Freshwater cyanobacteria blooms represent a risk to ecological and human health through induction of anoxia and release of potent toxins; both conditions require water management to mitigate risks. Many cyanobacteria taxa may produce microcystins, a group of toxic cyclic heptapeptides. Understanding the relationships between the abiotic drivers of microcystins and their occurrence would assist in the implementation of targeted, cost-effective solutions to maintain safe drinking and recreational waters. Cyanobacteria and microcystins were measured by flow cytometry and liquid chromatography coupled to tandem mass spectrometry in two interconnected reservoirs varying in age and management regimes, in southern Britain over a 12-month period. Microcystins were detected in both reservoirs, with significantly higher concentrations in the southern lake (maximum concentration >7 µg L⁻¹). Elevated microcystin concentrations were not positively correlated with numbers of cyanobacterial cells, but multiple linear regression analysis suggested temperature and dissolved oxygen explained a significant amount of the variability in microcystin across both reservoirs. The presence of a managed fishery in one lake was associated with decreased microcystin levels, suggestive of top down control on cyanobacterial populations. This study supports the need to develop inclusive, multifactor holistic water management strategies to control cyanobacterial risks in freshwater bodies.

Cyanobacterial blooms are often associated with the presence of harmful natural compounds which can cause adverse health effects in both humans and animals. One family of these compounds, known as anatoxins, have been linked to the rapid deaths of cattle and dogs through neurotoxicological action. Here, we report the findings resulting from the death of a dog at a freshwater reservoir in SW England. Poisoning was rapid following exposure to material at the side of the lake. Clinical signs included neurological distress, diaphragmatic paralysis and asphyxia prior to death after 45 min of exposure. Analysis by HILIC-MS/MS of urine and stomach content samples from the dog revealed the detection of anatoxin-a and dihydroanatoxin-a in both samples with higher concentrations of the latter quantified in both matrices. Detection and quantitative accuracy was further confirmed with use of accurate mass LC-HRMS. Additional anatoxin analogues were also detected by LC-HRMS, including 4-keto anatoxin-a, 4-keto-homo anatoxin-a, expoxy anatoxin-a and epoxy homo anatoxin-a. The conclusion of neurotoxicosis was confirmed with the use of two independent analytical methods showing positive detection and significantly high quantified concentrations of these neurotoxins in clinical samples. Together with the clinical signs observed, we have confirmed that anatoxins were responsible for the rapid death of the dog in this case.