Characterization of a new Baeyer-Villiger monooxygenase and conversion to a solely N-or S-oxidizing enzyme by a single R292 mutation

Article abstract of DOI:10.1016/j.bbapap.2016.06.010

Background Ar-BVMO is a recently discovered Baeyer-Villiger monooxygenase from the genome of Acinetobacter radioresistens S13 closely related to medically relevant ethionamide monooxygenase EtaA (prodrug activator) and capable of inactivating the imipenem antibiotic. Methods The co-substrate preference as well as steady-state and rapid kinetics studies of the recombinant purified protein were carried out using stopped-flow spectroscopy under anaerobic and aerobic conditions. K_d values were measured by isothermal calorimetry. Enzymatic activity was determined by measuring the amount of product formed using high pressure liquid chromatography or gas chromatography. Site-directed mutagenesis experiments were performed to decipher the role of the active site arginine-292. Results Ar-BVMO was found to oxidize ethionamide as well as linear ketones. Mechanistic studies on the wild type enzyme using stopped-flow spectroscopy allowed for the detection of the characteristic oxygenating C4a-(hydro)peroxyflavin intermediate, which decayed rapidly in the presence of the substrate. Replacement of arginine 292 in Ar-BVMO by glycine or alanine resulted in greatly reduced or no Baeyer-Villiger activity, respectively, demonstrating the crucial role of this residue in catalysis of ketone substrates. However, both the R292A and R292G mutants are capable of carrying out N- and S-oxidation reactions. Conclusions Substrate profiling of Ar-BVMO confirms its close relationship to EtaA; ethionamide is one of its substrates. The active site Arginine 292 is required for its Baeyer-Villiger activity but not for heteroatom oxidation. General significance A single mutation converts Ar-BVMO to a unique S- or N-monooxygenase, a useful biocatalyst for the production of oxidized metabolites of human drug metabolizing enzymes.

Full text of DOI:10.1016/j.bbapap.2016.06.010

Biochimica et Biophysica Acta 1864 (2016) 1177–1187
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbapap
Characterization of a new Baeyer-Villiger monooxygenase and
conversion to a solely N-or S-oxidizing enzyme by a single
R292 mutation
Gianluca Catucci a,1, Ivan Zgrablic a,1, Francesco Lanciani a, Francesca Valetti a, Daniela Minerdi a,
David P. Ballou b, Gianfranco Gilardi a, Sheila J. Sadeghi a,
⁎
a
Department of Life Sciences and Systems Biology, University of Torino, 10123 Turin, Italy
b
Department of Biological Chemistry, University of Michigan, Ann Arbor, MI 48109-0600, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 May 2016
Received in revised form 27 May 2016
Accepted 20 June 2016
Available online 23 June 2016
Background: Ar-BVMO is a recently discovered Baeyer-Villiger monooxygenase from the genome of Acinetobacter
radioresistens S13 closely related to medically relevant ethionamide monooxygenase EtaA (prodrug activator)
and capable of inactivating the imipenem antibiotic.
Methods: The co-substrate preference as well as steady-state and rapid kinetics studies of the recombinant puri-
fied protein were carried out using stopped-flow spectroscopy under anaerobic and aerobic conditions. Kd values
were measured by isothermal calorimetry. Enzymatic activity was determined by measuring the amount of prod-
uct formed using high pressure liquid chromatography or gas chromatography. Site-directed mutagenesis exper-
iments were performed to decipher the role of the active site arginine-292.
Results: Ar-BVMO was found to oxidize ethionamide as well as linear ketones. Mechanistic studies on the
wild type enzyme using stopped-flow spectroscopy allowed for the detection of the characteristic oxygenating
C4a-(hydro)peroxyflavin intermediate, which decayed rapidly in the presence of the substrate. Replacement of
arginine 292 in Ar-BVMO by glycine or alanine resulted in greatly reduced or no Baeyer-Villiger activity, respec-
tively, demonstrating the crucial role of this residue in catalysis of ketone substrates. However, both the R292A
and R292G mutants are capable of carrying out N- and S-oxidation reactions.
Keywords:
BVMO
Flavoprotein
Drug metabolites
Stopped-flow
Ethionamide
Isothermal calorimetry
Conclusions: Substrate profiling of Ar-BVMO confirms its close relationship to EtaA; ethionamide is one of its sub-
strates. The active site Arginine 292 is required for its Baeyer-Villiger activity but not for heteroatom oxidation.
General significance: A single mutation converts Ar-BVMO to a unique S- or N-monooxygenase, a useful biocata-
lyst for the production of oxidized metabolites of human drug metabolizing enzymes.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
All BVMO enzymes contain a flavin prosthetic group crucial for
their activity, and they use either NADH or NADPH as the electron
Baeyer-Villiger monooxygenases (BVMOs) are flavin-containing en-
zymes that mediate specific Baeyer-Villiger oxidations on the carbonyl
moiety of their substrates. These proteins have great value in the con-
version of ketones into corresponding esters and lactones, with wide
applications in bioremediation and sustainable biocatalysis. In addition
to Baeyer-Villiger oxidations, BVMO enzymes can also catalyze hetero-
atom oxidations (nitrogen and sulfur) [1] that can have interesting ap-
plications to the pharmaceutical industry, as shown by our recent
results on the oxidation of the antibiotic imipenem [2].
donor. BVMOs are abundant in bacteria, fungi, and plants but their
numbers are limited or zero in animal or human genomes, although
it has recently been discovered that some human flavin-containing
monooxygenases, such as FMO5, are capable of catalyzing a Baeyer-
Villiger oxidation reaction [3].
All BVMO enzymes characterized to date are able to incorporate one
atom from molecular oxygen into the substrate while converting the
other oxygen into water [4]. The catalytic mechanisms are similar to
those previously described in the literature for other flavin-containing
oxygenases. In the first step of the catalytic cycle, the cofactor undergoes
a 2-electron reduction supposedly via hydride ion transfer from
NAD(P)H. The reduced flavin reacts rapidly with molecular oxygen to
form a peroxyflavin that is stabilized by the bound NAD(P). After a suit-
able substrate enters the active site of the enzyme, the flavin peroxide
attacks the electrophilic carbonyl of the substrate, resulting in one
⁎
Corresponding author at: Department of Life Sciences and Systems Biology, University
of Torino, Via Accademia Albertina 13, 10123 Turin, Italy.
E-mail address: sheila.sadeghi@unito.it (S.J. Sadeghi).
These authors have contributed equally to this work.
1
http://dx.doi.org/10.1016/j.bbapap.2016.06.010
1570-9639/© 2016 Elsevier B.V. All rights reserved.

1178
G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
atom of molecular oxygen being transferred to the substrate with the
other forming water. The formation of the reactive peroxyflavin is inde-
pendent of substrate binding, distinguishing BVMO enzymes from other
flavin-containing hydroxylases and cytochromes P450. In the reaction,
the nucleophilic attack of the flavin peroxide on the carbonyl group of
the substrate forms a tetrahedral species that is known as a Criegee in-
termediate, and this rearranges to form an ester that is stereochemically
imposed by the enzyme. The product then leaves the active site
followed by the release of the NAD(P)+ and the cycle can start over
again [4–6].
ethyl acetate, methanol, n-hexane, and salts were purchased from
Sigma–Aldrich (Italy).
2.2. Cloning of Ar-BVMO gene in pT7 expression vector
A. radioresistens S13 was isolated from soil surrounding an activated
sludge pilot plant (Torino, Italy) and its Ar-BVMO gene was previously
isolated by our group from A. radioresistens S13 genome by PCR using
two primers designed on the basis of the almA gene of this species
[13]. The product of this PCR was used as the template for the insertion
of the desired restriction sites upstream and downstream of the gene for
cloning into pT7-7 [17,18].
The BVMO gene was amplified with a nested PCR using primers spe-
cifically designed on the basis of the Ar-BVMO coding gene, adding a 6
histidine tag at the C-terminus for ease of purification. The sequences
of the primers used were: 5′-AACCAACCAAGGATCCATGGATAAACAC
ATTGATGTTCTAATT-3′ and 5′-TTGGTTGGTTGAATTCTTAATGGTGATG
GTGATGGTGTGAAACCAGTTTAGGTTTACGTTC-3′. The PCR conditions
were as follows: denaturation at 94 °C for 3 min; 30 cycles at 94 °C for
30 s, 64 °C for 30 s, and 72 °C for 1 1/2 min; and a final extension at
72 °C for 10 min. A PCR product of the expected size (1541 bp) was ob-
tained. The PCR product was digested with EcoRI and BamHI restriction
enzymes. The linearized insert and pT7-7 vector [17,18], with compati-
ble sticky ends, were used in a ligation reaction to produce the
desired clone. Escherichia coli BL21 (DE3) competent cells were then
transformed with the ligation mixtures and plated on LB agar with
100 μg/mL ampicillin. Colonies with the insert were screened and sub-
sequently the entire pT7-BVMO sequenced.
The first BVMO to be studied in detail was cyclohexanone
monooxygenases (CHMO) from Acinetobacter sp. (NCIMB 9871) [7,8].
Although the genus Acinetobacter has been shown to be a source of sev-
eral BVMO proteins, many of these enzymes are not available in recom-
binant form and therefore have not yet been characterized. Using
sequence homologies and the presence of the conserved Baeyer-
Villiger sequence motif, we have recently discovered a novel BVMO,
Ar-BVMO, in the genome of Acinetobacter radioresistens S13 [9]. This
protein was found to share a 49.7% sequence identity and 71.0% se-
quence similarity with the ethionamide-activating (EtaA) BVMO from
Mycobacterium tuberculosis [10–12]. Furthermore, phylogenetic analysis
of the Ar-BVMO not only placed this protein in the same branch as EtaA
but also showed it to be closely related to the phase-1 drug metabolizing
enzyme, human flavin-containing monooxygenase 3 (FMO3), but only
distantly related to other canonical bacterial BVMO proteins [9,13].
Considering these relationships, initially we studied Ar-BVMO for its
possible activities with “unusual/non-typical” substrates and demon-
strated that it is not only capable of oxidizing two anticancer drugs
metabolized by human FMO3, Danusertib and Tozasertib [14,15], but
it can also oxidize other synthetic drugs such as imipenem [2].
Imipenem is a carbapenem, which is a clinically important antibiotic
family used in the treatment of MultiDrug-Resistant (MDR) bacterial in-
fections. Interestingly, an identical gene to that of Ar-BVMO also exists
in Acinetobacter Baumannii, one of the so-called superbugs [2]. In these
cases, Ar-BVMO seems to be responsible for the de-activation of this an-
tibiotic, and therefore it acts more like a drug metabolizing enzyme.
Chen and coworkers [16] have also shown that a FMO from Methylocella
silvestris is able to oxidize trimethylamine, which is also a typical sub-
strate of eukaryotic FMO enzymes. The reason(s) why prokaryotic flavin
monooxygenases should metabolize substrates that are typically
metabolized by the eukaryotic FMO enzymes but not present in
prokaryotic natural habitats is still not clear.
2.3. Expression of wild-type Ar-BVMO
The expression of the Ar-BVMO enzyme was carried out in a 5 L
Biostat A plus fermenter. Forty milliliters of overnight liquid culture of
E. coli BL21 (DE3) cells transformed with pT7-BVMO were used to inoc-
ulate 4 L LB medium supplemented with ampicillin (100 μg/mL) and ri-
boflavin (20 mg/L). The cells were grown at 37 °C and 200 rpm until
OD600 of 0.5 was reached. At this point, protein expression was induced
by the addition of IPTG and the temperature was lowered to 24 °C. The
cells were harvested 20 h post-induction by 20 min centrifugation at
4000 rpm at 4 °C.
In order to better understand these observations, we have undertak-
en the characterization of the wild type and mutant recombinant Ar-
BVMOs. Kinetic studies of Ar-BVMO, as well as investigations into the
roles of its active site amino acids thought to be important for BV reac-
tions were carried out. Comparisons of these properties with those of
canonical BVMO enzymes were used to try to understand how Ar-
BVMO is competent to carry out oxygenations with synthetic and
medically relevant drugs.
The role of the active site arginine 292 was also investigated in
relation to the enzyme activity towards linear ketones. Two mutants
R292A and R292G have no BV activity while still maintaining their S-
and N-oxygenation capabilities. These monooxygenation activities are
important for human drug metabolism, and therefore, these Ar-BVMO
mutants could be useful as bacterial biocatalysts for large-scale produc-
tion of human drug metabolites.
2.4. Purification of wild-type Ar-BVMO
All purification procedures were carried out at 4 °C. The pellet was
resuspended in 50 mL of Buffer A (50 mM KPi pH 7.4, 20% glycerol,
5 mM β-mercaptoethanol) supplemented with 0.5 mM of the protease
inhibitor phenylmethylsulfonyl fluoride (PMSF) and 1 mg/mL of lyso-
zyme. The resuspended cells were disrupted by sonication. The lysate
was supplemented with 1% Igepal (octylphenoxypolyethoxyethanol),
stirred for 30 min at 4 °C to obtain complete homogeneity, and then cen-
trifuged at 4000 rpm for 45 min. The clarified lysate was loaded onto a
DEAE-Sepharose Fast-Flow column pre-equilibrated with buffer A. The
bound BVMO protein was extensively washed with the same buffer be-
fore it was eluted by applying a gradient of 0–250 mM NaCl also in buff-
er A. Fractions containing the flavoprotein were then pooled and loaded
on a Nickel chelating Sepharose Fast-Flow column pre-equilibrated
with Buffer B (Buffer A supplemented with 250 mM NaCl). The bound
protein was washed with 5 column volumes of Buffer B, 5 column vol-
umes of Buffer B supplemented with 1 mM histidine and 2 column vol-
umes of Buffer B supplemented with 5 mM of histidine. The protein was
finally eluted with Buffer B supplemented with 40 mM histidine.
Fractions presenting the characteristic FAD peaks were pooled and
buffer exchanged with Amicon centrifugal units (30 kDa cut off) with
100 mM potassium phosphate buffer pH 7.4, 20% glycerol.
2. Materials and methods
2.1. Reagents
Riboflavin, NADPH, NADH, ethionamide, 4-phenyl-2-butanone,
phenethyl acetate, 2-octanone, hexyl acetate, 2-decanone, octyl acetate,
2-decanone, methyl undecanoate, decyl acetate, acetonitrile,

G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
1179
2.5. Determination of the extinction coefficient of Ar-BVMO
2.8. Mechanistic studies
Ar-BVMO was purified as a bright yellow protein containing the fla-
vin cofactor. Spectral analysis of the purified recombinant protein
displayed the characteristic flavin spectrum with absorbance maxima
at around 375 and 450 nm. The absorbance ratio 280:450 nm of the pu-
rified protein was around 10.
Following the method described by Sheng et al. [8], the purified Ar-
BVMO was denatured in the presence of 0.1% SDS and the concentration
of free FAD released from the enzyme was determined from its extinc-
tion coefficient of 11,300 M−1 cm−1 at 450 nm [19]. Using this proce-
dure the extinction coefficient of Ar-BVMO was determined to be
13,900 M−1 cm−1 at 450 nm.
The total amount of protein was also estimated using the Bradford
assay and was found to be very similar to the estimated concentration
of protein using the above extinction coefficient, indicating that the pro-
tein is fully occupied with FAD. The purity of the sample was estimated
from a SDS-PAGE gel stained with Coomassie blue.
A Hi-Tech scientific SF-61 (TgK scientific, UK) single mixing
stopped-flow system equipped with a diode array detector was used
for the mechanistic studies. All experiments were performed in
50 mM potassium phosphate buffer pH 7.4, at 15 °C in a glove box
(Belle Technologies, UK) which maintained an O2 concentration below
10 ppm. Anaerobic reduction of the FAD cofactor of the BVMO enzyme
by NADPH/NADH was followed in the stopped-flow apparatus. Each
shot mixed 50 μL of purified enzyme (50 μM) with 50 μL of buffer sup-
plemented with NADPH or NADH (100 μM); thus, samples were diluted
by half in the mixing cell during the kinetic measurements. Normalized
450 nm absorbance values were fitted with SigmaPlot 9 software to
exponential decay equations with 3 parameters (y = y0 + ae−bx
where b represents the measured relaxation rate.
)
Re-oxidation experiments were performed as follows: first, the
BVMO enzyme was fully reduced with NADPH (1 equivalent) or sodium
dithionite in anaerobiosis. The completely reduced enzyme in the
stopped-flow apparatus was then mixed with an equal volume of oxy-
genated buffer or with oxygenated buffer containing 1000 μM of the
substrate benzylacetone. Each shot was followed for 500 s for the co-
substrate reduced enzyme and 9.6 s for the sodium dithionite reduced
enzyme. Normalized 450 nm absorbance values were fitted with
SigmaPlot 9 software using exponential rise to maximum equations
with 3 parameters (y = y0 + a(1 − e−bx)) where b is the measured
rate.
2.6. Product detection: HPLC and GC analysis
All separations were performed using pure analytical standards
of Ethionamide, Ethionamide-SO, 4-phenyl-2-butanone, phenethyl
acetate, 2-octanone, hexyl acetate, 2-decanone, octyl acetate, 2-
dodecanone, methyl undecanoate, decyl acetate. Known concentrations
of analytical standards were also used to quantify the product forma-
tion. HPLC analysis was performed with an Agilent 1200 quaternary
pump HPLC System (Agilent Technologies, Italy) with a C18 reverse
phase column (LiChroCART 250-4 column packed with Lichrosphere
RP C18 5 μM). Previously published methods to separate Ethionamide,
Ethionamide-SO, Tozasertib and Tozasertib NO were used [2,9].
The GC analyses were performed with a 7890A GC System (Agilent
Technologies, Italy) with a DB-Wax column (Length 30 m; Diameter
0.320 mm; PEG film 0.25 μM). The 4-phenyl-2-butanone and phenethyl
acetate were separated following a published protocol [9].
The separation method of 2-octanone and hexyl acetate consisted in
the following steps: Oven start at 50 °C; to 80 °C at 10 °C/min; to 230 °C
at 25 °C/min; to 250 °C at 20 °C/min; 250 °C hold 5 min; the carrier gas
was He (at 3 mL/min). The FID (Flame Ionization Detector) was set at
275 °C. Using this method, the retention times were found to be
5.3 min and 5.1 min for 2-octanone and hexyl acetate, respectively.
The method to separate the 2-decanone and octyl acetate consisted
of the following steps: Oven start at 100 °C; to 250 °C at 10 °C/min. The
carrier gas was He (at 3 mL/min) and the FID (Flame Ionization Detec-
tor) was set to 275 °C. The retention times were 4.5 min and 4.2 min,
respectively.
2.9. Isothermal calorimetry (ITC)
ITC experiments were performed with an ITC-200 calorimeter
(Malvern Instruments, Worcestershire, UK). The volume of the calori-
metric cell was 0.2 mL, and the titrations were conducted by adding
the titrant in steps of 1 μL. Experiments were performed at 25 °C with
a 180 s initial delay. All solutions were thoroughly degassed to prevent
bubble formation in the cell due to stirring. Freshly prepared Ar-BVMO
was dialyzed extensively against degassed buffer containing 50 mM po-
tassium phosphate (pH 7.4 at 4 °C). The protein and NADP concentra-
tions were 8 μM and 800 μM, respectively. Ligand solution was
prepared in the flow-through of the dialyzed protein samples. Protein
and ligand samples were quickly pre-incubated to the required temper-
ature using a ThermoVac (Malvern Instruments) and loaded in the calo-
rimetric cell and titration syringe, respectively. The titration cell was
stirred continuously at 700 rpm. The binding isotherms were best fit
to a one-set binding site model by Marquardt nonlinear least-squares
analysis to obtain the binding stoichiometry (N), association constant
(KA), and thermodynamic parameters of the interaction using ORIGIN
software, Version 7 (Malvern Instruments).
The method to separate the 2-dodecanone, methyl undecanoate and
decyl acetate consisted in the following steps: Oven start at 100 °C; to
250 °C at 10 °C/min. The carrier gas was He (at 3 mL/min) and the FID
(Flame Ionization Detector) was set at 275 °C. The retention times
were 6.5 min, 6.4 min and 6.2 min, respectively.
2.10. H2O2 measurement
The Amplex Red assay (Thermo Fisher Scientific) was used to quan-
tify the H2O2 produced. Incubations contained 100 μM NADPH, 1 mM
Benzylacetone and 10 nM purified Ar-BVMO. Samples and standards
were analyzed in 96 well plates (50 μL/well). A working solution of
100 mM Amplex Red reagent, 0.2 U/mL horseradish peroxidase,
50 mM sodium phosphate pH 7.4 added to each well in a 1:1 volume
and incubated 60 min at 37 °C. The absorbance at 340 nm (NADPH con-
sumption) and 572 nm (H2O2 formation) was measured for 1 h at 30 s
time intervals.
2.7. Steady state parameters
The steady state kinetic parameters for 4-phenyl-2-butanone, 2-
octanone, 2-decanone and 2-dodecanone were determined spectropho-
tometrically by monitoring the NADPH consumption rates (AU/s) at
340 nm with a series of concentrations of substrates. Samples in a
total volume of 0.5 mL contained 0.5 μM enzyme and 150 μM of co-
substrate. Reactions were carried out for 5 min at 30 °C in 50 mM
potassium phosphate buffer, pH 7.4.
2.11. Site-directed mutagenesis
The primers used for the site-directed mutagenesis were:
The steady-state kinetic parameters for ethionamide or tozasertib as
substrate were obtained using HPLC methodology as described above,
with the reactions carried out at 37 °C.
Forward-R292G: 5′GCTATAATCCATGGGATCAAGGCCTTTGTGTAG
TGC-3′, Reverse-R292G: 5′GCACTACACAAAGGCCTTGATCCCATGGATTA
TAGC-3′, Forward-R292A: 5′GCTATAATCCATGGGATCAAGCCCTTTGTGT

1180
G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
AGTGC-3′ and Reverse-R292A: 5′GCACTACACAAAGGGCTTGATCCCAT
GGATTATAGC-3′.
correct, in frame insertion of the gene. Escherichia coli BL21 (DE3) was
chosen as expression system and was transformed with the pT7-
BVMO construct. Preliminary expression was carried out in flasks before
the process was scaled-up to a 5 L fermenter. An expression time course
was carried out to determine the best time for maximal Ar-BVMO pro-
tein expression. A clear overexpressed band at the appropriate molecu-
lar weight of 57 kDa was observed on SDS-PAGE gel starting from 2 h
post-induction with IPTG (data not shown). However, maximal protein
production was found to be at 20 h post-induction.
In order to investigate the cellular localization of the protein, the DAS
server (Dense Alignment Surface method) was used to predict the local
hydrophobicity of the protein, confirming the absence of any trans-
membrane domains (Fig. 1A) [20]. The primary sequence of the BVMO
enzyme was deduced in silico from the ORF of the gene using the Trans-
late tool of ExPASy Proteomics Server [21] and consisted of 502 amino
acids (including the engineered 6 histidine tag at the C-terminus)
with a predicted molecular weight of 57 kDa.
While the E. coli BL21 expression system is optimal for a fast and
high yield expression of proteins, it can also lead to both misfolded pro-
tein in inclusion bodies and enzymes predominantly in the apo-form. To
overcome these problems and to increase both the yield and quality of
the final recovered protein, the post-induction temperature was
lowered to 24 °C. The majority of the recombinant Ar-BVMO protein
was found in the cytosolic fraction but the addition of Igepal (a nonionic,
non-denaturing detergent) increased the final yield of the purified re-
combinant holo-enzyme by ~30%. Despite the in silico prediction of
being cytosolic, Ar-BVMO probably has some local hydrophobic surface
patches that render it membrane-associated, similarly to the EtaA
BVMO of Mycobacterium tuberculosis [11], since in the absence of the
nonionic Igepal from the purification buffers less protein was recovered.
The recombinant Ar-BVMO was purified using two chromatography
columns: first, a DEAE column and second, a Ni-affinity resin that tightly
retained the protein due to engineered His-tag at the C-terminus. The
eluted protein from this latter resin led to nearly 100% pure protein as
deduced from a Coomassie blue stained SDS-PAGE gel that revealed a
single band at approximately 57 kDa, in agreement with the in silico
prediction (Fig. 1B). The final yield of the pure protein was 8 2 mg
per liter of culture and was calculated based on the extinction coeffi-
cient of the protein at 450 nm (13,900 M−1 cm−1) together with the
Bradford protein assay.
PCR reactions were carried out in a final volume of 20 μL containing:
0.25 μM of each primer; 0.4 μL of template (50 ng/μL of pT7-BVMO vec-
tor), 1 mM MgSO4, 0.2 mM dNTPs, 2 μL of 10× KOD Buffer (Merck,
Germany), 0.02 U of KOD Hot Start DNA polymerase (Merck, Germany).
The PCR reactions were performed using a Techne TC-312 thermal cycler
(Bibby Scientific Limited, UK) with the following sequence: denaturation
at 95 °C for 2 min; 30 cycles at 95 °C for 1 min, 55 °C for 1 min and 70 °C
for 2 min and 30 s; this was followed by a final extension at 70 °C for
5 min. PCR products were digested for 2 h at 37 °C with DpnI enzyme
(Fermentas) to eliminate parental DNA. E. coli DH5α competent cells
were subsequently transformed with the DpnI-digested PCR product.
The transformation reactions were plated onto agar plates supplemented
with ampicillin (100 μg/mL). Plates were incubated at 37 °C for 16 h. The
colonies were initially screened for the presence of the BVMO gene and,
the R292G and R292A mutations were confirmed by DNA sequencing.
The two mutants were expressed and purified with the WT-Ar-BVMO
protocols described above.
2.12. Activity assay of the mutants
For all three enzymes (WT Ar-BVMO, R292G, R292A) the reactions
were performed with the same concentrations of each substrate in
50 mM potassium phosphate buffer pH 7.4. For ethionamide, the reac-
tions were performed in a final volume of 200 μL with 0.25 μM enzyme,
500 μM NADPH, and followed for 10 min at 37 °C. For 2-octanone, the
reactions were performed for 30 min at 37 °C in a final volume of
500 μL with 0.2 μM enzyme and 200 μM NADPH. For 4-phenyl-2-
butanone, the reactions were performed for 30 min at 37 °C in a final
volume of 500 μL with 0.1 μM enzyme and 200 μM NADPH.
3. Results and discussion
The newly discovered Ar-BVMO was heterologously expressed, pu-
rified, and characterized with respect to substrate and cofactor prefer-
ences, reaction intermediates and involvement of the active site R292
to the overall catalysis.
3.1. Cloning, expression and purification of recombinant Ar-BVMO
The ORF of the putative protein (1491 bp) was amplified via PCR
using primers with the restriction sites of BamHI and EcoRI. The ampli-
fied construct was cloned into the pT7 expression vector using the same
two restriction sites [17,18] and finally sequenced to confirm the
3.2. Search for Ar-BVMO substrates
Because the sequence of Ar-BVMO is more closely related to that of
EtaA than other known BVMO enzymes [9], known substrates of EtaA
Fig. 1. (A) Hydropathy plot of Ar-BVMO using the DAS-TM program. (B) 10% SDS-PAGE gel of purified Ar-BVMO: Lane 1 and 5 protein markers, Lane 2 crude extract; Lane 3 first
purification step (DEAE); Lane 4 second purification step (Ni-affinity).

G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
1181
from M. tuberculosis [12] were initially considered. Therefore, the EtaA
substrates, 2-octanone, 2-decanone and 2-dodecanone, were selected
and following incubation with purified Ar-BVMO they were found to
be converted into hexyl acetate, octyl acetate and a mixture of methyl
undecanoate (8 1%) and decyl acetate (92 1%), respectively. The
corresponding GC traces of the controls together with the reaction prod-
ucts are shown in Fig. 2.
The steady-state kinetic parameters for the conversion of 2-
octanone and 2-decanone by purified Ar-BVMO were determined spec-
trophotometrically at 340 nm to monitor the NADPH consumption
rates. Apparent Michaelis constants (Km) values of 351 60 μM and
66 11 μM together with turnover rates (kcat) of 1.3 s−1 and 1.6 s−1
were calculated for 2-octanone and 2-decanone, respectively. The
steady-state kinetics data are summarized in Table 1. Incubations
Fig. 2. GC chromatograms of incubations of 2-octanone, 2-decanone and 2-dodecanone in the absence (A, C, E) and presence (B, D, F) of purified Ar-BVMO.

1182
G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
Table 1
Steady-state kinetic parameters of Ar-BVMO determined by rates of NADPH consumption.
K
k
k
/K
m
cat
-1
cat m
Substrate
Reaction
3
-1 -1
M ]
[10
s
[µM]
[s
]
a
Tozasertib
54
8
0.11
1.95
a,b
Ethionamide
768 57
696 322
351 60
66 11
nd
0.3
0.3
1.3
1.6
nd
0.4
0.4
3.7
24
4-Phenyl-2-
Butanonea
2-Octanone
2-Decanone
2-Dodecanone
a
Substrates previously identified [2,14];
b
K
determined by quantifying the product formed using HPLC at different concentrations of substrate.
m
were performed with concentrations of substrate between 0–3 mM. As
can be seen in the Table 1, the best substrate among those tested is 2-
decanone with a catalytic efficiency of 2.4 × 104 M−1 s−1. Steady-
state kinetic parameters of Ar-BVMO were also measured in the absence
of substrate where it displayed a basal NADPH oxidase activity with a
kcat value of 0.09 s−1 and a Km of 14 μM.
The Km values measured for all these substrates were in the micro-
molar range with kcat values comparable to those reported for other pre-
viously studied BVMO enzymes [4,22]. As can be seen in the above table,
all of the turnover rates (kcat) for the tested substrates are quite similar
but the Km values differ. Similar observations have been made in other
BVMOs [4,8], suggesting that the maximal turnover rate does not de-
pend on the reactivity of the flavin intermediate with the substrate
but rather the release of NADP+ in the final step of the catalytic cycle.
As a consequence, the release of NADP+ limits the turnover number.
In order to test whether Ar-BVMO turnover is also limited by the
NADP+ release, the affinity of the enzyme for the oxidized form of
NADP+ was measured using ITC. As expected, Ar-BVMO showed a
high affinity for the oxidized cofactor (Fig. 3) with a calculated Kd
value of 8.3 μM.
Fig. 3. Isothermal calorimetric titration of NADP (800 μM) with purified Ar-BVMO
(12.5 μM) in 50 mM potassium phosphate buffer pH 8 at 25 °C. Heat variation generated
by each injection of titrant at each time interval (top panel) and the integration of each
peak and the amount of heat produced (lower panel).
observed in good yields with NADPH as reductant, suggesting that in
contrast to NADP+, NAD+ does not bind tightly to the reduced enzyme.
3.4. Measurement of the C4a-(hydro)peroxyflavin intermediate
The oxidative half-reaction of Ar-BVMO was studied by stopped-
flow techniques in an anaerobic glove box. The enzyme was reduced
by 1 equivalent of NADPH in one syringe and then mixed with air-
saturated buffer (50 mM potassium phosphate buffer, pH 7.4) from
the second syringe. Fig. 5A shows that the formation of the the C4a-
(hydro)peroxyflavin intermediate at 375 nm requires ~40 s to be com-
plete. This contrasts with CHMO, which forms the C4a-intermediate
within the dead time of the stopped-flow instrument, 1–2 ms [8]. The
intermediate then decays as shown by the increase in absorbance at
450 nm (Fig. 5A) in a time scale longer than 300 s.
3.3. Ar-BVMO reduction with NADH and NADPH
For the Group B flavoprotein monooxygenases [23], including BVMO
enzymes, the reduction of the FAD by the co-substrate occurs in a
substrate independent manner [24,25]. It is also known that class I
BVMOs are specific for NADPH and not NADH as the electron donor
for catalysis [5]. Stopped-flow experiments of the reductive half-
reactions of Ar-BVMO where carried out using each of the two co-
substrates. Although both NADPH and NADH reduced the enzyme,
NADPH was somewhat faster in reducing the FAD group. Reactions
with both co-enzymes are biphasic (Fig. 4) and were fitted to a double
exponential corresponding to a fast and a slow rate. The calculated
fast reduction rates were 12.8 2.0 s−1 and 21.2 4.0 s−1 with calcu-
lated slow rates of 0.15 0.03 and 0.16 0.01 for NADH and NADPH,
respectively. This observed biphasic nature of the reduction rates is
most probably due to some residual oxygen in the system.
If the same re-oxidation experiments are carried out in the presence
of a substrate (benzylacetone), the C4a-(hydro)peroxyflavin intermedi-
ate is formed within 0.5 s (Fig. 5B), and the re-oxidation is much faster,
with conversion to the fully oxidized form being complete within 7 s.
The presence of the substrate accelerates the decay of the C4a-
(hydro)peroxyflavin to the oxidized FAD by ~30-fold. These results
demonstrate that the reoxidation of the enzyme in the absence of an ap-
propriate substrate is slow, and occurs in a time scale of N2 min with an
apparent rate constant of 0.0129 0.0001 s−1 (Fig. 5C) indicating that
the decay of the C4a-(hydro)peroxyflavin is comparable to recent data
reported for another BVMO enzyme, BVMOAf1 from Aspergillus
fumigatus Af293 [26]. Under these conditions, the addition of a saturat-
ing concentration of the substrate (1000 μM benzylacetone) to the re-
duced enzyme increased the re-oxidation rate up to thirty-fold with
This difference is likely due to a more favored binding between
NADPH and BVMO. To confirm if the reduction of Ar-BVMO by each
co-enzyme subsequently led to oxygenation, reactions were performed
with each of the co-enzymes using 4-phenyl-2-butanone as substrate.
The ester product, phenethyl acetate, analyzed by GC, was only
the estimated rate constant of 0.40
0.02 s−1. Therefore, in the

G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
1183
Fig. 4. Ar-BVMO absorbance spectra collected during anaerobic reduction with 2 equivalents of either (A) NADH or (B) NADPH. The experiments were performed in 50 mM potassium
phosphate buffer pH 7.4 at 8 °C over 0.96 s at 10 ms intervals. (C) Kinetic analysis of the reduction of the flavin cofactor. Exponential decay fitting of the NADH (closed circles) and
NADPH mediated reduction (open circles) fitted to a double exponential: fast reduction rates were 21.2 4.0 s−1 and 12.8 2.0 s−1 with calculated slow rates of 0.16 0.01 and
0.15 0.03 for NADPH and NADH, respectively.
presence of the substrate, the absorbance at 450 nm increases very rap-
idly indicating that both formation and decay of the intermediate are ac-
celerated by the addition of a substrate (Fig. 5C). This finding is in line
with other Group B flavin monooxygenases that also show a faster
rate of flavin oxidation in the presence of a suitable substrate [27,28].
Previous structural as well as mutagenesis studies have concluded
that the C4a-(hydro)peroxyflavin intermediate is stabilized by NADP+
which remains bound until the last step of the catalytic cycle [8,29–
32]. However, if this intermediate is not stabilized or the NADP+ bind-
ing is compromised, the uncoupling of the enzyme results in hydrogen
peroxide release. In order to estimate the amount of any H2O2 produced
by Ar-BVMO during the catalytic cycle the AMPLEX RED assay was per-
formed in the presence of a saturating concentration of benzylacetone
(Fig. S1). The results show that the uncoupling of Ar-BVMO, evaluated
as the ratio between H2O2 formed and NADPH consumed, is about 7–
10%, and that the reaction leads to the formation of 3.16 nmol of
H2O2/min/nmol Ar-BVMO. Rauckman et al. [33] found that purified
pig liver FMO (FMO1) produced superoxide anion radical at the rate of
about 4% of the total NADPH oxidized. Tynes et al. [34] found hydrogen
peroxide produced by the purified rabbit lung FMO (FMO2) at up to 41%
of the total NADPH oxidized. More recently Siddens et al. [35] published
data on three human FMOs, where the percentage of molecular oxygen
consumed that appeared as H2O2 varied between 30 and 50% at a rate of
0.5–2.5 nmol/min/nmol FMO. However, all these uncoupling data are
for mammalian FMO enzymes and to our knowledge no such data has
been reported for the bacterial BVMO enzymes.
Finally, the oxidative half-reaction of Ar-BVMO was also monitored
in the absence of pyrnidine nucleotide by using sodium dithionite as
the reductant (Fig. 6A). In this case the re-oxidation is complete within
9.6 s, much faster than that observed with NADPH as the electron donor.
Moreover, no C4a-flavin peroxide is observed, and the re-oxidation rate
is not dependent on the presence of the substrate; re-oxidation rates in
the absence/presence of substrate were: 0.39 0.02 s−1 versus 0.43
0.02 s−1 (Fig. 6B). As mentioned earlier, the latter nearly identical rates
were expected and in line with other BVMO and FMO enzymes where in
normal catalysis the NADP+ remains bound and stabilizes the C4a-
intermediate. Therefore, NADPH plays a dual role both as a reductant
and in its oxidized form, a promoter of oxygenation, whereas dithionite
is solely a flavin reductant.
3.5. Identification of the catalytically determinant R292
In the absence of a crystal structure and in order to investigate the
putative active site of Ar-BVMO, a multi-alignment was performed
with amino acid sequences of different BVMOs including HAPMO
(Pseudomonas fluorescens), PAMO (Thermobifida fusca), CHMO
(Rhodococcus sp.), AlmA (Acinetobacter sp. DSM 17874), EtaA
(M. tuberculosis H37RV) and Ar-BVMO (A. radioresistens S13) as

1184
G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
Fig. 5. (A) Formation and decay of the C4a-(hydro)peroxyflavin within 300 s after mixing the NADPH reduced enzyme with air-saturated buffer (without substrate); (B) Formation and
decay of the C4a-(hydro)peroxyflavin within 7 s after mixing the NADPH reduced enzyme with air-saturated buffer (with 1000 μM benzylacetone). (C) Comparison of the decay of the
C4a-(hydro)peroxyflavin in the presence (red line) and absence of benzylacetone substrate (blue line) followed at 450 nm. The experiment was performed in 50 mM potassium
phosphate buffer pH 7.4 at 15 °C.
shown in Fig. 7. The conserved Baeyer-Villiger sequence motif
FXGXXXHXXXW(P/D) [36] is highlighted by the purple box together
with the known crucial catalytic conserved arginine residue (red
box). This arginine corresponds to R440 in HAPMO, R337 in PAMO
and R329 in CHMO. The marked sequence similarity between Ar-
BVMO, AlmA and EtaA is once again illustrated, as the conserved ar-
ginine is also within a conserved amino acid patch of around 5–10
amino acids (both upstream and downstream). The conserved
aspartyl residue, involved in proper positioning of NADPH in the ac-
tive site of these enzymes [29], is also highlighted in Fig. 7 (D66 in
Fig. 6. A) Spectra of the re-oxidation of Ar-BVMO after complete reduction with sodium dithionite and mixing the reduced enzyme with air-saturated buffer. The experiment was
performed in 50 mM potassium phosphate buffer pH 7.4 at 8 °C in a 9.6 s time scale. B) Kinetic analysis of the re-oxidation of the flavin cofactor. Exponential rise to maximum fitting
of the re-oxidation with oxygenated buffer only (filled circles) and oxygenated buffer with substrate (empty circles). The observed re-oxidation rate constants were 0.39 0.02 s−1
and 0.43 0.02 s−1, respectively.

G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
1185
PAMO, D192 in HAPMO, D58 in CHMO and D56 in AlmA, EtaA and Ar-
BVMO).
yield of approximately 4 mg per liter of culture. These two mutants
exhibited similar absorbance spectra to that of wild-type Ar-BVMO
and were able to oxidize NADPH. The thermal stability of the
mutants was also evaluated and compared to the wild type. The ob-
served melting temperatures for R292G and R292A were calculated
to be 48.9 0.2 °C and 49.1 0.2 °C, respectively. These values are
The important role played by the conserved arginine in different
BVMO enzymes has been previously demonstrated by site-directed mu-
tagenesis. For example, replacing R440 of HAPMO from Pseudomonas
fluorescens by an alanine resulted in an inactive enzyme [37] while
R337 of PAMO from Thermobifida fusca replaced by either alanine or ly-
sine still formed the C4a-peroxyflavin intermediate but the enzyme was
unable to perform either the Baeyer-Villiger reaction on phenylacetone
or the S-oxidation on benzyl methyl sulfide [38]. More recently, a high-
resolution crystal structure of PAMO from T. fusca [29] explained how
the arginine 337 of this enzyme contributes to ketone activation via hy-
drogen bonding to the carbonyl oxygen of the substrate, thus increasing
the electrophilicity of the carbonyl carbon to enhance the nucleophilic
attack by the C4a-peroxyflavin intermediate. Furthermore, the same
researchers also suggested that this positively charged amino acid to-
gether with the NADP ribose moiety, compensates for the negative
charge of the Criegee intermediate, hence stabilizing it [29].
in good agreement with the measured Tm value of 48.6
0.2 °C
for recombinant wild type Ar-BVMO, indicating that the mutants
are correctly folded and equally stable.
The Baeyer-Villiger activities of the two mutants, R292G and R292A,
were compared to that of the wild type Ar-BVMO enzyme using the two
previously identified ketone substrates (4-phenyl-2-Butanone and 2-
octanone) and measuring the products formed by GC. The results ob-
tained are shown in Table 2 where drastically reduced Baeyer-Villiger
activities of both mutants towards the ketone substrates is demonstrat-
ed. In the case of R292A, no product was detected with either 4-phenyl-
2-butanone or 2-octanone, whereas the R292G enzyme showed a small
residual activity of 1.5% and 4.5% with the same two ketones, respective-
ly. As mentioned earlier, the arginine to alanine mutations in HAPMO
and PAMO also resulted in inactive enzymes with no Baeyer-Villiger ac-
tivity [37,38]. The observed lack of Baeyer-Villiger activity of the two
mutants towards the tested ketones confirms the importance of this
arginine residue for catalysis [29].
In order to understand whether this conserved arginine (R292)
also plays an important role in Ar-BVMO, two mutants were de-
signed and generated by site-directed mutagenesis: R292G and
R292A. These mutants were expressed and subsequently purified
with the same protocol as the WT Ar-BVMO, although with a lower
Fig. 7. Multi-alignment of the primary sequences of HAPMO (P. fluorescens); PAMO (T. fusca); CHMO (Rhodococcus sp.); Ar-BVMO (A. radioresistens S13); AlmA (Acinetobacter sp. DSM
17874) and EtaA (M. tuberculosis H37RV). The BVMO motif, the conserved arginine and aspartic acid residues are indicated by purple, red and green boxes, respectively. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

1186
G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
In addition, the S-oxygenation and N-oxygenation activity of the
to assess the N-oxidation activity of the mutants. The R292A mutant
retained N60% of the activity of the wild type enzyme. The data clearly
show that the R292A mutant is not able to perform BV catalysis, but
can selectively perform S- or N-oxidations similar to human flavin-
containing monooxygenases.
Finally, since Ar-BVMO can be expressed in E. coli without inhibiting
the host bacterial growth, scaling up the expression of its mutants in fer-
menters can lead to high yields of these biocatalysts, which can be used
for large-scale production of oxidized human FMO metabolites.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.bbapap.2016.06.010.
mutant enzymes was also compared to that of the purified wild type
Ar-BVMO using ethionamide and tozasertib as substrates, quantifying
the S-oxide or N-oxide product by HPLC analysis. Both substrates are
also known to be metabolized by the human flavin-containing
monooxygenase 3 with the formation of the corresponding S- or N-ox-
ides [13–15]. R292A and R292G mutants showed a residual activity of
83.9% and 43.3% respectively when compared to the wild type Ar-
BVMO for ethionamide and 60% and 5.3% residual activity for Tozasertib,
as shown in Table 2. The fact that replacement of the conserved Arg res-
idue with Ala or Gly exerts an effect on the S- or N-oxygenation reaction
could mean that this residue is involved in more than just stabilization
of the Criegee intermediate. In this specific case, the size of the amino
acid residue seems to have a role in the S- or N-oxygenation activity
since the bulkier Ala residue has nearly double the activity of the smaller
Gly residue (83.9% versus 43.3% residual activity) suggesting a possible
role of Ala in helping the stabilization of the intermediate by excluding
water, which could otherwise break down the C4a-intermediate. The ef-
fect of the positive charge of the Arg residue is less important for the S-
or N-oxygenation reaction, with the activity of the R292A mutant de-
creased by 16% and 40% for N- and S-oxidation, respectively.
Transparency document
The Transparency document associated with this article can be
found, in online version.
References
[1] G. de Gonzalo, M.D. Mihovilovic, M.W. Fraaije, Recent developments in the applica-
tion of Baeyer-Villiger monooxygenases as biocatalysts, Chembiochem 11 (2010)
2208–2231.
[2] D. Minerdi, I. Zgrablic, S. Castrignanò, G. Catucci, C. Medana, M.E. Terlizzi, G.
Gribaudo, G. Gilardi, S.J. Sadeghi, Escherichia coli overexpressing a Baeyer-Villiger
monooxygenase from Acinetobacter radioresistens become resistant to imipenem,
Antimicrob. Agents Chemother. 60 (2016) 64–74.
[3] W.G. Lai, N. Farah, G.A. Moniz, Y.N. Wong, A Baeyer-Villiger oxidation specifically
catalyzed by human flavin-containing monooxygenase 5, Drug Metab. Dispos. 39
(2011) 61–70.
[4] N.M. Kamerbeek, D.B. Janssen, W.J.H. van Berkel, M.W. Fraaije, Baeyer–Villiger
monooxygenases, an emerging family of flavin-dependent biocatalysts, Adv.
Synth. Catal. 345 (2003) 667–678.
In summary, the data presented here confirm that the conserved ar-
ginine, R292, not only plays an important role in the activity of Ar-
BVMO but is also crucial for its Baeyer-Villiger reactions. Moreover,
the ability of the R292 mutants to catalyze S/N-oxidations is almost cer-
tainly associated with the nucleophilic nature of ketone oxygenations,
versus the electrophilic nature of heteroatom oxidations by these
enzymes.
[5] H. Leisch, K. Morley, P.C.K. Lau, Baeyer−Villiger monooxygenases: more than just
green chemistry, Chem. Rev. 111 (2011) 4165–4222.
4. Conclusions
[6] M.D. Mihovilovic, B. Mueller, P. Stanetty, Monooxygenase-mediated Baeyer–Villiger
oxidations, ChemInform 34 (2002) 3711–3730.
[7] C.C. Ryerson, D.P. Ballou, C.T. Walsh, Mechanistic studies on cyclohexanone oxygen-
Baeyer-Villiger monooxygenases are FAD-dependent enzymes with
broad substrate spectra, and in many cases they exhibit high stereospec-
ificity. Similar to other members of the BVMO group of enzymes, Ar-
BVMO shows a tightly but not covalently bound FAD cofactor that can
be reduced anaerobically with NADPH. With NADP+ product remaining
bound, the resulting FADH can react with O2 to form a flavin C4a-
peroxide. This “loaded gun” intermediate of the enzyme can bind any
of a variety of substrates, and the flavin C4a-peroxide can effect
Baeyer-Villiger reactions with carbonyl substrates or oxidations or oxy-
genations of thiols, thioethers, or amines. Ar-BVMO also seems to have
broad substrate tolerance, which is consistent with the kinetic data pre-
sented in this work. The enzyme C4a-(hydro)peroxyflavin intermediate
is relatively stable and does not convert rapidly to oxidized flavin rapid-
ly unless a suitable substrate is present. The initial reduction step by
NADPH is independent of the substrate and not strictly controlled, pre-
sumably because it would be difficult to effect substrate control with the
broad substrate acceptance of this enzyme.
ase, Biochemistry 21 (11) (1982) 2644–2655.
[8] D. Sheng, D.P. Ballou, V. Massey, Mechanistic studies of cyclohexanone
monooxygenases: chemical properties of intermediates involved in catalysis, Bio-
chemistry 40 (2001) 11156–11167.
[9] D. Minerdi, S.J. Sadeghi, G. Di Nardo, F. Rua, S. Castrignanò, P. Allegra, G. Gilardi,
CYP116B5: a new class VII catalytically self-sufficient cytochrome P450 from
Acinetobacter radioresistens that enables growth on alkanes, Mol. Microbiol. 95
(2014) 539–554.
[10] A.E. DeBarber, K. Mdluli, M. Bosman, L.-G. Bekker, C.E. Barry, Ethionamide activation
and sensitivity in multidrug-resistant Mycobacterium tuberculosis, Proc. Natl. Acad.
Sci. 97 (2000) 9677–9682.
[11] T.A. Vannelli, A. Dykman, P.R. Ortiz de Montellano, The antituberculosis drug ethion-
amide is activated by a flavoprotein monooxygenase, J. Biol. Chem. 277 (2002)
12824–12829.
[12] M.W. Fraaije, N.M. Kamerbeek, A.J. Heidekamp, R. Fortin, D.B. Janssen, The prodrug
activator EtaA from mycobacterium tuberculosis is
a
Baeyer–Villiger
monooxygenase, J. Biol. Chem. 279 (2004) 3354–3360.
[13] D. Minerdi, I. Zgrablic, S.J. Sadeghi, G. Gilardi, Identification of a novel Baeyer-Villiger
monooxygenase from Acinetobacter radioresistens: close relationship to the myco-
bacterium tuberculosis prodrug activator EtaA, Microb. Biotechnol. 5 (2012)
700–716.
The role of the active site Arginine 292 residue of Ar-BVMO in catal-
ysis was investigated by engineering two mutants, R292A and R292G.
The R292A variant had no measureable activity and the R292G variant
had very little activity in Baeyer-Villiger catalysis. However, the
monooxygenation activity with the sulfur-containing substrate ethion-
amide by the R292A variant was only slightly less than that of wild
type Ar-BVMO. An aurora kinase inhibitor, Tozasertib, was also tested
[14] G. Catucci, G. Gilardi, L. Jeuken, S.J. Sadeghi, In vitro drug metabolism by C-
terminally truncated human flavin-containing monooxygenase 3, Biochem.
Pharmacol. 83 (2012) 551–558.
[15] G. Catucci, A. Occhipinti, M. Maffei, G. Gilardi, S.J. Sadeghi, Effect of human flavin-
containing monooxygenase 3 polymorphism on the metabolism of aurora kinase in-
hibitors, Int. J. Mol. Sci. 14 (2013) 2707–2716.
[16] Y. Chen, N.A. Patel, A. Crombie, J.H. Scrivens, J.C. Murrell, Bacterial flavin-containing
monooxygenase is trimethylamine monooxygenase, Proc. Natl. Acad. Sci. U. S. A. 108
(2011) 17791–17796.
[17] H. Li, K. Darwish, T.L. Poulos, Characterization of recombinant bacillus meguterium
cytochrome P-450sM-a3nd its two functional domain, J. Biol. Chem. 266 (1991)
11909–11914.
Table 2
[18] F. Valetti, S.J. Sadeghi, Y.T. Meharenna, S.R. Leliveld, G. Gilardi, Engineering multi-
domain redox proteins containing flavodoxin as bio-transformer: preparatory stud-
ies by rational design, Biosens. Bioelectron. 13 (1998) 675–685.
[19] L.G. Whitby, A new method for preparing flavin-adenine dinucleotide, Biochem. J.
54 (1953) 437–442.
[20] M. Cserzo, E. Wallin, I. Simon, G. von Heijne, A. Elofsson, Prediction of transmem-
brane alpha helices in procariotic membrane proteins: the dense alignment surface
method, Protein Eng. 10 (1997) 673–676.
Activity of the two Ar-BVMO mutants with different substrates compared to the wild type
enzyme.
Substrate
BVMO
2-octanone
4-phenyl-2-butanone
Ethionamide
Tozasertib
WT
R292A
R292G
100%
nda
4.6%
100%
nda
1.5%
100%
83.9%
43.3%
100%
60%
[21] P. Artimo, M. Jonnalagedda, K. Arnold, D. Baratin, G. Csardi, E. de Castro, S. Duvaud,
V. Flegel, A. Fortier, E. Gasteiger, A. Grosdidier, C. Hernandez, V. Ioannidis, D.
Kuznetsov, R. Liechti, S. Moretti, K. Mostaguir, N. Redaschi, G. Rossier, I. Xenarios,
5.3%
a
Not detectable.

G. Catucci et al. / Biochimica et Biophysica Acta 1864 (2016) 1177–1187
1187
H. Stockinger, ExPASy: SIB bioinformatics resource portal, Nucleic Acids Res. 40
(W1) (2012) W597–W603.
monooxygenase reveal complex domain movements and a sliding cofactor, J. Am.
Chem. 131 (2009) 8848–8854.
[22] W.A. Duetz, J.B. van Beilen, B. Witholt, Using proteins in their natural environment:
potential and limitations of microbial whole-cell hydroxylations in applied biocatal-
ysis, Curr. Opin. Biotechnol. 12 (2001) 419–425.
[31] B.J. Yachin, T. Sprules, M.B. McEvoy, P.C.K. Lau, A.M. Berghuis, The substrate-bound
crystal structure of a Baeyer-Villiger monooxygenase exhibits a criegee-like confor-
mation, J. Am. Chem. Soc. 134 (2012) 7788–7795.
[23] W.J.H. van Berkel, N.M. Kamerbeek, M.W. Fraaije, Flavoprotein monooxygenases, a
diverse class of oxidative biocatalysts, J. Biotechnol. 124 (2006) 670–689.
[24] N.B. Beaty, D.P. Ballou, The reductive half-reaction of liver microsomal FAD-
containing monooxygenase, J. Biol. Chem. 256 (1981) 4611–4618.
[32] E. Malito, A. Alfieri, M. Fraaije, A. Mattevi, Crystal structure of a Baeyer–Villiger
monooxygenases, Proc. Natl. Acad. Sci. 101 (2004) 13157–13162.
[33] E.J. Rauckman, M.G. Rosen, B.B. Kitchell, Superoxide radical as an intermediate in the
oxidation of hydroxylamines by mixed function amine oxidase, Mol. Pharmacol. 15
(1979) 131–137.
[25] N.B. Beaty, D.P. Ballou, The oxidative half-reaction of liver microsomal FAD-
containing monooxygenase, J. Biol. Chem. 256 (1981) 4619–4625.
[34] R.E. Tynes, P.J. Sabourin, E. Hodgson, R.M. Philpot, Formation of hydrogen peroxide
and N-hydroxylated amines catalyzed by pulmonary flavin-containing
monooxygenases in the presence of primary alkylamines, Arch. Biochem. Biophys.
251 (1986) 654–664.
[26] M.L. Mascotti, M. Kurina-Sanz, M. Juri-Ayub, M.W. Fraaije, Insights in the kinetic
mechanism of the eukaryotic Baeyer-Villiger monooxygenase BVMOAf1 from
Aspergillus fumigatus Af293, Biochimie 104 (2014) 270–276.
[27] X. Dai, K. Mashiguchi, Q. Chen, H. Kasahara, Y. Kamiya, S. Ojha, J. DuBois, D.P. Ballou,
Y. Zhao, The biochemical mechanism of auxin biosynthesis by an Arabidopsis YUCCA
flavin-containing monooxygenase, J. Biol. Chem. 288 (2013) 1448–1457.
[28] J.A. Mayfield, R.E. Frederick, B.R. Streit, T.A. Wencewicz, D.P. Ballou, J.L. DuBois,
Comprehensive spectroscopic, steady state, and transient kinetic studies of a
representative siderophore-associated flavin monooxygenase, J. Biol. Chem. 285
(2010) 30375–30388.
[29] R. Orru, H.M. Dudek, C. Martinoli, D.E. Torres Pazmiño, A. Royant, M. Weik, M.W.
Fraaije, A. Mattevi, Snapshots of enzymatic Baeyer–Villiger catalysis: oxygen activa-
tion and intermediate stabilization, J. Biol. Chem. 286 (2011) 29284–29291.
[30] I. Ahmad Mirza, J.B. Yachnin, S. Wang, S. Grosse, H. Bergeron, A. Imura, H. Iwaki, Y.
Hasegawa, P.C.K. Lau, A.M. Berghuis, Crystal structures of cyclohexanone
[35] L.K. Siddens, S.K. Krueger, M.C. Henderson, D.E. Williams, Mammalian flavin-
containing monooxygenase (FMO) as a source of hydrogen peroxide, Biochem.
Pharmacol. 89 (2014) 141–147.
[36] M.W. Fraaije, N.M. Kamerbeek, W.J.H. van Berkel, D.B. Janssen, Identification of a
Baeyer–Villiger monooxygenase sequence motif, FEBS Lett. 518 (2002) 43–47.
[37] R.H.H. van den Heuvel, N. Tahallah, N.M. Kamerbeek, M.W. Fraaije, J. Willem, H. van
Berkel, D.B. Janssen, A.J.R. Heck, Coenzyme binding during catalysis is beneficial for
the stability of 4-hydroxyacetophenone monooxygenase, J. Biol. Chem. 280 (2005)
32115–32121.
[38] D.E. Torres Pazmiño, B.-J. Baas, D.B. Janssen, M.W. Fraaije, Kinetic mechanism of
phenylacetone monooxygenase from Thermobifida fusca, Biochemistry 47 (2008)
4082–4093.