Abstract

ZNF300 plays an important role in the regulation of HBV-related hepatocellular carcinoma. However, little is known about the role of ZNF300 in lipid metabolism and NAFLD. In the present study, we observed that ZNF300 expression was markedly decreased in free fatty acid (FFA)-induced fatty liver. Overexpressed ZNF300 alleviated hepatic lipid accumulation, whereas knockdown of ZNF300 enhanced the FFA-induced lipid accumulation. Investigations of the underlying mechanisms revealed that ZNF300 directly binds to and regulates the PPARα expression, thus promoting fatty acid oxidation. Furthermore, bisulfite pyrosequencing PCR (BSP) analysis identified the hypermethylation status of ZNF300 gene in FFA-treated hepatocytes. Importantly, the suppression of ZNF300 could be blocked by DNA methyltransferase inhibitor (5-azadC) or DNMT3a-siRNA. These results suggested that ZNF300 plays an important role in hepatic lipid metabolism via PPARα promoting fatty acid oxidation and this effect might be blocked by DNMT3a-mediated methylation of ZNF300. Therefore, in addition to ZNF300 expression levels, the methylation status of this gene also has a potential as a prognostic biomarker.

Introduction

Non-alcoholic fatty liver disease (NAFLD) is a common chronic liver disease with increasing prevalence as the numbers of obese or overweight adults and children increase [1,2]. It encompasses a wide spectrum of liver disorders and damage, including simple steatosis, inflammatory non-alcoholic steatohepatitis (NASH), liver cirrhosis and hepatocellular carcinoma [3,4]. The pathogenesis of NAFLD is not well understood, but dysregulation of lipid metabolic in the liver is known to play important roles. Hepatic lipid metabolism is controlled by a series of transcription factors, such as sterol regulatory element binding protein-1c (SREBP1c), liver X receptor (LXR) and proliferator-activated receptor alpha (PPARα) [5,6]. The activation of LXR plays a crucial role in the progression of NAFLD, and its inhibition in tissues or cells reduces the expression of SREBP1C and fatty acid synthase (FAS) in the liver [7,8]. PPARα, a member of the nuclear receptor family of ligand-activated transcription factors, is expressed predominantly in the liver [5,9] and controls the transcription of many genes involved in lipid catabolism [5,9,10]. The activation of PPARα increases the β-oxidation of fatty acids [11], ketogenesis and reduces cellular lipids [12,13]. Therefore, LXR inactivation and PPARα activation are considered important for the prevention or amelioration of metabolic syndrome.

Zinc finger protein 300 (ZNF300), encoding a KRAB domain and 12 C2H2 type zinc finger domain, belongs to the Krüpple-associated box zinc finger proteins (KRAB-ZFPs) and is exclusively expressed in tetrapod vertebrate [14]. KRAB-ZFPs constitute the largest individual family of transcriptional factors encoded by the higher organisms’ genome [15] and play an important role in regulating embryonic development, cell differentiation, proliferation and malignant transformation [16,17]. ZNF300 binds to C(t/a)GGGGG(g/c)G sequences that are found in the promoter regions of some genes, such as IL-2, IL-2Rβ, CD44, TNFα, and it can activate IL-2Rβ promoter activity [14,18]. Previous studies suggest that ZNF300 can enhance metastasis of cancer cells by activating NF-κB pathway [14] and plays crucial roles in leukemogenesis [18,19]. In addition, we also found that the zinc finger regions of ZNF300 gene can mediate the gene transcriptional repression through DNA methylation and histone acetylation [15]. Recently, one study showed that the hypermethylation of ZNF300 during HBV-related hepatocellular carcinogenesis (HCC) development is coupled with decreased gene expression [20]. Since the last stage of NAFLD development is HCC, whether ZNF300 is involved in the regulation of hepatic lipid metabolism during the development of NAFLD has not been investigated.

In the present study, we examined the expression of ZNF300 in the hepatocytes treated with free fatty acids (FFAs). To identify the role of ZNF300 in hepatic lipid metabolism, we evaluated alterations in lipid droplets and the contents of TG (triglyceride) in hepatocytes after ZNF300 overexpression and knockdown. We found that ZNF300 prevented FFA-induced hepatic steatosis mainly through promoting fatty acid oxidation. Then we preliminarily explored the molecular mechanism of the hepatic lipid metabolism modulated by ZNF300 and found ZNF300 could induce the expression of PPARα by directly binding to its promoter. In addition, we also found the ZNF300 methylation level dramatically increased in fatty liver. Collectively, the expression of ZNF300 is controlled by DNA methylation and ZNF300 regulates lipid metabolism through a PPARα-dependent manner in the liver.

Materials and methods

Animal experiments

All mice care and experimental procedures were approved by the Jiangsu Normal University. A mouse NAFLD model was established through feeding mice on HFD (60% kcal from fat; D-12492; Research Diets, NJ) for 18 weeks starting at the age of 8 weeks, whereas NCD-fed mice were maintained on a normal diet (10% kcal from fat, D-12450B). Food and water were provided ad libitum. Mice were kept at room temperature (22–24°C) with a 12 h light–dark cycle.

FFA preparation

The 50 mM palmitate (Sigma–Aldrich) and 50 mM oleic acid (Sigma–Aldrich) stock solutions were prepared in 50% ethanol and filtered through a 0.22 μm membrane. The 10% mass–volume fatty acid-free BSA solution was prepared. Finally, palmitate:oleate solution (FFAs) with a 1:2 ratio and the final concentration of 1% BSA were added to DMEM to prepare a 1 mM free fatty acid-BSA solution.

Cell culture and treatments

Human hepatocellular carcinoma cells HepG2, L02 and HEK (human embryonic kidney)-293 T cells were cultured in a complete Dulbecco's modified Eagle's medium (DMEM, Gibco BRL, NY, U.S.A.). The medium was supplemented with 10% (v/v) fetal bovine serum and 1% (v/v) penicillin/streptomycin (Gibco, NY, U.S.A.). All cells were cultured in a humidified chamber with 5% CO2 atmosphere at 37°C. To induce fatty liver model, HepG2 and L02 cells were starved for 12 h before stimulated with or without 1 mM PA/OA for 24 h.

Lentivirus-mediated gene transfer

HEK293 T cells were transfected with phage-6tag-ZNF300, empty vector, pLKO1-shZNF300-1#, pLKO1-shZNF300-2#, pLKO1-scramble along with the packing vectors pSPAX2 and pMD2.G. After transfection for 8 h, the medium was changed with fresh complete medium and 48 h later, the supernatants were harvested to infect HepG2 or L02 cells followed by puromycin selection for 1 week for various analyses.

Plasmid construction

The promoter of PPARα (−1000 to +150 relative to the transcription start site) was amplified by PCR using primers (5′-CCG CTCGAG GTCACGGCCC GAACAAAG-3′; 5′-CCC AAGCTT CGTCCGCCGC CCTCCG-3′) containing Xho I and Hind III restriction enzyme sites at each end from the human genome and subcloned into the pGL3-basic vector. The point mutation of the pGL3-PPARα vector was constructed by PCR amplifying the whole vector using primers with mutant sites (5′-GGCTCGCGCGGACCGGGGCAAAAAACGGGCCGAGG-3′; 5′-TTTTTTGCCCCGGTCCGCGCGAGCCAGTGTCCCG-3′). The deleted mutation of pGL3-PPARα vector was constructed using the over-lap PCR method. The all constructed vectors were identified by DNA sequencing.

RNA isolation and quantitative real-time PCR

Total RNA was extracted from cells using TRIzol reagent (Invitrogen, Grand Island, NY, U.S.A.) according to the manufacturer's instructions. cDNA was synthesized using the PrimeScript RT Reagent Kit with gDNA Eraser (Takara, Japan) according to the manufacturer's protocol. The quantitative real-time PCR (qRT-PCR) was performed with SYBR Green PCR Master Mix (DBI-2044, Germany) with triplicates on the ABI step one plus real-time PCR system. For each primer set, the Ct value was normalized to that of β-actin as the inner control, which was further normalized to that of the control sample. The relative quantitation of PCR products was measured using the comparative Ct method and presented as relative mRNA level.

Western blotting

Cellular proteins were extracted using RIPA lysis buffer. Subsequently, equal amounts of protein were separated by SDS–PAGE gels and transferred onto PVDF membranes (Millipore, U.S.A.). Primary antibodies against Flag (Sigma, F3165), ZNF300 (Sigma, SAB2102853), PPARα (Abcam, ab191226) and β-actin (CST, #4970) were incubated overnight at 4°C and visualized by ECL (Millpore).

RNA interference analysis

RNA interference (RNAi) experiments in HepG2 cells were performed by using Lipofectamine 2000 (Invitrogen, U.S.A.) according to the manufacturer's protocol. For DNMT1, DNMT3a and PPARα silencing, cells were transfected with DNMT1-siRNA, DNMT3a-siRNA, PPARα-siRNA or negative control siRNA (Sangon Biotech, China). The following siRNA sequences were used: DNMT1-siRNA (sense: 5′-CAGAACAAGAAUCGCA UCUUU-3′ and antisense: 5′-AGAUGCGAUUCUUGUUCUGUU-3′); DNMT3a-siRNA (sense: 5′-CUUUGAUGGAAUCGCUACAUU-3′ and antisense: 5′- UGUAG CGAUUCCAUCAAAGUU-3′); PPARα-siRNA (sense: 5′-CAUUUGCUGUG GAGA UCGUUU-3′ and antisense: 5′-ACGAUCUCCACAGCAAAUGUU-3′); siRNA-control with scrambled sequence: (sense: 5′-UUCUCCGAACGUGUCACGUUU-3′ and antisense: 5′-ACGUGACACGUUCGGAGAAUU-3′).

Oil Red O staining

Prior to staining, the stock solution of Oil Red O (Sigma–Aldrich) was diluted with deionized water at a 3:2 ratio. The cells were fixed with 3.7% formaldehyde for 1 h and washed three times with PBS. Before staining with Oil Red O solution, the cells were treated with 60% isopropyl alcohol for 2 min, and then stained for 10 min at room temperature. Finally, the cells were washed with 60% isopropyl alcohol for less than 5 s and with PBS washed several times. Stained lipid droplets were observed using a microscope and captured at 200× magnification. The stained lipid droplets were eluted with 100% isopropanol, and the optical density was detected using a spectrophotometer at the 520 nm wavelength.

Intracellular TG assay

For triglyceride detection, the treated hepatocytes were incubated with lysis buffer for 10 min and then heated at 70°C for 10 min, followed by centrifugation at 2000 rpm for 5 min at room temperature. The supernatant was evaluated using a triglyceride assay kit (Applygen-E1013, Beijing, China) at 550 nm.

Dual luciferase activity assay

HEK293 T cells or HepG2 cells (2 × 104) cultured in 24-well plates were transfected with the reporter plasmid (200 ng) in combination with other plasmid as indicated in figures. The vector pRL-TK (20 ng) expressing Renilla luciferase was used in all samples and served as internal control. Cells were lysed 24 h post-transfection with passive lysis buffer and the dual luciferase activity was assayed by the dual luciferase reporter assay kit (Promega) according to the manufacturer's instructions. The firefly luciferase activity was normalized to the renilla luciferase activity and presented as relative luciferase activity.

Chromatin immunoprecipitation assays

Cells were fixed with 1% formaldehyde and quenched by glycine and then washed three times with PBS and harvested in ChIP lysis buffer (50 mM Tris–HCl pH 8.0, 1% SDS, 5 mM EDTA) followed by sonication until DNA reached 400–600 bp. The lysates were centrifuged at 4°C for 15 min and ChIP dilution buffer (20 mM Tris–HCl pH 8.0, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100) was added to the supernatant. The resulted lysates were then incubated with the protein G beads and anti-Flag (Sigma, Cat#F3156) or control IgG at 4°C for 4 h. DNA was eluted using the ChIP elution buffer (0.1 M NaHCO3, 1% SDS, 30 μg/ml proteinase K) through incubation at 65°C overnight, and the DNA was purified with a DNA purification kit (TIANGE). The purified DNA was assayed by quantitative PCR.

Bisulfite sequencing PCR

DNA from cells was extracted using DNA extraction kit following the manufacturer's instructions. DNA bisulfite sequencing, wildly used to detect 5-methylcyosine in DNA, is conducted as described previously [15]. The primers for BSP at the locus of CpG island (5′-AAAGTGTGTTTTTTAATGTTTTTTT-3′; 5′-CATAACTACTCCCTAAAACT ATTTCC-3′) were designed for ZNF300 using the Methprime database (http://www.urogene.org/methprimer/). In brief, 500 ng genomic DNA was bisulfite converted using EZ DNA Methylation-Gold Kit™ according to the manufacturer's instructions. The converted DNA was used as the template for PCR amplification and the PCR products were subcloned into a pMD-19-T vector. For each gene, 10 clones were selected for sequencing and analyzing the sequence using DNAMan program.

Determination of S-adenosylmethionine and S-adenosylhomocysteine

The concentrations of S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH) were determined in HepG2 and L02 cells with or without FFA treatment for 24 h. The indicated cells (5 × 106 cells) were resuspended in 500 μl PBS and the cellular contents were released by repeated freeze–thaw. Homogenates were centrifuged at 3000 rpm/min for 20 min at 4°C, then the supernatants were collected. The concentrations of SAM and SAH in the supernatants were measured using the ELISA method (YI FEI XUE Bio TECH) according to the manufacturer's instructions.

Statistical analysis

Statistical analyses were performed using SPSS software (version 24.0). Data combined from three or more independent experiments are given as the mean ± SD. Statistical significance was determined by a one-way ANOVA (Dunnett's test) among more than two groups. Student's t-test (two-tailed, unpaired) was performed to compare the difference between the two groups. A P-value of 0.05 or less was considered significant.

Results

ZNF300 expression is decreased in FFA-induced hepatic steatosis

To examine the effect of FFAs on lipid accumulation in hepatocytes, HepG2 and L02 cells were treated with palmitate:oleate solution. Oil Red O staining showed that FFA treatment caused a time-dependent increase in lipid accumulation in HepG2 and L02 cells (Figure 1A,B). We also found the number and size of the lipid droplets significantly increased when cells were treated for 24 h. Moreover, the contents of TG in hepatocytes gradually increased with the extension of the induction time of FFAs (Figure 1C). These results showed that we have successfully constructed an in vitro fatty liver model. According to previous reports, ZNF300 plays a vital role in HBV-related HCC [20]. However, it is less known whether ZNF300 is involved in the development of NAFLD. To resolve this question, we first investigated the expression of ZNF300 in vitro fatty liver model. As shown in Figure 1D,E, the expression of ZNF300 at both the mRNA and protein levels was decreased in a time-dependent manner in hepatocytes, suggesting that ZNF300 may be involved in the development of hepatic steatosis.

ZNF300 expression is decreased in FFA-induced hepatic steatosis.

Figure 1.
ZNF300 expression is decreased in FFA-induced hepatic steatosis.

(A and B) Representative images of Oil Red O staining in L02 (upper panel) and HepG2 cells (bottom panel) (A), and quantified with a spectrophotometer at 520 nm (B). (C) The intracellular TG content was assayed using the triglyceride assay kit. (D and E) Quantitative PCR analysis (D) and Western blot (E) analysis showing the expression level changes of ZNF300 in the L02 and HepG2 cell lines, treated with 1 mM palmitate/oleate (PA + OA,1:2) for the indicated time. *P < 0.05, **P < 0.01 versus 0 h group.

Figure 1.
ZNF300 expression is decreased in FFA-induced hepatic steatosis.

(A and B) Representative images of Oil Red O staining in L02 (upper panel) and HepG2 cells (bottom panel) (A), and quantified with a spectrophotometer at 520 nm (B). (C) The intracellular TG content was assayed using the triglyceride assay kit. (D and E) Quantitative PCR analysis (D) and Western blot (E) analysis showing the expression level changes of ZNF300 in the L02 and HepG2 cell lines, treated with 1 mM palmitate/oleate (PA + OA,1:2) for the indicated time. *P < 0.05, **P < 0.01 versus 0 h group.

ZNF300 protects hepatocytes from FFA-induced steatosis

To investigate the role of ZNF300 in FFA-induced hepatic steatosis, we overexpressed ZNF300 using lentivirus gene transfer in hepatic cells (HepG2 and L02) and the overexpression efficiencies of ZNF300 were confirmed at mRNA and protein levels (Figure 2A). Compared with the control group, overexpression of ZNF300 ameliorates FFA-induced lipid droplet accumulation in L02 and HepG2 cells (Figure 2B,C). In addition, overexpression of ZNF300 also markedly decreased FFA-induced intracellular TG deposition (Figure 2D). We next measured the expression of related genes involved in lipid metabolism. As expected, FFAs could stimulate the gene expression involved in lipogenesis and inhibit the gene expression related to fatty acid oxidation, lipid secretion as well as lipolysis. In the presence of FFAs, ZNF300 could significantly enhance the expression of genes related to fatty acid oxidation and reduce the expression of a few lipogenic genes; however, it had no obvious effects on lipolysis (Figure 2E). Collectively, these data are indicative that ZNF300 overexpression alleviates the lipid accumulation induced by FFAs in hepatocytes mainly through regulating fatty acid oxidation and lipogenesis.

ZNF300 protects hepatocytes from FFA-induced steatosis.

Figure 2.
ZNF300 protects hepatocytes from FFA-induced steatosis.

(A) qRT-PCR analysis (left graph) of ZNF300 mRNA and immunoblot analysis (with anti-exogenous and endogenous ZNF300 protein, anti-β-actin protein) (right panel) of L02 and HepG2 cells stably transfected with an empty vector (Ctrl) or plasmid encoding ZNF300. ***P < 0.001 versus control group. (B and C) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs or BSA with no fatty acid for 24 h (B) and quantified with a spectrophotometer at 520 nm (C). (D) The intracellular TG content was assayed using the triglyceride assay kit. (E) Quantitative real-time RT-PCR analysis of genes involved in lipogenesis, fatty acid oxidation, lipolysis in indicated cells. *P < 0.05 versus control/BSA; #P < 0.05, ##P < 0.01 versus control/FFAs.

Figure 2.
ZNF300 protects hepatocytes from FFA-induced steatosis.

(A) qRT-PCR analysis (left graph) of ZNF300 mRNA and immunoblot analysis (with anti-exogenous and endogenous ZNF300 protein, anti-β-actin protein) (right panel) of L02 and HepG2 cells stably transfected with an empty vector (Ctrl) or plasmid encoding ZNF300. ***P < 0.001 versus control group. (B and C) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs or BSA with no fatty acid for 24 h (B) and quantified with a spectrophotometer at 520 nm (C). (D) The intracellular TG content was assayed using the triglyceride assay kit. (E) Quantitative real-time RT-PCR analysis of genes involved in lipogenesis, fatty acid oxidation, lipolysis in indicated cells. *P < 0.05 versus control/BSA; #P < 0.05, ##P < 0.01 versus control/FFAs.

Knockdown of ZNF300 aggravates hepatic lipid accumulation

To further validate the effect of ZNF300 on the hepatic lipid metabolism, we transfected hepatocytes with short hairpin RNA (shRNA) targeting ZNF300. The knockdown efficiencies of ZNF300 in L02 and HepG2 cells were confirmed by real-time PCR and Western blot (Figure 3A,B). We found that, compared with the scramble group, knockdown of ZNF300 significantly aggravated the FFA-induced hepatic steatosis in HepG2 and L02 cells (Figure 3C–E). Furthermore, the silence of ZNF300 further enhanced the expression of genes related to lipogenesis and also slightly decreased the expression levels of genes involved in the fatty acid oxidation (Figure 3F). Consistently, knockdown of ZNF300 still had no obvious effect on the lipolysis compared with the control group (Figure 3F). These data provided additional evidence that ZNF300 plays an important role in fatty acid metabolism in hepatocytes.

Knockdown of ZNF300 aggravates hepatic lipid accumulation.

Figure 3.
Knockdown of ZNF300 aggravates hepatic lipid accumulation.

(A and B) Western blot (A) and qRT-PCR (B) analysis of the expression of ZNF300 of L02 (left panel) and HepG2 cells (right panel) stably transfected with shRNA control vector (scramble) or shRNA targeting ZNF300 (shZNF300-1# or shZNF300-2#). **P < 0.01 versus scramble group. (C and D) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs or BSA with no fatty acid for 24 h (upper panel), and quantified with a spectrophotometer at 520 nm (bottom graph). #P < 0.05 versus scramble group. (E) The intracellular TG content was assayed using the triglyceride assay kit. (F) qRT-PCR analysis of genes involved in lipogenesis, fatty acid oxidation, lipolysis in indicated cells. #P < 0.05 versus scramble FFAs.

Figure 3.
Knockdown of ZNF300 aggravates hepatic lipid accumulation.

(A and B) Western blot (A) and qRT-PCR (B) analysis of the expression of ZNF300 of L02 (left panel) and HepG2 cells (right panel) stably transfected with shRNA control vector (scramble) or shRNA targeting ZNF300 (shZNF300-1# or shZNF300-2#). **P < 0.01 versus scramble group. (C and D) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs or BSA with no fatty acid for 24 h (upper panel), and quantified with a spectrophotometer at 520 nm (bottom graph). #P < 0.05 versus scramble group. (E) The intracellular TG content was assayed using the triglyceride assay kit. (F) qRT-PCR analysis of genes involved in lipogenesis, fatty acid oxidation, lipolysis in indicated cells. #P < 0.05 versus scramble FFAs.

ZNF300 directly regulates the expression of PPARα

Since PPARα expression was influenced by the alterations in ZNF300 expression even without the FFA treatment (Figures 2E, 3F and 4A), we speculate that ZNF300 may directly regulate PPARα expression. First, we examined the effect of ZNF300 on some response element related to metabolism or proliferation. As shown in Figure 4B,C, overexpression of ZNF300 significantly enhanced PPARα and slightly inhibited LXR activity in reporter assays, while knockdown of ZNF300 has an opposite effect. Next, we analyzed the sequences of PPARα and LXR promoters and found that there is one conserved ZNF300-binding site (CAGGGGGCG) in the upstream 142–150 bp of PPARα promoter (Figure 4D). Furthermore, there still exists five non-conserved sites in the PPARα promoter (+240 bp to −100 bp), which contains the core sequencing (GGGGG) of ZNF300 binding (Figure 4H). To further confirm whether ZNF300 was able to directly bind and regulate the PPARα, luciferase assay in HepG2 cells was performed. We observed that overexpression of ZNF300 increased the activity of PPARα in a dose-dependent manner (Figure 4E). Consistently, the activity of PPARα was increased by ZNF300 overexpression and reduced by the knockdown of ZNF300 in HepG2 cells (Figure 4F). In addition, we also found that mutation or deletion of the conserved binding sites of ZNF300 greatly inhibited the activity of PPARα (Figure 4G). To further substantiate this conclusion, an exogenous Flag-ZNF300 was overexpressed and chromatin immunoprecipitation (ChIP) assay with Flag antibody was performed to detect whether ZNF300 binds to these sites. As shown in Figure 4H, ZNF300 showed a strong binding activity to the conserved binding site and a weak binding activity to the non-conserved binding sites. These data together suggest that ZNF300 directly regulates the expression of PPARα.

ZNF300 regulates the expression of PPARα.

Figure 4.
ZNF300 regulates the expression of PPARα.

(A) ZNF300 was overexpressed (left panel) or knocked down (right panel) in L02 and HepG2 cells; Western blot was used to detect the protein levels of PPARα and ZNF300 in these cells treated with FFAs. (B and C) Luciferase reporter assays analyzing the effect of ZNF300 overexpression (B) or knockdown (C) on the activity of HepG2 cells transfected with indicated response element in the graph for 24 h. *P < 0.05,**P < 0.01 versus ctrl or scramble group. (D) Schematic illustration of human PPARα gene promoter (WT) (upper), point mutation (M-DBD) (middle) and deletion mutation (Δ-DBD) (bottom). (E) Luciferase reporter assays analyzing PPARα promoter activity (left graph) and immunoblot assay with anti-Flag (right graph) of HEK293 T cells transfected with empty vector or plasmid encoding Flag-ZNF300 (0.025–0.4 μg) for 24 h. #P < 0.05, ##P < 0.01 versus ZNF300 (0 μg) group. (F and G) ZNF300 regulates PPARα transactivation. HepG2 cells that stably expressed ectopic ZNF300, ZNF300-shRNA and their individual controls were transfected with the PPARα promoter luciferase reporter vector (F) as well as the mutation of PPARα promoter (G). After 24 h, luciferase activity was measured in these cells. **P < 0.01 versus control or scramble group; #P < 0.05 versus WT/ZNF300. (H) ChIP analysis of ZNF300 binding on the promoter of PPARα gene in 293 T cells transfected with Flag-ZNF300 (left graph); schematic analysis of the primer sites in the PPARα promoter (right graph).*P < 0.05, **P < 0.01 versus the IgG group.

Figure 4.
ZNF300 regulates the expression of PPARα.

(A) ZNF300 was overexpressed (left panel) or knocked down (right panel) in L02 and HepG2 cells; Western blot was used to detect the protein levels of PPARα and ZNF300 in these cells treated with FFAs. (B and C) Luciferase reporter assays analyzing the effect of ZNF300 overexpression (B) or knockdown (C) on the activity of HepG2 cells transfected with indicated response element in the graph for 24 h. *P < 0.05,**P < 0.01 versus ctrl or scramble group. (D) Schematic illustration of human PPARα gene promoter (WT) (upper), point mutation (M-DBD) (middle) and deletion mutation (Δ-DBD) (bottom). (E) Luciferase reporter assays analyzing PPARα promoter activity (left graph) and immunoblot assay with anti-Flag (right graph) of HEK293 T cells transfected with empty vector or plasmid encoding Flag-ZNF300 (0.025–0.4 μg) for 24 h. #P < 0.05, ##P < 0.01 versus ZNF300 (0 μg) group. (F and G) ZNF300 regulates PPARα transactivation. HepG2 cells that stably expressed ectopic ZNF300, ZNF300-shRNA and their individual controls were transfected with the PPARα promoter luciferase reporter vector (F) as well as the mutation of PPARα promoter (G). After 24 h, luciferase activity was measured in these cells. **P < 0.01 versus control or scramble group; #P < 0.05 versus WT/ZNF300. (H) ChIP analysis of ZNF300 binding on the promoter of PPARα gene in 293 T cells transfected with Flag-ZNF300 (left graph); schematic analysis of the primer sites in the PPARα promoter (right graph).*P < 0.05, **P < 0.01 versus the IgG group.

PPARα mediates the effects of ZNF300 on hepatic lipid metabolism

To explore whether PPARα mediates the effects of ZNF300 on hepatic lipid metabolism, we performed PPARα reverse experiments in HepG2 cells. We transfected the siRNA of PPARα into the HepG2 cells stably expressing ZNF300 and the inhibition efficiency of PPARα was shown in Figure 5A. We also found that silencing PPARα led to the inhibition of its target genes related to hepatic fatty acid oxidation (Figure 5B). Furthermore, Oil Red O staining showed that the FFA-induced lipid accumulation and the triglyceride contents in ZNF300 overexpression cells were reversed by si-PPARα (Figure 5C,D). Taken together, these data indicate that PPARα plays an important role in mediating the effects of ZNF300 on hepatic lipid metabolism.

PPARα mediates the effects of ZNF300 on hepatic lipid metabolism.

Figure 5.
PPARα mediates the effects of ZNF300 on hepatic lipid metabolism.

(A) ZNF300 was overexpressed in HepG2 cells with or without the knockdown of PPARα using si-PPARα; Western blot analysis of the protein levels of PPARα. (B) ZNF300 regulates the expression of PPARα target genes in HepG2 cells. The mRNA levels of PPARα target genes were measured by quantitative real-time RT-PCR. (C and D) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs for 24 h (C) and the intracellular TG content was assayed using the triglyceride assay kit (D).

Figure 5.
PPARα mediates the effects of ZNF300 on hepatic lipid metabolism.

(A) ZNF300 was overexpressed in HepG2 cells with or without the knockdown of PPARα using si-PPARα; Western blot analysis of the protein levels of PPARα. (B) ZNF300 regulates the expression of PPARα target genes in HepG2 cells. The mRNA levels of PPARα target genes were measured by quantitative real-time RT-PCR. (C and D) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs for 24 h (C) and the intracellular TG content was assayed using the triglyceride assay kit (D).

DNA methylation regulates the expression of ZNF300 in hepatic steatosis

We next attempted to gain insight into the mechanism of ZNF300 down-regulation in FFA-induced hepatic steatosis. TCGA data indicate that ZNF300 DNA methylation is significantly higher in the liver hepatocellular carcinoma (LIHC) cohort (0.282) than those in normal samples (0.19) (P < 0.05) (Figure 6A) and there is a strong negative correlation between ZNF300 methylation and expression (Figure 6B). To investigate whether DNA methylation controls variation of ZNF300 in the development of hepatic steatosis, we performed the BSP assay to examine the methylation status of ZNF300 CpG island, which is located at −123 bp to +93 bp from TSS and contains 25 CpG sites in total (Figure 6C). As shown in Figure 6D, the average methylation rate across the 25 CpG sites increased significantly in HepG2 cells with FFA treatment (19.29%), compared with the control group (3.21%). In addition, we also found FFA treatment increased SAM level and decreased SAH concentration, thereby increasing methylation potential in the hepatic steatosis (Figure 6E). To further confirm that the down-regulation of ZNF300 in hepatic steatosis is due to its hypermethylation status, hepatocytes were treated with 5AzadC (10 μM) and FFAs. When treated with 5-AzadC, hepatocytes showed a significant increase in ZNF300 expression compared with those with FFAs only (Figure 6F). In addition, 5-AzadC treatment also alleviates the lipid accumulation induced by FFAs in HepG2 cells (Figure 6G). In summary, these data suggest that DNA methylation regulates the expression of ZNF300 in hepatic steatosis.

DNA methylation regulates the expression of ZNF300 in hepatic steatosis.

Figure 6.
DNA methylation regulates the expression of ZNF300 in hepatic steatosis.

(A and B) Samples with different methylation level in liver hepatocellular carcinoma (LIHC) (A) and the relation between differential methylation and expression of ZNF300 promoter in LIHC (B), these results are from the cancer genome altas (TCGA) data. (C) Schematic displays of the GC content of ZNF300 promoter predicted using Methprimer software. (D) Bisulfite sequencing analysis of methylation status of CpG islands in the promoter of ZNF300 with or without FFA treatment in HepG2 cells. For each group, 10 clones were selected for sequencing. At each CpG site, the area filled with black represented the average percentage of methylation across all CpG sites examined in clones. The value of the average percentage was shown at the end of each row. (E) Concentrations of SAM (left panel) and SAH (right panel) in the L02/ HepG2 cells treated with or without FFAs were determined by ELIAC. (F) mRNA (upper panel) and protein levels (lower panel) analysis showing the expression level changes of ZNF300 in the L02 and HepG2 cell lines treated with FFAs or 5-AzadC. (G) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs or 5-AzadC.

Figure 6.
DNA methylation regulates the expression of ZNF300 in hepatic steatosis.

(A and B) Samples with different methylation level in liver hepatocellular carcinoma (LIHC) (A) and the relation between differential methylation and expression of ZNF300 promoter in LIHC (B), these results are from the cancer genome altas (TCGA) data. (C) Schematic displays of the GC content of ZNF300 promoter predicted using Methprimer software. (D) Bisulfite sequencing analysis of methylation status of CpG islands in the promoter of ZNF300 with or without FFA treatment in HepG2 cells. For each group, 10 clones were selected for sequencing. At each CpG site, the area filled with black represented the average percentage of methylation across all CpG sites examined in clones. The value of the average percentage was shown at the end of each row. (E) Concentrations of SAM (left panel) and SAH (right panel) in the L02/ HepG2 cells treated with or without FFAs were determined by ELIAC. (F) mRNA (upper panel) and protein levels (lower panel) analysis showing the expression level changes of ZNF300 in the L02 and HepG2 cell lines treated with FFAs or 5-AzadC. (G) The intracellular lipid droplets were stained with Oil Red O in indicated cells with FFAs or 5-AzadC.

Silencing of DNMT3a blocked FFA-induced down-regulation of ZNF300

In mammalians, DNA methylation is catalyzed by three different DNA methyltransferases: DNMT1, DNMT3a and DNMT3b [21]. First, we tested the expression level of these DNA methyltransferases in the progression of hepatic steatosis. As illustrated in Figure 7A,B, FFA treatment caused a significant increase in DNMT1 and DNMT3a expression in L02 and HepG2 cells. However, DNMT3b levels did not show significant variation. Next, we investigated which DNMT might be responsible for the inhibition of ZNF300 in hepatic steatosis by siRNA experiment. As compared with scrambled-siRNA, both siRNA (si-DNMT1 and si-DNMT3a) induced a significant decrease in FFA-induced L02 or HepG2 cells (Figure 7C–F). However, ZNF300 expression up-regulation was observed only in the cells transfected with si-DNMT3a (Figure 7D,F). These results indicated that DNMT3a might contribute to the down-regulation of ZNF300 through DNA methylation in FFA-induced hepatic steatosis.

Silencing of DNMT3a blocked FFA-induced down-regulation of ZNF300.

Figure 7.
Silencing of DNMT3a blocked FFA-induced down-regulation of ZNF300.

(A and B) Gene expressions of DNMT1, DNMT3a and DNMT3b were measured by real-time PCR (A) and Western blot (B) in FFA-treated L02 and HepG2 cells. *P < 0.05, **P < 0.01 versus the BSA group. (CF) L02 and HepG2 cells were transfected with si-DNMT1, si-DNMT3a or scrambled-siRNA, respectively. Whole cell extracts were isolated 48 h after transfection. mRNA (C and D) and protein (E and F) levels of DNMT1, DNMT3a, ZNF300 and β-actin were measured, *P < 0.05, **P < 0.01 versus BSA only group; #P < 0.05, ##P < 0.01 versus si-DNMT1/si-DNMT3a only group.

Figure 7.
Silencing of DNMT3a blocked FFA-induced down-regulation of ZNF300.

(A and B) Gene expressions of DNMT1, DNMT3a and DNMT3b were measured by real-time PCR (A) and Western blot (B) in FFA-treated L02 and HepG2 cells. *P < 0.05, **P < 0.01 versus the BSA group. (CF) L02 and HepG2 cells were transfected with si-DNMT1, si-DNMT3a or scrambled-siRNA, respectively. Whole cell extracts were isolated 48 h after transfection. mRNA (C and D) and protein (E and F) levels of DNMT1, DNMT3a, ZNF300 and β-actin were measured, *P < 0.05, **P < 0.01 versus BSA only group; #P < 0.05, ##P < 0.01 versus si-DNMT1/si-DNMT3a only group.

Discussion

NAFLD is defined as a fatty liver not due to excess alcohol consumption or other cause of steatosis [22], and it has emerged as one of the most common liver diseases worldwide. It encompasses a broad spectrum of conditions, from simple steatosis, through NASH, to fibrosis and ultimately cirrhosis and hepatocellular carcinoma [23]. As NAFLD is often associated with obesity, dyslipidemia, insulin resistance and type 2 diabetes [24], it is considered as the hepatic manifestation of metabolic syndrome [25]. The pathogenesis of NAFLD is thought to be a multifactorial, such as lifestyle habits, nutritional factors and genetics. However, the pathogenesis and underlying mechanism of NAFLD caused by genetics have not been fully elucidated [26].

ZNF300, which is located on chromosome 5q33.1, possesses an amino-terminal KRAB domain and 12 carboxyl-terminal C2H2 zinc finger motifs. ZNF300 is localized to the nucleus, and the KRAB domain of this protein has transcription repressor activities [15,18]. In addition, it has been reported to promote tumor development by activating the NF-κB pathway, in which it plays a vital role during tumor development in inflamed tissue [14]. The prevalence and molecular mechanisms underlying inflammation have been extensively studied because inflammation is a risk factor for a subset of diseases. Furthermore, one study indicated that the aberrant methylation of ZNF300 leads to the dysregulation of its expression, which contributes to HBV-related HCC development [20]. The injury of hepatic cells by FFA-induced lipid accumulation results in inflammation and may later progress to NASH even to HCC. However, little was known about the role of ZNF300 in the development of NAFLD.

In the present study, we described a novel role and mechanism for ZNF300 in the regulation of hepatic lipid metabolism. Our findings indicate that ZNF300 expression levels decreased in FFA-induced fatty liver models, and ZNF300 overexpression or down-regulation remarkably protected or aggravated the hepatocellular lipid accumulation. Excessive accumulation of hepatic triglycerides can occur as a result of increased fat synthesis or reduced fatty acid oxidation [4,27]. It is well established that PPARα plays a central role in the fatty acid catabolism by directly stimulating the transcription of genes involved in fatty acid oxidation [28]. In the present study, we found for the first time that ZNF300 could bind to the promoter of PPARα (Figure 4G,H) and positively regulate its expression (Figure 4A). As well, we also identified that ZNF300 was involved in the regulation of fatty acid oxidation mainly through PPARα activation (Figure 5). In addition, overexpression of ZNF300 could inhibit the activity of LXR to some extent; however, there are no binding sites of ZNF300 on the promoter of LXR gene, which suggests that ZNF300 may indirectly regulate the activity of LXR. Since LXR plays important roles in regulating the transcriptional effect on the enzymes involved in fatty acid synthesis and triglyceride synthesis, such as Fas, acetylCoA carboxylase (ACC) and steroyl-CoA desaturase 1 (Scd1) [29,30], this explained the phenomenon that ZNF300 could inhibit the lipogenesis (Figure 2E). Interestingly, we found ZNF300 could significantly inhibit AP1 activity in the reporter assays (Figure 4B,C). In the present study, hepatic AP1 mRNA expression was then analyzed. We found Fra-1 and JunD mRNA levels were markedly reduced in the FFA-induced liver cells, while the expressions of c-fos, FosB, Fra-2, c-Jun and JunB were not affected by the FFAs (Supplementary Figure S1A,B). Further studies indicated that ZNF300 could directly bind to the promoter of JunD and play a negative regulation role in its expression (Supplementary Figure S1C–E). Several studies have demonstrated that JunD is essential for NAFLD development [31,32], indicating ZNF300 may play a role in hepatic steatosis through AP1 pathway. Altogether, our data demonstrate that ZNF300 is an important regulator for the hepatic steatosis and the beneficial effects of ZNF300 are at least partially due to promoting fatty acid oxidation as well as the suppression of hepatic de novo lipogenesis. However, ZNF300 functional study in vivo has been impeded due to its lack of orthologous in mice.

The emerging field of epigenetics, an inheritable phenomenon that can affect gene expression without altering the DNA sequence, provides a new perspective on the pathogenesis of NAFLD [33]. DNA methylation, as one of the major epigenetic changes, plays a central role in the regulation of gene expression, representing a level of epigenetic regulation commonly associated with transcriptional repression [34]. The CpG island is commonly present with higher frequency at the promoter regions of the genes than at other DNA sites. Hypermethylation of CpG islands is generally related to gene silencing, whereas hypomethylation is associated with gene activation. Aberrant patterns of DNA methylation seem to be the starting point for carcinogenesis, especially in NAFLD-related liver diseases [35]. Recently, we reported that the abnormal DNA methylation of ZNF300 caused its expression dysregulation and led to leukemogenesis [15]. In the present study, we found that ZNF300 promoter methylation was increased in FFA-induced hepatic steatosis compared with that with no FFA treatment (Figure 6D). In addition, we also demonstrated that the expression of ZNF300 reduced owing to DNA methylation in hepatic steatosis (Figures 6F and 7D,F), which may be a critical event in NAFLD progression. A recent epigenomic study also suggested that DNA methylation is significantly associated with NAFLD activity score and it has shown obvious differences in several CpG sites [36]. It appears that the DNA methylation status at specific CpG could be useful for predicting the progression of NAFLD. So further studies are necessary to explore the DNA methylation status of ZNF300 at specific CpG using liver biopsy samples from NAFLD patients.

In the present study, we found that elevated DNMT3a was involved in the down-regulation of ZNF300 in FFA-induced hepatic steatosis (Figure 7). We also measured the methylation cycler metabolites and found that the concentration of SAM and the SAM: SAH ratio were increased (Figure 6E). Surprisingly, these results are opposite to that in the liver of HFD-fed mice (Supplementary Figure S2). This may be due to the difference between in vitro and in vivo as well as the different stimulus. For example, Glycine N-methyltransferase (GNMT) is a key protein in transmethylation by removing excess SAM and maintaining a constant SAM: SAH ratio [37,38], the deficiency of which in mice leads to the accumulation of hepatic SAM and the development of steatosis [39,40]. Nevertheless, various studies have shown that the GNMT is absent from patients with spontaneous liver disease and HCC cell lines [4042]. We suspect that the loss of GNMT in our cell lines may lead to the accumulation of SAM and increase in the SAM:SAH ratio. But the exact mechanism of SAM accumulation in our study is not clear and needs more investigations.

Collectively, our study and previous study suggest that either too much or too little hepatic SAM as well as abnormal SAM:SAH ratios may result in the development of NAFLD. Moreover, our work indicates that the methylation status of ZNF300 plays a key role in the hepatic steatosis by regulating the PPARα activation. Thus, the expression and methylation level of ZNF300 can be aimed as a marker for diagnosis in clinical practice.

Abbreviations

     
  • ACC

    acetylCoA carboxylase

  •  
  • BSP

    bisulfite pyrosequencing PCR

  •  
  • ChIP

    chromatin immunoprecipitation

  •  
  • DMEM

    Dulbecco's modified Eagle's medium

  •  
  • FAS

    fatty acid synthase

  •  
  • FFAs

    free fatty acids

  •  
  • GNMT

    glycine N-methyltransferase

  •  
  • HCC

    hepatocellular carcinogenesis

  •  
  • HEK

    human embryonic kidney

  •  
  • KRAB-ZFPs

    Krüpple-associated box zinc finger proteins

  •  
  • LIHC

    liver hepatocellular carcinoma

  •  
  • LXR

    liver X receptor

  •  
  • NAFLD

    Non-alcoholic fatty liver disease

  •  
  • NASH

    non-alcoholic steatohepatitis

  •  
  • qRT-PCR

    quantitative real-time PCR

  •  
  • RNAi

    RNA interference

  •  
  • RT

    real time

  •  
  • SAH

    S-adenosylhomocysteine

  •  
  • SAM

    S-adenosylmethionine

  •  
  • shRNA

    short hairpin RNA

  •  
  • SREBP1c

    sterol regulatory element binding protein-1c

  •  
  • TCGA

    the cancer genome altas

  •  
  • TG

    triglyceride

  •  
  • ZNF300

    Zinc finger protein 300

Author Contribution

F-J. Y. and Y-J. W. conceived and supervised the study; F-J. Y., Y-J. W. and Y-L. Z. designed experiments; F-J. Y., Y-J. W., S-R. Y. performed experiments; F-J. Y. and J. L. performed data analysis and interpretation; F-J. Y. and Y-J. W. drafted the paper and all authors read and approved the final manuscript.

Funding

This work was supported by the Grants from the National Natural Science Foundation of China [nos 81800521 and 31870778], Natural Science Foundation by Jiangsu Normal University [17XLR033 and 17XLR032].

Acknowledgement

We thank Zan Huang professor, Dr Zhou Jiang, Min Wang from the College of Life Science Wuhan University for their technical help.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Brunt
,
E.M.
,
Kleiner
,
D.E.
,
Wilson
,
L.A.
,
Belt
,
P.
,
Neuschwander-Tetri
,
B.A.
and
Network
,
N.C.R.
(
2011
)
Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings
.
Hepatology
53
,
810
820
2
Lonardo
,
A.
and
Targher
,
G.
(
2017
)
NAFLD: is there anything new under the sun?
Int. J. Mol. Sci.
18
PMID:
[PubMed]
3
Yan
,
F.-J.
,
Zhang
,
X.-J.
,
Wang
,
W.-X.
,
Ji
,
Y.-X.
,
Wang
,
P.-X.
,
Yang
,
Y.
et al.  (
2017
)
The E3 ligase tripartite motif 8 targets TAK1 to promote insulin resistance and steatohepatitis
.
Hepatology
65
,
1492
1511
4
Benedict
,
M.
and
Zhang
,
X.
(
2017
)
Non-alcoholic fatty liver disease: an expanded review
.
World J. Hepatol.
9
,
715
732
5
Zhao
,
Z.
,
Xu
,
D.
,
Wang
,
Z.
,
Wang
,
L.
,
Han
,
R.
,
Wang
,
Z.
et al.  (
2018
)
Hepatic PPARα function is controlled by polyubiquitination and proteasome-mediated degradation via the coordinated actions of PAQR3 and HUWE1
.
Hepatology
68
,
289
303
PMID:
[PubMed]
6
Chawla
,
A.
,
Repa
,
J.J.
,
Evans
,
R.M.
and
Mangelsdorf
,
D.J.
(
2001
)
Nuclear receptors and lipid physiology: opening the X-files
.
Science
294
,
1866
1870
7
Sim
,
W.-C.
,
Park
,
S.
,
Lee
,
K.Y.
,
Je
,
Y.T.
,
Yin
,
H.-Q.
,
Choi
,
Y.-J.
et al.  (
2014
)
LXR-α antagonist meso-dihydroguaiaretic acid attenuates high-fat diet-induced nonalcoholic fatty liver
.
Biochem. Pharmacol.
90
,
414
424
8
Huang
,
J.
,
Kang
,
S.
,
Park
,
S.-J.
and
Im
,
D.-S.
(
2017
)
Apelin protects against liver X receptor-mediated steatosis through AMPK and PPARα in human and mouse hepatocytes
.
Cell. Signal.
39
,
84
94
9
Kersten
,
S.
and
Stienstra
,
R.
(
2017
)
The role and regulation of the peroxisome proliferator activated receptor alpha in human liver
.
Biochimie
136
,
75
84
10
Gross
,
B.
,
Pawlak
,
M.
,
Lefebvre
,
P.
and
Staels
,
B.
(
2017
)
PPARs in obesity-induced T2DM, dyslipidaemia and NAFLD
.
Nat. Rev. Endocrinol.
13
,
36
49
PMID:
[PubMed]
11
Hong
,
A.E.
,
Ryu
,
M.S.
,
Kim
,
S.J.
,
Hwang
,
S.Y.
and
Lim
,
I.K.
(
2018
)
PPARα-target gene expression requires TIS21/btg2 gene in liver of the C57Bl/6 mice under fasting condition
.
Mol. Cells
41
,
140
149
PMID:
[PubMed]
12
Kersten
,
S.
(
2014
)
Integrated physiology and systems biology of PPARα
.
Mol. Metab.
3
,
354
371
13
Qin
,
S.
,
Yin
,
J.
and
Huang
,
K.
(
2016
)
Free fatty acids increase intracellular lipid accumulation and oxidative stress by modulating PPARα and SREBP-1c in L-02 cells
.
Lipids
51
,
797
805
14
Wang
,
T.
,
Wang
,
X.-G.
,
Xu
,
J.-H.
,
Wu
,
X.-P.
,
Qiu
,
H.-L.
,
Yi
,
H.
et al.  (
2012
)
Overexpression of the human znf300 gene enhances growth and metastasis of cancer cells through activating NF-kB pathway
.
J. Cell. Mol. Med.
16
,
1134
1145
15
Yan
,
F.-J.
,
Fan
,
J.
,
Huang
,
Z.
and
Zhang
,
J.-J.
(
2017
)
ZNF300 tight self-regulation and functioning through DNA methylation and histone acetylation
.
Cell Biosci.
7
,
33
16
Zhang
,
X.
,
Ding
,
C.
,
Tian
,
H.
,
Dong
,
X.
,
Meng
,
X.
,
Zhu
,
W.
et al.  (
2017
)
ZNF23 suppresses cutaneous melanoma cell malignancy via mitochondria-dependent pathway
.
Cell. Physiol. Biochem.
43
,
147
157
17
Cesaro
,
E.
,
Sodaro
,
G.
,
Montano
,
G.
,
Grosso
,
M.
,
Lupo
,
A.
and
Costanzo
,
P.
(
2017
)
The complex role of the ZNF224 transcription factor in cancer
.
Adv. Protein Chem. Struct. Biol.
107
,
191
222
18
Cai
,
J.
,
Gong
,
R.
,
Yan
,
F.
,
Yu
,
C.
,
Liu
,
L.
,
Wang
,
W.
et al.  (
2014
)
ZNF300 knockdown inhibits forced megakaryocytic differentiation by phorbol and erythrocytic differentiation by arabinofuranosyl cytidine in K562 cells
.
PLoS ONE
9
,
e114768
19
Xu
,
J.-H.
,
Wang
,
T.
,
Wang
,
X.-G.
,
Wu
,
X.-P.
,
Zhao
,
Z.-Z.
,
Zhu
,
C.-G.
et al.  (
2010
)
PU.1 can regulate the ZNF300 promoter in APL-derived promyelocytes hl-60
.
Leuk. Res.
34
,
1636
1646
20
Zhao
,
Y.
,
Xue
,
F.
,
Sun
,
J.
,
Guo
,
S.
,
Zhang
,
H.
,
Qiu
,
B.
et al.  (
2014
)
Genome-wide methylation profiling of the different stages of hepatitis B virus-related hepatocellular carcinoma development in plasma cell-free DNA reveals potential biomarkers for early detection and high-risk monitoring of hepatocellular carcinoma
.
Clin. Epigenetics
6
,
30
21
Massey
,
V.
,
Cabezas
,
J.
and
Bataller
,
R.
(
2017
)
Epigenetics in liver fibrosis
.
Semin. Liver Dis.
37
,
219
230
22
Neuschwander-Tetri
,
B.A.
and
Caldwell
,
S.H.
(
2003
)
Nonalcoholic steatohepatitis: Summary of an AASLD single topic conference
.
Hepatology
37
,
1202
1219
23
Eslam
,
M.
,
Valenti
,
L.
and
Romeo
,
S.
(
2018
)
Genetics and epigenetics of NAFLD and NASH: clinical impact
.
J. Hepatol.
68
,
268
279
24
Postic
,
C.
and
Girard
,
J.
(
2008
)
Contribution of de novo fatty acid synthesis to hepatic steatosis and insulin resistance: lessons from genetically engineered mice
.
J. Clin. Invest.
118
,
829
838
25
Mikolasevic
,
I.
,
Milic
,
S.
,
Turk Wensveen
,
T.
,
Grgic
,
I.
,
Jakopcic
,
I.
,
Stimac
,
D.
et al.  (
2016
)
Nonalcoholic fatty liver disease — a multisystem disease?
World J. Gastroenterol.
22
,
9488
9505
26
Hardy
,
T.
,
Oakley
,
F.
,
Anstee
,
Q.M.
and
Day
,
C.P.
(
2016
)
Nonalcoholic fatty liver disease: pathogenesis and disease spectrum
.
Annu. Rev.Pathol.
11
,
451
496
27
Samuel
,
V.T.
and
Shulman
,
G.I.
(
2018
)
Nonalcoholic fatty liver disease as a nexus of metabolic and hepatic diseases
.
Cell Metab.
27
,
22
41
28
Kersten
,
S.
,
Seydoux
,
J.
,
Peters
,
J.M.
,
Gonzalez
,
F.J.
,
Desvergne
,
B.
and
Wahli
,
W.
(
1999
)
Peroxisome proliferator-activated receptor alpha mediates the adaptive response to fasting
.
J. Clin. Invest.
103
,
1489
1498
29
Matsusue
,
K.
,
Aibara
,
D.
,
Hayafuchi
,
R.
,
Matsuo
,
K.
,
Takiguchi
,
S.
,
Gonzalez
,
F.J.
et al.  (
2014
)
Hepatic PPARγ and LXRα independently regulate lipid accumulation in the livers of genetically obese mice
.
FEBS Lett.
588
,
2277
2281
30
Aibara
,
D.
,
Matsusue
,
K.
,
Takiguchi
,
S.
,
Gonzalez
,
F.J.
and
Yamano
,
S.
(
2018
)
Fat-specific protein 27 is a novel target gene of liver x receptor alpha
.
Mol. Cell. Endocrinol.
474
,
48
56
PMID:
[PubMed]
31
Hasenfuss
,
S.C.
,
Bakiri
,
L.
,
Thomsen
,
M.K.
,
Williams
,
E.G.
Auwerx
,
J.
and
Wagner
,
E.F.
(
2014
)
Regulation of steatohepatitis and PPARγ signaling by distinct AP-1 dimers
.
Cell Metab.
19
,
84
95
32
Thepot
,
D.
,
Weitzman
,
J.B.
,
Barra
,
J.
,
Segretain
,
D.
,
Stinnakre
,
M.G.
,
Babinet
,
C.
et al.  (
2000
)
Targeted disruption of the murine junD gene results in multiple defects in male reproductive function
.
Development
127
,
143
153
PMID:
[PubMed]
33
Lee
,
J.
,
Kim
,
Y.
,
Friso
,
S.
and
Choi
,
S.-W.
(
2017
)
Epigenetics in non-alcoholic fatty liver disease
.
Mol. Aspects Med.
54
,
78
88
34
Del Campo
,
J.A.
,
Gallego-Duran
,
R.
,
Gallego
,
P.
and
Grande
,
L.
(
2018
)
Genetic and epigenetic regulation in nonalcoholic fatty liver disease (NAFLD)
.
Int. J. Mol. Sci.
19
PMID:
[PubMed]
35
Xu
,
R.-H.
,
Wei
,
W.
,
Krawczyk
,
M.
,
Wang
,
W.
,
Luo
,
H.
,
Flagg
,
K.
et al.  (
2017
)
Circulating tumour DNA methylation markers for diagnosis and prognosis of hepatocellular carcinoma
.
Nat. Mater.
16
,
1155
1161
36
Pirola
,
C.J.
,
Gianotti
,
T.F.
,
Burgueño
,
A.L.
,
Rey-Funes
,
M.
,
Loidl
,
C.F.
,
Mallardi
,
P.
et al.  (
2013
)
Epigenetic modification of liver mitochondrial DNA is associated with histological severity of nonalcoholic fatty liver disease
.
Gut
62
,
1356
1363
37
Graham
,
I.
(
1999
)
Homocysteine in health and disease
.
Ann. Internal Med.
131
,
387
388
38
Rowling
,
M.J.
,
McMullen
,
M.H.
,
Chipman
,
D.C.
and
Schalinske
,
K.L.
(
2002
)
Hepatic glycine N-methyltransferase is up-regulated by excess dietary methionine in rats
.
J. Nutr.
132
,
2545
2550
39
Martínez-Chantar
,
M.L.
,
Vázquez-Chantada
,
M.
,
Ariz
,
U.
,
Martínez
,
N.
,
Varela
,
M.
,
Luka
,
Z.
et al.  (
2008
)
Loss of the glycine N-methyltransferase gene leads to steatosis and hepatocellular carcinoma in mice
.
Hepatology
47
,
1191
1199
40
Varela-Rey
,
M.
,
Martínez-López
,
N.
,
Fernández-Ramos
,
D.
,
Embade
,
N.
,
Calvisi
,
D.F.
,
Woodhoo
,
A.
et al.  (
2010
)
Fatty liver and fibrosis in glycine N-methyltransferase knockout mice is prevented by nicotinamide
.
Hepatology
52
,
105
114
41
Chen
,
Y.-M.
,
Shiu
,
J.-Y.
,
Tzeng
,
S.J.
,
Shih
,
L.-S.
,
Chen
,
Y.-J.
,
Lui
,
W.-Y.
et al.  (
1998
)
Characterization of glycine-N-methyltransferase-gene expression in human hepatocellular carcinoma
.
Int. J. Cancer
75
,
787
793
42
Avila
,
M.A.
,
Berasain
,
C.
,
Torres
,
L.
,
Martín-Duce
,
A.
,
Corrales
,
F.J.
,
Yang
,
H.
et al.  (
2000
)
Reduced mRNA abundance of the main enzymes involved in methionine metabolism in human liver cirrhosis and hepatocellular carcinoma
.
J. Hepatol.
33
,
907
914

Supplementary data