The in-vitro genotoxicity of nanosized TiO(2) rutile and anatase was assessed in comparison with fine TiO(2) rutile in human bronchial epithelial BEAS 2B cells using the single-cell gel electrophoresis (comet) assay and the cytokinesis-block micronucleus test. BEAS 2B cells were exposed to eight doses (1-100 microg/cm(2)) of titanium(IV) oxide nanosized rutile (>95%, <5% amorphous SiO(2) coating; 10 x 40 nm), nanosized anatase (99.7%; <25 nm), or fine rutile (99.9%; <5 microm) for 24, 48, and 72 h. Fine rutile reduced cell viability at lower doses than nanosized anatase, which was more cytotoxic than nanosized rutile. In the comet assay, nanosized anatase and fine rutile induced DNA damage at several doses with all treatment times. Dose-dependent effects were seen after the 48- and 72-h treatments with nanosized anatase and after the 24-, 48- (in one out of two experiments), and 72-h treatments (one experiment) with fine rutile. The lowest doses inducing DNA damage were 1 microg/cm(2) for fine rutile and 10 microg/cm( 2) for nanosized anatase. Nanosized rutile showed a significant induction in DNA damage only at 80 microg/cm(2) in the 24-h treatment and at 80 and 100 microg/ cm(2) in the 72-h treatment (with a dose-dependent effect). Only nanosized anatase could elevate the frequency of micronucleated BEAS 2B cells, producing a significant increase at 10 and 60 microg/cm( 2) after the 72-h treatment (no dose-dependency). At increasing doses of all the particles, MN analysis became difficult due to the presence of TiO(2) on the microscopic slides. In conclusion, our studies in human bronchial epithelial BEAS 2B cells showed that uncoated nanosized anatase TiO(2) and fine rutile TiO(2) are more efficient than SiO( 2)-coated nanosized rutile TiO(2) in inducing DNA damage, whereas only nanosized anatase is able to slightly induce micronuclei.