Structure-guided evolution of carbonyl reductase for efficient biosynthesis of ethyl (: R)-2-hydroxy-4-phenylbutyrate

Article abstract of DOI:10.1039/d0cy01411g

Ethyl (R)-2-hydroxy-4-phenylbutanoate ((R)-HPBE) is an important versatile intermediate for the synthesis of angiotensin-converting enzyme inhibitors. Herein, a structure-guided rational design was adopted to improve the catalytic performance of carbonyl reductase from Gluconobacter oxydans (GoCR) for efficient production of (R)-HPBE at high substrate loading. To enhance the catalytic performance of GoCR, three sites (Cys93, Ile187 and Trp193) were identified based on a computational approach. Through single-site and cooperative mutation at these three sites, four variants with simultaneous increase in stereoselectivity and catalytic efficiency were obtained. Variants mut-W193L, mut-W193L/C93I, mut-W193L/I187L and mut-W193L/C93I/I187L exhibited 9.8-to 37.0-fold increase in catalytic efficiency (kcat/Km) compared to the wild-type enzyme. Meanwhile, the stereoselectivities of these variants were improved from 43.0% ee of wild-type GoCR to >99% ee. In addition, mut-W193L/C93I/I187L displayed improved thermostability simultaneously. Theoretical structural analysis revealed that the changes in the catalytic pocket microenvironment resulted in the concurrent improvement of enzyme activity and thermostability. In the batch production of (R)-HPBE, up to 371 g L-1 substrate loading was completely reduced by utilizing the most efficient variant mut-W193L/C93I/I187L at 40 °C, affording (R)-HPBE with >99% ee and a space-time yield of 540.4 g L-1 per day. This study provides a potential and attractive biocatalyst for the efficient synthesis of (R)-HPBE.

Full text of DOI:10.1039/d0cy01411g

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: W. Tang, L. Chen,
J. Deng, Y. Kuang, C. Chen, B. Yin, H. Wang, J. Lin and D. Wei, Catal. Sci. Technol., 2020, DOI:
1
0.1039/D0CY01411G.
Volume
Number
7
2
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
21
January 2017
Pages 313-534
Catalysis
Science &
Technology
rsc.li/catalysis
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2044-4761
PAPER
Chaoqiu Chen, Yong Qin et al.
Highly dispersed Pt nanoparticles supported on carbon
nanotubes produced by atomic layer deposition for hydrogen
generation from hydrolysis of ammonia borane
rsc.li/catalysis

Page 1 of 11
CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
View Article Online
DOI: 10.1039/D0CY01411G
ARTICLE
Structure-guided evolution of carbonyl reductase for efficient
biosynthesis of ethyl (R)-2-hydroxy-4-phenylbutyrate
a
a
a
a
b
c
a
a
Wen Tang , Lulu Chen , Jian Deng , Yuyao Kuang , Chao Chen , Bo Yin , Hualei Wang , Jinping Lin* ,
Received 00th January 20xx,
Accepted 00th January 20xx
a
Dongzhi Wei
DOI: 10.1039/x0xx00000x
Ethyl (R)-2-hydroxy-4-phenylbutanoate ((R)-HPBE) is an important versatile intermediate for the synthesis of angiotensin-
converting enzyme inhibitors. Herein, a structure-guided rational design was adopted to improve the catalytic performance
of carbonyl reductase from Gluconobacter oxydans (GoCR) for efficient production of (R)-HPBE at high substrate loading. To
enhance the catalytic performance of GoCR, three sites (Cys93, Ile187 and Trp193) were identified based on computational
approach. Through single-site and cooperative mutation at these three sites, four variants with simultaneous increase in
stereoselectivity and catalytic efficiency were obtained. Variant mut-W193L, mut-W193L/C93I, mut-W193L/I187L and mut-
m
W193L/C93I/I187L exhibited 9.8- to 37.0-fold increase in catalytic efficiency (kcat/K ) compared to wild-type enzyme.
Meanwhile, stereoselectivities of these variants were improved from 43.0% ee of wild-type GoCR to >99% ee. Besides, mut-
W193L/C93I/I187L displayed the improved thermostability simultaneously. Theoretical structural analysis revealed that the
changes of catalytic pocket microenvironment resulted in the concurrent improvement of enzyme activity and
thermostability. In the batch production of (R)-HPBE, up to 371 g/L substrate loading was completely reduced utilizing the
o
most efficient variant mut-W193L/C93I/I187L at 40 C, affording (R)-HPBE with >99% ee and a space-time yield of 540.4
g/(L·day). This study provides a potential and attractive biocatalyst for the efficient synthesis of (R)-HPBE.
An increasing number of carbonyl reductases and
dehydrogenases have been discovered in recent years to
synthesize (R)-HPBE, but most naturally occurring enzymes
Introduction
Chiral alcohols are important and useful building blocks for the
showed low catalytic activity, limited stereoselectivity, poor
stability, or poor tolerance for high substrate concentrations,
restricting their practical application 12-16. So far, only a few
examples of robust carbonyl reductases have been reported
that could catalyse OPBE at high substrate concentration. One
of the successful examples is a carbonyl reductase CgKR2 from
Candida glabrata which was found to completely reduce 206
g/L OPBE to (R)-HPBE with 99% ee 17. In addition, the
recombinant E. coli cells overexpressing a carbonyl reductase
IolS from B. subtilis and glucose dehydrogenase (GDH) were
employed in preparing (R)-HPBE with 99% ee and 100%
conversion at 330 g/L OPBE . However, these cases at present
still cannot meet the needs of practical production. These
indicate that developing and redesigning robust biocatalysts
and subsequently performing biocatalytic reduction process at
high substrate concentrations have always been an ongoing
subject.
synthesis of various agrochemicals, aroma chemicals, flavours,
food, pharmaceuticals and specialty materials 1-3. Among them,
ethyl (R)-2-hydroxy-4-phenylbutanoate ((R)-HPBE) is a versatile
key precursor for the synthesis of angiotensin-converting
4
enzyme (ACE) inhibitors, known as commercial “prils” . To date,
a series of methods have been developed for the preparation of
(
R)-HPBE, including chemical and enzymatic asymmetric
reduction of ethyl 2-oxo-4-phenyl-butyrate (OPBE) ,6, kinetic
5
7
,8
resolution of the corresponding racemate
esterification of 2-hydroxy-4-phenylbutanoic acid
chemical multi-step synthesis 0. In contrast, carbonyl
, enzymatic
9
and
1
1
1
reductase-mediated asymmetric reduction of OPBE to (R)-HPBE
has been an attractive method due to its significant advantages
such as high yield, high economic feasibility, environmental
benignity and under mild reaction conditions 4,11.
Protein engineering such as de novo enzyme design, structure-
guided (semi-) rational design and directed evolution, has been
employed to address these undesirable properties of enzymes
mentioned above and to remould the enzymatic property of
enzymes 18-21. With the rapid development of bioinformatics
data bases including computational chemistry and gene
sequence/three-dimensional structures of proteins, the
structure-guided (semi-)rational design strategy has been the
a. State Key Laboratory of Bioreactor Engineering, New World Institute of
Biotechnology, East China University of Science and Technology, No.130 Meilong
Road, Shanghai 200237, People's Republic of China. E-mail: jplin@ecust.edu.cn
b.
State Key Laboratory of Bioreactor Engineering, Biomedical Nanotechnology
Center, School of Biotechnology, East China University of Science and Technology,
Shanghai 200237, People's Republic of China.
National University of Singapore (Suzhou) Research Institute, No. 377 Linquan
c.
Street, Suzhou Industrial Park, Suzhou215123, Jiangsu, People's Republic of China.
Electronic Supplementary Information (ESI) available: [details of any supplementary
information available should be included here]. See DOI: 10.1039/x0xx00000x
Please do not adjust margins

CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Page 2 of 11
ARTICLE
Journal Name
preferred choice because of smaller mutant libraries and Site-directed mutagenesis was conducted using Vthieew ArmticuletaOtniloinne
reduced labor as well as process time. Regarding (semi-)rational primers (presented in Table S1) accordingDtOoI:t1h0e.10m39a/nDu0fCaYct0u14re11rG’s
design, various enzymes, including carbonyl reductases, have instructions of KOD-Plus-Mutagenesis Kit (Toyobo, Japan), and the
been successfully tailored to improve catalytic activity 22,23 or pET-28a (+) vector harbouring GoCR gene was used as template.
thermostability 24,25, to improve or invert stereoselectivity 6,23,26
and to broaden the scope of substrates accepted 23,27. As a
result, more efficient enzymes for chiral alcohol synthesis have
been designed and produced through structure-guided rational
design.
Previously, we identified a NAD(H)-dependent short-chain
dehydrogenase from Gluconobacter oxydans (GoCR), which
displayed a broad substrate spectrum, and in particular,
Subsequently, the recombinant plasmids harbouring GoCR or its
mutant gene were transformed into E. coli DH5α competent cells.
Afterwards, the colonies after transformation were selected for DNA
sequencing to obtain all designed variants. After that, the
recombinant plasmids harbouring designed variants were extracted
from E. coli DH5α cells and then transformed into E. coli BL21 (DE3)
competent cells for enzyme expression.
Protein expression and purification
exhibited the great potential in reduction of halogenated The E. coli BL21 (DE3) cells harbouring recombinant plasmids of GoCR
acetophenones to the corresponding (S)-alcohols with exclusive or its variants were cultured in 5 mL of LB medium containing 50
Prelog-selectivity (>99% ee) 26,28. To probe the molecular μg/mL kanamycin at 37°C. Subsequently, such overnight culture was
mechanism of stereoselectivity, we solved the crystal structure inoculated into 200 mL LB medium and cultured at 37°C. When the
of GoCR (PDB ID: 3WTC), and found Trp193 site played the optical density (OD600) of culture reached to around 0.6, isopropyl-β-
D-thiogalactopyranoside (IPTG) with 0.25 mM final concentration
was added to induce enzyme expression at 30°C for another 10 h.
After induction completion, the recombinant E. coli cells were
centrifuged at 4°C (6000 ×g, 15 min), washed with 0.9% NaCl solution
for several times and stored at -40°C.
critical role in determining stereoselectivity. Consequently, a
single-site mutant W193A was constructed by rational design,
and this mutant enzyme was proved to reduce OPBE to (R)-
HPBE with a significantly improved ee of >99% compared to 43.2%
for the wild type 26. Despite this success, the low activity of the
variant mut-W193A limited its application. Herein, we The collected recombinant E. coli cells were resuspended in PBS (100
mM, pH 7.4 ), and then disrupted through ultrasonication in an ice
water bath, followed by centrifugation (12000 ×g, 30 min, 4°C) to
collect supernatant. Subsequently, supernatant was filtered by 0.45
μm hydrophilic membranes, and then loaded onto a 5 mL HisTrap FF
crude column pre-equilibrated with 20 mM imidazole in PBS (100
mM, pH 7.4). The fractions were collected by gradient elution with
an increasing concentration (20-500 mM) of imidazole in PBS (100
mM, pH 7.4). Fractions containing the target protein were verified by
SDS-PAGE analysis and collected. The collections were dialyzed with
PBS (20 mM, pH 7.4) containing 200 mM NaCl and 5% (v/v)
glycerinum for desalting, and then concentrated by Millipore (3 KDa).
attempted to engineer GoCR via structure-guided semi-rational
design for improved catalytic performance in asymmetric
reduction of OPBE at high substrate loading. Computer-aided
structural analysis was also employed to understand the
potential mechanism for the improved performance of variants.
Finally, the production of (R)-HPBE employing E. coli cells co-
expressing GoCR variants and glucose dehydrogenase was also
evaluated.
Experimental
Materials
The purity of target protein was checked by SDS-PAGE. The enzyme
concentration was determined by the standard Bradford’s method 29.
Strains Genomic DNA of Gluconobacter oxydans strain 621H
was used as the template for PCR amplification of GoCR gene.
Escherichia coli (E. coli) BL21 (DE3) cells and pET-28a (+) vector
Enzyme activity assay, kinetic parameter and catalytic efficiency
determination
The enzyme activities of GoCR and its variants were determined by
monitoring the change in NADH absorbance at 340 nm using a
microplate reader (SpectraMax190, Molecular Devices, USA). The
standard assay mixture (0.2 mL) consisted of PBS (100 mM, pH 7.4),
(Novagen, Shanghai, China) were used for over-expression of
wild-type GoCR, its variants and glucose dehydrogenase.
Oligonucleotide primers used for mutagenesis were
synthesized at Sangni Biotech. (Shanghai, China). HPLC reagents
hexane, 2-propanol and chemicals ethyl 2-oxo-4-phenyl-
butyrate (OPBE) as well as ethyl (R)-2-hydroxy-4-
phenylbutanoate ((R)-HPBE) were obtained from Adamas
Reagent Co., Ltd. (Shanghai, China). Ethyl (S)-2-hydroxy-4-
phenylbutanoate ((S)-HPBE) was purchased from Yuanye Bio-
Technology Co. Ltd (Shanghai, China). Reduced β-nicotinamide
adenine diphosphate (NADH) and β-nicotinamide adenine
0
.2 mM NADH, 10 mM substrate and appropriate diluted enzyme. All
reactions except as specified were pre-mixed at 30°C and initiated by
adding NADH. One unit of enzyme activity was defined as the amount
of enzyme that catalyzes the oxidation of 1.0 μmol NADH per minute.
The kinetic parameters of purified GoCR and its variants towards
substrate OPBE were determined by measuring the activity at
different substrate concentrations (0.1-2.0 mM) with a fixed NADH
concentration of 0.2 mM. Initial velocity data obtained were fitted to
the Michaelis-Menten equation. The Michaelis constant K and the
m
maximum velocity Vmax were calculated by using Lineweaver-Burk
plots.
+
diphosphate (NAD ) were purchased from Jierui Reagent Co.,
Ltd. (Shanghai, China). Other chemicals involved were analytical
grade and commercially available.
Site-directed mutagenesis
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 11
CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Journal Name
ARTICLE
For comparison of stereoselectivity and catalytic activity of GoCR
A
one-plasmid co-expression system of GoCR VievwarAiratinclte Onalninde
variants, OPBE (0.5 M), D-glucose (0.6 M), lyophilized E. coli cells GDHE170K/Q252L was constructed using the priDmOeIr:s10li.s10te3d9/iDn0TCaYb0l1e41S12G.
harbouring GoCR variants (20 g/L), lyophilized E. coli cells harbouring Briefly, a thermostable glucose dehydrogenase gene (GDHE170K/Q252L
)
+
37,38
GDHE170K/Q252L (20 g/L), octanol (10%), NAD (0.5 mM) and phosphate
as constructed in vector pET-28a (+) (pET28a-GDH) using the
buffer (100 mM, pH 7.5) were combined. The reaction was carried primers listed in Table S2, entry 1, 2. Meanwhile, the gene (mutant)
o
out at 30 C and 200 rpm for 1 h. The conversion and stereoselectivity of GoCR mutant was cloned in vector pET-28a (+) (pET28a-mutant)
were determined by Gas Chromatography (GC) and High using the primers listed in Table S2, entry 3, 4. Subsequently, the
Performance Liquid Chromatography (HPLC), respectively. For easy GDHE170K/Q252L gene was amplified by PCR using primers with pET28a-
comparison purpose, conversion of the control (mut-W193L) was set GDHE170K/Q252L as a template and cloned into Hind III/Xhol I sites of
as 100%.
vector pET28a-mutant (Table S2, entry 5, 6). As a result, genes
mutant and GDHE170K/Q252L were arranged in tandem in one-plasmid
pET28a (+), which was under the control of a separate T7 promoter
and terminator, respectively.
Circular dichroism spectroscopy studies
Secondary structural analysis by circular dichroism (CD) spectra at
different temperatures can be used to provide information about
enzyme stability . The CD spectra of GoCR and its variants were
3
0
Asymmetric reduction of OPBE
measured by a Chirascan CD spectrometer (Applied Photophysics The whole-cell bioreduction of OPBE was conducted in a two-liquid
Ltd., Leatherhead, UK) equipped with a TC125 temperature-control (octanol : PBS = 15:85, v/v) 10-mL biphase reaction system which
system (Quantum Northwest, Liberty Lake, WA, USA). The unfolding contained octanol (15%), PBS (100 mM, pH 7.5), OPBE (1.5-1.8 M), D-
curves were measured from 200 to 260 nm against gradient glucose (1.2 equiv.), lyophilized E. coli cells co-expressing GoCR
+
temperatures, using the temperature scan mode with a gradient of variant and GDHE170K/Q252L (13.8-55 g/L) and NAD (0.5 mM). The
1
°C/min. The purified enzyme was diluted to 1.0-1.5 mg/mL in 20 reaction temperature was kept at 30°C or 40°C, and the pH was
mM sodium phosphate buffer (pH 7.4). The melting temperature (T adjusted to 7.5 by titrating Na CO
m
)
2
3
.
values were calculated by Global 3 analysis software (Applied
Photophysics, UK).
Samples were withdrawn periodically, and extracted with ethyl
acetate (EtAc). Centrifugation (12,000 × g, 10 min) was carried out to
facilitate the separation of two phases. The organic layer was
Computational analysis
According to previous studies 31-34, we chose the conformation of
2 4
dehydrated using anhydrous Na SO and then subjected to GC
analysis (Agilent 7820A system) with the HP-5 column to determine
the conversion and yield. Otherwise, the samples were heated to
remove EtAc, re-dissolved in isopropanol and then filtered with filter
membranes (0.22 μm) for chiral HPLC analysis with chiral column OD-
H for determination of enantiomeric excess (ee) value under the
following conditions: detection wavelength of 210 nm, flow rate of
1.0 mL/min, n-hexane : isopropanol volume ratio of 98:2. The
absolute configuration of products was compared according to
enzyme-substrate complex after energy minimization to calculate
the binding free energy. Here, considering the reason that reaction
enthalpy that is presumably driven by hydrogen bonding interaction
between enzyme and substrate is the major thermodynamic factor,
we used relative binding free energy (without entropy) to assess the
relative importance of different amino acid residues in substrate
3
5
binding, which had also been used in the previous study
.
Specifically, the relative binding free energy (△Gbind in kcal/mol) of
mut-W193L-OPBE complex after energy minimization was estimated
by the Molecular Mechanics Generalized Born Surface Area (MM-
GBSA) module implemented in AmberTools 14. Subsequently, the
per-residue binding free energy decomposition was performed to
identify the potential residues responsible for substrate binding
process.
1
1,26,39
authentic standards or literature reports
. The retention time
of (S)- and (R)-HPBE was 12.7 min and 18.4 min, respectively (Fig. S1).
The HPLC detection limit (DL) of (R)- and (S)-HPBE were 0.03 and 0.02
mM, and their quantitation limit (QL) were 0.09 and 0.08 mM,
respectively. The product from reduction of OPBE by mut-
1
W913L/C93I/I187L was confirmed by H NMR (CDCl
3
, 500 Hz): δ=1.38
CHH), 2.17-2.27 (1H, m;
), 3.16 (1H, s; OH), 4.25-4.36 (3H,
, CHOH) , 7.25-7.42 (5H, m; Ar-H) (Fig. S2).
(
3H, t; OCH
ArCH CHH), 2.80-2.93 (2H, m; ArCH
m; OCH CH
2 3 2
CH ), 1.99-2.10 (1H, m; ArCH
The virtual variants (mut-W193L, mut-W193L/C93I, mut-
W193L/I198L and mut-W193L/C93I/I187L) of GoCR (PDB ID: 3WTC)
were firstly constructed by free PyMoL-edu. Then, molecular docking
between GoCR or its variants and OPBE was conducted by the
AutoDock1.5.6 program. The detailed parameters were set according
2
2
2
3
Results and discussion
Computer-aided mutation of carbonyl reductase GoCR
2
6
to our previous study except the following modification . In
particular, the substrate OPBE, Cys93, Ile187 and catalytic tetrad
(
Asn113-Ser142-Tyr155-Lys159) were set as flexible, while the other Our previous study revealed that residue Trp193 located in the
residues in variants were set as rigid. The enzyme-substrate substrate-binding pocket plays a critical role in determining the
conformation with the lowest binding energy was further analyzed stereoselectivity of carbonyl reductase GoCR via a computational
2
6
by PyMoL-edu. Besides, the interactions of residues at positions 93, strategy
87 and 193 site with other residues in GoCR or its variants were stereoselectivity and activity were analysed in detail by
studied through the RING 2.0 web server
http://protein.bio.unipd.it/ring/)
. Herein, residue Trp193 that might affect the
1
3
6
(
.
Co-expressing of GoCR variant and glucose dehydrogenase in E. coli
cells
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Page 4 of 11
ARTICLE
Journal Name
Val188, Met192, Leu193, Ile196, Lys213, Lys214, Tyr215, Val216,
Leu256) in the vicinity of 4 Å from the substrate OPBE were selected
View Article Online
DOI: 10.1039/D0CY01411G
Table 1. Stereoselectivity and specific activity of wild-type GoCR
and its variants at 193 site.
for the per-residue binding free energy calculated by the MM-GBSA
method to discern the ‘hotspots’ in mut-W193L. According to the
reported study, binding free energy values △Gbind >-1.0 kcal/mol was
ee value
Specific activity
U/mg)
Improved
fold
Mutants
(
(
%, R)
43.0
99.3
99.1
99.2
99.1
96.3
98.6
98.2
97.9
98.4
98.3
97.7
98.5
98.6
98.5
4
0
identified as major contributors to the substrate binding
.
WT
3.7 ± 0.3
16.3 ± 0.6
18.4 ± 0.7
14.0 ± 1.1
14.7 ± 0.5
13.0 ± 0.8
3.8 ± 0.2
2.4 ± 0.5
9.3 ± 0.8
4.7 ± 0.3
5.9 ± 1.1
1.3 ± 0.2
3.1 ± 0.5
4.3 ± 0.7
4.8 ± 0.3
1
Accordingly, five residues (Cys93, Ser142, Tyr155, Ile187 and Leu193)
were chosen as candidate sites for affecting substrate binding due to
relatively high binding free energy values (△Gbind >-1.0 kcal/mol) (Fig.
W193A
W193L
W193I
W193V
W193F
W193M
W193N
W193P
W193Q
W193S
W193T
W193G
W193Y
W193C
4.4
5.0
3.8
4.0
3.5
1.0
0.6
2.5
1.3
1.6
0.4
0.8
1.2
1.3
1
b). Among these residues, Ser142 and Tyr155 are significant
catalytic residues in the conserved catalytic tetrad (Asn113-Ser142-
Tyr155-Lys159) of mut-W193L, and the Leu193 site has been
modified previously. These three amino acid residues were therefore
excluded from mutagenesis. Therefore, another two residues (Cys93
and Ile187) in the substrate-binding pocket (Fig. 1c) were chosen as
target sites for saturation mutagenesis and subsequent combination
of beneficial mutations.
After being expressed in E. coli BL21 (DE3), all variants were similar
to mut-W193L in protein expression, indicating that residue
substitutions of 93 site and 187 site did not affect protein expression
(Fig. S4 and S5). After screening of the saturation mutation library at
9
3
site, eight double-site variants (mut-W193L/C93I, mut-
W193L/C93L,
W193L/C93Q,
mut-W193L/C93F,
mut-W193L/C93T,
mut-W193L/C93S,
mut-W193L/C93Y,
mut-
mut-
W193L/C93G) exhibiting 1.1- to 2.2-fold improvements in catalytic
activity were identified compared with single-site variant mut-
W193L, indicating 93 residue could affect enzyme activity (Fig. 2a).
semisaturation mutagenesis. Considering that the substrate-binding For the mutants at 187 site, four hits (mut-W193L/I187F, mut-
pocket of GoCR is electrically neutral, a smart mutant library of non- W193L/I187L, mut-W193L/I187S, mut-W193L/I187V) exhibited
charged amino acid substitutions at 193 site was constructed and obvious enhancement in catalytic activity relative to single-site
screened for concurrent improvements in stereoselectivity and variant mut-W193L, which revealed that 187 site was another
enzyme activity. SDS-PAGE analysis showed that all variants were hotspot for affecting enzyme activity (Fig. 2b). These results further
expressed in soluble form in E. coli cells with molecular weights (MW) verified our above prediction that 93 and 187 sites could obviously
about 28 kDa, similar to the wild-type GoCR (Fig. S3). As shown in affect the interaction between enzyme and substrate. Among them,
Table 1, all variants in the mutant library exhibited significantly
improved stereoselectivity compared to wild-type GoCR. Among
2
6
them, in addition to the previously reported mut-W193A , three
variants, W193L (named as mut-W193L), W193I (mut-W193I) and
W193V (mut-W193V), also showed the highest stereoselectivity of
>99%. Moreover, it was noted that Trp193 residue was also
identified as a potential ‘hotspot’ for improving enzyme activity.
Most variants exhibited higher activity than the wild-type GoCR,
especially the variant mut-W193L had the highest enzyme activity of
1
8.4 U/mg, exhibiting a 5.0-fold improvement compared to the wild-
type GoCR. Taken stereoselectivity and enzyme activity into
consideration, mut-W193L was chosen to further improve its
catalytic efficiency towards OPBE.
Fig. 1. Residues chosen for mutation of mut-W193L. (a)
Residues (in orange) within the substrate-binding pocket of
mut-W193L within a 4 Å vicinity of the substrate OPBE. (b) Per-
residue binding free energy calculated by the MM-GBSA
method for mut-W193L-OPBE complex. Only negative △Gbind
value of residues contributing to the binding of OPBE are
displayed. (c) The candidate sites (Cys93 and Ile187) in
substrate-binding pocket of mut-W193L for mutation.
The computational method including molecular docking and
Molecular Mechanics Generalized Born Surface Area (MM-GBSA)
method was used to identify potential residues for improving
enzyme activity of mut-W193L. As shown in Fig. 1a, molecular
docking analysis revealed that the substrate OPBE was accurately
located at the substrate-binding pocket of mut-W193L, which further
confirmed that OPBE could be reduced by mut-W193L.
Subsequently, 22 residues (Cys93, Ser142, Ile143, Ala144, Gly145,
Glu147, Gly148, Tyr149, Leu152, Tyr155, Pro185, Gly186, Ile187,
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 11
CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/D0CY01411G
Fig. 2. Relative activities of mutants as compared to that of the mut-W193L. (a) C93 site mutants, (b) I187 site mutants and (c) optimal
double-site variants and three-site variant.
the best double-site variants mut-W193L/C193I and mut- mut-W193L/I187L)
W193L/I187L exhibited 2.2 and 1.6 times as catalytic activity as the W193L/C93I/I187L) exhibited a progressive increase in catalytic
mut-W193L, respectively. Cooperative mutational and additivity efficiency (kcat/K ) compared to GoCR. Moreover, the enhancement
effects are often adopted to further improve catalytic activity when in kcat/K was attributed to the decrease in K and the increase in
simplifying protein engineering 41. Hence, the mut-W193L/C93I and
cat. It was noted that the introduction of Leu substitution of Trp193
and
triple-site
mutation
(mut-
m
m
m
k
mut-W193L/I187L were attempted to combine for increasing made an important contribution to the increased activity and the
activity. As expected, variant mut-W193L/C93I/I187L displayed a 1.2- increased catalytic efficiency, affording a 5.0-fold increase in specific
and 1.6-fold further increase in catalytic activity in comparison to activity and 9.8-fold increase in kcat/K toward OPBE in comparison
m
mut-W193L/C93I and mut-W193L/I187L, respectively (Fig. 2c). The with wild-type GoCR (Table 2, entry 1, 3). Subsequently, compared
stereoselectivity of the best variant mut-W193L/C93I/I187L toward with the starting mut-W193L, mut-W193L/C93I and mut-
OPBE remained >99% ee.
W193L/I187L displayed 3.6- and 1.9-fold further increase in kcat/K
respectively (Table 2, entry 3, 4, 5). The enhanced kcat/K of mut-
W193L/C93I thus improved specific activity mainly benefit from the
decreased K and increased kcat. The combinational mutations of
m
,
m
Analysis of mutants with increased catalytic activity from kinetics
and structural perspective
m
To comparatively evaluate catalytic efficiency and elucidate the
reason for the increase in activity from kinetics perspective, the
specific activities and kinetic parameters of the purified wild-type
GoCR and its variants with increased catalytic activity were
determined (Table 2 and Fig. S6). In comparison with the wild-type
GoCR, all variants exhibited a great improvement in specific activity
toward OPBE (Table 2), which was consistent with the increased
catalytic activities of these variants (Fig. 2). Correspondingly, single-
site mutation (mut-W193L), double-site mutation (mut-W193L/C93I,
W193L/C93I with W193L/I187L resulted in clearly additivity effects
on activity. The resulting variant (mut-W193L/C93I/I187L) exhibited
the highest specific activity and the most enhanced kcat/K
m
value,
which mainly benefited from a decrease in K value together with an
m
increase in kcat value (Table 2, entry 6). Furthermore, mut-
W193L/C93I/I187L exhibited 5.3-fold higher catalytic efficiency
compared to that of reported mut-W193A (Table 2, entry 2, 6), due
m
to the decreased K and increased kcat. In general, mut-
W193L/C93I/I187L exhibited a 14.0-fold enhancement in specific
activity and a 37.0-fold increase in kcat/K
GoCR (Table 2, entry 1, 6).
m
compared to wild-type
Table 2. Specific activity and kinetic parameters of GoCR and its
variants towards OPBE.
Moreover, the mechanism for the enhancement of catalytic activity
regarding amino acid substitutions was also explored at atomic level.
Thus, molecular docking between GoCR or its variants and substrate
was performed. According to the carbonyl reductase GoCR reaction
mechanism, catalytic residue Tyr155 and NADH donated their
hydrogen atoms to the carbonyl oxygen and stereogenic carbon
atom of substrate OPBE, respectively, resulting in the reduction of
OPBE (Fig. 3a). Thus, the Distance 1 was defined as the distance
between the stereogenic carbonyl carbon atom of OPBE and a
Specific
activity
En
tr
y
K
m
k
cat
kcat/K
m
Mutants
-1
-1 -1
(
mM)
(s )
(mM s )
(
U/mg)
1
2
3
4
GoCR
3.7 ± 0.3
0.87 ± 0.08
0.53 ± 0.05
0.46 ± 0.04
0.27 ± 0.07
1.96 ± 0.43
8.29 ± 1.35
10.10 ± 1.5
21.21 ± 2.84
2.25
W193A
16.3 ± 0.6
18.4 ± 0.7
43.5 ± 2.2
15.64
21.96
78.56
W193L
4
hydrogen from the C atom of the nicotinamide ring of NADH. The
Distance 2 was defined as the distance between the carbonyl oxygen
atom of OPBE and the hydroxyl group of Tyr155. Both Distance 1 and
Distance 2 played key roles in determining the catalytic activity of
enzyme towards the substrate OPBE. For GoCR, the Distance 1 and
Distance 2 were 4.4 Å and 3.8 Å, respectively (Fig. 3b and Table S3,
entry 1), indicating the stable catalytic conformation was not easily
achieved. After mutation of Trp at 193 site to Leu, the Distance 1 and
W193L/C9
3
I
W193L/I1
7L
W193L/C9
I/I187L
5
6
33.9 ± 1.6
51.7 ± 2.9
0.44 ± 0.04
0.31 ± 0.03
18.52 ± 1.72
25.78 ± 2.11
42.09
83.16
8
3
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Page 6 of 11
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/D0CY01411G
Fig. 3. (a) Proposed catalytic mechanism of GoCR and its variants for the reduction of OPBE; Molecular docking of OPBE into the active
sites of GoCR and its variants towards OPBE. (b) GoCR, (c) mut-W193L, (d) mut-W193L/C93I, (e) mut-W193L/I187L and (f) mut-
W193L/C93I/I187L.
Distance 2 was shorten to 3.8 Å and 3.2 Å, respectively (Fig. 3c and Given that there is sometimes a trade-off between enzymatic activity
3
0,42
Table S3, entry 2). Subsequently, compared with mut-W193L, the and enzyme stability
, the thermostabilities in terms of melting
Distance 1 in mut-W193L/C93I, mut-W193L/I187L and mut- point (T ) of GoCR and four variants (mut-W193L, mut-W193L/C93I,
m
W193L/C93I/I187L was obviously shortened (Fig. 3d, 3e, 3f and Table mut-W193L/I187L and mut-W193L/C93I/I187L) were determined by
S3, entry 3-5), which benefited nucleophilic attack on stereogenic CD measurements. Fig. S7 and Fig. S8 showed that GoCR and its
carbonyl carbon atom of OPBE. Meanwhile, the Distance 2 in mut- variants exhibited double minimal negative peaks at approximately
W193L/C93I, mut-W193L/I187L as well as mut-W193L/C93I/I187L 208 nm and 222 nm, implying abundant α-helixs in GoCR and its
were also shortened to within 2.9 Å, which was beneficial for the variants, which are consistent with the crystal structure of GoCR (PDB
2
6
reduction of carbonyl oxygen atom of OPBE. Additionally, Ser142 can ID: 3WTC) . As the temperature increased, the CD values at 208 and
stabilize substrate or transition state of intermediate through 222 nm for the tested variants gradually decreased, indicating the
hydrogen bond (H-bond) between carbonyl oxygen atom of OPBE unfolding of protein increased, which was due to the destroyed
and hydroxyl groups of Ser142 (Fig. 3a). The GoCR could not form a secondary structure. The T
stable H-bond with substrate OPBE because the Distance 3 between W193L/I187L and mut-W193L/C93I/I187L were 62.4, 60.3 and
carbonyl oxygen atom of OPBE and hydroxyl groups of Ser142 was 70.7°C, respectively, higher than the mut-W193L (T =58.8°C) (Fig. 4a
.5 Å, which was longer than 3.5 Å that is the threshold for forming and 4b). Thus, both thermostability and catalytic activity of the
m
values of mut-W193L/C93I, mut-
m
5
a stable H-bond (Fig. 3b and Table S3, entry 1). Furthermore, for mut- engineered mut-W193L/C93I, mut-W193L/I187L and mut-
W193L, the stable H-bond between Ser142 and OPBE still could not W193L/C93I/I187L were improved compared to mut-W193L.
be formed (Fig. 3c and Table S3, entry 2), thus the stable transition Besides, although the thermostability of mut-W193L and mut-
states of intermediate could not occur. After further mutation at 93
site and 187 site, such distance in all three mutants was shortened to
within 3.3 Å (Fig. 3d, 3e, 3f and Table S3, entry 3-5), therefore stable
H-bond could be formed which could stabilize the transition states of
intermediate. It was also part of main reason for the increase in
catalytic activity. Therefore, mutants at the 93 site and 189 site (mut-
W193L/C93I, mut-W193L/I187L, mut-W193L/C93I/I17L) chosen by
structural-guide analysis significantly improved enzyme catalytic
activity by shortening the distances of OPBE with NADH and with key
Fig. 4. CD analysis of GoCR and its variants (a) Thermal
catalytic residues in substrate-binding pocket.
denaturation of GoCR and its variants by CD spectra; (b) Melting
Thermostability analysis of GoCR variants
temperature (T
m
) of GoCR and its variants.
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 11
CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Journal Name
ARTICLE
W193L/I187L had slightly decreased and mut-W193L/C93I did not formed by Ile187 with Tyr212 located on helix α8 (Fig. 5a and 5d).
View Article Online
o
DOI: 10.1039/D0CY01411G
change a lot compared to wild-type GoCR (T
W193L/C93I/I187L exhibited obviously increased thermostability stability of flexible loop8 and loop12 surrounding the active site
Fig. 4b). It is worth noting that mut-W193L/C93I/I187L with the residues, which might account for the significantly increased melting
m
= 62.1 C), mut- Together, these additional van der Waals interaction increased the
(
highest enzyme activity showed the highest thermostability. The T
value of mut-W193L/C93I/I187L was enhanced by 8.6 C compared to W193L/C93I/I187L in comparison with the mut-W193L. Similar
m
temperature of mut-W193L/C93I, mut-W193L/I187L and mut-
o
4
1
the wild-type GoCR. As we known, in addition to the increased phenomenon also could be observed in Gong’s study . Besides, the
catalytic activity, excellent thermostability of enzyme is also increased thermostability of mut-W193L/C93I/I187L in comparison
preferable in large-scale industrial processes, because higher with wild-type GoCR might be due to the further enhanced stability
reaction temperature often effectively lowered the risk of of flexible loop8 (Fig. 5d and Fig. S9), which resulted from the new
contamination, improved mass transfer rate and promoted the van der Waals interaction formed by Ile93 and Ile196 located on helix
4
3
solubility of poorly dissolved substrates . In the past few decades, α7.
numerous successful examples have been reported for improving the
Synthesis of (R)-HPBE using engineered variants
thermostability of enzymes by directed evolution or (semi-) rational
4
3-45
To further evaluate the feasibility of biocatalytic process at high
substrate loading, asymmetric reduction of OPBE to (R)-HPBE was
performed using lyophilized recombinant E. coli cells overexpressing
GoCR variants (mut-W193L, mut-W193L/C93I, mut-W193L/I187L or
mut-W193L/C93I/I187L) and GDHE170K/Q252L for the regeneration of
NADH in situ (Fig. 6a and Fig. S10). The variants mut-W193L/C93I,
mut-W193L/I187L and mut-W193L/C93I/I187L continued to display
pronounced catalytic activity towards OPBE compared to mut-
W193L which exhibited low conversion of about 68.5% within 22 h
design
. In these studies, the mutation sites selected are usually
far away from the active pocket or have no direct interaction with
the substrate. In our study, the best variant mut-W193L/C93I/I187L
with significant improvement in both catalytic efficiency and
thermostability was obtained by engineering the amino acid residues
in the substrate-binding pocket, which had direct interaction with
substrate.
In order to elucidate the mechanism of increased thermostability,
the major intramolecular interactions, such as the Van der Waals’s
interactions, salt bridges, hydrogen bonds and disulphide bond that
are considered as the dominant structural factors responsible for
protein thermostability, were analyzed using the RING 2.0 web
(Fig. 6b and Table 3, entry 3). Among these variants, mut-
W193L/C93I/I187L displayed the highest reaction rate towards 1.5 M
OPBE, completing conversion within 12 h, followed by mut-
W193L/C93I within 16 h. The mut-W193L/I187L also converted over
4
6
server . As shown in Fig. 5, 93 and 187 sites are located on flexible
loop8 and loop12 surrounding the active site residues, respectively.
Compared with the model structure of mut-W193L, mut-
W193L/C93I revealed that two new van der Waals interaction were
formed by Ile93 and Ile196 located on helix α7, and by Ile187 with
Leu193 located on helix α7 (Fig. 5a and 5b). Additional I187L
mutation in mut-W193L formed a new van der Waals interaction by
Ile187 with Leu193 located on helix α7 (Fig. 5a and 5c). All the above
van der Waals interactions were observed in mut-
W193L/C93I/I187L, besides, another new van der Waals interaction
9
9% of 1.5 M OPBE within 22 h (Fig. 6b and Table 3, entry 4-6). The
specific activity of mut-W193L/C93I/I187 was the highest among the
GoCR variants; it is therefore reasonable that its catalytic efficiency
were higher than those of the other three variants in the trial.
Moreover, mut-W193L/C93I/I187 also exhibited higher catalytic
efficiency than the reported mut-W193A, which was consistent with
the above results of specific activity (Table 3, entry 2, 6). Although all
variants exhibited different catalytic efficiency, they displayed
Fig. 6. Time courses of reduction of OPBE by variants. (a)
Representative asymmetric reduction of OPBE by variants; (b)
Time course of reduction of 1.5 M OPBE by mut-W193L, mut-
W193L/C93I, mut-W193L/I187L or mut-W193L/C93I/I187L;
(
c) Time courses of reduction of 1.8 M OPBE by mut-
Fig. 5. Local residue interactions of GoCR variants predicted by the
RING 2.0 web server. (a) mut-W193L, (b) mut-W193L/C93I, (c) mut-
W193L/I187L and (d) mut-W193L/C93I/I187L. The hydrogen bonds
and Van der Waals interactions are indicated by light blue and gray
lines, respectively.
W193L/C93I/I187L at different temperatures (30°C and 40°C).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Page 8 of 11
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/D0CY01411G
Table 3. Asymmetric reduction of OPBE catalyzed by GoCR or its variants.
S/C ratio (g/g)d
Temperature ( C) Time(h)
o
Conversion (%)
ee (%)
Entry
Biocatalyst
OPBE (M)
1
2
3
4
5
6
7
WTa
1.5
1.5
1.5
1.5
1.5
1.5
1.5
5.6
30
30
30
30
30
30
30
20
24
22
22
16
12
18
9.7
39.2
68.5
>99
>99
>99
>99
43.0
99.2
99.3
99.4
99.3
99.4
99.2
W193A
11.2
11.2
11.2
11.2
11.2
22.4
W193La
W193L/I187La
W193L/C93Ia
W193L/C93I/I187La
W193L/C93I/I187Lb
a
Reaction condition: OPBE (309 g/L), D-glucose (1.2 equiv.), lyophilized E. coli cells co-expressing GDHE170K/Q252L and GoCR (55 g/L) or its
+
o
variants (27.5 g/L), octanol (15%) and NAD (0.5 mM). The reaction temperature was kept at 30 C.
b
Reaction condition: OPBE (309 g/L), D-glucose (1.2 equiv.), lyophilized E. coli cells co-expressing GDHE170K/Q252L and mut-W193L/C93I/I187L
+
o
(
13.8 g/L), octanol (15%) and NAD (0.5 mM). The reaction temperature was kept at 30 C.
c
Reaction condition: OPBE (371 g/L), D-glucose (1.2 equiv.), lyophilized E. coli cells co-expressing GDHE170K/Q252L and mut-W193L/C93I/I187L
+
o
(
13.8 g/L), octanol (15%) and NAD (0.5 mM). The reaction temperature was kept at 30 or 40 C.
d
S/C: substrate/catalyst ratio
8
9
W193L/C93I/I187Lc
W193L/C93I/I187Lc
1.8
1.8
26.9
26.9
30
40
20
14
>99
>99
99.1
99.2
excellent stereoselectivity toward OPBE and afforded (R)-HPBE with result highlighted the considerable operational stability of mut-
99% ee (Table 3, entry 2-6). W193L/C93I/I187L under the reaction conditions, which is in
agreement with its high T value. To sum up, several mutants with
>
m
Although the variant mut-W193L/C93I/I187L displayed excellent
activity, the observed substrate/catalyst (S/C) ratio of 11.2 g/g was
still low (Table 3, entry 6). When the cell concentration was reduced
to 13.8 g/L of lyophilized E. coli cells, affording the increased S/C ratio
of 22.4 g/g, the reaction at a substrate loading of 1.5 M was
accomplished completely within 18 h by mut-W193L/C93I/I187L (Fig.
excellent catalytic efficiency were successfully constructed through
the semi-rational strategy. As a result, the catalyst loading S/C ratio
(gsubstrate/gcatalyst) was dramatically increased from 5.6 to 26.9 g/g, and
the substrate concentration which could be completely reduced by
the best mut-W193L/C93I/I187L was significantly enhanced to 1.8 M
371 g/L), resulting in >99% ee and 540.4 g/(L·day) space-time yield.
(
6
b and Table 3, entry 7). Excitingly, even at a higher substrate loading
It is noteworthy that a substrate loading with >99% conversion in this
study reached 371 g/L, which was higher than the highest record
reported so far for the reduction of OPBE to produce (R)-HPBE (Table
S4). These results showed the excellent catalytic activity and
potential application of variant mut-W193L/C93L/I187L for
production of (R)-HPBE.
of 1.8 M, a complete conversion was also achieved in 20 h with the
catalyst loading of 26.9 g/g (Fig. 6b and Table 3, entry 8).
Given that thermostability of mut-W193L/C93I/I187L was increased
significantly, and the half-life of mut-W193L/C93I/I187L at 40°C was
1
9.3 h (Fig. S11), we attempted to carry out the reduction of OPBE by
mut-W193L/C93I/I187L at higher temperature of 40°C, instead of
0°C, in order to shorten the reaction time. As shown in Fig. 6c and
Table 3, entry 8, 9, 1.8 M OPBE could be completely converted at
Substrate spectrum of mut-W193L/C93I/I187L
3
4
0°C within the shorter time of 14 h compared to 20 h at 30°C. This
Table 4. Carbonyl compounds with different molecular volumes catalyzed by mut-W193L/C93I/I187L.
Entry
Substrate
Volume (Å)
Conv.(%)
ee (%)
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 11
CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Journal Name
ARTICLE
a
View Article Online
DOI: 10.1039/D0CY01411G
1
2
2'-Fluoroacetophenone
127.18
>99
>99 (S)
1
43.31
>99a
>99 (S)
2'-Bromoacetophenone
1
37.62
>99a
>99 (S)
3
4
4'-Chloroacetophenone
Ethyl benzoylformate
1
1
78.53
96.37
>99b
>99a
>99 (R)
>99 (S)
5
3',5'-Bis(trifluoromethyl)acetophenone
ethyl 2-Oxo-4-phenyl-butyrate
2
1
3
10.04
70.58
42.82
>99a
>99 (R)
6
7
N.A.b,c
N.A.b,c
--
--
2-Oxo-4-phenylbutyric acid
(
4S)-3-[5-(4-fluorophenyl)-1,5-dioxophentyl]-4-phenyl-1,
8
3
-oxazolidin-2-one
a
b
c
1
1
00 mM substrate concentrate.
0 mM substrate concentrate.
No measurable activity.
In order to further evaluate the application potential of mut- activity and stereoselectivity toward OPBE compared with the wild-
W193L/C93I/I187L, we explored a panel of substrates and found that type enzyme were successfully obtained using single-site mutation
mut-W193L/C93I/I187L displayed a broad substrate spectrum, and cooperative mutation. The variants mut-W193L, mut-
notably exhibiting higher reduction activity towards α-diketones W193L/C93I, mut-W193L/I187L and mut-W193L/C93I/I187L showed
(Table S5). Compared with wild-type GoCR, mut-W193L/C93I/I187L the excellent stereoselectivities with >99% ee compared to 43.0% of
showed a general improvement in catalytic efficiency toward the the wild-type GoCR. Among these variants, mut-W193L/C93I/I187L
tested substrates, with the exceptions of 2, 3-Butanedione and 2- showed the highest specific activity and catalytic efficiency, which
Pentanone. In addition, we observed that mut-W193L/C93I/I187L were enhanced by 14.0- and 37.0-fold, respectively, compared to the
tended to reduce the ketones with bulky aromatic side chains. Thus, wild-type GoCR. It was noted that the simultaneous enhancement in
mut-W193L/C93I/I187L can serve as an attractive and versatile thermostability was occurred on the variant mut-W193L/C93I/I187L.
o
biocatalyst in the reduction of ketone substrates. Considering that The T
m
of mut-W193L/C93I/I187L was improved to 70.7 C from
o
mut-W193L/C93I/I187L exhibited high activity towards aromatic 62.1 C of the wild-type GoCR. Structural analysis of variant revealed
ketones, we also tested a panel of aromatic ketones with various that the obvious shortened distances of substrate with nicotinamide
molecular volumes as substrates to further examine the enzymatic cofactor NADH and with catalytic residues Ser142 and Tyr155 in
performance of mut-W193L/C93I/I187L. The results showed that these four variants accounted for the increased catalytic activity.
aromatic ketone and ketone ester substrates with volume ranging Additionally, three additional van der Waals interactions stabilize
o
from 127.18-210.04 Å could be efficiently and stereo-selectively flexible loop 8 and loop 12 around the active sites, leading to an 8.6 C
reduced by mut-W193L/C93I/I187L (Table 4), thereby indicating its improvement in thermostability when compared with wild-type
o
strong catalytic performance with bulky aromatic ketone and ketone GoCR. Most importantly, at an elevated temperature of 40 C, as high
ester substrates. In contrast, no detectable activity was observed as 1.8 M OPBE could be completely reduced with >99% ee and 540.4
towards aromatic keto acid (Table 4, entry 7) or the bulky diaryl g/(L·day) space-time yield, indicating the great potential of mut-
ketone (Table 4, entry 8). In conclusion, mut-W193L/C93I/I187L W193L/C93I/I187L for preparative synthesis of (R)-HPBE.
exhibited a broad substrate spectrum and high catalytic activity
towards bulky substrates, indicating its versatility and potential
applications as a catalyst in the preparation of chiral alcohols.
Conflicts of interest
There are no conflicts to declare.
Conclusions
Acknowledgements
This work was supported by Natural Science Foundation of Shanghai
In this study, the key residue Cys93 responsible for stereoselectivity
and enzyme activity, as well as Ile187 and Trp193 site that played the
role in determining the enzyme activity of GoCR were identified,
based on computational approach in combination with
semisaturation/saturation mutagenesis. Subsequently, four GoCR
variants that exhibited simultaneous enhancement in catalytic
(No. 19ZR1412700), the Fundamental Research Funds for the Central
Universities (No. 22221818014), Natural Science Foundation in
Jiangsu Province (BK20160373) and partially supported by Open
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

CP al et aa lsy es i ds oS c ni eo nt c ae d &j uTs et c mh na or gl oi ng sy
Page 10 of 11
ARTICLE
Journal Name
3
3
3
3
3
3
3
4
3 A. Weis, K. Katebzadeh, P. Soderhjelm, I. Nilsson aVniedw UAr.ticRleydOenl,inJe.
Funding Project of the State Key Laboratory of Bioreactor
Engineering.
Med. Chem. 2006, 49, 6596-6606.
DOI: 10.1039/D0CY01411G
4 K. Wichapong, A. Rohe, C. Platzer, I. Slynko, F. Erdmann, M.
Schmidt and W. Sippl, J. Chem. Inf. Model. 2014, 54, 881-893.
5 T. Li, M. Froeyen and P. Herdewijn, J. Mol. Graph. Model. 2008,
References
2
6, 813-823.
6 D. Piovesan, G. Minervini and S. C. E. Tosatto, Nucleic Acids Res.
016, 44, W367-W374.
7 B. Liang, Q. L. Lang, X. J. Tang and A. H. Liu, Bioresource Technol.
013, 147, 492-498.
1
B. S. Chen and F. Z. R. de Souza, RSC Adv. 2019, 9, 2102-2115.
Y. Nie, S. S. Wang, Y. Xu, S. G. Luo, Y. L. Zhao, R. Xiao, G. T.
Montelione, J. F. Hunt and T. Szyperski, ACS Catal. 2018, 8, 5145-
2
2
2
5
152.
8 E. Vazquez-Figueroa, J. Chaparro-Riggers and A. S. Bommarius,
ChemBioChem 2007, 8, 2295-2301.
3
Y. Zhou, Q. Y. Peng, L. Zhang, S. H. Cheng, L. T. Zeng, F. Dong and
Z. Y. Yang, Food Chem. 2019, 280, 27-33.
9 R. Chen, J. Deng, J. P. Lin, X. P. Yin, T. Xie, S. L. Yang and D. Z. Wei,
Biotechnol. Appl. Bioc. 2016, 63, 465-470.
4
5
G.-C. Xu and Y. Ni, Bioresour. Bioprocess. 2015, 2, 15.
P. Herold, A. F. Indolese, M. Studer, H. P. Jalett, U. Siegrist and H.
U. Blaser, Tetrahedron 2000, 56, 6497-6499.
0 G. De Simone, E. Langella, D. Esposito, C. T. Supuran, S. M. Monti,
J. Y. Winum and V. Alterio, J. Enzym. Inhib. Med. Ch. 2017, 32,
6
7
8
Z. G. Wang, S. Zhou, S. L. Zhang, S. Zhang, F. M. Zhu, X. L. Jin, Z.
M. Chen and X. L. Xu, Sci. Rep-UK 2017, 7, 4007-4017.
S. H. Huang and S. W. Tsai, J. Mol. Cata.l B-Enzym. 2004, 28, 65-
1
002-1011.
4
4
4
4
1 X. M. Gong, Z. Qin, F. L. Li, B. B. Zeng, G. W. Zheng and J. H. Xu,
ACS Catal. 2019, 9, 147-153.
6
9.
2 N. Tokuriki, F. Stricher, L. Serrano and D. S. Tawfik, PLoS Comput.
Biol. 2008, 4, e1000002-e1000008.
A. Liese, U. Kragl, H. Kierkels and B. Schulze, Enzyme Microb.
Tech. 2002, 30, 673-681.
3 L. J. Li, H. Liao, Y. Yang, J. Y. Gong, J. A. Liu, Z. D. Jiang, Y. B. Zhu,
A. F. Xiao and H. Ni, Int. J. Biol. Macromol. 2018, 112, 14-21.
4 W. Ren, L. Liu, L. Gu, W. L. Yan, Y. L. Feng, D. X. Dong, S. J. Wang,
M. S. Lyu and C. H. Wang, J. Agr. Food Chem. 2019, 67, 4355-
9
1
T. Sugai and H. Ohta, Agric. Biol. Chem. 1991, 55, 293-294.
0 W. Q. Lin, Z. He, Y. Jing, X. Cui, H. Liu and A. Q. Mi, Tetrahedron-
Asymmetr. 2001, 12, 1583-1587.
1
1
1
1
1
1
1
1
1
1 Y. Ni, Y. Su, H. Li, J. Zhou and Z. Sun, J. Biotechnol. 2013, 168, 493-
4
366.
4
98.
4
4
5 H. J. Wijma, R. J. Floor and D. B. Janssen, Curr. Opin. Struc. Biol.
2 Y. Chen, H. Lin, X. Xu, S. Xia and L. Wang, Adv. Synth. Catal. 2008,
50, 426-430.
2
013, 23, 588-594.
3
6 X. Y. Wang, R. Ma, X. M. Xie, W. N. Liu, T. Tu, F. Zheng, S. You, J.
3 I. Kaluzna, A. A. Andrew, M. Bonilla, M. R. Martzen and J. D.
Stewart, J. Mol. Catal. B-Enzym. 2002, 17, 101-105.
4 Y. G. Shi, Y. Fang, H. P. Wu, F. Li and X. Q. Zuo, Biocatal.
Biotransfor. 2009, 27, 211-218.
Z. Ge, H. F. Xie, B. Yao and H. Y. Luo, Sci. Rep-UK 2017, 7,15287-
1
5295.
5 I. Lavandera, A. Kern, M. Schaffenberger, J. Gross, A. Glieder, S.
de Wildeman and W. Kroutil, ChemSusChem 2008, 1, 431-436.
6 W. Zhang, Y. Ni, Z. H. Sun, P. Zheng, W. Q. Lin, P. Zhu and N. F. Ju,
Process Biochem. 2009, 44, 1270-1275.
7 N. D. Shen, Y. Ni, H. M. Ma, L. J. Wang, C. X. Li, G. W. Zheng, J.
Zhang and J. H. Xu, Org. Lett. 2012, 14, 1982-1985.
8 C. Li, R. Zhang, J. Wang, L. M. Wilson and Y. Yan, Trends
Biotechnol. 2020, 38, 729-744.
9 Z. Q. Liu, L. Wu, X. J. Zhang, Y. P. Xue and Y. G. Zheng, J. Agr. Food
Chem. 2017, 65, 3721-3729.
2
2
0 M. T. Reetz, J. Am. Chem. Soc. 2013, 135, 12480-12496.
1 I. Stanzione, C. Pezzella, P. Giardina, G. Sannia and A. Piscitelli,
Appl. Microbiol. Biot. 2020, 104, 915-924.
2
2
2
2
2
2
2 L. Ba, P. Li, H. Zhang, Y. Duan and Z. L. Lin, Biotechnol. Bioeng.
2
013, 110, 2815-2825.
3 C. Zhang, Y. Liu, C. Li, Y. Xu, Y. Su, J. Li, J. Zhao and M. Wu, Sci.
Rep-UK 2020, 10, 1680-1691.
4 O. Spadiut, C. Leitner, C. Salaheddin, B. Varga, B. G. Vertessy, T.
C. Tan, C. Divne and D. Haltrich, FEBS J. 2009, 276, 776-792.
5 Y. Y. Xia, W. J. Cui, Z. Y. Cheng, L. Peplowski, Z. M. Liu, M.
Kobayashi and Z. M. Zhou, ChemCatChem 2018, 10, 1370-1375.
6 J. Deng, Z. Yao, K. Chen, Y. A. Yuan, J. Lin and D. Wei, J. Biotechnol.
2
016, 217, 31-40.
7 J. Vina-Gonzalez, D. Jimenez-Lalana, F. Sancho, A. Serrano, A. T.
Martinez, V. Guallar and M. Alcalde, Adv. Synth. Catal. 2019, 361,
2
514-2525.
2
8 X. Liu, R. Chen, Z. W. Yang, J. L. Wang, J. P. Lin and D. Z. Wei, Mol.
Biotechnol. 2014, 56, 285-295.
2
3
9 M. M. Bradford, Anal. Biochem. 1976, 72, 248-254.
0 F.-F. Chen, G.-W. Zheng, L. Liu, H. Li, Q. Chen, F.-L. Li, C.-X. Li and
J.-H. Xu, ACS Catal. 2018, 8, 2622-2628.
3
3
1 T. J. Hou, J. M. Wang, Y. Y. Li and W. Wang, J. Comput. Chem.
2
011, 32, 866-877.
2 A. Lindstrom, L. Edvinsson, A. Johansson, C. D. Andersson, I. E.
Andersson, F. Raubacher and A. Linusson, J. Chem. Inf. Model.
2
011, 51, 267-282.
1
0 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 11
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY01411G
Graphical Abstract
This study reported a highly efficient engineered carbonyl reductase from Gluconobacter
oxydans (GoCR) through a structure-guided rational design to catalyze the synthesis of high
concentration (R)-HPBE, which provided an attractive biocatalyst for the efficient synthesis of
(R)-HPBE.