Reconstructing marine plankton food web interactions using DNA metabarcoding

Abstract

Knowledge of zooplankton in situ diet is critical for accurate assessment of marine ecosystem function and structure, but due to methodological constraints, there is still a limited understanding of ecological networks in marine ecosystems. Here, we used DNA‐metabarcoding to study trophic interactions, with the aim to unveil the natural diet of zooplankton species under temporal variation of food resources. Several target consumers, including copepods and cladocerans, were investigated by sequencing 16S rRNA and 18S rRNA genes to identify prokaryote and eukaryote potential prey present in their guts. During the spring phytoplankton bloom, we found a dominance of diatom and dinoflagellate trophic links to copepods. During the summer period, zooplankton including cladocerans showed a more diverse diet dominated by cyanobacteria and heterotrophic prey. Our study suggests that copepods present trophic plasticity, changing their natural diet over seasons, and adapting their feeding strategies to the available prey spectrum, with some species being more selective. We did not find a large overlap of prey consumed by copepods and cladocerans, based on prey diversity found in their guts, suggesting that they occupy different roles in the trophic web. This study represents the first molecular approach to investigate several zooplankton–prey associations under seasonal variation, and highlights how, unlike other techniques, the diversity coverage is high when using DNA, allowing the possibility to detect a wide range of trophic interactions in plankton communities.

DNA metabarcoding reveals trophic niche diversity of micro and mesozooplankton species

“…The association of Pseudocalanus with unidentified taxa might indicate that the components of its natural diet are not well represented in the taxonomic databases. The differences in trophic niches of copepods are supported by a more intense seasonal sample analysis of crustacean zooplankton described in the previous study [ 29 ] and more extensive sampling several locations in the Baltic Sea proper (Baptiste Serandour 2021, pers. comm).…”

“…With the high sensitivity of the polymerase chain reaction, metabarcoding requires very little biological material and is a non -a priori method with high taxonomic resolution. Metabarcoding allows for a food web-oriented approach as several zooplankton species can be investigated simultaneously [ 29 ], thereby providing detailed insights on trophic interactions and better linking the trophic niche diversity with energy flow.…”

A Portable Cruising Speed Net: Expanding Global Collection of Sea Surface Plankton Data

“…However, metabarcoding also presents some well acknowledged pitfalls, including possible biases due to primer or marker choice and the incompleteness of sequence reference databases (Harvey et al, 2017). The present study targeted the universal V4 region of 18S rRNA gene followed by sequence assignment using the PR2 database, enabling analysis of a broad range of taxonomic assemblages and including zoo-and phytoplanktonic taxa (De Vargas et al, 2015;Zamora-Terol et al, 2020). However, our data showed zooplankton (especially copepods) to be highly abundant compared with phytoplankton (e.g., Dinophyceae and Spirotrichea, see Supplementary Table 1) which were relatively scarce and mostly restricted to coastal sampling stations.…”

“…However, our data showed zooplankton (especially copepods) to be highly abundant compared with phytoplankton (e.g., Dinophyceae and Spirotrichea, see Supplementary Table 1) which were relatively scarce and mostly restricted to coastal sampling stations. Future studies aiming to capture a wider representation of phytoplankton taxa should consider using serial filtration methods before DNA extraction (Not et al, 2009) and/or include copepods-blocking oligos during PCR (Zamora-Terol et al, 2020). Additionally, simple DNA isolation methods should be considered while sampling plankton from sailing vessels with limited freezer (−20 • C) capacity.…”

To denoise or to cluster? That is not the question. Optimizing pipelines for COI metabarcoding and metaphylogeography

“…The idea that denoising should be used instead of clustering has been followed by some (e.g., Tapolczai et al 2019, Steyaert et al 2020, Zamora-Terol et al 2020, Pearman et al 2020, while other authors have combined the two approaches (e.g., Brandt et al 2020, Nguyen et al 2020, Laroche et al 2020. Indeed, denoising has the advantages of reducing the dataset and to ease pooling or comparing studies, which is necessary in long term biomonitoring applications.…”