The aim of this study was to determine if the cytotoxic and genotoxic responses of Allium cepa are effective biomarkers of harmful effects caused by polluted river water and if changes in the responses reflect seasonality in the harmful effects. Samples were collected in the dry season (August 2011 and 2012) and rainy season (February 2012 and 2013) at sampling points on the Jaguari River and the Ribeirão Lavapés, in Brazil. Allium cepa bulbs were exposed to the samples, to positive controls (15 µg/L methyl methanesulfonate), and to negative controls (tap water). Three root tips from each bulb were then stained using the Feulgen reaction, then the micronucleus frequency, the mitotic index, and mitotic anomalies were measured. The total number of anomalies (stickiness, c-mitosis, multipolarity, chromosome bridges, and unidentified anomalies) in the rainy season (8.61 ± 3.65) and dry season (7.07 ± 2.96) were significantly different (U = 11.31, p = 0.04). Toxicity, indicated by the formation of micronuclei and the mitotic index, was higher in the February 2012 samples than in the August 2012 samples. The mean manganese concentration (0.13 mg/L) in the rainy season samples was higher than the maximum concentration permitted by the Brazilian National Environmental Council (<0.1 mg/L) and the manganese concentrations positively correlated with chromosomal aberration induction (p = 0.01, r = 0.69). In conclusion, the rainy season samples were more toxic than the dry season samples. This was probably related to rain water carrying compounds with potentially negative impacts into the rivers. These findings highlight the importance of biomonitoring studies and of treating wastewater in urban areas.
Keywords: Allium cepa; Biomonitoring; Pollution; Toxicity.