Several clinical studies have shown that smoking in asthmatics and chronic obstructive pulmonary disease patients is closely associated with corticosteroid refractoriness. In this work, we have analyzed glucocorticoid insensitivity in human pulmonary artery endothelial cells (HPAECs) under cigarette smoke extract (CSE) exposure as well as the possible additive effects of the combination therapy with a phosphodiesterase (PDE)-4 inhibitor.
Interleukin (IL)-8 was measured in cell supernatants by ELISA. Histone deacetylase (HDAC), histone acetylase (HAT), and intracellular cAMP levels were measured by colorimetric assays and enzyme immunoassay, respectively. PDE4 isotypes and glucocorticoid receptor (GR)-α and β expression were measured by real-time RT-PCR.
The PDE4 inhibitor rolipram dose dependently inhibited the IL-8 secretion induced by CSE 5%. In contrast, dexamethasone 1 μM did not show inhibitory effect on IL-8 secretion. Combination of subeffective rolipram concentrations at 10 nM increased the inhibitory effect of dexamethasone to ~45% of inhibition. Cigarette smoke extract 5% inhibited HDAC activity and increased HAT activity generating glucocorticoid insensitivity. Rolipram did not modify the HDAC activity, however partially inhibited the increase in HAT activity at 1 μM. PDE4 isotypes were up-regulated by CSE 5% with the consequent cAMP down-regulation. Dexamethasone reduced all PDE4 isotypes expression and showed additive effects with rolipram enhancing cAMP levels. Furthermore, rolipram enhanced GR-α expression and inhibited the increase in GR-β induced by CSE.
Combination of rolipram and dexamethasone shows additive properties in HPAECs under glucocorticoid insensitive conditions. These results may be of potential value in future anti-inflammatory therapies using combination of PDE4 inhibitors and glucocorticoids.
Both asthma and chronic obstructive pulmonary disease (COPD) are chronic inflammatory disorders of the airways whose symptoms arise from airflow obstruction. Exposure to environmental cigarette smoke (second-hand smoke) has been linked to asthma in early childhood exposure, and active smoking has been associated with a rapid decline in lung function in asthma and COPD .
Although asthma and COPD represent two distinct diseases, these two conditions can occasionally coexist. This may be particularly seen in patients with asthma who have been exposed to cigarette smoke, who can develop fixed airflow limitation and a mixture of ‘asthma-like’ and ‘COPD-like’ inflammation .
It is therefore not surprising that treatments for both these diseases may often overlap. In addition to its effect on the course of the two diseases, cigarette smoke greatly reduces the efficacy of some of the important treatment regimens currently available for both conditions particularly inhaled corticosteroids [2, 3].
Recent evidence suggests that asthmatic and COPD patients characterized by elevated levels of tumor necrosis factor alpha (TNF-α), interleukin-8 (IL-8), and neutrophilic inflammation in the airways show glucocorticoid refractoriness [4-6]. Both TNF-α and IL-8 are pleiotropic pro-inflammatory molecules that exert their biological effects mainly by acting on the vascular endothelium , where they enhance inflammatory responses by up-regulating leukocyte recruitment into inflamed lung. Endothelial cells are particularly resistant to the anti-inflammatory effects of glucocorticoids  and endothelial dysfunction, and apoptosis may be enhanced by glucocorticoids under certain conditions [9, 10]. Furthermore, it has been well established that cigarette smoke induces endothelial dysfunction [11, 12], increasing vasoactive mediators and chemokines such as IL-8, thus increasing airway leukocyte cell recruitment into the lungs of asthmatic and COPD patients as well as increasing intimal hyperplasia and pulmonary hypertension in a relatively high proportion of patients with COPD . Because pulmonary artery endothelial cells play a key role in the pathophysiology of asthma and COPD and because glucocorticoids play a central role in the management of smoking asthmatics and COPD patients, it would be of potential value to discover new strategies to enhance their impaired anti-inflammatory effects on pulmonary endothelial cells.
One therapeutically attractive class of drug that may act in increasing the anti-inflammatory effects of glucocorticoids is the inhibitors of the enzyme phosphodiesterase (PDE)-4. These compounds inhibit the adenosine 3′,5′-cyclic monophosphate (cAMP) degradation enhancing anti-inflammatory effects of cAMP. Currently, one PDE4 inhibitor, roflumilast, is available for COPD, and others are in development for asthma . PDE4 inhibitors have shown potent anti-inflammatory effects on human neutrophils , and their combination with glucocorticoids appears to be additive in mononuclear cells . Furthermore, PDE4 inhibitors inhibit leukocyte–endothelial cell interactions, expression of adhesion molecules, and microvascular permeability following TNF-α or lipopolysaccharide (LPS) stimulation . Thus, one possible way to enhance the impaired anti-inflammatory effect of glucocorticoids could be to administer them in combination with a PDE4 inhibitor.
Currently, no data are available regarding the role of cigarette smoke on glucocorticoid insensitivity in pulmonary endothelial cells or the possible additive or synergistic effects of the PDE4 inhibitors. Therefore, we focused on endothelial cells in this study and hypothesized that cigarette smoke–mediated glucocorticoid insensitivity may be attenuated by PDE4 inhibitors. In addition, molecular mechanisms were also studied.
Materials and methods
Unless indicated otherwise, all reagents used were obtained from Sigma (Chemical Co, Madrid, Spain). Rolipram was dissolved in dimethyl sulfoxide (DMSO) at 10 mM stock concentration. Theophylline was dissolved in HCl 0.1 M at 100 mM stock concentration. Several dilutions of the stocks were performed with cell culture medium. The final concentrations of DMSO (0.1%) or HCl (10 μM) in the cell culture did not affect cellular functions. Other chemicals such as dbcAMP, N-acetyl-l-cysteine (NAC), or dexamethasone were dissolved in medium.
Isolation and culture of human pulmonary artery endothelial cells
Human lung tissue was obtained from patients who were undergoing transplant program and whose lungs were not available for transplant. All lung tissues (n = 10) were from nonsmokers, nonasthmatic, and non-COPD patients. The protocol for obtaining human tissue was approved by the local ethical review board for human studies (General Hospital of Valencia, Spain); the legal representative gave informed consent. Isolation and culture of human pulmonary artery endothelial cells (HPAECs) were performed using Dynabeads CD31 endothelial cell kit (Dynal Biotech, Hamburg, Germany) as previously outlined . Further information is provided in Online Repository (OR) in Data S1.
Preparation of cigarette smoke extract and incubations
Cigarette smoke extract (CSE) was obtained from research cigarette (2R4F, from Tobacco Health Research, University of Kentucky) as previously outlined . Further information is provided in OR.
IL-8 and cAMP measurements
The IL-8 supernatant content and intracellular cAMP levels were measured using commercially available enzyme-linked immunosorbent assay kits for IL-8 (R&D Systems, UK) and cAMP biotrack enzyme immunoassay (EIA) system (Amersham, Nottingham, UK) according to the manufacturer's protocol. Further information is provided in OR.
Measurement of histone deacetylase and histone acetylase activity
Histone deacetylase and histone acetylase (HAT) activities were measured on nuclear cell protein using colorimetric assay systems BML-AK501 and ALX-850–326–KI01, respectively (Enzo Life Sciences, Lause, Switzerland) according to the manufacturer's protocol. Further information is provided in OR.
PDE4A, PDE4B, PDE4C, and PDE4D subtypes and glucocorticoid receptor alpha (GRα) and glucocorticoid receptor beta (GRβ) were quantified using SYBR Green real-time PCR as previously outlined . Relative quantification of transcript levels (compared with control groups) was determined by evaluating the expression with the 2−ΔΔCT method using β-actin expression as endogenous control . Further information is provided in OR.
2′,7′-Dichlorofluorescein diacetate fluorescence measurement of reactive oxygen species
Intracellular reactive oxygen species (ROS) were determined using the fluorogenic substrate 2′,7′-dichlorofluorescein diacetate (DCFDA; Molecular proves, Nottingham, UK) as previously outlined . Further information is provided in OR.
To detect the early stages of injury, translocation of phosphatidylserine in the cell membrane was examined by the annexin V assay according to the manufacturer's instructions (Roche, Applied Science, Nottingham, UK). Human pulmonary artery endothelial cells necrosis was determined by propidium iodide (PI) cell incorporation as previously outlined . Further information is provided in OR.
Cell proliferation assay
Human pulmonary artery endothelial cells proliferation was measured by colorimetric immunoassay based on BrdU incorporation during DNA synthesis using a cell proliferation enzyme-linked immunosorbent assay BrdU kit (Roche, Mannheim, Germany) according to the manufacturer's protocol. Further information is provided in OR.
Data are presented as mean ± SEM of n experiments. Statistical analysis of data was carried out by analysis of variance (anova) followed by Bonferroni test (GraphPad Software Inc, San Diego, CA, USA). Significance was accepted when P <0.05.
Effects of dexamethasone and rolipram on CSE-induced IL-8 release in HPAECs
Cigarette smoke extract 5% significantly increased IL-8 secretion in HPAECs after 18 h of exposure (Fig. 1A,B). Thus, we selected these conditions for further experiments. The PDE4 inhibitor rolipram (10 nM–1 μM) attenuated the CSE-induced IL-8 release to almost 50% of control at 1 μM, effect that was mimicked by the cAMP analog dibutyryl adenosine 3′-5′ cyclic monophosphate (dbcAMP, 1 mM) (Fig. 1C). In a similar way, the antioxidant NAC (1 mM) suppressed the IL-8 release (Fig. 1D). Pre-incubation of HPAECs with dexamethasone 1 μM did not show significant reduction in IL-8 release. However, the combination of dexamethasone with noneffective rolipram concentration at 10 nM reached ~45% of IL-8 inhibition (Fig. 1D).
CSE down-regulates HDAC and increases HAT activities: role of dexamethasone and rolipram
In HPAECs, CSE 5% significantly reduced HDAC activity and increased HAT activity after 4 h and 18 h of exposure (Fig. 2A,C). Addition of rolipram (10 nM–1μM) or dbcAMP (1 mM) as well as dexamethasone 1 μM did not modify the CSE-induced HDAC inhibition. In contrast, rolipram 1 μM and dbcAMP inhibited by ~35% the increase in HAT. As positive control, theophylline 1 μM suppressed the effects of CSE on HDAC and HAT activity in a similar way that the antioxidant NAC (Fig 2B,D).
Effects of dexamethasone and rolipram on CSE-induced cAMP inhibition, PDE4 expression, and ROS production
Cigarette smoke extract 5% decreased intracellular cAMP content in HPAECs (Fig. 3A) and significantly increased PDE4A, PDE4B, and PDE4D expression after 18 h of exposure (Fig. 3B). Unlike dexamethasone, rolipram (10 nM–1 μM) restored cAMP levels in a similar way that dbcAMP and NAC treatments (Fig. 3A). In this line, the combination of rolipram 10 nM with dexamethasone 1 μM showed an additive effect of increasing cAMP to control levels (Fig. 3A). In other experiments, dexamethasone attenuated the CSE-induced PDE4A, PDE4B, and PDE4D overexpression (Fig. 3B). The addition of rolipram 10 nM to dexamethasone 1 μM suppressed all PDE4 isotypes under control levels, therefore showing additive effects (Fig. 3B). The antioxidant NAC and dbcAMP inhibited the CSE-induced PDE4 overexpression, suggesting a role of ROS on PDE4 overexpression. In HPAECs, CSE elevated intracellular ROS, reaching a peak value after 10 min, which was dose dependently inhibited by rolipram (10 nM–1 μM) as well as by the positive control NAC (Fig. 3A,D).
PDE4 inhibition increases GRα expression and inhibits the CSE-induced GRα down-regulation and GRβ up-regulation
Rolipram (10 nM–1 μM) dose dependently increased GRα expression for up to 2.5-fold over control after 18 h of exposure (Fig. 4A). This effect was also reproduced by dbcAMP. In contrast, dexamethasone down-regulated GRα expression to 0.5-fold under control levels (Fig. 4A). None of these drugs had any effect on GRβ expression (Fig. 4B). In other experiments, the exposure of HPAECs to CSE decreased GRα expression to 0.5-fold of control, which was prevented by rolipram and dbcAMP but not by dexamethasone (Fig. 4B). The addition of rolipram 100 nM to dexamethasone effectively increased GRα expression up to control levels. The mRNA expression of GRβ was increased following CSE exposure near to three-fold of control (Fig. 4B). Rolipram dose dependently inhibited CSE-induced GRβ expression similar to dbcAMP and dexamethasone. However, the combination of rolipram and dexamethasone showed an additive effect on GRβ down-regulation (Fig. 4B).
Rolipram, at low doses, enhances the anti-inflammatory effects of dexamethasone in HPAECs stimulated with TNF-α under oxidative stress conditions
Both smoking asthmatics and COPD patients show elevated oxidative stress as well as high levels of TNF-α in lung tissues. To represent this chronic inflammatory situation, HPAECs were pre-incubated in presence or absence of CSE for 18 h, treated with rolipram or dexamethasone for 1 h, and then stimulated with TNF-α for further 24 h until IL-8 levels were measured. In absence of CSE, dexamethasone 1 μM attenuated the TNF-α-induced IL-8 secretion by 43% of inhibition over maximal response (Fig. 5A). Human pulmonary artery endothelial cells pretreated with CSE were nearly insensitive to the inhibitory effect of dexamethasone on TNF-α-induced IL-8 secretion (Fig. 5A). In the CSE pretreatment protocol, the combination of dexamethasone (10 nM–1 μM) with noneffective rolipram concentration of 10 nM showed a dose-dependent inhibition of TNF-α-induced IL-8 reaching a 50% of inhibition at 1 μM (Fig. 5B). Similar results were observed using theophylline 1 μM in combination with dexamethasone (10 nM–1 μM) as positive control (Fig. 5C).
The lack of effect of dexamethasone on CSE-induced increase in apoptosis and decrease in proliferation is improved in combination with rolipram
Cigarette smoke extract exposure dose dependently enhanced early apoptosis, represented by annexin V–positive cells, reaching 40% of positive cells at CSE 5% after 24 h of exposure without affecting cell necrosis (Fig. 6A). In a similar way, CSE inhibited HPAEC proliferation (Fig. 6B). Rolipram dose dependently inhibited the cell apoptosis enhanced by CSE and increased cell proliferation close to control levels. Dexamethasone 1 μM did not affect the impaired apoptosis or proliferation induced by CSE. However, the combination of dexamethasone with rolipram, at noneffective doses of 10 nM, inhibited CSE-induced cell apoptosis by ~50% and increased cell proliferation close to control levels (Fig. 6C,D). The addition of the antioxidant NAC effectively suppressed the CSE-induced cell apoptosis and the impaired cell proliferation, suggesting a role for ROS on the effects of CSE.
Inhaled corticosteroids play a key role in the treatment of airway diseases such as asthma and COPD. However, several clinical studies have shown that smoking in asthmatics and COPD patients is closely associated with corticosteroid refractoriness [3, 20-22]. Therefore, there is a clinical need for additional therapies improving current inhaled corticosteroid. In this work, we provide new evidence that identifies the role and the mechanism of action of PDE4 inhibitors in preventing glucocorticoid insensitivity generated by cigarette smoke in HPAECs. Specifically, the present results demonstrate that: (i) CSE-mediated HPAEC inflammatory responses are insensitive to dexamethasone; (ii) Low doses of the PDE4 inhibitor rolipram show additive effects and reverse dexamethasone insensitivity; (iii) rational for additive effects between rolipram and dexamethasone is attributed to the increase in GRα by rolipram and by the down-regulation of PDE4 subtypes by dexamethasone as well as by the inhibition of CSE-induced GRβ up-regulation.
It is known that chronic smoking in asthma and COPD may promote shared cellular and molecular characteristics, including the presence of high levels of lung and systemic TNF-α, IL-8, and neutrophils as well as a poor response to corticosteroids [6, 23]. However, there is a lack of information about the effect of glucocorticoids on pulmonary endothelial cells in these conditions. In this regard, previous data showed that dexamethasone, at 1 μM, did not inhibit secretion and expression of IL-8, IL-6, GRO-α, and ICAM-1 in lung microvascular endothelial cells and HUVEC following TNF-α and IL1-β exposure [8, 24]. Furthermore, dexamethasone enhanced TNF-α-induced leukocyte adhesion to pulmonary microvascular endothelial cells . In the present work, we showed new evidence of IL-8 glucocorticoid insensitivity in HPAECs secondary to CSE exposure. Our results were also corroborated when HPAEC pre-exposed to CSE was stimulated with the pro-inflammatory TNF-α, thus showing a poor response to dexamethasone. Similar results have been described previously in alveolar macrophages in asthma and COPD [25, 26]. Perhaps, the most important mechanism that may explain the relative corticosteroid resistance in smokers with asthma and COPD is the reduction in the enzyme HDAC and the increase in HAT activity [26, 27]. For corticosteroids to exert their maximal effects in terms of pro-inflammatory cytokine suppression, HDAC activity is required. In this regard, theophylline has been proposed as an enhancer of glucocorticoids, because its effect increasing HDAC2 activity . Smoking reduces the activity of HDAC in inflammatory cells such as alveolar macrophages, which may explain the increased expression of inflammatory mediators seen in bronchoalveolar lavage samples of smokers . The decrease in HDAC activity with smoking is possibly a result of oxidative stress, which impairs its function as shown in vitro in primary airway epithelial cells from healthy volunteers . In this work, CSE significantly decreased HDAC and increased HAT in HPAECs. Thus, the lack of dexamethasone in preventing CSE-induced IL-8 release could be mediated by this molecular mechanism. In the effort to study new drugs to improve anti-inflammatory properties of glucocorticoids, we found that noneffective doses of the PDE4 inhibitor rolipram improved the inhibitory effect of dexamethasone on CSE-induced IL-8 release as well as in the CSE pre-exposure protocol followed by TNF-α-stimulation. However, this additive effect was not explained by the increase in HDAC activity, because rolipram did not improve the CSE-induced HDAC down-regulation. In this regard, previous reports have shown that PDE4 inhibitors are unable to up-regulate the cigarette smoke–impaired HDAC, confirming our results . However, rolipram, at higher doses, as well as dbcAMP was able to attenuate the increase in HAT induced by CSE, which may partially explain additive effects of rolipram and dexamethasone. However, this effect was only observed at relative high doses of rolipram, thus limiting its relevance. Supporting the hypothesis of oxidative stress as a molecular signal to decrease HDAC and increase HAT, we found that the antioxidant NAC inhibited HDAC down-regulation and HAT up-regulation induced by CSE. This interesting antioxidant hypothesis could be of relevance because rolipram showed antioxidant properties on HPAECs stimulated with CSE as previously we showed in bronchial epithelial cells .
However, a question that remains unanswered is how low doses of rolipram may enhance the anti-inflammatory properties of dexamethasone in our model. We and others showed that cigarette smoke and ROS may increase PDE4 isotypes that down-regulate cAMP levels resulting in inflammatory activation [31, 32]. As we previously observed in differentiated bronchial epithelial cells , the PDE4B isotype was the most induced in endothelial cells by cigarette smoke, followed by PDE4A, PDE4D, and by a slightly increase in PDE4C. Interestingly, PDE4B and PDE4D are the most expressed in endothelial cells, and PDE4A, PDE4B, and PDE4D are induced in endothelial cells by oxidative stress through the activation of NOX4 and NOX5 . Whether there are differences between PDE4 isotypes in the endothelial cell function or airway diseases such as asthma or COPD remains unclear. However, seems that PDE4B, PDE4D, and in a lesser extent PDE4A and C play an important role in airway inflammation (PDE4B and PDE4A), endothelial dysfunction (PDE4B and PDE4D) , and bronchial hyperresponsiveness (PDE4D) . In this respect, it has been shown that dexamethasone may down-regulate PDE4 expression in a variety of cells [34, 35]. Supporting these previous data, we showed that dexamethasone inhibited the CSE-induced PDE4 expression in HPAECs and that low doses of rolipram associated with dexamethasone diminished under control levels of all PDE4 isotypes. These results were translated in an increase in cAMP levels rescued from the CSE-induced cAMP down-regulation. These results are of potential value because PDE4 inhibitors showed dose-limiting gastro-intestinal side effects in clinics. Thus, the hypothesis generated from the PDE4 inhibitor–glucocorticoid association is that down-regulation of PDE4 isotypes induced by glucocorticoids allows effective inhibiting of PDE4 enzyme by low doses of PDE4 inhibitor, thus potentiating cAMP levels.
Another possibility that we explored was the modulation of GR by rolipram. In this regard, previous reports showed that PDE4 inhibitors induce the up-regulation of GR . In this work, we observed that rolipram was able to dose dependently increase GRα expression without affecting GRβ expression. Furthermore, CSE significantly decreased GRα expression and increased GRβ expression supporting a mechanistic explanation for glucocorticoid insensitivity. It is believed that overexpression of the GRβ isoform inhibits the action of the ligand-activated GRα, the functional isoform, through which the effects of glucocorticoids are mediated . Because we have observed that CSE may enhance GRβ expression and down-regulate GRα expression, this may explain the poor corticosteroid response observed in HPAECs. Interestingly, rolipram inhibited the CSE-induced GRβ overexpression and the down-regulation of GRα supporting the additive effects of rolipram.
It has been shown that PDE4 inhibitors reduce airway cell apoptosis induced by cigarette smoke in vivo  and enhance endothelial cell survival in vitro . Both early apoptotic signal and decrease in growth rate are signals of endothelial dysfunction [40, 41] and are present in small pulmonary arteries from patients with COPD contributing to leukocyte extravasation. Furthermore, both apoptotic and growth arrest have been described after cigarette smoke exposure [40, 41] as well as following corticosteroid treatment [9, 42], thus increasing endothelial dysfunction and permeability. Because PDE4 inhibitors have been shown to protect against endothelial cell apoptosis induced by cigarette smoke and because PDE4 inhibitors reduce leukocyte extravasation into the lung , it would be of potential interest to analyze possible additive effects of the combination of PDE4 inhibitors and corticosteroids. Thus, we observed that rolipram addition to dexamethasone showed additive effects on CSE-induced HPAEC apoptosis and growth arrest, which could be translated in protective effects of rolipram on endothelial dysfunction induced by cigarette smoke and dexamethasone.
In this respect, previous works showed that the combination of PDE4 inhibitors and corticosteroids showed additive properties in human alveolar macrophages and peripheral monocytes, which reinforce the hypothesis generated in this work [16, 44].
Whether there is a synergistic interaction between glucocorticoids and PDE4 inhibitors, as occurs with the enhanced activity of the association of β2-adrenergics and glucocorticoids in clinics , is nowadays being awaited from the results of a phase II study in which the effect of intranasal fluticasone propionate alone and in combination with GSK256066 (a PDE4 inhibitor) was examined in a cohort of subjects with seasonal allergic rhinitis (http://clinicaltrials.gov/ct2/show/NCT00612820).
In summary, we show for the first time an in vitro evidence of the additive effects of a PDE4 inhibitor and glucocorticoid in a situation of glucocorticoid insensitivity generated by cigarette smoke in HPAECs. Results observed in this work may be of potential value in future therapies directed to improve glucocorticoid efficacy such as in asthmatic smokers and COPD patients.
This work was supported by grants SAF2011-26443 (JC), FIS CP11/00293(JM), CIBERES (CB06/06/0027), and research grants from Regional Government (Prometeo/2008/045, ‘Generalitat Valenciana’; Research grants from ‘Generalitat Valencia’ AP-178/11 [JM]). Support from the CENIT programme (Spanish Government) was obtained.
JM, JLO, and JC contributed to conception and design. JM, JLO, JL, and CS involved in acquisition of data. JM, JLO, JC, and DDA analyzed and interpreted the data. JM, JLO, JL, DDA, CS, and JC drafted the article or revising it critically for important intellectual content. All authors approved this version of the manuscript.