Stability and structure studies on alisol A 24-acetate

Article abstract of DOI:10.1248/cpb.56.41

Alisol A 24-acetate is one of the main active triterpenoid compounds isolated from Rhizoma Alismatis, which is a famous Traditional Chinese Medicine, and has been determined for the quality control of this crude drug. In this study, alisol A 24-acetate was found to be unstable in solvents and its stability in different solvents was investigated in detail. The results showed that alisol A 24-acetate and 23-acetate inter-transformed in solvents and the transformation rate was more rapid in protic solvents than in aprotic solvents. Moreover, both alisol A 24-acetate and 23-acetate were deacetylated to yield alisol A when kept in methanol for a long time. This is the first report on the structural transformation between alisol A 24-acetate, alisol A 23-acetate and alisol A. In addition, the single crystal X-ray structure of alisol A 24-acetate and the NMR data of alisol A 23-acetate were also reported for the first time.

Full text of DOI:10.1248/cpb.56.41

January 2008
Chem. Pharm. Bull. 56(1) 41—45 (2008)
41
Stability and Structure Studies on Alisol A 24-Acetate
c
Bolat MAKABEL
,
a,b Yuying ZHAO,a Bin WANG,a Yanjing BAI,a Qingying ZHANG,*,a Li W
U
,c and Yang L
V
a Department of Natural Medicines and the State Key Laboratory of Natural and Biomimetic Drugs, School of
Pharmaceutical Sciences, Peking University Health Science Center; No. 38 Xueyuan road, Beijing 100083, PR China:
b Xinjiang Institute of Chinese Materia Medica and Ethnodrug; Urumqi 830002, PR China: and c Institute of Materia
Medica, Chinese Academy of Medical Sciences; Beijing 100050, PR China.
Received July 10, 2007; accepted October 10, 2007
Alisol A 24-acetate is one of the main active triterpenoid compounds isolated from Rhizoma Alismatis, which
is a famous Traditional Chinese Medicine, and has been determined for the quality control of this crude drug. In
this study, alisol A 24-acetate was found to be unstable in solvents and its stability in different solvents was inves-
tigated in detail. The results showed that alisol A 24-acetate and 23-acetate inter-transformed in solvents and the
transformation rate was more rapid in protic solvents than in aprotic solvents. Moreover, both alisol A 24-acetate
and 23-acetate were deacetylated to yield alisol A when kept in methanol for a long time. This is the first report
on the structural transformation between alisol A 24-acetate, alisol A 23-acetate and alisol A. In addition, the
single crystal X-ray structure of alisol A 24-acetate and the NMR data of alisol A 23-acetate were also reported
for the first time.
Key words alisol A 24-acetate; alisol A 23-acetate; stability; Rhizoma Alismatis; Alisma orientalis
MeOH was obtained by refluxing of MeOH with sodium (Na) and then dis-
tilling.
Rhizoma Alismatis, the dried rhizome of Alisma orientalis
(SAM.) JUZEP, is a famous Traditional Chinese Medicine
(TCM) which has been widely used for diuretic, hypolipi-
demic, anti-inflammatory and anti-diabetic purposes in
China for more than a thousand years. Protostane-type triter-
penes are the principal active constituents of Rhizoma Alis-
matis and more than 50 unique protostane-type triterpenes,
including alisols A, B and their monoacetates, have been iso-
Plant Material Rhizoma Alismatis was collected in August 2003 from
Fujian Province, PR China, and identified by Prof. Hubiao Chen, School of
Pharmaceutical Sciences, Peking University Health Science Center. A
voucher specimen was deposited at the Herbarium of the School of Pharma-
ceutical Sciences, Peking University Health Science Center, Beijing, PR
China.
Reference Compounds Alisol A 24-acetate and alisol A were obtained
from Rhizoma Alismatis by chromatography methods, and their structures
lated from this herbal drug.1—3) As the bioactive “marker were characterized by spectral methods, including MS, 1H-, 13C- and 2D-
NMR spectra.
compounds” of Rhizoma Alismatis, protostane-type triter-
penes, including alisol A 24-acetate, which is one of the
main active triterpenoid compounds isolated from Rhizoma
HPLC Analysis HPLC analysis was performed on a YMC analytical
column (Tokyo, Japan) with 5 mm C18-reversed phase material
(250ꢀ4.60 mm i.d.) using CH3CN/H2O (70% or 52%) as the mobile phase
Alismatis, have been determined for the quality control of
Rhizoma Alismatis.4,5) However, during the course of our re-
search on the quantification of alisol A 24-acetate, it was
found that alisol A 24-acetate was unstable and could be
transformed into other compounds in methanol.
In this paper, the stability of alisol A 24-acetate in differ-
ent solvents was described in detail. Furthermore, by using
LC-MS, NMR, and single crystal X-ray diffraction tech-
niques, the structures of the compounds transformed from al-
isol A 24-acetate and the single crystal X-ray structure of al-
isol A 24-acetate were elucidated for the first time.
with the flow rate of 1.0 ml/min at room temperature. The injection volume
was 20 ml. For UV detection, the detection wavelength were set at 210 nm,
while for ELSD detection, the tube temperature was set at 90 °C, and air was
used as the gas with the flow rate of 2.5 l/min.
LC-MS Analysis LC-MS analysis was performed on a YMC analytical
column (Tokyo, Japan) with 5 mm C18-reversed phase material (250ꢀ
4.60 mm i.d.) using CH3CN/H2O (52%, v/v) as the mobile phase. The flow
rate was 1.0 ml/min and split 0.4 ml/min to mass spectrometer. The injection
volume was 20 ml. Mass spectra were acquired using a MDS SCIEX QSTAR
mass spectrometer equipped with an ESI source. All the mass spectra were
acquired in negative or positive ion mode with ion source temperature at
400 °C, and detector voltage at 2200 V. The mass spectra were recorded in
the range m/z 100—1000 amu.
Single Crystal X-Ray Diffraction Single crystals of alisol A 24-acetate
were obtained from EtOAc solution of A-M2. A single crystal having di-
mensions of 0.15ꢀ0.30ꢀ0.50 mm was mounted on a Japan MAC DIP-
2030K area detector diffractometer with graphite monochromated MoKa ra-
diation (lꢁ0.71073 Å). The crystal data are as follows: monoclinic, space
group C2 (#5), aꢁ34.329(2), bꢁ7.550(1), cꢁ27.653(2) Å, bꢁ120.206(3)°,
Vꢁ6194.1(23) Å3, Zꢁ8, DXꢁ1.143 g/cm3. A total of 4720 independent re-
flections were collected at 23 °C by using the w scanning mode (2qꢂ50.0°),
of which 4662 intensity data with |F|2ꢃ2s|F|2 were taken as observed reflec-
tions. The structure was solved by direct methods using SHELXS-97 soft-
ware. The least-squares refinement converged with agreement factors of
R1ꢁ0.068 and wR2ꢁ0.207.
Experimental
General Experimental Procedures A Shimadzu 10A HPLC system
(Tokyo, Japan), equipped with a quaternary pump, a diode array spectropho-
tometric detector (DAD) and an Altech ELSD-2000ES evaporative light
scattering detector (ELSD), was used for HPLC analysis. NMR spectra were
recorded on a Bruker DRX-500 NMR spectrometer equipped with 5 mm
probes using TMS as internal standard. LC-MS analysis was performed on
an Agilent 1000 HPLC system (Palo Alto, CA, U.S.A.) equipped with a
MDS SCIEX QSTAR mass spectrometer with an ESI source. A single crys-
tal was mounted on a Japan MAC DIP-2030K area detector diffraction
meter (Tokyo, Japan). HPLC-grade acetonitrile (CH3CN) and methanol
(MeOH) were purchased from Fisher Scientific (Fair Lawn, NJ, U.S.A.).
HPLC-grade deionized water (H2O) was purified by Milli-Q Water purifica-
tion system (Millipore, MA, U.S.A.). MeOH, absolute ethanol (EtOH), 95%
EtOH, ethyl acetate (EtOAc), acetone (Me2CO), and chloroform (CHCl3) for
analysis were of analytic grade from Beijing Fine Chemical Company (Bei-
jing, PR China). The aqueous organics except for 95% EtOH were mixed
with the corresponding organics and HPLC-grade deionized H2O. Absolute
Results and Discussion
Discovery of the Instability of Alisol A 24-Acetate Al-
isol A 24-acetate (Fig. 1) was isolated from 95% EtOH ex-
traction of Rhizoma Alismatis through repeated column chro-
matography on silica gel with petrol ether/Me2CO (7 : 3) as
∗ To whom correspondence should be addressed. e-mail: qyzhang@bjmu.edu.cn
© 2008 Pharmaceutical Society of Japan

42
Vol. 56, No. 1
the mobile phase, and its structure was established by com- ing that involved during the above HPLC purification proce-
1
paring its spectroscopic data, including MS, H- and 13C- dure and then analyzed by HPLC right away. The result
NMR data, with those of published in cited references.2) showed that the fraction containing P1 revealed one peak at
However, it was found that its purity was not good enough to the same retention time of P1, while the fraction containing
be used as a reference compound for quantification. So this P2 yielded one peak at the same retention time of P2. How-
compound was further purified by semi-preparative HPLC ever, as time went on, P1 transformed to P2 and P2 trans-
eluting with 80% aqueous MeOH. It was interesting to find formed to P1 respectively (Fig. 3). Based on the above evi-
that in the HPLC two main peaks (peak 1 and peak 2) were dences, it could be concluded that alisol A 24-acetate (P1)
revealed after purification by HPLC, among which the reten- and another compound (P2) could be inter-transformed in
tion time of peak 1 (P1) was same as that of alisol A 24-ac- 80% MeOH, and this transformation was not caused but
etate, although there was only one main peak (P1) before the could be accelerated by heating. Additionally, alisol A 24-ac-
HPLC preparation (Fig. 2). This phenomenon suggested that etate was not found to be transformed to other compounds
alisol A 24-acetate might be transformed to another com- during the course of purification by column chromatography
pound (peak 2, P2) during the procedure of HPLC purifica- on silica gel eluting with petrol ether/Me2CO. So, the solvent
tion of alisol A 24-acetate. In order to clarify the reasons used might be an important factor that caused its instability,
causing the instability of alisol A 24-acetate, the sample con- and thus the stability of alisol A 24-acetate in different sol-
sisting of two peaks, named A-M2, was subjected to semi- vents was studied in detail.
preparative HPLC again eluting with 80% MeOH and the
Stability of Alisol A 24-Acetate in Different Solvents
fractions containing P1 or P2 were collected respectively. The Alisol A 24-acetate was divided into several portions and
collected fractions containing P1 or P2 were condensed prop- dissolved in different solvents, including absolute MeOH,
erly under nitrogen stream at room temperature to avoid heat- MeOH, absolute EtOH, 95% EtOH, Me2CO, CH3CN,
EtOAc, and CHCl3, to prepare the solution of different sol-
vents with the concentration of 0.5 mg/ml, respectively.
Then, the samples were stored at room temperature and ex-
amined by HPLC-ELSD in turn at different times. The re-
sults showed that the transformation of P1 (alisol A 24-ac-
etate) to P2 occurred in somewhat different degrees in all the
above solvents, and the transformation rate in absolute
MeOH, MeOH, 95% EtOH, especially in absolute MeOH,
was more rapid than that in Me2CO, CH3CN, EtOAc, and
CHCl3 (Fig. 4). In addition, it was also found that the trans-
formation rate of alisol A 24-acetate in absolute MeOH and
absolute EtOH was much different from that in their corre-
sponding solvent containing H2O. Considering that H2O was
the most common agent, the effect of H2O in organic sol-
vents on this transformation was studied. The results indi-
Fig. 1. Structures of Alisol A, Alisol A 23-Acetate and Alisol A 24-Ace-
tate
cated that except for in MeOH, H2O in the other organic sol-
vents studied accelerated the transformation of alisol A 24-
acetate (Fig. 5). Thus it could be concluded that in protic sol-
vents the transformation of alisol A 24-acetate was rapid,
while in aprotic solvents the transformation was relatively
slow. So we believed that protonation was involved in acetyl
transformation and the transformation rate in different sol-
vents depended on the protonation intensity of the solvents.
Additionally, it was also found that when alisol A 24-ac-
etate was kept in absolute MeOH or MeOH for a long time at
room temperature, e.g. over 14 d in MeOH, three peaks (P1,
P2 and P3) revealed in the HPLC chromatogram, and the re-
Fig. 2. HPLC-UV Chromatograms of Alisol A 24-Acetate before and
after HPLC Preparation
Fig. 3. HPLC-UV Chromatograms of Peak 1 and Peak 2 at Different Times

January 2008
43
Fig. 4. Structure Transformation of Alisol A 24-Acetate in Different Solvents at Different Times (A) in 48 h; (B) in 4 h; (C) in 40 d
AM: absolute MeOH, M: MeOH, AE: absolute EtOH, 95E: 95% EtOH, A: Me2CO, AT: CH3CN, EA: EtOAc, C: CHCl3.
Fig. 5. Structure Transformation of Alisol A 24-Acetate in Solvents with
and without H2O
Fig. 6. Structure Transformation of Alisol A 24-Acetate in MeOH during
Long Time
A: Me2CO; 50A: 50% aqueous Me2CO; AT: CH3CN; 70AT: 70% aqueous CH3CN.
tention times of P1, P2 and P3 were the same as those of alisol alisol A 24-acetate and P2 was deduced to be the isomer of
1
A 24-acetate, the transformed compound referred above and alisol A 24-acetate. The H- and 13C-NMR spectra of A-M2
alisol A respectively. Moreover, when alisol A 24-acetate was showed that it did comprise signals due to two compounds,
kept in absolute MeOH or MeOH for a longer time, e.g. and for the convenience of description later they were named
120 d in MeOH, P3 appeared as the main peak while P1 and as A-M2-1 and A-M2-2 respectively. The 1H-NMR spectrum
P2 reduced to very little amount (Fig. 6). However, P3 was of A-M2 revealed 16 singlets and 2 doublets for methyl
not yet be detected when alisol A 24-acetate was kept in groups, among which the signals at d 2.17 (3H, s) and 2.05
other solvents even for 40 d.
(3H, s) were the characteristic signals for acetyl group. Com-
1
Structure Identification of the Transformed Compound paring the H- and 13C-NMR data of A-M2 with those of al-
P2 From the above results of alisol A 24-acetate’s stability isol A 24-acetate, it showed that the H- and 13C-NMR data
1
in different solvents, we realized that alisol A 24-acetate was due to A-M2-1 were identical with those of alisol A 24-ac-
1
relatively stable in aprotic solvents. So A-M2 was subjected etate and the remnant H- and 13C-NMR signals due to A-
to silica-HPLC using hexane/EtOAc (75 : 25) as the eluent so M2-2 were also very similar with those of alisol A 24-acetate
as to try to obtain the pure P1 and P2 for further characteriz- except for a few signals ascribed to the side chain. Then 2D-
ing their structures. P1 was obtained as a pure compound and NMR, including 1H–1HCOSY, HMQC, HMBC, and NOESY,
1
elucidated as alisol A 24-acetate by comparing its retention were recorded and the H- and 13C-NMR signals of A-M2-1
1
time, H- and 13C-NMR data with those of standard of alisol and A-M2-2 were assigned completely (Table 1). By compar-
1
A 24-acetate. Unfortunately, P2 was obtained as a mixture of ing the H- and 13C-NMR data of A-M2-2 with those of al-
P1 and P2 and the intensity ratio of P1 to P2 was about 1 : 3 ac- isol A 24-acetate, the H-23 signal of A-M2-2 shifted down-
cording to the ratio of their HPLC peak areas. Therefore the field to d 4.91 (br t, 6.0) while the H-24 signal of A-M2-2
structure of P2 had to be characterized through elucidating shifted upfield to d 3.23 (s). The 1H–1HCOSY cross peaks of
the structures of the mixture A-M2.
H-23 (d 4.91) of A-M2-2 with d 1.77, 1.67 (H-22) and 3.23
LC-MS (ESI-TOF) analysis of A-M2 showed that both P1 (H-24), and the HMBC correlations of H-23 (d 4.91) of A-
and P2 revealed the same quasi-molecular ion peaks at m/z M2-2 with d 170.91 (COCH3), 37.90 (C-22), 28.72 (C-20),
555 [MꢄNa]ꢄ and 571 [MꢄK]ꢄ, the same major fragment and 72.55 (C-25), indicated that A-M2-2 should be a 23-ac-
ion peaks at m/z 515 [MꢅH2OꢄH]ꢄ in the positive mode, etate which was produced through acetyl transformation of
and the same quasi-molecular ion peaks at m/z 531 [MꢅH]ꢅ alisol A 24-acetate. As acetyl transformation is a popular re-
in the negative mode, suggesting that both the molecular action and the configuration of carbons does not change dur-
weights of P1 and P2 were 532, which were the same as that ing the procedure. So we deduced that the configuration at C-
of alisol A 24-acetate. Therefore, P1 was further proved to be 24 and C-23 of A-M2-2 must keep unchanged. Thus the

44
Vol. 56, No. 1
Table 1. 1H- (500 MHz) and 13C-NMR (125 MHz) Data of A-M2-1 and A-M2-2 in CDCl3 (ppm, J in Hz)
A-M2-1
A-M2-2
A-M2-1
A-M2-2
No.
No.
1H
13C
1H
13C
1H
13C
1H
13C
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
2.22, 2.09a)
2.33, 2.68a)
30.92
34.29
220.49
46.93
48.50
20.03
33.73
40.37
49.47
36.90
69.96
34.36
138.08
56.97
30.43
29.02
2.22, 2.09a)
2.33, 2.68a)
30.92
34.17
220.32
46.93
48.44
20.09
33.68
40.71
49.78
36.90
70.13
34.53
137.36
56.97
30.60
29.37
17
18
19
20
21
22
23
24
25
26
27
28
29
30
COCH3
COCH3
135.12
23.12
25.53
27.83
20.03
39.60
68.86
78.61
73.86
27.31
26.61
29.52
19.96
24.09
20.77
170.79
135.06
22.92
25.61
28.72
20.09
37.90
71.76
78.25
72.55
25.48
26.55
29.66
20.03
23.86
21.40
170.91
1.12, s
1.05, s
2.81a)
0.97, d (4.0)
1.24, 1.35a)
3.88, m
1.14, s
1.05, s
2.59a)
2.06a)
2.06a)
0.98, d (4.0)
1.77, 1.67
4.91, br t (6.0)
3.23, s
1.44, 1.25a)
1.23a)
1.45, 1.28a)
1.22a)
4.58, s
1.74, d (11.0)
1.72, d (10.5)
1.14, s
1.28, s
1.05, s
1.06, s
0.98, s
2.18, s
1.19, s
1.20, s
1.05, s
1.06, s
0.97, s
2.05, s
3.85a)
3.83a)
2.80, 2.01a)
2.60, 1.95a)
1.33, 1.87a)
2.13a)
1.33, 1.87a)
2.18a)
a) Overlapped signals.
Fig. 7. The Single Crystal Structure of Alisol A 24-Acetate
(A) single molecule; (B) 8 molecules in one unit cell; (C) superimpose of two molecules in one asymmetric unit.
structure of A-M2-2 was established as alisol A 23-acetate that shown in Figs. 1 and 7 on the basis of the absolute con-
(Fig. 1). By comparing the area ratio of P1 to P2 in the HPLC figuration of 19-bCH3 of steroids. Rings A and B were in
1
chromatogram and that of characteristic H-NMR signals of twist boat conformation, ring C in chair conformation, ring
A-M2-1 and A-M2-2, P1 was deduced to be corresponded to D in envelop conformation, the configurations of A/B and
A-M2-1 and P2 to A-M2-2, which was consistent with the B/C were linked in trans type, and the stereochemistry of the
above HPLC results.
side chain was as that shown in Figs. 1 and 7. In addition,
To further prove the correctness of structural elucidation there existed two molecules with the same configuration but
of alisol A 24-acetate and 23-acetate, X-ray diffraction analy- different conformation of the side chain in one asymmetric
sis was carried out using single crystals obtained from unit of the crystal cell because of the single bond rotation
EtOAc solution of A-M2. From the X-ray structure here ob- (Fig. 7). Crystallographic data for the structure of alisol A
tained, only the single crystal structure of alisol A 24-acetate 24-acetate reported in this paper had been deposited with the
was obtained and its stereochemistry was determined to be as Cambridge Crystallographic Data Centre (deposition num-

January 2008
45
ber: 633142). The supplementary crystallographic data can A 24-acetate might be transformed to alisol A 23-acetate
be obtained free of charge via www.ccdc.cam.au.uk/conts/ during the procedure of HIO4 oxidation through acetyl trans-
retrieving.html or from the CCDC, 12 Union Road, Cam- formation7) and their proposal was in good agreement with
bridge CB2 1EZ, U.K.; fax: ꢄ44 1223 336033; e-mail: de- our results in this paper.
posit@ccdc.cam.ac.uk.
Structure Identification of the Transformed Compound
A-M2 was proved to be a mixture of alisol A 24-acetate P3 As shown above that when alisol A 24-acetate was kept
and 23-acetate on the basis of HPLC and NMR evidences, in absolute MeOH and MeOH for a long time, P3 was pro-
however, the single crystal X-ray diffraction experiment duced, whose retention time was the same as that of alisol A.
showed that the single crystal obtained from A-M2 consisted The LC-MS (in positive and negative ion mode) analysis of
of alisol A 24-acetate only. The inconsistence of the results A-M3, i.e. the sample containing P1, P2 and P3, proved that
made us to the further analysis of the single crystal, which the rentention time and molecular weight of P1, P2 and P3
was used for single crystal X-ray diffraction, by HPLC. The were identical with that of alisol A 24-acetate, alisol A 23-
single crystal was dissolved in EtOAc and analyzed by acetate and alisol A, respectively. Furthermore, the LC-MS
HPLC-ELSD right away. The HPLC chromatogram showed data of P3 were the same as those of the reference compound
P1 as the main peak while P2 in very little amount. On the of alisol A. Based on the above evidences, P3 was elucidated
other hand, for the mother solution both P1 and P2 were re- as alisol A ultimately.
vealed in the HPLC chromatogram (Fig. 8). So, it could be
This is the first report on the instability of alisol A 24-ac-
concluded that only the pure crystal of alisol A 24-acetate etate and structural transformation between alisol A 24-ac-
was obtained from the mixture of alisol A 24-acetate and 23- etate, alisol A 23-acetate and alisol A. In addition, the single
acetate
crystal X-ray structure of alisol A 24-acetate and the NMR
Alisol A 24-acetate was reported to be isolated from the data of alisol A 23-acetate were also reported for the first
genus Alisma in numerous reports, but the inter-transforma- time.
tion of alisol A 24-acetate and 23-acetate had not been re-
ported previously. Alisol A 23-acetate was once reported to
Acknowledgements This work was financially supported by the Na-
tional Natural Science Foundation of China (grant no. 20432030) and the
Program for Changjiang Scholar and Innovative Team in University (grant
number: 985-2-063-112).
be obtained from Alisma plantago-aquatica L. var. orientale
in 1968,6) but revised to alisol A 24-acetate by themselves
very soon.7) Murata T. and the co-authors proposed that alisol
References
1) Yin R. J., Wu J. Z., Herald Med., 22, 295—296 (2003).
2) Nakajima Y., Satoh Y., Katsumata M., Tsujiyama K., Ida Y., Shoji J.,
Phytochemistry, 36, 119—127 (1994).
3) Kato T., Tomita M., Takigawam M., Iwasaki H., Hirukawa T., Yama-
hara J., Bull. Chem. Soc. Jpn., 67, 1394—1398 (1994).
4) Yoshikawa M., Yamaguchi S., Chatani N., Nishino Y., Matsuoka T.,
Yamahara J., Murakami N., Matsuda H., Kubo M., Yakugaku Zasshi,
114, 241—247 (1994).
5) Wen H. M., Peng G. P., Chi Y. M., Li W., Chin. J. Pharm. Anal., 18,
375—377 (1998).
6) Muruta T., Shinohara M., Hirata T., Kamiya K., Nishikawa M.,
Miyamoto M., Tetrahedron Lett., 1968, 103—108 (1968).
7) Muruta T., Shinohara M., Hirata T., Miyamoto M., Tetrahedron Lett.,
Fig. 8. HPLC-ELSD Chromatograms of A-M2
1968, 849—854 (1968).