3.1 TGF-β induces EMT of A549 cells
The A549 cells were first treated with various amounts of TGF-β at different time periods to determine the safe and effective concentration of TGF-β and incubation time required for investigating EMT in vitro. When A549 cells were treated with 1.0, 2.5, 5.0, 10 and 20 ng/ml TGF-β for 24 h, we observed a dose-dependent increase in the expression of the mesenchymal marker α-SMA and decrease in the expression of the mesenchymal marker E-cadherin using western blotting (Fig. 1a). Although α-SMA level increased in A549 cells after treatment with 20 ng/ml TGF-β, cell viability was poor. In addition, A549 cells were treated with 10 ng/ml TGF-β at different time periods (0, 6, 12, 24, and 48 h) to determine the appropriate time frame of treatment. We observed an increase in the expression of α-SMA in a time-dependent manner, which peaked at 48 h by 8.72 ± 0.45-fold compared with that of the control. Nevertheless, cell death was also evident. E-cadherin expression decreased significantly in a time-dependent manner after 24 h of exposure to TGF-β (Fig. 1b). Therefore, the optimal TGF-β concentration and incubation time were 10 ng/ml and 24 h, respectively. Indeed, the increase in the level of collagen-I (determined using ELISA) correlated with the degree of fibrosis in A549 cells exposed to TGF-β (Fig. 1c). Thus, we successfully established an in vitro model of EMT-induced fibrosis using A549 cells treated with optimal concentration of TGF-β for specific exposure duration.
3.2 Up-regulation of C/EBPβ is involved in TGF-β-induced EMT
Previous study has shown that phosphorylation of C/EBPβ is involved in pulmonary fibrosis in mice [19]. To investigate the precise roles of C/EBPβ in fibrosis, the relationship between C/EBPβ activation and TGF-β-induced EMT was investigated. In this study, we observed that TGF-β was up-regulated C/EBPβ mRNA level in dose-and time-dependent manner (Fig. 2a, b). Furthermore, western blot analysis showed that the elevated levels of C/EBPβ in A549 cells correlated closely with dose-and time of TGF-β treatment, as it peaked when 10 ng/ml TGF-β was used for 48 h (Fig. 2c, d). Obviously, the TGF-β-mediated increase in C/EBPβ expression correlated well with occurrence of EMT in A549 cells. Furthermore, we used luciferase reporter assay to better understand C/EBPβ activation after TGF-β-induced EMT. Indeed, exposure of A549 cells to TGF-β generated time-and dose-dependent increase in C/EBPβ-luciferase activity and exhibited a 2.83 ± 0.42-fold increase in expression compared with that of the control (Fig. 2e, f). Collectively, these observations suggested that TGF-β increased C/EBPβ expression and activation during the EMT.
3.3 Loss of C/EBPβ shifts TGF-β-induced collagen deposition following EMT
Previous studies have suggested that C/EBPβ activation is involved in the pulmonary fibrotic process [13, 14]. Other studies have reported that mice with C/EBPβ deficiency antagonise BLM-induced pulmonary fibrosis in vivo [13]. Hence, we investigated whether C/EBPβ is required for TGF-β-activated EMT and collagen-I deposition. In this regard, C/EBPβ siRNA (10 nM) was used to establish the reducing gene model in A549 cells. As shown in Fig. 3a, the C/EBPβ siRNA successfully decreased gene expression. A549 cells transfected with the C/EBPβ siRNA showed attenuation of TGF-β-induced α-SMA and collagen-I expression (Fig. 3a, b). Furthermore, TGF-β could not increase C/EBPβ-luciferase activity in A549 cells treated with the C/EBPβ siRNA (Fig. 3c). Taken together, our results suggest that C/EBPβ is a crucial factor for regulating TGF-β-induced EMT and collagen-I deposition in pulmonary fibrosis.
3.4 TGF-β induced C/EBPβ binding to α-SMA promoter in A549 cells
As mentioned above, our data confirmed that C/EBPβ played a pivotal role in EMT and pulmonary fibrosis in vitro. As an important transcription factor, C/EBPβ triggered the expression of downstream genes by binding to its cognate sites in the gene promoters. However, the C/EBPβ binding site on the α-SMA promoter region in A549 cells is not known. Therefore, to elucidate the molecular mechanism through which TGF-β regulates α-SMA expression, we checked the putative C/EBPβ-binding sites in the 5′ promoter region of human α-SMA gene. Also, the 5′ promoter region of the human, mouse and ratα-SMA gene was examined by using Multiple Sequence Alignment, We found a conserved putative C/EBPβ-binding motif TTGGGCAA in the 5′ promoter region within 200 bp from the transcription start site was identified (Figure 4A). Therefore, we hypothesized that the putative C/EBPβ-binding motif is a C/EBPβ-responsive cis-element that mediates the upregulation of the α-SMA gene, and that the activation of this cis-element is critical to the development of EMT. To test our hypothesis, we evaluated C/EBPβ binding to the putative binding motif present in the α-SMA promoter in A549 cells by using the chromatin immunoprecipitation (ChIP) assay. We observed that TGF-β-treated A549 cells showed increased C/EBPβ binding to the α-SMA promoter region (Fig. 4B). These results suggested that activated C/EBPβ accelerates TGF-β-induced EMT by binding to the α-SMA promoter region in A549 cells.
3.5 Role of acetylation of C/EBPβ in binding to the α-SMA promoter in A549 cells
Other reports have shown that phosphorylation of C/EBPβ plays a critical role in alveolar EMT and acts an essential step in pulmonary fibrosis. As a transcription factor, C/EBPβ activates its downstream signals via post-translational modification, such as phosphorylation, acetylation and methylation. However, reports that regard to the role of C/EBPβ acetylation in pulmonary fibrosis are lacking. As shown above, activated C/EBPβ binds to the α-SMA promoter region in TGF-β-treated A549 cells. Nonetheless, the mechanism connecting C/EBPβ binding to α-SMA expression is not clearly understood. Hence, we investigated the effect of TGF-β on C/EBPβ modification and observed that acetylation of C/EBPβ increased significantly by 6.42 ± 0.72 -fold in A549 cells treated with TGF-β (Fig. 5). Interestingly, enhanced α-SMA expression was observed in samples treated with the C/EBPβ antibody but not in TGF-β-treated cells with IgG. Collectively, our results showed that α-SMA expression is triggered by activated C/EBPβ, which is acetylated in an in vitro model of pulmonary fibrosis.
3.6 Acetylation and deacetylation of C/EBPβ in TGF-β-induced EMT and collagen deposition
In the present study, we used TGF-β-treated A549 cells to mimic the pulmonary fibrotic model in vitro. The binding of C/EBPβ to α-SMA and the increased expression and acetylation of C/EBPβ indicated that acetylated C/EBPβ was involved in EMT and pulmonary fibrosis. As shown in Fig. 6a, TGF-β treatment led to acetylation of C/EBPβ and accelerated EMT-induced collagen-I deposition. Acetylation of C/EBPβ is important for its activation. To clarify whether acetylation of C/EBPβ is necessary for pulmonary fibrosis, we subsequently investigated the effect of deacetylation of C/EBPβ on EMT using sirtuin1 (SIRT1). As reported, SIRT1, a class III histone deacetylase (HDAC), specifically deacetylates histone or non-histone proteins. C/EBPβ is one of the deacetylation targets of SIRT1 [20]. We observed that SIRT1 reversed TGF-β-induced C/EBPβ acetylation (Fig. 6a). Interestingly, C/EBPβ deacetylation significantly reversed the elevated expression of α-SMA and collagen-I in TGF-β-treated A549 cells (Fig. 6a, b). As shown above, SIRT1 stimulation also suppressed the increased C/EBPβ-luciferase activity in TGF-β-treated A549 cells (Fig. 6c). These observations suggested that acetylation and deacetylation are useful steps in regulating C/EBPβ functions. All these results confirm that C/EBPβ acetylation is a key player in alveolar EMT and that pulmonary fibrosis is blocked by its deacetylation.