prevention of Acetaminophen and Cocaine Hepatotoxicity in Mice by Cimetidine Treatment

GASTROENTEROLOGY 1983;85:122-9 LIVER AND BILIARY TRACT prevention of Acetaminophen and Cocaine Hepatotoxicity in Mice by Cimetidine Treatment FRANCIS J. PETERSON, ROBERT G. KNODELL, NANCY J. LINDEMANN, and NADINE M. STEELE Research Service and Gastroenterology Section. Veterans Administration Medical Center, Minneapolis, Minnesota Hepatotoxicity occurs in animals after administration of large doses of acetaminophen and cocaine and is thought to result from production of reactive metabolites of these parent drugs by cytochrome p 45' Because cimetidine binds to cytochrome P 45 and inhibits hepatic drug metabolism in both humans and animals, we determined the effects of cimetidine coadministration on acetaminophen and cocaine hepatotoxicity in mice. Marked elevations of serum glutamic pyruvic transaminase and severe peri central hepatocellular necrosis occurred in animals receiving intraperitoneal doses of 35 mg/kg acetaminophen or 35 mg/kg cocaine, while minimal serum glutamic pyruvic transaminase elevations and liver necrosis were seen in animals who also received 1 mg/kg cimetidine 1 h befote and 1 h after administration of either acetaminophen or cocaine. Consistent with the hypothesis that these in vivo protective effects resulted from interaction with cytochrome P 45, cimetidine inhibited in vitro hepatic microsomal metabolism of cocaine. However, despite its protective effect against acetaminopheninduced hepatic injury, concomitant administration of cimetidine did not significantly affect plasma pharmacokinetics of acetaminophen, prevent depletion of hepatic glutathione after acetaminophen administration, or alter in vivo covalent binding of ehlacetaminophen to hepatic proteins. These studies suggest that current theories regarding production of acetaminophen-induced liver damage re- Received February 3, 1982. Accepted December 31, 1982. Address requests for reprints to: Robert G. Knodell, M.D., Gastroenterology Section (llldj, Veterans Administration Medical Center, 54th Street & 48th Avenue South, Minneapolis, Minnesota 55417. This work was supported by the Research Service of the Veterans Administration and National Institutes of Health Grant No. 5R1 AM26697 (RGK). 1983 by the American Gastroenterological Association 16-585/83/$3. quire reexamination. The possibility that cimetidine treatment might be useful in preventing hepatic damage due to acetaminophen and other hepatotoxins in humans is intriguing and also warrants consideration. Drug-induced hepatic damage may result from metabolic activation of parent compounds to reactive intermediates by cytochrome P 45-dependent mixed function oxidases (1,2). These reactive metabolites are usually detoxified by hepatocyte defense mechanisms such as conjugation with glutathione. However, when large quantities of toxic intermediates are produced, this protective capacity of hepatocytes can be exceeded, and excess metabolites may react with hepatic macromolecules to cause cell necrosis. Acetaminophen, a widely used anajgesic, and cocaine, a powerful central nervous system stimulant, are two drugs that are thought to produce hepatocellular necrosis after metabolic activation to toxic intermediates. It has been postulated that acetaminophen is metabolized by cytochrome P 45 to a quinoneimine, which produces hepatic damage once glutathione has been depleted (3,4). Similarly, demethylation of cocaine and subsequent metabolism of norcocaine to a reactive intermediate, possibly a nitroxide radical of N-hydroxynorcocaine, is thought to be responsible for cocaine hepatotoxicity (5,6). Cimetidine, an antagonist of histamine Hz-receptors widely used to treat peptic ulcer disease, has been reported to decrease hepatic drug metabolism in both animals and humans (7-9). The mechanism of this decrease appears to be due to interaction between cimetidine and the heme moiety of cytoc chrome P 45 with subsequent inhibition of mixed function oxidase activity (1). Therefore, we examined whether or not cimetidine pretreatment of mice would inhibit formation of reactive metabolites of acetaminophen and cocaine and prevent the hepatic

July 1983 CIMETIDINE BLOCKS EFFECTS OF HEPATOTOXINS 123 necrosis normally associated with administration of large doses of these compounds. Materials and Methods Male CD-l mice, 23-25 g (Charles River Breeding Laboratory, Wilmington, Mass.) were used in all experiments. Animals received food and water ad libitum before and throughout all experiments. Drugs were administered intra peritoneally according to the following dose schedule: acetaminophen (Sigma Chemical Co., St. Louis, Mo.), 25 mg/ml in.9% basic saline, ph 11.3, 35 mg/kg body wt; cocaine hydrochloride (Merck Co., Inc., West Point, Pa.), 2.5 mg/ml in.9% saline, 35 mg/kg body wt; cimetidine and its inactive primary metabolite, cimetidine sulfoxide (Smith Kline & French Laboratories, Carolina, Puerto Rico), 15 mg/mt 1 mg/kg body wt 1 h before and 1 h after study drug or saline injection. All experiments were begun at 2 PM. This time interval is significantly removed from the end of the nocturnal feeding period of mice, and it was important to adhere to this time schedule to obtain uniformity of experimental results. We did not extensively investigate the reasons for the above observation, but hepatic glutathione levels in our animals were noted to be 46% lower at 3 PM than at 1 AM (5.99 ±.26 vs. 8.72 ± 1.2,umol/g liver, p <.2), even though food intake was not restricted. To obtain tissue for histology and serum for liver enzyme determinations, animals were killed 24 h after injection. Serum glutamic pyruvic transaminase (SGPT) was measured using a standardized kit (Beckman Instruments Co., Fullerton, Calif.). Liver tissue was fixed in 1% buffered formalin and stained with hematoxylineosin. Histologic examination of coded slides was performed by light microscopy and graded according to the following scale: no cell necrosis, ; <5% cell necrosis, 1; 6%-25% cell necrosis, 2; 26%-5% cell necrosis, 3; >5% cell necrosis, 4. Hepatic glutathione levels were measured 2 h after acetaminophen administration using Ellman's reagent (11). Addition of cimetidine, 5 mm, did not affect glutathione determinations, and in vivo administration of either cimetidine or basic saline alone had no effect on hepatic glutathione levels. Effects of cimetidine coadministration on plasma pharmacokinetics of [3H)acetaminophen (New England Nuclear, Boston, Mass.) were studied to determine if cimetidine might exert some extramicrosomal influence on acetaminophen metabolism. After intraperitoneal administration of [3H)acetaminophen, 35 mg/kg, samples of blood were drawn from the choroid plexus of animals at 3 min, 1 h, and 2 h. Unmetabolized [3H)acetaminophen was extracted from plasma with ethyl acetate (12) and pharmacokinetic parameters were calculated from plasma concentrationtime curves using standard formulas (13). In vivo covalent binding of [3H)acetaminophen to hepatic proteins was determined according to the method of Peterson (14) 2 h after i.p. administration of a 35-mg/kg dose of acetaminophen containing 1,uCi of labeled drug. Hepatic microsomes were prepared using standard methods (15). Hepatic microsomal N-demethylation of cocaine was measured by the rate of formaldehyde formation (16), and in vitro covalent binding of [3H)acetaminophen to microsomal protein was determined (14). Student's t-test was used to determine levels of significance for differences between the various experimental groups (17). Results Representative liver histology from the various treatment groups is shown in Figure 1. Marked necrosis of central hepatocytes was seen in livers of animals receiving either acetaminophen or cocaine alone, but histology remained normal in animals who were also treated with cimetidine. Serum glutamic pyruvic transaminase levels and mean histologic scores of liver tissue obtained in the various treatment groups are shown in Table 1. Serum glutamic pyruvic transaminase levels were greatly elevated in animals receiving either acetaminophen or cocaine alone, but remained relatively normal in animals who also received cimetidine 1 h before and 1 h after administration of these compounds. Concomitant cimetidine treatment also significantly decreased hepatic necrosis scores as compared with scores seen when either acetaminophen or cocaine was administered alone (p <.1 and.5, respectively). Serum SGPT levels and hepatic histology scores seen after acetaminophen plus cimetidine sulfoxide administration (2648 ± 119 IV/ml and 2.8 ±.5, respectively, n = 6) closely approximated those seen after acetaminophen administration alone, and values for both of these parameters were significantly greater than those seen in acetaminophen plus cimetidine experiments (p <.1 and.5, respectively). If cimetidine protects against cocaine hepatotoxicity in vivo by inhibiting cocaine metabolism, it would be expected to reduce in vitro hepatic microsomal demethylation of cocaine to norcocaine. As predicted, cimetidine noncompetitively inhibited demethylation of cocaine by hepatic microsomes (Figure 2). If cimetidine protects against acetaminophen-induced hepatotoxicity by inhibiting its cytochrome P 45-mediated metabolism, in vivo depletion of hepatic glutathione should have been prevented and covalent binding of acetaminophen to hepatic proteins should have been decreased. However, animals treated with cimetidine had both the same magnitude of, and time-course for, hepatic glutathione depletion found in animals receiving only acetaminophen (Figures 3 and 4). High concentrations of cimetidine did produce a modest, but statistically significant, decrease in in vitro covalent binding of [3HJacetaminophen to microsomal proteins (Figure 5). However, results for determination of covalent binding of [3HJacetaminophen to protein of liver homogenates prepared from animals killed 2 h after

124 PETERSON ET AL. GASTROENTEROLOGY Vol. 85, No.1 Figure 1. Typical liver histology seen for various treatment groups: a, acetaminophen; b, acetaminophen plus cimetidine; c, cocaine; d, cocaine plus cimetidine. Hematoxylin-eosin stain, magnification x 4.

July 1983 CIMETIDINE BLOCKS EFFECTS OF HEPATOTOXINS Figure 1. (continued) 125

126 PETERSON ET AL. GASTROENTEROLOGY Vol. 85, No.1 Table 1. Effect of Cimetidine Treatment on Serum Glutamic Pyruvic Transaminase Elevations and Hepatic Necrosis Produced by Administration of Acetaminophen and Cocaine to Mice U Treatment group Acetaminophen (n = 13) Acetaminophen + cimetidine (n = 9) Cocaine (n = 6) Cocaine + cimetidine (n = 6) Serum SGPT b (IUlml) 3447 ± 638 55 ± 13 2866 ± 359 16 ± 2 Histologic score b 2.6 ±.3.2 ±.2 3.3 ±.7. ±. a Mean values ± SEM. b P <.5 or greater for comparison between drug given singularly and drug plus cimetidine. SGPT = serum glutamic pyruvic transaminase. acetaminophen treatment plus/minus cimetidine treatment (Table 2) show that concentrations of cimetidine obtained in vivo produced no significant change in in vivo covalent binding of [3Hlacetaminophen to hepatic protein as compared with experiments using acetaminophen alone. Concomitant cimetidine administration also had no significant effect on in vivo plasma pharmacokinetics of acetaminophen (Table 3). Discussion Results of this study suggest that cimetidine may have potential as an agent that prevents hepatic damage when administered in close proximity to exposure of the liver to some hepatotoxic agents. Cimetidine effectively prevented cocaine- and acetaminophen-induced hepatic necrosis in our mouse model. Because both of these drugs are postulated to 6 require metabolic activation by the microsomal cytochrome P 45 system to reactive intermediates before they cause hepatotoxicity (3,5), a plausible mechanism for this observed protection would be that cimetidine interferes with cytochrome P 45-mediated activation of the parent compound. Data in this report are consistent with this hypothesis as the mechanism for cimetidine's prevention of cocaineinduced hepatotoxicity. Demethylation of cocaine to norcocaine is thought to be the first step in production of the hepatotoxic intermediate of cocaine (6), and in vitro inhibition of hepatic microsomal demethylation of cocaine to norcocaine by cimetidine (Figure 2) correlates well with the marked protection against cocaine hepatotoxicity provided by cimetidine treatment in vivo. The mechanism by which cimetidine protects against acetaminophen-induced hepatic necrosis appears to be more complex. Cimetidine sulfoxide is the primary metabolite of cimetidine and has been shown to have both low affinity for cytochrome P 45 and minimal inhibitory activity of in vivo and in vitro hepatic drug metabolism (18). The inability of cimetidine sulfoxide to prevent hepatic injury certainly suggests that some property of the unmetabolized parent compound, such as interaction of cimetidine with cytochrome P 45 and inhibition of microsomal drug metabolism, is important in preventing acetaminophen-induced hepatic injury. The failure of cimetidine sulfoxide to protect against Q; >.!!! " :;: : ~.!;., c 1: " ;; (3.!'! " -., I 1 6 4 2,. ~ o 1 2 3 4 Acetaminophen (mg/kg BW) o Figure 2..5 1. 1.5 2. Cocaine (mm) 1,.1 -I I[SJ [mm Cocaine] Hepatic microsomal demethylation of cocaine in the absence (e) and presence () of cimetidine,.5 mm. Each point represents the mean velocity for that particular cocaine concentration in three separate experiments. No difference in Km between the two groups was seen; V max was 4.38 and 2.25 nmollmin. mg protein for control and cimetidine experiments, respectively. Figure 3. Hepatic glutathione content in animals killed 2 h after intraperitoneal injection of graded doses of acetaminophen alone (e) or acetaminophen + cimetidine, 1 mg/kg, 1 h before and 1 h after acetaminophen administration (6). Each point represents the mean percent ± SEM of hepatic glutathione content for groups of 4 treated animals compared with hepatic glutathione content in saline-injected control animals who received no acetaminophen. Hepatic glutathione content in control animals for these experiments was 5.23 ±.45 JLmollg liver.

July 1983 CIMETIDINE BLOCKS EFFECTS OF HEPATOTOXINS 127 hepatic injury phis the lack of effect of cimetidine on plasma pharmacokinetics of acetaminophen also suggest that the protective effect of cimetidine is not mediated through major alterations in extramicrosomal metabolism of acetaminophen such as increased availability of sulfate for acetaminophen conjugation (19). However, the failure of cimetidine to prevent hepatic glutathione depletion or decrease in vivo covalent binding of [3Hlacetaminophen to hepatic proteins does not support the hypothesis that cimetidine protects against acetaminophen-induced hepatic necrosis by inhibition of cytochrome P 45-mediated metabolism of acetaminophen. The observation that cimetidine does not affect in vivo covalent binding of [3Hlacetaminophen also rules against the possibility that cimetidine acts as a surrogate glutathione and exerts its protective influence by binding reactive acetaminophen metabolites before they react with hepatic macromolecules. Such a mechanism has been postulated to account for the protective effect against acetaminophen-induced liver damage of another sulfur-containing compound, propylthiouracil (2,2 1). Two additional mechanisms should be considered to explain the protective effect of cimetidine against acetaminophen-induced liver damage. Labadarios et al. (22) have shown that a-mercaptopropionylglycine stabilizes hepatocyte constituents against the detrimental effects of covalent binding. In an attempt to determine if cimetidine might have such a general protective effect, we examined whether or not cimetidine protected against liver injury produced by galactosamine, a hepatotoxin that does not require cytochrome P 45 interaction to produce hepatic damage. Because mice are very resistant to galactosamine-induced hepatotoxicity (23), male Sprague- 1., co 1.c c. c: E!)--=u 8 «~ :--) C 6 I ~_. U ' c: ' :> 4 o ~ m ~.?>., c: 2 ~ u.5 1. 2. 3. 5. Cimelidine (mm) Figure 5. Effect of cimetidine on in vitro covalent binding of [3Hjacetaminophen to mouse hepatic microsomal protein. Incubations contained [3Hjacetaminophen (1.2 mm) and microsomal protein (2 mg/ml) in a volume of 3 ml and were run for 2 min. Each value represents the mean of three separate determinations. Covalent binding for control experiments in the absence of cimetidine was 36.5 ± 1.2 pmollmin mg protein. Binding was Significantly decreased (p <.1 ot greater) at all cimetidine concentrations except.5 mm (p =.59). Dawley rats (Bio-Lab, st. Paul, Minn.) weighing -25 g were used. Animals were killed 24 h after a 1.5-g/kg intraperitoneal dose of galactosamine (Sigma Chemical Co.). Controls received only galactosamine, while experimental animals received 1 mgl kg cimetidine 1 h before and 1 h after galactosamine administration. Mean serum glutamic oxaloacetic transaminase (SGOT) and SGPT values in 6 animals who also received cimetidine (824 ± 17 and 212 ± 32 IU/mI, respectively) were not significantly different from those seen in 5 control animals (587 ± 269 and 437 ± 298 IU/ml, respectively), and no significant difference in hepatic histology between the two groups was demonstrated. This observation is not conclusive proof that the mechanism by whieh cimetidine prevents liver damage after acetaminophen administration is not protection of hepatocytes against injury subsequent to covalent binding of reactive acetaminophen metabolites to hepatic proteins. However, it does suggest that cimetidine does o 2 4 6 8 1 Time (hours) Figure 4. Time-course for hepatic glutathione depletion and repletion in mice receiving 35 mg/kg of acetaminophen (e) or acetaminophen + cimetidine, 1 mg/kg, 1 h before and 1 h after acetaminophen injection (.6). Each point represents the mean percent ± SEM of hepatic glutathione content in 4 separate treated animals as compared with saline-injected control animals who received no acetaminophen. Table 2. Covalent Binding of eh]acetaminophen to Mouse Hepatic Proteins 2 h After Acetaminophen Administration o Treatment Acetaminophen (n = 5) Acetaminophen + cimetidine (n = 6) a Mean values ± SEM. Covalently bound acetaminophen (nmol/mg protein).693 ±.61.652 ±.47

128 PETERSON ET AL. GASTROENTEROLOGY Vol. 85, No.1 Table 3. Effects of Cimetidine Administration on Plasma Pharmacokinetics of eh]acetaminophen in Miceo Treatment To concentration (/Lg/ml) Volume of t1l2 distribution Clearance (h) (mll1 g) (mllh 1 g) Acetaminophen (n = 3) Acetaminophen + cimetidine (n = 3) 431 ± 19 (396-461) 447 ± 75 (318-576).93 ±.4 (.84-.99).98 ±.2 (.65-1.35).75 ±.4 8 ± 5 74 ± 6 (.7-.82) (71-89) (67-85).71 ±.16 66 ± 5 64 ± 11 (.51-1.7) (59-76) (49-86) a Mean values ± SEM (range). not have a broad cytoprotective influence on hepatocytes. Finally, recent studies by Wendel et al. (24) demonstrated that large doses of acetaminophen lead to lipid peroxidation in mice once glutathione has been depleted. These investigators have suggested that this phenomenon is the basic deteriorative mechanism responsible for acetaminophen-induced liver damage. We do not have any data on the effects of cimetidine on lipid peroxidation, and the effects of this compound on lipid peroxidation in relation to its protective effect against acetaminophen-induced liver damage remain to be elucidated. Mitchell et al. (25) have reported that cimetidine protected against acetaminophen hepatotoxicity in fasted rats pretreated with 3-methylcholanthrene. They postulated that this protective effect was due to cimetidine's ability to inhibit cytochrome P 45-mediated conversion of acetaminophen to glutathionedepleting reactive metabolite(s), but no measurements of hepatic glutathione levels in animals receiving acetaminophen or acetaminophen plus cimetidine were made to support or refute this hypothesis. Although cimetidine did not prevent hepatic glutathione depletion or decrease in vivo covalent binding of acetaminophen to hepatic protein in our studies, it did provide marked protection against acetaminophen-induced hepatotoxicity. It also protected against cocaine-induced liver damage and did inhibit hepatic microsomal metabolism of cocaine. It is possible that cimetidine prevented cytochrome P 45-mediated formation of some hepatotoxic metabolite but did not inhibit formation of all acetaminophen metabolites that bind to glutathione and other hepatic proteins; inhibition of some aspect of hepatic microsomal acetaminophen metabolism still seems to be the most likely mechanism for cimetidine's protective influence. However, these studies suggest that the mechanism for production of acetaminophen-related liver damage requires further examination and may not fit the commonly held theories concerning the pathogenesis of acetaminophen hepatotoxicity. Finally, although these animal observations cannot be transmitted to human experience, the possibility that inhibition of cytochrome P 45 by cimetidine treatment might be clinically useful in preventing hepatic damage due to acetaminophen and other compounds metabolized to hepatotoxic intermediates in humans is intriguing and warrants consideration. References 1. Mitchell JR, Jollow OJ. Metabolic activation of drugs to toxic substances. Gastroenterology 1975;68:392-41. 2. Zimmerman HJ. In: Hepatotoxicity. New York:Appleton-Century-Crofts, 1978:13-4. 3. Mitchell JR, Jollow OJ, Potter WZ, et al. Acetaminopheninduced hepatic necrosis. Role of drug metabolism. J Pharmacol Exp Ther 1973;187:185-94. 4. Corcoran GB, Mitchell JR, Vaishnav YN, et al. Evidence that acetaminophen and N-hydroxyacetaminophen form a common arylating intermediate N-acetyl-p-benzoquinoneimine. Mol PharmacoI198;17:536-42. 5. Shuster L, Quimby F, Bates A, et al. Liver damage from cocaine in mice. Life Sci 1977;2:135-42. 6. Evans MA, Johnson ME. The role of a reactive nitroxide radical in cocaine-induced hepatic necrosis. Fed Proc 1981; 4:638. 7. Pelkonen, Puurunen J. The effect of cimetidine on in vitro and in vivo microsomal drug metabolism in the rat. Biochem Pharmacol 198;29:375-8. 8. Serlin NJ, Mossman S, Sib eon RG, et al. Cimetidine: interaction with oral anticoagulants in man. Lancet 1979;ii:317-9. 9. Klotz U, Reimann 1. Delayed clearance of diazepam due to cimetidine. N Engl J Med 198;112-4. 1. Knodell RG, Holtzman JL, Cranks haw DL, et al. Drug metabolism by rat arid human hepatic microsomes in response to interaction with H2-receptor antagonists. Gastroenterology 1982 ;82 :84-8. 11. Ellman GL. Tissue sulfhydryl groups. Arch Biochem Biophys 1959;82:7-7. 12. Brodie BB, Axelrod J. The estimation of acetanilide and its metabolic products, aniline, N-acetyl-p-aminophenol and p aminophenol in biological fluids and tissues. J Pharmacol Exp Ther 1948;94:22-8. 13. Greenblatt OJ, Koch-Weser J. Clinical pharmacokinetics. N Engl J Med 1975;293:72-5. 14. Peterson FJ, Holloway DE, Erickson RR, et al. Ethanol induction of acetaminophen toxicity and metabolism. Life Sci 198;27:175-11. 15. Peterson FJ, Mason RP, Hovsepian J, et al. Oxygen-sensitive and insensitive nitroreduction by Escherichia coli and rat hepatic microsomes. J Bioi Chern 1979;254:49-14. 16. Nash T. The colorimetric estimation of formaldehyde by means of the Hautzsch reaction. Biochem J 1953;55:416-21. 17. Snedecor GW, Cochran WG. Statistical methods. Ames, Iowa: Iowa State University Press, 1967. 18. Speeg KV, Patwardhan RV, Avant GR, et al. Inhibition of

July 1983 CIMETIDINE BLOCKS EFFECTS OF HEPATOTOXINS 129 microsomal drug metabolism by histamine Hz-receptor antagonists studied in vivo and in vitro in rodents. Gastroenterology 1982;82:89-96. 19. Lin ]H, Levy G. Sulfate depletion after acetaminophen administration and replenishment by infusion of sodium sulfate or N-acetylcysteine in rats. Biochem Pharmacol 1981;3:2723-5. 2. Yamada T, Ludwig S, Kuhlenkamp }, et al. Direct protection against acetaminophen hepatotoxicity by propylthiouracil. J Clin Invest 1981;67:687-95. 21. Raheja KL, Linscheer WG, Cho C, et al. Protective effect of propylthiouracil independent of its hypothyroid effect on acetaminophen toxicity in the rat. J Pharmacol Exp Ther 1982;22:427-32. 22. Labadarios D, David M, Portman B. et al. Paracetamol-induced hepatic necrosis in the mouse-relationship between covalent binding, hepatic glutathione depletion and the protective effect of a-mercaptopropionylglycine. Biochem Pharmacol 1977;26:31-5. 23. Decker K, Keppler D. Galactosamine induced liver injury. In: Popper H, Schaffner F, eds. Progress in liver disease. Vol. IV. New York: Grune and Stratton, 1972:183-99. 24. Wendel A, Feuerstein S, Knoz K-H. Acute paracetamol intoxication of starved mice leads to lipid peroxidation in vivo. Biochem Pharmacol 1979;28:251-5. 25. Mitchell MC, Schenker S. Avant GR, et al. Cimetidine protects against acetaminophen hepatotoxicity in rats. Gastroenterology 1981;81:152-6.