Effect of triterpenes and triterpene saponins from the stem bark of Kalopanax pictus on the transactivational activities of three PPAR subtypes

Article abstract of DOI:10.1016/j.carres.2011.08.029

Kalopanax pictus (Araliaceae) is a deciduous tree that grows in East Asian countries. Its stem bark and leaves have been used in traditional medicine to treat rheumatic arthritis, neurotic pain, and diabetes mellitus. A phytochemical study on a methanol extract of the stem bark of K. pictus resulted in the isolation of three new compounds, 6β,16α-dihydroxy-hederagenin 3-O-β-d-glucuronopyranoside (1), 3-O-β-d-glucuronopyranosyl-28-O- β-d-glucopyranosyl-6β,16α-dihydroxy-oleanolic acid (2), and 3-O-β-d-galactopyranosyl(1→3)-α-l-arabinopyranosyl hederagenin 28-O-β-d-glucopyranosyl-(1→6)-β-d-glucopyranosyl ester (3), along with eight known compounds (4-11). Their structures were established on the basis of chemical and spectroscopic methods (IR, 1D and 2D NMR, and HRESITOFMS). Compounds 1-6 and 8-10 upregulated PPARs transcriptional activity in a dose-dependent manner in HepG2 cells, with EC₅₀ values in the range 0.20-15.5 μM. Moreover, the specific PPAR transactivational effects of compounds 1-6 and 8-10 on separate PPAR subtypes, PPARα, -γ, and -β(δ) were further investigated. Compounds 4, 5, 8, and 10 showed significant PPARα transactivational activity, with EC₅₀ values of 7.8, 8.0, 10.3, and 17.3 μM, respectively. Compounds 2, 4, 6, and 8-10 exhibited PPARγ dose-dependent transactivational activity, with EC ₅₀ values of 14.7, 15.5, 14.8, 10.9, 17.1, and 16.3 μM, whereas compounds 8 and 10 significantly upregulated PPARβ(δ) transcriptional activity, with EC₅₀ values of 15.7 and 17.7 μM, respectively.

Full text of DOI:10.1016/j.carres.2011.08.029

Carbohydrate Research 346 (2011) 2567–2575
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Effect of triterpenes and triterpene saponins from the stem bark of Kalopanax
q
pictus on the transactivational activities of three PPAR subtypes
Tran Hong Quang , Nguyen Thi Thanh Ngan , Chau Van Minh , Phan Van Kiem , Nguyen Phuong Thao a,b,
a,b
a
b
b
a
a,b
a
a,
⇑
Bui Huu Tai , Nguyen Xuan Nhiem , Seok Bean Song , Young Ho Kim
College of Pharmacy, Chungnam National University, Daejeon 305-764, Republic of Korea
a
b
Institute of Marine Biochemistry, Vietnam Academy of Science and Technology (VAST), 18 Hoang Quoc Viet, Caugiay, Hanoi, Viet Nam
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 June 2011
Received in revised form 12 August 2011
Accepted 25 August 2011
Available online 9 September 2011
Kalopanax pictus (Araliaceae) is a deciduous tree that grows in East Asian countries. Its stem bark and
leaves have been used in traditional medicine to treat rheumatic arthritis, neurotic pain, and diabetes
mellitus. A phytochemical study on a methanol extract of the stem bark of K. pictus resulted in the
isolation of three new compounds, 6b,16
-O-b- -glucuronopyranosyl-28-O-b- -glucopyranosyl-6b,16
-galactopyranosyl(1?3)- -arabinopyranosyl hederagenin 28-O-b-
a
-dihydroxy-hederagenin 3-O-b-
-dihydroxy-oleanolic acid (2), and 3-O-b-
-glucopyranosyl-(1?6)-b- -gluco-
D-glucuronopyranoside (1),
3
D
D
D
a
a
-
L
D
D
Keywords:
Kalopanax pictus
Araliaceae
Triterpene saponin
PPAR transactivational activity
pyranosyl ester (3), along with eight known compounds (4–11). Their structures were established on the
basis of chemical and spectroscopic methods (IR, 1D and 2D NMR, and HRESITOFMS). Compounds 1–6
and 8–10 upregulated PPARs transcriptional activity in a dose-dependent manner in HepG2 cells, with
EC50 values in the range 0.20–15.5
pounds 1–6 and 8–10 on separate PPAR subtypes, PPAR
pounds 4, 5, 8, and 10 showed significant PPAR transactivational activity, with EC50 values of 7.8, 8.0,
0.3, and 17.3 M, respectively. Compounds 2, 4, 6, and 8–10 exhibited PPAR dose-dependent transac-
tivational activity, with EC50 values of 14.7, 15.5, 14.8, 10.9, 17.1, and 16.3 M, whereas compounds 8 and
0 significantly upregulated PPARb(d) transcriptional activity, with EC50 values of 15.7 and 17.7 M,
respectively.
l
M. Moreover, the specific PPAR transactivational effects of com-
a
, -c, and -b(d) were further investigated. Com-
a
1
l
c
l
1
l
Ó 2011 Elsevier Ltd. All rights reserved.
1
. Introduction
Kalopanax pictus (Araliaceae) is a deciduous tree that grows in
glucose, lipid, and cholesterol metabolism by binding to specific
peroxisome proliferator response elements (PPREs) in the enhan-
7
–10
cer sites of regulated genes.
Accordingly, compounds that mod-
East Asian countries. The stem bark of K. pictus has been used in
traditional medicine to treat rheumatic arthritis, neurotic pain,
and diabetes mellitus.1 Phytochemical studies on the stem bark
of K. pictus demonstrated the presence of hederagenin glycosides,
syringin, and liriodendrin.2 The methanol extract of the stem bark
ulate the functions of PPARs are attractive for the treatment of type
2 diabetes, obesity, metabolic syndrome, inflammation, and car-
diovascular disease. Thus, we initially investigated the effects
of compounds 1–11 on the transcriptional activity of PPARs in hu-
man hepatocarcinoma (HepG2) cells.
11
,3
4
5
of K. pictus has been reported to possess cytotoxic, anti-diabetic,
PPAR
expressed in tissues that exhibit high levels of fatty acid catabo-
lism, such as the liver, heart, kidney, and skeletal muscle. PPAR
regulates the expression of genes involved in several aspects of li-
a, the first identified PPAR family member, is principally
and anti-inflammatory activities.6 In this study, three new com-
pounds (1–3) and eight known compounds (4–11) were isolated
from a methanol extract of the stem bark of K. pictus. Their struc-
tures were elucidated on the basis of chemical and spectroscopic
methods.
a
1
2
pid and glucose homeostasis. PPAR
a is the predominant thera-
peutic target of the fibrates, drugs that are widely used to lower
serum triglycerides and increase HDL cholesterol in patients with
dyslipidemia, atherosclerosis, coronary heart disease, and obesity.
The peroxisome proliferator-activated receptors (PPARs) form a
subfamily of the nuclear receptor superfamily, of which three iso-
forms, PPAR
a
, PPAR
c, and PPAR b(d) have been identified. PPARs
PPARc is expressed predominately in adipose tissue with lower
regulate the expression of genes involved in the regulation of
levels in heart, colon, spleen, intestine, skeletal muscle, kidney, li-
1
3
ver and macrophages. PPAR
pocyte differentiation and lipid metabolism. PPAR
therapeutic potential in the treatment of type 2 diabetes.
c
controls the genes involved in adi-
agonists have
c
q
PPAR = Peroxisome proliferator-activated receptor.
14
⇑
Corresponding author. Tel.: +82 42 821 5933.
E-mail address: yhk@cnu.ac.kr (Y.H. Kim).
PPARb(d) is ubiquitously expressed but displays a higher level of
0
008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.08.029

2
568
T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
expression in tissues controlling lipid metabolism, such as the adi-
pocytes, small intestine, heart, skeletal muscle, and macro-
phages.1
The molecular formula of compound 1 was determined to be
36 56
C H O12 from the pseudomolecular ion peak at m/z 681.3862
5
+
[M+H] (calcd for C36
H
57
O
12, 681.3850) in the HRESITOFMS. The
H NMR spectrum exhibited six tertiary methyl groups at d 1.03
(Me-24), 1.30 (Me-25), 1.07 (Me-26), 1.29 (Me-27), 0.84 (Me-29),
and 0.95 (Me-30), a broad singlet for an olefinic proton at d 5.32
proton in the pentacyclic triter-
penes, and a doublet of doublets at d 3.10 (1H, J = 13.2, 3.6 Hz)
1
Although the PPAR subtypes share a high level of sequence and
structural homology, each has distinct physiological functions, and
H
1
6
each PPAR subtype exhibits a unique tissue expression pattern.
H
1
2
Thus, with the aim to understand how specifically the compounds
modulate PPAR transcriptional activity, we further examined the
transactivational effects of the compounds on the individual PPAR
(H-12) characteristic of the D
H
1
2
17
1
attributable to H-18 of
D
oleanane skeleton. The H NMR spec-
4.41
1H, br s, H-6) and 4.38 (1H, br s, H-16), and an anomeric proton at
subtypes, PPAR
a, -
c, and -b(d).
trum further showed signals for two oxymethine groups at d
H
(
d
H
4.42 (1H, d, J = 7.8 Hz), indicating that 1 possessed a sugar moi-
2
. Results and discussion
ety. Acid hydrolysis of 1 with 10% HCl and analysis by GC led to the
identification of the sugar as D-glucuronic acid. The C NMR (Ta-
bles 1 and 2) and DEPT spectra exhibited 36 carbon signals, assign-
able to six methyls, nine methylenes, 12 methines, and nine
1
3
Compounds 1–3 were isolated as white, amorphous powders
Fig. 1). The monosaccharides obtained after aqueous acid hydroly-
(
sis of each compound were identified as glucose, glucuronic acid,
galactose, and arabinose by TLC comparison with authentic sam-
ples. The absolute configurations of the monosaccharides were
determined to be
and for arabinose by GC analysis of chiral derivatives of the
monosaccharides in the hydrolysate of each compound (see Sec-
tion 3). The relatively large coupling constants (J 7.8–8.4 Hz) for
the anomeric protons in the 1H NMR spectra of these compounds
suggested
and b- -configurations for the glucopyranosyl, glucuronopyrano-
syl, and galactopyranosyl moieties.
1
3
quaternary carbons. The C NMR spectrum showed signals for
two carboxyl [d 183.9 (C-28) and 177.2 (GlcA-6)], an anomeric
104.8, GlcA-1), an oxymethylene (64.7, C-23), and seven oxy-
methine groups [d 81.9 (C-3), 68.9 (C-6), 75.9 (C-16), 75.1 (GlcA-
), 78.3 (GlcA-3), 73.7 (GlcA-4), and 76.5 (GlcA-5)]. The HMBC
cross peak from H-1 of the glucuronic acid (d 4.42) to C-3 (d
1.9) indicated that the glucuronic acid attached to C-3 of the agly-
cone of 1 (Fig. 2). The upfield chemical shift of C-6 (d 68.9) and the
C
[d
C
D for glucose, glucuronic acid, and galactose,
L
C
2
H
C
8
a-L-configurations for the arabinopyranosyl moieties
C
downfield chemical shift of C-16 (d
signals of H-6 (d 4.41) and H-16 (d
droxyl groups at C-6 and C-16 are b- and a-oriented, respectively.
This was consistent with the ROE correlations between H-27/H-9,
C
75.9), and the broad singlet
H
4.38), revealed that the hy-
D
H
29
30
H-9/H-5, H-5/H-6, H-26/H-15a, and H-16/H-15a (Fig. 3). Based on
2
0
the above analyses, the structure of 1 was established as 6b,16
dihydroxy-hederagenin 3-O-b- -glucuronopyranoside.
The HRESITOFMS of compound 2 exhibited a pseudomolecular
a-
1
2
18
22
D
2
5
26
28
COOR5
R4
+
ion peak at m/z 827.4404 [M+H] (calcd for C42
consistent with the molecular formula of C42
spectrum showed the presence of seven tertiary methyl groups
at d 1.10 (Me-23), 1.19 (Me-24), 1.28 (Me-25), 1.02 (Me-26),
.29 (Me-27), 0.86 (Me-29), and 0.92 (Me-30), a broad singlet for
an olefinic proton at d 5.33 (H-12), and two broad singlets for
two oxymethine groups at d
4.45 (H-6) and 4.50 (H-16). The 1
NMR spectrum further revealed the presence of two anomeric pro-
tons at d 4.31 (1H, d, J = 8.4 Hz) and 5.32 (1H, d, J = 7.8 Hz) (Ta-
67
H O16, 827.4429),
H O16. The H NMR
66
14
16
1
1
9
3
5
7
27
H
1
R O
1
R2
2
3
24
R3
H
H
H
R1
R2
R3
R4
R5
H
ble 2), indicating that 2 possessed two sugar units. By acid
hydrolysis and GC analysis, the sugar units were identified as
1
GlcA
CH OH
OH OH
OH OH
H
2
1
3
D-glucose and D-glucuronic acid. The C NMR (Tables 1 and 2)
2
3
4
5
6
GlcA
CH3
Glc
and DEPT spectra exhibited 42 carbon signals, including seven
methyls, nine methylenes, 17 methines, and nine quaternary car-
3
6
Ara(p) Gal
CH OH
H
H
H
H
H
H
Glc Glc
2
bons. The downfield chemical shift of C-3 (d
C
91.0) and the upfield
13
H
CH OH
OH
H
2
chemical shift of C-28 of the aglycone (d 177.4) in the C NMR
C
spectra of 2 clearly revealed this compound to be bisdesmosidic
H
CH OH
H
H
H
H
2
13
saponin. The C NMR spectrum of 2 was found to be similar to that
Ara(p)
CH OH
of 1, except for the conversion of an oxymethylene (C-23) in 1 to a
2
methyl group (d
sugar moiety at C-28 in 2. The HMBC spectrum displayed the cross
peaks of H-1 of glucuronic acid with C-3 (d 91.0), and H-1 of glu-
cose (d 5.32) with C-28 (d 177.4), indicating that the glucuronic
C
28.3, C-23) in 2, and the additional presence of a
7
8
9
H
H
H
CH OH
H
H
H
H
H
Glc
2
6
CH OH
H
H
H
H
Glc Glc
C
2
H
C
6
4
CH3
Glc Glc Rha
acid and glucose were attached to C-3 and C-28 of the aglycone
of 2, respectively. The orientations of the hydroxyl groups at C-6
6
4
1
0
1
Ara(p)
CH OH
Glc Glc Rha
H
2
and C-16 were determined to be b and
the ROESY correlations between H-27/H-9, H-9/H-5, H-5/H-6, H-
6/H-15a, and H-16/H-15a (Fig. 3). Based on these data, the struc-
ture of 2 was identified as 3-O-b- -glucuronopyranosyl-28-O-b-
glucopyranosyl-6b,16 -dihydroxy-oleanolic acid.
a, respectively, based on
Ara(p) 2 Rha Xyl CH OH
3
1
2
2
D
D-
Ara(p): α-L-arabinopyranosyl
GlcA: β-D-glucuronopyranosyl
Rha : α-L-rhamnopyranosyl
a
Gal
Glc
: β-D-galactopyranosyl
: β-D-glucopyranosyl
The molecular formula of compound 3 was determined to be
ꢀ
C
1
53
H
86
O
23 from the pseudomolecular ion peak [MꢀH] at m/z
Xyl : β-D-xylopyranosyl
089.5449 (calcd for C53
85
H O23, 1089.5482) in the HRESITOFMS.
Figure 1. Chemical structures of compounds 1–11.
Acid hydrolysis of
3
with 10% HCl produced hederagenin,

T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
2569
Table 1
1
1
3
a
H
(
6
0
0
M
H
z
)
a
n
d
C
N
M
R
(
1
5
0
M
H
z
)
d
a
t
a
f
o
r
t
h
e
a
g
l
y
c
o
n
e
s
o
f
c
o
m
p
o
u
n
d
s
1
–
3
Position
1
2
3
b
b
b
d
H
d
C
d
H
d
C
d
H
d
C
1
2
0.95
.54
1.79
.92
41.9
1.52
1.67
1.93
1.96
41.8
1.57
0.93
2.03
1.74
39.7
26.3
1
26.4
27.2
1
3
4
5
6
3.62, dd (12.0, 4.2)
81.9
44.9
49.2
68.9
3.11, dd (12.0, 3.6)
91.0
41.2
57.5
68.9
3.51, dd (12.0, 4.2)
82.8
44.1
48.7
19.0
1.21, br s
4.41, br s
0.73, br s
4.45, br s
1.59
1.45
1.31
7
1.52
41.5
1.52
1.63
42.2
1.73
1.56
33.3
1.78
8
9
0
1
40.0
49.0
37.3
24.5
40.0
48.6
37.5
24.5
40.8
49.1
37.7
24.7
1.67, dd (11.4, 6.0)
1.86
1.57
1
1
1.90
2.02
2.00
1.98
1.86
1
1
1
1
2
3
4
5
5.32, br s
123.3
144.9
43.4
5.33, br s
124.1
143.8
43.2
5.22, br s
123.9
145.0
43.0
1.30
36.2
1.34, dd (15.0, 3.6)
1.87
4.50, br s
36.3
1.74
1.05
2.04
29.0
1.93
1
6
4.38, br s
75.9
75.0
24.2
1.67
1
1
1
7
8
9
48.7
42.7
48.1
50.0
42.3
47.8
48.2
42.6
47.3
3.10, dd (13.2, 3.6)
0.99, dd (13.2, 3.6)
2.98, dd (14.4, 4.2)
1.04
2.25, t (13.2)
2.93, dd (13.2, 4.2)
1.68
1.11
2.18, t (13.2)
2
2
0
1
31.4
36.7
31.3
36.5
31.6
35.0
1.10
.79
1.72
.82
3.44, d (11.4)
.71, d (11.4)
1.03, s
1.30, s
1.07, s
1.29, s
1.88
1.37, ddd (13.8, 13.8, 4.2)
1.17
1.73
1.21
3.74, d (11.4)
3.22, d (11.4)
0.69, s
0.95, s
0.76, s
1
1.34, dd (15.0, 3.6)
1.91
2
2
3
31.7
64.7
31.6
28.3
33.4
65.2
1
2
1.10, s
3
24
25
26
27
28
29
30
15.0
18.0
19.3
27.7
183.9
33.7
25.9
1.19, s
1.28, s
1.02, s
1.29, s
18.6
17.6
19.1
27.4
177.4
33.4
25.2
13.6
16.7
18.0
26.4
178.2
33.6
24.2
1.13, s
0.84, s
0.95, s
0.86, s
0.92, s
0.88, s
0.91, s
a
Spectra were recorded in CD
Protons are overlapped, unless otherwise stated (s, singlet; d, doublet; t, triplet).
3
OD. Coupling constants (J) are in Hz.
b
D
-glucose,
nals (Table 2) at
J = 7.8 Hz), 4.57 (1H, d, J = 7.8 Hz), and 5.32 (1H, d, J = 7.8 Hz)
gave correlations with anomeric carbon signals (Table 2) at d
04.7, 104.1, 105.8, and 95.8 in the HMQC spectrum, respec-
D
-galactose, and
L-arabinose. The anomeric proton sig-
identified by comparison of their spectroscopic data with those re-
ported in the literature (Fig. 1).
Currently, PPARs are therapeutic targets in the design and
development of agonists for the treatment of type 2 diabetes and
metabolic syndrome. The effects of compounds 1–11 on the activa-
tion of PPARs were evaluated using a PPRE luciferase reporter assay
d
H
4.32 (1H, d, J = 8.4 Hz), 4.50 (1H, d,
C
1
tively, indicating that 3 possessed four sugar units. The sequence
of the sugar residues was unambiguously defined by the HMBC
experiment. The HMBC cross peaks from H-1 of galactose (d
(
see Section 3.5.2). The results showed that compounds 1–6 and 8–
0 significantly activated the transcriptional activity of PPARs in a
dose-dependent manner (Fig. 4), with EC50 values ranging from 0.2
to 15.5 M, whereas compounds 7 and 11 were not active at the
H
1
4
4
.50) to C-3 of arabinose (d
C
83.6), and H-1 of arabinose (d
H
.57) to C-3 of the aglycone (d
C
82.8) revealed that the disaccha-
-galactopyranosyl-
-arabinopyranosyl. In the same way, the disaccharide
moiety of 3 linked to C-28 was identified as b- -glucopyrano-
syl-(1?6)-b- -glucopyranosyl based on the HMBC correlations
from H-1 of glucose (d 4.32) to C-6 of C-28-glucose (d 69.6),
and from H-1 of C-28-glucose (d 5.32) to C-28 of the aglycone
178.2). Thus, the structure of 3 was established as 3-O-b-
galactopyranosyl(1?3)- -arabinopyranosyl hederagenin 28-O-
b- -glucopyranosyl-(1?6)-b- -glucopyranosyl ester.
The known compounds, caulophyllogenin (4), hederagenin
l
ride moiety of 3 attached to C-3 was b-
1?3)-
D
tested concentrations (Table 3). Among the compounds tested,
compounds 3, 8, and 9 showed the most potent activity, with
(
a-L
D
EC50 values of 1.2, 1.0 and 0.2
activities of compounds 8 and 9 were more potent than that of
the positive control, benzafibrate (EC50 = 1.16 M).
lM, respectively. Remarkably, the
D
H
C
l
H
From these primary data, with the aim of understanding how
specifically the compounds modulate PPAR transcriptional activity,
we further examined the PPAR transactivational effects of the com-
(d
C
D-
a
-L
D
D
pounds on individual PPAR subtypes, including PPAR
b(d).
The expression of PPAR
to tissue, suggesting that the PPAR subtypes play different biolog-
ical roles. PPAR agonists have been reported to decrease hepatic,
a, -c, and
18
-
5),1 hederagenin 3-O-
8-O-b- -glucopyranosyl ester (7), dipsacussaponin A (8), cus-
sonoside B (9), cauloside D (10), and sapindoside B (11), were
9
a-
L
-arabinopyranoside (6), hederagenin
20
(
2
a, -c, and -b(d) varies widely from tissue
21
22
24
D
2
3
20
a

2
570
T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
Table 2
1
1
3
a
H
(
6
0
0
M
H
z
)
a
n
d
C
N
M
R
(
1
5
0
M
H
z
)
d
a
t
a
f
o
r
t
h
e
s
u
g
a
r
m
o
i
e
t
i
e
s
o
f
c
o
m
p
o
u
n
d
s
1
–
3
Position
1
2
3
b
d
H
d
C
d
H
d
C
d
H
d
C
3
1
2
3
4
5
-O-Ara
4.57, d (7.8)
3.56, dd (8.4, 7.8)
3.61, dd (8.4, 3.6)
3.81
105.8
73.6
83.6
70.1
68.3
3.78, dd (11.4, 3.0)
.46, br d (11.4)
3
3
1
2
3
4
5
6
-O-GlcA
4.42, d (7.8)
3.23, dd (8.4, 7.8)
3.36, t (8.4)
3.38, dd (9.0, 8.4)
3.56, d (9.0)
104.8
75.1
78.3
73.7
76.5
177.2
4.31, d (8.4)
3.22, dd (9.6, 8.4)
3.36
3.42, t (9.6)
3.52, d (9.6)
106.8
74.1
78.3
73.8
76.7
177.4
Gal
1
2
3
4
4.50, d (7.8)
3.29, dd (8.4, 7.8)
3.44
3.39
3.44
3.68
104.1
74.0
75.4
71.0
76.9
62.1
5
6
2
1
2
3
4
5
6
8-O-Glc
5.32, d (7.8)
3.23, dd (9.6, 7.8)
3.35, t (9.6)
3.31, dd (9.6, 8.4)
3.27
95.9
75.6
78.1
71.2
78.8
62.5
5.32, d (7.8)
3.45, dd (8.4, 7.8)
3.58, t (9.0)
3.80, dd (9.0, 8.4)
3.38
4.10, dd (11.4, 2.4)
3.73, br d (12.0)
95.8
74.9
78.4
72.3
78.1
69.6
3.78, br d (12.0)
3.64, dd (12.0, 4.2)
Glc
1
2
3
4
4.32, d (8.4)
104.7
75.2
78.3
71.6
77.9
62.8
3.18, dd (9.0, 8.4)
3.49, dd (9.0, 7.8)
3.39, dd (9.0, 8.4)
3.20
5
6
3.83, dd (12.0, 3.0)
3.80, br d (12.0)
a
3
Spectra were recorded in CD OD. Coupling constants (J) are in Hz.
Protons are overlapped, unless otherwise stated (s, singlet; d, doublet; t, triplet).
b
compounds tested, compounds 4, 5, 8, and 10 exhibited significant
effects on the PPAR transactivational activity in a dose-dependent
manner (Fig. 5), with EC50 values of 7.8, 8.0, 10.3, and 17.3 M,
respectively, whereas compound 9 showed moderate activity with
an EC50 value of 24.2 M. (Table 3).
Among PPARs, PPAR is the predominant molecular target for
a
l
l
O
c
insulin-sensitizing thiazolidinedione drugs such as troglitazone,
OH
pioglitazone, and rosiglitazone, all of which have been approved
HOOC
OH
2
5,26
for use in the treatment of type 2 diabetes patients.
of compounds 1–6 and 8–10 on the activation of PPAR
uated using a GAL-4-PPAR
Among the compounds tested, 2, 4, 6, and 8–10 showed significant
PPAR transactivation in a dose-dependent manner (Fig. 6), with
EC50 values of 14.7, 15.5, 14.8, 10.9, 17.1, and 16.3 M, respectively
Table 3).
Among the three PPAR subtypes, knowledge of the role of
The effects
O
HO
O
c
were eval-
HO
c
chimera assay (see Section 3.5.3).
OH
OH
OH
c
1
1
H- HCOSY
l
(
HMBC
Figure 2. Selective 1H– H COSY and HMBC correlations of compound 1.
1
PPARb(d) is lacking because of the late development of specific
agonists. Currently, the use of two synthetic selective agonists,
compounds GW501516 and GW610742 in animal and various cell
line models, and the concomitant development of new genetically
modified mouse models, are helping to unravel the role of
PPARb(d) in lipid and glucose homeostasis and inflammation. As
such, PPARb(d) has become a pharmacological target for the treat-
ment of metabolic disorders associated with the metabolic syn-
drome, including dyslipidaemia, obesity, and insulin resistance.15
muscle and pancreas lipotoxicity (steatosis), to reduce insulin
resistance, to normalize glucose levels, and to reduce cardiovascu-
lar risk factors, such as dyslipidaemia, inflammation, prothrombot-
ic state, and intima–media thickness.1 The effects of compounds
5
1
–6 and 8–10 on the activation of PPARa were evaluated using a
GAL-4-PPAR chimera assay (see Section 3.5.3). Among the
a

T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
2571
2
8
28
COOH
COOR5
H
H
1
1
5a
26
15a
13
2
5CH3
OH
26 CH3
H
3
H
25CH3
CH
H
H
3
12
12
18
18
24CH3
24CH3
OH
H
H
1
9
3
0CH3
9CH3
1
9
30CH3
R O
R O
1
6
16
5
H15b
5
H15b
3
6
3
6
HOH C23
H
H
27CH3
OH 2
H C23
H
H
27CH3
OH
2
3
29CH3
H
H
H
H
H
H
1
2
Figure 3. Selective ROESY correlations of compounds 1 and 2.
⁄
Figure 4. PPARs transactivational activity of compounds 1–11 in HepG2 cells. (ꢀ): vehicle group. (+): positive control (1
vehicle group (P <0.05).
l
M): benzafibrate. Significantly different from
Table 3
PPAR transactivational activities of compounds 1–11
Compound
EC50a
(lM)
PPARs
Gal4/PPAR
a-LBD
Gal4/PPAR
c
-LBD
Gal4/PPARb(d)-LBD
1
2
3
4
5
6
7
8
9
5.5 ± 0.8
4.0 ± 0.9
1.2 ± 0.1
6.1 ± 0.7
3.6 ± 0.6
5.4 ± 0.9
>40
>40
>40
7.8 ± 1.4
8.0 ± 1.7
>40
36.4 ± 3.3
14.7 ± 4.2
39.4 ± 3.2
15.5 ± 1.5
37.8 ± 8.7
14.8 ± 0.9
>40
31.3 ± 1.4
28.7 ± 2.8
>40
>40
31.6 ± 3.6
⁄
>40
1.0 ± 0.1
0.20 ± 0.05
15.5 ± 1.1
>40
10.3 ± 2.1
24.2 ± 2.6
17.3 ± 2.2
10.9 ± 1.3
17.1 ± 0.9
16.3 ± 1.5
15.7 ± 2.7
23.0 ± 3.4
17.7 ± 1.1
1
1
0
1
Benzafibrate
Ciprofibrate
Troglitazone
L-165.041
1.1 ± 0.2
0.9 ± 0.1
0.70 ± 0.05
0.50 ± 0.02
A compound is considered inactive with EC50 >40
l
M.
a
EC50: the concentration of test compound that gave 50% of the maximal reporter activity.
The effects of compounds 1–6 and 8–10 on the activation of
PPARb(d) were evaluated using a GAL-4-PPARb(d) chimera assay
compounds 2, 3, 6, and 9 showed moderate activity, with EC50 val-
ues of 31.3, 28.7, 31.6, and 23.0 M, respectively (Table 3).
Although PPAR activation is involved in the treatment of vari-
ous diseases, it can have a side effect of inducing weight gain.
PPAR dual agonists have been postulated to be a useful combi-
l
(
see Section 3.5.3). The results showed that compounds 1–3, 5, 6,
c
16
and 8–10 dose-dependently affected the PPARb(d) transactivation-
al activity (Fig. 7). Among the compounds tested, 8 and 10 exhib-
ited significant effects on PPARb(d) transactivational activity,
a/c
nation strategy to achieve a broad spectrum of metabolic effects
with EC50 values of 15.7 and 17.7 lM, respectively, whereas
and reduce mortality rates associated with type 2 diabetes by

2
572
T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
⁄
Figure 5. PPAR
from vehicle group (P <0.05).
a
transactivational activity of compounds 1–6 and 8–10 in HepG2 cells. (ꢀ): vehicle group. (+): positive control (1
l
M): ciprofibrate. Significantly different
⁄
Figure 6. PPAR
from vehicle group (P <0.05).
c
transactivational activity of compounds 1–6 and 8–10 in HepG2 cells. (ꢀ): vehicle group. (+): positive control (1
l
M): troglitazone. Significantly different
improving insulin resistance, hyperglycemia, and alleviating ath-
agents for the treatment of type 2 diabetes without inducing
weight gain. Combining the agonist activity of all three PPAR sub-
types (PPARs agonist) into a single chemical entity could poten-
tially deliver a drug that treats a broad spectrum of metabolic
diseases by improving insulin sensitization, obesity, dyslipidemia,
and hypertension as well as providing beneficial effects on inflam-
matory markers. Remarkably, among the compounds tested, 8
and 10 displayed significant PPAR transactivational activity on
three PPAR subtypes, suggesting that these compounds might be
useful agents for the treatment of the broad metabolic diseases.
Taken together, our findings have demonstrated a potential
mechanism for the traditional anti-diabetic and anti-inflammatory
1
6
erogenic dyslipidaemia. The propensity of PPAR
duce weight gain through its adipogenic effects may be offset by
the ability of PPAR activation to stimulate lipid catabolism, there-
by forming a combination that has reduced propensity for inducing
c activation to in-
a
1
6
undesired weight gain. Also, PPARc/b(d) dual agonists have been
1
6
postulated to be a beneficial combination therapy for type 2 diabe-
tes. Furthermore, the propensity of PPARb(d) activation to improve
insulin sensitivity and increase fatty acid oxidation suggests that a
dual PPAR
realized with selective PPAR
combinations such as PPAR
PPAR /b(d) agonists are emerging. Compounds 4, 8, and 10 were
found to show significant effects on both PPAR and transactiva-
tional activities, indicating that these compounds act as PPAR
dual agonists. Compounds 8 and 10 exhibited significant effects
on both PPARb(d) and transactivational activities, revealing that
these compounds act as PPAR /b(d) dual agonists. With the signif-
icant PPAR and /b(d) dual transactivational activities, these
data suggest that compounds 4, 8, and 10 may be promising new
c
/b(d) agonist could attenuate the undesired weight gain
agonists.16 Thus, the PPAR subtype
and /b(d) dual agonists and
c
a
/c
c
a/c
actions of the stem bark of K. pictus via activation of the PPARs, -
a,
a
c
-c, and -b(d) by triterpene saponin components. This study also
a
/c
provides scientific support to the use of the stem bark of K. pictus
for the treatment of diabetes, obesity, metabolic syndrome, inflam-
mation, and cardiovascular diseases. Besides, bioavailability is a
critical issue for saponins.
Cell viability, as measured by the MTS assay (see Experimental
Section 3.5.4), showed that compounds 1–11 had no significant
c
c
a
/c
c

T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
2573
⁄
lM): L-165.041. Significantly different
Figure 7. PPARb(d) transactivational activity of compounds 1–6 and 8–10 in HepG2 cells. (ꢀ): vehicle group. (+): positive control (1
from vehicle group (P <0.05).
cytotoxicity in HepG2 cells at tested concentrations (data not
shown).
fractions (C1–C5). Fraction C2 (2.2 g) was chromatographed over
silica gel column eluting with 17:1 CH Cl –MeOH to obtain 4
16 mg). Fraction C4 (3.5 g) was separated by reversed-phase chro-
2
2
(
2
matography using 3:1 MeOH–H O as eluent to give 5 (20 mg) and
3
3
. Experimental
7
(55 mg). Fraction C5 (2.6 g) was chromatographed over silica gel
column eluting with 10:1 CH
2
Cl
2
–MeOH, and further purified by
O as eluent
to afford 11 (10 mg), and 6 (6 mg). Fraction D was chromato-
graphed on a column of Diaion HP-20 and eluted with H O and
MeOH successively to give three fractions (D1–D3). Fraction D2
26 g) was chromatographed over silica gel, eluting with MeOH
in CH Cl (0–100%, stepwise) to provide five subfractions (D2A–
D2E). Subfraction D2A (4.3 g) was separated by reversed-phase
chromatography, using 2:1 MeOH–H O as eluent, and further puri-
fied by column chromatography over silica gel, eluting with 6:1:0.1
CH Cl –MeOH–H O to obtain 1 (8 mg) and 2 (10 mg). Subfraction
D2C (2.3 g) was separated by column chromatography over silica
gel, using 4:1:0.1 CH Cl –MeOH–H O as eluent, and further puri-
fied by reversed-phase chromatography, eluting with 1:1 MeOH–
O to give 3 (6 mg). Fraction D3 (35 g) was chromatographed over
silica gel, eluting with MeOH in CH Cl (0–100%, stepwise) to pro-
.1. General experimental procedures
Optical rotation was determined using a Jasco DIP-370 digital
reversed-phase chromatography using 2:1 MeOH–H
2
polarimeter. The IR spectra were measured using a Jasco Report-
00 infrared spectrometer. Electrospray-ionization (ESI) mass
2
1
(
spectra were obtained using an Agilent 1200 LC–MSD Trap spec-
trometer. Gas chromatography was carried out on a Shidmazu-
2
2
2
010 spectrometer, detector: FID, detection temperature: 300 °C,
column: SPB-1 (0.25 mm i.d. ꢁ 30 m), column temperature:
30 °C, carrier gas He (2 mL/min), injection temperature: 250 °C,
injection volume: 0.5 L. HRESITOF mass spectra were obtained
using a JEOL JMS-T100LC spectrometer. The NMR spectra ( H,
2
2
l
2
2
2
1
1
3
1
1
C, DEPT, HMQC, HMBC, H– H COSY, and ROESY) were recorded
2
2
2
on a Jeol ECA 600 spectrometer using TMS as an internal standard.
TLC was performed on Kieselgel 60 F254 (1.05715; E. Merck,
Darmstadt, Germany) or RP-18 F254s (E. Merck) plates. Spots were
visualized by spraying with 10% aq H SO , followed by heating.
2 4
Column chromatography (CC) was performed on silica gel (Kiesel-
gel 60, 70–230 mesh and 230–400 mesh, E. Merck) and YMC RP-18
H
2
2
2
vide four subfractions (D3A–D3D). Subfraction D3B (6.8 g) was
then separated by column chromatography over silica gel, using
5
:1:0.1 CH
versed-phase chromatography, eluting with 2:1 MeOH–H
afford 8 (15 mg). Subfraction D3D (5.6 g) was separated by re-
versed-phase chromatography, using 15:10 MeOH–H O as eluent,
and further purified by column chromatography over silica gel,
eluting with 4:1:0.1 CH Cl –MeOH–H O to obtain 9 (12 mg) and
2
Cl
2
–MeOH–H
2
O as eluents, and further purified by re-
O to
(
Cosmosil 75C18-PREP) resins.
2
3
.2. Plant materials
2
The stem bark of of K. pictus was purchased from a herbal mar-
2
2
2
1
0 (23 mg).
ket at Kumsan, Chungnam, Korea, in August 2009. The plant mate-
rial was identified by one of us (Y. H. Kim). A voucher specimen
3
.3.1. 6b,16
a
-Dihydroxy-hederagenin 3-O-b-D-
(
CNU09105) was deposited at the herbarium, College of Pharmacy,
glucuronopyranoside (1)
Chungnam National University.
2
D
5
White, amorphous powder; ½
aꢂ
ꢀ23.3 (c 0.12, MeOH); FTIR
ꢀ1 1
(
(
CH
CD
3
CN)
m
max 3434, 2913, 1613, 1417, 1372, 1016 cm
.
H NMR
3
.3. Extraction and isolation
1
3
3
OD, 600 MHz) and C NMR data (CD OD, 150 MHz), see Ta-
3
+
bles 1 and 2; HRESITOFMS m/z 681.3862 [M+H] (cacld for
The dried stem bark of K. pictus (3 kg) was extracted with hot
36 57
C H O12, 681.3850).
MeOH under reflux. After concentration, the MeOH extract
290 g) was suspended in water and then partitioned successively
with n-hexane, CH Cl , and EtOAc to give n-hexane (A, 50 g),
CH Cl (B, 30 g), EtOAc (C, 15 g) and water (D, 195 g) fractions,
respectively. Fraction C was chromatographed over silica gel, elut-
ing with MeOH in CH Cl (0–100%, step-wise), yielding five
(
3
6
.3.2. 3-O-b-
b,16 -dihydroxy-oleanolic acid (2)
White, amorphous powder; ½
CH CN) max 3379, 2926, 1733, 1606, 1429, 1024 cm
D-Glucuronopyranosyl-28-O-b-D-glucopyranosyl-
2
2
a
2
2
2
5
aꢂ
ꢀ12.8 (c 0.12, MeOH); FTIR
ꢀ1 1
D
(
3
m
.
H NMR
2
2

2
574
T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
(
CD
3
OD, 600 MHz) and 13C NMR data (CD
3
OD, 150 MHz), see Ta-
3.5.3. PPAR subtype specific transactivational assay
+
5
bles 1 and 2; HRESITOFMS: m/z 827.4404 [M+H] (calcd for
C
Human hepatoma cells (HepG2) were seeded at 1.5 ꢁ 10 cells
42
H
67
O16, 827.4429).
per well in 12-well plates and grown for 24 h before transfection.
Cells were transfected separately with one pGal4-PPAR subfamily
3
.3.3. 3-O-b-
D
-Galactopyranosyl(1?3)-
a
-L
-arabinopyranosyl
vector [pFA-Gal4-PPAR
a-LBD, pFA-Gal4-PPARc-LBD, or pFA-Gal4-
hederagenin 28-O-b-
glucopyranosyl ester (3)
D
-glucopyranosyl-(1?6)-b-D-
PPARb(d)-LBD expression plasmids, provided by Dr. Young Yang,
2
7
Sookmyung Women’s University, Seoul, Korea] together with
pFR-Luc using the WelFect M™ Gold transfection reagent (Wel-
GENE Inc, Daegu, South Korea), as described by the manufacturer.
After 24 h of transfection, the medium was changed to Opti-MEM
(Invitrogen, Carlsbad, CA) containing 0.1 mM NEAA, 0.5% char-
coal-stripped FBS, and each compound (test group), dimethyl sulf-
oxide (vehicle group), ciprofibrate (positive control group for
2
D
5
White, amorphous powder; ½
aꢂ
+4.5 (c 0.1, MeOH); FTIR
ꢀ
1 1
(
3
CH CN)
m
max 3345, 2934, 1717, 1601, 1419, 1039, 668 cm
. H
NMR (CD
3
OD, 600 MHz) and 13C NMR data (CD
3
OD, 150 MHz),
ꢀ
see Tables 1 and 2; HRESITOFMS: m/z 1089.5449 [MꢀH] (cacld
for C53
85
H O23, 1089.5482).
3
.4. Acid hydrolysis and sugar identification
PPAR
65.041 [positive control group for PPARb(d)]. The cells were then
cultured for 20 h, after which time the cells were washed with PBS
L). The intensity
of emitted luminescence was determined using a Centro LB 960
microplate luminometer (EG&G Berthold, Bad Wildbad, Germany)
by measuring light emission for 5 s.
a), troglitazone (positive control group for PPARc) or L-
1
Each saponin (3 mg) was heated in 3 mL of 10% 1:1 HCl–diox-
ane at 80 °C for 3 h. The solvent was removed in vacuo, and the res-
idue was partitioned between EtOAc and H O to give aglycone and
sugar, respectively. The sugar components in the aqueous layer
were analyzed by silica gel TLC by comparison with standard sug-
and harvested with 1ꢁ passive lysis buffer (200
l
2
ars. The solvent system was 2:1:0.2 CH
were visualized by spraying with 9:0.5:0.5 95% EtOH–H
aldehyde, then heated at 105 °C for 5 min. For sugars of the sapo-
nins, the R values of glucuronic acid, glucose, galactose, and
arabinose by TLC were 0.15, 0.30, 0.32, and 0.5, respectively. The
results were confirmed by GC analysis as follows. The aqueous
layer was evaporated to dryness to give a residue that was dis-
2
Cl
2
–MeOH–H
2
O, and spots
2
SO –anis-
4
3.5.4. Cell viability assay
An MTS assay (Promega Celltiter 96-Aqueous One Solution As-
say) was used to analyze the effect of compounds on cell viability.
f
4
Cells were cultured overnight in a 96-well plate (1 ꢁ 10 cells/
well). Cell viability was assessed after the addition of compounds
at the 10 lM concentrations for 24 h. The number of viable cells
solved in anhydrous pyridine (100
idine solution of 0.1 M -cysteine methyl ester hydrochloride
100 L). After warming at 60 °C for 2 h, trimethylsilylimidazole
l
L) and then mixed with a pyr-
was assessed by determination of the A490nm of the dissolved for-
mazan product after addition of MTS for 30 min as described by
the manufacturer (Promega, Madison, WI).
L
(
l
solution was added and warmed at 60 °C for 2 h. The mixture
was then evaporated in vacuo to give a dried product, which was
3.5.5. Statistical analysis
partitioned between n-hexane and H
filtered and analyzed by GC. The absolute configurations of the
monosaccharides were confirmed to be -arabinose, -glucuronic
acid, -glucose, and -galactose by comparison of the retention
2
O. The n-hexane layer was
All data represent the mean ± SD of at least three independent
experiments performed in triplicates. Statistical significance is
indicated as (p <0.05) as determined by one-way ANOVA followed
⁄
L
D
D
D
by Dunnett’s multiple comparison test.
times of persilylated monosaccharide derivatives with those of
standard samples.
Acknowledgments
3
3
.5. PPAR transactivational assay
This work was supported by a grant from the Korea Food and
Drug Administration for Studies on Standardization of Herbal Med-
icine (2010) and Priority Research Center Program through the Na-
tional Research Foundation of Korea (NRF) funded by the Ministry
of Education, Science and Technology (2009-0093815), Republic of
Korea.
.5.1. Cell culture and reagents
Human hepatocarcinoma HepG2 cells were maintained in
DMEM (Invitrogen, Carlsbad, CA) containing 10% heat-inactivated
fetal bovine serum (FBS), 100 U/mL of penicillin, and 10
streptomycin at 37 °C and 5% CO
l
g/mL
2
.
References
3
.5.2. PPRE-luciferase assay
Human hepatoma cells (HepG2) were seeded at 1.5 ꢁ 10 cells
5
1. Kim, T. J. In Korean Resources Plants; Seoul National University Press: Seoul,
per well in 12-well plates and grown for 24 h before transfection.
An optimized amount of DNA plasmid (0.5 g PPRE-Luc and
.2 g PPAR-inpCMV) was diluted in 100 L Dulbecco’s modified
1996; Vol. 2, pp 169–200.
2
.
Shao, C.-J.; Kasai, R.; Ohtani, K.; Tanaka, O.; Kohda, H. Chem. Pharm. Bull. 1990,
8, 1087–1089.
Sano, K.; Sanada, S.; Ida, Y.; Shoji, J. Chem. Pharm. Bull. 1991, 39, 865–870.
l
3
0
l
l
3
.
Eagle medium (DMEM). All cells were transfected with the plasmid
mixture using WelFect M™ Gold (WelGENE Inc, Daegu, South Kor-
ea) as described by the manufacturer. After 30 min of incubation at
4. Lee, K.-T.; Sohn, I.-C.; Park, H.-J.; Kim, D.-W.; Jung, G.-O.; Park, K.-Y. Planta Med.
000, 66, 329–332.
Park, H.-J.; Kim, D.-H.; Choi, J.-W.; Park, J.-H.; Han, Y.-N. Arch. Pharm. Res. 1998,
1, 24–29.
6. Lee, E. B.; Li, D. W.; Hyun, J. E.; Kim, I. H.; Whang, W. K. J. Ethnopharmacol. 2001,
7, 197–201.
2
5.
2
room temperature, the DNA plasmid solution (100 lL) was intro-
7
duced and mixed gently with cells. After 24 h of transfection, the
medium was changed to Opti-MEM (Invitrogen, Carlsbad, CA) con-
taining 0.1 mM NEAA, 0.5% charcoal-stripped FBS, and the individ-
ual compounds (test group), dimethyl sulfoxide (vehicle group), or
benzafibrate (positive control group). The cells were then cultured
for 20 h. Next, the cells were washed with PBS and harvested with
7
8
.
.
Berger, J.; Moller, D. E. Annu. Rev. Med. 2002, 53, 409–435.
Balint, B. L.; Nagy, L. Endocrinol. Metab. Immune Disord. Drug Targets 2006, 6,
33–43.
9.
Barish, G. D.; Narkar, V. A.; Evans, R. M. J. Clin. Invest. 2006, 116, 590–597.
1
1
0. Haluzik, M. M.; Haluzik, M. Physiol. Res. 2006, 55, 115–122.
1. Kuroda, M.; Mimaki, Y.; Honda, S.; Tanaka, H.; Yokota, S.; Mae, T. Bioorg. Med.
Chem. 2010, 18, 962–970.
12. Wu, P.; Peters, J. M.; Harris, R. A. Biochem. Biophys. Res. Commun. 2001, 287,
1
ꢁ passive lysis buffer (200
lL). The intensity of emitted lumines-
391–396.
cence was determined using an LB 953 Autolumat (EG&G Berthold,
Bad Wildbad, Germany). The PPRE-Luc plasmid was provided by
Dr. Kyoon E. Kim (Chungnam National University, Daejeon, Korea).
13. Escher, P.; Braissant, O.; Basu-Modak, S.; Michalik, L.; Wahli, W.; Desvergne, B.
Endocrinology 2001, 142, 4195–4202.
14. Zhang, B. B.; Moller, D. E. Curr. Opin. Chem. Biol. 2000, 4, 461–467.

T. H. Quang et al. / Carbohydrate Research 346 (2011) 2567–2575
2575
1
1
5. Gross, B.; Staels, B. Best Pract. Res. Clin. Endocrinol. Metab. 2007, 21, 687–710.
6. Shearer, B. G.; Billin, A. N. Biochim. Biophys. Acta – Mol. Cell Biol. Lipids 2007,
22. Kizu, H.; Hirabayashi, S.; Suzuki, M.; Tomimori, T. Chem. Pharm. Bull. 1985, 33,
3473–3478.
1
771, 1082–1093.
23. Dubois, M. A.; Ilyas, M.; Wagner, H. Planta Med. 1986, 52, 80–83.
24. Sawada, H.; Miyakoshi, M.; Isoda, S.; Ida, Y.; Shoji, J. Phytochemistry 1993, 34,
1117–1121.
1
1
7. Siddiqui, B. S.; Karzhaubekova, Z. Z.; Burasheva, G. S.; Sultanova, N. A. Chem.
Pharm. Bull. 2007, 55, 1356–1360.
8. Strigina, L. I.; Chetyrina, N. S.; Isakov, V. V.; Dzizenko, A. K.; Elyakov, G. B.
Phytochemistry 1974, 13, 479–480.
25. Kaplan, F.; Al-Majali, K.; Betteridge, D. J. J. Cardiovasc. Risk 2001, 8, 211–217.
26. Moller, D. E. Nature 2001, 414, 821–827.
1
2
9. Joshi, B. S.; Singh, K. L.; Roy, R. Magn. Reson. Chem. 1999, 37, 295–298.
0. Jhoo, J.-W.; Sang, S.; He, K.; Cheng, X.; Zhu, N.; Stark, R. E.; Zheng, Q. Y.; Rosen,
R. T.; Ho, C.-T. J. Agric. Food. Chem. 2001, 49, 5969–5974.
27. Park, Y. S.; Choi, J.; Kim, K.-Y.; Lim, J.-S.; Yoon, S.; Yang, Y. Biomol. Therapeut.
2010, 18, 77–82.
2
1. Panov, D. A.; Grishkovets, V. I.; Kachala, V. V.; Shashkov, A. S. Chem. Nat. Compd.
2
006, 42, 61–66.