Characterization of pediatric cystic fibrosis airway epithelial cell cultures at the air-liquid interface obtained by non-invasive nasal cytology brush sampling

Fifteen pediatric CF patients were recruited at the Children’s University Hospitals of Bern. The clinical characteristics of the CF patients are shown in Table 1. Seven healthy human volunteers were recruited with the following exclusion criteria: recurrent nasal bleeding, therapy with anticoagulants, and a current respiratory tract infection. The ethics committee of the Canton of Bern, Switzerland approved the study (authorization KEK/09) and informed consent was obtained from all patients/volunteers.

Primary nasal epithelial cells were obtained by nasal brushing as previously described [11, 15, 16]. Briefly, the nasal epithelial cells were obtained by brushing the inferior surface of the middle turbinate of both nostrils twice by using cytology brushes (Dent-o-care, London, UK). Next, the freshly brushed tissue was seeded in collagen-coated (Advanced BioMatrix Inc., San Diego, CA, USA) 12.5 cm² cell culture flasks (BD Bioscience, USA) in Bronchial Epithelial Growth Medium (BEGM, Lonza, Switzerland) supplemented with Single Quots (Lonza, Switzerland) and Primocin (100 μg/ml, InvivoGen, US) in a humidified incubator at 37 °C. The CF cells were additionally treated with Amphotericin B (250 μg/ml; Sigma Aldrich, US) and Ceftazidime (100 μg/ml, GlaxoSmithKline, Switzerland) during five days after sampling [16]. Out of 15 CF patients brushed, 5 cultures were lost due to poor cell growth during the expansion phase of the cultures. The 10 CF patients from which successful ALI cultures could be established and used for further investigation are presented in Table 1. We obtained 0.4 to 1.5 million viable cells per CF patient after brushing and the time to reach confluence during the expansion phase was 7 to 15 days. The time to confluence during the expansion phase was directly dependent on the number of cells obtained after brushing. The success rate of culture establishment for healthy donors was higher compare to CF patients (7 healthy donors brushed resulted in 6 cultures successfully established). Of note, when growing on the inserts, we obtained a 100% success in differentiation of the cultures. Once confluent under submerged condition, the cells where seeded at a density of 60,000 cells per insert onto 24-well inserts with a pore size of 0.4 μm at 37 °C, 5% CO₂ (Greiner Bio-One, Austria). Cells were grown on the insert membranes under submerged conditions by adding 200 μl of BEGM apically and 450 μl of BEGM in the basal chamber until they reached confluence (2–3 days post seeding). Cell cultures were then washed with phosphate buffered saline (PBS) 1X w/o Ca²⁺ and Mg²⁺ and culture medium was then changed to complete PneumaCult™-ALI Medium (Stemcell Technologies, CA) following the manufacturer’s instruction. Briefly, cell cultures were exposed to air on the apical side and provided with 450 μl of complete PneumaCult™-ALI Medium in the basal compartment to promote mucociliary differentiation. The basal medium was changed every second day. Twenty-one to 28 days post exposing the cells to air at 37 °C, 5% CO₂, fully differentiated primary CF airway epithelial cell cultures were obtained and showed evidence of mucus production and ciliary beating. According to the supplier’s specifications, the differentiated CF nasal epithelial cells were therefore used at 21 to 28 days post exposure to air for further experiments. The control bronchial ALI cultures form 2 different donors used for the evaluation of CFTR expression by confocal microscopy were obtained commercially (Epithelix Sàrl, Geneva, Switzerland).

Cell culture membranes were cut from the insert and fixed for 20 min in 70% ethanol. Samples were then stained with hematoxylin and eosin (H&E) and snap-frozen in Tissue-Tek O.C.T. Compound (Sakura Finetek USA Inc., US). Cryosections of 7 μm were cut (30,505 Cryostat, Leica Microsystems, Heerbrugg, Switzerland) and examined by brightfield microscopy.

Cells on membrane cut from the insert were fixed in 70% ethanol for 20 min and were permeabilized with 0.2% Triton X-100 (Sigma-Aldrich, US) for 15 min and washed with staining buffer (PBS containing 0.1% BSA and 0.001% NaN₃ (Sigma Aldrich, US) for 5 min at room temperature. The cells were then incubated with the primary antibody or dye in staining buffer for one hour at room temperature. Cells were stained with Phalloidin Rhodamine (1:100; Invitrogen, Switzerland), rabbit anti-MUC5AC (clone H-160, 1:100, Santa Cruz Biotechnologies, US) mouse anti-β-tubulin (clone 2 28 33, 1:100, Invitrogen, Switzerland), rabbit anti-p63 (clone EPR5701, 1:60, Abcam, UK), rabbit anti-ZO-1 (Zona occludens-1; ab59720, 1:50, Abcam, UK), and mouse anti-CFTR (clone 24–1, 1:100, R&D Systems, UK). After washing with staining buffer, the cells were incubated for one hour with the secondary antibody in staining buffer at room temperature: anti-mouse Alexa Fluor 488 (Invitrogen, US) and anti-rabbit Alexa Fluor 647 (Invitrogen, US). Cells were washed with staining buffer and then mounted in DAPI mounting media (4′,6-diamidino-2-phenylindole; Vector Laboratories Inc., US). A Zeiss Laser scanning microscope (LSM) 710 (Carl Zeiss AG, Switzerland) was used to take the optical sections and image processing was done using Imaris software (Bitplane AG, Switzerland).

Membranes with fully differentiated CF nasal epithelial cells were cut from inserts and were fixed in buffered 2.5% glutaraldehyde (Agar Scientific Ltd., Plano GmbH, Wetzlar, Germany) and post-fixed in buffered 1.0% osmium tetroxide (SPI Supplies, West Chester, USA). Preparations were then dehydrated in graded ethanol (Alcosuisse, Switzerland) and embedded in Epon (Fluka, Buchs, Switzerland) and left to harden at 60 °C for 5 days. Ultrathin sections (70–80 nm) were produced using an ultramicrotome UC6 (Leica Microsystems, Vienna, Austria) and mounted on single slot copper grids and subsequently stained with uranyl acetate and lead citrate utilizing an ultrostainer (Leica Microsystems, Vienna, Austria). Ultrathin sections were examined with a transmission electron microscope (TEM; CM12, Philips, Eindhoven, Netherlands).

Membranes were cut from inserts and placed on a glass slide mounted with culture medium and covered with a coverslip. Good contrast region of the differentiated cell culture with beating cilia was first selected using the OLYMPUS Vanox-S microscope with SPLAN Apo objectives (10X, 20X, 40X) with differential interference contrast (DIC/Nomarski). By means of a Point Grey Flea 3 (FL3-U3-13Y3M-C) B/W high speed CMOS camera (Point Grey Research, Inc., Canada) that was connected to the microscopy system, image series of beating cilia were taken serially at 300 frames per second. To perform ciliary beat frequency, one ALI insert per CF patient was used and 20 different spots per insert where evaluated (10 near the border and 10 in the center). The measurement where performed under constant temperature of 30 °C and to minimize any bias, the measurements where performed in a minimal amount of time. Image processing was done using the Image J software (US) and ciliary beating frequency (CBF) was analyzed by means of the power spectrum based on Fast Fourier Transform algorithm.

Cell layer integrity was measured in submerged cell cultures grown on inserts and in cell cultures grown on inserts at the ALI for one day to 24 days (fully differentiated) as follow: complete PneumaCult™-ALI Medium was removed from the basal chamber of CF nasal epithelial cells grown on 24-well inserts and cells were washed with PBS (Life Technologies, US). Two mg/ml of 4 kDa Fluorescein isothiocyanate-dextran (FD4, Sigma-Aldrich, US) was diluted in 150 μl complete PneumaCult™-ALI Medium and added apically to the cell cultures. The basal chamber was filled with fresh complete PneumaCult™-ALI Medium and cultures were then incubated for 4 h at 37 °C and 5% CO₂. The fluorescence intensity of the contents of the basal chamber was determined with the following wavelengths: excitation 485 nm, emission 544 nm. Of note, a 24-well insert only was used as reference value and percentage of fluorescence was expressed relative to this control.

Fully differentiated CF nasal epithelial cell cultures were washed with PBS and fresh complete PneumaCult™-ALI Medium was added to the apical and basal chamber for equilibration for 30 min at room temperature. Trans-epithelial electrical resistance (TEER) was then assessed three times with an EVOM meter with the STX2 “chopstick” electrodes of 4 mm width and 1 mm thick (World Precision Instruments, FL) and a mean resistance was calculated. Values were then corrected for fluid resistance (insert with no cells) and surface area.

Total RNA harvesting was performed on inserts at the ALI for 24 days with the RNA II kit (Macherey-Nagel, Oensingen, Switzerland) and complementary deoxyribonucleic acid (cDNA) synthesis was done with Omniscript RT Kit (Qiagen, Valencia, CA, USA). RT-PCR measurements were performed with the Taqman Fast Universal PCR Master Mix (Applied Biosystems, Foster City, CA, USA) on a 7500 Fast Real-Time PCR System (Applied Biosystems). Sequences of the primers and probe used for the detection of the CFTR transcript: forward primer: 5’-AGCTGTCAAGCCGTGTTCTAGATA-3′, reverse primer 5’-ATGAGGAGTGCCACTTGCAAA-3′, probe 5’-FAM-CACACGAAATGTGCCAATGCAAGTCCTT-TAMRA-3′. The CFTR mRNA levels where standardized to the housekeeping ribosomal 18S RNA by using the following primers/probe set: forward primer 5’-CGCCGCTAGAGGTGAAATTCT-3′, reverse primer 5’-CATTCTTGGCAAATGCTTTCG-3′, probe 5’-FAM-ACCGGCGCAAGACGGACCAGA-TAMRA-3′.

Basolateral protein levels of interleukin (IL)-6 and IL-8 were measured with the DuoSet ELISA Development kit (R&D, Minneapolis, MN, USA). In our experience, more reproducible cytokine levels in the basolateral vs. apical compartment are obtained due to the difficult-to-standardize apical washing procedure. Also, this approach prevents the perturbation of the epithelium. In line with our experimental approach, sampling of the basolateral compartment for the evaluation of IL-6 and IL-8 levels produced by airway epithelial ALI cultures is often selected [6, 10, 17‐19].

Data are presented as mean ± SEM. An unpaired t-test was used to determine differences between two groups. To conduct the statistical analysis, GraphPad Prism 5 software (GraphPad Software Inc., US) was used. A p value <0.05 was considered statistically significant.

Fully differentiated CF nasal epithelial cells (3 to 4 weeks post exposure to air) were fixed and stained with H&E to determine the structure of the differentiated epithelium. The stainings showed the structure of a pseudostratified epithelium with a ciliary brush border (Fig. 1a, b) indicative of a fully differentiated CF airway epithelium. Also, we noticed a productive secretion of mucus (see Additional file 1).

To investigate if the three main cell types of the native respiratory epithelium (ciliated, mucus-secreting and basal cells) were present in the differentiated CF nasal epithelial cell cultures, we analyzed immunofluorescence pictures at 3 to 4 weeks post exposure of cell cultures to air. The immunofluorescence pictures revealed fully differentiated CF cultures positive for ciliated cells (β-tubulin, green), mucus-secreting cells (MUC5AC, white) (Fig. 2a, b) and basal cells (p63, white) (Fig. 2c). Further a positive staining for ZO-1 (tight junction protein, white) suggested a tight, intact epithelial layer of the fully differentiated CF cell cultures (Fig. 2d).

Examination of the fully differentiated CF nasal epithelial cell cultures by TEM confirmed the pseudostratified organization of the epithelium and the mucociliary phenotype. Indeed, mucus-secreting cells with secretory granules and ciliated cells at the apical site and basal cells could be identified (Fig. 3a-d). Typical apical tight junctional complexes were observed between adjacent cells in the epithelium (Fig. 3e) and the cilia of ciliated cells showed the typical axonemal ultrastructure with its characteristic arrangement of microtubules and dynein arms (Fig. 3f).

Evaluation of the CBF is an important and reliable method to assess the physiological condition of the respiratory epithelium. It is also an essential tool of clinical relevance in CF since the functional and coordinated ciliary beat is a pivotal component of the mucociliary escalator, which is involved in the development and progression of CF lung disease [20‐22]. To investigate whether the ciliated cells showed a ciliary beat, the CBF of fully differentiated CF cell cultures established from three different CF patients was determined at 4 weeks post exposure to air. Three fully differentiated CF nasal epithelial cell cultures generated from 3 different CF patients revealed an average CBF of 9.67 ± 0.44 Hz. A recording of the ciliary beat of the CF cultures grown at the ALI is presented in Additional file 2.

In order to determine if the cell layer of well differentiated CF cells was functionally tight, we performed an epithelial cell layer integrity test in submerged CF cultures and at 1–24 days post exposure to air by apically applying FD-4 on CF cultures and assessed the fluorescence intensity in the basal chamber. We found that the relative fluorescence intensity was significantly ca. 10 times lower already one day post exposure to air in comparison to submerged cultures, indicative of a tight epithelium (p < 0.05; Fig. 4). We performed preliminary measurements of TEER in fully differentiated cultures at week 4 post-exposure to air and found values in the range of other investigators such as Sajjan et al. at 500 Ωcm² [4] and Wiszniewski et al. at 542 Ωcm² [5], which is suggesting tight junction formation.

We characterized further the nasal CF ALI cultures by measuring the CFTR transcript levels in CF compare to control ALI cultures. We obtained similar CFTR mRNA levels in both cultures (Fig. 5a). Of note, the CFTR levels of the 5 CF ALI cultures tested where originating from CF patients with a F508del/ F508del genotype. Several in vitro studies support the idea of dysregulated inflammatory response in CF airway cells. In this context, IL-8 and IL-6 are widely used to assess the inflammatory state in CF cells and to measure the inflammatory response of CF cells upon infection [15, 18, 23]. Also, primary nasal epithelial cells produce significant levels of those cytokine under baseline condition [14, 24]. Thus, we measured the baseline levels of IL-6 and IL-8 cytokines produced basolateraly by CF in comparison to control ALI cultures. We didn’t found any significant differences between control and CF ALI cultures in terms of IL-6 (Fig. 5b), and IL-8 (Fig. 5c) baseline production. Finally, we assessed CFTR protein expression levels by immunocytochemistry and confocal imaging in CF nasal (Fig. 5d) in comparison to control (Fig. 5e). Since bronchial cells are the gold standard in the field, we used bronchial healthy ALI cultures as a reference (Fig. 5f). Of note the representative micrographs of CF nasal ALI cultures presented in Fig. 5d/g were originating from patients with a F508del/F508del genotype. We didn’t observe a notable difference in the CFTR signal/level between the different cultures tested (Fig. 5g-i).