Cluster of differentiation (CD)36, sterol regulatory element-binding protein 1 (SREBP1) and PPARα antibodies were separately purchased from Santa Cruz Biotechnology, Inc., (Dallas, TX, USA). β-actin rabbit monoclonal antibody was purchased from Abcam (Cambridge, MA, USA). Secondary horseradish peroxidase-labeled goat anti-rabbit immunoglobulin G (H+L) was purchased from Novoprotein Scientific, Inc. (Summit, NJ, USA). GEM was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Oleic acid (OA) and bovine serum albumin (BSA; fatty acid free) were obtained from Sangon Biotech Co., Ltd., (Shanghai, China).

Human hepatoma SMMC-7721 cells were supplied by the Institute of Cell Biology (Shanghai, China). The cells were cultured in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (Biological Industry, Kibbutz Beit Haemek, Israel), 1% penicillin-streptomycin (10,000 U/ml penicillin and 10 mg/ml streptomycin; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) and maintained at 37°C with humidified air in 5% CO₂. SMMC-7721 cells were first exposed to GEM when they reached 75% confluence. Different dilutions (10–200 µM) of GEM were made with DMSO and 10 µl of GEM solutions were then added to 1 ml culture medium, respectively. To investigate the effect of OA on fat-overloading, cultures were exposed to different concentrations of OA, diluted in 10% BSA, ranging from 0.5–2 mM.

To determine the effect of GEM or OA on SMMC-7721 cell viability, the cells were treated with GEM at different concentrations (0, 50, 100 and 200 µM) for 24 and 48 h at 37°C. Following this, 1×10⁶ cells in 96-well plates were treated with OA at different concentrations (0, 0.5, 1, 1.5 and 2 mM) with 10% BSA overnight at 37°C, respectively. Cell viability was determined using Cell Counting kit-8 dye (Beyotime Institute of Biotechnology, Beijing, China), according to manufacturer's instructions. Absorbance was measured at 450 nm with a GENios multifunction-reader (Tecan GENios Pro; Tecan Group, Ltd., Männedorf, Switzerland).

Oil red O (Sangon Biotech Co. Ltd., Shanghai, China) was used to monitor the content of lipids in SMMC-7721 cells, according to the manufacturer's instructions. Cells were seeded in a 6-well plate at a density of 1.0×10⁵ cells/well. Following adherence, cells were treated with GEM at different concentrations (0, 10, 25, 50 and 100 µM) together with 1 mM OA for 24 h at 37°C, respectively. Subsequently, cells were fixed overnight at 37°C with 4% paraformaldehyde and stained with Oil red O at 37°C for 30 min. Images were photographed with an inverted fluorescent microscope (Nikon Eclipse TI; Nikon Corp., Tokyo, Japan). Following this, Oil red O was extracted using isopropanol (100 µl) for 1 h at 37°C. Then the extracted sample was moved to another 96-well plate. Absorbance was measured at 510 nm in a spectrophotometer for quantitative analysis (16).

For quantitative estimation of TG, lipids were extracted from cells using Triton X-100 (2%) for 30 min at 37°C. An enzymatic assay was then performed using an EnzyChromTM Triglyceride Assay kit (Bioassay Systems LLC, Hayward, CA, USA), according to the manufacturer's protocols. Total lipid extraction and separation was conducted by thin layer chromatography (TLC), according to previous methods (16). The cell pellets were harvested by centrifugation at 1,000 × g for 5 min at 37°C, washed twice with phosphate-buffered saline (PBS), snap-frozen and smashed in 1 ml methanol chloroform mix (v/v, 2/1). These components were mixed well and then centrifuged at 12,000 × g for 5 min at room temperature to allow phase separation. The chloroform phase was transferred to a new tube and blow-dried. The lipid fractions were separated by TLC using a developing solvent (hexane/diethyl ether/acetic acid; 40:80:2, v/v/v). To visualize different fractions of total lipids extracted from cells, the TLC plate (Sinopharm Chemical Reagent Co., Ltd., Beijing, China) was stained with iodine vapor at 60°C for 30 min and photographed using a DNR Bio-Imaging System, Ltd. (Neve Yamin, Israel).

Total RNA was extracted using RNAiso Plus (Takara Biotechnology Co. Ltd., Dalian, China), according to the manufacturer's instructions, and quantified using a NanoDrop 2000c (Thermo Fisher Scientific Inc., Waltham, MA, USA). First-strand cDNA synthesis (1 µg) and PCR reactions were performed using the PrimeScriptTM RT reagent kit with gDNA Eraser (Takara Biotechnology Co. Ltd., Dalian, China), according to the manufacturer's instructions. Following this, mRNA levels were determined by PCR (EmeraldAmp PCR MasterMix; Catalogue no. RR300A; Takara Biotechnology Co. Ltd., Dalian, China), as described in the study by Bergman et al (17). The primers were designed using Primer 5.0 software (Premier Biosoft International, Palo Alto, CA, USA) and are listed in Table I. 18S ribosomal (r)RNA was selected as an internal control. The PCR reaction conditions were as follows: 98°C for 10 sec, 55°C for 30 sec and 72°C for 1 min for a total 30 cycles. qPCR was performed in triplicate assays using SYBR Premix Ex Taq (Tli RNaseH Plus; catalogue no. RR420; Takara Biotechnology Co. Ltd., Dalian, China) in a CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The qPCR reaction conditions were as follows: Activation of the Taq DNA polymerase at 95°C for 30 sec, followed by 40 cycles of 95°C for 10 sec and 60°C for 32 sec. mRNA expression levels were analyzed by the 2^−ΔΔCq method (18), relative to 18S rRNA expression.

Cells were incubated with either OA (1 mM) or OA together with GEM (50 µM) for 24 h at 37°C and lysed with pre-chilled radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology) for 30 min on ice. Following this, lysates were centrifuged at 13,000 × g for 20 min at 4°C and quantified by a Bradford protein assay (Bio-Rad Laboratories, Inc.) (19). Proteins (50 µg) were separated by 8% SDS-PAGE and transferred to Amersham Hybond-P polyvinylidene fluoride membranes (GE Healthcare Life Sciences, Little Chalfont, UK). Membranes were blocked with 5% milk powder in PBS for 2 h at 37°C and then incubated with primary specific antibodies overnight at 4°C, including CD36 (catalogue no. sc-9154; 1:2,000), PPARα (catalogue no. sc-9000; 1:1,000), SREBP1 (catalogue no. sc-8984; 1:1,000) and β-actin (catalogue no. ab8226; 1:5,000) rabbit monoclonal antibodies. The membrane was placed in PBS-Tween 20 (PBST) and cleaned 3 times for 10 min. Then samples were incubated with secondary horseradish peroxidase (HRP)-labeled goat anti-rabbit immunoglobulin G (H+L) (catalogue no. L153B; 1:1,000) at room temperature for 1.5 h. After completion of secondary antibody incubation, wash 3 times in PBST for 10 min/time Target proteins on the membranes were visualized using Pro-light HRP Chemiluminescent kit (cat. no. PA112; Tiangen Biotech Co., Ltd., Beijing, China) following the manufacturers protocol. The protein bands were analyzed by densitometry using ImageJ software, version 1.37 (National Institutes of Health, Bethesda, MD, USA) after exposure of the membranes to gel capture software, version 2.0 (DNR Bio-Imaging System, Ltd., Jerusalem, Israel).

All experiments were performed at least three times. Values were expressed as the mean ± standard deviation. Statistical analysis was performed using the Student's t-test using SPSS version 19.0 (IBM Corp., Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

Using the WST-8 based Colorimetric Assay Cell Counting kit-8, the effect of GEM on cell viability was measured. As demonstrated in Fig. 1A, no significant cytotoxic effect was observed in cells following treatment with GEM. Cell viability of SMMC-7721 cells was inhibited by 4.22, 9.72 and 9.55% at GEM concentrations of 50, 100 and 200 µM after 24 h, respectively. Cells were also exposed to different concentration of OA. The concentrations of OA were chosen according to our previous work (15). As demonstrated in Fig. 1B, OA inhibited cell viability in a dose- and time-dependent manner. When cells were cultured in the presence of OA at 0.5 and 1 mM for 24 h, they accumulated intracellular lipids without acute cytotoxic effect, while cell viability was significantly inhibited when treated with 1.5 and 2 mM OA for 24 and 48 h compared with 0 mM OA (P<0.05). Cell viability was decreased by 25.7 and 46.9% at 1.5 mM OA for 24 and 48 h, respectively and 44 and 62.9% at 2 mM OA for 24 and 48 h, respectively. Therefore, cells treated with 1 mM OA were used as the cellular model of NAFLD.

To determine whether GEM affects lipid accumulation in SMMC-7721 cells, cells were incubated with different concentrations of GEM and 1 mM OA. As demonstrated in Fig. 2A, a clear dose-dependent decrease in lipid accumulation was observed in the cells under the microscope. Furthermore, the total lipid levels were decreased by 12.9, 21 and 16.5% in cells incubated with GEM concentrations of 25, 50 and 100 µM, respectively (Fig. 2B). The results indicated that 50 µM of GEM significantly reduced levels of intracellular total lipids and TG compared with cells treated with 1 mM OA only (P<0.05; Fig. 2B and C). Therefore 50 µM was used for subsequent experiments and assessments. The level of cellular TG was detected and was observed to be decreased by 43.8% at 50 µM GEM compared with cells treated with 1 mM OA only (Fig. 2C). To further confirm the lipid changes in the cellular model, TLC analysis was performed. TLC results suggested that there was a decrease in TG levels in cells treated with 50 µM GEM compared with those treated with OA only (Fig. 2D). Taken together, GEM may lower the lipid accumulation in an in vitro model of NAFLD. The optimal concentration is 50 µM.

RT-PCR and RT-qPCR analyses were employed to determine whether GEM was able to regulate lipid metabolism-related gene expression in SMMC-7721 cells. The results demonstrated that GEM (50 µM) significantly increased the levels of CD36, SREBP1 (P<0.05) and significantly reduced the levels of PPARα (P<0.01) in cells treated with 1 mM OA compared with cells treated with OA only (Fig. 3A and B). Subsequently, western blotting with specific antibodies was performed to detect the changes of related protein expression levels. The expression levels of SREBP1 and PPARα were markedly upregulated following treatment with GEM in the OA-overloaded cells. However, the level of CD36 in OA-overloaded cells treated with GEM showed no marked change compared with cells treated with OA only (Fig. 3C).

It is well recognized that lipogenic genes are commonly trans-activated by SREBP1, and this has critical central roles in the regulation of lipid synthesis (20). To validate the upregulation of SREBP1, the expression levels of its downstream target genes, such as LIPIN1, LIPIN2, diacylglycerol O-acyltransferase (DGAT)1 and DGAT2, were examined in the cells. The results demonstrated that mRNA levels of LIPIN1 and DGAT1 were markedly increased following treatment with GEM in cells treated with OA. However, mRNA expression levels of LIPIN2 and DGAT2 remained unchanged (Fig. 4A). As a transcription factor, PPARα has central roles in hepatic lipid oxidation, predominantly through regulating lipid target genes, such as carnitine palmitoyltransferase (CPT)1, CPT2, acyl-coA oxidase 1 (ACOX1) and hydroxyacyl-CoA dehydrogenase (HADHA) (21,22). Therefore, mRNA expression level changes of these genes were measured. The data demonstrated that mRNA levels of CPT2, ACOX1 and HADHA were significantly increased in the cellular model of NAFLD treated with GEM compared with the OA control, while the CPT1 mRNA expression level was not altered (Fig. 4B). Therefore, GEM may lower TG accumulation in OA-overloaded SMMC-7721 cells via the involvement of the PPARα and SREBP1 signaling pathways (Fig. 5).