Successful establishment of primary small airway cell cultures in human lung transplantation

Background The study of small airway diseases such as post-transplant bronchiolitis obliterans syndrome (BOS) is hampered by the difficulty in assessing peripheral airway function either physiologically or directly. Our aims were to develop robust methods for sampling small airway epithelial cells (SAEC) and to establish submerged SAEC cultures for downstream experimentation. Methods SAEC were obtained at 62 post-transplant bronchoscopies in 26 patients using radiologically guided bronchial brushings. Submerged cell cultures were established and SAEC lineage was confirmed using expression of clara cell secretory protein (CCSP). Results The cell yield for SAEC (0.956 ± 0.063 × 106) was lower than for large airway cells (1.306 ± 0.077 × 106) but did not significantly impact on the culture establishment rate (79.0 ± 5.2% vs. 83.8 ± 4.7% p = 0.49). The presence of BOS significantly compromised culture success (independent of cell yield) for SAEC (odds ratio (95%CI) 0.067 (0.01-0.40)) but not LAEC (0.3 (0.05-1.9)). Established cultures were successfully passaged and expanded. Conclusion Primary SAEC can be successfully obtained from human lung transplant recipients and maintained in culture for downstream experimentation. This technique will facilitate the development of primary in vitro models for BOS and other diseases with a small airway component such as asthma, cystic fibrosis and COPD.


Background
Although lung transplantation is a well-accepted therapeutic option for selected patients with advanced lung disease, long-term survival is limited largely by progressive and treatment refractory airflow limitation manifest clini-cally as bronchiolitis obliterans syndrome (BOS) [1]. The predominant histopathologic finding in patients with BOS is of fibro-proliferative small airway obliteration (obliterative bronchiolitis (OB)). Unfortunately, there has been no substantial improvement in the reported inci-dence of BOS over the last twenty years, despite improvements in immunosuppression, surgical techniques, and patient management [2,3]. Recognition of the central role of OB in limiting post-transplant survival has led to a large body of research aimed it improving management [4][5][6][7]. However, all in vitro human studies have used large airway epithelial cells (LAEC) despite OB being predominantly a small airway disease. The aim of this study was to develop methodology for the successful sampling and culture of small airway epithelial cells (SAEC) obtained from lung transplant patients at routine post-transplant bronchoscopy. The described techniques will provide a more relevant in vitro human cell based model to study the pathogenesis of OB.

Patients
A total of 62 bronchoscopies were performed in 26 patients (11 female; aged 18 to 64 years (median 51 years); 4 BOS). Patient demographics are summarized in Table 1. BOS was diagnosed and graded according to international guidelines [2]. The study was approved by the Royal Perth Hospital Human Research and Ethics Committee.

Bronchoscopy procedure
Human airway epithelial cells (AEC) were collected using a bronchial brush during routine surveillance and diagnostic post-transplant bronchoscopies. Bronchoscopy was conducted under general anaesthesia with a laryngeal mask or endotracheal tube. The bronchoscope(Olympus ® Evis EXERA II) was wedged in a suitable lateral segment of the right or left lower lobe. Prior to the acquisition of transbronchial biopsies, the sheathed nylon cytology brush (10 mm, 2 mm outer diameter, Olympus BC-25105, Waverley, Australia) was passed down the working  (Fig. 1).
Large airways brushings (*2-3) were obtained from segmental bronchi in the standard way. In both cases, brushings were collected into a tube containing 2 ml of RPMI on ice and the brush tip was also cut off and collected after the final brushing. After the completion of brushings, 20% (v/v) FCS was added to the tubes and processed immediately. Tubes and brushes were kept separate for large and small airways to prevent cross contamination. The lung apex was screened to exclude a pneumothorax as part of routine care following a transbronchial biopsy.

Establishment of cultures
During processing, the cells were fractionated for RNA archiving, cytospins and cell culture as previously described [8]. Successful culture establishment was assessed as previously described [8] and cultures were passaged at 90% confluence. Additionally, a human bronchial epithelial cell line (16HBE14o -; provided by Dieter Gruenert, University of California San Francisco, USA) was also utilised and maintained as previously described [8].

Small airway epithelial lineage verification
When cells reached 90% confluence in culture, they were trypsinised, collected and resuspended in 1 ml of RPMI. The cell suspension was then fractionated and a 350 μl aliquot was centrifuged and pelleted cells stored in RLT buffer (QIAGEN, Hamburg, Germany). RNA was extracted using QIAGEN RNA Easy Mini Kit. The remain- der of the cell suspension was seeded into pre-coated flasks to continue propagation of the culture. Lineage was verified by analysing mRNA from a representative group of cultures chosen at random from patients who were not diagnosed with BOS. Quantitative polymerase chain reaction (qPCR) was used to assess expression of the Clara Cell Secretory Protein (CCSP), which is uniquely expressed by small airway epithelial cells, and surfactant protein B (SP-B), which is commonly expressed by nonciliated bronchial epithelial cells [9] and type II alveolar cells [10]. Lineage verification was carried out on mRNA from cultures at p0 and p2. Gene expression was analyzed using two-step reverse transcription polymerase chain reaction (RT-PCR) and cDNA synthesized using hexanucleotide primers and Multiscribe™ Reverse

Results
Brushings were successfully obtained from both the small and large airways of the transplanted lung at all 62 bronchoscopies. The brushing method was well tolerated by all patients with no significant bleeding, pneumothorax or other adverse events being observed. The mean cell yield from the allograft was significantly higher for LAEC (1.306 ± 0.077 × 10 6 ) than for SAEC (0.956 ± 0.063 × 10 6 ; p < 0.01). No significant difference in yield was noted between BOS and non-BOS patients (Fig. 2).

Cell culture establishment
Cell cultures were successfully established from both large and small airway brushings with a similar success rate (83.8 ± 4.7% and 79.0 ± 5.2%; p = 0.49 respectively). Established cultures reached confluence within a median 21 days (range 13 -57 days) and maintained a polygonal, cobblestone appearance, typical of epithelial cells. No major morphological variations were observed between LAEC and SAEC over the life of the culture (Fig 3). Immunocytochemistry conducted on cells from passage 0 (p0) and passage 2 (p2) with epithelial, and mesenchymal markers confirmed the preservation of epithelial lineage of the cells.
Culture fates are presented in Table 2. Successful establishment was limited predominantly by superinfection by organisms colonizing the transplanted organ (2 patients, A. fumigatus and S. aureus) and low cell yield. The latter problem was confined to SAEC -failure of five of the cultures could be attributed to low cell yield and they were all from small airway brushings. The mean cell yield for the five failed cultures was 0.326 ± 0.055 × 10 6 cells, which is significantly lower than the cell yield for successful cultures (1.071 ± 0.070 × 10 6 cells; p < 0.01). The presence of BOS significantly compromised culture success for SAEC (odds ratio (95%CI) 0.067 (0.01-0.40)) but not for LAEC (0.3 (0.05-1.9)). Since the cell yield was not different for BOS SAEC, poor culture establishment did not appear to be related to low starting cell numbers.

Epithelial lineage verification
Morphological analysis of established cultures over repetitive passage showed that the typical cobblestone morphology indicative of epithelial cells was maintained over culture duration. Epithelial lineage was further verified at each passage via immunocytochemical staining. Cultured SAEC and LAEC stained intensely and exclusively for the epithelial specific marker, AE1-AE3 (cytokeratin) at both p0 and p2. No expression was observed for mesenchymal (Vimentin), macrophage (CD68), dendritic (CD1a) or endothelial (Von Willebrand Factor) lineage markers at either passage (Fig. 4).

Small airway epithelial lineage verification
Lineage verification of established cultures was then assessed using known and suggested markers of small airway epithelium [11,12]. Here, we assessed small airway gene expression of CCSP and SP-B using qPCR on RNA extracted from cultures at p0 and p2. CCSP was exclusively expressed in SAEC (4592 ± 743.4 fold normalized to 18 s at p0; 7148 ± 5385 fold normalized to 18 s at p2) compared to LAEC (11.56 ± 9.113 fold normalized to 18 s at p0, p = 0.0001; 235 ± 275.6 normalized to 18 s at p2, p = 0.0113). CCSP was not expressed in a LAEC immortalized cell line (16HBE14ocells (Fig. 5A &5B)). To exclude an alveolar source for the small airway brush cellular material, SP-B gene expression was also assessed in large and small airway cell cultures at p0 and p2. SP-B was expressed only at low levels in both SAEC and LAEC at p0 and p2. There was no difference between SAEC and LAEC Figure 3 Morphology of epithelial cells is maintained over passage. Phase contrast micrographs showing no morphological variation in bronchial epithelial cells cultured from small airway (SAEC) and large airway (LAEC) of a lung allograft, obtained during routine bronchoscopy. All cells exhibited a cobblestone morphology which was maintained over two consecutive passages (p0 to p2).

Discussion
We have developed a method for successfully collecting and establishing expandable primary cultures from human SAEC obtained bronchoscopically. The method was well tolerated and easy to perform. The cell yield using this collection method was lower in small airway brushings than the large airway brushings however this did not significantly compromise the culture establishment rate. Both LAEC and SAEC maintained their lineage over passage.
The inability to establish a suitable in vitro model using primary human cells has been a major impediment to research into post-transplant chronic allograft dysfunction. The most relevant in vitro work has been conducted on human LAEC despite BOS being a disease of small airways [7,13]. It is highly probable that in vitro work in LAEC can not be neatly extrapolated to SAEC. The described methods will facilitate the development of more relevant in vitro models not only for OB, but also for other diseases with small airway pathology. In the case of transplantation for instance, OB is the result of a range of alloreactive, infective and non-specific insults and recent evidence suggests that transforming growth factor β (TGFβ 1 ) driven epithelial mesenchymal transition (EMT) is the final common pathway to airway obstruction and fibrosis [14]. Using the model described herein, EMT can be induced by TGF-β 1 in vitro, with the ability to assess candidate compounds for therapeutic efficacy.
Several investigators have successfully established LAEC cultures from bronchial brushings, which have been used to study a wide range of diseases including OB [7,13], asthma [8], cystic fibrosis [15] and COPD [16]. The present study has extended these methods to SAEC. The LAEC collection and extraction methods described here are very similar to those reported by Forrest et. al. [17].
*Y -Successful Culture *N α -Unsuccessful culture due to superinfection with A. fumigatus *N β -Unsuccessful culture due to superinfection with S. aureus *N γ -Unsuccessful culture due to other causes *N δ = Unsuccessful culture due to low cell yield Characterisation of established epithelial cell cultures Figure 4 Characterisation of established epithelial cell cultures. Cytospins obtained from cells cultured from large airway bronchial brushing (LAEC; Fig. 4A) and small airway brushings (SAEC; Fig. 4B) from representative lung allograft samples (at p0 and p2) were incubated with primary antibodies specific for mesenchymal (Vimentin (Vim)), endothelial (von Willebrand factor (VWF)), macrophage (CD68), dendritic (CD1a), and epithelial lineages (AE1-AE3) for 24 hours at 4°C followed by fluorescently conjugated secondary antibodies for a similar period. The slides were counterstained with 4', 6-diamidino-2-phenylindole (DAPI), which illuminates cell nuclear material (blue). Results confirmed that established cultures were not contaminated by any other cell types since and were considered pure epithelial cultures by the sole expression of all cells with the epithelial lineage marker AE1-AE3 (magnification 400×).
Lineage verification of small airway epithelial cells The cell yield in this case was lower than that reported (~4.1 × 10 6 cells) but the number of brushings conducted was also lower in comparison (2-3 vs. 4-6) [17]. However, when compared to cell yields from non-transplant [18] or paediatric patients [8], yields are much lower. Reasons behind the reduced yield are unclear but may be specific to the post-transplant state or related to medication. Although large numbers of bronchial epithelial cells have been collected from sources such as surgically resected lung [19], explanted lung [20] and cadavers [21], sample availability severely limits the utility of this approach in high turnover laboratory projects. Conversely, using brushings from lungs from living patients not only allows the collection of a much higher numbers of samples, but also facilitates longitudinal analyses and the more rapid translation of in vitro studies to clinical practice.
As with LAEC, SAEC from whole lung or resection tissue have been successfully cultured using similar methodologies [20,21]. SAEC have also been obtained using an ultrathin fibrescope [22][23][24] or unguided bronchial brushings [12]. However, only one laboratory has successfully cultured human SAEC from bronchial brushings [25], derived from smokers, COPD patients and controls using an ultrathin fibrescope. These SAEC were directly cultured in 48 well plates and harvested for ICC. Although epithelial lineage was verified, no data was reported confirming small airway lineage [25], and the utility of this in vitro model was limited since cultures were not expanded through repeated passage. Expanded cultures facilitate varied experiments and the acquisition of multifaceted data including cellular morphology, gene and protein expression, and soluble protein expression in culture supernatant.
The culture establishment rate was higher than that reported by similar studies [17], but was still compromised by contamination with passenger organisms, insufficient cell yield and BOS. Bacterial and fungal contamination occurred despite the inclusion of antibiotic (gentamycin, penicillin and streptamycin) and antifungal (amphotericin B) agents. Attempts at increasing the doses of these agents in culture media were unsuccessful due to cytotoxicity (data not shown). Unfortunately endemic infection with opportunistic pathogens is common in this patient group and has been previously noted to compromise large airway cell culture [13,17]. In this study, we interestingly observed that establishing cell cultures from small airway (but not large airway) brushings of BOS patients was more difficult. Forrest et al did not report the effect of BOS grade on culture establishment [17], and we can find no other literature on the topic. Further investigation revealed revealed that the inability to establish cultures from BOS affected small airways was independent of both cell yield and the presence of passen-ger organisms (Table 2). Collectively, our results suggest that identifying the reasons for poor culture establishment may in fact provide insight into BOS pathogenesis. In this regard, our laboratory is currently investigating epithelial cell phenotype and function in BOS, the progenitor capacity and proliferative potential of these cells, as well as their tendency to undergo programmed cell death or senescence.

Conclusion
In conclusion, we have developed methodology for successfully collecting and culturing SAEC from humans during bronchoscopy. The techniques employed utilise commonly available equipment, facilitating easy and consistent sample collection. Given the importance of the small airways in a number of pulmonary diseases, including OB, the methods established here could facilitate several avenues of respiratory research. The present authors are using SAEC from transplanted lungs to develop an in vitro model for OB, however the sampling techniques described could be used to develop small airway submerged or air liquid interface culture models for diseases such as asthma, cystic fibrosis and COPD.