Uncovering the intricacies of microbial community dynamics at Helgoland Roads at the end of a spring bloom using automated sampling and 18S meta-barcoding.

• Main Authors: Katja Metfies, Johanna Hessel, Robin Klenk, Wilhelm Petersen, Karen Helen Wiltshire, Alexandra Kraberg
• Format: Article
• Language: English
• Published: Public Library of Science (PLoS) 2020-01-01
• Series: PLoS ONE
• Online Access: https://doi.org/10.1371/journal.pone.0233921

In May 2016, the remote-controlled Automated Filtration System for Marine Microbes (AUTOFIM) was implemented in parallel to the Long Term Ecological Research (LTER) observatory Helgoland Roads in the German Bight. We collected samples for characterization of dynamics within the eukaryotic microbial communities at the end of a phytoplankton bloom via 18S meta-barcoding. Understanding consequences of environmental change for key marine ecosystem processes, such as phytoplankton bloom dynamics requires information on biodiversity and species occurrences with adequate temporal and taxonomic resolution via time series observations. Sampling automation and molecular high throughput methods can serve these needs by improving the resolution of current conventional marine time series observations. A technical evaluation based on an investigation of eukaryotic microbes using the partial 18S rRNA gene suggests that automated filtration with the AUTOFIM device and preservation of the plankton samples leads to highly similar 18S community profiles, compared to manual filtration and snap freezing. The molecular data were correlated with conventional microscopic counts. Overall, we observed substantial change in the eukaryotic microbial community structure during the observation period. A simultaneous decline of diatom and ciliate sequences succeeded a peak of Miracula helgolandica, suggesting a potential impact of these oomycete parasites on diatom bloom dynamics and phenology in the North Sea. As oomycetes are not routinely counted at Helgoland Roads LTER, our findings illustrate the benefits of combining automated filtration with metabarcodingto augment classical time series observations, particularly for taxa currently neglected due to methodological constraints.