Biosensors and Bioelectronics 24 (2009) 34613466

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

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

Highly sensitive amperometric immunosensor for the detection of

Escherichia coli
K. Abu-Rabeah a,1 , A. Ashkenazi a,1 , D. Atias b , L. Amir c , R.S. Marks a,d,e,
a

Department of Biotechnology Engineering, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel
Department of Virology, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel
Department of Chemistry, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel
d
National Institute for Biotechnology in the Negev, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel
e
ILSE Katz Center for Meso and Nanoscale Science and Technology, P.O. Box 653, Beer Sheva 84105, Israel
b
c

a r t i c l e

i n f o

Article history:
Received 28 February 2009
Received in revised form 25 April 2009
Accepted 27 April 2009
Available online 3 May 2009
Keywords:
Escherichia coli (E. coli)
Electrochemistry
Immunosensor
-d-Galactosidase
Alginate-polypyrrole (Alg-Ppy)

a b s t r a c t
With the aim of detecting rapidly the presence of Escherichia coli (E. coli), a disposable amperometric
immunosensor was developed based on a double layered conguration at the transducer surface, consisting rst of a polypyrrole-NH2 -anti-E. coli antibody (PAE) inner layer followed by an alginate-polypyrrole
(Alg-Ppy) outer packing layer. In the presence of the substrate p-aminophenyl -d-galactopyranoside
(PAPG), the bacterial enzyme, -d-galactosidase produces the p-aminophenol (PAP) product, also generating an amperometric signal due to PAP electrooxidation by potentiostating the glassy carbon (GC)
electrode at 0.22 V. The operational procedure consists in rst adding the test sample containing the
bacteria, then coating it with Alg-Ppy to ensure the connement of the released enzyme and the analyte
(being generated by the enzymatic catalysis) to the electrode active surface. This procedure facilitates
the diffusion of the substrate within the complex and thus creates a higher oxidation level of the PAP
enabling a detection limit of 10 colony forming units (CFU)/ml. The immunosensor setup demonstrates
an improved detection limit of more than 10 times less bacteria detected than other immunosensing
techniques without the need for multi step pretreatments of the test sample and/or incubation as found
in some of the existing methods.
2009 Elsevier B.V. All rights reserved.

1. Introduction
Detection of microorganisms is important in food and water
safety, while the presence of Escherichia coli (E. coli) is used as a
potential marker for the presence of pathogens originating from
humans and warm-blooded animals (Tryland and Fiksdal, 1998).
Estimating the number of coliforms is essential due to enteric disease, such as enterohemorrhagic strains of E. coli (Buchanan, 1997;
Tokarskyy and Marshall, 2008) contracted from food or polluted
coliform water supplies which still constitute public health concerns (Lin et al., 2008; Tokarskyy and Marshall, 2008).
Conventional procedures for E. coli detection and monitoring
are mainly based on cultures grown on differential agar media followed by counting the number of target organisms in the sample,
a procedure that can take 13 days (George et al., 2000; Lin et al.,
2008). The time-consuming drawback increases the motivation for

Corresponding author at: Department of Biotechnology Engineering, BenGurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel.
Tel.: +972 8 6477182; fax: +972 8 6472857.
E-mail address: rsmarks@bgu.ac.il (R.S. Marks).
1
Both authors contributed equally to this work.
0956-5663/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.bios.2009.04.042

the development of rapid and sensitive methods aiming at estimating E. coli (nucleic acid assays and immunological techniques
with chromogenic, uorogenic, or chemiluminogenic substrates)
(Datsenko and Wanner, 2000; Feng and Hartman, 1982; Gehring et
al., 2004; Loge et al., 2002), usually showing a good linear correlation with those obtained from traditional methods (George et al.,
2000; Tryland and Fiksdal, 1998; Venkateswaran et al., 1996).
Biosensors are also being developed for the detection of
pathogenic bacteria including E. coli (Ivnitski et al., 1999; Leonard
et al., 2003; Tokarskyy and Marshall, 2008). Hence biosensors
using electrochemical (Abdel-Hamid et al., 1999a,b; Lin et al., 2008;
Mittelmann et al., 2002; Palenzuela et al., 2004), piezoelectric (Su
and Li, 2004), acoustic (Howe and Harding, 2000) and optical (Pyle
et al., 1995) transducers were applied for E. coli detection. The
number of electrochemical biosensing platforms (Bakker, 2004;
Mehrvar and Abdi, 2004) for the detection of coliform contamination have grown rapidly. For example, endogenous enzymatic
activity such as -d-galactosidase which participates in lactose
metabolism was used as a general marker for monitoring coliforms
(Mittelmann et al., 2002; Tryland and Fiksdal, 1998), while some
studies have used indirect sandwich enzyme-linked immunoassay
with amperometric immunosensing (Abdel-Hamid et al., 1999a,b;
Lin et al., 2008). In all these methods, E. coli bacteria could be

3462

K. Abu-Rabeah et al. / Biosensors and Bioelectronics 24 (2009) 34613466

detected at ranges of concentrations between 102 and 108 CFU/ml

without preincubation of the original sample. There is still room to
improve on the detection limits of environmental samples.
We report an alternative immunosensing approach for the
amperometric detection of E. coli. The method is based on electropolymerization and coating of glassy carbon (GC) electrodes
with polypyrrole-amine (Ppy-NH2 ). The immobilization of the antiE. coli antibody is carried out through carbodiimide chemistry
(Abu-Rabeah et al., 2005) later enabling the specic attachment of
E. coli. This step is followed by induced bacterial lysis and the subsequent release of -d-galactosidase which in the presence of PAPG
produces PAP that is then oxidized on the electrode thereby creating a current (Mittelmann et al., 2002). The system is entrapped
in an Alg-Ppy matrix (Ionescu et al., 2005) which will conne the
enzymatic catalysis to the electrode surface facilitating the diffusion and oxidation of the PAP. Electropolymerized Alg-Ppy lls the
porous alginate with polypyrrole laments reducing possible leaching of bacterial contents that was made available by lysis. Use of a
secondary antibody was not necessary because detection was based
on the activity of the intrinsic -d-galactosidase enzyme. This system was tested with a model E. coli strain K-12 MG1655, however,
introducing the appropriate capture antibodies would allow for the
detection of any wild type strains of E. coli including O157:H7. An
immunosensor that is very sensitive, specic, rapid, simple to fabricate and cost effective was thus developed.
2. Experimental

PGSTST30 electrochemical workstation: a conventional electrochemical cell (Metrohm) consisting of GC disk electrodes as
working electrode (diameter 3 mmpolished with 1 m diamond
paste MECAPREX Press PM), a saturated AgAgClKCl electrode
(Ag/AgCl) as reference electrode and a platinum wire as counter
electrode. The data acquisitions were achieved using GPES and FRA
software.
2.4. Permeability studies
The permeability of a GC bare electrode, coated with Ppy-NH2 ,
102 and 106 CFU/ml coatings were examined by a rotating-disk electrode (RDE) equipped with a rotation controller. The experiments
were conducted at different rotation speeds using FeCN (2 mM) in
0.1 M TrisHCl (pH 7) as a redox probe.
The permeability (Pm ) of the coatings was estimated using RDE
experiments at different rotation rates. The results were analyzed
and examined by applying Eqs. (1)(4) (Gough and Leypoldt, 1979).
A description of the variation of the steady state limiting current Ilim
is shown, with the mass transport for a RDE coated with different
electro-inactive membranes.
1
1
1
=
+
Ilim
Is
Im
2/3

Is = 0.62nFAC 0 DS

2.2. Bacterial culture

Bacterial strains of wild type E. coli K-12 MG1655, E. coli BW25113
with no -d-galactosidase activity and Pseudomonas aeruginosa
were grown over night in Luria Bertani (LB) medium at 37 C with
shaking. The experiment was begun by cultures inoculated into
Erlenmeyer asks containing 10 ml of LB medium, supplemented
with IPTG to a nal concentration of 0.5 mM. These cultures were
grown during 22.5 h at 37 C while shaking until reaching logarithmic phase. The cultures were then diluted with phosphate-buffered
saline (PBS), pH 7.2, to concentrations ranging from 1 CFU/ml up to
107 CFU/ml and analyzed later by the electrochemical amperometric technique.
2.3. Electrochemical instrumentation
The voltametric and the amperometric measurements as well
as the electropolymerization process were carried out using a

1/6 1/2

(2)

Im =

nFAKC 0 Dm
= NFC 0 Pm

(3)

Pm =

KDm
(cm s1 )
d

(4)

2.1. Chemicals and reagents

PAPG (A9545), isopropyl -d-thiogalactopyranoside (IPTG)
(16758),
N-ethyl-N -(3-dimethylaminopropyl)carbodiimide
hydrochloride (EDAC) (E1769), and TrisHCl (77-861) were
purchased from Sigma; pyrrole (py) (13170-9), lithium perchlorate
(431567), sodium p-toluene sulfonate (NaPTS) (152536) and 1,1 ferrocenedicarboxylic acid (FeCN) (1293-87-4) were purchased
from Aldrich; lysozyme (Amresco 0663), perchloric acid (17931)
and hydroxy-2,5-dioxopyrrolidine-3-sulfonicacid sodium salt
(sulfo-NHS) (5648) were purchased from Fluka; acetonitrile (7505-8) was purchased from Bio Lab; Rabbit anti-E. coli polyclonal
antibody (BO357) was purchased from Dako; Sodium alginate
(Alg) Protanal LF 10/60 (Laminaria hyperborea, average molecular
weight of 128 kDa, viscosity of 1% (w/v) solution is 39 cP) was
provided by FMC Biopolymer (Norway); and Alg-py and py-NH2
were synthesized according to the protocols described in literature
(Abu-Rabeah et al., 2005).

(1)

Terms Ds and Dm are the diffusion coefcients of the substrate in

the bulk solution and in the membrane respectively; , the kinematic viscosity of the solution; , the rotation rate of the RDE; , the
thickness of the membrane; K, the partition equilibrium constant
of the substrate between solution and membrane; A, the electrode
surface; n, the number of exchanged electrons and C0 , the substrate concentration. Eq. (1) is composed of two terms, where the
rst represents the current ow under the same conditions, in the
absence of a membrane, and is therefore characteristic of the diffusion of the substrate in the bulk solution (Levich current, Eq. (2)).
The second term accounts for the diffusion of the substrate in the
membrane and depends on the product of the partition equilibrium
constant K, the diffusion constant of the substrate in the membrane and the lm thickness. Only the rst term of the equation
is dependent upon the rotation rate of the RDE. Therefore, a plot of
1/Ilim versus 1/1/2 (Fig. 2) presents a linear behavior with the same
slope as for a bare electrode with a positive intercept, whose value
depends on the permeability Pm of the membranes (Eqs. (3) and
(4)). The relative deviation of the values was <5%. AFM micrographs
were taken with AFM systems manufactured by Nanonics Imaging
Ltd., Jerusalem.
2.5. Amperometric detection of -d-galactosidase activity
-d-Galactosidase activity was detected using PAPG as a substrate. The enzyme catalyzed PAPG to form PAP which is then
oxidized at the electrode surface. The amperometric detection was
carried out in 15 ml of an aqueous solution of 0.1 M TrisHCl buffer
electrolyte (pH 7). The modied electrode was potentiostated at
0.22 V versus the reference electrode until reaching equilibrium.
To start detection, the substrate PAPG in nal concentration of
0.8 mg/ml was introduced and the current resulting from -dgalactosidase activity was collected and measured by PGSTAT 30
equipped with GPES4 software.

K. Abu-Rabeah et al. / Biosensors and Bioelectronics 24 (2009) 34613466

2.6. Preparation of the amperometric E. coli immunosensor

The construction of the amperometric immunosensor consisted
in: (1) electropolymerizaion of the Ppy-NH2 on the surface of the
working electrode by applying a constant voltage of 0.93 V in a
15 ml solution composed of 0.1 M py-NH2 , 0.1 M pyrrole in double distilled water and acetonitrile (volume ratio 7:3 respectively),
0.1 M NaPTS and 1 M perchloric acid. Polymerization was monitored by charge accumulation on the electrode surface, set around
Q = 3 C and also by the formation of a thick and stable polymer
layer; (2) linking of anti-E. coli antibody to Ppy-NH2 by incubation of the polymerized electrode in 1 ml PBS solution containing
EDAC, 0.3 mg/ml and sulfo-NHS, 0.3 mg/ml and 30 l of polyclonal
antibody for 2 h at 37 C with mild shaking before a PBS wash.
The amount of linked antibody was calculated while considering
the charge density and the surface area of the electrode; (3) the
antibody-linked electrode was incubated in 1 ml solution of E. coli
in different concentrations for 1 h at 37 C with gentle shaking and
then washed with TrisHCl buffer; (4) lysis of the attached E. coli
was performed with 50 l of lysozyme solution (100 g/ml) during 1 h at 37 C enabling the release of -d-galactosidase from the
bacteria. Both lysis and antibody attachment procedures were performed in optimized conditions; (5) creation of an entrapment layer
on the surface was achieved by dropping 20 l of Alg-py (2% (w/v)
in 0.1 M TrisHCl buffer) and then cross-linking for 30 min with
a further drop of 10 l of 0.1 M CaCl2 . Adding a volume greater
than 20 l of Alg-Py will lead to an unstable entrapment layer;
(6) the Alg-Py matrix was electropolymerized on the surface of
the working electrode by applying a constant voltage of 0.93 V
in 15 ml of 0.1 M lithium perchlorate aqueous solution. Polymerization time was around 10 min and reaches a steady charge of
Q = 90 mC. These immunosensor working electrode probes were
then used in the amperometric detection of -d-galactosidase
activity to estimate the bacterial concentrations of the test samples.
3. Results and discussion
The new disposable immunosensor conguration exhibits a
double layer conguration in the transducer surface, namely, a
PAE inner capture layer and an Alg-Ppy outer connement layer
described in Fig. 1. The operational procedure consists in rst
adding the sample, and if the target bacteria are present then they
are specically attached by the PAE layer, and thereafter will be
coated with Alg-Ppy to conne the released enzyme and the ana-

3463

lyte (being generated by the enzymatic catalysis) to the electrode

active surface.
3.1. Electrochemical characterization of poly(Alg-py-PAE)
Electrochemical impedance spectroscopy (EIS) was used as a
sensitive method to probe the interface properties of surfacemodied electrodes (Geng et al., 2008), where the voltage
was set at 0.93 V to retain the polymer conductivity. Different
electrode congurations were investigated in the presence of
Fe(CN)6 3 /Fe(CN)6 4 as a redox probe, thereafter the values were
tted by randles circuit based on Nyquist plots over the frequency
range 1105 Hz, and presented in Fig. 2. The high frequency (low
impedance) semicircular regions in Ppy-NH2 modication have a
very small radius of curvature, indicating low resistance of the
charge transfer (Fig. 2(1a), inset) (R(p) = 0.5 103 ). But that of the
electrode coated with PAE (Fig. 2(1b), inset) showed an expected
higher value (R(p) = 1.7 103 ) resulting from the immobilization of macromolecules on the transducer surface which inhibit
the redox probe diffusion and increase the charge transfer resistance. The impedance values increased signicantly when the
electrode was treated with the entrapping layer of the Alg-Ppy
(Fig. 2(1c)). This setup showed higher charge transfer resistance
(R(p) = 4 103 ) than the setup without Alg-Ppy layer (Fig. 2(1b)).
The impedance of the setup with higher bacteria concentration
(106 CFU/ml) and entrapping Alg-py layer showed the highest
charge transfer resistance (R(p) = 9 103  (Fig. 2(1d)). It is noteworthy that the increase in the charge transfer resistance could be
correlated directly to diffusion limitations through the polymeric
coating matrix. This diffusion barrier will conne the resulting
products of the enzymatic catalysis for oxidization at the transducer
surface.
3.2. Permeability and AFM study
The permeability of the immunosensor with different concentrations of E. coli in the presence of FeCN as a redox probe
was measured showing the following results: 3.5 101 , 2 102
and 16 103 cm s1 for GC coated with Ppy-NH2 , 102 and
106 CFU/ml. respectively. The permeability of the electrode coated
with 102 CFU/ml (Fig. 2(3c)) was 13 times lower than the value calculated for the electrode coated with Ppy-NH2 only (Fig. 2(3a)), this
could be explained by the existence of a higher concentration of
bacteria reducing the diffusion of the redox probe to the oxidizing
sites on the transducer surface. The permeability of the electrode

Fig. 1. Schematic diagram of the fabrication and chemical processes of the electrochemical immunosensor.

3464

K. Abu-Rabeah et al. / Biosensors and Bioelectronics 24 (2009) 34613466

Fig. 2. (1) Impedance spectra of the immunosensor with different concentrations of E. coli in the presence of Fe(CN)6 3 /Fe(CN)6 4 as a redox probe. Inset plot: Impedance
spectra of: (a) electrode coated just with Ppy-NH2 . (b) Electrode coated with Ppy-NH2 , antibody and 106 CFU/ml of E. coli (no lysozyme and Alg-Ppy matrix). (c) 102 CFU/ml.
(d) 106 CFU/ml. (2) Equivalent electrical circuit of EIS data. (3) Permeability measurements in the presence of 2 mM FeCN in 0.1 M TrisHCl (pH 7) for GC bare electrode, GC
electrode coated with Ppy-NH2 , GC electrode coated with 102 CFU/ml and 106 CFU/ml and entrapped in Alg-Ppy layer. (4) AFM micrographs of (4.1) alginate coated in GC
electrode and (4.2) Alg-Ppy coated and polymerized upon GC electrode.

coated with 106 CFU/ml was further reduced (Fig. 2(3d)) by the
increase of the biological material, however the permeability value
still enabled a good diffusion compared to the values presented for
an electrode coated with only Ppy-NH2 . The permeability values
were further supported by the impedance charge resistance values
described before.
The electrode coated with alginate and Alg-Ppy was probed by
AFM techniques (Fig. 2(4)). The layer of Alg-Ppy was shown to be
more compact with smooth morphology and less thickness than
the layer of natural unmodied alginate. The dense morphology of
the Alg-Ppy tightens the polymeric matrix on the electrode surface and decreased the permeability as shown before. However,
the Ppy chains within the matrix improve the charge transfer and
present superior platform when compared with unmodied alginate.

Hamid et al., 1999a,b; Lin et al., 2008; Mittelmann et al., 2002).

Reducing the number of bacteria to 1 CFU/ml gave a 35% overlap
between the developed signal and the background noise of the
instrument. The experiments were run at three independent times
with immunosensors that were made in different days and still
showed good reproducibility.

3.3. Immunosensor calibration and sensitivity

Various E. coli concentrations (1107 CFU/ml) were tested and
investigated by the immunosensor probe to set a correlation
between the current level and the bacterial concentrations in order
to explore the sensitivity of the applied approach. These results
are depicted in Fig. 3. The linear correlation coefcient (0.95)
shows high matching between both current and concentration
when it was drawn on a logarithmic scale. The detection limit
of the immunosensor was 10 CFU/ml, which is sensitive to detect
the concentration of more than 10 times less bacteria than other
immuno-sensing techniques that were published recently (Abdel-

Fig. 3. Calibration curve of E. coli ranging from 1 CFU/ml until 107 CFU/ml (lled
squares). The current values were taken after 3 min from the addition time of PAPG.
Background levels of the current were also measured (lled triangle).

K. Abu-Rabeah et al. / Biosensors and Bioelectronics 24 (2009) 34613466

Fig. 4. Amperometric detection of 107 CFU/ml E. coli K-12 MG1655 () and
107 CFU/ml E. coli BW25113 (with no -d-galactosidase activity) () were analyzed
by the immunosensor with PAPG as a substrate. Inset plot: Control experiments
were preformed by the amperometric detection of 107 CFU/ml E. coli. K-12 MG1655
without lysozyme ().

3.4. Immunosensor performance in identifying E. coli cultures

PAPG was used as a substrate for -d-galactosidase, the potential
was set at 0.22 V, and the current response recorded as depicted in
the amperometric curves in Fig. 4. The resulting curves with a concentration of bacteria of 107 CFU/ml showed that the fast current
developed by the catalytic reaction of the enzyme clearly indicates a high current level measured after 3 min (1 A) that could be
directly related to the special construction of the immunosensor.
To check the high specicity of the E. coli immunosensor for -dgalactosidase activity, experiments were performed with a mutant
strain of E. coli (107 CFU/ml) lacking -d-galactosidase activity and
it showed no current (Fig. 4). E. coli that were analyzed by the
immunosensor without lysozyme (Fig. 4, inset), presented a similar current response shape to that of E. coli with lysozyme, but
approximately 600 times lower. The activity of -d-galactosidase
is dramatically increased from its basal levels by the lysis of the
cells followed by a large amount of secreted enzymes, contributing
to the sensitivity of the immunosensor.
The importance of the different modications to the
immunosensors sensitivity was further supported by running
several controls to examine the applied procedure (Table 1). The
chemical attachment of the antibody to the Ppy-NH2 lm polymerized onto the electrode surface and attachment of an anti-E.
coli antibody was examined by running a control experiment
without the coupling reagents, sulfo-NHS and EDAC. The resulting
current (1.5 nA) implied that no bacteria were attached as the
activating linkers and therefore also the capture antibodies were
Table 1
Effect of the different modications on the response current in the electrochemical
immunosensor.
Controls
Ppy-NH2
Anti-E. coli antibody
Sulfo-NHS and EDAC
E. coli K-12 MG1655 (107 CFU/ml)
Pseudomonas aeruginosa (107 CFU/ml)
Lysozyme
Alg-Ppy matrix
Current (nA)

+
+

+
+
1.5

+
+
+
+

+
1.7

Sensor
+
+
+
+

283

+
+
+

+
+
+
11.5

+
+
+
+

+
+
1000

Current values were taken after 3 min from the addition of the substrate PAPG to the
monitoring system.

3465

missing and all the surface was washed away. Similarly, when
we used P. aeruginosa that is phylogenetically remote from E. coli,
current levels dramatically decreased. Omitting the entrapping
layer of Alg-py caused a decrease of 3.5 times in the current
when compared to the complete modied sensor. In order to
increase assay sensitivity prior to the assay, it would be desirable
to concentrate the bacteria into a smaller volume which requires a
long period of time for such a multistep procedure. We sought to
increase sensitivity by coating the electrode with an Alg-Ppy layer
assigned to better entrap the enzyme onto the electrode surface
and conne the enzymatic reaction in close proximity to the
electrode surface (Ionescu et al., 2005). This coating layer is aimed
at increasing the sensitivity of the obtained current by preventing
the diffusion of the catalytic products (PAP) out into the phase
solution being agitated around and to allow them to be oxidized
with greater probability at the GC surface. Time is thus saved by
avoiding concentrating the bacteria. In addition, both polypyrrole
networks formed by electropolmerization and coating (Alg-py) at
the electrode surface, strengthen the structure mechanically from
moderate stirring.
In many analytical systems, the bacterial separation and
measurement are performed in several steps with a manual
transfer step in between resulting in sample loss and errors
due to contamination. Our approach does not require separation
(electrophoresis, chromatography or immunoltration) which dramatically improves speed, while still being selective and sensitive
with minimal sample manipulation and a short measurement time
(3 h) from the moment the modied electrode has been exposed
to the bacterial sample. The longtime stability of the immunosensor activity was examined in this study too. The polarization time
of the sensor was 3 h/day, and the PAPG response decreased to 45%
after 30 days. The experiment for long-time stability was conducted
daily for this time. When it was not in use, the immunosensor was
stored in a PBS buffer solution at 48 C.
Finally the reproducibility of the immunosensor fabrication was
evaluated via the comparison of the response current to 0.8 mg/ml
PAPG of different electrodes. Three different enzyme electrodes
were tested independently for the PAPG amperometric response,
providing a relative standard deviation (RSD) value of 10%. This indicates, in particular, an efcient and reproducible immobilization
process of the proposed biosensing matrix.
4. Conclusions
A rapid, specic and sensitive assay has been described for the
detection of intact cells of E. coli. The technique is based on the
immobilization of the bacterial sample via specic capture antibodies to a GC electrode surface. Then the attached cells are lysed
and entrapped by a conductive Alg-Ppy composite to conne the
catalytic reaction on the transducer surface. Electropolymerized
Alg-Ppy lls the porous alginate with polypyrrole laments reducing possible leaching of bacterial contents that was made available
by lysis. This immunosensor set up has shown a high sensitivity
(10 CFU/ml) as well as specicity and was shown to be stable up to
one month. The detection is done within a short 3 h period from
the dipping of the modied electrode into the bacterial sample
without the need for a preincubation period. It is expected that
this extremely sensitive immunosensor will be helpful for the easy
monitoring of water quality (Nicholas et al., 2001), food safety and
environmental control.
Acknowledgments
The authors gratefully acknowledge Dr. Boris Polyak from Drexel
University, College of Medicine, USA, for his valuable assistance and

3466

K. Abu-Rabeah et al. / Biosensors and Bioelectronics 24 (2009) 34613466

scientic discussions, Dr. Eliora Ron, Tel-Aviv University, Israel, for

providing the wild type strain of E. coli K-12 MG1655 and Dr. Ariel
Kushmaro, Ben-Gurion University of the Negev, Israel, for providing
the wild type strain of P. aeruginosa.
References
Abdel-Hamid, I., Ivnitski, D., Atanasov, P., Wilkins, E., 1999a. Biosens. Bioelectron. 14
(3), 309316.
Abdel-Hamid, I., Ivnitski, D., Atanasov, P., Wilkins, E., 1999b. Anal. Chim. Acta 399
(12), 99108.
Abu-Rabeah, K., Polyak, B., Ionescu, R.E., Cosnier, S., Marks, R.S., 2005. Biomacromolecules 6 (6), 33133318.
Bakker, E., 2004. Anal. Chem. 76 (12), 32853298.
Buchanan, R.L., 1997. Emerg. Infect. Dis. 3 (4), 517521.
Datsenko, K.A., Wanner, B.L., 2000. Proc. Natl. Acad. Sci. U.S.A. 97 (12), 66406645.
Feng, P.C., Hartman, P.A., 1982. Appl. Environ. Microbiol. 43 (6), 13201329.
Gehring, A.G., Irwin, P.L., Reed, S.A., Tu, S.I., Andreotti, P.E., Akhavan-Tafti, H., Handley,
R.S., 2004. J. Immunol. Methods 293 (12), 97106.
Geng, P., Zhang, X.N., Meng, W.W., Wang, Q.J., Zhang, W., Jin, L.T., Feng, Z., Wu, Z.R.,
2008. Electrochim. Acta 53 (14), 46634668.
George, I., Petit, M., Servais, P., 2000. J. Appl. Microbiol. 88 (3), 404413.

Gough, D., Leypoldt, J., 1979. Anal. Chem. 51 (3), 339444.

Howe, E., Harding, G., 2000. Biosens. Bioelectron. 15 (1112), 641649.
Ionescu, R.E., Abu-Rabeah, K., Cosnier, S., Marks, R.S., 2005. Electrochem. Commun.
7 (12), 12771282.
Ivnitski, D., Abdel-Hamid, I., Atanasov, P., Wilkins, E., 1999. Biosens. Bioelectron. 14
(7), 599624.
Leonard, P., Hearty, S., Brennan, J., Dunne, L., Quinn, J., Chakraborty, T., OKennedy, R.,
2003. Enzyme Microb. Technol. 32 (1), 313.
Lin, Y.H., Chen, S.H., Chuang, Y.C., Lu, Y.C., Shen, T.Y., Chang, C.A., Lin, C.S., 2008.
Biosens. Bioelectron. 23 (12), 18321837.
Loge, F.J., Thompson, D.E., Call, D.R., 2002. Environ. Sci. Technol. 36 (12), 27542759.
Mehrvar, M., Abdi, M., 2004. Anal. Sci. 20 (8), 11131126.
Mittelmann, A.S., Ron, E.Z., Rishpon, J., 2002. Anal. Chem. 74 (4), 903907.
Nicholas, J., Ashbolt, W.O.K.G.a., Snozzi, M., 2001. Indicators of Microbial Water Quality, Lorna Fewtrell and Jamie Bartram. IWA Publishing, London, UK.
Palenzuela, B., Simonet, B.M., Garcia, R.M., Rios, A., Valcarcel, M., 2004. Anal. Chim.
Acta 524 (12), 167174.
Pyle, B.H., Broadaway, S.C., McFeters, G.A., 1995. Appl. Environ. Microbiol. 61 (7),
26142619.
Su, X.L., Li, Y., 2004. Biosens. Bioelectron. 19 (6), 563574.
Tokarskyy, O., Marshall, D.L., 2008. Food Microbiol. 25 (1), 112.
Tryland, I., Fiksdal, L., 1998. Appl. Environ. Microbiol. 64 (3), 10181023.
Venkateswaran, K., Murakoshi, A., Satake, M., 1996. Appl. Environ. Microbiol. 62 (7),
22362243.