Semi-rational hinge engineering: modulating the conformational transformation of glutamate dehydrogenase for enhanced reductive amination activity towards non-natural substrates

Article abstract of DOI:10.1039/c9cy02576f

The active site is the common hotspot for rational and semi-rational enzyme activity engineering. However, the active site represents only a small portion of the whole enzyme. Identifying more hotspots other than the active site for enzyme activity engineering should aid in the development of biocatalysts with better catalytic performance. Glutamate dehydrogenases (GluDHs) are promising and environmentally benign biocatalysts for the synthesis of valuable chirall-amino acids by asymmetric reductive amination of α-keto acids. GluDHs contain an inter-domain hinge structure that facilitates dynamic reorientations of the domains relative to each other. Such hinge-bending conformational motions of GluDHs play an important role in regulating the catalytic activity. Thus, the hinge region represents a potential hotspot for catalytic activity engineering for GluDHs. Herein, we report semi-rational activity engineering of GluDHs with the hinge region as the hotspot. Mutants exhibiting significantly improved catalytic activity toward several non-natural substrates were identified and the highest activity increase reached 104-fold. Molecular dynamics simulations revealed that enhanced catalytic activity may arise from improving the open/closed conformational transformation efficiency of the protein with hinge engineering. In the batch production of three valuablel-amino acids, the mutants exhibited significantly improved catalytic efficiency, highlighting their industrial potential. Moreover, the catalytic activity of several active site tailored GluDHs was also increased by hinge engineering, indicating that hinge and active site engineering are compatible. The results show that the hinge region is a promising hotspot for activity engineering of GluDHs and provides a potent alternative for developing high-performance biocatalysts toward chirall-amino acid production.

Full text of DOI:10.1039/c9cy02576f

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: X. Yin, Y. Liu, L.
Meng, H. Zhou, J. Wu and L. Yang, Catal. Sci. Technol., 2020, DOI: 10.1039/C9CY02576F.
Volume
Number
21 January 2017
Pages 313-534
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.
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

Please do not adjust margins
Catalysis Science & Technology
Page 1 of 12
View Article Online
DOI: 10.1039/C9CY02576F
Catalysis Science & Technology
PAPER
Semi-Rational Hinge Engineering: Modulating the Conformational
Transformation of Glutamate Dehydrogenase for Enhanced
Reductive Amination Activity towards Non-Natural Substrates
Received 00th January 20xx,
Accepted 00th January 20xx
Xinjian Yin,† Yayun Liu, Lijun Meng, Haisheng Zhou, Jianping Wu,* and Lirong Yang,*
DOI: 10.1039/x0xx00000x
The active site is the common hotspot for rational and semi-rational enzyme activity engineering. However, the active site
represents only a small portion of the whole enzyme. Identifying more hotspots other than active site for enzyme activity
engineering should aid development of biocatalysts with better catalytic performance. Glutamate dehydrogenases (GluDHs)
are promising and environmentally benign biocatalysts for the synthesis of valuable chiral L-amino acids by asymmetric
reductive amination of α-keto acids. GluDHs contain an inter-domain hinge structure that facilitates dynamic reorientations
of the domains relative to each other. Such hinge-bending conformational motion of GluDHs play an important role in
regulating the catalytic activity. Thus, the hinge region represents a potential hotspot for catalytic activity engineering for
GluDHs. Herein, we report semi-rational activity engineering of GluDHs with the hinge region as the hotspot. Mutants
exhibiting significantly improved catalytic activity toward several non-natural substrates were identified and the highest
activity increase reached 104-fold. Molecular dynamics simulations revealed that enhanced catalytic activity may arise from
improving the open/closed conformational transformation efficiency of the protein with hinge engineering. In the batch
production of three valuable L-amino acids, the mutants exhibited significantly improved catalytic efficiency, highlighting
their industrial potential. Moreover, the catalytic activity of several active site tailored GluDHs was also increased by hinge
engineering, indicating that hinge and active site engineering are compatible. The results show that the hinge region is a
promising hotspot for activity engineering of GluDHs and provide a potent alternative for developing high-performance
biocatalysts toward chiral L-amino acid production.
studies have shown that mutation of residues distal from the
active site can affect the catalytic activity of enzymes
significantly.12-14 Nonetheless, identification of new engineering
hotspots that are not the active site of enzymes remains a long-
standing problem.
Introduction
Biocatalysis fulfils many key criteria of green chemistry, but
faces a major challenge of low catalytic performance of natural
enzymes in non-natural industrial environment.1-3 Rational
design and semi-rational design are two common molecular
engineering strategies used for enhancing the catalytic
performance of enzymes.4-6 These two strategies enable
efficient and precise engineering of enzymes by introducing
mutations into specific sites/regions of proteins. The selection
of the mutation sites is the core of such molecular engineering
strategies. The configuration and composition of the active site
are directly related to the catalytic function of enzymes. Thus,
rational and semi-rational activity engineering of enzymes have
primarily focused on residues located within or in close
proximity of the active site.7-11 However, the active site
represents only a small portion of the whole enzyme, and
Glutamate dehydrogenases (GluDHs, EC 1.4.1.2–1.4.1.4) are
ubiquitous enzymes that can catalyze the reversible oxidative
deamination of L-glutamate to a-ketoglutarate and
ammonia.15,16 Considering the richness of sources and diversity
of properties, GluDHs are particularly attractive and
environmentally friendly biocatalysts for the synthesis of
valuable optically pure L-amino acids by asymmetric reductive
amination of α-keto acids.17 However, their strict substrate
specificity toward native substrate α-ketoglutaric acid (α-KG)
limits their applications.18,19 Many rational and semi-rational
molecular engineering approaches have been performed to
enhance the catalytic activity of GluDHs toward non-natural
substrates.20-23 All of these engineering studies have focused on
the substrate-binding pocket as the mutation hotspot. Despite
some success, these active site tailored GluDHs show activity
enhancements for only several specific substrates. Moreover,
the catalytic activity of the acquired mutants toward extended
substrates is extremely low when compared with that of the
wild-type (WT) toward α-KG. Exploitation of new engineering
hotspots should facilitate the development of better
biocatalysts for the preparation of α-amino acids.
Institute of Bioengineering, College of Chemical and Biological Engineering,
Zhejiang University, Hangzhou, 310027, China. E-mail: lryang@zju.edu.cn,
wjp@zju.edu.cn
†
Current address: School of Marine Sciences, Sun Yat-sen University, Guangzhou,
510006, 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

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 12
Paper
Enzymes undergo multiple conformational changes in
Catalysis Science & Technology
A
pproximately 2000 colonies in the whole-gene random
View Article Online
DOI: 10.1039/C9CY02576F
solution, and these dynamics play a critical role in catalytic mutagenesis library of PpGluDH were screened using a high-
functions.24,25 Hinge-bending motions, resulting from reciprocal throughput screening method. After HPLC verification, 13
movements of dynamical quasi-rigid bodies, are always present mutants with significantly increased catalytic activity toward
in enzymes consisting of two domains.26-28 Such inter-domain PPO were obtained, including eight single-residue substituted
motions are important for substrate binding and product mutants, four two-residues substituted mutants, and one three-
release, while ensuring precise orientation of partners involved residues substituted mutant (Figure S1). The positive mutations
in the hydride transfer reaction. GluDHs are typical “hinge- that were primarily responsible for the improvement in activity
bending” enzymes.29-32 During catalysis, the cofactor-binding of the multi-residues substituted mutants were determined by
domain rotates with respect to the substrate-binding domain back-mutation analysis (Figure S2). Finally, we identified 11
around a α-helix structure, which is defined as the hinge, mutations that improved the PPO-oriented activity of PpGluDH
causing the subunits of the GluDHs transform between open (Table 1). The increases in specific activity of these mutations
and closed conformations.29,33 The hinge region is considered to varied from 1.52- to 8.15-fold, and the highest specific volume
play an important role in promoting or sustaining activity (per fermentation volume) reached 0.897 U/mL (I406F).
conformational changes and therefore in determining the These 11 mutations occurred at ten amino acid residues. The
catalytic activity of “hinge-bending” enzymes.34,35 In
a
distribution of these residues in the protein was analyzed using
thermostability engineering study of a GluDH, researchers the homology model of PpGluDH. As shown in Figure 1, five of
found that the introduction of a six-residue ion-pair network in these residues (Val29, Tyr81, Val194, Val380 and Met390) are
the hinge region could improve catalytic activity of the GluDH located in the substrate-binding domain of PpGluDH protein,
toward natural substrate-α-KG by about 64%.36 In view of this two residues (Thr352 and Leu354) are located in the cofactor-
dynamics-functions relationship, the hinge region can represent binding domain, and the other three residues (Asp400, Ile406
a potential hotspot for catalytic activity engineering of GluDHs. and His411) are located on the hinge structure that connects
However, to the best of our knowledge, no example of hinge the two domains. The most efficient mutation (i.e., I406F) is
engineering for improving catalytic activity of GluDHs toward located in the hinge region, indicating that a change in the hinge
non-natural substrates has been reported so far.
structure can significantly affect the catalytic activity of
In this study, we report semi-rational engineering of the
hinge region of GluDHs to increase their reductive amination
activity toward non-natural α-keto acids substrates. Molecular
dynamics (MD) simulations were conducted to rationalize the
improved activity of hinge-engineered mutants. In the batch
production of three valuable L-amino acids, these hinge-
engineered mutants exhibited marked improvement in catalytic
efficiency, highlighting the industrial potential of the developed
mutants. Importantly, catalytic activity of several active site
tailored GluDHs was also increased by hinge engineering. Our
results demonstrate that the hinge region of GluDHs is a
suitable engineering hotspot for activity enhancement,
providing a new perspective on the activity engineering of
“hinge-bending” enzymes.
Table 1. The positive mutations acquired by error-prone PCR
based directed evolution
Mutations
Activity (U/mL)
Relative activity (%)a)
WT
0.110±0.005
0.252±0.013
0.188±0.004
0.167±0.006
0.255±0.009
0.315±0.011
0.303±0.018
0.226±0.021
0.230±0.015
0.613±0.019
0.897±0.026
0.223±0.102
100
229
171
152
232
286
275
205
209
557
815
203
V29A
Y81C
V194A
T352A
L354P
V380G
M390L
D400G
I406T
I406F
H411Y
Results and discussion
Error-prone PCR Based Directed Evolution of PpGluDH.
In our previous work, we identified a NADP+-specific GluDH
(PpGluDH) from Pseudomonas putida that was able to catalyze
the
reductive
amination
of
acid
2-oxo-4-
(PPO) for
[(hydroxy)(methyl)phosphinyl]butyric
enantioselective synthesis of the competitive herbicide L-
phosphinothricin (L-PPT).37 Engineered by site-directed
mutagenesis based on multiple sequence alignment, the
specific activity of PpGluDH toward PPO increased 2.1-fold.
However, the activity of this mutant was still too low for
industrial application. Therefore, we used error-prone PCR-
based directed evolution to engineer PpGluDH for enhancing
the PPO-oriented activity.
a)Relative activity is expressed as a percentage of wild type (WT)
PpGluDH specific volume activity under the experimental
conditions.
2 |Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 3 of 12
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
Paper
View Article Online
DOI: 10.1039/C9CY02576F
Figure 1. Distribution of the mutation sites identified in error-prone PCR-based directed evolution. The substrate-binding domain is
shown in orange, the cofactor-binding domain is shown in blue and the hinge is shown in green. Red spheres represent the mutations
sites.
PpGluDH. Based on these observations, we carried out mutagenesis. Taking both screening effort and library coverage into
engineering of PpGluDH with the hinge region as the hotspot.
consideration, we applied NDT codon degeneracy (Phe, Leu, Ile, Val,
Tyr, His, Asn, Asp, Cys, Arg, Ser, Gly) for the construction of two-site
saturation mutagenesis and NNK codon degeneracy (all 20
proteinogenic amino acids) for the construction of single-site
saturation mutagenesis.38,39 Five focused libraries were constructed:
library A (I406X), library B (K402X), library C (A379X/L383X), library D
(T121X/L123X) and library E (I410X).
After high-throughput primary screening and HPLC verification, a
dozen mutants with improved PPO-oriented activity were identified
from four (not library E) of the focused libraries (Figure 3). Moreover,
all four libraries contain more than one positive mutant, indicating
that the catalytic activity of PpGluDH is vulnerable to the mutations
of hinge region. The specific volume activity and specific activity
(purified protein) of the best mutant in libraries A, B, C and D were
Semi-rational Engineering of the Hinge Region of PpGluDH.
Rational hinge engineering of PpGluDH is not possible without
structure-function information. Thus, a semi-rational approach was
used to engineer the hinge region of PpGluDH. As mentioned above,
mutation of Ile406, which is located in the hinge region, altered the
catalytic activity of PpGluDH significantly. Therefore, Ile406 and
spatially adjacent residues (< 4 Å, Thr121, Leu123, Ala379, Leu383,
Lys402 and Ile410) were selected as mutation targets (Figure 2). Two
pairs of adjacent residues, Thr121 and Leu123, and Ala379 and
Leu383, were mutated simultaneously to capture possible synergistic
conformational effects, while the other three residues (Lys402,
Ile406 and Ile410) were subjected to single-site saturation
Figure 2. Design of mutagenesis libraries for hinge region engineering. The target amino acid residues of libraries A to E are shown in
green, cyan, blue, pink and magenta, respectively.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 12
Paper
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C9CY02576F
Figure 3. The screening result of the hinge engineering libraries. (A) Library A, (B) library B, (C) library C and (D) library D. Relative activity is
expressed as the percentage of WT PpGluDH specific volume activity (0.11 U/mL) under the experimental conditions used.
determined (Table 2). Among these mutants, the T121N/L123Y mutant T121N/L123Y. This result shows that the activity
mutant screened from library D exhibited the highest activity increase of the double-mutant arises from the synergy of the
enhancement. The specific volume activity of T121N/L123Y two mutations made.
toward PPO was 11.21 U/mL and the specific activity of the
purified protein was 30.6 U/mg, which is 104-fold higher than Shedding Insights into the Origin of Enhanced Activity.
that of the WT enzyme (0.296 U/mg). The two-site combined
The subunits of GluDHs alternate between open and closed
mutants obtained from libraries C and D were analyzed by back
conformations in solution.29,30 In the absence of substrate and
mutation (Figure S3). Taking T121N/L123Y as an example, the
cofactor, the subunits of GluDH adopt the open conformation,
single-site mutations of T121N and L123Y increased the activity
and upon binding substrate and cofactor the subunits transform
of PpGluDH by 1.6- and 3.0-fold, respectively, which is much
to a closed conformation via hinge-bending motions, yielding
lower than the 104-fold increase observed for the combined
the active enzyme state. We hypothesize that mutations in the
hinge region affect the open/closed conformational transition
of PpGluDH and thereby affect its catalytic activity. To verify this
hypothesis, MD simulations were carried out to analyze
conformational differences between the hinge-engineered
mutants and WT PpGluDH in solution. Prior to carrying out the
MD simulations, we sought to define the open and closed
degree of PpGluDH. Thus, a homology three-dimensional model
of PpGluDH was generated using the solved structure of GluDH
Table 2. The most significant mutant in each library
Specific
Activity
(U/mL)
activity
(U/mg-
protein)
Fold
Mutants
Library
changeda)
WT
-
0.110±0.005 0.296±0.013
-
I406F
K402F
A
B
0.897±0.026
0.420±0.031
2.49±0.07
1.12±0.03
8.4
3.8
from Corynebacterium glutamicum (PDB ID: 5IJZ) as
a
template.30 The model contained subunits of "open" and
"closed" conformations. After systematic comparison, it was
found that the distance of the α carbons of Lys137 and Arg291
(Distance-K137/A291) is significantly different between the
“open” and “closed” conformations, i.e., 17.3 Å and 8.8 Å,
respectively (Figure S4). Therefore, the Distance-K137/A291
A379C/L
383C
C
2.25±0.10
5.74±0.21
19.4
104
T121N/L
123Y
D
11.21±0.57
30.67±0.72
a) Fold-change improvement in the specific activity of the purified
mutants over the WT.
4 |Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 5 of 12
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
Paper
View Article Online
DOI: 10.1039/C9CY02576F
Figure 4. Distance-K137/A291 distributions of WT and hinge-engineered mutants derived from 60 ns MD simulations. The Distance-
K137/A291 of 4000 conformers between 20–60 ns was calculated and counted.
was used as the indicator for judging the open/closed degree of catalytic activity by reducing the affinity of the enzyme protein
PpGluDH.
toward substrate (Km, Table 3). In summary, the enhanced catalytic
Sixty nanosecond MD simulations were performed for the WT activity of hinge-engineered mutants may be achieved because the
PpGluDH and hinge-engineered PpGluDH mutants (T121N/L123Y, open/closed conformational transformation efficiency is improved
A379C/L383C, I406F and K402F). The Distance-K137/A291 of 4000
conformers between 20–60 ns was then calculated and counted. As
shown in Figure 4, the average Distance-K137/A291 of the hinge-
engineered mutants is noticeably smaller than that of WT PpGluDH,
Table 3. Kinetic parameters of WT and hinge-engineered mutants
toward PPOa )
demonstrating that mutations in the hinge region decrease the open
amplitude of apo-PpGluDH in solution. A reduction in the open
amplitude enables a more efficient transition from the open state to
the closed conformation upon binding substrate and coenzyme,
thereby increasing the catalytic rate (kcat, Table 3). However, the
reduction of Distance-K137/A291 is not the greater the better.
Among the four hinge-engineered mutants, the open amplitude of
T121N/L123Y is not the smallest (Average Distance-K137/A291 =
22.25 Å) but the specific activity of this mutant was the highest. In
contrast, K402F has the smallest open amplitude (Average Distance-
K137/A291 = 14.26 Å) but showed the lowest specific activity among
the hinge-engineered mutants. Substrate and cofactor entry into the
active site and subsequent binding when the enzyme is in the open
conformation appears to be restricted when the Distance-
K137/A291 is too small for the K402F mutation, which hampers
kcat/ Km
(s-1M-1)
kcat (s-1)
Km (M)
Enzymes
WT37
6.33±0.12
598.56±20.31
388.02±13.12
246.28±7.79
-
0.30±0.02
0.29±0.03
1.52±0.17
2.47±0.21
-
21.06
T121N/L123Y
A379C/L383C
I406F
2062.07
255.27
99.71
-
K402Fb)
a)Kinetic data were obtained using purified proteins. All enzymatic
assays were performed at least triplicate and average values were
adopted.
b)Kinetic parameters of K402F can’t be accurately measured
because it’s Km is too high.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 12
Paper
Catalysis Science & Technology
by reducing the open amplitude of PpGluDH. In addition, to achieve for S1, only I406F and K402F showed an increase in catalytic activity,
View Article Online
the higher level of catalytic activity enhancement, reduction of the whereas for substrates S5, S6 and S7 all DfOouI:r10h.1i0n3g9e/-Ce9nCgYin0e2e57re6dF
open amplitude should not be too large such that it may hinders mutants showed an increase in catalytic activity. It should be noted
substrate binding.
that these hinge-engineered mutants were screened with PPO as the
target substrate. Thus, if we adopt specific screening processes for
other target substrates, mutants with higher catalytic activity toward
other targets should be obtained. Therefore, engineering of the
hinge region may serve as an efficient strategy to improve the
reductive amination activity of PpGluDH toward α-keto acids.
Substrate Specificity Analysis of Hinge-engineered Mutants.
The substrate specificities of WT PpGluDH and the hinge-engineered
mutants were characterized using a panel of α-keto acids with a
broad range of structural features. As shown in Table 4, the hinge-
engineered PpGluDHs exhibited improved catalytic activity toward
all the active α-keto acids. This is mainly because hinge engineering
altered the conformational transformation efficiency of PpGluDH,
which is essentially independent of the chemical structure of the
substrates. However, for different substrates, the increase of specific
activity varied between the hinge-engineered mutants. For example,
Preparative Application of Hinge-Engineered PpGluDH for the
Synthesis of L-Amino Acids.
The effectiveness of the hinge-engineered mutants in the batch
production of three valuable L-amino acids was studied by
coupling the mutants with glucose dehydrogenase/alcohol
dehydrogenase for NADPH regeneration.
Table 4. Specific activity of WT and hinge-engineered PpGluDH toward various α-keto acids.
O
O
O
O
O
O
O
O
O
P
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
HO
S7
PPO
S1
S2
S3
S4
S5
S6
O
O
O
O
O
O
COOH
COOH
COOH
COOH
COOH
S13
S8
S9
S10
S11
S12
Specific activity(U/g) a)
A379C/L383C
120.5±3.2
Substrate
WT
T121N/L123Y
496.1±27.2
138.6±6.97
4809±281
608.0±57.9
20161±1332
703.5±16.4
591.4±35.1
6369±140
446.7±23.7
108.8±9.25
264.8±6.9
282.2±19.8
N.A. c)
I406F
K402F
PPOb)
S1
4.52±0.81
160.2±14.4
5357±304
308.7±15.9
6618±323
67.10±2.83
51.95±2.76
568.3±28.9
53.78±5.96
49.39±3.69
116.3±6.7
26.18±2.80
N.A.c)
65.27±4.83
475.5±29.7
24642±1501
1972±125
29743±1256
494.7±23.9
471.9±23.7
3578±104
234.9±15.6
336.5±31.8
629.3±25.3
128.6±8.6
N.A. c)
13.27±1.21
409.9±20.5
8029±584
356.1±17.5
10124±997
185.9±9.7
148.6±11.6
1873±135
532.0±31.3
1052±84.6
2807±165
555.8±37.1
N.A. c)
136.5±14.4
4767±187
S2
S3
290.2±29.3
8515±384
S4
S5
129.2±5.6
S6
125.9±15.3
1025±61.4
65.91±4.6
S7
S8
S9
177.9±9.3
S10
S11
S12
S13
292.1±12.2
23.32±2.16
N.A. c)
75.31±6.61
286.6±10.7
632.5±34.5
2101±154
2264±136
a)
Specific activity was determined using purified proteins by monitoring the change in NADPH absorbance at 340 nm (ε = 6220 M−1
cm−1) during reductive amination. All enzymatic assays were performed in triplicate, at least, and the averaged values are reported.
b) Specific activities of PPO obtained here are lower than those determined using HPLC, owing to the different reaction systems.
c) N.A.= no measurable activity.
6 |Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 7 of 12
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
Paper
L-PPT
(L-2-amino-4-[hydroxy(methyl)phosphinoyl]butyric converted over 99% of 0.5 M PPO within 140 min, affording final
View Article Online
DOI: 10.1039/C9CY02576F
acid) is a green, broad-spectrum, non-selective herbicide. The L-PPT concentration of 80.58 g/L and space-time yield (STY) of
reductive amination of PPO using GluDHs is an ideal route for 828.82 g·L-1·d-1, which is marked higher than that of WT (40.96
the asymmetric synthesis of L-PPT.37 Figure 5A shows the time g·L-1·d-1) .
course of the conversion of 0.5 M PPO with WT PpGluDH and
L-Homophenylalanine is an important non-canonical amino
hinge-engineered mutants (T121N/L123Y, A379C/L383C, I406F acid used as the building block of angiotensin-converting
and K402F). Compared with the WT, all hinge-engineered enzyme inhibitors, such as lisinopril, enalapril and quinapril.40
mutants exhibited significantly higher catalytic efficiencies. The There are only a few reports describing the synthesis of L-
specific activity of T121N/L123Y is the highest of the mutants. It homophenylalanine using GluDHs, mainly because of the low
catalytic activity of WT GluDHs toward 2-oxo-phenylbutyric acid
(S13).20 Through hinge engineering the catalytic activity of
PpGluDH mutants toward S13 was improved significantly.
Among the mutants, the specific activity of I406F and K402F
toward S13 improved by 27.9-fold and 30.1-fold, respectively,
when compared with the WT. Figure 5b shows the time course
of the conversion of 0.1 M S13 with these two mutants. Both
mutants converted > 99% of 0.1 M S13 within 100 min, whereas
the WT only achieved 73.8% conversion after 300 min. The L-
homophenylalanine concentration in the final reaction mixture
was 17.13 g/L and the corresponding STY of the process was
246.67 g ·L–1·d–1.
L-2-Aminobutyric acid is an important precursor for the
synthesis of many chiral drugs, including the anti-tuberculotic
ethambutol,
brivaracetam
and
the
anti-epileptic
levetiracetam.22 Among the hinge-engineered mutants, I406F
showed a 4.6-fold increase in specific activity towards 2-
ketobutyric acid (S2). The reaction process of conversion of 0.15
M S2 with I406F was determined. As shown in Figure 5C, I406F
converted > 99% of 0.15 M S2 within 80 min, yielding 14.39 g/L
of L-2-aminobutyric acid and an STY of 259.02 g·L–1·d–1.
In the three groups of batch reactions, all hinge-engineered
mutants showed significant improvements in catalytic
efficiency when compared with WT. The ee values of the
resulting three L-amino acids were over 99% (Figure S5–S7),
showing strict stereoselectivity. These results indicate the
promising practical applications of hinge-engineered mutants in
the preparation of chiral L-amino acids.
Hinge Engineering of Active Site Tailored GluDHs.
In a previous study, the PPO catalytic activity of several GluDHs
including PpGluDH was markedly improved by tailoring the
“cave” structure located in the active site.41 We performed
hinge engineering to the active site tailored GluDHs to analyze
the compatibility of hinge and active site engineering. Taking
the active site tailored PpGluDH-A167G as the parent; residues
Thr121, Leu123, Ala379, Leu383 and Ile406 were chosen for
saturation mutagenesis. Three saturation mutagenesis libraries
were constructed: A167G/T121X/L123X (two-site combination
mutagenesis of Thr121 and Leu123), A167G/A379X/L383X
(two-site combination mutagenesis of Ala379 and Leu383) and
A167G/I406X (single-site mutagenesis of Ile406). High-
throughput primary screening and HPLC rescreening were
carried out with PPO as the substrate. As shown in Figure 6,
positive mutants were identified from all three libraries and
each of these libraries contained more than one positive
mutant, demonstrating that hinge engineering can further
Figure 5. Batch production of L-amino acids using WT and
hinge-engineered
GluDHs.
(A)
L-phosphinothricin
preparation, (B) L-homophenylalanine preparation and (C) L-
2-aminobutyric acid preparation.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 12
Paper
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C9CY02576F
Figure 6. The screening results of hinge engineering libraries constructed on the basis of the “cave-tailored” mutant PpGluDH-A167G.
Relative activity is expressed as the percentage of A167G specific volume activity (14.9 U/mL) under the experimental conditions used.
improve the catalytic activity of PpGluDH on the basis of active from Aladdin Reagents Co. Ltd (Shanghai, China). 2-Oxovaleric
site engineering. This result revealed that hinge engineering is acid (S4), 3-methyl-2-oxovaleric acid (S5), phenylpyruvic acid
compatible with active site engineering. Among the obtained (S12) and 2-oxo-4-phenylbutyric acid (S13) were purchased
mutants, the A167G/A379S/L383Y mutant screened from the from TCI Development Co., Ltd. (Shanghai, China). Nicotinamide
A167G/A379X/L383X library exhibited the highest catalytic adenine dinucleotide derivatives (NADH, NADPH, NAD+ and
activity toward PPO, reaching 73.37 U/mL, which is 4.9 times NADP+) were purchased from Bontac Bio-engineering Co., Ltd.
that of PpGluDH-A167G (14.9 U/mL). The enzyme was purified (Shenzhen, China). All other chemicals and reagents were
to homogeneity and its specific activity was determined to be chemically pure and were obtained commercially.
206.09 U/mg-protein, which is 5.4 times that of PpGluDH-
A167G (38.13 U/mg-protein),41 6.7 times that of T121N/L123Y Protein Expression and Purification.
and 696 times that of the WT enzyme. In addition to PpGluDH-
Protein expression and purification of the WT and mutants were
A167G, we also performed hinge engineering to two other
carried out using a protocol described previously.37 The purified
active site tailored GluDHs from different sources and with
proteins were stored in 20% (v/v) glycerol at -80 oC until further
different coenzyme specificity (Table S1). Through hinge
use. The purity of the purified GluDHs were analyzed by sodium
engineering, their catalytic activity toward PPO was also
dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE,
improved significantly (Figure S10), indicating the general
12%) with a 5% stacking gel. Protein concentrations were
applicability of this enzyme engineering method to enhance
determined using the Bradford protein assay kit (Quick StartTM
Bio-Rad, USA).
,
reductive amination activity of GluDHs.
Whole-gene Random Mutagenesis Library Construction.
Experimental Section
The whole-gene random mutagenesis library was generated by
the error-prone PCR. Plasmid pET-28a(+)-PpGluDH harboring
the PpGluDH gene was used as the template. The PpGluDH gene
library containing random mutations was amplified using
Microorganisms and Materials.
The recombinant GluDHs used in this work were constructed in
our previous work and were maintained in our laboratory.37,41
PPO,42 2-oxoheptanoic acid (S8), 2-oxooctanoic acid (S9) and 2-
oxononanoic acid (S10) 43,44 were synthesized in our laboratory
and were characterized by MS and NMR (Figure S15-S22). α-
Ketoglutaric acid, pyruvic acid (S1), 2-ketobutyric acid (S2), 3-
methyl-2-oxobutanoic acid sodium salt (S3), 4-methyl-2-
oxovaleric acid (S6), 2-ketohexanoic acid sodium salt (S7),
primers PpGluDH-F-BamHⅠ and PpGluDH-R-HindⅢ (sequences
shown in Supporting Information) with EasyTaq DNA
polymerase (TransGen Biotech, Beijing, China). For the 1.4-kb
PpGluDH gene, 0.8 mM MnCl2 was used to obtain the desired
level of mutagenesis rate (
∼1 to 3 amino acid substitutions).
The PCR product was digested with BamH
Ⅰ
and Hind and
Ⅲ
ligated into plasmid pET-28a(+) digested with the same two
restriction enzymes. The ligation products were transformed
into E. coli BL21(DE3), which were then plated on Luria-Bertani
phenylglyoxylic acid (S11)
, DL-homophenylalanine, DL-
phosphinothricin and DL-2-aminobutyric acid were purchased
8 |Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 9 of 12
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
Paper
(LB) agar plates containing 50 μg/ml kanamycin. Ten randomly Site-directed Mutagenesis.
View Article Online
The mutants of back-mutation analysis DwOeI:r1e0.c10o3n9s/tCr9uCcYt0e2d57b6yF
site-directed mutagenesis according to the standard
QuikChange Site-Directed Mutagenesis protocol.45 The required
mutations were confirmed by DNA sequencing. All the site-
picked clones from this library were sequenced to verify the
mutagenesis rate.
Saturation Mutagenesis Libraries Construction.
Saturation mutagenesis libraries were constructed using the directed mutagenesis primers used in this work are listed in
One-step Cloning Kit (Vazyme Biotech, Nanjing, China). Based Table S4.
on the sequences of template plasmids that harboring the
parent gene, mutation primers (Mutation-R and Mutation-F) Homology Modeling and MD Simulation.
and aid primers (Aid-F and Aid-R) were designed and
The three-dimensional homology model of WT PpGluDH was
synthesized (sequences shown in Supporting Information). The
generated
with
the
SWISS-MODEL
server
short gene fragment was amplified with Mutation-F/Aid-R as
primer pairs, and the long gene fragment was amplified with
Mutation-R/Aid-F as primer pairs. After gel-purified, the short
gene fragment was ligated to the long gene fragment by
recombination reaction. The system and conditions of the
recombination reaction was selected according the
manufacturer’s manual. The resulting products were
transformed into E. coli BL21(DE3), which were then plated on
LB agar plates containing 50 μg/ml kanamycin. Ten randomly
picked clones were sequenced to verify the quality of the
libraries.
(https://swissmodel.expasy.org/) using the solved structure of
GluDH from Corynebacterium glutamicum (PDB ID: 5IJZ,71.4%
sequence identity with PpGluDH) as the template.30 The
generated models were evaluated using PROCHECK, Verify-3D
and ERRAT (http://services.mbi.ucla.edu/SAVES/). Homology
modeling of mutants was performed using the Build Mutants
package in Accelrys Discovery Studio 3.0 (Accelrys Inc.) and was
based on the homology model of WT PpGluDH. The subunit at
open conformation (without binding of ligands) in each models
was chose to run MD simulations. All molecular structures were
visualized by PyMOL software (Version 1.5, Schrꢀdinger, LLC).
MD simulations were performed using the Accelrys Discovery
Studio 3.0 (Accelrys Inc.). The CHARMm force field and the
Screening Procedures.
Individual colonies were picked and inoculated into 200 μL of LB Momany-Rone partial charge were used to specify the atom
medium containing kanamycin (50 μg/mL) in separate wells of type. A solvent environment was added to make the simulation
96-well plates, with six parent clones included as controls in more realistic. MD simulations were performed according to
each plate. The plates were incubated at 37°C with shaking at standard dynamic cascade protocols. In the first and second
200 rpm for 8 h. Then, 50 μL of the precultures were inoculated steps, the maximum steps of the steepest descent minimization
into expression cultures in 96-well, 2-mL-deep well plates method and the conjugate gradient minimization method were
containing 450 μL of LB medium and 50 μg/mL kanamycin. The set to 10ꢁ000. The RMS gradients were set to 0.2 and 0.0001,
preculture plates were stored at -80°C with 20% (v/v) glycerol respectively. The temperature of the heating step was set to
for rescreening of positive clones. The expression cultures were 300 K for 100 ps. The equilibration and production steps were
incubated at 37°C with 200 rpm shaking for 3 h. Then, 50 μL of set to 1 ns and 60 ns, respectively. The tempering temperature
LB medium with 50 μg/mL kanamycin and 5 mM IPTG was for these two steps was 330 K. The type of production was NPT.
added into each well. After 12 h of growth at 37°C, 200 rpm, the The trajectory of the production step was saved every 10 ps,
o
plates were centrifuged at 10,000 × g, 4 C for 20 min, and the and 4000 frames between 20-60 ns were selected for
supernatant was carefully decanted. The collected cells were comparison analysis. Distance-K137/A291 (Distance between
frozen at -80°C until activity screening.
the α carbons of Lys137 and Arg291) of the frames was
After thawed at room temperature for 30 min
，
each cell calculated and analyzed using the Analyze Trajectory package in
pellet was resuspended in 300 mL lysis buffer (pH 7.5 10 mM Accelrys Discovery Studio 3.0.
phosphate buffer, 0.75 mg/mL lysozyme, 0.01 mg/mL DNase
Ⅰ
), and incubated at 37 °C with 200 rpm shaking for 60 min. Cell Determination of Kinetic Parameters.
debris was removed by centrifugation at 10,000 × g, 4°C for 20
The kinetic analysis of WT and mutants toward PPO was carried
min, and the crude extract was diluted to appropriate multiple
with pH 7.5 phosphate buffer (10 mM). A dose of 200 μL of the
diluted crude extract was added into 200 μL of the substrate
solution (pH 7.5 10 mM phosphate buffer, 2 mM NADPH, 20
mM PPO and 50 mM (NH4)2SO4). After incubation at 37 °C with
200 rpm shaking for 20 min, 100 μL of the reaction mixture was
sampled and diluted with 100 μL pH 7.5 10 mM phosphate
buffer. The absorbance at 340 nm was measured, and the
mutants with lower absorbance than WT enzyme were
selected. The activity of candidate mutants was verified by HPLC
method using the standard activity assay system described
previously.37
out using a protocol described previously.37 The kinetic
parameters were determined from three independent initial
rate measurements performed with the same batch of purified
+
enzyme over 0.04-0.4 M PPO at fixed NADPH (10 mM) and NH4
(0.5 M) concentrations. The concentration of NADPH and NH4
+
was chosen based on the results of saturation values
determination (Figure S13). Initial rate data were fitted to the
Michaelis-Menten equation, and the Km and kcat values were
calculated from the slope and y-intercept of the double
reciprocal plot (Figure S14).
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol

Please do not adjust margins
Catalysis Science & Technology
Page 10 of 12
Paper
Catalysis Science & Technology
Substrate Specificity Analysis.
pH 7.5 phosphate buffer/acetonitrile (0.8:0.05 v/v) as mobile
phase at a flow rate of 0.85 mL/min as mobile phase and
View Article Online
DOI: 10.1039/C9CY02576F
The specific activity of purified PpGluDH WT and mutants
detected at 210 nm. For S13, the column was eluted with 50
mM pH 7.5 phosphate buffer/acetonitrile (0.8:0.2 v/v) as
mobile phase at a flow rate of 1 mL/min as mobile phase and
detected at 220 nm. The column temperature was constant at
40°C.
toward
various
α-keto
acids
was
determined
spectrophotometrically by monitoring the change in
absorbance of NADPH (e= 6220 M-1 cm-1) at 340 nm during
reductive amination in phosphate buffer (pH 7.5) at 35 oC.41 The
assay mixture (1 mL) contained 10 mM (for PPO, S1–S11 and
+
S13) or 5 mM (for S12) α-keto acid, 0.12 mM NADPH, 0.5 M NH4
((NH4)2SO4) and a certain amount of enzyme. One unit of
enzyme activity was defined as the amount of enzyme required
for the oxidation of 1 mmol NADPH per minute.
Conclusions
In summary, we identified a new hotspot, the hinge region, for
activity engineering of GluDHs to enhance their reductive
amination activity toward non-natural substrates. Inspired by
the screening results of whole-gene random mutagenesis, we
selected specific residues located within or spatially near the
hinge region of PpGluDH for saturation mutagenesis. A dozen
positive mutants with significantly improved catalytic activity
toward the non-natural substrate PPO were obtained. MD
simulations were conducted to shed insights into the origin of
enhanced activity, revealing that the catalytic activity
enhancement may arises from the improvement of open/closed
conformational transformation efficiency of the protein.
Substrate specificity analysis showed that these hinge-
engineered mutants exhibited improved catalytic activity
toward all active α-keto acids. In the batch production of three
valuable chiral L-amino acids, L-PPT, L-homophenylalanine and
L-2-aminobutyric acid, these hinge-engineered mutants
exhibited markedly improved catalytic efficiency. The STY
varied from 246.67 to 828.82 g·L–1·d–1, highlighting the
industrial potential of the developed mutants. Finally, several
active-site tailored GluDHs underwent hinge engineering, which
resulted in a further improvement in their catalytic activity,
suggesting that hinge region engineering is compatible with
active-site engineering. Our results provide a new perspective
on the activity engineering of GluDHs and facilitate the
development of high catalytic performance biocatalysts in the
green production of chiral L-amino acid.
Preparative Application of Mutant GluDHs for the Synthesis of
L-amino acids.
Batch reactions were carried out at 35 °C by magnetic agitation
with water as the medium (30 mL). For L-PPT preparation, each
sample contained 0.5 M PPO, 0.25 M ammonium sulfate, 0.6 M
glucose, 0.5 mM NADP+, 1.25 g-DCW/L GluDH and 25 U/mL
glucose dehydrogenase; For L-homophenylalanine preparation
，
each sample contained 0.1 M S13, 0.25 M ammonium
sulfate, 0.12 M glucose, 0.2 mM NADP+, 1.25 g-DCW/L GluDH and
25 U/mL glucose dehydrogenase; For L-2-aminobutyric acid
preparation, each sample contained 0.15 M S2, 0.25 M
ammonium sulfate, 0.2 M isopropanol, 0.2 mM NADP+, 0.02 g-
DCW/L GluDH and 10 U/mL alcohol dehydrogenase. The glucose
dehydrogenase and alcohol dehydrogenase play a role in
coenzyme (NADPH) regeneration (amino acid sequences shown
in Supporting Information). The recombinant GluDH and
glucose/alcohol dehydrogenase cells were resuspended in
ddH2O and disrupted ultrasonically. The resultant slurry was
centrifuged at 12000 × g and the supernatant was used for the
batch reaction. The pH was automatically adjusted to 7.5 by
titrating with NH3·H2O. The reaction process was monitored by
the determination of residual substrate concentration using
HPLC, as described in the “Analysis” section. The formed
products were purified (purification process shown in
Supporting Information) and characterized using MS and NMR
analysis (Figure S23-S28).
Analysis.
Conflicts of interest
There are no conflicts to declare
The L-PPT and PPO were analyzed with the analytical method
described previously.37
The concentration and enantiomeric excess (ee) of L-
homophenylalanine and L-2-aminobutyric acid were
determined by analytical HPLC on a C18 column (Pntulips® QS-
C18, 5 μm, 4.6 × 250 mm) after derivatization with o-
phthalaldehyde and N-acetyl-L-cysteine and detected at 338
nm. For L-homophenylalanine, the column was eluted with 50
mM sodium acetate buffer/acetonitrile (0.8:0.3 v/v) as mobile
phase at a flow rate of 1.1 mL/min. For L-2-aminobutyric acid,
the column was eluted with 50 mM sodium acetate
buffer/acetonitrile (0.8:0.15 v/v) as mobile phase at a flow rate
of 0.95 mL/min. The column temperature was constant at 30°C;
The keto acids (S2 and S13) were analyzed by analytical HPLC on
the same column. For S2, the column was eluted with 50 mM
Acknowledgements
This work was financially supported by the National Natural
Science Foundation of China (No. 21476199, 21676240),
National Basic Research Program of China (973 Program, No.
2011CB710800).
References
10 |Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 11 of 12
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
Paper
1
2
3
4
5
S. Wenda, S. Illner, A. Mell and U. Kragl, Green Chemistry, 25 G. Alex and T. Janet, Febs Letters, 2004, 567, 67-73.
View Article Online
2011, 13, 3007-3047.
26 T. Knöfel and N. Sträter, Journal of Molecular Biology, 2001,
DOI: 10.1039/C9CY02576F
N. Ran, L. Zhao, Z. Chen and J. Tao, Green Chemistry, 2008, 10,
309, 255-266.
361-372.
27 C. M. Dobson, Nature, 1990, 348, 198-199.
S. Jemli, D. Ayadi-Zouari, H. B. Hlima and S. Bejar, Critical 28 R. D. Banks, C. C. F. Blake, P. R. Evans, R. Haser, D. W. Rice, G.
Reviews in Biotechnology, 2016, 36, 246-258.
L. Stefan, Current Opinion in Biotechnology, 2010, 21, 734-
743.
W. Hardy, M. Merrett and A. W. Phillips, Nature, 1979, 279,
773-777.
29 P. Prakash, N. S. Punekar and P. Bhaumik, Journal of Biological
Chemistry, 2018, 293, 6241-6258.
30 H. F. Son, I.-K. Kim and K.-J. Kim, Biochemical and biophysical
research communications, 2015, 459, 387-392.
R. A. Chica, N. Doucet and J. N. Pelletier, Current Opinion in
Biotechnology, 2005, 16, 378-384.
6
7
R. Chen, Trends in Biotechnology, 2001, 19, 13-14.
J. W. He, X. Q. Huang, J. Xue and Y. S. Zhu, Green Chemistry, 31 M. Nakasako, T. Fujisawa, S.-i. Adachi, T. Kudo and S. Higuchi,
2018, 20, 5484-5490.
Biochemistry, 2001, 40, 3069-3079.
8
9
F. F. Chen, G. W. Zheng, L. Liu, H. Li, Q. Chen, F. Li, C. X. Li and 32 M. W. Bhuiya, H. Sakuraba, T. Ohshima, T. Imagawa, N.
J. H. Xu, ACS Catalysis, 2018, 8, 2622-2628.
B. Xia, J. Xu, Z. Xiang, Y. Cen, Y. Hu, X. Lin and Q. Wu, ACS
Catalysis, 2017, 7, 4542-4549.
Katunuma and H. Tsuge, Journal of Molecular Biology, 2005,
345, 325-337.
33 T. J. Stillman, A. M. B. Migueis, X.-G. Wang, P. J. Baker, K. L.
Britton, P. C. Engel and D. W. Rice, Journal of molecular
biology, 1999, 285, 875-885.
10 M. T. Reetz, M. Bocola, J. D. Carballeira, D. Zha and A. Vogel,
Angewandte Chemie, 2005, 44, 4192-4196.
11 S. W. Han, J. Kim, H. S. Cho and J. S. Shin, ACS Catalysis, 2017, 34 F. Kazuyo, S. Y. Hong, Y. J. Yeon, J. C. Joo and Y. J. Yoo, Journal
7, 3752-3762.
of Industrial Microbiology & Biotechnology, 2014, 41, 1181-
1190.
12 K. L. Morley and R. J. Kazlauskas, Trends in Biotechnology,
2005, 23, 231-237.
13 M. Wilding, N. Hong, M. Spence, A. M. Buckle and C. J.
35 A. M. Schramm, R. Mehra-Chaudhary, C. M. Furdui and L. J.
Beamer, Biochemistry, 2008, 47, 9154-9162.
Jackson, Biochemical Society Transactions, 2019, 47, 701-711. 36 J. H. G. Lebbink, S. Knapp, J. van der Oost, D. Rice, R.
14 J. Lee and N. M. Goodey, Chemical reviews, 2011, 111, 7595-
Ladenstein and W. M. de Vos, Journal of Molecular Biology,
1998, 280, 287-296.
7624.
15 E. Santero, A. B. Hervás, I. Canosa and F. Govantes, 37 X. J. Yin, J. P. Wu and L. R. Yang, Applied Microbiology &
Dehydrogenases, 2012, 289-309.
16 N. M. Brunhuber and J. S. Blanchard, Critical reviews in 38 M. T. Reetz, D. Kahakeaw and R. Lohmer, ChemBioChem,
biochemistry and molecular biology, 1994, 29, 415-467. 2008, 9, 1797-1804.
Biotechnology, 2018, 102, 1-9.
17 Y. P. Xue, C. H. Cao and Y. G. Zheng, Chemical Society Reviews, 39 S. Kille, C. G. Acevedo-Rocha, L. P. Parra, Z. G. Zhang, D. J.
2018, 47, 1516-1561.
Opperman, M. T. Reetz and J. P. Acevedo, ACS Synthetic
Biology, 2013, 2, 83-92.
18 P. J. Baker, M. L. Waugh, X. G. Wang, T. J. Stillman, A. P.
Turnbull, P. C. Engel and D. W. Rice, Biochemistry, 1997, 36, 40 H. Joon-Young, P. Jihyang, S. Joo-Hyun, C. Minho, C. Byung-
16109-16115.
Kwan, K. Juhan and K. Byung-Gee, Biotechnology
Bioengineering, 2010, 102, 1323-1329.
&
19 K. L. Britton, P. J. Baker, P. C. Engel, D. W. Rice and T. J.
Stillman, Journal of Molecular Biology, 1993, 234, 938-945.
20 H. Li and J. C. Liao, ACS Synthetic Biology, 2014, 3, 13-20.
41 X. J. Yin, Y. Y. Liu, L. J. Meng, H. S. Zhou, J. P. Wu and L. R. Yang,
Advanced Synthesis & Catalysis, 2019, 361, 803-812.
21 L. L. Chen, J. L. Wang, Y. Hu, B. J. Qian, X. M. Yao, J. F. Wang 42 N. Minowa, N. Nakanishi and M. Mitomi, Patent US8017797
and J. H. Zhang, Journal of Molecular Modeling, 2013, 19,
1919-1927.
B2, 2011.
43 M. K. Wong, N. W. Chung, L. He, X. C. Wang, Z. Yan, Y. C. Tang
and D. Yang, Journal of Organic Chemistry, 2003, 68, 6321-
6328.
22 K. Zhang, H. Li, K. M. Cho and J. C. Liao, Proceedings of the
National Academy of Sciences of the United States of America,
2010, 107, 6234-6239.
44 R. J. Rapf, R. J. Perkins, H. Yang, G. M. Miyake, B. K. Carpenter
and V. Vaida, Journal of the American Chemical Society, 2017,
139, 6946-6956.
45 L. Zheng, U. Baumann and J. L. Reymond, Nucleic Acids
Research, 2004, 32, e115.
23 M. I. H. Khan, H. Kim, H. Ashida, T. Ishikawa, H. Shibata and Y.
Sawa, Bioscience Biotechnology and Biochemistry, 2005, 69,
1802-1805.
24 G. G. Hammes, Biochemistry, 2002, 41, 8221-8228.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol

Please do not adjust margins
Catalysis Science & Technology
Page 12 of 12
View Article Online
DOI: 10.1039/C9CY02576F
Catalysis Science & Technology
PAPER
Table of Contents Entry
Hinge region was identified to be a promising hotspot for activity engineering of GluDHs, providing a potent
alternative for developing high-performance biocatalysts toward valuable optically pure L-amino acids
production.
Please do not adjust margins