Structure of new monoterpene glycoside from sibiraea angustata Rchd. and its anti-obestic effect

Article abstract of DOI:10.1248/cpb.57.294

A new monoterpene glycoside (1) isolated from the aerial part of Sibirae angustata Rchd. (Rosaceae) was found to be 1-O-β-d-glucopyranosyl-geraniol- 5,10-olide and named as sibiskoside. Acute toxicity study revealed that oral administration of 1 (2.5 g/kg

Full text of DOI:10.1248/cpb.57.294

294
Notes
Chem. Pharm. Bull. 57(3) 294—297 (2009)
Vol. 57, No. 3
Structure of New Monoterpene Glycoside from Sibiraea angustata RCHD.
and Its Anti-obestic Effect
Yoshiaki ITO, ,a Satoshi KAMO,a Samir Kumar SADHU,b Takashi OHTSUKI,b Masami ISHIBASHI,b and
*
c
Yoshihiro KANO
a Iskra Industry Co., Ltd.; 1–14–2 Nihonbashi, Chuo-ku, Tokyo 103–0027, Japan: b Graduate School of Pharmaceutical
Sciences, Chiba University; 1–33 Yayoi-cho, Inage-ku, Chiba 263–8522, Japan: and c Institute of Natural Medicine,
University of Toyama; 2630 Sugitani, Toyama 930–0194, Japan.
Received October 8, 2008; accepted December 19, 2008; published online January 8, 2009
A new monoterpene glycoside (1) isolated from the aerial part of Sibirae angustata RCHD. (Rosaceae) was
found to be 1-O-b-D-glucopyranosyl-geraniol-5,10-olide and named as sibiskoside. Acute toxicity study revealed
that oral administration of 1 (2.5 g/kg body weight) to mice resulted in no death and no evidence of abnormalities
in internal organs. Its oral administration to the mice reared with high-fat diet resulted in weight-loss, which was
also reflected in serum triglyceride and sugar level, and the weight of abdominal fat. Sibiskoside could be consid-
ered to be an active ingredient of Liucha for exerting weight-loss effect in a drink of S. angustata.
Key words Sibiraea angustata; sibiskoside; monoterpene; serum triglyceride level; obesity-prevention
1
Sibiraea angustata RCHD. (Rosaceae) is a shrub found at sorption maximum at 220 nm (eꢁ13000). The H-NMR
high-elevation dryland around 3000 to 4000 m height in the spectrum indicated the presence of two tertiary methyl
areas of Xizang (Tibet), Sichuan, Qinghai, Gansu and Yun- groups [dH 1.61, 1.63, (each 3H, s)] and an anomeric proton
nan in China. The Tibetans believe that a long-term con- [dH 4.97 (1H, d, Jꢁ7.9 Hz)]. The 13C-NMR and DEPT spec-
sumption of the aerial part of S. angustata as tea has some tra suggested the presence of two tertiary methyl groups,
benefit of health in humans. And they avoid mixing this plant three methylenes, seven methines, three quaternary carbons
into sheep feed, since it has been practically known to cause including a carbonyl group. In addition, heteronuclear single
weight-loss during ingestion. It was no more than one of quantum correlation (HSQC) analysis suggested the presence
folklores before our study. These traditions were not of a methine (dC 124.1) overlapped with a signal due to the
recorded on any literature, but directly collected by our hear- solvent (pyridine) (Table 1).
ing from Tibetans. This folklore leads to the hypothesis that
The chemical structure of the compound 1 was elucidated
1
the aerial part of S. angustata might have preventive effects by H–1H correlation spectroscoopy (COSY), nuclear over-
on obesity. Then, we had interest and study about the proper- hauser effect spectroscopy (NOESY), HSQC and heteronu-
ties of S. angustata.
clear multiple bond correlation (HMBC) analyses (Fig. 1).
In the previous studies, 43 essential oils1) and glucityl fer- The connectives of these partial structures and the functional
ulate (sibirate)2) were identified from the aerial part of S. an- groups were determined by analysis of HMBC. As shown in
gustata. The extract of S. angustata improved immune sys- Fig. 1, key long range correlations were observed between
tem3) and prevented CCl4-induced liver diseases.4) However, the following proton and carbon signals: H-1 and C-2, C-3;
the active ingredients for these pharmacological effects have H-2 and C-4, C-10; H-4 (a and b) and C-3, C-6; H-6 and C-
not yet been identified.
7, C-8, C-9; H-8 and C-6, C-7, C-9; H-9 and C-6, C-7, C-8;
In this study, we present the isolation and structure deter- H-1ꢂ and C-1. Furthermore, NOESY spectrum exhibited
mination of a new compound (1), together with two known
compounds, isoferulic acid and quercetin 3-O-a-L-ara-
binopyranosyl(1—6)-b-D-galactopyranoside5) from the aerial
part of S. angustata, and the anti-obestic effect of 1 is also
discussed.
Table 1. 1H- and 13C-NMR Spectrum Data for 1 in Pyridine-d5
Position
dH
dC
Aglycone
1
2
5.30 (2H, m)
6.56 (1H, m)
66.3
139.7
126.6
35.7
Results and Discussions
3
The aerial part of S. angustata was boiled in water and the
decoction was spray dried. The powdered extract was ex-
tracted with EtOH under reflux, and EtOH soluble part was
then concentrated under reduced pressure to give a brown
solid material. It was applied upon a charcoal column and
eluted with stepwise gradients of H2O–EtOH. From the 20%
EtOH fraction, compound 1 was obtained as colorless amor-
phous powder by preparative HPLC. It was established to
have a molecular formula of C16H24O8 by high resolution-
positive FAB-MS spectrum. Its IR spectrum showed absorp-
tion maxima exhibiting a lactone carbonyl (1745 cmꢀ1) and
hydroxyl group (3354 cmꢀ1). UV spectrum exhibited an ab-
4
5
2.50, 2.95 (each, 1H, m)
5.21 (1H, m)
75.1
6
5.21 (1H, m)
124.1
139.7
25.5
7
8
9
1.63 (3H, s)
1.61 (3H, s)
18.2
10
1ꢂ
2ꢂ
3ꢂ
4ꢂ
5ꢂ
6ꢂ
169.9
104.5
75.0
78.3
71.4
Glc
4.97 (1H, d, Jꢁ8.0 Hz)
4.11 (1H, t, Jꢁ8.0 Hz)
4.28 (1H, m)
4.28 (1H, m)
3.95 (1H, m)
78.4
62.4
4.36, 4.52 (each, 1H, m)
∗ To whom correspondence should be addressed. e-mail: y.ito@iskra.co.jp
© 2009 Pharmaceutical Society of Japan

March 2009
295
cross-peaks between H-2 signal and both of signals due to H-
The experiment of evaluation of 1 for the effect on the
4 (a and b). The carbonyl group at C-10 (dC 169.9) was in- body weight of mice reared with high-fat diet was conducted.
ferred to be connected with an oxymethine carbon (C-5, dC Before mice were feeded with high-fat diet, there were no
75.1) through a lactone-linkage since the oxymethine proton significant differences of body weight among the 2 experi-
(H-5) resonated at significantly low field (dH 5.21). More- mental groups. Then, the mice were reared with high-fat diet
over, on acid hydrolysis, 1 afforded D-glucose which was with or without 1 as drinking water for 42 d. The intakes of
identified by TLC and [a]D. In order to determine the ab- the feed and water were not significantly different between
solute stereochemistry, 1 was converted into di-MTPA esters control and the group treated with 1 (Fig. 2). In the prelimi-
through 4-steps reactions (Chart 1). By applying Kusumi– nary study, the body weight of control group reared with
Mosher method6) for these MTPA esters (8a, b), the absolute high-fat diet was confirmed to be increased compared with
configuration of C-5 of 1 was suggested as R as shown in normal mice fed with usual diet (data not shown). The
Fig. 1. Thus, the structure of 1 was determined as 1-O-b-D- weight of mice treated with 1 was significantly lower than the
glucopyranosyl-geraniol-5,10-olide.
control group from day 21 to 42 (Fig. 3), suggesting that 1
Acute toxicity experiments revealed that 1 was extremely
low toxic. The oral LD50 value for mice was more than
2500 mg/kg. In both sexes, there were no difference in gen-
eral condition, body weight and food intakes in all groups
(data not shown). No deaths were observed in all groups.
Fig. 2. Profiles of Food (A) and Water (B) Intake in the Mice Rared with
High-Fat Diet with or without 1
Open circle, control group; closed circle, 1-treated group. Data were 1 d food intake
Fig. 1. Chemical Structure of 1
of 5 mice.
Chart 1. Conversion of 1 to MTPA Esters (8a, b) and Dd Values of 8a and 8b

296
Vol. 57, No. 3
Table 3. Effect of Compound 1 on Weight of Liver and Intra-abdominal
Fat
Weight of fat (g) around
Weight of
Group
liver (g)
Posterior
abdominal wall
Kidney
Epididymis
Control
1.40ꢅ0.22 0.116ꢅ0.113 0.291ꢅ0.120 1.540ꢅ0.559
Compound 1 1.12ꢅ0.12 0.035ꢅ0.010 0.186ꢅ0.096 0.445ꢅ0.215**
Data are expressed as meanꢅS.E. of five mice. ∗∗ pꢆ0.01 vs. control group.
fects on health, however, the effect of 1 has never been re-
ported. In this study, the possibility was suggested by new
molecular entity “compound 1” discovered from S. angustata
as a preventive medicine of the obesity syndrome. We are
currently running an additional study to evaluate the mecha-
nism of compound 1.
Fig. 3. Profiles of Body Weight of the Mice Treated with Compound 1
Open circle, control group; closed circle, compound 1-treated group. Data were ex-
pressed as meanꢅS.E. ∗ pꢆ0.05, ∗∗ pꢆ0.01 vs. control group.
Table 2. Effect of Compound 1 on Blood Biochemical Values
Experimental
General 1H- and 13C-NMR spectra were measured on Varian unity
Inova-500 (1H, 500 MHz; 13C, 125 MHz, Varian Co., CA, U.S.A.) spectrom-
eters. Chemical shift values are given in ppm using TMS as an internal stan-
dard. MS spectra were obtained using a JOEL JMS-HX110A mass spec-
trometer (JEOL Ltd., Tokyo). Optical rotations were measured on a JASCO
P-1020 polarimeter (JASCO Ltd., Tokyo). IR spectra were measured on a
JASCO FT-IR 230 spectrometer.
Plant Material The plant material of S. angustata cultivated in the field
of Apa in Sichuan province of China in 2002 was used in this study. This
plant was identified by professor Wang Tian Zhi in West China School of
Pharmacy, Sichuan University.
Blood biochemical values (mg/dl)
Group
Total
Blood glucose
level
Triglyceride
cholesterol
Control
161ꢅ30
94ꢅ22
228ꢅ20
Compound 1
118ꢅ22*
26ꢅ16**
183ꢅ36*
Data were expressed as meanꢅS.E. of five mice. ∗ pꢆ0.05 and ∗∗ pꢆ0.01 vs. control
group.
Extraction and Isolation One hundred kilograms of the aerial part of S.
angustata was extracted twice with boiling water of 1000 l for 1 h, and the
decoction was pulverized by spray dry method, yielding 33.7 kg of the ex-
tract. The powdered extract (3.0 kg) was eluted with EtOH (3 lꢃ3) under re-
flux, and the EtOH eluate was concentrated under reduced pressure to give a
brown solid material (1.3 kg). One kilogram of this material was applied
upon a charcoal (Charcoal Activated, Wako, Osaka) column chromatogra-
phy, and eluted successively with stepwise gradients of H2O–EtOH (20%,
40%, 60%, 90%, 100% EtOH). The 20% EtOH fraction was concentrated
under reduced pressure to give a brown solid material (98 g). Seventy-two
grams of the fraction was applied upon a silica-gel (Silicagel 60 N, Merck
Japan, Tokyo) column chromatography, and eluted with EtOAc to give frac-
tions (Fr. A-1—A-4). Fr. A-3 (15 g) was purified by preparative HPLC (col-
umn, Cosmosil 5C18-AR-II, 20fꢃ250 mm, Nacalai, Kyoto; mobile phase,
25% MeOH, 8 ml/min) to give compound 1 (1.2 g) at the retention time,
54 min. Furthermore, compound 2 (2.5 g) was obtained from 90% EtOH
has suppressive effect on the increase of body weight. At day
42, mice were sacrificed and the blood was corrected and the
weights of fat around the kidney, the posterior abdominal
wall and the epididymis were measured. As shown in Table
2, total cholesterol (TC), triglyceride (TG) and blood glucose
(BG) levels were significantly lower than those of control
group. Weights of liver, fat around the kidney and the poste-
rior abdominal wall were slightly lower than the control with-
out statistically significant difference. However, the weight of
fat around the epididymis was significantly lower than that in
control (Table 3). That is, it was suggested that 1 would have
the effect of lowering body fat, particularly, viscous fat, or
would have suppressive effect on the accumulation of fat. It fraction of charcoal column chromatography by repeated recrystallization
and identified as isoferulic acid by comparison of spectral data with those of
authentic sample (Tokyo Chemical Industry Co., Tokyo).
On the other hand, the EtOH eluate (1.4 kg) was applied upon a Diaion
HP-20 resin (Mitsubishi Chemical Co., Tokyo) column chromatography, and
is expected that 1 had an action of ameliorating obesity. Al-
though we are unable to clearly describe the mechanism be-
hind the ameliorating obesity effect of 1, the administration
of 1 did significantly affect the plasma levels of TC, TG, and
BG. It is speculated that compound 1 might suppress the in-
crease in body weight in the mice reared with high-fat diet by
the inhibition of TG absorption.
eluted successively with stepwise gradients of H2O–MeOH (20%, 50%,
80%, 100% MeOH) and acetone. The 50% MeOH fraction was concentrated
under reduced pressure to give a brown solid material (220 g). Fifty grams of
the material was applied upon a silica-gel column and eluted successively
with CHCl3 : MeOH : H2Oꢁ9 : 1 : 0.1, 8 : 2 : 0.1, 7 : 3 : 0.2, 7 : 3 : 0.5, 6 : 4 : 1,
5 : 5 : 1, and MeOH to give fractions (Fr. B-1—B-9). Compound 3 was ob-
tained from Fr. B-7 by recrystallization, and identified as quercetin 3-O-a-L-
arabinopyranosyl(1→6)-b-D-galactopyranoside (190 mg) by comparison of
spectral data.
Lipids are known to be an important energy source, but
excess intake may induce obesity and hyperlipidemia. Obe-
sity is an independent risk factor for cardiac disease7) which
has been associated conditions such as hypertension and dia-
betes. For example, as with other heart failure risk factors,
obesity frequently leads to left ventricular hypertrophy,8—11)
and because of its epidemic nature in developed societies,
obesity is a very important public health problem.12,13) Keep-
ing the body weight level within the normal range is impor-
tant for the prevention of variety diseases such as arterioscle-
rosis, cerebral apoplexy and myocardial infarction etc. So far,
1-O-b-D-Glucopyranosyl-geraniol-5,10-olide (1) Colorless amorphous
1
powder. [a]D22 ꢀ20° (cꢁ1.0, MeOH); H- and 13C-NMR spectra: see Table
1; HR-positive FAB-MS, m/z: 367.1369 [MꢄNa]ꢄ (Calcd for C16H25O8Na;
367.1368); UV (MeOH): lmax 220 (eꢁ13000); IR (NaCl, cmꢀ1): 3354
(OH), 1745 (CꢁO).
Acid Hydrolysis of 1 Compound 1 (50 mg) was heated in 1 N HCl
(3 ml) at 80 °C for 5 h and neutralized by passing through an ion-exchange
resin (Amberlite IRA410J, OHꢀ form, Organo Co., Tokyo) column. The re-
action mixtures were applied upon Wakogel 100C18 (Wako Pure Chemical
S. angustata has been known as the plant with beneficial ef- Industries, Osaka) column and eluted with H2O to give sugar fraction.

March 2009
297
The monosaccharide were identified as D-glucose by [a]D21 ꢄ66.9° (cꢁ subjected to a 12 h light–dark cycle and 50ꢅ20% relative humidity with
0.48, H2O) and TLC using Silica gel 60 (Merck Japan, Tokyo) with food and water available ad libitum. The mice were kept for 1 week before
EtOAc : AcOH : MeOH : H2Oꢁ12 : 3 : 3 : 2 to give Rf value 0.6, which is fit- the experiment. Thereafter, the obesity mice model was induced by feeding a
ted to standard D-glucose.
high fat diets (lard 40%, corn starch 10%, granulated sugar 9%, mineral mix
Preparation of MTPA Esters from 1 Compound 1 (140 mg) was sub- 4%, vitamin mix 1%, casein 29%, cellulose 5%, lactose 2%, Nosan Co.,
jected to enzyme hydrolysis by treatment with naringinase (700 mg) in ace- Yo k ohama, Day 0). For the treated group (nꢁ5) with compound 1, the com-
tone buffer (50 mM, pH 5.5; acetic acid 45.3 mg, anhydrous sodium acetate
348 mg and H2O up to 100 ml) at 37 °C for 16 h. The aqueous solution was
pound was suspended in water with the concentration prepared at the dose of
250 mg/kg/d (2.5 mg/ml, mice weight 20 g; 2.9 mg/ml, 30 g; 3.3 mg/ml,
extracted with EtOAc (50 mlꢃ3) followed by HPLC purification (YMC- 40 g), and administrated as drinking water freely from day 0 to 42. In control
Pack ODS-AM, 10ꢃ250 mm; flow rate, 2 ml/min; RI detection) to give group, water was administered instead of compound 1. The body weight of
aglycone (29.5 mg). The aglycone (20 mg) was treated with trimethyl- the mice was measured at a frequency of once a week during the feeding pe-
dimethylsilyl chloride (20 mg) in the presence of triethylamine (24 ml) and riod with the high fat diet. On day 42, mice were anesthetized with intraperi-
4-dimethylaminopyridine (2 mg) in CH2Cl2 (0.7 ml) at room temperature for toneal injection of pentobarbital (50 mg/kg), and whole blood was taken
18 h under argon atmosphere. The reaction mixture was evaporated and puri- from the inferior vena cava with a sterile syringe anticoagulated with he-
fied with a silica gel column chromatography with hexane/acetone (3 : 1) to parin lithium. Immediately after the collection of blood samples, the mice
give a TBDMS ether (6.2 mg), which was then reduced with LiAlH4 (4 mg) were exsanguinated and necropsied, and the liver, fat around kidney, poste-
in THF (0.4 ml) at room temperature for 10 min. After addition of 1 N HCl,
rior abdominal wall and epididymis were weighted. The plasma was ob-
extraction with EtOAc and evaporation of the organic layer afforded diol tained by centrifugation (1870ꢃg for 10 min) and stored at ꢀ80 °C until
(2.1 mg). A part of the diol (0.5 mg) was treated with (R)-MTPACl (5 ml) in use. The plasma clinical chemistry was analyzed using a Hitachi 7070 Auto-
pyridine (100 ml) at room temperature for 24 h. After addition of H2O, ex-
matic Serum Analyzer (Hitachi Ltd., Tokyo). The concentrations of TC, TG
traction with EtOAc followed by purification with silica gel column chro- and BG level were measured. The reagents used for the assays of TC, TG,
matography with CHCl3/acetone (9 : 1) afforded the (S)-MTPA ester (8a) and BG were purchased from Wako.
1
(0.4 mg). H-NMR (CDCl3) d 4.08 and 4.20 (each 1H, m, H2-1), 0.86 and
Statistical Analysis Data are shown as the meanꢅstandard error (S.E.),
1.28 (each 1H, m, H2-2), 1.51 (1H, m, H-3), 1.09 and 1.54 (each 1H, m, H2- and pꢆ0.05 statistically analysed by Student’s t-test was considered as sig-
4), 3.99 (1H, m, H-5), 5.10 (1H, m, H-6), 1.64 (3H, m, H3-9), 3.51 and 4.03 nificant.
(each 1H, m, H2-10), 1.23 (9H, s, t-Bu and 0.048 (6H, s, CH3ꢃ2). Treatment
of the diol (0.5 mg) with (S)-MTPACl by the same procedures afforded the
(R)-MTPA ester (8b) (0.4 mg). 1H-NMR (CDCl3) d 4.08 and 4.20 (each 1H,
m, H2-1), 0.86 and 1.27 (each 1H, m, H2-2), 1.51 (1H, m, H-3), 1.08 and
1.54 (each 1H, m, H2-4), 3.97 (1H, m, H-5), 5.10 (1H, m, H-6), 1.69 (3H, s,
H3-8), 1.64 (3H, s, H3-9), 3.51 and 4.01 (each 1H, H2-10), 1.23 (9H, s, t-Bu)
and 0.046 (6H, s, CH3ꢃ2).
Acknowledgements We thank Professor Wang Tian Zhi of West China
School of Pharmacy, Sichuan University for identification of the plant.
References
1) Tao T., Bai C., Wang T., Du L., Huaxi Yaoxue Zazhi, 19, 178—179
(2004).
2) Zhang C. Z., Li C., Feng S. L., Zhao C. L., Wang W. H., Acta Pharm.
Sin., 28, 798—800 (1993).
Compound 2: White amorphous solid, [a]D24 ꢀ0.79° (cꢁ2.0, MeOH);
1
HR-EI-MS, m/z: 182.0950 (Mꢄ) (Calcd for C10H14O3 ; 182.0943); H-NMR
data in CDCl3: d 1.72 (3H, s, H-9), 1.75 (3H, s, H-8), 2.63 and 3.06 (each
1H, m, H-4), 4.56 (2H, m, H-1), 5.20 (1H, m, H-5), 5.21 (1H, m, H-6), 6.37
(1H, m, H-2); 13C-NMR data in CDCl3: d 18.3 (C-9), 25.7 (C-8), 36.0 (C-4),
58.9 (C-1), 75.5 (C-5), 122.8 (C-6), 127.4 (C-3), 140.4 (C-7), 141.5 (C-2),
170.6 (C-10).
Acute Toxicity Test for Compound 1 Ten mice per group of each sex
weighing 18—22 g were used. Compound 1 suspended in water was orally
administered at the dose of 500—2500 mg/kg. Toxic sighs and mortality
were monitored up to day 14 when the test was terminated. LD50 value was
calculated by the Litchfield–Wilcoxon method. All procedures involving an-
imal studies were conducted in accordance with the Guidelines for Animal
Experiments by New Drug Development Research Center, Inc., and SunTen
Phytotech Co., Ltd.
3) Liu X., Yang P., Pan X., Mianyixue Zazhi, 12, 237—240 (1996).
4) Wang X., Yang B., Chin. J. Chin. Mater. Med., 20, 746—748 (1995).
5) Takemura M., Nishida R., Mori N., Kuwahara Y., Phytochemistry, 61,
135—140 (2002).
6) Ohtani I., Kusumi T., Kashman Y., Kakisawa H., J. Am. Chem. Soc.,
113, 4092—4096 (1991).
7) Sowers J. R., Am. J. Med., 115, 37S—41S (2003).
8) Paulson D. J., Tahiliani A. G., Life Sci., 51, 1557—1569 (1992).
9) Aipert M. A., Hashimi M. W., Am. J. Med. Sci., 306, 117—123 (1993).
10) Kasper E. K., Hruban R. H., Baughman K. L., Am. J. Cardiol., 70,
921—924 (1992).
11) Kenchaiah S., Evans J. C., Levy D., Wilson P. W., Benjamin E. J., Lar-
son M. G., Kannel W. B., Vasan R. S., N. Engl. J. Med., 347, 305—313
(2002).
Effect of Compound 1 on the Body Weight of Mice Reared with High-
Fat Diet Six week-old ddY male SPF mice weighing 25—30 g were pur-
chased from Japan SLC (Hamamatsu). The animals were housed at 22ꢅ3 °C
12) Bray G. A., Ryan D. H., Endocrine, 13, 167—186 (2000).
13) Pi-Sunyer F. X., Ann. Intern. Med., 119, 665—670 (1993).