Dissolution and bandgap paradigms for predicting the toxicity of metal oxide nanoparticles in the marine environment: an in vivo study with oyster embryos

Dissolution and bandgap paradigms have been proposed for predicting the ability of metal oxide nanoparticles (NPs) to induce oxidative stress in different in vitro and in vivo models. Here, we addressed the effectiveness of these paradigms in vivo and under conditions typical of the marine environment, a final sink for many NPs released through aquatic systems. We used ZnO and MnO2 NPs as models for dissolution and bandgap paradigms, respectively, and CeO2 NPs to assess reactive oxygen radical (ROS) production via Fenton-like reactions in vivo. Oyster embryos were exposed to 0.5-500 μM of each test NP over 24 h and oxidative stress was determined as a primary toxicity pathway across successive levels of biological complexity, with arrested development as the main pathological outcome. NPs were actively ingested by oyster larvae and entered cells. Dissolution was a viable paradigm for predicting the toxicity of NPs in the marine environment, whereas the surface reactivity based paradigms (i.e. bandgap and ROS generation via Fenton-like reaction) were not supported under seawater conditions. Bio-imaging identified potential cellular storage-disposal sites of solid particles that could ameliorate the toxicological behavior of non-dissolving NPs, whilst abiotic screening of surface reactivity suggested that the adsorption-complexation of surface active sites by seawater ions could provide a valuable hypothesis to explain the quenching of the intrinsic oxidation potential of MnO2 NPs in seawater.