Discriminating macromolecular interactions based on an impedimetric fingerprint supported by multivariate data analysis for rapid and label-free Escherichia coli recognition in human urine.

This manuscript presents a novel approach to address the challenges of electrode fouling and highly complex electrode nanoarchitecture, which are primary concerns for biosensors operating in real environments. The proposed approach utilizes multiparametric impedance discriminant analysis (MIDA) to obtain a fingerprint of the macromolecular interactions on flat glassy carbon surfaces, achieved through self-organized, drop-cast, receptor-functionalized Au nanocube (AuNC) patterns. Real-time monitoring is combined with singular value decomposition and partial least squares discriminant analysis, which enables selective identification of the analyte from raw impedance data, without the use of electric equivalent circuits. As a proof-of-concept, the authors demonstrate the ability to detect Escherichia coli in real human urine using an aptamer-based biosensor that targets RNA polymerase. This is significant, as uropathogenic E. coli is a difficult-to-treat pathogen that is responsible for the majority of hospital-acquired urinary tract infection cases. The proposed approach offers a limit of detection of 11.3 CFU/mL for the uropathogenic E. coli strain No. 57, an analytical range in all studied concentrations (up to 105 CFU/mL), without the use of antifouling strategies, yet not being specific vs other E.coli strain studied (BL21(DE3)). The MIDA approach allowed to identify negative overpotentials (-0.35 to -0.10 V vs Ag/AgCl) as most suitable for the analysis, offering over 80% sensitivity and accuracy, and the measurement was carried out in just 2 min. Moreover, this approach is scalable and can be applied to other biosensor platforms.