Chronic sublethal effects of pulsed versus constant pesticide exposures on the freshwater amphipod Gammarus fossarum: Behavioural and biochemical impairments under field-realistic conditions

• Pulsed vs constant pesticide exposures of Gammarus fossarum are compared in microcosms • Both exposure scenarios induce behavioural and biochemical impairments • Mobility, respiration and feeding shifts depend on the tested pesticide and scenario • Boscalid and imidacloprid induce oxidative stress, neurotoxicity and metabolism shifts • All biomarkers suggest gammarid is more affected by constant than pulsed imidacloprid After their application to agricultural soils, pesticides frequently reach surface waters, exposing aquatic organisms to fluctuating concentrations influenced by hydrological conditions. However, the influence of exposure dynamics on biological responses remains poorly investigated. In this study, the non-target amphipod Gammarus fossarum was exposed for four weeks in microcosms to environmentally realistic concentrations of three pesticide classes: the insecticide imidacloprid, the fungicide boscalid, and the herbicide bentazone. Two exposure scenarios were compared: (i) repeated short-term pulses mimicking runoff events and (ii) constant low-level exposures, both time-weighted average concentrations. Biochemical biomarkers (enzymes involved in essential functions and stress responses) and behavioural traits (locomotion, respiration, mating and feeding rate) were monitored to assess sublethal effects at cellular and individual levels. Both exposure regimes elicited changes in behaviour of animals and biochemical pathways. Notably, hyperactivity coupled with increased respiration occurred, with pesticide-specific intensity. Unlike repeated pulses, constant exposures to the insecticide and fungicide exerted more persistent inhibitory effects on feeding. Biochemical analyses revealed pesticide-specific disturbances in digestive metabolism, oxidative stress response, and neuronal activity, and suggested trade-offs in energy allocation under chronic chemical stress. Furthermore, these sublethal responses provide evidence of unintended impacts of these chemicals, even at concentrations compliant with environmental quality standards, to which wildlife is naturally exposed under various hydrological conditions. Since such disruptions in physiology and individual performance in this ecosystem engineer may precede population- and ecosystem-level consequences, our results highlight the need to integrate exposure dynamics and multi-level sublethal endpoints in ecotoxicological risk assessment to improve the protection of aquatic wildlife in agriculturally impacted freshwaters.