Long-Term Cultures of Polarized Airway Epithelial ... - ATS Journals

Long-Term Cultures of Polarized Airway Epithelial Cells from Patients with Cystic Fibrosis Ludovic Wiszniewski, Lan Jornot, Tecla Dudez, Alessandra Pagano, Thierry Rochat, Jean Silvain Lacroix, Susanne Suter, and Marc Chanson Laboratory of Clinical Investigation III, Department of Pediatrics, Division of Pulmonary Medicine, and Clinic of Oto-Rhino-Laryngology, Geneva University Hospitals, Geneva; Department of Pathology, Geneva Medical School, Geneva, Switzerland

The poor ability of respiratory epithelial cells to proliferate and differentiate in vitro into a pseudostratified mucociliated epithelium limits the general use of primary airway epithelial cell (AEC) cultures generated from patients with rare diseases, such as cystic fibrosis (CF). Here, we describe a procedure to amplify AEC isolated from nasal polyps and generate long-term cultures of the respiratory epithelium. AEC were seeded onto microporous permeable supports that carried on their undersurface a preformed feeder layer of primary human airway fibroblasts. The use of fibroblast feeder layers strongly stimulated the proliferation of epithelial cells, allowing the expansion of the cell pool with successive passages. AEC at increasing passage were seeded onto supports undercoated with airway fibroblasts and exposed to air. Either freshly isolated or amplified AEC could differentiate into a pseudostratified mucociliated epithelium for at least 10 mo. Thus, CF epithelia cultures showed elevated Na⫹ transport, drastic hyperabsorption of surface liquid, and absence of cAMP-induced Cl⫺ secretion as compared with non-CF cultures. They were also characterized by thick apical secretion that hampered the movement of cell surface debris by cilia. However, CF respiratory epithelia did not show increased production of mucins or IL-8. The method described here is now routinely used in our laboratory to establish long-term cultures of well differentiated respiratory epithelia from human airway biopsies. Keywords: airway epithelial cells; cystic fibrosis; epithelial–mesenchymal interactions; human cell model; long-term cell differentiation

Human conducting airways are lined with a tall, pseudostratified epithelium that includes several cell types, among which, basal, ciliated, and secretory cells are the most abundant. The developed respiratory tract epithelium serves to condition incoming air with moisture and salts and to orchestrate the pulmonary defense system. The regulation of composition and volume of the airway surface liquid (ASL) covering the epithelium is achieved by active transepithelial transport of electrolytes (1). In addition, the airway epithelium has the ability to restore its integrity after injury (1–3). The crucial roles of the mucociliated epithelium in maintaining healthy lungs are probably best illustrated for diseases such as asthma, chronic obstructive pulmonary disease, and cystic fibrosis (CF). In CF, mechanisms of airway protection may be defective, leading to increased coloni-

(Received in original form May 2, 2005 and in final form September 2, 2005) This work was supported by grants from the Swiss National Science Foundation (3100-067120.01 to M.C. and 3100A0-100621-1 to J.S.L.), the Swiss Cystic Fibrosis Foundation, and the French Association Vaincre la Mucoviscidose. Correspondence and requests for reprints should be addressed to Marc Chanson, Ph.D., Laboratory of Clinical Investigation III, Department of Pediatrics, HUG-P.O. Box 14, Micheli-du-Crest, 24, 1211 Geneva 14, Switzerland. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Cell Mol Biol Vol 34. pp 39–48, 2006 Originally Published in Press as DOI: 10.1165/rcmb.2005-0161OC on September 22, 2005 Internet address: www.atsjournals.org

zation by various pathogens, progressive destruction of the airways, and, ultimately, respiratory failure (4–6). A limitation in the understanding of the CF pathogenesis is the difficulty in studying the link between the mutant CF gene (cystic fibrosis transmembrane conductance regulator [CFTR]) expression and the associated molecular and cellular effects in the absence of an identified animal model of this disease (7). Although cell lines derived from airways of patients with CF have been established, they exhibit large heterogeneity, making it difficult to relate their properties solely to the CF phenotype (5). Thus, technical advances that permit the production of human airway epithelial cell (AEC) cultures resembling the in vivo epithelium are valuable in assessing native airway functions and to understand airway disease pathogenesis (6, 8, 9). Human AEC grown on permeable supports at the air–liquid interface develop a mucociliated morphology, thus providing a model for studying airway cell lineage, differentiation, and function (9–13). Although primary cells most closely represent in vivo physiology, they are only available intermittently, and variably contain cell contaminants. Although these problems may be eliminated by cell passage, AEC proliferate poorly and their ability to differentiate markedly decreases after the first passage. Indeed, most epithelial cells rapidly lose their differentiated features in vitro, probably because appropriate signals from extracellular matrix, growth factors, and hormones have not been fully defined (9, 11). The number of cultures that can be initiated is therefore limited by the yield of airway cells obtained from tissues or other commercially available sources. These difficulties have certainly thwarted researchers from using primary CF AEC cultures if their laboratories were not linked to large CF health centers. Here, we describe a novel approach to grow and differentiate in vitro human nasal AEC into polarized respiratory epithelia that retain CF features. It is recognized that mesenchymal–epithelial interactions have an instructive role in lung development and repair (3, 14, 15). Using similar methodologies as those previously reported to culture skin equivalents from isolated keratinocytes (16), we investigated the role of an airway-derived fibroblast feeding layer on proliferation and differentiation of respiratory cells obtained from nasal polypectomies. We show that well differentiated airway epithelium could be generated from serially passaged AEC and maintained in culture for up to 6–10 mo in the presence of either proliferation- or differentiationinducing subsets of airway fibroblasts. Long-term airway epithelium cultures generated from CF AEC showed an absence of cAMP-dependent Cl⫺ transport and dehydrated liquid surface, but no difference in the amounts of mucin and IL-8 secreted. The procedure described here requires few biopsies and no specific equipment, thus providing an efficient and reproducible method for generating large amounts of polarized respiratory epithelia in most laboratories. This will certainly accelerate in vitro studies on obtaining mechanistic insights in to rare pathologies of the human airways.

40

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 34

2006

MATERIALS AND METHODS Origin and Isolation of Airway Fibroblasts and Epithelial Cells Nasal and bronchial cells were obtained from patients undergoing surgical nasal polypectomy or partial or total lobectomy. All experimental procedures were explained in full, and all subjects provided informed consent. The study was conducted according to the declaration of Helsinki on biomedical research (Hong Kong amendment, 1989), and received approval from our local ethics commission. AEC were obtained from four patients who were homozygous for the ⌬F508 mutation of CFTR and from 27 individuals without CF. AEC were isolated exclusively from nasal polyps by overnight pronase (Roche, Mannheim, Germany) digestion at 4⬚C (13). Thereafter, the isolated AEC suspension was plated on tissue culture dishes at a density of 0.5–1 ⫻ 106 cells/ml in 5 ml of Dulbecco’s modified Eagles Medium (DMEM) supplemented with 10% FCS, and incubated in 5% CO2 at 37⬚C for 1–3 h to remove fibroblasts from the cell suspension. Nonadherent cells were then collected. To isolate fibroblasts, small explants (1–2 mm2) were prepared from nasal polyps or bronchi, allowed to attach on 10-cm Petri dishes, and covered with DMEM supplemented with 10% FCS. Fibroblasts migrated out of the explants within 2–3 wk. Fibroblasts were weekly dissociated by enzymatic treatment with 0.05% trypsin/0.02% EDTA, according to standard protocols. Each clone was first tested for its ability to divide without changes in morphologic appearance for at least 10 passages. Purity of the fibroblast clones was evaluated by positive immunolabeling for vimentin but absence of smooth muscle actin (data not shown). The proliferation of amplified pools of fibroblasts was arrested using 8 ␮g/ml mitomycin C (Sigma Chemical Co., Zu¨rich, Switzerland) for 5 h. Postmitotic fibroblasts were then rinsed with PBS and frozen in aliquots of 106 cells/ml of DMEM/FCS supplemented with 10% DMSO for later use.

AEC Colony-Forming Test The ability of fibroblast clones to stimulate proliferation of AEC was evaluated in a coculture assay. For each clone, 105 postmitotic fibroblasts were seeded on 60-mm Petri dishes. Two days later, isolated AEC were plated on top of the feeder layers at a density of 50 cells/cm2 in DMEM:F12 (3:1) medium supplemented with 10% FCS, 2 mM Lglutamine, 10 ng/ml epithelial growth factor, 1 ␮M hydrocortisone, 5 ␮g/ml insulin, 5 ␮g/ml transferrin, 30 nM triiodothyroine, 180 ␮M adenine, 5.5 ␮M epinephrine, 2.5 ␮g/ml fungizone, 100 U/ml penicillin, and 100 ␮g/ml streptomycin (hereafter referred to as growth medium). After 9 d in culture, cells were fixed with 4% paraformaldehyde (PFA) and stained with Azur blue (Sigma). AEC colonies formed were then viewed by a CCD camera connected to a personal computer, and their number and size determined by morphometric analysis using Scion Image 4.0.2 (Scion Corporation, Fredericks, MD). A colony-forming index for each fibroblast feeder layer was calculated by multiplying the number of colonies by their averaged size. Experiments were performed in triplicate for each fibroblast clones.

Airway Cell Cultures A schematic of the human airway cell model is shown in Figure 1. To prepare feeder layers, 5 ⫻ 104 and 1 ⫻ 105 postmitotic fibroblasts were seeded on the under surface of, respectively, 0.33- or 4.7-cm2 Transwell polyester membranes (Transwell 3470 or Transwell 3450, respectively; Corning Costar, Cambridge, MA). To this end, Transwell inserts were inverted and fibroblasts allowed to attach for 3 h at 37⬚C. The Transwell inserts were then turned back and transferred to 24- or 6-well plates, respectively. Dispersed AEC were plated on top of the porous membrane undercoated with postmitotic fibroblasts and allowed to attach for 24 h at 37⬚C (Figure 1). Proliferation of AEC was obtained by seeding 2.5 ⫻ 105 cells onto 4.7-cm2 Transwell membranes undercoated with fibroblasts. Cells were grown to confluence under submerged conditions in growth medium. Media were refreshed every 2 d, and confluence was reached within 5–7 d. Cells were then washed with PBS without Ca2⫹/Mg2⫹, dissociated with trypsin, and counted. For each filter, ⵑ 106 cells were recovered and equally seeded again on three new 4.7-cm2 Transwell inserts. This first enzymatic dissociation corresponds to passage 1 (P1), P0 referring to cultures initiated from AEC freshly isolated from polyps. AEC prolif-

Figure 1. Schematic representation of the human airway epithelium model. Freshly dissociated airway epithelial cells (AEC) are seeded onto porous membrane (PM) undercoated with postmitotic fibroblasts (PMF). In submerged conditions (“Medium” in the lower and upper chambers), cells proliferate and reach confluence. After one or several cycles of amplification, AEC are seeded onto porous supports undercoated with PMF for 24 h. The apical surface of epithelial cells is then exposed to air to induce their differentiation. Note that the pseudostratified architecture of the differentiated airway epithelium is not represented in the scheme.

eration was also studied in growth medium without fibroblasts or in commercially available airway cell growth media from Clonetics (Clonetics Corp., San Diego, CA) and PromoCell (PromoCell GmbH, Heidelberg, Germany). For the latter experiments, AEC were plated onto Petri dishes coated with 40 ␮g/ml Vitrogen 100 (Nutacon, Leimuiden, The Netherlands) according to standard procedures. For each culture condition, population doublings (PD) was calculated according to the formula: PD ⫽ ln(Nt/N0) ⫻ 1/ln2, where N0 is the number of cells at the time of plating and Nx is the number of cells at the time of harvest. Differentiation was induced by seeding 2–2.5 ⫻ 105 AEC onto 0.33-cm2 Transwell membranes undercoated with postmitotic fibroblasts. Cells were allowed to attach for 24 h before the apical medium was removed to obtain an air–liquid interface. The culture medium consisted of DMEM:F12 (3:1) containing 1.5% Ultroser G (Biosepra, Ciphergen Biosystems, Cergy-St.-Christophe, France), 2.5 ␮g/ml fungizone, 100 U/ml penicillin, and 100 ␮g/ml streptomycin. The criterion for differentiation was primarily the generation of a tight, tall, and ciliated pseudostratified airway epithelium. Cultures in the absence of fibroblasts were also generated. Freshly isolated or passaged AEC were seeded as described previously here on Transwell membranes coated with Vitrogen.

Immunofluorescence Microscopy Immunofluorescence analysis was performed according to standard protocols. Briefly, cells on culture inserts were fixed with methanol for 3 min (aquaporin 3, Na⫹-K⫹ ATPase ␣1, occluding, mucin 5AC) and 10 min (CFTR) at ⫺20⬚C or in PFA (Ki-67, ␤-tubulin, vimentin) overnight at 4⬚C. Methanol- and PFA-fixed cells were then rinsed with PBS and successively treated with 0.3% triton X-100 for 20 min, NH4Cl 0.5 M for 15 min, and 2% BSA prepared in PBS for 30 min. Cells were then incubated overnight at 4⬚C or 2 h at room temperature with either rabbit antibodies against occludin (Zymed Laboratories, South San Francisco, CA), aquaporin 3, and mucin 5AC (both kindly provided Dr. E. Puchelle, INSERM, Reims, France), or mouse antibodies against Ki-67 (clone Mib-1; Dako Cytomation, Blostrup, Denmark), CFTR (clone 24.1; R&D Systems GmbH, Wiesbaden, Germany), ␤-tubulin (Sigma), and Na⫹-K⫹ ATPase ␣1 (Upstate Biotechnology, Lake Placid, NY). An amplification step was used to detect CFTR, aquaporin 3, and Ki-67 by incubating cells with biotinylated goat antibodies for 1 h (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). After rinsing with PBS, cells were incubated for 1 h at room temperature using FITC/rhodamine–conjugated antibodies or streptavidin-FITC (Molecular Probes, Inc., Eugene, OR). Antibodies were used at dilutions ranging from 1:100 to 1:800. In some experiments, rhodaminephalloidin (Sigma) was used at 1:800 for 30 min to counterstain the cell preparation. After rinsing, the filters were cut out from their supports and mounted in Vectashield (Vector Laboratories, Burlingame, CA) before microscope examination.

Wiszniewski, Jornot, Dudez, et al.: Human Airway Cell Model

Light and Confocal Microscopy Preparations were examined using an inverted TMD300 microscope (Nikon AG, Ku¨snacht, Switzerland) equipped with a high-sensitivity black and white CCD Visicam camera (Visitron systems GmbH, Puchheim, Germany). Images were captured using the software Metafluor 4.01 (Universal Imaging Corp., Downington, PA). Preparations were also examined using a Zeiss LSM510 laser scanning confocal microscope (Carl Zeiss Inc., Oberkochen, Germany). Images were acquired and analyzed with LSM510 software or Imaris 3.2 (Bitplane AG, Zu¨rich, Switzerland). All images were processed using Adobe Photoshop 5.5 (Adobe Systems Inc., Mountain View, CA).

Movies Six-month-old AEC grown on 0.33-cm2 Transwell membranes were examined using the TMD300 microscope equipped with a high-sensitivity black and white CCD Visicam camera. Sequential images (1 image per second) were captured for ⵑ 8 min using Metafluor 4.01. Movies were then reconstituted using the MPEG-1 format with the same software.

Electron Microscopy AEC on culture inserts were fixed for 60 min at room temperature in 2.5% glutaraldehyde. Membrane supports with attached cells were detached from inserts. Preparations were post-fixed in 2% phosphatebuffered osmium tetroxide, dehydrated in graded ethanols, and then embedded in Epon. Thin sections were examined with a Philips CM10 400 electron microscope (Philips SA, Zurich, Switzerland) at 70 kV.

Ussing Chambers The bioelectric properties of AEC cultures were studied by placing the filters in Ussing chambers (Jim’s Instruments, Iowa City, IA). The apical and basal chambers were filled with a Krebs buffer containing (in mM): 135 NaCl, 2.4 KH2PO4, 0.6 KH2PO2, 1.2 CaCl2, 1.2 MgCl2, 5 glucose, and 5 Hepes (pH 7.4), maintained at 37⬚C and gassed with 100% O2. The transepithelial potential difference was voltage-clamped at zero, and the resulting short-circuit current (Isc) recorded continuously through Ag-AgCl electrodes and 3 M KCl agar bridges. The transepithelial resistance (R) was calculated using Ohm’s law from the recorded Isc changes resulting from 5-s square-voltage pulses of 5 mV imposed across the cell culture. The fibroblast feeder layer did not contribute to series resistance. After the voltage-clamp mode was established, 100 ␮M amiloride was applied to the apical solution to measure the fraction of the basal Isc due to amiloride-sensitive epithelial Na⫹ channels (ENaC). The cAMP agonists (10 ␮M forskolin ⫹ 50 ␮M 3-isobutyl1-methylxanthine [IBMX] or 10 ␮M isoproterenol) were added to the apical solution to stimulate transepithelial Cl⫺ through CFTR channels. Bumetanide (100 ␮M) was used to block Cl⫺ transport.

Western Blots AEC on culture inserts were rinsed with PBS and scraped into an ice-cold solubilization buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, and a mixture of protease inhibitors. After a 30-min incubation, the samples were centrifuged at 4⬚C for 10 min at 50,000 ⫻ g. Supernatants were recovered, and total amounts of protein were determined by a bicinchoninic acid quantification assay (Sigma). Equal amounts of protein were electrophoresed on a 7% SDS-PAGE and electrotransferred onto Immobilon-P polyvinylidene difluoride membranes (Millipore AG, Volketswill, Switzerland). Membranes were then soaked overnight at 4⬚C in a 2% defatted milk saturation buffer containing 10 mM Tris-HCl (pH 7.4), 2 mM EDTA, 133 mM NaCl, 0.05% Triton X-100, and 0.2% sodium azide. Proteins were then immunoblotted with antibodies against human CFTR (clone 24.1) or ␣ENaC (Affinity Bioreagents, Golden, CO). This step was followed by a 1-h incubation with goat anti-mouse or anti-rabbit IgG secondary antibodies conjugated to peroxidase (Jackson Laboratories). Immunoreactivity was detected through the Super Signal West Pico kit (Pierce, Rockford, IL).

Apical Surface Liquid Absorption To evaluate apical surface liquid absorption, 200 ␮l of NaCl 0.9% (pH 7.4), containing 10 mM Hepes (Sigma), was added to the apical surface of CF and non-CF epithelium equivalents. After 4 d, the

41

remaining liquid on the epithelium surface was recovered and its weight determined using a microbalance. For theses experiments, AEC obtained from three patients with CF and eight individuals without CF were used. Absorption was measured in triplicate on 3-mo-old airway epithelia generated from P0–P4 AEC. Results were expressed as mean rate of apical liquid absorption (␮l/d) ⫾ SEM and were compared using unpaired t tests.

Mucous Detection Quantitation of secreted airway mucins was performed by dot-blot analysis by Alcian blue and periodic acid-Schiff (PAS) staining. ASLs were collected with 400 ␮l of 0.9% NaCl. Samples were serially diluted in NaCl and applied to a prewetted Immobilon-P membrane (Millipore Corp., Bedford, MA) using a 96-well dot-blot filtration apparatus (Bio-Rad, Hercules, CA). Each well was washed twice with 50 mM sodium bicarbonate. The membranes were then removed and rinsed 2 ⫻ 5 min with Milli-Q water. To quantify Alcian blue–reactive material, membranes were incubated in a 5% (wt/vol) solution of BSA for 5 min, washed in Milli-Q water, and incubated 5 min with 1% Alcian blue, prepared in 3% acetic acid (pH 2.5). Membranes were then washed 3 ⫻ 5 min with Milli-Q water and dried in the dark. PAS-positive material was determined by incubating the membranes for 5 min with periodic acid solution. Membranes were rinsed and incubated for an additional 15 min with Schiff’s reagent to stain neutral mucins. Membranes were then washed 3 ⫻ 5 min with Milli-Q water and dried. Dot-blots were viewed with a CCD camera connected to a personal computer, and intensity of the staining in each well was determined using GeneTools software (Syngene, Cambridge, UK).

IL-8 Production IL-8 was measured using an ELISA kit (CLB, Amsterdam, The Netherlands) in basal media of non-CF and CF AEC cultures that were collected every 2 d. Only assays having standard curves with a calculated regression line value ⬎ 0.95 were accepted for analysis. To evaluate the response of cells to a proinflammatory cytokine, airway equivalents were stimulated for 2 h with 100 U/ml TNF-␣ (Bachem AG, Bubendorf, Switzerland). Thereafter, the basal medium was collected every 2 d for 8 d. Values are expressed as mean ⫾ SEM and compared using unpaired t tests.

RESULTS Selection of Airway Fibroblast Clones

In a first series of experiments, airway fibroblasts were isolated from non-CF nasal polyps (n ⫽ 17) and bronchial explants (n ⫽ 3). Among these clones, six were discarded because of contamination or weak growth properties (Table 1). The remaining 14 clones exhibited cells that divided without change in their morphology and that expressed vimentin, a positive marker for fibroblasts (data not shown). Fibroblast clones were amplified, treated with mitomycin C, and frozen until assayed. Ten fibroblast clones were selected for further examination of their influence on AEC proliferation and differentiation. Fibroblast clones were first tested for their ability to stimulate proliferation of AEC by calculating the index of colony-forming cells in a coculture system. Fibroblasts had variable effects on AEC growth, as illustrated in Figure 2 for three of the clones. According to the calculated index, a score was given to each fibroblast clone (Table 1). This approach allowed selection of 4 out of 10 clones that strongly stimulated AEC proliferation (hereafter referred to as “proliferating fibroblasts”). The efficiency of these clones to stimulate AEC proliferation was further investigated in a two-dimensional assay. As illustrated in Figure 1, dissociated AEC were seeded at a density of 50,000 cells/cm2 onto Transwell membranes undercoated with proliferating fibroblasts clones. Under submerged conditions, AEC formed confluent monolayers within 5–7 d. Proliferation of AEC was further confirmed by immunodetection of Ki-67, a nuclear antigen known to be expressed during the active phases of the cell cycle (data

42


2006

TABLE 1. SELECTION OF AIRWAY FIBROBLAST CLONES Fibroblast Clones None F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 F13 F14 F15 F16 F17 F18 F19 F20

Growth

Morphology

⫹⫹ ⫹ — — — ⫹⫹⫹ ⫹⫹⫹ — ⫹ — — ⫹ ⫹⫹ ⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹ ⫹ ⫹⫹⫹

⫹ ⫹ — — ⫹ ⫹⫹ ⫹⫹⫹ C — C C — ⫹ ⫹ ⫹⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹ ⫹⫹⫹

Epithelial Cell Proliferation (Submerged)

Epithelial Cell Differentiation (A-L Interface)

— ⫹

⫹

⫹ ⫹⫹⫹

⫹⫹⫹ —

⫹

⫹

⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹ — —

⫹ — — ⫹⫹⫹ ⫹ ⫹⫹⫹

Definition of abbreviations: A-L, air–liquid; C, contaminated cultures; F, fibroblast clones. Fibroblast clones isolated from airway biopsies were evaluated for their ability to divide without alteration of their morphology or expression of specific cell markers, and to support proliferation and differentiation of airway epithelial cells. Positive (⫹, ⫹⫹, ⫹⫹⫹) and negative (⫺) scores were given for this evaluation, according to the criteria described in RESULTS. The normal airway mucosa is a tall, pseudostratified epithelium in which every cell is attached to the basal membrane. A (⫹⫹⫹) score was given for airway epithelia exhibiting a pseudostratified morphology. A (⫺) score was given for epithelia exhibiting a pluristratified morphology (several layers of cells). A (⫹) score was given for epithelia in which cells grew as a flattened epithelium (all cells were attached to the basal membrane, but their height was not representative of the normal airway epithelium).

Figure 2. Test of colony-forming units to select fibroblast clones. (A ) The ability of different fibroblast clones (F7, F20, and F6) to stimulate AEC proliferation is compared. Whereas the F7 clone stimulated the formation of large AEC colonies (left image), F6 formed less colonies of moderate size (right image). The F20 clone stimulated the formation of small but numerous colonies (middle image). (B ) The number and the size of the colonies were measured to calculate the colony-forming index, as described in MATERIALS AND METHODS.

not shown). Under these conditions, airway cells could be dispersed by trypsin digestion and seeded again or frozen without contamination by the underlying fibroblastic cells. As compared with a classic procedure, which consists of growing AEC seeded on Vitrogen in the presence of defined growth media (Clonetics, PromoCell), the use of fibroblast subsets allowed much stronger amplification of the AEC pool. Indeed, a PD of 10.3 was obtained for AEC exposed to proliferating fibroblasts, as compared with 3.2 and 2.8 for AEC grown on Vitrogen in the presence of PromoCell and Clonetics media, respectively. We next evaluated the influence of the various fibroblast clones on morphology of AEC that have been cultured at the air–liquid interface for 30 d. As shown in Figure 3, images of phase-contrast and x-z confocal rhodamine-phalloidin–stained airway epithelia revealed different morphologies of the airway epithelia when AEC were cultured in the absence (Figures 3A and 3D) or presence (Figures 3B, 3C, 3E, and 3F) of fibroblasts. In the absence of fibroblasts, AEC cultured on Vitrogen exhibited a flattened morphology, growing as a cell monolayer (Figure 3D). In the presence of fibroblasts, airway cells grew in an apparently tighter epithelium. Depending on the fibroblast clone used, confocal microscopy revealed either areas of stratification (Figure 3E) or a pseudostratified appearance (Figure 3F). Under the latter conditions, round and columnar cells were indeed observed within a tall epithelium—a morphology resembling that of the respiratory epithelium. Table 1 summarizes the effects of the different feeder layers on airway cell differentiation. Of note, stratification of the epithelium was only observed in AEC that were cultured on proliferating fibroblasts, as confirmed by immunodetection of Ki-67–positive epithelial cells (data not


43

Figure 3. Effects of fibroblasts on AEC architecture. AEC at passage 2 were seeded onto porous supports and cultured at the air–liquid interface for 21–60 d. Phase-contrast images (A–C ) and x-z confocal fluorescent views of rhodamine-phalloidin–labeled cells (D–F ) showed the morphology of AEC grown in the absence (A and D ) and in the presence of proliferating (B and E ) or differentiating (C and F ) fibroblasts. Vitrogen was used in (A and D ), whereas fibroblast clones F7 and F20 were used in (B and E) and (C and F ), respectively. Scale bar: 50 ␮m in A–C, and 20 ␮m in D–F.

shown). Using this approach, 3 out of 10 fibroblast clones (hereafter referred to as “differentiating” fibroblasts) were selected for their influence on AEC differentiation toward a pseudostratified epithelium. Of note, the differentiating and proliferating fibroblasts represent a distinct population of cells. Microscopic Characterization of AEC Grown on Fibroblast Feeder Layers

The properties of the distinct subsets of fibroblasts were used to generate non-CF and CF airway epithelia. Briefly, AEC were subcultured on proliferating fibroblasts under submerged conditions and seeded at increasing passage on Transwell membranes undercoated with differentiating fibroblasts and cultured at the air–liquid interface. No apparent difference was observed in amplification or differentiation of CF AEC compared with nonCF cells. Under these culture conditions, a pseudostratified organization of the airway epithelium was confirmed by electron microscopy. Basal ciliated cells and mucus-secreting cells (Figure 4A) could be identified. Ciliated cells showed the apical junctional complex and cilia with typical ultrastructural features (Figures 4B and 4C). To quantify the proportion of ciliated cells, airway epithelia were subjected to immunolabeling for ␤-tubulin, a component of the cilia microtubule structure. As shown in Figure 4D, large areas of tubulin-labeled ciliated cells were detected by confocal microscopy. Cilia were usually detected within 21 d of culture. The proportion of ciliated cells increased with time in culture, reaching 90% after 5–6 mo, and ciliated cells were still observed in 9–10-mo-old epithelia. No difference in ciliogenesis, however, was observed between non-CF and CF cultures. Quantitative analysis on 30-d-old epithelia revealed that epithelia generated in the presence of fibroblasts were more ciliated (P ⬍ 0.001), with 425 ⫾ 14 ciliated cells/field (mean ⫾ SEM, n ⫽ 60) than airway cultures performed on Vitrogen coating (30 ⫾ 7 ciliated cells/field, n ⫽ 60). The intact epithelium contained polarized cells that line distinct apical and basolateral membranes with unique proteins. To further characterize the human cell model, airway cultures were immunolabeled with antibodies against several markers of intact epithelia. As shown in Figure 5, occludin was detected apically at the tight junctions, whereas Na⫹, K⫹-ATPase was found in basolateral membranes (Figure 5A). CFTR and aquaporin 3 (AQP3) were detected in apical and basal membranes, respectively (Figures 5B and 5C). The specific membrane expression of these proteins is characteristic of well polarized airway epithelia. Of note, similar membrane localization for Na⫹, K⫹-ATPase and AQP3 was observed in CF airway epithelia. Bioelectric Properties of AEC Grown on Fibroblast Feeder Layers

To examine for functional differentiation of the airway epithelium cultures, R, an index of barrier function, and Isc, an index of solute transport, were measured in a modified Ussing cham-

ber. A typical recording of Isc changes in response to successive application of various drugs is shown in Figure 6A (left panel). As shown, addition to the Na⫹-channel blocker, amiloride decreased basal Isc, whereas a cocktail of forskolin/IBMX increased Isc. The latter effect was abolished by addition of bumetanide (Figure 6A). These measurements were performed in cultures generated from freshly isolated AEC (P0) or AEC

Figure 4. Pseudostratified organization of the airway epithelium grown in the presence of fibroblasts. (A ) Electron micrograph of primary culture of cystic fibrosis (CF) AEC grown at their air–liquid interface in the presence of fibroblasts revealed the presence of basal (B), ciliated (C), and mucous-secreting (M) cells. Magnification, ⫻2,200. (B ) Cilia present their normal structure of nine outer doublet microtubules surrounding the central pair of single microtubules. Magnification, ⫻52,000. (C ) Apical intercellular tight junctional complex is present between adjacent non-CF airway cells. Magnification, ⫻28,500. (D ) 3-D projection of confocal microscope images of a non-CF airway epithelium immunolabeled with a ␤-tubulin antibody to visualize ciliated cells. Scale bar: 10 ␮m.

44


2006

Figure 5. AEC grown at the air–liquid interface in the presence of fibroblasts are well polarized. (A ) Fluorescent x-z confocal images of co-immunolabeled non-CF airway epithelia for occludin (red) and Na⫹, K⫹-ATPase (green). As expected, occludin is detected at the tight junctional complex and Na⫹, K⫹-ATPase at the basolateral membranes. (B ) Apical expression of cystic fibrosis transmembrane conductance regulator (CFTR; green). (C ) Basal expression of aquaporin 3 (green) and (D ) detection of airway cells secreting MUC5AC (green). The epithelium was counterstained with rhodamine-phalloidin (red). Scale bar: 10 ␮m.

serially amplified (P1–P3) and grown in the absence (but presence of Vitrogen) or presence of fibroblasts. All cultures were maintained at the air–liquid interface for 36 ⫾ 1 d. Of note, bioelectric properties could be measured in all cultures generated in the presence of fibroblasts, independent of the AEC passage. When amplified in the presence of proliferating fibroblasts but differentiated on Vitrogen, the number of AEC cultures that showed a measurable R decreased by 60% at each cell passage. The proportion of tight airway epithelia generated from AEC that have been amplified on Vitrogen in the presence of growth media and then differentiated at the air–liquid interface was also assessed. Under these conditions, none of the cultures generated from AEC at P2 or higher passage had a measurable R. Amiloride-induced Isc changes (⌬IscA) and forskolin/ IBMX–induced Isc changes (⌬IscF) were measured on a total of 239 airway epithelium cultures studied in Ussing chambers

(Figure 6B). All AEC were amplified in the presence of proliferating fibroblasts and differentiated either in the presence of differentiating fibroblasts or Vitrogen. Under both culture conditions, an inverse relationship between R and Isc was observed. Although R increased, Isc decreased with increasing passages (Figures 6A and 6B). At all AEC passages, the profile of ⌬IscA paralleled that of Isc, indicating high Na⫹ transport. Na⫹ transport was two to three times greater in airway epithelia cultured with fibroblasts compared with AEC grown on Vitrogen. Thus, AEC at passage 2 on differentiating fibroblasts exhibit similar bioelectric properties as freshly isolated AEC (P0) grown on Vitrogen. Under the latter conditions, ⌬IscA was virtually undetected in cultures from AEC at P3. Cl⫺ transport in response to cAMP elevation was maintained in epithelia at all AEC passages (P0–P3) when cultured in the presence of fibroblasts, whereas ⌬IscF decreased with passages for AEC seeded on Vitrogen,

Figure 6. Bioelectric properties of AEC grown at the air–liquid interface in the presence of Vitrogen or fibroblasts. (A ) Typical short-circuit recordings (Isc) from human AEC cultures at P0 (left panel) and P3 (right panel) in the presence of fibroblasts. The recordings illustrate the changes in Isc by sequential addition of amiloride, forskolin/IBMX, and bumetanide (arrows). The dashed lines indicate the zero current level. (B ) Quantitative analysis of the influence of airway cell passages (P0 to P3) on the resistance (R), Isc, amiloride-induced Isc changes (⌬IscA) and forskolin/IBMX-induced changes (⌬IscF), as measured in cultures grown at the air–liquid interface in the presence of fibroblasts (open bars) or Vitrogen (closed bars). Values are Mean ⫾ SEM of 10 and 18 measurements for cultures with or without fibroblasts at P0, of 26 and 12 at P1, of 13 and 5 at P2, and 16 and 4 at P3, respectively.


reaching 30% of its initial (P0) value for cultures at P3. To evaluate whether the decrease in Isc with AEC passage was caused by a change in the expression of ion channels, Western blot analysis for CFTR and the ␣ subunit of ENaC were performed. Although large variability was observed between airway cultures, no apparent relationship between the expression level of CFTR or ␣ENaC with increasing AEC passage was observed (Figure 7). Bioelectric properties of CF airway epithelia were also examined. As expected, amiloride decreased Isc current virtually to zero, whereas cAMP-elevating agents failed to increase Isc compared with non-CF cultures. Table 2 summarizes R, Isc, ⌬IscA, and ⌬IscF values measured in P2 and P3 cultures of non-CF and CF AEC grown at the air–liquid interface for ⬎ 30 d. Mucin and IL-8 Production by CF AEC Grown on Fibroblast Feeder Layers

Abnormalities in mucus properties and inflammation make a major contribution to the CF pathology. To test whether these defects are a direct result of CFTR mutations, we examined for mucin and IL-8 production by airway epithelia generated from CF AEC. The expression of mucin 5AC (MUC5AC), a human airway– specific mucin, was evaluated in our cell cultures. As shown in Figure 5D, MUC5AC was detected in some AEC–likely the mucous-secreting cells. The amount of mucous secreted by CF and non-CF airway epithelia was compared by dot-blot analysis of Alcian blue- and PAS-positive mucins. As shown in Figure 8A, acidic and neutral mucins from non-CF (n ⫽ 8) and CF (n ⫽ 30) airway epithelia were detected in similar amounts. Of note, CF airway cultures were characterized by the presence of sticky material at the apical surface. This can be appreciated in movies made on 6-mo-old non-CF and CF airway cultures (see NonCF.m1v and CF.m1v in the online supplement). Whereas beating of cilia with displacement of surface particles was clearly detected in non-CF cultures, the presence of thick mucous perturbed this observation in CF airway epithelia. Interestingly, this was correlated with enhanced (P ⬍ 0.01) airway liquid surface absorption of the CF epithelia (35.4 ⫾ 2.3 ␮l/d, n ⫽ 15) as compared with non-CF cultures (12.8 ⫾ 0.6 ␮l/d, n ⫽ 24). The production of the proinflammatory chemokine IL-8 was also evaluated in non-CF and CF airway epithelia. We observed that the release of IL-8 decreased with time after seeding, reaching steady-state levels within 10–14 d (data not shown). There-

Figure 7. Expression of CFTR and ␣ epithelial Na⫹ channel (ENaC) in airway cultures as a function of AEC passage. Western blots for CFTR (top panel) and ␣ENaC (bottom panels) were performed on total protein extracted from one airway epithelium culture generated from AEC at P0, P1, P2, or P3. A similar amount of CFTR or ␣ENaC was detected in these cultures, irrespective of AEC passage. Data are representative of at least three experiments.

45

fore, all IL-8 measurements were performed in airway epithelia that were maintained in culture for at least 2 wk. In non-CF cultures, the steady-state production of IL-8 was low, averaging 8.9 ⫾ 1.5 ng/ml (n ⫽ 15), and was not affected by the number of AEC passages (Figure 8B). In contrast, the production of IL-8 was significantly higher (17.3 ⫾ 4.5 ng/ml, n ⫽ 15) in CF respiratory epithelia generated from AEC at P0 and P1. Interestingly, this difference vanished for CF cultures generated from later AEC passages (Figure 8B). To determine whether long-term cultured airway epithelia were functional in terms of stimulus– secretion coupling, 90-d-old CF and non-CF respiratory epithelia generated from P0- to P3-airway cells were exposed to TNF-␣ for 2 h. Under these conditions, both non-CF and CF airway epithelia responded similarly to TNF-␣ by increasing their release of IL-8, which returned to basal levels within 4–6 d after treatment. The effect of TNF-␣ was observed in all cultures independently of the AEC passage (data not shown).

DISCUSSION We describe a novel procedure to stimulate proliferation and sustain differentiation of primary human AEC isolated from nasal polyps into functional mucociliated, pseudostratified respiratory epithelia. AEC from CF produced airway epithelia that retain typical features of the disease. The establishment of culture conditions that could improve proliferation of AEC without altering their differentiating abilities is essential for the generation of human airway cell models. The connective tissue is known to have a general supportive effect for the development of the overlying epithelium. Tissue fibroblasts can regulate the proliferation and differentiation of epithelial tissues, and have significant impact on cancer progression of adjacent epithelia (12, 14, 15, 17, 18). Fibroblast feeder layers have been shown to be applicable to AEC cultures. In most studies, commercialized embryonic mouse or human fetal lung fibroblasts were used. Although exhibiting good replicating abilities, AEC grown on these feeder layers rarely differentiated into pseudostratified epithelia (19, 20), even at the air–liquid interface (21). Fewer studies reported morphologic and functional features of reconstituted airway epithelia when both respiratory cells and fibroblasts were isolated from the same biopsy (22). Here, we show that subsets of primary airway fibroblasts are efficient in stimulating proliferation of non-CF and CF AEC, and a distinct population supported AEC differentiation into a pseudostratified epithelium. In agreement with previous reports, fibroblast-conditioned medium alone or in combination with complements was not sufficient to sustain AEC growth during successive passages (11, 19, 23). Studies on skin and airways have suggested that mesenchymal cells could stimulate epithelial cell growth and differentiation by elaborating a suitable biomatrix environment and by synthesis of diffusible factors (9, 11, 17). Therefore, the data indicate that bidirectional communications are established between fibroblasts and epithelial cells, thereby contributing to the regulation of airway cell growth or differentiation. The mechanisms underlying growth and differentiation of AEC by subpopulations of fibroblasts have not been investigated in this study. The long-term differentiation of airway cells could be achieved from subcultured AEC. In addition, AEC dissociated from well differentiated cultures could re-establish primary phenotype (data not shown). These observations suggest that progenitor-like cells are present in our airway cultures. Injury studies that target terminally differentiated cell populations have identified, on mouse models, the importance of the transit-amplifying (TA) progenitor cells in the rapid regeneration of a normal epithelium (24). It is therefore likely that TA cells with proliferating

46


2006

TABLE 2. BIOELECTRIC PROPERTIES OF AIRWAY EPITHELIA GENERATED FROM NON–CYSTIC FIBROSIS AND CYSTIC FIBROSIS AIRWAY EPITHELIAL CELLS

Non-CF CF

R (⍀.cm2)

Isc (␮A/cm2)

⌬IscA (␮A/cm2)

⌬IscF (␮A/cm2)

427 ⫾ 31 (78) 542 ⫾ 65 (26)

63 ⫾ 6 (78) 67 ⫾ 7 (26)

28 ⫾ 3 (77) 61 ⫾ 8 (25)

7 ⫾ 2 (75) 0.1 ⫾ 0.1 (25)

Definition of abbreviations: CF, cystic fibrosis; Isc, short-circuit current; ⌬IscA, amiloride-induced Isc changes; ⌬IscF, forskolin/ IBMX (3-isobutyl-1-methylxanthine)–induced Isc changes; R, transepithelial resistance. Values are mean ⫾ SEM of the number of measurements indicated in parenthesis. Data were obtained on ⬎ 30-d-old airway cultures generated from airway epithelial cells at P2 to P3 and grown in the presence of differentiating fibroblasts. CF airway epithelia exhibited no response to cAMP-elevating agents and increased amiloride-dependent inhibition of Isc.

and regenerating abilities were also present within the AEC population obtained from nasal polyps in the present study. To date, progenitor cells have been localized to submucosal gland ducts, to subsets of basal tracheal cells, and to neuroepithelial bodies of the mouse bronchiolar epithelium (24). Interestingly, our data indicate that the potential of human TA cells to grow or differentiate can be manipulated in vitro according to the airway origin of the feeder layers. These results are consistent with recent observations that the microenvironment may reprogram progenitor cells of one organ to repopulate and differentiate another organ (25). It is well established that exposure of the apical surface of airway cells to air is determinant for their differentiation into ciliated epithelia, although ciliogenesis is somehow limited in cells that have been subcultured (10, 26). Here, we show that

Figure 8. Mucin and IL-8 secretion by AEC grown at the air–liquid interface in the presence of fibroblasts. (A ) The presence of secreted mucins by CF and non-CF airway epithelia in airway surface liquid was determined by dot-blot analysis of Alcian blue- and PAS-positive mucins. Acidic and neutral mucins from non-CF (closed bars; n ⫽ 8) and CF (open bars; n ⫽ 30) airway epithelium equivalents were detected in similar amount. (B ) Steady-state production of IL-8 was measured in non-CF (closed circles) and CF (open circles) airway epithelia generated from AEC at P0, P1, P2, and P3. Symbols represent the production of IL-8 measured in separate experiments. Each value was obtained by averaging measurements performed at least in triplicate. Bars indicate median values.

distinct populations of fibroblasts support the differentiation of AEC that have been passaged up to three times. The morphology of these epithelia revealed a pseudostratified architecture with basal, ciliated, and secreting cells. After 30 d in culture, the percentage of ciliated and mucous-containing cells was ⵑ 15 times greater in airway equivalents grown in the presence of fibroblasts as compared with cultures initiated on Vitrogen coating. The airway epithelium was polarized, as evidenced by the detection of markers of intact airway epithelia (23). As expected, the tight junction protein occludin was detected apically at cell membrane contacts, whereas Na⫹, K⫹-ATPase was distributed in basolateral membranes. CFTR, a marker of ciliated airway cells, was localized in apical membranes, whereas AQP3, a protein expressed in basal cells (27), was detected in basal membranes. Finally, secreting cells were identified by expression of the airway-specific MUC5AC. In addition to the specific expression and membrane localization of these markers, the airway epithelia exhibited active transepithelial Na⫹ absorption and Cl⫺ secretion. Bioelectric properties were resolved in airway epithelia generated from AEC subcultured up to three times, with values comparable to those obtained in cultures initiated from freshly isolated AEC but grown in the absence of fibroblasts. Values for R, inhibition of amiloride-sensitive Na⫹ currents, and activation of cAMP-dependent Cl⫺ currents are in the range of already published values for human AEC in primary cultures (28–32). Although Isc decreased with increasing AEC passage, this observation was not associated with changes in the expression level of CFTR and ␣ENaC. These results point to the well differentiated architecture and function of the reconstituted epithelia established from P1- to P3-amplified AEC grown on a feeder layer of appropriate fibroblast clones. However, these properties could be maintained only exceptionally for later AEC passage, representing ⵑ 10% of the airway epithelia generated from AEC at passage 5. We hypothesize that Isc decreases reflect a change in the metabolic state or homeostasis of the airway epithelium with AEC subculture. Similarly, we generated respiratory epithelia from freshly isolated and passaged AEC obtained from patients with CF. The CF airway epithelia retained typical features of the airway disease, including absence of cAMP-dependent Cl⫺ transport. Interestingly, amiloride abolished the Na⫹ component of the Isc current in these cultures, whereas it only partially affected the Na⫹ current in non-CF airway epithelia. It has been proposed that increased Na⫹ absorption and decreased Cl⫺ secretion result in lowering the periciliary liquid layer and raising mucus viscosity (6). Consistent with this hypothesis, we observed that the rate of apical surface liquid absorption was 2.8 times larger in CF airway epithelia as compared with non-CF cultures. This was associated with the presence of dense material at the surface of the CF epithelia, which impaired clearance of particles normally observed in non-CF airway epithelia. Although the decreased clearance of airway surface particles may be caused by increased


mucin content of the CF mucus, this possibility appears unlikely, as no difference in the amount of secreted acid or neutral mucins could be detected in our CF or non-CF cultures (33). Therefore, the results indicate that a primary defect of CF airway epithelia is an increased rate of apical liquid absorption, an observation that is in agreement with the “low volume” hypothesis originally proposed by Boucher and collaborators (8). The airway epithelium plays a pivotal role in mediating the innate and adaptive immune response by secreting chemotactic factors for leukocytes and lymphocytes. There is evidence for aberrant production of various chemotactic factors by the CF airway epithelium, including IL-8 (34). We observed that IL-8 secretion was indeed increased in CF respiratory epithelia generated from P0 to P1 AEC as compared with non-CF cultures. This difference, however, vanished with higher AEC passages. This observation, which is in agreement with findings recently reported for P2-human tracheobronchial epithelial cell cultures (35), was not due to decreased ability of subcultured cells to secrete IL-8. Indeed, all airway epithelia of CF and non-CF origin responded to TNF-␣ independently of AEC passage. It is also possible that mutant CFTR is being processed more efficiently in later-passage CF cultures, thereby reducing inflammatory signaling. This possibility, however, appears unlikely, as we could detect neither partial correction of the cAMP-dependent membrane currents nor the presence of apical CFTR by immunofluorescence analysis of later-passage CF cultures. Alternatively, the CF airway epithelium may adapt to the proinflammatory environment by increasing its metabolic activity, a feature that is retained in cultures generated from early AEC passage. The results suggest, however, that there is no intrinsic CFTR-dependent anomaly of IL-8 secretion by CF AEC. In summary, the model described here presents several important advantages as compared with previously reported methods (35, 36). First, it allows stronger amplification of the AEC pool that is available after tissue digestion of biopsies. This will be advantageous for the study of pathogenesis of airway diseases, where materials from patients are limited (31, 32). Second, the amplified AEC can differentiate to respiratory epithelia at the air–liquid interface, allowing the generation of a large amount of culture that displays histologic and biochemical characteristics similar to those observed in vivo. Usually, the present method allows the generation of about 100 0.33-cm2 Transwell membranes from a tissue digestion that yields 500,000 AEC. All epithelia showed features of differentiated cells, including mucous and ciliated cells, ion transport, and barrier function. In contrast to previous reports (37), CF AEC cultures failed to show increased production of mucins and IL-8, but exhibited enhanced ASL absorption. This human airway cell model may ease phenotypic analysis of defined gene mutations, evaluation of bioelectric properties of airway epithelia reconstituted from various regions of the respiratory tract, and longitudinal studies of pharmacologic or genetic treatments. It may also represent a valuable tool for studies aimed at understanding the complex processes regulating airway epithelium renewal and repair, as well as for identifying unique markers for progenitor cells in airways. Conflict of Interest Statement : None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Acknowledgments : The authors thank Drs. Brenda Kwak, Song Huang, and Constance Barazzone for critical reading of the manuscript. The technical help of Assunta Caruso, Raphael Guanella, Philippe Henchoz, and Isabelle Scerri is acknowledged.

References 1. Knowles MR, Boucher RC. Mucus clearance as a primary innate defense mechanism for mammalian airways. J Clin Invest 2002;109:571–577.

47 2. Travis SM, Singh PK, Welsh MJ. Antimicrobial peptides and proteins in the innate defense of the airway surface. Curr Opin Immunol 2001; 13:89–95. 3. Whitsett JA. Intrinsic and innate defenses in the lung: intersection of pathways regulating lung morphogenesis, host defense, and repair. J Clin Invest 2002;109:565–569. 4. Ratjen F, Do¨ring G. Cystic fibrosis. Lancet 2003;361:681–689. 5. Verkman AS, Song Y, Thiagarajah JR. Role of airway surface liquid and submucosal glands in cystic fibrosis lung disease. Am J Physiol Cell Physiol 2003;284:C2–C15. 6. Boucher RC. New concepts of the pathogenesis of cystic fibrosis lung disease. Eur Respir J 2004;23:146–158. 7. Grubb BR, Boucher RC. Pathophysiology of gene-targeted mouse models for cystic fibrosis. Physiol Rev 1999;79:S193–S214. 8. Matsui H, Grubb BR, Tarran R, Randell SH, Gatzy JT, Davis CW, Boucher RC. Evidence for periciliary liquid layer depletion, not abnormal ion composition, in the pathogenesis of cystic fibrosis airways disease. Cell 1998;95:1005–1015. 9. Zabner J, Smith JJ, Karp PH, Widdicombe JH, Welsh MJ. Loss of CFTR chloride channels alters salt absorption by cystic fibrosis airway epithelia in vitro. Mol Cell 1998;2:397–403. 10. Gruenert DC, Basbaum CB, Widdicombe JH. Long-term culture of normal and cystic fibrosis epithelial cells grown under serum-free conditions. In Vitro Cell Dev Biol 1990;26:411–418. 11. Gruenert DC, Finkbeiner WE, Widdicombe JH. Culture and transformation of human airway epithelial cells. Am J Physiol 1995;268:L347– L360. 12. Burwinkel RP, Gruenert DC, Upshaw RT, Infeld MD. Promoting matrix invasion by human lung epithelial cells. In Vitro Cell Dev Biol Anim 1998;34:22–24. 13. Karp PH, Moninger TO, Weber SP, Nesselhauf TS, Launspach JL, Zabner J, Welsh MJ. An in vitro model of differentiated human airway epithelia: methods for establishing primary cultures. Methods Mol Biol 2002;188:115–137. 14. Hogan BLM. Morphogenesis. Cell 1999;96:225–233. 15. Wartburton D, Lee MK. Current concept on lung development. Curr Opin Pediatr 1999;11:188–192. 16. Limat A, Mauri D, Hunziker T. Successful treatment of chronic leg ulcers with epidermal equivalents generated from cultured autologous outer root sheath cells. J Invest Dermatol 1996;107:128–135. 17. Werner S, Smola H. Paracrine regulation of keratinocyte proliferation and differentitation. Trends Cell Biol 2001;11:143–146. 18. Bhowmick NA, Chytil A, Plieth D, Gorska AE, Dumont N, Shappell S, Kay Washington M, Neilson EG, Moses HL. TGF-␤ signaling in fibroblasts modulates the oncogenic potential of adjacent epithelia. Science 2004;303:848–851. 19. Kitamura H, Shibagaki T, Inayama Y, Ito T, Kanisawa M. Growth and differentiation of human distal airway epithelial cells in culture: effects of small amounts of serum in defined medium. Lab Invest 1990;63:420– 428. 20. Gray TE, Thomassen DG, Mass MJ, Barrett JC. Quantitation of cell proliferation, colony formation, and carcinogen induced cytotoxicity of rat tracheal epithelial cells grown in culture on 3T3 feeder layers. In Vitro 1983;19:559–570. 21. Choe M, Sporn PH, Swartz MA. An in vitro airway wall model of remodeling. Am J Physiol Lung Cell Mol Physiol 2003;285:L427–L433. 22. Goulet F, Boulet LP, Chakir J, Tremblay N, Dube J, Laviolette M, Boutet M, Xu W, Germain L, Auger FA. Morphologic and functional properties of bronchial cells isolated from normal and asthmatic subjects. Am J Respir Cell Mol Biol 1996;15:312–318. 23. You Y, Richer EJ, Huang T, Brody SL. Growth and differentiation of mouse tracheal epithelial cells: selection of a proliferative population. Am J Physiol Lung Cell Mol Physiol 2002;283:L1315–L1321. 24. Neuringer IP, Randell SH. Stem cells and repair of lung injuries. Respir Res 2004;5:6–15. 25. Blau HM, Brazelton TR, Weimann JM. The evolving concept of a stem cell: entity or function? Cell 2001;105:829–841. 26. de Jong PM, van Sterkenburg MA, Hesseling SC, Kempenaar JA, Mulder AA, Mommaas AM, Dijkman JH, Ponec M. Ciliogenesis in human bronchial epithelial cells cultured at the air–liquid interface. Am J Respir Cell Mol Biol 1994;10:271–277. 27. Kreda SM, Fynn MC, Fenstermacher DA, Boucher RC, Gabriel SE. Expression and localization of epithelial aquaporins in the adult human lung. Am J Respir Cell Mol Biol 2001;24:224–234. 28. Boucher RC, Stutts MF, Knowles MR, Cantley L, Gatzy JT. Na⫹ transport in cystic fibrosis respiratory epithelia: abnormal basal rate and

48

29. 30.

31.

32.

33.

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 34 response to adenylate cyclase activation. J Clin Invest 1986;78:1245– 1252. Smith JJ, Welsh MJ. Fluid and electrolyte transport by cultured human epithelia. J Clin Invest 1993;91:1590–1597. Mall M, Wissner A, Gonska T, Calenborn D, Kuehr J, Brandis M, Kunzelmann K. Inhibition of amiloride-sensitive epithelial Na⫹ absorption by extracellular nucleotides in human normal and cystic fibrosis airways. Am J Respir Cell Mol Biol 2000;23:755–761. Blouquit S, Morel H, Hinnrasky J, Naline E, Puchelle E, Chinet T. Characterization of ion and fluid transport in human bronchioles. Am J Respir Cell Mol Biol 2002;27:503–510. Lopez-Souza N, Avila PC, Widdicombe JH. Polarized cultures of human airway epithelium from nasal scrapings and bronchial brushings. In Vitro Cell Dev Biol Anim 2003;39:266–269. Leir SH, Parry S, Palmai-Pallag T, Evans J, Morris HR, Dell A, Harris A.

34.

35.

36.

37.

2006

Mucin glycosylation and sulphation in airway epithelial cells is not influenced by cystic fibrosis transmembrane conductance regulator expression. Am J Respir Cell Mol Biol 2005;32:453–461. Bonfield TL, Konstan MW, Berger M. Altered respiratory epithelial cell cytokine production in cystic fibrosis. J Allergy Clin Immunol 1999;104:72–78. Becker MN, Sauer MS, Muhlebach MS, Hirsh AJ, Wu Q, Verghese MW, Randell SH. Cytokine secretion by cystic fibrosis airway epithelial cells. Am J Respir Crit Care Med 2004;169:645–653. Randell SH, Walstad L, Schwab EE, Grubb BR, Yankaskas JR. Isolation and culture of airway epithelial cells from chronically infected human lungs. In Vitro Cell Dev Biol Anim 2001;37:480–489. Sajjan U, Keshayjee S, Forstner J. Responses of well-differentiated airway epithelial cell cultures from healthy donors and patients with cystic fibrosis to Burkholderia cenocepacia infection. Infect Immun 2004;72:4188–4199.