Free and immobilized Aspergillus niger epoxide hydrolase-catalyzed hydrolytic kinetic resolution of racemic p-chlorostyrene oxide in a neat organic solvent medium

Article abstract of DOI:10.1016/j.procbio.2009.09.009

An enantio- and regioselective hydrolytic kinetic resolution (HKR) of racemic p-chlorostyrene oxide (rac-pCSO) was achieved by epoxide hydrolase (EH) from recombinant Aspergillus niger in a selected neat organic solvent medium. The solid free EH was reuse

Full text of DOI:10.1016/j.procbio.2009.09.009

Process Biochemistry 45 (2010) 210–216
Contents lists available at ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Free and immobilized Aspergillus niger epoxide hydrolase-catalyzed
hydrolytic kinetic resolution of racemic p-chlorostyrene oxide in a neat
organic solvent medium
b
Salwa Karboune a, , Alain Archelas , Jacques C. Baratti
*
c,**
a Department of Food Science and Agricultural Chemistry, McGill University, 21,111 Lakeshore, Ste-Anne de Bellevue, Quebec, Canada H9X 3V9
b Universite´ Paul Ce´zanne, ISM2 UMR CNRS 6263, Faculte´ Sciences et Techniques de St Je´rome, Avenue Escadrille Normandie Niemen, 13397 Marseille Cedex 20, France
c Universite´ de la Me´diterrane´e, Groupe Biocatalyse et Chimie Fine, CNRS UMR 6098, Faculte´ des Sciences de Luminy, 163 avenue de Luminy, 13288 Marseille Cedex 9, France
A R T I C L E I N F O
A B S T R A C T
Article history:
An enantio- and regioselective hydrolytic kinetic resolution (HKR) of racemic p-chlorostyrene oxide (rac-
pCSO) was achieved by epoxide hydrolase (EH) from recombinant Aspergillus niger in a selected neat
organic solvent medium. The solid free EH was reused four times in repeated-batch reactors; however,
the relative activity as well as the enantiomeric ratio (E-value) of this EH decreased from 100 to 20% and
from 68 to 23, respectively. In order to overcome the diffusion hindrance, due to the accumulation of the
hydrophilic diol in the enzyme micro-environment, and thereby to improve the operational stability of
EH after recycling, strategies consisting of the immobilization of EH and the use of a binary organic
solvent as the reaction medium were successfully applied. Although the highest protein immobilization
yield (82%) and retention of EH activity (142%) in heptane were obtained upon the immobilization of EH
on Accurel EP, the E-value and the operational stability of the resulted EH immobilizate after recycling
were reduced as compared to the free EH. In contrast, the nonporous DEAE-cellulose improved the
operational stability of EH by more than twofold. On the other hand, both the HKR efficiency and the
operational stability of A. niger EH were found to be modest to excellent in various binary organic solvent
mixtures of heptane and dioxane, depending on their ratio resulting in different Log P.
Received 8 March 2009
Received in revised form 5 June 2009
Accepted 9 September 2009
Keywords:
Epoxide hydrolase
Aspergillus niger
Organic solvent
Immobilization
Resolution
Operational stability
ß 2009 Elsevier Ltd. All rights reserved.
1. Introduction
the chemical catalysts. Over the last years, a number of microbial
EHs with enantio- and regio-complementary characteristics have
Because of both the high demand for ‘‘environmentally clean’’
chemical processes and the critical need for optically pure active
molecules, the application of enzymes to optical resolutions of
racemates is an increasingly growing field in both organic
syntheses and biocatalysis [1,2]. For instance, the hydrolytic
kinetic resolution of racemic epoxides by epoxide hydrolases (EHs)
(EC 3.3.2.3) has been widely recognized as a valuable approach for
the preparation of enantiopure epoxides and vicinal diols [3–5].
These chiral compounds are essential building blocks for the
preparation of a wide range of fine chemicals and pharmaceutical
molecules [5,6]. In addition to the ability of EHs to discriminate
efficiently both enantiomers as a result of their high enantios-
electivity, they offer many advantages such as no requirement for
cofactors, a high catalytic efficiency, operations under mild
conditions and fewer undesirable side reactions as compared to
been discovered and studied [5,7–9]. In addition, the properties of
some of these biocatalysts have been improved by site-directed
mutagenesis and directed evolution [10–12]. In our previous
studies, the efficiency of enantioselective EH from recombinant
Aspergillus niger has been amply demonstrated throughout several
preparative-scale hydrolytic kinetic resolutions carried out at very
high substrate concentration (200–500 g/L), affording several
enantiopure epoxides and their corresponding diols in a nearly
quantitative yield and good to excellent enantiomeric excess [13–
15]. Unfortunately for commercial applications, most of these
hydrolytic kinetic resolutions are still limited by the low long-term
operational stability of A. niger EH, the high-cost associated with
their single use and the low productivity due to the low solubility
of hydrophobic epoxides in aqueous media. As a consequence,
there has been increasing interest in the immobilization of EHs for
enhancing their thermal stability and allowing their reusability in
batch and packed-bed reactors. EH from A. niger has been
immobilized by adsorption [16] and covalent binding [17,18].
During the last two decades, various solvent systems have been
used to perform biocatalysis in non-conventional media and it has
been reported that enzymes can maintain its active conformation
* Corresponding author. Tel.: +1 514 398 8666; fax: +1 514 398 7977.
** Corresponding author.
E-mail addresses: salwa.karboune@mcgill.ca (S. Karboune),
jbaratti@luminy.univ-mrs.fr (J.C. Baratti).
1359-5113/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2009.09.009

S. Karboune et al. / Process Biochemistry 45 (2010) 210–216
2.3. EH immobilization
211
in these nonaqueous media when the essential water layer in their
micro-environment is not stripped off by the organic solvent [19].
Despite the fact that the catalytic activity is often reduced in neat
organic solvent media as compared to that obtained in the aqueous
medium, their utilization as reaction media has been demon-
strated to be a useful approach for improving enzyme properties,
increasing the solubility of hydrophobic substrates and ensuring
ease of both product and enzyme recovery [19–21]. Additional
benefit of the use of neat organic solvent media includes the
possibility to reuse the biocatalyst, which minimizes downstream
processing costs [21]. Contrary to other hydrolases, the catalytic
and selectivity properties and the operational stability of EH have
been little-studied in non-conventional media. As far as the
authors are aware, our previous study [22] was the first reported
on the catalytic properties of EH from A. niger in a neat organic
solvent. This interesting report prompted us to investigate the
hydrolytic kinetic resolution of racemic p-chlorostyrene oxide
(rac-pCSO), in neat organic solvent media, by A. niger EH in free and
immobilized forms and to evaluate their operational stability
through its reuse in many consecutives batch reactors.
The immobilization of A. niger EH on DEAE-cellulose and Accurel EP 100 supports
was carried out at 4 8C using a protein loading of 10 mg/g wet support, according to
Karboune et al. [16] and Holmquist et al. [26], respectively. Prior immobilization,
DEAE-cellulose support was washed twice with Tris–HCl buffer (10 mM, pH 7.1) at
a ratio of 1:100 (w:v), whereas Accurel EP 100 support was suspended in ethanol at
a ratio of 1:20 (w:v). The immobilization mixtures were incubated for 45 min under
mild agitation at 4 8C. The support, containing the adsorbed enzyme, was
subsequently recovered by centrifugation (1000 rpm, 10 min), washed twice with
1.5 mL of the Tris–HCl buffer (10 mM, pH 7.1) to remove unbound proteins and
stored wet at 4 8C. Protein immobilization yield (%) was defined as the ratio of
protein, immobilized onto a support (mg), divided by the initial protein content
(mg) and multiplied by 100. The retention of enzyme activity (%) was defined as the
specific activity of the immobilized EH, divided by the specific activity of free EH
and multiplied by 100.
2.4. Repeated-use of EH
The operational stability of the free and immobilized EH was investigated by
performing a series of consecutive hydrolytic kinetic resolution in successive batch
reactors, using the above-defined conditions. When the reaction reached 50%
conversion, the immobilized EH was removed from the reaction medium by
centrifugation (1500 rpm, 10 min) and rinsed with 2 mL heptane before adding
30 mL of a fresh substrate solution. The initial enzymatic activity was measured as
described above.
2. Materials and methods
2.1. Preparation of the pre-purified EH enzymatic extract
3. Results and discussion
To produce now commercially available EH (Fluka, ref no. 71832), the
recombinant A. niger [23,24] overexpressed into another A. niger fungus was
cultivated in a simple medium composed of glucose (10 g/L) and corn steep liquor
(20 g/L) and harvested at the exponential phase. The mycelia were recovered by
vacuum filtration and re-suspended in 1.4 L of 10 mM Tris–HCl buffer, pH 7.1,
containing 1 mM EDTA, 1 mM cysteine and 0.3 mM phenylmethanesulfonyl
fluoride (buffer A). The suspension was homogenized using a mechanical grinder
and the cells were disrupted by pressure shock (27 kpsi, Cell Disrupter from
Constant System Ltd., Daventry Northants, United Kingdom) to liberate the
intracellular EH activity. The supernatant, containing the EH, was microfiltrated
3.1. Hydrolytic kinetic resolution of rac-pCSO in a selected neat
organic solvent
Our previous study [22] has shown no correlation between the
enzymatic activity or enantiomeric ratio (E) of EH from A. niger and
the Log P of solvent. However, heptane was identified to be the
most appropriate nonaqueous solvent for EH from A. niger, since it
led to the highest enzymatic activity and apparent enantioselec-
tivity. In addition of being reactant of EH-catalyzed hydrolysis of
epoxides, water controls the flexibility of the active conformation
of biocatalyst in organic media. Controlling the initial aw to 0.9 has
been revealed to be of fundamental importance for EH from A. niger
in order to exhibit its maximal hydrolytic activity and enantios-
electivity [22]. Contrary to E-value, the EH activity was also
dependent on the pH memory of the lyophilized enzymatic extract
through a membrane with a cut-off of 0.1
resulting microfiltrate, containing the soluble components, was concentrated by
tangential flow ultrafiltration through membrane with cut-off of 40 kDa
mm (Inceltech, PSFH, France). The
a
a
(Inceltech, 40UFIB/1/S2, France). The concentrated enzyme solution was then
loaded onto a DEAE-Sepharose column (2.5 cm ꢀ 30 cm, Pharmacia), previously
equilibrated with buffer A, and the unbound proteins were removed by washing
with 500 mL of buffer A. The adsorbed proteins were subsequently eluted with a
linear gradient of KCl ranging from 0 to 0.35 M in a total volume of 924 mL buffer A
at a flow rate of 3 mL/min. EH activity in the eluted fractions was determined using
the standard procedure described below. The active fractions were pooled, dialyzed
against water and lyophilized. All procedures were carried out at 4 8C. The degree of
purity of the resulted extract, which was calculated as the specific activity of the
partially purified EH extract over that of the pure enzyme, was estimated to be
around 27% (unpublished results).
2.2. General procedure for enzymatic hydrolysis of racemic pCSO in batch reactor
A batch reactor with a total volume of 30 mL was run at 27 8C using a rac-pCSO
as substrate at a concentration of 4 mM in heptane equilibrated at a aw value of 0.9.
The rac-pCSO epoxide substrate was prepared by epoxidation of the correspond-
ing olefin using m-chloroperoxybenzoic acid. The reaction was started by the
addition of 60 mg of solid EH equilibrated at a aw value of 0.9 under a constant
agitation of 750 rpm for 45–90 min. Control trials without the enzyme were
carried out in tandem with the enzymatic reactions to monitor for potential
chemical side reactions. At different time intervals, aliquots of 100
mL were
withdrawn and mixed with 150 L isooctane containing meta-bromoacetophe-
m
none as an internal standard. The mixtures were centrifuged (1000 rpm, 10 min,
room temperature) and the resulting clear supernatants were subjected to chiral
GC analysis (col. CP-Chirasil-Dex CB, 125 8C, (R) 12.4 and (S) 13.0 min) to quantify
the concentration of the remaining pCSO and its enantiomeric excess (ee). The
formed diol was extracted with 2 mL ethyl acetate after saturation of the aqueous
phase with NaCl, the quantification of its enantiomeric excess was performed after
derivatization into its acetonide which was analyzed by chiral GC (col. CP-Chirasil-
Dex CB, 140 8C, (R) 13.8 min and (S) 15.0 min). All the enzymatic reactions and
respective controls were run in triplicate. The enzymatic activity was calculated
from the slope, obtained by plotting the concentration of consumed epoxide
versus reaction time. One EH unit (U) was defined as the amount of biocatalyst
Fig. 1. Kinetic profile of the resolution of rac-pCSO catalyzed by the Aspergillus niger
EH in the selected neat organic solvent medium: enantiomeric excess (ee) of
remaining epoxide (*); ee of formed diol (*); conversion ratio (&); remaining (R)
(!) and (S) (5) epoxide concentrations. Experimental conditions: 27 8C; aw of 0.9;
heptane as reaction medium; pH memory of 7.
required for the hydrolysis of one
mmol of epoxide per minute. The E-value for
enantioselective resolution was calculated with the Sih’s equation [25] using the
degree of conversion and the ee of the remaining epoxide obtained after
hydrolysis of the rac-pCSO mixture.

212
S. Karboune et al. / Process Biochemistry 45 (2010) 210–216
Scheme 1. Regioselectivity coefficients of EH from Aspergillus niger in the neat organic solvent medium.
and a maximal value was obtained at pH 7.0. Using the optimal
condition described here above, EH from A. niger was able to
perform the resolution of rac-pCSO (4 mM, 27 8C) in a very efficient
and enantioselective manner in the selected neat organic solvent
medium. The results (Fig. 1) show that the concentration of the (R)-
epoxide decreased rapidly, whereas that of the (S)-epoxide
remained almost constant. The EH activity calculated as the initial
hydrolytic rate of rac-pCSO within the first 10 min of the time
course was estimated to be 10.9 U mgꢁ1 solid. After 28 min of
reaction, the ee of slow-reacting (S)-epoxide reached a value of 94%
with a conversion ratio of 51%. In parallel, the ee of the resulting
(R)-diol decreased from 96 to 87%, when the reaction time was
proceeded from 2 to 28 min. The E-value, calculated from the
remaining epoxide ee and the conversion ratio, was estimated at
about 68. This value is quite similar than that obtained for the same
enzyme in aqueous medium [16]. In order to elucidate the
regioselectivity of EH from A. niger in the neat organic solvent
resolution of rac-pCSO in the four cycles are shown in Fig. 2. The
results show that the required reaction time to reach a conversion
ratio of 50% increased from 30 to 340 min over the four cycles.
Table 1 shows that only 48 and 20% of the initial EH activity were
recovered after the third and fourth cycles, respectively. In
addition, the ee of the slow-reacting (S)-epoxide decreased from
94 to 76% at a conversion ratio of about 51%, over the four
resolution cycles; as a result, the E-value of EH from A. niger
decreased from 68 to 23. Several explanations could account for
these results: (a) a decrease in the hydration level of EH after
recycling, affecting its flexibility and, as a result, its activity and
enantioselectivity; (b) an enzyme thermal inactivation by a long-
term operational catalysis; (c) an enzyme inhibition by epoxide or
diol product [27].
To elucidate the influence of the hydration level of EH on its
operational stability and enantioselectivity, the recycled biocata-
lyst was equilibrated at a aw of 0.9 after each cycle. The results
(data not showed) reveal no improvement in the operational
stability of EH upon controlling the aw of the biocatalyst after each
cycle. Our previous study [22] has shown that the use of organic
solvent as the reaction medium increased the thermal stability of
EH by 2.0–13.8-fold as compared to the aqueous medium, at
incubation temperatures ranging from 27 to 57 8C. The low
operational stability obtained after recycling cannot be attributed
solely to the thermal inactivation of the biocatalyst. Moreover, the
enzyme inhibition by substrates or products is expected to
decrease the reaction rate of the racemate conversion, thereby
increasing the reaction time required to reach the 50% conversion
rate, but would not have an effect on the biocatalyst enantios-
electivity effectiveness [28]. Therefore, the enzyme inhibition
cannot be considered as an explanation in our case, since the low
operational stability of EH is associated with a continuous decrease
of its enantioselectivity. Another possible explanation of these
unexpected results is the possible accumulation of the formed diol
in the enzyme’s micro-environment, which may have led to the
formation of a hydrophilic hindrance layer around the solid
medium, the regioselectivity coefficients, a(R), b(R), a(S) and b(S),
were determined by performing a complete hydrolysis of each one
of the two enantiomers of rac-pCSO separately [7]. Similarly to the
aqueous media, the (R)-and (S)-enantiomers of pCSO were mostly
attacked by the nucleophylic enzyme residue at the terminal
carbon atom, with b(R) and b(S) > 98%, leading to the retention of
the configuration of the corresponding diol (Scheme 1). The overall
results indicate that the use of neat organic solvent did not modify
the global enantioselectivity and regioselectivity of EH from A.
niger as compared to the aqueous medium.
3.2. Repeated-use of free solid EH in successive batch reactors
Owing to the new possibilities provided by the use of neat
organic solvent media, recycling the solid EH was investigated by
performing a series of consecutive batch reactors using the same
biocatalyst. Indeed, after each cycle, the biocatalyst was recovered
by centrifugation and reintroduced in a fresh substrate solution
equilibrated at a aw of 0.9. The time courses of EH-catalyzed
Table 1
Effect of reuse of free A. niger EH in organic media on the enzymatic activity and E-value.
Number of cycle
1
2
3
4
Remaining specific activity (%)a
E-valueb
100 (ꢂ1.1)
88 (ꢂ2.3)
49 (ꢂ3.2)
20 (ꢂ1.4)
68
47
28
23
a
Remaining specific activity was calculated as the specific activity of reused biocatalyst divided by its initial activity, and multiplied by 100. The enzymatic activity was
assayed in organic media using the optimal conditions defined previously [22].
b
E-value (enantiomeric ratio E) was calculated from the degree of conversion and the enantiomeric excess of the remaining epoxide obtained after hydrolysis of the rac-
pCSO mixture.

S. Karboune et al. / Process Biochemistry 45 (2010) 210–216
213
hydrophobic interactions on Accurel EP 100 (a polypropylene
material). The immobilization of EH on the selected supports was
performed at its optimal pH of 7.1 and using an optimal protein
loading of 10 mg protein gꢁ1 wet support, defined previously [16],
at which the effect of the diffusional limitations was not
significant. Table 2 shows that the highest protein immobilization
yield (82%) and retention of EH activity in organic media (142%)
were obtained by the hydrophobic adsorption of EH on Accurel EP
100 support. The immobilization by ionic adsorption on DEAE-
cellulose resulted in a protein immobilization yield of 78% and
retention of EH activity of 91%. Numerous values of the EH activity
retention have been reported in the literature for the immobiliza-
tion of EH by adsorption and covalent bindings, ranging from 1 to
95%, depending on the nature of the support as well as on the
operational conditions [16–18,30]. As far as the authors are aware,
there is no report on the activity of immobilized EH in the neat
organic solvent media. An increase in the enzymatic activity in
organic media after immobilization has already been reported for
several lipases [31,32] and was attributed to the micro-environ-
ment effects of supports on the substrate partition [33,34]. The
Accurel EP-100 being hydrophobic support, it may have favored
more significantly, in the neat organic solvent, the partition of the
hydrophobic epoxide toward the enzyme’s micro-environment,
leading to an apparent catalytic activation. Reduced diffusional
limitations upon immobilization [34] and/or preferential adsorp-
tion of EH on the selected supports as compared to the other
contaminating proteins [35] could also explain our results. Indeed,
our previous work [22] has revealed the occurrence of the internal
diffusional limitations with the free EH in the neat organic solvent.
A better dispersion of the EH molecules may have been achieved
upon its immobilization, leading to apparent activation factors.-
Contrary to the organic media, only 1.5 and 67% of EH activity in
the aqueous medium were recovered upon its immobilization on
Accurel EP and DEAE-cellulose, respectively [16]. As a result, the
estimated apparent RO/A (ratio of activity in organic medium to
that in aqueous medium) was higher for the EH immobilized on
Accurel EP 100 support (16.8) than those of free and immobilized
on DEAE-cellulose (1.8 ꢀ 10ꢁ3 to 2.4 ꢀ 10ꢁ3). These results reveal
that the micro-environment effects of the Accurel EP 100 support,
in the neat organic solvent, on the EH active conformation as well
as on the substrate partition were different from those that took
place in the aqueous medium [19]. An increase in the intramo-
lecular stabilizing interactions between EH and Accurel EP 100 in
the neat organic solvent as compared to the aqueous medium may
have also contributed to its high catalytic activity in the first
medium [36]. Pencreac’h and Baratti [37] have reported that
among 32 lipase preparations, six of them showed RO/A values
higher than 1, demonstrating a higher activity in organic solvent
than in water.
Fig. 2. Time course for the resolution of rac-pCSO, in neat organic media, catalyzed
by the Aspergillus niger EH reused in four repeated-batch reactors (1st black, 2nd
red, 3rd blue and 4th brown). Epoxide enantiomeric excess (ee) (&) and conversion
ratio (*). Experimental conditions: 27 8C; aw of 0.9; heptane as reaction medium;
pH memory of 7. (For interpretation of the references to color in this figure caption,
the reader is referred to the web version of the article.)
enzyme particles and thereby limited the diffusion of the
hydrophobic epoxide substrates inside the biocatalyst particles
to the enzyme active site. This hypothesis was examined by
washing the reused EHs after four cycles with ethyl acetate to
remove the diol-based hydrophilic hindrance layer. This washing
attempt resulted in 80% recovery of the initial EH activity (data not
shown) and confirmed the effect of mass-diffusional and
accessibility limitations on the operational stability of EH in the
neat heptane. Other previous studies [28,29] have also reported
that in addition to the decrease in the catalytic performance, the
presence of mass-diffusional limitations could lead to a decrease in
the enzyme enantioselectivity. In order to improve the operational
stability and the enantioselectivity of EH after recycling, strategies
consisting of the immobilization of EH and the use of a binary
organic solvent as a reaction medium were undertaken.
3.3. Immobilization of EH and its application for the resolution
of rac-pCSO
Based on our previous work [16], EH from A. niger was
immobilized by ionic adsorption on DEAE-cellulose and by
Table 2
Immobilization of EH from Aspergillus niger and assessment of its hydrolytic activity in organic media.
E-valuef
e
Biocatalyst
Protein immobilization
yield (%)a
Retention of
activity (%)b
Specific activity
(U/mg protein)c
EH activity
(U/g support)d
RO/A
Free EH
100
142
90
0.043 (ꢂ0.0015)g
0.061 (ꢂ0.0032)
0.038 (ꢂ0.0025)
1.8 ꢀ 10ꢁ3
16.8
2.4 ꢀ 10ꢁ3
65
38
28
Immobilized EH on Accurel EP 100
Immobilized EH on DEAE-cellulose
82
78
0.50 (ꢂ0.023)
0.31 (ꢂ0.014)
a
Protein immobilization yield was defined as the ratio of immobilized protein relative to the initial amount of free protein suspended with support and multiplied by 100%.
Retention of activity was calculated as the ratio of the assayed activity of the immobilized EH to that of the theoretical activity in organic solvent media, and multiplied by
b
100.
c
Specific activity is expressed as
m
mol of produced diol per min of reaction per mg protein.
d
EH activity is expressed as
previously [22].
mmol of produced diol per min of reaction per g humid support and was assayed in organic media using the optimal conditions defined
e
RO/A was defined as the ratio of activity in organic media to that in aqueous media.
f
E-value (enantiomeric ratio E) in organic media was calculated from the degree of conversion and the enantiomeric excess of the remaining epoxide obtained after
hydrolysis of the rac-pCSO mixture.
g
Data are average of three determinations and the standard deviations are given.

214
S. Karboune et al. / Process Biochemistry 45 (2010) 210–216
Table 3
A. niger EH activity and its enantiomeric ratio E using heptane:dioxane organic
solvent mixture at different ratios as reaction medium.
Ratio of heptane:dioxane
solvent mixture (v:v)
Log Pa
Relative
EH activity (%)b
E-valuec
100:0
98:2
4
100 (ꢂ0.1)
99.5 (ꢂ5.1)
98.9 (ꢂ3.3)
85.2 (ꢂ4.8)
49.3 (ꢂ0.9)
14.5 (ꢂ0.4)
26.6 (ꢂ0.3)
60
58
56
59
62
44
14
3.83
3.59
3.20
2.49
0.80
ꢁ1.1
95:5
90:10
80:20
50:50
0:100
a
Log P value defined as the partition coefficient of the solvent mixture between
water and 1-octanol. Log P of the solvent mixture was calculated according to the
empirical formula Log P mixture = X1 Log P1 + X2 Log P2, in which X1 and X2 are the
mol fractions of solvents 1 and 2.
b
Relative specific activity was calculated as the specific activity of EH in organic
solvent mixture divided by that in heptane, and multiplied by 100.
c
E-value (enantiomeric ratio E) was calculated from the degree of conversion and
the enantiomeric excess of the remaining epoxide obtained after hydrolysis of the
rac-pCSO mixture.
Fig. 3. Reuse of immobilized EH onto DEAE-cellulose (&) and Accurel EP 100 (&) for
several reaction cycles of kinetic resolution of rac-pCSO in neat organic solvent
media.
an average of 60. However, when heptane and dioxane organic
solvent mixture of 50:50 (v:v) was used as the reaction medium,
the E-value decreased to 44 and low enantioselectivity (E-value of
14) and EH activity (27%) were displayed in the neat dioxane. It
should be emphasized that the EH preferred the (R)-epoxide in
both the hydrophilic and hydrophobic organic solvent mixtures.
The decrease in the EH activity with decreasing the Log P of solvent
mixtures may be explained [38] by (a) enzyme conformational
changes caused by the enzyme–solvent interactions; (b) the poor
solvation of hydrophobic epoxides, leading to a decrease in the
interactions between the biocatalyst and the substrates; and/or (c)
the desolvation of ES in the transition state leading to less enzyme–
substrate complexation. Contrary to the enzyme activity, the
enantioselectivity seems to be more or less unaltered with the
decrease in the solvent hydrophobicity until Log P was 2.0, and
then decreased with a further decrease of Log P. The results reveal
that the effects of dioxane on the molecular recognition and the
desolvation modes of the (S)- and (R)-enantiomers of epoxide in
the transition state were more significant in the organic solvent
mixtures with Log P less than 2 [39].
Contrary to the enzyme activity, the enantioselectivity of EH
from A. niger was altered after immobilization on DEAE-cellulose
and Accurel EP 100 (Table 2). Similar results have been reported for
the same enzyme in the aqueous medium [16] and were attributed
to the distinct alteration of the conformational structure of the
enzyme upon its immobilization on DEAE-cellulose, modifying the
binding of each enantiomer in its active site.
In order to assess the operational stability of the EH
immobilizates, repeated-batch reactors were performed using
the same biocatalyst and the optimized conditions defined
previously. Fig. 3 shows that the operational stability of the
immobilized EH on DEAE-cellulose was higher than that immo-
bilized on the Accurel EP 100. Immobilized EH on the DEAE-
cellulose still showed 81, 62, 52 and 41% of its initial activity after
the second, third, fourth and fifth cycles, respectively. Only 50 and
20% of the initial activity of the immobilized EH on the Accurel EP
100 were retained after the second and third cycles, respectively,
whereas no activity could be detected at the fourth cycle.
Unexpectedly, the Accurel EP 100 support seems to have promoted
a strong hindrance of the internal mass diffusion in the enzyme
micro-environment during recycling. These results may be
attributed to the internal morphology of the macroporous Accurel
EP 100 support, which has a high pore size (0.05–0.5
mm) that may
have favored the accumulation of diol in the EH micro-environ-
ment, leading to the observed apparent low operational stability.
In contrast, the nonporous DEAE-cellulose improved the opera-
tional stability of EH (binding in its external surface) during
recycling by more than twofold as compared to the free EH.
3.4. Effect of binary organic solvent mixtures
Although the polar organic solvent media may enhance the
solubility of the hydrophilic diol products, they have a greater
ability to strip off the essential water layer from the enzyme
molecule surface, leading to the enzyme deactivation [19]. To
overcome this shortcoming, series of binary organic solvent
mixtures consisting of hydrophobic heptane (Log P = 4) and
hydrophilic dioxane (Log P = ꢁ1.1) at different ratios were
investigated as reaction media for EH-catalyzed resolution of
rac-pCSO, using the optimal conditions defined previously (aw, 0.9;
T, 27 8C; pH memory of 7.1). The results (Table 3) show that
increasing the proportion of dioxane in the organic solvent mixture
from 2 to 20% resulted in a commitment decrease in the EH
activity, whereas the E-value remained relatively constant around
Fig. 4. Reuse of free A. niger EH for several reaction cycles of kinetic resolution of rac-
pCSO in heptane: dioxane organic solvent mixture at a ratio (v:v) of 90:10 (&),
80:20 ( ), 50:50 ( ) and 0:100 ( ).

S. Karboune et al. / Process Biochemistry 45 (2010) 210–216
215
use of heptane and dioxane organic solvent mixture at a ratio of
80:20 (v:v) provided a relatively high operational stability of the
solid EH from A. niger, as a result of the increase of the solubility of
diol, without affecting its enantioselectivity.
Acknowledgment
We would like to thank Dr Jaap Visser for transforming the
protease-deficient A. niger strain and the permission to use this
strain.
References
[1] Drauz K, Waldmann H. Enzyme catalysis in organic synthesis. Weinheim:
Wiley–VCH; 2002.
[2] Schoemaker HE, Mink D, Wubbolts MG. Dispelling the myths—biocatalysis in
industrial synthesis. Science 2003;299:1694–7.
[3] Archelas A, Furstoss R. Synthetic application of epoxide hydrolases. Curr Opin
Chem Biol 2001;5:112–9.
[4] de Vries EJ, Janssen DB. Biocatalytic conversion of epoxides. Curr Opin Bio-
technol 2003;14:414–20.
[5] Choi WJ, Choi YC. Production of chiral epoxides: epoxide hydrolase-catalyzed
enantioselective hydrolysis. Biotechnol Bioproc Eng 2005;10:167–79.
[6] Bottalla A-L, Ibrahim-Ouali M, Santelli M, Furstoss R, Archelas A. Epoxide
hydrolase-catalyzed kinetic resolution of a spiroepoxide, a key building block
of various 11-heterosteroids. Adv Synth Catal 2007;349:1102–10.
[7] Moussou P, Archelas A, Baratti J, Furstoss R. Microbiological transformations.
Part 39. Determination of the regioselectivity occurring during oxirane ring
opening by epoxide hydrolases: a theoretical analysis and a new method for its
determination. Tetrahedron Asymmetry 1998;9:1539–47.
[8] Steinreiber A, Mayer SF, Faber K. Asymmetric total synthesis of a beer-aroma
constituent based on enantioconvergent biocatalytic hydrolysis of trisubsti-
tuted epoxides. Synthesis 2001;13:2035–9.
[9] Hwang S, Choi CY, Lee EY. Enantioconvergent bioconversion of p-chlorostyrene
oxide to (R)-p-chlorophenyl-1,2-ethandiol by the bacterial epoxide hydrolase
of Caulobacter crescentus. Biotechnol Lett 2008;30:1219–25.
[10] Lingyun R, Li C, Wilfred C, Kenneth FR, Thomas KW. Protein engineering of
epoxide hydrolase from Agrobacterium radiobacter AD1 for enhanced activity
and enantioselective production of (R)-1-phenylethane-1,2-diol. Appl Environ
Biotechnol 2005;71:3995–4003.
[11] Reetz MT, Kahakeaw D, Lohmer R. Addressing the numbers problem in
directed evolution. Chem Bio Chem 2008;9:1797–804.
[12] Reetz MT, Wang L-W, Bocola M. Directed evolution of enantioselective
enzymes: iterative cycles of CASTing for probing protein-sequence space.
Angew Chem Int Ed Engl 2006;45:1236–41.
[13] Monfort N, Archelas A, Furstoss R. Enzymatic transformations. Part 55. Highly
productive epoxide hydrolase catalysed resolution of an azole antifungal key
synthon. Tetrahedron 2004;60:601–4.
[14] Doume`che D, Archelas A, Furstoss R. Enzymatic transformations 62. Prepara-
tive scale synthesis of enantiopure glycidyl acetals using an Aspergillus niger
epoxide hydrolase-catalysed kinetic resolution. Adv Synth Catal 2006;348:
1948–57.
[15] Deregnaucourt J, Archelas A, Barbirato F, Paris J-P, Furstoss R. Enzymatic
transformations 63. High-concentration two liquid–liquid phase Aspergillus
niger epoxide hydrolase-catalysed resolution: application to trifluoromethyl-
substituted aromatic epoxides. Adv Synth Catal 2007;349:1405–17.
[16] Karboune S, Archelas A, Furstoss R, Baratti J. Immobilization of epoxide
hydrolase from Aspergillus niger onto DEAE-cellulose: enzymatic properties
and application for the enantioselective resolution of a racemic epoxide. J Mol
Catal B Enzym 2005;32:175–83.
Fig. 5. Reuse of free A. niger EH for several reaction cycles of kinetic resolution of rac-
pCSO at a concentration of 100 mM in dioxane.
Fig. 4 summarizes the results of repeated-batch reactors carried
out using different binary organic solvent mixtures of heptane and
dioxane as the reaction medium. The use of 10% dioxane in heptane
did not improve the operational stability of EH after recycling as
compared to the free EH. However, higher operational stability was
observed in heptane and dioxane mixtures at a ratio of 80:20 and
50:50 (v:v); after the fifth cycle, 73 and 80% of the initial activity of
EH were still retained, respectively. Although the use of the neat
dioxane reduced the activity and the enantioselectivity of EH from
A. niger, it led to a high operational stability with more than 87% of
the initial activity of EH was still recovered after the fifth cycle.
Consequently, using the neat dioxane as reaction medium, the free
EH was also successfully recycled in a repeated-batch reactor with
a high concentration of pCSO substrate (100 mM, i.e. 15.6 g Lꢁ1) for
six cycles without significant decrease of its enantioselectivity and
only a low decrease of its initial activity after each run (5%) (Fig. 5).
Although the neat organic solvent media is not critical for an
epoxide like, rac-pCSO for which we can work in a biphasic
medium at a high substrate concentration, its use is a very
promising and attractive approach for other more hydrophobic
solid epoxides insoluble in aqueous medium. However,
a
compromise between the Log P of the reaction medium and the
catalytic efficiency as well as the operational stability of EH should
be considered in order to ensure an effective resolution of rac-
epoxides.
[17] Mateo C, Archelas A, Fernandez-Lafuente R, Guisan J-M, Furstoss R. Enzymatic
transformations. Immobilized A. niger epoxide hydrolase as a novel biocata-
lytic tool for repeated-batch hydrolytic kinetic resolution of epoxides. Org
Biomol Chem 2003;15:2739–43.
4. Conclusion
[18] Petri A, Marconcini P, Salvadori P. Efficient immobilization of epoxide hydro-
lase onto silica gel and use in the enantioselective hydrolysis of racemic para-
nitrostyrene oxide. J Mol Catal B Enzym 2004;32:219–24.
Using the optimal reaction conditions, the global enantioselec-
tivity and regioselectivity of EH from A. niger in the neat organic
solvent medium were similar to those obtained in the aqueous
medium. In order to improve the operational stability and the
enantioselectivity of EH after recycling, strategies consisting of the
immobilization of EH and the use of a binary organic solvent as a
reaction medium were successfully applied. The Accurel EP 100
support has favored the observed apparent low operational
stability during recycling due to the accumulation of diol in the
enzyme micro-environment, whereas the nonporous DEAE-cellu-
lose improved the operational stability of EH by more than twofold
as compared to the free EH. The overall results indicated that the
[19] Halling PJ. Enzymatic conversions in organic and other low-water media. In:
Drauz K, Waldmann H, editors. 2nd ed., Enzyme catalysis in organic synthesis
A comprehensive handbook, vol. I, 2nd ed. Weinheim, Germany: Wiley–VCH/
Verlag GmbH; 2002. p. 259–86.
[20] Klibanov AM. Improving enzymes by using them in organic solvents. Nature
2001;409:241–6.
[21] Carrea G, Riva S. Properties and synthetic applications of enzymes in organic
solvents. Angew Chem Int Ed 2000;39:2226–54.
[22] Karboune S, Archelas A, Baratti J. Properties of epoxide hydrolase from
Aspergillus niger for the hydrolytic kinetic resolution of epoxides in pure
organic media. Enz Microbiol Technol 2006;39:318–24.
[23] Arand M, Hemmer H, Durk H, Baratti J, Archelas A, Furstoss R, et al. Cloning and
molecular characterization of a soluble epoxide hydrolase from Aspergillus
niger that is related to mammalian microsomal epoxide hydrolase. Biochem J
1999;344:273–80.

216
S. Karboune et al. / Process Biochemistry 45 (2010) 210–216
[24] Naundorf A, Melzer G, Archelas A, Furstoss R, Wohlgemuth R. Influence of pH
on the expression of a recombinant epoxide hydrolase in Aspergillus niger.
Biotechnol J 2009;4:756–65.
[32] Pencreac’h G, Baratti J. Activity of Burkholderia cepacia lipase in organic media
is greatly enhanced after immobilisation on polypropylene support. Appl
Microbiol Biotechnol 1997;47:630–5.
[25] Chen CS, Fujimoto Y, Girdaukas G, Sih CJ. Quantitative analyses of biochemical
kinetic resolutions of enantiomers. J Am Chem Soc 1982;104:7294–9.
[26] Holmquist M, Martinelle M, Berglund P, Clausen IG, Patkar S, Svendsen A, et al.
Lipases from Rhizomucor miehei and Humicola lanuginosa: modification of the
[33] van Tol JBA, Stevens RMM, Veldhuizen WJ, Jongejan JA, Duine JA. Do organic
solvents affect the catalytic properties of lipase? Intrinsic kinetic parameters
of lipases in ester hydrolysis and formation in various organic solvents.
Biotechnol Bioeng 1995;47:71–81.
lid covering the active site alters enantioselectivity.
1993;12:747–55.
J
Protein Chem
[34] Ottolina G, Carrea G, Riva S, Sartore L, Veronese FM. Effect of the enzyme form
on the activity, stability and enantioselectivity of lipoprotein lipase in toluene.
Biotechnol Lett 1992;14:947–52.
[35] Fernandez-Lafuente R, Armisen P, Sabuquillo P, Fernandez-Lorente G, Guisan
JM. Immobilization of lipases by selective adsorption on hydrophobic sup-
ports. Chem Phys Lipids 1998;93:185–97.
[36] Hartsough DS, Merz KM. Protein dynamics and solvation in aqueous and
nonaqueous environments. J Am Chem Soc 1993;114:6529–37.
[37] Pencreac’h G, Baratti JC. Comparison of hydrolytic activity in water and
heptane for thirty-two commercial lipase preparations. Enzyme Microb Tech-
nol 2001;28:473–9.
[38] Hazarika S, Goswami P, Dutta NN, Hazarika AK. Ethyl oleate synthesis by
porcine pancreatic lipase in organic solvents. Chem Eng J 2002;85:61–8.
[39] Ke T, Wescott RC, Klibanov MA. Predictions of the solvent dependence of
enzymatic prochiral selectivity by means of structure-based thermodynamic
calculations. J Am Chem Soc 1996;118:3366–74.
[27] Toscano G, Pirozzi D, Maremonti M, Greco G. Solid-state enzyme deactivation
in air and in organic solvents. Biotechnol Bioeng 1994;44:682–9.
[28] Xiu G-H, Jiang L, Li P. Mass-transfer limitations for immobilized enzyme-
catalyzed kinetic resolution of racemate in a fixed-bed reactor. Biotechnol
Bioeng 2000;74:30–9.
[29] Jongejan JA, Heijnen JJ. Diffusion limitation causes decreased enantioselec-
tivity of esterification of 2-butanol by immobilized Candida Antarctica Lipase B.
Biocatal Biotrans 2000;19:1–19.
[30] Mateo C, Fernandez-Lafuente R, Archelas A, Guisan JM, Furstoss R. Preparation
of a very stable immobilized Solanum tuberosum epoxide hydrolase. Tetra-
hedron Asymmetry 2007;18:1233–8.
[31] Basri M, Ampon K, Wan Yunus WMZ, Razak CNA, Salleh AB. Enzymatic
synthesis of fatty esters by hydrophobic lipase derivatives immobilized on
organic polymer beads. J Am Oil Chem Soc 1995;72:407–11.