Dammarane glycosides from the root of Machilus yaoshansis

Article abstract of DOI:10.1021/np300310a

Nine new dammarane triterpene glycosides (1-3 and 8-13) and 12 known analogues have been isolated from an ethanol extract of the roots of Machilus yaoshansis. Compounds 1-7 have an uncommon 20,23-dihydroxydammar-24-en-21-oic acid-21,23-lactone moiety that

Full text of DOI:10.1021/np300310a

Article
pubs.acs.org/jnp
Dammarane Glycosides from the Root of Machilus yaoshansis
Maoluo Gan,†,‡,§ Mingtao Liu,†,§ Lishe Gan,⊥ Sheng Lin,*,† Bo Liu,† Yanling Zhang,† Jiachen Zi,†
Weixia Song,† and Jiangong Shi*,†
†State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica, Chinese Academy of
Medical Sciences and Peking Union Medical College, Beijing 100050, People’s Republic of China
‡Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050,
People’s Republic of China
⊥Institute of Modern Chinese Medicine, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058,
People’s Republic of China
S
* Supporting Information
ABSTRACT: Nine new dammarane triterpene glycosides
(1−3 and 8−13) and 12 known analogues have been isolated
from an ethanol extract of the roots of Machilus yaoshansis.
Compounds 1−7 have an uncommon 20,23-dihydroxydam-
mar-24-en-21-oic acid-21,23-lactone moiety that was previ-
ously reported in compounds isolated from Gynostemma
pentaphyllum. The configurations of the lactone moieties in 1−
3 were determined by comparison of the experimental ECD
spectra of 1−3 and the hydrolysates, 1a and 1b, with the
corresponding calculated ECD spectra. On the basis of NMR and ECD data analysis of 1−7, the previously reported C-20 and
C-23 configurations of 4−7 and related derivatives from Gynostemma pentaphyllum were revised. In addition, the application of
NMR data and Cotton effects to the determination of the relative and absolute configurations of the γ-lactone moiety in
3β,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone derivatives is discussed.
G. pentaphyllum were revised. In addition, the application of
NMR data and Cotton effects to the determination of the
relative and absolute configurations of the γ-lactone moiety in
3β,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone
derivatives is discussed.
pecies of the genus Machilus have long been used for the
Streatment of edema, abdominal distension, pain, and
inflammation in China and Southeast Asia.1 As part of a
program to assess the chemical and biological diversity of
Machilus species,2 we focused our study on Machilus yaoshansis
S. Lee et F. N. Wei, a plant that is widely distributed in the
south of China and used as a folk medicine by the ethnic
Zhuang in Guangxi Province for the treatment of rheumatism.
Our previous studies on the bark3 and the root4 of
M. yaoshansis led to the isolation and characterization of
several unusual cucurbitane derivatives and spirolactones with
cytotoxic activities. Continuing examination of the root extract
has resulted in the characterization of nine new (1−3 and
8−13) and 12 known dammarane glycosides. Herein, we discuss
the detailed structural determination of the isolates by extensive
spectroscopic analysis, including 1D and 2D NMR and
electronic circular dichroism (ECD). Compounds 1−7 possess
an unusual 20,23-dihydroxydammar-24-en-21-oic acid-21,23-
lactone moiety that was previously reported in compounds
isolated from Gynostemma pentaphyllum. The C-20 and C-23
configurations of the 20,23-dihydroxydammar-24-en-21-oic
acid-21,23-lactone moieties of 1−3 were established by
comparison of the ECD spectra calculated using the time-
dependent density functional theory (TDDFT) with the
experimental ECD spectra. On the basis of detailed NMR
and ECD data analysis of 1−7, the previously reported C-20
and C-23 configurations of 4−7 and related derivatives from
RESULTS AND DISCUSSION
■
Compound 1 was obtained as an amorphous powder, and its
molecular formula was determined to be C46H74O16 by
positive-ion HRESIMS data at m/z 905.4832 [M + Na]+
(calcd for C46H74O16Na, 905.4874), combined with the NMR
data (Tables 1 and 3). The IR spectrum suggested the presence
of OH (3383 cm−1) and γ-lactone (1760 cm−1) functionalities.
The 1H NMR spectrum showed resonances assignable to seven
tertiary methyl groups between δH 0.76 and 1.69 and five
deshielded methines (anomeric, olefinic, and oxygenated) at δH
6.13 (brs, H-1″), 5.59 (d, J = 8.8 Hz, H-24), 5.47 (ddd, J = 8.8,
7.6, 6.0 Hz, H-23), 4.92 (d, J = 5.2 Hz, H-1′), and 5.02 (d, J =
1
7.2 Hz, H-1″′). The H NMR spectrum also showed partially
overlapped signals due to oxymethine and oxymethylene
protons between δH 3.65 and 4.73. 13C NMR and DEPT
spectra showed 46 carbon resonances, of which three were
attributed to anomeric carbons [δC 105.2 (C-1″′), 104.8 (C-1′),
and 102.1 (C-1″)], two were attributed to a trisubstituted
Received: May 1, 2012
Published: July 10, 2012
© 2012 American Chemical Society and
American Society of Pharmacognosy
1373
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-
α-L-arabinopyranoside.
The configuration of the tetracyclic nucleus in 1 was
proposed to be identical to that in the natural dammarane
derivatives.6 The chemical shift and splitting pattern of H-3 (δH
1
3.29, dd, J = 12.0 and 4.0 Hz) in the H NMR spectrum
indicated that it was axially α-oriented.6a The configuration of
the γ-lactone in the side chain was established by analyses of the
NOESY and ECD data of 1 and its hydrolysates (1a and 1b), in
combination with the theoretical ECD spectra based on TDDFT,
a powerful tool for the configuration assignment of natural
products.7 In the NOESY spectrum, correlations of H-23/H-17
and H-22a revealed that these protons were cofacial on the
γ-lactone ring, while the correlation of H-24/H-22b indicated that
these protons were cofacial on the opposite side of the ring.
Acid hydrolysis of 1 with 1 M HCl at 60 °C produced two
isomers, 1a and 1b. The NOESY spectrum of 1a showed
correlations of H-23/H-17 and H-22a and of H-22b/H-24 and
OH-20. In contrast, 1b did not give the corresponding
correlations in its NOESY spectrum (Supporting Information,
Figures S20 and S26). This indicated that 1a had the same
configuration as 1, but 1b was an epimer of 1a with the
opposite configuration at C-20 or C-23. This was supported by
the similarity of the Cotton effects in the ECD spectra of 1 and
1a, while the ECD spectrum of 1b displayed Cotton effects
opposite those of 1 and 1a (Figure 1). The ECD spectra of 1a
and the (20R,23S)-isomer were calculated using TDDFT at the
B3LYP/6-311++G(2d,2p) level. The calculated ECD spectrum
of 1a matched the experimental spectra of 1a and 1. This indicated
the (20S,23R) configuration for 1 and 1a. Therefore, compound 1
was determined to be (3β,20S,23R)-3,20,23-trihydroxydammar-24-
en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-
[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranoside.
Compound 2 is an isomer of 1, as indicated by spectroscopic
data. Comparison of the NMR data of 2 and 1 indicated that H-
22a, H-22b, H-23, and H-24 in 2 were shifted by ΔδH −0.10,
−0.18, +0.24, and −0.18, respectively, while C-20, C-22, C-23,
and C-24 were shifted by ΔδC +2.1, −1.8, +1.1, and −1.4,
respectively. This suggested that 2 was the C-23 epimer of 1, as
supported by the ECD spectrum of 2, which displayed Cotton
effects opposite those of 1. Acid hydrolysis of 2 generated the
same compounds as those from 1, including the same sugars
and 1a and 1b. ECD calculations of 1b and the (20R,23R)-
isomer demonstrated that the calculated ECD curve for 1b was
consistent with the experimental ECD spectra of 1b and 2. This
confirmed that 2 was the C-23 epimer of 1. Therefore, compound
2 was determined to be (3β,20S,23S)-3,20,23-trihydroxydammar-
24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-
(1→2)]-[β-D-xylopyranosyl(1→3)]-α-L-arabinopyranoside.
double bond (δC 138.5 and 125.4), and one was attributed to a
lactone carbonyl carbon (δC 179.4). These spectroscopic data
suggested that 1 was a triglycosidic triterpenoid with γ-lactone
and olefinic subunits. The sugars obtained by acid hydrolysis of
1 were identified as L-arabinopyranose, D-xylopyranose, and
L-rhamnopyranose by GC analysis of the trimethylsilyl-L-cysteine
derivatives of the hydrolysate of 1 and the authentic sugars.5
The structure of 1 was finalized by analysis of the 2D NMR
data. The HSQC experiment allowed for the assignments of the
proton and protonated carbon resonances in the NMR spectra
1
of 1. In the H−1H COSY spectrum, the cross-peaks of H2-1/
H2-2/H-3; H-5/H2-6/H2-7; H-9/H2-11/H2-12/H-13/H-17/
H2-16/H2-15; and H2-22/H-23/H-24 demonstrated the
presence of vicinal coupling systems. Coupling constants of
the anomeric protons indicated α configurations for the
arabinopyranosyl (J1′,2′ = 5.2 Hz) and rhamnopyranosyl
(J1″,2″ ≈ 0 Hz) units and a β configuration for the xylopyranosyl
(J1″′,2″′ = 7.2 Hz) unit.6 In the HMBC spectrum, two- and three-
bond correlations of H3-18/C-7, C-8, C-9, and C-14; H3-19/C-1,
C-5, C-9, and C-10; H3-26, H3-27/C-24 and C-25; H3-28 and
H3-29/C-3, C-4, and C-5; H3-30/C-8, C-13, C-14, and C-15;
and H2-22/C-17, C-20, C-21, C-23, and C-24, together with
the chemical shifts of these protons and carbons and the
molecular formula, revealed that 1 had a 3,20,23-trihydrox-
ydammar-24-en-21-oic acid-21,23-lactone nucleus.6a,b In addi-
tion, HMBC correlations of H-1′/C-3, H-1″/C-2′, and H-1″′/
C-3′ demonstrated that the α-L-arabinopyranosyloxy unit was
located at C-3 and that the α-L-rhamnopyranosyloxy and
β-D-xylopyranosyloxy units were linked to C-2′ and C-3′ of the
arabinopyranosyloxy unit, respectively. Thus, 1 was determined
to be 3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone
Compound 3 exhibited IR and NMR spectroscopic features
similar to those of 2. Comparison of the NMR, HRESIMS, and
ECD data of 3 and 2 (Tables 1 and 3 and Experimental
Section) indicated that the α-L-arabinopyranosyl and α-L-
rhamnopyranosyl moieties in 2 were replaced by a β-D-
glucopyranosyl moiety in 3. This was confirmed by acid
hydrolysis of 3 and subsequent GC analysis of the hydrolysate,
according to the same protocol as that described for 1. In the
HMBC spectrum of 3, correlations of H-1′/C-3 and H-1″/C-3′
revealed the linkage of the two sugar units. Thus, compound 3
was assigned as (3β,20S,23S)-3,20,23-trihydroxydammar-24-en-
21-oic acid-21,23-lactone 3-O-β-D-xylopyranosyl-(1→3)-β-D-
glucopyranoside.
1374
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
a
1
Table 1. H NMR Data for Compounds 1−3, 8, and 9 in Pyridine-d5 (δ, mult., J in Hz)
b
no.
1
2
3
8
9
1
1.53 m, 0.81 m
2.05 m, 1.86 qd (13.2, 4.0)
3.29 dd (12.0, 4.0)
0.73 d (12.0)
1.46 m, 1.36 m
1.50 m, 1.21 m
1.30 m
1.56 m, 0.78 m
2.08 m, 1.80 m
3.31 dd (12.0, 4.0)
0.74 d (11.2)
1.48 m, 1.38 m
1.50 m, 1.23 m
1.31 m
1.48 m, 0.76 m
2.21 m, 1.82 m
3.37 dd (12.0, 4.0)
0.72 d (12.0)
1.50 m, 1.37 m
1.52 m, 1.22 m
1.28 m
1.52 m, 0.82 m
2.04 m, 1.85 m
3.28 dd (11.5, 4.0)
0.74 brd (12.0)
1.45 m, 1.36 m
1.52 m, 1.23 m
1.33 m
1.51 m, 0.80 m
2.04 m, 1.83 m
2
3
3.26 dd (11.5, 4.0)
0.72 brd (12.0)
1.44 m, 1.35 m
1.51 m, 1.21 m
1.29 m
5
6
7
9
11
12
13
15
16
17
18
19
21
1.46 m, 1.22 m
2.40 m, 1.34 m
2.07 m
1.50 m, 1.30 m
2.51 brd (12.0), 1.47 m
1.85 m
1.50 m, 1.29 m
2.51 brd (12.0), 1.46 m
1.84 m
1.49 m, 1.20 m
2.22 brd (11.0), 1.41 m
2.12 m
1.37 m, 1.11 m
2.13 m, 1.30 m
2.05 m
1.61 m, 1.12 m
2.02 m, 1.69 m
2.52 td (10.0, 4.8)
0.91 s
1.60 m, 1.10 m
2.05 m, 1.30 m
2.71 td (10.4, 5.6)
0.99 s
1.58 m, 1.09 m
2.07 m, 1.31 m
2.71 td (10.5, 4.5)
0.98 s
1.65 m, 1.12 m
2.00, 1.94 m
2.30 m
1.61 m, 1.08 m
1.92 m, 1.88 m
2.21 m
0.95 s
0.92 s
0.76 s
0.82 s
0.79 s
0.77 s
0.75 s
4.06 d (11.0)
4.00 d (11.0)
2.08 m
4.38 d (11.0)
4.01 d (11.0)
2.09 m
22
2.70 dd (13.6, 7.2)
2.30 dd (13.6, 6.0)
5.47 ddd (8.8, 7.6, 6.0)
5.59 d (8.8)
2.60 dd (13.2, 5.6)
2.12 dd (13.2, 10.0)
5.71 ddd (10.0, 8.8, 5.6)
5.41d (8.8)
2.59 dd (13.0, 5.5)
2.10 dd (13.0, 9.5)
5.70 ddd (9.5, 9.0, 5.5)
5.40 d (9.0)
1.95 m
1.90 m
23
24
26
27
28
29
30
1′
2.49 m, 2.40 m
5.33 t (6.0)
1.66 s
2.43 m, 2.31 m
5.27 t (7.0)
1.64 s
1.62 s
1.67 s
1.67 s
1.69 s
1.63 s
1.63 s
1.63 s
1.62 s
1.19 s
1.20 s
1.30 s
1.19 s
1.17 s
1.11 s
1.14 s
1.00 s
1.11 s
1.10 s
0.90 s
0.91 s
0.90 s
0.99 s
0.93 s
4.92 d (5.2)
4.92 d (5.2)
4.91 d (8.0)
4.92 d (5.0)
4.66 dd (7.0, 5.0)
4.30 dd (7.0, 3.0)
4.48 ddd (4.5, 3.0, 2.5)
4.32 dd (12.0, 4.5)
3.83 d (12.0, 2.5)
4.91 d (5.5)
4.65 dd (7.0, 5.5)
4.30 dd (7.0, 3.0)
4.47 ddd (5.0, 3.0, 2.5)
4.33 dd (12.0, 5.0)
3.83 dd (12.0, 2.5)
2′
4.67 dd (7.2, 5.2)
4.29 dd (7.2, 3.2)
4.48 ddd (4.8, 3.2, 2.4)
4.33 dd (11.2, 4.8)
3.81 dd (11.2, 2.4)
4.67 dd (7.2, 5.2)
4.31 dd (7.2, 3.2)
4.48 ddd (4.8, 3.2, 2.4)
4.33 dd (11.2, 4.8)
3.82 dd (11.2, 2.4)
4.06 dd (8.5, 8.0)
4.22 dd (8.5, 8.5)
4.11 dd (8.5, 8.5)
3.95 ddd (8.5, 5.5, 2.0)
3′
4′
5′a
5′b
6′a
6′b
1″
4.53 dd (12.5, 2.0)
4.33 dd (12.5, 5.5)
5.26 d (8.0)
6.13 brs
6.14 brs
6.13 brs
6.11 brs
2″
4.73 dd (3.6, 1.6)
4.57 dd (9.6, 3.6)
4.27 dd (9.6, 9.6)
4.73 dd (3.2, 1.2)
4.57 dd (9.6, 3.2)
4.28 dd (9.6, 9.6)
4.01 dd (8.0, 8.0)
4.14 dd (8.0, 8.0)
4.14 ddd (10.0, 8.0, 5.0)
4.29 dd (11.0, 5.0)
3.70 dd (11.0, 10.0)
4.73 brs
4.72 brs
3″
4.57 dd (9.0, 3.5)
4.27 dd (9.0, 9.0)
4.57 dd (9.0, 3.5)
4.26 dd (9.0, 8.5)
4″
5″a
5″b
6″
4.59 dq (9.6, 6.0)
1.62 d (6.0)
4.59 m
4.59 m
4.59 m
1.62 d (6.0)
1.62 d (6.0)
1.61 d (6.0)
1″′
2″′
3″′
4″′
5″′a
5″′b
5.02 d (7.2)
5.02 d (7.6)
5.01 d (7.5)
5.00 d (7.0)
3.92 dd (8.0, 7.2)
4.09 dd (8.4, 8.0)
4.11 ddd (10.8, 8.4, 4.8)
4.30 dd (10.8, 4.8)
3.65 dd (10.8, 10.8)
3.93 dd (8.0, 7.6)
4.08 dd (8.4, 8.0)
4.11 ddd (10.8, 8.4, 4.8)
4.30 dd (10.8, 4.8)
3.66 dd (10.8, 10.8)
3.93 dd (8.0, 7.5)
4.09 dd (8.0, 8.0)
4.11 ddd (9.5, 8.0, 4.5)
4.30 dd (11.0, 4.5)
3.65 dd (11.0, 9.5)
3.92 dd (8.0, 7.0)
4.08 dd (8.0, 8.0)
4.11 ddd (9.5, 8.0, 5.0)
4.30 dd (11.0, 5.0)
3.66 dd (11.0, 9.5)
a
1H NMR data were measured at 400 MHz for 1 and 2 and 500 MHz for 3, 8, and 9. The assignments were based on DEPT, 1H−1H COSY, HSQC,
b
and HMBC experiments. Data for Glc-21: δ 5.04 d (7.5 Hz, H-1″″), 4.08 dd (8.5 and 7.5 Hz, H-2″″), 4.20 t (8.5 Hz, H-3″″), 4.22 t (8.5 Hz, H-4″″),
3.97 m (H-5″″), 4.55 dd (11.5 and 2.5 Hz, H-6″″a), and 4.37 dd (11.5 and 4.0 Hz, H-6″″b).
The IR, NMR, and HRESIMS spectroscopic data for 4, 5, 6,
and 7 (Supporting Information, Tables S2 and S3) were
identical to those of (3β,20S,23S)-3,20,23-trihydroxydammar-
24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-
(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyra-
noside, (3β,20S,23S)-3,20,23-trihydroxydammar-24-en-21-oic
acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xy-
lopyranosyl-(1→3)]-β-D-glucopyranoside, (3β,20R,23R)-
3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone
3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→
3)]-β-D-6-O-acetylglucopyranoside, and (3β,20R,23R)-3,20,23-
trihydroxydammar-24-en-21-oic acid-21,23-lactone 3-O-[α-L-
rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-
glucopyranoside, respectively, which were isolated from
G. pentaphyllum.6b These compounds, along with five other
derivatives6a,b,8 from G. pentaphyllum, represent glycosides with
the 3β,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lac-
tone aglycone moiety. However, the initial assignment of
1375
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
a
1
Table 2. H NMR Data for Compounds 10−13 in Pyridine-d5 (δ, mult., J in Hz)
b
c
d
b
c
d
no.
10
11
12
13
no.
10
11
12
13
1
2
3
2.63 m, 0.71 m 2.45 m, 0.63 m 1.60 m, 0.96 m 1.38 m, 0.75 m
2.16 m, 1.67 m 2.23 m, 1.62 m 2.21 m, 1.87 m 2.23 m, 1.81 m
4′ 4.43 brs
3.98 dd (8.0,
8.0)
3.82 dd (9.0, 9.0) 4.00 dd (8.5,
8.5)
5′a 4.25 dd (11.6,
3.89 ddd (80,
5.5, 2.0)
3.97 ddd (9.0,
5.5, 2.0)
3.90 ddd (8.5,
5.5, 2.0)
3.38 dd (11.6,
3.6)
3.38 dd (12.0,
4.0)
3.32 dd (12.0,
4.0)
3.34 dd (11.5,
4.0)
4.0)
5
6
7
9
1.18 m
1.08 brs
0.76 brd (11.0) 0.67 brd (11.5)
5′b 3.80 brd (11.6)
6′a
1.85 m, 1.65 m 1.85 m, 1.60 m 1.47 m, 1.38 m 1.45 m, 1.35 m
1.62 m, 1.35 m 1.58 m, 1.31 m 1.52 m, 1.23 m 1.47 m, 1.19 m
4.49 dd (12.5,
2.0)
4.82 dd (12.0,
2.0)
4.50 dd (12.0,
2.0)
1.70 m
1.65 m
1.38 m
1.24 m
6′b
4.25 dd (12.6,
5.5)
4.70 dd (12.0,
5.5)
4.26 dd (12.0,
5.5)
11 1.72 m, 1.17 m 1.58 m, 1.06 m 1.49 m, 1.20 m 1.35 m, 1.14 m
12 2.17 m, 1.37 m 2.12 m, 1.29 m 2.24 m, 1.42 m 2.26 m, 1.35 m
1″ 5.26 d (7.6)
6.41 brs
6.44 brs
6.44 brs
2″ 4.00 dd (8.0, 7.6) 4.79 dd (3.5,
4.78 brd (3.5)
4.80 brd (3.5)
13 2.05 m
1.97 m
2.13 m
2.03 m
1.5)
15 1.59 m, 1.17 m 1.56 m, 1.12 m 1.66 m, 1.13 m 1.61 m, 1.08 m
16 2.02 m, 1.92 m 2.03 m, 1.92 m 2.00 m, 1.94 m 1.92 m, 1.81 m
3″ 4.14 dd (8.4, 8.0) 4.57 dd (9.5,
4.58 dd (9.5, 3.5) 4.59 dd (9.0,
3.5)
3.5)
17 2.27 m
2.23 m
2.30 m
2.31 m
4″ 4.19 dd (10.0,
4.26 dd (9.5,
9.5)
4.29 dd (9.5, 9.5) 4.29 dd (9.0,
9.0)
18 0.86 s
0.94 s
0.96 s
0.92 s
8.0, 4.8)
19 10.03 s
21 4.01 d (10.4)
3.94 d (10.4)
10.25 s
0.79 s
0.73 s
5″a 4.34 dd (11.2,
4.69 dq (9.5,
6.0)
4.74 dq (9.5, 6.0) 4.75 dq (9.0,
6.0)
4.34 d (11.0)
3.97 d (11.0)
4.06 d (11.0)
3.99 d (11.0)
4.29 d (11.0)
4.08 d (11.0)
4.8)
5″b 3.70 dd (11.2,
22 2.05 m, 1.93 m 2.04 m, 1.86 m 2.11 m, 1.96 m 2.01 m, 1.82 m
23 2.45 m, 2.38 m 2.44 m, 2.31 m 2.49 m, 2.39 m 2.49 m, 2.30 m
10.0)
6″
1.64 d (6.0)
4.99 d (8.0)
1.67 d (6.0)
4.98 d (7.5)
1.67 d (6.0)
4.99 d (8.0)
24 5.32 t (7.2)
26 1.67 s
5.27 t (7.0)
1.62 s
5.33 t (7.0)
1.66 s
5.27 t (7.0)
1.63 s
1″′
2″′
3.95 dd (8.0,
8.0)
3.96 dd (8.5, 7.5) 3.96 dd (8.5,
8.0)
27 1.61 s
1.61 s
1.62 s
1.65 s
28 1.33 s
1.27 s
1.24 s
1.22 s
3″′
4.07 dd (8.0,
8.0)
4.07 dd (9.0, 8.5) 4.07 dd (8.5,
8.5)
29 0.91 s
1.11 s
1.15 s
1.15 s
30 0.98 s
0.83 s
0.98 s
0.92 s
4″′
4.10 ddd (10.0, 4.11 ddd (10.5, 4.10 m
8.0, 5.0)
9.0, 5.0)
1′ 4.75 d (7.6)
4.86 d (7.5)
4.84 d (7.5)
4.88 d (7.5)
5″′a
5″′b
4.27 dd (11.0,
5.0)
4.27 dd (11.0,
5.0)
4.27 m
2′ 4.54 dd (8.8, 7.6) 4.22 d (8.0, 7.5) 4.21 dd (8.0, 7.5) 4.23 dd (8.0,
7.5)
3.68 dd (11.0,
10.0)
3.68 dd (11.0,
10.5)
3.68 dd (11.0,
10.0)
3′ 4.18 dd (8.8, 3.6) 4.16 dd (8.0,
4.15 dd (9.0, 8.0) 4.17 dd (8.5,
8.0)
8.0)
a
1
1H NMR data were measured at 400 MHz for 10 and 500 MHz for 11−13. The assignments were based on DEPT, H−1H COSY, HSQC, and
b
HMBC experiments. Data for Glc-21 of 11: δ 5.02 d (7.5 Hz, H-1″″), 4.08 dd (8.0 and 7.5 Hz, H-2″″), 4.21 dd (8.0 and 9.0 Hz, H-3″″), 4.23 t (9.0
c
d
Hz, H-4″″), 3.97 m (H-5″″), 4.55 dd (12.0 and 2.0 Hz, H-6″″a), and 4.36 dd (12.0 and 5.5 Hz, H-6″″b). Data for OAc of 12: δ 2.04 s. Data for inner
Glc-21 of 13: δ 4.95 d (8.0 Hz, H-1″″), 4.04 t (8.0 Hz, H-2″″), 4.17 t (8.0 Hz, H-3″″), 4.08 dd (8.0 and 9.0 Hz, H-4″″), 4.10 m (H-5″″), 4.87 brd (12.0
Hz, H-6″″a), and 4.22 m (H-6″″b); for terminal Glc-21 of 13: δ 5.00 d (8.0 Hz, H-1″″′),4.04 t (8.0 Hz, H-2″″′), 4.22 dd (8.0 and 9.0 Hz, H-3″″′), 4.20
dd (8.0 and 9.0 Hz, H-4″″′), 3.90 m (H-5″″′), 4.49 dd (12.0 and 2.0 Hz, H-6″″′a), and 4.35 dd (12.0 and 5.0 Hz, H-6″″′b).
(20S) and (20R) configurations for two reported C-20
epimers,6a made by comparison of the chemical shifts of C-13
and C-17 in these compounds with those in known ginseng
saponins,9 was ambiguous. This is because the substitution
patterns of these epimers differed significantly from those of the
reference ginseng saponins, and the configurations of the other
reported compounds were determined by comparison of their
NMR data with those of the two C-20 epimers. This, together
with the configuration assignments of 1−3, prompted us to re-
examine the configurations of the reported compounds by
comparing the NMR data of 1−3 with those of 4−7 and the
reported compounds.6a,b,8 Without exception, the data for the
aglycone moiety in 1 were consistent with those of the aglycone
moieties in 4 and 5 and the (20S,23S)-aglycone moiety in
the reported compounds. In addition, the data for the aglycone
moieties in 2 and 3 were similar to those of the aglycone
moieties in 6 and 7 and the (20R,23R)-aglycone moiety in the
reported glycosides. This, in combination with the config-
uration assignments of 1−3 made on the basis of NOESY and
ECD data, indicated that 4 and 5 had the same aglycone as 1
and that 6 and 7 had the same aglycone as 2 and 3. This was
proved by the Cotton effects in the ECD spectra of 4−7, as well
as by acid hydrolysis of 4−7, which generated the same
products (1a and 1b) as those from 1−3. These results also
indicated that the configuration of the reported (3β,20S,23S)-
and (3β,20R,23R)-3,20,23-trihydroxydammar-24-en-21-oic
acid-21,23-lactone derivatives6b,8 should be revised as
(3β,20S,23R) and (3β,20S,23S), respectively. This was further
supported by the [α]20D values of 4−7, 1a, and 1b, which were
consistent with those of the reported compounds having the
same gross structures.6b,10 In addition, the configurations at
C-20 of the (20R,23ξ)- and (20S,23ξ)-derivatives6a,10 did not
match the presented structures, and on basis of the above
revision, the configurations had to be revised as (20S,23R) and
(20S,23S), respectively. Therefore, 4, 5, 6, and 7 were defined
as (3β,20S,23R)-3,20,23-trihydroxydammar-24-en-21-oic acid-
21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyr-
anosyl-(1→3)]-β-D-6-O-acetylglucopyranoside, (3β,20S,23R)-3,
20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone 3-O-
[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-
D-glucopyranoside, (3β,20S,23S)-3,20,23-trihydroxydammar-
24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-
(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyra-
noside, and (3β,20S,23S)-3,20,23-trihydroxydammar-24-en-
21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-
xylopyranosyl-(1→3)]-β-D-glucopyranoside, respectively. The sugar
1376
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
a
Table 3. 13C NMR Data for Compounds 1−3 and 8−13 in Pyridine-d5
b
c
d
e
no.
1
2
3
8
9
10
11
12
13
1
39.6
26.8
88.3
39.8
56.6
18.4
35.7
40.8
51.2
37.1
21.8
27.3
43.3
50.7
31.7
25.8
45.8
15.6
16.5
79.0
179.4
40.8
74.1
125.4
138.5
25.6
18.1
27.9
16.8
16.6
104.8
74.7
81.5
68.3
64.8
39.8
26.8
88.2
39.6
56.6
18.4
35.7
40.7
51.2
37.1
21.8
28.0
45.0
50.2
31.7
26.2
45.3
15.6
16.6
81.1
178.3
39.0
75.2
124.0
139.4
25.6
18.1
27.9
16.8
16.2
104.7
74.7
81.5
68.3
64.8
39.7
26.7
89.1
39.3
56.4
18.4
35.7
40.7
51.1
37.0
21.8
27.9
45.0
50.2
31.7
26.2
45.3
15.6
16.5
81.1
178.3
39.0
75.3
124.0
139.4
25.6
18.1
28.0
16.8
16.2
106.5
74.7
87.8
69.6
78.1
62.7
106.3
75.2
78.2
70.9
67.4
39.8
26.8
88.3
39.6
56.6
18.5
35.7
40.8
51.2
37.0
21.9
28.1
41.8
50.5
31.8
24.8
46.3
15.8
16.6
76.6
66.8
36.6
23.3
126.2
130.8
25.8
17.7
28.0
16.7
16.8
104.8
74.7
81.6
68.3
64.8
39.8
26.8
88.3
39.5
56.5
18.4
35.8
40.7
51.1
37.0
21.8
27.8
41.7
50.5
31.6
24.7
46.1
15.8
16.6
76.4
76.3
36.6
23.3
126.0
130.8
25.8
17.8
27.9
16.6
16.8
104.7
74.6
81.4
68.2
64.8
33.5
27.7
87.5
40.1
54.6
17.7
34.7
40.4
52.9
52.8
22.4
27.9
41.6
50.3
32.1
24.7
46.1
16.6
205.7
76.4
66.6
36.6
23.2
126.2
130.2
25.8
17.7
26.5
15.9
17.2
107.4
71.8
83.5
69.4
67.1
33.6
27.7
88.1
40.4
54.9
17.6
34.6
40.0
52.8
52.7
22.3
27.9
41.6
50.3
32.0
24.7
46.0
16.0
205.5
76.3
76.2
36.5
23.3
126.0
130.9
25.8
17.8
26.3
16.5
17.2
104.9
76.9
87.6
69.9
78.2
62.6
101.8
72.4
72.6
73.9
69.7
18.6
104.9
74.8
78.3
70.6
67.3
39.7
26.8
88.6
39.7
56.8
18.5
35.7
40.8
51.2
37.1
21.9
28.1
41.8
50.5
31.7
24.8
46.3
15.8
16.6
76.6
66.8
36.6
23.3
126.2
130.8
25.8
17.7
27.8
16.7
16.8
105.0
76.6
87.8
69.8
74.4
64.1
101.8
72.4
72.5
73.9
69.9
18.6
104.9
74.8
78.3
70.6
67.3
39.7
26.9
88.9
39.7
56.6
18.5
35.6
40.7
51.0
37.0
21.8
27.7
41.9
50.3
31.5
24.9
46.5
15.8
16.6
76.3
77.8
35.8
23.1
126.0
130.9
25.8
17.8
27.9
16.7
16.9
104.9
77.0
88.2
69.8
78.0
62.6
101.8
72.4
72.5
73.9
69.7
18.6
105.0
74.8
78.2
70.6
67.3
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1′
2′
3′
4′
5′
6′
1″
2″
3″
4″
5″
6″
1″′
2″′
3″′
4″′
5″′
102.1
72.5
72.6
74.0
70.1
18.6
105.2
74.6
77.7
70.7
67.0
102.1
72.5
72.6
73.9
70.1
18.6
105.1
74.6
77.7
70.9
67.0
102.1
72.5
72.6
74.0
70.1
18.6
105.1
74.6
77.7
70.9
67.0
102.0
72.4
72.5
73.9
70.6
18.6
105.0
74.5
77.6
70.9
67.0
106.9
75.3
78.2
71.1
67.2
a
1
Data were measured at 100 MHz for 1, 2, and 10 and 125 MHz for 3, 8, 9, and 11−13. The assignments were based on DEPT, H−1H COSY,
b
c
HSQC, and HMBC experiments. Data for Glc-21 of 9: δ 106.2 (C-1″″), 75.5 (C-2″″), 78.7 (C-3″″), 71.1 (C-4″″), 78.6 (C-5″″), 62.8 (C-6″″). Data
d
e
for Glc-21of 11: δ 106.2 (C-1″″), 75.5 (C-2″″), 78.7 (C-3″″), 71.1 (C-4″″), 78.6 (C-5″″), 62.8 (C-6″″). Data for OAc of 12: δ 170.6, 20.8. Data for
inner Glc-21of 13: δ 106.3 (C-1″″), 75.4 (C-2″″), 78.5 (C-3″″), 71.7 (C-4″″), 77.0 (C-5″″), 70.1 (C-6″″); for terminal Glc-21 of 13: δ 105.2 (C-1″″′),
75.2 (C-2″″′), 78.3 (C-3″″′), 71.5 (C-4″″′), 78.3 (C-5″″′), 62.6 (C-6″″′).
units in 4−7 were confirmed by acid hydrolysis and subsequent GC
analysis of the hydrolysates using the protocol described earlier, and
their linkages were confirmed by 2D NMR data.
the γ-lactone moiety in the 3β,20,23-trihydroxydammar-24-en-
21-oic acid-21,23-lactone derivatives. In compounds with OH-20
and H-23 in the trans orientation [(20S,23R)- or (20R,23S)-
isomers], the H-24 resonance was deshielded by OH-20. Thus,
the chemical shift of H-24 (δH 5.60 0.02 for the glycosides in
Detailed NMR data analysis of 1−7 and the reported
compounds6a,b,8,10 indicated that the readily distinguishable
1
H-23 and H-24 resonances in the H NMR spectra should be
pyridine-d5; δH 5.31 and 5.26
0.01 for the aglycone in
applicable to the determination of the relative configuration of
acetone-d6 and CDCl3, respectively) was larger than that of
1377
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
Figure 1. (a) Experimental ECD spectra of 1 and 1a and the calculated ECD spectra of 1a and the (20R,23S)-isomer of 1a. (b) Experimental ECD
spectra of 2 and 1b and the calculated ECD spectra of 1b and the (20R,23R)-isomer of 1b.
H-23 (δH 5.47 0.02 for the glycosides in pyridine-d5; δH 5.23
and 5.12 for the aglycone in acetone-d6 and CDCl3,
respectively). On the other hand, in the compounds with
OH-20 and H-23 in the cis orientation [(20S,23S)- or
(20R,23R)-isomers], the H-23 resonance was deshielded by
OH-20 and the chemical shift of H-23 (δH 5.70 0.02 for the
glycosides in pyridine-d5; δH 5.31 and 5.34 for the aglycone in
acetone-d6 and CDCl3, respectively) was larger than that of
H-24 (δH 5.42 0.02 for the glycosides in pyridine-d5; δH 5.24
and 5.19 for the aglycone in acetone-d6 and CDCl3,
respectively). To eliminate errors, the chemical shift difference
Furthermore, comparison of the Cotton effects in the
experimental ECD spectra of 1−7 with those in the calculated
ECD spectra of 1a and the (20R,23S)-isomer and 1b and the
(20R,23R)-isomer revealed that the absolute configurations of
the γ-lactone moiety in the 3β,20,23-trihydroxydammar-24-en-
21-oic acid 21,23-lactone derivatives could be determined by
the Cotton effects. For the compounds with OH-20 and H-23
in the trans orientation, the (20S,23R)-isomers showed positive
Cotton effects at 227 1 nm and negative effects at 206 2 nm
(calculated at 231 and 208 nm), while the (20R,23S)-isomer
showed the reversed Cotton effects at the corresponding
wavelengths (calculated at 235 and 208 nm). For the
compounds with OH-20 and H-23 in the cis orientation, the
(20S,23S)-isomers showed negative Cotton effects at 230 2 nm
between the H-24 and H-23 resonances (ΔδH = δH‑24 − δH‑23
)
is proposed to be used in determination of the relative
configuration of the γ lactone moiety in the 3β,20,23-
trihydroxydammar-24-en-21-oic acid 21,23-lactone derivatives.
For compounds with OH-20 and H-23 in the trans orientation,
the ΔδH value is positive (approximately +0.13 0.02 ppm for
the glycosides in pyridine-d5; +0.08 0.01 and +0.14 0.01
ppm for the aglycone in acetone-d6 and CDCl3, respectively).
Conversely, for the compounds with OH-20 and H-23 in the cis
orientation, the ΔδH value is negative (around −0.30 0.02
ppm for the glycosides in pyridine-d5; −0.07 0.01 and −0.15
0.01 ppm for the aglycone in acetone-d6 and CDCl3,
respectively). In addition, the compounds with OH-20 and
H-23 in the trans and cis orientations showed different chemical
shifts for the C-20, C-21, C-22, C-23, C-24, and C-25
resonances in the 13C NMR spectra. In pyridine-d5, for the
glycosides with OH-20 and H-23 in the trans orientation, C-20,
C-23, and C-25 were shielded by ΔδC −2.1 0.1, −1.1 0.1,
and −1.0 0.1 ppm, respectively, as compared with those for
the compounds with OH-20 and H-23 in the cis orientation,
whereas C-21, C-22, and C-24 were deshielded by ΔδC +1.1
0.2, +1.6 0.2, and +1.4 0.2 ppm. To eliminate errors, the
chemical shift difference between the readily distinguishable
C-25 and C-24 resonances (ΔδC = δC‑25 − δC‑24) is suggested to
be used in assignment of the relative configuration of the
γ-lactone moiety in the 3β,20,23-trihydroxydammar-24-en-21-
oic acid-21,23-lactone derivatives. The ΔδC value is smaller
than +14.0 ppm for the glycosides with OH-20 and H-23 in the
trans orientation (+13.1 0.1 ppm, pyridine-d5), but for the
glycosides with OH-20 and H-23 in the cis orientation, the ΔδC
value is larger than +14.0 ppm (+15.6 0.2 ppm, pyridine-d5).
Similar chemical shift differences are observed for the aglycones
1a and 1b in CDCl3 (ΔδC: +17.0 0.2 ppm for 1a and +18.1
0.1 ppm for 1b) (Supporting Information, Table S1).10
and positive effects at 204
2 nm (calculated at 235 and
208 nm), while the (20R,23R)-isomer showed the reversed
Cotton effects at the corresponding wavelengths (calculated at 236
and 212 nm). This also demonstrated that the C-23 configuration
of the γ-lactone moiety in these compounds is determining the
signs of the Cotton effects, while the configuration change at C-20
induces only a variation in the intensity of the Cotton effects.
Comparison of the NMR spectra of compound 8 with those
of 1 (Tables 1 and 3) indicated that the only difference
between these two compounds was replacement of the
oxymethine (C-23) and carbonyl (C-21) units in 1 by
methylene and hydroxymethyl groups in 8, respectively. This
suggested that 8 had a 3,20,21-trihydroxydammar-24-ene
aglycone moiety,6a,8a,11 which was further supported by
HRESIMS and confirmed by 2D NMR analysis. Particularly,
HMBC correlations of H2-21/C-17, C-20, and C-22; H2-23/
C-20, C-22, C-24, and C-25; and H-1′/C-3, in combination
with their chemical shifts, verified the presence of hydroxy
groups at C-20 and C-21 in 8. The S configuration at C-20 was
proposed from the chemical shifts of C-20 (δ 76.6) and C-17 (δ
46.3).11a,12 This was confirmed by the in situ dimolybdenum
ECD method,13 which was reported for the assignment of the
configurations of acyclic 1,2-diols.4,14 According to the
empirical rule proposed by Snatzke,13,15 the bands around
310 nm (band IV) and 400 nm (band II) in the Mo2(AcO)4-
induced ECD spectrum, which have the same sign as the
O−C−C−O torsion angle in the favored conformation, allow
for the assignment of the absolute configuration.13b,15,16 Acid
hydrolysis of 8 yielded the aglycone 8a. In the Mo2(AcO)4-
induced ECD spectrum of 8a (Supporting Information, Figure
S68), positive Cotton effects at 309 and 386 nm supported the
1378
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
20S configuration. Therefore, compound 8 was determined as
(3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-[α-L-rhamno-
pyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyr-
anoside.
3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-
β-D-glucopyranosyl}-21-O-β-D-glucopyranosyl-(1→6)-β-D-gluco-
pyranoside.
The other known compounds were identified by comparison
of spectroscopic data with reported data as gylongiposide I,
gypenosides XLVIII11a and XLIX,12a (3β,20S)-3,20,21-trihy-
droxydammar-24-ene 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-
xylopyranosyl-(1→3)]-β-D-glucopyranoside,12c (3β,20S)-3,19,
20,21-tetrahydroxydammar-24-ene 3-O-{[α-L-rhamnopyranosyl-
(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranosyl}-21-O-
β-D-glucopyranoside, (3β,20S)-3,19,20,21-tetrahydroxydammar-
24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-
(1→3)]-β-D-glucopyranosyl}-21-O-β-D-glucopyranoside,
(3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-{[α-L-rhamno-
pyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyrano-
syl}-21-O-β-D-glucopyranoside, and (3β,20S)-3,20,21-trihydroxy-
dammar-24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylo-
pyranosyl-(1→3)]-β-D-6-O-acetylglucopyranosyl}-21-O-β-D-gluco-
pyranoside.11b
Similar dammarane derivatives from G. pentaphyllum have
been reported to possess biological activities such as protein
tyrosine phosphatase 1B (PTP1B) inhibitory activity8b,10 and
potential apoptotic effect17 or cytotoxic activity against various
cancer cell lines.18 The isolates from M. yaoshansis were tested
at 10 μM in preliminary assays to assess their PTP1B inhibitory
activity,19 cytotoxicity against A2780 ovary, HCT-8 colon, Bel-
7402 hepatoma, BGC-823 stomach, and A549 lung cancer cell
lines,20 and the TNF-α secretion inhibitory activity of mouse
peritoneal macrophages.21 However, all the isolates were found
to be inactive in all assays.
The spectroscopic data of compound 9 indicated that it was a
derivative of 8 with an additional β-glucopyranosyl unit.
Comparison of the NMR spectra of 9 and 8 demonstrated
that the H-21a, H-21b, and C-21 resonances in 9 were
deshielded by ΔδH +0.32, +0.01 and ΔδC +9.5 ppm,
respectively. This suggested that the β-glucopyranosyl unit
was located at C-21 in 9, which was confirmed by HMBC
correlations of H-1″″/C-21 and H2-21/C-1″″. Thus, compound
9 was determined as (3β,20S)-3,20,21-trihydroxydammar-24-
ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-
(1→3)]-α-L-arabinopyranosyl}-21-O-β-D-glucopyranoside.
Spectroscopic data analysis of compound 10 (C40H66O12)
indicated that it was another derivative of 8, with one less
rhamnopyranosyl unit and the methyl group replaced by a
formyl group (δH 10.03 and δC 205.7). In the HMBC spectrum,
correlations of H-5 and H-9/C-19 and H-19/C-1, C-9, and C-10
revealed that the formyl group was located at C-10 in 10,
while HMBC correlations of H-1′/C-3 and H-1″/C-3′ proved
the linkage of the sugar units. Thus, compound 10 was
determined as (3β,20S)-19-oxo-3,20,21-trihydroxydammar-24-
ene 3-O-β-D-xylopyranosyl-(1→3)-α-L-arabinopyranoside.
Comparison of the spectroscopic data of compound 11 with
those of 9 indicated that the methyl (CH3-19) and α-
arabinopyranosyl units in 9 were replaced by a formyl group
and a β-glucopyranosyl moiety in 11, respectively. This was
confirmed by 2D NMR analysis of 11, as well as by acid
hydrolysis of 11 followed by GC analysis of the hydrolysate
using the same protocol as described above. In particular,
HMBC correlations of H-5 and H-9/C-19; H-19/C-1, C-9, and
C-10; H-1′/C-3; H-1″/C-2′; H-1″′/C-3′; and H-1″″/C-21
verified the location of the formyl group and the linkage of
the sugar units in 11. Therefore, compound 11 was defined as
(3β,20S)-19-oxo-3,20,21-trihydroxydammar-24-ene 3-O-{[α-L-
rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glu-
copyranosyl}-21-O-β-D-glucopyranoside.
Compound 12 had the molecular formula C47H80O16 as
indicated by HRESIMS and NMR data. Comparison of the
NMR data of 12 and 8 suggested that the α-arabinopyranosyl
group in 8 was replaced by a β-D-6-O-acetylglucopyranosyl
moiety in 12. The suggestion was confirmed by 2D NMR
experiments and GC analysis of the hydrolysate of 12.
Particularly, in the HMBC spectrum of 12, correlations of
H-1′/C-3, H-1″/C-2′, and H-1″′/C-3′ confirmed the linkage of the
glycosyl moieties, while a correlation from the H2-6′ resonance
to the acetyl carbonyl carbon (δC 170.6) verified the location of
the acetyl group at C-6′. Thus, compound 12 was determined
as (3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-[α-L-rham-
nopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-ace-
tylglucopyranoside.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured using a Rudolph Research Autopol III automatic polar-
imeter. UV and ECD spectra were recorded using a JASCO J-815
spectropolarimeter. IR spectra were recorded using a Nicolet 5700 FT-
IR microscope spectrometer (FT-IR microscope transmission). 1D
1
and 2D NMR spectra were obtained at 400, 500, or 600 MHz for H
and 100, 125, or 150 MHz for 13C using a Varian 400, 500, or 600
MHz NMR spectrometer in pyridine-d5, acetone-d6, or CDCl3, with
solvent peaks used as references. ESIMS data were measured using a
Q-Trap LC/MS/MS (Turbo Ionspray Source) spectrometer.
HRESIMS data were measured using an AccuToFCS JMS-T100CS
spectrometer. Column chromatography was performed with HPD-100
macroporous adsorbent resin (Cangzhou Bonchem Co., Ltd., China),
silica gel (200−300 mesh, Qingdao Marine Chemical Inc., China), and
Sephadex LH-20 (GE Healthcare Bio-Sciences AB, Sweden). HPLC
was performed using a Waters 600 controller, a Waters 600 pump,
and a Waters 2487 dual λ absorbance detector with an Alltima (250 ×
10 mm i.d.) preparative C18 (5 μm) column. TLC was carried out with
glass precoated silica gel GF254 plates. Spots were visualized under UV
light or by spraying with 7% H2SO4 in 95% aqueous EtOH followed
by heating.
Plant Material. Roots of Machilus yaoshansis were collected at
Dayao Mountain, Guangxi, China, in December 2007. The plant was
identified by Mr. Guang-Ri Long (Guangxi Forest Administration,
Guangxi 545005, China). A voucher specimen (no. 07114) was
deposited at the Herbarium of Guangxi Forest Administration, China.
Extraction and Isolation. The air-dried roots of M. yaoshansis
(10 kg) were powdered and extracted with 95% aqueous EtOH (3 × 15 L)
at room temperature (3 × 48 h). The EtOH extract was evaporated
under reduced pressure to yield a dark brown residue (1050 g).
The residue was suspended in H2O (5 L) and partitioned with EtOAc
(5 × 5 L). The aqueous phase was loaded onto an HPD-100 macroporous
adsorbent resin (1500 g, dry weight) column. Successive elution
with H2O, 30% EtOH(aq), 70% EtOH(aq), and 95% EtOH(aq)
Compound 13 had the molecular formula C59H100O26, as
indicated by HRESIMS and NMR data. The NMR spectra of
13 were similar to those of 12, except for the presence of
resonances attributable to two additional β-glucopyranosyl
units and the absence of the acetyl resonances in 13. The NMR
resonance assignments for 13 were confirmed by 2D NMR data
analysis. Particularly, the HMBC correlations of H-1′/C-3,
H-1″/C-2′, H-1″′/C-3′, H-1″″/C-21, and H-1″″′/C-6″″ demonstra-
ted the linkage of the sugar units in 13. Therefore, compound 13
was determined as (3β,20S)-3,20,21-trihydroxydammar-24-ene
1379
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
(10 L each) and solvent removal yielded four corresponding fractions.
The fraction (170 g) eluted by 70% EtOH(aq) was separated over
silica gel, with elution using a gradient of increasing MeOH
concentration in CHCl3 (2−100%) to give eight fractions (A−H).
Fraction C (28 g) was fractionated via RP-MPLC using a preparative
C18 (5 μm) column, with elution using a gradient of increasing MeOH
concentration (0−85%) in H2O to give fractions C1−C7. Fraction C5
(2.9 g) was further chromatographed over Sephadex LH-20 (MeOH−
H2O, 1:1) to afford fractions C5‑1−C5‑4. Subsequent separation of C5‑2
by RP-HPLC using MeOH−H2O (85:15) as the mobile phase
afforded 1 (36 mg), 9 (15 mg), and 12 (9 mg), and separation of C5‑3
gave 2 (16 mg), 3 (33 mg), 8 (17 mg), and 10 (15 mg). Purification of
C5‑4 by RP HPLC using MeOH−H2O (83:17) as the mobile phase
gave 11 (12 mg) and 13 (35 mg).
IR νmax 3457, 3403, 1725, 1620, 1447, 1388 cm−1; 1H NMR (pyridine-
d5, 500 MHz) data, see Table 2; 13C NMR (pyridine-d5, 125 MHz)
data, see Table 3; ESIMS m/z 965 [M + Na]+; HRESIMS m/z
965.5454 [M + Na]+ (calcd for C49H82O17Na, 965.5450).
(3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-{[α-L-rhamno-
pyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyranosyl}-
21-O-β-D-glucopyranosyl-(1→6)-β-D-glucopyranoside (13): amor-
phous powder; [α]20 −9.3 (c 0.82, MeOH); IR νmax 3390, 1683,
D
1
1646, 1451, 1374 cm−1; H NMR (pyridine-d5, 500 MHz) data, see
Table 2; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS
m/z 1223 [M − H]−, 1247 [M + Na]+; HRESIMS m/z 1247.6408
[M + Na]+ (calcd for C59H100O26Na, 1247.6400).
Acid Hydrolysis of 1−13. Compounds 1−13 (5−10 mg) were
individually hydrolyzed with 1 N HCl−dioxane (1:1, 3 mL) at 60 °C
for 6 h. After dilution with H2O (5 mL), the reaction mixture was
extracted with EtOAc to yield separate EtOAc and H2O phases. The
organic layer was concentrated, and the residue was purified by HPLC
using 90% MeOH in H2O to afford 1a and 1b from 1−7, and 8a from
8, 9, 12, and 13. 1a: [α]20D −9.5 (c 0.1, MeOH); ECD (MeOH) 208
(3β,20S,23R)-3,20,23-Trihydroxydammar-24-en-21-oic acid-
21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyrano-
syl-(1→3)]-α-L-arabinopyranoside (1): amorphous powder; [α]20
D
−14.8 (c 0.49, MeOH); ECD (MeOH) 204 (Δε −3.46), 227
(Δε +1.66) nm; IR νmax 3383, 1760, 1647, 1549, 1449, 1377 cm−1; 1H
NMR (pyridine-d5, 400 MHz) data, see Table 1; 13C NMR (pyridine-
d5, 100 MHz) data, see Table 3; ESIMS m/z 905 [M + Na]+;
HRESIMS m/z 905.4832 [M + Na]+ (calcd for C46H74O16Na,
905.4874).
1
(Δε −1.06), 226 (Δε +1.82); H NMR (acetone-d6 and CDCl3, 600
MHz) and 13C NMR (CDCl3, 125 MHz) data, see Supporting
Information, Table S1; ESIMS m/z 495 [M + Na]+. 1b: [α]20 +41.3
D
(c 0.3, MeOH); ECD (MeOH) 206 (Δε +0.36), 230 (Δε −2.11); 1H
NMR (acetone-d6 and CDCl3, 600 MHz) and 13C NMR (CDCl3, 125
MHz) data, see Supporting Information, Table S1; ESIMS m/z 495
(3β,20S,23S)-3,20,23-Trihydroxydammar-24-en-21-oic acid-
21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-
xylopyranosyl(1→3)]-α-L-arabinopyranoside (2): amorphous pow-
der; [α]20D +7.1 (c 0.10, MeOH); ECD (MeOH) 203 (Δε +1.76), 231
(Δε −1.62) nm; IR νmax 3412, 1753, 1666, 1647, 1547, 1448, 1378
[M + Na]+. 8a: [α]20 +32.1 (c 0.3, MeOH); Mo2(OAc)4-induced
D
ECD (DMSO) 272 (Δε′ −0.26), 309 (Δε′ +0.59), 359 (Δε′ +0.09),
1
386 (Δε′ +0.11); H NMR (CDCl3, 400 MHz) data, see Supporting
1
cm−1; H NMR (pyridine-d5, 400 MHz) data, see Table 1; 13C NMR
Information, Table S2; 13C NMR (CDCl3, 100 MHz) data, see
Supporting Information, Table S3; ESIMS m/z 483 [M + Na]+, 459
[M − H]−.
(pyridine-d5, 100 MHz) data, see Table 3; ESIMS m/z 905 [M +
Na]+; HRESIMS m/z 905.4914 [M + Na]+ (calcd for C46H74O16Na,
905.4874).
The H2O layer was evaporated under reduced pressure. After
addition of H2O (5 mL), the acidic solution was evaporated again, and
this procedure was repeated until a neutral solution was obtained. The
neutral solution was evaporated and dried in vacuo to furnish a
monosaccharide residue. The residue was dissolved in pyridine (0.5 mL),
and 2 mg of L-cysteine methyl ester hydrochloride was added. The
mixture was maintained at 60 °C for 2 h, evaporated under a stream of
N2, and dried in vacuo. Next, 0.2 mL of N-trimethylsilylimidazole was
added, and the resultant reaction mixture was maintained at 60 °C for 1 h.
The mixture was partitioned between n-hexane and H2O (2 mL each),
and the n-hexane extract was analyzed by GC-MS under the following
conditions: capillary column, DB-5 (30 m × 0.25 mm × 0.25 μm);
detector, FID; detector temperature, 280 °C; injection temperature,
250 °C; initial temperature, 100 °C for 2 min and subsequent increase to
280 °C at a rate of 10 °C/min; final temperature, 280 °C for 5 min;
carrier, N2 gas. The absolute configurations of the sugars isolated from the
hydrolysates of 1−13 were determined by comparing the retention times
of their trimethylsilyl-L-cysteine derivatives with those of authentic sugars
prepared by a similar procedure. The retention times of the trimethylsilyl-
L-cysteine derivatives of the sugars were as follows: D-glucose, 19.55 min;
D-xylopyranose, 17.65 min; L-rhamnopyranose, 18.38 min; and L-
arabinopyranose, 17.79 min.
ECD Calculation. Conformational analyses of 1a and the
(20R,23S)-isomer were carried out via Monte Carlo searching in the
MMFF94 molecular mechanics force field using the SPARTAN 08
software.22 The two lowest energy conformers for 1a and the 12
lowest energy conformers for the (20R,23S)-isomer (Supporting
Information, Figure S2), whose relative energies were within 2 kcal/
mol, were considered for further DFT calculations. Subsequently, the
conformers were reoptimized using DFT at the B3LYP/6-31G(d)
level in the gas phase with the Gaussian 09 program.23 The B3LYP/
6-31G(d) harmonic vibrational frequencies were further calculated to
confirm their stability. The energies, oscillator strengths, and rotational
strengths of the first two electronic excitations of the conformers were
calculated using TDDFT methodology at the B3LYP/6-311++G-
(2d,2p) level in the gas phase, and the ECD spectra were simulated
by the GaussSum 2.25 program (σ = 0.8 eV).24 To obtain the final
spectra of 1a and the (20R,23S)-isomer, the simulated spectra of the
corresponding lowest energy conformations were averaged according
(3β,20S,23S)-3,20,23-Trihydroxydammar-24-en-21-oic acid-
21,23-lactone 3-O-β-D-xylopyranosyl-(1→3)-β-D-glucopyranoside
(3): amorphous powder; [α]20 +13.8 (c 0.65, MeOH); ECD
D
(MeOH) 205 (Δε +0.62), 231 (Δε −1.31) nm; IR νmax 3406, 1756,
1
1642, 1450, 1377 cm−1; H NMR (pyridine-d5, 500 MHz) data, see
Table 1; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS
m/z 789 [M + Na]+; HRESIMS m/z 789.4394 [M + Na]+ (calcd for
C41H66O13Na, 789.4401).
(3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-[α-L-rhamno-
pyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyrano-
side (8): amorphous powder; [α]20 +6.0 (c 0.42, MeOH); IR νmax
D
3401, 1643, 1562, 1451, 1377 cm−1; 1H NMR (pyridine-d5, 500 MHz)
data, see Table 1; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3;
ESIMS m/z 893 [M + Na]+; HRESIMS m/z 893.5279 [M + Na]+
(calcd for C46H78O15Na, 893.5238).
(3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-{[α-L-rhamno-
pyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyrano-
syl}-21-O-β-D-glucopyranoside (9): amorphous powder; [α]20D −13.6
(c 0.80, MeOH) IR νmax 3420, 1642, 1562, 1451, 1375 cm−1; 1H NMR
(pyridine-d5, 500 MHz) data, see Table 1; 13C NMR (pyridine-d5, 125
MHz) data, see Table 3; ESIMS m/z 1055 [M + Na]+; HRESIMS m/z
1055.5735 [M + Na]+ (calcd for C52H88O20Na, 1055.5767).
(3β,20S)-19-Oxo-3,20,21-trihydroxydammar-24-ene 3-O-β-D-xy-
lopyranosyl-(1→3)-α-L-arabinopyranoside (10): amorphous powder;
[α]20 +20.9 (c 0.85, MeOH); IR νmax 3379, 1700, 1646, 1594, 1443,
D
1
1377 cm−1; H NMR (pyridine-d5, 400 MHz) data, see Table 2; 13C
NMR (pyridine-d5, 100 MHz) data, see Table 3; ESIMS m/z 761
[M + Na]+; HRESIMS m/z 761.4424 [M + Na]+ (calcd for
C40H66O12Na, 761.4452).
(3β,20S)-19-Oxo-3,20,21-trihydroxydammar-24-ene 3-O-{[α-L-
rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyr-
anosyl}-21-O-β-D-glucopyranoside (11): amorphous powder; [α]20
D
1
+2.2 (c 0.38, MeOH); IR νmax 3406, 1702, 1647, 1448, 1378 cm−1; H
NMR (pyridine-d5, 500 MHz) data, see Table 2; 13C NMR (pyridine-d5,
125 MHz) data, see Table 3; ESIMS m/z 1099 [M + Na]+; HRESIMS
m/z 1099.5642 [M + Na]+ (calcd for C53H88O22Na,1099.5665).
(3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-[α-L-rhamno-
pyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylgluco-
pyranoside (12): amorphous powder; [α]20 −5.3 (c 0.32, MeOH);
D
1380
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
to the Boltzmann distribution theory, in which the relevant Gibbs free
energies (G) were adopted.
Zhang, Y.; Zi, J.; Song, W.; Fan, X.; Liu, Y.; Yang, Y.; Shi, J. Org. Lett.
2011, 13, 2856−2859. (c) Liu, M.; Lin, S.; Gan, M.; Chen, M.; Li, L.;
Wang, S.; Zi, J.; Fan, X.; Liu, Y.; Si, Y.; Yang, Y.; Chen, X.; Shi, J. Org.
Lett. 2012, 14, 1004−1007.
(4) Gan, M.; Liu, M.; Liu, B.; Lin, S.; Zhang, Y.; Zi, J.; Song, W.; Ye,
F.; Chen, X.; Shi, J. J. Nat. Prod. 2011, 74, 2431−2437.
(5) (a) Hara, S.; Okabe, H.; Mihashi, K. Chem. Pharm. Bull. 1987, 35,
501−506. (b) Kinjo, J.; Araki, K.; Fukui, K.; Higuchi, H.; Ikeda, T.;
Nohara, T.; Ida, Y.; Takemoto, N.; Miyakoshi, M.; Shoji, J. Chem.
Pharm. Bull. 1992, 42, 3269−3273.
(6) (a) Piacente, S.; Pizza, C.; De Tommasi, N.; De Simone, F. J. Nat.
Prod. 1995, 58, 512−519. (b) Yin, F.; Hu, L.-H. Helv. Chim. Acta 2005,
88, 1126−1134. (c) Diome, C.; Mitaine-Offer, A.-C.; Miyamoto, T.;
Delaude, C.; Mirjolet, J.-F.; Duchamp, O.; Lacaille-Dubois, M.-A. J.
Nat. Prod. 2007, 70, 1680−1682.
(7) (a) Ding, Y.; Li, X.-C.; Ferreira, D. J. Org. Chem. 2007, 72, 9010−
9017. (b) Gan, L.-S.; Zheng, Y.-L.; Mo, J.-X.; Liu, X.; Li, X.-H.; Zhou,
C.-X. J. Nat. Prod. 2009, 72, 1497−1501. (c) Li, X. C.; Ferreira, D.;
Ding, Y. Curr. Org. Chem. 2010, 14, 1678−1697. (d) Gan, M.; Zheng,
X.; Gan, L.; Guan, Y.; Hao, X. Q.; Liu, Y.; Si, S.; Zhang, Y.; Yu, L.;
Xiao, C. J. Nat. Prod. 2011, 74, 1142−1147. (e) Neff, S. A.; Lee, S. U.;
Asami, Y.; Ahn, J. S.; Oh, H.; Baltrusaitis, J.; Gloer, J. B.; Wicklow, D.
T. J. Nat. Prod. 2012, 75, 464−472.
(8) (a) Shi, L.; Cao, J.-Q.; Li, W.; Zhao, Y.-Q. Helv. Chim. Acta 2010,
93, 1785−1794. (b) Xu, J.-Q.; Shen, Q.; Li, J.; Hu, L.-H. Biorg. Med.
Chem. 2010, 18, 3934−3939.
(9) (a) Asakawa, J.; Kasai, R.; Yamasaki, K.; Tanaka, O. Tetrahedron
1977, 33, 1935−1939. (b) Junxian, W.; Liangyu, C.; Jufen, W.;
Friedrichs, E.; Jores, M.; Puff, H.; Chen Wei, s.; Breitmaier, E. Planta
Med. 1982, 45, 167−171. (c) Tori, M.; Matsuda, R.; Sono, M.;
Asakawa, Y. Magn. Reson. Chem. 1988, 26, 581−590.
(10) Hung, T. M.; Hoang, D. M.; Kim, J. C.; Jang, H.-S.; Ahn, J. S.;
Min, B.-S. J. Ethnopharmacol. 2009, 124, 240−245.
(11) (a) Yin, F.; Hu, L.; Pan, R. Chem. Pharm. Bull. 2004, 52, 1440−
1444. (b) Yin, F.; Hu, L.; Lou, F.; Pan, R. J. Nat. Prod. 2004, 67, 942−
952.
(12) (a) Takemoto, T.; Arihara, S.; Yoshikawa, K.; Hino, K.;
Nakajima, T.; Okuhira, M. Yakugaku Zasshi 1984, 104, 1155−1162.
(b) Kizu, H.; Koshijima, M.; Tomimori, T. Chem. Pharm. Bull. 1985,
33, 3176−3181. (c) Shi, L.; Cao, J. Q.; Li, W.; Zhao, H.; Zhao, Y. Q.
Chin. Chem. Lett. 2010, 21, 699−701.
(13) (a) Frelek, J.; Geiger, M.; Voelter., W. Curr. Org. Chem. 1999, 3,
117−146. (b) Di Bari, L.; Pescitelli, G.; Pratelli, C.; Pini, D.; Salvadori,
P. J. Org. Chem. 2001, 66, 4819−4825. (c) Frelek, F.; Klimek, A.;
Ruskowska, P. Curr. Org. Chem. 2003, 7, 1081−1104.
Conformational analyses in the MMFF94 force field showed three
and eight lowest energy conformers for 1b and the (20R,23R)-isomer,
respectively, whose relative energies were within 2.0 kcal/mol
(Supporting Information, Figure S4). The conformers were reopti-
mized using DFT at the B3LYP/6-31G(d) level in the gas phase. The
energies, oscillator strengths, and rotational strengths of the first four
electronic excitations for 1b and the first two electronic excitations for
the (20R,23R)-isomer were calculated using TDDFT methodology at
the B3LYP/6-311++G(2d,2p) level in the gas phase, and the ECD
spectra were simulated by the GaussSum 2.25 software (σ = 0.6 eV).
The final spectra of 1b and the (20R,23R)-isomer were obtained by
averaging the corresponding spectra according to their relative
conformational Gibbs free energies
PTP1B Inhibition Assay. See ref 19.
Cells, Culture Conditions, and Cell Proliferation Assay. See
ref 20.
TNF-α Secretion Inhibition Assay. See ref 21.
ASSOCIATED CONTENT
* Supporting Information
■
S
NMR data for compounds 1a and 1b (Table S1) and for 4−7
and 8a (Tables S2 and S3). ECD spectra calculation details of
1a and the (20R,23S)-isomer and 1b and the (20R,23R)-
isomer. Copies of MS, ECD, IR, and NMR spectra of
compounds 1−13. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Tel: 86-10-83154789. Fax: 86-10-63017757. E-mail: lsznn@
imm.ac.cn, shijg@imm.ac.cn.
■
Author Contributions
§These authors contributed equally to this study.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Professor J. Zhang (Institute of Medical Science,
School of Medicine, Shanghai Jiaotong University) for his kind
help on the calculation of ECD spectra. Financial support from
the National Natural Sciences Foundation of China (NNSFC;
grant nos. 30825044, 81102341, 20932007, and 21132009), the
Program for Changjiang Scholars and Innovative Research
Team in University (PCSIRT, grant no. IRT1007), and the
National Science and Technology Project of China (no.
2011ZX09307-002-01) is acknowledged.
(14) (a) Wu, X.; Lin, S.; Zhu, C.; Yue, Z.; Yu, Y.; Zhao, F.; Liu, B.;
Dai, J.; Shi, J. J. Nat. Prod. 2010, 73, 1294−1300. (b) Liu, J.; Liu, Y.; Si,
Y.; Yu, S.; Qu, J.; Xu, S.; Hu, Y.; Ma, S. Steroids 2009, 74, 51−61.
(c) Dong, S.-H.; Zhang, C.-R.; Dong, L.; Wu, Y.; Yue, J.-M. J. Nat.
Prod. 2011, 74, 1042−1048.
(15) Frelek, J.; Snatzke, G. Fresenius J. Anal. Chem. 1983, 316, 261−
264.
(16) (a) Frelek, J.; Ruskowska, P.; Suszczynska, A.; Szewczyk, K.;
Osuch, A.; Jarosz, S.; Jagodzinski, J. Tetrahedron: Asymmetry 2008, 19,
REFERENCES
■
(1) (a) Jiangsu New Medical College. Dictionary of Traditional
Chinese Medicine; Shanghai Science and Technology Publishing
House: Shanghai, 1977; pp 114, 1009, and 1423. (b) Cheng, M.-J.;
Tsai, I.-L.; Lee, S.-J.; Jayaprakasam, B.; Chen, I.-S. Phytochemistry
2005, 66, 1180−1185. (c) Park, E. Y.; Shin, S. M.; Ma, C. J.; Kim, Y.
C.; Kim, S. G. Planta Med. 2005, 71, 393−398. (d) Giang, P. M.;
Son, P. T.; Matsunami, K.; Otsuka, H. Chem. Pharm. Bull. 2006, 54,
380−383.
(2) (a) Cheng, W.; Zhu, C.; Xu, W.; Fan, X.; Yang, Y.; Li, Y.; Chen,
X.; Wang, W.; Shi, J. J. Nat. Prod. 2009, 72, 2145−2152. (b) Li, Y.;
Cheng, W.; Zhu, C.; Yao, C.; Xiong, L.; Tian, Y.; Wang, S.; Lin, S.; Hu,
J.; Yang, Y.; Guo, Y.; Yang, Y.; Li, Y.; Yuan, Y.; Chen, N.; Shi, J. J. Nat.
Prod. 2011, 74, 1444−1452.
(3) (a) Liu, M.; Lin, S.; Wang, Y.; He, W. e.; Li, S.; Wang, S.; Yang,
Y.; Shi, J. Org. Lett. 2007, 9, 129−132. (b) Liu, M.; Gan, M.; Lin, S.;
1709−1713. (b) Go
́
recki, M.; Jabłon
́
ska, E.; Kruszewska, A.;
Suszczynska, A.; Urban
́
́
W.; Szczepek, W. J.; Frelek, J. J. Org. Chem. 2007, 72, 2906−2916.
(17) (a) Chen, J.-C.; Lu, K.-W.; Lee, J.-H.; Yeh, C.-C.; Chung, J.-G.
Anticancer Res. 2006, 26, 4313−4326. (b) Wang, Q.-F.; Chiang, C.-W.;
Wu, C.-C.; Cheng, C.-C.; Hsieh, S.-J.; Chen, J.-C.; Hsieh, Y.-C.; Hsu,
S.-L. Planta Med. 2007, 73, 535−544.
(18) Ky, P. T.; Huong, P. T.; My, T. K.; Anh, P. T.; Kiem, P. V.; Van
Minh, C.; Cuong, N. X.; Thao, N. P.; Nhiem, N. X.; Hyun, J.-H.; Kang,
H.-K.; Kim, Y. H. Phytochemistry 2010, 71, 994−1001.
(19) Wang, Y.; Shang, X. Y.; Wang, S. J.; Mo, S. Y.; Li, S.; Yang, Y. C.;
Ye, F.; Shi, J. G.; He, L. J. Nat. Prod. 2007, 70, 296−299.
(20) Mo, S.; Wang, S.; Zhou, G.; Yang, Y.; Li, Y.; Chen, X.; Shi, J. J.
Nat. Prod. 2004, 67, 823−828.
1381
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products
Article
(21) Lin, S.; Wang, S.; Liu, M.; Gan, M.; Li, S.; Yang, Y.; Wang, Y.;
He, W.; Shi, J. J. Nat. Prod. 2007, 70, 817−823.
(22) Spartan 08; Wavefunction, Inc.: Irvine, CA.
(23) Gaussian 09, Revision A.1; Gaussian, Inc.: Wallingford, CT,
2009. A full list of authors can be found in the Supporting Information.
(24) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput.
Chem. 2008, 29, 839−845.
1382
dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382