3.6 Human Health Effects of Nanomaterials
There is a significant gap in our knowledge of the environmental, health, and ecological implications associated with nanotechnology (Dreher, 2004; Swiss Report, 2004; UK Royal Society, 2004; European NanoSafe, 2004; UK Health and Safety Executive, 2004). This section provides an overview of currently available information on the toxicity of nanoparticles; much of the information is for natural or incidentally formed nanosized materials, and is presented to aid in the understanding of intentionally produced nanomaterials.
3.6.1 Adequacy of Current Toxicological Database
The Agency’s databases on the health effects of particulate matter (PM), asbestos, silica, or other toxicological databases of similar or larger sized particles of identical chemical composition (U.S. EPA, 1986, 1996, 2004) should be evaluated for their potential use in conducting toxicological assessments of intentionally produced nanomaterials. The toxicology chapter of the recent Air Quality Criteria for Particulate Matter document cites hundreds of references describing the health effects of ambient air particulate matter including ultrafine ambient air (PM0.1), silica, carbon, and titanium dioxide particles (U.S. EPA, 2004). However, it is important to note that ambient air ultrafine particles are distinct from intentionally produced nanomaterials since they are not purposely engineered and represent a physicochemical and dynamic complex mixture of particles derived from a variety of natural and combustion sources. In addition, only approximately five percent of the references cited in the current Air Quality Criteria for Particulate Matter document describe the toxicity of chemically defined ultrafine particles, recently reviewed by Oberdörster et al. (2005a) and Donaldson et al. (2006).
A search of the literature on particle toxicity studies published up to 2005 confirms the paucity of data describing the toxicity of chemically defined ultrafine particles and, to an even greater extent, that of intentionally produced nanomaterials (Figure 20). The ability to assess the toxicity of intentionally produced carbon nanotubes by extrapolating from the current carbon-particle toxicological database was examined by Lam et al. (2004) and Warheit et al. (2004). Their findings demonstrate that graphite is not an appropriate safety reference standard for carbon nanotubes, since carbon nanotubes displayed very different mass-based dose-response relationships and lung histopathology when directly compared with graphite……
3.6.2 Toxicity and Hazard Identification of Engineered/Manufactured Nanomaterials
Studies assessing the role of particle size on toxicity have generally found that ultrafine or nanosize range (<100 nm) particles are more toxic on a mass-based exposure metric when compared to larger particles of identical chemical composition (Oberdörster et al., 1994; Li et al., 1999; Höhr et al., 2002). Other studies have shown that particle surface area dose is a better predictor of the toxic and pathologic responses to inhaled particles than is particle mass dose (Oberdörster et al., 1992; Driscoll, 1996; Lison et al., 1997; Donaldson et al., 1998; Tran et al., 2000; Brown et al., 2001; Duffin et al., 2002). Studies examining the pulmonary toxicity of carbon nanotubes have provided evidence that intentionally produced nanomaterials can display unique toxicity that cannot be explained by differences in particle size alone (Lam et al., 2004; Warheit et al., 2004). For example, Lam reported single walled carbon nanotubes displayed greater pulmonary toxicity than carbon black nanoparticles. Similar results have been obtained from comparative in vitro cytotoxicity studies (Jia et al., 2005). Muller et al. (2005) reported multi-walled carbon nanotubes to be more proinflammatory and profibrogenic when compared to ultrafine carbon black particles on an equivalent mass dose metric. Shvedova et al. (2005) reported unusual inflammatory and fibrogenic pulmonary responses to specific nanomaterials, suggesting that they may injure the lung by new mechanisms. Exposure of human epidermal keratinocyte cells in culture to single-walled carbon nanotubes was reported to cause dermal toxicity, including oxidative stress and loss of cell viability (Shvedova et al., 2003). The combination of small particle size, large surface area, and ability to generate reactive oxygen species have been suggested as key factors in induction of lung injury following exposure to some incidentally produced nanomaterials (Nel et al., 2006).
Contrary to other reports, Uchino et al. (2002), Warheit et al. (2006) and Sayes et al. (2006) have reported nanoscale titanium dioxide toxicity was not found to be dependent on particle size and surface area. These authors reported that specific crystal structure and the ability to generate reactive oxygen species are important factors to consider in evaluating nanomaterial toxicity. Similar to other reports, Warheit demonstrated that nanomaterial coating impacted toxicity (Warheit et al., 2005).
Studies have demonstrated that nanoparticle toxicity is extremely complex and multifactorial, potentially being regulated by a variety of physicochemical properties such as size and shape, as well as surface properties such as charge, area, and reactivity (Sayes et al., 2004; Cai et al., 1992; Sclafani and Herrmann, 1996; Nemmar et al., 2003; Derfus et al., 2004). The properties of carbon nanotubes in relation to pulmonary toxicology have recently been reviewed (Donaldson et al., 2006).
Toxicological assessment of intentionally produced nanomaterials will require information on the route (inhalation, oral, dermal) that carries the greatest risk for exposure to these materials, as well as comprehensive physicochemical characterization of them in order to provide information on size, shape, as well as surface properties such as charge, area, and reactivity. Establishment of dose-response relationships linking physicochemical properties of intentionally produced nanomaterials to their toxicities will identify the appropriate exposure metrics that best correlate with adverse health effects.
One of the most striking findings regarding particle health effects is the ability of particles to generate local toxic effects at the site of initial deposition as well as very significant systemic toxic responses (U.S. EPA, 2004). Pulmonary deposition of polystyrene nanoparticles was found to not only elicit pulmonary inflammation but also to induce vascular thrombosis (Nemmar et al., 2003). Pulmonary deposition of carbon black nanoparticles was found to decrease heart rate variability in rats and prolonged cardiac repolarization in young healthy individuals in recent toxicological and clinical studies (Harder et al., 2005; Frampton et al., 2004). Extrapulmonary translocation following pulmonary deposition of carbon black nanoparticles was reported by Oberdörster et al. (2004a, 2005a) Submicron particles have been shown to penetrate the stratum corneum of human skin following dermal application, suggesting a potential route by which the immune system may be affected by dermal exposure to nanoparticles (Tinkle et al., 2003; Ryman-Rasmussen et al., 2006). Zhao et al. (2005) have reported that in molecular dynamic computer simulations C 60 fullerenes bind to double and single-stranded DNA and note that these simulations suggest that C60 may negatively impact the structure, stability, and biological functions of DNA. It is clear that toxicological assessment of intentionally produced nanomaterials will require consideration of both local and systemic toxic responses (e.g., immune, cardiovascular, neurological toxicities) in order to ensure that that we identify the health effects of concern from these materials.