Process intensification for substrate-coupled whole cell ketone reduction by in situ acetone removal

Article abstract of DOI:10.1021/op700055e

Three different reactor configurations for in situ acetone removal in whole cell biotransformation processes with substrate-coupled cofactor regeneration were applied. The reduction of 2,5-hexanedione to the corresponding (2R,5R)-hexanediol was catalyzed by recombinant Escherichia coli cells expressing an alcohol dehydrogenase from Lactobacillus brevis. The reaction was carried out in a substrate-coupled cofactor regeneration approach using 2-propanol as redox equivalent for intracellular cofactor regeneration. In contrast to a process without acetone removal, where 54% yield could be reached, the yield was increased to >90% when a pervaporation system was applied or when acetone was removed by sparging air through the reaction mixture. In a third system, conversion was driven using a biphasic system to extract acetone continuously from the biocatalyst containing aqueous phase and to allow high concentrations of the hydrophobic substrate 1-phenyl-2-propanone. When methyl tert-butyl ether was applied as the non-aqueous phase, only 24% yield was achieved. When the ionic liquid 1-butyl-3-methylimidazolium bis((trifluoromethyl)sulfonyl)amide was applied as the non-aqueous phase, >95% yield was reached as a result of the preferential partitioning behaviour of acetone over 2-propanol into the ionic liquid.

Full text of DOI:10.1021/op700055e

Organic Process Research & Development 2007, 11, 836–841
Process Intensification for Substrate-Coupled Whole Cell Ketone Reduction by In
Situ Acetone Removal
Kirsten Schroer, Eva Tacha, and Stephan Lütz*
Institute of Biotechnology 2, Research Centre Jülich, 52425 Jülich, Germany
Abstract:
ketones with remarkable chemo-, regio-, and enantioselectivity.
In comparison to isolated enzymes, whole cell applications have
several advantages. Whole cell biocatalysts are usually more
stable due to the protective cell matrix envelope for the enzyme.
Furthermore, there is no need for enzyme purification when
Three different reactor configurations for in situ acetone removal
in whole cell biotransformation processes with substrate-coupled
cofactor regeneration were applied. The reduction of 2,5-hex-
anedione to the corresponding (2R,5R)-hexanediol was catalyzed
by recombinant Escherichia coli cells expressing an alcohol
dehydrogenase from Lactobacillus breWis. The reaction was carried
out in a substrate-coupled cofactor regeneration approach using
7
whole cells are used as biocatalysts.
ADHs are dependent on cofactors (NADH or NADPH) that
act as hydride donors in the reduction of ketones. Since these
8
cofactors are too expensive to be used stoichiometrically, there
2-propanol as redox equivalent for intracellular cofactor regenera-
is a significant interest in developing efficient cofactor regenera-
tion processes. The regeneration of cofactors can be carried out
in an enzyme-coupled or in a substrate-coupled approach. The
enzyme-coupled approach is characterized by the use of a
second enzyme, which catalyzes the oxidation of a cosubstrate
to regenerate the reduced form of the cofactor. Formate
dehydrogenase or glucose dehydrogenase can be applied as the
second enzyme for cofactor regeneration. In the substrate-
coupled approach the alcohol dehydrogenase that catalyzes the
reduction of the prochiral ketone to a chiral alcohol also
catalyzes the cofactor-regenerating reaction by oxidation of a
second cosubstrate. In most cases 2-propanol is used as
tion. In contrast to a process without acetone removal, where 54%
yield could be reached, the yield was increased to >90% when a
pervaporation system was applied or when acetone was removed
by sparging air through the reaction mixture. In a third system,
conversion was driven using a biphasic system to extract acetone
continuously from the biocatalyst containing aqueous phase and
to allow high concentrations of the hydrophobic substrate 1-phen-
yl-2-propanone. When methyl tert-butyl ether was applied as the
non-aqueous phase, only 24% yield was achieved. When the ionic
liquid 1-butyl-3-methylimidazolium bis((trifluoromethyl)sulfony-
l)amide was applied as the non-aqueous phase, >95% yield was
reached as a result of the preferential partitioning behaviour of
acetone over 2-propanol into the ionic liquid.
9,10
cosubstrate, which is oxidized to acetone.
The substrate-coupled approach leads to a thermodynamic
equilibrium between all four components since all reactions are
reversible. Thus the maximum yield is limited by the thermo-
dynamic potentials of the compounds. Furthermore acetone may
have a harmful effect on the activity and the stability of the
biocatalysts. To overcome such limitations, in situ (co)product
Introduction
Chiral compounds are important building blocks in the
chemical and pharmaceutical industry for the production of, for
example, chemical catalysts, liquid crystals, flavours, agro-
chemicals, or drugs. In particular, chiral alcohols are of interest
11
removal techniques can be applied.
Different strategies for in situ acetone removal have already
been published and are part of processes carried out in industrial
scale. Due to its low vapour pressure, acetone can easily be
stripped from a reactor system by gassing the reaction mixture
with pressurized air or any other inert gas. It has already been
applied in ketone-reducing processes catalyzed by isolated
1
as building blocks. There is a wide range of methods for the
production of chiral alcohols described in literature. One
chemical method is the use of chiral boranes as reagents for
the enantioselective reduction of ketones.2,3 Furthermore, chemi-
cal transition metal catalysts are applied for the asymmetric
4
synthesis of chiral alcohols.
12
enzymes. The applicability of this strategy for acetone removal
Biological methods apply isolated enzymes or whole resting
cells as biocatalysts for the synthesis of chiral alcohols. In
particular alcohol dehydrogenases (ADHs) have gained in-
creased interest for the commercial production of chiral
13
has also be shown for processes catalyzed by whole cells.
(
6) Liese, A.; Seelbach, K.; Buchholz, A.; Haberland, J. Processes. In
Industrial Biotransformations; Liese, A., Seelbach, K., Wandrey, C.,
Eds.; Wiley-VCH: Weinheim, 2006; pp 147–513.
5,6
alcohols. These biocatalysts, used as isolated enzymes or
whole cells, catalyze the stereoselective reduction of prochiral
(
7) Wichmann, R.; Vasic-Racki, D. AdV. Biochem. Eng. Biotechnol. 2005,
9
2, 225–260.
(
8) Wandrey, C. Chem. Rec. 2004, 4, 254–265.
*
To whom correspondence should be addressed. Phone: (+49) (0) 2461-
(9) Hummel, W.; Kula, M. R. Eur. J. Biochem. 1989, 184, 1–13.
(10) Findrik, Z.; Vasic-Racki, D.; Lütz, S.; Daussmann, T.; Wandrey, C.
Biotechnol. Lett. 2005, 27, 1087–1095.
6
1-4388. E-mail: s.luetz@fz-juelich.de.
(
(
(
(
1) Daussmann, T.; Lütz, S. CHEManager Eur. 2006, 3, 8.
2) Hirao, A. Chem. Commun. 1981, 7, 315–317.
(11) Lye, G. J.; Woodley, J. M. Trends Biotechnol. 1999, 17, 395–402.
(12) Stillger, T.; Bönitz, M.; Villela, M.; Liese, A. Chem- Ing.-Tech. 2002,
74, 1035–1039.
3) Corey, E. J. Am. Chem. Soc. 1987, 109, 7925–7926.
4) Tang, W.; Zhang, X. Chem. ReV. 2003, 103, 3029–3070.
(5) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.; Sturmer,
(13) Goldberg, K.; Edegger, K.; Kroutil, W.; Liese, A. Biotechnol. Bioeng.
2006, 95, 192–198.
R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43, 788–824.
8
36
•
Vol. 11, No. 5, 2007 / Organic Process Research & Development
10.1021/op700055e CCC: $37.00  2007 American Chemical Society
Published on Web 08/08/2007

Figure 1. Reduction of 2,5-hexanedione with substrate-coupled cofactor regeneration.
Julich Chiral Solutions and Wacker Chemistry have employed
reduced pressure systems to remove the coproduct acetone that
is formed during the synthesis of ꢀ-keto esters such as (R)-
1,14
methyl hydroxybutyrate and (R)-ethyl hydroxybutyrate. Also,
acetone removal by pervaporation has been described for
12
biotransformation processes catalyzed by isolated enzymes.
One problem of using biocatalysts for the reduction of
prochiral ketones is the low solubility of many ketones in
aqueous solutions. For that reason, only low substrate concen-
trations can be applied to the biocatalysts when the reaction is
carried out in an aqueous environment. A straightforward
15,16
solution to this problem is the application of biphasic systems.
In such a system the biocatalysts are located in the aqueous
phase while the hydrophic substrate is present in a high
concentration in the non-aqueous phase and in small concentra-
tions in the aqueous phase. While there is a yield of substrate
in the aqueous phase catalyzed by the cells, new substrate is
extracted continuously from the non-aqueous into the aqueous
phase. The non-aqueous phase can be used not only as a
substrate reservoir but also for the continuous extraction of
products or acetone in processes with substrate coupled cofactor
regeneration, according to the partition of these compounds
between the two phases.
Figure 2. Concentration time course for the reduction of 2,5-
hexanedione without acetone removal. Conditions: V ) 0.01
-
1
L, 40 °C, pH 6.0, 0.05 mol L potassium phosphate buffer, 50
-
1
-1
-1
g
2
cww
L
rec E. coli, 0.1 mol L 2,5-hexanedione, 1 mol L
-propanol. Dashed lines (- - -) are drawn as visual aides.
of whole cell biotransformation processes, in particular catalyzed
by recombinant microorganisms, and acetone removal techniques.
This work describes whole cell biotransformation processes
with recombinant Escherichia coli overexpressing recombinant
alcohol dehydrogenase from Lactobacillus breVis (LbADH)
for the production of several chiral alcohols in a substrate
coupled approach with in situ acetone removal. Acetone
removal has been achieved by stripping or pervaporation.
Furthermore whole cell biotransformation has been carried out
in a biphasic system with potassium phosphate buffer and the
In Situ Acetone Removal by Stripping
2,5-Hexanedione has been reduced to (2R,5R)-hexanediol
17
by recombinant E. coli cells expressing recombinant LbADH.
Because the conversion of 2,5-hexanedione to 2,5-hexanediol
is a two-step reaction, an excess of the cosubstrate 2-propanol
is necessary to achieve sufficient regeneration of cofactors
(Figure 1).
A batch reaction experiment without acetone removal
showed that there is a thermodynamic limitation that results in
low yield. In this batch experiment a high 2-propanol concen-
3 2 2
ionic liquid [BMIM][(CF SO ) N] (1-butyl-3-methylimidazo-
lium bis((trifluoromethyl)sulfonyl)amide) to utilize the partition
behaviour of acetone in order to remove it from the aqueous
-1
tration of 1 mol L was applied. Despite almost complete
18
phase. In all reactor systems higher yield could be achieved
compared to processes without in situ acetone removal. All
acetone removal techniques described in this work have already
been applied successfully for ketone reducing biotransformation
processes catalyzed by isolated enzymes. By contrast only a
few examples have been published concerning the combination
-1
consumption of the substrate, only 55 mmol L of the product
2R,5R)-hexanediol could be produced. The intermediate (R)-
-hydroxyhexane-2-one was not converted completely; a con-
(
5
-1
centration of 45 mmol L could be determined after equilib-
rium was reached (Figure 2).
The coproduct acetone is the most volatile compound in this
reaction system. Thus it is possible to remove acetone from
the reaction system by passing a continuous air stream through
the reaction mixture. The pressurized air was first passed
through a rotameter and then through a water/2-propanol
mixture in order to saturate the air stream with water and
(
(
(
14) Daussmann, T.; Rosen, T. C.; Dünkelmann, P. Eng. Life Sci. 2006, 6,
1
25–129.
15) Liese, A.; Zelinski, T.; Kula, M. R.; Kierkels, H.; Karutz, M.; Kragl,
U.; Wandrey, C. J. Mol. Catal. B: Enzymatic 1998, 4, 91–99.
16) Villela, M.; Stillger, T.; Müller, M.; Liese, A.; Wandrey, C. Ang.
Chem., Int. Ed. 2003, 42, 2993–2996.
(
(
17) Hummel, W. AdV. Biochem. Eng. Biotechnol. 1997, 58, 145–184.
2-propanol. Finally the air stream was passed into the reaction
18) Eckstein, M.; Villela, M.; Liese, A.; Kragl, U. Chem. Commun. 2004,
9
, 1084–1085.
solution thermostated at 40 °C.
Vol. 11, No. 5, 2007 / Organic Process Research & Development
•
837

Figure 3. Yield during production of (2R,5R)-hexanediol with
acetone removal by stripping. Conditions: V ) 0.05 L, 40 °C,
Figure 4. Yield during production of (2R,5R)-hexanediol with
acetone removal by pervaporation. Conditions: V ) 0.15 L, 40
-
1
pH 6.0, 0.05 mol
L
potassium phosphate buffer,
-1
°
5
C, pH 6.0, 0.05 mol L
potassium phosphate buffer,
-
1
-1
-1
5
2
0 gcww
L
rec E. coli, 0.1 mol L 2,5-hexanedione, 1 mol L
-1
-1
-1
0 gcww
L
rec E. coli, 0.1 mol L 2,5-hexanedione, 1 mol L
-1
-propanol, air stream 1 L min . Dashed lines (- - -) are drawn
2
-propanol, vacuum pressure 15 mbar. Dashed lines (- - -) are
as visual aides.
drawn as visual aides.
The batch reduction of 2,5-hexanedione was also carried out
in a batch experiment with acetone removal by stripping and
compared to a process without acetone removal (Figure 3). In
this two-step reduction yield was defined as amount of product
in a pervaporation process strongly depends on the temperature.
For that reason 40 °C was chosen as the reaction temperature,
which is an acceptable compromise between sufficient acetone
mass transfer and thermostability of the biocatalyst.
(2R,5R)-hexanediol divided through the initial amount of
Figure 4 shows the increasing yield in biotransformation
processes without acetone removal and with in situ acetone
removal by pervaporation. During the first 90 min, the course
of reaction was almost equal for both experiments. In the
experiment without acetone removal the yield just slightly
increased after 90 min. In the reactor system with the pervapo-
ration unit the reaction continued. In this reaction where
equilibrium is shifted to a much higher degree of conversion
as a result of the acetone removal by pervaporation, the yield
could be increased to >90%.
substrate (2,5-hexanedione):
ct(product)
yield )
(1)
c0(substrate)
During the first 90 min, the course of reaction was almost
equal for both experiments. In the experiment without acetone
removal a yield of 55% could be achieved when equilibrium
was reached. In the experiment with acetone removal by
stripping the reaction went on without deceleration until
completion. In comparison to the process without acetone
removal, thermodynamic limitation could be overcome and
almost quantitative yield of >95% was achieved.
Since the synthesis of (2R,5R)-hexanediol from the product
2,5-hexanedione is a two-step reaction there are 2 equiv of
acetone produced for each equivalent of ketone fully reduced.
In the experiment without acetone removal the acetone con-
-1
centration increased up to 180 mmol L (Figure 5). In the
biotransformation processes with in situ acetone removal, the
In Situ Acetone Removal by Pervaporation
-1
acetone concentration did not reach more than 75 mmol L
The reduction of 2,5-hexanedione (Figure 1) has also been
carried out in a reaction system with acetone removal by
pervaporation. The pervaporation process is characterized by a
over the course of the reaction. In general, the rate of acetone
removal did not significantly differ between the two methods.
From the concentration–time curve it becomes obvious that
there is a drastic increase in acetone concentration at the
beginning of the reaction due to the high catalytic activity of
the whole cell catalysts, even when in situ acetone removal
methods are applied. After 30 min, a decreasing acetone
concentration can be observed in the experiments with acetone
removal.
19
selective transfer of compounds through a membrane. Thereby
the physical conditions of these compounds change from liquid
to gaseous. In pre-investigations several organophilic mem-
branes have been characterized with respect to their ability to
remove acetone selectively from the reaction mixture. A
membrane composed of polydimethoxysiloxane was chosen for
further investigations due to the high acetone mass flow in
comparison to other tested membrane types. The mass transfer
of 2-propanol is 50% of that of acetone. Thus 2-propanol must
be added during the reaction in order to ensure sufficient
concentration of 2-propanol. The mass transfer of a compound
In Situ Acetone Removal by Extraction
The experiments with in situ acetone removal by extraction
were carried out using the substrate 1-phenyl-2-propanone,
which was reduced to 1-phenyl-2-propanol (Figure 6). Since
both compounds, substrate and product, are extremely hydro-
phobic compounds, they preferably remain in the non-aqueous
(
19) Lütz, S.; Rao, N.; Wandrey, C. Chem. Eng. Technol. 2005, 77, 1669–
1
682.
8
38
•
Vol. 11, No. 5, 2007 / Organic Process Research & Development

Figure 5. Acetone concentration during reduction of 2,5-
hexanedione with and without in situ acetone removal. Dashed
lines (- - -) are drawn as visual aides.
Figure 6. Reduction of 1-phenyl-2-propanone with substrate-
coupled cofactor regeneration.
Figure 7. Acetone concentration in both phases during reduc-
tion of 1-phenyl-2-propanone in (a) buffer/MTBE biphasic
1
8
Table 1. Partition coefficients for 2-propanol and acetone
system and (b) buffer/[BMIM][(CF
Conditions: V ) 0.005 L KP buffer pH 6.0, 0.005 L non-
aqueous phase, room temperature, 20 gcww
3 2 2
SO ) N] biphasic system.
system
MTBE/buffer
3 2 2
BMIM][(CF SO ) N]/buffer
2-propanol
acetone
i
-
1
L
rec E. coli, 0.02
1.0
0.4
1.1
2.0
-1
-1
mol L 1-phenyl-2-propanone, 0.4 mol L 2-propanol. Dashed
lines (- - -) are drawn as visual aides.
[
acetone prefers to remain in the aqueous phase according to its
partition coefficient (Figure 7a). In the biphasic system with
the ionic liquid [BMIM][(CF SO ) N] as non-aqueous phase,
3 2 2
the acetone concentration in the aqueous phase is only 50% of
that in the non-aqueous phase (Figure 7b). As a result of the
more effective extraction in the biphasic system with buffer
and [BMIM][(CF SO ) N], an increased yield of >95% could
3 2 2
be achieved. In the system with buffer and MTBE, a yield of
only 24% was observed (Figure 8).
phase. Thus their partitioning behaviour will not lead to any
shift in the thermodynamic equilibrium of the reaction, when
different non-aqueous phases are applied. However, calculating
equilibrium conversions in biphasic systems is a complex task.
Recently, new approaches have been reported to estimate
suitable reaction conditions for enzyme-catalyzed reactions in
biphasic systems and to obtain reliable data for calculation of
20,21
equilibrium constants.
It is known that the partition behaviour of 2-propanol and
acetone in a biphasic system containing buffer and the ionic
Stability Investigations/Comparison
To evaluate the stability of the biocatalysts applying different
methods of in situ acetone removal, the cells were incubated
under certain conditions and their remaining catalytic activity
was determined periodically. In a first experimental set-up the
stability of the cells was determined in processes applying the
acetone removal techniques. In a second set-up there was no
acetone removal but all other conditions (e.g., temperature,
stirring speed) were kept constant, so that the differences in
stability were influenced only by the acetone removal procedure.
In the case of the biphasic system the stability of the cells in
liquid [BMIM][(CF
partitioning behaviour in buffer/methyl tert-butyl ether (MTBE)
Table 1). Thus a positive effect on the biotransformation can
3 2 2
SO ) N] significantly differ from their
(
be expected due to the acetone extraction.
The reduction of 1-phenyl-2-propanone with substrate-
coupled cofactor regeneration was carried out in biphasic
systems containing MTBE or the ionic liquid
[
BMIM][(CF
3 2 2
SO ) N] as non-aqueous phase. In the biphasic
system including MTBE as non-aqueous phase, the produced
3 2 2
the [BMIM][(CF SO ) N]/buffer biphasic system was compared
(
(
20) Eckstein, M. F.; Peters, M.; Lembrecht, J.; Spiess, A. C.; Greiner, L.
AdV. Synth. Catal. 2006, 348, 1591–1596.
to that stability of the cells in an aqueous one-phase system
with the same parameters such as pH value and temperature.
The results can be seen from Table 2.
21) Eckstein, M. F.; Lembrecht, J.; Schumacher, J.; Eberhard, W.; Spiess,
A. C.; Peters, M.; Roosen, C.; Greiner, L.; Leitner, W.; Kragl, U.
AdV. Synth. Catal. 2006, 1597–1604.
Vol. 11, No. 5, 2007 / Organic Process Research & Development
•
839

tion processes with Saccharomyces cereVisiae or removal of
butanol and acetone from fermentation processes applying
22
Clostridium species. On the laboratory scale, this method had
been restricted to ketone-reducing processes using isolated
enzymes as biocatalysts. This is the first work that deals with
the combination of whole cell biotransformation and in situ
acetone removal by pervaporation.
The choice of biphasic systems is beneficial when hydro-
phobic substrates should be applied in biocatalytic reactions.
It could be shown that by using the ionic liquid
3 2 2
[BMIM][(CF SO ) N] as the non-aqueous phase hydrophobic
ketones such as 1-phenyl-2-propanone can be applied in high
concentrations. Furthermore this ionic liquid shows a favourable
partitioning behaviour for acetone and 2-propanol so that it can
be used not only as substrate reservoir but also for the in situ
extraction of acetone. The influence of biphasic systems
Figure 8. Yield during reduction of 1-phenyl-2-propanone in
biphasic systems. Conditions: V ) 0.005 L KP buffer pH 6.0,
.005 L non-aqueous phase, room temperature, 20 gcww
3 2 2
applying the ionic liquid [BMIM][(CF SO ) N] as the non-
i
-1
0
L
rec
aqueous phase on the stability and activity of the whole cell
biocatalyst has not been investigated satisfactorily until now.
In literature there are some other examples for biphasic systems
with ionic liquids in whole cell biotransformation processes.
Pfründer et al. described the reduction of 4-chloro acetophenone
using Lactobacillus kefir cells in biphasic systems consisting
-
1
-1
E. coli, 0.02 mol L 1-phenyl-2-propanone, 0.4 mol L
2
-propanol. Dashed lines (- - -) are drawn as visual aides.
It became obvious that the stripping procedure has the most
drastic effect towards the stability of the whole cell biocatalysts.
The relative activity was only 25% in comparison to the cells
incubated under the same conditions but without passing an
air stream through the reaction mixture. The pervaporation
process is a more gentle way for acetone removal since there
is a relative stability of 82% compared to cells incubated in a
process without applying the pervaporation procedure. The
relative stability of cells incubated in a biphasic system
23
of buffer and different ionic liquids. Other applications of ionic
liquids in biocatalysis, which mainly deal with isolated lipases,
24,25
have recently been reviewed several times.
Experimental Section
Bacterial Strains and Plasmids. Plasmid pBtacLB-ADH
(X-Zyme, Düsseldorf, Germany) carrying the adh gene and
3 2 2
consisting of [BMIM][(CF SO ) N] and buffer was determined
-1
ampicillin resistance was employed as expression vector. E. coli
BL21 Star (DE3) (Invitrogen, Karlsruhe, Germany) was used
for gene expression and whole cell biotransformation. The E.
coli strain was transformed by the method described by
to be 63% in comparison to cells incubated in 50 mmol L
potassium phosphate buffer, pH 6.0.
Conclusions and Summary
26
Hanahan.
Biomass Production. Recombinant E. coli were cultivated
By applying different in situ acetone removal techniques,
whole cell ketone reduction processes with substrate-coupled
cofactor regeneration can be intensified significantly. In par-
ticular for two-step reductions such as the reduction of 2,5-
hexanedione such a technique is an important tool since only
low yield could be obtained when the coproduct acetone is not
removed from the reaction mixture.
-1
on modified Luria Bertani (LB) medium: 10 g L casein
-
1
-1
-1
peptone, 5 g L yeast extract, 10 g L NaCl, 4 g L glucose.
-1
Then 50 mL of medium containing 50 mg L carbenicillin
was inoculated with 500 µL of a glycerine stock culture. After
incubation overnight at 30 °C and 150 rpm, 200 mL of the
same medium was inoculated with 500 µL of the preculture.
Main cultures were incubated for 6 h at 37 °C and 150 rpm.
Then gene expression was induced by adding 0.2 mmol L
IPTG. Afterwards the cultures were incubated for 18 h at 27
C and 150 rpm. The cells were harvested by centrifugation
Beckman Coulter Avanti J-20 XP, 8000 rpm, 20 min, 4 °C)
The reduction of 2,5-hexanedione by recombinant E. coli
cells resulted in almost similar yields when acetone was
removed by either applying the stripping procedure or a
pervaporation step. The pervaporation process is a very gentle
way of acetone removal which results in excellent stability of
the biocatalyst compared to processes applying acetone removal
by the stripping method. Nevertheless, the stripping process is
characterized by a simple reactor set-up that can be run cost-
effectively. In industry, there are also some ketone reducing
processes established using methods for acetone removal
-1
°
(
-1
and washed once with 50 mmol L potassium phosphate
buffer, pH 6.0. The cells were stored as a 100 g L cell
suspension in 50 mmol L potassium phosphate buffer, pH 6
-1
-1
at 4 °C.
1,14
employing evaporation by reduced pressure. To our knowl-
edge there are no processes established in industry using the
pervaporation method for in situ acetone removal in biotrans-
formation processes catalyzed by whole resting cells. The
majority of examples for in situ product removal in whole cell
biotechnology are dealing with product removal from fermenta-
tion processes, for example, removal of ethanol from fermenta-
(
(
(
(
22) Stark, D.; von Stockar, U. AdV. Biochem. Eng./Biotechnol. 2003, 80,
149–175.
23) Pfründer, H.; Amidjojo, M.; Kragl, U.; Weuster-Botz, D. Ang. Chem.,
Int. Ed. 2004, 43, 4529–4531.
24) Kragl, U.; Eckstein, M.; Kaftzik, N. Curr. Opin. Biotechnol. 2002,
13, 565–571.
25) Van Rantwijk, F.; Lau, R. M.; Sheldon, R. A. Trends Biotechnol. 2003,
2
1, 131–138.
(26) Hanahan, D. J. Mol. Biol. 1983, 166, 557–580.
8
40
•
Vol. 11, No. 5, 2007 / Organic Process Research & Development

Table 2. Relative stability of the biocatalyst and biocatalyst specific productivity during biocatalysis with in situ acetone
a
removal
biocatalyst specific productivity [gproduct g ]
cww d-1
-
1
method of acetone removal
relative stability [-]
without acetone removal
stripping
1
25.5
46.3
44.0
2.59
0.25
0.82
0.63
pervaporation
[
3 2 2
BMIM][(CF SO ) N] / buffer
a
Stripping and pervaporation experiments were carried out using 2,5-hexanedione as substrate whereas the biphasic system was applied for the reduction of the
hydrophobic substrate 1-phenyl-2-propanone.
Analytics. Quantification of all substrates and products of
Biphasic Systems. For the biotransformation process in
biotransformation processes was carried out on an Agilent HP-
biphasic systems, 5 mL of potassium phosphate buffer contain-
-1
6890A gas chromatograph with a Permabond Carbowax 20M
ing 40 gcww
L
biomass was introduced into a centrifuge tube
column (50 m × 0,32 mm i.d., Macherey-Nagel, Dueren,
Germany) with a flame ionization detector and helium as carrier
gas. To minimize the injection error, n-butanol was used as an
internal standard. All compounds from experiments with in situ
acetone removal were detected by a method using the following
together with 5 mL of the non-aqueous medium. The tubes were
mixed on a shaker (VXR basic IKA Vibrax, IKA Labortechnik,
Staufen, Germany) with 300 rpm.
Sampling. A 200 µL sample of reaction solution was
removed from the reaction system and centrifuged (Centrifuge
5415 D, Eppendorf, Hamburg, Germany) at 16000 rpm for 30 s
to remove the cells. To avoid further reaction catalyzed by
leaked enzyme, the supernatant was transferred into a second
tube and incubated for 60 s at 99 °C (ThermoStat plus,
Eppendorf, Hamburg, Germany). After cooling at 4 °C the
sample was prepared for gas chromatography analysis.
Tests for Stability. For the investigations on the catalyst
stability recombinant E. coli cells were incubated under the
conditions applied in biotransformation processes with in situ
acetone removal. Periodically the cells were tested for their
remaining catalytic activity. Therefore the cells were used as
biocatalysts for the reduction of methyl acetoacetate under the
following conditions: V ) 1 mL (Eppendorf tube), 30 °C, 300
rpm (Thermomixer compact 5350, Eppendorf, Hamburg, Ger-
-1
temperature programm: 5 min 50 °C, 40 °C min , 9 min 160
C. Typical retention times were acetone 2.5 min, 2-propanol
°
4.0 min, n-butanol 7.1 min, 1-phenyl-2-propanone 10.2 min,
1-phenylpropanol 10.9 min, 2,5-hexanedione 10.3 min, 5-hy-
droxyhexane-2-one 11.7 min, and 2,5-hexanediol 15.6 min. All
compounds from batch experiments for investigation on the
catalyst stability were detected by a method using the following
-1
temperature programm: 6 min 70 °C, 25 °C min , 3 min 160
C. Typical retention times were acetone 1.7 min, 2-propanol
.2 min, n-butanol 5.6 min, methylacetoacetate 9.9 min, and
°
2
methylhydroxybutyrate 10.6 min.
Reactor Performance. Stripping Experiments. Biotransfor-
mation processes with acetone removal by stripping were carried
out on a 50 mL scale in a jacketed glass reactor (glass workshop
of the Research Centre Juelich, Germany) with a porous glass
frit as bottom. The reactor solution was gassed from the bottom
-1
many) pH 6.0, 0.05 mmol L potassium phosphate buffer, 5
-
1
-1
g
L
cww
-1
L
rec E. coli, 0.1 mol L methyl acetoacetate, 0.2 mol
-1
with pressurized air at a flow rate of 1 L min . The flow rate
was measured by a rotameter (Sho-rate, Brooks Instrument,
Veenendaal, Netherlands). To avoid outgassing of 2-propanol,
the pressurized air was humidified with a 50% (v/v) mixture
of water and 2-propanol. One drop of antifoam (Biospumex
2-propanol. For calculation the catalytic activity the
increasing concentration of the product (R)-methyl hydroxy-
butyrate was measured by gas chromatography as described
above.
153K, Cognis GmbH, Monheim am Rhein, Germany) was
Acknowledgment
added to the reaction solution in order to avoid foam formation.
This work was financially supported by BMBF (Project no.
0313440C). The authors thank Prof. Dr. C. Wandrey for
ongoing support and Prof. Dr. A. Liese (TU Hamburg-Harburg)
for fruitful discussions. Help in plasmid amplification and strain
construction by Dr. Bringer-Meyer (IBT-1, Research Centre
Juelich) is gratefully acknowledged. Furthermore the authors
thank Dr. Matuschewski (PolyAN, Berlin, Germany) for
providing the pervaporation membrane.
-1
The biotransformation process was carried out in 50 mmol L
potassium phosphate buffer, pH 6.0, at 40 °C. The biomass
-1
concentration was 50 gcww L .
PerVaporation. Biotransformation processes with acetone
removal by pervaporation were carried out in a reverse osmosis
system (P-28, cmcelfa,wen-Schwyz, Switzerland). The stainless
2
steel cell provided 44.2 cm of active membrane area. The
polydimethoxysiloxane membrane PA-HP-02, (PolyAN, Berlin,
Germany) was used as pervaporation membrane. The reaction
solution was pumped through the reactor by a flow rate of 0.3
L min (flexible-tube pump 505U, Watson-Marlow, Rommer-
skirchen, Germany). The feed volume was 0.15 L. A vacuum
of 15 mbar was maintained throughout the whole reaction time
Supporting Information Available
Additional data on the impact of acetone on the biocatalysts,
reactor schemes, and stability measurements in the pervapora-
tion setup. This information is available free of charge via the
Internet at http://pubs.acs.org.
-1
(diaphragm vacuum pump MZ 2C, vacuubrand, Wertheim,
Germany). The biotransformation process was carried out in
-1
Received for review March 6, 2007.
OP700055E
5
0 mmol L potassium phosphate buffer, pH 6.0, at 40 °C.
-1
The biomass concentration was 50 gcww L .
Vol. 11, No. 5, 2007 / Organic Process Research & Development
•
841