Advancing Mechanistic Ecotoxicology with Lipidomics:

Ecotoxicology seeks to understand how chemicals affect organisms and ecosystems. Traditional environmental risk assessment (ERA) relies on endpoints such as survival and reproduction, but there is increasing demand for mechanistic tools that detect early signs of stress. This thesis explores untargeted lipidomics as a molecular approach to link early biochemical disruptions with life-history traits in soil and aquatic invertebrates, contributing to Next-Generation Risk Assessment (NGRA). The work examined sublethal effects of the insecticide teflubenzuron in three Collembola species and the antidepressant fluoxetine in the freshwater snail Lymnaea stagnalis. In Folsomia candida, teflubenzuron induced lipidomic alterations within 2–7 days, indicating disrupted energy metabolism and oxidative stress before reproduction declined. Time-resolved analyses revealed that early molecular changes propagated to life-history effects. Cross-species comparisons showed that bioaccumulation and lipid responses differed according to ecological traits, yet some lipid classes were consistently affected, suggesting potential cross-species indicators of exposure. In aquatic systems, fluoxetine did not affect adult survival but impaired egg development in a dose-dependent manner. Molecular analyses revealed disrupted neurotransmission and lipid balance in adults and embryos, linking biochemical imbalance to developmental arrest and demonstrating partial recovery when exposure ceased. Overall, lipidomic alterations frequently preceded phenotypic effects and reflected disruptions in energy storage, membrane integrity, and signaling pathways. By integrating molecular data with classical endpoints, this thesis demonstrates how lipidomics can serve as an early-warning and mechanistic tool in ERA. The findings support a transition toward predictive, mechanism-based frameworks for evaluating chemical safety across species and environmental compartments.