Enhancing the Enantioselectivity of an Epoxide Hydrolase by Directed Evolution

Article abstract of DOI:10.1021/ol035898m

(Matrix presented) The epoxide hydrolase (EH) from Aspergillus niger, which shows a selectivity factor of only E = 4.6 in the hydrolytic kinetic resolution of glycidyl phenyl ether, has been subjected to directed evolution for the purpose of enhancing enantioselectivity. After only one round of error-prone polymerase chain reaction (epPCR), enantioselectivity was more than doubled (E = 10.8). The improved mutant enzyme contains three amino acid exchanges, two of which are spatially far from the catalytically active center.

Full text of DOI:10.1021/ol035898m

ORGANIC
LETTERS
2004
Vol. 6, No. 2
177-180
Enhancing the Enantioselectivity of an
Epoxide Hydrolase by Directed
Evolution
,†
Manfred T. Reetz,* Claudia Torre,† Andreas Eipper,† Renate Lohmer,†
Marcus Hermes,† Birgit Brunner,† Andrea Maichele,† Marco Bocola,†
Michael Arand,‡ Annette Cronin,‡ Yvonne Genzel,§ Alain Archelas,§ and
Roland Furstoss§
Max-Planck-Institut fu¨r Kohlenforschung, Kaiser-Wilhelm-Platz 1,
45470 Mu¨lheim/Ruhr, Germany, Institut fu¨r Pharmakologie und Toxikologie,
UniVersita¨t Wu¨rzburg, Versbacher Strasse 9, 97078 Wu¨rzburg, Germany, and
Faculte´ des Sciences de Luminy, UMR CNRS 6111, Case 901, 163 AV. de Luminy,
13288 Marseille Cedex 9, France
reetz@mpi-muelheim.mpg.de
Received September 29, 2003
ABSTRACT
The epoxide hydrolase (EH) from Aspergillus niger, which shows a selectivity factor of only E ) 4.6 in the hydrolytic kinetic resolution of
glycidyl phenyl ether, has been subjected to directed evolution for the purpose of enhancing enantioselectivity. After only one round of
error-prone polymerase chain reaction (epPCR), enantioselectivity was more than doubled (E ) 10.8). The improved mutant enzyme contains
three amino acid exchanges, two of which are spatially far from the catalytically active center.
The hydrolytic kinetic resolution of chiral epoxides can be
performed either with synthetic transition metal catalysts
(Jacobsen catalysts)1 or enzymatically using epoxide hydro-
lases (EH).2,3 The former option constitutes one of the most
efficient and reliable methods for asymmetric transition metal
catalysis. The potential of the enzymatic approach was
recently increased by the commercialization of three epoxide
hydrolases, specifically the EH from Aspergillus niger and
Rhodococcus rhodochrous, as well as human microsomal
epoxide hydrolase.4 These and other EHs have been dem-
onstrated to be highly active and enantioselective catalysts
for the hydrolytic kinetic resolution of a variety of ep-
oxides.2,3 However, the substrate specificity of enzymes limits
the range of compounds that can be converted enantiose-
lectively. Moreover, some classes of epoxides such as trans-
1,2-disubstituted substrates react with unacceptably low rates,
a problem that has also been encountered in Jacobsen
chemistry. In principle, enzyme properties can be engineered
by “rational design” using site-specific mutagenesis, but in
the case of enantioselectivity, this is far from trivial. As a
viable alternative, we recently introduced the concept of
directed evolution of enantioselective enzymes.5 This strategy
comprises the proper combination of random mutagenesis
† Max-Planck-Institut fu¨r Kohlenforschung.
‡ Universita¨t Wu¨rzburg.
§ UMR CNRS 6111.
(1) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421-431.
(2) (a) Faber, K. Biotransformations in Organic Chemistry, 3rd ed.;
Springer: Berlin, 1997. (b) Drauz, K.; Waldmann, H. Enzyme Catalysis in
Organic Synthesis: A ComprehensiVe Handbook, 2nd ed.; VCH: Weinheim,
2002; Vol. I-III.
(3) (a) Archer, I. V. J. Tetrahedron 1997, 53, 15617-15662. (b) Orru,
R. V. A.; Faber, K. Curr. Opin. Chem. Biol. 1999, 3, 16-21. (c) Steinreiber,
A.; Mayer, S. F.; Faber, K. Synthesis 2001, 2035-2039. (d) Archelas, A.;
Furstoss, R. Curr. Opin. Chem. Biol. 2001, 5, 112-119. (e) Archelas, A.;
Furstoss, R. Trends Biotechnol. 1998, 16, 108-116. (f) Bellucci, G.;
Chiappe, C.; Cordoni, A.; Ingrosso, G. Tetrahedron Lett. 1996, 37, 9089-
9092. (g) Arand, M.; Mu¨ller, F.; Mecky, A.; Hinz, W.; Urban, P.; Pompon,
D.; Kellner, R.; Oesch, F. Biochem. J. 1999, 337, 37-43. (h) Arand, M.;
Wagner, H.; Oesch, F. J. Biol. Chem. 1996, 271, 4223-4229.
(4) Commercially available from Fluka/Aldrich/Sigma.
10.1021/ol035898m CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/20/2003

methods and protein expression6 coupled with efficient high-
throughput screening.7 As a first example, a highly enantio-
selective lipase variant was evolved as a catalyst in the
hydrolytic kinetic resolution of a chiral ester by applying
several cycles of error-prone polymerase chain reaction
(epPCR), saturation mutagenesis, and DNA shuffling.5 It was
shown that amino acid substitutions at positions far away
from the catalytically active site affect enantioselectivity,5e
a surprising phenomenon that was subsequently explained
by a novel relay mechanism.8 On the basis of these positive
results, we have started to apply the concept of directed
evolution of enantioselective enzymes to EHs.9
of a 1:1 mixture of (S)-1 and the pseudo-enantiomer (R)-
D5-1 (Scheme 2). Since the two starting compounds and the
Scheme 2
The hydrolytic kinetic resolution of glycidyl phenyl ether
(1) with formation of diol 2 was known to be catalyzed by
the EH from A. niger, although with low enantioselectivity
(Scheme 1).10 The selectivity factor E (reflecting the relative
two products (S)-2/(R)-D5-2 differ by five mass units, their
respective ratios can be determined by ESI-MS. To reduce
the extent of ESI-MS screening, a pretest14 for epoxide
hydrolase activity was routinely applied, which is based on
the known reaction of an epoxide with 4-p-nitrobenzyl-
pyridine leading to the rapid formation of a blue dye. It has
been parallelized to function efficiently on mictrotiter
plates.14 Thus, reaction mixtures in the wells of microtiter
plates (96- or 384-well format) in which a blue color is not
formed contain active EH mutants. These mutants were then
investigated for enantioselectivity using the ESI-MS assay.
Scheme 1
rate of the reaction of the two enantiomers) amounts to only
4.6, in slight favor of the (S)-product. We therefore chose
this as a model system to test the possibility of improving
enantioselectivity.
Several libraries of mutant EHs were prepared by applying
epPCR under various conditions and transforming into E.
coli BL21(DE3). Single transformants were cultured in liquid
media in a 96-well format, and the pretest for activity was
performed using bacterial cultures and the racemic epoxide
1. Of the first 20 000 clones that were screened, approxi-
mately 20% showed appreciable activity as revealed by the
pretest. The active mutants were then screened by our ESI-
MS-based ee assay leading to the identification of several
dozen mutants with improved enantioselectivity.9 Most of
these exhibited selectivity factors ranging between 5 and 7,
but several had distinctly higher E-values. The latter mutants
were subsequently studied in laboratory scale reactions, the
enantiopurity being determined by traditional HPLC analysis
using chirally modified columns. The E-values were subse-
quently derived using the formula of Sih15 (Table 1).
A prerequisite for the directed evolution of enantioselective
enzymes is the availability of an efficient expression system
and an appropriate high-throughput ee assay.5,7 An expression
system for the EH from A. niger based on the construct pGEF
Asp-EH in E. coli BL21(DE3) was previously described.11
Sufficient quantities of this recombinant wild-type EH, which
allows for commercialization, has also been achieved by
using a heterologous overexpression within another A. niger
strain.4,12 Moreover, in preliminary work, we developed an
ESI-MS-based high-throughput ee assay13 based on the use
(5) (a) Reetz, M. T.; Zonta, A.; Schimossek, K.; Liebeton, K.; Jaeger,
K.-E. Angew. Chem., Int. Ed. Engl. 1997, 36, 2830-2832. (b) Zha, D.;
Wilensek, S.; Hermes, M.; Jaeger, K.-E.; Reetz, M. T. Chem. Commun.
(Cambridge) 2001, 2664-2665. (c) Reetz, M. T.; Wilensek, S.; Zha, D.;
Jaeger, K.-E. Angew. Chem., Int. Ed. 2001, 40, 3589-3591. (d) Reetz, M.
T. Pure Appl. Chem. 2000, 72, 1615-1622. (e) Reetz, M. T. Tetrahedron
2002, 58, 6595-6602.
(6) (a) Arnold, F. H. Nature 2001, 409, 253-257. (b) Powell, K. A.;
Ramer, S. W.; del Cardayre´, S. B.; Stemmer, W. P. C.; Tobin, M. B.;
Longchamp, P. F.; Huisman, G. W. Angew. Chem., Int. Ed. 2001, 40, 3948-
3959. (c) Brakmann, S.; Johnsson, K. Directed Molecular EVolution of
Proteins (or How to ImproVe Enzymes for Biocatalysis); Wiley-VCH:
Weinheim, Germany, 2002. (d) Arnold, F. H.; Georgiou, G. Directed
Enzyme EVolution: Screening and Selection Methods; Humana Press:
Totowa, NJ, 2003; Vol. 230.
(7) Reetz, M. T. Angew. Chem., Int. Ed. 2001, 40, 284-310.
(8) Bocola, M.; Otte, N.; Jaeger, K.-E.; Reetz, M. T.; Thiel, W.
ChemBioChem, in press.
(9) Torre, C. Dissertation, Ruhr-Universita¨t Bochum, Bochum, Germany,
2003.
(10) Genzel, Y.; Archelas, A.; Furstoss, R. Unpublished results.
(11) Arand, M.; Hemmer, H.; Du¨rk, H.; Baratti, J.; Archelas, A.; Furstoss,
R.; Oesch, F. Biochem. J. 1999, 344, 273-280.
(12) Visser, J.; Archelas, A.; Arand, M.; Furstoss, R. Unpublished results.
(13) Schrader, W.; Eipper, A.; Pugh, D. J.; Reetz, M. T. Can. J. Chem.
2002, 80, 626-632.
The most selective EH variant (IS002B1) displays an
E-value of 10.8 in favor of (S)-2. Thus, enantioselectivity
has been more than doubled with respect to the wild-type
enzyme. This mutant, which also appears to be quite active,
is characterized by three amino acid exchanges, A217V,
K332E, and A390E. In 3 out of 11 cases under investigation,
plasmid DNA could not be recovered. This not only
prevented the determination of the respective amino acid
exchanges (Table 1) but also prohibited their use in
subsequent rounds of directed evolution.
(14) Zocher, F.; Enzelberger, M. M.; Bornscheuer, U. T.; Hauer, B.;
Schmid, R. D. Anal. Chim. Acta 1999, 391, 345-351.
(15) Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem.
Soc. 1982, 104, 7294-7299.
178
Org. Lett., Vol. 6, No. 2, 2004

Table 1. Improved Variants of the EH from Aspergillus niger as Catalysts in the Hydrolytic Kinetic Resolution of Epoxide 1
conversion [%]
variant
eep (S) [%]
(reaction time [min])
E
amino acid exchanges
IS002B1
IV001G2
IY001B1
IR003B1
IT001A1
II001G7
IB001C2
IE001H6
IS001H8
IL001D4
WT
74
73
69
69
66
68
64
65
67
62
56
39 (2)
24 (5)
35 (30)
31 (5)
38 (30)
30 (30)
38 (30)
34 (45)
23 (30)
30 (3)
10.8
8.1
7.7
7.4
7.2
6.8
6.6
6.4
6.2
5.5
4.6
A217V, K332E, A390E
not determined
not determined
R219G
F84L, M245L
not determined
F340Y
A327V
A327V
P222S
33 (45)
The X-ray crystal structure of the wild-type EH from A.
niger reveals a dimer comprised of identical subunits.16 Each
monomeric unit contains the catalytic triad (aspartate,
aspartate, histidine) typical for most EHs.2,3 Aspartate 192
attacks the epoxide nucleophilically with formation of a
glycol-monoester intermediate that is hydrolyzed in a second
step. Figure 1 shows the known three-dimensional structure
whereas A217V is somewhat closer. The side chain of the
amino acid at position 217 is located about 9 Å from the
nucleophilic active site D192. Molecular modeling (MOLOC
MD calculations) of the wild-type EH with both enantiomers
of 1 shows that the side chain of residue 217 is in van der
Waals contact with the phenyl-ring of the substrate. A
preliminary model suggests that the larger valine side chain
at residue 217 disfavors the (R)-substrate due to steric
interaction. The role of the remote substitutions K332E and
A390E is more difficult to reveal, requiring more extensive
theoretical investigations. Rare cases of remote effects with
respect to enzyme stability and/or activity were reported
previously by other groups.17
In summary, the present results show that it is possible to
apply the methods of directed evolution in the quest to
develop epoxide hydrolases having enhanced enantioselec-
tivity.18 Further rounds of epPCR, saturation mutagenesis,
and DNA shuffling now need to be applied. However, before
doing so, a problem that hampers systematic optimization
of the present reaction or of the kinetic resolution of other
more difficult substrates needs to be resolved, namely, our
observation that the expression system is genetically unstable.
Whereas this is of little concern when producing large
quantities of single enzyme variants, it causes difficulties
when performing directed evolution. Thus, our immediate
efforts are now directed toward improving the expression
system.19
Figure 1. Crystal structure16 of the monomer of the wild-type EH
from Aspergillus niger showing the catalytic triad (blue), the two
tyrosines that activate the epoxide by H-bonding (purple), and the
three amino acid substitutions (yellow).
(17) (a) Zhao, H.; Arnold, F. H. Prot. Eng. 1999, 12, 47-53. (b) Arnold,
F. H.; Wintrode, P. L.; Miyazaki, K.; Gershenson, A. Trends Biochem. Sci.
2001, 26, 100-106. (c) Iffland, A.; Tafelmeyer, P.; Saudan, C.; Johnsson,
K. Biochem. 2000, 39, 10790-10798. (d) Kumar, M.; Kannan, K. K.; Hosur,
M. V.; Bhavesh, N. S.; Chatterjee, A.; Mittal, R.; Hosur, R. V. Biochem.
Biophys. Res. Commun. 2002, 294, 395-401. (e) Agarwal, P. K.; Billeter,
S. R.; Rajagopalan, P. T. R.; Benkovic, S. J.; Hammes-Schiffer, S. Proc.
Natl. Acad. Sci. 2002, 99, 2794-2799.
(18) Other cases of improving enzyme enantioselectivity or substrate
acceptance by directed evolution:5 (a) Fong, S.; Machajewski, T. D.; Mak,
C. C.; Wong, C.-H. Chem. Biol. 2000, 7, 873-883. (b) May, O.; Nguyen,
P. T.; Arnold, F. H. Nat. Biotechnol. 2000, 18, 317-320. (c) Henke, E.;
Bornscheuer, U. T. Biol. Chem. 1999, 380, 1029-1033. (d) Horsman, G.
P.; Liu, A. M. F.; Henke, E.; Bornscheuer, U. T.; Kazlauskas, R. J. Chem.
Eur. J. 2003, 9, 1933-1939. (e) DeSantis, G.; Wong, K.; Farwell, B.;
Chatman, K.; Zhu, Z.; Tomlinson, G.; Huang, H.; Tan, X.; Bibbs, L.; Chen,
P.; Kretz, K.; Burk, M. J. J. Am. Chem. Soc. 2003, 125, 11476-11477.
of the monomer,16 displaying the active site, the two tyrosines
that bind and activate the epoxide via H-bonding and the
three mutations of the most enantioselective EH-variant
(IS002B1). It can be seen that two mutations (K332E and
A390E) occur at positions remote from the active center,
(16) (a) Zou, J. Y.; Hallberg, B. M.; Bergfors, T.; Oesch, F.; Arand, M.;
Mowbray, S. L.; Jones, T. A. Structure (London) 2000, 8, 111-122. (b)
http://www.pdb.ccdc.cam.ac.uk/pdb.
Org. Lett., Vol. 6, No. 2, 2004
179

Acknowledgment. Support from the European Com-
munity (Project QLRT-2000-00426) and the Fonds der
Chemischen Industrie is gratefully acknowledged.
Supporting Information Available: Details of mutagen-
esis methods, cloning, growth, and expression and type of
screening. This material is available free of charge via the
Internet at http://pubs.acs.org.
(19) Cedrone, F.; Niel, S.; Roca, S.; Bhatnagar, T.; Ait-Abdelkader, N.;
Torre, C.; Krumm, H.; Maichele, A.; Reetz, M. T.; Baratti, J. C. Biocatal.
Biotransform. Submitted.
OL035898M
180
Org. Lett., Vol. 6, No. 2, 2004