The Role of Erythrocyte Membrane Protein Band 4.1-like 3 in Idiopathic Pulmonary Fibrosis

Novel genetic and epigenetic factors involved in the development and prognosis of idiopathic pulmonary fibrosis (IPF) have been identified. We previously observed that erythrocyte membrane protein band 4.1-like 3 (EPB41L3) increased in the lung fibroblasts of IPF patients. Thus, we investigated the role of EPB41L3 in IPF by comparing the EPB41L3 mRNA and protein expression of lung fibroblast between patients with IPF and controls. We also investigated the regulation of epithelial–mesenchymal transition (EMT) in an epithelial cell line (A549) and fibroblast-to-myofibroblast transition (FMT) in a fibroblast cell line (MRC5) by overexpressing and silencing EPB41L3. EPB41L3 mRNA and protein levels, as measured using RT-PCR, real-time PCR, and Western blot, were significantly higher in fibroblasts derived from 14 IPF patients than in those from 10 controls. The mRNA and protein expression of EPB41L3 was upregulated during transforming growth factor-β-induced EMT and FMT. Overexpression of EPB41L3 in A549 cells using lenti-EPB41L3 transfection suppressed the mRNA and protein expression of N-cadherin and COL1A1. Treatment with EPB41L3 siRNA upregulated the mRNA and protein expression of N-cadherin. Overexpression of EPB41L3 in MRC5 cells using lenti-EPB41L3 transfection suppressed the mRNA and protein expression of fibronectin and α-SMA. Finally, treatment with EPB41L3 siRNA upregulated the mRNA and protein expression of FN1, COL1A1, and VIM. In conclusion, these data strongly support an inhibitory effect of EPB41L3 on the process of fibrosis and suggest the therapeutic potential of EPB41L3 as an anti-fibrotic mediator.


Introduction
Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive form of interstitial lung disease of unknown etiology characterized by progressive fibrosis and worsening lung function [1][2][3]. Although the disease course varies, IPF is usually progressive [4][5][6]. Several lines of evidence have suggested that genetic and epigenetic mechanisms play roles in the development and prognosis of IPF [7,8]. Global gene expression studies have identified several novel genes in lung tissues from IPF patients [9][10][11][12]. We also revealed that 178 of 15,020 genes are differentially expressed in lung fibroblasts from IPF patients compared with those from controls [13]. Among them, the mRNA level of erythrocyte membrane protein band 4.1-like 3 (EPB41L3) was 14-fold higher in IPF patients than controls.
EPB41L3 (also called DAL-1/4.1B) is on 18p11.31 and is believed to enable cytoskeletal protein-membrane anchor activity, cytoskeletal rearrangements, intracellular transport, and signal transduction [14]. EPB41L3 inhibits the progression and development of several cancers, including lung adenocarcinoma, meningioma, breast cancer, ovarian cancer, and prostate cancer [14]. Epithelial-mesenchymal transition (EMT) and fibroblastto-myofibroblast transition (FMT) are involved in the development and progression of cancer [15]. Multiple signaling pathways play roles in EMT and FMT [14]. The extracellular matrix (ECM) is degraded during the late stages of EMT through the increased expression of proteases, such as matrix metalloproteinases (MMPs), which play a crucial role in cancer metastasis and angiogenesis [16].
Several studies have revealed the involvement of EPB41L3 in EMT and FMT. Transforming growth factor (TGF)-β increased the mRNA and protein expression of EPB41L3 by the A549 cell line, and deleting EPB41L3 attenuated the TGF-β-induced EMT in a lung cancer cell line [17]. In addition, EPB41L3 suppressed tumor cell invasion and inhibited MMP2 and MMP9 expression in an esophageal squamous cancer cell line [18]. Most studies have focused on cancer. However, few have been conducted on pulmonary fibrosis. TGF-β1 expression is particularly high in fibrotic areas of the lungs of patients with IPF, and activated TGF-β1 drives ECM components, such as COL1A1 (collagen 1), FN1 (fibronectin), and ACTA2 (α-SMA), to generate a profibrotic environment in the injured area [19][20][21][22][23]. Therefore, regulating TGF-β1 activity with EPB41L3 is a potential therapeutic strategy for fibrosis in patients with IPF. In the present study, the role of EPB41L3 was investigated by comparing mRNA and protein expression between lung fibroblasts derived from patients with IPF and those of controls and by regulating the EMT of an epithelial cell line (A549) and FMT of a fibroblast cell line (MRC5) by overexpressing and silencing EPB41L3.

Changes in mRNA and Protein Expression of EPB41L3 and EMT-Related Genes by A549 Cells after Stimulation with TGF-β1
After A549 cells were cultured in tissue culture medium (TCM) containing 5 ng/mL of TGF-β1 for 24, 48, and 72 h, EPB41L3, N-cadherin, E-cadherin, COL1A1, and β-actin were quantified using real-time PCR. A 2-fold increase in the EPB41L3-to-β-actin mRNA ratio was observed after stimulation with TGF-β1 in a time-dependent manner. Moreover, the COL1A1and N-cadherin-to-β-actin mRNA expression ratios were 10-and 3-fold higher, respectively. In contrast, the E-cadherin-to-β-actin mRNA expression ratio decreased significantly after stimulation with TGF-β1 in a time-dependent manner ( Figure 2A). Western blot demonstrated similar changes in protein expression ( Figure 2B). The EPB41L3-, Ncadherin-, and collagen-1-to-β-actin protein expression ratios increased significantly after stimulation with TGF-β1 in a time-dependent manner, while the E-cadherin-to-β-actin ratio decreased significantly ( Figure 2C).   Figure S1 for RT-PCR and original Western blot images.)

Changes in mRNA and Protein Expression of EPB41L3 and EMT-Related Genes by A549 Cells after Stimulation with TGF-β1
After A549 cells were cultured in tissue culture medium (TCM) containing 5 ng/mL of TGF-β1 for 24, 48, and 72 h, EPB41L3, N-cadherin, E-cadherin, COL1A1, and β-actin were quantified using real-time PCR. A 2-fold increase in the EPB41L3-to-β-actin mRNA ratio was observed after stimulation with TGF-β1 in a time-dependent manner. Moreover, the COL1A1-and N-cadherin-to-β-actin mRNA expression ratios were 10-and 3-fold higher, respectively. In contrast, the E-cadherin-to-β-actin mRNA expression ratio decreased significantly after stimulation with TGF-β1 in a time-dependent manner ( Figure 2A). Western blot demonstrated similar changes in protein expression ( Figure 2B). The EPB41L3-, Ncadherin-, and collagen-1-to-β-actin protein expression ratios increased significantly after stimulation with TGF-β1 in a time-dependent manner, while the E-cadherin-to-β-actin ratio decreased significantly ( Figure 2C).  Basal levels (0 hour) were excluded from statistical analysis. Significant differences between groups were evaluated using two-way ANOVA with Bonferroni multiple comparisons test. * p < 0.05 and ** p < 0.01 compared to TGF-β1-stimulated cells. Black bars: untreated control group, bars with diagonals: TGF-β1-treated group. (See Figure S2 for original Western blot images.)

Effect of Overexpressing EPB41L3 on Epithelial-Mesenchymal Transition in A549 Cells
A549 cells were transfected with lenti-EPB41L3 or the control lentiviral vector. The mRNA levels of EPB41L3 (normalized to β-actin) measured using real-time PCR increased 48 h after transfection of lenti-EPB41L3 compared with those of the lentiviral vectors in the absence (4-fold) and presence (3-fold) of 5 ng/mL of TGF-β1 ( Figure 3(A-1)). Concomitantly, 5 ng/mL of TGF-β1 increased the N-cadherin and COL1A1 mRNA levels, which were significantly downregulated by transfected lenti-EPB41L3 ( Figure 3(A-2,A-4)). E-cadherin mRNA tended to be decreased by lenti-EPB41L3 ( Figure 3(A-3)). The protein levels Figure 2. Changes in mRNA and protein of EPB41L3 and EMT-related genes expressed by A549 cells after stimulation with TGF-β. A549 cells were cultured with or without 5 ng/mL of TGF-β1 for 24, 48, and 72 h. The EPB41L3-, N-cadherin-, E-cadherin-, and COL1A1-to-β-actin ratios were measured using (A) real-time PCR, (B) Western blot, and (C) densitometry of protein bands. They were normalized to β-actin and expressed as ratios. All experiments were performed separately six times, and the representative images are shown. Data are mean ± SE of the six independent experiments. Basal levels (0 h) were excluded from statistical analysis. Significant differences between groups were evaluated using two-way ANOVA with Bonferroni multiple comparisons test. * p < 0.05 and ** p < 0.01 compared to TGF-β1-stimulated cells. Black bars: untreated control group, bars with diagonals: TGF-β1-treated group. (See Figure S2 for original Western blot images).

EPB41L3 Gene Silencing during the EMT of A549 Cells
Treatment with siRNA (100 nM) for 48 h decreased the mRNA expression of EPB41L3 (normalized to the β-actin level) in the presence and absence of 5 ng/mL of TGF-β1 ( Figure  4A). Concomitantly, EPB41L3 siRNA significantly decreased the mRNA level of EPB41L3

EPB41L3 Gene Silencing during the EMT of A549 Cells
Treatment with siRNA (100 nM) for 48 h decreased the mRNA expression of EPB41L3 (normalized to the β-actin level) in the presence and absence of 5 ng/mL of TGF-β1 ( Figure 4A). Concomitantly, EPB41L3 siRNA significantly decreased the mRNA level of EPB41L3 to the untreated level ( Figure 4A). Western blot revealed similar changes in the protein levels ( Figure 4B). The protein level of EPB41L3 was decreased significantly, while the N-cadherin-to-β-actin protein expression ratio was increased significantly by EPB41L3 siRNA in the absence or the presence of TGF-β1 (Figure 4(C-1,C-2)), while the E-cadherin-to-β-actin ratio was not significantly changed (Figure 4(C-3)). siRNA in the absence or the presence of TGF-β1 (Figure 4(C-1,C-2)), while the E-cadherinto-β-actin ratio was not significantly changed (Figure 4(C-3)).

Effect of EPB41L3 Overexpression and Knockdown on the FMT in MRC5 Cells
Treatment with 5 ng/mL of TGF-β1 significantly increased the mRNA expression of FN1, ACTA2, N-cadherin, and COL1A1 (normalized to the β-actin level) of MRC-5 cells in a time-dependent manner ( Figure 5(A-2-A-5)). The protein levels of these genes showed the same pattern of change ( Figure 5B). The protein expression of EPB41L3, fibronectin, α-SMA, and collagen 1 increased in a time-dependent manner ( Figure 5(C-1-C-4)). When EPB41L3 was transfected into MRC5 cells with lenti-EPB41L3, the mRNA and protein expression of EPB41L3 increased 48 h after transfection (Figure 6(A-1,C-1)). Concomitantly, the mRNA and protein expression of FN1 (Figure 6(A-2,C-2)) and ACTA2 (Figure 6(A-3,C-3)) significantly decreased after transfection with lenti-EPB41L3 in the absence and presence of TGF-β1. Treatment with EPB41L3 siRNA (100 nM) for 48 h decreased the EPB41L3 Figure 4. Effect of silencing EPB41L3 on EMT-related genes in A549 cells. Changes of EPB41L3 expression in response to EPB41L3 siRNA in the presence and absence of 5 ng/mL of TGF-β1 for 48 h. Expression of mRNA and protein normalized to the β-actin level was measured using (A) real-time PCR, (B) Western blot, and (C-1-C-3) densitometry of protein bands. Independent experiments were analyzed using densitometry. They were normalized to β-actin and expressed as ratios. All experiments were performed separately six times, and the representative images are shown. Data are mean ± SE of 6 independent experiments. Significant differences between groups were evaluated using two-way ANOVA with Bonferroni multiple comparisons test. * p < 0.05. Black bar: control group, open bar: scramble group, Gray bar: EPB41L3-knockdown group. (See Figure S4 for original Western blot images).

Discussion
In the present study, the protein and mRNA expression of the EPB41L3 gene significantly increased in lung-tissue-derived fibroblasts from IPF patients compared to those from controls. This result is in line with the transcriptomic datasets of single-cell RNA-seq demonstrating the upregulation of EPB41L3 in the lung fibroblasts of IPF patients [24]. In A549 and MRC5, the overexpression of EPB41L3 suppressed EMT-and FMT-related genes, while EPB41L3 knockdown upregulated these genes. These data show for the first time that the EPB41L3 gene may play a role in the development of pulmonary fibrosis by regulating EMT and FMT.
EPB41L3 has been regarded as a tumor suppressor that inhibits the progression and development of several types of cancer, including lung adenocarcinoma, meningioma, breast cancer, ovarian cancer, and prostate cancer [25,26]. Loss of the 4.1B/DAL-1 protein Figure 7. Effect of silencing EPB41L3 on FMT-related genes in MRC5 cells. Changes in the expression of EPB41L3 and FMT-related genes in response to EPB41L3 siRNA. Expression of mRNA and protein were measured using (A-1-A-4) real-time PCR, (B) Western blot, and (C-1-C-4) densitometry of protein bands. Independent experiments were analyzed using densitometry. They were normalized to β-actin and expressed as ratios. All experiments separately were performed six times, and the representative images are shown. Data are mean ± SE of 6 independent experiments. Significant differences between groups were evaluated using two-way ANOVA with Bonferroni multiple comparisons test. * p < 0.05. Black bar: control group, open bar: scramble group, gray bar: EPB41L3-knockdown group. (See Figure S7 for original Western blot images).

Discussion
In the present study, the protein and mRNA expression of the EPB41L3 gene significantly increased in lung-tissue-derived fibroblasts from IPF patients compared to those from controls. This result is in line with the transcriptomic datasets of single-cell RNA-seq demonstrating the upregulation of EPB41L3 in the lung fibroblasts of IPF patients [24]. In A549 and MRC5, the overexpression of EPB41L3 suppressed EMT-and FMT-related genes, while EPB41L3 knockdown upregulated these genes. These data show for the first time that the EPB41L3 gene may play a role in the development of pulmonary fibrosis by regulating EMT and FMT.
EPB41L3 has been regarded as a tumor suppressor that inhibits the progression and development of several types of cancer, including lung adenocarcinoma, meningioma, breast cancer, ovarian cancer, and prostate cancer [25,26]. Loss of the 4.1B/DAL-1 protein leads to a substantial decrease in the expression of numerous EMT markers, including E-cadherin and β-catenin in lung cancer cell lines [27], which agrees with our EPB41L3 knockdown data. EPB41L3 maintains the epithelial cell phenotype and attenuates EMT by inhibiting Snail and PI3K/Akt/Mdm2/p53 signaling [14,27,28]. In a study using 83 osteosarcoma tissues [14], the expression of Snai1/Slug/Twist1 was correlated with that of EPB41L3. Thus, EPB41L3 may inhibit the progression of EMT by interacting with Snail1, 2, and Twist1 in patients with pulmonary fibrosis.
The mechanism underlying the increase in EPB41L3 in IPF has not been revealed. In our previous transcriptome analysis of 12 lung fibroblast samples derived from 4 controls and 8 IPF-patients (GSE71351), strong correlations were observed between the EPB41L3 levels and those of 344 among 15,020 genes (160 positive and 184 negative correlations) (Table S1). Among them, ZNF678, which is involved in transcriptional regulation, interacted with EPB41L3 in the STRING protein-protein interaction network analysis ( Figure S8). Among the 99 proteins interacting with ZNF687, 33 proteins, including MYC [29], CSNK2A1 [30], CDKN2A [31], HDAC2 [32], MI1 [33], SOX2 [34], and RNF2 [35], have been revealed to be associated with fibrosis. These data indicate that EPB41L3 may be regulated through interactions with ZNF687 and interacting genes.
In addition, CpG methylation is a common mechanism of genetic downregulation. The hypermethylation of DAL-1 is strongly correlated with the loss of DAL-1 and predicts short overall survival in patients with non-small cell lung cancer (NSCLC) [36]. The hypermethylation of CpG sites in the EPB41L3 promoter region leads to loss of expression, as observed in esophageal cancer [37]. In our previous methylation study (GSE107226) using the same fibroblasts [38], 21 CpG sites were identified on the EPB41L3 gene (5 on TSS1500, 5 on TSS200, 9 on the 5 UTR, 1 on the body, and 1 on the 3 UTR) ( Figure S9 and Table S2). Among them, different methylation levels were detected at four CpG sites (cg15528411, cg22335490, cg27082185, and cg14075742) in the fibroblasts of IPF patients compared with the controls. Significant positive correlations were observed at one CpG site (cg27082185), and a negative correlation was observed at one CpG site (cg00027083) compared with the transcriptome levels ( Figure S9 and Table S2). Although we did not conduct a functional study of these CpG sites, different CpG methylation sites may be responsible for the changes seen in the EPB41L3 transcriptome levels.
Our study has several limitations. First, control fibroblasts were obtained from normal portions of the resected cancer specimens. The gene expression profile of fibroblasts derived from lungs in which cancer developed may be different from that of fibroblasts derived from truly normal lungs. Secondly, we used the A549 and MRC-5 cell lines in the EMT and FMT studies instead of primary lung epithelial cells or fibroblasts. Thirdly, the EPB41L3 gene and protein levels were not measured using the samples, such as bronchoalveolar lavage or lung tissues. To reveal the exact role of EPB41L3 in IPF, the protein and gene levels of EPB41L3 should be evaluated in the lungs of the patients with IPF in terms of clinical manifestations by assessing the correlations of their levels with prognostic parameters, such as the longterm survival rate, in large number of patients. Additionally, one important limitation of our study is the lack of the functional study of EPB41L3 in IPF-related cellular processes, such as morphological changes, cell proliferation, migration, and invasion. Therefore, further functional studies are needed using EPB41L3 overexpressed and knocked cell lines.

Study Subjects
Lung fibroblasts and lung tissues of patients with IPF were obtained from the Biobank of Soonchunhyang University Hospital (Bucheon, South Korea) after approval of the study protocol by the Institutional Review Board of Soonchunhyang University (201910-BR-058). The lung fibroblasts were cultured from the surgical specimens of 14 patients with IPF and the normal lungs of 10 subjects, who underwent surgery to remove stage I or II lung cancer; their clinical characteristics were described previously [13]. The diagnostic criteria for IPF were based on 2011 and 2018 international consensus statements [39,40].

Lentiviral Transduction to Overexpress EPB41L3 in A549 and MRC5 Cells
Stable transfection was performed using the Open Reading Frame (ORF) Lentiviral Transduction Particles kit (ORIGENE, Rockville, MD, USA) according to the manufacturer's instructions. The EPB41L3 Human Tagged ORF (Cat: RC209736L3V; ORIGENE) and corresponding control lentiviral particles (Cat: PS100092V; ORIGENE) were used for stable transfection. Lentiviral transduction was performed when the cells reached 70% confluence. EPB41L3 Human Tagged ORF clone-expressing lentivirus or lenti-ORF control virus particles were added to the cells at a multiplicity of infection of 5 for A549 cells and 1 for MRC5 cells, along with 8 µg/ml polybrene (Sigma-Aldrich). The cells were maintained in 5 µg/mL puromycin (Sigma-Aldrich) for approximately 10 days to establish stable cell lines. Then, mRNA and protein overexpression of EPB41L3 was determined using real-time PCR and Western blot. 4.6. EPB41L3 Gene Silencing Using siRNA siRNAs against EPB41L3 and scrambled siRNA (Bioneer, Daejeon, South Korea) were transfected into A549 and MRC5 cells using Lipofectamine 2000 (Invitrogen), according to the manufacturer's protocol, when the cells were 70% confluent. The siRNA sequences used were as follows: siEPB41L3-sense 5 -GUG AAG ACG GAA ACC AUC A-3 and antisense 5 -UGA UGG UUU CCG UCU UCA C-3 . The siRNAs (100 nM) were incubated in Opti-MEM (Gibco) solution for 5 min; 4 µL of Lipofectamine 2000 was mixed into the solution, which was then incubated for 20 min. The mixtures were added to the cells and incubated for 6 h at 37 • C. The cells were cultured in 1% FBS-TCM with or without TGF-β1 (5 ng/mL) for 48 h. Knockdown efficiencies were quantified using real-time PCR and Western blot.

Statistical Analysis
Data were analyzed using SPSS (ver. 20.0; SPSS, Inc., Chicago, IL, USA). The distribution of data was assessed using the Shapiro-Wilk test, and Mann-Whitney U test was used for the comparison of non-parametric data between two groups. Two-way ANOVA with Bonferroni's post hoc test for multiple comparisons were used for comparison of parametric samples. Skewed data were expressed as medians with 25% and 75% quartiles, and normally distributed data were expressed as the means ± standard error of the mean. p-values < 0.05 were considered significant.

Conclusions
Fibroblasts from the lungs of IPF patients expressed high mRNA and protein levels of the EPB41L3 gene compared with controls. EPB41L3 gene and protein expression was induced during EMT and FMT. Overexpressing EPB41L3 suppressed EMT-and FMT-related genes and proteins, while silencing EPB41L3 upregulated them. These data demonstrate the inhibitory role of EPB41L3 in the fibrosing process. As TGF-β levels increase in the lungs of IPF patients [41], the phenotypes of the fibroblasts may be changed by long-term stimulation with an excessive amount of TGF-β in the lungs of IPF patients. Thus, a balance between the amount of TGF-β (fibrotic factor) and EPB41L3 (anti-fibrotic factor) may determine the progress of fibrosis in patients with IPF.