Assessment of evidence for nanosized titanium dioxide-generated DNA strand breaks and oxidatively damaged DNA in cells and animal models

<p>Nanosized titanium dioxide (TiO<sub>2</sub>) has been investigated in numerous studies on genotoxicity, including comet assay endpoints and oxidatively damaged DNA in cell cultures and animal models. The results have been surprisingly mixed, which might be attributed to physico-chemical differences of the tested TiO<sub>2</sub>. In the present review, we assess the role of certain methodological issues and publication bias. The analysis shows that studies on DNA strand breaks without proper assay controls or very low intra-group variation tend to show statistically significant effects. Levels of oxidatively damaged DNA, measured by the enzyme-modified comet assay, tend to show no effect in studies that have not included proper assay controls or they have uncertainty about the measurement. In addition, there are indications of publication and reporting bias. Nevertheless, the analysis shows that Aeroxide P25 generates DNA strand breaks in a concentration-dependent manner, which is not dependent on the duration of exposure. The standard comet assay seems to be able to discriminate between the genotoxicity of different types of TiO<sub>2</sub>, where anatase TiO<sub>2</sub> seems to be the form with strongest genotoxic potential. Cell culture studies also demonstrate increased levels of oxidatively damaged DNA after exposure to TiO<sub>2</sub>. There are relatively few studies on animal models where DNA strand breaks and oxidatively damaged DNA have been tested with reliable methods. Collectively, this review shows that exposure to nanosized TiO<sub>2</sub> is associated with genotoxicity in cells, whereas there are still too few reliable studies to assess the genotoxic potential in animal models.</p>