Regulatory Genomics of Ozone Air Pollution Response in Vitro and In Vivo

Project Summary Exposure to the ambient air pollutant ozone (O3) is associated with cardiopulmonary morbidity and mortality, rendering it an important public health issue. Controlled exposure studies show that acute O3 exposure causes airway inflammation, epithelial injury, and a transient decrease in lung function. These studies have also

demonstrated that subjects exhibit highly reproducible differences in O3 response, suggestive of gene-by- environment interactions (GxE). Candidate gene studies have provided evidence of GxE for a handful of genes, however, the role of genetic variants in the rest of the genome is largely unknown. This data gap limits our ability

to identify susceptible individuals and gain insight into mechanisms by which O3 causes adverse effects. Here, we put forth a proposal to address this data gap using human bronchial epithelial cells (hBECs) in vitro. hBECs are the first cells of the respiratory tract to interact with O3, and we have shown that hBECs exposed to O3 in vitro

upregulate the expression of key pro-inflammatory genes (e.g., CXCL8), mirroring the in vivo response. We hypothesize that variation in O3-induced inflammation is associated with differences in hBEC gene expression, and that inter-individual differences in gene expression at baseline and after O3 have a genetic basis, i.e., are

expression quantitative trait loci (eQTL). Further, we hypothesize that some eQTL are caused by single nucleotide polymorphisms (SNPs) that affect chromatin accessibility (caQTL). In Aim 1, we will establish well- differentiated hBEC cultures, grown at air-liquid interface, from 300 banked lung tissue donors of both sexes and

diverse ancestries, then expose them to O3 vs. filtered air (FA) and measure key hBEC O3 response phenotypes (e.g. IL-8 production, oxidative stress, lipid peroxidation, barrier function, and cytotoxicity). We will profile gene expression in FA and O3-exposed hBECs using both bulk RNA-seq and single cell RNA-seq to identify O3-

induced/repressed genes and their cell-type specificity. After genotyping, we will map eQTL at baseline (mRNAFA), response eQTL (mRNAO3-mRNAFA), and QTL for all hBEC O3 response phenotypes, then use mediation analyses to identify SNPs and genes fitting a putative causal model: O3+SNP → [mRNA] → hBEC O3

response phenotype. In Aim 2, we will perform ATAC-seq to characterize how O3 alters chromatin accessibility in hBECs, then map baseline and response caQTL. We will perform multi-omic data integration (eQTL, caQTL, QTL) to identify gene regulatory models of O3 response, i.e., O3+SNP→chromatin accessibility→[mRNA]→hBEC

O3 response phenotype. Finally, in Aim 3, we will validate novel genes and gene regulatory mechanisms underlying variation in O3 response in vitro and in vivo. We will determine how key SNPs affect gene regulation and whether knocking down the corresponding genes alters O3 response in vitro. For in vivo validation, we will

test for association between SNPs of interest and O3-induced neutrophil recruitment in a dataset of 191 human volunteers exposed to O3. In total, our work will identify genetic variants and gene regulatory mechanisms that influence susceptibility to O3-induced airway inflammation.