Skip to main content

Optical biosensing techniques for monitoring organic pollutants in the aquatic environment

Objective

The objective of this project is to develop optically-based analytical techniques for the detection and measurement of low concentrations of organic pollutants in the aquatic environment. The project, which will span three years, aims to investigate a number of techniques which will enable the development of highly sensitive biosensing instruments for the measurement of pesticides and herbicides in surface, ground and drinking waters.

Highly sensitive enzyme immunoassays for atrazine and simazine have been developed and primary antisera and hapten derivatives have been prepared.

Antisera to both atrazine and simazine showed very high titres, in excess of 1/60,000. The high titres were indicative of not only high concentration of specific antibodies in the sera but also of high affinity.

Where haptens with homologous bridge structures (aminohexanoic acid) were used to prepare plate coating antigens, the sensitivities of the assays for both triazines were poor. This was not unexpected since in common with all small size analytes the antibody binding site recognises the hapten as well as the bridge structures contained in the solid phase competing hapten when the bridge is homologous (same as the bridge in the immunogen). The analyte (which does not contain a bridge) provides poor competition for the antibody resulting in insensitive assays regardless of the level of the actual affinity of the antibody for the analyte molecule.

When haptens contained heterologous bridges were used to prepare plate coating antigens, assay sensitivities improved dramatically. The increase in sensitivity was relative to the degree of heterology of the bridge structures. Enzyme immunoassay calibration graphs for atrazine and simazine using heterologous bridges indicated that the least detectable amounts are about 2.5 pg per assay. This is equivalent to 25 ng/L or 0.025 ug/L. These detection limits are will within the European Community (EC) limits for drinking water.

Planar optical waveguide probes have been developed to determine the optical properties of attached sensing films. Three types of device are being investigated: surface plasmon resonance (SPR), directional coupler and chemiluminescence sensors. Techniques have been established for the deposition of compatible electrodes for electrochemical modulation of sensing reactions as an integral part of the devices, and equipment has been set up for the fabrication of waveguides in glass substrates by field assisted ion exchange. Interaction with partners has resulted in the establishment of standardized sensor chip formats and comparative evaluations of the sensors, using standardized sensing reactions, are in progress. Models of each type of sensor have been developed with a view to their use in the detection of pesticides in groundwater.

Electrochemiluminescent immunoassays on indium tin oxide coated waveguides have been developed.

Indium tin oxide coated glass was silanized by passing aminosilane vapour over the surface or by reflux in a solution of the aminosilane in dry toluene. Small fluorophores and large enzyme molecules can attach to the derivatized surface. The small fluorophore fluorescamine reacts rapidly and specifically with primary amines yielding a fluorescent product that can be measured in situ. Investigation of the 2 methods indicated that indium tin oxide silanized with aminosilane vapour had more amino groups on the surface. The enzyme horseradish peroxidase was attached with N-succinimidy 3-(2-pyridylthio)propionate (SPDP) or glutaraldehyde. The SPDP method gave more reliable results. Immobilized horseradish peroxidase can be used to generate a coloured dye and the amount of dye is a measure of the amount of enzyme attached to the surface. Results indicate that there is little difference between the 2 methods of silanizing.

An electrochemiluminescent immunoassay being developed is based on the generation of light when luminol is oxidized at an indium tin oxide coated electrode in the presence of hydrogen peroxide. Hydrogen peroxide can be generated using the enzyme glucose oxidase, which has been attached to antibodies specific for pesticides using glutaraldehyde.

Ovalbumin labelled with coumarin has been attached to indium tin oxide coated glass using the same method as for horseradish peroxidase. Antibodies to coumarin were labelled with horseradish peroxidase and incubated with ovalbumin derivatized indium tin oxide coated glass. Comparison with underivatized blanks showed that the antibodies had bound to immobilized coumarin.

Conditions for detecting the enzyme label are being optimized using glucose oxidase immobilized on indium tin oxide coated glass in a flow cell luminometer. The immobilization technique is identical to that used for horseradish peroxidase and ovalbumin. Pesticide immunoassays will be carried out following the reproducible electrochemiluminescent detection of glucose.

Work has been carried out on optimization of transducer structures capable of direct detection of an immunoreaction for analysis of pesticide levels. Two transducers have been investigated: a thin film transducer based on reflectometric interference spectroscopy (RIFS), that detects the immunoreaction as a change in the thickness of the active layer, and integrated optical devices (IO) that detect the immunoreaction by evanescent field effects.

In the RIFS transducer, light passing a thin transparent film causes 2 reflected partial beams. The interference of these beams leads to a periodic modulation of reflectance. This pattern can be acquired by a diode array spectrometer with high time resolution. From the reflectance spectrum of a small area the thickness of the thin film can be determined with very high resolution (10pm rms). Protein deposition during an immunoreaction leads to an increase in thickness and therefore RIFS allows dynamic monitoring of solid phase immunoassays. The limit of detection achieved with the RIFS transducer is very promising and first immunoassays carried out with pesticides have been encouraging. Further investigations will address long term stability and improvement of surface chemistry.

In the IO transducers, light is guided in a surface strip waveguide by total internal reflection. The interaction of the evanescent field with the superstrate affects the propagation of the guided mode. By a laterally structured protective layer, this interaction is confined to a measurement window. Deposition of protein at this window increases the effective refractive index of the measurement arm and leads to a phase shift between the 2 guided beams. The interferometer structure translates this phase shift into a periodic modulation of the transmission of the device. Good results have been obtained from modelling of device performance and have lead to a new chip design with improved transducer sensitivity. The complex handling of IO chips on an optical bench still poses a limit to extended tests, but first samples with polarisation maintaining fibre pigtails are available to aid further work.

A trial test has been performed on an immunoassay kit provided by Hirst Research Centre (HRC) for pesticide detection in groundwater. The immunoassay tests for simazine and atrazine were compared to tests with a previously validated technique, high performance liquid chromatography (HPLC) with a diode array detector. The HRC test consisted of an enzyme linked immunosorbent assay (ELISA).

Using the HRC immunoassay kit the simazine calibration curve was linear but the values of the optical density were quite scattered, mostly for the low values. The variability of the three replicates (relative standard deviation, RSD) was in the 2 to 17% range. The slope was rather low. The cross reactivity, as measured by the response of the simazine plate to the addition of atrazine, was significant. The atrazine calibration curve presented the same features as for the simazine, except for the cross reactivity. The addition of simazine to the atrazine plate does not induce any change in the optical density. This specificity, if confirmed by additional tests, would be a major advantage over the existing immunoassay kits commercially available. The day to day variability was tested by conducting a second set of trails 24 hours later. Most of the trends displayed one day also appeared the next, although the lack of cross reactivity of simazine towards atrazine, was not confirmed the day after.

In HPLC determination of simazine and atrazine, the calibration curves were perfectly linear over the complete range of concentrations. The slopes for simazine and atrazine were quite steep, roughly 10 times steeper than those of the immunoassays, indicative of good sensitivity. The addition of atrazine (simazine) to the simazine (atrazine) water sample had no effect on the calibration curve even when the 2 pesticides were present in a 1 to 5 ratio. This is due to the perfect separation of the 2 compounds by the chromatographic conditions used.

The HPLC results remained significantly better than the ELISA tests on the basis of sensitivity, selectivity and reproducibility but the HRC kits look promising especially the possible specificity of the ELISA test for atrazine.

Integrated optical interferometers have been developed which can be used as evanescent field sensors. The integrated optical Mach-Zehnder interferometer consists of 2 signals splitters just as the classic optical analogue. The first branch separates the measurement and reference signal. The waveguides are protected with a low refractive cover layer except in a window region. Due to the interaction with the superstrate in the window region the light in the measurement waveguide is shifted in phase relative to the light in reference waveguide. The superposition of both signals in the second branch produces an interference signal at the output port of the interferometer.

Integrated optical waveguides can be combined with biologically active adlayers consisting of antibodies, enzymes or cells which are immobilised on the surface of the integrated optical sensor chip. The biological macromolecules have high binding affinities to a special analyte and closely related molecules. Systems based on antibodies show little cross reactivity even for closely related analytes. Therefore antibody based sensitive layers can be used to develop sensors with high specifity. A Mach-Zehnder interferometer has been fabricated with a 15 mm long window where antibodies were immobilized on the waveguide surface. The interference signal of the integrated optical immunosensor comes from the successive covering of the waveguide surface during the antigen antibody interaction. The sensitivity met the requirements for biochemical sensors.
The project will concentrate on atrazine, simazine and aldrin which are identified as priority pollutants. The main emphasis of the research will be the development of immunosensing techniques capable of measuring concentrations lower than the EC limits set for the maximum admissible concentration of 0.lug/L (0.1 ppb) for pesticides and related substances. In addition, the proposed techniques will be designed to enhance selectivity and minimise measurement time to allow rapid screening of a large number of potential samples.

Funding Scheme

CSC - Cost-sharing contracts

Coordinator

Marconi Underwater Systems Ltd.
Address
Elettra Avenue
PO7 7XS Waterlooville
United Kingdom

Participants (4)

Bureau de Recherches Géologiques et Minières
France
Address
Avenue De Claude Guillemin 3
45060 Orléans
Entwicklungsgesellschaft für Integrierte Optik Technologie GmbH
Germany
Address
Bruchsaler Straße 22
68753 Waghäusel-kirrlach
GEC Marconi Ltd
United Kingdom
Address
Elstree Way
WD6 1RX Borehamwood
THE UNIVERSITY OF TUEBINGEN
Germany
Address
Auf Der Morgenstelle 8
72076 Tuebingen