Asymmetric reduction of ketones using recombinant E. coli cells that produce a versatile carbonyl reductase with high enantioselectivity and broad substrate specificity

Article abstract of DOI:10.1016/j.tet.2006.04.061

The gene encoding a versatile biocatalyst that shows high enantioselectivity for a variety of ketones, SCR (Saccharomyces cerevisiae carbonyl reductase), has been identified, cloned, and expressed in Escherichia coli. Two types of expression systems with high NADPH-regenerating capacities have been constructed. One is the tandem system, where the genes encoding SCR and GDH (glucose dehydrogenase) are located in the same plasmid, and the other is the two-plasmid system, where each of the SCR and GDH genes is located in separate plasmids that can coexist in one E. coli cell. Asymmetric reduction of ketones with the recombinant E. coli cells gave synthetically useful 20 alcohols, 11 of which were enantiomerically pure. The productivity of one of these products was as high as 41 g/L.

Full text of DOI:10.1016/j.tet.2006.04.061

Tetrahedron 62 (2006) 6143–6149
Asymmetric reduction of ketones using recombinant E. coli cells
that produce a versatile carbonyl reductase with high
enantioselectivity and broad substrate specificity
*
Tadashi Ema, Hideo Yagasaki, Nobuyasu Okita, Masahiro Takeda and Takashi Sakai
*
Division of Chemistry and Biochemistry, Graduate School of Natural Science and Technology, Okayama University,
Tsushima, Okayama 700-8530, Japan
Received 3 April 2006; revised 19 April 2006; accepted 21 April 2006
Available online 6 May 2006
Abstract—The gene encoding a versatile biocatalyst that shows high enantioselectivity for a variety of ketones, SCR (Saccharomyces
cerevisiae carbonyl reductase), has been identified, cloned, and expressed in Escherichia coli. Two types of expression systems with
high NADPH-regenerating capacities have been constructed. One is the tandem system, where the genes encoding SCR and GDH (glucose
dehydrogenase) are located in the same plasmid, and the other is the two-plasmid system, where each of the SCR and GDH genes is located
in separate plasmids that can coexist in one E. coli cell. Asymmetric reduction of ketones with the recombinant E. coli cells gave synthetically
useful 20 alcohols, 11 of which were enantiomerically pure. The productivity of one of these products was as high as 41 g/L.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
can be overcome by constructing an SCR gene expression
system. The synthetic power of recombinant carbonyl reduc-
tases from various origins has recently been demonstrated.5–13
In particular, a number of recombinant whole-cell reductions
with high efficiency have been reported.6–8 Recently, we
have identified and cloned the gene coding for SCR, and
have investigated the capabilities of the recombinant SCR
by conducting both enzymatic and whole-cell reductions
of various ketones.4c Here we report asymmetric reduction
of various ketones with recombinant Escherichia coli ex-
pressing the gene encoding SCR, demonstrating the utility
and power of the versatile biocatalyst.
Asymmetric synthesis with biocatalysts has attracted much
attention from the viewpoint of green chemistry.1–2
Among them, versatile biocatalysts capable of showing
high enantioselectivity and broad substrate specificity simul-
taneously have found broad utility because this characteris-
tic feature is useful for asymmetric synthesis of various
compounds needed in a laboratory as well as a quite impor-
tant compound.3 The discovery and development of a new
versatile biocatalyst with high efficiency is therefore an im-
portant subject, contributing to the significant enlargement
of the library of synthetically useful biocatalysts.
Carbonyl reductases from various microorganisms have
been used to prepare optically active alcohols from carbonyl
compounds.2 A Saccharomyces cerevisiae (bakers’ yeast)
carbonyl reductase (SCR) previously reported by us showed
catalytic activity for various ketones, such as a-chloro
ketones, a-acetoxy ketones, a-keto esters, b-keto esters,
g-keto esters, and b-diketones, and 13 out of 20 alcohols
obtained had enantiomeric purities of >98% ee.4a,b Despite
the potential of this enzyme as a versatile biocatalyst, its util-
ity was restricted by its low expression level in S. cerevisiae
and the laborious purification procedure. These drawbacks
2. Results
2.1. Identification, cloning, and expression of the gene
The amino-acid sequence analysis of purified SCR sug-
gested strongly that the gene encoding SCR is Gre2
(YOL151w). Stewart and co-workers have reported a library
of S. cerevisiae reductases, which includes Gre2.5 In this
study, the SCR gene was PCR-cloned from the genomic
DNA of an S. cerevisiae strain that we have used previously,
and an expression plasmid, pESCR, was constructed (Fig. 1).
As shown in Table 1, enzymatic activity (90 U per 1 g of
wet cells) of SCR heterologously expressed in E. coli
BL21(DE3) harboring pESCR was 800-fold higher than
that (0.11 U per 1 g of wet cells) of SCR in S. cerevisiae.
Keywords: Alcohol; Asymmetric synthesis; Biotransformation; Reduction.
* Corresponding authors. Tel.: +81 86 251 8091; fax: +81 86 251 8092;
e-mail: ema@cc.okayama-u.ac.jp
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.04.061

6144
T. Ema et al. / Tetrahedron 62 (2006) 6143–6149
Figure 2. SDS-PAGE of the cell lysates of E. coli BL21(DE3) transformed
with (A) pESCR, (B) pESCR+pABGD, (C) pACRGD, and (D) pABGD
after cultivation and induction.
pACRGD. Using E. coli BL21(DE3) harboring pACRGD,
a high level of production of SCR and GDH was confirmed
by SDS-PAGE (Fig. 2), and the SCR and GDH activities
were 60 and 15,000 U per 1 g of wet cells, respectively
(Table 1).
2.2. Asymmetric reduction of ketones with the
tandem system
The whole-cell reductions of various ketones were con-
ducted with the recombinant E. coli (tandem system). The
experimental procedure is described in Section 5.7, and
the results are summarized in Figure 3. Although 3 mmol
of ketone was typically used, the amount of a very reactive
one, ethyl pyruvate (7a), was increased to 6 mmol, and
that of some ketones with modest reactivity was decreased
to 1 mmol. Although the reaction did proceed without addi-
tion of NADP+, the addition of a catalytic amount
(0.4 mol %) of NADP+ increased the conversion (not opti-
mized). Figure 3 clearly shows that the inherent characteris-
tic features of SCR are retained in the recombinant SCR. The
absolute configurations of all the 16 alcohols 1b–16b ob-
tained with the recombinant E. coli were the same as those
obtained previously with SCR purified from S. cerevisiae.
The enantiomeric purities in the former are as high as those
in the latter in most cases. Both aliphatic and aromatic ke-
tones were successfully reduced. Not only enantioselectivity
but also regioselectivity for 2,4-hexanedione (16a) was com-
plete, with the less hindered carbonyl group being reduced
exclusively, as observed previously.4a,b
Figure 1. (A) Recombinant E. coli cell coexpressing SCR and GDH genes
to produce optically active alcohols. In this paper, E. coli harboring
pACRGD is called the tandem system, while that harboring pESCR and
pABGD is called the two-plasmid system. (B) Structures of expression
plasmids used in this study. SCR: SCR gene, GDH: GDH gene, AmpR:
ampicillin-resistance gene, CamR: chloramphenicol-resistance gene, PT7
:
T7 promoter, and TT7: T7 terminator.
Cofactor regeneration is essential for high productivity of
alcohols. Kataoka and co-workers have successfully
employed a glucose dehydrogenase (GDH) from Bacillus
megaterium to regenerate NADPH in vivo.6 We constructed
an expression plasmid having the SCR and GDH genes (tan-
dem system), pACRGD, as shown in Figure 1. pACYCDuet-
1 (Novagen) was selected as an expression vector because it
has two multiple cloning sites, each preceded by a T7 pro-
moter, which is suitable for coexpression of two genes.
The GDH gene was subcloned into pACYCDuet-1 to give
pABGD, into which the SCR gene was subcloned to give
As a control, E. coli BL21(DE3) harboring pACYCDuet-1
was used, and very low conversions (0% conversion in
many cases) were confirmed by H NMR in all cases. This
1
suggests that although E. coli produces its own enzymes,
its ketone-reducing and/or cofactor-regenerating capacity
for 1–6 mmol of ketone is very low. Furthermore, E. coli
BL21(DE3) strains harboring either pESCR or pABGD,
which lack either GDH or SCR gene, were also examined
for a-acetoxyacetophenone (5a) under the same reaction
conditions, which resulted in very low conversions (4%
and 0%). All these results indicate that SCR is responsible
for the reduction of all the ketones and that both SCR and
GDH are essential for the high-turnover catalysis.
Table 1. Enzymatic activity per 1 g of wet cells
Host
Plasmid
SCR GDH
activity (U) activity (U)
E. coli BL21(DE3) pESCR
E. coli BL21(DE3) pACRGD
90
60
—
15,000
5800
—
E. coli BL21(DE3) pESCR+pABGD 100
Bakers’ yeast 0.11
—

T. Ema et al. / Tetrahedron 62 (2006) 6143–6149
6145
Figure 3. Asymmetric reduction of ketones with SCR. The data are shown in the following order: isolated yield, enantiomeric purity, and absolute configuration.
The results obtained with recombinant E. coli (tandem system) and that (two-plasmid system) are shown in red and blue, respectively. The results obtained with
SCR purified from bakers’ yeast are taken from the lit.4b and are shown in black. Conditions for the tandem system: substrate (3.0 mmol for 1a–6a, 8a–9a, 11a,
and 15a–16a, 6.0 mmol for 7a, 1.0 mmol for 10a and 12a–14a), wet cells of E. coli BL21(DE3) harboring pACRGD (2 g), glucose (6.0 mmol), NADP+
(12 mmol), 0.1 M phosphate buffer (pH 7.0, 50 mL), 30 ꢀC. Conditions for the two-plasmid system: substrate (4.0 mmol for 17a, 1.0 mmol for 18a–20a),
wet cells of E. coli BL21(DE3) harboring pESCR and pABGD (2 g), glucose (2 equiv), NADP+ (12 mmol), 0.1 M phosphate buffer (pH 7.0, 10 mL), 30 ꢀC.
The pH of the reaction mixture was kept constant by adding 2 N NaOH. The experimental procedure is described in Section 5.7.
Mori and co-workers have demonstrated that (S)-3-hydroxy-
2,2-dimethylcyclohexanone ((S)-18b) can be used in the
total synthesis of more than 10 natural products.14 Based
on the stereochemical trend shown in Figure 3, we expected
that (S)-18b might be obtained by the asymmetric reduction
of 2,2-dimethylcyclohexane-1,3-dione (18a) using SCR. As
a result of the reduction with the recombinant E. coli (tan-
dem system), unfortunately, we detected only a trace amount
of product by 1H NMR.
system. Because the whole-cell reduction was found to be
faster in higher concentrations, the amount of buffer solution
was decreased from 50 mL to 10 mL (data not shown).
When the time courses of the reduction of a substrate with
moderate reactivity, 2,4-octanedione (17a), were compared
(Fig. 4), the two-plasmid system showed the best perfor-
mance, giving (S)-2-hydroxy-4-octanone ((S)-17b) with
complete enantioselectivity and regioselectivity in 71% iso-
lated yield, which corresponds to the relatively high produc-
tivity of 41 g/L. In contrast, E. coli transformants harboring
either pESCR or pABGD resulted in very low conversions
(Fig. 4), which indicates again that both SCR and GDH
are essential for the high-turnover catalysis.
2.3. Asymmetric reduction of ketones with the
two-plasmid system
In view of the very low conversion of 18a, we tried to im-
prove the reducing power by increasing the expression level
of SCR. We had two plasmids that can coexist in one E. coli
cell, pESCR and pABGD (Fig. 1). Because the copy number
of the former is higher than that of the latter, E. coli cells har-
boring the two plasmids (two-plasmid system) may produce
SCR more efficiently than that harboring pACRGD (tandem
system). As expected, enzymatic activity (Table 1) and SDS-
PAGE (Fig. 2) clearly indicate that SCR is produced more
efficiently in the two-plasmid system than in the tandem
Using the two-plasmid system, we conducted the reduction
of cyclic b-diketone 18a again, and found that the reaction
proceeded slowly. The product (S)-18b was isolated in
12% yield with 70% ee (Fig. 3). Despite the long reaction
time (24 h), the substrate 18a remained in the reaction mix-
ture (18% conversion). Using the two-plasmid system, we
further examined other substrates 19a and 20a, which con-
tain nitrogen and fluorine atoms, respectively, and obtained
optically pure alcohols (S)-19b and (S)-20b successfully
(Fig. 3).

6146
T. Ema et al. / Tetrahedron 62 (2006) 6143–6149
strain harboring pABGD exhibited 0% conversion because
of the absence of SCR.
The two-plasmid system showed a higher reducing power
than the tandem system did. Figure 4 shows that the initial
rate of the former is about 2-fold faster than that of the latter.
This is a result of the modulation of the production level of
SCR and GDH. Because glucose is the natural substrate for
GDH, GDH activity in the tandem system is sufficiently high
(Table 1). In contrast, because the ketones used in this study
are unnatural substrates for SCR, SCR activity is low in the
tandem system. Obviously, a lower amount of GDH and
a higher amount of SCR would be ideal. In the two-plasmid
system, the copy number of pESCR in the cell is higher than
that of pABGD. Indeed, SDS-PAGE (Fig. 2) indicated that
SCR was produced more efficiently in the two-plasmid sys-
tem than in the tandem system, and in the former case,
improved SCR activity (100 U per 1 g of wet cells) and
reduced GDH activity (5800 U per 1 g of wet cells) were
observed (Table 1). This modulation of the production level
of SCR and GDH led to the higher reducing power of the
two-plasmid system.
Figure 4. Time courses for reductions of 17a with E. coli BL21(DE3) trans-
formants harboring pACRGD (red), both pESCR and pABGD (blue),
pESCR (green), or pABGD (black). Conditions: 17a (4.0 mmol), wet cells
of E. coli BL21(DE3) transformant (2 g), glucose (8.0 mmol), NADP+
(12 mmol), 0.1 M phosphate buffer (pH 7.0, 10 mL), 30 ꢀC. The pH of the
reaction mixture was kept constant by adding 2 N NaOH. The reactions
were monitored by 1H NMR.
Some products were isolated in low yields with incomplete
enantiomeric purities (Fig. 3). For example, the useful syn-
thetic intermediate (S)-18b was isolated in 12% yield with
70% ee despite the use of the two-plasmid system. The struc-
tural alteration of SCR itself may be effective for further
improvement. For example, SCR can be used as a parent
enzyme for directed evolution or rational design approaches.
3. Discussion
Versatile biocatalysts capable of showing high enantioselec-
tivity and broad substrate specificity simultaneously are very
useful for asymmetric syntheses of various chiral com-
pounds. In this study, the gene encoding a versatile biocata-
lyst, SCR, has been identified, cloned, and expressed in
E. coli. Two types of systems coproducing SCR and GDH
(tandem and two-plasmid systems) have been constructed
(Fig. 1), and these recombinant E. coli cells afforded 20 op-
tically active alcohols, 11 of which had enantiomeric purities
of >98% ee (Fig. 3). Some of the obtained alcohols have
been used in the total synthesis of natural products and bio-
logically active compounds: e.g., (R)-11b for carnitine,15
(R)-12b for fluoxetine,16 (S)-13b for dihydrokawain,17 (S)-
14b for xestospongin A,18 and (S)-15b for pyrenophorin.19
(S)-18b is an extremely useful intermediate for the total syn-
thesis of various natural products such as polygodial, dihydro-
actinidiolide, O-methyl pisiferic acid, juvenile hormone III,
and (ꢁ)-K-76.14 SCR, which can produce these practical
alcohols, is a synthetically useful, versatile biocatalyst.
4. Conclusion
The gene encoding a versatile biocatalyst that shows high
enantioselectivity for a variety of ketones, SCR, has been
identified, cloned, and expressed in E. coli. Two types of ex-
pression systems with high NADPH-regenerating capacities
have been constructed. One is the tandem system, where the
genes encoding SCR and GDH are located in the same plas-
mid, and the other is the two-plasmid system, where each of
the SCR and GDH genes is located in separate plasmids that
can coexist in one E. coli cell. These recombinant E. coli
cells coproducing SCR and GDH are easy-to-use, syntheti-
cally useful biocatalysts to give a variety of optically active
alcohols, and 11 out of 20 alcohols obtained had enantio-
meric purities of >98% ee. When the two coexpression
systems were compared in terms of the conversion of 2,4-
octanedione, the two-plasmid system showed better perfor-
mance, giving (S)-2-hydroxy-4-octanone with complete
enantioselectivity and regioselectivity in 71% isolated yield,
which corresponds to the productivity of 41 g/L.
Coenzyme regeneration is essential for an efficient catalytic
reduction of ketones.6 NADPH is oxidized to NADP+ by the
SCR-catalyzed reduction of ketone, while NADP+ is re-
duced back to NADPH by the GDH-catalyzed oxidation of
glucose (Fig. 1). This catalytic cycle is realized only when
both the two enzymes are produced in vivo. Indeed, only
the tandem and two-plasmid systems, coexpressing the
two genes, gave the product in good yields (Fig. 4). In the
latter case, the relatively high productivity (41 g/L based
on the isolated yield) of (S)-17b was achieved; the produc-
tivity based on the conversion was 56 g/L. In contrast,
E. coli strain harboring pESCR showed much lower con-
versions, which suggests that NADP+ formed by the SCR-
catalyzed reaction was slowly reduced to NADPH by small
amount of enzymes originating from E. coli itself. E. coli
5. Experimental
The pressed bakers’ yeast was purchased from Oriental
Yeast, and SCR from bakers’ yeast was purified to homo-
geneity as described previously.4a Two internal amino-acid
sequences of SCR were determined at APRO Life Science
Institute, and the homology search was performed by
BLAST analysis. All the DNA manipulations and bacterial
transformations were carried out according to the standard

T. Ema et al. / Tetrahedron 62 (2006) 6143–6149
6147
protocols or manufacturers’ instructions, unless otherwise
stated.20 E. coli JM109 (Toyobo) and pGEM-T (Promega)
were routinely used as a host and a cloning vector,
respectively, and E. coli BL21(DE3) (Stratagene) was used
BL21(DE3) harboring pESCR was obtained by transforma-
tion of the competent cells.
5.2. Construction of a coexpression plasmid having SCR
and GDH genes, pACRGD
as
a host for pET-11a (Stratagene), pACYCDuet-1
(Novagen), and their derivatives. TaKaRa Ex Taq DNA
Polymerase was purchased from Takara Bio. Restriction en-
zymes were purchased from Toyobo, Stratagene, or Roche,
and T4 DNA ligase was purchased from Toyobo or New
England Biolabs. Agarose for electrophoresis was purchased
from Nacalai Tesque. QIAprep Spin Miniprep Kit,
QIAquick PCR Purification Kit, QIAquick Nucleotide
Removal Kit, and QIAquick Gel Extraction Kit (Qiagen)
were routinely used for DNA isolation and purification.
Synthetic oligonucleotides were obtained from Sigma
Genosys Japan. DNA sequencing was performed on an
ABI PRISM 310 Genetic Analyzer (Applied Biosystems)
at Advanced Science Research Center of Okayama
University, and then analyzed with MacVector 7.0 (Oxford
Molecular). All the genes cloned into the plasmids were
fully sequenced in both directions. NADPH and NADP+
were purchased from Oriental Yeast. The amount of proteins
was determined by the method of Bradford using BSA as
the standard.21 TLC was performed on Merck silica gel 60
F254, and silica gel column chromatography was performed
Genomic DNA of B. megaterium NBRC (formerly IFO)
15308 was isolated according to the lit.22 The glucose de-
hydrogenase (GDH) gene was amplified by PCR with the
primers BM-GD-1F (50-ATGTATACAGATTTAAAAGAT
AAAGTAGTT-30) and BM-GD-2R (50-TTAGCCTCTTC
CTGCTTGG-30).23 The conditions for the 100 mL PCR mix-
ture were as follows: 0.5 mM each primer, 0.2 mM each
dNTP, genomic DNA (80 ng), 5 U of TaKaRa Ex Taq
DNA Polymerase, 1.5 mM MgCl2, and 10 mL of PCR buffer.
PCR was done for 30 cycles of 98 ꢀC for 10 s and 50 ꢀC for
30 s followed by a final extension at 72 ꢀC for 5 min. The
amplified DNA fragment was directly cloned into pGEM-T
to yield pGBGD from transformed E. coli JM109 cells.
The GDH gene was subcloned into an expression vector,
pACYCDuet-1. The primers used for PCR are as follows:
BM-GD-10F (50-CCATGGCCGCGGGCATA-30) and BM-
GD-11R
(50-CGCGGTACCGATTTTAGCCTCTTC-30),
where the restriction site for KpnI is italicized. The condi-
tions for the 100 mL PCR mixture were as follows: 0.5 mM
each primer, 0.2 mM each dNTP, pGBGD (3.7 ng), 2.5 U
of TaKaRa Ex Taq DNA Polymerase, 1.5 mM MgCl2, and
10 mL of PCR buffer. PCR was done for 30 cycles of
98 ꢀC for 10 s and 48 ꢀC for 30 s followed by a final exten-
sion at 72 ꢀC for 5 min. The amplified DNA fragment was
digested with NdeI and KpnI, and then ligated into
pACYCDuet-1 that had been treated with the same restric-
tion enzymes. E. coli BL21(DE3) harboring pABGD was
obtained by transformation of the competent cells. The
SCR gene was then subcloned into pABGD. The primers
used for PCR are as follows: SC-CR-10F (50-CACGCCC
TTAATCATGAGTGTTTTTGTTTC-30) and SC-CR-6R (50-
CGGGATCCTTAAAGTTTATATTCTGCCCTC-30), where
the restriction sites for RcaI and BamHI are italicized. The
conditions for the 100 mL PCR mixture were as follows:
0.5 mM each primer, 0.2 mM each dNTP, pGSCR
(370 ng), 2.5 U of TaKaRa Ex Taq DNA Polymerase,
1.5 mM MgCl2, and 10 mL of PCR buffer. PCR was done
for 30 cycles of 98 ꢀC for 10 s and 55 ꢀC for 30 s followed
by a final extension at 72 ꢀC for 5 min. The amplified
DNA fragment was digested with RcaI and BamHI, and
then ligated into pABGD that had been treated with NcoI
and BamHI. E. coli BL21(DE3) harboring pACRGD was
obtained by transformation of the competent cells.
1
using Fuji Silysia BW-127 ZH (100–270 mesh). H NMR
spectra were measured in CDCl3 at 600, 500, 300, or
200 MHz.
5.1. Cloning of the SCR gene and construction of an
expression plasmid, pESCR
Two internal peptides of the purified SCR were sequenced:
AAWEFLEENR and DLLIPAVDGVK, both of which hit
Gre2 (YOL151w) of S. cerevisiae by BLAST analysis. The
SCR gene (GRE2, YOL151w) was amplified by PCR from
the genomic DNA of bakers’ yeast purchased from
Oriental Yeast. The primers used for PCR are as follows:
SC-CR-3F (50-CGATTTTCAAACAAACAGATAGCAG-30)
and
SC-CR-4R
(50-AAAATGCGCAGAGATGTACT
AGATG-30), which correspond to 50- or 30-noncoding
regions encompassing the gene. The conditions for the
100 mL PCR mixture were as follows: 0.5 mM each primer,
0.2 mM each dNTP, genomic DNA (400 ng), 5 U of
TaKaRa Ex Taq DNA Polymerase, 1.5 mM MgCl2, and
10 mL of PCR buffer. PCR was done for 30 cycles of
98 ꢀC for 10 s and 55 ꢀC for 30 s followed by a final exten-
sion at 72 ꢀC for 1 min. The amplified DNA fragment was
directly cloned into pGEM-T to yield pGSCR from trans-
formed E. coli JM109 cells. The SCR gene was subcloned
into an expression vector, pET-11a. The primers used for
PCR are as follows: SC-CR-5F (50-ATCACACGCCCTTA
CATATGTCAG-30) and SC-CR-6R (50-CGGGATCCTTAA
AGTTTATATTCTGCCCTC-30), where the restriction sites
for NdeI and BamHI are italicized. The conditions for the
100 mL PCR mixture were as follows: 0.5 mM each primer,
0.2 mM each dNTP, pGSCR (100 pg), 5 U of TaKaRa Ex
Taq DNA Polymerase, 1.5 mM MgCl2, and 10 mL of PCR
buffer. PCR was done for 30 cycles of 98 ꢀC for 10 s and
55 ꢀC for 30 s followed by a final extension at 72 ꢀC for
1 min. The amplified DNA fragment was digested with
NdeI and BamHI, and then ligated into pET-11a that had
been treated with the same restriction enzymes. E. coli
5.3. Preparation of E. coli harboring pESCR and
pABGD
E. coli BL21(DE3) harboring pABGD was transformed with
pESCR and grown on an LB plate containing ampicillin
(100 mg/mL) and chloramphenicol (34 mg/mL). The trans-
formation was confirmed by agarose gel electrophoresis
and enzymatic activity.
5.4. Sole expression of SCR gene
After E. coli BL21(DE3) cells harboring pESCR were grown
in LB medium (3 mL) containing ampicillin (100 mg/mL) at

6148
T. Ema et al. / Tetrahedron 62 (2006) 6143–6149
37 ꢀC for 16 h with shaking at 230 rpm, 2 mL of the culture
was transferred to the same medium (200 mL) in a 1-L
Erlenmeyer flask. The culture was shaken at 200 rpm at
37 ꢀC, and IPTG (0.1 mM) was added when OD600 reached
0.6–0.8. The cells were further incubated at 30 ꢀC for 16 h
with shaking at 200 rpm, harvested by centrifugation
(9000 rpm, 4 ꢀC, 5 min), and washed with 0.1 M phosphate
buffer (pH 7.0, 50 mL). The cell pellet was stored at ꢁ20 ꢀC
until it was used.
of the reaction mixture was kept constant by adding 2 N
NaOH. After an appropriate reaction time, the cells were
precipitated by centrifugation (10,000 rpm, 4 ꢀC, 10 min),
and solid NaCl was added. The products were extracted
with EtOAc or Et2O several times. The combined organic
layers were dried over MgSO4, filtered, and concentrated un-
der reduced pressure. The product was purified by column
chromatography and/or distillation. Although the control
reaction with E. coli BL21(DE3) harboring pACYCDuet-1
was performed in a similar way, the isolation of the product
was not attempted because of a very low conversion as deter-
mined by 1H NMR.
5.5. Coexpression of SCR and GDH genes
E. coli BL21(DE3) cells harboring pACRGD (tandem sys-
tem) were grown in LB medium (3 mL) containing chloram-
phenicol (34 mg/mL) at 37 ꢀC for 15 h with shaking at
230 rpm, and those harboring pESCR and pABGD (two-
plasmid system) were grown in LB medium (3 mL) contain-
ing ampicillin (100 mg/mL) and chloramphenicol (34 mg/
mL) in the same way. The culture (3 mL) was transferred
to the same medium (300 mL) in a 1-L Erlenmeyer flask,
and shaken at 200 rpm at 37 ꢀC. IPTG (0.1 mM) was added
when OD600 reached 0.6–0.8. The cells were further incu-
bated at 28 ꢀC for 18 h with shaking at 200 rpm, harvested
by centrifugation (7000 rpm, 4 ꢀC, 10 min), and washed
with 0.1 M phosphate buffer (pH 7.0, 50 mL). The cell pellet
was stored at ꢁ20 ꢀC until it was used for asymmetric
reduction.
5.8. Identification of products
Alcohols 1b–16b were characterized as described previ-
ously.4a,b 17b: 1H NMR (600 MHz, CDCl3) d 0.90 (t, J¼7.8 Hz,
3H), 1.18 (d, J¼6.0 Hz, 3H), 1.31 (sext, J¼7.8 Hz, 2H),
1.56 (quint, J¼7.8 Hz, 2H), 2.42 (t, J¼7.8 Hz, 2H), 2.50
(dd, J¼9.0, 17.4 Hz, 1H), 2.60 (dd, J¼3.0, 17.4 Hz, 1H),
3.14 (br s, 1H), 4.20–4.23 (m, 1H); 13C NMR (151 MHz,
CDCl3) d 13.8, 22.26, 22.27, 25.7, 43.3, 50.3, 63.8, 212.6;
[a]D27 +64.3 (c 1.84, CHCl3), >99% ee (S), lit.24 [a]D +50
(c 1.8) for (S)-17b with >99% ee. 18b: 1H NMR
(600 MHz, CDCl3) d 1.13 (s, 3H), 1.18 (s, 3H), 1.61 (br d,
J¼4.2 Hz, 1H), 1.64–1.70 (m, 1H), 1.81–1.86 (m, 1H),
2.00–2.07 (m, 2H), 2.37–2.45 (m, 2H), 3.70–3.73 (m, 1H);
13C NMR (151 MHz, CDCl3) d 19.5, 20.5, 22.7, 28.8,
37.1, 51.1, 77.7, 214.5; [a]2D7 +15.0 (c 0.80, CHCl3), 70%
ee (S), lit.14 [a]D21 +23.0 (c 2.0, CHCl3) for (S)-18b with
5.6. Enzyme activity
SCR activity was determined as described previously,4a after
cell disruption by sonication followed by centrifugation.
SCR activity of 1.0 U is defined as the amount of enzyme
that oxidizes 1.0 mmol of NADPH per minute at 25 ꢀC.4a
GDH activity was determined as follows. E. coli cells having
GDH activity were lysed by sonication for 1 min in an ice
bath (ꢂ10). Centrifugation (20,000 rpm, 4 ꢀC, 30 min)
gave a cell-free extract (CFE) as the supernatant. After ap-
propriately diluted, CFE (300 mL) was added to a solution
(2.5 mL) of glucose (100 mM) and NADP+ (2.0 mM) in
60 mM Tris buffer (pH 8.0) in a UV-cuvette thermostated
at 25 ꢀC. After the solution was quickly shaken, the reaction
rate was measured by following the increase in the absor-
bance of NADPH at 340 nm as a function of time. In this
paper, 1.0 U of GDH activity is defined as the amount of
enzyme that reduces 1.0 mmol of NADP+ per minute at
25 ꢀC under the above reaction conditions.
96.0–98.8% ee. 19b: H NMR (600 MHz, CDCl3) d 1.50
1
(d, J¼6.6 Hz, 3H), 4.30 (br s, 1H), 4.89 (q, J¼6.6 Hz, 1H),
7.18–7.21 (m, 1H), 7.27–7.28 (m, 1H), 7.67–7.70 (m, 1H),
8.53 (d, J¼4.2 Hz, 1H); 13C NMR (151 MHz, CDCl3)
d 24.2, 68.8, 119.8, 122.2, 136.8, 148.1, 163.0; [a]D25
ꢁ27.4 (c 1.03, CHCl3), >99% ee (S), lit.25 [a]D ꢁ24.5
1
(c 0.50, CHCl3) for (S)-19b with 98% ee. 20b: H NMR
(600 MHz, CDCl3) d 1.52 (d, J¼6.6 Hz, 3H), 1.95 (br s,
1H), 5.20 (q, J¼6.6 Hz, 1H), 7.00–7.03 (m, 1H), 7.14–
7.16 (m, 1H), 7.22–7.26 (m, 1H), 7.47–7.50 (m, 1H); 13C
NMR (151 MHz, CDCl3)
d 24.0, 64.6, 115.3 (d,
J¼21.9 Hz), 124.3 (d, J¼3.5 Hz), 126.6 (d, J¼4.7 Hz),
128.8 (d, J¼8.0 Hz), 132.6 (d, J¼13.3 Hz), 159.7 (d,
J¼245.7 Hz); 19F NMR (282 MHz, CDCl3) d ꢁ121.1 (m);
[a]2D5 ꢁ47.8 (c 1.46, CHCl3), >99% ee (S), lit.26 [a]D21
+47.2 (c 1.35, CHCl3) for (R)-20b with 97% ee.
5.9. Determination of enantiomeric purities
5.7. General procedure for whole-cell asymmetric
reduction
The enantiomeric purities of the obtained alcohols 2b, 3b,
5b–9b, 12b, 13b, 15b, and 16b were determined as reported
previously.4a,b The enantiomeric purities of 1b (acetate
form), 17b (acetate form), 18b (acetate form), 19b, and
20b were determined by capillary GC with a CP-cyclodextrin-
b-2,3,6-M-19 column (Chrompack, f 0.25 mmꢂ25 m). GC
for 1b (acetate form): Inj. 250 ꢀC, Col. 80 ꢀC, Det. 220 ꢀC,
(R) 32.7 min, (S) 37.8 min. GC for 17b (acetate form): Inj.
250 ꢀC, Col. 110 ꢀC, Det. 220 ꢀC, (S) 30.2 min, (R)
31.6 min. GC for 18b (acetate form): Inj. 250 ꢀC, Col.
110 ꢀC, Det. 220 ꢀC, (S) 45.0 min, (R) 47.5 min. GC for
19b: Inj. 300 ꢀC, Col. 70 ꢀC, Det. 250 ꢀC, (R) 82.6 min,
(S) 85.4 min. GC for 20b: Inj. 300 ꢀC, Col. 99 ꢀC, Det.
250 ꢀC, (R) 24.2 min, (S) 27.4 min. Alcohol 4b was
In the case of the tandem system, to a mixture of glucose
(1.08 g, 6.0 mmol), NADP+ (10 mg, 12 mmol), and E. coli
BL21(DE3) cells harboring pACRGD (2.0 g) in 0.1 M phos-
phate buffer (pH 7.0, 50 mL) was added the substrate (typi-
cally, 3.0 mmol). In the case of the two-plasmid system, to
a mixture of glucose (0.36 g, 2.0 mmol), NADP+ (10 mg,
12 mmol), and E. coli BL21(DE3) cells harboring pESCR
and pABGD (2.0 g) in 0.1 M phosphate buffer (pH 7.0,
10 mL) was added the substrate (typically, 1.0 mmol). The
mixture was stirred with a magnetic stirrer in a water bath
at 30 ꢀC. The progress of the reaction was monitored by
1
TLC and/or H NMR. In the two-plasmid system, the pH

T. Ema et al. / Tetrahedron 62 (2006) 6143–6149
6149
6. (a) Kataoka, M.; Yamamoto, K.; Kawabata, H.; Wada, M.; Kita,
K.; Yanase, H.; Shimizu, S. Appl. Microbiol. Biotechnol. 1999,
51, 486–490; (b) Kizaki, N.; Yasohara, Y.; Hasegawa, J.; Wada,
M.; Kataoka, M.; Shimizu, S. Appl. Microbiol. Biotechnol.
2001, 55, 590–595; (c) Yasohara, Y.; Kizaki, N.; Hasegawa,
J.; Wada, M.; Kataoka, M.; Shimizu, S. Tetrahedron:
Asymmetry 2001, 12, 1713–1718.
converted to the corresponding tosylate, which was analyzed
by HPLC with a chiral column (Daicel Chemical Industries):
Chiralpak AD-H, hexane/i-PrOH¼9/1, 0.5 mL/min, 254 nm,
(S) 21.2 min, (R) 22.1 min. The enantiomeric purities of
10b, 11b, and 14b were determined by 600 MHz H NMR
after conversion to the corresponding MTPA esters.
1
7. (a) Yamamoto, H.; Matsuyama, A.; Kobayashi, Y. Biosci.
Biotechnol. Biochem. 2002, 66, 481–483; (b) Yamamoto, H.;
Matsuyama, A.; Kobayashi, Y. Appl. Microbiol. Biotechnol.
2003, 61, 133–139; (c) Yamamoto, H.; Mitsuhashi, K.;
Kimoto, N.; Matsuyama, A.; Esaki, N.; Kobayashi, Y. Biosci.
Biotechnol. Biochem. 2004, 68, 638–649.
8. (a) Itoh, N.; Matsuda, M.; Mabuchi, M.; Dairi, T.; Wang, J. Eur.
J. Biochem. 2002, 269, 2394–2402; (b) Makino, Y.; Inoue, K.;
Dairi, T.; Itoh, N. Appl. Environ. Microbiol. 2005, 71, 4713–
4720.
Acknowledgements
We thank Dr. Hiroaki Yamamoto (Daicel Chemical
Industries) and Professor Tetsuo Toraya (Okayama
University) for the determination of the internal amino-acid
sequences of SCR and useful advice on gene expression,
respectively. This work was supported by a Grant-in-Aid
for Scientific Research from Japan Society for the Promo-
tion of Science (JSPS) and by a grant from Venture Business
Laboratory of Okayama University. We are grateful to the
SC-NMR Laboratory of Okayama University for the mea-
surement of NMR spectra.
9. Wada, M.; Yoshizumi, A.; Furukawa, Y.; Kawabata, H.; Ueda,
M.; Takagi, H.; Nakamori, S. Biosci. Biotechnol. Biochem.
2004, 68, 1481–1488.
10. Zhu, D.; Mukherjee, C.; Rozzell, J. D.; Kambourakis, S.;
Hua, L. Tetrahedron 2006, 62, 901–905.
11. Costello, C. A.; Payson, R. A.; Menke, M. A.; Larson, J. L.;
Brown, K. A.; Tanner, J. E.; Kaiser, R. E.; Hershberger,
C. L.; Zmijewski, M. J. Eur. J. Biochem. 2000, 267, 5493–5501.
12. Kao, C. M.; McPherson, M.; McDaniel, R. N.; Fu, H.; Cane,
D. E.; Khosla, C. J. Am. Chem. Soc. 1998, 120, 2478–2479.
13. (a) Anson, C. E.; Bibb, M. J.; Booker-Milburn, K. I.; Clissold,
C.; Haley, P. J.; Hopwood, D. A.; Ichinose, K.; Revill, W. P.;
Stephenson, G. R.; Surti, C. M. Angew. Chem., Int. Ed. 2000,
39, 224–227; (b) Booker-Milburn, K. I.; Gillan, R.;
Kimberley, M.; Taguchi, T.; Ichinose, K.; Stephenson, G. R.;
Ebizuka, Y.; Hopwood, D. A. Angew. Chem., Int. Ed. 2005,
44, 1121–1125.
14. Mori, K.; Mori, H. Organic Syntheses; Collect. Vol. VIII;
Wiley: New York, NY, 1993; pp 312–315.
15. (a) Zhou, B.; Gopalan, A. S.; VanMiddlesworth, F.; Shieh,
W.-R.; Sih, C. J. J. Am. Chem. Soc. 1983, 105, 5925–5926;
(b) Kitamura, M.; Ohkuma, T.; Takaya, H.; Noyori, R.
Tetrahedron Lett. 1988, 29, 1555–1556.
References and notes
1. Books: (a) Wong, C.-H.; Whitesides, G. M. Enzymes in
Synthetic Organic Chemistry; Elsevier: Oxford, 1994;
(b) Enzyme Catalysis in Organic Synthesis; Drauz, K.,
Waldmann, H., Eds.; Wiley-VCH: Weinheim, 2002; (c)
Faber, K. Biotransformations in Organic Chemistry, 5th ed.
Springer: Berlin, 2004; (d) Bommarius, A. S.; Riebel, B. R.
Biocatalysis; Wiley-VCH: Weinheim, 2004.
2. Recent reviews: (a) Sugai, T. Curr. Org. Chem. 1999, 3, 373–
406; (b) Stewart, J. D. Curr. Opin. Chem. Biol. 2001, 5, 120–
129; (c) Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada,
T. Tetrahedron: Asymmetry 2003, 14, 2659–2681; (d)
Kataoka, M.; Kita, K.; Wada, M.; Yasohara, Y.; Hasegawa, J.;
Shimizu, S. Appl. Microbiol. Biotechnol. 2003, 62, 437–445;
(e) Kroutil, W.; Mang, H.; Edegger, K.; Faber, K. Curr. Opin.
Chem. Biol. 2004, 8, 120–126.
3. (a) Ema, T.; Yamaguchi, K.; Wakasa, Y.; Yabe, A.; Okada, R.;
Fukumoto, M.; Yano, F.; Korenaga, T.; Utaka, M.; Sakai, T.
J. Mol. Catal., B Enzym. 2003, 22, 181–192; (b) Ema, T. Curr.
Org. Chem. 2004, 8, 1009–1025; (c) Ema, T. Tetrahedron:
Asymmetry 2004, 15, 2765–2770.
4. (a) Ema, T.; Sugiyama, Y.; Fukumoto, M.; Moriya, H.; Cui, J.-N.;
Sakai, T.; Utaka, M. J. Org. Chem. 1998, 63, 4996–5000; (b)
Ema, T.; Moriya, H.; Kofukuda, T.; Ishida, T.; Maehara, K.;
Utaka, M.; Sakai, T. J. Org. Chem. 2001, 66, 8682–8684;
(c) Ema, T.; Yagasaki, H.; Okita, N.; Nishikawa, K.;
Korenaga, T.; Sakai, T. Tetrahedron: Asymmetry 2005, 16,
1075–1078.
ˆ
16. Chenevert, R.; Fortier, G.; Rhlid, R. B. Tetrahedron 1992, 48,
6769–6776.
´
17. Spino, C.; Mayes, N.; Desfosses, H. Tetrahedron Lett. 1996, 37,
6503–6506.
18. Baldwin, J. E.; Melman, A.; Lee, V.; Firkin, C. R.; Whitehead,
R. C. J. Am. Chem. Soc. 1998, 120, 8559–8560.
19. Baldwin, J. E.; Adlington, R. M.; Ramcharitar, S. H. Synlett
1992, 875–877.
20. Sambrook, J.; Russell, D. W. Molecular Cloning: A Laboratory
Manual, 3rd ed.; Cold Spring Harbor Laboratory: New York,
NY, 2001.
21. Bradford, M. M. Anal. Biochem. 1976, 72, 248–254.
22. Marmur, J. J. Mol. Biol. 1961, 3, 208–218.
23. Baik, S.-H.; Ide, T.; Yoshida, H.; Kagami, O.; Harayama, S.
Appl. Microbiol. Biotechnol. 2003, 61, 329–335.
24. Ohta, H.; Ozaki, K.; Tsuchihashi, G. Agric. Biol. Chem. 1986,
50, 2499–2502.
´
5. (a) Rodrıguez, S.; Kayser, M. M.; Stewart, J. D. J. Am. Chem.
Soc. 2001, 123, 1547–1555; (b) Kaluzna, I.; Andrew, A. A.;
Bonilla, M.; Martzen, M. R.; Stewart, J. D. J. Mol. Catal., B
Enzym. 2002, 17, 101–105; (c) Kaluzna, I. A.; Matsuda, T.;
Sewell, A. K.; Stewart, J. D. J. Am. Chem. Soc. 2004, 126,
12827–12832; (d) Walton, A. Z.; Stewart, J. D. Biotechnol.
Prog. 2004, 20, 403–411; (e) Kayser, M. M.; Drolet, M.;
Stewart, J. D. Tetrahedron: Asymmetry 2005, 16, 4004–4009;
(f) Kaluzna, I. A.; Feske, B. D.; Wittayanan, W.; Ghiviriga,
I.; Stewart, J. D. J. Org. Chem. 2005, 70, 342–345.
25. Nakamura, K.; Takenaka, K.; Fujii, M.; Ida, Y. Tetrahedron
Lett. 2002, 43, 3629–3631.
26. Ohkuma, T.; Koizumi, M.; Doucet, H.; Pham, T.; Kozawa, M.;
Murata, K.; Katayama, E.; Yokozawa, T.; Ikariya, T.; Noyori,
R. J. Am. Chem. Soc. 1998, 120, 13529–13530.