Medicinal flowers. XXXI. Acylated oleanane-type triterpene saponins, sasanquasaponins I-V, with antiallergic activity from the flower buds of Camellia sasanqua

Article abstract of DOI:10.1248/cpb.58.1617

The methanolic extract and its 1-butanol-soluble fraction from the flower buds of Camellia sasanqua THUNB. were found to show inhibitory activities on the release of β-hexosaminidase from rat basophile leukemia (RBL-2H3) cells. From the 1-butanol-soluble

Full text of DOI:10.1248/cpb.58.1617

December 2010
Regular Article
Chem. Pharm. Bull. 58(12) 1617—1621 (2010)
1617
Medicinal Flowers. XXXI.1) Acylated Oleanane-Type Triterpene
Saponins, Sasanquasaponins I—V, with Antiallergic Activity from the
Flower Buds of Camellia sasanqua
Hisashi MATSUDA,a Seikou NAKAMURA,a Katsuyoshi FUJIMOTO,a Ryo MORIUCHI,a Yuta KIMURA,a
Noriko IKOMA,a Yuki HATA,a Osamu MURAOKA,b and Masayuki YOSHIKAWA*,a
a Kyoto Pharmaceutical University; Misasagi, Yamashina-ku, Kyoto 607–8412, Japan: and b Pharmaceutical Research and
Technology Institute, Kinki University; 3–4–1 Kowakae, Higashi-Osaka, Osaka 577–8502, Japan.
Received August 9, 2010; accepted September 13, 2010; published online September 14, 2010
The methanolic extract and its 1-butanol-soluble fraction from the flower buds of Camellia sasanqua THUNB.
were found to show inhibitory activities on the release of b-hexosaminidase from rat basophile leukemia (RBL-
2H3) cells. From the 1-butanol-soluble fraction, five new acylated oleanane-type triterpene saponins, sasanqua-
saponins I—V, were isolated together with a known saponin and their chemical structures were elucidated on the
basis of chemical and physicochemical evidence. The principal saponin constituents, sasanquasaponins I—III,
with an acyl group at the 22-position of the aglycon part showed the inhibitory effects on the release of b-hex-
osaminidase and some structure–activity relationships were reported.
Key words Camellia sasanqua; sasanquasaponin; medicinal flower; antiallergic activity; degranulation inhibitor; rat basophile
leukemia 2H3 cell
A Theaceae plant, Camellia (C.) sasanqua THUNB. (Japan- were found to exhibit antiallergic, antidiabetic, antiobestic,
ese name “sazanka”), which is native to Japan, has been gastroprotective, and platelet aggregation activities, etc.5—13)
widely cultivated as an ornamental plant in Japanese garden. As a continuation of our studies on bioactive constituents of
The flower buds of this plant are used for the similar purpose the flower buds of Camellia species, we found that the
to those of C. japonica (Japanese name “tsubaki”), which has methanolic extract and its 1-butanol-soluble fraction from the
been used for the treatment of blood vomiting and bleeding flower buds of C. sasanqua inhibited an immediate allergic
due to internal and external injury, and also as antiinflama- reaction by monitoring the release of b-hexosaminidase from
tory, tonic, and stomachic in Japanese folk medicine. As rat basophile leukemia (RBL-2H3) cells.14) From the 1-bu-
chemical constituents of this medicinal flower, several hy- tanol-soluble fraction, we have isolated new acylated
drolyzable tannins,2) acylated anthocyanins,3) and purin alka- oleanane-type triterpene saponins termed sasanquasaponins I
loids4) were reported. However, the pharmacological activi- (1), II (2), III (3), IV (4), and V (5), together with a known
ties of the flower buds of this plant have not yet character- saponin, primulagenin A 3-O-[b-D-glucopyranosyl(1→2)]
ized. Recently, we have reported the isolation and structure [a-L-rhammopyranosyl(1→2)-b-D-galactopyranosyl(1→3)]-
elucidation of triterpene saponins from the flower buds of C. b-D-glucuronopyranoside (6) (Chart 1).15) In addition, we ex-
japonica L. (camelliosides A—D),5,6) C. sinensis L. (chaka- amined the inhibitory effects of sasanquasaponins (1—5) and
saponins I—VI, floratheasaponins A—J),7—12) and C. oleifera 6 on the release of b-hexosaminidase from RBL-2H3 cells.
ABEL (yuchasaponins A—D).13) Furthermore, those saponins In this paper, we describe the isolation and structure elucida-
Chart 1. Saponin Constituents from the Flower Buds of C. sasanqua
∗ To whom correspondence should be addressed. e-mail: myoshika@mb.kyoto-phu.ac.jp
© 2010 Pharmaceutical Society of Japan

1618
Vol. 58, No. 12
Table 1. 13C-NMR (125 MHz) Data for 1—5
Carbon
1a)
2a)
3a)
4a)
5a)
5b)
Carbon
1a)
2a)
3a)
4a)
5a)
5b)
1
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
38.9
26.5
89.3
39.6
55.5
18.8
36.6
41.6
47.1
36.9
23.9
124.8
144.4
47.8
67.5
74.5
45.1
41.4
47.0
32.0
41.6
72.1
27.8
16.7
15.8
17.5
21.3
62.7
33.5
25.1
38.8
26.4
89.6
39.5
55.4
18.7
36.6
41.6
47.0
36.8
23.9
124.7
144.4
47.8
67.5
74.6
45.3
41.4
46.9
31.9
41.4
72.4
27.8
16.7
15.7
17.5
21.2
62.8
33.4
25.1
38.8
26.3
89.6
39.5
55.4
18.7
36.6
41.6
47.0
36.8
23.8
124.7
144.3
47.6
67.4
74.9
45.1
41.6
46.9
31.9
41.4
72.6
27.8
16.6
15.6
17.4
21.2
62.8
33.4
25.1
38.9
26.4
89.6
39.5
55.4
18.6
36.5
41.3
47.1
36.8
23.9
124.8
144.3
47.9
67.4
73.4
44.8
42.0
47.0
31.7
46.0
74.1
27.8
16.6
15.7
17.6
21.2
66.4
33.6
25.1
38.7
26.4
89.8
39.6
55.7
18.3
32.9
40.0
46.8
36.7
23.7
123.2
142.3
41.6
31.6
71.9
44.1
41.6
47.5
32.9
45.1
72.6
27.9
16.6
15.6
16.8
27.4
69.2
33.8
25.3
39.9
27.0
92.2
40.6
56.9
19.2
33.9
41.1
48.0
37.8
24.6
125.1
142.7
42.5
32.0
72.3
44.6
42.7
48.2
32.2
45.0
73.8
28.3
16.9
16.2
17.2
27.8
69.9
33.9
25.4
22-O-Angeloyl
1Љٞ
2Љٞ
3Љٞ
4Љٞ
5Љٞ
167.7
129.7
136.3
15.7
20.8
16-O-(2-Methylbutyloyl)
1Љٞ
2Љٞ
3Љٞ
4Љٞ
5Љٞ
175.8
42.2
27.1
12.0
16.5
177.6
43.2
27.8
12.2
16.5
3-O-b-D-Glucuronopyranosyl
1Ј
2Ј
3Ј
4Ј
5Ј
6Ј
105.4
79.4
82.4
71.2
77.1
172.3
105.3
79.3
82.5
71.2
76.9
172.3
105.3
79.3
82.5
71.1
76.9
172.3
105.4
79.4
82.4
71.2
77.1
172.3
105.9
79.4
82.5
71.2
77.0
172.2
105.9
79.0
81.1
71.7
77.0
172.1
2Ј-O-b-D-Glucopyranosyl
1Љ
2Љ
3Љ
4Љ
5Љ
6Љ
102.6
76.4
78.4
72.7
78.3
63.6
102.6
76.3
78.3
72.7
78.1
63.5
102.5
76.3
78.3
72.4
78.1
63.5
102.6
76.4
78.4
72.7
78.3
63.6
102.7
76.2
78.4
72.7
78.2
63.6
102.6
76.1
78.3
72.6
78.2
63.6
3Ј-O-b-D-Galactopyranosyl
1ٞ
2ٞ
3ٞ
4ٞ
5ٞ
6ٞ
101.1
76.1
76.0
71.2
77.4
62.1
101.1
76.2
76.0
71.1
77.3
61.9
101.2
76.1
75.9
71.2
77.3
61.8
101.1
76.1
76.0
71.2
77.4
62.1
101.3
76.2
76.0
71.2
77.4
61.9
100.9
75.9
75.9
71.7
77.9
62.8
22-O-(cis-2-Hexenoyl)
1Љٞ
2Љٞ
3Љٞ
4Љٞ
5Љٞ
6Љٞ
166.5
121.0
149.3
31.1
22.3
13.9
2ЉЉ-O-a-L-Rhamnopyranosyl
1ЉЉ
2ЉЉ
3ЉЉ
4ЉЉ
5ЉЉ
6ЉЉ
102.3
72.4
72.5
73.9
69.8
18.2
102.3
72.5
72.5
73.9
69.7
18.2
102.2
72.5
72.5
73.7
69.7
18.1
102.3
72.4
72.5
73.9
69.8
18.2
102.4
72.5
72.5
73.9
69.8
18.2
102.1
72.6
72.3
73.9
70.3
17.9
22- or 28-O-Tigloylc)
1Љٞ
2Љٞ
3Љٞ
4Љٞ
5Љٞ
167.7
129.7
135.7
13.9
167.7
129.1
137.1
14.1
12.2
12.2
a) Measured in pyridine-d5. b) Measured in CD3OD. c) 2: 22-O-Tig; 4: 28-O-Tig.
tion of the new saponins (1—5) and the inhibitory effects of sasanquasaponins I (1, 0.0032%), II (2, 0.015%), III (3,
the saponin constituents on the release of b-hexosaminidase 0.0057%), IV (4, 0.0010%), and V (5, 0.0012%) together
from RBL-2H3 cells.
with 6 (0.0037%).
The methanolic extract (10.41% from the fresh flower
Structures of Sasanquasaponins I—V (1—5) Sasan-
buds of C. sasanqua cultivated in Kyoto province in Japan) quasaponin I (1) was isolated as colorless fine crystals of mp
with the inhibitory effect on the release of b-hexosaminidase 213.0—215.0 °C (from aqueous MeOH) with negative opti-
from RBL-2H3 cells [inhibition (%): 36.7Ϯ1.5 (pϽ0.01) at cal rotation ([a]D20 Ϫ24.6° in MeOH). The IR spectrum of 1
100 mg/ml] was partitioned into an EtOAc–H2O (1 : 1, v/v) showed absorption bands at 3370, 1720, 1698, 1075, and
mixture to furnish an EtOAc-soluble fraction (1.51%) and 1047 cmϪ1 due to hydroxy, a,b-unsaturated ester, carboxy
aqueous layer. The aqueous layer was further extracted with and ether functions. In the positive-ion FAB-MS of 1, a qua-
1-butanol to give a 1-butanol- (3.90%) and an H2O- (5.00%) simolecular ion peak (MϩNa)ϩ was observed at m/z 1255
soluble fractions. The 1-butanol-soluble fraction inhibited the and the molecular formula C60H96O26 was determined by
degranulation in RBL-2H3 cells [inhibition (%) 53.3Ϯ2.9 high-resolution MS measurement of the quasimolecular ion
(pϽ0.01) at 30 mg/ml], but the EtOAc- and the H2O-soluble peak. Treatment of 1 with 10% aqueous KOH–50% aqueous
fractions showed weak [inhibition (%): 15.2Ϯ2.2 (pϽ0.01) 1,4-dioxane (1 : 1) mixture liberated a known compound, de-
at 100 mg/ml] or no activity, respectively. The 1-butanol- sacyl-boninsaponin A (1a).16) The 1H-NMR (pyridine-d5) and
soluble fraction was subjected to HP-20 and reversed-phase 13C-NMR (Table 1) spectra of 1, which were assigned by var-
ODS column chromatographies and repeated HPLC to give ious NMR experiments,17) showed signals assignable to a A1-

December 2010
1619
barrigenol part: seven methyls [d 0.79, 1.01, 1.05, 1.07, 1.12, 28-positions. The position of tigloyl group in 4 was con-
1.27, 1.87 (all s, H3-25, 26, 29, 24, 23, 30, 27)], a methylene firmed by a HMBC experiment, which showed a long-range
[d 3.59, 3.76 (both d, Jϭ10.3 Hz, H2-28)] and three methines correlation between the 28-methylene protons and the car-
bearing an oxygen function [d 3.21 (dd, Jϭ5.0, 12.0 Hz, H- bonyl carbon of tigloyl group. Comparison of the NMR data
3), 4.24 (d-like, H-15), 6.18 (dd, Jϭ5.5, 11.7 Hz, H-22)], an for 4 with those for 2 and 1a led us to confirm the structure
olefin [d 5.49 (br s, H-12)], and four glycopyranosyl moieties of 4 to be the 28-tigloyl isomer of 2.
{b-D-glucuronopyranosyl [d 4.84 (d, Jϭ7.0 Hz, H-1Ј)], b-D-
Sasanquasaponin V (5), obtained as colorless fine crystals
glucopyranosyl [d 5.94 (d, Jϭ7.6 Hz, H-1Љ)], b-D-galactopy- (mp 224.0—226.0 °C from aqueous MeOH) with negative
ranosyl [d 6.26 (d, Jϭ7.0 Hz, H-1ٞ)], and an a-L-rhamnopy- optical rotation ([a]D21 Ϫ32.4° in MeOH), showed absorption
ranosyl [d 6.24 (br s, H-1ЉЉ)] together with a cis-2-hexenoyl bands due to hydroxy, ester, carboxy and ether functions in
group [d 0.84 (t, Jϭ7.6 Hz, H3-6Љٞ)], 1.35 (m, H2-5Љٞ), 2.71, the IR spectrum (3370, 1730, 1717, 1078, 1047 cmϪ1). The
2.78 (both m, H2-4Љٞ), 5.79 (d, Jϭ11.0 Hz, H-2Љٞ), 6.09 (td, positive FAB-MS of 5 showed a quasimolecular ion peak at
Jϭ7.6, 11.0 Hz, H-3Љٞ)}.18) The position of the acyl group m/z 1227 (MϩNa)ϩ and the molecular formula C59H96O25
and the structure of the oligoglycoside moiety were con- was determined by high-resolution MS measurement. Alka-
firmed on the basis of a heteronuclear multiple bond connec- line hydrolysis of 5 provided ternstroemiaside A (5a)20) and
tivity spectroscopy (HMBC) experiment. Thus, the long- (S)-(ϩ)-2-methylbutyric acid, which was identified by HPLC
1
range correlations were observed between the following pro- analysis using an optical rotation detector. The H-NMR
ton and carbon pairs: H-22 and C-1Љٞ; H-1Ј and C-3; H-1Љ (pyridine-d5) and 13C-NMR (Table 1) spectra of 5 showed
and C-2Ј; H-1ٞ and C-3Ј; H-1ЉЉ and C-2ٞ. Furthermore, signals due to a ternstroemiaside A part and (S)-(ϩ)-2-methyl-
comparison of the 13C-NMR data for 1 with those for 1a re- butyroyl group [d 1.00 (t, Jϭ7.6 Hz, H3-4Љٞ), 1.30 (d, Jϭ
vealed an acylation shift around the 22-position of the A1- 6.9 Hz, H3-5Љٞ), 1.63, 1.94 (both m, H2-3Љٞ), 2.59 (m, H-2Љٞ)].
barrigenol part. On the basis of the above mentioned evi- The position of the acyl group in 5 was elucidated by a
dence, the chemical structure of sasanquasaponin I was de- HMBC experiment and comparison the 13C-NMR data for 5
termined to be 22-O-cis-2-hexenoyl-A1-barrigenol 3-O-[b-D- with those for 5a,20) so that the structure of sasanquasaponin
glucopyranosyl(1→2)][a-L-rhamnopyranosyl(1→2)-b-D- V (5) was characterized to be as shown.
galactopyranosyl(1→3)]-b-D-glucuronopyranoside (1).
Inhibitory Effects of Saponin Constituents (1—6) on
Sasanquasaponins II (2), III (3), and IV (4), which were the Release of b-Hexosaminidase from RBL-2H3 Cells
obtained as colorless fine crystals [2: mp 219.5—221.0 °C, 3: In the course of our studies on antiallergic constituents from
mp 227.0—229.0 °C, 4: mp 215.0—217.0 °C (from aqueous natural medicines,21—32) we previously reported that some
MeOH)] with negative optical rotation (2: [a]D20 Ϫ18.2°, 3: triterpenes and oleanane-type triterpene oligoglycosides
[a]D21 Ϫ18.5°, 4: [a]D20 Ϫ31.9° in MeOH), showed absorption showed inhibitory effects on histamine release from rat exu-
bands due to hydroxy, a,b-unsaturated ester, carboxy, and date cells induced by an antigen–antibody reaction33,34) and
ether functions in their IR spectra. The common molecular on b-hexosaminidase release induced by dinitrophenylated
formula C59H94O26 of 2, 3, and 4 was determined individu- bovine serum albumin (DNP-BSA) from RBL-2H3 cells sen-
ally from the quasimolecular ion peak [m/z 1241 (MϩNa)ϩ] sitized with anti-DNP immunoglobulin E (IgE).8,14) As a con-
in the positive-ion FAB-MS and by high-resolution MS tinuation of this study, we examined the effects of acylated
measurement. Alkaline treatment of 2, 3, and 4 yielded de- oleanane-type triterpene saponins, sasanquasaponins, on the
sacyl-boninsaponin A (1a) and organic acids (tiglic acid release of b-hexosaminidase from RBL-2H3 cells. As shown
from 2 and 4; angelic acid from 3), which were identical by in Table 2, sasanquasaponins I—III (1—3) with the 22-acyl
1
HPLC analysis of their p-nitrobenzyl derivatives.19) The H- group were found to exhibit strong activities regardless of the
NMR (pyridine-d5) and 13C-NMR (Table 1) spectra17) of 2 in- structures of their acyl groups. Whereas sasanquasaponin V
dicated the presence of desacyl-boninsaponin A part and a (5) with the 16-acyl group showed weak activity, and also
tigloyl group [d 1.43 (d, Jϭ6.9 Hz, H3-4Љٞ), 1.78 (s, H3-5Љٞ), sasanquasaponin IV (4) with the 28-acyl group and 6 showed
6.84 (q, Jϭ6.9 Hz, H-3Љٞ)]. On the other hand, the proton no activity. As a result, the 22-acyl group in the saponin
and carbon signals in the 1H- and 13C-NMR (Table 1) structures seems to be essential for the potent inhibitory
spectra17) of 3 were superimposable on those of 2, except for effect on b-hexosaminidase release from RBL-2H3 cells.
the signals due to the 22-acyl group of 3. The 1H-NMR (pyri-
In conclusion, five new acylated oleanane-type triterpene
dine-d5) and 13C-NMR spectra of 3 showed signals assigna- saponins, sasanquasaponins I—V (1—5), were isolated from
ble to an angeloyl group [d 1.82 (s, H3-5Љٞ), 2.02 (d, the flower buds of C. sasanqua and their chemical structures
Jϭ6.9 Hz, H3-4Љٞ), 5.83 (q, Jϭ6.9 Hz, H-3Љٞ)] together with were elucidated on the basis of chemical and physicochemi-
a desacyl-boninsaponin A part. The positions of the both acyl cal evidence. In addition, sasanquasaponins I—III (1—3)
groups in 2 and 3 were clarified by HMBC experiments, having the 22-acyl group were found to inhibit the b-hexo-
which showed long-range correlations between the 22-pro- saminidase release from RBL-2H3.
tons and the carbonyl carbons of the tigloyl and angeloyl
Experimental
1
groups. Furthermore, comparison of the H- and 13C-NMR
General Experimental Procedures The following instruments were
data for 2 and 3 with those for 1a revealed an acylation shift
around the 22-postion of the A1-barrigenol part. Conse-
quently, the chemical structures of sasanquasaponins II (2)
and III (3) were characterized to be as shown. The proton and
carbon signals in the 1H- and 13C-NMR spectra of 4 was also
used to obtain physical data: specific rotations, Horiba SEPA-300 digital po-
larimeter (lϭ5 cm); IR spectra, Shimadzu Fourier transform-infrared (FT-
IR)-8100 spectrometer; electron ionization-mass spectra (EI-MS) and high
resolution (HR)-EI-MS, JEOL JMS-GCMATE mass spectrometer; FAB-MS
1
and HR-FAB-MS, JEOL JMS-SX 102A mass spectrometer; H-NMR spec-
tra, JEOL JNM-EX 270 (270 MHz), JEOL JNM-LA 500 (500 MHz), and
similar to those of 2, except for the signals due to the 22- and JEOL JNM-ECA 600 (600 MHz) spectrometers; 13C-NMR spectra, JEOL

1620
Vol. 58, No. 12
Table 2. Inhibitory Effects of Sasanquasaponins (1—5) and 6 from C.
sasanqua on the Release of b-Hexosaminidase from RBL-2H3 Cells
Sasanquasaponin I {ϭ22-O-cis-2-Hexenoyl-A1-barrigenol 3-O-[b-D-Glu-
copyranosyl(1→2)][a -L-rhamnopyranosyl(1→2)-b -D-galactopyra
nosyl(1→3)]-b-D-glucuronopyranoside, 1}: Colorless fine crystals; mp
213.0—215.0 °C; [a]D20 Ϫ24.6° (cϭ0.85, MeOH); IR (KBr) nmax: 3370,
Conc.(mM)
Inhibition (%)
1720, 1698, 1075, 1047 cmϪ1 1H-NMR (pyridine-d5, 600 MHz) d: 0.79,
;
Sasanquasaponin I (1)
Sasanquasaponin II (2)
Sasanquasaponin III (3)
Sasanquasaponin IV (4)
Sasanquasaponin V (5)
6
0.3
1
3
0.3
1
3
0.3
1
3
1
3
6
3
6
10
3
11.6Ϯ5.2
41.3Ϯ2.9**
36.5Ϯ1.4**
10.7Ϯ4.6
17.9Ϯ5.6**
40.2Ϯ1.7**
14.9Ϯ6.0
24.4Ϯ3.0*
51.6Ϯ3.0**
9.4Ϯ4.5
1.01, 1.05, 1.07, 1.12, 1.27, 1.87 (all s, H3-25, 26, 29, 24, 23, 30, 27), 0.84
(t, Jϭ7.6 Hz, H3-6Љٞ), 1.35 (m, H2-5Љٞ), 2.71, 2.78 (both m, H2-4Љٞ), 3.21
(1H, dd, Jϭ5.0, 12.0 Hz, H-3), 3.59, 3.76 (both d, Jϭ10.3 Hz, H2-28), 4.24
(1H, d-like, H-15), 4.84 (d, Jϭ7.0 Hz, H-1Ј), 5.49 (1H, br s, H-12), 5.79 (d,
Jϭ11.0 Hz, H-2Љٞ), 5.94 (1H, d, Jϭ7.6 Hz, H-1Љ), 6.09 (td, Jϭ7.6, 11.0 Hz,
H-3Љٞ), 6.18 (1H, dd, Jϭ5.5, 11.7 Hz, H-22), 6.24 (1H, br s, H-1ЉЉ), 6.26
(1H, d, Jϭ7.0 Hz, H-1ٞ); 13C-NMR: given in Table 1; positive-ion FAB-MS
m/z: 1255 [MϩNa]ϩ; high-resolution positive-ion FAB-MS m/z: 1255.6080
(Calcd for C60H96O26Na [MϩNa]ϩ: m/z 1255.6088).
Sasanquasaponin II {ϭ22-O-Tigloyl-A1-barrigenol 3-O-[b-D-Glucopyra-
nosyl(1→2)][a-L-rhamnopyranosyl(1→2)-b-D-galactopyranosyl(1→3)]-b-
D-glucuronopyranoside, 2}: Colorless fine crystals; mp 219.5—221.0 °C;
[a]D29 Ϫ18.2° (cϭ0.43, MeOH); IR (KBr) nmax: 3370, 1720, 1684, 1075,
1045 cmϪ1; 1H-NMR (pyridine-d5, 600 MHz) d: 0.80, 1.02, 1.06, 1.06, 1.14,
1.28, 1.87 (all s, H3-25, 26, 29, 24, 23, 30, 27), 1.43 (3H, d, Jϭ6.9 Hz, H3-
4Љٞ), 1.78 (3H, s, H3-5Љٞ), 3.19 (1H, dd, Jϭ5.0, 11.0 Hz, H-3), 3.61, 3.77
(both d, Jϭ10.3 Hz, H2-28), 4.24 (1H, d-like, H-15), 4.84 (1H, d, Jϭ7.0 Hz,
H-1Ј), 5.47 (1H, br s, H-12), 5.92 (1H, d, Jϭ7.4 Hz, H-1Љ), 6.17 (1H, dd, Jϭ
5.5, 11.7 Hz, H-22), 6.21 (1H, d, Jϭ7.6 Hz, H-1ٞ), 6.25 (1H, br s, H-1ЉЉ),
6.84 (1H, q, Jϭ6.9 Hz, H-3Љٞ); 13C-NMR: given in Table 1; positive-ion
FAB-MS m/z: 1241 [MϩNa]ϩ; high-resolution positive-ion FAB-MS m/z:
1241.5923 (Calcd for C59H94O26Na [MϩNa]ϩ: m/z 1241.5931).
2.4Ϯ4.7
Ϫ13.1Ϯ5.7
0.7Ϯ3.2
27.0Ϯ7.2**
39.4Ϯ4.4**
Ϫ5.6Ϯ4.1
Ϫ8.3Ϯ6.3
Ϫ4.6Ϯ2.1
25.4Ϯ4.3**
44.5Ϯ4.4**
10.8Ϯ5.4
6
10
30
100
30
100
Tranilast32)
Ketorifen fumarate32)
30.6Ϯ4.8**
Sasanquasaponin III {ϭ22-O-Angeloyl-A1-barrigenol 3-O-[b-D-Glucopy-
ranosyl(1→2)][a-L-rhamnopyranosyl(1→2)-b-D-galactopyranosyl(1→3)]-
b-D-glucuronopyranoside, 3}: Colorless fine crystals; mp 227.0—229.0 °C;
[a]D21 Ϫ18.5° (cϭ0.72, MeOH); IR (KBr) nmax: 3370, 1720, 1699, 1076,
1043 cmϪ1: 1H-NMR (pyridine-d5, 600 MHz) d: 0.79, 1.00, 1.05, 1.06, 1.12,
1.28, 1.86 (all s, H3-25, 26, 29, 24, 23, 30, 27), 1.82 (3H, s, H3- 5Љٞ), 2.02
(3H, d, Jϭ6.9 Hz, H3-4Љٞ), 3.18 (1H, dd, Jϭ5.0, 11.0 Hz, H-3), 3.63, 3.79
(1H each, both d, Jϭ10.3 Hz, H2-28), 4.25 (1H, d-like, H-15), 4.89 (d, Jϭ
7.0 Hz, H-1Ј), 5.46 (1H, br s, H-12), 5.83 (1H, q, Jϭ6.9 Hz, H-3Љٞ), 5.93 (d,
Jϭ7.0 Hz, H-1Љ), 6.19 (1H, dd, Jϭ5.5, 11.7 Hz, H-22), 6.22 (d, Jϭ7.6 Hz,
H-1ٞ), 6.25 (br s, H-1ЉЉ); 13C-NMR: given in Table 1; positive-ion FAB-MS
m/z: 1241 [MϩNa]ϩ; high-resolution positive-ion FAB-MS m/z: 1241.5923
(Calcd for C59H94O26Na [MϩNa]ϩ: m/z 1241.5931).
Sasanquasaponin IV {ϭ28-O-Tigloyl-A1-barrigenol 3-O-[b-D-Glucopyra-
nosyl(1→2)][a-L-rhamnopyranosyl(1→2)-b-D-galactopyranosyl(1→3)]-b-
D-glucuronopyranoside, 4}: Colorless fine crystals; mp 215.0—217.0 °C;
[a]D21 Ϫ31.9° (cϭ0.17, MeOH); IR (KBr) nmax: 3370, 1716, 1697, 1078,
1045 cmϪ1: 1H-NMR (pyridine-d5, 600 MHz) d: 0.79, 1.05, 1.09, 1.12, 1.15,
1.21, 1.87 (all s, H3-25, 24, 29, 23, 26, 30, 27), 1.60 (3H, d, Jϭ6.8 Hz, H3-
4Љٞ), 1.84 (3H, s, H3-5Љٞ), 3.19 (1H, dd, Jϭ5.0, 11.7 Hz, H-3), 4.38, 4.61
(1H each, both d, Jϭ11.0 Hz, H2-28), 4.35 (1H, d-like, H-15), 4.59 (1H, dd,
Jϭ5.5, 11.7 Hz, H-22), 4.91 (d, Jϭ7.0 Hz, H-1Ј), 5.52 (1H, br s, H-12), 5.94
(d, Jϭ7.4 Hz, H-1Љ), 6.25 (d, Jϭ7.6 Hz, H-1ٞ), 6.26 (br s, H-1ЉЉ), 7.00 (1H,
q, Jϭ6.9 Hz, H-3Љٞ); 13C-NMR: given in Table 2; positive-ion FAB-MS m/z:
1241 [MϩNa]ϩ; high-resolution positive-ion FAB-MS m/z: 1241.5933
(Calcd for C59H94O26Na [MϩNa]ϩ: m/z 1241.5931).
Values represent the meansϮS.E.M. Significantly different from the control group,
∗ pϽ0.05, ∗∗ pϽ0.01.
JNM-EX 270 (68 MHz) JEOL JNM-LA (125 MHz), and JEOL JNM-ECA
600 (150 MHz) spectrometers with tetramethylsilane as an internal standard;
HPLC detector, Shimadzu RID-10A refractive index detector; and HPLC
column, COSMOSIL 5C18-MS-II (250ϫ4.6 mm i.d.) and (250ϫ20 mm i.d.)
columns were used for analytical and preparative purposes, respectively.
The following experimental materials were used for chromatography: nor-
mal-phase silica gel column chromatography, Silica gel BW-200 (Fuji Silysia
Chemical, Ltd., 150—350 mesh); reversed-phase silica gel column chro-
matography, Chromatorex ODS DM1020T (Fuji Silysia Chemical, Ltd.,
100—200 mesh); TLC, precoated TLC plates with Silica gel 60F254 (Merck,
0.25 mm) (ordinary phase) and Silica gel RP-18 F254S (Merck, 0.25 mm) (re-
versed phase); reversed-phase HPTLC, precoated TLC plates with Silica gel
RP-18 WF254S (Merck, 0.25 mm); and detection was achieved by spraying
with 1% Ce(SO4)2–10% aqueous H2SO4 followed by heating.
Plant Material C. sasanqua was cultivated in the medicinal plant gar-
den of Kyoto Pharmaceutical University and the flower buds were collected
in February, 2008. A voucher of the plant is on file in our laboratory (2008.
KPU-CS).
Extraction and Isolation The fresh flower buds (4.0 kg) were extracted
three times with methanol under reflux for 3 h. Evaporation of the solvent
under reduced pressure provided a methanolic extract (417.7 g, 10.41%).
The MeOH extract (400 g) was partitioned into an EtOAc–H2O (1 : 1, v/v)
mixture to furnish an EtOAc-soluble fraction (57.9 g, 1.51%) and an aque-
ous phase. The aqueous phase was further extracted with 1-butanol to give a
1-butanol-soluble fraction (149.5 g, 3.90%) and an H2O-solbule fraction
(191.7 g, 5.00%). A part of the 1-butanol-soluble fraction (143.5 g) was sub-
jected to HP-20 column chromatography (H2O–MeOH–acetone) to furnish
an H2O-eluted fraction (88.2 g, 2.40%), a MeOH-eluted fraction (54.1 g,
1.47%), and an acetone-eluted fraction (0.9 g, 0.024%). The MeOH-elutaed
fraction (54.1 g) was subjected to reversed phase silica gel column chro-
matography [H2O→MeOH–H2O (10 : 90→20 : 80→30 : 40→40 : 60→50 :
50→60 : 40→70 : 30→80 : 20→90 : 10, v/v)→MeOH] to give seven frac-
tions [Fr. 1 (0.83 g), Fr. 2 (3.28 g), Fr. 3 (27.96 g), Fr. 4 (6.14 g), Fr. 5
(8.89 g), Fr. 6 (0.88 g), Fr. 7 (0.64 g)]. Fraction 5 (8.89 g) was further sepa-
rated by reversed phase silica gel column chromatography [MeOH : H2O
(50 : 50→60 : 40→70 : 30→80 : 20→90 : 10)→MeOH] to give eight fractions
[Fr. 5-1, Fr. 5-2, Fr. 5-3, Fr. 5-4, Fr. 5-5 (3.69 g), Fr. 5-6 (3.05 g), Fr. 5-7, Fr.
5-8]. A part of fraction 5-5 (750 mg) was purified by HPLC [MeOH in 1%
AcOH : H2O (68 : 32)] to furnish sasanquasaponin II (2, 109 mg, 0.015%),
III (3, 42 mg, 0.0057%), and 6 (28 mg, 0.0037 %). A part of fraction 5-6
(1.50 g) was purified by HPLC [MeOH in 1% AcOH : H2O (73 : 27)] to give
sasanquasaponin I (1, 57 mg, 0.0032%), IV (4, 18 mg, 0.0010%), and V (5,
23 mg, 0.0012%).
Sasanquasaponin V {ϭ16-O-(S)-(ϩ)-2-Methylbutyroyl-camelliagenin A
3-O-[b-D-Glucopyranosyl(1→2)][a-L-rhamnopyranosyl(1→2)-b-D-galac-
topyranosyl(1→3)]-b-D-glucuronopyranoside, 5}: Colorless fine crystals;
mp 224.0—226.0 °C; [a]D21 Ϫ32.4° (cϭ0.31, MeOH); IR (KBr) nmax: 3370,
1730, 1717, 1078, 1047 cmϪ1 1H-NMR (pyridine-d5, 600 MHz) d: 0.75,
:
0.78, 1.05, 1.10, 1.12, 1.16, 1.54 (all s, H3-25, 26, 24, 29, 30, 23, 27), 1.00
(3H, t, Jϭ7.6 Hz, H3-4Љٞ), 1.30 (3H, d, Jϭ6.9 Hz, H3-5Љٞ), 1.63, 1.94 (1H
each, both m, H2-3Љٞ), 2.59 (1H, m, H-2Љٞ), 3.17 (1H, dd, Jϭ4.9, 11.1 Hz,
H-3), 3.66, 4.04 (1H each, both d, Jϭ10.3 Hz, H2-28), 4.09 (1H, dd like, H-
22), 4.90 (d, Jϭ7.0 Hz, H-1Ј), 5.32 (1H, br s, H-12), 5.94 (d, Jϭ7.7 Hz, H-
1Љ), 6.22 (m, H-16), 6.22 (d, Jϭ7.6 Hz, H-1ٞ), 6.25 (br s, H-1ЉЉ); 13C-NMR:
given in Table 2; positive-ion FAB-MS m/z: 1227 [MϩNa]ϩ; high-resolution
positive-ion FAB-MS m/z: 1227.6143 (Calcd for C59H96O25Na [MϩNa]ϩ:
m/z 1227.6138).
Alkaline Hydrolysis of Sasanquasaponins I—V (1—5) A solution of
sasanquasaponins I—V (1: 10 mg; 2–5: each 6.0 mg) was treated with 10%
aqueous KOH–1,4-dioxane (1 : 1, v/v, 1.0 ml) and the whole was stirred at
37 °C for 2 h, respectively. The reaction mixture was neutralized with Dowex
HCR W2 (Hϩ form) and the resin was removed by filtration. Evaporation of
the solvent from the filtrate under reduced pressure yielded a crude product,
whose most part was subjected to ordinary-phase silica gel column chro-

December 2010
1621
matography [CHCl3–MeOH–H2O (10 : 3 : 1→6 : 4 : 1, v/v/v)] to give 1a (5.1
mg from 1, 2.5 mg, 2.3 mg, 2.3 mg from 2—4, respectively) or 5a (2.4 mg
from 5). The desacyl-derivatives (1a, 5a) were identified by comparison of
their physical data with those of the reported values. A part (each 1.0 mg) of
crude product obtained from 2—4 was dissolved in (CH2)2Cl2 (0.5 ml), re-
spectively. The solution was treated with p-nitrobenzyl-N-NЈ-diisopyopy-
lisourea (4 mg), then the whole was stirred at 80 °C for 1 h. The reaction so-
lution was subjected to HPLC analysis [column: YMC-Pack ODS-A,
250ϫ4.6 mm i.d.; mobile phase: MeCN–H2O (50 : 50, v/v); detection: UV
(254 nm); flow rate: 1.0 ml/min; column temperature: room temperature] to
identify the p-nitrobenzyl esters of tiglic acid (tR 30.0 min, from 2, 4) and
angelic acid (tR 33.9 min, from 3). On the other hand, a part (1.0 mg) of
crude product obtained from 5 was subjected to HPLC analysis [column:
COSMOSIL 5C18-MS-II, 250ϫ4.6 mm i.d.; mobile phase: MeOH in 1%
AcOH–H2O (50 : 50, v/v); detection: optical rotation [Shodex OR-2 (Showa
Denko Co., Ltd., Tokyo, Japan)]; flow rate: 0.8 ml/min; column temperature:
room temperature] to identify the (S)-(ϩ)-2-methylbutyric acid. Identifica-
tion of (S)-(ϩ)-2-methylbutyric acid in the crude product obtained from 5
was carried out by comparison of its retention time and optical rotation with
References and Notes
1) Part XXX: Morikawa T., Wang L. B., Ninomiya K., Nakamura S.,
Matsuda H., Muraoka O., Wu L. J., Yoshikawa M., Chem. Pharm.
Bull., 57, 853—859 (2009).
2) Yoshida T., Chou T., Maruyama Y., Okuda T., Chem. Pharm. Bull., 38,
2681—2686 (1990).
3) Saito N., Yokoi M., Yamaji M., Honda T., Phytochemistry, 26, 2761—
2762 (1987).
4) Fujimori N., Ashihara H., Phytochemistry, 29, 3513—3516 (1990).
5) Yoshikawa M., Morikawa T., Fujiwara E., Ohgushi T., Asao Y., Ma-
tsuda H., Heterocycles, 55, 1653—1657 (2001).
6) Yoshikawa M., Morikawa T., Asao Y., Fujiwara E., Nakamura S., Ma-
tsuda H., Chem. Pharm. Bull., 55, 606—612 (2007).
7) Yoshikawa M., Morikawa T., Yamamoto K., Kato Y., Nagatomo A.,
Matsuda H., J. Nat. Prod., 68, 1360—1365 (2005).
8) Yoshikawa M., Nakamura S., Kato Y., Matsuhira K., Matsuda H.,
Chem. Pharm. Bull., 55, 598—605 (2007).
9) Yoshikawa M., Wang T., Sugimoto S., Nakamura S., Nagatomo A.,
Matsuda H., Harima S., Yakugaku Zasshi, 128, 141—151 (2008).
that of an authentic sample [tR: 9.2 min (positive optical rotation), commer- 10) Yoshikawa M., Sugimoto S., Nakamura S., Matsuda H., Chem. Pharm.
cial sample (Sigma-Aldrich, Japan)]. Bull., 56, 1297—1303 (2008).
Bioassay. Effects on the Release of b-Hexosaminidase from RBL-2H3 11) Sugimoto S., Yoshikawa M., Nakamura S., Matsuda H., Heterocycles,
Cells The inhibitory effects of the test samples on the release of b-hexo-
saminidase from RBL-2H3 cells [Cell No. JCRB0023, obtained from Health
Science Research Resources Bank (Osaka, Japan)] were evaluated by a
78, 1023—1029 (2009).
12) Yoshikawa M., Sugimoto S., Kato Y., Nakamura S., wang T., Ya-
mashita C., Matsuda H., Chem. Biodiv., 6, 903—915 (2009).
method reported previously21) with some modifications. Briefly, RBL-2H3 13) Sugimoto S., Chi G., Kato Y., Nakamura S., Matsuda H., Yoshikawa
cells were dispensed into 48-well plates at a concentration of 4ϫ104
cells/well using Eagle’s minimum essential medium (MEM, Sigma) contain- 14) Morikawa T., Nakamura S., Kato Y., Muraoka O., Matsuda H.,
ing fetal calf serum (10%), penicillin (100 units/ml), streptomycin (100 mg/ Yoshikawa M., Chem. Pharm. Bull., 55, 293—298 (2007).
M., Chem. Pharm. Bull., 57, 269—275 (2009).
ml), and 0.45 mg/ml of anti-DNP IgE, and these were incubated overnight at 15) Ohtani K., Mavi S., Hostettmann K., Phytochemistry, 33, 83—86
37 °C in 5% CO2 for sensitization of the cells. Then the cells were washed
twice with 200 ml of Siraganian buffer [119 mM NaCl, 5 mM KCl, 0.4 mM
MgCl2, 25 mM piperazine-N,NЈ-bis(2-ethanesulfonic acid) (PIPES), and
40 mM NaOH, pH 7.2], and incubated in 80 ml of Siraganian buffer [5.6 mM
glucose, 1 mM CaCl2, and 0.1% bovine serum albmin (BSA) were added] for
an additional 10 min at 37 °C. Aliquots (10 ml) of test sample solution were
added to each well and incubated for 10 min, followed by the addition of
10 ml of antigen (DNP-BSA, final concentration 10 mg/ml) at 37 °C for
(1993).
16) Kitagawa I., Yoshikawa M., Kobayashi K., Imakura Y., Im K. S., Ike-
nishi Y., Chem. Pharm. Bull., 28, 296—300 (1980).
1
17) The H- and 13C-NMR spectra of 1—5 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 connectivity spectroscopy (HMBC) experiments.
10 min to stimulate the cells to evoke allergic reactions (degranulation). The 18) Jiang Z., Gallard J.-F., Adeline M.-T., Dumontet V., Tri M. V., Sévenet
reaction was stopped by cooling in an ice bath for 10 min. The supernatant T., Païs M., J. Nat. Prod., 62, 873—876 (1999).
(40 ml) was transferred into a 96-well microplate and incubated with 40 ml of 19) Morikawa T., Li X., Nisida E., Nakamura S., Ninomiya K., Matsuda
substrate (1 mM p-nitrophenyl-N-acetyl-b-D-glucosaminide) in 0.1 M citrate
buffer (pH 4.5) at 37 °C for 2 h. The reaction was stopped by adding 200 ml
of stop solution (0.1 M Na2CO3/NaHCO3, pH 10.0). The absorbance was
measured using a microplate reader at 405 nm. The test sample was dis-
H., Oda Y., Muraoka O., Yoshikawa M., Helv. Chim. Acta, 93, 573—
587 (2010).
20) Shin M., Wang W., Nam K. I., Jo Y., Jung H. J., Im S. K., J. Nat. Prod.,
66, 1351—1355 (2003).
solved in dimethylsulfoxide (DMSO), and the solution was added to Sira- 21) Matsuda H., Morikawa T., Ueda K., Managi H., Yoshikawa M.,
ganian buffer (final DMSO concentration 0.1%). Bioorg. Med. Chem., 10, 3123—3128 (2002).
The percent inhibition of the release of b-hexosaminidase by the test ma- 22) Matsuda H., Morikawa T., Tao J., Ueda K., Yoshikawa M., Chem.
terial was calculated using the following equation:
Pharm. Bull., 50, 208—215 (2002).
23) Morikawa T., Matsuda H., Toguchida I., Veda K., Yoshikawa M., J.
Nat. Prod., 65, 1468—1474 (2002).
inhibition (%)ϭ[1Ϫ(TϪBϪN)/(CϪN)]ϫ100
24) Morikawa T., Matsuda H., Sakamoto Y., Ueda K., Yoshikawa M.,
Chem. Pharm. Bull., 50, 1045—1049 (2002).
25) Morikawa T., Tao J., Ueda K., Matsuda H., Yoshikawa M., Chem.
Pharm. Bull., 51, 62—67 (2003).
26) Tao J., Morikawa T., Ando S., Matsuda H., Yoshikawa M., Chem.
Pharm. Bull., 51, 654—662 (2003).
27) Matsuda H., Morikawa T., Xie H., Yoshikawa M., Planta Med., 70,
847—855 (2003).
28) Matsuda H., Morikawa T., Mangi H., Yoshikawa M., Bioorg. Med.
Chem. Lett., 13, 3197—3202 (2003).
29) Matsuda H., Tewtrakul S., Morikawa T., Nakamura S., Yoshikawa M.,
Bioorg. Med. Chem., 12, 5891—5989 (2004).
30) Sun B., Morikawa T., Matsuda H., Tewtrakal S., Wu L. J., Harima S.,
Yoshikawa M., J. Nat. Prod., 67, 1464—1469 (2004).
31) Morikawa T., Sun B., Matasuda H., Wu L. J., Harima S., Yoshikawa
M., Chem. Pharm. Bull., 52, 1194—1199 (2004).
32) Xu F., Matsuda H., Hata H., Sugawara K., Nakamura S., Yoshikawa
M., Chem. Pharm. Bull., 57, 1089—1095 (2009).
Control (C): DNP-BSA (ϩ), test sample (Ϫ); test (T): DNP-BSA (ϩ), test
sample (ϩ); blank (B): DNP-BSA (Ϫ), test sample (ϩ); normal (N): DNP-
BSA (Ϫ), test sample (Ϫ).
Under these conditions, it was calculated that 10—15% of b-hexo-
saminidase was released from the cells in the control groups by determina-
tion of the total b-hexosaminidase activity after treatment with 0.05% Triton
X-100.
In order to clarify that the antiallergic effects of samples were due to the
inhibition on hexosaminidase release, but not the false positive by the inhibi-
tion of b-hexosaminidase activity, the following assay was carried out. The
supernatant (36 ml) of the control group as an enzyme solution, the substrate
solution (40 ml), and test sample solution (4 ml) were transferred into a 96-
well microplate and enzyme activity was examined as described above.
Statistics Values were expressed as meansϮS.E.M. One-way analysis of
variance following Dunnett’s test was used for statistical analysis. Probabil-
ity (p) values less than 0.05 were considered significant.
Acknowledgments This research was supported by the 21st COE Pro-
gram, Academic Frontier Project, and a Grant-in Aid for Scientific Research
from the Ministry of Education, Culture, Sports, Science and Technology of
Japan and by Hoh-ansha Foundation, Japan. The authors thank Mr. Nobuyuki
Izumi for financial support.
33) Yoshikawa M., Shimada H., Komatsu H., Sakurama T., Nishida N.,
Shimoda H., Matsuda H., Tani T., Chem. Pharm. Bull., 45, 877—882
(1997).
34) Yoshizumi S., Murakami T., Kadoya M., Matsuda H., Yoshikawa M.,
Yakugaku Zasshi, 118, 188—192 (1998).