Abstract:

Although most experimental data suggest that exposure to nanomaterials may be dangerous, the matter of their toxicity in humans is still unresolved. A key question is whether they may be more toxic than their bulk counterpart. To clarify this issue, we perform an in-depth analysis of size-related toxicological properties and discuss the pathogenetic pathways activated by oxidative and non-oxidative stress. In addition, new insights on the complex molecular inter-relationships arising from ‘omics’ are examined, in light of the information they can provide on specific intracellular events elicited by nanomaterials. Finally, we describe the mechanism of action of fibre-like nanomaterials.

Key words

nanomaterials; toxicity; oxidative stress; physical interference; omics

2.1 Introduction

All substances, including water, may be toxic to humans depending on the amount and the circumstances of exposure (Farrell and Bower, 2003). Thus, it is neither alarming nor surprising that available in vitro and in vivo experiments, often performed at very high doses, show adverse effects of nanomaterials in several biological systems (Pietroiusti, 2012; Schrand et al., 2012). Reliable epidemiological data in humans are currently lacking. There are case reports in the literature highlighting the harmful consequences of accidental exposure to very high amounts of nanomaterials in occupational settings (Song et al., 2009; Toyama et al., 2009; Zeig-Owens et al., 2011). At the reported doses, however, adverse effects would have been expected also after exposure to the same material in its bulk form. This means that we do not know currently if nanomaterials pose a real hazard to humans after exposure to doses that are considered safe for the bulk material. Reasons for concern, however, are justified, because enhanced toxicity of nanomaterials may be predicted on the bases of size-related characteristics.

2.2 Size- and non-size-related toxicity mechanisms of nanomaterials

The main concern for potential nanomaterial toxicity is their small size. Although a critical threshold number of nanomaterials is required for the induction of any toxicological effect, some size-related toxicological properties are better understood by considering the behaviour of a single nanomaterial, whereas for others it is better to consider the behaviour of several nanomaterials. If we consider the journey of a single nanomaterial inside the human body, it appears as a sort of fantastic voyage compared with that of the bulk form. It may in fact reach biological sanctuaries such as the intracellular environment and neural axons (Elder et al., 2006), usually interdicted to the bulk material. Furthermore, it may cross biological barriers such as the blood–brain barrier and the blood-testis barrier and may therefore reach highly protected organs such as the brain and the testes (Pietroiusti et al., 2012); in pregnant women, it may translocate through the placenta and come in contact with fetal tissues, as recently reviewed by Campagnolo et al. (2012). Furthermore, it has been reported that the removal of nanomaterials from tissues is slower than that of bulk material, because nanomaterials are poorly taken over by macrophages (Kreyling et al., 2012). Thus, for a given level of toxicity, nanomaterials may cause injuy to biological structures spared by the bulk form and may reside longer in these compartments.

A single nanomaterial also has greater surface reactivity compared with the bulk form. This increased reactivity is related to the higher number of atoms at the surface, as a consequence of the increased surface-to-volume ratio (see below). It has been calculated that a particle of 300 nm in size has only 5% of the atoms at the surface, whereas for a particle of 30 nm, more than 50% of the atoms are at the surface. These atoms are bonded differently from those in the inner part of the particle, and, if present in large amounts, may confer to the nanomaterial properties that may be relevant for toxicology (Buzea et al., 2007). For example, electrons present in these atoms may be available for transfer to oxygen, leading to the formation of reactive oxygen species (ROS) and ultimately to oxidative stress and cellular damage, as will be discussed later (Nel et al., 2006).

Another size-related characteristic of single nanomaterials, relevant for toxicological effects, is the strong tendency to be coated with molecules present in biological fluids (mainly proteins in the blood, and lipids in the pulmonary environment) once the nanomaterial has gained access to the body. The formation of this nano/bio interface, or ‘corona’, is a complex dynamic process that affects biodistribution and cellular uptake (Mahon et al., 2012) and therefore indirectly modulates the toxicity. However, a protein corona may also be directly responsible for the toxic effect elicited by nanomaterials, such as the activation of complement and of the coagulation cascade, as reported by Simberg et al. (2009). Thus, the formation of a protein/lipid corona may represent a further element of increased toxicity.

Other properties of nanomaterials potentially relevant for toxicity include their very high surface area and the tendency to aggregation/agglomeration. Particles of small size have a highly increased surface area as a consequence of their high surface-to-volume ratio compared with larger particles. To understand the relevance of this concept, consider the sequential cutting of a cube. If we cut a cube into two pieces of equal dimensions, we will have two smaller cubes with a combined volume identical to that of the original cube, but with a larger surface area, because two additional surfaces have been exposed as a consequence of the cut. If we repeat this process until the edge of the smallest resulting cubes is less than 100 nm (i.e. until we obtain cubes of nanometre size), the combined volumes of these cubes will still be the same as that of the original cube, but the surface area will be enormously increased, because of the very large number of new surfaces exposed following each cut. From a biological perspective, this means that the contact area with biological structures (e.g. with pulmonary epithelium) will be much higher for nanomaterials compared with the same volume of bulk material. The clinical consequences of this concept are important. For example, a material with burning properties that came into contact with the skin would cause more extensive damage if made of nanoparticles. A burn to a small area of the skin generally does not have severe consequences; however, if it involves most of the body’s surface, the outcome could be death.

There is a strong tendency to aggregation/agglomeration in the presence of a large number of nanomaterials (Hotze et al., 2010); however, it is uncertain if this property confers specific toxicological attributes. The aggregation status probably influences cellular uptake in a specific manner according to different cell types (Albanese and Chang, 2011) and may therefore indirectly drive the target of nanomaterial toxicity.

Finally, nanomaterials share with the corresponding bulk material some properties that contribute to their overall toxicity, such as shape, biopersistence, surface charge, and loss of metal ions. Shape is very important, since some nanomaterials, such as carbon nanotubes, may have a fibre-like aspect, and this particular shape may be responsible for injury to biological systems via several mechanisms, which will be examined below.

The role of charge is less clear: in general, it seems that in biological systems, positively charged nanomaterials may be more dangerous than other nanomaterials, likely because of their ability to react with the negatively charged cellular membranes (Bhattacharjee et al., 2010). This observation, however, has not always been experimentally confirmed (Hoshino et al., 2004); in addition, there are some reports showing a more marked reactivity of negatively charged nanomaterials. These contrasting results may be dependent on other factors, which, in turn, may be related to the surface charge (e.g. changes in dispersibility or in composition of a protein corona), and which modulate nanomaterial toxicity, leading to a net effect that may be different from that expected of the surface charge per se.

Metal ions are sometimes the principal cause of nanomaterial toxicity. It has been shown that metal nanomaterials might release ions into the biological compartments, and this has been demonstrated not only for ions composed of the nanoparticles themselves (Pettibone et al., 2008) but also for those included in the nanomaterials during their synthesis (Pulskamp et al., 2007). Finally, biopersistence, defined as the unaltered long-term presence of a xenobiotic in a living organism, can be intuitively correlated to toxicity, since any given toxic effect becomes more severe if the exposure to the offending agent occurs over a long period. Carbon nanotubes and titanium dioxide are examples of nanomaterials with high biopersistence.

In summary, biological systems appears to be more affected by exposure to nanomaterials than to the corresponding bulk form for...