Mapping acute systemic effects of inhaled particulate matter and ozone: multiorgan gene expression and glucocorticoid activity

Abstract

Recent epidemiological studies have demonstrated associations between air pollution and adverse effects that extend beyond respiratory and cardiovascular disease, including low birth weight, appendicitis, stroke, and neurological/neurobehavioural outcomes (e.g., neurodegenerative disease, cognitive decline, depression, and suicide). To gain insight into mechanisms underlying such effects, we mapped gene profiles in the lungs, heart, liver, kidney, spleen, cerebral hemisphere, and pituitary of male Fischer-344 rats immediately and 24h after a 4-h exposure by inhalation to particulate matter (0, 5, and 50mg/m(3) EHC-93 urban particles) and ozone (0, 0.4, and 0.8 ppm). Pollutant exposure provoked differential expression of genes involved in a number of pathways, including antioxidant response, xenobiotic metabolism, inflammatory signalling, and endothelial dysfunction. The mRNA profiles, while exhibiting some interorgan and pollutant-specific differences, were remarkably similar across organs for a set of genes, including increased expression of redox/glucocorticoid-sensitive genes and decreased expression of inflammatory genes, suggesting a possible hormonal effect. Pollutant exposure increased plasma levels of adrenocorticotropic hormone and the glucocorticoid corticosterone, confirming activation of the hypothalamic-pituitary-adrenal axis, and there was a corresponding increase in markers of glucocorticoid activity. Although effects were transient and presumably represent an adaptive response to acute exposure in these healthy animals, chronic activation and inappropriate regulation of the hypothalamic-pituitary-adrenal axis are associated with adverse neurobehavioral, metabolic, immune, developmental, and cardiovascular effects. The experimental data are consistent with epidemiological associations of air pollutants with extrapulmonary health outcomes and suggest a mechanism through which such health effects may be induced.

Keywords: air pollution; glucocorticoid; hypothalamic-pituitary-adrenal axis; ozone; particulate matter; real-time PCR.; systemic effects.

Fig. 1.

Mapping pollutant effects across organs. Gene expression was assessed in all organs 0 and 24h after exposure of rats (n = 4–6 per group) to 50mg/m³ particulate matter (blue line), 0.8 ppm ozone (red line), or coexposure to 50mg/m³ particulate matter and 0.8 ppm ozone (green line). Vector diagrams display the fold change in response to treatment relative to the mean of the air control group for each gene. Genes that fell below the level of detection were given a value of 1. Statistical analyses for all genes are presented in Supplementary table 2. Metallothionein (MT), cytochrome p450 family (CYP), endothelin (ET), inducible nitric oxide synthase (iNOS), endothelial nitric oxide synthase (eNOS), hypoxia inducible factor (HIF), heme oxygenase (HMOX), NAD(P)H dehydrogenase quinone 1 (NQO1), glutathione reductase (GSR), glutathione S-transferase A2 (GSTA2), prostaglandin endoperoxide synthase 2 (PTGS2), interleukin (IL), tumour necrosis factor (TNF), chemokine (CC motif) ligand 2 (CCL2), histone H3 lysine-27 demethylase JMJD3, and vascular endothelial growth factor (VEGF).

Fig. 4.

Xenobiotic and antioxidant gene response in the lungs after particulate matter inhalation. Animals (n = 4–6 per group) were exposed to particulate matter (0, 5, and 50mg/m³ EHC-93), ozone (0, 0.4, and 0.8 ppm), or combinations of particles and ozone for 4h and euthanized 0 or 24h after exposure. Results are presented as the geometric mean ± geometric standard deviation. Letters over bars indicate statistical significance (Holm-Sidak, p < 0.05). (A) Lung cytochrome P450 family 1 member A1 (CYP1A1) mRNA immediately after exposure. EHC main effect, p < 0.001. ^a0, 5 versus 50mg/m³. (B) Lung NAD(P)H dehydrogenase quinone 1 (NQO1) mRNA immediately after exposure. EHC main effect, p = 0.025. ^a0, 5 versus 50mg/m³.

Fig. 2.

Regulation of inflammatory genes immediately following pollutant inhalation. (A) TNF, (B) IL-1β, and (C) MCP-1 were measured in organs immediately after exposure to ozone and particulate matter (EHC-93). To facilitate comparison of effects across multiple organs for a given treatment, only effects of exposure to EHC-93 (0, 5, and 50mg/m³) or ozone (0, 0.4, and 0.8 ppm) are displayed (n = 4–6 per group). Results are presented as the geometric mean ± geometric standard deviation. Asterisks denote statistical significance relative to control (Holm-Sidak, p < 0.05) as guided by statistical analyses (two-way ANOVA) conducted on the entire data set that includes coexposure to both ozone and particulate matter. Complete statistical analyses are presented in Supplementary table 2.

Fig. 3.

Antioxidant and inflammatory responses in the lungs after ozone inhalation. Animals (n = 4–6 per group) were exposed to particulate matter (0, 5, and 50mg/m³ EHC-93), ozone (0, 0.4, and 0.8 ppm), or combinations of particles and ozone for 4h and euthanized 0 or 24h after exposure. Results are presented as the geometric mean ± geometric standard deviation. Letters over bars indicate statistical significance (Holm-Sidak, p < 0.05). (A) Lung IL-6 mRNA immediately after exposure. Ozone main effect, p < 0.001. ^a0 versus 0.4 ppm ozone; ^b0, 0.4 versus 0.8 ppm ozone. (B) Lung metallothionein (MT)-2 mRNA immediately after exposure. Ozone main effect, p < 0.001. ^a0, 0.4 versus 0.8 ppm ozone.

Fig. 5.

Mapping pollutant effects by gene. Gene expression was assessed in all organs 0 and 24h after exposure to 50mg/m³ particulate matter, 0.8 ppm ozone, or coexposure to 50mg/m³ particulate matter and 0.8 ppm ozone (n = 4–6 per group). Expression of a given gene across all organs was plotted to enable direct comparison of interorgan differences in response. Statistical analyses for all genes are presented in Supplementary table 2. Lung (Lu), heart (He), liver (Li), kidney (Ki), spleen (Sp), cerebral hemisphere (CH), pituitary (Pi), cytochrome P450 family 1 member A1 (CYP1A1), metallothionein (MT), chemokine (C-C motif) ligand 2(CCL2)/monocyte chemotactic protein1 (MCP1), interleukin (IL) and prostaglandin endoperoxide synthase 2 (PTGS2).

Fig. 6.

Activation of the hypothalamic-pituitary-adrenal axis. Letters over bars indicate statistical significance (Holm-Sidak, p < 0.05). (A) Pituitary prostaglandin endoperoxide synthase 2 (PTGS2) mRNA (n = 4–6 per group). Relative mRNA expression was assessed by real-time PCR in rats exposed to the indicated pollutant exposures. Results are presented as the geometric mean ± geometric standard deviation. Ozone × Time interation, p < 0.001. ^a0 versus 0.8 ppm within 0h; ^b0 versus 24h within 0 ppm ozone; ^c0 versus 24h within 0.8 ppm ozone. (B) Plasma ACTH levels (n = 4–6 per group). Plasma ACTH was assessed by integrating the peak area of the analyte peak at m/z 2093.1 assigned to the ACTH(1–17) fragment following MALDI-TOF mass spectrometric analysis (relative units). Ozone main effect, p = 0.015. ^a0 versus 0.8 ppm within OZONE; two-way ANOVA, one-tailed α = 0.05. (C) Plasma corticosterone (n = 4 per group). Immunoreactive corticosterone was measured by ELISA assay. Ozone × Time interaction, p = 0.018; EHC main effect, p = 0.041. ^a0 versus 50mg/m³ within EHC; ^b0 versus 0.8 ppm within 0h; ^c0 versus 24h within 0.8 ppm ozone; three-way ANOVA, one-tailed α = 0.05.

Fig. 7.

Regulation of glucocorticoid-responsive genes by inhaled pollutants. (A) GILZ and (B) SGK mRNA levels were measured in organs immediately after exposure to ozone and particulate matter (EHC-93). To facilitate comparison of effects across multiple organs for a given treatment, only effects of exposure to EHC-93 (0, 5, and 50mg/m³) or ozone (0, 0.4, and 0.8 ppm) are displayed (n = 4–6 per group). Results are presented as the geometric mean ± geometric standard deviation. Asterisks denote statistical significance relative to control (Holm-Sidak, p < 0.05) as guided by statistical analyses (two-way ANOVA) conducted on the entire data set that includes coexposure to both ozone and particulate matter. Complete statistical analyses are presented in Supplementary table 2.

Fig. 8.

Distinct signalling pathways induced by pollutant exposure in the lungs and liver. Relative mRNA levels of GILZ and metallothionein (MT)-2 mRNA were plotted for each individual animal in the lungs and liver. Dotted lines indicate the group of animals exposed to 0.8 ppm ozone.