View Article Online / Journal Homepage / Table of Contents for this issue

Disposable potentiometric enzyme sensor for direct

determination of organophosphorus insecticides

Sonja Gäberlein,a Meinhard Knoll,a Friedrich Spenerb and Christiane Zaborosch*†a

a Institut für Chemo- und Biosensorik, Mendelstr. 7, 48149 Münster, Germany

b Institut für Biochemie, Westfälische Wilhelms-Universität Münster, Wilhelm-Klemm-Str. 2,
Published on 20 November 2000. Downloaded by University of Illinois at Chicago on 13/08/2013 12:49:36.

48149 Münster, Germany

Received 15th August 2000, Accepted 4th October 2000

First published as an Advance Article on the web 20th November 2000

A potentiometric disposable enzyme sensor for the direct and fast determination of organophosphorus (OP)
insecticides was developed by using an organophosphorus hydrolase (OPH) immobilized on an ion-selective
electrode. The disposable screen-printed transducer was based on double matrix membrane technology which
allows easy mass production. The potentiometric device consisted of a H+-sensitive electrode with integrated
Ag/AgCl reference electrode. The electrodes were prepared with N,N-dioctadecylmethylamine as H+-sensitive
ionophore and pH calibration resulted in slopes of 55 mV decade21 over a pH range from 11 to 6. OPH was
isolated from recombinant Escherichia coli DH5a and immobilized within poly(carbamoyl sulfonate) prepolymer
on the surface of the H+-sensitive electrode without any further fixation membrane. OPH catalyzes the hydrolytic
cleavage of OP compounds which releases protons in a concentration proportional to hydrolyzed substrate. Sensor
performance was investigated with regard to enzyme load, concentration, pH and temperature of the measuring
buffer using paraoxon as analyte. Best sensitivity and response time were obtained with sensors prepared with 250
U of OPH and measuring at 37 °C in 1.0 mM HEPES buffer, pH 9.3, containing 100 mM NaCl. The enzyme
sensor exhibited a linear calibration range of 0.01–0.15 mM chlorpyrifos, 0.05–0.35 mM diazinon, 0.05–0.4 mM
paraoxon and 0.007–0.05 mM parathion, respectively. For all these analytes response times to reach 95% of
maximum change in potential did not exceed 5 min. Sensors stored under dry conditions at 4 °C still showed 60%
of initial hydrolytic rate after 70 d. The sensors even when stored dry were ready for measurements after 5 min
incubation in measuring buffer. A range of putative interfering substances did not influence sensor response, and
suitability of measuring OPs in soil extracts was ascertained.

Introduction The use of the hydrolytic cleavage reaction of organophos-

Synthetic organophosphorus (OP) compounds are among the
most toxic substances known and represent the largest class of
insecticides widely used in agriculture. Target organisms are
affected by an almost irreversible inhibition of acetylcholines-
terase (AChE), which is also the primary cause for acute toxicity
in non-target organisms including mammals.
Growing public environmental concern as well as the
widespread use of OP compounds have intensified the need for
effective detection devices. Gas, liquid and thin-layer chroma-
tography are currently the most common analytical methods for
OP determination,1,2 being exact and reliable, but time-
consuming and expensive techniques. The use of immunochem-
ical assays with monoclonal antibodies provides the advantage
of less sample processing and high compound specificity,3 but
nevertheless limitations concerning analysis time and handling
occur. Another detection principle for pesticide analysis is
based on electrochemical biosensors.4 Biosensors are well
suited tools for on site environmental analysis and rapid and
direct detection of insecticides.5 Currently, the most common
detection principle of biosensors for OP compounds is based on
the inhibition of cholinesterase activity, using amperometric,6,8
potentiometric9,10 and fiber optic11 transducers. However
enzyme sensors based on inhibition reactions generally require
long incubation times for sufficient sensitivity and long
regeneration times of the inhibited enzyme after sample analysis
necessary for a repeated use.
† Present address: Departement Chemie, Zürcher Hochschule Winterthur,
Technikumstr. 9, Postfach, 8401 Winterthur, Switzerland.
phorus hydrolase (OPH) (EC 3.1.8.1) enables a new concept for
biosensor development.12,13 OPH hydrolyzes a wide range of
OP esters, e.g. parathion, paraoxon, diazinon and chemical
warfare agents like soman, sarin and VX.14–16 OPH-catalyzed
hydrolysis of one OP molecule goes along with the release of
two protons from the hydrolysis products thus providing the
measuring principle for a potentiometric biosensor. The ad-
vantage of OPH-based sensors over AChE-based sensors is the
simple, fast and specific measurement of OP insecticides.
Recently, we described the development of biosensors using
either whole cells of Flavobacterium sp., or cytoplasmic
membranes with associated OPH, respectively.17 The biological
components were immobilized within a hydrogel on the sensing
surface of a conventional glass pH electrode and the sensors
allowed the fast detection of OP compounds.
In the present paper we focused on a potentiometric enzyme
sensor with recombinant OPH isolated from E. coli DH5a cells.
It was based on a disposable H+-sensitive electrode with an
integrated Ag/AgCl reference electrode produced in the double
matrix membrane (DMM) technology.18,19 The ion-selective
polymeric membrane is formed by a casting technique using
filter paper as mechanical support for the H+-sensitive mem-
brane. The components of the poly(vinyl chloride) (PVC)
matrix, which includes the H+-sensitive ionophore, are dis-
solved within organic solvents and this membrane mixture is
homogeneously distributed on the filter paper by capillary
forces. After evaporation of the solvents, the ion-selective
DMM (filter paper and PVC membrane) is obtained. Pre-
viously, Na+- and K+-sensitive electrodes were developed using
this technology20,21 as well as ammonium sensitive electrodes,
which were used as transducers for disposable biosensors for

2274 Analyst, 2000, 125, 2274–2279 DOI: 10.1039/b006664h

This journal is © The Royal Society of Chemistry 2000
 View Article Online

the determination of urea in blood.22 In our device here, the OP
selective biocomponent was immobilized within the adhesive
poly(carbamoyl sulfonate) (PCS) prepolymer on the surface of
the heat sealing film adjacent to the H+-sensitive PVC
membrane. Integration of a disposable reference electrode on
the rear side of the H+-sensitive electrode resulted in a small,
compact sensor configuration for OP analysis. The fundamental
characteristics of this enzyme sensor are presented in the
following.
Published on 20 November 2000. Downloaded by University of Illinois at Chicago on 13/08/2013 12:49:36.
collected at a flow rate of 4 ml min21. The OPH-containing
fractions were combined and concentrated to about 14.5 mg
protein ml21 by ultrafiltration in an Amicon diaflo cell. The
concentrated enzyme solution was stored at 220 °C until use.
Progress of the purification process was monitored by determin-
ing specific activity and by analysis with sodium dodecyl
sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). En-
zymatic activity was measured by monitoring the change in
absorbance at 400 nm, when paraoxon was hydrolyzed to
diethyl phosphate and p-nitrophenol (e400 = 17 mM21 cm21) in
20 mM glycine buffer (pH 9.0), 100 mM sodium chloride. One

Experimental unit (U) of activity is defined as 1 mmol paraoxon hydrolyzed

per minute.
Chemicals

Paraoxon (O,O-diethyl-O-4-nitrophenylphosphate), parathion Preparation of disposable electrodes

(O,O-diethyl-O-4-nitrophenyl phosphorothioate) and diazinon
(O,O-diethyl-O-2-isopropyl-6-methylpyrimidin-4-yl phos-
phorothioate) were purchased from Dr. Ehrenstorfer Company
(Augsburg, Germany). Chlorpyrifos (O,O-diethyl-O-(3,5,6-tri-
chloro-2-pyridyl)phosphorothioate) was obtained from Frunol
Delicia GmbH (Delitzsch, Germany). The trivial names of these
organophosphorus insecticides are used as a matter of conven-
ience. Components for the H+-sensitive membranes, N,N-
dioctadecylmethylamine (H+-ionophore), bis(2-ethylhexyl)-
sebacate, sodium tetraphenylborate and poly(vinyl chloride)
(PVC) were obtained from Fluka (Buchs, Switzerland). The
poly(carbamoyl sulfonate) (PCS) prepolymer for immobiliza-
tion of organophosphorus hydrolase was a research product
from the Bundesforschungsanstalt für Landwirtschaft
(Braunschweig, Germany) and is available from SensLab
(Leipzig, Germany). Superdex 200 and diethylaminoethyl
(DEAE)–Sepharose were obtained from Amersham Pharmacia
(Freiburg, Germany). All other chemicals were purchased from
Sigma (Steinheim, Germany).
Fig. 1 shows a schematic view of the screen-printed H+-
sensitive electrode with integrated Ag/AgCl reference electrode
which was used as transducer for the disposable enzyme sensor.
The working electrodes were produced in a batch process
similar to that described previously.20,24 Conducting lines of the
electrodes were screen-printed with silver carbon ink (Auromal
L 180, Doduco, Pforzheim, Germany) on a sheet of a 150 mm
thick polyester–polyethylene heat sealing film (Team Codor,
DUBO-Schweitzer GmbH, Haltern, Germany). To prepare the
contact between the filter paper (Schleicher & Schuell, Dassel,
Germany) and the conducting line, silver was evaporated on the
rear side of the filter paper. For insulation a pre-perforated heat-
sealing film with holes of 2 mm diameter was used and the two
films were laminated together at 125 °C with the filter paper in
between. By this procedure the electrodes were completely
encapsulated by the heat sealing film except for the 2 mm
diameter hole in front of the filter paper for deposition of the H+-

Isolation and purification of organophosphorus hydrolase

(OPH)

OPH was purified from recombinant E. coli DH5a carrying the

plasmid pJK33 as described previously.23 E. coli cells were
grown on a rotary shaker overnight at 30 °C in 4 l culture
medium containing 12 g l21 peptone, 24 g l21 yeast extract, 80
mM K2HPO4, 20 mM KH2PO4, 4 ml l21 glycerol and 1 mM
CoCl2. Cells were harvested by centrifugation at 3000g at 4 °C
for 20 min, washed with 50 mM potassium phosphate buffer
(pH 7.2) and re-centrifuged at 3000g at 4 °C for 20 min. After
resuspension in potassium phosphate buffer (pH 7.2) (1 ml per
gram wet weight cells) cells were disrupted in a French pressure
cell at 1000 bar and 4 °C. The cell extract was ultracentrifuged
at 70 000g for 60 min at 4 °C. All subsequent steps were
performed on ice. The supernatant was mixed dropwise with a
solution of protamine sulfate (2% m/v) while stirring for 30 min
until the protamine sulfate concentration reached 0.4%. After
stirring for an additional 30 min the solution was centrifuged at
30 000g for 20 min. Ammonium sulfate fractionation was
performed with the supernatant by addition of solid ammonium
sulfate to 45% saturation while stirring for 30 min. After an
additional 30 min of stirring the solution was centrifuged at
30 000g for 20 min. The protein precipitate from ammonium
sulfate treatment was dissolved in 50 mM HEPES buffer (pH
8.5) (buffer A) and directly applied to a Superdex 200 prepgrade
column (2.6 3 60 cm), pre-equilibrated with the same buffer.
Fractions of 2 ml were collected at a flow rate of 4 ml min21 and
pooled on the basis of enzymatic activity and absorbance at 280
nm. Subsequently, a DEAE–Sepharose Fast Flow column (1.6
3 30 cm), pre-equilibrated with buffer A, was loaded with the Fig. 1 Schematic view of the disposable enzyme sensor based on H+-
OPH-containing fractions. Protein was eluted with a linear sensitive electrode with integrated Ag/AgCl reference electrode. A, Cross-
gradient using 1 M NaCl and buffer A. Fractions of 2 ml were sectional view. B, Front and rear view.

Analyst, 2000, 125, 2274–2279 2275


sensitive membrane and a contact pad of 5 mm length at the
electrode’s upper part.
The reference electrode was screen-printed on the back side
of the heat sealing film of the working electrode using screen-
printing ink (Elektrodag 6037 SS, Acheson, Scheemda, Nether-
lands) with Ag and AgCl in a ratio of 3+2.25 After curing for 1
h at 80 °C filter paper strips (4 3 20 mm, Schleicher & Schuell,
Dassel, Germany) were positioned on the conducting lines (Fig.
1). Encapsulation of the reference electrode was carried out by
lamination with a further heat sealing film at 125 °C leaving a
contact pad of 5 mm uncovered at the upper part of the
Published on 20 November 2000. Downloaded by University of Illinois at Chicago on 13/08/2013 12:49:36.
electrode. Upon immersion of the completed electrode in
measuring buffer, which contained 100 mM sodium chloride as
electrolyte, an opening at the lower part of the reference
electrode allowed the spontaneous filling of the filter paper by
capillary forces.
Preparation of H+-sensitive double matrix membranes
(DMM)
The working electrodes received the H+-sensitive character by
positioning a defined membrane mixture on to the filter paper.
The mixture contained the following components (wt.%): 1.0%
N,N-dioctadecylmethylamine (H+-ionophore), 67.0% bis(2-
ethylhexyl)sebacate, 0.3% sodium tetraphenylborate, 31.7%
poly(vinyl chloride) (PVC). The membrane components were
dissolved in 1.5 ml tetrahydrofuran and 0.5 ml cyclohexanone
by shaking vigorously overnight until a homogeneous solution
was obtained. A volume of 10 ml of the membrane mixture was
deposited on the filter paper of the working electrode in five
dispensing steps at intervals of 5 min in order to prevent an
overflow. The solvents were allowed to evaporate at room
temperature overnight leaving the H+-sensitive PVC-mem-
brane. The last step was the separation into single ion-selective
electrodes (ISE) with integrated reference electrodes, which
were stored dry at room temperature.
Preparation of enzyme sensors for OP detection
The immobilization procedure of OPH was carried out similarly
to previously described methods.17,26,27 In brief, 300 mg of PCS
prepolymer (33.4%) were mixed with 300 ml of 10.0 mM
glycine buffer (pH 9.0), 100 mM sodium chloride. Poly-
(ethyleneimine) (2.5% m/v in water) was added under constant
mixing to adjust the pH to 6.4–6.5. After centrifugation to
remove polymerized particles, PCS and OPH solution
(10 000–270 000 U ml21) were mixed at a ratio of 1+1. A total
volume of 3.0 ml of the mixture was deposited on the heat
sealing film in a spot directly adjacent to the H+-sensitive
membrane (Fig. 1) and allowed to polymerize for 30 min at
30 °C. Subsequently, the sensors were ready for use or stored
under dry conditions at 4 °C.
Experimental set-up for sensor measurements
All measurements were carried out in a stirred measuring cell
containing 2 ml of 1.0 mM HEPES buffer (pH 9.3) and 100 mM
sodium chloride at 37 °C. Stock solutions of OP compounds and
Soxhlet extraction of soil samples were prepared with methanol.
When adding OP insecticides the change in potential due to the
release of protons after hydrolysis was recorded with a
microprocessor pH analyser (microprocessor pH meter 3000
with multiplex 3000, WTW, Weilheim, Germany) connected to
a computer. The contact of the disposable H+-sensitive
electrode and the integrated reference electrode with the
potentiometer was provided by a bipolar plug.
2276 Analyst, 2000, 125, 2274–2279
View Article Online
Results and discussion
Characterization of disposable H+-sensitive electrodes
Investigation of electrochemical characteristics of the dispos-
able screen-printed electrodes demonstrated their suitability for
biosensor development. The pH calibration of ten H+-sensitive
electrodes revealed a slope with a mean value of 55.2 mV
decade21 (data not shown), which is well beyond the 48 mV
decade21 (80% of the theoretical Nernst value of 59 mV
decade21) considered for good ion-selective electrodes.28 The
disposable H+-sensitive electrodes showed a linear response
behaviour within a measuring range from pH 11 to 6 and the
standard deviation of 1.9% of the measurements (n = 10)
proved the suitability of the DMM technology for this biosensor
development. The electrodes responded within a few seconds to
changes in the H+ concentration.
Sensor performance
Different parameters influencing sensitivity and response
behaviour of the potentiometric enzyme sensor were studied
using paraoxon as analyte at a concentration of 0.1 mM. The
amount of enzyme immobilized on the electrode determines
whether the biosensor is working under kinetic or diffusional
control. Starting pH and concentration of the measuring buffer
as well as measurement temperature were additional parameters
under consideration. Data were evaluated by calculating the
initial hydrolytic rate of the biosensors upon addition of
paraoxon; response times necessary for this evaluation did not
exceed 60 s (results not shown).
Effect of enzyme loading. Sensors were prepared with
varying enzyme amounts from 15 to 400 U OPH in the
immobilization matrix PCS. An increase of the amount of OPH
on the electrode up to 200 U resulted in enhanced initial
hydrolytic rates of the sensors (Fig. 2). Up to this amount the
enzymatic hydrolytic reaction is the rate controlling step
influencing sensitivity of the sensor. An enzyme load over 200
units OPH did not further increase initial hydrolytic rates, which
means that sensitivity in this case was limited by the diffusion
of the substrate to the immobilized enzyme and/or of the protons
released to the sensing surface of the electrode. No bleeding-out
of OPH from the PCS hydrogel was observed when testing the
measuring solution by spectrophotometric investigation (p-
nitrophenol formation after addition of paraoxon). This con-
firmed that the hydrolysis reaction was solely due to the
immobilized OPH. Our biosensors reached maximum initial
hydrolytic rates DE/Dt of 100 mV min21 with 300 U OPH
immobilized in PCS on the disposable electrodes. When 300 U
OPH were immobilized on a conventional glass pH electrode
initial hydrolytic rates of only 2 mV min21 were achieved and
a maximum of 7 mV min21 with 775 U OPH immobilized.29
Fig. 2 Effect of OPH loading on initial hydrolytic rates of the enzyme
sensor measuring paraoxon (0.1 mM) in 1.0 mM HEPES buffer (pH 9.3),
100 mM NaCl, 37 °C (mean values ± s, n = 3).

The authors of the latter study used glutaraldehyde for
immobilization of OPH and a dialysis membrane for fixation of
the gel on the surface of the electrode. The lower initial
hydrolytic rates might be attributed to this procedure and to the
dialysis membrane which introduced an additional diffusion
barrier.
Effect of starting pH. Highest hydrolytic rates were reached
at pH 9.5 (Fig. 3A), which was 0.5 pH units below the optimum
of the non-immobilized enzyme (Fig. 3B). This effect might be
Published on 20 November 2000. Downloaded by University of Illinois at Chicago on 13/08/2013 12:49:36.
attributed to the immobilization procedure influencing the pH
optimum of the enzyme. All sensor measurements to follow
were carried out at pH 9.3.
Effect of buffer concentration and temperature. The
measuring signal of potentiometric sensors is strongly influ-
enced by the concentration of buffering substances. Protons
released during the enzymatic hydrolytic cleavage of OP
compounds are counteracted by buffering ions. Investigations
were carried out within a concentration range of 1 to 20 mM
using HEPES buffer (Fig. 4). The highest sensor signal was
attained with the buffer of lowest molarity (1 mM), which was
consequently used in further measurements. Lower buffer
concentrations resulted in drifting baselines presumably due to
the low buffer capacity. Sodium chloride addition (100 mM) to
the measuring buffer provided the conductivity necessary for
the integrated reference electrode of the disposable enzyme
sensor and neither influence electrode characteristics nor
enzymatic activity (data not shown). Nevertheless, with respect
to environmental samples, i.e. soil samples, difficulties arising
from weak buffer capacity may be of concern.
Fig. 3 Effect of starting pH of the measuring buffer on the initial rate of
paraoxon (0.1 mM) hydrolysis. A, Potentiometric measurements with
enzyme sensor containing 250 U OPH (sensor age: 2 d) at 37 °C. Measuring
buffers: 1.0 mM potassium phosphate (pH 7.0), 1.0 mM HEPES (pH 8.0 to
10.5), 1.0 mM glycine (pH 11.0 and 12.0) (mean values ± s, n = 3). B,
Spectrophotometric assay of p-nitrophenol liberated by non-immobilized
OPH (14.4 mg l21, 2.5 U) at 25 °C. Measuring buffers: 20 mM potassium
phosphate (pH 7.0), 20 mM HEPES (pH 8.0, 9.0 and 10.0), 20 mM glycine
(pH 11.0 and 12.0) (mean values ± s, n = 3).
Fig. 4 Effect of the buffer concentration on the initial hydrolytic rate of
enzyme sensors containing 250 U OPH (sensor age: 1 d) measuring
paraoxon (0.1 mM) in HEPES buffer (pH 9.3) at 37 °C (mean values ± s, n
= 3).
View Article Online
During all measurements the stirred measuring cell was kept
at 37 °C, a temperature where highest initial hydrolytic rates
were attained with the biosensors (data not shown).
Calibrations with organophosphorus compounds
Fig. 5 shows the calibration plots for chlorpyrifos, diazinon,
paraoxon and parathion, determined with the newly developed
disposable enzyme sensors. The data were calculated from 95%
of the maximum change in potential of the sensors to different
OP concentrations. Time ranges to reach these responses were
3 min for paraoxon and up to 5 min for chlorpyrifos. The
analytical characteristics (maximum sensitivity, linear range,
lower detection limit) for the different OP compounds measured
with the potentiometric enzyme sensors are summarized in
Table 1. Linear ranges covered almost one order of magnitude
with similar ranges for paraoxon (0.05–0.4 mM) and diazinon
(0.05–0.35 mM). The linear range for parathion (0.007–0.0 5
mM) was about one order of magnitude lower than that for
paraoxon, diazinon and for chlorpyrifos (0.01–0.15 mM).
Maximum sensitivity (mV decade21 OP concentration) of the
disposable enzyme sensors towards the OP compounds was in
the order paraoxon > diazinon > parathion > chlorpyrifos.
The lower detection limits (three times the background noise of
the electrode) of 0.005 mM for the four OP insecticides were
comparable to other described OPH based biosensors.13,29
These detection limits were generally higher than those of
AChE based sensors which were in the nM range,10,11,30 thus
limiting the application of OPH biosensors for environmental
analysis. For such applications samples need prehandling by
suitable extraction methods and subsequent concentration of the
OP compounds, allowing the use of OPH based sensors for off-
line monitoring processes. Furthermore the sensors developed
are well suited for on-line monitoring of detoxification
processes, e.g. of waste water during pesticide production and
consumption, or after inappropriate application of OP com-
pounds.
Fig. 5 Calibration plots for organophosphorus compounds chlorpyrifos
(2), diazinon (8), paraoxon (!) and parathion (:) with enzyme sensors
containing 500 U OPH (sensor age 2 d). Measuring buffer was 1.0 mM
HEPES (pH 9.3), 100 mM NaCl at 37 °C. Errors of sensor values were
below symbol size (n = 3). OP compounds were applied in methanol stock
solutions. Methanol concentrations did not exceed 2.5%.
Table 1 Analytical characteristics of disposable enzyme sensors contain-
ing 500 U OPH (sensor age: 2 d)
Maximum Linear range
sensitivity/mV corresponding
decade21 OP to maximum Lower detection
OP compound concentration sensitivity/mM limita/mM
Paraoxon 98 0.05–0.4 5
Diazinon 61 0.05–0.35 5
Parathion 37.7 0.007–0.05 5
Chlorpyrifos 24.6 0.01–0.15 5
a Defined as three times the background noise of the electrode.
Analyst, 2000, 125, 2274–2279 2277
 View Article Online

Specificity of the enzyme sensor
The influence of various substances on the sensor signal was
investigated with regard to OP detection in environmental
samples (Table 2). The addition of carbohydrates as well as
organic acids at 150-fold higher concentrations caused max-
imum signals of 5 mV at the sensor. Subsequent addition of
paraoxon (0.1 mM) resulted in changes of potential similar to
when measuring paraoxon (0.1 mM) in the absence of the
possibly interfering substances. As an exception, measurements
in the presence of acetate lead to a 32.5% increase of the sensor
Published on 20 November 2000. Downloaded by University of Illinois at Chicago on 13/08/2013 12:49:36.
with measuring buffer improved their performance, yet not
completely (Table 3).
Reproducibility, precision and long term storage stability
of the enzyme sensors
The excellent reproducibility of the disposable potentiometric
enzyme sensor was demonstrated by the low standard deviation
of 1.1% of the DE values for ten sensors with 250 U of OPH
(data not shown). The precision in case of the repeated use of

signal for paraoxon for reasons not known. The time ranges to
reach 95% of maximum change in potential did not exceed 2
min in all cases, underlining the fast sensor response. The
addition of humic acid, a substance which is to be expected in
extracts from soil samples, did not cause a signal at the sensor.
However, the time range to reach 95% of maximum response
was slightly prolonged (5 min). An incubation step of 5 min in
the measuring buffer after each measurement allowed the
electrode to return to the initial potential. This allows a
measurement frequency of 6 h21. All measurements presented
in Table 2 were performed with a single enzyme sensor. This
repeated use is not demanded for a disposable sensor, but shows
the suitability of our sensors for several measurements.
Investigation of soil samples
Environmental samples from sandy and humic soils were
prepared by Soxhlet extraction with methanol. Subsequently,
both extracts were concentrated, spiked with paraoxon and each
investigated with an enzyme sensor. DE values (95% of
maximum change in potential) were identical with the values
attained when paraoxon from stock solution was hydrolyzed
alone (Table 3). However, response times to reach DE for
identical paraoxon concentrations were higher in the presence
of soil extracts. Thus, Dt for paraoxon determination rose from
2 min to 10 min when humic soil extract served as matrix. This
was in line with our observation that humic acid addition
prolonged paraoxon determination to 5 min (Table 2). Ob-
viously the activity of immobilized OPH was negatively
affected by substances in soil extracts; even a memory effect
was observed as a subsequent 5 min treatment of the sensors
one sensor (250 U OPH) is documented by a low standard
deviation of 0.75% (n = 5).
Long term storage stability was investigated with sensors
prepared with 250 U of OPH. The sensors were stored under dry
conditions at 4 °C and the activity was determined from the
initial hydrolytic rate during paraoxon hydrolysis after different
storage times (Fig. 6). Each data point in Fig. 6 represents the
mean value of three measurements with a new sensor. The first
measurement was performed immediately after preparation of
the sensor and resulted in an initial hydrolytic rate DE/Dt of 100
mV min21 for 0.1 mM paraoxon. After a storage period of 70 d
still 60% of the initial activity was measured. After 110 d 26
mV min21 were reached representing 26% of the initial activity.
Allowing dry storage conditions, the disposable sensors on the
basis of screen-printed electrodes are superior to sensors based
on glass pH electrodes. Our disposable sensors, even when
stored dry, were ready for measurements after a 5 min
incubation step in measuring buffer, which allowed the
achievement of an initial potential and the filling of the filter
paper of the integrated reference electrode. Recently, the
application of screen-printed electrodes for amperometric
detection of OP compounds using either immobilized AChE30
or OPH31 was reported. A disposable multielectrode sensor on
the basis of four different types of AChE30 was used for the
analysis of binary reagent mixtures of paraoxon and the
carbamate insecticide carbofuran. Lower detection limits of
Table 3 Paraoxon determination in different matrices with an enzyme
sensor (500 U OPH, sensor age: 24 d). Investigation of sandy and humic
soils extracted with methanol and spiked with paraoxon. Measuring buffer
was 1.0 mM HEPES (pH 9.3), 100 mM NaCl at 37 °C (mean values ± s, n
= 3)

Measurementa DE/mV Dt/s

Table 2 Specificity of disposable enzyme sensors (500 U OPH, sensor 1 Paraoxon 31.7 ± 0.6 160 ± 5
age: 3 d). Investigation of the influence of various substances on the 2 Sandy soil extract spiked with paraoxon 31 ± 0.3 410 ± 20
response of enzyme sensors in HEPES buffer (1.0 mM, pH 9.3), 100 mM 3 Paraoxon 32 ± 0 255 ± 13
NaCl at 37 °C (mean values ± s, n = 3)
1 Paraoxon 32.2 ± 0.3 123 ± 5
Substancea DE/mV Dt/s 2 Humic soil extract spiked with paraoxon 32.8 ± 0.8 600 ± 20
3 Paraoxon 33 ± 1 185 ± 9
Paraoxon 40.0 ± 0.5 55 ± 0 a Paraoxon concentration = 0.1 mM.
Glucose 5.1 ± 0
Glucose + paraoxon 38.2 ± 1.2 55 ± 0
Sucrose 4.3 ± 0.3
Sucrose + paraoxon 41.2 ± 0.7 65 ± 0
Lactose 4.5 ± 0.3
Lactose + paraoxon 40.4 ± 0.8 100 ± 0
Maltose 5.0 ± 0.2
Maltose + paraoxon 40.4 ± 0.8 100 ± 0
Acetate 2.8 ± 0.1
Acetate + paraoxon 53.0 ± 4.0 115 ± 5
Gluconate 5.0 ± 0
Gluconate + paraoxon 41.0 ± 0.5 100 ± 0
Pyruvate 5.0 ± 0.05
Pyruvate + paraoxon 40.7 ± 1.2 110 ± 0
Glycerol 0
Glycerol + paraoxon 39.5 ± 1.0 100 ± 0
Humic acid 0 Fig. 6 Storage stability of disposable potentiometric enzyme sensors
Humic acid + paraoxon 39.5 ± 1.5 260 ± 15 (250 U OPH) at 4 °C under dry conditions. Determination of the initial
a Concentrations: paraoxon = 0.1 mM, humic acid = 20 mg l21, all other hydrolytic rate for 0.1 mM paraoxon in 1.0 mM HEPES buffer (pH 9.3), 100
substances 15 mM. mM NaCl at 37 °C. For each data point (mean values ± s, n = 3) new
sensors were used.

2278 Analyst, 2000, 125, 2274–2279

 View Article Online

about 1.0 nM for both analytes were better than those of all OPH
based sensors described so far. Nevertheless, analysis time with
the AChE sensor30 was up to 60 min per measurement, whereas
the present development allowed for determinations in the range
of a few minutes. The amperometric mode of thick film strip
electrodes using the hydrolytic reaction of OPH is based on the
anodic oxidation of enzymatically generated p-nitrophenol, yet
with a lower detection limit in the range of 0.1 mM.31 However,
this enzyme sensor is limited to the detection of OP compounds
releasing p-nitrophenol during enzymatic hydrolysis.
Published on 20 November 2000. Downloaded by University of Illinois at Chicago on 13/08/2013 12:49:36.
Germany) for helpful discussions. This paper is part of the
doctoral thesis of S. Gäberlein.
References
1 S. Yao, A. Meyer and A. Henze, Fresenius’ J. Anal. Chem., 1991,
339, 207.
2 J. Sherma, Anal. Chem., 1995, 67, 1R.
3 B. Hock, A. Dankwardt, K. Kramer and A. Marx, Anal. Chim. Acta,
1995, 311, 393.

4 M. Trojanowicz and M. L. Hitchman, Trends Anal. Chem., 1996, 15,

5 P. M. Paddle, Biosens. Bioelectron., 1996, 11, 1079.
Conclusions 6 P. Skladal and M. Mascini, Biosens. Bioelectron., 1992, 7, 335.
7 G. Palleschi, M. Bernabei, C. Cremisini and M. Mascini, Sens.
A disposable potentiometric biosensor for the fast and direct
detection of organophosphorus compounds was developed on 8 N. Mionetto, J. L. Marty and I. Karube, Biosens. Bioelectron., 1994,
the basis of recombinant organophosphorus hydrolase. A 9 S. Kumaran and C. Tran-Minh, Anal. Biochem., 1992, 200, 187.
screen-printed disposable H+-sensitive electrode was used as 10
transducer which allows mass production. The electrodes were
prepared with excellent precision and response behaviour over 11
a wide pH range. The polyurethane-based immobilization 12
matrix PCS has good adhesive properties and allows the simple
13
and direct immobilization of OPH on the electrode. The enzyme
sensors developed possessed good reproducibility, precision 14
and storage stability under dry conditions, additionally allowing
profound miniaturization and easier handling and storage in 15
comparison to sensors based on a glass pH electrode. Other
known methods of OP detection like immunoassays or AChE 16
based sensors require several reaction steps, while potentio- 17
metric OPH based sensors are applicable for the simple and fast
detection of organophosphorus triesters, which fits well with 18
practical applications where short analysis times are in demand. 19
Regardless of these advantages there are limitations of the
potentiometric system concerning the sensitivity of potentio- 20
metric responses to buffer capacity. In many practical applica- 21
tions such as sewage and subsoil water samples the content of
buffering substances is low and these are well-suited for 22
potentiometric analysis. Suitable extraction methods for enrich-
ment of OP compounds will also allow the OP detection in
environmental soil samples. 23
24
Acknowledgements 25
26
This work was supported by the scholarship program of the
Deutsche Bundesstiftung Umwelt (Osnabrück, Germany). We 27
are grateful to J. S. Karns (USDA, Beltsville, Maryland, USA)
for kindly providing us with the recombinant strain E. coli 28
DH5a carrying the plasmid pJK33. We thank K. D. Vorlop
(FAL Braunschweig, Germany) for providing us with the 29
immobilization matrix poly(carbamoyl sulfonate) prepolymer 30
and M. Borchardt (ICB, Münster, Germany) for kind assistance
concerning the layout of the disposable H+-sensitive electrode, 31
and we are grateful to H. Knipp (Frunol Delicia, Delitzsch,
38.
Actuators, B, 1992, 7, 513.
9, 463.
J. Diehl-Faxon, A. L. Ghindilis, P. Atanasov and E. Wilkins, Sens.
Actuators, B, 1996, 35–36, 448.
R. T. Andres and R. Narayanaswamy, Talanta, 1997, 44, 1335.
E. I. Rainina, E. N. Efremenco, S. D. Varfolomeyev, A. L. Simonian
and J. R. Wild, Biosens. Bioelectron., 1996, 11, 991.
A. Mulchandani, P. Mulchandani, I. Kaneva and W. Chen, Anal.
Chem., 1998, 70, 4140.
C. M. Serdar, D. T. Gibson, D. M. Munnecke and J. H. Lancaster,
Appl. Environ. Microbiol., 1982, 44, 246.
D. P. Dumas, J. R. Wild and F. R. Raushel, Biotechnol. Appl.
Biochem., 1989, 11, 235.
K. I. Dave, C. E. Miller and J. R. Wild, Chem.-Biol. Interact., 1993,
8, 55.
S. Gäberlein, F. Spener and C. Zaborosch, Appl. Microbiol.
Biotechnol., 2000, in the press.
M. Knoll, German Patent, P41 37 261.1, 1991.
C. Dumschat, M. Borchardt, C. Diekmann, J. Hepke, K. Cammann
and M. Knoll, Fresenius’ J. Anal. Chem., 1994, 348, 553.
M. Borchardt, C. Dumschat, K. Cammann and M. Knoll, Sens
Actuators, B, 1995, 24–25, 721.
C. Dumschat, M. Borchardt, C. Diekmann, K. Cammann and M.
Knoll, Sens. Actuators, B, 1995, 24–25, 279.
C. Eggenstein, M. Borchardt, C. Diekmann, B. Gründig, C.
Dumschat, K. Cammann, M. Knoll and F. Spener, Biosens.
Bioelectron., 1999, 14, 33.
G. A. Omburo, J. M. Kuo, L. S. Mullins and F. M. Raushel, J. Biol.
Chem., 1992, 267, 13278.
C. Eggenstein, M. Borchardt, C. Dumschat, B. Gründig, K.
Cammann, F. Spener and M. Knoll, Biosens. Bioelectron., 1995, 10,
595.
A. Mroz, M. Borchardt, C. Diekmann, K. Cammann, M. Knoll and C.
Dumschat, Analyst, 1998, 123, 1373.
A. Muscat, J. Beyersdorf and K. D. Vorlop, Biosens. Bioelectron.,
1995, 10, xi.
A. König, C. Zaborosch, A. Muscat, K. D. Vorlop and F. Spener,
Appl. Microbiol. Biotechnol., 1996, 45, 844.
K. Cammann, Working with ion-selective electrodes, Springer, New
York, 1979.
P. Mulchandani, A. Mulchandani, I. Kaneva and W. Chen, Biosens.
Bioelectron, 1999, 14, 77.
T. T. Bachmann and R. D. Schmid, Anal. Chim. Acta, 1999, 401,
95.
A. Mulchandani, P. Mulchandani and W. Chen, Anal. Chem., 1999,
71, 2246.

Analyst, 2000, 125, 2274–2279 2279