ToxSci Advance Access originally published online on April 15, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Toxicological Sciences 73, 220-234 (2003)
Copyright © 2003 by the Society of Toxicology
BIOTRANSFORMATION AND TOXICOKINETICS |
Type 1 Diabetic Mice Are Protected from Acetaminophen Hepatotoxicity
* Department of Toxicology, School of Pharmacy, The University of Louisiana at Monroe, Monroe, Louisiana 71209;
Department of Pharmaceutical Sciences, College of Pharmacy, University of Connecticut, Storrs, Connecticut 06269;
Arkansas Children’s Hospital Research Institute, Department of Pharmacology and Toxicology, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72202; and
Pathology Associates International, National Center for Toxicological Research, Jefferson, Arkansas 72079
Received November 21, 2002; accepted February 7, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Streptozotocin (STZ)-induced diabetic (DB) mice challenged with single ordinarily lethal doses of acetaminophen (APAP), carbon tetrachloride (CCl4), or bromobenzene (BB) were resistant to all three hepatotoxicants. Mechanisms of protection against APAP hepatotoxicity were investigated. Plasma alanine aminotransferase, aspartate aminotransferase, and liver histopathology revealed significantly lower hepatic injury in DB mice after APAP administration. HPLC analysis of plasma and urine revealed lower plasma t1/2, increased volume of distribution (Vd), and increased plasma clearance (CLp) of APAP in the DB mice and no difference in APAP-glucuronide, a major metabolite in mice. Interestingly, covalent binding of 14C-labeled APAP to liver target proteins; arylation of APAP to 58, 56, and 44 kDa acetaminophen binding proteins (ABPs); and glutathione (GSH) depletion in the liver did not differ between nondiabetic (non-DB) and DB mice in spite of downregulated hepatic microsomal CYP2E1 and 1A2 proteins in the DB mice, known to be involved in bioactivation of APAP. Compensatory cell division measured via 3H-thymidine pulse labeling and immunohistochemical staining for proliferating cell nuclear antigen (PCNA) indicated earlier onset of S-phase in the DB mice after exposure to APAP. Antimitotic intervention of liver cell division by colchicine (CLC) after administration of APAP led to significantly higher mortality in the DB mice suggesting a pivotal role of liver cell division and tissue repair in the protection afforded by diabetes. In conclusion, the resistance of DB mice against hepatotoxic and lethal effects of APAP appears to be mediated by a combination of enhanced APAP clearance and robust compensatory tissue repair.
Key Words: covalent binding; CYP2E1; diabetes; species differences; tissue repair.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Acetaminophen (APAP) causes fulminant hepatic necrosis in humans and experimental animals (Jollow et al., 1973
). APAP is bioactivated to N-acetyl-p-benzoquinoneimine (NAPQI), predominantly via hepatic cytochrome P450 2E1 (CYP2E1) and to minor extent by cytochrome P450 1A2 (CYP1A2; Zaher et al., 1998). There is general agreement that NAPQI-induced covalent binding and protein arylation, combined with depletion of cytosolic and mitochondrial glutathione (GSH), are pivotal in triggering the events culminating in loss of cellular homeostasis leading to liver injury (Jollow et al., 1973). At least 23 cytosolic, mitochondrial, and microsomal target proteins that are covalently modified after APAP treatment have been identified using matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS; Qiu et al., 1998). These include the cytosolic 58, 56, and 44 KDa acetaminophen binding proteins (ABPs), arylation of which shows substantial correlation with hepatic necrosis (Bartolone et al., 1988; Birge et al., 1990). Several studies have established the critical role of liver cell division and compensatory tissue repair response in determining the ultimate outcome of toxicity viz. survival or death (Chanda and Mehendale, 1996; Dalu and Mehendale, 1996; Mehendale, 1995; Rao et al., 1996; Soni and Mehendale, 1998).
Diabetic rats are highly sensitive to several hepatotoxicants, including carbon tetrachloride (CCl4), bromobenzene (BB), thioacetamide (TA), and chloroform (CHCl3; El-Hawari and Plaa, 1983; Hanasono et al., 1975; Wang et al., 2000a; Watkins et al., 1988). Earlier mechanistic studies focused on changes in bioactivation and metabolism of toxicants in relation to increased sensitivity to hepatotoxicants in diabetes. Compensatory tissue repair, which is markedly compromised in diabetic rats, is also a major factor leading to higher sensitivity of diabetic (DB) rats to TA (Wang et al., 2000a, 2001). In contrast, DB mice are in fact resistant to TA hepatotoxicity and survive a normally lethal challenge of TA (Shankar et al., 2003), revealing a major species difference. Diabetic C57BL6 mice exhibited considerably less liver injury after TA administration over 0 to 120 h, accompanied by an augmented repair response.
The objective of the present studies was to characterize the response of Type 1 diabetic mice to hepatotoxins other than TA and to identify mechanisms that render DB mice resistant to APAP hepatotoxicity. Here we report remarkable resistance of DB mice to several mechanistically different hepatotoxicants. Since in lethality experiments maximum protection was observed against APAP hepatotoxicity, we investigated possible mechanisms of this protection against APAP hepatotoxicity. Several mechanisms have been shown to protect from APAP hepatotoxicity. Broadly these mechanisms deal with inhibition of total or specific isozymes of P450s (Mitchell et al., 1973a; Wang et al., 1996; Zaher et al., 1998), enhanced detoxification via nontoxic conjugation (APAP-glucuronide or sulfate; Hazelton et al., 1986), supplementation with GSH or other sulfhydryl compounds (Corcoran and Wong, 1986; Corcoran et al., 1985), enhanced antioxidant status (Jaeschke, 1990; Kyle et al., 1987), or preplacement of cell division (Chanda et al., 1995; Shayiq et al., 1999). In our studies we have systematically examined changes in bioactivation and disposition of APAP, followed by the compensatory tissue repair response. Diabetes is known to modulate several P450 isozymes, including those involved in bioactivation of APAP (CYP2E1 and 1A2). Induction of these isozymes in Type 1 DB rats is thought to contribute to the higher bioactivation mediated liver injury of several hepatotoxicants. However, it is not clear if a similar modulation occurs in DB mice. Hence, we have examined the hypothesis that changes in these P450s in DB mice may mediate the protection from APAP-induced toxicity.
Our findings suggest that protection in the DB mice is not due to altered bioactivation but is mediated via a combination of enhanced clearance and augmented compensatory liver tissue repair after APAP-induced liver damage.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals and chemicals.
Male Swiss Webster mice (27–33 g) were obtained from Harlan Sprague-Dawley (Indianapolis, IN). Male Sprague-Dawley rats (250–300 g) were obtained from our central animal facility. Animals were housed in our central animal facility over sawdust bedding (Harlan Sani-Chips®, Harlan Sprague-Dawley, Indianapolis, IN) in a controlled environment (temperature 21 ± 1°C, humidity 50 ± 10%, and 12 h light/12 h dark, light cycle) and were acclimatized for a week before use in experiments. Animals had access to unlimited water and rodent chow (Harlan Teklad Rat Chow No. 7001, Madison, WI). Animal maintenance and treatments were conducted in accordance with the National Institute of Health Guide for Animal Welfare, as approved by Institutional Animal Care and Use Committee (IACUC). All chemicals and biochemical kits were obtained from Sigma Chemical Co. (St. Louis, MO). Perchloric acid and scintillation fluid were obtained from Fisher Scientific (Fair Lawn, NJ). [3H-Methyl] thymidine (3H-T; 2 Ci/mmol) was purchased from Moravek Biochemicals (Brea, CA).
Induction of diabetes.
Mice were made diabetic with a single dose of streptozotocin (STZ, 200 mg/kg, ip; Gaynes and Watkins, 1989; Jeffery et al., 1991; Rerup and Tarding, 1969); rats were also made diabetic with STZ injection (60 mg/kg, ip; Wang et al., 2000a) in 0.01 M citrate buffer (pH 4.3, 10 ml/kg). Vehicle control (nondiabetic group) mice and rats received citrate buffer alone. Blood samples were withdrawn retroorbitally under diethyl ether anesthesia to monitor plasma glucose, and diabetes was confirmed by plasma glucose values (>250 mg/dl) using Sigma kit no. 315-100. Ten days were allowed for complete removal of STZ (half-life of STZ is 5 min) before beginning of experiments. Glycated hemoglobin levels were determined on day 10 after STZ (or citrate buffer) treatment as an index of overall glucose levels using Sigma kit number 441-B.
Assessment of proliferating cell nuclear antigen (PCNA) and apoptosis during diabetes protocol.
At 0, 1, 3, 5, and 10 days after STZ treatment nondiabetic (non-DB) and DB mice were euthanized and blood and liver samples collected. Liver sections were fixed in formalin and processed as described below for PCNA and apoptosis immunohistochemistry.
Lethality study.
For lethality experiments, on day 10 of the diabetes protocol, non-DB and DB mice were divided into two groups each (n = 10 per group). One group of mice was treated with APAP (600 mg/kg, ip, warm basic saline, pH 8, 20 ml/kg), CCl4 (1 ml/kg, ip, corn oil, 5 ml/kg), or BB (0.5 ml/kg, ip, corn oil, 5 ml/kg). The other group received vehicle control saline (20 ml/kg) or corn oil (5 ml/kg). Mice were observed twice daily for 14 days and survival/mortality was recorded.
APAP treatment and collection of plasma and liver samples.
Assessments of plasma alanine aminotransferase (ALT), aspartate aminotransferase (AST), glucose, liver histopathology, 3H-thymidine incorporation and PCNA after APAP treatment were done in the same experiment. On day 10 after administration of STZ or citrate buffer, DB and non-DB mice received a single injection of either APAP (600 mg/kg, ip in basic saline, pH 8) or basic saline (20 ml/kg, ip) alone. Two h prior to sacrifice, 10 µCi of 3H-T (in 0.2 ml distilled water/mouse) was administered ip to each mouse in all groups. At 0, 6, 12, 24, 36, 48, 72, and 120 h after APAP treatment, blood samples for biochemical assays and glucose measurements and liver samples for pulse labeling metrics were collected from DB and non-DB mice under diethyl ether anesthesia. At 0, 6, 12, 24, 36, 48, and 72 h after APAP treatment to both groups, liver sections were collected and handled as described below for histopathology. Four mice were treated per group per time point, except for non-DB groups treated with APAP, in order to obtain an adequate number of surviving mice (n = 15 for 36, 48, 72, and 120 h). A minimum of four mice were surviving in each group at all time points.
In a separate experiment, APAP covalent binding was determined via immunochemical analysis. DB and non-DB mice were divided into two groups. One group of DB mice (n = 4) received APAP (600 mg/kg, ip in basic saline, pH 8) while the other group of DB mice received saline vehicle. Both groups of non-DB mice also received similar treatments. Animals were euthanized at 4 h after APAP or saline administration and liver samples were flash frozen in liquid nitrogen and stored at -80°C until needed for preparation of cytosolic fraction.
Plasma enzymes and glucose estimation.
Plasma was separated by addition of heparin followed by centrifugation. Plasma ALT (EC 2.6.1.2) and AST (EC 2.6.1.1) were measured as biochemical markers of liver injury using Sigma kit no. 59 UV and Sigma kit no. DG158-Kb, respectively. Glucose was measured using Sigma kit no. 315-100.
Histopathology.
Livers were surgically removed from mice under diethyl ether anesthesia. Portions of liver were taken from the left lateral lobes and washed with ice-cold normal saline (0.9% NaCl), cut into small pieces, and fixed immediately in 10% phosphate-buffered formalin for 48 h. The liver tissue was then transferred to 70% ethyl alcohol and stored until processed. The liver specimens were processed, embedded in paraffin, sectioned at 5 µm, and stained with hematoxylin and eosin (H&E) for histological examination under a light microscope. The extent of liver injury was estimated semiquantitatively and lesions scored as multifocal necrosis. Scoring was as follows: 0, no necrosis; 1, minimal, defined as only occasional necrotic cells in any lobule; 2, mild, defined as less than one third of the lobular structure affected; 3, moderate, defined as between one third and two thirds of the lobular structure affected; 4, severe, defined as greater than two thirds of the lobular structure affected; 5, more severe, defined as damage to most of the parenchyma of the liver (Wang et al., 2000a, 2001).
3H-T incorporation into hepatonuclear DNA.
DNA synthesis was measured by 3H-T incorporation into hepatonuclear DNA as described by Chang and Looney (1965). After a single ip injection of 3H-T (10 µCi in 0.2 ml distilled water/mouse) 2 h prior to sacrifice under diethyl ether anesthesia, the following procedure was used. Liver samples (0.5 g) were homogenized in 10 ml of 2.2 M sucrose on ice and centrifuged at 40,000 x g for 1 h to isolate hepatic nuclei. The nuclei were then washed, suspended in 0.25 M sucrose, and mixed with 0.5 ml of 0.6 N perchloric acid (PCA) on ice. The mixture was centrifuged at 10,000 x g for 20 min. Pellets were washed twice with 0.2 N PCA, resuspended in 3 ml of 0.5 N PCA, and heated at 80°C for 20 min. The suspension was centrifuged at 10,000 x g for 20 min. DNA was estimated by diphenylamine reaction (Burton, 1956) in the supernatant. Radiolabeled 3H-T in the supernatant was estimated using a liquid scintillation counter (Packard Instrument Co., Meriden, CT).
Pharmacokinetic parameters of APAP.
Blood (50 µl) was collected from non-DB and DB mice at 0, 2, 5, 10, 15, 30, 45, 60, 90, 120, 150, and 240 min after APAP administration (600 mg/kg, ip). To complete the entire time course once (n = 1), six mice had to be used. Three time points were collected from each mouse overlapping one time point between mice. Four such replicate time courses were performed for non-DB and DB mice. In a separate study (n = 20 and 6 for non-DB and DB mice, respectively), urine was collected at 0 and 24 h after APAP administration to non-DB and DB mice. Levels of APAP and AG were estimated using a reverse-phase HPLC analysis using 7% acetonitrile, 50 mM sodium sulfate, and 50 mM potassium phosphate buffer as the mobile phase (1 ml/min; Kulkarni et al., 2000). A C18 column (5 µm, Alltech, Deerfield, IL) was used to separate the components. APAP and AG (HPLC grade, Sigma Chemical Co.) were used as standards. AG and APAP were detected at 254 nm, with retention times of 2.6 and 6.9 min, respectively using a PDA-100 photodiode array detector (Dionex Systems, Sunnyvale, CA). Plasma half-life (t1/2) for APAP was computed using the elimination constant (ß), where t1/2 = 0.693/ß. The value of ß was estimated using a semi-logarithmic plot of time versus concentration (up to 180 min) using JMP-IN statistical software (SAS Institute, Cary, NC).
In a separate study, plasma APAP levels were quantitated using the above-mentioned HPLC assay after low dose of APAP (300 mg/kg, ip). Blood samples were drawn from non-DB and DB mice at 15, 30, 60, 90, and 120 min after APAP administration.
PCNA assay.
The PCNA immunohistochemical analysis was conducted as described by Greenwell et al.(1991). Briefly, the liver sections mounted on glass slides were first blocked with casein (0.5%) for 20 min and then reacted with monoclonal antibody (1:5000) to PCNA (PC.10, Dako Corporation, Carpentaria, CA) for 60 min. Antigen retrieval was performed by heating slides in 1% zinc sulfate solution for 6.5 min. The antibody was then linked with biotinylated goat anti-mouse IgG antibody (1:500 for 20 min, Boehringer/ Mannheim, Indianapolis, IN), which was then labeled with streptavidin-conjugated peroxidase (1:500 for 20 min, Jackson Immunoresearch, West Grove, PA). Brown color was developed by exposing the peroxidase to diaminobenzidine (one tablet in 10 ml of PBS, filtered and 3% hydrogen peroxide, Sigma Chemical Co.) for 10 min. The sections were counterstained with Gill’s hematoxylin. The nuclei of G0 cells were blue. G1 cells had light brown nuclei; S-phase nuclei stained dark brown. G2 cells showed brown cytoplasmic staining with or without brown speckling of the nucleus. M cells were identified by mitotic figures. For histomorphometric analysis, each section was observed for cells in the above gap phases of the cell cycle in 10 high power fields, as reported previously (Wang et al., 2000a).
Assessment of apoptosis.
Apoptosis was assessed via two independent methods. First, apoptotic hepatocytes were identified by morphological criteria, which included increased eosinophilic cytoplasm, darkened nucleus, and pyknotic separation of cytoplasmic membrane from neighboring hepatocytes (Herrman et al., 1999). Morphologically determined apoptotic cells were expressed as apoptotic index (number of apoptotic cells/total number of cells x 100). Secondly, apoptosis was confirmed via in situ end labeling of free 3'-hydroxyl ends generated during apoptosis using a commercial kit (ApopTag, Intergen, Purchase, NY). The sections were counterstained with Gill’s hematoxylin. Apoptotic bodies and cells appeared brown. At least 2000 cells were counted from liver sections mounted on each slide for both methods.
Preparation of microsomes for characterization of P450 isozymes.
Liver microsomes were prepared from livers of non-DB and DB, rats and mice, on day 10 after STZ (or citrate buffer) treatment by differential ultracentrifugation using the method of Chipman et al.(1979). Protein concentration of the microsomes was determined by the Bradford method using the Bio-Rad Protein Assay (Bio-Rad Laboratories, CA). Microsomes were stored at -80°C.
Analysis of CYP2E1, 1A1/1A2, and 3A proteins.
Microsomal protein (10 µg) from each animal was separated by SDS–PAGE and transferred to nitrocellulose membranes. For detecting CYP2E1, 1A1/1A2, and 3A proteins, membranes were incubated with primary antibodies against respective proteins. Primary antibodies were goat anti-rat CYP2E1 polyclonal antibody, goat anti-rat CYP1A2 polyclonal antibody (which recognizes both rat and mouse, CYP1A1 and 1A2; Gentest Corporation, Woburn, MA) and rabbit anti-human CYP3A4 polyclonal antibody (Research Diagnostics Inc., Flanders, NJ). The antibodies were cross-reactive for the corresponding mouse CYP2E1, 1A1/1A2, and 3A proteins. The blots were then further incubated with anti-goat (Sigma Biochemicals, St. Louis, MO)/anti-rabbit secondary antibody (Amersham Life Sciences, Piscataway, NJ) conjugated to horseradish peroxidase and visualized using a chemiluminescence kit obtained from Pierce Biochemicals (Rockford, IL).
Analysis of CYP2E1, 1A1/1A2, and 3A activities.
Activity of hepatic microsomal CYP2E1 was determined via formation of malondialdehyde (MDA) as a result of lipid peroxidation initiated by tricholromethyl free radical (•CCl3), a metabolite of CCl4. The formation of MDA was estimated spectrophotometrically by reaction with thiobarbituric acid (Hu et al., 1994). Since CCl4 is bioactivated only via CYP2E1, formation of MDA is specific for CYP2E1 activity. CYP2E1 activity was corroborated via p-nitrophenol hydroxylase activity (Reinke and Moyer, 1985). Ethoxyresorufin O-deethylase (EROD) activity is specific for CYP1A1 (Burke et al., 1994) and methoxyresorufin O-demethylase (MROD) as a marker of CYP1A2 (Rodrigues and Prough, 1991) were measured by following resorufin formation spectrofluorimetrically at 536 nm (excitation) and 586 (emission) using RF-5301PC scanning spectroflurometer (Shimadzu Scientific Instruments, Columbia, MD) under conditions of linearity of incubation time and protein. Cytochrome P450 3A (CYP3A) activity was measured as erythromycin N-demethylase. Erythromycin N-demethylase was measured according to the method of Werringloer (1978) with a 45 min incubation containing 12.5 mmol/l erythromycin in the presence of 0.5 mmol/l NADPH and 1 mg of microsomal protein. The rate of formaldehyde formation was determined spectrophotometrically at 412 nm using Nash reagent.
Covalent binding analyses of 14C-APAP.
Livers from non-DB and DB mice treated with 14C-APAP (600 mg/kg, ip, 4.3 mCi/mmol, 3 µCi/mouse, Sigma Chemical Co.) were flash frozen at 2 h after APAP treatment. Covalently bound 14C-APAP derived 14C label to liver macromolecules was estimated using procedure described by Jollow et al.(1973). Briefly, liver tissue (200 mg) was homogenized in 1 ml of 0.9% saline and total protein precipitated using 2 ml of 0.9 M trichloroacetic acid (TCA). Samples were centrifuged at 1000 x g for 15 min at room temperature. The supernatant was discarded and the protein precipitate was resuspended in 3 ml of 0.6 M TCA, mixed on a Vortex agitating mixer for 1 min and centrifuged at 1000 x g for 3 min. After two more washings with 0.6 M TCA (3 ml per wash), pellet was washed six times with 80% methanol (3 ml per wash). After six methanol washings radioactivity was no longer detected in the supernatant. The remaining pellet was dissolved in 2 ml of 1 M NaOH and an aliquot was used to estimate 14C using a liquid scintillation counter (Packard Instrument Co., Meriden, CT). Results are expressed as DPM/g of liver.
Assessment of Selective Protein Arylation Due to APAP
Preparation of cytosolic fraction for covalent binding studies.
Livers were homogenized (10% w/v) in a buffer containing 0.25 M sucrose, 10 mM Tris–HCl, 1 mM MgCl2, pH 7.4 (STM buffer). Homogenates were centrifuged at 9000 x g. The resultant supernatant was centrifuged at 105,000 x g for 1 h (Manautou et al., 1994, 1996). Cytosolic fractions were collected and stored at -80° C until needed. Total protein concentrations in the cytosol were determined using the Bio-Rad Protein Assay (Bio-Rad Laboratories, Hercules, CA).
Immunochemical assessment of APAP covalent binding and levels of 58 ABP.
Expression of 58-ABP (an APAP selective target protein) and APAP covalent binding to 58, 56, and 44 kDa ABPs were determined by Western analysis of cytosolic fractions. Assessment of expression of 58-ABP was done to examine if DB changes the levels of the target protein itself. Essentially the method of Bartolone et al.(1988) and Birge et al.(1989, 1990) were followed. Details of the method have also been described by Manautou et al.(1994). Briefly, proteins (30 µg per lane) were resolved by SDS–PAGE. Resolved proteins were transferred to nitrocellulose membranes. APAP-bound proteins or 58-ABP were analyzed by immunostaining with either affinity-purified anti-APAP antibody (Bartolone et al., 1988) or anti-58 antisera (Bartolone et al., 1992), respectively, followed with peroxidase-conjugated anti-rabbit IgG. Proteins were visualized using ECL Western blotting detection reagents (Amersham Life Sciences, Piscataway, NJ).
APAP-induced hepatic GSH depletion.
Total hepatic GSH was estimated in livers of non-DB and DB mice challenged with APAP (600 mg/kg, ip) at 0, 1, 2, and 12 h after APAP treatment. A commercially available kit (Cat. #703002) from Cayman Chemical Company (Ann Arbor, MI) was used. The kit is based on enzymatic recycling of GSH using GSH reductase followed by reaction of GSH with Ellman’s reagent. The colored product is read at 405 nm and quantitated using a standard curve.
Colchicine antimitotic intervention.
Colchicine (CLC) antimitotic intervention has been previously used to examine the role of cell division in ultimate outcome of hepatotoxic injury in several models (Chanda et al., 1995; Dalu and Mehendale, 1996; Kulkarni et al., 1997; Mangipudy et al., 1995, 1996; Rao and Mehendale, 1991a,b, 1993; Rao et al., 1996). Colchicine is known to exert its antimitotic effects by microtubule perturbations (Kirshner, 1978; Manfredi and Horwitz, 1984). In addition, CLC also inhibits S-phase DNA synthesis by inhibition of thymidine kinase and thymidylate synthetase (Tsukamoto and Kojo, 1989). At low doses (1 mg/kg) CLC is devoid of toxicity and does not cause any adverse effects on liver histology or on liver function (Rao and Mehendale, 1991b). On day 10 of diabetes two groups of DB mice were treated with APAP (600 mg/kg, ip, basic saline). One group received CLC (1 mg/kg, ip in saline) at 2 and 30 h after APAP treatment; the other group of diabetic mice received saline as control at the same time points. Another group of DB mice receiving saline (vehicle for APAP) was treated with two doses of CLC to control for any effects of CLC itself. Mice were observed twice daily for 14 days and survival/mortality was recorded.
Data and statistical analysis.
All Western blots were scanned using a GS-700 imaging densitometer (Bio-Rad, Hercules, CA) and quantitated using Quantity One® software (Bio-Rad, Hercules, CA). Data are expressed as means ± SE. SPSS software package version 11.0 (SPSS Inc., Chicago, IL) was used to perform all statistical tests. Chi-square test was used to analyze statistical significance between groups in lethality experiments. In all other experiments, statistical significance between DB and non-DB groups at the same time point was analyzed by Student’s t-test. Equality of variances was tested using Levene’s test for equality of variance. Statistical significance between DB and non-DB groups with respective controls (at 0 time point) was evaluated using one-way ANOVA, followed by Duncan’s multiple range test. The criterion for significance was p 
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
STZ-Induced Diabetic Mice Are Resistant to Mechanistically and Structurally Diverse Hepatotoxins
On day 10 after STZ treatment, DB mice showed significant elevation in plasma glucose (148 ± 26 in non-DB mice compared to 490 ± 37 mg/dl in DB mice) and glycated hemoglobin levels (2.2 ± 0.1 in non-DB mice compared to 5.3 ± 0.5 in DB mice). All three hepatotoxicants caused substantial mortality in the non-DB mice. APAP led to 70% mortality while CCl4 and BB caused 50 and 70% mortalities in the non-DB mice, respectively (Table 1
). In contrast, the DB mice showed lower mortality to all hepatotoxicants. No mortality resulted in DB mice treated with the same doses of APAP and CCl4 that were lethal to non-DB mice. BB treatment to the DB mice resulted in 30% mortality.
|
Temporal Assessment Revealed Significantly Lower Development of Liver Injury in the DB Mice after APAP Challenge.
Plasma ALT (Fig. 1A
) and AST (Fig. 1B) in the non-DB mice exhibited significant increase of both transaminases, which correlated with the high lethality observed in this group. The plasma ALT activity was ~6.5-fold lower in the DB mice at 12 h compared to non-DB mice consistent with the lack of any mortality in the DB mice. The transaminase activities for non-DB mice from 36 h onwards are representative of the 30% surviving mice. Liver sections stained with H&E were examined by light microscopy for multifocal necrosis, vacuolization, and inflammatory cell infiltration (Fig. 2, Table 2). Non-DB mice treated with APAP showed marked histopathology in the liver as early as 6 h. Intense accumulation of erythrocytes in the central vein characteristic of APAP-induced hepatotoxicity was evident. The necrosis and vacuolization spread toward the periportal area over time. On the other hand, DB mice treated with the same dose of APAP showed minimal necrosis and histological damage (Fig. 2, Table 2). By 72 h the surviving non-DB mice showed some mitotic figures and regression of multifocal necrosis. The diabetic mice staged greater numbers of mitotic figures earlier, consistent with earlier recovery.
|
|
|
Robust DNA Synthesis in the DB Mice after APAP Administration
S-phase DNA synthesis was higher and occurred much earlier at 6, 24, and 36 h in the DB mice compared to APAP treated non-DB cohorts (Fig. 3
). The DNA synthesis returned to basal levels in both groups by 120 h. PCNA immunohistochemical staining procedure helps to identify cells in different phases of the cell cycle. In the non-DB mice livers most hepatocytes were found in the quiescent G0 phase. After APAP treatment, hepatocytes entered the cell cycle and progressed to mitosis. However, remarkable differences were observed between the DB and non-DB mice. The DB mice exhibited significantly greater numbers of hepatocytes in S-phase as compared to the non-DB cohorts, at 36 and 48 h after APAP administration (0.65 and 2.5% in the non-DB group compared to 2.9 and 8% in the DB group, at 36 and 48 h after APAP treatment, respectively; Fig. 4). Significant numbers of hepatocytes in S-phase were seen at 72 h in the DB and non-DB mice (Fig. 4). It should be recalled that by 48 h only 30% of the non-DB mice survive, while none of the DB mice die. The DB mice exhibited earlier onset of the cell proliferation.
|
|
Plasma t1/2 of APAP Is Lower in DB Mice Along with Higher Clearance
Plasma t1/2 of APAP in non-DB mice was 61 ± 2 min compared to 40 ± 4 min in the DB mice (Fig. 5
, top; Table 3). Clearance of APAP from the plasma was also significantly increased (~twofold) in the DB mice. Surprisingly, plasma levels of acetaminophen-glucuronide (AG) were similar in both non-DB and DB mice (Fig. 5, bottom). The DB mice exhibited significant increases in urine volume (~11.2-fold) and in the fraction of dose excreted as APAP in the urine (~5.5-fold) (Table 3). Therefore, clearance of APAP was substantially increased in the DB mice even though glucuronide formation, which contributes to over 75% of APAP metabolism, was not affected.
|
|
To examine if changes in plasma t1/2 of APAP are related to higher injury in the non-DB mice, we employed a nontoxic dose of APAP (300 mg/kg, ip). At this dose, no liver injury was observed as measured by plasma ALT levels in either group (preliminary experiments). In spite of lack of injury, the DB mice showed a greater clearance of APAP from the plasma (Fig. 5A
), consistent with the hypothesis that the diabetic state is responsible for enhanced clearance of APAP.
Species Differences in CYP2E1 and 1A1/1A2 Modulation in Diabetic Mice versus Rats
Western blot analysis indicated ~fivefold induction of CYP2E1 protein (Fig. 6A) and a ~fourfold induction in CYP2E1 enzyme activity in the DB rats as compared to controls (Figs. 6B and 6C). By contrast, CYP2E1 protein decreased to 50% in the diabetic mice compared to non-DB mice (Fig. 6A). However, no significant change in the activity of CYP2E1 was evident in the DB mice (Figs. 6B and 6C). Diabetic rats showed a twofold induction in CYP1A1/1A2 proteins (Fig. 6D). EROD and MROD activities, specific for CYP1A1 and 1A2, respectively, were also increased in DB rats compared to non-DB controls (Fig. 6E). Similar to modulation of CYP2E1, DB mice showed 57% lower CYP1A1/1A2 proteins along with 35% lesser EROD and 51% lesser MROD activities, compared with respective non-DB mice (Figs. 6D, 6E, and 6F). Diabetic rats showed no change in CYP3A protein and activity (Figs. 6G and 6H). Activity of CYP3A was measured by N-demethylation of erythromycin. Diabetic mice showed decreased CYP3A protein as compared to non-DB mice (Fig. 6G). However, there was no change in the enzyme activity in the DB mice (Fig. 6H).
|
No Differences in 14C-APAP Covalent Binding between Non-DB and DB Mice Livers
Total covalent binding of 14C-APAP has been accepted to be suggestive of reactive intermediate production. Covalent binding sharply increases from 30 min to 2 h after APAP treatment and reaches a peak by 2 h (Jollow et al., 1973
). In our studies we found no differences in 14C-APAP covalent binding between non-DB and DB mice (39,902 ± 1069 and 32,676 ± 3340 DPM/g liver, respectively). These data suggest that bioactivation of APAP is similar in both non-DB and DB mice.
Arylation of APAP Target Proteins and Hepatic GSH Depletion Is Similar in Both Non-DB and DB Mice
Expression of the 58-ABP in the DB mice was similar to non-DB mice at both 0 and 4 h time points (Fig. 7A). Selective arylation of these target proteins was similar in both DB and non-DB mice treated with APAP in spite of significantly lower liver injury and histological damage in the DB mice (Fig. 7B). GSH depletion occurs subsequent to APAP administration, preceding development of necrosis. Since GSH levels have been shown to be affected early after APAP challenge, we estimated hepatic GSH levels 1 and 2 h after APAP treatment (Ruepp et al., 2002). Similar to 14C-APAP covalent binding and arylation of ABPs, depletion of hepatic GSH was also similar in both groups (Fig. 8). Consistent with earlier reports GSH depletion preceded 14C-APAP covalent binding (Jollow et al., 1973; Mitchell et al., 1973b).
|
|
Colchicine Antimitotic Intervention Results in Diminution of DB-Induced Resistance
Treatment of DB mice with CLC after APAP administration resulted in increased mortality in the DB mice (60% vs. 0%, in CLC treated group vs. CLC untreated group) suggesting a pivotal role of cell division in mediating the DB-induced resistance (Table 4
). Diabetic mice treated with saline + CLC alone did not show any mortality.
|
PCNA and Apoptotic Changes in Diabetes
PCNA in non-DB and DB mice examined at 0, 1, 3, 5, and 10 days after STZ treatment revealed that in non-DB mice, most cells were in the quiescent (G0) phase with only 0.075% cells in G2 phase of the cell cycle. On day 3 after induction of diabetes by STZ, 0.1% cells were in the G2 phase located around the centrilobular area. The number of G2 cells was higher than non-DB controls but not statistically different (Fig. 9
). However, on days 5 and 10, significantly higher numbers of G2 cells (0.45 and 1.26%, respectively) were found in the DB mice (Fig. 9). A minimum of 4000 cells was examined per group/time point.
|
No changes in apoptotic cells were seen at any time point between non-DB and DB mice (apoptotic indices of 0.05 ± 0.02 and 0.03 ± 0.01, in non-DB and DB mice, respectively). Assessment of apoptotic bodies under H&E stained liver sections was corroborated by immunohistochemical assessment of apoptosis. These results are similar to those reported by Herrman et al.(1999)
. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the present work, we report resistance of DB mice to three structurally and mechanistically dissimilar hepatotoxicants. Only a limited number of studies have employed the mouse model of diabetes. Previous studies have reported lower liver injury after APAP in DB mice (Gaynes and Watkins, 1989
; Jeffery et al., 1991). In separate studies, Watkins et al. have previously examined modulation of CCl4 hepatotoxicity of in both DB rats and DB mice and observed a species difference in hepatotoxic outcome (Gaynes and Watkins, 1989; Watkins et al., 1988). However, the mechanisms of this resistance remain to be investigated. No data were available on either temporal development of liver injury or the ultimate outcome of the toxic injury viz. survival or death in DB mice. The present study investigated both of these aspects.
APAP clearance and bioactivation were the two potential mechanisms selected for additional investigation. Firstly, our studies demonstrate 34% lower plasma half-life of APAP in the DB mice along with twofold higher clearance of APAP, presumably due to a tenfold increase in urine output in the DB mice. To test the possibility that the delay in APAP clearance in non-DB mice might be due to higher liver injury of APAP (600 mg/kg, ip), we used low dose of APAP (300 mg/kg) to examine the effect of diabetes alone on APAP clearance in the absence of any liver injury. Our data clearly suggest that the diabetic state enhances the clearance of APAP from the plasma irrespective of level of liver injury. Several changes in renal morphology and function reported in diabetes may account for this (Sharma et al., 1999). Glomerular hypertrophy, acute increase in basement membrane mass, proximal tubular changes, and microalbuminuria are some changes that occur in early stages of STZ-induced diabetes. Early stages of diabetes are also associated with increased glomerular area and hyperfiltration (Sharma et al., 1999). Polyuria due to a glucose-induced osmotic diuresis is common in patients with hyperglycemia (Spira et al., 1997). The roles of these mechanisms in protection from APAP toxicity are worthy of a more thorough investigation. The HPLC analysis in our study only presents APAP and the APAP-glucuronide data. Other metabolites such as the APAP-sulfate and the APAP-mercapturate conjugates were not measured. The APAP-glucuronide represents for about 70–75% of metabolism in the mice. The APAP-sulfate is only a minor metabolite in this species accounting for about 5–7% of the metabolism (Hazelton et al., 1986). From our data, the APAP-glucuronide and the parent compound account for ~80% of APAP metabolism. Quantitation of these metabolites might further elucidate the protection mechanisms in DB mice.
Furthermore, we investigated whether or not diabetes decreases bioactivation of APAP. Three independent indirect markers of actual reactive intermediate argue against this. First, covalent binding of 14C-APAP was not different between DB and non-DB mice. Second, selective protein arylation of cytosolic proteins has served as an indirect indicator of net availability of reactive electrophile in the liver (Bartolone et al., 1988; Birge et al., 1990; Manautou et al., 1996). Third, the degree of GSH depletion was also similar between non-DB and DB mice, further strengthening the concept that equal bioactivation occurred in both groups. Hence it appears that a decrease in CYP2E1 and 1A1/1A2-mediated bioactivation is not the mechanism by which the marked hepatotoxic protection against APAP toxicity is achieved in the DB mice.
Other examples of hepatoprotection in the absence of decreased covalent binding of APAP are available. Treatment with cimetidine (Peterson et al., 1983), dithiothreitol (Tee et al., 1986), disulfiram (Poulsen et al., 1987), and chlorpromazine (Saville et al., 1988) were shown to protect against APAP hepatotoxicity in the absence of significant reduction in covalent binding of APAP metabolite. Pretreatment with a single dose of peroxisome proliferator, clofibrate also protects against APAP toxicity without any change in selective protein arylation of target proteins or depletion of nonprotein sulfhydryl groups (Manautou et al., 1996).
Mitochondrial damage is an early event after APAP administration (Qiu et al., 1998; Ruepp et al., 2002), leading to decreased oxidative phosphorylation and energy production (Meyers et al., 1988) and decreased ATP content (Vendemiale et al., 1996). Liver mitochondria from diabetic animals are reported to be less susceptible to oxidative damage (Ferreira et al., 1999; Santos et al., 2001; Sukalski et al., 1993) and have a higher antioxidant capacity (Elangovan et al., 2000), critical in the elimination of mitochondrially generated reactive oxygen species. Diabetes also alters the mitochondrial permeability transition (MPT) due to changes in calcium uniporter functions and offers resistance of diabetic tissues to ischemia reperfusion injury (Kristal et al., 1996). Although speculative at this stage, this resistance of mitochondria from DB mice may significantly decrease initiation and progression of injury after APAP challenge without changes in arylation of cytosolic target proteins as observed in our studies.
In addition to the mechanisms that initiate toxicity, the dynamics of the compensatory tissue repair response influence the ultimate outcome of toxicity (survival or death) (Chanda and Mehendale, 1996; Mehendale, 1995; Soni and Mehendale, 1998). Timely onset of cell division and sustained continuation of the cell proliferative response are of pivotal importance for survival. Marked cell cycle changes in the present study observed in the DB mice are consistent with the early onset of tissue repair after APAP challenge. The movement of otherwise quiescent hepatocytes to G2 phase may be of critical importance to the DB mice. The number of cells in the G2 phase normally is small, yet these cells are an important defense against CCl4-induced hepatotoxicity (Calabrese and Mehendale, 1996). It appears that the greater number of cells in G2 phase in the DB mice provide the stimulus for early onset of tissue repair.
Antimitotic intervention has been shown to inhibit tissue repair and permit progression of liver injury initiated by CCl4 (Rao and Mehendale, 1993), TA (Mangipudy et al., 1996), and dichlorobenzene (Dalu et al., 1998; Kulkarni et al., 1997). Also, antimitotic intervention with CLC abolishes autoprotection by low priming doses of hepatotoxicants, against high doses of the same toxicant (Mangipudy et al., 1995; Rao and Mehendale, 1993) or heteroprotection against a different toxicant (Chanda et al., 1995). But how does absence of cell division lead to progression of liver injury, even after the toxicant has been cleared? Recent evidence suggests that this progression may be mediated via the action of lytic enzymes (e.g., calpain, phospholipaseA2, etc.) leaking from cells dying of cytotoxic injury (Limaye et al., 2002). New cells are resistant to the action of these lytic enzymes because they overexpress endogenous inhibitors of these enzymes. Hence, by replacing necrotic cells with new cells, further progression of liver injury is curbed (Limaye et al., 2002). In the absence of new cells (antimitotic action of colchicine) lytic enzymes lead to accelerated progressive injury (Mangipudy et al., 1996). In the study depicted in Table 4, progressive phase of injury (ALT) was accelerated upon CLC intervention of cell division in DB mice (data not shown).
What determines the extent of tissue repair? Two primary reasons may be considered to explain the observed increase in DNA synthesis in the DB mice. First, the lower liver injury in the DB mice might be considered as a reason for higher tissue repair. However, the tissue repair response, although initiated in response to injury, has been noted to be dependent on the physiological status of the animal and not on the initial extent of injury. For instance, diet restricted rats on treatment with TA, have markedly higher (2.5-fold higher) liver injury compared to ad libitum fed rats, but still survive due to increased tissue repair response (Apte et al., 2002; Ramaiah et al., 1998). Another illustration where the extent of initial injury does not predict the extent of tissue repair can be seen in DB rats. DB rats challenged with tenfold lower dose of TA as compared to non-DB rats attain same level of initial injury as the non-DB rats. In spite of equal initial injury, DB rats show significantly delayed tissue repair (Wang et al., 2000a).
The second possibility is that mechanisms that regulate cell division in DB mice are upregulated (Shankar et al., 2002). Although, mechanisms for the increased tissue repair remain unclear, altered hormonal regulation of tissue repair in DB mice is an attractive explanation. Growth hormone is known to accelerate hepatic regeneration. Hepatocyte growth factor gene expression and DNA synthesis in partial hepatectomized livers were both accelerated by treatment with growth hormone (Ekberg et al., 1992). Similarly, CCl4 hepatotoxicity is potentiated due to delayed compensatory liver tissue repair in the Mini rat model, a Wistar-derived transgenic rat in which the expression of growth hormone (GH) gene is suppressed by the presence of an antisense transgene (Shimizu et al., 2001). DB mice have about tenfold higher circulating levels of growth hormone (Grønbæk et al., 2002). Although speculative at this point, this marked increase in growth hormone may play a role in the increased repair response in the DB mice.
Another important finding from our studies is the species-specific modulation of P450 isozymes in DB mice versus rats. Type I diabetes induced via STZ or alloxan in rats is known to be accompanied by induction of a multitude of P450 isozymes including hepatic CYP2E1 (Barnett et al., 1994; Shimojo, 1994; Wang et al., 2000b). However, very little is known about such changes in P450 isozymes in DB mice. Similar to our findings, Sakuma et al.(2001) have also reported a species difference in the modulation of hepatic CYP2E1 and 1A2 between DB rats and mice. The mechanism of a dichotomous impact of diabetes on microsomal CYP2E1 in the two rodent species (rats vs. mice) is not known. Hypophysectomy induces CYP2E1 in rats but not in mice (Hong et al., 1990). Ketone bodies, thought to regulate CYP2E1 are increased in DB rats (Miller and Yang, 1984) but are unchanged in DB mice (Polotsky et al., 2001). These and other differences may collectively be responsible for differential modulation of CYP2E1 and other isozymes.
In conclusion, we report marked resistance of DB mice to several mechanistically and structurally diverse hepatotoxicants. Even though hepatoprotection against APAP is accompanied by downregulation of hepatic CYP2E1 and 1A2 proteins, protection against liver injury is not due to decreased covalent binding of the APAP reactive intermediate. Instead, protection appears to be due to accelerated clearance of APAP and upregulation of mechanisms that enhance early and substantial tissue repair, averting progressive injury.
|
ACKNOWLEDGMENTS |
|---|
These studies were partly supported by the Louisiana Board of Regents Support Fund through the University of Louisiana at Monroe, Kitty DeGree Endowed Chair in Toxicology.
|
NOTES |
|---|
1 To whom correspondence should be addressed at The University of Louisiana at Monroe, School of Pharmacy, Department of Toxicology, Sugar Hall 306, Monroe, LA 71209-0495. Fax: (318) 342-1686. E-mail: mehendale{at}ulm.edu.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Apte, U. M., Limaye, P. B., Ramaiah, S. K., Vaidya, V. S., Bucci, T. J., Warbritton, A., and Mehendale, H. M. (2002). Upregulated promitogenic signaling via cytokines and growth factors: Potential mechanism of robust liver tissue repair in calorie-restricted rats upon toxic challenge. Toxicol. Sci. 69, 448–459.
Barnett, C. R., Flatt, P. R., and Ioannides, C. (1994). Modulation of the rat hepatic cytochrome P450 composition by long-term streptozotocin-induced insulin-dependent diabetes. J. Biochem. Toxicol. 9, 63–69.[CrossRef][ISI][Medline]
Bartolone, J. B., Birge, R. B., Bulera, S. J., Bruno, M. K., Nishanian, E. V., Cohen, S. D., and Khairallah, E. A. (1992). Purification, antibody production, and partial amino acid sequence of the 58-kDa acetaminophen-binding liver proteins. Toxicol. Appl. Pharmacol. 113, 19–29.[CrossRef][ISI][Medline]
Bartolone, J. B., Birge, R. B., Sparks, K., Cohen, S. D., and Khairallah, E. A. (1988). Immunochemical analysis of acetaminophen covalent binding to proteins. Partial characterization of the major acetaminophen-binding liver proteins. Biochem. Pharmacol. 37, 4763–4774.[CrossRef][ISI][Medline]
Birge, R. B., Bartolone, J. B., Hart, S. G., Nishanian, E. V., Tyson, C. A., Khairallah, E. A., and Cohen, S. D. (1990). Acetaminophen hepatotoxicity: Correspondence of selective protein arylation in human and mouse liver in vitro, in culture, and in vivo. Toxicol. Appl. Pharmacol. 105, 472–482.[CrossRef][ISI][Medline]
Birge, R. B., Bartolone, J. B., McCann, D. J., Mangold, J. B., Cohen, S. D., and Khairallah, E. A. (1989). Selective protein arylation by acetaminophen and 2,6-dimethylacetaminophen in cultured hepatocytes from phenobarbital-induced and uninduced mice. Relationship to cytotoxicity. Biochem. Pharmacol. 38, 4429–4438.[CrossRef][ISI][Medline]
Burke, M. D., Thompson, S., Weaver, R. J., Wolf, C. R., and Mayer, R. T. (1994). Cytochrome P450 specificities of alkoxyresorufin O-dealkylation in human and rat liver. Biochem Pharmacol. 48, 923–936.[CrossRef][ISI][Medline]
Burton, K. (1956). A study of the conditions and mechanisms of the diphenylamine reaction for positive colorimetric estimation of DNA. Biochem. J. 62, 315–323.[ISI][Medline]
Calabrese, E. J., and Mehendale, H. M. (1996). A review of the role of tissue repair as an adaptive strategy: Why low doses are often non-toxic and why high doses can be fatal. Food Chem. Toxicol. 34, 301–311.[CrossRef][ISI][Medline]
Chanda, S., Mangipudy, R. S., Warbritton, A., Bucci, T. J., and Mehendale, H. M. (1995). Stimulated hepatic tissue repair underlies heteroprotection by thioacetamide against acetaminophen-induced lethality. Hepatology 21, 477–486.[CrossRef][ISI][Medline]
Chanda, S., and Mehendale, H. M. (1996). Role of nutrition in the survival after hepatotoxic injury. Toxicology 111, 163–178.