Development and testing of a fluorescence biosensor for glucose sensing

Rapid, accurate, and minimally-invasive biosensors for glucose measurement have the potential to enhance management of diabetes mellitus and improve patient outcome in intensive care settings. Recent studies have indicated that implantable biosensors based on Förster Resonance Energy Transfer (FRET) can provide high sensitivity in quantifying glucose concentrations. However, standard approaches for determining the potential for interference from other biological constituents have not been established. The aim of this work was to design and optimize a FRET-based glucose sensor and assess its specificity to glucose. A sensor based on competitive binding between concanavalin A and dextran, labeled with long-wavelength acceptor and donor fluorophores, was developed. This process included optimization of dextran molecular weight and donor concentration, acceptor to donor ratio, and hydrogel concentration, as well as the number of polymer layers for encapsulation. The biosensor performance was characterized in terms of its response to clinically relevant glucose concentrations. The potential for interference and the development of test methods to evaluate this effect were studied using a potential clinical interferent, maltose. Results indicated that our biosensor had a prediction accuracy of better than 11% and that the robustness to maltose was highly dependent on glucose level.