Cigarette Smoke Impairs A2A Adenosine Receptor Mediated Wound Repair through Up-regulation of Duox-1 Expression

Cigarette smoke (CS) exposure and intrinsic factors such as the NADPH oxidases produce high levels of reactive oxygen species (ROS), ensuing inflammatory tissue injury. We previously demonstrated that CS-generated ROS, particularly hydrogen peroxide (H2O2), impaired adenosine stimulated wound repair. We hypothesized that CS exposure modulates expression of Dual oxidase 1 (Duox-1), a NADPH oxidases known to generate H2O2. To test this hypothesis, we used human bronchial epithelial cell line Nuli-1 and C57BL/6 mice. Cells were treated with 5% CS extract (CSE) for various periods of time, and mice were exposed to whole body CS for six weeks. Both CSE and CS treatment induced increased expression of Duox-1, and silencing of Doux-1 improved the rate of cell wound repair induced by CSE treatment. Nuli-1 cells pretreated with thapsigargin but not calcium ionophore exhibited increased Duox-1 mRNA expression. CSE treatment stimulated PKCα activation, which was effectively blocked by pretreatment with diphenylene iodonium, a NADPH oxidase inhibitor. Compared to control, lungs from CS-exposed mice showed a significant increase in PKCα activity and Duox-1 expression. Collectively, the data demonstrated that CS exposure upregulates expression of Duox-1 protein. This further leads to H2O2 production and PKCα activation, inhibiting A2AAR-stimulated wound repair.

To determine the better concentration to activate A 2A AR, Nuli-1 cells were treated with 10 μ M, 100 nM or 10 nM CGS21680, a potent and selective A 2A AR agonist, for 24 h. The expression level of A 2A AR in cells treated with 10 μ M CGS21680 is significantly higher than 10 nM treated group, and observed the expression level of A 1 AR lower than 10 nM treated group (Supp 2). Stimulation of cells with CGS21680 10 μ M improves the rate of repair (Fig. 1D), indicating that A 2A AR activation mediates airway epithelial cell wound repair. However, the A 2A AR-mediated wound closure effect was abolished by CSE exposure (Fig. 1D). CSE-treated cells exhibited no significant release of lactate dehydrogenase compared to no-CSE treated cells (data not shown). To correlate the effect of CSE-generated oxidants on A 2A ARmediated wound closure ability, cells were exposed to H 2 O 2 100 μ M, electrically wounded, and stimulated with CGS21680. Cells exposed to H 2 O 2 and stimulated with CGS21680 had a marked decrease in TEER compared to the CGS21680 alone group (Fig. 1E). Interestingly, lower concentration of H 2 O 2 (1-10 μ M) had no effect in inhibiting CGS21680-stimulated wound closure (data not shown). Because micromolar concentration of CGS21680 may also activate A 2B AR, experiments were repeated using A 2B AR knockout Nuli-1 cells (Fig. 1F) treated with CSE or H 2 O 2 , in the presence or absence of CGS21680 stimulation. Figure 1G shows enhanced wound closure in A 2B AR knockout cells stimulated with CGS21680, which is consistent with the role of A 2A AR in repair after injury. Pre-exposure of A 2B AR knock-out Nuli-1 cells stimulated with CGS21680 to CSE or H 2 O 2 , decreased TEER compared to non-pre-exposed cells implying that A 2A AR but not A 2B AR is the target of inhibitory effect. These findings suggest that the balance of oxidants/antioxidants is central in the regulation of ADO-mediated repair processes and that CSE-generated oxidants play a role in impairing A 2A AR-mediated epithelial cell repair after injury.
Hydrogen Peroxide Activates PKC in Bronchial Epithelial Cells. We reported that ADO occupancy of A 2A AR stimulated wound closure in epithelial cells, and that this effect was mediated through cAMP-dependent kinase PKA pathway 10 . On the other hand, agents that activate protein kinase C (PKC), such as CSE and malondialdehyde-acetaldehyde adducted protein 9,11 , slowed cell migration in wound repair. Oxidative conditions have been demonstrated to selectively modify the regulatory domain of PKC, resulting in persistent active kinase 12,13 . To elucidate the mechanism of impaired A 2A AR-mediated wound healing by CSE and H 2 O 2 in epithelial cells, we investigated the involvement of PKC and oxidant species metabolizing enzymes. Wounded Nuli-1 cells were exposed to H 2 O 2 (1 μ M, a non-cytotoxic concentration), and PKC activity assessed. Some cells were pretreated with catalase (500 U/mL), an enzyme that primarily degrades H 2 O 2 , or with phorbol myristate acetate (PMA, a PKC activator, 100 nM; used as positive control). H 2 O 2 significantly enhanced PKC activation; this response was blocked in cells pretreated with catalase ( Fig. 2A). Because the precise mechanism involved in kinase activation in response to production of H 2 O 2 in airway epithelial cells is not fully understood, we explored the potential role of NADPH oxidases in CSE-stimulated PKC activation. Wounded bronchial epithelial cells were exposed to CSE, and PKC activity quantified; some cells were pretreated with DPI, a NADPH oxidase inhibitor (1 μ M), or with PMA. Figure 2B shows that CSE exposure significantly increased PKC activity. However, pretreatment with DPI markedly reduced CSE-stimulated PKC activation compared to CSE-treated cells. This inhibition of CSE-mediated PKC activation by DPI suggests that Duox oxidase enzymes are present in the airway epithelium.
CSE treatment induces an inflammatory environment. We investigated the role of NADPH oxidase in CSE-mediated release of the pro-inflammatory cytokines IL-6 and IL-8. CSE significantly enhanced both IL-6 and IL-8 release from epithelial cells ( Fig. 2C and D; ***P < 0.001). However, pretreatment with DPI significantly reduced CSE-mediated release of IL-6 and IL-8 ( # P < 0.001). Collectively, the data suggests that NADPH oxidases mediate PKC activation and promote secretion of pro-inflammatory cytokines. This provides evidence that CSE-generated oxidants may further potentiate CSE-mediated inhibition of A 2A AR-stimulated wound repair. CSE Modulates Duox-1 Expression in Nuli-1 Cells. It is well established that CSE exposure induces high levels of ROS formation, which has been linked to the activation of the Nox family of NADPH oxidases 3,4 . Duox-1 and 2 belong to the NADPH oxidases family and are known to generate H 2 O 2 either directly or indirectly by dismutation of superoxide 6 . To further understand the mechanism associated with the deregulation of ADO airway epithelial wound repair, we investigated the effect of CSE on Duox expression and function in the airway epithelial cells. Cells were treated with CSE for 6, 12, 24, 48 or 72 h, and Duox-1 protein expression evaluated. CSE significantly increased Duox-1 abundance at 48 h (Fig. 3A). Duox-2 showed a weaker signal than Duox-1 when exposed to CSE (not shown), suggesting that the predominant Duox protein in the airway epithelium induced by CSE treatment is Duox-1. Figure 3B confirms enhanced abundance of Duox-1 in the cytoplasm of human Nuli-1 cells exposed to CSE. Consistently, H 2 O 2 levels also increased in Nuli-1 cells treated with CSE for 48 h, as compared to untreated Nuli-1 cells (Fig. 3C); note an early peak of H 2 O 2 production at 6 h of CSE exposure. To determine whether Duox-1 function plays an essential role in ADO airway epithelial wound repair, we silenced Duox-1 expression (siDuox-1 cells) and examined the effect of decreased Duox-1 abundance on cell migration. Compared to siControl cells treated with CSE, the rate of TEER elevation is preserved in siDuox-1 cells exposed to CSE and it is similar to that in untreated cells (Fig. 3D). Moreover, stimulation of 5% CSE treated cells with CGS21680 improves the rate of repair; but in Duox-1 silenced Nuli-1 cells, CGS21680 treatment had no significant effect in increasing CSE-stimulated wound closure (Fig. 3E). Collectively, these results demonstrated that CSE stimulates expression of Duox-1, which is a major mediator of airway epithelial cell H 2 O 2 production and of impaired repair after cell injury.

Increased Levels of Intracellular Calcium Upregulate Duox-1 Expression and are Induced by CSE
Treatment. Duox-1 and Duox-2 contain EF motifs calcium-binding domain that can be directly activated by Ca ++ 14,15 . To examine the effect of calcium on Duox-1 expression, Nuli-1 cells were treated for 6 or 24 h with thapsigargin (1 μ M, a calcium pump blocker that inhibits re-entry of cytosolic calcium into the sarcoplasmic reticulum) or with the calcium ionophore A23187 (1 μ M). After 24 h, thapsigargin significantly increased Duox-1 mRNA abundance (P = 0.01) while A23187 treatment had no effect at either time point (Fig. 4A). This demonstrates that intracellular calcium mobilization stimulates Duox-1 gene expression. Treatment with 20% CSE induced a slower and transient increase of intracellular calcium concentration ([Ca ++ ]i) before returning to baseline levels ( Fig. 4B). Nuli-1 cells stimulated with ATP (1 μ M; positive control) exhibited a rapid transient increase of [Ca ++ ]i (data not shown). These studies suggest that increased intracellular calcium mobilization caused by CSE exposure may mediate Duox-1 activation.

Chronic CS Exposure Induces Inflammation, Increases PKC Activity, and Upregulates Duox-1 Expression in Mice.
We studied the effect of CS as a model of airway epithelium damage to elucidate the CS-oxidant driven mechanism(s) in mice exposed to whole body CS for six weeks. Real-time qPCR analysis shows no significant differences in total lung A 2A AR mRNA expression between the air-treated and CS-treated animals (Fig. 5A). Lung pathology revealed increased inflammatory cells and hypertrophy of the bronchial airway epithelium in CS-exposed mice compared to air exposed controls (Fig. 5B). Mice treated with CS showed a significant increase in the total number of BAL cells compared to control animals (Fig. 5C). In addition, IL-6 and CXCL1/KC amount in BAL fluid from CS-treated group was significantly elevated compared to the air group ( Fig. 5D and 5E; IL-6, P = 0.05; KC, P = 0.01). PKC activity in the CS-treated group as compared to the control group ( Fig. 5F; P = 0.001). Finally, immunohistochemical analysis showed Duox-1 expression in tracheobronchial epithelial cells of air-treated control mice, while the CS-treated group had a marked increase Duox-1 cellular localization (Fig. 6A). RT-PCR also revealed augmented Duox-1 transcriptional levels in the CS treated-group compared to controls (Fig. 6B, P = 0.001). Together, these results demonstrate that elevation of pro-inflammatory cytokine secretion, PKC activity, and Duox-1 expression observed in human cultured epithelial cells treated with CSE are recapitulated in mice exposed chronically to CS. We speculate that CS-mediated increase of Duox-1 may further lead to H 2 O 2 production and deregulation of ADO-mediated repair of the airway epithelium.

Discussion
CS is a major risk factor for a number of chronic respiratory diseases including asthma and chronic obstructive pulmonary disease (COPD). COPD is the third leading cause of mortality and morbidity in the United States (CDC Report, 2010), and is characterized by persistent inflammation and injury of both the airways and the parenchyma of the lung. CS exposure has been associated with high levels of oxidative stress and is involved in many biological processes, such as inflammation and carcinogenesis 16,17 . We previously demonstrated that ADO acts via the A 2A AR to promote repair after wound injury in bronchial epithelial cells 10,18 . Recently, we implicated Nuli-1 cells were exposed to 5% CSE for 48 h and Duox-1 visualized using a specific antibody followed by secondary antibody conjugated to Alex Fluor 647 (red); nuclear staining with DAPI (blue). Images are representative of at least four independent experiments. (C) CSE Exposure generates H 2 O 2 in Nuli-1-cells. Nuli-1 cells were exposed to CSE for the indicated time periods, and H 2 O 2 measured in the supernatants. (D) Duox-1 silencing ameliorates CSEmediated inhibition of Nuli-1 cellular migration. 2 × 10 6 cells were transfected with siControl or siDuox-1 siRNAs. After 48 h, cells were exposed to CSE for another 48 h, and wounds created. Resistance was measured in duplicate wells. A representative experiment is shown of two independently performed. (E) Duox-1 silencing impairs CGS21680 improved inhibition of CSE-treated Nuli-1 cellular migration.  CS-generated reactive oxidants in the CS impairment of ADO-mediated wound healing 2 . The present study was designed to further understand the mechanism underlying CS generation of H 2 O 2 and the dysregulation of ADO-stimulated wound repair of airway epithelial cells. We found that CS exposure induced cell calcium mobilization, upregulated mRNA and protein expression of the NADPH oxidase Duox-1 in human cells and mice lungs, and promoted Duox-1 activation. This further leads to H 2 O 2 production, with subsequent PKC activation and promotion of pro-inflammatory cytokines secretion and inhibition of A 2A AR-stimulated wound repair (Fig. 7).
The A 2A ARs can couple to both PKA and PKC dependent downstream cascades and these two transducing systems can crosstalk. Our earlier studies showed that stimulation of PKC blunted A 2A AR-mediated PKA activation 2 , that PKC activation retards wound repair 19,20 , and that exposure to CSE increases PKC activity 2 . In this study, we demonstrated that PKC activity is increased in mice expose to CS. Collectively, our findings suggest that CS inhibition of A 2A AR-mediated activation of PKA is mediated by activation of PKC both in vitro and in vivo.
We previously demonstrated that CS-generated H 2 O 2 is implicated in blunting ADO wound repair 1,2 . In this study, we observed that both CSE and H 2 O 2 significantly enhanced PKC activation, and that CSE stimulated the release of IL-6 and IL-8 in human cultured airway epithelial cells. These findings support our earlier studies in which CS-mediated IL-8 release blunted ciliary motility (a first line of defense in the airway) 21 and activation of PKC required CS-enhanced C5a-mediated release of IL-8 22 . In this study we observed that pretreating cells with CS can induce intrinsic free radical production, which, through Duox-1 and 2 action, is responsible for increased airway H 2 O 2 generation 6,15 . We showed that silencing expression of Duox-1 rendered epithelial cells partially resistant to the negative effect of CSE on wound healing after injury. Little is known about the regulation of dual oxidases in association with smoking and/or COPD. Lavigne et al., reported that Duox-1 mRNA expression did not change after NHBE cells were treated with CSE for 60 min 23 , and Nagai et al., reported that Duox-1 expression in tracheal and bronchial epithelial brushings was significantly down-regulated, although Duox-2 was up-regulated, in current smokers compared to individuals who never smoked 24 . We observed that expression of Duox-1 was temporally regulated by CS exposure (Fig. 3A) and evident in the lower airway of mice (Fig. 6A). Thus, Duox-1 mRNA expression was down-regulated within 12-24 h by CS, but after 48 h the expression of Duox-1 significantly increased, and by 72 h it was further down-regulated. This finding suggests a temporally differential effect of CS on Duox-1 expression, indicating that the observation from Lavigne's group was likely due to acute CS exposure, and it is consistent with Nagai's report on Duox 1 down-regulation in human tissue from current smokers. CS and Air Exposure. Mice were passively treated with CS using a whole-body smoke exposure system (Teague Industries, Davis, CA). Using the Teague device 25 , mice were exposed to smoke from sixty 3R4F reference cigarettes (University of Kentucky, Lexington, KY) per day. Mice receiving CS were gradually brought to their target exposure over a period of seven days, treated 5 days/week for 6 weeks. Control mice were exposed to air in the same manner in a similar apparatus for the same periods of time.

Bronchoalveolar Lavage Fluid (BALF).
Mice were euthanized by intraperitoneal injection of 75 mg/kg sodium pentobarbital (Nembutal; Abbott Labs, Chicago, IL) as previously described by Elliott et al. 27 . Tracheas were exposed, nicked at the bottom of the larynx and cannulated. The proximal ends of the tracheas were tied off and 1.0 mL of cold sterile PBS (Gibco, Grand Island, NY) was gently flushed into the lungs and recovered; a total of three washes were performed and combined. Collected BALF was centrifuged at 300 g for 7 min at 4 °C. The supernatant was stored at − 80 °C until use for measurement of cytokines concentration. Pelleted cells were resuspended in 1.0 ml of PBS. Total cell were counted on a hemocytometer, and 1-5 × 10 3 cells were spun onto glass microscope slides (cytospin 3; Shandon Scientific, Cheshire, UK). Cells were air dried for 24-36 h, fixed, and stained with a Diff-Quik stain set (Dade Behring, Newark, DE). Differential cell counts of at least 300 cells per slide were made according to morphological criteria. The number of cells recovered was calculated and expressed as absolute cell numbers.
Lung Collection and Histology. Whole lungs were excised and inflated to 10 cm H 2 O pressure with 10% formalin (Sigma, St. Louis, MO) to preserve pulmonary architecture. Lungs were embedded in paraffin, and sections (4-5 μ m) were cut and processed for hematoxylin and eosin staining.
Immunohistochemistry and Immunofluorescence. Lung sections were deparaffinized in xylene and rehydrated, and antigen retrieval was performed in PBS containing 0.1% trypsin for 1 h in a humidified chamber. Goat anti Duox-1 (1:500) IgG antibody was from Abcam (Cambridge, USA). Vectastain ABC kit was used to detect Doux-1 with diaminobenzidine (DAB; brown) and counterstained with hematoxylin control samples were blocked with 10% normal goat serum. Cells were seeded at a density of 20000 cells/cm 2 on collagen-coated coverslips overnight and fixed with 4% paraformaldehyde (Rockford, USA) for 15 min. The cells were blocked for 30 min with antibody diluting buffer (1.5% goat serum in PBS with 0.3% Triton X-100) and incubated with Duox-1 antibody (Abcam, Cambridge, USA) at 1:100 dilution overnight. Cells were incubated with Alex Fluor 594-CONJUGATED goat anti-rabbit IgG (1:100) (Life technologies, Eugene, USA) for 30 min. An extensive wash was performed between each step. The cells were then mounted with VECTASHIELD (Vector, Burlingame, USA) and imaged using an Olympus 3i Spinning Disk Confocal Microscope (Olympus, Center valley, PA). PKC Activity Assay. PKC activity was determined as previously described 28 using a reaction solution containing 24 μ g/ml PMA, 30 mM dithiotreitol, 150 μ M ATP, 45 mM Mg-acetate, PKC isoform-specific substrate peptide (Santa Cruz, Dallas, USA), and 10 μ Ci/mL [γ -32 P]-ATP Tris-HCl buffer (pH 7.5). Samples (20 μ l) were added to 40 μ l of the reaction solution and incubated for 15 min at 30 °C. This mixture (60 μ l) was then spotted onto P-81 phosphocellulose paper (Whatman, Clinton, NJ) to halt incubation. Immediately, papers were washed five times with 75 mM phosphoric acid for 5 min, once with 100% ethanol for 1 min, dried, and radioactivity counted in nonaqueous scintillant (National Diagnostics, Atlanta GA). Negative controls were carried out in similar assay conditions but without substrate. PKC activity was normalized to total protein assayed and expressed in picomoles of phosphate incorporated per minute per milligram of total protein. All samples were assayed in triplicate.
Electric Cell Substrate Impedance Sensing (ECIS) Wounding (Migration) Assay. Cell resistance was measured using the electric cell-substrate impedance sensing system (ECIS; Applied BioPhysics, Troy, NY). Cells are cultured onto gold electrodes and impedance values recorded, which are transformed to both resistance and capacitance values. As cells grow and become confluent, they constrict current flow through the electrodes and alter impedance 29 . Nuli-1 cells were grown to confluence on ECIS 96-well plate arrays (8W1E; Applied Biophysics, Troy NY). Cell monolayer were wounded using an elevated field pulse of 1400 μ A at 32,000 Hz applied for 20 sec, producing a uniform circular lesion 250 μ m in size. The wounds were tracked over a period of 24 h. The impedance was measured at 4000 Hz, and the transepithelial resistance (TEER; ohms) normalized relative to the value at the start of data acquisition previous to treatments, and plotted as a function of time.
Cytokines Analysis. IL-6 and IL-8 concentration in Nuli-1 cells exposed to CSE or air in the presence or absence of the NADPH oxidase inhibitor DPI was determined using Quantikine ELISA kits (R&D Systems, Minneapolis, MN). Similarly, IL-6 and CXCXI/KC levels in BALFs collected from mice exposed to CS or air were determined using Quantikine ELISA kits (R&D Systems, Minneapolis, MN). Samples were incubated for 2 h and washed with Quantikine wash buffer as per manufacture instruction. Absorbance was measured at 450 nm with a 540 nm correction, and concentrations calculated per the manufacturer's instructions.
Taqman Real-time RT-PCR. RNA was extracted using the Magmax 96 kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions. RNA concentration and purity was determined with NanoDrop spectrophotometer. cDNA was synthesized by using 100 ng of total RNA and TaqMan reverse transcription kit (Applied Biosystems, Foster City, CA). Real-time PCR reactions were prepared in triplicate using 1X TaqMan Master Mix (Applied Biosystems) and primers and probes, as previously described 18  siRNA. Accell TM -siRNAs were purchased from Dharmacon (GE Healthcare Dharmacon Inc., US). Nuli-1 cells were plated at a density of 30-40% confluency and incubated overnight. Accell Duox-1 siRNA or non-targeting siRNA was added to final concentration of 1 μ M. Transfected cells were incubated for 96 h before treatment.
Western Blot. Cells lysates were prepared in RIPA buffer (Cell signaling, Danvers, USA). Equal amounts of protein lysates were separated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene difluoride (PVDF) membranes. Membranes were blocked at room temperature for 1 h with 5% blotto (Santa Cruz, Dallas, USA) followed by exposure to Duox-1 antibody (1:500 dilution in 5% blotto) overnight at 4 °C. After washing with TBS plus 1% Tween-20, membranes were incubated with 1:5000 diluted goat anti rabbit IgG-HRP antibody for 1 h at room temperature. Membranes were incubated with Pierce Western Blotting Substrate (Pierce, Rockford, USA) and imaged. Antibody to GAPDH (Abcam, Cambridge, USA) was used as a loading control.
Detection of Hydrogen Peroxide. Hydrogen peroxide (H 2 O 2 ) concentrations were measured using a Hydrogen Peroxide Assay Kit (Abcam, Cambridge, USA), according to the manufacturer's protocol.

Measurement of Ca 2+
Release. Intracellular Ca 2+ mobilization was assessed using a modification of procedures previously described by Tian et al. 31 . Briefly, Nuli-1 cells were grown at 1.5 × 10 4 cells/cm 2 on placenta collagen type VI coated LabTek glass chambers (Nunc, NY, USA). Cells were incubated with 5 μ M Fluo-3 AM for 30 min at 37 °C, washed and placed on the stage of a Zeiss Confocal LSM 410 confocal microscope equipped with an Argon-Krypton Laser, 25 mW argon laser, 2% intensity (Thornwood, NJ). 0-20% CSE was manually added to a corner of the glass chamber and allowed to diffuse. Transient Ca 2+ signal (measured as change in fluorescence, Δ F) was recorded.

Statistics.
Results are expressed as mean ± SE. Data were statistically analyzed using Student's paired t-test followed by Tukey's multiple-comparison test. Statistical differences among groups were determined using one-way ANOVA followed by Tukey's multiple-comparison test (Graph-Pad Prism, version 5; Graph-Pad, San Diego, CA). Significance was assigned at P < 0.05.