Enantioselective synthesis of ethyl (S)-2-hydroxy-4-phenylbutyrate by recombinant diketoreductase

Article abstract of DOI:10.1016/j.tetasy.2009.10.036

Recombinant diketoreductase showed excellent stereoselectivity in the double reduction of β,δ-diketo esters. To investigate the substrate specificity and to broaden the applications of this new biocatalyst, a number of ketone substrates were used to evaluate the substrate spectrum and enantioselectivity of this enzyme in the present study. Among the ketone substrates tested, only this enzyme displayed high efficiency and excellent enantioselectivity in the reduction of ethyl 2-oxo-4-phenylbutyrate to ethyl (S)-2-hydroxy-4-phenylbutyrate. After optimizing the reaction conditions, the bio-reduction of ethyl 2-oxo-4-phenylbutyrate at a substrate concentration of 0.8 M (164.8 g/L) was achieved by the recombinant diketoreductase in an aqueous-toluene biphasic system coupled with formate dehydrogenase for the regeneration of cofactor, resulting in an overall hydroxyl product yield of 88.7% (99.5% ee). This new enzymatic transformation may offer a practical method for the preparation of this important chiral building block.

Full text of DOI:10.1016/j.tetasy.2009.10.036

Tetrahedron: Asymmetry 20 (2009) 2504–2509
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Enantioselective synthesis of ethyl (S)-2-hydroxy-4-phenylbutyrate
by recombinant diketoreductase
*
Xuri Wu, Yuchen Wang, Jianming Ju, Chen Chen, Nan Liu, Yijun Chen
Laboratory of Chemical Biology, School of Life Science and Technology, China Pharmaceutical University, 24 Tongjia Street, Nanjing,
Jiangsu Province 210009, 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:
Received 14 August 2009
Accepted 26 October 2009
Recombinant diketoreductase showed excellent stereoselectivity in the double reduction of b,d-diketo
esters. To investigate the substrate specificity and to broaden the applications of this new biocatalyst,
a number of ketone substrates were used to evaluate the substrate spectrum and enantioselectivity of
this enzyme in the present study. Among the ketone substrates tested, only this enzyme displayed high
efficiency and excellent enantioselectivity in the reduction of ethyl 2-oxo-4-phenylbutyrate to ethyl
(S)-2-hydroxy-4-phenylbutyrate. After optimizing the reaction conditions, the bio-reduction of ethyl 2-
oxo-4-phenylbutyrate at a substrate concentration of 0.8 M (164.8 g/L) was achieved by the recombinant
diketoreductase in an aqueous-toluene biphasic system coupled with formate dehydrogenase for the
regeneration of cofactor, resulting in an overall hydroxyl product yield of 88.7% (99.5% ee). This new
enzymatic transformation may offer a practical method for the preparation of this important chiral build-
ing block.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Recently, a diketoreductase was cloned from Acinetobacter bay-
lyi ATCC 33305, and this novel and unique enzyme was heteroge-
The stereoselective reduction of prochiral ketones is a very
useful approach in organic synthesis for producing chiral interme-
diates that typically are synthetic building blocks for enantiomeri-
cally pure pharmaceuticals, food and agrochemicals.1–3 For
instance, enantiopure 2-hydroxy-4-phenylbutanoic acid and its es-
ter have been utilized for the asymmetric synthesis of a wide vari-
ety of bioactive molecules, such as angiotensin-converting enzyme
(ACE) inhibitors.4,5 Recently, significant progress has been made in
the synthesis of these chiral intermediates by biocatalytic ap-
proaches.6,7 One of the widely used methods is the kinetic resolu-
tion of racemic alcohols.8,9 A lipase from Pseudomonas cepacia was
employed to enantioselectively hydrolyze racemic ethyl 2-hydro-
xy-4-phenylbutyrate to give 33% yield of the (S)-alcohol.10 Since
only a 50% theoretical yield could be achieved by kinetic resolu-
tion, the bio-reduction of prochiral ketones to chiral alcohols
exhibits obvious advantages with a 100% theoretical yield after
completing the reactions. For example, a dehydrogenase from
Enterobacter sp. BK2 K was able to reduce 2-oxo-4-phenylbutanoic
acid to its S-hydroxyl product with a conversion of 96% and ee va-
lue of 94%,11 although the relatively low substrate loading (0.1 M)
and moderate enantioselectivity of this reaction do limit its poten-
tial as a practical biocatalyst.
neously expressed in Escherichia coli.12,13 The recombinant
diketoreductase (rDKR) could stereoselectively reduce ethyl-6-
(benzyloxy)-3,5-dioxohexanoate to ethyl (3R,5S)-6-(benzyloxy)-
3,5-dihydroxy-hexanoate, a valuable intermediate in the synthesis
of statin drugs. Although the bio-reduction of the diketo ester by
rDKR has been extensively studied, its substrate specificity to-
wards other ketone substrates, especially mono-keto compounds,
remains unclear. Herein, we conducted a survey on the substrate
specificity of rDKR against different ketones. Among the ketone
substrates examined, rDKR was found to be the only one capable
of efficiently converting ethyl 2-oxo-4-phenylbutyrate 1 to ethyl
(S)-2-hydroxy-4-phenylbutyrate 11a with high conversion rate
and excellent enantioselectivity (Scheme 1). Consequently, an
aqueous-toluene biphasic system was established to increase the
substrate concentration to a practical level (0.8 M), which provides
another useful application of the diketoreductase.
2. Results and discussion
2.1. Reductions of ketone substrates with rDKR
Various ketones with carbonyl group(s) in the side chains
(Fig. 1) were examined with rDKR for the conversion and stereose-
lectivity. The reaction was coupled with formate dehydrogenase
(FDH) for NADH regeneration,14 while reverse phase high perfor-
* Corresponding author. Tel.: +86 25 83271045; fax: +86 25 83271249.
E-mail address: yjchen@cpu.edu.cn (Y. Chen).
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.10.036

X. Wu et al. / Tetrahedron: Asymmetry 20 (2009) 2504–2509
2505
OH
O
O
rDKR
O
O
O
ee = 99.5%
NAD+
NADH
HCOOH
CO2
FDH
Scheme 1. The reduction of 1 catalyzed by rDKR. Reduction of 1 catalyzed by rDKR gave 11a with ee value of 99.5%. FDH was used for the regeneration of NADH.
O
OH
O
OH
O
O
O
O
O
O
O
O
2
(S)-11a
OH
(R)-11b
1
OH
O
OH
O
O
O
O
O
O
O
O
O
(S)-13a
3
(R)-12b
(S)-12a
OH
O
O
O
OH
O
O
OH
O
O
O
(S)-14b
(R)-14a
4
(R)-13b
O
O
O
O
OH OH
O
O
OH OH
O
O
O
O
O
O
O
5
(3S,5S)-15a
(3R,5R)-15b
OH OH
OH OH
O
O
O
O
O
O
Cl
O
O
(3S,5R)-15c
OH
6
(3R,5S)-15d
OH
OH
OH
O
O
(S)-17b
(S)-16b
(R)-17a
(R)-16a
8
7
OH
OH
O
OH
OH
O
N
N
N
(S)-19b
(R)-18a
Figure 1. Structures of ketones and their hydroxyl products.
(R)-19a
9
(S)-18b
10
mance liquid chromatography (HPLC) and chiral HPLC were used
to assay the conversion and stereoselectivity, respectively.
As shown in Table 1, rDKR showed better conversion in the
acetophenone 7, propiophenone 8 and benzylacetone 9, were
served as substrates, the length of the side chain from the phenyl
ring significantly affected the bioconversion by rDKR, in which
the conversion rates varied from 10.6% to 88.4%. The bioconversion
could be considerably improved by increasing the length of the
side chain. Furthermore, the bio-reduction was completely dimin-
ished by replacement of the phenyl ring by pyridine in the mono-
reduction of
a-keto esters than that of b-keto esters. The conver-
sions of 1, ethyl benzoylformate 2 and methyl benzoylformate 3
were greater than 99%, whereas the conversion of ethyl 3-oxo-3-
phenylpropanoate 4 was only 46.1%. When aryl ketones, such as

2506
X. Wu et al. / Tetrahedron: Asymmetry 20 (2009) 2504–2509
Table 1
As shown in Figure 2, both rDKR and FDH could be well toler-
Reduction of ketones catalyzed by rDKR
ated against a variety of organic solvents. In the presence of 20%
(v/v) of organic solvents, such as ethanol, DMSO and toluene, the
reaction system consisting of rDKR and FDH was stable and per-
formed well. However, the conversion rate decreased dramatically
when either ethanol or toluene was used at a higher concentration
for the reaction, whereas DMSO could reach values of up to 40% (v/
v) in the reaction mixture without a negative impact (Fig. 2). Com-
pared to these solvents, n-butanol, tetrahydrofuran, ethyl acetate
and n-hexane did not show an improvement in the conversion.
The enantioselectivity remained unchanged in all cases. Since the
practical utility of many enzymes, such as alcohol dehydrogenases
from Pseudomonas fluorescens16,17 and from Rhodococcus erythropo-
lis,15 is restricted by their poor tolerance against organic solvents,
our results indicated that the complement between rDKR/FDH
and organic solvents could be particularly useful for the synthesis
of chiral alcohols.
Substrate
Conversiona (%)
eeb (%)
deb (%)
1
2
3
4
5
6
7
8
9
99.5
99.1
99.0
46.1
51.7
0
10.6
43.8
88.4
0
99.5 (S)
31.9 (S)
73.3 (S)
99.5 (S)
98.8 (3R,5S)
0
30.7 (R)
0.4 (R)
36.7 (R)
0
N/Ac
N/A
N/A
N/A
86.5 (3R,5S)
0
N/A
N/A
N/A
N/A
10
All the reactions coupled with formate dehydrogenase (FDH) for NADH regenera-
tion were performed at room temperature for 18 h.
Conversion of ketones was determined by achiral HPLC.
The ee and de values of hydroxyl products were determined by Chiralcel OD-RH
column.
a
b
c
Not applicable.
2.3. Reduction of 1 at high substrate concentration
ketones and by the substitution of benzyloxy group by chlorine in
t-butyl 3,5-diketo-6-benzyloxyhexanoate 5, suggesting that the
phenyl ring might be essential for the binding between rDKR and
substrate. Meanwhile, rDKR generally displayed poor enantioselec-
tivity in the reduction of aryl ketones at a concentration of 2 mg/
ml. However, when compounds 1 and 4 were used as substrates,
excellent enantioselectivities were observed with different conver-
sion rates (Table 1). The complete conversion (99.5%) and excellent
enantioselectivity (99.5% ee) for the transformation of 1 to 11a
prompted us to investigate this enzymatic reaction further as a
valuable biocatalytic application.
According to the results of organic solvent tolerance, ethanol,
DMSO and toluene were selected as co-solvents for further inves-
tigation. The reactions were carried out at different concentrations
of 1 with 20% of ethanol/toluene or 40% of DMSO. The conversion
rates and ee values were examined by HPLC and chiral HPLC. When
ethanol was used as a co-solvent, the conversion was only 47.7% at
a concentration of 0.05 M of 1 (Fig. 3). Meanwhile, with 20% (v/v)
toluene as a co-solvent the conversion was 99.1% at the same sub-
strate concentration. In the case of 40% (v/v) DMSO, the conversion
reached 99.6%, and a similar conversion rate was obtained even at
1.2 M of substrate concentration (Fig. 3). Although DMSO showed
excellent properties with respect to high substrate concentration,
the safety issues and high boiling point of this solvent prohibited
its practical use as a co-solvent for industrial applications. There-
fore, toluene was ultimately the best choice among the solvents
examined. Since toluene is not water miscible, a biphasic system
is necessary for the bio-reduction. When whole cells containing
rDKR were examined with the same biphasic systems using glu-
cose as a co-substrate for cofactor regeneration, the bio-reduction
at a substrate concentration of 10 mg/ml (0.049 M) was only 3.5%
after 18 h. On the other hand, evaluation of a biphasic system con-
sisting of an aqueous buffer solution and 20% toluene resulted in
95.2% conversion at substrate concentration of 0.8 M (164.8 g/L)
after 9 h (Fig. 4), which shows higher efficiency and productivity
than previous reports on (S)-alcohol dehydrogenase from R. ery-
2.2. Optimization of the reaction media for the reduction of 1
A common obstacle of enzymatic conversion on an industrial
scale is the low solubility of hydrophobic substrates in an aqueous
media. As a result, enzymatic reactions are usually carried out at a
low substrate concentration (5–10 mM), which leads to low effi-
ciency and productivity. One strategy to overcome such problems
is to identify an appropriate organic solvent as a co-solvent, in
which the hydrophobic substrates can be dissolved without inacti-
vating the enzyme.15 Therefore, in order to increase the substrate
concentration for practical utility, various organic solvents were
screened to assess the tolerance of the reaction system consisting
of rDKR and FDH in the reduction of 1. The conversion and enanti-
oselectivity were determined by HPLC.
100
80
60
40
20
0
1
2
3
4
5
6
7
Organic solvents
10% (v/v)
40% (v/v)
20% (v/v)
30% (v/v)
50% (v/v)
Figure 2. Reduction of 1 in different organic solvents. (1) Ethanol; (2) n-butanol; (3) THF; (4) ethyl acetate; (5) toluene; (6) n-hexane and (7) DMSO.

X. Wu et al. / Tetrahedron: Asymmetry 20 (2009) 2504–2509
2507
100
80
60
40
20
0
1
2
3
4
5
6
7
8
9
Substrate concentration (M)
20% ethanol (v/v)
20% toluene (v/v)
40% DMSO (v/v)
Figure 3. Reduction of 1 at different substrate concentrations. The reduction of 1 at different substrate concentrations was conducted with 20% ethanol, 20% toluene or 40%
DMSO as a co-solvent, respectively. Substrate concentrations: (1) 0.05 M; (2) 0.1 M; (3) 0.2 M; (4) 0.4 M; (5) 0.6 M; (6) 0.8 M; (7) 1.0 M; (8) 1.2 M and (9) 1.5 M.
of 1. Coupling with FDH for NADH regeneration, rDKR can effi-
ciently convert 1 to its hydroxyl product 11a in an aqueous-
toluene biphasic system. More importantly, the substrate concen-
100
80
60
40
20
0
tration of
1 in this system was demonstrated to be 0.8 M
(164.8 g/L). A practical and simple method for the preparation of
11a has been developed and optimized, which further broadens
the biocatalytic applications of rDKR.
4. Experimental
4.1. Materials and instruments
NAD+ and IPTG were purchased from Sigma (USA). HPLC-grade
acetonitrile was obtained from TEDIA (USA). All ketone substrates
and alcohol standards were purchased from Sigma–Aldrich Chem-
ical Co. (USA), ChemPacific (USA) or Nanjing Chemlin Chemical
Industry Co., Ltd (China). t-Butyl 3,5-diketo-6- benzyloxyhexano-
ate 5 and t-butyl 3,5-dihydroxy-6-benzyloxyhexanoate 15 were
prepared according to the procedures described in the literature.19
The HPLC analyses were performed on a Shimadzu 2010A HT with
UV detector. Enantiomeric excess (ee) was determined by chiral
HPLC with a Chiralcel OD-RH column (Daicel Instruments, Ltd,
USA). FDH was expressed and prepared as described by Shuber
et al.20
0
1
2
3
4
5
6
7
8
9
10 11
Time (h)
Figure 4. Time courses of the reduction of 1 by rDKR at substrate concentration of
0.8 M. Substrate 1 at 0.8 M (164.8 g/L) was transformed by rDKR coupled with FDH
for NADH regeneration. All data represent the averages of three experiments. Open
diamond; closed diamond: 11a.
thropolis for the reduction of p-chloroacetophenone (0.2 M) in the
aqueous-heptane (4:1) biphasic media.15,18 Hence, a 50 ml pre-
parative scale with substrate 1 concentration of 0.8 M (164.8 g/L)
was performed to confirm this biocatalytic method at this scale.
As expected, a conversion rate of 91.8% and 99.5% ee value were
obtained after 9 h. After extraction and purification, the desired hy-
droxyl product 11a was obtained with an isolated yield of 88.7%.
Thus, the aqueous-toluene biphasic system consisting of rDKR
and FDH could be used as a practical and efficient method for
the preparation of 11a from 1.
4.2. Expression and enzyme activity of rDKR
rDKR was over-expressed in E. coli using LB medium at 15 °C,
220 rpm for 14 h after induction by 50 lM IPTG as previously de-
scribed.12,13 After expression, the cells were collected by centrifu-
gation. Next, recombinant E. coli cells (5.0 g) were resuspended in
25 ml of 50 mM potassium phosphate buffer (pH 6.0). The cell sus-
pensions were disrupted by passing through a high-pressure cell
press at 1000 psi. Cell-free extract was obtained by centrifuging
the mixture at 13,500g for 30 min. The absorbance change at
340 nm for the oxidation of NADH was used to determine the
activity of rDKR with ethyl-6-(benzyloxy)-3, 5-dioxohexanoate as
a substrate by spectrophotometric methods at 40 °C. Assay mixture
contained 0.15 mM NADH, 0.25 mM ethyl-6-(benzyloxy)-3,5-
3. Conclusions
Previous studies have shown that rDKR was able to reduce a
diketo ester substrate with excellent stereoselectivity. Herein, the
substrate specificity of rDKR on a number of ketones was investi-
gated. An aryl mono-ketone 1 was found to be a suitable substrate
for this enzyme. After evaluating different organic solvents, tolu-
ene was identified for its usefulness in the biocatalytic reduction
dioxohexanoate, 5–50 lg DKR and 0.1 M potassium phosphate
buffer (pH 6.0) in a final volume of 1.0 ml. For all assays, enzyme
activity was defined as one unit representing the oxidation of
1 lmol of NADH per minute per milligram protein. The enzyme

2508
X. Wu et al. / Tetrahedron: Asymmetry 20 (2009) 2504–2509
activity of rDKR was approximately 5
lmol/min/mg with a protein
4.5. Chiral HPLC
concentration of 10.5 mg/ml, which was determined by Bradford
method.
Chiral HPLC methods were performed on a Chiralcel OD-RH col-
umn (5
l and a flow rate of 0.5 ml/min.
Method 1 was used for the chiral analysis of 11 with a gradient
l
m, 150 Â 4.6 mm) at 25 °C with an injection volume of
5
l
4.3. Achiral HPLC
of 30–40% B in 25 min, and then 40% B was kept for additional
5 min. Retention times of 11a and 11b were 21.1 min and
23.6 min, respectively.
Method 2 was used for the chiral analysis of 12 with a gradient
of 15–30% B in 25 min, and then 30% B was kept for additional
5 min. Retention times of 12a and 12b were 23.4 min and
26.7 min, respectively.
Method 3 was used for the chiral analysis of 13 with a gradient
of 15–25% B in 25 min, and then 25% B was kept for additional
5 min. Retention times of 13a and 13b were 17.5 min and
20.3 min, respectively.
Method 4 was used for the chiral analysis of 14 with a gradient
of 25–30% B in 25 min, and then 30% B was kept for additional
5 min. Retention times of 14a and 14b were 17.9 min and
20.9 min, respectively.
Method 5 was used for the chiral analysis of the reduction prod-
ucts of 15 with a gradient of 32–40% B in 25 min, and then the gra-
dient of 40% B was kept for additional 5 min. Retention times of
dihydroxy ester stereoisomers of 15a, 15b, 15c and 15d were
18.4 min, 22.6 min, 24.6 min and 27.2 min, respectively.
Method 6 was used for the chiral analysis of 16 with a gradient
of 15–30% B in 25 min, and then 30% B was kept for additional
5 min. Retention times of 16a and 16b were 21.1 min and
22.9 min, respectively.
Method 7 was used for the chiral analysis of 17 with a gradient
of 32–40% B in 25 min, and then 40% B was kept for additional
5 min. Retention times of 17a and 17b were 13.7 min and
15.1 min, respectively.
Achiral HPLC methods were performed with various gradients
of mobile phase A (0.1% trifluoroacetic acid in water) and mobile
phase B (0.1% trifluoroacetic acid in acetonitrile) at 40 °C with UV
detection at 220 nm. The analyses were performed on a Waters
ODS column (5
10 l and a flow rate of 1 ml/min.
Method 1 was used to determine the conversion of 1 with a gra-
l
m, 150 Â 4.6 mm) with an injection volume of
l
dient from 15% to 80% B in 14 min. The retention times of racemic
11 and 1 were 10.4 min and 12.2 min.
Method 2 was used to determine the conversion of 2 with a gra-
dient from 10% to 80% B in 14 min. The retention times of racemic
ethyl mandelate 12 and 2 were 8.7 min and 11.8 min.
Method 3 was used to determine the conversion of 3 with a gra-
dient from 10% to 80% B in 14 min. The retention times of racemic
methyl mandelate 13 and 3 were 7.2 min and 10.6 min.
Method 4 was used to determine the conversion of 4 with a gra-
dient from 15% to 85% B in 14 min. The retention times of racemic
ethyl-3-hydroxy-3-phenylpropionate 14 and 4 were 8.4 min and
9.8 min.
Method 5 was used to determine the conversion of 5 with a gra-
dient from 30% to 90% B in 14 min. Retention times of racemic t-
butyl anti-3, 5-dihydroxy-6-benzyloxyhexanoate, racemic t-butyl
syn-3, 5-dihydroxy-6-benzyloxyhexanoate and 5 were 7.9 min,
8.3 min and 11.3 min, respectively.
Method 6 was used to determine the conversion of 7 with a gra-
dient from 15% to 65% B in 13 min. The retention times of racemic
1-phenylethanol 16 and 7 were 10.9 min and 12.8 min.
Method 7 was used to determine the conversion of 8 with a gra-
dient from 20% to 80% B in 13 min, and then 80% B was kept for
additional 3 min. The retention times of racemic 1-phenyl-1-pro-
panol 17 and 8 were 11.2 min and 13.3 min.
Method 8 was used for the chiral analysis of 18 with a gradient
of 32–40% B in 25 min, and then 40% B was kept for additional
5 min. Retention times of 18a and 18b were 17.5 min and
19.2 min, respectively.
Method 8 was used to determine the conversion of 9 with a gra-
dient from 20% to 65% B in 13 min. Then, the gradient was in-
creased to 90% B and kept for additional 3 min. The retention
times of racemic 4-phenyl-2-butanol 18 and 9 were 13.4 min and
14.8 min.
Method 9 was used to determine the conversion of 2-acetyl-
pyridin 10 with a gradient from 5% to 50% B in 14 min. The reten-
tion times of racemic 1-(2-pyridyl) ethanol 19 and 10 were 2.9 min
and 6.8 min.
4.6. Ketone reduction catalyzed by rDKR
The enzymatic reduction of aryl ketones was performed using
rDKR and FDH for NADH regeneration. 1 ml of reaction mixture
contained different ketones (2 mg/ml), 5% ethanol (v/v), NAD+
(0.5 mM), FDH (4 U/ml), sodium formate (0.2 M), rDKR (5 U/ml)
and 0.1 M potassium phosphate buffer (pH 6.0). The reaction mix-
tures were incubated at room temperature, 200 rpm for 18 h. Sub-
sequently, all reactions were terminated by the addition of 1 ml
ethanol. After vortexing and centrifugation, the resulting superna-
tants were subjected to HPLC analysis to determine the conversion
and enantioselectivity. The absolute configurations of alcohols
were confirmed by chiral HPLC with authentic samples.
4.4. Analysis of the reduction of t-butyl 6-chloro-3,5-
dioxohexanoate
The HPLC analysis of the reduction of t-butyl 6-chloro-3,5-
dioxohexanoate 6 was performed on a Waters LC/MS system that
consisted of a quaternary pump, an in-line degasser, an autosam-
pler, a column thermostat, and an Micromass Quattro MicroTM
series model Ion Trap SL mass spectrometer detector electrospray
positive ionization. The chromatographic separation of the reac-
tion mixture of 6 was conducted on a Shimadzu ODS column
4.7. Reduction of 1 in aqueous organic solvent mixtures
The reduction of 1 in an organic solvent was conducted in the
presence of various concentrations of organic solvents ranging
from 10% to 50% (v/v). Reaction mixture (1 ml) contained 1
(2 mg/ml), NAD+ (0.5 mM), FDH (4 U/ml), sodium formate
(0.2 M), rDKR (5 U/ml) and 0.1 M potassium phosphate buffer
(pH 6.0). The reaction mixtures were incubated at room tempera-
ture, 200 rpm for 18 h and were terminated by the addition of eth-
anol (1 ml). The resulting supernatants were subjected to HPLC
analyses to determine the conversion and enantioselectivity.
(5
10
l
m, 250 Â 4.6 mm) at 40 °C with an injection volume of
ll and a flow rate of 1 ml/min. Elution was achieved with a
gradient of 15–90% B (0.1% trifluoroacetic acid in acetonitrile)
in 14 min and kept at 90% B for additional 3 min. No hydroxyl
products from the reduction of 6 catalyzed by rDKR were de-
tected, except for m/z 235 of 6 (M+H) at 8.6 min.

X. Wu et al. / Tetrahedron: Asymmetry 20 (2009) 2504–2509
2509
4.8. Reduction of 1 at different substrate concentrations
and dried over Na2SO4. After filtration and evaporation under re-
duced pressure a yellow oily residue was obtained. Chromatogra-
phy on silica gel (ethyl acetate/petroleum ether 5:95 v/v) gave
7.3 g 11a (88.7% isolated yield) as a colorless oil.
Substrate concentrations from 0.05 to 1.5 M were used in the
reaction media consisting of aqueous buffer solution and 20%
ethanol/toluene or 40% DMSO. The reaction mixture (1 ml) con-
tained different concentrations of 1, sodium formate (2 M), NAD+
(0.5 mM), FDH (4 U/ml), organic solvent (20% or 40% v/v), rDKR
(5 U/ml), 0.1 M potassium phosphate buffer (pH 6.0). The reactions
were performed at room temperature, 200 rpm for 18 h. The con-
version rate and enantiomeric purity were analyzed by HPLC.
Acknowledgements
We are grateful for the technical assistance from Ms. Lili Wang.
This work was supported by the ‘111 Project’ from the Ministry of
Education of China and the State Administration of Foreign Expert
Affairs of China (No. 111-2-07) and the Key Project of R&D from the
Ministry of Education of China (No. 108070).
4.9. Reduction of 1 by whole cells containing rDKR
Recombinant E. coli cells (0.2 g) containing rDKR were sus-
pended in 1 ml of 0.1 M phosphate buffer (pH 6.0). Glucose
(75 mg) was then added along with 10 mg of 1. The bioconversion
of 1 was conducted by shaking at 200 rpm at room temperature for
18 h and was terminated by the addition of 1 ml ethanol. After
vortexing and centrifugation, the resulting supernatants were fil-
tered and analyzed by HPLC to determine the conversion and
enantioselectivity.
References
1. Patel, R. N. Curr. Org. Chem. 2006, 10, 1289–1321.
2. Patel, R. N. Coord. Chem. Rev. 2008, 252, 659–701.
3. Müller, M. Angew. Chem., Int. Ed. 2005, 44, 362–365.
4. Lin, W. Q.; He, Z.; Jing, Y.; Cui, X.; Liu, H.; Mi, A. Q. Tetrahedron: Asymmetry 2001,
12, 1583–1587.
5. Dao, D. H.; Kawai, Y.; Hida, K.; Hornes, S.; Nakamura, K.; Ohno, A.; Okamura,
M.; Akasaka, T. Bull. Chem. Soc. Jpn. 1998, 71, 425–432.
6. Chadha, A.; Baskar, B. Tetrahedron: Asymmetry 2002, 13, 1461–1464.
7. Oda, S.; Inada, Y.; Kobayashi, A.; Ohta, H. Biosci., Biotechnol., Biochem. 1998, 62,
1762–1767.
4.10. Time course for the reduction of 1 in the biphasic system
8. Kalaritis, P.; Regenye, R. W.; Partridge, J. J.; Coffen, D. L. J. Org. Chem. 1990, 55,
812–815.
9. Huang, S.-H.; Tsai, S.-W. J. Mol. Catal. B: Enzym. 2004, 28, 65–69.
10. Liese, A.; Kragl, U.; Kierkels, H.; Schulze, B. Enzyme Microb. Technol. 2002, 30,
673–681.
11. Cha, M.; Kim, E. J.; Yun, H.; Cho, B.-K.; Kim, B.-G. Biotechnol. Prog. 2007, 23, 606–
612.
12. Wu, X. R.; Liu, N.; He, Y. M.; Chen, Y. J. Nat. Proc., 2008. http://hdl.handle.net/
10101/npre.2008.1697.1.
13. Wu, X. R.; Liu, N.; He, Y. M.; Chen, Y. J. Acta Biochem. Biophys. Sinica 2009, 41,
163–170.
The reduction of 1 were performed at a concentration of 0.8 M
in a system containing sodium formate (2 M), NAD+ (0.5 mM), FDH
(4 U/ml), DKR (5 U/ml), toluene (20%, v/v) and 0.1 M potassium
phosphate buffer (pH 6.0). The reaction mixtures were incubated
at room temperature at 200 rpm. Samples were taken at 1 h inter-
vals for 11 h and were terminated by the addition of ethanol. After
centrifugation, the resulting supernatants were subjected to HPLC
analyses.
14. Geertman, J. M.; van Dijken, J. P.; Pronk, J. T. FEMS Yeast Res. 2006, 6, 1193–
1203.
15. Gröger, H.; Hummel, W.; Buchholz, S.; Drauz, K.; Nguyen, T.; Rollmann, C.;
Hüsken, H.; Abokitse, K. Org. Lett. 2003, 5, 173–176.
16. Hildebrandt, P.; Riermeier, T.; Altenbuchnerc, J.; Bornscheuera, U. T.
Tetrahedron: Asymmetry 2001, 12, 1207–1210.
17. Zhu, D. M.; Malik, H. T.; Hua, L. Tetrahedron: Asymmetry 2006, 17, 3010–3014.
18. Gröger, H.; Hummel, W.; Rollmann, C.; Chamouleau, F.; Hüsken, H.; Werner,
H.; Wunderlich, C.; Abokitse, K.; Drauz, K.; Buchholz, S. Tetrahedron 2004, 60,
633–640.
19. Guo, Z. W.; Chen, Y. J.; Goswami, A.; Hanson, R. L.; Patel, R. N. Tetrahedron:
Asymmetry 2006, 17, 1589–1602.
20. Shuber, A. P.; Orr, E. C.; Recny, M. A.; Schendel, P. F.; May, H. D.; Schauer, N. L.;
Ferry, J. G. J. Biol. Chem. 1986, 261, 12942–12947.
4.11. Reduction of 1 at a preparative scale
A 50 ml reaction mixture contained substrate 1 (8.25 g), HCO-
ONaÁ2H2O (10.4 g), NAD+ (16.6 mg), FDH (200 U), toluene (10 ml),
DKR (250 U) and 0.1 M potassium phosphate buffer (pH 6.0). The
reaction mixture was incubated at room temperature, at 200 rpm
for 9 h. Then, the toluene layer was separated from the aqueous
phase, and the aqueous phase was extracted with ethyl acetate
for three times (3 Â 50 ml). The organic phases were combined