Glucose-lowering activity of novel tetrasaccharide glyceroglycolipids from the fruits of Cucurbita moschata

Article abstract of DOI:10.1016/j.bmcl.2010.12.030

Two new tetrasaccharide glyceroglycolipids were obtained from pumpkin. The structures of the two compounds were determined using chemical methods and spectroscopic analysis. Both compounds demonstrated significant glucose-lowering effects in streptozotocin- and high-fat-diet-induced diabetic mice.

Full text of DOI:10.1016/j.bmcl.2010.12.030

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1001–1003
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Glucose-lowering activity of novel tetrasaccharide glyceroglycolipids
from the fruits of Cucurbita moschata
⇑
Zhiguo Jiang, Qizhen Du
Institute of Food and Biological Engineering, Zhejiang Gongshang University, 149 Jiaogong Road, Hangzhou 310035, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Two new tetrasaccharide glyceroglycolipids were obtained from pumpkin. The structures of the two
compounds were determined using chemical methods and spectroscopic analysis. Both compounds dem-
onstrated significant glucose-lowering effects in streptozotocin- and high-fat-diet-induced diabetic mice.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 21 September 2010
Revised 18 November 2010
Accepted 7 December 2010
Available online 10 December 2010
Keywords:
Tetrasaccharide galactolipid
Cucurbita moschata
Pumpkin
Glucose-lowering
The fruit of the pumpkin, Cucurbita moschata, is considered to
be a health food in China.1 Preliminary investigations indicated
that a pumpkin-rich diet could reduce blood glucose,2 and pro-
tein-bound polysaccharides from pumpkin reduced blood glucose
levels and improved glucose tolerance.3 However, lower molecular
weight compounds from pumpkin have rarely been screened for
glucose-lowering activity, possibly because no small molecular
weight compounds of interest have been isolated. To clarify the
bioactive constituents of pumpkin, we systematically isolated its
chemical constituents and identified two new tetrasaccharide
glyceroglycolipids which belong to glycolipids possessing possible
antithrombotic,4 antiviral,5,6 anticancer,7–10 or antiinflamma-
tory11,12 activities. In the present study we investigated the
glucose-lowering activity of these new tetrasaccharide glycerogly-
colipids from pumpkin.
Two new tetrasaccharide glyceroglycolipids, QGMG-3 and
QGMG-2, were obtained from pumpkin by extraction and separa-
tion. Freeze-dried and powdered pumpkin (5 kg) was extracted
with 95% ethanol (10 l ꢀ 3) at room temperature. After solvent re-
moval, the crude extract (350 g) was suspended in H2O (3 l), then
extracted with EtOAc and n-BuOH (5 ꢀ 500 ml each) to produce
two extracts. The BuOH-soluble fraction (100 g) was subjected to
CH2Cl2/n-hexane/EtOH/H2O (6:2:2:4, v/v), by eluting the lower
organic phase at 1.5 ml/min and 800 rpm to give a mixture of
QGMG-3 and QGMG-2 (200 mg). This was further applied to pre-
parative RP-18 HPLC (CH3CN/H2O 5.5:4.5, v/v) to afford com-
pounds QGMG-3 (40 mg) and QGMG-2 (50 mg). Their structures
were determined by chemical and spectral methods (IR, 1H and
13C NMR, DEPT, TOCOSY, HSQC, HMBC, and MS spectral data are
included in Supplementary data.).
QGMG-3 was obtained as an optically active, white amorphous
powder ð½a 2D0
ꢁ
¼ ꢂ18:1Þ. The HRESIMS m/z 1023.4628 ([M+Na]+;
calcd 1023.4624) revealed a molecular formula of C45H76O24
,
inferring the existence of eight degrees of unsaturation. The IR
absorption bands at 3420 and 1722 cmꢂ1 were attributed to the
hydroxyls and ester carbonyl, respectively. The interpretation of
NMR spectra of QGMG-3 indicated that it was a glyceroglycolipid.
Analysis of the acid hydrolysis product by TLC showed that the su-
gar moiety of QGMG-3 contained only galactose. Alkaline hydroly-
sis of QGMG-3 yielded linolenic methyl ester, identified by GC–MS
analysis. The 13C NMR spectrum showed the presence of four ano-
meric carbons (dC 106.0, 105.8, 105.8, and 105.8), one carbonyl (dC
173.8), three olefinic (dC 132.3, 130.7, 128.9 (2 ꢀ C), 128.3, and
127.7), and one terminal methyl (dC 14.6). In the 1H NMR spectrum,
four anomeric proton signals (dH 4.85 (1H, d, J = 7.7 Hz); 4.75 (1H,
d, J = 7.6 Hz); 4.75 (1H, d, J = 7.5 Hz); 4.75 (1H, d, J = 7.6 Hz)) indi-
cated that the configurations of all three galactopyranosyls were
b-type. Six olefinic protons and one methyl were also observed in
the 1H NMR spectrum. In addition, the characteristic proton signals
observed at dH 2.92 (4H, m) and 2.32 (2H, t, J = 7.5 Hz) were easily
assigned as the methylene proton signals inserted between double
MCI CHP20P gel (75–150 lm, Mitsubishi Chemical Industries Ltd,
Tokyo, Japan) column chromatography by eluting with MeOH/
H2O (2:8–8:2, v/v) to give two major fractions, Fr.1 (3.2 g) and
Fr.2 (7.5 g). Fr. 2 was purified by high speed countercurrent
chromatography using a two-phase solvent system composed of
bonds and the methylene proton signals next to carbonyl13
respectively. The downfield chemical shift of C-600(dC 69.7),
,
⇑
Corresponding author. Tel./fax: +86 571 88071024 8575.
E-mail address: qizhendu@163.com (Q. Du).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.12.030

1002
Z. Jiang, Q. Du / Bioorg. Med. Chem. Lett. 21 (2011) 1001–1003
Figure 1. Chemical structures of compounds 1 and 2 with key HMBC correlations (H?C).
Table 1
Serum glucose levels in streptozotocin- and high-fat-diet-induced diabetic mice treated with glyceroglycolipids and metformin (mmol/L, mean SD).
Mice (n = 10)
Day 0
7 days
14 days
5.02 0.65⁄⁄
22. 17 5.63
17.09 3.19⁄⁄⁄
19.09 4.03
18.27 3.33⁄⁄
21 days
28 days
Control
4.90 0. 32⁄⁄
19.21 2.35
19.12 2.56
18.99 2.37
19.17 3.45
4.73 0.51⁄⁄
22.65 6.16
16.12 3.25⁄⁄
21.47 5.05
20.89 5.50⁄
5.15 0.72⁄⁄
23.89 6.98⁄⁄
16.97 3.03
4.87 0.54⁄⁄
23.77 6.12⁄⁄
15.27 2.79
Diabetic group
Metformin group
QGMG-2 group
QGMG-3 group
18.17 3.58⁄⁄
16.36 3.01⁄⁄
15.73 2.66⁄⁄
14.61 2.31⁄⁄
⁄⁄P <0.01, ⁄P <0.05 versus diabetic group.
C-6000(dC 69.6), and C-60000(dC 69.6) compared with the carbon signal
of C-60000(dC 62.5), indicated that the three sugars likely formed
a linear linkage between C-10000 and C-60000, C-10000 and C-6000, and
between C-1000 and C-600 by the formation of ether bonds14, as
confirmed by HMBC spectroscopy. The connectivity of sugars,
glycerol, and acyl parts was investigated by HMBC analysis; the
carbonyl at dC 173.8 (C-10) showed cross-peaks with the H-C (dH
4.63 (dd, 10.3, 5.9); 4.61 (dd, 10.3, 5.2)); the anomeric proton sig-
nal at d 4.75 (H-10000) correlated with the carbon signal at d 69.6 (C-
60000); the anomeric proton signal at d 4.75 (H-10000) correlated with
the carbon signal at d 69.6 (C-6000); the anomeric proton signal at d
4.75 (H-1000) correlated with the carbon signal at d 69.7 (C-600); and
the anomeric proton signal at dH 4.85 (H-C (100)) correlated with
C-3 (dC 72.8). The structure of QGMG-3 was therefore identified
To investigate the effects of QGMG-2 and QGMG-3 on blood
glucose levels, serum glucose levels in streptozotocin- and high-
fat diet-induced diabetic mice treated with the glyceroglycolipids
were compared with those after metformin treatment.15 Blood
glucose levels were monitored at 7-day intervals in diabetic mice
after intraperitoneal (ip) injection of QGMG-3 and QGMG-2 at
doses of 50 mg/kg body weight daily for 4 weeks. After the treat-
ment of 4-weeks, the values of blood glucose of QGMG-3 and
QGMG-2 were 14.61 2.31 and 5.73 2.66 mmol/L, respectively,
while that of metformin was 15.27 2.79 mmol/L. It is clear that
the two tetrasaccharide glyceroglycolipids showed stronger blood
glucose-lowering activity similar to that of metformin (250 mg/
kg body weight) (Table 1). In comparison, QGMG-3, with a three
double-bond side chain, exhibited higher activity than QGMG-2,
with a two double-bond side chain. Experimental result also
showed that body weight was unaffected by both QGMG-3 and
QGMG-2, which showed the two compounds were possibly no po-
tential cytotoxicity.
as 1-O-(9Z,12Z,15Z-octadecatrienoyl)-3-O-[b-
syl-(1?6)-O-b- -galactopyranosyl-(1?6)-O-b- -galactopyranosyl-
(1?6)-O-b- -galactopyranosyl] glycerol (Fig. 1).
D-galactopyrano-
D
D
D
QGMG-2 had the molecular formula C45H78O24, based on HR-
ESI-MS analysis (m/z 1025.4777 ([M+Na]+; calcd 1025.4780), infer-
ring the existence of seven degrees of unsaturation, one less than
that of QGMG-3. The compound was also obtained as a white amor-
phous powder. The NMR spectra of its sugar and glycerol moieties
were almost identical to those of QGMG-3; the only difference was
the presence of one less double bond in the acyl part, judging from
the existence of two less olefinic protons in QGMG-2. Alkaline
hydrolysis of QGMG-2 yielded a fatty acid methyl ester, which was
identified as linoleic methyl ester by GC-MS analysis. The structure
of QGMG-2 was thus characterized as 1-O-(9Z,12Z-octadecadie-
In conclusion, this study demonstrated that two tetrasaccha-
ride glyceroglycolipids from pumpkin possess glucose-lowering
activities. These glyceroglycolipids could represent suitable candi-
dates for the treatment of type II diabetes, though further investi-
gations into the mechanisms responsible for the glucose-lowering
effects of these glyceroglycolipids are required.
Acknowledgment
Financial support by
Sino-German Center for Research Promotion is gratefully acknowl-
edged.
a grant (GZ051-4 (148)) from the
noyl)-3-O-[b-
(1?6)-O-b- -galactopyranosyl-(1?6)-O-b-
glycerol (Fig. 1).
D
-galactopyranosyl-(1?6)-O-b-
D
-galactopyranosyl-
D
D-galactopyranosyl]

Z. Jiang, Q. Du / Bioorg. Med. Chem. Lett. 21 (2011) 1001–1003
1003
15. Animals and experimental procedures: Eight-week-old male C57/BL6J mice were
purchased from the Laboratory Animal Center of the Academy of Zhejiang
Medical Sciences (Hangzhou, China). The animals were maintained under
standard conditions (12 h light–dark cycle, 24 °C) with free access to water and
standard laboratory chow. All mice were cared for in accordance with the
standards of the Guide for the Care and Use of Laboratory Animals. Diabetes
was induced according to the procedure reported by Lian et al.16 In brief, the
animals were fed high-fat diets for 4 weeks before ip injection of 120 mg/kg
streptozotocin (STZ, Sigma, USA). An equal volume of vehicle was injected into
control mice (n = 10). Serum glucose levels were determined using a glucose
measuring kit (Biosino Bio-technology and Science, China), using the glucose
oxidase method. Mice with fasting glucose levels >11 mmol/l were considered
to be diabetic. Diabetic mice with consecutive 10-day hyperglycemia
(P11 mmol/l) were used for the experiment. The diabetic mice were
randomly divided into four groups: QGMG-3-treated, QGMG-2-treated, and
metformin-treated groups (n = 10 each), and diabetic group (n = 10). The mice
in the QGMG-3- and QGMG-2-treated groups were injected ip with QGMG-3
and QGMG-2, respectively, at doses of 50 mg/kg body weight daily for 4 weeks.
The metformin-treated group was injected i.p. with metformin at a dose of
250 mg/kg body weight daily for 4 weeks. The diabetic group was injected ip
with vehicle solvent (sterilized 0.9% NaCl). An age- and weight-matched
normal group (n = 10) was used as the control group. Body weight and food
consumption in each group were monitored daily. Blood glucose levels were
checked every 7 days. Blood samples were obtained from the tail vein, and
serum glucose levels were determined, as described above. The results are
presented as means SEM. The statistical methods used to analyze the data in
this study were unpaired Student’s t-test (two-tailed) using MS-Excel software
program. Comparisons with P values <0.05 were considered to be statistically
significant.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.12.030.
References and notes
1. Fu, C.; Shi, H.; Li, Q. Plant Food Hum. Nutr. 2006, 61, 70.
2. Li, Q. H.; Fu, C. L.; Rui, Y. K.; Hu, G. H.; Cai, T. Y. Plant Food Hum. Nutr. 2005, 60,
13.
3. Fu, C. L.; Tian, H. J.; Cai, T. Y.; Liu, Y.; Li, Q. H. Food Chem. 2007, 100, 944–947.
4. Karanionis, H. C.; Antonopoulou, S.; Demopoulos, C. A. J. Agric. Food Chem. 2002,
50, 1150.
5. Reshef, V.; Mizrachi, E.; Maretzki, T.; Silberstein, C.; Loya, S.; Hizi, A.; Carmeli, S.
J. Nat. Prod. 1997, 60, 1251.
6. Nakata, K.; Guo, C. T.; Matsufuji, M.; Yoshimoto, A.; Inagaki, M.; Higuchi, R.;
Suzuki, Y. J. Biochem. 2000, 127, 191.
7. Colombo, D.; Compostella, F.; Ronchetti, F.; Scala, A.; Toma, L.; Kuchide, M.;
Tokuda, H.; Nishino, H. Cancer Lett. 2000, 161, 201.
8. Pérez Gutierrez, R. M.; Lule, P. R. Pharm. Biol. 2005, 43, 313.
9. Franchini, L.; Toma, L.; Ronchetti, F.; Tanaka, R.; Takayasu, J.; Nishino, H.;
Tokuda, H. Eur. J. Med. Chem. 2006, 41, 1456.
10. Hou, C.-C.; Chen, Y.-P.; Wu, J.-H.; Huang, C.-C.; Wang, S.; Yang, N.-S.; Shyur,
L.-F. Cancer Res. 2007, 67, 6907.
11. Larsen, E.; Kharazmi, A.; Christensen, L. P.; Christensen, S. B. J. Nat. Prod. 2003,
66, 994.
12. Bruno, A.; Rossi, C.; Marcolongo, G.; Lena, A. D.; Venzo, A.; Berrie, C. Eur. J.
Pharmacol. 2005, 524, 159.
13. Jung, J. H.; Lee, H.; Kang, S. S. Phytochemistry 1996, 42, 447.
14. Zhan, Z. J.; Yue, J. M. Lipids 2003, 38, 1299.
16. Lian, J.; Xiang, Y.; Guo, L.; Hu, W.; Ji, W.; Gong, B. Scand. J. Lab. Anim. Sci. 2007,
34, 21.