Toxicology of high aspect ratio nanomaterials based on the fibre pathogenicity paradigm structure-activity relationship of pathogenic fibres

Abstract

Carbon nanotubes (CNT) are a new form of industrially relevant nano-scale particle and are seen as the cutting edge of the burgeoning nanotechnology revolution which promises to impact on all our lives. Due to high length to diameter ratio, CNT are perhaps the most well known of a growing collection of high aspect ratio nanoparticles (HARN). However the production and use of carbon nanotubes has presented an interesting toxicological question based on its structure and raised the question ‘are carbon nanotubes like asbestos?’. Few people are unaware of the devastating global pandemic of diseases caused by asbestos and similarities in needle-like shape between asbestos and nanotubes have raised fears that nanotubes may mimic asbestos-type disease. The purpose of this study was to investigate this link, based on the wealth of information known about the toxic effects of certain forms of fibre on the respiratory system. From this we hope to identify those carbon nanotubes which are hazardous whilst not prejudicing the use of those industrially relevant materials which can be used safely. Within fibre toxicology there exists a central paradigm which outlines the main properties a fibrous particle must possess if it is to exert pathogenic effects in the body. This paradigm outlines the importance of length, thinness and biopersistence to a fibre and an absence of one or more of these attributes results in a loss of pathogenicity. We took this paradigm and, using suitable asbestos and non-asbestos controls, applied it various morphological forms of carbon nanotubes using an in vivo model. The resultant data demonstrates for the first time that asbestos-like pathogenic behaviour associated with carbon nanotubes is closely linked to the morphology of the nanotubes and their aggregates. Specifically our results showed that CNT which possessed a long, straight length were highly inflammogenic and fibrogenic within the peritoneal cavity of mice; a model sensitive to the pathogenic effects of fibres. As well as length, the importance of biopersistence in the pathogenesis of fibrous particles has been known for many years and is a central attribute affecting the pathogenicity of fibres. Amphibole asbestos is known to be durable, a commercially exploited attribute and as such is biopersistent in the lung which is a key feature of its pathogenicity. Glass fibre on the other hand is bio-soluble, and whilst long and inhalable, does not cause significant disease due to its lack of biopersistence. Based on the grapheme structure of CNT which impart exceptional strength and rigidity and the chemical inertness of carbon we would hypothesis that CNT would be biopersistent and therefore fulfil another of the criteria of the fibre pathogenicity paradigm (FPP). Our aim therefore has been to establish the durability of CNT against fibrous particles of known durability using a synthetic solution maintained at a pH to simulate the lung environment. Using a range of 4 CNT and using both durable and non-durable fibres such as amphibole asbestos and glass fibre to bench mark our result; we demonstrated that 3 of the 4 CNT tested displayed exceptional durability whilst the fourth lost approximately 30% of its mass during the experiment with concomitant reduction in pathogenicity. As well as length and biopersistence, the surface of a particle has been shown to contribute to the overall toxicity of a particle and in certain circumstances, such as that of quartz, the surface of the particle can be the biologically active component. In the case of carbon nanotubes, surface functionalisation is commonly used for various endpoints including the addition of various tags and labels for tracking. As such our further aim was to investigate the relationship between the length-dependent pathogenicity of a fibre sample and the surface of the fibre. By using different forms of functional groups attached to the surface of a pathogenic carbon nanotube we aim to critically test if the level of inflammation and fibrosis triggered in vivo can be altered by simple alteration of the surface. Our results showed that surface modification of CNT could alter the inflammogenic and fibrogenic effects of CNT which may have important implications when considering the hazard assessment of functionalised HARN. As CNT are not the only form of fibrous nanomaterial and within this project we also attempted to determine the applicability of the FPP to further high aspect ratio nanomaterials. In order to do this we set out to determine the generality of this hypothesis by asking whether nickel nanowires, a radically different form of HARN to CNT, show length-dependent pathogenicity. Nickel oxide nanowires synthesised to be predominantly long (>20 μm) act similarly to amphibole asbestos in showing the ability to elicit strong inflammation in the mouse peritoneal model in a dose dependent manner; inflammation was not seen with the short (<5 μm) nanowires. In summation, the results from this study are the first to show that long HARN can indeed behave like asbestos when in contact with the sensitive mesothelium. This study suggests a potential link between inhalation exposure to long nanotubes and asbestos-related disease, especially mesothelioma and as such this may have immediate implications across many disciplines if care is to be taken to avoid a long term legacy of harm.