Epidemiologic studies have shown causal relationships between air pollution particles and adverse health effects in susceptible subpopulations. Fly ash particles (containing water-soluble and insoluble metals) are a component of ambient air particulate pollution and may contribute to particulate-induced health effects. Some of the pathological effects after inhalation of the particles may be due to reactive oxygen species (ROS) produced by metal-catalyzed reactions. In this investigation, we analyzed emission source particulates oil fly ash (OFA) and coal fly ash (CFA) for metal content and solubility in relation to their ability to induce 2'-deoxyguanosine (dG) hydroxylation and DNA damage as measured by 8-oxo-dG formation by HPLC/UV-electrochemical detection (ECD). Water-soluble vanadium and nickel were present at the highest concentrations, and iron was present in trace amounts in OFA (5.1% V, 1.0% Ni, and 0.4% Fe by weight). In contrast, CFA comprised mostly of water-insoluble aluminosilicates and iron (9.2% Al, 12.2% Si, and 2.8% Fe by weight). As a first approach to gain insight into the mode of action of these particulates, we examined metal species-catalyzed kinetics of dG hydroxylation. Metal species at a concentration of 0.1 mM in the incubation mixture containing 0.1 mM dG under ambient air at room temperature catalyzed maximum 8-oxo-dG formation at 15 min with yields ranging from 0.05 to 0.17%, decreasing in the following order: vanadium(IV) > iron(II) > vanadium(V) > iron(III) > or = nickel(II). Insoluble Fe(III) oxide (Fe(2)O(3)) under similar conditions had no effect. Consistent with these results, OFA rich in vanadium and nickel concentrations showed a dose-dependent increase in the level of dG hydroxylation to 8-oxo-dG formation at particulate concentrations of 0.1-1 mg/mL (p < 0.05). In contrast, CFA with high concentrations of aluminosilicates and iron did not result in a significant increase in the level of 8-oxo-dG over that of the control, i.e., dG (p > 0.05). DMSO, a (*)OH scavenger, inhibited OFA-induced 8-oxo-dG formation, and metal ion chelators, deferoxamine (DFX), DTPA, and ferrozine blocked OFA-induced 8-oxo-dG formation. OFA and CFA induced 8-oxo-dG formation in a pattern similar to that observed for dG hydroxylation when calf thymus DNA was used as a substrate. Treatment of OFA particles with DFX before reacting with DNA or addition of a catalase in the incubation mixture significantly suppressed 8-oxo-dG formation (p < 0.05). These results suggest that metal availability, but not the concentration of metals present in CFA and OFA, is critical in mediating molecular oxygen-dependent dG hydroxylation and DNA base damage.