Atmospheric particulate matter (PM) is genotoxic and recently was classified as carcinogenic to humans by the International Agency for Research on Cancer. PM chemical composition varies depending on source and atmospheric conditions. The Salmonella/microsome assay is the most used mutagenicity test and can identify the major chemical classes responsible for observed mutagenicity. The objective of this work was to characterize the mutagenicity of PM samples from a countryside city, Limeira, Brazil, which is influenced by heavy traffic and sugar cane biomass burning. Six samples of total PM were collected. Air mass backward trajectories were calculated. Organic extracts were assayed using the Salmonella/microsome microsuspension mutagenicity assay using TA98, YG1041, and TA1538, with and without metabolic activation (S9). YG1041 was the most sensitive strain and mutagenicity reached 9,700 revertants per m(3) without metabolic activation. Potency for TA1538 was higher than TA98, indicating that this strain should be considered in air mutagenicity studies. The increased response to YG1041 relative to TA98, and the decreased response with S9, suggests that nitroaromatics are the major contributors. Limeira is among the most mutagenic cities in the world. High mutagenicity in Limeira seems to occur when the air mass from the area of sugarcane production is mixed with air from the region impacted by anthropogenic activities such as traffic. An increase in the formation of nitro-polycyclic aromatic hydrocarbons may result from longer contact time between the aromatic compounds and the atmosphere with high NOx and ozone concentration, although more studies are required to confirm this hypothesis.

The Salmonella/microsome assay is the most used assay for the evaluation of air particulate matter (PM) mutagenicity and a positive correlation between strain TA98 responses and benzo[a]pyrene (B[a]P) levels in PM has been found. However, it seems that the major causes of PM mutagenicity in this assay are the nitro and oxy-PAHs. Salmonella YG5161, a 30-times more responsive strain to B[a]P has been developed. To verify if YG5161 strain was sufficiently sensitive to detect mutagenicity associated with B[a]P mutagenicity, PM samples were collected in Brazil and Sweden, extracted with toluene and tested in the Salmonella/microsome microsuspension assay. PAHs and B[a]P were determined and the extracts were tested with YG5161 and its parental strain TA1538. The extracts were also tested with YG1041 and its parental strain TA98. For sensitivity comparisons, we tested B[a]P and 1-nitropyrene (1-NP) using the same conditions. The minimal effective dose of B[a]P was 155 ng/plate for TA1538 and 7 ng/plate for YG5161. Although the maximum tested dose, 10 m(3) /plate containing 9 ng of B[a]P in the case of Brazilian sample, was sufficient to elicit a response in YG5161, mutagenicity was detected at a dose as low as 1 m(3) /plate (0.9 ng). This is probably caused by nitro-compounds that have been shown to be even more potent than B[a]P for YG5161. It seems that the mutagenicity of B[a]P present in PM is not detectable even with the use of YG5161 unless more efficient separation to remove the nitro-compounds from the PAH extract is performed.