Asymmetric synthesis of the chiral synthon ethyl (S)-4-chloro-3- hydroxybutanoate using Lactobacillus kefir

Article abstract of DOI:10.1016/j.tetasy.2005.01.013

Lactobacillus kefir was used as the whole cell biocatalyst for the asymmetric reduction of ethyl 4-chloro acetoacetate 1 to the chiral synthon ethyl (S)-4-chloro-3-hydroxybutanoate 2. Ketoester 1 was obtained as micro-droplets, without the use of an organic solvent as substrate reservoir. 2 (1.2 M) was produced using 2-propanol as co-substrate with a final yield of 97% within 14 h. A high space-time yield and a high specific product capacity of 85.7 mmol/L h and of 24 mmol/g_DCW were measured. The enantiomeric excess of the (S)-alcohol 2 was 99.5%.

Full text of DOI:10.1016/j.tetasy.2005.01.013

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 899–901
Asymmetric synthesis of the chiral synthon ethyl
(S)-4-chloro-3-hydroxybutanoate using Lactobacillus kefir
Maya Amidjojo and Dirk Weuster-Botz*
Lehrstuhl fu¨r Bioverfahrenstechnik, Technische Universita¨t Mu¨nchen, Boltzmannstr. 15, D-85748 Garching, Germany
Received 30 November 2004; accepted 10 January 2005
Abstract—Lactobacillus kefir was used as the whole cell biocatalyst for the asymmetric reduction of ethyl 4-chloro acetoacetate 1 to
the chiral synthon ethyl (S)-4-chloro-3-hydroxybutanoate 2. Ketoester 1 was obtained as micro-droplets, without the use of an
organic solvent as substrate reservoir. 2 (1.2 M) was produced using 2-propanol as co-substrate with a final yield of 97% within
14 h. A high space-time yield and a high specific product capacity of 85.7 mmol/Lh and of 24 mmol/gDCW were measured. The enan-
tiomeric excess of the (S)-alcohol 2 was 99.5%.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
2 with an enantiomeric excess >99%.9 Herein 15 mM
of CAAE 1 were converted with a yield of 100% into
the corresponding (S)-CHBE 2. This (S)-CHBE 2 con-
centration is rather low and thus not relevant for further
industrial applications.
Chiral C3- and C4- building blocks are important inter-
mediates for the synthesis of pharmaceuticals and fine
chemicals.1 Ethyl (S)-4-chloro-3-hydroxybutanoate 2
((S)-CHBE), for instance, is a key intermediate for
HMG-CoA reductase inhibitors, a compound class
which lowers the cholesterol level in human blood.2
We herein report the stereoselective reduction of 1 to 2
using wildtype L. kefir aiming at (S)-CHBE 2 concentra-
tions comparable to existing yeast processes (see Fig. 1).
Microbial transformations using whole cells as biocata-
lysts for these bioreductions have been well investigated.
For the stereoselective reduction of the prochiral keto-
ester ethyl 4-chloro acetoacetate (CAAE) 1 to the corre-
sponding (S)-CHBE 2 various processes using yeasts
have been published.1,3,4 Nevertheless, a main problem
with yeast bioreductions is the low enantiomeric excess
of the alcohols, due to many reductases with competitive
stereoselectivities present in the yeasts.5 To overcome
this problem various approaches have been suggested,
such as heat treatment of the cells or their preparation
with acetone,1,3,6 the use of a two-phase organic/aque-
ous-system,5 a slow release of the ketoester substrates,5
the selection of a suitable co-substrate,5 the addition
of enzyme inhibitors7 and the use of recombinant
microorganisms.4,8
Figure 1. Scheme of the asymmetric reduction of ethyl 4-chloro
acetoacetate 1 to ethyl (S)-4-chloro-3-hydroxybutanoate 2 by Lacto-
bacillus kefir with regeneration of the cofactor NADPH using 2-
propanol as co-substrate.
The wildtype bacterium Lactobacillus kefir was proven
to perform the asymmetric synthesis of the (S)-alcohol
2. Results and discussion
*
For the conversion of CAAE 1 (300 mM) the specific
product capacity and the enantiomeric excess were
Corresponding author. Tel.: +49 89 2891 5712; fax: +49 89 2891
5714; e-mail: d.weuster-botz@lrz.tum.de
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.01.013

900
M. Amidjojo, D. Weuster-Botz / Tetrahedron: Asymmetry 16 (2005) 899–901
compared at a co-substrate concentration of 450 mM
glucose and 20% v/v 2-propanol, respectively. The con-
version stopped with both co-substrates after 5 h. A
yield of 100% was measured using 2-propanol as co-sub-
strate. The yield with glucose as co-substrate was much
lower (22%). The same difference was observed with re-
spect to the specific product capacity (see Fig. 2). With
2-propanol as co-substrate (S)-CHBE 2 (5.7 mmol)
was formed by 1 g of the biocatalyst (dry cell weight,
DCW). The specific product capacity was much lower
(1.3 mmol/gDCW) for the conversion with glucose as
co-substrate. Furthermore the enantiomeric excess was
increased from 87% with glucose as co-substrate to
>99% if 2-propanol was used.
Figure 3. (a) Concentrations of CAAE 1 (h) and (S)-CHBE 2 (j) in a
stirred-tank reactor (potassium phosphate buffer, 0.2 M, pH 6.5, 5 mM
MgCl2, 30 ꢁC). CAAE 1 was added at t = 0 h and t = 3.5 h, respec-
tively. (b) Conversion rates (,) and estimation of the conversion rates
based on a first order decay of the Lactobacillus kefir cells (drawn line).
Figure 2. Comparison of the specific product capacity and the
enantiomeric excess of 2 for the cofactor regeneration with glucose
and 2-propanol, respectively.
Investigations with different 2-propanol concentrations
between 5% and 20% v/v and 450 mM CAAE 1 showed
that a 2-propanol concentration of 5% v/v was sufficient
to achieve a yield of 100% within 3 h.
Our process with L. kefir resulted in an improvement to
the final (S)-CHBE 2 concentration, as well as in stereo-
selectivity, if compared to yeast bioreductions per-
formed in organic/aqueous two-phase systems with
NADP addition. For the synthesis of 2 with the yeast
Candida magnoliae a final product concentration of
680 mM was reported at complete conversion. The
enantiomeric excess was 96%.3 Using a recombinant
Pichia pastoris strain, 350 mM of 1 were converted with
a yield of 91% at an enantiomeric excess of 95%.4
Based on these results the asymmetric reduction of
CAAE 1 to the corresponding (S)-alcohol 2 was carried
out in a stirred-tank reactor on a 200 mL-scale with 5%
v/v 2-propanol (see Fig. 3a). CAAE 1 was added in two
steps with a concentration of 620 mM at a time, due to
its instability in aqueous media and an assumed decrease
of stereoselectivity at too high ketoester 1 concentra-
tions.5 A final (S)-CHBE 2 concentration of 1.2 M and
an enantiomeric excess of 99.5% was achieved within
14 h. The final yield was 97%. The space-time yield
High stereoselectivity comparable to our L. kefir conver-
sion was reported using a recombinant E. coli with glu-
cose as co-substrate.8 Herein the final (S)-CHBE 2
concentration was 1.25 M in aqueous media and
2.58 M in an organic/aqueous two-phase system. Never-
theless the addition of NADP was necessary in both
cases, because the E. coli cells needed to be permeabi-
lized by a pretreatment with Triton X-100.
and specific product capacity were 85.7 mmol/(L h)
and 24 mmol/gDCW, respectively.
*
Figure 3b shows the conversion rates of the bioreduc-
tion. The drawn lines represent first order decay of the
biocatalyst. At the end of the first section, a CAAE 1
limitation could be observed. This is indicated by an in-
crease of the conversion rate after the second CAAE 1
dosage at 3.5 h. A half-life of 2.9 h was estimated
for L. kefir based on the measured (S)-CHBE 2
concentrations.
3. Conclusion
L. kefir was proven to be an effective wildtype biocata-
lyst for the asymmetric reduction of the ketoester 1 to
its corresponding (S)-alcohol 2 without making use of
an organic solvent. Compared to all the other reported
processes, no addition of cofactor was necessary,
due to the successful use of the intracellular cofactor
regeneration system with 2-propanol as co-substrate.
Compared to existing (S)-CHBE 2 syntheses the
Compared to the reference conversion with L. kefir re-
ported by Arragozini et al. the final concentration of 2
was improved 80-fold.9 Despite the often observed de-
creases in stereoselectivity for high ketoester concentra-
tions in the aqueous phase,5,10 no losses in stereo-
selectivity were observed with L. kefir.

M. Amidjojo, D. Weuster-Botz / Tetrahedron: Asymmetry 16 (2005) 899–901
901
bioreduction process by whole cells of L. kefir is thus
a competitive alternative considering the process para-
meters stereoselectivity, yield and final product
concentration.
AG, Bottmingen, Switzerland). Ketoester CAAE 1 was
provided as micro-droplets. The power input for the
bioreduction was 0.54 kW/m3. The estimated mean
droplet diameter for 1 was <43 lm. The conversion of
1 and the production of 2 was monitored by GC analysis
throughout the whole process. for all bioreductions per-
formed. The conversion rate was estimated based on the
measured (S)-CHBE 2 concentrations using the central
difference approximation.
4. Experimental
4.1. Chemicals
All chemicals were purchased from VWR/Merck except
ethyl 4-chloro 3-hydroxybutanoate [((R)- and (S)- enan-
tiomers)] used for analytical purposes, which were ob-
tained from Sigma-Aldrich.
4.4. Analysis
The concentrations of CAAE 1 and (S)-CHBE 2 and
(R)-CHBE were determined by chiral gas chromato-
graphy on a CP-3800 system (Varian, Darmstadt,
Germany) equipped with a Chiraldex-GTA column
(20 m · 0.25 mm ID, Astec Inc., Whippany, USA) and
an FID-detector (Varian, Darmstadt, Germany). Ethyl
acetate was used as the organic solvent for the extraction
of the components from the aqueous phase. For this
800 lL of ethyl acetate and 500 lL of glass beads
(0.2 mm diameter) were added to 400 lL of reaction sus-
pension. For the extraction, the samples were mixed
using a Thermomixer (Eppendorf AG, Hamburg. Ger-
many) for 15 min at 1400 minÀ1. After centrifugation
at 13,000 rpm for 5 min the ethyl acetate supernatant
was diluted in anhydrous ethyl acetate (1:5 v/v). The di-
luted ethyl acetate phase could then be used for GC
analytics.
4.2. Microorganism
The biocatalyst (L. kefir DSMZ 20587) was purchased
from Juelich Fine Chemicals, Juelich, Germany and
stored at À20 ꢁC.
4.3. Bioreductions
The bioreductions were performed with resting cells of
L. kefir. Prior to the conversions, the cells were washed
once in 0.2 M potassium phosphate buffer, pH 6.5. After
centrifugation at 4500 rpm for 10 min (Hettich GmbH
& Co.AG, Tuttlingen, Germany) the supernatant was
removed.
For the comparison of co-substrates potassium phos-
phate buffer with 0.2 M, pH 6.5, 5 mM MgCl2,
450 mM glucose and 20% v/v 2-propanol, respectively,
were used. The cell pellet was suspended in the buffer
(25 mL) in a way that 50 gDCW/L was achieved. The
reaction was started by adding 1 (1 mL). The bioreduc-
tions were performed at 30 ꢁC in stirred flasks (multi-
point magnetic stirrer, H + P Labortechnik AG,
Oberschleissheim, Germany). The conversion of 1 and
the production of 2 was monitored by GC analysis
throughout the whole process.
References
1. Shimizu, S.; Kataoka, M. Adv. Biochem. Eng./Biotechnol.
1999, 63, 109–123.
2. Thottathil, J. K.; Pendri, Y.; Li, W.-S.; Kronenthal, D.
United States Patent 1994; US5278313.
3. Yasohara, Y.; Kizaki, N.; Hasegawa, J.; Takahashi, S.;
Wada, M.; Kataoka, M.; Shimizu, S. Appl. Microbiol.
Biot. 1999, 51, 847–851.
4. Engelking, H.; Pfaller, R.; Wich, G.; Weuster-Botz, D.
Tetrahedron: Asymmetry 2004, 15, 3591–3593.
5. Sybesma, W. F. H.; Straathof, A. J. J.; Jongejan, J. A.;
Pronk, J. T.; Heijnen, J. J. Biocatal. Biotransform. 1998,
16, 95–134.
6. Patel, R. N. Enzyme Microb. Technol. 2002, 31, 804–
826.
7. Nakamura, K.; Inoue, K.; Ushio, K.; Oka, K.; Ohno, A.
Chem. Lett. 1987, 9, 679–682.
8. Kizaki, N.; Yasohara, Y.; Hasegawa, J.; Wada, M.;
Kataoka, M.; Shimizu, S. Appl. Microbiol. Biotechnol.
2001, 55, 590–595.
9. Arragozini, F.; Valenti, M.; Santaniello, E.; Ferraboschi,
P.; Grisenti, P. Biocatalysis 1992, 5, 332–352.
10. Kometani, T.; Yoshii, H.; Kitatsuji, E.; Nishimura, H.;
Matsuno, R. J. Ferment. Technol. 1993, 76, 33–37.
The investigations on initial 2-propanol concentrations
followed the same procedure. The same buffer (25 mL)
with 5% v/v to 20% v/v 2-propanol was applied. The
reduction was started by addition of 1.5 mL of 1.
The cells for the bioreduction with dual CAAE 1 addi-
tion were dissolved in 183.5 mL buffer as described be-
fore with 5% v/v 2-propanol. The reaction was started
by adding 16.5 mL of 1. The initial L. kefir concentra-
tion was adjusted to 50 gDCW/L. After 3.5 h, 16.5 mL
of the ketoester 1 was added again. The bioreductions
were performed at 30 ꢁC in a stirred tank reactor (Infors