Activation of carbohydrate response element binding protein (ChREBP) by ethanol

Abstract

Carbohydrate response element binding protein (ChREBP) is a transcription factor involved in hepatic lipogenesis. Its function is in part under the control of AMP-activated protein kinase (AMPK) and protein phosphatase 2A (PP2A). Given known effects of ethanol on AMPK and PP2A, it is plausible that ethanol might enhance fatty acid synthesis by increasing the activity of ChREBP. We hypothesized that another potential pathway of ethanol-induced hepatic steatosis is mediated by activation of ChREBP.

Methods

The effects of ethanol on ChREBP were assessed in hepatoma cells and in C57BL/6J mice fed with the Lieber-DeCarli diet.

Results

When the cells were exposed to ethanol (50 mM) for 24 hrs, the activity of a liver pyruvate kinase (LPK) promoter-luciferase reporter was increased by ~4-fold. Ethanol feeding of mice resulted in the translocation of ChREBP from cytosol to the nucleus. PP2A activity was increased in the liver of ethanol-fed mice by 22%. We found no difference in the levels of hepatic Xu-5-P between ethanol-fed mice and controls. Transfection of a constitutively active AMPK expression plasmid suppressed the basal activity of the LPK luciferase reporter and abolished the effect of ethanol on the reporter activity. However, transfection of rat hepatoma cells with a dominant negative AMPK expression plasmid induced basal LPK luciferase activity by only ~20%. The effect of ethanol on ChREBP was attenuated in the presence of okadaic acid, an inhibitor of PP2A.

Conclusions

The effects of ethanol on AMPK and PP2A may result in activation of ChREBP, providing another potential mechanism for ethanol-induced hepatic steatosis. However, additional okadaic acid-insensitive effects appear to be important as well.

INTRODUCTION

Hepatic steatosis is the most common and earliest response of the liver to heavy alcohol consumption ¹. Although previously considered to be a benign consequence of alcohol use, accumulating evidence indicates that fatty livers are unusually susceptible to the toxic effects of endotoxin that is involved in alcoholic hepatitis and fibrosis ². Clinically, a subset of individuals with alcoholic fatty liver goes on to develop fibrosis, cirrhosis, and liver failure ³. Therefore, understanding the pathogenesis of alcoholic fatty liver may improve prevention and provide promising prospects for more effective treatments.

Several new concepts have evolved during the past few years on how alcohol regulates enzymes and transcription factors leading to hepatic steatosis. Alcohol exerts effects on transcription factors which control fatty acid oxidation, peroxisome proliferator-activated receptor α, and fatty acid synthesis through sterol regulatory element binding protein-1c (SREBP-1c) ^4–6. Alcohol also inhibits AMP-activated protein kinase (AMPK), leading to increasing activity of acetyl-CoA carboxylase (ACC), the rate-limiting enzyme in fatty acid biosynthesis in liver and other tissues ⁵. Recently, we found that the inhibitory effect of ethanol on AMPK phosphorylation is mediated in part through increased activity of protein phosphatase 2A (PP2A)^7;8.

Besides SREBP, studies of the regulation of genes involved in lipogenesis in isolated hepatocytes identified another DNA-binding site promoting the transcription of these lipogenic enzymes, specifically after stimulation by high glucose, which was designated the carbohydrate response element, ChoRE ⁹. Uyeda and his colleagues have purified the 100 KDa protein, carbohydrate response element binding protein (ChREBP), which binds to the ChoRE of the promoter region of the liver pyruvate kinase (LPK) gene ^9;10. Recently, ChREBP was found to directly promote transcription of the genes for lipogenic enzymes, such as ACC and FAS ¹¹. The increase in expression of multiple hepatic lipogenic enzyme mRNAs elicited by feeding a high carbohydrate diet, as well as that of LPK mRNA, is markedly reduced in mice lacking ChREBP gene expression (ChREBP −/−) in comparison to wild type mice ^12;13.

ChREBP contains multiple phosphorylation sites that regulate its activity and cellular localization. Phosphorylation of ChREBP by AMPK activates export of the protein from the nucleus and decreases its DNA-binding activity ¹⁴. 5-amino-4-imidazolecarboxamide ribotide (AICAR), a specific activator of AMPK, inhibited LPK transcription in ChREBP-overexpressing hepatocytes by simulation of phosphorylation of Ser⁵⁶⁸ in ChREBP ¹⁴. ChREBP is also under the control of xylulose-5-phosphate (Xu-5-P). Xu-5-P activates a unique form of protein phosphatase PP2A, which in turn activates ChREBP by dephosphorylating multiple Ser/Thr sites and promotes its cytosol-to-nucleus translocation and DNA binding ¹⁵. Hence, the activity of ChREBP can be modulated by the relative activities of AMPK and PP2A; it is also subject to phosphorylation at a number of other residues by cAMP-dependent protein kinase (PKA) at residues Ser¹⁹⁶, Ser⁶²⁶, and Thr⁶⁶⁶, which also inhibit transcriptional activation ¹⁶.

The effect of ethanol on ChREBP activity and cellular location has not been studied. Given the effect of ethanol on AMPK and PP2A ^5;7;8, it is plausible that ethanol might enhance fatty acid synthesis by increasing the activity of ChREBP, in addition to its previously reported effect on SREBP-1c ⁶. In this study, we explored the effect of ethanol on ChREBP, hypothesizing that this factor represents another potential mechanism for ethanol-induced hepatic steatosis.

MATERIALS AND METHODS

Materials

Most chemicals were purchased from Sigma Chemical Co. (St. Louis, MO). Trypsin and tissue culture media were from Invitrogen. Delipidated fetal bovine serum was purchased from Sigma. Rat hepatoma cell line (H4IIEC3) was purchased from the American Type Culture Collection (ATCC, Manassas, VA). Most antibodies were from Cell Signaling Technology (Beverly, MA) unless stated otherwise. Actin and ChREBP antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and AbCam (Cambridge, MA), respectively. AMPKα1³¹², a constitutively active form of AMPK, and AMPKα1DN, a dominant-negative mutant of AMPK which were used in our previously published study ⁵, were kind gifts of Dr. David Carling (Hammersmith Hospital, London, United Kingdom) and Dr. Jin-Zhong Zhang (Case Western Reserve University, Cleveland, OH), respectively.

Transfection of tissue culture cells

Transfection in H4IIEC3 cells were performed with lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. H4IIEC3 cells were grown in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, antibiotics (63 μg/ml of penicillin G and 100 μg/ml streptomycin), 100 nM dexamethasone, and 5 μg/ml of insulin. The cells were plated in 24-well plates at 50%–70% confluence 24 h prior to transfection. The cells were then washed with phosphate buffer saline (PBS) and switched to DMEM without glucose, antibiotics, 100 nM dexamethasone, and 5 μg/ml of insulin. For each transfection reaction in each well, 0.5 μg of the reporter plasmid (liver pyruvate kinase), 0.5 μg of a ChREBP expression plasmid (a kind gift from Dr. K. Uyeda, University of Texas Southwestern Medical Center, TX), and 0.1 μg of Renilla luciferase reporter were used. Twenty four hours after transfection, the cells were washed with PBS and fresh medium including charcoal dextran were added. They were then treated with ethanol as indicated in the text.

Animals and diets

The experimental protocols were approved by the Indiana University School of Medicine Animal Care and Use Committee. Animal feeding protocol was described in our previous studies. Liquid diets provide 1 kcal/mL (prepared by Dyets, Inc.; Bethlehem, Philadelphia, PA) and were based on the Lieber-DeCarli formulation. Protein content was constant at 18% of calories, and each diet had identical mineral and vitamin content. In this study, six- to 8-week-old male C57BL/6J mice (Jackson Laboratory, Bar Harbor, ME) were divided into 2 dietary groups: (a) control diet (fat comprising 10% of total calories, 6% from cocoa butter and 4% from safflower oil, 72% of calories as carbohydrate); and (b) ethanol-containing diet (identical to the control diet but with ethanol added to account for 27.5% of total calories and the caloric equivalent of carbohydrate [maltose-dextrin] removed). The animals were pair fed for 4 weeks and killed. At the time of sacrifice, liver tissues were harvested, as rapidly as possible, immediately freeze-clamped with Wollenberger tongs at the temperature of liquid nitrogen, powdered under liquid nitrogen with a mortar and pestle, and stored at −80°C for analysis. For hepatic triglyceride measurements, twenty mg of liver tissue powder prepared under liquid nitrogen was used. Lipids were extracted using isopropanol. Hepatic triglyceride content was measured using Wako L-type TG H assay (Wako Diagnostics, Richmond, VA).

Histology of Liver

A part of the sliced liver tissues was fixed in 10% formalin solution for routine hematoxylin and eosin staining. Frozen sections of the liver were stained with Oil Red O.

Immunoblot analysis

Nuclear and cytosolic protein extracts from mouse livers were prepared using the Active Motif Nuclear extract kit (Catalog number 40010, Carlsbad, CA). For other experiments, sixty mg of whole liver tissue powder prepared under liquid nitrogen was homogenized with radioimmunoprecipitation assay (RIPA) buffer containing 50 mM Tris/HCl, pH 7.4, 150 mM NaCl, 0.25 % (w/v) deoxycholic acid, 1 % (v/v) Nonidet P-40, 1 mM EDTA, 10 μM tosyl phenylalanyl chloromethyl ketone, 10 μg/ml trypsin inhibitor, 1 mM phenylmethylsulfonyl fluoride, 2 μg/ml aprotinin, 3.5 mM bis-benzamidine, 50 mM potassium fluoride, and 0.4 mM sodium orthovanadate. Tissue extracts were obtained by centrifugation for 10 min at 4 °C and 14000 × g. Protein concentrations were determined by the Bio-Rad assay. Measurement of ChREBP nuclear and cytosol protein levels were performed using 100 μg of liver nuclear protein, separated by electrophoresis in an 8% SDS-polyacrylamide gel. For other experiments, equal amounts of protein (normally 20 μg of protein, unless otherwise stated) were separated on SDS-polyacrylamide gels, transferred to a nitrocellulose membrane by the wet blotting method, and probed with antibodies as indicated. The amounts of bound antibodies were accessed by the peroxidase activity of horseradish peroxidase-conjugated secondary antibody as detected by chemiluminescence with Lumi-light western blotting substrate (Amersham Biosciences, Piscataway, NJ). The intensity of the individual bands on Western blots were measured by PhosphoImager and analyzed with ImageQuant (Amersham Biosciences) software analysis.

Total RNA isolation and qRT-PCR

Total RNA was prepared from liver tissue using an Absolutely RNA RT-PCR Miniprep kit (Stratagene, Cedar Creek, TX). Reverse transcription of1 μg total RNA to cDNA was performed using the StrataScript qPCR cDNA synthesis kit (Stratagene). Real-time quantitative polymerase chain reaction (qRT-PCR) amplification was performed in a Stratagene MX 3005P thermal cycler(La Jolla, CA) using RT² SYBR Green qPCR Master Mix. Primers for SYBR Green-based real-time PCR were purchased from SA Bioscience (Frederick, MD). The following primers were used: Pklr (PPM05120, liver pyruvate kinase), Acaca (PPM05109, acetyl-Coenzyme A carboxylase), Srebf1 (PPM05094, sterol regulatory element binding transcription factor 1), Fasn (PPM03816, fatty acid synthase), ChREBP (MLX interacting protein-like, PPM 28173B)and GAPDH (PPH00150, glyceraldehydes-3-phosphate dehydrogenase). The relative amount of target mRNA was calculated using the comparative cycle threshold (C_t) method and normalizing each target gene with C_t of housekeeping gene, GAPDH.

Measurement of hepatic protein phosphatase 2A activity

The activity of PP2A was measured with the PP2A immunoprecipitation phosphatase assay kit (Millipore, CA). Threonine phosphopeptide (K-R-Pt-I-R-R) was used as the PP2A substrate. In brief, the cells were harvested in lysis buffer (0.5 M Tris-HCl, pH 7.4, 1.5 M NaCl, 2.5% deoxycholic acid, 10% NP-40, 10 mM EDTA), 1 mM PMSF, and protease inhibitors. Supernatants were incubated with anti-PP2A [C subunit, clone 1D6] and protein A agarose at 4°C for 2 hours with constant rocking. The immunoprecipitates were then washed 3 times with Tris-buffered saline and diluted phosphopeptide (final concentration 750 μM) as well as Ser/Thr assay buffer were added. The mixtures were incubated for 10 minutes at 30°C in a shaking incubator; then briefly centrifuged, and 25 μl of the samples were transferred to 96-well microtiter plate. PP2A activities were determined by the addition of the Malachite Green Phosphate Detection Solution into the mixtures and measuring the absorbance at 650 nm. The absorbance values of each sample were compared to negative controls containing no PP2A enzyme activity.

Streptavidin-Biotin Complex DNA Binding Assay

Liver pyruvate kinase carbohydrate response element (LPK ChoRE) sense (5′ – GGG CGC ACG GGG CAC TCC CGT GGT TCC- 3′), antisense (5′ – GGA ACC ACG GGA GTG CCC CGT GCG CCC – 3′) and biotin LPK ChoRE sense oligonucleotides were from Invitrogen (Carlsbad, CA). Fifty μg of nuclear protein and annealed LPK ChoRE oligonucleotides were incubated overnight at 4°C in DNA pull down buffer (25 mM HEPES, 15 mM NaCl, 0.5 mM DTT, 0.5% NP-40, 0.1 mM EDTA at pH 7.5, and 10% glycerol). They were then precipitated with streptavidin-agarose (Upstate, Catalog number 16-126) for 2 hrs at 4 °C, centrifuged and washed with DNA pulldown buffer. The bound materials were eluted and separated with boiling and brief centrifugation and then resolved by SDS-PAGE for Western blot analysis.

Xylulose-5-Phosphate measurement

Xylulose-5-phosphate measurement was performed using LC/MS-MS at the Proteomics/Metabolomics Core Facilities, Purdue University, West Lafayette, Indiana. One hundred and fifty mg of liver tissue powder prepared under liquid nitrogen was homogenized in 4 ml/gram of cold methanol ¹⁷. Samples were kept on ice and then centrifuged for 5 min at 14,000g at 4 °C to remove precipitated protein and tissue debris. Supernatant was then transferred into Eppendorf tubes and keep it in −80C until analyses. The HPLC–ESI-MS system consisted of a capillary HPLC system (1100 series LC, Agilent) and an electrospray ionization (ESI) source of time-of-flight (TOF) mass spectrometer (MSD TOF, Agilent). The system was controlled by ChemStation software (Agilent). The autosampler was set at 10 °C. Separations were performed on a Zorbax C8 column (2.1 mm × 150 mm, Agilent). The elution started from 95% mobile phase A (5 mM TBA aqueous solution, adjusted to pH 5.0 with acetic acid) and 5% mobile phase B (100% ACN), raised to 70% B in 25 min, further raised to 100% B in 2 min, and then held at 100% B for 3 min. The flow rate was set at 0.3 mL/min with injection volume as 20 μL. The column was preconditioned by pumping the starting mobile phase mixture for 10 min. LC–ESI-MS chromatograms were acquired in negative ion mode under the following conditions: capillary voltage of 4000 V and fragmentor of 165 V, dry temperature at 300 °C, dry gas flow maintained at 8.0 L/min, and an acquisition range of m/z 150–1000 with LC/MS-MS.

Data analysis

All data are presented as the mean ± SE. Statistical significance was calculated with the Student t test or ANOVA analysis, followed by post hoc testing with least squares difference (LSD), when appropriate. P < .05 was considered statistically significant.

RESULTS

Effects of Ethanol on Transcription of ChoRE-containing Promoters in Vitro

We first studied the effect of ethanol on ChoRE-containing promoters in H4IIEC3 cells. The cells were transfected with the reporter (driven by the LPK promoter) and the internal control plasmid (Renilla luciferase reporter) and exposed to various concentrations of glucose and ethanol as indicated. As shown in Figure 1, the reporter activity was doubled by incubation the H4IIEC3 cells with glucose (25 mM). When the cells were exposed to ethanol (50 mM) for 24 hrs, the activity of the ChREBP reporter was increased by ~4-fold compared to controls. Our results suggested that ethanol increased transcription of ChoRE-containing promoters in vitro.

Figure 1. Ethanol increased the ChRE reporter activity.

The effect of ethanol on a ChREBP reporter construct was studied in H4IIEC3 cells. The cells were transfected with the reporter (liver pyruvate kinase, LPK, lucifierase), a ChREBP expression plasmid, and the internal control plasmid (Renilla luciferase reporter) and exposed to glucose and ethanol. The reporter activity was induced by incubation of the cells with glucose (25 mM). Ethanol markedly increased the activity of the ChREBP reporter by 4-fold when compared to controls. The data are expressed as relative luciferase activity compared to controls (mean ± S.E.) from at least six experiments performed in duplicate. *, p < 0.05; compared with control.

Effects of Ethanol Feeding on ChREBP in vivo

To determine the effect of ethanol on ChREBP in vivo, we first established the effects of ethanol feeding of mice using the usual liquid diet pair feeding protocol as described above. Histological analysis by Oil Red O staining showed prominent accumulation of lipid droplets in the livers of ethanol-fed mice, whereas lipid droplets were rare in the livers of control groups (Figure 2A). As we previously published, a steady average 2- to 3-g increase in the body weight was observed in both groups during the entire 4-week study. The body weights did not vary between these two groups, but the liver weights and liver/body weight ratio were significantly increased in the low fat diet plus ethanol group (Figure 2B).

Figure 2.

(2A and 2B) Six to eight week old male C57BL/6J mice were fed with a low fat Lieber-DeCarli diet with and without ethanol as described in the text for 4 weeks. Hepatic histology (H&E and Oil red O stains) showed significant steatosis in ethanol-fed group. (2C snd 2D) Immunoblot analysis of ChREBP in nuclear and cytoplasmic extracts from livers of mice fed a low fat diet with or without ethanol. Ten male C57BL/6J mice were fed a low fat diet with (LF + E) or without ethanol (LF) for 4 weeks. Aliquots of nuclear (80 μg of protein) and cytosol (40 μg of protein) from livers of each group were electrophoresed on 8% SDS-polyacrylamide gels, transferred to nitrocellulose membranes, and immunostained with anti-ChREBP. Addition of ethanol into low fat diet resulted in the preferential localization of ChREBP protein in the nucleus. (2E) DNA binding ability of ChREBP to its canonical response element. Hepatic nuclear extract was used to determine the ability of ChREBP to bind to its response element using a streptavidin-biotin complex DNA binding assay. Non-biotinylated LPK-ChoRE (Lane 5) significantly competed with the binding of biotinylated LPK to nuclear protein, demonstrating the specificity of the binding. Ethanol feeding significantly increased DNA binding ability of ChREBP when compared to pair-fed controls (Lanes 5 and 6). (2F) Hepatic tissue expression of enzyme involved in lipogenesis. mRNA levels were determined by real time reverse transcription-PCR from liver tissues of C57BL/6J mice after 4 weeks of LF or LF diet with ethanol feeding. All data are shown as the mean ± S.D. when compared to LF controls, *, p < 0.05.

Immunoblot analysis of the liver nuclear and cytolsolic extracts was performed. As shown in Figs. 2C and D, ethanol feeding resulted in the translocation of ChREBP protein from cytosol to the nucleus. We next determined the ability of ChREBP to bind to its response element using streptavidin-biotin complex DNA binding assay. As shown in Figure 2E, ethanol feeding significantly increased DNA binding ability of ChREBP when compared to pair-fed controls. Two molecular forms of ChoRE-binding activity were observed which were eliminated by the presence of non-biotinylated oligonucleotides. The larger form is very close to the predicted size of the ChREBP protein (865 aa, or approximately 95 kDa). The identity of the lower band is uncertain. No isoforms of ChREBP are reported; however, humans express 4 alternatively spiced mRNA variants of differing in length. To determine whether these effects of ethanol on ChREBP were associated with an increase in the expression of genes known to be regulated by ChREBP, real-time quantitative polymerase chain reactions were performed. As shown in Figure 2F, ethanol feeding increased the expression of mRNA for liver pyruvate kinase by ~ 4-fold when compared to pair-fed controls. Additionally, other target genes including fatty acid synthase and acetyl Co-A carboxylase were also increased with ethanol feeding, as previously reported. The level of Srebf1 mRNA was not altered by ethanol. In addition, the mRNA levels of hepatic ChREBP did not differ between controls and ethanol fed mice (1± 0.3 vs. 1.4±0.5, p = 0.28); suggesting that the effect of ethanol on ChREBP is at post translational levels.

Mechanism for the activation of ChREBP by ethanol

The regulation of ChREBP nuclear translocation and transcriptional activity involves, among other things, the balance between dephosphorylation of the protein by members of the PP2A family of protein phosphatases and phosphorylation by AMPK or PKA. PP2A activity was increased in the liver of ethanol-fed mice by 22% when compared to pair-fed controls (Figure 3A). The pentose phosphate intermediate, Xu-5-P, has been shown to activate one isoform of PP2A which can dephosphorylate ChREBP, thereby allowing its translocation into the nucleus. We found no difference in the levels of hepatic Xu-5-P between ethanol-fed mice and controls (1.2 ± 0.5 vs. 2.3 ± 1.4 nmol/g, respectively, p = 0.15, Figure 3B), suggesting that the effect of ethanol on PP2A is mediated though a Xu-5-P independent pathway. We previously showed that ethanol activated PP2A by increasing intracellular levels of ceramide ⁷. We next determined whether ceramide and PP2A could be involved in the ethanol effect on ChREBP. As shown in Figure 3C, treating H4IIEC3 cells with ceramide increased LPK luciferase activity by 80–100%. If ceramide and ethanol are affecting ChREBP action by way of PP2A, this should be sensitive to inhibitors of PP2A. Okadaic acid is a potent inhibitor of protein phosphatases, inhibiting PP2A completely and relatively specifically at 1–2 nM ⁸. As shown in Figure 3C, okadaic acid significantly but incompletely attenuated the stimulatory effect of ethanol on ChREBP transcriptional activity. We found no effect of okadaic acid alone on the activity of the ChREBP reporter (data not shown).

Figure 3.

(3A–3B) Effect of ethanol on PP2A activity and hepatic xylulose-5-P concentration. Hepatic tissues from mice fed with low fat and low fat with ethanol diets were prepared as described in text and measured for PP2A activity (3A) and xylulose-5-P concentrations (3B) and. *, p < 0.05, when compared to low fat fed controls. (3C–3D) Effects of AMPK expression plasmids, ceramide, and okadaic acid on LPK reporter and PP2A activity. The hepatoma cells were transfected as described for Figure 1. In addition, AMPK expression plasmids or ceramide were tested for their effect on expression of the ChREBP reporter plasmid. (Fig 3C) Effect of ceramide on LPK luciferase activity. Ceramide increased the activity of LPK luciferase activity by 80–100%. Okadaic acid significantly attenuated the stimulatory effect of ethanol on liver pyruvate kinase luciferase activity. (Fig 3D) Effect of constitutively active AMPKα (AMPK α1³¹²) and dominant negative AMPK (DN-AMPKα) on the activity of the LPK reporter in H4IIEC3 cells. Constitutively active AMPK suppressed basal LPK luciferase activity by ~ 50% and significantly reduced the effect of ethanol on its activity. Expression of a dominant negative subunit of AMPKα1 (DN-AMPK α, 1 μg) induced basal LPK luciferase activity by ~20%. There was no additive effect with ethanol. *, p < 0.05, when compared to controls, ^$ when compared to ethanol-treated cells.

AMPK has been reported to reduce the activity of ChREBP in the liver. Hence, we hypothesized that one of the possible mechanisms by which ethanol increased ChREBP activity was mediated by inhibition of AMPK, a well known effect of ethanol ⁵ that is partly mediated by activation of PP2A. We tested the effect of a catalytically inactive mutant of the subunit AMPKα1 (α1DN AMPK), which has a dominant-negative effect on both α1 and α2 AMPK activities by competing for the binding of the β and γ subunits, and the small molecule inhibitor of AMPK, compound C. Transfection of rat hepatoma cells with α1DN AMPK induced basal LPK luciferase activity by ~20% and there was no additive effect ethanol on activation of the liver pyruvate kinase reporter (Figure 3D). The effect of the DN-AMPK was not as robust as that of ethanol, possibly due to the transfection efficiency. We also attempted to inhibit AMPK with the direct inhibitor compound C ⁵, but found unacceptable cell toxicity during 48 hr of culture post transfection.

Conversely, we tested whether co-transfection of a constitutively active form of AMPK, AMPKα1³¹², could reverse the effect of ethanol on the reporter. Truncation of AMPKα1 at the first 312 residues produces an enzyme that retains significant kinase activity without requiring interactions with the β and γ subunits. Additionally, mutation of threonine 172 within the truncated α subunit to an aspartic acid residue mimics the effect of phosphorylation, creating a constitutively active enzyme. We have previously tested this expression plasmid and found that transfection of AMPKα1³¹² resulted in ~ 2-fold increase in AMPK activity in H4IIEC3 cells ⁵. Transfection of the AMPKα1³¹² expression plasmid suppressed the basal activity of the LPK luciferase reporter by ~ 50% and abolished the effect of ethanol on ChREBP activity (Figure 3D).

DISCUSSION

The prominent ability of ethanol to induce hepatic lipogenesis has been attributed in part to activation of SREBP ^6;18. We now provide evidence that ethanol also activates the transcription factor ChREBP. SREBP-1c binds to the sterol response element DNA sequence, upregulating the synthesis of lipogenic enzymes. SREBP-1c is a major mediator of insulin ¹⁹ and ethanol ⁶ effects on lipogenic genes. However, SREBP-1c is not the only transcription factor involved in this process, as shown by several lines of evidence. First, the increase in the activity of SREBP-1c is not sufficient to account for the stimulation of glycolytic and lipogenic gene expression in the high carbohydrate-induced lipogenesis model ¹⁰. Second, the induction of liver pyruvate kinase, LPK, the rate-limiting enzyme of glycolysis and thus required for conversion of glucose to lipid, with high carbohydrate feeding, is not dependent on SREBP-1c ⁹. Lastly, SREBP-1c knockout mice fed a high carbohydrate diet had only a 50% reduction in hepatic fatty acid synthesis ²⁰. The transcription factor ChREBP appears to account for these phenomena by acting in concert with SREBP.

ChREBP is a glucose-responsive basic/helix–loop–helix/leucine zipper (bHLH/LZ) transcription factor ²¹. The regulation of the activity and expression of ChREBP involves nuclear/cytosolic partitioning and the ability of this protein to bind to its response element. This process is modulated by the phosphorylation/dephosphorylation of the transcription factor. AMPK phosphorylates ChREBP at Ser⁵⁶⁸, hindering cytosol-to-nuclear translocation and thus reducing its transcriptional activity ¹⁴. AMPK activity is sensitive to the energy state of the cells and various stresses. Under low energy conditions, AMPK is active and ChREBP is retained in the cytosol. When glucose is abundant, the activity of AMPK is low, leading to nuclear translocation of ChREBP and induction of glucose responsive genes. ChREBP is also under the control of PP2A. This phosphatase dephosphorylates several phosphorylation sites on ChREBP, such as Ser¹⁹⁶, Thr⁶⁶⁶, and Ser^{568 21}. Uyeda et al. reported that specific isoforms of nuclear and cytosolic PP2As are activated by Xu-5-P, an intermediate in glucose metabolism by the pentose shunt pathway. In the presence of high glucose, glucose is converted to Xu-5-P leading to activation of enzyme PP2A and dephosphorylation of ChREBP, thus increasing its activity by promoting nuclear translocation ¹⁵.

Treatment of hepatoma cells with physiologically relevant concentrations of ethanol strongly activated ChREBP (Figure 1). Ethanol caused nuclear translocation of ChREBP protein and increased its ability to bind to its response element, leading to an increase the expression of genes (LPK, ACC, FAS) known to be regulated by ChREBP, establishing this new mechanism for alcohol induced steatosis. We then explored the mechanisms underlying the ability of ethanol to activate ChREBP. Ji et al. found that intragastric alcohol infusion to rats for 4 weeks significantly increased hepatic mRNA and protein expression of ChREBP compared to pair-fed controls ¹⁸. This may reflect the presence of a sterol-response element in the promoter of the ChREBP gene ²². In addition to this induction of ChREBP by ethanol, our study provides additional mechanisms for increased ChREBP transcriptional activity. These mechanisms appear to involve the ability of ethanol to activate PP2A, since okadaic acid partially reversed this effect. This effect likely involves AMPK as well, since PP2A can also control the activity of this enzyme (see below).

PP2A can be activated by Xu-5-P and ceramide. We measured levels of these metabolites in livers of ethanol-fed mice. There were no differences in the hepatic levels of Xu-5-P compared to controls, suggesting that the effect of ethanol on ChREBP is Xu-5-P independent, and in fact is consistent with the fact that the ethanol-containing diet contains less carbohydrate than the control diet (in which ethanol calories are replaced with maltose-dextrin). It is possible that there would be additive effects of ethanol in the setting of a high carbohydrate diet on the development of fatty liver. On the other hand, ethanol increased ceramide levels and exogenous ceramide increased ChREBP activity. However, the effect of ceramide on ChREBP was not as great as that of ethanol. This might, of course, relate to differences in the intracellular sites of ceramide generation and the fate of extracellular ceramide vs intracellularly generated ceramide. However, the modest effect of added ceramide and the inability of okadaic acid to completely block the effect of ethanol, suggests that ethanol has additional effects on ChREBP.

AMPK may be responsible for the okadaic acid-insensitive effects of ethanol. We suspect that the modest effect of transfection of the dominant negative AMPK expression plasmid was due to only a fraction of the cells being transfected and expressing this protein; unfortunately we were unable to use the small molecule inhibitor compound C to better block AMPK activity because of its toxicity. The ability of ethanol to inhibit AMPK is ameliorated by okadaic acid or silencing PP2A (13), but ethanol may reduce AMPK activity by having additional actions not involving PP2A. If ethanol reduces AMPK activity in the cells even in the presence of okadaic acid, there may be addition effects on ChREBP due to reduced phosphorylation at critical regulatory sites. In addition. AMPK may alter chromatin and ChREBP acetylation. Recent work showed that ChREBP activates transcription in a complex formed on the L-PK promoter with HNF4, and either CBP ²³ or p300 ²⁴, and that in the presence of high glucose, the H3 and H4 histones associated with the L-PK promoter are hyperacetylated. Conversely, increased cAMP levels interfered with the assempbly of this complex and reduced histone acetylation ²³. Chronic ethanol treatment of cultured rat hepatocytes increased rather than decreased glucagon-stimulated cAMP levels ²⁵, suggesting this is not a mechanism by which ethanol activates ChREBP-mediated reporter activity. However, very recently the salt-inducible kinase 2 (SIK2), a member of the AMPK family, was shown to inhibit the transciprtional activity of ChREBP. The mechanism of this effect is interesting: SIK2 phosphoryates p300 at Ser89, inhibiting its histone acetylase activity and its ability to acetylate ChREBP on Lys672 ²⁴. Acetylation of this residue activates ChREBP transcriptional activity. Presumedly, SKI2 phosphorylation of p300 also reduces histone acetylation in the L-PK promoter region. Of note, ethanol has been reported to increase histone acetylase activity in liver cells ²⁶, and AMPK was reported to phosphorylate p300 on Ser89 ²⁷. This might be another mechanism whereby activation of AMPK using AICAR ¹⁴ or transfection of a constitutively active AMPK expression plasmid (Figure 3D) blocked the effect of ethanol on ChREBP activity. Thus, the presence of ethanol may increase histone (and ChREBP) acetylase activity of p300 or CBP in the cells, leading to increased activity of ChREBP in the transfection studies.

While it would be interesting to examine the ability of genetic knockouts of ChREBP to prevent ethanol-induced steatosis, such experiments could be difficult to interpret. The knockout mice are intolerant of dietary carbohydrate, have reduced body adipose tissue, and increased liver glycogen owing to impaired glycolysis. The impaired glycolysis results from markedly reduced LPK levels (mRNA levels were only 27% of normal). Due to the intolerance of the knockout mice to simple sugars, high starch diets had to be used to demonstrate the effect of the knockout on hepatic lipogenesis, and this was accompanied by elevated blood glucose and insulin levels, as well as reduced free fatty acid levels. Thus, while it is likely that ethanol-induced triglyceride synthesis and steatosis would be reduced in ChREBP-knockouts, it would likely be difficult to understand the mechanism responsible for this observation ¹³.

In summary, the results of the current study demonstrate an additional mechanism, activation of ChREBP (Figure 4), for the development of alcoholic fatty liver, and have significant implications for therapy. The epidemiological assocation of more severe alcoholic liver disease with obesity (21) might reflect synergism between high alcohol and carbohydrate intake, leading to a greater activation of ChREBP and worse steatosis. Activation of AMPK or inhibition of ceramide generation might prove a rational approach to controlling alcoholic fatty liver. In fact, we previously showed that the effect of ethanol on SREBP was also mediated in part through inhibition of AMPK (5,6). Thus, AMPK activators seems to be attractive compounds for the treatment of alcoholic fatty liver disease. The role of PP2A inhibitors and compounds that interfere with ceramide synthesis on alcohol-induced hepatic steatosis deserve further study.

Figure 4. Proposed mechanism on ethanol effects on ChREBP.

The inhibitory effects of ethanol on AMPK phosphorylation are secondary to i) direct inhibition of upstream kinases such as PKCζ and LKB1 ⁸ and ii) increased intracellular ceramide concentrations leading to increased PP2A activity ⁷. Increases in PP2A and decreases in AMPK activity lead to dephosphorylation of ChREBP, thus increasing its activity and its nuclear localization. Inhibition of AMPK might also reduce the phosphorylation of p300/CBP, leading to increased acetylase activity of this protein and altering chromatin and ChREBP acetylation. In turn, this could activate ChREBP transcriptional activity.