SummaryBloom-forming cyanobacteria Planktothrix agardhii and P. rubescens are regularly involved in the occurrence of cyanotoxin in lakes and reservoirs. Besides microcystins (MCs), which inhibit eukaryotic protein phosphatase 1 and 2A, several families of bioactive peptides are produced, thereby resulting in impressive secondary metabolite structural diversity. This review will focus on the current knowledge of the phylogeny, morphology, and ecophysiological adaptations of Planktothrix as well as the toxins and bioactive peptides produced. The relatively well studied ecophysiological adaptations (buoyancy, shade tolerance, nutrient storage capacity) can partly explain the invasiveness of this group of cyanobacteria that bloom within short periods (weeks to months). The more recent elucidation of the genetic basis of toxin and bioactive peptide synthesis paved the way for investigating its regulation both in the laboratory using cell cultures as well as under field conditions. The high frequency of several toxin and bioactive peptide synthesis genes observed within P. agardhii and P. rubescens, but not for other Planktothrix species (e.g. P. pseudagardhii), suggests a potential functional linkage between bioactive peptide production and the colonization potential and possible dominance in habitats. It is hypothesized that, through toxin and bioactive peptide production, Planktothrix act as a niche constructor at the ecosystem scale, possibly resulting in an even higher ability to monopolize resources, positive feedback loops, and resilience under stable environmental conditions. Thus, refocusing harmful algal bloom management by integrating ecological and phylogenetic factors acting on toxin and bioactive peptide synthesis gene distribution and concentrations could increase the predictability of the risks originating from Planktothrix blooms.

Cyanobacteria are frequently involved in the formation of harmful algal blooms wherein, apart from the toxic microcystins, other groups of bioactive peptides are abundant as well, such as anabaenopeptins (APs). The APs are synthesized nonribosomally as cyclic hexapeptides with various amino acids at the exocyclic position. We investigated the presence and recombination of the AP synthesis gene cluster (apnA-E) through comparing 125 strains of the bloom-forming cyanobacterium Planktothrix spp., which were isolated from numerous shallow and deep water habitats in the temperate and tropical climatic zone. Ten ecologically divergent strains were purified and genome sequenced to compare their entire apnA-E gene cluster. In order to quantify apn gene distribution patterns, all the strains were investigated by PCR amplification of 2 kbp portions of the entire apn gene cluster without interruption. Within the 11 strains assigned to P. pseudagardhii, P. mougeotii, or P. tepida (Lineage 3), neither apnA-E genes nor remnants were observed. Within the P. agardhii/P. rubescens strains from shallow waters (Lineage 1, 52 strains), strains both carrying and lacking apn genes occurred, while among the strains lacking the apnA-E genes, the presence of the 5′end flanking region indicated a gene cluster deletion. Among the strains of the more derived deep water ecotype (Lineage 2, 62 strains), apnA-E genes were always present. A high similarity of apn genes of the genus Planktothrix when compared with strains of the genus Microcystis suggested its horizontal gene transfer during the speciation of P. agardhii/P. rubescens. Genetic analysis of the first (A1-) domain of the apnA gene, encoding synthesis of the exocyclic position of the AP molecule, revealed four genotype groups that corresponded with substrate activation. Groups of genotypes were either related to Arginine only, the coproduction of Arginine and Tyrosine or Arginine and Lysine, or even the coproduction of Arginine, Tyrosine, and Lysine in the exocyclic position of the AP-molecule. The increased structural diversity resulted from the evolution of apnA A1 genotypes through a small number of positively selected point mutations that occurred repeatedly and independently from phylogenetic association.

Harmful algal blooms formed by colony-forming cyanobacteria deteriorate water resources by producing cyanotoxins, which frequently occur at high intracellular concentrations. We aimed to localize toxic microcystins (Mcs) and bioactive anabaenopeptins (Aps) at the subcellular level under noninvasive conditions. Since both metabolites are synthesized nonribosomally, the relaxed specificity of key enzymes catalyzing substrate activation allowed chemical labeling through a standard coppercatalyzed click chemistry reaction. the genera Planktothrix and Microcystis specifically incorporated unnatural amino acids such as N-propargyloxy-carbonyl-L-lysine or O-propargyl-L-tyrosine, resulting in modified AP or MC peptides carrying the incorporated alkyne moiety. The labeled cells were quantitatively differentiated from the unlabeled control cells. MCs and APs occurred intracellularly as distinct entities showing a cell-wide distribution but a lowered spatial overlap with natural autofluorescence. Using the immunofluorescence technique, colocalization with markers of individual organelles was utilized to relate the distribution of labeled Mcs to cellular compartments, e.g., using RbcL and FtsZ (cytosol) and PsbA (thylakoids). The colocalization correlation coefficients calculated pairwise between organelles and autofluorescence were highly positive as opposed to the relatively low positive indices derived from labeled MCs. The lower correlation coefficients imply that only a portion of the labeled Mc molecules were related spatially to the organelles in the cell.

It has been generally hypothesized that mobile elements can induce genomic rearrangements and influence the distribution and functionality of toxic/bioactive peptide synthesis pathways in microbes. In this study, we performed in depth genomic analysis by completing the genomes of 13 phylogenetically diverse strains of the bloom-forming freshwater cyanobacteria Planktothrix spp. to investigate the role of insertion sequence (IS) elements in seven pathways. Chromosome size varied from 4.7–4.8 Mbp (phylogenetic Lineage 1 of P. agardhii/P. rubescens thriving in shallow waterbodies) to 5.4–5.6 Mbp (Lineage 2 of P. agardhii/P. rubescens thriving in deeper physically stratified lakes and reservoirs) and 6.3–6.6 Mbp (Lineage 3, P. pseudagardhii/P. tepida including planktic and benthic ecotypes). Although the variation in chromosome size was positively related to the proportion of IS elements (1.1–3.7% on chromosome), quantitatively, IS elements and other paralogs only had a minor share in chromosome size variation. Thus, the major part of genomic variation must have resulted from gene loss processes (ancestor of Lineages 1 and 2) and horizontal gene transfer (HGT). Six of seven peptide synthesis gene clusters were found located on the chromosome and occurred already in the ancestor of P. agardhii/P. rubescens, and became partly lost during evolution of Lineage 1. In general, no increased IS element frequency in the vicinity of peptide synthesis gene clusters was observed. We found a higher proportion of IS elements in ten breaking regions related to chromosomal rearrangements and a tendency for colocalization of toxic/bioactive peptide synthesis gene clusters on the chromosome.