A Positive Feed Forward Loop between Wnt/ β-Catenin and NOX4 Promotes Silicon Dioxide-Induced Epithelial-Mesenchymal Transition of Lung Epithelial Cells

Abstract

Silicosis is a chronic fibrotic lung disease caused by the accumulation of silica dust in the distal lung. Canonical Wnt signaling and NADPH oxidase 4 (NOX4) have been demonstrated to play a crucial role in the pathogenesis of pulmonary fibrosis including silicosis. However, the underlying mechanisms of crosstalk between these two signalings are not fully understood. In the present study, we aimed to explore the interaction of Wnt/β-catenin and NOX4 of human epithelial cells in response to an exposure of silica dust. Results demonstrated an elevated expression of key components of Wnt/β-catenin signaling and NOX4 in the lungs of silicon dioxide- (SiO₂-) induced silicosis mice. Furthermore, the activated Wnt/β-catenin and NOX4 signaling are accompanied by an inhibition of cell proliferation, an increase of ROS production and cell apoptosis, and an upregulation of profibrogenic factors in BEAS-2B human lung epithelial cells exposed to SiO₂. A mechanistic study further demonstrated that the Wnt3a-mediated activation of canonical Wnt signaling could augment the SiO₂-induced NOX4 expression and reactive oxygen species (ROS) production but reduced glutathione (GSH), while Wnt inhibitor DKK1 exhibited an opposite effect to Wnt3a. Vice versa, an overexpression of NOX4 further activated SiO₂-induced Wnt/β-catenin signaling and NFE2-related factor 2 (Nrf2) antioxidant response along with a reduction of GSH, whereas the shRNA-mediated knockdown of NOX4 showed an opposite effect to NOX4 overexpression. These results imply a positive feed forward loop between Wnt/β-catenin and NOX4 signaling that may promote epithelial-mesenchymal transition (EMT) of lung epithelial cells in response to an exposure of silica dust, which may thus provide an insight into the profibrogenic role of Wnt/β-catenin and NOX4 crosstalk in lung epithelial cell injury and pathogenesis of silicosis.

Figure 1

An elevated Wnt/β-catenin signaling activity and expression of NOX4 in lungs of mice exposed to silica (SiO₂) dust. The lungs of C57BL/6 mice were harvested at 2 weeks post the exposure of saline control or silica dust (sizes ranged 0.5-10 μm) for histological analysis by HE staining (a–f), examining alternations of the expression of signaling molecules by immunoblotting assay (g) and immunofluorescent staining (h). (a–f) Representative images of lungs of mice exposed to saline control (a–c) and silica dust (d–f). Abundant silicosis nodules were observed in lungs of mice challenged with silica dust (a–c), but not in the saline control group (d–f). (B, C, E, and F) Frame insets depict the regions of corresponding images shown in (b, c, e, f). (g) Immunoblotting analysis demonstrated an increased expression of NOX4 protein and an enhanced Wnt/β-catenin signaling activity as accessed by the increased abundance of active β-catenin (ABC) and Axin2 but a decreased Wnt inhibitor DKK1, accompanied by increased deposition of profibrogenic epithelial-mesenchymal transition (EMT) proteins alpha smooth muscle actin (α-SMA) and vimentin, in lungs exposed to silica. (h) Semiquantitative analysis of the expression of indicated proteins in (g) by evaluating the relative densitometric densities. (i) The increased expression of Wnt/β-catenin signaling ligand Wnt3a, NOX4, α-SMA, and vimentin was also predominantly detected in the silicosis nodules of silica-challenged lungs (top panel), compared to the saline (bottom panel) as ascertained by immunofluorescent staining (IF). DAPI was used for nuclear staining. Vmt: vimentin in (h). Bars in (a, d) 100 μm; (b, e) 40 μm; (c, f) 20 μm; (i) 50 μm. Data in (h) represented the mean ± SEM of 9 mice from 3 independent experiments. Compared to the control group, ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.

Figure 2

SiO₂ activates Wnt/β-catenin signaling in lung epithelial BEAS-2B cells. The lung epithelial BEAS-2B cells were exposed to different doses of SiO₂ (2 nm in size) for 24 h or 48 h, the cell viability (a) and Wnt/β-catenin signaling activity were examined. (a) A dose-dependent inhibition of cell viability was observed in BEAS-2B cells exposed to SiO₂ for 24 h and 48 h as determined by a CCK8 assay. (b) Immunofluorescent staining further corroborated the increased expression of Wnt3a and active β-catenin (ABC) proteins in cells treated with 100 μg/cm² of SiO₂ for 48 h in comparison with the saline control. (c) Dual luciferase reporter assay demonstrated a significantly enhanced SiO₂-activated Wnt/β-catenin signaling in BEAS-2B cells exposed to 100 μg/cm² of SiO₂ for 48 h compared with the saline control (CTRL) (p < 0.01). (d) Immunoblotting assay also revealed an increased abundance of Wnt3a and ABC in BEAS-2B cells exposed to 100 μg/cm² of SiO₂ for 48 h relative to that of CTRL. (e) Semiquantitative analysis of the expression of indicated proteins in (d) by evaluating the relative densitometric densities. Bars in (b) 20 μm. Compared to the CTRL, ^∗p < 0.05 and ^∗∗p < 0.01.

Figure 3

SiO₂ alters the NOX-mediated ROS production and promotes EMT and cell apoptosis in lung epithelial BEAS-2B cells. The lung epithelial BEAS-2B cells were exposed to different doses of SiO₂ (2 nm in size) for 24 h or 48 h; the alteration of ROS production (a) and expression of NOX and EMT proteins were examined. (a) A dose- and time-dependent increase of ROS production induced by SiO₂ was determined in BEAS-2B cells at 48 h but not 24 h post exposure to silica dust. (b) Immunoblotting assay revealed an increased expression of NOX4 and α-SMA but reduced expression of E-cadherin in BEAS-2B cells exposed to 100 μg/cm² of SiO₂ for 48 h relative to the saline. (c) Semiquantitative analysis of the expression of NOX1, NOX4, and NOX5 proteins in (b) by evaluating the relative densitometric densities. (d) Semiquantitative analysis of the expression of EMT-related proteins E-cadherin, α-SMA, and vimentin in (b) by evaluating the relative densitometric densities. (e) Representative images of IF exhibited more abundant NOX4, α-SMA, and vimentin proteins in BEAS-2B cells exposed to 100 μg/cm² of SiO₂ for 48 h as compared with the saline control. Bars in (c) 20 μm. Compared to the CTRL, ^∗∗p < 0.01; ^∗∗∗p < 0.001.

Figure 4

ROS scavenger NAC inhibits SiO₂-induced Wnt/β-catenin signaling activity and expression of fibrogenic factors. The lung epithelial BEAS-2B cells were exposed to 100 μg/cm² of SiO₂ (2 μm in size) for 48 h in the presence or absence of ROS scavenger NAC; the abundance of the indicated Wnt/β-catenin signaling key molecules, fibrogenic factors, and signaling of caspase-3-mediated cell apoptosis were examined by an immunoblotting assay. (a) Representative immunoblots showed that the presence of NAC reduced the SiO₂-induced Wnt/β-catenin signaling ligand Wnt3a and mediator nuclear ABC but increased the expression of Wnt inhibitor DKK1 (top panel); suppressed the SiO₂-induced fibrogenic factors MMP2, vimentin, and α-SMA but increased the abundance of epithelial cell marker E-cadherin (middle panel); and decreased the SiO₂-induced abundance of cleaved caspase-3 and BAX but increased the expression of antiapoptotic protein Bcl-2 in BEAS-2B cells exposed to 100 μg/cm² of SiO₂ for 24 h relative to the saline control (bottom panel). (b) Semiquantitative analysis of the expression of Wnt signaling proteins Wnt3a, ABC, and DKK1 in (a) by evaluating the relative densitometric densities using arbitrary units (A.U.). (c) Semiquantitative analysis of the expression of fibrogenic factors MMP2, vimentin, α-SMA, and E-cadherin in (a) by evaluating the relative densitometric densities using arbitrary units (A.U.). (d) Semiquantitative analysis of the expression of apoptotic marker cleaved caspase-3 and BAX and antiapoptotic protein Bcl-2 in (a) by evaluating the relative densitometric densities using arbitrary units (A.U.). Data were expressed as the mean ± SEM from three independent experiments. Compared to the CTRL, ^∗∗p < 0.01; ^∗∗∗p < 0.001. WL-ABC: whole cell lysate active beta-catenin; Nu-ABC: nuclear active beta-catenin.

Figure 5

Impacts of Wnt/β-catenin signaling on the expression of NOX4 and ROS production in lung epithelial cells in response to SiO₂ exposure. The lung epithelial BEAS-2B cells were infected with adenoviral vectors at multiplicity of infection of 1000 for 24 h before they were exposed to SiO₂ and cultured for additional 48 h, the abundance of NOX4 and fibrogenic factors were examined by an immunoblotting assay (a–c), and production of ROS (d, e) and GSH (f) was determined by FACS and/or fluorescent staining. (a) Representative images of immunoblots of indicated proteins of interest showed the, respectively, increased and decreased expression of NOX4 and fibrogenic factors in cells infected with AdWnt3a and AdDKK1 regardless of SiO₂ exposure, in comparison with cells infected by AdC control virus (CTRL). (b) Semiquantitative analysis of the expression of Wnt signaling proteins Wnt3a, ABC, DKK1, and NOX4 in (a) by evaluating the relative densitometric densities using arbitrary units (A.U.). (c) Semiquantitative analysis of the expression of fibrogenic factors α-SMA and vimentin in (a) by evaluating the relative densitometric densities using arbitrary units (A.U.). (d) Representative images of cells treated with indicated conditions and stained with 5 mmol/L CellROX® Orange Reagent images displayed more robust ROS staining in cells exposed to SiO₂ compared to cells without SiO₂ exposure; a, respectively, increased and decreased intensive ROS staining was observed in AdWnt3a and AdDKK1-infected cells relative to AdC-infected cells. (e) Quantitative analysis showed that the Wnt3a-mediated activation of Wnt/β-catenin signaling increased the ROS production in lung epithelial cells, regardless of the exposure to SiO₂, and inhibition of the Wnt signaling by AdDKK1 infection dramatically reduced the ROS production, including the SiO₂-induced ROS in these cells, as compared with the AdC-infected cell controls. (f) Effects of Wnt/β-catenin signaling in the production of reduced glutathione (GSH) in lung epithelial cells. In addition to the reduction of GSH by SiO₂, an activation of Wnt/β-catenin signaling by Wnt3a further reduced the production of GSH, and the DKK1-mediated inhibition of Wnt signaling increased the GSH production in lung epithelial cells, regardless of the exposure to SiO₂, as compared with the AdC-infected cells. Data are presented as the mean ± SEM of at least three repeat experiments. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 compared with AdC-infected cells without SiO₂ exposure; ^#p < 0.05, ^##p < 0.01, and ^###p < 0.001 compared to AdC-infected cells with SiO₂ exposure. Bars in (d) 50 μm.

Figure 6

Effects of NOX4 on Wnt/β-catenin signaling and EMT of lung epithelial cells in response to SiO₂ exposure. The lung epithelial BEAS-2B cells were infected with indicated adenoviral vectors at multiplicity of infection of 1000 for 24 h before they were exposed to SiO₂ and cultured for additional 48 h; the expression of Wnt/β-catenin signaling key molecules and fibrogenic factors (a, b) and the production of GSH (c) were examined. (a) Representative images of immunoblots of indicated proteins of interest. An overexpression of NOX4 in BEAS-2B cells resulted in an increased abundance of Wnt/β-catenin signaling proteins Wnt3a, ABC, cyclin D1, and Axin2, but less abundant Wnt inhibitor DKK1, along with an increased expression of fibrogenic factors, and oxidative stress-related proteins HO-1 and Nrf-2. A shRNA-mediated knockdown of NOX4 led to an opposite effect of the overexpression of NOX4. (b) Semiquantitative analysis of the expression of proteins of Wnt signaling cassette (top panel), NOX4, HO-1, and Nrf-2 (bottom left panel) and fibrogenic factors (bottom right panel) in (a) by evaluating the relative densitometric densities using arbitrary units (A.U.). (c) Representative images of IF exhibited an activation of Wnt signaling mediated by NOX4 in BEAS-2B cells regardless of the exposure of SiO₂ as ascertained by the expression of ABC and cell proliferation by EdU incorporation, while shRNA-mediated knockdown of NOX4 acted an opposite effect, as compared with the AdC control. (d) Representative images of ROS staining showed more robust ROS staining in cells infected with AdNOX4, while shRNA-mediated knockdown of NOX4 exhibited an opposite effect to AdNOX4, as compared with the saline and AdC controls. (e) Quantitative analysis showed the NOX4 further SiO₂-induced increased the ROS production in lung epithelial cells infected with AdNOX4, while shRNA-mediated knockdown of NOX4 exhibited an opposite effect to AdNOX4, as compared with the saline and AdC controls. (f) Impacts of NOX4 in the production of reduced glutathione (GSH) in lung epithelial cells. The overexpression of NOX4 further reduced the production of GSH, and the shRNA-mediated suppression of NOX4 increased the GSH production in lung epithelial cells, as compared with the saline treatment and AdC-infected cells. Data are presented as the mean ± SEM of at least three repeat experiments. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 compared with AdC-infected cells without SiO₂ exposure; ^#p < 0.05, ^##p < 0.01, and ^###p < 0.001 compared to AdC-infected cells with SiO₂ exposure.

Figure 7

A schematic mechanism of interaction between Wnt/β-catenin and NOX4 signaling in SiO₂-induced epithelial-mesenchymal transition in lung epithelial cells. The exposure of silica dust activated Wnt/β-catenin signaling and induced NOX4 expression in lung epithelial cells, substantially promoted the expression of Wnt target genes for epithelial cell proliferation and injury repair, and increased ROS production and expression of extracellular matrix (ECM) and fibrogenic factors to promote fibrogenesis. However, the activation of Wnt/β-catenin signaling also induced NOX4 expression that resulted in an overwhelming ROS production and further caused epithelial cell injury. Vice versa, the NOX4 (or ROS) was able to in turn enhanced Wnt/β-catenin signaling which further induced NOX4 expression. In this context, the Wnt/β-catenin signaling and NOX4 formed a positive forward loop to promote lung epithelial injury and EMT, ECM deposition, and fibrogenesis in response to a continuous exposure of silica dust.