Efficient enzymatic preparation of esculentoside B following condition optimization by response surface methodology

Article abstract of DOI:10.1016/j.molcatb.2016.04.013

Esculentoside B (EsB, also named phytolaccagenin 3-O-β-d-xylopyranoside), a pentacyclic triterpene isolated from herbal medicine Radix phytolaccae, has been found to possess multiple pharmacological activities. Nonetheless, the low content in nature and the difficulties in the total synthesis of EsB strongly limit its extensive investigations and further development as a drug candidate. This study aims to provide a practical method for highly efficient preparation of EsB using esculentoside A (EsA, phytolaccagenin 3-O-β-d-glucopyranosyl (1 → 4)-β-d-xylopyranoside) as the starting material. β-d-glucosidase from snailase was used to catalyze the formation of EsB, and the product was then purified and fully characterized by both HRMS and NMR. To prepare EsB in a more cost-effective way, response surface methodology (RSM) was used to explore the potential effects of the reaction conditions (such as reaction temperature, pH, enzyme load, and reaction time) on the conversion rates of EsA. The highest EsB yield of 0.66 mg/ml was obtained experimentally under optimized conditions as follows: temperature 48.28 °C, pH 6.4, enzyme load 4.43%, and reaction time 2.73 h. This result agreed well with the predicted yield of 0.68 mg/ml by RSM. The enzymatic kinetics of this biotransformation was characterized at the optimum pH and temperature. The S₅₀ value was evaluated as 167.4 μM, while the V_max value was 345.6 nmol/min/mg. In summary, this study provided a mild and practical method for the highly efficient preparation of EsB from EsA, which held great promise for large scale production of EsB.

Full text of DOI:10.1016/j.molcatb.2016.04.013

Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Efficient enzymatic preparation of esculentoside B following
condition optimization by response surface methodology
Pan Cuia,b,1, Tong-Yi Doua,b,1, Yan-Ping Suna, Shi-Yang Lia, Lei Fenga, Li-Wei Zoua,
Ping Wanga, Da-Cheng Haoc, Guang-Bo Gea,∗, Ling Yanga,d,∗∗
a Laboratory of Pharmaceutical Resource Discovery, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China
b University of Chinese Academy of Sciences, Beijing 100043, China
c Dalian Jiaotong University, Dalian 116028, China
d Jiangxi University of Traditional Chinese Medicine, Nanchang 330004, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Esculentoside B (EsB, also named phytolaccagenin 3-O-␤-d-xylopyranoside), a pentacyclic triterpene
isolated from herbal medicine Radix phytolaccae, has been found to possess multiple pharmacological
activities. Nonetheless, the low content in nature and the difficulties in the total synthesis of EsB strongly
limit its extensive investigations and further development as a drug candidate. This study aims to provide
a practical method for highly efficient preparation of EsB using esculentoside A (EsA, phytolaccagenin 3-O-
␤-d-glucopyranosyl (1 → 4)-␤-d-xylopyranoside) as the starting material. ␤-d-glucosidase from snailase
was used to catalyze the formation of EsB, and the product was then purified and fully characterized by
both HRMS and NMR. To prepare EsB in a more cost-effective way, response surface methodology (RSM)
was used to explore the potential effects of the reaction conditions (such as reaction temperature, pH,
enzyme load, and reaction time) on the conversion rates of EsA. The highest EsB yield of 0.66 mg/ml was
obtained experimentally under optimized conditions as follows: temperature 48.28 ◦C, pH 6.4, enzyme
load 4.43%, and reaction time 2.73 h. This result agreed well with the predicted yield of 0.68 mg/ml by RSM.
The enzymatic kinetics of this biotransformation was characterized at the optimum pH and temperature.
The S50 value was evaluated as 167.4 ␮M, while the Vmax value was 345.6 nmol/min/mg. In summary, this
study provided a mild and practical method for the highly efficient preparation of EsB from EsA, which
held great promise for large scale production of EsB.
Received 24 December 2015
Received in revised form 25 March 2016
Accepted 28 April 2016
Available online 9 May 2016
Keywords:
Esculentoside A
Esculentoside B
Response surface methodology (RSM)
Enzymatic hydrolysis
Condition optimization
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
the anti-inflammatory activity [3], anti-rheumatic effect, as well
as the inhibition against agrobacterial plant transformation [5],
Radix phytolaccae, the dried root of Phytolacca acinosa Roxb or
Phytolacca americana L., is one of the most commonly used tra-
ditional Chinese medicine against edema, mastitis, and lymphatic
congestion [1,2]. Radix phytolaccae contained a number of saponins
that showed potent analgesic and antidiuretic effects, fungistatic
activity as well as the anti-inflammatory effect [2]. Among these
bioactive saponins, esculentoside B (EsB, also named phytolacca-
genin 3-O-␤-d-xylopyranoside; Fig. 1) had drawn much attention
in recent years for its excellent biological activities. EsB exhibited
and showed much broader spectrum of antifungal activity among
the esculentosides reported, including plant pathogens and human
opportunistic yeasts and molds [6,7]. In respect of toxicity, EsB
showed little hemolytic activity [7]. The excellent pharmacological
activity and low toxicity make it a promising drug candidate.
Unfortunately, the in vivo studies regarding EsB are insufficient
due to its very low content in Radix phytolaccae, and it is chal-
lenging to extract this compound with ample amounts. It is worth
noting that EsB is a hydrolytic product of esculentoside A (EsA,
also named phytolaccagenin 3-O-␤-d-glucopyranosyl (1 → 4)-␤-d-
xylopyranoside; Fig. 1), a naturally abundant component in Radix
phytolaccae [2]. In contrast to EsA, EsB showed much higher antifun-
gal activity [6], anti-inflammatory activity, as well as more potent
cytotoxicity against VA-13, HepG2 and A549 cells [8]. In addition,
EsA also exhibited very strong haemolytic activity which might fur-
ther result in its toxicity because of the sugar moiety [8,9]. Recently,
∗
Corresponding author.
Corresponding author at:Laboratory of Pharmaceutical Resource Discovery,
∗∗
Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023,
China.
E-mail addresses: geguangbo@dicp.ac.cn (G.-B. Ge), yling@dicp.ac.cn (L. Yang).
These two authors contributed equally to this work.
1
http://dx.doi.org/10.1016/j.molcatb.2016.04.013
1381-1177/© 2016 Elsevier B.V. All rights reserved.

26
P. Cui et al. / Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
Fig. 1. Scheme of esculentoside B using esculentoside A as the starting material.
several researchers tried to hydrolyze EsA by the acid and micro-
bial biotransformation [9,10]. However, acid hydrolysis produced
corresponding byproduct and brought environmental problems.
Furthermore, these methods were time-consuming, of poor selec-
tivity, and often accompanied by the complicated post-treatment
process. Therefore, it is highly desirable to develop a practical and
proficient method for the large-scale preparation of EsB using EsA
as the starting material.
Biocatalysis is increasingly recognized as an important man-
ufacturing strategy in chemical and pharmaceutical industries,
especially in the preparation of pharmaceutical intermediates [12].
Enzyme can be utilized to produce useful products for its efficient
catalytic capability and perfect selectivity. For instance, a great
number of glucosidases have been characterized and exploited for
the production of desired pharmaceutical aglycones in recent years
[13–15]. Previously, we have successfully isolated and character-
ized several glucosidases from snailase, an easily available enzyme
preparation which contains multiple glucosidase isoenzymes, and
prepared some rare saponins, such as ginsenoside Rd, compound
K, Mx, and Mc, by the enzymatic hydrolysis [16,17]. However, until
now, no report has been provided to achieve the efficient prepara-
tion of EsB by enzymatic hydrolysis.
In this study, a highly efficient method to prepare EsB via
enzymatic hydrolysis of EsA was reported for the first time. ␤-d-
glucosidase isolated from snailase was utilized to remove the C-3
glucosyl group of EsA to obtain EsB. To obtain the optimal condi-
tions for this enzymatic hydrolysis, response surface methodology
(RSM) combined with central composite rotatable design (CCD) was
performed to investigate the potential effects of different reaction
parameters on the conversion rate of EsA. A series of factors on
enzymatic hydrolysis, including reaction temperature, pH, enzyme
load, and reaction time, are estimated by this model. Under the opti-
mal conditions, a very high conversion rate was achieved, which
showed great potential in the large-scale production. Furthermore,
the kinetics behavior of EsA hydrolysis was characterized to deeply
understand the catalytic properties of this enzymatic process.
2.2. Separation and assay of ˇ-d-glucosidase fractions
Pre-experiment had indicated that snailase could hydrolyze
both glycosyl groups in EsA, as shown in Supplemental Fig. S1.
To obtain ␤-d-glucosidase fractions from snailase that could selec-
tively remove the glucosyl group from EsA to form EsB, gel filtration
chromatography was performed, and sample fractions were kept at
4 ◦C for further use. The buffer used was 50 mM Tris-HCl (pH 7.5)
(buffer A).
One hundred milligrams of the viscera powder from China white
jade snail were dissolved in 5 ml buffer A, and then filtrated through
a 0.22 ␮m microfilter. Then the crude enzyme was loaded onto
a Sephacryl S-200HR column(Ø1.6 cm × 80 cm) previously equili-
brated with buffer A. The proteins were eluted in the same buffer
with flow rate of 0.7 ml/min. Fractions with glycosidase activity
were collected and concentrated for further use.
Assay of the ␤-d-glucosidase activity of these fractions were per-
formed in a 0.2 ml reaction system containing 0.02 ml of 40 mM
pNPG (dissolved in deionized water), 0.005 ml of enzyme fraction
(dissolved in buffer A), and 0.175 ml of 50 mM Tris-HCl buffer (pH
7.5). The hydrolytic product of pNPG was identified referencing to
Akao T [18] with some modifications. In brief, the absorbance was
read online for 20 min at 405 nm and 37 ◦C by Synergy H1Hybrid
Multi-Mode Microplate Reader (BioTek), with time intervals of half
a minute. The control incubation without enzyme fractions was
performed under the same conditions to ensure that the absorbance
produced was enzyme-dependent. One unit of the enzyme activ-
ity was defined as the amount of enzyme liberating 1 ␮mol of
p-nitrophenol per minute under the above conditions.
Protein concentration was determined according to the Brad-
ford method [19] using bovine serum albumin (BSA) as a standard.
The actual chromatogram for snailase separation by Sephacryl S-
200HR column was recorded at 280 nm.
2.3. Enzymatic hydrolysis of EsA
To remove the C-3 glucosyl group from EsA, enzymatic reac-
tion was carried out using the isolated ␤-d-glucosidase fractions.
In brief, the mixture (0.2 ml) containing ␤-d-glucosidase fractions
(0.035 mg protein/ml), 10 mM CaCl2, 500 ␮M EsA, and 50 mM Tris-
HCl buffer (pH 7.5) were incubated together. After incubation at
37 ◦C for 2 h in shaking water bath, the reaction was terminated
by the addition of 0.4 ml ice-cold methanol, and then the super-
natant acquired by centrifugation at 20,000g for 20 min was used
for HPLC analysis. Blank controls without enzyme sources or sub-
strate were carried out to ensure that the product formation was
enzyme dependent.
2. Experimental
2.1. Materials and reagents
Esculentoside A (Purity > 98%) was obtained from Chengdu Pufei
De Biotech Co., Ltd. pNP-␤-d-glucopyranoside (pNPG) was pur-
chased from Sigma-Aldrich (St. Louis, MO); Snailase preparation
was purchased from Beijing BioDee Biotechnology Co., Ltd. as the
enzyme source. Sephacryl S-200HR was purchased from GE Health-
care, and home-packed. Methanol and acetonitrile (HPLC grade)
were from Tedia (Fairfield, USA) and millipore water (Millipore,
Bedford, MA) was employed. All the other reagents were of AR
grade.
2.4. LC analytical conditions
EsA and its product were analyzed by LC system (Shimadzu,
Kyoto, Japan), which contains a SCL-10A system controller, two LC-

P. Cui et al. / Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
27
Table 1
dle level), 1 (higher level), 2 (highest level). The experiment design
contains 30 points with 6 replications determined by the formula
(2k + 2k + 6), in which k is the number of factors [25,26]. With the
concentration of EsB as the response value, the experimental results
of RSM were fitted using the following second order polynomial Eq.
(1) [25,27].
Independent variables and their coded levels for the CCD.
Variables
Symbols
Levels
−2
−1
0
1
2
Temperature (Temp; ◦C)
pH
Enzyme load (El; %)
Reaction time (Rt; h)
X1
X2
X3
X4
30
6
40
6.5
50
7
3.34
3
60
70
8
5.57
5
7.5
4.46
4
ꢀ
ꢀ
ꢀꢀ
1.11 2.23
1
k
k
k
2
Y = ˇ0
+
ˇiXi +
ˇiiXi2
+
ˇijXiXj
(1)
i=1
i=1
i<j=1
where Y is the predicted response; Xi, Xj are independent variables;
10AT pumps, a SIL-10A auto-injector and a SPD-10AVP UV detector.
An ODS C18 column (150 × 4.6 mm, 5 um; Kromasil) were adopted
to separate EsA and its product.
Chromatographic separation was conducted as follows: The
mobile phase was acetonitrile (A) and water(B) with a gradient pro-
gram: 0–4.5 min, 70% B-50% B; 4.5–6 min, 50% B-5% B; 6–8 min, 5%
B; 8–10 min, balance to 70% B and column was kept at 40 ◦C. The
flow rate was set to 1 ml/min, and the detector was set at 203 nm.
For hydrolysate analysis, 10 ␮l of the sample was injected into LC.
is the constant coefficient; is the linear coefficient; is the
ˇ0
ˇi
ˇii
quadratic coefficient and
number of factors.
the interaction coefficient; k is the
ˇij
The software Design Expert 8.05b was adopted to perform
the experimental design, data analysis, model fitting, and graph
plotting. Each response could be obtained according to the
corresponding experimental design, and a multiple non-linear
regression was made using the variables and responses, expressed
in the form of a 3D surface to illustrate the effects of variables or
their interaction on response.
2.5. Structural characterization
2.7. Kinetic analysis
EsA and its product were analyzed by a high resolution mass
spectrometry (HRMS). The mass detection was performed on
an Orbitrap Elite mass spectrometer (Thermo Scientific, Bremen,
Germany) equipped with an ESI source. MS spectra were recorded
in negative-ion mode from m/z 400–1000. The detection conditions
were as follows: ion spray voltage, 3.5 kV; capillary temperature,
300 ◦C. Nitrogen was used as the sheath gas (25 arbitrary units) and
auxiliary gas (11 arbitrary units), and helium served as the collision
gas. The tube lens and capillary voltages were set to 80 and 35 V,
respectively. Collision-induced dissociation (CID) was conducted
with an isolation width of 2 Da, and the activation time was set at
30 ms. CID was conducted in the dynamic exclusion mode, with
repeat counts of 2, repeat duration of 30 s, exclusion list size of
50, and exclusion duration of 180 s. All ion acquisition experiments
were performed using the Orbitrap mass spectrometer, and scan-
ning was performed at a resolution of 30,000 for both full-scan MS
and data-dependent scan MS.
To estimate the kinetic parameters for the formation of EsB, EsA
(50–1000 ␮M) was incubated with ␤-d-glucosidase (3.8 ␮g/ml) for
10 min at the optimum temperature and pH, which were deter-
mined by RSM. The formation of EsB with ␤-d-glucosidase was
linear with respect to the incubation time and ␤-d-glucosidase con-
centration over ranges relevant to this study. All incubations were
carried out in triplicate. The experimental data was fit to a particu-
lar model using nonlinear regression analysis according to the Hill
Eq. (2) [28,29] with Origin Pro 8.0. Kinetic constants were recorded
as the value S.D. of the parameter estimates.
Vmax × Sn
v =
(2)
S5n0 + Sn
where v is the initial rate, Vmax is the maximum velocity of reaction,
S50 is the concentration at half Vmax, S is the substrate concentra-
tion, and n the Hill coefficient.
EsB biotransformed using ␤-d-glucosidase was characterized by
NMR. In brief, 500 ␮M EsA was incubated with ␤-d-glucosidase
fraction (0.05 mg protein/ml), 10 mM CaCl2, and 50 mM Tris-HCl
buffer (pH 7.5) in 20 ml final incubations for 4 h at 37 ◦C. After
centrifugation at 20,000g for 20 min at 4 ◦C, the combined super-
natants were concentrated to give the residue, which was dissolved
in 1 ml methanol, and purified by preparative reverse chromatogra-
phy. The entire process was monitored by TLC with the developing
solvent (methylene chloride: methanol: water = 9:1:0.1). The struc-
ture of product was determined on a Varian INOVA-400 NMR
spectrometer (Varian, Palo Alto, CA). The purified product was
stored at −20 ◦C before dissolving in DMSO-d6 (Euriso-Top, Saint-
Aubin, France) for NMR analysis. Chemical shifts were given on ␦
scale and referenced to tetramethylsilane (TMS) at 0 ppm for 1H
NMR (400 MHz) and 13C NMR (100 MHz).
3. Results and discussion
3.1. Separation and assay of the ˇ-d-glucosidase fraction
After loading the snailase sample to Sephacryl S-200HR column
according to Section 2.1.3, at least two different chromatographic
peaks are resolved. The major peak exhibited a strong hydroly-
sis activity against the synthesized ␤-d-glucoside pNPG, as shown
in Fig. 2. The active fractions 26–32 against pNPG were pooled
and utilized as ␤-d-glucosidase. For the minor peak, however, no
hydrolysis activity against pNPG was detected.
This spearation step resulted in an approximately 2.74-fold
purification, with 86.1% of the total ␤-d-glucosidase activity
present in the final preparation. The specific activity of this prepa-
ration was 0.0322 U/mg (Table 3).
2.6. Experimental design by RSM
The central composite design (CCD), a statistically-based exper-
imental design approach, has been widely used for estimating
second order response surfaces [20–24]. In this work, it is intro-
duced to study the interaction of different parameters to determine
the optimum conditions of the EsB preparation from EsA. Based
on the single-factor experiment results, temperature, pH, enzyme
load, and reaction time are subjected to the further optimization.
As presented in Table 1, each of four independent variables had
five coded levels, i.e., −2 (lowest level), −1 (lower level), 0 (mid-
3.2. LC profiles of EsA and its product
After co-incubation with EsA (500 ␮M) and ␤-d-glucosidase
(0.035 mg/ml), the hydrolytic products were analyzed by LC-UV.
Only one product was detected in the incubation system (Fig. 3).
The retention times were approximately 3.98 min and 4.72 min
for EsA and its product M1 respectively, and the latter was in
good agreement with EsB (Supplemental Fig. S2). It was noted
that ␤-d-glucosidase was highly reaction specific, as no additional

28
P. Cui et al. / Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
Table 2
Central composite design arrangement and the corresponding value for response
variables.
Run
Coded level of variable
Y (mg/ml)
order
Temp( ◦C)
pH
El (%)
Rt (h)
Actual
Predicted
1
2
3
4
5
6
7
8
−1
1
−1
−1
1
−1
−1
−1
−1
1
1
1
1
−1
−1
−1
−1
1
1
1
1
0
−1
−1
−1
−1
−1
−1
−1
−1
1
1
1
1
1
1
1
1
0
0
0
0
0
0.51
0.62
0.32
0.30
0.60
0.60
0.44
0.43
0.56
0.37
0.56
0.32
0.66
0.55
0.56
0.45
0.38
0.19
0.59
0.45
0.45
0.59
0.47
0.55
0.64
0.58
0.59
0.64
0.64
0.60
0.51
0.54
0.35
0.32
0.57
0.62
0.41
0.41
0.55
0.39
0.53
0.32
0.63
0.49
0.61
0.43
0.39
0.23
0.66
0.43
0.46
0.63
0.50
0.56
0.61
0.61
0.61
0.61
0.61
0.61
−1
1
1
−1
1
−1
−1
1
−1
1
1
9
−1
1
−1
−1
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
−1
1
1
−1
1
−1
−1
1
1
0
Fig. 2. Sephacryl S-200HR size-exclusion chromatography. (᭿) LC-UV profile at
280 nm, (᭹)␤-d-glucosidase activity.
−1
1
−2
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
−2
2
0
0
0
0
0
0
0
0
−2
2
0
0
0
0
0
0
0
0
0
−2
2
0
0
0
0
0
0
0
0
0
bon region in the spectrum of EsB disappeared, which could be
assigned as the terminal glucose of EsA. The carbon signal of C-4ꢀ
in EsB displayed a downfield shift by −7.38 ppm, while the C-3ꢀ
and C-5ꢀ appeared at ı75.23 and 63.66 shifted upload by +1.91 and
+2.44 ppm, respectively. This is probably due to the deprivation of
glucose in 4ꢀ-substituted ␤-d-xylopyranoside. These characteristic
chemical shifts indicated that the ␤-d-glucose located at C-4ꢀ␤-
d-xylose of EsA was removed and the product was identified as
EsB.
Fig. 3. Representative chromatographic profiles of EsA hydrolysis whith (B) or
wihtout (A) ␤-d-glucosidase.
peaks were identified compared with the blank incubations with-
out enzymes or without EsA.
3.4. Response surface analysis
3.3. Structural characterization
Table 2 presented the results of CCD experiments investigating
the influence of independent variables on EsB production, along
with the corresponding predicted value and actual value. The EsB
yield ranged from 0.19 to 0.66 mg/ml. The experimental data of CCD
design could be fitted with the following second order polynomial
equation:
The product generated by ␤-d-glucosidase was extracted by
ethyl acetate and purified by LC. Then this product was identified
by HRMS in negative ion mode, which is more sensitive than the
positive ion mode for the identification of M1. The molecular ion
peak of M1 is located at m/z 663.3824 and 709.3877 (Supplemen-
tal Fig. S3), corresponding to [M1−H]− and [M1 + HCOO]− of EsB
(C36H56O11).
Y = 0.61 − 0.0390X1 − 0.0570X2 + 0.0420X3 + 0.0140
X4 − 0.0120X1X2 + 0.0054X1X3 − 0.0460X1X4
+ 0.0014X2X3 + 0.0370X2X4 + 0.0056X3X4
The hydrolysis product was further purified by LC, and then well
characterized by NMR. The 1H NMR spectrum data for EsA and EsB
were listed below and the 13C NMR data was shown in Table 4
(Supplemental Fig. S4-S7):
2
2
2
2
− 0.0770X1 − 0.0180X2 − 0.0180X3 − 0.0210X4
1H NMR (400 MHz, DMSO-d6):ı 0.70 (3H, s, H-26), 0.82 (3H, s, H-
29), 1.09 (3H, s, H-27), 1.10 (3H, s, H-24), 1.20 (3H, s, H-25), 3.63 (3H,
s, OMe), 4.25 (1H, d, J = 7.8 Hz, H-3), 4.29 (1H, d, J = 7.4 Hz, Glc-H),
5.19 (1H, brs, H-12);
where Y is the respected EsB yield, X1 is reaction temperature, X2
is pH, X3 is enzyme load, X4 is reaction time.
The evaluation of the polynomial model equation significance
was carried out by Fisher’s statistical test for the analysis of vari-
ance (ANOVA; Table 5). The F value of 12.26 for EsB suggested that
this model was statistically significant; the p value less than 0.01%
implied no more than 0.01% chance that insignificance of “Model
F-Value” could occur due to noise. The coefficient of determination
(R2) was 0.9196, indicating that this model could explain 91.96%
1H NMR (400 MHz, DMSO-d6):ı 0.70 (3H, s, H-26), 0.81 (3H, s,
H-29), 1.08 (3H, s, H-27), 1.10 (3H, s, H-24), 1.20 (3H, s, H-25), 3.63
(3H, s, OMe), 4.25 (1H, d, J = 7.0 Hz, H-3), 5.19 (1H, brs, H-12).
In comparison of 13C NMR spectrum of EsA with that of EsB,
nearly all signals of the aglycone moiety in both compounds
appeared at the same position. A series of signals of the sugar car-

P. Cui et al. / Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
29
Table 3
Purification of ␤-d-glucosidase from China white jade snail.
Purification
steps
Total protein
(mg)
Total activity
(Units)
Specific activity
(Units/mg)
Yield
(%)
Purification
fold
Crude enzyme
Sephacryl S-200
61.27
19.25
0.72
0.62
0.0117
0.0322
100
86.11
1
2.74
Table 4
13C NMR data for EsA and EsB.
Esculentoside A
13C
Esculentoside B
13C
Esculentoside A
13C
Esculentoside B
Position
␦
␦
Position
␦
␦
13C
1
2
3
4
5
6
7
8
43.72
70.19
80.71
42.56
46.59
17.33
32.46
40.59
47.85
36.36
23.17
122.56
144.05
41.84
27.83
23.48
45.40
42.14
42.07
44.08
30.15
43.72
70.08
80.73
42.56
46.59
17.33
32.45
40.89
47.83
36.36
23.17
122.56
144.04
41.84
27.64
23.45
45.40
42.13
42.06
44.06
30.13
22
23
24
25
26
27
28
29
30
31 (OMe)
Xyl-1
2
3
4
5
33.73
63.66
14.60
16.88
17.33
26.00
178.66
28.26
176.81
52.06
105.44
73.92
75.23
77.51
63.66
102.07
73.13
76.81
70.62
76.89
61.60
33.72
63.73
14.61
17.00
17.33
25.99
178.66
28.25
176.81
52.07
105.68
74.07
77.14
70.13
66.10
9
10
11
12
13
14
15
16
17
18
19
20
21
Glc-1
2
3
4
5
6
Table 5
ANOVA for the quadratic polynomial model for level optimization of EsB production using ␤-d-glucosidase.
Source
Sum of Square
Df
Mean
F value
p-value
Model
X1-Temp
X2-pH
X3-El
X4-Rt
X1X2
X1X3
X1X4
X2X3
X2X4
0.39
0.037
0.078
0.042
14
1
1
1
1
1
1
1
1
1
1
1
1
1
1
15
10
5
0.028
0.037
0.078
0.042
12.26
16.50
34.77
18.55
2.14
1.07
0.21
14.84
9.179 × 10−3
9.53
0.23
71.64
4.03
4.02
5.47
< 0.0001
0.0010
< 0.0001
0.0006
0.1643
0.3174
0.6530
0.0016
0.9249
0.0075
0.6412
< 0.0001
0.0630
0.0633
0.0336
Significant
4.817 × 10−3
2.409 × 10−3
4.739 × 10−4
0.033
4.817 × 10−3
2.409 × 10−3
4.739 × 10−4
0.033
2.067 × 10−5
0.021
2.067 × 10−5
0.021
X3X4
5.096 × 10−4
0.16
5.096 × 10−4
0.16
2
X1
X2
X3
X4
2
9.079 × 10−3
9.056 × 10−3
0.012
9.079 × 10−3
9.056 × 10−3
0.012
2
2
Residual
0.034
0.029
2.252 × 10−3
2.947 × 10−3
8.635 × 10−4
Lack of Fit
Pure Error
Cor Total
3.41
0.0939
Not significant
4.318 × 10−3
0.42
29
of the total variations. The adjusted R2 value 0.8446 implied the
good fitting of the regression model. For an approved model, “Adeq
precision” larger than 4 is desirable, and the high value of 12.72 in
this model indicated an adequate signal, implying that this model
is competent for the design optimization.
contrast, reaction time showed a less significant effect on the yield
of EsB.
Three dimension surface plots were made to investigate the
relationship between different variables and response, in order to
obtain the optimal hydrolysis conditions that would maximize the
yield of EsB (Fig. 4).
The production of EsB increased with the temperature chang-
ing from 30 to 50 ◦C, but a gradual decrease occurred over 50 ◦C
(Fig. 4a–c). Temperature usually influences the reaction process
by altering the collision frequency between enzyme and sub-
strate [27], and influences the enzyme activity by changing the
enzyme conformation [27,31]. Under a certain temperature range,
an increase in temperature could contribute to the improvement
The p-value was an important reference to determine the sig-
nificance of each coefficient, and this value less than 0.05 indicated
statistically significant for a model term [27,30]. From Table 5 and
the model equation, it could be concluded that the model linear
terms X1, X2, and X3, the interaction terms X1X4 and X2X4, and
2
2
the quadratic terms X1 and X4 could significantly affect the EsB
yield (p < 0.05). Among these statistically significant variables, pH
(X2) was the most important factor that influenced the hydroly-
sis process, followed by enzyme load (X3) and temperature (X1). In

30
P. Cui et al. / Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
Fig. 4. The contributions and relationships of the factors that may affect the yields of EsB.
in enzyme activity. However, any further increase in temperature
above the optimum value might lead to enzyme inactivation.
The effects of pH values on EsB yield followed a similar trend.
pH functioned mainly by modification of enzyme conformation
[25], and the native conformation was preferred to reach the maxi-
mum enzyme activity. Fig. 4d and 4e indicated that EsB production
increased with the decrease of pH from 8.0 to 6.5, but decreased
with the further reduction below pH 6.5 due to enzyme inactiva-
tion.
97.5% and the yield of EsB was 0.66 0.03 mg/ml, which was in
good accordance with the predicted value. Furthermore, this model
was also validated with additional experiments under the selected
random conditions while fixing the reaction time at the optimized
value (Table 6). Therefore, the predicted model was considered to
be reliable and robust.
3.5. Kinetic analysis
It was obvious that the response gradually enhanced with the
increase of enzyme load and reached a platform after the enzyme
load surpassed 4.5% (Fig. 4f). The excessive enzymes were not con-
ducive to the increase in the yield. In fact, the response could reach
the maximum value within 3 h at the above enzyme load, and the
excessive reaction time would otherwise decrease the yield of EsB.
Moreover, the strong interaction was detected between reac-
tion time and other two variables (reaction temperature and pH;
Fig. 4c,e). At the lower temperature level, a prolongation in the reac-
tion time would contribute to the production of EsB, which may be
due to a lower reaction rate at this condition. However, higher tem-
perature adversely affected the yield of EsB, which reduced as the
reaction continued (Fig. 4e). The interaction between reaction time
and pH (Fig. 4c) nearly followed the same pattern.
To gain a deeper understanding of the catalytic properties of
␤-d-glucosidase mediated biotransformation of EsA to EsB, the
enzyme kinetic analysis was performed using various substrate
concentrations incubated with ␤-d-glucosidase. Prior to the kinetic
analysis, the incubation conditions were optimized to ensure that
the formation rates of EsB were within the linear range in relation
According to the optimization of the regression model, the opti-
mum conditions giving the maximal EsB yield were as follows:
temperature 48.28 ◦C, pH 6.4, enzyme load 4.43%, and reaction time
2.73 h. Under this condition, the actual conversion rate is up to
Table 6
Validation of experimental data.
Serial no.
Temp(◦C)
pH
El (%)
Actual (mg/ml)
Predicted (mg/ml)
1
2
3
4
48.3
35.5
50.5
65.5
6.4
6.0
7.5
6.8
4.43
5.15
4.43
3.22
0.66 0.03
0.53 0.01
0.57 0.01
0.44 0.02
0.68
0.51
0.55
0.41
Fig. 5. Kinetic analysis of EsA hydrolysis.

P. Cui et al. / Journal of Molecular Catalysis B: Enzymatic 130 (2016) 25–31
31
to the incubation time and protein concentration. The formation
References
rate of EsB was linear up to 20 min of incubation and linear up
to 9.5 ␮g/ml ␤-d-glucosidase under the optimized reaction condi-
tions (pH 6.4 and 48.3 ◦C). In this case, the kinetic parameters were
determined using the final enzyme concentration of 3.8 ␮g/ml fol-
lowing 10 min incubation at 48.3 ◦C in Tris-HCl buffer (pH 6.4). EsA
hydrolysis exhibited atypical (sigmoidal) kinetics, as depicted by
the ‘hockey stick’ Eadie-Hofstee plot (Fig. 5). The data well fitted
with Hill plot, while the degree of sigmoidicity was determined by
Hill coefficient, as n = 2.42 0.09. The S50 value and the Vmax value
for the formation of EsB were also determined as 167.4 3.2 ␮M
and 345.6 3.0 nmol/min/mg protein, respectively. These results
demonstrated that ␤-d-glucosidase from snailase exhibited very
high reactivity towards EsA hydrolysis, and the formation rate of
EsB were more than 300 nmol/min/mg protein at high substrate
concentrations (>0.5 mM).
[1] F. Wei, R.S. Singh, M. Fueth, S. Swarts, P. Okunieff, H. Derendorf, J. Pharm.
Biomed. Anal. 107 (2015) 82–88.
[2] H.M. Gao, J.X. Liu, Z.M. Wang, W.H. Wang, J. Asian Nat. Prod. Res. 11 (5) (2009)
433–438.
[3] H.J. Chi, H.S. Kim, Arch. Pharm. Res. 8 (1) (1985) 15–20.
[5] H. Kanzaki, T. Kagemori, Y. Yamachika, T. Nitoda, K. Kawazu, Biosci.
Biotechnol. Biochem. 63 (9) (1999) 1657–1659.
[6] A.M. Escalante, C.B. Santecchia, S.N. López, M.A. Gattuso, A. Gutiérrez Ravelo,
F. Delle Monache, M. Gonzalez Sierra, S.A. Zacchino, J. Ethnopharmacol. 82 (1)
(2002) 29–34.
[7] A. Escalante, M. Gattuso, P. Pérez, S. Zacchino, J. Nat. Prod. 71 (10) (2008)
1720–1725.
[8] L. Wang, L. Bai, T. Nagasawa, T. Hasegawa, X. Yang, J.-i. Sakai, Y. Bai, T. Kataoka,
S. Oka, K. Hirose, A. Tomida, T. Tsuruo, M. Ando, J. Nat. Prod. 71 (2008) 35–40.
[9] W. Gong, Z.H. Jiang, P. Sun, L. Li, Y.S. Jin, L.C. Shao, W. Zhang, B.S. Liu, H.W.
Zhang, H. Tang, Chem. Biodivers. 8 (10) (2011) 1833–1852.
[10] Y.Y. Zhu, L.W. Qian, J. Zhang, J.H. Liu, B.Y. Yu, Tetrahedron 67 (23) (2011)
4206–4211.
[12] M.E. Hegazy, T.A. Mohamed, A.I. ElShamy, A.E. Mohamed, U.A. Mahalel, E.H.
Reda, A.M. Shaheen, W.A. Tawfik, A.A. Shahat, K.A. Shams, N.S. Abdel-Azim,
F.M. Hammouda, J. Adv. Res. 6 (1) (2015) 17–33.
4. Conclusion
[13] T.Y. Dou, H.W. Luan, X.B. Liu, S.Y. Li, X.F. Du, L. Yang, Biotechnol. Lett. 37 (9)
(2015) 1905–1910.
In this study, ␤-d-glucosidase from snailase was used for the
preparation of EsB by direct hydrolysis of EsA. The reaction condi-
tions were carefully optimized and determined by RSM combined
with central composite rotatory design. Under optimal conditions
(reaction temperature 48.28 ◦C, pH 6.4, enzyme load 4.43%, reac-
tion time 2.73 h), the maximum conversion of EsA is up to 97.5%
and the yield of EsB is 0.66 0.03 mg/ml, which agreed well with
the RSM prediction. The kinetic behaviors of this biotransformation
demonstrated that ␤-d-glucosidase from snailase had very high
reactivity towards EsA hydrolysis. ␤-d-glucosidase from snailase
could be used as a highly efficient catalyst to prepare EsB from EsA,
and the enzymatic method holds great promise for the large-scale
production of EsB.
[14] Q.X. Yang, L. Wang, L.X. Zhang, H.B. Xiao, J. Mol. Catal. B: Enzym. 90 (2013)
132–138.
[15] C.S. Park, M.H. Yoo, K.H. Noh, D.K. Oh, Appl. Microbiol. Biotechnol. 87 (1)
(2010) 9–19.
[16] H.W. Luan, X. Liu, X.H. Qi, Y. Hu, D.C. Hao, Y. Cui, L. Yang, Process Biochem. 41
(9) (2006) 1974–1980.
[17] Y. Hu, H.W. Luan, D.C. Hao, H.B. Xiao, S.L. Yang, L. Yang, Enzyme Microb.
Technol. 40 (5) (2007) 1358–1366.
[18] T. Akao, Q.M. Che, K. Kobashi, L. Yang, M. Hattori, T. Namba, Appl. Environ.
Microbiol. 60 (3) (1994) 1041–1043.
[19] M.M. Bradford, Anal. Biochem. 72 (1) (1976) 248–254.
[20] P. Mahapatra, A. Kumari, V. Kumar Garlapati, R. Banerjee, A. Nag, J. Mol. Catal.
B: Enzym. 60 (1–2) (2009) 57–63.
[21] T. Ponrasu, B. Manohar, S. Divakar, Process Biochem. 44 (2) (2009) 251–255.
[22] U. Pongket, W. Piyatheerawong, S. Thapphasaraphong, A. HKittikun,
Biotechnol. Biotec. Eq. 29 (5) (2015) 926–934.
[23] J.N. Hu, X.M. Zhu, K.T. Lee, Y.N. Zheng, W. Li, L.K. Han, Z.M. Fang, L.J. Gu, B.S.
Sun, C.Y. Wang, Biol. Pharm. Bull. 31 (10) (2008) 1870–1874.
[24] M.A. Bezerra, R.E. Santelli, E.P. Oliveira, L.S. Villar, L.A. Escaleira, Talanta 76 (5)
(2008) 965–977.
Acknowledgement
[25] B. Das, A.P. Roy, S. Bhattacharjee, S. Chakraborty, C. Bhattacharjee, Ecotoxicol.
Environ. Saf. 121 (2015) 244–252.
[26] K. Karthikeyan, K. Nanthakumar, K. Shanthi, P. Lakshmanaperumalsamy, Iran.
J. Microbiol. 2 (4) (2010) 213.
[27] S. Zhu, Y. Li, C.Y. Ma, Z.X. Lou, S.W. Chen, J. Dai, H.X. Wang, J. Mol. Catal. B:
Enzym. 97 (2013) 87–94.
This work was supported by the National Basic Research Pro-
gram of China (2013CB531805), and NSF of China (81573501,
81473181, 81302793, 81273590).
[28] M.G. Soars, B.J. Ring, S.A. Wrighton, Drug Metab. Dispos. 31 (6) (2003)
762–767.
Appendix A. Supplementary data
[29] L.E. Witherow, J.B. Houston, J. Pharmacol. Exp. Ther. 290 (1) (1999) 58–65.
[30] W. Li, W. Du, D. Liu, J. Mol. Catal. B: Enzym. 45 (3–4) (2007) 122–127.
[31] N.A. Kabbashi, N.I. Mohammed, M.Z. Alam, M.E.S. Mirghani, J. Mol. Catal. B:
Enzym. 116 (2015) 95–100.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcatb.2016.04.
013.