Oxidative Stress in Ozone-Induced Chronic Lung Inflammation and Emphysema: A Facet of Chronic Obstructive Pulmonary Disease

Oxidative stress plays an important role in the pathogenesis of chronic obstructive pulmonary disease (COPD) caused by cigarette smoke and characterized by chronic inflammation, alveolar destruction (emphysema) and bronchiolar obstruction. Ozone is a gaseous constituent of urban air pollution resulting from photochemical interaction of air pollutants such as nitrogen oxide and organic compounds. While acute exposure to ozone induces airway hyperreactivity and neutrophilic inflammation, chronic ozone exposure in mice causes activation of oxidative pathways resulting in cell death and a chronic bronchial inflammation with emphysema, mimicking cigarette smoke-induced COPD. Therefore, the chronic exposure to ozone has become a model for studying COPD. We review recent data on mechanisms of ozone induced lung disease focusing on pathways causing chronic respiratory epithelial cell injury, cell death, alveolar destruction, and tissue remodeling associated with the development of chronic inflammation and AHR. The initial oxidant insult may result from direct effects on the integrity of membranes and organelles of exposed epithelial cells in the airways causing a stress response with the release of mitochondrial reactive oxygen species (ROS), DNA, and proteases. Mitochondrial ROS and mitochondrial DNA activate NLRP3 inflammasome and the DNA sensors cGAS and STING accelerating cell death pathways including caspases with inflammation enhancing alveolar septa destruction, remodeling, and fibrosis. Inhibitors of mitochondrial ROS, NLRP3 inflammasome, DNA sensor, cell death pathways, and IL-1 represent novel therapeutic targets for chronic airways diseases underlined by oxidative stress.


INTRODUCTION
Ozone is a gaseous constituent of urban air pollution that is generated by interaction of rising constituents of air pollution such as nitrogen oxide and organic compounds, induced by sunlight. Early studies have documented the effects of a short acute exposure to ozone in inducing airway inflammation and bronchial hyperreactivity in humans and in various species including mice, rats, guinea pigs, and dogs, at levels of ozone that were much higher than those measured in a highly polluted traffic-dense environment during the summer months. More than 92% of the world's population are regularly exposed to unhealthy levels of ozone and among the world's 11 most populous countries, population-weighted seasonal ozone concentrations range from about 45 ppb in Brazil to 68 ppb in China. In Europe, the average ozone levels are reported to be highest in Italy with an annual average of 65 ppb (https://www. eea.europa.eu/publications/air-quality-in-europe-2019). Ozone concentration EU targets for the protection of human health has been set not to exceed 60 ppb but these targets are very often exceeded in several EU countries ( Table 1). The European Air Quality Report highlights that 17 EU member states and six other reporting countries registered concentrations of ozone above the target value more than 25 times.
Over the recent years, there has been growing evidence from epidemiological and human exposure studies of the detrimental effects of ozone on respiratory health. Thus, short-term changes in ozone levels have been associated with increased mortality (1,2), and a positive association between ozone and hospital admissions for asthma and COPD in the elderly and between ozone levels and asthma emergency visits in children (3,4). Indeed, more recent studies have linked long-term exposure to ozone with reduced lung function and an increasing risk of developing emphysema irrespective of being a cigarette smoker (5,6).
The effects of ozone on inflammation have already been reviewed although these have usually looked at mainly acute exposures (7). One of the long-term interests of the effect of ozone has been to determine whether exposure to ozone can represent a good model for airways disease. Acute exposure to ozone in rats or mice and later in humans have been used as a model of bronchial hyperresponsiveness underlined by a neutrophilic inflammation, the mechanisms of this effect having been reviewed (7). However, it is the chronic exposure to ozone that led to a model of emphysema and chronic airway inflammation that has been of the most interest in recent years (8). This represented the most direct proof that oxidative stress could induce features similar to that seen in the condition labeled as chronic obstructive pulmonary disease (COPD).
In this review, we will consider by what mechanisms this oxidant stress can lead to chronic bronchial inflammation and lung destruction of emphysema, and how models of oxidative stress such as exposure to ozone particularly on a long term basis can inform us on the nature of COPD and can be used to examine potentially new treatments.

CHRONIC OBSTRUCTIVE PULMONARY DISEASE (COPD)
COPD is defined as a common, preventable, and treatable disease that is characterized by persistent respiratory symptoms and airflow obstruction that is due to airway and/or alveolar abnormalities usually caused by significant exposure to noxious particles or gases (9,10). This is a disease with high degree of morbidity and mortality throughout the world. While initially, this was considered to be a disease mainly found or caused by cigarette smoking, it is now agreed that exposure to biomass fuel and increasingly to environmental pollution can also contribute to the pathogenesis of COPD (11).
Some of the pathological features of smoking-induced COPD consist of severe airflow obstruction which is associated partly to the inflammation in the small airways particularly by neutrophils, macrophages, and lymphocytes (12). In addition, there is airway wall remodeling of the small airways, which involves all the components of the airway wall including epithelium, lamina propria, and airway smooth muscle (13,14). Finally, there is the presence of emphysema, with the destruction of alveolar airspaces, which has been attributed to an imbalance of protease and anti-protease activity with enhanced protease activity associated with neutrophil activation (15). Airway hyperreactivity (AHR) prevalence and pathology in COPD was found to be present in one out of two COPD patients with its presence influenced by decline in lung function and smoking status (16). As such, AHR can be seen as an independent predictor and a contributing factor to COPD development.
Various pathological pathways have been proposed including increased apoptosis mechanisms (17,18) in addition to inflammageing and autoimmunity (19). There is growing evidence that COPD may represent a cellular senescence program associated with secretion of pro-inflammatory cytokines that cause chronic inflammation, that leads to the increased levels of reactive oxygen species that can induce oxidative stress (20). Oxidative stress mechanisms can also lead to activation of pro-inflammatory pathways in addition to causing DNA damage (21).
An increasing body of evidence has been accumulating linking COPD with the process of autoimmunity, with the presence of autoantibodies in the sera of COPD patients, and with some of the antibodies correlating with particular disease phenotypes (19). Thus, it has been proposed that autoimmunity may play an important role in the pathogenesis of COPD. Interestingly, ozone-exposed mice for 6 weeks exhibited increased antibody titres to carbonyl-modified protein, as well as activated antigenpresenting cells in lung tissue and splenocytes sensitized to activation of carbonyl-modified protein (22). This is evidence of oxidative stress induced antibodies supporting an auto-immune component, as has been described in COPD with the elevation of carbonyl-modified self-protein that correlated with the severity of disease. Cigarette smoke models have also supported this effect of ozone as has been shown in matrix metalloproteinase 12 (MMP12)-generated elastin fragments serving as a selfantigen and driving the cigarette smoke-induced autoimmune processes in mice that result in a bronchitis-like phenotype and airspace enlargement (23). These elastin autoantibodies and others have been described in advanced COPD (24). The fact that autoantibodies in animal models of COPD such as after ozone exposure are capable of inducing a COPD-like disease phenotype indicate an autoimmune mechanisms perhaps involving B cells, plasma cells, and B cell-rich lymphoid follicles that are present in COPD (25).

MECHANISMS OF OZONE-INDUCED OXIDATIVE STRESS ON INFLAMMATION AND AIRWAY REMODELING
It is quite clear that ozone reacts with cellular membranes and with the epithelial lung lining fluid to generate bioactive mediators that cause oxidative stress, innate immune responses, and signaling (26,27). When inhaled, there is direct contact between ozone and the first level of cells in the airway surface such as airway and alveolar epithelial cells, and airway macrophages. These cells release reactive oxygen species and various other inflammatory mediators including cytokines and lipids from oxidative damage to the airway epithelium (28,29). In addition, ozone exposure has been shown to impair macrophage phagocytic and efferocytosis function (28,30) which can cause prolonged injury and inflammation. The mechanisms underlying these changes have been previously reviewed in terms of the transcriptional effects of ozone in the lungs and airways (31,32).

Effects of Ozone Exposure on Oxidative Stress and Intracellular Signaling Pathways
The increased oxidative stress as a direct effect of ozone is often investigated in relation to cellular dysfunction. Ozone affects several intracellular pathways in different cell types, but not always in the same way. Ozone exposure in cultured alveolar epithelial cells results in cytotoxicity but does not always affect the production of cytokines (33). However, when conditioned medium from ozone-exposed alveolar macrophages is added to alveolar epithelial cells, the cytokines chemokine (C-X-C) ligand 1 (CXCL1) and C-C motif chemokine ligand 2 (CCL2) were induced, mediated through interleukin-1α (IL-1α) (34). On the other hand, it has been reported that, although human alveolar macrophages are much more sensitive to ozone than epithelial cells, they do not produce increased amounts of IL-6, IL-8, or fibronectin following ozone exposure. This increased sensitivity of alveolar macrophages was observed in the form of increased cell death with a third of the cells lost after a low ozone exposure dose. However, whether macrophages are more sensitive to ozone in vivo is not known, but bronchial airway epithelial cells produce substantially more of all three proteins following ozone exposure, and both IL-6 and fibronectin are secreted vectorially toward the apical side at least in the first 4 and 24 h after ozone exposure (29). Ozone exposure has been shown to induce the expression of intracellular adhesion molecule-1 (ICAM-1) and neutrophil adhesion to human airway epithelial cells, which is mediated through the Canonical Transient Receptor Potential Channel 6 (TRPC6) (35). TRPC6 also impacts neutrophilic inflammation by nuclear factor-κB (NF-κB) activation and therefore promoting transcription of inflammatory mediators (35). Therefore, cellular interactions such as those between alveolar macrophages and, airway and alveolar epithelial cells which are the initial airway cells encountered by ozone may be important in the induction of its effects. The role of TRPC6 during oxidative stress conditions has been investigated in several cell types. Oxidative stress induces TRPC6 expression and function in podocytes (36), HEK293T cells (37), vascular myocytes (38), neutrophils (35), and macrophages (39). In a recent study, TRPC6 in bronchial epithelium cells, was shown to act as an oxidative stress sensor where the TRPC6-mediated calcium cascade leads to the activation of the extracellular signal-regulated kinase (ERK) pathway and inflammation (40). This could be a possible inflammatory response pathway in several cell types in response to ozone, but this has yet to be described in COPD.
Ozone exposure activates several types of innate immune cells which play a role in both the Th1 and Th2 inflammatory responses. While the role of neutrophils as part of the Th1 response is well-established, other innate cell such NK cells (41) and innate lymphoid type 2 (ILC2) cells (42) may also contribute. Surfactant protein D released by exposure to ozone has been shown to stimulate interferon (IFN)-γ by NK cells to initiate IFNγ, IL-12 feedback circuit (41). Continuous exposure to ozone has been shown to result in the conversion of an initial neutrophilic inflammation to an eosinophilic inflammation in the nasal mucosa of mice (43). This was accompanied by an overexpression of type-2 cytokines in the nasal epithelium. The role of ILC2 cells was confirmed by showing that the epithelial-derived alarmins, IL-33, IL-25, and thymic stromal lymphopoietin (TSLP) that were increased in the nasal epithelium was dependent on ILC-2 cells, but not on adaptive T or B lymphoid cells (44). The involvement of ILC2 cells is also supported by the observation that IL-13, a product of such cells, can augment the effects of single ozone exposure, namely AHR and inflammation (45). In addition, ovalbumin sensitization and challenge can also augment ozone exposure effects (46).
Several intracellular pathways are involved in the effects of ozone exposure. Prolonged exposure to ozone induced gene expression levels of several hypoxia inducible factor-1α (HIF-1α) target genes including histone deacetylase 2 (HDAC2), vascular endothelial growth factor (VEGF), Kelch-like ECH-associated protein 1 (Keap1), and Macrophage migration inhibitory factor (MIF), in addition to a decrease in the antioxidative stress response as indicated by an increase in nuclear erythroidrelated factor 2 (Nrf2) activity and protein level (47). The Toll like receptors (TLRs) and downstream adaptor proteins myeloid differentiation marker 88 (MyD88) and toll-interleukin 1 receptor domain containing adaptor protein (TIRAP) appear to be involved in the inflammation response to ozone (48,49). Ozone is able to induce TLR4 signaling through MyD88 in normal Tlr4 expressing C3H/HeOuJ mice. This in contrast to C3H/HeJ mice that express a defective dominant negative mutant Tlr4 gene (50). TLR4 can be activated by ligand binding and subsequently recruits adapter proteins including MyD88. The MyD88-dependant pathway signals through several signal transduction pathways including MAPK, NF-κB, and AP-1 to induce cytokine gene expression contributing to the inflammation response toward ozone. Heath shock protein 70 (HSP70) was identified as a downstream mediator to have a role the TLR4 mediated effects by ozone (50). The ozoneinduced oxidative stress response might be involved in the activation of the NACHT, LRR, and PYD domains-containing protein 3 (NLRP3) inflammasome complex (51). Both the acute and chronic exposure of mice to ozone induces lung inflammation through activation of ROS and of the NLRP3 inflammasome with caspase activation (52,53) and release of the highly inflammatory mediator IL-1β, but also IL-1α (54), which are involved in the ozone injury response (55). Uric acid crystals cause NLRP3-dependent lung inflammation upon injury (56) and therefore it is not excluded that crystals formed during injury may contribute to ozone induced inflammation (57).
How this cytoplasmic multiprotein complex is activated upon ozone exposure remains to be determined. One of the mechanisms might involve reactive oxygen species (ROS) and oxidative stress induced danger signals as reviewed recently (58). The transcriptional effects of ozone and the subsequent impact on airway inflammation have recently been reviewed in an article in this series (31).

Reactive Oxygen Species-Mediated Barrier Disruption, Inflammation, and Emphysema
Ozone reacts with cellular membranes and with the epithelial lung lining fluid to generate bioactive mediators that cause oxidative stress, innate immune responses, and signaling (26,27). Ozone-induced ROS at the cell membrane influences membrane integrity with the disruption of tight junction, cell stress, and death with leak of the respiratory barrier within 1-2 h. This first phase is independent of cellular ROS, inflammatory mediators, and inflammatory cells (58). These effects might be the result of ozone exposure induced changes in the expression of claudin (CLDN) proteins which are protein components of tight junctions. In mice exposed to ozone an increased expression levels of CLDN3 and CLDN4 while a reduction in CLDN14 was reported (59). These workers also showed that bronchial epithelial cell integrity was disrupted and that the trans-electrical resistance between cells was decreased leading to cell disintegration. These changes were associated with elevated ROS and increased expression of antioxidant defense system involving Nrf2 and Keap1 (59). Another potential mechanism is lipid peroxidation, which is induced by ozone in human alveolar epithelial cells (60) and in lung surfactant (61). Phospholipids and cholesterol in cell membranes and surfactant can react with ozone to form cytotoxic products which activate second messenger systems involving free arachidonic acid (62), plateletactivating factor (PAF) (63) and prostaglandin E2 (PGE2) (64). Ozone exposure induced changes to cell membrane integrity might involve several pathways, dependent on dose and duration of exposure.
The acute respiratory barrier disruption by ozone has been recently reviewed in an accompanying review in this series (65). This first phase is followed by a robust barrier injury with cell death, protein leak, and influx of ROS expressing myeloid cells including neutrophils, IL-1α and IL-33 production by epithelial and myeloid cells within 6-12 h. Thus, a biphasic response is observed, an immediate direct membrane damage by ROS and a second phase mediated by myeloid cells, which aggravates the damage of the epithelium and inflammation. In support of this hypothesis is the fact that neutrophil depletion attenuates the second phase (58). Oxidative stress peaks at ∼18 h after a single ozone exposure during the second phase with increased ROS positive neutrophils and epithelial cells (66). This increase in ROS subsequently causes cell injury, mitochondrial dysfunction, formation, and release of toxic metabolites and even DNA damage. These events have been associated with and contribute to the development of tissue destruction leading to emphysema and lung remodeling.
The toxic effect of ozone was also observed in the distal part of the lung, the alveoli, with death of alveolar epithelial cells with defective repair resulting in enlarged air spaces and emphysema (8). Emphysema upon chronic ozone exposure is commonly found in mice and is dependent on ROS-dependent inflammation (8), but this does not involve IL-17 (67). Understanding the initial physico-chemical and molecular events of ozone-induced cell membrane injury in vitro and in vivo is an area of intense research leading to cell death will be important. Inflammation and cell death are also associated with neutrophil extracellular trap (NET) formation with chromatin externalization (68) and release of extracellular DNA, which are inflammatory activating the DNA sensing pathways cGAS/STING as reviewed for lung inflammation (69). Aerosol particles such as silica and likely others including ozone cause extracellular release of nuclear and mitochondrial DNA from dying cells, which drive cGAS/STING-type I interferon dependent inflammation, which is inhibited by DNase administration or blockade of STING (70).
In ozone-exposed mice, ROS-induced cell-death was dependent on interactions of Keap1, the cellular ROS sensor and the phosphatase phosphoglycerate mutase family member 5 (PGAM5) with antioxidant function and apoptosis inducing factor mitochondria associated 1 (AIFM1), a pro-apoptotic factor. At high ROS concentration PGAM5 is released from the complex and activates the pro-apoptotic factor AIFM1 inducing apoptotic pathway resulting in a new form of cell death known as oxeiptosis (71)(72)(73). However, the contribution of the different cell death pathways such as necroptosis, apoptosis, and efferocytosis by ozone is yet resolved (74). These cell death pathways are likely to be responsible for the emphysematous process seen on oxidant-induced emphysema.

EFFECT OF OZONE EXPOSURE ON OXIDATIVE STRESS AND MITOCHONDRIAL DYSFUNCTION
Ozone exposure in ex vivo and in vivo experimental models has focussed recently on the mitochondrial effects of ozone. Mitochondria are double membrane bound organelles that exist in most eukaryotic organisms. Mitochondria play an important role in production of adenosine triphosphate (ATP) and mitochondrial ROS production (75). Dysfunctional mitochondria influence airway contractility, gene expression, oxidative stress, cell proliferation, apoptosis and metabolism, and immune and inflammatory responses that are all implicated in airway diseases including COPD (76).

Influence of Age on Acute and Chronic Ozone Exposure
The toxic effects of ozone on mitochondria are closely dependent upon the dose and duration of ozone exposure and age of animals ( Table 2). Short-term ozone exposure (2 ppm for 8 h or 4 ppm for 4 h) to adult rats caused decreases in mitochondrial O 2 consumption, mitochondrial ATP synthesis and oxidation of thiol groups in mitochondria in lung, and an increase in lung mitochondrial permeability (77). In vitro mitochondrial respiration in lung homogenates and in lung mitochondria from adult rats (2-3 months) was decreased after acute ozone exposure (15 ppm, 20 min). Subacute ozone exposure (0.8 ppm, for 10-20 days) to adult rats led to increased mitochondrial O 2 consumption in lung homogenates, increased mitochondrial number and mitochondrial respiratory activity in alveolar type II cells in the lung (81). Acute ozone exposure (3 ppm for 8 h) in both adult (4-6 months) and old rats (24-26 months) decreased mitochondrial respiration and mitochondrial O 2 consumption in isolated lung mitochondria (78). In young (3 weeks) and adult rats (6 months), a more chronic ozone exposure (0.5 ppm, 12 h, 7 days) did not affect lung mitochondrial O 2 consumption. However, in aged rats (20 months), ozone increased the mitochondrial O 2 consumption and H 2 O 2 release, which means increased mitochondrial ROS activity (80). This indicates that ozone may accelerate senescence process in the elderly.
In mice, a single ozone exposure (3 ppm for 3 h) resulted in enhanced cellular and mitochondrial ROS levels and a reduced mitochondrial membrane potential ( m) in lung. (83). Upon acute and chronic ozone exposures, increased mitochondrial ROS levels, decreased ATP content, decreased electron transport chain (ETC) complex I enzyme activity, and reduced expression of ETC complex I, complex III, and complex V in the lungs of 1-and 6-week ozone-exposed mice (3 ppm, 3 h, twice a week) have been observed. Furthermore, ozoneinduced inflammation, airway hyperreactivity, mitochondrial dysfunction, and ROS levels were reduced when mice were administered the mitochondrial antioxidant MitoQ (79). In addition, ozone induced inflammation, increased mitochondrial ROS, and expression of ETC complex II and IV in lung mitochondria in 6-week ozone exposed mice (2.5 ppm, 3 h, twice a week) was reduced by treatment with MitoTEMPO, another mitochondria-targeting antioxidant (82). Therefore, targeting the mitochondrial dysfunction to prevent or treat the ozoneinduced associated inflammation, airway hyperresponsiveness, and oxidative stress could be a promising treatment strategy. AHR is a hallmark of ozone effects on the airways (66, 67) and is exacerbated in allergen-sensitized mice (84)(85)(86). AHR is due to a direct effect of ROS on bronchial smooth muscle cells (79), but also neuronal cells (87). The effect of ROS on cell membrane integrity is likely due to a direct disruption of the cell membrane structure which needs further studies (88). This is further supported by the effect of N-acetylcysteine (NAC) in reversing established AHR after chronic exposure to ozone (89). Studies of the airway smooth muscle from mice exposed to ozone showed that the hyperresponsiveness to cholinergic contractile agents can be reproduced in vitro and that this oxidant stressinduced hyperresponsiveness was dependent on the activation of the p38 mitogen-activated kinase (MAPK) and inhibited by corticosteroids (52). The AHR induced by chronic ozone exposure was also dependent on IL-17 (67), as was also shown by Pichavant et al. who reported that NK cells producing IL-17 was important for the maintenance of ozone-induced AHR (86). The pulmonary inflammation induced by subacute ozone exposure has been reported to require γδ T cells and tumor necrosis factorα (TNFα)-dependent recruitment of IL-17A + γδ T cells to the lung (90).

THERAPEUTIC STRATEGIES TARGETING INFLAMMATION, OXIDATIVE STRESS, AND MITOCHONDRIA
Several studies have investigated the effects of novel treatment strategies relating to the effects of ozone exposure in animal models ( Table 3). Targeting inflammation, oxidative stress, and mitochondria could be part of future treatment strategies for COPD.

Corticosteroids and ISO-1
Corticosteroids can attenuate the single exposure effects of ozone including AHR and lung inflammation (95)(96)(97). This includes the inhibition of expression of macrophage inflammatory protein 2 (MIP-2), inducible nitric oxidase synthase (iNOS) (98,99) and NFκB (96), and the increased proliferation of the airway epithelium (91). However, chronic exposure to ozone itself can induce a state of corticosteroid insensitivity similar to what has been observed in COPD with reduced or little effect in preventing AHR, inflammation, and emphysema (94,100). Corticosteroid insensitivity occurs in COPD through mechanisms induced by reactive oxygen species (101,102). Several such mechanisms have been postulated, including a reduction in HDAC2 activity and expression, impaired corticosteroid activation of the glucocorticoid receptor (GR) and increased pro-inflammatory signaling pathways. Therefore, the chronic model of ozone exposure represents a good model of COPD to study the mechanisms of corticosteroid insensitivity. Macrophage migration inhibitory factor (MIF) has been implicated as a driver of inflammation in COPD, and possibly as a driver of corticoidsteroid insensitivity in COPD. Using (S,R)3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid methyl ester (ISO-1) which inhibits MIF tautomerase activity, the corticosteroid-insensitive lung inflammation and AHR after chronic ozone exposure was blocked (94). Thus, inhibition of MIF which is elevated in COPD may provide a novel antiinflammatory approach in COPD. However, the contribution of mitogen-activated protein kinase phosphatase-1 (MKP-1) that has been proposed to underlie CS insensitivity in COPD was found to be negligible in the chronic ozone model (100). Single ozone exposure aggravated airway inflammation, airway remodeling, activation of p38 MAPK, and downregulation of MKP-1 in ovalbumin (OVA)-sensitized and -challenged mice, an effect that was ineffectively controlled by corticosteroids (46), this also supported the role of p38 MAPK activation as a likely pathway involved in corticosteroid insensitivity (103).

Hydrogen Sulfide (H 2 S)
Hydrogen sulfide (H 2 S), a metabolic product of methionine, is synthesized from L-cysteine primarily by three key enzymes: cystathionine-c-lyase (CGL), cystathionine-b-synthetase (CBS) and 3-mercaptypyruvate sulfurtransferase (MPST). Identified as the third gasotransmitter, along with nitric oxide and carbon monoxide, H 2 S modulates a variety of physiological functions including anti-oxidative stress, anti-senescence/aging, and anti-apoptotic effects (104)(105)(106). H 2 S has been proposed as serving as a potent antioxidant through reactive oxygen species/reactive nitrogen species scavenging, or through posttranslational modification of proteins by addition of a thiol (-SH) group onto reactive cysteine residues (107). In the lungs, H 2 S suppresses the airway smooth muscle proliferation and cytokine release, an effect that is less effective in muscle from COPD patients (108).
In addition, H 2 S content is reduced in the lungs of smokers and COPD patients (109), and H 2 S attenuates nicotineinduced endoplasmic reticulum stress and apoptosis in bronchial epithelial 16HBE cells (110). H 2 S is able to prevent and treat the development of inflammation, AHR, and oxidative stress in acute ozone-exposed mice (92). In addition, the ozone-induced increase in p38 MAK signaling was reduced in mice treated with H 2 S indicating that this intracellular signaling pathway might be involved (92). In chronic ozone-exposed mice, H 2 S is also able to prevent the inflammation, AHR, and remodeling of the lung but is not able to reverse these hallmarks of the model (93). Induction of p38 MAPK signaling is also reversed in both treatment strategies in the chronic ozone exposure model. In addition to this, the activation of the NLRP3 inflammasome and the ratio between cleaved caspase-1 to pro-caspase-1 were positively correlated with changes in lung function parameters and structural changes in the lung. H 2 S was able to prevent and treat the changes observed in NLRP3 activation (93). Taken together, the main difference between the acute and chronic ozone exposure models on the treatment effects of H 2 S on the ozone-induced changes suggests that H 2 S treatment only affects the damage inducing pathways (oxidative stress) and not the regenerative pathways of the lung (92,93). It is of interest that H 2 S donor NaHS significantly inhibits cigarette smoke-induced mitochondrial dysfunction, oxidative damage, cell senescence, and apoptosis in alveolar epithelial A549 cells (111). These findings provide novel mechanisms underlying the protection of H 2 S against ozone and cigarette smoke-induced COPD and suggest that H 2 S donors targeted toward mitochondria may be beneficial in the treatment of COPD.

NLRP3 Inflammasome Inhibitor, Belnacasan (VX-765)
In addition to H 2 S, Belnacasan or VX-765 inhibits NLRP3 inflammasome activation effects by inhibiting caspase-1. In acute ozone exposed mice VX-765 is able to prevent bronchoalveolar lavage (BAL) inflammatory markers, and AHR. Mitochondrial oxidative stress was reduced and this was associated with lower expression levels of dynamin-related protein 1 (DRP1) and mitochondrial fission factor (MFF), and increased expression of Mitofusin 2 (MFN2) proteins involved in mitochondrial fission and fusion, respectively (53). Similar effects were observed in chronic ozone exposed mice were VX-765 is able to prevent inflammation, emphysema, airway remodeling, and oxidative stress while it decreased the expression of the fission protein DRP1 and MFF with affecting proteins involved in fusion dynamics. (82). Mitochondrial oxidative stress and NLRP3 inflammasome were driving ozone-induced inflammation processes and targeting these specifically might have therapeutic value in COPD.

Mitochondrial-Targeted Antioxidants
Apocynin, reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase inhibitor decreased the proliferation in bronchial epithelium after an acute exposure to ozone but not the inflammation (91). Ozoneinduced inflammation, airway hyperreactivity, mitochondrial dysfunction, and ROS levels were reduced when chronicallyexposed mice were pre-administered the mitochondrial directed antioxidant, MitoQ (79). Similarly, ozone-induced inflammation, increased mitochondrial ROS, and expression of ETC complex II and IV in lung mitochondria in 6-week ozone exposed mice was reduced by treatment with mitoTEMPO, another mitochondriatargeting antioxidant (82). However, airway remodeling and airflow obstruction were not (82). Similarly, in single exposure to ozone, mitoTEMPO inhibited mitochondrial ROS without affecting inflammation and bronchial hyperresponsiveness (53).
Mucolytic/antioxidant agents such as erdosteine, carbocysteine, and NAC reduced the risk of acute exacerbations in patients with COPD (112). In the chronic ozone exposure model, preventive NAC reduced the number of BAL macrophages and airway smooth muscle (ASM) mass while therapeutic NAC reversed AHR, and reduced ASM mass and apoptotic cells (89). Thus, NAC could represent a treatment for protecting against the oxidative effects of ozone and other pollutants, as well as an agent for reducing exacerbations of COPD.

Downstream Signaling Pathways
As documented above, several intracellular pathways have been implicated in the effects of single or multiple exposures to ozone in the mouse. These pathways are related to the control of several key transcription regulatory factors including NF-κB, antioxidant factors such as Nrf2, the p38 MAPK, and priming of the immune system by up-regulating toll-like receptor expression. Thus, in the single ozone exposure model, AHR and inflammation was inhibited by a c-jun NH2 terminal kinase (JNK) inhibitor (SP600125) (113), p38 MAPK inhibitor (SD282) (46,114) and NF-κB inhibitor (115). VX-765, an inhibitor of NLRP3 inflammasome, prevented lung inflammation and AHR caused by acute exposure to ozone (53), and also inhibited lung inflammation and emphysema from chronic exposure (82). The biphasic response to ozone starts with an immediate intracellular reactive oxygen species (ROS) and inflammation independent phase that is induced by (Continued) Frontiers in Immunology | www.frontiersin.org 8 September 2020 | Volume 11 | Article 1957 FIGURE 1 | extracellular oxidative stress. ROS induces membrane damage with changes in cell membrane integrity, disruption of tight junctions, epithelial cell stress, and death. Inflammatory mediators including IL-25 and IL-33 are released and attract innate immune cells such as natural killer (NK) cells and innate lymphoid type 2 (ILC2) cells. Cell death and release of dsDNA can induce an interferon (IFN) inflammation response via the cGAS/STING pathway. Processes activated during the intracellular ROS and inflammation-dependent phase of the effects of ozone include transcription factor-mediated inflammatory response and the activation of the antioxidant defense mechanism. Several inhibitors have been shown to prevent or treat the pro-inflammatory gene expression and subsequently inhibit the inflammatory response, including the JNK inhibitor, SP600125, and the p38 MAPK inhibitor, SD282. In addition, activation of the endogenous antioxidant defense system involving HADC2, Keap1, and Nrf2 may be sufficient to counteract the oxidant stress during acute ozone exposure but may be overwhelmed during chronic ozone exposure. Treatment with the antioxidant N-acetylcysteine (NAC) also reduces the inflammatory response, by scavenging of intracellular ROS with subsequent reduction of cytokine and chemokine production. Mitochondrial oxidative stress and mitochondrial dysfunction contribute to apoptotic processes and the activation of the NLRP3 inflammasome further enhancing the inflammatory response. Several treatment strategies targeting the mitochondria have been able to reduce or prevent the mitochondrial oxidative-induced dysfunction. These include several mitochondrial-targeted antioxidants such as MitoQ, MitoTEMPO, and SS-31. In addition, stem cell therapy with induced pluripotent stem cell-derived mesenchymal stem cells (iPSC-MSCs) prevented ozone-induced mitochondrial dysfunction and inflammation which may result from direct interaction and mitochondrial transfer between the iPSC-MSCs and airway cells. Treatment with the caspase-1 inhibitor VX765 and hydrogen sulfide (H 2 S) prevent the activation of the inflammasome and reduces inflammation and mitochondrial oxidative stress. The MIF inhibitor ISO-1 blocks the ozone exposure induced inflammation and airway hyperreactivity and might have an impact on the corticosteroid insensitivity present in chronic ozone exposed lungs. Corticosteroids reduce inflammation induced by acute ozone exposures but fail to affect these processes in the steroid insensitive chronic ozone exposed lung. The ozone exposure driven intracellular processes contribute to the inflammatory cytokine and chemokine production, immune cell recruitment, and eventually the development of airway hyperreactivity, airway obstruction, airway remodeling, emphysema, autoimmunity, and steroid insensitivity which are hallmarks of COPD.

iPSC-MSC Mitochondrial Transfer
Mitochondrial transfer from induced pluripotent stem cellderived mesenchymal stem cell (iPSC-MSC) offered protection against oxidative stress-induced mitochondrial dysfunction in human airway smooth muscle cells (ASMC) and in mouse lungs exposed to ozone while reducing airway inflammation and hyperresponsiveness (83). Direct co-culture of ASMCs with iPSC-MSCs protected the former from cigarette smoke-induced mitochondrial ROS production, mitochondrial depolarization, and apoptosis. When ASMCs were exposed to supernatants from iPSC-MSCs or transwell inserts with iPSC-MSCs, only cigarette smoke-induced mitochondrial ROS, but not mitochondrial depolarization and apoptosis in ASMCs, were improved, indicating that soluble factors from iPSC-MSCs reduced production of mitochondrial ROS. When there was direct contact between iPSC-MSCs and ASMCs, mitochondria were transferred from iPSC-MSCs to ASMCs, possibly through formation of tunneling nanotubes, an effect that was enhanced by cigarette smoke medium (CSM) treatment. iPSC-MSCs prevented, but did not reverse, ozone-induced mitochondrial dysfunction, AHR, and airway inflammation in the mouse model of single ozone exposure, an effect resulting from direct interaction and mitochondrial transfer between iPSC-MSCs and airway cells. Therefore, transfer of mitochondria from IPSC-MSC cells to replace damaged mitochondria by oxidative stress may present a novel approach to treating conditions such as COPD.

CONCLUSION
As an important component of air pollution, ozone has been closely related to the development of COPD. The link comes from two sides: on the one hand, chronic exposure to ozone in murine model of lung inflammation and emphysema and on the other, long-term exposure to ambient air pollutant such as ozone, has been associated with increases in emphysema evaluated by computed tomographic imaging with chronic airflow obstruction. The mechanisms of ozone-induced lung and airway changes are the release of inflammatory factors such as IL-1α, IL-6, IL-8, CXCL-12, CCL2, ICAM-1, KEAP-1, and MIF, the activation of intracellular pathways such as the MAPK pathway, TLR, cell death pathways, NLRP3 inflammasome, and NF-κB, the induction of oxidative stress through a decrease in the antioxidative response and an increase in the production of ROS, with a detrimental effect on mitochondrial function such as increased mitochondrial ROS, decreased ATP content and abnormal ETC complex (Figure 1). Other mechanisms include the disruption of airway epithelial barrier, the development of AHR and emphysema and the state of CS insensitivity. Inhibitors of mitochondrial ROS, NLRP3 inflammasome, DNA sensor, cell death pathways and IL-1, including hydrogen sulfide, apocynin, mitochondrial-targeted antioxidants such as MitoQ and MitoTEMPO, mucolytic/antioxidant agents such as NAC, VX765, and iPS-MSCs mitochondrial transfer represent novel therapeutic options for treating oxidative stress-induced COPD.

AUTHOR CONTRIBUTIONS
CW, FL, DT, and KC wrote the review. KC and BR inspired the work and corrected the manuscript. All authors have read the contribution.