Sleep Breath (2015) 19:297–306 DOI 10.1007/s11325-014-1019-4

ORIGINAL ARTICLE

Airway cell involvement in intermittent hypoxia-induced airway inflammation C. Philippe & Y. Boussadia & V. Prulière-Escabasse & J F. Papon & C. Clérici & D. Isabey & A. Coste & E. Escudier & M P. d’Ortho

Received: 31 December 2013 / Revised: 18 May 2014 / Accepted: 3 June 2014 / Published online: 4 July 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract Purpose Respiratory inflammation has been described in patients with obstructive sleep apnea syndrome, but it is unknown whether the increased neutrophil and interleukin (IL)-8 levels observed in induced sputum reflect systemic or local airway inflammation. We assessed the potential role of resident cells in intermittent hypoxia-induced airway inflammation. C. Philippe : J. F. Papon : D. Isabey : A. Coste INSERM, U955, Equipe 13, Créteil, France C. Philippe (*) APHP, Groupe Hospitalier Pitié Salpêtrière, Unité des Pathologies du Sommeil, 47-83, Bd de l’hôpital, 75651 Paris, France e-mail: [email protected] Y. Boussadia : C. Clérici INSERM, U773, Equipe 8, Paris, France V. Prulière-Escabasse INSERM, U955, Equipe 11, Créteil, France V. Prulière-Escabasse : J. F. Papon : A. Coste Centre Hospitalier Intercommunal de Créteil, Service d’ORL, Créteil, France V. Prulière-Escabasse : J. F. Papon : A. Coste Université Paris Est, Faculté de Médecine, Créteil, France E. Escudier INSERM, UMRS 933, Paris, France E. Escudier APHP, Hôpital Armand-Trousseau, Service de Génétique et Embryologie Médicales, Paris, France C. Clérici : M. P. d’Ortho APHP, Hôpital Bichat, Service de Physiologie - Explorations Fonctionnelles, DHU FIRE, Paris, France C. Clérici : M. P. d’Ortho Université Paris Diderot, Sorbonne Paris Cité, 75018 Paris, France

Methods Airway epithelial cells (AEC) and bronchial smooth muscle cells (BSMC) were exposed to intermittent hypoxia (IH) in vitro. Cell supernatants were assessed for matrix metalloproteinase, growth factor, and cytokine expression. The role of IH on neutrophil and BSMC migration capacities was evaluated, and the effect of supernatants from IH-exposed or control AEC was tested. Results Compared to normoxic conditions, 24 h of exposure to IH induced a significant increase of MMP-9 and MMP-2 expression and pro-MMP-9 activation (p<0.05), and IL-8 (p < 0.05), platelet-derived growth factor (PDGF)-AA (p<0.05), and vascular endothelial growth factor (VEGF) (p<0.05) expression by AEC and VEGF expression (p= 0.04) by BSMC. Neutrophil chemotaxis and BSMC migration were enhanced by IH and supernatants of IH-exposed AEC (112.00±4.80 versus 0.69±0.43 %, p=0.0053 and 247±76 versus 21±23, p=0.009 respectively). This enhanced BSMC migration was totally abolished in the presence of an antibody blocking PDGF-AA. Conclusions These data suggest a specific inflammatory response of airway cells to IH, independently of systemic events. Keywords Airway epithelial cells . Airway smooth muscle cells . Neutrophil migration . Obstructive sleep apnea syndrome Abbreviations AECs Airway epithelial cells BSMCs Bronchial smooth muscle cells FCS Fetal calf serum IH Intermittent hypoxia IL Interleukin MMPs Matrix metalloproteinases Nx Normoxia OSAS Obstructive sleep apnea syndrome PBS Phosphate buffered saline

298

PDGF SDSPAGE VEGF

Sleep Breath (2015) 19:297–306

Platelet-derived growth factor Sodium dodecyl sulfate polyacrylamide gel electrophoresis Vascular endothelial growth factor

Introduction The desaturation-reoxygenation sequence is a typical pattern coupled with a majority of respiratory events, resulting in intermittent hypoxia (IH). The most common form of IH is observed in obstructive sleep apnea syndrome (OSAS). IH is notably implicated in the cardiovascular complications of OSAS by triggering activation of pro-inflammatory pathways [1]. It may directly promote oxidative stress with formation of reactive oxygen species (ROS) [2], cytokine production, and inflammatory circulating cell activation [3]. Although systemic inflammation has been extensively studied in clinical cohorts and in vivo and in vitro models of IH [4], little is known about the role of IH in local airway inflammation. Clinical studies conducted on sputum or breath condensates concluded on the existence of inflammation based on increased neutrophil percentages [5, 6] and increased levels of interleukin (IL)-8 [6, 7], IL-6, and oxidative stress marker 8-isoprostane [8]. A study conducted in an in vivo model of chronic IH in mice investigated the role of the proinflammatory NF-κB pathway in various tissues and demonstrated increased NF-κB expression in lungs in response to IH [9]. However, it is unknown whether lung inflammation is exclusively due to systemic inflammation or results from activation of specific airway cells. The objective of the present study was to evaluate the potential role of airway cells in initiating IH-induced inflammation. An in vitro model was used to evaluate local effects independently of systemic inflammation. We measured cytokines, growth factor secretion, and matrix metalloproteinase (MMP) secretion under conditions of normoxia (Nx), IH, and sustained hypoxia (SH) in conditioned media of primary cultures of two types of airway cells, i.e., airway epithelial cells (AEC) and bronchial smooth muscle cells (BSMC). We also analyzed the effect of AEC, exposed to IH compared to Nx, on neutrophil and BSMC migration capacities.

was established on the basis of clinical history, endoscopic findings, and computed tomography results. Patients were asked to stop systemic and/or topical nasal treatment 1 month before surgery. The study protocol was approved by our institution’s institutional review board (Comité Consultatif de Protection des Personnes, Hôpital Henri Mondor). Informed consent was obtained from all patients before the study. Nasal polyposis samples were cultured at an air-liquid interface adapted from a culture model originally described for human tracheobronchial cells [10] and further developed and characterized [11]. Briefly, samples were immediately placed in DMEM-Ham’s F-12 medium (Invitrogen Life Technologies, Cergy-Pontoise, France) containing antibiotics (100 U/mL penicillin, 100 mg/mL streptomycin, 2.5 μg/mL amphotericin B, and 100 mg/mL gentamicin), at 4 °C, for transportation. The samples were then rinsed in phosphate buffered saline (PBS) with dithiothreitol (5 nM) and placed overnight at 4 °C in a PBS antibiotics solution containing 0.1 % pronase (Sigma-Aldrich, St Quentin Fallavier, France). After pronase neutralization by incubation in DMEM/F12 with 5 % fetal calf serum (FCS, Invitrogen, Gibco), the cell pellets were resuspended in 0.25 % trypsinEDTA solution for 3 min and incubated in DMEM/F12-antibiotics with 5 % FCS before centrifugation (300×g, 7 min) and resuspension. Finally, AECs were plated (1×106 cells/insert) on permeable Transwell® supports (Transwell Clear, Costar, USA) previously coated with collagen IV (10–20 μg/cm2). For the first 24 h, AECs were incubated in 1 mL of DMEM/F12antibiotics with 2 % Ultroser G outside the insert and DMEM/F12-antibiotics with 5 % FCS inside the insert. After 24 h, the medium inside the insert was then removed to place the cells at an air-liquid interface. The medium outside the insert was then changed daily. Transepithelial resistance (about 1,100 Ω cm2, indicating the development of tight junctions) and transepithelial potential difference (−40 mV from the fourth day onward, indicating the existence of ion transport) were measured every 3 days using a microvoltmeter (EVOM®, World Precision Instrument, GB). A well-differentiated state was reached after the first week, with the presence of ciliated and secretory cells. All experiments were performed using 14- to 21-day-old AEC cultures, presenting stable cell differentiation.

Material and methods BSMC Primary cell cultures AEC AEC were obtained from 18 adult patients undergoing surgery for nasal polyposis. The diagnosis of primary nasal polyposis

Human primary BSMC (Lonza, Basel, Switzerland) were used for all experiments. Upon reception, cryopreserved cells were cultured in T-25 flasks in Smooth muscle Basal Medium (SmBM), supplemented with fibroblast growth factor 2 (FGF2, 2 ng/mL), epidermal growth factor (EGF, 0.5 ng/mL), and

Sleep Breath (2015) 19:297–306

insulin (5 μg/mL), as well as a mixture of antibiotics, including gentamicin (100 ng/mL) and amphotericin B (0.1 ng/mL) and 5 % FCS. Thawing, subculturing, and harvesting procedures were performed according to the manufacturer’s instructions. Experiments were performed with cells at passages 2 to 4. Neutrophil preparation Neutrophils were obtained from peripheral blood of six healthy nonsmokers. Blood was drawn in sterile disposable syringes containing EDTA. Plasma and neutrophils were separated using Ficoll-plaque™ (GE Healthcare, Paris, France) according to the manufacturer’s instructions. Neutrophil purity was assessed using the May-Grünwald Giemsa technique. Neutrophils were resuspended in serum-free medium at a concentration of 2.5×106 cells/mL. IH model Cell exposure to IH Cell cultures were exposed to six alternating cycles consisting of 0.5 to 1 % O2 for 30 min followed by 21 % O2 for 30 min at 37 °C in a 13-L volume chamber. The chamber was equilibrated with a hypoxic gas mixture of 0 % O2, 5 % CO2, and 95 % N2 (Air Products SAS, Paris, France), until an O2 fraction (FO2) of 0.5 to 1.0 % was achieved. The O2 fraction and partial pressure of O2 were monitored at various timepoints of each cycle to ensure that a stable FO2 of 1 % was achieved. An FO2 of 1 %, corresponding to a partial pressure of 50–53 mmHg, was consistently achieved within 7 min and remained stable until return to normoxia. This 0.5–1 % ambient FO2 induced a partial pressure of 94 to 96 mmHg in the cell culture media, versus 160 mmHg during standard normoxic culture conditions A normoxic control experiment was run in parallel by maintaining the cells under normoxia throughout the 6-h period. SH was obtained by maintaining the cells under 1 % O2 hypoxia for 6 h. Supernatant samples were collected either at the end of IH (6 h) or at 24 h and were stored at −80 °C. At least five different cultures were used for each condition. Soluble mediators: cytokines and gelatinases Cytokine assays IL-8, IL-1β, IL-6, platelet-derived growth factors (PDGFs) AA and AB, TGFα, TGFβ (-1, -2, and -3), and vascular endothelial growth factor (VEGF) were assayed in supernatants using the Milliplex Human Cytokine panel (Millipore, Molsheim, France) and the microassay platform at the Institut de Recherche en Santé, IFR65, Paris, France. Data were

299

expressed as protein concentrations (pg/mL). Assays were performed in duplicate. Gelatinase expression Basal AEC gelatinase expression was detected using zymography on gelatin SDS-PAGE as previously described [12]. This method detects both activated and latent forms of MMPs. Ten to 20 μL of supernatants were loaded onto the lanes of an 8 % SDS-polyacrylamide gel with 1 mg/mL gelatin. After electrophoresis, the gel was then washed for 30 min in 2.5 % Triton X-100 and incubated overnight at 37 °C in a buffer (100 mM Tris/HCl, 10 mM CaCl2, pH 7.6). After 0.5 % Coomassie Brilliant Blue R250 staining, gelatindegrading enzymes were identified as clear zones against a blue background. Quantitative evaluation was performed using Image J software (NCBI, National Center for Biotechnology Information, NIH, Bethesda, USA). Effect of IH on neutrophil and BSMC migration This set of experiment was performed in Transwell® inserts (8.0 μm pore size) (Dutsher Dominique, Brumath, France). The 2.5×106 neutrophil or 2.5×104 BSMC was added to the upper chamber filled with 250 μL of serum-free medium. Neutrophils and BSMC were allowed to migrate for 3 and 6 h, respectively, under either normoxic or IH conditions. After incubation, the cells were fixed with cold methanol for 30 min, then washed three times with PBS, and stained by May-Grünwald Giemsa (Sigma-Aldrich). Cells on the upper surface of the membrane were removed by gently scraping with a cotton swab. Cells that had migrated through the membrane and were adherent to the lower part of the filter were observed under the microscope (Inverted optical microscope, Axiovert 10, ZEISS, Carl Zeiss, Le Pecq, France) and counted in three different fields (×100 magnification) for each of the triplicate membranes. Migration was evaluated as the number of Neutrophils or BSMCs per field at the lower surface of the filter, expressed as percentage of controls. Positive controls were IL-8 (25 ng/mL) (R&D Systems, Lille, France) for neutrophils and PDGF-AB (10 ng/mL) (SigmaAldrich) for BSMC. Actin staining of BSMC was performed under Nx, IH, SH, or after stimulation of BSMC with 5 ng/mL PDGF-AB under Nx. BSMC were plated in Lab-Tek (Dutsher Dominique) dedicated for immunofluorescence. After rinsing with PBS, cells were fixed with 1 % glutaraldehyde at 37 °C, then rinsed again and permeabilized with 2 % Triton X-100 for 20 min. Staining was performed with Alexa Fluor 568 Phalloidin (Invitrogen Life Technologies) for 30 min. Nuclei were stained with SYTOX® Green (Invitrogen Life Technologies). F-actin was visualized by confocal microscopy (Carl Zeiss) (×40 magnification).

300 Fig. 1 a, b, c Effects of intermittent hypoxia (IH) (gray) compared to normoxia (Nx) (light gray) and sustained hypoxia (SH) (dark gray) on secretion by human airway epithelial cells (AECs) after 24 h. IL-8 (a), PDGF-AA (b), VEGF (c). Data are expressed as mean ± SD. Statistically significant differences are indicated (*p<0.05). d Effects of IH on gelatinase expression by human airway epithelial cells (AECs) after 6 and 24 h. Left: representative zymogram for AEC supernatants, with enzyme molecular masses on the left. Bands at 88 and 92 kDa: active MMP-9 and latent MMP-9 (proMMP-9), respectively. MMP-2 appears as a 72-kDa band. Table shows quantitative zymogram analysis showing significant increases in MMP-9 and MMP-2 after 24 h (AU)

Sleep Breath (2015) 19:297–306

(A) IL-8 (pg/ml)

(B) PDGF AA (pg/ml)

*

10000

800

8000

*

600

6000

400

4000

200

2000

0 Nx

0 Nx

IH

SH

IH

SH

(C) VEGF (pg/ml) 8000

*

7000

*

6000

*

5000 4000 3000 2000 1000 0 Nx

IH

SH

(D) MMP : 24 h

Nx

IH

MMP9 (AU) 3718+/-1354 7589+/-4770 * MMP2 (AU) 2182+/-1467 4921+/-5129 *

Effects of AEC conditioned medium on neutrophil and BSMC migration Neutrophil migration assay: chemotaxis on agarose Agarose plates were assessed using the Nelson RD method [13]. Agarose (Invitrogen) was dissolved in sterile distilled water by heating in a boiling water bath for 10 min. After cooling to 48 °C, the agarose was mixed with an equal volume of MEM 2 × 137 mg of MgCl 2 , and 10 % FCS. Five milliliters of agarose medium was poured into a tissue culture dish (3 cm in diameter), allowed to harden, and transferred at 4 °C for 30 min. Twelve equidistant wells were then created using a biopsy punch. The center well received 4-μL of the cell suspension containing 106 purified neutrophils. The inner well received 4 μL of PBS (control medium) and the outer well received either 4 μL of 1 μM formyl-Met-Leu-Phe (fMLP) as positive control or 4 μL of supernatant. After 2 h of incubation at 37 °C in a humidified atmosphere containing 5 % CO2 in air, the cells were fixed by addition of 3 mL of absolute methanol for 30 min. The gel

was then removed and the plates were stained using the May-Grünwald Giemsa technique. The linear distance traveled by the cells from the margin of the center well toward the chemotactic factor or toward control medium was measured. Migration was expressed as the ratio of the distance traveled from the center well to the test well over the total distance between these two wells. Results were expressed as the mean of three independent experiments, each conducted in quadruplicate.

BSMC migration assay: chemotaxis across a membrane These experiments were performed as described in paragraph 4. In order to test AEC-induced BSMC migration, the lower chambers were filled with non-conditioned AEC basal medium (control) or AEC supernatants obtained under normoxic or IH conditions for 3 h. A full set of additional experiments was conducted to test the role of PDGF-AA in BSMC migration induced by IHexposed AEC, using a blocking antibody against PDGF-AA (Millipore). The antibody was preincubated with supernatants for 1 h at a concentration of 10 μg/mL [14].

Sleep Breath (2015) 19:297–306

Statistical analysis

301

(A) VEGF (pg/ml) *

Data were expressed as mean ± SD. The Wilcoxon rank sum test was used to assess differences between nonparametric data. The significance of between-group differences for quantitative variables was determined using the Kruskal-Wallis test followed by the Mann-Whitney U test. All statistical analyses were performed with Statview 5.0 (SAS Institute Inc, Cary, NC, USA). p values less than or equal to 0.05 were considered to be statistically significant.

Nx

*

IH

SH

(B) IL-8 (pg/ml) Results IH-induced airway cell inflammatory response IH-induced AEC inflammatory response Cytokine and growth factor secretion No differences were observed between Nx and IH in terms of TGFα, TGFβ, IL1β, or IL-6 secretion. Figure 1a shows that IL-8 secretion by AECs was increased 18 h after the end of IH compared to Nx (24 h after the beginning of the experiment), while no difference was observed at the end of the exposure to hypoxia (6 h). Similarly, IH exposure increased the release of PDGF-AA and VEGF in the supernatants (Fig. 1b, c). When AECs were exposed to SH for 6 h, the only significant difference versus Nx after 24 h was an increase in VEGF secretion (726.2± 538 pg/mL versus 272.5±259, p<0.05, Fig. 1c). VEGF expression was much more markedly enhanced by exposure to IH compared to SH (Fig. 1c). Gelatinase expression After 6 h, MMP-9 was detected in its latent 92-kDa form with no significant difference between Nx and IH. MMP-2 was not detected in either condition. After 24 h, 92 kD pro-MMP-9 was significantly increased (p<0.05) and its active 88-kDa form was present in IH-exposed supernatants but not in normoxic supernatants (Fig. 1d). IH was also associated with a significant increase in 72-kD pro-MMP2 expression compared to normoxia (p<0.05). IH-induced BSMC inflammatory response Cytokine and growth factor secretion. No differences were observed between Nx and IH conditions in terms of IL-1β, TGFβ, and PDGF secretion. At 24 h, a significant increase in VEGF (217 pg/ml±155 versus 98±66, p=0.04) (Fig. 2a) and a trend toward an increase in IL-8 (161±92 pg/ml versus 102 ±49 pg/ml, p=0.079) (Fig. 2b) were observed with IH compared to Nx. Under SH conditions, only a significant increase in VEGF (275±187 pg/ml, p=0.04) was observed, with no statistical difference compared to IH.

Nx

IH

SH

Fig. 2 Effects of intermittent hypoxia (IH) (gray) compared to normoxia Nx (light gray) and sustained hypoxia (SH) (dark gray) on cytokine secretion by bronchial smooth muscle cells (BSMCs) after 24 h. VEGF (a), IL-8 (b). Data are expressed as mean ± SD. Statistically significant differences are indicated (*p<0.05)

Gelatinase expression In line with previous studies, zymographic analysis of gelatinase activities in BSMCsupernatants only demonstrated a 72-kDa band at baseline, corresponding to the molecular mass of latent MMP-2 [15]. No modification was observed in response to IH, highlighting the different responses of HNEC and BSMC to IH. IH-induced migration IH induction of neutrophil migration IH significantly enhanced migration of neutrophils from healthy controls compared to Nx. Adding IL-8 significantly increased neutrophil migration and potentiated the effect of IH (Fig. 3a). Induction of BSMC migration by IH Analysis of BSMC migration across a membrane IH induced increased migration of BSMC across the membrane of Transwell® inserts, compared to the slight spontaneous migration observed under Nx conditions (62±55 versus 9±7, p= 0.018. Of note, IH-induced BSMC migration was greater than

302

(A) *

*

%

* *

(B)

*

*

*

%

Fig. 3 Effects of IH on neutrophil (a) and BSMC (b) migration. Data are expressed as a percentage relative to Nx. Statistically significant difference are indicated (*p<0.05). Positive control is IL-8 (25 ng/ml) for neutrophils and PDGF (10 ng/ml) for BSMC. c Effects of IH on BSMC morphology . Staining of F-actin showing stress fiber formation in response to IH and SH compared to NX. Stress fiber formation in response to IH was more pronounced than that observed under SH conditions or than induced by 5 ng/ml PDGF

Sleep Breath (2015) 19:297–306

*

(C)

Nx

Nx + PDGF IH

SH

Sleep Breath (2015) 19:297–306

that observed under SH condition (62±55 versus 44±52, p= 0.028) (Fig. 3b). Analysis of BSMC morphology In line with these enhanced migration capacities, actin stress fiber formation was observed in BSMC under IH conditions (Fig. 3c) and was more pronounced than that observed under SH conditions or than induced by 5 ng/mL PDGF.

303

(A) fMLP or AEC supernatants Neutrophils

PBS

Neutrophil and BSMC migration induced by IH-exposed AEC supernatants

Supernatants of IH-exposed AEC induced strong chemoattraction of neutrophils compared to supernatants of Nx-exposed AEC (112.00±4.80 versus 0.69±0.43 %, p= 0.0053). This chemotactic response was similar to that induced by fMLP (Fig. 4b). BSMC migration induced by IH-exposed AEC supernatants Supernatants of IH-exposed AEC induced a marked increase in BSMC migration capacities compared to supernatants of Nx-exposed AEC (247±76 versus 21±23, p=0.009). This increased migration was blunted by the use of an antiPDGF-AA (247±76 versus10±2, p=0.014) (Fig. 5).

(B) 1.4

*** ***

1.2

distance ratio

Neutrophil migration induced by IH-exposed AEC supernatants

1 0.8 0.6 0.4 0.2 0

PBS

Discussion The main findings of this study are increased IL-8 and PDGFAA release induced by AEC exposed to IH and increased neutrophil and BSMC migration under IH and induced by conditioned media of AEC exposed to IH. We also showed that increased BSMC migration was PDGF-dependent. These results illustrate the direct effect of IH on airway cells, suggesting that the respiratory inflammation observed in OSAS patients may reflect a local response as well as being a possible consequence of systemic inflammation and may also contribute to systemic inflammation. We acknowledge certain limitations of this study. Firstly, the IH model consisted of six 30-min cycles of hypoxia followed by 30 min of Nx, which certainly does not reflect the IH cycling observed during OSAS. However, a similar cycling pattern has been used in several previous studies by other groups [3, 16, 17]. In vitro models of IH have been developed to obtain more rapid cycles, using preconditioned hypoxic medium [18], or to ensure proper diffusion of oxygen through the culture medium [19]. However, these methods have their own limitations, as changes of culture medium may

fMLP

Nx

IH

Fig. 4 a Schematic representation of the experimental setup. b Chemotaxis of human neutrophils induced by human airway epithelial cell (AEC) supernatants. Migration was expressed as the ratio of the distance traveled by neutrophils from the center well to the test well over the total distance between these two wells. Positive control was fMLP (1 μM). Data are expressed as mean ± SD. Statistically significant differences are indicated (***p=0.005)

be responsible for shear stress that per se can induce an inflammatory cascade and interfere with the effects of IH. In addition, this method does not allow the collection of supernatants, which was essential for our experiments. Another limitation concerns the low O2 fraction used (0.5–1 %). However, this low FO2 is in accordance with previous in vitro studies on intermittent hypoxia [18, 20] and, under our experimental conditions, induced a partial pressure of 94 to 96 mmHg in the cell culture medium, while the O2 partial pressure was 160 mmHg under normoxic conditions. These low partial pressures are consistent with clinical observations showing that some patients experience very low PaO2, as reflected by SpO2 values as low as 60–70 %, during apneas. A major strength of our study is the use of primary human cell cultures, obtained from either tissue samples (AEC) or commercial sources (BSMC). The use of human cells eliminated a potential bias due to interspecies differences and the use of

304

(A) *

*

Cell number / field

Fig. 5 Effects of human airway epithelial cell (AEC) supernatants on BSMC migration after 3 h. a Effects of supernatants of AEC exposed to Nx (light gray) compared to supernatants of AEC exposed to IH (gray). Addition of a PDGF-AA blocking antibody (10 μg/ml) completely abolished BSMC migration (dark gray). Data were expressed as the number of BSMCs/field. Statistically significant differences are indicated (*p<0.05). b Optical microscopy of BSMC migration induced by supernatants of AEC exposed to either Nx or IH, showing a marked increase in the number of cells adherent to the lower part of the Transwell® insert in response to IH-exposed AEC, with a typical pattern of migrating cells

Sleep Breath (2015) 19:297–306

(B)

primary cultures instead of cell lines ensured that the cell responses were not linked to transformation, a key point when studying cell response to low O2 partial pressure. Lastly, AEC cultures were conducted at an air-liquid interface, a culture condition representative of their position in the respiratory tract, allowing the airway epithelium to be consistently studied in vivo physiologic conditions. To our knowledge, this is the first in vitro study to address the issue of a specific local respiratory inflammation in response to IH independently of any systemic contribution, although the role of IH in inducing systemic inflammation has been widely described, notably in endothelial cells [18]. IH-induced inflammation in the lung has been demonstrated in vivo with increased NFkB expression, but the assay was performed on total lung homogenates, preventing any discrimination between systemic or local processes and this study also did not investigate the role of resident respiratory cells [9]. Airway inflammation in OSAS patients has been previously reported. A preliminary study reporting bronchial inflammation demonstrated by increased neutrophils in patient sputum [5] was confirmed by further studies showing oxidative stress and inflammatory cytokine production in breath condensates, independently of obesity [8], and increased IL-8 in induced sputum and NO exhalation [6]. Very recently, a study focusing on the relationships between systemic inflammation and airway inflammation in OSAS demonstrated increased IL-8 and VEGF in sputum [21]. This increased secretion of

Nx

IH

IH + antiPDGF-AA

cytokines in sputum was related to the severity of OSAS, while the increased secretion of cytokines in serum was predominantly related to obesity. The present study demonstrated increased IL-8 production by both AEC and BSMC exposed to IH. In addition, the IL-8 concentration in conditioned media of AEC was 20-fold higher than that of BSMC, consistent with a concentration gradient allowing recruitment of neutrophils from the vascular compartment, via the bronchial subepithelial compartment to the airway lumen, highlighting the participation of these two cell types in local inflammation, in line with the current theory concerning lung inflammation [22]. AEC- and BSMCinduced migration of neutrophils into the airway lumen could also be potentiated by direct activation of neutrophils by IH, with increased spontaneous chemotaxis and IH potentiation of the response to IL-8. This activation of neutrophils chemotaxis is in line with Dyugovskaya’s results showing activation of circulating neutrophils and delayed apoptosis induced by IH [3]. In the airways, as in the endothelium and blood vessels, this type of microenvironment could promote epithelial cell damages via activated neutrophils [3]. Increased production and activation of matrix metalloproteinase gelatinases was also observed in AEC exposed to IH, consistent with the production of other inflammatory mediators, IL-8, and PDGF-AA. This increased production of gelatinases could also contribute to airway epithelial damage, as suggested by previous studies conducted in various models [23, 24]. In particular, IH-induced MMP and cytokine

Sleep Breath (2015) 19:297–306

expression differed between AEC and BSMC demonstrating the cell specificity of this response. Spontaneous BSMC migration was enhanced in response to both IH and supernatant of AEC previously exposed to IH. In line with this enhanced migration, we observed increased formation of stress fibers after exposure to IH, comparable to that induced by PDGF-AA. BSMC migration induction by IH and AEC supernatants reflects activation of BSMC by IH and the PDGF-dependent interplay between these two cell types, as confirmed by total inhibition of migration by PDGF-AA blocking antibodies. No histological evidence of airway remodeling in OSAS is available to date, but decreased airway obstruction [25] and nonspecific bronchial hyperreactivity [6] have been reported, consistent with airway inflammation. An increased prevalence of asthma and poorer asthma control have also been described in OSAS patients, especially in children but also in adults [26], whereas experimental data obtained in transgenic mice showed that NFkB induces inflammation and hyperresponsiveness of airway epithelium [27]. It is noteworthy that increased morbidity and mortality are observed in COPD patients with coexisting OSAS compared to COPD patients without sleep apnea [28]. The most obvious cause for this increased risk are the cardiovascular consequences of OSAS [29]. However, worsening of respiratory inflammation as a result of IH could predispose to exacerbations and poorer disease control. Lastly, together with other tissues, such as adipose tissue [30], the inflammatory response of resident airway cells, AEC and BSMC, to IH could contribute to systemic inflammation. Acknowledgments We are very grateful to D. Bokar-Thire and L. Margarit for their technical assistance and to Dr A. Saul for his English expertise. We are very grateful to Bernadette Lescure at the microassay platform (Institut de Recherche en Santé, IFR65, Paris, France) for cytokines assays. This work was funded by the nonprofit Air-Liquide Foundation and INSERM. Conflict of interest The authors have no conflicts of interest to declare.

References 1. Garvey JF, Taylor CT, McNicholas WT (2009) Cardiovascular disease in obstructive sleep apnoea syndrome: the role of intermittent hypoxia and inflammation. Eur Respir J 33:1195–1205 2. Lavie L (2003) Obstructive sleep apnoea syndrome—an oxidative stress disorder. Sleep Med Rev 7:35–51 3. Dyugovskaya L, Polyakov A, Lavie P, Lavie L (2008) Delayed neutrophil apoptosis in patients with sleep apnea. Am J Respir Crit Care Med 177:544–554 4. Levy P, Tamisier R, Minville C, Launois S, Pepin JL (2011) Sleep apnoea syndrome in 2011: current concepts and future directions. Eur Respir Rev 20:134–146 5. Salerno FG, Carpagnano E, Guido P, Bonsignore MR, Roberti A, Aliani M, Vignola AM, Spanevello A (2004) Airway inflammation

305 in patients affected by obstructive sleep apnea syndrome. Respir Med 98:25–28 6. Devouassoux G, Levy P, Rossini E, Pin I, Fior-Gozlan M, Henry M, Seigneurin D, Pepin JL (2007) Sleep apnea is associated with bronchial inflammation and continuous positive airway pressure-induced airway hyperresponsiveness. J Allergy Clin Immunol 119:597–603 7. Carpagnano GE, Spanevello A, Sabato R, Depalo A, Palladino GP, Bergantino L, Foschino Barbaro MP (2010) Systemic and airway inflammation in sleep apnea and obesity: the role of ICAM-1 and IL8. Transl Res 155:35–43 8. Carpagnano GE, Kharitonov SA, Resta O, Foschino-Barbaro MP, Gramiccioni E, Barnes PJ (2002) Increased 8-isoprostane and interleukin-6 in breath condensate of obstructive sleep apnea patients. Chest 122:1162–1167 9. Greenberg H, Ye X, Wilson D, Htoo AK, Hendersen T, Liu SF (2006) Chronic intermittent hypoxia activates nuclear factor-kappaB in cardiovascular tissues in vivo. Biochem Biophys Res Commun 343: 591–596 10. Gruenert DC, Finkbeiner WE, Widdicombe JH (1995) Culture and transformation of human airway epithelial cells. Am J Physiol 268: L347–L360 11. Coste A, Brugel L, Maitre B, Boussat S, Papon JF, Wingerstmann L, Peynegre R, Escudier E (2000) Inflammatory cells as well as epithelial cells in nasal polyps express vascular endothelial growth factor. Eur Respir J 15:367–372 12. d’Ortho MP, Clerici C, Yao PM, Delacourt C, Delclaux C, FrancoMontoya ML, Harf A, Lafuma C (1997) Alveolar epithelial cells in vitro produce gelatinases and tissue inhibitor of matrix metalloproteinase-2. Am J Physiol 273:L663–L675 13. Nelson RD, Quie PG, Simmons RL (1975) Chemotaxis under agarose: a new and simple method for measuring chemotaxis and spontaneous migration of human polymorphonuclear leukocytes and monocytes. J Immunol 115:1650–1656 14. Bosse Y, Thompson C, Stankova J, Rola-Pleszczynski M (2006) Fibroblast growth factor 2 and transforming growth factor beta1 synergism in human bronchial smooth muscle cell proliferation. Am J Respir Cell Mol Biol 34:746–753 15. Nishihara-Fujihara M, Shoji S, Maeda I, Shimoda T, Nishima S, Okamoto K (2010) Involvement of fibronectin and matrix metalloproteinases in airway smooth muscle cell migration for the process of airway remodeling. Allergol Int 59:267–275 16. Gozal E, Sachleben LR Jr, Rane MJ, Vega C, Gozal D (2005) Mild sustained and intermittent hypoxia induce apoptosis in PC-12 cells via different mechanisms. Am J Physiol-Cell PH 288:C535–C542 17. Lattimore JD, Wilcox I, Nakhla S, Langenfeld M, Jessup W, Celermajer DS (2005) Repetitive hypoxia increases lipid loading in human macrophages—a potentially atherogenic effect. Atherosclerosis 179:255–259 18. Ryan S, Taylor CT, McNicholas WT (2005) Selective activation of inflammatory pathways by intermittent hypoxia in obstructive sleep apnea syndrome. Circulation 112:2660–2667 19. Baumgardner JE, Otto CM (2003) In vitro intermittent hypoxia: challenges for creating hypoxia in cell culture. Respir Physiol Neurobiol 136:131–139 20. Yuan G, Nanduri J, Bhasker CR, Semenza GL, Prabhakar N (2005) Ca2+/calmodulin kinase-dependent activation of hypoxia inducible factor 1 transcriptional activity in cells subjected to intermittent hypoxia. J Biol Chem 280(6):4321–4328 21. Aihara K, Oga T, Chihara Y, Harada Y, Tanizawa K, Handa T, Hitomi T, Uno K, Mishima M, Chin K (2013) Analysis of systemic and airway inflammation in obstructive sleep apnea. Sleep Breath 17(2): 597–604 22. Munakata M (2006) Airway remodeling and airway smooth muscle in asthma. Allergol Int 55:235–243 23. Calvet JH, Planus E, Rouet P, Pezet S, Levame M, Lafuma C, Harf A, D'Ortho MP (1999) Matrix metalloproteinase gelatinases in sulfur

306 mustard-induced acute airway injury in guinea pigs. Am J Physiol 276:L754–L762 24. Guignabert C, Taysse L, Calvet JH, Planus E, Delamanche S, Galiacy S, d'Ortho MP (2005) Effect of doxycycline on sulfur mustardinduced respiratory lesions in guinea pigs. Am J Physiol Lung Cell Mol Physiol 289:L67–L74 25. Zerah-Lancner F, Lofaso F, Coste A, Ricolfi F, Goldenberg F, Harf A (1997) Pulmonary function in obese snorers with or without sleep apnea syndrome. Am J Respir Crit Care Med 156:522–527 26. Teodorescu M, Polomis DA, Hall SV, Teodorescu MC, Gangnon RE, Peterson AG, Xie A, Sorkness CA, Jarjour NN (2010) Association of obstructive sleep apnea risk with asthma control in adults. Chest 138: 543–550

Sleep Breath (2015) 19:297–306 27. Pantano C, Ather JL, Alcorn JF, Poynter ME, Brown AL, Guala AS, Beuschel SL, Allen GB, Whittaker LA, Bevelander M, Irvin CG, Janssen-Heininger YM (2008) Nuclear factor-kappa b activation in airway epithelium induces inflammation and hyperresponsiveness. Am J Respir Crit Care Med 177:959–969 28. Marin JM, Soriano JB, Carrizo SJ, Boldova A, Celli BR (2010) Outcomes in patients with chronic obstructive pulmonary disease and obstructive sleep apnea: the overlap syndrome. Am J Respir Crit Care Med 182:325–331 29. Drager LF, Polotsky VY, Lorenzi-Filho G (2011) Obstructive sleep apnea: an emerging risk factor for atherosclerosis. Chest 140:534–542 30. Jun J, Polotsky VY (2009) Metabolic consequences of sleepdisordered breathing. ILAR J 50:289–306