miRNA‐130b‐5p promotes hepatic stellate cell activation and the development of liver fibrosis by suppressing SIRT4 expression

Abstract Liver fibrosis is a progressive disease accompanied by the deposition of extracellular matrix (ECM). Numerous reports have demonstrated that alterations in the expression of microRNAs (miRNAs) are related to liver disease. However, the effect of individual miRNAs on liver fibrosis has not been studied. Hepatic stellate cells (HSCs), being responsible for producing ECM, exert an important influence on liver fibrosis. Then, microarray analysis of non‐activated and activated HSCs induced by transforming growth factor β1 (TGF‐β1) showed that miR‐130b‐5p expression was strongly up‐regulated during HSC activation. Moreover, the progression of liver fibrosis had a close connection with the expression of miR‐130b‐5p in different liver fibrosis mouse models. Then, we identified that there were specific binding sites between miR‐130b‐5p and the 3′ UTR of Sirtuin 4 (SIRT4) via a luciferase reporter assay. Knockdown of miR‐130b‐5p increased SIRT4 expression and ameliorated liver fibrosis in mice transfected with antagomiR‐130b‐5p oligos. In general, our results suggested that miR‐130b‐5p promoted HSC activation by targeting SIRT4, which participates in the AMPK/TGF‐β/Smad2/3 signalling pathway. Hence, regulating miR‐130b‐5p maybe serve as a crucial therapeutic treatment for hepatic fibrosis.

essential role during the progression of liver fibrosis. 4 In the presence of external stimuli and liver injury, these cells go from quiescent to active. In activated HSCs, fibrotic marker gene expression is increased. 5,6 Currently, the mechanism of HSC activation is unclear and needs to be explored.
MicroRNAs (miRNAs), a subclass of short, non-coding RNAs, could regulate target mRNA degradation and protein translation. 7 In recent years, numerous reports have indicated that miRNAs are important regulators of gene expression in various liver diseases. 8 For example, exosomal miR-21 contributes to tumour progression in hepatocellular carcinoma. 9 Alcohol regulates miR-148a expression to induce hepatocyte pyroptosis via FoxO1. 10 miR-192-5p increases the degree of liver damage. 11 Then, how miRNAs involved in HSC activation and the relationship between miRNAs and HSCs have been reported in several studies. 12,13 Therefore, miRNAs play a key role in HSC activation.
The aim of this study was to explore the effect of miR-130b-5p on liver fibrosis. First, the data suggested that miR-130b-5p expression was significantly up-regulated during the process of HSC activation using microarray analysis. Next, we detected the expression of miR-130b-5p in experimental hepatic fibrosis models and in clinical patient samples. Consistently, miR-130b-5p expression was increased. Then, we found that SIRT4 may be the target of miR-130b-5p according the predicting software. Recent study had been demonstrated that SIRT4 could alleviate the HFD-induced liver fibrosis. EX-527, a SIRT1 inhibitor, could inhibit the progression of high-fat diet (HFD)-induced hepatic fibrosis via up-regulating SIRT4. 14 In this study, we identified a novel regulatory mechanism of miR-130b-5p-mediated SIRT4/ AMPK inactivation on the HSC activation in liver fibrosis diseases.
Further experiments revealed that miR-130b-5p promoted liver fibrosis, mechanistically activating HSCs and promoting its proliferation by repressing SIRT4 expression. In addition, in vivo administration of antagomiR-130b-5p relieved hepatic fibrosis. All in all, these results identify that the miR-130b-5p/SIRT4 axis as a possible novel regulator of HSC-mediated liver fibrosis and a potential therapeutic target.

| Clinical liver sample collection
Liver specimens were collected from patients with fibrosis or other liver diseases (the remote tissues of liver haemangioma) in the First Affiliated Hospital of Nanjing Medical University. All samples were instantly store in liquid nitrogen for rapid freezing after surgical resection. Informed consent was obtained from all subjects. All methods followed the relevant guidelines and regulations of Ethics Committee of Affiliated Hospital of Nanjing Medical University.

| Cell line and cell culture
HSC-T6 cells were obtained from the Cell Center of Shanghai Institutes for Biological Sciences. HSCs were maintained in DMEM (KeyGEN Biotech, China) with 10% foetal bovine serum (FBS: WISENT, Canada) and incubated 5% CO 2 at 37°C. HSCs were cultured in serum-free DMEM supplemented with TGF-β1 (10 ng/mL) for 0, 3, 6, 12 or 24 hours for activation.

| Western blotting analysis
Protein was obtained from cells in RIPA lysis buffer (Beyotime Biotechnology) with protease inhibitors and phosphatase inhibitors.
Extracted protein was subjected to SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride (PVDF) membranes (Bio-Rad). After blocking at room temperature for 2 hours with 5% non-fat dried milk, membranes were incubated with primary antibody overnight at 4°C. Membranes were then incubated with the appropriate secondary antibodies for 2 hours at room temperature.

| RNA extraction and quantitative RT-PCR
Total RNA was extracted from tissue or cells with TRIzol reagent (Invitrogen). The Reverse Transcription Kit (Vazyme) was performed to reverse RNA and synthesize cDNA. SYBR green (Vazyme) was used to perform mRNA amplification and cDNA quantification.
The levels of U6 and miR-130b-5p were detected by using TaqMan miRNA assay system (Life Technologies Corporation). Gene expression was normalized to snRNA U6 or β-actin expression, respectively. Primer sequences are shown in Table S1.

| miR-130b-5p and SIRT4 transfection in vitro or in vivo
miR-130b-5p mimics, scrambled miRNA (SCR-miRNA), antagomiR control (NC-miR), antagomiR-130b-5p, SIRT4 lentivirus (shSIRT4), SIRT4 lentivirus (LV-SIRT4) and control lentivirus were purchased from GenePharma. NC-miR and antagomir-130b-5p were transfected into HSC-T6 cells for 48 hours, and cells were collected for further experiment. HSC-T6 cells were transfected with lentiviruses using Lipofectamine 2000 for 48 hours (Invitrogen). Mice were treated with NC-miR and antagomiR-130b-5p (Gema) at a concentration of 20 nmol/200 µL via the tail vein injection. The mice were first treated with CCl 4 or treated by HFD for 2 weeks and then injected with NC-miR and antagomiR-130b-5p twice a week. Mice were killed after the end of treatment. Mice that underwent BDL were treated with antagomiR-130b-5p and NC-miR twice a week via the tail vein injection. Control mice were treated with NC-miR and antagomiR-130b-5p via tail vein twice a week. Mice were killed after the end of treatment. Liver tissues were collected for further experiments.

| Microarray analysis
Total RNA was isolated from quiescent (induced with TGF-β1 for 0 hour) and activated (induced with TGF-β1 for 24 hours) HSC-T6 cells, respectively, with TRIzol reagent (Life Technologies). RNA was purified with the RNA-quick Purification kit (YiShan Biotech).

Microarray analysis was performed by Shanghai Biotechnology
Corporation. The analysis of scanned images was conducted by using anscaAgilent Feature Extraction software (version 10.7).

| Mouse liver fibrosis models
Eight-week-old male C57BL/6 mice were obtained from the Animal Center of Nanjing Medical University (NJMU) and were housed under a specific pathogen-free conditions. For the olive oil group and CCl 4 group, mice received olive oil (2 mL/kg) or CCl 4 (10% in olive oil, 2 mL/kg) by intraperitoneally injection twice a week for 8 weeks. For the sham group and BDL group, male mice were subjected to sham or BDL surgery. For the chow group (normal diet) and HFD group, healthy mice were fed a high-fat diet (protein, 18

| Liver histopathology and fibrosis measurement
Formaldehyde-fixed, paraffin-embedded liver sections were stained with Masson and Sirius red to evaluate the extent of liver fibrosis. Dehydrated slides were scanned and imaged. The staining was further quantified by a pathologist. The positive staining area of Masson, Sirius red and α-SMA immunohistochemistry staining was measured using ImageJ software and quantified in a bar graph.
The percentages of the Masson-positive, Sirius red-positive and α-SMA-positive areas were calculated from five fields for each liver slice.

| Immunohistochemistry
Formalin-fixed and paraffin-embedded liver samples were deparaffinized, rehydrated and subjected to antigen retrieval. Samples

| Immunofluorescence (IF) staining
HSC-T6 cells were incubated with antiα-SMA primary antibody (Cell Signaling Technology), and then, the cells were treated with secondary goat anti-rabbit Texas Red-conjugated IgG (Sigma) and DAPI.
After washing the slides twice with pre-cooled PBS, we used confocal microscopy (ZEISS, Oberkochen) to obtain fluorescence images of the cells according to the manufacturer's instructions.

| Statistical analysis
Statistical analyses were performed using GraphPad Prism software or Statistical Package for the Social Sciences (SPSS) software version 19.0 and displayed as means ± standard deviation (SD). Differences between two groups were analysed using Student's t test. For multiple comparisons, the results were corrected by the Bonferroni method. P < .05 was considered statistically significant (*P < .05, **P < .01, ***P < .001).  Figure 1E). These findings indicated that miRNA-130b-5p expression is up-regulated during HSC-T6 cell activation.

| Up-regulated miRNA-130b-5p expression in multiple liver injury models
The mice were exposed to CCl 4 subcutaneous injection and subjected to BDL surgery. Since non-alcoholic fatty liver disease (NASH) has been one of the important causes that led to hepatic fibrosis, we used HFD to induce liver fibrosis model. The degree of liver fibrosis was also validated by H&E, Masson and α-SMA staining, which revealed obvious fibrous tissues (Figure 2A). In addition, the RT-PCR data revealed that the level of a-SMA, Collagen I and TIMP-1, which are genes associated with fibrosis, were significantly increased in liver tissues of treated mice compared with those of control mice ( Figure 2B-D). Furthermore, miRNA-130b-5p expression was strongly up-regulated in the livers with fibrosis ( Figure 2E). A recent study indicated that down-regulated miR-130b-5p expression could reduce hepatic lipid accumulation. 15 These results support that miRNA-130b-5p likely has a significant effect on the development of liver fibrosis.
In order to assess the effect of the transfection treatment on the results of cell proliferation, cell apoptosis and immunofluorescence, the experiment were made and data were presented in Figure S3.

| miR-130b-5p suppresses SIRT4 expression
To study the molecular mechanism by which miR-130b-5p regulates HSC activation, the predicting software (miRbase, miRwalk and TargetScan) was used to find the target genes of miR-130b-5p ( Figure S4). According to the results, SIRT4 may be the target of miR-130b-5p. Then, we examined the protein expression of SIRT4 in HSC-T6 cells activated by TGF-β1 for 0 hour and 24 hours. SIRT4 expression was clearly reduced in the activated HSCs ( Figure 4A). A previous study showed that the overexpression of SIRT4 can prevent HFD-induced liver fibrosis. 14 Moreover, whether miR-130b-5p can regulate SIRT4 expression during HSC activation has not yet been illustrated. A luciferase reporter assay suggested that miR-130b-5p mimics could repress luciferase activity of SIRT4 with a wild-type 3′-UTR (WT SIRT4 3′-UTR), compared with a mutant 3′-UTR (MUT SIRT4 3′-UTR) ( Figure 4B,C). As shown in Figure 4D, knockdown of miR-130b-5p increased the expression of SIRT4 in HSC-T6 cells induced with 10 ng/mL TGFβ for 24 hours. ( Figure 4D). The SIRT4 expression was reduced in the livers with fibrosis compared with F I G U R E 1 Up-regulated expression of miR-130b-5p during HSC activation. A, The protein expression level of α-SMA was increased in activated HSC-T6 cells induced by 10 ng/mL TGF-β1 at different time points. B, The mRNA expression of α-SMA was increased in activated HSC-T6 cells induced by 10 ng/mL TGF-β1 at different time points. C, α-SMA (red) was identified by immunofluorescence assays in HSC-T6 cells induced by 10 ng/mL TGF-β1 at different time points. D, MicroRNA microarray analysis results in quiescent and activated HSCs. E, The expression level of miR-130b-5p was detected by quantitative real-time PCR. F, The expression level of miR-130b-5p in activated HSCs was detected at different time points. Data represent means ± SEM of at least three independent experiments. *P < .05, **P < .01 and ***<.001 controls ( Figure 4E). Moreover, the protein expression levels of SIRT4 were lower in the liver fibrosis sample from patients than in those of normal controls ( Figure 4F). Then, the immunochemistry result demonstrated that expression of SIRT4 was down-regulated in the liver sample from patients with fibrosis ( Figure 4G). These observations suggested that miR-130b-5p might target SIRT4 and regulate its expression.

| miR-130b-5p regulates HSCs activation, proliferation and apoptosis via SIRT4
Previous study had reported that SIRT4 depletion augmented mTOR signalling via inactivating AMPKα. 16 Moreover, AMPK is associated with TGFβ/Smads pathway. 17 Therefore, we detected the protein and mRNA expression of AMPK, p-AMPK, TGFβ, Smad2, p-Smad2, Smad3 and p-Smad3. The expression of TGFβ, p-Smd2 and p-Smad3 was significantly reduced following the knockdown of miR-130b-5p in HSC-T6 cells induced with 10 ng/mL TGFβ for 24 hours, whereas SIRT4 and p-AMPK expression was increased ( Figure 5A). To further elucidate the possible mechanism by which miR-130b-5p modulates the activation and proliferation of HSCs by targeting SIRT4, the lentivirus was used to knockdown SIRT4 in HSC-T6 cells transfected with antagomiR-130b-5p. We performed Western blot and RT-qPCR to evaluate the efficiency of SIRT4 knockdown. The results showed that the expression level of SIRT4 in HSCs transfected with shSIRT4 is significantly lower than that of HSCs transfected with shctrl ( Figure S5). TGFβ, p-Smad2 F I G U R E 2 miR-130b-5p is up-regulated in multiple liver fibrosis models. A, Liver sections stained with H&E and Masson staining for assessing liver fibrosis in mice treated with olive oil or CCl 4 for 8 wk, sham or BDL for 2 wk and chow or HFD for 24 wk (n = 6/group; original magnification ×100; scale bars, 200 μm). The quantification of Masson-positive, α-SMA-positive and Sirius red-positive fibrosis areas in the livers from mice. B, The mRNA level of α-SMA was measured in liver tissue of treated mice by quantitative real-time. n = 6/ group. C, Collagen I expression was measured in liver tissue of treated mice at mRNA level. n = 6/group. D, TIMP-1 was detected in liver tissue of treated mice at mRNA level. n = 6/group. E, The expression level of miR-130b-5p was examined in liver tissue of treated mice by quantitative real-time PCR. n = 6/group. Graph represents mean ±SEM. *P < .05, **P < .01, and ***P < .001 Data represent means ± SEM of at least three independent experiments. *P < .05, **P < .01 and ***<.001. *P < .05, **P < .01 and ***P < .001 F I G U R E 4 miR-130b-5p targets SIRT4 and regulate its expression. A, The protein levels and mRNA expression of SIRT4 in HSC-T6 cells induced by 10 ng/mL TGFβ for 0 and 24 h. B, Predicted miR-130b-5p targeting sequence in SIRT4 3′UTR (WT SIRT4 3′UTR). Target sequences of SIRT4′UTR were mutated (MUT SIRT4 3′UTR). C, Dual-luciferase reporter assay of HSC-T6 cells transfected with WT SIRT4 3′UTR reporter together with miR-130b-5p mimics or NC-miR. D, The protein levels and mRNA expression of SIRT4 were detected by Western blotting in HSC-T6 cells transfected with antagomir or NC under the treatment of 10 ng/mL TGFβ for 24 h. E, Western blotting analysis for the expressions of SIRT4 in liver from mice with fibrosis. F, The protein levels of SIRT4 were examined in normal liver and liver fibrosis from patients. n = 6/group. G, Immunohistochemical staining of Masson, Sirius red and SIRT4 in liver section from patients with liver fibrosis or other liver disease (n = 6/group; original magnification ×200; scale bars, 100 μm). Data represent means ± SEM of at least three independent experiments. *P < .05, **P < .01 and ***<.001. *P < .05, **P < .01 and ***P < .001  Figure S7). Collectively, these results suggest that miR-130b-5p/SIRT4 axis may exert an influence on regulating HSCs activation, proliferation and apoptosis.

| AntagomiR-130b-5p ameliorates liver fibrosis in mice
Our results demonstrated that miR-130b-5p inhibited HSC activation and proliferation in vitro. Therefore, to verify whether antagomiR-130b-5p can prevent hepatic fibrosis in vivo, NC-miR or antagomiR-130b-5p was injected into control mice and CCl 4 -, BDL-and HFD-treated mice. The hepatic expression of miR-130b-5p was measured by RT-PCR ( Figure 6A). The results of H&E, Masson and Sirius red staining suggested that the extent of liver injury and liver fibrosis was heavier in CCl 4 /BDL/HFD group mice than in control group mice ( Figure 6B-D). In general, our study indicated that antagomiR-130b-5p can ameliorate liver fibrosis in mice.
F I G U R E 7 miR-130b-5p reduces the expression of SIRT4 in hepatic fibrosis tissues. A, α-SMA staining of α-SMA in liver sections from representative mice from each group. n = 6 mice for each group. B, The quantification of α-SMA-positive area in the livers from mice. n = 6 mice for each group. C, The fibrotic marker gene protein expression was detected in liver sections from mice transfected with NC-miR or antagomir-130b-5p by Western blotting. D, The mRNA levels of α-SMA were examined by quantitative real-time PCR. E, The mRNA levels of Collagen I was detected by quantitative real-time PCR. F, The mRNA levels of TIMP-1 were detected by quantitative real-time PCR. G, The protein levels of SIRT4, AMPK, p-AMPK, Smad2, p-Smad2, TGFβ, Smad3 and p-Smad3 in liver section was measured by western blotting. Data represent means ± SEM of at least three independent experiments. *P < .05, **P < .01 and ***<.001

| miR-130b-5p reduces the expression of SIRT4 in hepatic fibrosis tissues
Next, analysis of α-SMA expression indicated that knockdown of miR-130b-5p attenuated liver fibrosis in vivo ( Figure 7A,B). Consistently, in accordance with the reduction in miR-130b-5p levels, the expression levels of fibrotic gene were decreased ( Figure 7C-F). While the activation of Smad2, Smad3 and AMPK was inhibited, the SIRT4 protein expression was increased in the liver tissues of mice with fibrosis after treatment with antagomiR-130b-5p ( Figure 7G). In conclusion, our findings demonstrated that miR-130b-5p promotes liver fibrosis by regulating SIRT4 via the AMPK/TGFβ/Smad2/3 signalling pathway.

| D ISCUSS I ON
Hepatic fibrosis is a progressive disease accompanied by the deposition of ECM. Exposed to long-lasting damage, liver fibrosis could result in cirrhosis, liver failure, liver cancer and ultimately death. 18 HSCs are mainly responsible for liver fibrosis. In response to liver injury, quiescent HSCs are activated and transdifferentiate into proliferative, fibrogenic myofibroblasts, which produce most of the ECM. 19,20 Growing evidence has shown that the miRNAs are key factors in the progress of fibrosis. 21 Aberrant miRNA levels could regulate HSC activation and liver fibrosis. 22 Recent studies have shown that miR-130b-5p has different functions in diverse types of tumours. 23,24 In addition, miR-130b-5p was reported to ameliorate activated microglia-induced neuronal injury via the TLR4/NF-κB signalling pathway. 25 A report has demonstrated that microRNA-130b-5p as a regulator for inhibition and treatment of NAFLD. 15 However, the underlying mechanism of miR-130b-5p in HSC activation is poorly understood.
In this study, the data indicated that miR-130b-5p expression was significantly increased during HSC activation. Next, we observed that knockdown of miR-130b-5p prevented HSC activation and proliferation and promoted apoptosis of HSCs. Moreover, miR-130b-5p expression was up-regulated in hepatic fibrosis in CCl 4 , BDL, and HFD models and in patients with fibrosis. Compared with controls, mice transfected with antagomiR-130b-5p could ameliorate liver fibrosis. Our findings in vitro and in vivo suggested that miR-130b-5p may play a promoting role in liver fibrogenesis and that it acts as a positive regulator in HSCs activation.
Bioinformatics analysis revealed that SIRT4 might be a direct target of miR-130b-5p. SIRT4, which resides mitochondria and plays a significant role in cellular metabolic processes. 26 EX-527, as a SIRT1-selective inhibitor, relieve hepatic fibrosis by up-regulating SIRT4 expression. 14 A previous study indicated that SIRT4 silencing in tumour-associated macrophages promotes HCC development. 27 SIRT4 depletion augmented mTOR signalling via inactivating AMPKα. 16 AMPK, which consists of α, β and γ subunits, is a member of a serine/threonine (Ser/Thr) kinase family and is expressed in multiple organs. 28 The luciferase reporter assay showed that SIRT4 is a target that is bound by miR-130b-5p. Former studies have verified that overexpression of SIRT4 can activate AMPKα. 16 As expected, we found that the expression levels of p-AMPK were decreased in HSCs transfected with antagomiR-130b-5p. The previous report have proved that AMPK protects against fibrosis in the heart, 9 liver, 29 lung, 30 kidney 31 and skin. 32 Activation of AMPK can repress fibroblast proliferation and extracellular matrix accumulation and inhibit the transforming growth factorβ (TGFβ)/Smads signalling pathway. 17 TGFβ is an important cytokine that performs a key role in regulating tissue development and homeostasis. 33 Moreover, the TGFβ signalling pathway is closely related to the activation of HSCs 34 and induces ECM synthesis, 35 and its effector Smad proteins (Smad2, Smad3, Smad4) exert various and even opposing functions in the regulation of fibrosis. 35 Smad2, an antifibrotic role, is a major regulator upon the pro-fibrotic function of Smad3. 35 Smad4, as a pro-fibrotic protein, can promote collagen I promoter activity. 36 Our study indicated that knockdown of miR-130b-5p can alleviate hepatic fibrosis by targeting SIRT4 via the AMPK/TGFβ/Smad2/3 signalling pathway.
Collectively, our study revealed that the expression of miR-130b-5p was significantly up-regulated in activated HSCs and liver fibrosis models. miR-130b-5p, as a pro-fibrotic factor, promoted the expression of pro-fibrotic markers and HSC activation by targeting SIRT4 via the AMPK/TGFβ/Smad2/3 signalling pathway ( Figure 8).
Furthermore, antagomiR-130b-5p ameliorated liver fibrosis and promoted SIRT4 expression in animal models. Indeed, lack of F I G U R E 8 Schematic diagram showing the mechanism of miR-130b-5p-mediated liver fibrosis. miR-130b-5p could be upregulated in liver fibrosis models, miR-130b-5p overexpression inhibited SIRT4 expression and phosphorylation of AMPK, and it resulted in promoting phosphorylation of Smad2 and Smad3 to accelerate the progression of liver fibrosis substantial molecular mechanism, how SIRT4 activate AMPK, is one of the limitations of our study. We would address this question in our future research. In general, our study revealed a mechanism by which miR-130b-5p is involved in the activation of HSCs, and it offers a possible miRNA-based therapeutic strategy for alleviating hepatic fibrosis.

DATA AVA I L A B I L I T Y S TAT E M E N T
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.