Mobilomics of toxin production in cyanobacteria

Harmful algal blooms (HABs) formed by freshwater cyanobacteria have been repeatedly associated with the toxification of humans and livestock. Up to date only one genome from toxin-producing cyanobacteria has been completely elucidated. Efforts to form one contiguous sequence representing the whole genome have been hampered by the high amount of repetitive elements such as mobile transposable elements (TEs). Previous research has revealed that the genetic basis of secondary metabolite (SM) production is frequently influenced by TEs, however, only the quantitative comparison of complete genomes can reveal the influence of TEs on the interrelation between SM production and ongoing speciation processes. We have totally sequenced the genome of the bloom-forming and toxic cyanobacterium Planktothrix agardhii, which will serve as the reference genome. In this project we propose the sequencing of eight additional clonal isolates (strains) of Planktothrix spp. representing eight different phylogenetic lineages comprising 140 strains which have been characterized genetically by genotyping (using Multi Locus Sequence Typing) as well as for SM production by chemical analytical techniques. The strains have been collected from 40 waterbodies located in 17 countries on three different continents (Europe, North America, Africa) and evolved both in the presence and in the absence of toxins. The project will find out (1) the TE induced variation and innovation in SM production in specific Planktothrix lineages; (2) SM synthesis genes which have not been affected by TE inactivation and which are therefore of potential selective importance; (3) the activity and specificity of frequent TEs through in vitro experiments; (4) the relationship between the frequency of active TEs and the SM synthesis recombination potential in populations in the wild. While the total genome assembly will reveal the full potential of present TEs, the comparative analysis of the 140 strains will show, which TEs and which TE induced mutations are specific to individual lineages. The activity and the specificity of TEs will be probed in vivo and in vitro. Since the TE sequences are known, the TE insertion sites can be elucidated by sequencing the flanking regions directly from field populations. This cultivation independent approach would elucidate TE induced mutations in toxin production at an early stage. This project is the first to study speciation within a well defined HAB genus distributed over a wide geographical scale in relation to the occurrence of TEs in cyanobacteria. Since this project will reveal the entire set of SM gene clusters of different lineages it will go far beyond monocausal gene vs. environment studies and will further elucidate the relationship between toxigenicity and other ecophysiological characters. Thus it will be elucidated which toxin-producing ecotypes are most likely to profit from the predicted scenario of a global increase in HAB occurrence.

Cyanobacterial harmful algal blooms are increasing in frequency on a global scale. The secondary metabolites produced by widely distributed, bloom-forming fresh-water cyanobacteria such as Planktothrix can make freshwater dangerous for drinking and recreation. The regulation of toxin and bioactive peptide production has become a significant research task. Isolates (strains) contain an impressive metabolic diversity of toxic/bioactive peptides resulting in a variable spatial/temporal distribution of (non)toxic genotypes. Despite their importance to human wellbeing, such genotype diversity so far has not been mapped at scales relevant to nature. In this project, we genome- (re)sequenced ten strains of the genus Planktothrix, representing three distantly related phylogenetic lineages. These high-quality draft genomes provide insights into location and physical proximity of individual peptide synthesis gene clusters on the chromosome. Furthermore, a large number of isolates (more than hundred strains), originating from a variety of shallow/deep lake habitats of temperate and tropical climatic zone from three continents (Europe, North America, Africa), were quantitatively analyzed for microcystin/peptide synthesis gene clusters as well as the distribution and activity of transposable elements (TEs). Most importantly, these empirical data provide evidence for the long- standing question of how the distribution and inheritance of genes encoding microcystin/bioactive peptide synthesis correlate with speciation processes and ecological diversification. In particular, the findings show that microcystin/bioactive peptide production are primarily related to speciation processes, while within a phylogenetic lineage the probability that strains differ in peptide composition increases with geographic distance. Other results indicate the role of horizontal gene transfer (HGT) and chromosomal inversion that finally led to the co-localization and the formation of a large meta peptide synthesis gene cluster, and selective advantage in the natural habitat resulted in its phylogenetic fixation. The high frequency of several microcystin/bioactive peptide synthesis genes observed among bloom-forming P. agardhii/rubescens but not among other morphologically similar Planktothrix species suggests a potential functional linkage between bioactive peptide production and the colonization potential and possible dominance in habitats. Some TEs showed a pronounced dependence in phylogenetic distribution, i.e. occurred among strains of phylogenetic lineage 2 (isolates from deep lakes of temperate climate zone). In a second step, the activity of specific abundant TE was observed in real-time in a large number of individually selected filaments from five lakes in the Alps. The results show that microcystin synthesis genes are frequently and nonrandomly mutated either by transposable elements or partial gene deletions which could be linked to repetitive nucleotide sequence motifs (REPs) predicted to form stem-loop secondary structures. Our project also pinpoints spatial-temporal variation in abundance of these mutations which might be caused by variable activity of the transposable element. In summary, TE insertion events occurring within cyanotoxin genes are predictable due to their correlation with REPs, which can induce the rather common mutation events as compared with other toxic cyanobacteria.