This study presents the results of applying deep learning methodologies within the ecotoxicology field, with the objective of training predictive models that can support hazard assessment and eventually the design of safer engineered nanomaterials (ENMs). A workflow applying two different deep learning architectures on microscopic images of Daphnia magna is proposed that can automatically detect possible malformations, such as effects on the length of the tail, and the overall size, and uncommon lipid concentrations and lipid deposit shapes, which are due to direct or parental exposure to ENMs. Next, classification models assign specific objects (heart, abdomen/claw) to classes that depend on lipid densities and compare the results with controls. The models are statistically validated in terms of their prediction accuracy on external D. magna images and illustrate that deep learning technologies can be useful in the nanoinformatics field, because they can automate time‐consuming manual procedures, accelerate the investigation of adverse effects of ENMs, and facilitate the process of designing safer nanostructures. It may even be possible in the future to predict impacts on subsequent generations from images of parental exposure, reducing the time and cost involved in long‐term reproductive toxicity assays over multiple generations.
Engineered nanoparticles (NPs) undergo physical, chemical, and biological transformation after environmental release, resulting in different properties of the “aged” versus “pristine” forms. While many studies have investigated the ecotoxicological effects of silver (Ag) NPs, the majority focus on “pristine” Ag NPs in simple exposure media, rather than investigating realistic environmental exposure scenarios with transformed NPs. Here, the effects of “pristine” and “aged” Ag NPs are systematically evaluated with different surface coatings on Daphnia magna over four generations, comparing continuous exposure versus parental only exposure to assess recovery potential for three generations. Biological endpoints including survival, growth and reproduction and genetic effects associated with Ag NP exposure are investigated. Parental exposure to “pristine” Ag NPs has an inhibitory effect on reproduction, inducing expression of antioxidant stress related genes and reducing survival. Pristine Ag NPs also induce morphological changes including tail losses and lipid accumulation associated with aging phenotypes in the heart, abdomen, and abdominal claw. These effects are epigenetic remaining two generations post‐maternal exposure (F2 and F3). Exposure to identical Ag NPs (same concentrations) aged for 6 months in environmentally realistic water containing natural organic matter shows considerably reduced toxicological effects in continuously exposed generations and to the recovery generations.
Pristine engineered nanomaterials (NMs) entering the aquatic environment become ‘aged’ during their lifetime via chemical, physical and/or biological process. Therefore, traditional ecotoxicology tests which were designed for soluble chemicals prior to the emergence of NMs, use pristine NMs and salt-only media which are not representative of realistic NM exposure scenarios. Exposure medium and NM ageing moderation of NM toxicity were explored using Daphnia magna multigenerational studies aiming to determine whether the daphnids adapted to continuous exposure and/or if parent-only exposure resulted in epigenetic effects in subsequent generations. Daphnids were continuously, or parent-only, exposed to pristine and aged titanium dioxide (TiO2) and silver (Ag) NMs, in a standard high hardness culture media and synthetic European Class V lowland water. Pristine NMs in the standard culture medium had the most severe toxic consequences, and displayed negative effects in two generations post exposure. NMs aged in the class V water had fewer overall toxic consequences on growth and longevity across all generations in both continuous and parent-only exposure scenarios. Overall, in the absence of environmentally relevant media and aged NMs, standardised Daphnia tests currently overestimate NM toxicity, and fail to consider potential impacts in subsequent generations. The results demonstrate the importance of updating standard testing to reflect scientific advances and increase stakeholder trust in regulation.
Secreted corona composition of Ag and TiO 2 NMs, evaluated after 7 days of continuous exposure to daphnids, provides important insights into the biological pathways perturbed by the NMs.