Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Metallothionein (MT) is a low molecular weight protein ubiquitous in the animal kingdom (Kägi, 1993). MT has an unusual amino acid composition, that is no aromatic amino acids, and one-third of its residues are cysteines. The cysteine content and the sulfhydryl motifs in MT are highly conserved. These cysteine residues bind and store metal ions, thus MT has been proposed to play an important role in essential metal, such as zinc and copper, homeostasis and in the detoxication of heavy metals, such as cadmium (Kägi, 1993, Klaassen and Liu, 1998).

Because of its high sulfhydryl content, MT has also been suggested to react with free radicals and electrophiles (Klaassen and Cagen, 1981; Lazo and Pitt, 1995). Indeed, MT can serve as a sacrificial scavenger for hydroxyl radicals in vitro (Thornalley and Vasäk, 1985), and thus protect against free radical-induced DNA damage (Abel and Ruiter, 1989; Chubatsu and Meneghini, 1993; Schwarz et al., 1995). MT can also assume the function of superoxide dismutase in yeast (Tamai et al., 1993), and protect against lipid peroxidation in erythrocyte ghosts produced by xanthine oxidase-derived superoxide anion and hydrogen peroxide (Thomas et al., 1986). MT is induced by oxidative stress-producing chemicals (Bauman et al., 1991), and has been proposed to protect against oxidative damage (Sato and Bremner, 1993) and the toxicity of alkylating anticancer drugs (Lazo and Pitt, 1995).

Acetaminophen is a widely used analgesic drug, and is biotransformed and eliminated mainly as nontoxic conjugates with glucuronic acid and sulfate (Nelson, 1995). Only a small portion (~4%) of acetaminophen is bioactivated by cytochrome P-450 yielding N-acetyl-p-benzoquinone imine (NAPQI), a reactive toxic intermediate (Dahlin et al., 1984). After an overdose of acetaminophen, both glucuronidation and sulfation are saturated (Hjelle and Klaassen, 1984) and the formation of NAPQI is increased; this produces liver injury via a chain of cellular events: 1) depletion of cellular glutathione and covalent binding to cellular proteins (Hinson et al., 1995; Cohen and Khairallah, 1997), 2) recruitment and activation of macrophages (Laskin and Pendino, 1995), 3) initiation of oxidative stress and oxidation of protein thiols (Tirmenstein and Nelson, 1990; Jaeschke, 1990), 4) alteration of calcium homeostasis, and 5) damage to nuclear DNA (Corcoran and Ray, 1992). Thus, both covalent and noncovalent interactions are involved in acetaminophen-induced hepatotoxicity.

Based on the proposed antioxidant properties of MT, we hypothesize that intracellular MT may provide an additional detoxication mechanism in protection against acetaminophen toxicity by reducing acetaminophen-induced oxidative damage. In the present study, MT-I/II null mice (Michalska and Choo, 1993; Masters et al., 1994), which are "normal" except for lack of MT protein (Michalska and Choo, 1993; Masters et al., 1994; Liu et al., 1996), were used to test this hypothesis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Homozygous MT-I/II null mice (129/Ola × C57BL/6J background) were described previously (Michalska and Choo, 1993; Liu et al., 1996). The control mice were bred to match the corresponding background (129/Ola mice from Harlan Laboratory, Indianapolis IN, were bred to C57BL/6J mice from Jackson Laboratory, Bar Harbor, ME). Animals were housed in American Association for the Accreditation of Laboratory Animal Care accredited facilities at 21 ± 1°C with a 12-h light/dark cycle. Food (Purina Laboratory Rodent Chow, St. Louis, MO) and tap water were provided ad libitum. Adult male and female mice (8-10 weeks old) were used for this study. Another colony of MT-I/II knockout mice (129/SvPCJ background; Masters et al., 1994) and their corresponding controls were also used in some studies (Zn protection, GSH depletion, and isolated hepatocytes). Both MT-null mouse colonies gave similar results.

Chemicals. Acetaminophen and NAPQI were purchased form Sigma Chemical Co. (St. Louis, MO), and 2',7'-dichlorofluorescin diacetate was obtained from Molecular Probes (Eugene, OR). Polyclonal antibody against 4-hydroxynonenal- and malondialdehyde-protein adducts were kindly provided by Dr. Petersen and Dr. Hartley (University of Colorado Health Science Center, Denver, CO). All other reagents were commercially available and of reagent grade.

Hepatotoxicity Studies. For the dose-response studies, animals were given acetaminophen (150-500 mg/kg i.p.) or saline (10 ml/kg) for 24 h. Mice were decapitated and blood was collected for the preparation of serum. The livers were removed and weighed, and a portion of liver was fixed in 10% buffered formalin (pH 7.4). Liver samples were processed by standard histopathological techniques, stained with hematoxylin and eosin, and examined for morphologic evaluation of liver injury. Biochemical evaluation of liver function was determined by measuring serum enzyme activities of alanine aminotransferase (ALT) and sorbitol dehydrogenase, using commercially available kits from Sigma (St. Louis, MO).

Acetaminophen Bioactivation Studies. Hepatic microsomal protein concentration was measured by a dye-binding method (Bradford method), using a commercial kit from Bio-Rad (Hercules, CA). Total cytochrome P-450 was determined from the carbon monoxide difference spectrum of dithionite-reduced microsomes, based on an extinction coefficient of 91 mM¹cm¹. Cytochrome b₅ was determined from the difference spectrum of NADH-reduced versus oxidized microsomes, based on an extinction coefficient of 185 mM¹cm¹. NADPH-cytochrome c reductase activity was determined at 550 nm, based on an extinction coefficient of 19.1 mM¹cm¹ for reduced minus oxidized cytochrome c. Acetaminophen metabolism was determined in bile duct-cannulated mice under pentobarbital anesthesia as described previously (Liu et al., 1993). Acetaminophen was dissolved in saline and injected into the tail vein at a dose of 150 mg/kg, and bile and urine were collected for 2 h in 30-min periods. Acetaminophen and its metabolites in bile and urine were analyzed by HPLC, and the four major metabolites, i.e., acetaminophen-glutathione conjugate, acetaminophen-cysteine conjugate, acetaminophen-glucuronide, and acetaminophen-sulfate conjugate were quantified.

Acetaminophen Covalent Binding. Control and MT-I/II null mice were treated with acetaminophen (100-300 mg/kg i.p.) or saline (20 ml/kg i.p.). Four hours after acetaminophen administration, mice were euthanized and livers were removed. Livers were homogenized in 0.25 M sucrose buffer containing 10 mM Tris-HCl (pH 7.4) and 1 mM MgCl₂, and were fractioned by differential centrifugation. Selective acetaminophen arylation of cytosolic and microsomal proteins were determined immunochemically using specific antiacetaminophen antibody (Bartolone et al., 1988). Briefly, proteins (30 µg/lane) were resolved on discontinuous 10% SDS-polyacrylamide gel electrophoresis slab gels using a 3% stacking gel, followed by transblotting onto nitrocellulose membranes. Membranes were subsequently incubated in a 1:10 dilution of affinity-purified antiacetaminophen antibody overnight at 4°C, followed by a 4-h incubation with secondary antibody (peroxidase-conjugated anti-rabbit IgG). Immunoreactive proteins were detected by using enhanced chemiluminescence Western blotting detection reagents (Amersham, Arlington Heights, IL), and visualized by exposure of Kodak XAR-5 film. The immunoreactive intensity of Western blots was semiquantified using a PDI image analyzer (Protein and DNA ImageWare Systems, Huntington Station, NY).

Hepatic Glutathione Concentration and Lipid Peroxidation. Control and MT-I/II null mice were given acetaminophen (100, 200, and 300 mg/kg i.p.). At 1, 2, 3, and 4 h after acetaminophen administration, mice were euthanized and livers removed. Glutathione concentration in livers were determined by the method of Tietze (1969). Briefly, the livers were homogenized with 3% sulfosalicylic acid (1:10, w/v) on ice, followed by centrifugation (10,000g for 10 min at 4°C). The resultant supernatant was mixed with 0.1 M sodium phosphate (pH 7.4) containing 10 mM 5,5'-dithiobis-(2-nitrobenzoic acid), 10 U/ml glutathione reductase, and 2 mM NADPH, and the change in absorbance at 412 nm was quantified. Reduced glutathione was used as the standard.

Mouse Hepatocyte Isolation. Hepatocytes were isolated from control and MT-I/II null mice by a two-stage single pass perfusion method (Zheng et al., 1996). Briefly, a calcium- and magnesium-free Hanks' solution supplemented with EGTA (0.5 mM) and Tris (25 mM, pH 7.4) was perfused through the liver via the portal vein for 15 min at 37°C. The liver was then perfused with media containing 0.05% collagenase (hepatocyte qualified; Gibco, Long Island, NY) at a flow rate of 5 ml/min for 20 min. After enzymatic digestion, the liver was removed, minced, and filtered. The parenchymal cells were separated from nonparenchymal cells and debris by low speed (50g, 1 min) centrifugation. Cell viability exceeded 80% as determined by trypan-blue exclusion. Cytotoxicity was assessed by measuring lactate dehydrogenase (LDH) leakage into the medium, using a commercial LDH kit (DG1340-K; Sigma), and oxidative stress was determined by oxidation of 2',7'-dichlorofluorescin (DCF) diacetate to DCF in the cell. The fluorescence was monitored for 30 min after addition of the 10 uM DCF diacetate into cuvettes and the change in fluorescence was recorded on Perkin-Elmer Luminescence Spectrometer LS 50B (Perkin-Elmer Cetus Instruments, Norwalk, CT).

Statistics. The .05 level of probability was used as the criteria of significance. Comparison between control and MT-I/II null mice was made by Student's t test. Comparison between two or more treatments was made by ANOVA, followed by Duncan's new multiple range test. The 2 × 2 table -square test was used for survival studies.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

MT-I/II Null Mice Are More Susceptible to Acetaminophen-Induced Lethality and Hepatotoxicity. Acetaminophen administration produced a dose-dependent lethality. MT-I/II null mice were more sensitive than control mice to acetaminophen-induced lethality. For example, acetaminophen at doses of 300 and 350 mg/kg caused 30 and 70% mortality in MT-I/II null mice, respectively, as compared with 10 and 30% mortality in control mice.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Dose-response of acetaminophen-induced hepatotoxicity in control and MT-I/II null mice. Mice were given an i.p. injection of acetaminophen (50-400 mg/kg for 24 h). The liver injury in surviving mice was measured by ALT and serum sorbitol dehydrogenase activity. Values are mean ± S.E. of 10 to 16 mice. *Significantly different from control mice (P < .05).

View larger version (160K): [in this window] [in a new window]   Fig. 2.   Photomicrographs of mouse liver sections 24 h after acetaminophen (250 mg/kg i.p.) administration. Sections were stained with hematoxylin and eosin. A, liver of a control mouse treated with acetaminophen, exhibiting swelling of parenchymal cells and sporadic necrotic injury. B, liver of a MT-I/II null mouse treated with acetaminophen, exhibiting foci of necrotic parenchymal cells. Arrows indicate necrosis. Magnitude ×200.

Acetaminophen Metabolism Is Not Altered in MT-I/II Null Mice. Hepatic cytochrome P-450, b₅ concentration, and cytochrome-c reductase activity were determined. In MT-I/II null mice, total cytochrome P-450 was about 20% lower than that of controls (570 versus 690 pmol/mg protein), while cytochrome b₅ concentration and NADPH cytochrome-c reductase activity were slightly lower, but were not statistically different than controls (data not shown). Analysis of acetaminophen metabolites in bile and urine were performed by HPLC (Table 2). No difference in biliary and urinary excretion of acetaminophen metabolites was observed between control and MT-I/II null mice after acetaminophen administration (150 mg/kg i.p., 2 h). Thus, the sensitivity of MT-I/II null mice to acetaminophen-induced hepatotoxicity does not appear to be due either to an increase in P-450-mediated acetaminophen bioactivation or a decrease in acetaminophen detoxification by phase-2 conjunction.

Acetaminophen-Induced GSH Depletion and Covalent Binding in Control and MT-I/II Null Mice. Hepatic GSH was determined by the enzymatic assay. No difference in hepatic GSH depletion was observed between control and MT-I/II null mice after acetaminophen administration (100, 200, and 300 mg/kg at 1, 2, and 4 h; Fig. 3). At a higher dose (300 mg/kg) of acetaminophen, GSH was decreased similarly ~90% in both control and MT-I/II null mice. The recover rate of cellular GSH seems to be slower in MT-I/II null mice, but was not statistically different from controls. Treatment of mice with Zn had little effect on acetaminophen-induced GSH depletion.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Hepatic concentration of glutathione in control and MT-I/II null mice 1 to 4 h after acetaminophen (100, 200, and 300 mg/kg i.p.) administration. Values are mean ± S.E. of four to six mice. *Significantly different from controls (P < .05).

View larger version (79K): [in this window] [in a new window]   Fig. 4.   Acetaminophen covalent binding to liver cytosolic 44- and 58-kDa proteins. Liver cytosols from control (MT) and MT-I/II null (MT) mice were prepared 4 h after acetaminophen (100-250 mg/kg i.p.) administration. Cytosolic protein (30 µg/lane) was separated by SDS-polyacrylamide gel electrophoresis, followed by immunoblotting with an affinity-purified antiacetaminophen antibody as described in Materials and Methods

MT-I/II Null Hepatocytes are More Susceptible to Acetaminophen and NAPQI-Induced Oxidative Stress. Hepatic lipid peroxidation after acetaminophen administration was determined via the abundance of TBARS (Fig. 5). MT-I/II null mice had more lipid peroxidation than did control mice at 1 h (2×), 2 h (3×), and 4 h (4×) after acetaminophen injection. The localization of hepatic lipid peroxidation was performed using polyclonal antibody against 4-hydroxynonenal-protein adducts (Fig. 6). MT-null mice had more positive staining around the central vein than did control mice (arrows). Immunohistochemistry using polyclonal antibody against malondialdehyde-protein adducts showed similar results (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Hepatic concentration of TBARS in control and MT-I/II null mice 1 to 4 h after acetaminophen (100 mg/kg i.p.) administration to control and MT-null mice. Values are mean ± S.E. of four to six mice. *Significantly different from controls. (P < .05).

View larger version (162K): [in this window] [in a new window]   Fig. 6.   Photomicrographs showing the immunohistochemical localization of polyclonal antibody against 4-hydroxynonenal-protein adducts. A, liver section of control mouse 4 h after 200 mg/kg acetaminophen, showing paucity of positive cells. B, liver section of MT-I/II null mice 4 h after the same dose of acetaminophen; note the abundance of positive cells around the portal vein. Arrows indicate positive straining. Magnitude ×200.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Top, representative dichlorofluoroscein fluorescence intensity graph after exposure of cultured hepatocytes to NAPQI (200 µM). More oxidative stress was consistently observed in MT-I/II null hepatocytes than controls after NAPQI (50-200 µM) exposure. Bottom, cytotoxicity of NAPQI in cultured hepatocytes isolated from control and MT-I/II null mice. Values are mean ± S.E. of four hepatocyte preparations. *Significantly different from controls.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study demonstrates that MT-I/II null mice are more susceptible than controls to acetaminophen-induced lethality and liver injury. Furthermore, induction of MT by Zn protected against acetaminophen toxicity in control mice, but not in MT-I/II null mice, suggesting that Zn-induced protection may be due to increased cellular MT. These data support the hypothesis that intracellular MT not only plays a protective role against heavy metal toxicity, but also plays an important role in protecting against organic chemical-induced liver injury (Klaassen and Liu, 1998).

Acetaminophen-induced hepatotoxicity is thought to be mediated by a cytochrome P-450-generated intermediate, NAPQI (for review, see Cohen et al., 1998). Therefore, our initial efforts were directed to determine whether MT-I/II null mice have an altered metabolism of acetaminophen. Our results on hepatic cytochrome P-450 analysis and quantification of the urinary and biliary metabolites of acetaminophen indicate that MT-I/II null mice metabolize acetaminophen similarly to controls (Table 1). No differences are found in cytochrome P-4502E1 activity between MT-I/II null and control mice (Rofe et al., 1998). Thus, the greater susceptibility of MT-I/II null mice to acetaminophen is due neither to increased production of toxic acetaminophen metabolites nor to decreased glucuronidation and sulfation of acetaminophen.

The depletion of glutathione and arylation of proteins by acetaminophen are important events leading to acetaminophen hepatotoxicity. In the present study, MT-I/II null mice were as equally susceptible as control mice to acetaminophen-induced glutathione depletion, in agreement with a recent report (Rofe et al., 1998). Further studies did not reveal any apparent differences in covalent binding of acetaminophen to cytosolic 44- and 58-kDa proteins between MT-I/II null and control mice (Fig. 4). The subcellular distribution of acetaminophen is also not altered by MT inducers, such as Zn (Chengelis et al., 1986) and oleanolic acid (Liu et al., 1993). Thus, the greater susceptibility of MT-I/II null mice does not appear to be due to the altered availability of the acetaminophen intermediate. The glutathione depletion and covalent binding may be necessary, but not sufficient, for the development of hepatocellular injury, because some antioxidants protect against liver injury produced by acetaminophen or NAPQI, without markedly preventing the depletion of glutathione (Jaeschke, 1990; Sakaida et al., 1995) or affecting covalent binding (Cohen et al., 1998).

Oxidative damage has been proposed as a contributing mechanism of acetaminophen hepatotoxicity (Mason and Fisher, 1986; Cohen et al., 1998). Acetaminophen may oxidize protein sulfhydryls in addition to forming covalent adducts (Birge et al., 1988; Tirmenstein and Nelson, 1990). After acetaminophen-induced depletion of glutathione, lipid peroxidation occurs (Adamson and Harman, 1993; Amimoto et al., 1995). In the present study, MT-I/II null mice were more susceptible than controls to acetaminophen- and NAPQI-induced lipid peroxidation, as evidenced by increased thiobarbituric acid substances, and by increased oxidative potential produced by NAPQI in vitro. The immunohistochemical localization of lipid peroxidation adducts (MDA and 4-HNE) at early time-points (4 h) correlates with necrosis of hepatocytes around the central zone, and the increased cellular oxidation of DCF correlates with cell cytotoxicity. All these data suggest that the greater susceptibility of MT-I/II null mice may be due to the increased oxidative damage produced by acetaminophen.

Oxidative stress occurs in cells when there is disruption of cellular redox balance. Acetaminophen-induced oxidative stress results in lipid peroxidation, oxidation of protein thiols, mitochondrial injury, altered calcium homeostasis, and DNA damage (Corcoran and Ray, 1992; Nelson, 1995; Cohen et al., 1998). Cellular defenses against oxidative damage include superoxide dismutase, glutathione peroxidase, catalase, glutathione, vitamin C, and vitamin E (Sies, 1993). Recently, cellular MT has been included as an important nonenzymatic antioxidant (Bray and Bettger, 1990; Sato and Bremner, 1993). MT can be oxidized in vitro by oxidative stress, and the Zn released during the process may play an important role in the cellular defense (Bray and Bettger, 1990; Maret and Vallee, 1998).

The susceptibility of MT-I/II null mice to acetaminophen hepatotoxicity may also be associated with reduced hepatic Zn concentration and energy metabolism, as hepatic concentrations of glycogen and glucose are much lower in MT-I/II null mice compared with control mice, after acetaminophen administration (Rofe et al., 1998). Disrupted energy metabolism could compromise the cellular detoxication mechanisms.

In conclusion, this study demonstrated that MT-I/II null mice were more vulnerable to acetaminophen-induced hepatotoxicity. The increased sensitivity does not appear to be due to reduced acetaminophen activation, glutathione depletion, or covalent binding, but may be associated with a loss of the antioxidant role of MT.