Species and strain differences in xenobiotic-mediated induction of hepatic drug metabolizing enzymes

The nature of extent of drug metabolizing enzyme induction by exogenous compounds is often dependent on the animal being investigated, the chemical nature of the xenobiotic and the specific dosing protocol. Herein, the effects of both traditional and nontraditional inducing agents on the biotransformation of isozyme-selective substrates by hepatic drug metabolizing enzymes were investigated in the rat, hamster and three mouse strains. The N-substituted imidazoles clotrimazole, N-benzylimidazole and nafimidone were more potent in inducing Phase I monooxygenase (cytochrome P450) and Phase II conjugative (microsomal UDP-glucuronosyltransferase and cytosolic glutathione S-transferase) activities in the rat than in the hamster and mouse. These species differences were especially noteworthy for the O-deethylation of ethoxyresorufin, a monooxygenase activity purported to be induced by xenobiotic binding of a cytosolic receptor. In mice or hamsters, N-benzylimidazole and nafimidone caused only small increases in ethoxyresorufin O-deethylase activity, compared to the major inductions seen in rats. Marked species and strain variation was also observed in the induction caused by traditional inducing agents, including dexamethasone. In one strain of mouse, dexamethasone displayed a Phase I 'phenobarbital-like' inductive property and, in another strain, caused a Phase II '$/beta$-naphthoflavone-like' induction. Dexamethasone also induced hamster monooxygenase activity in a manner similar to that reported for ethanol-mediated induction in rats. Prolonged enteral exposure to the glutathione-depleting agent buthionine sulfoximine selectively induced Phase II UDP-glucuronosyltransferase and glutathione S-transferase activities in rats. The Phase II selective effect caused by buthionine sulfoximine was not evident in mice. In buthionine sulfoximine-treated rats, the increase in in vitro UDP-glucuronosyltransferase activity was paralleled by enhanced partial clearance and urinary recovery of acetaminophen-glucuronide following intravenous acetaminophen administration. These findings suggest that the ostensible isozyme-specific xenobiotic substrate assays that have been used and characterized, historically, in rats, to model induction may be inappropriate in other species. Furthermore, the induction of Phase II enzymes by buthionine sulfoximine has not been previously reported and is a significant effect to be considered in studies using prolonged buthionine sulfoximine treatments. The possible relationship between glutathione availability and the induction of conjugative enzymes is discussed.

SPECIES AND STRAIN DIFFERENCES IN XENOBIOTIC-MEDIATED INDUCTION OF HEPATIC DRUG METABOLIZING ENZYMES by Bradford W. Manning A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Pharmacology Department of Pharmacology and Toxicology The University of Utah March 1991 Copyright © Bradford W. Manning 1991 All Rig hts Reserved 'II IF UNIVFRSITY OF UTAII CRADUATE SCIIOOL S lJ I' 14~ lZ V I S() l{ \' (:0 l\Il'vi 1'1" 1 'r:E AIJI)I{() V Al-l o[ a dissctlalioll SlIlllllitlcd by Bradford Wayne Manning This dissertatioll 11;ls 1)('('11 I(';,d by callI nl('lIlhcr or the following supcrYisory COIlllllillcc (lJld by 1Il;~jorit Y vole has heen fOll nd to be satisl'a<lory. Chair: Michael R. Franklin /~~'-''''' , . ~ / David E. Moody - 3 -(to H. Steve White /~ 1J . -.--d-~---~? ~~~--=---.c4!:::.-'-· -'-- l /' Garold S. Yost' THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the University of Utah: I have read the thesis of B r a d for d W. Ma n n in g in its final form and have found that (1) its format, citations and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables and charts are in place; and (3) the final manuscript is satisfactory to the supeIVisory committee and is ready for submission to The Graduate School. February 6, 1991 Date Michael R. Franklin Chair. Supervisory Comminee Approved for the Major Department Approved for the Graduate Council B. Gale Dick Dean of The Graduate School ABSTRACT The nature and extent of drug metabolizing enzyme induction by exogenous compounds is often dependent on the animal being investigated, the chemical nature of the xenobiotic and the specific dosing protocol. Herein, the effects of both traditional and nontraditional inducing agents on the biotransformation of isozyme-selective substrates by hepatic drug metabolizing enzymes were investigated in the rat, hamster and three mouse strains. The N-substituted imidazoles clotrimazole, N-benzylimidazole and nafimidone were more potent in inducing Phase I monooxygenase (cytochrome P450) and Phase II conjugative (microsomal UDP-glucuronosyltransferase and cytosolic glutathione S-transferase) activities in the rat than in the hamster and mouse. These species differences were especially noteworthy for the O-deethylation of ethoxyresorufin, a monooxygenase activity purported to be induced by xenobiotic binding of a cytosolic receptor. In mice or hamsters, N-benzylimidazole and nafimidone caused only small increases in ethoxyresorufin O-deethylase activity, compared to the major inductions seen in rats. Marked species and strain variation was also observed in the inductio n caused by traditio nal inducing age nts, includi ng dexamethasone. In one strain of mouse, dexamethasone displayed a Phase I "phenobarbital-like" inductive property and, in another strain, caused a Phase II "B-naphthoflavone-like" induction. Dexamethasone also induced hamster monooxygenase activity in a manner similar to that reported for ethanol-mediated induction in rats. Prolonged enteral exposure to the glutathione-depleting agent buthionine sulfoximine selectively induced Phase II UDP-glucuronosyltransferase and glutathione S-transferase activities in rats. The Phase II selective effect caused by buthionine sulfoximine was not evident in mice. In buthionine sulfoximine-treated rats, the increase in in vitro UDP-glucuronosyltransferase activity was paralleled by enhanced partial clearance and urinary recovery of acetaminophen-glucuronide following intravenous acetaminophen administration. These findings suggest that the ostensible isozyme-specific xenobiotic substrate assays that have been used and characterized, historically, in rats, to model induction may be inappropriate in other species. Furthermore, the induction of Phase II enzymes by buthionine sulfoximine has not been previously reported and is a significant effect to be considered in studies using prolonged buthionine sulfoximine treatments. The possible relationship between glutathione availability and the induction of conjugative enzymes is discussed. v TABLE OF CONTENTS ABSTRACT ......................................................................................................... iv LIST OF FIGURES ............................................................................................ viii LIST OF TABLES ............................................................................................... ix ABBREVIATIONS FOR HEPATIC DRUG METABOLISM PARAMETERS ........................................................... xi ACKNOWLEDGEMENTS ................................................................................ xii Chapter 1 INTRODUCTION ..................................................................................... 1 Biological Significance of Xenobiotic Metabolism ........................... 1 Nitrogen-substituted Imidazoles as Inducers of Drug Metabolizing Enzymes........... ...... ... ... ...... .... ...... ..................... 6 Species and Strain Variation in Drug Metabolizing Enzyme Induction by Traditional Inducers ... .................. ... ............ 11 Toxicologic and Pharmacologic Importance of Drug Metabolizing Enzyme Induction: Species and Strain Differences ...................................................... 1 6 2 METHODS AND MATERIALS ............................................................. 21 Animals .................................................................................................... 21 Chemicals ................. ............ ... ... ....................... ..................................... 21 Xenobiotic Treatments. ........ ......... ... ...... ........... ........ ........ ..... .... ........... 22 Hepatic Cell Fraction Preparation ...................................... ..... ........... 23 Glutathione Concentration ................................................................... 23 Microsomal and Cytosolic Parameter Analyses .............................. 24 Acetaminophen Disposition Studies .................................................. 25 Data Analysis...... ................................. .................. ................................. 25 3 SPECIES AND STRAIN COMPARISONS IN XENOSIOTIC-MEDIATED INDUCTION ............................................. 28 N-Substituted Imidazoles and B-Naphthoflavone ............................ 28 Phenobarbital and Dexamethasone .................................................. 41 4 EFFECT OF SSO TREATMENT ON XENOSIOTIC-MEDIATED INDUCTION IN THE SPRAGUE-DAWLEY RAT ......... 44 N-8enzylimidazole ................................................................................ 44 B-Naphthoflavone .................................................................................. 4 7 5 SPECIES- AND DOSE-DEPENDENCIES FOR SSO-MEDIATED INDUCTION ............................................................ 50 Sprague-Dawley Rat ............................................................................ 50 02 and S6 Strains of Mice .................................................................. 56 6 EFFECT OF SSO TREATMENT ON THE DISPOSITION OF ACETAMINOPHEN IN TWO STRAINS OF RAT ......................... 58 Sprague-Dawley .................................................................................... 58 Fischer 344 ............................................................................................. 60 7 DISCUSSION ......................................................................................... 69 Species and Strain Differences in N-Substituted Imidazole-Mediated Induction .......................................................... 69 Species and Strain Differences in Traditional Inducer-Mediated Induction .............................................................. 73 Species Differences in SSO-Mediated Induction ............................ 77 Summary ................................................................................................ 80 REFERENCES .................................................................................................. 82 vii LIST OF FIGURES Rgure 1 Chemical structures of compounds used as inducing agents ......... 3 2 Effect of two clotrimazole dosing regimens on hepatic drug metabolizing enzymes in the CF-1 mouse .......................................... 29 3 Effect of four N-benzylimidazole dosing regimens on hepatic drug metabolizing enzymes in the CF-1 mouse .................................. 31 4 The temporal-dependent effects of clotrimazole and N-benzyl­imidazole on CF-1 mouse hepatic cytochrome P450 concentration and p-nitroanisole demethylase activity ................................................ 34 5 The effect of 3-day exposure to various concentrations of buthionine sulfoximine in drinking water on Sprague-Dawley rat hepatic microsomal U DP-glucuronosyltransferase activities ..... 54 6 Plasma concentrations of acetaminophen (A), acetaminophen­glucuronide (8) and acetaminophen-sulfate (C) in acetaminophen­and buthionine sulfoximine+acetaminophen-treated Sprague-Dawley rats ................................. ,., ........................................... 61 7 Plasma concentrations of acetaminophen (A), acetaminophen­glucuronide (8) and acetaminophen-sulfate (C) in acetaminophen­and buthionine sulfoximine+acetaminophen-treated Fischer 344 rats ......................................................................................... 66 LIST OF TABLES Hepatic parameters associated with drug metabolism in untreated mouse, hamster and rat .................................................... 26 2 Clotrimazole-mediated induction of mouse and rat hepatic drug metabolizing enzyme activities ...................................................... 36 3 The induction of mouse, hamster and rat hepatic drug metabolizing enzymes by N-benzylimidazole and B-naphthoflavone .............................................................................. 38 4 Nafimidone-n1ediated induction of mouse and rat hepatic drug metabolizing enzyme activities ..................................................... 40 5 The induction of mouse, hamster and rat hepatic drug metabolizing enzymes by phenobarbital and dexamethasone ....... ......... .................... .................... ......... ....... ........ 42 6 The effect of buthionine sulfoximine and N-benzylimidazole, alone and in combination, on Sprague-Dawley rat hepatic drug metabolizing enzyme activities ..................................................... 45 7 The effect of buthionine sulfoximine and B-naphthoflavone, alone and in combination, on Sprague-Dawley rat hepatic drug metabolizing enzyme activities ..................................................... 48 8 Hepatic glutathione concentration and Phase I and Phase II parameters in Sprague-Dawley rats following buthionine sulfoximine exposure by various routes ........................... 51 9 The effect of 3-day exposure to various concentrations of buthionine sulfoximine in the drinking water on Sprague-Dawley rat hepatic parameters ............................................. 53 10 The effect of intragastric administration of buthionine sulfoximine on mouse hepatic glutathione concentration and Phase I and Phase II enzyme parameters ........................................... 57 11 Hepatic glutathione concentration and Phase I and Phase II parameters in Sprague-Dawley rats treated with acetaminophen alone and in combination with buthionine sulfoximine ............................................................................. 59 1 2 The effect of buthionine sulfoximine treatment on acetaminophen elimination kinetics and urinary excretion products in Sprague-Dawley rats .......................................................... 63 1 3 Hepatic glutathione concentration and Phase I and Phase II parameters in Fischer 344 rats treated with acetaminophen alone and in combination with buthionine sulfoximine ............................................................................ 64 1 4 The effect of buthionine sulfoximine treatment on acetaminophen elimination kinetics and urinary excretion products in Rscher 344 rats ................................................................... 68 x GSH P450 pNA deM EROD PROD EYdeM pNPH UDPGt-pNP -N -M -E -T GSHt PSt ABBREVIATIONS FOR HEPATIC DRUG METABOLISM PARAMETERS Glutathione concentration Cytochrome P450 p-Nitroanisole demethylase Ethoxyresorufi n O-deethylase Pentoxyresorufin O-dealkylase Erythromycin demethylase p-Nitrophenol hydroxylase UDP-glucuronosyltransferase activity toward p-nitrophenol UDP-glucuronosyltransferase activity toward 1-naphthol UDP-glucuronosyltransferase activity toward morphine UDP-glucuronosyltransferase activity toward estrone UDP-glucuronosyltransferase activity toward testosterone Glutathione S-transferase activity toward 1-chloro-2,4- dinitrobenzene Phenol sulfotransferase activity toward p-nitrophenol ACKNOWLEDGEMENTS I wish to express my profound gratitude to and respect for my mentor, Dr. Michael R. Franklin. The importance of his support and guidance during the more trying times of my training cannot be overestimated. Additionally, my thanks to the other members of my committee; Drs. David E. Moody, Jeanette C. Roberts, H. Steve White and Gary S. Yost. I am also grate'ful to Janice Gilson and Tracy Gibb, whose technical assistance was often tendered and always accepted. Most important is Janice Thornton-Manning, whose patience and understanding were critical for perservering through the bad times and who made the good times infinitely better. CHAPTER 1 INTRODUCTION Biological Sit) nificance of Xenobiotic Metabolism The biotransformation of drugs and environmental contaminants by drug metabolizing enzymes frequently influences their biological activities. Because the liver receives the blood that perfuses the splanchnic area, the enzyme systems that catalyze the metabolism of foreign compounds are primarily hepatic in location. The metabolism of exogenous chemicals is commonly categorized into Phase I and Phase II enzyme systems (Williams, 1959). The Phase I enzymes oxidize, hydrolyze or reduce the substrate, thereby introducing a polar functional group onto the xenobiotic. An important Phase ,. enzyme system is the family of cytochrome P450 (P450) isozymes. These hemoproteins are found within the endoplasmic reticular (microsomal) membrane and require molecular oxygen and an electron source (NADPH) to function as catalysts in the oxidation of substrates. Also required is the flavoprotein NADPH cytochrome P450 reductase that funnels electrons from NADPH to the cytochrome. Since P450-mediated oxidation results in the incorporation of one atom of molecular oxygen into the drug, these enzymes are often referred to as monooxygenases. The Phase II reactions involve conjugation of functional groups with an endogenous substance, e.g., glucuronic acid, glutathione, sulfate or various amino acids, by the action of corresponding transferase enzymes. The result of combined Phase I and Phase II xenobiotic metabolism is often the production of innocuous hydrophilic compounds 2 with enhanced potential for elimination. For some xenobiotics, however, P450-mediated oxidation has been implicated in the production of reactive electrophilic intermediates capable of interacting with cellular macromolecules (e.g., proteins, nucleic acids, lipids) and eliciting a diversity of pharmacodynamic and toxicologic responses. Such deleterious effects may be mitigated by Phase II enzyme-catalyzed conjugation of the reactive intermediate. However, Phase II reactions are not always detoxifying and conjugation can result in the bioactivation of a chemical to a toxic compound. Hence, the activities of hepatic Phase I and Phase II enzymes are frequently critical in the efficacy and/or toxicity of drugs and other xenobiotics. Increases in P450 monooxygenase activities are common following administration of a wide range of foreign chemicals, with the pattern and extent of induction dependent on the nature of the xenobiotic (Conney, 1967). Agents capable of increasing hepatic P450 concentration and activity were traditionally classified as phenobarbital- (Pb, see Figure 1 for chemical structures), polycyclic aromatic hydrocarbon- or mixed- (e.g., Aroclor 1254) type inducers (Snyder and Remmer, 1982). The P450llB1 and P450llB2 isozymes are preferentially induced by Pb, whereas polycyclic aromatic hydrocarbons are selective in their induction of the P450lA1 and P450lA2 forms. However, other compounds, e.g., isosafrole (Dickens et aI., 1978), clofibrate (Tamburini et aI., 1984), ethanol (Reinke and Moyer, 1985), isoniazid (Ryan et aI., 1985), and the steroids dexamethasone (DEX, Figure 1) and pregnenolone-16-alpha-carbonitrile (Wrighton et aI., 1985a), preferentially induce P450 isozymes different from those induced by Pb and/or polycyclic aromatic hydrocarbons. Clofibrate selectively induces P450lVA 1, while both ethanol and isoniazid induce P4501lE1. Isozymes of the P4501llA subfamily are induced by DEX and pregnenolone-16-alpha-carbonitrile. Hence, the traditional classification system, while useful, was incomplete. The induction phenomenon is most 3 Figure 1. Chemical structures of compounds used as inducing agents. N-Substituted Imidazoles yel < ) r < > N (It N Clotrimazole (Clotz) H-C-H I N (? N-benzylimidazole (NBI) Traditional Inducers C=O I H-C-H I N C'« Nafimidone ~-N a phthoflavone (BNF) o Phenobarbital Dexamethasone (Pb) (DEX) Buthionine Sulfoximine (BSO) 4 5 often a consequence of de novo protein synthesis via transcriptional activation of P450 genes rather than stabilization of constitutive enzymes or a decrease in the rate of protein degradation (Haugen et aI., 1976; Adesnik and Atchison, 1986). Increased levels of specific m RNAs are detectable soon after treatment with either Pb (Adesnik et aI., 1981; Omiecinski, 1986) or the polycyclic aromatic hydrocarbon, 3-methylcholanthrene (Tukey et aI., 1981). DEX has also been demonstrated to increase the steroid-inducible concentration of P4501liA (P450p) in rats by enhancing protein synthesis (Watkins et aI., 1986). The induction of P450 by traditional inducers is often accompanied by increases in certain Phase II activities, although the extent to which P450 activity is induced is frequently greater than that of Phase II enzymes (Okey, 1990). Microsomal UDP-glucuronosyltransferase activity towards planar monohydroxylated aglycones (e.g., p-nitrophenol, 1 ~naphthol, 4-methylumbelliferone) is preferentially enhanced after treatment of rats with the polycyclic aromatic hydrocarbon 3-methylcholanthrene, whereas the transferase that catalyzes the glucuronidation of bulkier molecules (morphine, monoterpenoid alcohols, 4-hydroxybiphenyl) is selectively induced by Pb (Watkins et aI., 1982; Bock et aI., 1983; Thomassin et aI., 1986). Both 3-methylcholanthrene and Pb also induce cytosolic glutathione S-transferase enzymes (Hales and Neims, 1977; Pickett et aI., 1982). Specific inducers of cytosolic sulfotransferases are not known. However, certain agents, e.g., the antioxidant butylated hydroxyanisole (Cha and Bueding, 1979; Thompson et aI., 1982; Watkins et aI., 1982), the dental analgesic and food-flavor additive eugenol (4-allyl-2-methoxyphenol) (Boutin et aI., 1983; Yokota et aI., 1988) and flavanone (Siess et aI., 1989) induce UDP-glucuronosyltransferase(s) and/or glutathione S-transferase activities, while having little or no effect on P450 concentration or monooxygenase activity. Nitrogen-substituted Imidazoles as Inducers of Drug Metabolizing Enzymes 6 Several N-substituted imidazoles are employed clinically in the treatment of mycotic infections (Feldman, 1986), as antithrombitic agents (Randall et aI., 1981; Fischer et aI., 1983), or have been proposed for use as anticonvulsants (Walker et aI., 1981). The antimycotic imidazoles, e.g., clotrimazole (Clotz, Figure 1), miconazole, ketoconazole, econazole and tioconazole, exert their fungistatic properties by inhibiting the P450-dependent 14-alpha-demethylation of lanosterol, thereby blocking synthesis of the essential fungal membrane component ergosterol (Borgers, 1980; Beggs et aI., 1981; Henry and Sisler, 1984; Vanden-Bossche et aI., 1984). However, the antifungal imidazoles are also potent inhibitors of mammalian P450-mediated steroidogenesis in a number of tissues (Loose et aI., 1983; Mason et aI., 1985; Sikka et aI., 1985), as well as inhibitors of the hepatic microsomal oxygenation of a variety of xenobiotics (Kahl et aI., 1980; Sheets and Mason, 1984; Meredith et aI., 1985; Rodrigues et aI., 1987). For example, Clotz, miconazole and ketoconazole inhibit the hepatic oxidative metabolism by two subfamilies of P450: Pb-inducible P450llB1 [P450b; 7-pentoxyresorufin O-dealkylase (PROD) activity] and 3-methylcholanthrene-inducible P450lA 1 [P450c; 7 -ethoxyresorufin O-deethylase (EROO) activity] in rat microsomes and elicit type II difference spectra, Le., peak at 430 nm and trough at 393 nm, with both isozymes (Rodrigues et aI., 1987). Type II spectral characteristics result from the binding of the unhindered nitrogen of the imidazole ring (N-3) to the heme ferric iron of P450 (Imai and Sato, 1967), thereby suppressing the catalytic activity of the enzyme. Of these three N-substituted imidazoles, ketoconazole was the weakest inhibitor of rat monooxygenase activity, but all were better inhibitors of the PROD activity of purified P450llB1 than the EROO activity of purified P450lA1 (Rodrigues et aI., 1987). In contrast, in 7 vivo studies in rats have reported that ketoconazole, after a single dose, was a more selective inhibitor of P450lA 1-catalyzed caffeine N-demethylation than the P4501lB 1-catalyzed N-demethylation of aminopyrine (Meredith et aI., 1985). N-substituted imidazoles, as a class, are among the most potent P450 inhibitors known (Leibman and Ortiz, 1973; Wilkinson et aI., 1974). However, not all N-substituted imidazoles are good inhibitors of P450 activity. In vitro studies comparing the inhibitory effects of two antithrombitic (dazmegrel and dazoxiben) and four antimycotic (ketoconazole, econazole, miconazole and Clotz) N-substituted imidazoles on rat hepatic microsomal P450 monooxygenases have shown the antithrombitic agents to be ineffective as inhibitors of both steroid and xenobiotic oxidations, despite their ability to suppress platelet P450-thromboxane synthetase (Murray and Zaluzny, 1988). As expected, the antifungal N-substituted imdazoles were all potent inhibitors of monooxygenase activity. Binding studies revealed that all six N-substituted imidazoles produced a type II difference spectrum and exhibited a high binding affinity for ferric P450. Structurally, both dazmegrel and dazoxiben, but not the antifungal N-substituted imidazoles, possess a carboxylate group on their N-arylalkyl substituents, a feature that causes the antithrombitic compounds to be predominantly ionized at physiological pH. The investigators suggested that, while the N-3 atom of dazmegrel and dazoxiben may interact with the P450 heme iron to cause the observed type " binding spectra, their N-substituents lack sufficient hydrophobic character to effectively interact with substrate binding sites adjacent to the heme moiety. Conversely, the ionized carboxylate group on dazmegrel and dazoxiben may be important for coordination of the two drugs with the active site on P450-thromboxane synthetase. In addition to their inhibitory effects, many N-substituted imidazoles can, depending on the treatment regimen, also induce drug metabolizing 8 enzymes. Neither the 4,5-disubstituted imidazole cimetidine nor imidazole itself are especially effective in inhibiting or inducing drug metabolizing enzymes (Heusner and Franklin, 1985; Ritter and Franklin, 1987a). Furthermore, generally only those imidazoles with N-substituents possess both potent Phase I inhibitory and inductive properties. This has lead to the hypothesis that sustained inhibition of P450 activity can trigger compensatory P450 induction (Ritter and Franklin, 1987b). A corollary is found in the suicide substrate allylisopropylacetamide, which inhibits P450 by causing a mechanism-based heme destruction, but which can also act as an inducing agent (Deloria et aI., 1980). However, the inactivation of P450 by another suicide substrate, 1-aminobenzotriazole, does not cause subsequent induction (Ortiz de Montellano and Costa, 1986). Hence, the relationship between the incapacitation of P450 activity and its induction is not clear. The involvement of both temporal- and dose-dependencies in the inhibition and induction of hepatic P450 activity caused by N-substituted imidazoles are documented. The anticonvulsant N-substituted imidazole nafimidone is a potent inhibitor of hepatic Phase I drug metabolism in rats and mice (Rush et aI., 1987). Inhibition in rats was detected in vivo following acute administration as increased hexobarbital sleep time and, in YitrQ, as decreased ethylmorphine N-demethylase and aniline hydroxylase activities. Administration of single daily doses of nafimidone or its aliphatic alcohol metabolite for 4 days caused induction of murine hepatic drug metabolism, reflected in a shortening of pentobarbital .. induced sleep time. In rats, daily administration of nafimidone alcohol for 4 days increased P450 concentration 2-fold. Daily nafimidone treatment also produced a 30-fold induction of EROD activity but did not affect aniline hydroxylation or Phase II microsomal UDP-glucuronosyltransferase activity towards 4-hydroxybiphenyl or 4-methylumbelliferone. Treatment of mice with miconazole also caused a biphasic effect on barbiturate-induced sleep 9 time (Andre et aI., 1984). The pentobarbital-induced sleep time was prolonged after an acute dose, implying an inhibitory effect on P450 activity, while an inductive effect (decreased sleep time) was observed after 3 or 5 days of daily miconazole treatment. Lavrijsen et al. (1986) reported that rats displayed both dose~ and time-dependent inhibition and induction patterns in response to treatment with miconazole or ketoconazole. Acute treatment (160 mg/kg 1 hr before sacrifice) with either compound inhibited microsomal p-nitroanisole demethylase activity. At 160 mg/kg/day, both drugs increased P450 concentration and were effective inducers of Phase I p-nitroanisole demethylation and Phase " p-nitrophenol glucuronosyltransferase activity when determined 23 hr after dosing for 7 days. Miconazo!e, but not ketoconazole, also induced aminopyrine and N,N-dimethylaniline N-demethylase activities. Neither drug affected the rate of aniline hydroxylation. An intermediate dose (40 mg/kg/day for 7 days) of miconazole, but not ketoconazole, increased P450 concentration but to a lesser extent than observed at the highest dose. No effect on drug metabolizing enzymes was detected for either antimycotic at a dose of 10 mg/kg/day, independent of the duration of exposure. Based on a comparison of induced enzyme activities and sodium dodecyl sulfate-polyacrylamide gel electrophoretograms with Pb­and 3-methylcholanthrene-treated animals, the investigators concluded that miconazole behaved as a Pb-type inducer, whereas ketoconazole was unlike either Pb or 3-methylcholanthrene in its inductive properties. The inhibitory and Pb-like inductive properties of miconazole have also been reported by Ritter and Franklin (1987a). However, Clotz, at 75 mg/kg/day for 3 days followed by a 2-day wait prior to sacrifice, induced microsomal Phase I and Phase II drug metabolizing enzyme activities in rats in a manner similar to that elicited by both Pb (e.g., induced PROD activity and enhanced morphine glucuronidation) and DEX [e.g., induced erythromycin demethylase (EY deM) activity] (Ritter and Franklin 1987b; 10 1987c). Glotz also exhibited dose-differentiated inductive properties. At doses of less than 25 mg/kg/day for 3 days, the parameters indicative of Pb-like induction were increased, while at 75 mg/kg/day for 3 days, the drug also enhanced monooxygenase activities similar to those induced by DEX. The selective changes in Phase I enzyme activities in rats caused by high doses of Glotz have been confirmed by increases in hybridizable P4501liA (P450p; DEX-inducible) and P450llB1111B2 (P450b/e; Pb-inducible) mRNAs (Hostetler et aI., 1989) and by immunoquantification with isozyme-specific monoclonal antibodies (Khan et aI., 1989). Glotz and miconazole also increased cytosolic glutathione S-transferase activity toward 1-chloro-2,4-dinitrobenzene, but decreased cytosolic p-nitrophenol sulfotransferase activity. In contrast to the Pb- and OEX-like induction profiles mediated by Glotz, the N-substituted imidazoles tioconazole (150 mg/kg/day) and N-benzylimidazole [(NBI, Figure 1); 75 mg/kg/day], both administered as single daily doses for 3 days followed by 2 days without treatment before sacrifice, showed microsomal Phase II UOP-glucuronosyltransferase induction patterns in -the rat that paralleled those caused by both Pb (increased morphine glucuronidation) and B-naphthoflavone [(BNF, Figure 1); preferential increases in p-nitrophenol and 1-naphthol glucuronidations)] (Ritter and Franklin 1987a; Papac and Franklin 1988). NBI also increased P450 concentration and PROD and EROO activities by 3- (from 0.88 to 2.82 nmoles/mg), 16- (from 0.01 to 0.16 nmoles/min/mg protein) and 56- (from 0.04 to 2.24 nmoles/min/mg protein) fold, respectively, whereas tioconazole was only marginally effective as a monooxygenase inducer. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed similar patterns for NBI- and Aroclor 1254-induced microsomes, Aroclor 1254 being a documented mixed Pb-/poJycycJic aromatic hydrocarbon-like inducer. Thus, the NBI-mediated induction of rat Phase I hepatic drug metabOlizing enzymes exhibits Pb- and polycyclic 1 1 aromatic hydrocarbon-like characteristics. As with Clotz and miconazole, both tioconazole and NBl increased glutathione S-transferase activity and decreased p-nitrophenol sulfotransferase activity. Species and Strain Variation in Drug Metabolizing Enzyme Induction by Traditional Inducers A number of endogenous and exogenous factors may influence the manner in which drug metabolizing enzymes respond to exogenous chemical exposure, including the genetic composition and nutritional status of the animal, as well as the route, site, duration and frequency of xenobiotic treatment. A well-documented example is the importance of genetic variation in drug metabolizing enzyme induction by BNF and nonhalogenated polycyclic aromatic hydrocarbons, e.g., 3-methylcholanthrene, benzo[a]pyrene and benz[a]anthracene, between C57BU6 (86) and DBAl2 (D2) mice. The 86 mice display major induction of hepatic P4S0lA 1 (P1450; equivalent to rat P4S0c) and P4S0lA2 (P3450; equivalent to rat P450d) and are termed aromatic hydrocarbon-(Ah) responsive. Collectively, induced P4S0lA1 and P4S0lA2 have been referred to in past reports (see below) as P448 because of the hypsochromic shift of the ferrous P4S0-carbon monoxide complex to an absorbance maximum of 448 nm. The D2 mice are classified as Ah-nonresponsive since 3-methylcholanthrene and other nonhalogenated polycyclic aromatic hydrocarbons and BNF do not induce either P4S0lA 1 or P450lA2 in this strain of mouse (Poland and Knutson, 1982; Eisen et aI., 1983). The environmental pollutant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induces P4S0lA 1 and P4S0lA2 in both mouse strains, but the EDSO is 20 nmoles/kg for D2 mice compared to 1 nmole/kg in 86 mice (Poland et aI., 1974), suggesting that Ah-nonresponsive mice harbor the structural and regulatory genes necessary for induction. The relative xenobiotic inductive potencies have, in part, led to the conclusion that B6 12 mice possess high-affinity Ah receptors, while 02 mice have low amounts of the receptor and/or binding sites with low affinity (Eisen et aL, 1983). These strain differences are linked to the Ah locus, which encodes for a cytosolic Ah receptor. The Ah receptor has been detected in a variety of mammalian cells, ranging from rodents to humans (Okey et aL, 1983; Gasiewicz and Rucci, 1984; Okey et aL, 1984; Denison et aI., 1985; Denison and Wilkinson, 1985), and in chick embryos (Denison et aL, 1985). All rat strains (Nebert et aL, 1982) and golden Syrian hamsters (Blaich et aI., 1988; Lubet et aI., 1990) are Ah-responsive, although the inducibility of P448 in hamsters is less than in rats and Ah-responsive mice. There is also evidence for heritable variation in P448 inducibility in humans (Kellermann et aL, 1973; Atlas et aI., 1976; Thorgeirsson and Nebert, 1977), as well as for the inductive effects on P448 caused by various environmental contaminants; e.g., tobacco smoke (Welch et aL, 1968; Kapitulnik et aL, 1976) and charcoal-cooked foods (Canney et aL, 1976; Pantuck et aL, 1976). Early studies on the substrate specificity of P448 utilized "aryl hydrocarbon (benzo[a]pyrene) hydroxylase" (AHH) activity to monitor induction. While this assay is still commonly employed, there is also evidence supportive of EROD activity as reflecting only P448-nlediated metabolism in Ah-responsive mice, hamsters and rats (Burke et aL, 1977 I 1985; Kelley et aL, 1990; Lubet et aL, 1990). Extensive studies using purified P450 proteins from rat livers have shown that only P450lA 1 and P450lA2 can catalyze the O-deethylation of ethoxyresorufin (Guengerich et aI., 1982a; Astrom and DePierre, 1985). Similar observations have been made with the orthologous rabbit isozymes, forms 6 (P450 IA 1) and 4 (P450IA2) (Phillipson et aI., 1984). Additionally, treatment of golden Syrian hamsters with polycyclic aromatic hydrocarbon-type inducers increases hepatic EROD, but not AHH, activity (Chiang and Steggles, 1983; Smith et aL, 1986). EROD activity is also associated with the murine 13 Ah locus, as illustrated in studies with Ah-responsive and Ah-nonresponsive mice (Lum et aI., 1986). Enzymes other than P450 monooxygenases that are induced by Ah receptor agonists and that appear to be associated with the dominant somatic allele (Ahb) encoding for the Ah receptor include microsomal UOP-glucuronosyltransferase and two cytosolic enzymes; NAO(P)H :menadione oxidoreductase and ornithine decarboxylase (Nebert et aI., 1981). Some investigators have also reported cytosolic glutathione S-transferase activity to be induced in Ah-responsive rodents by 3-methylcholanthrene (Hales and Neims, 1977; Pickett et aI., 1982), benz[a]anthracene (Sherratt et aI., 1990) and TCOO (Ryan et aI., 1976), whereas other studies have concluded that the transferase is not significantly affected by 3-methylchotanthrene (Mukhtar and Bresnick, 1976a), TCDD (Kouri et aI., 1978) or BNF (Mukhtar and Bresnick, 1976b). In genetic backcross experiments with progeny from Ah-responsive B6 and Ah-nonresponsive 02 mice, Felton et al. (1980) claimed that the increase in glutathione S-transferase activity towards various substrates, including 1-chloro-2,4-dinitrobenzene, following 3-methylcholanthrene treatment did not correlate with the Ah locus. However, the 5f-flanking region of the rat liver gene encoding for the Va subunit of the dimeric glutathione S-transferase proteins has been reported to contain a polycyclic aromatic hydrocarbon regu latory sequence for inducible expression by 3-methylcholanthrene (Telakowski-Hopkins et aI., 1988). The mechanism by which the Ah receptor mediates expression of the polycyclic aromatic hydrocarbon-inducible genes is not fully understood, but appears to be similar to that of glucocorticoid-inducible systems (Okey et aI., 1979; Eisen et aI., 1983; Poellinger et aI., 1983; Nebert et aI., 1984; Gustafsson et aI., 1987). After initial binding to a cytosolic form of the Ah receptor, the ligand-receptor complex undergoes a temperature-dependent translocation to the nucleus. Interaction of the 14 inducer-receptor complex with an enhancer region on DNA upstream from the promoter activates gene transcription (Nebert and Gonzalez, 1987). The P448 cytochromes can also be induced by polychlorinated biphenyls lacking ortho substituents (Luster et aL, 1982; Parkinson et aI., 1983; Safe et aI., 1985), ellipticines (Cresteil et aI., 1982; Phillipson et aI., 1982), aminoazobenzenes (Degawa et aI., 1985;1986) and polychlorinated dibenzofurans (Bandiera et aI., 1984). Although Poellinger et al. (1983) claimed that Sprague-Dawley rat and B6 mouse livers contain an identical specific cytosolic binding site for TCDD, it has been reported that the Ah receptors of rats and B6 mice are similar but not identical, since differences were observed in molecular size, subunit dissociation properties and ligand binding preferences (Denison et aL, 1986; Piskorska-Pliszcynska and Safe, 1988; Raha et aI., 1990). Nevertheless, Fernandez et aL (1988) demonstrated that the binding characteristics of isolated Sprague-Dawley rat and B6 mouse Ah receptors were similar when assessed by competition between TCDD, 2,3,7,8-tetrachlordibenzofuran, 3-methylcholanthrene, benzo[a]pyrene, BNF and ellipticines with radiolabeled 3-methylcholanthrene or TCDD. A good correlation between the competitive potency of the compounds and their ability to induce AHH activity was also shown. Additionally, for both Sprague-Dawley rats and B6 mice, treatment of the Ah receptor with thiol-modifying reagents dramatically decreases receptor-ligand binding affinity (Denison et aI., 1987; Kester and Gasiewicz, 1987), suggesting that the cytosolic Ah receptor, like the cytosolic glucocorticoid receptor, contains sulfhydryl group(s) essential for maintaining its agonist binding integrity. The cytosolic Ah and glucocorticoid receptors share many physiochemical properties (Gustafsson et aI., 1987). They are structurally and mechanistically similar and may belong to the same family of regulatory proteins. In cultured fetal rat hepatocytes, the synthetic glucocorticOid DEX potentiates the induction of P450lA 1 by benz[a]- 15 anthracene (Mathis et aI., 1986, 1989; Sherratt et aI., 1990). DEX alone does not induce P450lA 1 in this system and the cytosolic Ah receptor has no detectable affinity for DEX (Eisen et aI., 1983). The synergistic interaction between the two xenobiotics appears to involve binding of DEX to the glucocorticoid receptor, followed by interaction of the DEX-glucocorticoid receptor complex with specific regions of the P450lA 1 gene and enhanced transcription. Indeed, glucocorticoid regulatory element(s) have been shown to be located within intron 1 of the P450lA 1 gene (Hines et aI., 1988; Mathis et aI., 1989). Sherratt et al. (1989) have also reported enhancement of 3-methylcholanthrene-induced hepatic P450lA 1 concentration and EROD activity by the simultaneous treatment of rats with DEX. The potentiation of 3-methylcholanthrene-induced P450lA 1 activity was tissue specific since DEX did not augnlent the 3-methylcholanthrene induction of EROD activity in lung or kidney. The synthetic steroid did not enhance the liver concentrations of 3-methylcholanthrene-induced P450lA2 or Pb-induced P450llB1 proteins. However, DEX has recently been shown to increase P450lA2 mRNA in rat hepatocyte cultures treated with 3-methylcholanthrene while having little effect on P450lA2 gene transcription (Silver et aI., 1990). DEX also exhibits a synergistic effect on the benz[a]anthracene-mediated induction of glutathione S-transferase activity toward 1-chloro-2,4-dinitrobenzene in cultured fetal rat hepatocytes (Sherratt et aI., 1990). For other traditional inducers, there is no convincing evidence for specific receptor-based mechanisms of induction and the physiological parameters governing enzyme induction are unknown. However, a characteristic of Pb and Pb-like inducers, including other barbiturates, phenytoin and dichlorodiphenyltrichloroethane, commonly known as DDT (Conney, 1967), and polychlorinated biphenyls with ortho chlorines (Safe et aI., 1985), is that they have relatively low potency. For example, TCDD is an effective inducer of P448 in Ah-responsive animals when admini- 16 stered in the nmole/kg dose range. In contrast, the effective inducing dose for Pb and related inducers is in the J.1mole/kg range. Hence, if a specific Pb receptor exists, it would be expected to have a comparatively low affinity for the inducer. Such a low affinity protein would be difficult to detect with conventional receptor assay procedures. Poland et al. (1980) synthesized 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP), a Pb-like inducer with a much greater inductive potency than Pb in mice, in an attempt to identify a Pb receptor. However, TCPOBOP did not behave as a Pb-like inducer in rats and there was no indication that TCPOBOP interacted with a specific receptor in mice (Poland et aI., 1981). Other mechanisms that have been proposed include the suggestions that Pb and related compounds act, via their physical properties, on membranes, or that the P450 proteins themselves act as "receptor" sites (Poland et aI., 1980). According to the latter hypothesis, inducers of P450 are poor substrates for P450, thereby uncoupling the monooxygenation reaction and leading to an accumulation of singlet oxygen, which represents the true inducer (Paine et aI., 1978). To date there is no compelling evidence for either of these proposed alternate models. Toxicologic and Pharmacologic Importance of Drug Metabolizing Enzyme Induction: Species and Strajn Differences The biological potency and/or efficacy of drugs or other xenobiotics may be intluenced by the existing state of drug metabolizing enzyme activities. Furthermore, the activities of drug metabolizing enzymes can be altered by interdependent factors such as the nature of the xenobiotic exposure and the animal being exposed. Formation of the carcinogenic metabolite of the Ah receptor agonist benzo[a]pyrene, for example, is favored when the P450lA 1 to total P450 ratio is high. The bioactivation of benzo[a]pyrene has been associated with the Ah locus as evidenced by 17 the greater incidence of benzo[a]pyrene-initiated subcutaneous fibrosarcomas in the Ah-responsive 86 mouse stra.in compared to Ah-nonresponsive 02 mice (Kouri and Nebert, 1977). The differences in EO 50 for P448 induction by TCOO in Ah-responsive and Ah-nonresponsive mice also correlates with differences in the pleiotropic responses symptomatic of TCOO exposure, e.g., chloracne, thymic atrophy, immunodeficiency (Eisen et aI., 1983). The tumorigenesis caused by 2-acetylaminofluorene is dependent on P450-mediated N-hydroxylation to the proximate carcinogenic species (Thorgeirsson et aI., 1973). The 3-methylchlolanthrene-inducible N-hydroxylase activity is associated with the Ahb allele in mice (Felton et aI., 1976). Chen et al. (1983) screened several therapeutic agents for differences in toxicity between 86 and 02 mice pretreated with 8NF. The 86 strain was markedly more resistant to the lethal effects of the vasodilator isoxsuprine and the anthelmintic niridazole. Neither drug was able to displace TCOO from the 86 mouse Ah receptor, implying that a direct drug-Ah receptor interaction was not involved in the reported differences but, rather, that a strain difference in drug metabolizing enzyme activity subsequent to 8NF treatment was critical in the observed toxicities. A strong correlation has also been reported between both acetaminophen- and naphthalene-mediated lenticular opacification and the Ah locus (Shichi et aI., 1978, 1980; Shichi and Nebert, 1985). 80th Ahb homozygous and Ahb heterozygous mice, but not mice homozygous for the Ah-nonresponsive allele, when pretreated with 3-methylcholanthrene or 8NF, developed cataracts following administration of acetaminophen or naphthalene. The over-the-counter analgesic/antipyretic drug acetaminophen is a useful compound for the study of Phase I and Phase II drug metabolism. At therapeutic doses, and in the absence of perturbations in enzyme activities or cofactor availability, both sulfate and glucuronide esterification of the phenolic moiety of acetaminophen yields inactive and highly excretable 18 metabolites. At higher doses, glutathione conjugates of P450-oxidized acetaminophen are found. Following the depletion of hepatic glutathione that occurs with excessive doses, however, acetaminophen can produce liver necrosis in both humans and laboratory animals (Hinson, 1980). The present consensus is that acetaminophen is bioactivated through a P450-catalyzed oxidation to the toxic, electrophilic intermediate N-acetyl-p-benzoquinoneimine (Dahlin et aI., 1984; Potter and Hinson, 1987). However, the hepatotoxicity of acetaminophen varies considerably among species (Davies et aI., 1974; Siegers et aI., 1978). Hamsters and mice are more sensitive to the hepatotoxic properties of acetaminophen than rats, rabbits and guinea pigs. In studies comparing acetaminophen metabolism between these species, Gregus et al. (1988) showed that the variation in species sensitivity to acetaminophen-induced liver necrosis is a consequence of differences in the balance of toxication and detoxication metabolic pathways. The effect of traditional inducing agents on acetaminophen bioactivation also varies between species. Treatment of mice and rats with Pb or 3-methylcholanthrene enhances the toxicity of acetaminophen (Jollow et aI., 1973; Mitchell et aI., 1973; Gregus et aI., 1990). In hamsters, 3-methylcholanthrene also increases the bioactivation of acetaminophen, but Pb has little or no effect (Jollow et al. 1982; Lupo et aI., 1987; Madhu et aI., 1989). Ethanol, isoniazid, butylated hydroxyanisole and pregnenolone-16alpha-carbonitrile are also ineffective in enhancing acetaminophen bioactivation in hamsters (Madhu et al., 1989). Conversely, treatment of rats with pregnenolone-16alpha-carbonitrile increases formation of the acetaminophen-glutathione metabolite, indicative of acetaminophen bioactivation (Gregus et aI., 1990). Thus, there are significant differences between species in the inducibility of P450 isozyme(s) involved in the oxidative activation of acetaminophen. Because of its ability to inhibit gamma-glutamylcysteine synthetase 19 and, hence, glutathione synthesis (Griffith and Meister, 1979; Griffith, 1982), without apparently affecting drug metabolizing enzyme activities, buthionine sulfoximine (BSO, Figure 1) has become the agent of choice for investigating the role of glutathione in xenobiotic biotransformation and toxicity. The use of diethylmaleate to deplete glutathione has the disadvantages of its inhibitory and enhancing effects on monooxygenase activities (Anders, 1978), its stin1ulation of microsomal heme oxygenase and aminolevulinate synthetase (Maines, 1982) and its inhibitory effect on protein synthesis (Costa and Murphy, 1986). BSO is most commonly used on an acute basis in studies examining the metabolism and disposition of xenobiotics. Drew and Miners (1984) showed that the amount of acetaminophen excreted unchanged in the urine, as well as that excreted as the glucuronide and sulfate conjugates, were not altered in mice that had received intraperitoneal (ip) doses of BSO 45 min before and 6 hr after acetaminophen administration. Not unexpectedly, the acetaminophen recovered as metabolites derived from glutathione conjugation was decreased. Acute intragastric (ig) administration of BSO also had no effect on hexobarbital sleep time. Following a single ip BSO dose that acutely depleted liver glutathione concentration, the investigators reported little or no change in hepatic glutathione S-transferase, p-nitrophenol sulfotransferase and p-nitrophenol UDP-g lucuronosyltransferase activities, and P450 concentration and activity for up to 24 hr after treatment. White et al. (1984) have also demonstrated BSO to have no effect on murine glutathione S-transferase activities, several monooxygenase activities and NADPH-P450 reductase when assayed 2 hr after a single ip dose. Sun et a.1. (1985) reported there to be no difference in murine P450 concentration between control mice and those maintained on drinking water supplemented with BSO at a concentration of 30mM. The effect of such long-term BSO exposure on monooxygenase and conjugative enzyme activities was not explored. 20 An appreciation of the potential biological significance of changes in xenobiotic metabolizing enzyme activities caused by many of the compounds to which humans are intentionally or unintentially exposed is primarily due to experimental work with whole animals or mammalian cell cultures. However, the effect a given chemical has on drug metabolism may vary between the species and strains commonly employed in the laboratory. This inconsistency of response to exogenous chemicals complicates the processes of screening and developing therapuetic agents and the assessment of potential health risk posed by environmental contaminants. While the disparity between species and strains in their response to xenobiotics is generally recognized, it is often ignored. The present studies demonstrate the need for caution in extrapolating data derived from a single species to possible effects in humans. CHAPTER 2 METHODS AND MATERIALS Animals All animals employed in the current research were male. Mice (20-30 g), hamsters (100-140 g) and Sprague-Dawley rats (150-370 g) were purchased 'from Simonsen Laboratories, Gilroy, CA. Fischer 344 rats (200-300 g) were obtained 'from colonies maintained for the National Institute on Aging (Indianapolis, IN). All animals were housed under a 12-hr light-dark cycle and allowed free access to food and water. Temperature and humidity remained constant and all animals were placed on sawdust bedding. Rats and hamsters were maintained in cages housing no more than 4 animals per cage, except for rats treated with acetaminophen that were housed in individual metabolism cages. Mice were housed at 10 or fewer animals per cage. For all studies, xenobiotic-treated animals were housed separate from the untreated controls. Chemicals Clotz, DEX, BNF, BSO, acetaminophen, 1-naphthol, estrone and testosterone were purchased from Sigma Chemical Co. (St. Louis, MO). Morphine sulfate was obtained from Merck Chemical Division (Rahway, NJ). NBI was purchased from Aldrich Chemical Co. (Milwaukee, WI) and nafimidone was a gift from Syntex Laboratories (Palo Alta, CA). Pb was obtained from Gane's Chemical Works (New York, NY). Authentic 22 standards of the sulfate and glucuronide conjugates of acetaminophen were obtained from McNeil Consumer Products Co. (Fort Washington, PAl. Ethoxyresorufin and pentoxyresorufin were purchased from Pierce Chemical Co. (Rockford, IL) and Molecular Probes (Eugene, OR), respectively. Erythromycin was obtained from Abbott Laboratories (North Chicago, IL). All high-performance liquid chromatography eluents were of analytical reagent grade. Xenobiotic Treatments The N-substituted imidazoles were administered by intragastric injection. Clotz and nafimidone were suspended in 30% polyethylene glycol 400 given and given at doses of 75 or 150 mg/kg/day. Polyethylene glycol 400 has no effect on rat hepatic drug metabolizing parameters when administered alone (Ritter and Franklin, 1987a). NSI was administered as an aqueous solution at doses of 75, 125, 150 or 250 mg/kg/day. All traditional inducing agents were administered by intraperitoneal injection. Animals treated with SNF received 40 or 80 mg/kg/day in a corn oil suspension. Pb was given as an aqueous solution at 80 mg/kg/day and DEX was administered at a dose of 100 mg/kg/day in a 20/0 Tween 80 suspension. The N-substituted imidazoles, SNF, Pb and DEX were administered in a volume not exceeding 1 ml. Intragastric and intraperitoneal treatments with SSO were at a dose of 6 mmoles (1.3 g)/kg/day in an aqueous solution, injected as single doses of 2 ml or less. Sprague-Dawley rats were also exposed to SSO in the drinking water at concentrations of 1, 5, 15 or 30 mM, while Fischer 344 rats were treated with SSO via an indwelling gastrotomy at a dose of 2 mmoles/kg once every 12 hr. Acetaminophen was given as a single intravenous dose of 150 mg/kg in 20% (v/v) propylene glycol/saline. Untreated animals served as controls. 23 Hepatic Cell Fraction Preparation Following treatment, animals were sacrificed by decapitation and the livers perfused in situ via the hepatic portal vein with ice-cold isotonic saline. Gall bladders were removed from mouse and hamster livers. The livers from three to six mice were combined to yield single microsomal and cytosolic samples. Single rat and hamster livers were used for preparation of microsomal and cytosolic fractions. The cellular fractions were prepared according to a published method (Franklin and Estabrook, 1971). Liver samples were homogenized in 0.25 M sucrose (25% w/v) and the homogenate subjected to two low speed centrifugations to remove nuclear and mitochondrial fractions. The resultant supernatant, containing both microsomes and cytosol, was subjected to ultracentrifugation (106,000 X gmax for 55 min), with the cytosolic supernatant collected and stored (-700 C) for the assay of glutathione S-transferase and p-nitrophenol sulfotransferase activities. The pellet was resuspended in 0.15 M potassium chloride and centrifuged at 106,000 X gmax for 40 min. The microsomal fraction was recovered and homogenized in 0.05M Tris (pH 7.4)-sucrose (0.25 M). Microsomal and cytosolic protein concentrations were determined by the method of Lowry et al. (1951) using bovine serum albumin as a standard. Microsomal samples were adjusted to a final concentration of 30 mg/ml. All microsomal and cytosolic assays were performed after storage of the fractions at -700C for 18 to 72 hr. Glutathione Concentration Total hepatic glutathione concentration (reduced plus oxidized) was determined by the kinetic method of Akerboom and Sies (1981). Briefly, 30-50 mg samples were removed from the excised livers and homogenized in 3% trichloroacetic acid. The homogenates were centrifuged at 1000 rpm for 10 min and aliquots of the supernatants analyzed for glutathione content at 412 nm in the presence of 5,5'-dithio- 24 bis-(2-nitrobenzoic acid), glutathione reductase and NADPH. All glutathione determinations were conducted on animals sacrificed between 9:00 and 10:00 A.M. to minimize errors arising from the diurnal variation in hepatic glutat~lione concentration (Beck et aI., 1958). Microsomal and Cytosolic Parameter Analyses Hepatic P450 concentration was quantified from the dithionite-reduced carbon monoxide difference spectrum (450-490 nm) using an extinction coefficient of 91 mM-1 cm-1 (Omura and Sato, 1964). p-Nitroanisole demethylase activity at 250C was determined 'from the rate of p-nitrophenol generation as continuously monitored at 417 nm (Netter and Seidel, 1964). E ROD (Klotz et aI., 1984) and PROD (Lubet et aI., 1985) activities at 250C were determined from the rate of resorufin production detected at a wavelength of 572 nm. EY deM activity, at 370C and in the presence of 1 mM erythromycin, was determined by monitoring the generation of formaldehyde at 412 nm (Nash, 1953). p-Nitrophenol hydroxylase activity at 250C was determined by monitoring the formation of 4-nitrocatechol at 515 nm using a modified method (2 mM concentration of the substrate in a potassium phosphate buffer, pH 6.8) of Koop (1986) and Reinke and Moyer (1985). Microsomal UDP-glucuronosyltransferase activity toward p-nitrophenol was determined in the presence of 0.05% of the nonionic detergent Triton X-100, 1.25 mM uridine diphosphoglucuronic acid and a 0.1 mM concentration of the aglycone. Detergent-activated UDP-glucuronosyltransferase activities toward 1-naphthol (0.6 mM), morphine (2.5 mM), estrone (0.15 mM) and testosterone (0.6 mM) were determined by the method of Liu and Franklin (1984). The glucuronides were separated by reverse phase high-performance liquid chromatography on a 0.26 X 25 cm ODS/SIL-X column (Perkin-Elmer, Norwalk, CT) using a mobile phase consisting of 160/0 acetonitrile and 8% methanol in 20 mM potassium phosphate (pH 2.1 ) and 2.5 mM sodium 25 dodecyl sulfate. The eluate was continuously monitored at 220 nm. p-Nitroanisole served an internal standard. Cytosolic glutathione S-transferase activity toward 1-ct"lloro-2,4-dinitrobenzene at 250C was analyzed by formation of the glutathione conjugate at 340 nn1 (Habig and Jakoby, 1981). The disappearance of p-nitrophenol at 400 nm was monitored and used to assess cytosolic sulfotransferase activity at 250C (Ritter and Franklin, 1987a). Acetaminophen Disposition Studies Prior to the acetaminophen disposition studies, one group of Sprague-Dawley rats was exposed to 30mM SSO-supplemented drinking water for 6 days and one group of Fischer 344 rats was administered 2 mmoles SSO/kg once every 12 hr over 5.5 days via an indwelling feeding gastrotomy. One day before the acetaminophen pharmacokinetic studies, all rats were placed under mild methoxyflurane anesthesia and an indwelling cannula inserted in the right jugular vein (Weeks and Davis, 1964). Animals were housed individually in metabolism cages. After intravenous acetaminophen administration (150 mg/kg), blood samples (0.25 ml) were collected via the jugular vein cannula over the following 4 hours, immediately transferred to heparinized glass capillary tubes and centrifuged to yield plasma. Urine was collected for 24 hr after acetaminophen dosing. The concentrations of acetaminophen, acetaminophen glucuronide and acetaminophen sulfate in plasma and urine were determined by the high-performance liquid chromatographic method of Corcoran et al. (1985). Data Analysis The values for the hepatic parameters of control (untreated) mice, hamsters and rats are shown in Table 1. The control values shown in Table 1 are used throughout the thesis except for the acetaminophen Table 1. Hepatic parameters associated with drug metabolism in untreated mouse, hamster and rat. ~12~~i~§ Mous~ Hamster .B9t Strain QE-] 0, Be gQlden S~[ian ScraQue- Dawle~ Earamete[ GSHa nd 7.03±0.07(4) 5.72±D.17(3) nd 6.93±0.27 (16) Phase Ib P450 1.04±0.05(18) 0.78±0.03(12) 0.66±0.05(12) 0.97±0.06 (5) 0.79±0.04 (15) pNA deM 2.15±0.14(15) 1.13±0.13(14) 1.33±0.11 (13) 1.01±0.05 (5) 0.65±O.03 (16) EROD 0.03±0.01 (16) 0.03±0.01 (12) 0.02±0.01 (12) 0.04±0.01 (5) O.OO±O.OO (9) PROD 0.00±0.00(15) 0.00±0.00(13) 0.00±0.00(12) O.OO±O.OO (5) O.OO±O.OO (8) EYdeM 1.15±0.08(15) 0.93±0.06(11 ) 0.81±0.06(12) 1 .11 ±O .15 (5) 0.52±0.07(4) pNPH 0.62±0.04(14) 0.50±0.04(14) O. 69±0.07( 10) 1.29±0.04 (5) 0.47±O.03 (10) Phg§e lib UDPGt-pNP 10.96±0.44(16) 7.39±0.34(14) 9.37±0.52(14) 8.47±0.25 (5) 6.83±O.37 (16) -N nd 15.66±2.26(8) 7.48±2.36(5) nd 63.98±2.86 (16) -M nd 15.78±1 .00(8) 13.87±1.97(5) nd 15.44±1.35 (16) -E nd 0.11±0.01 (8) 0.12±0.02(5) nd 0.17±0.01 (16) -T nd 1.42±0.20(8) 1.55±O.38(5) nd 4.44±0.14 (16) GSHt 5133± 255(17) 4802±298(13) 3886±243( 13) 10740±1050(5) 1666±136 (16) PSt nd 0.70±0.06(4) 0.96±O.21 (3) nd 1.82±0.12 (16) aJlmoles/g liver bnmoles/mg protein or nmoles/min/mg protein ± SE; number of determinations are given in parentheses. nd = not determined N 0) 27 studies. Unless stated otherwise, subsequent data are expressed as mean percent of control ± percent SE of the mean (calculated as the treated SE divided by the mean control value X 100), except for EROD and PROD activities, which are given as activities (nmoles or pmoles/min/mg protein). Statstical comparisons between control and treated groups were performed using a two-tailed Student t-test for unpaired sample means. Because of the often large number of control determinations, differences are reported as significant if p ~ 0.03 so as to avoid the implication of biological significance to minor statistical changes. CHAPTER 3 SPECIES AND STRAIN COMPARISONS IN XENOBIOTIC-MEDIATED INDUCTION N-Substjtuted Imjdazoles and B-Naphthoflayone Treatment of mice (CF-1 strain) with Clotz at 75 mg/kg/day for 3 days followed by a 2-day wait before sacrifice caused only small elevations in hepatic Phase I and Phase II parameters (Figure 2). DOSing at 150 mg/kg/day resulted in dose-dependent increases in P450 concentration and p-nitroanisole demethylase [a P450 isozyme nonspecific activity (Guengerich et aI., 1982b)], EROD, PROD and EY deM activities. However, the relative increases were small and not proportional to the doubling of the daily' Clotz dose. p-Nitrophenol hydroxylase activity [catalyzed by the ethanol- and isoniazid-inducible P450llE1 isozyme] (Reinke and Moyer, 1985) and microsomal Phase II p-nitrophenol UDP-glucuronosyltransferase activity tended to be decreased at the higher dose compared to the lower dose. Cytosolic glutathione S-transferase activity was not affected by either Clotz dose. Similarly, 3 daily doses of NBI at 75, 125, 150 or 250 mg/kg followed by 2 days without treatment before sacrifice resulted in minor dose-related increases in P450 concentration and p-nitroanisole demethylase, EROD and EY deM activities (Figure 3). Administration of a single 300 mg/kg dose of NBI resulted in death within 6 hr after dosing. No dose-dependent effects were observed for p-nitrophenol hydroxylase, p-nitrophenol UDP-glucuronosyltransferase and glutathione S-transferase activities. 29 Figure 2. Effect of two clotrin1azole dosing regimens on hepatic drug metabolizing enzymes in the CF-1 mouse. Mice were treated with clotrimazole for 3 days at 75 or 150 mg/kg/day and sacrificed 2 days after the final dose. The hepatic Phase I and Phase II parameters were determined as described in Methods and Materials. The nUrTlber of determinations for both treatment regimens ranged from 3 to 5. Data are expressed as a percent of the untreated control values shown in Table 1 ± SE except EROD and PROD, which are shown as activities (pmoles/min/mg protein). * significantly different at p ~ 0.03 compared to untreated controls (Table 1 ). o o ~ C\J C\J + +('1) ('I) .. .. 0) 0).;:£ ._;_: £-0­) O)E Eo L()L() f'..r- •• o o M o o C\J o o r- 30 lHS8 HdNd V\lap A3 OOtJd 00tJ3 OSvd o 31 Figure 3. Effect of four N-benzylimidazole dosing regimens on hepatic drug metabolizing enzymes in the CF-1 mouse. Mice were treated with N-benzylimidazole for 3 days at 75, 125, 150 or 250 mg/kg/day and sacrificed 2 days after the final dose. The hepatic Phase I and Phase II parameters were determined as described in Methods and Materials. The number of determinations for each treatment regimen ranged from 3 to 6. Data are expressed as a percent of the untreated control values shown in Table 1 ± SE except EROD and PROD, which are shown as activities (pmoles/min/mg protein). * significantly different at p ~ 0.03 compared to untreated controls (Table 1 ). C\JC\JC\J C\J+++ +('t')('t')('t') ('t') .. .. .. ... 0) 0) 0) 0)..::£ ..::£ ..::£ ~0l0l0l O)EEE LEO LC\JO LOO OLO r-.....~~C\J o o C\I o a or- 32 IHS8 HdNd LAJ8p A3 OOtid 00tl3 09vd a 33 PROD activity was not detected in microsomes from naive or NBI-treated mice, even at the highest nonlethal dose. To determine if the 2-day delay between the final Clotz or NBI dose and sacrifice was sufficient for clearance of residual inhibitor and, hence, appropriate for determination of P450-dependent activities in the CF-1 mouse, mice were treated with single 150 mg/kg doses of either Clotz or NBl, followed by a wait of between 12 and 72 hr before sacrifice (Figure 4). Following Clotz administration, total hepatic P450 concentration tended to be slightly elevated for up to 48 hr and then declined toward basal concentrations. p-Nitroanisole demethylase activity was inhibited at 12 hr but gradually increased over the following 36 hr to achieve a maximum mean activity of 3.50 nmoles/min/mg protein at 48 hr. Subsequent to NBl treatment, P450 concentration and p-nitroanisole demethylase activity reached maximum values at 36 hr and remained essentially constant over the following 12 hr. p-Nitroanisole demethylase activity was inhibited for up to 24 hr after NBI treatment, but steadily increased thereafter to achieve a 36% elevation in activity (to a mean of 3.10 nmoles/min/mg protein) at 36 and 48 hr. Hence, the 48-hr wait between dosing and sacrifice appears to allow for proper determination of Clotz- and NBI-mediated induction of P450 in mice. All further studies in mice and hamsters investigating the effects of N-substituted imidazoles were performed using treatment regimens of 3 daily 150 mg/kg/day doses, followed by a 48-hr interval after the final dose prior to sacrifice. Treatment of rats with Clotz, NBI or nafimidone also involved a 48-hr interval but with daily dosing at 75 mg/kg for 3 days. Other studies have shown marked inductions of rat hepatic drug metabOlizing enzymes by Clotz and NBI at this dose (Ritter and Franklin,1987a; Papac and Franklin, 1988). Table 2 provides a direct comparison between CF-1 mice and Sprague-Dawley rats for the induction of hepatic Phase I and Phase II drug metabolizing enzymes by Clotz. In the rat, Clotz treatment caused a 4-fold 34 Figure 4. The temporal-dependent effects of clotrimazole and N-benzylimidazole on CF-1 mouse hepatic cytochrome P450 concentration and p-nitroanisole demethylase activity. Mice were treated with a single dose of clotrimazole (150 mg/kg) or N-benzylimidazole (150 mg/kg) and sacrificed at various times after dosing. Hepatic cytochrome P450 concentration and p-nitroanisole demethylase activity were determined as described in Methods and Materials. Values for untreated mice (time "0") are shown on the left axis of each panel and each point on the graphs was derived from a single microsomal preparation. 2.5 2.5 ''eQ2i 2.0 .C~ 2.0 r:J a. a. C) C) 1.5i B -E 1.5 .E r:J II) Q) IQI)) 0E 1.0 0E 1.°1 r:J ",L0¢O s -L0cO EIB'lIl a.. 0.5 a"¢.. 0.5 0.0 0.0 0 12 24 36 48 60 72 0 12 24 36 48 time (hours) after Clotz dosing time (hours) after NBI dosing '2 4 C 4 '(i '(i EI e e a. a. C) 3 C) E E 3 C -c '-E .~ II) II) Q) 2 Q) 0 0 E E ,s ,s :::E :E Q) Q) -0 -0 zor.1 : «z &g a. a. 0 0 0 12 24 36 48 60 72 0 12 24 36 48 time (hours) after Clotz dosing time (hours) after NBI dosing w U1 36 Table 2. Clotrimazole-n1ediated induction of mouse and rat hepatic drug metabolizing enzyme activities. Species Strain Dosea Parameterb Phase I P450 pNA deM EROD PROD EYdeM pNPH Phase II UDPGt-pNP GSHt Mouse .c.E:.1 150,3+2 162 ± 8 * 173 ± 12 * 0.06 ± 0.01 0.05 ± 0.01 339 ± 10* 127 ± 13 76±8 * 120 ± 13 * flatC Sprague-Dawley 75,3+2 410 ± 20 * 410 ± 10 * 0.13 ± 0.01 0.70 (n=2) 1202 ± 59 * nd 121 ± 6 224 ± 10 * * a Mice (n=3) and rats (n=5) were treated (ig) with 150 and 75 mg Clotz/kg/day, respectively, for 3 days and sacrificed 2 days after the final dose. b Data are expressed as mean percent of control ± SE except for EROD and PROD, which are given as activities (nmoles/min/mg protein). c Data from Ritter and Franklin (1987a and 1987c). *significantly different at p ~ 0.03 compared to untreated controls (Table 1). nd = not determined 37 increase in P450 concentration and p-nitroanisole demethylase activity. Of the Phase I isozyme-specific substrate activities employed, PROD and EY deM activities were notably enhanced by Glotz in rats. PROD activity is considered a selective marker for Pb-mediated induction of P450 in rats and mice (Burke et aL, 1985; Kelly et aI., 1990; Lubet et aL, 1990; Lum et aL, 1986; de Waziers et aI., 1990), while EY deM activity is a selective probe for glucocorticoid-mediated induction in rats (Wrighton et aI., 1985a, 1985b). Hence, Glotz exhibits a mixed Pb-/glucocorticoid-like induction profile in Sprague-Dawley rats. In contrast, the GF-1 mouse was much less responsive to Glotz-mediated induction of Phase I activities, with PROD and EY deM activities being 14- and 3.5-fold less than seen in rats following Glotz administration. Moreover, whereas Glotz induced cytosolic glutathione S-transferase activity 2-fold in rats, the N-substituted imidazole failed to induce this transferase in GF-1 mice. UDP-glucuronosyltransferase activity toward p-nitrophenol was not induced in either species by Glotz treatment. Microsomes prepared from Sprague-Dawley rats treated with NBI for 3 days followed by a 2-day delay before sacrifice showed a 3-fold increase in P450 concentration (Table 3) and a shift in the ferrous P450-carbon monoxide absorbance maximum to 448 nm. The N-substituted imidazole can be characterized as possessing mixed polycyclic aromatic hydrocarbon-/Pb-type inductive properties in the rat due to its ability to markedly increase Phase I EROD and PROD activities, respectively, and to induce the Phase II glucuronidations of p-nitrophenol, 1-naphthol and morphine. For those parameters suggestive of Ah receptor activation (Le., EROD and p-nitrophenol and 1-naphthol UDP-glucuronosyltransferase activities), NBI was as effective an inducer as the recognized Ah receptor agonist BNF. However, NBI caused only trivial changes in EROD and UDP-glucuronosyltransferase activities in the Ah-responsive B6 mouse. Furthermore, no spectral shift away from 450 nm for the reduced P450- Table 3. The induction of mouse, hamster and rat hepatic drug metabolizing enzymes by N-benzylimidazole and (3-naphthoflavone. ~ Mouse ~ RaP Strain CF-1 .Q2 .eg golcx:m Syrian SQrwl&~y ~ m EN= N3I fH: I\BI .Ei£ ~ EJ\F I\BI EN= rP 3 6 4 4 6 6 4 S 11 8 Phase iG P-450 128±10* B3±5 163±2S* 108±4 13S±8* 270±20· 169±12* 143±7* 320±16* 162±6* pNAdeM 146±14· 109±S 164±16(S)* 141±11 164±24· 398±41* 2S0±20* 123±8* 768±82* 382±4S" EROD 0.1 0:t0.0 lit 0.20:t0.03 1t 0.13:t0.02" O. 09:.t0. 01 It O. OS:.tO.OI " 2. 62:t0. 08 It 0.06:t0.01 0.29:tD. 02 It 2. 24:.t0.17 .. I. 66:t0.2 I " PROD O.OOM.OO 0.00:t0.00 0.02:t0.01* O.00:t0.00 O.00:.t0.00 0.00:t0.00 0.00:1:0. 00 O.00:tD.00 0.16:.t0.03(7r 0.03:t0.01 EYdeM 161±28* 99±8 417±S· 126±16 202±23* 134±27(4) 90±7 44±5· 200±18(4)* 72±8 pNPH 100±10 110±7 100±13(6) 93±11 90±14 14S±7(4)* 112±6 90±8 nd rd Phase liG UDPGt-pNP 97±11 101±6 85±S(5) 124±6* 96±9(7) 231±26· 132±25 112±5 308±4S" 49S±16(7)* -N nd rd 74±10 70±3 168±40(4) 354±123 nd nd 250±4S* 2S0±47(7)* -M rd rd 106±1O 110±6 136±11(4) 179±20* nd nd 222±29* 129±8(7) -E nd rd 114±10 102±4 118±18(4) 150±16 nd nd 89±S 137±11(7)* -T rd rd 120±26 133±14 106±30(4) 100±20 nd nd 120±16 100±18(7) GSHt 110±2 74±8· 93±12 92±9 118±9(7) 139±14* 10S±20 66±8* 198±12* 148±11 (7)* aMice and hamsters were treated ig with 1S0 mg NBllkg once a day for 3 days and sacrificed 2 days after the final dose. Rats were similarly treated with NBI, but at 7S mglkg. All animals were administered BNF ip at 80 mglkg once a day for 4 days and sacrificed 1 day after the final dose. bNumber of determinations unless indicated in parentheses or not determined (nd); "significantly different at p ~ 0.03 from untreated controls (Table 1). CData are expressed as mean percent of control (Table 1) ± SE except for EROD and PROD, which are given as activities (nmoles/minlmg protein). w OJ dData from Papac and Franklin (1988). 39 carbon monoxide complex was detected in the microsomes from NBI-treated B6 mice, in contrast to that observed for rats. The Ah-responsive golden Syrian hamster was also unresponsive to NBI-induction of EROD and p-nitrophenol UDP-glucuronosyltransferase activities and NBI treatment did not change the ferrous P450-carbon monoxide absorbance maximum. The order of magnitude lesser response of EROD activity induction by BNF in hamsters compared to rats has also been reported for 3-methylcholanthrene (Blaich et aI., 1988; Lubet et aI., 1990). Neither NBI nor BNF caused major increases in EROD or UDP-glucuronosyltransferase activities in the Ah-nonresponsive CF-1 and D2 mouse strains compared to the Sprague-Dawley rat. The most pronounced effect of NBI on the investigated drug metabolizing enzyme activities in rodents other than rats was a 4-fold increase in EY deM activity in D2 mice. Nafimidone induced D2 and B6 mouse and Sprague-Dawley rat hepatic P450 concentration by approximately 2-fold and p-nitrophenol demethylase activity from 2.5- to 4.6-fold (Table 4). Nafimidone was a 3-fold better inducer of EROD activity (Table 4) than NBI (Table 3) in the Ah-responsive B6 mouse. However, the rat displayed a conspicuously greater sensitivity to nafimidone-mediated induction of E ROD activity, despite being treated with half the dose administered to the B6 mice. Nearly equivalent increases in EY deM activity were detected in CF-1, D2 and B6 mice and Sprague-Dawley rats following nafimidone administration. UDP-glucuronosyltransferase activity toward p-nitrophenol was also increased by nafimidone in the rat (2.5-fold) and to lesser extents in B6 (1.5-fold) and CF-1 (1.4-fold) mice. While minor in magnitude, these nafimidone-mediated increases in mouse p-nitrophenol glucuronidation contrast with the absence of increases in NBI-treated mice (Table 3). Only small increases in glutathione S-transferase activity were observed in rats and the three mouse strains following nafimidone treatment. The effect of 40 Table 4. Nafimidone-mediated induction of mouse and rat hepatic drug metabolizing enzyme activities.a SC~Qi~S Mouse .Bat Strain QE:1 D2 .62 SQrggu~-Dgwle~ Ebas~ Ib P450 131 194 236 219 ± 17 pNA deM 161 247 383 460 ± 64 EROD 0.08 0.06 0.23 1.45 ± 0.17 PROD 0.00 0.01 0.07 0.05±0.04 EYdeM 175 160 (n=1) 214 225 ± 59 pNPH 52 nd 100 88± 10 Phg~~ lib UDPGt-pNP 138 102 154 255 ± 30 GSHt 126 125 123 122 ± 21 a Mice (n=2) and rats (n=3) were treated (ig) with 150 and 75 n1g nafimidone/kg/day, respectively, for 3 days and sacrificed 2 days after the final dose. b Data are expressed as mean percent of control (Table 1) ± SE except for EROD and PROD, which are given as activities (nmoles/min/mg protein). nd = not determined 41 nafimidone on hamster drug metabolizing enzymes was not determined. Phenobarbital and Dexamethasone The inductive effects of Pb and DEX between Sprague-Dawley rats, golden Syrian hamsters and three mouse strains are compared in Table 5. Not unexpectedly, Pb induced PROD activity in the rat and CF-1 mouse. The effect of Pb administration on PROD activity in the other two mouse strains and in hamsters was not determined. In the CF-1 mouse, however, DEX was as effective an inducer of PROD activity as Pb. Additionally, while DEX was an excellent inducer of EY deM activity in rats and all strains of mice, it failed to increase this purported steroid-selective inducible isozyme activity in hamsters. Rather, the synthetic steroid caused a 3-fold increase in p-nitrophenol hydroxylation, an activity reflective of ethanol (Reinke and Moyer, 1985) and isoniazid (Ryan et aI., 1985) induction in rats. Also surprising was the 3-fold DEX-mediated increase in the Phase II glucuronidation of 1-naphthol in B6 mice, a reaction indicative of polycyclic aromatic hydrocarbon- or BNF-like induction in rats (Table 3; Watkins et aI., 1982; Bock et aI., 1983). The DEX-mediated increase in UDP-glucuronosyltransferase activity toward 1-naphthol in the B6 mouse was of the same magnitude as that observed following BNF administration, but unlike BNF, DEX did not induce p-nitrophenol UDP-glucuronosyltransferase activity. To determine if DEX could facilitate NBI induction of EROD activity in the Ah-responsive B6 mouse in a manner analogous to that reported for 3-nlethylcholanthrene induction in Ah-responsive rats (Sherratt et aI., 1989), DEX was administered concomitantly with NBI for 3 days (Table 5). DEX alone was given for 1 day following the last NBI dose and the mice were sacrificed 1 day after the final DEX dose. The combined DEX plus NBI treatment regimen did not induce EROD activity in the B6 mouse, unlike the previous report on the in vivo potentiation of polycyclic aromatic Table 5. The induction of mouse, hamster and rat hepatic drug metabolizing enzymes by phenobarbital and dexamethasone. Species Mouse Hamster Rat Strain CF-l .Q.2 66 golden S!lriao Sprague-Dawle!l Inducer Fb DEX DEX DEX DEX± NBI Pbd DEX Pb DEX Dosea 80.4+ 1 100.4+ 1 100.4+ 1 100.4+ 1 100.4+1;150,3+2 80.4+ 1 100,4+1 80,3+1 100,3+ 1 nb 4 4 3 4 2 3 5 4 3 Phase IC P-450 186±14* 242±5* 247t8* 262±29* 348 lS2±8* 74t8* 271±18* 286±4* pNA deM 2S8±26* 173±18* 189±27- 281±26* . 281 326±18* 120±7(4) 708±47* 382±10* EROD O.14±0.02" O. 08:tO. 0 1* 0.04±O.01" 0.061:0.01(5)* 0.00 nd O.04±O.OI 0.10±O.06 0.08±0.06 PROD 0.16:tO.OI* 0.25±O.01* 0.09:t0.02* O.I2:tO.OI(S)* 0.23 nd O.OOIO.OO 0.48:t0.OB" 0.031:0.01 EYdeM 296±1 a* 585±43* 804±8* 867t181* 1207 134±13 109±18 94±58* 1366±4* pNPH 162±19* 139±10* 13a±16 107±2l nd nd 30S±34* 294±52* 304±32* Phase IIc UDPGt-pNP 148±13(3)* 92±17(3) 10l±a 100±6(3) 67 159±8 - 116±2* 178±27 a5±8 -N nd nd 132±11 306±33(3)* nd 122±7 nd 137±14 96±11 -M nd nd 55±S* 103±S(3) nd 218±24* nd 643±73 * 342±44* -E nd nd 88±13 148±64(3) nd 108±a nd 134±22 lS3±6* -T nd nd 46±12 77±2(3) nd 210±21" nd 141±19 122±4 GSHt 134±15* 100±20 86±10 134±3(6)* 162 122±8 78tS 180±13* l68±25* PSt nd nd 1S4±31 104±10(5) 96 ao±a nd 107±3 63±11 amglkg, days of treatment + period (in days) between last dose and sacrifice. bNumber of determinations unless indicated in parentheses or not determined (nd). CData are expressed as mean percent of control (Table 1) ±.SE except for EROD and PROD, which are given as activities (nmoleslmin/mg protein). dData from Ritter and Franklin, (1987d). +::> *significantly different at p $ 0.03 compared to untreated controls (Table 1) N 43 hydrocarbon-inducible EROD activity by DEX in rats. The increases in P450 concentration and PROD and EY deM activities in the DEX+NBI group compared to mice treated with DEX or NBI alone could be suggestive of a greater than additive effect but, because of the limited number of determinations (n = 2), attributing a DEX+NBI inductive synergisism for these parameters would be presumptive. CHAPTER 4 EFFECT OF BSO TREATMENT ON XENOBIOTIC­MEDIATED INDUCTION IN THE SPRAGUE-DAWLEY RAT N-Benzylimidazole Analogous to the cytosolic glucocorticoid receptor, the cytosolic Ah receptor appears to possess sulfhydryl groups critical for receptor-ligand binding integrity (Denison et aI., 1987; Kester and Gasiewicz, 1987). Since one of the many functions of glutathione is to maintain cellular protein thiols in the reduced state, it was hypothesized that prolonged depletion of hepatic glutathione might diminish the binding affinity of the Ah receptor by allowing oxidation of Ah receptor sul'fhydryl groups, thereby rendering the Ah-responsive rat unresponsive to receptor agonists. The gamma-glutamylcysteine synthetase inhibitor BSO (Griffith and Meister, 1979; Griffith, 1982) was chosen to deplete glutathione since other glutathione-depleting agents are less specific in their biological effects (Anders, 1978; Maines, 1982; Costa and Murphy, 1986). Table 6 summarizes the effect of BSO treatment on NBI-mediated induction of Sprague-Dawley rat Phase I and Phase II hepatic drug metabolism parameters. Exposure to BSO was accomplished by adding the sulfoximine to the drinking water and allowing access to the supplemented drinking water for 13 consecutive days prior to sacri'fice (the BSO group). Dosing with NBI was initiated after 8 days on 30mM BSO drinking water (the BSO+NBI group) or in the absence of BSO exposure (the NBI group). 45 Table 6. The effect of buthionine sulfoximine and N-benzylimidazole, alone and in combination, on Sprague-Dawley rat hepatic drug metabolizing enzyme activities. Treatment as.Qb Paramelera GSH 18 ± 2· Pbas~ I P450 94±8 pNA deM 95 ± 11 EROD O.OO± 0.00 PROD o.oo± 0.00 EYdeM 111 ± 15 pNPH 102 ± 2 Pha~e II UDPGt-pNP 291 ± 31 • -N 190 ± 11 • -M 175 ± 18 • -E 112 ± 6 -T 121 ± 7 • GSHt 172 ± 8 * PSt 81 ± 15 NSlc 104 ± 10 340 ± 22 * 360 ± 152 * 1.41 ± 0.16 O.OO± 0.00 236 ± 48 • 76±26 135 (2) 130 ± 22 170 ± 33 * 64± 12 105 ± 17 191 ± 7 * 76±8 * * 27± 3· 289 ± 14 * 691 ± 38 * 1.36 ± 0.30 0.12 ± 0.04 296 ± 54 • 187 ± 12 * 350 ± 25 * 293 ± 39 * 297 ± 15 * 151±10* 164 ± 5 * 301 ± 11 * 52±3 * • • a Data are expressed as mean percent of control (Table 1) ± SE except for EROD and PROD, which are given as activities (nmoles/min/mg protein); n = 3 unless indicated in parentheses. b Rats were maintained on drinking water containing 30mM SSO for 13 consecutive days prior to sacrifice. The mean daily SSO dose was 6.3 mmoles/kg. c Rats were treated (ig) with 75 mg NSllkg once a day for 3 days and sacrificed 2 days after the thi rd dose. d Rats were maintained on 30mM SSO-drinking water for 8 days prior to initiating NSI treatment and were continued on SSO until sacrifice. The mean daily SSO dose was 5.5 mmoles/kg. * significantly different at p ~ 0.03 compared to untreated controls (Table 1); statistical differences between treatment groups not shown. 46 Continuous access to the BSO-supplemented drinking water was provided for the BSO+NBI group until sacrifice and all animals were sacrificed on the same day. Hepatic glutathione concentrations at the time of sacrifice in the BSO and BSO+NBI groups were diminished by 82% and 730/0, respectively (Table 6). As with a single ip dose of BSO in mice (Drew and Miners, 1984; White et aI., 1984), 13-day 30mM BSO drinking water exposure (the BSO group) did not affect a range of rat hepatic monooxygenase activities. BSO treatment had no statistically significant effect (p > 0.03) on NBI-mediated induction of P450 concentration and most monooxygenase activities, including p-nitroanisole demethylase and PROD activities. However, combined BSO+NBI treatment approximately doubled p-nitrophenol hydroxylation above the activities in the untreated control (Table 1), BSO alone and NBI alone groups. Additionally, the lack of PROD induction by NBI alone is contrary to that reported by Papac and Franklin (1988) shown in Table 3. In contrast to the lack of effect on Phase I activities, there were marked increases in several Phase II activities attributable to BSO exposure. Microsomal UDP-glucuronosyltransferase activity towards p-nitrophenol was induced 3-fold and both 1-naphthol and morphine glucuronidation rates were induced 2-fold in the BSO group. A lesser, but statistically significant increase (21 %) in testosterone glucuronidation was also found. Cytosolic glutathione S-transferase activity was also induced (72%) in the BSO group, while p-nitrophenol sulfotransferase activity was not increased. The greater increases in glucuronosyltransferase and glutathione S-transferase activities in the BSO+NBI group compared to the BSO group appear to reflect more of an additive, rather than synergistic, effect between the two xenobiotics. 47 B-Naphthoflavone In the, above BSO+NBI study, it was assumed that the induction of EROO activity by NBI in the rat required binding of the N-substituted imidazole with the cytosolic Ah receptor. Given the lack of EROO activity induction following NBI treatment of Ah-responsive B6 mice, such an assumption may not be valid. Therefore, the hypothesis that sustained depletion of glutathione by BSO may mitigate the Ah receptor-dependent induction of rat liver EROO activity was further pursued with the acknowledged Ah receptor agonist BNF (Table 7). Anin1als receiving only BSO (the BSO group) were exposed to drinking water containing BSO at a concentration of 30mM for 5 consecutive days before sacrifice, as compared to the 13-day exposure employed in the above NBI study. Rats treated with BNF alone (the BNF group) were administered the flavone at 40 mg/kg once a day for 2 days and sacrificed 24 hr after the second dose. The animals receiving both compounds (the BSO+BNF group) were started on the BSO-supplemented drinking water 3 days prior to initiating BNF treatment and were maintained on the BSO solution until sacrifice. All rats were sacrificed on the same day. Similar to the previous experiment, liver glutathione concentrations in the BSO and BSO+BNF groups were depleted by 83% and 76%, respectively (Table 7). As was generally the case in the BSO and NBI experiment, BSO treatment had no effect on the BNF-mediated induction of rat hepatic EROO activity and other Phase I parameters. However, in contrast to the induction of p-nitrophenol hydroxylase activity subsequent to combined BSO+NBI treatment (Table 6), treatment of rats with BSO+BNF was not able to overcome the 500/0 decrease in p-nitrophenol hydroxylase activity resulting from BNF administration alone (Table 7). Again, treatment with 30mM BSO drinking water alone (in this experiment for 5 days) had little or no effect on the Phase I parameters but caused, with the exception of morphine glucuronidation, a similar enhancing effect 48 Table 7. The effect of buthionine sulfoximine and B-naphthoflavone, alone and in combination, on Sprague-Dawley rat hepatic drug metabolizing enzyme activities. Treatment na Parameterb GSH Phase I P450 pNA deM EROD PROD EYdeM pNPH Phase II UDPGt-pNP -N GSHt PSt -M -E -T ~c 3 72± 1 112 ± 3 0.01± 0.00 O.OO±O.OO 121 ± 12 83±6 383 ± 20* 172 ± 6 * 124±3 79±7 134 ± 4* 173 ± 9* 85±4 BNFd 3 99 ± 11 (7) 128 ± 15 (7) 295 ± 17 (6)* 1.16 ± 0.08 (7) O.OO±O.OO 104 (2) 50±6 * 189 ± 20 (7) 86 ± 14 39±2 * 78±8 69 ± 14* 130 ± 11 (7) 155 ± 7* * * BSO+BNFe 4 102 ± 9 317 ± 25* 1.22 ± 0.13* 0.00 ± 0.00 110± 19 79±8 485± 8 * 158 ± 14 76 ± 12 85±4 96±4 167 ± 8* 76±30 * a Number of determinations unless indicated in parentheses. b Data are expressed -as n1ean percent of control (Table 1) ± SE except for EROO and PROD, which are given as activities (nmoles/min/mg protein). c Rats were maintained on drinking water containing 30mM BSO for 5 consecutive days prior to sacrifice. The mean daily BSO dose was 3.5 mmoles/kg. d Rats were treated (ip) with 40 mg BNF/kg once a day for 2 days and sacrificed 1 day after the second dose. e Rats were maintained on 30mM BSO-drinking water for 3 days prior to initiating BNF treatment and were continued on BSO until sacri'fice. The mean daily BSO dose was 3.0 mmoles/kg. * significantly different at p ~ 0.03 compared to untreated controls (Table 1); statistical differences between treatment groups not shown. 49 on Phase II UDP-glucuronosyltransferase and glutathione S-transferase activities as found for the 13-day exposure regimen. Treatment of rats with 40 mg BNF/kg/day for 2 days (the BNF group) increased p-nitrophenol glucuronidation by only 90 % , did not induce 1-naphthol UDP-glucuronosyltransferase activity and decreased morphine and testosterone glucuronosyltransferase activity to 40% and 70% of control, respectively, effects differing from those observed following treatment with 80 mg BNF/kg/day for 4 days (Table 3). The increases in glutathione S-transferase and Phase I EROD activities following 40 mg BNF/kg treatment for 2 days were similar to those observed with administration of 80 mg BNF/kg for 4 days. A comparison between these data sets implies a dose-dependency for the BNF-nlediated induction of rat UDP-glucuronosyltransferase isozymes. The combined effect of BSO and BNF on Phase II activities was generally reflective of how both xenobiotics changed the activities individually. The approximate 5-fold increase in UDP-glucuronosyltransferase activity toward p-nitrophenol found in the BSO+BNF group was similar to the sum of the increases caused by BSO and BNF alone, while 1-naphthol, morphine and testosterone glucuronidation rates in the BSO+BNF group were intermediate of those observed in the BSO alone and BNF alone groups. Estrone glucuronosyltransferase activity was decreased by approximately 20% in all three treatment groups. Glutathione S-transferase activity following combined treatment was essentially the same as seen in the BSO alone group, while BSO appeared to prevent the increase in p-nitrophenol sulfotransferase activity caused by BNF. CHAPTER 5 SPECIES- AND DOSE-DEPENDENCIES FOR SSO-MEDIATED INDUCTION Sprague-Dawley Bat To further investigate the apparent Phase II-selective inductive effect of SSO, Sprague-Dawley rats were administered SSO by different routes and at various daily doses. The effects of ip, ig and drinking water SSO treatments on rat hepatic glutathione and P450 concentrations and selected monooxygenase activities are compared in Table 8. As expected, hepatic glutathione concentration was diminished by the 3 day ip and ig (6 mmoles SSO/kg/day) dosing regimens and following 3 and 7 consecutive days of exposure to 30mM SSO drinking water. The glutathione concentration rebounded to 126% of control in rats exposed to drinking water containing SSO for 6 days but returned to unsLipplemented drinking water for 1 day prior to sacri'fice. In general, only minor and inconsistent changes in the Phase I parameters were observed following any of the diverse SSO exposures. For example, the 7-day 30mM SSO drinking water exposure increased P450 concentration by 290/0 but failed to induce p-nitroanisole demethylase, EROD or PROD activities and both ip and ig administration of SSO caused 25% decreases in p-nitroanisole demethylase activity without affecting P450 concentration. In contrast, several Phase II activities were conSistently induced by ip, ig and/or drinking water administration of SSO. Microsomal UDP-glucuronosyltransferase activity toward p-nitrophenol was induced 51 Table 8. Hepatic glutathione concentration and Phase I and Phase II parameters in Sprague-Dawley rats following buthionine sulfoximine exposure by various routes. a Exposure Regimen 3 day ip 3 day ig 3 day dw 6+1 day dw 7 day dw 3 4 4 4 4 Pa[am~t~[ GSHc 47±11 * 30±7 * 22±2 * 126±5* 32±2* nlicrosomal protein yieldd 75 80 99 102 95 PhgS~ IC P450 99±1 91±8 99±4 122±9 129±4* pNAdeM 75±2* 75±6 * 111±6 102±9 115±6 EROD nd O.OO±O.OO nd nd O.OOM.00 PROD nd O.OO±O.OO nd nd O.OOM.00 Phase IIc UDPGt-pNP 177±17* 244±15* 215±19 * 263±12* 246±14 * -N 98±6 122±7 214±22* 231±18* 211±20* -M 146±7 165±7* 143±10* 162±13* 145±16* -E 41±12 *- 65±12 * 88±6 112±12 94±6 -T 96±5 75±17* 120±6 152±10* 145±2 * GSHt 122±3 143±3* 113±13 177(2) 174±8* PSt 90±4 70±7 78±7 103(2) 108±5 aTwo groups of rats were subjected to 3 daily ip or ig doses of 6 mmoles SS~/kg and sacrificed 1 day after the third dose. Two other groups were provided 30 mM SSO-supplemented drinking water (dw) for 3 and 7 days. Another group (6+1) was allowed access to 30 n1M SSO dw for 6 days but returned to unsupplemented drinking water for 1 day prior to sacri'fice. Daily ingestion of SSO for dw exposures ranged from 3.2 to 4.4 mmoles/kg/day. bNumber of determinations unless indicated in parentheses. CData are expressed as mean percent of control (Table 1) ± SE except for EROD and PROD, which are given as activities (nmoles/min/mg protein). dData are expressed as mean percent of control (74 mg protein/liver; n=17). *significantly different at p ~ 0.03 compared to untreated controls (Table 1). nd=not determined. 52 (1.8- to 2.6-fold) by all of the SSO exposure regimens. All enteral BSO treatments induced morphine UDP-glucuronosyltransferase activity and drinking water exposures were effective in enhancing 1-naphthol glucuronidation. Intraperitoneal administration tended to be the least effective route of SSO exposure for inducing the glucuronosyltransferases. UDP-glucuronosyltransferase activity toward estrone was decreased after ip and ig BSO injections, as was the glucuronidation of testosterone after ig SSO administration, but drinking water exposure tended to increase testosterone glucuronidation. Glutathione S-transferase, but not p-nitrophenol sulfotransferase, activity was increased after ig and most drinking water SSO treatments. The increases in p-nitrophenol, 1-naphthol, morphine and testosterone glucuronosyltransferase activities and glutathione S-transferase activity after 7 consecutive days of exposure to 30mM BSO drinking water were similar to those of rats exposed for 6 days followed by a return to unsupplemented drinking water for 1 day (the "6+ 1 II group). Since the hepatic glutathione concentrations in the 7- and "6+ 1 "-day groups at the time of liver excision were 32% and 126% of control, respectively, the determined levels of induction were independent of glutathione concentratration at the time of sacrifice. To determine the sensitivity of the drug metabolism parameters to changes mediated by BSO, Sprague-Dawley rats were exposed to drinking water containing SSO at concentrations ranging 'from 1 mM to 30mM for 3 days (Table 9 and Figure 5). Hepatic glutathione concentration at the time of sacrifice was diminished by 30, 59, 75 and 78% following the 1, 5, 15 and 30mM BSO drinking water exposures, respectively. Neither P450 concentration nor p-nitroanisole demethylase activity were affected by drinking SSO of any concentration. Glutathione S-transferase and p-nitrophenol sulfotransferase activities were significantly increased (40%) and decreased (30%), respectively, following 3 days of the highest dose (30mM drinking water), effects that were more 53 Table 9. The effect of 3-day exposure to various concentrations of butt"iionine sulfoximine in the drinking water on Sprague-Dawley rat hepatic parameters. BSOb GSH 1mM 70±5 * 5mM 41 ±3 * 15mM 25±5 * 30mM 22±3 * P450 107 ± 2 100 ± 7 100 ± 5 96±3 Parameterfi pNA deM 91 ± 4 99 ± 14 88± 6 85 ± 6 GSHt 102 ± 7 132 ± 8 127 ± 5 140 ± 7* PSt 88 ± 7 73 ±9 83 ±6 70 ±7 * a Data are expressed as mean percent of control (Table 1) ± SE, n = 5. b The mean daily doses of BSO were 0.2, 0.9, 2.4 and 5.3 mmoles/kg for the 1, 5, 15 and 30mM solutions, respectively. * significantly different at p :5; 0.03 compared to untreated controls (Table 1). 54 Figure 5. The effect of 3-day exposure to various concentrations of buthionine sulfoximine in the drinking water on Sprague-Dawley rat hepatic microsomal UDP-glucuronosyltransferase activities. Rats were maintained on drinking water supplemented with buthionine sulfoximine at concentrations of 1 mM, 5mM, 15mM and 30mM for 3 days. Hepatic microsomal UDP-glucuronosyltransferase activities were determined as described in Methods and Materials. The number of determinations was 5 for each buthionine sulfoximine treament. Values for untreated rats (concentration "0") are shown on the left of each panel. * significantly different at p $; 0.03 compared to untreated controls (Table 1 ). 200 "'6 E 8 100 * "'6 .b c 200 8 100 rf!. o o g ac o * 300 200 100 o LlDPGt-1-naphthol o 5 UDPGt~p~nitrophenol o 5 15 30 200 UDPGt~morphine g c 8 100 * o 15 30 o 5 200 UDPGt-estrone UDPGt·testosterone g c 8 100 * o o 5 15 30 0 5 concentration of BSO (mM) in drinking water 55 15 30 15 30 56 pronounced than those observed in the previous experiment (Table 8). All UDP-glucuronosyltransferase activities were induced by a BSO drinking water concentration of 5mM, with the increases being similar to those seen after exposure to 30mM BSO drinking water. The 1 mM BSO concentration, while sufficient to lower liver glutathione concentration by 30%, was ineffective in enhancing UDP-glucuronosyltransferase and glutathione S-transferase activities, except for a small (160/0) increase in testosterone glucuronidation. 02 and B6 Strains of Mice While ig treatment of rats with 6 mmoles BSO/kg/day for 3 days lowered hepatic glutathione by 70% (Table 8), identical treatment of D2 and B6 mice din1inished glutathione concentration by only 40 and 360/0, respectively (Table 10). Other differences observed in the two strains of mice compared to rats (Table 8) following ig BSO treatment include the lack of significant increases in morphine UDP-glucuronosyltransferase activity and the absence of decreases in estrone glucuronidation. Both mouse strains were also much less responsive than rats to the induction of UDP-glucuronosyltransferase activity toward p-nitrophenol. The increases in murine 1-naphthol glucuronosyltransferase activity were slightly greater than that seen in rats following ig BSO administration, although the B6 mouse value failed to achieve significance. The increases in glutathione S-transferase activity were similar between both mouse strains and the Sprague-Dawley rat. However, ig BSO treatment increased B6 mouse p-nitroanisole demethylase and EY deM activities and D2 and B6 mouse p-nitrophenol hydroxylase activity to greater extents than observed in rats following any of the BSO exposure regimens (Tables 6, 7 and 8). Hence, it appears that the heretofore unrecognized selective Phase II inductive effects of BSO may be unique to rats. The effect of BSO on hamster hepatic drug metabolizing enzymes was not investigated. 57 Table 10. The effect of intragastric buthionine sulfoximine administration on mouse hepatic glutathione concentration and Phase I and Phase II enzyme parameters.a Mouse Strai n nb ParameterC GSH Phase I P450 pNA deM EROD PROD EYdeM pNPH Phase II UDPGt-pNP GSHt PSt -N -M -T 60±7 * 99±6 113 ± 4 (3) 0.02:t 0.00 0.00 0.00 100 ± 3 148 ± 12 (3) * 113± 12 171 ± 19 * 120 ± 11 115 ± 5 78±3 137 ± 11 * 116±19 .62 3 64±9 * 113 ± 2 177 ± 12 * 0.03:t 0.00 0.00:t 0.00 175 ± 5 * 207 (2) 134 ± 7 * 189 ± 32 134 ± 10 144 ± 9 92±8 156 ± 1 * 92±8 a SSO was administered as single daily doses of 6 mmoles/kg for 3 days. All animals were sacrificed 1 day after the third dose. b Nurnber of determinations unless indicated in parentheses. c Data are expressed as mean percent of control (Table 1) ± SE except for EROD and PROD, which are given as activities (nmoles/min/mg protein). * significantly different at p s; 0.03 compared to untreated controls (Table 1). CHAPTER 6 EFFECT OF SSO TREATMENT ON THE DISPOSITION OF ACETAMINOPHEN IN TWO STRAINS OF RATS Sprague-Dawley Two groups of rats, one receiving 30mM SSO-supplemented drinking water and another maintained on normal drinking water, were sacrificed 24 hr after being given acetaminophen at 150 mg/kg by intravenous injection. Exposure to SSO reduced hepatic glutathione concentration to 22% of that observed in rats treated only with acetaminophen (Table 11). The values for the Phase I and Phase II drug metabolism parameters in the animals treated with acetaminophen alone were lower than those of naive Sprague-Dawley rats (Table 1), possibly due to the surgical manipulations that preceded acetaminophen administration (see Methods and Materials). Nonetheless, the relative effects of prolonged SSO exposure on hepatic drug metabolizing enzymes were similar to those previously shown in rats, Le., no effect on P450 concentration or p-nitroanisole demethylase activity and a 300% induction of UDP-glucuronosyltransferase activity toward p-nitrophenol. The glucuronidations of 1-naphthol and morphine were also increased in the SSO+acetaminophen group but the increase in 1-naphthol glucuronidation failed to achieve significance (p > 0.03). UDP-glucuronosyltransferase activities towards estrone and testosterone were unchanged by the SSO treatment. Combined SSO and acetaminophen treatment resulted in a 52% increase in cytosolic 59 Table 11. Hepatic glutathione concentration and Phase I and Phase II parameters in Sprague-Dawley rats treated with acetaminophen alone and in combination with buthionine sulfoximine.a Ac~taminQQh~n BSO+Ac~taminQQh~n Pa[am~l~[ GSHb 5.32±0.65 1.17±0.17 ** Phase IC P450 0.60±0.03 0.57±0.03 pNA deM 0.38±0.03 0.31±0.02 PhgS~ IIc UDPGt-pNP 1.81 ±0.06 7.38±0.11 * -N 23.50±3.28 37.85±2.83 -M 4.97±0.33 8.25±0.64* -E 0.07±0.01 0.08±0.01 -T 2.97±0.01 3.12±0.29 GSHt 1137±57 1733±72* PSt 1.48±0.03 1.32±0.06 a Both acetaminophen- (n = 3) and BSO+acetaminophen-treated rats (n = 4) were administered (iv) a 150 mg/kg dose of acetaminophen. The BSO+acetaminophen rats were given 30mM BSO-drinking water for 6 days prior to acetaminophen treatment and maintained on BSO until sacrifice 24 hr after dosing with acetaminophen. The mean daily BSO dose was 3.4 mmoles/kg. b /lmoles/g liver c nmoles/mg protein or nmoles/min/mg protein * significantly increased at p $; 0.03 compared to rats treated with acetaminophen alone ** significantly decreased at p $; 0.03 compared to rats treated with acetaminophen alone 60 glutathione S-transferase activity compared to the acetaminophen group, but cytosolic p-nitrophenol sulfotransferase activity was unaffected. Thus, the apparent Phase II-selective inductive effects of extended sse drinking water exposure in acetaminophen-treated Sprague-Dawley rats are similar to those seen in otherwise untreated rats. The time courses of acetaminophen, acetaminophen-glucuronide and acetaminophen-sulfate concentrations in plasma are shown in Figure 6. The plasma concentration of acetaminophen in both treatment groups declined exponentially with no effect of sse treatment on the total and renal clearance of unchanged acetaminophen (Table 12). However, sse treatment significantly increased the partial clearance of acetaminophen to the glucuronide and increased the fraction of the dose recovered in the urine as this conjugate. The partial clearance of acetaminophen to the sulfate was decreased in the SSe+acetaminophen group, as was the fraction of the dose recovered in the urine as the sulfate conjugate, although the former did not achieve statistical significance. Fischer 344 Fischer rats were administered sse (2 mmoles/kg every 12 hr) via an indwelling feeding gastrotomy for 6 days. Acetaminophen (150 mg/kg) was given 24 hr before sacrifice and the final sse dose was administered 12 hr before sacrifice. The animals receiving only acetaminophen were given an equal volume of water instead of the sse solution. In contrast to the continuous drinking water sse exposure of acetaminophen-treated Sprague-Dawley rats, an analogous sse daily dose (4 mmoles/kg/day) administered at 12-hr intervals by gastrotomy to acetaminophen-treated Fischer 344 rats did not significantly decrease liver glutathione concentration at the time of sacrifice (Table 13) and also failed to induce microsomal morphine glucuronosyltransferase and cytosolic glutathione S-transferase activities. However, as with continuous sse 61 Figure 6. Plasma concentrations of acetaminophen (A), acetaminophen-glucuronide (B) and acetaminophen-sulfate (C) in acetaminophen- and buthionine sulfoximine+acetaminophen-treated Sprague-Dawley rats. The BSO+acetaminophen rats (closed squares, n = 4) were given 30mM BSO-supplemented drinking water for 6 days prior to acetaminophen administration (150 mg/kg, iv). Rats receiving only acetaminophen (open squares, n = 3) were maintained on normal drinking water. Access to the BSO-supplemented drinking water was continued for the BSO+acetaminophen group until all animals were sacrificed 24 hr after acetaminophen treatment. E 0, ::1. C .Q ro ~ E Q) u c 0 u E- C> ::1. C .Q ro ~ E Q) cu 0 u E- C> ::1. C .Q ro ~ E Q) u c 0 u 1000 A 100 10 0 60 120 180 240 Time, minutes 30 B 20 10 0 0 60 120 180 240 Time, minutes 30 20 10 o~------~~------~------~~------- o 60 120 Time, minutes 180 240 62 63 Table 12. The effect of buthionine sulfoximine treatment on acetaminophen elimination kinetics and urinary excretion products in Sprague-Dawley rats. a Elimination Kineticsb Acetaminophen BSOtAcetamjnophen Total Clearance 7.39±0.44 7.17±0.49 Partial Clearance to Acetaminophen-Sulfate 4.46±0.36 3.39±0.41 Partial Clearance to Acetaminophen-Glucuronide 1.29±0.05 1.90±O.12 * Renal Clearance of Acetaminophen 1.15±0.01 1.03±0.06 Urine CompositionC Acetaminophen-Sulfate 60.3±1.4 46.B±2.4 ** Acetaminophen-Glucuronide 17.6±1.4 26.5±0.3 * Acetaminophen 15.6±O.8 14.4±0.7 Total Recove ry 93.5±1.2 87.7.±2.0 a Both acetaminophen- (n = 3) and BSOtacetaminophen-treated rats (n = 4) were administered (iv) a 150 mg/kg dose of acetaminophen. The BSOtacetaminophen rats were given 30mM BSO-drinking water for 6 days prior to acetaminophen treatment and maintained on BSO until sacrifice 24 hr after dosing with acetaminophen. The mean daily BSO dose was 3.4 mmoles/kg. b Clearances are expressed as ml/min/kg ± SE. c Data are expressed as mean percent of the administered acetaminophen dose recovered in urine over 24 hr ± SE. * significantly increased at p $ 0.03 compared to rats treated with acetaminophen alone ** significantly decreased at p $ 0.03 compared to rats treated with acetaminophen alone 64 Table 13. Hepatic glutathione concentration and Phase I and Phase II parameters in Fischer 344 rats treated with acetaminophen alone and in combination with buthionine sulfoximine.a AcetaminQQhen SSO+AcelaminQQhen Pa[aOJfll~r GSHb 5.47±0.12 3.99±0.78 Phase IC P450 0.57±0.02 0.47±0.O4 pNA deM 0.50±0.04 0.42±0.03 Phase IIc UDPGt-pNP 1.72±0.24 4.96±O.76 * -N 21.20±O.85 36.70±5.25* -M 7.18±0.26 8.43±1.02 -E 0.26±0.02 0.29±O.04 -T 2.39±0.36 2.72±O.31 GSHt 1365±178 1546±144 PSt 1.74±0.26 1.42±O.26 a SSO+acetaminophen-treated rats received a SSO dose of 2.0 mmoles/kg via an indwelling gastrotomy once every 12 hr for a total of 11 doses prior to iv injection of a 150 mg/kg dose of acetaminophen and once after acetaminophen administration. All rats (n = 4 for both groups) were sacrificed 24 hr after acetaminophen treatment. b Jlmoles/g liver c nmoles/mg protein or nmoles/min/mg protein * significantly increased at p ~ 0.03 compared to rats treated with acetaminophen alone 65 drinking water exposure in the Sprague-Dawley strain, p-nitrophenol and 1-naphthol glucuronidations were increased in the BSO+acetaminophen group and Phase I parameters and cytosolic p-nitrophenol sulfotransferase were not significantly changed compared to treatment with acetaminophen alone. Plasma concentration time courses of acetaminophen, acetaminophen-glucuronide and acetaminophen-sulfate for the acetaminophen- and BSO+acetaminophen-treated Fischer 344 rats are shown in Figure 7. The concentration of unchanged acetaminophen declined in a similar exponential manner in both groups. Similar to the SSO effects seen in Sprague-Dawley rats, BSO enhanced the partial clearance of acetaminophen to the glucuronide and decreased partial clearance to the sulfate conjugate, although the latter was not significant at p ~ 0.03 (Table 14). These changes in rates of metabolite formation are mirrored by the significant increase and decrease, respectively, in the fraction of the acetaminophen dose recovered in the urine as acetaminophen-glucuronide and acetaminophen-sulfate. 66 Figure 7. Plasma concentrations of acetaminophen (A), acetaminophen-glucuronide (B) and acetaminophen-sulfate (C) in acetaminophen and buthionine sulfoxi mine+acetaminophen-treated Fischer 344 rats. The BSO+acetaminophen rats (closed squares, n = 4) were administered 2 mmoles BSO/kg every 12 hr via an indwelling feeding gastrotomy for a total of 12 doses, 11 given prior to acetaminophen treatment and one after. The final BSO dose was administered 12 hr before sacrifice. Rats-receiving only acetaminophen (open squares, n = 4) were given an equa.l volume of water instead of the BSO solution. Both treatment groups were sacri'ficed 24 hr after intravenous injection of 150 mg acetaminophen Ikg. E ~ c o .~ .... E CD u c o u E ~ c- 0 .~ .... E Q.) u c 0 u E a, ~ c 0 .~ .... E CD <..> c 0 <..> 1000 A 100 10 0 60 120 180 240 Time, minutes 50 B 40 30 20 10 0 0 60 120 180 240 Time, minutes 20 C 15 10 5 04-----------------~------__ ------~ o 60 120 Time, minutes 180 240 67 68 Table 14. The effect of buthionine sulfoximine treatment on acetaminophen elimination kinetics and urinary excretion products in Fischer 344 rats.a Elimination Kineticsb Acetamjnophen BSOtAcetamjnophen Total Clearance 9.00±0.78 9.01±0.42 Partial Clearance to Acetaminophen-Sulfate 4.36±0.32 3.04±0.26 Partial Clearance to Acetaminophen-Glucuronide 1.61±0.55 3.08±.Q.S5 * Renal Clearance of Acetaminophen 1.48±0.21 1.53±0.38 Urine CompositionC Acetaminophen-Sulfate 48.8±2.8 29.7±t 4.8 ** Acetaminophen-Glucuronide 18.2±O.9 33.3±4.0 * Acetaminophen 12.7±3.6 20.7±4.3 Total Recovery 79.7±4.8 83.7±3.8 aBSOtacetaminophen-treated rats received a BSO dose of 2.0 mmoles/kg via an indwelling gastroton1Y once every 12 hr for a total of 11 doses prior to iv injection of a 150 mg/kg dose of acetaminophen and once after acetaminophen administration. All rats (n = 4 for both groups) were sacrificed 24 hr after acetaminophen treatment. b Clearances are expressed as ml/min/kg ± SE. c Data are expressed as mean percent of the administered acetaminophen dose recovered in urine over 24 hr ± SE. * significantly increased at p ~ 0.03 compared to rats treated with acetaminophen alone ** significantly decreased at p ~ 0.03 compared to rats treated with acetaminophen alone CHAPTER 7 DISCUSSION Species and Strain Differences in N-Substituted Imidazole-Mediated Induction Neither Clotz, NBI nor nafimidone were able to induce mouse and hamster hepatic drug metabolizing enzymes to the extent observed in rats. Although definitive explanations for such differences are not possible based solely on the present data, comparisons between the current data and the literature permit discussion of possible reasons for the species dissimilarities. In the rat, the major increases in EROD activity (Tables 3 and 4) and shifts to 448 nm spectral maxima for the ferrous P450-carbon monoxide complex following NBI and nafimidone administration are suggestive of increases in a P450 isozyme that is normally induced via an Ah receptor-mediated induction mechanism. The marked increase in EROD activity after 5 days of treatment of Wistar rats with NBI reported by Magdalou et al. (1988), coupled with the present observations in Sprague-Dawley rats, demonstrate that these two strains of rat respond similarly to NBI. The polycyclic aromatic hydrocarbon-like induction of EROD activity by nafimidone in Sprague-Dawley rats has also been shown previously (Rush et aI., 1987). The inability of a 200-fold molar excess of NBI to displace benzo[a]pyrene from the rat cytosolic Ah receptor in vitro (Magdalou et aI., 1988), however, casts doubt on the premise of Ah receptor involvement in the induction of EROD activity by NBI and the 70 structurally similar nafimidone. This doubt is enhanced by the current findings that treatment of the Ah-responsive B6 strai n of mouse and Ah-responsive hamster with NBI and the B6 mouse with nafimidone yielded only minor increases in EROO activity compared to rats and did not shift the reduced P450-carbon monoxide absorbance maximum. That the small increases in EROO activity observed in the B6 strain were also seen in the Ah-nonresponsive CF-1 and 02 mouse strains following NBI or nafimidone administration further undermines the presumption of Ah receptor involvement in the observed changes. Concurrent OEX treatment synergistically increases the 3-methylcholanthrene induction of EROO activity in rats because of interaction of the OEX-bound glucocorticoid receptor with an upstream glucocorticoid responsive element enhancing P450lA 1 gene transcription (Sherratt et aI., 1989). OEX has also been shown to potentiate the benz[a]anthracene induction of Phase II cytosolic glutathione S-transferase activity in cultured rat hepatocytes (Sherratt et aI., 1990). To determine whether glucocorticoid-dependent factors might be involved in the species differences for induction of EROO and glutathione S-transferase activities by NBI (Table 3), B6 mice were simultaneously treated with OEX and NBI (Table 5). The combined xenobiotic treatment failed to potentiate the minor increases observed following treatment with NBI alone. Hence, the lack of EROO and glutathione S-transferase activity induction by NBI in B6 mice compared to rats does not appear to be due to species differences in glucocorticoid availability. Whether or not species variation in the presence and/or function of glucocorticoid responsive element(s) might account for these differences is not known. The relatively minor increases in EROO activity caused by NBI and nafimidone in Ah-responsive rodent species other than rats has several possible explanations. The rat Ah receptor might possess a greater binding affinity for NBI and nafimidone than the Ah receptor in the B6 71 mouse and hamster. The inability of excess NBI to displace benzo[a]pyrene from the rat Ah receptor (Magdalou et aI., 1988) indicates that if the rat receptor possesses some affinity for NBI it is low compared to polycyclic aromatic hydrocarbons, but sufficient to allow induction of EROD activity by daily NBI doses as low as 25 mg/kg (Papac and Franklin, 1988). Alternatively, the NBI- and nafimidone-mediated inductions of EROD activity in the rat could occur through a mechanism different from that of recognized Ah receptor agonists and one that is not operative in mice and hamsters. Also possible is that EROD activity is not as sensitive an indicator of P448 induction by NBI and nafimidone in mice and hamsters as it is in rats, analogous to the relatively minor induction of EROD activity seen in Ah-responsive hamsters following BNF (Table 3) or 3-methylcholanthrene (Blaich et aI., 1988; Lubet et aI., 1990) treatment. Another possibility is that metabolites of NBI and nafimidone, and not the parent compounds, mediate the induction of EROD activity, either by Ah receptor ligation or by some alternate mechanism, and that these metabolites of the two N-substituted imidazoles are only produced in the rat. The investigation of this hypothesis would require identification and quantification of NBI and nafimidone metabolites present in the livers of the animals and a search for metabolites unique to the rat. While few studies have been conducted on species differences in the biotransformation of N-substituted imidazoles, select species-dependent metabolism of nafimidone has been characterized. Nafimidone undergoes rapid reductive metabolism in the dog, monkey, rat and man to an aliphatic alcohol metabolite (Graham et aI., 1983, 1987; Rush et aI., 1990), followed by some conjugation to the O-B-glucuronide. The dog is unique among these species in that there are no metabolites resulting from oxidation of the naphthalene moiety, but whether or not that correlates with a lack of induction by nafimidone in this species is unknown. As noted earlier, it has been hypothesized that the induction of P450 by N-substituted imidazoles 72 is a response to the sustained inhibition of P450 activity caused by binding of the imidazole ring N-3 atom with the heme ferric iron of P450. If the N-3 glucuronidation of nafimidone reported to occur in monkeys, dogs and man (Rush et aI., 1990) also occurred in the hamster and mouse, but not in the rat, it would be expected to diminish such an interaction and, hence, lessen the inductive response. The decrease in lipophilicity accompanying either glucuronidation or naphthalene ring oxidation could also render nafimidone less capable of Ah receptor binding. However, an increased capacity for xenobiotic glucuronidation in rats caused by BSO treatment did not diminish NBI-mediated induction (Table 6). This suggest that either glucuronidation is not important in determining the inductive potency of N-substituted imidazoles or that BSO does not induce the glucuronosyltransferase(s) that mediate the conjugation of N-substituted imidazoles. In addition to the above reactions, oxidation and subsequent opening of the imidazole ring has been reported for the N-substituted imidazoles ketoconazole, econazole and midaglizole in dogs and man (Midgley et aI., 1981; Heel, 1982; Nakaoka and Hakusui, 1987). Loss of the imidazole ring has also been reported for the antifungal N-substituted imidazole croconazole in rabbits (Nakano et aI., 1989; Nakano and Mizojiri, 1989), although via a mechanism other than imidazole ring oxidation. Whether or not these or other metabolic differences exist between rats, hamsters and mice for nafimidone, NBI and Clotz deserves investigation. Such putative differences could account for the species-related dissimilarity in the observed inductive responses. The species differences in polycyclic aromatic hydrocarbon-like induction of hepatic Phase I enzymes by imidazole-containing compounds is not restricted to N-substituted imidazoles. A substituted benzimidazole, omeprazole, is used therapeutically as an inhibitor of gastric acid secretion but is also an inhibitor and inducer of P450-mediated monooxygenase reactions in certain animal species, including humans (Gugler and Jensen, 73 1985; Jensen and Gugler, 1986; Chenery et aI., 1988; Diaz et aI., 1990). Treatment of Wistar rats with omeprazole for 14 days increased hepatic P450 concentration and ethoxycoumarin O-deethylation, an activity that shows some selectivity for P450lA 1 in rats, by 1.5- and 2.5-fold, respectively, but did not increase the Pb- and DEX-inducible N·demethylation of ethylmorphine (Chenery et aI., 1988). Omeprazole also exhibits polycyclic aromatic hydrocarbon-type induction in human liver both in vitro and in vivo, enhancing EROD and AHH activities and P450lA 1 and P450lA2 gene transcription (Diaz et aI., 1990). However, in both Ah·responsive 86 mice and Ah·nonresponsive D2 mice, omeprazole failed to induce either P450lA 1 or P4501A2. Attempts to induce these proteins with omeprazole in rabbits were also unsuccessful (Diaz et aI., 1990). Thus, in common with the N-substituted imidazoles NBI and nafimidone, the polycyclic aromatic hydrocarbon-like induction caused by omeprazole appears to be species-specific, with the rat being the most responsive rodent species among those investigated. Species and Strain pifferences in Traditional Inducer·Mediated Induction The polycyclic aromatic hydrocarbon-inducible P450 isozymes exhibit a high degree of sequence homology between the rat, rabbit and mouse, as do the Pb-inducible P450 isozymes (Heinemann and Ozols, 1983; Cheng et aI., 1984; Thomas et aI., 1984; Adesnik and Atchison, 1986). Proteins similar to glucocorticoid-inducible rat P4501liA are inducible by DEX in the mouse, hamster, gerbil and rabbit liver based on such criteria as triacetyloleandomycin-P450 metabolic-intermediate complex formation, EY deM activity, immunoblot analysis, apparent molecular weight, inducibility by other agents known to induce P4501liA in the rat (e.g., pregnenolone-16·alpha-carbonitrile) and mRNAs of sufficient homology to hybridize to cloned cDNAs of P4501liA mRNA (Wrighton et aI., 1985b). 74 Thus, the P450 subfamilies inducible by polycyclic aromatic hydrocarbons, Pb and glucocorticoids appear to be conserved in many rodent species. PROD activity is considered a selective marker for Pb-mediated induction of P450 in rats and mice (Burke et aI., 1985; lum et aI., 1986; Kellyet aI., 1990; lubet et aI., 1990; de Waziers et aI., 1990). In agreement with these reports, the present data show that Pb treatment induced PROD activity in the Sprague-Dawley rat and CF-1 mouse (Table 5). However, PROD activity was also markedly enhanced by DEX in two strains of mice (Table 5; Meehan et aI., 1988), an effect not seen in two strains of rats (Table 5; Meehan et aI., 1988) or in hamsters (Table 5). Determination of specific mRNA concentrations indicated that the species difference between rats and mice was at the level of transcription (Meehan et aI., 1988). In contrast to the above information, Yamazoe et al. (1987) showed that a similar DEX treatment regimen significantly increased the liver concentrations of the Pb-inducible P450llB1 and P450llB2 isozymes and enhanced PROD activity by 2-fold in rats. The Pb-like induction of P450 by DEX was proposed to be due to a suppression in the release of pituitary hormone(s) reported to accompany DEX treatment of rats (Wakabayashi et aI., 1971; Kokka et aI., 1972; Oosterom et aI., 1983), since similar increases in P450llB1 and P450llB2 concentrations were seen in hypophysectomized animals. Administration of growth hormone, but not prolactin, to hypophysectomized rats decreased the concentrations of both P4501lB1 and P45011B2. Hypophysectomy also induced liver P450llB 1 +P4501lB2 mRNA in the mouse (Meehan et aI., 1988) and increased hepatic hexobarbital hydroxylase and aminopyrine N-demethylase activities, the latter effects being prevented by the administration of growth hormone (Macleod and Shapiro, 1989). Thus, it seems plausible that the induction of CF-1 mouse PROD activity by DEX documented in Table 5 could be the result of a DEX .. mediated decrease in growth hormone release from the pituitary. Other mechanisms suggested 75 to be involved in the DEX-mediated increases in P450llB1 and/or P450llB2 include stabilization of Pb-inducible P450 mRNA (Whitlock, 1986) and transcriptional activation caused by interaction of the DEX-glucocorticoid complex with a glucocorticoid response element proximal to the P450llB2 gene (Jaiswal et aI., 1990). The reason(s) for the apparent species differences in the Pb-like induction of P450 caused by DEX, as well as for the markedly different findings between laboratories on the Pb-like inductive effects of DEX in rats, remain uncertain. EY deM activity is used as a selective probe for steroid-mediated induction of P450 in rats (Wrighton et aI., 1985a, 1985b). In contrast to the marked induction of EY deM in rats and mice, DEX did not induce EY deM activity in hamsters (Table 5). A minor increase (25%) in EY deM activity has been shown in hamsters following DEX treatment but at a dose three times that used in the present study (Wrighton et aI., 1985b). While differences in dose and route of administration (gavage versus ip) may account for this discrepency, the reason(s) for the comparative lack of EY deM activity induction in hamsters is unknown. Unexpectedly, DEX treatment caused a 3..;fold increase in hamster p-nitrophenol hydroxylation (Table 5), a reaction catalyzed by the ethanol- (Reinke and Moyer, 1985) and isoniazid- (Ryan et aI., 1985) inducible P450llE1 isozyme in rats. DEX also increased rat p-nitrophenol hydroxylase activity 3-fold but in this species it was accompanied by a large (13-fold) increase in EY deM activity. Only minor (s; 400/0) increases in p-nitrophenol hydroxylase activity were detected in the three mouse strains following DEX administration and, as in the rat, these changes were accompanied by major inductions of EY deM activity. In a recent investigation, hypophysectomy of Sprague-Dawley rats caused significant increases in both liver P450llE1 protein content (2-fold, as determined by immunoblot analysis) and mRNA levels (Hong et aI., 1990). In BALB/c mice, hypophysecton1Y had no effect on P450llE1 content when analyzed with antibodies against rat hepatic 76 P4501lE1. Given the aforementioned DEX-mediated decrease in pituitary hormone release (Wakabayashi et aI., 1971; Kokka et aI., 1972; Oosterom et aI., 1983), the increase in p-nitrophenol hydroxylase activity observed in hamsters and rats, but not in mice, following DEX administration (Table 5) may be due to a species-selective effect of diminished circulating pituitary hormone(s) on liver P450llE1 synthesis and/or degradation. Moreover, when considered together with the current data regarding the effect of DEX on PROD activity, there also appears to be an isozyme-selectivity within a given species for the anomalous P450 inductive properties of DEX. While the synthetic glucocorticoid induced p-nitrophenol hydroxylase activity in hamsters and rats but not in mice, it increased PROD activity in mice but not in hamsters and rats. Thus, DEX, either directly or via its suppression of pituitary hormone release, appears to have dissimilar effects on the expression of different P450 genes. Support for a differential regulatory effect of growth hormone on P450 gene expression is found in a recent study showing that incubation of rat hepatocyte cultures with physiologic concentrations of growth hormone completely blocks the Pb-mediated induction of P450llB1 and P450llB2 mRNAs, suggesting that this inductive process is under direct inhibitory control by growth hormone (Schuetz et aI., 1990). Conversely, growth hormone had little effect on P450lliA mRNA induction by pregnenolone-16-alpha-carbonitrile or DEX. DEX treatment also displayed a species- and strain-selective effect on Phase " microsomal UDP-glucuronosyltransferase activity toward 1-naphthol (Table 5). DEX increased this polycyclic aromatic hydrocarbon-inducible Phase " activity 3-fold in the Ah-responsive B6 mouse but caused no change in activity in the rat, which is also Ah-responsive. DEX also failed to elicit an inductive response in the Ah-nonresponsive D2 mouse strain. The DEX-mediated increase in UDP-glucuronosyltransferase activity toward 1-naphthol was of the same magnitude in the B6 mouse as that caused by the Ah receptor agonist BNF 77 (Table 3). DEX administration also caused a 3-fold increase in Pb-inducible morphine glucuronidation in rats but this increase was only approximately 50% of that elicited by Pb (Table 5). Thus, rats appear to be less responsive than S6 mice to the induction of 1-naphthol glucuronosyltransferase activity by DEX. This is in contrast to the generally greater sensitivity of rats compared to mice for the induction of hepatic drug metabolizing enzymes by many of the xenobiotics investigated herein. Species Differences in SSQ-Mediated Induction Extended exposures to the gamma-glutamylcysteine synthetase inhibitor BSO (Griffith and Meister, 1979; Griffith, 1982) induced both microsomal UDP-glucuronosyltransferase and cytosolic glutathione S-transferase activities without causing major changes in P450 concentration or monooxygenase activity in rats (Tables 6-9,