Electron transfer mediated by glucose oxidase at the liquid/liquid interface.

In order to establish an experimental basis for exploring the reactivity of membrane-bound redox enzymes using electrochemistry at an organic/aqueous interface, the reactivity of glucose oxidase adsorbed at the dichloroethane/water interface has been studied. Turnover of glucose in the aqueous phase mediated by dimethyl ferricenium electrogenerated in the organic phase was measured by measuring the feedback current caused by recycling the mediator as the generator electrode approached close to the interface from the organic side. An unexpected self-exchange reaction of the ferrocene at the interface was suppressed by adsorption of a surfactant. The interfacial enzyme reaction could be distinguished from reaction within the bulk of the aqueous phase. Reaction within a protein-surfactant film formed at the interface is conjectured.

Development and comparison of biosensors for in-vivo applications.

Electrochemical biosensors have been of increasing interest, especially those developed to be directly applied in diagnostic areas, such as neuroscience. We have been interested in developing a range of biosensors for monitoring glucose, lactate, pyruvate, and glutamate in order to study on-line both brain function in the laboratory and to monitor brain health in neurointensive care. For a biosensor to function effectively in these situations, it has to combine the following characteristics: quick response and high sensitivity, good reproducibility and adequate stability. In this study we compared the performance of a number of different amperometric biosensors strategies. These included ferrocene mediation of immobilised enzymes (system A), a redox hydrogel based system (system B), and a conducting polymer approach using polyaniline (system C). All assays were operated as flow-injection systems with upstream immobilised enzyme beds if necessary. When calibrated for H2O2 systems A and B reacted quickly enough to give quantitative conversion up to 0.2 mM. Above this concentration the response was limited by horseradish peroxidase enzyme kinetics and eventually enzyme loading. System C showed a restricted H2O2 response. When calibrated for glucose (by use of immobilised glucose oxidase) system B exhibited the highest sensitivity but its analytical range was restricted because the system became limited by H2O2 response. System A had low sensitivity for analyte compared to H2O2 and system B, but a greater useful range. Problems of mediator cycling between the immobilised enzymes are discussed. System C gave an excellent linear range but sensitivity was limited by background noise. Stability and reproducibility of the systems are also described. In conclusion, from this study the ferrocene system proved to be overall most useful and has now been used in the first dual on-line monitoring of glucose and lactate in patients in neurointensive care.