Lack of phylogeographic structure in the freshwater cyanobacterium Microcystis aeruginosa suggests global dispersal

Abstract

Background: Free-living microorganisms have long been assumed to have ubiquitous distributions with little biogeographic signature because they typically exhibit high dispersal potential and large population sizes. However, molecular data provide contrasting results and it is far from clear to what extent dispersal limitation determines geographic structuring of microbial populations. We aimed to determine biogeographical patterns of the bloom-forming freshwater cyanobacterium Microcystis aeruginosa. Being widely distributed on a global scale but patchily on a regional scale, this prokaryote is an ideal model organism to study microbial dispersal and biogeography.

Methodology/principal findings: The phylogeography of M. aeruginosa was studied based on a dataset of 311 rDNA internal transcribed spacer (ITS) sequences sampled from six continents. Richness of ITS sequences was high (239 ITS types were detected). Genetic divergence among ITS types averaged 4% (maximum pairwise divergence was 13%). Preliminary analyses revealed nearly completely unresolved phylogenetic relationships and a lack of genetic structure among all sequences due to extensive homoplasy at multiple hypervariable sites. After correcting for this, still no clear phylogeographic structure was detected, and no pattern of isolation by distance was found on a global scale. Concomitantly, genetic differentiation among continents was marginal, whereas variation within continents was high and was mostly shared with all other continents. Similarly, no genetic structure across climate zones was detected.

Conclusions/significance: The high overall diversity and wide global distribution of common ITS types in combination with the lack of phylogeographic structure suggest that intercontinental dispersal of M. aeruginosa ITS types is not rare, and that this species might have a truly cosmopolitan distribution.

Figure 1. Map indicating the origin of the Microcystis ITS sequences used in this study.

Figure 2. Predicted secondary structure of the spacer region of the rrn operon transcript.

The rrn operon transcript shown in the figure is the 5′ leader sequence upstream from the 16S rRNA, the 16S–23S rRNA ITS and 23S-5S rRNA spacer of Microcystis strain NIES843 (EMBL accession number AP009552, ITS sequence identical to BG08 of the present study). P1 indicates the position of the promotor. Locations of the 16S rRNA, 23S rRNA and 5S rRNA are represented by triangles. The antiterminator box B stems and box A sequences are indicated in the leader and spacer domains, in accordance with Rocap et al. . The conserved motifs (D1–D5, defined by Iteman et al. [73]) are marked. Positions of the hypervariable regions, identified in the present study, are indicated by a black background.

Figure 3. 95% probability parsimony network of Microcystis rDNA ITS sequences based on the full ITS alignment.

Colours indicate region of origin: blue: Europe, yellow: Africa, red: Asia, green: South America, grey: North America, pink: Oceania. A line between ITS types represents one mutational step, open circles represent ITS types not present in the sample.

Figure 4. 95% probability parsimony network of Microcystis rDNA ITS sequences based on the stripped ITS alignment.

Colours indicate region of origin: blue: Europe, yellow: Africa, red: Asia, green: South America, grey: North America, pink: Oceania. A line between ITS types represents one mutational step, open circles represent ITS types not present in the sample. Radius of the circles represent number of sequences, numbers 1–4 indicate subclusters.