Multiparameter Optimization in Directed Evolution: Engineering Thermostability, Enantioselectivity, and Activity of an Epoxide Hydrolase

Article abstract of DOI:10.1021/acscatal.6b01113

The challenge of optimizing several parameters in the directed evolution of enzymes remains a central issue. In this study we address the thermostability, enantioselectivity, and activity of limonene epoxide hydrolase (LEH) as the catalyst in the hydrolytic desymmetrization of cyclohexene oxide with formation of (R,R)- and (S,S)-cyclohexane-1,2-diol. Wild type LEH shows a thermostability of T₅₀³⁰ = 41 °C and an enanioselectivity of 2% ee (S,S). Two approaches are described herein. In one strategy, the mutations generated previously by Janssen, Baker, and co-workers for notably increased thermostability are combined with mutations evolved earlier for enhanced enantioselectivity. Although highly enantioselective R,R and S,S variants (92-93% ee) with increases in T₅₀³⁰ by 10-11 °C were obtained, relative to wild type LEH the tradeoff in activity was significant. The second strategy based on the simultaneous optimization of both parameters using iterative saturation mutagenesis (ISM) with minimized tradeoff in activity proved to be superior. Several notably improved variants were observed, a reasonable "compromise" being R,R- and S,S-selective LEH variants (80-94% ee) showing enhanced thermostability by 5-10 °C and still reasonable levels of activity. Analysis of the X-ray structure of the S,S variant (94% ee) with and without diol product sheds light on the origin of altered stereoselectivity.

Full text of DOI:10.1021/acscatal.6b01113

Subscriber access provided by UOW Library
Article
Multi-Parameter Optimization in Directed Evolution: Engineering
Thermostability, Enantioselectivity and Activity of an Epoxide Hydrolase
Guangyue Li, Hui Zhang, Zhoutong Sun, Xinqi Liu, and Manfred T. Reetz
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b01113 • Publication Date (Web): 02 May 2016
Downloaded from http://pubs.acs.org on May 7, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
Multi-Parameter Optimization in Directed Evolution: Engineering
Thermostability, Enantioselectivity and Activity of an Epoxide
Hydrolase
1,2,§
3,§
1,2
3,
1,2,
Guangyue Li , Hui Zhang , Zhoutong Sun , Xinqi Liu * and Manfred T. Reetz *
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr,
Germany
2
Fachbereich Chemie der Philipps-Universität, Hans-Meerwein-Strasse, 35032 Marburg, Germany
3
State Key Laboratory of Medicinal Chemical Biology, College of Life Sciences, Nankai University, 300071
Tianjin, China
ABSTRACT: The challenge of optimizing several parameters in directed evolution of enzymes remains a
central issue. In this study we address thermostability, enantioselectivity and activity of limonene
epoxide hydrolase (LEH) as the catalyst in the hydrolytic desymmetrization of cyclohexene oxide with
30
ο
formation of (R,R)- and (S,S)-cyclohexane-1,2-diol. Wildtype LEH shows a thermostability of T50 = 41 C
and an enanioselectivity of 2% ee (S,S). Two approaches are described herein. In one strategy, the
mutations generated previously by Janssen, Baker and coworkers for notably increased thermostability
are combined with mutations evolved earlier for enhanced enantioselectivity. Although highly
30
ο
enantioselective (R,R)- and (S,S)-variants (92-93% ee) with increases in T50 by 10-11 C were obtained,
relative to Wildtype LEH the tradeoff in activity was significant. The second strategy based on the
simultaneous optimization of both parameters using iterative saturation mutagenesis (ISM) with
minimized tradeoff in activity proved to be superior. Several notably improved variants were observed, a
reasonable “compromise” being (R,R)- and (S,S)-selective LEH variants (80-94% ee) showing enhanced
ο
thermostability by 5-10 C and still reasonable levels of activity. Analysis of the X-ray structure of the
(
S,S)-variant (94% ee) with and without diol product sheds light on the origin of altered
stereoselectivity.
KEYWORDS: directed evolution, enantioselectivity, thermostability, activity, saturation mutagenesis
3,4
INTRODUCTION
especially in industrial processes.
It has been
5
pointed out that stability promotes evolvability.
Many different strategies for thermostabilization
have been reported, the common feature being
Directed evolution is
engineering the catalytic profiles of enzymes as
catalysts in organic chemistry and biotechnology,
the control of stereo- and regioselectivity being of
particular interest. It constitutes a prolific source
of selective biocatalysts which complement chiral
man-made transition metal catalysts and
a
versatile tool for
1
3,4
the predominance of surface mutations. In most
2
cases the tradeoff in terms of catalytic properties
such as activity remained small or negligible,
which does not surprise since the mutations are
remote from the active site. However, these
additional properties are generally not improved
in the process of thermostabilization. Various
approaches to the optimization of two parameters
2
organocatalysts. The degree of success depends
upon the right choice of the mutagenesis method
and the amount of experimental work that the
researcher is willing to invest. Since the screening
step is the bottleneck, much effort has been
invested in order to maximize the efficacy of
directed evolution.
Directed evolution has also been applied in
order to enhance protein thermostability,
in specific cases have been reported, but general
1-4,6
guidelines remain to be issued.
In most cases
1
,2
substantial experimental efforts had to be
invested. Two general options are possible,
sequential or simultaneous optimization of two or
more parameters.
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Here we describe the directed evolution of source. In a recent study we evolved (R,R)- and
thermostability and enantioselectivity of an (S,S)-selective LEH variants for the model reaction
9
enzyme, while attempting to maintain activity as (Scheme
1).
Structure-guided
saturation
much as possible. Limonene epoxide hydrolase mutagenesis based on three amino acids (in
7
from Rhodococcus erythropolis (LEH) was chosen addition to WT) as building blocks was employed
2
as the catalyst in the hydrolytic desymmetrization at large CAST sites lining the binding pocket in a
of cyclohexene oxide (1) with formation of (R,R)-2 process
and (S,S)-2 (Scheme 1). The goal was to evolve mutagenesis (TCSM).
robust and active (R,R)- and (S,S)-selective harbored both (R,R)- and (S,S)-selective variants,
dubbed
triple
code
saturation
9
The initial libraries
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
variants on an optional basis.
making iterative saturation mutagenesis (ISM)
9
superfluous. The variants were shown to be
distinctly better than those generated in earlier
directed evolution study utilizing other saturation
mutagenesis strategies. Two of the best variants
were identified as:
OH
OH
OH
OH
LEH
O
H O
2
1
0
(S, S)-2
1
(R, R)-2
SZ529 (4 mutations; (R,R)-selective; 97% ee):
M32V/M78V/I80V/L114F and
SZ348 (3 mutations; (S,S)-selective; 99% ee):
I80Y/L114V/I116V
Scheme 1. Hydrolytic desymmetrization catalyzed
by the limonene epoxide hydrolase LEH.
8
In previous studies, thermostability and
Combining the mutations of the variants
evolved for thermostability and enantioselectivity
provided four variants: RR-8 (LEH-P + SZ529),
RR-11 (LEH-F1b + SZ529), SS-8 (LEH-P + SZ348)
and SS-11 (LEH-F1b + SZ348). The catalytic profiles
of the four new variants in terms of
enantioselectivity and thermostability are listed in
Table 1. Thermostability was assessed by
9
enantioselectivity were evolved separately. It was
therefore logical to combine the respective point
8,9
mutations of the best variants, which forms the
first part of the present study. In the second part,
we present a different approach in which these
two parameters in addition to activity are
optimized simultaneously.
30
measuring T50 values, the temperature at which
RESULT AND DISCUSSION
5
0% of enzyme activity is lost following a heat
3
treatment for 30 minutes. The results of kinetic
studies are summarized in Table 2.
Combining mutations previously evolved
for
enhanced
thermostability
and
It can be seen that tradeoffs of some sort
occur in all of the four new variants. If only
enantioselectivity. Janssen, Baker and coworkers
have shown that the thermostability of LEH can
be notably increased by applying a technique
dubbed FRESCO (Framework for Rapid Enzyme
thermostability
and
enantioselectivity
are
considered, the (R,R)-selective variant RR-11 and
the (S,S)-selective variant SS-8 appear to be
acceptable compromises (Table 1, entries 4 and 8,
respectively). However, inspection of Table 2
shows that catalytic efficiency relative to WT LEH
8
Stabilization by Computational libraries). It is a
promising computationally guided approach to
protein thermostabilization in which Rosettaddg,
FoldX and DD (Disulfide Discovery) software
packages are employed. In the case of LEH, several
multi-site mutants resulted, showing an
impressive increase in apparent melting
as measured by kcat/K
Table 2, entries 4 and 8, respectively).
Evolving three parameters
m
is diminished considerably
(
ο
ο
simultaneously. In view of the limited success of
the above approach, a different approach was
needed. In the interest of efficacy in directed
evolution, we intended to keep the screening
effort at a minimum. As in the previous LEH
temperature relative to WT LEH from 50 to 85 C
and a more than 250-fold longer half-life. We
chose two of the best variants for our present
study:
LEH-P (8 mutations):
S15P/A19K/E45K/T76K/T85V/N92K/Y96F/E124D
LEH-F1b (11 mutations):
I5C/S15P/A19K/T76K/E84C/T85V/G89C/S91C/N92
K/Y96F/E124D
Mutations for enhanced and inverted
enantioselectivity were taken from a different
9
study, we employed Reymond’s adrenaline
1
1
on-plate assay. It is a convenient on-plate pre-test
which rapidly identifies active mutants on the
basis of a color change by visual inspection or
1
1
semi-quantitatively by a UV/Vis plate reader.
2
ACS Paragon Plus Environment

Page 3 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
Table 1. Catalytic profiles of the mutants formed by combining mutations of variants previously evolved
8
9
a)
for thermostabilization and enantioselectivity, respectively. Data refers to the model reaction 1 →
R,R)-2 + (S,S)-2 (Scheme 1). The bold faced enzymes are the designations of the new variants formed by
combining mutations.
(
30
50
Entry
Enzyme
T
ee (%) Preferred
enantiomer
Mutations
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ο
1
2
3
WT LEH 41 C
2
(S,S)
(R,R)
(R,R)
ο
SZ529
RR-8
40 C 97
M32V/M78V/I80V/L114F
S15P/A19K/M32V/E45K/T76K/M78V/I
ο
51 C
93
8
0V/T85V/N92K/Y96F/L114F/E124D
ο
4
RR-11
58 C 89
(R,R)
I5C/S15P/A19K/M32V/T76K/M78V/I8
0
V/E84C/T85V/G89C/S91C/N92K/Y9
F/L114F/E124D
6
ο
5
LEH-P
54 C
3
3
(S,S)
(S,S)
S15P/A19K/E45K/T76K/T85V/N92K/Y
6F/E124D
9
ο
6
LEH-F1b 62 C
I5C/S15P/A19K/T76K/E84C/T85V/G89
C/S91C/N92K/Y96F/E124D
ο
7
8
9
SZ348
SS-8
38 C 99
(S,S)
(S,S)
(S,S)
I80Y/ L114V/ I116V
ο
52 C 92
S15P/A19K/E45K/T76K/T85V/N92K/Y
9
6F/E124D/ I80Y/L114V/I116V
ο
SS-11
63 C 87
I5C/S15P/A19K/T76K/I80Y/E84C/T85
V/G89C/S91C/N92K/Y96F/L114V/I116
V/E124D
a)
30
Explanatory notes: ee = enantiomeric excess; T50 : the temperature at which 50% of enzyme activity is
lost following a heat treatment for 30 minutes. The enzyme was incubated at different temperatures
(
30–70°C) for 30 min followed by measuring the residual activity by monitoring the conversion of
substrate by GC using 1- heptanol as the internal standard.
a)
Table 2. Results of kinetic experiments using the model reaction (Scheme 1).
-1
Entry
Enzyme
K
m
(mM)
k
cat (S )
kcat/K
m
-1
-1
(
S M )
1
WT LEH
SZ529
RR-8
RR-11
LEH-P
LEH-F1b
SZ348
SS-8
6.70 ±0.45
10.12 ±0.81
16.26 ±1.28
17.70 ±1.66
7.74 ±0.42
8.79 ±0.65
19.11 ±1.93
15.30 ±1.63
19.32 ±1.18
0.82 ±0.017
0.18 ±0.0051
0.41 ±0.013
0.51 ±0.02
1.03 ±0.018
1.31 ±0.032
0.11 ±0.0048
0.13 ±0.0057
0.19 ±0.0049
122.39
17.78
25.21
28.81
133.07
149.03
5.75
2
3
4
5
6
7
8
8.50
9.84
9
SS-11
a)
The kinetic parameters were measured by monitoring the conversion of substrate by GC using
-heptanol as the internal standard.
1
About 30-40 96-format plates can be assayed per transformants are active, depending upon the
9
day, the active hits then being analyzed for particular randomization site. In the present
enantioselectivity
(Supplementary Table S8). Previous experience
has shown that about 10-30% of the LEH
by
automated
GC project, we decided to limit the screening effort to
9
,10
about 8,000 transformants.
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We planned to apply saturation mutagenesis enantioselectivity and/or activity. CAST is
at two types of LEH residues (Fig. 1): Those that convenient acronym for randomization at sites
can be expected to be sensitive to changes in lining the binding pocket (Combinatorial
a
thermostability on the basis of the previous study Active-site
S15, A19, E45, T76, T85, N92, Y96, E124], and distinguishing it from saturation mutagenesis at
Saturation
Test),
thereby
8
[
those that can be expected to be sensitive to other sites for different purposes.
changes in enantioselectivity and/or activity on
Table 3. Grouping of the 16 chosen LEH residues
into four randomization sites A, B, C and D and
the respective triple codes used in saturation
mutagenesis. The green numbers denote
the basis of the other previous study [M32, L74,
M78, I80, L114, I116, F139, L147]. For clarity, these
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
are mapped in Figure 1 which is based on the
1
2
crystal structure of WT LEH.
“
“
thermostability” positions, the red ones
enantioselectivity” positions.
The 16 positions were grouped into four
potential randomization sites, A, B, C and D, each
containing representatives from each of the two
types of residues (Table 3). The plan was to apply
ISM in the expectation that all three parameters
could be positively influenced, thermostability,
enantioselectivity and activity. Randomization of a
Site
Triple code
A: S15, A19, L74, M78
B: T76, T85, L114, I116
C: N92, Y96, F139, L147
D: E45, E124, M32, I80
K-P-D / V-F-Y
K-P-D / V-F-Y
K-P-D / V-F-Y
K-P-D / V-F-Y
4
-residue site using NNK codon degeneracy
encoding all 20 canonical amino acids would
entail for 95% library coverage excessive screening
Following
saturation
mutagenesis,
the
supernatants of the lysates were placed in the
wells of 96-format micro-titer plates, which were
subjected to a heat treat in an oven at 70 C for 20
minutes. Under these conditions WT LEH loses
about 90% of its activity. Thus, this procedure was
designed so that less stable LEH variants would be
eliminated, although possibly having high
6
2,13
(
3 x 10 transformants). Therefore, triple code
9
saturation mutagenesis (TCSM) based on reduced
amino acid alphabets was considered, which
would require the screening of about 770
transformants for 95% library coverage. When
ο
applying
saturation
mutagenesis
at
a
randomization site, two different triple codes were
chosen in one and the same experiment: One to
enantioselectivity.
A
three-phase screening
procedure was applied: On-plate pre-test for
control
designed
thermostability” residues, and the other at the
enantioselectivity” residues. For this purpose
mutagenesis
at
the
10
activity,
automated
GC
analysis
for
“
“
enantioselectivity followed by automated GC
analysis for conversion. The latter requires a
longer elution time. The best variants were
characterized by sequence determination and
kinetics. Whereas saturation mutagenesis at sites
C and D did not lead to notably improved
variants, the initial libraries at sites A and B
harbored a number of hits. Thereafter, ISM was
applied using the best variants as templates. All
theoretically possible pathways were explored, the
process terminating whenever a given library
failed to harbor improved variants. In these cases
local minima on the fitness pathway landscape
triple codes Lys-Pro-Asp (K-P-D) and Val-Phe-Tyr
V-F-Y) were chosen, respectively. These choices
(
were made on the basis of structural information
and previous studies in which thermostability and
8,9
stereoselectivity were evolved separately.
1
4
were encountered. Although it is possible to
escape from such “dead ends” by utilizing a
non-improved or even inferior mutant as the
14
template in the subsequent ISM step, this
procedure was not attempted in the present case.
Scheme 2 summarizes all results. During the
Figure 1. LEH residues chosen for saturation
entire
directed
evolution
process,
8448
mutagenesis marked in the crystal structure of
transformants were screened in the first phase, to
be followed by 676 and 106 transformants in the
second and third phase, respectively.
1
2
WT LEH. The green-colored residues indicate
positions at which potential mutations may
enhance thermostability, while the red-colored
residues mark CAST positions lining the binding
pocket where potential mutations may influence
4
ACS Paragon Plus Environment

Page 5 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
Tables 4 and 5 display data concerning increase in catalytic efficiency relative to the
thermostability and enantioselectivity of the best previous (S,S)-selective variant SZ348. When
9
variants as well as the results of kinetic comparing variant J-7-A12 with the previous
3
0
experiments, respectively. The assignment of (R,R)-selective variant SZ529, an increase in T50
ο
“
best” hit was made on the basis of improved by about 6 C and a 5-fold enhanced catalytic
enantioselectivity as the main requirement, efficiency are observed. In order to get additional
30
thermostabilization being the second parameter insight concerning the improved T50 , the melting
that was considered.
points of WT LEH (50.3℃) and LEH mutant
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ο
In all of the variants some sort of tradeoff H-2-H5 (92.4 C) were measured by differential
occurs, which is not unexpected. The best scanning calorimetry (DSC) (Fig. 2). Variant
“
T
9
(
compromises” are variants H-2-H5 showing a H-2-H5 shows a significant increase in T (~42.1℃
m
30
ο
50 value of 51 C and an enantioselectivity of ) compared to WT. These results are consistent
30 ο
30
4% ee (S,S), and J-7-A12 (T50 = 46 C; 80% ee with the improvement of the T50 value, and
R,R)). This is not ideal, but significantly better indicates that the mutational changes do not
8,9
than previous variants. Variant H-2-H5 shows a primarily
induce
faster
refolding.
ο
30
13 C improvement in the T value and an 8-fold
50
Scheme 2. ISM exploration of LEH as catalyst in the model reaction (Scheme 1) focusing on
enantioselectivity. A, B, C and D denote 4-residue randomization sites as defined in Table 3.
a)
Table 4. Profiles of LEH variants as catalysts in the model reaction (Scheme 1).
3
0
Entry Enzyme
T
50
ο
ee (%)
Preferred
enantiomer
Mutations
1
WT LEH 41 C
2
(S,S)
(S,S)
(S,S)
(S,S)
ο
2
3
4
B-1-F12
F-6-A9
G-7-H2
44 C
69
80
92
T76K/L114V/I116V
ο
45 C
T76K/L114V/I116V/N92K/F139V/L147F
T76K/L114V/I116V/N92K/F139V/L147F/S
ο
51 C
1
5D/A19K/L74F/M78F
T76K/L114V/I116V/N92K/F139V/L147F/S
5D/A19K /L74F/M78F/ E45D
S15P/M78F
S15P/M78F/N92K/F139V
ο
5
H-2-H5
51 C
94
(S,S)
1
ο
6
7
8
9
A-5-C1
E-5-A8
I-4-H10
J-7-A12
47 C
34
39
45
80
(R,R)
(R,R)
(R,R)
(R,R)
ο
48 C
ο
44 C
S15P/M78F/N92K/F139V/T76K/T85K
S15P/M78F/N92K/F139V/T76K/T85K/E
ο
46 C
4
5D/I80V/E124D
30
Explanatory notes: ee = enantiomeric excess; T50 : the temperature at which 50% of enzyme activity is
lost following a heat treatment for 30 minutes. The enzyme was incubated at different temperatures
(
30–70°C) for 30 min followed by measuring the residual activity by monitoring the conversion of
substrate by GC using 1- heptanol as the internal standard.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a)
Table 5. Kinetic characterization of best LEH variants as catalysts in the model reaction (Scheme 1).
-
1
-1 -1
kcat/K (S M )
m
Entry
Enzyme
K
m
(mM)
k
cat (S )
1
WT LEH
B-1-F12
F-6-A9
G-7-H2
H-2-H5
A-5-C1
E-5-A8
I-4-H10
J-7-A12
6.70 ±0.45
14.41 ±1.40
15.44 ±1.01
19.41 ±1.18
16.10 ±1.08
16.97 ±0.81
16.06 ±0.50
8.23 ±0.43
6.39 ±0.36
13.09±1.76
6.96±0.39
0.82 ±0.017
1.55 ±0.06
0.76 ±0.02
0.82 ±0.02
0.84 ±0.02
1.25 ±0.02
0.86 ±0.01
0.59 ±0.01
0.64 ±0.01
0.93±0.05
0.84±0.01
122.39
107.56
49.22
42.25
52.17
73.66
53.54
71.69
100.15
71.05
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b)
1
0
H-2-H5 (37ºC)
b)
11
J-7-A12 (37ºC)
120.68
a)
The kinetic parameters in entries 1-9 were measured by monitoring the conversion of substrate at 30°C
b)
by GC using 1-heptanol as the internal standard. The kinetic parameters were measured at 37 ºC.
We also performed kinetic experiments and enzyme holds two water molecules by residues
ee-determinations using variants H-2-H5 and R99, D101 and D132, which are located on one side
J-7-A12 at a higher temperature (37 C). The result of the binding pocket. In contrast, the two water
was a 37% and 20% increase in catalytic efficiency molecules are absent in the enzyme/product
relative to the measurements at 30 C, respectively complex, and the binding pocket is occupied by
(Table 5), while maintaining high stereoselectivity; the product diol (S,S)-2. In order to shed light on
H-2-H5: 92% ee (S,S), and J-7-A12 (79% ee (R,R) the reaction process, substrate cyclohexene oxide
(1) was docked into the position of (S,S)-2, the
Relative to the good performance of the structure then being refined for three cycles by
previously reported epoxide hydrolases from PHENIX in order to minimize the steric clashes. It
ο
ο
(Table S3).
1
5
Sphingomonas sp. HXN-200 and Rhodotorula was found that the substrate fits well into the
glutinis, variants H-2-H5 and J-7-A12 show product position. Moreover, the orientation of
1
6
comparable or higher enantioselectivities.
oxygen atom is consistent with the oxygen in
S,S)-2 diol. With this composed structure, we
proceeded to analyze the possible reaction process
catalyzed by the H-2-H5 mutant.
When comparing the structure of the WT
LEH/1-complex obtained by previous docking
experiments, the substrate is now located in a
new pose because the phenyl ring of residue F134
has “turned away” (Fig. 3C and 3E). Additionally,
the sidechain of residue R99 occupies a new
position, thereby also providing more space for
two water molecules to be bound in this region of
the binding pocket (Fig. 3C and 3F). With the
relocation of the sidechains of the two residues,
these new water molecules are tightly coordinated
by R99, D101, D132 and positioned close to the
substrate’s prochiral C1-atom so that the (S,S)-diol
is formed preferentially. In this model, it makes no
(
1
0a
Figure 2. DSC curves of WT LEH (blue) and LEH
mutant H-2-H5 (red).
In order to gain some insight regarding the
source of enantioselectivity improvement, the
crystal structures of the (S,S)-selective variant difference which one of the two water molecules
H-2-H5 with and without complexed product
S,S)-2 were solved at 1.89 Å and 2.30 Å resolution,
participates in the S
N
2 reaction, since both of
(
them are closer to C1 leading to (S,S)-2. It is also
respectively (Fig. 3A and 3B). The two structures noteworthy that the O-atom of the substrate still
are very similar and can be superimposed well, forms the usual hydrogen bond with protonated
showing negligible conformational differences at
most residues. At the active site, the substrate-free
aspartate D101 which is necessary for activation.
We propose that the combination of hydrogen
6
ACS Paragon Plus Environment

Page 7 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
bond network helping to bind and activate LEH, in which the nucleophilic water molecule is
1
0a
substrate 1, two well positioned water molecules, positioned and activated by D132, N55, and Y53.
and certain LEH sidechains accounts for the origin Unfortunately, some residues in the N- and
of the observed (S,S)-selectivity of variant H-2-H5 C- termini are missing in our crystal structures.
Fig. 3D). This model implies catalytic These residues are cleaved off during
(
a
machinery that is somewhat different from WT
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Analysis of active site of LEH variant H-2-H5 in comparison with WT LEH, based on X-ray
structures. A) Active site of variant H-2-H5. B) Structure of active site of variant H-2-H5 complexed with
(S,S)-cyclohexane-1,2-diol. C) Model of substrate 1 and water molecule in active site of WT LEH; WO
represents water molecule. D) Model of substrate 1 and two water molecules in active site of variant
H-2-H5. W1 and W2 represent two separate water molecules. The black and blue dashed lines represent
the hydrogen bond network and the distances from W1, W2 to C1 and C2, respectively. The distances are
labeled in angstrom (Å). E) Model of the alteration of the F134 rotamer conformation resulting in the
change of the location of substrate 1 and water molecules. Orange and light blue sticks represent WT
and H-2-H5, respectively. F) Model of the alteration of the R99 rotamer conformation resulting in the
change of the location of substrate 1 and water molecules. Orange and light blue sticks represent WT
and H-2-H5, respectively. The critical residues are shown as sticks in all of these pictures.
crystallization as judged by SDS-PAGE analysis Additionally, we constructed
a new mutant
using protein crystal samples (Supplementary Fig. H-2-H5-1(E45D/L74F/T76K/M78F/N92K/L114V/I11
S10). During the ISM process, both of the two 6V) lacking the N- and C- terminal mutations, and
improved variants B-1-F12 (T76K/L114V/I116V) and this mutant displays 86% ee in favor of (S,S)-2.
B-4-B11 (T76D/L114V/I116V), which lack the N- Based on these control experiments, it can be
and
enantioselectivity of 71% ee in favor of (S,S)-2 show minimal effect on the enantioselectivity.
Table S2). Similarly, other positive mutants In order to gain some insight regarding the
F-1-E4 (T76K/L114V/I116V/N92D/F139V/L147F), source of enhanced thermostability, we analyzed
F-2-H10 (T76K/L114V/I116V /F139V/L147F) and the stability-related mutations on the crystal WT
F-6-A9 (T76K/L114V/I116V/N92K/F139V/L147F) structure (Fig. 4). Accordingly, mutations E45D,
C-
terminal
mutations,
maintain concluded that the N- and C- terminal mutations
(
lacking the N-terminal mutations, show T76K and N92K are located on or near the surface
enantioselectivity of 82%, 82% and 83% ee, of LEH. It is likely that these mutations stabilize
respectively, in favor of (S,S)-2 (Table S2). the protein by optimizing the distribution of
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. The evolved thermostability-related mutations shown as spheres in WT (left) and variant
H-2-H5 (right). Gray: uncharged amino acids; purple: negatively charged amino acid; red: positively
charged amino acids.
charges on the enzyme surface, which is an mutagenesis (ISM) and reduced amino acid
3, 17
established method of protein stabilization.
alphabets based on triple code saturation
9
Furthermore, S15D may form an ionic bond with mutagenesis (TCSM),
better results were
A19K, thereby stabilizing a flexible N-terminal obtained. Insight regarding the origin of
loop. However, since mutation S15D and A19K in enantioselectivity was gained by analyzing the
the N-terminus are missing in our crystal crystal structures of the (S,S)-selective variant
structure, the respective interaction cannot be with and without complexed product. Due to the
visualized.
CONCLUSION
mutations introduced by TCSM, the substrate and
the nucleophilic water were re-positioned for
smooth and stereoselective ring-opening. This
insight was gained by the analyzing the X-ray
structure of variant H-2-H5, which reveals a
hydrogen bond network instrumental in binding
and activating the substrate. The comparison of
the traditional with the present approach in one
and the same enzyme system may be of help when
performing future multi-parameter optimization
by directed evolution while minimizing the
screening effort. Sequential optimization, e.g.,
directed evolution of thermostability followed by
protein engineering of stereoselectivity and
This study demonstrates once more the challenge
of optimizing several enzyme parameters when
performing directed evolution. It is one of the
1-5
remaining problems in directed evolution for
which a reliable and general strategy still needs to
be developed, especially when considering
industrial applications. In the present study we
attempted to evolve thermostability and
enantioselectivity, while also considering activity.
Two approaches were explored using the epoxide
hydrolase LEH as the catalyst in the hydrolytic
desymmetrization of cyclohexene oxide (1), the
goal being the evolution of both (R,R)- and
1,3
activity also needs increased attention. Finally,
protein engineering of stereoselectivity and
activity followed by stabilization for industrial
(
S,S)-selective variants. In one strategy, mutations
from two previous studies reporting enhanced
thermostability and increased as well as inverted
enantioselectivity, respectively, were combined in
19
applications by means of immobilization also
deserves mention as a possible alternative.
MATERIAL AND METHODS
a
“conventional” approach. This traditional
strategy proved to be only partially successful,
KOD Hot Start DNA Polymerase was
because activity was somewhat reduced. This obtained from Novagen. Restriction enzyme Dpn I
finding may have different reasons, including the was bought from NEB. The oligonucleotides were
possibility that mutational effects on several synthesized by Life Technologies. Plasmid
enzyme properties are not necessarily additive, as preparation kit was ordered from Zymo Research,
1
8
already demonstrated in one-parameter systems.
Upon applying an alternative strategy in QIAGEN. DNA sequencing was conducted by
and PCR purification kit was bought from
which
simultaneously
the
parameters
using
were
iterative
optimized GATC Biotech. All commercial chemicals were
saturation purchased from Sigma-Aldrich or Tokyo Chemical
8
ACS Paragon Plus Environment

Page 9 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
Industry (TCI) or Alfa Aesar. Lysozyme and DNase Amplification of the short fragments of LEH using
I were purchased from AppliChem.
mixed primers F1/R1, F2/R2 and F3/R3 for Library
PCR based methods for site-directed A, B and C respectively. Amplification of the short
mutagenesis and library construction of LEH. fragments of LEH using mixed primers F4/R4 and
Libraries were constructed using the Over-lap PCR F5/R5 for Library D; 2) Over-lap PCR using the
and megaprimer approach with KOD Hot Start PCR products of F4/R4 and F5/R5 as template and
polymerase. 50 µL reaction mixtures typically mixed primers F4/R5; 3) Amplification of the
contained 30 µL water, 5 µL KOD hot start whole plasmid pET22bLEHwt using the PCR
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
polymerase buffer (10×), 3 µL 25 mM MgSO
4
, 5 µL products of F1/R1, F2/R2 and F3/R3 and over-lap
2
mM dNTPs, 2.5 µL DMSO, 0.5 µL (50~100 ng) PCR product of step2 as megaprimers, leading to
template DNA, 100 µM primers Mix (0.5 µL each) the final variety plasmids for library generation.
and 1 µL KOD hot start polymerase. The PCR Primers are listed in Supplementary Table S6. The
conditions for short fragment: 95 °C 3 min, (95 °C PCR products were digested by Dpn I and
3
°
0 sec, 56 °C 30 sec, 68 °C 40 sec) × 30 cycles, 68 transformed
C 120 sec. For mega-PCR: 95 °C 3 min, (95 °C 30 coliBL21(DE3) to create the library for screening.
sec, 60 °C 30 sec, 68 °C 5 min 30 sec) × 24 cycles, Other Libraries were created using the same
8 °C 10 min. The PCR products were analyzed on procedure as mentioned above. All the primers
agarose gel by electrophoresis and purified using a used are listed in Supplementary Table S7.
Qiagen PCR purification kit. 2 µL NEB CutSmart™ Screening Procedures. Colonies were
into
electro-competent
E.
6
Buffer and 2 µL Dpn I were added in 50 µL PCR picked up and inoculated into deep-well plates
reaction mixture and the digestion was carried out containing 300 µL LB medium with 50 µg/ml
at 37 °C for 7 h. After Dpn I digestion, the PCR carbenicillin and cultured overnight at 37 °C with
products 1.5 µL were directly transformed into shaking. An aliquot of 120 µL was transferred to
electro-competent E. coli BL21(DE3) to create the glycerol stock plate and stored at -80 °C. Then,
final library for Quick Quality Control and 800 µL TB medium with 0.5% (m/v) lactose and 50
screening.
Primer
µg/mL carbenicillin was added directly to the
design
and
site-directed culture plate for 8 h at 28 °C with shaking for
mutagenesis of LEH. Primer design depend protein expression. The cell pellets were harvested
upon the particular amino acid residues sites and and washed with 400 µL 50 mM pH 7.4 potassium
mutagenesis, and in the case of LEH mutant phosphate buffer and centrifuged for 10 min 4000
LEH-P creation eight residues which were divided rpm at 4 °C. Then, the pellets were resuspended in
into four groups (Supplementary Fig.S1): 1) 400 µL of the same buffer with 6 U/mL DNase I
Amplification of the short fragments of LEH using and 1 mg/mL lysozyme for breaking the cell at 30
mixed
LEH-2-F/LEH-3-R
primers
and
LEH-1-F/LEH-2-R, °C for 1 h with shaking. The crude lysate was
LEH-3-F/LEH-4-R, centrifuged for 30 min 4000 rpm at 4°C. 50 µL of
respectively; 2) Over-lap PCR using the products the
supernatant
was
transferred
into
of step as template and mixed primers 96-well-micrioplate and heated at 70°C for 20
1
LEH-1-F/LEH-4-R; 3) Amplification of the whole minutes at oven (The 96-well-micrioplate has
plasmid pET22bLEHwt using the products of step protection function for the inside supernatant.
2
as megaprimers, leading to the final mutagenesis After heated at 70°C for 20 minutes at oven, the wt
plasmids for library generation. Primers are listed LEH remain about 10% activity, so we select this
in Supplementary Table S5. The PCR products condition for the high-throughput screening of
were digested by Dpn I and transformed into thermostability).
electro-competent E. coliBL21(DE3) to create the supernatant 50 ul was used for an adrenaline
library. assay. 110 µL potassium phosphate buffer (50 mM,
Mutants LEH-F-b, SS-8, RR-8, SS-11, RR-11 pH 7.4) with 5% acetonitrile and substrate 1 (final
and H-2-H5-1 were created using the same concentration 10 mM) were added. The plates
procedure as mentioned above (Supplementary were incubated at 30 °C, 500 rpm, 1 h. Afterwards,
Then
the
heat-treated
Figure S2-S7). All the primers used are listed in NaIO
Supplementary Table S5. was added and the plates were further incubated
4
solution 20 µL (77 mg in 24 mL of water)
Primer design and library creation of at 30 °C, 500 rpm, 10 min. Subsequently,
LEH. Primer design and library construction adrenaline solution 20 µL (epinephrine 132 mg,
depend upon the particular amino acid chosen, water 24 mL, concentrated HCl 5 drops for
and in the case of LEH this involves sixteen solubilizing the adrenaline) was added, which
residues which were divided into four groups caused the immediate formation of a red color in
(Supplementary Fig. S8 and Fig. S9): 1) the inactive reactions. Active variants gave lighter
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
colour relative to WT. 300 µL rest supernatant of min. The EtOAc layer was analyzed by GC (The
the active transformants was transferred into new column is Hydrodex-β-TBDAc, 25 m x 0.25 mm
deep-well plates for reaction with 10 mM substrate ID).
1
and 5% acetonitrile as co-solvent for 5 h at 30 °C
Effects of temperature on enzyme activity.
8
00rpm, the final volume was 400 µL. The product For thermostability, the enzyme was incubated at
and remaining substrate were extracted using different temperatures (30–70°C) for 30 min
equal volumes of ethyl acetate (EtOAc) for GC followed measuring the residual activity by
analysis by chiral column (Supplementary Table monitoring the conversion of substrate by GC
S8). The screening results were shown in using 1- heptanol as the internal standard.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Supplementary Table S2.
Differential scanning calorimetry. The
Protein expression and purification. The protein samples of WT LEH and LEH mutant
WT and positive mutants were inoculated in 5 mL H-2-H5 were dissolved in the 50 mM potassium
LB containing 50 µg/mL carbenicillin and cultured phosphate buffer (pH 8.0) with
a
final
overnight at 37 °C with shaking, and then scaled concentration of 12mg/mL. DSC examinations of
up to 500 mL TB containing 50 µg/mL the sample were carried out using the instrument
carbenicillin. When the OD600 reached 0.8, 0.4 Rigaku Thermo plus EVO2 DSC8231. DSC scans
mM IPTG was added to induce the protein were taken from 20 to 100°C at a heating rate of
expression. The cell cultures continued to grow for 5°C/min. The 50 mM potassium phosphate buffer
10 hours at 25 °C before being harvested by (pH 8.0) was used as a reference. The heat flow
centrifugation at 6,000 × g and resuspended in a versus temperature plot of the WT LEH and LEH
PBS buffer (20 mM, pH 7.4) containing 500 mM mutant H-2-H5 was obtained after subtraction of
NaCl, 20 mM imidazole. The cell pellets were the buffer signal. The peak value of the DSC plot
8
disrupted by sonication and the cell debris was was considered as the melting point (T
m
).
removed by centrifugation at 15,000 × g for 60
X-ray structural analysis. The H-2-H5
min. The soluble protein sample was loaded onto crystals were grown in 0.1 M HEPES (pH 7.5), 2%
a nickel affinity column (GE Healthcare) and v/v Polyethylene glycol 400 and 2.0 M ammonium
washed with 20~500 mM imidazole solution sulfate by the sitting-drop vapor diffusion method
containing 500 mM NaCl and 20 mM PBS buffer at 20 °C. Proteins at 10 mg/mL were mixed in a 1:1
(
pH 7.4). Proteins from the flow through were ratio with the reservoir solution in a final volume
pooled and concentred, and then loaded onto a of 2 μL and equilibrated against the reservoir
HiTrap™ Q HP 5 ml column (GE Healthcare) and solution. Single crystals of H-2-H5 were obtained
eluted with 0~1000 mM Nacl solution containing directly, whereas the complex structures of
5
00 mM Tris buffer (pH 8.0). The eluted protein H-2-H5-(S,S)-cyclohexane-1,2-diol were obtained
sample was pooled and concentred, and then by soaking the crystals in growth buffer
loaded onto a Hiload™ 16/600 Superdex™ 200 pg containing 100 mM of epoxide for 1 min. All
column (GE Healthcare) and eluted with 50 mM crystals were mounted in nylon loops and
Tris (pH 8.0), 500 mM Nacl. Fractions containing flash-frozen in liquid nitrogen. Diffraction data of
highly pure LEH proteins were pooled and all crystals was collected at the wave length of
concentrated with Centricon filtration devices 0.9792 Å using an ADSC Quantum 315r detector at
(
Amicon) to 10 mg
were determined by Bradford.
The measurement of kinetic parameters were indexed, integrated, and scaled using the
/
ml. The protein concentrations beamline BL17U of Shanghai Synchrotron
Radiation Facility (SSRF) at 100 K. All data sets
2
0
of purified enzyme. The enzymatic hydrolysis HKL2000 package. The structures of H-2-H5 and
rate was measured by monitoring the conversion H-2-H5-(S,S)-cyclohexanediol complex were
of substrate by GC using 1-heptanol as the internal solved by molecular replacement method using
2
1
standard. Pure enzymes were kept at 30°C or 37°C the program PHASER and the coordinate of
for 10min and added to potassium phosphate wild-type LEH (PDB code 1NU3) as a search
buffer (50 mM, pH 7.4) with a total volume of 400 model. Rounds of automated refinement were
2
2
μL containing epoxide of varied concentration performed with PHENIX and the models were
2
3
(2–64 mM) and 2.5% (v/v) of acetonitrile, and the extended and rebuilt manually with COOT. The
reaction was performed at 30°C or 37°C for 10min structures of H-2-H5 and
with shaking (800 rpm). The reaction was H-2-H5-(S,S)-cyclohexane-1,2-diol complex have
terminated by the addition of 400 μL EtOAc been refined to 2.30 and 1.89 Å, respectively. The
containing 2.0 mM 1-heptanol. The mixture was statistics for data collection and crystallographic
vigorously mixed and the organic layer was refinement are summarized in Supplementary
separated by centrifugation at 4,000 × rpm for 15
10
ACS Paragon Plus Environment

Page 11 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
Table S4. All structural figures were prepared Clark, D. S. Annu. Rev. Chem. Biomol. 2012, 3, 77-102; (d)
Suplatov, D.; Voevodin, V.; Švedas, V. Biotech. J. 2015,
using Pymol (http://www.pymol.org/).
10, 344-355; (e) Socha, R. D.; Tokuriki, N. FEBS. J. 2013,
2
80, 5582-5595; (f) Bommarius, A. S.; Paye, M. F. ChSRv.
ASSOCIATED CONTENT
2013, 42, 6534-6565; (g) Yu, H.; Huang, H. Biotechnol.
Supporting Information: The figures of primer Adv. 2014, 32, 308-315.
(4) (a) Rathi, P. C.; Jaeger, K.-E.; Gohlke, H. PLoS. ONE.
design and creation of LEH mutants and libraries,
results of libraries screening, data collection and
refinement statistics of H-2-H5 crystal structure,
list of primers, SDS-PAGE of H-2-H5 and analytic
conditions of GC are available free of charge via
the Internet at http://pubs.acs.org.
2015, 10, e0130289; (b) Yong, K.; Scott, D. Biotechnol.
Bioeng. 2015, 112, 438-446; (c) Huang, L.; Xu, J.-H.; Yu,
H.-L. J. Biotechnol. 2015, 203, 54-61; (d) Fulton, A.;
Frauenkron-Machedjou, V. J.; Skoczinski, P.; Wilhelm,
S.; Zhu, L.; Schwaneberg, U.; Jaeger, K. E.
ChemBioChem. 2015, 16, 930-936; (e) Bednar, D.;
Beerens, K.; Sebestova, E.; Bendl, J.; Khare, S.;
Chaloupkova, R.; Prokop, Z.; Brezovsky, J.; Baker, D.;
Damborsky, J. PLoS. Comput.Biol. 2015, 11, e1004556. (f)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Author
Lamazares,
Velázquez-Campoy, A.; Sancho, J. Sci. Rep. 2015, 5, 9129.
5) (a) Bloom, J. D.; Labthavikul, S. T.; Otey, C. R.;
Arnold, F. H. Proc. Nat. Acad. Sci. USA. 2006, 103,
869-5874. (b) Tokuriki, N.; Tawfik, D. S. Curr. Opin.
E.;
Clemente,
I.;
Bueno,
M.;
*
Correspondence should be addressed to (M.T.R.)
reetz@mpi-muelheim.mpg.de and (Xinqi Liu)
liu2008@nankai.edu.cn
(
5
Author Contributions
Struct. Biol. 2009, 19, 596-604.
§
These authors contributed equally.
(6) (a) Li, T.; Liang, J.; Ambrogelly, A.; Brennan, T.;
Gloor, G.; Huisman, G.; Lalonde, J.; Lekhal, A.; Mijts, B.;
Muley, S. J. Am. Chem. Soc. 2012, 134, 6467-6472; (b)
Savile, C. K.; Janey, J. M.; Mundorff, E. C.; Moore, J. C.;
Tam, S.; Jarvis, W. R.; Colbeck, J. C.; Krebber, A.; Fleitz,
F. J.; Brands, J. Science. 2010, 329, 305-309; (c) Zhang,
D.; Chen, X.; Chi, J.; Feng, J.; Wu, Q.; Zhu, D. ACS. Catal.
Notes
The authors declare no competing financial
interest.
ACKNOWLEDGEMENTS
2015, 5, 2452-2457; (d) DeSantis, G.; Wong, K.; Farwell,
Support from the Max-Planck-Society and the
B.; Chatman, K.; Zhu, Z.; Tomlinson, G.; Huang, H.; Tan,
LOEWE Research Cluster SynChemBio is gratefully X.; Bibbs, L.; Chen, P. J. Am. Chem. Soc. 2003, 125,
acknowledged. We thank Dr. Richard Lonsdale for 11476-11477; (e) Yamada, R.; Higo, T.; Yoshikawa, C.;
constructing Fig. 1.
China, H.; Ogino, H. J. Biotechnol. 2014, 192, 248-254; (f)
Qian, C.; Liu, N.; Yan, X.; Wang, Q.; Zhou, Z.; Wang, Q.
Enzyme. Microb. Technol. 2015, 70, 35-41; (g) Liang, C.;
Gui, X.; Zhou, C.; Xue, Y.; Ma, Y.; Tang, S.-Y. Appl.
Microbiol. Biotechnol. 2015, 99, 2673-2682; (h) Asako,
H.; Shimizu, M.; Itoh, N. Appl. Microbiol. Biotechnol.
2008, 80, 805-812; (i) Boehlein, S. K.; Shaw, J. R.;
Stewart, J. D.; Sullivan, B.; Hannah, L. C. Arch. Biochem.
Biophys. 2015, 568, 28-37; (j) Gong, Y.; Xu, G.-C.; Chen,
Q.; Yin, J.-G.; Li, C.-X.; Xu, J.-H. Catal. Sci. Technol.
2016, DOI: 10.1039/C5CY01723H.
REFERENCES
(
1) Recent reviews of directed evolution: (a) Bommarius,
A. S. Annu. Rev. Chem. Biomol. Eng. 2015, 6, 319-345; (b)
Gillam, E. M. J.; Copp, J. N.; Ackerley, D. F. Directed
Evolution Library Creation. Methods in Molecular
Biology, Humana Press, Totowa, 2014; (c) Denard, C. A.;
Ren, H.; Zhao, H. Curr. Opin. Chem. Biol. 2015, 25, 55-64;
(d) Reetz, M. T. Directed Evolution of Enzymes. Enzyme
Catalysis in Organic Synthesis, Wiley-VCH, Weinheim,
2
012, Vol. 1, pp.119-190; (e) Jäckel, C.; Hilvert, D. Curr.
(7) (a) van der Werf, M. J.; Overkamp, K. M.; de Bont, J.
A. J. Bacteriol. 1998, 180, 5052-5057; (b) Vanderwerf, M.;
Orru, R.; Overkamp, K.; Swarts, H.; Osprian, I.;
Steinreiber, A.; de Bont, J.; Faber, K. Appl. Microbiol.
Biotechnol. 1999, 52, 380-385.
Opin. Biotechnol. 2010, 21, 753-759; (f) Brustad, E. M.;
Arnold, F. H. Curr. Opin. Chem. Bio. 2011, 15, 201-210; (g)
Goldsmith, M.; Tawfik, D. S. Curr. Opin. Struct. Biol.
2012, 22, 406-412; (h) Widersten, M. Curr. Opin. Chem.
Biol. 2014, 21, 42-47; (i) Currin, A.; Swainston, N.; Day, P.
J.; Kell, D. B. Chem. Soc. Rev. 2015, 44, 1172-1239; (j)
Turner, N. J. Nat. Chem. Biol. 2009, 5, 567-573; (k)
Siloto, R. M. P.; Weselake, R. J. Biocatal. Agric.
Biotechnol. 2012, 1, 181-189; l) Lutz, S.; Bornscheuer, U. T.
Protein Engineering Handbook. Wiley-VCH, Weinheim,
(8) Wijma, H. J.; Floor, R. J.; Jekel, P. A.; Baker, D.;
Marrink, S. J.; Janssen, D. B. Protein. Eng. Des. Sel. 2014,
27, 49-58.
(9) Sun, Z.; Lonsdale, R.; Lian, W.; Li, G.; Li, A.; Wang, J.,
Zhou, J., Reetz, M. T. ACS. Catal. 2016, 6, 1590-1597.
(10) (a) Sun, Z.; Lonsdale, R.; Kong, X. D.; Xu, J. H.;
2
009; (m) Sun, Z.; Wikmark, Y.; Bäckvall, J.-E., Reetz, M. Zhou, J.; Reetz, M. T. Angew. Chem. Int. Ed. 2015, 54,
T. Chem. Eur. J. 2016, 22, 5046-5054.
12410-12415; (b) Zheng, H.; Reetz, M. T. J. Am. Chem.
(
(
2) Reetz, M. T. Angew. Chem. Int. Ed. 2011, 50, 138-174.
3) (a) Eijsink, V. G.; Gåseidnes, S.; Borchert, T. V.;
Soc. 2010, 132, 15744-15751.
(11) (a) Wahler, D.; Reymond, J. L. Angew. Chem. Int. Ed.
2002, 41, 1229–1232; (b) Reymond, J. L. Enzyme Assays:
High-Throughput Screening, Genetic Selection and
Fingerprinting, Wiley-VCH, Weinheim, 2006.
vandenburg, B. Biomol. Eng. 2005, 22, 21-30; (b) Polizzi,
K. M.; Bommarius, A. S.; Broering, J. M.;
Chaparro-Riggers, J. F. Curr. Opin. Chem. Biol. 2007, 11,
220-225; (c) Liszka, M. J.; Clark, M. E.; Schneider, E.;
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
12) Arand, M.; Hallberg, B. M.; Zou, J.; Bergfors, T.;
(18) Review of non-additive mutational effects: Reetz,
M. T. Angew. Chem. Int. Ed. 2013, 52, 2658-2666.
(19) (a) Choi, J. M.; Han, S. S.; Kim, H. S. Biotechnol.
Adv. 2015, 33, 1443-1454; (b) Guisan, J. M. (ed.).
Immobilization of Enzymes and Cells, 3rd ed.,
Springer-Verlag, New York, 2013.
Oesch, F.; van der Werf, M. J.; de Bont, J. A.; Jones, T.
A.; Mowbray, S. L. EMBO. 2003, 22, 2583-2592.
(
13) (a) Patrick, W. M.; Firth, A. E. Biomol. Eng. 2005, 22,
105-112; (b) Denault, M.; Pelletier, J. N. Protein
Engineering Protocols, Humana Press, NJ, 2007, Vol.
352, pp 127-154; (c) Nov, Y. Appl. Environ. Microbiol.
2012, 78, 258-262.
(20) Otwinowski, Z.; Minor, W. Method. Enzymol. 1997,
276, 307-326.
(
14) Gumulya, Y.; Sanchis, J.; Reetz, M. T. ChemBioChem. (21) Mccoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.;
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
(
7
012, 13, 1060-1066.
15) Wu, S.; Li, A.; Chin, Y.S.; Li, Z. ACS. Catal. 2013, 3,
52-759.
16) Weijers, C.A. Tetrahedron: Asymmetry 1997, 8,
39-647.
17) (a) Gribenko, A.V.; Patel, M. M.; Liu, J. J.;
M.D., W.; Storoni, L. C.; Read, R. J., J. Appl. Crystallogr.
2007, 40, 658-674.
(22) Adams, P. D.; Afonine, P. V.; Bunkóczi, G.; Chen, V.
B.; Davis, I. W.; Echols, N.; Headd, J. J.; Hung, L.-W.;
Kapral, G. J.; Grosse-Kunstleve, R. W. Acta. Crystallogra.
Sect. D-Biol. Crystallogr. 2010, 66, 213-221.
(23) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K.
Acta Crystallogra. Sect. D-Biol. Crystallogr. 2010, 66,
486-501.
(
6
(
McCallum, S. A.;Wang, C. Y.; Makhatadze, G. I. Proc.
Nat. Acad. Sci. USA 2009, 106, 2601-2606; (b) Lee, C. F.;
Makhatadze, G. I.; Wong, K. B. Biochemistry 2005, 44,
1
6817-16825; (c) Loladze.; V. V., Ibarra, M, B.;
Sanchez-Ruiz, J.M.; Makhatadze, G. I. Biochemistry
999, 38, 16419-16423 (1999).
1
12
ACS Paragon Plus Environment

Page 13 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment