The Th2 cytokines interleukin (IL)-4 and IL-13 are thought to play critical roles in the airway inflammation and hyperresposiveness that characterize asthma. Recent evidence indicates that IL-13 can mediate these effects by acting directly on airway epithelial cells. Here we evaluated early [signal transducer and activator of transcription (STAT)6 phosphorylation] and delayed [granulocyte/macrophage colony-stimulating factor (GM-CSF) and transforming growth factor-β2 (TGF-β2) secretion] responses of airway epithelial cells to IL-4 and IL-13 stimulation and the dependence of these responses on the culture technique employed. As expected, normal human bronchial epithelial cells grown on microporous inserts at an air-liquid interface (ALI) expressed a well-differentiated mucociliary phenotype; in contrast, cells grown on plastic in submerged cultures were poorly differentiated. When stimulated with IL-4 or IL-13, the magnitude and duration of STAT6 phosphorylation under the differing culture conditions were statistically indistinguishable. In contrast, cytokine secretion responses to IL-4 and IL-13 were highly dependent on the culture technique; cells cultured on plastic exhibited significant concentration-dependent increases in GM-CSF and TGF-β2 secretion, whereas cells grown at ALI showed no statistically significant response. These results demonstrate that the coupling between early signal transduction responses to IL-4 and IL-13 and downstream functions such as cytokine secretion may be critically dependent on the cell culture technique employed and the resulting differentiation status of bronchial epithelial cells.
- bronchial epithelial cells
- cell differentiation
- granulocyte/macrophage colony-stimulating factor
- transforming growth factor-β
the Th2 cytokines interleukin (IL)-4 and IL-13 have been shown to play critical roles in allergic inflammation and the development of airway hyperresponsiveness (AHR; see Ref. 32). Although both cytokines play key roles in T-cell phenotypic commitment and IgE production (32, 34), recent evidence demonstrates that IL-13 can mediate the development of AHR by direct action on airway epithelial cells (13). These results underscore the need to understand the gene and protein responses evoked in the airway epithelium by IL-4 and, especially, IL-13.
In vivo, the airway epithelium is composed of a mixed population of ciliated, nonciliated, and mucous-secreting cells (10). Studies using bronchial epithelial cell cultures differentiated to induce a mucociliary phenotype have shown that IL-4 and IL-13 stimulation modulate basic cell functions such as proliferation (3), ciliary beating, and mucous production (2, 14, 25). In contrast, our growing understanding of the ability of Th2 cytokines to stimulate gene expression (15) and production of soluble factors (4, 17, 20, 24, 31) is derived primarily from study of submerged cultures of airway epithelial cells grown on plastic, conditions that result in a poorly differentiated, proliferating cell population. Among the soluble factors evoked by IL-4/13 stimulation under these conditions, of particular interest are granulocyte/macrophage colony-stimulating factor (GM-CSF) and transforming growth factor (TGF)-β2 (17, 20, 24, 31). These factors are of special interest because of their purported role in the biology of asthma (22, 27) and AHR (11, 30).
We investigated signal transducer and activator of transcription (STAT)6 phosphorylation and secretion of GM-CSF and TGF-β2 in airway epithelial cells grown using the two most commonly employed culture techniques, monolayer submerged cultures on plastic and ALI cultures on microporous substrates, to identify the dependence of IL-4/13-induced signal transduction and soluble protein secretion on the culture technique employed. Our results indicate that, although early signal transduction responses to IL-4/13 stimulation such as STAT6 phosphorylation may be independent of the culture conditions employed, the transduction of these early responses into downstream responses such as GM-CSF and TGF-β2 secretion may depend strongly on the culture technique and the resulting polarity and differentiation status of the cultures.
MATERIALS AND METHODS
Normal human bronchial epithelial (NHBE) cells were purchased from Clonetics-BioWhittaker (San Diego, CA) and grown on tissue culture-treated flasks (5% CO2, 37°C) in bronchial epithelial growth medium (BEGM; Clonetics), a defined medium provided by Clonetics. We further supplemented this medium with BSA (1.5 μg/ml) and retinoic acid (50 nM) following the method of Gray and colleagues (6). Passage 4 cells were plated on either uncoated, tissue culture-treated polyester microporous inserts (0.4 μm pore size, Transwell Clear; Costar, Cambridge, MA) or on tissue culture-treated plastic (MULTIWELL; Becton-Dickinson Labware, Franklin Lakes, NJ). Cells on plastic were fed with BEGM supplemented with BSA and retinoic acid as above until 80–90% confluent. Approximately 16 h before each experiment, the culture medium was changed to a minimal medium containing bronchial epithelial basal medium (BEBM; Clonetics) supplemented with only insulin (5 μg/ml), transferrin (5 μg/ml), penicillin (100 U/ml), and streptomycin (100 μg/ml).
Cells plated on microporous inserts were fed with a 1:1 mixture of BEGM and DMEM (Life Technologies, Frederick, MD) with the same final concentration of media additives as used for plastic wells. Medium was applied both apically and basally until cells reached confluence, and then basally after an air-liquid interface (ALI) was established at approximately day 7. The cells were maintained until a differentiated cell population with prominent mucus secretion and early cilia development was present (days 10–14; see Fig. 1). Approximately 16 h before each experiment, the culture medium was changed to a minimal medium containing a 1:1 mixture of BEBM and DMEM supplemented with only insulin (5 μg/ml), transferrin (5 μg/ml), penicillin (100 U/ml), and streptomycin (100 μg/ml). Therefore, all media supplements were identical in the two culture conditions, with the only difference being the use of 50% DMEM in the base medium for cells cultured on microporous inserts. In some experiments designed to demonstrate the role of the base medium, the medium used during cytokine stimulation was reversed (BEBM on microporous inserts, 1:1 BEBM-DMEM on plastic wells). In experiments designed to define the role of the culture surface, cells were grown on microporous inserts to subconfluence and stimulated under submerged culture conditions using medium identical to that used for plastic well experiments. Cells from two different donors were used throughout these experiments.
Immunocytochemistry for cell differentiation markers.
Differentiation status of NHBE cells in the two different culture conditions was confirmed by immunofluorescent analysis. NHBE cells cultured on a plastic well or at ALI were fixed in 3% paraformaldehyde for 5 min and then treated with 1% Triton X-100 in PBS for 5 min. Nonspecific immunogloblin binding was blocked by incubation with 10% normal goat serum in PBS with 2% BSA for 1 h at room temperature. Mouse monoclonal antibodies against E-cadherin (1:500 dilution; Zymed Laboratories, South San Francisco, CA), β-tubulin IV (1:250 dilution; Sigma, St. Louis, MO), and mucin 5AC (MUC5AC, 1:500 dilution; Neo Markers, Fremont, CA) were diluted in PBS with 2% BSA and incubated at 4°C overnight. To visualize the primary antibodies, the samples were then incubated in the secondary antibody [goat anti-mouse IgG conjugate; Molecular Probes, Eugene, OR] diluted 1:1,000 in PBS at room temperature for 30 min. To visualize F-actin, cells were incubated for 30 min at room temperature with a 1:40 dilution of Alexa Fluor 488-conjugated phalloidin (Molecular Probes) in PBS with 2% BSA. For nuclear counterstaining, cells were incubated in 1 μg/ml propidium iodide (Sigma) in PBS for 5 min. Controls in the absence of primary antibodies confirmed the specificity of the immunolabeling. Conventional fluorescence images were obtained with a Leica DMLB microscope (Leica) using a ×20 objective, and x-z cross sections were constructed from image stacks obtained using a Sarastro 2000 confocal laser-scanning microscope.
Stimulation of NHBE cells with IL-4 or IL-13.
After 16 h in the appropriate minimal medium, fresh minimal medium alone or with various concentrations (0.1–100 ng/ml) of recombinant human IL-4 (R&D Systems, Minneapolis, MN) or IL-13 (R&D Systems) was added to the cells and incubated up to 48 h at 37°C in an atmosphere of 5% CO2. Cytokines were applied apically to the cells grown in plastic wells and to the basal medium of cells grown at ALI on microporous inserts. In a subset of experiments, the IL-13 stimulus was applied to the apical surface of ALI cultures. In the experiments using submerged cultures on microporous inserts, IL-13 was applied in both the apical and basal medium.
After stimulation with IL-4 or IL-13 as described above, NHBE cell lysates were collected in SDS sample buffer containing 62.5 mM Tris·HCl (pH 6.8 at 25°C), 2% wt/vol SDS, 10% glycerol, 50 mM dithiothreitol, and 0.01% wt/vol bromophenol blue. The collected cell lysates were boiled for 8 min and then cooled rapidly on ice. STAT6 and the phosphorylated, active form of STAT6 (p-STAT6) were measured by Western blotting. After centrifugation for 5 min at 11,000 g, the supernatants (20 μl/lane for p-STAT6 and 5 μl/lane for STAT6) were loaded on a 7.5% Tris·HCl gel (Bio-Rad). Solubilized proteins were separated by SDS-PAGE and transferred electrophorically to a polyvinylidene fluoride microporous membrane (Immobilon-P; Millipore, Bedford, MA). The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.1% Tween 20 (TBST) for 1 h at room temperature. The blots were probed with rabbit polyclonal anti-human p-STAT6 antibody (Tyr641; Cell Signaling, Beverly, MA) or anti-human STAT6 antibody (20-778; Upstate Biotechnology, Lake Placid, NY) using a 1:1,000 dilution for both in TBST with 5% BSA. Blots were washed in TBST and subsequently incubated for 1 h at room temperature in TBST with 5% nonfat dry milk containing horseradish peroxidase-conjugated goat anti-rabbit IgG (1:2,000 dilution). The proteins were visualized by light emission on film with enhanced chemiluminescent substrate (Cell Signaling). The bands of p-STAT6 and STAT6 were visualized at ∼100–110 kDa and were quantified by standard densitometry.
Measurement of GM-CSF and TGF-β2.
After stimulation with IL-4 or IL-13 as described above, the apical medium from plastic wells and the basal medium from ALI wells were removed and reserved at −80°C for later analysis. Release of GM-CSF and TGF-β2 in medium was measured using commercially available ELISA kits according to the manufacturer's instructions (Quantikine; R&D Systems). Just before the ELISA for TGF-β2, all samples were activated using 1 N HCl and then neutralized using 1.2 N NaOH in 0.5 M HEPES to activate latent TGF-β2 to the immunoreactive form. In these assays, the minimum detectable concentrations of human GM-CSF and TGF-β2 are 3 and 7 pg/ml, respectively. In some experiments, cells were detached from microporous inserts or plastic wells by incubation with 0.5 mg/ml trypsin-EDTA (GIBCO, Grand Island, NY) at 37°C for 5–10 min and then neutralized by an equal volume of trypsin-neutralizing solution (Clonetics). Cells were counted, and the results of cytokine production were then normalized by the cell number per well and expressed as picograms per 105 cells.
Quantitative real-time PCR.
Total RNA was purified from NHBE cell lysates after exposure to IL-13 (10 ng/ml) or control medium for 48 h (RNeasy; Qiagen, Valencia, CA). Equal amounts of RNA were reverse transcribed to cDNA using Ready-to-Go RT-PCR beads (Amersham, Piscataway, NJ). Real-time PCR reactions were performed using SYBR Green Master Mix (Bio-Rad, Hercules, CA) in an iCycler PCR System (Bio-Rad). The degree of change in transcript levels was calculated relative to control treatment using the delta delta cycle threshold (Ct) method. The expression of housekeeping gene GAPDH was used as the reference standard. Real-time PCR primers (Invitrogen) targeting GM-CSF, TGF-β2, and GAPDH were designed using Primer Express software (Applied Biosystems, Foster City, CA) with similar melting point temperatures, primer lengths, and amplicon lengths to obtain similar PCR efficiency. Each primer was tested against GAPDH over a range of concentrations to ensure similar PCR efficiency. The respective forward and reverse primers for GM-CSF, TGF-β2, and GAPDH were CCTTGACCATGATGGCCAG and TGGAGGGCAGTGCTGTTTG, GCTGGAGCATGCCCGTATT and ACCCTGCTGTGCTGAGTGTCT, and TGGGCTACACTGAGCACCAG and GGGTGTCGCTGTTGAAGTCA, respectively.
Data are presented as means ± SE. Data were evaluated by repeated-measures ANOVA in experiments examining the dose-dependent effect of cytokines on the ratio of p-STAT6 to STAT6. In the experiments examining dose-dependent effect of cytokines on GM-CSF and TGF-β2, one-way ANOVA was performed. If the overall F-statistic was significant at the 0.05 level, subsequent intergroup significance testing was assessed post hoc by the Dunnett's F-test for experiments examining the dose-dependent effect of cytokines on GM-CSF and TGF-β2. Experiments to compare baseline cytokine release and percent change in cytokines on plastic with those at ALI were evaluated by unpaired t-test. The reversed media and substrate experiments, and experiments for donor comparison, were evaluated by paired t-test.
As expected, and as shown in Fig. 1, cells grown as submerged monolayers on plastic differed greatly from cells grown at ALI on microporous inserts in their morphology and distribution of differentiation markers. The most obvious difference, visualized by nuclear counterstaining, was the marked increase in cell density attained under ALI conditions. As seen by staining with phalloidin and the antibody to E-cadherin, cells grown at ALI were highly organized, exhibiting mature cell-cell junctions and a network of pericellular actin staining. In contrast, cells grown on plastic exhibited little or no cell-cell adherence and a heterogeneous pattern of actin staining characterized by dense cytoplasmic stress fibers. Cells grown on plastic exhibited a diffuse, weak pattern of cytoplasmic staining for mucin (Muc)5AC, a marker of mucous differentiation, that was fairly uniform from cell to cell (data not shown). In contrast, cells grown at ALI exhibited a heterogeneous pattern of Muc5AC staining marked by the absence of staining in some cells and prominent staining of others near the apical surface, indicating the presence of Muc5AC-positive secretory granules in mucous-secreting goblet cells. The β-tubulin IV antibody was used as a marker of ciliary differentiation. Cells grown on plastic demonstrated a punctate cell surface staining pattern that was uniform over the entire cell population (data not shown). Cells grown at ALI, in contrast, demonstrated a heterogeneous pattern with positive staining of distinct elongated cilia-like structures at the apical surface in a subset of the cell population.
IL-4/13-dependent phosphorylation events.
Cells grown on both plastic wells and under ALI conditions exhibited a robust increase in STAT6 phosphorylation in response to stimulation with IL-4 and IL-13 (Fig. 2A). The peak in p-STAT6 protein occurred at ∼30–60 min, and sustained elevation of STAT6 phosphorylation was maintained up to 24 h. Cells grown at ALI did not respond when stimulated with IL-13 applied apically, indicating that the expression of IL-4/13 receptors in these cells is likely limited to the basolateral aspects of the cells. Compared with unstimulated controls, IL-4 and IL-13 significantly increased the p-STAT6-to-STAT6 ratio in a concentration-dependent manner for each culture technique (P < 0.0001 for each, Fig. 2B). Despite the dramatic phenotypic changes associated with differentiation (Fig. 1), there were no significant differences in the STAT6 phosphorylation response of NHBE cells cultured under the two conditions (Fig. 2B).
IL-4/13-dependent GM-CSF and TGF-β2 production.
To characterize soluble factors released by NHBE cells under the influence of IL-4/13 stimulation, medium was collected from NHBE cells after stimulation for 48 h with varying concentrations of IL-4 and IL-13. As shown in Table 1, NHBE cells cultured on plastic produced approximately three to five times more GM-CSF and TGF-β2 under baseline (0 ng/ml) conditions than did cells cultured at ALI. For NHBE cells cultured on plastic, production of both cytokines increased in a concentration-dependent manner with IL-4 and IL-13 stimulation (P = 0.0035, GM-CSF release by IL-4 stimulation; P = 0.0039, GM-CSF release by IL-13 stimulation; P = 0.0313, TGF-β2 release by IL-13 stimulation). In contrast, NHBE cells at ALI did not exhibit a statistically significant increase in production of either GM-CSF or TGF-β2 in response to IL-4 or IL-13 stimulation (Table 1).
To normalize the production of soluble factors in each culture condition, total cell numbers were recorded for each (Table 1), and soluble proteins were expressed per 105 cells. Cell density in ALI wells was much greater than on plastic, but neither IL-4 nor IL-13 produced a statistically significant change in cell number per well in either culture condition after stimulation for 48 h. Normalization of GM-CSF and TGF-β2 production per cell amplified the differences in baseline (unstimulated) production of each factor (Fig. 3A). When compared with baseline production, the differences in culture technique-dependent responsiveness to IL-4/13 were differentiated easily (Fig. 3B).
The culture condition-dependent IL-13 effects on protein secretion were paralleled by changes in mRNA levels. Transcripts for GM-CSF and TGF-β2 were elevated by IL-13 treatment in plastic wells relative to control in two independent experiments (range: 2.1- to 3.2-fold change in GM-CSF, 1.4- to 1.7-fold change in TGF-β2); in contrast, IL-13 had little or negative effect on message levels when added to ALI cultures (range: 0.4- to 0.5-fold change in GM-CSF, 0.9- to 1.4-fold change in TGF-β2).
Influence of base media and cell donor.
To exclude the possibility that the difference in responsiveness of NHBE cells to IL-4/13 in the two different conditions might derive from the different base media used for each condition (BEBM for NHBE cells cultured on plastic and 1:1 BEBM-DMEM for NHBE cells cultured at ALI), we ascertained the responsiveness of NHBE cells stimulated with IL-13 in media reversed from the normal culture conditions. For both culture techniques, the baseline TGF-β2 production and effect of IL-13 were independent of base media (Fig. 4A).
To confirm that the differences in responsiveness of NHBE cells between plastic and ALI culture were not dependent on the cell donor, cells from a second donor (donor 2) were stimulated by IL-13 (10 ng/ml) for 48 h. Baseline and IL-13-induced production of GM-CSF and TGF-β2 were nearly identical in the cells from the two donors, and the culture technique-dependent responsiveness was maintained across donors (Fig. 4B).
Relative importance of substrate and duration in ALI.
To examine whether the lack of responsiveness of NHBE cells to IL-4/13 stimulation is dependent on expression of the mature mucociliary phenotype, we examined cells stimulated with IL-13 beginning on day 3 after institution of ALI (denoted early ALI), when the cells do not yet possess mature mucous secretion capabilities and apical cilia. As shown in Fig. 5, direct comparison with plastic and mature mucociliary cultures (late ALI) shows that early ALI cells exhibited baseline and IL-13-induced GM-CSF and TGF-β2 production most similar to mature ALI wells. Although TGF-β2 production from early ALI cells was statistically increased by IL-13 stimulation, the effect was muted, and the baseline production closely matched that in mature mucociliary epithelium (Fig. 5). Baseline GM-CSF secretion in the early ALI cells was suppressed compared with plastic wells, and IL-13 had no effect on GM-CSF secretion, a response identical to that in cells cultured for 10–14 days at ALI (late ALI) to express the mature mucociliary phenotype.
Given the early loss of IL-13 responsiveness after initiation of ALI, we also examined the response of subconfluent NHBE cells grown in conditions identical to those for plastic wells, with the exceptions that the medium and IL-13 stimulation were applied to both the apical and basal surface and the substrate was the porous polyethylene used for the ALI studies. As in the case of the early ALI wells, cells grown submerged on the microporous Transwell inserts (pre-ALI) exhibited a muted response to IL-13 stimulation that, although statistically significant, was markedly reduced compared with cells grown on plastic wells (Fig. 5).
Human bronchial epithelial cells are known to express subunits of the IL-4/13 receptor, including the IL-4Rα, common γ-chain, IL-13Rα1, and IL-13Rα2 (17, 29), and respond to IL-4/13 stimulation with robust increases in STAT6 phosphorylation, the major intracellular effector of the IL-4/13 signaling pathway (28). We show here that this early response to IL-4/13 is present in both submerged monolayer cultures of NHBE cells that are poorly differentiated and in mucociliary ALI cultures (Fig. 2); the similar responses are present despite differences in culture substrate, polarity of stimulation, and differentiation status of the cells. Although we did not directly evaluate the surface expression of receptor subunits in the two differing culture techniques, we were able to confirm that the preponderance of functional receptors was on the basolateral surface of ALI cultures (Fig. 2A), since apical stimulation with IL-13 did not provoke a measurable increase in STAT6 phosphorylation. In addition to altering the localization of IL-4/13 receptors, it is probable that cellular differentiation modulates the density and distribution of IL-4/13 receptor subunits. Nevertheless, we have shown here that IL-4/13 stimulation in the two culture conditions resulted in a nearly identical time course, concentration dependence, and magnitude change in STAT6 phosphorylation. In agreement with previous findings (7, 9), we also found small increases in STAT1, ERK, JNK, and p38 phosphorylation after IL-13 stimulation (data not shown); as with STAT6 phosphorylation, these events were independent of the culture technique employed. These results strongly indicate that IL-4 and IL-13 evoke nearly identical early signaling events in NHBE cells under the conditions employed.
The culture technique independence of early IL-4/13 responses did not extend to later responses. Although airway epithelial cells are well recognized as potent sources of cytokines such as IL-6, IL-8 (5, 19), IL-16 (1), GM-CSF (5, 19, 26), TGF-β (12, 18), endothelin-1 (23), RANTES (8), and eotaxin (16), little is known regarding the effects of culture technique and cellular differentiation on the capacity of the airway epithelium to produce these factors under baseline conditions, or in response to cytokines present in the asthmatic airway. Whereas previous reports have demonstrated that NHBE cells produce GM-CSF and TGF-β2 under baseline conditions, and increase production of both under the influence of Th2 cytokines (20, 31), these reports have been limited to submerged monolayer cultures. Here we showed that differentiation in ALI conditions to a mature mucociliary phenotype significantly reduced baseline production of both GM-CSF and TGF-β2 on an absolute and per cell basis relative to cells grown in submerged conditions on plastic. Furthermore, we showed that IL-4/13-induced secretion of TGF-β2 and GM-CSF (12, 33) was completely suppressed upon mucociliary differentiation, demonstrating for the first time that NHBE responses to Th2 cytokine stimulation depend critically on the culture technique employed and the resulting differentiation status of the cells.
To more closely examine the effect of substrate, polarity, and differentiation, we stimulated cells grown on microporous inserts at subconfluence and early after the transition to ALI and compared their responses with cells grown on plastic and those grown to a mature mucociliary phenotype at ALI. The comparison revealed that switching from a plastic substrate to a microporous substrate (accompanied by a switch in polarity of stimulation from apical to apical + basal) attenuated the secretory response to IL-13, although in both cases the response to IL-13 continued to reach statistical significance. Among the Transwell groups, the comparison of baseline unstimulated secretion is complicated by the fact that cell densities increased rapidly as wells progressed from pre-ALI to early ALI to late ALI (data not shown); thus, the decrease in baseline secretion per cell from pre-ALI to late ALI was more gradual than appears in Fig. 5. Nevertheless, our results suggest a degree of complexity in the time course of loss of responsiveness to IL-13 stimulation, since the GM-CSF response was completely lost in early-ALI cells but a significant responsiveness for TGF-β2 was maintained until late ALI.
Our results demonstrate that, whereas cultured NHBE cells can mount identical early signal transduction responses to IL-4/13 stimulation independent of the culture technique, changes in the substrate, cell polarity, and differentiation status that begin with the onset of ALI modify the processing of the IL-4/13 signal and the generation of later responses. Context-dependent gene expression has been previously demonstrated for NHBE cells by modulating retinoic acid-dependent cell differentiation (21), and our results extend this concept to culture context-dependent cytokine-induced responses. These results emphasize the caution required in designing and interpreting studies aimed at understanding the responses of the bronchial epithelium to Th2 cytokines.
In conclusion, the present study demonstrates that previously identified responses to IL-4/13 stimulation in NHBE cells can be uncoupled from this stimulus depending on the culture technique employed and that this decoupling process begins rapidly after the institution of ALI culture conditions. Unraveling the factors that regulate the complex interplay between cellular differentiation and cytokine stimulation (e.g., context-dependent signal transduction, expression of transcriptional cofactors or repressors, differences in chromatin state, posttranscriptional processing) may yield new insights into important Th2-dependent effects on the airway epithelium in the asthmatic airway.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-33009. D. J. Tschumperlin is a Parker B. Francis Fellow in Pulmonary Research.
We thank Jean Lai for performance of confocal imaging.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society