Toxicological assessment of divalent ion-modified ZnO nanomaterials through artificial intelligence and in vivo study.

ABSTRACT

The nanotechnology-driven industrial revolution widely relies on metal oxide-based nanomaterial (NM). Zinc oxide (ZnO) production has rapidly increased globally due to its outstanding physical and chemical properties and versatile applications in industries including cement, rubber, paints, cosmetics, and more. Nevertheless, releasing Zn2+ ions into the environment can profoundly impact living systems and affect water-based ecosystems, including biological ones. In aquatic environments, Zn2+ ions can change water properties, directly influencing underwater ecosystems, especially fish populations. These ions can accumulate in fish tissues when fish are exposed to contaminated water and pose health risks to humans who consume them, leading to symptoms such as nausea, vomiting, and even organ damage. To address this issue, safety of ZnO NMs should be enhanced without altering their nanoscale properties, thus preventing toxic-related problems. In this study, an eco-friendly precipitation method was employed to prepare ZnO NMs. These NMs were found to reduce ZnO toxicity levels by incorporating elements such as Mg, Ca, Sr, and Ba. Structural, morphological, and optical properties of synthesized NMs were thoroughly investigated. In vitro tests demonstrated potential antioxidative properties of NMs with significant effects on free radical scavenging activities. In vivo, toxicity tests were conducted using Oreochromis mossambicus fish and male Swiss Albino mice to compare toxicities of different ZnO NMs. Fish and mice exposed to these NMs exhibited biochemical changes and histological abnormalities. Notably, ZnCaO NMs demonstrated lower toxicity to fish and mice than other ZnO NMs. This was attributed to its Ca2+ ions, which could enhance body growth metabolism compared to other metals, thus improving material safety. Furthermore, whether nanomaterials' surface roughness might contribute to their increased toxicity in biological systems was investigated utilizing computer vision (CV)-based AI tools to obtain SEM images of NMs, providing valuable image-based surface morphology data that could be correlated with relevant toxicology studies.