Human flavin-containing monooxygenase 3 on graphene oxide for drug metabolism screening

Article abstract of DOI:10.1021/ac504535y

Human flavin-containing monooxygenase 3 (hFMO3), a membrane-bound hepatic protein, belonging to the second most important class of phase-1 drug-metabolizing enzymes, was immobilized in its active form on graphene oxide (GO) for enhanced electrochemical response. To improve protein stabilization and to ensure the electrocatalytic activity of the immobilized enzyme, didodecyldimethylammonium bromide (DDAB) was used to mimic lipid layers of biological membranes and acted as an interface between GO nanomaterial and the hFMO3 biocomponent. Grazing angle attenuated total reflectance Fourier transform infrared (GATR-FT-IR) experiments confirmed the preservation of the protein secondary structure and fold. Electrochemical characterization of the immobilized enzyme with GO and DDAB on glassy carbon electrodes was carried out by cyclic voltammetry, where several parameters including redox potential, electron transfer rate, and surface coverage were determined. This systems biotechnological application in drug screening was successfully demonstrated by the N-oxidation of two therapeutic drugs, benzydamine (nonsteroidal anti-inflammatory) and tamoxifen (antiestrogenic widely used in breast cancer therapy and chemoprevention), by the immobilized enzyme.

Full text of DOI:10.1021/ac504535y

Article
pubs.acs.org/ac
Human Flavin-Containing Monooxygenase 3 on Graphene Oxide for
Drug Metabolism Screening
Silvia Castrignano,† Gianfranco Gilardi,†,‡ and Sheila J. Sadeghi*,†,‡
̀
†Department of Life Sciences and Systems Biology, University of Torino, via Accademia Albertina 13, 10123 Torino, Italy
‡Centre for Nanostructured Interfaces and Surfaces, University of Torino, via Pietro Giuria 7, 10125 Torino, Italy
S
* Supporting Information
ABSTRACT: Human flavin-containing monooxygenase 3
(hFMO3), a membrane-bound hepatic protein, belonging to
the second most important class of phase-1 drug-metabolizing
enzymes, was immobilized in its active form on graphene oxide
(GO) for enhanced electrochemical response. To improve
protein stabilization and to ensure the electrocatalytic activity of
the immobilized enzyme, didodecyldimethylammonium bro-
mide (DDAB) was used to mimic lipid layers of biological
membranes and acted as an interface between GO nanomaterial
and the hFMO3 biocomponent. Grazing angle attenuated total
reflectance Fourier transform infrared (GATR-FT-IR) experi-
ments confirmed the preservation of the protein secondary
structure and fold. Electrochemical characterization of the
immobilized enzyme with GO and DDAB on glassy carbon
electrodes was carried out by cyclic voltammetry, where several parameters including redox potential, electron transfer rate, and
surface coverage were determined. This system’s biotechnological application in drug screening was successfully demonstrated by
the N-oxidation of two therapeutic drugs, benzydamine (nonsteroidal anti-inflammatory) and tamoxifen (antiestrogenic widely
used in breast cancer therapy and chemoprevention), by the immobilized enzyme.
chemistry in terms of its catalytic activity toward several drugs
mong the five different functional human flavin-containing
Amonooxygenases (hFMOs) belonging to the mammalian including benzydamine, sulindac sulfide, tamoxifen, and aurora
kinase inhibitors.9−12
FMO family, hFMO3 is described as the most important
isoform present in the adult liver,1 having a prominent role in
In this work we developed a hFMO3 bioelectrode by
the metabolism of drugs and chemicals.2 FMO enzymes are
immobilizing hFMO3 protein in a nanoscaled system, based on
graphene oxide (GO) in the presence of DDAB on glassy
carbon electrodes. Owing to its attractive properties, graphene
has been used as a transducer to develop a number of different
biosensing methodologies: biofield-effect transistors, electro-
chemical biosensors, impedance biosensors, electrochemilumi-
nescence, and fluorescence biosensors, as well as biomolecular
labels.13−15
Among graphene derivatives, GO plays a major role in the
development of carbon nanomaterials. GO can be produced by
complete exfoliation of graphite oxide, obtaining a single
graphene sheet widely functionalized.16Unlike graphene, whose
structure is only composed of sp2 hybridized carbon atoms, GO
has a carbon structure that is interspersed by a range of oxygen-
containing functional groups, giving rise to some degree of sp3
hybridization. Experimental data support the model proposed
by Lerf and colleagues17,18 in which the basal plane of the
carbon sheet binds hydroxy- and epoxy-functional groups.
microsomal, flavin adenine dinucleotide-containing, nicotina-
mide adenine dinucleotide phosphate-dependent,1−4 and able
to use molecular oxygen to catalyze the oxygenation of a variety
of structurally different xenobiotics, including many therapeutic
drugs.5,6 All soft nucleophilic molecules possessing highly
polarizable electron lone pairs on their heteroatoms are putative
substrates for these enzymes.1,3,6 hFMO3 catalyzes the selective
oxygenation of the heteroatom by incorporation of one atom of
molecular oxygen.4,6 These reactions are usually considered to
be part of the phase-1 drug metabolism detoxification process
whereby compounds are transformed into highly polar and
excretable molecules.7 Unlike cytochromes P450, the hFMO3
activity produces few toxic metabolites and therefore could be
used to control drug clearance with significant clinical
advantage.2 Therefore, the design of drugs that are metabolized
by hFMO3 rather than by cytochromes P450 is considered
extremely advantageous.2 Moreover, the hFMO3 is not readily
induced or inhibited, like cytochromes P450, consequently
minimizing potential adverse drug−drug interactions.8 Pre-
viously, we have described the engineering of a solubilized form
of hFMO3 that has been characterized by direct electro-
Received: December 6, 2014
Accepted: January 29, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry
Article
Carbonyl groups are mainly present as carboxylic groups at the
edge of the sheet.19 The abundance of oxygen functional
groups are advantageous for GO electrochemical applica-
tions20,21 and are expected to promote the electron transfer
between electrode substrates and enzyme molecules for the
study of their direct electrochemistry.22 Oxygen-containing
groups also allow GO to be suspended in water and make it
feasible for use in a variety of chemical reactions and
compatible with biological applications not available to
graphene, because it is insoluble in water and needs to be
supported on a substrate.23,24
A key issue in the development of GO-based bioanalytical
systems is the stabilization of the biomolecule together with the
preservation of its native structure and activity. The latter
becomes crucial when dealing with enzyme-based electro-
chemical systems in which the catalytic activity is strongly
dependent on the correct fold of the enzyme. In this regard, the
use of a stabilizing component that could support both the
preservation of enzymatic structure and activity, and the
electrochemical interaction between the biocomponent and the
GO based transducer could be greatly advantageous. To this
end, surfactant films of synthetic lipid like didodecyldimethy-
lammonium bromide (DDAB) provide a biomembrane-like
microenvironment containing enough water for supporting
structure and activity of proteins on electrode surfaces by acting
as stable lyotropic liquid crystal coats.25
drop of DDAB chloroform solution and of GO water
dispersion onto an amorphous carbon film supported on a
copper mesh grid and drying in air.
Fourier Transform Infrared Spectroscopy. Infrared
spectra of hFMO3, in both the presence and absence of GO,
were acquired using the grazing angle attenuated total
reflectance (GATR) tool. Human FMO3−DDAB samples
were prepared on gold−polyethylene terephthalate (PET) flat-
surface substrates following the same procedure described for
glassy carbon electrodes and compared with hFMO3 samples
prepared by gently mixing equal volumes of hFMO3 solution
and buffer (50 mM phosphate buffer pH 7.4 with 20%
glycerol). Before the Fourier transform infrared (FT-IR)
analysis, samples were kept overnight at 4 °C. All spectra
were collected from 4000 to 800 cm−1 using a Bruker model
Tensor 27 FT-IR spectrometer (Bruker Instruments, Billerica,
MA) with a scan velocity of 10 kHz and a resolution of 4 cm−1.
During data acquisition, the spectrometer was continuously
purged with nitrogen at room temperature. Data were collected
in triplicate, and spectra were averaged using the Opus software
(Bruker Instruments, Billerica, MA). Spectra of the protein
were corrected by subtraction of the control samples acquired
under the same scanning and temperature conditions. In
particular, IR spectra of buffer, DDAB, and DDAB−GO were
used with the same dilution as background for hFMO3,
hFMO3−DDAB, and hFMO3−DDAB−GO samples, respec-
tively. Information on the number and location of components
for the amide I band was provided by the Fourier self-
deconvolution conducted on the average spectra, using a
deconvolution factor of 50 and a noise reduction factor of 0.8.
Subsequently, curve fitting was performed using PeakFit
software (SPSS Inc., U.S.A.).
Electrode Preparation. Glassy carbon electrodes were
modified with 10 μL of 20 mM DDAB chloroform solution or
DDAB plus 5 μL of GO water dispersion and then left at room
temperature for 10 min to allow solvent evaporation. Five μL of
purified hFMO3 solution (100 μM) or free FAD solution were
added, and the modified glassy carbon electrodes were kept at 4
°C for 2 h before any further experimental procedure.
Cyclic Voltammetry and Chronoamperometry. All
electrochemical experiments were carried out at room temper-
ature (25 °C) and in 50 mM phosphate buffer pH 7.4,
containing 100 mM KCl as supporting electrolyte, using an
Autolab PGSTAT12 potentiostat (Ecochemie, The Nether-
lands) controlled by GPES3 software. A conventional three-
electrode glass cell, equipped with a platinum wire counter
electrode, an Ag/AgCl (3 M NaCl) reference electrode, and 3
mm diameter glassy carbon working electrode (BASi, U.S.A.),
was also used.
Here we report the development of a GO-based nano-
structured electrode system exploitable for the pharmacological
screening of novel hFMO3 metabolized drugs by immobilizing
this enzyme on GO modified glassy carbon in the presence of
DDAB as a biomembrane-like surfactant. To our knowledge
this is the first report of the use of GO with this class of human
enzymes.
EXPERIMENTAL SECTION
■
Chemicals. GO (4 mg/mL, water dispersion) was
purchased from Graphenea (Spain). Analytical-grade chemicals
were used with no further purification. All solutions were
prepared with ultrapure deionized water. DDAB, benzydamine
(hydrochloride), benzydamine N-oxide (hydrogen maleate),
and tamoxifen were purchased from Sigma-Aldrich, and
tamoxifen N-oxide was purchased from Biozol (Germany).
Their solutions were prepared immediately before use by
dissolving the adequate amount in the appropriate solvent.
Expression and Purification of Wild-Type Human
FMO3. Wild-type hFMO3 protein was expressed in Escherichia
coli (JM109) and purified from the membrane fractions via
nickel affinity chromatography, following the procedure
described by Catucci and co-workers.26 After the purification,
the protein was visualized in a 10% sodium dodecyl sulfate
(SDS)−polyacrylamide gel and stained with Coomassie Blue to
verify its purity. The hFMO3 protein concentration was
calculated assuming a molecular mass of 56 kDa, a molar
content equal to that of flavin adenine dinucleotide (FAD), and
an extinction coefficient of 11 300 M−1 cm−1 at 450 nm.27 The
activity of the solubilized enzyme was measured by its N-
oxidation of benzydamine where a KM of 22 μM was measured,
which is similar to previously published values.28
Electrochemical investigation of hFMO3 properties, both in
the presence and in the absence of GO was performed by cyclic
voltammetry in a nitrogen atmosphere within a glovebox (Belle
Technologies, U.K.). Cyclic voltammograms were recorded
between 0 and −750 mV at increasing scan rates.
Electrochemically driven substrate oxygenation by the
hFMO3−DDAB−GO was performed using chronoamperom-
etry with an applied potential bias of −650 mV for 15 min. To
allow substrate permeation into the enzymatic layer and
minimize mass transport influence at the transducer surface,
hFMO3 was immobilized through DDAB−GO on glassy
carbon rotating disk electrodes. All electrocatalysis experiments
were performed using a BASi RDE-2 rotator system (BASi,
U.S.A.) at 200 rpm rotation speed. Chronoamperometric
Transmission Electron Microscopy. High-resolution
images of GO in the presence of DDAB were collected by
transmission electron microscopy (TEM) (JEOL-3010 UHR,
Jeol Ltd., Japan, operating at 300 keV) at room temperature.
Specimens for TEM observation were prepared by casting one
B
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry
Article
procedure was applied on freshly prepared hFMO3−DDAB−
GO electrodes in the presence of increasing concentrations of
benzydamine or tamoxifen, and the product(s) obtained after
the 15 min reaction was immediately injected for high-
performance liquid chromatography (HPLC) for separation
and quantification. Product quantification was achieved using
calibration curves obtained both for benzydamine N-oxide and
for tamoxifen N-oxide, injecting standard solutions of known
concentrations.
molecules should be able to directly interface with the GO
layer.
The preservation of the hFMO3 3D fold in the presence of
DDAB and GO was investigated by FT-IR spectroscopy
(Figure 2). As can be seen, the collected spectra are effectively
High-Performance Liquid Chromatography. The quan-
tification of electrocatalysis product formed was performed by
HPLC coupled with a diode array UV detector (Agilent-1200,
Agilent Technologies, U.S.A.) equipped with a 4.6 × 150 mm, 5
μm Eclipse XDB-C18 column, as previously reported.11
Benzydamine and benzydamine N-oxide were separated using
a linear gradient elution programmed as follows: linear gradient
elution from 20% to 40% acetonitrile and from 80% to 60% 50
mM KH2PO4, 0−10 min; isocratic elution of 40% acetonitrile
and 60% 50 mM KH2PO4, 10−12 min; linear gradient elution
from 40% to 60% acetonitrile and from 60% to 40% 50 mM
KH2PO4, 12 min−end of the run. The flow rate was set at 1
mL/min, and the detection wavelength was 308 nm. Retention
times were 17.5 and 20 min for benzydamine N-oxide and
benzydamine, respectively. Tamoxifen and tamoxifen N-oxide
were separated using an isocratic elution of 82% methanol and
18% triethylamine (1%) and a flow rate of 0.8 mL/min. The
detection wavelength was set at 278 nm. Retention times were
7 and 15 min for tamoxifen N-oxide and tamoxifen,
respectively.
Figure 2. FT-IR spectra of DDAB-stabilized hFMO3 protein in the
absence and presence of GO compared to that of only hFMO3 protein
solution.
similar to amide I, II, and III bands, which are characteristics of
protein IR spectra. Amide I band (∼1650 cm−1) is related to
the stretching vibration of the peptidic carbonyl group with a
minor contribution from the out-of-phase C−N stretching
vibration.32 This band results from the overlapping of a group
of signals containing information on the secondary structure
composition of the enzyme. Bands centered at ∼1545 cm−1 can
be assigned to the amide II band, which is due to the out-of-
phase combination of the peptidic N−H in-plane bend and the
C−N stretching vibration. The peak at 1517 cm−1 is due to the
C−C stretch and C−H bend of side-chain tyrosine residues.32
In addition, a set of bands can be observed in the region
between 1200 and 1400 cm−1 that can be assigned to the amide
III mode, associated with the in-phase combination of the
peptidic N−H with C−N bending vibration.32 In general the
infrared spectral features of hFMO3 protein do not seem to be
affected by the presence of GO with the enzyme structural
integrity and fold preserved.
RESULTS AND DISCUSSION
■
TEM images of GO in the presence of DDAB were collected to
investigate the morphological features of GO layers distribution
(Figure 1). As can be seen in the figure, GO is well-distributed
in a two-dimensional manner as a foil-like material, with
exfoliated sheets characterized by wrinkling and waviness, in
agreement with previously published experimental data.29−31
TEM results indicate that GO sheets are distributed throughout
the electrode surface and that DDAB-stabilized protein
A deeper investigation of the secondary structure composi-
tion of hFMO3 protein was performed by fitting the amide I
band with component bands after identification of their
position through the deconvolution of the amide I peak
(Figure 3A and B). Amide I component bands were then
assigned to the corresponding protein secondary structure,32,33
and their contribution to the secondary structure composition
of hFMO3 enzyme was calculated as percentages for the
protein in the presence and in the absence of GO.
The secondary structure composition of hFMO3 is
compared in Table 1 to the one measured in the presence of
DDAB and DDAB−GO. As seen in the table, results obtained
are highly comparable in terms of both peak positions and
percentage contributions. Moreover, these results are also in
agreement with previously published data on the hFMO3
secondary structure calculation by far-UV circular dichroism
and FT-IR.11,26
Cyclic voltammetry of hFMO3 was carried out both on
DDAB and on DDAB−GO glassy carbon electrodes. The cyclic
Figure 1. TEM images of GO in the presence of DDAB taken from
four different regions of the modified grid.
C
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry
Article
Figure 4. Anaerobic cyclic voltammograms of hFMO3 on DDAB
glassy carbon electrodes in the absence (dashed line) and in the
presence (solid line) of GO. Scan rate: 120 mV s−1. Inset: plot of
cathodic (filled symbols) and anodic (empty symbols) peak currents
versus scan rate for hFMO3 on DDAB glassy carbon electrodes in the
absence (triangles) and in the presence (circles) of GO. R2 > 0.99.
Table 2. Redox Properties of hFMO3 in the Presence and
a
Absence of GO Compared to Those of Free FAD
E1/2/mV
ΔE/mV
Γ/pmol cm−2
GC/hFMO3−
−363.2 1.0
61.0 1.7
19.5 2.7
DDAB
(
GC/hFMO3−
−354.7 0.3*
−353.7 3.5#
−344.3 4.0*,#
64.0 1.0
50.7 9.6
49.3 1.5
32.1 1.9 **)
DDAB−GO
(#)
(#)
(#)
GC/FAD−
27.6 0.2
Figure 3. Amide I band with individual components fitted for hFMO3
in the absence (A) and in the presence (B) of GO.
DDAB
(
GC/FAD−
DDAB−GO
64.9 6.9 ***,##)
a
voltammograms shown in Figure 4 demonstrate the electro-
chemical reversibility of immobilized hFMO3 both in the
presence and in the absence of GO. No voltammetric peaks
were observed in the absence of the enzyme (data not shown).
Redox peak potential values of hFMO3−DDAB and hFMO3−
DDAB−GO electrodes are summarized and compared to those
of free FAD in Table 2. Interestingly, it was found that the
presence of GO shifts the redox potential, calculated as the
midpoint potential (E1/2), toward more positive values both for
hFMO3 bound and free FAD. As expected, the redox potential
of the protein-bound FAD is shifted toward more negative
values compared to free FAD, both in the presence and in the
absence of GO. No significant differences were found when
comparing peak-to-peak separation values due to GO. The
presence of peak current values were found to be linearly
dependent on the scan rate up to 120 mV (Figure 4, inset). As
*P < 0.05, **P < 0.01, and ***P < 0.001 when comparing data
obtained in the absence of GO. #P < 0.05 and ##P < 0.001 when
comparing data obtained with hFMO3 protein.
stated by Laviron’s theory,34 this property is characteristic of
thin-film confined electroactive species that are not under
diffusion control. A significant increase of the redox peak
current values was also observed due to the presence of GO (P
< 0.001). Apparent surface coverage (Γ) of the electroactive
hFMO3 bound and free FAD was calculated from the peak
current plot versus scan rate by applying Faraday’s law.
Apparent surface coverage was increased by GO presence both
in the case of protein bound and of free FAD. The observed
increases were statistically significant (statistical analysis was
performed by one-way ANOVA followed by Student−New-
man−Keuls post hoc test).
Table 1. Calculated Secondary Structure Content of Immobilized hFMO3 from the Deconvolution of the Amide I Band of FT-
a
IR Spectra
hFMO3
hFMO3−DDAB
peak position
hFMO3−DDAB−GO
peak position
structure
peak position
area
area
area
α helix
1658 cm−1
1666 cm−1
1630 cm−1
1674 cm−1
1685 cm−1
1648 cm−1
34.9%
1658 cm−1
1668 cm−1
1629 cm−1
1677 cm−1
1688 cm−1
1647 cm−1
36.7%
1657 cm−1
1666 cm−1
1629 cm−1
1676 cm−1
1688 cm−1
1647 cm−1
34.5%
β sheet
20.9%
21.5%
20.2%
20.4%
20.7%
23.6%
Turn
unordered
22.7%
22.7%
21.3%
a
Fit quality (R2) between original and fitted spectra ≥0.99.
D
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry
Article
The electron transfer rate constant (ks) was also calculated
from empirically fitted Laviron’s equations.34 To this end,
irreversible electrochemistry of hFMO3 protein was studied in
cyclic voltammetry by increasing the scan rate from 10 to 22 V
s−1, and cathodic and anodic peak potentials were plotted
versus the logarithm of the scan rate. The ks values were
calculated from the intercept of the E plot versus the natural
logarithm of scan rate and were determined to be 38.6 2.3 s−1
for hFMO3−DDAB and 48.1 3.0 s−1 for hFMO3−DDAB−
GO electrodes.
Direct electrochemical investigation of hFMO3 on DDAB−
GO glassy carbon electrodes revealed an improvement of the
electrochemical performance of the enzyme due to the
presence of GO in terms of electron transfer rate. Moreover,
GO increases the redox potential of both hFMO3 bound and
free FAD, confirming that FAD acquires a higher tendency to
be reduced in DDAB−GO glassy carbon electrodes. This is
highly advantageous for the electrocatalysis performance of
hFMO3 protein because a protein that is more prone to
reduction is likely more catalytically efficient. GO also enhances
peak currents of hFMO3 enzyme, leading to an improvement
of the signal-to-noise ratio. Moreover, a higher apparent surface
coverage value was also observed for hFMO3−DDAB−GO
electrodes, proving that in the presence of GO the amount of
electroactive enzyme molecules on the electrode surface are
increased.
Catalytic activity of hFMO3 on GO electrodes was examined
using benzydamine and tamoxifen by chronoamperometry on
glassy carbon rotating disk electrodes. Benzydamine is a
nonsteroidal anti-inflammatory drug that is metabolized to its
N-oxide by hFMO3.35
Benzydamine is a nonsteroidal anti-inflammatory drug that
has been mainly used as a marker substrate of hFMO3, and its
N-oxidation is usually referred to as an indicator of hFMO3
catalytic activity. Tamoxifen is an antiestrogenic drug widely
used in breast cancer therapy and chemoprevention that is also
N-oxygenated by hFMO3 to its N-oxide.36,37 The calculation of
apparent kinetic parameters for the N-oxidation of these two
drugs was performed by electrocatalysis followed by HPLC
separation of the products. Reaction rate values were estimated
after the quantification of the product formed during the
electrocatalysis experiment and corrected by subtraction of the
relevant controls (Figure 5A and B).
Figure 5. Bioelectrocatalytically produced benzydamine N-oxide (A)
and tamoxifen N-oxide (B) by hFMO3−GO−DDAB glassy carbon
electrodes: plot of reaction velocity versus substrate concentration
fitted to the Michaelis−Menten equation. Data are shown as mean
SD of three different electrode measurements.
value of 16.4 min−1 is lower than the only other reported value
of 61 min−1.38 In this case, it was more difficult to compare our
values with published results due to the limited availability of
data regarding kinetics of hFMO3 turnover in the presence of
tamoxifen, especially regarding recombinant systems. Further-
more, it must be taken into account that in our method
reducing equivalents are provided by the electrode transducer
and not NADPH and that the enzyme velocity is directly
measured by estimating product formation, whereas the only
available data about hFMO3 turnover in the presence of
tamoxifen is based on the NADPH consumption assay (which
provides an indirect measurement of enzyme catalysis).
When comparing benzydamine electrocatalysis results for
hFMO3 in the presence of GO with our previously published
data about hFMO3 electrocatalysis, a significant improvement
of the enzyme catalytic efficiency, expressed as apparent
turnover number, was observed. Finally, the reproducibility of
the results obtained from different electrodes expressed as
residual standard deviation was calculated to be 15%.
Control experiments were performed on DDAB−GO
rotating disk electrodes using denatured hFMO3 protein
(treated with 8 M guanidinium chloride for 30 min). Apparent
Michaelis−Menten constant (K′M) and turnover number (k′cat)
values were calculated for benzydamine and tamoxifen N-
oxygenation by fitting velocity plot versus substrate concen-
tration with Michaelis−Menten-type curves. Calculated K′M
and k′cat parameters for benzydamine electrocatalysis were 52.9
6.4 μM and 62.3
1.6 min−1, respectively, whereas for
tamoxifen electrocatalysis K′M and k′cat values were estimated to
be 5.8 1.0 μM and 16.4 0.6 min−1, respectively.
Apparent kinetic values obtained for hFMO3 in the presence
of GO were compared with previously published data for both
benzydamine11,26,28,38 and tamoxifen36,37and are reported in
Table 3. Regarding benzydamine catalysis, our results are
generally in good agreement with previously published data,
especially the calculated K′M values (∼50−60 μM). For
tamoxifen, our electrochemically determined K′M value of 5.8
μM is much lower than the published values (121 and 1430
μM), which shows a better affinity. However, the calculated k′cat
CONCLUSIONS
■
The biocompatibility of inorganic nanomaterials is a crucial
issue for the successful development of protein-based high-
throughput electrochemical assay systems. Because of its
interesting electrochemical and mechanical properties, GO
can be considered as an attractive nanomaterial and has been
E
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry
Article
Table 3. Comparison of Apparent Kinetic Parameters for Benzydamine and Tamoxifen N-Oxidation with Previously Published
Results
expression system
enzymatic activity assay
KM (μM)
kcat (min‑1)
ref
benzydamine
E. coli
electrode immobilization
electrode immobilization
solution
58.9 9.1
52.1 9.5
63.2 2.2
present work
E. coli
0.47 0.02
̀
Castrignano et al., 2012
E. coli
56
51 19
80
8
9
2
Catucci et al., 2012
Shimizu et al., 2007
Lang and Rettie, 2000
present work
E. coli
solution
300 27
36
baculovirus
E. coli
solution
8
2
tamoxifen
electrode immobilization
NADPH consumption
NADPH consumption
5.8 1.1
121 25
1430
16.4 0.6
61 20
n.d.
baculovirus
E. coli
Krueger et al., 2006
Hodgson et al., 2000
used as a high-performance electrochemical transducer to
develop a number of bioanalytical paradigms. However, enzyme
stabilization is essential in order to optimize electrocatalytic
performance and electrochemical tuning between the nano-
structured transducer and the protein biocomponent.
REFERENCES
■
(1) Cashman, J. R. Drug Discovery Today 2004, 9, 574−581.
(2) Cashman, J. R. Expert Opin. Drug Metab. Toxicol. 2008, 4, 1507−
1521.
(3) Cashman, J. R. Chem. Res. Toxicol. 1995, 8, 166−181.
(4) Krueger, S. K.; Williams, D. E. Pharmacol. Ther. 2005, 106, 357−
387.
(5) Phillips, I. R.; Dolphin, C. T.; Clair, P.; Hadley, M. R.; Hutt, A. J.;
McCombie, R. R.; Smith, R. L.; Shephard, E. A. Chem.-Biol. Interact.
1995, 96, 17−32.
(6) Cashman, J. R. Drug Metab. Rev. 2002, 34, 513−521.
(7) Cashman, J. R. Curr. Drug Metab. 2000, 1, 181−191.
(8) Cashman, J. R. Biochem. Biophys. Res. Commun. 2005, 338, 599−
604.
In this work, the unique features of GO have been applied to
hFMO3 electrochemical properties using the stabilizing
capability of the synthetic lipid DDAB to develop a sensitive,
high-performing bioelectrochemical assay for the screening of
drugs metabolized by hFMO3. The direct electrochemical
response of hFMO3 was enhanced by the presence of GO in
terms of electron transfer rate, current signal-to-noise ratio, and
apparent surface coverage. Moreover, catalytic efficiency of the
enzyme in the presence of GO, determined by the kinetic
parameters K′M and k′cat for benzydamine and tamoxifen, were
found to be comparable with previously published data. Both
FTIR and electrochemical characterization of hFMO3−
DDAB−GO glassy carbon electrodes were found to be
consistent with the preservation of the protein structure and
fold. Moreover, compared to our previously obtained electro-
catalysis results, GO improves the electrocatalytic efficiency of
hFMO3 enzyme.
(9) Sadeghi, S. J.; Meirinhos, R.; Catucci, G.; Dodhia, V. R.; Di
Nardo, G.; Gilardi, G. J. Am. Chem. Soc. 2010, 132, 458−459.
(10) Castrignano,
2010, 398, 1403−1409.
(11) Castrignano, S.; Sadeghi, S. J.; Gilardi, G. Biochim. Biophys. Acta
̀
S.; Sadeghi, S. J.; Gilardi, G. Anal. Bioanal. Chem.
̀
2012, 1820, 2072−2078.
(12) Catucci, G.; Occhipinti, A.; Maffei, M.; Gilardi, G.; Sadeghi, S. J.
Int. J. Mol. Sci. 2013, 14, 2707−2716.
(13) Pumera, M. Mater. Today 2011, 14, 308−315.
(14) Artiles, M. S.; Rout, C. S.; Fisher, T. S. Adv. Drug. Delivery Rev.
2011, 63, 1352−1360.
The use of GO in the presence of DDAB can be considered
as a novel clinically relevant biotechnological approach with
potential application not only in high-throughput screening of
new chemical entities as possible drugs metabolized by hFMO3
but also in deciphering the implications of hFMO3 poly-
morphism in drug metabolism, i.e., personalized medicine.
(15) Kuila, T.; Bose, S.; Khanra, P.; Mishra, A. K.; Kim, N. H.; Lee, J.
H. Biosens. Bioelectron. 2011, 26, 4637−4648.
(16) He, H. Y.; Riedl, T.; Lerf, A.; Klinowski, J. J. Phys. Chem. 1996,
100, 19954−19958.
(17) Lerf, A.; He, H.; Forster, M.; Klinowski, J. J. Phys. Chem. B 1998,
102, 4477−4482.
(18) He, H. Y.; Klinowski, J.; Forster, M.; Lerf, A. Chem. Phys. Lett.
1998, 287, 53−56.
ASSOCIATED CONTENT
■
S
* Supporting Information
Electron transfer rate constants (k’s) calculated from
empirically fitted Laviron’s equations. This includes the plot
of cathodic and anodic peak potentials versus the logarithm of
the scan rates in the range of 10−22 V s−1 where irreversible
electrochemistry of hFMO3 protein was observed using cyclic
voltammetry. This material is available free of charge via the
Internet at http://pubs.acs.org/.
(19) Compton, O. C.; Nguyen, S. T. Small 2010, 6, 711−723.
(20) McCreery, R. L. Chem. Rev. 2008, 108, 2646−2687.
(21) Shan, C.; Yang, H.; Song, J.; Han, D.; Ivaska, A.; Niu, L. Anal.
Chem. 2009, 81, 2378−2382.
(22) Banks, C. E.; Davies, T. J.; Wildgoose, G. G.; Compton, R. G.
Chem. Commun. 2005, 7, 829−841.
(23) Galande, C.; Gao, W.; Mathkar, A.; Dattelbaum, A. M.;
Narayanan, T. N.; Mohite, A. D.; Ajayan, P. M. Part. Part. Syst.
Charact. 2014, 31, 619−638.
(24) Dreyer, D. R.; Todd, A. D.; Bielawski, C. W. Chem. Soc. Rev.
2014, 43, 5288−5301.
(25) Rusling, J. F. Acc. Chem. Res. 1998, 31, 363−369.
(26) Catucci, G.; Gilardi, G.; Jeuken, L.; Sadeghi, S. J. Biochem.
Pharmacol. 2012, 83, 551−558.
(27) Macheroux, P. In Flavoprotein Protocols: Methods in Molecular
Biology; Chapman, S. K., Reid, G. A., Eds; Humana Press: Totowa, NJ,
1999; pp 1−7.
(28) Lang, D. H.; Rettie, A. E. Br. J. Clin. Pharmacol. 2000, 50, 311−
314.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: sheila.sadeghi@unito.it. Tel.: +39-011-6704528. Fax:
+39-011-6704643.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors wish to acknowledge financial support from the
Progetto Ateneo-San Paolo 2012.
(29) Zhou, M.; Zhai, Y.; Dong, S. Anal. Chem. 2009, 81, 5603−5613.
F
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry
Article
(30) Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.;
Dommett, G. H.; Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S. Nature
2007, 448, 457−460.
(31) Wang, K.; Ruan, J.; Song, H.; Zhang, J.; Wo, Y.; Guo, S.; Cui, D.
Nanoscale Res. Lett. 2011, 6, 8−15.
(32) Barth, A.; Zscherp, C. Q. Rev. Biophys. 2002, 35, 369−430.
(33) Barth, A. Biochim. Biophys. Acta 2007, 1767, 1073−1101.
(34) Laviron, E. J. Electroanal. Chem. 1979, 101, 19−28.
(35) Hamman, M. A.; Haehner-Daniels, B. D.; Wrighton, S. A.;
Rettie, A. E.; Hall, S. D. Biochem. Pharmacol. 2000, 60, 7−17.
(36) Hodgson, E.; Rose, R. L.; Cao, Y.; Dehal, S. S.; Kupfer, D. J.
Biochem. Mol. Toxicol. 2000, 14, 118−120.
(37) Krueger, S. K.; Vandyke, J. E.; Williams, D. E.; Hines, R. N. Drug
Metab. Rev. 2006, 38, 139−147.
(38) Shimizu, M.; Yano, H.; Nagashima, S.; Murayama, N.; Zhang, J.;
Cashman, J. R.; Yamazaki, H. Drug Metab. Dispos. 2007, 35, 328−330.
G
DOI: 10.1021/ac504535y
Anal. Chem. XXXX, XXX, XXX−XXX