This Mutagenesis special issue is on the topic of nanogenotoxicology. It unites a collection of reports that provide insight into: (i) the properties of engineered nanomaterials (ENMs) that contribute to genotoxicity, (ii) the genotoxic mechanisms associated with DNA damage observed in both in vitro and in vivo tests and (iii) the future test systems that will provide more accurate prediction of ENM genotoxicity to support regulatory hazard assessment frameworks. The contributions within therefore provide collective oversight of our current understanding, coupled to future perspectives aimed at overcoming technical hurdles and describing novel analytical methods to further advance the field.
The nanotechnology industrial sector is delivering significant scientific, economic and societal benefits. The industry is therefore expanding, leading to increased human exposure to engineered nanomaterials (ENMs) through direct application (e.g. in cosmetics products, nanomedicine, food additives) and indirectly, due to their increasing abundance in the workplace and environment. Despite this growth, there remain limitations in our knowledge on the human health and environmental impacts of ENM exposure that affect the public’s trust in this new technology. Over recent years, there has been increasing momentum in nanosafety research, boosting our understanding of key factors that govern ENM toxicity and, importantly, the DNA damaging capacity of ENMs. Understanding genotoxicity is vital as substances that damage DNA commonly lead to carcinogenesis. Accumulation of DNA damage in somatic cells is also related to degenerative conditions such as immune dysfunction, while in germ cells DNA damage is associated with malformation or heritable damage in subsequent generations. It is therefore imperative that we reach a detailed understanding of the factors that orchestrate genotoxicity, in addition to determining the modes of action (MoA) and types of DNA damage induced by ENMs in order to support future hazard and risk assessment.
An additional complication, however, is that it has become increasingly evident that our current safety testing regimes may require adaptation to predict ENM genotoxicity more accurately. There is also a lack of standardised safety testing protocols suitable for ENMs. Thus, improved test systems are required with a particular focus on in vitro assays to reduce the necessity for animal testing. This special issue therefore unites key elements of strategic research that provide insight into: (i) the properties of ENMs that contribute to genotoxicity, (ii) the genotoxic mechanisms contributing to DNA damage observed within both in vitro and in vivo tests and (iii) the future test systems that will provide more accurate prediction of ENM genotoxicity to support regulatory hazard assessment frameworks.
Of the 20 publications in this special issue dedicated to the genotoxicity of ENMs, eight focus on in vivo studies (1–8), nine are in vitro studies (9–17) and three are reviews highlighting future perspectives (18–20). Mechanisms of ENM genotoxicity and their adverse outcome pathways (AOP) have been investigated in recent European and national research projects. Several results from European Commission 7th Framework Programme for Research and Technological Development (FP7) projects such as ENPRA (9), SUNPAP (3), NANoREG (2,10,11) and NanoMILE (14) are included in this issue sharing identical reference ENMs with the same intrinsic properties.
Nanotechnology has applications in a very wide range of industrial sectors. Titanium dioxide (TiO2) is the most highly used ENM worldwide, occurring in paints, plastics, papers, inks, foods, pharmaceuticals, toothpaste and cosmetics. There has therefore been extensive research assessing its genotoxicity and MoA; indeed TiO2 is the most frequently investigated ENM in this issue. Three manuscripts describe studies with TiO2 exposure in rats and mice (4–6), and six papers report in vitro effects of TiO2 in several cell lines (9–14). Silver ENMs were evaluated in four in vitro studies (9,10,13,17), while carbon-based ENMs (fullerenes, single and multiwall carbon nanotubes) were assessed both in vivo (7,8) and in vitro (9,15,16). Other materials considered in the special issue include silica (1), barium sulfate (BaSO4) (2), cerium oxide (CeO2) (2,10), zinc oxide (ZnO) (9,10), iron oxide (17) and cellulose (3). In these studies, a range of DNA damage endpoints were considered including DNA strand breaks (SBs) measured with the comet assay (with the option of detecting oxidised purines using formamidopyrimidine DNA glycosylase, Fpg), DNA repair, gene mutations in vivo and in vitro, chromosomal damage measured by the micronucleus (MN) assay and cell transformation. Furthermore, novel approaches are presented in several papers including epigenetics (8,16), toxicogenomics approaches (1,4,6,8,14,16), high-throughput assays and miniaturisation (10,11,19).
Data generated by the in vivo studies provide insight into both the level of genotoxicity induced in several species by ENMs and their underlying MoA. Pfuhler et al. (1) investigated the genotoxicity of silica ENMs by examining DNA damage and transcriptional regulation in the liver after intravenous administration; an increase in DNA base oxidation was consistent with a MoA involving reactive oxygen species (ROS). Histopathology showed liver damage and neutrophil involvement, while changes in expression of key genes indicated that inflammation and oxidative stress were the primary response, with DNA damage resulting as a secondary effect. The authors also suggested that the concept of a threshold might be applicable in the risk assessment of these materials. In contrast to silica, Cordelli et al. (2) reported that inhalation of CeO2 and BaSO4 ENMs by rats for up to 6 months caused no genotoxicity in a range of blood-based tests (comet assay, flow cytometric Pig-a gene mutation assay and MN assay). It remains to be seen whether the full 2-year exposure to these two ENMs will result in lung tumour formation.
Of the in vivo studies conducted in mice, the association between inflammation and oxidative stress with genotoxicity was a key focus. Catalan et al. (3) investigated possible genotoxicity of nanofibrillated cellulose administrated to mice by a single pharyngeal aspiration. This ENM increased the recruitment of inflammatory cells to the lungs, and caused a dose-dependent increase in mRNA expression of tumour necrosis factor α, interleukins and chemokine and neutrophilic accumulation in the alveolar lung space. DNA damage was seen in lung cells, but there was no systemic genotoxic effect in the bone marrow MN assay. Three other studies conducted in mice investigated TiO2 genotoxicity. Li et al. (4) evaluated the effect of anatase TiO2 after i.p. injection using the comet assay and whole-genome microarray technology to analyse gene expression patterns in the liver and lungs. DNA SBs were only detected in the liver, but DNA base oxidation and distinct gene expression patterns were observed in both liver and lungs. The influence of rutile TiO2 surface properties on the inflammatory response, acute phase response and genotoxicity was investigated by Wallin et al. (5) in mice following a single intratracheal instillation. DNA SBs were assayed in bronchial alveolar lavage (BAL) cells, lung and liver tissues, and the hepatic acute phase response was analysed by real-time quantitative PCR. Exposure to TiO2 ENMs with both negative and positively charged surface modifications induced DNA SBs in lung tissue, but differences were found in the liver, indicating that surface functionalisation inconsistently influenced toxicity in different tissues. An interesting comparative study presents a comprehensive toxicogenomic analysis of lung responses in mice exposed to a single intratracheal instillation of six individual TiO2 ENMs with different sizes (8, 20 and 300 nm), crystalline structure (anatase, rutile, anatase/rutile) and surface modifications (hydrophobic, hydrophilic) (6). Although all TiO2 ENMs induced lung inflammation as measured by the neutrophil influx in BAL fluid, rutile-type TiO2 ENM induced the greatest degree of inflammation and the largest number of differentially expressed genes. Increased collagen staining and fibrosis-like changes were seen in lung sections following exposure to the rutile TiO2 ENM at the highest dose tested. Among the anatase ENMs, the smallest were most reactive. The results suggest that the severity of lung inflammation is property-specific; however, the underlying mechanisms (genes and pathways perturbed) leading to inflammation were the same for all ENM types. Collectively, these investigations therefore highlight the importance of an inflammatory response and its association with secondary genotoxicity.
The murine in vivo studies are complemented by two marine invertebrate studies using mussels and zebrafish to explore the effects of exposure to carbon-based ENMs with molecular markers at different levels of biological organization (7,8). Mussels were exposed to C60 fullerene, either alone or in combination with benzo(a)pyrene (7). The exposure induced genotoxic responses (increased DNA SBs and gene expression alterations) that were reversible after a recovery period, suggesting the ability of mussels to cope with these exposures. Zebrafish were also exposed to C60 fullerene, single-walled carbon nanotubes (SWCNTs), short or long multiwalled carbon nanotubes (MWCNTs) all at low doses via the diet (8). The subsequent consequences of exposure were then investigated in the brain, gonads and gastrointestinal tract using global genomic methylation and ‘omic’ approaches. Even at low concentrations, carbon nanotubes (CNTs) were capable of inducing significant cellular and genomic modifications in a range of tissues.
With the increasing production of ENMs, and their wide ranging applications impacting many aspects of human life, there is a great demand for rapid and reliable safety testing. However, this cannot be achieved with in vivo systems as they are expensive, time consuming and raise ethical concerns. In vitro approaches in contrast are high-throughput, cost-effective and are more desirable alternatives to in vivo hazard assessment tests. The in vitro studies utilized in this special issue employ a range of mammalian cell models derived from lung (A549, BEAS2B) kidney (HK-2), liver, (HepG2), blood (TK6, THP-1, L5178Y), gut (Caco-2, HCT116) and fibroblast (Bhas42) cells. These cells were then applied to evaluate several genotoxic endpoints resulting from exposure to an array of different ENMs.
Two studies evaluated the genotoxicity of several ENMs (up to 10 different types) in multiple cell lines (9,10). Both studies demonstrated that DNA damage induction varied considerably with the cell line selected, exposure time and use of surfactants. These studies highlight the importance of selecting appropriate test cells, as use of a cell line merely on the basis of its origin may provide poor insight into ENM genotoxic hazard. Similarly, Li et al. (13) demonstrated that cells harbouring p53 gene mutations were more sensitive to micronucleus induction by certain ENMs than the p53 wild-type cells. These studies also demonstrated that aligned testing protocols were of critical importance as slight variations introduced significant differences in genotoxic outcome. Di Bucchianico et al. (11) stressed the necessity of avoiding light exposure when assessing photocatalytic ENMs to prevent false positive results following their evaluation of DNA strand breaks, oxidised bases and micronuclei induced by three reference TiO2 ENMs (two anatase and one rutile) in BEAS-2B cells.
Significant effort has gone into increasing the robustness of genotoxicity assays for ENMs over recent years and several papers in this special issue focused on that particular point. For DNA SB detection, El Yamani et al. (10) applied the high throughput miniaturized comet assay with 12 minigels on one slide and Di Bucchianico et al. (11) used a similar, slightly modified 8 gel version, to increase throughput, reduce variability and increase precision of the results generated. The inclusion of Fpg in the comet assay also allowed detection of indirect genotoxic effects via oxidative stress (10). With respect to the MN assay, caution was recommended when analysing the induction of micronuclei using the flow cytometry based method as ENM spectral properties (e.g. high adsorption capacity, fluorescence or optical properties) can lead to unexpected interactions with experimental detection systems (13,19). Additionally, confounding factors affecting cytotoxicity and genotoxicity assessment, including ENM coatings, use of cytochalasin B and presence of fetal bovine serum in cell treatment medium were emphasised (13). Collectively, the data presented by these papers highlight the importance of methodological adaptations required to more accurately measure ENM genotoxicity.
As with the in vivo studies, several in vitro based genotoxicity studies considered the impact of TiO2 exposure. In addition to the in vitro studies mentioned above, Proquin et al. (12) investigated TiO2 used as a food additive (E171—a mixture of micro- and nanoparticles) with the standard comet assay and MN tests in human Caco-2 and HCT116 cells; they also studied ROS formation in a cell-free environment (by electron spin/paramagnetic resonance (ESR/EPR) spectroscopy). E171 was found to induce ROS formation and both its micro- and nano-fractions displayed genotoxic potential, raising concern about potential adverse effects associated with E171 and TiO2 in food. TiO2 ENMs were also found to induce comparable cyto- and genotoxic responses in BEAS-2B and A549 cells, which were associated with a general down-regulation of DNA repair processes linked to increased methylation of the promoter regions of these genes (14). Similarly, CNTs were found to induce differential methylation of genes involved in several signalling cascades including carcinogenesis, VEGF and platelet-activation pathways (16). This study further demonstrated that in THP-1 cells SWCNTs induced higher cytotoxicity and genotoxicity than MWCNTs. This difference could be explained by the findings of Borghini et al. (15), who demonstrated that MWCNT exposure significantly increased DNA repair activity and enhanced telomere shortening (which promotes replicative senescence) in A549 cells, accounting for the seemingly low mutagenicity of these ENMs despite their pro-oxidant behaviour. Interestingly, this also correlates with the conclusion presented by Thongkam et al. (9), who reported that measurement of the intrinsic oxidant generating capacity of ENM is a poor predictor of genotoxicity, most likely as this does not take into account cell homeostatic mechanisms such as DNA repair and methylation profiles.
The MN and comet assays are the most popular in vitro ENM genotoxicity studies, measuring gross chromosomal damage and DNA SBs respectively. However there has been less focus on the ability of ENM to induce point mutations—a critical endpoint to consider when undertaking a comprehensive genotoxicity evaluation. Furthermore, few carcinogenicity studies have been conducted. In this special issue, Gabelova et al. (17) investigated the capacity of two reference nanosilvers (spherical and fibrous shape) and three magnetite ENMs with different surface properties to induce thymidine kinase (Tk+/−) mutations in L5178Y mouse lymphoma cells and transformed foci in the Bhas 42 cell transformation assay (CTA). Spherical AgNM300 showed neither mutagenic nor carcinogenic potential. In contrast, silver nanorods/wires (AgNM302) significantly increased the induction of both gene mutations and transformed foci. The results revealed that fibrous shape underlies the mutagenic and carcinogenic potential of nanosilver while surface chemistry affects the biosafety of magnetite ENMs.
The special issue concludes with three articles presenting future perspectives for ENM genotoxicity evaluation, considering in depth the following key points: (i) the role of DNA repair in governing genotoxic outcome (18), (ii) a critical review of novel and emerging analytical platforms to increase throughput of genotoxicity testing (19) and (iii) methods to increase the physiological relevance and enhance the reporting accuracy of in vitro genotoxicity test systems to minimize the requirement for animal testing and support the international ‘3Rs’ efforts (20). Together these papers highlight the advantages and limitations of varying experimental approaches, with specific reference to their application for evaluating ENMs. They also provide recommendations to assist the scientific community in making progress towards the goal of a more comprehensive understanding of nanogenotoxicity and associated MoA driving this genetic damage, and in defining the methods capable of supporting future ENM regulatory frameworks. This has clearly become a fast-paced scientific field, with exciting developments that will be important in informing protective measures for both human health and the environment.
Conflict of interest statement: None declared.