Electrochemical detection of human cytochrome P450 2A6 inhibition: A step toward reducing dependence on smoking

Article abstract of DOI:10.1021/ac4041839

Inhibition of human cytochrome P450 2A6 has been demonstrated to play an important role in nicotine metabolism and consequent smoking habits. Here, the "molecular Lego" approach was used to achieve the first reported electrochemical signal of human CYP2A6 and to improve its catalytic efficiency on electrode surfaces. The enzyme was fused at the genetic level to flavodoxin from Desulfovibrio vulgaris (FLD) to create the chimeric CYP2A6-FLD. Electrochemical characterization by cyclic voltammetry shows clearly defined redox transitions of the haem domain in both CYP2A6 and CYP2A6-FLD. Electrocatalysis experiments using coumarin as substrate followed by fluorimetric quantification of the product were performed with immobilized CYP2A6 and CYP2A6-FLD. Comparison of the kinetic parameters showed that coumarin catalysis was carried out with a higher efficiency by the immobilized CYP2A6-FLD, with a calculated k_catvalue significantly higher (P < 0.005) than that of CYP2A6, whereas the affinity for the substrate (K _M) remained unaltered. The chimeric system was also successfully used to demonstrate the inhibition of the electrochemical activity of the immobilized CYP2A6-FLD, toward both coumarin and nicotine substrates, by tranylcypromine, a potent and selective CYP2A6 inhibitor. This work shows that CYP2A6 turnover efficiency is improved when the protein is linked to the FLD redox module, and this strategy can be utilized for the development of new clinically relevant biotechnological approaches suitable for deciphering the metabolic implications of CYP2A6 polymorphism and for the screening of CYP2A6 substrates and inhibitors.

Full text of DOI:10.1021/ac4041839

Article
pubs.acs.org/ac
Electrochemical Detection of Human Cytochrome P450 2A6
Inhibition: A Step toward Reducing Dependence on Smoking
†
†
†,‡
†
†
̀
Silvia Castrignano, Alex Ortolani, Sheila J. Sadeghi, Giovanna Di Nardo, Paola Allegra,
,
†,‡
and Gianfranco Gilardi*
†
Department of Life Sciences and Systems Biology, University of Torino, 10123 Torino, Italy
‡
Centre for Nanostructured Interfaces and Surfaces, University of Torino, via Pietro Giuria 7, 10125 Torino, Italy
ABSTRACT: Inhibition of human cytochrome P450 2A6 has
been demonstrated to play an important role in nicotine
metabolism and consequent smoking habits. Here, the
“
molecular Lego” approach was used to achieve the first
reported electrochemical signal of human CYP2A6 and to
improve its catalytic efficiency on electrode surfaces. The
enzyme was fused at the genetic level to flavodoxin from
Desulfovibrio vulgaris (FLD) to create the chimeric CYP2A6-
FLD. Electrochemical characterization by cyclic voltammetry
shows clearly defined redox transitions of the haem domain in
both CYP2A6 and CYP2A6-FLD. Electrocatalysis experiments
using coumarin as substrate followed by fluorimetric
quantification of the product were performed with immobilized
CYP2A6 and CYP2A6-FLD. Comparison of the kinetic parameters showed that coumarin catalysis was carried out with a higher
efficiency by the immobilized CYP2A6-FLD, with a calculated kcat value significantly higher (P < 0.005) than that of CYP2A6,
whereas the affinity for the substrate (K ) remained unaltered. The chimeric system was also successfully used to demonstrate
M
the inhibition of the electrochemical activity of the immobilized CYP2A6-FLD, toward both coumarin and nicotine substrates, by
tranylcypromine, a potent and selective CYP2A6 inhibitor. This work shows that CYP2A6 turnover efficiency is improved when
the protein is linked to the FLD redox module, and this strategy can be utilized for the development of new clinically relevant
biotechnological approaches suitable for deciphering the metabolic implications of CYP2A6 polymorphism and for the screening
of CYP2A6 substrates and inhibitors.
uman cytochrome P450 2A6 (CYP2A6) is a polymorphic
phase I drug metabolizing enzyme representing up to
In addition, grapefruit juice, whose inhibitory characteristic of
CYPs is well-established, has been proposed as a natural
remedy, because its consumption can inhibit coumarin
H
1
0% of total CYPs present in the liver, where it is mainly
1
6
expressed. There is considerable interest in CYP2A6 primarily
due to its leading role in the human metabolism of nicotine, the
addictive agent present in tobacco. About 80% of nicotine is
metabolized by CYP2A6, and a clear relationship between
CYP2A6 polymorphism and dependence from smoking has
been demonstrated, together with the consequent risk of lung
metabolism in vivo. Other attempts have also been made to
produce synthetic CYP2A6 inhibitors by mimicking the
7
chemical structure of nicotine. The discovery and improve-
ment of new powerful and selective CYP2A6 inhibitors with
low or no risk of undesirable side effects is an ongoing research
area. Therefore, the development of new bioanalytical methods
suitable for the screening of CYP2A6 substrates and inhibitors
together with metabolic implications of CYP2A6 poly-
morphism is extremely desirable and of high clinical relevance,
and this is the aim of the present work.
2
,3
cancer. It has been widely reported that the individual
variability in nicotine metabolism rate, including ethnic group
and gender-related differences, is highly correlated to CYP2A6
2
genetic polymorphism. In particular, it has been observed that
A well-known substrate of CYP2A6 is coumarin, the simplest
subjects with a genetic basis for slow CYP2A6 metabolism
smoke fewer cigarettes and are more likely to stop smoking,
8
member of plant-derivative benzopyrones. In humans around
2,3
8
0% of coumarin is metabolized by CYP2A6, and its 7-
compared to individuals with faster metabolism. For these
reasons, the use of CYP2A6 inhibitors has been suggested as a
hydroxylation is commonly used both in vitro and in vivo as a
1
,8
4
probe of CYP2A6 activity.
Coumarin derivatives are
possible treatment strategy for tobacco-dependent smokers.
9,10
extensively used as anticoagulant and antithrombotic agents.
Among drugs and xenobiotics that can be suitable for
CYP2A6 inhibition, methoxsalen, a clinically approved drug for
the treatment of psoriasis, has been proposed to reduce
Received: December 23, 2013
Accepted: February 5, 2014
Published: February 5, 2014
5
smoking behavior, but it may not be selective for CYP2A6 and
could increase the risk of undesirable drug−drug interactions.
©
2014 American Chemical Society
2760
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766

Analytical Chemistry
Article
Due to their biotechnological potential and relevance to drug
metabolism, CYPs have received considerable attention for the
development of bioelectrochemical platforms suitable for
This was done with the aim of improving the expression and
28
stability of the construct, as previously reported. The cyp2A6
gene was amplified by PCR using the forward and reverse
primers (restriction enzyme sites shown in bold) and
subsequently ligated into the pCW expression vector. Primers
were 5′ CAA GAT ACT ACC CAT ATG GCG AAA AAG
ACC TCG AGC AAG GGG AAG CTG 3′ (2A6 forward) and
5′ TTC CTG CCC CGC CAC CAT CAC CAT TGA AAG
CTT GCG AGG GCT 3′ (2A6 reverse).
Cyp2A6-FLD was cloned into pCW expression vector
starting from CYP2A6 clone. Two primers, reported below,
were used to amplify and modify the existing Cyp2A6 in order
to delete the stop codon and to insert a unique AvrII restriction
site. The AvrII restriction site allowed for the cloning of the
11−20
substrate and inhibitor screening.
Particular interest has
been focused on the improvement of the electronic coupling of
the enzyme with the electrode surface and of its catalytic
efficiency. In a fully coupled system, consumption of reducing
equivalents and molecular oxygen is stoichiometrically
equivalent to the product formed, whereas in uncoupled
systems, reducing equivalents are wasted in superoxide and
hydrogen peroxide production, resulting in an impairment of
16
the enzyme catalytic activity. In order to address these
difficulties, significant improvement has been made by
optimizing advanced methods for electrode surface modifica-
tion in conjunction with the development of appropriate
modified Cyp2A6 gene into the pCW expression vector already
1
6,17,19
23,24
protein engineering strategies.
To this end, our group has
containing the sequence coding for the FLD gene.
. Primers
previously developed and implemented the so-called “molecular
were 5′ CAA GAT ACT ACC CAT ATG GCG AAA AAG
ACC TCG AGC AAG GGG AAG CTG 3′ (2A6 forward) and
5′ CCA CGA AAC TAC ACC ATG AGC TTC CTG CCC
CGC CCT AGG AAT TAA 3′ (2A6 reverse).
2
1−25
Lego” approach
transfer modules such as Desulfovibrio vulgaris flavodoxin
FLD) to the CYP of interest resulting in active multidomain
based on the genetic fusion of electron
(
P450 enzymes. In particular, it has been demonstrated that
using artificial redox chains, it is possible to regulate the
electron flow from the electrode to the catalytic domain by
mediation of the artificial reductase domain, thus enhancing
Both clones were sequenced in order to confirm the correct
2
3,24
amino acid sequence. As previously reported,
CYP2A6 and
CYP2A6-FLD fusion proteins were expressed in E. coli DH5α
cells with a six histidine tag at the C-terminus. The resulting
recombinant proteins were purified by coupling ionic exchange
(DEAE fast flow) and Ni-affinity chromatography. Both
CYP2A6 and CYP2A6-FLD were stored in 50 mM phosphate
buffer pH 7.4 with 20% glycerol at −20 °C.
1
6,26
coupling efficiency and catalytic activity.
Here, we apply the “molecular Lego” approach to improve
the catalytic efficiency of CYP2A6 by fusing it at the genetic
level with the FLD redox module, which is chosen for its high
sequence similarity to the FMN binding domain of the class II
Fourier Transform Infrared Spectroscopy (FTIR). Infra-
red spectra of CYP2A6 were acquired using the attenuated total
reflectance (ATR) tool (Harrick Scientific Products, U.S.A.).
Protein samples were prepared on gold-PET flat surface
substrates following the same procedure described for glassy
carbon electrodes and compared with samples prepared by
gently mixing equal volumes of enzyme solution and buffer (50
mM phosphate buffer pH 7.4 with 20% glycerol). Before the
FTIR analysis, samples were kept overnight at 4 °C. All spectra
were collected from 4000 to 800 cm− using a Bruker Model
Tensor 27 FT-IR spectrometer (Bruker Instruments, U.S.A.)
2
1
CYP reductases. Electrochemical study of CYP2A6-FLD is
performed by immobilizing the fusion protein on a poly
diallyldimethylammonium chloride (PDDA)-modified glassy
carbon electrode, and results obtained are compared to those
relative to CYP2A6 enzyme as a control. The rate of electron
flow through the CYP2A6-FLD redox chain is also estimated by
27
applying Laviron’s theory using cyclic voltammetry. Fur-
thermore, the catalytic performance of CYP2A6-FLD is
investigated using the well-known substrate coumarin. We
examine CYP2A6-FLD turnover efficiency in the presence of
coumarin by chronoamperometric electrocatalysis on the glassy
carbon immobilized enzyme in comparison to CYP2A6. In
addition, coupling efficiency of the enzyme is examined during
coumarin electrocatalysis by online measurement of hydrogen
peroxide formed using a screen-printed platinum electrode.
Finally, the inhibition of immobilized CYP2A6-FLD in the
presence of tranylcypromine, a potent and selective CYP2A6
inhibitor, is demonstrated toward coumarin and nicotine
electrocatalysis.
1
−1
29
with a scan velocity of 10 kHz and a resolution of 4 cm .
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, U.S.A.). Spectra of the protein were
collected in triplicate, subsequently averaged, and then
corrected by subtraction of control samples acquired under
the same scanning and temperature conditions. In particular, IR
spectra of buffer or PDDA were used with the same dilution as
background for enzyme samples.
EXPERIMENTAL SECTION
■
Immobilization on Glassy Carbon Electrodes. Both
CYP2A6 and CYP2A6-FLD proteins were immobilized on
PDDA-modified glassy carbon electrodes (diameter of 0.07
Chemicals. Analytical grade chemicals were used with no
further purification. All solutions were prepared with ultrapure
deionized water. Coumarin, 7-hydroxy-coumarin, nicotine,
tranylcypromine (Sigma-Aldrich), and nicotine-Δ1′(5′)-
iminiumdiperchlorate salt (Biozol) solutions were prepared
just before use by dissolving the adequate amount in the
appropriate solvent. Poly diallyldimethylammonium chloride
2
cm , BASi, U.K.). To this end, a 1:1 mixture of PDDA/protein
was cast onto clean glassy carbon electrodes and allowed to set
overnight at 4 °C.
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 a 3
(
PDDA) (Sigma-Aldrich) was used without dilution.
Engineering, Expression, and Purification of CYP2A6
and CYP2A6-FLD. The Cyp2A6 gene was modified at the 5′-
end, corresponding to the N-terminus of the protein, in order
to remove the hydrophobic membrane anchor by deleting the
first 22 amino acids using the primer 2A6 FW reported below.
2
761
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766

Analytical Chemistry
Article
mm diameter glassy carbon working electrode (BASi, U.K.),
was also used.
detected maximum emission at 450 nm. Excitation and
emission slit sizes were set at 4 and 6 nm, respectively.
Electrochemical investigation of CYP2A6 and CYP2A6-FLD
properties was performed by cyclic voltammetry, and the
potential was scanned between 400 and −850 mV at a scan rate
of 120 mV/s. Cyclic voltammetry experiments were performed
in a nitrogen atmosphere within a glovebox (Belle Tech-
nologies, U.K.).
HPLC. The quantification of nicotine electrocatalysis
product formation was achieved 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. For this purpose, standard solutions of nicotine and
Δ1′(5′)-iminium ion were separated using a linear gradient
elution programmed as follows: linear gradient elution from
Electrochemically driven substrate catalysis by CYP2A6 and
CYP2A6-FLD was performed using chronoamperometry with
an applied potential bias of −650 mV for 30 min. In order to
ensure the substrate permeation into the enzymatic layer and
minimize mass transport influence at the transducer surface,
both CYP2A6 and CYP2A6-FLD were immobilized through
PDDA on glassy carbon rotating disk electrodes. All
chronoamperometric electrocatalysis experiments were per-
formed at 200 rpm rotation speed using a BASi RDE-2 rotator
system (Bioanalytical Systems, Inc., U.S.A.). The chronoam-
perometric procedure was applied on freshly prepared
electrodes in the presence of increasing concentrations of
substrate, and the product obtained after the 30 min reaction
was immediately quantified by fluorescence spectroscopy
detection in the case of coumarin and by high performance
liquid chromatography (HPLC) in the case of nicotine.
The coupling efficiency of the immobilized enzymes was
measured during the electrocatalysis of coumarin (saturating
concentration = 10 μM) using the above-mentioned procedure.
In particular, the amount of hydrogen peroxide produced was
estimated using miniaturized screen-printed platinum electro-
des (BVT Technologies, Czech Republic) by sampling 30 μL of
the electrocatalysis solution every 5 min. Hydrogen peroxide
quantification was performed on a platinum screen-printed
electrode, after calibration with standard solutions, in
chronoamperometry by applying a potential of 600 mV for 5
min and by measuring the steady-state current. Hydrogen
peroxide concentration was estimated using the calibration
curve after volume-appropriate correction. As a control
experiment, ethanol (used for dissolving coumarin), instead
of the substrate, was added to both CYP2A6 and CYP2A6-FLD
electrodes.
2
0% to 50% acetonitrile and 80% to 50% 20 mM phosphate
buffer pH 7, 0−10 min; isocratic elution of 50% acetonitrile and
5
elution to 20% acetonitrile and 80% 20 mM phosphate buffer
pH 7, 24 min or end of the run. Flow rate was set to 1 mL/min.
Detection wavelength was set to 257 nm. Retention times were
0% 20 mM phosphate buffer pH 7, 10−24 min; linear gradient
3
and 21 min for Δ1′(5′)-iminium ion and nicotine,
respectively.
RESULTS AND DISCUSSION
■
CYP2A6 and CYP2A6-FLD were expressed in E. coli and
purified using the protocols established in our lab for other
23
CYPs and reported previously. Reduction of the haem
followed by bubbling of carbon monoxide in both purified
proteins showed a complete spectral shift of the Soret peak at
4
18 nm to the expected 450 nm, indicating the presence of an
active enzyme.
Prior to the electrochemical experiments, ATR-FTIR analysis
was carried out to assess the structural integrity of the catalytic
domain in the presence and absence of the cationic
polyelectrolyte PDDA (poly diallyldimethylammonium chlor-
ide). In particular, the spectral region included between 1450
and 1750 cm− was considered, which contains information on
protein secondary structure composition. As shown in Figure 1,
1
Inhibition studies were carried out on CYP2A6-FLD glassy
carbon electrodes using chronoamperometry, with the same
parameters mentioned above for coumarin and nicotine
electrocatalysis. Inhibition experiments were performed in the
presence of coumarin (10 μM) or nicotine (150 μM) as the
substrate and increasing concentrations of tranylcypromine
(
from 0.01 to 100 μM) as the inhibitor. The decrease in the
amount of 7-hydroxy coumarin formed in the presence of the
tranylcypromine inhibitor during coumarin electrocatalysis was
calculated by fluorescence spectroscopy. The decrease in the
amount of nicotine-Δ1′(5′)-iminium ion formed in the
presence of the tranylcypromine inhibitor during nicotine
electrocatalysis was estimated by HPLC.
Figure 1. FTIR spectra of (A) CYP2A6 and (B) CYP2A6 in the
presence of PDDA. Spectra were collected in triplicate and
subsequently averaged and corrected by background subtraction.
Fluorescence Spectroscopy. Following the termination of
the chronoamperometric experiments in the presence of
different concentrations of the substrate coumarin, the
electrochemical cell contents were transferred to a fluorimeter
for the subsequent 7-hydroxy coumarin product quantification.
Fluorescence spectra were recorded at room temperature with a
Perkin-Elmer LS 55 fluorescence spectrometer (Perkin-Elmer,
U.S.A.). The excitation wavelength was set at 325 nm, and
emission was monitored between 400 and 600 nm with a
spectra obtained for both CYP2A6 in the absence and in the
presence of PDDA are essentially similar and exhibit the amide
I and II bands, which are characteristics of the secondary
30
−1
structure of proteins. The amide I band (∼1650 cm )
consists of a group of overlapped signals, mainly due to the
peptidic carbonyl group stretching mode with a minor
3
0
contribution from the out-of-phase C−N stretching vibration.
Grouped bands centered at around 1550 cm− are commonly
1
2
762
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766

Analytical Chemistry
Article
assigned to the amide II band, which corresponds to the out-of-
phase combination of the peptidic N−H in-plane bend and the
−
1
C−N stretching vibration. Additionally, the peak at 1517 cm
can be assigned to the C−C stretch and C−H bend of side
chain tyrosine residues. Moreover, a set of bands that can be
assigned to the amide III are also clearly detectable in the
30
−1
region between 1200 and 1400 cm . The latter signals are
related to the in-phase combination of the peptidic N−H and
30
C−N bonds bending vibration.
The ATR-FTIR data show that the catalytic domain is not
denatured upon immobilization and that the secondary
structure of CYP2A6 is not affected by the presence of
PDDA. Subsequently, direct electrochemistry of both enzymes
was performed using cyclic voltammetry on PDDA-modified
glassy carbon electrodes. Cyclic voltammograms of CYP2A6
and CYP2A6-FLD collected under anaerobic conditions are
reported in Figure 2, where a redox couple corresponding to
III/II
haem Fe
transition with a midpoint potential of about −230
mV is shown.
An additional redox couple is observed in the cyclic
voltammogram of CYP2A6-FLD and is assigned to the
semiquinone−hydroquinone transition of the FMN cofactor
with a midpoint potential of about −660 mV. No voltammetric
peaks are observed in the absence of enzyme (data not shown).
III/II
Peak-to-peak separation of about 130 mV for the haem Fe
transition and 180 mV for the FLD semiquinone−hydro-
quinone indicate a quasi-reversible electrochemical processes
typical of slow electron transfer kinetics. The electron transfer
rate constant (k ) was also calculated from empirically fitted
s
27
Laviron’s equations. In particular, irreversible electrochemis-
try of both CYP2A6 and CYP2A6-FLD PDDA-immobilized
proteins was studied by cyclic voltammetry with an increase in
−1
the scan rate from 0.2 to 0.5 V s , and the cathodic and anodic
peak potentials were plotted versus the logarithm of the scan
rate. The k values were calculated from the intercept of the
s
−1
plot of E versus ln ν and were determined to be 0.13 s for
CYP2A6 and 0.08 s for CYP2A6-FLD. As previously
−
1
1
9
reported, low electron transfer rate is essential for electro-
chemically driven catalysis of P450 systems, because it allows
for the fine-tuning of the electron delivery from the reductase
to the haem catalytic site. A summary of electrochemical data
calculated for CYP2A6 and CYP2A6-FLD is reported in Table
1.
The activity of both CYP2A6 and CYP2A6-FLD using
coumarin as the marker substrate was investigated by
chronoamperometric delivery of reducing equivalents to the
immobilized enzymes in the presence of increasing concen-
trations of this substrate. Fluorescence detection was then used
Figure 2. Anaerobic cyclic voltammograms of CYP2A6 (A) and
CYP2A6-FLD (B) on PDDA-modified glassy carbon electrodes.
Shown are the original and baseline corrected (upper traces) cyclic
voltammograms. Scan rate: 120 mV s in in 50 mM potassium
phosphate buffer pH 7.4 with 100 mM KCl at 25 °C.
−1
3
1
to quantify the 7-hydroxy coumarin product with an
excitation wavelength of 325 nm and an emission wavelength
of 463 nm, as described in the Experimental Section. The
results obtained using CYP2A6 and CYP2A6-FLD glassy
carbon electrodes in the presence of increasing concentrations
of coumarin are shown in Figure 3A,B, respectively. Michaelis−
Menten curves for the turnover of both CYP2A6 and CYP2A6-
FLD in the presence of coumarin were obtained by plotting the
rate of product formation versus the relative coumarin
concentrations (Figure 3C). As can be seen from the figure, a
significant difference between CYP2A6 and CYP2A6-FLD was
observed in the presence of coumarin, which was confirmed by
a one-way ANOVA followed by a post hoc Tukey test (P <
for CYP2A6 and 1.00 ± 0.28 μM and 0.28 ± 0.01 min− for
CYP2A6-FLD. Results from a student’s t test showed no
differences when evaluating KM values, whereas a significant
difference between CYP2A6 and CYP2A6-FLD was found
when comparing kcat values. In particular, the kcat value of
coumarin electrocatalysis was significantly higher for CYP2A6-
FLD than for CYP2A6 (P < 0.005).
1
Kinetic values obtained for CYP2A6 have been compared in
Table 2 with previously published results for coumarin
32−35
catalysis.
Because to our knowledge this is the first report
of the immobilization of CYP2A6, the values reported in the
latter table are those of the recombinant enzyme free in
solution and not immobilized. Regarding the KM values, our
0
.001). K and k values for the electrocatalysis of coumarin
M cat
−1
were calculated to be 1.03 ± 0.05 μM and 0.17 ± 0.01 min
2
763
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766

Analytical Chemistry
Article
Table 1. Redox Properties of CYP2A6 and CYP2A6-FLD Immobilized on Glassy Carbon Electrodes Modified with PDDA
Ec (mV)
Ea (mV)
Em (mV)
ΔE (mV)
ks (s−1)
CYP2A6
−294 ± 8
−321 ± 12
−174 ± 2
−128 ± 4
−572 ± 6
−234 ± 5
−225 ± 4
−662 ± 6
121 ± 7
134 ± 15
181 ± 10
0.13
0.08
CYP2A6-FLD
−753 ± 7
Figure 3. Formation of 7-hydroxycoumarin by CYP2A6 (A) and CYP2A6-FLD (B) immobilized on PDDA glassy carbon electrodes detected by
fluorescence spectroscopy. (C) Plot of reaction velocity versus substrate concentration fitted to a Michaelis−Menten equation for CYP2A6 (circles)
and CYP2A6-FLD (triangles). Data are shown as mean ± SD of three different electrodes.
Table 2. Comparison of Calculated Kinetic Parameters for
Coumarin Catalysis by Immobilized CYP2A6 and Literature
Data Determined on Recombinant Enzyme (Rec) Free in
Solution
electrodes does affect the turnover number, because not all
the enzymes are within distances that allow for electron
transfer, and this is not unique to the system described here.
More interesting though is the comparison between the two
immobilized proteins where the chimeric enzyme shows a
higher catalytic efficiency (k /K ) to that of the CYP2A6 (i.e.,
KM (μM)
kcat (min−1)
variant, host
cat
M
present work:
−1
−1
0
.28 versus 0.17 min μM ).
An evaluation of the coupling efficiency of immobilized
2
2
A6
1.03 ± 0.05
1.00 ± 0.28
0.17 ± 0.01 rec, E. coli
0.28 ± 0.01 rec, E. coli
A6-FLD
CYP2A6 and CYP2A6-FLD was also carried out by measuring
the hydrogen peroxide production during coumarin electro-
catalysis. The chronoamperograms of sampled solutions during
electrocatalysis of CYP2A6 and CYP2A6-FLD obtained using
screen-printed platinum electrodes are shown in Figure 4A,B.
Hydrogen peroxide production profiles (Figure 4C) were
obtained by plotting the calculated hydrogen peroxide
concentration values versus time for both coumarin electro-
catalysis and control experiments (in the presence of only
ethanol−coumarin solvent). The different hydrogen peroxide
production profiles were compared by two-way ANOVA
followed by a post hoc Tukey test, which showed that
hydrogen peroxide produced during coumarin electrocatalysis
was significantly higher for immobilized CYP2A6 than
3
2
Soucek, 1999
1.48 ± 0.37
2.17 ± 0.63
3.36 ± 0.18 rec, E. coli
3
3
He et al., 2004
3.23
rec, Spodoptera
frugiperda
rec, E. coli
rec, E. coli
3
4
Fukami et al., 2005
1.00 ± 0.10
1.80 ± 0.70
5.2 ± 0.3
4.5 ± 0.5
3
5
Yun et al., 2005
results were found to be in the same range as those in the
literature (1−2 μM), confirming the preservation of protein
affinity for the substrate. However, the electrochemical kcat
values for both proteins were found to be lower than the values
reported for CYP2A6 free in solution. It must be noted that in
general the nonoriented immobilization of enzymes on
Figure 4. Estimation of the amount of hydrogen peroxide produced during coumarin electrocatalysis by CYP2A6 (A) and CYP2A6-FLD (B)
performed on screen-printed platinum electrode. (C) Hydrogen peroxide production profile during coumarin electrocatalysis (filled symbols) and
control experiments (empty symbols) on immobilized CYP2A6 (circles) and CYP2A6-FLD (triangles) on glassy carbon electrodes.
2
764
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766

Analytical Chemistry
Article
CYP2A6-FLD (P < 0.001). Furthermore, on CYP2A6 glassy
carbon electrodes, hydrogen peroxide was produced in
statistically higher amounts during coumarin electrocatalysis
than during control experiments in the presence of only buffer
oxidation mediated by CYP2A6. Second, the Δ1′(5′)-iminium
ion product of this reaction is further metabolized to cotinine
6
,39,40
by aldehyde oxidase.
Here, we first calculated the KM
value by performing electrocatalysis of nicotine on CYP2A6-
FLD glassy carbon electrodes. The estimated KM value was
131.4 ± 52.2 μM, which is in agreement with previously
(P < 0.001). However, no significant differences were found
when comparing hydrogen peroxide production during
coumarin electrocatalysis, and control experiments on
CYP2A6-FLD glassy carbon electrodes confirmed that the
chimeric protein is more coupled. The latter observation is also
supported by the fact that the immobilized CYP2A6-FLD yields
double the amount of 7-hydroxycoumarin product when
compared to CYP2A6 alone, as mentioned earlier (Table 2).
The data obtained demonstrate that the regulation of
electron flow to the haem of immobilized CYP2A6 when
genetically fused to D. vulgaris FLD significantly enhanced both
coupling and catalytic efficiencies of this enzyme toward
coumarin hydroxylation.
40
published results on microsomal CYP2A6 protein. Sub-
sequently, we used CYP2A6-FLD glassy carbon electrodes to
investigate the inhibition of nicotine metabolism by tranylcy-
promine. The inhibition profile of CYP2A6-FLD nicotine 5′-
oxidation by tranylcypromine is shown in Figure 6.
Finally, the improved catalytic efficiency of CYP2A6 when
genetically fused to FLD was examined toward enzyme
inhibition in the presence of tranylcypromine, a potent and
1
relatively selective inhibitor of CYP2A6. Tranylcypromine is a
nonhydrazine monoamine oxidase inhibitor used in psychia-
4
0
1
try and has proven inhibitory activity toward many CYPs.
Previously published studies on human liver microsomes or
cDNA-expressing microsomes have reported the inhibition of
CYP2A6 activity toward coumarin 7-hydroxylation by
tranylcypromine, with calculated K values ranging from 0.04
i
1
,36,37
to 0.42 μM.
Here, we used CYP2A6-FLD glassy carbon
electrodes to investigate the inhibition of coumarin metabolism
by tranylcypromine.
The inhibition profile of CYP2A6-FLD coumarin 7-
hydroxylation by tranylcypromine is shown in Figure 5. The
Figure 6. Tranylcypromine concentration-dependent inhibition of
nicotine 5′-oxidation by CYP2A6-FLD-modified glassy carbon electro-
des. Data are shown as mean ± SD of three different electrodes.
The calculated IC50 value was 3.7 ± 0.2 μM. Using the
38
Cheng and Prusoff relationship for competitive inhibition, the
calculated IC value was also converted to K , obtaining a value
50
i
of 1.72 ± 0.09 μM. No data regarding tranylcypromine
inhibition of CYP2A6 5′-oxidation of nicotine are found in the
literature to compare with our results.
CONCLUSIONS
■
In conclusion, for the first time, a bioelectrochemical setup
suitable for the investigation of CYP2A6 catalytic activity has
been developed by immobilizing on glassy carbon electrodes
the chimeric form of the enzyme CYP2A6-FLD obtained by
genetically fusing CYP2A6 with FLD. The fusion protein
showed an improved catalytic efficiency in agreement with
16
previously reported data on CYP3A4, where the regulation of
the electron flow to the haem through FLD resulted in higher
amounts of product due to an enhancement of the coupling
efficiency. It was also demonstrated that the increase in catalytic
activity observed for CYP2A6-FLD when compared to
CYP2A6 could be related to the improved control of the
enzyme coupling efficiency and to a more refined electron flow
tuning. The CYP2A6-based bioelectrochemical platform was
also successfully used to test enzymatic inhibition in the
presence of its well-known inhibiting drug, proving the
suitability of this system for the screening of inhibitors of this
enzyme, which is highly relevant to nicotine addiction
treatment. Finally, this molecular engineering approach can
aid in the development of improved electrochemical sensors
Figure 5. Tranylcypromine concentration-dependent inhibition of
coumarin 7-hydroxylation by CYP2A6-FLD-modified glassy carbon
electrodes. Data are shown as mean ± SD of three different electrodes.
calculated IC50 value was 3.1 ± 0.6 μM. Using the Cheng and
Prusoff relationship for competitive inhibition, the calculated
IC value was also converted to K , obtaining a value of 0.29 ±
50
i
0
.06 μM which is in good agreement with previously published
1,36,37
results using microsomes.
In addition, inhibition experiments in the presence of
tranylcypromine were also performed using nicotine as
substrate. In humans, about 70 to 80% of nicotine is converted
to cotinine in two enzymatic steps. The first is nicotine 5′-
2
765
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766

Analytical Chemistry
Article
based on CYP enzymes for drug screening and metabolic
profiling.
(24) Fairhead, M.; Giannini, S.; Gillam, E. M.; Gilardi, G. J. Biol.
Inorg. Chem. 2005, 10, 842−853.
(
25) Sadeghi, S. J.; Gilardi, G. Biotech. Appl. Biochem. 2013, 60, 102−
1
10.
AUTHOR INFORMATION
■
(26) Degregorio, D.; Sadeghi, S. J.; Di Nardo, G.; Gilardi, G.; Solinas,
Corresponding Author
S. P. J. Biol. Inorg. Chem. 2011, 16, 109−116.
(
(
27) Laviron, E. J. Electroanal. Chem. 1979, 101, 19−28.
28) Yano, J. K.; Hsu, M. H.; Griffin, K. J.; Strout, C. D.; Johnson, E.
*
E-mail: gianfranco.gilardi@unito.it. Fax: +39-011-6704643.
Tel.: +39-011-6704593.
F. Nat. Struct. Mol. Biol. 2005, 12, 822−823.
Author Contributions
(
29) Castrignano, S.; Sadeghi, S. J.; Gilardi, G. Biochim. Biophys. Acta
̀
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
2012, 1820, 2072−2078.
(30) Barth, A. Biochim. Biophys. Acta 2007, 1767, 1073−1101.
(
31) Kim, D.; Wu, Z. L.; Guengerich, F. P. J. Biol. Chem. 2005, 280,
4
(
(
0319−40327.
Notes
32) Soucek, P. Arch. Biochem. Biophys. 1999, 370, 190−200.
The authors declare no competing financial interest.
33) He, X. Y.; Shen, J.; Hu, W. Y.; Ding, X.; Lu, A. Y.; Hong, J. Y.
Arch. Biochem. Biophys. 2004, 427, 143−153.
34) Fukami, T.; Nakajima, M.; Higashi, E.; Yamanaka, H.; Sakai, H.;
McLeod, H. L.; Yokoi, T. Drug Metab. Dispos. 2005, 33, 1202−1210.
35) Yun, C.-H.; Kim, K.-H.; Calcutt, M. W.; Guengerich, F. P. J. Biol.
Chem. 2005, 280, 12279−12291.
36) Taavitsainen, P.; Juvonen, R.; Pelkonen, O. Drug Metab. Dispos.
001, 29, 217−222.
37) Zhang, W.; Kilicarslan, T.; Tyndale, R. F.; Sellers, E. M. Drug
Metab. Dispos. 2001, 29, 897−902.
38) Copeland, R. A. Enzymes: A Practical Introduction to Structure,
Mechanism, and Data Analysis, 2nd ed.; John Wiley & Sons: New York,
000.
39) Denton, T. T.; Zhang, X.; Cashman, J. R. Biochem. Pharmacol.
004, 67, 751−756.
40) Murphy, S. E.; Raulinaitis, V.; Brown, K. M. Drug Metab. Dispos.
005, 33, 1166−1173.
(
ACKNOWLEDGMENTS
■
This work was supported by the Region Piedmont CIPE 2006
CYP-TECH project, Italy) and Progetto Ateneo-San Paolo
012 (Torino, Italy awarded to S.J.S.).
(
(
2
(
2
(
REFERENCES
■
(
1) Di, Y. M.; Chow, V. D.; Yang, L. P.; Zhou, S. F. Curr. Drug Metab.
009, 10, 754−780.
2) Malaiyandi, V.; Sellers, E. M.; Tyndale, R. F. Clin. Pharmacol.
(
2
(
2
(
2
(
Ther. 2005, 77, 145−158.
(
(
3) Benowitz, N. L. Annu. Rev. Pharmacol. Toxicol. 2009, 49, 57−71.
4) Sellers, E. M.; Tyndale, R. F.; Fernandes, L. C. Drug Discovery
Today 2003, 8, 487−493.
5) Hukkanen, J.; Jacob, P., III; Benowitz, N. L. Clin. Pharmacol. Ther.
006, 80, 522−530.
6) Yano, J. K.; Denton, T. T.; Cerny, M. A.; Zhang, X.; Johnson, E.
F.; Cashman, J. R. J. Med. Chem. 2006, 49, 6987−7001.
7) Kontogiorgis, C.; Detsi, A.; Hadjipavlou-Litina, D. Expert Opin.
2
(
2
(
(
Ther. Pat. 2012, 22, 437−454.
(
8) Mueller, R. L. Best Pract. Res. Clin. Haematol. 2004, 17, 23−53.
(
9) Fareed, J.; Thethi, I.; Hoppensteadt, D. Annu. Rev. Pharmacol.
Toxicol. 2012, 52, 79−99.
10) Fylaktakidou, K. C.; Hadjipavlou-Litina, D. J.; Litinas, K. E.;
Nicolaides, D. N. Curr. Pharm. Des. 2004, 10, 3813−3833.
11) Yarman, A.; Wollenberger, U.; Scheller, F. W. Electrochim. Acta
013, 110, 63−72.
12) Fantuzzi, A.; Capria, E.; Mak, L. H.; Dodhia, V. R.; Sadeghi, S. J.;
Collins, S.; Somers, G.; Huq, E.; Gilardi, G. Anal. Chem. 2010, 82,
(
(
2
(
1
(
(
0222−10227.
13) Schneider, E.; Clark, D. S. Biosens. Bioelectron. 2013, 39, 1−13.
14) Fantuzzi, A.; Fairhead, M.; Gilardi, G. J. Am. Chem. Soc. 2004,
1
26, 5040−5041.
15) Fantuzzi, A.; Meharenna, Y. T.; Briscoe, P. B.; Sassone, C.;
Borgia, B.; Gilardi, G. Chem. Commun. 2006, 12, 1289−1291.
16) Dodhia, V. R.; Sassone, C.; Fantuzzi, A.; Di Nardo, G.; Sadeghi,
S. J.; Gilardi, G. Electrochem. Commun. 2008, 10, 1744−1747.
17) Ferrero, V. E. V.; Andolfi, L.; Di Nardo, G.; Sadeghi, S. J.;
Fantuzzi, A.; Cannistraro, S.; Gilardi, G. Anal. Chem. 2008, 80, 8438−
446.
18) Mak, L. H.; Sadeghi, S. J.; Fantuzzi, A.; Gilardi, G. Anal. Chem.
010, 82, 5357−5362.
19) Sadeghi, S. J.; Fantuzzi, A.; Gilardi, G. Biochim. Biophys. Acta
011, 1814, 237−248.
20) Sadeghi, S. J.; Ferrero, S.; Di Nardo, G.; Gilardi, G.
Bioelectrochem. 2012, 86, 87−91.
21) Valetti, F.; Sadeghi, S. J.; Meharenna, Y. T.; Leliveld, S. R.;
Gilardi, G. Biosens. Bioelectron. 1998, 13, 675−685.
22) Sadeghi, S. J.; Meharenna, Y. T.; Fantuzzi, A.; Valetti, F.; Gilardi,
G. Faraday Discuss. 2000, 116, 135−153.
23) Dodhia, V. R.; Fantuzzi, A.; Gilardi, G. J. Biol. Inorg. Chem. 2006,
1, 903−916.
(
(
(
8
(
2
(
2
(
(
(
(
1
2
766
dx.doi.org/10.1021/ac4041839 | Anal. Chem. 2014, 86, 2760−2766