Artificial antibodies as receptor layers for chemical sensor

Chemical sensors are miniaturized devices capable of continuously recognizing chemical compounds. Biosensors with high selectivities have inadequate robustness for process control or applications under harsh environmental conditions. Furthermore, the availability of biological components, such as antibodies, is limited. Our project aims at the development of robust synthetic materials with high selectivities for industrially applicable sensors. "Molecular imprinting" of polymers and sol-gels enables us to combine the advantages of biosensors and chemical sensors in the form of so-called "synthetic antibodies".These artificial receptors will be used to selectively detect various forms of analytes having nanometer or micrometer dimensions. Analytes of interest are: complex mixtures, enzymes, viruses, microorganisms as well as tumor and blood cells. We will characterize complex mixtures via their "fingerprints" using "mixture imprinted polymers". The selective detection of enzymes, viruses, microorganisms and mammalian cells is a new frontier for MIP sensor applications and has been successfully demonstrated in the case of yeasts. Here, we are using stamping procedures to generate honeycomb-like cell moulds. Amongst others, mass-sensitive transducers, such as shear transverse wave resonators, normally used as electronic components in cellular phones e.g., will be used to detect even single cells. The transducers coated with artificial antibodies will be highly adaptable and can be used for environmental monitoring or food control. Additionally, we want to demonstrate the technical feasibility of this new type of sensor for industrial manufacturing.

Within this project we succeeded in rationally designing artificial, polymer-based "antibodies" as sensor materials against analytes ranging from sub-nanometre to micrometre in size. Combined with different sensor devices, such as e.g. the mass-sensitive quartz crystal microbalance (QCM) or surface transverse wave resonator (STW), such materials yield excellent analytical tools. We developed all sensitive materials by the means of imprinting processes, where recognition sites are generated by self-organzing the growing polymer material around a template compound, usually the analyte-to-be. This strategy leads to excellent selectivities, as we could show by the example of the three xylene isomers, which can be distinguished from each other with selectivity factors in the range of 10 to 100. Detecting a wide variety of bioanalytes was a fundamental part of the project. The initial experiments were done with yeast cells self-assembled on a stamp and then pressed into a film of the forming polymer. This leads to surface cavities that selectively re-incorporate the template cells. When applying high-frequency STWs, even single cell recognition can be detected. With red blood cells (erythrocytes) imprinting leads to blood-group selective surfaces. As erythrocytes of different blood groops only differ by the chemical pattern on the cell surface, we proved that our imprints both shape-selective and depict the chemical composition of the cell wall and therefore lead to antibody-like behaviour. Viruses represent an especially important class of analytes, because they are usually analyzed by infectivity tests that require time and high effort. With our MIP-technique we succeeded to coat QCM surfaces with virus-sensitive polymers. When templating with tobacco mosaic virus (TMV) the materials e.g. show selectivity factors of 104 between TMV and a human rhinovirus (HRV). This excellent shape-selectivity and surface functionality enabled us to detect TMV infection of plants by screening the crude plant sap. Virus imprints also show very pronounced chemical functionality: layers templated with different serotypes of HRV (e.g HRV-2 and HRV-14) favour their respective template over the other serotype by a selectivity factor of five. Given the fact that the shape of the different serotypes is exactly the same, this is even more remarkable. Selectivities and sensitivities thus indicate that the analyte and the layer undergo a variety of non-covalent interactions that lead to the sensor effects.