Eight- to ten-week-old specific pathogen-free C57BL/6 male mice were maintained under specific pathogen-free conditions, and offered free access to sterile water and food. Their weight ranged between 20 and 25 g. Animal experiments were approved by the Institutional Animal Care and Use Committee of Jining Medical University.

After a week of adaptive feeding, the mice were randomly assigned to a sham (n = 10) or BDL group (n = 20). BDL was performed as previously described [20]. Briefly, the mice were anesthetized with 2% pentobarbital sodium (40 mg/kg) (Sigma-Aldrich, St. Louis, USA). We first isolated the common bile duct and the left and right hepatic ducts, and then ligated the left and right hepatic ducts and the hepatic portal and duodenal portions of the common bile duct, respectively. At last, the abdomen was closed. The sham group was used as a control and underwent laparotomy without ligation. After 7 or 28 days, the mice were killed and their liver tissues were harvested. It was further manipulated for qRT-PCR, western blotting, histological, and immunohistochemical analyses.

Paraformaldehyde-fixed liver tissue samples were cut into 4-$\mu$m-thick paraffin-embedded sections and stained with hematoxylin-eosin. They were viewed under a light microscope to assess the fibrosis stage according to the METAVIR scoring system [21]. Before immunostaining, antigen retrieval was performed with 10 mM sodium citrate buffer in microwave. Sequentially, permeabilization was performed with TritonX-100 (Solarbio, Beijing, China). Slices were incubated with 5% Bovine Serum Albumin (BSA) in phosphate buffer saline (PBS) for 30 min at room temperature and then incubated at 37 °C for 1 h with a primary anti-MEF2A antibody (Abcam, Cambridge, UK, 1:500). After rinsing 3 times with PBS, slices were incubated with fluorescently conjugated secondary antibodies at room temperature for 15 min. The sections were washed 5 times in PBS and 3 times in 5% BSA, followed by adding 15 $\mu$L mounting media containing DAPI stain and add a coverslip. Scanning of slices was performed using a laser scanning confocal microscope (Zeiss LSM800, Oberkochen, Germany).

Lentiviral vectors expressing an shRNA targeted against the murine MEF2A transcript (LV-shMEF2A) or negative control (LV-NC) were constructed by GenePharma (China). The shMEF2A targeting sequence was 5′-GCAGCCAGCTCAACGTTAACA-3′, and that of the negative control was 5′-TTCTCCGAACGTGTCACGT-3′.

Human intrahepatic biliary epithelial cells (HIBECs) were purchased from the BeNa Culture Collection (China). They were cultured in RPMI 1640 medium (Gibco-BRL) supplemented with 10% FBS (Gibco-BRL) at 37 °C and 5% CO${}_2$. The cells (1 $\times$ 10${}^5$ per well) were transfected with LV-shMEF2A or LV-NC following the manufacturer’s protocol (multiplicity of infection = 30). In brief, LV-shMEF2A or LV-NC was mixed with 10 mg/mL polybrene and added to the cell suspension. The cell culture medium was replaced with fresh RPMI 1640 12 to 16 h after transfection, and the transfected cells were cultured continuously. Green fluorescent protein was detected by fluorescence microscopy to assess the transfection efficiency 72 h after transfection.

Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, USA). Complementary DNA (cDNA) was reverse transcribed with an All-In-One 5$\times$ RT reagent kit (abm, Zhenjiang, Jiangsu, China) following the manufacturer’s protocol, and was stored at –20 °C until use. The quantitative PCR assay was performed using an UltraSYBR Mixture (CWBIO, Taizhou, Jiangsu, China) with the following conditions: 1 min at 95 °C, 40 cycles at 95 °C, 15 s; and 40 cycles at 60 °C, 30 s. The primers were synthesized by ShengGong Biotech (Shanghai, China), and their sequences are listed in Table 1. The relative expression of GAPDH was used as an endogenous reference to quantify the relative mRNA expression of target genes. The transcript level of each gene was calculated using the 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ method.

Cells were lysed using a RIPA lysis buffer containing 10% phosphatase and proteinase inhibitor (Beyotime, Shanghai, China). An equal amount of protein was loaded and separated on 10% sodium dodecyl sulphate-polyacrylamide gels. Proteins were transferred to polyvinylidene fluoride membranes (Millipore, Bedford, MA, USA). They were blocked with 5% non-fat milk for 2 h at room temperature and incubated overnight at 4 °C with primary antibodies: anti-MEF2A (Abcam, 1:1000), anti-E-cadherin (Abcam, 1:1000), anti-N-cadherin (Abcam, 1:1000), anti-vimentin (Abcam, 1:1000), anti-$\alpha$-SMA (Abcam, 1:1000), anti-P21 (Abcam, 1:1000), anti-p38 (Danvers, MA, USA, 1:1000), anti-JNK (CST, 1:1000), anti-ERK (CST, 1:1000), or anti-GAPDH (Boster Bio, Shanghai, China, 1:1000). The HRP-conjugated secondary antibodies were added to the membranes and incubated for 2 h at room temperature. The membranes were visualized in a drak room using an ECL reagent (Beyotime) following the manufacturer’s instructions.

The SA-$\beta$-Gal activity was assessed using a senescence $\beta$-galactosidase staining kit (Beyotime) by the manufacturer’s instructions. Briefly, the cells were washed with PBS and fixed with SA-$\beta$-Gal staining fixative for 15 min at room temperature. After rewashing twice with PBS, the SA-$\beta$-Gal staining solution was added, and the cells were incubated overnight at 37 °C in the dark. The cells were observed under a light microscope, and the senescent cells were identified by blue color.

Cells were fixed with 4% paraformaldehyde for 15 min at room temperature. They were permeabilized with 0.01% Triton X-100 (Solarbio, Beijing, China) for 5 min and blocked in 10% fetal bovine serum for 1 h at room temperature. For the staining, the cells were incubated with a primary rabbit anti-$\alpha$-SMA (Abcam, 1:500) overnight at 4 °C, and then incubated with an Alexa Fluor 488- or 590-conjugated secondary antibody (Santa Cruz Biotechnology, California, USA, 1:200) for 1 h. Nuclei were visualized with 4′-6-diamidino-2-phenylindole (Beyotime, 5 mg/mL), and images were captured using a confocal microscope (Zeiss LSM800, Oberkochen, Germany, Germany).

Serum MEF2A levels of 15 patients with PBC and 15 healthy controls (HCs) was analyzed in this study. All participants were inpatients from the Department of Gastroenterology, Affiliated Hospital of Jining Medical University, China. All fulfilled the requirements for PBC diagnosis specified in the European Association for the Study of the Liver clinical practice guidelines [22]. The participants gave their written informed consent to partake in the study. The Institutional Review Board for Clinical Research of the Affiliated Hospital of Jining Medical University approved the study. Clinical characteristics of included patients are shown in Table 2.

Serum MEF2A levels were tested using a human MEF2A ELISA kit (FineTest, China). Peripheral blood was collected in EDTA anti-coagulated tubes and centrifuged at 1000 $\times$ g and 4 °C for 20 min. The supernatant was aspirated, and the assay was performed following the manufacturer’s instructions. Briefly, the capture antibody was incubated in 96-well plates at 4 °C overnight. After blocking with assay diluents, the standards and samples were added to the designated wells and incubated at 37 °C for 2 h. The plates were incubated with the detection antibody for 1 h and Horseradish Peroxidase (HRP) for 30 min. They were washed with 0.05% Tween-PBS before adding the 3,3',5,5'-tetramethylbenzidine sulfate (TMB) substrate to develop color. Absorbance was measured using a microplate spectrophotometer (BioTek, Vermont, USA).

All experiments were performed in triplicates, and data were expressed as the mean $\pm$ standard deviation (SD). Differences between two groups were analyzed with the t test. A one-way ANOVA was recruited to compare differences among multiple groups. Statistical Package for Social Sciences (v 20.0) (Chiago, IL, USA) was used for all data analyses. Significant differences were considered when p 0.05.

Bile duct-ligated mice are a widely used model to study cholestatic liver fibrosis, including PBC. Therefore, we examined the expression of MEF2A in the livers of BDL mice. They were sacrificed 7 or 28 days after BDL surgery, and their liver tissues were harvested. Histological examination of liver sections showed that BDL surgery induced liver fibrosis in mice (Fig. 1A), confirming the successful establishment of the BDL model. Fluorescent immunohistochemistry analysis localized MEF2A expression in the liver and revealed that the percentage of MEF2A${}^{+}$ cells in BDL-mice was markedly higher than the control (Fig. 1B,C). We also examined protein levels of MEF2A and TGF-$\beta$1 in the liver by western blotting. In BDL-mice liver tissues, which were substantially increased compared with those of the control (Fig. 1D). Moreover, MEF2A mRNA levels were consistent with protein levels (Fig. 1E).

Fig. 1.

MEF2A is overexpressed in livers of BDL mice and positively correlates with degree of fibrosis. Bile duct-ligated model was established using C57BL/6 mice to induce liver fibrosis. Sham mice were used as control. Mice were sacrificed 7 or 28 days after BDL surgery, and liver tissues were harvested. (A) Histological examination of liver sections with H&E staining. Scale bar represents 50 $\mu$m. (B) Expression of MEF2A in liver tissues of BDL and sham mice detected with immunohistochemical staining using second antibodies labeled with fluorochromes. Original magnification $\times$ 200. Scale bar represents 50 $\mu$m. (C) Percentages of MEF2A${}^{+}$ cells in mice livers shown in (B). **p 0.01, ***p 0.001. (D) MEF2A and TGF-$\beta$1 protein expression in liver tissues of BDL and sham mice identified by western blotting. Reference protein was GAPDH. (E) Relative MEF2A mRNA expression in liver tissues of BDL and sham mice examined by qRT-PCR. Gene expression was normalized to GAPDH in each group. **p 0.01, ***p 0.001.

Evidence suggests that TGF-$\beta$1 induces EMT in cultured cholangiocytes [23]. Hence, we assessed whether silencing MEF2A affects TGF-$\beta$1-induced EMT. We treated HIBECs with TGF-$\beta$1 or in combination with a lentiviral vector expressing shRNA against MEF2A (LV-shMEF2A) or the empty vector (LV-NC). Western blotting and qRT-PCR assays verified the silencing of MEF2A with approximately 50% knock-down efficiency compared with control and LV-NC groups (Fig. 2). In addition, TGF-$\beta$1 significantly decreased the expression of E-cadherin but increased N-cadherin and vimentin, which were significantly altered in the presence of LV-shMEF2A (Fig. 2). These results indicate that MEF2A acts as a positive modulator of EMT in HIBECs.

Fig. 2.

Silencing MEF2A inhibits TGF-$\beta$1-induced EMT in HIBECs. HIBECs were cultured and treated with TGF-$\beta$1 (15 ng/mL) or in combination with LV-shMEF2A (lentiviral vector expressing shRNA against MEF2A) or LV-NC (empty vector). (A) Expression of MEF2A protein and EMT markers in HIBECs assessed by western blotting. Reference protein was GAPDH. (B) Quantification of MEF2A and EMT markers in cells depicted in (A). *p 0.05, **p 0.01. (C) Relative expression of MEF2A and EMT markers evaluated by qRT-PCR. Gene expression was normalized to GAPDH in each group. *p 0.05, **p 0.01, ***p 0.001.

Cholangiocyte senescence contributes to the pathogenesis of PBC and liver fibrosis [24]. HIBECs were transfected with TGF-$\beta$1 in combination with LV-shMEF2A or LV-NC to investigate the effect of MEF2A on their senescence. About 96 h post-transfection, cells were stained with SA-$\beta$-gal assay, and those positive for SA-$\beta$-gal activity were determined. TGF-$\beta$1 substantially increased the number of senescent HIBECs than that in the control group. When MEF2A was silenced, the proportion of the positive cells in the LV-shMEF2A-transfected group was significantly lower than that in LV-NC-transfected group (Fig. 3A). Thus, silencing MEF2A decreases the number of senescent HIBECs. Moreover, qRT-PCR and western blotting revealed that transcript and protein levels of p21, a marker of senescence [25], significantly decreased after the MEF2A knockdown (Fig. 3B,C). These data suggest that MEF2A silencing reduces HIBEC senescence.

Fig. 3.

Silencing MEF2A decreases TGF-$\beta$1-induced cellular senescence in HIBECs. HIBECs were cultured and treated with TGF-$\beta$1 (15 ng/mL) or in combination with LV-shMEF2A (lentiviral vector expressing shRNA against MEF2A) or LV-NC (empty vector). Cells were harvested 96 h posttreatment to determine senescence. (A) SA-$\beta$-galactosidase staining of the senescent cells (blue). Original magnification $\times$ 100. Scale bar = 50 $\mu$m. (B) Expression of p21, a senescence marker, in HIBECs detected by western blotting. Reference protein was GAPDH. *p 0.05, **p 0.01. (C) Quantification of p21 levels in cells shown in (B). (D) Relative expression of p21 mRNA evaluated by qRT-PCR. Gene expression was normalized to GAPDH in each group. *p 0.05, **p 0.01.

TGF-$\beta$1 is considered the most potent cytokine that maintains the fibrogenic response in the liver. We treated HIBECs with TGF-$\beta$1 or in combination with LV-shMEF2A or LV-NC to explore whether MEF2A affects the TGF-$\beta$1-promoted fibrogenic response in these cells. Expression levels of profibrogenic marker $\alpha$-SMA were examined by qRT-PCR, western blotting, and immunofluorescence assays. The TGF-$\beta$1 treatment significantly increased protein and mRNA levels of $\alpha$-SMA in HIBECs compared with the control (Fig. 4A–C). Conversely, MEF2A knockdown significantly decreased the levels of $\alpha$-SMA induced by TGF-$\beta$1 (Fig. 4A–C). Immunofluorescence staining of HIBECs also demonstrated elevated $\alpha$-SMA protein under the TGF-$\beta$1 treatment and its reduction upon MEF2A knockdown (Fig. 4D). These results imply that MEF2A promotes fibrogenesis in HIBECs.

Fig. 4.

Silencing MEF2A alleviates fibrogenesis in HIBECs. HIBECs were cultured and treated with TGF-$\beta$1 (15 ng/mL) or in combination with LV-shMEF2A (lentiviral vector expressing shRNA against MEF2A) or LV-NC (empty vector). (A) Expression of $\alpha$-SMA protein in HIBECs identified by western blotting. (B) Quantification of $\alpha$-SMA protein levels in cells shown in (A). (C) Relative expression of $\alpha$-SMA mRNA in HIBECs evaluated with qRT-PCR. *p 0.05, **p 0.01 vs. negative control. (D) Immunofluorescence assay demonstrating $\alpha$-SMA expression in HIBECs. Original magnification $\times$ 200. Scale bar = 50 $\mu$m. Red, $\alpha$-SMA${}^{+}$ cells; blue, DAPI-counterstained nuclei.

TGF-$\beta$1 is a central regulator of EMT, senescence, and liver fibrosis. We stimulated HIBECs with TGF-$\beta$1 to determine whether it could induce MEF2A expression and cytoplasm translocation. We determined its expression and subcellular localization with qRT-PCR, western blotting and immunofluorescence assays. The TGF-$\beta$1 treatment increased the MEF2A expression in HIBECs in a concentration- and time-dependent manner (Fig. 5A–F). Furthermore, TGF-$\beta$1 induced its translocation from the nucleus to the cytoplasm (Fig. 5G,H).

Fig. 5.

Regulating MEF2A expression and cytoplasm translocation by TGF-$\beta$1 in HIBECs. (A–C) HIBECs were treated for 48 h with TGF-$\beta$1 at various concentrations: 0, 5, 10, and 15 ng/mL. Expression levels of MEF2A protein and mRNA were examined across different TGF-$\beta$1 concentrations by western blotting and qRT-PCR. *p 0.05, **p 0.01 vs. control. (D–F) HIBECs were treated with TGF-$\beta$1 (15 ng/mL) at various time points: 0, 12, 24, 48, and 72 h. MEF2A protein and mRNA expression levels were analyzed at each time point by western blotting and qRT-PCR. *p 0.05, **p 0.01 vs. control. (G) Immunofluorescence analysis of MEF2A localization in HIBECs. Red, MEF2A; blue, DAPI-counterstained nuclei (scale bar = 50 $\mu$m). (H) Western blotting assay demonstrating MEF2A levels in cytoplasm and nuclei of HIBECs.

The MAPK pathway may play a role in the pathogenesis of tissue fibrosis, acting downstream of TGF-$\beta$1 [26], and p38 directly activates MEF2A, promoting its nuclear activation [27]. Hence, we tested the roles of MAPK signalings (p38, JNK, and ERK) in TGF-$\beta$1-induced MEF2A in HIBECs. The results showed that TGF-$\beta$1 causes an increase in the levels of phosphorylated p38 (p-p38), but not in that of p-JNK or p-ERK, demonstrating activation of p38 MAPK (Fig. 6A). We further examined whether altering the p38 MAPK signaling could modulate the TGF-$\beta$1-induced MEF2A expression, EMT, and senescence in HIBECs. Indeed, inhibiting phosphorylation of p38 by its inhibitor SB 203580 was sufficient to prevent the increase of MEF2A, p21, $\alpha$-SMA and suppress the decrease of E-cadherin induced by TGF-$\beta$1 (Fig. 6B). These data suggest that suppressing the TGF-$\beta$1-activated p38 MAPK signaling contributes to the anti-EMT and anti-senescence properties of MEF2A inhibition in HIBECs.

Fig. 6.

Regulation of the expression of MEF2A, E-cadherin, p21, and $\alpha$-SMA by TGF-$\beta$1-p38 MAPK in HIBECs. (A) HIBECs were treated with TGF-$\beta$1 for different lengths of time (0, 30 min, and 60 min). Levels of phosphorylated p38 (p-p38), total p38 (t-p38), p-JNK, t-JNK, p-ERK and t-ERK were determined by western blotting. (B) Western blotting assay showing p-p38, p38, MEF2A, E-cadherin, p21, and $\alpha$-SMA levels in HIBECs after TGF-$\beta$1 treatment with or without SB 203580 inhibitor.

Since MEF2A plays a key role in the fibrogenic response, we aimed to answer whether it contributes to PBC fibrosis. We examined the expression of MEF2A in the serum of patients with PBC and found its levels were significantly high in the serum of patients versus HCs (Fig. 7A). To assess whether serum MEF2A levels were also associated with the extent of liver fibrosis, we determined the liver stiffness measurement (LSM). We used a transient elastographic instrument for the noninvasive evaluation of liver fibrosis called Fibroscan [28]. Next, we analyzed the correlation of serum MEF2A expression with LSM in patients with PBC. Interestingly, MEF2A expression in serum from PBC patients positively correlated with LSM (Fig. 7B). These data indicate that MEF2A expression in serum of patients with PBC positively correlates with the extent of liver fibrosis.

Fig. 7.

MEF2A expression increases in patients with PBC and correlates with the extent of liver fibrosis. (A) Peripheral blood samples were collected from patients with PBC (n = 15) and healthy controls (HCs) (n = 15). Expression of serum MEF2A was detected by ELISA. **p 0.01. (B) Correlation analysis was performed between liver stiffness measurement (LSM) and MEF2A expression in serum from patients with PBC. **p 0.01.