RESUMO
The effects of modifications to an existing protocol for a surface plasmon resonance biosensor-based inhibition immunoassay for progesterone in cow's milk with a sensitivity of 3.5 ng/mL were examined to establish if the detection limit could be further reduced to broaden the potential applications of the assay. The mean relative standard deviation of duplicate measurements was 0.62% and the high precision resulted in very low values for the lower detection limits. Hence, the standard concentrations giving 95% maximum binding [effective dose (ED 95)] were compared instead. The ED 95 was not affected within a running temperature range of 20-37 degrees C, or at a flow rate and a contact time above 20 microL/min and 90 s, respectively. Increasing both the absolute sample volume and the antibody dilution improved sensitivity. However, there was a simultaneous reduction in the working range when the assay was applied to milk due to nonspecific binding. Less antibody was associated with large decreases in the maximum binding, but because the high precision extended over a broad analytical range, an ED 95 of 0.4-0.6 ng/mL in milk and 35-60 pg/mL in HBS-EP buffer were achieved. Thus, simple procedural modifications with the same sensor chip can alter performance characteristics of the assay as required for different applications.