Evaluation of quantitative real-time PCR to characterise spatial and temporal variations in cyanobacteria, Cylindrospermopsis raciborskii (Woloszynska) Seenaya et Subba Raju and cylindrospermopsin concentrations in three subtropical Australian reservoirs

Three drinking water storage reservoirs in subtropical southeast Queensland, Australia have regular blooms of the toxic cyanobacterium Cylindrospermopsis raciborskii that can produce cylindrospermopsins. We tested water samples from 16 sites in 3 reservoirs on 2 sampling occasions during a bloom of C. raciborskii in the austral summer and autumn of 2007. Using a range of parameters including quantitative real-time PCR, microscope cell counts and HPLC–MS/MS we correlated the 16S ribosomal RNA gene with total cyanobacteria, the rpoC1 gene with C. raciborskii cell concentrations, and the cyrC gene with cylindrospermopsin concentrations to assess spatial and temporal variability within and between reservoirs. While the correlation between cyrC and cylindrospermopsin cell quotas was good (mean r 2 = 0.61 for February samples and 0.75 for March samples), the correlation between total cyanobacteria and the 16S ribosomal RNA gene, and between C. raciborskii and the rpoC1 gene were poor indicating that further work is needed to develop these novel molecular methods. Spatial and temporal analysis of the distribution of rpoC1, cylindrospermopsin cell quotas, and a range of physical and chemical water quality parameters showed the greatest variation occurred between reservoirs, and within the largest and most spatially diverse reservoir. This suggests that populations of C. raciborskii strains with inherently different cylindrospermopsin cell quotas may be an important driver of toxicity in these reservoirs. An outcome of this study was the observation that deoxycylindrospermopsin always exceeded the cylindrospermopsin cell quota by up to 5-fold, and that a peak cell quota of 60 fg (cylindrospermopsin + deoxycylindrospermopsin) cell −1 was measured.