Human Toxicology and Effects of Nanoparticles - Environmental and Human Health Impacts of Nanotechnology - page 367

suggesting iron content is important in determining redox dependent effects in
macrophages. The bioavailability of iron in SWCNT was investigated by Guo et al.
(2007). These authors were able to mobilise redox active iron from a diverse array
of commercially available nanotubes. The amount of iron that could be mobilised
could not be predicted from the total amount of iron present in the sample, sug-
gesting that they differ in terms of the amount trapped within the nanotube
structure.
With particulate air pollution there is a clear link between the respiratory expo-
sure and subsequent cardiovascular effects, although the mechanism remains
unclear (Donaldson et al. , 2001c ; Mills et al. , 2005 ). Li et al. (2007) exposed mice to
SWCNT (10 and 40 g/mouse) and found induction of oxidative markers in the

lung, aorta and heart tissue. In addition, the SWCNT accelerated plaque formation
in ApoE - / - mice fed an atherogenic diet. This requires further investigation in order
to prevent increased risk of cardiovascular disease as a result of occupational expo-
sure and use of nanotubes.

µ

(ii) Dermal effects of carbon nanotubes
The studies conducted to investigate the potential dermal toxicity of nanotubes have
been conducted using either in vitro cell line models, or implantation into the skin
of rodents. The main reason for conducting such studies is due to the potential for
graphite and other carbon materials (e.g. carbon fi bres) to induce dermatitis.
Shvedova et al. (2003) found that treatment of a human keratinocyte cell line with
SWCNT (30% iron) (0.6 mg/ml for 18 hours) led to oxidative stress and cytotoxicity.
Monteiro - Riviere et al. (2005) exposed human epidermal keratinocytes to MWCNT
for up to 48 hours at relatively high concentrations of 100, 200 and 400 g/ml. The

MWCNT were observed by transmission electron microscopy (TEM) within cyto-
plasmic vacuoles as early as one hour after exposure. All concentrations resulted in
IL8 (interleukin 8) production by eight hours and signifi cant cytotoxicity at 24 hours.
In animal studies, Koyama et al. (2006) implanted both SWCNT and MWCNT
into the subcutaneous tissue of mice for up to three months. The authors noted that
the CNT induced toxicity, but that this subcutaneous toxicity was relatively low in
comparison to asbestos. The toxicity took the form of activation of antigen/antibody
response systems, in that after one week the SWCNT induced activation of major
histocompatibility complex (MHC) class 1 of CD4+/DC8+ T-cells. After two weeks
all CNT samples activated MHC class II markers in CD4+/DC8+ cells and in CD4+
cells. Granulomatous tissue encapsulating nanotube agglomerates was observed in
both the SWCNT and MWCNT treated animals. A similar study was conducted by
Yokoyama et al. (2005) using hat - stacked carbon nanofi bres instead of carbon
nanotubes. In this study the implantation was into rats; infl ammatory response
generated was described by the authors as mild, in that it did not generate wide-
spread necrosis, but it did induce the appearance of granuloma like structures.

µ

Paradigm II: CNT and the fi bre paradigm
Our own group has found that the physical structure of MWCNT greatly infl uences
their interaction with macrophages. Long (50 m) rigid MWCNT formed fi bre - like

aggregates that were too large to be taken up by macrophages in vitro , resulting in
ROS production, frustrated phagocytosis and TNF

µ

α production. While entangled