The Thau lagoon, characterised by a high abundance of small eukaryotes and by recent in situ changes in phytoplankton structure due to water temperature increase , is an interesting ecosystem to investigate the responses of small eukaryotes to climatic and anthropogenic regulatory factors. Our experimentation does not intend to predict the impact of long-term global change on the structure of small planktonic eukaryotes. Indeed, only a combination of approaches including laboratory studies on model microbes, microcosm and mesocosm experiments, and in situ comparative studies would help to provide realistic predictions of the effects of environmental changes [23, 54]. Our goal was to reveal the potential rapid responses of small eukaryote assemblage (using molecular and morphological methods) during the productive spring season when plankton may be particularly vulnerable to elevated temperature and UVBR .
Molecular analyses revealed the presence of various phylogenetic groups within the “black box” of small eukaryotes, especially non-pigmented eukaryotes (poorly discriminated by microscopy). Some limitations in the PCR-based methods are recognized, for instance, the over-representation of Alveolata (particularly Dinoflagellates and Ciliates) in 18S rRNA gene clone libraries due their high SSU rRNA gene copy number [50–52]. However, the molecular methods used here enabled the description of the diversity within dominant eukaryotic populations, and allowed examination of the effects of regulatory factors by considering both the dynamics of OTUs (using the sequencing and fingerprinting datasets) and the comparison of phylogenetic composition obtained for all treatments (using the sequencing data). The impact of temperature nutrients and UVBR explained 18.8%, 11.0% and 8.4% of the variance of the small eukaryotes structure respectively. While Bouvy et al. (2011) could not detect any significant responses of pico- or nano-eukaryotic plankton in the same experimental conditions, we demonstrated here, at a different taxonomic resolution, that small eukaryotes community structure was actually affected by this multi-factorial pressure.
The simultaneous use of molecular and morphological methods was therefore essential to provide evidence of rapid shifts that occur at various taxonomic levels (abundance of large groups or community composition at OTU level) under the influence of temperature, UVBR and nutrient treatments.
Among the 3 regulatory factors tested, both sequencing and CE-SSCP demonstrated that increased temperature had the greatest influence on the small eukaryote community structure and composition. The single effect of temperature (without any significant interaction with UVBR and nutrients) on total pigmented eukaryote abundance was observed by microscopy. Considering the different phylogenetic groups within pigmented eukaryotes, complex interaction effects were also suggested. For instance, our results showed that under multi-factorial environmental changes, the general impact on the molecular diversity and abundance of pigmented Dinophyceae resulted from complex interactive (non-additive) effects. Multi-factorial interactions were also apparent for Cryptophyceae which experienced antagonistic effects of nutrient addition (significantly negative impact) and temperature (positive impact on relative abundance).
In addition to the manipulated factors (temperature, UVBR and nutrients), some biotic interactions such as predation, viral lysis and competition, are involved in the responses observed in this experiment. For example, the general reduction of Mamiellophyceae (Micromonas and Ostreococcus) in all treatments might be linked to (i) manipulation effects since these fragile cells might have been affected by filtration steps, (ii) limitation by inorganic nutrients under the rather low orthophosphate concentrations at T96h (from 0.05 to 0.08 μM of PO4), (iii) the grazing impact of heterotrophic flagellates: these microorganisms are known to play a significant role in the regulation of Ostreococcus populations in the Thau lagoon  and were shown to exert a strong control of bacterioplankton during the study period . We could not detect a link between the dynamics of Micromonas/Ostreococcus and viruses. Since biological descriptors can explain some of the variance of small eukaryote phylogenetic structure, the observations made here regarding the effect of temperature, UVBR, and nutrients have to be considered in view of possible biological effects. Predation by zooplankton and competition with larger phytoplanktonic species were not considered in our size fractionated approach and should be taken into account, especially if long-term extrapolation of in situ responses of small eukaryotes is considered.
Our data provide further illustration of the need to consider the taxonomic and functional diversity of heterotrophic flagellates. The lack of discrimination between heterotrophic bacterivores and parasitic/saprotrophic zoospores within the non-pigmented flagellates can lead to misinterpretation of the functioning and responses of planktonic food webs. Indeed, while microscope observations did not allow us to detect changes in the abundance and structure of non-pigmented eukaryotes, a structuring impact of manipulated factors (especially temperature) was observed through sequencing results on taxa affiliated to parasitic and saprotroph groups (particularly Syndiniales and Hyphochytrids). The existence of eukaryotic parasites among small-size plankton was recently re-discovered by molecular environmental surveys, and the ecological significance of these groups has been highlighted by several authors [57, 58]. The ‘Fungi-like’ Hyphochytrids possess many morphological and ecological similarities to chytrids [58, 59], and their role as saprotrophs and/or parasites is unclear [60, 61], whereas the Amoebophrya are well recognized as a widely distributed parasitic order within the Dinophyceae . Amoebophrya and Hyphochytrids emerged in clone libraries at T96 h and were presumably present among the rare species at T0. The taxa found to be phylogenetically close to Amoebophrya particularly emerged in treatments with increased temperature (Figure 5), along with their hosts (pigmented Dinoflagellates). This observation supports Guillou et al.’s  suggestion that warming could promote rapid infection cycles of Amoebophrya. However, broad extrapolation would need to take into account various aspects of the host-parasite relationships, such as the mechanisms underlying the parasitic specificity. In contrast to the Amoebophrya, hyphochytrids were associated with all treatments except those with increased temperature (Figure 5). From our results, we hypothesized that not only parasite communities, but also saprotroph communities would be shaped by temperature and UVBR conditions, as already described in other ecosystems . The responses of saprotrophs to these drivers may result from direct and/or indirect effects as demonstrated in soils ; further research is probably needed on the saprotrophs in aquatic systems since changes in their assemblages may influence organic matter decomposition and nutrient cycling.