Peroxisome Proliferator-Activated Receptor α Activation Induces Hepatic Steatosis, Suggesting an Adverse Effect

Non-alcoholic fatty liver disease (NAFLD) is characterized by hepatic triglyceride accumulation, ranging from steatosis to steatohepatitis and cirrhosis. NAFLD is a risk factor for cardiovascular diseases and is associated with metabolic syndrome. Antihyperlipidemic drugs are recommended as part of the treatment for NAFLD patients. Although fibrates activate peroxisome proliferator-activated receptor α (PPARα), leading to the reduction of serum triglyceride levels, the effects of these drugs on NAFLD remain controversial. Clinical studies have reported that PPARα activation does not improve hepatic steatosis. In the present study, we focused on exploring the effect and mechanism of PPARα activation on hepatic triglyceride accumulation and hepatic steatosis. Male C57BL/6J mice, Pparα-null mice and HepG2 cells were treated with fenofibrate, one of the most commonly used fibrate drugs. Both low and high doses of fenofibrate were administered. Hepatic steatosis was detected through oil red O staining and electron microscopy. Notably, in fenofibrate-treated mice, the serum triglyceride levels were reduced and the hepatic triglyceride content was increased in a dose-dependent manner. Oil red O staining of liver sections demonstrated that fenofibrate-fed mice accumulated abundant neutral lipids. Fenofibrate also increased the intracellular triglyceride content in HepG2 cells. The expression of sterol regulatory element-binding protein 1c (SREBP-1c) and the key genes associated with lipogenesis were increased in fenofibrate-treated mouse livers and HepG2 cells in a dose-dependent manner. However, the effect was strongly impaired in Pparα-null mice treated with fenofibrate. Fenofibrate treatment induced mature SREBP-1c expression via the direct binding of PPARα to the DR1 motif of the SREBP-1c gene. Taken together, these findings indicate the molecular mechanism by which PPARα activation increases liver triglyceride accumulation and suggest an adverse effect of fibrates on the pathogenesis of hepatic steatosis.


Introduction
Non-alcoholic fatty liver disease (NAFLD) represents a spectrum of diseases ranging from hepatic steatosis to steatohepatitis and cirrhosis. The hallmark of NAFLD is excess triglyceride accumulation within hepatocytes [1]. NAFLD is the most common liver disease in Western countries; approximately one third of all Western populations are affected, and the prevalence of these diseases continues to progressively increase [2]. Emerging evidence suggests that NAFLD is the hepatic manifestation of metabolic syndrome and is a risk factor for cardiovascular diseases [3]. Antihyperlipidemic drugs are recommended as part of the treatment for patients with NAFLD [4]. Fibrates are synthetic ligands of peroxisome proliferator-activated receptor a (PPARa), and they serve as first-line drugs for reducing serum triglyceride levels [5]. The lipid-lowering action of fibrates in the blood is mediated through the activation of PPARa and lipoprotein lipase and the suppression of apolipoprotein C-III, among other proteins. [6][7][8][9]. Theoretically, fibrates might be beneficial for the treatment of NAFLD.
However, no definitive conclusion on the efficacy of PPARa agonists in the treatment of NAFLD can be drawn based on the available clinical data [10]. Some studies have suggested that PPARa activation might have protective and therapeutic effects against NAFLD [10], while others have reported contrasting findings. Fenofibrate, one of the most commonly used fibrates, was reported to exert no beneficial effect on liver steatosis, as assessed using MRI [11]. In 16 patients with biopsy-confirmed NAFLD, 48 weeks of therapy with fenofibrate did not reveal any significant change in the grade of steatosis, lobular inflammation, fibrosis, or the NAFLD activity score when determined by liver histology [12]. Another study investigated liver biopsies before and after 12 months of clofibrate treatment and revealed no improvement in the histological grade of steatosis, inflammation, or fibrosis [13]. We conducted preliminary experiments exploring the effect of fenofibrate as a monotherapy on NAFLD in several patients. Notably, MRI did not reveal any significant change in the steatosis score (data not shown). Interestingly, experiments with mice have shown that fenofibrate can increase hepatic triglyceride synthesis [14]. However, the detection of liver steatosis was not discussed in previous studies. Thus, there is a great need to determine the effect of fibrates on hepatic steatosis as well as the mechanism underlying its effects.
Based on the evidence obtained from previous studies, we hypothesized that PPARa activation induces, rather than improves, hepatic steatosis. In the present study, we showed that fenofibrate treatment increased hepatic steatosis and the liver triglyceride content through the up-regulation of mature SREBP-1c (a key lipogenic transcription factor) expression via the direct binding of PPARa to the DR1 motif of the SREBP-1c gene [15]. These findings indicate an adverse effect of fibrates on the pathogenesis of hepatic steatosis. Therefore, the proper use of fibrates should be considered, particularly for the treatment of fatty liver disease.

Ethics statement
The use of animals in this study was in compliance with the relevant federal guidelines and institutional policies, and the animal protocol was approved by the Animal Care and Use Committee of Shandong Provincial Hospital affiliated with Shandong University (approval number: No. 2013-087). All surgical procedures were performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.

Animal experiments
All animals were housed in a temperature-controlled room (22-23uC) under a 12-h light/12-h dark cycle and given free access to food and water.
Eight-week-old C57BL/6J male mice (Vital River Company, Beijing, China) were used. The mice were randomly divided into three groups (6 each), including the control group. Fenofibrate (Sigma-Aldrich, St. Louis, MO) suspended in a 1% carboxymethylcellulose solution of gum Arabic was administered through daily gavage for 10 days at a dose of 0.04 g/kg/day or 0.5 g/kg/day. Animals receiving vehicle alone were used as controls.
The mice were fasted for 6 h and then euthanized using pentobarbital sodium. Serum was collected immediately prior to sacrificing the mice. The livers were immediately harvested and frozen in liquid nitrogen for further experiments. Part of the liver was frozen in optimal cutting temperature compound (OCT) embedding medium in liquid nitrogen.

Human hepatic cell lines and mouse primary hepatocyte isolation and culture
The HepG2 human hepatocellular carcinoma cell line was obtained from the Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China. HepG2 cells were routinely maintained in MEM/EBSS (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (Gibco BRL, Gaithersburg, MD, USA), 100 U/mL penicillin, and 100 mg/mL streptomycin at 37uC in a humidified atmosphere of 5% CO 2 .
Hepatocytes were isolated from C57BL/6J male mice using the two-step collagenase perfusion protocol [17]. Briefly, mice were anesthetized with sodium pentobarbital (30 mg/kg intraperitoneally), and the portal vein was cannulated under aseptic conditions. The liver was perfused with 0.9% saline containing 0.5 mM EDTA and low-glucose DMEM (HyClone) containing 100 CDU/ ml collagenase type IV (Sigma-Aldrich). The isolated mouse hepatocytes were then cultured at 80%-90% confluence in DMEM media (HyClone) containing 10% FBS in rat-tail collagen type I (Sigma-Aldrich) coated plates. The cells were then incubated overnight at 37uC in a humidified atmosphere of 5% CO 2 .
When treated with fenofibrate, the cells were washed twice with PBS and then starved in serum-free medium overnight before treatment. The cells were cultured in serum-free medium during treatment.

Dual luciferase activity assays
Cells were cotransfected with 0.4 mg of luciferase reporter plasmid and 20 ng of Renilla luciferase plasmid pRL-SV40 (Promega, Madison, WI) as an internal control. Ten hours after transfection, the medium was changed, and the cells were allowed to recover for an additional 8 h. The cells were treated with fenofibrate in serum-free medium for 24 h. The cells were then harvested, and luciferase activity was measured using a dualluciferase reporter assay system (Promega). Data represent the amount of firefly luciferase activity, normalized to that of Renilla luciferase activity.

RNA isolation and quantitative RT-PCR
Total RNA from cells and mouse liver tissue was isolated using TRIzol reagent (Takara, Tokyo, Japan) following the manufacturer's instructions. The RT reaction was performed using 1 mg of total RNA. Real-time PCR was performed with a Light Cycler 480 (Roche Applied Science, Indianapolis, IN) [18]. The PCR primers are shown in Table 1 (for human) and Table 2 (for mice). b-actin was employed as an endogenous control for normalization.

Quantification of the triglyceride content
The triglyceride content was measured using a colorimetric assay (Applygen Technologies Inc., Beijing, China) as previously reported [19][20][21]. Briefly, the liver homogenate was prepared after homogenizing the tissue (,50 mg) in 1 ml of standard diluent. The samples were centrifuged at 2000 g for 10 min, and the supernatant was collected. The absorbance at 550 nm is proportional to the concentration of triglycerides of each sample. All samples were determined in duplicate, and the triglyceride values were expressed as mmol of triglycerides/g of protein (for liver) or mmol/L (for serum).

Oil red O staining
The cultured cells were washed twice with PBS and then fixed with 4% paraformaldehyde for 15 min and stained for 15 minutes in a freshly diluted oil red O (Sigma-Aldrich) solution. The cells were counterstained with hematoxylin for 10 sec. To evaluate hepatic lipid accumulation, sections of the liver (10 mm) frozen in OCT embedding medium were stained with oil red O for 10 minutes and then washed and counterstained with hematoxylin for 20 seconds. Representative photomicrographs were captured using a system incorporated into the microscope (Axiovert 100 M Zeiss, Zeppelinstrasse, Germany).

Electron microscopy
Cells were first fixed with 3.5% (v/v) glutaraldehyde in phosphate buffer (pH 7.2) at room temperature overnight and then post-fixed using 1% osmic acid, dehydrated through an ethanol series, and embedded in Spurr's low-viscosity resin.
Transverse ultrathin were prepared and contrasted with saturated uranyl acetate and lead citrate. Microphotographs were taken using a Jeol 1200X electron microscope (Jeol System Co., Akishima, Tokyo, Japan).

Nuclear and cytoplasmic protein extraction and Western blotting
Nuclear and cytoplasmic extracts from cultured hepatocytes and mouse livers were prepared using the NE-PER nuclear and cytoplasmic extraction reagent kit (Pierce Biotechnology, Rockford, IL) according to the manufacturer's instructions. Protein content was determined using a BCA Protein Assay Kit (Shenergy Biocolor Bioscience & Technology Company, Shanghai, China). Protein from nuclear extracts (40-60 mg) or cytoplasmic extracts (60-80 mg) was electrotransferred onto a polyvinylidene fluoride membrane (Millipore, Billerica, MA, USA), and after incubation in 5% BSA for one hour, the blots were probed with the following antibodies at the dilution indicated: SREBP-1 (1:200; Santa Cruz) and PPARa (1:1000; Millipore) at 4uC for the entire night. Mouse anti-LMB1 antibody and anti-GAPDH antibody were obtained from Cwbiotech (Beijing, China) and were used to target endogenous control proteins in the nuclear and cytosolic fractions, respectively. After incubation with the appropriate secondary antibodies conjugated to horseradish peroxidase (HRP) (Amersham, Little Chalfont Bucks, UK) at 1:5,000 for one hour at room temperature, the membranes were visualized using a HyGLO HRP detection kit (Denville, NJ, USA). Quantification of Western blots was performed using ImageJ software (developed at the National Institutes of Health, Bethesda, Maryland).

Immunofluorescence
Cells attached to coverslips were washed with PBS and fixed in 4% paraformaldehyde. The cells were then blocked with 10% normal goat serum for 30 minutes and then incubated with primary antibodies (rabbit anti-SREBP1 or mouse anti-PPARa, both 1:100 dilution) overnight at 4uC, followed by a 1 h incubation at room temperature with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (1:100 dilutions). The nuclei of hepatocytes were stained with DAPI. Specimens were imaged under a confocal fluorescence microscope (Axiovert 100 M Zeiss, Zeppelinstrasse, Germany).

Statistical analysis
The data were analyzed using SPSS 17.0 and are expressed as the mean 6 standard deviation. Differences between two groups were compared using an unpaired Student's t-test. ANOVA was used to compare the means of multiple groups. All of the calculated P values are two-sided. Differences were considered significant at P,0.05.

Results
Liver triglyceride homeostasis was disrupted through pharmacological treatment with fenofibrate The PPARa agonist fenofibrate has been commonly used in humans and animals in previous studies [22]. As shown in Table 3, fenofibrate treatment induced weight loss in the mice. However, glutamic-pyruvic transaminase (ALT), glutamic-oxaloacetic transaminase (AST), and the ratio of liver weight to body weight were increased in mice treated with high doses of fenofibrate compared with the control mice. The liver of fenofibrate-fed mice was pale in color, suggesting increased lipid storage (Fig. 1A). To determine the role of PPARa in triglyceride metabolism, the serum and hepatic triglyceride contents were measured. Consistent with previous studies [23], the serum triglyceride levels were significantly reduced in response to fenofibrate treatment (Fig. 1B), whereas the hepatic triglyceride contents increased in a dosedependent manner after fenofibrate gavage (Fig. 1C). Oil red O staining of liver sections from fenofibrate-fed mice showed accumulated neutral lipids (Fig. 1D), which was consistent with the gross morphological appearance. These results indicated that fenofibrate treatment induced triglyceride accumulation in hepatocytes, which leads to liver steatosis.
The effect of PPARa activation was confirmed in HepG2 cells treated with fenofibrate. Cellular triglyceride contents increased in a concentration-dependent manner in response to fenofibrate treatment ( Fig. 2A). In addition, oil red O staining demonstrated the accumulation of neutral lipids in fenofibrate-treated cells (Fig. 2B). Furthermore, the electron microscopic analysis of HepG2 cells revealed only a few lipid droplets in the control cells. Fenofibrate treatment, however, induced the production of a large number of medium-to-large lipid droplets (Fig. 2C). These data confirmed that fenofibrate increased the intracellular triglyceride content.

PPARa activation induced SREBP-1 gene expression in vivo and in vitro
To elucidate the mechanisms underlying the influence of fenofibrate on the triglyceride content in the liver, we examined the expression of the genes involved in triglyceride metabolism. As shown in Fig. 3A, the expression of Cpt1a, which is directly regulated through PPARa, was increased upon fenofibrate treatment, indicating the activation of PPARa. Subsequently, the expression of SREBP-1c, a key regulatory molecule involved in lipogenesis, was significantly increased in the livers of fenofibratetreated mice, and SREBP-1a expression was not significantly affected. Expression of the key genes associated with lipogenesis including ACC (acetyl-CoA carboxylase), FASN (fatty acid synthase), SCD1 (stearoyl-CoA desaturase 1), and GPAT (glycerol phosphate acyltransferase), was also increased in the fenofibratetreated mouse livers. Interestingly, the transcription level of these genes in response to fenofibrate treatment showed a dosedependent increase in parallel with the level of SREBP-1c expression. The expression of fatty acid-binding protein 1 (FABP1), which regulates the cellular uptake of long-chain fatty acids, was enhanced, and the expression of apoB, which regulates triglyceride exportation from the liver, was reduced in fenofibratetreated mouse livers. These findings are consistent with the results of a previous study [24].
To further evaluate whether the expression of SREBP-1c was induced during the lipogenesis resulting from fenofibrate treatment, we examined liver extracts using Western blotting. Notably, prominent increases in the precursor and mature forms of SREBP-1 proteins were observed in fenofibrate-treated mouse livers (Fig. 3B).
To reconfirm the effect of PPARa activation on the induction of SREBP-1 gene expression, we treated HepG2 cells with fenofibrate. Notably, fenofibrate increased the expression of SREBP-1c protein in a dose-dependent manner in HepG2 cells treated for 48 h (Fig. 4A). Immunofluorescence analysis of mouse primary hepatocytes revealed strong SREBP-1 staining in the nucleus and cytoplasm of these cells (Fig. 4B). Fenofibrate incubation increased SREBP-1 expression in the cytoplasm and promoted the translocation of this gene to the nuclei. In addition, real-time PCR analysis revealed prominent elevations in SREBP-1c and its downstream molecules, such as FASN, ACC, and SCD1, while SREBP-1a showed no change (Fig. 4C). Interestingly, the expression of both the precursor and mature forms of SREBP-1 correspondingly decreased when PPARa was silenced by siRNA in HepG2 cells (Fig. 4D).
To determine whether the induction of SREBP-1 expression observed in fenofibrate-treated mice is dependent on PPARa activation, we used Ppara 2/2 mice. As shown in Fig. 4E, fenofibrate gavaging increased SREBP-1 protein expression in Ppara +/+ mouse livers, whereas this effect was abolished in the PPARa 2/2 mice.

Direct regulation of SREBP-1c by PPARa and SREBP-1c was indispensable in PPARa-induced liver triglyceride accumulation
Studies have reported that SREBP-1c expression is reduced in Ppara 2/2 mice compared with wild-type mice [14,25,26]. Indeed, PPARa agonists enhance the activity of the Srebp-1c promoter through direct binding with the DR1 motif [15]. Using the fulllength SREBP-1c promoter (1564/+1)-driven luciferase construct, we observed that luciferase activity was significantly increased by fenofibrate treatment in a dose-dependent manner, indicating that SREBP-1c expression is directly regulated through PPARa (Fig. 5A).
To determine the indispensable role of SREBP-1c in PPARainduced hepatic triglyceride accumulation, we used a plasmid encoding DN-SREBP-1c. DN-SREBP-1c contains a tyrosine 320 to arginine mutation on the truncated nuclear form of rat SREBP-1c (1-403R), which disrupts the binding of SREBP-1c to the SRE motif [27]. Interestingly, DN-SREBP-1c completely inhibited the fenofibrate-mediated increase in the hepatic triglyceride content   (Fig. 5B). These results suggest that SREBP-1c is necessary for PPARa-induced liver lipid accumulation.

Discussion
Using a series of in vivo and in vitro experiments, we confirmed that PPARa activation through fenofibrate increased liver triglyceride synthesis, leading to hepatic steatosis. The effect of fenofibrate was observed at both low and high doses. Fenofibrate treatment induced mature SREBP-1c expression via the direct binding of PPARa to the DR1 motif of the SREBP-1c gene, which up-regulates the expression of the key genes associated with lipogenesis. These findings suggest a molecular mechanism that underlies specific clinical findings, showing that fibrates cannot improve hepatic steatosis in patients with NAFLD [11][12][13]. Based on these results and previous clinical findings, the efficacy of fibrates, particularly in the treatment of fatty liver disease, should be re-evaluated, indicating a need for large prospective studies and a full assessment of liver histology.
Fenofibrate is available for oral administration at a daily dose of 200-300 mg (4-5 mg/kg/day for 60 kg) in adult patients in the clinic [28], and a previous study reported that the blood concentration reached 30 mM after fenofibrate treatment at 200 mg daily for 7 days [29]. Based on these data, we adopted 0.04 g/kg daily as a low in vivo (therapeutic dose) dosage and 0.5 g/kg daily as a high in vivo (overdose) dosage for treating mice; we also used 50 and 100 mM concentrations in vitro to stimulate hepatocytes. The results showed that fenofibrate increased the expression of the genes involved in triglyceride synthesis and fatty acid uptake, transport, synthesis, and b-oxidation, increasing the triglyceride content in the liver, which is consistent with previous studies [23]. The induction of weight loss by a high dose of fenofibrate was observed in the present and previous studies [14]. Elevated plasma ALT and AST levels were also observed. However, it appears unlikely that the induction of liver steatosis by fenofibrate was the result of liver damage. Indeed, treatment with the low dose of fenofibrate, in which ALT and AST remained normal, also induced liver triglyceride accumulation, indicating a direct role of fenofibrate in liver steatosis. In addition, Nakajima T et al also showed remarkable differences in bezafibrate action on PPARa activation and reactive oxygen species generation between conventional experimental high doses and clinically relevant low doses in wild-type mice [30]. Thus, despite the use of a different molecule, these findings support the differences observed in the present study.
Some clinical studies have assessed the effects of fenofibrate on biochemical and imaging surrogates of NAFLD [11][12][13]. Indeed, recent preclinical studies have strongly suggested that PPARa activation increases liver lipid synthesis. Treatment with a PPARa agonist promotes 3 H 2 O incorporation into hepatic lipids in wildtype mice but not in Ppara 2/2 mice [31]. Additionally, fenofibrate-treated mice show strong acetyl-CoA incorporation into hepatic fatty acids [14]. The normal circadian rhythms of hepatic lipogenic FASN and ACC expression are disturbed in Ppara 2/2 mice [25]. Moreover, studies have reported that SREBP-1c mRNA levels are decreased in Ppara 2/2 mice compared with wild-type mice, suggesting the PPARa-dependent induction of hepatic fatty acid synthesis and SREBP-1c activation [14,25,26]. These findings are consistent with the results of the present study, which showed that PPARa activation induced hepatic triglyceride accumulation through the up-regulation of mature SREBP-1c expression. Notably, compared with previous studies [32], we administered both a therapeutic dose and an overdose of fenofibrate. Moreover, we focused on the effect of fenofibrate on hepatic steatosis, while previous studies did not present similar results. Morphological observations and oil red O staining were used to examine liver steatosis in mice. The effects of fenofibrate on liver lipid accumulation were reconfirmed using electron microscopy.
These findings suggest a direct regulatory effect of PPARa on SREBP-1c. A PPARa response element (DR1) in the promoter of the human SREBP-1 gene has been identified and is involved in PPARa protein binding [33,34]. Using the dual-luciferase reporter assay system, we demonstrated that fenofibrate treatment enhanced the activity of the human SREBP-1c promoter in a dose-dependent manner. Furthermore, we found that SREBP-1c expression was reduced after the HepG2 cells were treated with PPARa siRNA. Therefore, it is reasonable to conclude that the increased levels of SREBP-1c mRNA and mature protein following PPARa activation were induced by fenofibrate treatment. Although a DR1 motif has not been found in the mouse SREBP-1 promoter [31], the induction of SREBP-1 mRNA expression observed in fenofibrate-treated mice could be due to different molecular mechanisms, which require further study: 1. A PPARa binding site other than DR1 may exist on the mouse SREBP-1c promoter. 2. PPARa exerts an indirect regulatory effect on SREBP-1c in mice. In the present study, the requirement of PPARa for the induction of SREBP-1 was tested in a Ppara 2/2 mouse model. The up-regulation of SREBP-1 expression was observed in fenofibrate-treated Ppara +/+ mice, and this effect was strongly impaired in Ppara 2/2 mice. The results indicate that the induction of SREBP-1 expression observed in fenofibrate-treated mice is dependent on PPARa activation, similar to the changes observed in other studies [25,26,35,36]. Fibrates also stimulate the b-oxidation of fatty acids, leading to fatty acid depletion, which increases SREBP-1c expression [15]. Notably, PPARa-induced SREBP-1c expression might not occur secondary to fatty acid depletion because treatment with etomoxir, an inhibitor of fatty acid oxidation, does not abolish the effect of WY 14,643 (a PPARa agonist) on the incorporation of 3 H 2 O into fatty acids [25,26]. Interestingly, a DR1 element has been found in the promoter region of other lipogenic genes regulated by SREBP-1 [37], and they are under the direct control of PPARa. This is helpful for explaining the development of steatosis observed in fenofibrate-treated mice. The molecular mechanism by which PPARa regulates the mouse SREBP-1c expression remains to be elucidated.
However, some studies have suggested that hepatic triglyceride accumulation might be a protective mechanism through which the toxic effects of free fatty acids are prevented [38] [39]. Moreover, previous studies have demonstrated that PPARa activation might be protective and therapeutic against NAFLD [40][41][42]. This benefit has been associated with improved fatty acid turnover and the anti-inflammatory and anti-oxidant properties of PPARa [43][44][45]. In these studies, the data obtained suggested a role for fenofibrate under conditions of high-fat diet, obesity, insulin resistance, and type 2 diabetes mellitus. In the present study, we administered fenofibrate to normal adult mice, which presented normal serum lipid levels before treatment. The discrepancy between these results and those of previous studies likely reflects the different animal models employed. PPARa activation exerted a synergistic effect on lipid metabolism, which involved accelerated lipid mobilization in white adipose tissue, liver free fatty acid uptake, DNL, fatty acid b-oxidation, and exportation. The disease models might perturb this balance, contributing to a different effect of fenofibrate on the hepatic triglyceride content. However, this controversy should be further assessed.
In conclusion, the results of the present study showed that PPARa activation through fenofibrate treatment increased liver triglyceride synthesis, leading to hepatic steatosis. The underlying mechanism involves the induction of mature SREBP-1c expression via the direct regulation of SREBP-1c through PPARa, which further up-regulates the expression of genes associated with lipogenesis. These findings are consistent with the results of previous clinical studies showing that fibrates do not improve hepatic steatosis in patients with NAFLD. Thus, there is a need for large prospective studies and a full assessment of liver histology to reevaluate the efficacy of fibrates, particularly for the treatment of fatty liver disease.  PPARa-dependent regulation of the SREBP-1c promoter was analyzed through luciferase assays using the human SREBP-1c-1564/+1luc construct in the presence of the indicated amounts of fenofibrate for 24 h. The activities of the constructs transfected into HepG2 cells under the indicated dosages are shown. The results represent relative Firefly/Renilla luciferase activities compared with the wild-type construct under basal conditions as the reference value. (B) HepG2 cells were transfected with the empty plasmid pcDNA3.1 or DN-SREBP-1c and subsequently treated with or without 50 mM fenofibrate in serumfree medium for an additional 48 hours. The cells were collected and assayed to determine the intracellular TG content. The content was normalized to the total protein from the same group. The data were calculated from 3 independent experiments and are expressed as the mean 6 SD. * P,0.05 versus the control group. # P,0.05 versus the fenofibrate 10 mM group. doi:10.1371/journal.pone.0099245.g005