ABSTRACT
Increasing concern for human health related to cyanotoxin exposure imposes the identification of pattern and level of exposure; however, current monitoring programs, based on cyanobacteria cell counts, could be inadequate. An integrated approach has been applied to a small lake in Italy, affected by Planktothrix rubescens blooms, to provide a scientific basis for appropriate monitoring program design. The cyanobacterium dynamic, the lake physicochemical and trophic status, expressed as nutrients concentration and recycling rates due to bacterial activity, the identification/quantification of toxic genotype and cyanotoxin concentration have been studied. Our results indicate that low levels of nutrients are not a marker for low risk of P. rubescens proliferation and confirm that cyanobacterial density solely is not a reliable parameter to assess human exposure. The ratio between toxic/non-toxic cells, and toxin concentrations, which can be better explained by toxic population dynamic, are much more diagnostic, although varying with time and environmental conditions. The toxic fraction within P. rubescens population is generally high (30-100%) and increases with water depth. The ratio toxic/non-toxic cells is lowest during the bloom, suggesting a competitive advantage for non-toxic cells. Therefore, when P. rubescens is the dominant species, it is important to analyze samples below the thermocline, and quantitatively estimate toxic genotype abundance. In addition, the identification of cyanotoxin content and congeners profile, with different toxic potential, are crucial for risk assessment.
