Toxicological Effects of Metal Oxide Nanomaterials - Safety of Nanomaterials along Their Lifecycle - page 164

organic surface modification did not trigger ROS formation in 10 different cell lines
tested, whereas nonmodified TiO 2 nanoparticles of similar size and surface area and
an even lower anatase content induced a concentration-dependent increase in ROS
in all cell lines (Kroll et al. 2011).
These findings suggest that particle coatings can have a strong impact on the
biological effects of nanoparticles. In line with that notion, Pan et al. reported that
intracellular ROS formation was reduced to control levels when rutile TiO 2 particles
were coated with a dense polymer brush (Pan et al. 2009).
In a comprehensive study conducted to identify physicochemical properties that
modulate nanoparticle toxicity, four nearly identical CeO 2 nanoparticles (A-D) from
the same production process with only slight variations were generated and assessed
for their ROS generating potential (Kroll et al. 2011). Whereas three of the four
CeO 2 nanoparticle dispersions (A-C) significantly induced ROS in the majority of
the cell lines tested, one CeO 2 nanoparticle (D) failed to evoke a significant increase
in ROS in any of the cell lines tested. CeO 2 -D nanoparticles were almost identical
to CeO 2 -A nanoparticles regarding size and surface area but displayed slight differ-
ences in the surface chemistry due to some variation in gassing with carbon dioxide
during the production process. These differences in surface chemistry may lead to
direct or indirect effects. Among the latter changes in agglomeration and gravita-
tional settling of particles should be considered (see Chapter 7, Schippritt et al.). In
addition, a different adsorption of serum proteins from the cell culture medium may
be involved. It has been speculated that the composition of the nanoparticle protein
corona influences the cytotoxicity of nanoparticles (Lundqvist et al. 2008). Indeed,
it was demonstrated that the absorption of CeO 2 -A and CeO 2 -D nanoparticles to
bovine serum albumin differed significantly (Schaefer et al. 2012). Thus, a disparity
of the protein composition adsorbed by CeO 2 -A and CeO 2 -D nanoparticles may be
responsible for their different effects on cells.
In the nanoGEM project, cellular ROS formation in response to metal oxide
nanomaterial exposure was tested based on the in vitro test matrix using the DCF
assay. An increase in intracellular ROS levels could not be observed directly after
exposure to any of the naked or surface-modified SiO 2 or ZrO 2 nanomaterials in
any of the cell lines tested (see Table 8.1). In contrast, P25 TiO 2 , which was used
as a reference nanomaterial at a concentration of 32 µg/mL, was shown to increase
intracellular ROS levels in most cell lines of the test matrix with the exception of
HaCaT (nanoGEM, unpublished results). In nanoGEM, we futhermore established
an indirect measure of oxidative stress through the detection of oxidative protein
modifications (i.e., protein carbonylation). This assay proved to be highly sensi-
tive and was well suited to analyze the potential of nanoparticles for triggering
oxidative stress. Protein carbonyls are detected via coupling with dinitrophenyl
(DNP) hydrazine. The resulting hydrazone can be detected in a 1D or 2D western
blot using an anti-DNP antibody (Robinson et al. 1999). In some studies such an
approach was already successfully used (Haase et al. 2011; Haase et al. 2012; Sun
et al. 2013). Here, we applied this assay for assessing the oxidative stress poten-
tial of all nanoGEM nanomaterials in NRK-52E cells. The results showed that 8
out of 16 nanoGEM nanoparticles induced protein carbonylation (Driessen et al.,
unpublished).