A novel method for the highly efficient biotransformation of genistein from genistin using a high-speed counter-current chromatography bioreactor

Article abstract of DOI:10.1039/c8ra10629k

Genistein, an important soybean isoflavone compound, has gained attention for its significant properties. Compared with the glycone form of genistin, low content of genistein limits the use in food and pharmaceutical fields. In this study, a novel bioreactor with high-speed counter-current chromatography (HSccc) was built for the highly efficient biotransformation of genistein from genistin. The solvent system for the bioreactor was selected according to the K_D values. The selected solvent system was evaluated by the enzyme activity of β-glucosidase. An ethyl acetate/buffer solution was selected as the preferred solvent system for the HSccc bioreactor. Optimum reactor parameters were selected according to the retention of the stationary phase. The HSccc bioreactor was operated using different flow rates, and 2.0 mL min^-1 was chosen as the optimal flow rate with a conversion rate of over 90% within 24 h; the novel bioreactor easily immobilized and recycled the enzyme and could be applied in the preparation of genistein.

Full text of DOI:10.1039/c8ra10629k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
A novel method for the highly efficient
biotransformation of genistein from genistin using
Cite this: RSC Adv., 2019, 9, 4892
a high-speed counter-current chromatography
bioreactor
a
Daijie Wang,
Muhammad Shafiq Khan,b Li Cui,a Xiangyun Song,a Heng Zhu,a
Tianyu Ma,ac Xiaoyu Li and Rong Sun
c
def
*
Genistein, an important soybean isoflavone compound, has gained attention for its significant properties.
Compared with the glycone form of genistin, low content of genistein limits the use in food and
pharmaceutical fields. In this study, a novel bioreactor with high-speed counter-current chromatography
(HSCCC) was built for the highly efficient biotransformation of genistein from genistin. The solvent
system for the bioreactor was selected according to the KD values. The selected solvent system was
evaluated by the enzyme activity of b-glucosidase. An ethyl acetate/buffer solution was selected as the
preferred solvent system for the HSCCC bioreactor. Optimum reactor parameters were selected
according to the retention of the stationary phase. The HSCCC bioreactor was operated using different
flow rates, and 2.0 mL minꢀ1 was chosen as the optimal flow rate with a conversion rate of over 90%
within 24 h; the novel bioreactor easily immobilized and recycled the enzyme and could be applied in
the preparation of genistein.
Received 29th December 2018
Accepted 22nd January 2019
DOI: 10.1039/c8ra10629k
rsc.li/rsc-advances
stronger antioxidant potency in DPPH radical-scavenging and
LDL oxidation assays than genistin.10 However, genistein is
1. Introduction
Genistein (40,5,7-trihydroxyisoavone), an important isoavone
compound, is widely distributed in leguminous plants such as
soybeans, Pueraria lobata (Willd.) Ohwi and Sophora japonica L.
Genistein is structurally similar to estrogen and has a number
of biological properties that can prove benecial in alleviating
a variety of health conditions including kidney disease, meno-
pausal symptoms, neuronal loss, Alzheimer's disease, and
cancer.1–7 Because of this broad range of biological properties,
genistein has gained attention as a food additive and dietary
supplement. Currently, there is an enormous demand for gen-
istein in health and food products.
mainly present as a glycoside form of genistin in soybeans.
Therefore, the conversion of genistein from genistin is very
important. There are many methods for the transformation of
genistin to genistein, which include alkaline/acidic hydrolysis
and microbial and enzymatic transformations, as shown in
Fig. 1.11–13 Alkaline or acidic hydrolysis always results in a strong
reaction and causes environmental pollution. On the other
hand, biotransformation using enzymes or microorganisms has
many advantages including simplicity of procedure, mild
conditions, low cost, and less pollution of the environment. To
provide efficient biotransformation conditions, many types of
reactors need to be built. These include a membrane, a packed
bed, and an immobilized enzyme reactor.14,15 There are also
many obstacles that need to be overcome such as physical
damage to the membrane, membrane pollution, cleaning of the
immobilized enzyme, and enzyme recycling.
Genistein is a glycone form of genistin (40,5,7-trihydroxyiso-
avone-b-D-glucopyranoside) with a higher potential for intes-
tinal absorption than genistin.8,9 Genistein has also shown
aKey Laboratory of TCM Quality Control, Shandong Analysis and Test Center, Qilu
University of Technology (Shandong Academy of Sciences), Jinan 250014, P. R. China
bDepartment of Biotechnology and Bioinfomatics, International Islamic University,
Islamabad 44000, Islamic Republic of Pakistan
cCollege of Pharmacy, Shandong University of Traditional Chinese Medicine, Jinan,
China
dInstitute of Advanced Medical Science, Shandong University, Jinan 250012, P. R.
China. E-mail: sunrong@sdu.edu.cn
eThe Second Hospital of Shandong University, Jinan 250033, P. R. China
fThe Post of Taishan Scholar in Traditional Chinese Medicine Pharmacology and
Toxicology, Jinan 250033, P. R. China
Fig. 1 Chemical structures of the compounds and a schematic of
enzymolysis.
4892 | RSC Adv., 2019, 9, 4892–4899
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
High-speed counter-current chromatography (HSCCC) is China). HPLC-grade methanol was purchased from the Fisher
a liquid–liquid partition chromatographic technique that Company (Fairlawn, NJ, USA). Water used was deionized by an
eliminates the possibility of irreversible sample adsorption on osmosis Milli-Q system (Millipore, Bedford, MA, USA). Reverse
a solid support. Furthermore, HSCCC provides high recovery osmosis Milli-Q water (Millipore, USA) was used in this study.
and rapid and effective separation, is easily scaled-up and The purities of genistin and genistein were over 98% (HPLC,
simple to operate, and has low solvent consumption. It has Desite Biotech Co., Ltd. Chengdu, China).
recently become a useful tool for the preparative isolation and
purication of various natural products.16–19 The HSCCC
2.2 Apparatus
column is a hollow polytetrauoroethylene coil twining around
The HSCCC equipment was TBE-300A (Shanghai, Tauto
the HSCCC column holder. The two-phase solvent system is set
Biotech, China) with three 300 mL multi-layer coil separation
at a high distribution rate of over 13 times per second (a rota-
columns (diameter of the PTFE tube was 1.6 mm) as well as
tion speed of 800 rpm) (Fig. 2). When the enzyme or microor-
a 20 mL manual sample loop. The b-values of the HSCCC
ganism remains in the stationary phase of the HSCCC solvent
system and the substrate continues to be eluted in the mobile
phase, a novel biotransformation reactor is constructed. There
are many advantages of this novel reactor such as a decrease in
substrate inhibition, contact between the enzyme and the
substrate, and high purity of products outowing in the mobile
phase.
In this study, a novel HSCCC bioreactor was built for the
highly efficient biotransformation of genistin to genistein. The
bioreactor parameters, the type of enzyme used, and the ow
rate of the mobile phase were optimized. To the best of our
knowledge, to date, an HSCCC bioreactor has not been used for
the biotransformation of genistein. In this study, an attempt
was made to perform a biotransformation of genistin to genis-
column ranged from 0.47 (internal) to 0.73 (external) (b ¼ r/R,
where r is the rotation radius or the distance from the coil to the
holder sha, and R (R ¼ 7.5 cm) is the revolution radius or the
distance between the holder axis and the central axis of the
centrifuge). The HSCCC apparatus was equipped with four
other instrument modules, i.e. a 8823A-UV monitor at 280 nm
(Beijing Emilion Technology, Beijing, China), a Model 3057
portable recorder (Yokogawa, Sichuan Instrument Factory,
Sichuan, China), a TBP-5002 constant-ow pump and a DC-0506
low constant temperature bath (Tauto Biotechnique, Shanghai,
China), to maintain the temperature at 40 C. The spectropho-
tometer used was the Genesys 10S UV-Vis scanning spectro-
ꢁ
photometer (Thermo Fisher Scientic, Waltham, MA, USA).
HPLC separation was performed using the Agilent 1260 system
tein using HSCCC. This method could be used to transform
(Agilent Technologies, USA) consisting of a quaternary solvent
other compounds and enriches the types of bioreactors.
delivery system, vacuum degasser, diode array detector (DAD),
auto-sampler and column oven connected to the Agilent
2. Materials and methods
2.1 Reagents and materials
ChemStation soware. The HPLC column used was the Waters
Symmetry C18 column (250 mm ꢂ 4.6 mm, i.d. 5 mm, USA).
n-Hexane, methanol, citric acid, sodium citrate, ethyl acetate, n-
butanol, acetic acid and dichloromethane were analytical grade
2.3 Determination of the partition coefficient
(Sinopharm Chemical Reagent Co., Ltd, Shanghai, China). p- The partition coefficient (KD) of the substrate and the product in
Nitrophenyl b-D-glucopyranoside (p-NPG) and b-glucosidase different solvent systems was determined by HPLC. The
were biochemical grade (Sangon Biotech Co., Ltd. Shanghai, substrate (genistin) and the product (genistein) at appropriate
Fig. 2 The instrument construction of the HSCCC reactor. (A) Type-J planetary motion of the HSCCC multi-layer coil column. (B) Schematic of
the distribution and motion of two phases in the HSCCC column.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 4892–4899 | 4893

View Article Online
RSC Advances
Paper
amounts were placed in 5 mL test tubes, dissolved with the in a separatory funnel, to which 120 mg of enzyme was added.
stationary phase of the two-phase solvent system, and detected The funnel was placed on a shaker for uninterrupted mixing.
by HPLC. The peak areas AS1 and AP1 were obtained. An equal Every 4 h, 6 mL of the upper and lower phase solutions were
volume of the mobile phase solution was added to the lower sampled. The organic phase was discarded, and the aqueous
phase solution followed by vigorous shaking to ensure complete phase was divided into three equal portions. Each portion was
mixing. Aer the equilibrium of the partition was reached, the mixed with 2 mL of 0.8 mg mLꢀ1 p-NPG solution and heated in
stationary phase was analyzed by HPLC to obtain the peak areas a 45 ꢁC water bath for 30 min. Next, 2 mL of 1.0 mol Lꢀ1 Na2CO3
AS2 and AP2. The KD value was dened as the peak area of the aqueous solution was added to terminate the reaction, and
target compound in the stationary phase divided by the peak changes in the color of the solution were observed. The absor-
area of the target compound in the mobile phase. The partition bance values AY1, AY2, and AY3 were measured at 400 nm, and the
ꢀ
coefficients of the substrate and product were calculated as average absorbance value AY was calculated.
follows: KSD ¼ (AS1 ꢀ AS2)/AS2, KPD ¼ (AP1 ꢀ AP2)/AP2
.
The blank and reference solution were prepared by mixing
The partition coefficient of b-glucosidase was determined in 2 mL of 0.4 mg mLꢀ1 initial and inactivated b-glucosidase buffer
a two-phase solvent system and analyzed as follows: 2 mg of b- solution with 2 mL of 0.8 mg mLꢀ1ꢁ p-NPG solution, which were
glucosidase was fully dissolved in 6 mL of the stationary phase then heated in a water bath at 45 C for 30 min. The reactions
solution of the two-phase solvent system to obtain solution A. were terminated by the addition of 2 mL of 1 mol Lꢀ1 Na2CO3
Then, 3 mg of p-NPG was fully dissolved in 1 mL of the solution. Moreover, the absorbances AM (blank) and A0 (refer-
stationary phase solution to obtain solution B. Furthermore, ence) at 400 nm were measured by a spectrophotometer. Each
2 mL of solution A and 1 mL of solution B were heated in a 45 ꢁC measurement was averaged three times, and the average
water bath for 30 min, and 1 mL of 1 mol Lꢀ1 Na2CO3 solution enzyme activity was calculated using the following formula:
ꢀ
was added to terminate the reaction. The absorbance E1 was enzyme activity/% ¼ (AY ꢀ A0) ꢂ 100%/(AM ꢀ A0).
measured by a spectrophotometer at 400 nm. Then, 4 mL of
solution A was mixed with an equal volume of the mobile phase
solution of the two-phase solvent system followed by vigorous
shaking to ensure complete mixing. Aer the equilibrium of the
partition was reached, 2 mL of the lower phase solution was
heated with 1 mL of solution B in a 45 ꢁC water bath for 30 min.
The reaction was terminated by adding 1 mL of 1 mol Lꢀ1
Na2CO3 solution, and the absorbance E2 was measured by
a spectrophotometer at 400 nm. The reference solution was
prepared by heating an inactivated b-glucosidase solution A
(2 mg of inactivated b-glucosidase in 6 mL of the stationary
phase solution) with 1 mL of solution B in a 45 ꢁC water bath for
30 min. The reaction was terminated by adding 1 mL of
1 mol Lꢀ1 Na2CO3 solution, and the absorbance E0 was
measured by a spectrophotometer at 400 nm. The partition
coefficient of b-glucosidase in the two phases was calculated
using the following formula: KD ¼ (E1 ꢀ E2)/(E2 ꢀ E0).
2.5 Optimization of the HSCCC reactor parameters
During the process of counter-current chromatography, there were
two modes of rotation direction for the main unit: forward rotation
(FWD) and reverse rotation (REV). The input port could serve as the
inlet (IN) or the outlet (OUT) of the main unit. Therefore, there
were four elution modes for the counter-current chromatography:
REV-IN, FWD-OUT, FWD-IN, and REV-OUT. In this experiment, we
tested the retention of the stationary phase using an ethyl acetate/
buffer solution (1 : 1, v/v) solvent system, four elution modes of the
counter-current chromatography (REV-IN, FWD-OUT, FWD-IN,
and REV-OUT), the lower phase as the stationary phase, and the
upper phase as the mobile phase. At rst, the entire length of the
Teon tubing of the high-speed counter-current chromatography
system was lled with the stationary phase solution at the ow rate
of 30 mL minꢀ1; once the main unit was initiated, the rotation
speed was slowly adjusted to 800 rpm, and the temperature was
maintained at 40 ꢁC. Aer the rotation speed stabilized, the mobile
phase was pumped in at 2 mL minꢀ1 until equilibrium was
reached. Aer the rotation was stopped, the two-phase solvent
system in the separation column was transferred under nitrogen to
a graduated cylinder to calculate the retention of the stationary
phase (Sf) such that Sf (%) ¼ (VS/VT) ꢂ 100%, where VS is the
volume of the stationary phase and VT is the total volume of the
HSCCC column.
2.4 Determination of the maximum solubility and enzyme
activity
Aer 2.0 mg of b-glucosidase was accurately weighed, the
stationary phase of the two-phase solvent system was gradually
added under sonication. When the solution became clear, its
volume was determined, and the maximum solubility of b-
glucosidase was determined to be 750 mg Lꢀ1. Then, 2.0 mg of
genistin was accurately weighed, and the mobile phase of the
two-phase solvent system was gradually added under sonica-
tion. When a clear solution was obtained, its volume was
determined, and the maximum solubility of genistin was
In this experiment, Sf was measured in parallel at 550, 600,
650, 700, 750, and 800 rpm. The measurement method was the
same as that used for different elution modes, and each
measurement was averaged three times.
determined to be 62.5 mg Lꢀ1
.
2.6 Operation of the HSCCC bioreactor
Aqueous solutions of citric acid and sodium citrate at
0.10 mol Lꢀ1 were separately prepared, mixed, and ltered, and
2.6.1 Preparation of the two-phase solvent and sample
a citric acid/sodium citrate blank buffer solution (pH ¼ 5) was solutions. The ethyl acetate/buffer solution (1 : 1, v/v) solvent
obtained. Then, 600 mL of a two-phase solvent system consisting system was placed in a separatory funnel, fully shaken, and
of equal volumes of ethyl acetate and buffer solution was placed allowed to stand for phase separation. Aer the two phases
4894 | RSC Adv., 2019, 9, 4892–4899
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
reached equilibrium, the upper and lower phases were sepa- processing the active ingredients in natural medicines. A suit-
rately sampled and degassed by ultrasonication. According to able two-phase solvent system requires that (1) the enzymes are
the requirements of the counter-current chromatography mainly distributed in the stationary phase since distribution in
bioreactor, the following three solutions were prepared: (1) the mobile phase results in a loss of enzymes; (2) the product is
a portion of the upper phase solution was used as the mobile mainly distributed in the mobile phase and can be rapidly
phase of the counter-current chromatography system, to which eluted with the mobile phase; and (3) the substrate has good
genistin was added and allowed to fully dissolve under ultra- solubility in the mobile phase. Based on the principles of
sonication to obtain a substrate solution of 62.5 mg Lꢀ1; (2) the enzymatic reactions, the larger contact area between the
lower phase was the stationary phase, to which b-glucosidase substrate and the enzyme could enhance the catalytic efficiency.
was added and fully mixed to prepare a 750 mg Lꢀ1 bioreactor However, HSCCC is a partition chromatography in which the
enzyme solution; and (3) a portion of the upper phase solution substrate undergoes vigorous mixing and partitioning inside
was stored as the blank mobile phase. The abovementioned the column. Therefore, if the substrate is adequately distributed
ꢁ
three solutions were kept in a 40 C water bath before use.
in the stationary phase of the selected two-phase solvent system,
2.6.2 Operation of the HSCCC bioreactor. At rst, an the HSCCC bioreactors can be implemented.
enzyme solution of the bioreactor was pumped into the entire
As shown in Table 1, the distribution of the substrate and
HSCCC column at the ow rate of 30 mL minꢀ1. Aer the loading product was analyzed in the n-hexane/buffer solution (1 : 1, v/v),
was completed, the main unit was started. Using the FWD-OUT ethyl acetate/buffer solution (1 : 1, v/v), dichloromethane/buffer
mode (rotating in a clockwise direction), the mobile phase was solution (1 : 1, v/v), and n-butanol/buffer (1 : 1, v/v) solution.
pumped in from the end of the HSCCC column. The rotation The results showed that the KD values in the n-hexane/buffer
speed was adjusted to 800 rpm, and the blank mobile phase was solution solvent system were below 0.1, indicating that both
introduced at the end of the counter-current chromatography the substrate and product were mainly distributed in the
separation column at a certain ow rate. Aer the two-phase stationary phase, and the product could not be continuously
solvent system reached hydrodynamic equilibrium, the substrate eluted with the mobile phase aer enzymatic hydrolysis.
mobile phase solution containing genistin was pumped in. The Therefore, the n-hexane/buffer solution was not suitable as
detector was turned on, the detector wavelength was set at 254 nm, a solvent system for the HSCCC reactors. The remaining solvent
and the recorder was started to initiate data acquisition. The eluted systems provided satisfactory KD values for the substrate and
solution was obtained in 10 mL test tubes.
product and could be further tested for enzymatic activity as
Aer biotransformation of genistin by the HSCCC bioreactor, suitable two-phase solvent systems.
1 mL of the eluted solution was periodically obtained, dried with
nitrogen, re-dissolved with methanol, and ltered with a 0.45 mm
3.2 Effects of the solvent system on the enzyme activity
needle lter. The conversion rate of the target peak was analyzed
by HPLC. Aer the conversion dropped below a certain value, the
run was stopped, the liquid in the separation column was
extracted with nitrogen, and Sf was calculated.
Fig. 3 shows activity changes for b-glucosidase in three solvent
systems: ethyl acetate/buffer (1 : 1, v/v), dichloromethane/buffer
(1 : 1, v/v), and n-butanol/buffer (1 : 1, v/v) solutions. The results
showed that enzyme activity in the dichloromethane/buffer
solution solvent system decreased to less than 1% at 4 h and
to 0% at 8 h. We suspect that dichloromethane readily dena-
tures b-glucosidase and thus destabilizes the enzyme structure.
Therefore, dichloromethane is not suitable as a solvent for the
HSCCC bioreactor. The ethyl acetate/buffer and n-butanol/
buffer solution solvent systems provided higher stability of
the enzyme activity during the test. The enzyme activity was
above 99% at 48 h, indicating that the ethyl acetate and n-
2.7 HPLC analysis condition and structural identication
The mobile phase comprised methanol/water (70 : 30, v/v), with
isocratic elution at the ow rate of 1.0 mL minꢀ1 and a column
temperature of 30 ꢁC. The detection wavelength was set at
254 nm, and the injection volume was 10 mL.
The identication of the converted product was performed
via mass spectrometry (ESI-MS) with electrospray ionization
(ESI)/quadrupole (Q) ESI-MS data (quadrupole LC/MS 6120,
Agilent Technologies, Palo Alto, CA) in the direct injection
mode, and the MS conditions were as follows: mass range: 100–
1000 m/z, capillary temperature: 300 ꢁC, spray voltage: 4 kV,
Table 1 The KD-values for resveratrol in a HSCCC bioreactor with
different solvent systemsa
skimmer voltage: 40 V, auxiliary gas (nitrogen) ow: 6 L minꢀ1
,
KD-values
and positive mode. NMR spectra were obtained using the
Bruker AV-400 spectrometer (Bruker BioSpin, Rheinstetten,
Germany) with TMS as the internal standard.
Solvent systems
Genistin
Genistein
n-Hexane/buffer solution (1 : 1, v/v)b
Ethyl acetate/buffer solution (1 : 1, v/v)b
n-Butanol/buffer solution (1 : 1, v/v)b
Dichloromethane/buffer solution (1 : 1, v/v)c
<0.1
0.66
3.05
>10
<0.1
>10
>10
<0.1
3. Results and discussion
3.1 Selection of the solvent system
a
The buffer solution contained citric acid (0.10 M) and a sodium citrate
b
c
The selection of an appropriate two-phase solvent system is one
of the key steps in the application of HSCCC as a bioreactor for
buffer solution (pH 5). Lower phase as the stationary phase. Upper
phase as the stationary phase.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 4892–4899 | 4895

View Article Online
RSC Advances
Paper
suitable elution modes for the HSCCC reactors. The FWD-IN
and REV-OUT modes, under which the Sf was zero, were not
suitable for the HSCCC reactors.
Fig. 4B shows the effect of different rotation speeds (i.e. 550,
600, 650, 700, 750, and 800 rpm) on the Sf in the FWD-OUT
mode. As shown, the Sf was 0 when the rotation speed was
550 rpm, and it signicantly increased with the increasing
rotation speed. The stationary phase retention rate increased to
as high as 64.5% as the rotation speed reached 800 rpm.
Therefore, the FWD-OUT mode and 800 rpm rotation speed
were chosen as the optimal parameters for the HSCCC
bioreactor.
Fig. 3 Effect of the solvent system on enzyme activity. EtOAc: ethyl
acetate/buffer solution (1 : 1, v/v), BuOH: n-butanol/buffer solution
(1 : 1, v/v), and CH2Cl2: dichloromethane/buffer solution (1 : 1, v/v).
3.4 Optimization of the ow rate for the HSCCC bioreactor
Fig. 5 shows the HSCCC chromatograms and purity results for
the biotransformation of genistin to genistein in the novel
butanol systems did not readily cause enzyme denaturation. In
the n-butanol/buffer solution solvent system, the n-butanol
phase was saturated with a large amount of water. This caused
b-glucosidase to elute with the mobile phase and lowered the
enzyme concentration during the operation of the bioreactor.
Eventually, the ethyl acetate/buffer solution was selected as the
preferred two-phase solvent system for the HSCCC bioreactor.
bioreactor at the ow rates of 4.0, 3.0, 2.0, and 1.0 mL minꢀ1
.
When the ow rate was 3.0 mL minꢀ1 (Fig. 5A), the Sf was 42%,
and the conversion rate signicantly declined within 3 h. It
remained stable at 60–65% during the 3–12 h time period.
When the ow rate continued to decrease to 2.5 mL minꢀ1
(Fig. 5B), the retention of the stationary phase was 51%, the
absorption curve showed signicant uctuations, and the
conversion rate of genistin was initially higher and then
decreased over time. The conversion rate of genistin began to
rapidly drop aer 8 h and was eventually reduced to 70% at 14 h.
When the ow rate was 2.0 mL minꢀ1 (Fig. 5C), the Sf was 62%,
and the conversion rate of genistin was initially higher and then
decreased over time. The conversion rate was over 90% within
24 h. When the ow rate was 1.0 mL minꢀ1 (Fig. 5D), the Sf was
75%. The absorption value increased in the time period of 1.6–
2.4 h, and the conversion rate for genistin reached 100%, fol-
lowed by a slow decline. This may be attributed to the fact that
during the initial introduction of the substrate, the reaction
time for the substrate is longer, allowing the substrate to be
fully converted in the HSCCC reactor. Within 24 h of the reac-
tion time, genistin was efficiently processed by b-glucosidase
with over a 99% conversion rate (Fig. 6).
3.3 Optimization of the HSCCC reactor parameters
Fig. 4A shows the Sf of the ethyl acetate/buffer solution solvent
system in four HSCCC elution modes (FWD-OUT, REV-IN, FWD-
IN, and REV-OUT). The results showed that the Sf was higher in
the FWD-OUT (64.5%) and REV-IN (64.6%) modes, which were
Based on the abovementioned results, the Sf of HSCCC
decreased as the ow rate increased. On the one hand, the
increase in the ow rate correlated with the increase in the
amount of the substrate entering the reactor. If the substrate
was completely converted, the product yield produced per unit
of time increased. On the other hand, a decrease in Sf decreased
b-glucosidase, resulting in a decreased amount of the enzyme
and an incomplete conversion of the substrate. Consequently,
in the operation of the HSCCC bioreactor, a certain balance
between the ow rate and Sf was required to achieve a suffi-
ciently high product concentration at the outlet. Based on this
study, a ow rate of 2.0 mL minꢀ1 was chosen as the optimal
ow rate, which provided a conversion rate of over 90% within
24 h of the reaction. Moreover, the purity of the product
appeared to gradually decrease as the reaction time was
extended. A possible explanation for this effect is that aer
genistin is hydrolyzed, the detached glucose remains in the
Fig. 4 The results of the optimized HSCCC bioreactor parameters. (A)
different separation modes. (B) Different rotated speed. Solvent
system: ethyl acetate/buffer solution (1 : 1, v/v). Rotation speed:
800 rpm. Flow rate: 2.0 mL minꢀ1. Temperature: 40 ꢁC.
4896 | RSC Adv., 2019, 9, 4892–4899
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Fig. 5 Effect of the HSCCC bioreactor at different flow rates. (A) 3.0 mL minꢀ1, (B) 2.5 mL minꢀ1, (C) 2.0 mL minꢀ1, and (D) 1.0 mL minꢀ1. Solvent
system: ethyl acetate/buffer solution (1 : 1, v/v). Enzyme: 750 mg Lꢀ1 of b-glucosidase. Substrate: 62.5 mg Lꢀ1 of genistin. Separation mode:
FWD-OUT. Rotation speed: 800 rpm. Detection wavelength: 254 nm. Biotransformation temperature: 40 ꢁC.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 4892–4899 | 4897

View Article Online
RSC Advances
Paper
Fig. 6 HPLC chromatograms of the genistin and genistein. (A) Mixture of genistin and genistein, (B) genistin, and (C) genistein. Experimental
conditions: Waters Symmetry C18 column (5 mm, 4.6 mm ꢂ 250 mm, i.d.), mobile phase: methanol/water (70 : 30, v/v), flow-rate: 1.0 mL minꢀ1
detection wavelength: 254 nm, and injection volume: 10 mL.
,
stationary phase, accumulates to high concentration, and
results in the by-product inhibition of b-glucosidase.
Acknowledgements
The authors gratefully acknowledge the nancial support
provided by the National Natural Science Foundation of China
(21506119), Shandong Province Major Scientic and Techno-
logical Innovation Project (2017CXGC1301, 2017CXGC1308),
Primary Research & Development Plan of Shandong Province
(2017GSF216002), and Natural Science Foundation of Shan-
dong Province (ZR2017BC061).
3.5 Identication of the product
ESI-MS m/z: 271.2415 [M + H]+. 1H-NMR (DMSO-d6, 400 MHz) d:
12.93 (1H, s, OH-5), 10.83 (1H, s, OH-7), 9.55 (1H, s, OH-40), 8.29
(1H, s, H-2), 7.35 (2H, d, J ¼ 8.5 Hz, H-20, 60), 6.80 (2H, d, J ¼
8.5 Hz, H-30, 50), 6.36 (1H, d, J ¼ 2.0 Hz, H-8), 6.20 (1H, d, J ¼
2.0 Hz, H-6). Aer comparing the data with the reported
values,20 the chemical structure was identied as genistein.
References
1 C. M. Bargues, M. Ingles, L. G. Mallench, J. S. Ros,
V. B. Costa, V. H. Perez, P. G. Tarraga, M. Dromant,
C. Borras and J. M. G. Verdugo, Clearing Amyloid-
4. Conclusions
In the present study, a novel HSCCC bioreactor was established
for the efficient biotransformation of genistein from genistin. A
suitable solvent system for the HSCCC bioreactor was chosen
according to the enzyme activity (i.e. KD values for the substrate
and product). The HSCCC bioreactor parameters were selected
as FWD-OUT and 800 rpm with a retention of the stationary
phase as the index. Using the optimal bioreactor parameters,
different ow rates for the substrate solution were investigated,
and 2.0 mL minꢀ1 was chosen as the optimal ow rate with
a conversion rate of over 90% within 24 h. The established
bioreactor was a novel type of immobilized enzyme reactor with
the advantages of simple operation as well as the ability to
immobilize and recycle the enzyme; moreover, the established
bioreactor can be applied in the preparation of genistein and
other biotransformations of natural products.
b
through PPARg/ApoE Activation by Genistein is
a Treatment of Experimental Alzheimer's Disease, Free
Radic. Biol. Med., 2017, 108, S44.
2 M. Imaisumida, S. Majid, P. Dasgupta, P. Kulkarni, S. Saini,
D. Bhagirath, T. Kato, S. Maekawa, Y. Hashimoto and
M. Shiina, Abstract 3449: genistein inhibits renal cancer
progression through long non-coding RNA HOTAIR
suppression, Cancer Res., 2017, 77, 3449.
¨
3 S. Jaskulski, A. Y. Jung, A. Rudolph, T. Johnson, K. Thone,
E. Herpel, P. Sinn and J. Chang-Claude, Genistein and
enterolactone in relation to Ki-67 expression and HER2
status in postmenopausal breast cancer patients, Mol. Nutr.
Food Res., 2017, 61(11), 1700449.
4 W. F. Li, K. Yang, P. Zhu, H. Q. Zhao, Y. H. Song, K. C. Liu
and W. F. Huang, Genistein Ameliorates Ischemia/
Reperfusion-Induced Renal Injury in a SIRT1-Dependent
Manner, Nutrients, 2017, 9(4), 403–423.
Conflicts of interest
5 T. R. Moore, R. B. Franks and C. Fox, Review of Efficacy of
Complementary and Alternative Medicine Treatments for
All authors declare no conicts of interest.
4898 | RSC Adv., 2019, 9, 4892–4899
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Menopausal Symptoms, J. Midwifery Women's Health, 2017, 14 S. T. Morthensen, A. S. Meyer, H. Jørgensen and M. Pinelo,
62(7), 286–297.
Signicance of membrane bioreactor design on the
biocatalytic performance of glucose oxidase and catalase:
free vs. immobilized enzyme systems, Biochem. Eng. J.,
2017, 17, 41–47.
6 M. I. Sumida, Y. Tanaka, R. Dahiya and S. Yamamura,
Abstract 4430: genistein regulates non-coding RNA
HOTAIR and epithelial-to-mesenchyme transition in renal
cancer cells, Cancer Res., 2018, 78, 4430.
7 X. Zhang, K. L. Cook, A. Warri, I. M. Cruz, M. Rosim,
J. Riskin, W. Helferich, D. Doerge, R. Clarke and
15 J. Xu, C. Liu, M. Wang, L. Shao, L. Deng, K. Nie and F. Wang,
Rotating packed bed reactor for enzymatic synthesis of
biodiesel, Bioresour. Technol., 2017, 224, 292–297.
L. Hilakivi-Clarke, Lifetime genistein intake increases the 16 T. Chen, H. Li, D. Zou, Y. Liu, C. Chen, G. Zhou and Y. Li,
response of mammary tumors to tamoxifen in rats, Clin.
Cancer Res., 2017, 23(3), 814–824.
Separation of three anthraquinone glycosides including
two isomers by preparative high-performance liquid
8 T. Izumi, M. K. Piskula, S. Osawa, A. Obata, K. Tobe, M. Saito,
S. Kataoka, Y. Kubota and M. Kikuchi, Soy isoavone
aglycones are absorbed faster and in higher amounts than
chromatography
chromatography from Rheum tanguticum Maxim, J. Sep.
Sci., 2016, 39(16), 3105–3112.
and
High-speed
counter-current
their glucosides in humans, J. Nutr., 2000, 130(7), 1695–1699. 17 T. Chen, P. Wang, N. Wang, C. Sun, X. Yang, H. Li, G. Zhou
9 M. K. Piskula, J. Yamakoshi and Y. Iwai, Daidzein and
genistein but not their glucosides are absorbed from the
rat stomach, FEBS Lett., 1999, 447(2–3), 287–291.
and Y. Li, Separation of three polar compounds from Rheum
tanguticum by high-speed countercurrent chromatography
with an ethyl acetate/glacial acetic acid/water system, J.
Sep. Sci., 2017, 225, 55–61.
10 C. H. Lee, L. Yang, J. Z. Xu, S. Y. V. Yeung, Y. Huang and
Z. Y. Chen, Relative antioxidant activity of soybean 18 Q. Gao, R. Ma, L. Chen, S. Shi, P. Cai, S. Zhang and H. Xiang,
isoavones and their glycosides, Food Chem., 2005, 90(4),
735–741.
11 W. D. Chiang, C. J. Shih and Y. H. Chu, Optimization of acid
Antioxidant proling of vine tea (Ampelopsis grossedentata):
off-line coupling heart-cutting HSCCC with HPLC-DAD-
QTOF-MS/MS, Food Chem., 2017, 225, 55–61.
hydrolysis conditions for total isoavones analysis in 19 H. Zhao, X. Hu, X. Chen, S. Shi, X. Jiang, X. Liang, W. Chen
soybean hypocotyl by using RSM, Food Chem., 2001, 72(4),
499–503.
12 S. J. Yeom, B. N. Kim, Y. S. Kim and D. K. Oh, Hydrolysis of
isoavone glycosides by a thermostable b-glucosidase from
Pyrococcus furiosus, J. Agric. Food Chem., 2012, 60(6), 1535–
1541.
and S. Zhang, Analysis and improved characterization of
minor antioxidants from leaves of Malus doumeri using
a combination of major constituents' knockout with high-
performance liquid chromatography–diode array detector–
quadrupole time-of-ight tandem mass spectrometry, J.
Chromatogr. A, 2015, 1398, 57–65.
13 D. Y. Zhang, Y. G. Zu, Y. J. Fu, M. Luo, W. Wang, C. B. Gu and 20 C. Lang'at Thoruwa, T. T. Song, J. Hu, A. L. Simons and
X. H. Yao, Application of immobilized enzymes to accelerate
the conversion of; genistin to genistein in pigeon pea root
extracts and the evaluation; their antioxidant activity, Ind.
Crops Prod., 2013, 42(3), 409–415.
P. A. Murphy, A Simple Synthesis of 7,4'-Dihydroxy-6-
methoxyisoavone, Glycitein, the Third Soybean
Isoavone, J. Nat. Prod., 2003, 66(1), 149–151.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 4892–4899 | 4899