Assessing genetic diversity and identification of Microcystis aeruginosa strains through AFLP and peRRFLP analyses.

English: There are 150 cyanobacterial genera and approximately 2 000 species known in the world. More than 40 of these have toxin producing strains. Cyanobacteria, commonly known as blue-green algae, are often present in small numbers together with a diverse assemblage of other photosynthetic algae that naturally occur in surface water worldwide. However, under conditions of warm temperatures, minimal water movement and elevated concentrations of phosphorus in a water body, cyanobacteria may frequently become dominant and form thick scums of floating algal cells. These dense aggregations of floating cells, termed 'blooms', presents a number of water quality problems; most often offensive odours and tastes, and sometimes biotoxins that can be divided into alkaloid neurotoxins and cyclic peptide hepatotoxins, commonly from the genus Microcystis and released in waterbodies. The neurotoxins act chiefly at neuromuscular junctions and cause rapid death because of respiratory paralysis. The hepatotoxins act on the hepatocyte cytoskeleton and cause intrahepatic haemorrhage and centrilobular necrosis. Clinically the hepatotoxin most often causes peracute or acute death, or subacute poisoning with signs such as icterus and hepatogenous photosensitivity. Currently cyanobacterial taxonomy does not provide an unequivocal system for the identification of toxigenic and bloom-forming genus Microcystis. The ambiguities that exist in the cyanobacterial taxonomy are due to the expressed variability, minor morphological and developmental characteristics that are used for identification. In this study geographically unrelated axenic strains of Microcystis aeruginosa were obtained from the Pasteur Institute, France (PCC); the National Institute for Environmental Studies, Japan (NIES); the Institute of Freshwater Ecology, UK (CCAP); the Pflanzen Physiologisches Institut, Universitat Gottingen, Germany (SAG) and the University of the Free State, South Africa (UV) culture collections. Nonaxenic strains were collected from Hartbeespoort, Rietvlei and Roodeplaat Dams in South Africa. After screening 20 primer combinations on a subset of strains eight IRDye700™-labeled EcoR1 primer pairs were selected for amplified fragment length polymorphism (AFLP) analysis to determine the genetic relationship of these geographically unrelated strains. A total of 909 bands were amplified from the eight primer combinations, of which 665 were informative, 207 non-informative and 37 monomorphic, with an average of 83.12 polymorphic bands per primer combination. The genetic relationship among all the Microcystis aeruginosa strains based on the combination of data obtained with the eight primer combinations was analysed employing the Unweighted Pair Group Method using Arithmetic Means (UPGMA) algorithm and presented as a dendrogram. In the dendrogram, the strains from Rietvlei (UP01) and Hartbeespoort Dams (UP04) grouped together and were thus genetically closer to each other, than to the strain from the Rhoodeplaat Dam (UP03). The Japanese strains (NIES88, NIES89, NIES90, NIES99, NIES299) also grouped separate from the other strains, with NIES90 and NIES299, genetically closest to each other. Interestingly, Microcystis aeruginosa strain PC7806 that originated from The Netherlands, also grouped within this group. Microcystis aeruginosa strains CCAP1450/1 (UK), UV027 (South Africa) and PC7813 grouped together, and are genetically closer to the UP-strains, than any of the other strains. In the present study, AFLP analysis proved useful for the identification of genetic diversity and analysis of population structure within Microcystis aeruginosa. In order to link the identification of strains with toxicity, the utility of the mcyB gene sequence for identification of strains was tested. Based on conserved motifs present in known sequences of mcyB four primer pairs were designed. Using the primer pairs Tax 3P/2M, Tax 1P/1M, Tax 7P/3M and Tax 10P/4M, the mcyB gene from PCC7813 and UV027 were sequenced, resulting in fragments of 2174 and 2170 base pairs in size, respectively. The obtained sequences were analyzed using nucleotide BLASTN annotation of the Basic Local Alignment Search Tool (BLAST). The sequence alignment indicated high homology to other published sequences in GenBank (AY034601 for pee7813 and AY034602 for UV027; e-value = 0.0). Upon further analysis of the sequences it was obvious that there are several base differences between the sequences of the two strains, which led us to investigate the potential of using differences in restriction sites, and thus insertions/deletions (indels) in nucleotide sequence to discriminate between the other M. aeruginosa strains, as well as using the mcyB gene to discern between M. aeruginosa and M. wesenbergii in raw water samples. A vast number of restriction sites were identified with differences followed by restriction digest of the specific polymerase chain reaction (PCR) mcyB gene fragment. This work demonstrates that PCR assays provide a useful indicator of toxicity as well as the identification of taxonomical characteristics between laboratory cultures and environmental isolates. A number of questions arise from the present study and future research therefore needs to address the following issues: • Are there more than one Microeystis aeruginosa strain / "population" present at a given time in a specific water reservoir? Do these populations change through the season? What role does the individual populations play in a cyanobacterial bloom? Thus, the dynamics and structure of populations need to be clarified. • Which mcy gene in the cluster is mostly responsible for toxin production? Does the expression of the genes correlate with gene structure/sequence? What role does the environment play in determining the level of expression, and thus toxin production?