Differentiation between microcystin contaminated and uncontaminated fish by determination of unconjugated MCs using an ELISA anti‐adda test based on receiver‐operating characteristic curves threshold values: Application to Tinca tinca from natural ponds

The aim of this study was to evaluate whether the enzyme‐linked immunosorbent assay (ELISA) anti‐Adda technique could be used to monitor free microcystins (MCs) in biological samples from fish naturally exposed to toxic cyanobacteria by using receiver operating characteristic (ROC) curve software to establish an optimal cut‐off value for MCs. The cut‐off value determined by ROC curve analysis in tench (Tinca tinca) exposed to MCs under laboratory conditions by ROC curve analysis was 5.90‐μg MCs/kg tissue dry weight (d.w.) with a sensitivity of 93.3%. This value was applied in fish samples from natural ponds (Extremadura, Spain) in order to asses its potential MCs bioaccumulation by classifying samples as either true positive (TP), false positive (FP), true negative (TN), or false negative (FN). In this work, it has been demonstrated that toxic cyanobacteria, mainly Microcystis aeruginosa, Aphanizomenon issatchenkoi, and Anabaena spiroides, were present in two of these ponds, Barruecos de Abajo (BDown) and Barruecos de Arriba (BUp). The MCs levels were detected in waters from both ponds with an anti‐MC‐LR ELISA immunoassay and were of similar values (between 3.8–6.5‐μg MC‐LR equivalent/L in BDown pond and 4.8–6.0‐μg MC‐LR equivalent/L in BUp). The MCs cut‐off values were applied in livers from fish collected from these two ponds using the ELISA anti‐Adda technique. A total of 83% of samples from BDown pond and only 42% from BUp were TP with values of free MCs higher than 8.8‐μg MCs/kg tissue (d.w.). © 2009 Wiley Periodicals, Inc. Environ Toxicol 26: 45–56, 2011.