Acylated oleanane-type triterpene saponins with acceleration of gastrointestinal transit and inhibitory effect on pancreatic lipase from flower buds of Chinese tea plant (Camellia sinensis)

Article abstract of DOI:10.1002/cbdv.200800153

The MeOH extract and its BuOH-soluble fraction (crude saponin fraction) from the flower buds of Chinese tea plant (Camellia sinensis (L.) O. Kuntze; Fujian Province) were found to exhibit accelerating effects on gastrointestinal transit in mice and inhibi

Full text of DOI:10.1002/cbdv.200800153

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
903
Acylated Oleanane-Type Triterpene Saponins with Acceleration of
Gastrointestinal Transit and Inhibitory Effect on Pancreatic Lipase from
Flower Buds of Chinese Tea Plant (Camellia sinensis)1)
by Masayuki Yoshikawa*, Sachiko Sugimoto, Yasuyo Kato, Seikou Nakamura, Tao Wang,
Chihiro Yamashita, and Hisashi Matsuda
Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto 607-8412, Japan
(phone: þ81-75-5954633; fax: þ81-75-5954768; e-mail: myoshika@mb.kyoto-phu.ac.jp)
The MeOH extract and its BuOH-soluble fraction (crude saponin fraction) from the flower buds of
Chinese tea plant (Camellia sinensis (L.) O. Kuntze; Fujian Province) were found to exhibit accelerating
effects on gastrointestinal transit in mice and inhibitory effects against pancreatic lipase. From the
BuOH-soluble fraction, three new acylated oleanane-type triterpene oligoglycosides, chakasaponins I, II,
and III (1–3, resp.), were isolated together with 13 known compounds. The chemical structures 1–3 were
elucidated on the basis of chemical and physicochemical evidence. Compounds 1–3 showed accelerating
effects on gastrointestinal transit in mice and inhibitory effects against porcine pancreatic lipase (IC50
150–530 mm).
¼
Introduction. – During the course of our characterization studies on the bioactive
saponin constituents from Camellia (C.) sinensis (Theaceae), we have reported the
isolation and structure elucidation of theasaponins A1 –A7, B1, B5, C1, E1 –E13, F1 –F3,
G1, G2, and H1 [2–7], assamsaponins A–I [8][9], and camelliasaponins B1 and C1 [10]
from the seeds of Japanese C. sinensis (L.) O. Kuntze and Sri Lankan C. sinensis L. var.
assamica Perre, and foliatheasaponins I–V from the leaves of Japanese C. sinensis
[11]. In addition, floratheasaponins A–I [12][13] were isolated from the flower buds of
C. sinensis in Japan and China (Anhui Province), and these acylated polyhydroxy-
oleanane-type triterpene oligoglycosides showed inhibitory effects on serum triglycer-
ide elevation in olive oil-treated mice [12] and the release of b-hexosaminidase from
RBL-2 H3 cells [13]. Recently, floratheasaponins were found to exhibit inhibitory
effects on EtOH- and indomethacin-induced gastric mucosal lesions in rats and on
serum glucose elevation in sucrose-loaded rats [1]. As a continuation of our studies on
bioactive constituents of medicinal flowers [12–19], the MeOH extract from the flower
buds of C. sinensis grown in China (Fujian Province) was found to exhibit accelerating
effect on gastrointestinal transit in mice and inhibitory effect against pancreatic lipase.
We have isolated three new acylated oleanane-type triterpene oligoglycosides,
chakasaponins I, II, and III (1–3, resp.), together with 13 known compounds.
Furthermore, we examined the effects on gastrointestinal transit and the inhibitory
effects against pancreatic lipase of chakasaponins 1–3. Here, we describe the isolation
and structure elucidation of the new saponins 1–3, and their accelerating effects on
gastrointestinal transit as well as the inhibitory effects against pancreatic lipase.
1
)
Medicinal Flowers, Part 19; for Part 18, see [1].
ꢀ 2009 Verlag Helvetica Chimica Acta AG, Zꢁrich

904
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Results and Discussion. – Chemistry. Since the MeOH extract (31.1%) from the
dried flower buds of C. sinensis cultivated in Fujian Province of China showed the
accelerating effect on gastrointestinal transit (Table 1) and the inhibitory effect against
pancreatic lipase (Table 2), it was partitioned with AcOEt/H2O 1 :1 (v/v) to furnish an
AcOEt-soluble fraction (3.2%) and an aqueous layer. The aqueous layer was further
extracted with BuOH to give BuOH- and H2O-soluble fractions (16.4 and 11.5%,
resp.). The BuOH-soluble fraction accelerated gastrointestinal transit, while the
AcOEt- and BuOH-soluble fractions inhibited pancreatic lipase activity, but the H2O-
soluble fraction did not show the effect. The AcOEt-soluble fraction was subjected to
normal-phase and reversed-phase (RP) column chromatographies, and finally HPLC
gave ten known compounds, kaempferol 3-O-b-d-galactopyranoside [20], kaempferol
3-O-b-d-glucopyranoside [21], (þ)-gallocatechin [22], (ꢀ)-epicatechin 3-O-gallate
[23], (ꢀ)-epigallocatechin 3-O-gallate [23], benzyl alcohol [24], (S)-1-phenylethyl b-d-
glucopyranoside [25], trans-cinnamic acid2), protocatechuic acid methyl ester [26], and
caffeine2). Since phenolic compounds such as flavonoids and catechins were known to
show inhibitory effects on lipase [27], the inhibitory activity of the AcOEt-soluble
fraction on pancreatic lipase seems to be due to those phenolic compounds. The BuOH-
soluble fraction was also subjected to normal-phase and RP silica-gel column
chromatographies, and repeated HPLC gave 1, 2, and 3 together with eight known
compounds, (ꢀ)-epicatechin 3-O-gallate [23], (ꢀ)-epigallocatechin 3-O-gallate [23],
(ꢀ)-epicatechin [22], kaempferol 3-O-b-d-glucopyranosyl(1!3)-a-l-rhamnopyranos-
yl(1!6)-b-d-glucopyranoside [28], benzyl b-d-glucopyranoside [29], (S)-1-phenyl-
ethyl b-d-glucopyranoside [25], and caffeine2).
2
)
Those known compounds were identified by comparison of their physical data with those of
commercial compounds.

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
905
Table 1. Effects of the MeOH Extract of Camellia sinensis, and Its BuOH-Soluble Fraction on
Gastrointestinal Transit in Mice
Treatment
Dose [mg/kg], p.o.
n
Gastrointestinal transit [%]a)
Control
MeOH Extract
–
200
500
7
6
6
49.2ꢁ4.1
66.5ꢁ9.2
79.9ꢁ5.4*
Control
BuOH-Soluble fraction
–
7
6
6
7
40.0ꢁ2.5
51.4ꢁ3.9
52.4ꢁ2.4
78.2ꢁ5.7**
50
100
200
a) Mean values ꢁS.E.M. Significantly different from the control group: *: p<0.05, **: p<0.01.
Table 2. Inhibitory Effects of the MeOH Extract of Camellia sinensis, and Its AcOEt-, BuOH-, and H2O-
Soluble Fractions against Pancreatic Lipase
Conc. [mg/ml] Inhibition [%]a)
MeOH Extract AcOEt-Soluble fraction BuOH-Soluble fraction H2O-Soluble fraction
50
100
200
400
800
3.9ꢁ3.8
8.6ꢁ3.5
26.3ꢁ2.4**
48.5ꢁ1.8**
94.5ꢁ3.2**
99.4ꢁ2.3**
101.9ꢁ1.1**
20.4ꢁ17.1
13.8ꢁ8.8
5.5ꢁ12.2
10.8ꢁ14.2
13.8ꢁ13.9
21.6ꢁ15.8
61.3ꢁ2.4**
20.8ꢁ1.7
49.2ꢁ3.4**
87.3ꢁ8.1**
29.5ꢁ8.0
66.3ꢁ11.1**
100.2ꢁ0.3**
IC50
0.40 mg/ml
0.091 mg/ml
0.31 mg/ml
0.69 mg/ml
a) Mean values ꢁS.E.M (n¼4). Significantly different from the control group: **: p<0.01.
Chakasaponin I (1) was isolated as colorless fine crystals from CHCl3/MeOH (m.p.
214.0–216.08) and exhibited a negative optical rotation ([a]2D6 ¼ ꢀ9.0 (c¼0.32,
MeOH)). The IR spectrum of 1 showed absorption bands at 1736, 1718, and
1664 cmꢀ1 due to a,b-unsaturated ester, carboxy, and olefin functions, and strong broad
absorption bands at 3468 and 1078 cmꢀ1 suggesting an oligoglycosidic structure. In the
FAB-MS (negative-ion mode) of 1, a quasi-molecular-ion peak was observed at m/z
1215 ([MꢀH]ꢀ ) in addition to fragment-ion peaks at m/z 1083 ([MꢀHꢀC5H8O4]ꢀ )
and 1053 ([MꢀHꢀC6H10O5]ꢀ ), whereas the FAB-MS (positive-ion mode) displayed
a quasi-molecular-ion peak at m/z 1239 ([MþNa]þ ). The molecular formula C59H92O26
of 1 was determined by HR-FAB-MS measurement of the quasi-molecular-ion peak
[MþNa]þ . Treatment of 1 with 10% aq. KOH/50% aqueous 1,4-dioxane 1:1 liberated
a known compound, desacylassamsaponin E (1a) [9], and two organic acids, AcOH and
tiglic acid, which were identified by HPLC analysis of their 4-nitrobenzyl derivatives
[9][10]. Acid hydrolysis of 1a with 5% aqueous H2SO4/1,4-dioxane 1:1 (v/v) furnished
l-arabinose, d-galactose, d-glucuronic acid, and d-xylose, the absolute configurations
(d- or l-) of which were established by GLC analysis of their thiazolidine derivatives
1
[30–32]. The H- and 13C-NMR spectra of 1 (Tables 3 and 4, resp.), which were

906
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
assigned by various NMR experiments3), showed signals assignable to a desacylas-
samsaponin E part: seven Me singlets at d(H) 0.72, 0.78, 1.06, 1.21, 1.24, 1.24, and 1.82
(Me(25), Me(26), Me(29), Me(24), Me(23), Me(30), and Me(27)); signals of an O-
bearing CH2 and four O-bearing CH groups at d(H) 3.20 (dd-like, HꢀC(3)), 3.20 and
3.65 (2d, J¼10.9 each, CH2(28)), 4.30 (br. s, HꢀC(16)), and 6.18 and 6.51 (2d, J¼10.3
each, HꢀC(22), HꢀC(21)); of an olefin at d(H) 5.30 (br. s, HꢀC(12)); and of four
glycopyranosyl moieties at d(H) 4.87 (d, J¼7.5, HꢀC(1’)), 5.69 (d-like, HꢀC(1’’)), 5.72
(d-like, HꢀC(1’’’)), and 4.93 (d, J¼7.6, HꢀC(1’’’’)), together with that of an Ac group
at d(H) 2.05 (s) and of a tigloyl (Tig; 2-methylbut-2-enoyl) group at d(H) 1.57 (d, J¼
7.6, Me), 1.87 (s-like, Me), 7.03 (qd-like, ¼CH). The positions of the AcO and TigO
groups in 1 were determined on the basis of a HMBC experiment, which showed 1H!
13C long-range correlations between the following pairs: HꢀC(21) and (C(1) of TigO
(d(C) 168.0); HꢀC(22) and H of Ac, and C¼O of Ac (d(C) 171.0) (Fig.). Furthermore,
comparison of the 13C-NMR data for 1 with those for desacylassamsaponin E (1a)
revealed acylation shifts around the 21- and 22-positions of the aglycon moiety. On the
basis of the above-mentioned evidence, the structure of chakasaponin I was determined
to be 22-O-acetyl-21-O-tigloyltheasapogenol B 3-O-b-d-galactopyranosyl(1!2)[b-d-
xylopyranosyl(1!2)-a-l-arabinopyranosyl(1!3)]-b-d-glucopyranosiduronic
acid
(1).
Chakasaponin II (2), obtained as colorless fine crystals from CHCl3/MeOH (m.p.
213.0–215.08) and with a negative optical rotation ([a]2D6 ¼ ꢀ17 (c¼0.41, MeOH)),
showed absorption bands at 3470, 1736, 1718, 1646, and 1048 cmꢀ1 due to OH, a,b-
unsaturated ester, carboxy, olefin, and ether functions in the IR spectrum. The
negative- and positive-ion mode FAB-MS of 2 showed quasi-molecular-ion peaks at
m/z 1271 ([MꢀH]ꢀ ) and 1139 ([MꢀHꢀC5H8O4]ꢀ ), and 1295 ([MþNa]þ ), and the
molecular formula C62H96O27 was determined by HR-FAB-MS measurement of the
quasi-molecular-ion peak [MþNa]þ . Treatment of 2 with 10% aqueous KOH/50%
aqueous 1,4-dioxane 1:1 liberated a known compound, desacylfloratheasaponin B (2a)
[12], together with tiglic acid, which was identified by HPLC analysis of its 4-
nitrobenzyl derivative [9][10]. Acid hydrolysis of 2a with 5% aqueous H2SO4/1,4-
dioxane 1:1 (v/v) furnished l-arabinose, d-galactose, d-glucuronic acid, and d-xylose,
which were identified by GLC analysis of their (trimethylsilyl)thiazolidine derivatives
1
[30–32]. The H- and 13C-NMR spectra of 2 (Tables 3 and 4, resp.) indicated the
presence of a desacylfloratheasaponin B part: seven Me singlets at d(H) 0.75, 0.93, 1.06,
1.07, 1.20, 1.28, and 1.80 (Me(25), Me(26), Me(29), Me(24), Me(23), Me(30), and
Me(27)); signals of an O-bearing CH2 and five O-bearing CH groups at d(H) 3.20 (dd-
like, HꢀC(3)), 3.39 and 3.65 (2 br. d, J ꢂ 10 each, CH2(28)), 4.21 (br. s, HꢀC(16)),
4.40 (m, HꢀC(15)), and 6.27 and 6.67 (2d, J¼10.3 each, HꢀC(22), HꢀC(21)); of an
olefin at d(H) 5.43 (br. s, HꢀC(12)); and of four glycopyranosyl moieties at d(H) 4.87
(d, J¼7.6, HꢀC(1’)), 5.70 (d, J¼7.6, HꢀC(1’’)), 5.72 (d, J¼7.2, HꢀC(1’’’)), and 4.96 (d,
J¼7.6, HꢀC(1’’’’)), together with those of two Tig groups at d(H) 1.24 (d, J¼6.8, Me of
The 1H- and 13C-NMR spectra of 1–3 were assigned with the aid of distortionless enhancement by
polarization transfer (DEPT), double-quantum filter correlation spectroscopy (DQF COSY),
heteronuclear multiple quantum coherence spectroscopy (HMQC), and heteronuclear multiple-
bond correlation spectroscopy (HMBC) experiments.
3
)

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
907
Figure. DQF COSY and HMBC data of chakasaponins I–III (1–3, resp.)

908
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Table 3. 1H-NMR Data of Chakasaponins I–III (1–3, resp.). At 600 MHz, in (D5)pyridine; d(H) in
ppm, J in Hz. Arbitrary C-atom numbering (see Formula)
1
2
3
HꢀC(3)
HꢀC(12)
HꢀC(16)
HꢀC(18)
HꢀC(21)
HꢀC(22)
Me(23)
3.20 (dd-like)
5.30 (br. s)
4.30 (br. s)
3.02 (dd-like)
6.51 (d, J ¼ 10.3)
6.18 (d, J ¼ 10.3)
1.24 (s)
3.20 (dd-like)
5.43 (br. s)
4.21 (br. s)
3.06 (dd-like)
6.67 (d, J ¼ 10.3)
6.27 (d, J ¼ 10.3)
1.20 (s)
3.20 (dd-like)
5.45 (br. s)
4.10 (br. s)
3.01 (dd-like)
6.51 (d, J ¼ 10.3)
6.19 (d, J ¼ 10.3)
1.21 (s)
Me(24)
1.21 (s)
1.07 (s)
1.11 (s)
Me(25)
0.72 (s)
0.75 (s)
0.79 (s)
Me(26)
0.78 (s)
0.93 (s)
0.98 (s)
Me(27)
1.82 (s)
1.80 (s)
2.08 (s)
CH2(28)
3.20 (d, J ¼ 10.9),
3.65 (d, J ¼ 10.9)
1.06 (s)
3.39 (br. d, J ꢂ 10),
3.65 (br. d, J ꢂ 10)
1.06 (s)
3.40 (d, J ¼ 11.0),
3.58 (d, J ¼ 11.0)
1.08 (s)
Me(29)
Me(30)
HꢀC(1’)
HꢀC(1’’)
HꢀC(1’’’)
HꢀC(1’’’’)
21-TigO:
HꢀC(3)
Me(4)
1.24 (s)
1.28 (s)
1.29 (s)
4.87 (d, J ¼ 7.5)
5.69 (d-like)
5.72 (d-like)
4.93 (d, J ¼ 7.6)
4.87 (d, J ¼ 7.6)
5.70 (d, J ¼ 7.6)
5.72 (d, J ¼ 7.2)
4.96 (d, J ¼ 7.6)
4.96 (d, J¼7.5)
5.70 (d, J ¼ 7.2)
5.72 (d, J ¼ 7.2)
4.93 (d, J ¼ 7.3)
7.03 (qd-like)
1.57 (d, J ¼ 7.6)
1.87 (br. s-like)
7.06 (qd-like)
1.53 (d, J ¼ 6.9)
1.84 (br. s-like)
7.04 (qd-like)
1.57 (d, J ¼ 7.6)
1.89 (br. s-like)
Me(5)
AcOꢀC(22):
Me
2.05 (s)
2.08 (s)
22-TigO:
HꢀC(3)
Me(4)
6.72 (qd-like)
1.24 (d, J ¼ 6.8)
1.69 (br. s-like)
Me(5)
22-TigO), 1.53 (d, J¼6.9, Me of 21-TigO), 1.69 (s-like, Me of 22-TigO), 1.84 (s-like, Me
of 21-TigO), 6.72 (qd-like, ¼CH of 22-TigO), 7.06 (qd-like, ¼CH of 21-TigO). The
positions of the two TigO groups in 2 were established on the basis of HMBC
1
13
experiment, which showed H! C long-range correlations between the following
pairs: HꢀC(21) and C(1) of 21-TigO (d(C) 168.1); HꢀC(22) and C(1) of 22-TigO
(d(C) 168.4) (Fig.). Furthermore, comparison of the 13C-NMR data for 2 with those for
2a revealed acylation shifts around the 21- and 22-positions of the aglycon moiety.
Consequently, the structure of chakasaponin II was elucidated as 21,22-di-O-tigloyl-R1-
barrigenol
3-O-b-d-galactopyranosyl(1!2)[b-d-xylopyranosyl(1!2)-a-l-arabino-
pyranosyl(1!3)]-b-d-glucopyranosiduronic acid (2).
Chakasaponin III (3), obtained as colorless fine crystals from CHCl3/MeOH (m.p.
219.0–221.08) with a negative optical rotation ([a]2D6 ¼ ꢀ8.2 (c¼0.83, MeOH)),
showed absorption bands at 3469, 1736, 1723, 1655, and 1047 cmꢀ1 in the IR spectrum.
The negative- and positive-ion mode FAB-MS of 3 showed quasi-molecular-ion peaks
at m/z 1231 ([MꢀH]ꢀ ) and 967 ([MꢀHꢀC10H16O8]ꢀ ), and 1255 ([MþNa]þ ), and

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
909
Table 4. 13C-NMR Data of Chakasaponins I–III (1–3, resp.). At 150 MHz, in (D5)pyridine; d(C) in
ppm, J in Hz. Arbitrary C-atom numbering (see Formula)
Position
d(C)
Position
d(C)
1
2
3
1
2
3
1
2
3
4
5
6
7
8
9
38.8
26.6
89.6
39.6
55.7
18.3
33.1
40.1
46.9
36.8
23.8
123.8
142.8
41.7
34.6
68.0
48.0
40.0
47.2
36.5
79.0
74.3
28.0
16.8
15.6
16.9
27.4
63.7
29.5
20.1
39.0
26.6
89.6
39.6
55.5
18.8
36.7
41.4
47.1
37.0
24.0
125.4
143.7
48.6
73.8
67.6
47.8
40.9
46.9
36.5
78.3
73.4
28.0
16.8
15.8
17.6
21.3
62.8
29.5
20.1
39.0
26.6
89.5
39.6
55.6
18.8
36.7
41.4
47.1
37.0
24.0
125.4
143.7
48.3
74.0
67.5
47.8
40.9
46.9
36.4
79.3
72.6
28.0
16.8
15.8
17.5
21.1
63.1
29.5
20.0
22-AcO:
1
2
22-TigO:
1
2
3
171.0
20.9
170.8
20.7
168.4
129.0
137.0
14.2
4
5
12.3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
21-TigO:
1
b-d-GlcA:
1’
2’
3’
4’
5’
6’
b-d-Gal:
1’’
2’’
3’’
4’’
105.7
79.0
84.0
71.1
77.4
172.1
105.7
79.1
84.1
71.0
77.4
172.0
105.7
79.1
84.0
71.1
77.4
172.0
103.5
73.8
75.1
70.1
76.4
61.9
103.5
73.7
75.1
70.1
76.4
62.0
103.5
73.7
75.2
70.1
76.4
61.9
5’’
6’’
a-l-Ara:
1’’’
2’’’
3’’’
4’’’
5’’’
b-d-Xyl:
1’’’’
2’’’’
3’’’’
101.8
82.0
73.3
68.3
65.9
101.8
82.0
73.4
68.2
65.9
101.8
82.0
73.3
68.3
65.9
106.9
75.7
78.2
70.7
67.5
106.9
75.6
78.3
70.7
67.5
106.9
75.8
78.3
70.8
67.5
168.0
129.5
136.9
14.2
168.1
129.4
136.8
14.2
168.1
129.4
137.0
14.2
2
3
4
4’’’’
5’’’’
5
12.4
12.4
12.4
the molecular formula C59H92O27 was determined by HR-FAB-MS. Treatment of 3 with
10% aqueous KOH/50% aqueous 1,4-dioxane 1:1 liberated desacylfloratheasaponin B
(2a) [12], together with AcOH and tiglic acid, which were identified by HPLC analysis
of their 4-nitrobenzyl derivatives [9][10]. The signals in the 1H- and 13C-NMR spectra
of 3 (Tables 3 and 4, resp.) were superimposable on those of 2, except for the signals
1
due to the 22-acyl moiety; i.e., the H- and 13C-NMR spectra of 3 showed signals
assignable to an Ac and a Tig moiety: d(H) 2.08 (s, Me of AcO), 1.57 (d, J¼7.6, Me of
TigO), 1.89 (s-like, Me of TigO), 7.04 (qd-like, ¼CH of TigO) together with the signals

910
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
of a desacylfloratheasaponin B part. The positions of the Ac and Tig groups in 3 were
1
13
established on the basis of HMBC experiment, which showed H! C long-range
correlations between the following pairs: HꢀC(21) and C(1) of TigO (d(C) 168.1);
HꢀC(22) and C¼O of Ac (d(C) 170.8) (Fig.). Furthermore, comparison of the
13C-NMR data of 3 with those for 2a revealed acylation shifts around the 21- and 22-
positions. Consequently, the structure of chakasaponin III was elucidated as 22-O-
acetyl-21-O-tigloyl-R1-barrigenol 3-O-b-d-galactopyranosyl(1!2)[b-d-xylopyrano-
syl(1!2)-a-l-arabinopyranosyl(1!3)]-b-d-glucopyranosiduronic acid (3).
We previously isolated the nine acylated oleanane-type triterpene oligoglycosides,
floratheasaponins A–I [12][13], from the flower buds of C. sinensis grown in Japan and
China (Anhui Province). Those structures were similar to those of chakasaponins I, II,
and III (1–3, resp.), except for the acyl moieties or the terminal glycosyl residue.
Interestingly, chakasaponins 1–3 were only isolated from C. sinensis grown in China
(Fujian Province), while floratheasaponins A–I were isolated from tea plants of
Japanese and Chinese (Anhui Province) provenances. This finding is interesting from
the perspective of chemotaxonomy of C. sinensis.
Biology. Gastrointestinal transit accelerating effects are thought to be beneficial for
the prevention and treatment of the inhibition of gastrointestinal transit such as affects
of the ileus in clinical situations [33][34]. In the present study, the MeOH extract from
the flower buds of C. sinensis (Fujian Province) was found to exhibit accelerating effect
on gastrointestinal transit in mice. Therefore, the effects of acylated oleanane-type
triterpene oligoglycosides, 1–3, were examined. As a result, all chakasaponins (1–3,
resp.) showed the accelerating effects at a dose of 100 mg/kg, p.o. (Table 5).
Table 5. Effects of Chakasaponins I–III (1–3, resp.) on Gastrointestinal Transit in Mice
Treatment
Dose [mg/kg, p.o.]
n
Gastrointestinal transit [%]a)
Control
Chakasaponin I (1)
–
25
50
100
25
50
100
25
50
7
8
7
7
5
5
5
7
7
7
40.0ꢁ2.5
43.8ꢁ1.3
49.1ꢁ4.3
81.6ꢁ3.3**
64.9ꢁ10.8
52.1ꢁ5.8
Chakasaponin II (2)
Chakasaponin III (3)
75.6ꢁ11.1**
51.0ꢁ3.4
64.2ꢁ6.3*
100
68.5ꢁ2.5**
a) Mean values ꢁS.E.M. Significantly different from the control group: *: p<0.05, **: p<0.01.
Next, the inhibitory effect of the MeOH extract against pancreatic lipase, which is
well-known to play an important role in lipid digestion, was examined. As shown in
Table 2, the MeOH extract inhibited the enzyme activity of pancreatic lipase. In
addition, Han et al. reported that a mixture of teasaponins, which were acylated
oleanane-type triterpene oligoglycosides, showed an anti-obese effect in high-fat-diet-
treated mice and pancreatic lipase inhibitory effect [35]. Therefore, the inhibitory
effects of the isolated chakasaponins 1–3 and desacylfloratheasaponin B (2a) against

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
911
pancreatic lipase were examined (Table 6). A lipase inhibitor (positive control),
orlistat, showed a potent effect in this assay model (IC50 ¼36 nm). Chakasaponins 1–3
also inhibited the pancreatic lipase activity with IC50 values of 0.17–0.53 mm, although
their activities were much weaker than that of orlistat. On the other hand,
desacylfloratheasaponin B (2a) did not show such effect. These findings suggested
that the 21- and 22-acyl groups are essential for the effect (Table 6). Consequently,
chakasaponins 1–3 may be useful for the prevention of obesity, but further experiment
need to be performed to assess their therapeutic potential.
Table 6. Inhibitory Effects of Chakasaponins I–III (1–3, resp.) and Desacylfloratheasaponin B (2a)
against Pancreatic Lipase
Conc. [mm] Inhibition [%]a)
Chakasaponin I (1) Chakasaponin II (2) Chakasaponin III (3) Desacylflora-
theasaponin B (2a)
50
100
200
300
400
500
600
ꢀ4.2ꢁ6.2
ꢀ7.0ꢁ6.1
64.4ꢁ2.5**
88.1ꢁ1.7**
93.8ꢁ0.7**
96.0ꢁ1.1**
96.6ꢁ0.9**
ꢀ1.0 ꢁ8.1
ꢀ1.2 ꢁ7.2
ꢀ0.8ꢁ6.2
3.5ꢁ2.4
2.3ꢁ5.4
2.3ꢁ5.6
2.7ꢁ5.0
1.3ꢁ4.5
3.6ꢁ5.4
3.0ꢁ4.4
1.8ꢁ4.1
ꢀ6.2ꢁ7.7
59.9ꢁ8.3**
83.7ꢁ3.8**
93.4ꢁ1.7**
95.1ꢁ1.8**
96.1ꢁ0.9**
4.4ꢁ5.4
23.7ꢁ1.3**
40.9ꢁ1.6**
63.4ꢁ1.9**
IC50
0.17 mm
0.18 mm
0.53 mm
–
a) Mean values ꢁS.E.M (n¼4). Significantly different from the control group: **: p<0.01. IC50 of
orlistat (positive control): 36 nm.
M. Y., S. N., and H. M. were supported by the 21st Century COE Program, the Academic Frontier
Research Project, and a Grant in Aid for Scientific Research from the Ministry of Education, Culture,
Sports, Science and Technology of Japan.
Experimental Part
General. Normal-phase silica-gel column chromatography (CC): BW-200 silica gel (SiO2; 150–350
mesh; Fuji Silysia Chemical, Ltd., Japan). Reversed-phase (RP) silica-gel CC: Chromatorex ODS
DM1020T (100–200 mesh; Fuji Silysia Chemical, Ltd.). HPLC: Shimadzu RID-6A refractive-index and
SPD-10A UV/VIS detectors, LC-6AD pump, and C-R6A Chromatopac and YMC-Pack ODS-A (250ꢃ
4.6 mm; 5 mm; for anal. purpose; and 250ꢃ20 mm; 5 mm; for prep. purpose) columns. M.p.: Yanagimoto
MP-500D micro melting-point apparatus; uncorrected. Optical rotations: Horiba SEPA-300 digital
1
polarimeter (l ¼ 5 cm). IR Spectra: Shimadzu FTIR-8100 spectrometer; ˜n in cmꢀ1. H- and 13C-NMR
spectra: JEOL JNM-LA500 (500 MHz) and ECA-600K (600 MHz) spectrometers; at 500 and 125 MHz
and 600 and 150 MHz, resp., d in ppm rel. to Me4Si, J in Hz. FAB-MS and HR-FAB-MS: JEOL JMS-SX
102A mass spectrometer (matrix: glycerol); in m/z. EI-MS and HR-EI-MS: JEOL JMS-GCMATE mass
spectrometer; in m/z.
Plant Material. Flower buds of Camellia sinensis, which was cultivated in Fujian Province of China,
were collected in 2006. The botanical identification was undertaken by one of the authors (M. Y.). A
voucher of the plant is on file in our laboratory (2006. China-06F).
Extraction and Isolation. The dried flower buds of C. sinensis (1.5 kg, Fujian Province, China) were
finely cut and extracted three times with MeOH (3ꢃ15 l) for 3 h under reflux. Evaporation of the solvent

912
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
under reduced pressure provided a MeOH extract (467 g, 31.1%). A portion (450 g) of the MeOH
extract was partitioned in AcOEt (3ꢃ2 l)/H2O (2 l) to furnish an AcOEt-soluble fraction (46 g, 3.2%)
and aq. phase, which was extracted with BuOH (3ꢃ2 l) to give BuOH- and H2O-soluble fractions (237 g
(16.4%) and 166 g (11.5%), resp.).
The AcOEt-soluble fraction (40 g) was subjected to CC (2.0 kg of normal-phase SiO2; hexane/
AcOEt 10 :1!5 :1!1:1, AcOEt, CHCl3/MeOH 20 :1! 10 :1! 5 :1!1:1, MeOH) to give seven
fractions, Fr. 1, Fr. 2, Fr. 3 (3.1 g), Fr. 4 (8.0 g), Fr. 5 (8.1 g), Fr. 6 (4.4 g), and Fr. 7. Fr. 3 (3.1 g) were
separated by CC (300 g of RP SiO2; MeOH/H2O 30 :70!50 :50!70 :30, MeOH, CHCl3) to give seven
fractions, Fr. 3.1 (37 mg), Fr. 3.2 (159 mg), Fr. 3.3 (255 mg), Fr. 3.4, Fr. 3.5, Fr. 3.6, and Fr. 3.7. Fr. 3.1
(37 mg) and Fr. 3.2 (159 mg) were purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/min) to give PhCH2OH
(15.1 mg, tR 21.7 min) and protocatechuic acid methyl ester (11.3 mg, tR 38.0 min). A portion (250 mg) of
Fr. 3.3 was purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/min) to give (E)-cinnamic acid (9.8 mg, tR
24.3 min). Fr. 4 (8.0 g) was separated by CC (480 g of RP SiO2; MeOH/H2O 10 :90!30 :70!50 :50!
70 :30, 9.0 ml/min, MeOH, CHCl3) to yield eight fractions, Fr. 4.1 (655 mg), Fr. 4.2 (2.1 g), Fr. 4.3, Fr. 4.4,
Fr. 4.5, Fr. 4.6, Fr. 4.7, and Fr. 4.8. Fr. 4.1 (655 mg) were purified by HPLC (MeOH/H2O 10 :90, 9.0 ml/
min) to give (þ)-gallocatechin (14.2 mg, tR 37.8 min). A portion (213 mg) of Fr. 4.2 and a portion
(700 mg) of Fr. 4.3 were purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/min) to give (ꢀ)-epigallocatechin
3-O-gallate (294 mg, tR 18.2 min) and (ꢀ)-epicatechin 3-O-gallate (160 mg, tR 37.2 min). Fr. 5 (8.1 g) was
separated by CC (120 g of RP SiO2; MeOH/H2O 10 :90!30 :70!50 :50!70 :30, MeOH, CHCl3) to
afford ten fractions, Fr. 5.1, Fr. 5.2, Fr. 5.3 (1.9 g), Fr. 5.4, Fr. 5.5 (340 mg), Fr. 5.6, Fr. 5.7, Fr. 5.8, Fr. 5.9,
and Fr. 5.10. A portion (200 mg) of Fr. 5.3 was purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/min) to give
caffeine (49.1 mg, tR 20.6 min) and (S)-1-phenylethyl b-d-glucopyranoside (17.2 mg, tR 43.3 min). Fr. 5.5
(340 mg) was purified by HPLC (MeOH/H2O 45 :55, 9.0 ml/min) to give kaempferol 3-O-b-d-
galactopyranoside (34.3 mg, tR 37.7 min) and kaempferol 3-O-b-d-glucopyranoside (28.0 mg, tR
41.7 min). Fr. 6 (4.4 g) was separated by CC (120 g of RP SiO2; MeOH/H2O 10 :90!30 :70!
50 :50!70 :30, MeOH, CHCl3) to give nine fractions, Fr. 6.1, Fr. 6.2, Fr. 6.3, Fr. 6.4 (942 mg), Fr. 6.5,
Fr. 6.6, Fr. 6.7, Fr. 6.8, and Fr. 6.9. Fr. 6.4 (942 mg) was purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/
min) to give (ꢀ)-epigallocatechin 3-O-gallate (94.1 mg, tR 18.7 min).
The BuOH-soluble fraction (170 g) was subjected to CC (3 kg normal-phase SiO2, CHCl3, CHCl3/
MeOH/H2O 30 :10 :1!10 :3 :1!7:3 :1!6 :4 :1, MeOH) to give 6 fractions, Fr. 1, Fr. 2 (25.5 g), Fr. 3
(10.3 g), Fr. 4 (115.0 g), Fr. 5, and Fr. 6. A portion (10.0 g) of Fr. 2 was separated by CC (300 g of RP
SiO2; MeOH/H2O 30 :70!40 :60!50 :50!60 :40!70 :30, MeOH, CHCl3) to give six fractions, Fr. 2.1
(4.5 g), Fr. 2.2, Fr. 2.3, Fr. 2.4, Fr. 2.5, and Fr. 2.6. A portion (300 mg) of Fr. 2.1 was purified by HPLC
(MeOH/H2O 30 :70, 9.0 ml/min) to give (S)-1-phenylethyl b-d-glucopyranoside (12.4 mg, tR 39.1 min)
and caffeine (81.1 mg, tR 18.5 min). Fr. 3 (10.3 g) was separated by CC (300 g of RP SiO2; MeOH/H2O
10 :90!30 :70!50 :50!70 :30, MeOH, CHCl3) to yield four fractions, Fr. 3.1, Fr. 3.2 (4.8 g), Fr. 3.3,
and Fr. 3.4. A portion (300 mg) of Fr. 3.2 was purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/min) to give
(ꢀ)-epicatechin (0.8 mg, tR 20.0 min), benzyl b-d-glucopyranoside (1.2 mg, tR 20.6 min), (S)-1-phenyl-
ethyl b-d-glucopyranoside (33.0 mg, tR 37.2 min), and caffeine (1.1 mg, tR 18.5 min). Fr. 4 (115 g) was
subjected to CC (530 g of RP SiO2; MeOH/H2O 10 :90!30 :70!50 :50!70 :30, MeOH) to give nine
fractions, Fr. 4.1 (6.6 g), Fr. 4.2, Fr. 4.3 (9.2 g), Fr. 4.4 (13.4 g), Fr. 4.5 (6.7 g), Fr. 4.6 (12.5 g), Fr. 4.7
(24.1 g), Fr. 4.8 (23.3 g), and Fr. 4.9. A portion (655 mg) of Fr. 4.1 was purified by HPLC (MeOH/H2O
10 :90, 9.0 ml/min) to give (þ)-gallocatechin (14.2 mg, tR 37.8 min). A portion (500 mg) of Fr. 4.3 was
purified by HPLC (MeOH/H2O 20 :80, 9.0 ml/min) to give (ꢀ)-epigallocatechin 3-O-gallate (119 mg, tR
31.5 min). A portion (500 mg) of Fr. 4.4 was purified by HPLC (MeOH/H2O 30 :70, 9.0 ml/min) to give
(ꢀ)-epigallocatechin 3-O-gallate (29.3 mg, tR 31.5 min). A portion (200 mg) of Fr. 4.5 was separated by
HPLC (MeOH/H2O 50 :50, 9.0 ml/min) to give kaempferol 3-O-b-d-glucopyranosyl(1!3)-a-l-
rhamnopyranosyl(1!6)-b-d-glucopyranoside (43.5 mg, tR 18.2 min). Fr. 4.6 (12.5 g) was subjected to
CC (370 g of RP SiO2; MeOH/H2O 50 :50!60 :40, MeOH) to give five fractions, Fr. 4.6.1, Fr. 4.6.2,
Fr. 4.6.3 (6.2 g), Fr. 4.6.4, Fr. 4.6.5, and Fr. 4.6.6. A portion (1 g) of Fr. 4.6.3 was further separated by
HPLC (MeOH/H2O 70 :30, 9.0 ml/min) to give four fractions, Fr. 4.6.3.1, Fr. 4.6.3.2 (159.7 mg),
Fr. 4.6.3.3, and Fr. 4.6.3.4. Fr. 4.6.3.2 (159.7 mg) was purified by HPLC (MeOH/H2O 65 :35, 9.0 ml/
min) to afford chakasaponin III (3; 21.8 mg, tR 60.2 min). A portion (1.0 g) of Fr. 4.7 was purified by

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
913
HPLC (MeOH/H2O 70 :30, 9.0 ml/min) to give chakasaponin I (1; 212 mg, tR 30.0 min). A portion
(200 mg) of Fr. 4.8 was purified by HPLC (MeOH/H2O 70 :30, 9.0 ml/min) to give chakasaponin II (2;
61.3 mg, tR 57.7 min). The known compounds were identified by comparison of their physical data ([a]D,
1
MS (FAB-MS for 4, 5, and 10, and EI-MS for 6–9 and 11–16), and H- and 13C-NMR) with reported
values.
Chakasaponin I (¼22-O-Acetyl-21-O-tigloyltheasapogenol B 3-O-b-d-Galactopyranosyl(1!2)[b-
d-xylopyranosyl(1!2)-a-l-arabinopyranosyl(1!3)]-b-d-glucopyranosiduronic Acid; 1). Colorless fine
crystals. M.p. 214.0–216.08 (CHCl3/MeOH). [a]2D6 ¼ ꢀ9.0 (c¼0.32, MeOH). IR (KBr): 3468, 1736, 1718,
1078. 1H- and 13C-NMR: Tables 3 and 4, resp. FAB-MS (pos.): 1239 ([MþNa]þ ). FAB-MS (neg.): 1215
([MꢀH]ꢀ ), 1083 ([MꢀHꢀC5H8O4]ꢀ ), 1053 ([MꢀHꢀC6H10O5]ꢀ ). HR-FAB-MS (pos.): 1239.5764
([MþNa]þ , C59H92NaO2þ6 ; calc. 1239.5775).
Chakasaponin II (¼21,22-Di-O-tigloyl-R1-barrigenol 3-O-b-d-Galactopyranosyl(1!2)[b-d-
xylopyranosyl(1!2)-a-l-arabinopyranosyl(1!3)]-b-d-glucopyranosiduronic Acid; 2). Colorless fine
crystals. M.p. 213.0–215.08 (CHCl3/MeOH). [a]2D6 ¼ ꢀ17.1 (c¼0.41, MeOH). IR (KBr): 3470, 1718,
1646, 1078. 1H- and 13C-NMR: Tables 3 and 4, resp. FAB-MS (pos.): 1295 ([MþNa]þ ). FAB-MS (neg.):
1271 ([MꢀH]ꢀ , 1139 ([MꢀHꢀC5H8O4]ꢀ ), 1007 ([MꢀHꢀC10H16O8]ꢀ ), 845 ([MꢀHꢀC16H26O13]ꢀ ).
HR-FAB-MS (pos.): 1295.6029 ([MþNa]þ , C62H96NaO2þ7 ; calc. 1295.6037).
Chakasaponin III (¼22-O-Acetyl-21-O-tigloyl-R1-barrigenol 3-O-b-d-Galactopyranosyl(1!2)[b-d-
xylopyranosyl(1!2)-a-l-arabinopyranosyl(1!3)]-b-d-glucopyranosiduronic Acid; 3). Colorless fine
crystals. M.p. 219.0–221.08 (CHCl3/MeOH). [a]2D6 ¼ ꢀ8.2 (c¼0.83, MeOH). IR (KBr): 3470, 1722, 1656,
1082. 1H- and 13C-NMR: Tables 3 and 4, resp. FAB-MS (pos.): 1255 ([MþNa]þ ). FAB-MS (neg.): 1231
([MꢀH]ꢀ ), 967 ([MꢀHꢀC10H16O8]ꢀ ). HR-FAB-MS (pos.): 1255.5715 ([MþNa]þ , C62H96NaOþ27
;
calc. 1255.5624).
Alkaline Hydrolysis of 1–3. A soln. of 1–3 (19.0, 6.4, and 3.0 mg, resp.) in 50% aq. 1,4-dioxane
(0.5 ml) was treated with 10% aq. KOH (0.5 ml), and the mixture was stirred at 378 for 1 h. The mixture
was neutralized with Dowex HCR W2 (Hþ form), and the resin was removed by filtration. Evaporation
of the solvent from the filtrate under reduced pressure yielded a reaction product. A part of the product
was dissolved in ClCH2CH2Cl (2 ml), and the soln. was treated with 4-nitrobenzyl-N,N’-diisopropyli-
sourea (10 mg), then the soln. was stirred at 808 for 1 h. The soln. was subjected to HPLC analysis (YMC-
Pack ODS-A, 250ꢃ4.6 mm, MeCN/H2O 50 :50, 1.0 ml/min, UV detection at 254 nm) to identify the 4-
nitrobenzyl esters of AcOH (tR 8.9 min) from 1 and 3, and of tiglic acid (tR 35.1 min) from 1–3. The rest of
the reaction product was subjected to CC (50 mg of normal-phase SiO2; CHCl3/MeOH/H2O 10 :3 :1
lower layer!6 :4 :1) to give desacylassamsaponin E [8] (1a; 14.2 mg) from 1 or desacylfloratheasaponin
B [11] (2a; 4.2 and 2.0 mg) from 2 and 3, resp.
Acid Hydrolysis of 1a and 2a. A soln. of 1a or 2a (both 2 mg) in 5% aq. H2SO4/1,4-dioxane 1:1 (v/v,
1 ml) was heated under reflux for 2 h. After cooling, the mixture was neutralized with Amberlite IRA-400
(OHꢀ form), and the resin was removed by filtration. Evaporation of the solvent from the filtrate under
reduced pressure gave a product, which was subjected to a Sep-Pack C18 cartridge with H2O and MeOH.
The H2O eluate was concentrated, and the residue was treated with l-cysteine methyl ester
hydrochloride (1.0 mg) in pyridine (0.1 ml) at 608 for 1 h. After reaction, the soln. was treated with
N,O-bis(trimethylsilyl)trifluoroacetamide (0.1 ml) at 608 for 1 h. The supernatant was then subjected to
GLC analysis to identify the derivatives of d-glucuronic acid (i), d-galactose (ii), l-arabinose (iii) and d-
xylose (iv) (Supelco STB-1 (30 mꢃ0.25 mm) cap. column; 2308 column temp.; N2 carrier gas; tR for i)
26.5 min, ii) 25.6 min, iii) 15.1 min, iv) 19.3 min).
Test Animals. Male ddY mice, weighing ca. 30 g, were purchased from Kiwa Laboratory Animal Co.,
Ltd., Wakayama, Japan. The animals were housed at a constant temp. of 23ꢁ28 and were fed a standard
laboratory chow (MF, Oriental Yeast Co., Ltd., Tokyo, Japan). The animals were fasted for 18–20 h prior
to the beginning of the experiment, but were allowed free access to tap water. Each test sample was
suspended in 5% acacia soln., and the soln. was orally administered at 10 ml/kg in each experiment, while
the vehicle was administered orally at 10 ml/kg in the corresponding control groups. The experimental
protocols were approved by Experimental Animal Research Committee at Kyoto Pharmaceutical
University.

914
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Effects on Gastrointestinal Transit in Mice. A charcoal meal containing a soln. of 1.5% carboxymethyl
cellulose sodium salt (CMC-Na) and 5% charcoal as a marker was intragastrically given (0.2 ml/mouse)
to conscious mice. Fifteen min later, mice were sacrificed by cervical dislocation. The abdominal cavity
was opened, and the gastrointestinal tract was removed. The traveled distance of the marker was
measured and expressed as a percentage of the total length of the small intestine from pylorus to caecum.
The test samples were orally given by means of a metal orogastric tube 60 min prior to the administration
of the charcoal meal.
Effect on Pancreatic Lipase Activity. A suspension of triolein (80 mg), phosphatidylcholine (10 mg),
and sodium taurocholate (5 mg) in 9 ml of 0.1m Tris·HCl buffer (pH 7.0) containing 0.1m NaCl was
sonicated for 10 min. This sonicated substrate suspension (0.1 ml) in a test tube was pre-incubated with
5 ml of test sample in DMSO and 95 ml of the Tris·HCl buffer for 3 min at 378. An aliquot of porcine
pancreatic lipase (250 mg/ml, type II, Sigma Chemical Co.; 50 ml) or the Tris·HCl buffer (50 ml) as a blank
test was then added to start the reaction. After 30 min of incubation, the test tube was immediately
immersed in boiling water for 2 min to stop the reaction, then cooled with water. Free fatty acid
concentration was determined with a commercial kit (NEFA C-test Wako, Wako Pure Chemical
Industries, Ltd.).
Statistics. Values were expressed as means ꢁS.E.M. For statistical analysis, one-way analysis of
variance followed by Dunnettꢀs test was used. Probability (p) values less than 0.05 were considered
significant.
REFERENCES
[1] M. Yoshikawa, T. Wang, S. Sugimoto, S. Nakamura, A. Nagatomo, H. Matsuda, S. Harima,
Yakugaku Zasshi 2008, 128, 141.
[2] I. Kitagawa, K. Hori, T. Motozawa, T. Murakami, M. Yoshikawa, Chem. Pharm. Bull. 1998, 46, 1901.
[3] M. Yoshikawa, T. Morikawa, N. Li, A. Nagatomo, X. Li, H. Matsuda, Chem. Pharm. Bull. 2005, 53,
1559.
[4] T. Morikawa, H. Matsuda, N. Li, S. Nakamura, X. Li, M. Yoshikawa, Heterocycles 2006, 68, 1139.
[5] T. Morikawa, N. Li, A. Nagatomo, H. Matsuda, X. Li, M. Yoshikawa, J. Nat. Prod. 2006, 69, 185.
[6] M. Yoshikawa, T. Morikawa, S. Nakamura, N. Li, X. Li, H. Matsuda, Chem. Pharm. Bull. 2007, 55,
57.
[7] T. Morikawa, H. Matsuda, N. Li, X. Li, M. Yoshikawa, Helv. Chim. Acta 2007, 90, 2342.
[8] T. Murakami, J. Nakamura, H. Matsuda, M. Yoshikawa, Chem. Pharm. Bull. 1999, 47, 1759.
[9] T. Murakami, J. Nakamura, T. Kageura, H. Matsuda, M. Yoshikawa, Chem. Pharm. Bull. 2000, 48,
1720.
[10] M. Yoshikawa, T. Murakami, S. Yoshizumi, N. Murakami, J. Yamahara, H. Matsuda, Chem. Pharm.
Bull. 1996, 44, 1899.
[11] T. Morikawa, S. Nakamura, Y. Kato, O. Muraoka, H. Matsuda, M. Yoshikawa, Chem. Pharm. Bull.
2007, 55, 293.
[12] M. Yoshikawa, T. Morikawa, K. Yamamoto, Y. Kato, A. Nagatomo, H. Matsuda, J. Nat. Prod. 2005,
68, 1360.
[13] M. Yoshikawa, S. Nakamura, Y. Kato, K. Matsuhira, H. Matsuda, Chem. Pharm. Bull. 2007, 55, 598.
[14] M. Yoshikawa, F. Xu, T. Morikawa, Y. Pongpiriyadacha, S. Nakamura, Y. Asao, A. Kumahara, H.
Matsuda, Chem. Pharm. Bull. 2007, 55, 308.
[15] M. Yoshikawa, S. Sugimoto, S. Nakamura, H. Matsuda, Chem. Pharm. Bull. 2007, 55, 571.
[16] S. Nakamura, S. Sugimoto, H. Matsuda, M. Yoshikawa, Heterocycles 2007, 71, 577.
[17] M. Yoshikawa, T. Morikawa, Y. Asao, E. Fujiwara, S. Nakamura, H. Matsuda, Chem. Pharm. Bull.
2007, 55, 606.
[18] M. Yoshikawa, S. Sugimoto, S. Nakamura, H. Sakumae, H. Matsuda, Chem. Pharm. Bull. 2007, 55,
1034.
[19] S. Nakamura, S. Sugimoto, H. Matsuda, M. Yoshikawa, Chem. Pharm. Bull. 2007, 55, 1342.
[20] O. Barbera, J. F. Sans, J. Sanchez-Parareda, Phytochemistry 1986, 25, 2361.

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
915
[21] K. R. Markham, B. Ternai, R. Stanley, H. Geiger, T. J. Mabry, Tetrahedron 1978, 34, 1389.
[22] C. Nishino, N. Enoki, S. Tawata, A. Mori, K. Kobayashi, M. Fukushima, Agric. Biol. Chem. 1987, 51,
139.
[23] G. Nonaka, O. Kawahara, I. Nishioka, Chem. Pharm. Bull. 1986, 31, 3906.
[24] V. I. Tararov, R. K. Kadyrov, T. H. Rieneier, Adv. Synth. Catal. 2002, 344, 200.
[25] Y. S. Lee, D. I. Ito, K. Koya, S. Izumi, T. Hirata, Chem. Lett. 1996, 719.
[26] M. Miyazawa, T. Oshima, K. Koshio, Y. Itsuzaki, J. Anzai, J. Agric. Food Chem. 2003, 51, 6953.
[27] M. Nakai, Y. Fukui, S. Asami, Y. Toyoda-Ono, T. Iwashita, H. Shibata, T. Mitsunaga, F. Hashimoto,
Y. Kiso, J. Agric. Food Chem. 2005, 53, 4593.
[28] A. Finger, U. H. Engelhardt, J. Sci. Food Agric. 1991, 55, 313.
[29] T. N. Ly, R. Yamaguchi, M. Shimoyamada, K. Kato, J. Agric. Food Chem. 2002, 50, 4919.
[30] S. Hara, H. Okabe, K. Mihashi, Chem. Pharm. Bull. 1986, 34, 1843.
[31] T. Murakami, A. Emoto, H. Matsuda, M. Yoshikawa, Chem. Pharm. Bull. 2001, 49, 54.
[32] S. Nakamura, T. Murakami, J. Nakamura, H. Kobayashi, H. Matsuda, M. Yoshikawa, Chem. Pharm.
Bull. 2006, 54, 1545.
[33] H. Fukuda, C. Chen, C. Mantyh, K. Ludwig, T. N. Pappas, T. Takahashi, J. Surg. Res. 2006, 131, 290.
[34] E. P. Bouras, M. Camilleri, D. D. Burton, G. Thomforde, S. McKinzie, A. R. Zinsmeister,
Gastroenterology 2001, 120, 354.
[35] L.-K. Han, Y. Kimura, M. Kawashima, T. Takaku, T. Taniyama, T. Hayashi, Y.-N. Zheng, H. Okuda,
Int. J. Obesity 2001, 25, 1459.
Received April 21, 2008