Selective Glutathione Depletion of Mitochondria by Ethanol Sensitizes Hepatocytes to Tumor Necrosis Factor

Transcription

1 GASTROENTEROLOGY 1998;115: Selective Glutathione Depletion of Mitochondria by Ethanol Sensitizes Hepatocytes to Tumor Necrosis Factor ANNA COLELL,* CARMEN GARCÍA RUIZ,* MERCE MIRANDA,* ESTHER ARDITE,* MONTSE MARÍ,* ALBERT MORALES,* FERNANDO CORRALES, NEIL KAPLOWITZ, and JOSÉ C. FERNÁNDEZ CHECA* *Liver Unit, Department of Medicine, Hospital Clinic i Provincial and Instituto Investigaciones Biomédicas, Consejo Superior Investigaciones Cientificas, Barcelona, Spain; Division of Hepatology and Gene Therapy, Department of Medicine, University of Navarra, Pamplona, Spain; and Center for Liver Disease Research, University of Southern California School of Medicine, Los Angeles, California Background & Aims: Tumor necrosis factor (TNF)- induces cell injury by generating oxidative stress from mitochondria. The purpose of this study was to determine the effect of ethanol on the sensitization of hepatocytes to TNF-. Methods: Cultured hepatocytes from ethanol-fed (ethanol hepatocytes) or pair-fed (control hepatocytes) rats were exposed to TNF-, and the extent of oxidative stress, gene expression, and viability were evaluated. Results: Ethanol hepatocytes, which develop a selective deficiency of mitochondrial glutathione (mgsh), showed marked susceptibility to TNF-. The susceptibility to TNF-, manifested as necrosis rather than apoptosis, was accompanied by a progressive increase in hydrogen peroxide that correlated inversely with cell survival. Nuclear factor B activation by TNF- was significantly greater in ethanol hepatocytes than in control hepatocytes, an effect paralleled by the expression of cytokine-induced neutrophil chemoattractant. Similar sensitization of normal hepatocytes to TNF- was obtained by depleting the mitochondrial pool of GSH with 3-hydroxyl-4- pentenoate. Restoration of mgsh by S-adenosyl-Lmethionine or by GSH ethyl ester prevented the increased susceptibility of ethanol hepatocytes to TNF-. Conclusions: These results indicate that mgsh controls the fate of hepatocytes in response to TNF-. Its depletion caused by alcohol consumption amplifies the power of TNF- to generate reactive oxygen species, compromising mitochondrial and cellular functions that culminate in cell death. The pathogenesis of alcoholic liver disease (ALD) has not been fully characterized. In addition to several factors 1 7 that have been shown to contribute to the development of this illness, such as peroxidation of membrane lipids and the oxidative stress caused by the oxidative metabolism of alcohol, previous studies have reported impairment in the methionine adenosyl transferase (MAT) activity and subsequent depletion in S- adenosyl-l-methionine (SAM) levels. 8,9 Furthermore, oral administration of SAM to patients with ALD has been shown to normalize hepatic SAM concentration, restoring the MAT activity, through increased hepatic glutathione (GSH) levels. 9 Participation of inflammatory cytokines, such as tumor necrosis factor (TNF- ), has been proposed to play a critical role in the development of ALD. Circulating levels of TNF- and other inflammatory cytokines are increased in patients with acute alcoholic hepatitis and chronic ALD; furthermore, TNF- is overexpressed in the intragastric infusion model of ALD. 1,10,11 Among the several mechanisms underlying the toxic effects of TNF-, overproduction of reactive oxygen species (ROS) appears to play a key role in mediating TNF- induced cytotoxicity and gene regulation; these ROS can lead to cell death subsequent to the alteration of critical macromolecules such as DNA, lipids, and proteins Previous studies have provided evidence that ROS produced by TNF- originated from mitochondria because of the interaction of ceramide, a signaling molecule, with specific components of this subcellular compartment, highlighting the importance of this organelle in mediating the toxic effects of the cytokine. GSH in mitochondria comprises a minor fraction of the total cellular pool (10% 15%) and is determined by the activity of a specific carrier located in the inner membrane that transports GSH from cytosol into the mitochondrial matrix Previous studies have shown that chronic ethanol exposure resulted in a selective decrease ( 45% 60%) in the mitochondrial pool of GSH because of impaired uptake of GSH from cytosol Abbreviations used in this paper: ALD, alcoholic liver disease; BHT, butylated hydroxytoluene; CINC, cytokine-induced neutrophil chemoattractant; DCFDA, 2,7 -dichlorofluorescindiacetate; DCF, 2,7 dichlorofluorescein diacetate; GSH, glutathione; GSH-EE, GSH ethyl ester; GST, glutathione S-transferase; HP, (R,S)-3-hydroxy-4- pentenoate; MAT, methionine adenosyl transferase; NAC, N- acetylcysteine; NF- B, nuclear factor B; ROS, reactive oxygen species; SAH, S-adenosyl-homocysteine; SAM, S-adenosyl-L-methionine; TNF-, tumor necrosis factor by the American Gastroenterological Association /98/$3.00

2 1542 COLELL ET AL. GASTROENTEROLOGY Vol. 115, No. 6 into the matrix Such a defect, which occurs preferentially in the mitochondria of the perivenous population of liver cells, 26,27 causes greater susceptibility of alcohol-fed hepatocytes to lethal oxidative stress induced by exogenous oxidants, suggesting that this pool of GSH plays a key contributory role in the early manifestations of ethanol-induced cell injury. 28 In addition to the ability of SAM treatment to attenuate alcohol-induced liver injury, 8,9,29 its administration to long-term ethanol-fed rats has been shown to prevent depletion of GSH in mitochondria by maintaining the appropriate fluidity of the inner mitochondrial membrane. 27,30 The effect of TNF- in hepatocytes from long-term ethanol-fed rats has not been reported. Because mitochondrial GSH is crucial for maintenance of a critical functional organelle and cell viability, and because longterm ethanol consumption selectively depletes this pool of cellular GSH, the purpose of this study was to determine the role of mitochondrial GSH and the effect of SAM on the susceptibility of hepatocytes from longterm ethanol-fed rats to the cytotoxicity of TNF-. Materials and Methods Materials and Animal Model GSH ethyl ester (GSH-EE), butylated hydroxytoluene (BHT), vitamin E, N-acetylcysteine (NAC), and Hoechst were obtained from Sigma Chemical Co. 2,7 -dichlorofluorescindiacetate (DCFDA) was from Molecular Probes (Eugene, OR). SAM was a gift from Europharma, S. A. (Madrid, Spain). (R,S)-3-hydroxy-4-pentenoate (HP) was a generous gift from Dr. W. Anders (University of Rochester, NY). Male Sprague Dawley rats (Pan Lab, Barcelona, Spain) were fed the Lieber DeCarli liquid diet, in which ethanol composed 36% of total calories; pair-fed animals received the same diet except that carbohydrates isocalorically replaced ethanol SAM and NAC were administered at 0.4 mg/ml and mg/ml, respectively, in the liquid diet, as previously described. 27,30 Animals were pair-fed for up to 4 weeks. Experimental protocols met the guidelines of the Animal Care Committee of the Hospital Clinic Universidad de Barcelona, Barcelona, Spain. Preparation of Mitochondria, Functional Integrity, and Determination of GSH Rat liver mitochondria were isolated from liver homogenates by differential centrifugation 19 or by rapid centrifugation through Percoll density gradient as described in detail previously Enrichment and recovery of mitochondria, ascertained by the specific activity of succinic dehydrogenase as described previously, 26,27 was fold vs fold and 85% 5% vs. 92% 7% for pair-fed and ethanol-fed rats, respectively. Mitochondrial purity was confirmed by estimating the contamination with other subcellular organelles such as sinusoidal and canalicular plasma membrane, microsomes, and lysosomes assessed by the activity of Na,K - adenosine triphosphatase (ATPase), Mg 2 -ATPase, G-6-Pphosphatase, and acid phosphatase, respectively. 29 Mitochondrial integrity was determined by the acceptor control ratio as the adenosine diphosphate (ADP)-stimulated oxygen consumption over its absence as described previously. 26,27 GSH was measured from mitochondria and cytosol fractions after trichloroacetic acid precipitation by high-performance liquid chromatography as described previously Levels of GSH in mitochondria in individual preparations were corrected by the recovery of mitochondrial fraction. Isolation and Culture of Rat Hepatocytes Rat hepatocytes from control and ethanol-fed rats were isolated by collagenase as described previously. 26,27 Cell viability was determined by trypan blue exclusion and by the release of cellular enzymatic activities such as lactate dehydrogenase or GSH S-transferases (GST). Recombinant human TNF- (Promega; sp act, U/mg protein) was added to serum-deprived cells at a dose of ,000 U/mL for 12 hours in cell culture or for 4 hours in cell suspension studies. In some cases, normal hepatocytes were incubated with HP at 1 mmol/l for 5 minutes followed by washing of cells. Hepatocytes were incubated with antioxidants, BHT (dimethyl sulfoxide), or vitamin E (vegetable oil), 50 µmol/l for 30 minutes, before the addition of TNF-. Hepatocytes from pair- and ethanol-fed rats were preincubated with GSH-EE (2 mmol/l for minutes) to increase the mitochondrial pool of GSH. Cells were fractionated to obtain mitochondria and cytosol as described above using Percoll gradient centrifugation. Determination of MAT Activity and SAM Levels MAT activity and levels of SAM and S-adenosylhomocysteine (SAH) from extracts of hepatocytes from pair-fed and ethanol-fed rats were determined as previously described. 8,29 Determination of Hydrogen Peroxide Production of ROS, mainly hydrogen peroxide and other organic peroxides, 19,31 was monitored spectrofluorometrically by DCFDA. Before cell harvesting (1 hour), DCFDA (2 µg/ml) was added to culture plates, followed by washing and determination of 2,7 -dichlorofluorescein (DCF) fluorescence at 529 nm/503 nm (emission/excitation). Preparation of Nuclear Extracts and Electrophoretic Mobility Shift Assay for Nuclear Factor B Preparation of nuclear extracts and assay of nuclear factor B (NF- B) activity using a 32 P-labeled B oligonucleotide (5 -AGTTGAGGGGACTTTCCCAGGC-3 ) were done as described previously. 31 To test binding specificity, a mutated oligonucleotide probe (5 -AGTTGAATTCACTTTCCCAGGC- 3 ) was used. Probes were labeled at the 5 end with T4 kinase and [ - 32 P]adenosine triphosphate (ATP; 3,000 Ci/mmol).

3 December 1998 MITOCHONDRIAL GSH AND TNF CYTOTOXICITY 1543 Proteins were separated through native 6% polyacrylamide gel electrophoresis and visualized by autoradiography. In some cases, supershift assays were performed by incubation of nuclear extracts with antibodies to different subunits of the NF- B/Rel protein family generously provided by Dr. Rice (Bethesda, MD). 32 Analysis of Cytokine-Induced Neutrophil Chemoattractant mrna Levels A complementary DNA (cdna) probe for cytokineinduced neutrophil chemoattractant (CINC) was prepared by reverse-transcription polymerase chain reaction from messenger RNA (mrna) using the following specific primers: 5 -TGAGCTGCGCTGTCAGTCAGTGCCTGCAGA and 5 - TTACTTGGGGACACCTTTTAGCATCTTTTGGACA, corresponding to nucleotides 87 to 112 and 254 to 292 of a previously reported published sequence. 33,34 Total cellular RNA was extracted, denatured, filtered under vacuum on nylon membrane, and fixed with UV light. The membranes were prehybridized at 65 C, and hydridizations were performed using the 32 P-labeled CINC cdna. The level of CINC mrna was calculated in relationship to the 18S band used as housekeeping reference and expressed as a percentage of control. Cytochemical Staining of Apoptotic Cells Morphological changes in the nuclear chromatin of cells undergoing cell death were determined by staining with the DNA-binding fluorchrome Hoechst Cultured hepatocytes ( ) from pair- and ethanol-fed rats were incubated with TNF- for 12 hours. Before staining, cells were trypsinized and resuspended for fixation in 200 µl of 3% of paraformaldehyde in phosphate-buffered saline (PBS; ph 7.8) at 4 C for 10 minutes. After washing with PBS, cells were stained with Hoescht (8 µg/ml) for 15 minutes at room temperature. Cells were observed under an Olympus fluorescence microscope, using nm filter for excitation and 460 nm for emission. Incidence of apoptotic chromatin changes include condensation of chromatin, its margination along the periphery of the nucleus, and segmentation of the nucleus into more than 3 fragments. Determination of Ceramide Levels Levels of ceramide produced in hepatocytes by stimulation with TNF- were quantified by use of the diacylglycerol kinase assay as described previously. 19,35 Ceramide 1-phosphate was resolved by thin-layer chromatography on silica gel 60 plates (Whatman) using chloroform:methanol:acetic acid (65: 15:5) and detected by autoradiography; incorporated 32 P was measured by liquid scintillation counting. Statistical Analyses Statistical analyses of mean values for multiple comparisons between hepatocyte preparations from different groups were made by one-way analysis of variance followed by the Fisher exact test. Results Long-term Ethanol Intake Sensitizes Hepatocytes to TNF- Cytotoxicity Because one of the major mechanisms in TNFinduced cytotoxicity is the mitochondrial generation of ROS, 15,36 we used a model of specific mitochondrial GSH depletion to investigate the role of ROS in conferring sensitivity of hepatocytes to TNF-. As shown in Figure 1, freshly isolated hepatocytes from long-term ethanolfed rats showed the reported depletion in concentration of GSH in mitochondria with sparing of cytosol GSH. Cultured cells retained the same selective depletion of mitochondrial GSH as freshly isolated cells. Cells from control rats survived TNF- exposure, but the viability of hepatocytes from ethanol-fed rats decreased progressively with increasing doses of TNF- (Figure 2). At the highest dose of 10,000 U/mL (370 ng/ml), 20% 30% of hepatocytes from ethanol-treated rats survived, compared with 80% 90% of cells from control rats. Parallel aliquots were taken to determine the burst of ROS induced by TNF-. As shown in Figure 2B, the level of DCF increased with TNF- in a dose-dependent fashion only in hepatocytes from long-term ethanol-fed rats. Indeed, ROS generation and loss of viability correlated inversely, suggesting that the latter was a consequence of the former (Figure 3). Time-dependent studies showed that the increase in DCF in long-term ethanol-fed hepatocytes induced by TNF- (10,000 U/ml) occurred by 2 3 hours before the onset of toxicity. To show that the TNF- induced cytotoxicity in ethanol-fed hepatocytes was dependent on ROS generation, we tested the effect of antioxidants. As seen in Figure 1. Compartmentation of cellular GSH in hepatocytes from pair-fed and ethanol-fed rats. Freshly isolated hepatocytes from pairand ethanol-fed rats with or without SAM supplementation in the liquid diet were fractionated as indicated in Materials and Methods to obtain the cytosol and mitochondria. Total GSH in these fractions was determined as described in Materials and Methods. Results are means SD of pair-fed (n 5) and ethanol-fed (n 6) rats without SAM and for each group with SAM supplementation. *P 0.05 vs. pair-fed without SAM; **P 0.05 vs. ethanol-fed without SAM.

4 1544 COLELL ET AL. GASTROENTEROLOGY Vol. 115, No. 6 Figure 2. Survival and hydrogen peroxide generation in hepatocytes exposed to TNF- and effect of SAM supplementation. Hepatocytes isolated from pair-fed ( ) or ethanol-fed ( ) rats with or without SAM supplementation were cultured overnight., Pair-fed SAM;, ethanol-fed SAM. Cells were then incubated with increasing concentrations of TNF- (250 10,000 U/mL; ng/ml) for 12 hours. (A) Cell viability was determined by trypan blue exclusion and GST activity released in extracellular medium. (B) Parallel culture dishes were incubated with DCFDA to determine DCF fluorescence to monitor the levels of hydrogen peroxide generation as described in Materials and Methods. Results are means SD of pair-fed (n 5) and ethanol-fed (n 6) rats without SAM and for each group with SAM supplementation. Data from the ethanol-fed group were significantly different from pair-fed control and ethanol-fed SAM-supplemented values. Figure 4, BHT and vitamin E prevented both oxidative stress (DCF fluorescence) and cytotoxicity of TNF- in hepatocytes from ethanol-fed rats. Long-term alcohol ingestion has been shown to impair MAT activity, leading to decreased SAM levels 8,9,29 ; therefore, we determined MAT activity in hepatocytes Figure 3. Relationship of survival and hydrogen peroxide generation in hepatocytes exposed to TNF-. Survival and DCF fluorescence from the indicated groups of rats, determined as described in Figure 2, were plotted, and linear regression was performed by computer analyses. Survival and DCF data for the pair-fed group with and without SAM administration were similar and hence were pooled. Slopes were , , and for pair-fed ( ), ethanol and SAM supplementation ( ), and ethanol-fed ( ), respectively. Results are means SD for pair-fed (n 5) and ethanol-fed (n 6) rats without SAM and for each group with SAM supplementation. from rats fed ethanol. MAT activity and SAM/SAH were similar in hepatocytes from pair-fed and long-term ethanol-fed rats (Table 1). Because TNF- has been shown to induce both apoptosis and necrosis, 15 17,37 39 we determined the type of death induced by TNF- in hepatocytes from longterm ethanol-fed rats. TNF- induced necrosis, as opposed to apoptosis, was judged by examination for typical features of apoptosis. Nonviable hepatocytes did not show signs of apoptosis, such as chromatin disruption and DNA fragmentation (Figure 5). Effect of SAM Administration on the Susceptibility of Hepatocytes From Long-term Ethanol-Fed Rats to TNF- We next examined the effect of SAM, which prevents ethanol-induced depletion of the mitochondrial pool of GSH, on TNF- induced cytotoxicity in hepatocytes from long-term ethanol-fed rats. As shown in Figure 1, administration of SAM to long-term ethanolfed rats resulted in a significant increment of both cytosol and mitochondrial pool of GSH, consistent with previous reports. 27,30 The levels of mitochondrial GSH of ethanolfed rats supplemented with SAM were similar to those of pair-fed controls. The survival and generation of hydrogen peroxide of hepatocytes from ethanol-fed rats with SAM supplementation were similar to those of cells from pair-fed rats (Figures 2 and 3). The levels of ceramide generated in response to TNF- in hepatocytes from ethanol-fed rats were similar regardless of SAM administration ( and

5 December 1998 MITOCHONDRIAL GSH AND TNF CYTOTOXICITY 1545 Figure 4. Effect of antioxidants on TNF-induced oxidative stress. Freshly isolated hepatocytes from pair-fed and ethanol-fed rats were incubated with 50 µmol/l BHT and 50 µmol/l vitamin E or vehicle, dimethyl sulfoxide, or vegetable oil for 30 minutes and then exposed to 10,000 U/mL TNF- for the indicated time in Fisher s medium in the continued presence of antioxidants. Aliquots of cells were taken to determine DCF fluorescence, and cell survival was assessed by trypan blue exclusion. DCF and viability of pair-fed cells were unchanged with TNF- exposure in the presence of BHT and vitamin E. Results are means SD of 3 cell preparations for each group. *P 0.05 vs. pair-fed control and ethanol-fed plus BHT and vitamin E. nmol/10 6 cells, with or without SAM supplementation, vs nmol/10 6 cells for untreated cells). These levels did not differ from values of cells from pair-fed rats after exposure to TNF- ( nmol/10 6 cells vs nmol/10 6 cells for untreated cells). Supplementation of ethanol-fed rats with the GSH precursor NAC was ineffective in restoring the mitochondrial pool of GSH and preventing the sensitization of hepatocytes to the cytotoxicity to TNF- (Table 2). However, under these circumstances, NAC resulted in a marked increase in cytosol GSH levels but did not increase mitochondrial GSH levels, strongly suggesting that mitochondrial GSH controls ROS overproduction and subsequent cell survival. Table 1. MAT Activity and SAM/SAH Levels in Hepatocytes From Pair- and Ethanol-Fed Rats MAT SAM/SAH Pair-fed Ethanol-fed NOTE. Total extracts of hepatocytes from pair-fed and ethanol-fed rats were used for determination of MAT and SAM/SAH. The values of MAT activity are expressed in pmol min 1 mg protein 1. SAM levels were and nmol/10 6 cells for pair- and ethanol-fed hepatocytes, respectively. Results are means SD of 3 cell preparations for pair- and ethanol-fed rats. Restoration of Mitochondrial GSH Protects Hepatocytes From Long-term Ethanol-Fed Rats to TNF- To examine the role and significance of mitochondrial GSH in regulating the cell death induced by TNF-, we used a permeable form of GSH, GSH-EE, that diffuses readily into cells and intracellular compartments. GSH-EE displayed an increased level of mitochondrial GSH not different from that of pair-fed controls (Figure 6), whereas the cytosol pool of GSH increased by 20% 30% above basal levels. In this paradigm, hepatocytes from ethanol-fed rats were resistant to TNF- (Figure 7A). This resistance to killing of hepatocytes was paralleled by a reduction in DCF fluorescence, indicating attenuation of ROS (Figure 7B). Because GSH-EE generates ethanol, we tested the effect of an equimolar dose of ethanol on the survival of hepatocytes from long-term ethanol-fed rats to TNF-. As shown in Figure 7, ethanol (2 mmol/l) did not modulate susceptibility to TNF-. These findings indicate that restoration of mitochondrial GSH pool size, irrespective of the strategy used, determines a resistance to the toxic effects of TNF-. Activation of NF- B by TNF- in Hepatocytes From Long-term Ethanol-Fed Rats Because TNF- activates the transcription factor NF- B, we examined its pattern of activation in hepatocytes from control and ethanol-fed rats. As shown in Figure 8, the magnitude of activation of NF- B by TNF- was greater in nuclear extracts of hepatocytes from ethanol-fed rats. Activation of NF- B by TNF- was specific as an excess of unlabeled oligonucleotide displaced the retarded band (not shown). Furthermore, to determine the composition of the activated NF- B, supershift assays were performed by incubation of nuclear extracts with antibodies against p50 and p65, showing that both antibodies resulted in a further shifted band (Figure 8B).

6 1546 COLELL ET AL. GASTROENTEROLOGY Vol. 115, No. 6 Figure 5. Morphological appearance and DNA integrity of hepatocytes incubated with TNF-. (A) Cells from pair-fed and ethanol-fed rats were cultured and incubated with 10,000 U/mL TNF- (a and b, respectively) for 12 hours and then exposed to Hoechst and observed under a fluorescence microscope to determine chromatin appearance. (B) Parallel aliquots were taken to estimate DNA fragmentation (Mrk., DNA size markers). The positive control for apoptosis (c) shows the appearance of cultured hepatocytes whose total GSH levels were depleted by preincubation with diethyl maleate (0.8 mmol/l for 15 minutes followed by washing to remove excess diethyl maleate) before treatment of cells with 10,000 U/mL TNF- and stained with the same fluorochrome. cn, condensed nuclei; fn, fragmented nuclei. (Original magnification 200.) We next examined the pattern of NF- B activation by TNF- in hepatocytes from ethanol-fed rats supplemented with SAM. As shown in Figure 9, although there was activation of the p50 homodimer, the magnitude of activation by TNF- of the heterodimer p65/p50 was significantly attenuated compared with cells from rats fed ethanol without SAM administration (Figure 8). After activation of the transcriptional active p65/p50 heterodimer, expression of B-dependent genes should ensue. We examined the expression of the cytokine CINC in hepatocytes from ethanol-fed rats with and without SAM supplementation in response to TNF-. The pattern of CINC expression by TNF- treatment in hepatocytes from ethanol-fed rats was paralleled by the activa- Table 2. Effect of NAC on GSH Compartmentation and TNF- Cytotoxicity of Ethanol-Fed Hepatocytes Cytosol GSH (nmol/10 6 cells) Mitochondrial GSH (nmol/10 6 cells) Viability Pair-fed Ethanol-fed a 53 7 a Pair-fed NAC b Ethanol-fed NAC b a 55 5 a NOTE. Hepatocytes from pair- and ethanol-fed rats with and without NAC supplementation were isolated and fractionated into cytosol and mitochondria. Cells were then exposed to TNF- (10,000 U/mL), and viability was examined after 4 hours of incubation. Results are means SD of 4 cell preparations for each group of rats. a P 0.05 vs. pair-fed cells. b P 0.05 vs. the corresponding group in the absence of NAC. Figure 6. Restoration of mitochondrial pool of GSH by GSH-EE. Freshly isolated rat hepatocytes from pair-fed and ethanol-fed rats were fractionated into cytosol and mitochondria for GSH determination as described for Figure 1. Parallel cell suspensions were incubated with GSH-EE at 2 mmol/l for minutes and then washed to remove excess ester. Cells were subsequently fractionated to isolate cytosol and mitochondrial fraction by Percoll centrifugation as described in Materials and Methods, and GSH in each compartment was measured. Results are means SD of 3 and 4 cell preparations from pair-fed and ethanol-fed rats, respectively. *P 0.05 vs. pair-fed in the absence of GSH-EE; **P 0.05 vs. ethanol-fed without GSH-EE; # P 0.05 vs. pair-fed.

7 December 1998 MITOCHONDRIAL GSH AND TNF CYTOTOXICITY 1547 Figure 7. (A) Survival and (B) hydrogen peroxide generation induced by TNF- and effect of GSH-EE. Hepatocytes from pair-fed and ethanol-fed rats were incubated in the presence of 10,000 U/mL TNF- or its vehicle for 4 hours in Fisher s medium. Cell suspensions for each group were incubated with GSH-EE (2 mmol/l) to increase the cellular GSH levels or ethanol (2 mmol/l). Compartmentation of cytosol and mitochondrial GSH after incubation with GSH-EE was determined as in Figure 6. At the end of the incubation period, viability and DCF fluorescence were determined as described in Materials and Methods. Results are means SD of 3 and 4 cell preparations from pair-fed and ethanol-fed rats, respectively. *P 0.05 vs. pair-fed and ethanol-fed groups in the absence of TNF- ; **P 0.05 vs. ethanolfed group in the presence of TNF-. tion of NF- B, with a greater expression in cells from ethanol-fed rats (Figures 8 and 9). SAM administration to ethanol-fed rats decreased CINC expression in response to TNF-. Effect of HP on Susceptibility of Normal Hepatocytes to TNF- To validate the results obtained with hepatocytes from long-term ethanol-fed rats suggesting that GSH in mitochondria controls the survival in response to TNF-, we used HP, an agent reported to deplete the mitochondrial pool of GSH with minimal depletion of cytosol GSH. 40 As shown in Figure 10, the mitochondrial pool of GSH was depleted by 70% in hepatocytes incubated with HP, compared with the modest depletion (20% 25%) of cytosol GSH. Subsequent exposure of these cells to TNF- resulted in increased generation of hydrogen peroxide, which preceded the onset of cytotoxicity. Similar results were observed when HP-treated hepatocytes were exposed to acidic sphingomyelinase instead of TNF- (not shown). Discussion The present study was undertaken to assess the role of mitochondrial GSH in modulation of the oxidative stress and survival of hepatocytes from long-term ethanolfed rats in response to TNF-. Our study provides evidence for the first time that ethanol intake sensitizes rat hepatocytes to the toxic effects of TNF-. The ability of added antioxidants to attenuate the TNF- induced cell injury supports the theory that the cytotoxicity of TNF- to hepatocytes is dependent on ROS generation, in agreement with previous reports. 17 Our data point to the limitation of the mitochondrial pool of GSH caused by ethanol intake as a vital determinant of this response, because this antioxidant acts as the only defense available in mitochondria against hydrogen peroxide. In support for this, cytosol GSH levels remained unchanged in hepatocytes from control and ethanol-fed rats, yet these cells showed a marked differential sensitivity to TNF- cytotoxicity. To further document that a lower mitochondrial GSH pool size is a critical factor in controlling the survival of cells in response to TNF-, different approaches to restoration of the mitochondrial pool of GSH in hepatocytes from ethanol-fed rats were used, and the resulting survival was examined in response to TNF- exposure. Supplementation of ethanol-fed rats with SAM, which resulted in almost complete normalization of the mitochondrial pool of GSH, prevented the susceptibility of hepatocytes from ethanol-fed rats to TNF-. Furthermore, these findings indicate that the increased susceptibility to TNF- seen in cells from ethanol-fed rats was not caused by culture conditions of hepatocytes, nor was it a consequence of ethanol withdrawal. Supplementation of ethanol-fed rats with NAC, a precursor of GSH, failed to restore the mitochondrial pool of GSH and was ineffective in protecting these cells from TNF-, yet this treatment resulted in a significant increase in the cytosol pool of GSH. These findings provide additional support for the vital role of the mitochondrial pool of GSH, as opposed to cytosol GSH, in the control of processes that culminate in cell death, and suggest that the ability of TNF- to kill hepatocytes depends largely on the availability of the mitochondrial pool of GSH. In addition, the beneficial effect of SAM supplementation in protecting hepatocytes of ethanol-fed rats against TNF- was not attributable to interference of SAM with ethanol metabolism or TNF- signaling because the blood ethanol levels 27,29 and magnitude of ceramide generation in response to TNF- were unchanged by SAM treatment. Our findings show that neither MAT deficiency nor SAM depletion appears to be involved in the susceptibility of hepatocytes from long-term ethanol-fed rats to TNF- and, consequently, that the beneficial effect afforded by SAM supplementation is not related to the ability of exogenous SAM to restore MAT activity, as

8 1548 COLELL ET AL. GASTROENTEROLOGY Vol. 115, No. 6 Figure 8. TNF-induced NF- B activation and expression of CINC in hepatocytes. Hepatocytes from each group were cultured and exposed to the indicated dose of TNF- overnight. (A) Nuclear extracts were isolated and analyzed by electrophoretic mobility shift assay to detect NF- B activation. (B) Parallel nuclear extracts were incubated with the indicated antibodies against p65 and p50 of NF- B/rel members and analyzed by electrophoretic mobility shift assay for detection of supershifted bands denoted as SS. (C) Total RNA was isolated and hybridized with CINC cdna, and its levels were expressed in relationship to 18 S used as reference control. Expression of CINC was quantitated by densitometry. Results for CINC expression are means SD of 3 and 4 cell preparations from pair-fed and ethanol-fed rats, respectively. *P 0.05 vs. pair-fed cells without TNF; **P 0.05 vs. corresponding pair-fed hepatocytes incubated with TNF-. Figure 9. Effect of SAM supplementation on NF- B activation and CINC expression induced by TNF-. Hepatocytes from pair-fed and ethanol-fed rats supplemented with SAM during feeding were cultured and treated with TNF- as in Figure 8. (A) Nuclear extracts were then examined for activation of p50/p65 heterodimer and p50 homodimers of NF- B. (B) Total RNA was isolated and hybridized with CINC cdna, (C) and its levels were expressed in relationship to 18 S used as reference control. Expression of CINC was quantitated by densitometry. Results for CINC expression are means SD of 3 and 4 cell preparations from pair-fed and ethanol-fed rats, respectively.

9 December 1998 MITOCHONDRIAL GSH AND TNF CYTOTOXICITY 1549 Figure 10. Sensitization of normal hepatocytes by HP to TNF-. Freshly isolated normal hepatocytes were incubated with HP (1 mmol/l for 5 minutes) and washed thoroughly. Cells were then fractionated into (A) cytosol and (B) mitochondria to determine GSH content in these fractions. Subsequently, cells were exposed to 10,000 U/mL TNF- for the indicated time. Aliquots were taken for determination of (C) DCF fluorescence and (D) viability. Cytosol and mitochondrial GSH in the absence of TNF- did not change over the course of the incubation. Results are means SD of 3 cell preparations. *P 0.05 vs. control cells. reported in patients and experimental animal models of ALD. 8,9,29 However, these results do not challenge the significance of the reported impairment of MAT activity by long-term ethanol abuse, because in most of these reports the impairment in MAT activity was associated with severe GSH depletion. Indeed, as demonstrated by Pajares et al., 41 MAT activity is subject to redox regulation by the intracellular levels of GSH. Our findings in this regard indicate that the levels of cytosol GSH remained unchanged in hepatocytes from long-term ethanol-fed rats, probably being sufficient to maintain an appropriate environment for optimal functioning of MAT activity and maintenance of SAM homeostasis. Thus, at present we do not understand the mechanism for the protective effect of SAM in prevention of altered mitochondrial fluidity and GSH transport. To critically examine the role of mitochondrial GSH in conferring sensitivity to TNF-, the mitochondrial pool of GSH of hepatocytes from ethanol-fed rats was restored by GSH-EE. Similar to the outcome obtained with SAM supplementation, replenishment of the mitochondrial pool of GSH blocked susceptibility to TNF-. In addition, the findings observed with GSH-EE in restoring the mitochondrial GSH and resistance to TNF- in vitro provide direct support for a causal relationship between prevention of mitochondrial GSH depletion by SAM in vivo and susceptibility to TNF-. Thus, although we do not understand how SAM prevents the altered fluidity and GSH sequestration in mitochondria, this outcome of SAM treatment probably explains its restoring the resistance of ethanol-fed hepatocytes to TNF-. To validate our observations in hepatocytes from long-term ethanol-fed rats that limitation of the mitochondrial pool of GSH controls the susceptibility to TNF-, the levels of mitochondrial GSH of normal hepatocytes were depleted using HP. The predominant metabolism of HP within mitochondria by the (R)-3-hydroxybutanoate: NAD oxidoreductase generates a Michael acceptor that reacts with GSH, resulting in its depletion. 40 Findings using this agent show a similar sensitization of normal hepatocytes to TNF-, indicating that the susceptibility of hepatocytes to TNF- by alcohol intake is not an artifact, but rather is causally related to the alcoholinduced limitation of mitochondrial pool of GSH. The amplification of oxidative stress by TNF- in ethanol-fed

10 1550 COLELL ET AL. GASTROENTEROLOGY Vol. 115, No. 6 hepatocytes showing depleted mitochondrial GSH not only was a major factor in cytotoxicity, but also exerted an effect on activation of NF- B and gene expression, which were inhibited by SAM. Although TNF- is known to use several signaling mechanisms to activate NF- B, a major contribution to that effect in ethanol-fed hepatocytes is the enhanced oxidative stress caused by depletion of mitochondrial GSH. An interesting aspect of our findings is the indication that the sensitization of hepatocytes by ethanol occurred despite increased activation of NF- B, which has been shown to protect cells against TNF Although these reports indicate that NF- B protects against apoptosis, it is conceivable that it may be inefficient against necrosis, a distinct form of cell death. Indeed, recent studies showed that NF- B does not afford protection against necrosis induced by hyperoxia. 45 Our data on the type of death of hepatocytes from long-term ethanol-fed rats indicate that these cells died by necrosis as opposed to apoptosis. Such an outcome is consistent with the findings of others who have identified factors that shape the type of death. Recent observations by Leist et al. 46 indicate that the intracellular ATP concentration acts as a molecular switch in determining the type of death. Decreased ATP concentration in hepatocytes is a common effect of ethanol consumption, 4,26 indicating that the lack of apoptosis of these susceptible hepatocytes in response to TNF- may be caused by energy limitation. In summary, our results show a critical role of mitochondrial overproduction of ROS in the cascade of cell death. A defect in transporting GSH from cytosol into mitochondria caused by long-term ethanol intake, resulting in depletion of the mitochondrial pool of GSH, renders ethanol-fed hepatocytes more susceptible to the cytotoxic effects of inflammatory cytokines, e.g., TNF-, by amplifying the generation of ROS within mitochondria. This effect also contributes to enhanced expression of oxidative stress-responsive genes. Thus restoration of mitochondrial GSH by SAM or GSH-EE and scavenging of ROS by antioxidants are potential therapeutic strategies for attenuation of alcohol-induced liver injury. Further studies are needed to fully evaluate the role of defective transport of GSH in mitochondria in patients with ALD, although the present studies and previous reports with animal models indicate that selective mitochondrial GSH depletion precedes the worsening and progression of ALD. References 1. Ashak KG, Zimmerman HJ, Ray MB. Alcoholic liver disease: pathologic, pathogenic and clinical aspects. Alcohol Clin Exp Res 1991;15: Lin RC, Smith RS, Lumeng L. Detection of a protein-acetaldehyde adduct in the liver of rats fed alcohol chronically. J Clin Invest 1988;81: Lieber CS. Metabolic effects of acetaldehyde. Biochem Soc Trans 1988;16: Cunningham CC, Coleman X, Spach V. The effects of chronic ethanol consumption on hepatic mitochondrial energy metabolism. Alcohol Alcohol 1990;25: Fernández-Checa JC, Takeshi H, Tsukamoto H, Kaplowitz N. Mitochondrial glutathione depletion and alcoholic liver disease. Alcohol 1993;10: Cederbaum AI, Rubin E. Molecular injury to mitochondria produced by ethanol and acetaldehyde. Fed Proc 1975;34: Thayer WS, Rubin E. Effects of chronic ethanol intoxication on oxidative phosphorylation in rat liver submitochondrial particles. J Biol Chem 1979;254: Martín Duce A, Ortiz P, Cabrero C, Mato JM. S-Adenosyl-Lmethionine synthetase and phospholipid methyltransferase are inhibited in human cirrhosis. Hepatology 1988;8: Vendemiale G, Altomare E, Trizio T, LeGrazie C, Di Padova C, Salerno MT, Carrieri V. Effects of oral S-adenosyl-L-methionine on hepatic glutathione in patients with liver disease. Scand J Gastroenterol 1989;18: Tsukamoto H, French SW, Benson N. Severe and progressive steatosis and focal necrosis in rat liver induced by continuous intragastric infusion of ethanol and low fat diet. Hepatology 1985;5: Tsukamoto H, Towner SJ, Ciofalo LH, French SW. Ethanol-induced liver fibrosis in rats fed high fat diet. Hepatology 1986;6: Beutler B, Cerami A. Tumor necrosis, cachexia, shock and inflammation: a common mediator. Annu Rev Biochem 1988;57: Fiers W. Tumor necrosis factor: characterization at the molecular, cellular and in vivo level. FEBS Lett 1991;285: Fernández-Checa JC, Kaplowitz N, García-Ruiz C, Colell A, Miranda M, Marí M, Ardite E, Morales A. GSH transport in mitochondria: defense against TNF-induced oxidative stress and alcoholinduced defect. Am J Physiol 1997;273:G7 G Schulze-Osthoff K, Bakker AC, Vanhaesebroeck B, Beyaert R, Jacob WA, Fiers W. Cytotoxic activity of tumor necrosis factor is mediated by early damage to mitochondrial functions. J Biol Chem 1992;267: Schulze-Ostholl K, Beyaert R, Vandevoorde V, Haegeman G, Fiers W. Depletion of the mitochondrial electron transport abrogates the cytotoxic and gene-inductive effect of TNF. EMBO J 1993;12: Goosens V, Grooten J, de Vos J, Fiers W. Direct evidence for tumor necrosis factor induced mitochondrial reactive oxygen intermediates and their involvement in cytotoxicity. Proc Natl Acad Sci USA 1995;92: Adamson GH, Billings RE. Tumor necrosis factor induced oxidative stress in isolated mouse hepatocytes. Arch Biochem Biophys 1992;294: García-Ruiz C, Colell A, Marí M, Morales A, Fernández-Checa JC. Direct effect of ceramide on the mitochondrial electron transport chain leads to generation of reactive oxygen species. Role of mitochondrial glutathione. J Biol Chem 1997;272: Griffith OW, Meister A. Origin and turnover of mitochondrial glutathione. Proc Natl Acad Sci USA 1985;82: Martersson J, Lai JCK, Meister A. High-affinity transport of glutathione is part of a multicomponent system essential for mitochondrial function. Proc Natl Acad Sci USA 1990;87: García-Ruiz C, Morales A, Colell A, Rodés J, Yi J-R, Kaplowitz N, Fernández-Checa JC. Evidence that the rat hepatic mitochondrial

11 December 1998 MITOCHONDRIAL GSH AND TNF CYTOTOXICITY 1551 carrier is distinct from the sinusoidal and canalicular transporters for reduced glutathione. J Biol Chem 1995;270: Fernández-Checa JC, García-Ruiz C, Ookhtens M, Kaplowitz N. Impaired uptake of GSH by hepatic mitochondria from chronic ethanol-fed rats. Tracer kinetic studies in vitro and in vivo and susceptibility to oxidant stress. J Clin Invest 1991;87: Fernández-Checa JC, Ookhtens M, Kaplowitz N. Effect of chronic ethanol feeding on rat hepatocytic GSH: compartmentation, efflux and response to incubation with ethanol. J Clin Invest 1987;80: Fernández-Checa JC, Ookhtens M, Kaplowitz N. Effect of chronic ethanol feeding on rat hepatocytic GSH: relationship of cytosolic GSH to efflux and mitochondrial sequestration. J Clin Invest 1989;83: García-Ruiz C, Morales A, Colell A, Ballesta A, Rodés J, Kaplowitz N, Fernández-Checa JC. Effect of chronic ethanol feeding on glutathione and functional integrity of mitochondria in periportal and perivenous rat hepatocytes. J Clin Invest 1994;94: García-Ruiz C, Morales A, Colell A, Ballesta A, Rodés J, Kaplowitz N, Fernández-Checa JC. Feeding S-adenosyl-L-methionine attenuates both ethanol-induced depletion of mitochondrial glutathione and mitochondrial dysfunction in periportal and perivenous rat hepatocytes. Hepatology 1995;21: Takeshi H, Kaplowitz N, Kamimura T, Tsukamoto H, Fernández- Checa JC. Hepatic mitochondrial GSH depletion and progression of experimental alcoholic liver disease in rats. Hepatology 1992; 16: Lieber CS, Casini A, DeCarli LM, Kim CL, Lowe N, Sasaki R, Leo MA. S-adenosyl-L-methionine attenuates alcohol-induced liver injury in the baboon. Hepatology 1990;11: Colell A, García-Ruiz C, Morales A, Ballesta A, Ookhtens M, Rodés J, Kaplowitz N, Fernández-Checa JC. Transport of reduced glutathione in hepatic mitochondria and mitoplasts from ethanol-fed treated rats: effect of membrane physical properties and S- adenosyl-l-methionine. Hepatology 1997;26: García-Ruiz C, Colell A, Morales A, Kaplowitz N, Fernández-Checa JC. Role of oxidative stress generated from the mitochondrial electron transport chain and mitochondrial glutathione status in loss of mitochondrial function and activation of the transcription factor NF- B: studies with isolated mitochondria and rat hepatocytes. Mol Pharmacol 1995;48: Rice NR, MacKichan ML, Israell A. The precursor of NF- B p50 has I B-like functions. Cell 1992;71: Wu X, Wittwer AJ, Carr LS, Crippes BA, DeLarco JE, Lefkowitz JB. Cytokine-induced neutrophil chemoattractant mediates neutrophil influx in immune complex glomerulonephritis in rat. J Clin Invest 1994;94: Huang S, Paulauskis JD, Kobzik L. Rat KC cdna cloning and mrna expression in lung macrophages and fibroblasts. Biochem Biophys Res Commun 1992;184: Dressler KA, Kolesnick RN. Ceramide 1-phosphate, a novel phospholipid in human leukemia (HL-60) cells: synthesis via ceramide from sphingomyelin. J Biol Chem 1990;265: Morales A, García-Ruiz C, Miranda M, Marí M, Colell A, Ardite E, Fernández-Checa JC. Tumor necrosis factor increases hepatocellular glutathione by transcriptional regulation of the heavy subunit chain of -glutamylcysteine synthetase. J Biol Chem 1997;272: Shinagawa T, Yoshioka K, Kakumu S, Wakita T, Ishikawa T, Itoh Y, Takayanagi M. Apoptosis in cultured rat hepatocytes: the effects of TNF and interferon. J Pathol 1991;165: Leist M, Gantner F, Jilg S, Wendel A. Activation of the 55 kda TNF receptor is necessary and sufficient for TNF-induced liver failure, hepatocyte apoptosis and nitrite release. J Immunol 1995;154: Wang JH, Redmond HP, Watson RWG, Bouchier-Hayes D. Role of lipopolysaccharide and tumor necrosis factor in induction of hepatocyte necrosis. Am J Physiol 1995;269:G297 G Shan X, Jones DP, Hashmi M, Anders MW. Selective depletion of mitochondrial glutathione concentrations by (R,S)-3-hydroxy-4- pentenoate potentiates oxidative cell death. Chem Res Toxicol 1993;6: Pajares AM, Duran C, Corrales F, Pliego MM, Mato JM. Modulation of rat liver S-adenosyl methionine synthetase activity by GSH. J Biol Chem 1992;267: Beg AM, Sha NC, Bronson RT, Gosh S, Baltimore D. Embryonic lethality and liver degeneration in mice lacking the Rel A component of NF- B. Nature 1995;376: Van Antwerp DJ, Martin SJ, Kafri T, Green DR, Verma IM. Suppression of TNF-induced apoptosis by NF- B. Science 1996;274: Iimuro Y, Nishiura T, Hellerbrand C, Behrns KE, Schoonhoven R, Grisham JW, Brenner DA. NF-kB prevents apoptosis and liver dysfunction during liver regeneration. J Clin Invest 1998;101: Li Y, Zhang W, Mantell LI, Kazzaz JA, Fein AM, Horwitz S. Nuclear factor-kb is activated by hyperoxia but does not protect from cell death. J Biol Chem 1997;272: Leist M, Single M, Castoldi AF, Kuhnle S, Nicotera P. Intracellular adenosine triphosphatase (ATP) concentration: a switch in the decision between apoptosis and necrosis. J Exp Med 1997;185: Received February 2, Accepted August 25, Address requests for reprints to: José C. Fernández-Checa, Ph.D., Instituto Investigaciones Biomédicas, CSIC and Liver Unit, Hospital Clinic i Provincial, Villarroel 170, Barcelona, Spain. checa@medicina.ub.es; fax: (34) Supported by National Institute of Alcohol Abuse and Alcoholism (AA09526); Dirección General Política Científica y Técnica (PM ); Fondo Investigaciones Sanitarias (FISS /01); Plan Nacional de I D grant SAF C01; and Europharma. Dr. García-Ruiz is supported by a postdoctoral contract from the Ministerio de Educación y Ciencia, Spain. Anna Colell, Merce Miranda, and Montserrat Marí are supported by Europharma, and Esther Ardite is supported by Madaus Cerafarma. The authors thank Dr. N. Rice for providing antibodies to the members of Rel/NF- B, Dr. W. Anders for the generous gift of HP, and Dr. J. M. Mato for valuable comments.