Two new competitive ELISA methods for the determination of caffeine and cotinine in wastewater and river waters
Two enzyme-linked immunosorbent assays (ELISAs) for the determination of caffeine and cotinine in water samples were validated and evaluated for the analysis of river and wastewater samples. The caffeine ELISA shows an IC50 ranging from 2.21 to 2.73 ng ml−1, a limit of detection (LOD) of 0.135 ng ml−1 and cross-reacts with paraxanthine (CR% = 63), theophylline (CR% = 5.2), and theobromine (CR% = 2.8); while the cotinine ELISA shows an IC50 ranging from 0.38 to 0.45 ng ml−1, a LOD of 0.047 ng ml−1 and cross-reacts with 3′-hydroxycotinine (CR% = 32) and nicotine (CR% = 1.1). Calibration curves were constructed in buffer in the concentration range 0.175–5 ng ml−1 for the caffeine ELISA and 0.05–5 ng ml−1 for the cotinine ELISA. Precision and accuracy were also evaluated. For the two assays, repeatability and reproducibility, expressed as the relative standard deviation of the absorbance, were within 3% and 8%, respectively. From the spiking experiment, the recovery of caffeine found in river and wastewater was between 94–101% and 89–110%, respectively; while the recovery of cotinine found in river and wastewater was between 95–105% and 99–115%, respectively. Fifteen river water samples and four wastewater samples from different wastewater treatment plants (WWTPs) were analyzed by ELISA and confirmed by LC-MS/MS. Comparing the two methods a correlation coefficient of 0.935 was obtained, showing a small overestimation of the ELISAs with respect to LC-MS/MS.