Evolving P450pyr hydroxylase for highly enantioselective hydroxylation at non-activated carbon atom

Article abstract of DOI:10.1039/c2cc30779k

Directed evolution of a monooxygenase to achieve very high enantioselectivity for hydroxylation at non-activated carbon atoms is demonstrated for the first time, where a triple mutant of P450pyr hydroxylase is obtained via determination of enzyme structure, iterative saturation mutagenesis, and high-throughput screening with a MS-based ee assay to increase the product ee from 53% to 98% for the hydroxylation of N-benzyl pyrrolidine to (S)-N-benzyl 3-hydroxypyrrolidine. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2cc30779k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 4618–4620
www.rsc.org/chemcomm
COMMUNICATION
Evolving P450pyr hydroxylase for highly enantioselective hydroxylation
at non-activated carbon atomw
Son Quang Pham,a Guillaume Pompidor,b Ji Liu,a Xiao-Dan Lic and Zhi Li*a
Received 3rd February 2012, Accepted 2nd March 2012
DOI: 10.1039/c2cc30779k
Directed evolution of a monooxygenase to achieve very high
enantioselectivity for hydroxylation at non-activated carbon
atoms is demonstrated for the first time, where a triple mutant
of P450pyr hydroxylase is obtained via determination of enzyme
structure, iterative saturation mutagenesis, and high-throughput
screening with a MS-based ee assay to increase the product ee
from 53% to 98% for the hydroxylation of N-benzyl pyrrolidine
to (S)-N-benzyl 3-hydroxypyrrolidine.
Regio- and stereoselective hydroxylation at non-activated carbon
atom is a useful reaction for the functionalization of alkanes and
Fig. 1 (a) P450pyr-catalyzed enantioselective hydroxylation of N-benzyl
the preparation of valuable chiral alcohols. However, it remains a
pyrrolidine; (b) structure of P450pyr hydroxylase. Yellow: heme, green: the
significant challenge in classic chemistry. Monooxygenase could
catalyze this type of reaction with molecular oxygen as green
oxidant. Many monooxygenase-containing microorganisms were
reported for the hydroxylations,1 and some monooxygenases
(MMO,2 AlkB,3 P4504) were well characterized and engineered
for special hydroxylations. Nevertheless, the enantioselectivities in
many cases need further improvement for practical application.
Recently, directed evolution has become a useful tool for improving
enzyme enantioselectivity.5,6 Pioneering work in biohydroxylation
was performed on the directed evolution of P450BM3 mono-
oxygenase,7 a class II P450 with a single protein combining all
three catalytic functions. Special difficulties in evolving a
monooxygenase for highly enantioselective biohydroxylation
are the involvement of enantiotopic discrimination and the
lack of an applicable high-throughput ee assay. Thus, it is still
a significant challenge to evolve a monooxygenase giving a
hydroxylation product with Z 98% ee, the required ee value
for chiral pharmaceutical intermediates.
3 mutation sites leading to the highest enantioselectivity.
as an active and selective catalyst for the regio- and stereo-
selective hydroxylation at non-activated carbon atom with a
broad substrate range and broad applications.9 However, in
the case of hydroxylation of N-benzyl pyrrolidine9a,b to
prepare (S)-N-benzyl 3-hydroxy-pyrrolidine as useful pharma-
ceutical intermediate,10 the product ee was only 53%. To
improve the enantioselectivity, we previously performed the
directed evolution of P450pyr hydroxylase by identifying key
amino acid residues for evolution based on a homology
structure model and using a colorimetric ee screening assay.11
P450pyr mutant N100S/T186I with reversed and improved
enantioselectivity was obtained to give (R)-3-hydroxypyrrolidine in
83% ee.11 Here we report our recent success in evolving P450pyr
for the highly enantioselective hydroxylation of N-benzyl
pyrrolidine to reach 98% ee of the (S)-product by structure-
based selection of key amino acid residues, iterative saturation
mutagenesis, and high-throughput screening with an accurate
and sensitive MS-based ee assay.
Recently, we discovered P450pyr hydroxylase8 from
Sphingomonas sp. HXN-200, a class I P450 requiring ferredoxin
(Fdx) and ferredoxin reductase (FdR) for electron transfer (Fig. 1a),
To determine the structure of P450pyr hydroxylase, an
Escherichia coli Top10 pTr99A-his-P450pyr was engineered to
express P450pyr with a poly-histidine tag at the N-terminal. The
his-tagged P450pyr was purified by fast protein liquid chromato-
graphy (FPLC) using a Ni-NTA column and a size-exclusion
column, with 495% purity and a specific activity of 1595 U gÀ1
protein for the hydroxylation of N-benzyloxycarbonyl pyrrolidine
in the presence of NADH, together with the purified his-tagged
ferredoxin and his-tagged ferredoxin reductase.
a 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
b Swiss Light Source, Paul Scherrer Institut, CH-5232 Villigen PSI,
Switzerland
c Biomolecular Research, Department of Biology and Chemistry, Paul
Scherrer Institut, CH-5232 Villigen PSI, Switzerland
w Electronic supplementary information (ESI) available: Cloning,
purification, crystallization, and X-ray structure of his-tagged
P450pyr; directed evolution protocols; high throughput ee assay;
HPLC chromatograms; molecular modeling; kinetic study of
P450pyr and the triple mutant. See DOI: 10.1039/c2cc30779k
His-tagged P450pyr was crystallized, and its structure was
determined by X-ray diffraction (PBD ID 3RWL). As shown
in Fig. 1b, the structure exhibits an overall fold typical of
c
4618 Chem. Commun., 2012, 48, 4618–4620
This journal is The Royal Society of Chemistry 2012

View Article Online
P450 hydroxylases. The heme is located at the bottom of a
gorge leading to the active site. The large size of the pocket
explains the broad substrate range for the biohydroxylation.
The final model is lacking residues 88–96 for which no electron
density is visible. These residues are located in a loop on the
top of the gorge leading to the active site. The same disorder is
observed in the equivalent loop of the structure of the cytochrome
P450cam in the ‘‘open state’’ corresponding to the substrate-free
state.12 20 amino acids located either in the vicinity of the active
site or in the gorge leading to the heme were selected for mutation:
A77, I82, I83, L98, P99, N100, I102, A103, S182, D183, T185,
T186, L251, V254, G255, D258, T259, L302, M305, and F403.
A simplified version of Iterative Saturation Mutagenesis5a
was applied to the evolution. Each of the selected amino acid
sites was subjected to saturation mutagenesis; all mutants were
screened for better enantioselectivity; and the best mutant was
used as the template for the next round of evolution in which
each of other selected residues was subjected to saturation
mutagenesis and screening. For saturation mutagenesis, random
mutations at the desired position of the P450pyr gene were
generated by PCR using NNK degenerated codon. PCR products
were then transformed in E. coli BL21(DE3) containing ferredoxin
and ferredoxin reductase genes to create P450pyr mutants. To
cover all possible 19 mutants at the selected site, 188 colonies were
picked up and grown in two 96-deep well plates. For high-
throughput screening of enantioselective mutants, our previously
developed assay based on mass spectrophotometer13 was
applied (Fig. 2a). The cells in each well were harvested and then
equally resuspended in potassium phosphate buffer in two wells.
Table 1 Directed evolution of P450pyr hydroxylase for hydroxylation
of N-benzyl pyrrolidine 1 to (S)-N-benzyl 3-hydroxypyrrolidine 2
No. of
sites
Round saturated screened
No. of
mutants
Activityc Rel.
Positive eea eeb [U gÀ1 activityd
mutant [%] [%] cdw]
[%]
100
Wild Nil
type
Nil
Nil
49 53 6.4
72 78 5.3
M305A 76 64 0.6
1
9
1692
3572
I83H
83
9
2
19
I83H/
M305Q
I83H/
A77S
I83H/
M305Q/
A77S
470 94 4.1
470 85 6.2
470 98 5.2
64
97
81
3
5
940
a
Determined by using LC-MS assay. b Determined by chiral HPLC
analysis with a concentrated product sample from a 10 mL-scale biotrans-
formation. c Determined from 30 min biotransformation of 10 mM
substrate 1 with 1 g cdw LÀ1 of cells of E. coli (P450pyr or mutant) at
30 1C and 300 rpm. d Related to the specific activity of E. coli (P450pyr).
(R)- and (S)-deuterated N-benzyl pyrrolidines 1 (1 mM) were
added, respectively, and the aqueous biotransformation product
after 1 h was analyzed by LC-MS (1 sample per min) to determine
the product ee. An example is shown in Fig. 2b(i). For promising
mutants selected by the ee assay, further tests were performed by
biotransformation of N-benzyl pyrrolidine 1 with the resting cells
of the mutant on a 10 mL scale. The product after 1 h reaction was
analyzed by chiral HPLC to determine the product ee [Fig. 2b(ii)]
and by reversed-phase HPLC to obtain accurate conversion and
regioselectivity [Fig. 2b(iii)]. As demonstrated in Fig. 2b(i) and (ii),
the ee value obtained from LC-MS assay was nearly the same as
the value determined by chiral HPLC, thus proving the accuracy
of the LC-MS ee assay.
The results of directed evolution are summarized in Table 1.
In the first round of mutagenesis, 9 of 20 selected amino acid
positions were subjected to saturation since other positions
were screened in a previous study.11 Mutant I83H was found
to increase the product ee to 78%(S) with 83% activity related
to the activity of the wild-type P450pyr. This mutant was used
as a template for the second round evolution in which all other
19 selected amino acid residues were subjected to saturation
and screening, respectively. In this and later rounds, only
mutants with 470% ee(S) by LC-MS assay were selected
for further investigations. Mutant I83H/M305Q was found to
give a product ee of 94%(S) with 64% relative activity. Thus, a
big improvement of enantioselectivity from product ee of
53%(S) to 94%(S) was achieved after only 2 rounds of
mutagenesis and screening of 5264 colonies. The third
round evolution was conducted using mutant I83H/M305Q
as template with saturation at only 5 amino acid residues A77,
P99, N100, S182, and L302, respectively, since mutations at
these positions gave good results in previous rounds (data not
shown). Triple mutant I83H/M305Q/A77S was found to give
the product in 98% ee(S) (Fig. 2c), with a specific activity of
5.2 U gÀ1 cdw (81% activity of P450pyr). No by-products
were formed, suggesting also clean reaction and excellent
regioselectivity for the triple mutant.
Fig. 2 (a) High-throughput ee assay based on the use of a deuterated
substrate and MS detection;13 (b) example of screening an enantio-
selective mutant (I83H). (i) MS spectra for fast ee determination,
(ii) chiral HPLC chromatogram for accurate ee analysis, (iii) reversed-
phase HPLC chromatogram for conversion analysis; (c) chiral HPLC
chromatogram of the purified product 2 from biohydroxylation of
N-benzyl pyrrolidine 1 with best mutant I83H/M305Q/A77S.
To investigate the kinetics, P450pyr triple mutant, ferredoxin,
and ferredoxin reductase, all with polyhistidine at the N-terminal,
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4618–4620 4619

View Article Online
guide the evolution. It provides with a solid basis for further
engineering of P450pyr monooxygenase for enantioselective
hydroxylations of other types of substrates as well as for other
types of enantioselective oxidations. The high-throughput
MS-based ee assay has been proven to be practical for evolution
and will become a valuable tool for evolving other enantio-
selective hydroxylating enzymes.
This work was supported by Science & Engineering Research
Council of A*STAR, Singapore, through a research grant
(project No. 0621010024).
Fig. 3 Substrate–enzyme binding pose. (a) Wild type P450pyr; (b) triple
mutant I83H/M305Q/A77S. The mutation sites are illustrated in yellow.
The distance between the hydrogen atom of substrate 1 and the heme-
oxygen atom is denoted by a dashed line and represented in angstrom.
Notes and references
1 Z. Li and D. Chang, Curr. Org. Chem., 2004, 8, 1647; Z. Li,
J. B. Beilen, W. A. Duetz, A. Schmid, A. Raadt, H. Griengl and
B. Witholt, Curr. Opin. Chem. Biol., 2002, 6, 136; G. Braunegg,
A. Raadt, S. Feichtenhofer, H. Griengl, I. Kopper, A. Lehmann
and H. Weber, Angew. Chem., Int. Ed., 1999, 38, 2763.
2 M. Merkx, D. A. Kopp, M. H. Sazinsky, J. L. Blazyk, J. Muller
and S. J. Lippard, Angew. Chem., Int. Ed., 2001, 40, 2782.
3 J. B. Van Beilen, J. Kingma and B. Witholt, Enzyme Microb.
Technol., 1994, 16, 904; R. T. Ruettinger, G. R. Griffith and
M. J. Coon, Arch. Biochem. Biophys., 1977, 183, 528.
4 P. R. Ortiz de Montellano, Cytochrome P450: Structure, Mechanism,
and Biochemistry, Springer, Berlin, 3rd edn, 2005; A. W. Munro,
H. M. Girvan and K. J. McLean, Nat. Prod. Rep., 2007, 24, 585;
A. Glieder, E. T. Farinas and F. H. Arnold, Nat. Biotechnol., 2002,
20, 1335.
were engineered, produced and purified. Hydroxylation of N-benzyl
pyrrolidine 1 at different concentrations was performed with
P450pyr mutant or P450pyr, ferredoxin, and ferredoxin reductase
at a molar ratio of 1 : 10 : 1 in the presence of NADH for 5 min to
obtain the kinetic data. The P450pyr mutant gave slightly higher
apparent kcat (26.9 vs. 21.9 minÀ1) and Km (15.6 vs. 13.6 mM) than
P450pyr and nearly the same catalytic efficiency (kcat/Km) as
P450pyr. Thus, the triple mutant is as efficient as P450pyr for the
target hydroxylation. The product ee obtained with the triple
mutant in vitro was also found to be 98%(S).
Docking models of N-benzyl pyrrolidine on the MD simulated
structures of P450pyr and the triple mutant were established. In
P450pyr, the benzene ring of the substrate is stabilized by
hydrophobic I82, I102 and L251, while the pyrrolidine ring has
only weak interaction with L302 and thus certain flexibility
(Fig. S18, ESIw). Through the mutation of I83H and A77S,
hydrogen bonds are formed within a big loop (S75–D105), which
moved the loop closer to the heme with I102 occupying the
original binding position of the benzene ring of the substrate
(Fig. S19, ESIw). In addition, the mutation of M305Q moves
L302 and V404 closer to the heme. As a result, the benzene ring
of the substrate is located within a hydrophobic cleft (V254,
V404) above the heme and the pyrrolidine ring is restrained by
I102 and L302 in the vicinity of the heme in the mutant (Fig. S20,
ESIw). As shown in Fig. 3, the distance between the heme
O-atom and the pro-S or pro-R H-atom at C(3) of the
substrate is 2.8 A or 3.7 A for P450pyr and 2.4 A or 3.6 A
for the triple mutant, respectively. Thus, the preference of
inserting an O-atom into the pro-S C–H bond over the pro-R
C–H bond is much higher in the mutant than in P450pyr. This
explains the significantly improved enantioselectivity for the
hydroxylation via only three mutations.
5 For recent reviews, see: (a) M. T. Reetz, Angew. Chem., Int. Ed.,
2011, 50, 138; (b) N. J. Turner, Nat. Chem. Biol., 2009, 5, 567.
6 S. Kille, F. E. Zilly, J. P. Acevedo and M. T. Reetz, Nat. Chem.,
2011, 3, 738; M. T. Reetz and S. Wu, J. Am. Chem. Soc., 2009,
131, 15424; O. May, P. T. Nguyen and F. H. Arnold, Nat.
Biotechnol., 2000, 18, 317; M. Landwehr, L. Hochrein,
C. R. Otey, A. Kasrayan, J. E. Backvall and F. H. Arnold,
¨
J. Am. Chem. Soc., 2006, 128, 6058; R. J. Fox, S. C. Davis,
E. C. Mundorff, L. M. Newman, V. Gavrilovic, S. K. Ma,
L. M. Chung, C. Ching, S. Tam, S. Muley, J. Grate, J. Gruber,
J. C. Whitman, R. A. Sheldon and G. W. Huisman, Nat. Biotechnol.,
2007, 25, 338; F. Escalettes and N. J. Turner, ChemBioChem, 2008,
9, 857; R. Carr, M. Alexeeva, A. Enright, T. S. C. Eve, M. J. Dawson
and N. J. Turner, Angew. Chem., Int. Ed., 2003, 42, 4807.
7 S. T Jung, R. Lauchli and F. H. Arnold, Curr. Opin. Biotechnol., 2011,
22, 809; M. W. Peters, P. Meinhold, A. Glieder and F. H. Arnold,
J. Am. Chem. Soc., 2003, 125, 13442; D. F. Muenzer, P. Meinhold,
M. W. Peters, S. Feichtenhofer, H. Griengl, F. H. Arnold, A. Glieder
and A. de Raadt, Chem. Commun., 2005, 2597.
8 D. Chang, Diss. ETH No. 15380, 2003, p. 168.
9 (a) Z. Li, H. J. Feiten, J. B. van Beilen, W. Duetz and B. Witholt,
Tetrahedron: Asymmetry, 1999, 10, 1323; (b) Z. Li, H. J. Feiten,
D. Chang, W. A. Duetz, J. B. van Beilen and B. Witholt, J. Org.
Chem., 2001, 66, 8424; (c) D. Chang, B. Witholt and Z. Li, Org.
Lett., 2000, 2, 3949; (d) D. Chang, H. J. Feiten, K. H. Engesser,
J. B. van Beilen, B. Witholt and Z. Li, Org. Lett., 2002, 4, 1859;
(e) D. Chang, H. J. Feiten, B. Witholt and Z. Li, Tetrahedron:
Asymmetry, 2002, 13, 2141; (f) W. Zhang, W. L. Tang, Z. Wang
and Z. Li, Adv. Synth. Catal., 2010, 352, 3380.
10 K. Tamazawa, H. Arima, T. Kojima, Y. Isomura, M. Okada,
S. Fujita, T. Furuya, T. Takenaka, O. Inagaki and M. Terai,
J. Med. Chem., 1986, 29, 2504; A. Graul and J. Castaner, Drugs
Future, 1996, 21, 1105; T. Nagahara, Y. Yokoyama, K. Inamura,
S. Katakura, S. Komoriya, H. Yamaguchi, T. Hara and
M. Iwamoto, J. Med. Chem., 1994, 37, 1200.
A highly enantioselective P450pyr hydroxylase has been
engineered for the hydroxylation of N-benzyl pyrrolidine to give
the (S)-3-hydroxypyrrolidine in 98% ee. This is a breakthrough in
directed evolution of an enantioselective monooxygenase for
asymmetric hydroxylation at non-activated carbon atom: for the
first time, a monooxygenase has been obtained via evolution to
give very high product ee (Z98% ee) and clean biohydroxylation;
both high enantioselectivity and high whole-cell activity have been
achieved with the engineered P450pyr triple mutant. The X-ray
structure of P450pyr has been obtained and successfully used to
11 W. L. Tang, Z. Li and H. Zhao, Chem. Commun., 2010, 46, 5461.
12 Y. T. Lee, R. F. Wilson, I. Rupniewski and D. B. Goodin,
Biochemistry, 2010, 49, 3412.
13 Y. Chen, W. L. Tang, J. Mou and Z. Li, Angew. Chem., Int. Ed.,
2010, 49, 5278.
c
4620 Chem. Commun., 2012, 48, 4618–4620
This journal is The Royal Society of Chemistry 2012