Synthesis of enantiopure epoxide by 'one pot' chemoenzymatic approach using a highly enantioselective dehydrogenase

Article abstract of DOI:10.1016/j.tetlet.2016.01.048

Enantiopure α-phenethyl alcohols, including aromatic halohydrins, are important chiral building blocks. One of the best approaches to synthesise α-phenethyl alcohols is asymmetric reduction of prochiral ketones by alcohol dehydrogenases (ADHs). The obtained enantiopure halohydrin could be directly used to produce enantiopure epoxide through a base-induced ring-closure reaction, which is an attractive 'one pot' chemoenzymatic method for producing high-yield epoxide. In this study, a novel medium-chain dehydrogenase (KcDH) from Kuraishia capsulate CBS1993 was identified and characterised to show its broad substrate scope and excellent enantioselectivity. KcDH showed activities on 25 substrates of the 26 tested aromatic ketones and heteroaryl ketones, with an enantiomeric excess (ee) >99% and the highest relative activity observed with para-nitro acetophenone. Due to its high enantioselectivity for α-haloketones, a chemoenzymatic method for the synthesis of enantiopure styrene oxide (SO) and phenyl glycidyl ether (PGE) was developed through a base-induced ring-closure reaction on enantiopure halohydrin obtained with KcDH. (R)-SO and (S)-PGE were obtained in 86% and 94% analytical yield, respectively, and both epoxides were obtained with ee >99%. Thus, our results suggested that KcDH may be a promising biocatalyst for the production of multiple enantiopure α-phenethyl alcohols and epoxides.

Full text of DOI:10.1016/j.tetlet.2016.01.048

Tetrahedron Letters 57 (2016) 899–904
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of enantiopure epoxide by ‘one pot’ chemoenzymatic
approach using a highly enantioselective dehydrogenase
⇑
⇑
Kai Wu, Lifeng Chen, Haiyang Fan, Zhiqiang Zhao, Hualei Wang , Dongzhi Wei
State Key Laboratory of Bioreactor Engineering, New World Institute of Biotechnology, East China University of Science and Technology, Shanghai 200237, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Enantiopure
One of the best approaches to synthesise
a
-phenethyl alcohols, including aromatic halohydrins, are important chiral building blocks.
-phenethyl alcohols is asymmetric reduction of prochiral
Received 17 November 2015
Revised 11 January 2016
Accepted 13 January 2016
Available online 14 January 2016
a
ketones by alcohol dehydrogenases (ADHs). The obtained enantiopure halohydrin could be directly used
to produce enantiopure epoxide through a base-induced ring-closure reaction, which is an attractive ‘one
pot’ chemoenzymatic method for producing high-yield epoxide. In this study, a novel medium-chain
dehydrogenase (KcDH) from Kuraishia capsulate CBS1993 was identified and characterised to show its
broad substrate scope and excellent enantioselectivity. KcDH showed activities on 25 substrates of the
26 tested aromatic ketones and heteroaryl ketones, with an enantiomeric excess (ee) >99% and the
highest relative activity observed with para-nitro acetophenone. Due to its high enantioselectivity for
Keywords:
Alcohol dehydrogenase
Enantioselectivity
Epoxide
a-haloketones, a chemoenzymatic method for the synthesis of enantiopure styrene oxide (SO) and phenyl
Halohydrin
glycidyl ether (PGE) was developed through a base-induced ring-closure reaction on enantiopure halohy-
drin obtained with KcDH. (R)-SO and (S)-PGE were obtained in 86% and 94% analytical yield, respectively,
and both epoxides were obtained with ee >99%. Thus, our results suggested that KcDH may be a promising
biocatalyst for the production of multiple enantiopure
a-phenethyl alcohols and epoxides.
Ó 2016 Elsevier Ltd. All rights reserved.
Introduction
(S)-1-octene oxide, respectively, using a base-induced ring-closure
reaction.7,8
Optically pure aromatic secondary alcohols, such as
a
-phe-
Expanding the ADH toolbox remains an important topic in
redox biocatalysis. Based on the increased availability of public
genome information, many putative ADHs can be obtained from
Genbank.9 In this study, one putative ADH from Kuraishia capsulate
CBS1993 (KcDH) was heterologously overexpressed in Escherichia
coli, and its catalytic properties were studied in detail, revealing
a broad substrate specificity and excellent enantioselectivity. Fur-
thermore, due to its high enantioselectivity to halohydrin, we per-
formed a chemoenzymatic assay to synthesise two representative
epoxides, which showed promising results.
nethyl alcohol and its substitutive derivatives, represent an impor-
tant class of chiral building blocks that are versatile intermediates
for the synthesis of a vast number of bioactive compounds, includ-
ing agrochemicals, b-blockers, HIV-protease inhibitors and
anti-asthmatic agents.1 Additionally, the important intermediate
halohydrin can be converted into epoxide by adding alkali into
the reaction medium after reduction.2 Epoxides are also attractive
synthons used for fine organic synthesis due to their chemical
versatility.3
Asymmetric reduction of prochiral ketones using ADHs is an
attractive synthesis approach to producing
a-phenethyl alcohols,
Materials and methods
Materials
because it gives a 100% theoretical yield. Examples of industrial
ADHs include those from Thermoanaerobacter brockii ATCC
53556,4 Rhodococcus ruber DSM 44541,5 and Rhodococcus erythro-
polis DSM43297,6 with each demonstrating excellent properties
of enantioselectivity and substrate scope. Among these reported
ADHs, those from R. ruber DSM 44541 and Lactobacillus kefir were
successfully used in the synthesis of (R)-1,2-epoxyoctane and
All ketones, styrene oxide (SO), phenyl glycidyl ether (PGE) and
1-phenylethane-1,2-diol and 3-phenoxy-1,2-propanediol used in
this study are commercially available. The racemic sec-alcohol
standards of acetophenone-substitutive derivatives were obtained
by the reduction of ketone with 1.2 equiv of NaBH4 in 1% NaOH.
The four chloroalcohols were synthesised by reduction with corre-
sponding ketones in the presence of NaBH4 and CaCl2, with excess
⇑
Corresponding authors. Tel.: +86 21 64251803; fax: +86 21 64250068.
E-mail addresses: hlwang@ecust.edu.cn (H. Wang), dzhwei@ecust.edu.cn (D. Wei).
http://dx.doi.org/10.1016/j.tetlet.2016.01.048
0040-4039/Ó 2016 Elsevier Ltd. All rights reserved.

900
K. Wu et al. / Tetrahedron Letters 57 (2016) 899–904
HCl added during the reaction according to literature.10 The gen-
ome of Kuraishia capsulate CBS1993 was stored in our lab. E. coli
on KcDH activity were examined over a temperature range of 25 °C
to 50 °C in 50 mM potassium phosphate buffer (pH 7.0).
DH5
a
and E. coli BL21 (DE3) cells were grown in Luria–Bertani
CaCl2, CoCl2, CuCl2, FeCl3, MgCl2, MnCl2, NiCl2 and ZnCl2 were
used to test the effects of metal ions, and EDTA was used to test
whether KcDH requires essential metal ions to maintain its activ-
ity. The influences of various metal ions and EDTA on enzyme
(LB) medium and used as the cloning and expression hosts, respec-
tively. Column SB-AQ (4.6 mm ꢀ 250 mm; Agilent Technologies,
Santa Clara, CA, USA) was used for conversion-ratio determination.
The enantiomeric excess (ee) of sec-alcohol products PGE and SO
were analysed by high-performance liquid chromatography (HPLC)
or gas chromatography (GC) equipped with a chiral column: Chi-
ralcel OB-H column, OD-H column (250 mm ꢀ 4.6 mm; Daicel
Corp., Hyogo, Japan), Supelco fused-silica capillary column b-Dex
120 (Supelco, 30 m ꢀ 0.25 mm ꢀ 0.25
Louis, MO, USA,) and
(30 m ꢀ 0.25 mm ꢀ 0.25 m; Agilent Technologies).
activity were investigated by pre-incubating the enzyme (70 lg/
ml) at a certain concentration of each different compound in
50 mM potassium phosphate buffer (pH 7.0) for 20 min at 35 °C.
The enzyme activity was estimated using the previously described
standard assay protocol. Relative activity was expressed as a per-
centage of the activity in the absence of any test compound.
lm; Sigma–Aldrich, St.
CP-Chirasil-Dex CB
l
Substrate scope and enantioselectivity determination
Cloning, expression and purification of the KcDH gene
The relative activities of 26 substrates were measured using the
previously described assay protocol with adjusted ratio of enzyme
The full-length KcDH gene (accession number: CDK24134.1)
flanked by NdeI and HindIII restriction enzyme sites was amplified
by polymerase chain reaction using forward (50-GGAATTCCA-
TATGTCTGCTCTCTCCAAAACCCAGG-30) and reverse primers (50-
CCCAAGCTTTCAGCCCGCGGGGTGGTTCTCC-30). The obtained DNA
fragment was digested and ligated into the correspondingly
digested pET-28a(+) plasmid. The resulting recombinant plasmid
was then transformed into BL21 (DE3) cells for KcDH expression.
A single transformant was cultured at 37 °C for 12 h, and then
transferred to 100 mL fresh LB medium supplemented with kana-
and substrate concentration. The
was assumed 100%.
a-chloroacetophenone activity
Enantioselectivity was determined by examining the reduction
of aromatic ketones using an NADH-regeneration system consist-
ing of the purified KcDH and glucose dehydrogenase (GDH) from
Bacillus subtilis CGMCC 1.1398. The 1-mL reaction mixture con-
tained 0.5 mM NAD+, 10 mM ketone, 1 U KcDH, 50 mg glucose
and 2 U GDH in 50 mM potassium phosphate buffer (pH 7.0). After
16 h, the reaction sample was equally separated into two parts,
with one terminated by adding an equal volume of methanol, fol-
lowed by HPLC analysis to determine the conversion ratio, and the
other extracted with ethyl acetate, followed by ee analysis. Meth-
ods used for analysing chiral products using HPLC or GC are
described in Supplementary Table S1.
mycin (50 lg/mL) and cultured at 37 °C. The culture was induced
by the addition of 0.1 mM isopropyl b-D-1-thiogalactopyranoside
after reaching an OD600 of 0.6. After induction at 20 °C for 20 h,
cells were harvested by centrifugation at 8500g for 10 min, and
resuspended in 50 mM Tris–HCl (pH 8.0). After disrupting the cells
by sonication and removing cell debris/inclusion bodies by cen-
Epoxidation after reduction
trifugation, the soluble cell-free extract was filtered (0.22-lm fil-
ter; EMD Millipore, Billerica, MA, USA) and loaded onto a nickel
column pre-equilibrated with 50 mM Tris–HCl (pH 8.0) binding
buffer. After washing with the binding buffer, the bound recombi-
nant enzyme was eluted by applying binding buffer with increas-
ing concentrations of imidazole (20–200 mM). Pure KcDH was
obtained following the addition of 200 mM imidazole. The expres-
sion and purity of the protein were identified by sodium dodecyl
sulfate polyacrylamide gel electrophoresis on 12% gels.
a
-Chloroacetophenone and 3-chloro-1-phenoxy-2-propanone
were selected as typical substrates for chemoenzymatic epoxida-
tion. Resting E. coli cells containing KcDH (50 mg/mL) were used
as catalysts for this procedure, and 5% isopropanol was used as
the co-substrate for NADH regeneration. After reduction, 25 lL of
6 M NaOH was added to the 1-mL reaction medium (containing
10 mM product) to a final NaOH concentration of 146 mM. Three
samples were collected at different time points (5, 15 and
30 min), and epoxidation was terminated by adjusting the pH
between 8.0 and 9.0. The analytical yield was measured by HPLC
analysis with the solvents methanol and H2O at a ratio of 2:3.
The methods used to assay the ee of SO and PGE are illustrated
in Supplementary Table S1.
Determination of enzyme activity and kinetic parameters
KcDH activity was assayed spectrophotometrically at 35 °C by
measuring the change in absorbance at 340 nm of NADH in
1 min. The standard assay for the reduction reaction was per-
formed by adding 4
196 L of preheated assay mixture containing 14
0.5 mM NADH in 50 mM potassium phosphate buffer (pH 7.0).
The assay for oxidation activity was performed by adding 16
of 200 mM isopropanol to 184 L of preheated assay mixture
containing 4.2 g KcDH and 0.5 mM NAD in 50 mM potassium
phosphate buffer (pH 7.0). The kinetic parameters of -chloroace-
tophenone and isopropanol were determined by measuring initial
velocities at different substrate concentrations using a Linewea-
ver–Burk plot (1/v vs 1/[S]).
l
L of 200 mM
a
-chloroacetophenone to
Scale-up preparation of (R)-SO and (S)-PGE
l
l
g KcDH and
The reaction started at 30 °C by adding 624 mg a-chloroacetophe-
lL
none or 738 mg 3-chloro-1-phenoxy-2-propanone into 100 ml PBS
buffer (pH7.6) containing 100 mg/ml KcDH resting cell, 1 ml iso-
propanol, and 20 mg NAD. After bio-reduction was completed,
2.5 ml of 6 M NaOH was added for epoxidation. The epoxidation
was terminated after 30 min by neutralisation with 6 M HCl. The pro-
duct SO and PGE was extracted with hexane three times. After drying
over anhydrous sodium sulphate, solvents were removed under vac-
uum. The obtained SO and PGE were analysed by chiral HPLC and
GC, and the spectral data are presented in Supplemental material.
l
l
a
Effects of pH, temperature and metal ions on KcDH activity
The optimal pH for purified KcDH activity was investigated by
measuring its activity in 0.1 M of the following buffers with pH
ranging from 4 to 10.5: citrate buffer (pH 4.0, 5.0 and 6.0), phos-
phate buffer (pH 6.0, 7.0 and 8.0), Tris–HCl buffer (pH 8.0, 9.0
and 9.5) and carbonic buffer (pH 10.5). The effects of temperature
Results and discussion
A novel biocatalyst, KcDH from Kuraishia capsulate CBS1993,
was identified by genomic data mining and cloned, characterised,

K. Wu et al. / Tetrahedron Letters 57 (2016) 899–904
901
and overexpressed in E. coli. KcDH belongs to the medium-chain
dehydrogenase/reductase super family, which contains two types
of proteins: metalloproteins containing zinc and nonmetallopro-
teins without zinc.11 KcDH contains 341 amino acids and has
two typical Zn-binding regions that coordinate catalytic Zn
(Cys45, Ser47, His67, Asp158) and structural Zn (Cys99, Cys102,
Cys105, Cys113), respectively, implicating KcDH as a Zn-depen-
dent enzyme. The protein exhibiting the highest homology was a
secondary alcohol dehydrogenase from Candida auris, which
shared 49% amino acid sequence identity. After purification, KcDH
of KcDH to reduce 10 mM
a
-chloroacetophenone with 5% iso-
propanol, and observed complete conversion. Furthermore, the
kinetic parameters associated with -chloroacetophenone and iso-
propanol were obtained, indicating a Km and Vmax for -chloroace-
tophenone of 0.33 mM and 0.39
mol minꢁ1 mgꢁ1, respectively,
Km and Vmax for isopropanol of 1.5 mM and
mol minꢁ1 mgꢁ1, respectively. These data suggested iso-
a
a
l
and
a
4.13
l
propanol as a potential co-substrate for use in reduction reactions
catalysed by KcDH.
Metal ions may exhibit substantial effects on KcDH activity,
therefore, we investigated the effects of metal ions and EDTA on
enzyme activity (Table 1). Among the nine compounds tested,
KcDH displayed a slightly higher activity following the addition
of Mg2+ and Mn2+. There were no obvious differences on KcDH
activity when increasing the concentration of Mg2+ and Mn2+ to
5 mM. However, the presence of 1 mM Co2+, Cu2+ and Zn2+ consid-
erably diminished enzyme activity, and more serious inhibitions
were observed by increasing their concentrations to 5 mM. Fe3+
and Ni2+ completely inhibited KcDH activity, and Ca2+ demon-
strated no effect on KcDH activity.
The presence of the chelating agent EDTA also diminished
enzyme activity. The activity of KcDH decreased with the increas-
ing concentration of EDTA and was totally inhibited in 100 mM
EDTA (Fig 2), suggesting that KcDH required metals to sustain
the enzyme activity (Fig. 2). As a predicted Zn-dependent enzyme,
the loss of essential metal ions was assumed as the explanation for
the observed reduction in activity.12 All the tested reagents have no
effect on enantioselectivity.
enantioselectivity was first tested on
a-chloroacetophenone, and
showed ee >99%, which encouraged further investigation.
The optimal temperature and pH for activity were determined
as 35 °C and pH 8.0, respectively (Fig. 1), with KcDH showing the
highest activity level in phosphate buffer. Under optimal reaction
conditions, the specific activity of KcDH on a-chloroacetophenone
was measured at 0.38 U/mg. KcDH only used the less expensive
NADH instead of NADPH as a cofactor, which lowered the cost
associated with bioprocessing. NADH can be regenerated via sim-
ple hydrogen transfer using isopropanol as the hydrogen donor,
making the reaction system less complex as compared to the sys-
tem using GDH and glucose. Therefore, we investigated the ability
To map the substrate spectrum of KcDH, various ketones with
diverse structures were employed (Table 2), with four types of sub-
strates corresponding to high-value enantiopure alcohols investi-
gated. Type I substrates consisted of
a-substituted phenyl ethyl
ketones .The derivatives of type I products are building blocks used
for many adrenergic drugs, such as salbutamol, sotalol and nife-
nalol.13 For these drugs, the (R)-enantiomer exhibits much higher
physiological activity as compared to the (S)-enantiomer. Type II
substrates contain acetophenone ring-substitutive derivatives,
the products of which are generally used as building blocks for
selective histamine H1-antagonists14 and antisepsis agents.15 Type
III substrates represent heteroaryl ketones, of which the (ꢁ)-(S)-4-
pyridylethanol is used for the synthesis of multiple b3-blockers.
The (R)-alcohols of type IV substrates are important precursors
for adrenergic b-blockers.16 After epoxidation, (R)-1-chloro-3-phe-
noxy-2-propanol is converted into (S)-PGE (with a switch in Cahn-
Ingold-Prelog priority), which is used for the synthesis of adrener-
gic b-blockers.
The relative activity and ee of the products are illustrated in
Table 2. KcDH showed activity on 25 of the 26 substrates tested,
with relative activities ranging from 0% to 1779%, suggesting that
different influences on enzyme activity were caused by various
substituents. It was notable that all of the products were obtained
Table 1
Effect of metal ions and additives on reductase activity
Metal ions
1 mM
5 mM
Relative activity (%)
ee (%)
Relative activity (%)
ee (%)
Control
Ca2+
100
100
16
2.5
0
108
119
0
>99
>99
>99
—
100
100
0
0
0
109
116
0
>99
>99
—
—
—
>99
>99
—
Co2+
Cu2+
Fe3+
—
Figure 1. Effects of temperature and pH on KcDH activity. (A) The activities of
TpEH1 were tested at different temperatures. The value at 35 °C was set as 100%. (B)
pH ranging from 4 to 10.5: citrate buffer (pH 4, 5 and 6); phosphate buffer (pH 6, 7
and 8); Tris–HCl buffer (pH 8, 9 and 9.5); and carbonic buffer (pH 10.5). The value at
pH 8.0 was set as 100%. All experiments were performed in triplicate.
Mg2+
Mn2+
Ni2+
>99
>99
—
Zn2+
15
>99
5
—

902
K. Wu et al. / Tetrahedron Letters 57 (2016) 899–904
sion analysis to explain the correlation of electronic effects of ring
substituents and enzyme activity. They found that the electron-
withdrawing characteristic of the substituent enhanced the activ-
ity of certain ketoreductases,17 which was in agreement with our
results.
Enantiopure SO and PGE are two representative high-value
epoxides.3 The two corresponding optically pure epoxides, (R)-SO
and (S)-PGE, are the bioactive enantiomers required for the synthe-
sis of adrenergic receptor agonists and b-blockers, respectively.
Many chemical-based asymmetric synthesis methods for these ter-
minal epoxides have been described in recent years, such as Jacob-
sen asymmetric epoxidation and hydrolytic kinetic resolution.18
However, these elegant chemical methods suffer from potentially
toxic heavy-metal-based catalysts, very low reaction temperatures
(ꢁ20 °C),19 low ee,20 or low yield,21 especially for preparation of
enantiopure SO.
An alternative approach involves the biocatalytic asymmetric
reduction of
under basic conditions as a second step (Scheme 1). After biocat-
alytic reduction of -chloroacetophenone and chloro-1-phenoxy-
a-chloroketone, coupled with chemical ring-closure
Figure 2. KcDH activity assay of samples treated with different concentrations of
EDTA.
a
2-propanone, optically pure halohydrins could be easily converted
into epoxides by adding excessive base and maintaining its ee. We
developed this chemoenzymatic method using KcDH, based on its
high enantioselectivity, and obtained two epoxides with a high ee
and in a high yield. To increase the epoxide yield, NaOH concentra-
tion and reaction time for ring closure of halohydrin were investi-
gated. The optimal results are illustrated in Figure 3. The epoxide
yield reached maximum after 30 min at 146 mM NaOH. Under
these conditions, the analytical yield of (R)-SO was 86%, with ee
>99% and the analytical yield of (S)-PGE was 94%, with ee >99%.
Because the addition of base catalysed (R)-SO hydrolysis without
strict regioselectivity,22 vicinal diols were produced as by-prod-
ucts, with only 30% ee and 13% analytical yield. In order to explore
the application potential of the chemoenzymatic approach, we fur-
ther performed this chemoenzymatic method with 40 mM sub-
strate concentration in 100 ml volume. The (R)-SO was obtained
with 350 mg in nearly enantiopure form (98.0% ee) and the
with ee >99% after complete conversion. The high enantioselectiv-
ity and wide substrate scope demonstrated that KcDH is an excel-
lent biocatalyst for the production of optically pure alcohols. Based
on our results, we observed that KcDH activity increased gradually
according to the substituents in the order ortho- < meta- < para-.
Additionally, the strong electron-withdrawing groups, such as the
nitro group, resulted in a much higher activity relative to the other
substituents. We inferred that electronic and steric effects of phe-
nyl-substituents caused the phenomenon together. The typical
result was that para-nitro acetophenone demonstrated the highest
activity, while no activity was detected for ortho-nitro
acetophenone.
Some previous reports discussed the impact of phenyl sub-
stituents on catalytic efficiency. Dunming et al. investigated 24 dif-
ferent ketoreductases and 15 acetophenone derivatives,
introducing the Hammett substituent constant and using regres-
Table 2
Substrate
X
Ar
Relative activity
(%)
ee (%), Absolute
configuration
Substrate
X
Ar
Relative activity
(%)
ee (%), Absolute
configuration
I
S15
S16
S17
S18
H
H
H
H
40-Me-Ph
20-NO2-Ph
30-NO2-Ph
40-NO2-Ph
280
0
762
1779
>99, (S)
—
>99, (S)
>99, (S)
S1
S2
S3
Cl Ph
100
180
120
>99, (R)
>99, (R)
>99, (R)
Cl 40-Cl-Ph
Cl 40-MeO-
Ph
II
S19
S20
III
S21
S22
S23
S24
H
H
30-MeO-Ph
40-MeO-Ph
98
116
>99, (S)
>99, (S)
S4
S5
S6
S7
S8
S9
H
H
H
H
H
H
Ph
108
90
200
240
176
240
>99, (S)
>99, (S)
>99, (S)
>99, (S)
>99, (S)
>99, (S)
20-Cl-Ph
30-Cl-Ph
40-Cl-Ph
20-Br-Ph
30-Br-Ph
H
H
H
H
Pyridin-2-yl
Pyridin-3-yl
Pyridin-4-yl
Thiophen-2-
yl
77
30
156
12
>99, (S)
>99, (S)
>99, (S)
>99, (S)
S10
S11
S12
S13
S14
H
H
H
H
H
40-Br-Ph
30-F-Ph
460
128
212
40
>99, (S)
>99, (S)
>99, (S)
>99, (S)
>99, (S)
IV
S25
S26
Cl Ph-O
Cl 4-Me-Ph-O
220
108
>99,(R)
>99,(R)
40-F-Ph
20-Me-Ph
30-Me-Ph
240

K. Wu et al. / Tetrahedron Letters 57 (2016) 899–904
903
Scheme 1. Synthesis of enantiopure epoxide by chemoenzymatic approach. ‘one-pot two-steps’ cascade-reaction consisting of a biocatalytic reduction of
a-halo ketones
(step 1) and in situ base-induced ring closure to furnish the corresponding epoxide (step 2). The remaining base catalyses the hydrolysis of epoxide which results in the
byproduct vicinal diol.
Figure 3. Analytical yield of enantiopure SO and PGE. (A) Analytical yield of three components during SO epoxidation: 1-phenylethane-1,2-diol (light grey); (R)-2-chloro-1-
phenylethanol (grey); (R)-SO (black). (B) Analytical yield of three components during PGE epoxidation: 3-phenoxy-1,2-propanediol (light grey), (R)-1-chloro-3-phenoxy-2-
propanol (grey); (S)-PGE (black).
isolated yield was 73.1%. The (S)-PGE was obtained with 498 mg
(99.2% ee) and the isolated yield was 83.0%. The enantiopure epox-
ide could be easily extracted from reaction medium using hexane
and the byproduct remained in aqueous phase,23 which made the
downstream purification much easier.
report involved using ADH from Lactobacillus kefir, which was used
to prepare 1 M (S)-1-Octene oxide by adding 10 equiv of NaOH
after 1-Bromo-2-octanone was reduced to halohydrin. These
results indicated that this chemoenzymatic method using ADH
was a promising approach.
Reduction of
a-chloroacetophenone coupled with isopropanol
for NADH regeneration could lead to quasi-quantitative conver-
sion. During this investigation, we found that only 2.6 equiv of iso-
propanol was enough for >99% conversion of the substrate, which
further lowered the cost of this procedure. Furthermore, this ‘one
pot, two steps’ procedure using inexpensive alkali for epoxidation
was less complex as compared to previous methods using
two-enzyme cascades.24 The two-enzyme cascade using ADH and
halohydrin dehalogenase to produce epoxides must overcome
equilibrium issues by adding anion exchanging resins, and requires
excellent organic tolerance for both enzymes.
Conclusion
Here, the novel enzyme KcDH was expressed and identified, and
the basic catalytic properties were studied using
tophenone as a substrate. KcDH showed a wide substrate scope
and excellent enantioselectivity for 25 substrates, demonstrating
a-chloroace-
potential for the production of enantiopure
a-phenethyl alcohols
and aromatic halohydrins. Furthermore, we optimised the ‘one-
pot, two-steps’ chemoenzymatic approach to produce enantiopure
epoxides, obtaining desirable analytical yield and ee. Although the
substrate scope and ee for KcDH was satisfactory, the quantified
enzyme activity was still low. Improving KcDH activity through
protein engineering, and diminishing by-product formation will
constitute our future work on this process.
Besides ADHs, biocatalysts, including epoxide hydrolase,25 halo-
hydrin dehalogenases,26 cytochrome P45027 and styrene monooxy-
genase,28 were developed as alternatives for chemical methods.
Among these enzymes, epoxide hydrolase suffered a limitation of
50% theoretical yield, and halohydrin dehalogenases suffered an
equilibrium problem, requiring the removal of halide ions from
the reaction medium.24 Although cytochrome P450 displayed
100% theoretical yield, poor operational stability and low activity
remain a major challenge. The styrene monooxygenase from Pseu-
domonas sp. strain VLB120 was a promising biocatalyst that
showed bioprocesses superior to two chemical-process alterna-
tives, and slightly inferior to a third. The use of an additional toxic
bis(2-ethylhexyl)phthalate phase was the major drawback of this
process.29 Although few reports using ADH to produce enantiopure
epoxide exist,8,30,31 promising results were obtained. The best
1H NMR data of the synthesised enantiopure products
(R)-2-Chloro-1-phenylethanol: 1H NMR (500 MHz, CDCl3) d
7.50–7.27 (m, 5H, Ar), 4.90 (dd, J = 8.6, 2.4 Hz, 1H, CH), 3.74 (dd,
J = 11.2, 3.3 Hz, 1H, CH2Cl), 3.65 (dd, J = 11.1, 9.0 Hz, 1H, CH2Cl),
2.69 (s, 1H, OH).
(R)-1-Chloro-3-phenoxy-2-propanol: 1H NMR (500 MHz, CDCl3)
d 7.31 (t, J = 7.8 Hz, 2H, Ar), 7.00 (t, J = 7.3 Hz, 1H, Ar), 6.93 (d,
J = 8.1 Hz, 2H, Ar), 4.27–4.17 (m, 1H, CH), 4.15–4.03 (m, 2H,

904
K. Wu et al. / Tetrahedron Letters 57 (2016) 899–904
8. Berkessel, A.; Rollmann, C.; Chamouleau, F.; Labs, S.; May, O.; Gröger, H. Adv.
Synth. Catal. 2007, 349, 2697–2704.
9. Morales, L.; Noel, B.; Porcel, B.; Marcet-Houben, M.; Hullo, M.-F.; Sacerdot, C.;
Tekaia, F.; Leh-Louis, V.; Despons, L.; Khanna, V.; Aury, J.-M.; Barbe, V.; Couloux,
A.; Labadie, K.; Pelletier, E.; Souciet, J.-L.; Boekhout, T.; Gabaldon, T.; Wincker,
P.; Dujon, B. Genome Biol. Evol. 2013, 5, 2524–2539.
10. Wang, M.; Wu, J.; Yang, L.; Chen, X. Chin. J. Org. Chem. 2005, 25, 1548–1551.
11. Persson, B.; Hedlund, J.; Jornvall, H. Cell. Mol. Life Sci. 2008, 65, 3879–3894.
12. Man, H.; Loderer, C.; Ansorge-Schumacher, M. B.; Grogan, G. ChemCatChem
2014, 6, 1103–1111.
13. Yang, W.; Xu, J.-H.; Xie, Y.; Xu, Y.; Zhao, G.; Lin, G.-Q. Tetrahedron: Asymmetry
2006, 17, 1769–1774.
OCH2), 3.76 (ddd, J = 29.5, 11.2, 5.4 Hz, 2H, CH2Cl), 2.81 (d,
J = 3.2 Hz, 1H, OH).
(S)-Phenyl glycidyl ether: 1H NMR (500 MHz, CDCl3) d 7.30 (t,
J = 7.7 Hz, 2H, Ar), 7.09–6.80 (m, 3H, Ar), 4.22 (dd, J = 10.9, 2.6 Hz,
1H, OCH2), 3.97 (dd, J = 10.9, 5.6 Hz, 1H, OCH2), 3.39–3.33 (m, 1H,
CH), 2.91 (t, J = 4.2 Hz, 1H, CH2), 2.83–2.62 (m, 1H, CH2).
(R)-Styrene oxide: 1H NMR (500 MHz, CDCl3) d 7.52–7.11 (m,
5H, Ar), 3.86 (s, 1H, CH), 3.15 (dd, J = 4.7, 4.7 Hz, 1H, CH2), 2.80
(dd, J = 5.2, 2.0 Hz, 1H, CH2).
14. Li, A.; Wu, S.; Adams, J. P.; Snajdrova, R.; Li, Z. Chem. Commun. (Camb.) 2014,
8771–8774.
15. Yamada, M.; Ichikawa, T.; Yamano, T.; Kikumoto, F.; Nishikimi, Y.; Tamura, N.;
Kitazaki, T. Chem. Pharm. Bull. 2006, 54, 58–62.
Acknowledgements
This work was supported by the National Natural Science Foun-
dation of China (No. 21406068/B060804), the Fundamental
Research Funds for the Central Universities, and National Basic
Research Program of China (No. 2012CB721103).
16. Zhao, J.; Chu, Y.-Y.; Li, A.-T.; Ju, X.; Kong, X.-D.; Pan, J.; Tang, Y.; Xu, J.-H. Adv.
Synth. Catal. 2011, 353, 1510–1518.
17. Zhu, D.; Rios, B. E.; Rozzell, J. D.; Hua, L. Tetrahedron: Asymmetry 2005, 16,
1541–1546.
18. Hwang, S.; Choi, C. Y.; Lee, E. Y. J. Ind. Eng. Chem. 2010, 16, 1–6.
19. Matsumoto, K.; Oguma, T.; Katsuki, T. Angew. Chem., Int. Ed. 2009, 48, 7432–
7435.
20. Talsi, E. P.; Samsonenko, D. G.; Bryliakov, K. P. Chemistry 2014, 20, 14329–
14335.
Supplementary data
21. Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277, 936–
938.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2016.01.
048.
22. Wang, Z.; Cui, Y. T.; Xu, Z. B.; Qu, J. J. Org. Chem. 2008, 73, 2270–2274.
23. Chen, W.; lou, W.; Wang, X. Chin. J. Catal. 2011, 1557–1563.
24. Schrittwieser, J. H.; Lavandera, I.; Seisser, B.; Mautner, B.; Lutje Spelberg, J. H.;
Kroutil, W. Tetrahedron: Asymmetry 2009, 20, 483–488.
25. Wu, K.; Wang, H.; Sun, H.; Wei, D. Appl. Microbiol. Biotechnol. 2015, 1–11.
26. You, Z. Y.; Liu, Z. Q.; Zheng, Y. G. Appl. Microbiol. Biotechnol. 2013, 97, 9–21.
27. O’Reilly, E.; Kohler, V.; Flitsch, S. L.; Turner, N. J. Chem. Commun. (Camb.) 2011,
2490–2501.
28. Lin, H.; Qiao, J.; Liu, Y.; Wu, Z.-L. J. Mol. Catal. B Enzym. 2010, 67, 236–241.
29. Kuhn, D.; Kholiq, M. A.; Heinzle, E.; Bühler, B.; Schmid, A. Green Chem. 2010, 12,
815.
30. Bisogno, F. R.; Cuetos, A.; Orden, A. A.; Kurina-Sanz, M.; Lavandera, I.; Gotor, V.
Adv. Synth. Catal. 2010, 352, 1657–1661.
References and notes
1. Zhu, D.; Mukherjee, C.; Hua, L. Tetrahedron: Asymmetry 2005, 16, 3275–3278.
2. Fardelone, L. C.; Rodrigues, J. A.; Moran, P. J. Enzyme Res. 2011, 2011, 976368.
3. Kotik, M.; Archelas, A.; Wohlgemuth, R. Curr. Org. Chem. 2012, 16, 451–482.
4. Bogin, O.; Peretz, M.; Burstein, Y. Protein Sci. 1997, 6, 450–458.
5. Edegger, K.; Gruber, C. C.; Poessl, T. M.; Wallner, S. R.; Lavandera, I.; Faber, K.;
Niehaus, F.; Eck, J.; Oehrlein, R.; Hafner, A.; Kroutil, W. Chem. Commun. (Camb.)
2006.
6. Abokitse, K.; Hummel, W. Appl. Microbiol. Biotechnol. 2003, 62, 380–386.
7. Poessl, T. M.; Kosjek, B.; Ellmer, U.; Gruber, C. C.; Edegger, K.; Faber, K.;
Hildebrandt, P.; Bornscheuer, U. T.; Kroutil, W. Adv. Synth. Catal. 2005, 347,
1827.
31. Aguirre-Pranzoni, C.; Bisogno, F. R.; Orden, A. A.; Kurina-Sanz, M. J. Mol. Catal. B
Enzym. 2015, 114, 19–24.