Asymmetric synthesis of lipitor chiral intermediate using a robust carbonyl reductase at high substrate to catalyst ratio

Article abstract of DOI:10.1016/j.molcatb.2015.11.001

An NADPH-dependent carbonyl reductase (RpCR) from Rhodococcus pyridinivorans was discovered by genome mining for the asymmetric reduction of ethyl 4-chloro-3-oxo-butanoate (COBE). RpCR has been soluble expressed in Escherichia coli BL21(DE3). The highest activity is determined at pH 5.0 and 50 °C toward COBE. The apparent K_m and k_cat/K_m are 0.39 mM and 1747 s^-1 mM^-1, endowing RpCR with high catalytic efficiency in reduction of COBE. Employing merely 0.1 g recombinant RpCR-GDH in a toluene-aqueous biphasic system, as much as 7.0 g COBE could be asymmetrically reduced into ethyl (S)-4-chloro-3-hydroxybutanoate [(S)-CHBE] (>99% ee) without addition of external cofactor, achieving molar isolation yield of 91%, substrate to biocatalyst ratio of 70 and space-time yield of 1480 g L^-1 d^-1. Our results indicate the robust RpCR could be potentially applied in the preparation of optically pure (S)-CHBE.

Full text of DOI:10.1016/j.molcatb.2015.11.001

Accepted Manuscript
Title: Asymmetric synthesis of lipitor chiral intermediate using
a robust carbonyl reductase at high substrate to catalyst ratio
Author: Guo-Chao Xu Ming-Hui Tang Ye Ni
PII:
S1381-1177(15)30095-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2015.11.001
MOLCAB 3265
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Accepted date:
24-10-2015
1-11-2015
Please cite this article as: Guo-Chao Xu, Ming-Hui Tang, Ye Ni, Asymmetric
synthesis of lipitor chiral intermediate using a robust carbonyl reductase at
high substrate to catalyst ratio, Journal of Molecular Catalysis B: Enzymatic
http://dx.doi.org/10.1016/j.molcatb.2015.11.001
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Submitted to: Journal of Molecular Catalysis B: Enzymatic
Asymmetric synthesis of lipitor chiral intermediate using a
robust carbonyl reductase at high substrate to catalyst ratio
Guo-Chao Xu, Ming-Hui Tang, Ye Ni*
The Key Laboratory of Industrial Biotechnology, Ministry of Education, School of Biotechnology,
Jiangnan University, Wuxi 214122, Jiangsu, China
Corresponding author. Tel/Fax: +86-510-85329265; E-mail: yni@jiangnan.edu.cn
1

Graphical Abstract
2

Highlights



RpCR shares low sequence identity with known COBE carbonyl reductases.
RpCR exhibits a high kcat/Km of 1747 s−1·mM−1.
The substrate to catalyst ratio and space-time yield of RpCR reached 70 and 1480
g·L−1·d−1.
3

Abstract
An NADPH-dependent carbonyl reductase (RpCR) from Rhodococcus pyridinivorans was
discovered by genome mining for the asymmetric reduction of ethyl 4-chloro-3-oxo-butanoate
(COBE). RpCR has been soluble expressed in Escherichia coli BL21(DE3). The highest activity is
determined at pH 5.0 and 50 °C towards COBE. The apparent Km and kcat/Km are 0.39 mM and
1747 s−1·mM−1, endowed RpCR with high catalytic efficiency in reduction of COBE. Employing
merely 0.1 g recombinant RpCR-GDH in a toluene-aqueous biphasic system, as much as 7.0 g
COBE could be asymmetrically reduced into ethyl (S)-4-chloro-3-hydroxybutanoate [(S)-CHBE]
(>99% ee) without addition of external cofactor, achieving molar isolation yield of 91%, substrate
to biocatalyst ratio of 70 and space-time yield of 1480 g·L−1·d−1. Our results indicate the robust
RpCR could be potentially applied in the preparation of optically pure (S)-CHBE.
Keywords: Asymmetric reduction; Carbonyl reductase; Ethyl (S)-4-chloro-3-hydroxybutanoate;
Substrate to catalyst ratio
4

1. Introduction
Asymmetric reduction of ketones is one of the most important and practical approaches for
the production of chiral secondary alcohols, which is essential for the synthesis of industrially
important chemicals such as pharmaceuticals, agrochemicals and natural products [1–3]. As an
alternative to chemical methodologies, bioreductive preparation of (R) or (S)-enantiomers of
alcohols using isolated enzymes or whole-cell systems has been extensively investigated due to its
high enantio-, regio- and chemoselectivities, mild conditions, reproducibility and easy operation
[4,5]. However, two major challenges for the scale-up application of bioreduction systems are the
lack of efficient biocatalysts and the necessity of expensive cofactors such as NAD(P)H/NAD(P)+
[6,7]. The employment of robust biocatalysts with high cofactor utilization efficiency could reduce
or even avoid the addition of external cofactors. Consequently, exploration of robust biocatalysts
is of special interests to solve above issues [8,9].
Ethyl (S)-3-hydroxyl-4-chlorobutanoate [(S)-CHBE] is a versatile and important chiral
intermediate for the production of chiral drugs, including the cholesterol lowering
3-hydeoxy-3-methyl-glutaryl CoA (HMG-CoA) reductase inhibitors (namely statins) [10,11],
which rank the class I best-selling drugs due to its excellent therapeutic effect and low side effect.
The asymmetric reduction of ethyl 3-oxo-4-chlorobutanoate (COBE) into (S)-CHBE was the most
promising approach. In recent years, various microorganisms have been identified for the efficient
synthesis of (S)-CHBE and well-reviewed by Ye et al [11]. However the application of wild-type
strains is often hindered by their complicated dehydrogenases/reductases systems with variable
stereospecificity and low expression level of key reductases [11,12]. An effective solution is
discovering novel reductases by genome mining and their heterogeneous overexpression [11,13].
5

A number of enantioselective carbonyl reductases have been identified with S-selectivity in the
asymmetric reduction of COBE, including S1 from Candida magnolia [14], ARII from
Sporobolomyces salmonicolor [15], CPE from Candida parapsilosis [16], PsCRI and PsCRII from
Pichia stipitis [17,18], ScCR from Streptomyces coelicolor [19], SOU1 from Candida albicans
[20] and DhCR from Debaryomyces hansenii [21]. The highest catalytic efficiency (Vmax = 349
μmol·min−1·mg−1) was reported for ARII, however at a high substrate concentration (Km = 1.49),
and the ee was relatively lower for application [15]. A carbonyl reductase CPE identified by Wang
and coworkers displayed 99% ee and a low Km (0.19 mM), while its Vmax (200 μmol·min−1·mg−1)
was not as high as ARII [19].
Besides high stability and enantioslelectivity, the lower Km and higher Vmax (or higher kcat/Km)
is one of the crucial parameters of biocatalysts in large-scale application. Biocatalysts with high
kcat/Km value could often reach the maximum velocity and maintain the activity for a longer time
even at a low substrate concentration [22]. In current study, a novel carbonyl reductase from
Rhodococcus pyridinivorans (RpCR) was identified by genome mining, and soluble expressed in
Escherichia coli BL21(DE3). RpCR exhibited a high enantioselectivity, lower Km than CPE and
higher catalytic efficiency in the asymmetric reduction of COBE. Furthermore, the synthesis of
(S)-CHBE employing RpCR at high substrate loading was achieved in an organic solvent-aqueous
biphasic system to evaluate its potential for industrial applications.
2. Experimental procedures
2.1 Cloning and expression of RpCR coding gene in Escherichia coli BL21(DE3)
Genomic DNA was extracted from Rhodococcus pyridinivorans using a TIANamp Bacteria
6

DNA Kit from Tiangen (Shanghai). Primers with HindIII and XhoI restriction sites were designed
according to the RpCR coding gene (rpcr) sequence (GenBank accession No.: EHK80525.1). The
PCR product of rpcr was double digested with HindIII and XhoI and then inserted into the
expression vector pET28a. The resultant plasmid, pET28-rpcr, was transformed into E. coli
BL21(DE3). The cells were cultivated at 37 °C in LB medium supplemented with 50 μg/mL
kanamycin. When OD600 of the culture reached 0.6, isopropyl-β-D-thiogalactoside (IPTG) was
added to a final concentration of 0.2 mM, and the culture was further cultivated at 25 °C for 12 h.
2.2 Purification of RpCR
Cells were harvested by centrifugation (8000 × g, 10 min), washed twice with saline and
resuspended in buffer A (20 mM PBS, 500 mM NaCl, 10 mM imidazole, pH 7.4), followed by
disruption by ultrasonication (400 W, work 3 s and stop 2 s for 15 min). The cell lysate was
centrifuged (10000 × g, 30 min) at 4 °C. Afterwards, the supernatant was loaded onto a Histrap
column (1 mL, GE Corp.) pre-equilibrated with buffer A, and the proteins were eluted with an
increasing imidazole gradient from 10 to 500 mM with buffer B (20 mM PBS, 500 mM NaCl, 500
mM imidazole, pH 7.4) at a flow rate of 1 mL/min. Then the collected eluents were desalted and
concentrated at 4 °C. The purity of fractions was determined by SDS-PAGE. The purified RpCR
was stored at −80 °C with 20% glycerol for further use.
2.3 Enzyme activity assay protocol
Standard enzyme activity assay was performed spectrophotometrically by monitoring the
changes in absorbance of NADPH at 340 nm and 30 °C. The reaction mixture for RpCR consisted
7

of 2 mM COBE, 0.5 mM NADPH in 190 μL PBS buffer (pH 7.0, 100 mM) and 10 μL enzyme
solution with appropriated concentration. One unit of enzyme activity was defined as the amount
of enzyme that catalyzed the oxidation of 1 μmol NADPH per minute under standard condition.
2.4 Characterization of purified RpCR
2.4.1 Optimum pH and temperature
The optimum temperature of RpCR was determined under above mentioned standard
condition at various temperatures (25–70 °C). The pH-profile of purified RpCR was determined in
the following buffers (final concentration 100 mM): sodium citrate (pH 3.0–6.0), sodium
phosphate buffer (pH 6.0–8.0), Glycine-NaOH (pH 8.0–10.0). All the activities were assayed in
triplicates.
2.4.2 Effect of metal ions and additives on RpCR activity
The effect of various metal ions and additives (including Mn2+, Mg2+, Zn2+, Ca2+, Co2+, Ag+,
Fe3+, Ni2+, Cu2+, Al3+, EDTA, DTT, imidazole, β-mercaptoethanol, SDS, Tris, Triton-X100 and
Tween 20) on RpCR activity was examined by adding each compound (final concentration of 1
mM) in the reaction mixture for 60 min at 30 °C, and the residual enzyme activity was measured
under the standard condition. Control was performed in the absence of any test compound. All the
activities were assayed in triplicates.
2.4.3 Effect of organic solvents on RpCR activity
The influence of organic solvents on RpCR activity was performed by added with 50% v/v
8

organic solvents (n-pentane, cyclohexane, n-hexane, n-heptane, n-decane, n-nonane, isooctane,
toluene, ethyl lactate, ethyl acetate and glycerol), or 10% m/v PEG derivatives (PEG400, 600,
1000, 2000, and 4000), or 20% ionic liquid v/v (glycerol, choline chloride (ChCl), ChCl/glycerol,
ChCl/urea, ChCl/formic acid, ChCl/acetic acid, ChCl/oxalic acid, ChCl/malonic acid and
ChCl/citric acid) in the enzyme solutions (0.1 mg·mL−1), and incubated at 30 °C for 12 h. Then
the mixtures were centrifuged at 8000 × g and 4 °C for 5 min. The activity of the enzyme aqueous
phase was determined using standard activity protocol. Protein concentration was measured using
Bradford method with BSA as standard protein.
2.4.4 Kinetic analysis
The kinetic parameters of the purified RpCR were determined by determining the activity at
different COBE concentrations (0.25–5.0 mM) and a fixed NADPH concentration of 1.0 mM.
Similarly, the apparent Km value for NADPH was determined at different NADPH concentrations
(0.1–1 mM) and a fixed COBE concentration of 2 mM. All the activity was assayed in triplicates.
The apparent Km and Vmax values of the purified RpCR were calculated according to
Lineweaver-Burk plot [23].
2.4.5 Substrate specificity
Substrate profile of purified RpCR was determined using standard activity assay method,
except different prochiral ketone substrates including diketone, aromatic ketone, α-ketoester and
β-ketoester (final concentration, 2 mM) were used. All the activities were assayed in triplicates.
9

2.4.6 Enantioselectivity
Enantioselectivity of RpCR in the asymmetric reduction of COBE was performed using
chiral GC analysis as previously described [21].
2.5 Asymmetric reduction of COBE to (S)-CHBE employing RpCR
2.5.1 Construction of RpCR and GDH coexpression strain.
Glucose dehydrogenase coding gene (gdh) was cloned from Bacillus megaterium and
inserted into pACYDuet for expression as previously described [21]. The coexpression of RpCR
and glucose dehydrogenase (GDH) in Escherichia coli was developed simply by transformation of
pET28-rpcr and pACYDuet-gdh into E. coli BL21(DE3), and screening for double resistance
against kanamycin and chloramphenicol. Recombinant strain E. coli BL21(DE3) harboring
pET28-rpcr and pACYDuet-gdh (RpCR-GDH) was cultivated in LB medium supplemented with
50 μg/mL kanamycin and chloramphenicol at 37 °C and 180 rpm. When OD600 reached 0.6, 0.5
mM IPTG was added and further cultivated at 25 °C and 180 rpm for 12 h. The cells were
harvested by centrifugation followed by lyophilization. The dry cells of recombinant E. coli were
stored at 4 °C and used as biocatalyst.
2.5.2 Optimization of (S)-CHBE production by RpCR
The effect of reaction pH on the performance of RpCR was investigated with 1.65 g COBE,
3.0 g glucose and 0.05 g dry cells (RpCR-GDH), in 5 mL PBS buffer (pH 6.0, 6.5, 7.0 and 7.5,
100 mM) and 5 mL toluene at 30 °C and 180 rpm for 6 h. Then the reaction was stopped and
extracted with equal volume of ethyl acetate and the conversion and ee were analyzed by chiral
10

GC analysis.
2.5.3 Bioreductive preparation of (S)-CHBE with substrate fed-up
A 10 mL reaction was performed by adding 0.1 g dry cells and 5 mL PBS buffer (pH 7.0, 0.1
M), 1.8 g glucose (9.2 mmol) in 5 mL PBS buffer (0.1 M, pH 7.0) and pre-incubated at 30 °C for
10 min. Then the reaction was started by addition of 1.0 g COBE (6.1 mmol) dissolved in 5 mL
toluene into the reaction mixture, and then conducted at 30 °C and 180 rpm. The reaction pH was
maintained at pH 7.0 by titration with 0.5 M K2CO3. About 20 μL sample was taken intermittently ,
diluted with 480 μL PBS buffer (0.1 M, pH 7.0), and then extracted with 500 μL ethyl acetate
supplemented with 10 mM dodecane as internal standard. When the reaction was completed, 1.0 g
COBE (6.1 mmol) and 1.2 g glucose were fed each time for six times. Then the mixture was
extracted with equal volume of ethyl acetate and centrifuged (8000 × g, 10 min). The organic
phase was combined and dried over anhydrous Na2SO4. Afterwards, the extract was evaporated
and further dried in drying cabinet for 48 h.
3. Results and discussion
3.1 Screening of recombinant reductases
Genome mining strategy was adopted to explore novel enzymes capable of reducing COBE.
Twenty oxidoreductases sharing 40–60% amino acid sequence identities with known COBE
reductases were selected from NCBI databases. All the potential COBE reductases were
heterogeneous expressed in E. coli. The carbonyl reductase (GenBank accession No.:
EHK80525.1) from Rhodococcus pyridinivorans displayed the highest activity and
11

enantioselectivity towards COBE, which was designated as RpCR and chosen for further studies
(Table S1). As shown in Fig. 1, majority of RpCR was expressed in soluble form.
3.2 Multiple sequence alignment and sequence analysis
BLAST analysis in protein database revealed that RpCR shares high identity (over 55%) with
several hypothetical proteins. The highest identity of 43% is found with FabG from Bacillus
subtilis, which is reported to be
a robust reductase in the preparation of ethyl
(S)-2-hydroxy-4-phenylbutanoate [(S)-HPBE] [24]. Key motif analysis reveals that RpCR is a
potential member of short-chain dehydrogenases/reductases (SDR). Sequence-search against
Hidden Markov Model library of SDR database (http://www.sdr-enzymes.org/) returns that RpCR
belongs to cD2 subfamily of SDR [25]. Moreover, the secondary structure elements of RpCR was
predicted by the online program psipred (http://bioinf.cs.ucl.uk/psipred). Conserved patterns of
RpCR were identified with classic SDR as shown in Table 1. Compared with the Rossmann fold
for NADPH binding of TGxxxGhG in classic SDRs [26], a highly conserved TGASSGLG was
found in RpCR. In all the homogenous SDR proteins, there is a conserved aspartic acid residue in
the loop region between β3 and α3, which is essential for the stabilization of adenine ring of
cofactor. The conserved NNAG motif of RpCR is vital for the stabilization of central β-sheets [27].
The active sites in α4-β5-α5 are also conserved in RpCR. Furthermore, a proline residue in the β6,
involving in structural and reaction direction, is also found in region 183–195 of RpCR.
Consequently, RpCR is a member of classic SDR family.
12

3.3 Protein purification
RpCR with an N-terminal His-tag was purified to homogeneity by nickel affinity
chromatography. The specific activity of the purified RpCR is 223 U·mg−1, representing a
purification fold of 4.6 compared with crude extract. The purified enzyme was confirmed by
SDS-PAGE as a single band of about 29 kDa (Fig. 1), which is in accordance with the calculated
molecular mass based on gene sequence. According to gel-exclusion chromatography, the
molecular weight of RpCR is 118 kDa, suggesting RpCR is a homotetramer, with subunit of 29.5
kDa.
3.4 Temperature and pH profiles of purified RpCR
The optimum temperature for RpCR was determined by measuring activity at 25–70 °C as
illustrate in Fig. 2(A). The RpCR exhibits the highest activity at 50 °C. Remarkably, RpCR
retained 60% of activity at temperature as high as 65 °C. The pH profiles of RpCR was measured
at various pH ranging from 3.0 to 10.0. According to Fig. 2(B), the optimum pH of RpCR is
observed at pH 5.0. And the activity of purified RpCR decreases sharply at pHs higher or lower
than 5.0.
3.5 Effect of various metal ions and additives on RpCR
Various metal ions were evaluated as shown in Table 2. No metal ions is found with ability in
activating RpCR. The existence of Ag+, Fe3+, and Al3+ show inhibitory effect on the activity of
RpCR. The effect of some additives was also examine. SDS is extremely toxic to RpCR. Addition
of EDTA has no influence on the activity, indicating RpCR is not a metal ion dependent enzyme,
13

which is similar to most of other SDR members.
3.6 Effect of organic solvents and ionic liquids on RpCR
Since COBE is unstable in aqueous system and could be toxic to reductases, employing
organic solvents or ionic liquids in the reaction system could eliminate undesired side reactions
[19]. Ten organic solvents with different LogP values were examined. As shown in Table 3, RpCR
retains higher residual activity in aliphatic alkane and toluene after incubation for 12 h. Although
pentane, hexane, heptane and nonane are more biocompatible with RpCR, their partition
coefficients for COBE are lower. Among all the organic solvent test, toluene is regarded as the
preferred one as other reports [19].
Some PEG derivatives and deep eutectic solvents were also attempted to reduce the
spontaneous hydrolysis of COBE. However, after incubation for 12 h, the residual activity of
RpCR is decreased to less than 10% in most of the solvents. Although RpCR could retain higher
activity in ChCl/glycerol (135%) and glycerol (122%), the low solubility of COBE is not
advantageous for the bioreduction.
3.7 Substrate specificity
As shown in Table 4, RpCR could catalyze the reduction of all the tested diketones. Along
the chain length (1–3) increase, the specific activity also increases. The location of the carbonyl
group in the substrates is also important for the activity of RpCR. The activity towards
2,3-hexanedione (4) is higher than that of 3,4-hexanedione (5). With regard to the four tested
α-ketoesters, the existence of aromatic side-chain seems to be beneficial to the activity, except for
14

ethyl phenyl pyruvate (7). The specific activity of RpCR towards ethyl 4-phenyl-2-oxobutanoate
(8) is 92.1 U·mg−1. While the highest activity is found with COBE (117 U·mg−1), a β-ketoester.
No activity is detected towards aromatic ketones. A possible explanation is that the native
substrates of 3-ketoacyl-(acyl carrier protein) reductases (FabG) contain ester bonds, and therefore
the activity is also dependent on the recognition of ester bonds or alphatic diketones [28].
Enantioselectivity analysis reveals that RpCR follows anti-Prelog’s rule in the asymmetric
reduction of prochiral ketones and could reduce COBE into (S)-CHBE.
3.8 Kinetic properties
The purified RpCR displays no activity with NADH and full activity with NADPH,
suggesting that RpCR is NADPH-dependent. The kinetic parameters of purified RpCR were also
determined (Table 5). The Km and Vmax of RpCR towards COBE are 0.39±0.03 mM and 335±6
μmol·min−1·mg−1, while the Km and Vmax to NADPH were 0.17±0.01 mM and 398±7
μmol·min−1·mg−1. As summarized in Table 5, the Km and Vmax of RpCR are comparable with the
Km of CPE and Vmax of PsCRI respectively [16,17]. However, the Vmax of CPE (200
μmol·min−1·mg−1) and Km of PsCRI (4.9 mM) are not as beneficial as those of RpCR.
Consequently, RpCR displays the highest kcat/Km value (1747 s−1·mM−1), and is supposed to be a
highly efficient enzyme in the asymmetric reduction of COBE.
3.9 Effect of pH on the asymmetric reduction COBE
The RpCR and GDH were co-expressed in recombinant E. coli BL21(DE3), with specific
activities of 52 and 18 U·mg−1 crude extract at pH 5.0 and 8.0 respectively. However the
15

recombinant RpCR and GDH display different pH-profiles. As shown in Fig. 2(B), RpCR exhibits
the highest activity at an acidic pH of around 5.0. However, glucose dehydrogenase (GDH)
displays a basic pH optimum of 8.0–8.5 [29], and only 34% activity was remained at pH 5.0.
Consequently, the pH of reaction system is critical for the asymmetric reduction catalyzed by E.
coli coexpressing RpCR and GDH. To achieve the balance of RpCR and GDH activity in the
coexpression strain, the pH of the reaction mixture was optimized over a pH range from 6.0 to 7.5
(Fig. 3). After 6 h of reaction, the highest conversion is observed at pH 7.0, when RpCR retained
70% activity and GDH also displayed high activity. Besides, pH does not affect the
enantioselectivity of RpCR (data not shown). Therefore, the optimum pH of the reaction is 7.0.
3.10 (S)-CHBE production with substrate feeding
To eliminate the toxicity and spontaneous hydrolysis of COBE and improve the productivity,
a substrate feeding strategy in a 10-mL toluene/aqueous (1:1, v/v) biphasic system was employed
(Fig. 4). Substrate COBE (total 7.0 g) was fed at 1.0 g for each batch. At the same time, the
reaction system was maintained at pH 7.0 by titration with 0.5 M K2CO3 to neutralize the gluconic
acid generated in glucose oxidation. To avoid the extra addition of expensive NADP+, the amount
of biocatalyst loading was optimized. And the results show that 10 g·L−1 DCW is sufficient for the
complete conversion of 100 g·L−1 COBE within 3 h (data no shown).
Within 1 h, 1.0 g COBE (6.1 mmol) was entirely reduced into (S)-CHBE with >99% ee. Then
1.0 g COBE (6.1 mmol) and 1.2 g glucose was added. The second batch was completed within 70
min. Afterwards, another 1.0 g COBE (6.1 mmol) and 1.2 g glucose was fed when COBE was
converted thoroughly. At the 7th batch, at least 3 h was required for a complete reaction. All 7.0 g
16

COBE was fully reduced into (S)-CHBE by merely 0.1 g RpCR-GDH dry cells within 11 h
without addition of external NADP+. After extraction, about 6.29 g (S)-CHBE (>99% ee) was
recovered, with a molar yield of about 91%. The substrate to catalyst ratio and the space-time
yield were about 70 and 1480 g·L−1·d−1, which was attributed to the high catalytic efficiency of
RpCR in the biphasic system with substrate-feeding. Table 4 shows a comparison of the
asymmetric reduction of COBE to (S)-CHBE by various recombinant E. coli strains. It has been
reported that S1 could accumulate 2580 mM (430 g·L−1) (S)-CHBE in the organic phase with a
molar isolation yield of 89% and substrate to catalyst ratio of 27.7 [30, 31], however under the
addition of 0.13 mM expensive NADP+. Although PsCRI and PsCRII display higher Vmax values,
the substrate to biocatalyst ratios are 7.7 and 5.1 respectively, which are a result of the higher Km
towards COBE [32,33]. The substrate to biocatalyst ratios of ScCR and SOU1 are higher than 30.
Notably, the application potential of ScCR has been successfully scaled up at a 40-L volume
loaded with 100 g·L−1 COBE under assistance of 0.3 mM NAD+ [34]. The RpCR displays better
performance in the preparation of (S)-CHBE with regard to the cofactor addition, kcat/Km value and
substrate to biocatalyst ratio. As much as 7.0 g COBE can be completed reduced into (S)-CHBE
by merely 0.1 g recombinant whole cells without addition of external cofactor. The substrate to
biocatalyst ratio of RpCR (70) in this study ranks the highest record due to its high kcat/Km value of
1747 compared with other COBE reductases (Table 5). To further decrease the addition of
biocatalyst and improve the productivity, protein engineering of RpCR is undergoing in our
laboratory for lowering the Km value towards NADP+.
17

4. Conclusions
The COBE reductase RpCR from Rhodococcus pyridinivorans was discovered by genome
mining. RpCR displays high substrate affinity and catalytic efficiency towards COBE. RpCR is a
promising biocatalyst for the synthesis of (S)-CHBE considering its high stereoselectivity, yield
and final product concentration. The substrate to catalyst ratio and the space-time yield of RpCR
in asymmetric reduction of COBE are about 70 and 1480 g·L−1·d−1, respectively. To our best
knowledge, this is the first report on newly identified RpCR serving as an efficient reductase for
asymmetric preparation of (S)-CHBE.
Acknowledgements
We are grateful to National Natural Science Foundation of China (21276112, 21506073), Natural
Science Foundation of Jiangsu Province (BK20150003), the Fundamental Research Funds for the
Central Universities (JUSRP51409B), the Program of Introducing Talents of Discipline to
Universities (111-2-06), and a project funded by the Priority Academic Program Development of
Jiangsu Higher Education Institutions for the financial support of this research.
18

References
[1] W. Kroutil, H. Mang, K. Edegger, K. Faber, Curr. Opin. Chem. Biol. 8(2004):120–126.
[2] Y. Ni, J. H. Xu, J. Mol. Catal. B:Enzym. 18(2002):233–241.
[3] G. C. Xu, Y. Ni, Bioresour. Bioprocessing 2(2015):15.
[4] Y. Ni, J. H. Xu, Biotechnol. Adv. 30(2012):1279–1288.
[5] U. T .Bornscheuer, G. W. Huisman, R. J. Kazlauskas, S. Lutz, J. C. Moore, K. Robins, Nature
485(2012):185–194.
[6] J. C. Moore, D. J. Pollard, B. Kosjek, P. N. Devine, Acc. Chem. Res. 40(2007):1412–1419.
[7] F. Hollmann, I. W. C. E. Arends, D. Hotmann, Green Chem. 13(2011):2285–2314.
[8] K. Nakamura, R. Yamanaka, T. Matsuda, T. Harada, Tetrahedron:Asymmetry 14(2003):2659–
2681.
[9] T. Matsuda, R. Yamanaka, K. Nakamura, Tetrahedron:Asymmetry 20(2009):513–557.
[10] S. K. Ma, J. Gruber, C. Davis, L. Newman, D. Gray, A. Wang, J. Grate, G. W. Huismann, R.
A. Sheldon, Green Chem. 12(2010):81–86.
[11] Q. Ye, P. K. Ouyang, H. J. Ying, Appl. Microbiol. Biotechnol. 89(2011):513–522.
[12] K. Goldberg, K. Schroer, S. Lütz, A. Liese, Appl. Microbiol. Biotechnol. 76(2007):249–255.
[13] K. Goldberg, K. Schroer, S. Lütz, A. Liese, Appl. Microbiol. Biotechnol. 76(2007):237–248.
[14] M. Wada, M. Kataoka, H. Kawabata, Y. Yasohara, N. Kizaki, J. Hasegawa, S. Shimizu,
Biosci. Biotechnol. Biochem. 62(1998):280–285.
[15] K. Kita, K. Nakase, H. Yanase, M. Kataoka, S. Shimizu, J. Mol. Catal. B:Enzym.
6(1999):305–313.
[16] Q. Y. Wang, L. H. Shen, T. T. Ye, D. Cao, R. Chen, X. L. Pei, T. Xie, Y. Li, W. B. Gong, X. P.
Yin, Bioresour. Technol. 123(2012):690–694.
[17] Q. Ye, M. Yan, Z. Yao, L. Xu, H. Cao, Z. J. Li, Y. Chen, S. Y. Li, J. X. Bai, J. Xiong, H. J.
Ying, P. K. Ouyang, Bioresour. Technol. 100(2009):6022–6027.
[18] H. Cao, , M. Lan, Q. Ye, G. Zheng, M. Yan, Y. Wang, Y. Y. Zhang, X. M. Li, L. Xu, J. Xiong,
P. K. Ouyang, H. J. Ying, Bioresour. Technol. 102(2011):1733–1739.
[19] L. J. Wang, C. X. Li, Y. Ni, J. Zhang, X. Liu, J. H. Xu, Bioresour. Technol. 102(2011):7023–
7028.
[20] P. Cai, M. D An, S. Xu, M. Yan, N. Hao, Y. Li, L. Xu, Biosci. Biotechnol. Biochem.
19

79(2015):1090–1093.
[21] G. C. Xu, H. L. Yu, Y. P. Shang, J. H. Xu, RSC Adv. 5(2015):22703–22711.
[22] Y. Ni, Y. N. Su, H. D. Li, J. Y. Zhou, Z. H. Sun, J. Biotechnol. 168(2013):493–498.
[23] H. Lineweaver, D. Burk, J. Am. Chem. Soc. 25(1934):658–666.
[24] Y. Ni, C. X. Li, J. Zhang, N. D. Shen, U. T. Bornscheuer, J. H. Xu, Adv. Synth. Catal.
353(2011):1213–1217.
[25] U. Oppermann, C. Filling, M. Hult, N. Shafqat, X. Wu, M. Lindh, J. Shafqat, E. Nordling, Y.
Kallberg, B. Persson, H. Jornvall, Chem. Biol. Interact. 2003:247–253.
[26] H. Jornavall, J. O. Hoog, B. Persson, FEBS Lett. 445(1999):261–264.
[27] C. Filling, K. D. Berndt, J. Benach, S. Knapp, T. Prozorovski, E. Nordling, R. Ladenstein, H.
Jornvall, U. Oppermann, J. Biol. Chem. 277(2002):25677–25684.
[28] K. Hölsch, J. Havel, M. Haslbeck, D. Weuster-Botz, Appl. Environ. Microbiol.
74(2008):6697–6702.
[29] T. Mitamura, I. Urabe, H. Okada, Eur. J. Biochem. 186(1989):389–393.
[30] N. Kizaki, Y. Yasohara, J. Hasegawa, M. Wada, M. Kataoka, S. Shimizu, Appl. Microbiol.
Biotechnol. 55(2001):590–595.
[31] Y. C. He, Z. C. Tao, X. Zhang, Z. X. Yang, J. H. Xu, Bioresour. Technol. 161(2014):461–464.
[32] Q. Ye, H. Cao, M. Lan, M. Yan, Y. Wang, Q. T. He, J. Li, L. Xu, Y. Chen, J. Xiong, P. K.
Ouyang, H. J. Ying, Bioresour. Technol. 101(2010):8911–8914.
[33] Q. Ye, H. Cao, M. Lan, M. Yan, Y. Wang, Y. Y. Zhang, X. M. Li, L. Xu, J. Xiong, P. K.
Ouyang, H. J. Ying, Appl. Microbiol. Biotechnol. 88(2010):1277–1285.
[34] J. Pan, G. W. Zheng, Q. Ye, J. H. Xu, Org. Process Res. Dev. 18(2014):739–743.
20

Figure 1. SDS-PAGE analysis of the purified RpCR. Lane 1, supernatant; Lane 2, precipitant;
Lanes 3 & 4, purified RpCR with different dilution fold; Lane M, protein molecular mass markers.
21

120
100
80
60
40
20
0
(A)
20
30
40
50
60
70
80
Temperature /oC
120
100
80
60
40
20
0
(B)
2
4
6
8
10
12
pH
Figure 2. Effect of temperature and pH on the activities of purified RpCR. (A)
Activity-temperature profile was estimated at various temperatures (25–70 °C) in PBS buffer (pH
7.0, 100 mM). (B) Activity-pH profile was determined using standard protocol in the following
100 mM buffers: (i) citrate (pH 3.0–6.0), (ii) PBS buffer (pH 6.0–8.0) and (iii) Gly-NaOH (pH
8.0–10.0). Relative activity was expressed as a percentage of the maximum activity under
experimental conditions.
22

100
90
80
70
60
50
40
30
20
10
0
6.0
6.5
7.0
7.5
pH
Figure 3. Effect of pH on the asymmetric reduction of COBE employing recombinant E. coli
BL21 coexpressing RpCR-GDH. Reactions were carried out with 10 mmol COBE, 15 mmol
glucose, and 0.1 g dry cells in 10 mL PBS buffer (pH 6.0, 6.5, 7.0, 7.5; 100 mM) at 30 °C and 180
rpm for 12 h.
23

45
40
35
30
25
20
15
10
5
100
80
60
40
20
0
0
0
1
2
3
4
5
6
7
8
9
10 11
Time /h
Figure 4. Time course of (S)-CHBE production catalyzed by recombinant E. coli coexpressing
RpCR-GDH in a toluene/aqueous biphasic system with substrate feeding. () production of
(S)-CHBE; (▲) ee. Reduction of COBE using 0.1 g dry cells was conducted at 30 °C in 5 mL PBS
buffer (pH 7.0, 100 mM) and 5 mL toluene containing 1.0 g COBE and 1.8 g glucose; 1.0 g COBE
and 1.2 g glucose were fed at the end of each batch.
24

Table 1. The secondary structure element motifs in classic SDR and RpCR.
Secondary
structure
element
Conserved motifs of
classical SDR
Function
Motifs of RpCR
Position in RpCR
β1-α1
β3-α3
β4
TGxxxGhG
Dhx[cp]
Coenzyme binding region
TGASSGLG
DIAD
17–24
Adenine ring binding of coenzyme
68–71
GxhDhhhNNAGh
hNhxG
Structural role in stabilizing central
β-sheet
GRVDILINNAGI
INLNG
87–98
α4
Part of active site
Part of active site
Part of active site
118–122
131–140
159–163
β5
GxhhxhxSSh
Yx[A/S][S/T]K
GRVMQPGSSI
YSASK
α5
β6
h[K/R]h[N/S]xhxPGxxxT Structural role, reaction direction
IRVNAIAPGFFRT 183–195
Note: in the motifs: ‘h’, a hydrophobic residue; ‘c’, a charged residue; ‘p’, a polar residue; ‘x’, any residue.
Conserved amino acids are underlined. Alternative ones are given within brackets.
25

Table 2. Effect of metal ions and additives on the activity of RpCR.
Metal ion
Control
Mn2+
Mg2+
Zn2+
Ca2+
Co2+
Ag+
Fe3+
Ni2+
Cu2+
Relative activity /%
100
Additive
EDTA
Relative activity /%
98.8 ± 1.7
88.7 ± 4.7
81.3 ± 1.8
63.0 ± 3.3
91.7± 3.2
83.0 ± 0.8
14.4 ± 0.5
17.1 ± 1.6
74.1 ± 1.4
96.9 ± 0.6
28.2 ± 1.5
DTT
79.8 ± 1.7
Imidazole
69.1 ± 0.8
β-mercaptoethanol 85.0 ± 3.4
SDS
1.1 ± 0.8
Tris base
Triton-X100
Tween 20
83.0 ± 0.8
91.1 ± 4.6
106.7 ± 1.9
Al3+
26

Table 3. Effect of organic solvents, ionic liquids and PEG on the activities of RpCR.
Organic solvent Residual activity /% Organic solvent
Residual activity /%
10.0 ± 0.9
Control
73.2 ± 2.9
102 ± 3.2
93.8 ± 8.8
101 ± 4
PEG400
n-Pentane
Cyclohexane
n-Hexane
n-Heptane
n-Decane
n-Nonane
iso-Octane
Toluene
PEG600
5.62 ± 0.52
15.1 ± 0.3
PEG1000
PEG2000
4.02 ± 0.06
0.87 ± 0.15
135 ± 1
103 ± 3
PEG4000
85.9 ± 0.5
103 ± 4
ChCl/glycerol
ChCl/Urea
85.9 ± 2.9
91.1 ± 7
ChCl/formic acid
ChCl/acetic acid
ChCl/oxalic acid
2.79 ± 0.11
1.50 ± 0.11
3.87 ± 0.46
98.3 ± 3.6
0.76 ± 0.53
73.2 ± 8.8
122 ± 3
Ethyl lactate
Ethyl acetate
Glycerol
ChCl/malonic acid 1.80 ± 0.20
ChCl/citric acid 1.45 ± 0.06
ChCl
65.6 ± 5.4
27

Table 4. Substrate specificity of RpCR towards various prochiral ketones.
Substrate
Specific activity /U·mg−1
1
2
3
4
5
6
8.37 ± 0.22
11.0 ± 0.4
26.0 ± 0.5
9.81 ± 0.22
4.59 ± 0.14
5.69 ± 0.03
7
0.51 ± 0.01
8
9
92.1 ± 0.6
117 ± 1
10
0.99 ± 0.02
11
n. d. a
n. d.
12
a n. d.: no activity was detected.
28

Table 5. Comparison of various enzymes capable of reducing COBE into (S)-CHBE.
Enzyme a
S1
ARII
CPE
PsCRI
PsCRII
ScCR
Q9EX28 P87219
26
SDR
SOU1
DhCR
Q6BQ25 H0JYI9
33.6
SDR
RpCR
Accession No. b
MW of subunit (kDa)
Protein family
Q9C4B3 Q9UUN9 G8B541 A3GF07 A3GF05
32
SDR
37
AKR
37.7
SDR
33
SDR
30.8
SDR
31
SDR
29.5
SDR
Cofactor
NADPH NADPH
NADPH NADPH NAD(P)H NADH
NADPH NADPH NADPH
Optimum pH
5.5
55
4.6
270
5.5
40
1.49
349
5.5
40
0.19
200
6
6
6.5
45
0.49
43.7
6.2
–
–
6.5
55
1.3
16.6
5
50
0.39
335
Optimum Temp.
Km for COBE (mM)
Vmax for COBE
(μmol·min−1·mg−1)
kcat/Km (s−1·mM−1)
Stereoselectivity (%)
Gene size (bp)
Molar yield (%)
Substrate/catalyst
References
30
4.9
337
30
3.3
224
–
110
100
849
89
24.7
[14,30]
380
>93
1032
–
–
[15]
1675
>99
1029
91.1
–
125
>99
849
90
7.7
[17]
132
>99
855
91
5.1
[18]
206
>99
798
93
36.8
[19]
–
22.8
>99
840
92.5
16.5
[21]
1747
>99
762
91
70
>99
846
82.6
30.9
[20]
[16]
This work
29