IL-4 and IL-13 elicit several important responses in airway epithelium including chemokine secretion and mucous secretion that may contribute to airway inflammation, cell migration, and differentiation. These cytokines have overlapping but not identical effector profiles likely due to shared subunits in their receptor complexes. These receptors are variably described in epithelial cells, and the relative expression, localization, and function of these receptors in differentiated and repairing epithelial cells are not clear. We examined IL-4/IL-13 receptor expression and localization in primary airway epithelial cells collected from normal human lungs and grown under conditions yielding both undifferentiated and differentiated cells inclusive of basal, goblet, and ciliated cell phenotypes. Gene expression of the IL-4Rα, IL-2Rγc, IL-13Rα1, and IL-13Rα2 receptor subunits increased with differentiation, but different patterns of localization and protein abundance were seen for each subunit based on both differentiation and the cell subtypes present. Increased expression of receptor subunits observed in more differentiated cells was associated with more substantial functional responses to IL-4 stimulation including increased eotaxin-3 expression and accelerated migration after injury. We demonstrate substantial differences in IL-4/IL-13 receptor subunit expression and responsiveness to IL-4 based on the extent of airway epithelial cell differentiation and suggest that these differences may have functional consequences in airway inflammation.
the airway epithelium is both a target of inflammatory and physical insults and an effector of ongoing airway inflammation (2, 20, 31, 37, 45). As such, it plays a critical role in diseases such as chronic asthma. Epithelial injury leads to release of chemokines such as IL-8 (7, 8, 13), GM-CSF (13, 42), and eotaxin (65) that then stimulate inflammatory cell infiltration. Injury also leads to disordered regulation of submucosal myofibroblasts (19) and subsequent subbasement membrane fibrinogenesis (40). Persistent epithelial injury (31, 39, 56) is a characteristic part of airway remodeling (9) in asthma.
Activated lymphocytes are a prominent feature in asthmatic airways (5, 6, 32). BAL fluid obtained from asthmatics is enriched in both IL-4 (50) and IL-13 (21, 22) but not IFN-γ (50), indicating the presence of T helper type 2 (Th2) subclass CD4+ cells. Both IL-4 and IL-13 may stimulate epithelial cells to produce chemokines such as eotaxin-3 (28, 55, 65), GM-CSF (42), IL-8 (8, 28), and RANTES (17, 46) and growth factors such as transforming growth factor-β (TGF-β) (50, 57) and the EGF receptor-binding factor, TGF-α (4). Overexpression of either IL-4 or IL-13 in murine airways elicits inflammation, subepithelial fibrosis, and mucous cell metaplasia (49, 67).
IL-4 and IL-13 have overlapping but not identical effector profiles likely due to shared subunits in their receptor complexes (43). IL-4 can bind to two distinct receptors: the type I receptor consisting of the IL-4Rα chain dimerized with the common γ-chain (γc) of the IL-2 receptor or the type II receptor with IL-4Rα dimerized to the IL-13Rα1 chain (34, 52). Similarly, IL-13 can bind to either the type II receptor or the separate IL-13Rα2 receptor (16, 24). These receptor subunits are described variously in airway epithelium. IL-4Rα can be demonstrated in situ in both normal and asthmatic airways (29, 54). One study examining cultured primary human airway epithelial cells (AEC) grown in submersion culture demonstrated the presence of all four subunits (17), whereas another could demonstrate substantial IL-4Rα and IL-13Rα1 expression but γc expression only at low levels (59). One recent study could not demonstrate the presence of IL-13Rα2 in human AEC in submersion culture (36).
Multiple signaling pathways are activated by these receptors. Dimerization of either the type I or type II receptor causes the cytoplasmic tails to associate with tyrosine kinases of the Janus family (JAK1–3 and the related TYK2) (43, 44). IL-4Rα associates with JAK1 (44) or JAK2 (51), γc with JAK3, and IL-13Rα1 with either JAK2 or the related TYK2 (44, 51). Dimerization of the receptor chains enhances JAK activity and leads to phosphorylation of tyrosine residues in the cytoplasmic domain of IL-4Rα. These residues then act as docking sites for signaling molecules that contain Src homology 2 (SH2) domains (27) such as STAT6.
Another pathway that mediates IL-4 and IL-13 signaling from the type I or type II receptor is via either insulin receptor substrate (IRS)-1 or IRS-2. IL-4Rα contains a motif in the I4R region that is highly homologous to sequences within insulin and insulin growth factor-1 receptors that bind IRS-1 (25, 61) and IRS-2 (60).
A potential effector from the IL-13Rα1 subunit is STAT3. STAT3 is activated by either IL-4 or IL-13 in airway fibroblasts stimulated to release eotaxin (11). IL-4 treatment of aberrant, but not normal, glioma cells similarly activates this signaling molecule (48).
The receptor unique to IL-13, IL-13Rα2, binds IL-13 with high affinity (16, 24) and has been considered a soluble decoy receptor (1, 66). One recent report suggests that although the intracytoplasmic region of the receptor is short, it has a motif to activate an activator protein-1 (AP-1) variant containing c-Jun and Fra-2, which then activates the TGF-β promoter (12) to induce collagen synthesis and lung fibrosis. IL-13Rα2 may also associate physically with the IL-4Rα subunit, blocking its activity and thus downregulating IL-4-stimulated effects (1). These studies suggest multiple potential roles for IL-13Rα2 in mediating effects on AEC, but whether this receptor is expressed in differentiated airway epithelium is not clear.
Given the multiplicity of IL-4/IL-13 receptors, one potential mechanism by which IL-4 responsiveness could be controlled in both fully differentiated and in repairing AEC would be by regulating expression of receptor subunits. It is known, for example, that the lack of IL-2Rγc leads to greater IL-13 sensitivity compared with IL-4 in mouse macrophages (23), presumably because the lack of the type I receptor results in less efficient binding of IL-4. Expression of the IL-13Rα1 receptor subunit is also upregulated in human eosinophils by inflammatory proteins such as TGF-α and TGF-β (41).
It is not clear, however, whether expression of IL-4/IL-13 receptor subunits differs in AEC based on the state of differentiation of the cells and whether changes in expression have functional consequences. To evaluate this, we examined both gene and protein expression of receptor subunits in culture models of AEC that allow for differentiation of these cells to appropriate an in vivo state. We then examined two measures of IL-4 responses in AEC, eotaxin-3 expression (28) and cell migration (63), to examine the functional significance of receptor subunit expression based on AEC differentiation. Our data demonstrate that IL-4/IL-13 receptor expression in fully differentiated AEC correlate with significant functional differences in eotaxin-3 expression and in cell migration after injury. These results suggest that IL-4 responsiveness in airway epithelium may be due, in part, to differences in receptor subunit expression based on the state of AEC differentiation.
MATERIALS AND METHODS
Medium A consisted of BEBM (CC-3171) and a SingleQuot Kit (CC-4171) both purchased from Lonza (Walkersville, MD.). Medium B consisted of medium A plus 1.25 μg/ml amphotericin, 100 μg/ml ceftazidime, 80 μg/ml tobramycin, 100 μg/ml vancomycin, 100 U/ml penicillin, 100 μg/ml streptomycin, 100 U/ml nystatin, and 50 μg/ml gentamicin. DMEM medium consisted of DMEM (Cellgro 10-017-CM) supplemented with 10% FCS, 100 g/ml streptomycin, and 100 U/ml penicillin G. ALI medium consisted of 1:1 BEBM to DMEM. Every 500 ml were supplemented with a SingleQuot Kit, 130 μg/ml bovine pituitary extract, 25 ng/ml EGF, 50 nM retinoic acid, 0.5 mg/ml low-endotoxin bovine serum albumin, and 20 U/ml nystatin. Antibodies directed against β-tubulin were obtained from Abcam (Cambridge, MA). Antibodies directed against IL-4/IL-13 receptor subunits and antibodies directed against cytokeratin 5 (CK5) and mucin 5AC (MUC5AC) used in confocal microscopy were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). All secondary antibodies used in confocal microscopy were obtained from Invitrogen (Carlsbad, CA). For antibodies in Western blots, a rabbit polyclonal anti-IL-4Rα was obtained from Strategic Diagnostics (Newark, DE), a goat monoclonal anti-IL-13Rα2 was obtained from R&D Systems (Minneapolis, MN), and mouse monoclonal anti-IL-13Rα1, anti-IL-2Rγc, and anti-GAPDH were obtained from Abcam. All other reagents were obtained from Sigma-Aldrich (St. Louis, MO).
We (62) have described these methods previously. The use of primary human AEC was approved by the University of Chicago Institutional Review Board. Primary cells were obtained from two sources. Cells were purchased from Lonza. These were grown per instructions on collagen IV-coated containers in medium A in 5% CO2 atmosphere at 37°C. Cells also were collected from lungs provided by the Regional Organ Bank of Illinois (Elmhurst, IL). Mucosal membranes from central airways were dissected and incubated in 1% protease at 37°C for 2 h. Epithelial cells then were removed by pipetting gently up and down several times and then repeating this operation 4–5 times with fresh medium A. Cell suspensions were centrifuged at ∼460 g for 3 min at 4°C. Cells were grown on collagen IV-coated containers in medium B for 1–3 days in 5% CO2 atmosphere at 37°C. Then, the strength of the antibiotics and antimycotics in medium B were reduced progressively every day to 80, 50, 25, and then 0%. No differences were noted in cell migration, eotaxin expression, or receptor expression based on the source of cells.
Cells were replated onto collagen IV-coated 12-mm Transwell membranes in submersion culture for 2–4 days in ALI medium. Cells then were changed to air-liquid interface (ALI) culture conditions for an additional 1 or 3 wk, changing the medium every other day on the basal side only. Some cells were kept in submersion culture for 4 days and used in experiments when confluent. All cells, whether in submersion or ALI culture, were used at 100% confluence.
We (62, 63) have described this method previously. Cells were fixed in 4% paraformaldehyde for 15 min and stored in PBS plus 0.1% sodium azide at 4°C until use. Cells were washed, blocked with donkey serum as appropriate, and then stained with two primary antibodies. Antibodies were selected based on the vendor's recommendation and preliminary experiments (data not shown). The first was directed against CK5 (mouse anti-human, sc-32721, or goat anti-human, sc-17090) to mark basal cells (14, 58), MUC5AC (rabbit anti-human, sc-20118, or goat anti-human, sc-16910) to mark goblet cells (3, 35), or β-tubulin (rabbit anti-human, ab6046, or goat anti-human, ab21057) to mark ciliated cells (30). The second were antibodies directed against subunits of the IL-4/IL-13 receptors: either mouse anti-human (sc-28361) or rabbit anti-human (sc-684) IL-4Rα, either rabbit anti-human (sc-25849) or goat anti-human (sc27861) IL-13Rα1, rabbit anti-human (sc-667) IL-2Rγc, and mouse anti-human IL-13Rα2. Antibodies were selected to avoid species overlap. Each primary antibody was added at appropriate concentrations with incubation overnight at 4°C. Cells then were washed, and secondary antibodies directed against each primary were added at room temperature for 1 h. Cells were imaged using an Olympus DSU disk scanning confocal system on an IX81 microscope platform (Olympus America, Melville, NY). Background was calculated from images collected after staining that omitted the primary antibody; the background was then subtracted using ImageJ (W. S. Rasband, National Institutes of Health, Bethesda, MD; http://rsbweb.nih.gov/ij/). To do this, a histogram was constructed of the LUT values for each control image in which the primary antibody was omitted. The LUT value representing the 99th percentile of LUT values from this histogram then was used as the minimum for the display range in processing subsequent z-stacks from that experimental series. Contrast, brightness, and threshold in each z-stack were otherwise not changed. For each z-stack generated, x-y slices were selected starting from the membrane on which cells were grown and progressing to the top of the cells; these slices then were combined using a maximum intensity protocol (ImageJ) to generate final images. At least three separate experiments were done at each stage of differentiation and for each set of antibodies used, and representative images were selected.
Total RNA was isolated from cells using a PerfectPure RNA 96 Cell Kit (5 Prime, Gaithersburg, MD) following the manufacturer's protocol. Samples were treated with DNase I (5 Prime). Total RNA was reverse-transcribed using random primers and Superscript II reverse transcriptase (Invitrogen). Real-time RT-PCR was performed using a Bio-Rad iCycler iQ PCR Detection System using iQ Supermix (Bio-Rad, Hercules, CA) and gene-specific primers for each receptor subunit and for eotaxin-3 (Table 1).
Migration after injury.
We (63, 64) have published details of this assay previously. Briefly, confluent monolayers were washed twice and placed in serum-free or defined medium appropriate for the cells being studied. Mediators or control diluent were added as appropriate. Linear wounds of ∼1.0-mm width were made with a stylet. Wound closure was measured serially for 18 h starting immediately after wound creation. Wounds were photographed edge to edge. Measurements were made using ImageJ software.
In some experiments, cells were fixed and labeled for the presence of receptor subunits and for the presence of CK5 6 or 18 h after injury and then imaged by confocal microscopy.
Western blot analysis.
We (63, 64) have published details of this method previously. Briefly, immunoblotting was done for each IL-4/IL-13 receptor subunit using total protein lysates collected from cells in submersion culture or ALI culture for 1 or 3 wk. Antibodies were selected based on the vendor's recommendation and preliminary experiments (data not shown). A positive control using lysates provided by each vendor was run in a separate lane on the same gels. Membranes were probed using an antibody for GAPDH to control for differences in protein loading. Densitometry was done using ImageJ software for both receptor and GAPDH blots; the ratio of receptor blot density to GAPDH blot density reflected the relative abundance of the receptor subunit in each lysate.
Real-time RT-PCR data are expressed as comparative cycle time ratios using 18S rRNA as an internal standard. Data for eotaxin-3 expression were log-transformed before statistical analysis. Wound repair data are expressed as remaining wound width (means ± SE) compared with width at time 0. Densitometry values from Western blots are expressed as means ± SE. Differences were examined by ANOVA; when significant differences were found, post hoc analysis was done using Fisher protected least significant difference test. In some comparisons, a paired t-test was used. Differences were considered significant when P < 0.05.
Primary cells grown in submersion culture stained positive for CK5 and were almost always negative for both β-tubulin and MUC5AC, as expected in cells with no significant differentiation (Fig. 1). By 21 days, cells in ALI culture displayed characteristics of differentiated epithelium, and numerous ciliated cells could be identified by phase-contrast microscopy. An increased proportion of cells labeling for either MUC5AC or β-tubulin were seen at 21 days compared with cells 7 days in ALI cultures (Fig. 1). Few (<2%) cells were labeled as “intermediate” or “indeterminate” cells with localization of more than one cell subtype antibody.
Expression of IL-4/IL-13 receptor subunits.
We first examined expression of each subunit in each culture condition using real-time RT-PCR. Clear differences were seen in the gene expression of IL-4 and IL-13 receptor subunits as a function of differentiation. For each receptor subunit, expression was modest in primary cells grown in submersion culture (Fig. 2). Differentiation for 1 wk in ALI culture conditions did not change gene expression significantly, but by 3 wk in ALI culture expression of IL-4Rα, IL-13Rα1, and IL-2Rγc had increased significantly.
Expression of IL-13Rα2 also was low in cells grown in both submersion culture and in ALI culture for 1 wk (Fig. 2). Differentiation in ALI culture for 3 wk increased IL-13Rα2 gene expression, but there was significant variance in the samples such that there was no significant difference in expression compared with the other groups.
These data suggested that although differentiation was associated with increased expression of all IL-4/IL-13 receptor subunits, gene expression for the specific subunits comprising the type I and type II IL-4 receptors did not occur in a fixed ratio at any point of differentiation in AEC.
Localization and protein abundance of IL-4/IL-13 receptor subunits.
We then examined relative abundance and localization of each receptor subunit in AEC at each point in differentiation using both confocal microscopy (for abundance and localization) and Western blot analysis (for overall abundance only). As with gene expression, striking differences were found for each subunit, particularly as related to cellular state of differentiation. Table 2 summarizes the changes in receptor protein localization noted in confocal microscopy.
IL-4Rα abundance was very low in cells grown in submersion culture as assessed by confocal microscopy (Fig. 3A); these data correlated well with gene expression (Fig. 2). In cells grown under ALI conditions for 1 wk, abundance was low and was seen primarily in basal (CK5-positive) and goblet (MUC5AC-positive) cells. When grown to 3 wk in ALI conditions, IL-4Rα showed the greatest overlap with CK5 expression (yellow in the images) in basal cells, with less overlap in both columnar (β-tubulin-positive) and goblet cells. However, in Western blot analysis, less IL-4Rα abundance was noted overall in protein lysates obtained from fully differentiated AEC grown to 3 wk in ALI conditions (Fig. 3B). As the number of goblet and columnar cells increased substantially as expected with differentiation, our data suggest that whereas overall abundance decreased in these cell subtypes, basal cells preferentially expressed more of the IL-4Rα subunit.
In contrast, abundance and localization of IL-13Rα1 revealed a different pattern. This subunit had widespread expression in cells grown in submersion culture as assessed by both confocal microscopy and by Western blot analysis (Fig. 4, A and B). By confocal microscopy, in cells grown in ALI culture for 1 wk, there was modest abundance in all three cell types, whereas in more differentiated cells grown in ALI culture for 3 wk, less IL-13Rα1 could be found in basal and columnar cells and instead was seen in goblet cells. By Western blot, no significant differences were seen in IL-13Rα1 receptor subunit expression. These data suggest that the relative decrease in localization seen in both basal and ciliated cells was balanced by increased abundance in goblet cells.
Although gene expression of IL-2Rγc was low in all cell types and culture conditions compared with the other two IL-4-associated receptor subunits, protein expression by confocal microscopy was clearly present in each culture condition and was seen in all three major cell types (Fig. 5A). This was confirmed by Western blot (Fig. 5B), which demonstrated relatively constant abundance of the receptor subunit in both undifferentiated and differentiated cells.
The abundance and localization of IL-13Rα2 was different than that for other receptor subunits. By confocal microscopy, the receptor subunit was seen only in the nuclear and perinuclear regions in cells grown in either submersion or ALI culture for 1 wk (Fig. 6A), whereas in more differentiated cells grown in ALI culture for 3 wk, IL-13Rα2 abundance was greater and was seen in both basal and goblet cells, with modest staining in ciliated cells. By Western blot, no significant differences were seen in lysates collected from any group (Fig. 6B), suggesting that overall abundance was similar even as the localization of the protein changed substantially.
Expression of eotaxin-3 in differentiating cells.
Does the difference in IL-4 receptor subunit expression lead to differences in function after IL-4 treatment? To answer this question, we first examined expression of eotaxin-3 after IL-4 treatment. This chemokine has been demonstrated to be secreted by primary AEC and by epithelial cell lines upon IL-4 stimulation (17, 28, 55, 65), and we reasoned that this would be an appropriate marker to examine functional differences due to receptor expression. Cells were grown either in submersion culture or in ALI culture for 1 or 3 wk and then treated with 10 ng/ml IL-4 for 24 h. Eotaxin-3 expression then was assessed by real-time RT-PCR. As shown in Fig. 7, there was no significant baseline (unstimulated) expression in any group. Treatment with IL-4 elicited significant expression of eotaxin-3 after IL-4 stimulation in each group, and more differentiated cells expressed significantly more eotaxin-3 compared with undifferentiated cells grown in submersion culture.
Migration after injury in differentiating cells.
We then examined whether differences in IL-4 receptor subunit expression translated into differences in migration after injury. We (63) have demonstrated previously that IL-4 stimulates migration in fully differentiated AEC after mechanical injury. In the present study, we examined migration in cells grown either in submersion culture or in ALI culture for 1 or 3 wk. Cells were treated with 10 ng/ml IL-4 immediately after generation of linear wounds, and wound closure was examined over 18 h. In untreated cells, migration was most rapid in cells grown in submersion culture and least rapid in cells grown in ALI culture for 3 wk (Fig. 8). Treatment with IL-4 accelerated migration only in cells grown in ALI culture for 3 wk: 18 h after injury, migration was 24 ± 5% of original wound width (OWW) in treated cells vs. 46 ± 3% in untreated cells (P = 0.03; n = 8; Fig. 8). In contrast, IL-4 treatment had only a modest effect in cells grown in ALI culture for 1 wk: 18 h after injury, migration was 19 ± 5% of OWW in treated cells vs. 35 ± 8% in untreated cells (P = 0.09; n = 12; Fig. 8). In cells grown in ALI culture, treatment with IL-4 did not accelerate migration in cells grown in submersion culture (n = 8; Fig. 8).
Localization of IL-4/IL-13 receptor subunits after injury.
We then examined localization of each receptor subunit in differentiated AEC after mechanical injury. In these experiments, linear wounds were generated in cells grown in ALI culture for 1 or 3 wk, and cells were fixed 6 or 18 h later. Localization of CK5 to demonstrate basal epithelial cells within the wound region demonstrated that >95% of cells were CK5-positive in all cases (data not shown). This finding is consistent with basal cells being the migratory cell responsible for initial wound closure.
Localization of IL-4Rα by confocal microscopy in either partially (1-wk ALI) or more completely (3-wk ALI) differentiated cells demonstrated increased abundance in cells at and near the wound edge compared with cells further away, although the receptor subunit was present in nearly all cells (Fig. 9). No significant differences were seen in either 1- or 3-wk ALI cells.
Localization of IL-13Rα1 also demonstrated a substantial increase in cells at and near the wound edge. As expected from previous experiments, in 1-wk ALI cells, localization was modest in cells away from the edge, whereas in 3-wk ALI cells, localization was seen near and beyond the wound edge (Fig. 9). Likewise, localization of IL-2Rγc, although less abundant than seen for IL-4Rα and IL-13Rα1, was noted at and near the wound edge in both 1- and 3-wk ALI cells (Fig. 9).
The localization of IL-13Rα2 after injury and during wound closure differed from that of the other three subunits. IL-13Rα2 abundance was modest at/near the wound edge in cells grown in ALI culture for 1 wk 6 h after injury, but relatively more was seen at 18 h after injury (Fig. 9). In cells grown in ALI culture for 3 wk, more IL-13Rα2, like other receptor subunits, was concentrated more at and near the wound edge 6 and 18 h after injury, although its expression was observed at a low level in other cells.
Epithelial cells stimulated with either IL-4 or IL-13 express several cytokines and chemokines that trigger and maintain airway inflammation (8, 28, 42, 55, 65). Both IL-4 and IL-13 elicit changes in epithelial structure, morphology, and differentiation that may contribute to both airway inflammation and airway remodeling (49, 57, 67). Although it is logical to assume that IL-4/IL-13 receptor components may be regulated, their expression and presence in airway epithelium have implications for understanding the contribution made by both cytokines to the pathogenesis of asthma.
We demonstrate for the first time that IL-4 and IL-13 receptor subunit expression and localization depend on the state of epithelial cell differentiation. Furthermore, these differences correlate to differences in important inflammatory and reparative responses such as eotaxin-3 secretion and migration after injury.
We selected three points of differentiation (phenotype shift) in cultured AEC: submersion culture, in which cells are undifferentiated, and cells in ALI culture when partly (1 wk) and more completely (3 wk) differentiated, as demonstrated by the appearance of markers for goblet cells (MUC5AC) and ciliated, columnar cells (β-tubulin). Undifferentiated and partly differentiated epithelial cells are observed in airways following injury (10, 26). Thus it is possible that the differential contribution of IL-4 and IL-13 to chemokine secretion, repair, and differentiation could be based on the relative expression of receptors for these cytokines in lesser and more fully differentiated cells.
As shown in our current study, gene expression of each receptor subunit differed with the phenotype of each set of cells. For each subunit, gene expression was low in undifferentiated cells grown in submersion culture and was highest in more fully differentiated cells grown for a longer period in ALI culture. Furthermore, whereas gene expression for both IL-4Rα and IL-13Rα1 expression was comparable at each stage of differentiation, IL-2Rγc expression was modest at all stages, and modest expression of the IL-13Rα2 subunit was noticeable mainly in cells grown in ALI culture for 3 wk. These data would predict that more differentiated cells would respond more robustly to IL-4 stimulation, and this indeed was the case: for both eotaxin-3 expression, a useful marker of cell chemokine secretory function, and wound closure, a useful marker of cell migration after injury, more differentiated cells had a greater response to treatment with IL-4. These data suggest a clear change not only in expression, but also in function of IL-4/IL-13 receptor subunits that is related to differentiation.
Different patterns of localization of each subunit were seen, although protein localization (confocal microscopy) and protein expression (Western blot) did not necessarily correlate with overall gene expression. IL-2Rγc protein expression was relatively constant and was localized to basal cells at all levels of differentiation and also to goblet and ciliated cells at 1 and 3 wk of ALI culture. This finding suggests that the IL-2Rγc protein is stable in cells, with expression potentially regulated by a posttranscriptional or posttranslational mechanism.
In contrast, protein expression of IL-13Rα1 as measured by Western blot was relatively constant over time, even as its gene expression increased, and changes in protein localization in different cell subtypes were noted: localization was decreased in basal cells and increased in goblet cells by 3-wk ALI culture. These data suggest that gene expression corresponding to this subunit may be increasing only in goblet cells as they become more differentiated.
Protein localization of the common partner for both the type I and type II receptor, IL-4Rα, suggested a sparse localization in undifferentiated cells shifting to a prominent expression in basal cells in more differentiated culture. This matched the pattern of gene expression, which also increased over time in culture. IL-4Rα abundance was modest in goblet cells, and the substantial numbers of these cells in differentiated culture (Fig. 1) likely accounts for the overall decrease in the abundance of this receptor subunit as assessed by Western blot. Likewise, by confocal microscopy, IL-13Rα2 was clearly expressed in all sets of cells, but the pattern was different: it was cytoplasmic in differentiated cells but perinuclear in cells grown in submersion culture. Whereas both gene expression and protein abundance increased over time, localization shifted so that the receptor subunit was expressed most abundantly in basal and goblet cells in 3-wk ALI culture.
The observed correlation in the prominent expression of both IL-4Rα and IL-13Rα2 in more differentiated cells and the shift in their localization are of particular interest in light of the finding that increased levels of IL-13Rα2 inhibit both IL-13- and IL-4-mediated epithelial responses (1). Since it has been shown that these two receptor subunits can form a complex (47), the coincident expression of these receptor subunits only in differentiated cells may provide a regulatory mechanism that inhibits fully differentiated cells from responding to IL-13 or IL-4. Hence, such regulation would reserve these cytokines to function as stimulators of cellular responses needed for repair and differentiation only in injured epithelia. We (63) previously demonstrated that IL-4 could stimulate migration of fully differentiated AEC after injury. We now show that less-differentiated AEC migrate less well in response to IL-4. As migration is an early event after injury, less responsiveness in partly differentiated cells may be a helpful regulatory mechanism to limit migration to only the time period required, early after injury, for coverage of a damaged basement membrane.
Differences in the magnitude and location of IL-4/IL-13 receptor subunit expression based on cell subtype and differentiation may have important implications not only in repair and remodeling, but also in airway inflammation. A more differentiated, intact epithelium may respond with greater chemokine expression on IL-4 or IL-13 stimulation, whereas a damaged epithelium with a less-differentiated phenotype may be less able to do so. Likewise, fully differentiated cells at or near the site of injury may in the initial stages be stimulated to migrate and spread, whereas cells shifted to a less-differentiated phenotype (e.g., those that have already migrated into a site of injury) may not respond. IL-13 induces goblet cell differentiation both in vivo (15, 67) and in vitro (33). In our cell model system, the localization of IL-13Rα1, a necessary partner in the type II (IL-4Rα and IL-13Rα1) receptor that binds IL-13, exhibits high expression in basal cells in submersion and early ALI culture when these cells could be induced to shift their phenotype to goblet cells. In more differentiated culture layers, IL-13Rα1 localization was abundant in goblet cells, as would be predicted given that IL-13 can induce enhanced mucin secretion.
It is not difficult to envision how perturbation of the normal patterns of expression of the IL-13/IL-4 receptor subunits could result in pathogenesis. It is known that ingestion of alcohol leads to changes in the gene expression patterns of the IL-13Rα1 and IL-13Rα2 receptors in rat airways, with an increase in the expression of the IL-13Rα1 subunit and a decrease in the IL-13Rα2 subunit (38). Examined in light of our current findings, the alcohol-induced expression pattern resembles that observed in less-differentiated (submerged or 1 wk) cells rather than that normally observed in highly differentiated epithelia. Thus these cells may be more susceptible to modulation by IL-13 and more likely to develop a remodeled, asthmatic phenotype.
Differences in receptor expression, whether as a function of differentiation or as a response to injury, may also lead to differences in effector pathway signaling. We (63) have demonstrated previously that IL-4-stimulated migration of AEC requires signaling from the receptor to IRS-1 and IRS-2 such that signaling via STAT6 is irrelevant to migration. One recent study suggests that only type I receptor, γc-dependent signaling induced efficient activation of IRS-2 (18). Regulation of receptor subunit expression then may alter the activation of effector pathways and thus provide another opportunity for fine-tuning epithelial cell responsiveness in a given state. We also note the possibility of differences in signaling intermediates downstream of these receptors (STAT3, STAT6, and IRS-1/2) as a function of differentiation that may also regulate epithelial cell responsiveness either beyond or in place of that seen with surface receptors.
As our data are derived using an in vitro cell culture model, important influences on IL-4 and IL-13 receptor subunit expression may be absent, including local and circulating factors and the influence of cells beneath the basement membrane such as fibroblasts. However, the use of ALI culture permits examination of receptor expression during specific stages of differentiation that cannot be done solely using submersion culture techniques and that is nearly impossible to control in vivo. Use of the ALI culture system also permits careful examination of endogenous cellular influences on receptor expression. Future experiments can examine the role of paracrine and circulating factors added to this system. Receptor expression may also change as a function of disease state and inflammation in airways. Careful comparison and correlation then will be required to understand specific receptor expression changes in a developing or repairing epithelium in vivo.
One important limitation of the ALI culture system is that there may be some imprecision in labeling cell subtypes, particularly in partly differentiated cells. Although we observed relatively few (<2%) indeterminate cells, such cells have been observed in other AEC culture systems (53) and may confound efforts to localize receptors to epithelial cell subtypes, particularly when ALI conditions are varied.
The apparent discordance between receptor protein expression by Western blot analysis and receptor localization by confocal microscopy can be explained, in part, by the fact that protein lysates are “averaged” across all cells collected from a culture container, whereas confocal microscopy can examine individual cell subtypes. In turn, confocal microscopy generally is less quantitative and may not distinguish between internally stored receptor and receptor on the cell surface, although cytoplasmic vs. perinuclear/nuclear localization sometimes can be determined. The disparity in overall protein abundance of IL-4Rα (Western blot) compared with both its gene expression (real-time RT-PCR) and localization (microscopy), for example, suggests a shift in localization and preferential expression in at least one cell subtype; increased localization of the subunit is observed in basal cells in the 3-wk ALI cultures. We recognize that caution is required in interpreting these apparent shifts in localization vs. overall abundance given the limits of resolution of both Western blot and confocal microscopy.
In summary, we demonstrate increased expression of IL-4 and IL-13 receptor subunits as a function of the degree of differentiation in cultured, primary human AEC and after mechanical injury. Increased expression of specific subunits may lead to increasing and differential responses to both IL-4 and IL-13 in AEC. Such changes in expression may regulate, along with ligand availability and regulation of downstream effector pathways, cellular responses to these cytokines.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-080417.
No conflicts of interest, financial or otherwise, are declared by the author(s).
We thank Stage Marroquin for technical assistance.
This work was presented, in part, at the international meeting of the American Thoracic Society on May 18, 2008.
- Copyright © 2010 the American Physiological Society