Structure-Guided Triple-Code Saturation Mutagenesis: Efficient Tuning of the Stereoselectivity of an Epoxide Hydrolase

Article abstract of DOI:10.1021/acscatal.5b02751

The directed evolution of enzymes promises to eliminate the long-standing limitations of biocatalysis in organic chemistry and biotechnology - the often-observed limited substrate scope, insufficient activity, and poor regioselectivity or stereoselectivity. Saturation mutagenesis at sites lining the binding pocket with formation of focused libraries has emerged as the technique of choice, but choosing the optimal size of the randomization site and reduced amino acid alphabet for minimizing the labor-determining screening effort remains a challenge. Here, we introduce structure-guided triple-code saturation mutagenesis (TCSM) by encoding three rationally chosen amino acids as building blocks in the randomization of large multiresidue sites. In contrast to conventional NNK codon degeneracy encoding all 20 canonical amino acids and requiring the screening of more than 10¹⁵ transformants for 95% library coverage, TCSM requires only small libraries not exceeding 200-800 transformants in one library. The triple code utilizes structural (X-ray) and consensus-derived sequence data, and is therefore designed to match the steric and electrostatic characteristics of the particular enzyme. Using this approach, limonene epoxide hydrolase has been successfully engineered as stereoselective catalysts in the hydrolytic desymmetrization of meso-type epoxides with formation of either (R,R)- or (S,S)-configurated diols on an optional basis and kinetic resolution of chiral substrates. Crystal structures and docking computations support the source of notably enhanced and inverted enantioselectivity.

Full text of DOI:10.1021/acscatal.5b02751

Subscriber access provided by Weizmann Institute of Science
Article
Structure-Guided Triple Code Saturation Mutagenesis: Efficient
Tuning of the Stereoselectivity of an Epoxide Hydrolase
Zhoutong Sun, Richard Lonsdale, Lian Wu, Guangyue Li,
Aitao Li, Jianbo Wang, Jiahai Zhou, and Manfred T. Reetz
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b02751 • Publication Date (Web): 22 Jan 2016
Downloaded from http://pubs.acs.org on January 28, 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 10
ACS Catalysis
1
2
3
4
5
6
7
8
9
Structure-Guided Triple Code Saturation Mutagenesis: Efficient Tun-
ing of the Stereoselectivity of an Epoxide Hydrolase
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
Zhoutong Sun1,2, Richard Lonsdale1,2, Lian Wu3, Guangyue Li1,2, Aitao Li1,2, Jianbo Wang1,2, Jiahai
Zhou3,*, Manfred T. Reetz1,2,*
1Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr, Germany
2Fachbereich Chemie der Philipps-Universität, Hans-Meerwein-Strasse, 35032 Marburg, Germany
3State Key Laboratory of Bio-organic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry,
Chinese Academy of Sciences, Shanghai 200032, China
ABSTRACT: Directed evolution of enzymes promises to eliminate the long-standing limitations of biocatalysis in organic
chemistry and biotechnology - the often observed limited substrate scope, insufficient activity and poor regio- or stere-
oselectivity. Saturation mutagenesis at sites lining the binding pocket with formation of focused libraries has emerged as
the technique of choice, but choosing the optimal size of the randomization site and reduced amino acid alphabet for
minimizing the labor-determining screening effort remains a challenge. Here we introduce structure-guided triple-code
saturation mutagenesis (TCSM) by encoding three rationally chosen amino acids as building blocks in the randomization
of large multi-residue sites. In contrast to conventional NNK codon degeneracy encoding all 20 canonical amino acids and
requiring the screening of more than 1015 transformants for 95% library coverage, TCSM requires only small libraries not
exceeding 200~800 transformants in one library. The triple-code utilizes structural (X-ray) and consensus-derived se-
quence data, and is therefore designed to match the steric and electrostatic characteristics of the particular enzyme. Us-
ing this approach, limonene epoxide hydrolase has been successfully engineered as stereoselective catalysts in the hydro-
lytic desymmetrization of meso-type epoxides with formation of either (R,R)- or (S,S)-configurated diols on an optional
basis and kinetic resolution of chiral substrates. Crystal structures and docking computations support the source of nota-
bly enhanced and inverted enantioselectivity.
KEYWORDS: amino acid alphabet, directed evolution, epoxide hydrolases, saturation mutagenesis, stereoselectivity,
triple code.
INTRODUCTION
Since screening is the labor-intensive step (bottleneck)
when evolving enhanced or inverted stereoselectivity,1,8
various attempts have been made for generating higher-
quality mutant libraries,1,9 including the use of reduced
amino acid alphabets in SM,1e,10,11 Application of the Pat-
rick/Firth algorithm12 or the Nov metric13 enables the es-
timation of oversampling as a function of the size of the
randomization site and of the number of amino acids
used as building blocks1e,10,11(Supporting Information Table
S1). Extensive multi-residue sites in combination with
large amino acid alphabets (up to 20 canonical amino
acids as defined by NNK codon degeneracy) ensure max-
imal structural diversity, but such scenarios maximize the
screening effort, e.g., up to ≈1015 transformants in the case
of a 10-residue site for 95% library coverage. To date,
uncertainty persists concerning the optimal choice of a
reduced amino acid alphabet and the number of residues
in a randomization site. When opting for reduced amino
acids in SM experiments, two different strategies can be
applied: 1) Use of one and the same codon degeneracy for
the whole multi-residue randomization site,10a-b or 2) use
of a different codon degeneracy at each position of such a
site.10c,11
Directed evolution of stereoselective enzymes constitutes
a prolific source of biocatalysts for asymmetric reaction.1
In this endeavor the most commonly applied gene muta-
genesis methods are error-prone polymerase chain reac-
tion (epPCR), DNA shuffling and saturation mutagenesis
(SM).1 SM at sites lining an enzyme binding pocket is a
particularly powerful way to evolve stereo- and regioselec-
tive enzymes as catalysts in fine chemicals manufactur-
ing.1a,e If the initial libraries fail to harbor hits showing
optimal catalytic profiles, iterative saturation mutagenesis
(ISM) can be invoked, which consists of consecutive
rounds of SM. SM has been used in other important ap-
plications, including the degradation of phosphorus-
based pesticides and chemical warfare agents,2,3 and in
potential universal blood production.4 When evolving
stereo- and/or regioselectivity, selection platforms5 or
FACS-based display systems,6 normally capable of “han-
dling” super large libraries (≈1010), are generally not possi-
ble,7 which means that screening assays need to be ap-
plied.7,8
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 10
sen for each of the randomization sites: A (I80/V83/L114/I116,
B (L74/M78/L147) and C (M32/L35/L103).
In the present study we apply “triple code saturation mu-
tagenesis” (TCSM) in SM of relatively large multi-residue
randomization sites using strategy 1 as a practical strategy
for generating focused high-quality libraries requiring
minimal screening (Fig. 1). Utilizing structural, mechanis-
tic and/or sequence10c data of an enzyme under study, a
rational choice is made regarding three appropriate ami-
no acids as building blocks in SM, in addition to the
wildtype (WT) amino acid. Then, 10 residues surrounding
the binding pocket are identified on the basis of the crys-
tal structure followed by grouping them into typically 3-
or 4-residue randomization sites. This is necessary be-
cause application of TCSM at a 10-residue site requires the
screening of 3.14 x 106 transformants for 95% library cov-
erage. These choices constitute a viable “compromise”
between structural diversity and amount of screening as
shown by statistical analyses.12 Following SM and screen-
ing, ISM can be applied if necessary (Fig. 1). In order to
illustrate the efficacy of this approach, we have applied
the method to limonene epoxide hydrolase (LEH)14 which
has been characterized by X-ray crystallography.15 The
choice of this particular enzyme as the model system was
made in view of previous directed evolution studies of
LEH using other SM-based strategies,16,17 thereby allowing
direct comparison of the different approaches.
1
2
3
4
5
6
7
8
RESULTS AND DISCUSSION
Evolving enantioselective mutants of the epoxide
hydrolase LEH. As the model reaction we chose the
hydrolytic desymmetrization of cyclohexene oxide (1)
with formation of (R,R)- and (S,S)-2 (Scheme 1). WT LEH
results in minimal enantioselectivity (ee = 4%) in favor of
(S,S)-2.
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
Scheme 1. Hydrolytic desymmetrization of substrate 1
catalyzed by LEH.
Based on the crystal structure of LEH,15 10 residues lining
the binding pocket were chosen for potential SM. Apply-
ing TCSM over such an extensive site would require the
screening of about 106 transformants for 95% library cov-
erage. Therefore, they were grouped into three randomi-
zation sites A (I80/V83/L114/I116, B (L74/M78/L147) and C
(M32/L35/L103) (Fig. 2).
I80
L103
M78
L114
V83
L147
1
I116
L74
L35
M32
Figure 2. The 10 residues lining the binding pocket of LEH,
assigned to three randomization sites A (blue), B (green) and
C (yellow). Residues were selected based on docking of sub-
strate 1 to the WT X-ray structure (PDB: 1NU3).15
At this point a decision had to be made concerning the
choice of the three amino acids as building blocks in
TCSM. We were primarily guided by the LEH crystal
structure15 which shows that the binding pocket is lined
by residues having hydrophobic character. This suggests
that amino acids with hydrophobic sidechains should be
chosen, as in previous SM-based studies.16,17 In order to
gain further confidence, we applied the consensus ap-
proach in directed evolution18 by first aligning 100 limo-
nene epoxide hydrolases and then focusing on the respec-
tive 10 residues lining the corresponding binding pockets.
Figure 1. Steps in structure-guided triple code saturation
mutagenesis (TCSM) as illustrated in the directed evolution
of limonene epoxide hydrolase (LEH) as the catalysts in
stereoselective transformations. The green letters in the
parentheses (top) indicate the chosen triple code which
defines the reduced amino acid alphabet used in SM of LEH.
In this work, one and the same triple code (V-F-Y) was cho-
ACS Paragon Plus Environment

Page 3 of 10
ACS Catalysis
Figure 3a, the data for the other two libraries are summa-
The dominance of hydrophobic amino acids was clearly
1
2
3
4
5
6
7
8
confirmed (Fig. S1), such amino acids as valine, leucine,
isoleucine and phenylalanine being typical. Tyrosine is
not included in the list, but we considered it as well be-
cause its para-hydroxyl group could engage in H-bonding.
In our previous study,16 substitution using this amino acid
had moderately positive effects.
rized in the SI (Table S2). It can be seen that three vari-
ants show (S,S)-selectivity in the range of ee = 96-99%,
while the best (R,R)-selective variant I80V/L114F is not
quite as selective (ee = 89%). Therefore, the latter was
used as a template for ISM at sites B and C employing the
triple code Val/Phe/Tyr, which led to distinctly improved
(R,R)-selective mutants (ee = 97%) (Fig. 3b). Kinetic data
of the best mutants can be found in the SI Table S3. The
total search required the screening of only 1728 trans-
formants. As shown below in the Conclusions and Per-
spectives, the TCSM method is more efficient than previ-
ous approaches based on other SM techniques for engi-
neering stereoselective LEH variants.16,17
Based on the sum of these information-based guides, we
chose valine (V), phenylalanine (F) and tyrosine (Y) as the
triple code. Randomization at sites A, B, and C requires a
screening effort for 95% library coverage of only 576, 192
and 192 transformants, respectively.
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
Catalytically active variants in libraries A, B and C were
rapidly identified by the adrenaline on-plate pretest,19
followed by GC analysis for enantioselectivity of the active
hits. Library A contained the best mutants as shown in
Figure 3. (a) LEH variants originating from library A as catalysts in the model reaction of epoxide 1. (b) Improved (R,R)-selective
variants resulting from ISM at sites B (purple arrow) and C (light green arrow).
Understanding the source of evolved LEH enantiose-
lectivity. Structural and theoretical analyses of the
evolved mutants provide detailed insight into the reshap-
ing of the active site and the effect on substrate binding.
X-ray structures were obtained for the greatly improved
(S,S)-selective variant SZ348 (I80Y/L114V/I116V - 3.0 Å
resolution) and the stereo-inverted (R,R)-selective variant
SZ529 (M32V/M78V/I80V/L114F - 2.25 Å resolution).
Structural comparison between the wild-type (WT, PDB
code 1NU3) and the two mutants shows very little differ-
ence in tertiary structure (SI Fig. S2). The overall struc-
tural elements can be superimposed with a root-mean-
square deviation (r.m.s.d.) of 0.24 and 0.688 Å for all
atoms between the WT and the SZ348 and SZ529 mutants,
respectively. The greater deviation in the case of SZ529 is
mainly due to the C-terminal helix (residues 133-143 are 6-
residues shorter than the corresponding helix residues
133-149 in WT and SZ348), which has rotated by ~ 35 de-
grees towards the active site pocket (Fig. 4a and SI Fig.
S2). Figures 4c and 4d reveal that the variants SZ529 and
SZ348 have undergone a significant change in shape of
Figure 4. (a) An enlarged view of the structural alignments
of WT, SZ529 and SZ348. The highlighted C-terminal helix
was colored in magenta for WT, cyan for SZ529, and green
for SZ348, respectively. Mutations resulted in 6 residues
missing and a 35° rotation of the C-terminal helix in the
their active sites relative to WT LEH (Fig. 4b). These re-
sults clearly show that the correlation between reshaped
binding pockets and altered stereoselectivity. Changes in
hydrophobicity at the residues lining the binding pocket
are shown in the SI (Fig. S3).
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 10
SZ529 structure. Alterations in the shape of the LEH binding
atom of the epoxide ring at which nucleophilic attack
1
2
3
4
5
6
7
8
pocket induced by SM-based directed evolution as revealed
by crystal structures. Surface representation of the binding
pockets of WT LEH (b) and variants SZ529 (c) and SZ348 (d)
reveal the changes in shape; D101 marks the aspartate that is
known to form an activating H-bond to the O-atom of epox-
ides.
occurs determines the stereochemistry of the diol product.
In the case of both mutants, the epoxide carbon atom
closest to the nucleophilic water molecule in the docking
poses is consistent with the experimentally observed ste-
reoselectivity. The positioning of Y80 in the SZ348 mutant
(Fig. 5b) suggests that this residue may hydrogen bond to
the epoxide oxygen, in addition to D101, further affecting
the binding of the substrate in this mutant. Docking of
substrate 1 was also performed to the WT enzyme: Two
reactive docking poses were observed with similar scores,
one favoring formation of (R,R)-2 and the other (S,S)-2 (as
shown in SI Fig. S4). This is consistent with the low stere-
oselectivity observed for reaction of substrate 1 with WT
LEH.
In order to shed light on the origin of reversed enantiose-
lectivity, substrate 1 was docked into the active site of the
SZ529 and SZ348 crystal structures. In order for the sub-
strate to undergo epoxide ring-opening by the active site
water molecule, activation must occur via a hydrogen
bond from the protonated D101 to the epoxide oxygen.
For each of the mutants, only a single reactive docking
pose was found in which this requirement is satisfied (as
shown in Fig. 5). As discussed previously,15-17, 20 the carbon
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
a)
b)
V80
V78
Y80
V114
F114
D101
D101
D132
2
V116
2
1
1
V32
D132
H2O
H2O
Y53
N55
N55
Y53
Figure 5. Highest ranked docking poses for substrate 1 in the crystal structures of the SZ529 (a) and SZ348 (b) mutants of LEH.
Nucleophilic attack at C1 and C2 of substrate 1 results in formation of the (R,R)-2 and (S,S)-2, respectively. Geometric parameters
for these complexes are displayed in SI Table S4.
Expanding the substrate scope of LEH. Some of the
best variants evolved for substrate 1 were tested as cata-
lysts in the hydrolytic desymmetrization of structurally
different epoxides 3, 5 and 7 without performing any addi-
tional mutagenesis experiments (Scheme 2). It can be
seen that in most, but not in all cases, high stereoselectiv-
ity was achieved (Table 1). The sterically smaller cyclopen-
tene oxide (3) is not accepted by any of the variants, an
observation that was made previously when testing LEH
variants that had been evolved for desymmetrization of
substrate 1.16
Scheme 2. Hydrolytic desymmetrization of further meso-
epoxides as substrates using the best variants evolved for
epoxide 1.
HO OH
O
+
+
O
(R)-9
(R)-10
LEH
H2O
HO OH
O
9
rac-
(S)-9
10
(S)-
Scheme 3. Hydrolytic kinetic resolution of styrene oxide
(rac-9) using the best variants evolved for epoxide 1.
The best variants were also tested for hydrolytic kinetic
resolution of styrene oxide (rac-9) (Scheme 3). All of them
induce reversed enantioselectivity compared to WT, vari-
ACS Paragon Plus Environment

Page 5 of 10
ACS Catalysis
E = 43 (Table 2).
ant SZ351 reaching maximum (S)-selectivity amounting to
1
2
3
4
Table 1. Hydrolytic desymmetrization of different epoxides catalyzed by the best LEH variants evolved for substrate 1.
5
1
3
5
7
6
7
8
9
Code
ee%
c%
>99
91
ee%
2 (R,R)
nd
c%
80
<5
5
ee%
c%
97
16
ee%
c%
>99
7
WT
4 (S,S)
17 (S,S)
81 (S,S)
98 (S,S)
93 (S,S)
68 (R,R)
86 (R,R)
95 (R,R)
nd
94 (R,R)
0.2 (R,R)
4 (S,S)
14 (S,S)
>99 (R,R)
>99 (R,R)
>99 (R,R)
nd
SZ347
SZ348
SZ351
SZ366
SZ369
SZ380
SZ386
SZ388
SZ389
SZ390
SZ391
SZ398
SZ529
96 (S,S)
99 (S,S)
97 (S,S)
93 (R,R)
95 (R,R)
96 (R,R)
94 (R,R)
96 (R,R)
97 (R,R)
95 (R,R)
95 (R,R)
96 (R,R)
97 (R,R)
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
97
10 (S,S)
10 (S,S)
nd
81
89
93
<5
28
73
98
5
63
8
97
<5
<5
<5
<5
<5
<5
<5
<5
<5
8
98
nd
26
74
<5
46
80
5
98
nd
91
nd
<5
49
68
<5
57
<5
92
>99
>99
96
nd
87 (R,R)
91 (R,R)
93 (R,R)
85 (R,R)
95 (R,R)
94 (R,R)
>99 (R,R)
>99 (R,R)
nd
10 (R,R)
nd
98
10 (R,R)
nd
58
36
83
97 (R,R)
nd
98
>99
10 (R,R)
>99 (R,R)
nd: not determined.
Table 2. Hydrolytic kinetic resolution of epoxide rac-9
with best LEH variants evolved for substrate 1.
The results of the present study contrast with those of
the initial report utilizing conventional ISM, in which 8
CAST residues in WT LEH were chosen for randomization
and grouped into four 2-residue sites.17 Unfortunately,
such a scenario poses the problem of determining which
of the 4! = 24 ISM pathways should be considered, since
exploring all of them requires excessive screening.21 Sever-
al pathways were arbitrarily chosen, and following the
screening of 5,000 transformants, which is twice as much
as in the present study, the best hits showed the following
improved but not fully satisfactory catalytic profiles.17 For
epoxide 1 : 90% ee (R,R); 97% ee (S,S); epoxide 5 : 77% ee
(R,R); 98% ee (S,S); epoxide 7 : 99% ee (R,R) versus 92%
ee of WT LEH.
Favored
Code
ee%
c%
E
enantiomer
WT
26
92
94
39
51
17
(R)
(S)
(S)
(S)
(S)
(S)
1.8
28
SZ348
SZ351
SZ386
SZ390
SZ391
15
23
36
46
20
43
2.8
4.6
2.9
44
The present approach also stands in contrast to the most
recent study in which a 10-residue site in LEH was ran-
domized using a single amino acid alphabet in what can
be called single code saturation mutagenesis (in that case
valine), which required more than three times the screen-
ing effort for 95% library coverage (>3072 transformants)
and provided only mediocre enantioselectivities (82-86%
ee for (S,S)-variants and 70-76% ee for (R,R)-variants) in
one and the same library.16 About 828 additional trans-
formants had to be screened to obtain variants that show
96% ee (R,R-2) and 92% ee (S,S-2). A comparison with the
previously published results16 based on single code satura-
tion mutagenesis is presented in Table 3.
CONCLUSIONS AND PERSPECTIVES
The purpose of this study was to increase the efficacy of
directed evolution of stereo- and regioselective enzymes.
Lessons have been learned in terms of methodology de-
velopment and in uncovering the reasons for enhanced
and reversed enantioselectivity of LEH mutants. We con-
clude that triple code saturation mutagenesis (TCSM) is a
simple and effective means to explore protein sequence
space at large randomization sites lining the binding
pockets of enzymes, provided guidance is sought from
crystal structures, consensus sequence data and mecha-
nistic considerations. The inherent knowledge-driven
strategy constitutes a practical compromise between
restricted structural diversity and screening effort.
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 10
Table 3. Comparison of screening effort of this work with
single code saturation mutagenesis used in previous study.
when water is the nucleophile, in contrast to other O- or
N-nucleophiles.25 The same applies to chiral organocata-
lysts26 The state of the art appears to be List’s chiral or-
ganocatalysts based on BINOL-derived chiral phosphoric
acids as catalysts using benzoate as the nucleophile at low
temperatures (-5 to -40ο C) and reaction times of several
days.26c For example, in the desymmetrization of epoxides
1, 5 and 7, enantioselectivities of 93%, 91% and 82% were
reported, as assessed by measuring the respective mono-
benzoates of the 1,2-diols. Base-mediated deprotection
then provides the respective 1,2-diols. Since EHs do not
deliver O-protected derivatives,27 which may actually be
desired in some circumstances, chiral organocatalysts and
transition metal catalysts can be considered to be com-
plementary in this type of asymmetric catalysis.
1
2
3
4
5
6
7
8
Library Residues
Library
Best
ee Reference
included
(code
size for obtained
in initial 95%
usage)
library
coverage
Valine
L74,F75,
3072
86% (S,S)
76% (R,R)
16
M78,I80,
L103,L114,
9
I116,F134,
F139,L147
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
ISM
A
828
576
384
92% (S,S)
16
96% (R,R)
99% (S,S)
89% (R,R)
97% (R,R)
I80,V83,
This work
This work
(V-F-Y) L114,I116
ISM
MATERIALS AND METHODS
Materials
KOD Hot Start DNA Polymerase was obtained from No-
vagen. Restriction enzyme Dpn I was bought from NEB.
The oligonucleotides were synthesized by Life Technolo-
gies. Plasmid preparation kit was ordered from Zymo
Research, and PCR gel extraction kit was bought from
QIAGEN. DNA sequencing was conducted by GATC Bio-
tech. All commercial chemicals were purchased from
Sigma-Aldrich, Tokyo Chemical Industry (TCI) or Alfa
Aesar. Lysozyme and DNase I were purchased from Ap-
pliChem.
Full library coverage is not mandatory in any SM experi-
ment, but it gives the researcher confidence that the best
variants have not been missed.1e,13 TCSM-designed librar-
ies are so small that they can be screened by automated
GC or HPLC, not only for notably reduced library cover-
age (e.g., 50%10a,b corresponding to the 53rd best variant
according to the Nov algorithm13,22), which would require
considerably less screening, but even when aiming for
essentially full coverage (95%). Thus, the bottleneck of
directed evolution of enzymes is no longer the screening
step. When an on-plate pretest is used for identifying
active variants, as demonstrated in this study, the actual
ee-screening effort is even less.
PCR based methods for library construction
Libraries were constructed using the Over-lap PCR and
megaprimer approach with KOD Hot Start polymerase.
50 µL reaction mixtures typically contained 30 µL water, 5
µL KOD hot start polymerase buffer (10×), 3 µL 25 mM
MgSO4, 5 µL 2 mM dNTPs, 2.5 µL DMSO, 0.5 µL (50~100
ng) template DNA, 100 µM primers mix 0.5 µL each and 1
µL KOD hot start polymerase. The PCR conditions for
short fragment: 95 °C 3 min, (95 °C 30 sec, 56 °C 30 sec,
68 °C 40 sec) × 32 cycles, 68 °C 120 sec, 16 °C 30 min. For
mega-PCR: 95 °C 3 min, (95 °C 30 sec, 60 °C 30 sec, 68 °C 5
min 30 sec) × 24 cycles, 68 °C 10 min, 16 °C 30 min. The
PCR products were analyzed on agarose gel by electro-
phoresis and purified using a Qiagen PCR gel extraction
kit. 2 µL NEB CutSmart™ Buffer and 2 µL Dpn I were add-
ed in 50 µL PCR reaction mixture and the digestion was
carried out at 37 °C for more than 3 h. After Dpn I diges-
tion, the PCR products (1 µL) were directly transformed
into electro-competent E. coliBL21(DE3) to create the final
library for Quick Quality Control28 and screening.
Relative to other methods such as epPCR or DNA shuf-
fling, CAST-based SM defines pre-determined boundaries
in the vast protein sequence space ensuring high-quality
focused libraries. Among the various SM strategies pres-
ently available,1a,e,10,11,16 structure-driven TCSM appears to
be a logical choice. The number of recursive cycles of
mutagenesis/expression/screening is limited. We recom-
mend that step-economy, a term borrowed from synthetic
organic chemistry,23 should be considered in future di-
rected evolution studies more so than in the past. In fu-
ture TCSM-based work, the choice of the triple code can
also be guided by in silico design and screening.
Crystal structure analyses of stereoselective variants de-
rived from directed evolution are rarely performed,1,16,24
yet they provide a unique opportunity to analyze on a
more sound basis the effect of the mutations on the cata-
lytic profile. The crystal structures reported in this study,
combined with docking computations, shed light on the
origin of enhanced and reversed stereoselectivity and on
the intricacies of the catalytic machinery of LEH.
Primer design and Library creation
Finally, we comment on the hydrolytic desymmetrization
of meso-configurated epoxides as a synthetic method for
accessing enantiopure vicinal diols in organic chemistry.
This asymmetric transformation remains a challenge
using chiral transition metal complexes such as the Jacob-
sen chiral cobalt-catalysts, because these do not work well
Primer design and library construction depend upon the
particular amino acid chosen, and in the case of LEH this
involves ten residues which were divided into three
groups (SI Fig. S5): 1) Amplification of the short fragments
of LEH using mixed primers F1/R1, F2/R2 and F3/R3 for
Library A, B and C, respectively; 2) Amplification of the
ACS Paragon Plus Environment

Page 7 of 10
ACS Catalysis
31,000 × g for 40 min. The soluble protein sample was
whole plasmid pET22bLEHwt16 using the products of step
1
2
3
4
1 as megaprimers, leading to the final variety plasmids for
library generation. Primers are listed in SI Table S5. The
PCR products were digested by Dpn I and transformed
into electro-competent E. coli BL21(DE3) to create the
library for screening.
loaded onto a nickel affinity column (GE Healthcare) and
washed with 10~500 mM imidazole solution containing
300 mM NaCl and 25 mM Tris-HCl (pH 8.0). Proteins
from the flow through were pooled and concentrated, and
then loaded onto a Superdex 75 Hiload 16/60 column (GE
Healthcare) and eluted with 25 mM Tris-HCl (pH 7.5), 150
mM NaCl and 1 mM DTT. Fractions containing highly
pure LEH proteins were pooled and concentrated with
Centricon filtration devices (Amicon). The protein con-
centrations were determined by the Bradford procedure.
5
6
7
8
9
Libraries A-B, A-C, A-B-C1, A-B-C2 and A-C-B were creat-
ed using the same procedure as mentioned above. All the
primers used are listed in SI Table S6.
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
Screening Procedures
Measurement of kinetic parameters
Colonies were picked and transferred into deep-well
plates containing 300 µL LB medium with 50 µg/ml car-
benicillin and cultured overnight at 37 °C with shaking.
An aliquot of 120 µL was transferred to glycerol stock
plate and stored at -80 °C. Then, 800 µL TB medium with
0.5% (m/v) lactose and 50 µg/mL carbenicillin was added
directly to the culture plate for 8 h at 28 °C with shaking
for protein expression. The cell pellets were harvested and
washed with 400 µL 50 mM pH 7.4 potassium phosphate
buffer and centrifuged for 10 min 4000 rpm at 4 °C. Then,
the pellets were resuspended in 400 µL of the same buffer
with 6 U DNase I and 1 mg/mL lysozyme for breaking the
cell at 30 °C for 1 h with shaking. The crude lysate was
centrifuged for 30 min 4000 rpm at 4 °C. 40 µL of the
supernatant was used for an adrenaline assay,19 then 110
µL potassium phosphate buffer (50 mM, pH 7.4) with 5%
acetonitrile and substrate 1 (final concentration 10 mM)
were added. The plates were incubated at 30 °C, 500 rpm,
3~4 h. Afterwards, NaIO4 solution (20 µL, 77 mg in 24 mL
of water) was added and the plates were further incubat-
ed at 30 °C, 500 rpm, 10 min. Subsequently, adrenaline
solution 20 µL (epinephrine 132 mg, water 24 mL, concen-
trated HCl (5 drops) for solubilizing the adrenaline) was
added, which caused the immediate formation of a red
color in the inactive reactions. Active variants gave color-
less wells. 300 µL rest supernatant of the active trans-
formants was transferred into new deep-well plates for
reaction with 5 mM substrate 1 and 5% acetonitrile as co-
solvent for 14~16 h at 30 °C 800 rpm, the final volume was
400 µL. The product and remaining substrate were ex-
tracted using equal volumes of ethyl acetate (EtOAc) for
GC analysis by chiral column (SI Table S7). The screening
results are shown in SI Table S2.
The enzymatic hydrolysis rate was measured by moni-
toring the conversion of substrate by GC using 1-heptanol
as the internal standard. Pure enzymes were added to
potassium phosphate buffer (50 mM, pH 7.4) with a total
volume of 400 μL containing epoxide of varied concen-
tration (2-64 mM) and 2.5% (v/v) of acetonitrile, and the
reaction was performed at 30°C for 10 min with shaking
(800 rpm). The reaction was terminated by the addition
of 400 μL EtOAc containing 2.0 mM 1-heptanol. The
mixture was vigorously mixed and the organic layer was
separated by centrifugation at 4,000 rpm for 15 min. The
EtOAc layer was analyzed by GC, the results are shown in
Table S3.
X-ray structural analysis
The SZ529 crystals were grown in 1.4 M sodium citrate
and 0.1 M Hepes (pH 7.0) by the sitting-drop vapor diffu-
sion method at 4 °C. The SZ348 variant was crystallized in
2.4 M sodium/potassium phosphate, 0.1 M Tris-HCl
(pH8.5) by the sitting-drop vapor diffusion method at 18
°C. Proteins (SZ529 at 21 mg/mL while SZ348 at 16
mg/mL) were mixed in a 1:1 ratio with the reservoir solu-
tion in a final volume of 4 μL and equilibrated against the
reservoir solution. Single crystals of SZ529 were obtained
directly, whereas single crystals of SZ348 were achieved
by adding additive such as 3% xylitol or 3% D-Sorbitol. All
crystals were mounted in nylon loops and flash-frozen in
liquid nitrogen. The cryoprotectant for SZ529 crystals
contained 1.6 M sodium citrate, 0.1 M Hepes (pH7.0) and
30% sucrose while the cryoprotectant for SZ348 crystals
consisted of 2.4 M sodium/potassium, 0.1 M Tris-HCl
(pH8.5). Diffraction data of SZ529 were collected at the
wavelength of 1.5418 Å on a Raxis IV++ imaging plate de-
tector at 100 K. Diffraction data of SZ348 was collected at
the wave length of 0.97853 Å using a Pilatus detector at
beamline BL19U of Shanghai Synchrotron Radiation Facil-
ity (SSRF) at 100 K. All data sets were indexed, integrated,
and scaled using the HKL2000 package.29 The structures
of SZ529 and SZ348 were solved by molecular replace-
ment method using the program PHASER30 and the coor-
dinate of wild-type LEH (PDB code 1NU3) as a search
model. Rounds of automated refinement were performed
with PHENIX31 and the models were extended and rebuilt
manually with COOT.32 The structures of SZ529 and
SZ348 were refined to 2.25 and 3.0 Å, respectively. The
statistics for data collection and crystallographic refine-
Protein expression and purification
LEH and mutants SZ348 and SZ529 were inoculated
into 250 mL shaking flasks of 40 mL LB containing 50
µg/mL ampicillin and cultured overnight at 37 °C with
shaking, and then scaled up to 4 flasks of 500 mL TB con-
taining 50 µg/mL ampicillin. When the OD600 reached to
0.8, 0.4 mM IPTG was added to induce the protein ex-
pression. The cell cultures continued to grow overnight at
16 °C before being harvested by centrifugation at 6,000 × g
and resuspended in a Tris lysis buffer (25 mM, pH 8.0)
containing 300 mM NaCl, 10 mM imidazole and 5 mM β-
mercaptoethanol. The cell pellets were disrupted by soni-
cator and the cell debris was removed by centrifugation at
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 10
(3) Gupta, R. D.; Goldsmith, M.; Ashani, Y.; Simo, Y.; Mullo-
ment are summarized in Table S8. All structural figures
were prepared using Pymol (http://www.pymol.org/)
except for Figure S3 which was drawn by Discovery Studio
Visualizer.33
1
2
3
4
5
6
7
8
kandov, G.; Bar, H.; Ben-David, M.; Leader, H.; Margalit, R.;
Silman, I.; Sussman, J. L.; Tawfik, D. S. Nat. Chem. Biol. 2011, 7,
120-125.
(4) Kwan, D. H.; Constantinescu, I.; Chapanian, R.; Higgins, M.
A.; Kötzler, M. P.; Samain, E.; Boraston, A. B.; Kizhakkedathu, J.
N.; Withers, S. G. J. Am. Chem. Soc. 2015, 137, 5695-5705.
(5) Taylor, S. V.; Kast, P.; Hilvert, D. Angew. Chem. Int. Ed. .
2001, 40, 3310-3335.
(6) Yang, G.; Withers, S. G. Chembiochem. 2009, 10, 2704-2715.
(7) Acevedo-Rocha, C. G.; Agudo, R.; Reetz, M. T. J. Biotechnol.
2014, 191, 3-10.
(8) Reymond, J. L. Enzyme Assays: High-throughput Screening,
Genetic Selection and Fingerprinting; Wiley-VCH, Weinheim ,
2006.
(9) Lutz, S.; Patrick, W. M. Curr. Opin. Biotechnol. 2004, 15, 291-
297.
(10) (a) Reetz, M. T.; Carballeira, J. D. Nat. Protoc. 2007, 2, 891-
903; (b) Reetz, M. T.; Kahakeaw, D.; Lohmer, R. Chembiochem.
2008, 9, 1797-1804 (c) Reetz, M.T.; Wu, S. Chem. Commun. 2008,
43, 5499-5501.
(11) (a) Evans, B. S.; Chen, Y.; Metcalf, W. W.; Zhao, H.; Kelleher,
N. L. Chem. Biol. 2011, 18, 601-6o7; (b) Chen, M. M. Y.; Snow, C.
D.; Vizcarra, C. L.; Mayo, S. L.; Arnold, F. H. Prot. Eng. Des. Sel.
2012, 25, 171-178; (c) Sandström, A. G.; Wikmark, Y.; Engström,
K.; Nyhlén, J.; Bäckvall, J.-E. Proc. Natl. Acad. Sci. USA. 2012, 109,
78-83; (d) Swe, P. M.; Copp, J. N.; Green, L. K.; Guise, C. P.;
Mowday, A. M.; Smaill, J. B.; Patterson, A. V.; Ackerley, D. F.
Biochem. Pharmacol. 2012, 84, 775-783; (e) Dudek, H. M.; Fink,
M. J.; Shivange, A. V.; Dennig, A.; Mihovilovic, M. D.;
Schwaneberg, U.; Fraaije, M. W. Appl. Microbiol. Biotechnol.
2014, 98, 4009-4020; (f) Wikmark, Y.; Svedendahl Humble, M.;
Bäckvall, J.-E. Angew. Chem. Int. Ed. 2015, 54, 4284-4288; (g)
Zhang, L.; Lu, L.; Fan, S.; Jin, L.; Gu, G.; Xu, L.; Xiao, M. RSC Adv.
2015, 5, 22361-22364; (h) Chuang, H.-Y.; Suen, C.-S.; Hwang, M.-
J.; Roffler, S. R. Prot. Eng. Des. Sel. 2015, 28, 519-530.
(12) (a) Patrick, W. M.; Firth, A. E. Biomol. Eng. 2005, 22, 105-112;
(b) Denault, M.; Pelletier, J. N. In Protein Engineering Protocols;
Arndt, K. M.; Müller, K. M., Eds.; Humana Press, NJ, 2007, Vol.
352, pp 127-154.
Docking calculations
The crystal structure of WT LEH was used as the starting
structure for the docking of substrate 1 to the WT (PDB
entry 1NU3).15 The apo crystal structures of the SZ529 and
SZ348 mutants were used for docking to the respective
mutants. The protein structures were prepared for dock-
ing using the Protein Preparation Wizard in Maestro.34
The positions of all hydrogen atoms were energy mini-
mized using Prime. Preparation of substrate 1 for docking
was performed using LigPrep.35 Docking was performed
using Glide,36 with a rigid protein and standard precision
(SP) settings. The docking calculations were restricted
such that a hydrogen bond must be formed between the
substrate and the protonated D101. A maximum of 10
docking poses were allowed for each model, but no more
than 8 were found for a given mutant.
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
ASSOCIATED CONTENT
Supporting Information: including figures and tables for
multi-sequence alignment, library design and creation, pri-
mers list, X-ray structure, docking analysis, kinetic parame-
ters measurement and GC analytical conditions. This materi-
al is available free of charge via the Internet at
http://pubs.acs.org.”
AUTHOR INFORMATION
Corresponding Author
* Correspondence should be addressed to (M.T.R.)
reetz@mpi-muelheim.mpg.de or (J.Z.) jiahai@mail.sioc.ac.cn.
(13) Nov, Y. Appl. Environ. Microbiol. 2012, 78, 258-262.
(14) van der Werf, M.J.; Overkamp, K. M.; de Bont, J. A. M. J.
Bacteriol. 1998, 180, 5052-5057.
(15) Arand, M.; Hallberg, B. M.; Zou, J.; Bergfors, T.; Oesch, F.;
van der Werf, M. J.; de Bont, J. A. M.; Jones, T. A.; Mowbray, S. L.
EMBO. J. 2003, 22, 2583-2592.
(16) Sun, Z.; Lonsdale, R.; Kong, X.-D.; Xu, J.-H.; Zhou, J.; Reetz,
M. T. Angew. Chem. Int. Ed. 2015, 54, 12410-12415.
(17) Zheng, H.; Reetz, M. T. J. Am. Chem. Soc. 2010, 132, 15744-
15751.
(18) Examples of the consensus approach to protein engineering:
(a) Steipe, B.; Schiller, B.; Plückthun, A.; Steinbacher, S. J.
Mol.Biol. 1994, 240, 188-192; (b) Vázquez-Figueroa, E.; Chaparro-
Riggers, J.; Bommarius, A. S. ChemBioChem. 2007, 8, 2295-2301.
(19) Wahler, D.; Reymond, J. L. Angew. Chem. Int. Ed. 2002, 41,
1229-1232.
(20) Theoretical studies of LEH: (a) Hou, Q. Q.; Sheng, X.;
Wang, J. H.; Liu, Y. J.; Liu, C. B. Biochim. Biophys. Acta Proteins
Proteomics 2012, 1824, 263-268; (b) Lind, M. E. S.; Himo, F. An-
gew. Chem. Int. Ed. 2013, 52, 4563-4567.
(21) Gumulya, Y.; Sanchis, J.; Reetz, M. T. ChemBioChem 2012, 13,
1060-1066.
(22) Hoebenreich, S.; Zilly, F. E.; Acevedo-Rocha, C. G.; Zilly,
M.; Reetz, M. T. ACS Synth. Biol. 2015, 4, 317-331.
(23) Wender, P. A.; Verma, V. A.; Paxton, T. J.; Pillow, T. H. Acc.
Chem. Res. 2008, 41, 40-49.
ACKNOWLEDGMENT
Support from the Max-Planck-Society and the LOEWE Re-
search cluster SynChemBio is gratefully acknowledged in
addition to grants from Science and Technology Commission
of Shanghai Municipality (15JC1400403).
REFERENCES
(1) Recent reviews of directed evolution: (a) Bommarius, A. S.
Annu. Rev. Chem. Biomol. Eng. 2015, 6, 319-345; (b) Currin, A.;
Swainston, N.; Day, P. J.; Kell, D. B. Chem. Soc. Rev. 2015, 44,
1172–239; (c) Denard, C. A.; Ren, H.; Zhao, H. Curr. Opin. Chem.
Biol. 2015, 25, 55-64; (d) Gillam, E. M. J.; Copp, J. N.; Ackerley, D.
F. Methods in Molecular Biology, Humana Press, Totowa, 2014,
Vol. 1179 ; (e) Goldsmith, M.; Tawfik, D. S. Methods. Enzymol.
2013, 523, 257-283; (f) Brustad, E. M.; Arnold, F. H. Curr. Opin.
Chem. Biol. 2011, 15, 201–10; (g) Reetz, M. T. Angew. Chem. Int.
Ed. 2011, 50, 138–74; (h) Jäckel, C.; Hilvert, D. Curr. Opin. Bio-
technol. 2010, 21, 753–9; (i) Turner, N. J. Nat. Chem. Biol. 2009, 5,
567–73; (j) Lutz, S.; Bornscheuer, U. T. Protein Engineering
Handbook, Wiley-VCH, Weinheim, 2009.
(2) Hill, C. M.; Li, W. S.; Thoden, J. B.; Holden, H. M.; Raushel,
F. M. J. Am. Chem. Soc. 2003, 125, 8990-8991.
ACS Paragon Plus Environment

Page 9 of 10
ACS Catalysis
(24) Recent examples of X-ray structures of stereoselective
(28) (a) Bougioukou, D. J.; Kille, S.; Taglieber, A.; Reetz, M. T.
1
2
3
4
5
6
7
8
variants generated by directed evolution: (a) Pompeu, Y. A.;
Sullivan, B.; Stewart, J. D. ACS Catal. 2013, 3, 2376-2390; (b)
Reetz, M. T.; Bocola, M.; Wang, L.-W.; Sanchis, J.; Cronin, A.;
Arand, M.; Zou, J.; Archelas, A.; Bottalla, A.-L.; Naworyta, A.;
Mowbray, S. L. J. Am. Chem. Soc. 2009, 131, 7334-7343.
Adv. Synth. Catal. 2009, 351, 3287-3305; (b) Acevedo-Rocha, C.
G.; Kille, S.; Reetz, M. T. In Methods in Molecular Biology,
Gillam, E. M. J.; Copp, J. N.; Ackerley, D. F. Eds.; Humana Press,
Totowa, 2014, Vol, 1179, pp. 103-128.
(29) Otwinowski, Z.; Minor, W. Methods in Enzymology, Carter
Jr, C. W.; Sweet, R. M. Eds.; Academic Press, New York, 1997,
Vol. 276, pp 307-326.
(30) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn,
M. D.; Storoni, L. C.; Read, R. J. J. Appl. Crystallogr. 2007, 40,
658-674.
(31) Adams, P. D.; Grosse-Kunstleve, R. W.; Hung, L.-W.;
Ioerger, T. R.; McCoy, A. J.; Moriarty, N. W.; Read, R. J.; Sacchet-
tini, J. C.; Sauter, N. K.; Terwilliger, T. C. Acta Crystallogr. D.
2002, 58, 1948-1954.
(25) (a) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421–431; (b)
Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.; Han-
sen, K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am.
Chem. Soc. 2002, 124, 1307–1315; (c) Jacobsen, E. N.; Kakiuchi, F.;
Konsler, R. G.; Larrow, J. F.; Tokunaga, M. Tetrahedron Lett.
1997, 38, 773–776; (d) Kawthekar, R. B.; Kim, G.-J. Helv. Chim.
Acta 2008, 91, 317– 332; (e) Ready, J. M.; Jacobsen, E. N. J. Am.
Chem. Soc. 2001, 123, 2687–2688; (f) Ready, J. M.; Jacobsen, E. N.
Angew. Chem. 2002, 114, 1432–1435. Angew. Chem., Int. Ed. 2002,
41, 1374-1377; (g) Private communication of Jacobsen, E. N. (Har-
vard University, Cambridge/MA) regarding the performance of
Jacobsen-catalysts in the hydrolytic desymmetrization of meso-
epoxides (2010).
(26) Reviews and key papers of organocatalyzed desymmetriza-
tion of meso-epoxides: (a) Wang, P. A.; Beilstein. J. Org. Chem.
2013, 9, 1677-1695; (b) Wang, Z.; Law, W. K.; Sun, J. Org. Lett.
2013, 15, 5964-5966; (c) Monaco, M. R.; Prévost, S.; List, B. An-
gew. Chem. Int. Ed. 2014, 53, 8142-8145.
(27) Recent reviews of EHs: (a) Wu, S.; Li, A.; Chin, Y. S.; Li, Z.
ACS Catal. 2013, 3, 752-759. (b) Schober, M.; Faber, K.; Trends
Biotechnol. 2013, 31, 468-478. (c) Kotik, M.; Archelas, A.;
Wohlgemuth, R. Curr. Org. Chem. 2012, 16, 451-482; (d)
Widersten, M.; Gurell, A.; Lindberg, D. Biochim. Biophys. 2010,
1800, 316-326;
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
(32) Emsley, P.; Cowtan, K. Acta Crystallogr. D. 2004, 60, 2126-
2132.
(33) Discovery Studio Modeling Environment, v. 4.0 (Accelrys
Software Inc., San Diego, 2007).
(34) (a) Protein Preparation Wizard, Schrödinger Suite v. 2014-4
(Schrödinger, LLC, New York, 2014); (b) Epik v. 3.0 (Schrödinger,
LLC, New York, 2014); (c) Impact v. 6.5 (Schrödinger, LLC, New
York, 2014); (d) Prime v. 3.8 (Schrödinger, LLC, New York, 2014).
(35) LigPrep, v. 3.2 (Schrödinger, LLC, New York, 2014).
(36) (a) Glide, v. 6.5 (Schrödinger, LLC, New York, 2014); (b)
Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.;
Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T.
J. Med. Chem. 2006, 49, 6177–6196; (c) Friesner, R. A.; Banks, J.
L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Re-
pasky, M. P.; Knoll, E. H.; Shaw, D. E.; Shelley, M.; Perry, J. K.;
Francis, P.; Shenkin, P. S. J. Med. Chem. 2004, 47, 1739-1749.
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 10
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
ACS Paragon Plus Environment