In the last years, a number of in vitro studies have been performed to assess the genotoxic activity of titanium dioxide (TiO ). To resolve the contradictory results, in this study, we investigated the genotoxic activity of commercial TiO nanoparticles (NPs) and microparticles of different forms (anatase, rutile and mix of both). We evaluated micronucleus formation in stimulated lymphocytes, as well as DNA strand breaks and 8-oxo-7,8-dihydro-2'-deoxyguanosine in peripheral blood mononuclear cells (PBMCs), a mixed population of lymphocytes and monocytes. Different responses to TiO exposure were obtained depending on the assay. Both TiO NPs and microparticles and all the crystalline forms elicited a significant increase in 8-oxo-7,8-dihydro-2'-deoxyguanosine and DNA strand breaks in the whole PBMC population, without a concurrent increase of micronuclei in proliferating lymphocytes. The distribution of DNA damage in PBMCs, detected by the comet assay, that measures DNA damage at level of single cells, indicated the presence of a more susceptible cell subpopulation. The measurement of side scatter signals by flow cytometry highlighted the preferential physical interaction of TiO particles with monocytes that also displayed higher reactive oxygen species generation, providing a mechanistic explanation for the different responses observed in genotoxicity assays with PBMCs and lymphocytes. This study confirmed the suitability of human PBMCs as multi-cell model to investigate NP-induced DNA damage, but suggested some caution in the use of stimulated lymphocytes for the assessment of NP clastogenicity.