Stereoselective reduction of ethyl 4-chloro acetoacetate with recombinant Pichia pastoris

Article abstract of DOI:10.1016/j.tetasy.2004.09.021

A recombinant strain of Pichia pastoris coexpressing the carbonyl reductase of Candida magnoliae and the glucose dehydrogenase of Bacillus subtilis was used for the stereoselective reduction of ethyl 4-chloro acetoacetate 1 to (S)-4-chloro-3-hydroxybutanoate 2. The bioreduction was performed in a two-phase reaction system with n-butyl acetate as organic solvent. Complete conversion of 350 mM 1 with a final yield of 91% and an enantiomeric excess of 95% was achieved by continuous feeding of cell suspension and ketoester 1.

Full text of DOI:10.1016/j.tetasy.2004.09.021

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 3591–3593
Stereoselective reduction of ethyl 4-chloro acetoacetate with
recombinant Pichia pastoris
Helge Engelking,a Rupert Pfaller,b Gunther Wichb and Dirk Weuster-Botza,*
¨
aLehrstuhl fu¨r Bioverfahrenstechnik, Technische Universita¨t Mu¨nchen, Boltzmannstraße 15, D-85748 Garching, Germany
bConsortium fu¨r Elektrochemische Industrie GmbH, Zielstattstraße 20, D-81379 Mu¨nchen, Germany
Received 19 August 2004; accepted 14 September 2004
Available online 22 October 2004
Abstract—A recombinant strain of Pichia pastoris coexpressing the carbonyl reductase of Candida magnoliae and the glucose dehy-
drogenase of Bacillus subtilis was used for the stereoselective reduction of ethyl 4-chloro acetoacetate 1 to (S)-4-chloro-3-hydroxy-
butanoate 2. The bioreduction was performed in a two-phase reaction system with n-butyl acetate as organic solvent. Complete
conversion of 350mM 1 with a final yield of 91% and an enantiomeric excess of 95% was achieved by continuous feeding of cell
suspension and ketoester 1.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Herein we report the stereoselective reduction of 1 to the
corresponding (S)-alcohol 2 using a recombinant strain
of Pichia pastoris (Fig. 1). The carbonyl reductase of
C. magnoliae and a glucose dehydrogenase derived from
Bacillus subtilis for improved cofactor regeneration were
coexpressed under control of the methanol induced
AOX promotor.
Optically active alcohols can easily be obtained by
microbial reduction of prochiral ketones. The stereose-
lective reduction of ethyl 4-chloro acetoacetate (CAAE)
1 to (S)-4-chloro-3-hydroxybutanoate [(S)-CHBE] 2, a
chiral building block in the pharmaceutical industry,
can be catalyzed by various yeasts.1,2 The enantiomeric
excess (ee) is often low due to many different stereoselec-
tive reductases present in yeasts.3,4 The use of recombi-
nant microorganisms holds great potential to overcome
these limitations.3,5
O
O
OH
O
Pichia pastoris
Cl
Cl
OEt
OEt
Ketoester 1
(S)-Alcohol 2
Ketoester 1 is unstable in aqueous media. A two-phase
organic-solvent reaction systems can be used to prevent
the decomposition of 1 by minimizing its concentration
in the aqueous phase.5–7 At the same time, the inhibitory
effect of the ketoester on the biocatalyst is reduced.
Figure 1. Reduction of ethyl 4-chloro acetoacetate 1 to (S)-4-chloro-3-
hydroxybutanoate 2 by recombinant Pichia pastoris.
2. Results and discussion
The yeast Candida magnoliae reduces ketoester 1 to the
corresponding alcohol 2 with an enantiomeric excess of
96%.8 This NADPH-depending carbonyl reductase with
(S)-directed stereoselectivity has been isolated, charac-
terized and expressed in E. coli.9 By coexpressing a glu-
cose dehydrogenase of Bacillus megaterium for cofactor
regeneration, the (S)-alcohol 2 was obtained with an ee
of 100% and a yield of 85%.9
The bioreduction of 1 was carried out in aqueous media
as well as in organic-solvent/water two-phase systems.
The addition of small amounts of NADP (0.1mM)
was necessary to achieve sufficient reduction rates. In
the aqueous reaction system, a final concentration of 2
of 21mM and an enantiomeric excess of 97% was
achieved within 24h. The reaction was limited by the
inhibitory effects on the biocatalyst and the decomposi-
tion of ketoester 1. To overcome these limitations, four
biphasic reaction systems were studied by adding water
immiscible organic solvents to the reaction medium.
*
Corresponding author. Tel.: +49 89 289 15712; fax: +49 89 289
15713; e-mail: d.weuster-botz@lrz.tum.de
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.09.021

3592
H. Engelking et al. / Tetrahedron: Asymmetry 15 (2004) 3591–3593
With n-butyl acetate as the organic solvent, product
concentration was increased by 50% when compared
to the reduction in an aqueous reaction system with
the enantiomeric excess reaching >98% (Fig. 2). (S)-
alcohol 2 (4.1mmol) was formed by 1g of biocatalyst
(dry cell weight).
suspension was fed for 20h. Complete conversion of 1
(350mM) was achieved within 24h. The yield of the
reaction and the enantiomeric excess was 91% and
95%, respectively. Compound 1 (5.8mmol) was formed
by 1g of biocatalyst (dry cell weight). This is an
improvement of 41% when compared to the batch reac-
tion. The yield was higher than reported for bioreduc-
tions with recombinant E. coli being 85%, while the
enantiomeric excess did not reach 100%.5
160
140
120
100
80
100
98
96
94
92
90
3. Conclusion
The recombinant yeast P. pastoris is a useful tool for the
stereoselective reduction of b-keto ester 1 to the corre-
sponding (S)-alcohol 2. The decomposition of the sub-
strate and the inhibitory effect on the biocatalyst were
significantly reduced in a two-phase reaction system
containing n-butyl acetate as the organic phase and by
continuously feeding substrate 1 and the biocatalyst.
60
TMBE
hexene
aqueous
ethyl acetate
n-butyl acetate
Figure 2. Effect of the organic solvents on the concentration of 2 and
the enantiomeric excess. The relative concentration of 2 is calculated
relative to the bioreduction in a monophasic aqueous reaction system.
4. Experimental
4.1. Chemicals
Previously published results have demonstrated that
the carbonylreductase is most stable in the presence of
n-butyl acetate when compared to other organic
solvents.10 Furthermore, n-butyl acetate shows good
extraction properties towards 1 and 2.
All chemicals were purchased from VWR/Merck except
yeast nitrogen base (YNB), which was obtained from
Difco.
4.2. Microorganism
The reduction efficiency was further improved by adding
6.5% v/v Triton X100 thus increasing the yield by 60%
compared to the aqueous reaction system. Triton X100
leads to a permeabilization of the cells and therefore
to an increased mass transfer.11 Furthermore, the sur-
factant reduces the equilibrium droplet size of the
n-butyl acetate in water and therefore increases the
interfacial area.
The methylotrophic yeast strain Pichia pastoris GC909
over-expressing the glucose dehydrogenase gene of
Bacillus subtilis and the carbonyl reductase gene of
Candida magnoliae, both under control of the AOX
promoter, was provided by the Consortium fur
¨
Munchen,
¨
Elektrochemische
Germany.
Industrie
GmbH,
Lysis was observed in the presence of n-butyl acetate. In
order to maintain the stability and activity of the biocat-
alyst cell suspension and substrate solved n-butyl acetate
were fed into the reactor.
4.3. Growth conditions
Cells were grown in 1L shaking flasks with 200mL com-
plex media (10g/L yeast extract, 20g/L peptone, 13.4g/L
yeast nitrogen base, 10g/L glycerol, 0.4mg/L biotin,
0.1mol/L potassium phosphate puffer, pH6.0) at 28ꢁC
for three days on a rotary shaker at 250rpm. The expres-
sion of recombinant proteins was induced by adding
methanol (2mL) every 24h for three days. The biomass
was collected by centrifugation.
The feeding rate of the substrate led to an accumulation
of 1 during the first 15h of the process (Fig. 3). The cell
400
300
200
100
0
200
150
100
50
Feed of11
4.4. Bioreductions
For batch reductions, cells were dissolved in 20mL reac-
tion buffer (1M glucose, 0.1M sodium chloride, 0.1M
potassium phosphate buffer, pH7.0 or 0.1mM NADP).
The reaction was started by adding 1 (0.5mL) and
1.5mL of ethyl acetate, n-butyl acetate, tert-butyl
methyl ether (TBME) or hexane, respectively. Flasks
were incubated on a magnetic stirrer at 30ꢁC for 24h.
0
0
6
12
Time [h]
18
24
Figure 3. Concentration of CAAE 1, (R)-CHBE and (S)-CHBE 2 in a
fedbatch reduction. CAAE was fed for 15h and cell suspension for
20h; dotted line: expected course of concentrations.
For fedbatch bioreductions, cells were dissolved in reac-
tion buffer as described above to a final volume of

H. Engelking et al. / Tetrahedron: Asymmetry 15 (2004) 3591–3593
3593
180mL and 60g/L dry cell weight. Sixty millilitres of the
References
suspension was used as starting volume and 120mL of
the suspension continuously fed for 22h at 5.5mL/h
using a cassette pump (Watson-Marlow, England).
Ketoester 1 (10mL) was added to n-butyl acetate
(30mL) and continuously fed for 15h (2.6mL/h) using
a parallel titration system (FedBatch Pro, Dasgip, Ger-
many). The pH was controlled at pH7.0 by adding 5M
NaOH. The reaction vessel was magnetically stirred and
incubated at 30ꢁC for 24h. NADP was added to a final
concentration of 0.1mmol/L after 6h.
1. Shimizu, S.; Kataoka, M.; Kita, K. Appl. Environ.
Microbiol. 1990, 56, 2374–2377.
2. Kita, K.; Kataoka, M.; Shimizu, S. J. Biosci. Bioeng. 1999,
88, 591–598.
´
3. Rodrıguez, S.; Kayser, M. M.; Steward, J. D. J. Am.
Chem. Soc. 2001, 123, 1547–1556.
4. Katz, M.; Frejd, T.; Hahn-Ha¨gerdal, B.; Gorwa-Grausla-
und, M. F. Biotechnol. Bioeng. 2003, 84, 573–582.
5. Kizaki, N.; Yasohara, Y.; Hasegawa, J.; Wada, M.;
Kataoka, M.; Shimizu, S. Appl. Microbiol. Biotechnol.
2001, 5, 590–595.
6. Nakamura, K.; Kondo, S.; Kawai, Y.; Ohno, A. B. Chem.
Soc. Jpn. 1993, 66, 2738–2743.
7. Rotthaus, O.; Kruger, D.; Demuth, M.; Schaffner, K.
¨
4.5. Analytics
The reaction mixture (0.5mL) was extracted by adding
ethyl acetate (1mL), phosphoric acid (0.1mL) (45% v/
v) and 0.2mL glass beads (0.5mm in diameter). The sam-
ples were mixed in a ball mill (Mixer Mill MM200,
Tetrahedron 1997, 53, 935–938.
8. Yasohara, Y.; Kizaki, N.; Hasegawa, J.; Takahashi, S.;
Wada, M.; Kataoka, M.; Shimizu, S. Appl. Microbiol.
Biotechnol. 1999, 5, 847–851.
9. Wada, M.; Kataoka, M.; Kawabata, H.; Yasohara, Y.;
Kizaki, N.; Hasegawa, J.; Shimizu, S. Biosci. Biotechnol.
Biochem. 1998, 62, 280–285.
10. Shimizu, S.; Kataoka, M.; Katoh, M.; Morikawa, T.;
Miyoshi, T.; Yamada, H. J. Mol. Catal. B 1998, 5, 321–
325.
11. Laouar, L.; Lowe, K. C.; Mulligan, B. J. Enzyme Microb.
Technol. 1996, 18, 433–438.
Retsch GmbH, Haan, Germany) for 10min at 30sÀ1
.
The samples were centrifuged at 12,000g for 5min for
separating the phases. The supernatant was diluted in
anhydrous ethyl acetate (1:5v/v). The concentrations of
ketoester 1, (S)-alcohol 2 and (R)-alcohol were measured
using chiral gas chromatography (Lipodex E column,
25m · 0.32mm i.d., Macherey Nagel, Duren, Germany).
¨
Acetophenone was used as the internal standard.