Astragalosides isolated from the root of astragalus radix inhibit the formation of advanced glycation end products

Article abstract of DOI:10.1021/jf9007168

Because advanced glycation end product (AGE) inhibitors such as pyridoxamine significantly inhibit the development of retinopathy and neuropathy in the streptozotocin-induced diabetic rat, treatment with AGE inhibitors is believed to be a potential strate

Full text of DOI:10.1021/jf9007168

7666 J. Agric. Food Chem. 2009, 57, 7666–7672
DOI:10.1021/jf9007168
Astragalosides Isolated from the Root of Astragalus Radix
Inhibit the Formation of Advanced Glycation End Products
†
K
M
EITA
M
OTOMURA,† YUKIO
F
UJIWARA,‡ NAOKO
K
IYOTA,† KEIICHIRO
§
T
SURUSHIMA
,
,†
OTOHIRO
T
AKEYA,‡ TOSHIHIRO
N
OHARA,# RYOJI
NAGAI
,
AND
T
SUYOSHI
I
KEDA
*
†Department of Natural Medicine, Graduate School of Medical and Pharmaceutical Sciences, Kumamoto
University, Oe-honmachi 5-1, Kumamoto 862-0973, Japan, ‡Department of Cell Pathology, Graduate
School of Medical and Pharmaceutical Sciences, Kumamoto University, Honjo 1-1-1, Kumamoto
860-8556, Japan, §Department of Food and Nutrition, Japan Women’s University, Mejirodai 2-8-1,
Bunkyo-ku, Tokyo 112-8681, Japan, and #Faculty of Pharmaceutical Sciences, Sojo University, Ikeda
4-22-1, Kumamoto 860-0082, Japan
Because advanced glycation end product (AGE) inhibitors such as pyridoxamine significantly inhibit
the development of retinopathy and neuropathy in the streptozotocin-induced diabetic rat, treatment
with AGE inhibitors is believed to be a potential strategy for the prevention of lifestyle-related
diseases such as diabetic complications. A crude extract of Astragali Radix (AR; roots of Astragalus
membranaceus) inhibits the formation of Nε-(carboxymethyl)lysine (CML) and pentosidine during
the incubation of bovine serum albumin with ribose. In the present study, compounds were isolated
from AR that prevented CML and pentosidine formation. Astragalosides significantly inhibited the
formation of both CML and pentosidine, and astragaloside V had the strongest inhibitory effect
among all if the isolated compounds. These data suggest that AR and astragalosides may be a
potentially useful strategy for the prevention of clinical diabetic complications by inhibiting AGEs.
KEYWORDS: Advanced glycation end products; N ε-(carboxymethyl)lysine; pentosidine; Astragali
Radix; astragaloside
INTRODUCTION
hypochlorous acid (14), thus suggesting that CML is an impor-
tant biological marker of oxidative stress in vivo. Furthermore,
pentosidine is also an oxidation-dependent AGE structure be-
cause its formation is inhibited by antioxidative conditions (5).
Pyridoxamine, which was originally proposed to be an inhi-
bitor of the oxidative degradation of fructosamine to AGEs (15),
inhibits the formation of AGEs and lipid peroxidation pro-
ducts (16). Pyridoxamine inhibits the development of retinopathy
and neuropathy in streptozotocin (STZ)-induced diabetic
rats (17, 18). Therefore, treatment with AGE inhibitors and
antioxidants may be a potential strategy for the prevention of
clinical diabetic complications. However, the screening of new
AGE inhibitors by instrumental analyses such as high-perfor-
mance liquid chromatography (HPLC) is required for several
preparation steps, and it is difficult to estimate the inhibitory
effects of a great number of candidate compounds. In fact,
although the inhibitory effects of several crude extracts from
natural products on AGE formation have been documented, the
active compound(s) in those crude extracts and AGE structure(s)
that are inhibited have not yet been identified (19).
Long-term incubation of proteins with glucose leads, through
the formation of early products such as Schiff base and Amadri
products, to the formation of advanced glycation end products
(AGEs) (1). Nε-(Carboxymethyl)lysine (CML), a major antigenic
AGE structure, accumulates in several human and animal tissues
during aging (2) and various diseases including diabetic nephro-
pathy (3). Cellular interactions with CML-modified proteins
induce several biological responses, which are involved in the
development of diabetic vascular complications (3). Pentosidine
is an AGE structure that possesses fluorescent and cross-linking
properties (4) and can be a glycoxidation marker for AGEs in
proteins (5). Pentosidine has been identified in skin collagen (1,4),
glomerular basement membranes (4, 6), and plasma proteins of
diabetic patients (7). The pentosidine level in skin is correlated
with the severity of diabetic complications (8). Furthermore, a
dramatic increase in pentosidine levels is observed in the plasma
of patients with end-stage renal failure (7, 9), β2-microglobulin
amyloid deposits (10), and skin collagens (11), irrespective of the
presence or absence of diabetes. Pentosidine has thus been
implicated in tissue damage in not only diabetic patients but also
in hemodialysis patients with end-stage renal failure.
In the present study an assay system was developed to estimate
the formation of CML and pentosidine using antibodies, and
then the inhibitory effect of 50 crude extracts from natural
products and Chinese herbs such as Epimedii Herba, Artemisiae
Capillari Flos, Foeniculi Fructus, and Astragali Radix were
measured. Under the assay conditions employed, Astragali Radix
(AR; roots of Astragalus membranaceus Bunge, Leguminosae
family) extract significantly inhibited CML and pentosidine
CML is generated by the oxidative cleavage of Amadori
products by the hydroxyl radical (12), peroxynitrite (13), and
*Author to whom correspondence should be addressed (telephone
þ81-96-371-4738; fax þ81-96-362-7799; e-mail tikeda@kumamoto-u.
ac.jp).
pubs.acs.org/JAFC
Published on Web 08/14/2009
© 2009 American Chemical Society

Article
J. Agric. Food Chem., Vol. 57, No. 17, 2009 7667
Figure 1. Time dependency of CML and pentosidine formation. BSA and
ribose were incubated in phosphate buffer for 7 days followed by
determination of CML (A) and pentosidine (B) formation by noncompe-
titive ELISA and HPLC analysis as described under Materials and
Methods.
formation. Furthermore, compounds such as astragalosides and
isoastragalosides were isolated from AR, and the inhibitory effect
of those compounds on CML and pentosidine formation was
also measured. As a result, astragaloside V significantly inhibited
CML and pentosidine formation. These results suggest that
astragaloside V could be a new therapeutic compound inhibiting
the development of diabetic complications such as diabetic
nephropathy, retinopathy, and neuropathy by inhibiting AGE
formation.
Figure 2. Inhibitory effect of AR extract on CML and pentosidine forma-
tion. BSA and ribose were incubated with 1 mg/mL natural products (EH,
Epimedii Herba; ACF, Artemisiae Capillari Flos; FF, Foeniculi Fructus; SR,
Scutellariae Radix; PC, Phellodendori Cortex; CR, Coptidis rhizoma; CF,
Catalpae Fructus) extract in phosphate buffer for 7 days followed by
determination of CML (A) and pentosidine (B) by noncompetitive ELISA
as described under Materials and Methods. Data are presented as the
mean ( SD. /, P < 0.01 versus control.
MATERIALS AND METHODS
pentosidine in the fraction eluted with 40% methanol. The crude product
was then purified according to a modification of the HPLC system
reported by Grandhee and Monnier (21). All other chemicals were of
the best grade available from commercial sources.
Chemicals. Astragalus Radix (lot SU262625) inspected and certified
by regulation of the Japanese Pharmacopeia was purchased from Uchida
Wakan-yaku Co. Ltd. (Tokyo, Japan). Fatty acid-free BSA and methyl
hydroquinone were purchased from Wako (Osaka, Japan). Ribose was
purchased from Acros. Horseradish peroxidase (HRP)-conjugated goat
anti-mouse IgG antibody was purchased from Kirkegaard Perry Labora-
tories. Monoclonal anti-pentosidine antibody, 1C12, was prepared by
immunizing diacetylpentosidine-conjugated human serum albumin
HSA (20). The reaction of the antibody with pentosidine-HSA was
effectively inhibited by free pentosidine; the concentration for 50%
Preparation of Crude Extracts. Astragali Radix, Epimedii Herba,
Artemisiae Capillari Flos, Foeniculi Fructus, Scutellariae Radix, Phello-
dendori Cortex, Coptidis rhizoma, Catalpae Fructus, Gambir, Mume
Fructus, Corydalis Tuber, Cuculus canorus, Puerariae Radix, Zingiberis
Rhizoma, Aurantii Fructus, Immaturus, Armeniacae Semen, Sophorae
Radix, Schizonepetae Spica, Cinnamomi Cortex, Magnoliae Cortex, Schi-
sandrae Fructus, Asiasari Radix, Rehmanniae Radix, Paeoniae Radix,
Cimicifugae Rhizoma, Cnidii Rhizoma, Atractylodis Lanceae, Zizyphi
Fructus, Alismatis Rhizoma, Ophiopogonis Tuber, Atractylodis Rhizoma,
Hoelen, Coicis Semen, Polygalae Radix, Glycyrrhizae Radix, Anemarrhe-
nae Rhizoma, Zedoariae Rhizoma, Cassiae Semen, Pharbitidis Semen,
Cyperi Rhizoma, Schisandrae Fructus, Bupleuri Radix, Lithospermi Radix,
Plantaginis Semen, Houttuyniae Herba, Nupharis Rhizoma, Polyporus,
Aurantii Nobilis Pericarpium, Angelicae Radix, and Angelicae Dahuricae
Rhizoma were extracted with MeOH (three times) by refluxing for 2 h, and
the extracts were concentrated in vacuo to afford residues. The residues
were loaded onto a Diaion HP-20 column and eluted by H2O and MeOH.
Those MeOH eluates were used as crude extracts in the AGE inhibi-
tion assay.
inhibition (IC50) was 6 nM. In contrast, L-arginine and L-lysine, starting
materials for pentosidine formation, were not recognized by the antibody.
Furthermore, other compounds, such as creatine, creatinine, N-mono-
methylarginine, and N-propylpyridinium bromide, did not compete with
the antibody for binding to pentosidine-HSA. We have confirmed that
this monoclonal anti-pentosidine antibody does not cross-react with CML.
Microtitration plates (96-well, Nunc Immunoplate II) were purchased
from Nippon Iner Med (Tokyo, Japan). CML standard was prepared as
described previously (13). Briefly, 0.26 M N R-acetyllysine was incubated
overnight with 0.13 M glyoxylic acid in the presence of 0.65 M NaCNBH3
in 0.5 mL of 0.1 M sodium carbonate buffer (pH 10.0) at room
temperature. Pentosidine standard was prepared as described pre-
Extraction and Isolation of Astragalosides and Isoastragalosides
from AR. AR (1.3 kg) was extracted with MeOH (three times) by
refluxing for 2 h, and the extract was concentrated in vacuo to afford
residues (311.5 g). The residues were loaded onto a Diaion HP-20 column
(60 ꢀ 300 mm) and eluted by a stepwise gradient of H2O/MeOH (100:0,
20:80, 40:60, 60:40, 80:20, 100:0). The obtained fractions, 20% MeOH
(5.4 g), 40% MeOH (5.6 g), 60% MeOH (8.3 g), 80% MeOH (6.1 g), and
100% MeOH (0.65 g), were tested for inhibitory effect on CML and
pentosidine formation. The 80% MeOH eluate (0.47% from AR) was the
viously (20). Briefly, 2.19 g of
D-ribose, 2.5 g of N-acetyllysine, and
3.69 g of N-acetylarginine were dissolved in 58 mL of 100 mM sodium
phosphate, the pH was adjusted to 9.0 by the addition of 5 M sodium
hydroxide, and the mixture was stirred for 48 h at 65 °C. The reaction
mixture was diluted 5-fold with water and passed through 800 mL of
Diaion HP-20 (Mitsubishi Chemical, Tokyo) cation exchange resin. The
column was eluted with 2 L of water, 5% methanol, 10% methanol,
and 20% methanol, followed by elution with 2 L of 40% methanol.
Our analytical HPLC system described below detected N,N φ-diacetyl

7668 J. Agric. Food Chem., Vol. 57, No. 17, 2009
Motomura et al.
Figure 4. Inhibitory effect of the saponin fraction and the flavonoid fraction
on CML and pentosidine formation. BSA and ribose were incubated with
the saponin fraction and the flavonoid fraction in phosphate buffer for
7 days followed by determination of CML (A) and pentosidine (B) by
noncompetitive ELISA as described under Materials and Methods. Data
presented as the mean ( SD. /, P < 0.01; //, P < 0.001; ///, P < 0.0001
versus control.
Figure 3. Inhibitory effect of the fractions isolated by Diaion HP-20 column
chromatography on CML and pentosidine formation. BSA and ribose were
incubated with indicated MeOH eluted fractions in phosphate buffer for
7 days followed by determination of CML (A) and pentosidine (B) by
noncompetitive ELISA as described under Materials and Methods. Data
are presented as the mean ( SD. /, P < 0.01; //, P < 0.001; ///, P <
0.0001 versus control.
reversed-phase column (L 30 ꢀ 150 mm, Chromatorex ODS), using
MeOH/H2O (gradient elution, from 50 to 60% MeOH) to give six
fractions (18-1-18-6). Fraction 18-4 (65.1 mg) was subjected to evaluation
by a silica gel column (L 18 ꢀ 250 mm) using C/M/W (gradient elution,
from 8:2:0.2 to 7:3:0.5) to give OG-9 (27.2 mg, 0.00209%). The structures
of the obtained astragalosides OG-1-9 were identified to be astragaloside
IV (22), astragaloside III (23), astragaloside I (22), isoastragalside I (22),
astragaloside II (22), isoastragalside II (22), astragaloside V (23,24), astra-
galoside VI (23, 24), and astragaloside VII (23, 24), respectively, based on
the findings of a combined spectroscopic analysis. The NMR data of these
compounds closely correlated with those reported previously (22-24).
OG-1 (astragaloside IV): white amorphous powder; [R]1D8 þ28.2° (c 0.1,
MeOH); negative FAB-MS m/z 783 [M - H]-; 1H NMR (pyridine-d5) δ
0.21 (1H, d, J = 4.3 Hz, H-19A), 0.56 (1H, d, J=4.3 Hz, H-19B), 0.96, 1.31
ꢀ 2, 1.37, 1.42, 1.59, 2.03 (each 3H, s, tert-CH3), 2.53 (1H, d, J=3.7 Hz, H-
17), 3.52 (1H, dd, J=4.3, 11.6 Hz, H-3), 3.89 (1H, m, H-24), 4.85 (1H, d,
J = 7.9 Hz, xyl H-1), 4.90 (1H, d, J=7.9 Hz, glc H-1), 5.00 (1H, m, H-16).
OG-2 (astragaloside III): brown amorphous powder; [R]1D8 þ13.8°
(c 0.1, MeOH); negative ESI-MS m/z 783 [M - H]-; 1H NMR (pyridine-d5)
δ 0.29 (1H, m, H-19A), 0.58 (1H, m, H-19B), 1.01, 1.31, 1.32, 1.44 ꢀ 2, 1.63,
1.95 (each 3H, s, tert-CH3), 3.56 (1H, m, H-3), 4.92 (1H, d, J=6.7 Hz, xyl H-
1), 5.41 (1H, d, J=7.9 Hz, glc H-1).
OG-3 (astragaloside I): white amorphous powder; [R]1D8 þ9.6° (c 0.1,
MeOH); negative FAB-MS m/z 867 [M - H]-; 1H NMR (pyridine-d5) δ
0.22 (1H, m, H-19A), 0.57 (1H, m, H-19B), 0.94, 1.27, 1.31 ꢀ 2, 1.42, 1.59,
1.79 (each 3H, s, tert-CH3), 1.97 (3H, s, Ac), 2.03 (3H, s, Ac), 3.40 (1H, m,
H-3), 4.82 (1H, d, J=7.9 Hz, xyl H-1), 4.93 (1H, d, J=7.3 Hz, glc H-1).
OG-4 (isoastragaloside I): white amorphous powder; [R]1D8 þ21.1°
(c 0.1, MeOH); negative FAB-MS m/z 867 [M - H]-; 1H NMR (pyridine-
d5) δ 0.19 (1H, m, H-19A), 0.56 (1H, m, H-19B), 0.93, 1.27, 1.31 ꢀ 2, 1.42,
1.59, 1.82 (each 3H, s, tert-CH3), 1.96 (3H, s, Ac), 2.03 (3H, s, Ac), 3.41
(1H, m, H-3), 4.81 (1H, d, J=7.9 Hz, xyl H-1).
most effective of all fractions and was further separated by Sephadex LH-
20 column (L 40 ꢀ 300 mm, 80% MeOH) to give a saponin fraction (5.2 g,
0.40%) and a flavonoid fraction (0.7 g, 0.054%). By comparison of the
saponin and flavonoid fractions, the saponin fraction was found to be
significantly effective (see Figure 4). As a result, the saponin fraction was
further purified with a silica gel column (L 45 ꢀ 410 mm) and eluted by a
stepwise gradient of CHCl3/MeOH/H2O (C/M/W) (9:1:0.1, 8:2:0.2,
7:3:0.5, 6:4:1). The eluates (20 mL in each test tube) were combined into
22 fractions (1-22) on the basis of silica gel TLC. Fraction 5 (751.3 mg)
was chromatographed on a silica gel column (L 26 ꢀ 290 mm) using C/M/
W (gradient elution, from 9:1:0.1 to 8:2:0.2) to give 10 fractions (5-1-
5-10). Fraction 5-6 (116.8 mg) was subjected to evaluation by a reversed-
phase column (L 30 ꢀ 150 mm, Chromatorex ODS), using MeOH/H2O
(gradient elution, from 70 to 75% MeOH) to afford three fractions (5-6-
1-5-6-3). Fraction 5-6-2 (90.0 mg) was subjected to a silica gel column
(L 20 ꢀ 330 mm) using C/M/W (gradient elution, from 10:1:0 to 8:2:0.2) to
furnish OG-3 (21.8 mg, 0.00168%) and OG-4 (16.1 mg, 0.00124%).
Fraction 8 (183.7 mg) was chromatographed on a reversed-phase column
(L 38 ꢀ 90 mm, Chromatorex ODS), using MeOH/H2O (gradient elution,
from 60 to 70% MeOH) to give seven fractions (8-1-8-7). Fraction 8-5
(123.6 mg) was subjected to a silica gel column (L 20 ꢀ 310 mm) using
C/M/W (gradient elution, from 9:1:0.1 to 8:2:0.2) to afford OG-6 (17.3 mg,
0.00 133%), OG-5 (47.4 mg, 0.00365%), and OG-1 (22.4 mg, 0.00172%).
Fraction 11 (200.8 mg) was chromatographed on a reversed-phase column
(L 30 ꢀ 150 mm, Chromatorex ODS), using MeOH/H2O (gradient
elution, from 60 to 70% MeOH) to give six fractions (11-1-11-6).
Fraction 11-4 (101.7 mg) was subjected to a silica gel column (L 18 ꢀ
500 mm) using C/M/W (gradient elution, from 9:1:0.1 to 25:5:0.3) to give
OG-2 (20.9 mg, 0.00161%) and OG-1 (72.9 mg, 0.00561%). Fraction
16 (142.6 mg) was chromatographed on a reversed-phase column (L 20 ꢀ
330 mm, Chromatorex ODS), using MeOH/H2O (gradient elution,
from 60 to 65% MeOH) to give OG-7 (9.6 mg, 0.00074%) and OG-8
(43.9 mg, 0.00338%). Fraction 18 (372.5 mg) was chromatographed on a
OG-5 (astragaloside II): white amorphous powder; [R]1D8 þ30.4° (c 0.1,
MeOH); negative FAB-MS m/z 825 [M - H]-; 1H NMR (pyridine-d5) δ
0.20 (1H, d, J = 3.7 Hz, H-19A), 0.57 (1H, d, J = 3.7 Hz, H-19B), 0.95,
1.28, 1.30, 1.31, 1.41, 1.56, 1.81 (each 3H, s, tert-CH3), 2.05 (3H, s, Ac),

Article
J. Agric. Food Chem., Vol. 57, No. 17, 2009 7669
Determination of the Inhibitory Effect of Plant Extracts and
Natural Compounds on Pentosidine Formation. BSA (2 mg/mL) and
ribose (100 mM) were incubated with plant extracts (1 mg/mL) or natural
compounds (1 mM) in 200 mM sodium phosphate buffer at 60 °C for
7 days, followed by the determination of pentosidine by both noncompe-
titive ELISA and HPLC analyses.
Determination of the Inhibitory Effect of Astragaloside V on
CML Formation from Amadori Products. Glycated HSA (2 mg/mL)
was incubated with astragaloside V (1 mM) in PBS at 37 °C for 5 days,
followed by the determination of CML by ELISA.
Determination of the Inhibitory Effect of Astragaloside V on
CML Formation from Glyoxal Pathway. BSA (2 mg/mL) and glyoxal
(5 mM) were incubated with astragaloside V (1 mM) in PBS at 37 °C for
2 days, followed by the determination of CML by ELISA.
ELISA. ELISA was performed as described previously (25). Briefly,
each well of a 96-well microtiter plate was coated with 100 μL of the sample
to be tested in PBS, blocked with 0.5% gelatin, and washed three times
with PBS containing 0.05% Tween 20 (washing buffer). The wells were
incubated with 0.1 mL of anti-CML antibody, 6D12 (1 μg/mL), or
anti-pentosidine antibody (1 μg/mL) dissolved in washing buffer for 1 h.
The wells were then washed with washing buffer three times and reacted
with HRP-conjugated anti-mouse IgG antibody, followed by reaction
with 1,2-phenylenediamine dihydrochloride. The reaction was terminated
by the addition of 0.1 mL of 1 M sulfuric acid, and the absorbance at 492
nm was read by a micro-ELISA plate reader.
Detection of CML by HPLC. The CML content of the samples were
quantified by acid hydrolysis with 6 N HCl for 24 h at 110 °C, followed by
amino acid analysis on a Hitachi L-8500A instrument equipped with an
ion-exchange HPLC column (2622 SC, 4.6 ꢀ 80 mm; Hitachi) and a
ninhydrin postcolumn detecztion system, as described previously (26).
Detection of Pentosidine by HPLC. The pentosidine content of the
samples were quantified by acid hydrolysis with 6 N HCl for 24 h at 110 °C,
followed by an analysis using HPLC equipped with an ODS column
(Capcell pak C18 UG-80, 4.6 ꢀ 250 mm; Shiseido) and fluorescence
detector (ex/em = 335/385 nm) as described previously (20).
Statistical Analysis. All data are expressed as the mean ( SD.
Differences between groups were examined for statistical significance
using the Mann-Whitney U test and the nonrepeated measures ANOVA.
A p value of <0.05 denoted the presence of a statistically significant
difference.
Figure 5. Structures of astragalosides and isoastragalosides.
3.41 (1H, m, H-3), 4.79 (1H, d, J = 7.9 Hz, xyl H-1), 4.91 (1H, d, J = 7.3
Hz, glc H-1), 5.54 (1H, br, xyl H-2).
RESULTS AND DISCUSSION
OG-6 (isoastragaloside II): white amorphous powder; [R]1D8 þ15.1°
(c 0.1, MeOH); negative ESI-MS m/z 825 [M - H]-; 1H NMR (pyridine-d5)
δ 0.22 (1H, m, H-19A), 0.59 (1H, m, H-19B), 0.96, 1.31, 1.33, 1.43, 1.59, 1.67,
1.97 (each 3H, s, tert-CH3), 2.00 (3H, s, Ac), 3.50 (1H, m, H-3), 4.84 (1H, d,
J= 7.3 Hz, xyl H-1), 5.91 (1H, d, J= 7.3 Hz, glc H-1), 5.71 (1H, m, xyl H-3).
OG-7 (astragaloside V): brown amorphous powder; [R]1D8 þ9.2° (c 0.1,
MeOH); negative FAB-MS m/z 945 [M - H]-; 1H NMR (pyridine-d5) δ
0.27 (1H, d, J = 3.7 Hz, H-19A), 0.56 (1H, d, J = 3.7 Hz, H-19B), 0.94,
1.29, 1.35, 1.42, 1.43, 1.67, 1.94 (each 3H, s, tertiary-CH3), 3.55 (1H, dd,
J = 4.3, 11.6 Hz, H-3), 4.91 (1H, d, J = 6.1 Hz, xyl H-1), 5.06 (1H, m, glc3
H-1), 5.40 (1H, d, J = 7.3 Hz, glc1 H-1).
OG-8 (astragaloside VI): brown amorphous powder; [R]1D8 þ17.2°
(c 0.1, MeOH); negative FAB-MS m/z 945 [M - H]-; 1H NMR (pyridine-
d5) δ 0.18 (1H, d, J = 4.3 Hz, H-19A), 0.66 (1H, d, J = 4.3 Hz, H-19B),
0.98, 1.26, 1.30, 1.42, 1.43, 1.58, 1.91 (each 3H, s, tert-CH3), 3.42 (1H, m,
H-3), 3.77 (1H, m, H-6), 4.86 (1H, d, J = 7.9 Hz, xyl H-1), 4.97 (1H, m,
glc2 H-1), 5.34 (1H, d, J = 7.9 Hz, glc1 H-1).
OG-9 (astragaloside VII): brown amorphous powder; [R]1D8 þ11.2° (c
0.1, MeOH); negative FAB-MS m/z 945 [M - H]-; 1H NMR (pyridine-d5)
δ 0.18 (1H, d, J = 4.3 Hz, H-19A), 0.58 (1H, d, J = 4.3 Hz, H-19B), 0.92,
1.26, 1.33, 1.36, 1.42, 1.65, 2.01 (each 3H, s, tert-CH3), 3.52 (1H, m, H-3).
Determination of the Inhibitory Effect of Plant Extracts and
Natural Compounds on CML Formation. BSA (2 mg/mL) and ribose
(33 mM) were incubated with either plant extracts (1 mg/mL) or natural
compounds (1 mM) in PBS at 37 °C for 7 days, followed by the
determination of CML by a noncompetitive enzyme-linked immunosor-
bent assay (ELISA) and HPLC analysis.
BSA was incubated with ribose in sodium phosphate buffer,
and CML and pentosidine formation was measured. As shown in
Figure 1, the reactivity of anti-CML antibody increased in a time-
dependent manner during incubation of BSA (2 mg/mL) with 33
mM ribose in PBS at 37 °C. A similar tendency was observed
when CML content was measured by HPLC (Figure 1A). Simi-
larly, the reactivity of anti-pentosidine antibody increased in a
time-dependent manner during incubation of BSA (2 mg/mL)
with 100 mM ribose in 200 mM sodium phosphate buffer at
60 °C, and those reactivities were correlated with the finding of
the HPLC analysis (Figure 1B).
We first measured the inhibitory effect of 50 crude extracts
from natural products and Chinese herbs on CML and pentosi-
dine formation. Incubation of BSA with ribose increased CML
and pentosidine formation. Under the assay conditions em-
ployed, AR extract inhibited CML and pentosidine formation
(Figure 2). These data suggest that AR contains compound(s)
that can inhibit AGE formation. Next, the AR extract was
applied to Diaion HP-20 column chromatography and eluted
with20-100%MeOHtogivefivefractions. Asshown inFigure3,
the 80 and 100% MeOH fractions significantly inhibited both
CML and pentosidine formation. Because the extractive content
of the 80% MeOH fraction was 10-fold higher than that of the
100% MeOH fraction, the 80% MeOH fraction was further
separated into a saponin fraction and a flavonoid fraction. As a

7670 J. Agric. Food Chem., Vol. 57, No. 17, 2009
Motomura et al.
Figure 6. Inhibitoryeffectof astragalosidesandisoastragalosideson CMLand pentosidine formation. BSAand ribosewereincubated with astragalosidesand
isoastragalosides in phosphate buffer for 7 days followed by determination of CML and pentosidine by noncompetitive ELISA (A, C, E, and F) and HPLC (B
and D) as described under Materials and Methods. Data are presented as the mean ( SD. /, P < 0.05; //, P < 0.0005; ///, P < 0.0001 versus control.
result, the saponin fraction inhibited CML and pentosidine
formation, whereas the flavonoid fraction slightly inhibited
CML formation (Figure 4). Then, the saponin fraction was
further applied to ODS column chromatography and/or silica
gel column chromatography to yield astragalosides and isoas-
tragalosides (Figure 5). As shown in Figure 6, isolated compounds
other than astragaloside IV and astragaloside VI inhibited CML
formation. Among those compounds, astragaloside I, isoastra-
galoside I, and astragaloside V have inhibitory effects on CML
formation the same as that of pyridoxiamine (Figure 6A).
Furthermore, all compounds inhibited pentosidine formation
(Figure 6C), and astragaloside V and astragaloside VI showed
the highest inhibitory effect on pentosidine formation, whereas
pyridoxiamine did not inhibit pentosidine formation (Figure 6C).
Moreover, astragaloside V inhibited both CML and pentosidine
formation in a dose-dependent manner (Figure 6E,F). These
results suggest that those compounds in the AR extract played
a key role in the inhibition of CML and pentosidine formation
and that astragalosides are more effective AGE inhibitors than
pyridoxamine. We next measured the inhibitory mechanism of
astragaloside V on CML formation. It is known that CML is
generated from the oxidation of Amadori products and glucose
autoxidation-derived glyoxal pathway. Astragaloside V inhibited
CML formation from incubation of BSA with glyoxal
(Figure 7B), whereas it did not inhibit CML formation from
Amadori products (Figure 7A). These data indicate that astraga-
loside V inhibits CML formation by inhibiting the glyoxal
pathway.
AGEs are a heterogeneous group of reaction products that
form between a protein’s primary amino group and a carbohy-
drate-derived aldehyde group. A substantial number of previous
papers have indicated that AGEs exacerbate and accelerate the
changes associated with aging while also contributing to the early
phases of age-related disease, including atherosclerosis, cataracts,
neurodegenerative disease, renal failure, arthritis, and age-related
macular degeneration (2,3). CML and pentosidine are known as
AGE structures associated with pathogenesis and development of
these diseases. For these reasons, it is believed that the use of AGE
inhibitors may therefore be a potentially effective strategy to
prevent the pathogenesis of age-related diseases. In this study,
CML and pentosidine inhibitors were screened from 50 crude
extracts of natural products and Chinese herbs. Among those
extracts, AR extract significantly inhibited CML and pentosidine
formation (Figure 2). On the other hand, extracts such as EH,
ACF, PC, CR, and CF enhanced CML and pentosidine forma-
tion (Figure 2). Akagawa et al. (27) demonstrated that some
polyphenols such as epicatechin gallate and epigallocatechin
gallate generate hydrogen peroxide, and the EH, ACF, PC,
CR, and CF extracts contain plenty of polyphenols and may
have enhanced CML and pentosidine formation by producing

Article
J. Agric. Food Chem., Vol. 57, No. 17, 2009 7671
which has the same aglycon as astragaloside V, inhibited only
pentosidine formation, thus suggesting that a difference in the
binding position of glucose (C-6 and C-25) between astragaloside
V and astragaloside VI might affect the inhibition of CML
formation. In traditional Chinese medicine, AR is often mixed
withother herbs, such as angelicaand ginseng, invarious complex
prescription formulas. Such herbal formulas have been used for
centuries in Asia to treat cancers, diabetes, kidney infections,
strokes, and many other diseases (32, 33). For this reason, the
astragalosides in AR may play an important role in those
beneficial effects by inhibiting AGE formation. Therefore, as-
tragalosides could be candidate agents for use as a potential new
medicine.
ACKNOWLEDGMENT
We are very grateful to Makiko Yoshitomi, Katsumi Mera,
and Mime Nagai for their valuable collaborative endeavors.
LITERATURE CITED
Figure 7. Inhibitory effect of astragaloside V on CML formation from
Amadori products and glyoxal pathway. Glycated HSA (2 mg/mL) was
incubatedwithastragalosideV(1 mM) inPBSat37 °C for5 days (A). BSA
(2 mg/mL) and glyoxal (5 mM) were incubated with astragaloside V
(1 mM) in PBS at 37 °C for 2 days (B), followed by the determination of
CML by ELISA.
(1) Sell, D. R.; Monnier, V. M. End-stage renal disease and diabetes
catalyze the formation of a pentose-derived crosslink from aging
human collagen. J. Clin. Invest. 1990, 85 (2), 380–384.
(2) Araki, N.; Ueno, N.; Chakrabarti, B.; Morino, Y.; Horiuchi, S.
Immunochemical evidence for the presence of advanced glycation
end products in human lens proteins and its positive correlation with
aging. J. Biol. Chem. 1992, 267 (15), 10211–10214.
(3) Vlassara, H.; Bucala, R.; Striker, L. Pathogenic effects of advanced
glycosylation: biochemical, biologic, and clinical implications for
diabetes and aging. Lab Invest. 1994, 70 (2), 138–151.
(4) Sell, D. R.; Monnier, V. M. Structure elucidation of a senescence
cross-link from human extracellular matrix. Implication of pentoses
in the aging process. J. Biol. Chem. 1989, 264 (36), 21597–21602.
(5) Baynes, J. W. Role of oxidative stress in development of complica-
tions in diabetes. Diabetes 1991, 40 (4), 405–412.
(6) Monnier, V. M.; Sell, D. R.; Nagaraj, R. H.; Miyata, S.; Grandhee,
S.; Odetti, P.; Ibrahim, S. A., Maillard reaction-mediated molecular
damage to extracellular matrix and other tissue proteins in diabetes,
aging, and uremia. Diabetes 1992, 41 (Suppl. 2), 36-41.
(7) Odetti, P.; Fogarty, J.; Sell, D. R.; Monnier, V. M. Chromatographic
quantitation of plasma and erythrocyte pentosidine in diabetic and
uremic subjects. Diabetes 1992, 41 (2), 153–159.
(8) Sell, D. R.; Lapolla, A.; Odetti, P.; Fogarty, J.; Monnier, V. M.
Pentosidine formation in skin correlates with severity of complica-
tions in individuals with long-standing IDDM. Diabetes 1992, 41
(10), 1286–1292.
(9) Taneda, S.; Monnier, V. M. ELISA of pentosidine, an advanced
glycation end product, in biological specimens. Clin. Chem. 1994, 40
(9), 1766–1773.
(10) Miyata, T.; Taneda, S.; Kawai, R.; Ueda, Y.; Horiuchi, S.; Hara, M.;
Maeda, K.; Monnier, V. M. Identification of pentosidine as a native
structure for advanced glycation end products in β-2-microglobulin-
containing amyloid fibrils in patients with dialysis-related amyloi-
dosis. Proc. Natl. Acad. Sci. U.S.A. 1996, 93 (6), 2353–2358.
(11) Hricik, D. E.; Schulak, J. A.; Sell, D. R.; Fogarty, J. F.; Monnier, V.
M. Effects of kidney or kidney-pancreas transplantation on plasma
pentosidine. Kidney Int. 1993, 43 (2), 398–403.
(12) Nagai, R.; Ikeda, K.; Higashi, T.; Sano, H.; Jinnouchi, Y.; Araki, T.;
Horiuchi, S. Hydroxyl radical mediates N ε-(carboxymethyl)lysine
formation from Amadori product. Biochem. Biophys. Res. Commun.
1997, 234 (1), 167–172.
hydrogen peroxide. AR is a traditional Chinese medicinal herb
that originated in northern China and has been used for more
than 2000 years (28). AR contains 2040-dihydroxy-5,6-dimethox-
yisoflavone, kumatakenin, choline, betaine, polysaccharides,
saponins, glucuronic acid, sucrose, amino acids, traces of folic
acid, and astraisoflavanin (29). AR is a widely used Chinese
medicinal herb that is well-known for its vital-energy tonifying,
skin reinforcing, diuretic, and tissue generative actions. Further-
more, AR has been extensively used as a tonic to enhance the
body’s defense system (30). Evidence has indicated the impor-
tance of AR polysaccharide fractions in the modulation of
immunefunctions inbothhumansand experimentalanimals (31).
The current study demonstrated that astragalosides, isolated
from AR by successive purification with Diaion HP-20 and
Sephadex LH-20 column chromatography, inhibited AGE for-
mation (Figure 6). Among these astragalosides, astragaloside V
significantly inhibited both CML and pentosidine formation
(Figure 6). However, astragaloside VI significantly inhibited only
pentosidine formation (Figure 6). Both CML and pentosidine are
known as oxidation-dependent AGE structures. However, those
formation pathways are different from each other. CML is
generated from the oxidation of Amadori products and the
glucose autoxidation-derived glyoxal pathway. On the other
hand, pentosidine is generated from oxidation of glycated protein
formed by cyclization reaction of ribose with lysine and arginine.
Therefore, these results suggest that astragalosides V and VI
showed different inhibitory effects on AGE formation because of
differences in inhibitory mechanism and chemical structure
between astragalosides V and VI. Astragalosides are known as
cycloartane-type triterpenoid compounds. In a previous study,
several types of triterpenoid compounds such as oleanane-type,
ursane-type, and lupane-type triterpenoid compounds could
hardly inhibit AGE formation under our assay conditions (data
not shown). However, astragalosides significantly inhibited CML
and pentosidine formation in this study, suggesting that the
cycloartane skeleton may play an important role in the inhibition
of AGE formation. Furthermore, astragaloside V inhibited both
CML and pentosidine formation, whereas astragaloside VI,
(13) Nagai, R.; Unno, Y.; Hayashi, M. C.; Masuda, S.; Hayase, F.;
Kinae, N.; Horiuchi, S. Peroxynitrite induces formation of
N ε-(carboxymethyl) lysine by the cleavage of Amadori product
and generation of glucosone and glyoxal from glucose: novel path-
ways for protein modification by peroxynitrite. Diabetes 2002, 51 (9),
2833–2839.
(14) Mera, K.; Nagai, R.; Haraguchi, N.; Fujiwara, Y.; Araki, T.;
Sakata, N.; Otagiri, M. Hypochlorous acid generates

7672 J. Agric. Food Chem., Vol. 57, No. 17, 2009
Motomura et al.
N ε-(carboxymethyl)lysine from Amadori products. Free Radical
Res. 2007, 41 (6), 713–718.
(15) Booth, A. A.; Khalifah, R. G.; Todd, P.; Hudson, B. G. In vitro
kinetic studies of formation of antigenic advanced glycation end
products (AGEs). Novel inhibition of post-Amadori glycation path-
ways. J. Biol. Chem. 1997, 272 (9), 5430–5437.
(16) Nakamura, Y.; Kawakami, M.; Yoshihiro, A.; Miyoshi, N.; Ohiga-
shi, H.; Kawai, K.; Osawa, T.; Uchida, K. Involvement of the
mitochondrial death pathway in chemopreventive benzyl isothio-
cyanate-induced apoptosis. J. Biol. Chem. 2002, 277 (10), 8492–8499.
(17) Degenhardt, T. P.; Alderson, N. L.; Arrington, D. D.; Beattie, R. J.;
Basgen, J. M.; Steffes, M. W.; Thorpe, S. R.; Baynes, J. W.
Pyridoxamine inhibits early renal disease and dyslipidemia in the
streptozotocin-diabetic rat. Kidney Int. 2002, 61 (3), 939–950.
(18) Stitt, A.; Gardiner, T. A.; Alderson, N. L.; Canning, P.; Frizzell, N.;
Duffy, N.; Boyle, C.; Januszewski, A. S.; Chachich, M.; Baynes, J.
W.; Thorpe, S. R. The AGE inhibitor pyridoxamine inhibits devel-
opment of retinopathy in experimental diabetes. Diabetes 2002, 51
(9), 2826–2832.
(19) Yamabe, N.; Kang, K. S.; Goto, E.; Tanaka, T.; Yokozawa, T.
Beneficial effect of Corni Fructus, a constituent of Hachimi-jio-gan,
on advanced glycation end-product-mediated renal injury in
streptozotocin-treated diabetic rats. Biol. Pharm. Bull. 2007, 30 (3),
520–526.
(20) Miyazaki, K.; Nagai, R.; Horiuchi, S. Creatine plays a direct role as a
protein modifier in the formation of a novel advanced glycation end
product. J. Biochem. 2002, 132 (4), 543–550.
(21) Grandhee, S. K.; Monnier, V. M. Mechanism of formation of the
Maillard protein cross-link pentosidine. Glucose, fructose, and
ascorbate as pentosidine precursors. J. Biol. Chem. 1991, 266 (18),
11649–11653.
(22) Kitagawa, I.; Wang, H.; Takagi, A.; Fukuda, M.; Miura, I.;
Yoshikawa, M. Saponin and sapogenol. XXXV. Chemical consti-
tuents of Astragali Radix, the root of Astragalus membranaceus
Bunge. (2). Astragalosides I, II and IV, acetylastragaloside I and
isoastragalosides I and II. Chem. Pharm. Bull. 1983, 31 (2), 698–708.
(23) Kitagawa, I.; Wang, H.; Saito, M.; Yoshikawa, M. Saponin and
sapogenol. XXXVI. Chemical constituents of Astragali Radix, the
root of Astragalus membranaceus Bunge. (3). Astragalosides III, V,
and VI. Chem. Pharm. Bull. 1983, 31 (2), 709–715.
(25) Nagai, R.; Hayashi, C. M.; Xia, L.; Takeya, M.; Horiuchi, S.
Identification in human atherosclerotic lesions of GA-pyridine, a
novel structure derived from glycolaldehyde-modified proteins.
J. Biol. Chem. 2002, 277 (50), 48905–48912.
(26) Nagai, R.; Araki, T.; Hayashi, C. M.; Hayase, F.; Horiuchi, S.
Identification of N ε-(carboxyethyl)lysine, one of the methylglyoxal-
derived AGE structures, in glucose-modified protein: mechanism for
protein modification by reactive aldehydes. J. Chromatogr., B: Anal.
Technol. Biomed. Life Sci. 2003, 788 (1), 75–84.
(27) Akagawa, M.; Shigemitsu, T.; Suyama, K. Production of hydrogen
peroxide by polyphenols and polyphenol-rich beverages under quasi-
physiological conditions. Biosci., Biotechnol., Biochem. 2003, 67 (12),
2632–2640.
(28) Shao, B. M.; Xu, W.; Dai, H.; Tu, P.; Li, Z.; Gao, X. M. A study on
the immune receptors for polysaccharides from the roots of Astra-
galus membranaceus, a Chinese medicinal herb. Biochem. Biophys.
Res. Commun. 2004, 320 (4), 1103–1111.
(29) Ma, X. Q.; Shi, Q.; Duan, J. A.; Dong, T. T.; Tsim, K. W. Chemical
analysis of Radix Astragali (Huangqi) in China: a comparison with
its adulterants and seasonal variations. J. Agric. Food Chem. 2002,
50, 4861–4866.
(30) Kusum, M.; Klinbuayaem, V.; Bunjob, M.; Sangkitporn, S. Pre-
liminary efficacy and safety of oral suspension SH, combination of
five chinese medicinal herbs, in people living with HIV/AIDS ; the
phase I/II study. J. Med. Assoc. Thai 2004, 87 (9), 1065–1070.
(31) Wang, R. T.; Shan, B. E.; Li, Q. X. [Extracorporeal experimental
study on immuno-modulatory activity of Astragalus memhranaceus
extract]. Zhongguo Zhong Xi Yi Jie He Za Zhi 2002, 22 (6), 453–456.
(32) Zhang, Y. W.; Xie, D.; Chen, Y. X.; Zhang, H. Y.; Xia, Z. X.
Protective effect of Gui Qi mixture on the progression of diabetic
nephropathy in rats. Exp. Clin. Endocrinol. Diabetes 2006, 114 (10),
563–568.
(33) Li, S.; Zhang, Y.; Zhao, J. Preparation and suppressive effect of
astragalus polysaccharide in glomerulonephritis rats. Int. Immuno-
pharmacol. 2007, 7 (1), 23–28.
Received March 3, 2009. Revised manuscript received July 24, 2009.
Accepted July 28, 2009. This work was supported in part by Grants-in-
Aid for Scientific Research from the Cosmetology Research Foundation
(to Y.F.), a Grant-in-Aid for Scientific Research (No. 21590340 to
R.N.) from the Ministry of Education, Science, Sports and Cultures of
Japan, and Grants-in-Aid for Scientific Research from the Takeda
Science Foundation (to T.I.).
(24) Kitagawa, I.; Wang, H.; Yoshikawa, M. Saponin and sapogenol.
XXXVII. Chemical constituents of Astragali Radix, the root of
Astragalus membranaceus Bunge. (4). Astragalosides VII and VIII.
Chem. Pharm. Bull. 1983, 31 (2), 716–722.