A multidisciplinary approach to characterise exposure risk and toxicological effects of PM₁₀ and PM₂.₅ samples in urban environments.

Urban aerosol samples collected in Barcelona between 2008 and 2009 were toxicologically characterised by means of two complementary methodologies allowing evaluation of their Reactive Oxidative Stress (ROS)-generating capacity: the plasmid scission assay (PSA) and the dichlorodihydrofluorescin assay (DCFH). The PSA determined the PM dose able to damage 50% of a plasmid DNA molecule (TD(50) values), an indication of the ability of the sample to exert potential oxidative stress, most likely by formation of ·OH. This toxicity indicator did not show dependency on different air mass origins (African dust, Atlantic advection), indicating that local pollutant sources within or near the city are most likely to be mainly responsible for PM health effect variations. The average TD(50) values show PM(2.5-0.1) samples to be more toxic than the PM(10-2.5) fraction, with doses similar to those reported in previous studies in polluted urban areas. In addition, the samples were also evaluated using the oxidant-sensitive probe DCFH confirming the positive association between the amount of DNA damage and the generation of reactive oxidant species capable of inducing DNA strand break. Results provided by the PSA were compared with those from two other different methodologies to evaluate human health risk: (1) the toxicity of particulate PAHs expressed as the calculated toxicity equivalent of benzo[a]pyrene (BaPteq) after application of the EPA toxicity factors, and (2) the cancer risk assessment of the different PM sources detected in Barcelona with the receptor model Positive Matrix Factorisation (PMF) and the computer programme Multilinear Engine 2 (ME-2) using the organic and inorganic chemical compositions of particles. No positive associations were found between PSA and the toxicity of PAHs, probably due to the inefficiency of water in extracting organic compounds. On the other hand, the sum of cancer risk estimates calculated for each of the selected days for the PSA was found to correlate with TD(50) values in the fine fraction, with fuel oil combustion and industrial emissions therefore being most implicated in negative health effects. Further studies are necessary to determine whether toxicity is related to PM chemical composition and sources, or rather to its size distribution.