An unusual (R)-selective epoxide hydrolase with high activity for facile preparation of enantiopure glycidyl ethers

Article abstract of DOI:10.1002/adsc.201100031

A novel epoxide hydrolase (BMEH) with unusual (R)-enantioselectivity and very high activity was cloned from Bacillus megaterium ECU1001. Highest enantioselectivities (E>200) were achieved in the bioresolution of ortho-substituted phenyl glycidyl ethers and para-nitrostyrene oxide. Worthy of note is that the substrate structure remarkably affected the enantioselectivities of the enzyme, as a reversed (S)-enantiopreference was unexpectedly observed for the ortho-nitrophenyl glycidyl ether. As a proof-of-concept, five enantiopure epoxides (>99% ee) were obtained in high yields, and a gram-scale preparation of (S)-ortho-methylphenyl glycidyl ether was then successfully performed within a few hours, indicating that BMEH is an attractive biocatalyst for the efficient preparation of optically active epoxides. Copyright

Full text of DOI:10.1002/adsc.201100031

FULL PAPERS
DOI: 10.1002/adsc.201100031
An Unusual (R)-Selective Epoxide Hydrolase with High Activity
for Facile Preparation of Enantiopure Glycidyl Ethers
Jing Zhao,a Yan-Yan Chu,b Ai-Tao Li,a Xin Ju,a Xu-Dong Kong,a Jiang Pan,a
Yun Tang,b,* and Jian-He Xua,*
a
Laboratory of Biocatalysis and Synthetic Biotechnology, State Key Laboratory of Bioreactor Engineering, East China
University of Science and Technology, 130 Meilong Road, Shanghai 200237, Peopleꢀs Republic of China
Fax : (+86)-21-6425-0840; e-mail: jianhexu@ecust.edu.cn
Department of Pharmaceutical Sciences, School of Pharmacy, East China University of Science and Technology, 130
b
Meilong Road, Shanghai 200237, Peopleꢀs Republic of China
Fax : (+86)-21-6425-3651; e-mail: ytang234@ecust.edu.cn
Received: January 14, 2011; Revised: March 21, 2011; Published online: June 16, 2011
Abstract: A novel epoxide hydrolase (BMEH) with proof-of-concept, five enantiopure epoxides (>99%
unusual (R)-enantioselectivity and very high activity ee) were obtained in high yields, and a gram-scale
was cloned from Bacillus megaterium ECU1001. preparation of (S)-ortho-methylphenyl glycidyl ether
Highest enantioselectivities (E>200) were achieved was then successfully performed within a few hours,
in the bioresolution of ortho-substituted phenyl gly- indicating that BMEH is an attractive biocatalyst for
cidyl ethers and para-nitrostyrene oxide. Worthy of the efficient preparation of optically active epoxides.
note is that the substrate structure remarkably affect-
ed the enantioselectivities of the enzyme, as a re- Keywords: enantioselectivity; epoxide hydrolase;
versed (S)-enantiopreference was unexpectedly ob- glycidyl ethers; kinetic resolution; molecular model-
served for the ortho-nitrophenyl glycidyl ether. As a ing
Introduction
cillus subtilis EH (Bsueh)[8] with a specific activity of
merely 0.01 mmol minÀ1 mgÀ1, which could not meet
The enantioselective hydrolysis of racemic epoxides the requirement for practical application. In our labo-
catalyzed by epoxide hydrolases (EHs) is one of the ratory, a strain of B. megaterium ECU1001, which is
promising approaches to obtain enantiopure aryl gly- highly selective towards (R)-PGE, was successfully
cidyl ethers,[1] which are important building blocks for isolated from soil samples.[9] It was the distinguished
the production of bioactive compounds such as chiral property of EHs in B. megaterium ECU1001 that at-
amino alcohols[2] and b-blockers.[3] However, only few tracted our interest in cloning them for further inves-
preparative-scale examples using EHs have been re- tigation and application.
ported due to the quite low activities of the wild-type
whole cells.[4] Thus the task of searching for novel bio- tioselectivity and high activity toward (R)-glycidyl
catalysts still remains a challenge. ethers, which has the potential to be used as an indus-
Herein, we report a novel EH with unusual enan-
EHs are ubiquitously found in nature and have trial biocatalyst for the production of enantiopure ep-
been identified in many organisms including mam- oxides and diols.
mals, plants, insects and various micro-organisms.[5]
Some of them exhibit modest enantioseletivities
toward phenyl glycidyl ether (PGE), e.g., Aspergillus Results and Discussion
niger EH[4c] and Agrobacterium radiobacter EH.[6]
Thus protein engineering has been further applied to The whole genome sequence of B. megaterium QM
enhance the enantioselectivity of existing EHs.[7] B1551 was recently released in the GenBank database
However, most of them are (S)-selective, and only a (entry number CP001983.1). However, no putative
few could preferentially hydrolyze (R)-PGE, retaining EH gene was available to date. Therefore, we predict-
the useful (S)-epoxide for the synthesis of b-block- ed the EH gene based on the conserved regions
ers.[8] Unfortunately, the activities of the current (R)- shared in the a/b-hydrolase fold EHs. As a result, an
selective EHs are generally low, e.g., recombinant Ba- open reading frame designated as bmeh was identified
1510
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1510 – 1518

An Unusual (R)-Selective Epoxide Hydrolase with High Activity
Figure 1. Sequence alignment of EHs. The protein accession numbers are: B. megaterium ECU1001 (BMEH, this paper); B.
subtilis (yfhM, O31581); Glycine max (GmEH, Q39856); Solanum tuberosum (StEH, Q41413); human sEH (EPHX2,
P34913). Dark and light colours indicate sequence identity and similarity, respectively. The secondary structures are marked
under the sequences. Regions of putative motif are boxed, and the functionally essential residues are marked by stars. The
dashed line boxes represent the NC- and cap-loops, respectively.
as a putative EH. As shown in Figure 1, BMEH con- substrate 1 was submitted to enzymatic hydrolysis. As
tains all the conserved motifs and shares similar sec- expected, the BMEH is highly (R)-selective (E=58)
ondary structures with other known EHs. This gene and exhibited a pretty high activity (Table 1), indicat-
was then successfully cloned from the strain B. mega- ing a good potential for industrial applications.
terium ECU1001, and the sequencing result indicates
To the best of our knowledge, the enantioselectivity
that it consists of 864 bp, encoding 287 amino acids is the highest among all the known native EHs for the
with a molecular weight of 33,580 Da (GenBank ac- bioresolution of rac-1 (Table 1). Hence, this encourag-
cession number HQ436037). Noteworthy, a BLAST ing result prompted us to expand the substrate scope
search against the Protein Data Bank reveals that for further understanding of the interaction between
human sEH shares the closest structural homology the enzyme and substrate. A range of racemic phenyl
with a sequence identity of only 24%. The gene was glycidyl ethers, substituted with electron-donating and
overexpressed in Escherichia coli followed by a series electron-withdrawing groups, were employed for bio-
of optimizations to obtain soluble expression. The hydrolysis using BMEH (Scheme 1, substrates 2–
BMEH was then purified to an apparent homogeneity 10).[12]
with a specific activity of 83 U/mg protein by His-tag
affinity chromatography (Figure 2).
As shown in Table 2, the position and electronic
properties of the substituents were of great influence
To investigate whether the BMEH had the same on the activity and enantioselectivity. The resolution
enantiopreference as the wild-type strain, the model of epoxides 1–8 proceeded with varying degrees of
Adv. Synth. Catal. 2011, 353, 1510 – 1518
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1511

Jing Zhao et al.
FULL PAPERS
Scheme 1. Epoxide substrates used in this study.
the enantiopreference was opposite. These results
were also confirmed in the following preparative-
scale experiments. To the best of our knowledge, such
an interesting phenomenon has been observed for the
first time. In contrast, the EH from B. alcalophilus[4a]
Figure 2. Purification of BMEH. Lane 1: the protein size
standard; Lane 2: soluble fraction; Lane 3: insoluble frac-
tion; Lane 4: the purified BMEH.
enantioselectivity (E=4 to >200), and a general exhibited no inversed enantioselectivity for rac-5, and
trend can be found that the E value increases as the no activity towards rac-6 even after 24 h reaction
substituent on the phenyl ring is shifted from the time.
para- to the ortho-position. We also noticed that a
To determine which epoxide carbon was preferen-
longer (9) or a larger side-chain (10) could decrease tially attacked for substrates 2–5, the regioselectivity
the E value to some extent, which is in agreement coefficients aS and aR were calculated using Eq.
with reported results.[4a] Interestingly, the substitution (1).[13] aS and aR are regioselectivity coefficients relat-
pattern of the nitro group remarkably affected not ed to the attack at the more substituted carbon atom
only the level of selectivity but also the configuration (C-2) of the S and R enantiomers, respectively. Nor-
of the unreacted epoxide. As in the case of rac-5, the mally, EH attack occurs preferentially at the less sub-
remained enantiomer was (R)-configuration which stituted carbon atom (C-1), as demonstrated in the
was opposite to all the other tested glycidyl ethers. cases of compounds 2–4 (Table 3). However, for com-
Surprisingly, the unexpected excellent E values (> pound 5, it was noticed with surprise that the oxirane
200) were observed in the cases of 5 and 6, although ring was preferentially attacked at the C-2 for both
Table 1. Comparison of PGE resolution between BMEH and other EHs.
Epoxide hydrolase Catalyst
Catalyst concentra- PGE concentra-
Time
[min]
ee [%]/config- Yield
E
Ref.
source
form
tion [gLÀ1]
tion [mM]
uration
[%]
[4c]
[6]
Aspergillus niger
Agrobacterium ra- purified
diobacter enzyme
Trichosporon lou- wet cells
bierii
Rhodobacterales
bacterium
Bacillus alcalophi- growing
lus cells
Bacillus megateri- crude
wet cells
75
20
1
240
n.a.
100/(R)
>99/(R)
26
28
n.a.[a]
12
0.025-0.125
[10]
[11]
[4a]
50
67
270
20
>99/(R)
100/(R)
>99/(S)
>99/(S)
35
20
purified
enzyme
0.02
n.a.
1.28
29.2
6.6
20
38.4
27
38.4
27
1440
2
44
58
this
um
enzyme
study
[a]
n.a.: not available.
1512
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1510 – 1518

An Unusual (R)-Selective Epoxide Hydrolase with High Activity
Table 2. Results for enzymatic resolution of epoxides 1–16 on an analytical scale.
Substrate Concentration [mM] Enzyme used [U][a] Time [min] eeS [%][b]/configuration Conversion [%][c] E value[d]
1
2
3
4
5
6
7
8
20
10
10
10
10
10
10
10
10
10
5
5
5
5
20
10
8
9
9
9
18
18
108
18
18
108
400
400
4
2
>99/(S)
>99/(S)
77/(S)
56
50
50
52
50
50
54
52
69
58
51
57
51
54
67
70
58
>200
19
30
30
50
150
30
120
30
120
120
5
73/(S)
11
>99/(R)
>99/(S)[e]
67/(S)
>200
>200
7
25
4
25
>200
4
3
2
87/(S)[e]
77/(S)
9
10
11
12
13
14
15
16
98/(S)
>99/(R)
53/(R)
30
2
10
5
39/n.d.[f]
400
8
400
31/
20/
60/
U
1.4
3
8
[a]
[b]
[c]
[d]
[e]
[f]
Lyophilized enzyme powders or cell-free extracts were used for substrates 1–10 and 11–16, respectively.
Determined by chiral HPLC or GC as described in the Experimental Section.
Conversion was calculated from the remaining substrate concentration and the initial substrate concentration.
E values were calculated by the following equation: E=ln
The absolute configurations were confirmed by comparing the optical rotations reported in the literature.
n.d.: not determined.
[(1Àc)
N
U
ACHTUNGRTEN(NUNG 1+ees)].
of the enzyme. A similar result has been described by
Furstoss et al.[14] for Beauveria sulfurescens EH.
During the biohydrolysis of para-substituted styrene
enantiomers, resulting in configuration inversion of oxide derivatives, they found that a p-NO2 substituent
the corresponding diol. This is consistent with the fact in the substrates caused a switch of EH enantioprefer-
that at total conversion, the formed diol exhibited an ence.
ee of about 10%. This phenomenon might be due to
Homology modeling and molecular docking were
the change of electronic properties caused by ortho- performed to gain insights into the binding mode of
nitro substitution.[13] Interestingly, it was observed the substrates and the origin of inverted enantioselec-
that ortho-nitro substitution on the phenyl ring affects tivity.[15] As a target template structure, a human solu-
the enantiopreference as well as the regioselectivity ble epoxide hydrolase with an inhibitor bound (PDB
entry code: 3I28), showing 24% sequence identity and
52% similarity to the BMEH, was selected after a
PSI-BLAST search. Systematic analysis of the se-
quence and structure of EHs shows that although the
Table 3. Regioselectivity of BMEH for biohydrolysis of rac-
emic epoxides 2–5.
[a]
[a]
Substrate aS
aR
[%]
Absolute configuration[b]
sequence identity of EHs is rather low, the cap region
and the conserved a/b-hydrolase fold of EHs could be
modeled reliably.[16] Two loops called NC-loop (resi-
dues 121–127) and cap-loop (residues 358–369)
(Figure 1), whose length determines the feasibility of
modeling, should be taken into consideration.[16] The
cap-loop of BMEH is much shorter than that of
human EH, which was refined sufficently. Procheck
and Verify-3D were used to evaluate the modeled
structure. The results indicate that the BMEH model
is reliable (see Experimental Section).
The binding modes of the substrates in the active
site from molecular docking are shown in Figure 3,
and the related parameters including energies, distan-
ces and angles are summarized in Table 4. As shown
in Figure 3, all the substrates form hydrogen bonds
[%]
residual epox-
ide
diol
formed
2
3
4
5
2.1
6.2
0
2.9
0
0
S
S
S
R
R
R
R
R
95
85
[a]
aS and aR were calculated from three sets of data (con-
version, ees, eep) at seven different time points during the
biohydrolysis of racemic epoxides using the non-linear
curve fitting function of Origin software according to Eq.
(1).
The absolute configurations were established by chiral
HPLC analyses or the optical rotations, for details see
Experimental Section.
[b]
Adv. Synth. Catal. 2011, 353, 1510 – 1518
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1513

Jing Zhao et al.
FULL PAPERS
Figure 3. Docking of compounds 2–5 (A–D) to the binding pocket of BMEH, respectively. The BMEH protein was shown in
white cartoon, and the residues of the active pocket were shown by sticks in light grey carbon. The dark and light grey car-
bons shown in ball and stick represent the (R) and (S) enantiomers, respectively. And hydrogen bonds were labeled by
dashed lines, (R) in black and (S) in grey.
with Y144 and Y203 by the epoxide oxygen. The cata- sis can be related to NACs as postulated by Bruice[15a]
lytic efficacy of the enzyme is dependent on how in a molecular dynamics (MD) study of another EH.
often the nucleophile and electrophile are present in We noticed that, for both enantiomers of compound
near attack conformations (NACs).[15a] As Table 4 5, the C-2 position is much more favorable to be at-
shows, the favored enantiomers have relatively lower tacked than the C-1 position considering both the dis-
binding free energies (DG), shorter distance (d value) tances and the angles (data not shown), which is also
and larger angles (a1 and a2), corresponding to NACs in agreement with the measured regioselectivity coef-
or more generally to productive positions,[15d] as ex- ficients.
pected. It is also observed that (S)-5 is more likely to
What about other types of epoxides using BMEH?
produce reactive conformations, which might be the Thus, we further tested some other typical compounds
reason for the inverted enantioselectivity. This analy- 11–16 (Scheme 1).[17] Unfortunately, low E values
1514
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1510 – 1518

An Unusual (R)-Selective Epoxide Hydrolase with High Activity
Table 4. Results of docking experiments.[a]
Substrate
ÀDG [kcalmolÀ1][b]
d [ꢂngstroms][c]
a1 [deg][c]
a2 [deg][c]
R
S
R
S
R
S
R
S
2
3
4
5
25.3
23.4
23.0
27.5
22.8
22.6
21.4
29.0
3.1
3.2
3.2
3.2
3.7
3.8
3.6
2.9
136.8
136.9
128.3
121.9
106.8
110.1
105.1
153.7
85.1
84.2
74.6
73.4
55.0
61.7
51.8
99.9
[a]
The data shown are the average values of three independent docking simulations.
The binding free energy was calculated by MM-GBSA of the compounds.
d: the distance between the Asp97 oxygen and the attacked epoxide carbon; a1: the angle from the Asp97 oxygen via the
attacked epoxide carbon to the epoxide oxygen; a2: the angle from the Asp97 oxygen via the attacked epoxide carbon to
the other epoxide carbon.
[b]
[c]
Table 5. Semi-preparative scale resolution of epoxides with Conclusions
BMEH.
In summary, the newly cloned BMEH from B. mega-
terium ECU1001 was demonstrated to be a very effi-
cient biocatalyst for the kinetic resolution of rac-gly-
cidyl ethers. It was observed that the o-NO2 substitu-
ent causes a dramatic switch in both the enantioselec-
tivity and regioselectivity of the enzyme. Molecular
docking was then performed to uncover the origin of
the inverted enantioselectivity. Given the low identity
to other currently known EH structures (<25%), we
hope that the crystal structure of BMEH might pro-
vide deeper insights into the mechanism of EHs, and
BMEH might become an attractive catalyst for facile
Substrate Time
[min]
Conversion eeS [%]/config- Yield
[%][a]
uration
[%][b]
1
2
5
6
5
58
52
52
51
50
>99/(S)
>99/(S)
>99/(R)
>99/(S)
>99/(R)
31
40
41
45
47
15
90
30
150
11
[a]
[b]
Determined by HPLC analysis.
Isolated yield.
were observed for most of them, while in the case of preparation of enantiopure epoxides.
11, an unexpected excellent E value (>200) was ob-
served in favor of (S)-11. Encouraged by this result,
we decided to further investigate various styrene Experimental Section
oxide derivatives,[12,18] and this work is in progress.
In order to confirm the results from the analytical General Remarks
scale experiments, semi-preparative 200-mg scale res-
The racemic epoxides 3–7, 10–11 and 15–16 were synthe-
olutions of epoxides 1, 2, 5, 6 and 11 were carried out
using lyophilized BMEH powder (crude enzyme). As
a consequence, enantiopure epoxides (ee>99%) were
obtained with good yields (Table 5). To the best of
our knowledge, this is the first report of preparing
(S)-6 by bioresolution which is the solely useful anti-
pode for b-blocker synthesis.[19] Subsequently, without
any optimization, a gram-scale preparation of (S)-2
was achieved within 6 h at an elevated substrate con-
centration of 30 g/L. This afforded (S)-2 in a nearly
enantiopure form (98% ee) in 32% isolated yield and
the antipodal (R)-diol (87% ee) in 40% isolated yield.
In contrast, the biohydrolysis of 2 with growing cells
of B. alcalophilus afforded (S)-2 with 99% ee after
24 h at a low concentration of 1 g/L,[4a] indicating that
BMEH is a more efficient catalyst. Further studies on
optimizing the reaction conditions are in progress, in
order to improve the utility of BMEH for large-scale
application.
sized according to the methods previously reported with
some modifications. All other chemicals were obtained com-
mercially and used without further purification. The
1H NMR spectra were recorded on a Brucker 500 MHz
spectrometer, using the d scale (ppm) for chemical shifts.
GC was performed on a Shimadzu GC-14C, and HPLC was
performed on a Shimadzu LC-10AT. Thin layer chromatog-
raphy (TLC) was carried out on RSG F254 silica gel sheets
using a mobile phase of petroleum ether and ethyl acetate
(EtOAc). Flash chromatography was performed on silica gel
(300–400 mesh). Optical rotations were measured in a 10-
cm cell on a Rudolph Research Autopol I automatic polar-
imeter.
Identification and Cloning of the BMEH Gene
To clone EHs from B. megaterium ECU1001, sequence
searches (H-G-X-P and Sm-X-d-X-Sm-Sm motif) against
ORFs of B. megaterium QM B1551 were performed using
the GLIMMER program of NCBI and the basic local align-
ment search tool (BLAST) program. The pairwise compari-
sons of candidate EHs and reported EHs were performed
Adv. Synth. Catal. 2011, 353, 1510 – 1518
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1515

Jing Zhao et al.
FULL PAPERS
with the CLUSTAL W program. The resulting candidates
were manually confirmed for the presence of the putative
EHs active-site residues. Sequences that contained the cata-
lytic triad, HGXP motif, Sm-X-D-X-Sm-Sm (Sm=small res-
idue and X=any residue) motif and at least one ring-open-
ing tyrosine were selected and aligned together with the
known EHs sequences, and the distribution of the secondary
structures was also labeled.
ACHTUNGTRENNUNG
tained as a colourless oil after column chromatography;
yield: 68%. H NMR (500 MHz, CDCl3, TMS): d=2.33 (s,
3H, CH3), 2.74–2.76 (m, 1H, CH2), 2.89–2.91 (m, 1H, CH2),
3.33–3.36 (m, 1H, CH), 3.95 (dd, J=11.0, 5.6 Hz, 1H, CH2),
4.19 (dd, J=11.0, 3.2 Hz, 1H, CH2), 6.72–6.79 (m, 3H,
ArH), 7.15–7.18 (m, 1H, ArH).
1
AHCTUNGTREG(NNNU R,S)-1,2-Epoxy-3-(4-methylphenoxy)propane (4): Ob-
Genomic DNA of B. megaterium ECU1001 was isolated
using the Genomic DNA extraction kit (TIANGEN, Bei-
jing, China) following the manufacturerꢀs instructions. The
full-length of bmeh gene flanked by BamHI and SalI sites
tained as a colourless oil after column chromatography;
yield: 71%. H NMR (500 MHz, CDCl3, TMS): d=2.28 (s,
3H, CH3), 2.74–2.75 (m, 1H, CH2), 2.88–2.90 (m, 1H, CH2),
3.33–3.35 (m, 1H, CH), 3.93 (dd, J=11.0, 5.6 Hz, 1H, CH2),
4.18 (dd, J=11.0, 3.1 Hz, 1H, CH2), 6.75–6.83 (m, 2H,
ArH), 7.07–7.09 (m, 2H, ArH).
1
was amplified by
(bmehF: 5’-CACGGATCCATGAGTAAA CAGTATATAAACGT-3’)
and reverse primer (bmehR:
5’-GGCGTCGACTTACTTATTTAAAAAATTCCACAT-3’). The
a PCR with the forward primer
AHCTUNGTREG(NNUN R,S)-1,2-Epoxy-3-(2-nitrophenoxy)propane (5): Obtained
as a light yellow solid after recrystallization; yield: 31%.
1H NMR (500 MHz, CDCl3, TMS): d=2.87–2.88 (m, 1H,
CH2), 2.92–2.94 (m, 1H, CH2), 3.40 (s, 1H, CH), 4.14–4.17
(m, 1H, CH2), 4.41 (dd, J=11.2, 2.5 Hz, 1H, CH2), 7.06–
7.14 (m, 2H, ArH), 7.52–7.55 (m, 1H, ArH), 7.85–7.86 (m,
1H, ArH).
italic sequences indicate the BamHI site in the forward
primer and the SalI site in the reverse primer. The amplified
DNA fragment was digested with BamHI and SalI, the frag-
ment was ligated to BamHI/SalI-digested plasmid pET28a
(+) and then the recombinant plasmid was transformed into
E. coli DH5a cells. The recombinant plasmid was intro-
AHCTUNGTREG(NNUN R,S)-1,2-Epoxy-3-(3-nitrophenoxy)propane (6): Obtained
duced into E. coli BL21
confirmation.
G
as a light yellow solid after recrystallization; yield: 56%.
1H NMR (500 MHz, CDCl3, TMS): d=2.79–2.81 (m, 1H,
CH2), 2.94–2.96 (m, 1H, CH2), 3.39–3.40 (m, 1H, CH), 4.00
(dd, J=11.0, 6.0 Hz, 1H, CH2), 4.37–4.39 (m, 1H, CH2),
7.27–7.29 (m, 1H, ArH), 7.43–7.47 (m, 1H, ArH), 7.76–7.80
(m, 1H, ArH), 7.85–7.89 (m, 1H, ArH).
Preparation and Purification of BMEH
Preparative-scale production of proteins was achieved by
cultivating the recombinant E. coli in a 5-L fermentor
(Baoxing Bioengineering Equipment Co. Ltd, Shanghai,
China) as described previously.[20] The cells were harvested
by centrifugation and used directly for the enzyme lyophili-
zation or protein purification. The harvested cells were re-
suspended in potassium phosphate buffer (100 mM, pH 7.0)
AHCTUNGTREG(NNUN R,S)-1,2-Epoxy-3-(4-nitrophenoxy)propane (7): Obtained
as a light yellow solid after recrystallization; yield: 24%.
1H NMR (500 MHz, CDCl3, TMS): d=2.78–2.80 (m, 1H,
CH2), 2.95–2.96 (m, 1H, CH2), 3.37–3.40 (m, 1H, CH), 4.01
(dd, J=11.1, 6.0 Hz, 1H, CH2), 4.38 (dd, J=11.1, 2.6 Hz,
1H, CH2), 7.00 (d, J=9.2 Hz, 2H, ArH), 8.22 (d, J=9.2 Hz,
2H, ArH).
and disrupted twice by
AHCTUNGTREG(NNNU R,S)-1,2-Epoxy (1-napthoxy)propane (10): Obtained as a
(AH110B, ATS Engineering Inc.). Cell debris was removed
by centrifugation (10,000 g, 48C, 15 min). To the supernatant
a lactose solution of 1% (w/v) was added for enzyme protec-
tion before lyophilization. As a result, 40 g of powder with a
specific activity of 12,500 U/g powder were obtained and
stored at 48C for the following reactions unless otherwise
stated. The purification of BMEH was performed with Ni2+-
NTA beads according to the standard protocol. The purity
of the protein was examined by SDS-PAGE using a 15%
separating and 4% stacking gel.
light yellow oil after column chromatography; yield: 34%.
1H NMR (500 MHz, CDCl3, TMS): d=2.84–2.85 (m, 1H,
CH2), 2.95–2.97 (m, 1H, CH2), 3.49–3.50 (m, 1H), 4.14 (dd,
J=10.9, 5.6 Hz, 1H, CH2), 4.39 (dd, J=11.0, 3.0 Hz, 1H,
CH2), 6.80 (d, J=7.6 Hz, 1H, ring naphthyl), 7.34–7.37 (m,
1H, ring naphthyl), 7.44–7.51 (m, 3H, ring naphthyl), 7.79–
7.80 (m, 1H, ring naphthyl), 8.29–8.31 (m, 1H, ring naph-
thyl).
Synthesis of Substrate 16
rac-16 was prepared from benzaldehyde and methyl chloro-
AHCTUNGTREGaNNNU cetate in the presence of sodium methoxide, and purified
General Procedure for Synthesis of 3–7 and 10
rac-Glycidyl ethers were synthesized from corresponding
phenols and rac-epichlorohydrin according to the method
previously reported with some modifications.[21] Phenol
(50 mmol) was dissolved in 75 mL of 0.8M sodium hydrox-
ide and the mixture was stirred for 30 min. Epichlorohydrin
(5.85 mL, 75 mmol) was added and the mixture was stirred
vigorously at ambient temperature for about 5 h. The homo-
genous solution was then extracted three times with 15 mL
dichloromethane and the organic fractions were combined
and washed with 5% sodium hydroxide followed by brine.
The dichloromethane was dried over anhydrous sodium sul-
fate and removed under reduced pressure, giving a solid or
oil. The crude products were recrystallized from EtOAc/pe-
troleum ether or purified by flash chromatography.
by flash chromatography according to the method previous-
ly reported.[22] The product was obtained as a colourless oil
1
after column chromatography. H NMR (500 MHz, CDCl3,
TMS): d=3.51–3.52 (m, 1H, CH), 3.82 (s, 3H, CH3), 4.10–
4.10 (m, 1H, CH), 7.28–7.31 (m, 2H, ArH), 7.33–7.38 (m,
3H, ArH).
General Procedure for Enzymatic Reactions on an
Analytical Scale
To a stirred solution of BMEH powder (amount added as
indicated in Table 2) in potassium phosphate buffer
(100 mM, pH 7.0), the appropriate amount of epoxide (final
concentration 5–20 mM) was added from a stock solution in
1516
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1510 – 1518

An Unusual (R)-Selective Epoxide Hydrolase with High Activity
DMSO to a total volume of 0.5 mL. The final concentration
of DMSO was 5–10%. The reaction mixture was stirred in a
thermoshaker (Eppendorf, Germany) at 308C for the appro-
priate time. The reaction mixture was extracted twice with
0.5 mL of EtOAc, the combined organic layers were dried
over anhydrous magnesium sulfate, filtered, and analyzed by
chiral HPLC or GC to determine enantioselectivity.
Molecular Modeling
The crystal structure of human soluble epoxide hydrolase
with an inhibitor bound was retrieved from the Protein
Data Bank (PDB, entry code: 3I28) and chosen as the tem-
plate. The EH models were constructed using Modeller 9v8
with loops refined sufficiently. Several models were obtained
and validated by Procheck and Profile-3D.[26] The best
model was shown in the Ramachandran plot, where 90.6%
residues were located in the most favoured regions, 8.6% in
the additional allowed regions, 0.8% in the generously al-
lowed regions and no residues located in disallowed regions.
The RMSD between the model and 3I28 crystal structure
was 1.24 ꢂ. The alignment score was 0.06 calculated by pro-
tein structure alignment panel of Schrçdinger software pack-
age. The results from Profile-3D showed that overall com-
patibility score for this model was 107.8 in the scale of the
expected value 58.75–130.55. All these data indicated that
the model was reasonable and reliable.
Before carrying out molecular docking, all the tested sub-
strates were treated using the program LigPrep of Maestro
9.0 (Schrodinger LLC) to generate lowest energy conforma-
tions for all enantiomers. And the molecular docking was re-
peated three times using Glide (version 5.5) in standard pre-
cision (SP) mode. The Schrçdingerꢀs proprietary GlideScore
was utilized to score the binding mode. One best binding
pose for each enantiomer was written out and analyzed.[27]
General Procedure for Semi-Preparative Scale
Biohydrolysis of Epoxides 1–2, 5–6 and 11
BMEH powder (285 mg for 1 and 2, 570 mg for 5, 6 and 11)
was rehydrated in potassium phosphate buffer (90 mL,
100 mM, pH 7.0) for 30 min on a shaker (180 rpm, 308C).
Then 10 mL DMSO containing 2 mmol of the substrate
were added and the mixture was agitated at 308C. The reac-
tion progress was monitored by TLC or HPLC, and termi-
nated when the ee value of the residual epoxide reached
99% by adding 150 mL of EtOAc. Then the reaction mix-
ture was extracted twice again with EtOAc (150 mL) after
saturation with NaCl. After drying over anhydrous sodium
sulfate, solvents were removed under vacuum. Purification
was performed by flash chromatography.
(S)-1,2-Epoxy-3-phenoxypropane
(1):
Yield:
31%
(93.1 mg); >99% ee; [a]3D0: +3.0 (c 1.0, CHCl3), Lit.[19] [a]2D0:
+3.16 (c 0.95, CHCl3).
(S)-1,2-Epoxy-3-(2-methylphenoxy)propane (2): Yield:
40% (130.4 mg); >99% ee; [a]3D0: +13.0 (c 1.0, EtOH),
Lit.[23] [a]D: +14.8 (c 0.5, EtOH).
Assignment of Absolute Configuration
Absolute configurations were determined by chiral HPLC
or GC analysis by comparison of retention times with refer-
ence materials except for (S)-6 and (S)-8. In the case of ep-
oxide 6, the absolute configuration of (S)-6 was established
by comparisons of the specific rotation with the literature
values. The absolute configuration of (S)-8 was determined
from the sign of its optical rotation. Particularly, the abso-
lute configuration of (R)-5 was confirmed by the following
two methods: (i) chiral HPLC analysis by comparison of re-
tention time with reference values;[4a] (ii) deduced from the
sign of optical rotation based on the fact that all the (R)-ep-
oxides from various substituted phenyl glycidyl ether deriva-
tives (of known absolute configuration) always exhibit nega-
tive rotations.
(R)-1,2-Epoxy-3-(2-nitrophenoxy)propane (5): Yield: 41%
(160.0 mg); >99% ee; [a]3D0: À9.7 (c 1.0, CHCl3).
(S)-1,2-Epoxy-3-(3-nitrophenoxy)propane (6): Yield: 45%
(175.6 mg); >99% ee; [a]3D0: +2.0 (c 1.0, CHCl3), Lit.[19]
[a]20: +1.22 (c 1.02, CHCl3).
(DR)-2-(4-nitrophenyl)oxirane (11): Yield: 47% (154.5 mg);
>99% ee; [a]30: À41.0 (c 1.0, CHCl3), Lit.[24] [a]D: À34.7 (c
1.0, CHCl3) foDr 91% ee.
General Procedure for Preparative Scale
Biohydrolysis of Epoxide 2
In a 250-mL reactor, to rac-2 (18.3 mmol) dissolved in
DMSO (10 mL) was added a solution of BMEH powder
(1.0 g) in demineralized water (90 mL) then the mixture was
stirred at 500 rpm with a mechanical engine, and maintained
at 308C. The reaction was stopped when the ee of the resid-
ual epoxide reached 98% by adding 150 mL of dichlorome-
thane. The aqueous layer was saturated with NaCl then ep-
oxide and diol were extracted twice again with dichlorome-
thane (150 mL). After drying over anhydrous sodium sul-
fate, the solvents were removed under vacuum. Purification
was performed by flash chromatography. The diol was ob-
tained as a white solid; yield: 40%; 87% ee; [a]2D5: +18.0 (c
1.0, hexane/i-PrOH, 4:1), Lit.[25] [a]2D0: +19.3 (c 1.2, hexane/i-
PrOH, 4:1). 1H NMR (500 MHz, CDCl3, TMS): d=2.00–
2.03 (m, 1H, OH), 2.21–2.24 (m, 3H, CH3), 2.56 (d, J=
5.0 Hz, 1H, OH), 3.77–3.82 (m, 1H, CH2), 3.85–3.90 (m,
1H, CH2), 4.04–4.09 (m, 2H, CH2), 4.12–4.17 (m, 1H, CH),
6.83 (d, J=7.9 Hz, 1H, ArH), 6.88–6.91 (m, 1H, ArH),
7.15–7.18 (m, 2H, ArH).
Enzyme Assay
To a stirred solution of purified BMEH in potassium phos-
phate buffer (100 mM, pH 7.0), the appropriate amount of
PGE (final concentration 2 mM) was added from a stock so-
lution in DMSO to a total volume of 1 mL and then incu-
bated at 308C. During incubation, the samples were with-
drawn periodically, and the harvested mixtures were extract-
ed with ethyl acetate. The resulting extracts were analyzed
by chiral HPLC analysis. One EH unit (U) was defined as
the amount of enzyme required for the hydrolysis of 1 mmol
of PGE (1) per minute under the assay conditions.
Chiral Analysis of Epoxides and Diols
The enantiomeric excess of the epoxides and diols was de-
termined using the following columns: Chiralcel OD-H
(Daicel, Japan): epoxides 1–4 and 9–10; Chiralpak AD-H
Adv. Synth. Catal. 2011, 353, 1510 – 1518
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1517

Jing Zhao et al.
FULL PAPERS
(Daicel, Japan): epoxides 5, 7, 11 and 15–16; BETA DEXTM
120 (Sulpeco Inc, USA): epoxides 12 and 13; Chiralcel OJ-
H (Daicel, Japan): epoxides 6, 8 and 14. The antipodal diols
of 2–4 and 5 were analyzed using the Chiralcel OD-H and
Chiralpak AD-H column, respectively.
[8] B. van Loo, J. Kingma, M. Arand, M. G. Wubbolts,
D. B. Janssen, Appl. Environ. Microbiol. 2006, 72,
2905–2917.
[9] Y.-F. Tang, J.-H. Xu, Q. Ye, B. Schulze, J. Mol. Catal.
B: Enzym. 2001, 13, 61–68.
[10] Y. Xu, J.-H. Xu, J. Pan, Y.-F. Tang, Biotechnol. Lett.
2004, 26, 1217–1221.
[11] J.-H. Woo, J.-H. Kang, Y.-O. Hwang, J.-C. Cho, S.-J.
Kim, S. G. Kang, J. Biosci. Bioeng. 2010, 109, 539–544.
[12] P. Moussou, A. Archelas, J. Baratti, R. Furstoss, J. Org.
Chem. 1998, 63, 3532–3537.
Acknowledgements
This work was financially supported by the National Natural
Science Foundation of China (Nos. 20902023 and 31071604),
Ministry of Science and Technology, P.R. China
(2011CB710800 and 2009ZX09501-016), and China National
Special Fund for State Key Laboratory of Bioreactor Engi-
neering (No. 2060204). Cordial thanks are also given to Prof.
Dr. Manfred T. Reetz at Max-Planck-Institut fꢀr Kohlenfor-
schung for his constructive discussion, and to Dr. Jie Sun at
ECUST for her valuable suggestions in the manuscript prepa-
ration.
[13] P. Moussou, A. Archelas, J. Baratti, R. Furstoss, Tetra-
hedron: Asymmetry 1998, 9, 1539–1547.
[14] S. Pedragosa-Moreau, C. Morisseau, J. Zylber, A. Ar-
chelas, J. Baratti, R. Furstoss, J. Org. Chem. 1996, 61,
7402–7407.
[15] a) B. Schiøtt, T. C. Bruice, J. Am. Chem. Soc. 2002, 124,
14558–14570; b) S. Henrich, I. Feierberg, T. Wang, N.
Blomberg, R. C. Wade, Proteins Struct. Funct. Bioinf.
2010, 78, 135–153; c) A. Tafi, A. van Almsick, F. Corel-
li, M. Crusco, K. E. Laumen, M. P. Schneider, M.
Botta, J. Org. Chem. 2000, 65, 3659–3665; d) H.
Zheng, M. T. Reetz, J. Am. Chem. Soc. 2010, 132,
15744–15751; e) M. T. Reetz, M. Bocola, L.-W. Wang,
J. Sanchis, A. Cronin, M. Arand, J. Zou, A. Archelas,
A.-L. Bottalla, A. Naworyta, S. L. Mowbray, J. Am.
Chem. Soc. 2009, 131, 7334–7343.
References
[1] Y. Simeꢃ, K. Faber, Tetrahedron: Asymmetry 2006, 17,
402–409.
[2] A. Kamal, Y. Damayanthi, M. V. Rao, Tetrahedron:
[16] S. Barth, M. Fischer, R. D. Schmid, J. Pleiss, Proteins
Asymmetry 1992, 3, 1361–1364.
Struct. Funct. Bioinf. 2004, 55, 846–855.
[17] A. Steinreiber, K. Faber, Curr. Opin. Biotechnol. 2001,
12, 552–558.
˙
´
[3] D. Zelaszczyk, K. Kiec-Kononowicz, Curr. Med. Chem.
2007, 14, 53–65.
[4] a) N. Bala, S. S. Chimni, H. S. Saini, B. S. Chadha, J.
Mol. Catal. B: Enzym. 2010, 63, 128–134; b) Y. Xu, J.-
H. Xu, J. Pan, L. Zhao, S.-L. Zhang, J. Mol. Catal. B:
Enzym. 2004, 27, 155–159; c) W. J. Choi, E. C. Huh,
H. J. Park, E. Y. Lee, C. Y. Choi, Biotechnol. Tech.
1998, 12, 225–228.
[18] M. Cleij, A. Archelas, R. Furstoss, J. Org. Chem. 1999,
64, 5029–5035.
[19] M. G. Perrone, L. Bleve, E. Santandrea, P. Vitale, M.
Niso, A. Scilimati, ChemMedChem 2009, 4, 2080–2097.
[20] X. Liu, J.-H. Xu, J. Pan, J. Zhao, Appl. Biochem. Bio-
technol. 2010, 162, 1574–1584.
[21] a) B. M. Khadilkar, P. M. Bendale, Synth. Commun.
1997, 27, 2051–2056; b) J. L. Wright, T. F. Gregory,
T. G. Heffner, R. G. MacKenzie, T. A. Pugsley, S. V.
Meulen, L. D. Wise, Bioorg. Med. Chem. Lett. 1997, 7,
1377–1380.
[22] a) P. Crotti, M. Ferretti, F. Macchia, A. Stoppioni, J.
Org. Chem. 1986, 51, 2759–2766; b) T. R. Hoover, J. A.
Groeper, R. W. Parrott II, S. P. Chandrashekar, J. M. Fi-
nefield, A. Dominguez, S. R. Hitchcock, Tetrahedron:
Asymmetry 2006, 17, 1831–1841.
[23] A. Bisi, A. Rampa, R. Budriesi, S. Gobbi, F. Belluti, P.
Ioan, E. Valoti, A. Chiarini, P. Valenti, Bioorg. Med.
Chem. 2003, 11, 1353–1361.
[24] C. Barbieri, L. Bossi, P. D’Arrigo, G. P. Fantoni, S.
Servi, J. Mol. Catal. B: Enzym. 2001, 11, 415–421.
[25] A. A. Bredikhin, Z. A. Bredikhina, V. G. Novikova,
A. V. Pashagin, D. V. Zakharychev, A. T. Gubaidullin,
Chirality 2008, 20, 1092–1103.
[26] S. Khajamohiddin, P. S. Babu, D. Chakka, M. Merrick,
A. Bhaduri, R. Sowdhamini, D. Siddavattam, Biochem.
Biophys. Res. Commun. 2006, 351, 675–681.
[27] S. J. Taylor, F. Soleymanzadeh, A. B. Eldrup, N. A.
Farrow, I. Muegge, A. Kukulka, A. K. Kabcenell, S. D.
Lombaert, Bioorg. Med. Chem. Lett. 2009, 19, 5864–
5868.
[5] M. S. Smit, M. Labuschagnꢄ, Curr. Org. Chem. 2006,
10, 1145–1161.
[6] J. H. L. Spelberg, R. Rink, R. M. Kellogg, D. B. Jans-
sen, Tetrahedron: Asymmetry 1998, 9, 459–466.
[7] a) M. T. Reetz, J. Org. Chem. 2009, 74, 5767–5778;
b) F. H. Arnold, Nature 2001, 409, 253–257; c) L. Rui,
L. Cao, W. Chen, K. F. Reardon, T. K. Wood, Appl. En-
viron. Microbiol. 2005, 71, 3995–4003; d) B. van Loo,
J. H. L. Spelberg, J. Kingma, T. Sonke, M. G. Wubbolts,
D. B. Janssen, Chem. Biol. 2004, 11, 981–990; e) M. T.
Reetz, C. Torre, A. Eipper, R. Lohmer, M. Hermes, B.
Brunner, A. Maichele, M. Bocola, M. Arand, A.
Cronin, Y. Genzel, A. Archelas, R. Furstoss, Org. Lett.
2004, 6, 177–180; f) J. H. L. Spelberg, R. Rink, A. Ar-
chelas, R. Furstoss, D. B. Janssen, Adv. Synth. Catal.
2002, 344, 980–985; g) M. T. Reetz, L.-W. Wang, M.
Bocola, Angew. Chem. 2006, 118, 1258–1263; Angew.
Chem. Int. Ed. 2006, 45, 1236–1241; h) E. Y. Lee, M. L.
Shuler, Biotechnol. Bioeng. 2007, 98, 318–327; i) R.
Kourist, P. Domꢅnguez de Marꢅa, U. T. Bornscheuer,
ChemBioChem 2008, 9, 491–498; j) H. Jochens, K.
Stiba, C. Savile, R. Fujii, J.-G. Yu, T. Gerassenkov, R. J.
Kazlauskas, U. T. Bornscheuer, Angew. Chem. 2009,
121, 3584–3587; Angew. Chem. Int. Ed. 2009, 48, 3532–
3535.
1518
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1510 – 1518