Evaluating metal oxide nanoparticle (MeOx NP) toxicity with different types of nano descriptors mainly focusing on simple periodic table-based descriptors: a mini-review

Abstract

Given the rapid growth of nanotechnology, it is essential to know the hazardous effects of metal oxide nanoparticles (MeOx NPs) posed to living organisms within the ecosystem. With the availability of experimental data and the rapid development of different modeling algorithms, quantitative structure–activity relationship (QSAR) models play an essential role in modern toxicological research. Molecular descriptors are the core features in determining the performance of a QSAR model. Classical QSAR descriptors remain unable to describe the structural details of metal oxide nanostructures, and due to the lack of sufficiently good quality descriptors, appropriate modeling of properties and toxicities of nanoparticles becomes complex. Conventionally, theoretical descriptors and/or experimentally derived properties are used as descriptors for nano-QSAR modeling. To simplify the calculation of nanostructure properties and to determine the nano-level interaction without computationally expensive quantum-chemical modeling and/or performing further experiments, simple periodic table-based (PT) descriptors may be successfully employed to model various endpoints of metal oxide NPs. Studying the toxicity of nanomaterials (NMs) is undoubtedly challenging due to the lack of standardized data and specific, universal nano descriptors. This poses significant obstacles in accurately simulating NM toxicity on a realistic and spatial scale. However, this review provides a comprehensive view of the available nano descriptors and their respective pitfalls in nanotoxicology studies, which aims to alleviate this issue and improve the accuracy of such studies. Here, we have thoroughly discussed the advantages of incorporating PT descriptors in nano toxicological modeling, especially in terms of their impact on the environment. In addition, we have demonstrated how nano-QSAR models based on PT descriptors can effectively fill the gaps in toxicity and ecotoxicity data for various NMs, such as nano-pesticides, nano-medicines, and nano-mixtures. Moreover, we have explored the potential of developing universal descriptors that consider more realistic biological systems and ecosystems, thereby paving the way for the prospects of new descriptors.