Engineering receptors and antibodies for biosensors

Abstract

Biosensor sensitivity and selectivity depend essentially on the properties of the biorecognition elements to be used for analyte binding. Two principally different applications are considered, (1) effects monitoring with biological components as targets for bioeffective substances, among them endocrine disruptors; and (2) immunochemical analysis employing antibodies as binding proteins for a wide variety of analytes such as pesticides. Genetic engineering provides an elegant way not only for providing unlimited amounts of biorecognition molecules but also for the alteration of existing properties and the supplementation with additional functions. Instrumental applications were carried out with the optical sensor BIAcore™. The first example deals with the characterization of receptors. For this purpose, the human estrogen receptor α was used. Binding studies were carried out with natural as well as xenoestrogens. An equilibrium dissociation constant K d of 2.3×10 −10 (M) was derived for 17β-estradiol. A competition assay was performed with a bovine serum albumin (BSA)–17β-estradiol conjugate, immobilized at the optical sensor surface, and the free estrogen. The signals obtained represent estradiol equivalents. This format was transferred to a microplate-based enzyme-linked receptor assay. It reached a detection limit of 0.02 μg l −1 17β-estradiol and proved suitable for the detection of natural and synthetic estrogens as well as xenoestrogens in field studies. The second example is targeted at kinetic and affinity measurements of recombinant antibody fragments derived from antibody libraries with s-triazine selectivities. Different strategies for the synthesis of antibody fragment libraries, followed by the selection of specific antibody variants, were examined. An antibody library was derived from a set of B cells. Chain shuffling of the heavy and light chains provided the best binders. An enzyme linked immunosorbent assay (ELISA) was achieved for atrazine with an IC 50 of 0.9 μg l −1 and a detection limit of 0.2 μg l −1. The close relations between the optimization of recombinant antibodies by evolutionary strategies and genetic algorithms are considered.