A novel computer-generated procedure for fabrication of an electrochemical biosensor and its application to glucose determination in three-dimensional cell culture based on exploiting first-order advantage

In this study, we developed a groundbreaking electrochemical biosensor by modifying a glassy carbon electrode (GCE) with multiwalled carbon nanotubes-ionic liquid (MWCNTs-IL), chitosan-ionic liquid (Ch-IL), and electrochemically synthesized alloy nanoparticles made of gold, platinum, and palladium (AuPtPd NPs). The composite of AuPtPd NPs/Ch-IL/MWCNTs-IL/GCE was treated with glutaraldehyde, facilitating covalent cross-linking between the amine groups in chitosan and the free aldehyde groups in glutaraldehyde. This process constructed a three-dimensional network that enabled the immobilization of glucose oxidase (GO), resulting in the final biosensor structure: GO/AuPtPd NPs/Ch-IL/MWCNTs-IL/GCE. This structure served as a rotating working electrode to determine glucose levels in three-dimensional cell cultures, leveraging first-order advantages amidst the substances present in the cell culture matrix. To optimize the biosensor design and its response, we employed a two-step experimental design approach. We evaluated its amperometric responses using various first-order multivariate calibration algorithms, including PLS-1, rPLS, CPR, LS-SVM, RBF-PLS, RBF-ANN, BP-ANN, and PCR, ultimately identifying the best-performing algorithm for our biosensor. Our findings indicated that the biosensor's prediction performance was maximized with LS-SVM support. Additionally, the biosensor utilizing RBF-ANN effectively detected glucose concentrations ranging from 0.5 to 35 fM, with a limit of detection of 0.21 fM and sensitivity measured at 0.9931 μA/fM. The calculated Michaelis-Menten constant was determined to be 0.38 fM, showcasing the remarkable capability of the biosensor for glucose determination in three-dimensional cell cultures. • A novel enzymatic electrochemical biosensor was fabricated. • First-order advantage was exploited for determination of glucose in three-dimensional cell cultures. • PLS-1, rPLS, CPR, LS-SVM, RBF-PLS, RBF-ANN, BP-ANN, and PCR were used. • The best performance was observed by assisting the biosensor with RBF-ANN. • Response of the biosensor was accurate, repeatable, reproducible, stable and precise.