The contribution of the neuroendocrine system to adaption after repeated daily ozone exposure in rats

Abstract

Ozone-induced lung injury/inflammation dissipates despite continued exposure for 3 or more days; however, the mechanisms of adaptation/habituation remain unclear. Since ozone effects are mediated through adrenal-derived stress hormones, which also regulate longevity of centrally-mediated stress response, we hypothesized that ozone-adaptation is linked to diminution of neuroendocrine stress-axes activation and glucocorticoid levels. Male Wistar-Kyoto-rats (12-week-old) were injected with vehicle or a therapeutically-relevant dexamethasone dose (0.01-mg/kg/day; intraperitoneal) for 1-month to determine if suppression of glucocorticoid signaling was linked to adaptation. Vehicle- and dexamethasone-treated rats were exposed to air or 0.8-ppm ozone, 4 hours/dayx2 or 4 days to assess the impacts of acute exposure and adaptation, respectively. Dexamethasone reduced thymus and spleen weights, circulating lymphocytes, corticosterone and increased insulin. Ozone increased lavage-fluid protein and neutrophils and decreased circulating lymphocytes at day-2 but not day-4. Ozone-induced hyperglycemia, glucose intolerance and inhibition of beta-cell insulin release occurred at day-1 but not day-3. Ozone depleted circulating prolactin, thyroid-stimulating hormone, and luteinizing-hormone at day-2 but not day-4, suggesting central mediation of adaptation. Adrenal epinephrine biosynthesis gene, Pnmt, was induced after ozone exposure at both timepoints. However, genes involved in glucocorticoid biosynthesis were induced after day-2 but not day-4, suggesting that acute 1- or 2-day ozone-mediated glucocorticoid increase elicits feedback inhibition to dampen hypothalamic stimulation of ACTH release in response to repeated subsequent ozone exposures. Although dexamethasone pretreatment affected circulating insulin, lymphocytes and adrenal genes, it had modest effect on ozone adaptation. In conclusion, ozone adaptation likely involves lack of hypothalamic response due to reduced availability of circulating glucocorticoids.

INTRODUCTION

Air pollution is now the 4^th leading cause of human morbidity and mortality world-wide (Health Effects Institute, 2020). Specifically, this HEI report shows that ozone-related mortality and morbidity are high in the south Asian countries where ozone increases have been highest. And that ambient ozone levels are now 30% to 70% higher than those estimated 100 years ago. Increased ambient temperature due to climate change and enhanced release of industrial chemicals contribute to ozone formation and as a result will have an increased global health impact (Health Effects Institute, 2020).

Ozone exposures have been linked to exacerbation of chronic obstructive pulmonary disease and asthma-related inflammatory flare-ups in susceptible individuals (Zielinski et al., 2018; Rosenquist et al., 2020; Zheng et al., 2021). In healthy individuals, ozone exposure causes acute lung injury and inflammation (Rich et al., 2020). It is well established that upon repeated daily ozone exposure in individuals and laboratory animals, pulmonary effects are reversible; generally referred to as adaptation or habituation (Farrell et al., 1979). In our sub chronic ozone exposure studies, we have shown that by the third consecutive day of ozone exposure, no pulmonary or systemic health effects are apparent. However, if animals are re-exposed to ozone after a period of 4-5 days of no exposure, the adaptation response is lost and the effects of ozone are once again readily apparent (Miller et al., 2016c). These observations imply that the longevity of adaptation likely depends on the type, and severity of exposure.

Recent research has demonstrated the mechanistic role of neuroendocrine stress pathways in mediating ozone biological effects, both in the lung and systemically through circulating adrenal-derived stress hormones (Miller et al., 2015; 2016a; 2016b; Henriquez et al., 2019a; Snow et al., 2018; Kodavanti, 2019). Adaptation and resiliency to psychosocial and physical stressors, or alternatively, failure to adapt to such stressors, has been shown to be mediated through mechanisms regulated by circulating glucocorticoids binding to their receptors and inducing corticotropin releasing factor release (CRF) in the hippocampus and hypothalamus (Herman et al., 2020; McEwen and Akil, 2020). While glucocorticoids and catecholamines released from adrenals in response to centrally perceived acute stress mediate peripheral immunological and metabolic response to stress through binding to their receptors, these hormones are also involved in stress adaptation (McEwen and Akil, 2020; Herman et al, 2020). Since ozone mediates its pulmonary injury and inflammatory effects in part through release of adrenal-derived glucocorticoids and catecholamines, we postulated that adaptation to repeated daily ozone exposure for more than 2 consecutive days (we used 4 days herein) would involve the availability of circulating stress hormones, especially glucocorticoids. Based on the evidence that low level DEX, a potent glucocorticoid agonist treatment for 3-4 weeks causes central hypocorticosteroid state in the rat brain (Karssen et al., 2006) and insulin resistance (Severino et al., 2002), we hypothesized that DEX would impair the peripheral adaptation response to ozone through downregulating endogenous glucocorticoid bioavailability and subsequent pulmonary and systemic metabolic effects. Herein we examined ozone effects and adaptation in relation to neuroendocrine involvement and adrenal production of glucocorticoids in rats pretreated with vehicle (VEH) or DEX. We show that the diminution of ozone-induced pulmonary and systemic effects after 4 consecutive days of exposure is linked to (1) adaptations in levels of neuroendocrine hormones and (2) to altered expression of enzymes involved in synthesis of adrenal corticosterone, but not to that of catecholamine synthesis. We further demonstrate that pretreatment with DEX for up to 1 month (1-month DEX treatment) had a minimal effect on preventing ozone adaptation in the lung, but did result in impaired glucose metabolic adaptation.

MATERIALS AND METHODS

Animals.

For this study we purchased 11–12-week-old male Wistar Kyoto (WKY) rats from Charles River Laboratories (Raleigh, NC). Upon arrival, rats were pair-housed in polycarbonate cages containing hardwood chip bedding. These animals were maintained at 21°C with relative humidity of 55-65%, and 12h light/dark cycle. The animal facility is approved by the Association for Assessment and Accreditation of Laboratory Animal Care. Rats were fed Purina 5001 rat chow (Brentwood, MO) and provided tap water, ad libitum. At the beginning of the experimentation, rats were randomized by body weight and allocated to each treatment, exposure condition and time (8/group). The U.S. EPA, Center for Public Health and Environmental Assessment’s Institutional Animal Care and Use Committee approved the experimental protocol (#19-05-002), and we followed National Institutes of Health guide for the care and use rats (NIH Publications No. 8023). The experiments were conducted in accordance with ARRIVE guidelines (https://arriveguidelines.org).

Dexamethasone (DEX) treatment.

We used an immunosuppressive dose of DEX (Cuzzocrea et al., 2005) in the present study. Thirty-two rats received 1 mL/kg saline injections daily intraperitoneally and 32 rats 0.01 mg/kg DEX hydrochloride daily between 6-7 am for 30 days and during ozone exposure (Figure 1). DEX treatment was continued during days of ozone exposure to assure the persistency of DEX-mediated effect on glucocorticoid sensitivity without its reversibility, especially at later timepoint of 4-day ozone exposure. Thirty-two saline and 32 DEX-treated rats were randomized to respective air and ozone groups and 2-day or 4-day time points (n=8/group).

Figure 1.

Experimental design. The times of vehicle (VEH) or dexamethasone (DEX)-pretreatment and air or ozone exposures are shown by a horizontal arrow on the top. The in-life assessment times as well as necropsies are indicated by vertical arrows. Different cohorts of animals were used for 2-day (D+2) and 4-day (D+4) exposures. Eight animals per group (for each exposure/time) were analyzed totaling 32 rats per timepoint. Tissue collection was performed within 30 minutes (right after plethysmography) to 2 hours once exposure was terminated. ITT, insulin tolerance test; GTT, glucose tolerance test.

Ozone exposure.

On day 31, rats while on respective treatment regimen were exposed 4-hours to air or 0.8 ppm ozone daily for 2 or 4 consecutive days (Figure 1), as done in our previous studies (Henriquez et al., 2018; 2021a). We have shown ozone as an environmental stressor induces lung injury and inflammation associated with neuroendocrine effects that are readily apparent on day 1 and day 2, while pulmonary neutrophil response is maximum on day 2 (Miller et al., 2015). Moreover, it has been shown that if exposure occurred for 3 days or more, the pulmonary and systemic effects fade (Farrell et al., 1979; Hamade and Tankersley, 2009; Miller et al., 2016c). Rats were exposed to filtered air (0.0 ppm), or 0.8 ppm ozone for 4 hours (Figure 1), for 2 consecutive days to assess acute ozone effects and for 4 consecutive days to assess adaptation. Exposure occurred in a staggered manner to accommodate for necropsy and sample processing times. As reported in our previous studies (Miller et al., 2015; Henriquez et al., 2021a), ozone was generated from oxygen using a silent arc discharge generator (OREC, Phoenix, Arizona). Mass flow controllers (Coastal Instruments Inc., Burgaw, North Carolina) were used for regulating its concentration, which was monitored by photometric analyzers (API Model 400, Teledyne, San Diego, California). Air and ozone chamber mean temperature, relative humidity, and air flow were monitored hourly during exposure.

Whole-body plethysmography (WBP) measurements.

To (1) non-invasively monitor the influence of adaption on ozone-induced ventilatory changes, and to (2) determine the potential impact of DEX pretreatment on adaption, WBP was performed following each day of exposure for all animals assigned to the D+4 group (n=8/group). Immediately after exposure, animals were transported to a nearby quiet room and placed in WBP chambers to assess differences in their ventilatory rate and pattern. In a staggered manner, 2 animals from each group were assessed per run (n=8 rats/run), and rats were rotated across the different chambers each day. After a 2 min acclimation period, ventilatory parameters were recorded for 5 minutes. All data were acquired, stored, viewed, and analyzed using EMKA iox2 software (EMKA Technologies, Falls Church, VA). Additional details have been previously published (Henriquez et al., 2018). In brief, herein we show changes in breathing frequency (f), expiratory time (Te), tidal volume (TV), peak expiratory flow (PEF), and Penh, an index of airflow limitation (Hamelmann, 1997). To better ensure that only valid breaths were included in these assessments, all recorded WBP pseudo-flow waves were visually inspected. To minimize the influence of overt exploratory sniffing activity (i.e., rapid breathing related to odor discrimination), breaths with f ≥ 375 breaths/min were excluded. Furthermore, only breaths with inspiratory and expiratory volumes within 40% were accepted.

Glucose tolerance test (GTT), blood collection and insulin tolerance test (ITT).

A single ozone exposure induces glucose intolerance (Miller et al., 2015). To determine if this systemic metabolic response also exhibits adaptation, GTT was performed immediately following day 1 and day 3 in animals assigned to D+4 exposure. Since GTT was performed immediately after ozone exposure and plethysmography, animals were not provided food for ~6 hours prior to testing. Briefly, blood glucose levels were measured at baseline (0 min) and rats were injected intraperitoneally with a bolus dose of glucose (D-glucose 20% V/V; Henry Schein, Dublin, OH; 2.0 g/kg). Then, glucose measurements were repeated every 30 min until 120 min. Blood glucose levels were assessed from a tail prick using a Bayer Contour glucometer (Bayer Corp., Leverkusen, Germany). To determine glucose mediated insulin release, ~100 μL of blood samples were collected from the tail vain at baseline and 30 min after glucose injection. Blood samples were then centrifuged, and serum aliquots were stored at −80°C for insulin measurement.

Animals assigned to D+2 exposures underwent insulin tolerance test immediately following day 1 exposure (exposure and preparatory periods accounted for fasting). After determining baseline base line glucose levels from the tail prick as in the case of GTT, rats were injected intraperitoneally with 0.75 IU Humulin^R/kg body weight in 1 mL saline (Lilly USA, LLC, Indianapolis, IN). Blood glucose was measured at 15, 30, 60, 90 and 120 min after insulin injection.

Necropsy and samples collection.

Rats from each group were staggered for exposures to accommodate for necropsies and sampling. Rats were further staggered during necropsies to reduce sampling variability. Between 30 minutes to 2 hours post exposure, rats were euthanized with an overdose of sodium pentobarbital (Fatal Plus, Virbac AH, Inc., Fort Worth, TX; >200 mg/kg, intraperitoneal). As we have done in our previous studies (Henriquez et al., 2018), blood samples were collected from the abdominal aorta directly in one EDTA and one serum separator vacutainer tube. Blood collected in EDTA tubes was used for complete white blood cell count on a Beckman-Coulter AcT blood analyzer (Beckman-Coulter Inc., Fullerton, California). Then, both the EDTA and serum separator tubes were centrifuged (3500 x g for 10 min) to collect plasma and serum. These samples were stored at −80°C for further analysis. The right lung was lavaged with Ca²⁺ and Mg²⁺ free PBS (pH 7.4, 37 °C) at 60% of 28 mL (estimated total lung capacity)/kg body weight as we have done in our previous studies (Miller et al., 2015; Henriquez et al., 2018). Un-lavaged left lungs and adrenal glands were removed and snap frozen in liquid nitrogen and stored at −80 °C for later RNA extraction.

BALF analysis of lung injury and inflammation markers.

Total cells in bronchoalveolar lavage fluid (BALF) were counted using Z1 Coulter Counter (Coulter, Inc., Miami, FL) and cytospin slides were prepared using a Shandon cytocentrifuge (Thermo Fisher Scientific, Waltham, MA). The slides were dried and stained with Diff-Quik (Thermo Fisher Scientific) for cell differentials using light microscopy (>300 cells/slide). After centrifugation, the cell-free BALF was used to measure total protein (Coomassie plus Protein Assay Kit, Pierce, Rockford, IL), and albumin (DiaSorin, Stillwater, MN). These assays were performed using a Konelab Arena 30 clinical analyzer (Thermo Chemical Lab Systems, Espoo, Finland).

Immunoassays for BALF and serum markers.

BALF and serum cytokines were analyzed using the V-PLEX proinflammatory panel 2 (rat) kit. Manufacturer’s protocol was followed for these assays except for using 200 microliters of BALF sample volume (Mesoscale Discovery Inc., Rockville, MD). The resulting electrochemiluminescence signals were detected using the MESO QuickPlex SQ 120 platform (Mesoscale Discovery Inc., Rockville, MD). For animals in which cytokine levels were below assay detection limits, the values were substituted with the lowest quantified value for the given cytokine in the analysis. Plasma samples prepared from EDTA tubes were used for analysis of adrenal-derived epinephrine (Rocky Mountain Diagnostics, Colorado Springs, CO) and corticosterone (Arbor Assays, Ann Arbor, MI). Manufacturer’s protocols were used for the assay and the ELISA plates were read on a SpectraMax i3x Multi-Mode Microplate Reader (Molecular Devices, San Jose, CA). Pituitary hormone levels in the serum (adrenocorticotropic hormone, ACTH; thyroid stimulating hormone, TSH; prolactin, PRL; luteinizing hormone, LH) were analyzed using MILLIPLEX MAP Rat Pituitary Magnetic Bead Panel following manufacturer’s protocol (Merck-Millipore, Burlington, MA). Serum insulin levels were determined using a rat-specific chemiluminescence assay kit (Millipore, Billerica, MA) following the manufacturer’s instructions. Serum leptin levels were analyzed using kits and protocols from Mesoscale Discovery® Multi-Spot® assay system (Rockville, MD).

Lung and adrenal RNA isolation and real time PCR (qRT-PCR).

Uniform portions of frozen un-lavaged left lung lobes or whole adrenal glands were used for total RNA extraction using RNeasy mini kits (Qiagen, Valencia, CA) following manufacturer protocols. Total RNA was quantified using a Qubit 2.0 fluorimeter (Thermo Fisher, Waltham, MA). For PCR, Qscript Supermix (Quanta Biosciences, Beverly, MA) was used for cDNA synthesis. PCR primers for adrenal catecholamine and glucocorticoid synthesis genes and lung markers were designed using Rattus Norvegicus sequences annotated in NCBI and obtained from Integrated DNA Technologies, Inc. (Coralville, IA). The primer sequences are shown in Table 1. SYBR Green PCR Master Mix (Thermo Fisher, Waltham, MA) was used for qRT-PCR and the PCR products were detected using the Applied Biosystems 7900HT Sequence Detection System (Foster City, CA). Relative mRNA expression was calculated using β-actin as the normalization transcript and the ΔΔCt method was employed where the 2-day air-exposed group served as a control.

Table 1.

Primer sequences for genes used in qRT-PCR.

Gene markers	Sequences for forward and reverse primers
Interleukin 6 (Il6)	f - CTTCACAAGTCGGAGGCTTAAT, r - GCATCATCGCTGTTCATACAATC
TSC22 domain family member 3 (Tsc22d3)	f - CCGAATCATGAACACCGAAATG, r - CAGAGAAGAGAAGAAGGAGATG
Hydroxysteroid 11-beta dehydrogenase 1 (Hsd11b1)	f - CTCCACTTCTGCTTGGGAAT, r - CTCAGGAGTTCCTAGTTGCTTAC
Hydroxysteroid 11-beta dehydrogenase 2 (Hsd11b2)	f - CTTCATCAGTCACCTTCTACCC, r - TGTGGTTCACAGAGCTGATAC
Phenylalanine hydroxylase (Pah)	f - TTGCTGGCTTACTGTCATCTC, r - GGCTTCGATCCATGCCTAAT
Tyrosine hydroxylase (Th)	f - CCCTACCAAGATCAAACCTACC, r - CTGGATACGAGAGGCATAGTTC
Dopamine beta-hydroxylase (Dbh)	f - GGTGAACAGAGACAACCACTAC, r - GCACGAAGTGATGAGGACAT
Phenyl ethanolamine-N-methyltransferase (Pnmt)	f - GGATGACGTCAAGGGTATCTTC, r - CACTTCGGGTGATAAGGAGTTAG
Cytochrome P450 family 11 subfamily A member 1 (Cyp11a1)	f – AGAACGGCACACACAGAAT, r - CCTTAGGGTCCAGGATGTAAAC
Hydroxy-delta-5-steroid dehydrogenase, 3 beta (Hsd3b1)	f - TTCCTGCTGCGTCCATTT, r - GATCTCTCTGAGCTTTCTTGTAGG
Cytochrome P450 family 21 subfamily A member 1 (Cyp21a1)	f - CCACATCTATGCCACCTTATCC, r - TGGGAGAGGACATCACTTCA
Cytochrome P450 family 11 subfamily B member 1 (Cyp11b1)	f - CCTTTGAGTGTGAGGCAGTATAG, r - GAGTAGGCACAACCCAGTAATC
Cytochrome P450 family 11 subfamily B member 2 (Cyp11b2)	f - CCTTTGAGTGTGAGGCAGTATAG, r - GAGTAGGCACAACCCAGTAATC
β-Actin (Actβ)	f-CAACTGGGACGATATGGAGAAG r-GTTGGCCTTAGGGTTCAGAG

Statistics.

Based on the incidence of spontaneous cardiac hypertrophy in about 10% of WKY rats, as explained previously (Henriquez et al., 2018), data from WKY rats demonstrating hypertrophy were excluded from further consideration. Hence, our group size for statistical analysis ranged from n=6-8. Data were analyzed using a two-way analysis of variance (ANOVA) where exposure (air or ozone) and treatment (vehicle or DEX) were the factors. D+2 and D+4 data were analyzed separately. Holm Sidak’s test was used to correct for multiple comparisons. Group differences were considered significant when p values were ≤ 0.05. GraphPad prism 9.1.2 and Statext 2.7 software were used for statistical analysis. For GTT, area under the curve was calculated using the trapezoidal method. For gene expression analysis, outliers were identified and discarded using the ROUT method (robust regression and outlier removal, Motulsky and Brown 2006) prior to statistical analysis.

RESULTS

Adaptation of acute ventilatory changes upon repeated ozone exposure

It is well established that acute respiratory physiology changes induced after acute ozone exposure in humans and in animals are reversible not only upon termination of exposure but also upon continuation of exposure (Dimeo et al., 1981; Tepper et al., 1989). This short-lived adaptation is somewhat dependent upon experimental protocol. To ensure that our study design and ozone exposures were consistent with previous reports, and likewise, resulted in adaptation in the D+4 group, we compared ventilatory changes in these WKY rats immediately after air or ozone exposure for 4 consecutive days.

As expected, on Day 1 and Day 2, the ozone-exposed VEH-pretreated group exhibited the greatest changes in their breathing patterns (Fig. 2). These changes were characterized by significant increases in the peak expiratory flow (PEF) rates and corresponding changes in Penh (enhanced pause), a unitless parameter that increases during labored or obstructive breathing. On Day 1, the ozone-exposed VEH-pretreated rats exhibited increased breathing frequency (f), as compared to the air-exposed VEH-pretreated rats. Although the f increase was not statistically significant, it was consistent with rats experiencing a “shortened or shallow” breathing pattern on Day 1 as they sought to minimize discomfort associated with taking a deep inspiration acutely after ozone exposure. By Day 2, the ozone-exposed rats continued to exhibit high PEF rates, and developed even greater increases in Penh. They also exhibited a lower breathing f, allowing for disproportionately increased expiratory time (Te). Such findings are consistent with development of airway obstruction by Day 2 such that during each breath, prolongation of Te would allow trapped air to be more fully expired. Notably, within the air-exposed rats, DEX-pretreatment did not appear to alter breathing rate, effort, or pattern. Similarly, within the ozone-exposed groups, DEX-pretreatment minimally altered ozone exposure response, although this group did have a significantly increased Penh (on Day 1) and Te (on Day 2). By Day 3 and Day 4, all ventilatory rate or pattern changes appeared to be resolved.

Figure 2.

Differences in WBP-derived ventilatory parameters of vehicle (VEH) and dexamethasone (DEX)-treated rats exposed to air or ozone (0.8 ppm x 4 h) daily for 4 consecutive days. Respiratory measures include breathing frequency (A), tidal volume (B), peak expiratory flow (C), expiratory time (D), and Penh (E). Each exposure group includes VEH or DEX-pretreated rats (mean ± SEM of n=6-8 rats/group). Black asterisks (*) represent a significant ozone effect (P ≤ 0.05) in VEH group relative to time-matched VEH-pretreated air group. Red asterisks (*) indicate a significant ozone effect (P ≤ 0.05) in DEX group relative to time-matched DEX-pretreated air group.

Effect of DEX on lymphoid organ weights and adaptation from acute ozone-induced lung injury/ inflammation

DEX pretreatment was associated with decreases in thymus and spleen weights but not adrenal weights (Table 2) indicating the effectiveness of DEX on lymphoid organs linked immunosuppression. Ozone did not have significant effects except for a further drop in thymus weight in DEX pretreated animals (Table 2). Ozone-induced increases in BALF protein and albumin, markers of vascular leakage, were observed at D+2 in VEH and DEX groups but this increase was nearly absent on D+4 VEH-pretreated group. DEX animals showed slightly lesser attenuation of ozone effects than VEH-pretreated rats (Figure 3A–B). BALF alveolar macrophages were depleted after ozone exposure at D+2, likely indicating increased adherence of activated macrophages leading to smaller recovery in BALF. This effect was not seen at D+4. Rather, ozone-induced increase in BALF macrophages occurred in VEH- but not DEX-pretreated animals (Figure 3C). As we have noted in our earlier studies (Miller et al., 2015), ozone exposure was associated with marked neutrophilic inflammation in both VEH- and DEX-pretreated rats at D+2, however, this effect was absent in VEH-pretreated group at D+4. A small degree of neutrophilic inflammation persisted in DEX-pretreated group at D+4 (Figure 3D).

Table 2.

Relative thymus, spleen, and adrenals weights after D+2 and D+4 ozone exposure in animals pretreated with vehicle (VEH) or dexamethasone (DEX) (organ weight / body weight x 1000).

	Day 2				Day 4
	Air		Ozone		Air		Ozone
	VEH	DEX	VEH	DEX	VEH	DEX	VEH	DEX
Thymus	0.71 ± 0.04	0.51 ± 0.04†	0.69 ± 0.04	0.50 ± 0.03*†	0.64 ± 0.08	0.47 ± 0.07†	0.56 ± 0.10	0.53 ± 0.05
Spleen	2.53 ± 0.19	1.83± 0.09†	2.40 ± 0.17	2.07 ± 0.24†	2.27 ± 0.23	1.70 ± 0.11†	2.2 ± 0.21	1.71 ± 0.11†
Adrenals	0.19 ± 0.02	0.18 ± 0.01	0.20 ± 0.02	0.19 ± 0.02	0.20 ± 0.01	0.20 ± 0.02	0.20 ± 0.02	0.40 ± 0.57

The data show mean ± SD (n=6-8 rats/group).

^*Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air-exposed group.

^†Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

Figure 3.

Lung injury, inflammation, and lymphopenia at D+2 but not D+4 after ozone exposure and the effect of dexamethasone (DEX) pretreatment. Lung injury and inflammation markers assessed in bronchoalveolar lavage fluid (BALF) include protein (A), albumin (B), alveolar macrophages (C), neutrophils (D) and circulating inflammatory markers include total white blood cells (WBC) (E), and lymphocytes (F). The data show mean ± SEM of n=6-8 rats/group. *Significant ozone effect (P ≤ 0.05) in vehicle (VEH) or DEX-pretreated group relative to time-matched air-exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

The majority of circulating white blood cells in rats are lymphocytes and thus, the change in white blood cells generally reflects the change in lymphocyte population. Ozone-induced lymphopenia after ozone and DEX pretreatment is well characterized in our previous studies (Henriquez et al., 2018, 2019b). Consistent with those observations, we noted decreases in total white blood cells and lymphocytes in all animals pretreated with DEX regardless of ozone, however, in VEH-pretreated animals, this effect of ozone was observed only at D+2 but not at D+4, suggesting adaptation (Figure 3E–F).

BALF and circulating cytokines indicative of inflammation in the lung and systemically

BALF TNF-α was increased after ozone exposure only in DEX-pretreated rats on D+2 and D+4, suggesting a long-term DEX effect on the lung making it susceptible to ozone (Figure 4A). BALF IL-6 levels increased after ozone exposure in both VEH and DEX-pretreated animals at D+2, however, this effect was modestly attenuated at D+4 (Figure 4B). Increased levels of IL-1β in DEX-pretreated animals was enhanced after ozone at D+2, but not at D+4 (Figure 4C). At D+2, BALF KC-GRO increased in DEX-pretreated ozone-exposed animals only, but this effect was absent at D+4 (Figure 4D). Two critical circulating cytokines, TNF-α and IL-6, were determined in the serum (Figure 4E–F). While no treatment or exposure related changes were noted in circulating IL-6 levels (Figure 4F), serum TNF-α levels were decreased in all DEX-pretreated animals regardless of ozone exposure and in VEH-pretreated ozone-exposed animals. This ozone effect occurred only at D+2 but not at D+4, suggesting adaptation (Figure 4E).

Figure 4.

Cytokines in the bronchoalveolar lavage fluid (BALF) and serum after D+2 and D+4 ozone exposure in animals pretreated with vehicle (VEH) or dexamethasone (DEX). BALF cytokines include TNF-α, tumor necrosis factor α (A); IL-6, interleukin-6 (B); IL-1β, interleukin-1β (C); KC-GRO, keratinocyte chemoattractant/human growth-regulated oncogene (D); and serum cytokines include TNF-α (E) and IL-6 (F). The data show mean ± SEM of n=6-8 rats/group. *Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air-exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

Anterior pituitary, adrenal-derived and metabolic hormones, and ozone adaptation

To determine if adaptation to repeated ozone exposure is modulated by the activity of the neuroendocrine system, we determined several neuroendocrine hormones at D+2 and D+4 in our model (Figure 5). While ACTH increase during ozone inhalation occurs rapidly (peaking at 1 hour) and is reversible at 4 hr (Henriquez et al., 2021b), we noted no ozone-exposure-related changes in ACTH at either time points in VEH-pretreated rats, however, as expected, DEX pretreatment was associated with a significant drop in circulating ACTH regardless of ozone exposure as it can inhibit pituitary ACTH release through feedback inhibition (Figure 5A). The levels of serum PRL, TSH and LH, which are catabolic hormones of the anterior pituitary, were markedly inhibited by ozone in both VEH and DEX-pretreated animals at D+2 but not D+4, indicating complete adaptation upon repeated ozone exposure (Figure 5B–D). The adaptation response of TSH at D+4 was less remarkable in DEX-pretreated ozone-exposed animals relative to VEH-treated rats (Figure 5C). At D+2 we have seen variable results with corticosterone in terms of ozone effects (Henriquez et al., 2018). Likewise, we noted that corticosterone in VEH-pretreated ozone-exposed rats tended to be higher at D+2 when compared to air-exposed group; however, this change was not significant. DEX treatment markedly decreased the levels of circulating corticosterone in all air and ozone-exposed groups at both time points, but the levels were still significantly higher at D+2 in ozone-exposed animals relative to the air, suggesting that even in the presence of DEX, ozone increased corticosterone at D+2. This effect was absent at D+4 suggesting adaptation in ozone-induced corticosterone increase (Figure 5E). Plasma epinephrine levels tended to be higher in all DEX-pretreated animals and ozone-exposed animals, but these changes did not reach significance (Figure 5F).

Figure 5.

Serum/plasma levels of pituitary, adrenal-derived stress hormones and metabolic hormones are changed at D+2 but not D+4 after ozone exposure in rats pretreated with vehicle (VEH) or dexamethasone (DEX). These hormones include ACTH, adrenocorticotropic hormone (A); TSH, thyroid stimulating hormone (B); PRL, prolactin (C); LH, luteinizing hormone (D); corticosterone (E), epinephrine (F), insulin (G) and leptin (H). The data show mean ± SEM of n=6-8 rats/group. *Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air-exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

Based on the evidence that long-term DEX pretreatment has been linked to insulin resistance (Severino et al., 2002), at necropsy we assessed circulating insulin levels. In general, we noted that DEX pretreatment of WKY rats increased serum insulin levels in air exposed animals, whereas in ozone-exposed rats, insulin levels tended to decrease in both VEH and DEX pretreated animals at D+2 (Figure 5G). Results suggest independent effects of catecholamine and glucocorticoids (Miller et al., 2016c). By day 4 insulin levels were not depressed by ozone in DEX-pretreated rats suggesting adaptation (Figure 5G). The levels of leptin also tended to be higher in all DEX-pretreated animals exposed to air and ozone, but this effect was only significant for ozone-exposed D+4 group. Moreover, as we have noted in our previous studies (Miller et al., 2015), leptin levels increased significantly at D+2, however decreases were noted at D+4 in VEH-pretreated rats (Figure 5H).

Adaptation in ozone-induced glucose metabolic changes and effects of DEX

In addition to adipose lipolysis, acute ozone exposure has been shown to increase blood glucose, induce gluconeogenesis and inhibit glucose-mediated secretion of insulin from pancreas (Miller et al.,2016c). Here we performed GTT on day 1 and day 3 (immediately post ozone exposure on each day) in animals assigned to D+4 group. We have previously noted that glucose intolerance is readily apparent in animals at the first and second day of consecutive ozone exposure, however, this effect is diminished on the third day (Miller et al., 2016c). Here we confirmed that ozone-induced glucose intolerance noted at day 1 was nearly absent at day 3 (Figure 6A–B). DEX did not have significant effect on ozone-induced hyperglycemia and glucose intolerance, however, in air-exposed rats on day 1, DEX pretreatment induced a small degree of glucose intolerance (Figure 6C). We have previously shown that ozone does not impair insulin-mediated peripheral glucose uptake as determined by ITT (Miller et al., 2016c). Nevertheless, we performed ITT at day 1 on D+2 group to determine if DEX-pretreatment produced insulin resistance. We noted that neither ozone nor DEX-pretreatment caused significant insulin resistance in WKY rats with our treatment protocol, as hyperglycemia induced by ozone on D+1 was effectively corrected by insulin injection (Figure 6D). To determine the influence of DEX pretreatment and ozone on glucose-mediated insulin secretion from pancreas, we performed beta cell insulin secretion test on day 1 and day 3, immediately after ozone exposure along with GTT. As observed in our previous study, we noted that at day 1 ozone completely inhibited the ability of beta cells to secrete insulin in response to injected glucose. We also noted that DEX-pretreated animals exposed to air but not ozone had significantly higher levels of baseline insulin relative to vehicle-pretreated rats (this trend was only noted at day 1, shown in figure 6E). In DEX-pretreated animals this increase was not evident after ozone exposure, suggesting that ozone dampened DEX-induced increase in insulin (Figure 6E). This ozone effect on insulin in DEX and in vehicle-pretreated rats was nearly eliminated at day 3 showing adaptation in metabolic processes. This was evident by higher levels of insulin at baseline in DEX-pretreated animals regardless of exposure and that injection of glucose led to significant increases in circulating insulin in all animals (Figure 6F).

Figure 6.

Effect of ozone on glucose and insulin metabolic effects at day 1 and day 3 after ozone exposure in rats pretreated with vehicle (VEH) or dexamethasone (DEX). Glucose tolerance test (GTT) was performed after day 1 and day 3 (A and B shows GTT data while C shows area under the curve, AUC values). Insulin tolerance test (ITT) was performed at day 1 after ozone exposure (D). Glucose-mediated insulin release was determined at day 1 and day 3 (E and F, respectively). The data show mean ± SEM of n=6-8 rats/group. *Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air -exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

Pulmonary and adrenal gene expression reveals the role of glucocorticoids in ozone adaptation

The roles of glucocorticoids in adaptation or resiliency at the central level have been well established, although, the precise pathways continued to be further examined (McEwen and Akil, 2020). With understanding that there is likely a key role for glucocorticoid feedback, we wanted to determine if peripheral endogenous glucocorticoid availability was critical in tissue adaptation after ozone exposure. Therefore, we quantified gene expression in the lung to demonstrate glucocorticoid influence of ozone and DEX and markers of epinephrine and glucocorticoids biosynthesis pathway in the adrenal glands.

As we have observed in earlier studies (Henriquez et al., 2018), lung Il6 mRNA increased after ozone exposure at D+2 but this effect was dampened at D+4 in both VEH- and DEX-pretreated rats (Figure 7A). Tsc22d3 (also known as Gilz), a glucocorticoid responsive gene induced after ozone exposure as reported in our previous study (Henriquez et al., 2021a) was increased after ozone exposure in VEH-pretreated rats but only at D+2. All DEX pretreated rats had slightly higher levels of Tsc22d3 expression in both air and ozone-exposed groups (Figure 7B). This suggested that at D+2, glucocorticoid effect of ozone exposure was occurring at lung level, however there was an adaptation at D+4 when no ozone effect was evoked.

Figure 7.

The expression of genes for cytokines, and glucocorticoid response in the lungs at D+2 and D+4 after ozone exposure in rats pretreated with vehicle (VEH) or dexamethasone (DEX). Il6, interleukin-6 (A); Tsc22d3, TSC22 Domain Family Member 3 (Gilz) (B); Hsd11b1, hydroxysteroid 11-Beta Dehydrogenase 1 (C); Hsd11b2, hydroxysteroid 11-Beta Dehydrogenase 2 (D).The data show mean ± SEM of n=6 rats/group. *Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air-exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

At a cellular level in rats, glucocorticoid intracellular concentrations are regulated by enzymatic interconversion between active corticosterone and inactive 11-dehydrocorticosterone. Hsd11b2 (hydroxysteroid 11-beta dehydrogenase 2) gene encodes for an enzyme involved in converting corticosterone to its inactive form 11-dehydrocorticosterone while Hsd11b1 (hydroxysteroid 11-beta dehydrogenase 1) encodes for an enzyme that participates in the reverse step converting 11-dehydrocorticosterone to its active form - corticosterone. This regulation of glucocorticoid activity reduces overt or long-term effects of the presence of glucocorticoids in cells (Seckl, 2004). In the lung tissue, we noted that mRNA expression of Hsd11b1 was not changed by ozone or DEX pretreatment at any time point, however Hsd11b2 expression was significantly inhibited by ozone in VEH-pretreated animals at D+2 and not at D+4, indicating adaption (Figure 7C–D). This inhibition tended to be preserved for DEX-pretreated group at D+2.

Adrenal medulla synthesizes catecholamines, primarily epinephrine from phenylalanine upon activation of SAM axes (Figure 8E), which is released into the circulation whereas, adrenal cortex upon activation of HPA axis in response to ACTH, synthesizes corticosterone from cholesterol which is also released in the circulation (Figure 9G). When stress is encountered, both pathways are activated. We evaluated mRNA expression for enzymes involved in pathways for epinephrine and corticosterone biosynthesis. Expression of mRNA for phenylalanine hydroxylase (Pah) involved in the first step of converting phenylalanine to tyrosine, tyrosine hydroxylase (Th), which converts tyrosine to DOPA and dopamine beta-hydroxylase (Dbh), which converts dopamine to nor-epinephrine were not consistently changed by either ozone exposure or DEX pretreatment at any time point (Figure 8A–C). However, mRNA of phenyl ethanolamine N-methyltransferase (Pnmt), which is a rate limiting enzyme in the synthesis of epinephrine from nor-epinephrine, was significantly increased after ozone exposure in both VEH and DEX-pretreated animals at D+2. This ozone effect was only partially dampened at D+4 in VEH-pretreated animals, suggesting a modest ozone adaptation in epinephrine synthesis pathway. DEX group did not show any ozone related increases at D+4 (Figure 8D).

Figure 8.

Expression of genes coding for enzyme proteins involved in catecholamine synthesis at D+2 and D+4 after ozone exposure in rats pretreated with vehicle (VEH) or dexamethasone (DEX). Pah, phenylalanine hydroxylase (A); Th, tyrosine hydroxylase (B); Dbh, dopamine beta-hydroxylase (C); Pnmt, phenylethanolamine N-methyltransferase (D). Right schematic shows steps involved in biosynthesis of nor-epinephrine and epinephrine while left side shows the level of expression for genes coding some of these proteins in adrenal glands (E). The data show mean ± SEM of n=6 rats/group. *Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air-exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

Figure 9.

Expression of genes coding for enzyme proteins involved in glucocorticoid synthesis at D+2 and D+4 after ozone exposure in rats pretreated with vehicle (VEH) or dexamethasone (DEX). Cyp11A1, Cytochrome P450 Family 11 Subfamily A Member 1 (A); Hsd3b1, hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 1 (B); Cyp21A1, cytochrome P450, family 21, subfamily a, polypeptide 1 (C); Cyp11B1, cytochrome P450 family 11 subfamily B member 1 (D); Cyp11B2, cytochrome P450 family 11 subfamily B member 2 (E); Hsd11b2, hydroxysteroid 11-beta dehydrogenase 2 (F). Right schematic shows steps involved in biosynthesis of corticosterone while left side shows the level of expression for genes coding these proteins in adrenal glands (G). The data show mean ± SEM of n=6 rats/group. *Significant ozone effect (P ≤ 0.05) in VEH or DEX-pretreated group relative to time-matched air-exposed group. †Significant DEX effect (P ≤ 0.05) in air or ozone-exposed group relative to time-matched VEH-pretreated group.

Next, we assessed mRNA expression for enzymes involved in corticosterone synthesis from cholesterol (Figure 9G) in adrenal glands as glucocorticoids play a central role in not only mediating stress response but also in terminating this response even after continuation of stressor exposure. Expression of mRNA for the first enzyme, Cyp11a1 involved in converting cholesterol to pregnenolone tended to increase in ozone-exposed rats pretreated with VEH, however, this difference did not reach significance and DEX pretreatment in ozone-exposed groups tended to decrease its expression (Figure 9A). The mRNA for the second enzyme Hsd3b1, which converts pregnenolone to progesterone was significantly increased in ozone exposed VEH- and DEX-pretreated animals at D+2 but not at D+4 (Figure 9B). Expression for Cyp21a1, which converts progesterone to 11-deoxycorticosterone was also significantly increased after ozone exposure in both VEH and DEX-pretreated animals at D+2 but not at D+4 (Figure 9C). Likewise, the mRNA expression for Cyp11b1, which converts 11-deoxycorticosterone to corticosterone, was significantly increased after ozone exposure in both VEH and DEX groups at D+2 but not at D+4 (Figure 9D). Cyp11b2 catalyzes the synthesis of aldosterone - an important regulator of salt/water balance and blood pressure - from 11-deoxycorticosterone. The mRNA expression of this gene was also significantly increased after ozone exposure at D+2 in both VEH and DEX-pretreated rats but this effect was nearly eliminated at D+4 (Figure 9E). Similarly, Hsd11b2, which inactivates corticosterone into 11-dehydrocorticosterone, was induced by ozone at D+2 but not D+4 in adrenal glands (Figure 9F).

DISCUSSION

Our recent work demonstrated the contribution of components of the neuroendocrine system (especially adrenal gland-derived stress hormones), in mediating ozone effects. Since glucocorticoids mediate feedback regulation, resiliency, and adaptation to stress through their central effects (McEwen and Akil, 2020), we hypothesized that the adaptation that occurs with daily repeated exposure to ozone in rats will involve neuroendocrine system activation and the availability of glucocorticoids in the circulation. Additionally, based on the evidence that long-term treatment with DEX, a glucocorticoid agonist can result in a hypo corticosteroid state in the brain (Karssen et al., 2007) and peripheral insulin resistance (Severino et al., 2002), we presumed that it would modify ozone adaptation in systemic metabolic and pulmonary effects.

By exposing rats for 4 consecutive days to 0.8 ppm ozone, 4 hours/day, we confirmed marked ozone-induced breathing alterations, metabolic impairment, and lung injury/inflammation on day 1 and day 2 and near complete adaptation in these parameters at day 3 and day 4 (Table 3). We further demonstrated that ozone-induced changes in anterior pituitary hormones in D+2 group, were not observed in D+4 group, concurrent to the lack of glucocorticoid increase, indicating that ozone-adaptation is likely regulated centrally and involves the availability of endogenous glucocorticoids. Further examination of gene expression of targets involved in epinephrine and corticosterone biosynthesis in adrenal glands indicated modest adaptation at D+4 for Pnmt (a rate limiting enzyme in epinephrine synthesis), which was induced at D+2 after ozone exposure. However, all enzymes involved in corticosterone and aldosterone synthesis induced at D+2 after ozone exposure showed no changes at D+4, suggesting a near complete adaptation which may suggest dampening of hypothalamic trigger to increase corticosterone synthesis (Table 3). Although, DEX pretreatment had an inhibitory effect on the adrenal expression of several genes involved in corticosterone biosynthesis and produced immunosuppressive effects in lymphoid organs, in general, it had modest effects on ozone-adaptation of glucose metabolic alterations and pulmonary injury/inflammation (Table 3). We demonstrate here the possible contribution of HPA axis involving the central effect of adrenal-glucocorticoid production in adaptation after repeated ozone exposure (Figure 10).

Table 3.

Summary of findings related to acute ozone effects and adaptation in control and DEX-pretreated rats.

Biological process	Endpoint affected by ozone or DEX	DEX-pretreatment effects in air group	Ozone effects in control rats (day 1 or 2)	Ozone effects in control rats (adaptation; day 3 or 4)	Ozone effects in DEX-treated rats (day 1 or 2)	Ozone effects in DEX-treated rats (adaptation; day 3 or 4)
Plethysmography measures	PEF (ml/sec), ET (msec), PenH	---	↑↑↑	---	↑↑↑	---
Lung injury	BALF protein and albumin (μg/ml)	---	↑↑↑	---	↑↑↑	↑
Lung inflammation	BALF macrophages/ml	---	↓	---	↑	---
	BALF neutrophils/ml	---	↑↑↑	---	↑↑↑	↑
Systemic inflammation	WBC x 106/ml	↓↓↓	↓↓	---	↓↓↓	↓↓↓
	Lymphocytes x 106/ml	↓↓↓	↓↓	---	↓↓↓	↓↓↓
Systemic cytokines	Serum TNF-α (pg/ml)	↓↓	↓↓	---	↓↓↓	↓↓
	Serum IL-6 (pg/ml)	---	---	----	---	---
BALF cytokines	TNF-α (pg/ml)	---	↑↑↑	---	↑	↑
	IL-6 (pg/ml)	---	↑↑↑		↑↑↑	↑
	IL-1β (pg/ml)	---	---	---	↑	↑
	KC-GRO (pg/ml)	---	---	---	↑	---
Lung mRNA for stress markers*	Il6	---	↑↑	↑	↑↑	---
	Tsc22d3 (Gilz)	---	↑↑	---	---	---
	Hbd11b1	---	---	---	---	---
	Hbd11b2	---	↓↓	---	---	---
Pituitary hormones-serum	ACTH (pg/ml)	↓	---	---	---	↓↓
	TSH, LH (pg/ml)	---	↓↓↓	---	---	---
	FSH (pg/ml)	↑	---	---	---	---
	BDNF (pg/ml)	↓	---	---	↓	↓
Adrenal hormones-serum	Epinephrine (pg/ml)	---	---	---	---	↑
	Corticosterone (pg/ml)	↓↓	---	---	↑	↓↓
Metabolic hormones-serum	Insulin (ng/ml)	↑↑	---	---	↑↑	---
	Leptin (pg/ml)	---	↑↑	↓	---	---
Systemic glucose metabolism	Hyperglycemia, glucose intolerance	↑	↑↑↑	---	↑↑↑	---
	Insulin intolerance	----	---	---	---	---
	Glucose-mediated insulin release	---	↓↓	---	↓↓	---
Adrenal mRNA-catecholamine synthesis enzymes*	Pah	---	↓↓	---	---	---
	Pnmt	---	↑↑	↑↑	↑↑	↓
Adrenal mRNA-glucocorticoid synthesis enzymes*	Hsd3b1, Cyp21a1, Cyp11b1,Cyp11b2, Hsd11-b2	---	↑↑↑	---	↑↑↑	↓

This summary table illustrates changes observed in most endpoints relative to control rats exposed to air at a given time point. No significant effects are shown by broken straight lines (---), small increases are shown as one arrow pointing up (↑), large increases are shown as two arrows pointing up (↑↑) and largest increases are shown as three arrows pointing up (↑↑↑). Likewise, small decreases are shown as one arrow pointing down (↓), large decreases are shown as two arrows pointing down (↓↓) and largest decreases are shown as three arrows pointing down (↓↓↓).

^*mRNA expressions are evaluated as relative fold change from control air group.

Figure 10.

A flow diagram of proposed mechanism and sequence of events after one- or two-day ozone exposure leading to pulmonary and systemic effects, and upon continued exposure, subsequent adaptation/habituation. The numbers indicate the sequence of major events where yellow highlighted number refer to one- or two-day ozone exposure effects and blue highlights show glucocorticoid feedback inhibition-related lack of ozone effects after 3 or more days of exposure (adaptation). The event numbers refer to: (1) Ozone inhalation (acute first and second day) leading to activation of stress centers in the brain (hypothalamus); (2) activation of SAM and HPA and inhibition of HPT and HPG axes; (3) activation of SAM and HPA axes leading to synthesis and release of adrenal catecholamines and glucocorticoids in the circulation; (4) pulmonary and systemic immune and metabolic effects mediated through changes in circulating adrenal and pituitary hormones; (5) glucocorticoid feedback inhibition of stress centers in the brain resulting in no activation of follow-up events upon repeated ozone exposure; including no glucocorticoid effect on the lung and possible lack of activation of hypothalamic stress response through lung; (6) no activation of SAM and HPA or inhibition of HPT or HPG axes; (7) no adrenal glucocorticoid or catecholamines production; (8) no pulmonary, systemic metabolic or immune effects leading to adaptation despite continued exposure. How lung communicates with brain is not fully understood but vagal sensory innervation might be involved. Ozone exposure has been shown to activate stress responsive regions in the brain (Gackière et al., 2011). SAM, sympathetic adrenal medullary; HPA, hypothalamic pituitary adrenal; HPT, hypothalamic pituitary thyroid, and HPG, hypothalamic pituitary gonadal axes. ACTH, adrenocorticotropic hormone; TSH, thyroid stimulating hormone, PRL, prolactin; LH, luteinizing hormone, FSH, follicle stimulating hormone.

Twenty months of high concentration continuous ozone exposure induces bronchiolar cell metaplasia and terminal bronchiolar collagen accumulation (Stockstill et al., 1995); however, it is not understood how initial neutrophil inflammatory response that occurs after ozone exposure is also linked to increases in alveolar macrophages and interstitial lymphocytes during long-term exposure. Although it is not clear if adaptation from initial lung injury and neutrophilic inflammation will continue with daily ozone exposure in a chronic scenario (over several months), it is well established that when ozone exposure occurs for 2–3 days per week followed by 4–5 days no exposure period, the adaptation response is lost on the subsequent week of exposure, suggesting that it is short-lived and may depend on the experimental protocol of differential exposure durations, concentration, and the degree of baseline response of an animal model (Hamade and Tankersley, 2009; Bass et al., 2013; Miller et al., 2016c; Snow et al., 2016). For example, ozone exposure for 2-days per week for 13 weeks diminishes the initial response to ozone at week 1 in terms of glucose intolerance in Brown Norway rats (Bass et al., 2013), however, in WKY rats, no such diminution in lung injury or glucose intolerance was observed with 3 days/week exposure (Miller et al., 2016c). We have noted concentration-dependent effects of ozone in pituitary and adrenal hormones during a single ozone exposure (Henriquez et al., 2021). In this study, 0.8 ppm ozone exposure for 4 hours led to adaptation in on day 3, however, at lower or higher ozone concentrations the dynamicity and the degree of adaptation might vary. Our data in WKY rats show that assessing lung injury/inflammation at D+2 and at D+4 provides a simplistic model to study ozone adaptation.

As far as lung injury and inflammation are concerned, the adaptation response we observed at D+4 was not substantially changed by DEX pretreatment, except for a small dampening of adaptation measured in lung vascular leakage (BAL protein and albumin) and inflammation (neutrophils). The precise mechanisms for this effect cannot be ascertained since DEX pretreatment, based on the inducibility of glucocorticoid responsive genes in the lung (Mittelstadt and Ashwell, 2001), did not affect pulmonary glucocorticoid response. For example, Tsc22d3 was induced in the lung with DEX regardless of ozone, suggesting active glucocorticoid signaling at cellular level. Nevertheless, ozone decreased the expression of Hsd11b2 despite increased Tsc22d3 in DEX-pretreated rats at D+2, which suggests that ozone exposure was associated with increased local glucocorticoid availability and its activity in the lung at D+2 but not at D+4. This together with adaptation in inflammatory gene expression and lung injury/inflammation, as well as lung functional outcomes at D+4 indicates the role of endogenous glucocorticoid release in adaptation response, likely at tissue and central levels.

The involvement of hypothalamic stress response pathway in ozone adaptation is further confirmed by the assessment of anterior-pituitary-derived hormones. The activation of SAM and HPA has been well established after ozone inhalation (Kodavanti 2019; Henriquez et al., 2021a). Ozone and other environmental stressors and non-chemical stressors have been shown to impact hypothalamic-pituitary-thyroid (HPT) and hypothalamic-pituitary-gonadal (HPG) axes, and depending on the type of stressor, there is activation or inhibition (Clemons and Garcia, 1980; Corbacho et al., 2004; Kirk et al., 2017; Kirby et al., 2009). We have shown that ozone activates anabolic neuroendocrine axes, such as SAM and HPA but inhibits HPT and HPG (catabolic) axes (Henriquez et al., 2021a and b). Because these changes are not observed in adrenalectomized rats (Henriquez et al., 2018b), one can presume that adrenal-derived hormones may modulate hypothalamic response to ozone. To further support this notion, we show in this study that pituitary hormones associated with HPT (such as TSH) and HPG (such as PRL and LH) were inhibited by ozone on D+2 but this inhibition was not noted at D+4, indicating that release of pituitary hormones is likely regulated and mediated in the hypothalamus through glucocorticoids.

Acute metabolic changes from ozone exposure have been recently noted in several our studies in rodents and humans, which are mediated through the activation of SAM and HPA axes (Miller et al., 2015; 2016b, 2016c). We show here that if daily ozone exposure continues at day 3, no hyperglycemia or glucose intolerance are noted. More importantly, ozone-induced abolition of beta cell insulin secretion in response to glucose observed after day 1 was lost at day 3, suggesting that the ability of adrenal-derived stress hormones to block pancreatic insulin secretion was lost, despite continuous ozone exposure. In our study, a 1-month low dose DEX, which does not penetrate brain but can exert feedback inhibitory effect on the pituitary gland (Karssen et al., 2005), and is shown to induce insulin resistance (Severino et al., 2002), did not affect ozone adaptation response. This suggests that ozone-induced metabolic changes might be mediated by other adrenal hormones (catecholamines) in addition to glucocorticoids.

While DEX pretreatment tended to increase epinephrine levels in the circulation, it diminished levels of corticosterone. Importantly however, in ozone exposed rats for D+2 group, the DEX-mediated inhibition of circulating corticosterone was not evident. This implies that DEX pretreatment did not inhibit ozone-mediated corticosterone increase at D+2, but this ozone effect was absent at D+4 even in DEX-pretreated rats, suggesting that pretreatment with DEX did not alter ozone adaptation in glucocorticoid increase. Since DEX has been shown to have a wide array of metabolic effects through peripheral glucocorticoid receptors stimulation, we wanted to determine how it influenced ozone-mediated glucose metabolic alterations. As has been reported in other studies (Severino et al, 2002; Chruvattil et al., 2017), DEX pretreatment at 0.01 mg/kg dose level increased circulating insulin, and caused a small degree of glucose intolerance but did not induce hyperglycemia in air-exposed rats. At double the concentration and 2 days of treatment, DEX was associated with hyperglycemia in our previous study in the same rat model (Henriquez et al., 2020). Surprisingly, at day 1 and/or 2, ozone exposure not only diminished DEX-mediated insulin increases but also abolished pancreatic release in response to a bolus glucose injection, suggesting that ozone effect on insulin release from pancreas are likely mediated through central regulation that may involve the role of catecholamines, while low-dose DEX might induce peripheral insulin resistance. These ozone-induced changes were absent at day 3, suggesting that adaptation is likely regulated centrally.

The mechanisms by which specific stressors may induce peripheral responses, and how glucocorticoids, in concert with different neural circuitries in brain centers determine the longevity of the stressor impact are highly complex (McEwen and Akil, 2020). We initially asked if reduced glucocorticoid availability was involved in ozone adaptation in our experimental setting. We have shown that if adrenal glands (where the major portion of circulating catecholamines and glucocorticoids are synthesized) were removed from animals, there was no effect of acute ozone exposure in the lung or in the brain (Miller et al., 2016b; Henriquez et al., 2019a), implying that neuroendocrine hormones play a key role in driving these effects. Unlike adrenalectomy, however, the adaptation response may involve the lack of stimuli for adrenals to produce higher than base-line levels of catecholamines and glucocorticoids in response to ozone. To determine the relative contribution of catecholamines and glucocorticoids in adaptation, we first analyzed transcription levels of genes coding enzymes that are involved in the synthesis of epinephrine and nor-epinephrine from phenylalanine. Our data show that the expression of Pnmt, the rate limiting enzyme in epinephrine production (Hu et al., 2012) was increased after ozone exposure on D+2 and this induction was also evident at D+4 (in VEH-treated rats only). Although in DEX-pretreated rats at D+4, the Pnmt expression was not increased, these data imply that the increase in epinephrine did not show adaptation at D+4 of ozone exposure. We next assessed mRNA expression for enzymes involved in corticosterone synthesis. Most transcripts of enzymes involved in corticosterone biosynthesis in rats were increased after ozone exposure on D+2 but not at D+4. This lack of induction coincided with failure of ozone to increase circulating corticosterone at D+4. Thus, these data show that ozone adaptation is linked to a lack of glucocorticoid response. It is noteworthy that in adrenalectomy, circulating glucocorticoid levels are diminished, however, during ozone adaptation, base-line physiological levels are preserved.

While circulating levels of glucocorticoids often fluctuate on an hourly basis due to ultradian changes and handling stress on animals (Joëls et al., 2012), the most consistent indication of increased glucocorticoid activity in our studies with ozone has been the depletion of circulating lymphocytes. DEX has potent immunosuppressive effects through its peripheral influence on lymphoid organs (Roholl et al., 1983; Taves and Ashwell, 2021), and therefore, we postulated that in the presence of DEX, adaptation from ozone-induced lymphopenia at D+4 will not be evident. We show here that ozone-induced leukopenia and lymphopenia observed at D+2 in VEH-pretreated animals was not observed at D+4, indicating adaptation. However, DEX caused leukopenia and lymphopenia regardless of ozone, suggesting a direct effect of DEX in peripheral lymphoid organs, likely masked any ozone-induced adaptation. This is further evident by decreases in circulating TNF-α by DEX at both time points and by ozone only at D+2 in vehicle-treated rats together with DEX-induced reduction in the weights of lymphoid organs. Since Tnfα gene transcription is regulated by glucocorticoids through glucocorticoid response element in lymphocytes (Esparza and Arch, 2006), it is likely that circulating TNF-α might originate from immune cells. This is unlike its levels in BALF, where increases were noted in DEX-pretreated rats. The inflammatory response to ozone in the lung is mediated through increased circulating adrenal-derived catecholamines, especially epinephrine (Henriquez et al., 2018), thus, might result in increased BALF TNF-α levels despite glucocorticoid influence.

This study neither confirmed if the observed ozone adaptation response could be replicated in other exposure scenarios, nor examined how brain changes in glucocorticoid mechanisms might mediate ozone adaptation. Moreover, it is also not clear if the adaptation upon discontinuation of ozone exposure would involve similar mechanism as the adaptation during continuous exposure. Furthermore, we did not study sex differences in the adaptation response and that in what conditions this adaptation response may be delayed or inhibited in females. Ozone exposure has been shown to have a greater pulmonary effect in males when compared to females (Osgood et al., 2019). Understanding how the impairment of stress axis in susceptible hosts with psychosocial, and environmental stressors might lead to chronic neurological and peripheral diseases remains an area of interest.

We conclude that adaptation upon continued ozone exposure occurs in not only pulmonary injury and inflammation but also systemic effects. Adaptation is evident in metabolic effects of ozone, including inhibition of pancreatic insulin secretion in response to glucose. And that this adaptation results in dampening of ozone effects on neuroendocrine hormones upon continued exposure for 3 or more days. We show that endogenous glucocorticoid levels despite being inhibited by DEX shows rebound upon ozone exposure at D+2 but fail to increase at D+4, which is tied with adaptation in expression of markers involved in glucocorticoid biosynthesis in adrenals, suggesting glucocorticoid availability is likely driving the adaptation response. Overall, our data suggests that ozone adaptation response might be orchestrated centrally, and that HPA axis-mediated glucocorticoid availability could play a key role (Figure 10).