Biosensors and Bioelectronics 28 (2011) 105111

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journal homepage:www.elsevier.com/locate/bios

Electrochemical sensing of trimethylamine based on polypyrroleflavincontaining monooxygenase (FMO3) and ferrocene as redox probe for
evaluation of fish freshness
Sondes Bourigua

a,c

Serge Dzyadevych

, Sarra El Ichi

a,b

, Hafsa Korri-Youssoufi

, Nicole Jaffrezic Renault

b,

, Abderrazak Maaref ,

a,

Universit de Lyon, Laboratoire des Sciences Analytiques, Universit Claude Bernard, Lyon 1, 43 Boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex, France
CNRS UMR-8182, Institut de Chimie Molculaires et Matriaux dOrsay, Equipe de Chimie Bioorganique et Bioinorganique, Universit Paris-Sud, Bt 420, 91405 Orsay,
c
France Laboratoire de Physique et Chimie des Interfaces, Facult des Sciences de Monastir, Avenue de lEnvironnement, 5019 Monastir, Tunisia
d
Laboratory of Biomolecular Electronics, Institute of Molecular Biology and Genetics, National Academy of Sciences of Ukraine,150 Zabolotnogo St., Kiev 03143, Ukraine
b

article info
Article history:
Received 31 May 2011
Accepted 4 July 2011
Available online 13 July 2011
Keywords:
Trimethylamine biosensor Flavin-containing monooxygenase 3 Polypyrrole
Ferrocene Redox probe Impedance Amperometry Fish
freshness

and
micr
obiol
ogic
1. Introduction
al
form
Fish is one of the most highly perishable
s of
food products. Dur-ing handling and storage,
dete
quality deterioration of fresh fish rapidly occurs
riora
and limits the shelf life of the product (Sallam,
tion
2007). The quality of fish degrades, due to a
are
complex process in which physical, chemical
impli

abstra
ct
Amperometr
ic
and
impedimetric
biosensor
for detecting
trimethylami
ne
(TMA)
which
represents
good
parameters
for
estimating
fish
freshness
has
been
developed.
The
biosensor is
based on a
conducting
polypyrrole
substituted
with
ferrocenyl,
where flavincontaining
monooxyge
nase
3
(FMO3)

enzyme was immobilised by covalent bonding.

FMO3 catalyzes the monooxygenation TMA to
trimethylamine N-oxide (TMO). For catalysis FMO
require flavin adenine (FAD) as a prosthetic group,
NADPH as a cofactor and molecular oxygen as
cosubstrate. Ferrocenyl group substituted on the
polypyrrole matrix will serve as redox probe for
monitoring the response of the biosensor to TMA.
The construction of the biosen-sor was
characterized by FT-IR, cyclic voltammetry and
impedance measurements. Detection is done
through the analysis of the current of oxidation
signal of the ferrocenyl groups and compared to the
measurement of impedance related to the electrical
properties of the layers. Amperometric and impedimetric response were measured as a function of
TMA concentration in range of 0.4 g mL
1

80 g

mL (6.5 mol L 1.5 mmol L ). Amperometric

measurements show a decrease in current
response which is in correlation with the increase of
the charge transfer resistance demonstrated by
impedance. Cali-bration curve obtained by
impedance spectroscopy shows a high sensitivity
with a dynamic range from (0.4 g mL

to 80 g

mL ). We demonstrated, using ferrocene as redox

probe for catalytic reaction of FMO3, that high
sensitivity and dynamic range was obtained. The
biosensor was stable during 16 days. The
biosensor shows high selectivity and its sensitivity
to TMA in real samples was evaluated using fish
extract after deterioration during storage.

2011 Elsevier B.V. All rights reserved.

cated.
Because
of Corresponding
consumer demand for
fresh refrigerated foods author. Tel.: +33 1
with extended shelf life, 69154723; fax: +33 1
considerable research has
69157281.
Corr
been directed toward using 1
esponding
author.
+33 +33
4 Tel.:
72448306;
fax:
4 E7231206.
mail
addresses:
hafsa.korriyoussoufi@u
-psud.fr
(H.
KorriYoussoufi),
various
preservation
technologies to preserve or
nicole.jaffrezic@univprolong the
lyon1.fr (N. Jaffrezic
Renault).
0956-5663/$

see front
matter 2011
Elsevier B.V.
All rights
reserved. doi:
10.1016/j.bios.
2011.07.005

men
t of
fish
shelf life, while ensuring the safety, of fresh
quali
foods,
including
fishery
products
(ty
duri
Siripatrawan et al., 2009). So the quality of
ng
fish product is influenced by its freshness.
stor
Therefore, the rapid and accurate deterage
mination of freshness is desired for quality and
(Gill,
processing controls (Ozogul et al., 2005).
199
Many indicators were used for the assess2).

Among
all
of
them,concentration rapidly
trimethylamine (TMA) isincreases in products
one
of
the
volatileafter the death of
compounds that can bemicroorganisms
measured in order to(Hamada-Sato et al.,
estimate fish freshness2005).
(Pacquit et al., 2006). It is
produced
by
the Flavin containing
3
decomposition
ofmonooxygenase
is
the
trimethylamine
N-oxide(FMO3)
that
(TMAO)
byenzyme
microorganisms and itsbreakdown

trimethylamine (TMA)
into trimethylamine
oxide
(TMAO)
through a process
called
Noxygenation.
The
flavin-containing
monooxygenase 3 is
an
NADPHdependent enzyme
that

106

S. Bourigua et al. / Biosensors and Bioelectronics 28 (2011) 105111

ferrocene group could play

catalyzes the oxygenation of a widethe role of redox probe as
variety of nucleophilic com-poundsit electrochemical response
sensitive
to
any
containing sulfur, nitrogen or phosphorusis
in
their
atoms. Oxidation of TMA is catalyzed by modification
FMO3 enzyme with NADPH as a co- environment. Amperometric biosensor from the
factor following the reaction.
response of the ferrocene
+
FMO3
will be measured and
PH+H +O
TMAO+H
+
compared
to
the
O+NADP 2
2
The FMO enzymes metabolizeimpedance measurement
xenobiotic substrates, convert-ing the of the polypyrroles layer
substrate to a more polar product, aiding after such reaction. The
in its excretion from the body. Theyperformances of the new
demonstrate broad substrate specificitybiosensor were evaluated
and have considerable overlap infor the detection of TMA in
function with the cytochrome P450PBS solution and in real
enzymes. The FMO oxygen transfer
sample.
Different
potential is much lower than that of the
parameters have been
cytochrome P450 enzymes, limiting FMO
investigated to determine
substrates to heteroatom nucleophiles
the
optimum
working
with the oxidative reaction resulting in the
conditions.
formation of N-oxides. The FMO3
mechanism differs from that of
cytochrome P450 enzymes by binding 2. Experiments
and activating molecular oxygen prior to
binding of the substrate (Cashman,2.1. Reagents
2005). NADPH binds to the FMO3
The gold electrodes
enzyme and reduces the TMA to TMAO.
fabricated
by
Various analytical techniques havewere
evaporated
gold
layer
of
been employed for the deter-mination of
TMA. These techniques include gas300 nm thickness on
chromatography (Mushiroda et al.,silicon substrate using
1999), Colorimetric assay (Pena-Pereiratitanium as adhesion layer
et al., 2010), flow injection analysis of 30 nm thickness. These
electrodes
were
(Dhaouadi et al., 2007; Sadok et al.,gold
1996) or detection using cell (Gamati etprovided by the LAAS,
al., 1991; Li et al., 1994) or recombinantCNRS Toulouse. Flavincell (Suska et al., 2009). Biosensorscontaining
based in catalytic reaction of FMO3 hasmonooxygenase 3 was
purchased from Sigma and
also been developed for TMA (is a human recombinant
Mitsubayashi et al., 2004) and othersmicrosomes
(EC
volatiles compounds ( Mitsubayashi and1.14.13.8), 2.5 mg/mL
Hashimoto, 2002). Catalytic oxygenationprotein with specific activity
reaction of such com-pounds by FMO3 of 44,000 units/mg protein.
leads to better sensitivity than the othersIt was stored in solution of
mM
potassium
kinds of TMA biosensors (Fillit et al.,100
2008). The problem in enzymaticphosphate buffer, pH 7.4.
biosensors is the lost of activity when the Trimethylamine (25 wt.% in
enzyme is immobilised on solid state water) was purchased from
beside their low stability with timeSigmaAldrich.
(Sheldon, 2007). Conducting polymers The monomer 3-N(CPs)
such
as polypyrrole
andhydroxyphtalimide-pyrrole
polyaniline have been demonstrated as a (py-NHP) was synthesized
good matrix for the construction offrom 3 acetic acid methyl
following
the
enzymatic biosensor as they preserveester
the catalytic activity of the enzyme. CPs procedure described in a
provides a controlled method of previous work (Godillot et
localizing an enzyme in a defined areaal., 1996). The product was
obtained
after
various
on the electrodes (Ahuja et al., 2007).
In this work we report thesynthesis steps. The first
development of an electrochemicalstep is the
biosensor for TMA detection, based on
FMO3
enzyme,
flavin-containing
monooxygenase,
immobilised
on
electropolymerized
functionalised
polypyrroles
with
ferrocene.
The

acetylation reaction of Nprotected pyrrole followed

by its conver-sion to 3
acetic-acid methyl ester
using
thallium
based
oxidative transposition. The
hydrolysis of 3-acetic acid
methyl ester in basic
condition gives the 3-acetic
acid pyrrole with a 70%
yield. Grafting of Nhydroxyphtalamide
was
realised
using
N,N
dicyclohexyl car-bodiimide
which leads to 3-acetate Nhydroxyphtalimido pyrrole
(Py-NHP) with a 60% yield.
The purity of product was
characterized by NMR and
mass spectrometry. 1-Ethyl
amino-ferrocene was synthesized by reduction of
cyanomethyl
ferrocene
(purchased from Aldrich)
using mixed metal hybride
halide LiAlH4 AlCl3 (1:1),
fol-lowing a procedure
described in our previous
work (Godillot et al., 1996).
The product obtained with
a 60% yield and was
characterized by NMR and
mass spectrometry.
The buffer solution
used for all experiments
was
phosphate
buffer
saline (PBS) containing
137 mM NaCl, 2.7 mM KCl,
0.01 M KH2 PO4 and 0.01
M K2 HPO4 , pH 7. All
solutions were made up in
ultrapure water (resistance
1
18.2 M cm ) produced by
a Millipore Milli-Q system.
2
.
2
.
P
r
e
p
a
r
a
t
i
o
n
o
f
a
c
o
p

olymer poly
(Nhydroxyphtali
mide-pyrroleco-pyrrole)

allows
a
less
steric
hindrance
for
protein
immobilisation (L et al.,
2010). The solution was
degassed with nitrogen for
10 min prior to use. The
potential applied for the
In order to control the thickness of the
elec-tropolymerization
copolymer
film
and
then
the
reaction was 0.9 V/SCE
reproducibility of polypyrroles layers
and was stopped when we
chronoamperom-etry technique was
reach the desired charge
chosen for the film electrodeposition on
2
the
gold
electrode.
Beforeof 25 mC cm . After
electropolymerization, the gold elec-trode electropolymer-ization, the
was first polished using diamond slurries electrodes were washed
(6, 3, 1, 0.25 m, respectively). Thewith acetonitrile and water.
electrode was washed in an ultrasonic
bath for 10 min in water and ethanol and
dried under a dry N2 flow. After cleaning,2.3. Enzyme attachment
the gold electrodes were immediately
The enzyme, FMO3
placed as working electrode in a threeelectrode electrochemical cell at roomwas attached to the
surface
by
temperature. The copolymer poly(N-polymer
covalent binding of the
hydroxyphtalimide-pyrrole-co-pyrrole)
film was electrochemically prepared onamine group from lysine to
the N-hydroxy phtalamide
gold electrode from a solution containing
reactive group by the
8
mM
pyrrole,
2
mM
Nformation of amide link.
hydroxyphtalimide-pyrrole and 0.5 MThe
copolymer-modified
LiClO4 as supporting elec-trolyte inelectrode was immersed
acetonitrile. The concentration wasafter electropoly-merization
chosen according to the previous studyin an enzyme solution of
where we demonstrate that the film0.5 mg mL1 FMO3 in a
obtain with this ratio of concentration

phosphate buffer solution

(8 mM, pH 7.0) and was

left overnight at +4 C. The

electrode was then washed
with a phosphate buffer
solution pH 7.0 to remove
the unbound enzyme.
2.4. Covalent attachment
of ethyl amine ferrocene
After immobilization of
the enzyme FMO3 onto the
copolymer
layer,
the
electrode was dipped for 1
h
at
a
controlled

temperature of 20 C in a
solution
of
phosphate
buffer pH 8.5 containing
ethyl amine ferrocene. The
pH of 8.5 was used in
order to avoid the protonation of the amine on
the ferrocenyl group. This
solution was realised by
dissolving 1 mg of the ethyl
amine ferrocene in 10 L of
acetonitrile and diluted in
100 L of PBS pH 8.5. The
electrode was rinsed with a
pH 7.0 phosphate buffer

solution and stored at 4 C

until use.

S. Bourigua et al. / Biosensors and Bioelectronics 28 (2011) 105111

Scheme 1. Representation of the chemical process in the formation of the biosensor based on functionalised polypyrrole.

were performed using a

Voltalab 80, model PGZ
2.5. TMA extraction from fish
402,
controlled
by
Voltamaster 4 software.
200 g of a sample of horse-mackrel,The
measurements

merlin, were kept at 25 C for 12 h and thenwere performed in a 8

mM solution of PBS
brown with moltar before being stirred with
buffer at pH 7.
water during 1 h. The resulting suspension
solution was centrifuged. The obtained
supernatant solution of the fish extract was2.8. Electrochemical

then filtered and kept in the freezer at 20 Cimpedance

until use. The presence of TMA in sample was spectroscopy (EIS)
confirmed by gas chromatography coupled to
Three electrode cell
the mass spectroscopy using ionisation
impact. Concentration of the extract wasconfiguration was used
measured by HPLC. HPLC was realised usingfor EIS measure-ments.
reversed phase column of C18 and the mobile The impedance analysis
phase was the mixture of acetonitrilewater inwas performed with a
gradient elution mode, with the flow rate of 1 Voltalab model PGZ 402
mL/min.
Meanwhile,
the
acetonitrileimpedance analyzer in
concentration in the mobile phase was linearly the frequency range
increased from 0 to 60% at 2.5 min, the mobile between 100 mHz and
phase composition was kept constant until 3.5100 kHz, and with an
min, and then the percentage of acetonitrile
amplitude of alter-native
was increased to 70% at 10 min and to 100%
voltage of 10 mV.
at 15 min.
Measurements
were
performed in a PBS
buffer solution at pH 7.0.
2.6. Fourier-transform infrared spectroscopy The
measured
(FT-IR)
impedance spectra were
A Brucker IFS 66 spectrometer with MCT
detector was used to acquire infrared spectra.
The spectra was recorded in reflectance mode
with Attenuated Total Reflectance (ATR)
technique using a germanium crystal. The film
was deposited onto the gold surface and the
spectrum was recorded before and after
reaction with the protein FMO3.
2.7. Cyclic voltammetry (CV)
Cyclic voltammetry and amperometric
measurements were done in a conventional
three electrode cell configuration consisting of
2

a working gold electrode (0.11 cm ), a

saturated calomel ref-erence electrode (SCE)
2

and platinum plate (0.5 cm ) as a counter

electrode. All cyclic voltammetry experiments

analysed in terms of
electrical
equivalent
circuits
using
the
analysis
program
Z
view.
3. Results and
discussion
3.1. Construction of the
biosensor and its
characterisation
The
electrodeposition of the
copolymer
poly(Nhydroxyphtalimidepyrrole-co-pyrrole) onto
the surface of the gold
electrode was monitored
using
chronoamperometry
tech-nique.
The
Chronoamperogram
shows an initial abrupt
increase of the recorded
current due to the
diffusion
controlled
monomer oxidation in
solution. It is followed by
a slight increase in
current due to an
electrochemical
nucleation of oligomers
produced in solution and
growth of copolymer
onto the gold electrode.
The schematic steps for
the formation of the
biolayer and the enzyme
recognition are shown in
Scheme 1. After the
electrodeposition of the
copolymer
film,
the
electrode
surface
became
rich
with
activated
ester
(pyNHP).
This
functionalised surface is
able to form covalent
binding to enzyme and
ethyl amine ferrocene by

107

with
the
their reaction with terminal amine group (NH 2 )associated
amide
band
of
the
of both protein and ferrocenyl group. The
modified electrode was analysed by FT-IR toprotein and at 1650
1
associated with
determine the efficiency of their covalentcm
attachment. Spectra were recorded afterthe amide bond of the
polymer
formation
and
after
proteinferrocenyl group.
attachment (figure not shown). Spectra of
polypyrrole bearing activated ester shows the
band assigned to polypyrrole backbone at
Modified
electrode
1

1631, 1564 and 1475 cm corresponding towas

electrochemically
C C stretching vibrations and broad bands atcharacterized
in
1
buffer
1182 and 1134 cm
assigned for NCphosphate
stretching band. Beside well definedsolution at pH 7.0 by
pyrrolidinedione stretching vibration of C Ocyclic voltammetry. Fig.
1
are observed at 1818 and 1784 cm which is1 shows the gold
the signature of the activated ester groupssurface modified by the
functionalised polypyrrole (Korri-Youssoufi etcopolymer layer after
al., 2001). After the covalent attachment of covalent attachment of
protein FMO3 and ferrocenyl group newFMO3 and ferrocenyl
1
stretching band appear at 1526 cm groups. The covalent
108

S. Bourigua et al. / Biosensors and Bioelectronics 28 (2011) 105111

-5

2,0x10

3
.
2
b
.

-5

1,5x10

-5

1,0x10

Current (A)

attachment of ferrocenyl
group lead to the
appearance
of
two
peaks, where the anodic
is at 0.45 V and the
cathodic at 0.25 V
attributed to the redox
couple of ferrocenyl
group. The intensity of
redox wave confirms the
high electroactivity of
ferrocene
groups
attached to conducting
polypyrrole.
Nyquist plot of the
gold surface modified
with
the
poly(Nhydroxyphtalimidepyrrole-co-pyrrole) and
after FMO3 and ethyl

-6

5,0x10

0,0

-6

-5,0x10

-5

-1,0x10

-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6

R
e
s
ap
o
n
s
e
o
f

tection of TMA is
performed
by
following the signal
of the oxidation of
the redox ferrocenyl
groups. This later
reacts
as
redox
probe
which
is
sensitive
to
the
variation of chemical
properties of the
layer. In fact the
enzymatic reaction
of FMO3 involves
the cofac-tor NADPH
and transform it to
+

t NADP . The NADP

h remains bound to
e the enzyme FMO3
and
molecular
b oxygen binds to the
i prosthetic group and
o forms
the
s peroxyflavin species.
e Monooxygenation of
n the substrate and
s
release
of
a
o
molecule of water
r
leaves the flavin in
its oxidized state.
t
+
NADP
o The
molecule dissociates
T and the enzyme is
M ready for another
A catalytic cycle (see
Eqs.
(1)(4))
s (Phillips
and
u Shephard, 2008).
b
s
t
r
a
t
e
T
h
e
d
e

+ 2e

FAD + NADPH + H FADH2 + NADP

Potential (V/SCE)
F
i

(1)

FADH2 + O2 FADHOOH

(2)

FADHOOH + Me3 N FADHOH + Me3 NO

(3)

FADHOH FAD + H2 O
r matic
t electr
a
odes
m
in the
absen
ce
and
prese
nce of
trimet
hylam
ine
were
then
analy
sed
(Fig.
3A)
and
show
ed a
decre
ase in
the
curre
nt
value
of the
oxidat
ion
peak
of
ferroc
enyl
F
group
i
after
3 min
of
cataly
tic
reacti
on of
TMA
with
the
biose
nsor.
The
curre
nt

decreaseto
TMA
was found
is
dependinto be linear
the
g on thein
range
of
TMA
concen- 2.510 g
1

with
tration inmL
regression
the
sample. coefficient
99%.
Differ of
The
ent
amounts sensitivity
of TMAcalculated
the
varying from
of
between slope
2.5 andthe
50
gcalibration
curve
is
1
mL
evaluated
were
to
6
tested
2
( A/cm )/
with the
1
.
biosenso g mL
r. UponThe
addition detection
of
thelimit of 0.4
1
TMA, forg
mL
all
the
was calcumodified
lated
electrode
taking into
s,
the
account
current
the signal
decrease
to
noise
s
ratio of 3.
continuo
This value
usly as a
is
lower
function
than
the
of TMA
detection
concentr
limit
ation
demonstra
(Fig. 3B).
ted
in
The
others
response
biosensors
of
of
TMA
enzyme
based on
modified
FMO3. For
electrode
example,
formed
within the
with
catalytic
poly(pyrr
reac-tion
oleof FMO3,
FMO3the TMA
ferrocen
was
e-comonitored
pyrrole)

(4)
by
feroocenyl
measuring groups as
the oxygenredox
consumed probe.
during the
Trimeth
reaction by
ylamine
clark
FMO3
oxygen
electrode catalytic
and
areaction
detection was
limit of 1monitored
mM (59 gby
1
mL
) (impedance
Mitsubaya spectrosco
py as this
shi et al.,
analytical
2004) wasmethod
demonstra
could be
ted.
more
sensitive
A
than
highly
amperome
sensitive
tric
amperome
measurem
tric
ents (Tlili
biosensor et
al.,
could be2008).
designed Measurem
thanks,
ents have
firstly,
tobeen
the
achieved
developed with
approach varying
consisting frequencie
s from 100
in
to
immobiliz- mHz
ing
the100 kHz in
PBS
FMO3
enzyme onsolution at
pH
7.
the
polypyrrole Systematic
ally,
the
matrix
modified
which
electrode
preserves
was
its catalytic
equilibrate
properties
d with a
and,
range
of
secondly, concentrato the usetions of the
of
TMA from

0.4 to 160
1

g mL
.
The
impedance
spectra
(Nyquist
plot)
obtained
after
addition of
different
concentrati
ons
of
TMA are
shown in
Fig.
4A
where
Re(z)
is
the
real
part
and
Im(z) is the
imaginary
part of the
complex
impedance
Z.
Immedi
ately,
at
low
frequencie
s,
the
impedance
modulus |
Z(f)|
increases
clearly with
the
increase of
TMA
concentrati
on, which
indicates
that
a
larger
amount of
TMA has
reacted on
the
surface.

S. Bourigua et al. / Biosensors and Bioelectronics 28 (2011) 105111

-5

2,0x10

a
-5

1000

b
c

1,5x10

800

-5

Current A

- Im (Z) (Ohm.cm )

1,0x10

-6

5,0x10

0,0

-6

-5,0x10

-1,0x10

600

400

200

-5

-0,4

-0,2

0,0

0,2

0,4

abc

0,6
0

Potential (V/SCE)

ij

200

400

600

g
800

h
1000

Re (Z) (Ohm.cm2)

2,7
2,6

5
2,5
2,4

2,3

log Z

current A

2,2

2,1

2,0

-1

1,9

10

20

30

40

50
-20

-1

TMA concentration (g.mL )

previously which was

Fig. 3. (A) Cyclic voltammograms of poly(Py-FMO31
ferrocene-co-py) in the absence
from 2 to 40 g mL
(a) and (be) presence of various concentration TMA (b) 5 g
1
1
(Fillit et al., 2008).
mL , (c) 10 g mL ,
1
1
(d) 20 g mL , (e)50 g mL . The measurement was realised3.3. Optimisation of
in PBS solution at scan rate of 0.1 V/s (B) The relationshipbiosensor
between the measured current and the trimethylamine
concentration at fixed potential of 0.45 V/SCE.

In order to optimize
the
sensitivity
of
This reaction leads to a decrease in the kinetic of
electrochemical
electron transfer of the redox probe ferrocenyl
measure-ment
for
group to the electrode which induces an increase detection of TMA, the
in the charge transfer resistance of the layereffect
of
various
which is in concordance with the decrease inparameters on the
amperometric measurement. The calibrationamperometric
curves for TMA detection obtained by measuring biosensor
response
the value of modulus |Z(f)| at frequency range ofwas analysed. The
0.5 Hz where the max-imum variation was effect
of
the
obtained was shown in Fig. 4B. Linear range forconcentration
of
1
TMA detection was obtained from 0.4 g mL toenzyme solution was
investigated
by
1
10 g mL
with coefficient regression of 0.95 varying the FMO3
which is the same as previous amper-ometricenzyme
measurements. However enhanced of dynamicconcentrations
from
range of TMA detection, corresponding to the0.5 to 1 mg mL1
response of the sensor before it reach saturation,
during the covalent
was obtained by the impedance method which is
attachment.
Results
1
observed in the range of 0.480 g mL . Thisshow that a larger
range is larger than that obtained in the case of linear range of the
amperometric
detection
and
also
for trimethyleamine
conductimetric
TMA
biosensor
developeddetection is obtained
with the concentration

of

20

40

[TM
A]
(g/
ml)

60

80

100

120

140

160

180

Fig. 4. (A) Nyquist plot of poly(Py-FMO3- ferrocene-co-py) temperature on the

during reaction of var-ious trimethylamine concentration in activity of the FMO3
1
1
solution PBS pH 7.0, (a) 0 g mL TMA; (b) 0.4 g mL TMA;enzyme
in
this
1
1
1
(c) 1 g mL TMA; (d) 2 g mL TMA; (e) 5 g mL TMA; (f) 10biosensor
was
1

g mL ; (g) 20 g mL TMA; (h) 40 g mL TMA; (i) 80 g mL determined using 8

1
TMA; (j) 160 g mL TMA. Spectra were recorded in the mM phosphate buffer
frequency range between 100 mHz and 100 KHz, using a at
pH
7.0.
The
modulation voltage of 10 mV. (B) Calibration curves of the
response
of
the
TMA biosensor obtained from impedance spectra.

enzyme electrode was

mea-sured
at
FMO3 of 0.5 mg mL during the biosensor set-temperatures between
up. The excess of the immobilised protein may10 and 40 C. The
leads to enzyme agglomeration and then to a immobilised FMO3 in
limited access to its active site.
polypyrrole
layer
The effect of the total phosphate concentrationshowed
maximum
in buffer solu-tion on the response of the enzyme

activity at 20 C. The
electrode was determined by keeping the pH
thermal denaturation
value at 7.0. The concentration effect of the bufferof the enzyme caused
solutions was studied between 4 and 32 mM. The a decrease in the
optimum working value of the buffer solution was response
of
the
found as 8 mM.
enzyme electrode for
Enzymes are proteins that become denaturedhigher temperatures.
at high temper-atures. The effect of the
1

3.4. Reproducibility
and storage stability of
the biosensor
The reproducibility
of the biosensor and
the response of the
enzyme
electrode
were
investigated
within
6
modified
electrodes
and
a
standard deviation of
11% was obtained.
Comparing
the
reproducibility of our
sensor
to
those
described in literature,
the same value was
obtained
with
potentiometric
biosensor based

110

S. Bourigua et al. / Biosensors and Bioelectronics 28 (2011) 105111

2,0x10
1,5x10

Current A

1,0x10
5,0x10

con-centration of 500 L
extract solution was diluted
in 5 mL PBS solution and
measured
within
the
biosensor. Fig. 5 shows the
results obtained after 5 min
of the reaction of the fish
extract sample with the
biosensor. Decrease in
redox activity of ferrocenyl
group was observed as
demonstrated
previously
with TMA sample. The
amperometric
analysis
allows evaluation of the
concentration
of
TMA

-5

-5

-5

-6

0,0
-5,0x10
-1,0x10
-1,5x10

-6

-5
-5

-0,4

-0,2

0,0

extract to 2 g mL .
To
confirm
the
potenti presence of TMA in fish
al
extract analysis of fish
V/SCE
sample was performed by
Fig. 5. Cyclic voltammograms of poly(Py-FMO3- GC/MS and compared to
ferrocene-co-py) in the absence
pure TMA. GC shows a
(a) and (b) with extract of 0.5 mL of TMA diluted in high peak obtained at
5 mL of PBS solution.
retention times of 1.4 in
both sample pure TMA and
on microbial sensor where relativefish extract corresponding
standard deviation was deter-mined to to retention time of TMA.
be 14% (Gamati et al., 1991). HighThis peak is the most
reproducibility are generally obtained important observed in the
with
volatile
compounds
when GC curve of fish extract
derivatization procedure was performedcompared to the other
which maintain the amine more stableobtained
peaks.
The
and less likely to be lost as gases during identification of volatile
measurement and storage (DaCosta etamine in fish sample was
al., 1990). These methods of detection confirmed by the presence
are not direct and need the chemicalof ion fragment at m/z of
reaction
before
measurement
by58 for TMA in both TMA
coupling volatile amine with other sample and in fish extract
figure
in
products such as 9-fluorenylmethyl chlo-(see
roformate or ethyl bromoacetate. Highsupplementary
reproducibility was also obtained wheninformation).
To
evaluate
the
flow system was used where the
concentration of volatile compound wasaverage recovery of such
sensor for TMA eval-uation
maintained stable during measure-ment (
in fish sample. The TMA
Mitsubayashi et al., 2004). Thewas evaluated by HPLC as
reproducibility obtained within this sensor another method. The TMA
fish
extract
was
using direct method without anyin
derivatiza-tion procedure and any flowmeasured with the present
system is acceptable for a volatile sensor and compared with
the value obtained with
compound.
HPLC. For this
The operational lifetime of the
biosensor, defined as the period during
which the sensitivity remains constant,
was also investigated by checking
periodically with the same sensor, its

purpose, a sample of fish

extract with unknown TMA
concentration was used for
the measurement and the
obtained value was compared to the calibration
curve
determined
by
measuring
known
concentrations of TMA
1
between 1 and 10 g mL .
TMA in fish sample was
found to be 1.9 0.1 g
1
mL . Comparative value
of TMA measured with
1
present sensor of 2 g mL
against HPLC as reference
method
demonstrates
average recovery of 105%.
4. Conclusion
In this study, we have
described,
the
development of a new
enzyme
biosensor
for
trimethylamine
detection
based on the covalent
attachment of the flavin
containing monooxygenase
and ferrocenyl groups to an
electropolymerized
polypyrrole
film.
We
demonstrated
that
the
ferrocenyl groups are able
to act as redox probe and
monitor
the
catalytic
reaction of FMO3 to TMA.
Amper-ometric biosensor
of TMA was obtained
leading to a sensitivity of 6
1

A g mL cm . Monitoring
the
TMA
by
electrochemical impedance
spectroscopy shows the
same
sensitivity
and
improves
the
dynamic
range of TMA detection. In
fact, a low detection limit of
1

0.4 g mL of TMA and a

dynamic range from 0.4 to
1

80 g mL
were obtained
by impedance.
This novel biosensor for
TMA analysis with high
sensitivity, rapid response
and large dynamic range
seems to be a very promising analytical tool for the
evaluation of fish freshness
in real samples.

amperometric sensitivity for 10 g mL of

TMA. The biosensor shows a decrease
of 10% of it response after 5 days and
remains stable during two weeks.

3.5. Evaluation of fish freshness

The evaluation of fish freshness was
realised by measuring the extract sample
from horse-mackerel fish. The unknown

Acknowledgements
Sondes Bourigua is
thankful and grateful to the
Ministry of Education of
Tunisia for the alternation
grant and to Rhone-Alpes
Region for the grant in the
framework of MIRA 2008
program.
The
authors

thank
EC
for
support
through
NANOBIOSENS Project (PIRSES-GA2008-2300802). The authors thank Dr. D.
Clainquart from UFR de Chimie
University Paris-Diderot for GC-MS
measure-ment and Dr. L. Salmon from
ICMMO
University
Paris-Sud
for
assistance in HPLC measurement.
Appendix A. Supplementary data
Supplementary data associated with
this article can be found, in the online
version,
at
doi:10.1016/j.bios.2011.07.005.
References
Ahuja, T., Mir, I.A., Rajesh Kumar, D.,
2007. Biomaterials 28, 791805.
Cashman, J.R., 2005. Biochem.
Biophys. Res. Commun. 338, 599604.
DaCosta, K., Vrbanac, J.J., Zeisel, S.,
1990. Anal. Biochem. 187, 234239.
Dhaouadi, A., Monser, L., Sadok, S., Adhoum, N.,
2007. Food Chem. 103, 10491053. Fillit, C., JaffrezicRenault, N., Bessueille, F., Leonard, D., Mitsubayashi,
K. Tardy, 2008.

J. Mater. Sci. Eng. C 28, 781786.

Gamati, S., Luong, J.H.T., Mulchandani, A., 1991.
Biosens. Bioelectron. 6, 125131. Gill, T.A., 1992. In:
Huss, H.H., Jakobsen, M., Liston, J. (Eds.), Quality
Assurance in the

Fish Industry. Development in Food Science,

vol. 30. Elsevier, Amsterdam. Godillot, P., KorriYoussoufi, H., Srivastava, P., El Kassmi, A.,
Garnier, F., 1996. Synth.
Met. 83, 117123.
Hamada-Sato, N., Kazushige, U., Kobayashi, T.,
Imada, C., Watanabe, E., 2005. Food Control
16, 301307.
L, H.Q.A., Sauriat-Dorizon, H., Korri-Youssoufi, H.,
2010. Anal. Chim. Acta 674, 18. Li, N., Endo, H.,
Hayashi, T., Fujii, T., Takai, R., Watanabe, E., 1994.
Biosens. Bioelectron.

9, 593599.
Mitsubayashi, K., Kubotera, Y., Yano, K.,
Hashimoto, Y., Kon, T., Nakakura, S., Nishi, Y.,
Endo, H., 2004. Sens. Actuators B 103, 463
467.
Mitsubayashi, K., Hashimoto, Y., 2002. Sens.
Actuators B 83, 3540.
Mushiroda, T., Yokoi, T., Takahara, E., Nagata, O.,
Kato, H., Kamatakia, T., 1999. J. Chromatogr.
B 734, 319323.
Ozogul, Y., Ozyurt, G., Oogul, F., Kuley, E., Polat, A.,
2005. Food Chem. 92, 745751. Pacquit, A., Lau, K.T.,
McLaughlin, H., Frisby, J., Quilty, B., Diamond, D.,
2006. Talanta

69, 515520.
Pena-Pereira, F., Lavilla, I., Bendicho,
C., 2010. Food Chem. 119, 402407.
Phillips, I.R., Shephard, E.A., 2008.
Trend Pharmacol. Sci. 29, 294301.
Sadok, S., Uglow, R.F., Haswell, S.J.,
1996. Anal. Chim. Acta 32, 6974.

S. Bourigua et al. / Biosensors and Bioelectronics 28 (2011) 105111

Sallam, K.I., 2007. Food Chem. 101, 592600.

Sheldon, R.A., 2007. Adv. Synth. Catal. 349,
12891307.
Siripatrawan, U., Sanguandeekul, R., Narakaew,
V., 2009. LWT - Food Sci. Technol. 42, 343
349.

111
Suska, A., Ibnez, A.B.,
Lundstrm, I., Berghard, A.,
2009. Biosens. Bioelectron.
25, 715720.
Tlili, C., Jaffrezic-Renault, N.,
Martelet, C., Korri-Youssoufi,
H., 2008. Mater. Sci. Eng. C
28, 848854.