Increased catalyst productivity in α-hydroxy acids resolution by esterase mutation and substrate modification

Article abstract of DOI:10.1021/cs401183e

Optically pure α-hydroxy acids and their derivatives are versatile chiral building blocks in the pharmaceutical industry. In this study, the potential of a recombinant Pseudomonas putida esterase (rPPE01) for the enzymatic resolution of α-acetoxy acids was significantly improved by combinatorial engineering of both the biocatalyst and substrate. Semirational design based on homologous modeling and molecular docking provided a single-point variant, W187H, whose k_cat/K_M for sodium 2-acetoxy-2-(2′-chlorophenyl)acetate (Ac-CPA-Na) was increased 100-fold, from 0.0611 to 6.20 mM^-1 s^-1, while retaining its excellent enantioselectivity and broad substrate spectrum. Biocatalyst deactivation under the operating conditions was decreased by using the potassium salt of Ac-CPA instead of Ac-CPA-Na. With 0.5 g L^-1 of lyophilized cells containing rPPE01-W187H, 500 mM (R,S)-Ac-CPA-K was selectively deacylated with 49.9% conversion within 15 h, giving satisfactory enantiomeric excesses (ee) for both the S product (>99% ee) and the remaining R substrate (98.7% ee). Consequently, the amount of (S)-2-hydroxy-2-(2′-chlorophenyl)acetate prepared per unit weight of lyophilized cells was improved by a factor of 18.9 compared with the original productivity of the wild-type esterase. Further enzymatic resolution of other important hydroxy acids at the 100 mL scale demonstrated that the rPPE01-W187H-based bioprocess is versatile and practical for the large-scale preparation of chiral α-hydroxy acids.

Full text of DOI:10.1021/cs401183e

Research Article
pubs.acs.org/acscatalysis
Increased Catalyst Productivity in α‑Hydroxy Acids Resolution by
Esterase Mutation and Substrate Modification
Bao-Di Ma, Xu-Dong Kong, Hui-Lei Yu,* Zhi-Jun Zhang, Shuai Dou, Yan-Peng Xu, Yan Ni,
and Jian-He Xu*
Laboratory of Biocatalysis and Synthetic Biotechnology, State Key Laboratory of Bioreactor Engineering, East China University of
Science and Technology, 130 Meilong Road, Shanghai 200237, China
S
* Supporting Information
ABSTRACT: Optically pure α-hydroxy acids and their
derivatives are versatile chiral building blocks in the
pharmaceutical industry. In this study, the potential of a
recombinant Pseudomonas putida esterase (rPPE01) for the
enzymatic resolution of α-acetoxy acids was significantly
improved by combinatorial engineering of both the biocatalyst
and substrate. Semirational design based on homologous
modeling and molecular docking provided a single-point
variant, W187H, whose kcat/KM for sodium 2-acetoxy-2-(2′-
chlorophenyl)acetate (Ac-CPA-Na) was increased 100-fold,
from 0.0611 to 6.20 mM−1 s−1, while retaining its excellent
enantioselectivity and broad substrate spectrum. Biocatalyst deactivation under the operating conditions was decreased by using
the potassium salt of Ac-CPA instead of Ac-CPA-Na. With 0.5 g L−1 of lyophilized cells containing rPPE01-W187H, 500 mM
(R,S)-Ac-CPA-K was selectively deacylated with 49.9% conversion within 15 h, giving satisfactory enantiomeric excesses (ee) for
both the S product (>99% ee) and the remaining R substrate (98.7% ee). Consequently, the amount of (S)-2-hydroxy-2-(2′-
chlorophenyl)acetate prepared per unit weight of lyophilized cells was improved by a factor of 18.9 compared with the original
productivity of the wild-type esterase. Further enzymatic resolution of other important hydroxy acids at the 100 mL scale
demonstrated that the rPPE01-W187H-based bioprocess is versatile and practical for the large-scale preparation of chiral α-
hydroxy acids.
KEYWORDS: catalyst productivity, deacylation, enzymatic resolution, α-hydroxy acid, product:catalyst ratio, semirational design
consumption [0.98 g of product were obtained per gram of
catalyst using the most efficient Lipase PS (Amano) to acetylate
substituted mandelic acids],12 which is unsuitable for industrial-
scale applications.15
INTRODUCTION
■
α-Hydroxy acids are widely used as chiral synthons in the
chemical and pharmaceutical industries.1,2 Enantiopure 2-
hydroxyphenylacetic acid and its derivatives represent the
most important class of hydroxy acids because of their high
commercial value.3−5 (R)-2-Hydroxy-2-(2′-chlorophenyl)acetic
acid is the key chiral intermediate in the synthesis of (S)-
clopidogrel, a platelet aggregation inhibitor with a large market
share, used for heart attack and stroke treatment. Various
biocatalytic approaches to the synthesis of enantiopure α-
hydroxy acids have been developed,1 including reductase-
catalyzed asymmetric reduction of the corresponding keto
ester,3,4,6 nitrilase-catalyzed enantioselective hydrolysis of
mandelonitrile and its derivatives,7,8 and esterase or lipase
catalyzed resolution of the O-acetates or esters of hydroxy
acids.9,10 Among these pathways, the enzymatic resolution of
O-acetylated hydroxy acids by esterase or lipase (Scheme 1)
shows great potential for application due to their practical use
in chemical industry11 and has been extensively studied.10,12−14
Several commercially available lipases [e.g., Pseudomonas sp.
In our previous work, an esterase producer, Pseudomonas
putida ECU1011, was isolated from soil for the enantioselective
deacylation of (S)-2-acetoxy-2-(2′-chlorophenyl) acetate [(S)-
Ac-CPA] from its racemate.16 Subsequently, the corresponding
esterase (rPPE01) was successfully cloned and heterogeneously
expressed in Escherichia coli.14 The excellent thermostability
and high enantioselectivity (E > 200) of rPPE01 toward α-
acetoxycarboxylates indicated its great potential in industrial
applications. However, practical applications of rPPE01-
mediated bioprocesses were impeded by its relatively low
enzymatic activity and poor operational stability. Further
enhancement of the activity and stability are therefore highly
desirable to achieve industrial applications of rPPE01 in the
enzymatic resolution of α-hydroxy acids.
Lipase (Rohm GmbH) and Lipase PS (Amano)] have been
used in such biotransformations. However, most of them show
poor productivity, resulting in high costs in terms of catalyst
Received: December 12, 2013
Revised: February 13, 2014
Published: February 17, 2014
̈
© 2014 American Chemical Society
1026
dx.doi.org/10.1021/cs401183e | ACS Catal. 2014, 4, 1026−1031

ACS Catalysis
Research Article
Scheme 1. Esterase or Lipase Catalyzed Resolution of Racemic α-Acetoxycarboxylates
Protein engineering is frequently used to tailor native
enzymes for industrial purposes, and great progress has been
made in recent decades.17 Two different and complementary
strategies have been developed for protein engineering:
directed evolution followed by high-throughput screening of
the numerous variants18 and rational design based on detailed
information on the three-dimensional structure and structure−
function relationship of the enzyme.19 Recently, with the rapid
increase in the number of enzyme structures available from the
Protein Data Bank (PDB) and highly developed bioinformatic
tools, rational design has become an effective strategy for
designing desired enzymes. A BLAST search of the PDB
revealed that three crystal structures from a hormone-sensitive
lipase family showed considerable sequence identities (>40%)
with rPPE01, which made it possible to predict the three-
dimensional structure of rPPE01 by homology modeling.
Because it is a member of the α/β-hydrolase fold family,
another advantage of the rational design of rPPE01 was the in-
depth insights available into its reaction mechanism.20 Rational
design of the “hot spots” that might potentially affect the
enzymatic activity would significantly increase the efficiency of
desired mutant hits. A semirational design based on the
homology-modeled structure of rPPE01 was therefore chosen
to engineer rPPE01, i.e., to identify hot spots for subsequent
mutations.
The poor operational stability of biocatalysts is a common
restriction in the industrialization of enzymatic processes.
Several strategies have been developed to stabilize enzymes for
industrial use, such as immobilization and use of additives.21,22
Dissolved salts are important additives for enzyme stabilization.
However, the effects of stabilization with different salts may
differ.23 The stabilization effects of various cations and anions at
moderate or high concentrations (ionic strength >0.01 mol
kg−1) are ranked qualitatively in the Hofmeister series.24 The
substrate Ac-CPA exists in the ionic form in our study,14 and a
high substrate concentration would result in a high ionic
strength in the reaction system. Cations at such a high salt
concentration might therefore significantly affect the biocatalyst
in this study. This might provide a simple and direct method of
relieving biocatalyst deactivation by choosing a suitable
substrate counterion (cation).
variant with superior activity was obtained and used to reduce
the biocatalyst load. The productivity of the engineered
biocatalyst in the reaction process was further improved by
the stabilizing effect on the enzyme obtained by switching the
substrate counterion. An rPPE01-W187H-based bioprocess
with significantly increased catalyst productivity, defined as the
weight gain ratio of the product to the catalyst (P/C), was
achieved in this study.
EXPERIMENTAL SECTION
■
Chemicals. Racemic 2-hydroxy-3-phenylpropanoic acid was
purchased from the Yurlic Chemical S&T Co., Ltd. (Shanghai,
China). Racemic 2-bromonaphthalene was purchased from the
Kangtuo Chemical Co., Ltd. (Shanghai, China). Racemic 2-
hydroxy-2-phenylacetic acid, 2-hydroxy-2-(2′-chlorophenyl)-
acetic acid, 2-hydroxy-2-(3′-chlorophenyl)acetic acid, and 2-
hydroxy-2-(4′-chlorophenyl)acetic acid were obtained from the
Guangde Chemical Co., Ltd. (Anhui, China). Unless otherwise
stated, all other chemicals and reagents were obtained
commercially. The synthesis of 2-hydroxy-4-phenylbutyrate
and 3-hydroxy-4-phenylbutyrate as well as the further
acetylation of the hydroxy acids were conducted according to
previously published protocols.14,25
Homology Modeling and Molecular Docking. The
crystal structures in the PDB showing high sequence identities
with rPPE01 (>40%) were selected as templates. The rPPE01
structure was modeled with a multiple-template-based homol-
ogy modeling method, using the program Modeller 9.11, aided
by the graphical interface EasyModeller.26 After further energy
minimization using the same software package, the quality of
the modeled structure was assessed using the Procheck
program (http://nihserver.mbi.ucla.edu/SAVES/). Docking
experiments were performed using the program AutoDock
Vina, and the images were created using PyMOL.27
Mutant Construction and Esterase Expression. Muta-
tions were introduced using a Mut Express II Fast Mutagenesis
Kit (Vazyme Biotech Co., Ltd.), according to the manufac-
turer’s instructions and confirmed by sequencing. The primer
sequences are listed in the Supporting Information (SI). The
plasmids with specific mutations were then transformed into E.
coli BL21 (DE3) for esterase expression. Cultivation of the
recombinant E. coli cells expressing rPPE01 or its mutants and
enzyme purification using Ni2+ affinity chromatography were
performed as described previously.14
In light of the above analysis, the semirational design of an
esterase and substrate−cation optimization were used to
improve the enzymatic activity and operational stability,
respectively. On the basis of our smart design, a single-point
1027
dx.doi.org/10.1021/cs401183e | ACS Catal. 2014, 4, 1026−1031

ACS Catalysis
Research Article
Enzymatic Activity and Enantioselectivity Assay. The
enzymatic activity, conversion, and enantioselectivity toward
acetoxy acids were analyzed using chiral HPLC, as described
previously.14 One unit of enzyme activity was defined as the
amount of enzyme releasing 1.0 μmol of hydroxy acid per
minute under the assay conditions. The residual activity was
analyzed spectrophotometrically by measuring the hydrolysis
rate of 1 mM p-nitrophenyl acetate to p-nitrophenol28 and
expressed as a percentage of the initial activity measured at 0 h.
Characterization of Esterase Variant rPPE01-W187H.
The initial hydrolysis rates of various α-acetoxycarboxylates by
purified rPPE01-W187H were measured using the same
method as that used with rPPE01-WT.14 The initial rates of
Ac-CPA-Na deacylation under various reaction conditions were
measured to evaluate the effects of temperature and pH on the
enzymatic activity. The thermostabilities of rPPE01-W187H
and its wild type were investigated by measuring the residual
activities after incubation of the purified esterase (0.40 mg
mL−1) at different temperatures (30, 40, and 50 °C) for 12 h.
Enzymatic Resolution of α-Acetoxy Acid by rPPE01-
W187H or its Wild Type. The reaction, using 50 mM
phosphate buffer (pH 6.5), lyophilized E. coli cells expressing
rPPE01-W187H or its wild type, Ac-CPA (300 mM), and 0.5
equiv of Na2CO3, was performed using a total volume of 10 mL
at 30 °C. The pH of the reaction mixture was automatically
maintained at 6.5 by titrating with 0.5 M Na2CO3 during the
process.
Optimization of Substrate Counterions. To identify the
best cation for the substrate in the enzymatic resolution of Ac-
CPA by rPPE01-W187H, various alkalis (Na2CO3, K2CO3, or
NH3·H2O) were added to neutralize the substrate and acetic
acid produced in the reaction. The enzymatic reactions were
carried out at a substrate concentration of 500 mM, as
described above. Samples were taken periodically to measure
the conversions and residual activities.
Preparation of Enantiopure Hydroxy Acids. For larger-
scale preparation, α-acetoxy acid (20 or 50 mmol) and 0.5
equiv of K2CO3 (10 or 25 mmol) were dissolved in 100 mL of
phosphate buffer (50 mM, pH 6.5). Lyophilized recombinant E.
coli cells (0.05 or 0.2 g) were added to initiate the reaction at 30
°C. When the conversion reached ∼50%, the reaction mixture
was acidified with 2 M H2SO4 and extracted twice with ethyl
acetate. Then the combined extracts were dried over anhydrous
sodium sulfate for product isolation. Enantiopure (R)-α-acetoxy
acids and (S)-α-hydroxy acids were obtained after subsequent
chromatographic separation, and the (R)-α-acetoxy acids were
chemically deacylated in aqueous NaOH solution (20%, w/v)
at room temperature. After complete hydrolysis, the products
were acidified again and extracted as described above. The
conversion and ee values for the substrates and products during
the process were analyzed using chiral HPLC as described in
the SI. The isolated products were validated based on specific
rotation and 1H NMR spectra, and the data are also given in the
SI.
Figure 1. Close-up view of rPPE01 active site in complex with the
most favorable docked position of (S)-Ac-CPA (yellow sticks). The
predicted hot spots, W187 and D287, are highlighted as green sticks.
Table 1. Steady-State Kinetic Parameters and
Enantioselectivities of rPPE01-WT and Variants for
Substrate Ac-CPA-Na
c
enzyme
KM (mM) kcat (s−1
)
kcat/KM (mM−1 s−1
)
E
a
rPPE01-WT
146
19.9
4.30
8.90
123
21.6
0.0611
6.20
>200
>200
8.7
b
rPPE01-W187H
a
rPPE01-D287A
5.01
a
The kinetic parameters were determined at its optimal pH (pH 6.2,
b
50 mM sodium phosphate buffer) and temperature (50 °C). The
kinetic parameters were determined at its optimal pH (pH 6.5, 50 mM
c
sodium phosphate buffer) and temperature (40 °C). The enzymes are
selective toward the S substrate, producing the (S)-2-hydroxy-2-(2′-
chlorophenyl) acetic acid in large excess.
Table 2. Substrate Specificity of rPPE01-W187H and Its
Wild Type
initial rate (μmol min−1 mg protein−1
)
a
b
substrate
rPPE01-WT
rPPE01-W187H
rac-1a
rac-1b
rac-1c
rac-1d
rac-1e
rac-1f
rac-1g
rac-1h
1.37 0.09
1.17 0.06
1.47 0.20
76.7 3.1
15.4 0.1
1.34 0.10
c
c
n.d.
n.d.
2.09 0.01
1.06 0.10
1.44 0.06
75.2 0.6
51.7 4.3
40.2 0.3
3.72 0.20
0.40 0.02
a
b
Data cited from reference 14. The initial rates were measured using
purified enzyme under its optimal reaction condition. Values represent
the mean of three replicates standard deviation. n.d.: not detectable.
RESULTS AND DISCUSSION
■
c
On the basis of the PDB BLAST search, three crystal structures
showing high sequence identities with rPPE01, from
Alicyclobacillus acidocaldarius (PDB entry: 2HM7, 42%
identity), archaeon Pyrobaculum calidifontis VA1 (PDB entry:
2YH2, 42% identity), and a metagenomic library (PDB entry:
2C7B, 41% identity), were selected as templates. The multiple-
template-based homology-modeled structure of rPPE01 pre-
dicted by Modeller was validated using Procheck to ensure its
reliability. The rPPE01 model consists of a core domain,
belonging to the canonical α/β-hydrolase topological fold, and
two separate helical regions.29 A detailed inspection of the
solvent-accessible surface around the conserved catalytic triad
1028
dx.doi.org/10.1021/cs401183e | ACS Catal. 2014, 4, 1026−1031

ACS Catalysis
Research Article
of speculation that the high KM value (146 mM) and low kcat
value (8.90 s−1) of rPPE01 were caused by inappropriate
accommodation of the negatively charged substrate. It was
observed that the side chain of D287 partially occupied the
alcohol-group-binding pocket, and its substitution by alanine
was designed to enlarge the binding pocket and to eliminate
possible repulsion between the carboxyl group of aspartic acid
and the hydrophobic phenyl moiety or the negatively charged
carboxyl group of the substrate. The other residue, W187, was
located at the bottom of the alcohol-binding pocket, toward the
negatively charged carboxyl group of the substrate. The
substitution of W187 by the polar and smaller histidine was
designed to increase the local polarity and to expand the
binding pocket for facile accommodation of the carboxyl group
from the substrate. Two designed mutants (D287A and
W187H) were constructed and subjected to subsequent
enzymatic activity and enantioselectivity analysis.
Figure 2. Bioresolution of Ac-CPA-Na at 300 mM by rPPE01-W187H
(red dot) and rPPE01-WT (black solid square) at a catalyst load of 0.5
g L−1 of lyophilized cells. The reaction was conducted at 30 °C, and
the pH was maintained at 6.5 by automatically titrating 0.5 M Na2CO3.
The steady-state kinetic parameters for the Ac-CPA substrate
(shown in Table 1) indicated that the catalytic efficiencies of
both the constructed mutants were significantly higher than
that of wild-type rPPE01. For the W187H variant, a 7.4-fold
decrease in KM combined with a 13.9-fold increase in kcat
contributed to a total 100-fold increase in kcat/KM. For the
D287A mutant, the 81.6-fold increase in kcat/KM was ascribed
to the 34.0-fold decrease in KM and the 2.4-fold increase in kcat.
The W187H variant retained excellent enantioselectivity (E >
200). However, the enantioselectivity of the D287A variant
decreased significantly (the E value decreased from >200 to
8.7); this might be explained by excessive enlargement of the
binding pocket. The W187H variant exhibited significantly
increased catalytic efficiency and retained excellent enantiose-
lectivity toward rac-Ac-CPA-Na. The potential application of
the W187H variant to the production of chiral hydroxy acids
was therefore further investigated.
The initial reaction rates of rPPE01-W187H and rPPE01-
WT toward a series of α-acetoxycarboxylates (shown in Scheme
1) revealed that the mutant showed increased activities toward
most of the substrates (as shown in Table 2), with excellent
enantioselectivities (E > 200, data not shown). The influence
on the activity of the distance between the phenyl ring and the
acetoxy group was more significant for rPPE01-W187H than
for the wild type. The enzymatic activity of rPPE01-W187H
decreased sharply with increasing distance (1a < 1b < 1c), but
the influence of distance was negligible for the wild type. The
excellent enzymatic activity and enantioselectivity of rPPE01-
W187H toward a series of α-acetoxycarboxylates indicated that
rPPE01-W187H could act as a versatile biocatalyst for the
production of enantiopure α-hydroxy acids.
Figure 3. Enzymatic resolution of Ac-CPA at 500 mM, with different
salts, by rPPE01-W187H. Shown are conversions of substrate in the
form of sodium salt (black solid square), ammonium salt (blue solid
triangle), or potassium salt substrate (red dot) and the residual
enzyme activity with the substrate of sodium salt (black open square),
ammonium salt (blue open triangle), or potassium salt (red circle).
(Ser159, Asp256, and His286) and oxyanion hole shows two
regions of substrate binding sites, one for the acyl group and
the other for the alcohol group of the ester substrate. (S)-Ac-
CPA was docked into the binding pocket, and the most
favorable substrate-binding position was analyzed to identify
potential hot spots, as shown in Figure 1. The residues W187
and D287 were chosen as potential mutation sites on the basis
a
Table 3. Preparation of Important Enantiopure α-Hydroxy Acids with rPPE01-W187H
b
substrate
concn (mM, g L−1
)
cell load (g L−1
)
t (h)
conv (%)
(S)-hydroxy acid yield (%), ee (%)
(R)-hydroxy acid yield (%), ee (%)
rac-1a
rac-1b
rac-1e
500, 97.4
500, 104
500, 114
200, 45.6
200, 45.6
0.5
2.0
0.5
0.5
0.5
15
10
15
8
50.0
49.5
49.5
49.7
49.4
43.1, 98.9
44.5, >99.0
44.1, 98.8
43.4, >99.0
39.3, >99.0
41.1, 97.8
42.5, 97.7
42.4, 97.0
41.2, 97.8
40.1, 97.7
c
rac-1f
c
rac-1g
12
a
The enzymatic resolutions were conducted at a substrate concentration of 200−500 mM in the 50 mM KPB buffer at 100 mL scale. The reaction
mixtures were stirred by magnetic agitation, and the pH was automatically maintained at 6.5 by titrating 0.5 M K2CO3. The conversion and ee values
were analyzed by HPLC as described in the Experimental Section. The (R)-hydroxy acids were obtained after alkaline hydrolysis of the residual
(R)-O-acetate conducted in aqueous NaOH solution (20%, wt%) at room temperature. The reactions were conducted at 200 mM substrate
b
c
concentration owing to substrate inhibition.
1029
dx.doi.org/10.1021/cs401183e | ACS Catal. 2014, 4, 1026−1031

ACS Catalysis
Research Article
Table 4. Comparison of Esterase rPPE01-W187H with Other Reported Counterparts for Preparation of (R)-2-Hydroxy-2-(2′-
chlorophenyl)acetic acid
catalyst
CgKR1
substrate
substrate concn (g L−1
)
catalyst load(g L−1
)
conv (%)
ee (%)
P/C (g/g)
ref
a
c
o-chlorobenzoyl formate
o-chlorobenzoyl formate
o-chlorobenzoyl formate
o-chloromandelonitrile
Ac-CPA-Na
300
198
500
50.3
68.4
114
100
>99
98
98.7
>99
>99
96.0
>99
98.8
15
4
3
8
7
a
c
SCR
200
5.0
b
YtbE
50
>99
94.5
4.4
10
b
LaN
52
1.0
5.1
96
b
rPPE01
rPPE01-W187H
0.5
this study
this study
b
Ac-CPA-K
0.5
49.5
a
b
c
The wet cells were used as catalyst. The lyophilized cells were used as catalyst. The weight of the wet cells was divided by 5 to estimate the dry
weight for calculating the P/C ratio.
The catalytic properties of rPPE01-W187H were investigated
to explore its potential as a candidate biocatalyst for the up-
scaled enzymatic resolution of α-hydroxy acids. Both the pH−
activity and temperature−activity profiles of rPPE01-W187H
were different from those of rPPE01-WT. The optimal pH for
W187H was around 6.5, a slight shift from the pH of 6.2 for the
wild-type enzyme (SI Figure S1A). The mutant W187H
showed maximum activity at 40 °C and retained >92% of its
maximum activity over a broad temperature range from 30 to
50 °C (SI Figure S1B), whereas the wild-type rPPE01 showed
maximum activity at 50 °C, with a narrow optimum
temperature range. The broad temperature profile of rPPE01-
W187H allows enzymatic resolution to be conducted at 30 °C
(exhibiting 94.6% of its maximum activity at 40 °C), avoiding
enzyme deactivation as a result of high concentrations of
substrate and/or product at high temperature, as observed in
our previous study.14 rPPE01-W187H lost the majority of its
activity at 50 °C for 12 h, whereas its wild type retained 96% of
the initial activity under the same conditions (SI Figure S2),
However, no obvious inactivation was observed for either the
mutant or the wild type after incubations at 30 and 40 °C for
12 h.
To verify the improvements offered by the mutant W187H
in practical applications, biocatalytic resolution of Ac-CPA-Na
(300 mM) with either mutant or wild-type rPPE01 was
conducted with a catalyst loading of 0.5 g L−1 of lyophilized
cells. As shown in Figure 2, the reaction catalyzed by the
mutant W187H reached 49.6% conversion after 3 h, whereas
the conversion in the case of the wild type was only 4.4% after 3
h. Obviously, the mutant W187H was much more active than
rPPE01-WT, and its enantioselectivity was similar to that of the
parent (>99% eep) during the whole reaction process. These
results are in accordance with the kinetic data obtained, further
demonstrating that rPPE01-W187H is a more competent
biocatalyst for large-scale production of enantiopure α-hydroxy
acids.
counterions might affect the stability and activity of rPPE01-
W187H, and the optimum cation might serve as an inherent
additive, stabilizing the enzyme against substrate/product
deactivation. On the basis of the Hofmeister series22 of cations
and anions, we chose the potassium and ammonium salts to
provide potential substrate counterions instead of the sodium
salt previously used. Conversion of the substrate with
potassium ions reached ∼50% after 15 h, probably as a result
of decreased deactivation during the catalytic process, as
confirmed by the higher residual activity (Figure 3).
Ammonium cations reduced deactivation slightly, but con-
version of the substrate reached only 9.4% after 20 h. This
phenomenon might be explained by possible activity inhibition
by the ammonium salt in the reaction system. Generally, by
simply switching the substrate counterion from Na+ to K+, a
500 mM substrate (Ac-CPA-K) could be easily deacetylated
with excellent eep (>99%) and ees (98.7%) within 15 h,
affording satisfactory results at high substrate concentration.
To confirm the feasibility of the biotransformation process
on a large scale, several important α-hydroxy acids, that is, 2-
hydroxy-3-phenylpropanoic acid, 2-hydroxyphenylacetic acid,
and their derivatives, were prepared at the 100 mL scale under
the optimized reaction conditions. After the conversion reached
∼50%, the reaction mixture was acidified and extracted using
the normal workup; chromatographic separation gave enantio-
pure (R)-2-acetoxy acids and (S)-2-hydroxy acids. Finally, both
enantiomers of the hydroxy acids were obtained with high ee
values after further alkaline hydrolysis of the resultant (R)-2-
acetoxy compounds (Table 3).
CONCLUSION
■
In summary, combinatorial engineering approaches to the
biocatalyst and substrate counterion were successfully used to
tailor-make the “best” biocatalyst and substrate form for the
economical production of enantiopure α-hydroxy acids. A
semirational design based on analysis of structure−activity
relationships was effective in significantly improving the
esterase activity, providing a superior mutant with a 100-fold
increase in kcat/KM for Ac-CPA. Potassium ions as an inherent
additive to stabilize the enzyme against deactivation were found
to be the best substrate counterion for improving enzyme use.
Compared with other reported bioprocesses aimed at
producing optically active 2-hydroxy-2-(2′-chlorophenyl)acetic
acid (listed in Table 4), rPPE01-W187H-mediated enzymatic
resolution gave significantly higher catalyst productivity.
Accordingly, the cost of enzyme consumption could be greatly
reduced, and the emulsification caused by biocatalyst overload
could be avoided. The economically feasible and technologi-
cally efficient bioprocess described in this study demonstrates
The substrate concentration was then increased to 500 mM
at the same catalyst loading to further improve the substrate or
product loads and catalyst productivity, with a view to industrial
applications.15 A higher substrate or product concentration
would reduce the cost of product isolation to a large extent, and
a high catalyst productivity can reduce the cost of enzyme
consumption. A lower conversion (39.0%) was observed even
when the reaction time was prolonged to 20 h (as shown in
Figure 3); this was ascribed to poor substrate tolerance of the
esterase. A similar phenomenon was also observed for an
rPPE01-WT-catalyzed resolution reaction at a high substrate
concentration.14 In our study, the high concentration (500
mM) of substrate in the ionic form would result in a high ionic
strength (∼0.5 mol kg−1). We suspect that various substrate
1030
dx.doi.org/10.1021/cs401183e | ACS Catal. 2014, 4, 1026−1031

ACS Catalysis
Research Article
(17) (a) Kazlauskas, R. J.; Bornscheuer, U. T. Nat. Chem. Biol. 2009,
5, 526−529. (b) Bornscheuer, U. T.; Pohl, M. Curr. Opin. Chem. Biol.
2001, 5, 137−143. (c) Quin, M. B.; Schmidt-Dannert, C. ACS Catal.
2011, 1, 1017−1021.
the possibility of establishing an ideal bioprocess and solving
specific problems using goal-oriented engineering strategies.
ASSOCIATED CONTENT
* Supporting Information
■
(18) (a) Schmidt, M.; Hasenpusch, D.; Kahler, M.; Kirchner, U.;
̈
S
Wiggenhorn, K.; Langel, W.; Bornscheuer, U. T. ChemBioChem 2006,
7, 805−809. (b) Moore, J. C.; Arnold, F. H. Nat. Biotechnol. 1996, 14,
458−467.
Additional experimental results including the characterization
1
of the rPPE01-W187, H NMR spectra, and other data. This
(19) (a) Aehle, W.; Sobek, H.; Amory, A.; Vetter, R.; Wilke, D.;
Schomburg, D. J. Biotechnol. 1993, 28, 31−40. (b) Chica, R. A.;
Doucet, N.; Pelletier, J. N. Curr. Opin. Biotechnol. 2005, 16, 378−384.
(20) (a) Jochens, H.; Hesseler, M.; Stiba, K.; Padhi, S. K.; Kazlauskas,
R. J.; Bornscheuer, U. T. ChemBioChem 2011, 12, 1508−1517.
(b) Østerlund, T. Eur. J. Biochem. 2001, 268, 1899−1907.
(21) (a) Stepankova, V.; Bidmanova, S.; Koudelakova, T.; Prokop, Z.;
Chaloupkova, R.; Damborsky, J. ACS Catal. 2013, 28, 424−2836.
(b) Tran, D. N.; Balkus, K. J., Jr. ACS Catal. 2011, 1, 956−968.
(c) Madern, D.; Zaccai, G. Eur. J. Biochem. 1997, 249, 607−611.
(22) Iyer, P. V.; Ananthanarayan, L. Process Biochem. 2008, 43, 1019−
1032.
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
■
*Fax: (+86)-21-6425 0840. E-mail: huileiyu@ecust.edu.cn.
*Fax: (+86)-21-6425 0840. E-mail: jianhexu@ecust.edu.cn.
Notes
The authors declare no competing financial interest.
(23) Laskin, A. I. Advances in Applied Microbiology; Klibanov, A. M.,
Eds.; Elsevier Science Inc.: New York, 1983, Vol. 29, p 17−18.
(24) Bommarius, J. M.; Bommarius, A. S. J. Phys. Chem. B 2005, 109,
20612−20619.
(25) Basavaiah, D.; Krishna, P. R. Tetrahedron 1995, 51, 2403−2416.
(26) Sanchez, R.; Sali, A. Methods Mol. Biol. 2000, 143, 97−129.
(27) (a) Trott, O.; Olson, A. J. J. Comput. Chem. 2010, 31, 455−461.
(b) Seeliger, D.; de Groot, B. L. J. Comput. Aid. Mol. Des. 2010, 24,
417−422.
ACKNOWLEDGMENTS
■
This work was financially supported by the National Natural
Science Foundation of China (No. 21276082), Ministry of
Science and Technology, People’s Republic of China (Nos.
2011CB710800, 2011AA02A210, and 2012AA022201), Shang-
hai Commission of Science and Technology (No.
11431921600), and Huo Yingdong Education Foundation.
(28) Gao, L.; Xu, J.-H.; Li, X.-J.; Liu, Z.-Z. J. Ind. Microbiol. Biotechnol.
2004, 31, 525−530.
(29) Holmquist, M. Curr. Protein Pept. Sci. 2000, 1, 209−235.
REFERENCES
■
(1) Groger, H. Adv. Synth. Catal. 2001, 343, 547−558.
(2) Tang, L.; Deng, L. J. Am. Chem. Soc. 2002, 124, 2870−2871.
(3) Ema, T.; Ide, S.; Okita, N.; Sakai, T. Adv. Synth. Catal. 2008, 350,
2039−2044.
(4) Ma, H.-M.; Yang, L.-L.; Ni, Y.; Zhang, J.; Li, C.-X.; Zheng, G.-W.;
Yang, H.-Y.; Xu, J.-H. Adv. Synth. Catal. 2012, 354, 1765−1772.
(5) Xue, Y.-P.; Xu, M.; Chen, H.-S.; Liu, Z.-Q.; Wang, Y.-J.; Zheng,
Y.-G. Org. Process. Res. Dev. 2013, 17, 213−220.
(6) Ni, Y.; Pan, J.; Ma, H.-M.; Li, C.-X.; Zhang, J.; Zheng, G.-W.; Xu,
J.-H. Tetrahedron Lett. 2012, 53, 4715−4717.
(7) Zhang, Z.-J.; Pan, J.; Liu, J.-F.; Xu, J.-H.; He, Y.-C.; Liu, Y.-Y. J.
Biotechnol. 2011, 152, 24−29.
(8) Zhang, C.-S.; Zhang, Z.-J.; Li, C.-X.; Yu, H.-L.; Zheng, G.-W.; Xu,
J.-H. Appl. Microbiol. Biotechnol. 2012, 95, 91−99.
(9) (a) Adam, W.; Lazarus, M.; Boss, B.; Saha-Moller, C. R.; Humpf,
H.-U.; Schreier, P. J. Org. Chem. 1997, 62, 7841−7843. (b) Adam, W.;
Lazarus, M.; Schmerder, A.; Humpf, H.-U.; Saha-Moller, C. R.;
Schreier, P. Eur. J. Org. Chem. 1998, 2013−2018. (c) Miyazawa, T.;
Imagawa, K.; Yamada, T. Biocatal. Biotransform. 2006, 24, 291−298.
(d) Ma, J.-B.; Wu, L.; Guo, F.; Gu, J.; Tang, X.-L.; Jiang, L.; Liu, J.;
Zhou, J.-H; Yu, H.-W Appl. Microbiol. Biotechnol. 2013, 97, 4897−
4906.
(10) Jeromin, G. E.; Albertz, M. J. Parkt. Chem./Chem.-Ztg. 1992,
334, 526−528.
(11) (a) Bornscheuer, U. T. FEMS Microbiol. Rev. 2002, 26, 73−81.
(b) Kirk, O.; Borchert, T. V.; Fuglsang, C. C. Curr. Opin. Biotechnol.
2002, 13, 345−351.
(12) Campbell, R. F.; Fitzpatrick, K.; Inghardt, T.; Karlsson, O.;
Nilsson, K.; Reilly, J. E.; Yet, L. Tetrahedron Lett. 2003, 44, 5477−
5481.
(13) (a) Sugai, T.; Ohta, H. Agric. Biol. Chem. 1991, 55, 293−294.
(b) Chadha, A.; Manohar, M. Tetrahedron: Asymmetry 1995, 6, 651−
652.
(14) Ma, B.-D.; Yu, H.-L.; Pan, J.; Liu, J.-Y.; Ju, X.; Xu, J.-H. Bioresour.
Technol. 2013, 133, 354−360.
(15) Luetz, S.; Giver, L.; Lalonde, J. Biotechnol. Bioeng. 2008, 101,
647−653.
(16) Ju, X.; Yu, H.-L.; Pan, J.; Wei, D.-Z.; Xu, J.-H. Appl. Microbiol.
Biotechnol. 2010, 86, 83−91.
1031
dx.doi.org/10.1021/cs401183e | ACS Catal. 2014, 4, 1026−1031