Multi-tissue transcriptomic and serum metabolomic assessment reveals systemic implications of acute ozone-induced stress response in male Wistar Kyoto rats

doi:10.21203/rs.3.rs-3178715/v1

Download PDF

Research Article

Multi-tissue transcriptomic and serum metabolomic assessment reveals systemic implications of acute ozone-induced stress response in male Wistar Kyoto rats

https://doi.org/10.21203/rs.3.rs-3178715/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 10 Sep, 2023

Read the published version in Metabolomics →

You are reading this latest preprint version

Air pollutant exposures have been linked to systemic disease; however, the underlying mechanisms between responses of the target tissue and systemic effects are poorly understood. A prototypic inducer of stress, ozone causes respiratory and systemic multiorgan effects through activation of a neuroendocrine stress response. The goal of this study was to assess transcriptomic signatures of multiple tissues and serum metabolomics to understand how neuroendocrine and adrenal-derived stress hormones contribute to multiorgan health outcomes. Male Wistar Kyoto rats (12–13 weeks old) were exposed to filtered air or 0.8 ppm ozone for 4-hours, and blood/tissues were collected immediately post-exposure. Each tissue had distinct expression profiles at baseline. Ozone changed 1,640 genes in lung, 274 in hypothalamus, 2,516 in adrenals, 1,333 in liver, 1,242 in adipose, and 5,102 in muscle (adjusted p-value < .1, absolute fold-change > 50%). Serum metabolomic analysis identified 863 metabolites, of which 447 were significantly altered in ozone-exposed rats (adjusted p-value < .1, absolute fold change > 20%). A total of 6 genes were differentially expressed in all 6 tissues. Glucocorticoid signaling, hypoxia, and GPCR signaling were commonly changed, but ozone induced tissue-specific changes in oxidative stress, immune processes, and metabolic pathways. Genes upregulated by TNF-mediated NFkB signaling were differentially expressed in all ozone-exposed tissues, but those defining inflammatory response were tissue-specific. Upstream predictor analysis identified common mediators of effects including glucocorticoids, although the specific genes responsible for these predictors varied by tissue. Metabolomic analysis showed major changes in lipids, amino acids, and metabolites linked to the gut microbiome, concordant with transcriptional changes identified through pathway analysis within liver, muscle, and adipose tissues. The distribution of receptors and transcriptional mechanisms underlying the ozone-induced stress response are tissue-specific and involve induction of unique gene networks and metabolic phenotypes, but the shared initiating triggers converge into shared pathway-level responses. This multi-tissue transcriptomic analysis, combined with circulating metabolomic assessment, allows characterization of the systemic inhaled pollutant-induced stress response.

metabolic response

neuroendocrine

stress response

transcriptome

metabolome

The surge of mental health crises and peripheral chronic disease epidemics of the 21st century have been attributed to increased stressors of physiological, psychosocial, and environmental nature (Reuben et al., 2022). The autonomic sensory and motor nervous system, in conjunction with neuroendocrine-mediated responses, are responsible for the recognition of stress encounters and subsequent orchestration of systemic, tissue-specific coordinated biological responses to restore homeostasis. The same mechanisms are also responsible for returning these stress-induced changes to a stable homeostatic setpoint. The general scientific consensus is that the immediate stress response following a toxicant encounter is non-specific and reversible, with no long-term consequences (Everds et al., 2013). The ability of an organism to induce a stress response to an environmental exposure might have implications in individual variations in susceptibility (de Kloet & Joëls, 2023).

Inhaled air pollutants with oxidant and reactive properties have been extensively studied in relation to lung cellular injury mechanisms and cardio-physiological outcomes. However, the contribution of the neuroendocrine stress mechanisms in producing pulmonary, as well as extra-pulmonary effects, has only been recently recognized (Kodavanti et al., 2023). These new findings demonstrate that air pollutants, regardless of their ability to systemically translocate, induce myriad effects in distant organs. We have demonstrated that ozone exposure induces a neurally-mediated systemic stress response through the activation of sympathetic adrenal medullary (SAM) and hypothalamic pituitary adrenal (HPA) axes together with inhibition of hypothalamic pituitary thyroid (HPT) and hypothalamic pituitary gonadal (HPG) axes. These dynamic neurohormonal changes are associated with lung injury and inflammation, systemic metabolic alterations, and lung, hypothalamic, and liver gene expression changes. Interventional approaches have shown that adrenal-derived catecholamines and glucocorticoids mediate ozone-induced pulmonary and systemic alterations (Henriquez et al., 2018; Jackson et al., 2022; Miller, Snow, et al., 2016).

The neurohormones released in response to stress at the sympathetic nerve terminals and from neuroendocrine organs such as adrenals, mediate multiorgan effects at cellular levels. The physiological and stress response effects of these mediators on tissues are dynamic and dependent on respective receptor distribution with co-regulatory mechanisms that facilitate cell signaling changes at cytoplasmic and nuclear levels. Adrenal-derived glucocorticoids bind to ubiquitously distributed glucocorticoid receptors (GRs) in tissues and regulate expression of nearly 20% of the human genome through nuclear translocation and transcriptional activation and/or repression (Oakley & Cidlowski, 2013; Xavier et al., 2016). Glucocorticoid binding to GRs and related transcriptional activity is tightly spatially and temporally regulated to adjust the physiological stress response to maximize resiliency and prevent unwanted cellular changes (Joëls et al., 2018). Adrenal hormones are released in the circulation within an hour into ozone exposure and mediate effects through their receptor binding (Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022).

Although the biology of ARs and GRs has been extensively characterized with respect to therapeutic potential, minimal attention has focused on their role in mediating responses to environmental stressors and the contribution of their malfunction in toxicology. Likewise, the contribution of these stress hormones and receptors in mediating multiorgan effects of inhaled pollutants have not been characterized using a systems-level approach. Understanding how cellular changes caused by activation of ARs and GRs in multiple organ systems can lead to systemic toxicity is essential to evaluating the stress response and inter-individual differences in susceptibility.

The goal of the present study was to examine shared and unique transcriptional response to acute ozone inhalation as a model environmental stressor and evaluate whether those transcriptional changes were linked to systemic metabolic responses. Characterizing organ-specific effects is important in understanding potential interactive downstream effects of psychosocial and environmental stressors. Using a systems biology approach, we exposed male WKY rats to 0.8ppm ozone and examined the subsequent transcriptional response in six tissues and global metabolomic assessment of serum. The novel integration of multi-tissue transcriptomic and circulating metabolomic response to a well-characterized stressor allows us to explore conserved, systemic responses and those that may be limited to only some organ systems.

Animals

Physiological data from this cohort of animals has been previously published (Henriquez, Snow, Jackson, House, Alewel, et al., 2022). Male Wistar-Kyoto (WKY) rats were purchased from Charles River Laboratories, Inc., Raleigh, NC at 4 weeks of age. Animals were double-housed in cages with hardwood chip bedding and EnviroDri® enrichment material. The animal rooms were maintained at 21–22°C, ~ 55% relative humidity, and 12h light/dark cycle. Animals were fed Purina 5001 rat chow (Brentwood, MO) and drank filtered tap water ad libitum unless otherwise noted for 8 weeks. Rats were randomized by body weight using a custom Excel macro and allocated to air or ozone exposure. The EPA animal facility is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. All procedures were approved by U.S. EPA’s Institutional Animal Care and Use Committee following the experimental protocol and guidelines of the National Institute of Health for the care and use of rats (NIH Publications No. 8023).

Ozone generation and animal exposures

Ozone was employed as an air pollutant known to induce a canonical stress response when animals were 13 weeks of age, as previously described (Henriquez, Snow, Jackson, House, Alewel, et al., 2022). A silent arc discharge generator (OREC, Phoenix, AZ) generated ozone from oxygen. Mass flow controllers were used to regulate the entry of ozone into the Rochester style “Hinners” chambers. Photometric ozone analyzers (API Model 400) were used to monitor the ozone concentrations in the chambers, and rats were exposed to filter air or ozone (0.8 ppm) for 4 h followed immediately by necropsy. During exposure, all rats were individual housed in wire-mesh cages without access to food or water. Exposure chamber conditions, including temperature, relative humidity, and airflow, along with actual ozone concentrations achieved have been previously described (Henriquez, Snow, Jackson, House, Alewel, et al., 2022). Ozone concentrations are an order of magnitude higher than encountered in non-attainment areas in the US (Perdigones et al., 2022), but this concentration in resting rats is comparable to levels used in human clinical experiments during intermittent exercise based on airway dose deposition (Hatch et al., 1994).

Necropsy and tissue collection

Animals were fasted for the duration of exposure and until necropsies were complete (total fast of 5–6 h). Exposures were staggered over several days to assure consistent diurnal timing for stress sensitive endpoints. Intraperitoneal injections of sodium pentobarbital (Fatal Plus, Virbac AH, Inc., Fort Worth, TX; >200 mg/kg) was used for euthanasia. Blood samples were collected in glass EDTA and serum separator vacutainer tubes from the abdominal aorta. Left lung, hypothalamus, both adrenal glands, a piece of the right median lobe of the liver, a piece of white abdominal tissue from the right side, and a piece of gracilis leg muscle from left side of the animal were removed, frozen in liquid nitrogen, and stored at -80°C.

RNA isolation and mRNA sequencing

Total RNA was isolated from each tissue using Qiagen RNAeasy Mini kits (liver, lung), Qiagen Fibrous Tissue Mini kits (muscle), Qiagen Lipid Tissue Mini kits (adipose; Germantown, MD), or Direct-Zol RNA MiniPrep kits from Zymo Research (hypothalamus, adrenals; Irvina, California). RNA quality was examined using an Agilent Bioanalyzer (Agilent Technologies, Santa Clara, California). RNA samples were processed and run in Apollo324 automated sample processing system for mRNA selection (Wafergen Biosystems, Fremont, California). RNA-Seq library was prepared with Wafergen’s PrepX mRNA 48 Protocol. cDNA libraries were amplified and dsDNA from each library was quality checked by Agilent Bioanalyzer with an RIN cutoff of 7 (Agilent Technologies, Santa Clara, California). Each library was sequenced according to Illumina NextSeq 500 protocols (1.8 pM + 5% PhiX from Illumina, Inc, San Diego, California). An n = 6 per tissue per exposure were included, except for adipose where only 5 air and 5 ozone samples passed QC.

Nextseq data normalization, statistics, and bioinformatics

Sequenced mRNA reads were mapped to the rat genome (rn7) using ensemble release 105 in the Partek Flow suite, R version 4.1.3. In total, 28,760 genes were mapped. Genes that did not have at least 1 read in at least 10 of the 70 samples were removed, leaving 19,465 genes. An average of 13.8 million reads were mapped per sample with a standard deviation of 2.4 million reads. The average counts per gene per sample were 709. Principal components analysis was performed to identify if any samples should be removed. All samples clustered by tissue. When clustering was examined within each tissue for air and ozone-exposed rats, two samples (one adrenal air, one muscle ozone) did not cluster by treatment and were removed. Differential gene expression was assessed in DESeq2 package in R (v1.34.0) (Love et al., 2014) and normalized reads were assessed for ozone effects using various bioinformatics approaches described further below. The Illumina mRNAseq data are deposited at the Gene Expression Omnibus Web site (accession number GSE234042).

Gene expression was considered different when fold change (FC) was more than 1.5 (upregulation) or less than 0.667 (downregulation) and the Benjamini-Hochberg afalse discovery rate was ≤ .1. Gene expression values were used for biologic pathway analysis using the eXploring Genomic Relations (XGR) R package (Fang et al., 2016) to analyze overexpression of Reactome and Hallmark pathways. Gene lists for heatmaps were generated using the Reactome content service (Fabregat et al., 2018) and the Molecular Signatures Database (MSigDB) (Subramanian et al., 2005). Pathway heatmaps were generated using the heatmap.2 function from the gplots package in R. The UpSetR package (Conway et al., 2017) was used to generate a plot to visualize the intersection of DEG within tissues visualizing set data for intersecting sets of tissues.

Upstream predictor analysis was conducted using an absolute fold-change cutoff of 50% with Ingenuity Pathways Analysis (IPA, QIAGEN Redwood City, www.qiagen.com/ingenuity; last accessed January 27, 2023) and adjusted p-values ≤ .1 (Krämer et al., 2014).

Metabolomics sample preparation

Rat serum metabolomics was performed as described previously (Bridgewater BR, 2014). One hundred microliters of sample were used for each analysis. Samples were prepared using the automated MicroLab STAR® system from Hamilton Company. Several recovery standards were added prior to the first step in the extraction process for QC purposes. To remove protein, small molecules bound to protein or trapped in the precipitated protein matrix were dissociated, and to recover chemically diverse metabolites, proteins were precipitated with methanol under vigorous shaking for 2 min (Glen Mills GenoGrinder 2000) followed by centrifugation (Ford et al., 2020). Internal standards specific for each chromatographic method were also added to the methanol extract.

Serum Metabolomics – LC/MS/MSⁿ analyses

All methods utilized a Waters ACQUITY ultra-performance liquid chromatography (UPLC) and a Thermo Scientific Q-Exactive high resolution/accurate mass spectrometer interfaced with a heated electrospray ionization (HESI-II) source and Orbitrap mass analyzer operated at 35,000 mass resolution (Bridgewater BR, 2014; Ford et al., 2020). The dried sample extract aliquots were reconstituted in solvents compatible to each of four LC-MS/MS methods with a series of isotopically labeled standards at fixed concentrations to monitor injection and chromatographic consistency and to align chromatograms. Separate aliquots were analyzed by two reverse phase positive ion methods, one reverse phase negative ion method, and one hydrophilic interaction liquid chromatographic method (Bridgewater BR, 2014). Raw data files were archived, and data extracted. Metabolites were identified by matching the ion chromatographic retention index, accurate mass, and mass spectral fragmentation signatures with a reference library consisting of over 4,000 entries created from authentic standard metabolites under the identical analytical procedure as the experimental samples (DeHaven et al., 2010). Identification of compounds was based on the match of its retention time, parent ion accurate mass, and MS/MS fragmentation spectrum to an authentic standard representing Tier 1 identification (Sumner et al., 2007). Some compounds were identified based on parent ion accurate mass and MS/MS fragmentation mass spectral data without a reference standard (i.e., Tier 2 identification). Molecules not detected in a sample were below the limit of detection.

All samples were run as a single batch and randomized within 24 samples for each of the MS/MS methods. A single aliquot of each sample was run in the study. Matrix samples, which are technical replicates of pooled samples, were run as technical replicates throughout the study to monitor process variability and quality. Within each sample batch, 3 process blanks and 5 quality control samples were run and analyzed. For this analysis, there were two sample batches resulting in 6 process blanks and 10 quality control samples run in total. The median relative standard deviation (RSD) was calculated for all standards and endogenous biochemicals using median scaled values. The median RSDs for the internal standards and endogenous biochemicals were 3% and 8%, respectively.

Metabolomics data normalization, statistics, and bioinformatics

In total, 1,002 metabolites were detected and 887 were identified. Peaks were quantified using the trapezoidal method of area under the curve calculation. Prior to statistical analysis, each compound was corrected in run-day blocks by registering the medians to equal 1.00 and normalizing each data point proportionately (termed the “block correction”). Following normalization, any missing values were imputed with sample set minimums on a per biochemical basis. An average of 46.5 metabolites were imputed for each sample, with 30 metabolites excluded from subsequent statistical analysis because > 50% of samples were imputed. Each metabolite was analyzed using a linear model with ozone as the main effect. Pairwise comparisons were performed using the emmeans (estimated marginal means) package in R followed by a Benjamini-Hochberg false-discovery rate (FDR) correction, with FDR < 0.1 considered statistically significant.

Principal components analysis was performed to identify if any samples should be removed, and all samples clustered by exposure. Metabolites were considered differentially expressed when fold change (FC) was greater than 1.2 (upregulation) or less than 0.833 (downregulation) and the Benjamini-Hochberg adjusted p value was ≤ .1. For each metabolite, knowledge-based pathway annotations from the metabolomics platform were used (Metabolon, Inc.), and these annotations draw primarily from the human metabolome database (HMDB) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (Kanehisa & Goto, 2000; Wishart et al., 2007). Each known metabolite was annotated with a subpathway representing metabolic pathways or biochemical subclasses of compounds. Each subpathway is further assigned to a broad superpathway. The full profile of metabolites and their annotations can be found in Supplemental Table S1.

Pathway enrichment scores were calculated to display the number of statistically significantly different compounds relative to all detected compounds in a subpathway compared to the total number of statistically significantly different compounds relative to all compounds detected in the study. From the full set of known metabolites, those characterized as members of partially characterized subpathways or chemicals were removed prior to over-enrichment analysis. Further, only subpathways where 3 or more metabolites were detected were included for over-enrichment analysis. 810 metabolites were included in over-enrichment analysis. The enrichment value (EV) for each subpathway was calculated as:

EV = (k/m)/(n-k)/(N-m)

where m = the number of included metabolites in the pathway, k = the number of significantly modified metabolites included in the pathway, n = the total number of significantly modified metabolites overall, and N = the total number of metabolites included (810). A pathway enrichment value greater than one indicates that the pathway contains more significantly changed compounds relative to the study overall. A fisher’s exact test was performed for each pathway. Metabolomic pathway over-enrichment scores can be found in Supplemental Table S2.

Ozone-induced -omics changes are tissue-specific and robust

To determine multi-tissue transcriptional responses to acute ozone-induced activation of neuroendocrine stress pathways (Henriquez, Snow, Jackson, House, Alewel, et al., 2022; Miller, Ghio, et al., 2016), we analyzed mRNA expression of 6 tissues in air- and ozone-exposed male rats. We additionally performed serum metabolomic analysis similarly to our previous analysis (Miller et al., 2015) to assess coherency and integration with multi-tissue transcriptional response to ozone. Overall ozone-induced log2 transformed fold change for every gene that is differentially expressed in at least one tissue highlight the effects of ozone on gene expression and illustrates the differential responsiveness by tissue (Supplemental Fig. 1). Further, hierarchical clustering of differentially expressed genes reveal that each tissue has a distinct profile of genes that are physiologically transcribed in basal conditions, suggesting a role in maintenance of tissue-specific homeostatic function.

As expected, PCA analysis of the 500 most expressed transcripts revealed primary clustering of transcriptomic data by tissue (Fig. 1A). Metabolomics data was only performed in serum, and serum metabolomics clustered by exposure (Fig. 1B). Within each tissue, PCA revealed clear clustering between ozone- and air-exposed rats in lung (Fig. 1C), adrenal (Fig. 1E), Liver (Fig. 1F), adipose (Fig. 1G), and muscle (Fig. 1H), but only modest clustering by exposure in hypothalamus (Fig. 1D).

Within each tissue, there were notable ozone-induced transcriptomic changes using an absolute fold change cutoff of 50% and an adjusted p-value ≤ 0.1. Of the 19,645 genes identified with at least one read in 10 or more of the 70 total samples, ozone collectively resulted in 8,637 differentially expressed genes (DEG) in at least one tissue. Ozone resulted in 1,640 DEG in lung (978 down-, 662 up-regulated; Fig. 2A), 274 DEG in hypothalamus (125 down-, 149 up-regulated; Fig. 2B), 2,516 DEG in adrenal (1,267 down-, 1,249 up-regulated; Fig. 2C), 1,333 DEG in liver (610 down-, 723 up-regulated; Fig. 2D), 1,242 DEG in adipose (623 down-, 619 up-regulated; Fig. 2E), and 5,102 DEG in muscle (3,098 down-, 2004 up-regulated; Fig. 2F). In the serum metabolomics analysis, using an absolute fold change cutoff of 20% and an adjusted p-value ≤ 0.1, 447 of the 863 included metabolites were significantly altered (201 down-, 246 up-regulated) by ozone exposure (Fig. 2G).

An UpSet plot (Fig. 3) was generated that visualizes the sets of each tissue showing all DEG, as well as the unique intersections between tissues. Only those unique intersections with at least 50 members are shown. The overlap of different tissues used to create the UpSet plot is shown in Supplemental Table S3. The total number of DEG for a given tissue or overlap of tissues is shown in the Total DEG column, and the combsize for the UpSet plot is shown in the combsize column. Combsize indicates the number of DEG that are unique to that intersection of tissues.

Six DEG were shared between all 6 tissues (AABR07006311.1, Cdkn1a, Gpatch4, Mt1m, Mt2a, Pcdh12), with 3 (Cdkn1a, Gpatch4, and Pcdh12) sharing directionality between all 6 tissues, AABR07006311.1 up in all tissues except hypothalamus, and Mt1m/Mt2a up in all tissues except adrenal. To examine any overlap between tissues in the hypothalamic-pituitary-adrenal axis, DEG within adrenal and hypothalamus (HT) were considered. There were 65 DEG shared between the adrenal and HT (Supplemental Table S3).

To explore similarity in metabolically active tissues, overlap between liver, muscle, and adipose was examined. 135 DEG were shared between these three metabolically active tissues (Supplemental Table S3).

Pathway analyses reveals tissue concordance in ozone response

For the 6-tissue transcriptomic analysis, unbiased Reactome + Hallmark pathway analysis was performed with an adjusted p-value cutoff of .1 and an absolute fold change cutoff of 50%. There were 98 Reactome pathways (Supplemental Table S4) and 37 Hallmark pathways (Supplemental Table S5) significantly altered by ozone exposure in at least one tissue.

From these pathways, overall patterns showed that, across tissues, the ozone-induced transcriptomic changes revolved around hypoxia and polyamine metabolism, adrenergic stress response, steroid metabolic pathways, stress metabolic pathways, and stress immune pathways. Specific components of these pathways were explored further due to their biological relevance. Although the proportion of shared DEG identified in the UpSet plot and associated tables was not usually high, the patterns of pathway alterations were much more consistent across tissues, with 2 pathways (NF-kB response to TNF and hypoxia) in the top 5 most significant Hallmark pathways of every tissue and several other pathways (inflammatory response, IFNG, STAT5) in the top 5 most significant Hallmark pathways for 4–5 tissues. This suggests that different genes between tissues within the same pathways are modified by ozone exposure; the specific genes changed may vary by tissue based on basal expression and responsiveness to the stressor, or potentially due to timing of measurement in relation to the stressor exposure.

For serum metabolomics analysis, pathway over-enrichment analysis identified 36 pathways with an enrichment value greater than 1, indicating ozone-induced differentially changed metabolites within these pathways are enriched relative to the overall metabolomic changes. Notably, these included changes consistent with those resulting from multi-tissue transcriptomic changes, including markers of hypoxia and polyamine changes, G-protein coupled receptor (GPCR) and adrenergic signaling, steroid metabolic pathways, stress metabolic pathways, and stress immune pathways. Fisher’s exact test found that 9 of these pathways were significantly over-enriched relative by ozone exposure, including those related to poly- and mono-unsaturated fatty acids, amino acids, and diacylglycerol (Supplemental Table S2).

Integrated serum metabolomic and transcriptomic analysis for each tissue revealed 22 KEGG and 77 Reactome pathways shared with lung, 24 KEGG and 150 Reactome pathways shared with HT, no KEGG and 8 Reactome pathways shared with adrenal, no KEGG and 9 Reactome shared with liver, 2 KEGG and 20 Reactome shared with adipose, and 35 KEGG and 303 Reactome shared between muscle and the serum metabolome (Supplemental Tables S6 and S7). These differences in tissue-specific concordance between serum metabolite changes and transcriptional changes in different tissues suggest that ozone-induced metabolic changes did not result from altered liver or adrenal transcriptional changes and that rather the circulating metabolites might have induced changes in the tissue transcriptome.

Upstream predictor analyses reveal tissue concordance in ozone response

Upstream predictor analysis revealed 7,040 unique upstream transcriptional regulators that can explain changes in genes to be modified in at least one tissue. Of those, 3,591 were predicted to be modified in lung, with 256 predicted as activated and 150 as inhibited (Supplemental Table S8), 2,395 in hypothalamus, with 37 predicted as activated and 29 inhibited (Supplemental Table S9), 1,946 in adrenal, with 154 predicted activated and 98 inhibited (Supplemental Table S10), 2,305 in liver, with 82 predicted activated and 89 inhibited (Supplemental Table S11), 3,023 in adipose, with 113 predicted activated and 72 inhibited (Supplemental Table S12), and 3,758 in muscle, with 418 predicted activated and 747 inhibited (Supplemental Table S13). Across all tissues, 563 upstream regulators were predicted to be modified by ozone exposure, including corticosterone, Creb, cholesterol, cytokine, dexamethasone, Fulvestrant, insulin, Smad, testosterone, Tgf beta, WNT1, numerous interleukins, and several microRNA. Prediction of involvement of glucocorticoid agonists across all tissues indicated the contribution of HPA-mediated corticosterone increase following ozone exposure.

Hypoxia and polyamines

Various pathways associated with hypoxia and polyamine metabolism were modified in the 6-tissue transcriptome. A total of 160 DEG across tissues were identified in at least one tissue in the hypoxia pathway (Fig. 4A; Supplemental Table S14), 46 DEG in the polyamine metabolism pathway {Supplemental Table S15), and 252 DEG in amino acid metabolism pathways (Supplemental Table S16).

Pathway analysis of serum metabolomics revealed significant overenrichment in a subpathway associated with food components and plants, with an EV = 1.33 and 29 of 43 metabolites significantly altered by ozone exposure (Supplemental Table S2), indicative of changes in microbiome and potentially hypoxia (Han et al., 2021). Although polyamine metabolism is enriched in the serum metabolomics (EV = 1.22, with 5 of 8 metabolites significantly modified by ozone), this pathway is not significantly altered by ozone exposure (Supplemental Table S2). Branched-chain amino acids, leucine, isoleucine, and valine metabolism was significantly enriched (EV = 1.54, 21 of 27 metabolites) by ozone exposure. Numerous aromatic amino acids were also modified by ozone exposure, including tyrosine and tryptophan metabolites (Supplemental Table S1).

Multi-omics analysis of the transcriptome and metabolome also identified numerous Reactome pathways involved in translation of proteins, peptide chain elongation, metabolism of proteins, amino acid and derivative metabolism, selenoamino acid metabolism, selenocysteine synthesis, and several other pathways involved in polyamine signaling shared between the hypothalamic transcriptome and circulating metabolome. Some of the translation, amino acid metabolism, and protein signaling changes were also shared between the muscle transcriptome and circulating metabolome (Supplemental Table S6 and S7).

Genes associated with GPCR/adrenergic and glucocorticoid receptor signaling

Pathways involved with glucocorticoid receptor signaling (GR; Fig. 4B) and adrenergic signaling (Fig. 4C) were modified by ozone in the transcriptome of multiple tissues (Supplemental Table S17-S19). Across all tissues, GPCR signaling pathway had 314 DEG (Supplemental Table S17), adrenoreceptor signaling had 6 DEG (Supplemental Table S18),, and general GR signaling had 44 DEG (Supplemental Table S19.

Serum metabolomics found that numerous markers of adrenergic and, glucocorticoid signaling, as well as lipoxygenases such as 12-HEPE, and 12-HETE were altered by ozone exposure (Supplemental Table S1).

Metabolic pathways

Steroid-associated metabolic pathways were significantly altered in the transcriptome of numerous tissues. There were 60 DEG related to the androgen receptor signaling pathway (Supplemental Table S20), 124 DEG related to the early estrogen response (Supplemental Table S22), and 42 DEG related to FGFR signaling (Supplemental Table S21) across all tissues. It is noteworthy that some of the DEGs identified in all 6 tissues are not the same gene transcripts, rather those include the tissue-specific gene sets induced by identified steroid-associated metabolic pathways.

Stress-associated metabolic and immune pathways were significantly altered in the transcriptome of numerous tissues. There were 151 DEG related to TNFα and NFκB signaling pathway (Supplemental Table S23), 151 DEG related to carbohydrate metabolism (Supplemental Table S24), and 348 DEG related to lipid metabolism (Supplemental Table S25) across all tissues.

Pathway analysis of serum metabolomics revealed significant overenrichment in subpathways associated with fatty acid metabolism (EV = 1.96, 6 of 6 metabolites), long chain polyunsaturated fatty acids (EV = 1.87, 17 of18 metabolites), long chain saturated fatty acids (EV = 1.96, 7 of 7 metabolites), long chain monounsaturated fatty acids (EV = 1.68, 6 of 7 metabolites), and diacylglycerol (EV = 1.43, 21 of 29 metabolites; Supplemental Table S2).

Multi-omics analysis of the transcriptome and metabolome identified Reactome pathways related to the metabolism of RNA and rRNA processing that were significantly altered by ozone exposure in adipose, hypothalamus, and muscle and shared with the circulating metabolome (Supplemental Tables S6 and S7). This is likely reflective of the fact that during an acute stress response, rRNA processing is reduced (Grummt, 2013). Further, in the hypothalamus transcriptome, ozone exposure resulted in shared responses with the circulating metabolome on the metabolism reactome pathway and numerous pathways involved in metabolism of proteins and cellular responses to stimuli and stress (Supplemental Table S6 and S7).

Stress immune pathways

Stress-associated immune pathways were disrupted in the transcriptome of multiple tissues. Across all tissues, there were 42 DEG related to chemokines (Supplemental Table S27), 9 DEG related to the acute inflammatory response (Supplemental Table S26), and 53 DEG related to immunoregulatory interactions (Supplemental Table S28).

In addition to the previously mentioned adrenergic signaling metabolites, numerous glutamate and glutathione serum metabolites were significantly modified by ozone exposure (Supplemental Table S1).

Although air pollution exposures are associated with multiorgan toxicity, the mechanisms by which pollutants, upon encountering the lung, trigger substantial systemic and distal effects is poorly understood. We have previously demonstrated that activation of the neuroendocrine system is critical for the systemic response (Henriquez et al., 2018, 2021; Henriquez, Snow, Dye, Schladweiler, Alewel, et al., 2022a, 2022b; Henriquez, Snow, Jackson, House, Alewel, et al., 2022; Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022; Jackson et al., 2022; Miller, Snow, et al., 2016). Here, we analyzed multi-tissue transcriptional signatures and circulating metabolites following an acute ozone exposure to reveal tissue-specific effects and potential mechanistic insights on shared biological mediators. Although each tissue responded uniquely to ozone exposure, conserved pathway-level responses provide insights into the underlying mechanisms of systemic ozone toxicity.

Transcriptome analysis identified strongest clustering by tissue (lung, hypothalamus, adrenals, liver, adipose, and muscle) representing the highly specific basal tissue expression reflective of homeostatic function of a given organ (Lilja et al., 2023). Assessment of ozone-induced differentially expressed genes revealed that the greatest impact of ozone was not in lung, but in skeletal muscle, with ozone-induced transcriptomic changes in order of muscle (5,102 DEG), adrenal (2,516 DEG), lung (1,640 DEG), liver (1,333 DEG), adipose (1,242 DEG), and hypothalamus (274 DEG). Comparative assessment of multi-tissue transcriptional response revealed that only six genes were changed in common between all 6 tissues, with additional overlap identified with combinations of 2–5 tissues. Functional differences between organ systems, potentially stemming from differences in adrenergic and glucocorticoid receptor distribution or activation, are causing tissue-specific responses to the ozone stressor (Kalinyak et al., 1987; Paik et al., 2020). The few genes commonly changed by ozone in all tissues include several responsible for responding to cell stress, including Cdkn1a, involved in DNA damage response and cell cycle progression (Dutto et al., 2015), Gpatch4, important in nucleosome response to stress (Hirawake-Mogi et al., 2021), Mt1m and Mt2a, responders to metal and non-metal stress (Miyazaki & Asanuma, 2023), and Pcdh12, critical in cell adhesion (Rampon et al., 2005). Because ozone itself is unlikely to translocate beyond the lung upon inhalation, the conserved response across diverse organ systems is likely due to a systemic stress response. Systemic transcriptomic and circulating metabolomic analysis suggest that hypoxia is likely a primary driver of the ozone-induced systemic stress response, in conjunction with glucocorticoid and GPCR signaling.

Reactome and Hallmark pathway transcriptome analyses revealed common pathways enriched across all tissues, including hypoxia, TNF-mediated NFkB signaling, and steroid hormone (including glucocorticoid, androgen, and estrogen) activity, suggesting the contribution of neurotransmitters and hormones in mediating common effects. These tissue-specific transcriptional response corroborated > 51% of serum metabolites changed in ozone-exposed animals consistent with previous work that found substantial dysregulation of the serum metabolome in ozone-exposed WKY rats (Miller et al., 2015). These data suggest that the tissue-specific and systemic metabolic effects of ozone involve common triggers through a stress response. Our previous studies have demonstrated the involvement of pituitary- and adrenal-derived hormones in mediating ozone-induced stress response (Henriquez et al., 2019, 2021; Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022). Tissue-specific enrichment of most pathways involved metabolic regulation by amino acids, inflammatory processes, cell cycle control, and m-TORC signaling, and further implies that the response generated by the activation of ozone-induced neuroendocrine stress results in tissue-specific modulation of cellular pathways. Adrenal-derived stress hormones effects are likely modulated by tissue-specific receptor subtype distribution and density, as well as regulatory mechanisms that differentially modulate cellular activity and transcriptional processes. The contribution of the stress-induced tissue-specific and general processes in ultimate pathology will need to be taken into consideration when examining long-term impacts of stressors.

To understand how ozone-induced release of neuroendocrine hormones affect each tissue, we performed upstream predictor analysis, which takes publicly available expression data of myriad chemicals and compares transcriptome results to those identified in the current study. The extensiveness of transcriptional changes in each tissue correlated with the list of predicted similarities to other chemicals/molecules as activated or inhibited. In all tissues, the list of predicted transcriptional regulators were substantial, suggesting major signaling events that involved mediation through transcriptional regulators. The most common pathways predicted to be upregulated included those induced by steroidal hormones and cAMP-activating chemicals such as forskolin, suggesting the activation of glucocorticoid and adrenergic signaling in all tissues. Specifically in the lung, ozone’s target organ, forskolin was predicted to be strongly activated along with steroidal drugs and inflammatory mediators. The prediction of steroidal hormones and forskolin as activated, even in the hypothalamus, suggests that both glucocorticoids and catecholamines have significant impacts on regulation of the dynamicity of ozone-induced stress response, consistent with other stress conditions (Bouchez et al., 2012; Joëls et al., 2012). Significant ozone effects on genes involved in interferon signaling, together with the prediction of inhibition of interferon action in most tissues examined, emphasizes that increased circulating glucocorticoids cause generalized immune suppression systemically (Shimba & Ikuta, 2020).

Glucocorticoids are released into circulation following a single ozone exposure in a dynamic manner along with catecholamines (Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022) and therefore, we assessed if circulating glucocorticoids and/or catecholamines mediate tissue-specific changes in the lung and in other tissues. There are cell- and tissue-specific distributions of GRs and adrenergic GPCRs (Barnes, 2004; Hodge et al., 2021), which could underlie how ozone might induce cellular changes through activation of these receptors. Receptor binding activates transcriptional machinery with a variety of chaperone proteins and can cause transcriptional repression and/or enhancement (Oakley & Cidlowski, 2013; Xavier et al., 2016). The transcriptional activity of glucocorticoids and/or catecholamines regulating metabolic and immune-related cellular signaling might show consistency in gene changes across tissues based on the evidence that GR are distributed in virtually all tissues and cells. However, Reactome and Hallmark pathway analysis revealed remarkable tissue-specificity in ozone effects on glucocorticoid- and catecholamine- regulated genes, despite identification of glucocorticoid-modulating chemicals as upstream mediators across all tissues, which emphasizes that glucocorticoid and catecholamine transcriptional activities are multifaceted. Interestingly, GPCR signaling, which modulates cyclic-AMP-mediated action (Wright et al., 2015), was changed only in the lung and in the hypothalamus, suggesting its potential contribution in inflammation in the lung and in feedback inhibition of a stress response within hypothalamus. Pathway analysis indicated that genes linked to adrenergic/GPCR signaling were widely impacted and were identified in a variety of immune signaling pathways but did not show coherent tissue-specific patterns. Further, although immunosuppression through inhibition of interferon signaling genes was common across multiple tissues, many glucocorticoid-signaling processes were tissue-specific, suggesting tissue-level regulation.

Ozone induces inflammation in the lung and associated immune changes in immunomodulatory tissues such as spleen, thymus and lymphatics, in addition to causing lymphopenia and monocyte changes in circulation (Feng et al., 2006; Francis et al., 2017; Li & Richters, 1991). Assessing pathway-specific heatmaps of 6 tissues, lung, adrenal, and adipose tissue showed upregulation of more immunomodulatory genes, whereas in the liver and muscle, many of the same genes were downregulated. Tissue-level regulation of immune response to stress could vary mechanistically and/or temporally through different cellular and transcriptional control mechanisms across tissues even though both, catecholamines and glucocorticoids likely contribute to these effects. The duality of stress hormones impact on physiological processes and differences in stress-impacted transcriptional processes might contribute to upregulation of genes in one tissue concomitant with downregulation of the same genes in other. Nevertheless, it was apparent that a variety of kinases involved in signaling processes linked to inflammatory mechanisms appeared to be a generalized target of signaling (Meng et al., 2023; Sasse et al., 2019). It would be worthwhile to explore the tissue-level response of immune tissues including thymus and spleen in future studies.

Consistent with our prior study (Miller et al., 2015), numerous metabolic pathways were modified in the circulating metabolome by acute ozone exposure. Pathways related to fatty acid metabolism, and specifically polyunsaturated fatty acids, were significantly over-enriched, suggesting that fatty acids were mobilized during the acute stress response to provide energy for the “fight-or-flight” response (Prentice et al., 2019). Adipose lipolysis was likely induced since Reactome pathway analysis demonstrated significant increases in FGFR-associated genes to be upregulated in adipose tissue, which is known to stimulate lipolysis (Hotta et al., 2009). Considering that the transcriptomic response of pathways involving fatty acids, lipids and lipoproteins, and amino acids all occurred predominantly in the liver, it is likely that the timing of this assessment (immediately after the acute 4-hour exposure) contributed to the hepatic response as the liver is in the process of responding to the altered circulating metabolites. We have shown that hormonal changes occur in a dynamic manner after ozone exposure (Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022) and the dynamicity of cellular processes will need to be considered in future studies that involve stress mechanisms.

As noted previously, inhibition of multiple miRNA were predicted for lung and liver (Jackson et al., 2022), but surprisingly, there were dozens of miRNA predicted to be activated in muscle. The robust prediction of miRNA in muscle suggests influence by locally-produced miRNAs. Skeletal muscle has been recently shown to release miRNAs in the tissue microenvironment, but not systemically, that regulate local myogenesis (Watanabe et al., 2022). Incidentally, notable changes of glucocorticoid signaling, which induces muscle atrophy through NFkB pathway activation (Fry et al., 2016), was supported by remarkable changes in ozone-induced NFkB-TNF signaling genes in muscle (Supplemental Table S4). However, the predicted involvement of steroidal hormones (deoxycorticosterone acetate, desoxycorticosterone, norepinephrine), together with inhibitory effects of norepinephrine linked to SAM activation in muscle, could also explain how muscle played a distinct role in supporting the stress response. The other neural or hormonal influence, or even overriding effect of the parasympathetic system over generalized sympathetic activation, could contribute to remarkable transcriptional changes that occur in skeletal muscle after ozone exposure. The predicted inhibition of levothyroxine was consistent with depletion of TSH and T3/T4 after ozone exposure in rodents (Clemons & Wei, 1984; Henriquez et al., 2019). The role of muscle in observed hypothermia during ozone exposure (Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022) is of further interest as a regulatory mechanism. Metabolomic changes associated with the release of branched chain amino acids in circulation suggest the contribution of muscle protein catabolism.

Considering the transcriptomic and metabolomic changes together, it is likely that animals exposed to ozone for 4 hours experienced changes in the dynamics of oxygen levels in blood and at distal tissues. Hypoxia can lead to the activation of the neuroendocrine axis (Fournier et al., 2007; Mikhailenko et al., 2009), and it is likely that ozone exposure caused acute hypoxia and contributed to the systemic stress response. Ozone-induced transcriptomic changes in the Hallmark pathway “Genes up-regulated in response to low oxygen levels (hypoxia)” in all 6 tissues were over-represented, suggesting that the hypoxic response was systemic and not specific to any given tissues. Serum metabolomic signatures in our study further supported this concept, consistent with reported hypoxia-induced changes to the gut microbial populations of laboratory mice. Intermittent hypoxic exposure in laboratory mice has been documented to induce changes in the gut microbial community (Moreno-Indias et al., 2015). Metabolites identified in the serum may be a product of either host or microbial metabolism. For example, significant alteration of secondary bile acids between groups may be indicative of alterations to the gut microbial environment, considering these metabolites are produced through the action of gut microbes on primary bile acids, but these metabolites can also be recycled through enterohepatic recirculation. In response to acute ozone exposure, several benzoate and aromatic amino acid metabolites, as well as several plant and food components were significantly altered in the serum. Ozone exposure has been demonstrated to induce changes in gut microbiota (Tashiro & Shore, 2021). Considering dietary input was consistent across all groups until the beginning of exposure, which was followed by ~ 5–6 hours fasting prior to necropsy, alterations to these metabolites between exposure groups, along with changes to secondary bile acids in the serum is consistent with alterations in the gut microbial environment, as we have previously reported (Miller et al., 2015). Our transcriptomic and metabolomic analyses support the potential contribution of hypoxia as a mediator of gut microbiota-linked metabolite changes.

Polyamine metabolites were also significantly altered between animals that were exposed to ozone compared to those that were not. Polyamines are produced in most mammalian cells but are also generated by gut bacteria from dietary arginine. Polyamine accumulation in the lung has been previously investigated in response to ozone, and high levels of polyamines are also present in serum of asthmatics during exacerbations (Elsayed et al., 1990; Kurosawa et al., 1992; North et al., 2013). Together, alterations of these metabolites in the serum of ozone exposed rats indicate that acute ozone exposure can influence lung function and oxygen availability, and these pathways can further be influenced by changes to the gut microbial environment.

Although the ozone concentration employed here is above that encountered environmentally, this concentration in resting rats is comparable to concentrations used in clinical studies where humans are exposed during intermittent exercise which leads to increase in the lung dosimetry (Hatch et al., 1994). Acute 0.8ppm ozone exposure can be used as a tool to gain mechanistic understanding of the systemic response to stress. As we have observed previously, the ozone-induced stress response is dynamic and reversible, despite continued exposure (Henriquez, Snow, Dye, Schladweiler, Alewel, et al., 2022b; Henriquez, Snow, Jackson, House, Motsinger-Reif, et al., 2022). The assessment of temporality of changes is necessary in understanding how impairment of stress response centrally or peripherally at tissue level might be linked to susceptibility variations.

Often, toxicological studies focus on limited endpoints and prioritize the assessment of a single tissue or organ system. The data presented here demonstrate that the transcriptomic response to an acute ozone stressor is widespread across all organ systems, is incredibly tissue-specific, and that assessment of only a single tissue is likely to overlook critical endpoints. Further, analysis of circulating metabolomic data identified changes that were corroborated with transcriptional changes in metabolically active organs and identified additional changes that were not directly captured by transcriptional changes. The ozone-induced transcriptional changes reflected the involvement of common mediators of transcriptional response among many tissues, such as hypoxia, GPCR signaling, and glucocorticoid signaling. However, these changes were associated with tissue-specific alterations in gene networks. This pattern of changes suggests that an inhaled pollutant-induced stress response has organ-specific mechanisms for their transcriptional activities resulting in the activation/suppression of different sets of genes allowing a coherent homeostatic regulation tailored to the specific stress condition. By analyzing multi-tissue transcriptome and integrating with the circulating metabolome, we are better able to understand the dynamicity of this response and envision how inhaled pollutants could be contributing to diseases of distant organs. The homeostatic stress response, mediating both neural and peripheral signals, could be impaired in individuals with neuropsychiatric and/or chronic diseases, thereby exacerbating these ailments when individuals are exposed to air pollutants.

Declaration of Conflicting interests:

KMPJ is an employee of Metabolon Inc (Morrisvlle, NC) and, as such, has an affiliation with or financial involvement with Metabolon, Inc.

Acknowledgements:

The authors would like to thank Drs. Jacqueline Bangma, Steve Gavett, and Ian Gilmour of the U.S. Environmental Protection Agency and Dr. Darryl C. Zeldin of the National Institute of Environmental Health Sciences for their critical review of the manuscript. We acknowledge the help of Dr. Mark Higuchi for performing inhalation exposures and Mr. Abdul Malek Khan in finalizing ozone exposure parameters.

Funding:

This research was supported by the intramural research program of the U.S. Environmental Protection Agency (EPA). Partial support also came through an appointment of T.W.J., A.R.H., and D.I.A. to the U.S. EPA Research Participation Program administered by the Oak Ridge Institute for Science and Education (ORISE) through an interagency agreement between the U.S. Department of Energy (DOE) and the U.S. EPA. ORISE is managed by ORAU under DOE contract number DE-SC0014664. In addition, this research was supported in part by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences.

Author Contributions:

Conceptualization: T.W.J, J.S.H, A.R.H, U.P.K.; Methodology: T.W.J., J.S.H., A.R.H., M.C.S., K.M.P.J., A.A.F., S.J.S., D.I.A., A.M-R., U.P.K.; Software: T.W.J., J.S.H.; Validation: T.W.J., J.S.H., A.R.H., M.C.S., U.P.K.; Formal analysis: T.W.J.; Investigation: A.R.H., M.C.S., U.P.K., Resources: T.W.J., J.S.H., M.C.S., U.P.K.; Data Curation: T.W.J., J.S.H., A.R.H., M.C.S., A.A.F., U.P.K.; Writing – original draft: T.W.J., J.S.H., U.P.K.; Writing – review & editing: T.W.J., J.S.H., A.R.H., M.C.S., K.M.P.J., A.A.F., S.J.S., D.I.A., A.M-R., U.P.K; Visualization: T.W.J., J.S.H., U.P.K; Supervision: U.P.K; Project administration: U.P.K; Funding acquisition: U.P.K.

Data availability:

All data and code are available either in supplemental tables, publicly accessible via repositories (transcriptomics: GEO accession GSE234042) or via request to the corresponding author.

Thomas W. Jackson - https://orcid.org/0000-0002-7996-0412

John S. House - https://orcid.org/0000-0002-8447-7871

Andres R. Henriquez - https://orcid.org/0000-0002-1917-4153

Mette C. Schladweiler – https://orcid.org/0009-0005-4562-4479

Samantha J. Snow - https://orcid.org/0000-0003-1812-8582

Devin I. Alewel - https://orcid.org/0000-0003-3415-9477

Allison Motsinger-Reif - https://orcid.org/0000-0003-1346-2493

Urmila P. Kodavanti - https://orcid.org/0000-0001-6333-1024

Disclaimer: The research described in this article has been reviewed by the U.S. Environmental Protection Agency (EPA) and the National Institute of Environmental Health Sciences (NIEHS) and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the EPA, NIH, NIEHS, or the U.S. Government, nor does the mention of trade names, commercial products, or organizations constitute endorsement or recommendation for use.

Barnes, P. J. (2004). Distribution of receptor targets in the lung. Proceedings of the American Thoracic Society, 1(4), 345–351. https://doi.org/10.1513/pats.200409-045MS.
Bouchez, G., Millan, M. J., Rivet, J. M., Billiras, R., Boulanger, R., & Gobert, A. (2012). Quantification of extracellular levels of corticosterone in the basolateral amygdaloid complex of freely-moving rats: A dialysis study of circadian variation and stress-induced modulation. Brain Research, 1452, 47–60. https://doi.org/10.1016/j.brainres.2012.01.010.
Bridgewater, B. R., E. A (2014). Spectrometry in High-Throughput Profiling Metabolomics. Journal of Postgenomics Drug & Biomarker Development, 04(02), https://doi.org/10.4172/2153-0769.1000132. High Resolution Mass Spectrometry Improves Data Quantity and Quality as Compared to Unit Mass Resolution Mass.
Clemons, G. K., & Wei, D. (1984). Effect of short-term ozone exposure on exogenous thyroxine levels in thyroidectomized and hypophysectomized rats. Toxicology and Applied Pharmacology, 74(1), 86–90. https://doi.org/10.1016/0041-008x(84)90273-4.
Conway, J. R., Lex, A., & Gehlenborg, N. (2017). UpSetR: An R package for the visualization of intersecting sets and their properties. Bioinformatics (Oxford England), 33(18), 2938–2940. https://doi.org/10.1093/bioinformatics/btx364.
de Kloet, E. R., & Joëls, M. (2023). The cortisol switch between vulnerability and resilience. Molecular Psychiatry. https://doi.org/10.1038/s41380-022-01934-8.
DeHaven, C. D., Evans, A. M., Dai, H., & Lawton, K. A. (2010). Organization of GC/MS and LC/MS metabolomics data into chemical libraries. Journal of Cheminformatics, 2(1), 9. https://doi.org/10.1186/1758-2946-2-9.
Dutto, I., Tillhon, M., Cazzalini, O., Stivala, L. A., & Prosperi, E. (2015). Biology of the cell cycle inhibitor p21(CDKN1A): Molecular mechanisms and relevance in chemical toxicology. Archives of Toxicology, 89(2), 155–178. https://doi.org/10.1007/s00204-014-1430-4.
Elsayed, N. M., Ellingson, A. S., Tierney, D. F., & Mustafa, M. G. (1990). Effects of ozone inhalation on polyamine metabolism and tritiated thymidine incorporation into DNA of rat lungs. Toxicology and Applied Pharmacology, 102(1), 1–8. https://doi.org/10.1016/0041-008x(90)90078-9.
Everds, N. E., Snyder, P. W., Bailey, K. L., Bolon, B., Creasy, D. M., Foley, G. L., Rosol, T. J., & Sellers, T. (2013). Interpreting stress responses during routine toxicity studies: A review of the biology, impact, and assessment. Toxicologic Pathology, 41(4), 560–614. https://doi.org/10.1177/0192623312466452.
Fabregat, A., Korninger, F., Viteri, G., Sidiropoulos, K., Marin-Garcia, P., Ping, P., Wu, G., Stein, L., D’Eustachio, P., & Hermjakob, H. (2018). Reactome graph database: Efficient access to complex pathway data. PLoS Computational Biology, 14(1), e1005968. https://doi.org/10.1371/journal.pcbi.1005968.
Fang, H., Knezevic, B., Burnham, K. L., & Knight, J. C. (2016). XGR software for enhanced interpretation of genomic summary data, illustrated by application to immunological traits. Genome Medicine, 8(1), 129. https://doi.org/10.1186/s13073-016-0384-y.
Feng, R., He, W., Ochi, H., & Castranova, V. (2006). Ozone exposure impairs antigen-specific immunity but activates IL-7-induced proliferation of CD4-CD8- thymocytes in BALB/c mice. Journal of Toxicology and Environmental Health Part A, 69(16), 1511–1526. https://doi.org/10.1080/15287390500468696.
Ford, L., Kennedy, A. D., Goodman, K. D., Pappan, K. L., Evans, A. M., Miller, L. A. D., Wulff, J. E., Wiggs, B. R., Lennon, J. J., Elsea, S., & Toal, D. R. (2020). Precision of a Clinical Metabolomics Profiling Platform for Use in the Identification of Inborn Errors of Metabolism. The Journal of Applied Laboratory Medicine, 5(2), 342–356. https://doi.org/10.1093/jalm/jfz026.
Fournier, S., Allard, M., Gulemetova, R., Joseph, V., & Kinkead, R. (2007). Chronic corticosterone elevation and sex-specific augmentation of the hypoxic ventilatory response in awake rats. The Journal of Physiology, 584(Pt 3), 951–962. https://doi.org/10.1113/jphysiol.2007.141655.
Francis, M., Sun, R., Cervelli, J. A., Choi, H., Mandal, M., Abramova, E. V., Gow, A. J., Laskin, J. D., & Laskin, D. L. (Eds.). (2017). Editor’s Highlight: Role of Spleen-Derived Macrophages in Ozone-Induced Lung Inflammation and Injury. Toxicological Sciences: An Official Journal of the Society of Toxicology, 155(1), 182–195. https://doi.org/10.1093/toxsci/kfw192.
Fry, C. S., Nayeem, S. Z., Dillon, E. L., Sarkar, P. S., Tumurbaatar, B., Urban, R. J., Wright, T. J., Sheffield-Moore, M., Tilton, R. G., & Choudhary, S. (2016). Glucocorticoids increase skeletal muscle NF-κB inducing kinase (NIK): Links to muscle atrophy. Physiological Reports, 4(21), e13014. https://doi.org/10.14814/phy2.13014.
Grummt, I. (2013). The nucleolus—Guardian of cellular homeostasis and genome integrity. Chromosoma, 122(6), 487–497. https://doi.org/10.1007/s00412-013-0430-0.
Han, N., Pan, Z., Liu, G., Yang, R., & Yujing, B. (2021). Hypoxia: The “Invisible Pusher” of Gut Microbiota. Frontiers in Microbiology, 12, 690600. https://doi.org/10.3389/fmicb.2021.690600.
Hatch, G. E., Slade, R., Harris, L. P., McDonnell, W. F., Devlin, R. B., Koren, H. S., Costa, D. L., & McKee, J. (1994). Ozone dose and effect in humans and rats. A comparison using oxygen-18 labeling and bronchoalveolar lavage. American Journal of Respiratory and Critical Care Medicine, 150(3), 676–683. https://doi.org/10.1164/ajrccm.150.3.8087337.
Henriquez, A. R., House, J. S., Snow, S. J., Miller, C. N., Schladweiler, M. C., Fisher, A., Ren, H., Valdez, M., Kodavanti, P. R., & Kodavanti, U. P. (2019). Ozone-induced dysregulation of neuroendocrine axes requires adrenal-derived stress hormones. Toxicological Sciences: An Official Journal of the Society of Toxicology, kfz182. https://doi.org/10.1093/toxsci/kfz182.
Henriquez, A. R., Snow, S. J., Dye, J. A., Schladweiler, M. C., Alewel, D. I., Miller, C. N., & Kodavanti, U. P. (2022a). The contribution of neuroendocrine system in adaptation after repeated daily ozone exposure in rats (preprint). BioRxiv.
Henriquez, A. R., Snow, S. J., Dye, J. A., Schladweiler, M. C., Alewel, D. I., Miller, C. N., & Kodavanti, U. P. (2022b). The contribution of the neuroendocrine system to adaption after repeated daily ozone exposure in rats. Toxicology and Applied Pharmacology, 447, 116085. https://doi.org/10.1016/j.taap.2022.116085.
Henriquez, A. R., Snow, S. J., Jackson, T. W., House, J. S., Alewel, D. I., Schladweiler, M. C., Valdez, M. C., Freeborn, D. L., Miller, C. N., Grindstaff, R., Kodavanti, P. R. S., & Kodavanti, U. P. (2022). Social isolation exacerbates acute ozone inhalation induced pulmonary and systemic health outcomes. Toxicology and Applied Pharmacology, 457, 116295. https://doi.org/10.1016/j.taap.2022.116295.
Henriquez, A. R., Snow, S. J., Jackson, T. W., House, J. S., Motsinger-Reif, A. A., Ward-Caviness, C. K., Schladweiler, M. C., Alewel, D. I., Miller, C. N., Farraj, A. K., Hazari, M. S., Grindstaff, R., Diaz-Sanchez, D., Ghio, A. J., & Kodavanti, U. P. (2022). Stress Drivers of Glucose Dynamics during Ozone Exposure Measured Using Radiotelemetry in Rats. Environmental Health Perspectives, 130(12), 127006. https://doi.org/10.1289/EHP11088.
Henriquez, A. R., Snow, S. J., Schladweiler, M. C., Miller, C. N., Dye, J. A., Ledbetter, A. D., Richards, J. E., Hargrove, M. M., Williams, W. C., & Kodavanti, U. P. (2018). Beta-2 Adrenergic and Glucocorticoid Receptor Agonists Modulate Ozone-Induced Pulmonary Protein Leakage and Inflammation in Healthy and Adrenalectomized Rats. Toxicological Sciences: An Official Journal of the Society of Toxicology, 166(2), 288–305. https://doi.org/10.1093/toxsci/kfy198.
Henriquez, A. R., Williams, W., Snow, S. J., Schladweiler, M. C., Fisher, C., Hargrove, M. M., Alewel, D., Colonna, C., Gavett, S. H., Miller, C. N., & Kodavanti, U. P. (2021). The dynamicity of acute ozone-induced systemic leukocyte trafficking and adrenal-derived stress hormones. Toxicology, 458, 152823. https://doi.org/10.1016/j.tox.2021.152823.
Hirawake-Mogi, H., Thanh Nhan, N. T., & Okuwaki, M. (2021). G-patch domain-containing protein 4 localizes to both the nucleoli and Cajal bodies and regulates cell growth and nucleolar structure. Biochemical and Biophysical Research Communications, 559, 99–105. https://doi.org/10.1016/j.bbrc.2021.04.026.
Hodge, M. X., Henriquez, A. R., & Kodavanti, U. P. (2021). Adrenergic and Glucocorticoid Receptors in the Pulmonary Health Effects of Air Pollution. Toxics, 9(6), 132. https://doi.org/10.3390/toxics9060132.
Hotta, Y., Nakamura, H., Konishi, M., Murata, Y., Takagi, H., Matsumura, S., Inoue, K., Fushiki, T., & Itoh, N. (2009). Fibroblast growth factor 21 regulates lipolysis in white adipose tissue but is not required for ketogenesis and triglyceride clearance in liver. Endocrinology, 150(10), 4625–4633. https://doi.org/10.1210/en.2009-0119.
Jackson, T. W., Henriquez, A. R., Snow, S. J., Schladweiler, M. C., Fisher, A. A., Alewel, D. I., House, J. S., & Kodavanti, U. P. (2022). Adrenal stress hormone regulation of hepatic homeostatic function after an acute ozone exposure in Wistar-Kyoto male rats. Toxicological Sciences: An Official Journal of the Society of Toxicology, kfac065. https://doi.org/10.1093/toxsci/kfac065.
Joëls, M., Karst, H., & Sarabdjitsingh, R. A. (2018). The stressed brain of humans and rodents. Acta Physiologica (Oxford England), 223(2), e13066. https://doi.org/10.1111/apha.13066.
Joëls, M., Sarabdjitsingh, R. A., & Karst, H. (2012). Unraveling the time domains of corticosteroid hormone influences on brain activity: Rapid, slow, and chronic modes. Pharmacological Reviews, 64(4), 901–938. https://doi.org/10.1124/pr.112.005892.
Kalinyak, J. E., Dorin, R. I., Hoffman, A. R., & Perlman, A. J. (1987). Tissue-specific regulation of glucocorticoid receptor mRNA by dexamethasone. The Journal of Biological Chemistry, 262(22), 10441–10444.
Kanehisa, M., & Goto, S. (2000). KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Research, 28(1), 27–30. https://doi.org/10.1093/nar/28.1.27.
Kodavanti, U. P., Jackson, T. W., Henriquez, A. R., Snow, S. J., Alewel, D. I., & Costa, D. L. (2023). Air Pollutant impacts on the brain and neuroendocrine system with implications for peripheral organs: A perspective. Inhalation Toxicology, 35(3–4), 109–126. https://doi.org/10.1080/08958378.2023.2172486.
Krämer, A., Green, J., Pollard, J., & Tugendreich, S. (2014). Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics (Oxford England), 30(4), 523–530. https://doi.org/10.1093/bioinformatics/btt703.
Kurosawa, M., Shimizu, Y., Tsukagoshi, H., & Ueki, M. (1992). Elevated levels of peripheral-blood, naturally occurring aliphatic polyamines in bronchial asthmatic patients with active symptoms. Allergy, 47(6), 638–643. https://doi.org/10.1111/j.1398-9995.1992.tb02388.x.
Li, A. F., & Richters, A. (1991). Ambient level ozone effects on subpopulations of thymocytes and spleen T lymphocytes. Archives of Environmental Health, 46(1), 57–63. https://doi.org/10.1080/00039896.1991.9937430.
Lilja, S., Li, X., Smelik, M., Lee, E. J., Loscalzo, J., Marthanda, P. B., Hu, L., Magnusson, M., Sysoev, O., Zhang, H., Zhao, Y., Sjöwall, C., Gawel, D., Wang, H., & Benson, M. (2023). Multi-organ single-cell analysis reveals an on/off switch system with potential for personalized treatment of immunological diseases. Cell Reports Medicine, 4(3), 100956. https://doi.org/10.1016/j.xcrm.2023.100956.
Love, M. I., Huber, W., & Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550. https://doi.org/10.1186/s13059-014-0550-8.
Meng, Q., Bai, M., Guo, M., Li, Z., Liu, W., Fan, X., Sun, R., Yang, X., Yuan, D., Shi, Y., & Deng, H. (2023). Inhibition of Serum- and Glucocorticoid-Regulated Protein Kinase-1 Aggravates Imiquimod-Induced Psoriatic Dermatitis and Enhances Proinflammatory Cytokine Expression through the NF-kB Pathway. The Journal of Investigative Dermatology, S0022-202X(23)00007 – 6. https://doi.org/10.1016/j.jid.2022.12.013.
Mikhailenko, V. A., Butkevich, I. P., Bagaeva, T. R., Makukhina, G. V., & Otellin, V. A. (2009). Short- and long-term influences of hypoxia during early postnatal period of development on behavioral and hormonal responses in rats. Neuroscience Letters, 464(3), 214–217. https://doi.org/10.1016/j.neulet.2009.08.047.
Miller, D. B., Ghio, A. J., Karoly, E. D., Bell, L. N., Snow, S. J., Madden, M. C., Soukup, J., Cascio, W. E., Gilmour, M. I., & Kodavanti, U. P. (2016). Ozone Exposure Increases Circulating Stress Hormones and Lipid Metabolites in Humans. American Journal of Respiratory and Critical Care Medicine, 193(12), 1382–1391. https://doi.org/10.1164/rccm.201508-1599OC.
Miller, D. B., Karoly, E. D., Jones, J. C., Ward, W. O., Vallanat, B. D., Andrews, D. L., Schladweiler, M. C., Snow, S. J., Bass, V. L., Richards, J. E., Ghio, A. J., Cascio, W. E., Ledbetter, A. D., & Kodavanti, U. P. (2015). Inhaled ozone (O3)-induces changes in serum metabolomic and liver transcriptomic profiles in rats. Toxicology and Applied Pharmacology, 286(2), 65–79. https://doi.org/10.1016/j.taap.2015.03.025.
Miller, D. B., Snow, S. J., Schladweiler, M. C., Richards, J. E., Ghio, A. J., Ledbetter, A. D., & Kodavanti, U. P. (2016). Acute Ozone-Induced Pulmonary and Systemic Metabolic Effects Are Diminished in Adrenalectomized Rats. Toxicological Sciences: An Official Journal of the Society of Toxicology, 150(2), 312–322. https://doi.org/10.1093/toxsci/kfv331.
Miyazaki, I., & Asanuma, M. (2023). Multifunctional Metallothioneins as a Target for Neuroprotection in Parkinson’s Disease. Antioxidants (Basel Switzerland), 12(4), 894. https://doi.org/10.3390/antiox12040894.
Moreno-Indias, I., Torres, M., Montserrat, J. M., Sanchez-Alcoholado, L., Cardona, F., Tinahones, F. J., Gozal, D., Poroyko, V. A., Navajas, D., Queipo-Ortuño, M. I., & Farré, R. (2015). Intermittent hypoxia alters gut microbiota diversity in a mouse model of sleep apnoea. The European Respiratory Journal, 45(4), 1055–1065. https://doi.org/10.1183/09031936.00184314.
North, M. L., Grasemann, H., Khanna, N., Inman, M. D., Gauvreau, G. M., & Scott, J. A. (2013). Increased ornithine-derived polyamines cause airway hyperresponsiveness in a mouse model of asthma. American Journal of Respiratory Cell and Molecular Biology, 48(6), 694–702. https://doi.org/10.1165/rcmb.2012-0323OC.
Oakley, R. H., & Cidlowski, J. A. (2013). The biology of the glucocorticoid receptor: New signaling mechanisms in health and disease. The Journal of Allergy and Clinical Immunology, 132(5), 1033–1044. https://doi.org/10.1016/j.jaci.2013.09.007.
Paik, D. T., Tian, L., Williams, I. M., Rhee, S., Zhang, H., Liu, C., Mishra, R., Wu, S. M., Red-Horse, K., & Wu, J. C. (2020). Single-Cell RNA Sequencing Unveils Unique Transcriptomic Signatures of Organ-Specific Endothelial Cells. Circulation, 142(19), 1848–1862. https://doi.org/10.1161/CIRCULATIONAHA.119.041433.
Perdigones, B. C., Lee, S., Cohen, R. C., Park, J. H., & Min, K. E. (2022). Two Decades of Changes in Summertime Ozone Production in California’s South Coast Air Basin. Environmental Science & Technology, 56(15), 10586–10595. https://doi.org/10.1021/acs.est.2c01026.
Prentice, K. J., Saksi, J., & Hotamisligil, G. S. (2019). Adipokine FABP4 integrates energy stores and counterregulatory metabolic responses. Journal of Lipid Research, 60(4), 734–740. https://doi.org/10.1194/jlr.S091793.
Rampon, C., Prandini, M. H., Bouillot, S., Pointu, H., Tillet, E., Frank, R., Vernet, M., & Huber, P. (2005). Protocadherin 12 (VE-cadherin 2) is expressed in endothelial, trophoblast, and mesangial cells. Experimental Cell Research, 302(1), 48–60. https://doi.org/10.1016/j.yexcr.2004.08.024.
Reuben, A., Manczak, E. M., Cabrera, L. Y., Alegria, M., Bucher, M. L., Freeman, E. C., Miller, G. W., Solomon, G. M., & Perry, M. J. (2022). The Interplay of Environmental Exposures and Mental Health: Setting an Agenda. Environmental Health Perspectives, 130(2), 25001. https://doi.org/10.1289/EHP9889.
Sasse, S. K., Gruca, M., Allen, M. A., Kadiyala, V., Song, T., Gally, F., Gupta, A., Pufall, M. A., Dowell, R. D., & Gerber, A. N. (2019). Nascent transcript analysis of glucocorticoid crosstalk with TNF defines primary and cooperative inflammatory repression. Genome Research, 29(11), 1753–1765. https://doi.org/10.1101/gr.248187.119.
Shimba, A., & Ikuta, K. (2020). Control of immunity by glucocorticoids in health and disease. Seminars in Immunopathology, 42(6), 669–680. https://doi.org/10.1007/s00281-020-00827-8.
Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette, M. A., Paulovich, A., Pomeroy, S. L., Golub, T. R., Lander, E. S., & Mesirov, J. P. (2005). Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America, 102(43), 15545–15550. https://doi.org/10.1073/pnas.0506580102.
Sumner, L. W., Amberg, A., Barrett, D., Beale, M. H., Beger, R., Daykin, C. A., Fan, T. W. M., Fiehn, O., Goodacre, R., Griffin, J. L., Hankemeier, T., Hardy, N., Harnly, J., Higashi, R., Kopka, J., Lane, A. N., Lindon, J. C., Marriott, P., Nicholls, A. W., & Viant, M. R. (2007). Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI). Metabolomics: Official Journal of the Metabolomic Society, 3(3), 211–221. https://doi.org/10.1007/s11306-007-0082-2.
Tashiro, H., & Shore, S. A. (2021). The Gut Microbiome and Ozone-induced Airway Hyperresponsiveness. Mechanisms and Therapeutic Prospects. American Journal of Respiratory Cell and Molecular Biology, 64(3), 283–291. https://doi.org/10.1165/rcmb.2020-0288TR.
Watanabe, S., Sudo, Y., Makino, T., Kimura, S., Tomita, K., Noguchi, M., Sakurai, H., Shimizu, M., Takahashi, Y., Sato, R., & Yamauchi, Y. (2022). Skeletal muscle releases extracellular vesicles with distinct protein and microRNA signatures that function in the muscle microenvironment. PNAS Nexus, 1(4), pgac173. https://doi.org/10.1093/pnasnexus/pgac173.
Wishart, D. S., Tzur, D., Knox, C., Eisner, R., Guo, A. C., Young, N., Cheng, D., Jewell, K., Arndt, D., Sawhney, S., Fung, C., Nikolai, L., Lewis, M., Coutouly, M. A., Forsythe, I., Tang, P., Shrivastava, S., Jeroncic, K., Stothard, P., & Querengesser, L. (2007). HMDB: The Human Metabolome Database. Nucleic Acids Research, 35(Database issue), https://doi.org/10.1093/nar/gkl923. D521-526.
Wright, P. T., Schobesberger, S., & Gorelik, J. (2015). Studying GPCR/cAMP pharmacology from the perspective of cellular structure. Frontiers in Pharmacology, 6, 148. https://doi.org/10.3389/fphar.2015.00148.
Xavier, A. M., Anunciato, A. K. O., Rosenstock, T. R., & Glezer, I. (2016). Gene Expression Control by Glucocorticoid Receptors during Innate Immune Responses. Frontiers in Endocrinology, 7, 31. https://doi.org/10.3389/fendo.2016.00031.

Competing interest reported. KMPJ is an employee of Metabolon Inc (Morrisvlle, NC) and, as such, has an affiliation with or financial involvement with Metabolon, Inc.

Download PDF

Journal Publication

published 10 Sep, 2023

Read the published version in Metabolomics →

Editorial decision: Major revision
03 Aug, 2023
Reviews received at journal
31 Jul, 2023
Reviewers agreed at journal
28 Jul, 2023
Reviewers agreed at journal
25 Jul, 2023
Reviewers invited by journal
19 Jul, 2023
Submission checks completed at journal
18 Jul, 2023
Editor assigned by journal
18 Jul, 2023
First submitted to journal
17 Jul, 2023

You are reading this latest preprint version

Multi-tissue transcriptomic and serum metabolomic assessment reveals systemic implications of acute ozone-induced stress response in male Wistar Kyoto rats

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and Methods

Animals

Ozone generation and animal exposures

Necropsy and tissue collection

RNA isolation and mRNA sequencing

Nextseq data normalization, statistics, and bioinformatics

Metabolomics sample preparation

Serum Metabolomics – LC/MS/MSⁿ analyses

Metabolomics data normalization, statistics, and bioinformatics

Results

Ozone-induced -omics changes are tissue-specific and robust

Pathway analyses reveals tissue concordance in ozone response

Upstream predictor analyses reveal tissue concordance in ozone response

Hypoxia and polyamines

Genes associated with GPCR/adrenergic and glucocorticoid receptor signaling

Metabolic pathways

Stress immune pathways

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1

Multi-tissue transcriptomic and serum metabolomic assessment reveals systemic implications of acute ozone-induced stress response in male Wistar Kyoto rats

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and Methods

Animals

Ozone generation and animal exposures

Necropsy and tissue collection

RNA isolation and mRNA sequencing

Nextseq data normalization, statistics, and bioinformatics

Metabolomics sample preparation

Serum Metabolomics – LC/MS/MSn analyses

Metabolomics data normalization, statistics, and bioinformatics

Results

Ozone-induced -omics changes are tissue-specific and robust

Pathway analyses reveals tissue concordance in ozone response

Upstream predictor analyses reveal tissue concordance in ozone response

Hypoxia and polyamines

Genes associated with GPCR/adrenergic and glucocorticoid receptor signaling

Metabolic pathways

Stress immune pathways

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1

Serum Metabolomics – LC/MS/MSⁿ analyses