A2: Analysis of dynamics and interactions within planktonic microbial communities in Lake Constance by next-generation sequencing

David Schleheck & Peter G. Kroth

State of the art

Contemporary next-generation sequencing (NGS) methods are increasingly being used to study detailed successional patterns in composition and diversity of bacterioplankton communities, while the classical microscopical methods often ignore the huge taxonomical and functional diversity of the bacterioplankton. In contrast, successional patterns of, e.g., phytoplankton and ciliates are still being estimated mostly by microscopic counting (but see Eiler et al. 2013; Xiao et al. 2014), which is strongly dependent on the taxonomic expertise of the observers. NGS instead allows for a demonstration of algal succession patterns in a much more robust way, enabling a comparison of these patterns with those established by microscopical counting. Furthermore, a much more fine-scaled correlation analyses is feasible regarding positive and negative interactions between phytoplankter, other protists and bacteria, reflecting competition, grazing, and other interactions. Finally, in conjunction with labelling methods, NGS can be utilised for studying predator-prey interactions within the plankton community directly (Moreno et al. 2010; Fay et al. 2013). For example, labelling bacteria in a mesotrophic lake with bromodeoxyuridine (BrdU) (Fay et al. 2013) suggested a much larger diversity of bacterivores than initially realised. Hence, such NGS approaches with BrdU or 13C-labelling might allow us to identify predators and preys within the plankton in Lake Constance, and to address questions such as: What is the relative importance of heterotrophic versus mixotrophic bacterial consumers? Which of the potentially hetero/mixotrophic consumers contribute substantially to bacterial mortality? Which bacterial taxa represent the major link in carbon and phosphorous flow within the classical food chain and the microbial loop?

Preparatory work

Two GOS-metagenome datasets (454 technology) were generated by the Craig Venter Institute from samples taken at two different depths, and comprising subsets of data derived of three different size classes (see Rusch et al. 2007). This preliminary exploration of Lake Constance plankton by NGS for a single time point, has revealed a typical oligotrophic freshwater bacterioplankton community (Kesberg and Schleheck, unpublished). Proteobacteria, Actinobacteria, Bacteroidetes, Cyanobacteria and Verrucomicrobia were the most abundant phyla; expected high abundance of typical ‘freshwater bacterial tribes’ (Newton et al. 2011) and of the LD12 freshwater sister clade of SAR11 (Salcher et al. 2011) was confirmed. Further, we established a size-class filtration and DNA extraction protocol, which recovers high-molecular weight DNA of the plankton-size classes 30µm - 5µm, 5µm - 1µm, and 1µm - 0.1µm (Kesberg and Schleheck 2013); a first set of samples collected from winter to summer (16 time points) has now been sequenced by 16S rRNA-gene amplicon sequencing (Illumina). Further, in previous work PGK contributed to the analyses of the diversity of a protist seas ice community from the Central Arctic Ocean.

Proposed project and role within the RTG

In this project, a doctoral researcher will perform both 16S and 18S rRNA-gene amplicon sequencing to study successional patterns (identity and abundance of bacteria, algae and other protists) within the plankton community in Lake Constance for the first time. We will focus on the recurrence and resilience of taxa within the annual plankton succession at the routine sampling site at Wallhausen in Upper Lake Constance, but we plan to sample also the seasonal dynamics in Lower Lake Constance; the two basins differ in their morphology and physicochemical (nutritional) conditions, and we expect that the plankton succession in both basins differ due to differences in abiotic and biotic conditions. We will target the 16S rRNA gene (bacteria, plastids of most eukaryotic organisms), the 18S rRNA gene (dynamics of the eukaryotic/protist contingent at a higher taxonomic resolution), and more specific markers (to identify groups of protists) (Eiler et al. 2013). The 16S and 18S data, and the data on physicochemical and nutritional parameters from Upper and Lower Lake Constance, will be analysed for correlations. For example, we will try to trace synergistic and antagonistic relationships of pro- and eukaryotic taxa, which may indicate positive or negative interactions between these organisms.

The doctoral researcher will also share his/her genomic tools with the doctoral researchers of projects B3 and C4, allowing studies on e.g. the role of metazoan grazers as direct and indirect (via trophic cascades) regulators of algae and bacteria. In this context (projects A2, C4 and B3), we plan to feed 13C/BrdU-labelled bacteria to eukaryotic plankton in order to define important members the bacterivorous contingent. The data generated in this project will also serve as important reference for projects A1, B4, C1, C2, and for the modelling approaches A4 and B1.