Human Peripheral Blood-derived Mast Cells Contribute to Epithelial to Mesenchymal Transition in Bronchial Epithelial Cells in the Presence of IL-1β

Background: Bronchial epithelial to mesenchymal transition (cid:0) EMT (cid:0) is an important mechanism for the airway remodeling in asthmatics. Mast cell is one of the critical effector cells in pathogenesis of asthma. Although mast cells have been shown to release a plethora of pro-brotic factors with the potential to induce EMT, it is not clear whether mast cells also directly have an impact on the bronchial EMT. In this study, we investigated the contribution of human mast cells to EMT in human bronchial epithelial cell line 16-HBE. Methods: Human peripheral blood-derived mast cells were co-cultured with 16-HBE cells. The protein and mRNA expression of E-cadherin and vimentin in 16-HBE cells were analyzed by Western blot and quantitative real-time PCR. A scratch wound assay was performed to evaluate the migratory properties of the 16-HBE cells. Results: Mast cells alone failed to produce signicant effects on the epithelial morphology, mobility, and expression of E-cadherin and vimentin. However, mast cells in combination of interleukin (IL)-1β signicantly decreased E-cadherin expression but increased vimentin expression in the co-cultured 16-HBE cells, which exhibited a spindle-like appearance with reduced cell junctions and enhanced migration. The down-regulation of E-cadherin expression and up-regulation of vimentin expression were not abrogated by the transforming growth factor (TGF)-β1 neutralizing antibody. Conclusion: Mast cells combined with IL-1β, not mast cells alone, were able to induce EMT in 16-HBE cells through a TGF-β1-independent mechanism. The results of in vitro culture suggest the possibility that mast cells contribute to human bronchial epithelial EMT in the asthmatic airway tissues with high level of IL-1β.

Eosinophils, mast cells and T lymphocytes are the prime in ammatory cells in ltrated in asthmatic airways. Mast cells are believed to be key effector cells that are responsible for elicitation and maintenanceof airway in ammation through the releasing ofan array of in ammatory mediators [15,16].
Mast cells are also the important source of the pro-brotic cytokines and growth factors as mentioned above [16,17]. Although these pro-brotic factors, especially the growth factor TGF-β1, have been proved to be EMT inducers, it is not clear whether mast cells also directly have an impact on the bronchial EMT. IL-1β is one of the most important in ammatory factors implicated in the in ammatory response in the airways of asthmatics [10,[18][19][20]. In addition to powerful proin ammatory and pro-brotic effects, IL-1β is also capable of activating human mast cells to produce multiple cytokines, such as IL-6, 8, 13 and monocyte chemoattractant protein-1 (MCP-1) [21,22]. In this study, we investigated the effect of mast cells on EMT by examining the alteration of morphology, expression of EMT marker proteins and mobility in human bronchial epithelial cells co-cultured with human peripheral blood-derived mast cells in the presence and absence of IL-1β.

Mast Cell Cultures
Human mast cells were cultured according to the methods reported by Wang et al [23]. Brie y, mononuclear cells were separated from peripheral blood of healthy volunteers by density-gradient centrifugation. Lineage-negative cells were then puri ed from the mononuclear cells by depletion of cells expressing a panel of lineage antigens, CD2, CD3, CD11b, CD14, CD15, CD16, CD19, CD56, CD123, and CD235a using human Lineage Cell Depletion kit (Miltenyi Biotech, Bergisch Gladbach, Germany) according to the manufacturer's instructions. The isolated lineage-negative cells were cultured in serumfree methylcellulose medium containing 200 ng/mL SCF, 50 ng/mL IL-6 and 5 ng/mL IL-3. After 1 week of culture, fresh methylcellulose medium containing 100 ng/mL SCF and 50 ng/mL IL-6 was layered over the methylcellulose cultures every 7 days. After 4 ~ 6 weeks, cells were collected and maintained in complete IMDM supplemented with 100 ng/mL SCF, 50 ng/mL IL-6 and 10% FCS, and the culture medium was replaced weekly. The purity of the cultured mast cells (> 98%) was determined by calculating the percentage of positive cells stained with anti-human mast cell tryptase mAb (Chemicon International, UK) [23].

Cell line
Human bronchial epithelial cell line 16-HBE-14o (16-HBE) was used in the present study. 16-HBE cells were maintained in IMDM medium supplemented with 10% heat-inactivated FCS, 100 µg/ml penicillin and streptomycin, at 37 °C and 5% CO 2 in humidi ed atmosphere. 16-HBE cells were used for the next experiments when they reached 80% con uence.
Cell co-culture experiments 16-HBE cells were cultured in 6-well plates at 5 × 10 4 cells per well with IMDM medium containing 100 ng/mL SCF, 50 ng/mL IL-6 and 10% FCS. For the co-culture experiments, mast cells (2.5 × 10 5 /well) with or without 10 ng/mL of IL-1β were added to 16-HBE cell cultures and incubated for 3 days. In some experiments, co-culture of 16-HBE cells and mast cells were performed in the presence of mouse anti-TGF-β1 mAb (2 µg/ml). The total number of cells in each well is the same. After 3 days of co-culture, the suspended mast cells, adherent 16-HBE cells, and cell medium in plates were collected for the subsequent experiments.
Western blot 16-HBE cells in 6-well plates were collected by digesting with trypsin and lysed in 50 µL lysis buffer (1 mL RIPA buffer containing 10 µL PMSF). Lysates were centrifuged at 1200 rpm for 15 min at 4 °C. Total protein concentrations in supernatants were quanti ed by the BCA Protein Assay Kit (Beyotime Biotechnology, Shanghai, China) according to the manufacturer's instructions. Denatured total protein of 50 µg per well was separated by SDS-PAGE (10% separation gel, 5% stacking gel), followed by transfer onto polyvinylidene uoride membranes (Millipore Corporation). The membranes were incubated with Tris-base sodium solution containing Tween-20 and 5% fat-free milk for 1 h at room temperature to block non-speci c binding sites, then incubated with primary mAb against E-cadherin (1:100) or vimentin (1:100) overnight at 4 °C. Next, the membranes were washed and incubated with horseradish-peroxidase conjugated goat anti-rabbit/mouse secondary antibody (1:3000 dilution, Beyotime Biotechnology, Shanghai, China) for 1 h at room temperature. The immunoblots were detected with enhanced chemiluminescence (Servicebio, Wuhan, China) and exposed to GeneGnome HR Model (Synoptics Ltd, Cambridge, UK). The protein levels were quanti ed as ratios to the GAPDH band intensities by using GeneTools software.
Measurement of TGF-β1 in supernatants TGF-β1 contents in the culture supernatants were measured using a commercial enzyme linked immunosorbent assay kit (Boster, Wuhan, China) according to the manufacturer's instructions.

Scratch wound healing assay
Wound healing assay was performed as described by Doernerand Zuraw [13]. Brie y,16-HBE cells were seeded in 6-well plates at a density of 5 × 10 4 /well and cultured in IMDM containing 100 ng/mL SCF, 50 ng/mL IL-6 and 10% fetal calf serum. After 24 hours, mast cells (5 × 10 5 /well) with or without 10 ng/mL of IL-1β were added to 16-HBE cultures. When 16-HBE cells grew to 90% con uence, a straight wound line was made across the cell monolayer using a sterile 10 µl pipette tip. Wells were washed with PBS and serum-free IMDM was applied to the cells. Images were captured at 0 and 48 hours after wound creation. The area between the wound edges was measured at each time point using Image J software. The remaining wound areas were expressed as a percentage of area at time 0 and calculated by the formula: % of wound closure = (measurement at 48 h/measurement at time 0 h) * 100.

Data analysis
Each result is expressed as the mean ± SEM for n independent experiments. Statistical analysis was performed using one-way analysis of variance with Bonferroni's post hoc test. GraphPad Prism 5.0 was used to analyze data and P < 0.05 was considered signi cant.

Morphological changes in16-HBE cells
Under phase contrast microscopy the 16-HBE cells cultured alone or the co-cultured 16-HBE cells with mast cells in the absence of IL-1β displayed a typical epithelial cobblestone-like shape and were attached to each other. In contrast, the 16-HBE cells stimulated with IL-1β or co-cultured with mast cells and IL-1β were elongated and exhibited a spindle-shape or broblast-like morphology, and the cell-cell contacts were decreased or disappeared (Fig. 1).
Protein and mRNA expression of E-cadherin was signi cantly down-regulated, but vimentin was signi cantly up-regulated in 16-HBE cells treated with IL-1β or the co-cultured 16-HBE with mast cells and IL-1β when compared with that in 16-HBE cells alone or the co-cultured 16-HBE cells with mast cells. The combination of mast cells and IL-1β also signi cantly enhanced the decrease of E-cadherin expression and increase of vimentin expression in 16-HBE cells induced by IL-1β alone ( Fig. 2A, B). The changes of Ecadherin and vimentin expression in the IL-1β-treated 16-HBE cells were completely abrogated in the presence of anti-TGF-β1 mAb. However, in the co-cultured 16-HBE cells with mast cells and IL-1β, anti-TGF-β1 mAb did not signi cantly inhibit the decrease of E-cadherin expression and the increase of vimentin expression (Fig. 3A, B). There was no signi cant difference in the expression of E-cadherin and vimentin between 16-HBE cells alone and the co-cultured 16-HBE cells with mast cells (Fig. 2A, B). The results indicate that IL-1β instead of mast cells alone could down-regulate E-cadherin expression while up-regulate vimentin expression in 16-HBE cells, but mast cells combined IL-1β produced a synergetic effect on E-cadherin or vimentin expression in  Production and mRNA expression of TGF-β1 As shown in Fig. 4A, a low concentration of TGF-β1 could be detected in the culture supernatant of 16-HBE cells or mast cells. By comparison, the level of TGF-β1 was signi cantly increased in the supernatant from the IL-1β-stimulated 16-HBE cells or mast cells. The TGF-β1 production from the co-culture of 16-HBE cells with mast cells and IL-1β was also signi cantly greater than that from 16-HBE cells or mast cells treated with IL-1β. RT-PCR analysis showed that IL-1β signi cantly promoted TGF-β1 expression in 16-HBE cells (Fig. 4B) or in mast cells (Fig. 4C). The combination of 16-HBE cells and mast cells with IL-1β signi cantly enhanced TGF-β1 expression either in 16-HBE cells (Fig. 4B) or in mast cells (Fig. 4C). The production and mRNA expression of TGF-β1 in the co-cultured 16-HBE cells (Fig. 4A, B) or in mast cells ( (Fig. 4A, C) had no signi cantly differences in contrast with 16-HBE cells or mast cells alone.

16-HBE cell migration
The wound area in the scratch wound healing assay indirectly re ected the migratory ability of 16-HBE cells. The smaller area means the cultured 16-HBE cells have more powerful mobility. As illustrated in Discussion E-cadherin is a marker protein of epithelium and is indispensable for the maintenance of the epithelial phenotype [24,25]. Suppression of E-cadherin expression leads to the disassembly of cell-cell adhesion and subsequent loss of epithelial polarity, and initiates the differentiation of epithelial cells into broblasttype mesenchymal cells [25][26][27]. Vimentin is a protein expressed in mesenchymal cells, but it can also be expressed in epithelial cells where EMT occurs. Epithelial cells undergoing EMT are also characteristic of irregularly spindle-shaped appearance and powerful migration capacity [28]. In the present study, we found that mast cells alone did not affect signi cantly E-cadherin and vimentin expression in the cocultured 16-HBE cells, nor did affect the epithelial morphology and migration ability, suggesting that mast cells alone have not the potential to induce EMT in 16-HBE cells. Unlike mast cells, IL-1β signi cantly down-regulated E-cadherin expression while signi cantly upregulated vimentin expression in 16-HBE cells. Moreover, the IL-1β-treated 16-HBE cells exhibited the morphological features of mesenchymal cells and an increasing mobility. The results indicate that IL-1β is able to induce 16-HBE cells to undergo EMT.
TGF-β1is the most potent and most well described inducer of EMT identi ed so far [28]. In the normal airways, a low of level of TGF-β1 can be secreted by airway epithelial cells and other structural cells, and is an essential growth factor for the maintaining of epithelial integrity [10]. However, repeated aggression of in ammation or exogenous irritants (allergens, infections and cigarette smoke) leads to the release of large amounts of TGF-β1 from the damaged epithelial cells in asthmatic airways [10,29]. Increased levels of TGF-β1 have been reported in bronchoalveolar lavage uid and bronchial biopsies of asthmatic patients [30,31]. A lot of studies have shown that TGF-β1 could directly induce EMT in human bronchial epithelial cells [11][12][13][14]. Yasukawa et al. reported that eosinophils induce EMT in airway epithelial cells via increasing TGF-β1 production [32]. In our experiment, we observed that epithelial cells cultured alone produced only a small amount of TGF-β1, however, mRNA expression and protein production of TGF-β1 were signi cantly increased in IL-1β-treated 16-HBE cells compared with 16-HBE cells alone. In addition, in the presence of anti-TGF-β1 mAb, IL-1β-induced suppression of E-cadherin expression and enhancement of vimentin expression in 16-HBE cells were abrogated completely. These results indicate that the IL-1β could stimulate 16-HBE cells to produce TGF-β1 which mediated the conversion of 16-HBE cells to mesenchymal cells.
Previous reports have demonstrated that human mast cell line LAD2 and the cultured mast cells from the progenitors in human cord blood or peripheral blood can constitutively express mRNA for TGF-β1 and produce bioactive TGF-β1 [33,34]. In our study, the peripheral blood-derived mast cells could releasea small amount of TGF-β1, but IL-1β stimulation signi cantly enhanced expression of both TGF-β1 mRNA and protein in the cultured mast cells. The results provide further evidence that mast cells are also a potential source of TGF-β1 and in ammatory stimulation is able to activate mast cells to release more TGF-β1. It is likely that the amount of TGF-β1 produced by 16-HBE cells and/or mast cells in a quiescent state is too small to activate TGF-β signaling pathway to affect the expression of E-cadherin and vimentin in 16-HBE cells, therefore, mast cells failed to induce EMT in the co-cultured 16-HBE cells.
Our study found that mast cells and 16-HBE cells in the co-culture were incapable of interacting with each other in the expression of TGF-β1, however, when IL-1β was added to the co-culture, TGF-β1 expression was signi cantly higher in the co-cultured mast cells than in IL-1β-stimulated mast cells, and higher in the co-cultured 16-HBE cells than in IL-1β-stimulated 16-HBE cells. The result shows that addition of IL-1β could signi cantly enhance mRNA expression and protein production of TGF-β1 in both 16-HBE cells and mast cells. Since activated epithelial cells or mast cells also have potential to produce a variety of biologically active mediators such as IL-4, TNF-α and IL-1β, which in turn further promote TGF-β1 expression in mast cells and epithelial cells [10,16,35], therefore, the enhancing TGF-β1 expression in the co-cultured cells with IL-1β may result from the stimulation of active mediators other than TGF-β1 secreted by 16-HBE cells and/or mast cells in an autocrine/paracrine fashion.
In our experiment, mast cells alone had no effect on E-cadherin or vimentin expression in the co-cultured 16-HBE cells, however, when IL-1β was added to the co-cultures, E-cadherin expression was signi cantly decreased while vimentin expression was signi cantly increased in 16-HBE cells. The result indicates that EMT could be induced in the 16-HBE cells co-cultured with mast cells and IL-1β. When compared with IL-1β-induced EMT of 16-HBE cells, EMT in co-cultured 16-HBE cells with mast cells was signi cantly enhanced by the addition of IL-1β, and could not be abrogated signi cantly by TGF-β1 neutralizing antibody, suggesting that the EMT was independent on TGF-β1. Given that IL-1β itself induced EMT of 16-HBE through a TGF-β1-dependent mechanism, mast cells may play a vital role in promoting the transformation of 16-HBE cells to mesenchymal cells in the present of IL-1β. As mentioned above, in addition to TGF-β1, activated mast cells can also release multiple other pro brotic factors such as epidermal growth factor, connective tissue growth factor, broblast growth factor-2, IL-6, IL-1β and TNF-α, which have been reported to participate in epithelial EMT [7,13,14,16,29]. Thus, it is likely that mast cells induced the TGF-β1-independent EMT in the co-cultured 16-HBE cells by releasing some unknown EMT inducers in the presence of IL-1β.
In conclusion, we for the rst time demonstrated that IL-1β alone induced a TGF-β1-dependent EMT in 16-HBE cells, but human peripheral blood-derived mast cells alone failed to induce EMT of 16-HBE cells. Mast cells combined with IL-1β induced EMT in 16-HBE cells through a TGF-β1-independent mechanism. Our results suggest the possibility that mast cells contribute to EMT in human bronchial epithelial cells in the in ammatory airway tissues of asthmatics.