Air pollution is a major environmental health risk and it contributes to respiratory and cardiovascular diseases and excess mortality worldwide. The adverse health effects have been associated with the inhalation of fine particulate matter (PM2.5) and induction of respiratory oxidative stress. In this work, we quantified the oxidative potential (OP) of PM2.5 from several Canadian cities (Toronto, Hamilton, Montreal, Vancouver) using a recently developed bioanalytical method which measures the oxidation of lung antioxidants, glutathione, cysteine, and ascorbic acid, the formation of glutathione disulfide and cystine, and the related redox potential (RP) in a simulated epithelial lining fluid (SELF). We evaluated the application of empirical SELF RP as a new metric for aerosol OP. We further investigated how PM2.5 chemical composition and OP are related across various emission source sectors and whether these features are linked to specific properties of aerosol aqueous phase, such as pH and metal-ligand complexation. The OP indicators including SELF RP were strongly correlated among each other, indicating that the empirical RP could be used as a reliable metric in future studies. OP based on ascorbic acid showed dependency on the emission source sectors, most likely due to variation in the solubility of Fe. Traffic emissions resulted in the highest OP, followed by industrial emissions and resuspended crustal matter. OP presented low correlation with PM2.5 concentrations, low-moderate correlation with the aerosol organic matter, and moderate-strong association with black carbon and transition metals across the sites. We did not find strong association between the concentration of biomass burning tracers and OP. Copper was the only metal that showed high association with OP across all sites, whereas the correlation with other metals, such as iron, manganese, and titanium, showed clear dependency on the source sectors. The aerosol pH correlated negatively with ambient temperature and positively with biomass burning tracers and the levels of nitrate, ammonium, and aerosol liquid water content. The solubility of Fe was associated with sulfate and aerosol pH at most sites, suggesting the involvement of proton-mediated dissolution pathway, while this was not visible at the site influenced by industrial emission, most likely due to the abundance of pyrogenic Fe. The effect of metal-ligand complexation on the solubility of transition metals, in particular Fe, was clearly observed at all sites, whereas a combined effect with aerosol pH, and a subsequent impact on OP, was only seen at the traffic site in Toronto. The enhanced solubility of Fe due to proton- and ligand-mediated dissolution pathways and subsequent formation of reactive oxygen species may in part explain the health effects of PM2.5 seen in previous epidemiological studies.