The mosaic of habitats of the Seine estuary: Insights from food-web modelling and network analysis

Ecological network analysis was applied in the Seine estuary ecosystem, northern France, integrating ecological data from the years 1996 to 2002. The Ecopath with Ecosim (EwE) approach was used to model the trophic flows in 6 spatial compartments leading to 6 distinct EwE models: the navigation channel and the two channel flanks in the estuary proper, and 3 marine habitats in the eastern Seine Bay. Each model included 12 consumer groups, 2 primary producers, and one detritus group. Ecological network analysis was performed, including a set of indices, keystoneness, and trophic spectrum analysis to describe the contribution of the 6 habitats to the Seine estuary ecosystem functioning. Results showed that the two habitats with a functioning most related to a stressed state were the northern and central navigation channels, where building works and constant maritime traffic are considered major anthropogenic stressors. The strong top-down control highlighted in the other 4 habitats was not present in the central channel, showing instead (i) a change in keystone roles in the ecosystem towards sediment-based, lower trophic levels, and (ii) a higher system omnivory. The southern channel evidenced the highest system activity (total system throughput), the higher trophic specialisation (low system omnivory), and the lowest indication of stress (low cycling and relative redundancy). Marine habitats showed higher fish biomass proportions and higher transfer efficiencies per trophic levels than the estuarine habitats, with a transition area between the two that presented intermediate ecosystem structure. The modelling of separate habitats permitted disclosing each one's response to the different pressures, based on their a priori knowledge. Network indices, although non-monotonously, responded to these differences and seem a promising operational tool to define the ecological status of transitional water ecosystems.

β-N-methylamino-l-alanine (BMAA) and isomers: Distribution in different food web compartments of Thau lagoon, French Mediterranean Sea

The neurotoxin BMAA (β-N-methylamino-l-alanine) and its isomer DAB (2,4-diaminobutyric acid) have been detected in seafood worldwide, including in Thau lagoon (French Mediterranean Sea). A cluster of amyotrophic lateral sclerosis (ALS), a neurodegenerative disease associated with BMAA, has also been observed in this region. Mussels, periphyton (i.e. biofilms attached to mussels) and plankton were sampled between July 2013 and October 2014, and analyzed using HILIC-MS/MS. BMAA, DAB and AEG (N-(2-aminoethyl)glycine) were found in almost all the samples of the lagoon. BMAA and DAB were present at 0.58 and 0.83, 2.6 and 3.3, 4.0 and 7.2 μg g−1 dry weight in plankton collected with nets, periphyton and mussels, respectively. Synechococcus sp., Ostreococcus tauri, Alexandrium catenella and eight species of diatoms were cultured and screened for BMAA and analogs. While Synechococcus sp., O. tauri and A. catenella did not produce BMAA under our culture conditions, four diatoms species contained both BMAA and DAB. Hence, diatoms may be a source of BMAA for mussels. Unlike other toxins produced by microalgae, BMAA and DAB were detected in significant amounts in tissues other than digestive glands in mussels.