Analysis of phytoplankton assemblage structure in the Mediterranean Sea based on high-throughput sequencing of partial 18S rRNA sequences

Abstract

Studying taxonomic and ecological diversity of phytoplankton assemblages is often difficult because morphological analysis cannot provide a complete description of their composition. Therefore, more robust and feasible approaches have to be chosen to elucidate the interactions between environmental and human pressures and phytoplankton assemblages. The Ocean Sampling Day (OSD) allowed collecting seawater samples from a wide range of oceanic regions including the Mediterranean Sea. In this study, a total of 754,167 V4-18S ribosomal DNA (rDNA) metabarcodes derived from 20 plankton samples collected at 19 sampling sites across the coastal areas of the Mediterranean Sea were analyzed to explore the relationships between phytoplankton assemblages' composition, sub-regional environmental features and human pressures. We reduced the whole set of autotroph plankton (1398 OTUs) to a smaller number of ecologically relevant entities (205 taxa) and used the latter for analysing the structure of phytoplankton assemblages. Chaetoceros was the only genus occurring in all the samples, while the number of taxa was maximum in the W Mediterranean. Based on the assigned OTUs, the structure of E Mediterranean phytoplankton was the most homogeneous. Further, phytoplankton assemblages from the three Mediterranean sub-regions (Western, Adriatic and Eastern) were significantly different (R = 0.25, p = 0.0136) based on Jaccard similarity. We also observed that phytoplankton diversity and human impact on marine ecosystems were not significantly related to each other based on Mantel's test.

Introduction

Protists are a paraphyletic group including a wide range of microorganisms among which phytoplankton assemblages act as a primary producer. Large-scale analysis of protistans diversity is difficult owing to the huge underlying heterogeneity, which reflects their extremely broad distribution and involvement in multiple ecological and functional processes, such as global biological and geochemical processes. The taxonomic and ecological diversity study of unicellular eukaryotic primary producers is often difficult, because it is extremely time-consuming and expert based when deep morphological classification is needed. Moreover, morphological analyses alone cannot provide a complete description of the huge phytoplankton diversity (Penna et al., 2013, de Vargas et al., 2015, Malviya et al., 2016). The current high-throughput sequencing of ribosomal DNA (rDNA) genes coupled with metabarcoding allowed discovering a huge diversity at large scale in various marine ecosystems. Data sets obtained from this approach are more complete, quickly available and especially it is not based on taxonomic skills. In marine unicellular eukaryotes the hypervariable regions of 18S rDNA has been long proposed as barcode (Amaral-Zettler et al., 2009, de Vargas et al., 2015).

In the Mediterranean Sea, subregional seas and coastal areas are different due to their hydrographical and climatological regimes. Those differences imply that trophic conditions and productivity are also different, as well as the structure and dynamics of autotrophic plankton assemblages. In an oligotrophic system, such as the Mediterranean Sea, coastal productivity largely depends on inputs from rivers and areas of high productivity are mainly restricted to waters close to major freshwater inputs of rivers (D'Ortenzio and Ribera d'Alcalà, 2009, Ludwig et al., 2009). Eutrophication, i.e. anthropogenic nutrient inputs, can cause considerable shifts in coastal ecosystem structure and function (Smith and Schindler, 2009), with effects spreading along the pelagic food web (Rissik et al., 2009). Moreover, costal ecosystems impacted by anthropogenic nutrient inputs can be influenced by hydrographic variability and local conditions, which can trigger site-specific effects and patterns. While environmental drivers can select species and functions of phytoplankton assemblages (Boyd et al., 2010, Edwards et al., 2013), human activities and global climate changes are now potent new drivers that significantly affect the functioning of coastal and offshore marine ecosystems (Hallegraeff, 2010, Huertas et al., 2011, Sunday et al., 2014). The Mediterranean Sea is a critical large marine ecosystem for many reasons: it is a hot-spot of biodiversity, it is dramatically impacted by resource exploitation, maritime traffic, and, last but not least by coastal urbanization with high density population pressure (Coll et al., 2010, Micheli et al., 2013, Newton et al., 2014). These major disturbances, in addition to allochthonous species introduction, severely affect the natural balance of the Mediterranean ecosystems and have already resulted in an extensive loss of biodiversity (Occhipinti-Ambrogi, 2007, Lejeusne et al., 2010, Glibert et al., 2014). Other effects include diseases or mass mortality events that affect marine organisms, in some cases with consequent disruption of the benthic-pelagic coupling. Cascading effects induce changes in fluxes of matter and energy in trophic webs, in some case affecting also their first level through top-down effects (Scheffer et al., 2005, Winder and Sommer, 2012, Vergés et al., 2014). Consequently, the study of phytoplankton diversity composition and its fluctuations over time and space is of paramount importance in order to assess the consequences of anthropogenic perturbations to marine ecosystems.

To elucidate the relationships between composition of phytoplankton assemblages, subregional environmental features and human impact in the Mediterranean Sea, we used the metabarcode 18S rRNA amplicons in combination with some data analysis techniques commonly used in ecological research. We analyzed 18S ribosomal DNA sequences across the smallest phototroph unicellular eukaryotes (about 0.8 μm) to larger ones (about 180 μm) from coastal seawater samples collected at the summer solstice 2014 during the OSD Sampling Day (OSD) survey in the Mediterranean Sea under the EU funded Marine Microbial Biodiversity, Bioinformatics, Biotechnology Project (Micro-B3) (Kopf et al., 2015). In fact, the OSD sampling sites are located in coastal regions within exclusive economic zones (EEZ). Therefore, the OSD data set provided the opportunity to better understand the potential of a high-throughput approach in the analysis of phytoplankton assemblages in a large marine ecosystem that is comprised of very diverse subregions and exposed to relevant anthropic pressures, especially along its northern shore.