Current investigations into the genotoxicity of zinc oxide and silica nanoparticles in mammalian models in vitro and in vivo: carcinogenic/genotoxic potential, relevant mechanisms and biomarkers, artifacts, and limitations

Abstract

Engineered nanoparticles (NPs) are widely used in many sectors, such as food, medicine, military, and sport, but their unique characteristics may cause deleterious health effects. Close attention is being paid to metal NP genotoxicity; however, NP genotoxic/carcinogenic effects and the underlying mechanisms remain to be elucidated. In this review, we address some metal and metal oxide NPs of interest and current genotoxicity tests in vitro and in vivo. Metal NPs can cause DNA damage such as chromosomal aberrations, DNA strand breaks, oxidative DNA damage, and mutations. We also discuss several parameters that may affect genotoxic response, including physicochemical properties, widely used assays/end point tests, and experimental conditions. Although potential biomarkers of nanogenotoxicity or carcinogenicity are suggested, inconsistent findings in the literature render results inconclusive due to a variety of factors. Advantages and limitations related to different methods for investigating genotoxicity are described, and future directions and recommendations for better understanding genotoxic potential are addressed.

Keywords:

• carcinogenicity
• exposure assessment
• genotoxicity
• nanoparticles
• risk evaluation

Introduction

The rapidly growing nanotechnology industry will have significant economic and scientific impact in areas such as aerospace engineering, nanoelectronics, environmental remediation, and health care.¹ The design and development of novel engineered nanoparticles (NPs) are important to the industry due to beneficial physicochemical features that have led to over 800 NP-containing consumer products.² Hence, human exposure is high and continues to increase dramatically.

Due to their small size and great surface area coupled with physicochemical characteristics such as metal contaminations and charged surfaces, NPs may exhibit unpredictable genotoxic properties. Indirect DNA damage may be caused by induction of oxidative stress and inflammatory responses. Small NPs may cross cellular membranes and access the nucleus, where direct DNA interaction may result in damage. If NPs accumulate within a cell but do not readily gain access to the nucleus, direct DNA contact is possible during mitosis, when the nuclear membrane breaks down and gives rise to opportunity for DNA aberrations.

Occupational exposure of workers in the semiconductor, automotive, and aerospace industries is a primary concern, but NPs are also widely used in numerous cosmetics (such as lipstick, sunscreen, and antiaging creams), as well as medical sources (such as debris from dental prosthesis and orthopedic implants).^3–5 There is promise that NPs could be administered as diagnostic aids, drug carriers, and therapeutic treatments for patients.^6,7 With current and near-future exposure scenarios, workers in nanotechnology industries have the highest levels of chronic exposure to NPs, patients receiving NP-based treatments would have high-to-medium exposures over a limited duration, and consumers would likely have low, chronic exposures.

Among a variety of NPs, zinc oxide (ZnO) and silica NPs are in the most attractive positions for advanced nanotechnology industries and their potential applications, especially biomedical and pharmaceutical fields. They are also continuously directed for the advent of novel devices with multifunctionalities and multiple purposes, providing great benefits to human health. In particular, mesoporous ZnO and silica NPs have striking characteristics for application as drug carriers. They exhibit high surface area and porous interiors serving as reservoirs of drug molecules. The pore size and surrounding environment can influence the storage of various drugs of interest, whereas the size and shape of NPs can affect cellular uptake.⁸ For example, mesoporous silica has been successfully utilized for delivery of ibuprofen into the pores via hydrogen bond interaction between ibuprofen and the silanol functional groups in the pore wall.⁹ Moreover, highly mesoporous spherical three-dimensional ZnO nanoassemblies have been accomplished for loading doxorubicin hydrochloride as a model drug.¹⁰ Both silica and ZnO NPs are generally regarded as essentially nontoxic and nonirritant beyond oral and topical pharmaceutical applications. In clinical use, individuals who may be more vulnerable to NP toxicity due to their pre-existing medical conditions thus require the examination of genotoxic potential and the underlying mechanism of action.

Key characteristics of nanoparticles

Due to large surface-area-to-volume ratio, NPs exhibit distinct physicochemical (eg, optical, magnetic, and electrical) and catalytic properties, rendering higher numbers of atoms binding on particle surfaces than their bulk counterparts.^11,12 These characteristics of NPs promote their diffusion, reactivity, hardness, dimensionality, and suspension ability.

Usually, optical features of NPs are attributed to their ability to confine electrons to a very tiny size and to generate quantum effects. These optical absorption properties are related to their structure and shape. For instance, the yellow color of silver suspension in nanoform becomes a blue color in clustered form. Likewise, the color of gold NPs changes from blue to green to magenta, corresponding to their size and shape.

The suspension formation of NPs is also unique due to the great interaction force between their surface and suspension media, enabling density differences to be overcome.¹³ In contrast, interactions of bulk material often result in either sinking or floating in liquid media. In aqueous media, NPs are dispersed due to electrostatic and steric repulsion of their surface charge (positive/negative).¹⁴ Brownian motion and collision also have a crucial role in dispersion. As surface charges of NPs skew toward zero value, repulsive forces between NPs become decreased, eventually leading to their sedimentation by gravitational forces. The agglomeration process involves adhesion toward particles, mostly due to van der Waals forces resulting from their large surface-area-to-volume ratio at nanoscale (Figure 1).¹⁵ Due to agglomeration/aggregation, the physicochemical properties (eg, surface charge, size, size distribution, surface-area-to-volume ratio, surface reactivity) of NPs become altered, leading to mediation of their bioavailability and toxicities.^16,17

Figure 1 Various states of nanoparticles in different forms of dry powder and liquid in suspension media.

In addition, diffusion of NPs is unique because it regulates their behavior in the surrounding environment. Indeed, particle diffusion coefficient is negatively proportional to the particle diameter. The smaller the particle size, the higher the diffusional forces, presenting the behavior tendency of gas or vapor.^18,19 Thus, NPs with high diffusion coefficient display high mobility and consequently mix quickly in an aerosol. After their release in the environment, atmospheric diffusion rapidly promotes the migration of NPs, leading to them quickly traveling a long distance from the source and increasing detrimental health risks.²⁰

Additionally, other predominant properties of NPs are quantum confinement in semiconductors (eg, triple quantum dot silicon-based semiconductor, transition metal-doped ferromagnetic semiconducting silicon nanotubes, and ZnO semiconductor),^21–23 surface plasmon resonance in particular metal NPs (eg, doped silicon nanocrystals, ZnO, and copper NPs),^24,25 and superparamagnetism in magnetic materials (eg, multifunctional silica nanocomposites, gadolinium complexes, fluorophores, cell-penetrating peptides, and transition metal-doped ferromagnetic semiconducting silicon nanotubes).^22,26,27 For instance, ferroelectric materials (<10 nm in size) can switch their magnetization direction using room temperature thermal energy, rendering them inappropriate for memory storage. Copper NPs (>50 nm in size) are also regarded as superior durable materials that do not confer the same malleability and ductility as their bulk counterpart.²⁸

Regarding the unique features of NPs, their physicochemical properties should definitely be characterized prior to investigating the impact on human and environmental health. They include size, shape, structure, composition, purity, aggregation/agglomeration (size distribution), particle number, mass concentration, surface area, porosity, roughness, morphology, surface charge and chemistry, crystallinity, dispersity, and solubility. Measurement of primary particle size, hydrodynamic diameter, size distribution, zeta potential (surface charge), dispersity, concentration, and period of time in which agglomeration occurs provide better understanding of NP behavior relative to their cytotoxic and genotoxic responses.

Recent genotoxicological studies of metal oxide nanoparticles in in vitro and in vivo mammalian models

Zinc oxide nanoparticles

ZnO NPs are used in applications such as cosmetics, paints, drug carriers, and fillings in medical materials.²⁹ Also used as ultraviolet (UV) blocking materials, especially for UVA, their high catalytic activity in oxidation and photochemical reactions limits their use as UV blockers.³⁰ ZnO NPs are thought to be nontoxic and biocompatible.³¹ Exposure to ZnO NPs has been associated with inflammatory responses³² and cytotoxicity.^33–35 Little work regarding the genotoxic potential of ZnO NPs has been conducted. A previous investigation used Chinese hamster ovary cells to study chromosomal aberrations induced by ZnO NPs with a mean diameter of 100 nm promoted by UV light, finding increased clastogenicity under preirradiation and simultaneous irradiation conditions than in the dark.²⁹ This study indicates that ZnO NPs may cause photogenotoxicity, but a lack of information of the physicochemical properties makes the validity of the experiments questionable. Further assessment to fully investigate genotoxicity with a focus on size dependence and physiochemical features is required.

In vivo, the genotoxic potential of ZnO NPs has been investigated in animal models with systematic administration for 14 consecutive days,³⁶ as well as for 90 days.³⁷ The exposures increased liver enzyme and oxidative DNA breakage.

Recently, in vitro and in vivo, the mammalian toxicity of ZnO NPs as well as their toxicokinetics in various types of cells and animal models have been summarized (Table 1).^{34,35,38–41} In vitro, comet assays and the cytokinesis-blocked micronucleus present genotoxicity. Moreover, lung cells with in vitro exposure show cytotoxicity, increased oxidative stress, decreased mitochondrial membrane potential, and production of interleukin-8. Likely, the ZnO NPs are phagocytosed by macrophages and dissolved in lysosomes. In vivo, the ZnO NPs exhibit systemic distribution in target organs, including the liver, spleen, lung, kidney, and, in some cases, heart.

Table 1 Genotoxicity studies of ZnO and silica nanoparaticles using in vitro and in vivo mammalian models

Nanoparticles	Toxicological effect		Reference
ZnO	Genotoxicity in vivo	Using in vivo micronucleus test, no genotoxic effect was observed in lung cells from rats exposed to triethozycaprylylsilane-coated ZnO by inhalation	Landsiedel et al^Citation63
		50 nm ZnO had not induced micronucleus in the animal model at a concentration of up to 5 g/kg body weight	Li et al^Citation64
		60–200 nm ZnO did not induce genotoxicity in the in vivo system	Monteiro-Riviere et al^Citation65
	Genotoxicity in vitro	Genotoxic potential was observed in ZnO exposed cells by alkaline standard comet assay	Gopalan et al^Citation66
		Using comet assay, significant DNA damage was induced by 30 nm ZnO in a dose-dependent manner	Sharma et al^Citation38
		Induction of DNA damage was observed significantly in 10 nm and 20 nm ZnO exposure of Caco-2 cells with and without Fpg enzyme	Gerloff et al^Citation67
		Significant DNA damage was observed in 19.6±5.8 nm at 5 μg/mL and 10 μg/mL	Yang et al^Citation68
		Water-soluble ZnO nanoparticles have no mutagenic potential in Ames test	Yoshida et al^Citation69
		Diemthyoxydiphenylsilane/triethoxycaprylylsilane crosspolymer-coated ZnO was evaluated as nongenotoxic substance in Ames test	Landsiedel et al^Citation63
		Genotoxicity was observed by comet assay and micronucleus test in HEp-2 cells exposed to ZnO	Osman et al^Citation70
		Poly methyl acrylic acid coated ZnO induced significantly increased genotoxicity compared with uncoated ZnO measured by micronucleus test in WIL2-NS human lymphoblastoid cells	Yin et al^Citation71
		A significant increase in DNA damage was observed in 30 nm ZnO exposed cells	Sharma et al^Citation72
		30 nm ZnO nanoparticles induced DNA damage in in vitro system	Sharma et al^Citation73
		DNA damage measuring by comet assay was observed in human nasal mucosa exposed to ZnO repetitively	Hackenberg et al^Citation74
	Oxidative stress	Poly methyl acrylic acid-coated ZnO showed decreased cytotoxicity and ROS generation compared with uncoated ZnO in WIL2-NS human lymphoblastoid cells	Yin et al^Citation75
		ZnO induced mitochondrial dysfunction, morphological modification, and apoptosis in human fetal lung fibroblast	Zhang et al^Citation76
		ZnO led to cellular oxidant injury, inflammation, and cell death in in vitro system	Xia et al^Citation77
		Oxidative stress and cytotoxicity were induced by ZnO in human colon carcinoma cells	De Berardis et al^Citation78
		ZnO induced oxidative DNA damage and ROS-mediated apoptosis in human liver cells	Sharma et al^Citation79
		Induction of oxidative stress, DNA damage, and apoptosis were observed in a malignant human skin melanoma cell line exposed to ZnO	Alarifi et al^Citation80
		ZnO induced ROS-mediated cytotoxic effect in rat retinal ganglion cells	Guo et al^Citation81
Silica	Genotoxicity in vivo	No induction of hypoxanthine phosphoribosyltransferase-encoding gene (HPRT) mutation frequency was observed in rats exposed to silica for 13 weeks	Johnston et al^Citation82
		Inhalation of 37 nm and 83 nm SiO2 did not induce genotoxicity in rat lung	Sayes et al^Citation83
	Genotoxicity in vitro	A weak induction of micronuclei was observed in V79 cells at highly cytotoxic doses	Liu et al^Citation84
		No mutagenic potential was observed in Ames test with and without metabolic activation	ECETOC,^Citation85
		Also, no induction of chromosomal aberrations was observed in mammalian cells	EPA,^Citation86 OECD^Citation87
		No induction of genotoxicity was detected by comet assay in mouse fibroblasts exposed to silica	Barnes et al^Citation51
		A very slight DNA damage was observed in silica-exposed primary mouse embryo fibroblast cells by comet assay	Yang et al^Citation68,Citation88
		No significant induction of genotoxicity was observed in A549 cells exposed to amorphous silica particles for 40 hours	Gonzalez et al^Citation89
		Significant increase of micronuclei was induced in mouse fibroblast cells exposed to 80 nm silica nanoparticles	Park et al^Citation90,Citation91
	Oxidative stress	SiO2 induced cytotoxicity via production of oxidative stress in human embryonic kidney cells	Wang et al^Citation92
		P53 and Bax-mediated apoptosis was induced by SiO2 exposure in human hepatic cell line	Ye et al^Citation93
		P53 and p21-mediated G1 phase arrest was observed in myocardial cells	Ye et al^Citation94
		Endocytosis-dependent ROS generation and DNA damage was induced by nanosilica in human keratinocytes	Nabeshi et al^Citation62
		20 nm silica induced cytotoxic effects via induction of ROS and lipid peroxidation in kidney cells	Passagne et al^Citation61
		Nanosized silica induced developmental neurotoxicity via production of oxidative stress in PC12 cells	Wang et al^Citation59
		Hepatotoxicity was induced by SiO2 in Kupffer cells	Chen et al^Citation95
		SiO2 led to cutaneous toxicity via ROS generation	Park et al^Citation96

Abbreviations: ROS, reactive oxygen species; SiO₂, silicon dioxide; ZnO, zinc oxide.

The current review focuses on nanocosmetic ZnO sunscreens that have been thought not to be toxic, irritating, sensitizing, or photosensitizing after topical application.⁴² The toxicity issue has also been examined by the European Commission, where sunscreen preparations containing ZnO NPs were reviewed.⁴² Current studies have suggested that NPs do not exhibit increased penetration.^43,44 However, populations with unhealthy skin or wounds still need to be cautious about long-term topical use of nanoscreens.

Silica nanoparticles

Silica induces inflammatory response via nuclear factor kappa B activation and oxidative stress responses both in vivo and in vitro,^32,45,46 but cytotoxicity is observed only at high concentrations.^47,48 Silica NPs have been shown to enter the cell nucleus to potentially bind to the DNA phosphate backbone.⁴⁹ The silica NPs induce the reactive oxygen species (ROS),⁴⁶ especially the hydroxyl radical, a highly reactive molecule that may induce DNA strand breaks and oxidized bases.⁵⁰

Silica NPs have an impact on nuclear integrity, forming intranuclear protein aggregates and resulting in inhibition of replication, transcription, and cell proliferation.⁴⁹ Moreover, decreased replication activity as well as transcriptional activity were found for cells exposed to silica NPs. NPs of size >200 nm fail to penetrate the nucleus and do not alter nuclear structure and function or interfere with gene expression.⁴⁹ Nevertheless, there is limited evidence showing the genotoxic potential of silica NPs.^48,51 A micronucleus assay report shows that these NPs do induce chromosomal damage.⁵² To better indicate genotoxic potential, a battery of standardized tests quantifying different types of genetic aberrations are needed to cover all potential forms of inducible DNA damage as a result of exposure to NPs.

Recent work assessing systemic toxicological mechanisms of silica NPs in terms of cytotoxicity, genotoxicity, and phototoxicity has been reviewed and summarized (Table 1).^{34,35,53–57} Based on extensive physicochemical characterization, ecotoxicology, toxicology, safety, and epidemiology data, environmental and health risks seem to be unassociated with these particles if produced or utilized under current hygiene standards and recommendations. Silica NP interactions with membranes may induce the release of endosomal substance, ROS, cytokines, and chemokines, resulting in inflammatory responses.^58–62 Silica NP toxicity is likely linked to mechanisms of interaction with outer and inner membranes, signaling responses, and vesicle trafficking pathways, but human health and environmental risks and the mechanisms of toxicity are not fully elucidated.

Taken together, genotoxicological investigations of ZnO and silica NPs have been carried out using in vitro and in vivo mammalian models, as summarized in Table 1.

Key mechanisms underlying nanoparticle-induced DNA damage

If NPs are able to enter the body through inhalation, dermal, or oral routes, direct and indirect mechanisms exist to stimulate DNA damage.^34,35,97 NPs may be able to penetrate into the cell, and subsequently the nucleus, through a number of routes (Figure 2).¹ If NPs are located within the nucleus, direct interaction with DNA or DNA-associated proteins is possible. Indeed, silica NPs can enter the nucleus,^98,99 inducing intranuclear protein aggregates and resulting in inhibition of replication, transcription, and cell proliferation.^49,100,101 Quantum dots have also been shown to penetrate the nucleus via the nuclear pore complexes¹⁰² and interact with histone proteins.

Figure 2 Scheme illustrating possible routes of cellular uptake, including passive diffusion, receptor-related endocytosis, and clarthrin- or caveolae-dependent endocytosis. In brief, nanoparticles are in the correct size and shape. They may dock on membrane receptors, facilitating receptor-mediated endocytosis. Alternatively, clathrin- or caveolae-mediated endocytosis may occur, which results in the formation of pits in the region of 120 nm or up to 80 nm, respectively, which regulates the size of the material they are able to enclose.

Genotoxicity may arise through indirect mechanisms where NPs do not physically interact with the DNA molecule but with other cellular components, such as those involved in the cell division process. Other cellular responses may be induced and give rise to genotoxicity, such as oxidative stress induction, inflammatory response, and aberrant signaling responses (Figure 3).^1,35,97 Moreover, putative mechanisms underlying the detrimental effects of ZnO and silica NPs are proposed (Figure 4).

Figure 3 Key indirect mechanisms underlying nanogenotoxicity. Nanoparticles (NPs) may cause oxidative stress induction, inflammatory responses, or aberrant cellular signaling. These responses may be implicated in cancer risk.

Abbreviations: NPs, nanoparticles; ROS, reactive oxygen species.

Figure 4 Putative mechanisms underlying the detrimental effects of zinc oxide and silica nanoparticles. These nanoparticles dissolve in the extracellular milieu, giving rise to increased extracellular metallic cations. This leads to increased intracellular respective metallic cations, resulting in decreased activity of particular enzymes and transcription factors. Moreover, this event can induce ROS generation and resulting oxidative stress, as well as stimulate various cytokine production and inflammatory responses. These phenomena, in turn, render membrane damage, DNA breakage, mitochondrial dysfunction, and lysosome destabilization.

Abbreviations: ROS, reactive oxygen species; RNS, reactive nitrogen species; IL, interleukin; TNF, tumor necrosis factor; NF-κB, nuclear factor kappa-light-chain-enhancer of activated B cells; MAPK, mitogen-activated protein kinases; AP-1, activator protein 1.

Oxidative stress

Oxidative stress is a redox imbalance within cells as a consequence of increased intracellular ROS and decreased antioxidants. ROS-induced DNA damage is categorized by single- and double-stranded DNA breaks, base modifications (eg, formation of 8-hydroxydeoxyguanosine adducts), and DNA crosslinks, all of which may be implicated to initiate and promote carcinogenesis if unrepaired.^103,104

The transition metals ions (such as iron [Fe] and zinc) released from certain NPs are capable of converting cellular oxygen metabolic products (such as H₂O₂ and superoxide anions) to hydroxyl radicals and to DNA damaging species. Fe(II) can cause the production of H₂O₂ from molecular O₂, which can diffuse through the cellular and nuclear membranes to react with DNA-bound Fe and lead to radical production, crosslinking thymine–tyrosine (DNA-histone protein) chromatin.⁵⁰ Free Fe ions can cause OH-mediated purine and pyrimidine modifications.¹⁰⁵ As a result, Fe-containing NPs are a concern as a surplus source of Fe within the cells and fuelling the generation of highly reactive hydroxyl radicals via the Fenton reaction.

In addition to the elemental and ionic composition of NPs, the inherent high surface area can enhance the production of ROS. The smaller the NP, the higher the oxidative stress produced.^106–110 Researchers have reported that silica NP toxicity is dependent on size and may be associated with ROS generation.^54,61,111

Oxidative DNA damage, as determined by the comet assay and micronucleus assay,^108,112,113 has been demonstrated by several studies of the genotoxic effects of NPs. Oxidative stress activates specific signaling pathways, including mitogen-activated protein kinase and nuclear factor kappa B,¹¹⁴ together with interference of antioxidant defenses, resulting in release of proinflammatory cytokines.¹¹⁵ This signaling cascade is a key trigger of inflammation, a defensive reaction that leads to further ROS release from inflammatory cells (eg, neutrophils).^116,117

Inflammation

Inflammation is an important physiological process in response to tissue injury mediated by inflammatory cells secreting cytokines (eg, interleukins and tumor necrosis factor protein families), migration inhibition factors, RNS (reactive nitrogen species), and ROS. These factors are involved in protective defense against infection and/or tissue injury. They promote DNA damage in the form of point mutations, DNA adducts, and chromosomal fragmentation, as well as inhibit DNA repair and induce aberrant methylation patterns.^60,61 Expectedly, chronic inflammation has been associated with carcinogenesis.^118,119

At present, a number of studies have shown that NPs can exhibit inflammatory responses. Their small size and great surface area are involved in facilitating inflammation, as previous studies demonstrate that ultrafine NPs display higher inflammatory potency in the lungs of rats following intratracheal instillation.^120,121 The composition of the particle may be a determinant factor affecting the extent of the inflammatory response induced. For instance, in vitro, induced inflammation seems to follow the trend of silica and ZnO NPs.^111,122,123 Some NPs have oxidative DNA damaging potential via excessive formation of ROS and the release of metal ions, but also as a consequence of chronic inflammatory responses.

DNA damage responsive signaling

Exposure to NPs is associated with induction of oxidative stress, leading to damage to cellular components, most importantly DNA. Consequently, this damage can affect several cellular responses, including cell cycle arrest, apoptosis, and DNA repair. The DNA repair system is responsible for genetic stability and cell survival, and if repair fails to occur during or before replication of damaged DNA, mutagenic and possibly carcinogenic incidences may occur.

Once DNA is damaged, a key effector molecule, p53, is activated. Tumor suppressor gene p53 has been recognized as “the guardian of the genome” because of its essential role in arresting the cell cycle, activating transcription of genes that mediate DNA repair, and preventing the incidence of mutagenic conversion.¹²⁴ If DNA damage is extensively accumulated, p53 triggers apoptosis to eliminate the individual cell for the benefit of the organism. When these protective factors are compromised, stable heritable changes may undertake cellular transformation and, ultimately, carcinogenesis.

The pro-oxidative and proinflammatory properties have been found after exposure to various metal oxide NPs. ROS generation by ZnO NPs stimulates cellular processes: specifically, oxidant injury, inflammatory response, and cell death in different cell types such as mouse macrophages and human bronchial epithelial cells.⁷⁷ Moreover, ZnO NPs are thought to be linked to the incidence of metal fume fever. A previous study has indicated that pro-oxidant activity of ZnO NPs is attributable to particle dissolution.¹²⁵ Prevention of ZnO NP dissolution via Fe doping could decrease the pro-oxidative and proinflammatory effects of these particles.¹²⁵ Other NPs (silica, cationic polystyrene, and C60 fullerene) have been reported to exhibit pro-oxidative and proinflammatory properties in vitro and in vivo, including extensive accumulation of ROS, induction of oxidative stress, and stimulation of antioxidant and signaling pathways.^126–128

Indeed, several reports have demonstrated that exposure to asbestos fibers in unregulated workplaces risks pleural and lung fibrosis (asbestosis), lung cancer, and pleural and peritoneal malignant mesothelioma.^129–131 Asbestos may be regarded as a tumor promoter or cocarcinogen in the induction of lung cancers, particularly representing synergistic effects with chemical carcinogens in cigarette smoke.¹³² Asbestos fibers are naturally occurring in rocks and soils and comprise six different types: amphibole types (crocidolite, amosite, anthophyllite, tremolite, and actinolite), which are rod-shaped and have higher durability relative to the only serpentine type (chrysotile).¹³³ Much evidence has been presented that exposure to asbestos gives rise to a spectrum of asbestos-related diseases, including malignant pleural mesothelioma. This suggests an obvious relationship between the specific NP and disease, probably due to ROS accumulation and resulting oxidative stress.^134–137 However, there is no currently published report addressing a definite correlation between a disease outcome and exposure to a specific type of newly developed NP in humans. An active approach should be taken as a precaution. One effective strategy is to identify biomarkers associated with NP exposure.¹³⁸ Development of a panel of biomarkers as indicators of exposure-specific disease outcomes will require further time but would be well worth the effort for the identification of early biological responses related to current knowledge-based injury pathways.

Previous reports of metal fume fever, a flu-like illness with characteristics of self-limiting inflammation and oxidative stress response in the lung, indicate that it is caused by inhalation of highly concentrated metal oxide particles, particularly ZnO.^139–143 Given the increase in application of NPs and the uncertainty of their potential health impacts, health surveillance of workers frequently exposed to NPs in the occupational setting is important.

Using liquid chromatography–tandem mass spectrometry analysis, more than 30 proteins were thought to be responsible for incidental NP-induced oxidative stress. Some of these proteins may serve as markers for exposure to pro-oxidative substances.^144,145 Alterations of other particle-induced proteomes include modification of nitrotyrosine-based protein, activation of unfolding protein response, and incremental expression of ATF4, an endoplasmic reticulum stress-related transcription factor.^144,146 In animal studies, oxidative stress-altered proteomic profiles were found in the bronchoalveolar lavage fluid and lung tissue in mouse asthma models.^147,148 A more recent study demonstrates the expression of polymeric immunoglobulin receptor, complement C3, neutrophil gelatinase-associated lipocalin, chitinase 3-like protein 3, chitinase 3-like protein 4, and acidic mammalian chitinase in the lung to be associated with the adjuvant effect of ultrafine particles on the oxidant activity and the primary immune response (allergic sensitization).^148,149 Increased chitinase 3-like protein 3 expression is associated with the boosting of ambient ultrafine particles on the secondary immune response of inhalation exposure.¹⁵⁰ Furthermore, alteration of oxidative stress-associated proteome was observed in the bronchoalveolar lavage fluid from C57BL/6 mice exposed to ZnO NP via pharyngeal aspiration, indicating that proteomics may be used to identify biomarkers related to the exposure of certain NPs. Because oxidative stress and inflammatory response are also responsible for the toxicity of various NPs, the technology of proteomics has the potential to identify the biomarkers associated with NP exposure and the resulting deleterious effects of injury pathways.

The ideal biomarkers for evaluating environmental and occupational exposures should provide a mechanistic, molecular, and biological basis for the diseases and be exposure specific to reflect early adverse health effects, have clinical relevance, and be easy to use. Although identification of biomarkers that meet all these criteria is a challenge, it is feasible to study NP exposure-associated early biological events such as oxidative stress and inflammation.

Artifacts and limitations influencing nanogenotoxicity studies

Although little is known about the toxicokinetics of NPs, key factors gaining attention are the physicochemical properties influencing cellular uptake and subsequent physiological consequences. Parameters involved in genotoxicological responses are uncertain, and the evidence points to different factors participating in modulating molecular interactions.

Size, shape, and surface area

The nanometer size (<100 nm) of particles is considered a primary feature representing unique properties over bulk counterparts. Decreased size increases the number of particles per unit mass, but small size can also demonstrate a health hazard due to their interference with biological components once internalized. As a consequence, the size influences absorption, distribution, metabolism, and excretion kinetics, the driving force behind the development of new nanomedicines and nanodevices for clinical health care.^6,7 With regard to cellular uptake, size is a key factor in the different internalization mechanisms. Perhaps the most prominent mechanism is diffusion across the plasma membrane (either directly across the membrane or through membrane channels 10–30 nm wide), endocytosis, or energy-dependent mechanisms via a number of different routes (Figure 1).

In addition to cellular uptake, size does influence toxicological outcome, as reported by many studies focusing on ZnO and silica NPs versus microparticles. For instance, inhalation studies show that NPs penetrate deeper into the lungs and become localized within various cell types, indicating a greater inflammatory response that is markedly associated with potential toxicity in comparison with their fine-sized counterparts.^98,151–154 Although genotoxic potential was not examined in these investigations, the established relationship between chronic inflammation, DNA damage, and carcinogenesis may provide insight into the adverse health effects of long-term NP inhalation.

Shape or morphology is another matter related to negative cellular effects of NPs. Fewer studies focusing on the toxicological relationship associated with this parameter have been established, though a previous publication has demonstrated that removal of structural defects from a particular NP was sufficient to substantially reduce inflammatory response and overall toxicity.¹⁵⁵ Previous studies have revealed that shape of NPs strongly governs uptake rate. Spherical NPs exhibit higher uptake than nanorods, whereas internalization of cylindrical materials is strongly influenced by a high aspect ratio.^156,157

When the number of particles per mass unit increase, the overall surface area will also increase. The shape of NPs contributes to the overall surface area, such that spherical NPs have slightly smaller surface areas than an octagonal structure of the same size. This greater surface area promotes catalytic activity of the material, allowing an increase in its reactivity due to unsatisfied high energy bonds of surface atoms.¹²⁰ If NPs are able to gain access to the cellular milieu, the large surface area will give rise to more reactivity with biological components, resulting in unwanted cellular damage and oxidative stress.

Purity

Purity of NPs is a concern, as contamination of residual metal may cause stronger (geno) toxicological responses than the actual nanomaterial. Most metal catalysts are removed by postproduction processes. Purified NPs may contain up to 15% metal residual bymass. Efforts to purify NPs are under way to limit effects of impurities on toxicity.

Numerous studies undertake the synthesis of lead-rich carbon nanotubes for use as X-ray protection shields. However, Fe is one of the primary sources of damage via oxidative stress, resulting in Fenton or Haber–Weiss reactions.^158–161 Indeed, Fe contaminants on carbon nanotubes have reportedly been shown to cause a substantial loss of glutathione and increased lipid peroxidation in alveolar macrophages, indicators of oxidative stress.¹⁶² Conversely, a previous study found that single-walled carbon nanotubes induced dose-dependent lung lesions (granulomas) in mice, irregardless of purity.¹⁶³ Similarly, nickel and yttrium catalyst impurities entrapped within single-walled carbon nanotubes and multiwalled carbon nanotubes do not seem to be responsible for potential toxicity associated with these materials.¹⁶⁴ In general, in vitro studies utilizing metal chelators could provide more insight into the role of such impurities. However, the conflicting information may be attributed to the other physicochemical characteristics of NP used in the toxicological studies, and thus emphasizes the importance of full characterization and standardization of NPs.

Agglomeration (size distribution)

An inherent feature of several NPs is their hydrophobicity and tendency to agglomerate, especially under physiological conditions. Upon exposure to biological systems, most NPs will form aggregates rather than remain monodisperse. In genotoxicity testing, in vivo or in vitro dosing in an aqueous carrier or into an aqueous environment (with the exception of dust inhalation studies) is usually established, and exposure responses relate to the degree of agglomerated NP form.¹⁶⁵

Although NPs have a tendency to form larger aggregates, fibrous NPs represent a more complex situation. In addition to aggregation, the fibers may form a tangled structure depending upon rigidity, leading to a change in the dimensions and surface area of the original structure. Although fiber rigidity is dependent on the synthesis method, rigid NPs are attracted to one another by van der Waals forces, with a tendency to curve and twist, forming bundles.^166,167 Agglomerates are larger and often more rigid than individual NPs, leading to new causes of toxicity.¹⁶⁸

Different approaches are being undertaken to improve the hydrophilicity of NPs. Concerning dissolution of NPs in suspension media, aggregation/agglomeration is still an obstacle due to van der Waals forces and resulting adhesion toward particles at nanoscale. Hence, efforts on dispersal methods are established to improve their solubilities by using various dispersing agents of both inorganic and organic stabilizers (eg, fetal serum, organosulfur compounds, polyethyleneglycol, dextran, liposomes, micelles) or chemically modified functionalization of particle surface (eg, polymeric macromolecules).^{100,169–171}

Indeed, the aggregate states can be overcome by use of organic small molecules harboring multiple functional groups such as carboxyl (COOH), amine (NH₂), thiol (SH), phosphate, and sulfates. These stabilizers can be tailored for dispersibility into aqueous media or other biocompatible fluids.^172,173 The organosulfur compound 2,3-meso dimercaptosuccinic acid (DMSA) containing two carboxylic and two thiol groups has been widely applied as a dispersing stabilizer. Magnetic NPs have been stabilized with DMSA for tissue- and cell-targeted delivery of therapeutic drugs in the lung.¹⁷⁴ In particular, the mechanism of proinflammatory effects of magnetic NPs and DMSA has been examined. Also, the postfunctionalization of NPs using methoxy polyethyleneglycol (PEG) 2000 silane has been successfully developed for stabilizing free thiols onto the surface of metal oxide NPs under physiological pH.¹⁷⁵

Moreover, various polymer molecules have been employed for steric stabilization of oxide NPs in aqueous and high ionic strength media.^176–178 These polymeric stabilizers can affect the performance of nanomaterials, depending on the chemical nature of the polymer (ie, hydrophilicity/hydrophobicity, biocompatibility, and biodegradation), the molecular weight of the polymer, the conformation of the polymer, and the degree of particle surface. Other stabilizers, such as the amphiphilic molecules (eg, liposomes and micelles), have been successfully utilized to improve hydrophilicity of oxide NPs in biological or physiological media.^179,180

The features of solubility or dispersity usually influence cytotoxic and genotoxic impact; however, these issues should be treated with caution. Previous reports have demonstrated that surface modification-solubilized NPs can alleviate toxicity as a result of the functionalization.^181,182 Conversely, several studies have shown the surfactant-stabilized NPs to be more cytotoxic, and coating NPs with gold and silver apparently augments cytotoxicity in comparison with nondispersed or noncoated counterparts.¹⁸³

Extensive effort is necessary to refine the interplay between agglomeration, dispersal methods, and negative cellular effects. The degree of agglomeration under experimental conditions when conducting risk assessment may provide clearer interpretation of results.^184,185 Given that agglomerated structure may not be in the nanoscale, risks regarding exposure may be substantially decreased due to reduced cellular uptake or an inability to cross biological barriers. However, overlooking the agglomeration-related issues in many studies may be a primary cause for the lack of consistency and often conflicting reports.

Surface charge and chemistry

An understanding of surface properties of NPs is essential for providing insight into their behavior under different experimental conditions. Surface charge and chemistry will influence the production of agglomerates and resulting toxicities according to factors such as the pH or ionic strength in the aqueous environment.¹⁸⁶ Thus, aggregation/disaggregation kinetics may be useful to decipher toxicities that may occur during the course of an in vitro experiment or according to the specific biological compartment where the NPs may become concentrated.

Surface charge has an important role in regulating cellular uptake of NPs. The plasma membrane is negatively charged (due to the phospholipids on the outer surface), as well as the intracellular environment. Thus, anionic NPs may be endocytosed at a lower rate compared with cationic NPs. Although this has been observed in practice using PEGylated polylactide and hydrogel NPs of similar sizes but varying charges,^157,187 this effect does not preclude the uptake of negatively charged NPs.¹⁸⁸ However, cationic surface charges show greater cytotoxic responses as compared with those with anionic charges, although it is uncertain whether cell death is a direct consequence of surface charge or a result of increased uptake often associated with cationic NPs.¹⁸⁹ Furthermore, DNA is negatively charged. Thus, cationic NPs appear to interact more significantly with genetic material.

In addition to functionalizing the surface to promote their solubility, surface chemistry may be modified to attach biological components such as peptides for cell targeting or pharmaceuticals for drug delivery. These modifications will have toxicological impact toward the resultant NPs because cytotoxicity is strongly associated with coating the functional group.¹⁹⁰ Information on surface charge and chemistries is significant in elucidating uptake mechanisms and predicting biological interactions for the evaluation of toxicity.

Concluding remarks for nano(geno) toxicological studies

Currently, inadequate information on the genotoxic potential of NPs and the impact on persistent exposure to human health is a concern. Numerous literature reports focus on the cytotoxicity of NPs, specific aspects of physicochemical characteristics, association with other potential toxic effects, as well as consideration of DNA damage and cellular uptake, bioaccumulation, distribution, and retention. A growing body of studies point to particular NPs eliciting DNA damage potential, but inconclusive reports necessitate a discussion within the scientific community for clearer and more informative reports. Previous reviews have addressed potential considerations, including material characterization, use of standardized experimental methods, and association between in vitro and in vivo results,^191,192 but should account for the following.

1. In toxicological studies of any NPs, detailed physicochemical characterization under real experimental conditions is preferential. Measurements should include size distribution, morphology, surface area, charge, surface modifications, chemical composition, crystallinity, and agglomeration. Moreover, information on the fabrication process should be provided.

2. Appropriate controls (positives and negatives) and standards need to be established. Sources of metal ions or the use of metal chelators should be addressed in experimental design to elucidate whether biological effects are the consequence of NP interactions, impurities, or degradation products released during exposure.

3. If the functionalized form of NPs is being assessed, the unfunctionalized form should also be included to directly investigate the effect of the surface modification.

4. NP dynamics during the period of genotoxicity assay need to be a concern: eg, what about the degree of agglomeration and size distribution? In vitro, do NPs sediment or remain in suspension? Metrological techniques required to provide clearer information are currently not available and need to be developed.

5. Once internalized, the long-term fate of NPs needs to be considered. Physicochemical properties before and after experimentation should be detailed but may require the development of novel methodologies to form answers.

6. A battery of genotoxicity tests with varying end points should be utilized to provide insight into the mechanism of action and to ensure a comprehensive view of the reactivity of NPs.

7. In addition to somatic cells, the genotoxic potential of NPs on germ cells should be conducted.

8. Extension of in vitro experiments over 24 hours might be necessary if longer treatment times are more informative on the genotoxic potential of NPs.

Well-designed experiments are needed to enable a concerted effort to better exploit NP-mediated hazards and to define similarities enabling further extrapolations. The underlying mechanisms responsible for the exposure effect observed is also important in effectively informing the safer design of future NP systems, ensuring biocompatibility with minimum deleterious health risks. Biomarker studies provide valuable information identifying early biological events associated with adverse health effects of engineered nanomaterials before the manifestation of clinical outcomes, potentially helping health surveillance of workers at higher risk due to their occupational settings.

Disclosure

The authors report no conflicts of interest in this work.

Acknowledgments

This research was supported by a grant (10182MFDS991) from the Ministry of Food and Drug Safety in 2010–13 and ‘The Ecoinnovation Project’ (412-112-110) by the Korean Ministry of Environment. This research project is also supported by a grant from Mahidol University, Thailand, and a grant from the Faculty of Environment and Resource Studies, Mahidol University, Thailand.