Cluster analysis of fine particulate matter (PM_2.5) emissions and its bioreactivity in the vicinity of a petrochemical complex

This study evaluated associations between the bioreactivity of PM2.5 in vitro and emission sources in the vicinity of a petrochemical complex in Taiwan. The average PM2.5 was 30.2 μg/m3 from 9 February to 23 March 2016, and the PM2.5 was clustered in long-range transport (with major local source) (12.8 μg/m3), and major (17.3 μg/m3) and minor industrial emissions (4.7 μg/m3) using a k-means clustering model. A reduction in cell viability and increases in the cytotoxicity-related lactate dehydrogenase (LDH), oxidative stress-related 8-isoprostane, and inflammation-related interleukin (IL)-6 occurred due to PM2.5 in a dose-dependent manner. The PM2.5 from major industrial emissions was significantly correlated with increased 8-isoprostane and IL-6, but this was not observed for long-range transport or minor industrial emissions. The bulk metal concentration was 9.52 ng/m3 in PM2.5. We further observed that As, Ba, Cd, and Se were correlated with LDH in the long-range transport group. Pb in PM2.5 from the major industrial emissions was correlated with LDH, whereas Pb and Se were correlated with 8-isoprostane. Sr was correlated with cell viability in the minor industrial emissions group. We demonstrated a new approach to investigate particle bioreactivity, which suggested that petrochemical-emitted PM2.5 should be a concern for surrounding residents' health. The significance and novelty of this study was that a new approach (source-toxicity apportionment) is developed for determining environmental health by integration of source apportion modeling with environmental toxicity.