Directed evolution of hybrid enzymes: Evolving enantioselectivity of an achiral Rh-complex anchored to a protein

Article abstract of DOI:10.1039/b610461d

The concept of utilizing the methods of directed evolution for tuning the enantioselectivity of synthetic achiral metal-ligand centers anchored to proteins has been implemented experimentally for the first time. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b610461d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Directed evolution of hybrid enzymes: Evolving enantioselectivity of an
achiral Rh-complex anchored to a protein
Manfred T. Reetz,* Je´roˆme J.-P. Peyralans, Andrea Maichele, Yu Fu and Matthias Maywald
Received (in Cambridge, UK) 21st July 2006, Accepted 16th August 2006
First published as an Advance Article on the web 5th September 2006
DOI: 10.1039/b610461d
system (Scheme 1). As in the case of directed evolution of
enzymes,1 multiple cycles are possible which may be needed.6
The experimental implementation of this concept is challenging
because several non-trivial prerequisites need to be fulfilled: (1) The
host protein should be robust; (2) An efficient expression system,
delivering sufficient protein in miniaturized and parallelized form
has to be available; (3) The host protein mutants need to be
purified in parallel; (4) Chemical modification with introduction of
a metal–ligand entity should be essentially quantitative; (5) A high-
throughput screening system has to be available for a given
reaction of interest.8
Parallel to efforts employing covalent chemical modification,6
we proposed in 2002 the non-covalent variant as an alternative,6b
specifically the use of the Whitesides system9 based on the strong
interaction of a chemically modified biotin with the host protein
avidin. In that system biotin is first attached covalently to an
achiral diphosphine–Rh complex via a spacer to form complex 1
which then binds with high affinity to avidin. The wild-type (WT)
avidin–1 complex was used as a single catalyst in the Rh-catalyzed
hydrogenation of a-acetamido-acrylic acid, the resulting ee ranging
between 33 and 44% depending upon the conditions used.9a Ward
has used chemical tuning and rational protein design to improve
the enantioselectivity significantly in the same reaction,10 which is
fundamentally different from our Darwinian approach. In the
present project we likewise employed the biotinylated dipho-
sphine–Rh complex 1, but chose to use the esterified substrate 2
because this facilitates parallel analysis by gas chromatography
(Scheme 2). The reaction mixtures can be extracted with ethyl
acetate in a parallel manner, whereas the acid of the Whitesides–
Ward system is accessible efficiently only by continuous extraction.
The concept of utilizing the methods of directed evolution for
tuning the enantioselectivity of synthetic achiral metal–ligand
centers anchored to proteins has been implemented experimen-
tally for the first time.
Directed evolution is a powerful method for enhancing the
stability, activity and/or selectivity of enzymes.1 This approach to
protein engineering comprises repeating cycles of appropriate
random gene mutagenesis, expression of mutant enzymes and
high-throughput screening (or selection) for a given catalytic
property. In 1997 we showed that this strategy can be used to
enhance the enantioselectivity of enzymes catalyzing the transfor-
mation of unnatural substrates;2 later reversal of enantioselectivity
was demonstrated.3 This novel approach to asymmetric catalysis
has since been applied to many different types of enzymes.4
However, enzymes are incapable of catalyzing many if not most
reactions that chemists have invented using transition metals.5
Therefore, we have previously proposed the use of directed
evolution as a means to tune hybrid catalysts.6 This concept is
based on the well-known fact that it is possible to chemically
modify a protein by covalent or non-covalent anchoring of a
metal–ligand moiety to produce an individual potential catalyst.7
In our approach a library of protein mutants is first produced by
random mutagenesis and then subjected to an appropriate post-
translational chemical modification. This provides a pool of
mutant hybrid catalysts, each having a synthetic metal center in a
different protein environment (Scheme 1).
Subsequent screening for enantioselectivity leads to the
identification of an improved mutant (hit), and the mutant gene
encoding the respective protein can then be used as a template for
another cycle, thereby imparting evolutionary pressure on the
Scheme 1 Directed evolution of hybrid catalysts in which metal (M)–
ligand is anchored non-covalently.
Max-Planck-Institut fu¨r Kohlenforschung, D-45470 Mu¨lheim/Ruhr,
Germany. E-mail: reetz@mpi-muelheim.mpg.de; Fax: +49 208 306 2985
Scheme 2 Rh-catalyzed hydrogenation of a-acetamido-acrylic acid ester.
This journal is ß The Royal Society of Chemistry 2006
4318 | Chem. Commun., 2006, 4318–4320

View Article Online
Although several expression systems for avidin have been
described, production of eukaryotic proteins is both time
consuming and low yielding. We therefore turned to streptavidin,
a genetically unrelated bacterial protein which also binds biotin
with high affinity. Several expression systems for streptavidin
have been described,11 and some of them were compared.
Unfortunately, problems arose with the expression level and
purification in parallel form. The best solution for our needs
is based on pET11b-sav.12 This construct encodes 12 residues of
T7-tag followed by Asp and Gln and residues 15 to 159 of the
mature streptavidin. Esherichia coli strain BL21(DE3) transformed
with this plasmid and grown in Studier’s auto-induction media,13
ZYP5052, requires less monitoring than conventional induction
with IPTG at the mid-log phase and thus allows multiple
unattended overnight cultures. This adaptation was crucial in
simplifying the screening task. However, the system is still not fully
suited for screening thousands of mutants since it requires a
150 mL culture scale to provide a sufficient amount of streptavidin.
Typically we used 1.04 6 1027 mol of binding site, which is
equivalent to y1.7 mg protein (based on a MW of 16.5 kDa per
monomer and expecting 3.8–3.9 free binding sites per tetramer as
obtained for the WT). However, when lower amounts of
streptavidin were obtained for a given mutant, the culture scale
was increased by up to five-fold and/or the amount of hybrid
catalyst used in the reaction was decreased to 0.1%. Titrated
mutant streptavidins were transferred into glass vessels of an in-
house adapted reactor block14 for the Chemspeed Accelerator2
SLT 100 Synthesizer.
Fig. 1 Selected close (purple) and distal (blue) sites from Rh(I) centers
(red) of two important calculated conformers of the complex.
In some cases protein variants were obtained in amounts too
small for fulfilling the experimental prerequisites for reproducible
hydrogenation, and these were not considered for further study.
Nevertheless, a few mutants showing enantioselectivity different
from the WT were observed in a reproducible manner, the best
variant I leading to 35% ee (R). It is characterized by mutation
Ser112Gly. However, a single round of saturation mutagenesis
does not yet constitute an evolutionary process. Therefore, iterative
CASTing was performed using the gene that encodes mutant I and
saturating at position 49. This led to a double mutant II having
mutations Asn49His/Ser112Gly and showing an ee-value of 54%
(R) in the model reaction (Scheme 3). Finally, a third-generation
saturation experiment was performed using the gene which
encodes mutant II and focusing once more on position 112. This
experiment was designed to test whether glycine at position 112 is
really the best choice when combining with histidine at position 49.
An improved mutant III was identified leading to an ee of 65% (R)
in which, surprisingly, the original mutation Ser112Gly was
We therefore settled for only a few hundred mutants in each
mutagenesis experiment and proceeded with directed evolution on
a ‘‘small scale’’ which in fact was still labor-intensive. In an initial
experiment, the WT-streptavidin–1 was found to be a poor catalyst
in the hydrogenation of 2, leading to an ee of only 23% in favour
of (R)-3. Rather than targeting the whole protein for amino acid
substitution by error-prone PCR,1 we applied CASTing (CAST =
Combinatorial Active-site Saturation Test),15 in which appropriate
amino acid sites next to the binding pocket are randomized by
saturation mutagenesis, specifically in an iterative manner.16 In the
present case an X-ray structure of the conjugate has not been
reported. Therefore, the CAST sites for amino acid randomization
were chosen on the basis of modeling the biotinylated Rh-complex
1 into the X-ray structure of streptavidin–biotin.17 Fig. 1 shows an
excerpt of the modeled structure and the amino acid sites that
appeared to be appropriate for CAST experiments.
The positions that we considered for saturation mutagenesis can
be classified into two categories. The first are ‘‘close’’ positions, for
example Asn49, Leu110, Ser112 and Leu124, which are located
˚
about 4 to 6 A away from the Rh(I) of the two calculated major
conformers. These could influence directly the conformation of the
catalyst or catalyst–substrate complex. The second type of
positions are ‘‘distal’’: Glu51, Tyr54, Trp79, Asn81, Arg84,
Asn85, His87, which are located further away from Rh(I). These
‘‘second sphere’’ CAST-positions could influence the structure of
the enzyme as a whole because they are involved in hydrogen
bonding between secondary elements.
We started saturation mutagenesis at positions 110, 112, and
124 using the QuikChange method (Stratagene) and pET11b-
sav.12 In each saturation experiment about 200–300 clones
(oversampling for .95% coverage) were harvested and tested.
Scheme 3 Directed evolution of hybrid catalysts comprising streptavi-
din–1, the Rh-catalyzed hydrogenation 2 A 3 serving as the model
reaction (40–90% yield).
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4318–4320 | 4319

View Article Online
4 Comprehensive review of directed evolution of enantioselective enzymes:
M. T. Reetz, in Advances in Catalysis, ed. B. C. Gates and H. Kno¨zinger,
Elsevier, San Diego, 2006, vol. 49, pp. 1–69.
reverted back to serine (Scheme 3). Thus, the best variant of
this study is characterized by a single mutation (Asn49Val).
The increase in selectivity corresponds to about DDG?
=
5 (a) G. W. Parshall and S. D. Ittel, Homogeneous Catalysis; The
Application and Chemistry of Catalysis by Soluble Transition Metal
Complexes, 2nd edn, John Wiley & Sons, Inc., New York, 1992; (b)
B. Cornils and W. A. Herrmann, Applied Homogeneous Catalysis with
Organometallic Compounds, Wiley-VCH, Weinheim, 1996, vol. 1–2.
6 (a) M. T. Reetz, DE-A 101 29 187.6, 2001; (b) M. T. Reetz, Tetrahedron,
2002, 58, 6595–6602; (c) M. T. Reetz, M. Rentzsch, A. Pletsch and
M. Maywald, Chimia, 2002, 56, 721–723.
2.7 kJ 6 mol21. Mutagenesis experiments at positions 51, 54,
79, 81, 84, 85 and 87 were not successful because no soluble
protein was obtained. In contrast, saturation mutagenesis at
position 49 on WT template led directly to mutants III and IV.
The latter is characterized by Asn49His. This variant has lower
enantioselectivity than the WT (ee = 8% (R)), suggesting the
possibility of inverting stereoselectivity. The plasmid encoding
variant IV was utilized as a template for saturation mutagenesis.
Indeed, upon focusing on position 124, mutant V (Asn49His/
Leu124Phe) was identified which is (S)-selective, although not by a
great degree (ee = 7%). The essential evolutionary steps are
summarized in Scheme 3. We observed no clear relationship
between enantioselectivity and rate, although differences in activity
were observed.
7 (a) D. Qi, C.-M. Tann, D. Haring and M. D. Distefano, Chem. Rev.,
2001, 101, 3081–3111; (b) L. Polgar and M. L. Bender, J. Am. Chem.
Soc., 1966, 88, 3153–3154; (c) P. G. Schultz, Science, 1988, 240, 426–433;
(d) K. Khumtaveeporn, G. DeSantis and J. B. Jones, Tetrahedron:
Asymmetry, 1999, 10, 2563–2572; (e) H. B. Smith and F. C. Hartman,
J. Biol. Chem., 1988, 263, 4921–4925; (f) K. M. Nicholas, P. Wentworth,
Jr., C. W. Harwig, A. D. Wentworth, A. Shafton and K. D. Janda,
Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 2648–2653; (g) I. Hamachi and
S. Shinkai, Eur. J. Org. Chem., 1999, 539–549; (h) Y. Lu, Curr. Opin.
Chem. Biol., 2005, 9, 118–126; (i) Y. Lu and J. S. Valentine, Curr. Opin.
Struct. Biol., 1997, 7, 495–500; (j) Y. Lu, S. M. Berry and T. D. Pfister,
Chem. Rev., 2001, 101, 3047–3080; (k) P. D. Barker, Curr. Opin. Struct.
Biol., 2003, 13, 490–499, see also: (l)E. T. Kaiser, Angew. Chem., 1988,
100, 945–955, (Angew. Chem., Int. Ed. Engl., 1988, 27, 913–922); (m)
D. Hilvert and E. T. Kaiser, Biotechnol. Genet. Eng. Rev., 1987, 5,
297–318; (n) J. G. de Vries and L. Lefort, Chem.–Eur. J., 2006, 12,
4722–4734; (o) C. A. Kruithof, M. A. Casado, G. Guillena,
M. R. Egmond, A. van der Kerk-van Hoof, A. J. R. Heck, R. J. M.
Klein Gebbink and G. van Koten, Chem.–Eur. J., 2005, 11, 6869–6877;
(p) L. Panella, J. Broos, J. Jin, M. W. Fraaije, D. B. Janssen,
M. Jeronimus-Stratingh, B. L. Feringa, A. J. Minnaard and J. G. de
Vries, Chem. Commun., 2005, 5656–5658; (q) B. G. Davies, Curr. Opin.
Biotechnol., 2003, 14, 379–386.
8 M. T. Reetz, Angew. Chem., 2001, 113, 292–320, (Angew. Chem., Int.
Ed., 2001, 40, 284–310).
9 (a) M. E. Wilson and G. M. Whitesides, J. Am. Chem. Soc., 1978, 100,
306–307, see also: (l) C.-C. Lin, C.-W. Lin and A. S. C. Chan,
Tetrahedron: Asymmetry, 1999, 10, 1887–1893.
10 (a) J. Collot, J. Gradinaru, N. Humbert, M. Skander, A. Zocchi and
T. R. Ward, J. Am. Chem. Soc., 2003, 125, 9030–9031; (b) C. M. Thomas
and T. R. Ward, Chem. Soc. Rev., 2005, 34, 337–346.
11 (a) E. A. Bayer, H. Ben-Hur and M. Wilchek, Methods Enzymol., 1990,
184, 80–89; (b) L. D. Thompson and P. C. Weber, Gene, 1993, 136,
243–246; (c) T. Sano and C. R. Cantor, Proc. Natl. Acad. Sci. U. S. A.,
1990, 87, 142–146; (d) S.-C. Wu, M. H. Qureshi and S. L. Wong, Protein
Expression Purif., 2002, 24, 348–356; (e) S.-C. Wu and S.-L. Wong,
Appl. Environ. Microbiol., 2002, 68, 1102–1108.
In summary, our work demonstrates that it is possible to apply
the methods of directed evolution to increase and/or to invert
enantioselectivity of a hybrid catalyst composed of a synthetic
achiral transition metal catalyst anchored to a host protein. Due to
the problems associated with the expression system, the full
potential of this novel approach to asymmetric catalysis was not
tested in the present system. However, proof-of-principle has been
achieved for the first time, providing incentive for designing and
testing other systems.18
We thank T. Sano (Columbia University) for plasmid pUC-SZ.
Support from EU MRTN-CT-2003-505020 (IBAAC) and the
Fonds der Chemischen Industrie is gratefully acknowledged.
Notes and references
1 (a) Methods and Molecular Biology, (Directed Enzyme Evolution:
Screening and Selection Methods), ed. F. H. Arnold and G. Georgiou,
Humana Press, Totowa, New Jersey, 2003, vol. 230; (b) Evolutionary
Methods in Biotechnology (Clever Tricks for Directed Evolution), ed.
S. Brakmann and A. Schwienhorst, Wiley-VCH, Weinheim, Germany,
2004; (c) S. V. Taylor, P. Kast and D. Hilvert, Angew. Chem., 2001, 113,
3408–3436, (Angew. Chem., Int. Ed., 2001, 40, 3310–3335); (d)
K. A. Powell, S. W. Ramer, S. B. del Cardayre´, W. P. C. Stemmer,
M. B. Tobin, P. F. Longchamp and G. W. Huisman, Angew. Chem.,
2001, 113, 4068–4080, (Angew. Chem., Int. Ed., 2001, 40, 3948–3959); (e)
N. J. Turner, Trends Biotechnol., 2003, 11, 474–478; (f) S. Lutz and
W. M. Patrick, Curr. Opin. Biotechnol., 2004, 15, 291–297.
2 (a) M. T. Reetz, A. Zonta, K. Schimossek, K. Liebeton and K.-E.
Jaeger, Angew. Chem., 1997, 109, 2961–2963, (Angew. Chem., Int. Ed.
Engl., 1997, 36, 2830–2832); (b) K. Liebeton, A. Zonta, K. Schimossek,
M. Nardini, D. Lang, B. W. Dijkstra, M. T. Reetz and K.-E. Jaeger,
Chem. Biol., 2000, 7, 709–718; (c) M. T. Reetz, S. Wilensek, D. Zha and
K.-E. Jaeger, Angew. Chem., 2001, 113, 3701–3703, (Angew. Chem., Int.
Ed, 2001, 40, 3589–3591); (d) M. T. Reetz, Proc. Natl. Acad. Sci. U. S. A.,
2004, 101, 5716–5722.
12 A. Gallizia, C. de Lalla, E. Nardone, P. Santambrogio, A. Brandazza,
A. Sidoli and P. Arosio, Protein Expression Purif., 1998, 14, 192–196.
13 F. W. Studier, Protein Expression Purif., 2005, 41, 207–234.
14 M. Maywald, Dissertation, Ruhr-Universita¨t Bochum, Germany, 2005.
15 M. T. Reetz, M. Bocola, J. D. Carballeira, D. Zha and A. Vogel,
Angew. Chem., 2005, 117, 4264–4268, (Angew. Chem., Int. Ed., 2005, 44,
4192–4196).
16 M. T. Reetz, L.-W. Wang and M. Bocola, Angew. Chem., 2006, 118,
1258–1263, (Angew. Chem., Int. Ed., 2006, 45, 1236–1241), and
corrigendum, M. T. Reetz, L.-W. Wang and M. Bocola, Angew.
Chem., 2006, 118, 2556, (Angew. Chem., Int. Ed., 2006, 45, 2494).
17 P. C. Weber, D. H. Ohlendorf, J. J. Wendoloski and F. R. Salemme,
Science, 1989, 243, 85–88.
3 (a) D. Zha, S. Wilensek, M. Hermes, K.-E. Jaeger and M. T. Reetz,
Chem. Commun., 2001, 2664–2665; (b) see also: O. May, P. T. Nguyen
and F. H. Arnold, Nat. Biotechnol., 2000, 18, 317–320.
18 M. T. Reetz and N. Jiao, Angew. Chem., 2006, 118, 2476–2479, (Angew.
Chem. Int. Ed., 2006, 45, 2416–2419).
4320 | Chem. Commun., 2006, 4318–4320
This journal is ß The Royal Society of Chemistry 2006