BackgroundTraffic particulate matter (PM) comprises a mixture of particles from fuel combustion and wear of road pavement, tires and brakes. In countries with low winter temperatures the relative contribution of mineral-rich PM from road abrasion may be especially high due to use of studded tires during winter season. The aim of the present study was to sample and characterize size-fractioned PM from two road tunnels paved with different stone materials in the asphalt, and to compare the pro-inflammatory potential of these fractions in human bronchial epithelial cells (HBEC3-KT) in relation to physicochemical characteristics.MethodsThe road tunnel PM was collected with a vacuum pump and a high-volume cascade impactor sampler. PM was sampled during winter, both during humid and dry road surface conditions, and before and after cleaning the tunnels. Samples were analysed for hydrodynamic size distribution, content of elemental carbon (EC), organic carbon (OC) and endotoxin, and the capacity for acellular generation of reactive oxygen species. Cytotoxicity and pro-inflammatory responses were assessed in HBEC3-KT cells after exposure to coarse (2.5–10 μm), fine (0.18–2.5 μm) and ultrafine PM (≤ 0.18 μm), as well as particles from the respective stone materials used in the pavement.ResultsThe pro-inflammatory potency of the PM samples varied between road tunnels and size fractions, but showed more marked responses than for the stone materials used in asphalt of the respective tunnels. In particular, fine samples showed significant increases as low as 25 µg/mL (2.6 µg/cm2) and were more potent than coarse samples, while ultrafine samples showed more variable responses between tunnels, sampling conditions and endpoints. The most marked responses were observed for fine PM sampled during humid road surface conditions. Linear correlation analysis showed that particle-induced cytokine responses were correlated to OC levels, while no correlations were observed for other PM characteristics.ConclusionsThe pro-inflammatory potential of fine road tunnel PM sampled during winter season was high compared to coarse PM. The differences between the PM-induced cytokine responses were not related to stone materials in the asphalt. However, the ratio of OC to total PM mass was associated with the pro-inflammatory potential.