The study was conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, after approval by the Institutional Review Board of Kyungpook National University; every effort was made to minimize both the number of animals used and their suffering. Nine-week-old male Sprague-Dawley (SD) rats were unilaterally nephrectomized under ketamine (150 mg/kg; Yuhan, Seoul, Korea) and xylazine (18 mg/kg; Bayer, Seoul, Korea) anesthesia, as previously described [21]. Rats received subcutaneous infusion of ACTH (40 ng/kg/day) via an osmotic mini-pump. Sodium valproate (VPA) was purchased from Sigma-Aldrich (St. Louis, MO, USA). VPA (0.71% w/v) was administered via the drinking water for 4 weeks. Rats were anesthetized with sodium pentobarbital (50 mg/kg intraperitoneally). Tissues were frozen in liquid nitrogen and stored at −80℃ until further study.

HepG2 cells were purchased from the American Type Culture Collection (Manassas, VA, USA) and maintained under 5% CO₂ at 37℃ in Dulbecco's modified Eagle's medium with L-glutamine and 4,500 mg/L glucose containing 10% fetal bovine serum (Invitrogen, Carlsbad, CA, USA). They were sub-cultured every 3 or 4 days with trypsin (0.25%) and EDTA (0.02%). A fraction of the detached cells was transferred to normal Tyrode's solution and kept at 4℃ in microtubes. We treated HepG2 cells with the HDA-Ci; VPA (Sigma-Aldrich Inc., MO, USA), SAHA (Sigma-Aldrich Inc., MO, USA), TSA (Sigma-Aldrich Inc., MO, USA), MS275 (Selleckchem, TX, USA) and MC1568 (Santa Cruz Biotechnology, CA, USA), then incubated the cells with or without DEX.

For trichrome, Oil Red O, and hematoxylin&eosin (H&E) stains, liver tissues were fixed in 4% formalin overnight, then dehydrated and embedded in paraffin. The paraffin-embedded samples were sectioned at a thickness of 3 mm. The slides were examined using light microscopy.

Briefly, to examine the accumulation of fat, HepG2 cells were rinsed with cold phosphate-buffered saline (PBS) and fixed in 10% paraformaldehyde for 30 min. After the cells were washed with 60% isopropanol, they were stained for at least 1 h in a freshly diluted Oil Red O solution (6 parts 0.5% Oil Red O in isopropanol stock solution and 4 parts H₂O). After the stain was removed and the cells were washed with 60% isopropanol, images of each group of cells were acquired (Nikon Imaging Korea, Seoul, Korea).

RT-qPCR was performed using an ABI PRISM® 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) as previously described [21]. Amplification reactions containing 10 µL of SYBR® Green PCR Master Mix (Applied Biosystems), 4 µL of cDNA, and the indicated primer set (200 nM) were performed in a total volume of 20 µL. All samples were amplified in triplicate in a 96-well plate, with the following cycling conditions: 2 min at 50℃, 10 min at 95℃, and 40 cycles at 95℃ for 15 s followed by 1 min at 60℃. The relative mRNA expression level was determined by calculating the values of Δcycle threshold (ΔCt) by normalizing the average Ct value compared to its endogenous control (Gapdh), then calculating 2^−ΔΔCt. Statistical analyses were performed using one-way analysis of variance (ANOVA). The primer sets are given in the supplementary data (Table 1S).

The frozen tissues were homogenized in RIPA buffer containing protease inhibitors. The cell lysates were precleared with protein G agarose at 4℃ for 2 h. The supernatants were incubated overnight with 1 mg of GR antibody (Santa Cruz Biotechnology, USA), or acetyl-lysine (ac-K) antibody (Cell Signaling Technology, USA), at 4℃. The immunocomplexes were washed 3 times with lysis buffer (20 mM Tris, 150 mM NaCl, 1% NP-40, 0.5% SDS, 0.1% SDS, and 1 proteinase inhibitor cocktail), then subjected to western blot analysis. For western blots, protein-matched samples (Bradford assay) were subjected to SDS-PAGE, then transferred to nitrocellulose membranes. The nitrocellulose membranes were blocked with 5% skim milk in TBS (25 mM Tris base and 150 mM NaCl) for 2 h at room temperature, then incubated with 1 mg/mL of GR or ac-K antibody at 4℃ overnight. Then, the membranes were incubated in secondary antibody (diluted 1:5000) at room temperature for 1 h, and washed 3 times for 10 min each in TBST. The target proteins were detected with Enhanced Chemiluminescence Plus detection reagents (Amersham, Pittsburgh, PA, USA). The expression levels were quantified by optical densitometry (ImageJ software; http://rsbweb.nih.gov). We performed western blotting with the following antibodies: GR antibody (Santa Cruz Biotechnology, CA, USA), SREBP1 antibody (Santa Cruz Biotechnology, CA, USA), ac-K antibody (Cell Signaling Technology, MA, USA), ACC antibody (Thermo Fisher, MA, USA) FAS antibody (Thermo Fisher, MA, USA), GAPDH antibody (Thermo Fisher, MA, USA), SCD1 antibody (Abcam, Cambridge, UK), and β-actin antibody (Sigma-Aldrich Inc., MO, USA).

Oligonucleotides were either commercially synthesized or, in the case of the oligonucleotide glucocorticoid response element (GRE), obtained from Genotech (Pohang, Korea). The oligonucleotides that we used in the EMSAs included 35 nucleotides of GCs. Complementary oligonucleotides in equimolar amounts were heated to 100℃, cooled overnight to 25℃, divided into aliquots, and stored at −20℃ before use. Double-stranded oligonucleotide probes were end-labeled using T4 polynucleotide kinase and [γ-³²P]-ATP. Binding reactions were performed in a 20-mL volume containing approximately 20,000 cpm of labeled probe, 6~12 mg of nuclear extract, 20 mM HEPES (pH 7.9), 60 mM KCl, 5 mM MgCl₂, 2 mM dithiothreitol, 10% glycerol, 200 ng poly(dIzdC), 1 mg bovine serum albumin, and unlabeled competitor oligonucleotides. We followed the instructions provided by the manufacturer for EMSAs with the human recombinant GR. The binding reaction mixtures were incubated at 25℃ for 30 min, then loaded onto 6% non-denaturing polyacrylamide gels in 25 mM Tris (pH 8.3), 25 mM boric acid, and 0.5 mM EDTA. For supershift experiments, incubations with polyclonal anti-human GR antibody from Abcam (Cambridge, UK) or preimmune antiserum (diluted 1:500) were performed for 30 min at room temperature after addition of radiolabeled GRE 5 and nuclear extract. Gels were dried and autoradiographed overnight at 270℃, using Fujifilm RX-Fuji medical X-ray film and intensifying screens. Epstein-Barr Nuclear Antigen (EBNA) extract was used as a positive control.

ChIP assays were performed according to the manufacturer's instructions, with minor modifications, using the EZ-ChIP™ kit (Upstate Biotechnology, Lake Placid, NY, USA) as previously described [22]. In brief, tissues were fixed with 1% formaldehyde and washed with ice-cold PBS. After homogenization, tissues were incubated in SDS lysis solution for 10 min on ice. The lysates were sonicated with 15 cycles of 100 amplitude for 10 s followed by cooling on ice for 50 s. The lysates were precleared with protein G agarose beads for 2 h. Then, antibodies were added and incubated at 4℃ overnight. The antibody against GR was obtained from Abcam. Anti-polymerase II antibodies were obtained from Upstate Biotechnology. To reverse the crosslinking between DNA and chromatin, elutes were incubated at 65℃ for 5 h after the addition of NaCl to a final concentration of 0.2 M. The proteins were eliminated by digestion with proteinase K at 45℃ for 2 h and the DNA was purified with a spin column. Specific promoter DNA was quantified by qPCR. All samples were amplified in triplicate in 96-well plates, with the following cycling conditions: 2 min at 50℃, 10 min at 95℃, and 40 cycles at 95℃ for 15 s followed by 1 min at 60℃. The primer sets are given in the supplementary data (Table 2S).

For the luciferase assay, HepG2 cells were seeded on collagen I-coated 12-well plates, cultured overnight, and transfected with vector constructs using Superfect Transfection Reagent (Qiagen, Hilden, Germany), according to the manufacturer's recommendation. The next day, cells were exposed to dexamethasone (DEX, 10 nmol/L) for 24 h with or without pretreatment with VPA (10 mmol/L). The cells were lysed with a lysis buffer and the luciferase activity was analyzed with a Bright-Glo™ Luciferase Assay System kit (Promega, Madison, WI, USA) according to the manufacturer's instructions. Luminescence was measured with a Veritas™ Microplate Luminometer (Turner Biosystems, Sunnyvale, CA, USA).

Results are expressed as mean±standard error (SE). Data were analyzed with the Kruskal-Wallis test or one-way ANOVA followed by post hoc Tukey's comparison test; differences were considered significant at p<0.05. Student's t-test was applied for the analysis of the significance of differences between 2 groups. The procedures were performed using SPSS software (release 19.0, SPSS, Chicago, IL, USA).

To determine whether VPA has an anti-steatotic effect in vivo, we administered VPA into control SD rats or SD rats subcutaneously infused with ACTH for 4 weeks. Upon assessment of liver morphology and Oil Red O staining, we found that the administration of VPA considerably reduced ACTH-induced steatosis. We also examined collagen deposition and fibrosis, which would appear blue after trichrome staining. ACTH infusion did not affect liver collagen deposition and fibrosis (Fig. 1). We analyzed the expression of the lipogenic genes Acc, Fasn, Scd1, and Srebp1c by RT-qPCR. The ACTH-induced increase in the expression of Acc, Fasn, Scd1, and Srebp1c was attenuated by VPA (Figs. 2A~D). We confirmed the effect at the protein level by western blot analysis (Fig. 2E). Importantly, hepatic steatosis was reduced by VPA, which was associated with significantly reduced mRNA expression and protein levels of lipogenesis in the livers. These results suggested that HDAC inhibition may prevent hepatic steatosis in ACTH-infused rats by suppressing lipogenesis.

We next analyzed the enrichment of GR and Pol II on the promoters for Acc, Fasn, Scd1, and Srebp1c by ChIP assay followed by qPCR (Fig. 3). The qPCR results showed that VPA attenuated the ACTH-induced enrichment of GR and Pol II at the Acc, Fasn, and Srebp1c promoters in rats compared with controls (Figs. 3A, 3B and 3D). VPA did not affect the enrichment of GR induced by ACTH at the Scd1 promoter (Fig. 3C). These findings suggested that treatment with an HDACi decreased the recruitment of GR and Pol II to the promoters of lipogenesis-related genes in vivo.

To further explore the molecular mechanism of the HDACi-induced reduction of liver steatosis, we developed an in vitro model of hepatic steatosis by exposing HepG2 cells to DEX (10 nmol/L). We investigated the expression of Acc, Fasn, Scd1, and Srebp1c using RT-qPCR. In HepG2 cells, Oil Red O stain indicated that DEX induced lipid accumulation in a dose-dependent manner (Fig. 4A). In addition, DEX increased the expression of lipogenesis genes in a dose- and time-dependent manner (Figs. 4B and C). Thus, DEX increased lipogenesis and lipid accumulation in HepG2 cells.

We used VPA to determine whether HDAC inhibition could influence the DEX-mediated induction of lipogenesis. Pretreatment with VPA for 6 h attenuated DEX-induced expression of Acc (Fig. 5A), Fasn (Fig. 5B), Scd1 (Fig. 5C), and Srebp1c (Fig. 5D). DEX increased expression of lipogenesis genes, which was attenuated by VPA in HepG2 cells. Thus, we consider VPA a potential therapeutic agent for the treatment of lipogenesis. However, VPA is a pan-HDACi. To identify which type of HDACi affected DEX-induced lipid accumulation, we treated HepG2 cells with the pan-HDACi VPA, SAHA, and TSA; the HDAC class I-specific inhibitor MS275; and the HDAC class II a-specific inhibitor MC1568 for 6 h, then incubated the cells with or without DEX for 48 h. Pan-HDACi and the HDAC class I-specific inhibitor decreased DEX-induced lipid accumulation (Fig. 6A) and expression of Srebp1c (Fig. 6B) in HepG2 cells.

We analyzed the enrichment of GR and Pol II at the promoters of Srebp1c by ChIP assay followed by qPCR in HepG2 cells (Fig. 7A). The qPCR results revealed that the enrichment of GR and Pol II on the Srebp1c promoters induced by DEX, compared with the control, was attenuated by VPA. To examine the effect of VPA on DEX-induced GR DNA binding activity, we performed EMSAs. Treatment with DEX (10 nM) induced the formation of a strong binding complex (Fig. 7B, lane 4). Formation of the binding complex was blocked by treatment with VPA (10 mM) for 6 h (lane 5). These findings suggested that VPA decreased the recruitment of GR and Pol II to the promoters of lipogenesis genes, as well as the GR DNA binding activity, in vitro.

To evaluate the transcriptional activity of Fasn in response to DEX, we performed a luciferase assay. HepG2 cells were transfected with a luciferase vector driven by the Fasn promoter. Treatment with DEX increased the transcriptional activity of Fasn, which was significantly decreased by VPA only in cells transfected with wild-type GR (Fig. 6A). We also used 2 GR mutants, in which 2 lysines in the GR acetylation site were substituted with glutamine (QQ) or arginine (RR) (Vigene Biosciences, CA, USA). We found that the DEX-induced transcriptional activity of Fasn was not decreased by VPA in cells transfected with the RR-mutant (Fig. 8). We investigated GR acetylation by blotting with an anti–ac-K antibody after immunoprecipitation with an anti-GR antibody (Fig. 9). GR acetylation was analyzed by western blot with anti–ac-K antibody after immunoprecipitation with anti-GR antibody (Figs. 9A and C). The relative protein expression was quantified by optical densitometry (Figs. 9B and D). GR acetylation was significantly increased in HepG2 cells and ACTH-infused rats after treatment with VPA, compared to controls (Figs. 9B and D). These findings indicated that VPA decreased the DEX-induced transcriptional activity of GR by acetylation.