Defining the mechanistic toxicity of metal oxide nanoparticles in vitro, under physiologically relevant oxygen concentrations

Metal oxide nanoparticles (MONPs) are intended for use in numerous consumer applications, leading to inevitable human exposure. Previous work conducted in hyperoxic cell culture conditions (21% O2, 5% CO2) with nanoparticles (NPs) has proven the ability of some material types to induce genotoxicity and inflammatoxicity. Alteration in intracellular calcium [i(Ca2+)] signalling is involved in facilitating toxicity through the alteration of signal-transduction pathways, but there is less understanding of the impact of NPs exposure upon changes in such signalling pathways. Furthermore, whilst human cells cultured in ambient air may induce a particular toxicity profile, this may not be the same under the physiologic oxygen conditions experienced in the human body. Therefore, the aim of this study was to assess the impact of anatase TiO2 (NM-102), Rutile TiO2 (NM-104) and dextran coated superparamagnetic Fe3O4 (dSPIONs) upon monocytes (THP-1), macrophages (dTHP-1) and hepatocarcinoma (HepG2) cells in both an in vivo-resembling physioxia environment (5%O2, 5%CO2) and hyperoxic cell culture conditions (21%O2, 5%CO2). Their impact on i(Ca2+) homeostasis and how it relates to their potential genotoxic potential was also evaluated.Due to the importance of different physicochemical characteristics for the facilitation of toxicity, all MONPs were characterized. MONPs hydrodynamic diameter (HD) and ζ-potential (ζ) in PBS were identified using dynamic light scattering: NM-102: HD=391.9nm, ζ =7.1±2.0mV; NM-104: HD=255nm, ζ=14.6 +/- 2.1mV; dSPIONs: HD=88.6nm, ζ =10.4±1.3mV. The possible toxic effect of NPs depends on their concentration and duration of their interaction with cells. Therefore, following 24h exposure to dSPIONs (0-100µg/ml), concentration-dependent and cell-type-dependent (dTHP1>THP-1>HepG2) significant increases in NP-cellular interaction were observed, which was significantly greater in the physioxic culture environment. Concurrent, significant loss of dSPION-associated cell proliferation (evaluated using relative population doubling) in all cell lines and significant increases in DNA damage was also identified in HepG2 cells (using the cytokinesis block micronucleus assay), albeit only in physioxia. Exposure to ≥10ug/ml NM-102 and NM-104 resulted in significant, two-fold increases in micronuclei formation in HepG2 in both environments. All MONPs induced a significant increase in tumour necrosis factor-α and interleukin-8 secretion in all cell lines and oxygen culture environments. Increase in the production of the chemokines was correlated with the observed HepG2 cell genotoxicity. In all cell lines and cell culture environments, treatment for up to 5h with NM-102 or dSPIONs triggered cell type specific increases in i(Ca2+) that correlated with the reduction of cellular antioxidant glutathione (measured after 5h treatment with all the MONPs). After pre-treatment of the cell lines with antioxidant trollox in all cell culture environments i(Ca2+) appeared to be increased independently from the change of cellular redox status. Environment-specific biological interaction and impacts with regard to NP uptake, genotoxic effects, and consequence on cellular signaling mechanisms were only observed with dSPIONs in a physioxic culture environment, while NM-102 and NM-104 induced similar effects in both environments. The results presented in this study allow the conclusion that the environmental oxygen content has an impact on the NP toxicity profiles although it is NP dependent.