Development of an improved Amplex Red peroxidation activity assay for screening cytochrome P450 variants and identification of a novel mutant of the thermophilic CYP119

Article abstract of DOI:10.1007/s00775-020-01816-w

Abstract: Biocatalysts are increasingly utilized in the synthesis of drugs and agrochemicals as an alternative to chemical catalysis. They are preferred in the synthesis of enantiopure products due to their high regioselectivity and enantioselectivity. Cytochrome P450 (P450) oxygenases are valuable biocatalysts, since they catalyze the oxidation of carbon–hydrogen bonds with high efficiency and selectivity. However, practical use of P450s is limited due to their need for expensive cofactors and electron transport partners. P450s can employ hydrogen peroxide (H₂O₂) as an oxygen and electron donor, but the reaction with H₂O₂ is inefficient. The development of P450s that can use H₂O₂ will expand their applications. Here, an assay that utilizes Amplex Red peroxidation, to rapidly screen H₂O₂-dependent activity of P450 mutants in cell lysate was developed. This assay was employed to identify mutants of CYP119, a thermophilic P450 from Sulfolobus acidocaldarius, with increased peroxidation activity. A mutant library of CYP119 containing substitutions in the heme active site was constructed via combinatorial active-site saturation test and screened for improved activity. Screening of 158 colonies led to five mutants with higher activity. Among improved variants, T213R/T214I was characterized. T213R/T214I exhibited fivefold higher k_cat for Amplex Red peroxidation and twofold higher k_cat for styrene epoxidation. T213R/T214I showed higher stability towards heme degradation by H₂O₂. While the K_m for H₂O₂ and styrene were not altered by the mutation, a fourfold decrease in the affinity for another substrate, lauric acid, was observed. In conclusion, Amplex Red peroxidation screening of CYP119 mutants yielded enzymes with increased peroxide-dependent activity. Graphic abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s00775-020-01816-w

JBIC Journal of Biological Inorganic Chemistry
https://doi.org/10.1007/s00775-020-01816-w
ORIGINAL PAPER
Development of an improved Amplex Red peroxidation activity assay
for screening cytochrome P450 variants and identifcation of a novel
mutant of the thermophilic CYP119
M. Semih Başlar1 · Tuğçe Sakallı2 · Gülce Güralp1 · Ekin Kestevur Doğru1 · Emre Haklı2 · Nur Basak Surmeli1
Received: 27 April 2020 / Accepted: 30 August 2020
© Society for Biological Inorganic Chemistry (SBIC) 2020
Abstract
Biocatalysts are increasingly utilized in the synthesis of drugs and agrochemicals as an alternative to chemical catalysis. They
are preferred in the synthesis of enantiopure products due to their high regioselectivity and enantioselectivity. Cytochrome
P450 (P450) oxygenases are valuable biocatalysts, since they catalyze the oxidation of carbon–hydrogen bonds with high
efciency and selectivity. However, practical use of P450s is limited due to their need for expensive cofactors and electron
transport partners. P450s can employ hydrogen peroxide (H2O2) as an oxygen and electron donor, but the reaction with
H2O2 is inefcient. The development of P450s that can use H2O2 will expand their applications. Here, an assay that utilizes
Amplex Red peroxidation, to rapidly screen H2O2-dependent activity of P450 mutants in cell lysate was developed. This
assay was employed to identify mutants of CYP119, a thermophilic P450 from Sulfolobus acidocaldarius, with increased
peroxidation activity. A mutant library of CYP119 containing substitutions in the heme active site was constructed via com-
binatorial active-site saturation test and screened for improved activity. Screening of 158 colonies led to fve mutants with
higher activity. Among improved variants, T213R/T214I was characterized. T213R/T214I exhibited fvefold higher kcat for
Amplex Red peroxidation and twofold higher kcat for styrene epoxidation. T213R/T214I showed higher stability towards
heme degradation by H2O2. While the Km for H2O2 and styrene were not altered by the mutation, a fourfold decrease in
the afnity for another substrate, lauric acid, was observed. In conclusion, Amplex Red peroxidation screening of CYP119
mutants yielded enzymes with increased peroxide-dependent activity.
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s00775-020-01816-w) contains
supplementary material, which is available to authorized users.
* Nur Basak Surmeli
nursurmeli@iyte.edu.tr
1
Department of Bioengineering, İzmir Institute
of Technology, Gülbahçe, Urla, Izmir, Turkey
2
Program in Biotechnology and Bioengineering, İzmir
Institute of Technology, Gülbahce, Urla, Izmir, Turkey
Vol.:(0123456789)
1 3

JBIC Journal of Biological Inorganic Chemistry
Graphic abstract
Keywords Cytochrome · Biocatalysis · Protein engineering · Enzyme kinetics · Heme
Abbreviations
P450s have attracted attention due to their capability to
catalyze the oxidation of “unactivated” hydrocarbons with
high efciency and selectivity. Indeed, P450 enzymes are
currently used in the synthesis of steroids and statins [2, 3].
P450s most commonly catalyze the insertion of a sin-
gle oxygen atom into an organic substrate using molecu-
lar oxygen (O2) and the hydride (H−) donor nicotinamide
adenine dinucleotide phosphate (NAD(P)H). P450s require
protein partners to deliver the electrons from the expensive
cofactor NAD(P)H. Since the expression and purifcation
of electron transfer proteins’ cost resources and energy,
whole-cell catalysis is generally employed for the synthesis
of pharmaceuticals [4, 5]. This limits the P450 enzymes that
can be utilized, since not all host organisms are amenable
to industrial processes. P450s can also employ hydrogen
peroxide (H2O2) as an oxygen donor and electron acceptor
[6]. This eliminates the need for expensive cofactors and
electron transfer partners. Nevertheless, the reaction with
H2O2 is inefcient, except for a few P450s such as P450BSβ
[7]. General acid/base residues at the heme distal site are
crucial for the generation of compound I, which is respon-
sible for H2O2-dependent oxidation [6]. The development
of P450 enzymes that can utilize H2O2 efciently will lead
to an expansion of the application of P450s as biocatalysts.
To date, protein engineering approaches have mostly
focused on the bacterial CYP102A1 (P450BM3) and
CYP101A1 (P450CAM), because they are highly soluble
and active when expressed in E. coli [8]. Since it is advan-
tageous to obtain P450 enzymes that can utilize H2O2,
laboratory evolution of P450CAM to enhance the rate of
P450
Cytochrome P450
HRP
Horseradish peroxidase
WT
Wild type
IPTG
NAD(P)H
Isopropyl β-d-1-thiogalactopyranoside
Nicotinamide adenine dinucleotide
phosphate
CAST
PMSF
EDTA
DMSO
TBHP
FAD
Combinatorial active-site saturation test
Phenylmethylsulfonyl fuoride
Ethylenediaminetetraacetic acid
Dimethyl sulfoxide
Tert-butyl hydroperoxide
Flavin adenine dinucleotide
Flavin mononucleotide
FMN
RCSB PDB The Research Collaboratory for Structural
Bioinformatics Protein Data Bank
PDB ID
REU
Protein data bank identity
Rosetta energy unit
Introduction
Biocatalysis is the chemical process whereby enzymes cata-
lyze reactions between organic components. Most signifcant
features of enzymes are their high regioselectivity and enan-
tioselectivity. These properties have expanded the utilization
of enzymes in the pharmaceutical and agrochemical indus-
try [1]. Developments in biotechnology have led to engi-
neered biocatalysts capable of performing reactions that are
challenging with chemical catalysts. Amongst biocatalysts,
1 3

JBIC Journal of Biological Inorganic Chemistry
H2O2-driven naphthalene hydroxylation has been attempted
[9]. However, isolation and characterization of this mutant
have revealed that it generates only trace amounts of hydrox-
ylated naphthalene [10]. P450BM3 is expressed with the
electron transfer FAD/FMN reductase domain and the P450
monooxygenase domain on a single polypeptide, which
makes optimization of P450BM3 challenging as mutations
can independently afect coupling efciency and activity.
CYP119 is a P450 from the thermoacidophilic archaeon
Sulfolobus acidocaldarius. Sulfolobus sp. are sulfur utilizing
organisms that have optimum growth conditions of between
78 and 86 °C and pH 2–4 [11, 12]. Since the thermophilic
enzymes have increased stability toward mutations, high
temperature, and extreme pH, they have more potential
applications compared to mesophilic enzymes. CYP119 is
one of the most extensively studied thermostable P450s and
its molecular structure has been determined by X-ray crys-
tallography [13]. The natural substrates and redox partners
of CYP119 are currently unknown. However, CYP119 can
oxidize lauric acid using the auxiliary redox partner proteins
putidaredoxin and putidaredoxin reductase with NADH as
an electron donor [14]. CYP119 can also use H2O2 as an
oxidant via the peroxide shunt pathway with low efciency.
Previous studies have shown that CYP119 can catalyze
epoxidation of styrene, hydroxylation of lauric acid, chemi-
cal dehalogenation, and electrochemical reduction of nitrite,
nitric oxide, and nitrous oxide [15]. Here, CYP119 mutants
were screened for peroxidation activity to develop a biocata-
lyst that can utilize H2O2 more efciently.
this approach pairs of amino acids within a defned proxim-
ity—sufciently close to interact—are randomized pairwise
[16]. The efectiveness of the CAST method depends highly
on the selection of the right amino acids to be randomized.
Previous studies have shown that Thr213 and Thr214 resi-
dues play important roles in the active site, but they are not
critical for thermal stability [11]. Therefore, these residues
were selected for targeted mutation.
Here, we report the generation of a mutant library of
CYP119 containing substitutions at the Thr213 and Thr214
positions as well as the development of an efective screen-
ing method utilizing Amplex® Red as a substrate to meas-
ure improved H2O2-dependent catalytic activity of CYP119
variants. Screening of the mutant library resulted in fve
mutants with improved activity compared to wild-type (WT)
CYP119. Among fve improved mutants, further characteri-
zation of the T213R/T214I was performed. T213R/T214I
showed improved activity toward Amplex® Red peroxida-
tion and styrene epoxidation and exhibited improved stabil-
ity against H2O2 degradation.
Materials and methods
Expression of WT CYP119 and T213R/T214I mutant
Expression vector containing WT CYP119
(pET11a + CYP119) was a gift from Teruyuki Nagamune
(Addgene plasmid # 66131) [22]. For expression of WT
CYP119, pET11a+CYP119 plasmid was transformed into
E. coli BL21 (DE3) cells. Recombinant E. coli BL21 (DE3)
was cultured in 10 mL of Lysogeny Broth (LB, 10 g L−1
tryptone, 5 g L−1 yeast extract, and 10 g L−1 NaCl) with
0.1 g L−1 ampicillin. The culture was incubated overnight at
37 °C and 220 rpm in a shaking incubator. Overnight culture
(5 mL) was inoculated to 500 mL of 2xYT Medium Broth
(16 g L−1 tryptone, 10 g L−1 yeast extract, and 5 g L−1 NaCl
at pH 6.8) with 0.1 g L−1 ampicillin and grown at 37 °C
with 220 rpm shaking until optical densities reached~0.7
at OD600. Expression of WT CYP119 was induced with
1 mm isopropyl β-d-1-thiogalactopyranoside (IPTG) for
32 h at 30 °C. The cells were harvested by centrifugation
at 3800 rpm for 30 min and pellets were stored at − 80 °C
until purifcation.
Even though natural redox partner and substrate(s) of
CYP119 are unknown, previous studies have shown that
CYP119 can catalyze the peroxidation of Amplex® Red
(N-acetyl-3,7-dihydroxyphenoxazine) in the presence of
H2O2 [15, 17, 18]. Amplex® Red is a non-fuorescent rea-
gent, which is converted by oxidation to a red fuorescent
product resorufn that has excitation maximum at 571 nm
and emission maximum at 585 nm [19–21]. Amplex® Red
has commonly been used as a probe in the detection of H2O2
in conjunction with horseradish peroxidase (HRP) [28].
However, due to the complex mechanism of Amplex® Red
oxidation and low activity of Amplex® Red with P450s, it
has never been used in cell lysates to screen for activity of
mutant P450s. We optimized the Amplex® Red screen to
assess CYP119 peroxidase activity. Improved catalytic activ-
ity of the mutant library was measured by an assay involving
the conversion of Amplex® Red to resorufn in the presence
of H2O2.
Enrichment of WT CYP119
A mutant library of CYP119 was generated to screen for
increased peroxidation activity using the optimized assay.
Since mutations in the active site are expected to have a
much larger efect on activity, we decided to have a targeted
approach. The combinatorial active-site saturation test
(CAST) method was used to generate the mutant library. In
The frozen cell pellets obtained were dissolved in lysis bufer
(0.1 m NaCl, 0.2 mm phenylmethylsulfonyl fuoride (PMSF),
1 mm benzamidine HCl, and 50 mm potassium phosphate
at pH 7.5). Cells were ultrasonicated, and cell lysate was
incubated at 65 °C for 1 h, followed by centrifugation at
1 3

JBIC Journal of Biological Inorganic Chemistry
12,000 rpm for 20 min at room temperature. WT CYP119
was enriched in the collected supernatant.
mixture 1 (1 mm H2O2, 50 µm Amplex® Red, and 50 mm
potassium phosphate bufer, pH 7.4).
Harvesting of protein by SoluLyse treatment was done
by dissolving pellets in SoluLyse reagent (300 µL) and then
centrifuged to obtain proteins in the supernatant. Collected
supernatants were used in the Amplex® Red peroxidation
activity assay. A reaction volume of 100 µL contains 50 µL
of extracted enzyme and 50 µL of reaction mixture 1.
For both methods, proteins extracted from E. coli BL21
(DE3) cells that contain pET20b were used as a negative
control group and HRP was used as a positive control group.
Mixtures were inoculated for 30 min at 25 °C; fuorescence
emission values of resorufn between 550 and 700 nm were
measured with excitation at 530 nm. Each reaction was
done quintuplicate; graphs were created using mean values
of duplicates.
Generation of mutant library
Generation of mutant library was accomplished with the
polymerase chain reaction (PCR) using Q5 Site-Directed
Mutagenesis Kit (BioLabs). Degenerate primers that con-
tain NDT codons at Thr213–Thr214 positions (GGGTAA
TGAGNDTNDTACTAACTTAATATCAAACTC TGTTAT
TG and GCTATGAGAAGTAAAATAATGTATC) were
used. CYP119 mutant library was created by heat-shock
transformation of the plasmids containing mutant genes to
E. coli BL21 (DE3). Colonies were randomly selected from
the library; their plasmids were isolated and then sequenced
to confrm mutations at desired sites.
Rapid screening of CYP119 mutant library
Peroxidase activity of WT CYP119 and T213R/T214I
mutant with Amplex® Red
The rapid screening method used for CYP119 mutant library
is summarized in Fig. 1. 96-well cell culture plates that con-
tain 140 µL LB media with 0.1 g L−1 ampicillin in each well
were inoculated with E. coli BL21 (DE3) bacteria that have
plasmids of WT or mutant CYP119 or pET20b. Plates were
incubated overnight at 37 °C with moderate shaking. 3 µL of
each culture was transferred to 96 deep well plate in a trip-
licate manner, each well containing 500 µL of 2xYT media
with 0.1 g L−1 ampicillin. Cultures were incubated at 37 °C
with moderate shaking until reaching OD600 ~0.7 value. A
fnal concentration of 1 mm IPTG was added to each culture
in two of the copy plates to induce protein expression and
Amplex® Red (Thermo Fisher Scientific) peroxidation
experiments were carried out as previously described [15,
18]. Briefy, reaction mixtures included 10 μm Amplex®
Red, 1.5 mm H2O2, 1 mm ethylenediaminetetraacetic acid
(EDTA), and 1.5 μm WT CYP119 or variants in 50 mm
potassium phosphate at pH 7.4. The reactions occurred
at room temperature. Production of resorufin was ana-
lyzed by monitoring fuorescence emission (570 nm exci-
tation and 585 nm emission) or absorbance at 570 nm
(ε570nm =54 mm−1 cm−1).
Development of screening method
E. coli BL21 (DE3) cells that contain plasmids with WT
or mutant CYP119 and empty pET20b genes were grown
overnight, in LB media (5 mL) containing 0.1 g L−1 ampi-
cillin, at 37 °C with moderate shaking. Each culture (30
µL) was inoculated to 2xYT (8 mL) media containing 0.1 g
L−1 ampicillin and were grown until reaching OD600 ~0.7 at
37 °C with moderate shaking. A fnal concentration of 1 mm
IPTG was added to each culture media. Protein expression
was induced for 32 h at 30 °C with moderate shaking. Ali-
quots from each culture (3 mL) were harvested by centrifu-
gation and kept at − 80 °C for SoluLyse (protein extraction
reagent, Genlantis) treatment or sonication processes.
Harvesting of protein by sonication was done by dissolv-
ing cell pellets in lysis bufer (300 µL) and disrupting cells
by sonication. Lysed cells were heat-treated at 65 °C for 1 h
and then centrifuged to enrich CYP119 in supernatant. Col-
lected supernatants from each culture were used in Amplex®
Red peroxidation activity assay. A reaction volume of 100
µL contains 50 µL of extracted enzyme and 50 µL of reaction
Fig. 1 Steps of the rapid screening method. 1. Transfer of mutant col-
onies from LB agar plates into 96-well plates that contain LB media
in each well. 2. Transfer of grown cultures to 96 deep well plates that
contain 2xYT media in each well. 3. Visualization of peroxidation
activity of CYP119 variants. 4. Glycerol cell stocks of grown cultures
for further use of variants
1 3

JBIC Journal of Biological Inorganic Chemistry
cultures were incubated at 30 °C for 20 h with moderate
shaking. Culture plates were centrifuged at 15,000 rpm for
5 min. Pellets were kept in − 80 °C freezer overnight to
help with cell lysing. Pellets were suspended in 50 µL of
SoluLyse reagent. Duplicates were merged in a microcentri-
fuge tube and incubated at 25 °C with moderate shaking for
10 min. Proteins were enriched in the supernatant by cen-
trifugation at 15,000 rpm for 10 min. Collected supernatants
were used in Amplex® Red peroxidation activity assay. The
third copy plate was kept at − 80 °C after 50% glycerol was
added to cultures in a 1:1 volume ratio for traceback.
10 mm imidazole, 0.2 mm PMSF, 1 mm benzamidine HCl,
50 mm potassium phosphate bufer, and pH 7.5). Cells were
lysed by sonication and incubated in water bath at 60 °C for
1 h. Lysate was then centrifuged at 3900 rpm for 1.5 h at
4 °C. WT CYP119 was isolated from the supernatant by pre-
cipitation with ammonium sulfate (60%) and dissolved with
potassium phosphate bufer (50 mm potassium phosphate
bufer, 20 mm NaCl, 5% glycerol, and pH 7.5) and dialyzed
against the same bufer. T213R/T214I mutant pellets were
dissolved with triethanolamine bufer (50 mm) at pH 7.3
and desalted with PD 10 desalting column. 10 μL samples
of WT and T213R/T214I were taken for purity analysis by
SDS-PAGE (Figure S1). WT CYP119 and the T213R/T214I
mutant concentrations were calculated based on the extinc-
tion coefcient of CYP119 (ε415nm =104 mm−1 cm−1).
Amplex® Red peroxidation activity assay
in the rapid screening
Amplex® Red peroxidation activity assay has a total of
100 µL reaction volume that consist of 50 µL of enzyme
extracted with SoluLyse reagent and 50 µL of reaction mix-
ture 2 (1 mm H2O2, 10 µm Amplex® Red, and 50 mm potas-
sium phosphate bufer, pH 7.4). 1:1 v/v SoluLyse reagent
and potassium phosphate bufer were used as blank, 1:1 v/v
potassium phosphate bufer, and reaction mixture 2, and
proteins extracted from E. coli BL21 (DE3) cells that con-
tain pET20b mixed with reaction mixture 2 were used as
negative controls. The positive control was 1 µU of HRP in
1:1 v/v potassium phosphate bufer and reaction mixture 2.
Samples were incubated at 25 °C for 10 min. Fluorescence
emission by resorufn was measured at 595 nm with FLU-
Ostar Omega (BMG LABTECH).
Electronic spectroscopy
All spectra were recorded on a UV-1600PC Scanning Spec-
trophotometer at room temperature. The UV–Visible spectra
of WT CYP119 and T213R/T214I mutant were obtained in
50 mm potassium phosphate bufer at pH 7.4.
Kinetic analysis of Amplex® Red peroxidation by WT
CYP119 and T213R/T214I mutant
Amplex® Red peroxidation was performed with WT
CYP119 and the T213R/T214I mutant and absorbance
change was followed at 570 nm. Reaction mixtures included
10 μm Amplex® Red, 1.5 mm H2O2, 1 mm EDTA, and
1.5 μm of enzyme (CYP119 or T213R/T214I mutant) in
50 mm potassium phosphate at pH 7.4. UV–Visible spectra
were taken at 0, 2, 5, 10, 20, 30, 45, 60, 90, and 120 min
after the addition of H2O2. Amplex® Red Reactions with
WT CYP119 and T213R/T214I mutant were analyzed
under diferent H2O2 concentrations (0.5–2.5 mm) for the
determination of kinetic parameters. The initial rate was
calculated for each H2O2 concentration by linear ftting. A
Lineweaver–Burk plot is shown for comparison, and kinetic
parameters were obtained by nonlinear ftting of the data to
Michaelis–Menten equation.
Confrmation of improved activity mutant enzymes
Mutant enzymes that exhibited improved activity were
picked and grown from the third replicate plate. LB media
(5 mL) containing 0.1 g L−1 ampicillin were inoculated
with improved activity mutant enzymes. Cultures were
grown overnight at 37 °C with moderate shaking. 2xYT
media (8 mL) containing 0.1 g L−1 ampicillin were inocu-
lated with 30 µL of cultures. Upon reaching OD600 ~0.7, a
fnal concentration of 1 mm IPTG was added to cultures.
3 mL of cultures were incubated for 25 h at 30 °C with
moderate shaking. Harvesting was done by centrifugation at
15,000 rpm for 10 min. Pellets were resuspended in 300 µL
of SoluLyse reagent and incubated at 25 °C for 10 min with
moderate shaking. Proteins were enriched in the supernatant
by centrifugation. Collected supernatants were subjected to
Amplex® Red peroxidation activity assay.
Substrate binding study
Binding constants of WT CYP119 and T213R/T214I mutant
were determined at room temperature by diference spectros-
copy. Enzyme solutions for WT CYP119 (1.5 µm) or T213R/
T214I (1.5 µm) were prepared in 50 mm potassium phosphate
bufer at pH 7.4. Enzyme solutions were kept for 10 min at
room temperature and divided into two glass cuvettes. Lau-
ric acid in a dimethyl sulfoxide (DMSO) stock solution was
titrated into the sample cuvettes containing WT CYP119
and T213R/T214I mutant. The concentration of DMSO in
Isolation and purifcation of WT CYP119 and T213R/
T214I mutant
The frozen cell pellets of WT CYP119 and T213R/T214I
mutant were resuspended with lysis bufer (150 mm NaCl,
1 3

JBIC Journal of Biological Inorganic Chemistry
the cuvette did not exceed 1% of the initial volume. Equal
amounts of DMSO were added to the reference cuvette. The
absorbance shift between 386 and 418 nm was followed. The
shift was plotted against substrate concentration by nonlinear
ftting. Kd value was calculated by plotting the shift against
substrate concentration to quadratic equation [23]:
to FastRelax protocol using “REF2015” as energy function
before ligand docking. Ligand docking was performed with
the DockMCMProtocol of PyRosetta. 1000 rounds of dock-
ing were performed for each structure and docking scores
were calculated according to PyRosetta energy function,
“ligand”. Dock models with the lowest Rosetta Energy Unit
(REU) scores were selected to analyze ligand–protein inter-
actions. All images were created with the UCSF Chimera
software [25].
�
(
)
⎛
⎜
⎜
⎞
⎟
⎟
Kd + [E] + [L] −
Kd + [E] + [L] 2 − 4[E][L]
ΔA = A
max
2[E]
⎜
⎝
⎟
⎠
Results
Styrene epoxidation
Peroxidation of Amplex® Red by CYP119
Styrene epoxidation was performed at room temperature in
closed glass vials for 10 min. 100 µl reaction mixture con-
tained 12.5 µm CYP119, variable concentrations of styrene
(3–7 mm), and 7 mm tert-butyl hydroperoxide (TBHP) as
oxidant in 50 mm phosphate bufer (pH 7.4). The reaction
was stopped by adding acetonitrile (900 µL). The reaction
mixture was analyzed by high-performance liquid chroma-
tography (HPLC) on a Thermo Scientifc Ultimate 3000
coupled with a UV detector fxed at 220 nm. 25 cm Nucle-
odur C18 column (Macherey–Nagel) was used with 3:7
ddH2O:acetonitrile as the mobile phase. The retention times
of styrene and styrene oxide were determined as 6.5 min and
4.27 min, respectively. Solutions containing known amounts
of ( )-styrene oxide were used to generate standard curves
for product quantitation.
Previous studies have shown that CYP119 has peroxidase
activity with Amplex® Red as a substrate and this activ-
ity can be monitored by fuorescence spectroscopy [15, 18].
Oxidation of Amplex® Red by peroxidase results in the for-
mation of the fuorescent compound resorufn, which emits
fuorescence at 585 nm when excited at 571 nm.
Peroxidase activity of CYP119 was assayed in cell
extracts after cell lysis by sonication and enrichment of
CYP119 by heat treatment. WT CYP119 can be obtained
approximately 70–80% pure through this method (Fig-
ure S2). Enriched CYP119 (2.5 µm) showed oxidation of
Amplex® Red (50 µm) using H2O2 as an oxidant (1 mm) at
room temperature in sodium phosphate bufer (50 mm) at
pH 7.4. The fuorescence spectra of the resorufn product
obtained are shown in Fig. 2. Based on this result, the oxi-
dation of Amplex® Red by CYP119 was selected for library
screening for peroxidation activity.
Reactions of WT CYP119 and T213R/T214I mutant
with hydrogen peroxide
Hydrogen peroxide (H2O2) reactions were performed
with WT CYP119 and T213R/T214I mutant, and absorb-
ance was followed at 414 nm. Reaction mixtures included
1.5 mm H2O2, 1 mm EDTA, and 1.5 μm of enzyme (CYP119
or T213R/T214I mutant) in 50 mm potassium phosphate at
pH 7.4. UV–Visible spectra were taken at 0, 2, 5, 10, 20, 30,
and 45 min after the addition of H2O2.
Molecular modeling and docking of Amplex® Red
to WT CYP119 and T213R/T214I mutant
The crystal structure of WT CYP119 (PDB ID: 1F4T) was
obtained from the RCSB PDB database [13]. The PyRo-
setta program was used for creating mutations on the struc-
ture of WT CYP119 enzyme [24]. Mutations were applied
to WT CYP119 structure using the TaskFactory module
of PyRosetta. Conformation of mutant residues was opti-
mized with the PackRotamersMover function according to
the “REF2015” score function. WT CYP119 and T213R/
T214I mutant structures were energy minimized according
Fig. 2 The fuorescence spectra of the resorufn produced by the
reaction of enriched WT CYP119 (2.5 µm), Amplex® Red (50 µm),
and H2O2 (1 mm) in 50 mm sodium phosphate bufer at pH 7.4. The
resorufn spectra are shown for 570 nm excitation (dash) and 585 nm
emission (solid), respectively
1 3

JBIC Journal of Biological Inorganic Chemistry
Screening method for selection of CYP119 mutants
for improved peroxidation activity
Screening of CYP119 mutant library
After the construction of the mutant library, CYP119
mutants were screened for improved peroxidation activity
as described in the Materials and method section (Fig. 4).
After 10 min of incubation at room temperature, fuores-
cence emitted by resorufn in each well was measured.
Incubation for longer than 10 min did not result in signif-
cant changes in fuorescence (results not shown). HRP (1
µUnits) with the reaction mixture 2 (10 µm Amplex® Red
and 1 mm H2O2 in 50 mm pH 7.4 potassium phosphate
bufer) was used as a positive control. Cells harboring
empty pET20b plasmid in reaction mixture 2 were used
as the negative control.
Obtaining enriched CYP119 by sonication and heat treat-
ment is time-consuming for a rapid activity assay. To develop
a feasible screening method, we decided to use SoluLyse
treatment as described in the Materials and methods section.
The peroxidase activity of enriched CYP119 obtained by
sonication as described above was compared with activity
observed in the cell lysate obtained with SoluLyse method
(Figure S3). In both methods, proteins extracted from E. coli
BL21 (DE3) containing WT or mutant CYP119 genes were
mixed with a reaction mixture (1 mm H2O2, 10 µm Amplex®
Red, and 1 mm EDTA in 50 mm pH 7.4 potassium phosphate
bufer) and incubated at room temperature. Cells that have
empty pET20b plasmid were used as a control group. Fluo-
rescence emission of the product resorufn was followed at
585 nm for all samples. As seen in Figure S3, the fuores-
cence obtained by the SoluLyse method was comparable to
enrichment treatment for WT, mutant CYP119, and pET20b
control. Therefore, SoluLyse treatment method was used in
our assay.
Peroxidation activity assay with Amplex® Red was
performed on 158 mutants to determine variants with
improved activity. As shown in Fig. 4 among 158 mutants,
there were 13 mutants that showed improved activity.
Then colonies showing the improved activity of CYP119
mutants were grown on a bigger scale to confrm their
improved activities and for further characterization. The
peroxidation activity of these mutants was confrmed by
larger scale expression tests (described in the Materials
and method section). An increase in peroxidase activity
was confrmed for six colonies (Figure S4) in larger scale
and these mutants were sent for sequence analysis. Among
these mutants, we decided to further characterize the novel
T213R/T214I variant of CYP119 that has not been char-
acterized before.
Construction of CYP119 mutant library for screening
peroxidase activity
Previous studies have shown that residues Thr213 and
Thr214 were important for the peroxidase activity of
CYP119 [11]. Therefore, CYP119 mutant library was cre-
ated by applying the CAST method to the amino acids
Thr213 and Thr214 [16]. The nucleotides coding the two
threonine residues were mutated to NDT codons coding
for 12 amino acids (Phe, Leu, Ile, Val, Tyr, His, Asn, Asp,
Cys, Arg, Ser, and Gly). To test if there is any mutational
bias in the library, 31 colonies were selected at random and
sequenced. The sequences obtained were analyzed by sub-
stitution analysis, as seen in Fig. 3, no mutational bias was
observed in the mutant library.
Fig. 4 Relative activities of WT and variants of CYP119, and
pET20b samples obtained through Solulyse treatment in 96-well
plates plotted in descending order. Relative fuorescence activity
obtained from reactions occurred in a single well for each sample.
WT CYP119 is set as baseline at 1 (dash line) and T213R/T214I
(DM) is indicated with an arrow
Fig. 3 Substitution analysis of 31 variants of CYP119, at Thr213
(637-639), and Thr214 (640-642) mutation sites [11]. Mutations cre-
ated using NDT codons. Created using WebLogo3 [26]
1 3

JBIC Journal of Biological Inorganic Chemistry
WT or T213R/T214I mutant CYP119 at 25 °C are shown
in Fig. 6. An increase at 570 nm can be observed during the
reaction, this can be attributed to resorufn formation. At the
end of the reaction, 7% of Amplex® Red was oxidized by
WT in comparison to 60% of Amplex® Red by the T213R/
T214I mutant, based on the fnal concentration of resorufn
using the extinction coefcient at 570 nm [16]. Therefore,
there was an eightfold increase in the yield of Amplex® Red
oxidation by the T213R/T214I mutation.
Spectral characterization of the T213R/T214I
mutant
The optical spectra of WT and T213R/T214I mutant
CYP119 as isolated are shown in Fig. 5. WT CYP119
shows maximum Soret absorbance at 414 nm and split
alpha–beta (α/β) bands at 531 and 564 nm, similar to previ-
ously reported spectra [27]. The T213R/T214I mutation did
not result in a signifcant Soret shift with a Soret maximum
at 415 nm; however, there was a broadening of the Soret
absorbance and α/β bands (Fig. 5).
Kinetic analysis of Amplex® Red oxidation by WT
CYP119 and T213R/T214I mutant
Peroxidase activity of WT CYP119 and T213R/T214I
mutant with Amplex® Red
The kinetic parameters for Amplex® Red oxidation cata-
lyzed by these proteins were investigated (Fig. 7). When
Amplex® Red oxidation was monitored in the presence of
increasing concentrations of H2O2, the kcat obtained was
The peroxidase activity of purifed WT and T213R/T214I
mutant of CYP119 was investigated by following the oxida-
tion of Amplex® Red by H2O2 with UV–Visible spectros-
copy [15]. The changes in the UV–Visible spectra during the
reaction of 10 µm Amplex® Red, 1.5 mm H2O2, and 1.5 µm
Fig. 7 Determination of the kinetic parameters of Amplex® Red oxi-
dation by H2O2 in the presence of WT CYP119 (empty circle) and
T213R/T214I variant (DM, flled triangle). Reactions contained
10 μm Amplex® Red, 1.5 μm T213R/T214I or WT CYP119, and 0.5–
2.5 mm of H2O2 in 50 mm potassium phosphate bufer, pH 7.4
Fig. 5 The comparison of UV–visible spectra of T213R/T214I
mutant and WT CYP119 (solid: WT CYP119; dash: T213R/T214I
CYP119). Inset: The changes observed in the alpha–beta region
Fig. 6 Changes in the UV–visible spectra during Amplex® Red
oxidation by H2O2 in the presence of WT (a) and T213R/T214I
mutant (b); the UV–visible spectra taken at 0, 2, 5, 10, 20, 30, and
45 min after addition of H2O2. The reaction components were 10 μm
Amplex® Red, 1.5 mm H2O2, and 1.5 μm enzyme, 1 mm EDTA in
50 mm potassium phosphate bufer at pH 7.4. Time course of changes
in resorufn formation at 570 nm for WT CYP119 (empty circle) and
T213R/T214I variant (DM, flled triangle) (c) in the same reaction
1 3

JBIC Journal of Biological Inorganic Chemistry
2.1 0.3×10−3 s−1 for WT and 1.1 0.13×10−2 s−1 for the
T213R/T214I mutant. There was a fvefold increase in the
kcat observed in the T213R/T214I mutant. On the other hand,
the Km observed for H2O2 was 0.82 0.46 mm for WT com-
pared to 1.32 0.3 mm for T213R/T214I mutant; showing
that the Km for H2O2 was not signifcantly altered by the
mutation.
the aid of calibration curves of pure styrene and styrene
epoxide. Kinetic constants were obtained by nonlinear ft-
ting analysis with variable concentrations of the substrate
styrene (Fig. 8). The kcat obtained were 5 0.1 × 10−2
s−1 and 11.7 0.5 × 10−2 s−1 for WT and T213R/T214I
mutant, respectively. While the Km obtained for styrene
were 14.8 3.2 mm and 14.0 8.3 mm for WT and T213R/
T214I, respectively. Therefore, T213R/T214I mutant showed
increased kcat of styrene epoxidation without altering the Km.
Styrene epoxidation by WT CYP119 and T213R/
T214I mutant
Reactions of WT CYP119 and T213R/T214I mutant
with hydrogen peroxide
We also investigated the changes in the styrene epoxidation
activity of WT and the T213R/T214I mutant of CYP119.
The styrene epoxidation was carried out at 25 °C and pH
7.4 in the presence of TBHP as an oxidant. The rate of sty-
rene epoxidation was determined by HPLC analysis with
To study the reaction of H2O2 with WT and T213R/T214I
CYP119, we monitored the changes in heme Soret during
the reaction of the P450s with H2O2. The reaction of 1.5 µm
WT with 1.5 mm H2O2 resulted in a decrease in the Soret
absorbance at 414 nm (Fig. 9a). A 15% decrease in absorb-
ance at 414 nm was observed after incubation of WT with
H2O2 for 45 min at 25 °C. The observed rate constant for
the decrease in Soret absorbance was 1.8 0.6×10−3 s−1.
The reaction of T213R/T214I with H2O2 was also fol-
lowed by UV–Vis spectroscopy. The reaction of 1.5 µm
T213R/T214I with 1.5 mm H2O2 did not result in a signif-
cant decrease in maximum Soret absorbance but instead, a
narrowing of Soret band was observed (Fig. 9b). The difer-
ence spectra show a decrease in 390 nm (Fig. 9b inset). As
seen in Fig. 9c the heme Soret maximum 414 nm did not
change signifcantly during the reaction for T213R/T214I,
while, for the WT, a signifcant decrease was observed.
Substrate binding by WT CYP119 and T213R/T214I
mutant
Fig. 8 Determination of the kinetic parameters of styrene epoxida-
tion by variable concentrations of styrene (3–7 mm) in the presence of
WT CYP119 (empty circle) and T213R/T214I (DM, flled triangle).
Reactions contained variable concentration styrene (3–7 mm), 7 mm
TBHP, and 12.5 µm enzyme in 50 mm potassium phosphate bufer at
pH 7.4 and 25 °C for 10 min
Residues Thr213 and Thr214 are located at the active site
of the enzyme; therefore, the efects of mutation on the sub-
strate binding to CYP119 were also investigated. Lauric
acid was investigated as substrate and the Kd values were
Fig. 9 Changes in the UV–visible spectra during H2O2 reaction with
WT (a) and T213R/T214I mutant (DM) (b). Observed spectra at 0,
2, 5, 10, 20, 30, and 45 min. Insets show diference spectra. Reaction
contained 1.5 mm H2O2, and 1.5 μm enzyme, 1 mm EDTA in 50 mm
potassium phosphate bufer at pH 7.4. (c) Decrease in 414 nm of WT
(empty circle) and DM (flled triangle)
1 3

JBIC Journal of Biological Inorganic Chemistry
determined from the magnitude of the spin state change
which is linked to the magnitude of the absorbance shift
(Fig. 10). A larger Soret shift was observed for T213R/
T214I mutant compared to WT. In addition, T213R/T214I
mutant showed fourfold lower affinity for lauric acid
(Kd =59.1 30.2 μm) compared to WT (Kd =16.8 4.5 μm).
mutant (Fig. 11) which indicates that Amplex® Red was
located closer to the heme group in T213R/T214I mutant.
As shown in Fig. 11, Amplex® Red also forms a hydro-
gen bond with the backbone of Val258 of T213R/T214I
mutant. 1000 rounds of ligand docking were performed
for both WT CYP119 and the T213R/T214I mutant. Score
distribution histograms are shown in Figure S5. Since the
PyRosetta scoring function is a combination of physics-
based and knowledge-based information about the struc-
ture, the calculated energy scores have the generic unit
of REU. Models that have more negative REU scores are
generally accepted to be more native-like [28], and thus,
docked models with the lowest REU scores were selected
to analyze ligand–protein interactions. Lowest REU scores
for Amplex® Red docked with WT CYP119 and T213R/
T214I mutant models are -517 REU and -1186 REU,
respectively.
Molecular modeling and docking of Amplex® Red
to WT CYP119 and T213R/T214I mutant
Ligand docking of Amplex® Red molecule into the WT
CYP119 (PDB ID: 1F4T) and the T213R/T214I mutant
was performed by using the PyRosetta program [24].
The distance between the heme iron atom and the tar-
geted nitrogen atom of Amplex® Red was measured as
6.57 Å for WT CYP119 and 6.26 Å for T213R/T214I
Fig. 10 UV–Visible and diference spectra of WT (a) and T213R/
T214I mutant (b) titrated with saturated lauric acid. Overlay of UV–
Visible spectra of 1.5 µm WT CYP119 or T213R/T214I mutant with
diferent concentrations of lauric acid (0.5–50 μm). The insets show
diference spectra with increasing lauric acid concentrations. c The
absorbance shifts observed in 386 nm and 418 nm for lauric acid
binding titration of WT CYP119 (empty circle) and T213R/T214I
mutant (DM, flled triangle). The absorbance diference between
386 nm and 418 nm [ΔΔAbs (386–418 nm)] versus concentrations of
lauric acid plotted by nonlinear ftting
Fig. 11 Docking results of Amplex® Red (pink) with WT CYP119
(PDB ID: 1F4T) and T213R/T214I mutant. a WT CYP119 with
Amplex® Red. Thr213 (green) and Thr214 (orange) residues are
labeled. b T213R/T214I mutant with Amplex® Red. Mutant residues
are labeled as T213R (green) and T214I (orange). Hydrogen bond
between Amplex® Red and Val258 shown is shown as a black line
1 3

JBIC Journal of Biological Inorganic Chemistry
due to the complex mechanism of Amplex® Red peroxida-
tion and the low activity of CYP119 with Amplex® Red
(kcat for WT CYP119 is 2 × 10−3 s−1, Fig. 7) [31]. Indeed,
our initial attempts at monitoring Amplex® Red peroxi-
dation with whole-cell catalysis in cells expressing WT
CYP119 did not result in any signifcant product forma-
tion (results not shown). The addition of SoluLyse reagent
allowed rapid lysis of cells and substrate conversion by
CYP119. In addition, steps in the assay (steps 2 and 3,
Fig. 1) had to be optimized to maximize WT or mutant
CYP119 concentration in the assay to ensure a signifcant
increase in fuorescence compared to the control. The fnal
CYP119 concentration in the assay was determined to be
around 0.5 µm by the Soret absorbance at 415 nm (results
not shown). The assay was also optimized by lowering the
background fuorescence observed in the negative control,
cells containing empty pet20b plasmid, via the addition
of 1 mm EDTA (Figure S7). These optimizations allowed
rapid screening of mutant colonies for peroxidase activity;
to our knowledge, this is the frst application of Amplex®
Red to detect peroxidase activity of mutants in cell lysate.
We have shown that the conversion of Amplex® Red to
resorufn by CYP119 can also be detected in enriched cell
lysate without further purifcation of the protein (Fig. 2).
In addition, enrichment of CYP119 was not essential for
monitoring activity, simple cell lysis by SoluLyse method
was enough for detection of activity (Figure S3). This new
method signifcantly increased the throughput of the assay
and 158 mutants of CYP119 were screened for improved
peroxidation activity (Fig. 4). This screening led to the iden-
tifcation of six colonies with increased CYP119 peroxida-
tion activity (Figure S4). The six colonies contained fve
mutations: T213R/T214I, T213R/T214Y, T213D/T214I,
T213D/T214N, and T213C/T214C (T213R/T214Y muta-
tion was observed in two diferent colonies). Among these
improved variants, T213R/T214I was selected for further
characterization.
Discussion
The development of novel P450 enzymes that can utilize
H2O2 efciently will open the way for the far wider appli-
cation of P450s as biocatalysts in industry. Here, we devel-
oped an assay that utilizes Amplex® Red peroxidation, to
rapidly screen H2O2-dependent activity of P450 mutants
in cell lysate. This assay was applied to screen a CYP119
mutant library to increase its activity with H2O2. Combina-
torial Active-site Saturation Test (CAST) method was used
to generate the CYP119 mutant library [16]. CYP119 vari-
ants with higher activity using H2O2 as an oxidant were
selected by the rapid screen for Amplex® Red peroxida-
tion. A novel high activity CYP119 variant (T213R/T214I)
obtained from the screen was characterized chemically and
spectroscopically to gain insights into the mechanism of
H2O2 reactions with P450 enzymes.
In most activity screens, only a small fraction of all
the possible mutated proteins can be screened. There-
fore, the CAST method was used to generate a targeted
mutant library. Previous studies have shown that Thr213
and Thr214 residues in the active site have key roles for
enzyme activity, but they are not critical for thermal sta-
bility [11]. Therefore, these residues were selected for
targeted mutation.
To improve the catalytic activity of CYP119 with
H2O2, we developed an activity assay with a substrate
whose product can be detected rapidly using fluores-
cence. Amplex® Red has been used as a highly sensitive
and chemically stable fuorogenic probe for enzymatic
detection for H2O2 [29]. Previous studies have shown that
CYP119 can catalyze Amplex® Red peroxidation [15,
18]. The fuorescent product of Amplex® Red peroxida-
tion, resorufn, has a high extinction coefcient and good
chemical and photostability; therefore, it is an ideal can-
didate for rapid screening.
Amplex® Red has commonly been used in assays for
the detection of H2O2 in conjunction with HRP [29]. Rabe
et al. also developed an assay using Amplex® Red with
catalase as a reporter enzyme to screen enzyme/peroxide/
substrate combinations and assess P450 reactivity with
various substrates [17]. The assay developed by Rabe et al.
determined activity by monitoring peroxide depletion
during the reaction, which makes it prone to interference
by other enzymes in the cell lysate; consequently, it can
only be used with isolated enzymes. Another Amplex®
Red-peroxidase assay was developed for screening the
efects of additives and cosolvents on P450 enzymes [30],
this assay was also performed with isolated enzymes at
micromolar concentrations. The Amplex® Red assays in
the literature were not compatible for rapid screening of
CYP119 mutants, because they require enzyme isolation
Previous studies have indicated that the distal water
ligand is more tightly bound in CYP119 compared to other
P450s; which makes CYP119 low spin and conversion
to high spin is much more difcult [11]. Replacement of
Thr214 and Thr213 with large amino acids considerably
increased the proportion of high spin. The improved vari-
ant, T213R/T214I, showed broadening of the Soret peak,
which can be explained by changes in the spin state of the
heme iron (Fig. 5).
When substrate binding of T213R/T214I mutant is com-
pared to WT, in addition to changes in afnity for lauric
acid, binding of substrate to T213R/T214I mutant results
in a more signifcant increase in 390 nm absorbance com-
pared to WT (Fig. 9). This shift in Soret absorbance can
also be attributed to changes in the spin state of the heme
iron to high spin. Taken together, the conversion of the
1 3

JBIC Journal of Biological Inorganic Chemistry
heme iron in T213R/T214I to high spin is easier and the
protein is most likely isolated in a mixed spin state rather
than low spin. This result is consistent with the previous
studies that indicated that the Thr214 residue is involved
in controlling the spin state of the heme iron [11]. In addi-
tion, alignment of the active site of WT CYP119 with
modeled T213R/T214I variant (Figure S6) shows that
the T213R mutation displaces the distal water molecule,
which will also lead to changes in the resting spin state of
the heme iron.
(kcat=1×10−2 s−1 with H2O2 as the oxidant at 30 °C [11],
kcat=1.3 s−1 with TBHP as the oxidant at 70 °C [15]).
In previous studies, styrene epoxidation of CYP119
has also been signifcantly improved by single mutation
of T213E [32]. Indeed, the kcat for styrene epoxidation
by T213E CYP119 was shown to be 0.27 s−1 at 70 °C in
the presence of 60 mm H2O2. The observed kcat for T213E
appears to be twofold higher than for T213R/T214I; how-
ever, the reactions were performed under diferent conditions
(25 °C vs 70 °C and oxidant 7 mm TBHP vs 60 mm H2O2).
Consequently, it is not possible to compare these mutants
directly. Evidence from previous studies have shown that
changing oxidant and increasing temperature can lead to a
tenfold increase in kcat for CYP119 [11, 14].
To test if the T213R/T214I variant will show increased
Amplex® Red peroxidation activity when isolated and to
investigate the changes in Amplex® Red peroxidation
mechanism, the reaction of WT and T213R/T214I mutant
with Amplex® Red and H2O2 was investigated. The mutant
showed an eightfold increase in Amplex® Red peroxidation
yield compared to WT. When the reaction kinetics were fol-
lowed with increasing concentrations of H2O2, a fvefold
increase in kcat was observed for the T213R/T214I mutant
(Fig. 7). On the other hand, the Km for H2O2 was not afected
by the mutation.
In this study, the CYP119 mutant library was generated
using the NDT codon degeneracy which utilizes 12 amino
acids (Phe, Leu, Ile, Val, Tyr, His, Asn, Asp, Cys, Arg, Ser,
and Gly). Therefore, the mutant library does not contain
the T213E mutation, but the T213D mutation was tested.
Indeed, two of the fve mutants that show higher activity
from the Amplex® Red screen are the T213D/T214I and
T213D/T214N mutants (Figure S4). While the activity of
T213D CYP119 has not been tested previously, the Ala245
residue is located at the same position in P450SPα, and muta-
tional studies by Shoji et al. have shown that A245D P450SPα
also has increased styrene epoxidation activity (but not as
high as A245E P450SPα) [32]. The increased activity in
A245E P450SPα was explained by the carboxylate group of
A245E in the distal pocket acting as an acid–base catalyst in
the generation of the active species during the reaction with
H2O2 similar to the substrate-bound state of P450SPα [32].
The carboxylate group of the substrate in P450SPα is thought
to abstract a peroxide proton leading to the heterolytic cleav-
age of the peroxide to form the active oxygen species com-
pound I [33]. While A245E P450SPα and T213E CYP119
can accelerate the reaction via a similar mechanism, T213R
CYP119 must have a diferent mechanism.
The screen was designed to select for increased Amplex®
Red peroxidation by H2O2. The reasoning behind the screen
was to utilize the excellent probe properties of resorufn to
identify general increased reactivity towards oxidations by
H2O2. Increased Amplex® Red peroxidation activity can
stem from two changes in the active site: increased activ-
ity towards H2O2 and increased afnity for Amplex® Red.
Since kinetic analysis of the peroxidation reaction was not
performed under saturating concentrations of Amplex®
Red, the changes in kcat can be due to changes in afnity for
Amplex® Red. Docking studies were performed to test for
changes in afnity for Amplex® Red. Indeed, docking results
showed lower binding energy (− 1186 REU for T213R/
T214I vs -517 REU for WT), therefore, increased afnity
for Amplex® Red for the improved variant. In addition, the
distance between the heme iron and targeted nitrogen atom
of Amplex® Red decreased for the improved mutant (6.57
Å for WT vs 6.26 Å for T213R/T214I, Fig. 11), which may
lead to more efcient electron transfer.
It is important to note that two of the fve mutants that
show higher activity contain the T213R mutation, T213R/
T214I and T213R/T214Y, (Figure S4), we can thus con-
clude that the T213R mutation is important for increased
activity. One possible explanation for the increase in activ-
ity could be that Arg213 is involved in an interaction with
a close-by carboxylic acid similar to Arg241 in P450SPα
with the carboxylic acid group of the substrate. However, a
detailed analysis of the active site did not show any evidence
of any such interaction. One conceivable outcome of the
T213R mutation is an increase in the polarity of the heme
pocket; peroxidases and P450s that employ H2O2 as an oxi-
dant such as P450SPα and P450BSβ possess more polar heme
pocket compared to the nonpolar heme pockets observed in
other P450s, such as P450BM3 [34]. While Arg213 cannot
abstract a proton from the peroxide, its interactions with
To understand if the mutant CYP119s selected by the
assay also show increased activity for other substrates,
epoxidation of styrene by the T213R/T214I mutant was
investigated. Previous studies have shown that TBHP is a
better electron acceptor than H2O2 for styrene epoxidation
by CYP119; therefore, TBHP was used as an oxidant rather
than H2O2 in styrene epoxidation [15]. Under these condi-
tions T213R/T214I showed twofold higher catalytic activ-
ity for styrene epoxidation (kcat =1.2×10−1 s−1 for T213R/
T214I vs 5×10−2 s−1 for WT). While the afnity for styrene
was not altered by mutation (Km =14 mm for T213R/T214I
vs 14.8 mm for WT). The styrene epoxidation kinetics of
WT CYP119 is within the range of previous observations
1 3

JBIC Journal of Biological Inorganic Chemistry
water molecules in the distal pocket may lead to acid–base
catalysis that promotes heterolysis of the peroxide O–O
bond. In peroxidases such as HRP and cytochrome c per-
oxidase, the distal arginine is involved in stabilizing the
developing negative charge on the leaving oxygen atom thus
promoting heterolysis. The arginine mutants in peroxidases
also show diferent reactivity and stability of compound I
[34]. More mechanistic studies are necessary to understand
the role of the T213R mutation in the peroxidase activity
of CYP119.
mutant was selected for further characterization. T213R/
T214I displayed increased activity for styrene epoxidation in
addition to Amplex® Red peroxidation. In addition, the heme
iron in T213R/T214I mutant demonstrated higher stability
towards H2O2-dependent degradation. While T213R/T214I
is not the frst CYP119 mutant that shows increased peroxi-
dation activity, the activation mechanism of T213R mutation
is novel and future studies are necessary to elucidate the
molecular mechanism. Taken together, P450 mutants with
higher peroxidation activity can be developed by screening
with Amplex® Red as a substrate; this method was used to
identify a novel variant of CYP119 with signifcantly higher
peroxidation activity. Characterization of the higher activity
T213R/T214I variant will contribute to our understanding
of P450s H2O2-dependent oxidation reactions.
The reaction of T213R/T214I with H2O2 was also inves-
tigated to understand how the mutation afected reactivity
towards peroxides in the absence of other substrates. T213R/
T214I heme showed increased stability towards degradation
by H2O2 since no signifcant decrease in Soret absorbance is
observed during its reaction with H2O2 (Fig. 9). This obser-
vation is consistent with the previous studies where Thr213
and Thr214 mutations were heme bleached more slowly by
H2O2 [11]. Therefore, increased Amplex® Red peroxidation
activity was due to both changes in substrate afnity and
reactivity towards peroxides.
Acknowledgements This work was supported by The Scientifc and
Technological Research Council of Turkey [TUBİTAK, 116Z380]. We
thank Ali Oğuz Büyükkileci for his help with the HPLC analysis and
Çağlar Karakaya for allowing access to ultrasonicator instrument.
Compliance with ethical standards
The role of Thr213 in the activation of ferric peroxide
complex during the reaction of CYP119 with H2O2 is not yet
completely understood. Previous studies indicated that the
accessibility of H2O2 was more crucial than the spin state of
the heme iron [11]. However, our results show that the pres-
ence of a bulky charged arginine residue at Thr213 position
can also increase peroxidation activity; therefore, accessibil-
ity to H2O2 may not be as important as previously thought.
Taken together, Amplex® Red peroxidation activity assay
can select for enzymes that show higher activity with per-
oxides. However, because Amplex® Red is used as a sub-
strate inherently the assay will also select for enzymes with
altered substrate selectivity. This may be an asset as one of
the limitations of microbial P450s is their limited substrate
scope, and future studies will focus on changes in substrate
selectivity of the improved variant T213R/T214I.
Conflict of interest The authors declare no confict of interest.
References
1. Denard CA, Ren H, Zhao H (2015) Improving and repurpos-
ing biocatalysts via directed evolution. Curr Opin Chem Biol
25:55–64
2. McLean KJ et al (2015) Single-step fermentative production of
the cholesterol-lowering drug pravastatin via reprogramming of
Penicillium chrysogenum. Proc Natl Acad Sci 112(9):2847–2852
3. Bracco P, Janssen DB, Schallmey A (2013) Selective steroid oxy-
functionalisation by CYP154C5, a bacterial cytochrome P450.
Microb Cell Fact 12(1):95
4. Duport C et al (1998) Self-sufcient biosynthesis of pregne-
nolone and progesterone in engineered yeast. Nat Biotechnol
16(2):186–189
5. Szczebara FM et al (2003) Total biosynthesis of hydrocor-
tisone from a simple carbon source in yeast. Nat Biotechnol
21(2):143–149
Conclusion
6. Hrycay EG, Bandiera SM (2012) The monooxygenase, peroxi-
dase, and peroxygenase properties of cytochrome P450. Arch
Biochem Biophys 522(2):71–89
This study entails the development of an assay to detect the
peroxidation activity of P450s in cell lysate and screening
CYP119 mutants to improve its H2O2-dependent activity.
CAST method was employed in the generation of a CYP119
mutant library by targeting Thr213 and Thr214 residues
for mutation. An efective screening method that utilized
Amplex® Red as a substrate was developed to determine var-
iants displaying higher peroxidation activity while employ-
ing H2O2 as oxidant. This allowed the rapid screening of 158
variants of CYP119 in 96-well plates; fve mutants showed
higher peroxidase activity compared to WT CYP119.
Among the improved variants, the novel T213R/T214I
7. Shoji O, Watanabe Y (2014) Peroxygenase reactions catalyzed by
cytochromes P450. J Biol Inorg Chem 19(4–5):529–539
8. Lee JW et al (2011) Microbial production of building block chem-
icals and polymers. Curr Opin Biotechnol 22(6):758–767
9. Joo H, Lin Z, Arnold FH (1999) Laboratory evolution of
peroxide-mediated cytochrome P450 hydroxylation. Nature
399(6737):670–673
10. Matsuura K et al (2004) Structural and functional characterization
of “laboratory evolved” cytochrome P450cam mutants showing
enhanced naphthalene oxygenation activity. Biochem Biophys Res
Commun 323(4):1209–1215
11. Koo LS et al (2000) The active site of the thermophilic CYP119
from Sulfolobus solfataricus. J Biol Chem 275(19):14112–14123
1 3

JBIC Journal of Biological Inorganic Chemistry
25. Pettersen EF et al (2004) UCSF Chimera—a visualization sys-
tem for exploratory research and analysis. J Comput Chem
25(13):1605–1612
12. Brock TD et al (1972) Sulfolobus: a new genus of sulfur-oxidizing
bacteria living at low pH and high temperature. Archiv für Mik-
robiologie 84(1):54–68
26. Schneider TD, Stephens RM (1990) Sequence logos: a new
way to display consensus sequences. Nucleic Acids Res
18(20):6097–6100
13. Yano JK et al (2000) Crystal structure of a thermophilic
cytochrome P450 from the archaeon Sulfolobus solfataricus. J
Biol Chem 275(40):31086–31092
27. McLean MA et al (1998) Characterization of a cytochrome P450
from the acidothermophilic Archaea Sulfolobus solfataricus. Bio-
chem Biophys Res Commun 252(1):166–172
14. Lim YR et al (2010) Regioselective oxidation of lauric acid by
CYP119, an orphan cytochrome P450 from Sulfolobus acidocal-
darius. J Microbiol Biotechnol 20(3):574–578
28. Misura KM et al (2006) Physically realistic homology models
built with ROSETTA can be more accurate than their templates.
Proc Natl Acad Sci 103(14):5361–5366
15. Rabe KS, Kiko K, Niemeyer CM (2008) Characterization of the
peroxidase activity of CYP119, a thermostable P450 from Sul-
folobus acidocaldarius. ChemBioChem 9(3):420–425
16. Reetz MT et al (2005) Expanding the range of substrate accept-
ance of enzymes: combinatorial active-site saturation test. Angew
Chem Int Ed 44(27):4192–4196
29. Zhou M et al (1997) A stable nonfuorescent derivative of resoru-
fn for the fuorometric determination of trace hydrogen peroxide:
applications in detecting the activity of phagocyte NADPH oxi-
dase and other oxidases. Anal Biochem 253(2):162–168
30. Rabe KS et al (2010) Peroxidase activity of bacterial cytochrome
P450 enzymes: modulation by fatty acids and organic solvents.
Biotechnol J 5(8):891–899
17. Rabe KS et al (2009) Screening for cytochrome p450 reactiv-
ity by harnessing catalase as reporter enzyme. ChemBioChem
10(4):751–757
18. Aslantas Y, Surmeli NB (2019) Efects of N-terminal and C-ter-
minal polyhistidine tag on the stability and function of the ther-
mophilic P450 CYP119. Bioinorg Chem Appl 2019:8080697
19. Ivanec-Goranina R, Kulys J (2008) Kinetic study of peroxidase-
catalyzed oxidation of 1-hydroxypyrene. Development of a
nanomolar hydrogen peroxide detection system. Central Eur J
Biol 3(3):224
31. Towne V et al (2004) Complexities in horseradish peroxidase-
catalyzed oxidation of dihydroxyphenoxazine derivatives: appro-
priate ranges for pH values and hydrogen peroxide concentrations
in quantitative analysis. Anal Biochem 334(2):290–296
32. Shoji O et al (2016) A substrate-binding-state mimic of
H
2O2-dependent cytochrome P450 produced by one-point
20. Mishin V et al (2010) Application of the Amplex red/horserad-
ish peroxidase assay to measure hydrogen peroxide generation
by recombinant microsomal enzymes. Free Radic Biol Med
48(11):1485–1491
mutagenesis and peroxygenation of non-native substrates. Catal
Sci Technol 6:5806–5811
33. Lee DS et al (2003) Substrate recognition and molecular mecha-
nism of fatty acid hydroxylation by cytochrome P450 from Bacil-
lus subtilis. Crystallographic, spectroscopic, and mutational stud-
ies. J Biol Chem 278(11):9761–9767
21. Zhao B, Summers FA, Mason RP (2012) Photooxidation of
Amplex Red to resorufn: implications of exposing the Amplex
Red assay to light. Free Radic Biol Med 53(5):1080–1087
22. Suzuki R, Hirakawa H, Nagamune T (2014) Electron donation
to an archaeal cytochrome P450 is enhanced by PCNA-mediated
selective complex formation with foreign redox proteins. Biotech-
nol J 9(12):1573–1581
34. Poulos TL (2014) Heme enzyme structure and function. Chem
Rev 114(7):3919–3962
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
23. Koo LS et al (2002) Enhanced electron transfer and lauric acid
hydroxylation by site-directed mutagenesis of CYP119. J Am
Chem Soc 124(20):5684–5691
24. Chaudhury S, Lyskov S, Gray JJ (2010) PyRosetta: a script-based
interface for implementing molecular modeling algorithms using
Rosetta. Bioinformatics 26(5):689–691
1 3