Induction of cytochrome P450 2E1 increases hepatotoxicity caused by Fas agonistic Jo2 antibody in mice

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Cytochrome P450 2E1 (CYP2E1) may be a central pathway in generating oxidative stress, reactive oxygen species, and causing hepatotoxic injury by alcohol and various hepatotoxins. This study evaluated the ability of CYP2E1 to potentiate or synergize the hepatotoxicity of Fas in vivo. C57BL/6 mice were injected intraperitoneally with pyrazole (Pyr) to induce CYP2E1. Then, 16-hour fasted mice were administered agonistic Jo2 anti-Fas antibody ip. Other mice were treated with Pyr or Jo2 alone. Levels of serum aminotransferase were 8.3- and 6.3-fold higher in the Pyr/Jo2 group compared with Jo2 alone, respectively. Histological evaluation of liver showed more extensive acidophilic necrosis and severe pathological changes in the Pyr/Jo2-treated mice. DNA fragmentation and caspase-8 and -3 activities were more elevated in the Pyr/Jo2 group compared with Jo2 alone. CYP2E1 activity and protein levels were higher in the Pyr/Jo2 group than in Jo2 alone. Levels of inducible nitric oxide synthase, 3-nitrotyrosine protein adducts, malondialdehyde, and protein carbonyls were also higher in the Pyr/Jo2 group compared with Jo2 alone. Glutathione and activities of catalase and Cu-Zn superoxide dismutase were decreased in the Pyr/Jo2 group. Administration of chlormethiazole, an inhibitor of CYP2E1, to the Pyr/Jo2-treated mice caused a significant decrease of alanine aminotransferase and liver pathological changes in association with a decrease in CYP2E1 protein and activity. In conclusion, enhanced hepatotoxicity of Fas was found in mice with elevated levels of CYP2E1. We speculate that overexpression of CYP2E1 might synergize and increase the susceptibility to Fas induced-liver injury. (HEPATOLOGY 2005;42:400–410.)

Alcoholic liver disease includes alcoholic fatty liver (steatosis), alcoholic hepatitis, and alcoholic cirrhosis and its complications: intrahepatic cholestasis, hepatocellular carcinoma, and iron overload.1 Alcoholic liver disease is a result of complex pathophysiological events involving various types of liver cells and injurious factors such as oxidative stress, nitrosative stress, apoptotic-inducing factors, endotoxins, and cytokines.2 The liver is an important target of the toxicity of drugs, toxins, xenobiotics, and oxidative products. The ethanol-induced cytochrome P450 2E1 (CYP2E1) plays a key role in metabolism and activation of toxic substrates, such as ethanol, carbon tetrachloride, acetaminophen, and N-nitrosodimethylamine.3–6 CYP2E1 may be an important determinant of human susceptibility to toxicity and carcinogenicity of industrial and environmental chemicals. CYP2E1 may be a central pathway in oxidative stress, production of reactive oxygen species, and hepatotoxic injury, especially in the presence of CYP2E1 inducers.7–9

Hepatic apoptosis has been shown to occur in both experimental and clinical alcoholic liver injury and acute liver failure.10, 11 There are two major pathways leading to hepatic apoptosis: (1) the exogenous pathway, which is mediated by death signal and receptor systems including Fas/Fas ligand and tumor necrosis factor (TNF)/TNF receptor; and (2) the endogenous pathway mediated by intracellular stress signals including mitochondrial cytochrome c release.12, 13 The Fas/Fas ligand system plays a central role in ethanol-induced hepatic apoptosis.14–16 Chronic ethanol feeding to rats for 7 weeks did not alter Fas ligand messenger RNA expression, whereas treatment with lipopolysaccharide increased Fas ligand content to similar extents in hepatocytes and Kupffer cells of the alcohol-fed and pair-fed control rats.17 Apoptosis induced by low concentrations of ethanol in HepG2 cells was associated with Fas-receptor activation and subsequent caspase-8 activation.18

In view of these reported ethanol–Fas interactions, and the contributions that CYP2E1 plays in alcoholic liver disease and other drug-induced liver injury, including nonalcoholic steatohepatitis,19 it is of interest to evaluate whether CYP2E1 contributes or potentiates Fas-mediated liver injury. The overall objective of this study was to explore the susceptibility and possible synergistic effect of CYP2E1 overexpression to Fas antibody hepatotoxicity and assess the involvement of CYP2E1 in the increase of hepatotoxicity of Fas antibody-induced liver injury following pyrazole pretreatment in mice. These studies may provide an experimental model to better understand the mechanism of ethanol-induced liver damage.

Abbreviations

CYP2E1, cytochrome P450 2E1; Pyr, pyrazole; TNF, tumor necrosis factor; Sal, saline; ALT, alanine aminotransferase; AST, aspartate aminotransferase; iNOS, inducible NO synthase; 3-NT, 3-nitrotyrosine; TUNEL, terminal deoxynucleotidyl transferase–mediated dUTP nick-end labeling; GSH, glutathione; SOD, superoxide dismutase; CMZ, chlormethiazole.

Materials and Methods

Animals and Treatments.

Experimental animals were male C57BL/6 mice (20-23 g, 6-8 weeks of age) purchased from Charles River Breeding Laboratory (Boston, MA) and housed in a facility approved by the American Association for Accreditation of Laboratory Animal Care. Experiments were performed with approval of Mount Sinai's Animal Use and Committee. Mice were divided into saline (Sal), Jo2 alone (Jo2), pyrazole (Pyr) alone, and combined administration of Jo2 plus Pyr (Pyr/Jo2) groups, respectively (four groups of 8-10 mice each). Mice were injected intraperitoneally with Pyr (Sigma, St. Louis, MO), 120 mg/kg body weight, once a day for 2 days to induce CYP2E1. After 16-hour fasting (for maximal CYP2E1 induction), mice were administered intraperitoneally with Sal or with agonistic Jo2 hamster anti-mouse Fas monoclonal antibody (BD Pharmingen, San Diego, CA), 0.2 μg/g body weight.20 A Jo2 concentration of 0.2 μg per gram body weight or about 4 μg per mouse was found to initiate mild toxicity, so this was the dose of Jo2 chosen for the subsequent experiments.

At 8 hours after administration of Jo2 or saline, mice were bled from the retro-orbital venous sinus for measurement of serum aminotransferases and TNF-α. The liver was rapidly excised, and specimens were immediately cut into small fragments and placed in fixative for histopathological and immunohistochemical assessment. The remaining liver samples were immediately frozen in liquid nitrogen and stored at −70°C in aliquots for preparation of homogenates and further use. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured using a diagnostic kit (ThermoDMA, Louisville, CO) and kinetically following changes in absorbance at 340 nm. The serum TNF-α level was assayed using a mouse TNF-α ELISA kit (Pierce Biotechnology, Rockford, IL).21

Liver Histopathology and Immunohistochemistry.

Small liver fragments were immediately cut into 1-mm3 blocks, fixed in ice-cold 3% glutaraldehyde in phosphate-buffered saline for transmission electron microscopy (TEM) processing, or fixed in 10% buffered formalin and processed into paraffin sections for hematoxylin-eosin staining and histopathological observation. The morphological changes of liver tissues were observed by two pathologists who were blinded from the experimental information. All changes of degeneration and necrosis were graded as none (0), mild (<25%), moderate (25%-50%), and severe (>75%).

Immunohistochemical staining was performed with the ImmunoCruz Rabbit ABC Staining System Kit (Santa Cruz Biotechnology, Santa Cruz, CA) for CYP2E1 with polyclonal rabbit–anti-CYP2E1 antibody (1:200) (a gift from Dr. Jerome Lasker, Hackensack Biomedical Research Institute, Hackensack, NJ), for 3-nitrotyrosine (3-NT) protein adducts with polyclonal rabbit–anti–3-NT antibody (1:100) (Upstate USA Inc., Lake Placid, NY), and for inducible NO synthase (iNOS) with polyclonal rabbit–anti-iNOS antibody (1:200) (Chemicon, Temecula, CA). Slides were visualized with 3,3-diaminobenzidine, and positive staining was reflected by a brownish-yellow color. Immunofluorescence staining of Fas receptor in liver sections was performed using a rabbit anti-Fas antibody (1:40) (Oncogene, San Diego, CA) and an anti-rabbit–fluorescein–immunoglobulin G (1:10) (Biomeda, Foster City, CA). Positive staining was reflected by a yellowish-green fluorescence. In each case, a negative control (nonimmune serum) was used. The evaluation of a specific positive reaction was marked as negative (−), weakly positive (+), moderately positive (++), and strongly positive (+++).

Cytochrome P450 2E1 Activity and Lipid Peroxidation.

Liver homogenates were freshly prepared in a 5-10 volume of ice-cold 150 mmol/L 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 obtained from BioRad (Hercules, CA). CYP2E1 activity was measured in liver microsomes by assaying the oxidation of p-nitrophenol to p-nitrocatechol.22

The production of thiobarbituric acid reactive substances, expressed as malondialdehyde equivalents, was assayed in liver mitochondria, microsomes and total liver homogenate fractions by the spectrophotometric analysis at 535 nm of the formation of thiobarbituric acid-reactive components as previously described.23 The concentration of malondialdehyde was calculated using an extinction coefficient of 1.56 × 105 mol/L/cm and expressed as picomoles per milligram of protein.

Western Blot Analysis of Protein Expression and Carbonyls.

Levels of CYP2E1, iNOS, and Fas receptor protein in 10-50 μg of protein samples from freshly prepared microsome or cytosol fractions were determined using Western blot analysis with anti-CYP2E1 antibody (1:10,000), anti-iNOS antibody (1:2,000), and anti-Fas antibody (1:200) (Santa Cruz Biotechnology), respectively, followed by incubation with horseradish peroxidase conjugated to goat anti-rabbit immunoglobulin G (1:5,000) (Sigma). Chemiluminscence reaction using an enhanced chemiluminescence kit (Amersham Biosciences, Buckinghamshire, England) was performed for 1 minute followed by exposure to Kodak BioMax film (Kodak Industrie, France). Protein carbonyl adducts were assayed in liver homogenates using 20 μg of protein samples and the OxyBlot Protein Oxidation Detection Kit (Chemicon). The sample loading was controlled by addition of equal concentration of protein samples. All specific bands of protein adducts detected via Western blot analysis were quantitated with the Automated Digitizing System (UN-SCAN-IT gel programs, version 5.1, Silk Scientific Corp., Orem, UT).

Terminal Deoxynucleotidyl Transferase–Mediated dUTP Nick-End Labeling and DNA Fragmentation Analysis.

Apoptosis was assessed via terminal deoxynucleotidyl transferase–mediated dUTP nick-end labeling (TUNEL) assay using the ApopTag in situ apoptosis detection kit (Serological Corp., Atlanta, GA). The labeled DNA was visualized with horseradish peroxidase–conjugated anti-digoxigenin antibody. For a negative control, the terminal deoxynucleotidyl transferase enzyme was omitted from the reaction mixtures. Slides were counterstained with 0.5% methyl green. The quantitative analysis of positive nuclei with DNA fragmentation or apoptosis was performed by counting the average number of apoptotic nuclei per visual field (original magnification ×400 [25 visual fields]). The formation of a DNA fragmentation ladder was assayed via agarose gel electrophoresis. DNA was loaded onto a 1.5% agarose gel containing ethidium bromide, electrophoresed in Tris-acetate–EDTA buffer (40 mmol/L Tris-acetate, 1 mmol/L EDTA [pH 8.0]) at 80 V and photographed under UV illumination.24

Caspase Activities.

Caspase-8, -3, -9, and -6 activity was determined in liver tissue homogenates by measuring proteolytic cleavage of the proluminescent substrates (0.1 mmol/L) Z-IETD-AFC, AC-DEVD-AMC, AC-LEHD-AFC (Calbiochem, La Jolla, CA), and AC-VEID-AFC (MP Biomedicals, Aurora, OH), which are cleaved by caspase-8, caspase-3, caspase-9, and caspase-6, respectively. Fluorescence was determined with a spectrofluorometer (PerkinElmer, Wellesley, MA) based on the amount of released AFC (caspase-8, -9, -6 activities, λex = 400, λem = 505) or AMC (caspase-3 activity, λex = 380, λem = 460). The results were expressed as arbitrary units of fluorescence per milligram of protein.

Antioxidant Levels.

Glutathione (GSH) was analyzed in liver tissue homogenates via fluorescence assay with the proluminescent substrate o-phthalaldehyde (1 mg dissolved in 1 mL of 50 mmol/L sodium phosphate). Fluorescence was determined with λex = 350 and λem = 420 nm. Catalase activity was assayed by measuring the decomposition of H2O2 at 240 nm.25 The activity of Cu/Zn-superoxide dismutase (SOD) was determined by the kinetic measurement of the absorbance at 525 nm from the Vs/Vc ratio of the autoxidation rates as a function of Cu/Zn-SOD in cytosol (Calbiochem).

Evaluation of Chlormethiazole Inhibition In Vivo.

Chlormethiazole (CMZ), a drug used for alcohol withdrawal and agitation, and an efficient inhibitor of CYP2E1 activity,26 was used to try to prevent the Jo2 agonist Fas antibody–induced liver injury found in Pyr-treated mice. Mice were divided into Sal (n = 3), Pyr/Jo2 (n = 6), and Pyr/Jo2 plus CMZ (n = 16) groups, respectively. CMZ was injected intraperitoneally at a dose of 75 mg/kg body weight 30 minutes after Jo2 injection. At 8 hours after the administration of Jo2, mice were bled from the retro-orbital venous sinus for measurement of serum aminotransferases. The liver was rapidly excised and specimens were immediately cut into small fragments and placed in a fixative for histopathological observation. The remaining liver samples were immediately frozen in liquid nitrogen and stored at −70°C in aliquots for biochemical experiments.

Statistical Analysis.

Values reflect the mean ± SEM. For statistical analysis, one-way ANOVA (and subsequent post hoc comparisons) was performed by Excel 2000 data analysis toolpac. The Student t test was used for the TUNEL assays after square root calculation of positive cell counts. P values of less than .05 were considered statistically significant.

Results

Liver Pathological Changes.

In initial experiments, we first tested the effect of various doses of the Jo2 antibody alone on liver injury to select a concentration that would cause only a mild toxic effect, thereby setting the stage to try to potentiate this mild injury (via Pyr treatment to induce CYP2E1). Serum ALT and AST activities were slightly increased 8 hours after administration of 2 and 4 μg Jo2, but were more dramatically elevated after giving 6 and 8 μg of the Jo2 anti-Fas antibody per mouse (Fig. 1A). The low toxic dose of 4 μg per mouse (0.2 μg per g body weight) was chosen for all the following studies. Mice were treated with Sal, Pyr alone, Jo2 alone, or Pyr plus Jo2. Eight hours after Jo2 administration in mice pretreated with Pyr once a day for 2 days, serum ALT and AST activities were significantly higher in the combined Jo2/Pyr treatment than that in the mice given only Jo2 or only Pyr (P < .001) (Fig. 1B). These concentrations of Jo2 and Pyr were chosen because they did not produce significant toxicity by themselves. After observation under light microscopy, severe pathological changes were detected in the combined Jo2/Pyr treatment group as many hepatocytes appeared to display extensive acidophilic necrosis (apoptosis) and focal hemorrhages in the hepatic centrilobular zone or intermediate lobular zone; however, there was no marked infiltration of inflammatory cells (Fig. 2B-C compared with Fig. 2A). Electron microscopy showed severe mitochondrial swelling, endoplasmic reticulum dilation, and numerous lipid droplets in hepatocytes (Fig. 2E-F compared with Fig. 2D). Some mild lesions were observed in the mice administrated only Jo2, mainly including dilation and congestion in the sinusoid, swelling and focal necrosis of hepatocytes in the centrilobular area, and slight mitochondrial swelling and endoplasmic reticulum dilation in hepatocytes (Fig. 2A,D). Only mild swelling of hepatocytes was found after Pyr administration alone (data not shown) while the morphology of Sal control was normal (data not shown). The expression in situ of the Fas receptor was increased in the Pyr/Jo2 group compared with the other groups (Fig. 3A, panels 3 and 4 compared with panels 1 and 2). The protein levels of Fas receptor as detected via Western blot analysis were also higher in the Pyr/Jo2 group than in the other groups (Fig. 3B). These results suggest that Fas receptor expression gradually increased with increased pathological changes. The content of TNF-α in serum was higher in both the Pyr/Jo2 and Jo2 alone group compared with the Pyr group and Sal control, but there was no significant difference between the Pyr/Jo2 and Jo2 alone groups (P > .05) (Fig. 3C).

Figure 1.

Levels of serum ALT and AST after treatment with Jo2, Pyr, or Jo2 plus Pyr. (A) Effects of increasing doses of Jo2 anti-Fas antibody on serum ALT and AST activity. Mice fasted for 16 hours were treated with saline or the Jo2 antibody (1, 2, 4, 6, and 8 μg per mouse). Serum ALT and AST were measured at 8 hours after Jo2 administration (mean ± SEM for 8-10 mice). *P < .05; **P < .01 Jo2 vs. saline control. (B) Mice were pretreated with pyrazole once a day for 2 days (120 mg/kg body weight) or saline, fasted for 16 hours, and treated with Jo2 anti-Fas antibody (0.2 μg/g body weight) or saline. Serum ALT and AST were measured at 8 hours after the treatment with Jo2 anti-Fas antibody or saline. ***P < .001 for Pyr/Jo2 vs. the other groups. ALT, alanine aminotransferase; AST, aspartate aminotransferase; Sal, saline; Pyr, pyrazole.

Figure 2.

Morphological changes of liver tissues observed under (A-C) light microscopy and (D-F) transmission electron microscopy. Panels A and D are from the Jo2-treated mice; panels B, C, E, and F are from the Pyr/Jo2-treated mice. (A) Slight sinusoid dilation and congestion, swelling, and focal necrosis of hepatocytes in centrilobular zone (arrows) (hematoxylin-eosin stain; original magnification ×200). (B-C) Many hepatocytes show extensive acidophilic necrosis (apoptosis) and severe hemorrhages but no marked infiltration of inflammatory cells in hepatic centrilobular zone or intermediate lobular zone (arrows) (hematoxylin-eosin stain; original magnification ×200). (D) Limited mitochondrial swelling and endoplasmic reticulum dilation, and no marked lipid droplets in hepatocytes (arrows) (transmission electron microscopy; original magnification ×12,000). (E-F) Severe mitochondrial swelling, endoplasmic reticulum dilation, and numerous lipid droplets in hepatocytes (arrows) (panel E, transmission electron microscopy, original magnification ×12,000; panel F, transmission electron microscopy, original magnification ×3,000).

Figure 3.

Fas receptor expression and serum TNF-α contents. (A) Immunofluorescence staining of Fas receptor in liver sections. Panel A1 is from the Jo2-treated mice, panel A2 is from the Pyr-treated mice, and panels A3 and A4 are from the Pyr/Jo2-treated mice. Staining intensity was +++ for panels A3 and A4; ++ for panel A1; and + for panel A2 (arrows) (original magnification ×200). (B) Western blot analysis for expression of the Fas receptor. Numbers under the blots refer to the ratio of Fas/β-actin. (C) Serum TNF-α level was assayed using a mouse TNF-α ELISA kit. **P < .01 for Pyr/Jo2 or for Jo2 vs. saline control; #P > .05 for Pyr/Jo2 vs. Jo2-treated. TNF, tumor necrosis factor; Sal, saline; Pyr, pyrazole.

CYP2E1 Protein Expression and Catalytic Activity.

There was an approximately twofold increase in CYP2E1 activity in the Pyr alone group compared with the Sal controls (P < .01) and an approximately 1.8-fold increase in the Pyr/Jo2 group (P < .05) compared with the Sal controls (Fig. 4A). Increases of 2.9- and 2.5-fold of CYP2E1 protein expression were detected via Western blot analysis in the Pyr/Jo2 and Pyr alone groups compared with the Sal controls, respectively (Fig. 4C). Immunohistochemistry confirmed that expression of CYP2E1 in situ was markedly higher in the Pyr/Jo2 and Pyr alone groups (++-+++) compared with the Sal controls (0-+) (Fig. 4B). The highest level of CYP2E1 was in the centrilobular zone of the liver acinus, the zone in which liver toxicity was greatest. Interestingly, increases of activity and protein expression of CYP2E1 were also found in the Jo2 alone group. The mechanism for this will require future study.

Figure 4.

Activity and protein expression of CYP2E1. (A) CYP2E1 activity was measured by evaluating the oxidation of p-nitrophenol as described in Materials and Methods. **P < .01 Pyr vs. saline-treated, *P < .05 Pyr/Jo2 vs. saline-treated. (B) Immunohistochemical staining showing in situ expression of CYP2E1 in hepatocytes. Panel B1 shows Sal controls; panel B2 shows Jo2 alone; panel B3 shows Pyr alone; panel B4 shows Pyr/Jo2 administration (arrows indicate the centrilobuar zone of the liver) (original magnification ×200). (C) The levels of CYP2E1 in 10- or 30-μg microsome fractions were determined via Western blot analysis. CYP2E1, cytochrome P450 2E1; Sal, saline; Pyr, pyrazole.

Apoptotic Liver Injury.

To assess apoptosis, caspases activities, DNA fragmentation, and DNA ladder formation were determined. Both caspase-8 and caspase-3 activity was significantly higher in the Pyr/Jo2 group and the Jo2 alone group than that in Sal controls, and there was a significant difference in the Pyr/Jo2 group compared with the Jo2 alone group (P < .01, P < .05) (Table 1). The activities of caspase-9 and caspase-6 were also significantly higher in both the Pyr/Jo2 and Jo2 alone group than that in Sal controls, but there was no significant difference between the two (P > .05) (Table 1). Pyr treatment had no significant effect on any of the caspase activities. TUNEL results showed that there were more hepatocytes with positive staining nuclei in the Pyr/Jo2 group than in the Jo2 group (Fig. 5A). After quantitative analysis of TUNEL-positive nuclei, there was a significant increase in the Pyr/Jo2 group (27.20 ± 5.76) compared with the Jo2 alone group (19.8 ± 3.54) (P < .01) (Fig. 5B). Pyr treatment alone caused a small increase in TUNEL-positive nuclei compared with Sal control. DNA ladder analysis showed that a DNA ladder was observed only in the Jo2 alone and Pyr/Jo2 groups, but not the Pyr alone group (Fig. 5C).

Table 1. Activities of Caspase-8, -3, -9, and -6 in Liver Homogenates
 SalineJo2PyrazolePyrazole/Jo2
  • NOTE. Caspase-8, -3, -9, and -6 activity was determined in liver tissue homogenates by measuring proteolytic cleavage of the specific substrate as described in Materials and Methods. The results are expressed as arbitrary units of fluorescence per milligram of protein.

  • *

    P < .01,

  • **

    P < .05 Jo2 vs. saline control; pyrazole/Jo2 vs. saline control.

  • P < .01,

  • P < .05 pyrazole/Jo2 vs. Jo2 alone.

Caspase-8506 ± 1723,484 ± 1,017**783 ± 1136,767 ± 692**
Caspase-3254 ± 482,940 ± 1,267*516 ± 2305,458 ± 1,498*
Caspase-9512 ± 993,476 ± 765*452 ± 964,911 ± 1,652*
Caspase-61,592 ± 2107,605 ± 1,484*1,729 ± 5998,594 ± 673*
Figure 5.

DNA fragmentation analysis. (A) Apoptosis of liver was assessed based via TUNEL assay using the ApopTag in situ apoptosis detection kit as described in Materials and Methods. Panel A1 shows Jo2 alone; panel A2 shows Pyr alone; panel A3 shows Pyr/Jo2 administration (arrows) (original magnification ×200). (B) The quantitative analysis of positive nuclei with DNA fragmentation was performed by counting the average number of apoptotic nuclei per visual field (original magnification ×400, 25 visual fields). **P < .01 Pyr/Jo2 vs. Jo2- or Pyr-treated, #P < .01 Jo2- vs. Pyr-treated. (C) The observation of a DNA fragmentation ladder was performed by agarose gel electropherosis as described in Materials and Methods. Sal, saline; Pyr, pyrazole.

Liver Lipid Peroxidation and Protein Carbonyls.

To assess the production of lipid peroxidation and protein carbonyls, levels of thiobarbituric acid reactive substances and protein carbonyl adducts in different fractions of liver homogenates were analyzed. The amount of malondialdehyde, an end product of lipid peroxidation, was significantly higher in the mitochondrial and microsomal fractions of liver from both the Pyr/Jo2 group and the Pyr alone group than that in Sal controls or the Jo2 alone group (Table 2). The levels of protein carbonyls were also higher in the Pyr/Jo2 and Pyr alone groups than in the Sal controls or Jo2 alone group (Fig. 6B). Jo2 alone did not increase either lipid peroxidation or protein carbonyls.

Table 2. Lipid Peroxidation in Different Fractions of Liver Tissues
 SalineJo2PyrazolePyrazole/Jo2
  • NOTE. The production of thiobarbituric acid-reactive substances, expressed as malondialdehyde equivalents, was assayed as described in Materials and Methods. The results are expressed as picomoles per milligram of protein.

  • *

    P < .01,

  • **

    P < .05 pyrazole vs. saline control; pyrazole/Jo2 vs. saline control.

  • P < .01, pyrazole/Jo2 vs. Jo2 alone.

Total liver malondialdehyde163 ± 39173 ± 31377 ± 80*221 ± 112
Mitochondrial malondialdehyde690 ± 1061,232 ± 4252,549 ± 601**3,184 ± 1,609*
Microsome malondialdehyde2,090 ± 5792,569 ± 2795,833 ± 439**3,704 ± 118**
Figure 6.

Levels of 3-NT and protein carbonyl adducts. (A) Immunohistochemical staining was performed using the ImmunoCruz ABC kit for 3-NT protein adducts. In each case, a negative control (nonimmune serum) was used. Panel A1 shows Sal control; panel A2 shows Jo2 alone; panel A3 shows Pyr alone; panel A4 shows Pyr/Jo2 administration (arrows) (original magnification ×200). (B) The protein carbonyl adducts level was assayed in liver homogenates as described in Materials and Methods. In each case, a negative control (derivatization–control solution instead of the DNPH solution) and a positive control (dinitrophenylated standard protein) were used. Sal, saline; Pyr, pyrazole; NT-SM, nontreated standard molecule.

Expression of iNOS and 3-NT Protein Adducts.

NO and NO-derived products such as peroxynitrite (ONOO) may play a key role in stress-induced injuries and certain liver diseases,27 including alcoholic liver disease.28 Immunohistochemical observation showed that 3-NT protein adducts were mainly expressed in situ in hepatocytes in the central lobular zone of the liver or in the area subject to injury (Fig. 6A). The positive expression of 3-NT adducts was higher in the Pyr/Jo2 group (+++) than in the Jo2 alone group (+) (Fig. 6A). Similarly, expression of iNOS was mainly found in the centrilobular zone and nearby areas showing injury in the Pyr/Jo2 group, and iNOS levels were higher in the Pyr/Jo2 group (+++) than in the Jo2 alone or Pyr alone groups (+) (Fig. 7A). The expression of iNOS protein in liver cytosol fractions was also increased in the Pyr/Jo2 group and the Pyr alone group compared with the Sal control, or the Jo2 alone group (Fig. 7B). Jo2 had little effect on iNOS levels or 3-NT protein adducts.

Figure 7.

Levels of iNOS. (A) Immunohistochemical observation showed that the iNOS protein was mainly expressed in situ in hepatocytes in the centrilobular zone or in the surrounding vessel area. Panel A1 shows Sal control; panel A2 shows Jo2 alone; panel A3 shows Pyr alone; panel A4 shows Pyr/Jo2 administration (arrows) (original magnification ×200). (B) The levels of iNOS protein in 50 μg of protein samples from freshly prepared cytosol fractions were determined via Western blot analysis as described in Materials and Methods. Sal, saline; Pyr, pyrazole; iNOS, inducible NO synthase.

Changes of Antioxidant Levels.

GSH levels were slightly increased in the Jo2 and Pyr alone groups compared with Sal controls but were markedly decreased in the Pyr/Jo2 group (P < .01) (Fig. 8A). Catalase activity gradually decreased in all experimental groups compared with Sal controls; the lowest activity was observed in the Pyr/Jo2 group (P < .01) (Fig. 8B). Cu/Zn-SOD activities slightly but not significantly decreased in the Jo2 or Pyr alone groups; the lowest Cu/Zn-SOD activity was in the combined Pyr/Jo2 group; however, because of variability, these changes were not considered significant (P > .05) (Fig. 8C).

Figure 8.

Changes in antioxidant levels. (A) GSH was analyzed in liver tissue homogenates. The concentration of GSH was calculated from a GSH standard curve and results were expressed as nmol per mg of protein. *P < .05 for Pyr/Jo2 vs. Jo2- or Pyr-treated. (B) Catalase activity was assayed by measuring the decomposition of H2O2 at 240 nm. The activity of catalase was calculated using a formula of activity (units/mg protein) =OD240/43.6/mg protein × 103. *P < .05 for Jo2 vs. saline. **P < .01 for Pyr vs. saline. ***P < .001 for Pyr/Jo2 vs. saline. ##P < .01 for Pyr/Jo2 vs. Jo2. (C) The activity of SOD was determined as the kinetic measurement of the 525 nm absorbance from the Vs/Vc ratio of the autoxidation rates as a function of Cu/Zn-SOD in the cytosol. Results are expressed as units per mg of protein. *P < .05 for Pyr/Jo2 vs. saline control. #P > .05 for Pyr/Jo2 vs. Jo2-treated. GSH, glutathione; Sal, saline; Pyr, pyrazole; SOD, superoxide dismutase.

Protective Effect of CMZ.

The hypothesis being evaluated in this study was that induction of CYP2E1 by treatment with Pyr may potentiate or synergize with Fas to promote liver injury. It was therefore important to provide evidence that inhibition of CYP2E1 would prevent the enhanced toxicity found in the Pyr/Jo2 group. CMZ was used to inhibit CYP2E1, because this agent was shown to prevent alcoholic liver disease in the intragastric infusion model.29 Serum ALT was significantly lower in the Pyr/Jo2/CMZ group compared with the Pyr/Jo2 group (P < .001) (Fig. 9A). The activity of CYP2E1 was also lower in the Pyr/Jo2/CMZ group compared with the Pyr/Jo2 group (P < .01), and there was a 50% decrease in CYP2E1 protein level after treatment with CMZ (Fig. 9B-C). Histopathological evaluation showed that there was only restricted focal necrosis and slight congestion in the hepatic centrilobular zone in the Pyr/Jo2/CMZ group (mild or moderate changes), while extensive acidophilic necrosis and severe degeneration was observed in the hepatic centrilobular zone of the Pyr/Jo2 group (Fig. 9D).

Figure 9.

Protective effect of CMZ against Jo2 plus pyrazole-induced liver damage. (A) CMZ was injected intraperitoneally at a dose of 75 mg/kg body weight 30 minutes after Jo2 injection into Pyr-treated mice. Serum ALT was measured at 8 hours after the addition of Jo2 anti-Fas antibody. ***P < .001 Pyr/Jo2/CMZ vs. Pyr/Jo2. (B) CYP2E1 activity was measured in liver microsome fractions via oxidation of p-nitrophenol. **P < .001 Pyr/Jo2/CMZ vs. Pyr/Jo2. (C) The levels of CYP2E1 in 50 μg of microsomal protein were determined via Western blot analysis. (D) Liver pathology. Panel 1 shows extensive acidophilic necrosis and severe degeneration of hepatocytes in the hepatic centrilobular or intermediatelobular zone in mice 8 hours after Jo2 administration following Pyr pretreatment for 2 days (arrows) (hematoxylin-eosin stain, original magnification ×200). Panel 2 shows restricted focal necrosis and slight congestion in the hepatic centrilobular zone in Jo2/Pyr-treated mice after administration of CMZ (arrows) (hematoxylin-eosin stain, original magnification ×200).

Discussion

The biochemical and toxicological effects of CYP2E1 have been studied in HepG2 cells engineered to express this cytochrome P450 and in cultured hepatocytes from pyrazole-treated rats with high levels of CYP2E1.30 To further explore these biological effects in vivo, the current study was designed to evaluate the synergistic toxicity and potential susceptibility of mice induced to express CYP2E1 to Fas antibody-induced liver damage.31 Hepatotoxicity significantly increased in the Pyr/Jo2 group compared with the other groups. There were also low hepatotoxic levels in the Pyr alone group. Subtoxic administration of Pyr was able to potentiate the hepatotoxicity caused by suboptimal administration of Jo2 antibody. To produce at least a moderate CYP2E1 induction with the least toxicity of the inducer itself, we adjusted the concentration of Pyr administration to 120 mg/kg body weight in this experimental model.

Binding of Fas to its ligand or anti-Fas antibody results in receptor cross-linking and apoptosis of Fas-positive cells. Growing evidence suggests that the Fas/Fas ligand system is one of the most important signaling pathways mediating activation of caspases and apoptosis in the liver. The Fas receptor is constitutively expressed in hepatocytes, while Fas ligand is expressed in activated T lymphocytes. Fas-mediated apoptosis may occur in an autocrine or paracrine fashion via a soluble form of the ligand. Fas-mediated killing is not restricted to lymphocytes but may also occur in nonlymphoid cells such as hepatocytes (e.g., after bile duct ligation).32–34 The intracellular pathway of apoptosis includes receptor oligomerization and recruitment of the Fas-associated protein with death domain, which eventually leads to the activation of caspase 8 and downstream caspases.35–37 Cell death stimulated by agonist Fas antibodies is characteristic of apoptosis, suggesting that the lethal effects are a result of interaction of the antibody with a functional Fas antigen as opposed to complement-mediated lysis.21, 38 Apoptosis was produced in the Jo2-treated mice, and the apoptosis was significantly higher in the Jo2/Pyr group as reflected by TUNEL assays of DNA fragmentation and caspase-8 and -3 activities. The Pyr treatment did not cause DNA fragmentation or increase activities of caspases. The toxicity found after Pyr/Jo2 treatment did not appear to reflect an inflammatory response as significant infiltration of inflammatory cells into the liver was not observed. TNF-α levels were elevated in the Pyr/Jo2 and Jo2 groups, so this cytokine may be important for the Jo2 toxicity and the Jo2 contribution to the overall toxicity in the Pyr/Jo2 group. Toxic interactions between CYP2E1 and TNF-α have been reported.39, 40 Further studies on Kupffer cell activation and cytokine production are planned in these models. That Jo2 causes apoptosis and increases TNF-α but does not evoke an inflammatory response is in agreement with a previous report.41 These results suggest that Fas antibody was able to synergize with CYP2E1 to directly result in the death of hepatocytes expressing the Fas receptor.

CYP2E1 plays a role in ethanol or hepatotoxin-induced oxidative stress and lipid peroxidation.42, 43 Increased lipid peroxidation and protein oxidation appear to be critical features in ethanol-induced liver injury.8–10, 44, 45 Nitrosative stress also causes cellular dysfunction, DNA damage, and cell death.46, 47 Peroxynitrite, formed by the rapid reaction of NO and superoxide, produces nitrated tyrosine protein adducts, eventually resulting in the development of tissue necrosis, which can be correlated with nitration of tyrosine.48, 49 These data suggest that peroxynitrite may play a critical role in the development of hepatotoxicity. The levels of iNOS, 3-NT protein adducts, malondialdehyde, and protein carbonyl formation were higher in the Jo2/pyr group than in the Jo2 alone group. This suggests that oxidative and nitrosative stress occur to a greater extent in the Pyr/Jo2 group. Future studies will evaluate the possible protection by antioxidants, ONOO scavengers, and inhibitors of iNOS on the liver toxicity found in the Pyr/Jo2 group. We speculate that the increase in iNOS, which produces NO, and the increase in CYP2E1, which produces superoxide, sets the stage for the generation of ONOO and subsequently formation of 3-NT protein adducts.

Modulation and adaptation of antioxidants and antioxidant enzymes is critical to protect cells against oxidant stress. Levels of several antioxidants were elevated in HepG2 cells overexpressing CYP2E1, and this was shown to be an important adaptation to protect the cells against CYP2E1-dependent oxidant stress.30 Levels of antioxidants and antioxidant enzymes are gradually decreased because of their depletion in reacting to the early oxidative stress and scavenging of activated reactive oxygen species. Levels and activities of GSH and Cu-Zn-SOD did not decrease in the Jo2 or Pyr alone groups compared with Sal controls but did decrease 2- to 5-fold in the Jo2/Pyr group, which may reflect antioxidant mechanisms to remove CYP2E1 plus Jo2-derived oxidants. Such decreases may contribute to the developing toxicity in the Pyr/Jo2-treated mice. Future experiments will evaluate the ability of antioxidants and iNOS inhibitors to prevent this toxicity.

CYP2E1 levels and activities were increased in the Pyr and Pyr/Jo2 mice, and the Fas-induced liver pathology correlates with CYP2E1 levels and elevated lipid peroxidation. The histopathological localizations of liver injury were in agreement with the distributions of CYP2E1, iNOS, and 3-NT protein adducts. CMZ protected the liver from free radical damage by inhibiting CYP2E1 activity, thereby ameliorating pathological changes and attenuating fibrosis.50 There was a significant decrease in ALT, CYP2E1 activity, and protein level, as well as prevention of the liver pathology in the Pyr/Jo2 mice treated with CMZ. CMZ could effectively prevent the increase of Jo2 Fas antibody-induced hepatotoxicity following pretreatment with Pyr, most probably by inhibiting the activity and overexpression of CYP2E1 in the liver.

In conclusion, the CYP2E1 inducer Pyr could potentiate the hepatotoxicity caused by the Jo2 agonist Fas antibody, suggesting that overexpression of CYP2E1 might contribute to the synergy and susceptibility of the liver to Fas-induced injury. Potential mechanisms involved in the potentiated toxicity include elevated oxidative stress, nitrosative stress, lipid peroxidation, and apoptosis. Thus, induction of CYP2E1 can synergize with other hepatotoxins such as Fas to promote liver injury. It is interesting to speculate that some of the interactions between alcohol and Fas may involve the alcohol-inducible CYP2E1, but this has yet to be validated.

Ancillary