Freshwater harmful algal blooms: Beyond biomass as a predictor of toxicity

Background/Question/Methods Lakes within the upper mid-western United States are a valuable source of water, seafood and recreation. However, as more of our water bodies are becoming eutrophic, these lakes increasingly pose serious health risks due to toxic cyanobacteria. Local health departments close popular beaches, swimming locations, or entire lakes in order to protect public health. Typically closings are initiated based on monitoring of cyanobacterial biomass and unsolicited reports (i.e. media and direct call-ins). We have measured biomass, cyanobacterial genotype composition, abundance of Microcystis species, and four microcystin variants at twelve locations in four popular recreational eutrophic lakes in the capital city of Madison, WI on a weekly basis throughout the open water season. Biomass was estimated from the concentration of phycocyanin. Microcystin-LR, LA, YR, and RR were measured by triple quadrupole mass spectrometry with electrospray ionization. Cyanobacterial genotype composition was assessed by automated phycocyanin interspacer sequence analysis (APISA), which delineates cyanobacterial genera at the sub-genus level. The abundance of Microcystis and microcystin toxin genes was based on quantitative PCR analysis of the Microcystis 16S ribosomal RNA genes and the mcyA/mcyE genes, respectively. We also gathered reports of human illnesses reported to the Dane County local health department and news media thought to be due to cyanobacterial toxins.

Results/Conclusions

We found that microcystin dynamics did not correlate (R< 0.3) with typical chemical and physical limnological measurements (i.e. nutrients, temperature, lake stability), cyanobacterial biomass, Microcystis abundance, or the abundance of toxin genes in all lakes and locations. For example, at times microcystin-LR was elevated (>10 mg/L), even during times of relatively low cyanobacterial biomass or Microcystis abundance. We found that 80-100% of Microcystis cells at all locations contained toxin genes while no other microcystin producers were present. In addition, the relative abundance of certain Microcystis genotypes increased during peaks in toxin concentration indicating these genotypes may be responsible for toxin production. Altogether the data suggest that the microcystin content of individual Microcystis genotypes, may contribute to high toxin levels even when overall cyanobacterial biomass is low. Furthermore, the toxin containing genotype varies both spatially and temporally within lakes. As a consequence, beach and/or lake closings based on monitoring of total cyanobacterial biomass are likely to be insufficient for protecting public health. In this study microcystin concentrations were elevated above acceptable recreational limits as stated by the World Health Organization when lakes were not closed, resulting in seven reported human illnesses.