Comparison of three thermostable β-glucosidases for application in the hydrolysis of soybean isoflavone glycosides

Article abstract of DOI:10.1021/jf1046915

A novel thermostable β-glucosidase (Te-BglA) from Thermoanaerobacter ethanolicus JW200 was cloned, characterized and compared for its activity against isoflavone glycosides with two β-glucosidases (Tm-BglA, Tm-BglB) from Thermotoga maritima. Te-BglA exhib

Full text of DOI:10.1021/jf1046915

ARTICLE
pubs.acs.org/JAFC
Comparison of Three Thermostable β-Glucosidases for Application
in the Hydrolysis of Soybean Isoflavone Glycosides
Xiangfei Song,† Yemin Xue,*,‡ Qilei Wang,† and Xixi Wu†
†Jiangsu Key Laboratory for Microbes and Functional Genomics, College of Life Science, Nanjing Normal University, Nanjing, PR China
210046
‡Department of Food Science and Nutrition, GinLing College, Nanjing Normal University, Nanjing, PR China 210097
ABSTRACT: A novel thermostable β-glucosidase (Te-BglA) from Thermoanaerobacter ethanolicus JW200 was cloned, character-
ized and compared for its activity against isoflavone glycosides with two β-glucosidases (Tm-BglA, Tm-BglB) from Thermotoga
maritima. Te-BglA exhibited maximum hydrolytic activity toward pNP-β-D-glucopyranoside (pNPG) at 80 °C and pH 7.0, was
stable for a pH range of 4.6-7.8 and at 65 °C for 3 h, and had the lowest Km for the natural glycoside salicin and the highest relative
substrate specificity (kcat/Km)(salicin)/(kcat/Km)(pNPG) among the three enzymes. It converted isoflavone glycosides, including
malonyl glycosides, in soybean flour to their aglycons more efficiently than Tm-BglA and Tm-BglB. After 3 h of incubation at 65 °C,
Te-BglA produced complete hydrolysis of four isoflavone glycosides (namely, daidzin, genistin and their malonylated forms),
exhibiting higher productivity of genistein and daidzein than the other two β-glucosidases. Our results suggest that Te-BglA is
preferable to Tm-BglA and Tm-BglB, but all three enzymes have great potential applications in converting isoflavone glycosides into
their aglycons.
KEYWORDS: thermostable β-glucosidase, hydrolysis, soybean flour, isoflavone glycosides
’ INTRODUCTION
soybean seedling with high specificity toward soybean isoflavone
conjugates, and the temperature optimum was 40 °C. Xie et al.19
reported that they isolated one enzyme from an Absidia sp. R
from Liquor Qu that could hydrolyze soybean isoflavone con-
jugates, and the temperature optimum was 50 °C. Two β-glucosi-
dases from the legumes Dalbergia cochinchinensis and Dalbergia
nigrescens were compared for their ability to hydrolyze soybean
isoflavone conjugates from soybean.8 Recently, comparison of two
β-glucosidases from a thermophilic fungus, Paecilomyces thermo-
phila J18, and almond was also reported.20 However, there are
few reports on β-glucosidase from the thermophilic bacteria with
preferential activity toward isoflavone glucosides.8,9 The advan-
tages of thermostable enzymes in industrial processes include
reduced risk of contamination, increased substrate solubility, higher
stability of enzymes against denaturing agents and proteolytic
enzyme, and reduced cost of external cooling.21 Therefore, in
view of the commercial importance of β-glucosidase enzymes,
there is still a great need for finding better β-glucosidases than
those currently tested. High activity and thermostability of β-
glucosidase should attract considerable attention as character-
istics of an efficient enzyme.
Isoflavones, which are abundant in soybeans, reportedly have
various health benefits related to their estrogenic activities, including
reduced risk of cardiovascular disease, lower rates of prostate,
breast and colon cancers and improved bone health.1 Isoflavones
in soybeans exist predominantly in glucoside forms and are rarely
found in aglycon forms.2 The content and composition of isofla-
vones vary in soybean-based foods, depending on processing
techniques such as heat treatment, defoaming, soaking in water,
enzymatic hydrolysis and microbial fermentation.3-5 Numerous
studies have demonstrated that isoflavone aglycons are superior
to isoflavone glycosides in various bioactivities, due to their
effective absorption into the human body.2,6,7 Therefore, there is
much interest in increasing the amounts of isoflavone aglycons in
soy products and converting isoflavone glycosides into their
aglycons.
The formation of free aglycons should be associated with the
isoflavone conjugates hydrolyzing β-glucosidase.8,9 The abilities
of many bacterial and fungal β-glucosidase for converting isoflavone
glycosides into the aglycons have been extensively studied in
recent years,10-13 due to their importance in the preparation of
isoflavone aglycons. Tsangalis et al. reported that β-glucosidase
from Bifidobacteriain soy milk was capable of converting glyco-
sides to their aglycons.14 Ming et al.15 partially purified and char-
acterized β-glucosidase from soybean that could hydrolyze soybean
isoflavone conjugates, and the temperature optimum was 30 °C.
Liu et al.16 reported that an isoflavone conjugate hydrolyzing β-
glucosidase (ICHG) was purified from endophytic bacterium,
Pseudomonas ZD-8, and the temperature optimum was 40 °C.
Kuo et al.17 reported that two β-glucosidases from Bacillus subtilis
had high specific activity toward soybean isoflavones. Suzuki
et al.18 reported one β-glucosidase (GmICHG) from the roots of
Thermoanaerobacter ethanolicus JW200 grows optimally 58-
78 °C and has two different exoacting, β-specific glycosyl hydro-
lases, including a β-xylosidase/R-arabinofranosidase22 and one
β-glucosidase (Te-BglA). Thermatoga maritima is of considerable
interest because of the hyperthermostability of its enzymes, which
includes two different β-glucosidases (Tm-BglA and Tm-BglB).23,24
Received: July 27, 2010
Accepted: January 4, 2011
Revised:
December 21, 2010
Published: February 02, 2011
r
2011 American Chemical Society
1954
dx.doi.org/10.1021/jf1046915 J. Agric. Food Chem. 2011, 59, 1954–1961
|

Journal of Agricultural and Food Chemistry
ARTICLE
In our previous study, T. maritima Tm-BglA was found to have
the ability to hydrolyze isoflavone glucosides and their malony-
lated forms, and potential to increase the isoflavone aglycons in
soybean flour.13,25 Although the Tm-BglB had been reported
previously, there have been no reports regarding its applications
in the hydrolysis of isoflavone glycosides. To get highly active and
efficient isoflavone conjugates hydrolyzing β-glucosidase, we
selected three thermostable β-glucosidases from two thermo-
philic bacterias to evaluate their potentials as catalysts.
In this study, we report the cloning and characterization of a
novel thermostable Te-BglA from T. ethanolicus JW200, and com-
parison of three thermostable β-glucosidases (Tm-BglA, Tm-
BglB and Te-BglA) for soybean glycoside hydrolysis, in order to
find a better catalyst for converting isoflavone glycosides into
their aglycons. Their hydrolysis of the artificial substrate pNP-
glucopyranoside (pNPG) and the natural glycoside salicin, pH
profiles, and thermal stabilities also are compared.
50 μL reaction containing 0.2 mM dNTPs each, 20-35 pmol of each
primer and 0.2 μg of chromosomal DNA template. The Tm-bglB gene
was ligated into pET-20b at the NdeI and XhoI restriction sites, resulting
in the recombinant plasmid pET-20b-Tm-bglB, while the Te-bglA gene
was ligated into pET-20b at XbaI and XhoI restriction sites, resulting in
the recombinant plasmid pET-20b-Te-bglA. These expression plasmids
were sequenced and characterized by restriction analysis. Oligonucleo-
tide primer synthesis was performed by Sangon (Shanghai, China).
Purification of Recombinant Enzymes. All E. coli JM109
(DE3) containing pET-20b-Tm-bglA, pET-20b-Tm-bglB and pET-
20b-Te-bglA were grown in LB with ampicillin at 37 °C to OD600 of
0.8, and incubated further with isopropylthio-β-galactoside (IPTG,
0.8 mM) for 10 h. The cells were harvested by centrifugation, and washed
twice with water, resuspended in 20 mL of 5 mM imidazole, 0.5 mM
NaCl, and 20 mM Tris-HCl buffer (pH 7.9), and fragmentated by three
passes through a French press. The cell extracts were heat-treated
(70 °C, 20 min), then cooled in an ice bath, and centrifuged (9600g,
4 °C, 30 min). The resulting supernatants were purified by Ni-affinity
chromatography (Novagen) to homogeneity as determined by sodium
dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using
10% polyacrylamide running gels with 4% polyacrylamide stacking gels.
Protein Determination and Enzyme Assay. Total protein
concentration was determined by the Bradford method,28 using bovine
serum albumin (BSA, Sigma) as a standard. Enzyme activity was quan-
tified by p-nitrophenol (pNP) release from pNPG. The reaction mixture
contained 10 μL of suitably diluted enzyme, 170 μL of 0.1 M potassium
phthalate buffer (PPB, pH 7.0 for Tm-BglA and Te-BglA; pH 5.8 for
Tm-BglB) and 20 μL of substrate solution (20 mM pNPG). After
incubation at 65 °C for 5 min, the reaction was stopped by adding 0.6 mL
of 1 M Na2CO3, and the A405 was read. A standard curve was prepared
using pNP. One unit was determined as the amount of enzyme pro-
ducing 1 μmol of pNP/min. All reactions were done in triplicate.
β-Glucosidase activity toward the available natural glycoside salicin
(Sigma) was determined by assaying the amount of glucose released.
The reaction mixture (0.1 mL) containing 1 mM salicin, 50 mM PPB,
and appropriately diluted enzyme in the same buffer was incubated at the
respective temperature for 10 min, and was stopped by adding 0.75 mL
of 3,5-dinitrosalicylic acid.29 The reaction mixture was then boiled for
5 min and cooled in an ice water bath, and adding 1.075 mL of ddH2O, at
last the A520 was then read.25 Enzyme activities are reported in inter-
national units (U), the amount of enzyme that releases 1 μmol of glucose
in a minute.
Kinetic Parameters. For determinations of the kinetic parameters
(Km, Vmax, kcat/Km), the purified enzymes were assayed at substrate
concentrations ranging from 1.0 to 10 mM for salicin, and from 0.2 to
2.0 mM for pNPG at their optimal conditions. Kinetic parameters (Km
and Vmax) were determined by the Eadie-Hofstee representation of the
Michaelis-Menten model. The specificity constant, kcat/Km, was calcu-
lated to determine the substrate specificity of each enzyme. To calculate
the catalytic constant, apparent kcat, the amount of protein was divided
by the subunit molecular mass of 52,338 Da for Tm-BglA, 52,639 Da for
Te-BglA or 81,846 Da for Tm-BglB. Each experiment was done in
duplicate. The standard error was recorded to be <2%.
HPLC Analysis. HPLC analysis was used to measure the amounts
of daidzein, genistein, daidzin (Din), genistin (Gin), malonylgenistin
(MGin), and malonyldaidzin (MDin) to quantify changes in isoflavone
content of crude extracts treated with the three recombinant β-gluco-
sidases. Separation and quantification of isoflavones were achieved with
an Agilent HC-C 18 (4.6 ꢀ 250 mm, 5 μm) reverse phase column on an
HP series 1100 HPLC (Agilent Corp., Palo Alto, CA) with the UV
detector set at a wavelength of 260 nm in a manner similar to that
Chuankayan et al. described for the separation of soybean isoflavones.8,9
A linear HPLC gradient was employed. Solvent A was 0.1% phosphoric
acid in water, and solvent B was acetonitrile. Solvent B was increased
’ MATERIALS AND METHODS
Materials. In the present study, soybeans were purchased from Yan
Zhi Fang Co. (Hefei, China). The p-nitrophenyl (pNP) glycoside
substrate pNP-β-D-glucopyranoside (pNPG) was purchased from Sigma
Chemical Co. (St. Louis, MO). High-performance liquid chromatogra-
phy (HPLC) grade methanol and acetonitrile were purchased from
Fisher Scientific (Hanover Park, IL). Isoflavone standards of daidzin
(Din), daidzein, genistin (Gin), and genistein were purchased from
Sigma Chemical Co. (St. Louis, MO); malonyl genistin (MGin) and
malonyl daidzin (MDin) were purchased from Wako Chemical Co.
(Wako Pure Chemical Industries, Ltd., Japan). Thermoanaerobacter
ethanolicus JW200 (ATCC31550) and Thermotoga maritima (ATCC-
43589) were gifts from Dr. J. Wiegel (University of Georgia, USA) and
Dr. Weilan Shao (Nanjing Normal University, China). T. ethanolicus
JW200 and T. maritima cells used for isolation of genomic DNA were
grown in serum bottles containing 100 mL of prereduced clostridial
media at 60 °C,22 and anaerobically at 80 °C in modified Luria-Bertani
(LB) medium,26 respectively. The recombinant plasmid pET-20b-Tm-
bglA previously described was used in preparing purified Tm-BglA.13,25
Escherichia coli JM109 (DE3) (Promega, Madison, WI) was used as host
for the expression of Tm-bglA, Tm-bglB and Te-bglA via the T7 RNA
polymerase expression system with pET-20b plasmids (Novagen). Cells
were cultured at 37 °C in LB supplemented with 100 μg of ampicillin/mL.
All other chemicals used were analytical grade reagents unless otherwise
stated.
DNA Manipulation. Routine DNA manipulations were carried
out essentially as described.27 Plasmid DNA and PCR products were
purified using QIAGEN plasmid kit and PCR purification kit (Germany).
PCR reactions were performed in a PE Applied Biosystems 9700 ther-
mal cycler (Foster City, CA) using standard reaction conditions. DNA
modifying enzymes and polymerases were purchased from TAKARA
(Dailian, China).
Construction of Recombinant Plasmid. Based on the DNA
sequence of the β-glucosidase B (TM0025) from T. maritima, two
synthetic oligodeoxyribonucleotides P1 (50-CCCCATATGATGGAA-
AGGATCGATGAAA-30) and P2 (50-CCGCTCGAGTGGTTTGAA-
TCTCTTCTC-30) were used as primers. The PCR primers P3 (50-GC-
TCTAGAAATAATTTTGTTTAACTTTAAGAAGGAGATATACA-
TATG ATAAAATTTCCAAAAGA-30) and P4 (50-CCCCTCGAGCT-
CAATAGAATTTTTCTGT-30) were designed and synthesized based
on the bglA gene (ZP_00777761) sequence from Thermoanaerobacter
pseudethanolicus ATCC 33223, and used to amplify the Te-bglA gene
from T. ethanolicus JW200 genomic DNA. PCR amplification 30 cycles
with Pyrobest DNA polymerase (TAKARA, China) were carried out in a
1955
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 1. Comparison of the amino acid sequences of Te-BglA to those of other thermophilic bacteria β-glucosidases. Abbreviations: Te,
Thermoanaerobacter ethanolicus JW200 (GU391423); Tp, Thermoanaerobacter pseudethanolicus ATCC 33223 (ZP_00777761); Tb, Thermoanaerobacter
brockii (Z56279.1); Tm, Thermotoga maritima (X74163). The conserved and similar amino acids in these β-glucosidases are indicated by solid and gray
boxes, respectively. The conserved motifs surrounding the nucleophile are boxed, while the respective catalytic residue is underlined by an asterisk.
Sugar-binding amino acid residues that are highly conserved are indicated by arrowheads.
1956
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961

Journal of Agricultural and Food Chemistry
ARTICLE
GU391423) of 447 amino acids (Figure 1). A predicted molecular
weight of 52,639 Da was calculated for Te-BglA and was in agree-
ment with the SDS-PAGE analysis of the recombinant protein
(Figure 2). As expected from the analysis of the Te-bglA sequence,
the translated Te-bglA sequence has the highest similarity with that
of the β-glucosidase gene bglA of Thermoanaerobacter pseudethano-
licus ATCC 33223 (98% identity) (ZP_00777761.1)30 and the β-
glucosidase gene bglA of Thermoanaerobacter brockii (95% identity)
(Z56279.1),31 but shares 53% identity with Tm-BglA (X74163)
from T. maritima.23 It contains the highly conserved peptide motifs
Asn-Glu-Pro (residues 164-166) and Ile-Thr-Glu-Asn-Gly (res-
idues 351-355) (Figure 1), in which the Glu (E165, E353)
residues are typical catalytic residues of GH1 enzymes.18 Sugar-
binding amino acid residues that are highly conserved among all
GH1 enzymes could also be identified in the translated Te-bglA
sequence: Gln-19, His-119, Asn-164, Glu-165, Glu-353, Trp-400,
Glu-407, and Trp-408. These results show that Te-BglA is a ther-
mophilic bacteria member of the GH1 family.
Figure 2. SDS-PAGE of purified enzymes: lane M, the molecular mass
standards (M); lane 1,Tm-BglA; lane 2, Tm-BglB; lane 3, Te-BglA.
from 15 to 35% over 45 min. The solvent flow rate was 0.8 mL/min.
Peaks of soy isoflavone glucosides (Gin, Din, MGin, and MDin) and
aglycons (genistein and daidzein) were identified by matching retention
times with isoflavonoid standards. Relative amounts were calculated
from relative peak areas, because all peaks were well within the linear
range of the instrument.
The three recombinant enzymes were purified to gel electro-
phoretic homogeneity by Ni-affinity chromatography after heat
treatment (Figure 2), and had apparent molecular masses of
52 kDa for Tm-BglA and Te-BglA and 82 kDa for Tm-BglB on
SDS-PAGE, which were in accordance with the theoretically
calculated molecular mass, respectively.
Hydrolysis of Soybean Flour. First, soybean flour (SF) was
extracted with three volumes of n-hexane by stirring for 30 min at room
temperature for defatting, precipitated by centrifugation at 9600g for
20 min and air-dried. To compare the hydrolysis efficiencies of the three
β-glucosidases toward the isoflavonoid glycosides in SF, 10% (w/v) SF
was incubated with the purified Tm-BglA, Tm-BglB and Te-BglA in 0.1
M PPB buffer (pH 7.0 for Tm-BglA and Te-BglA, pH 5.8 for Tm-BglB)
in a thermostatically controlled incubator at 65 °C. The enzyme concen-
tration for each purified enzyme was 50 U per gram SF (activity was
determined with pNPG as substrate at 65 °C). The reactions were
stopped by ice water cooling before analysis.13 The hydrolyzed solutions
were centrifuged at 13000g for 20 min to collect the supernatants and
pellets, respectively. The pellets were then extracted with 80% methanol
at 30 °C for 2 h, and the methanol extract was collected by centrifuga-
tion. The reaction supernatant were dried by speed vacuum and then
resuspended in 80% methanol. The methanol extracts were collected by
centrifugation and filtered through a 0.45 μm filter for the quantification
of isoflavones using HPLC. A control reaction of crude extract without
enzyme at 65 °C was set up in the same manner in 0.1 M PPB buffer of
pH 7.0 and pH 5.8, respectively. Each assay was done in duplicate, and
20 μL of each sample was injected for HPLC analysis. The percent
enzymatic hydrolysis of the three β-glucosidases was evaluated by the
following equation:
Biochemical and Kinetic Parameters. The biochemical
properties of Te-BglA, Tm-BglA and Tm-BglB were evaluated.
The temperature dependence of the three enzyme activities was
determined with pNPG and salicin as substrates (Figure 3). The
optimal temperatures using pNPG as a substrate were 90 °C for
Tm-BglA, and 80 °C for Tm-BglB and Te-BglA (Figure 3A);
those for Tm-BglA, Tm-BglB and Te-BglA using salicin as a
substrate were 85 °C, 90 °C and 75 °C, respectively (Figure 3C).
The optimal temperature of GH1 Tm-BglA and Te-BglA for
pNPG was 5 °C higher than that for salicin, and that of GH3 Tm-
BglB for pNPG was 10 °C lower than that for salicin. Analysis of
enzyme activity using the substrate pNPG demonstrated that the
three purified enzymes were most stable at 65 °C for 3 h
(Figure 3E); when incubated for 5 h, Tm-BglB and Tm-BglA
retained more than 80% of their activity, but Te-BglA lost 90% of
its activity.
Next, the effects of pH on the three enzyme activities on the
substrates, pNPG and salicin, were determined at 37 °C in
various buffers ranging from pH 3.8-8.2 (Figure 3B,D). The
optimal pH with pNPG as a substrate was 6.2 for Tm-BglA, 4.2
for Tm-BglB, and 7.0 for Te-BglA, and that using salicin as a
substrate was 7.0 for Tm-BglA, 5.8 for Tm-BglB, and 7.0 for Te-
BglA, indicating that the optimum pH for Tm-BglA and Tm-
BglB differed with the different substrates. Analysis of enzyme
activity with the substrate pNPG showed that pH stability for Te-
BglA and Tm-BglA was higher than that of Tm-BglB over the pH
range from 5.4 to 8.2, and the three enzymes were stable over a
pH range of 4.6 to 7.8 (Figure 3F).
hydrolysis ð%Þ ¼100 - 100 ꢀ ðisoflavone content in enzyme-
hydrolyzed samples = isoflavone content in
control samples without enzymeÞ
The amounts of isoflavone aglycons produced from SF were expressed
as mg per 100 g of sample.
DNA Sequence Analysis. DNA sequencing was performed by
Sangon (Shanghai, China), and the sequences were analyzed with
Dnaman, version 6.0, of the sequence analysis software package (Lynnon
Biosoft, USA). The homology was analyzed in GenBank with the
BLAST program.
Nucleotide Sequence Accession Number. The nucleotide
sequence of the β-glucosidase gene bglA from T. ethanolicus JW200 was
deposited in the GenBank database under accession number GU391423.
Concerning kinetic properties, the Michaelis constants for the
three enzymes were measured with pNPG and salicin as sub-
strates. The apparent Km value of Tm-BglB for pNPG was lower
than those of Tm-BglA and Te-BglA (Table 1). Its kcat/Km value
was the highest due to a minimum in the Km value and a max-
imum in the kcat value compared to those of Tm-BglA and Te-
BglA, indicating that Tm-BglB had the most efficient pNPG
hydrolysis. For the hydrolysis of salicin, the Km value of Te-BglA
was 4.7 ( 0.04 mM, which was significantly lower than those of
Tm-BglA and Tm-BglB; thus, its kcat/Km value was higher than
’ RESULTS
Sequence Analysis, Expression and Purification. Sequence
analysis of the Te-bglA gene from the T. ethanolicus JW200 revealed
that its open reading frame encodes a protein (accession number,
1957
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 3. The optimal temperature (A, C) and pH (B, D) indicated the three recombinant β-glucosidases' activity against pNPG (A, B) and salicin (C, D) as
the substrate. The optimal pH was determined in 0.1 M PIB from pH 3.4 to 8.2 at 80 °C. Optimal temperatures were detemined at the enzymes’ optimal pH
values, aliquots of purified proteins were incubated at various temperatures (50 to 95 °C), and β-glucosidase activities were assayed as described in Materials and
Methods. The highest residual activity was defined as 100%. The thermostability (E) and pH stability (F) profiles of the three recombinant β-glucosidases with
pNPG as substrate. The purified enzyme in its optimal pH was preincubated at 65 and 70 °C for different times in the absence of substrate, respectively, and
these enzyme activities were then assayed as previously indicated in optimal conditions after cooling in an ice water bath. For pH stability experiments, the
purified enzymes were preincubated in 0.1 M PIB from pH 3.8 to 8.2 for 1 h at 37 °C; then aliquots were transferred in standard reaction mixture to determine
the amount of remaining activity. The activity determined prior to the preincubation was taken as 100%.
that of Tm-BglA, and was almost the same as that of Tm-BglB. It
appeared that Te-BglA and Tm-BglB had higher catalytic effi-
ciency for salicin hydrolysis than Tm-BglA.
Isoflavone Aglycon Production from Soy Flour. Because
Din, Gin, MDin, and MGin were the predominant isoflavones
in the soybean flour, the hydrolysis of these four isoflavone
1958
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961

Journal of Agricultural and Food Chemistry
ARTICLE
Table 1. Kinetic Parameters of the Three Recombinant β-Glucosidases
substrate
enzyme
Km (mM)
Vmax (U mg-1
)
kcat (s-1
)
kcat/Km (M-1 s-1
)
3
pNPG
Tm-BglA
Tm-BglB
Te-BglA
Tm-BglA
Tm-BglB
Te-BglA
0.63 ( 0.01
0.22 ( 0.01
0.79 ( 0.01
12.0 ( 0.3
9.8 ( 0.1
307.3 ( 0.7
1134.0 ( 1.4
67.0 ( 0.7
268.1 ( 0.6
1547.6 ( 2.9
58.8 ( 0.6
4.3 ( 0.08 ꢀ 105
7.2 ( 0.2 ꢀ 106
7.5 ( 0.1 ꢀ 104
1.6 ( 0.06 ꢀ 104
2.6 ( 0.01 ꢀ 104
2.3 ( 0.04 ꢀ 104
salicin
217.5 ( 2.6
191.4 ( 2.5
125.8 ( 3.5
189.7 ( 2.3
261.1 ( 3.5
110.4 ( 3.0
4.7 ( 0.04
The kinetic parameters were determined at their optimal pH and temperature for the substrate concentrations ranging from 1.0 to 10 mM for salicin, and
from 0.2 to 2.0 mM for pNPG, using the standard assay as described in Materials and Methods.
glycosides and the isoflavone aglycons produced after reactions
with or without enzyme at different times were analyzed by
HPLC, in order to evaluate the ability of three β-glucosidases for
hydrolyzing soybean isoflavone glycosides. As shown in Tables 2A
and 3, the hydrolysis of Din and Gin for Te-BglA and Din by Tm-
BglA and Tm-BglB was complete after reaction for 10 min, how-
ever, 10.5% and 6.6% of Gin remained in the reactions containing
Tm-BglA and Tm-BglB, respectively. Simultaneously, Te-BglA,
Tm-BglB and Tm-BglA were found to hydrolyze 59.5% and
81.4%, 37.4% and 75.9%, and 17.8% and 18.6% of the hydrolysis
of MDin and MGin, respectively (Table 3), indicating that Te-
BglA had the most efficient malonyl glycoside hydrolysis. After
3 h of incubation, the remaining MDin and MGin were com-
pletely hydrolyzed by Te-BglA, however, 6.2% of MDin and
12.2% MGin remained in the reaction with Tm-BglB, and 3.7% of
Gin, 26.5% of MDin and 14.5% of MGin remained in the reaction
with Tm-BglA (Table 2A). It appeared that Te-BglA hydrolyzed
both conjugated (MGin and MDin) and nonconjugated (Gin and
Din) glycosides in SF more efficiently than Tm-BglB and Tm-BgA.
Subsequently, the production of aglycon isoflavones with the
three recombinant β-glucosidases was evaluated (Table 2B). In
10 min reaction containing Te-BglA and Tm-BglB, the amount of
two isoflavone aglycons increased rapidly to 68.1 (20.4 ( 1.8 mg of
daidzein, 47.7 ( 0.2 mg of genistein) and 54.2 (13.8 ( 0.1 mg of
daidzein, 41.4 ( 0.1 mg of genistein) mg/100 g, respectively, and
achieved 82.8% and 75.2% of their amounts in 3 h reaction,
respectively. In contrast, Tm-BglA produced 39.4 mg (12.0 (
0.2 mg of daidzein, 27.4 ( 0.4 mg of genistein) of two isoflavone
aglycons per 100 g of SF after 10 min, which hold only 59.2% of
their amounts in 3 h reaction. The hydrolysis of Te-BglA and
Tm-BglB resulted in increases of 0.7-fold and 0.2-fold in daidzein
and 0.7-fold and 0.5-fold in genistein within 10 min of incuba-
tion, respectively. It was exhibited that Te-BglA had faster con-
versions from isoflavone glucosides to aglycons and higher yield
of two aglycon isoflavones than Tm-BglA and Tm-BglB.
properties. Properties corresponding optimal pH and pH stabi-
lity ranges for Te-BglA and Tm-BglA were found to be higher
than those of Tm-BglB, indicating that the pKa values of Tm-
BglA and Te-BglA were higher than that of Tm-BglB. The
optimum pH of Tm-BglA and Tm-BglB differed with the
different substrates, a possible explanation being that the inter-
actions of Tm-BglA and Tm-BglB with different substrates
influence the pKa values of the glutamic acid proton donor.33,34
The optimal temperatures of Tm-BglA and Tm-BglB were higher
than those of Te-BglA, but their temperature optimum differed
with the different substrates. The explanation for this observation
is the possibility that the interactions of the three β-glucosidases
with different substrates influence their conformational forms
with thermostability properties. Despite a previous study that
showed that the thermal stability of recombinant enzyme was
affected by fusing his-tag, resulting in an obvious enzyme
inactivation in the temperature range 65-95 °C,25 the three
recombinant β-glucosidases that had six extra histidines added at
the C-terminal of the enzyme in this study, with temperature
optimum over 75 °C, exhibited high thermostability at 65 °C,
which makes them highly suitable for industrial biological pro-
cesses. As demonstrated by the kinetic results of three enzymes,
Tm-BglB exhibited the lowest Km for pNPG and the most efficient
pNPG hydrolysis, but Te-BglA showed the lowest Km for natural
glycoside salicin and the highest relative substrate specificity
(kcat/Km)(salicin)/(kcat/Km)(pNPG). Moreover, Te-BglA and Tm-
BglB had higher catalytic efficiency for salicin hydrolysis than
Tm-BglA, exhibiting a good liberation of glucose from the natural
glycoside salicin.
The three recombinant β-glucosidases were evaluated for
hydrolysis of isoflavone glycosides and isoflavone aglycons pro-
duced from SF. Of three enzymes, the conversions from iso-
flavone glucosides to their aglycons for Te-BglA and Tm-BglB
were found to be higher than that for the Tm-BglA. However, the
hydrolyses of Din, MDin and MGin for Te-BglA were higher
than those of Tm-BglB and Tm-BglA, resulting in complete
hydrolysis of four isoflavone glycosides (Din, Gin, MDin
and MGin) after 3 h of incubation. Furthermore, Te-BglA
increased the isoflavone aglycons from SF more rapidly than
Tm-BglB and Tm-BgA and produced the highest yield of each
isoflavone aglycon. It is known that the predominant soybean
isoflavone forms are malonyl glycosides, which made up
about 70% of the total soybean isoflavone in soybean.20,35
Most β-glucosides are not effective in hydrolyzing the con-
jugated isoflavones (malonyl and acetyl glycosides) to their
aglycon.10,18,20 In our study, three thermostable β-glucosides
could hydrolyze isoflavone glucosides, including their malo-
nylated forms, but Te-BglA converted malonyl glycosides to
’ DISCUSSION
For effective conversion of isoflavone glycosides into their
aglycons with high productivity and low cost, three recombinant
thermostable β-glucosidases were targeted. The thermostable
enzymes exhibit high hydrolysis yields due to their high initial
velocities, contamination prevention abilities, high substrate sol-
ubility, and increased half-lives.32 In this study, to get a better
catalyst for converting isoflavone glycosides into the aglycons, we
reported the cloning of a novel thermostable β-glucosidase
(Te-BglA) from T. ethanolicus JW200 and comparison of three
thermostable β-glucosidases (Tm-BglA, Tm-BglB and Te-
BglA) for soybean glycoside hydrolysis and their physiochemical
1959
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961

Journal of Agricultural and Food Chemistry
ARTICLE
Table 2. Comparison of Soybean Isoflavone Glycosides in the Hydrolyzed Samples and Isoflavone Aglycons Produced by the
Three Recombinant β-Glucosidasesa
(A) rel amt of isoflavone (%):b isoflavone glycosides
Din Gin MDin MGin
100 100
(B) isoflavone aglycons (mg/100 g):c isoflavone aglycons
enzyme
time
daidzein
genistein
not treated
100
100
0.7
0.1
control (pH 5.8) 10 min
28.5 ( 0.5
40.0 ( 1.0
57.5 ( 0.5
73.5 ( 0.5
0.0
67.0 ( 2.0
138.0 ( 4.0
116.0 ( 1.0
123.0 ( 5.0
10.5 ( 0.5
3.7 ( 0.1
6.6 ( 0.3
0.0
65.5 ( 0.5
59.0 ( 1.0
68.0 ( 1.0
61.0 ( 2.0
55.5 ( 0.5
26.5 ( 0.5
41.0 ( 1.0
6.2 ( 0.2
27.5 ( 0.5
0.0
76.5 ( 1.5
72.0 ( 1.0
76.5 ( 1.5
71.5 ( 1.5
62.0 ( 1.0
14.5 ( 0.5
18.5 ( 0.5
12.2 ( 0.05
14.2 ( 0.05
0.0
1.8 ( 0.05
3.2 ( 0.05
0.6 ( 0.01
1.8 ( 0.07
12.0 ( 0.2
17.7 ( 1.1
13.8 ( 0.1
23.4 ( 1.1
20.4 ( 1.8
26.2 ( 0.3
3.4 ( 0.1
4.0 ( 0.1
1.4 ( 0.05
2.1 ( 0.1
27.4 ( 0.4
48.9 ( 0.4
41.4 ( 0.1
48.7 ( 0.9
47.7 ( 0.2
56.1 ( 1.1
3 h
control (pH 7.0) 10 min
3 h
Tm-BglA
Tm-BglB
Te-BglA
10 min
3 h
0.0
10 min
3 h
0.0
0.0
10 min
3 h
0.0
0.0
0.0
0.0
a Soy flours were incubated in 0.1 M PPB (pH 5.8 or 7.0) at 65 °C for 10 min and 3 h, respectively. Samples that did not contain the enzyme were used as
control, and the amounts at time 0 were set at 100%. Values represent the means ( standard deviation; n = 2 (p < 0.05). b Relative amounts (relative
HPLC peak areas) at 10 min and at 3 h and with three enzymes were determined by HPLC, as described in Materials and Methods. c The isoflavone
aglycon content is expressed as mg per 100 g of sample.
Table 3. Comparison of the Percent Hydrolysis of the Three Recombinant β-Glucosidasesa
percentage enzymatic hydrolysis (%): isoflavone glycosides
enzyme
time
Din
Gin
MDin
MGin
Tm-BglA
10 min
3 h
100
100
100
100
100
100
91.0 ( 0.7
97.0 ( 0.2
89.6 ( 0.3
17.8 ( 0.4
56.6 ( 0.5
37.4 ( 1.1
89.5 ( 0.5
59.5 ( 0.5
18.6 ( 0.8
79.7 ( 0.4
75.9 ( 0.3
83.0 ( 0.3
81.4 ( 0.5
Tm-BglB
Te-BglA
10 min
3 h
100
10 min
100
100
3 h
100
100
a Values represent the means ( standard deviation; n = 2.
their aglycons more efficiently than Tm-BglB and Tm-BgA.
Therefore, T. ethanolicus JW200 β-glucosidase is preferable to
the two β-glucosidases from T. maritima in enzymatic hydro-
lysis for converting isoflavone glycosides into their aglycons,
and all three enzymes have great potential applications in food
and medicine production. The three-dimensional structure
of β-glucosidase A from T. maritima has been published.36
Further work to improve isoflavone glycoside transformation via
the structural modification of the enzymes by molecular modeling
on the basis of the solved three-dimensional structures is in
progress.
In conclusion, a novel thermostable β-glucosidase (Te-BglA)
gene from T.ethanolicus JW200 was cloned and sequenced. It
encodes a protein of 447 amino acids that belongs to the gly-
coside hydrolase family 1. The enzyme could more effectively hy-
drolyze isoflavone glucosides of soy flour to their aglycons than
could two β-glucosidases from T. maritima. The study suggests
that Te-BglA is preferable to Tm-BglA and Tm-BglB, and is a
good candidate for use in industrial conversion of isoflavone
glycosides into their aglycons.
Funding Sources
This was work project BK2010545 supported by NSF of Jiangsu
Province of China.
’ ABBREVIATIONS USED
BSA, bovine serum albumin; HPLC, high performance liquid
chromatography; Din, daidzin; Gin, genistin; GH1, family 1 gly-
cosyl hydrolases; GH3, family 3 glycosyl hydrolases; IPTG, iso-
propylthio-β-galactoside; MDin, malonyl daidzin; MGin, malonyl
genistin; DMSO, dimethyl sulfoxide; pNP, p-nitrophenyl; pNPG,
pNP-β-D-glucopyranoside; PPB, potassium phthalate buffer;
SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electro-
phoresis; SF, soybean flour; Tm-BglA, β-glucosidase A from Ther-
motoga maritima MS8; Tm-BglB, β-glucosidase B from Thermo-
toga maritima; Te-BglA, β-glucosidase from Thermoanaerobacter
ethanolicus JW200
’ REFERENCES
(1) Ishimi, Y.; Miyaura, C.; Ohmura, M.; Onoe, Y.; Sato, T.;
Uchiyama, Y.; Ito, M.; Wang, X.; Suda, T.; Ikegami, S. Selective effects
of genistein, a soybean isoflavone, on beta-lymphopoiesis and bone loss
caused by estrogen deficiency. Endocrinology 1999, 140, 1893–900.
(2) Izumi, T.; Piskula, M. K.; Osawa, S.; Obata, A.; Tobe, K.; Saito,
M.; Kataoka, S.; Kubota, Y.; Kikuchi, M. Soy isoflavone aglycones are
absorbed faster and in higher amounts than their glucosides in humans.
J. Nutr. 2000, 130, 1695–1699.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: þ86-25-85794123. Fax: þ86-25-883707623. E-mail:
xueyemin@njnu.edu.cn).
1960
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961

Journal of Agricultural and Food Chemistry
ARTICLE
(3) Wu, J. P.; Muir, A. D. Isoflavone during protease hydrolysis of
defatted soybean meal. Food Chem. 2010, 118, 328–332.
(4) Anderson, R.; Wolf, W. Composition changes in trypsin inhi-
bitor, phytic acid, saponins and isoflavones related to soybean proces-
sing. J. Nutr. 1995, 125, 581S–588S.
(5) Jackson, C.; Dini, J.; Avandier, C.; Rupasinghe, H.; Faulkner, H.;
Poysa, V. Effects of processing on the content and composition of
isoflavones during manufacturing of soy beverage and tofu. Process
Biochem. 2002, 37, 1117–1123.
(6) Nielsen, I. L.; Williamson, G. Review of the factors affecting
bioavailability of soy isoflavones in humans. Nutr. Cancer 2007, 57, 1–10.
(7) Walsh, K. R.; Haak, S. J.; Bohn, T.; Tian, Q.; Schwartz, S. J.;
Failla, M. L. Isoflavonoid glucosides are deconjugated and absorbed in
the small intestine of human subjects with ileostomies. Am. J. Clin. Nutr.
2007, 85, 1050–1056.
(24) Goyal, K.; Selvakumar, P.; Hayashi, K. Characterization of a
thermostable beta-glucosidase (Bg1B) from Thermotoga maritima showing
transglycosylation activity. J. Mol. Catal B: Enzym. 2001, 15 (1-3), 45–53.
(25) Xue, Y. M.; Song, X. F.; Yu, J. J. Overexpression of β-glucosidase
from Thermotoga maritima for the production of highly purified agly-
cone isoflavones from soy flour. World J. Microbiol. Biotechnol. 2009, 25,
2165–2172.24.
(26) Xue, Y. M.; Shao, W. L. Expression and characterization of a
thermostable beta-xylosidase from the hyperthermophile, Thermotoga
maritima. Biotechnol. Lett. 2004, 26 (19), 1511–1515.
(27) Sambrook, J.; Fritsch, E.; Maniatis, T. Molecular cloning: a
laboratory manual, 2nd ed.; Cold Spring Harbor Laboratory Press: 1989.
(28) Bradford, M. M. A rapid and sensitive method for the quantita-
tion of microgram quantities of protein utilizing the principle of protein-
dye binding. Anal. Biochem. 1976, 72, 248–254.
(8) Chuankhayan, P.; Rimlumduan, T.; Svasti, J.; Cairns, J. Hydro-
lysis of soy isoflavonoid glycosides by Dalbergia β-glucosidases. J. Agric.
Food Chem. 2007, 55, 2407–2412.
(9) Kuo, L. C.; Cheng, W. Y.; Wu, R. Y.; Huang, C. J.; Lee, K. T.
Hydrolysis of black soybean isoflavone glycosides by Bacillus subtilis
natto. Appl. Microbiol. Biotechnol. 2006, 73, 314–320.
(10) Ismail, B.; Hayes, K. β-Glycosidase activity toward different
glycosidic forms of isoflavones. J. Agric. Food Chem. 2005, 53, 4918–4924.
(11) Chun, J. Y.; Kim, J. S.; Kim, J. H. Enrichment of isoflavone
aglycones in soymilk by fermentation with single and mixed cultures of
Streptococcus infantarius 12 and Weissella sp 4. Food Chem. 2008, 109,
278–284.
(29) Miller, G. Use of dinitrosalicylic acid reagent for determination
of reducing sugars. Anal. Chem. 1959, 31, 426–428.
(30) Mitchell, R. W.; Hahn-Haegerdal, B.; Ferchak, J. D.; Pye, E. K.
Characterization of beta-1,4-glucosidase activity in Thermoanaerobacter
ethanolicus. Biotechnol. Bioeng. Symp. 1982, 12, 461–467.
(31) Breves, R.; Bronnenmeier, K.; Wild, N.; Lottspeich, F.;
Staudenbauer, W. L.; Hofemeister, J. Genes encoding two different
beta-glucosidases of Thermoanaerobacter brockii are clustered in a
common operon. Appl. Environ. Microbiol. 1997, 63, 3902–3910.
(32) Haki, G. D.; Rakshit, S. K. Developments in industrially
important thermostable enzymes: a review. Bioresour. Technol. 2003,
89, 17–34.
(12) Yang, S. Q.; Jiang, Z. Q.; Yan, Q. J.; Zhu, H. F. Characterization
of a thermostable extracellular beta-glucosidase with activities of exoglu-
canase and transglycosylation from Paecilomyces thermophila. J. Agric.
Food Chem. 2008, 56, 602–608.
(33) Roberge, M.; Dupont, C.; Morosoli, R.; Shareck, E; Kluepfei, D.
Asparagine-127 of xylanase A from Streptomyces lividans, a key residue in
glycosyl hydrolases of superfamily 4/7: kinetic evidence for its involve-
ment in stabilization Stabilization of the catalytic intermediate. Protein
Eng. 1997, 10, 399–403.
(34) Joshi, M. D.; Sidhu, G.; Pot, I.; Brayer, G. D.; Withers, S. G.;
McIntosh, L. P. Hydrogen bonding and catalysis: a novel explanation for
how a single amino acid substitution can change the pH optimum of a
glycosidase. J. Mol. Biol. 2000, 299, 255–279.
(35) Zhang, C.; Yang, X. Q.; Kong, H. Q.; Lu, Q. Y.; Li, X.; Liu, M. L.
Current studies on stability of soybean isoflavones in different process
conditions. Soybean Sci. 2006, 25, 73–76.
(36) Zechel, D. L.; Boraston, A. B.; Gloster, T.; Boraston, C. M.;
Macdonald, J. M.; Tilbrook, D. M.; Stick, R. V.; Davies, G. J. Iminosugar
glycosidase inhibitors: structural and thermodynamic dissection of the
binding of isofagomine and 1-deoxynojirimycin to beta-glucosidases.
J. Am. Chem. Soc. 2003, 125, 14313–14323.
(13) Xue, Y. M.; Yu, J. J.; Song, X. F. Hydrolysis of soy isoflavone
glycosides by recombinant beta-glucosidase from hyperthermophile
Thermotoga maritima. J. Ind. Microbiol. Biotechnol. 2009, 36, 1401–1408.
(14) Tsangalis, D.; Ashton, J. F.; McGill, A. E. J.; Shah, N. P.
Enzymatic transformation of isoflavone phytoestrogens in soymilk by
β-glucosidase-producing bifidobacteria. J. Food Sci. 2002, 67, 3104–3113.
(15) Ming, C. H.; Terrence, L. G. Partial purification and character-
ization of a soybean β-glucosidase with high specific activity towards
isoflavone conjugates. Phytochemistry 2001, 58, 995–1005.
(16) Liu, Y.; Zhou, S. N.; Chen, Z. S.; Zhong, Y. C.; Liu, Y. H.
Purification and characterization of an isoflavone-conjugates-hydrolyz-
ing R-glucosidase from endophytic bacterium. J. Agric. Food Chem. 2004,
52, 1940–1944.
(17) Kuo, L. C.; Lee, K. T. Cloning, expression, and characterization
of two β-glucosidases from isoflavone glycoside-hydrolyzing Bacillus
subtilis natto. J. Agric. Food Chem. 2008, 56, 119–125.
(18) Suzuki, H.; Watanabe, R.; Fukushima, Y.; Fujita, N.; Noguchi,
A.; Yokoyama, R.; Nishitani, K.; Nishino, T.; Nakayama, T. An
isoflavone conjugate-hydrolyzing β-glucosidase from the roots of soy-
bean (Glycine max) seedlings. J. Biol. Chem. 2006, 4, 30251–30259.
(19) Xie, M. J.; Song, M.; Zou, C. X.; Lu, M. C.; Jin, F. X. Purification
and characterization of an isoflavone-conjugates-hydrolyzing β-glucosi-
dase. Biol. Eng. 2006, 22 (4), 635–638.
(20) Yang, S. Q.; Wang, L. J.; Yan, Q. J.; Jiang, Z. Q.; Li, L. T.
Hydrolysis of soybean isoflavone glycosides by a thermostable beta-
glucosidase from Paecilomyces thermophila. Food Chem. 2009, 115 (4),
1247–1252.
(21) Claire, V.; Gregory, J. Z. Hyperthermophilic Enzymes: Sources,
Uses, and Molecular Mechanisms for Thermostability. Microbiol. Mol.
Biol. Rev. 2001, 65, 1–43.
(22) Mai, V.; Wiegel, J.; Lorenz, W. W. Cloning, sequencing, and
characterization of the bifunctional xylosidase-arabinosidase from the
anaerobic thermophile Thermoanaerobacter ethanolicus. Gene 2000, 247
(1-2), 137–143.
(23) Gabelsberger, J.; Liebl, W.; Schlifer, K. H. Purification and proper-
ties of recombinant β-glucosidase of the hyperthermophilic bacterium
Thermotoga maritima. Appl. Microbiol. Biotechnol. 1993, 40, 44–52.
1961
dx.doi.org/10.1021/jf1046915 |J. Agric. Food Chem. 2011, 59, 1954–1961