Biosynthesis of glycyrrhetic acid 3-O-mono-β-d-glucuronide catalyzed by β-d-glucuronidase with enhanced bond selectivity in an ionic liquid/buffer biphasic system

Article abstract of DOI:10.1016/j.procbio.2010.03.015

The bond selective hydrolysis of glycyrrhizin (GL) to glycyrrhetic acid 3-O-mono-β-d-glucuronide (GAMG) catalyzed by recombinant β-d-glucuronidase from Escherichia coli BL21 (PGUS-E) was successfully performed in an ionic liquid (IL)/buffer biphasic system. Five ILs were analyzed, however, a hydrophobic IL 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM]PF₆) showed the best biocompatibility with PGUS-E. An obvious enhancement in the initial reaction rate, substrate conversion, GAMG yield and chemical bond selectivity (S_cb) was observed using 40% (v/v) [BMIM]PF₆/buffer as the reaction medium when compared to the acetate buffer medium. Under the optimized conditions (pH 6.0, temperature 50°C, substrate concentration 6mM and shaking speed 200rpm), the initial reaction rate, the GAMG yield and the S_cb reached 3.15mMh^-1, 74.36% and 98.12%, respectively. The recyclability of [BMIM]PF₆ was also studied and found to be reusable for five batches with high recovery percentage (≥92%). Furthermore, the desired product and byproduct were easily separated since they were distributed in different phases. Additionally, higher V_max (3.14 versus 2.24mMh^-1), lower apparent K_m (1.21 versus 1.80mM) and E_a (25.97 versus 32.60kJmol^-1) were achieved in [BMIM]PF₆/buffer biphasic system than that in monophasic buffer system.

Full text of DOI:10.1016/j.procbio.2010.03.015

Process Biochemistry 45 (2010) 1916–1922
Contents lists available at ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Biosynthesis of glycyrrhetic acid 3-O-mono-␤-d-glucuronide catalyzed by
␤-d-glucuronidase with enhanced bond selectivity in an ionic liquid/buffer
biphasic system
Dong-Mei Hea, Imdad Kaleemb, Si-Ying Qinb, Da-Zhang Daib,∗, Gui-Yan Liub, Chun Lia,b,∗
a School of Chemistry and Chemical Engineering, Shihezi University, Shihezi 832003, Xinjiang, PR China
b School of Life Science and Technology, Beijing Institute of Technology, Beijing 100081, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The bond selective hydrolysis of glycyrrhizin (GL) to glycyrrhetic acid 3-O-mono-␤-d-glucuronide
(GAMG) catalyzed by recombinant ␤-d-glucuronidase from Escherichia coli BL21 (PGUS-E) was suc-
cessfully performed in an ionic liquid (IL)/buffer biphasic system. Five ILs were analyzed, however, a
hydrophobic IL 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM]PF6) showed the best bio-
compatibility with PGUS-E. An obvious enhancement in the initial reaction rate, substrate conversion,
GAMG yield and chemical bond selectivity (Scb) was observed using 40% (v/v) [BMIM]PF6/buffer as the
reaction medium when compared to the acetate buffer medium. Under the optimized conditions (pH
6.0, temperature 50 ◦C, substrate concentration 6 mM and shaking speed 200 rpm), the initial reaction
rate, the GAMG yield and the Scb reached 3.15 mM h−1, 74.36% and 98.12%, respectively. The recyclability
of [BMIM]PF6 was also studied and found to be reusable for five batches with high recovery percent-
age (≥92%). Furthermore, the desired product and byproduct were easily separated since they were
distributed in different phases. Additionally, higher Vmax (3.14 versus 2.24 mM h−1), lower apparent Km
(1.21 versus 1.80 mM) and Ea (25.97 versus 32.60 kJ mol−1) were achieved in [BMIM]PF6/buffer biphasic
system than that in monophasic buffer system.
Received 22 November 2009
Received in revised form 17 February 2010
Accepted 12 March 2010
Keywords:
Ionic liquid
Bond selectivity
␤-d-Glucuronidase
Glycyrrhizin
Glycyrrhetic acid
3-O-mono-␤-d-glucuronide
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
required for the large scale production of GAMG in industry. Biosyn-
thesis of GAMG has several advantages over chemical methods
Glycyrrhizin (GL) is a major ingredient of liquorices root which
is one of the most widely used medicinal herbs in China [1]. As
a popular food additive, the sweetness of GL is 170-fold higher
than that of sucrose. GL is also well known for its physiological
functions, such as anti-inflammation, anti-tumor, anti-allergic and
anti-viral activities [2]. GL can be modified to glycyrrhetic acid
3-O-mono-␤-d-glucuronide (GAMG) by hydrolyzing the distal gly-
cosidic bond or glycyrrhetic acid (GA) by removal of two molecules
of glucuronide (Fig. 1). GAMG is an important derivative of GL and
displays stronger physiological functions as compared to GL, espe-
cially it has an evident anti-tumor effect and also its sweetness is
5-fold higher than that of GL [3]. So GAMG is expected to be a new
and better food additive and therapeutic agent having more com-
mercial potential than GL. Considering the great importance and
the expanding market of GAMG, the efficient methodologies are
due to its high specificity, efficiency, and also an eco-friendly sta-
tus. ␤-d-Glucuronidase (PGUS, EC 3.2.1.31), a glycosidase enzyme,
is widely used as a biocatalyst for hydrolyzing glycosidic bonds.
The ␤-d-glucuronidases produced by Streptococcus LJ-22 [4], Cryp-
tococcus magnus MG-27 [5] and mixed Eubacterium sp. GLH and
Ruminococcus sp. PO1–3 [6] are generally used for the production
of GAMG, but due to high cost value, low activity and selectivity,
are not conducive to industrial scale production.
Previously, our research team reported about the fungus Peni-
cillium purpurogenum Li-3 [7] that was involved in producing
specific ␤-d-glucuronidase with high chemical bond selectivity
which could directly hydrolyze GL to GAMG when it was grown in
a liquid culture medium containing GL, where GL was the inducer
as well as the best and sole carbon source. However, the amount of
enzyme produced was relatively low. Therefore, we cloned the gene
pgus (GenBank Accession No. EU095019) and overexpressed it in
Escherichia coli BL21 for the high production of ␤-d-glucuronidase
(PGUS-E) [8]. Although a large amount of PGUS-E was produced, but
still the chemical bond selectivity was much lower than that of the
corresponding wild strains. Hence, it is necessary to develop new
potential approaches to enhance the chemical bond selectivity.
∗
Corresponding authors at: School of Life Science and Technology, Beijing Insti-
tute of Technology, Beijing 100081, PR China. Tel.: +86 10 68913171;
fax: +86 10 68913171.
E-mail addresses: daidazhang@bit.edu.cn (D.-Z. Dai), lichun@bit.edu.cn (C. Li).
1359-5113/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2010.03.015

D.-M. He et al. / Process Biochemistry 45 (2010) 1916–1922
1917
Fig. 1. Hydrolysis of GL to GAMG catalyzed by PGUS-E in the IL/buffer systems.
The use of ionic liquids (ILs, salts consist of organic cations
2. Materials and methods
and inorganic anions and are liquid at temperatures below 100 ◦C)
in enzymatic reactions leads to remarkably improved thermal
stability, stereoselectivity, regioselectivity and inhibition of side
reactions [9–11], due to their advantageous properties includ-
ing the negligible vapor pressure, high thermal and chemical
stability, and also their ability to dissolve highly polar and
nonpolar compounds. Thus enzymes can play a tremendous cat-
alytic role in ILs as they can maintain their structure and active
site conformation, which have been reviewed by several groups
[12–14]. In recent years, many enzymatic transformations in ILs
or other systems containing ILs have been conducted success-
fully, where ILs have been used as pure solvents [15–18], as
co-solvents in the aqueous phase [19–21], or as biphasic sys-
tems together with other solvents [22,23,10,24] and even in
combination with supercritical carbon dioxide (scCO2) [25,26].
Cull et al. were first to use an IL in two-phase for the bio-
transformation of 1,3-dicyanobenzene to 3-cyanobenzamide and
3-cyanobenzoic acid catalyzed by nitrile hydratase [27]. The greater
advantage of the biphasic systems in ILs is the easy removal of
upper phase containing product and unreacted substrate, while
the lower phase having the enzyme and IL can be used again
in several cycles. Such systems help to meet the increasing
demand for clean technologies in industrial biochemical processes
[28]. The most frequently used ILs in biphasic system are 1-
butyl-3-methylimidazolium hexafluorophosphate ([BMIM][PF6])
and bis(trifluoromethanesulfonyl)imide ([BMIM][Tf2N]). The enzy-
matic processes in [BMIM][PF6]/alcohol [22], [BMIM][PF6]/buffer
[23,10], [BMIM][PF6]/hexane and [BMIM][PF6]/scCO2 [25,26] have
already been successfully designed and performed, which dis-
played excellent results. In such systems, enzymes including
lipases, proteases, esterases and dehydrogenases are mostly used
as well as the whole cells. So far only ␤-galactosidase from
Bacillus circulans [29] and ␤-glycosylhydrolase from Pyrococcus
furiosus [30] of glycosidases have been investigated in the pres-
ence of [MMIM][MeSO4]/water co-solvent system, but there are no
reports about the applicability of IL based biphasic system to ␤-d-
glucuronidase which is an important enzyme for the commercial
production of GAMG.
2.1. Chemicals and microorganism
Glycyrrhizin (GL) and glycyrrhetic acid (GA) were purchased from Sigma Chemi-
cal Co. (USA). Glycyrrhetic acid 3-O-mono-␤-d-glucuronide (GAMG) was generously
donated by Nanjing University of Technology (China). Methanol was of chromato-
graphic grade. All other chemicals were of analytical grade from commercial sources.
The microorganism used, recombinant E. coli BL21 expressing ␤-d-
glucuronidase from P. purpurogenum Li-3, was preserved by the lab of
biotransformation and microecology (Beijing Institute of Technology, China).
Cells used in the biotransformation experiments were inoculated into the
kanamycin-containing LB medium in 500 ml Erlenmeyer flasks, which was
operated at 170 rpm at 37 ◦C. After 3 h of inoculation at an OD600 of 0.6, the
isopropyl-␤-d-thiogalactopyranoside (IPTG) was added as an inducer. The incu-
bated cells were harvested by centrifugation at the end of the exponential growth
phase, washed twice with distilled water, separated from the aqueous medium
by centrifugation and resuspended in 50 mM acetate buffer (pH 5.0) for ultrasonic
fragmentation and then ammonium sulfate fractionation. Thus the crude enzyme
was obtained and stored at 4 ◦C for further experiments.
The PGUS-E activity was determined in the presence of 40 ␮l of 1.25 mM p-
nitrophenyl-␤-d-glucopyranoside (pNPG, Sigma, France) as a substrate in 50 mM
acetate buffer (pH 5.0), containing 10 ␮l crude enzyme. The reaction mixture was
kept at 40 ◦C for 10 min. The reaction was stopped by adding 200 ␮l of 0.4 M sodium
carbonate solution (pH 5.0) and the amount of p-nitrophenol (pNP) was measured
at 405 nm on an ELISA reader (Bio-Tek ELX 808, USA). One unit corresponds to the
amount of enzyme producing 1 ␮mol pNP per hour.
2.2. Synthesis and purification of ionic liquids
2.2.1. 1-Butyl-3-methylimidazolium bromide ([BMIM]Br)
A
mixture of N-methylimidazole (79.7 ml, 1.0 mol) and 1-bromobutane
(107.6 ml, 1.0 mol) was put in a three-neck round-bottom flask (500 ml) equipped
with a reflux condenser and kept it in a water containing ultrasonic cleaner (super-
sonic frequency 40 kHz, power 100 W), the temperature was adjusted between 35
and 40 ◦C for the working time period of 20 min and then waited for 5 min for the
completion and cool down of reaction mixture, this cycle was repeated again and
again until viscous oil/semi-solid was obtained which was purified by recystaliza-
tion from acetonitrile and then repeatedly recystallized from ethyl acetate to yield
a refined off-white solid (yield: 201 g, 92%).
2.2.2. 1-Butyl-3-methylimidazolium hexafluorophosphate ([BMIM]PF6)
Potassium hexafluorophosphate KPF6 (92 g, 0.5 mol) was added to a vigorously
stirred solution of [BMIM]Br (117 g, 0.5 mol) in H2O (500 ml) at room temperature.
The mixture was stirred for 24 h and then allowed to be stratified, the upper aqueous
phase was decanted. The lower ionic liquid phase was washed with distilled water
(8× 500 ml) until the residual bromine was free when determined by silver nitrate
test. The final product was dried in vacuum at 80 ◦C for 12 h. A colorless transparent
thick fluid [BMIM]PF6 (yield: 123.1 g, 87%) was obtained. 1H NMR (400 MHz, CDCl3,
25 ◦C), ı (ppm): 8.439 (s, 1H), 7.393 (s, 1H), 7.363 (s, 1H), 4.169 (t, 2H), 3.913 (s,
3H), 1.866 (m, 2H), 1.343 (m, 2H), 0.931 (t, 3H); 13C NMR (400 MHz, CDCl3, 25 ◦C), ı
(ppm): 135.491, 123.451, 122.201, 49.464, 35.760, 31.495, 19.017, 12.982.
Other four investigated ILs 1-butyl-3-methylimidazolium tetrafluorobo-
rate ([BMIM]BF4), 1-ethyl-3-methylimidazolium tetrafluoroborate ([EMIM]BF4),
N-butyl-pyridinium tetrafluoroborate ([BPy]BF4) and N-ethyl-pyridinium tetraflu-
oroborate ([EPy]BF4) were synthesized and characterized by the same method as
described above.
In this paper, we have attempted to resolve the problem that
the chemical bond selectivity of recombinant ␤-d-glucuronidase
was much lower than that of the corresponding wild strains. So
we have introduced the use of an IL in a biphasic system for the
target-oriented hydrolysis of GL to GAMG catalyzed by PGUS-E
(Fig. 1) to make full use of the advantages of biphasic reaction con-
ditions along with enhanced chemical bond selectivity. The effects
of several parameters such as buffer pH, reaction temperature, sub-
strate concentration and shaking speed on the hydrolysis of GL have
been investigated comprehensively. In addition, comparison of the
hydrolysis in an IL/buffer biphasic system and a monophasic buffer
system has also been evaluated.
2.3. Enzymatic hydrolysis
Like a typical analytical procedure, 2 ml of the aqueous acetate buffer (50 mM,
pH 5.0) monophasic system and also 2 ml of the [BMIM]PF6/buffer (50 mM, pH

1918
D.-M. He et al. / Process Biochemistry 45 (2010) 1916–1922
5.0) biphasic system both containing a fixed amount of GL were pre-incubated in
10 ml stopper flasks, and then added a fixed amount of PGUS-E to them to initi-
ate the biotransformation processes. These mixtures were shaken in a water-bath
shaker at various temperatures and shaking speeds under normal pressure. Aliquots
(20 ␮l) were withdrawn after specified time intervals from the aqueous phase of the
monophasic buffer system and also from the lower and upper phase of the IL/buffer
biphasic system. Then the samples were analyzed by HPLC. Details of the substrate
concentration, buffer pH, reaction temperature and shaking speed are specified for
each case. The substrate concentration throughout the paper refers to substrate con-
centration present in the buffer phase for IL/buffer biphasic system unless specified.
2.4. Determination of partition coefficients
At the end of each reaction, the biphasic system can be considered in a relatively
equilibrium state, so the partition coefficients (K) can be defined as the ratio of the
concentration of the product or the substrate between the lower IL phase and the
upper aqueous phase. The concentration of the product or the substrate in each
phase was determined by HPLC as described in Section 2.6.
2.5. Reuse of IL in repeated batch process
Fig. 2. Effect of contents of various ILs on the hydrolysis of GL to GAMG catalyzed
by PGUS-E in different IL/buffer systems with pure buffer as a control. The reactions
were performed in 2 ml acetate buffer (50 mM, pH 5.0) containing 4 mM GL, 18.6 U
PGUS-E and various amounts of ILs at 40 ◦C, 180 rpm.
After each cycle of reaction, the biphasic mixture was separated to an upper
aqueous and a lower IL phase. The upper aqueous phase containing the unreacted
substrate and GAMG was removed while the lower IL phase having the GA was
treated under vacuum and heated to vaporize the water present in the IL. Only IL
left behind due to its higher boiling point and was used again in consecutive new
reactions with fresh buffer, substrate and enzyme.
the enzyme, also the high viscosity of the reaction mixture caused
by high content of IL may limit the mass transfer of the substrate
and product to and from the active sites of the enzyme, and thus
may contribute to the falling of the initial reaction rate. Moreover,
the Scb observed in IL-containing system was generally higher than
that in pure buffer (Table 1), the existence of ILs during the biotrans-
formation process may be partially responsible for the directional
hydrolysis of GL to GAMG.
2.6. HPLC analysis
The concentrations of GL, GAMG and GA were determined by HPLC. The
chromatographic conditions were as follows: ODS column (Shim-pack, VP-ODS,
4.6 mm × 250 mm, Shimadzu Corporation, Kyoto, Japan); UV detector; detection
wavelength 254 nm; flow rate 1.0 ml min−1; mobile phase, water (pH 2.85 with 0.6%
(v/v) acetic acid) and methanol at 19:81 (v/v); injection volume 10 ␮l. The retention
times for GL, GAMG and GA were around 7.5, 15.9 and 22.7 min, respectively.
The initial reaction rate (V0), conversion (c), yield and chemical bond selectivity
(Scb) were determined as follows:
3.2. Optimization of the hydrolysis of GL to GAMG catalyzed by
PGUS-E in [BMIM]PF6/buffer biphasic system
(S0 − St )
V0
=
(1)
(2)
t
ꢀ
ꢁ
All five IL-containing systems were tested, but the system with
40% (v/v) [BMIM]PF6/buffer was found to be the most suitable
one, in which the hydrolysis of GL to GAMG catalyzed by PGUS-
E occurred efficiently. So it was of great importance to optimize the
reaction conditions to obtain further improved results with respect
to the initial reaction rate and the Scb. Therefore, a systematic study
was made to check the effects of key factors such as buffer pH, reac-
tion temperature, substrate concentration and shaking speed on
the reaction. The four reaction parameters had a noticeable effect
on the initial reaction rate but hardly affected the Scb except the
buffer pH showed significant impact on both systems (Fig. 3).
It is well known that buffer pH plays a key role in enzymatic
reactions in IL-containing medium, and variation of buffer pH alters
the ionic state of the substrate and the charging enzyme involved in
the reaction and also affects the local polarity of the enzyme’s active
site, thus leading to changes in activity and selectivity of enzyme
[20]. The same reason could be applied to enzymatic reactions in
IL/buffer biphasic systems. There was a significant effect of buffer
pH on the activity and Scb of hydrolysis of GL to GAMG catalyzed
St
c (%) = 1 −
× 100
× 100
S0
PGAMG
Yield (%) =
(3)
(4)
S0
PGAMG
Scb (%) =
× 100
)
(PGAMG + PGA
where S0 and St are concentrations of substrate at time 0 and time t, and PGAMG and
PGA are concentrations of GAMG and GA, respectively.
3. Results and discussion
3.1. Selection of suitable IL
Five ILs of imidazolium or pyridinium type, considered as
suitable solvents for biotransformation process according to the
previous research reports, were selected and used in the experi-
ments as shown in Fig. 2. The addition of imidazolium-based ILs into
aqueous buffer seemed to increase the biocompatibility of PGUS-E
when the content of ILs was 10% (v/v), because the initial reaction
rate was clearly higher than that observed in pyridinium-based
ILs/buffer systems. It may be due to the reason that pyridinium-
based ILs have residual pyridine which could have strong toxic
effects on the enzyme. Obviously, the hydrophobic IL [BMIM]PF6
supported the fastest reaction rate, the highest yield, substrate con-
version and Scb when its content increased up to 40% (v/v) (Table 1),
thus it was regarded as the most suitable IL and 40% (v/v) was
the optimal content for subsequent investigations. As expected, the
initial reaction rate was strongly dependent on IL content. The ini-
tial reaction rate dropped sharply when the content of other four
hydrophilic ILs was over 10% (v/v), which may be attributed to
the following reasons [31]: high content of IL causes higher ionic
strength in the reaction medium, which may partially inactivate
Table 1
Comparison of hydrolysis of GL to GAMG catalyzed by PGUS-E in six different media.
Media
V0 (mM h−1
)
Yield (%)
c (%)
Scb (%)
Pure buffer
1.49 0.07 37.25 0.57 46.01 0.96 80.96 1.71
[BMIM]PF6 (40%)a 2.90 0.11 71.71 0.88 72.62 1.64 98.76 2.22
[BMIM]BF4 (10%)a 1.67 0.14 39.58 0.79 41.97 0.48 94.30 2.54
[EMIM]BF4 (10%)a 2.16 0.13 52.52 1.03 54.23 1.11 96.84 2.13
[BPy]BF4 (10%)a
[EPy]BF4 (10%)a
0.90 0.11 22.56 0.63 23.39 0.36 96.45 2.80
0.74 0.09 18.42 0.45 18.67 0.49 98.66 3.01
The reactions were performed in 2 ml of acetate buffer (50 mM, pH 5.0) solutions
of different ILs or acetate buffer (50 mM, pH 5.0) containing 4 mM GL and 18.6 U
PGUS-E at 40 ◦C, 180 rpm.
a
The optimum content for this reaction.

D.-M. He et al. / Process Biochemistry 45 (2010) 1916–1922
1919
Fig. 3. Effects of buffer pH (a), reaction temperature (b), GL concentration (c) or shaking speed (d) on the hydrolysis of GL to GAMG catalyzed by PGUS-E in [BMIM]PF6/buffer
biphasic system and monophasic buffer system. The reactions were performed in 2 ml buffer (50 mM, various pHs) or 2 ml 40% (v/v) [BMIM]PF6/buffer, both containing 18.6 U
PGUS-E and different concentrations of GL at various temperatures and shaking speeds. Symbols: (᭹, ꢀ) the initial reaction rate; (ꢀ, ꢁ) the chemical bond selectivity (Scb);
full symbols refer to the [BMIM]PF6/buffer biphasic system; open symbols refer to the monophasic buffer system.
by PGUS-E in 40% (v/v) [BMIM]PF6/buffer biphasic system when
compared with pure buffer system (Fig. 3a). As can be observed,
variation in pH of buffer phase ranging from 3.6 to 9.0 altered the
ionic state of substrate and enzyme, thus leading to changes in the
initial reaction rate and Scb of the reaction. The initial reaction rate
increased with the buffer pH from 3.6 to 6.0, but further enhance-
ment of buffer pH resulted in a sharp decline of initial reaction rate.
It may be due to the reason that the increase in basic environment
makes the deactivation of enzyme and the instability of substrate
while the moderate to low acidic conditions (pH 3.6–6.0) activates
the enzyme and stabilizes the substrate and hence increases the
initial reaction rate [21]. The Scb value remained unchanged at
buffer pH in the range of 4.6–7.0 both in [BMIM]PF6/buffer bipha-
sic system and pure buffer system. However, lower Scb values
were observed at pH below 4.6 or above 7.0. It is possibly because
of an inactivation of the enzyme and non-enzymatic secondary
hydrolysis of GAMG caused by strong acidic conditions or slight
alkaline environments [21]. Considering the initial reaction rate
and Scb, pH 6.0 was found to be the optimum pH for the biphasic
system.
Temperature has a great effect on the activity, chemical bond
selectivity, and stability of the enzyme as well as on the equilib-
rium state of a reaction. An increase in reaction temperature clearly
boosted the initial reaction rate up to 50 ◦C, and further rise in tem-
perature led to a decline in initial reaction rate (Fig. 3b). It is due
to the reason that high temperature denatures the enzyme and
decreases its activity. As the temperature exceeded 50 ◦C, the Scb
also decreased and it could be explained by the acceleration of non-
enzymatic hydrolysis at higher temperature. In this case, 50 ◦C was
regarded as the optimum temperature for obtaining high initial
Further analysis of buffer pH on the reaction was performed
and the effect of [BMIM]PF6 content on the pH of buffer phase in
biphasic system was examined. The pH of buffer phase was slightly
increased from 5.0 to 7.0 within the [BMIM]PF6 content range from
0% to 70% (v/v), while no further increase in the pH of buffer phase
was observed when the content of [BMIM]PF6 was above 70% (v/v)
(Fig. 4). At 40% (v/v) of [BMIM]PF6 content, the pH of buffer phase
was 6.0, which was in accordance with the result obtained previ-
ously from Fig. 3a.
Fig. 4. Effect of [BMIM]PF6 content on the pH of the buffer phase.

1920
D.-M. He et al. / Process Biochemistry 45 (2010) 1916–1922
reaction rate and Scb in [BMIM]PF6/buffer biphasic system. While
the optimum temperature for PGUS-E in monophasic buffer was
45 ◦C. This study also determines the higher thermal stability (50
versus 45 ◦C) of PGUS-E in [BMIM]PF6/buffer biphasic system than
that in monophasic buffer system.
Thermodynamically, high substrate concentration increases the
reaction rate, however, too high substrate concentration leads
to the substrate inhibition to synthesize the product and thus
decreases the reaction rate [32]. Therefore, the effect of substrate
concentration on the reaction was examined and it was found that
the initial reaction rate significantly increased with the increas-
ing concentration of GL up to 6 mM and afterwards decreased
slightly (Fig. 3c). This might be due to the excessive substrate
concentration leading to an inhibitory effect on the enzyme and
mass transfer limitation of the substrate into the active centre of
enzyme [31]. Thus, the optimal concentration of GL was 6 mM in
[BMIM]PF6/buffer biphasic system, which was higher than that in
pure buffer system (4 mM), indicating that the solubility of GL in
buffer was increased to some extent and the reaction efficiency
was substantially enhanced with the participation of [BMIM]PF6.
Shaking speed affects the diffusion and partition of substrate
and product in reaction system and consequently influences the ini-
tial reaction rate and chemoselectivity, especially in IL-containing
systems, as high viscosity of IL limits the diffusion of substrate
and product to and from the active site of enzyme [33]. The effect
of shaking speed on the hydrolysis of GL to GAMG catalyzed by
PGUS-E in 40% (v/v) [BMIM]PF6/buffer biphasic system was inves-
tigated and it was found that the reaction accelerated rapidly with
the increase in shaking speed up to 200 rpm (Fig. 3d), suggesting
that mass transfer was the rate-limiting step. The optimal shaking
speed was found to be 200 rpm in both systems, above which no
significant variation in the initial rate was observed. Also, shaking
speed showed little influence on the chemical bond selectivity, with
changes in Scb value between 94% and 97%.
Fig. 5. Lineweaver–Burk plot for the determination of apparent kinetic parameters
of the hydrolysis of GL to GAMG catalyzed by PGUS-E in monophasic buffer system
and [BMIM]PF6/buffer biphasic system. The reactions were performed in 2 ml buffer
(50 mM, pH 6.0) or 2 ml 40% (v/v) [BMIM]PF6/buffer, both with different concentra-
tions of GL and 18.6 U PGUS-E at 50 ◦C, 200 rpm. Symbols: (ꢀ) the monophasic buffer
system; (᭹) the [BMIM]PF6/buffer biphasic system.
Table 2
Comparison of apparent kinetic parameters and activation energy for the hydrolysis
of GL to GAMG catalyzed by PGUS-E in [BMIM]PF6/buffer biphasic system and pure
buffer system.
b
b
c
Media
Km (mM)
Vmax
(mM h−1
Vmax/Km (h−1 Ea (kJ mol−1
) )
)
Pure buffer
1.80 0.03 2.24 0.11 1.24 0.07
32.60 1.09
25.97 1.27
[BMIM]PF6 (40%)a 1.21 0.05 3.14 0.14 2.60 0.13
When the reaction was carried out under the determined opti-
mum conditions described above and at normal pressure, the initial
reaction rate, the GAMG yield and the Scb were 3.15 mM h−1, 74.36%
and 98.12%, respectively. These values were much higher than that
in aqueous monophasic system, due to the excellent biocompat-
ibility of [BMIM]PF6 with the PGUS-E and moderately increasing
solubility of GL as well as the possible specificity of PGUS-E caused
by [BMIM]PF6.
The reactions were performed under the same conditions as Figs. 5 and 6.
a
The optimum content for this reaction.
The apparent Km and Vmax values were determined by Lineweaver–Burk plot.
The apparent Ea values were calculated by Arrhenius plot.
b
c
for the increased activity and specificity as exhibited by PGUS-E in
40% (v/v) [BMIM]PF6/buffer biphasic system when compared with
pure buffer system. Moreover, an increase in initial reaction rate
might be the result of a decline in activation energy (Ea) of the enzy-
matic reaction. So the Ea of PGUS-E-catalyzed hydrolysis of GL to
GAMG was determined by liner regression analysis of the Arrhenius
plot (Fig. 6). As can be seen in Table 2, the Ea value of the PGUS-
E-catalyzed hydrolysis performed in 40% (v/v) [BMIM]PF6/buffer
biphasic system was much lower than that achieved in pure buffer
system, suggesting that [BMIM]PF6 could make the formation of
the “enzyme–substrate complex” easier and more stable due to its
ionic nature [36].
3.3. Kinetic parameters and activation energy
To get a better insight into the influence of the IL [BMIM]PF6
on the hydrolysis of GL to GAMG catalyzed by PGUS-E in the
biphasic system, a comparative study of the apparent kinetic
parameters (Km, Vmax) of the reaction was made between 40% (v/v)
[BMIM]PF6/buffer biphasic system and monophasic buffer system.
In the two media the hydrolysis was regarded as a monosubstrate
reaction since water was in excess amount. The Lineweaver–Burk
plot (1/S ∼ 1/V) suggested that the hydrolysis of GL to GAMG cat-
alyzed by PGUS-E follows the Michaelis–Menten equation (Fig. 5).
The values of the apparent Km and Vmax are presented in Table 2.
As expected, the higher Vmax and the lower Km were observed in
40% (v/v) [BMIM]PF6/buffer biphasic system, where the ratio of
Vmax/Km was 2-fold higher than that observed in pure buffer. De
Diego [34] and Micaêlo [35] described that hydrophilic solvents
strip essential water from the enzyme, which leads to the unfold-
ing of the molecule with the exposure of the inner hydrophobic
residues, while hydrophobic solvents keep enzyme flexible and
in active conformation thereby an enhancement in the affinity
of enzyme towards the substrate. Therefore, it appears that the
hydrophobic IL [BMIM]PF6 causes an enhancement in the affin-
ity of the PGUS-E towards the substrate, which partially accounts
Table 3
Partition coefficients of substrate and products in [BMIM]PF6/buffer biphasic system
at different temperatures.
T (◦C)
KGL
KGAMG
KGA
55.290
30
40
50
60
0.017
0.020
0.010
0.017
0.006
0.010
0.004
0.003
86.706
98.931
100.801
The reactions were performed in 2 ml buffer (50 mM, pH 6.0) containing 40% (v/v)
[BMIM]PF6, 6 mM GL and 18.6 U PGUS-E, shaking at 200 rpm and at various temper-
atures.

D.-M. He et al. / Process Biochemistry 45 (2010) 1916–1922
1921
Table 4
Recovery of [BMIM]PF6 and the effect of using the recycled [BMIM]PF6 on the hydrolysis of GL to GAMG catalyzed by PGUS-E in [BMIM]PF6/buffer biphasic system in each
batch.
Batch
1
2
3
4
5
Recovery percentage (%)
GAMG yield (%)
Scb (%)
100
74.4 1.05
98 3.19
95 3.12
71.6 2.44
95 2.18
96 3.07
73.1 1.89
96 2.94
94 2.83
72.7 1.57
97 2.68
92 2.46
73.9 2.14
95 3.23
The reactions were performed in 2 ml buffer (50 mM, pH 6.0) containing 40% (v/v) [BMIM]PF6, 6 mM GL and 18.6 U PGUS-E at 50 ◦C, 200 rpm. In the later cycles the upper
buffer phase was removed and the lower IL phase was treated at 80 ◦C in vacuum and reused again in new reactions with fresh buffer, substrate and enzyme.
3.4. Distribution of product and substrate in the biphasic system
showed that [BMIM]PF6 could be recycled at very high recovery
efficiency (over 92%) and the recycled [BMIM]PF6 could be reused
for the hydrolysis of GL. Nearly the same yields and Scb were
obtained in the biphasic system with recycled [BMIM]PF6. Thus
the operating cost of the hydrolysis of GL catalyzed by PGUS-E
in [BMIM]PF6/buffer biphasic system can be reduced due to the
reusability of [BMIM]PF6.
Since the reaction mixture was a biphasic system, the distri-
bution of product and substrate between the two phases were
thoroughly investigated, and the results are summarized in Table 3.
The solubility of the desired product GAMG and unreacted sub-
strate GL in [BMIM]PF6 was very low at the temperatures tested,
which can be seen from the low KGAMG and KGL. Nearly all molecules
of byproduct (GA) of the hydrolysis were found to be distributed
in the IL phase and were barely detectable in buffer phase. There
would be only GAMG in the buffer phase if GL was fully hydrolyzed.
So it is the best and easy method for the separation of desired
product GAMG and byproduct GA which were automatically got
separated. Glucuronide, a hydrophilic group presented in both
GAMG and GL, may be a hindrance for GAMG or GL entering into the
hydrophobic IL [BMIM]PF6. While GA is hydrophobic in nature as it
has no glucuronide group, so it can be easily dissolved in [BMIM]PF6
since the values of KGA were so high.
4. Conclusions
Briefly, this work reports for the first time that ␤-d-
glucuronidase catalyzes the biotransformation of GL to GAMG in
the [BMIM]PF6/buffer biphasic system. The initial reaction rate and
chemical bond selectivity were much higher than those observed
in monophasic buffer system. The excellent yield and fast initial
reaction rate were obtained at optimum conditions. The separa-
tion of product is much easier since the desired product (GAMG)
was distributed in the upper buffer phase while the byproduct (GA)
was found in the lower IL phase. Importantly, the hydrophobic
IL [BMIM]PF6 can be used repeatedly, which determines its eco-
nomical and eco-friendly status for future industrial applications.
All these advantages make the biphasic system more favorable
than monophasic system for the commercial biosynthesis of many
important products.
3.5. Recovery of IL
One of the most important properties of ILs is their reusabil-
ity, which determines their eco-friendly status for future industrial
applications. There are two methods to investigate this key fac-
tor: recycling the IL itself or recycling the IL together with enzyme
[22]. In this research work, recycling the IL was investigated in an
IL/buffer biphasic system. Table 4 illustrates the recovery percent-
age of [BMIM]PF6 and the effect of using the recycled [BMIM]PF6
on the hydrolysis of GL to GAMG catalyzed by PGUS-E. Results
Acknowledgements
This research was financially supported by National Natural
Science Foundation of China (20776017 and 20976014), Natu-
ral Science Foundation of Beijing (5072028), National High-Tech
Project (2009AA02Z202) and National Basic Research Program
(2007CB714300), PR China.
References
[1] Cinatl J, Morgenstern B, Bauer G, Chandra P, Rabenau H, Doerr H. Glycyrrhizin,
an active component of liquorice roots, and replication of SARS-associated
coronavirus. Lancet 2003;361(9374):2045–6.
[2] Mizutani K, Kambara T, Masuda H, Tamura Y. Glycyrrhetic acid monoglu-
curonide (MGGR): biological activities. Int Congress Ser 1998;1157:225–35.
[3] Mizutani K, Kuramoto T, Tamura Y, Ohtake N, Shiqeki D, Nakura M, et al. Sweet-
ness of glycyrrhetic acid 3-O-␤-d-monoglucuronide and the related glycosides.
Biosci Biotechnol Biochem 1994;58(3):554–5.
[4] Kim D-H, Lee K, Han K. Biotransformation of glycyrrhetinic acid-3-O-␤-d-
glucuronide by Streptococcus LJ-22, a human intestinal bacterium. Biol Pharm
Bull 1999;22(3):320–2.
[5] Kuramoto T, Ito Y, Oda M, Tamura Y, Kitahara S. Microbial production of gly-
cyrrhetic acid 3-O-mono-␤-d-glucuronide from glycyrrhizin by Cryptococcus
magnus Mg-27. Biosci Biotechnol Biochem 1994;58(3):455–8.
[6] Akao T. Competition in the metabolism of glycyrrhizin with glycyrrhetic acid
mono-glucuronide by mixed Eubacterium sp. GLH and Ruminococcus sp. PO1–3.
Biol Pharm Bull 2000;23(2):149–54.
[7] Feng SJ, Li C, Xu XL, Wang XY. Screening strains for directed biosynthesis of
␤-d-mono-glucuronide–glycyrrhizin and kinetics of enzyme production. J Mol
Catal B: Enzyme 2006;43:63–7.
[8] Song ZK, Wang XY, Chen GQ, Li C. Clonging and prokaryotic expression of
␤-glucuronidase from Penicillium purpurogenum Li-3. J Chem Ind Eng (China)
2008;59(12):3101–6.
[9] Matsuda T, Yamagishi Y, Koguchi S, Iwai N, Kitazume T. An effective method
to use ionic liquids as reaction media for asymmetric reduction by Geotrichum
candidum. Tetrahedron Lett 2006;47(27):4619–22.
Fig. 6. Arrhenius plot for the calculation of apparent activation energy of the
hydrolysis of GL to GAMG catalyzed by PGUS-E in monophasic buffer system and
[BMIM]PF6/buffer biphasic system. The reactions were performed in 2 ml buffer
(50 mM, pH 6.0) or 2 ml 40% (v/v) [BMIM]PF6/buffer, both with 4 mM GL and 18.6 U
PGUS-E at various temperatures, 200 rpm. Symbols: (ꢀ) the monophasic buffer
system; (᭹) the [BMIM]PF6/buffer biphasic system.

1922
D.-M. He et al. / Process Biochemistry 45 (2010) 1916–1922
[10] Singh M, Singh RS, Banerjee UC. Stereoselective synthesis of (R)-1-chloro-3(3,4-
difluorophenoxy)-2-propanol using lipases from Pseudomonas aeruginosa in
ionic liquid-containing system. J Mol Catal B: Enzyme 2009;56(4):294–9.
[11] Kahveci D, Guo Z, Özc¸ elik B, Xu X. Lipase-catalyzed glycerolysis in
ionic liquids directed towards diglyceride synthesis. Process Biochem
2007;42(12):1649–54.
[24] Eckstein M, Filho MV, Liese A, Kragl U. Use of an ionic liquid in a two-phase sys-
tem to improve an alcohol dehydrogenase catalysed reduction. Chem Commun
2004;4(9):1084–5.
[25] Reetz MT, Wiesenhöfer W, Franciò G, Leitner W. Biocatalysis in ionic liquids:
batchwise and continuous flow processes using supercritical carbon dioxide as
the mobile phase. Chem Commun 2002;2(9):992–3.
[12] Kragl U, Eckstein M, Kaftzik N. Enzyme catalysis in ionic liquids. Curr Opin
Biotechnol 2002;13:565–71.
[13] van Rantwijk F, Lau RM, Sheldon RA. Biocatalytic transformations in ionic liq-
uids. Trends Biotechnol 2003;21:131–8.
[26] Lozano P, De Diego T, Sauer T, Vaultier M, Gmouh S, Iborra J. On the importance
of the supporting material for activity of immobilized Candida antarctica lipase
B in ionic liquid/hexane and ionic liquid/supercritical carbon dioxide biphasic
media. J Supercrit Fluids 2007;40:93–100.
[14] van Rantwijk F, Sheldon RA. Biocatalysis in ionic liquids. Chem Rev
2007;107:2757–85.
[15] Weuster-Botz D. Process intensification of whole-cell biocatalysis with ionic
liquids. Chem Rec 2007;7:334–40.
[27] Cull SG, Holbrey JD, Vargas-Mora V, Seddon KR, Lye GJ. Room-temperature
ionic liquids as replacements for organic solvents in multiphase bioprocess
operations. Biotechnol Bioeng 2000;69(2):227–33.
[28] DeSimone JM. Practical approaches to green solvents. Science
2002;297:799–803.
[16] Ha SH, Lee SH, Dang DT, Kwon MS, Chang WJ, Yu YJ, et al. Enhanced sta-
bility of Candida antarctica lipase
2008;25(2):291–4.
[17] Adamczak M, Bornscheuer UT. Improving ascorbyl oleate synthesis catalyzed
by Candida antarctica lipase B in ionic liquids and water activity control by salt
hydrates. Process Biochem 2009;44(3):257–61.
B
in ionic liquids. Korean
J
Chem Eng
[29] Kaftzik N, Wasserscheid P, Kragl U. Use of ionic liquids to increase yield
and enzyme stability in the ␤-galactosidase catalysed synthesis of N-
acetyllactosamine. Org Proc Res Dev 2002;6(4):553–7.
[30] Lang M, Kamrat T, Nidetzky B. Influence of ionic liquid cosolvent on trans-
galactosylation reactions catalyzed by thermostable ␤-glycosylhydrolase
CelB from Pyrococcus furiosus. Biotechnol Bioeng 2006;95(6):1093–
100.
[31] Lou WY, Zong MH, Liu YY, Wang JF. Efficient enantioselective hydrolysis of d,l-
phenylglycine methyl ester catalyzed by immobilized Candida Antarctica lipase
B in ionic liquid containing systems. J Biotechnol 2006;125:64–74.
[32] Lou WY, Zong MH, Wu H. Enhanced activity, enantioselectivity and stability of
papain in asymmetric hydrolysis of d,l-p-hydroxyphenylglycine methyl ester
with ionic liquid. Biocatal Biotrans 2004;22(3):171–6.
[18] Katsoura MH, Polydera AC, Katapodis P, Kolisis FN, Stamatis H. Effect of
different reaction parameters on the lipase-catalyzed selective acylation
of polyhydroxylated natural compounds in ionic liquids. Process Biochem
2007;42(9):1326–34.
[19] Lou WY, Wang W, Zong MH. Efficient enantioselective reduction of 4ꢁ-
methoxyacetophenone with immobilized Rhodotorula sp. AS2. 2241 cells
in
a hydrophilic ionic liquid-containing co-solvent system. J Biotechnol
2009;143(3):190–7.
[20] Wang W, Zong MH, Lou WY. Use of an ionic liquid to improve asymmetric
reduction of 4ꢁ-methoxyacetophenone catalyzed by immobilized Rhodotorula
sp AS2.2241 cells. J Mol Catal B: Enzyme 2009;56:70–6.
[33] Lou WY, Zong MH, Wu H, Xu R, Wang JF. Markedly improving lipase-mediated
asymmetric ammonolysis of d,l-p-hydroxyphenylglycine methyl ester by
using an ionic liquid as the reaction medium. Green Chem 2005;7:500–6.
[34] De Diego T, Lozano P, Gmouh S, Vaultier M, Iborra JL. Understanding
structure–stability relationships on Candida antarctica lipase B in ionic liquids.
Biomacromolecules 2005;6:1457–64.
[35] Micaêlo NM, Soares CM. Protein structure and dynamics in ionic liq-
uids. Insights from molecular dynamics simulation studies. J Phys Chem B
2008;112:2566–72.
[21] He JY, Zhou LM, Wang P, Zu L. Microbial reduction of ethyl acetoacetate
to ethyl(R)-3-hydroxybutyrate in an ionic liquid containing system. Process
Biochem 2009;44:316–21.
ˇ
[22] Fehér E, Illeová V, Kelemen-Horváth I, Bélafi-Bakó K, PolakoviC M, Gubicza L.
Enzymatic production of isoamyl acetate in an ionic liquid–alcohol biphasic
system. J Mol Catal B: Enzyme 2008;50(1):28–32.
[23] Bräutigam S, Bringer-Meyer S, Weuster-Botz D. Asymmetric whole cell bio-
transformations in biphasic ionic liquid/water-systems by use of recombinant
Escherichia coli with intracellular cofactor regeneration. Tetrahedron: Asym-
metry 2007;18:1883–7.
[36] Lou WY, Zong MH, Zhang YY, Wu H. Efficient synthesis of optically active
organosilyl alcohol via asymmetric reduction of acyl silane with immobilized
yeast. Enzyme Microb Technol 2004;35:190–6.