan, C. The antioxidative and immunostimulating
properties of D-glucosamine. Int. Immunopharm. 2007, 7, 29–35.
(10) Yan, Y.; Wan-Shun, L.; Bao-Qin, H.; Hai-Zhou, S. Antioxidative
properties of a newly synthesized 2-glucosamine-thiazolidine-
4(R)-carboxylic acid (GlcNH2Cys) in mice. Nutr. Res. (N.Y.)
2006, 26, 369–377.
(24) Ruperez, P.; Ahrazem, O.; Leal, J. A. Potential antioxidant capacity
of sulfated polysaccharides from the edible marine brown seaweed
Fucus versiculosus. J. Agric. Food Chem. 2002, 50, 840–845.
(25) Takebayashi, J.; Tai, A.; Gohda, E.; Yamamoto, I. Characterization
of the radical-scavenging reaction of 2-O-substituted ascorbic acid
derivatives, AA-2G, AA-2P, and AA-2S: a kinetic and stoichio-
metric study. Biol. Pharm. Bull. 2006, 29, 766–771.
Received for review October 16, 2008. Revised manuscript received
February 19, 2009. Accepted February 23, 2009.
(11) Frankel, E. N.; Meyer, A. S. The problems of using one-
dimensional methods to evaluate multifunctional food and biologi-
cal antioxidants. J. Sci. Food Agric. 2000, 80, 1925–1941.
JF804020U