Andreas Brecht

Optical probes and transducers

Biosensors are by definition a combination of a biological receptor compound and a physical or physicochemical transducer. Therefore, the transducing structure is a critical part of every biosensor. In the development of new and improved biosensing layers the importance of the transducing structure is not restricted to the substrate to which biological structures have to be coupled. A field of even greater importance is the use of transducers as probes providing information on the structure and function of biosensing layers, and their relation to a transducer surface. The aim of this paper is to give an overview on optical transducer principles and optical (surface) analytical techniques relevant as part of biosensing structures as well as probes in the development and optimisation of biosensing layers. Categories discussed are basic optical effects, materials involved, surface chemistry, the principal and technological limits of spatial resolution, and sensitivity. The intimate relation between the spatial resolution of a probe, the resulting size of interaction areas, and the feasibility of array structures is pointed out. Two interferometric methods are presented in principle, and their application to biosensing and some results are discussed in detail. The necessity to characterise receptor layers to get detailed information about the interaction process is pointed out. The close relationship between optimal characterisation of layers by selection of adequate probe technologies and improvement of probe performance, and the development of new biosensing layers is discussed. Finally, an outlook is given for future aspects of improved spatial resolution and multianalyte detection.

A high-density poly(ethylene glycol) polymer brush for immobilization on glass-type surfaces

Label-free heterogeneous phase detection critically depends on the properties of the interfacial layer. We have obtained high-density monomolecular poly(ethylene glycol) (PEG) layers by solvent-free coupling of homo-bifunctional PEGs (2,000 g/mol) at 75 degrees C to silica surfaces silanized with glycidyloxipropyltrimethoxysilane (GOPTS). Characterization by ellipsometry and contact angles revealed that PEG layers up to 3.4 ng/mm2 with low roughness and flexibility were obtained. Specific and non-specific binding at these PEG surfaces was monitored by reflectometric interference spectroscopy (RIfS). No significant non-specific adsorption upon incubation of 1 mg/ml ovalbumin was detectable (< 10 pg/mm2), and 150 pg/mm2 upon incubation of 10% calf serum, less than 10% of the amount adsorbed to the solely silanized surfaces. The terminal functional groups of the PEG layers were utilized to couple ligands and a protein. Specific protein interaction with these immobilized compounds was detected with saturation loadings in the range of protein monolayers (2-4 ng/mm2). The excellent functional properties, the high stability of the layers, the generic and practical coupling procedure and the versatility for immobilizing compounds of very different functionality make these PEG layers very attractive for application in label-free detection with silica or metal-oxide based transducers.

Chemical and biochemical sensors based on interferometry at thin (multi-) layers☆

Abstract Spectral interferometry is presented as a tool to monitor the swelling of polymers caused by organic gases or hydrocarbons in waste water as well as the adsorption and interaction of antigens and antibodies in immunoreactions. Modern diode-array technology allows the consequent observation of changes in optical pathlength on a fractional nanometer scale with subsecond repetition times. The theory of multiple-reflection principles in white-light interferometry determines the possibilities and limitations of this method. The optical set-up and some applications in gas sensing and label-free immunosensing are discussed with respect to the sensitivity, selectivity and limits of detection at present.

Surface modification for direct immunoprobes

The modification of glass-type surfaces by several hydrophilic polymers of different molecular masses and functional properties [chitosan, dextran, poly(oxyethylene), poly(ethyleneimine) and poply(acrylamide)] with respect to the application for direct immunoprobes was investigated. Activation of the surface was carried out by silanisation and the polymers were coupled to the surface via amide bonds. The carboxyl derivative of a hapten was attached to the functional groups of the polymers by carbodiimide-activated coupling. As a reference system, the ligand was directly coupled to the silanised surface. Non-specific protein adsorption, specific binding of antibodies and regeneration were monitored by evaluation of reflectance spectra obtained by white light interference at a thin silica layer (RifS). All polymer modified layers showed improved properties compared to those with direct attachment of the hapten. The non-specific adsorption was reduced to 5-50%. Binding of a specific antibody was significantly increased by the polymer modification: Mass transport limited binding of the specific antibody in low concentrations (30 nM) up to a surface coverage value of 2 ng/mm2 and a maximum surface coverage in the range of a monolayer of IgG (5-6 ng/mm2) was observed for most of the polymers. The surface coverage found for IgG bound specifically to the dextran-modified surface exceeded a protein monolayer.

A direct optical immunosensor for atrazine detection

Abstract Immunoanalytical techniques represent one of the most important applications of biomolecules in analytical procedures. Direct monitoring of immunoreactions by an analytical device is a particularly attractive approach to environmental sensing as it offers speed, a simple test scheme and does not require labelled compounds. Target limits of detection for pesticides are imposed by the EU drinking water act (0.1 μg/l for a single pesticide). A competitive test format with surface immobilised antigen is common for pesticide detection with direct immunosensors. Free pesticide binds to antibody in solution decreasing the amount of antibody binding to the transducer. As the concentration of antibody in a competitive test must be comparable to the concentration of the analyte (0.1 ppb atrazine ≈ 0.5 nM), the transducer must be able to resolve changes in antibody concentration below 1 nM. A prototype atrazine sensor based on reflectometric interference spectroscopy (RIFS) was investigated. The slope of the binding curve was used as a measure for antibody concentration. Binding curves of monoclonal anti-atrazine to an atrazine-modified surface were recorded at different concentrations. A reproducibility of 10% was found for the determination of antibody concentrations (100 to 1000 ng/ml). In competitive binding experiments with free atrazine inhibition of 50% was observed at 0.5 ppb atrazine (500 ng/ml antibody) and at 1 ppb (1 μg/ml antibody). The results are discussed with respect to the theoretical performance of the device and practical requirements. A comparison to other optical transducers and a critical discussion about the feasibility of direct optical immunosensors in pesticide detection is given.

Optical multiple-analyte immunosensor for water pollution control.

A prototype of a portable optical immunosensor (called river analyser) has been developed. It can be applied for the monitoring of surface water quality. Antibodies carrying a fluorescent label are used for the specific recognition of pollutants, such as frequently applied pesticides. The transduction principle is based on total internal reflection fluorescence (TIRF). The outstanding advantage of the river analyser is, that at least three analytes can be detected simultaneously in one sample. Test cycles and fluid handling are automated and enable unattended measuring.

Determination of simazine in water samples by waveguide surface plasmon resonance

We assessed a new sensing device based on the monitoring of immunobinding reactions using waveguide surface plasmon resonance (WSPR) for the determination of simazine in water samples. Standard solutions between 0.1 and 1.0 μg l−1 analysed in triplicate showed a mean within-day variability of 5%. Calibration curves for the same standards conducted on five consecutive days showed a 14% mean day-to-day variability. The detection limit calculated as three standard deviations below the mean blank value was 0.2 μg l−1. The upper limit of the working range calculated as a 90% decrease in the blank signal was 2.4 μg l−1. The cross-reactivity of atrazine and terbuthylazine was 61 and 63%, respectively. The recovery from spiked natural ground- and surface-water samples ranged from 55 to 153% for spikes ranging from 0.1 to 1.0 μg l−1. For the 11 surface- and 8 ground-water samples tested, the correlation coefficient between WSPR and high pressure liquid chromatography/gas chromatography (HPLC/GC) values was significant (p<0.05) when the chromatography values were calculated as the weighted sum of simazine and atrazine, taking into account the predetermined cross-reactivity of the latter in the WSPR determination. The present system is therefore better suited for screening groups of pesticides than for the determination of a single molecule. An attempt at analysing a soil water sample proved unsuccessful due to interference probably resulting from strong non-selective polyanion-polycation binding to the transducer surface which includes a basic amino dextran. The total duration of one determination, 22 min, enables almost immediate measurements without any sample pretreatment other than 0.45 μm filtration. No significant alteration of the sensor was observed after 200 determinations.

Integrated optical surface plasmon resonance immunoprobe for simazine detection

This paper presents the detailed design and characterisation of a regenerable integrated optical surface plasmon resonance immunoprobe as a detector for the triazine herbicide simazine. A sensor design theoretically optimised for use in the aqueous environment is presented and its fabrication described. Experimental results on the sensitivity to changes in bulk refractive index of the analyte and on non-specific binding of ovalbumin are presented. Binding inhibition immunoassays were conducted for simazine and the lower limit of detection determined to be 0.16 microgram/l using anti-simazine IgG antibodies and 0.11 microgram/l using anti-simazine Fab fragments. A sample test cycle of 20 min was established.

Protein interactions in covalently attached dextran layers

Abstract Protein interactions with polymeric carbohydrates play an important role for application in chromatography, biomaterials and biophysics. In this study, we present a detailed morphological and functional characterization of covalently side-bound dextran layers by spectroscopic ellipsometry (SE) and reflectometric interference spectroscopy (RIfS). The surface chemistry was monitored step-by-step by ellipsometric characterization of the surface loading. Dextrans of various molecular masses (10–2000 kD) were immobilized leading to surface loadings between 3 and 8 ng mm−2. The refractive indices of the covalently attached dextran layers under atmospheric conditions (nD=1.51) were very close to the refractive index of a spin-coated dextran layer (nD=1.52) indicating dense and homogeneous coverage achieved by the coupling chemistry. Under buffer solution, refractive indices between 1.34 and 1.365 and thicknesses between 20 and 40 nm of these dextran layers were determined. A dextran concentration in the hydrated layers of 0.05–0.21 g cm−3 was estimated from the refractive index. The density and the thickness of the hydrated layers increased with molecular mass of the dextran. Non-specific binding was strongly reduced by the dextran layers and decreased with increasing thickness and density of the layer. Specific antibody binding to haptens immobilized in the dextran layer lead to an increase of both the density and the thickness of the layers. Time resolved detection by RIfS indicated significant decrease of protein mobility in the dextran layer. From these results we conclude that the functional properties of dextran layers with respect to protein interactions are determined by their effective pore-size, which is controlled by the number of bonds, the surface loading and the concentration of charged groups in the polymer.

Quartz crystal microbalances for quantitative biosensing and characterizing protein multilayers

The use of quartz crystal microbalances (QCMs) for quantitative biosensing and characterization of protein multilayers is demonstrated in three case studies. Monolayers of QCM-based affinity biosensors were investigated first. Layers of a thiol-containing synthetic peptide constituting an epitope of the foot-and-mouse-disease virus were formed on gold electrodes via self-assembly. The binding of specific antibodies to epitope-modified gold electrodes was detected for different concentrations of antibody solutions. Oligolayers were studied in a second set of experiments. Dextran hydrogels were modified by thrombin inhibitors. The QCM response was used in a competitive binding assay to identify inhibitors for thrombin at different concentrations. Multilayers of proteins formed by self-assembly of a biotin-conjugate and streptavidin were investigated next. The QCM frequency response was monitored as a function of layer thickness up to 20 protein layers. A linear frequency decay was observed with increasing thickness. The decay per layer remained constant, thus indicating perfect mass coupling to the substrate. Frequency changes a factor of four higher were obtained in buffer solution as compared to measurements in dry air. This indicates a significant incorporation of water (75% weight) in the protein layers. This water behaves like a solid concerning the shear mode coupling to the substrate. The outlook discusses briefly the need for controlled molecular engineering of overlayers for subsequent QCM analysis, and the importance of an additional multiparameter analysis with other transducer principles and with additional techniques of interface analysis to characterize the mechanical coupling of overlayers as biosensor coatings. A promising trend concerns the use of QCM-arrays for screening experiments.