Unmasking the identity of toxigenic cyanobacteria driving a multi-toxin bloom by high-throughput sequencing of cyanotoxins genes and 16S rRNA metabarcoding.

Cyanobacterial harmful algal blooms (CyanoHABs) are complex communities that include coexisting toxic and non-toxic strains only distinguishable by genetic methods. This study shows a water-management oriented use of next generation sequencing (NGS) to specifically pinpoint toxigenic cyanobacteria within a bloom simultaneously containing three of the most widespread cyanotoxins (the hepatotoxins microcystins, MCs; and the neurotoxins anatoxin-a, ATX, and saxitoxins, STXs). The 2013 summer bloom in Rosarito reservoir (Spain) comprised 33 cyanobacterial OTUs based on 16S rRNA metabarcoding, 7 of which accounted for as much as 96.6% of the community. Cyanotoxins and their respective biosynthesis genes were concurrently present throughout the entire bloom event including: MCs and mcyE gene; ATX and anaF gene; and STXs and sxtI gene. NGS applied to amplicons of cyanotoxin-biosynthesis genes unveiled 6 toxigenic OTUs, comprising 3 involved in MCs production (Planktothrix agardhii and 2 Microcystis spp.), 2 in ATX production (Cuspidothrix issatschenkoi and Phormidium/Tychonema spp.) and 1 in STXs production (Aphanizomenon gracile). These toxigenic taxa were also present in 16S rRNA OTUs list and their relative abundance was positively correlated with the respective toxin concentrations. Our results point at MC-producing P. agardhii and ATX-producing C. issatschenkoi as the main contributors to the moderate toxin concentrations observed, and suggest that their distribution in Southern Europe is broader than previously thought. Our findings also stress the need for monitoring low-abundance cyanobacteria (<1% relative abundance) in cyanotoxicity studies, and provide novel data on the presence of picocyanobacteria and potentially ATX-producing benthic taxa (e.g., Phormidium) in deep thermally-stratified water bodies. This study showcases a straightforward use of amplicon metagenomics of cyanotoxin biosynthesis genes in a multi-toxin bloom thus illustrating the broad applicability of NGS for water management in risk-oriented monitoring of CyanoHABs.