|Home | About | Journals | Submit | Contact Us | Français|
We have previously shown that treatment of mice with pyrazole or acute ethanol potentiated Fas agonistic Jo2 antibody-induced liver injury by a mechanism involving induction of CYP2E1 and elevated oxidative stress. The current study evaluated whether chronic alcohol feeding potentiates Fas-induced liver injury and whether CYP2E1 plays a role in any enhanced hepatotoxicity. Wild type and CYP2E1 knockout mice were fed ethanol or isocaloric dextrose for four weeks followed by a single treatment with either saline or Jo2. Mice were killed 8 h after the Jo2 challenge. There were 3–5 fold increases of transaminases and more extensive eosinophilic necrosis, hemorrhage and infiltration of inflammatory cells in the central zone of the hepatic lobule in the ethanol-fed mice treated with Jo2 compared to the dextrose/Jo2 or ethanol/saline treated mice. Liver injury was blunted in ethanol-fed CYP2E1 knockout mice treated with Jo2. The chronic ethanol feeding produced steatosis, elevation of CYP2E1 and oxidative stress in wild type but not CYP2E1 knockout mice. These changes in wild type mice fed ethanol were similar after saline or Jo2 treatment. The Jo2 treatment produced activation of JNK and p38 MAP kinase, increased activity of caspases 8 and 3, and lowered hepatic GSH levels in both the dextrose- and alcohol-fed mice. JNK was activated at early times after Jo2 treatment in the ethanol-fed mice. Serum TNF-α levels were strikingly elevated in the wild type ethanol/Jo2 group which showed liver injury compared to all the other groups which did not show liver injury. Inhibition of JNK or p38 MAPK partially, but not completely, prevented the elevated liver injury in the wild type ethanol/Jo2 mice. These results show that chronic ethanol feeding enhances Fas-induced liver injury by a mechanism associated with induction of CYP2E1, elevated serum TNF-α levels and activation of MAPK.
Alcoholic liver disease (ALD) is a common clinical complication of long-term alcohol abuse. Its morphological features include alcoholic fatty liver (steatosis), alcoholic hepatitis, and alcoholic cirrhosis. The pathogenesis of ALD involves multifactorial processes such as genetic, nutritional and environmental factors, and a number of injurious factors such as oxidative/nitrosative stress, bacterial lipopolysaccharide and cytokines [1,2]. Accumulating evidence suggests that oxidative stress and lipid peroxidation play a key role in mechanisms of alcohol-induced liver injury. Many pathways participate in mechanisms as to how ethanol induces oxidative stress, including induction of CYP2E1, mitochondrial dysfunction, activation of MAP kinases, elevated cytokine formation, elevated expression of the inducible isoform of nitric oxide synthase (iNOS), and decreases in hepatic antioxidant defense [3–7]. Induction of CYP2E1 contributes to the alcohol-mediated hepatotoxicity in cultured cell models such as ethanol-sensitive E47 HepG2 human hepatoma cells and RALA hepatocytes transduced with ethanol-inducible CYP2E1, and in the intragastric ethanol infusion model of ALD [8–10]. However, there are reports that CYP2E1 may not play a role in alcohol liver injury based on studies with gadolinium chloride or CYP2E1 knock-out mice [11,12]. Therefore, there is a need to further understand the role and significance of CYP2E1 induction via ethanol exposure so as to explore the mechanisms of ALD and provide a basis for preventing ALD complications and severe liver damage.
The Fas/Fas ligand system may play a central role in ethanol-induced hepatic apoptosis [13–15]. Galle, et al  showed up-regulation of Fas ligand mRNA expression in hepatocytes of alcohol-damaged liver. Apoptosis induced by low concentrations of ethanol in HepG2 cells was associated with Fas-receptor activation and subsequent caspase-8 and caspase-3 activation . Minana, et al  showed that ethanol and acetaldehyde caused apoptosis in hepatocytes after chronic ethanol feeding and that acetaldehyde elevated Fas ligand levels; they suggested that Fas may play a role in ethanol-induced apoptosis. However, Nakayama, et al  reported that in HepG2 cells, ethanol-induced apoptosis was not mediated via Fas or TNF-α receptors, and that ethanol did not up-regulate Fas expression. Hepatocytes were recently shown to express both Fas and Fas ligand which can induce death of other co-cultured cells, and it is interesting to speculate that apoptosis by autocrine or paracrine mechanisms involving Fas may play a role in alcohol-induced liver injury . Pyrazole treatment to induce CYP2E1 potentiated Fas-mediated liver injury via elevated oxidative and nitrosative stress .
Our previous data showed that acute ethanol pretreatment potentiated Fas agonistic antibody Jo2-induced hepatic toxicity by elevating CYP2E1-dependent oxidative/nitrosative stress, suggesting that acute ethanol treatment increases sensitivity to a low-toxic Jo2 challenge resulting in severe synergistic liver damage . Other studies reported that fatty liver or obesity or chronic ethanol consumption cause increased sensitivity to hepatotoxins such as LPS [22,23], suggesting the multifactorial nature and complex interactions among primary mechanistic factors and between primary and secondary factors as the basis for elucidation of mechanisms of ALD . In this study we evaluated whether chronic ethanol consumption potentiates the hepatotoxicity of Jo2 Fas agonistic antibody. We fed wild type mice containing CYP2E1 and CYP2E1 knockout mice with ethanol chronically, followed by subliminal Jo2 antibody treatment to evaluate whether an increased hepatotoxicity involves CYP2E1.
Animal experiments were approved by the Laboratory of Animal Care and Use Committee of the Mount Sinai School of Medicine and carried out under the guidelines of the National Institutes of Health for the humane use of animal subjects. CYP2E1 knockout (−/−) mice, Cyp2e1 tm1Gonz, weighing 17.8±2.7 g at 6–8 weeks of age, were kindly provided by Dr. Frank Gonzalez, National Institutes of Health and were bred and maintained in the animal center at Mount Sinai. The wild-type (CYP2E1+/+) 129/SV mice, weighing 16.9±1.6 g at 6–8 weeks of age, were purchased from Charles River Breeding Laboratory (Boston, MA). The CYP2E1 −/− (KO, n=16) and +/+ (WT, n=16) mice were pair-fed a liquid ethanol diet (ETOH) or control dextrose diet (Dex) (Bio-Serv, Frenchtown, NJ) at ethanol calorie of 10%, 15%, 20%, 25% each for 5 days, respectively and 35% for 10 days. After the 4 weeks feeding period, body weights of the dextrose-fed wild type and CYP2E1 knockout mice were 18–20 g (gains of 2–3 g body wt) while body weights of the ethanol-fed wild type and CYP2E1 knockout mice were 14–15 g (loss of body wt of about 2–3 g). Mice were withdrawn from ethanol and food and received intraperitoneally one dose of 0.1 µg of Jo2 anti-Fas antibody (Jo2) (BD Pharmingen, San Diego, CA) or saline control (Sal). At 8 hours after administration of Jo2 or saline, the mice were sacrificed and serum and liver samples were collected. Blood ethanol levels were similar in the wild type and the CYP2E1 knockout mice fed ethanol. In some experiments assaying the time course of injury, ethanol or dextrose-fed wild-type (CYP2E1+/+) 129/SV mice were sacrificed at 2 (n=4), 4 (n=4) and 8 (n=4) hours after administration of Jo2 or saline.
Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were assayed using a diagnostic kit (ThermoDMA, Louisville, CO). The serum TNF-a level was assayed using a mouse TNF-α ELISA kit (Pierce Biotichnology, Rockford, IL). A total of 50 µl of Standards (standard curve concentration: 0, 50, 350 and 2450 pg/ml) or serum were added into each well in duplicate. The absorbance was measured at 450 minus 550 nm and the results were calculated using the standard curve.
Small pieces of liver tissue were fixed in 10% neutral buffered formalin and processed into paraffin sections for hematoxylin-eosin (HE) or immunohistochemical staining. Morphological changes were observed by two pathologists who were blinded from the experimental information. All pathological changes of fatty degeneration, apoptosis and necrosis by HE staining and light microscope were graded as none (0), mild (<25%), moderate (25–50%), and severe (>75%). Immunohistochemical staining was performed by using the IHC Select Detection System (Chemicon, Temecula, CA) for MDA with rabbit-anti-MDA antibody (1:100). Slides were visualized with 3,3-diaminobenzidine (DAB) and positive staining was reflected by the brownish yellow color. In each case, a negative control (non-immune serum) was used. The evaluation of a specific positive reaction was marked as negative (−), weakly positive (+), moderately positive(++), and strongly positive (+++).
Liver tissue homogenates were freshly prepared in 5–10 volume of ice-cold 150 mM KCl. Microsomes, mitochondria and the cytosol fractions were prepared using differential centrifugation. The protein concentration of the different fractions was determined using a protein assay kit based on the Lowry assay (BioRad, Hercules, CA). The levels of CYP2E1, FAS, cFLIP, JNK and P38 MAPK protein in 50 µg of protein samples from freshly prepared microsome, cytosol or homogenate fractions were determined by Western blot analysis with anti-CYP2E1 (1:10000) (a gift from Dr. Jerome Lasker, Hackensack Biomedical Research Institute, Hackensack, NJ), anti-FAS (1:2000) (Santa Cruz Biotechnology, Santa Cruz, CA), anti-cFLIP (1:1000) (Stressgen Biotechnology, British Columbia, Canada), anti-p-JNK and anti-JNK (1:1000), anti-p-P38 and anti-P38 antibody (1:1000) (Santa Cruz Biotechnology), respectively. All specific bands of proteins detected by Western blot were quantitated with the Automated Digitizing System (ImageJ gel programs, version 1.34S, National Institute of Health).
CYP2E1 activity was measured in liver microsome fractions by the spectrophotometric analysis at 510 nm of the oxidation of p-nitrophenol to p-nitrocatechol in the presence of NADPH and oxygen . Protein carbonyl adducts were assayed in liver homogenates using 20 µg of protein samples and the OxyBlot Protein Oxidation Detection Kit (Chemicon, Temecula, CA). The DNP-derivatized protein samples were separated by SDS-PAGE followed by Western blot as described above. Reduced non-protein thiols, mainly glutathione (GSH) was analyzed in liver tissue homogenates by a fluorescence assay with the proluminescent substrate o-phthalaldehyde.
Caspase-8, caspase-3, caspase-6 and caspase-9 activities were determined in liver tissue homogenates by measuring proteolytic cleavage of the proluminescent substrates Z-IETD-AFC, AC-DEVD-AMC, AC-VEID-AFC and AC-LEHD-AFC (Calbiochem, La Jolla, CA). The fluorescence was determined based on the amount of released AFC (caspase-8, −6, −9, λex=400, λem=505) or AMC (caspase-3, λex=380, λem=460). The results were expressed as arbitrary units of fluorescence (AUF) per milligram of protein.
In order to assess the possible protection against Jo2 hepatoxicity by JNK or P38 MAPK inhibitors--- SP600125 (Sigma, St Louis, MO) or SB203580 (Tocris, Ellisville, MO), wild type 129/SV mice were fed the liquid ethanol diet or control dextrose diet at total ethanol calorie of 10% and 20% for 10 days, respectively and 35% for 10 days. Mice received intraperitoneally one dose of 0.1 µg of Jo2 anti-Fas antibody plus SP (15 mg/kg, body wt) (n=6) , or Jo2 plus SB (15 mg/kg, body wt) (n=6) or Jo2 plus saline control (n=6). At 8 hours after administration of Jo2 plus SP or plus SB or saline control, the mice were sacrificed and serum and liver samples were collected for detection of serum transaminase, liver morphology or p-JNK/JNK and p-P38/P38 MAPK protein levels as described above.
Values reflect means±SD. One-way ANOVA (post-hoc test Tukey HSD comparisons) analysis was performed by SPSS 10.0 statistic software. Statistical data were corrected for multiple group comparison. P values of less than 0.05 were considered statistically significant.
Eight groups of mice were studied in this report. Wild type mice were fed dextrose or ethanol and after 4 weeks treated with either saline or Jo2; these are referred to as WT Dex/Sal, WT Dex/ Jo2, WT ETOH/Sal and WT ETOH/Jo2. Similarly, CYP2E1 knockout mice were fed dextrose or ethanol and after 4 weeks treated with either saline or Jo2; these are referred to as CYP2E1 KO Dex/Sal, CYP2E1 KO Dex/Jo2, CYP2E1 KO ETOH/Sal and CYP2E1 KO ETOH/Jo2. Treatment with Jo2 elevated ALT and AST levels in dextrose-fed WT mice compared to saline treated dextrose-fed mice. A similar increase by Jo2 was found in CYP2E1 KO mice fed dextrose (Fig.1A,1B). Thus Jo2 causes some liver injury in dextrose-fed mice by a CYP2E1-independent pathway. In ethanol-fed mice, Jo2 administration produced a high increase in serum ALT and AST levels compared to saline treated ethanol-fed mice. This large increase by Jo2 was blunted in the CYP2E1 KO mice (Fig.1A,1B). Elevated steatosis and macrovesicular fat were observed in the WT ETOH mice treated with either saline or Jo2 (Fig. 1C3, C4) compared to the CYP2E1 KO ETOH mice treated with either saline or Jo2 (Fig.1 C7, C8). More severe pathological changes were observed in the WT ETOH/Jo2 (Fig.1 C4) than that in the CYP2E1 KO ETOH/Jo2 (Fig.1 C8) group; in the WT ETOH/Jo2 group, many hepatocytes displayed extensive eosinophilic necrosis, hemorrhage and infiltration of inflammatory cells in the central zone of the hepatic lobule. Jo2 treatment produced some hepatocyte degeneration or focal necrosis in both the dextrose-treated wild type and CYP2E1 knockout groups compared to the saline-treated WT dextrose and KO dextrose groups (Fig.1 panels, C2 and C6 compared to C1 and C5); however, the injury by Jo2 in the WT dextrose-fed mice was much less than that in the WT ethanol-fed mice (C2 compared to C4). Thus, chronic ethanol feeding potentiated Jo2-induced liver injury in WT mice but not in CYP2E1 KO mice.
Experiments were carried out to evaluate whether the enhanced hepatotoxicity found in the WT ETOH/Jo2 group was associated with an elevated formation of reactive oxygen species and whether CYP2E1 contributes to such an increase. The level of protein carbonyl adducts was higher in the WT ETOH/Jo2 or WT ETOH/Sal group compared to WT Dex/Jo2 or WT Dex/Sal group (Fig.2A). The ethanol feeding also elevated protein carbonyls in the CYP2E1 knockout mice compared to the dextrose fed KO mice (Fig.2A). This suggests that ethanol elevates carbonyl adducts by a CYP2E1-independent mechanism. Nevertheless, greatest levels of protein carbonyls were found in the ethanol-fed wild type mice, suggesting that CYP2E1 contributes to the increase in protein carbonyl adducts by chronic ethanol feeding with or without Jo2 administration. The Jo2 treatment did not elevate protein carbonyl levels in any group; indeed, carbonyls were less in Jo2-treated than saline-treated groups e.g. WT Dex and WT ETOH. The reason for this decrease by Jo2 is unknown. MDA was assayed as one index of lipid peroxidation; MDA expression in situ was mainly found in areas of steatosis or in macrovesicular fat in the centrilobular zone or portal area of the liver. The MDA levels were highest in the WT ETOH/Jo2 or WT ETOH/Sal group (+++) (Fig.2B3, B4); smaller increases were also found in all the other groups (+) (Fig.2B2,2B6,2B7,2B8) except for the lack of obvious positive staining in the WT Dex/Sal (Fig.2B1) or KO Dex/Sal group (Fig.2B5). These results, similar to the protein carbonyl adduct data, show that ethanol treatment elevates MDA levels, CYP2E1 contributes to this increase, but Jo2 treatment does not further elevate MDA levels over the chronic ethanol increase.
After 4 weeks of chronic ethanol feeding, the oxidation of PNP, a reflection of CYP2E1 catalytic activity, was significantly higher in the WT ethanol-fed mice treated with either saline or Jo2 compared to the WT dextrose-fed mice treated with either saline or Jo2 (Fig.3A). In the CYP2E1 KO ETOH/Sal or KO ETOH/Jo2 group, the PNP activity was very low, validating the absence of CYP2E1 (Fig.3A). The slight increase in PNP oxidation in the ethanol-fed CYP2E1 KO mice compared to dextrose-fed mice might reflect ethanol induction of other CYPs capable of oxidizing PNP e.g. CYP3A. Increases of 1.7-fold of CYP2E1 protein expression were detected by western blot analysis in the WT ETOH/Sal group compared to the WT Dex/Sal group, while the CYP2E1 level in the WT ETOH/Jo2 group was 3.8-fold higher than the CYP2E1 level in the WT Dex/Jo2 group. This appears to be due to a decrease in CYP2E1 by Jo2 in the WT Dex (Fig.3D). Thus, Jo2 administration does not alter the increase in CYP2E1 activity or even increases CYP2E1 protein levels about 2-fold after ethanol feeding. In the CYP2E1 KO groups, PNP activity or CYP2E1 protein expression could barely be detected compared to the WT groups (Fig.3A, 3D). Administration of Jo2 lowered the GSH level in the WT dextrose-fed and ethanol-fed mice compared to the WT saline-treated dextrose or ethanol-fed mice, respectively (Fig.3C). However, Jo2 did not lower hepatic GSH levels in the CYP2E1 knockout mice (Fig.3C), analogous to the low extent of increase in protein carbonyls and MDA by ethanol in CYP2E1 KO mice.
The content of TNF-α in serum was markedly higher in the WT ETOH/Jo2 group compared to the WT Dex/Jo2 group or WT ETOH/Sal group (Fig.3B). Importantly, it was found that the serum TNF-a level was significantly lower in the CYP2E1 KO ETOH/Jo2 group compared to the WT ETOH/Jo2 group (Fig.3B). Jo2 treatment lowered the level of cFLIP in both the WT and the KO group compared to the Sal-treated mice in both groups (Fig.3D). In the WT ETOH/Jo2 and the KO ETOH/Jo2 groups, the cFLIP level was significantly lower than that in the WT Dex/ Jo2 groups (Fig.3D). The decline in cFLIP level was similar in the WT ETOH/Jo2 group and the KO ETOH/Jo2 group (Fig.3D), thus appears to be independent of CYP2E1. There was no difference in Fas levels in all groups either in the WT mice or in the KO mice (Fig.3E). Thus, the enhanced hepatotoxicity found in the WT ETOH/Jo2 group is not due to altered levels of Fas.
The activity of caspase-8 was significantly higher in the Jo2-treated group compared to the Sal-treated group either fed the dextrose diet or the ETOH diet. There was no significant difference in caspase-8 activity in the Jo2-treated WT dextrose or ethanol-fed mice compared to the Jo2-treated dextrose or ethanol-fed CYP2E1 KO mice (Fig.4A). Similarly, the activity of caspase-3 was significantly higher in the Jo2-treated WT dextrose or ethanol-fed mice compared to the Sal-treated groups. There was also no significant difference in caspase-3 activity in the WT Dex/Jo2 or WT ETOH/Jo2 group compared to the KO Dex/Jo2 or KO ETOH/Jo2 group (Fig.4B). Activity of caspase-6 or caspase-9 was not significantly different between all the groups or between WT versus KO groups (Fig.4C, 4D).
Immunoblot assay showed that the ratio of phosphorylated p-JNK to total JNK was elevated 1.9-fold in the Dex/Jo2 groups but only 1.2-fold in the ETOH/Jo2 compared to the ETOH/Sal group (Fig.5A). Note that JNK was activated by the chronic ethanol feeding even in the absence of Jo2: 1.7-fold ratio of p-JNK/JNK in the ETOH/Sal group compared to a ratio of 1.0 in the Dex/Sal group (Fig.5A). There were increases of 1.5-fold and 1.3-fold in the ratio of phosphorylated p-P38 to total P-38 MAPK in the Dex/Jo2 or ETOH/Jo2 group compared to the Dex/Sal or ETOH/Sal group, respectively (Fig.5A). Thus, Jo2 activated JNK and p38 MAPK in both the dextrose-fed mice and alcohol-fed mice. Time course experiments indicated that JNK was activated at 2 h in the ethanol-fed mice compared to the dextrose-fed mice (Fig.5B). The JNK activation began to decline at 8 h after Jo2 treatment in ethanol-fed mice and began to increase in the dextrose-fed mice (Fig.5B). P38 MAPK was activated only at 8 h after administration of Jo2 in the ethanol-fed mice, in contrast to the early activation of JNK by Jo2.
A time course for Jo2 hepatotoxicity was carried out. Levels of serum ALT or AST were significantly elevated in the ethanol/Jo2 (and to a small extent in the dextrose/Jo2) group at 8 h but not at 2 or 4 h after Jo2 administration (Fig.6A,6B). Thus, the time point where elevated hepatotoxicity is observed in the ethanol-fed mice treated with Jo2 occurs at times where JNK and a lesser extent p38 MAPK are activated.
To initiate studies on the significance of the elevation of JNK or P38 MAPK in the enhanced toxicity of chronic ethanol feeding plus Jo2 treatment, the ability of SP600125 or SB203580 to protect against this toxicity were carried out. After administration of SP, the ratio of phosphorylated p-JNK to total JNK was significantly lower in the ETOH or Dex group at 8 h after Jo2 treatment (Fig.5B). Similarly, the ratio of phosphorylated p-P38 to total P38 MAPK was also significantly lower in the ETOH group 8 h after administration of SB (Fig.5B). Levels of serum AST were significantly lowered in the ETOH/Jo2/SP or ETOH/Jo2/SB group compared to the ETOH/Jo2 group (Fig.6D). The level of serum ALT was also lowered in the ETOH/Jo2/SP or ETOH/Jo2/SB group compared to the ETOH/Jo2 group but because of variability, p values were 0.05<p<0.1 (Fig.6C). Histopathology showed that severe injury was observed in the ETOH/ Jo2 group at the 8 h time point (Fig.6E3) including fatty accumulation, extensive eosinophilic necrosis and some infiltration of inflammatory cells as compared to 2 or 4 h after Jo2 administation (Fig.6E1, E2). There was a partial amelioration of liver injury, including limited necrotic focus, in the ETOH/Jo2/SP (Fig.6E4) or ETOH/Jo2/SB (Fig.6E5) group compared to the ETOH/Jo2 group (Fig.6E3).
CYP2E1 is induced under a variety of pathophysiological conditions such as fasting, diabetes, obesity and high-fat diet [26–28]. CYP2E1 protein expression and activity are elevated after ethanol exposure [29,30]. CYP2E1 is responsible for metabolizing and activating a large number of chemical solvents, industrial monomers and precarcinogens  and is active in catalyzing the production of ROS and lipid peroxidation in microsomal membranes [32–34]. In vitro, ethanol-mediated production of ROS, cell death and GSH depletion were observed in both the cytosolic and mitochondrial fraction of CYP2E1 transfected cell lines [8,35]. The in vivo role of CYP2E1 in alcohol-induced liver injury is controversial, especially in the early stage of experimental ALD. In this study, we used a chronic ethanol feeding plus Jo2 treatment model in wild type mice and in CYP2E1 knockout mice to evaluate whether chronic ethanol treatment increases Jo2 hepatotoxicity, whether CYP2E1 or CYP2E1-mediated ROS contributes to an increased liver injury formed after chronic ethanol consumption, and what key factors participate in this potentiated hepatotoxicity. Hepatotoxicity occurs in the wild type mice fed the alcohol diet chronically and treated with Jo2 compared to the wild type mice fed dextrose and treated with Jo2 or CYP2E1 knockout mice fed the dextrose or alcohol diet followed by treatment with Jo2. Steatosis was more severe in the CYP2E1 wild-type mice than that in the CYP2E1 knockout mice after alcohol feeding and treatment with either saline or Jo2. We have previously shown that CYP2E1 plays a role in ethanol-induced fatty liver . Interestingly, no significant toxicity was found in the saline treated ethanol-fed groups, suggesting that steatotitic histopathological changes alone did not promote further liver injury. However, the fatty liver may provide a basis for a second toxic challenge e.g. with Jo2. CYP2E1 protein and catalytic activity were highly induced in wild type mice compared to the CYP2E1 knockout mice fed alcohol and treated with either saline or Jo2 treatment. These results show that a) chronic ethanol feeding produced more steatosis in wild type mice than in CYP2E1 knockout mice; b) the ethanol feeding elevated CYP2E1 in wild type mice; c) the increase in fatty liver and in CYP2E1 are not sufficient by themselves to result in significant liver injury i.e. necrosis; d) however, the increases in fatty liver and CYP2E1 may set the stage for the liver to become sensitive to a second challenge as provided by Jo2 administration at concentrations which do not cause liver injury to dextrose-fed mice without fatty liver or elevated CYP2E1.
Experiments were subsequently carried out to evaluate possible mechanisms which play a role in the elevated Jo2 toxicity in the ethanol-fed WT mice. Since CYP2E1 contributes to this enhanced toxicity, our initial hypothesis was that oxidative stress was elevated in the ethanol plus Jo2-treated wild type mice. Indeed, protein carbonyl, MDA or lipid peroxidation and GSH depletion were highly elevated in the wild type mice fed ethanol and treated with Jo2, as compared to the CYP2E1 knockout mice fed ethanol and treated with Jo2. However, similar increases in these markers of oxidative stress also occur in the saline-treated ethanol-fed mice (without Jo2 treatment) as compared to saline-treated ethanol-fed CYP2E1 knockout mice. These results show that CYP2E1 plays a major role in the elevation of oxidative stress produced by chronic ethanol feeding. However, since this elevation was not different between the saline-ethanol-fed mice which do not show liver injury, and the Jo2-ethanol-fed mice which do show liver injury, differences in the extent of oxidative stress alone can not explain the liver injury formed in the ethanol/Jo2 mice compared to ethanol/saline mice (although differences in oxidative stress can contribute to the liver injury in the ethanol/Jo2 mice compared to the dextrose/Jo2 mice).
Abnormal cytokine metabolism is a major feature of ALD, especially LPS mediated-hepatitis [8,37,38], and increased serum TNF-α levels are associated with the disease severity and mortality of alcoholic hepatitis . Rats after chronic alcohol feeding are more sensitive to the hepatotoxic effects of administration of LPS and have higher plasma levels of TNF-α than control rats [23,40,41]. Gadolinium chloride was found to protect against alcoholic liver injury via down-regulation of TNF-α production in Kupffer cells . Anti-TNF-α antibody prevented alcohol liver injury in rats , and mice lacking the TNF-R1 receptor did not develop alcohol liver injury . These studies clearly indicate TNF-α as a major risk factor for the development of alcohol liver injury. Could differences in TNF-α levels be important for the liver injury found in the ethanol/Jo2 group compared to the other groups ? There was a small increase in serum TNF-α levels in wild type mice and CYP2E1 knockout mice fed the dextrose diet followed by Jo2 treatment as compared to the wild type or knockout mice fed the dextrose diet and treated with saline. Thus, Jo2 elevates TNF-α comparably in the absence or presence of CYP2E1 in dextrose fed mice. Chronic ethanol feeding alone did not elevate serum TNF-α levels in wild type or knockout mice compared to the dextrose controls, showing that fatty liver or elevated CYP2E1 are not sufficient to increase TNF-α in this model. However, Jo2 elevated serum TNF-α levels to high levels in wild type mice fed ethanol; this increase was not observed in the ethanol/Jo2-treated knockout mice indicating a role for CYP2E1 in the ethanol/Jo2-mediated elevation of TNF-α. CYP2E1 is present in hepatic Kupffer cells and is inducible by ethanol . How CYP2E1 and Jo2 may interact to elevate production of TNF-α will require further study. Interestingly, the extent of TNF-α elevation is associated with the severity of hepatotoxicity being highest in CYP2E1 wild-type mice fed alcohol which display liver injury, and low in the 3 other groups which do not show liver injury. These results suggest that the elevation of TNF-α may contribute to the increased hepatotoxicity in the wild type ethanol/Jo2 mice.
The Fas/Fas-L system may play a key role in ethanol-induced hepatic apoptosis [13–15]. Fas recognizes Fas ligand (FasL) or Fas antibody to initiate the receptor cross-linking and cell apoptosis via receptor oligomerization and recruitment of the Fas-associated protein with death domain, which eventually leads to the activation of caspase-8 and downstream caspases such as caspase-3 [45–47]. Liver is particularly sensitive to Fas-mediated toxicity and the injection of Fas antibody in mice results in extensive hepatocyte necrosis and even fulminant liver failure . The current study shows that Jo2 Fas antibody-mediated toxicity was potentiated by chronic alcohol feeding with development of extensive acidophilic necrosis, hemorrhage and infiltration of inflammatory cells in damaged areas. Levels of Fas were similar in all groups. While there were increases of caspase-8, and caspase-3 activities in the alcohol/Jo2 mice, there were similar increases in the dextrose/Jo2 mice, thus elevated caspase activities cannot explain the differences in liver injury between the alcohol/Jo2 versus the dextrose/Jo2. There was no significant difference in the Jo2 elevation of caspase 3 or 8 activities between wild type mice and CYP2E1 knockout mice, thus the Jo2 elevation of caspase 3 or 8 activities is independent of CYP2E1 status. c-FLIP has been identified as a regulator of death ligand-induced apoptosis downstream of death receptors and FADD [49,50]. c-FLIP was decreased after alcohol feeding plus Jo2 treatment compared to the dextrose feeding plus Jo2 treatment. However, there was no significant difference in downregulation of c-FLIP between wild type alcohol/Jo2 mice with liver injury and CYP2E1 knockout alcohol/Jo2 mice without liver injury. These results suggest that the enhanced hepatotoxicity found in the CYP2E1 wild type mice fed alcohol followed by Jo2 treatment is not due to the altered levels of Fas, the upregulation of caspase-8, −3 and the decreased levels of cFLIP. These results do not rule out an important role for any of the above in the overall injury, but appear to rule out an important role for any of the above in explaining the differential liver injury produced by Jo2 in the ethanol-fed mice.
Mitogen activated protein kinases (MAPKs), JNK, p38 and ERK, are important cellular signaling molecules that convert various extracellular signals into intracellular responses via serial phosphorylation [51–54]. Basically, prolonged activation of either JNK or p38 MAPK promotes cell death with an associated decrease in mitochondial membrane protential, whereas ERK activation may serve as a cell survival factor [55–57]. But the exact role of MAPK in the cellular response in different hepatotoxicity models remains controversial. As to the role of MAPK in CYP2E1–dependent toxicity, Hoek and colleagues found that ethanol increased TNF-α toxicity in CYP2E1 expressing liver cells via a p38 MAPK-dependent pathway . Czaja and colleagues showed that increased CYP2E1 expression sensitized hepatocytes to TNF-α toxicity through prolonged activation of JNK . CYP2E1 plus arachidonic acid toxicity was mediated either by ERK or p38 MAPK [59, 60]. Koteish, et al  reported that chronic ethanol exposure potentiated LPS liver injury in mice despite inhibiting JNK activity whereas Chung, et al  reported that six months of ethanol treatment in rats increased JNK activation. With respect to the role of MAPK on Fas-induced liver damage, Brunner, et al reported that TRAIL-induced amplification of Fas-induced liver damage required JNK activation and phosphorylation of its downstream target and proapoptotic Bcl-2 homolog Bim . We evaluated the possible role of MAPK on the ethanol/Jo2 enhanced liver injury. Jo2 alone was able to promote the activation of JNK and p38 MAPK in both the ethanol fed and dextrose fed mice; there was no further increase in activation of JNK or p38 MAPK in the ethanol/Jo2 mice which display elevated hepatotoxicity. In order to explore the exact role and dynamic changes of JNK or p38 MAPK activation on the enhanced hepatotoxicity, time course experiments and inhibitors experiments were carried out. The results showed that chronic ethanol feeding was able to significantly promote the activation of JNK at an early stage after Jo2 treatment, with activation reaching a peak value at 2 h followed by gradual decline at 4 and 8 h. A time course of liver injury in the ethanol-fed mice showed injury occurring at 8 h but not at 2 or 4 h after addition of Jo2. Thus, the liver injury occurs after the activation of JNK and it is possible that the early activation of JNK in the ethanol/Jo2 mice may play a role in the developing liver injury. Treatment with inhibitors of JNK or p38 MAPK partially but not completely prevented the liver injury in the ethanol/Jo2 mice. These results suggest that JNK or p38 MAPK activation, at least partially, contribute to the enhanced Jo2-mediated hepatotoxicity following chronic alcohol feeding, however, detail mechanisms of the enhanced hepatotoxicity remain to be further defined.
In conclusion, chronic alcohol feeding leading to induction of CYP2E1 and to liver steatosis potentiates Fas Jo2-mediated hepatotoxicity in mice. Chronic alcohol feeding produces extensive steatosis and over-expression of CYP2E1 in liver, events which promote the early stage of ALD. CYP2E1 plays an important role in the steatosis, mediating oxidative stress and lipid peroxidation, events independent of Jo2 administration. Jo2 activates Fas internalization and increases activation of caspases 8 and 3, activation of JNK and P38 MAPK, and decreases GSH, events independent of CYP2E1. The combination of alcohol induction of CYP2E1 plus Jo2 administration leads to an increase in TNF-α levels. A scheme to summarize the enhanced hepatotoxicity and events envisioned to occur in the chronic alcohol diet feeding plus Jo2 treatment model is shown in Fig.7. Neither of the factors associated with the induction of CYP2E1 (fatty liver, oxidative stress) or addition of Jo2 (activation of MAPK, caspase activation, decline in GSH) are sufficient to promote significant liver injury. However, the combination of alcohol, CYP2E1 and Jo2 is sufficient to promote liver injury. Thus, the enhanced hepatotoxicity is associated with the over-expression of CYP2E1, increase of TNF-α production and upregulation of MAPK activation especially JNK.
This work was supported by USPHS Grants AA 017425 and AA 03312 from the National Institute on Alcohol Abuse and Alcoholism. We thank Dr. Frank J. Gonzalez (Laboratory of Metabolism, National Cancer Institute, Bethesda, MD) for providing breeder CYP2E1 knockout mice used for this study and Dr. Syllvan Wallenstein (Biostatistician, Department of Community and Preventive Medicine, Mount Sinai School of Medicine, New York, NY) for help with the statistics in the manuscript preparation.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.