Chronic ethanol consumption can cause protein. Proteasome Inhibition Potentiates CYP2E1-Mediated Toxicity in HepG2 Cells

Transcription

1 Proteasome Inhibition Potentiates CYP2E1-Mediated Toxicity in HepG2 Cells María JoséPérez and Arthur I. Cederbaum Chronic ethanol consumption causes increased oxidative damage in the liver. Induction of CYP2E1 is one pathway involved in how ethanol produces oxidative stress. Ethanol can cause protein accumulation, decreased proteolysis, and decreased proteasome activity. The objective of this study was to investigate the effect of inhibition of the proteasome activity on CYP2E1-dependent toxicity. HepG2 cells over-expressing CYP2E1 (E47 cells) were treated with arachidonic acid (AA) plus iron, agents important in development of alcoholic liver injury and which are toxic to E47 cells by a mechanism dependent on CYP2E1, oxidative stress, and lipid peroxidation. Addition of various proteasome inhibitors was associated with significant potentiation of the loss of cell viability caused by AA plus iron. Potentiation of toxicity was associated with increased oxidative damage as reflected by an increase in lipid peroxidation and accumulation of oxidized and nitrated proteins in E47 cells and an enhanced decline in mitochondrial membrane potential. Antioxidants prevented the loss of viability and the potentiation of this loss of viability by proteasome inhibition. CYP2E1 levels were elevated about 3-fold by the proteasome inhibitors. Inhibition of proteasome activity also potentiated toxicity of AA alone and toxicity after treatment to remove glutathione (GSH). Similar results were found in hepatocytes from pyrazole-treated rats with high levels of CYP2E1. In conclusion, proteasome activity plays an important role in modulating CYP2E1-mediated toxicity in HepG2 cells by regulating CYP2E1 levels and by removal of oxidized proteins. Such interactions may be important in CYP2E1-catalyzed toxicity of hepatotoxins and in alcohol-induced liver injury. (HEPATOLOGY 2003;37: ) Abbreviations: ROS, reactive oxygen species; AA, arachidonic acid; BSO, buthionine sulfoximine; GSH, glutathione; PUFA, polyunsaturated fatty acids; MEM, minimal essential medium; Fe-NTA, Fe-nitrilotriacetic acid; MTT, 3(4,5-dimethylthiazole-2-yl) 2,5-diphenyltetrazolium bromide; MDA, malondialdehyde; TBARS, thiobarbituric acid reactive substances; PI, propidium iodide; Rh123, rhodamine 123; MG115, Cbz-leu-leu-norvalinal; MG132, Cbz-leu-leu-leucinal; ALLN, acetyl-leu-leu-norleucinal. From the Department of Pharmacology and Biological Chemistry, Mount Sinai School of Medicine, New York, NY. Received January 14, 2003; accepted March 9, Supported by USPHS grants AA07311 and AA03312 from The National Institute on Alcohol Abuse and Alcoholism. Address reprint requests to: Arthur I. Cederbaum, Ph.D., Department of Pharmacology and Biological Chemistry, Box 1603, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY arthur.cederbaum@mssm.edu; fax: Copyright 2003 by the American Association for the Study of Liver Diseases /03/ $30.00/0 doi: /jhep Chronic ethanol consumption can cause protein accumulation in the liver, and this contributes to the ethanol-induced liver enlargement. 1 Alcoholinduced protein accumulation is due to a decrease in the protein catabolism. 2,3 Eukaryotic cells contain 2 major systems for protein degradation: the lysosomal apparatus, involved in long-lived protein degradation, 4 and the proteasome pathway, involved in the removal of short-lived proteins and proteins modified by adduct formation or oxidative stress. The bulk of proteins in mammalian cells are degraded by the proteasome pathway. 5,6 Some studies have suggested that decreased hepatic lysosomal proteolysis is a primary factor responsible for ethanol-induced hepatic protein accumulation. 7 However, it has also been reported that, in patients with alcoholic liver disease, ubiquitin levels are elevated in serum 8 as well as in hepatocytes. 9 Similar results were found in rodents after oral ethanol administration. 10 Intragastric ethanol infusion in rats caused inhibition of ubiquitin synthesis and reduced the proteolytic activity of the proteasome by 43%. 11 The reduction in proteasome activity and the subsequent accumulation of not only native but also damaged proteins (e.g., oxidized) may play a pathogenic role in experimental alcoholic liver disease. 12 An inverse correlation between proteasome activity and the degree of hepatic lipid peroxidation has been found in rats intragastrically fed ethanol. 11 The mechanism responsible for the decrease in proteasome activity by intragastric ethanol feeding is still not clear, but one factor may be the elevated levels of CYP2E1 1395

2 1396 PÉREZ AND CEDERBAUM HEPATOLOGY, June 2003 induced by chronic ethanol administration. 11,13 Moreover, CYP2E1 degradation involves the ubiquitin-proteasome pathway, 14,15 and its induction by ethanol occurs, in part, via stabilization of the enzyme preventing its degradation by the proteasome. 16 These data suggest that ethanol-dependent induction of CYP2E1 is associated with decreased activity of the proteasome complex in the liver. This is important because the accumulation of CYP2E1 promotes toxicity of various compounds, e.g., acetaminophen and CCl 4, and increases generation of reactive oxygen species (ROS) but also because the proteasome constitutes a secondary antioxidant defense system that protects critical regulatory molecules from permanent oxidative damage. 17 The selective degradation of oxidized proteins by the proteasome complex prevents the accumulation of modified and nonfunctional proteins in the cells. 18,19 These findings suggest a possible link between proteasome activity, CYP2E1 induction, ROS generation, and ethanol-induced liver injury. The aim of this study was to investigate whether inhibition of proteasome activity could modulate CYP2E1-dependent toxicity using models such as arachidonic acid (AA) plus iron, AA, and buthionine sulfoximine (BSO) toxicity in HepG2 cells over expressing CYP2E These models of toxicity were utilized because they appear to reproduce several of the key features associated with liver injury induced in the ethanol intragastric infusion model such as induction of CYP2E1, toxicity by polyunsaturated fatty acids (PUFA) or iron or GSH depletion, 28 and elevated lipid peroxidation. Materials and Methods Chemicals. Most reagents were purchased from Sigma Chemical Co. (St. Louis, MO). The protein DC-20 assay kit was from Bio-Rad (Hercules, CA). Culture and Treatment of Cells. HepG2 cells overexpressing CYP2E1 (E47 cells) 29 were cultured in minimal essential medium (MEM) containing 10% fetal bovine serum and 0.5 mg/ml of G418 (Invitrogen, Carlsbad, CA) supplemented with 100 units/ml of penicillin, 100 g/ml of streptomycin, and 2 mmol/l L- glutamine in a humidified atmosphere in 5% CO 2 at 37 C. CYP2E1 activity was monitored once a week by assaying the oxidation of p-nitrophenol. For each experiment, cells were plated and incubated in MEM overnight. For AA plus iron treatments, cells were treated with 20 mol/l AA for 14 hours, culture medium was replaced with fresh MEM (to remove unincorporated AA), and cells were incubated in the presence or absence of the proteasome inhibitors for 30 minutes; 25 mol/l Fe-nitrilotriacetic acid (Fe-NTA) was then added to initiate the toxicity phase (time 0), and cells were collected for various assays at different time points. For the experiment involving addition of adenovirus, 24 hours before the above treatment, E47 cells were infected with adenovirus containing no cdna (Ad-Null), adenovirus containing the cdna-encoding catalase (AdcCAT), or adenovirus containing the cdna-encoding catalase with a manganese-superoxide dismutase mitochondrial signal peptide, which translocates the catalase into the mitochondria (Ad-mCAT). 30 For AA alone (absence of iron) toxicity experiments, the proteasome inhibitors were added to the cells 30 minutes before adding the AA. For BSO toxicity experiments, cells were treated with 0.2 mmol/l BSO for 16 hours, to deplete GSH, and then the proteasome inhibitors were added. Cell viability was determined at varying time points (3-12 hours). Hepatocytes were isolated from pyrazole-treated rats as described previously. 31 The cells were incubated for 3 hours after seeding, then 10 mol/l lactacystin was added for 30 minutes before the addition of 60 mol/l AA. Viability was determined 6 hours after the addition of AA. Cytotoxicity Assays. Cell toxicity was routinely measured by assaying the reduction of 3(4,5-dimethylthiazole-2-yl) 2,5-diphenyltetrazolium bromide (MTT). 21,22 Changes in the mitochondrial membrane potential were examined by monitoring cell fluorescence after double staining with rhodamine 123 (Rh123) and propidium iodide (PI). 21,22,30 Proteasome Peptidase Activity Assay. The proteasomal chymotrypsin-like activity was determined by following the cleavage of the fluorogenic substrate Suc-Leu-Leu- Val-Tyr-AMC (SLLVY-AMC). Assays were performed at 37 C in a shaking water bath for 1 hour using 0.1 mol/l Tris-HCl, ph 8.0, 5 mmol/l MgCl 2, 50 mol/l SLLVY-AMC, and 50 g cell lysate protein. Enzymatic reactions were terminated by adding absolute ethanol. Fluorescence of 4-aminomethylcoumarin was measured at an excitation wavelength of 390 nm and an emission wavelength of 440 nm. Results are expressed as arbitrary units of fluorescence per g of cell protein. Lipid Peroxidation Assay. Production of malondialdehyde (MDA) was determined by assaying for thiobarbituric acid reactive substances (TBARS). 32 The amount of MDA produced was calculated using an extinction coefficient of M 1 cm 1. Protein Carbonyls Measurement. The amount of protein carbonyls was determined spectrophotometrically in cell lysates using the 2,4-dinitrophenylhydrazine (DNPH) assay. 33 The net absorption at 370 nm (DNPH-

3 HEPATOLOGY, Vol. 37, No. 6, 2003 PÉREZ AND CEDERBAUM 1397 treated sample minus sample blank) was determined. Carbonyl content was calculated using the extinction coefficient of 22,000 M 1 cm 1 for aliphatic hydrazones. 33 Immunoblot Analysis. CYP2E1 levels were detected by Western blot of cell lysates (20 g of protein) using 8% polyacrylamide gels. Polyclonal monospecific CYP2E1 IgG raised in rabbits was kindly provided by Dr. Jerome Lasker (Hackensack Medical Center, NJ). Nitrotyrosine residues were analyzed by slot blots on nitrocellulose membranes using 5 g of total protein from cell lysates. Membranes were blocked with nonfat milk, incubated with either anti-cyp2e1 polyclonal antibody or antinitrotyrosine polyclonal antibody (Upstate Biotechnology, Lake Placid, NY), washed, and incubated with peroxidase-conjugated goat anti-rabbit antibody (Chemicon International, Temecula, CA). Detection by the chemiluminescence reaction was carried out using the ECL kit (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom). Statistics. Data are expressed as mean standard deviation. The number of experiments is indicated in the legends to Figs. and Tables. Groups were compared among themselves using Student s t test for unpaired data to evaluate results in the absence and presence of proteasome inhibitors. Results Potentiation by Proteasome Inhibitors of the Synergistic Toxicity of Arachidonic Acid Plus Iron. AA plus iron produces a synergistic toxicity in E47 cells that is greater than that found in control C34 cells or in HepG23A4 cells, which express a different P450, CYP3A4. 23 This toxicity was previously shown to be necrotic, with no increase in DNA ladder formation or caspase 3 activity. 23 To characterize the effect of proteasome inhibitors on this toxicity, time course and dose-dependent experiments were conducted using Cbzleu-leu-norvalinal (MG115) or Cbz-leu-leu-leucinal (MG132) as proteasome inhibitors. Figure 1A shows that the loss of viability induced by AA plus iron in E47 cells was time dependent over a 9-hour incubation. MG115 itself had little effect on cell viability but potentiated the toxicity of iron plus AA in a concentration-dependent manner. A dose-dependent experiment was carried out assessing the effectiveness of MG132 on CYP2E1-dependent toxicity (Fig. 1B). The incubation with Fe-NTA for 6 hours after AA preincubation induced a loss of viability of about 35%. MG132 potentiated this toxicity in a dosedependent manner, reaching values from 50% to 75% loss of viability at concentrations from 5 to 50 mol/l, Fig. 1. Time course and concentration curve for the potentiation of arachidonic acid plus iron-induced cytotoxicity by MG115 or MG132. (A) E47 cells were preincubated with medium containing 20 mol/l arachidonic acid (AA) for 14 hours. The medium was then replaced, and the cells were incubated in the absence or presence of either 1 or 10 mol/l of MG115 for 30 minutes before adding 25 mol/l Fe-NTA. Cell viability was determined after 3, 6, or 9 hours of incubation by the MTT assay. (B) E47 cells were preincubated with medium containing 20 mol/l AA for 14 hours. The medium was then replaced, and the cells were incubated in the absence or presence of the indicated concentrations of MG132 before adding 25 mol/l Fe-NTA. Cell viability was determined after 6 hours of incubation by the MTT assay. The percentage of viability was calculated in comparison with the no addition control, which was taken as the 100% viability value. Results are expressed as mean SD and are from 4 experiments. *P.05, **P.01 versus the corresponding control cells; #P.05, ##P.01 versus the AA Fe-treated cells in the absence of the inhibitors. respectively. The effect of other proteasome inhibitors was also assayed. AA plus Fe-NTA induced about a 34% loss of viability after 6 hours of incubation (Fig. 2). The addition of 10 mol/l MG115, 10 mol/l MG132, 130 mol/l acetyl-leu-leu-norleucinal (ALLN), or 25 mol/l lactacystin significantly increased this toxicity, resulting in losses of viability of 54%, 67%, 70%, and 48%, respectively. No toxic effect was observed by these proteasome inhibitors alone under the assayed conditions. In these experiments, the MG115, MG132, and ALLN were dissolved in DMSO, final concentration of 0.1% or 0.2% vol/vol. Controls also were treated with 0.1% or 0.2% DMSO. Lactacystin was dissolved in water. Proteasome Activity. To validate the effectiveness of the proteasome inhibitors, the chymotrypsin-like peptidase activity of the proteasome was assayed after incuba-

4 1398 PÉREZ AND CEDERBAUM HEPATOLOGY, June 2003 Fig. 2. Potentiation of arachidonic acid plus iron cytotoxicity by proteasome inhibitors. E47 cells were preincubated with medium, containing 20 mol/l arachidonic acid (AA) for 14 hours. The medium was then replaced, and the cells were incubated in the absence or presence of the proteasome inhibitors: 10 mol/l MG115, 10 mol/l MG132, 130 mol/l ALLN, or 25 mol/l lactacystin for 30 minutes before adding 25 mol/l Fe-NTA. After 6 hours of incubation, cell viability was determined by the MTT assay. The percentage of viability was calculated in comparison with the no addition control, which was taken as the 100% viability value. Results are expressed as mean SD and are from 4 experiments. *P.05, **P.01 versus the corresponding control cells; #P.05 versus the AA Fe-treated cells in the absence of the inhibitors. Table 1. Proteasome Activity in the Absence or Presence of Proteasome Inhibitors Addition Proteasome Activity (AU/ g Protein) mol/l MG * 25 mol/l Lactacystin NOTE. E47 cells were incubated in the absence or presence of MG115 or lactacystin for 6 hours. Cell lysates were then collected, and proteasome activity was determined using the fluorogenic peptide Suc-Leu-Leu-Val-Tyr-AMC as described in the Materials and Methods section. Results are expressed as mean SD. *P.05. P.01 versus no addition. Fig. 3. Effect of MG115 on arachidonic acid plus iron-induced lipid peroxidation and levels of protein carbonyls. E47 cells were preincubated with medium, containing 20 mol/l arachidonic acid (AA) for 14 hours. The medium was then replaced, and the cells were incubated in the absence or presence of 10 mol/l MG115 for 30 minutes before adding 25 mol/l Fe-NTA. After 6 hours of incubation, the cells were harvested by scraping, and cell lysates were collected and assayed for the production of MDA, utilizing the TBARS assay (A), or for protein carbonyls (B) as described in the Materials and Methods section. Results are expressed as mean SD and are from 4 experiments. *P.05, **P.01 versus the corresponding control cells; #P.05 versus the AA Fetreated cells in the absence of MG115. tion with MG115 or lactacystin for 6 hours; 10 mol/l MG115 reduced the proteasome activity about 50%, whereas 25 mol/l lactacystin produced about a 90% loss of activity (Table 1). The lesser extent of observed inhibition by MG115 compared with lactacystin is likely areflection of the fact that MG115 is a reversible inhibitor, whereas lactacystin is an irreversible inhibitor of the proteasome; therefore, some of the inhibition by MG115 is attenuated by dilution during preparation of the cell extracts and by the assay buffer and 60-minute assay incubation time. This in vitro loss of inhibition does not occur with the irreversible inhibitor lactacystin. Effect of Lysosomal Inhibitors. Although lactacystin appears to be a relatively specific inhibitor of the proteasome complex, the peptide aldehyde proteasome inhibitors can also inhibit lysosomal cysteine proteases. To study whether lysosomal proteases could be involved in the potentiation of AA plus iron toxicity by these inhibitors, the effect of inhibitors of lysosomal proteases such as 50 mmol/l ammonium chloride, 50 mol/l pepstatin, 50 mol/l aprotinin, and 50 mol/l leupeptin were tested. None of these compounds affected AA plus ironinduced loss of viability, under conditions in which 10 mol/l MG115 potentiated the toxicity (data not shown). Effect of MG115 on Arachidonic Acid Plus Iron- Induced Lipid Peroxidation and Levels of Protein Carbonyls. AA plus iron increased lipid peroxidation in the E47 cells, and this played a central role in the developing toxicity Lipid peroxidation was assessed by measuring production of MDA by the TBARS assay. AA plus Fe-NTA induced a 3-fold increase in MDA in E47 cells (Fig. 3A). Preincubating with 10 mol/l MG115

5 HEPATOLOGY, Vol. 37, No. 6, 2003 PÉREZ AND CEDERBAUM 1399 Fig. 4. Levels of nitrotyrosine adducts in E47 cells treated with arachidonic acid plus iron in the absence or presence of MG115. E47 cells were preincubated with medium, containing 20 mol/l arachidonic acid (AA) for 14 hours. The medium was then replaced, and the cells were incubated in the absence or presence of 10 mol/l MG115 for 30 minutes before adding 25 mol/l Fe-NTA. Cell lysates were collected after 6 hours of incubation and analyzed by slot blot for nitrotyrosine residues. (A) Typical slot blot of triplicate aliquots. (B) Bar graph depicting the quantification by densitometry of different blots from 4 experiments. Results are expressed as mean SD. *P.05, **P.01 versus the corresponding control cells; ##P.01 versus the AA Fe-treated cells in the absence of the inhibitor. before adding the Fe-NTA resulted in a further doubling of MDA levels, resulting in an overall 6-fold increase of lipid peroxidation. Protein oxidative damage was assayed by measuring oxidized protein levels. Treatment with AA plus iron induced a small, but significant increase in the protein carbonyl content (Fig. 3B). Levels of protein carbonyl were further enhanced in the presence of the proteasome inhibitor MG115 (Fig. 3B). Levels of Nitrotyrosine Protein Adducts. Oxidative damage was also studied by measurement of the levels of nitrated protein, assaying for nitrotyrosine adducts by slot blot. Representative blots as well as the densitometric quantification of several blots are shown in Fig. 4A and B. Treatment with AA plus iron induced a slight increase in protein nitrotyrosine adduct levels; however, AA plus iron in the presence of MG115 elevated nitrotyrosine levels about 8-fold. MG115 alone had no effect on nitrotyrosine levels. Effect of Proteasome Inhibitors on CYP2E1 Levels. To study mechanisms by which the proteasome inhibitors were potentiating the AA plus iron toxicity, the effect of these compounds on the content of CYP2E1 was assayed. Two separate cultures of E47 cells were incubated with or without MG115 while 2 other E47 cultures were incubated with or without lactacystin for 6 hours, the time point at which most toxicity studies were carried out. Incubation with MG115 or lactacystin produced a 3-fold increase in CYP2E1 levels (Fig. 5). Protection by Antioxidants. Antioxidants protected against the toxicity of AA plus iron in E47 cells, and it was important to evaluate whether antioxidants would protect against the enhanced toxicity found in the presence of proteasome inhibitors. The antioxidant Trolox or infection with adenoviral vectors that express catalase protected the E47 cells against the loss of viability induced by AA plus iron (Table 2). Cytosolic or mitochondrial catalase delivered by adenovirus (but not Ad-Null) afforded about 75% protection, whereas Trolox was completely protective against the enhanced toxicity induced by AA plus iron plus MG115 (Table 2). Effect of MG115 on Changes in Mitochondrial Membrane Potential. Damage to mitochondria and decreases in mitochondrial membrane potential are important in CYP2E1-dependent toxicity. 34 Mitochondrial membrane potential was assayed by flow cytometry after staining with Rh123. Untreated E47 cells displayed strong rhodamine fluorescence (M2 population), reflective of viable, intact cells (Fig. 6A). A small percentage of cells (23%) were located in the M1 population (low rhodamine fluorescence), reflective of damaged cells. Although some effect on the mitochondrial membrane potential may be due to CYP2E1 even in the absence of added prooxidants, these cells are somewhat labile and are damaged by the trypsinization process, sample workup, and resuspension. MG115 alone had no effect on the Fig. 5. Effect of proteasome inhibitors on CYP2E1 levels. E47 cells were incubated in the absence or presence of 10 mol/l MG115 or 10 mol/l lactacystin for 6 hours. Cell lysates were then collected and analyzed by Western blot. Western blots of CYP2E1 are shown for 2 separate experiments, from 2 different cultures, for all treatments. Results are expressed as arbitrary units (AU) of densitometry for each representative blot.

6 1400 PÉREZ AND CEDERBAUM HEPATOLOGY, June 2003 Table 2. Effect of Antioxidants on Arachidonic Acid Plus Iron Cytoxicity in the Absence or Presence of MG115 Addition Viability (%) MG * Trolox * MG115 Trolox Ad-Null MG115 Ad-Null * Ad-cCat MG115 Ad-cCat Ad-mCat MG115 Ad-mCat NOTE. E47 cells were preincubated with medium containing 20 mol/l arachidonic acid for 14 hours. The medium was then replaced, and, where indicated, 100 mol/l Trolox or 10 mol/l MG115 was added, and, after 30 to 60 minutes of incubation, 25 mol/l Fe-NTA was added to initiate the toxicity phase. Another set of cells were infected with adenovirus containing cytosolic or mitochondrial catalase or with the null adenovirus before the arachidonic acid plus iron additions. After 6 hours of incubation, cell viability was determined by the MTT assay. The percentage of viability was calculated in comparison with the no addition control, which was taken as the 100% viability value. Results are expressed as mean SD and are from 4 experiments. *P.01 versus AA Fe-treated cells in the absence of MG115. P.01 versus AA Fe-treated cells in the presence of MG115. P.05 versus AA Fe-treated cells in the absence of MG115. mitochondrial membrane potential. E47 cells treated with AA and iron displayed lower rhodamine fluorescence, increasing the M1 population of cells to 64%. Treatment with AA plus iron plus MG115 further increased the percentage of cells in the M1 population to 88% (Fig. 6A). Quantification of these changes for several experiments is shown in Fig. 6B. The insets of the histographs display the PI fluorescence (Y axis, FL2-H fluorescence) indicative of damaged, permeable cells. The numbers in the upper left quadrant reflect damaged cells with low rhodamine fluorescence and confirm the toxicity produced by AA plus iron and the further increase in toxicity by AA plus iron in the presence of MG115. Potentiation by Proteasome Inhibitors of Arachidonic Acid and BSO Cytotoxicities. The effect of MG115 and lactacystin was determined in other toxicity models in E47 cells, the toxicity produced by AA alone, in the absence of added iron, and toxicity produced when cellular GSH levels were depleted after treatment with BSO. 20,29 Treatment with 60 mol/l AA for 6 hours produced a 20% not significant loss of viability. However, the presence of AA plus MG115 resulted in a 55% significant loss of viability (Fig. 7A). Incubation with 60 or 30 mol/l AA for 6 or 12 hours, respectively, did not induce a significant loss of viability, but the presence of AA plus lactacystin induced a significant loss of viability of 35% and 58%, respectively (Fig. 7B). Similar results were found in the BSO toxicity model. Incubation with MG115 or lactacystin for 3 or 6 hours, after a 16-hour preincubation with BSO, induced a loss of viability of about 60% and 70%, respectively, compared with a 35% loss when incubated for 3 or 6 hours without the proteasome inhibitor (Fig. 7C) Effect of Lactacystin on Arachidonic Acid Cytoxicity in Hepatocytes Isolated From Pyrazole-Treated Rats. To extend the results in the E47 HepG2 cells to primary hepatocytes, hepatocytes were isolated from pyrazole-treated rats, and toxicity of AA was evaluated in the absence and presence of lactacystin. Pyrazole treatment elevates the levels of CYP2E1 about 3- to 4-fold over those found with saline controls. Hepatocytes isolated from pyrazole-treated rats showed a 30% loss of viability after incubation with 60 mol/l AA for 6 hours (Fig. 8). Lactacystin increased the AA-induced loss of viability to about 60% (Fig. 8). Lactacystin alone had no effect on cell viability. Discussion The main goal of this study was to investigate the effect of inhibition of proteasome activity on CYP2E1 toxicity. For this purpose, toxicity was induced in HepG2 cells over-expressing CYP2E1 (E47 cells) by iron and AA. The results obtained demonstrate that inhibition of proteasome activity potentiates this toxicity. The iron plus AAinduced decrease in vital dye reduction and decrease in the mitochondria membrane potential were enhanced by proteasome inhibitors. Potentiation of toxicity was associated with an increased oxidative damage as reflected by increased MDA, protein carbonyls, and protein nitrotyrosine adducts. The synergistic toxicity was related to enhanced oxidative stress because antioxidants such as trolox or catalase prevented both the AA plus iron toxicity and the potentiated toxicity found in the presence of the proteasome inhibitors. Inhibition of proteasome activity also potentiated the AA- and BSO-dependent toxicities in the E47 cells. Similar results were found in hepatocytes from pyrazole-treated rats with high levels of CYP2E1. Induction of CYP2E1 by ethanol is one of the pathways by which ethanol produces increased oxidative stress in the liver and how ethanol potentiates the toxicity of numerous toxicants. CYP2E1 degradation involves the ubiquitin-proteasome pathway, 14,15 and its induction by ethanol occurs, in part, via stabilization of the enzyme preventing its degradation by the proteasome. 15 Proteasome activity is significantly decreased in the liver of intragastric ethanol-fed animals. 11,13,35 Several types of low-molecular-weight inhibitors of the proteasome have been identified that can readily enter cells and selectively inhibit this degradative pathway. The most widely used are the peptide aldehydes, such as MG132, MG115, and ALLN. These inhibitors block

7 HEPATOLOGY, Vol. 37, No. 6, 2003 PÉREZ AND CEDERBAUM 1401 Fig. 6. Flow cytometry analysis of the mitochondrial membrane potential. E47 cells were incubated with MEM only (no addition) or were preincubated with medium containing 20 mol/l arachidonic acid (AA) for 14 hours. The medium was then replaced, and the cells were incubated in the absence or presence of 10 mol/l MG115 for 30 minutes before adding 25 mol/l Fe-NTA. After 6 hours of incubation, the cells were incubated with medium containing 5 g/ml Rh123 for 1 hour. The cells were harvested by trypsinization, resuspended in 1 ml of medium containing 5 g/ml PI, and the intensity of fluorescence from Rh123 was determined by flow cytometry. M1 and M2 are 2 populations of cells with low (M1) and high (M2) Rh123 fluorescence (A). The insets of A show a typical histogram depicting the fluorescence (FL1-H) associated with Rh123 fluorescence and the fluorescence (FL2-H) associated with PI fluorescence in E47 cells. (B) Bar graph summarizing the percentage of cells in the M1 fraction from 4 experiments. Results are expressed as mean SD. *P.05 versus the corresponding control cells; #P.05 versus the AA Fe-treated cells in the absence of MG115. proteolytic activity of the 26S proteasome without influencing its ATPase or isopeptidase activities. These proteasome inhibitors were utilized in this study, and, after exposure to these agents alone for 6 hours, cell viability and growth were not affected, but toxicity by AA plus iron or AA or BSO were increased. However, the peptide aldehydes also inhibit certain lysosomal cysteine proteases. 36 However, some selective inhibitors of lysosomal function such as ammonium chloride, pepstatin, aprotinin, and leupeptin were also tested, and no effect was found on the AA plus iron toxicity. Moreover, a similar potentiating effect on toxicity occurred with a more specific proteasome inhibitor, lactacystin, that does not affect lysosome proteases. Because CYP2E1 is degraded by the proteasome, one possible mechanism involved in the potentiation of the AA plus iron toxicity by proteasome inhibition in E47 cells is an increase in CYP2E1 levels. Indeed, CYP2E1 levels were increased about 3-fold after treatment with proteasome inhibitors. CYP2E1 turnover was previously characterized by pulse-chase analysis in HepG2 cells, and a half-life of about 2.5 hours was found in the absence of stabilizing substrate or ligand. 37 Thus, the 3-fold increase in CYP2E1 levels by MG115 and lactacystin after a 6-hour incubation (about 2 half-lives) is consistent with the 2.5-hour half-life of CYP2E1. Increased levels of CYP2E1 in the presence of proteasome inhibitors will exacerbate the oxidative stress produced by iron and

8 1402 PÉREZ AND CEDERBAUM HEPATOLOGY, June 2003 ethanol has been proposed to play a role in the decrease in proteasome activity found in the liver of these animals. Indeed, CYP2E1 inhibitors ameliorated the decrease in proteasome activity and improved the liver pathology. 11,13 Proteasome activity was reduced in wild-type mice intragastrically infused with ethanol (CYP2E1 was increased 6-fold by ethanol), whereas no change in proteasome activity was found in ethanol-fed CYP2E1 knockout mice. 38 Increased oxidative damage can be due not only to an increased generation of ROS but also to decreased clearance of oxidatively modified biomolecules. Because the proteasome is a major clearing machinery for short-lived proteins as well as damaged proteins, inhibition of its function can lead to an accumulation of not only native proteins (e.g., CYP2E1) but damaged (e.g., oxidized) proteins. The 20S proteasome is involved in selective recognition and degradation of proteins that have been oxidized by exposure to superoxide anion radical, hydrogen peroxide, hydroxyl radical, or peroxynitrite. 39 A single dose of ethanol or H 2 O 2, in HepG2 cells, caused an increase in protein oxidation followed by an increase in proteasome-dependent intracellular degradation. 40 Evidence of oxidative protein damage was found in ethanolinfused rats 12 as well as increased levels of protein carbonyls, and the extent of accumulation of oxidized protein in the liver was associated with the extent of ethanol-induced liver injury. Chronic oral feeding of ethanol significantly increased both cytosolic and mitochondrial concentrations of protein carbonyls. 41 AA plus iron induced an increased accumulation of oxidized proteins in E47 cells. Reduction of proteasome activity led to an additional ac- Fig. 7. Potentiation of arachidonic acid or BSO toxicity by proteasome inhibitors. (A) E47 cells were preincubated in the absence or presence of 10 mol/l MG115 for 30 minutes before adding 60 mol/l arachidonic acid (AA). Cell viability was determined 6 hours after the addition of the AA. (B) E47 cells were preincubated in the absence or presence of 25 mol/l lactacystin for 30 minutes before adding 30 mol/l or 60 mol/l AA. Cell viability was determined after 12 (30 mol/l AA) or 6 (60 mol/l AA) hours of incubation. (C) E47 cells were preincubated with medium containing 0.2 mmol/l BSO for 16 hours. The cells were then incubated in the absence or presence of 25 mol/l lactacystin or 10 mol/l MG115 for either 3 or 6 hours, followed by determination of cell viability by the MTT assay. The percentage of viability was calculated in comparison with the no addition control, which was taken as the 100% viability value. Results are expressed as mean SD and are from 4 experiments. *P.05, **P.01 versus the corresponding control cells; #P.05, ##P.01 versus the AA- or BSO-treated cells in the absence of the inhibitors. PUFA or after GSH depletion and is likely to play a key role in the potentiation of oxidative damage and cellular toxicity. Oxidative stress coupled to elevated levels of CYP2E1 induced in the livers of animals chronically fed Fig. 8. Potentiation of arachidonic acid-induced toxicity by lactacystin in hepatocytes from pyrazole-treated rats. Hepatocytes from pyrazoletreated rats were preincubated in the absence or presence of 25 mol/l lactacystin for 30 minutes before adding 60 mol/l arachidonic acid (AA) for 6 hours. After this incubation, cell viability was determined by the MTT assay. The percentage of viability was calculated in comparison with the no addition control, which was taken as the 100% viability value. Results are expressed as mean SD and are from 4 experiments. *P.05, **P.01 versus the corresponding control cells; #P.05 versus the AA-treated cells in the absence of the inhibitor.

9 HEPATOLOGY, Vol. 37, No. 6, 2003 PÉREZ AND CEDERBAUM 1403 cumulation of oxidized proteins. We have not yet assayed cellular distribution of the increased protein carbonyls or other oxidized products (MDA, nitrotyrosine adducts) in the presence of the proteasome inhibitors. This may be important in view of the potentiation in the decline in the mitochondrial membrane potential caused by the proteasome inhibitors. Protein carbonyls can arise both from direct oxidative damage to amino acid residues and also from the binding of certain end products of lipid peroxidation to proteins. 42 Rises in protein carbonyls and lipid peroxidation products can be related. The elevated levels of MDA induced by AA plus iron in E47 cells were further increased in the presence of MG115. Proteasome chymotrypsinlike activity and the degree of hepatic lipid peroxidation were found to be inversely correlated in the ethanol intragastric infusion model. 11 Nitrated proteins, another index of oxidative damage, were measured as nitrotyrosine adducts. Elevated levels of nitrotyrosine adducts were found in AA plus iron-treated E47 cells. Reduction of proteasome activity led to a further increase in this accumulation. Rats fed ethanol intragastrically displayed nitrotyrosine formation in the liver and increased plasma NO. 43 Chronic oral feeding of ethanol to rats also showed increased nitration of proteins. 44 Elevated levels of oxidized and nitrated proteins can be an indicator of high levels of reactive nitrogen species such as peroxynitrite because this agent is a potent oxidant and nitrating agent. 45 Peroxynitrite is formed in the presence of high levels of superoxide, which reacts rapidly with nitric oxide. The increase in CYP2E1 levels, after inhibition of the proteasome, can lead to an increase in ROS generation, and enhanced superoxide formation can lead to the formation of peroxynitrite after reaction with nitric oxide. It is not known whether peroxynitrite contributes to the AA plus iron dependent toxicity in the E47 cells or whether the increase in nitrotyrosine adducts by MG115 is due to increased production of nitrotyrosine or decreased removal of the adducts because of inhibition of the proteasome, or both. A preliminary experiment showed that 2 mmol/l L-NAME, an inhibitor of nitric oxide synthase, afforded some protection (about 30%) against the AA plus iron toxicity. Further studies are planned to evaluate the role of NO and peroxynitrite in the CYP2E1-dependent toxicity. Proteasome inhibition, by lactacystin or epoxomicin in the absence of other additions, led to increases in the levels of oxidative protein damage, lipid peroxidation, and 3-nitrotyrosine in a human teratocarcinoma and a neuroblastoma cell line. 46 No changes in these parameters were observed when the E47 cells were incubated with the proteasome inhibitors alone, which likely reflects the fact that the time of incubation in our model was a maximum of 12 hours as compared with days in the Lee et al. 46 study. In summary, inhibition of the activity of the proteasome in E47 cells led to increases in the levels of CYP2E1, followed by increases in lipid peroxidation, oxidized and nitrated proteins, and damage to mitochondria after treatment with prooxidants such as AA plus iron, ultimately potentiating the decline in cell viability. Elevated levels of ROS and oxidative damage cause mitochondrial injury that can lead to cell death. These results suggest that proteasome activity plays an important role in CYP2E1-mediated toxicity in HepG2 cells (and hepatocytes from pyrazole-treated rats). Such interactions coupled to the key role of the proteasome in degrading CYP2E1 and in removing oxidized cellular proteins may be important in CYP2E1-catalyzed toxicity of hepatotoxins such as acetaminophen, CCl 4, and others and may play an important role in alcohol-induced liver injury. References 1. Baraona E, Leo MA, Borowsky SA, Lieber CS. Alcoholic hepatomegaly: accumulation of protein in the liver. Science 1975;90: Donohue TM Jr, Zetterman RK, Tuma DJ. Effect of chronic ethanol administration on protein catabolism in rat liver. Alcohol Clin Exp Res 1989;13: Poso AR, Hirsimaki P. Inhibition of proteolysis in the liver by chronic ethanol feeding. Biochem J 1991;273: Henell F, Berkenstam A, Ahlberg J, Glaumann H. Degradation of shortand long-lived proteins in perfused liver and in isolated autophagic vacuoles-lysosomes. Exp Mol Pathol 1987;46: Goldberg AL. Functions of the proteasome: the lysis at the end of the tunnel. Science 1995;268: Ciechanover A. The ubiquitin-proteasome proteolytic pathway. Cell 1994;79: Kharbanda KK, McVicker DL, Zetterman RK, Donohue TM Jr. Ethanol consumption reduces the proteolytic capacity and protease activities of hepatic lysosomes. Biochim Biophys Acta 1995;1245: Takagi M, Yamauchi M, Toda G, Takada K, Hirakawa T, Ohkawa K. Serum ubiquitin levels in patients with alcoholic liver disease. Alcohol Clin Exp Res 1999;23:76S-80S. 9. Lowe J, Blanchard A, Morrell K, Lennox G, Reynolds L, Billett M, Landon M, et al. Ubiquitin is a common factor in intermediate filament inclusion bodies of diverse type in man, including those of Parkinson s disease, Pick s disease, and Alzheimer s disease, as well as Rosenthal fibres in cerebellar astrocytomas, cytoplasmic bodies in muscle, and mallory bodies in alcoholic liver disease. J Pathol 1988;155: Born LJ, Kharbanda KK, McVicker DL, Zetterman RK, Donohue TM Jr. Effects of ethanol administration on components of the ubiquitin proteolytic pathway in rat liver. HEPATOLOGY 1996;23: Fataccioli V, Andraud E, Gentil M, French SW, Rouach H. Effects of chronic ethanol administration on rat liver proteasome activities: relationship with oxidative stress. HEPATOLOGY 1999;29: Rouach H, Fataccioli V, Gentil M, French SW, Morimoto M, Nordmann R. Effect of chronic ethanol feeding on lipid peroxidation and protein oxidation in relation to liver pathology. HEPATOLOGY 1997;25: Gouillon Z, Lucas D, Li J, Hagbjork AL, French BA, Fu P, Fang C, et al. Inhibition of ethanol-induced liver disease in the intragastric feeding rat model by chlormethiazole. Proc Soc Exp Biol Med 2000;224: Roberts BJ. Evidence of proteasome-mediated cytochrome P-450 degradation. J Biol Chem 1997;272:

10 1404 PÉREZ AND CEDERBAUM HEPATOLOGY, June Goasduff T, Cederbaum AI. NADPH-dependent microsomal electron transfer increases degradation of CYP2E1 by the proteasome complex: role of reactive oxygen species. Arch Biochem Biophys 1999;370: Banerjee A, Kocarek TA, Novak RF. Identification of a ubiquitinationtarget/substrate-interaction domain of cytochrome P-450 (CYP) 2E1. Drug Metab Dispos 2000;28: Davies KJA. Intracellular proteolytic systems may function as secondary antioxidant defenses: an hypothesis. J Free Radic Biol Med 1986;2: Grune T, Reinheckel T, Joshi M, Davies KJ. Proteolysis in cultured liver epithelial cells during oxidative stress. Role of the multicatalytic proteinase complex, proteasome. J Biol Chem 1995;270: Grune T, Reinheckel T, Davies KJ. Degradation of oxidized proteins in mammalian cells. FASEB J 1997;11: Chen Q, Galleano M, Cederbaum AI. Cytotoxicity and apoptosis produced by arachidonic acid in HepG2 cells overexpressing human cytochrome P4502E1. J Biol Chem 1997;272: Pérez MJ, Cederbaum AI. Spin trapping agents (Tempol and POBN) protect HepG2 cells overexpressing CYP2E1 against arachidonic acid toxicity. Free Radic Biol Med 2001;30: Pérez MJ, Cederbaum AI. Antioxidant and pro-oxidant effects of a manganese porphyrin complex against CYP2E1-dependent toxicity. Free Radic Biol Med 2002;33: Caro AA, Cederbaum AI. Synergistic toxicity of iron and arachidonic acid in HepG2 cells overexpressing CYP2E1. Mol Pharmacol 2001;60: Takahashi H, Johansson H, French SW. Effect of dietary fat composition on activities of the microsomal ethanol oxidizing system and ethanolinducible cytochrome P450 (CYP2E1) in the liver of rats chronically fed ethanol. Pharmacol Toxicol 1992;70: Tsukamoto H, Horne W, Kamimura S, Niemela O, Parkkila S, Yla-Herttuala S, Brittenham GM. Experimental liver cirrhosis induced by alcohol and iron. J Clin Invest 1995;96: Nanji A, Zhao S, Lamb RG, Sadrzadeh SMH, Dannenberg AJ, Waxman DJ. Changes in microsomal phospholipases and arachidonic acid in experimental alcoholic liver injury: relationship to cytochrome P-450 2E1 induction and conjugated diene formation. Alcohol Clin Exp Res 1993;17: Morimoto M, Hagbjork AL, Nanji AA, Ingelman-Sundberg M, Lindros KO, Fu PC, Albano E, et al. Role of cytochrome P4502E1 in alcoholic liver disease pathogenesis. Alcohol 1993;10: Fernandez-Checa JC, Hirano T, Tsukamoto H, Kaplowitz N. Mitochondrial glutathione depletion in alcoholic liver disease. Alcohol 1993;10: Chen Q, Cederbaum AI. Cytotoxicity and apoptosis produced by cytochrome P450 2E1 in Hep G2 cells. Mol Pharmacol 1998;53: Bai J, Cederbaum AI. Adenovirus-mediated overexpression in the cytosolic or mitochondrial compartment protects against cytochrome P4502E1- dependent toxicity in HepG2 cells. J Biol Chem 2001;276: Wu D, Cederbaum AI. Ethanol and arachidonic acid produce toxicity in hepatocytes from pyrazole-treated rats with high levels of CYP2E1. Mol Cell Biochem 2000;204: Niehaus WG, Samuelsson B. Formation of malonaldehyde from phospholipid arachidonate during microsomal lipid peroxidation. Eur J Biochem 1968;6: Reznick AZ, Packer L. Oxidative damage to proteins: spectrophotometric method for carbonyl assay. Methods Enzymol 1994;233: Wu D, Cederbaum AI. Cyclosporine A protects against arachidonic acid toxicity in rat hepatocytes: role of CYP2E1 and mitochondria. HEPATOL- OGY 2002;35: Donohue TM Jr, Zetterman RK, Zhang-Gouillon ZQ, French SW. Peptidase activities of the multicatalytic protease in rat liver after voluntary and intragastric ethanol administration. HEPATOLOGY 1998;28: Rock KL, Gramm C, Rothstein L, Clark K, Stein R, Dick L, Hwang D, et al. Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC class I molecules. Cell 1994;78: Yang MX, Cederbaum AI. Characterization of cytochrome P4502E1 turnover in transfected HepG2 cells expressing human CYP2E1. Arch Biochem Biophys 1997;341: Bardag-Gorce F, Yuan QX, Li J, French BA, Fang C, Ingelman-Sundberg M, French SW. The effect of ethanol-induced cytochrome p4502e1 on the inhibition of proteasome activity by alcohol. Biochem Biophys Res Commun 2000;279: Dunlop RA, Rodgers KJ, Dean RT. Recent developments in the intracellular degradation of oxidized proteins. Free Radic Biol Med 2002;33: Pirlich M, Muller C, Sandig G, Jakstadt M, Sitte N, Lochs H, Grune T. Increased proteolysis after single-dose exposure with hepatotoxins in HepG2 cells. Free Radic Biol Med 2002;33: Bailey SM, Patel VB, Young TA, Asayama K, Cunningham CC. Chronic ethanol consumption alters the glutathione/glutathione peroxidase-1 system and protein oxidation status in rat liver. Alcohol Clin Exp Res 2001; 25: Stadtman ER, Berlett BS. Reactive oxygen-mediated protein oxidation in aging and disease. Drug Metab Rev 1998;30: Nanji AA, Jokelainen K, Lau GK, Rahemtulla A, Tipoe GL, Polavarapu R, Lalani EN. Arginine reverses ethanol-induced inflammatory and fibrotic changes in liver despite continued ethanol administration. J Pharmacol Exp Ther 2001;299: Baraona E, Zeballos GA, Shoichet L, Mak KM, Lieber CS. Ethanol consumption increases nitric oxide production in rats, and its peroxynitritemediated toxicity is attenuated by polyenylphosphatidylcholine. Alcohol Clin Exp Res 2002;26: Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am J Physiol 1996;271:C1424-C Lee MH, Hyun DH, Jenner P, Halliwell B. Effect of proteasome inhibition on cellular oxidative damage, antioxidant defences and nitric oxide production. J Neurochem 2001;78:32-41.