Rearranged oleanane type saponins, astraisoolesaponins A1–A3and B, from the stems of Astragalus membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao

Article abstract of DOI:10.1016/j.tet.2016.09.034

Four novel saponins, astraisoolesaponins A₁–A₃and B (1–4) were isolated from the stems of Astragalus membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao. Spectroscopic analysis indicated that 1–4 are the first example of saponins possessing a 28-methyl-rearranged carbon skeleton. Their absolute configurations were determined by electronic circular dichroism (ECD) calculations. Both 3 and 4 displayed triglyceride accumulation inhibitory effects in HepG2 cells and glucose uptake increase effects on L6 cells.

Full text of DOI:10.1016/j.tet.2016.09.034

Tetrahedron 72 (2016) 7008e7013
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Rearranged oleanane type saponins, astraisoolesaponins A eA and
1
3
B, from the stems of Astragalus membranaceus (Fisch.) Bge. var.
mongholicus (Bge.) Hsiao
a
b
a
a
a
a
Yi Zhang , Yuan-Qiang Guo , Xiao-Xia Li , Jing-Ya Ruan , Ting-Ting Wang , Jian Li ,
a
a
a,
*
Li-Feng Han , Hai-Yang Yu , Tao Wang
Tianjin Key Laboratory of TCM Chemistry and Analysis, Institute of Traditional Chinese Medicine, Tianjin University of Traditional Chinese Medicine,
a
3
12 Anshan Road, Nankai District, Tianjin, 300193, China
Tianjin Key Laboratory of Molecular Drug Research, College of Chemistry, Nankai University, Tianjin, 300071, China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 July 2016
Received in revised form 13 September 2016
Accepted 18 September 2016
Available online 19 September 2016
Four novel saponins, astraisoolesaponins A eA3 and B (1e4) were isolated from the stems of Astragalus
membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao. Spectroscopic analysis indicated that 1e4 are
the first example of saponins possessing a 28-methyl-rearranged carbon skeleton. Their absolute con-
figurations were determined by electronic circular dichroism (ECD) calculations. Both 3 and 4 displayed
triglyceride accumulation inhibitory effects in HepG2 cells and glucose uptake increase effects on L6
cells.
1
Keywords:
Ó 2016 Elsevier Ltd. All rights reserved.
Astragalus membranaceus (Fisch.) Bge. var.
mongholicus (Bge.) Hsiao
28-Methyl-rearranged saponins
Electronic circular dichroism
Triglyceride accumulation
Glucose uptake
HepG2 cells
L6 cells
1. Introduction
electronic circular dichroism (ECD) spectra. In this paper, we de-
scribe their isolation, structure elucidation, and biological activities.
Astragalus genus plants, especially Astragalus membranaceus,
Astragalus mongholicus, and Astragalus complanatus have been
mainly used in folk medicine for their antidiabetic, antioxidative,
anti-inflammatory, anti-cancer, cardioprotective, hepatoprotective,
and antiviral properties, which are owing to the biological active
secondary metabolites, such as saponins, flavonoids, and poly-
saccharides in them.1
2
. Results and discussion
2
The 70% EtOH extract of AM (8.0 kg) was partitioned with H O
and CHCl
3
, the H
2
O layer was subjected to D101 macroporous resin
O and 95% EtOH,
column chromatography (CC) and eluted with H
successively. Then 95% EtOH eluted fraction was isolated by Silica
gel, ODS, Sephadex LH-20 and preparative HPLC CC to afford
astraisoolesaponins A
and B (4, 13.7 mg) (Fig. 1).
Astraisoolesaponin A
2
In our process of continuing to study triglyceride (TG) accu-
mulation inhibitory effective metabolites from the stems of A.
1 2 3
(1, 31.5 mg),A (2, 35.7 mg),A (3, 10.7 mg),
2
e4
membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao (AM),
four novel rearranged oleanane type saponins with 28-methyl-
rearranged carbon skeleton, astraisoolesaponins A (1), A (2), A
3), and B (4) were obtained, which were firstly found in natural
1
(1), was obtained as white powder with
ꢀ73.5 , in MeOH). The molecular formula,
25
ꢁ
1
2
3
negative rotation ([
C H O21, of 1 was established by negative-ion HRESI-TOF-MS (m/z
48 72
a
]
D
(
products. The structures of them were established by analysis of
their NMR spectroscopic data and experimental and calculated
e
1
019.4238 [MþCl] , calcd for C48
72
H O
21Cl, 1019.4260). The IR
ꢀ1
spectrum showed absorption bands due to hydroxyl (3400 cm ),
ꢀ
1
ꢀ1
carboxyl group (1719 cm ), olefinic bond (1642 cm ), and O-
glycosidic linkage (1073 cm ). The UV spectrum displayed maxi-
mum absorption at 294 nm, which indicated the presence of
ꢀ
1
*
Corresponding author. Fax: þ86 22 5959 6163; e-mail address: wangtao@
tjutcm.edu.cn (T. Wang).
a,b,g,d-unsaturated carboxyl group. Acid hydrolysis of 1 with 5%
http://dx.doi.org/10.1016/j.tet.2016.09.034
0
040-4020/Ó 2016 Elsevier Ltd. All rights reserved.

Y. Zhang et al. / Tetrahedron 72 (2016) 7008e7013
7009
spectrum of 1 indicated the presence of seven partial structures
written in bold lines (Fig. 2). The planar structure of the aglycon was
Fig. 1. Structure of compounds 1e4.
aqueous H
-glucose, and
determined by GCeMS analysis of their trimethylsilyl thiazolidine
2
SO
4
e1,4-dioxane (1:1, v/v) afforded
D-glucuronic acid,
D
L-rhamnose, whose absolute configurations were
2
13
5 5
derivatives. The C NMR spectrum (C D N, Table 1) displayed 48
Table 1
1
5 5
H NMR data (500 MHz) for 1 in C D N (J in Hz)
No.
d
H
No.
24
d
H
1
2
3
4
5
6
7
8
9
1.00, m; 1.55, m
1.95, m; 2.38, m
3.40, dd (4.5, 11.5)
d
3.32, d (12.0)
4.32, d (12.0)
0.70, s
0.91, s
1.74, s
25
26
27
28
29
30
0.97, dd (3.0, 11.5)
1.21, m; 1.57, m
1.36, m; 1.54, m
d
6.01, s
d
1.48, s
0
0
0
0
0
0
2.08, br d (ca. 13)
d
1
2
3
4
5
6
5.02, d (7.0)
4.41, dd (7.0, 9.0)
4.71, t (9.0)
4.49, dd (9.0, 9.5)
4.67, d (9.5)
d
1
1
1
1
1
1
0
1
2
3
4
5
1.66, m; 1.95, m
5.33, br s
d
Fig. 2. Key 1He1H COSY and HMBC correlations of 1e4.
d
0
0
0
0
0
0
0
0
0
0
0
0
1.40, m
1
2
3
4
5
6
5.86, d (7.5)
4.34, dd (7.5, 9.0)
4.25, t (9.0)
4.51, t (9.0)
3.62, m
4.28, dd (5.0, 11.5)
4.41, br d (ca. 12)
6.40, br s
4.77, br d (ca. 3)
4.72, dd (3.0, 9.0)
4.37, t (9.0)
5.08, m
determined based on the key HMBC correlations from H
17, C-18, C-28; H -19 to C-13, C-17, C-18, C-21; H -21 to C-22, C-28,
C-29; H -23 to C-3eC-5, C-24; H -25 to C-1, C-5, C-9, C-10; H -26 to
C-7e9, C-14; H -27 to C-8, C-13e15; H-28 to C-16, C-18, C-21; H
0 to C-19eC-21, C-29 (Fig. 2). Moreover, the H and C NMR data
2
-16 to C-
1
.86, dd (12.0, 14.0)
2.16, br d (ca. 12)
.54, t (12.0)
16
2
2
2
d
3
3
3
1
1
1
7
8
9
3
3
-
d
1
13
3
3.02, d (14.0)
3
0
0
0
0
0
0
00
00
00
00
00
00
.44, d (14.0)
1
2
3
4
5
6
on A, B rings and glycosyl groups in 1 were superimposable on
5
2
2
0
1
d
those of azukisaponin V, which indicated 1 was an oleanane type
2.70, d (12.5)
3
saponin with 3
correlations between H-3 and H-5, H
-27; H -24 and H -25; H -25 and H
observed in the 2D NOESY experiment, suggested A, B, C rings were
in chair conformations and an -orientation for H-3, H-5, H-9, 23-
CH , 27-CH , as well as a -orientation for 24-CH OH, 25-CH , and
6-CH . On the other hand, -orientation of H-12 was proved by the
NOE correlation between H-11 and H-12 and the coupling con-
stant between H-11 and H-12 (near to zero). The NOE correlation
between H -27 and H -16 was clearly observed, which suggested
the conformation of D ring was semichair. The geometric configu-
b
-hydroxyl-24-hydroxymethyl. Meanwhile, NOE
-23; H-5 and H-9; H-9 and
-26, H-11 ; H -26 and H-11
.33, d (12.5)
3
2
2
2
3
d
H
3
2
3
3
3
b
3
b
1.48, s
1.83, d (6.0)
a
3
3
b
2
3
2
3
b
carbon signals. In addition to the carbon signals represented by the
b
above mentioned sugars, the other 30 ones indicated 1 was a tri-
1
13
terpenoid saponin. The analysis of the H, C and HSQC NMR data
suggested the presences of five methyl groups [ 0.70, 0.91, 1.48,
.48, 1.74 (3H each, all s, H -25, 26, 23, 30, 27)], one oxygenated
methylene proton [ -24],
3.32, 4.32 (1H each, both d, J¼12.0 Hz), H
and two oxygenated methine protons [
3
2
d
D
D
1
3
rations for olefinic bonds of 13(18)E and 17(28)Z were certified by
NOE correlations found between H-12 and H -19; H -16 and H-28,
respectively. Meanwhile, a -orientation for 30-CH was indicated
by the NOE correlations between H-12 and H -30, H -19; H -30 and
H-19 . Then, the relative configuration of 1 was assigned un-
ambiguously (Fig. 3). The absolute configuration,
d
2
2
2
d
3.40 (1H, dd, J¼4.5, 11.5 Hz,
b
3
H-3), 5.33 (1H, br s, H-12)], together with nine methylenes, three
3
2
3
methines, and ten quaternary carbons for the aglycon of 1. And the
b
presences of
were further certified by NMR data [
C-13), 159.5 (C-17), 130.7 (C-18), 126.7 (C-28)]. The He H COSY
a,
b,
g,d
-unsaturated carboxyl group and olefinic bond
D
D
d
H
6.01 (1H, s, H-28); 150.7
d
C
3S,4S,5R,8R,9R,10R,12S, 13(18)E,14S, 17(28)Z,20S of it was de-
1
1
(
termined by comparing the experimental with the computational

7010
Y. Zhang et al. / Tetrahedron 72 (2016) 7008e7013
1H, 13C NMR (DMSO-d
, Table 2) and 1He H COSY, HSQC, HMBC,
1
6
and NOESY spectra indicated the aglycon of 4 was similar to that of
13
1
e3. According to the main C NMR data differences between 4
and 1e3 aglycons (Table 2, in DMSO-d ) displayed at C-8/C-9/C-11/
6
C-12/C-18/C-27 taking C ring as the centre, we could deduce that
the methine at C-12 was replaced by methylene. The correlations
1
1
observed in He H COSY between
H-9), 2.15, 2.64 (H -12), and in HMBC from
126.7 (C-18), 150.3 (C-13) confirmed the correctness of the above
d
H
1.26, 1.68 (H
2
-11) and
d
H
1.56
(
d
2
d
H
2.15, 2.64 (H
2
-12) to
C
conjecture. The NOE correlations in NOESY (Fig. 3) spectra and the
Cotton effect showed in CD spectra proved the absolute configu-
D
D
Fig. 3. Key NOESY correlations for aglycons of 1 and 4.
ration in 4 was 3S,4S,5R,8R,9R,10R, 13(18)Z,14S, 17(28)Z,20S.
Liver and skeleton muscles are major organs in energy metab-
olism. In insulin resistance condition, insulin action is delayed and
the ability of insulin to stimulate glucose uptake is blunted in
skeleton muscle, insulin fails to suppress blood sugar production by
the liver while allowing the production of hepatic triglycerides.
This combination results in high blood sugar and fatty liver dis-
electronic circular dichroism (ECD) spectra (Fig. 4).6,7 Furthermore,
the connectivity of oligoglycoside moiety to the aglycon part was
characterized by HMBC experiment on 1, in which long-range
0
00
0
000
correlations were found from H-1 to C-3, H-1 to C-2 , H-1 to
0
0
C-2 .
The molecular formula for astraisoolesaponins A
8
ease. Compounds 1e4 showed significant inhibitory effects on
oleic acid induced triglyceride accumulation in HepG2 cells, espe-
cially 3 and 4, suppressing 20% intracellular triglyceride levels at
M (Fig. 5). Moreover, compared with control group, compounds
and 4 significantly increased glucose uptake on L6 cell, but 1 and 2
had no effect on it (Fig. 6).
2
(2) and A
3
(3)
17Cl,
16 {m/z 843.3562 [MþCl] (calcd for
41 60
H O16Cl, 843.3575)} obtained from HRESI-TOF-MS, re-
e
were C42
H
62
O
17 {m/z 873.3651 [MþCl] (calcd for C42
62
H O
e
8
C
60
73.3681)} and C41H O
1
3
0
m
1
13
spectively. The UV, CD, ECD, IR, H, C NMR (DMSO-d
6
, Table 2) and
2
D-NMR spectra (Fig. 2) suggested the aglycon of them was the
same as that of 1. Sugar analysis result indicated D-glucuronic acid
and
D
-glucose presented in 2, meanwhile D-glucuronic acid and D-
3. Conclusion
2
xylose presented in 3. The linkage positions of sugar parts with
aglycon were elucidated by the HMBC data, which showed long-
In summary, our current study describes the discovery of four
novel saponins, astraisoolesaponins A eA (1e3) and B (4) from
1 3
0
range correlations from
d
H
4.45 (d, J¼7.5 Hz, H-1 ) to
d
C
89.5 (C-
4.40 (d,
79.4
00
0
3
);
d
H
4.62 (d, J¼7.5 Hz, H-1 ) to
d
C
79.9 (C-2 ) for 2, and
d
H
AM, which possess a 28-methyl-rearranged carbon skeleton. Their
absolute configurations were determined by ECD calculations.
Moreover, compounds 3 and 4 displayed TG inhibitory effects in
HepG2 cells and glucose uptake increase effects on L6 cells. The
results of the present study indicated that 3 and 4 could be possible
lead candidates for improvement of insulin resistance.
0
00
J¼8.0 Hz, H-1 ) to
d
C
89.4 (C-3);
d
H
4.49 (d, J¼7.5 Hz, H-1 ) to
d
C
0
(
C-2 ) for 3.
The molecular formula of astraisoolesaponin B (4) was found to
be C42 16 determined by negative-ion HRESI-TOF-MS (m/z
21.3987 [MeH] , calcd for C42
62
H O
e
8
61
H O16, 821.3965). The UV, IR, CD,
Fig. 4. Calculated (both the isolates and their enantiomers) and experimental ECD spectra of 1e4.

Y. Zhang et al. / Tetrahedron 72 (2016) 7008e7013
7011
Table 2
spectrophotometer. IR spectra were obtained on a Varian 640-IR
FTIR spectrophotometer. Optical rotations were determined on
a Rudolph Autopol IV automatic polarimeter. NMR spectra were
1
3C NMR data of 1e4
No. 1a
1b
2b
3b
4b
No. 1a
1b
2b
3b
4b
Ò
measured on a Bruker AVANCE III 500 MHz and/or 400 MHz NMR
spectrometer with TMS as an internal standard. Negative-ion
HRESI-TOF-MS were determined on an Agilent 6520 Accurate-
Mass Q-Tof MS spectrometer. Column chromatographies (CC)
were performed on macroporous resin D101 (Haiguang Chemical
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
38.6 37.9 38.0 38.3 38.0 25 16.3 15.6 15.6 15.5 15.7
26.6 25.6 25.6 25.6 25.6 26 17.6 16.9 16.9 17.0 16.7
91.5 90.4 89.5 89.4 89.5 27 22.9 21.7 21.7 21.7 20.1
43.8 42.8 42.9 43.2 42.9 28 126.7 125.3 125.3 125.3 124.3
56.4 55.3 55.4 55.7 55.1 29 179.0 177.1 177.0 177.0 177.0
18.5 17.7 17.9 18.1 17.8 30 25.5 24.5 24.5 24.5 24.4
0
0
34.4 33.5 33.5 33.6 33.3
41.4 44.1 44.1 44.1 40.7
45.3 44.1 44.2 44.2 49.1
36.6 35.7 35.8 35.9 36.2
30.5 29.6 29.5 29.6 21.6
65.6 64.1 64.1 64.1 25.6
150.7 149.8 149.7 149.8 150.3
45.8 44.9 44.9 44.9 45.5
32.7 31.8 31.8 31.8 30.8
31.5 30.3 30.3 30.3 30.6
159.5 159.4 159.3 159.3 159.8
130.7 129.2 129.2 129.1 126.7
40.9 40.4 39.4 39.4 39.7
45.6 44.1 44.1 44.1 44.4
53.3 52.2 52.1 52.2 52.0
200.0 198.7 198.7 198.7 198.7
22.8 22.0 21.9 21.9 21.9
63.4 62.0 62.0 61.5 62.0
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
105.3 103.7 103.0 103.3 103.0
78.5 76.4 79.9 79.4 80.0
78.6 76.2 76.4 76.2 76.4
73.8 71.8 71.2 71.4 71.3
77.3 74.9 75.1 75.1 75.0
173.1 170.5 170.4 170.3 170.6
102.0 99.9 103.0 103.7 103.0
77.8 76.7 74.0 74.0 74.1
79.2 77.4 76.4 76.4 76.4
69.8 68.7 68.7 69.1 68.7
78.0 76.4 76.7 65.6 76.7
Co., Ltd., Tianjin, China), Silica gel (48e75
mm, Qingdao Haiyang
Chemical Co., Ltd., Qingdao, China), ODS (40e63
mm, YMC Co., Ltd.,
Tokyo, Japan), and Sephadex LH-20 (Ge Healthcare Bio-Sciences,
Uppsala, Sweden), and Preparative HPLC (PHPLC) column, Cosmo-
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
sil 5C18-MS-II (20 mm i.d.ꢂ250 mm, 5
mM, Nakalai Tesque, Inc.,
Tokyo, Japan) were used to purify the constituents.
0
0
0
0
0
0
00
00
00
00
00
00
4.2. Plant material
61.4 60.0 60.2
102.0 100.0
72.4 70.5
60.2
The stems of A. membranaceus (Fisch.) Bge. var. mongholicus
(Bge.) Hsiao. were collected from Gansu province, China, and
identified by Dr. Li Tianxiang. The voucher specimen was deposited
at the Academy of Traditional Chinese Medicine of Tianjin Univer-
sity of TCM.
72.7 70.5
74.4 72.2
69.5 67.9
19.0 17.8
a
4.3. Extraction and isolation
Determined in C
Determined in DMSO-d
5
D
5
N.
b
; 1 , 2 e4 were measured at 125 MHz and 1b at
a
b
b
6
125 MHz.
The stems of AM (8.0 kg) were cut and refluxed with 70% etha-
nolewater. Then, the 70% ethanolewater extract (580 g) was
partitioned in a CHCl
O (390 g) layers were given. Then the H
subjected to D101 macroporous resin CC (H
a result, H O (179 g) and 95% EtOH (116 g) eluted fractions were ob-
tained. The 95% EtOH eluate (90.6 g) was subjected to [CHCl eMeOH
100:5, v/v)/CHCl eMeOHeH (10:3:1/7:3:1/6:4:1, v/v/v,
lower layer)/MeOH] to yield fourteen fractions (Fr.1e14). Fraction 9
7.0 g) was isolated by ODS CC [MeOHeH (10:90/
0:80/30:70/40:60/50:50/70:30/80:20/100:0, v/v)] to
afford twenty fractions (Fr. 9-1e9e20). Fraction 9e12 was separated
by PHPLC [MeOHeH O (40:60, v/v)] to obtain six fractions (Fr. 9e12-
e9e12-6). Fractions 9e12-4 (56.5 mg) and 9e12-5 (81.0 mg) were
further purified by PHPLC [CH CNeH
O (24:76, v/v)þ1% HAc] to give
astraisoolesaponins A (2, 35.7 mg) and A (3, 10.7 mg), respectively.
Fraction 9e14 (734.0mg) wasisolated by PHPLC [MeOHeH O (55:45,
v/v)þ1% HAc] to yield seven fractions (Fr. 9e14-1e9e14-7). Fraction
e14-3 (47.2 mg) was subjected to [CHCl eMeOHeH
10:3:1/7:3:1, v/v/v, lower layer)], and three fractions (Fr. 9e14-3-
e9e14-3-3) were obtained. Fraction 9e14-3-2 (54.2 mg) was puri-
fied by PHPLC [CH CNeH
O (26:74, v/v)þ1% HAc] to afford astrai-
soolesaponin B (4, 13.7 mg). Fraction 12 (8.2 g) was separated by
PHPLC [CH CNeH
O (30:70/40:60/50:50/60:40/80:20, v/v)þ
% HAc] to give nineteen fractions (Fr. 12-1e12e19). Fraction 12e8
687.8mg)wasisolated byPHPLC [MeOHeH
O(32:68, v/v)þ1%HAc],
to gain six fractons (Fr.12e8-1e12e8-6). Fraction 12e8-6 (237.8 mg)
was subjected to SiO gel CC [CHCl eMeOHeH O (7:3:1/6:4:1, v/v/
3
eH
2
O mixture (1:1, v/v), and CHCl
O layer (300 g) was
O/95% EtOH). As
3
(190 g) and
H
2
2
2
2
3
(
3
2
O
(
2
2
O
2
1
Fig. 5. Inhibitory effects of compounds 1e4 on oleic acid induced triglyceride accu-
mulation in HepG2 cells. N: Normal group; C: Control group; ORLI: Orlistat 0.5
treated group; 1e4: 1e4 10
M treated group. Values represent the meanꢃSD of six
determinations. *P<0.05; **P<0.01, ***P<0.001 versus control group. N¼6.
3
2
mM
2
3
m
2
9
(
1
3
2
O
3
2
3
2
1
(
2
2
3
2
v, lower layer)], and three fractions (Fr. 12e8-6-1e12e8-6-3) were
obtained. Fraction 12e8-6-3 (123.6 mg) was further purified by
PHPLC [MeOHeH
2
O (23:77, v/v)þ1% HAc] to afford astraisoolesapo-
Fig. 6. Effects of compounds 1e4 on glucose uptake in L6 cells. N: Normal group; P:
Cytochalasin B 10 M treated group; Ins: insulin 10 g/mL treated group. 1e4: 1e4
treated group in different dose. Values represent the meanꢃSD of six determinations.
P<0.05 versus control group. N¼6.
nin A (1, 31.5 mg).
1
m
m
2
5
ꢁ
4
.3.1. Astraisoolesaponin
max (KBr) cm : 3400, 2928, 2858, 1719, 1642,
max (MeOH) nm (log ):
nm): þ0.49 (211), e0.51
): 0.92, 1.64 (1H
-1), 1.70, 1.91 (1H each, both m, H -2), 3.26 (1H, dd,
J¼4.0, 12.4 Hz, H-3), 0.93 (1H, dd, J¼2.8, 11.6 Hz, H-5), 1.24, 1.58 (1H
each, both m, H -6), 1.31, 1.43 (1H each, both m, H -7), 1.76 (1H, dd,
A
1
(1). White powder.
[a]
D
ꢀ73.5
*
ꢀ
1
(
1
c¼0.65, MeOH); IR
n
4
4
. Experimental
455, 1405, 1269, 1170, 1128, 1073, 1041; UV
l
3
294 (4.08). CD (CH
3
CN, c¼0.0011) mdeg (
l
1
.1. General
(
294), þ0.25 (356). H NMR (400 MHz, DMSO-d
6
d
each, both m, H
2
2
The following instruments were used to obtain physical data:
UV spectra were recorded on
a Varian Cary 50 UVevis
2
2

7012
Y. Zhang et al. / Tetrahedron 72 (2016) 7008e7013
J¼6.0,10.4 Hz, H-9),1.46,1.64 (1H each, both m, H
H-12), 1.34, 1.75 (1H each, both m, H -15), [2.17 (1H, br d, ca.
J¼12 Hz), 2.45 (1H, m), H -16], 2.35, 3.01 (1H each, both d, J¼8.0 Hz,
-19), 2.33, 2.62 (1H each, both d, J¼12.0 Hz, H -21), 1.11 (3H, s, H
3), 3.10, 3.89 (1H each, both d, J¼10.8 Hz, H -24), 0.74 (3H, s, H -25),
.77 (3H, s, H -26), 1.38 (3H, s, H
2
-11), 4.74 (1H, br s,
(CH
(375). H NMR (500 MHz, DMSO-d
-1), 1.70, 1.90 (1H each, both m, H
3
CN, c¼0.0013) mdeg (
l
nm): þ0.27 (240), e0.63 (298), þ0.17
1
2
6
): 1.00, 1.64 (1H each, both m,
d
2
H
2
2
-2), 3.27 (1H, dd, J¼5.0, 11.0 Hz,
H
2
0
2
2
3
-
H-3), 0.90 (1H, dd, J¼3.0,11.0 Hz, H-5),1.35,1.56 (1H each, both m, H
2
-
2
3
6),1.33,1.38 (1H each, both m, H
1.26, 1.68 (1H each, both m, H
2
-7),1.56 (1H, dd, J¼6.0,11.0 Hz, H-9),
3
3
-27), 5.65 (1H, s, H-28), 1.14 (3H, s,
2
-11), 2.15, 2.64 (1H each, both m, H
2
-
0
0
0
0
H
3
-30), 4.27 (1H, d, J¼6.4 Hz, H-1 ), 3.44 (2H, m, H-2 and 3 ), 3.22
12), 1.40, 1.71 (1H each, both m, H
2.33 (1H, m), H
-16], 2.38 2.75 (1H each, both d, J¼14.0 Hz, H
2.34, 2.60 (1H each, both d, J¼12.0 Hz, H -21),1.10 (3H, s, H -23), 3.15,
-24), 0.74 (3H, s, H -25), 0.80 (3H,
2
-15), [2.16 (1H, br d, ca. J¼13 Hz),
0
(
1H, dd, J¼8.0, 9.2 Hz, H-4 ), 3.26 (1H, d, J¼9.2 Hz, H-5 ), 4.81 (1H, d,
2
2
-19),
0
0
00
J¼7.6 Hz, H-1 ), 3.17 (1H, dd, J¼7.6, 8.0 Hz, H-2 ), 3.29 (1H, t,
2
3
0
0
00
J¼8.0 Hz, H-3 ), 3.19 (1H, dd, J¼8.0, 10.4 Hz, H-4 ), 3.01 (1H, m, H-
3.87 (1H each, both d, J¼10.5 Hz, H
2
3
0
0
00
5
), [3.45 (1H, dd, J¼5.6, 11.6 Hz), 3.61 (1H, br d, ca. J¼12 Hz), H
2
-6 ],
s, H
3
-26),1.19 (3H, s, H
3
-27), 5.59(1H, s, H-28),1.20 (3H, s, H
3
-30), 4.42
0
00
000
0
0
5
.03 (1H, br s, H-1 ), 3.68 (1H, br d, ca. J¼4 Hz, H-2 ), 3.54 (1H, dd,
(1H, d, J¼8.0 Hz, H-1 ), 3.35 (1H, dd, J¼8.0, 9.0 Hz, H-2 ), 3.45 (1H, t,
0
00
000
000
0
0
0
J¼4.0, 8.0 Hz, H-3 ), 3.42 (1H, t, J¼8.0 Hz, H-4 ), 3.98 (1H, m, H-5 ),
J¼9.0 Hz, H-3 ), 3.32(1H, t, J¼9.0 Hz, H-4 ), 3.60(1H, d, J¼9.0 Hz, H-5 ),
0
00 13
00
00
1
.12 (3H, d, J¼6.0 Hz, H
3
-6 ); C NMR (100 MHz, DMSO-d
6
) spec-
4.60 (1H, d, J¼7.5 Hz, H-1 ), 2.97 (1H, dd, J¼7.5, 9.0 Hz, H-2 ), 3.14 (2H,
0
0
00
00
troscopy data, see Table 2. HRESI-TOF-MS: Negative-ion mode m/z
m, H-3 and 4 ), 3.06 (1H, m, H-5 ), [3.51 (1H, dd, J¼4.0,12.0 Hz), 3.61
e
00 13
2 6
-6 ]; C NMR (125 MHz, DMSO-d ) spec-
1019.4238 [MþCl] (calcd for C48
72
H O21Cl 1019.4260).
(1H, br d, ca. J¼12 Hz), H
troscopy data, see Table 2. HRESI-TOF-MS: Negative-ion mode m/z
2
5
ꢀ31.1ꢁ
e
4
(
1
3
.3.2. Astraisoolesaponin (2). White powder.
A
2
[
a]
D
821.3987 [MꢀjꢀH] (calcd for C42
61
H O16 821.3965).
ꢀ
1
c¼0.84, MeOH); IR
458, 1402, 1276, 1199, 1168, 1074, 1039; UV
): 293 (3.89). CD (CH
CN, c¼0.0019) mdeg (
e4.69 (292), þ0.92 (365). H NMR (500 MHz, DMSO-d
1H each, both m, H -1), 1.70, 1.87 (1H each, both m, H
dd, J¼4.5, 11.5 Hz, H-3), 0.91 (1H, dd, J¼3.0, 11.0 Hz, H-5), 1.32, 1.57
1H each, both m, H -6), 1.33, 1.42 (1H each, both m, H -7), 1.76 (1H,
dd, J¼6.0, 9.0 Hz, H-9), 1.48, 1.64 (1H each, both m, H -11), 4.74 (1H,
br s, H-12), 1.34, 1.75 (1H each, both m, H -15), [2.17 (1H, br d, ca.
J¼13 Hz), 2.46 (1H, m), H -16], 2.38, 2.98 (1H each, both d,
nmax (KBr) cm : 3339, 2924, 2855, 1715, 1637,
l
max (MeOH) nm (log
nm): þ2.42 (236),
): 0.92,1.64
-2), 3.31 (1H,
4.4. Acid hydrolysis of 1e4
3
l
1
6
d
Solution of new compounds 1e4 (each 2.0 mg) in 5% aqueous
H SO -1,4-dioxanewere treatedbyusingthe knownmethod: heated
2 4
2
(
2
2
under reflux for 1 h, respectively, neutralized with Amberlite IRA-400
e
(
2
2
(OH form), removed by filtration, dealed with ODS column
2
2
(H O/MeOH), reacted with L-cysteine methyl ester hydrochloridein
2
pyridine and N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA), suc-
cessively. Finally, the reaction product was determined by GCeMS
analysis (GC conditions, column: RESTEK Rxi-5ms, 30 mꢂ0.25 mm
2
J¼10.0 Hz, H
2
-19), 2.33, 2.63 (1H each, both d, J¼12.0 Hz, H
2
-21),
-24),
ꢁ
1.11 (3H, s, H
3
-23), 3.15, 3.89 (1H each, both d, J¼11.0 Hz, H
2
(i.d.) capillary column; column temperature: 230 C; carrier gas: He.).
0
.76 (3H, s, H
3
-25), 0.77 (3H, s, H
3
-26), 1.38 (3H, s, H
3
-27), 5.65 (1H,
As result, the identification for derivatives of
glucose, and -rhamnose from 1; -glucuronic acid and
2 and 4; -glucuronic acid and -xylose from 3 present in the super-
D-glucuronic acid,
D-
0
s, H-28),1.15 (3H, s, H
3
-30), 4.45 (1H, d, J¼7.5 Hz, H-1 ), 3.37 (1H, dd,
L
D
D-glucosefrom
0
0
J¼7.5, 9.0 Hz, H-2 ), 3.47 (1H, t, J¼9.0 Hz, H-3 ), 3.35 (1H, dd, J¼9.0,
D
D
0
0
9
1
3
.5 Hz, H-4 ), 3.65 (1H, d, J¼9.5 Hz, H-5 ), 4.62 (1H, d, J¼7.5 Hz, H-
natant were performed by comparison of their retention times with
the hydrolysate with their authentic samples treated in the same way,
0
0
00
00
), 2.98 (1H, dd, J¼7.5, 9.0 Hz, H-2 ), 3.14 (1H, t, J¼9.0 Hz, H-3 ),
0
0
00
0
.15 (1H, t, J¼9.0 Hz, H-4 ), 3.07 (1H, m, H-5 ), [3.52 (1H, dd, J¼4.0,
R
t : (i) 13.3 min (D-glucuronic acid), (ii) 11.4 min (
D-glucose), (iii)
0
13
1
1.5 Hz), 3.62 (1H, br d, ca. J¼12 Hz), H
2
-6 ]; C NMR (125 MHz,
7.7 min ( -rhamnose), and (IV) 6.4 min (D
L
-xylose).
DMSO-d
6
) spectroscopy data, see Table 2. HRESI-TOF-MS: Negative-
e
ion mode m/z 873.3651 [MþCl] (calcd for C42
62
H O
17Cl 873.3681).
4.5. Computations
2
5
ꢀ51.2ꢁ
4
(
1
(
(
.3.3. Astraisoolesaponin (3). White powder.
A
3
[
a]
D
The ECD spectra for the optimized conformers were calculated
at the CAM-B3LYP/SVP level with a CPCM solvent model in aceto-
nitrile, and the calculated ECD spectra of different conformers were
simulated with a half bandwidth of 0.3e0.4 eV. The ECD curves
were extracted by SpecDis 1.62 software. The overall ECD curves of
all the compounds were weighed by Boltzmann distribution after
UV correction.
ꢀ1
c¼0.13, MeOH); IR
460, 1401, 1170, 1046; UV
CH
CN, c¼0.0014) mdeg (
370). H NMR (500 MHz, DMSO-d
-1), 1.74, 1.83 (1H each, both m, H
H-3), 0.88 (1H, dd, J¼3.0, 11.5 Hz, H-5), 1.35, 1.58 (1H each, both m,
-6), 1.33, 1.43 (1H each, both m, H
-7), 1.75 (1H, dd, J¼6.0, 11.0 Hz,
H-9), 1.46, 1.65 (1H each, both m, H -11), 4.74 (1H, br s, H-12), 1.37
.74 (1H each, both m, H
-15), [2.16 (1H, br d, ca. J¼13 Hz), 2.46 (1H,
m), H -19), 2.33, 2.63
-16], 2.39 2.98 (1H each, both d, J¼9.5 Hz, H
1H each, both d, J¼12.0 Hz, H -21), 1.08 (3H, s, H -23), 3.17, 3.76
1H each, both d, J¼9.0 Hz, H -24), 0.77 (3H, s, H -25), 0.78 (3H, s,
-26), 1.37 (3H, s, H -27), 5.65 (1H, s, H-28), 1.15 (3H, s, H -30),
.40 (1H, d, J¼8.0 Hz, H-1 ), 3.33 (1H, dd, J¼8.0, 9.0 Hz, H-2 ), 3.41
nmax (KBr) cm : 3430, 2924, 2854, 1719, 1631,
l
max (MeOH) nm (log
nm): þ2.25 (236), e1.84 (294), þ0.51
): 0.93, 1.64 (1H each, both m,
-2), 3.21 (1H, dd, J¼5.0, 11.0 Hz,
3
3
l
1
6
d
H
2
2
H
2
2
2
4.6. Bioassay
1
2
2
2
4.6.1. Inhibitory effects on oleic acid induced triglyceride accumula-
tion in HepG2 cells. The hepatic cell line HepG2 (IBMS, CAMS/
PUMC, Beijing China) were maintained in high glucose Minimum
Essential Medium (MEM) supplemented with 10% fetal bovine se-
rum (FBS) and 1% penicillin-streptomycin under a humidified at-
(
(
H
4
(
2
3
2
3
3
3
3
0
0
0
0
1H, t, J¼9.0 Hz, H-3 ), 3.33 (1H, m, overlapped, H-4 ), 3.62 (1H, m,
2
mosphere of 5% CO in air. After growth to 80% confluence, cells
0
00
4
overlapped, H-5 ), 4.49 (1H, d, J¼7.5 Hz, H-1 ), 2.97 (1H, dd, J¼7.5,
were seeded at 1ꢂ10 cells/mL on 48-well dish. After 24 h in-
cubation, the medium was switched to high glucose MEM and
supplemented with 10% FBS and 0.2 mM oleic acid sodium salt,
together with sample DMSO solution (final concentration of DMSO
was less than 0.1%). After 48 h incubation, the amount of in-
tracellular triglycerides was determined with the Triglycerides kit
0
0
00
00
9
.0 Hz, H-2 ), 3.08 (1H, t, J¼9.0 Hz, H-3 ), 3.25 (1H, m, H-4 ), [2.98
13
00
(
(
1H, t, J¼11.0 Hz), 3.62 (1H, m, overlapped), H
2
-5 ]; C NMR
125 MHz, DMSO-d ) spectroscopy data, see Table 2. HRESI-TOF-
MS: Negative-ion mode m/z 843.3562 [MþCl] (calcd for
16Cl 843.3575).
6
e
41 60
C H O
(
BioSino Bio-technology and Science Inc., China) after cell lysis.
2
5
ꢁ
4
.3.4. Astraisoolesaponin B (4). White powder. [
a
]
D
ꢀ92.1 (c¼0.84,
ꢀ
1
MeOH); IR
n
max (KBr) cm : 3339, 2933,1715,1635,1604, 1404, 1381,
4.6.2. Increase effects on glucose uptake in L6 cells. Differentiated L6
mature myotubes were cultured on 48-well plates (with 2ꢂ10 cells/
4
1270, 1192, 1069, 1040; UV
lmax (MeOH) nm (log
3

Y. Zhang et al. / Tetrahedron 72 (2016) 7008e7013
7013
mL), incubated in glucose and serum-free HEPES-buffered saline
HBS) for 2 h. The medium was switched to HBS containing 1 mg/mL
Supplementary data
(
glucose, together with sample DMSO solution (final concentration of
DMSO was less than 0.1%). After 4 h incubation, the amount of
glucose in culture supernatant was determined with the glucose kit
Supplementary data (NMR spectroscopic data for all new com-
pounds) associated with this article can be found in the online
version, at http://dx.doi.org/10.1016/j.tet.2016.09.034.
(
BioSino Bio-technology and Science Inc., China). Insulin (10
mg/mL)
was used as positive control and Cytochalasin B (10
transporter inhibitor, was used as negative control.
mM), a glucose
References and notes
1
. Li, X.; Qu, L.; Dong, Y.; Han, L.; Liu, E.; Fang, S.; Zhang, Y.; Wang, T. Molecules 2014,
19, 18850e18880.
4
.6.3. Statistical analysis. Values are expressed as meanꢃS.D. All
2
3
. Zhang, Y.; Li, X.; Ruan, J.; Wang, T.; Dong, Y.; Hao, J.; Liu, E.; Han, L.; Gao, X.;
Wang, T. Fitoterapia 2016, 109, 99e105.
. Hao, J.; Li, J.; Li, X.; Liu, Y.; Ruan, J.; Dong, Y.; Zhang, Y.; Wang, T. Molecules 2016,
21, http://dx.doi.org/10.3390/molecules21030354 pii: E354.
the grouped data were statistically performed with SPSS 11.0. Sig-
nificant differences between means were evaluated by one-way
analysis of variance (ANOVA) and Tukey’s Studentized range test
was used for post hoc evaluations. P<0.05 was considered to in-
dicate statistical significance.
4. Wang, T.; Ruan, J.; Li, X.; Chao, L.; Shi, P.; Han, L.; Zhang, Y.; Wang, T. J. Nat. Med.
016, 70, 198e206.
. Tsunoda, Y.; Okawa, M.; Kinjo, J.; Ikeda, T.; Nohara, T. Chem. Pharm. Bull. 2008, 56,
138e1142.
6. Li, X. C.; Ferreira, D.; Ding, Y. Curr. Org. Chem. 2010, 14, 1678e1697.
. Mazzeo, G.; Santoro, E.; Andolfi, A.; Cimmino, A.; Troselj, P.; Petrovic, A. G.;
2
5
1
Acknowledgements
7
Superchi, S.; Evidente, A.; Berova, N. J. Nat. Prod. 2013, 76, 588e599.
. Korenblat, K. M.; Fabbrini, E.; Mohammed, B. S.; Klein, S. Gastroenterol. 2008, 134,
1369e1375.
This research was supported by Projects NCET-12-1069 and
TD12-5033.
8