Glucosylation of steroidal saponins by cyclodextrin glucanotransferase

Article abstract of DOI:10.1055/s-0030-1249938

It is known that the sugar chains of steroidal saponins play an important role in the biological and pharmacological activities. In order to synthesize steroidal saponins with novel sugar chains in one step for further studies on pharmacological activity, we here describe the glucosylation of steroidal saponins, and 5 compounds, timosaponin AIII (1), saponin Ta (2), saponin Tb (3), trillin (4) and cantalasaponin I (5), were converted into their glucosylated products by Toruzyme 3.0L, a cyclodextrin glucanotransferase (CGTase). 12 glucosylated products were isolated and their structures elucidated on the basis of spectral data; they were all characterized as new compounds. The results showed that Toruzyme 3.0L had the specific ability to add the -D-glucopyranosyl group to the glucosyl group linked at the sugar chains of steroidal saponins, and the glucosyl group was the only acceptor. This is the first report of steroidal saponins with different degrees of glucosylation. The substrates and their glucosylated derivatives were evaluated for their cytotoxicity against HL-60 human promyelocytic leukemia cell by MTT assay. The substrates all exhibited high cytotoxicity (IC₅₀ <10μ mol/L), excluding compound 5 (IC₅₀ >150μ mol/L), and the cytotoxicity of most of the products showed no obvious changes compared with those of their substrates. Georg Thieme Verlag KG Stuttgart - New York.

Full text of DOI:10.1055/s-0030-1249938

1724 Original Papers
Glucosylation of Steroidal Saponins by
Cyclodextrin Glucanotransferase
Authors
Yong-ze Wang1,2*, Bing Feng1*, Hong-zhi Huang1,2, Li-ping Kang1, Yue Cong1, Wen-bin Zhou1, Peng Zou1,2
,
Yu-wen Cong1, Xin-bo Song2*, Bai-ping Ma1
1
Affiliations
Beijing Institute of Radiation Medicine, Beijing, Peopleʼs Republic of China
Tianjin University of Traditional Chinese Medicine, Tianjin, Peopleʼs Republic of China
2
Key words
Abstract
ability to add the α-D-glucopyranosyl group to the
glucosyl group linked at the sugar chains of steroi-
dal saponins, and the glucosyl group was the only
acceptor. This is the first report of steroidal sapo-
nins with different degrees of glucosylation. The
substrates and their glucosylated derivatives
were evaluated for their cytotoxicity against HL-
60 human promyelocytic leukemia cell by MTT
assay. The substrates all exhibited high cytotoxic-
ity (IC50 < 10 µmol/L), excluding compound 5 (IC50
> 150 µmol/L), and the cytotoxicity of most of the
products showed no obvious changes compared
with those of their substrates.
"
l biotransformation
!
"
l glucosylation
It is known that the sugar chains of steroidal sa-
ponins play an important role in the biological
and pharmacological activities. In order to syn-
thesize steroidal saponins with novel sugar chains
in one step for further studies on pharmacological
activity, we here describe the glucosylation of
steroidal saponins, and 5 compounds, timosapo-
nin AIII (1), saponin Ta (2), saponin Tb (3), trillin
(4) and cantalasaponin I (5), were converted into
their glucosylated products by Toruzyme 3.0 L, a
cyclodextrin glucanotransferase (CGTase). 12 glu-
cosylated products were isolated and their struc-
tures elucidated on the basis of spectral data; they
were all characterized as new compounds. The re-
sults showed that Toruzyme 3.0 L had the specific
"
l cyclodextrin glucano-
transferase
l steroidal saponins
"
"
l cytotoxicity
Supporting information available online at
http://www.thieme-connect.de/ejournals/toc/
plantamedica
Introduction
Cyclodextrin glucanotransferase (CGTase) is an
enzyme that is capable of catalyzing intramolecu-
lar (cyclization) and intermolecular (coupling and
disproportionation) transglycosylations as well as
the hydrolysis of starch and cyclodextrin. CGTase
has been used widely in the industrialized pro-
duction of cyclodextrin (CD) to produce α-, β- or
γ-CD [18], and also in the production of glucosy-
lated modifications of natural compounds, such
as hesperidin, hydroquinone, rutin, and glycosyl-
glycerol [19–21]. A requirement for the accep-
tance by CGTase depends on the structure of a D-
glucopyranose with equatorial hydroxy groups at
C2, or C3 and C4 [22]. However, the effect of sub-
stitution by other glycosyl groups of the hydroxy
groups in the essential glucosyl group (acceptor)
at C2-OH, C3-OH on the glucosylation has rarely
been studied.
!
Steroidal saponins are one of the main classes of
active ingredients in traditional Chinese medicine
with various pharmacological and biological ac-
tivities and are distributed in a wide range of
plant species [1–7]. It is reported that the sugar
chain of steroidal saponins plays an important
role in the pharmacological and biological activ-
ities [8,9].
Bioconversions are chemical reactions catalyzed
by cells, organs or enzymes, and may be used for
specific conversions of complex substrates includ-
ing hydrolysis, hydroxylation, and glycosylation
[10–13]. Among these pathways, glycosylation is
a key pathway in the processes of plant metabo-
lism [14–16]. Glycosylation can alter the hydro-
philicity, contribute to the chemical stabilization,
and improve the bioavailability of natural prod-
ucts [15,17].
received
revised
accepted
Nov. 11, 2009
April 6, 2010
April 16, 2010
Bibliography
DOI http://dx.doi.org/
10.1055/s-0030-1249938
Published online May 19, 2010
Planta Med 2010; 76:
1724–1731 © Georg Thieme
Verlag KG Stuttgart · New York ·
ISSN 0032‑0943
Correspondence
Prof. Dr. Bai-ping Ma
Beijing Institute of Radiation
Medicine
27# Tai-ping Road
Hai-dian District
In previous work, we found that the spirostanol
skeleton and its C3-O-glycoside moiety were es-
sential for platelet aggregation and cytotoxicity
against HL-60 [23,24]. Increasing the number of
monosaccharides in the C3-O-glycoside moiety
Beijing 100850
Peopleʼs Republic of China
Phone: + 861068210077
ext. 930265
Fax: + 861068214653
mabaiping@sina.com
* These authors contributed equally to this study.
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

Original Papers 1725
could enhance the induction of platelet aggregation [23], and at-
taching an α-L-rhamnopyranosyl group to the glucopyranosyl
moiety at the C2-OH of the C3-O-glycoside moiety led to a re-
markable increase in the activity [24]. Otherwise, spirostanol
saponins are generally water-insoluble and are thus hardly ab-
sorbed by the human body. In order to synthesize steroidal sapo-
nins with novel sugar chains for further pharmacological re-
search and to improve their poor solubility, 9 steroidal saponins
with various aglycones and sugar chains were employed as sub-
strates to be glucosylated by Toruzyme 3.0 L (a kind of CGTase).
The effects of reaction media (buffer or methanol-water), the
amount of enzyme, incubation period, sugar donors, and metal
ions on enzymatic activity were investigated. The biotransforma-
tion pathways of compounds 3, 4, and 5 were studied. The sub-
strates and their glucosylated products were evaluated for their
cytotoxicity against HL-60 cells.
(1 mg) and dextrin (5 mg) were dissolved in 1 mL of 70% metha-
nol, 30 µL of Toruzyme solution were added to the substrate solu-
tion and allowed to react at 50 1°C for 24 h. The reaction mix-
tures were boiled to stop the reaction.
Effects of various factors on enzymatic activity: The effects of var-
ious factors on the enzymatic activity on 3 as a substrate were in-
vestigated. The reaction media (buffer, 30% methanol, 50% meth-
anol, 70% methanol and pure methanol), the amount of enzyme
ranging from 3 µL to 30 µL, the time course (6 h, 24 h, 48 h, 96 h,
and 144 h), the metal ions solution [CaCl2, MgCl2, CuSO4, CdCl2,
BaCl2, FeSO4, CoCl2, ZnSO4, Pb(CH3COO)2, MnSO4, NaCl2, KCl2,
and (NH4)Cl, 0.02 M], and the sugar donors (starch, corn starch,
cyclodextrin, dextrin, lactose, sucrose, maltose, and glucose; Bei-
jing Chemical Reagents Company) were used as effect factors on
enzymatic activity.
Extraction, purification and identification: The reaction mixtures
were extracted three times with n-butanol, and the n-butanol
layer was then washed twice with distilled water to remove cy-
clodextrin and any surplus dextrin. The n-butanol layer was con-
centrated in a vacuum until dry. The residues were separated by
PTLC and recovered with methanol, using an RP‑C18 column and
eluted with CH3OH‑H2O (in a gradient manner from 40% to 90%,
at a flow rate of 1.5 mL/min), respectively. All the products were
identified on the basis of their spectroscopic data (MS, 1D and
2D NMR).
Materials and Methods
!
Apparatus and reagents
1H‑NMR, 13C‑NMR and 2D‑NMR (1H-1H COSY, HSQC, HMBC and
HSQC-TOCSY) spectra were run in pyridine-d5 solutions with a
Varian UNITY INOVA 600 spectrometer. Chemical shifts were re-
corded in ppm (δ). FAB mass spectra were carried out on a Micro-
mass Zabspec, and HR‑ESI mass spectra were obtained with a
Bruker 9.4 TQ‑FT‑MS Apex Qe instrument. HPLC analyses were
performed on an Agilent 1100 unit with an Alltech 2000 Evapo-
rative Light Scattering Detector. The procedure of biotransforma-
tion was carried out in an HZS‑H shake incubator (Donglian Elec-
tric Technique Co.). Thin layer chromatography (TLC) analyses
were carried out on precoated silica gel GF254 plates (0.25 mm
thick; Qingdao Marine Chemical Group Co.). Preparative thin
layer chromatography (PTLC, 0.5 mm thick, 25 × 25 cm) was per-
formeed with silica gel H (Qingdao Marine Chemical Group Co.)
by our group. The developing solvents A [CHCl3-CH3OH‑H2O
(70:30:6, v/v/v), lower layer], B [CHCl3-CH3OH‑H2O (65:25:4,
v/v/v)], and C [CHCl3-CH3OH‑H2O (70:25:5, v/v/v, lower phase)]
were prepared by our group. Isolation and purification of prod-
ucts were performed on an RP‑C18 column (ODS‑A 12 mm S-50;
5409; YMC Co.). All chemicals used were of analytical grade. The
following materials and reagents were used for cell culture and
the cytotoxicity assay: microplate reader (EL311S; BIO‑TEK In-
struments, Inc.); 96-well plates (VWR); 3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide (MTT; Amresco); cis-
diamminedichloroplatinum (DDP; Qilu Pharmaceutical Co., Ltd);
and HL-60 cells (Department of Biology, Beijing Institute of Radi-
ation Medicine).
Analysis
TLC analysis: The extracts were spotted on silica gel plates which
were developed by developing solvents A, B, and C, and visualized
by spraying with 10% H2SO4 solution, followed by heating at
120°C for 5 min.
HPLC analysis: The samples were analyzed on an Agilent 1100 se-
ries HPLC equipped with an Alltech 2000 Evaporative Light Scat-
tering Detector (temperature: 100°C, gas flow: 2.4 L/min) using a
Hanbon-RP‑C18 column (5 µm, 250 × 4.6 mm). The mobile phase
consisted of 70–90% methanol. The liquid flow rate was set at
1.0 mL/min with an injection volume of 50–100 µL.
Identification of products: Incubation of compounds 1–5 with Tor-
uzyme yielded 12 glucosylated products. Their structures were
identified on the basis of their spectral data (1H- and 13C‑NMR,
COSY, HSQC, HMBC, and HSQC-TOCSY). Analysis of the 1H-1H shift
correlation spectroscopy (1H-1H COSY), heteronuclear single
quantum correlation (HSQC) and heteronuclear multiple bond
correlation (HMBC) spectra confirmed the sequential assign-
ments of all the resonances for each monosaccharide and their
linkage sites.
Cell culture and cytotoxicity assay
HL-60 cells in the log phase of their growth cycle (105 cells/mL)
were added to each well of the 96-well plates (90 µL/well), and
then treated, in four replicates with various concentrations of
the samples (substrates and their derivatives), and incubated in
a humidified atmosphere of 5% CO2 for 48 h at 37°C. After stop-
ping the cell culture and adding 10 µL of MTT solution (5 mg/mL)
to each well, and the plates were incubated for a further 4 h in a
humidified atmosphere of 5% CO2 at 37°C. A two-system solution
(10% sodium dodecyl sulfate-0.012 M hydrochloric acid) was
then added to each well (100 µL/well). After 12 h at room temper-
ature, the optical density (OD) of each well was measured with a
microplate reader at a wavelength of 570 nm. A dose-response
curve was plotted for each sample, and the concentrations that
caused 50% inhibition (IC50) were calculated. In these experi-
Enzyme and substrates
The enzyme, Toruryme 3.0 L, was obtained from Novozymes, Chi-
na. Compounds 1, 2, 3, and 5 were isolated from Anemarrhena as-
phodeloides Bunge, Paris polyphylla Smith var. yunnanensis
(Franch) Hand Mazz, and Agave sisalana Perrine by the authors
[25,26]. Compound 4 was synthesized by Prof. Yu-guo Du in the
Research Center for Eco-Environmental Sciences, Chinese Acade-
my of Sciences. Their purities are above 98% by HPLC analysis.
Biotransformation
General methods: Compounds 1–5 were used as substrates, a
methanol-water cosolvent acted as the reaction medium to im-
prove the solubility, and dextrin was used as a donor. Substrates
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

1726 Original Papers
Table 1 The structures of substrates and products.
No.
Substrates
Products and yield
Molecular formula, molecular weight, and references
Compound
C45H74O18; [M + Na]+ m/z: = 925.4764; sarsasapogenin-3-O-α-D-
glucopyranosyl-(1 → 4)-β-D-glucopyranosyl-(1 → 2)-β-D-galactopy-
ranoside [26–28]
1
Timosaponin AIII
1-1 (17%)
2-1 (3%)
2-2 (1%)
Compound
C45H72O17; [M + Na]+ m/z = 907.4652; diosgenin-3-O-α-D-glucopy-
ranosyl-(1 → 3)-[α-L-rhamnopyranosyl-(1 → 2)]-β-D-glucopyranoside
[29]
2
SaponinTa
C
51H82O22; [M + Na]+ m/z = 1069.4345; diosgenin-3-O-α-D-gluco-
pyranosyl-(1 → 4)-α-D-glucopyranosyl-(1 → 3)-[α-L-rhamnopyrano-
syl-(1 → 2)]-β-D-glucopyrannoside
Compound
C45H72O18; [M + Na]+ m/z = 923.4600; pennogenin-3-O-α-D-gluco-
pyranosyl-(1 → 3)-[α-L-rhamnopyranosyl-(1 → 2)]-β-D-glucopyrano-
side [30]
3
3-1 (3%)
SaponinTb
C
51H82O22; [M + Na]+ m/z = 1069.4345; pennogenin-3-O-α-D-glu-
copyranosyl-(1 → 4)-α-D-glucopyranosyl-(1 → 3)-[α-L-rhamnopy-
ranosyl-(1 → 2)]-β-D-glucopyrannoside
3-2 (2%)
Compound
C39H62O13; [M + Na]+ m/z = 761.4090; diosgenin-3-O-α-D-glucopy-
ranosyl-(1 → 4)-β-D-glucopyranoside [31]
4
4-1 (26%)
Trillin
C45H72O18; [M + Na]+ m/z = 923.4595; diosgenin-3-O-α-D-glucopyr-
anosyl-(1 → 4)-α-D-glucopyranosyl-(1 → 4)-β-D-glucopyranoside
4-2 (7%)
Compound
C45H74O20; [M+Na]+ m/z = 957.4666; (25R)-3β,6α,23α-trihydroxy-
spirost-5α 3-O-{α-D-glucopyranosyl-(1 → 4)-β-D-glucopyranoside}-6-
O-β-D-glucopyranoside [32]
5
5-1 (6%)
Cantalasaponin I
C45H74O20; [M + Na]+ m/z = 957.4666; (25R)-3β,6α,23α-trihydroxy-
spirost-5α-3-O-β-D-glucopyranoside-6-O-[α-D-glucopyranosyl-
(1 → 4) -β-D-glucopyranoside]
(continued)
5-2 (11%)
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

Original Papers 1727
Table 1 Continued
No.
Substrates
Products and yield
Molecular formula, molecular weight, and references
51H84O25; [M + Na]+ m/z = 1119.5203; (25R)-3β,6α,23α-trihy-
C
droxyspirost-5α-3-O-[α-D-glucopyranosyl-(1 → 4)-α-D-glucopyrano-
syl-(1 → 4)-β-D-glucopyranoside]-6-O-β-D-glucopyranoside
5-3 (1%)
C51H84O25; [M + Na]+ m/z = 1119.5203; (25R)-3β,6α,23α-tri-
hydroxyspirost-5α-3-O-[α-D-glucopyranosyl-(1 → 4)-β-D-glucopyra-
noside]-6-O-[α-D-glucopyranosyl-(1 → 4)-β-D-glucopyranoside]
5-4 (0.8%)
C51H84O25; [M + Na]+ m/z = 1119.5203; (25R)-3β,6α,23α-trihy-
droxyspirost-5α-3-O-β-D-glucopyranoside- 6-O-[α-D-glucopyranosyl-
(1 → 4)-α-d-glucopyranosyl-(1 → 4)-β- D-glucopyranoside]
5-5 (1%)
Notes: The structures of products were elucidated by HR‑ESI‑MS, FAB‑MS, 1H‑NMR, 13C‑NMR, COSY, and HSQC
ments, the negative reference agent was 0.1% dimethyl sulfoxide
(DMSO), and DDP was used as the positive reference control. Data
are the mean values of three experiments performed in triplicate.
strates have different aglycones, sarsasapogenin (compound 1),
diosgenin (compounds 2 and 4), pennogenin (compound 3), and
hongguanggenin (compound 5), and could all be glucosylated by
Toruzyme. The results confirmed that the enzyme had no selec-
tivity to the aglycones mentioned above.
Supporting information
The TLC and HPLC analyses, as well as the spectral data of the glu-
cosylated products are available in the Supporting Information.
To investigate the substrate specificity of Toruzyme towards ste-
roidal saponins in glucosylation, other saponins, dioscin (6), sa-
ponin Pa (7), and progenin (9) with the C4-glycosyl substituents
of the glucopyranosyl groups, were used as substrates under the
same experimental conditions. The results showed that they
Results and Discussion
!
"
could not be glucosylated. The structures are shown in l Fig. 1.
In the screening test, we found that compounds 1–5 could be
transformed into more polar products by Toruzyme 3.0 L (the
profiles are showed in Supporting Information, Fig. 1S and
Fig. 2S). All the products were elucidated as new compounds,
and had the aglycone intact with different degrees of glucosyla-
tion as established by HRMS and MS and based on the data of
1H- and 13C‑NMR, COSY, HSQC, HMBC, and HSQC-TOCSY experi-
ments. A list of substrates and products, yield, molecular formula,
molecular weight, methods of identification and relative refer-
Comparing their structures with those of compounds 2 and 3
with the C2-glycosyl substituents of the glucopyranosyl groups,
it is quite clear that the C4-OH of the glucopyranosyl group was
the major location for glucosylation. It is of much interest in this
work that gracillin (8) with an internal glucopyranosyl group and
"
ences is shown in l Table 1.
Compounds 1–5 are steroidal saponins all bearing a glucosyl
group as the sugar chain. Toruzyme showed the specific ability
to add an α-D-glucopyranosyl group to the glucosyl groups of
compounds 1–5 and the glucosyl group was the only acceptor
for forming glucosylation products. Compounds 1, 4, and 5 which
bear a terminal glucopyranosyl group or a monoglucopyranosyl
group were glucosylated at the C4-OH of the glucosyl group to
give products with an α-D-1,4-glucopyranosyl group. Compounds
2 and 3 have the same sugar chains with an internal glucopyrano-
syl group and a terminal rhamnosyl group; the first glucose was
linked preferentially to the C3-OH of the glucopyranosyl group
instead of the C4-OH to give the products with an α-D-1,3-gluco-
pyranosyl group, and the second glucose was attached to the C4-
OH of an α-D-1,3-glucopyranosyl group. These characters sug-
gested that the enzyme had a strict regioselectivity. The 5 sub-
Fig. 1 The structures of compounds which were incapable of glucosyla-
tion by Toruzyme 3.0 L
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

1728 Original Papers
Fig. 2 a The biotransformation pathway of com-
pound 3. b The biotransformation pathway of
compound 4. c The biotransformation pathway of
compound 5.
a terminal glucopyranosyl group could not be glucosylated by
5-1, and 5-2) were used as substrates for the reaction. The results
showed that 4-1 could be catalyzed into 4-2, 5-1 could be con-
verted into 5-3 and 5-4, and 5-2 could be transformed into 5-4
and 5-5. Product 3-1, however, could not be converted to 3-2. The
results suggested that 4 and 5 with monoglucosyl groups were
transformed to their two-glucose products in one step as well as
in two step reactions; while compound 3 with an internal gluco-
syl group was converted into the two-glucose product in one-
step. The proposed biotransformation pathways of compounds
Toruzyme. As a matter of fact, we found that many microorgan-
isms and enzymes were all incapable of catalyzing the glucosyla-
tion of this compound. For example, Curvularia lunata, which is
able to selectively hydrolyze the α-1,2-rhamnosyl residues of
steroidal saponins, was incapable of hydrolyzing the α-1,2-rham-
nosyl residue of 8 [33]. What causes compound 8 to be so diffi-
cultly catalyzed by most biocatalysts? This problem needs further
investigation.
"
In order to investigate the biotransformation pathway of glucosy-
lation catalyzed by Toruzyme, the one-glucose products (3-1, 4-1,
3, 4 and 5 are shown in l Fig. 2a–c.
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

Original Papers 1729
Fig. 3 Effect of reaction media on enzymatic activity. The effect of reac-
tion media on products was investigated by using compound 3. 30 µL of
enzyme were incubated with compound 3 (1 mg) and dextrin (5 mg) in
1 mL of buffer, 30%, 50%, 70%, and 100% methanol at 50 1°C for 24 h.
The reaction mixtures were determined by measuring the increase of the
product by HPLC. HPLC analysis was carried out on an Agilent 1100 unit
with an Alltech Evaporative Light Scattering Detector 2000 by using an All-
tech-Apolloo-C18 column (5 µm, 250 × 4.6 mm).
Fig. 4 Effect of the amount of enzyme on products. The effect of the
amount of enzyme on products was investigated by using compound 3.
3 µL, 10 µL, and 30 µL of enzyme were incubated with compound 3 (1 mg)
and dextrin (5 mg) in 1 mL of 70% methanol at 50 1°C for 24 h. The reac-
tion mixtures were determined by measuring the increase of the product
by HPLC. HPLC analysis was carried out on an Agilent 1100 unit with an All-
tech Evaporative Light Scattering Detector 2000 by using an Alltech-Apol-
loo-C18 column (5 µm, 250 × 4.6 mm).
Most spirostanol saponins are water insoluble, which has a detri-
mental effect on the yield. Most enzymatic activities, however,
are usually higher in native aqueous media than in organic me-
dia. In order to improve the yield of products and the water solu-
bility of the substrates, pure methanol, methanol-water cosol-
vent, and buffer were used as reaction media. It was demonstrat-
ed that 70% methanol cosolvent could increase the yield. Pure
methanol as a reaction medium, however, induced a rapid inacti-
vation of Toruzyme due to enzymatic denaturation and confor-
"
mational rigidity (determined by HPLC, l Fig. 3). The effects of
substrate concentration/amount of enzyme and time course on
the different degrees of glucosylation and yield were also investi-
"
gated. l Fig. 4 and Fig. 5 (determined by HPLC) show the changes
in products, whereby one-glucose products increased gradually,
while the two- or three-glucose products decreased and disap-
peared rapidly as the amount of enzyme and reaction time
courses increased. The results suggested that the products with
more than one additional glucose could be converted into the
products with only one additional glucose.
Fig. 5 Reaction time course of production. The effect of reaction time
course on products was investigated by using compound 3. 30 µL of en-
zyme were incubated with compound 3 (1 mg) and dextrin (5 mg) in 1 mL
of 70% methanol-water at 50 1°C for 6 h, 24 h, 48 h, 96 h, and 144 h. The
reaction mixtures were determined by measuring the increase of the prod-
uct by HPLC. HPLC analysis was carried out on an Agilent 1100 unit with an
Alltech Evaporative Light Scattering Detector 2000 by using an Alltech-
Apolloo-C18 column (5 µm, 250 × 4.6 mm).
These results confirmed once more our deductions in the inves-
tigation of the biotransformation pathway of compound 3
"
(l Fig. 2a). It was demonstrated that the Toruzyme-catalyzed
glucosylation could use exogenous free glucose as a donor, and
also the glucose linked to the products as a donor. 8 diverse sugar
donors were used in this work. A soluble starch and a corn starch
are high polymers having D-glucopyranose units connected by α-
1,4-glycosidic linkages. β-Cyclodextrin is a cyclic oligomer with 7
α-D-glucopyranoses, and dextrin is one of a number of carbohy-
drates with the same general formula as starch but is a smaller
and less complex molecule. Maltose is a disaccharide composed
of two glucoses connected by α-1,4-glycosidic linkages. Toruzyme
was able to transfer α-1,4-D-glucosyl residues from starch, cyclo-
dextrin, dextrin, and maltose to a distinct substrate. A lactose
consists of a glucose and a galactose connected by a β-1,4-glyco-
sidic linkage, and a sucrose is composed of a glucose and a fruc-
tose connected by a α-1,2-glycosidic linkage. However, Toruzyme
was not able to utilize the β-1,4- or α-1, ‑glycosyl residue as a
donor or the monoglucose as a donor. The effect of 13 metal ions
Toruzyme 3.0 L compared with water, but NH4+ inhibited slightly
the activity, and only one product was detected.
The in vitro cytotoxicity of the substrates and their products
against HL-60 cells was evaluated by the MTT assay. As shown in
"
l Table 2, compounds 1–4 showed considerable cytotoxicity
with IC50 values of 5.98, 9.59, 5.95 and 9.93 µmol/L respectively,
and compound 5 did not show any cytotoxicity against HL-60
cells (IC50 > 150). Compared with their substrates, the cytotoxic-
ity of derivatives of 1, 3, and 5 was not changed, the derivatives
of 2 were slightly enhanced with IC50 values of 2.82 and
3.17 µmol/L, and derivatives of 4 were decreased with IC50 values
of 40.95 and 50.88 µmol/L. The results implied that most of the α-
D-glucosylation derivatives of steroidal saponin did not exhibit
any great changes in cytotoxicity against HL-60 cells. In this work,
it was found that the aglycone could also effect a change in activ-
ity, but less than that of the glycone. For example, on comparing 2
on the enzymatic activity showed that Ca2+, Mg2+, Cu2+, Cd2+, Ba2+
Fe2+, Co2+, Zn2+, Pb2+, Mn2+, Na+, or K+ did not inhibit the activity of
,
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

1730 Original Papers
Substrates
IC50 (µmol/L)
Products
1-1
IC50 (µmol/L)
Table 2 A comparison of cytotox-
icity levels between substrates and
their products.
DDP
0.38 0.08
5.98 0.14
Compound 1
(Timosaponin AIII)
Compound 2
(SaponinTa)
Compound 3
(SaponinTb)
Compound 4
(Trillin)
9.51 1.71
9.59 0.99
5.95 0.37
9.93 0.61
2-1
2-2
3-1
3-2
4-1
4-2
5-1
5-2
5-3
5-4
5-5
2.82 0.39
3.17 0.27
6.80 0.63
8.54 1.30
40.95 4.31
50.88 4.98
Compound 5
(Cantalasaponin I)
> 150
> 150
109 14
> 150
–
> 150
– Without activity assay; Note: IC50 data are given as average SD
6 Akihito Y, Yoshihiro M, Yutaka S. Spirostanol saponins from the rhi-
zomes of Tacca chantrieri and their cytotoxic activity. Phytochemistry
2002; 61: 73–78
7 Hu GH, Lu YH, Mao RG, Wei DZ, Ma ZZ, Zhang H. Aphrodisiac properties
of Allium tuberosum seeds extract. J Ethnopharmacol 2009; 122: 579–
582
8 Mimaki Y, Yokosuka A, Kuroda M, Sashida Y. Cytotoxic activities and
structure – cytotoxic relationships of steroidal saponins. Biol Pharm
Bull 2001; 24: 1286–1289
9 Trouillas P, Corbière C, Liagre B, Duroux JL, Beneytout JL. Structure-func-
tion relationship for saponin effects on cell cycle arrest and apoptosis
in the human 1547 osteosarcoma cells: a molecular modelling ap-
proach of natural molecules structurally close to diosgenin. Bioorg
Med Chem 2005; 13: 1141–1149
10 Giri A, Dhingra V, Giri CC, Singh A, Ward OP, Narasu ML. Biotransforma-
tions using plant cells, organ cultures and enzyme systems: current
trends and future prospects. Biotechnol Adv 2001; 19: 175–199
11 Su JH, Xu JH, Lu WY, Lin GQ. Enzymatic transformation of ginsenoside
Rg3 to Rh2 using newly isolated Fusarium proliferatum ECU2042. J Mol
Catal B Enzym 2006; 38: 113–118
and 3 which have the different aglycones and the same glycones
with their derivatives, the cytotoxicity of derivatives of 2 had a
change, but derivatives of 3 had no change. On comparing 2 and
4 which have the same aglycons and different glycones with their
derivatives, the cytotoxicties of all derivatives showed a consider-
able change.
This is the first report of steroidal saponins with different degrees
of glucosylation prepared by CGTase. These studies provide both
useful and practical information about the chemoselectivity, re-
gioselectivity and stereoselectivity of Toruzyme 3.0 L in the gluco-
sylation of steroidal saponins. And it is also worthwhile to evalu-
ate steroidal saponins and their glucosylated products for cyto-
toxicity against HL-60 human promyelocytic leukemia cells,
which help us to understand the role of the glycones in the phar-
macological activity.
12 Zhan JX, Guo HZ, Dai JG, Zhang YX, Guo D. Microbial transformations of
artemisinin by Cunninghamella echinulata and Aspergillus niger. Tetra-
hedron Lett 2002; 43: 4519–4521
Acknowledgements
!
13 Montero E, Alonso J, Cañada FJ, Fernández-Mayoralas A, Martín-Lomas
M. Regioselectivity of the enzymatic transgalactosidation of D- and L-
xylose catalysed by 3-galactosidases. Carbohydr Res 1998; 305: 383–
391
14 Bowles D, Isayenkova J, Lim EK, Poppenberger B. Glycosyltransferases:
managers of small molecules. Curr Opin Plant Biol 2005; 8: 254–263
15 Gachon CMM, Langlois-Meurinne M, Saindrenan P. Plant secondary me-
tabolism glycosyltransferases: the emerging functional analysis.
Trends Plant Sci 2005; 10: 542–549
16 Kramer CM, Prata RTN, Willits MG, De Luca V, Steffens JC, Graser G. Clon-
ing and regiospecificity studies of two flavonoid glucosyltransferases
from Allium cepa. Phytochemistry 2003; 64: 1069–1076
17 Li W, Koike K, Asada Y, Yoshikawa T, Nikaido T. Biotransformation of
umbelliferone by Panax ginseng root cultures. Tetrahedron Lett 2002;
43: 5633–5635
18 Alexandra T. Bacterial cyclodextrin glucanotransferase. Enzyme Microb
Technol 1998; 22: 678–686
19 Zhu BR, Shi ZQ, He BL. The progress in the study on the property, appli-
cations and Immobilization of cyclodextrin glycosyltransferase. Amino
Acids Biotic Res 1997; 19: 36–39
20 Shimoda K, Hara T, Hamada H, Hamada H. Synthesis of curcumin β-
maltooligosaccharides through biocatalytic glycosylation with Stro-
phanthus gratus cell culture and cyclodextrin glucanotransferase.
Tetrahedron Lett 2007; 48: 4029–4032
21 Nakano H, Kiso T, Okamoto K, Tomita T, Manan MBA, Kitahata S. Synthe-
sis of glycosyl glycerol by cyclodextrin glucanotransferases. J Biosci
Bioeng 2003; 95: 583–588
We are grateful to Mr. He-bing Chen and Miss Yan Xue at the Na-
tional Center of Biomedical Analysis for NMR spectra and FAB
mass spectral measurements, and acknowledge Prof. Yu-guo
Duʼs group at the Research Center for Eco-Environmental Scien-
ces, Chinese Academy of Sciences for provision of a substrate.
The project was supported by National Natural Science Founda-
tion of China (NSFC; 30973632) and Major Program of Municipal
Natural Science Foundation of Beijing (7090001).
References
1 Mahato SB, Ganguly AN, Sahu NP. Steroidal sapinins. Phytochemistry
1982; 21: 956–978
2 Itabashi M, Segawa K, Ikeda Y, Kondo S, Naganawa H, Koyano T, Umeza-
wa K. A new bioactive steroidal saponin, furcreastain, from the plant
Furcraea foetida. Carbohydr Res 2000; 323: 57–62
3 Sparg SG, Light ME, van Staden J. Biological activities and distribution of
plant saponins. J Ethnopharmacol 2004; 94: 219–243
4 Ohtsuki T, Sato M, Koyano T, Kowithayakorn T, Kawahara N, Goda Y, Ish-
ibashi M. Steroidal saponins from Calamus insignis, and their cell
growth and cell cycle inhibitory activities. Bioorg Med Chem 2006;
14: 659–665
5 Ghosh P, Besra SE, Tripathi G, Mitra S, Vedasiromoni JR. Cytotoxic and
apoptogenic effect of tea (Camellia sinensis var. assamica) root extract
(TRE) and two of its steroidal saponins TS1 and TS2 on human leuke-
mic cell lines K562 and U937 and on cells of CML and ALL patients.
Leuk Res 2006; 30: 459–468
́
22 Martın MT, Cruces MA, Alcalde M, Plou FJ, Bernabé M, Antonio B. Synthe-
sis of maltooligosyl fructofuranosides catalyzed by immobilized cyclo-
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

Original Papers 1731
dextrin glucosyltransferase using starch as donor. Tetrahedron 2004;
60: 529–534
23 Fu YL, Yu ZY, Tang XM, Zhao Y, Yuan XL, Wang S, Ms BP, Cong YW. Penno-
genin glycosides with a spirostanol structure are strong platelet ago-
nists: structural requirement for activity and mode of platelet agonist
synergism. J Thromb Haemost 2008; 6: 524–533
24 Zhao Y, Kang LP, Liu YX, Liang YG, Tan DW, Yu ZY, Cong YW, Ma BP. Ste-
roidal saponins from the rhizome of Paris polyphylla and their cyto-
toxic activities. Planta Med 2009; 75: 356–363
29 Singh SB, Thakur RS. Plant saponins. Part 3. Costusoside-I and costuso-
side-J, two new furostanol saponins from the seeds of Costus speciosus.
Phytochemistry 1982; 21: 911–915
30 Hirai Y, Sanada S, Ida Y, Shoji J. Studies on the constituents of Palmae
plants III. The constituents of Chamaerops humilis L. and Trachycarpus
wagnerianus BECC. Chem Pharm Bull 1986; 34: 82–87
31 Agrawal PK, Jain DC, Gupta RK, Thakur RS. Carbon-13 NMR spectro-
scopy of steroidal sapogenins and steroidal saponins. Phytochemistry
1985; 24: 2479–2496
25 Ma BP, Dong JX, Wang BJ, Yan XZ. Studies on the furostanol saponins
from Anemarrhena asphodeloides Bunge. Acta Pharm Sin 1996; 31:
271–277
26 Zhao Y, Kang LP, Liu YX, Zhao Y, Xiong CQ, Ma BP. Three new steroidal
saponins from the rhizome of Paris polyphylla. Magn Reson Chem
2007; 45: 739–744
27 Dini I, Tenore GC, Trimarco E, Dini A. Furostanol saponins in Allium cae-
pa L. Var. tropeana seeds. Food Chem 2005; 93: 205–214
28 Agrawal PK. NMR spectroscopy in the structural elucidation of oligo-
saccharides and glycosides. Phytochemistry 1992; 31: 3307–3330
32 Husáková L, Riva S, Casali M, Nicotra S, Kuzma M, Hunková Z, Kren V.
Enzymatic glycosylation using 6-O-acylated sugar donors and accep-
tors: β-N-acetylhexosaminidase-catalysed synthesis of 6-O,N,N'-tri-
acetylchitobiose and 6′-O,N,N'-triacetylchitobiose. Carbohydr Res
2001; 331: 143–148
33 Feng B, Kang LP, Ma BP, Quan B, Zhou WB, Wang YZ, Zhao Y, Liu YX, Wang
SQ. The substrate specificity of a glucoamylase with steroidal saponin-
rhamnosidase activity from Curvularia lunata. Tetrahedron 2007; 63:
6796–6812
Wang Y-Z et al. Glucosylation of Steroidal… Planta Med 2010; 76: 1724–1731

Copyright of Planta Medica is the property of Georg Thieme Verlag Stuttgart and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.