Coupling of biocatalytic asymmetric epoxidation with NADH regeneration in organic-aqueous emulsions

Article abstract of DOI:10.1002/anie.200353338

Cell-free enzymatic oxidations: Styrene monooxygenase (StyA) was used as a reagent for the gram-scale preparation of enantiopure epoxides. The catalyst is highly stable in a biphasic system and results in conversions of more than 88%. R¹ = H, C

Full text of DOI:10.1002/anie.200353338

Angewandte
Chemie
Asymmetric Synthesis
separated from complex mixtures that contain not only
substrate, products, and a single catalyst, but also whole
Coupling of Biocatalytic Asymmetric Epoxidation cells and cell debris. Thus, as biocatalytic processes develop
with NADH Regeneration in Organic–Aqueous
Emulsions**
further, isolated oxygenases coupled with NAD(P)H regen-
eration may well become a viable alternative to whole-cell
applications. Several challenges are identified and become an
[
9]
issue in cell-free oxygenations. The supply of reducing
equivalents can, in principle, be solved as several regeneration
Karin Hofstetter, Jochen Lutz, Irene Lang,
Bernard Witholt, and Andreas Schmid*
[
10]
systems become available.
In contrast, oxygenase avail-
ability and stability under process conditions represent major
obstacles towards preparative cell-free biocatalytic oxygen-
ations and so far remain unsolved.
The pivotal role of chiral oxyfunctionalized hydrocarbons as
building blocks for pharmaceuticals or agrochemicals has
driven the research for effective methods to obtain them in
enantiopure form. Over the past few decades, numerous
chemical routes have been developed for catalytic asymmet-
Herein we report the first example of the preparation and
synthetic application of a bacterial monooxygenase as a
reagent for asymmetric cell-free epoxidation. As catalyst, we
chose the soluble flavin- and NADH-dependent styrene
monooxygenase (StyAB) from Pseudomonas sp. VLB120.
StyAB catalyzes the enantiospecific epoxidation of a broad
range of styrene-type substrates, such as 1,2-dihydronaph-
[
1,2]
III
ric epoxidations.
Amongst them, chiral salen–Mn com-
plexes and metalloporphyrins have emerged as efficient
catalysts for syntheses of optically active epoxides. Despite
their applicability in synthetically useful organic oxidations
and related reactions, these catalysts are often limited by their
low stability against autoxidation, resulting in low turnover
[
11]
thalene and indene. This two-component enzyme consists
of the actual oxygenase subunit (StyA) and a reductase
[
3]
[12]
numbers.
(StyB).
Alternatively, enzymes can be applied, as they are
StyA was obtained in multigram amounts from a 30-L-
promising catalysts owing to their regio- and enantiodiscri-
scale fermentation of recombinant Escherichia coli JM101
À1
[13]
mination, high turnover numbers (k = 1–20 s ), broad
(pSPZ10), yielding 1.2kg of wet cells.
Overall, approx-
cat
[
4]
substrate spectra, and environmentally friendly reaction
conditions. Chiral epoxides can be obtained through various
biocatalytic methods. The most elegant approach is the
asymmetric synthesis from the corresponding olefins by using
imately 15 g of technical-grade StyA was obtained in one
purification step from 280 g of wet cells (Table 1). This
[
5]
[
6]
Table 1: Parameters for StyA enrichment during expanded bed chroma-
tography.
[
7,8]
oxidoreductases.
Oxygenases are particularly interesting
biocatalysts because of their ability to activate molecular
oxygen in situ. Reactive oxygen donors such as peracids,
ozone, and iodosylbenzene are thus becoming obsolete.
The specific technical approach to a particular biocatalytic
process must take into account the performance of the
relevant enzyme under process conditions. For oxygenases,
the best catalysts developed thus far are whole cell systems,
which stabilize the often complex heteromultimeric enzyme
Sample
Protein StyA Specific
Total
activity [%]
[kU]
Yield
[a]
[g]
[g]
activity
[Umg
À1 [b]
]
Crude cell extract
Fractions containing StyA
59.7
25.3
16.1 0.57
15.2 0.89
24.3
22.5
100
93
[c]
[d]
[a] Calculated from purities in the enzyme samples. [b] 1 U (international
unit) corresponds to 1 mmol product formed per minute. [c] Upon
stepwise elution with KCl two StyA-containing fractions were obtained
(
up to four individual enzyme subunits) and enable coenzyme
(45% and 70% purity). [d] Average activity of the two different pools
regeneration based on cell metabolism. At the same time, the
use of whole cells introduces further complexity in a
biocatalytic process owing to the frequently observed over-
oxidation and the functional necessity of cellular and
metabolic integrity of the biocatalyst. Products have to be
obtained after expanded bed chromatography.
corresponds to more than 22000 U. Thus a total of 60 g of
catalyst (almost 100000 U) can be obtained from a single 30-L
fermentation within one week. Lyophilization was used to
attain a stable biocatalyst preparation, and full specific
[*] Dipl.-Natw. K. Hofstetter, Dipl.-Biotech. J. Lutz, Dipl.-Ing. I. Lang,
[
14]
enzyme activity was retained for at least three months.
Prof. Dr. B. Witholt, Dr. A. Schmid
This biocatalyst powder was used for the production of
selected aromatic epoxides. Previous results (unpublished
data) indicated that StyA is deactivated in the presence of
high substrate and product concentrations, possibly caused by
the covalent attachment of the epoxides to nucleophilic
amino acid residues of proteins. To circumvent StyA deacti-
vation, a biphasic reaction setup was applied in which the
organic phase serves as substrate reservoir and product sink
Institute of Biotechnology, HPT
Swiss Federal Institute of Technology Zurich
Wolfgang-Pauli Strasse 16
8093 Zurich (Switzerland)
Fax: (+41)16-331-051
E-mail: andreas.schmid@biotech.biol.ethz.ch
[
**] StyB was kindly provided by Dr. Katja Otto. Dipl.-Ing. Ulrich Bauer is
gratefully acknowledged for excellent technical assistance. The
authors thankDr. FrankHollmann and Dr. Bruno Bühler for critical
reading of the manuscript. The project was financially supported by
the EU (QLRT-1999-00439).
(Figure 1).
Dodecane was used as the organic phase so that the
reactant concentrations in the aqueous, enzyme-containing
phase could be maintained at values that did not significantly
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2004, 43, 2163 –2166
DOI: 10.1002/anie.200353338
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2163

Communications
Figure 1. Reaction pathway during biocatalytic epoxidation in a biphasic system. The organic
phase serves as a substrate reservoir and product sink. In the aqueous phase, FDH and for-
mate were used for the regeneration of NADH. StyB transfers the reducing equivalents from
1
NADH to FAD. FADH and oxygen are cosubstrates for olefin epoxidation by StyA. R =H,Cl;
2
2
3
R =R =H,CH .
3
[
15]
affect StyA activity.
An emulsion was formed to avoid
mass-transfer limitations across the phase boundary. Enzyme
denaturation at the liquid–liquid interface was apparent from
the formation of white protein aggregates, but could be
almost completely avoided by the addition of bovine serum
albumin (BSA). This prolongs StyA epoxidation activity from
several minutes to hours (data not shown). StyA was supplied
with the reducing equivalents needed for the activation of
molecular oxygen by its native NADH:flavin reductase
[
16]
StyB. Concomitant NADH regeneration was achieved by
the formate/formate dehydrogenase (FDH) system of Pseu-
domonas sp. 101, which has a higher solvent tolerance and
Figure 2. Biocatalytic epoxidation of different styrene derivatives by
StyAB in a biphasic system. The time courses of substrate depletion
) are shown for the epoxidation of 3-
chlorostyrene (a), a-methylstyrene (b), and trans-b-methylstyrene (b).
The substrate was pulsed at the time points indicated by arrows in
reactions a (a total of 1.9 mmol) and c (2.5 mmol). In the beginning of
reaction a (enclosed in box), oxygen was supplied at a low rate. Imme-
diately after increasing aeration (180 min), the epoxide formation rate
also increased.
&
( ) and product formation (~
[
17]
high substrate affinity than other FDHs.
activity relative to epoxidation activity was achieved by
Maximal StyA
[
18]
adding a four–fivefold excess of FDH. Figure 2shows the
time course of three preparative epoxidation reactions with 3-
chlorostyrene (1), a-methylstyrene (2), and trans-b-methyl-
styrene (3) as substrates. Reaction parameters are given in
Table 2.
Aeration was detrimental to enzyme activity, as protein
denaturation was observed at the gas–liquid interface at high
aeration rates. At the same time, oxygen limitation should be
avoided, as O is a cosubstrate in the epoxidation reaction.
The influence of aeration on StyA epoxidation activity is
shown in Figure 2a (box). A diminished activity was observed
2
Table 2: Data for the conversion of different vinyl aromatics into their enantiopure epoxides in a cell-free biphasic system with isolated styrene
monooxygenase.
[
c]
[d]
[e]
Substrate
Product
Substrate [mm] Conversion [%] Initial
Average
TNStyA
TNNAD
ee [%] Yield [%]
product-formation product-formation
À1 [a]
À1 [b]
rate [Ug
108
]
rate [Ug
48
]
[
f]
5
5
5
0 (+2.5)
90.5
87.9
95.3
2171
1844
2867
66
56
87
>99.9 73
98.1 75
99.7 87
0
112
118
44
66
[
f]
0 (+1.9)
[a] Activity measured for the first 30 min. [b] Average activity measured over the entire reaction period. [c] TNStyA: Total turnover number for StyA=
(
moles of product formed)/(moles of StyA). The amount of StyA was calculated based on the purity (70%) of the lyophilized sample. [d] TNNAD: Total
+
+
turnover number for NAD =(moles of product formed)/(moles of NAD in the reaction). [e] Yields of isolated product after purification. [f] In
parenthesis, the amount of substrate (in mmol) pulsed during the reaction is given.
2
164
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 2163 –2166

Angewandte
Chemie
at a low aeration rate at the beginning of the reaction. StyA
activity could be recovered by increasing the air supply. A
compromise with respect to the aeration rate was found by
regulating the dissolved oxygen in the reaction mixture at
by SDS-PAGE (sodium dodecyl sulfate–polyacrylamide gel electro-
[
25]
phoresis), and StyA activity was measured.
StyA-containing
samples were supplemented with 1% (v/v) sucrose and 2% (v/v)
mannitol and lyophilized in a Lyovac GT 3 Freeze-Drying Plant
(Leybold-Heraeus GmbH, Germany).
1
0% of oxygen saturation. Under these conditions, gram-
scale production of each epoxide with ee values higher than
8% and yields up to 87% were achieved; no by-products
were detected besides trace amounts of diols (from sponta-
Biphasic reaction: The preparative biphasic conversion of the
different styrene derivatives was performed in 350-mL Sixfors
reactors (Infors HT, Switzerland) with pH and temperature control
(pH 7.5, 308C) at a stirrer speed of 400 rpm. The aqueous phase
9
À1
À1
[
19]
contained StyA (2gL ), StyB (0.03 gL ), FAD (0.1 mm), sodium
neous, non-enzymatic hydrolysis), phenethyl alcohols, and
phenylacetaldehydes.
À1
formate (50 mm), FDH (8 UmL ; Jülich Fine Chemicals, Germany),
Catalase (250 UmL ; Fluka AG, Switzerland), BSA (2gL ; Sigma-
Aldrich Chemie GmbH, Germany), and NAD (1 mm) in a total
À1
À1
We showed the simple and fast preparation of an easily
applicable oxygenase in multigram amounts. Despite the
early development stage, we were able to use this enzyme to
produce enantiopure epoxides in a simple and scalable
+
volume of 100 mL in Tris buffer (50 mm; pH 7.6). The organic phase
was composed of dodecane (100 mL) that contained the substrate
(50 mm). During the reaction, formate (1 mmol) was added every 1 h.
After reaction termination, the phases were separated by centrifuga-
tion and the enantiomeric excess was determined by normal-phase
reaction setup. Volumetric product formation rates
À1 À1
(
ꢀ 1 gL h ) are already in the range of those reported for
[
26]
[11,20,21]
high-pressure liquid chromatography (HPLC).
whole-cell oxygenations.
For the first time, nearly
Product purification: A silica-gel column (Fluka AG, Switzer-
land) was used to separate the product from substrate and dodecane.
The substrate was eluted with hexane containing 1% triethylamine,
and the product was eluted with hexane containing 1% triethylamine
and 10% diethyl ether to afford 1a (0.78 g), 2a (0.51 g), and 3a
complete conversion was achieved with high reactant con-
centrations, which was not possible so far in other applications
[22]
with isolated oxygenases in a biphasic system. For StyA,
total catalyst turnover numbers and average reaction rates of
À1
(0.88 g).
1
800–2800 and 3–4.3 min , respectively, were achieved under
process conditions. These values are already promising and
compare well with chemical catalysts for which total turn-
Received: November 17, 2003 [Z53338]
À1
overs below 1000 and frequencies mostly below 1 min have
[
23]
Keywords: asymmetric synthesis · biphasic catalysis ·
been reported. Furthermore, the purification of the prod-
ucts from the reaction mixture is simple. Overall, StyA was
shown to epoxidize sterically hindered C-C double bonds,
especially trans olefins, complementing the product spectrum
.
coenzymes · enzyme catalysis · epoxidation
[
1] T. Katsuki, Coord. Chem. Rev. 1995, 140, 189 – 214.
[
1,24]
of commonly applied chemical catalysts.
We observed a
[
2] a) T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102,
significant decrease of StyA activity (eightfold lower than in
biphasic short-term assays). Further investigations will there-
fore focus on increasing the turnover number of StyA.
Overall, the modular character of the presented biocata-
lytic reaction allows the substitution of single components to
achieve not only epoxidations but also specific hydroxyla-
tions, as well as Baeyer–Villiger or heteroatom oxidations.
This study has shown that cell-free oxygenase catalysis can be
a versatile tool for the asymmetric synthesis of enantiopure
oxyfunctionalized hydrocarbons.
5
974 – 5976; b) E. N. Jacobsen in Catalytic Asymmetric Synthesis,
VCH, New York, 1993, pp. 159 – 201; c) J. T. Groves, R. S. Myers,
J. Am. Chem. Soc. 1983, 105, 5786 – 5791; d) C. Bolm, Angew.
Chem. 1991, 103, 414 – 415; Angew. Chem. Int. Ed. Engl. 1991,
3
0, 403 – 404.
3] a) J. R. L. Smith, G. Reginato, Org. Biomol. Chem. 2003, 1,
543 – 2549; b) J.-L. Zhang, Y.-L. Liu, C.-M. Che, Chem.
[
2
Commun. 2002, 23, 2906 – 2907; c) E. BrulØ, Y. R. de Miguel,
Tetrahedron Lett. 2002, 43, 8555 – 8558; d) B. Clapham, T. S.
Reger, K. D. Janda, Tetrahedron 2001, 57, 4637 – 4662.
[4] S. M. Resnick, K. Lee, D. T. Gibson, J. Ind. Microbiol. 1996, 17,
38 – 457.
4
[
5] R. B. Silverman, The Organic Chemistry of Enzyme-Catalyzed
Reactions, 1st ed., Academic Press, San Diego.
Experimental Section
[6] For a general overview, see: a) E. J. de Vries, D. B. Janssen, Curr.
Opin. Biotechnol. 2003, 14, 414 – 420; b) A. Archelas, R.
Furstoss, Top. Curr. Chem. 1999, 200, 160 – 191; c) J. A. M.
de Bont, Tetrahedron: Asymmetry 1993, 4, 1331 – 1340.
[7] a) E. J. Allain, L. P. Hager, L. Deng, E. N. Jacobsen, J. Am.
Chem. Soc. 1993, 115, 4415 – 4416; b) S. Colonna, N. Gaggero, L.
Casella, G. Carrea, P. Pasta, Tetrahedron: Asymmetry 1993, 4,
1325 – 1330; c) F. P. Guengerich, Arch. Biochem. Biophys. 2003,
409, 59 – 71; d) M. Sono, M. P. Roach, E. D. Coulter, J. H.
Dawson, Chem. Rev. 1996, 96, 2841 – 2887; e) M. G. Wubbolts, S.
Panke, J. B. van Beilen, B. Witholt, Chimia 1996, 50, 436; f) B. J.
Wallar, J. D. Lipscomb, Chem. Rev. 1996, 96, 2625 – 2658; g) E. I.
Solomon, T. C. Brunold, M. I. Davis, J. N. Kemsley, S.-K. Lee, N.
Lehnert, F. Neese, A. J. Skulan, Y.-S. Yang, J. Zhou, Chem. Rev.
2000, 100, 235 – 349; h) P. D. Gennaro, A. Colmegna, E. Galli, G.
Sello, F. Pelizzoni, G. Bestetti, Appl. Environ. Microbiol. 1999,
65, 2794 – 2797.
Chemicals: All substrates were obtained from Sigma-Aldrich Chemie
GmbH, Germany. Epoxide standards were prepared and character-
[
11]
ized as described in the literature.
Flavin adenine dinucleotide
(
(
FAD), sodium formate, b-nicotinamide adenine dinucleotide
NAD ), buffer components, and solvents were obtained from
+
Fluka AG, Switzerland.
Cell Cultures: E. coli JM101 containing the plasmid pSPZ10
carrying the styrene monooxygenase genes styAB) were grown as
[
8]
(
[
21]
described in the literature,
but in a single-phase rather than a
biphasic medium. The expression of styAB was induced with
dicyclopropylketone (0.05% v/v; Fluka AG, Switzerland).
Large-scale enrichment of StyA: Wet cells (280 g) were disrupted
in a bead mill and the resulting crude extract was loaded onto 500 mL
of a Streamline DEAE matrix in the expanded-bed mode through a
5
0
Streamline column. After the flowthrough had become clear, the
packed-bed mode was applied. A stepwise salt gradient at 0, 0.16, and
0
.24 m KCl was used for elution. During elution 100-mL fractions
[8] S. Panke, M. G. Wubbolts, A. Schmid, B. Witholt, Biotechnol.
Bioeng. 2000, 69, 91 – 100.
were collected. The amount of StyA in every fraction was determined
Angew. Chem. Int. Ed. 2004, 43, 2163 –2166
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2165

Communications
[
9] V. L. Vilker, V. Reipa, M. Mayhew, M. J. Holden, J. Am. Oil
Chem. Soc. 1999, 76, 1283 – 1289.
10] a) F. Hollmann, B. Witholt, A. Schmid, J. Mol. Catal. B 2002, 19–
0, 167 – 176; b) C.-H. Wong, G. M. Whitesides, J. Am. Chem.
[
2
Soc. 1981, 103, 4890 – 4899; c) H. K. Chenault, G. M. Whitesides,
Bioorg. Chem. 1989, 17, 400 – 409; d) W. A. van der Donk, H.
Zhao, Curr. Opin. Biotechnol. 2003, 14, 421 – 426; e) Z. Shaked,
G. M. Whitesides, J. Am. Chem. Soc. 1980, 102, 7104 – 7105.
11] A. Schmid, K. Hofstetter, H.-J. Feiten, F. Hollmann, B. Witholt,
Adv. Synth. Catal. 2001, 343, 732– 737.
12] K. Otto, K. Hofstetter, B. Witholt, A. Schmid in Flavins and
Flavoproteins (Eds.: S. Chapman, R. Perham, N. Scrutton),
Rudolf Weber, Berlin, 2002, pp. 1027 – 1033.
[
[
[
13] Efficient heterologous expression of oxygenase genes is ach-
ieved using the alk regulatory system of Pseudomonas putida
GPo1: a) I. E. Staijen, J. B. van Beilen, B. Witholt, Eur. J.
Biochem. 2000, 267, 1957 – 1965; b) S. Panke, V. de Lorenzo, A.
Kaiser, B. Witholt, M. G. Wubbolts, Appl. Environ. Microbiol.
1999, 65, 5619 – 5623.
[
[
14] Data are available as Supporting Information.
15] The partition coefficient for styrene and styrene oxide in
dodecane as organic solvent are 500 and 50, respectively. Thus,
only minor product inhibition on StyA epoxidation was expected
during the reaction.
[
[
16] Reducing equivalents for the epoxidation reaction can also be
supplied by other NAD(P)H:flavin reductases or chemical
mediators. StyB can easily be prepared from recombinant
E. coli by inclusion body isolation and refolding (see refer-
ence [12]).
17] V. I. Tishkov, A. G. Galkin, G. N. Marchenko, O. A. Egorova,
D. V. Sheluho, L. B. Kulakova, L. A. Dementieva, A. M. Egorov,
Biochem. Biophys. Res. Commun. 1993, 192, 976 – 981.
18] Data on StyA:FDH ratio are available as Supporting Informa-
tion.
19] Data on the stability of the epoxides are available as Supporting
Information.
20] a) S. D. Doig, L. M. O'Sullivan, S. Patel, J. M. Ward, J. M.
Woodley, Enzyme Microb. Technol. 2001, 28, 265 – 274; b) B.
Bühler, I. Bollhalder, B. Hauer, B. Witholt, A. Schmid,
Biotechnol. Bioeng. 2003, 82, 833 – 842.
[
[
[
[
[
21] S. Panke, M. Held, M. G. Wubbolts, B. Witholt, A. Schmid,
Biotechnol. Bioeng. 2002, 80, 33 – 41.
22] a) A. Schmid, I. Vereyken, M. Held, B. Witholt, J. Mol. Catal. B
2001, 11, 455 – 462; b) J. Lutz, V. V. Mozhaev, Y. L. Khmelnitsky,
B. Witholt, A. Schmid, J. Mol. Catal. B 2002, 19–20, 177 – 187;
c) U. Schwarz-Linek, A. Krödel, F.-A. Ludwig, A. Schulze, S.
Rissom, U. Kragl, V. I. Tishkov, M. Vogel, Synthesis 2001, 6, 947 –
951.
[
23] a) P. Pietikäinen, J. Mol. Catal. A 2001, 165, 73 – 79; b) G. Pozzi,
M. Cavazzini, F. Cinato, F. Montanari, S. Quici, Eur. J. Org.
Chem. 1999, 1999, 1947 – 1995; c) S. H. R. Abdi, R. I. Kureshy,
N. H. Khan, M. M. Bhadbhade, E. Suresh, J. Mol. Catal. A 1999,
150, 185 – 194.
[
24] a) N. Hosoya, A. Hatayama, R. Irie, H. Sasaki, T. Katsuki,
Tetrahedron 1994, 50, 4311 – 4322; b) R. Zhang, W.-Y. Yu, K.-Y.
Wong, C.-M. Che, J. Org. Chem. 2001, 66, 8145 – 8153; c) K.-H.
Ahn, S. W. Park, S. Choi, H.-J. Kim, C. J. Moon, Tetrahedron
Lett. 2001, 42, 2485 – 2488.
[
[
25] Standard assay conditions are available as Supporting Informa-
tion.
26] Standard conditions for the determination of ee values are
available as Supporting Information.
2
166
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 2163 –2166