MicroRNA-145 induces the senescence of activated hepatic stellate cells through the activation of p53 pathway by ZEB2

2.1 Animal studies

Adult (8-week-old) male C57BL/6J mice from the Experimental Animal Center of Anhui Medical University were used for carbon tetrachloride (CCl₄) liver fibrosis model mice. All experimental procedures were approved by the institutional and local committee on the care and use of animals of Anhui Medical University. Liver fibrosis was generated by the biweekly intraperitoneal dose of a 20% solution of CCl₄ in olive oil (0.02 ml/g for a mouse) for the first 6 weeks and then twice a week as previously described. After 8 weeks, the liver was fixed in 4% buffered paraformaldehyde for histological analysis of liver fibrosis and immunostaining analysis or HSCs were isolated for western blot analysis. Serum was harvested for the further analysis.

All experiments were performed according to the institutional ethical guidelines for laboratory animal care and use of Anhui Medical University. Procedures involve animals and their cares were conducted in conformity with NIH guidelines and was approved by the Animal Care and Use Committee (number: LLSC20150348).

2.2 Cell culture

Rat HSC cell line HSC-T6 was purchased from Cell Bank of Chinese Academy of Sciences (Shanghai, Chinese). The human HSC line (LX-2) was purchased from Xiang Ya Central Experiment Laboratory (China). HSC-T6 cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM; Invitrogen, Grand Island, NY) with 10% fetal bovine serum (FBS; Sijiqing Biological Engineering Materials, Hang Zhou, China) and 1% antibiotics. LX-2 cells were cultured in DMEM (Gibco, Cambridge, MA) supplemented with 10% FBS (Termo). Cells were grown in a 95% air and 5% CO₂ humidified atmosphere at 37°C.

2.3 Immunofluorescence analysis

Immunofluorescence staining with liver tissues or treated cells were performed as we had previously reported, the cells were incubated with the primary antibodies against p53 (Santa Cruz Biotechnology, Dallas, TX), α-smooth muscle actin (α-SMA; Santa Cruz Biotechnology), or ZEB2 (Santa Cruz Biotechnology) and then incubated with the second antibodies. 4′,6-Diamidino-2-phenylindole (DAPI) was used to stain the nucleus in liver tissues. Hoechst reagent was used to stain the nucleus of HSCs in vitro. All the images were captured with the fluorescence microscope (Olympus, Tokyo, Japan) and representative images were shown. The red represents ZEB2 or p53 and α-SMA as green fluorescence.

2.4 Primary HSCs isolation

Primary HSCs were extracted via pronase/collagenase perfusion and differential centrifugation. The mice were treated with hydrolyzed trichloroacetaldehyde (Aladdin, Beijing, China). After placing a line around the portal vein, a special needle was inserted into the vein. Then perfusion buffer perfused liver via the portal vein. After the liver swelled, the cannula in the inferior vena cava was opened to allow draining. Subsequently, the liver was perfused in turn with enzymes fix after the blood was washed away and then the liver turns white. Finally, the liver was gently removed from the abdominal cavity and placed in a sterile dish. Then the livers were disrupted to prepare cells suspension which was filtered to remove residual debris. After cell suspension centrifugation by several times, the cell suspension density was adjusted with Nycodenz (Doobio, Beijing) mixture. The primary HSCs were acquired from the gradient centrifugation.

2.5 Analysis of HSC senescence

The senescence-associated β-galactosidase (SA-β-Gal) assay was performed using a senescence staining kit (Beyotime, Shanghai, China). Briefly, cells were plated onto 6-well culture plates, transfected with small interfering RNA (siRNA) or plasmid and inhibitor or mimics, then washed once with cleaning solution, and fixed with a fixation solution for 5 min at room temperature. Cells were washed three times with an acidic solution and incubated with the staining solution overnight at 37°C before microscopic analysis. All the images were captured with a light microscope and representative images were shown. Results were from triplicate experiments.

2.6 Cell cycle analysis by flow cytometry

Distribution of cell cycle was determined by propidium iodide (PI) staining and flow cytometry analysis. HSCs were seeded in 6-well plates. Cells were collected into flow cytometry tubes and centrifuged at 2,000 rpm for 5 min to obtain cell pellets. For cell cycle analysis, collected cells were washed with cold PBS and fixed in 70% ethanol at 4°C for 1 hour or at −20°C overnight. The fixed cells were washed with PBS and incubated with RNAase A (0.1 mg/ml) for 30 min followed by incubation with PI (50 mg/ml) for 30 min at room temperature. Cell cycle analysis was performed with a Coulter Epics XL Flow Cytometry System (Beckman-Coulter, Miami, FL). Cell cycle phases were determined using the Modfit software (Verity Software House, Topsham, ME).

2.7 Liver histopathology

Harvested liver tissues were fixed in 4% neutral-buffered formalin and embedded in paraffin. Liver slices of 50-μm thick were prepared and stained with hematoxylin and eosin stain (H&E) and Masson's trichrome stain by using standard methods. For Sirius red collagen staining, thin sections were deparaffinized and stained with picrosirius red for 1 hr at room temperature. After washes, sections on the slides were dehydrated in 100% ethanol and in xylene, and then they were mounted in permount. Photographs were taken in a blinded fashion at random fields. Representative views of liver sections are shown.

2.8 Serum measurements

Serum was collected and assayed for alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), and albumin (ALB) and was measured by using VetTest Chemistry analyzer (IDEXX Laboratories, Westbrook, ME) according to the manufacturer's instruction.

2.9 Luciferase reporter assay

The luciferase reporter plasmid P53 (pYr-PromDetect-TP53) was purchased from Changsha Yingrun Biotechnology Co. Ltd. (Hunan, China). Cells were seeded into a 24-well plate and collected after cotransfecting luciferase report vectors (pYr-ZEB2–3U and pYr-TP53p) with either miR-145 mimics, ZEB2-siRNA, or control vector by using Lipofectamine™ 2000 (Invitrogen, Carlsbad, CA). Luciferase activity was detected by using a Dual-Luciferase Reporter Assay Kit (Beyotime Biotechnology, Shanghai, China) according to the manufacturer's protocol. Luciferase activity was normalized by cotransfecting with pRL-TK (Promega, Madison, WI). All experiments were repeated three times.

2.10 Cell transfection

To downregulate or upregulate the expression of miR-145, ZEB2, and p53, miR-145 mimics, inhibitor, ZEB2 and p53-siRNA, or corresponding control RNA were used. The transfection was performed by using Lipofectamine™ 2000 (Invitrogen) according to the manufacture's protocols. Transfection after 6 hr, the culture medium was changed and TGF-β1 was added. Cells were reincubated in complete medium at 37°C for an additional 24 hr. After 3 days, cells were collected for quantitative real-time polymerase chain reaction (qRT-PCR) or western blot analysis. MiR-145 mimics, miR-145 inhibitor, ZEB2-siRNA, and p53-siRNA were purchased from GenePharma (Shanghai, China). The sequences are as following:

miR-145 inhibitor: 5′-AGGGAUUCCUGGGAAAACUGGAC-3′,

miR-145 mimics: 5′-GUCCAGUUUUCCCAGGAAUCCCU

GGAUUCCUGGGAAAACUGGACUU-3′; p53-siRNA (sense): 5′-CCACCAUCCACUACAACUATT-3′, p53-siRNA (antisense): 5′-UAGUUGUAGUGGAUGGUGGTT-3′; ZEB2-siRNA: 5′-CCAUCUCUGCUCAGAGUCCAATT-3′, 5′-UUGGACUCUGAGCAGCAGAGCAGAUGGTT-3′; negative control: 5′-UUCUCCGAACGUGUCACGUTT-3′, 5′-ACGUGACACGUUCGGAGAATT-3′.

2.11 Plasmid transfection

Overexpression plasmid of ZEB2 (ZEB2-OE) was obtained by amplifying complementary DNA (cDNA) coding for ZEB2 cDNA and inserting cDNA coding for ZEB2 into destination vectors by Gateway cloning (Invitrogen). The N-terminal region of ZEB2 coding region containing the predicted CARD domain was cloned into pEGFP-C2 vector by using restriction sites XbaI and BamHI. In addition, the p53 plasmid (p53-OE) was purchased from GenePharma (Shanghai, China). ZEB2 and p53 plasmid were transfected by using Lipofectamine™ 2000 (Invitrogen) in accordance with the manufacturer's instructions.

2.12 RNA extraction and qRT-PCR

Total cellular RNAs in cells were isolated with TRIzol Reagent (Invitrogen) according to the manufacturer's protocol. Reverse transcription was conducted with a PrimeScript RT Reagent Kit with gDNA eraser (TaKaRa, Dalian, China). The qRT-PCR was performed using the ABI Prism 7500 Sequence Detection System (Applied Biosystems, Foster City, CA). PCR was catalyzed by SYBR Green Master (TAKARA BIO Inc, Shiga, Japan). Glyceraldehyde phosphate dehydrogenase (GAPDH) was used as the invariant control. Fold changes in the mRNA levels of target genes related to the invariant control GAPDH were calculated according to the suggested method. U6 was used as the internal control for the expression of miRNA. The relative expression of each gene was calculated using the $$ method. The following primers of genes were used:

ZEB2 (human) (forward: 5′-CCCTTCTGCGACATAAATACGA-3′,

reverse: 5′-TGTGATTCATGTGCTGCGAGT-3′);

P53 (human) (forward: 5′-CGAAAGAAGACAGGCAGACTTTTCG-3′,

reverse: 5′-GAAGGTAAGGATAGGTCGGCGGTTC-3′);

GAPDH (human) (forward: 5′-ACCACAGTCCATGCCATCAC-3′,

reverse: 5′-TCCACCACCCTGTTGCTGTA-3′).

2.13 Western blot analysis

Protein extraction and subsequent analysis was performed as previously described (Yang et al., 2017; Zhou et al., 2016). Densitometric analysis of individual bands was performed using Image J software (National Institutes of Health, Bethesda, MD). Primary antibodies were as follows: ZEB2 (Santa Cruz Biotechnology), p16, p21, p53, Col. I (Abcam), α-SMA (Sigma, St. Louis, MO), and β-actin (Bio world, Beijing, China).

2.14 Statistical analysis

PRISM Software (Graphpad, San Diego, CA) was used for data analysis. Results are expressed as mean ± SD from multiple experiments. Student t tests were used to compare two groups. Variables with values of p < 0.05 by univariate analysis were used in subsequent multivariate analysis based on the Cox proportional hazards model. The differences were considered statistically significant at p < 0.05.

3.1 MiR-145 was downregulated in CCl₄-treated liver tissues, primary HSCs, and TGF-β1-induced HSCs

Mice treated with CCl₄ by using intraperitoneal injection were characterized by steatosis, expansion of cytoplasm, gradual conversion to myofibroblasts, and inflammatory infiltration. The pathological change of liver fibrosis tissues was detected by H&E staining, Masson staining, and picrosirius red staining. Histopathological analysis results showed that CCl₄ treatment resulted in serious hepatic steatosis, necrosis, and formation of regenerative nodules and fibrotic septa compared to normal groups (Figure 1a). In addition, serum AST and ALT activities, ALB and TBIL levels were detected by using kits. The results demonstrated that serum AST and ALT activities as well as TBIL levels were significantly increased in liver fibrosis induced by CCl₄ compared with normal groups. However, the serum ALB was decreased in CCl₄-induced liver fibrosis compared with normal groups (Figure 1b). We also examined the protein expression of α-SMA and Col. I by using western blot analysis, which are widely accepted as fibrosis markers in liver fibrosis. As shown in Figure 1c, expression of α-SMA and Col. I was noticeably increased in liver fibrosis induced by CCl₄ compared with normal groups. Based on these observations, the liver fibrosis models were established successfully. Thus we examined the expression of miR-145 in liver fibrosis tissues, primary HSCs, and TGF-β1-induced (10 ng/ml) HSCs. The qRT-PCR results showed that the expression of miR-145 was significantly lower in CCl₄-induced liver fibrosis compared with normal groups. Additionally, HSC-T6 cells and LX-2 cells were treated with TGF-β1 (10 ng/ml) for 24 hr. The results showed that the expression level of miR-145 was downregulated in HSC-T6 cells and LX-2 cells treated with TGF-β1 compared with not treated with TGF-β1 (10 ng/ml; Figure 1d). Next, we further measured whether there is any change in senescence between nonactivated HSCs and activated HSCs, particularly in the development of liver fibrosis. We found that the protein level of p16 and p21 were decreased in primary HSCs compared with the normal groups (Supplementary Information Figure 1a). In general, the finding revealed that miR-145 was decreased in activated HSCs.

3.2 MiR-145 inhibits fibrosis by promoting senescence in activated LX-2 cells

To investigate the effect of miR-145 on activated HSCs, we established stably downexpression or overexpression of miR-145 by using miR-145 inhibitor and miR-145 mimics in LX-2 cells treated with TGF-β1 (10 ng/ml). The flow cytometry results showed that ectopic expression of miR-145 in LX-2 cells significantly increased the number of cells in the G1 phase but decreased the number in LX-2 cells in the S phase compared with the negative control-transfected cells (Figure 2a). To further investigate the impact of miR-145 on senescence, SA-β-gal, a senescence-associated marker, was used to examine senescence in LX-2 cells after miR-145 inhibitor or miR-145 mimics treatment. As anticipated, we observed high numbers of SA-β-Gal-positive cells treatment with miR-145 mimics (Figure 2b). Western blot analyses of senescence-associated genes consistently showed that miR-145 mimics treatment upregulated the expression of senescence markers p16 and p21. Interestingly, silencing of miR-145 remarkably decreased the protein level of p16 and p21 in activated LX-2 cells (Figures 2c,d).

As modulation of senescence of activated HSCs could be an important complementary pathway in preventing liver fibrosis, we examined whether the expression level of miR-145 was associated with the severity of liver fibrosis. As shown in Figure 3e, the miR-145 inhibitor increased the levels of α-SMA and Col. I protein compared with normal cells without inhibitor transfection in activated LX-2 cells. Whereas, overexpression of miR-145 suppressed the expressions of α-SMA and Col. I (Figure 2f). Collectively, these data demonstrated that miR-145 may be inducing senescence in activated HSCs, thereby, potentially slowing down liver fibrogenesis.

3.3 ZEB2 is a direct target of miR-145 in HSC cells

To understand the potential mechanisms by which miR-145 regulates HSCs senescence, we used bioinformatics tools (TargetScan, miRDB, and miRanda) to predict the target genes of miR-145. The miRNA target prediction-algorithm predicted that the 3′-UTR of ZEB2 mRNA contains putative miR-145 binding sites (Figure 3a). To determine whether miR-145 regulates ZEB2 by binding to the corresponding 3′-UTRs, we cloned the 3′-UTR from ZEB2 into the pmirGLO luciferase reporter vector and cotransfected these vectors with miR-145 or control mimics into LX-2 cells, and evaluate ZEB2 response to miR-145 by using double luciferase reporter assay. The results showed that miR-145 could significantly reduce ZEB2 luciferase activity in LX-2 cells (Figure 3b). Furthermore, we found that overexpression of miR-145 decreased the level of ZEB2 mRNA in LX-2 cells, whereas transfection with miR-145 inhibitor increased the level of ZEB2 mRNA in LX-2 cells (Figure 3c). There results were confirmed by western blot analysis (Figure 3d). Next, we further measured whether manipulation of ZEB2 influences the induction of senescence by microRNA-145 (miR-145) in HSCs, the protein expression of p16 and p21 were detected by using western blot in activated LX-2 cells that was cotransfection of miR-145 inhibitor and ZEB2-siRNA, compared with miR-145 inhibitor. We found that the protein level of p16 and p21 was increased compare with miR-145 NC or miR-145 inhibitor (Supplementary Information Figure 1B). Taken together, these results provided evidence that miR-145 specifically targets the 3′-UTR regions of ZEB2 and thus inhibits the expression of genes.

3.4 ZEB2 suppresses HSCs senescence in liver fibrosis

3.4.1 Increased ZEB2 expression in vivo and in vitro

We examined expression of ZEB2 in liver fibrosis tissues. The colabeling ZEB2 and α-SMA was performed for colocalization. CCl₄-induced liver fibrosis showed a significant increase in ZEB2 compared with normal liver tissues, as examined by immunofluorescence (Figure 4a) and western blot analysis (Figure 4c). Importantly, the double immunofluorescence staining showed that the expression of ZEB2 was primarily colocalized with α-SMA, indicating that HSCs can be considered as one of the main sources of the ZEB2 levels present in CCl₄-induced liver fibrosis (Figure 4a). We further confirmed the expression of ZEB2 in TGF-β1-induced LX-2 cells and HSC-T6 cells. Interestingly, TGF-β1 could simultaneously increase the level of ZEB2 in LX-2 cells and HSC-T6 cells, as examined by using immunofluorescence (Figure 4b) and western blot analysis (Figure 4d). Collectively, these data suggested that ZEB2 is significantly increased in liver fibrosis tissues, LX-2 cells, and HSC-T6 cells.

3.4.2 ZEB2 suppresses cell senescence in activated LX-2 cells

To analyze the role of ZEB2 in the progression of liver fibrosis, we further assessed the effects of ZEB2 silence and overexpression on HSCs senescence in vitro. ZEB2-siRNA and ZEB2-plasmid were transfected into activated LX-2 cells. We found that downregulation of ZEB2 significantly increased the proportion of cells in the G1 phase of the cell cycle by using flow cytometry (Figure 5a). Moreover, SA-β-Gal staining showed that the SA-β-Gal activities induced by ZEB2-siRNA and inhibited by ZEB2-OE (Figure 5b). To further confirm the above results, HSCs were activated by TGF-β1 (10 ng/ml) and directly used for western blot analyses of p16 and p21. The results indicated that ZEB2-siRNA markedly increased the expression of p16 and p21 in activated LX-2 cells (Figure 5c). Opposite results were obtained by western blot analysis of p16 and p21 in LX-2 cells treated with ZEB2 plasmid (Figure 5d).Then we further examined the expression of α-SMA and Col. I at protein levels. Results of western blot analysis showed that ZEB2 plasmid resulted in increased protein levels of α-SMA and Col. I (Figure 5e). Whereas, the protein levels of α-SMA and Col. I were markedly decreased by ZEB2-siRNA treatment (Figure 5f). These data collectively suggested that ZEB2 may be regulating the senescence of activated LX-2 cells.

3.5 ZEB2 attenuates activities of p53 promoter via binding to E-box in LX-2 cells

To further characterize the mechanism by which ZEB2 acted on HSCs senescence. We used luciferase reporter assays to identify the role of ZEB2 in the activities of the p53 promoter. The p53 promoter has an E-box and can be potentially repressed by E-box binding transcriptional repressors like ZEB2. Interestingly, the results showed that an increase of luciferase activity in the p53 promoter was observed upon downregulation of ZEB2 in LX-2 cells compared with the empty vector (Figure 6a). In contrast, ZEB2 overexpression suppressed p53 promoter activity in activated LX-2 cells (Figure 6b). At the same time, the results from qRT-PCR showed that the p53 mRNA level was significantly inhibited in cells transfected with ZEB2 plasmid than in the cells in the negative control group, whereas transfection with ZEB2-siRNA increased the levels of p53 mRNA in cells (Figure 6c). There results were confirmed by using western blot analysis (Figure 6d). Altogether, the results supported that ZEB2 suppresses p53 promoter activity and thus inhibit the expression of p53.

3.6 P53 promotes the senescence of activated HSCs in liver fibrosis

3.6.1 The expression of p53 was downregulated in vivo and in vitro

To identify a potential relationship between liver fibrosis and the transcription factor p53, we first examined expression of p53 in CCl₄-induced liver fibrosis. Immunofluorescence double staining showed chronic CCl₄ intoxication reduced the production of p53 in mouse fibrotic liver (Figure 7a). In addition, results from western blot analysis indicated that the level of p53 was also decreased in liver fibrosis induced by CCl₄ (Figure 7c). Apart from these observations, we examined the expression of p53 in activated HSCs. Fluorescence staining showed that p53 was reduced in the nuclear upon TGF-β1 stimulation in HSCs (Figure 7b). In agreement with the observed protein changes, the p53 protein levels were downregulated upon TGF-β1 stimulation in HSCs (Figure 7d). Taken together, the results revealed that p53 was significantly downregulated by HSCs activation, suggesting that transcription factor p53 might play a role in HSCs activation.

3.6.2 P53 promotes activated HSCs senescence in vitro

P53 is the major mediator of cell cycle arrest and senescence in response to a series of cellular damage. Next, we examined whether disruption of p53 could affect the senescence of activated HSCs in vitro. The p53 overexpression and knockdown were introduced into LX-2 cells with p53-siRNA or p53-OE by lipidosome. As expected, flow cytometer indicated that the overexpression of p53 increased the ratio of cells in G1 phase in LX-2 cells compared with the vector (Figure 8a). We further examined the senescence marker SA-β-gal activity. As shown in Figure 8b, treatment with p53-OE significantly increased the number of SA-β-gal-positive cells in LX-2 cells compared with by vector treatment. In contrast, knockdown of p53 did not appear to enhance SA-β-gal activity. To further confirm the above results, LX-2 cells used for western blot analyses of p16 and p21. The results indicated that overexpression of p53 markedly increased the number of senescence marker-positive HSCs. Intriguingly, in contrast to the p53-OE-treated LX-2 cells, silencing of p53 inhibited the expression of p16 and p21 (Figure 8c,d). Finally, we analyzed the effect of p53 on HSCs activation. The results demonstrated that interference of p53 significantly elevated the α-SMA and Col. I in activated LX-2 cells. In contrast, the overexpression of p53 led to inhibition of expression of α-SMA and Col. I (Figure 8e,f). Overall, these results indicated that p53 promotes senescence in activated HSCs and thus potentially slowing down liver fibrosis.