Enantioselective synthesis of (S)-phenylephrine by whole cells of recombinant Escherichia coli expressing the amino alcohol dehydrogenase gene from Rhodococcus erythropolis BCRC 10909

Article abstract of DOI:10.1016/j.procbio.2010.06.003

(R)-phenylephrine [(R)-PE] is an α₁-adrenergic receptor agonist that is widely used in over-the-counter drugs to treat the common cold. We found that Rhodococcus erythropolis BCRC 10909 can convert detectable level of 1-(3-hydroxyphenyl)-2-(methylamino) ethanone (HPMAE) to (S)-PE by high performance liquid chromatography tandem mass spectrometry analysis. An amino alcohol dehydrogenase gene (RE_AADH) which possesses the ability to convert HPMAE to (S)-PE was then isolated from R. erythropolis BCRC 10909 and expressed in Escherichia coli NovaBlue. The purified RE_AADH, tagged with 6×His, had a molecular mass of approximately 30kDa and exhibited a specific activity of 0.19μU/mg to HPMAE in the presence of NADPH, indicating this enzyme could be categorized as NADP⁺-dependent short-chain dehydrogenase reductase. E. coli NovaBlue cell expressing the RE_AADH gene was able to convert HPMAE to (S)-PE with more than 99% enantiomeric excess (ee), 78% yield and a productivity of 3.9mmol(S)-PE/Lh in 12h at 30°C and pH 7. The (S)-PE, recovered from reaction mixture by precipitation at pH 11.3, could be converted to (R)-PE (ee>99%) by Walden inversion reaction. This is the first reported biocatalytic process for the production of (S)-PE from HPMAE.

Full text of DOI:10.1016/j.procbio.2010.06.003

Process Biochemistry 45 (2010) 1529–1536
Contents lists available at ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Enantioselective synthesis of (S)-phenylephrine by whole cells of recombinant
Escherichia coli expressing the amino alcohol dehydrogenase gene from
Rhodococcus erythropolis BCRC 10909
Wei-De Lina,b,c, Chien-Yu Chen , Huei-Chung Chen , Wen-Hwei Hsu
a
a
a,∗
a
Institute of Molecular Biology, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 402, Taiwan
School of Post Baccalaureate Chinese Medicine, China Medical University, Taichung 404, Taiwan
Department of Medical Research, China Medical University Hospital, Taichung 404, Taiwan
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
(R)-phenylephrine [(R)-PE] is an ␣ -adrenergic receptor agonist that is widely used in over-the-counter
drugs to treat the common cold. We found that Rhodococcus erythropolis BCRC 10909 can convert
detectable level of 1-(3-hydroxyphenyl)-2-(methylamino) ethanone (HPMAE) to (S)-PE by high perfor-
mance liquid chromatography tandem mass spectrometry analysis. An amino alcohol dehydrogenase
gene (RE AADH) which possesses the ability to convert HPMAE to (S)-PE was then isolated from R. ery-
thropolis BCRC 10909 and expressed in Escherichia coli NovaBlue. The purified RE AADH, tagged with
1
Received 25 March 2010
Received in revised form 10 May 2010
Accepted 7 June 2010
Keywords:
Rhodococcus erythropolis
Amino alcohol dehydrogenase
Phenylephrine
Enantioselective synthesis
Bioconversion
6
×His, had a molecular mass of approximately 30 kDa and exhibited a specific activity of 0.19 ␮U/mg
+
to HPMAE in the presence of NADPH, indicating this enzyme could be categorized as NADP -dependent
short-chain dehydrogenase reductase. E. coli NovaBlue cell expressing the RE AADH gene was able to
convert HPMAE to (S)-PE with more than 99% enantiomeric excess (ee), 78% yield and a productivity of
◦
3
.9 mmol (S)-PE/L h in 12 h at 30 C and pH 7. The (S)-PE, recovered from reaction mixture by precipitation
at pH 11.3, could be converted to (R)-PE (ee > 99%) by Walden inversion reaction. This is the first reported
biocatalytic process for the production of (S)-PE from HPMAE.
©
2010 Elsevier Ltd. All rights reserved.
1
. Introduction
metric hydrogenation and chemical resolution [8–10]. However, a
racemic mixture typically results from symmetric hydrogenation,
and repeat kinetic resolution to obtain the desired optical product
often is both expensive and tedious. In recent years, several chem-
ical asymmetric hydrogenation methods have been developed to
allow for more direct (R)-PE synthesis [11–14]; but these pro-
cesses usually require high pressure, high temperature, and several
organic solvents that are not environmentally friendly.
Phenylephrine (PE), a sympathomimetic agonist, has a wide
variety of surgical applications, such as counteracting hypotension.
It also is an active ingredient used in eye drops to dilate the pupil
and in topical nasal decongestants [1,2]. Although the systemic
bioavailability of PE in nasal decongestants is only about 38% that of
orally administered d-pseudoephedrine (d-PDE) [3,4], PE does not
cause the release of endogenous noradrenaline [5] and is less likely
tocause side effectslikecentralnervous system stimulation, insom-
nia, anxiety, irritability and restlessness [2]. Moreover, the Combat
Methamphetamine Epidemic Act of 2005 permits the banning of
over-the-count sales of medications containing d-PDE, in order to
control the clandestine manufacture of methamphetamine from d-
PDE in the USA. This has resulted in the substitution of (R)-PE for
d-PDE in many nasal decongestants [6,7].
Chiral aryl alcohols are important chiral building blocks for
the production of optically active drugs in the pharmaceutical
and agrochemical industries [15–17]. They can be obtained by
asymmetric reduction of prochiral aryl ketones using oxidoreduc-
tases. Chiral 2-chloro-1-phenylethanol is a key intermediate during
the preparation of anti-depressants and potential therapeutic
agents of cocaine-abuse [18]. It can be converted from 2-
chloro-1-phenylethanone by asymmetric reduction, using Candida
+
PE contains a chiral center in the C␣ of the side chain, and (R)-
PE exhibits more potent than the S enantiomer in activating of
␣1
magnoliae NADP -dependent carbonyl reductase [19] or Ralstonia
+
sp. NADP -dependent alcohol dehydrogenase [20]. Corynebac-
+
-adrenergic receptors [2]. (R)-PE can be synthesized by sym-
terium sp. ST-10 with NAD -dependent phenylacetaldehyde
reductase activity can be used in the conversion of 2-chloro-1-
(
3-chlorophenyl)ethanone to 2-chloro-1-(3-chlorophenyl)ethanol,
a precursor for the synthesis of ␤3 adrenergic receptor agonists
21,22]. Very recently, Escherichia coli cells that express the l-1-
∗ Corresponding author. Tel.: +886 4 2285 1885; fax: +886 4 2287 4879.
E-mail address: whhsu@dragon.nchu.edu.tw (W.-H. Hsu).
[
1
359-5113/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2010.06.003

1
530
W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
Fig. 1. Schematic diagram showing the conversion of HPMAE to (S)-PE by asymmetrical biocatalysis and the conversion of (S)-PE to (R)-PE by Walden inversion reaction.
amino-2-propanol dehydrogenase (AADH) gene from Rhodococcus
erythropolis MAK154 have been demonstrated to possess the capac-
ity to reduce (S)-1-phenyl-2-methylamino-propan-1-one (MAK) to
d-PDE [23].
PE and d-PDE are almost identical in their structure, except that
PE contains one hydroxyl group in the phenyl moiety, and d-PDE
contains one methyl group at C␤ in the side chain. To the best
of our knowledge, no bioconversion method for the production
of chiral PE from 1-(3-hydroxyphenyl)-2-(methylamino) ethanone
2.3. Screening of bacterial strains capable of converting HPMAE to PE
Bacterial colonies were inoculated into 50 ml Luria-Bertani (LB) medium and
incubated at a given temperature for each microorganism with vigorous shaking
until an OD600 of approximately 2.0 was achieved. Cells were harvested by centrifu-
gation at 12,000 × g for 10 min. Cell pellets were resuspended in 10 ml of 100 mM
sodium phosphate buffer (pH 7) or sodium citrate buffer (pH 6) containing 2% glu-
◦
cose and 10 mM HPMAE, and then incubated at 28 or 37 C for 2 h. Once the reaction
was completed, the cells were removed by centrifugation. The reaction solution was
filtered through a 0.22-␮m membrane and subjected to HPLC–MS-MS analysis.
(
HPMAE) has been reported. In this study, we describe that E.
2.4. Expression of the RE AADH gene in E. coli
coli cells expressing amino alcohol dehydrogenase gene (RE AADH)
from R. erythropolis BCRC 10909 are able to produce (S)-PE from
HPMAE, and the resulting (S)-PE could subsequently be converted
to (R)-PE, a clinically useful sympathomimetic agonist, by Walden
inversion reaction (Fig. 1).
Genomic DNA of R. erythropolis BCRC 10909 was isolated using a MasterPure
Gram Positive DNA Purification Kit (Epicentre Biotechnologies, Madison, WI). The
open reading frame (ORF) of RE AADH was obtained by PCR using R. erythropolis
genomic DNA as the template, and primers designed on the basis of the R. erythropolis
ꢀ
MAK154AADHgene[23]: forward primer (5 -CGGGATCCATGTTCAACTCCATTGAAG),
which introduces the sequence for a unique Bam HI site (underlined); and
ꢀ
reverse primer (5 -CCCAAGCTTTTACAGTTCGCCGAGCGCCAT), which incorporates
2
.
Materials and methods
.1. Chemicals, bacterial strains and media
R)-PE and other general chemicals were purchased from Sigma–Aldrich (St.
sequences from a unique Hind III site (underlined). The PCR product was cloned
as a Bam HI-Hind III fragment into the corresponding site of pQE-30 to generate
the plasmid pQE-aadh, thereby allowing high-level expression of the N-terminal
2
6
×His-tagged RE AADH in E. coli. The recombinant plasmid was introduced into
(
E. coli NovaBlue. For high-level expression of the RE AADH gene, E. coli NovaBlue
Louis, MO). All solvents used in HPLC analysis were of LC grade and purchased from
Merck (Darmstadt, Germany). HPMAE and (S)-PE were obtained from Industrial
Technology Research Institute (Hsinchu, Taiwan). The chemicals required for protein
assay and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
were obtained from Bio-Rad (Hercules, CA). Culture media were obtained from Bec-
ton, Dickinson and Company (Sparks, MD). Bacterial strains were purchased from
the Food Industry Research and Development Institute (Hsinchu, Taiwan). Restric-
tion enzymes and T4 DNA ligase were purchased from New England Biolabs (Beverly,
MA). The expression vector pQE-30 was purchased from QIAGEN (Hilden, Germany).
◦
(
pQE-aadh) was grown in LB broth containing ampicillin (100 ␮g/ml) at 37 C for
1
3
6 h with shaking. The culture was diluted (1:100) with 100 ml of LB and grown at
7 C to an OD600 of approximately 0.6–0.8. To induce gene expression, isopropyl-
◦
␤
-d-thiogalactopyranoside (IPTG) was added to a final concentration of 0.5 mM and
◦
the cultivation continued at 30 C for an additional 6 h. Cells were collected by cen-
◦
trifugation at 12,000 × g and 4 C for 10 min. Bacterial pellets were washed twice
with 25 ml of ice-cold 100 mM sodium phosphate buffer (pH 7.0) and resuspended
in the same buffer. The IPTG-induced cells were disrupted by sonication and the
◦
cell debris pelleted by centrifugation at 4 C for 15 min. The resulting supernatant
was subjected to a Ni-NTA column (QIAGEN, Hilden, Germany), and the purifica-
tion procedure carried out in accordance with the manufacturer’s instructions. The
expression level of the RE AADH gene was analyzed by SDS-PAGE. The concentra-
tion of purified protein was measured with a protein assay kit (Bio-Rad, Hercules,
CA), using bovine serum albumin as the standard. Reduction activity of RE AADH
was measured by determining the production of PE from HPMAE by chiral-HPLC.
The assay mixture (500 ␮l), containing the appropriate amount of enzyme, 5 mM
2
.2. Analysis of HPMAE and PE
A
high performance liquid chromatography tandem mass spectrometry
(
HPLC–MS-MS) system consisting of two Perkin-Elmer Series 200 Micro LC pumps,
a Series 200 Autosampler (Perkin-Elmer Co., Waltham, MA), and an AB-Sciex
API-2000 triple quadrupole mass spectrometer with TurboIonSpray probe (Applied-
Biosystems, Foster City, CA), was used to rapidly identify HPMAE and PE in the
reaction mixture. Data processing was performed using Analyst version 1.3.1 soft-
ware (Applied-Biosystems). HPLC analysis was performed on a reversed-phase
Polaris C-18A column (2 mm i.d. × 50 mm; particle size, 3 ␮m) (Varian Inc. Palo Alto,
CA). The mobile phase was 0.1% formic acid and methanol at a ratio of 15:85, and
the flow rate was 100 ␮l/min. For MS-MS detection, the TurboIonSpray, orifice volt-
HPMAE, and 1 mM NADPH in 100 mM sodium phosphate buffer (pH 6.0), was incu-
bated at 30 ◦C. The enzyme activity of one unit (U) was defined as the amount of
enzymes that result in the formation of 1 ␮mol PE from HPMAE per minute at 30 ◦C.
2.5. Production of (S)-PE using recombinant E. coli cells
ages, temperature, collision energy, and entrance potential were set at 5500 V, 80 V,
A glass vessel (250 ml) equipped with a propeller-type impeller and an auto-
matic pH controller (pH/ORP controller PC310, Suntex CO., Taipei, Taiwan) was used
for the conversion of HPMAE to (S)-PE. A reaction mixture (100 ml) containing 1 g
(wet weight) of IPTG-induced E. coli NovaBlue (pQE-aadh), 2% glucose, 20 mM KCl,
◦
2
00 C, 28 eV, and −9 V, respectively. The collision gas (nitrogen) was maintained
−5
at a pressure of 2.3 × 10 torr. The positive ion mode and multiple-reaction moni-
toring (MRM) mode were used to detect the HPMAE (m/z 166.1–148.1) and PE (m/z
1
68.1–150.1), respectively. Total analysis time was 5 min for each sample.
The chirality of PE was analyzed by HPLC with a chiral column (chiral-HPLC).
18 mM NH Cl, and various concentrations of HPMAE in 100 mM sodium phosphate
4
buffer (pH 6.5 or 7.0) was incubated at 30 ◦C. The consumption of HPMAE and the
Analytical HPLC was performed utilizing a Waters 2695 LC/Waters 996 Photodi-
ode Array detector system with a CYCLOBOND I 2000 AC column (3.2 × 250 mm,
Astec Inc., Whippany, NJ), the mobile phase consisted of methanol and 0.5%
sodium acetate (pH 5.5) at a ratio of 5:95 under the following conditions at
room temperature: flow rate of 0.7 ml/min, detection at 215 nm, scanning from
production of (S)-PE were measured using chiral-HPLC.
2
.6. Recovery of (S)-PE and Walden inversion reaction
Recovery of (S)-PE from the reaction mixture and conversion of (S)-PE to (R)-
PE were performed mainly according to the method described by Dorokhova et al.
[24]. Briefly, 200 ml solution containing 48 mM (S)-PE, which were prepared from
2
00 to 400 nm for compound identification, and retention times of 13.5 min for
HPMAE, 14.8 min for (R)-PE, and 18.6 min for (S)-PE. Concentrations of HPMAE,
S)-PE and (R)-PE in the reaction mixture were also measured by chiral-HPLC,
which was calibrated using several standard solutions of HPMAE, (S)-PE and
R)-PE, with their concentrations varied from 1 to 100 mM. The standard solu-
(
the reaction mixture containing 60 mM HPMAE, was centrifuged at 12,000 × g and
4 ◦C for 20 min. The supernatants were filtered with a 0.22 ␮m membrane. The pH
(
of the filtrate was adjusted to 10.5–11.5 using 10 N NaOH. As a result, the (S)-PE
precipitated out, which was collected by centrifugation and then washed with cold
water. Washed precipitates were dried by vacuum evaporator for Walden inversion
reaction. To perform the Walden inversion reaction, (S)-PE, acetic anhydride, and
tions were prepared freshly prior to each experiment. The values of enantiomeric
excess of (S)-PE were calculated using the equation: ee(S)-PE = [(S-enantiomer) − (R-
enantiomer)]/[(S-enantiomer) + (R-enantiomer)] × 100%.

W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
1531
sulfuric acid were mixed in a molar ratio of 1:15:1. The reaction mixture was heated
ever, HPLC-UV involves complicated purification procedures and
requires a long separation time to avoid interference from other
compounds. Recently, the HPLC–MS-MS method, which boasts
higher sensitivity and specificity relative to HPLC-UV, was estab-
lished for the determination of PE in human plasma [26]. In this
study, a simple and fast HPLC–MS-MS method was developed to
analyze HPMAE and PE in the reaction mixture.
◦
at 105 C for 12 h, diluted to 25% of the original concentration using water, and
◦
then incubated at 95 C for 3 h. Reaction products in the solution were recovered by
alkaline precipitation as described above. Chirality and ee value of the product were
determined by chiral-HPLC.
2.7. Sequence accession number
The nucleotide sequence of the RE AADH gene from R. erythropolis BCRC 10909
has been deposited in the GenBank nucleotide database under accession number
GQ280388.
In the preliminary experiment using the full scan mode for MS
+
analysis, the most abundant [M+H] ions for HPMAE and PE were
m/z 166 and 168.1, respectively, which were selected as the pre-
cursor ions in MS-MS analysis. As shown in Fig. 2, the product ion
spectra for HPMAE and PE were obtained by infusion in positive
ion mode and the major product ions of HPMAE and PE were m/z
3
. Results and discussion
3.1. Analysis of HPMAE and PE using HPLC–MS-MS and
1
48 and 150.1, respectively. The MS-MS transitions of m/z 166–148
chiral-HPLC
and m/z 168.1–150.1 then were selected for HPMAE and PE analysis
using MS-MS with the MRM mode. To determine the concentra-
tions of HPMAE and PE using HPLC–MS-MS analysis, the diluted
reaction mixtures were separated by a Polaris C-18A column and
introduced into MS-MS. The PE and HPMAE were eluted at retention
PE in pharmaceutical formulations usually is analyzed via the
HPLC-UV method with a detection limit of 120 ng/ml [25]. How-
Fig. 2. Positive electron spray ionization product ion mass spectra of HPMAE (A) and
PE (B) and mass chromatogram of HPMAE and PE (C). Standard solution (10 ␮M) of
HPMAE and PE in 100 mM sodium phosphate buffer (pH 7.0) was separated by a
Polaris C-18A column and detected under MRM mode (HPMAE: m/z 166–148; PE:
m/z 168.1–150.1). Mobile phase: 0.1% formic acid/methanol 85:15 (v/v). (D) Mass
chromatogram of PE converted from HPMAE by R. erythropolis BCRC 10909. The
Fig. 3. Chiral-HPLC analysis of a PE enantiomer converted from HPMAE. (A) HPMAE,
(R)-PE and (S)-PE standards. (B) Products recovered from reaction mixture by alka-
line precipitation. (C) Inversion of (S)-PE to (R)-PE. To convert HPMAE to (S)-PE, the
reaction mixture (200 ml) containing 100 mM phosphate buffer (pH 7.0), 2% glucose,
60 mM HPMAE and 1% (wet weight) of IPTG-induced E. coli NovaBlue (pQE-aadh)
screening reaction using 100 mM phosphate buffer (pH 7.0) containing 2% glucose,
◦
◦
1
0 mM HPMAE, and 1% (wet weight) of cells was performed at 28 C for 2 h.
was incubated at 30 C for 12 h. The resulting solution was used for the inversion of
(
S)-PE to (R)-PE.

1
532
W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
Fig. 4. Effect of pH on the stability of HPMAE (A) and (S)-PE (B). Buffer systems: (♦) sodium citrate (pH 4.0); (ꢀ) sodium citrate (pH 5.0); (ꢁ) sodium citrate (pH 6.0); (ꢁ)
sodium phosphate (pH 7.0); (ꢂ) sodium phosphate (pH 8.0); (᭹) Tris–HCl (pH 8.0); (ꢂ) Tris–HCl (pH 9.0). The reaction was performed in different buffers containing 10 mM
◦
HPMAE or (S)-PE at 30 C for 24 h.
times of 2.05 and 2.3 min in the MRM chromatogram, respectively
Fig. 2C). The detection limit for PE was 0.1 ␮M (16.8 ng/ml, signal-
to-noise ratio >3).
Agrobacterium, Bacillus, Corynebacterium, Deinococcus, Enterobacter,
Mesorhizobium, Pseudomonas, Rhodococcus and Streptomyces were
screened for their capacity to convert HPMAE to PE by HPLC–MS-MS
analysis, and only R. erythropolis BCRC 10909 exhibited a detectable
level of (S)-PE production (Fig. 2D).
(
Although HPLC–MS-MS method is fast and high sensitive, it
cannot determine the enantiomeric structure of PE. The absolute
configurations of (R)-PE and (S)-PE were therefore determined
by HPLC method using a CYCLOBOND I 2000 AC column. The
CYCLOBOND I 2000 AC column, which is based upon the acety-
lated beta cyclodextrin form, has the potential to resolve many
aromatic alcohols or amines that are chiral at the alpha or beta car-
bon [27–29]. According to the chromatogram of chiral-HPLC, the
retention times for HPMAE, (R)-PE, and (S)-PE were 13.5, 14.8, and
3.4. Cloning and expression of the RE AADH gene
+
NADP -dependent-l-1-amino-2 propanol dehydrogenase from
R. erythropolis MAK154 possesses the ability to convert (S)-MAK to
d-PDE, and the structure of HPMAE is similar to (S)-MAK. Thus, the
ORF of the R. erythropolis BCRC 10909 RE AADH gene was amplified
by the PCR method using primers that were designed based upon
the R. erythropolis MAK154 AADH gene. The amplified product was
cloned into pQE-30 and introduced into E. coli NovaBlue to evaluate
its capacity to convert HPMAE into PE.
The inserted DNA contained a 780-bp ORF encoding a 259-
residue protein with a predicated molecular mass of 26.58 kDa. The
DNA sequence of the RE AADH gene demonstrated 99.5% homol-
ogy with the R. erythropolis MAK154 AADH gene, with two amino
acid residues Ala119 and Val188 in RE AADH different from the
Thr119 and Gly188 in R. erythropolis MAK154 AADH. RE AADH
exhibited 34–64% sequence identities with short-chain dehydro-
genase reductase (SDR) from Mycobacterium vanbaalenii PYR-1;
1
8.6 min, respectively (Fig. 3A).
3.2. pH stability of substrate and product
The substrate, HPMAE, is synthesized under strong acid condi-
tions and stable in solutions with a low pH. However, when the
whole cells are used as a catalyst in the bioconversion reaction, the
reaction is usually performed at pH 5–8 [30–32]. Thus, the effects
of pH on the stabilities of HPMAE and PE were evaluated. As shown
in Fig. 4A, HPMAE was stable at pH 4 and completely degraded in
the buffer at pH higher than 8 after 24 h incubation. In contrast, PE
was stable at all pH levels examined (Fig. 4B). PE has been reported
◦
to be stable in condition of stress caused by heat (80 C), high acid
␤-ketoacyl-(acyl carrier protein) reductase (Protein Data Bank code
(
(
pH 2.0), high base (pH 10.0), UV radiation (254 nm) and oxidation
H O ) treatments. Except under oxidative stress, less than 1.3% of
2
NMO) from Streptomyces coelicolor A3(2) [36]; 3-oxoacyl-(acyl
2
2
carrier protein) reductase (Protein Data Bank code 2UVD) from
Bacillus anthracis (Ba3989) [37]; and ␤-ketoacyl-(acyl carrier pro-
tein) reductase (Protein Data Bank code 1I01) from E. coli [38]. SDRs
usually require a nicotinamide ribose, like NAD(H) or NADP(H),
as a coenzyme in enzyme reactions. Alignment of the RE AADH
PE is degraded over 24 h [33]. It is possible that the hydrogen of the
hydroxyl group in the benzene ring of HPMAE can be removed in
neutral or basic conditions following electron transfer between the
benzene ring and the ketone in the C␣ position, which may con-
tribute to the structural rearrangement and instability of HPMAE.
However, the ketone at C␣ can be reduced to hydroxyl group to
make PE stable.
+
sequence with those of some NADP -dependent SDR enzymes of
known three-dimensional structure revealed that the nucleotide-
binding motif (TGxxxGxG) for coenzyme binding [39] is conserved
in the N-terminal region (residues 13–20) of RE AADH (Fig. 5). The
conserved catalytic tetrad, Asn115-Ser143-Tyr157-Lys161, with
the YxxxK active site motif for the SDRs [40,41], also was identi-
fied in RE AADH. These results clearly imply that RE AADH should
3.3. Screening of bacterial strains capable of converting HPMAE
to PE
+
+
Several microorganisms have been demonstrated to possess the
be categorized among the NAD /NADP -dependent SDRs [42]. SDR
genes are widely distributed in the genomes of humans, animals,
plants and microorganisms, and are involved in the metabolism of
steroid hormones, prostaglandins, carbohydrates, fatty acids and
retinoids, as well as in the metabolism of xenobiotics, drugs and
capacity to reduce various kinds of ketone group to a hydroxyl
group with enantioselectivity [34,35]. However, the bioprocess
for the reduction of HPMAE to PE has never been reported.
In this study, about 100 strains of bacteria from the genera of

W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
1533
Fig. 5. Alignment of the RE AADH with the short-chain dehydrogenase/reductases with known three-dimensional structures, from Streptomyces coelicolor (SC ACP), Bacillus
anthracis Ba3989 (BA ACP), E. coli (EC FabG) and Lactobacillus brevis (LB RADH). The glycine-rich consensus sequence (TGxxxGxG) is shaded. The catalytic tetrad (Asn115,
Ser143, Tyr157 and Lys161) is depicted on a black background.
carcinogens [43]. However, the physiological roles of this enzyme
in R. erythropolis BCRC 10909 remain to be elucidated.
interacts with small amino acid residues, like Gly37 in LB RADH
[41]. The Ala39 residue of RE AADH, corresponding to the Gly37
of LB RADH, also is a small amino acid, suggesting that RE AADH
+
could be categorized as a NADP -dependent SDR.
3.5. Expression of RE AADH gene in E. coli
For high-level expression of the RE AADH gene, the ORF of the
RE AADH gene was amplified by PCR and cloned into the E. coli
expression vector pQE-30 to generate pQE-aadh. The recombinant
E. coli NovaBlue (pQE-aadh) was incubated in LB medium contain-
◦
ing ampicillin and induced with 0.5 mM IPTG at 30 C for 6 h. The
6
×His-tagged enzymes were purified to homogeneity by metal-
chelate affinity chromatography on a Ni-NTA column. SDS-PAGE
analysis of the crude extract and purified protein revealed a major
protein band corresponding to a molecular mass of approximately
3
0 kDa (Fig. 6), comparing well with the calculated mass (28 kDa)
of the 6×His-tagged translated product of the corresponding gene.
The purified recombinant RE AADH catalyzed the conversion of
HPMAE to (S)-PE with a specific activity of 0.19 ␮U/mg protein in
the presence of NADPH, while no conversion activity was detected
when NADH was used as the cofactor. The chiral-HPLC analysis
revealed that the conversion product was (S)-PE with more than
+
9
9% ee. Structural study of NADP -dependent (R)-specific alcohol
dehydrogenase from Lactobacillus brevis (LB RADH) has shown that
the residue Gly37 plays a critical role in determining the prefer-
+
+
ence of SDRs for NADP [41]. In several NAD -dependent SDRs, the
residue corresponding to Gly37 of LB RADH is highly conserved
with aspartate, and the negative side-chain of aspartate forms
ꢀ
hydrogen bonds with the 2 -hydroxyl group of the adenine ribose
Fig. 6. SDS-PAGE analysis of RE AADH purified from E. coli NovaBlue (pQE-aadh) by
Ni-NTA column. Lane M: protein markers; lane C: crude extract; lane 1: wash flow;
lane 2: purified RE AADH.
moiety of NAD+ [44,45]. In NADP -dependent SDR enzymes, the
protruding phosphor moiety of NADP occupies more space and
+
+

1
534
W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
Fig. 7. Effect of pH on the cell enzyme activity of IPTG-induced E. coli NovaBlue
(
pQE-aadh) cells. Symbols: (ꢀ) Sodium citrate buffer; (ꢂ) sodium phosphate buffer;
(
ꢂ) Tris–HCl buffer. The reaction was performed in 100 mM buffer containing 2%
Fig. 8. Effect of pH on the synthesis of (S)-PE from HPMAE using E. coli NovaBlue
(pQE-aadh) as catalysts. Symbols: (ꢀ) HPMAE at pH 6.5; (ꢁ) (S)-PE at pH 6.5; (ꢂ)
HPMAE at pH 7; (ꢁ) (S)-PE at pH 7. A reaction mixture (10 ml) containing 0.1 g E.
coli NovaBlue (pQE-aadh) cells (wet wt.), 10 mM HPMAE and 2% glucose in 100 mM
glucose, 10 mM HPMAE, and 1% (wet weight) of IPTG-induced E. coli NovaBlue (pQE-
◦
aadh) at 30 C for 8 h.
◦
phosphate buffer was incubated at 30 C.
3.6. Production of PE by recombinant E. coli cells
Since RE AADH requires NADPH as a cofactor for its reduc-
tion activity, whole cells that express RE AADH instead of purified
RE AADH would be the better choice for the bioconversion process,
as NADPH may be continuously supplied by catabolic pathways of
the cells during the course of the reaction [15]. To evaluate the
feasibility of using whole cells expressing RE AADH in the produc-
tion of PE from HPMAE, the effect of pH on the RE AADH activity
of IPTG-induced E. coli NovaBlue (pQE-aadh) was determined. As
shown in Fig. 7, the optimal pH for reduction was 7.0 in the sodium
phosphate buffer, and the specific activity of whole cells expressing
RE AADH was 0.42 U/g wet cells.
To determine the effect of the substrate concentration on (S)-PE
production, the HPMAE concentration was increased up to 70 mM
and the pH of the reaction mixture maintained at 7.0 (Fig. 10). As
shown in Fig. 10B, a maximal concentration of (S)-PE (46.8 mM)
and productivity of 3.9 mmol/L h with a conversion yield of 78%
could be obtained from 60 mM HPMAE in a 12 h reaction. When
the concentration of HPAME was increased to 70 mM, the viable
10
cell count of recombinant E. coli was decreased from 1.6 × 10 to
6
9
× 10 CFU/ml and the production of (S)-PE was also decreased
significantly (Fig. 10C), indicating a toxicity of high HPMAE concen-
tration to E. coli cells. Therefore, continuous feeding to maintain an
appropriate substrate concentration of HPMAE in the reaction mix-
ture would be a preferred strategy to optimize (S)-PE production.
To determine the enantioselectivity of the RE AADH, the reac-
tion mixture was centrifuged and filtered to remove cells and
debris. The conversion products in the supernatant were analyzed
using chiral-HPLC. The retention time of the examined PE was
1
8.8 min, indicating that the PE obtained from our biocatalysis pro-
cess was S-form.
RE AADH activity of recombinant E. coli cells was most active
at pH 7 in the sodium phosphate buffer and decreased more than
5
0% at pH below 6 in the sodium citrate buffer (Fig. 7). However,
HPMAE was unstable at the pH higher than 7.0 (Fig. 4A), but consid-
erable stable at pH below 6.0. Therefore, the production of (S)-PE
from HPMAE using IPTG-induced E. coli NovaBlue (pQE-aadh) was
performed at pH 6.5 and 7.0 in 100 mM phosphate buffer. As shown
in Fig. 8, about 7.5 mM PE was obtained in a 4-h reaction at pH 7.0,
while only about 6.0 mM PE was obtained at pH 6.5. Obviously, pH
7
.0 was better than pH 6.5 for the conversion of HPMAE to (S)-PE by
cells expressing RE AADH. We also found that the pH of the reac-
tion mixture decreased over the course of the reaction (data not
shown). To obtain a high yield of (S)-PE using a whole cell reac-
tion system, the reaction mixture should be kept at a desired pH
level. Therefore, a 250 ml glass vessel equipped with a propeller-
type impeller and pH controller was used for (S)-PE production.
The reaction was performed using 1% (wet weight) E. coli cells that
◦
express RE AADH and 10 mM HPMAE at pH 7.0 and 30 C. As shown
in Fig. 9, the conversion of HPMAE to (S)-PE could be significantly
improved by using a reaction mixture with pH control, and a pro-
ductivity of 1.88 mmol PE/L h, with a conversion yield of about 75%,
can be obtained in a 4-h reaction at pH 7.0. Obviously, cell enzyme
activity would benefit from pH control, though maintaining a pH
of 7.0 might cause HPMAE degradation.
Fig. 9. Synthesis of (S)-PE from HPMAE using E. coli NovaBlue (pQE-aadh) cells as
catalysts. Symbols: (ꢀ) HPMAE consumption without pH control; (ꢁ) (S)-PE pro-
duction without pH control; (ꢂ) HPMAE consumption with pH control at 7; (ꢁ)
(
S)-PE production with pH control at 7. A reaction mixture (100 ml) containing 1 g E.
coli NovaBlue (pQE-aadh) cells (wet wt.), 10 mM HPMAE and 2% glucose in 100 mM
◦
phosphate buffer (pH 7) was incubated at 30 C.

W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
1535
Fig. 10. Effect of HPMAE concentration on (S)-PE production by E. coli NovaBlue (pQE-aadh) cells. A reaction mixture (100 ml) containing 1 g IPTG-induced E. coli NovaBlue
◦
(
pQE-aadh) cell (wet wt.), 2% glucose and different concentrations of HPMAE in 100 mM phosphate buffer (pH 7) was incubated at 30 C. Reaction pH was maintained at 7.0
by a pH controller. Initial HPMAE concentrations: (A) 40 mM, (B) 60 mM, and (C) 70 mM. Symbols: (ꢂ) HPMAE consumption; (ꢁ) (S)-PE production.
3
.7. Conversion of (S)-PE to (R)-PE by Walden inversion reaction
resistance to the high concentration of HPMAE, might lead to a
bioconversion process with high yield of (S)-PE. Nevertheless, we
found that about 20% of (S)-PE was lost in the alkaline precipita-
tion step for the inversion of (S)-PE to (R)-PE. To avoid the need for
Walden inversion reaction, experiments are currently in progress
to develop a process for direct conversion of HPMAE to (R)-PE using
other amino alcohol dehydrogenases.
A reaction mixture (200 ml) containing 2 g (wet weight) of IPTG-
induced E. coli NovaBlue (pQE-aadh), 2% glucose, 20 mM KCl, 18 mM
NH Cl and 60 mM HPMAE in 100 mM sodium phosphate buffer (pH
7
4
◦
.0) was incubated at 30 C for 12 h. To perform Walden inversion,
recombinant E. coli cells in 200 ml resulting solution containing
4
8
8 mM of (S)-PE (1.60 g) were removed by centrifugation. About
1% of (S)-PE (1.42 g) was recovered from supernatant by alkaline
Acknowledgment
precipitation (Fig. 3B). Resulting (S)-PE was subjected to inversion
reaction. The (R)-PE was obtained in 99% reaction yield (1.41 g) and
more than 99% ee (Fig. 3C). The overall yield of Walden inversion
was about 80%, in which the decrease in the yield is almost resulted
from the process of alkaline precipitation of (S)-PE from the reaction
mixture.
This work was supported by a research grant of NSC95-2313-
B005-005 from National Science Council, Taiwan.
References
[
[
[
1] Hanna PE. Adrenergic agents. In: Delgado JN, Remers WA, editors. Wilson and
Gisvold’s textbook of organic medicinal and pharmaceutical chemistry. 9th ed.
Philadelphia: Lippincott-Raven; 1991. p. 413–42.
4
. Conclusion
2] Westfall TC, Westfall DP. Adrenergic agonists and antagonists. In: Brunton LL,
Lazo JS, Parker KL, editors. Goodman and Gilman’s the pharmacological basis
of therapeutics. 11th ed. New York: McGraw-Hill; 2006. p. 237–95.
Our study has shown that E. coli expressing RE AADH can be used
as biocatalyst for the enantioselective synthesis of (S)-PE (ee > 99%)
from HPMAE, and resulting (S)-PE can be converted to give a reac-
tion yield of 81% and an ee of >99% of (R)-PE using the Walden
inversion reaction. Several improvements, such as the protein engi-
neering of RE AADH to increase biocatalytic activity of recombinant
E. coli cells and the screening of recombinant E. coli cells with
3
3] Hengstmann JH, Goronzy J. Pharmacokinetics of H-phenylephrine in man. Eur
J Clin Pharmacol 1982;21:335–41.
[4] Kanfer I, Dowse R, Vuma V. Pharmacokinetics of oral decongestants. Pharma-
cotherapy 1993;13:116S–28S.
[
5] Hoffman BB. Adrenoceptor—activating and other sympathomimetic drugs. In:
Katzung BG, editor. Basic & clinical pharmacology. 10th ed. New York: McGraw-
Hill; 2007. p. 121–40.

1
536
W.-D. Lin et al. / Process Biochemistry 45 (2010) 1529–1536
[
6] Tanne JH. Methamphetamine epidemic hits middle America. BMJ
[28] Warnke MM, Mitchell CR, Rozhkov RV, Emrich DE, Larock RC, Armstrong DW.
Use of native and derivatized cyclodextrin based and macrocyclic glycopeptide
based chiral stationary phases for the enantioseparation of pterocarpans by
HPLC. J Liq Chromatogr Relat Technol 2005;28:823–34.
2
006;332:382.
[
[
7] Coombes R. Cold Turkey. BMJ 2007;334:1190–2.
8] Bergmann ED, Sulzbacher M. A new synthesis of 1-(m- and p-hydroxyphenyl)-
2
-methylaminoethanol (m- and p-sympathol). J Org Chem 1951;16:84–9.
[29] Huang K, Breitbach ZS, Armstrong DW. Enantiomeric impurities in chi-
ral synthons, catalysts, and auxiliaries: Part 3. Tetrahedron: Asymmetry
2006;17:2821–32.
[30] Katz M, Frejd T, Hahn-Hägerdal B, Gorwa-Grauslund MF. Efficient anaerobic
whole cell stereoselective bioreduction with recombinant Saccharomyces cere-
visiae. Biotechnol Bioeng 2003;84:573–82.
[31] Hsu SK, Lo HH, Kao CH, Lee DS, Hsu WH. Enantioselective synthesis of l-
homophenylalanine by whole cells of recombinant Escherichia coli expressing
l-aminoacylase and N-acylamino acid racemase genes from Deinococcus radio-
durans BCRC12827. Biotechnol Prog 2006;22:1578–84.
[
9] Russell PB, Childerss ST. New route to phenylephrine. J Pharm Sci 1961;50:713.
[
[
10] Britten AZ. A new route to d,l-phenylephrine. Chem Ind 1968;24:771–2.
11] Takeda H, Tachinami T, Aburatani M, Takahashi H, Morimoto M, Achiwa K. Prac-
tical asymmetric synthesis of (R)-(−)-phenylephrine hydrochloride catalyzed
by (2R,4R)-MCCPM-rhodium complex. Tetrahedron Lett 1989;30:367–70.
12] Klinger FD, Wolter L, Dietrich W. Method for preparing of l-phenylephrine
hydrochloride. U.S. Patent 6,187,956 (2001).
[
[
[
[
[
13] Pandey RK, Upadhyay PK, Kumar P. Enantioselective synthesis of (R)-
phenylephrine hydrochloride. Tetrahedron Lett 2003;44:6245–6.
14] Kumar P, Naidu V, Gupta P. Application of hydrolytic kinetic resolution (HKR)
in the synthesis of bioactive compounds. Tetrahedron 2007;63:2745–85.
15] De Wildeman SM, Sonke T, Schoemaker HE, May O. Biocatalytic reductions:
from lab curiosity to “first choice”. Acc Chem Res 2007;40:1260–6.
16] Kroutil W, Mang H, Edegger K, Faber K. Recent advances in the biocatalytic
reduction of ketones and oxidation of sec-alcohols. Curr Opin Chem Biol
[32] Lee YC, Chien HC, Hsu WH. Production of N-acetyl-d-neuraminic acid by
recombinant whole cells expressing Anabaena sp. CH1 N-acetyl-d-glucosamine
2-epimerase and Escherichia coli N-acetyl-d-neuraminic acid lyase. J Biotechnol
2007;129:453–60.
[33] Marín A, Barbas C. LC/MS for the degradation profiling of cough-cold products
under forced conditions. J Pharm Biomed Anal 2004;35:1035–45.
[34] Kataoka M, Kita K, Wada M, Yasohara Y, Hasegawa J, Shimizu S. Novel biore-
duction system for the production of chiral alcohols. Appl Microbiol Biotechnol
2003;62:437–45.
2
004;8:120–6.
[
[
17] Patel R, Hanson R, Goswami A, Nanduri V, Banerjee A, Donovan MJ, et al. Enzy-
matic synthesis of chiral intermediates for pharmaceuticals. J Ind Microbiol
Biotechnol 2003;30:252–9.
18] Hsin LW, Dersch CM, Baumann MH, Stafford D, Glowa JR, Rothman RB, et
al. Development of long-acting dopamine transporter ligands as potential
cocaine-abuse therapeutic agents: chiral hydroxyl-containing derivatives of
[35] Nakamura K, Yamanaka R, Matsuda T, Harada T. Recent developments in
asymmetric reduction of ketones with biocatalysts. Tetrahedron: Asymmetry
2003;14:2659–81.
[36] Tang Y, Lee HY, Tang Y, Kim CY, Mathews I, Khosla C. Structural and func-
tional studies on SCO1815: a beta-ketoacyl-acyl carrier protein reductase from
Streptomyces coelicolor A3(2). Biochemistry 2006;45:14085–93.
[37] Zaccai NR, Carter LG, Berrow NS, Sainsbury S, Nettleship JE, Walter TS, et al.
Crystal structure of a 3-oxoacyl-(acylcarrier protein) reductase (BA3989) from
Bacillus anthracis at 2.4-Å resolution. Proteins 2008;70:562–7.
[38] Price AC, Zhang YM, Rock CO, White SW. Structure of beta-ketoacyl-[acyl car-
rier protein] reductase from Escherichia coli: negative cooperativity and its
structural basis. Biochemistry 2001;40:12772–81.
[39] Oppermann U, Filling C, Hult M, Shafqat N, Wu X, Lindh M, et al. Short-
chain dehydrogenases/reductases (SDR): the 2002 update. Chem Biol Interact
2003;143–144:247–53.
[40] Filling C, Berndt KD, Benach J, Knapp S, Prozorovski T, Nordling E, et al. Critical
residues for structure and catalysis in short-chain dehydrogenases/reductases.
J Biol Chem 2002;277:25677–84.
[41] Schlieben NH, Niefind K, Müller J, Riebel B, Hummel W, Schomburg D. Atomic
resolution structures of R-specific alcohol dehydrogenase from Lactobacillus
brevis provide the structural bases of its substrate and cosubstrate specificity.
J Mol Biol 2005;349:801–13.
[42] Kavanagh KL, Jörnvall H, Persson B, Oppermann U. The SDR superfamily: func-
tional and structural diversity within a family of metabolic and regulatory
enzymes. Cell Mol Life Sci 2008;65:3895–906.
[43] Hoffmann F, Maser E. Carbonyl reductases and pluripotent hydroxysteroid
dehydrogenases of the short-chain dehydrogenase/reductase superfamily.
Drug Metab Rev 2007;39:87–144.
1
1
2
-[2-[bis(4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)piperazine and
-[2-(diphenylmethoxy)ethyl]-4-(3-phenylpropyl)piperazine.
002;45:1321–9.
J Med Chem
[
[
[
[
19] Zhu D, Yang Y, Hua L. Stereoselective enzymatic synthesis of chiral alcohols
with the use of a carbonyl reductase from Candida magnoliae with anti-Prelog
enantioselectivity. J Org Chem 2006;71:4202–5.
20] Lavandera I, Kern A, Ferreira-Silva B, Glieder A, de Wildeman S, Kroutil W. Stere-
oselective bioreduction of bulky-bulky ketones by a novel ADH from Ralstonia
sp. J Org Chem 2008;73:6003–5.
21] Itoh N, Matsuda M, Mabuchi M, Dairi T, Wang J. Chiral alcohol production by
NADH-dependent phenylacetaldehyde reductase coupled with in situ regener-
ation of NADH. Eur J Biochem 2002;269:2394–402.
22] Uehling DE, Donaldson KH, Deaton DN, Hyman CE, Sugg EE, Barrett DG, et al.
Synthesis and evaluation of potent and selective beta(3) adrenergic receptor
agonists containing acylsulfonamide, sulfonylsulfonamide, and sulfonylurea
carboxylic acid isosteres. J Med Chem 2002;45:567–83.
23] Kataoka M, Ishige T, Urano N, Nakamura Y, Sakuradani E, Fukui S, et al.
Cloning and expression of the l-1-amino-2-propanol dehydrogenase gene from
Rhodococcus erythropolis, and its application to double chiral compound pro-
duction. Appl Microbiol Biotechnol 2008;80:597–604.
24] Dorokhova MI, Smolina NE, Tikhonova Ya.O., Mikhalev VA. Inversion of con-
figuration of optically active 1-m-nitrophenyl-2-methylaminoethanols. Pharm
Chem J 1974;8:209–11.
25] Marín A, García E, García A, Barbas C. Validation of a HPLC quantification of
acetaminophen, phenylephrine and chlorpheniramine in pharmaceutical for-
mulations: capsules and sachets. J Pharm Biomed Anal 2002;29:701–14.
[
[
[
[
[44] van der Werf MJ, van der Ven C, Barbirato F, Eppink MH, de Bont JA, van Berkel
WJ. Stereoselective carveol dehydrogenase from Rhodococcus erythropolis
DCL14. A novel nicotinoprotein belonging to the short chain dehydroge-
nase/reductase superfamily. J Biol Chem 1999;274:26296–304.
[45] Pennacchio A, Pucci B, Secundo F, La Cara F, Rossi M, Raia CA. Purification and
characterization of a novel recombinant highly enantioselective short-chain
NAD(H)-dependent alcohol dehydrogenase from Thermus thermophilus. Appl
Environ Microbiol 2008;74:3949–58.
26] Ptácek P, Klíma J, Macek J. Development and validation of
a liquid
chromatography–tandem mass spectrometry method for the determination
of phenylephrine in human plasma and its application to a pharmacokinetic
study. J Chromatogr B Analyt Technol Biomed Life Sci 2007;858:263–8.
27] Armstrong DW, Lee JT, Chang LW. Enantiomeric impurities in chiral cata-
lysts, auxiliaries and synthons used in enantioselective synthesis. Tetrahedron:
Asymmetry 1998;9:2043–64.
[