Multi-dimensional fluorescence spectroscopy coupled with chemometrics in rapid antibiotic detection and discrimination

Abstract

Antibiotic residues in animal-derived foods pose risk to human health including chemical poisoning and antimicrobial resistance, and therefore, constant monitoring throughout the food supply chain is important. In the present study multi-dimensional fluorescence spectroscopy (3D and 2D) coupled with chemometric tools were tested for potential application to detect, discriminate and quantify penicillin G (PG), sulfadiazine (SF) and tetracycline (TC) in milk by direct measurement. Qualitative and quantitative calibration models were developed for prediction of antibiotic residues. Results demonstrated optimal discrimination of milk samples on the basis of antibiotic type and concentration with close to 100% of accuracy. Negative correlation between antibiotic concentration and fluorescence peak height was displayed (r ≥ 0.963 and p ≤ 0.002). A great potential for quantitative antibiotic determination was established with R2 > 0.9 and low standard errors of estimation indicating acceptable precision for the developed technique. Fluorescence spectroscopy demonstrated high specificity and sensitivity with detection limit below the maximum residue limit of PG, SF and TC in milk. Therefore, fluorescence spectroscopy can be used as an alternative method for rapid screening of antibiotic residues in milk at collection centers and processing plants to ensure product quality and safety.