TOXICOLOGICAL EFFECTS OF GRAPHENE FAMILY NANOMATERIALS AFTER CUTANEOUS EXPOSURE

Graphene and the so-called Graphene Family Nanomaterials (GFNs) have unique physicochemical properties, making them promising candidates for a wide range of applications in several fields. However, they may pose health risks, and little is known about their potential human toxicity, so far. Skin is the largest organ of human body and the cutaneous exposure represents one of the major exposure routes to GFNs. Even though cutaneous contact to other related materials, such as graphite and carbon nanomaterials, has been associated with increased incidence of skin diseases, such as airborne irritant contact dermatitis, hyperkeratosis and naevi, the toxicity of GFNs at the skin level remains largely unexplored. Thus, a study was carried out using HaCaT skin keratinocytes to investigate the effects of a research grade few layer graphene (FLG) and three graphene oxides (GOs): a research grade GO (GO1) and two commercial GOs (GO2 from Antolin Group and GO3 from Graphenea). At concentrations higher than 30 μg FLG/mL and 1 μg GO/mL, these GFNs induced significant mitochondrial and plasma-membrane damages with variable potencies, depending on GFNs oxidation state: the less (FLG) and the most (GO3) oxidized compounds were the less and the most cytotoxic, respectively. The damage at the plasma membrane level was confirmed by epifluorescence microscopy and confocal analysis, demonstrating that GFNs were strongly bound to cell membranes. Moreover, long-term exposures (up to 14 days) to low GFNs concentrations (0.1 μg/mL) showed only slight reductions of mitochondrial activity, being significant only after 10 days exposure. The effects on mitochondrial damage induced by the less (FLG) and the most (GO3) cytotoxic GFN were further investigated, demonstrating a concentration- and time-dependent mitochondrial depolarization. This effect was not dependent on mitochondrial permeability transition pores formation but appeared to be dependent on a significant concentration- and time-dependent reactive oxygen species (ROS) production, mainly induced by the activation of flavoprotein-based oxidative enzymes, such as NADH dehydrogenase and xanthine oxidase. In the second part of the project, the inflammatory effects of FLG and GO3 at the skin level were evaluated. Sub-cytotoxic concentrations of both compounds (0.1 – 1 μg/mL) induced a significant release of pro-inflammatory mediators (granulocyte macrophage colony stimulating factor, interleukin-1α, -6 and -8, and tumor necrosis factor α) from HaCaT cells, mainly after a short exposure time to GFNs (4 h), followed by long recovery times in GFNs-free media (20 or 68 h). However, the conditioned media, obtained after exposure of HaCaT cells to FLG or GO3 under these conditions, did not induce a significant differentiation of THP-1 monocytes towards macrophages or dendritic cells. Similarly, they did not induce any significant release of inflammatory mediators by these cells, suggesting only a moderate inflammatory reaction. These results were confirmed by the investigation of the sensitizing potential of FLG and GO3 on THP-1 monocytes (according to the Organisation for Economic Co-operation and Development, OECD, guideline n°442E), demonstrating that these nanomaterials are not skin sensitizers. On the whole, these results suggest that, even though these compounds are able to induce significant mitochondrial and plasma membrane damage after long exposure times to high concentrations, FLG and GO induce only minor toxic effects at the skin level.