New neolignan component from camellia amplexicaulis and effects on osteoblast differentiation

Article abstract of DOI:10.1248/cpb.57.65

A new neolignan named "camellioside A" (1) has been isolated from the leaves and branches of Camellia amplexicaulis, along with seven known glycosides (2-8). Their structures were determined by a variety of spectroscopic analyses. Among them, compounds 2,

Full text of DOI:10.1248/cpb.57.65

January 2009
Notes
Chem. Pharm. Bull. 57(1) 65—68 (2009)
65
New Neolignan Component from Camellia amplexicaulis and Effects on
Osteoblast Differentiation
Nguyen Huu TUNG,a,b Yan DING,a Eun Mi CHOI,a Chau Van MINH,b and Young Ho KIM*,a
a College of Pharmacy, Chungnam National University; Daejeon 305–764, Korea: and b Institute of Natural Products
Chemistry, Vietnamene Academy of Science and Technology; 18 Hoang Quoc Viet, Nghiado, Caugiay, Hanoi, Vietnam.
Received May 26, 2008; accepted October 28, 2008; published online October 30, 2008
A new neolignan named “camellioside A” (1) has been isolated from the leaves and branches of Camellia
amplexicaulis, along with seven known glycosides (2—8). Their structures were determined by a variety of spec-
troscopic analyses. Among them, compounds 2, 3, 4, and 6 significantly (pꢄ0.05) increased the alkaline phos-
phatase activity and the mineralization of the nodules of the MC3T3-E1 osteoblastic cells compared to those of
the control.
Key words Camellia amplexicaulis; Theaceae; neolignan; camellioside A; osteoblast differentiation; MC3T3-E1 cell
Osteoporosis is a major health concern for aging commu- O-b-D-glucopyranoside (2),5) 7R,8R-dihydrodehidrodiconife-
nities. The associated progressive decrease in bone mass ryl alcohol-9ꢀ-O-b-D-glucopyranoside (3),6) urolignoside
leads to an increased susceptibility to fractures, which results (4),7) junipetrioloside A (5),8) isolarciresinol (6),9) corchoio-
in substantial morbidity and mortality.1) Bone tissue is con- noside C (7),10) and blumeon C glycoside (8).11) This paper
stantly being resorbed and rebuilt in a coupled process describes the isolation method, the structure elucidation of
known as remodeling. The bone remodeling process is con- camellioside A, and the evaluation of the effects of com-
trolled by the rates of bone formation and bone resorption.2) pounds 1—8 on osteoblast differentiation in MC3T3-E1 cells.
A rapid rate of bone resorption that exceeds the rate of bone
formation results in bone abnormalities, such as osteoporo- Results and Discussion
sis. Since many osteoporotic patients have already lost a sub-
stantial amount of bone, a method of increasing bone mass Its HR-ESI-MS showed a [MꢁH2OꢂNa]ꢂ peak at m/z
by stimulating new bone formation is required.3)
531.1824 (Calcd for C25H32O11Na, 531.1842), corresponding
Compound 1 was obtained as a white amorphous powder.
In our ongoing study to find active natural compounds to to the molecular formula C25H34O12Na lost of a water mole-
1
differentiate MC3T3-E1 osteoblastic cells, potent activity was cule. The H- and 13C-NMR spectra of 1 displayed typical
found in the MeOH extract of the aerial parts of Camellia features that were similar to those of 2, which was first iso-
amplexicaulis (Theaceae). This plant is only really known for lated from Lonicera gracilipes var. glandulosa MAXIM.5) This
its flowers, which are a favorite choice for New Year decora- suggests that 1 was a 8-O-4ꢀ neolignan-type glycoside. The
tions. To date, there appears to have been no study that fo- 1H-NMR (CD3OD) spectrum of 1 showed six aromatic pro-
cussed on the phytochemistry and bioactivity of this plant.4) tons [d 6.68 (dd, Jꢃ8.0, 2.0 Hz), 6.75 (d, Jꢃ2.0 Hz), 6.81 (d,
Based on a bioactivity-guided fractionation and isolation Jꢃ8.0 Hz), 6.83 (d, Jꢃ8.0 Hz), 6.87 (dd, Jꢃ8.0, 2.0 Hz),
method, one new neolignan glycoside named “camellioside 6.97 (d, Jꢃ2.0 Hz)] revealing two ABX proton systems of
A” (1) has been isolated from the leaves and branches of C. two 1,3,4-trisubstituted aromatic rings, one methoxy proton
amplexicaulis, along with seven known compounds including at d 3.84 (s), two methylen protons [d 1.87 (m, H-8ꢀ), 2.60
7S,8S-4,9,9ꢀ-trihydroxy-3,3ꢀ-dimethoxy-8-O-4ꢀ-neolignan-7- (t, Jꢃ6.4 Hz, H-7ꢀ)], two methine protons [d 4.84 (over-
Fig. 1. Structures of 1—8
∗ To whom correspondence should be addressed. e-mail: yhk@cnu.ac.kr
© 2009 Pharmaceutical Society of Japan

66
Vol. 57, No. 1
lapped, H-7), 3.98 (1H, m, H-8)], and a glucose anomeric based on the CD spectrum the stereochemistry of C-7 and C-
proton at d 4.22 (d, Jꢃ7.6 Hz, H-1ꢅ). However, in CD3OD, 8 were both assigned to be R according to the study of re-
the proton signal H-7 of 1 was buried in the solvent signal. lated structures.5,14—16) The CD spectrum of 1 showed a
1
Thereby, H-NMR spectrum of 1 was checked in the other strong negative Cotton effect at 233 nm (De ꢁ14.45), oppo-
NMR solvent, DMSO-d6; in which, the proton signal H-7 site of that of 7S,8S-4,7,9,9ꢀ-tetrahydroxy-3,3ꢀ-dimethoxy-8-
was appeared as a doublet at d 4.96 (Jꢃ4.8 Hz), consider- O-4ꢀ-neolignan-9ꢀ-O-b-D-glucopyranoside which reported
ately. The 13C-NMR (CD3OD) spectrum of 1 showed 25 car- recently from Iodes cirrhosa.14) Based on these evidence,
bon signals, including one methoxy carbon at d 55.3, 18 car- compound 1 has been identified as 7R,8R-3ꢀ,4,7,9-tetrahy-
bons of two phenyl propanoid units forming neolignan, and droxy-3-methoxy-8-O-4ꢀ-neolignan-9ꢀ-O-b-D-glucopyra-
six carbons of an O-glucose unit (d 103.2, 74.0, 76.9, 70.5, noside, which has been named “camellioside A”.
76.7, 61.6).12) In addition, it was possible to assign the car-
To evaluate the effect of the isolated compounds on
bon signals corresponding to each proton signal using the MC3T3-E1 differentiation, the alkaline phosphatase (ALP)
obtained HMQC spectrum. Furthermore, the partial struc- activity and calcium deposition were examined. Osteoblastic
tures of 1 were confirmed from COSY and HMBC NMR MC3T3-E1 cells, at confluence, were cultured with a differ-
spectra; key correlations for these experiments are illustrated entiation-inducing medium, and then incubated in a medium
in Fig. 2. In the COSY spectrum, correlations of H-5/H-6, H- containing the individual isolated compounds. Among the
7/H-8, H-8/H-9, H-5ꢀ/H-6ꢀ, H-7ꢀ/H-8ꢀ, and H-8ꢀ/H-9ꢀ were eight compounds tested, compounds 2, 3, 4, and 6 sig-
obsevered. The HMBC spectrum revealed correlations of H- nificantly increased the ALP activity. Compounds 3 and
7/C-1,2; H-8/C-4ꢀ; H-7ꢀ/C-1ꢀ,2ꢀ. Moreover, further analysis 4 showed the most potent activity. At concentrations of
of the HMBC spectrum found the correlations from 0.01 mM, compounds 3 and 4 increased the ALP activity up
anomeric proton H-1ꢅ to C-9ꢀ at d 68.7 and H-9ꢀ to C-1ꢅ at d to 152.2 and 142.3%, respectively, compared to that of the
103.2, indicating the linkage position with glucose unit at C- control and stronger than the positive control, 17b-estradiol.
9ꢀ. The mode of glycosidic linkage were determined to be b Compounds 1, 5, 7, and 8 showed comparatively weaker ef-
form due to the large coupling constant of the anomeric pro- fect (Table 1). The effect of the isolated compounds on min-
ton doublet at d 4.22 (Jꢃ7.6 Hz).
eralization was then examined by measuring the calcium
Enzymatic hydrolysis of 1 gave D-glucose (determined by deposition by Alizarin Red staining. As was found for the
1
a GC experiment) and an aglycone (1a). In the H-NMR ALP activity study, compounds 2, 3, 4, and 6 showed signifi-
spectrum of 1a in CDCl3, the proton H-7 was a doublet with cant stimulatory effects on mineralization (Table 2). They in-
a large coupling constant (Jꢃ8.0 Hz) at d 4.92. Braga et al. creased the mineralization to 115, 118, 124, and 113%, re-
reported that large and small J values for H-7 and H-8 of 8- spectively, compared to that of the control at a concentration
O-4ꢀ neolignan diastereoisomers correspond to the threo and of 10 mM. When the cell viability was measured using the
erythro forms, respectively, in terms of the possible staggered MTT assay, no toxic effects were observed after incubation
conformers with intramolecular hydrogen bonding of the of the cells with 0.01 to 10 mM of the eight compounds (data
benzylic hydroxyl and aryloxyl groups.13) Accordingly, 1 was not shown). This result indicated that the compounds were
determined to have relative threo configuration. Furthermore, not toxic at the concentrations used in this study.
Of the eight isolated compounds, compounds 1—4, and 6
belong to the lignan group. As the results, lignans 2, 3, 4, and
6 showed more significant stimulatory effects in ALP activity
and calcium deposition than did the rest of the compounds.
Recently, lignans have received considerable attention due to
their noteworthy role in stimulating osteoblast differentiation
both in in vitro and in vivo studies.17) For instance, lignans
from Machilus thunbergii SIEB. et ZUCC. (Lauraceae) were
found to increase osteoblast differentiation in primary cul-
tures of mouse osteoblasts.18) Another study reported that
Fig. 2. Selected HMBC and COSY Correlations of 1
secoisolariciresinol diglycoside from flaxseed showed anti-
Table 1. Effects of Compounds 1—8 on the ALP Activity of MC3T3-E1 Cells
ALP activity (% of control)
Compounds
0.01 mM
0.1 mM
1.0 mM
10 mM
1
2
3
4
5
6
7
8
113.5ꢆ5.5*
110.7ꢆ7.3
152.2ꢆ8.8*
142.3ꢆ3.7*
129.6ꢆ15.1
137.4ꢆ14.3*
119.7ꢆ2.6
103.6ꢆ1.8
123.0ꢆ4.7*
112.0ꢆ13.6
117.9ꢆ5.1*
121.0ꢆ8.1*
136.9ꢆ12.9*
102.9ꢆ2.9
137.9ꢆ12.4*
111.4ꢆ6.2
110.1ꢆ8.8
100.9ꢆ4.3
126.2ꢆ18.3*
120.8ꢆ3.1*
103.9ꢆ5.8
121.0ꢆ12.1*
109.9ꢆ9.9
108.3ꢆ1.8
114.0ꢆ3.3*
104.7ꢆ3.1
102.6ꢆ1.9
114.8ꢆ13.2
101.6ꢆ4.0
112.5ꢆ2.4*
134.8ꢆ12.9
121.0ꢆ15.4
129.6ꢆ20.2*
113.3ꢆ5.1*
98.0ꢆ3.1
17b-Estradiola)
114.0ꢆ6.5*
Data are the meanꢆS.D. (nꢃ5), expressed as a percentage of the control. The control ALP activity was 0.193ꢆ0.001 Unit/106 cells. a) Positive control. ∗ pꢄ0.05 vs. control.

January 2009
67
Table 2. Effects of Compounds 1—8 on the Mineralization of MC3T3-E1 Cells
Mineralization (% of control)
Compounds
0.01 mM
0.1 mM
1.0 mM
10 mM
1
2
3
4
5
6
7
8
104.7ꢆ5.3
100.3ꢆ3.9
100.5ꢆ4.6
107.8ꢆ2.1*
100.5ꢆ1.7
101.1ꢆ1.5
101.4ꢆ1.2
105.9ꢆ3.9
126.9ꢆ6.0*
103.5ꢆ5.4
105.2ꢆ1.5
105.9ꢆ2.1
107.1ꢆ2.2*
101.3ꢆ1.9
100.9ꢆ1.2
102.7ꢆ1.3
101.6ꢆ2.3
124.3ꢆ7.7*
102.7ꢆ4.0
106.3ꢆ1.8
104.2ꢆ2.0
113.4ꢆ3.4*
105.2ꢆ3.5*
106.2ꢆ2.5
104.7ꢆ2.2
101.9ꢆ3.4
120.5ꢆ8.7*
109.7ꢆ6.3
115.0ꢆ3.5*
117.7ꢆ1.6*
124.0ꢆ5.8*
100.8ꢆ2.1
113.1ꢆ3.6*
103.4ꢆ1.7
105.4ꢆ4.2
114.6ꢆ12.8
17b-Estradiola)
Data are the meanꢆS.D. (nꢃ5), expressed as a percentage of the control. The control mineralization value was 0.302ꢆ0.011 O.D. a) Positive control. ∗ pꢄ0.05 vs. control.
HR-ESI-MS m/z: 531.1824 [MꢁH2OꢂNa]ꢂ (Calcd for C25H32O11Na,
osteoporotic activity in postmenopausal women.19) Also, iso-
531.1842); 1H-NMR (CD3OD, 400 MHz) d: 1.87 (2H, m, H-8ꢀ), 2.60 (2H, t,
taxiresinol from Taxus yunnanensis has been found to pre-
Jꢃ6.4 Hz, H-7ꢀ), 3.16 (1H, dd, Jꢃ7.6, 8.4 Hz, glc H-2ꢅ), 3.25—3.38 (3H,
overlapped, glc H-3ꢅ,4ꢅ,5ꢅ), 3.44 (1H, dd, Jꢃ4.8, 12.4 Hz, H-9), 3.52 (1H,
vent bone loss in an ovariectomized model.20) This current
study suggests that, among the compounds obtainable from
C. amplexicaulis, the lignans play an important role in pro-
moting MC3T3-E1 osteoblast cell differentiation in vitro.
m, H-9ꢀ), 3.63 (1H, dd, Jꢃ6.4, 12.4 Hz, H-9), 3.65 (1H, dd, Jꢃ4.0, 12.4 Hz,
glc H-6ꢅ), 3.82 (1H, m, glc H-6ꢅ), 3.84 (3H, s, 3-OCH3), 3.88 (1H, m, H-9ꢀ),
3.98 (1H, m, H-8), 4.22 (1H, d, Jꢃ7.6 Hz, glc H-1ꢅ), 4.84 (1H, overlapped,
H-7), 6.68 (1H, dd, Jꢃ2.0, 8.0 Hz, H-6ꢀ), 6.75 (1H, d, Jꢃ2.0 Hz, H-2ꢀ), 6.81
(1H, d, Jꢃ8.0 Hz, H-5), 6.83 (1H, d, Jꢃ8.0 Hz, H-5ꢀ), 6.87 (1H, dd, Jꢃ2.0,
1
Experimental
8.0 Hz, H-6), 6.97 (1H, d, Jꢃ2.0 Hz, H-2); H-NMR (DMSO-d6, 400 MHz)
General Experimental Procedures Optical rotations were determined
using a JASCO DIP-360 digital polarimeter. UV spectra were recorded
using a Beckman Du-650 UV–VIS recording spectrometer. CD spectra were
recorded using a JASCO J-700 spectropolarimeter. IR spectra were meas-
ured by a Perkin-Elmer 577 spectrometer. NMR spectra were recorded on
Bruker DRX 400 and 500 NMR spectrometers. The HR-ESI-MS spectra
were obtained using a JMS-T100TD spectrometer with an electrospray ion
source (Tokyo, Japan). Column chromatography (CC) was performed on sil-
ica gel (70—230 and 230—400 mesh, Merck), YMC RP-18 resins (30—
50 mm, Fuji Silysia Chemical Ltd.), and HP-20 dianion (Mitshubishi Chemi-
cal Corporation). Thin layer chromatography (TLC) was performed on
Kieselgel 60 F254 (Merck 1.05715) or RP-18 F254s (Merck) plates. Spots were
visualized by spraying the plates with 10% aqueous H2SO4 solution and
heating them for 5 min.
d: 1.76 (2H, m, H-8ꢀ), 2.32 (2H, t, Jꢃ6.4 Hz, H-7ꢀ), 3.76 (3H, s, 3-OCH3),
4.04 (1H, m, H-8), 4.12 (1H, d, Jꢃ7.6 Hz, glc H-1ꢅ), 4.96 (1H, d, Jꢃ4.8 Hz,
H-7), 6.68—6.98 (6H, aromatic protons); and 13C-NMR (CD3OD, 100 MHz)
d: 128.5 (C-1), 110.8 (C-2), 147.1 (C-3), 55.26 (3-OCH3) 141.8 (C-4), 115.1
(C-5), 120.5 (C-6), 76.5 (C-7), 78.6 (C-8), 61.00 (C-9), 135.3 (C-1ꢀ), 116.6
(C-2ꢀ), 143.8 (C-3ꢀ), 148.0 (C-4ꢀ), 116.4 (C-5ꢀ), 121.3 (C-6ꢀ), 31.2 (C-7ꢀ),
31.5 (C-8ꢀ), 68.7 (C-9ꢀ), 103.2 (glc C-1ꢅ), 74.0 (glc C-2ꢅ), 76.9 (glc C-3ꢅ),
70.5 (glc C-4ꢅ), 76.7 (glc C-5ꢅ), 61.6 (glc C-6ꢅ).
Enzymatic Hydrolysis b-Glucosidase (10 mg, from almond) was added
to a suspension of 1 (2 mg) in 50 mM phosphate buffer (5 ml) adjusted to pH
7.0 by NaOH. The resulting mixture was stirred at 37 °C for 48 h. The reac-
tion mixture was extracted with EtOAc (5 mlꢇ3), and the organic layer was
evaporated in vacuo. The residue was subjected to preparative TLC with
CHCl3–MeOH (10 : 1) to afford aglycone 1a (1 mg). The sugar residue, ob-
tained by concentration of the water layer, was dissolved in dry pyridine
(0.1 ml). Then L-cysteine methyl ester hydrochloride in pyridine (0.06 M,
0.1 ml) was added to the solution. After heating the reaction mixture at
60 °C for 2 h, 0.1 ml of trimethylsilylimidazole was added. Heating at 60 °C
was continued for a further 2 h. The dried product was partitioned between
hexane and H2O. The hexane layer was analyzed by GC (Shimadzu-2010)
using a DB-05 capillary column (0.5 mm IDꢇ30 ml) (column temperature:
210 °C; detector temperature: 300 °C; injector temperature: 270 °C; He gas
flow rate: 30 ml/min).21) The peak of the hydrolysate of 1 was detected at tR
14.12 min for D-glucose. The retention times for the authentic samples
(Sigma), after being treated in the similar manner, were 14.12 min (D-glu-
cose) and 14.25 min (L-glucose). Co-injection of the hydrolysates of 1 with
Plant Material The leaves and branches of C. amplexicaulis were col-
lected in Tamdao National Park, Vinhphuc province, Vietnam, in April
2006. They were identified by Dr. Tran Huy Thai, Institute of Ecology and
Biological Resources, VAST. A voucher specimen (No. 20060408) was de-
posited at the Institute of Natural Products Chemistry, VAST.
Extraction and Isolation The air-dried sample of C. amplexicaulis
(5 kg) were extracted with methanol (10 lꢇ3 times). Evaporation of the com-
bined MeOH extracts in vacuo gave a residue (300 g), which was then parti-
tioned between CH2Cl2 and H2O. The H2O layer was fractionated over a Di-
anion HP-20 column eluted with H2O and MeOH–H2O (20%, 50%, 70%,
and 100% MeOH) to give four fractions: 1A (9 g), 1B (18 g), 1C (40 g), and
1D (60 g). Fractions 1A and 1B were combined due to their similar TLC be-
haviour. They were then chromatographed over a silica gel column with
CH2Cl2–MeOH (10 : 1—0 : 1), gradiently, to give five fractions (2A—E).
Fraction 2C (8 g) was rechromatographed on a silica gel column with
CHCl3–MeOH–H2O (70 : 30 : 4) to give six subfractions (3A—F). Com-
pound 7 (12 mg) was obtained from subfraction 3A (30 mg) by silica gel col-
umn chromatography with CHCl3–MeOH–H2O (5 : 1 : 0.1). Fraction 2D
(9 g) was chromatographed on a YMC RP-18 column (MeOH–H2O, 4 : 1) to
afford 5 (100 mg). Fraction 1D was subjected to silica gel column chro-
matography (CHCl3–MeOH, 10 : 1—0 : 1) to give ten fractions (4A—J).
Fraction 4A was purified using a YMC RP-18 column (MeOH–H2O, 1 : 1.5)
to give 8 (8.3 mg). Fraction 4D (3 g) was rechromatographed on a YMC RP-
18 column (MeOH–H2O, 1 : 4) to give 1 (30 mg) and 3 (10 mg). Fraction 4H
(300 mg) was rechromatographed on a YMC RP-18 column (MeOH–H2O,
1 : 2) to give 4 (5.2 mg). Fraction 4I was rechromatographed on a YMC RP-
18 column (MeOH–H2O, 1 : 1) to afford 2 (3.3 mg). Fraction 4J was rechro-
matographed on a YMC RP-18 column (MeOH–H2O, 1 : 2) to give 6
(5.5 mg).
1
standard D-glucose gave single peaks. 1a: H-NMR (400 MHz, CDCl3) d:
1.90 (2H, m, H-8ꢀ), 2.64 (2H, t, Jꢃ7.2 Hz, H-7ꢀ), 3.86 (3H, s, OCH3), 4.02
(1H, m, H-8), 4.92 (1H, d, Jꢃ8.0 Hz, H-7), 6.60—7.00 (6H, aromatic pro-
tons).
Cell Culture MC3T3-E1 cells (RCB1126, an osteoblast-like cell line
from C57BL/6 mouse calvaria) were obtained from the RIKEN Cell Bank
(Tsukuba, Japan), and cultured at 37 °C in a 5% CO2 atmosphere, in an a-
modified minimal essential medium (a-MEM; GIBCO). Unless otherwise
specified, the medium contained 10% heat-inactivated fetal bovine serum
(FBS), 100 U/ml penicillin, and 100 mg/ml streptomycin.
Alkaline Phosphatase Activity The cells were treated, at 90% conflu-
ence, with a culture medium containing 10 mM b-glycerophosphate and
50 mg/ml ascorbic acid to initiate in vitro mineralization. The medium was
changed every 2—3 d. After 6 d, the cells were cultured with a medium con-
taining 0.3% bovine serum albumin (BSA) and the isolated compounds for
3 d. On harvesting, the medium was removed and the cell monolayer was
gently washed twice with phosphate buffered saline (PBS). The cells were
lysed with 0.2% Triton X-100, with the lysate centrifuged at 14000ꢇg for
5 min. The clear supernatant was used to measure the ALP activity, which
was determined using an ALP activity assay kit (Asan Co., Korea).
Camellioside
A
White amorphous powder; [a]D25 ꢂ14.8 (cꢃ0.3,
MeOH); IR nmax cmꢁ1 (KBr): 3415, 1605, 1514, 1276, 1032; UV lmax nm
(log e) (MeOH): 227 (4.2), 277 (3.8); CD nm (De) (MeOH): 233 (ꢁ14.5);

68
Vol. 57, No. 1
Calcium Deposition Assay The cells were treated, at 90% confluence,
1009—1014 (2006).
with a culture medium containing 10 mM b-glycerophosphate and 50 mg/ml
ascorbic acid to initiate in vitro mineralization. After 10 d, the cells were cul-
tured with a medium containing 0.3% BSA and the isolated compounds in-
dividually for 3 d. On harvesting, the cells were fixed with 70% ethanol for
1 h, and then stained with 40 mM Alizarin Red S for 10 min with gentle shak-
ing. To quantify the bound dye, the stain was solubilized with 10%
7) Shen Y. C., Hsieh P. W., Kuo Y. H., Phytochemistry, 48, 719—723
(1998).
8) Comte G., Vercauteren J., Chulia A. J., Allais D. P., Delage C., Phyto-
chemistry, 45, 1679—1682 (1997).
9) Guyot R. B., Trendel J. M., Albrecht P., Schaeffer A., J. Agric. Food
Chem., 48, 6178—6182 (2000).
cetylpyridinum chloride by shaking for 15 min, while shielded from light. 10) Yoshikawa M., Shimada H., Saka M., Yoshizumi S., Yamahara J., Mat-
The absorbance of the solubilized stain was measured at 561 nm. suda H., Chem. Pharm. Bull., 45, 464—469 (1997).
Statistics The results are expressed as the meanꢆS.D. (nꢃ5). Statistical 11) Takeda Y., Zhang H., Masuda T., Honda G., Otsuka H., Sezik E., Yesi-
analysis was performed using a one-way ANOVA (pꢄ0.05) with SAS statis-
tical software.
lada E., Sun H., Phytochemistry, 44, 1335—1337 (1997).
12) Agrawal P. K., Phytochemistry, 31, 3307—3330 (1992).
13) Braga A., Zacchino S., Badano H., Sierra M. G., Ruveda E. A., Phyto-
chemistry, 23, 2025—2028 (1984).
Acknowledgements This work was supported by the Korea Foundation
for International Cooperation of Science & Technology (KICOS) through a 14) Gan M., Zhang Y., Lin S., Liu M., Song W., Zi J., Yang Y., Fan X., Shi
grant provided by the Korean Ministry of Science & Technology (MOST) in J., Hu J., Sun J., Chen N., J. Nat. Prod., 71, 647—654 (2008).
Korea (No. K2072100000207B010000210). We are greatful to KBSI for 15) Huo C., Liang H., Zhao Y., Wang B., Zhang Q., Phytochemistry, 69,
performing the NMR experiments.
788—795 (2008).
16) Arnoldi A., Merlini L., J. Chem. Soc. Perkin Trans. I, 1985, 2555—
2557 (1985).
References
1) Riggs B. L., Melton L. J. III, N. Engl. J. Med., 327, 620—627 (1992).
2) Roodman G. D., Endocrinol. Rev., 17, 308—332 (1996).
3) Mohan S., Kutilek S., Zhang C., Shen H. G., Kodama Y., Srivastava A.
17) Dixon R. A., Annu. Rev. Plant Biol., 55, 225—261 (2004).
18) Lee M. K., Yang H., Ma C. J., Kim Y. C., Biol. Pharm. Bull., 30,
814—817 (2007).
K., Wergedal J. E., Bermer W. G., Baylink D. J., Bone, 27, 471—478 19) Kim M. K., Chung B. C., Yu V. Y., Nam J. H., Lee H. C., Huh K. B.,
(2000). Lim S. K., Clin. Endocrinol., 56, 321—328 (2002).
4) Ho P. H., “An Illustrated Flora of Vietnam,” Tre Publishing House, 20) Yin J., Tazeka Y., Shi L., Nobukawa M., Nobukawa T., Kadota S., Phy-
Hanoi, 1999.
tomedicine, 13, 37—42 (2006).
5) Matsuda N., Kikuchi M., Chem. Pharm. Bull., 44, 1676—1679 (1996).
6) Dong L. P., Ni W., Li J. Z., Chen C. X., Liu H. Y., Molecules, 11,
21) Min B. S., Nakamura N., Miyashiro H., Kim Y. H., Hattori M., Chem.
Pharm. Bull., 48, 194—200 (2000).