Epigenetic silencing of LncRNA ANRIL enhances liver fibrosis and HSC activation through activating AMPK pathway

Abstract Long non‐coding RNAs (LncRNAs) and DNA methylation are important epigenetic mark play a key role in liver fibrosis. Currently, how DNA methylation and LncRNAs control the hepatic stellate cell (HSC) activation and fibrosis has not yet been fully characterized. Here, we explored the role of antisense non‐coding RNA in the INK4 locus (ANRIL) and DNA methylation in HSC activation and fibrosis. The expression levels of DNA methyltransferases 3A (DNMT3A), ANRIL, α‐Smooth muscle actin (α‐SMA), Type I collagen (Col1A1), adenosine monophosphate‐activated protein kinase (AMPK) and p‐AMPK in rat and human liver fibrosis were detected by immunohistochemistry, qRT‐PCR and Western blotting. Liver tissue histomorphology was examined by haematoxylin and eosin (H&E), Sirius red and Masson staining. HSC was transfected with DNMT3A‐siRNA, over‐expressing ANRIL and down‐regulating ANRIL. Moreover, cell proliferation ability was examined by CCK‐8, MTT and cell cycle assay. Here, our study demonstrated that ANRIL was significantly decreased in activated HSC and liver fibrosis tissues, while Col1A1, α‐SMA and DNMT3A were significantly increased in activated HSC and liver fibrosis tissues. Further, we found that down‐regulating DNMT3A expression leads to inhibition of HSC activation. Reduction in DNMT3A elevated ANRIL expression in activated HSC. Furthermore, we performed the over expression ANRIL suppresses HSC activation and AMPK signalling pathways. In sum, our study found that epigenetic DNMT3A silencing of ANRIL enhances liver fibrosis and HSC activation through activating AMPK pathway. Targeting epigenetic modulators DNMT3A and ANRIL, and offer a novel approach for liver fibrosis therapy.


| INTRODUC TI ON
Hepatic fibrosis is the wound-healing process in response to chronic liver injury. 1 Hepatic stellate cell (HSC) constitutes the accumulation of extracellular matrix (ECM) once they activated. 2 The activated HSC expresses a variety of factors such as transforming growth factor-β1 (TGF-β1), which stimulate the HSC activation, and secrete collagens and smooth muscle α-action (α-SMA). 3 This study discusses the molecular and cellular mechanisms of HSC activation and offers a novel approach for liver fibrosis therapy.
Currently, it is known that epigenetic modifications of liver fibrosis-related genes in liver fibrosis development. [4][5][6] Epigenetic provides to a heritable modulation in gene expression that does not alter the DNA itself. 7,8 Epigenetic influences generally refer to aberrant DNA methylation and non-coding RNA (ncRNA) modifications. 9,10 With regard to the latter, de novo DNA methylation activity catalysed by DNA methyltransferase 3A (DNMT3A) is methylated by addition of transfer methyl groups to the C-5 position in the cytosine ring. 11,12 DNA methylation can establish a docking site for transcriptional repressors to permanent gene silencing. 13 Long non-coding RNAs (LncRNAs) are well-known to interact with components of the epigenetic machinery. LncRNAs are longer than 200 nucleotides, which were protein-non-coding genes. 14 LncRNA (antisense non-coding RNA in the INK4 locus) ANRIL has been demonstrated to play an important role in fibrosis disease. 15 However, the molecular mechanisms of LncRNA ANRIL in liver fibrosis remain largely unknown.
Here, we document DNA methylation modification and their regulatory enzyme (DNMT3A) and LncRNA ANRIL that accompany liver fibrosis and HSC activation. DNMT3A silencing of LncRNA ANRIL regulates hepatic stellate cell activation through adenosine monophosphate-activated protein kinase (AMPK) pathway. Our study provides new understanding of epigenetic changes during liver fibrosis and HSC activation. Our findings suggest that epigenetic DNMT3A silencing of ANRIL enhances liver fibrosis and HSC activation through activating AMPK pathway, and offer a novel approach for liver fibrosis therapy.

| Animal models of liver fibrosis
Sprague-Dawley rats (Forty) were obtained from the Anhui Medical University, Experimental Animal Center. SD rats were intraperitoneally injected twice-weekly for 12 weeks with a mixture of carbon tetrachloride (CCl 4 )/olive oil in a 1:1 (vol/vol) ratio at 1 mL/kg. 16 All animal experiments were approved by the Institutional Animal Care and Use Committee of Anhui Medical University. Twelve weeks later, the animals were anaesthetized, the rats were then sacrificed and liver tissues collected.

| Histological analyses
Human and rat liver tissue sections were prepared, cut at 5 μm and stained with H&E, Sirius red and Masson's trichrome using standard histological techniques. Each section was assessed under light microscopic fields. Semi quantitatively be measured in five randomly selected fields of each liver samples using Image J software.

| Immunohistochemistry
Human and rat liver tissue sections were prepared, cut at 5 μm and slides were then permeabilized with 1% saponin/0.5% BSA/PBS for 10 minutes at room temperature, washed, blocked with 3% BSA for

| Immunofluorescence
Hepatic stellate cells then were fixed in 4% paraformaldehyde, and non-specific sites were blocked with 10% FBS. Cells then were incubated overnight at 4ºC in primary antibody to detect collagen I, α-SMA and DNMT3A. After 24 hours of culture, cells were washed with PBS and incubated with TGF-β1 (10 ng/mL) for 48 hours. After washing, cells were incubated in fluorochrome-coupled secondary antibody diluted in 1× phosphate-buffered saline for 1 hour at room temperature. DAPI (Genview Inc) was employed for staining. Finally, fluorescence was visualized with a microscope.

| MTT assays
According to the protocol of the manufacturer, cell proliferation assays were carried out by the use of MTT solution. HSC-T6 (5 × 10 3 /mL) was cultured with concentrations of 10 ng/mL TGF-β1 for 24, 48 hours in 96-well plates. According to the requirements of experiments, cell proliferation was observed at 24, 48 hours. Before the observation, MTT (0.5 mg/mL) was helped for incubation of these cells for about three hours at 37°C. After medium was removed, DMSO solution was added into the formazan crystal to dissolve and measure in triplicate at 490 nm wavelength using a Thermomax microplate reader (bio-tekEL). All experiments were performed in triplicate and repeated at least three times.

| CCK-8 assay
According to the protocol of manufacturer, cell proliferation assays were carried out by the use of CCK-8 (Dojindo Laboratories). After treatment, cells were seeded in six-well plates at approximately 2000 per well. After being cultured for 24, 48 hours, cells were added in 10 mL CCK-8 solutions. After incubating, the OD value was measured at 450 nm wavelengths using a Thermomax microplate reader (bio-tekEL).

| QRT-PCR
Total RNA was isolated from liver tissues or cells using TRIzol reagent

| Western blotting
Total proteins were extracted from liver tissues and cells. Proteins subjected to 10% SDS-PAGE for separation, and then they were transferred onto PVDF membranes. Following antibodies were used in this study: antibodies of DNMT3A, α-SMA, TGF-β1, Col1A1, AMPK, p-AMPK and GAPDH were diluted in 1:200-1:1000. After being blocked using nonfat milk, the membranes were incubated at 4°C with primary antibodies overnight, rinsed with Tris-buffered saline containing Tween 20 and further incubated with secondary antibody at room temperature for 1 hour. When the washing was finished, chemiluminescence detection kit was employed to examine signals from different proteins.

| Statistical analyses
Data were analysed by SPSS17.0 statistical software (SPSS Inc), where measurement data were represented by mean ± SD.
Comparison was tested by t test, and comparisons among groups were analysed by one-way analysis of variance (ANOVA). If P < .05, it was considered to be statistically significant.

| Pathological change of CCl 4 -induced experimental liver fibrosis
To begin to determine whether fibrosis is associated with pathological changes in the liver tissues, we examined the hepatic expression of TGF-β1, α-SMA and Col1A1. TGF-β1, Col1A1 and α-SMA are increased in liver fibrosis tissues ( Figure 1A-C). H&E staining found inflammatory infiltration, steatosis and fibrosis in the liver fibrosis tissues ( Figure 1D). Moreover, Masson's trichrome staining demonstrated that collagen deposition significantly increased in the liver fibrosis tissues ( Figure 1E). Furthermore, representative stains for Sirius red confirmed liver fibrosis in response to CCl 4 ( Figure 1F). In sum, 12 weeks after CCl 4 -treated, increased fibrosis, collagen deposition was observed in liver fibrosis tissues.

| Fibrosis liver displays aberrant DNMT3A and ANRIL expression while TGF-β1 causes the similar alterations in HSC
Given our interest in epigenetic mediators and, specifically, DNA methylation and LncRNA modifications, we examined differences in this domain. DNMT3A proteins and mRNAs were increased in rat fibrosis livers (Figure 2A-C). We also found that, in particular, DNMT3A was over expressed after TGF-β1 treatment HSC ( Figure 2D-F). Interestingly, we determined the expression of ANRIL in liver fibrosis tissues. ANRIL expression was significantly down-regulated when HSC treated with TGF-β1 in a time-dependent manner ( Figure 2G,H). All together, these results suggest that TGF-β1 suppresses ANRIL likely associated with aberrant DNMT3A expression in fibrosis liver.

| Epigenetic DNMT3A inhibition attenuates TGF-β1-dependent HSC activation in vitro
We next sought to examine whether DNMT3A inhibition, either with pharmacologic or genetic approaches, would modulate HSC activation in vitro. Firstly, we used an epigenetic compound, and α-SMA gene expression is classic events associated with fibroblasts proliferation, both were repressed in DNMT3A-siRNAtreated cultures ( Figure 3H).

| Epigenetic silencing of LncRNA ANRIL is required for HSC proliferation
We were therefore interested to determine whether HSC proliferation is associated with alterations in DNA methylation machinery and DNMT3A marks. HSC transfected with DNMT3A-siRNA expressed higher levels of LncRNA ANRIL relative to cells transfected with a scrambled siRNA ( Figure 4A). Moreover, HSC was treated with 5-AzadC to promote LncRNA ANRIL expression ( Figure 4B). HSC

| LncRNA ANRIL depletion triggers HSC activation through AMPK pathway
To gain the roles of LncRNA ANRIL in regulating HSC activation, we tested the effect of LncRNA ANRIL down-expression on the HSC proliferation. We found that the expression of LncRNA ANRIL significantly decreased in the HSC transfected with LncRNA ANRIL-siRNA ( Figure 5A). The HSC that was transfected with LncRNA ANRIL-siRNA had a significantly higher proliferation than NC and vehicle ( Figure 5B,C). Moreover, we confirmed the significant downregulation of LncRNA ANRIL, which led to a subsequent increase in α-SMA and Col1A1 ( Figure 5D,E). AMPK signal pathway plays a key role in HSC activation. We next wanted to conclusively determine whether AMPK is a key gene in LncRNA ANRIL-mediated HSC activation. HSC transfected with pcDNA3.1-ANRIL expressed lower levels of phosphorylated AMPK relative to cells transfected with a vector, while HSC that was transfected with LncRNA ANRIL-siRNA expressed higher levels of phosphorylated AMPK relative to cells transfected with a scrambled siRNA ( Figure 5F,G). These results suggested that LncRNA ANRIL depletion triggers HSC activation through AMPK pathway.

| Epigenetic alterations in human chronic liver diseases of distinct aetiologies resemble those associated with rat liver fibrosis
We were next interested to discover whether similar epigenetic changes occur in fibrosis human livers. H&E staining found inflammatory infiltration, steatosis and fibrosis in human liver fibrosis tissues ( Figure 6A). Moreover, Masson's trichrome staining demonstrated that collagen deposition significantly increased in the human liver fibrosis tissues ( Figure 6B). Furthermore, representative stains for Sirius red confirmed in human liver fibrosis tissues ( Figure 6C). DNMT3A, α-SMA, Col1A1 and p-AMPK were increased in human fibrosis livers, while LncRNA ANRIL was decreased in human fibrosis livers ( Figure 6D-F). These data, collected from mechanistically distinct examples of chronic human liver disease, reveal common fibrosis-associated epigenetic changes including increased DNMT3A expression, and reduced LncRNA ANRIL that are also observed as characteristic epigenetic features of experimental liver fibrosis in rats.

| D ISCUSS I ON
Hepatic stellate cell activation plays a key role in the pathogenesis of liver fibrosis. 17 It is known that TGF-β1 triggers AMPK signal pathway activation in liver fibrosis and HSC activation. 18 The implication of DNA methylation and LncRNAs in liver fibrosis has been increasingly discovered. 19,20 In this study, we have demonstrated LncRNA ANRIL, which is involved in the attenuation of HSC activation in liver fibrosis. To our knowledge, this F I G U R E 4 Epigenetic silencing of LncRNA ANRIL is required for HSC proliferation. A, HSC transfected with si-DNMT3A or si-scrambled, ANRIL expression was analysed by qRT-PCR. B, HSC treatment with TGF-β1 and 5-AzadC, ANRIL expression was analysed by qRT-PCR. C, HSC transfected with pcDNA3.1-ANRIL, ANRIL expression was analysed by qRT-PCR. D, HSC transfected with pcDNA3.1-ANRIL, cell proliferation was measured by MTT assay. E, HSC transfected with pcDNA3.1-ANRIL, cell proliferation was measured by CCK-8 assay. F, HSC transfected with pcDNA3.1-ANRIL, α-SMA and Col1A1 mRNA expression was analysed by qRT-PCR. G, HSC transfected with pcDNA3.1-ANRIL, α-SMA and Col1A1 protein expression was analysed by Western blotting. Data are representative of at least three separate experiments. *P < .05, **P < .01 vs Control or Vector is the first study demonstrating DNMT3A silencing of LncRNA ANRIL regulates HSC activation, thus providing novel mechanistic insights into a critical role for LncRNA ANRIL in liver fibrosis pathogenesis.
LncRNA ANRIL is a recently discovered LncRNA that consistently correlated with fibrosis disease as well as other human diseases like cardiovascular, tumour and so on. Presently, we uncovered significant down-regulation of LncRNA ANRIL in activated HSC, as well as low expression of this gene in liver fibrosis.
We found that LncRNA ANRIL depletion triggers HSC activation.
Moreover, LncRNA ANRIL could negatively inhibit the activation of AMPK pathway and the excretion of collagens. In addition, the over expression of LncRNA ANRIL showed an opposite effect in the activated HSC. Interestingly, our results largely implied that LncRNA ANRIL deficient in activated HSC was more likely to be a cellular compensatory mechanism.
There is a growing interest in epigenetic molecular mechanisms for investigations and therapeutic purpose. Aberrant DNMT expressions have been found in liver fibrosis conditions, 21  Moreover, we used 5-AzadC, which specifically inhibitors the catalytic subunit of DNMT3A. Inhibition of DNMT3A in HSC-treated 5-AzadC and TGF-β1 led to a significant decrease in α-SMA and Col1A1. We also have been showed that DNMT3A mediates HSC activation in liver fibrosis by down-regulating the expression of LncRNA ANRIL. At the same time, HSC was treated with 5-AzadC to promote LncRNA ANRIL expression. We also discovered similar epigenetic changes occur in fibrosis human livers. These data, collected from mechanistically distinct examples of chronic human liver disease, reveal common fibrosis-associated epigenetic changes including increased DNMT3A expression, and reduced LncRNA ANRIL F I G U R E 5 LncRNA ANRIL depletion triggers HSC activation through AMPK pathway. A, HSC transfected with LncRNA ANRIL-siRNA, ANRIL expression was analysed by qRT-PCR. B, HSC transfected with LncRNA ANRIL-siRNA, cell proliferation was determined using MTT assay. C, HSC transfected with LncRNA ANRIL-siRNA, cell proliferation was determined using CCK-8 assay. D, HSC transfected with LncRNA ANRIL-siRNA, α-SMA and Col1A1 mRNA expression was analysed by qRT-PCR. E, HSC transfected with LncRNA ANRIL-siRNA, α-SMA and Col1A1 protein expression was analysed by Western blotting. F, HSC transfected with LncRNA ANRIL-siRNA, AMPK and p-AMPK protein expression was analysed by Western blotting. G, HSC transfected with pcDNA3.1-ANRIL, AMPK and p-AMPK protein expression was analysed by Western blotting. Results shown were representative of three independent experiments. *P < .05, **P < .01 vs Control or Vector or Scrambled F I G U R E 6 Epigenetic alterations in human chronic liver diseases of distinct aetiologies resemble those associated with rat liver fibrosis. A, Human liver fibrosis tissue was fixed with formalin, and then it was embedded in paraffin. Thin sections were cut and stained with haematoxylin & eosin (H&E), B, Masson's trichrome stain and (C) Sirius red staining. D, DNMT3A, Col1A1, α-SMA and ANRIL expression were analysed by qRT-PCR. E, DNMT3A, Col1A1, α-SMA, AMPK and p-AMPK expression were analysed by Western blotting. F, The DNMT3A, Col1A1 and α-SMA expression were evaluated by immunohistochemistry. Data are representative of at least three separate experiments. *P < .05, **P < .01 vs vehicle that are also observed as characteristic epigenetic features of fibrosis human livers.
In conclusion, our study demonstrates that epigenetic silencing of LncRNA ANRIL regulates HSC activation though AMPK pathway is crucial pro-fibrogenic components forming an epigenetic cascade promoting liver fibrosis. This is a novel understanding of the role of DNMT3A and LncRNA ANRIL in HSC proliferation and the mechanism involved.

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest.

AUTH O R ' S CO NTR I B UTI O N
Jing-Jing Yang wrote the paper; Yang Yang and Chong Zhang participated in the design of the study and performed the statistical analysis, Yan Yang and Jun Li instructed the paper. All authors read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.