Contrasting neurotoxic pathways triggered by PM10 in relation with organic molecular markers in suburban and rural sites in Catalonia.

Abstract

Air pollution, particularly particulate matter (PM), is a major driver of global morbidity and mortality, with increasing evidence linking it to neurological disorders. This study investigates the chemical composition and neurotoxic potential of PM collected simultaneously in three sites in Catalonia (Spain): Bellver de Cerdanya (rural background), Manlleu (suburban), and Mollet del Vallès (suburban-industrial). Fifty-four filter samples collected in 2022 were analyzed by GC-MS for 30 organic molecular tracers, including polycyclic aromatic hydrocarbons (PAHs) and levoglucosan. Extracts were tested in SH-SY5Y human neuroblastoma cells across six toxicity endpoints: cell viability, reactive oxygen species (ROS), acetylcholinesterase (AChE) activity, antioxidant response, xenobiotic response, and p53 activation (DNA damage response). Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) on the combined chemical-biological dataset resolved four components: a winter biomass burning component enriched in levoglucosan, dehydroabietic acid, and PAHs, inducing strong cytotoxicity, oxidative stress, and xenobiotic responses; a traffic component present throughout the year; a spring-summer secondary organic aerosol (SOA) component associated with selective AChE inhibition without cytotoxicity; and a summer primary organic aerosol (POA) component. Partial Least Squares (PLS) regression linked PM10 composition with toxicity responses. Five of six models were statistically significant (R2CV = 0.53-0.75), with the highest performance for ROS, p53 activation, and cell death (R2CV ≥ 0.62). Biomass burning markers and PAHs were the main predictors of oxidative stress and cytotoxicity, whereas biogenic SOA tracers showed low importance. These findings link specific PM10 sources to distinct neurotoxic effects and highlight the importance of controlling winter emissions.