Metabolic fingerprinting applied in diatom taxonomy

Abstract

The main aim of this thesis was to investigate if and how metabolic fingerprinting can be applied in diatom taxonomy. Even though both morphology and gene sequences have been shown to be appropriate tools in diatom taxonomy there are cases where these give contradicting results, like in the case of cryptic species. Cryptic species have similar morphology but are genetically different. Another issue with these two tools is that they do not offer much information about the function of a species, information that is interesting in light of for example ecology and management. Metabolomics investigates the metabolites synthesized by an organism. The metabolites synthesized at a certain moment in time will be a reflection of what genes are expressed at that time and will be a product of the organisms response to the environmental and biological conditions prevailing. Direct injection mass spectrometry was used to investigate the metabolic fingerprints of different, commonly occurring, northern diatom species. The method produces mass-to-charge ratios (markers) normally with a mass precision of four decimals. Reproducibility of the method was 80% with the direct injection method applied, using a decimal precision of 0.1. The results of the analysis showed that the different species shared between 26-67% of the total markers. Even species of the same genera showed a high diversity. The two species Chaetoceros furcellatus and Chaetoceros socialis only shared 30% of the total markers. For four out of six species the difference between species increased with decreasing temperature. The expected phylogeny of these six species could not be reflected by the metabolite data. Cryptic diversity has been documented within the so-called cosmopolitan species Chaetoceros socialis. We investigated this diversity between strains collected from two geographic areas; the northeast Atlantic and Arctic and from Mediterranean waters. Monoclonal cultures were cultivated at three different temperatures, and analyzed with the aid of morphology, LSU rRNA sequencing, growth rates, photosynthetic maximum quantum yield and metabolic fingerprinting. Comparison of gene sequences of the two groups showed an unequivocal difference, while only small morphological differences in spore morphology (not in the morphology of the vegetative cells) could be found between the two groups. At all three temperatures there were clear differences in growth and maximum quantum yield. The results from the metabolic fingerprinting also supported these findings. The clear genetic as well as functional differences does not support the cosmopolitan distribution of C. socialis and we therefore conclude that this species should be revised. The results, both from the comparison of metabolic fingerprinting between diatom species as well as within a pseudo cryptic diatom species, in my opinion, is in support of the use of metabolomics in diatom taxonomy. Our results underline the need of metadata, e.g. growth rates, in metabolomics studies. I also think that increased knowledge of functional traits of species, like metabolomics, could be implemented in ecological modeling, building a bridge between taxonomy and ecology. The results of this thesis are also relevant to bioprospecting. The higher chemical diversity between species found at the lower temperatures, would indicate that it could be beneficial to cultivate diatoms at low temperatures, close to zero, in search for bioactivities.