Inverting the enantioselectivity of P450pyr monooxygenase by directed evolution

Article abstract of DOI:10.1039/c0cc00735h

We report the first example of directed evolution of a P450 monooxygenase with inverted enantioselectivity for asymmetric biohydroxylation. The biohydroxylation product of the best mutant 1AF4A has an ee of 83% (R) compared to the wild type's ee of 43% (S). The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c0cc00735h

View Article Online / Journal Homepage / Table of Contents for this issue
This article is part of the
2
010 ‘Enzymes & Proteins’
web themed issue
This issue showcases high quality research in the field of enzymes and
proteins.
Please visit the website to access the other papers in this issue:-
http://www.rsc.org/chemcomm/enzymesandproteins

COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Inverting the enantioselectivity of P450pyr monooxygenase
by directed evolutionwz
Weng Lin Tang, Zhi Li* and Huimin Zhao*
ab
b
a
Received 2nd April 2010, Accepted 13th May 2010
First published as an Advance Article on the web 3rd June 2010
DOI: 10.1039/c0cc00735h
We report the first example of directed evolution of a P450
monooxygenase with inverted enantioselectivity for asymmetric
biohydroxylation. The biohydroxylation product of the best
mutant 1AF4A has an ee of 83% (R) compared to the wild
type’s ee of 43% (S).
monooxygenases have focused on altering their catalytic
6
efficiency, substrate specificity, regioselectivity, solvent
7
8
9
10
tolerance, and thermostability. There have also been many
successful attempts to engineer an enzyme for the reversal in
1
1
enantioselectivity, but none of these studies were done with
the P450 monooxygenases and very few on asymmetric
catalysis. While screening a library of self-sufficient P450
BM-3 mutants for changes in regioselectivity, Arnold and
coworkers have also identified some mutants with altered
Enantioselective hydroxylation of non-activated carbon atoms
1
represents a significant challenge in classical organic chemistry.
However, nature has offered us a solution via biocatalysis with
monooxygenases using molecular oxygen as oxidant. Cytochrome
P450 monooxygenases (also known as CYP) constitute the
largest family of heme-containing monooxygenases that can
oxidize a broad range of substrates, often at non-reactive
1
2
enantioselectivity based on an activity assay. Here, we report
the first example where directed evolution, coupled with a
high-throughput enantioselectivity screening system, has
been applied to improve the biohydroxylation enantio-
selectivity of a class I P450 monooxygenase using a prochiral
substrate.
2
carbon centers. Of particular interest is a novel P450pyr
3
enzyme from Sphingomonas sp. HXN-200 that belongs to
the class I P450 proteins and requires the presence of
an electron-delivering protein system (i.e. ferredoxin and
ferredoxin reductase) to effect its monooxygenase activity.
The P450pyr monooxygenase was found to catalyze the
hydroxylations of N-substituted pyrrolidines, piperidines,
azetidines, 2-pyrrolidinones, and 2-piperidinones with high
activity, excellent chemo- and regioselectivity, and good to
In this work, we have evolved both (S)- and (R)-selective
P450pyr by iterative targeted site-saturation mutagenesis using
the prochiral substrate 1. The development of an efficient
high-throughput enantiomeric excess (ee)-screening method
was no trivial task. In fact, most of the directed evolution
studies found in the literature focused on improving the
enantioselectivity in the kinetic resolution of the enzyme,
and very few for asymmetric transformations due to the
complexities in developing an ee-assay. A high-throughput
two-enzyme based colorimetric assay was applied to semi-
quantitatively determine the enantiomeric ratio of the racemic
product. The concept of utilizing two complementary alcohol
4
excellent enantioselectivity. The enzymatic hydroxylation of
N-benzylpyrrolidine 1 to its corresponding (R)- and (S)-N-
benzyl-3-hydroxypyrrolidines 2 is of great interest since these
enantiomers are important pharmaceutical intermediates in
the synthesis of a k-receptor agonist, an antibacterial agent,
3
13
carbapenem antibiotics, and a 5-HT1Da receptor agonist.
dehydrogenases has been described, but the colorimetric
Unfortunately, the enantioselectivity of the wild type (WT)
P450pyr is unsatisfactory and therefore it is desirable to
improve the P450pyr enantioselectivity via protein engineering.
Directed evolution and rational design have been used to
enhance the enantioselectivity of enzymes, such as lipases,
esterases, hydantoinases, nitrilases, epoxide hydroxylases,
phosphotriesterases, aminotransferases, aldolases, cyclohexanone
and cyclopentanone monooxygenases, and monoamine
determinations make it the first example of a high-throughput
enantioselectivity assay developed and used in the directed
evolution of a monooxygenase for asymmetric transformations.
After an extensive literature search, we found two alcohol
1
4
dehydrogenases, i.e. BRD from Micrococcus luteus and
1
5
RDR from Devosia riboflavina, that were highly specific for
(S)- and (R)-2, respectively. The genes encoding these two
alcohol dehydrogenases were cloned and overexpressed in
Escherichia coli BL21(DE3). The resulting purified enzymes
were used in the high-throughput screening method
5
oxidases. Previous directed evolution studies on P450
(
Scheme 1). We assayed these purified enzymes with different
a
Department of Chemical and Biomolecular Engineering,
University of Illinois at Urbana-Champaign,
mixtures of (S)- and (R)-2 and found that the ee correlated to
the color intensity of the assay (See Fig. S1 in Supplementary
Informationz).
600 S. Mathews Avenue, Urbana, IL 61801, USA.
E-mail: zhao5@illinois.edu; Fax: +1-217-333-5052;
Tel: +1-217-333-2631
Department of Chemical and Biomolecular Engineering,
National University of Singapore, 4 Engineering Drive 4, Singapore
117576. E-mail: chelz@nus.edu.sg; Fax: +65-67791936;
Tel: +65-65168416
Whole cell assay was necessary in this case as our P450pyr
system requires the close interaction of the ferredoxin and
ferredoxin reductase to form an electron transport system,
unlike P450BM3 which has its hydroxylase and reductase
domains on a single polypeptide. Also, the whole cell assay
approach removed the need for the cell lysis step, and the
addition of NADH and other auxiliary protein components,
b
w This article is part of the ‘Enzymes and Proteins’ web-theme issue for
ChemComm.
z Electronic supplementary information (ESI) available: Materials
and experimental procedures. See DOI: 10.1039/c0cc00735h
This journal is ꢀc The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5461–5463 | 5461

˚
homology model, 17 residues were identified within 5 A of the
heme-docked substrate (See Fig. S2(a) in Supplementary
Informationz) and were subjected to individual site saturation
mutagenesis. C366 and G256 were not subjected to randomiza-
tion because they are highly conserved residues involved in the
heme coordination and catalytic activity of the P450pyr,
respectively. For each site, 32 distinct variant possibilities exist
(
32 possible codon substitutions). Hence, the screening of 180
transformants per site should be more than enough to provide
comprehensive coverage of all created variants. This protein
engineering approach involves screening very manageable
size of mutant libraries which target the residues within the
substrate binding pocket.
In the first round of saturation mutagenesis, using the WT
P450pyr as a template, we screened 180 mutants per site,
which translates into a total of 3060 mutants (17 sites)
screened. Two interesting mutants with preference for
Scheme 1 A high-throughput two-enzyme based colorimetric ee
assay for asymmetric biohydroxylation of prochiral substrate 1 to
(
S)- and (R)-2 were identified (Table 1). The (S)-selective
(
R)- and (S)-2.
mutant 11BB12 (F403L) displayed an ee of 65% which was
a 1.5 times improvement over that of the WT P450pyr. Mutant
1AF4 (N100S), on the other hand, displayed a complete
reversal of enantioselectivity with an ee of 42% with a
preference for (R)-2. In fact, it was identified twice in the
screening of the same 3060-mutant library. It was intriguing
that only one mutational change at position 100 (replacing
asparagine with serine) could induce a total inversion of the
P450pyr enantioselectivity.
thus simplifying our screening protocol. Previously, the
Pseudomonas putida strain expressing the P450pyr together with
the ferredoxin gene (Fdx) from Sphingomonas sp. HXN200
and the ferredoxin reductase gene (FdR) from Mycobacterium
1
6
sp. HXN1500 was developed but showed low activity.
Hence, we constructed a recombinant E. coli BL21(DE3)
strain containing a dual plasmid system (pRSFDuet P450pyr
and pETDuet Fdx FdR) which showed much higher hydro-
xylation activity and was thus designated as the parent strain.
Our high-throughput screening protocol is described as
follows. Firstly, the mutants were inoculated and expressed
in TB medium in 96-deep well plates, in the presence of
d-aminolevulinic acid (d-ALA) to increase the expression level
of P450pyr. The biohydroxylation assay to convert 1 to (R)-
and (S)-2 was performed with resting cells in 96-deep well
plates with shaking at 1100 rpm to facilitate better mixing and
mass transfer. When the biohydroxylation reaction was carried
out at low shaking speed of 250 to 500 rpm, negligible
substrate conversion was observed. After the biohydroxylation
with whole cells, the deep well plates were centrifuged and
aliquots (80 ml) of the supernatant containing the unreacted 1
and its racemic product 2 were pipetted into two separate
Encouraged by the success in the first round, F403L was
used as a template for a second round of saturation mutagenesis
on the remaining 16 residue sites. Unfortunately, we failed to
discover any (S)-selective mutants with significantly higher ee
than the parent 11BB12. Hence, we decided to identify more
(R)-selective mutants instead. Using N100S as a template, a
third round of saturation mutagenesis was conducted on the
remaining 16 residue sites. In this round, we discovered four
mutants with improved (R)-enantioselectivity ranging from ee
of 60% to 83%. The best mutant, 1AF4A, was sequenced and
it was found that the mutational change at position 186
(isoleucine substitution of threonine) was responsible for the
further increase in the (R)-enantioselectivity of the mutant
P450pyr.
9
6-well microtiter plates. In each plate, purified BRD or RDR
The conversion of 1 by the P450pyr mutants is
and its corresponding cofactor were added. By utilizing the
nitro blue tetrazolium (NBT)-phenazine methosulfate (PMS)
comparatively lower than that of the WT P450pyr (Table 1).
The conversion of the best mutant 1AF4A was more than 50%
lower than the WT P450pyr. The introduction of the mutation
in the active pocket of the P450pyr has probably introduced a
destabilizing effect in the enzyme, thus negatively affecting its
activity, although the enantioselectivity of the enzyme has
been significantly improved. For future work, we could
perhaps take a step back and apply error-prone PCR (epPCR)
to the full-length gene encoding the 1AF4A mutant to improve
its activity to a level equal to or higher than the WT P450pyr.
The roles played by individual mutations towards altering
the enantioselectivity of the P450pyr were not clear. The
inversion of enantioselectivity can arise by the insertion of
the O-atom either from the opposite direction into the C–H
bond at the original 3-position or from the same direction into
the C–H bond at the adjacent position due to the special C(2v)
symmetry nature of the pyrrolidine substrate. From the
17
colorimetric assay, the production of NAD(P)H was monitored
by observing the formation of formazan at OD580, which
corresponded to the activity of the dehydrogenases that in
turn correlated to the concentration of each enantiomer in the
racemic product. Hence, the estimated ee of each mutant could
be compared against the WT P450pyr. Mutants that showed a
greater ee than that of the WT were selected and subjected to a
larger-scale biohydroxylation assay in shaking flasks to verify
the enantioselectivity of the mutants by chiral HPLC analysis.
To identify the residues for directed evolution, we used
molecular modeling to build a structural model of P450pyr
based on the crystal structure of CYP119 from Sulfolobus
1
8
solfactaricus (1F4T) and P450st from Sulfolobus tokodaii
1
9
1UE8). The mutant library was created using iterative
(
2
targeted site-saturation mutagenesis. Based on the P450pyr
0
5
462 | Chem. Commun., 2010, 46, 5461–5463
This journal is ꢀc The Royal Society of Chemistry 2010

Table 1 Hydroxylation of N-benzylpyrrolidine 1 by engineered cytochrome P450pyr variants
a
b
Description
Mutant
Amino acid changes
Conversion (%)
ee (%)
WT P450pyr
Round 1
NA
1AF4
NA
N100S
F403L
55
33
47
43 (S)
42 (R)
65 (S)
1
1BB12
Round 2
Round 3
No significantly improved mutants
1AF4A N100S, T186I
23
16
19
27
83 (R)
63 (R)
71 (R)
60 (R)
1
1
1
AF4B
AF4C
AF4D
N100S, T259S
N100S, L302V
N100S, V404K
a
Conversion determined based on substrate consumption with whole cell assay (10 g cell dry weight (cdw) per liter) using starting substrate
b
concentration of 5 mM. Biohydroxylation time was 4 h. Determined by chiral HPLC analysis using the Chiralcel OB-H (250 mm ꢁ 4.6 mm)
column. All verification experiments were carried out in shaking flasks in triplicate.
homology model, a large cavity allowing substrate access
to the heme site was observed, with hydrophobic residues lining
the entrance to the cavity. An interesting observation was that
the mutation sites 100, 403, and 186 were located near the
entrance of the cavity, suggesting that a benzyl ring–protein
interaction furthest from the heme may induce enantio-
selectivity towards 1 (See Fig. S2(b) in Supplementary
Informationz). The single amino acid substitution of the
Asn100 residue by Ser that caused a complete inversion of
the P450pyr enantioselectivity is of great interest. We reasoned
that changing the bigger hydrophilic Asn residue to the smaller
neutral Ser residue would increase the active-site volume and
alter the conformation of the helix containing the N100S upon
substrate binding, which may be responsible for the reversal of
enantioselectivity. When the neutral Thr residue was also
mutated to a hydrophobic Ile, the cooperative effect of the
two mutations increased the enantioselectivity towards the
4 D. Chang, H.-J. Feiten, K.-H. Engesser, J. B. van Beilen,
B. Witholt and Z. Li, Org. Lett., 2002, 4, 1859–1862; D. Chang,
H.-J. Feiten, B. Witholt and Z. Li, Tetrahedron: Asymmetry, 2002,
1
3
3, 2141–2147; D. Chang, B. Witholt and Z. Li, Org. Lett., 2000, 2,
949–3952; Z. Li, H.-J. Feiten, D. Chang, W. A. Duetz, J. B. van
Beilen and B. Witholt, J. Org. Chem., 2001, 66, 8424–8430.
M. T. Reetz, in Asymmetric Organic Synthesis with Enzymes,
ed. V. Gotor, I. Alfonso and E. Garcia-Urdiales, Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim, 2008, pp. 21–56.
5
6 X. Feng, G. B. Stephen, L. Jaka, I. Andrew, R. Zihe and W.
Luet-Lok, Angew. Chem., Int. Ed., 2005, 44, 4029–4032;
A. Glieder, E. T. Farinas and F. H. Arnold, Nat. Biotechnol.,
2
002, 20, 1135–1139.
7
8
9
R. Fasan, Y. T. Meharenna, C. D. Snow, T. L. Poulos and
F. H. Arnold, J. Mol. Biol., 2008, 383, 1069–1080.
O. Lentz, A. Feenstra, T. Habicher, B. Hauer, R. D. Schmid and
V. B. Urlacher, ChemBioChem, 2006, 7, 345–350.
T. S. Wong, F. H. Arnold and U. Schwaneberg, Biotechnol.
Bioeng., 2004, 85, 351–358.
10 O. Salazar, P. C. Cirino and F. H. Arnold, ChemBioChem, 2003, 4,
91–893.
8
1
1 S. Bartsch, R. Kourist and U. T. Bornscheuer, Angew. Chem., Int.
Ed., 2008, 47, 1508–1511; M. Ivancic, G. Valinger, K. Gruber and
H. Schwab, J. Biotechnol., 2007, 129, 109–122; Y. Koga,
K. Kato, H. Nakano and T. Yamane, J. Mol. Biol., 2003, 331,
(
(
R)-enantiomer by almost 2 fold. Interestingly, in both cases
N100S and T186I), the residue was replaced by a more
hydrophobic residue. The Phe403 in the WT P450pyr is
located in the middle of the cavity to the active site pocket,
thus providing a hydrophobic barrier for the hydrophilic
directing groups in incoming substrates. In the case of the
5
85–592; A. O. Magnusson, M. Takwa, A. Hamberg and K. Hult,
Angew. Chem., Int. Ed., 2005, 44, 4582–4585; O. May,
P. T. Nguyen and F. H. Arnold, Nat. Biotechnol., 2000, 18,
317–320; R. H. van Den Heuvel, M. W. Fraaije, M. Ferrer,
A. Mattevi and W. J. van Berkel, Proc. Natl. Acad. Sci. U. S. A.,
(
S)-enantioselective mutant, 11BB12, the aromatic hydro-
2
000, 97, 9455–9460.
phobic side chain of Phe was replaced by the aliphatic side
chain of Leu, thus slightly reducing the hydrophobicity which
may have directly or indirectly increased the mutant’s enantio-
selectivity towards the (S)-enantiomer due to conformational
changes at the entrance to the enzyme’s active site.
1
2 M. W. Peters, P. Meinhold, A. Glieder and F. H. Arnold, J. Am.
Chem. Soc., 2003, 125, 13442–13450.
3 S. Dey, D. R. Powell, C. Hu and D. B. Berkowitz, Angew. Chem.,
Int. Ed., 2007, 46, 7010–7014; Z. Li, L. Butikofer and B. Witholt,
Angew. Chem., Int. Ed., 2004, 43, 1698–1702.
1
1
1
1
4 N. Kizaki, Y. Yasohara and J. Hasegawa, 2006, US patent
7,033,808 B2.
5 N. Y. Kizaki, N. Nagashima and J. Hasegawa, J. Mol. Catal. B:
Enzym., 2008, 51, 73–80.
6 J. B. van Beilen, E. G. Funhoff, A. van Loon, A. Just, L. Kaysser,
M. Bouza, R. Holtackers, M. Rothlisberger, Z. Li and B. Witholt,
Appl. Environ. Microbiol., 2006, 72, 59–65.
In summary, we have demonstrated the first example of
improving the biohydroxylation enantioselectivity of a class I
P450 monooxygenase by iterative targeted site-saturation
mutagenesis combined with a colorimetric ee high-throughput
screening assay. Interestingly, one mutation at position 100
was all it took to invert the enantioselectivity of the P450pyr.
Financial support by the Science and Engineering Research
Council of A*STAR Singapore through a research grant
1
7 R. Woodyer, W. A. van der Donk and H. Zhao, Biochemistry,
2003, 42, 11604–11614.
8 J. K. Yano, L. S. Koo, D. J. Schuller, H. Li, P. R. Ortiz de
Montellano and T. L. Poulos, J. Biol. Chem., 2000, 275,
1
3
9 Y. Oku, A. Ohtaki, S. Kamitori, N. Nakamura, M. Yohda,
1086–31092.
(
Project No. 0621010024) is gratefully acknowledged.
1
H. Ohno and Y. Kawarabayasi, J. Inorg. Biochem., 2004, 98,
1
Notes and references
194–1199.
1
2
3
J. A. Labinger and J. E. Bercaw, Nature, 2002, 417, 507–514.
Z. Li and D. Chang, Curr. Org. Chem., 2004, 8, 1647–1658.
Z. Li, H.-J. Feiten, J. B. van Beilen, W. Duetz and B. Witholt,
Tetrahedron: Asymmetry, 1999, 10, 1323–1333.
20 M. T. Reetz, D. Kahakeaw and J. Sanchis, Mol. BioSyst., 2009, 5,
115–122; K. Chockalingam, Z. Chen, J. A. Katzenellenbogen and
H. Zhao, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 5691–5696;
N. U. Nair and H. Zhao, ChemBioChem, 2008, 9, 1213–1215.
This journal is ꢀc The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5461–5463 | 5463