Elevated PTEN Levels Account for the Increased Sensitivity of Ethanol-exposed Cells to Tumor Necrosis Factor-induced Cytotoxicity*

Tumor necrosis factor (cid:1) (TNF) is known to be one of the primary causative cytokines inflicting the characteristic damage to hepatocytes seen in alcoholic liver disease. TNF activates both cell survival and death-in-ducing signaling pathways. The balance between these two prongs determines the fate of the cell and the onset of disease. Ethanol exposure has been shown to alter mitochondrial function, decreasing their threshold for injury. Importantly, mitochondrial injury is a necessary end point of TNF-induced cell killing. It has been shown that ethanol exposure increases the sensitivity of hepatocytes and HepG2E47 cells to TNF-mediated death. The cumulative and terminal effect of the increased sensitivity to TNF caused by ethanol is an induction of a mitochondrial permeability transition. TNF brings about the mitochondrial permeability transition in eth-anol-exposed cells due to amplification in the activity of the p38 stress kinase and a diminution in the activity of the antiapoptotic Akt/PKB kinase. The present report identifies an increase of PTEN expression in ethanol-exposed cells as the main causative factor in altering the balance between prosurvival and prodeath signals initiated by TNF. Suppression of the elevated PTEN levels found in ethanol-exposed HepG2E47 cells through the use of RNA interference reversed the ethanol-induced alterations to TNF signaling, resulting in a preservation of mitochondrial function and cell viability. and cytotoxicity imposed by the elevated PTEN levels induced by ethanol exposure. In normal non-ethanol-exposed cells, the TNF re- ceptor activates a survival pathway mediated by NF- (cid:2) B and Akt activation. Ethanol exposure increases PTEN levels, causing an inhibition of Akt activation by TNF. The resulting suppression by PTEN of Akt activity results in a blunting of the survival pathway mediated by NF- (cid:2) B and a potentiation of ASK-1 activation, resulting in an augmented p38 response. In turn, p38 promotes the translocation of Bax to the mitochondria with subsequent induction of mitochondrial injury and cell death.

Tumor necrosis factor ␣ (TNF) is known to be one of the primary causative cytokines inflicting the characteristic damage to hepatocytes seen in alcoholic liver disease. TNF activates both cell survival and death-inducing signaling pathways. The balance between these two prongs determines the fate of the cell and the onset of disease. Ethanol exposure has been shown to alter mitochondrial function, decreasing their threshold for injury. Importantly, mitochondrial injury is a necessary end point of TNF-induced cell killing. It has been shown that ethanol exposure increases the sensitivity of hepatocytes and HepG2E47 cells to TNF-mediated death. The cumulative and terminal effect of the increased sensitivity to TNF caused by ethanol is an induction of a mitochondrial permeability transition. TNF brings about the mitochondrial permeability transition in ethanol-exposed cells due to amplification in the activity of the p38 stress kinase and a diminution in the activity of the antiapoptotic Akt/PKB kinase. The present report identifies an increase of PTEN expression in ethanolexposed cells as the main causative factor in altering the balance between prosurvival and prodeath signals initiated by TNF. Suppression of the elevated PTEN levels found in ethanol-exposed HepG2E47 cells through the use of RNA interference reversed the ethanol-induced alterations to TNF signaling, resulting in a preservation of mitochondrial function and cell viability.
Alcoholic liver disease (ALD) 1 encompasses the spectrum of fatty liver, alcoholic hepatitis, and alcoholic cirrhosis. Fatty liver is a benign and reversible condition that develops in response to short term alcohol abuse. Prolonged alcohol abuse of 80 g of alcohol per day, which is the equivalent of 6 -8 drinks daily, predisposes to alcoholic hepatitis that can progress to cirrhosis. Cytokines are critical in initiating and perpetuating the injury that leads to ALD. ALD is associated with an increase in the activity and infiltration of Kupffer cells and neutrophils into the liver parenchyma (1,2). At this locale, inflammatory cells release a number of proinflammatory cytokines and chemokines. One of the most prevalent and best studied is tumor necrosis factor ␣ (TNF). TNF is a central inciting agent in the genesis of ALD (3)(4)(5)(6)(7). This is suggested by the ability of Kupffer cell depletion to prevent the hepatic damage associated with an ethanol intragastric feeding model in rats (8). Additionally, antibodies neutralizing TNF prevented the onset of liver damage (9). In cases of ALD, a correlation was established between TNF levels and measurements of liver damage, such as serum alanine aminotransferase, creatinine, and bilirubin, in addition to a greater likelihood of mortality in patients with higher serum TNF values (10).
The binding of the TNF ligand to its receptor (TNFR) activates a series of intracellular signaling modules, resulting in a broad range of outcomes including increased cell survival, proliferation, or death. The interplay of prosurvival and death signals that are engaged by TNFR activation and how this is integrated by the cell into a phenotypic response are murky. In ALD, the hepatocyte sustains the bulk of the damage evoked by TNF, mediated predominantly by the 55-kDa TNF receptor (TNFR55). Similar to other cell types, hepatocytes are mostly resistant to the death-inducing effects of TNF. This is due to the fact that TNF activates signaling pathways resulting in protection from the cytotoxic features of TNF. Such pathways include activation of the NF-B transcription factor and the phosphatidylinositol (PI) 3-kinase-Akt signaling pathway (11,12). However, when exposed to ethanol, hepatocytes exhibit an increased sensitivity to TNF-induced cell killing (13,14). Ethanol's effects on hepatocyte sensitivity to TNF are multifaceted, including an increase of reactive oxygen species formation and an increase in the levels of the TNF receptor. Additionally, mitochondria isolated from the livers of ethanol-fed rats exhibit an increased propensity for triggering of the mitochondrial permeability transition (MPT) (15,16). The MPT results in the collapse of the electrochemical gradient across the inner mitochondria membrane from the formation of pores, which can cause the release of intermembrane space proteins such as cytochrome c to the cytoplasm. Importantly, development of the MPT has been shown to be a common nidus to which an array of TNF-initiated signaling pathways converge. Such studies point to a lowering of the threshold to induction of the MPT evoked by ethanol. This may be partly explained by the changes caused in the mitochondria by ethanol exposure, such as a decrease of mitochondrial glutathione or a lowering of mitochondrial ribosomal and DNA content (17)(18)(19). However, it is unclear how ethanol exposure modifies upstream signaling * This study was supported by National Institute on Alcohol Abuse and Alcoholism Grants K01-AA-00330-1 and 1-R01-12897-01A2. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  events initiated by TNF to induce the MPT in the compromised mitochondria.
Utilizing both primary hepatocytes isolated from control-fed and ethanol-fed rats and the hepatoma cell line HepG2E47, expressing cytochrome P4502E1, we have shown that ethanol exposure alters the intracellular signaling response to TNF (20). Specifically, ethanol-exposed cells exhibit an increase and decrease of p38 and Akt activity, respectively, upon TNF treatment. The increased p38 activity is responsible for mediating the translocation of Bax from the cytosol to the mitochondria. Bax has been shown to bring about mitochondrial damage, including induction of the MPT (21)(22)(23). Inhibition of the intensified p38 response prevented Bax activation, mitochondrial dysfunction, and the cell killing induced by TNF in ethanolexposed cells (20). Conversely, increasing the activity of Akt in ethanol-treated cells by the use of a constitutively active construct of Akt preserved cell viability. Cell surface receptors such as the TNFR activate phosphatidylinositol 3-kinase, which phosphorylates PIP 2 to generate phosphatidylinositol 3,4,5-trisphosphate (PIP 3 ). PIP 3 is necessary for Akt activation. PTEN (phosphatase and tensin homologue deleted from chromosome 10) is a tyrosine phosphatase with dual protein and lipid phosphatase activity. PTEN recognizes PIP 3 as a substrate and removes the D3 phosphate from the inositol ring. Loss of PTEN as seen in certain cancers leads to the accumulation of PIP 3 , resulting in the constitutive activation of the phosphatidylinositol 3-kinase/Akt pathway. By contrast, an increase of PTEN levels would be expected to decrease the concentration of PIP 3 and cause an inhibition of receptor-stimulated Akt activity. The present report identifies an increase of PTEN expression caused by ethanol exposure as being accountable for mediating the array of alterations in TNF signaling seen in ethanol-exposed cells, including the accentuated p38 response, inhibition of Akt and NFB signaling, and ultimately the mitochondrial damage culminating in cell death.

MATERIALS AND METHODS
Cell Culture and Treatments-Hepatoma G2E47 (HepG2E47) cells that express cytochrome P-4502E1 (CYP2E1) (kindly provided by Dr. Arthur I. Cederbaum) were maintained in 25-cm 2 polystyrene culture flasks with 5 ml of minimal essential medium containing 100 units/ml penicillin, 0.1 mg/ml streptomycin, 0.4 mg/ml G418, 1 mM pyruvate, and 10% heat-inactivated fetal bovine serum (complete minimal essential medium) incubated under an atmosphere of 95% air, 5% CO 2 at 37°C. Cells were subcultured 1:5 once a week. The cells were treated with 25 mM ethanol for 48 h before treatment with TNF. The culture medium was replaced every 24 h with fresh medium containing ethanol. To prevent evaporation of ethanol, the flasks or plates were placed in a plastic dessicator in the incubator containing a mixture of water and ethanol. The level of ethanol in the culture medium was monitored spectophotometrically by an alcohol dehydrogenase assay. The level of NADH was measured by an increase in absorbance at 340 nM. After 24 h of ethanol exposure, the cells (control or ethanol-treated) were trypsinized, counted in a hemocytometer, and plated at 1.0 ϫ 10 5 cells into 1.88-cm 2 wells of a 24-well plate for cell viability and membrane potential measurements or into 6-well, 9.3-cm 2 plates at 1.0 ϫ 10 6 cells for determination of caspase-3 or -8 activity. The cells were allowed to attach and spread for 24 h. When present, ethanol was added back to the cells during this time (total time of ethanol exposure, 48 h). On the day of the experiment, the cells were washed and placed in minimal essential medium without serum and in the absence of ethanol. TNF was dissolved in PBS and added to the wells in a 0.2% volume to give a final concentration of 10 ng/ml (220 units/ml).
Preparation of Hepatocytes and Ethanol Diet-Male Sprague-Dawley rats (140 -160 g) were obtained from Charles River Laboratories (Raleigh, NC). Nutritionally adequate liquid diets were formulated according to the method of Lieber and DeCarli (51). The animals were maintained on the ethanol-containing or control diets for 8 weeks. The ethanol-containing diet consisted of 18% total calories as protein, 35% as fat, 11% as carbohydrate, and 36% as ethanol. In the control diet, ethanol was replaced isocalorically with maltose-dextrin. The rat had a daily alcohol intake of 10 -12 g of ethanol/kg/day. This intake results in an average blood ethanol concentration of 25-30 mM. Isolated hepatocytes were prepared by collagenase perfusion according to the method of Seglen (30). Yields of 2-4 ϫ 10 8 cells/liver with 90 -95% viability as assessed by trypan blue exclusion were routinely obtained. Hepatocytes were either plated in 24-or 6-well plates at 1.0 ϫ 10 5 or 1.0 ϫ 10 6 cells/well, respectively.
Measurements of Cell Viability-Cell viability was determined by trypan blue exclusion and the ability of cells to reduce 3-(4,5dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS). For trypan blue exclusion, 10 l of a 0.5% solution of trypan blue was added to the wells. For each data point, we have randomly selected eight microscopic fields and blindly counted a representative group of cells in each well for viability. Each experiment was carried out a minimum of three times. For the MTS assay, the reaction was started by the addition of MTS and phenazine methosulfate (PMS). The absorbance change obtained upon reduction of MTS by viable cells was read 90 min later with a plate reader at 490 nm. 100% cell death was determined by the addition of Triton X-100 to a final concentration of 0.5%, 30 min before MTS and PMS addition. The MTS assay and trypan blue exclusion gave identical results.
Measurement of Mitochondrial Energization-Mitochondrial energization was determined as the retention of the dye DiOC 6 (3) (Molecular Probes, Inc., Eugene, OR). Cells were loaded with 40 nM of DiOC 6 (3) during the last 30 min of treatment. The cells were then washed twice in PBS. The level of retained DiOC 6 (3) was measured on a Bio-Tek HTS plate reader at 488-nm excitation and 500-nm emission.
Detection of Caspase-3 and Caspase-8 Activity-The assay is based on the ability of the active enzymes to cleave the chromophore pNA from the enzyme substrates DEVD-pNA (caspase-3) or IETD-pNA (caspase-8). Cell extracts were prepared and diluted 1:1 with 2ϫ reaction buffer (10 mM Tris, pH 7.4, 1 mM dithiothreitol, 2 mM EDTA, 0.1% CHAPS, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml pepstatin, 10 g/ml leupeptin). DEVD-pNA or IETD-pNA was added to a final concentration of 50 M, and the reaction was incubated for 1h at 37°C. Samples were then transferred to a 96-well plate, and absorbance measurements were made in a 96-well plate reader at 405 nm.
Isolation of Cytosolic and Mitochondrial Fractions-Cells were plated in 25-cm 2 flasks at 5.0 ϫ 10 6 cells/flask. After treatments, the cells were harvested by trypsinization followed by centrifugation at 600 ϫ g for 10 min at 4°C. The cell pellets were washed once in PBS and then resuspended in 3 volumes of isolation buffer (20 mM HEPES, pH 7.4, 10 mM KCl, 1.5 mM MgCl 2 , 1 mM sodium-EDTA, 1 mM dithiothreitol, 10 mM phenylmethylsulfonyl fluoride, and 1.0 g/ml leupeptin, 1.0 g/ml aprotinin in 250 mM sucrose). After being chilled on ice for 3 min, the cells were disrupted by 40 strokes of a glass homogenizer. The homogenate was centrifuged twice at 2,500 ϫ g at 4°C to remove unbroken cells and nuclei. The mitochondria were then pelleted by centrifugation at 12,000 ϫ g at 4°C for 30 min. The supernatant was removed and filtered through 0.2-m and then 0.1-m Ultrafree MC filters (Millipore Corp.) followed by centrifugation at 100,000 ϫ g at 4°C to yield cytosolic protein. Cytosolic fractions (25 g of protein) were separated on 12% SDS-polyacrylamide gels and electroblotted onto nitrocellulose membranes. Cytochrome c was detected by a monoclonal antibody (Pharmingen, San Diego, CA) at a dilution of 1:5,000. Secondary goat anti-mouse horseradish peroxidase-labeled antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) diluted at 1:10,000 was detected by enhanced chemiluminescence.
Western Blot Analysis-After various treatments, cells were harvested by trypsinization The cells were then counted, and equal numbers of cells were centrifuged at 500 ϫ g for 5 min and washed once with PBS. Cell pellets were then lysed in SDS sample buffer. Samples were separated on 12% SDS-PAGE and electroblotted onto nitrocellulose membranes. Phospho-ASK-1 was detected by rabbit polyclonal antibody at a dilution of 1:500 (Cell Signaling Technology, Beverly, MA). Bax was detected with an antibody that binds to the NH 2 terminus N-20 (Santa Cruz Biotechnology). PTEN was detected by a rabbit polyclonal antibody (Cell Signaling Technology, Beverly, MA). Lamin and actin were detected with a mouse monoclonal antibody (Affinity BioReagents and BD Biosciences Pharmingen, respectively). In each case, the relevant protein was visualized by staining with the appropriate secondary horseradish peroxidase-labeled antibody (1:10,000) and was detected by enhanced chemiluminescence.
Enzyme-linked Immunosorbent Assay of Akt, p38 MAPK, and NF-B Activation-Cells were collected in PBS by centrifugation. The cells were then lysed in cell extraction buffer (10 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM NaF, 20 mM Na 4 P 2 O 7 , 2 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, and protease inhibitor mixture plus 1% Triton X-100, 10% glycerol, 0.1% SDS, 0.5% deoxy-cholate). Standards, samples, or controls were added to microtiter wells coated with the appropriate antibody (anti-p38 MAPK, anti-Akt, or anti-p65). The wells were covered and incubated for 2 h at room temperature. The wells were then decanted and thoroughly washed four times. 100 l of an antibody, either anti-p38 MAPK (pTpy180/182), anti-p38 MAPK, anti-Akt (pS473), anti-Akt, or anti-p65, was added to the wells and incubated for 1 h at room temperature. The solution was decanted, and the wells were washed four times. Afterward, 100 l of horseradish peroxidase-conjugated anti-rabbit IgG was added to each well. The wells were incubated for an additional 30 min. The wells were decanted and washed four times. 100 l of stabilized chromogen was added to each well and incubated for 30 min in the dark; 100 l of stop solution was added to each well. The absorbance of each well was then read at 450 nm on a Synergy HTS microplate reader.
RNA Interference-The Dharmacon SMART selection and SMART pooling technologies are utilized for the RNA interference studies. The SMART selection uses an algorithm composed of 33 criteria and parameters that attempt to limit nonfunctional small interfering RNAs (siR-NAs). SMART pooling combines four SMART-selected siRNA duplexes in a single pool to increase the probability of reducing mRNA levels by at least 75%. In addition, the siRNA duplexes comprising the pool were tested individually for their effect on PTEN expression and phenotype of the cells. The siRNAs 1, 2, and 3 all decreased PTEN expression with siRNA 1 giving the most consistent results. Therefore, siRNA 1 was used in subsequent experiments to assess the effects of decreasing PTEN levels on the sensitivity of ethanol-exposed cells to TNF. The siRNA was delivered by a lipid-based method supplied from a commercial vendor (Gene Therapy Systems) at a final siRNA concentration of 100 nM. After formation of the siRNA-liposome complexes, the mix was added to the control or ethanol-exposed HepG2E47 cells in serum-free medium for 4 h. Afterward, the medium was aspirated, and complete medium was added back with or without the addition of 25 mM ethanol for a further 24 h.

RESULTS
Ethanol Exposure Increases PTEN Levels and PTEN Promoter Activity-HepGE247 cells were exposed to ethanol at a concentration of 25 mM over a 48-h time frame. The level of ethanol in the cell culture medium over a 24-h time course is shown in Fig. 1A. As can be seen, the ethanol level remained relatively constant, reaching a minimum of 22 mM ethanol at 24 h, at which point the cells were washed and fresh medium was added containing 25 mM ethanol, for a total exposure time of 48 h. The level of PTEN expression was then determined by Western blotting. As shown in Fig. 1B, HepG2E47 cells exposed to ethanol exhibited a progressive increase in PTEN expression with levels reaching 7-fold over control levels at 48 h of exposure, as determined by densitometric analysis. The cells were FIG. 1. Ethanol exposure induces an elevation of PTEN levels and PTEN promoter activation. A, ethanol levels in the cell culture medium of HepG2E47 cells were assessed at the times indicated spectrophotometrically by an alcohol dehydrogenase assay as described under "Materials and Methods." B, HepG2E47 cells were either left untreated or exposed to 25 mM ethanol for the times indicated. The cells were then harvested, and the levels of PTEN expression were assessed by Western blotting. Actin levels were assessed as loading controls. Alternatively, in C, hepatocytes were isolated from either control or ethanol-fed animals, and PTEN levels also were assessed by Western blotting. D, control or HepG2E47 cells exposed to 25 mM ethanol for 24 h where transiently transfected with the native luciferase reporter construct or the luciferase reporter driven by the 1359-bp PTEN promoter. After a further 24 h of ethanol exposure, the cells were harvested, and the luciferase activity in the cell lysates was determined. Alternatively, HepG2E47 cells were withdrawn from ethanol for 24 h, and PTEN promoter activity was determined. Luciferase activity in cell lysates was normalized to ␤-galactosidase expression and is expressed relative to values obtained using the empty reporter plasmid. Results shown are the mean of three independent experiments. subsequently washed and placed in medium without ethanol. Ethanol withdrawal resulted in a sharp decrease of PTEN levels by 16 h, with the PTEN levels returning to near control values by 24 h postexposure and going below them by 48 h. Also shown in Fig. 1C, PTEN levels were found to be elevated in hepatocytes isolated from ethanol-fed rats compared with their control-fed counterparts. This result is consistent with previous studies showing an elevation of PTEN in the livers of ethanol-fed rats (24). The ability of ethanol to increase PTEN levels is partly due to a stimulation of the PTEN promoter. A reporter plasmid was constructed containing 1350 base pairs of the human PTEN promoter (NCBI accession number NM_000314) inserted upstream of the luciferase gene. The plasmid was transfected into HepG2E47 cells, and the luciferase assay was performed at various time points during ethanol exposure. As can be seen in Fig. 1D, there is an ϳ5-fold increase in PTEN promoter activation after 24 h of ethanol exposure. Conversely, there was a rapid drop in PTEN promoter activity upon ethanol withdrawal, with levels decreasing by half at 24 h after ethanol treatment.
Selective Suppression of PTEN by siRNA-The method of RNA interference was utilized to suppress PTEN levels in ethanol-treated cells. A pool of four siRNA were designed by computer algorithm taking into account 34 separate variables. HepG2E47 cells were either left untreated or exposed to ethanol for 24 h. The cells were transfected with siRNA targeting PTEN or luciferase. The cells were then incubated for another 24 h in the presence or absence of 25 mM ethanol. As can be seen in Fig. 2A, 48 h post-transfection, the siRNAs targeting PTEN ablated its expression in both control (lane 3) and ethanol-exposed HepG2E47 cells (lane 4). By contrast, siRNA targeting luciferase had no effect on PTEN levels (lanes 5 and 6). The individual siRNAs constituting the pool were then tested for their ability to decrease expression of PTEN. PTEN levels were repressed by siRNAs 1, 2, and 3 (Fig. 2B). In addition, Fig.  2C demonstrates an increase of PTEN mRNA levels induced by ethanol exposure and its elimination by siRNA 1 targeting PTEN. A siRNA targeting lamin A/C was without effect on PTEN expression but did ablate the expression of lamin A/C (Fig. 2D). Importantly, siRNA 1 against PTEN did not alter the expression of lamin but did down-regulate expression of PTEN (Fig. 2D). Additionally, a nontargeting control siRNA that possesses no homology to any known gene had no effect on the levels of both PTEN and lamin A/C (Fig. 2D). These experiments validate the efficacy and selectivity of the siRNA targeting PTEN to decrease PTEN levels.
Elevated PTEN Levels Inhibit TNF Activation of Akt in Ethanol-exposed Cells-The phosphatidylinositol 3-kinase-Akt signaling pathway is critical in the inhibition of apoptosis. TNF activates Akt, and this is stunted in ethanol-exposed cells. As can be seen in Fig. 3A, TNF treatment of HepG2E47 cells results in a rapid increase in Akt activation, starting as early as 15 min following TNF addition and peaking at 10-fold above basal levels at 60 min. By contrast, HepG2E47 cells exposed to 25 mM ethanol for 48 h exhibited a restrained activation of Akt upon TNF treatment, with a delay in both the onset and extent of Akt activation, with peak levels of only 3-fold above basal levels at 60 min following TNF addition. HepG2E47 cells were

FIG. 2. Suppression of PTEN levels by RNA interference.
In all of the panels, HepG2E47 cells were either left untreated or exposed to 25 mM ethanol for 24 h. The cells were then transfected with siRNAs targeting PTEN, luciferase, or lamin A/C. The cells were then cultured for another 24 h in the presence or absence of ethanol. The cells were then harvested, and the levels of PTEN were determined by Western or Northern blotting. A, a Western blot showing the down-regulation of PTEN levels by a pool of four siRNAs targeting PTEN and the lack of effect of siRNAs against luciferase on PTEN levels. B, the down-regulation of PTEN levels by three separate siRNAs constituting the pooled siRNAs against PTEN. In addition, the lack of effect of siRNA targeting lamin A/C on PTEN levels is shown. C, Northern blot analysis demonstrates an increase of PTEN mRNA levels promoted by ethanol exposure and its prevention by siRNA 1 targeting PTEN. D, the specificity of the siRNA targeting PTEN is demonstrated by its lack of effect on lamin A/C levels. Conversely, siRNA targeting lamin A/C results in lamin A/C down-regulation but no effect on PTEN expression. A nontarget control siRNA having no homology to any known gene had no effect on the levels of PTEN or lamin. Results shown are typical of three independent experiments. exposed for 24 h to 25 mM of ethanol followed by transfection with siRNA 1 targeting PTEN or siRNA against lamin A/C, whereupon they were exposed to ethanol for an additional 24 h. The cells were then treated with TNF, and Akt activation was determined. As can be seen in Fig. 3A, suppression of the ethanol-induced elevation of PTEN restored the ability of TNF to activate Akt, with levels of Akt phosphorylation reaching those seen in cells not exposed to ethanol. By contrast, siRNA to lamin A/C did not restore the ability of TNF to activate Akt in cells treated with ethanol. Importantly, as shown in Fig. 3B, the level of Akt protein was the same in control, ethanolexposed cells and in ethanol-exposed cells treated with siRNA against PTEN, thus showing that the measured decrease in the activation of Akt by TNF in ethanol-treated cells and its restoration by siRNA targeting PTEN cannot be accounted for by a variability of total Akt levels.
Suppression of PTEN Restores TNF-induced NF-B Activation in Ethanol-exposed Cells-The transcription factor NF-B is a critical mediator in the protective pathways initiated by TNF. Akt has been shown to augment the activation of NF-B in some contexts (12,25). As shown in Fig. 4, TNF treatment led to a rapid and robust activation of NF-B in control HepG2E47 cells over a 2-h time course. By contrast, in cells exposed to ethanol for 48 h, TNF treatment induced only a modest increase of NF-B activity. However, as with TNF stimulation of Akt, the ability of TNF to activate NF-B was restored in both rate and extent to that seen in control cells not exposed to ethanol when PTEN levels were ablated by siRNA. By contrast, ethanol-exposed cells transfected with siRNA to lamin A/C displayed no reversal of the inhibition of TNFinduced NFB activation induced by ethanol. In order to establish the importance of Akt in mediating TNF activation of NF-B, stable transfectants expressing a constitutively activated form of Akt (myristoylated Akt) were generated. As can be seen in Fig. 4, ethanol exposure was unable to prevent TNF from activating NF-B in the cells expressing constitutively active Akt. TNF stimulated NF-B to the same rate and extent in ethanol-treated cells expressing constitutively activated Akt as in cells not exposed to ethanol. These results indicate that the elevation of PTEN levels evoked by ethanol exposure and the resultant repression of TNF-induced Akt activity are responsible for the decreased stimulation of NF-B by TNF.
Elevated PTEN Levels Potentiate TNF Activation of p38 in Ethanol-exposed Cells-We have shown that ethanol exposure results in a magnified response of the stress kinase, p38, to TNF treatment. These results were reproduced and are shown in Fig. 5. TNF stimulates an initial moderate activation of p38 in control cells that peaks at 15 min and rapidly decays after 1 h. TNF stimulated p38 to the same rate and extent in ethanol-exposed cells as in nonexposed controls during the initial 45 min of treatment. However, there was a second and much greater increase of p38 activation induced by TNF in the ethanol-exposed cells, with p38 activation reaching 6-fold above basal levels at 2 h and being sustained for up to 4 h. By contrast, in ethanol-exposed cells in which PTEN expression was suppressed by siRNA, there was a restrained stimulation of p38 by TNF treatment, with the cells exhibiting only the early onset and modest p38 activation seen in control cells and FIG. 3. Elevated PTEN levels in ethanol-exposed cells blunt activation of Akt by TNF. A, HepG2E47 cells were either exposed to 25 mM ethanol for 24 h or left untreated. The cells were then transfected with siRNA targeting PTEN or lamin A/C. After a further 24 h of culture in the presence or absence of ethanol, the cells were treated with 10 ng/ml TNF. The cells were harvested at the indicated time points. The level of activated Akt phosphorylated on serine 473 was determined by enzyme-linked immunosorbent assay. B, the level of total Akt was determined in control, HepG2E47 cells exposed to ethanol for 48 h and in ethanol-exposed HepG2E47 cells treated with siRNA targeting PTEN. Results shown are the mean of three independent experiments.
not the later and more robust p38 activation stimulated by TNF in cells exposed to ethanol. By contrast, siRNA to lamin A/C had no effect on the augmented activation of p38 by TNF in ethanol-treated cells. Furthermore, as shown in Fig. 5, expression of constitutively active Akt (Myr-Akt) was also able to suppress the augmented stimulation of p38 brought about by TNF in ethanol-exposed cells. Such results indicate that the elevated PTEN levels characteristic of ethanol exposure inhibit the TNF-initiated Akt activity that is necessary to dampen the response of p38 to TNF treatment.
PTEN Prevents Inactivation of ASK-1 by TNF in Ethanoltreated Cells-Apoptosis signaling kinase-1 (ASK-1) is a MAPK kinase kinase that is capable of activating the p38 stress kinase pathway. It is regulated by both oxidative stress and phosphorylation. Akt negatively regulates ASK-1. Akt phosphorylates ASK-1 on serine 83, preventing it from activating p38. As seen in Fig. 6, TNF stimulates phosphorylation of ASK-1 on serine 83 in control cells not exposed to ethanol (Fig. 6A, lane 1 versus  lane 2). By contrast, ASK-1 phosphorylation on serine 83 did not occur in ethanol-exposed cells treated with TNF (Fig. 6A,  lane 3). The elevation of PTEN levels in ethanol-exposed cells was responsible for the suppression of ASK-1 phosphorylation by TNF. As shown in Fig. 6A, lanes 4 and 7, ethanol-exposed cells transfected with siRNAs targeting PTEN exhibited a restoration of ASK-1 phosphorylation induced by TNF, whereas siRNA targeting lamin A/C has no ability to reverse the inhibition of ASK-1 phosphorylation caused by ethanol treatment (lane 5). Similar to suppression of PTEN levels, expression of constitutively active Akt (Myr-Akt) also resulted in a restoration of ASK-1 phosphorylation on serine 83 in TNF-treated cells exposed to ethanol (Fig. 6A, lane 6). Such results indicate that the increased PTEN levels resulting from ethanol exposure are responsible for preventing the inhibitory phosphoryl-

FIG. 4. Inhibition of Akt stimulation by increased PTEN expression in ethanol-exposed cells prevents TNF-induced activation of NF-B.
HepG2E47 cells were either left untreated or exposed to 25 mM ethanol for 24 h. The cells were then either left untreated or transfected with siRNA 1 targeting PTEN or siRNA targeting lamin A/C. Following a further 24 h of ethanol exposure, the cells where treated with 10 ng/ml TNF. At the time points indicated, the cells were harvested, and nuclear extracts were prepared. Similarly, HepG2E47 cells expressing constitutively active Akt (Myr-Akt) were either left untreated or exposed to 25 mM ethanol for 48 h. The cells where treated with 10 ng/ml TNF. At the time points indicated, the cells were harvested, and nuclear extracts were prepared. The levels of the activated form of NF-B in the nuclear extracts were measured by an ability to bind the NF-B consensus site coated in a 96-well plate. After binding and washing, the p65 subunit was detected by primary and secondary antibodies, providing a colorimetric readout. Results shown are the mean of three independent experiments.

FIG. 5. Potentiation of TNF-induced p38 activity in ethanolexposed cells is dependent on an elevation of PTEN.
HepG2E47 cells were either left untreated or exposed to 25 mM ethanol for 24 h. The cells were then either left alone or transfected with siRNA 1 targeting PTEN or siRNA against lamin A/C. After a further 24 h of incubation in 25 mM ethanol, the cells were treated with 10 ng/ml TNF. Alternatively, HepG2E47 cells expressing constitutively active myristoylated Akt (Myr-Akt) were also either left untreated or exposed to 25 mM ethanol for 48 h, after which they were treated with 10 ng/ml TNF. At the indicated time points, the cells were harvested, and the lysates were assayed using an enzyme-linked immunosorbent assay for active p38 phosphorylated on tyrosine 180 and 182. Results shown are the mean of three independent experiments.  PTEN (lane 7), or lamin (lane 5), followed by exposure to ethanol for a further 24 h. Alternatively, HepG2E47 cells expressing constitutively active Akt (Myr-Akt) were exposed to ethanol for 48 h and then treated with TNF (lane 6). All of the cells were then treated with 10 ng/ml TNF and harvested after 30 min, and the level of ASK-1 phosphorylated on serine 83 was determined by Western blotting utilizing a phosphospecific antibody. B, HepG2E47 cells were either left untreated or transfected with siRNA targeting ASK-1 or PTEN. The levels of total ASK-1 were determined by Western blotting. C, HepG2E47 cells were either left untreated or exposed to 25 mM ethanol for 24 h. The cells were then either left alone or transfected with siRNA targeting ASK-1. For all conditions, at the indicated time points, the cells were harvested, and the lysates were assayed using an enzyme-linked immunosorbent assay for active p38 phosphorylated on tyrosine 180 and 182. Results shown are the mean of three independent experiments. ation of ASK-1 brought about by Akt during TNF treatment. Such a finding may account for the amplified pattern of p38 activity seen in TNF-treated cells exposed to ethanol.
To confirm that ASK-1 was the upstream kinase causing the magnified response of p38 to TNF treatment in ethanol-exposed cells, siRNA against ASK-1 was utilized. As seen in Fig.  6B, lanes 1 and 2, the siRNA targeting ASK-1 decreased its level 24 h post-transfection. Importantly, the siRNA to PTEN had no effect on ASK-1 levels (Fig. 6B, lane 3), and conversely the siRNA to ASK-1 did not influence PTEN expression (result not shown). As shown in Fig. 6C, suppression of ASK-1 levels did not prevent the initial response of p38 to TNF but did prevent the second and much greater phase of p38 stimulation seen only in ethanol-exposed cells. These results indicate that the first phase of p38 activation induced by TNF is independent of ASK-1, but that the second much larger p38 activation phase seen in ethanol-exposed cells treated with TNF is ASK-1-dependent.

Suppression of PTEN Prevents TNF from Promoting Baxinduced Mitochondrial Injury and Cell Killing in Ethanol-exposed Cells-
We have shown that cell death in ethanol-exposed cells treated with TNF is independent of caspase-8 but is accompanied by caspase-3 stimulation. As can be seen in Fig. 7A, suppression of PTEN did not decrease substantially the activation of caspase-8 by TNF treatment in ethanol-exposed cells but did prevent the elevation of caspase-3 activity provoked by TNF. The increase of caspase-3 activity in ethanol-exposed cells treated with TNF is dependent on mitochondrial injury. As can be seen in Fig. 7B, in ethanol-treated cells there is a small increase of the proapoptotic protein Bax in the mitochondria (Fig. 7B, lane 2). However, treatment of ethanol-exposed cells with TNF induces a rapid and robust translocation of Bax from the cytosol to the mitochondria in ethanol-exposed cells (Fig. 7B, lane 3), a translocation prevented by suppressing PTEN levels (Fig. 7B, lane 4). Concomitantly, lowering PTEN levels prevented the mitochondrial damage promoted by TNF in ethanol-exposed cells. As shown in Fig. 7C, ethanol exposure resulted in a TNF-induced release of the intermembrane space protein, cytochrome c, from the mitochondria and its accumulation in the cytosol. Repressing PTEN levels in ethanol-exposed cells treated with TNF prevented the loss of cytochrome c from mitochondria and its localization to the cytosol (Fig. 7C). Similarly, PTEN repression prevented mitochondrial membrane depolarization and cell death induced by TNF in ethanolexposed cells. TNF treatment of ethanol-exposed cells caused a rapid drop in mitochondrial membrane potential that was complete by 3 h (Fig. 8A, circles). The siRNA against PTEN prevented mitochondrial depolarization in ethanol-exposed cells treated with TNF (Fig. 8A, squares). TNF also induced cell killing in ethanol-exposed cells over a 6-h time course, with 80% of the cells losing viability by 6 h (Fig. 8B, circles). Suppression of PTEN levels substantially inhibited TNF-induced cell killing, with less than 20% of the ethanol-exposed cells losing viability at 6 h post-TNF treatment (Fig. 8B, squares). DISCUSSION The present study demonstrates that the alterations to TNF signaling pathways induced by ethanol exposure are largely attributable to an increase of PTEN levels. Suppression of the elevated PTEN levels induced by ethanol through the use of RNA interference resulted in a reversal in the responses of ethanol-exposed cells to TNF to those characteristic of cells not exposed to ethanol, including a restoration of TNF-induced Akt and NF-B activation, a decrease in the exaggerated p38 re- FIG. 7. Suppression of elevated PTEN levels in ethanol-exposed cells prevents the TNF-induced activation of Bax and caspase-3 and the release of cytochrome c into the cytosol. HepG2E47 cells were either left untreated or exposed to 25 mM ethanol for 24 h. The cells were then transfected with siRNA targeting PTEN and exposed to ethanol for a further 24 h. A, the cells were treated with 10 ng/ml TNF for the indicated time periods and harvested. The cell lysates were assayed for caspase-8 and -3 activity as described under "Materials and Methods." In B and C, at 2 h post-TNF treatment, the cells were harvested, and cytosolic (Cytosol) and mitochondrial (Mito.) fractions were prepared as described under "Materials and Methods." The levels of cytochrome c (Cyt. C) or Bax were assessed by Western blotting. The results shown are typical of three independent experiments. sponse, inhibitory phosphorylation of ASK-1, inhibition of caspase-3 activation, and a preservation of mitochondrial integrity and cell viability.
The phosphatidylinositol 3-kinase-Akt pathway is a central regulator of cell survival (12,26,27). Akt activation inhibits apoptosis through a multitude of mechanisms. Akt has been shown to phosphorylate and inactivate the proapoptotic proteins BAD, caspase-9, and Bax. In addition, a major part of TNF-induced cytoprotection is attributed to the activation of NF-B, which occurs as a result of the phosphorylation and degradation of the inhibitory factor, IB (12,25,28,29). Activated NF-B induces the expression of multiple proteins that protect against cell death, including antiapoptotic proteins of the Bcl-2 family (Bcl-2 and Bcl-XL) and inhibitors of caspases such as FLIP and XIAP (X-linked inhibitor of apoptosis protein) (31,32). Akt has been shown to be necessary for the activation of the p65 subunit of NFB (28). Indeed, PTEN has been shown to block TNF-stimulated NF-B-dependent gene transcription and sensitize cells to TNF-induced cell killing (33). Interestingly, in the present studies, TNF was unable to activate NF-B in ethanol-exposed cells. However, suppression of the elevated PTEN levels restored TNF-dependent NF-B activation, suggesting that the increase of PTEN levels may be a key mechanism through which ethanol inhibits the protective pathways in cells activated by TNF.
Besides being critical for protective mechanisms, Akt is also necessary to restrain signaling pathways that cause cell damage. The stress kinase p38 is activated by a number of noxious insults, including TNF. TNF treatment of ethanol-exposed cells causes a more sustained and greater increase of p38 activity than that seen in control cells not exposed to ethanol. Inhibition of the increased p38 response prevents the TNF-provoked cytotoxicity in ethanol-exposed cells (20). Similarly, inhibition of p38 has been shown to rescue cells from demise in other instances of toxic insult (34 -40). The increased p38 activity caused by TNF in ethanol-exposed cells is associated with the translocation of Bax from the cytosol to the mitochondria, with the inhibition of p38 activity preventing the TNF-induced Bax translocation and the resulting mitochondrial injury. Importantly, Akt has been shown to restrain p38 activity (41)(42)(43).
The adenoviral early region 1A protein mediates sensitization to different cell death-inducing stimuli including TNF (43). E1A expression was shown to result in a down-regulation of Akt activity and increase of p38 activity, inhibition of which prevents sensitization by E1A to TNF. Similarly, in the present study, the restoration of TNF-initiated Akt activation by repression of PTEN levels in ethanol-exposed cells diminished the p38 response to TNF, resulting in an inhibition of the translocation of Bax to the mitochondria and its attendant damage. In addition, constitutively active Akt that is unresponsive to PTEN inhibition prevented the magnified p38 activation in ethanol-exposed cells treated with TNF, thereby also preserving mitochondrial integrity and cell viability. Interestingly, the only parameter unaltered by the expression of PTEN was the activation of caspase-8. This observation suggests that the activation of caspase-8 by TNF in and of itself does not cause cell death in ethanol-exposed hepatocytes.
The increased p38 activity seen in ethanol-exposed cells is associated with a decreased phosphorylation of ASK-1 by Akt. ASK-1 is a MAPK kinase kinase that can activate the c-Jun N-terminal kinase and p38 MAPK cascades (44). By phosphorylating ASK-1 on serine 83, Akt negatively regulates ASK-1 activity (45). In control cells not exposed to ethanol, TNF treatment induced an increased phosphorylation of ASK-1 on serine 83. By contrast, TNF was unable to induce ASK-1 phosphorylation in ethanol-exposed cells, a feature reversed by suppression of PTEN levels and expression of constitutively active Akt. These results suggest that the elevated PTEN level found in ethanol-treated cells is responsible for preventing the negative regulation of ASK-1 by Akt. The importance of ASK-1 in instigating the augmented p38 response to TNF during ethanol treatment is shown by the ability of siRNA targeting ASK-1 to reduce p38 activity in ethanol-exposed cells treated with TNF.
The response of p38 to TNF treatment of ethanol-exposed cells is biphasic. In the first phase, there is a modest increase of p38 activity, starting at 15 min and regressing by 45 min. By contrast, the second phase of the p38 response to TNF is seen only in ethanol-exposed cells, starting at 60 min and being maintained at levels 2-fold above control through 4 h posttreatment. It is this second phase of p38 activation that appears to be critical in mediating increased cell sensitivity to TNF, since the first phase is also observed in control cells not exposed to ethanol and resistant to TNF-induced cytotoxicity. We have shown previously that the first phase of p38 activation is dependent on the formation of reactive oxygen species but that the second is not (20). Importantly, antioxidants delayed but did not prevent the increased cell killing seen in ethanolexposed cells treated with TNF. In that study, such results were seen in both HepG2E47 and primary hepatocytes isolated from ethanol-fed rats.
Interestingly, elimination of ASK-1 expression did not prevent the first phase of p38 activity but did prevent the second phase seen only in ethanol-exposed cells. The increased p38 activity evoked by TNF treatment can be partly explained by the inability of Akt to phosphorylate and inhibit ASK-1 due to the ethanol-induced elevation of PTEN. However, this does not account for how ASK-1 is stimulated to begin with and whether ethanol exposure has any influence on this aspect of ASK-1 control. ASK-1 binds to thioredoxin and is regulated by reactive oxygen species (46). However, ASK-1 can also be controlled by reactive oxygen species-independent mechanisms such as through the formation of ceramide that is triggered by TNF (47). Additionally, the question remains as to how ethanol exposure activates the PTEN promoter as was shown in the present study (Fig. 1D). Unfortunately, little is known about the transcriptional elements in the PTEN promoter. However, it has recently been shown that the transcription factors p53 and Erg-1 (early growth response gene) can up-regulate PTEN expression (48 -50). Both of these transcription factors possess growth-suppressing activities and are triggered by cellular stresses. It is possible that ethanol can impose a stress on the cell that can lead to activation of such or similar transcription factors through as yet uncharacterized processes. It must also be acknowledged that the elevated PTEN levels induced by ethanol exposure may be due to a diminution of degradation of the PTEN mRNA. The resulting up-regulation of PTEN levels brings about the observed alterations to the p38 and Akt signaling pathways that elicit a sensitization to TNF cytotoxicity (Fig. 9).
FIG. 9. Alterations to TNF-initiated signaling pathways and cytotoxicity imposed by the elevated PTEN levels induced by ethanol exposure. In normal non-ethanol-exposed cells, the TNF receptor activates a survival pathway mediated by NF-B and Akt activation. Ethanol exposure increases PTEN levels, causing an inhibition of Akt activation by TNF. The resulting suppression by PTEN of Akt activity results in a blunting of the survival pathway mediated by NF-B and a potentiation of ASK-1 activation, resulting in an augmented p38 response. In turn, p38 promotes the translocation of Bax to the mitochondria with subsequent induction of mitochondrial injury and cell death.