Steroselective hydrolysis of DL-beta-acetylthioisobutyramide catalyzed by genetically engineered E. coli immobilized on Celite 580 in a packed bed bioreactor

Article abstract of DOI:10.1002/jccs.200700227

Pseudomonas putida IFO12996 catalyzes the stereoselective hydrolysis of methyl DL-β-acetylthioisobutyramide (DL-ATIA) to form D-β- acetylthioisobutyric acid (DAT), a key intermediate for synthesis of a series of angiotensin converting enzyme inhibitors. The esterase gene of Pseudomonas putida IFO12996 was cloned and expressed in Escherichia coli which was further immobilized and retained on a packed bed bioreactor filled with Celite 580. The packed bed bioreactor was used to conduct the stereoselective hydrolysis of DL-ATIA and to give DAT with a yield of 34.5%, enantiometric excess value of 97% and enantioselectivity value > 150. The optimal pH and temperature for the reaction were 9.0 and 57 °C ～ 67 °C, respectively. The kinetic constants (Km and Vmax) of immobilized cells were found to be 372.5 mM and 285.7 μmol min^-1 (g cell)^-1, respectively. The immobilized cells retained over 60% of the initial catalytic activity after 5 batch cycles of production. This paper presents a simple, practical and economical process of immobilization of genetically engineered E. coli on a novel packed bed bioreactor for production of DAT.

Full text of DOI:10.1002/jccs.200700227

Journal of the Chinese Chemical Society, 2007, 54, 1607-1613
1607
Steroselective Hydrolysis of DL-beta-acetylthioisobutyramide Catalyzed by
Genetically Engineered E. coli Immobilized on Celite 580 in a Packed Bed
Bioreactor
a,b,
Shyh-Yu Shaw * (
a
), Yu-Jen Chen (
),
a
Jung-Jung Ou (
c
) and Lewis Ho (
)
Department of Chemistry, National Cheng Kung University, Tainan, Taiwan, R.O.C.
a
b
Institute of Biotechnology, National Cheng Kung University, Tainan, Taiwan, R.O.C.
c
IDL Bioservices, LLC, Waterford, CT 06385, U.S.A.
Pseudomonas putida IFO12996 catalyzes the stereoselective hydrolysis of methyl DL-b-acetylthio-
isobutyramide (DL-ATIA) to form D-b-acetylthioisobutyric acid (DAT), a key intermediate for synthesis
of a series of angiotensin converting enzyme inhibitors. The esterase gene of Pseudomonas putida
IFO12996 was cloned and expressed in Escherichia coli which was further immobilized and retained on a
packed bed bioreactor filled with Celite 580. The packed bed bioreactor was used to conduct the stereo-
selective hydrolysis of DL-ATIA and to give DAT with a yield of 34.5%, enantiometric excess value of
9
7% and enantioselectivity value > 150. The optimal pH and temperature for the reaction were 9.0 and 57
C ~ 67 °C, respectively. The kinetic constants (Km and Vmax) of immobilized cells were found to be
°
-
72.5 mM and 285.7 mmol min (g cell) , respectively. The immobilized cells retained over 60% of the
1
-1
3
initial catalytic activity after 5 batch cycles of production.
This paper presents a simple, practical and economical process of immobilization of genetically engi-
neered E. coli on a novel packed bed bioreactor for production of DAT.
Keywords: Whole cell immobilization; Genetically-engineered E. coli;
DL-b-Acetylthioisobutyramide (DL-ATIA); D-b-Acetylthioisobutyric acid (DAT);
Packed bed bioreactor.
INTRODUCTION
Enzymes and whole-cell biocatalysts have become
aceous earth which is primarily used as a filter aid for filtra-
tion application in industry. Celite has a void fraction of
75% to 80% of particle volume. The majority of pores on
Celite range from 1 to 15 mm in diameter with an average
economically and environmentally attractive because they
catalyze reactions with high stereoselectivity under mild
1
conditions. In order to repeatedly use biocatalysts, biocat-
6
pore size of 4.3 mm. As a result, it has also been used for
7,8
cell immobilization by adsorbing or entrapping the cells
2
alysts can be immobilized on solid supports. In general,
inside of the Celite. In this study we took advantage of not
only this porous property, but also primarily the filter aid
function of Celite to retain a massive quantity of cells in the
reactor bed and to establish an easy, efficient and economi-
cal bioreactor.
immobilization of purified enzymes is far more expensive
than immobilization of whole cells. Therefore, immobi-
lized whole cell technology has generated practical interest
3
,4
for industrial applications.
Whole cell immobilization has been studied for many
years. Nevertheless, commercialization of immobilized
whole cells is still rather limited, partially because of its
high cost of immobilization relative to its reusability. Re-
cently, there has been interest in using inert porous solid
materials (e.g., silica and glass) for immobilization of mi-
Captopril is a drug commonly used against hyperten-
9
sion by specific competitive inhibition of angiotensin I
converting enzyme (ACE), the enzyme responsible for the
1
0
conversion of angiotensin I to angiotensin II. Because
Captopril has two optical isomers, research has shown that
in relation to the a position, the D form has 100 times the
5
11
croorganisms. Celite is an inexpensive form of diatom-
reactivity of the L form. Therefore D-b-acetylthioisobu-
*
Corresponding author. Tel: +886-6-2080476; Fax: +886-6-2740552; E-mail: syshaw@mail.ncku.edu.tw

1
608 J. Chin. Chem. Soc., Vol. 54, No. 6, 2007
Shaw et al.
1
4
tyric acid (DAT) also known as S-(-)-3-acetylthio-2-meth-
ously. The plasmid pET-PpEST was transformed into E.
coli BL21 (DE3) competent cells. A single colony was
picked up and grown in LB medium containing 100 mg
ylpropionic acid has become the desired form of precursor
1
2
for synthesis of Captopril. Recently, we have cloned an
esterase (PpEST) gene from Pseudomonas putida IFO12996
and over-expressed it in Escherichia coli (E. coli). PpEST
is a non-disulfide-linked homotrimer with a monomer mo-
lecular weight of 33,000, and it catalyzes the stereoselec-
tive hydrolysis of methyl DL-b-acetylthioisobutyrate (DL-
MATI) and DL-b-acetylthioisobutyramide (DL-ATIA) to
give DAT with an enantiometric excess value (e.e. value) of
about 97% and an enantioselectivity value (E-value) > 150,
-
1
mL ampicillin. The overnight culture was diluted 1:100
-
1
with LB medium containing 100 mg mL ampicillin and al-
lowed to grow until A600 reached 0.3. IPTG was added at a
final concentration of 0.1 mM and the cells were harvested
after 3 hours. Cells were collected from the culture by
centrifugation at 10,800 ´ g at 4 °C for 20 min. Cell pellets
were suspended with 20 mL Tris-HCl buffer (0.5 mM, pH
7.0) per gram of cell pellet. It was stored at 4 °C before im-
mobilization. Specific enzymatic activity of one unit is de-
fined as the amount of enzyme that catalyzes the formation
of 1 mmole of DAT in 1 min at 780 mM DL-ATIA at 37 °C
1
3,14
respectively.
However, DL-ATIA has shown better sta-
bility and is higher in solubility than DL-MATI and thus be-
comes the choice for production of DAT by the enzymatic
1
resolution method (Scheme I).
4
15
(pH 7.0).
In this report, we described a method to immobilize
and retain genetically-engineered E. coli. in a packed bed
bioreactor filled with Celite 580, used the bioreactor to per-
form stereoselective hydrolysis of DL-ATIA and examined
the characteristics of the immobilized cells.
Synthesis of DL-ATIA from methacrylamide
1
4
DL-ATIA was synthesized as previously described
(Shaw et al. 2006). In brief, 17.2 g (0.2 mol) methacryl-
amide (Acros, Geel, Belgium) was dissolved in 100 mL
n-propyl alcohol, then 30.4 g (0.4 mol) thioacetic acid
(
TCI, Tokyo, Japan) and 1 mL (10 mmol) piperidine were
EXPERIMENTAL SECTION
added. The mixture reacted at room temperature for 4 h. To
the reaction mixture was then added 150 mL 0.5 M Tris-
HCl buffer (pH 7) and the pH was adjusted to 1 with con-
Strains, plasmids and growth medium
Pseudomonas putida IFO12996 was from the Insti-
tute for Fermentation, Osaka (Osaka, Japan). Escherichia
coli BL21 (DE3) was used for gene cloning and expression.
Microorganisms were cultured on Luria-Bertani (LB) me-
dium at 30 °C for Pseudomonas putida IFO12996 and at
2 4
centrated H SO . The mixture was extracted with 100 mL
hexane 3 times, and most of the unreacted thioacetic acid
was extracted to the hexane layer. Then the aqueous solu-
tion was adjusted to pH 7 with 10 N NaOH and extracted
with 100 mL ethyl acetate 3 times, and the DL-amide was
extracted to the ethyl acetate layer. Condensation of the
ethyl acetate extract followed to give the crude DL-amide,
and pure DL-ATIA was prepared by crystallization and
gave 20.1 g (yield of 62.4%).
-
1
3
7 °C for E. coli. In selective medium, 100 mg mL of
ampicillin was used. pET21d (Invitrogen) was used as
cloning and expression vector. Celite 580 was obtained
from Celite Corp. (Lompoc, CA).
Preparation of recombinant E. coli cells
Immobilization of cells
The recombinant PpEST from Pseudomonas putida
IFO12996 was cloned and expressed as described previ-
The laboratory bioreactor used in this study consisted
of a glass tube column (140 mm length ´ 26 mm diameter)
Scheme I Cell-catalyzed optical resolution of DL-ATIA

Enzymatic Resolution in a Packed Bed Bioreactor
J. Chin. Chem. Soc., Vol. 54, No. 6, 2007 1609
with a glass fiber on the bottom to hold Celite, a peristaltic
pump and a 250 mL bottle with a magnetic stirrer (see Fig.
(v/v) glutaraldehyde in pH 7.0 phosphate buffer (0.1 M) for
3 h at 4 °C and dehydrated in ethanol (from 30% to 70%)
7
and then in acetone (from 70% to 100%) (Llanes et al.
1
). 1.0 g of Celite 580 was suspended in 50 mL of Tris-HCl
buffer (0.5 M, pH 7.0) and pumped through the column to
form a precoat. 1.5 g cells (wet weight, 322 U) and 2.0 g of
Celite 580 were suspended and well mixed in a 50 mL
Tris-HCl buffer (0.5 M, pH 7.0). Then the mixture was
pumped through the precoated column slowly and recircu-
lated until the solution became clear. The cells were thus
immobilized and retained in the reactor. The assembly of
the reactor system was then placed in a thermostatic incu-
bator set at 37 °C.
2001). The samples were dried in a critical point dryer
(Hitachi HCP-2) and covered with a gold film. SEM Im-
ages were taken using a JEOL JSM-5310 LV scanning elec-
tron microscope at 20 kEv and a working distance of 25
mm.
DAT production in the bioreactor
9 g DL-ATIA was added to the bottle and stirred till
the DL-ATIA was dissolved. The peristaltic pump was
started to recirculate the solution in the system. The solu-
tion was sampled frequently and analyzed by an HPLC sys-
tem with a reverse phase C-18 column. Injected samples
were eluted with isocratic 50% solvent A (25 mM phos-
phate buffer, pH 6.5) and solvent B (acetonitrile) and moni-
tored at UV 229 nm. The eluted samples were quantified
for both concentrations of DL-ATIA and DAT. The reaction
was terminated as the reduction of DL-ATIA or production
of DAT ceased.
9
g DL-ATIA was added to the bottle and completely
dissolved in the buffer at a saturated concentration of 780
mM, and then the peristaltic pump was started to com-
mence the reaction. The hydrolytic reaction mixture was
sampled and detected by a reverse phase HPLC for assay of
DL-ATIAand DAT. As a batch process, the reaction was ter-
minated when the substrate was mostly consumed.
Scanning electron microscopy (SEM)
SEM was used to examine how the cells were immo-
bilized on the Celite 580. Samples of Celite with immobi-
lized cells in the bioreactor were taken and fixed in 4%
Recovery and purification of DAT from the reaction
mixture
After the reaction, unreacted DL-ATIA and L-ATIA in
the supernatant was extracted with 50 mL ethyl acetate 3
times. pH of the aqueous layer of the resultant extract was
adjusted with diluted sulfuric acid to 2.0. The mixture was
then extracted with 50 mL ethyl acetate 3 times to give a
4
DAT extract. Anhydrous MgSO was added to the extract.
After dehydration, the solvent was vaporized to give crude
DAT and then distilled under reduced pressure to give pure
DAT. The optical rotation of DAT was used to calculate the
e.e. value and E-value according to a previously described
1
6
method.
Effect of pH and temperature on the reactivity of
immobilized cells
The reaction mixtures (1 mL) containing 100 mg
Celite with retained and immobilized cells (13 U), 780 mM
DL-ATIA and buffered solution with different pHs (from
3
.0 to 11.0) were incubated at 37 °C for 20 min and the con-
centration of DAT was determined as described above.
The reaction mixtures (1 mL) containing 100 mg
Celite with retained and immobilized cells (13 U) 780 mM
DL-ATIA were incubated in 0.5 M Tris-HCl buffer (pH 7.0)
at various temperatures of 17, 27, 37, 47, 57, 67, and 77 °C
Fig. 1. Schematic representation of the Celite-immo-
bilized E. coli bioreactor for the stereoselective
hydrolysis of DL-ATIA.

1
610 J. Chin. Chem. Soc., Vol. 54, No. 6, 2007
Shaw et al.
for 20 min, and the concentration of DAT was determined
as described above.
100 mL Tris-HCl (0.5 M, pH 7.0) containing DL-ATIA(780
mM) at 37 °C. The hydrolytic reaction mixture was sam-
pled and the concentration of DL-ATIA and DAT were de-
tected by a reverse phase HPLC to determine the activity of
the immobilized cells. After each cycle of reaction (240
mins), the reaction was stopped and the cells in the bio-
reactor were washed with 100 mL Tris-HCl buffer (0.5 M,
pH 7.0) and then a 100 mL fresh DL-ATIA (780 mM) solu-
tion was added to start the next cycle.
Thermal stability of the immobilized cells
The thermal stability measurement for the immobi-
lized cells was done by incubating the Celite with cells (13
U) at temperatures of 57 °C, 67 °C, 72 °C and 77 °C. Every
1
7
5 min 1 mL Tris-HCl buffer (0.5 M, pH 7.0) containing
80 mM DL-ATIAwas added into a 100 mg Celite of immo-
bilized cells (13 U) solution at 37 °C for 20 min. The re-
maining enzyme activity was determined by measuring
DAT as described above.
RESULTS AND DISCUSSION
Kinetics of the immobolized cells in the bioreactor
The kinetic studies of the immobilized cells were per-
formed by direct extraction of the data from the enzymatic
reaction in the packed bed bioreactor with an initial con-
centration of DL-ATIA at 780 mM and at a constant temper-
ature of 37 °C, and pH of 7.0. The concentrations of DL-
ATIA/L-ATIA and DAT at different time intervals were de-
termined by the HPLC system as described above. The
non-linear equations presenting substrate and product pro-
files were curve-fitted by a least-squares method using
Cell immobilization
The scanning electron micrograph of Celite-immobi-
lized E. coli cells showed that some of the cells were ad-
sorbed on the surface and some of them were entrapped in-
side the pores of Celite (Fig. 2). It has been reported that
Mycobacterium sp. NRRL B-3805 was able to be absorbed
on the surface of Celite and catalyze a bioconversion reac-
7
tion. In this study, E. coli was not only absorbed on the sur-
face but also entrapped inside the pores of Celite. This is
probably due to that the loading capacity of cells in this
7
study was significantly higher than the previous one. How-
1
7
Microsoft Office program. From the equation the rates of
reaction at all substrate and product concentrations can be
obtained. The kinetic constants (Km and Vmax) were fur-
ther obtained by Lineweaver-Burk plot.
ever, the optimal cell load in a packed bed bioreactor would
be achieving a maximum load while maintaining a good
flow rate of the mixture of reactant and product and retain-
ing the cells in the reactor without leakage.
Reusability of the immobilized cell
The operational stability of the immobilized cells on
Celite 580 was evaluated in a repeated batch process. 1.5 g
of cells (322 U) was packed in the bioreactor as described
above and the reaction was performed in final volumes of
Effect of pH on reactivity
Fig. 3 shows the pH-activity profiles of immobilized
cells and free cells. The highest activity of immobilized
cells was seen at pH 9.0 which is similar to that of free cells.
Fig. 2. Scanning electron micrographs of Celite-immobilized E. coli cells. Two different views of micrographs were shown
in (a) and (b). The solid arrow points at entrapped cells and the white arrow points at surface adsorbed cells.

Enzymatic Resolution in a Packed Bed Bioreactor
J. Chin. Chem. Soc., Vol. 54, No. 6, 2007 1611
2
from the best curve fitting has an R value of 0.9932. The
Effect of temperature on reactivity
The effect of temperature on the reactivity of immobi-
lized cells and free cells was studied. Results shown in Fig.
entire reaction remained at pH ~ 7.0 without requirement of
pH control because of the formation of ammonia from the
substrate which neutralized the acid formation. DAT was
recovered and purified from the reacted mixture with an
enantiometric excess value of 97% (e.e.) and enantioselec-
tivity value (E-value) > 150. The yield was 34.5%.
4
indicate that the optimal temperatures for the immobi-
lized cells and free cells were about the same at around 57
C ~ 67 °C.
°
Thermal stability of immobilized cells
As to the thermal stability of the enzyme (Fig. 5), it is
expected that the stability was greater at lower tempera-
tures. The immobilized cells remained stable for 1 hour at
Determination of kinetic constants
The kinetic constants (Km and Vmax) were obtained
by fitting experimental data with the Microsoft Office pro-
gram that performs nonlinear regression based on the least-
5
7 °C but lost its activity completely in 45 min at 77 °C.
1
7
squares method. Kinetic parameters for the whole cell
were determined using DL-ATIA as the substrate whose
concentration was 780 mM.
Production of DAT in the packed bed reactor
In Fig. 6 shows the course of enzymatic reaction for
DAT production from a saturated solution of DL-ATIA in
the packed bed reactor. The non-linear equation derived
In Fig. 7 shows the Lineweaver-Burk plot from the
data derived from the equation shown in Fig. 6. The Km
and Vmax values of the immobilized cell was 372.5 mM
-
1
-1
and 285.7 mmol min (g cell) , respectively.
Fig. 3. Effect of pH on relative enzyme activities in the
immobilized cells and free cells.
Fig. 5. Thermal stability of immobilized cells.
Fig. 6. Course of DAT production from DL-ATIA:
Fig. 4. Effect of temperature on relative enzyme activ-
ities in the immobilized cells and free cells.
DL-ATIA (¨) and DAT (n).

1
612 J. Chin. Chem. Soc., Vol. 54, No. 6, 2007
Shaw et al.
Reusability
The operational stability of the immobilized cells on
effective method was developed to immobilize and retain
genetically-engineered E. coli cells on a packed bed reactor
filled with Celite 580 for production of DAT by enzymatic
hydrolysis of DL-ATIA. The substrate concentration allow-
ing for maximum reaction rate was around 780 mM, and
the yield of DAT was at 34.5% with optical purity of 97%
e.e. and E-value > 150. The optimal conditions of the bio-
reactor was found to be at 57 ~ 67 °C and pH 8 ~ 10. The ki-
netic constants of Km and Vmax was found to be 372.5 mM
Celite 580 was evaluated in a repeated batch process. Re-
sults shown in Fig. 8 revealed that the immobilized cells
slowly decreased their activity in higher batch cycles and
retained 60% and 40% of their original activity after 5 and
1
0 batch cycles of usage. Since the immobilization of cells
on Celite 580 was mainly via adsorption and entrapment,
the loss of activity was probably due to some cells leaking
from the packed bed and being removed from each cycle. It
has been reported that Mycobacterium sp. NRRL B-3805
cells immobilized on Celite were able to convert b-sito-
sterol but they were only able to retain 40% activity after 4
batch cycles (Llanes et al. 2001). The difference of the re-
usability in these two studies is probably due to using dif-
ferent bacteria and different immobilization methods.
-
1
-1
and 285.7 mmol min (g cell) , respectively. The opera-
tional stability of the small laboratory reactor maintained
60% of its activity after 5 cycles of use. The operational sta-
bility or reusability was expected to be significantly im-
proved in a larger scale of operation where better or a finer
precoat of Celite in the column can be applied and cell
loss/leakage can be minimized. Nevertheless, the simplic-
ity of the method makes the immobilization of geneti-
cally-engineered E. coli for enzymatic reactions become
economical and practical.
CONCLUSIONS
E. coli has been widely used as a host to produce re-
combinant proteins of interest. In this report, a simple and
ACKNOWLEDGEMENTS
We gratefully acknowledge the financial support
from National Research Council of the Republic of China
(
NSC-96-2113-M-006-006).
Received March 7, 2007.
REFERENCES
Fig. 7. Lineweaver-Burk plot for Km and Vmax.
1. Panke, S.; Held, M.; Wubbolts, M. Curr. Opin. Biotechnol.
2
004, 15, 272.
2
3
. Gill, I.; Ballesteros, A. Trends Biotechnol. 2000, 18, 282.
. Jones, A.; Wood, D. N.; Razniewska, T.; Gaucher, G. M.
Can. J. Chem. Eng. 1986, 64, 547.
4
5
6
7
8
. Kim, C. J.; Chang, Y. K.; Chun, G. T.; Jeong, Y. H.; Lee, S. J.
Biotechnol. Prog. 2001, 17, 453.
. El-Sayed, A. H. M. M.; Mahmoud, W. M.; Coughlin, R. W.
Biotechnol. Bioeng. 1990, 36, 83.
. Wang, S. D.; Wang, D. I. C. Biotechnol. Prog. 1989, 34,
1
261.
. Llanes, N.; Fernandes, P.; Leon, R.; Cabral, J. M. S.;
Pinheiro, H. M. J. Mol. Catal. B: Enzym. 2001, 11, 523.
. Peres, C. M.; Aken, B. V.; Naveau, H.; Agathos, S. N. Appl.
Microbiol. Biotechnol. 1999, 52, 440.
Fig. 8. Reusability of immobilized cells.

Enzymatic Resolution in a Packed Bed Bioreactor
J. Chin. Chem. Soc., Vol. 54, No. 6, 2007 1613
9
. Ondetti, M. A.; Rubin, B.; Cushman, D. W. Sci 1977, 196,
41.
28, 57.
4
13. Elmi, F.; Lee, H. T.; Huang, J. Y.; Hsieh, Y. C.; Wang, Y. L.;
Chen, Y. J.; Shaw, S. Y.; Chen, C. J. J. Bacteriol. 2005, 187,
8470.
1
0. Skiles, J. W.; Suh, J. T.; Williams, B. E.; Menard, P. R.;
Barton, J. N.; Love, B.; Jones, H.; Neiss, E. S.; Schwab, A.;
Mann, W. S.; Khandwala, A.; Wolf, P. S.; Weinryb, I. J. Med.
Chem. 1986, 29, 784.
14. Shaw, S. Y.; Chen, Y. J.; Ou, J. J.; Ho, L. J. Mol. Catal. B:
Enzym. 2006, 38, 163.
1
1. Sakimae, A.; Ozaki, E.; Toyama, H.; Ohsuga, N.; Numazawa,
15. Shin, J. S.; Kim, B. G.; Shin, D. H. Enzyme Microb. Technol.
2001, 29, 232.
R.; Muraoka, I.; Hamada, E.; Ohnishi, H. Biosci. Biotech.
Biochem. 1993, 57, 782.
16. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294.
1
2. Suh, J. T.; Skiles, J. W.; Williams, B. E.; Youssefyeh, R. D.;
Jones, H.; Love, B.; Neiss, E. S.; Schwab, A.; Mann, W. S.;
Khandwala, A.; Wolf, P. S.; Weinryb, I. J. Med. Chem. 1985,
17. Dhar, A.; Dhar, K.; Rosazza, J. P. N. J. Ind. Microbiol.
Biotechnol. 2002, 28, 81.