“Unravelling the mystery” of Marine eDNA: A Comparative Metabarcoding Analysis and Protocols Validation

Monitoring environmental changes in aquatic habitats – such as biodiversity loss, shifts in distribution and migratory patterns, spread of invasive species – has become an urgent task in the context of climate change. In this framework, the analysis of environmental DNA (eDNA) – as the DNA released by organisms in the environment and extracted from water and sediments - appears as an innovative and cost-effective tool for gathering useful data for monitoring biodiversity profiles, searching for target species, and emerging pathogens. Although this technique has been extensively used for several purposes and in different environments – for instance, in oceans, rivers, and lakes – its application remains limited by the lack of standardized methods and a fragmentary understanding of the variables influencing detection and data interpretation. This multi-phase project investigates how sampling design, laboratory protocols, and environmental conditions affect eDNA metabarcoding in marine ecosystems. In the first phase, protocols of DNA extraction, sequencing, and bioinformatic analysis were tested in controlled settings using fish mock communities. After adjusting the protocols under laboratory conditions, they were implemented in a semi-controlled environment (Atlantic salmon fish farm) to assess eDNA transport and the influence of water temperature, currents, and fish biological activity on the detectability rate. Data were compared with results from a species-specific qPCR assay to explore the limits of sensitivity for the metabarcoding approach. Finally, the entire eDNA workflow is applied to natural marine environments for monitoring a local population of bottlenose dolphins (Tursiops truncatus) and their feeding preferences, providing also a local snapshot of biodiversity. After delving into eDNA dynamics and exploring factors influencing its detectability, the eDNA speciesspecific approach is assessed for the detection of the invasive Chinese mitten crab (Eriocheir sinensis). Overall, this thesis project expands the knowledge on the applicability of the eDNA workflow in aquatic ecosystems. Sampling methods (active vs. passive), primer set choice, and hydrodynamic conditions emerged as major drivers in eDNA detection. Metabarcoding provides useful low-cost multi-species data but still requires 17 considering all the influencing methodological and environmental variables during data interpretation. The implementation of eDNA into the monitoring of invasive species represents an efficient approach for an early warning system; however, the sensitivity of the method has to be investigated deeply.