Aging of nanosized titanium dioxide modulates the effects of dietary copper exposure on Daphnia magna - an assessment over two generations

Nanosized titanium dioxide (nTiO2) is widely used in products, warranting its discharge from various sources into surface water bodies. However, nTiO2 co-occurs in surface waters with other contaminants, such as metals. Studies with nTiO2 and metals have indicated that the presence of natural organic matter (NOM) can mitigate their toxicity to aquatic organisms. In addition, "aging" of nTiO2 can affect toxicity. However, it is a research challenge, particularly when addressing sublethal responses from dietary exposure over multiple generations. We, therefore exposed the alga Desmodesmus subspicatus to nTiO2 (at concentrations of 0.0, 0.6 and 3.0 mg nTiO2/L) in nutrient medium aged for 0 or 3 days with copper (Cu) at concentrations of 0 and 116 mu g Cu/L and with NOM at concentrations equivalent to 0 and 8 mg total organic carbon (TOC) per litre. Subsequently, the exposed alga was fed to Daphnia magna for 23 days over two generations and survival, reproduction and body length were assessed as endpoints of toxicity. In parallel, Cu accumulation and depuration from D. magna were measured. The results indicate that the reproduction of D. magna was the most sensitive parameter in this study, being reduced by 30% (at both parental (F0) and filial (F1) generations) and 50% (at F0 but not F1) due to the dietary Cu exposure in combination with nTiO2 for 0 and 3 days aging, respectively. There was no relationship between the effects observed on reproduction and Cu body burden in D. magna. Moreover, D. magna from the F1 generation showed an adaptive response to Cu in the treatment with 3.0 mg nTiO2/L aged for 3 days, potentially due to epigenetic inheritance. Unexpectedly, the presence of NOM hardly changed the observed effects, pointing towards the function of algal exopolymeric substances or intracellular organic matter, rendering the NOM irrelevant. Ultimately, the results indicate that the transferability of the impacts observed during the F0 to the responses in the F1 generation is challenging due to opposite effect directions. Additional mechanistic studies are needed to unravel this inconsistency in the responses between generations and to support the development of reliable effect models.