Untargeted Metabolomics Analysis of Wood Smoke and Diesel Exhaust Particulate Matter

Abstract

TPS 711: The exposome, Exhibition Hall, Ground floor, August 26, 2019, 3:00 PM - 4:30 PM Background: Untargeted metabolomics analysis is a valuable tool in the field of metabolic profiling of biological samples, but it has been less widely applied to characterization of environmental mixtures. The current study explores using untargeted analysis of environmental samples of air pollution for component analysis and source appointment. Methods: Particulate matter (PM) in wood smoke and diesel exhaust were collected in laboratory settings at the University of Washington. Total suspended particles were collected using 37 mm Teflon filtration membranes (PALL Corporation, USA) in open-face cassettes attached to air sampling pumps (AirChek XR5000, SKC Inc., USA). After sample extraction (Methanol), concentration and reconstitution (25μL Methanol and 25μL 0.4% (v/v) acetic acid), PM samples were analyzed by accurate-mass Q-TOF (Agilent 6500) for separation and qualification. We used MetaboAnalyst (version 4.0) to perform principal component analysis (PCA) to visualize the data and t-tests to compare chemical features between wood smoke and diesel exhaust groups. Results: We collected and analyzed six wood smoke PM samples (weight = 65 ± 3 μg) and five diesel exhaust PM samples (average weight = 150 ± 1 μg). Using Q-TOF HPLC/MS, we detected 1513 chemical features from the PM samples. In a volcano plot, a total of 112 chemical features were identified that exhibited differences (fold-change > 2, and p-value < 0.05) between PM from the wood smoke and diesel exhaust. PCA showed that 81.3% of the variance was explained when including the 5 most significant principal components (PC). Both the 2-D scores plot and 3-D score plot between selected PCs totally separated wood smoke and diesel exhaust PM samples. Conclusion: Untargeted metabolomics analysis detected numerous chemical features from wood smoke and diesel exhaust PM samples. The untargeted analysis has the potential to be used in air pollution studies for components analysis and source appointment of PM.