Effect-directed analysis of toxic organics in PM 2.5 exposure to the cellular bioassays in vitro: Application in Shanxi of China.

To optimize the effect-directed analysis (EDA) approach to identify the fine particulate matter (PM) bound organic toxicants, Jinzhong city, in the Shanxi Province of China, was selected as the object of our study. First, PM samples were collected and their organic extracts were separated out in 9 fractions (F1-F9) using reversed-phase high performance liquid chromatography after purification using gel permeation chromatography. Second, the toxicity effects of each fraction were measured by human bronchial epithelial cells (BEAS-2B) in vitro. And toxicity effects included antioxidant stress (ROS, LDH, and CAT) and an inflammatory response (IL-6, IL-1β, and TNF-α). The results showed that the scores of the toxicity effects on multiple lines of evidence were the highest in the F3 and F4 fractions compared with those of the control. Subsequently, the main poisons, o-cymene, p-cymene, benzene, ethylbenzene, xylene, and styrene, were identified using GC×GC-TOF/MS. Finally, to confirm the above possible candidates, (1) the levels of o-cymene, p-cymene and BTEXS in daily PM were measured using GC-MS in November 2020, and the rates of detection of these pollutants were 100% in PM. Among them, o-cymene and p-cymene were first reported as the key toxic substances of PM, and their average concentration values were 0.16 ± 0.11 and 0.18 ± 0.15 ng‧m, respectively. (2) the toxicity of p-cymene may be no less than that of other benzene derivatives according to their LC50 in Daphnia magna. (3) based on canonical correlation analysis, the exposure to p-cymene, benzene, and styrene in PM was most likely associated with the toxicity effects (CAT, IL-6, and TNF-α), which in turn caused the observed toxicity. In conclusion, p-cymene, benzene, and styrene were found to be the key toxic organics in PM for cells in vitro. EDA technology avoids the limitations of chemical analysis and uncertainty of the biological testing and adds new toxicants to the control list of PM, contributing to this study field. However, the application of EDA to PM still faces challenges such as the selection of biological effects, loss of toxicity with the separation process, influence of the dosing method, and identification of the unknown effects of pollutants.