Molecular (PCR-DGGE) versus morphological approach: analysis of taxonomic composition of potentially toxic cyanobacteria in freshwater lakes
Springer Science and Business Media LLC -- Aquatic Biosystems
DOI 10.1186/2046-9063-10-2
Keyword(s)
  1. Cyanobacteria
  2. DGGE
  3. ITS
  4. mcy genes
  5. Microcystins
  6. Microscopic analysis
Abstract(s)

Background

The microscopic Utermöhl method is commonly used for the recognition of the presence and taxonomic composition of potentially toxic cyanobacteria and is especially useful for monitoring reservoirs used as drinking water, recreation and fishery resources. However, this method is time-consuming and does not allow potentially toxic and nontoxic cyanobacterial strains to be distinguished. We have developed a method based on denaturing gradient gel electrophoresis (DGGE) of the marker gene ITS and the mcy-gene cluster, and DNA sequencing. We have attempted to calibrate the DGGE-method with a microscopic procedure, using water samples taken in 2011 from four lakes of the Great Mazurian Lakes system.

Results

Results showed that the classic microscopic method was much more precise and allowed the classification of the majority of cyanobacterial taxa to the species or genus. Using the molecular approach, most of the sequences could only be assigned to a genus or family. The results of DGGE and microscopic analyses overlapped in the detection of the filamentous cyanobacteria. For coccoid cyanobacteria, we only found two taxa using the molecular method, which represented 17% of the total taxa identified using microscopic observations. The DGGE method allowed the identification of two genera of cyanobacteria (Planktothrix and Microcystis) in the studied samples, which have the potential ability to produce toxins from the microcystins group.

Conclusions

The results confirmed that the molecular approach is useful for the rapid detection and taxonomic distinction of potentially toxic cyanobacteria in lake-water samples, also in very diverse cyanobacterial communities. Such rapid detection is unattainable by other methods. However, with still limited nucleotide sequences deposited in the public databases, this method is currently not sufficient to evaluate the entire taxonomic composition of cyanobacteria in lakes.