A new waveguide interferometer for the label free detection of biomolecules

The rapid development in life science research requires continually new methods for the analysis of biomolecular interactions. Optical detection systems based on evanescent field sensing for biomolecular binding studies are marked by a fast increase of their applications in fields like diagnostics, health care, screening or assay development. For a successful implementation the optical biosensors need to be highly sensitive, selective and accurate while allowing a wide range of applications. In this thesis a highly sensitive label-free detection method based on the optical principle of a Young interferometer and a description of the design and realization of the system is presented including an experimental characterization and a validation of the biosensor system by application measurements. In chapter 2, the theoretical basis of the propagation of light in planar waveguides and a background of evanescent field sensors is provided as well as calculations concerning the optimization of waveguide sensitivity for the interferometric biosensor proposed in this thesis. One section is devoted to the theoretical basis of interferometry, especially the optical principle of the Young interferometer since the proposed design for the interferometric biosensor is based on the Young configuration. For signal evaluation, algorithms based on Fast Fourier Transform are described and the implementation in the biosensor system proposed along with system principles and relationships and signal filtering methods for noise reduction. Chapter 3 is devoted to theoretical considerations on the biochemistry for waveguide surface functionalization. The kinetic theory of biochemical interactions such as antibody-antigen binding events is described in detail with respect to the evaluation of the application measurements in chapter 5. Furthermore a biochemical background on the application measurement reagents is given. The design of the interferometric biosensor is presented in chapter 4, describing two different readout schemes: a flow-cell system implementing a two-channel flow cell as fluidic element, and a system setup designed for the readout of microtiterplate (MTP)-formatted wellplates. The system components and system designs of the interferometric biosensor and their performance are discussed in detail. From the analysis of system components several conclusions can be made, such as the choice of waveguide chip materials, the light source and the CCD detector. The characteristics of the chosen components influence the final design and setup, leading to an optimal set of parameters for the interferometric biosensor. Further a detailed system characterization by refractometric test measurements is provided including an evaluation of the experimental results compared to theoretical calculations and simulations. Initial refractometric test measurements on the flow-cell system with glycerin show an effective refractive index resolution of ≈ 6. 0 · 10-8, corresponding to a refractive index resolution of ≈ 7. 5 · 10-7 for TE mode and ≈ 2. 7 · 10-7 for TM mode, respectively, at a sampling rate of 1 Hz. It is also found that the CCD sensor is the limiting system component concerning phase noise, whereas the phase noise is approximately equal for all three light sources tested and therefore the light source can be excluded as limiting factor. Signal filtering methods such as average filters are described and analyzed concerning their suitability for phase noise reduction. The experimentally derived sensitivity constants of the interferometer system are in very good accordance with the theoretical values derived from waveguide theory in chapter 2. The flow-cell system shows a good long-term stability with a typical drift of < 1 · 10-6/h in neff. This chapter also presents the system design and describes the realized setup of the interferometric biosensor with an MTP-formatted 8-well frame as fluidic element and newly introduced system components that are different from the flow cell system are discussed. Suitable system components are chosen for an optimal set of design parameters for the interferometric biosensor. The proposed biosensor system is characterized by refractometric test measurements with ethanol solutions and the experimental results are evaluated and compared to theory and the results obtained with the flow cell system. The refractometric test measurements on the microplate system show an effective refractive index resolution of ≈ 1. 0 · 10-7, corresponding to a refractive index resolution of ≈ 1. 2 · 10-6 for TE mode and ≈ 4. 5 · 10-7 for TM mode, respectively, at a sampling rate of 1 Hz. We found that, apart from the CCD sensor being the limiting system component concerning phase noise, the resolution of the microplate system is finally limited by the repeatability of the recorded data values. The experimentally derived sensitivity constants of the microplate interferometer system are also in good accordance with the theoretical values derived from waveguide theory in chapter 2. The system shows a long-term stability with a typical drift of < 5 · 10-5/h in neff and a repeatability of < 2 · 10-7 in neff. Chapter 5 presents a validation of the interferometric biosensor by biological application measurements on both systems and their discussion. In this chapter a detailed description of the materials and methods used for application and test measurements, a characterization of the surface chemistry on the Ta2O5 waveguide chips and finally application measurements on different affinity systems is provided. Differently functionalized surfaces are characterized by contact angle measurements and compared to literature values. Experiments with fluorescently labeled streptavidin allow a detailed characterization of different surface modifications and a direct comparison to measurements with the interferometric biosensor with the aim to find a stable and robust surface functionalization for the immobilization of biotinylated reagents. The immunoassay protein G -IgG is tested on the flow-cell system as well as on the microplate system, yielding affinity rate constants that are in good agreement with values found in literature. Experimental data obtained with the two-channel flow-cell system allow an evaluation of the reaction kinetics, while the microplate system is suited for the parallel detection of several analyte concentrations. Furthermore the direct detection of IgG by immobilized protein A is shown and the affinity rate constants determined. Test measurements with biotinylated cytochrome c on a streptavidin-functionalized surface compared directly to the same assay performed on the commercial biosensor system BIACORE 1000 show the suitability of the developed surface chemistry for the Ta2O5 waveguide chips implemented in the interferometric biosensor. The interferometric biosensor has been successfully used for the detection of the affinity system -NPT IgG -E2-NPT developed at Novartis in Basel. We tested the assay on both the interferometric biosensor (flow-cell system) and the commercial BIACORE 2000 system provided by Novartis. A comparison of the surface chemistries used on both biosensor systems is presented, and a series of measurements to detect the analyte from different sample buffers. We found that the performance of the interferometric biosensor system proposed in this thesis is approximately equal to the BIACORE system with a detection limit for the analyte in the low picomolar range. Measurements in cell lysate as sample matrix show that with the interferometric biosensor the analyte can be detected even out of a complex sample matrix without significant unspecific binding. In chapter 6, conclusions concerning the development of the interferometric biosensor and the results from the experimental characterization and validation are presented. A comparison between the proposed interferometric biosensor and other label-free biosensors is given. The chapter concludes with an outlook concerning further improvements of the resolution and surface chemistry and further possibilities for developments to adapt the sensor design to high throughput applications.