Engineering P450 Peroxygenase to Catalyze Highly Enantioselective Epoxidation of cis-β-Methylstyrenes

Article abstract of DOI:10.1002/chem.201601176

P450 119 peroxygenase and its site-directed mutants are discovered to catalyze the enantioselective epoxidation of methyl-substituted styrenes. Two new site-directed P450 119 mutants, namely T213Y and T213M, which were designed to improve the enantioselectivity and activity for the epoxidation of styrene and its methyl substituted derivatives, were studied. The T213M mutant is found to be the first engineered P450 peroxygenase that shows highly enantioselective epoxidation of cis-β-methylstyrenes, with up to 91 % ee. Molecular modeling studies provide insights into the different catalytic activity of the T213M mutant and the T213Y mutant in the epoxidation of cis-β-methylstyrene. The results of the calculations also contribute to a better understanding of the substrate specificity and configuration control for the regio- and stereoselective peroxygenation catalyzed by the T213M mutant.

Full text of DOI:10.1002/chem.201601176

DOI: 10.1002/chem.201601176
Full Paper
&
Enantioselective Epoxidation
Engineering P450 Peroxygenase to Catalyze Highly
Enantioselective Epoxidation of cis-b-Methylstyrenes
Chun Zhang,[a] Ping-Xian Liu,[a] Lu-Yi Huang,[b] Si-Ping Wei,[a] Li Wang,[a] Sheng-Yong Yang,[b]
Xiao-Qi Yu,[c] Lin Pu,*[a, d] and Qin Wang*[a]
Abstract: P450 119 peroxygenase and its site-directed mu-
tants are discovered to catalyze the enantioselective epoxi-
dation of methyl-substituted styrenes. Two new site-directed
P450 119 mutants, namely T213Y and T213M, which were
designed to improve the enantioselectivity and activity for
the epoxidation of styrene and its methyl substituted deriva-
tives, were studied. The T213M mutant is found to be the
first engineered P450 peroxygenase that shows highly enan-
tioselective epoxidation of cis-b-methylstyrenes, with up to
91% ee. Molecular modeling studies provide insights into
the different catalytic activity of the T213M mutant and the
T213Y mutant in the epoxidation of cis-b-methylstyrene. The
results of the calculations also contribute to a better under-
standing of the substrate specificity and configuration con-
trol for the regio- and stereoselective peroxygenation cata-
lyzed by the T213M mutant.
Introduction
the oxygenation of substrates through a so-called shunt path-
way in the absence of NAD(P)H and redox protein partners.[5]
Anaerobic archaeal CYP119 cloned from Sulfolobus acidocal-
darius, as one of the first thermophilic P450 enzymes, was
found to catalyze the peroxygenation of styrene, substituted
styrenes, and laurate in the presence of H2O2 and other perox-
ides,[6–9] which makes it an attractive system for the formation
of the Compound I species by the shunt pathway in the P450s
reaction mechanism.[10–15] The rigid structure of P450 119 per-
oxygenase also provides a unique model for improving the
substrate specificity and stereoselectivity control through pro-
tein engineering. The available crystal structure of P450 119 is
helpful to define the determinants of substrate specificity;
moreover, rational modification of the heme active site can
lead to improved control of stereoselectivity through protein
engineering.[16]
Cytochrome P450 (P450 or CYP) enzymes are versatile heme-
containing proteins that catalyze a diverse range of specific ox-
idations and accept a wide range of given substrates, as exem-
plified by the regio- and stereoselective epoxidation of olefins
and hydroxylation of alkanes.[1–3] These properties make P450s
highly promising catalysts in drug discovery, bioremediation
and fine chemical synthesis.[4] For many oxidative reactions,
P450 enzymes typically catalyze the insertion of one atom of
molecular oxygen (monooxygenation) into exogenous and en-
dogenous substrates through a classical two-electron oxo-
transfer pathway. However, the requirement for expensive co-
factors and the need for complicated electron-transfer systems
are major drawbacks for applications of P450 monooxygenases
in synthetic chemistry. Unique P450 peroxygenases do not
follow the classical O2-based reaction cycle, but employ H2O2
and other peroxides as surrogate oxygen donors to catalyze
Asymmetric epoxidation of alkenes is of particular interest
because the formed chiral epoxides are very useful in the syn-
thesis of fine chemicals and biologically active molecules. For
example, certain epoxidation products of cis-b-methylstyrenes
such as b-asarone and cis-anethole are versatile precursors for
the synthesis of pharmaceutically and biologically active mole-
cules.[17–19] Although extensive experimental studies on the ep-
oxidation of styrene and its derivatives by using the P450
monooxygenases have been reported,[1,3] and computational
data is available on olefin epoxidation by using P450 en-
zymes,[2,20] highly enantioselective epoxidation of some specific
substrates such as cis-b-methylstyrene remains a significant
challenge in the field of P450 catalysis. For example, Oritiz de
Montellano reported that the wild-type P450cam monooxyge-
nase catalyzes the epoxidation of styrene and b-methylstyrene,
but only 78% ee was observed for the asymmetric epoxidation
of cis-b-methylstyrene.[21] After the natural isozyme P450 BM3
monooxygenase from Bacillus megaterium was engineered for
[a] Dr. C. Zhang, P.-X. Liu, Dr. S.-P. Wei, Dr. L. Wang, Prof. L. Pu, Prof. Q. Wang
Department of Medicinal Chemistry, Southwest Medical University
No. 319, Zhongshan Road, Luzhou, Sichuan, 646000 (P. R. China)
E-mail: wq_ring@hotmail.com
[b] L.-Y. Huang, Prof. S.-Y. Yang
State Key Laboratory of Biotherapy, West China Hospital
Sichuan University, No.17 People’s South Road
Chengdu, Sichuan 610041 (P. R. China)
[c] Prof. X.-Q. Yu
College of Chemistry, Sichuan University
No. 29 Wangjiang Road, Chengdu, Sichuan 610064 (P. R. China)
[d] Prof. L. Pu
Department of Chemistry, University of Virginia
Charlottesville, Virginia 22903 (USA)
E-mail: lp6n@virginia.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201601176.
Chem. Eur. J. 2016, 22, 10969 – 10975
10969
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
enantioselective epoxidation of simple terminal alkenes, two
P450 BM3 variants were found to convert a range of terminal
alkenes into epoxides with up to 83% ee.[22] A triple mutant of
P450pys monooxygenase was found to show high enantiose-
lectivity for the epoxidation of para-substituted styrenes.[23] Be-
sides P450 119 peroxygenase, there are only a few cases re-
ported for the asymmetric epoxidation of styrene and methyl-
substituted styrene by using P450 peroxygenase.[5–8] To our
knowledge, no highly enantioselective epoxidation of cis-b-
methylstyrene has been achieved by using P450 peroxygenase.
Thus, non-terminal aromatic alkenes bearing a cis-methyl sub-
stituent are poor substrates for the enantioselective epoxida-
tion catalyzed by P450 enzymes.
Figure 1. a) The relative position of Thr213 and Thr214 residues located in
P450 119 with respect to the heme. b) The relative position of Thr252 resi-
due located in P450cam with respect to the heme.
active pocket of P450 119 to interact with the FeIII(O2H2) inter-
mediate by adjusting the size of the Thr213 site to control the
stereoselectivity. Therefore, we propose that the site-directed
mutation of the Thr213 residue in P450 119 to a large residue
with a polar side chain such as a tyrosine residue or a sulfur-
containing amino acid could improve the steric control around
the active pocket for the asymmetric epoxidation of the cis-b-
methylstyrenes.
Given that the endogenous substrates and the natural redox
partners of P450 119 peroxygenase are not known, the initial
studies for the epoxidation of styrene and the substituted styr-
enes including cis- and trans-b-methylstyrene were carried out
with hydrogen peroxide to assay the catalytic activity of the
wild-type P450 119.[6] Low activity and (S)-enantioselectivity
were observed with a rate of 0.6 minÀ1 and ca. 50% ee for the
styrene epoxidation. Only the product of the cis-isomer was
identified from the epoxidation of cis-b-methylstyrene, indicat-
ing retention of the configuration in the reaction.[6] Whereas
the site-directed mutation of Thr214 to valine in P450 119,
giving the T214V mutant, was found to increase the styrene
epoxidation rate by about threefold, the (S)-enantioselectivity
was not improved.[6] Recently, we reported that introduction of
glycine into the Thr213 residue of the wild-type P450 119,
giving the T213G mutant, enhances both the turnover rate
and (S)-enantioselectivity in the asymmetric epoxidation of sty-
rene with kcat value of 51.2 minÀ1 and ee value of 66% in the
presence of tert-butyl hydroperoxide (TBHP).[8] This indicates
that the Thr213 site can play a major role in improving catalyt-
ic activity and controlling stereoselectivity. However, no turn-
over rate was observed for the double T213G/T214V mutant
under the same conditions.[8] The above results suggest that at
least one polar threonine residue in the adjacent Thr213 and
Thr214 residues in this enzyme is essential for the catalytic ac-
tivity of the alkene epoxidation, probably through hydrogen-
bonding interactions. This is similar to the hydrogen peroxide
shunt pathway of P450cam and its Thr252 mutant in reaction
networks in which the FeIII(O2H2) species is found to be a requi-
site intermediate and Compound I is the sole and final oxidant
generated though an alternative coupling-II pathway.[24]
Figure 1 presents the relative position of the Thr213, Thr214,
and Thr252 residues located in the P450 119 and P450cam
with respect to the heme, where hydrogen-bonding interac-
tions can stabilize the FeIII(O2H2) intermediate with consequent
Compound I formation.
Herein, we report an enantioselective epoxidation of methyl-
substituted styrenes catalyzed by the wild-type P450 119 per-
oxygenase and its mutants. Besides the use of the previously
reported T213G and T214V mutants, we have designed two
new site-directed mutants, namely T213Y and T213M, to im-
prove the enantioselectivity and activity for the asymmetric ep-
oxidation of styrene and its methyl-substituted derivatives. The
P450 119 T213M mutant is discovered to be the first engi-
neered P450 peroxygenase to catalyze the highly enantioselec-
tive epoxidation of cis-b-methylstyrenes, with up to 91% ee.
Molecular docking studies provide insight into the enantiose-
lective control and specific substrate recognition in the asym-
metric epoxidation catalyzed by the T213M mutant.
Results and Discussion
We first investigated the asymmetric epoxidation of the
methyl-substituted styrenes 2a–5a catalyzed by the wild-type
P450 119 peroxygenase, and by the two previously reported
T213G and T214V mutants (Scheme 1) at 358C and pH 7.5 in
the presence of TBHP, according to the optimized reaction
conditions reported earlier;[8] the results are summarized in
Table 1. As shown in Table 1 (entries 1–12), relatively low enan-
On the basis of the second coupling pathway, we speculate
that the P450 119 peroxygenase and the methyl-substituted
styrene substrates could work together to exert great influence
on the stability of the FeIII(O2H2) intermediate. On one hand,
P450 119 can employ the hydrogen-bonding interaction pro-
vided by the Thr213 site to prevent H2O2 release and success-
fully generate the FeIII(O2H2) intermediate. On the other hand,
a methyl-substituted styrene could be oriented properly in the
Scheme 1. Asymmetric epoxidation of styrene and the methyl-substituted
styrenes catalyzed by P450 119 and its mutants.
Chem. Eur. J. 2016, 22, 10969 – 10975
www.chemeurj.org
10970
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ure 1b).[24] In P450 119 peroxygenase, the conserved Thr213
catalytic residue is followed by the adjacent Thr214 residue, as
shown in Figure 1a. In our previous study,[8] the increased turn-
over rate in the asymmetric epoxidation catalyzed by using the
P450 119 T213G mutant might be mainly due to the introduc-
tion of the nonpolar, small glycine residue into the Thr213 site,
which should improve the access of the substrate and TBHP to
the catalytically active pocket. Meanwhile, the polar Thr214
residue in the T213G mutant might become essential for the
catalytic activity of the asymmetric epoxidation in replacement
of the Thr213 residue in the wild-type P450 119. However, the
T213G mutant does not significantly improve the enantioselec-
tivity for the asymmetric epoxidation of the methyl-substituted
styrenes (Table 1, entries 2, 5, 8, and 11). In addition, mutation
of the conserved Thr213 residue to a large hydrophobic resi-
due such as tryptophan or phenylalanine was found to de-
crease the peroxide-dependent epoxidation activity of P450
119 peroxygenase.[6] Therefore, we propose that the site-direct-
ed mutation of the Thr213 residue in P450 119 to a large resi-
due with a polar side chain may increase steric hindrance
around the active pocket as well as maintain high catalytic ac-
tivity in the asymmetric epoxidation of the methyl-substituted
styrenes, as was demonstrated with P450cam and its Thr252
mutant in the second coupling pathway.[24]
Table 1. Epoxidation of methyl-substituted styrenes catalyzed by
P450 119, the T213G mutant and the T214V mutant.
Entry
Substrate
Enzyme
ee [%][a]
Conv. [%][b]
Prod. (Conf.)[c]
1
2
3
4
5
6
7
8
2a
2a
2a
3a
3a
3a
4a
4a
4a
5a
5a
5a
WT
20.9
20.6
19.0
23.6
17.8
33.2
32.4
35.1
30.3
61.4
47.2
46.5
57.8
35.0
34.0
55.7
74.3
50.6
77.1
70.4
58.0
56.2
70.0
47.7
2b (S)
2b (S)
2b (S)
3b (R)
3b (R)
3b (R)
4b (1S,2S)
4b (1S,2S)
4b (1S,2S)
5b (1S,2R)
5b (1S,2R)
5b (1S,2R)
T213G
T214V
WT
T213G
T214V
WT
T213G
T214V
WT
T213G
T214V
9
10
11
12
[a] Determined by chiral-phase HPLC analysis. [b] Determined by GC anal-
ysis. [c] Product configuration given in parentheses was determined by
comparison with reported data.
tioselectivities (18–61% ee) were observed for the epoxidation
of the methyl-substituted styrenes in comparison with the ep-
oxidation of styrene reported earlier.[8] These reactions also
give low substrate conversion of 34–77%. The epoxidations of
4-methyl styrene 2a afforded the corresponding (S)-epoxide
2b with very low enantioselectivities (19–21% ee; entries 1–3),
and the product maintained the same (S)-configuration as that
of styrene. The inverted (R)-configuration of the epoxide prod-
uct was observed for the epoxidation of a-methyl styrene 3a
(18–33% ee; entries 4–6). The reaction of trans-b-methyl sty-
rene 4a gave 30–35% ee for the (1S,2S)-epoxide product 4b
(entries 7–9). The P450 119 peroxygenase and its mutants
show higher enantioselectivity (47–61% ee) for the epoxidation
of cis-b-methyl styrene 5a than for trans-b-methyl styrene with
a preference for formation of the (1S,2R)-enantiomer 5b (en-
tries 10–12). The epoxidation of the two non-terminal aromatic
alkenes 4a and 5a, by using the catalytic system of P450 119
peroxygenase and its mutants, gave low enantioselectivity, but
the corresponding (1S)-configuration in the two diastereomeric
epoxides was still generated. Although the alkene epoxidation
catalyzed by P450s has not been probed as extensively as the
hydroxylation of hydrocarbon chains, a concerted mechanism
of Compound I formation is supported by the finding that the
P450-catalysed olefin epoxidation always proceeds with com-
plete retention of the olefin stereochemistry.[3] As illustrated by
the use of the wild-type P450 119 peroxygenase, it converted
cis-b-methylstyrene exclusively into its cis-epoxide isomer with
no trans-epoxide isomer formation.[6] The preference for the
formation of the epoxide with (1S)-configuration should be de-
termined by the specific orientation of the given substrate in
the heme active site of P450 119 peroxygenase. This is the first
example for the enantioselective epoxidation of non-terminal
aromatic alkenes catalyzed by P450 119 peroxygenase and its
mutants.
Thus, we conducted an investigation to replace the Thr213
residue of the P450 119 peroxygenase with a large polar tyro-
sine residue that should be capable of hydrogen bonding. This
T213Y mutant was constructed by using the Quick-change
Lighting Site-directed Mutagenesis Kit. The site-directed muta-
tion enzyme was obtained after overexpression and purifica-
tion by an optimized expression method. As shown in the UV/
Vis absorption spectra (see the Supporting Information), the
T213Y mutant maintains a normal absorption maximum at
450 nm in the reduced CO-complexed state. The asymmetric
epoxidation of styrene and its methyl-substituted derivatives
catalyzed by the T213Y mutant was examined. As shown in
Table 2, the T213Y mutant retained the P450 119 peroxygenase
catalytic activity in the asymmetric epoxidation of styrene 1a
and its methyl substituted derivatives 2a–5a. The epoxidations
of trans-b-methylstyrene 4a and cis-b-methylstyrene 5a (en-
Table 2. Epoxidation of methyl-substituted styrenes catalyzed by the
T213Y mutant and the product obtained from the epoxidation of the
mutant.
Entry
Substrate
1a
Enzyme
ee [%][a]
Conv. [%][b]
Prod. (Conf.)[c]
1
2
3
4
5
6
7
8
9
10
T213Y
T213M
T213Y
T213M
T213Y
T213M
T213Y
T213M
T213Y
T213M
55.9
60.0
38.5
29.7
50.4
63.2
58.9
62.7
77.0
90.6
42.2
61.3
34.7
47.7
36.6
55.6
54.7
69.9
43.7
74.7
1b (S)
1b (S)
2b (S)
2b (S)
3b (R)
3b (R)
4b (1S,2S)
4b (1S,2S)
5b (1S,2R)
5b (1S,2R)
2a
3a
4a
5a
Recent mechanistic studies on P450s showed that the con-
served Thr252 residue is crucial in hydrogen peroxide shunting
of the cycle of the wild-type P450cam by hydrogen-bonding
interactions to stabilize the FeIII(O2H2) intermediate (Fig-
[a] Determined by chiral-phase HPLC analysis. [b] Determined by GC anal-
ysis. [c] Product configuration given in parentheses was determined by
comparison with reported data.
Chem. Eur. J. 2016, 22, 10969 – 10975
www.chemeurj.org
10971
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tries 7 and 9) gave 59 and 77% ee, respectively, which are
higher than those of the terminal aromatic alkenes including
styrene 1a (56% ee), 4-methylstyrene 2a (39% ee), and a-
methylstyrene 3a (50% ee) (entries 1, 3, and 5). Low conver-
sions of the substrates (35–55%) were observed with the use
of the T213Y mutant. The epoxide product configurations ob-
tained by using the T213Y mutant were the same as those ob-
tained by using P450 119 and its aforementioned mutants. For
example, in the epoxidation of cis-b-methylstyrene by using
T213Y, the (1S,2R)-enantiomer was obtained with 77% ee and
44% conversion (entry 9).
Table 3. Kinetic constants for epoxidation of cis-b-methylstyrene cata-
lyzed by P450 119 T213Y and T213M mutants.
Entry Enzyme kcat [minÀ1
]
Km [mm]
Hydrogen bond length [Š]
1
2
T213Y
T213M
0.04Æ0.002 7.33Æ0.67 2.8
32.93Æ3.9 5.23Æ1.20 1.9
To further understand the different catalytic activity between
the T213M mutant and the T213Y mutant for the epoxidation
of cis-b-methylstyrene, we modeled the hydrogen-bonding in-
teraction of the FeIII(O2H2) intermediate with the Met213 resi-
due and Tyr213 residue in the two mutants, respectively. As
shown in Figure 2, the distal H of the FeIII(O2H2) complexes
forms a hydrogen bond with the sulfoxide oxygen of the me-
thionine-sulfoxide in the Met213 site of the T213M mutant by
the shunt pathway, whereas the T213Y mutant has a hydrogen
bond donated by Tyr213 to the distal O of H2O2 complexes.
The hydrogen-bonding interaction from Met213 in the T213M
mutant might provide better stabilization than that in the
T213Y mutant because of the shorter hydrogen-bond length
obtained by the modeling study, which could contribute to
more facile Compound I formation.
Although cysteine has traditionally been classified as a large
polar amino acid in comparison with serine, it is often consid-
ered to have hydrophobic nature. In P450s, cysteine thiolate as
a proximal ligand to heme iron plays an important role in the
spin state of the resting P450 enzyme. The free cysteine resi-
dues are mostly covalently bonded to other cysteine residues
to form disulfide bonds, which is critical for the folding and
stability of the P450s enzyme. Therefore, mutation of Thr213
to a cysteine residue is unsuited to improve the peroxygenase
activity of P450 119 for the asymmetric epoxidation of the
methyl-substituted styrenes. Methionine is a nonpolar sulfur-
containing amino acid that cannot form disulfide bonds. It was
reported that methionine is apt to undergo oxidation to me-
thionine-sulfoxide by peroxide species, which contributes to
hydrogen-bonding interactions in aqueous reaction net-
works.[25] Thus, we mutated the Thr213 to a methionine residue
in P450 119 peroxygenase. Similar to the construction, expres-
sion, and purification of the T213Y mutant, the T213M mutant
was obtained with a typical P450s character in the UV/Vis ab-
sorption spectrum (see the Supporting Information). The asym-
metric epoxidation catalyzed by the T213M mutant was carried
out. As shown in Table 2 (entries 2, 4, 6, 8, and 10), the epoxi-
dations of styrene 1a, a-methyl styrene 3a, trans-b-methyl sty-
rene 4a, and cis-b-methyl styrene 5a catalyzed by the T213M
mutant gave higher enantioselectivities (60–91% ee) than by
the T213Y mutant. Especially, the T213M mutant showed excel-
lent enantioselectivity for the epoxidation of cis-b-methyl sty-
rene (91% ee) to give the (1S,2R)-epoxide product (entry 10).
To our knowledge, this is the first highly enantioselective epox-
idation of non-terminal aryl alkenes catalyzed by the engi-
neered P450 peroxygenase. The T213M mutant also exhibits
increased substrate activity in comparison with the T213Y
mutant.
Figure 2. a) Hydrogen-bonding interaction of the FeIII(O2H2) intermediate
with the Met213 residue in the T213M mutant. b) Hydrogen-bonding inter-
action of the FeIII(O2H2) intermediate with the Tyr213 residue in the T213Y
mutant.
Chiral b-methylstyrene (propenylbenzene) epoxides are ver-
satile precursors for the synthesis of pharmaceutically and bio-
logically active molecules. For example, b-asarone (1-propenyl-
2,4,5-trimethoxy-benzene), isolated from plants such as Acorus
Linnaeus, exhibits a wide spectrum of biological activities by its
epoxidation to form (Z)-asarone epoxides.[17,18] trans-Anethole
(1-propenyl-4-methoxy-benzene), which is found in the essen-
tial oils of several plants including fennel and Chinese star ani-
seed, has potent pharmacological effect by its metabolism to
the epoxides.[18,19] Therefore, we studied the use of the T213M
mutant and the T213Y mutant of P450 119 peroxygenase to
catalyze the asymmetric epoxidation of the cis-b-methylstyr-
enes with substituents on the benzene ring. Scheme 2 shows
the synthesis of a selection of cis-b-methyl styrene derivatives
by hydrogenation of phenylpropynes in the presence of Lind-
lar catalyst. The results obtained for the T213M and T213Y
mutant-catalyzed asymmetric epoxidation of the cis-b-methyl
styrene derivatives to afford the corresponding epoxides are
summarized in Table 4. The T213M mutant shows much higher
The kinetic constants for the asymmetric epoxidation of cis-
b-methylstyrene catalyzed by both the T213Y and T213M mu-
tants were calculated from the resulting regression of the Mi-
chaelis–Menten equation. As shown in Table 3 (entry 1), the
T213Y mutant converted cis-b-methylstyrene into the corre-
sponding epoxide with a kcat value of 0.04 minÀ1 and a Km
value of 7.33 mm based on the double-reciprocal plot. The ki-
netic constants kcat and Km for the asymmetric epoxidation of
cis-b-methylstyrene catalyzed by the T213M mutant were
32.93 minÀ1 and 5.23 mm, respectively, as shown in Table 3
(entry 2). Thus, the T213M mutant shows higher catalytic activi-
ty for the epoxidation of cis-b-methylstyrene than the T213Y
mutant.
Chem. Eur. J. 2016, 22, 10969 – 10975
www.chemeurj.org
10972
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
substrate, which prevents it docking into the active pocket of
the enzyme.
To ascertain the regio- and stereochemistry changes in the
peroxygenation of P450 119, we also modeled the docking of
the prochiral styrene 1a and cis-b-methylstyrene 5a into the
catalytically active pocket of the T213M mutant. As shown in
Figure 3, the average distances of the olefin double bond of
styrene in the re- and si-face from Compound I are 2.85 and
3.50 Š, respectively (Figure 3a and b), whereas those of cis-b-
Scheme 2. Synthesis and asymmetric epoxidation of substrates 6a–13a.
Table 4. Enantioselective epoxidation of cis-b-methyl styrenes catalyzed
by P450 119 T213M and T213Y mutant.
Entry Sub. T213M
T213Y
Prod. (Conf.)[c]
ee [%][a] Conv. [%][b] ee [%][a] Conv. [%][b]
1
2
3
4
5
6
7
8
9
5a
6a
7a
8a
9a
10a 79.6
11 a 83.1
12a 65.3
13a 58.0
90.6
83.5
82.4
76.9
71.7
74.7
67.3
40.5
57.5
68.6
64.8
48.0
57.7
60.4
77.0
76.9
71.9
69.2
52.6
33.5
55.8
34.7
–
43.7
35.6
20.8
33.2
56.3
35.5
41.8
31.8
–
5b (1S,2R)
6b (1S,2R)
7b (1S,2R)
8b (1S,2R)
9b (1S,2R)
10b (1S,2R)
11 b (1S,2R)
12b (1S,2R)
13b (1S,2R)
Figure 3. The molecular docking of the prochiral styrene and cis-b-methyl-
styrene into the catalytically active pocket of the T213M mutant: a) the re-
face of styrene, b) the si-face of styrene, c) the re-face of cis-b-methylstyrene,
d) the si-face of cis-b-methylstyrene.
[a] Determined by chiral-phase HPLC analysis. [b] Determined by GC anal-
ysis. [c] Product 5b configuration given in parentheses was determined
by comparison with reported data.
methylstyrene are closer to Compound I, with 2.75 and 3.35 Š,
respectively (Figure 3c and d). This suggests that the T213M
mutant should have much higher catalytic activity for the spe-
cific substrates such as cis-b-methylstyrene than styrene. Given
that the average distances of the vinyl groups of styrene and
cis-b-methylstyrene from the re faces to the the iron-oxo unit
are 2.85 and 2.75 Š, respectively (Figure 3a and c), and those
of the si faces are 3.50 and 3.35 Š, respectively (Figure 3b and
d), the preference for the formation of the (1S)-epoxide prod-
ucts could be explained by the shorter distances for the re
face interaction with Compound I. Comparing the average dis-
tance from Compound I to the double bond of the olefinic
moiety with that to the allylic CÀH bond (Figure 3c), we could
also obtain a clue to understand why the epoxidation of cis-b-
methylstyrene predominates over the allylic hydroxylation.
(1S,2R)-enantioselectivity (58–91% ee) and conversion (41–
75%) for the asymmetric epoxidation of most substrates than
the T213Y mutant. Generally, the epoxidations of the para-sub-
stituted cis-b-methylstyrenes 6a–9a (entries 2–5; 72–84% ee)
catalyzed by the T213M mutant give much higher enantiose-
lectivity than those of the meta-substituted cis-b-methylstyr-
enes 10a–13a (entries 6–9; 58–83% ee). In comparison with
the epoxidations of the cis-b-methylstyrenes 9a and 12a, con-
taining a methyl substituent on the benzene, the epoxidations
of the substrates 6a–8a and 10a–11 a, containing functional
groups including methoxyl-, chloro- and bromo-substituents
on the benzene ring, usually gave higher enantioselectivity. For
example, the T213M mutant shows a high enantioselectivity of
84% ee for the epoxidation of para-methoxyl cis-b-methylstyr-
ene 6a (i.e., cis-anethole). However, the epoxidation of meta-
methoxyl cis-b-methylstyrene 13a catalyzed by the T213M
mutant gave only 58% ee and 60% conversion, and neither
enantioselectivity nor conversion was observed for the T213Y
mutant. We also attempted the epoxidation of b-asarone cata-
lyzed by the T213M and T213Y mutants, but no epoxide was
found. This may be attributed to the much greater size of the
Conclusion
We have found that P450 119 peroxygenase and its site-direct-
ed mutants can catalyze the enantioselective epoxidation of
methyl-substituted styrenes. On the basis of the second cou-
pling pathway in the P450s peroxide shunting mechanism, it is
proposed that the introduction of a large amino acid residue
with a polar side chain into the Thr213 site in P450 119 should
both enhance its steric control and improve its catalytic activi-
ty for the asymmetric epoxidation of the methyl-substituted
Chem. Eur. J. 2016, 22, 10969 – 10975
www.chemeurj.org
10973
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tagenic PCR was applied on pET30a-CYP 119 plasmid constructed
previously as the template DNA. For the T213M mutant, the for-
ward and reverse primers were 5’-TTCTCATAGCGGGTAATGAGATGA-
CAACTAACTTAAT ATCAAA-3’, 5’-TTTGATATTAAGTTAGTTGTCATCT-
CATTACCCGCTA TGAGAA-3’. For the T213Y mutant, the primers
were 5’-GTAATGAG TACACAACTAACTTAATATCAAACTCTGT-3’, 5’-GA-
TATTAAGTTAGT TGTGTACTCATTACCCGCTATGAG-3’ (bold italic face
indicates the positions of the mutations). All enzymes were overex-
pressed in E.coli BL21 (DE3) plysS cells, induced by isopropyl-b-d-
thiogalactopyranoside as described in a previous study.[8] The
purity of the CYP119 mutants was determined by SDS-polyacryl-
amide gel electrophoresis, with a single 45 kDa band eluted from
the Ni-NTA column. Protein concentration was calculated by using
a molar extinction coefficient of e415 =104 mmÀ1. The reduced-CO
complex formation of CYP 119 mutants was monitored by UV/Vis
spectroscopy.
styrenes. We have thus designed two new site-directed mu-
tants, namely T213Y and T213M, to catalyze the asymmetric
epoxidation. In general, the T213M mutant shows higher enan-
tioselectivity and catalytic activity for the epoxidation of cis-b-
methyl styrenes than that of the T213Y mutant. Especially, the
T213M mutant is discovered to be the first engineered P450
peroxygenase that shows highly enantioselective epoxidation
of cis-b-methylstyrenes. Our modeling studies reveal a different
hydrogen-bonding interaction of the FeIII(O2H2) intermediate
with the Met213 residue and Tyr213 residue of the mutants,
which provides insights into the origin of the different catalytic
activity of the T213M mutant and the T213Y mutant in the ep-
oxidation of cis-b-methyl styrene. The molecular docking of
the prochiral substrates into the catalytically active pocket also
allows a better understanding of the configuration control and
substrate specificity for the regio- and stereoselective peroxy-
genation catalyzed by the T213M mutant. Currently, we are
modifying the structure of P450 119 peroxygenase further and
exploring their application for the asymmetric epoxidation of
special substrates such as b-asarone.
Synthesis of the standard samples of the epoxide products:
meta-Chloroperbenzoic acid (m-CPBA, 12 mmol) was added to
a stirred solution of each substrate (10 mmol) except cis-4-meth-
oxy-b-methyl styrene (see below) and Na2CO3 (0.8 g) in CH2Cl2
(10 mL) on ice. After the mixture was stirred at RT for 2 h, sodium
hydroxide was added to neutralize the unreacted m-CPBA. The or-
ganic phase was separated and washed with saturated NaHCO3
(3”10 mL), washed with saturated NaCl solution (3”10 mL), and
dried over Na2SO4 overnight. After filtration, the solvent was re-
moved by rotary evaporation. The epoxide products were purified
by flash chromatography on a silica gel column and characterized
by 1H and 13C NMR analyses. These compounds were used as
standards in the HPLC and GC-MS analyses for the identification of
the enzymatic reaction products.
Experimental Section
Materials: Escherichia coli strain BL21 (DE3) plysS and pET30a
vector were obtained from Novagen (La Jolla, CA). All restriction
enzymes and ExTaq Polymerase were purchased from TaKaRa Bio-
technology (Liaoning China), Water was generated by using a Milli-
Q-Gradient purification system (Millipore). Medium components
tryptone and yeast extract were purchased from Oxoid. Isopropyl-
b-d-1-thiogalactopyranoside (ITPG, >99%), ampicillin sodium salt
(>99%), kanamycin sulfate (>99%), styrene (1a), 4-methyl-phenyl-
ene (4-methyl styrene, 2a), 2-phenyl-1-propene (a-methyl-styrene,
3a), trans-b-methylstyrene (4a), cis-b-methyl styrene (5a), and
meta-chloroperbenzoic acid were purchased from Sigma–Aldrich;
mesotetraphenylporphinato-manganese (III) chloride (Mn-TPP) and
benzyldimethylhexadecylammonium chloride were purchased from
Adamas-beta. Other chemicals were purchased from J&K Scientific,
China.
Synthesis of the standard sample of cis-4-methoxy-b-methylstyr-
ene epoxide (6b): The reaction was performed in a CH2Cl2 (2.5 mL)
solution containing meso-tetra-phenylporphinatomanganese (III)
chloride (Mn-TPP) (6 mmol), 4-methylpyridine (0.2 mmol), benzyldi-
methylhexadecyl-ammonium chloride (10 mmol), and cis-4-meth-
oxy-b-methylstyrene (6a; 1 mmol). NaOCl (6% aqueous solution, 4
mol per mol of alkene) was then added and the reaction was initi-
ated by vigorous stirring at RT for 1 h. Samples were then extract-
ed with CH2Cl2 (2”2.5 mL) and the combined CH2Cl2 layers were
separated and purified by column chromatograph on silica gel (pe-
troleum ether/ethyl acetate, 30:1). The epoxide products were
1
characterized by H and 13C NMR analyses.
Synthesis of substrates cis-b-methyl styrenes 6a–13a: To
a flame-dried round-bottom flask under argon was added phenyl-
acetylene (5 mL, 46 mmol, 1 equiv.) or its derivatives followed by
dioxane (300 mL, 0.15m). The flask was placed in an ice water/salt
bath and NaH (276 mmol, 6 equiv.) and iodomethane (5.0 equiv.)
were added slowly. The reaction was stirred at RT for 320 min and
then stirred at 608C for several hours until the reaction was com-
plete as indicated by TLC analysis. The reaction mixture was cooled
to RT, filtered, and the residue was washed with n-hexane. The fil-
trate was washed with dilute hydrochloric acid aqueous solution
and water. The organics were dried over MgSO4 and the solvent
was removed under pressure. The residue was dissolved in n-
hexane, and Lindlar catalyst (0.1 equiv.) was added. The mixture
was stirred at RT under 1 atm hydrogen gas, until the 1-arylpro-
pyne disappeared as shown by TLC analysis. The reaction mixture
was filtered, and the residue was washed with n-hexane. The sol-
vent was removed under pressure and the residue was distilled
under pressure to give cis-b-methyl styrenes 6a–13a in 41–72%
overall yield.
Asymmetric epoxidation of substrates catalyzed by CYP 119 and
its mutants: Asymmetric epoxidation of substrates by CYP119 and
its mutants was carried out by incubating a reaction mixture
(200 mL) containing potassium phosphate buffer (pH 7.5) (50 mm),
CYP 119 enzyme (12.5 mm), substrate (4 mm), and TBHP (4 mm) at
358C for 10 min. The reaction mixture was then extracted with
hexane (2”200 mL) and the enantiomeric excess of the product
was analyzed by using chiral HPLC on a Waters1525 Breezeꢁ with
a UV detector at 220 nm and a chiral column (250”4.6 mm, 5 mm)
at 258C. The same reaction was extended to 1 h and the reaction
mixture was then extracted with dichloromethane (2”200 mL) for
the determination of conversion. The CH2Cl2 layer was submitted
to GC-MS analysis with a CP-chirasl-Dex CB column (25 m ”
0.25 mm inner diameter), Thermo Scientific ITQ900. All the experi-
ments were carried out in triplicate at least.
Activity determination: The activity of the CYP 119 T213M mutant
was determined with cis-b-methylstyrene as substrate. The reaction
mixture consisted of 12.5 mm CYP119 mutant, variable concentra-
tions of the substrate (cis-b-methylstyrene was added as a 0.25m
solution in acetonitrile) and TBHP (the same concentration as cis-b-
methylstyrene). Phosphate buffer (pH 7.5) was added to the final
Construction, expression, and purification of CYP119 mutants:
Mutation of CYP119 was carried out by using the quickchange
Lighting Site-directed Mutagenesis Kit (Agilent Technologies). Mu-
Chem. Eur. J. 2016, 22, 10969 – 10975
www.chemeurj.org
10974
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] J. T. Groves in Cytochrome P450 Structure, Mechanism, and Biochemistry
3rd ed. (Eds.: P. R. Ortiz de Montellano), Kluwer: New York, 2005, pp. 1–
44.
[2] S. Shaik, S. P. De Visser in Cytochrome P450 Structure, Mechanism, and
Biochemistry 3rd ed. (Eds.: P. R. Ortiz de Montellano), Kluwer: New York,
2005, pp. 45–86.
[3] P. R. Ortiz de Montellano, J. J. De Voss in Cytochrome P450 Structure,
Mechanism, and Biochemistry 3rd ed. (Eds.: P. R. Ortiz de Montellano),
Kluwer: New York, 2005, pp. 183–245.
[4] V. B. Urlacher, S. Eiben, Trends Biotechnol. 2006, 24, 324–330.
[5] a) E. G. Hrycay, S. M. Bandiera, Arch. Biochem. Biophys. 2012, 522, 71–
89; b) A. S. Faponle, M. G. Quesne, S. P. De Visser, Chem. Eur. J. 2016, 22,
5478–5483.
volume at 80 mL. The reaction mixtures were incubated at 358C for
30 s in a closed glass vials and stopped by addition of acetonitrile
(720 mL). The activity of the CYP119 T213Y mutant was determined
as described for the T213M mutant, except that the reaction time
was increased to 30 min. The products were analyzed by HPLC
(Agilent 1260 Infinity) connected to a UV-detector set to 220 nm
on a Waters Symmetry C18 column (5 mm ” 150 mm) with ddH2O
(45%)/acetonitrile (55%) as the mobile phase at a flow rate of
1 mLminÀ1. The purified cis-b-methylstyrene epoxide was used as
the authentic standard to identify the retention time and to pre-
pare the calibration curve. The retention times found under these
conditions were ca. 11.22 min for cis-b-methylstyrene and ca.
4.55 min for cis-b-methylstyrene epoxide. The initial rates for the
formation of cis-b-methylstyrene oxide were estimated at various
concentrations of the substrate and analyzed by using the double-
reciprocal plot method to evaluate the kinetic parameters Km and
kcat of the reaction.
[6] L. S. Koo, R. A. Tschirret-Guth, W. E. Straub, P. Monne-Loccoz, T. M.
Loehr, P. R. Ortiz de Montellano, J. Biol. Chem. 2000, 275, 14112–14123.
[7] K. S. Rabe, K. Kiko, C. M. Niemeyer, ChemBioChem 2008, 9, 420–425.
[8] C. Zhang, J. Li, B. Yang, F. He, S.-Y. Yang, X.-Q. Yu, Q. Wang, RSC Adv.
2014, 4, 27526–27531.
[9] L. S. Koo, C. E. Immoos, M. S. Cohen, P. J. Farmer, P. R. Ortiz de Montella-
no, J. Am. Chem. Soc. 2002, 124, 5684–5691.
Molecular modeling and docking study: The crystal structure of
the wild-type P450 119 (PDB code: 1F4T) and P450cam (PDB code:
1DZ9) with a resolution of 1.93 Š was downloaded from the pro-
tein data bank database (http://www.pdb.org/). The structure of
CYP119 was then taken as a template for building the model mu-
tants. We used the MODELER program in Discovery Studio 3.1 to
mutate the specific residues and optimized the conformation of
both the mutated residues and any surrounding residues that lie
within 5 Š. The best mutant model structures with minimum DOPE
score were then obtained. Hydrogen peroxide, heme, and oxygen
atom were maintained in these models. Subsequently, a variety of
styrene derivatives and styrene were prepared and docked into the
mutant and the wild-type CYP119 by using GOLD 5.2. Binding
modes between the re face as well as the si face of the substrates
and the CYP119 mutants were studied by analyzing poses and the
average distances from the vinyl groups to the oxygen ion species.
The figures were generated by using Pymol.
[10] D. G. Kellner, S.-C. Hung, K. E. Weiss, S. G. Sligar, J. Biol. Chem. 2002, 277,
9641–9644.
[11] X. Sheng, J. H. Horner, M. Newcomb, J. Am. Chem. Soc. 2008, 130,
13310–13320.
[12] Q. Wang, X. Sheng, J. H. Horner, M. Newcomb, J. Am. Chem. Soc. 2009,
131, 10629–10636.
[13] X. Yuan, Q. Wang, J. H. Horner, X. Sheng, M. Newcomb, Biochemistry
2009, 48, 9140–9146.
[14] Y. Jiang, S. Sivaramakrishnan, T. Hayashi, S. Cohen, P. Monne-Loccoz, S.
Shaik, P. R. Ortiz de Montellano, Angew. Chem. Int. Ed. 2009, 48, 7193–
7195; Angew. Chem. 2009, 121, 7329–7331.
[15] J. Rittle, M. T. Green, Science 2010, 330, 933–937.
[16] J. K. Yano, L. S. Koo, D. J. Schuller, H. Li, P. R. Ortiz de Montellano, T. L.
Poulos, J. Biol. Chem. 2000, 275, 31086–31092.
[17] X.-L. Feng, Y. Yu, D.-P. Qin, Gao, X.-S. Yao, RSC Adv. 2015, 5, 5173–5182.
[18] A. T. Cartus, S. Stegmuoller, N. Simson, A. Wahl, S. Neef, H. Kelm, D.
Schrenk, Chem. Res. Toxicol. 2015, 28, 1760–1773.
[19] S. G. Kim, A. Liem, B. C. Stewart, J. A. Miller, Carcinogenesis 1999, 20,
1303–1307.
[20] a) S. P. De Visser, F. Ogliaro, P. K. Sharma, S. Shaik, J. Am. Chem. Soc.
2002, 124, 11809–11826; b) S. P. De Visser, F. Ogliaro, N. Harris, S. Shaik,
J. Am. Chem. Soc. 2001, 123, 3037–3047; c) S. P. De Visser, F. Ogliaro, S.
Shaik, Angew. Chem. Int. Ed. 2001, 40, 2871–2874; Angew. Chem. 2001,
113, 2955–2958.
[21] J. A. Fruetel, J. R. Collins, D. L. Camper, G. H. Loew, P. R. Ortiz de Montel-
lano, J. Am. Chem. Soc. 1992, 114, 6987–6993.
[22] T. Kubo, M. W. Peters, P. Meinhold, F. H. Arnold, Chem. Eur. J. 2006, 12,
1216–1220.
[23] A. Li, J. Liu, S. Q. Pham, Z. Li, Chem. Commun. 2013, 49, 11572–11574.
[24] a) B. Wang, C. Li, K. D. Dubey, S. Shaik, J. Am. Chem. Soc. 2015, 137,
7379–7390; b) M. Altarsha, T. Benighaus, D. Kumar, W. Thiel, J. Am.
Chem. Soc. 2009, 131, 4755–4763.
[25] J. W. Chu, B. R. Brooks, B. L. Trout, J. Am. Chem. Soc. 2004, 126, 16601–
16607.
Acknowledgements
This project was supported by the National Natural Science
Foundation of China (No. 21072088, No. 81001700), the Pro-
gram for New Century Excellent Talents in University (NCET-10-
0945), the Key Project (211160) and the Project Sponsored by
the Scientific Research Foundation for the Returned Overseas
Chinese Scholars (2012-1707) from the Ministry of Education of
China, the Scientific Fund of Sichuan Province (2014JQ0052,
14JC0139, 14JC0137), the Program of the Education Depart-
ment of Sichuan Province (14TD0017), and the Science and
Technology Project of Luzhou City (2013LZLY-K71, 2013LZLY-
K58).
Received: March 11, 2016
Revised: April 25, 2016
Keywords: cytochrome P450 · enantioselectivity · enzyme
catalysis · epoxidation · protein engineering
Published online on June 30, 2016
Chem. Eur. J. 2016, 22, 10969 – 10975
www.chemeurj.org
10975
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim