Ebselen reduces cigarette smoke‐induced endothelial dysfunction in mice

Background and Purpose It is well established that both smokers and patients with COPD are at a significantly heightened risk of cardiovascular disease (CVD), although the mechanisms underpinning the onset and progression of co‐morbid CVD are largely unknown. Here, we explored whether cigarette smoke (CS) exposure impairs vascular function in mice and given the well‐known pathological role for oxidative stress in COPD, whether the antioxidant compound ebselen prevents CS‐induced vascular dysfunction in mice. Experimental Approach Male BALB/c mice were exposed to either room air (sham) or CS generated from nine cigarettes per day, 5 days a week for 8 weeks. Mice were treated with ebselen (10 mg·kg−1, oral gavage once daily) or vehicle (5% w/v CM cellulose in water) 1 h prior to the first CS exposure of the day. Upon killing, bronchoalveolar lavage fluid (BALF) was collected to assess pulmonary inflammation, and the thoracic aorta was excised to investigate vascular endothelial and smooth muscle dilator responses ex vivo. Key Results CS exposure caused a significant increase in lung inflammation which was reduced by ebselen. CS also caused significant endothelial dysfunction in the thoracic aorta which was attributed to a down‐regulation of eNOS expression and increased vascular oxidative stress. Ebselen abolished the aortic endothelial dysfunction seen in CS‐exposed mice by reducing the oxidative burden and preserving eNOS expression. Conclusion and Implications Targeting CS‐induced oxidative stress with ebselen may provide a novel means for treating the life‐threatening pulmonary and cardiovascular manifestations associated with cigarette smoking and COPD.


| INTRODUCTION
Chronic obstructive pulmonary disease (COPD) is a major incurable global health burden and is currently the fourth largest cause of death in the world (The World Health Organisation, 2018). Approximately 50% of COPD patients will die from a cardiovascular event (Sin & Man, 2005), and consequently, the pathobiological mechanisms linking COPD to cardiovascular disease are now an area of intensive research. Each puff of cigarette smoke contains >10 16 free radicals per puff, driving oxidative stress and tissue damage (Bartalis et al., 2007). Such oxidative stress and inflammation alter pulmonary blood vessel structure, through driving vascular remodelling, and promoting arterial stiffness and atherosclerosis (Sin et al., 2006). Vascular tone is controlled by vasoactive substances such as NO and PGs, and their secretion can be maintained by circulating oxygen levels (Chan & Vanhoutte, 2013). Under hypoxic conditions, as those seen in COPD, dysregulation of this vascular homeostatic balance occurs, due to oxidative damage to the vascular endothelial cells (VECs) leading to impaired NO production, thereby promoting endothelial dysfunction (Chan & Vanhoutte, 2013). NO is a key vasodilator produced by endothelial NOS (eNOS). Under normal physiological conditions, increased sheer stress on the vascular endothelium stimulates mechanosensitive ion channels, triggering a rapid influx of Ca 2+ into the cytoplasm of the VECs. This increases eNOS activity via myoendothelial gap junctions, that transmit vasodilatory NO signals to the underlying smooth muscle cells. VECs appear to be sensitive to oxidative damage, which may be the result of the conversion of NO to peroxynitrite (ONOO − ) in the presence of the harmful ROS; superoxide anion (O 2 − ) that ultimately reduces vascular NO bioavailability (Endemann & Schiffrin, 2004;Kolluru et al., 2012;Tabit et al., 2010). An altered oxidative balance in VECs has been demonstrated to favour cardiovascular events such as atherosclerosis, myocardial infarction (MI) and stroke Endemann & Schiffrin, 2004;Kolluru et al., 2012;Tabit et al., 2010).
Although cardiovascular co-morbidities are the largest cause of mortality in COPD, the detrimental effects of CS and its associated oxidative stress on the systemic vasculature remain largely unknown.
Given the deleterious role of oxidative stress in COPD, antioxidant treatment may be a viable therapeutic approach to treat the cardiovascular manifestations associated with this disease. We have previously shown that the antioxidant ebselen (2-phenyl-1,2-benzisoselenazol-3(2H)one), an organoselenium GSH peroxidase (Gpx) mimetic, inhibits CS-induced lung inflammation in mice (Duong et al., 2010). Ebselen treatment may also be an effective therapeutic in chronic diseases such as atherosclerosis, thrombosis, and stroke where oxidative stress and inflammation play a crucial role (Azad & Tomar, 2014;Sarker et al., 2003;Sarma & Mugesh, 2008;Takasago et al., 1997). Moreover, studies have shown that Gpx-1 deficient mice have enhanced pulmonary inflammation (Duong et al., 2010), as well as worsened cardiovascular outcomes including a larger infarct volume following ischaemic stroke (Crack et al., 2001;Duong et al., 2010), suggesting that Gpx-1 (or compounds which mimic its actions, such as ebselen) may exhibit protective effects.
Of interest, Gpx-1 activity is elevated in smokers, as a potential mechanism to counteract the harmful oxidative stress. However, Gpx-1 is severely depleted in the lungs of COPD patients, resulting in an overstated inflammatory response and oxidative burden Santos et al., 2004;Tkacova et al., 2007;Vlahos et al., 2010). A study by Chew et al. has shown therapeutic effects of ebselen in Gpx-1 knockout mice, with the study finding that synthetic repletion of Gpx activity in these diabetic mice produced atheroprotective effects in vivo (Chew et al., 2010). Exogenous repletion of Gpx-1 with compounds like ebselen may have therapeutic potential in not only treating the pulmonary manifestations of COPD but perhaps its cardiovascular co-morbidities.
In the present study, we investigated whether chronic CS exposure in a preclinical mouse model of COPD impaired vascular function and whether ebselen could prevent CS-induced vascular dysfunction in mice.

| Animals
All animal care and experimental procedures were conducted in accor-

| Cigarette smoke exposure and ebselen treatment
Mice were placed into an 18-L Perspex chamber (The Plastic Man, Huntingdale, Victoria, Australia) in a standard fume cabinet (Aircare Extraction Systems LTD, Clayton, Victoria, Australia) and exposed to cigarette smoke generated from 9 Winfield Red Cigarettes (total particulate matter of 419 mgÁm −3 , 16 mg or less of tar, 1.2 mg or less of nicotine and 15 mg or less of CO, Philip Morris, Moorabbin, Australia) for 5 days a week for 8 weeks. Mice were exposed to CS (n = 10 per treatment group) delivered three times per day with three cigarettes delivered at 9 a.m., 12 noon, and 3 p.m., over a 1 h time-period. CS was generated in 60 ml tidal volumes over 10 s, via a timed draw-back to mimic normal smoking inhalation and burn rates. We have previously shown that this CS exposure protocol in Balb/C mice replicates key clinical traits of early stage COPD in humans, including lung inflammation and pathology (mucus hypersecretion and impaired lung function), increased lung and systemic oxidative stress and co-morbidities including skeletal muscle wasting (Austin et al., 2016;Chan et al., 2019;Vlahos & Bozinovski, 2014). Thus, this model provides a robust and clinically relevant platform to test therapies for COPD and its co-morbidities. Sham-exposed mice are placed into an identical 18-L Perspex chamber but do not receive cigarette smoke. Mice were weighed every second day (prior to initial CS exposure) up to and including the end of the experiment. For the ebselen treatment studies, mice were administered 10 mgÁkg −1 of ebselen (Sapphire Bioscience, Australia) prepared in 5% w/v CM-cellulose in water (Sigma-Aldrich, USA) or vehicle treated with 5% CM-cellulose in water alone. Treatments were administered via oral gavage once daily, 1 h prior to the initial CS exposure.

| Bronchoalveolar lavage and lung collection
Animals were killed at the end of the experimental protocol via intraperitoneal injection of sodium pentobarbitone (240 mgÁkg −1 ; Virbac, NSW, Australia). Lungs were then lavaged in situ via a surgical tracheotomy with 0.4 ml of chilled PBS initially followed by 0.3 ml PBS three times, with $1 ml of BAL fluid (BALF) retrieved per mouse as previously described (Vlahos et al., 2006). Total viable cell numbers in the BALF were determined using 50 μl of BALF diluted with 50 μl of acridine orange/ethidium bromide (AO/EB) (Invitrogen, USA). Cell counting was carried out using a standard Neubauer haemocytometer, under fluorescent light on an Olympus BX53 microscope (Olympus, Japan). Right ventricular perfusion with 6-7 ml of PBS was then performed to clear whole lungs from blood, and the lungs then excised, rinsed in PBS, snap-frozen in liquid nitrogen and stored at −80 C until required.

| Differential cell counting
To differentiate the various cell populations in the BALF, cytocentrifuged cell preparations (Shandon Cytospin 3, 18.06 g, 10 min) were used, with $5 × 10 4 cells per slide. Dried cytospots were subjected to a Shandon Kwik-Diff® Fixative (Thermo Fischer Scientific, USA), followed by Hemacolour® eosin and thiazine differential stains (Merck, USA) as outlined in the manufacturer's instructions. Cell types (i.e. macrophages, neutrophils, and lymphocytes) were identified according to standard morphological criteria, using the above-mentioned microscope, with at least 500 cells counted per slide.

| Quantitative real-time PCR (RT-qPCR)
Total RNA was extracted from approximately 10 mg of whole lung tissue using a RNeasy® Mini Kit (Qiagen, Germany).

| Vascular reactivity
To assess the effect of both CS-exposure and ebselen treatment on vessel function, the thoracic aorta was excised from each mouse and all perivascular fat removed. Vessels were placed into carbogen-

| Data and statistical analysis
The data and statistical analysis comply with the recommendations of the

| Materials
The suppliers of the following compounds are as follows: Winfield

| Chronic CS exposure causes endothelial dysfunction in mouse thoracic aorta
The effect of chronic CS exposure on vascular function was examined to establish if the enhanced oxidative stress and inflammation arising from smoking has detrimental effects on blood vessel function in vivo.

| CS exposure drives pulmonary immune cell infiltration, pro-inflammatory and oxidative stress gene expression
CS exposure caused a significant increase in the total number of immune cells infiltrating the lungs when compared to sham-exposed control mice (Figure 2a). This increase in total cell number was attributed to a significant increase in macrophages, neutrophils, and lymphocytes (Figure 2b-d). To better understand the mechanism(s) underlying the increased BALF inflammation, the pulmonary expression of both pro-inflammatory and oxidative stress genes was examined. CS exposure caused a significant increase in mRNA expression of the pro-inflammatory cytokines TNF-α (2.2-fold) and IL-6 (4.7-fold), as well as the key oxidative stress enzyme NOX-2 (1.5-fold; Figure 2f,g).

| Endothelial NOS expression is downregulated due to increased vascular ROS in COPD
Given that chronic CS exposure causes endothelium-dependent vascular dysfunction, we then went on to investigate the potential underlying mechanism driving this impaired vascular function. Endothelial expression of the key vascular tone regulator eNOS and a marker of oxidative stress, 3-nitrotyrosine (3-NT), were quantified. CS exposure significantly reduced expression of eNOS by $70% (Figure 3a). In addition, the level of 3-NT expression was significantly up-regulated by $3.2-fold following CS exposure (Figure 3b), indicative of enhanced vascular oxidative stress which may be responsible for the reduced expression of eNOS.

| Ebselen prevents CS-induced endothelial dysfunction
As chronic CS exposure caused significant lung inflammation as well as heightened lung and vascular oxidative stress, we sought to investigate the effect of antioxidant treatment on vascular endothelial function. Sham + Veh-treated mice showed an $90% R max , while CS + Veh-treated mice showed an $40% R max to ACh (Figure 4a), confirming CS-induced endothelial dysfunction as also demonstrated in Figure 1. CS-exposed mice treated with ebselen were completely protected from the CS-induced endothelial dysfunction, as shown by the maximal relaxation of $90% to ACh. In addition, ebselen did not affect vascular endothelial function in sham-exposed mice ( Figure 4a), suggesting that its protective effects are specific to CS exposure. As shown in Figure 1, smooth muscle relaxant responses to SNP were unaltered, regardless of CS exposure or ebselen treatment (Figure 4b).

| Ebselen reduces pulmonary immune cell infiltration but has no effect on pro-inflammatory and oxidative stress gene expression
Given that ebselen was able to protect against CS-induced vascular dysfunction, we next sought to determine the effect of ebselen on CS-induced lung inflammation. Consistent with Figure 2, CS caused a significant increase in BALF total cells, macrophages, neutrophils, F I G U R E 1 CS exposure causes endothelial dysfunction in the thoracic aorta. Cumulative concentration response curves to (a) ACh and (b) sodium nitroprusside (1 × 10 −8 M to 1 × 10 −5 M) to assess both endothelial-dependent and smooth muscle-dependent vasodilatory responses in mouse thoracic aorta (n = 8 per group) following either chronic CS or sham exposure, respectively. Results are expressed as mean percentage relaxation relative to precontraction ± SEM. *P < .05, significantly different as indicated; two-way ANOVA with Tukey's multiple comparisons F I G U R E 2 Chronic CS exposure increases BAL fluid cellularity and enhances both lung pro-inflammatory and oxidative stress mediator gene expression. The lungs of mice exposed to CS were lavaged for the assessment of total cells (a), macrophages (b), neutrophils (c), and lymphocytes (d) (n = 10). Whole lungs excised from mice were then used to measure mRNA expression by RT-qPCR of TNFα (e) (n = 10), IL-6 (f) (n = 10), and NOX-2 (g) (n = 9). Gene expression data are expressed as fold change relative to the sham group. All data are expressed as mean + SEM. *P < .05, significantly different as indicated; Student's unpaired t-test and lymphocytes (Figure 5a-d). Interestingly, ebselen significantly reduced CS-induced increases in BALF total cell and neutrophil counts but not macrophage and lymphocyte numbers (Figure 5a-d).
Moreover, while CS exposure increased TNF-α and NOX-2 mRNA expression in the lungs, no detectable attenuations were found after ebselen administration (Figure 5e,f). Ebselen administration also appears to be ineffective in preserving lung Gpx-1 mRNA expression against CS exposure (Figure 5g) but rather mimics its antioxidant activity.

| Ebselen diminishes CS-induced vascular oxidative stress and down-regulation of eNOS
We next sought to define the underlying mechanism by which ebselen prevented CS-induced endothelial dysfunction. Mice chronically exposed to CS had increased vascular endothelial oxidative stress as measured by 3-NT expression ( Figure 6). However, ebselen administration significantly reduced CS-induced increases in 3-NT expression to baseline sham levels ( Figure 6). Moreover, ebselen treatment F I G U R E 3 Immunofluorescent staining of endothelial NOS and 3-nitrotyrosine in the thoracic aorta of mice exposed to 8 weeks of CS or room air. Immunofluorescent quantification of eNOS and 3-NT expression in either sham or 8-week CS-exposed mice. Green staining indicates the presence of eNOS (a) or 3-NT (b) and blue staining denotes the nuclear counterstain, DAPI. Representative photographs of immunofluorescent staining in (a) sham-exposed or (b) CS-exposed mice. After normalisation to the relevant negative control, the expression of eNOS (n = 5-6 mice per group) and 3-NT (n = 5 mice per group) are presented as either percentage (%) change of or fold change from the shamtreated group. Scale bar represents 50 μM. Data are expressed as mean + SEM. *P < .05, significantly different as indicated; Student's unpaired t-test completely prevented the down-regulation of eNOS as a result of CSexposure ( Figure 7). Collectively, these findings suggest that ebselen reduced CS-induced vascular oxidative stress and subsequently prevented the down-regulation of eNOS within the thoracic aorta.

| DISCUSSION
In the present study, the mechanisms underlying systemic endothelial dysfunction, the key driver of co-morbid CVD-associated mortality are revealed and the effects of CS exposure on blood vessel function in our preclinical murine model of COPD were assessed. We found that 8 weeks of CS exposure caused significant pulmonary inflammation and oxidative stress in mice. Pulmonary immune cell infiltration is believed to be the underlying factor driving increased NOX-2 expression and ROS formation. Studies from our group have shown that mice treated with influenza A virus had increased levels of ROS production in the lungs as a consequence of increased pulmonary inflammation (macrophages and neutrophils) and NOX-2 expression (To et al., 2017;Vlahos et al., 2011;Vlahos & Selemidis, 2014). In this study, we also showed increased NOX-2 expression in the lung which is presumably a result of increased pulmonary macrophage and neutrophil numbers in response to CS exposure. We also found a significant increase in the expression of the pro-inflammatory mediators TNF-α and IL-6 in response to CS, which is consistent with our previously published work (Hansen et al., 2013;Vlahos et al., 2006). TNF-α is largely secreted by stimulated macrophages (e.g. in response to CS), driving the inflammatory response and intracellular ROS production, while down-regulating antioxidant activity (Mukhopadhyay et al., 2006). IL-6 has also been implicated in the pathophysiology of pulmonary diseases. Thus, these pro-inflammatory mediators may contribute to pulmonary inflammation and reduced lung function observed in COPD patients (Rincon & Irvin, 2012).
It was clear from the present study that CS exposure significantly impaired vasodilation of mouse thoracic aorta to ACh and that this was specifically attributed to endothelial dysfunction without affect- Having shown that CS causes endothelium-dependent vascular dysfunction, we next investigated whether this was attributed to changes in eNOS expression. We found that CS caused an $60% Cumulative concentration response curves to ACh (a) and sodium nitroprusside (b) (1 × 10 −8 M to 1 × 10 −5 M) to assess endothelial and smooth muscle-dependent vasodilation in mouse thoracic aorta (n = 6) following either chronic CS or sham exposure in either ebselen or vehicle-treated mice, respectively. Results are expressed as mean percentage relaxation relative to pre-constriction ± SEM. *P < .05, significantly different as indicated; two-way ANOVA with Tukey's multiple comparisons F I G U R E 5 Effect of chronic CS and ebselen treatment on BAL fluid cellularity and pulmonary pro-inflammatory and oxidative stress gene expression. The lungs of mice exposed to CS were lavaged for the assessment of total cells (a), macrophages (b), neutrophils (c), and lymphocytes (d) (n = 10). Whole lungs excised from mice were then used to measure mRNA expression by RT-qPCR of TNF-α (e), NOX-2 (f), and GPX-1 (g) (n = 10). Responses are expressed as fold change relative to the sham + vehicle-treated group, post normalisation to GAPDH (housekeeping gene). All data are expressed as mean + SEM. *P < .05, significantly different as indicated; two-way ANOVA and Tukey's post hoc analysis vascular complications through its free radical scavenging activity, could prevent CS-induced vascular dysfunction. Moreover, de Haan and Cooper (2011) have proposed that deficiencies in the antioxidant enzyme GPX and an enhanced oxidative burden promotes endothelial dysfunction leading to DM-related microvascular and macrovascular complications. As such, targeted antioxidant replenishment therapy using GPX-mimetics (i.e. ebselen) may be effective in reducing the cardiovascular manifestations in disease states, such as DM.
In the present study, we showed that ebselen completely Consistent with our previous studies, ebselen significantly reduced CS-induced BALF inflammation which was largely attributed to a reduction in neutrophilic infiltration (Duong et al., 2010;Oostwoud et al., 2016). Excess neutrophils play a detrimental role in COPD particularly during periods of acute exacerbation, as they can directly induce protease-mediated tissue damage, that has been directly correlated to worsening of emphysema in these patients (Oostwoud et al., 2016;Pesci et al., 1998). MMP activation drives a loss of lung integrity and an increase in permeability which may facilitate the spill over of proinflammatory mediators into the systemic circulation.
It was interesting to note that the whole lung gene expression of the pro-inflammatory mediator TNFα and the oxidative stress enzyme NOX-2, induced by CS exposure, were not reduced by ebselen pretreatment. Although not investigated in the present study, it would be worth exploring whether TNF-α protein expression is altered F I G U R E 6 Immunofluorescent staining of 3-nitrotyrosine in the thoracic aorta of mice exposed to chronic CS and treated with ebselen. Immunofluorescent quantification of 3-NT expression in either sham or 8-week CS-exposed mice that were treated with either vehicle or ebselen. Green staining indicates the presence of 3-NT specific and blue staining denotes the nuclear counterstain, DAPI. Representative photographs of immunofluorescent staining in (a) sham-exposed vehicle treated (n = 5), (b) sham-exposed ebselen treated (n = 5), (c) CSexposed vehicle treated (n = 5), and (d) CSexposed ebselen treated mice (n = 6). After normalisation to the relevant negative control, the expression of 3-NT is presented as fold change from the sham vehicle-treated group. Scale bar represents 50 μM. Data are expressed as mean + SEM. *P < .05, significantly different as indicated; two-way ANOVA with Tukey's multiple comparisons following ebselen administration. Similarly, it would be worth investigating whether ebselen can directly impede the activity of the regulatory p47phox and other subunits of the NOX-2 enzyme, ultimately reducing superoxide production, as this has been previously shown (Smith et al., 2012). The ROS scavenging properties of ebselen within the lung have also been established in the context of asthma, with Zhang et al. showing that following ovalbumin challenge, guinea pigs showed significantly enhanced pulmonary superoxide and hydrogen peroxide concentration, which was decreased in ebselen treated animals (Zhang et al., 2002), reinforcing the powerful antioxidant properties of ebselen in CS-induced lung inflammation and oxidative stress.
Expression of the Gpx-1 gene is up-regulated in the lungs of smokers (Barnes & Celli, 2009), which may be a compensatory antioxidant mechanism in response to noxious effects of CS. Conversely, smokers and patients with established COPD have reduced Gpx activity (James & Wenzel, 2007;Versari et al., 2009;Vlahos et al., 2006), contributing to an over-exuberant oxidative burden in the lungs of these patients (Geraghty et al., 2013). In the present study, we found that whole lung Gpx-1 mRNA expression was significantly down-regulated in mice exposed to CS, irrespective of ebselen treatment, further reinforcing the suggestion that loss of GPX protein would increase lung oxidative stress and inflammation. Blunted Gpx expression has also been implicated as a contributing factor in driving endothelial dysfunction, inducing apoptosis and promoting atherosclerosis systemically (Geraghty et al., 2013).
Ebselen treatment has shown promising effects on the vasculature in this study by completely preventing endothelial dysfunction in CS-exposed mice as well as reducing BALF cellularity attributed to neutrophilic infiltration. It has been established that eNOS can undergo oxidative modification as a direct result of the heightened oxidative burden in smokers (Arunachalam et al., 2010;Edirisinghe & Rahman, 2010;Li & Forstermann, 2014;Zhang et al., 2006). Findings from this study showed that eNOS expression as quantified through immunofluorescent staining was significantly down-regulated as a result of CS exposure. However, pretreatment with ebselen prevented CS-induced down-regulation of eNOS and was the most likely mechanism by which ebselen restored vascular function in CS-exposed mice.
It was also interesting to note that ebselen significantly reduced F I G U R E 7 Immunofluorescent staining of endothelial NOS in the thoracic aorta of mice exposed chronically to CS and treated with ebselen Immunofluorescent staining of eNOS in either sham or 8-week CS-exposed mice with or without ebselen administration. Green staining detects the presence of eNOS and blue staining denotes the nuclear counterstain, DAPI. Representative photographs of immunofluorescent staining in (a) sham-exposed vehicle treated (n = 5), (b) sham-exposed ebselen treated (n = 5), (c) CSexposed vehicle treated (n = 7), and (d) CSexposed ebselen treated mice (n = 7). After normalisation to the relevant negative control, the expression of eNOS is presented as percentage (%) change of the sham vehicletreated group. Scale bar represents 50 μM. Data are expressed as mean ± SEM. *P < .05, significantly different as indicated; two-way ANOVA with Tukey's multiple comparisons CS-induced 3-NT staining in the thoracic aorta indicating that ebselen completely prevented enhanced oxidative stress within the vascular endothelium, leading to sustained eNOS levels and normal vascular function in CS-exposed and ebselen treated mice.
While it is clear from this study that ebselen can prevent CS-induced lung inflammation and vascular dysfunction, it would be worth investigating whether ebselen can stop the progression of disease or could reverse established vascular dysfunction when administered therapeutically, that is, during established disease. We would also like to investigate the oxidative mechanisms that deplete vascular eNOS expression, as studies have shown that eNOS can become uncoupled under highly oxidative conditions (Heiss & Dirsch, 2014;Karbach et al., 2014). Direct measurement of markers of systemic inflammation including CRP or IL-6, as well as blood leukocyte numbers would also be beneficial in the understanding of the pathology observed within the vasculature as a study by Zeng et al. has shown that serum levels of IL-6 are significantly enhanced in patients with COPD, when compared to control subjects, although levels of CRP and TNFα remained unchanged (Zeng et al., 2019). This study has implications for people with COPD as it sheds light on the mechanisms that could explain why co-morbid CVD is the largest killer of COPD patients and may lead to the development of novel, life-saving therapeutics.
In conclusion, we found that chronic CS exposure in mice caused endothelial dysfunction, as a direct result of enhanced vascular oxidative stress leading to a down-regulation of eNOS. In addition, ebselen administration significantly reduced CS-induced lung inflammation and vascular oxidative stress leading to restored vascular endothelial function in CS-exposed mice. Collectively, the data from the present study suggest that ebselen may be a novel therapeutic approach to the treatment of both the pulmonary manifestations and cardiovascular co-morbidities associated with CSinduced COPD.