Choose your cell model wisely: The in vitro nanoneurotoxicity of differentially coated iron oxide nanoparticles for neural cell labeling

Currently, there is a large interest in the labeling of neural stem cells (NSCs) with iron oxide nanoparticles (IONPs) to allow MRI-guided detection after transplantation in regenerative medicine. For such biomedical applications, excluding nanotoxicity is key. Nanosafety is primarily evaluated in vitro where an immortalized or cancer cell line of murine origin is often applied, which is not necessarily an ideal cell model. Previous work revealed clear neurotoxic effects of PMA-coated IONPs in distinct cell types that could potentially be applied for nanosafety studies regarding neural cell labeling. Here, we aimed to assess if DMSA-coated IONPs could be regarded as a safer alternative for this purpose and how the cell model impacted our nanosafety optimization study. Hereto, we evaluated cytotoxicity, ROS production, calcium levels, mitochondrial homeostasis and cell morphology in six related neural cell types, namely neural stem cells, an immortalized cell line and a cancer cell line from human and murine origin. The cell lines mostly showed similar responses to both IONPs, which were frequently more pronounced for the PMA-IONPs. Of note, ROS and calcium levels showed opposite trends in the human and murine NSCs, indicating the importance of the species. Indeed, the human cell models were overall more sensitive than their murine counterpart. Despite the clear cell type-specific nanotoxicity profiles, our multiparametric approach revealed that the DMSA-IONPs outperformed the PMA-IONPs in terms of biocompatibility in each cell type. However, major cell type-dependent variations in the observed effects additionally warrant the use of relevant human cell models. Statement of Significance Inorganic nanoparticle (NP) optimization is chiefly performed in vitro. For the optimization of iron oxide (IO)NPs for neural stem cell labeling in the context of regenerative medicine human or rodent neural stem cells, immortalized or cancer cell lines are applied. However, the use of certain cell models can be questioned as they phenotypically differ from the target cell. The impact of the neural cell model on nanosafety remains relatively unexplored. Here we evaluated cell homeostasis upon exposure to PMA- and DMSA-coated IONPs. Of note, the DMSA-IONPs outperformed the PMA-IONPs in each cell type. However, distinct cell type-specific effects were witnessed, indicating that nanosafety should be evaluated in a human cell model that represents the target cell as closely as possible. (C) 2017 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.