effects of induction and inhibition of cytochromes p450 on the

TOHCOLOGICAL SCIENCES 46, 1 8 5 - 1 % (1998)

ARTICLE NO. TX982513

Effects of Induction and Inhibition of Cytochromes P450on the Hepatotoxicity of Methapyrilene

G. S. Ratra, S. Cottrell, and C. J. Powell1

Department of Toxicology, St Bartholomew's and the Royal London School of Medicine and Dentistry,Charterhouse Square, London, EC1M 6BQ, United Kingdom

Received December 16, 1997; accepted May 29, 1998

Effects of Induction and Inhibition of Cytochromes P450 on theHepatotoxicity of Methapyrilene. Ratra, G. S., Cottrell, S., andPowell, C. J. (1998). ToxicoL Sci. 46, 185-196.

The mechanisms by which the antihistamine drug metha-pyrilene causes acute periportal hepatotoxicity in rats are not yetelucidated. This study investigated the effects of modulators ofcytochrome P450 (CYP) activity on the hepatotoxicity of metha-pyrilene and also the effect of methapyrilene on hepatic CYP.Pretreatment of male Han Wistar rats with p-naphthoflavone,phenobarbitone, butylated hydroxytoluene, piperonyl butoxide,Aroclor 1254, or cobalt protoporphyrin IX, agents known to mod-ify hepatic CYP, all afforded some degree of protection against ahepatotoxic dose of methapyrilene (150 mg/kg x 3 days p.o.), asassessed by clinical chemistry and histology. Total hepatic CYPdepletion by cobalt protoporphyrin IX treatment indicated CYP-mediated bioactivation was a prerequisite for methapyrilene-in-duced hepatotoxicity. Protection against hepatic damage wasstrongly associated with /3-naphthoflavone induction of CYP1Aand phenobarbitone-associated CYP2B induction. However, therole of CYP3A, which is constitutively expressed in the liver andinduced by piperonyl butoxide, butylated hydroxytoluene, or Aro-clor 1254, was unclear. Modulation of FAD monooxgenase activ-ity by methimazole pretreatment was not associated with increasedmethapyrilene-induced hepatotoxicity. Methapyrilene treatmentalone specifically decreased microsomal enzyme activity markersfor CYP2C11, CYP3A, and CYP2A and pretreatment with all thehepatic enzyme-inducing agents specifically prevented the loss ofCYP2C11. Together this suggested that CYP2C11 was responsiblefor the suicide substrate bioactivation of methapyrilene and thetoxicologic outcome largely relied upon an abundance of detoxi-fying enzymes present in the liver, o

Methapyrilene is an ethylenediamine H, histamine receptorantagonist which has been used for over 20 years, principallyas a sleep aid and in cold and allergy formulations. Subse-quently, methapyrilene was withdrawn from clinical use whenit was found to be hepatocarcinogenic, following chronic ad-ministration in rats (Lijinsky et ai, 1980), although not in

' T o whom correspondence should be addressed. Fax: 0171 982 6135.E-mail: [email protected].

mice, guinea pigs, or hamsters (Brennan and Creasia, 1982;Lijinsky et ai, 1983). In the rat, acute administration of metha-pyrilene has been found to damage the periportal region of theliver (Graichen et ai, 1985); however, the underlying bio-chemical mechanisms responsible are poorly understood.

Methapyrilene undergoes extensive metabolism in vivo, incultured hepatocytes and in microsomes (Kammerer andSchmitz, 1986, 1987; Singer et ai, 1987; Kelly et al., 1992).While many metabolites of methapyrilene have been identified,the metabolite(s) responsible for hepatic damage in rats re-mains elusive. Species differences in the metabolism of metha-pyrilene seem largely quantitative rather than qualitative (Ka-mmerer and Schmitz, 1987; Lampe and Kammerer, 1990). It isplausible, therefore, that the species differences in metha-pyrilene-induced hepatotoxicity may be explained by differ-ences in the metabolic activation or detoxication pathways.

Many drugs undergo oxidative metabolism, to either more orless toxic moieties, via the cytochrome P450 (CYP)2 family ofisoenzymes. The profile of isozymes present in the liver can beselectively altered by chemical pretreatments (Schenkman etal, 1991). The aim of this study was to chemically modulatethe profile of CYP in the liver, in order to elucidate the possiblerole of this enzyme system on the hepatotoxicity of metha-pyrilene and, conversely, to see the effects of methapyrileneon CYP.

MATERIALS AND METHODS

Male Han Wistar rats (viral antibody free, 200-250 g, from B&K UniversalLtd.. Grimston, North Humberside, UK) were housed in solid-bottomed cageson sawdust bedding with controlled temperature, lighting, and humidity (20 ±2°C, 12:12 h light and dark, 50 ± 10%, respectively) and allowed access tostandard R/M 1 (C 3/8) pelleted diet (SDS Ltd., Witham, Essex, UK) and tap

2 Abbreviations used: ALP, alkaline phosphatase; ALT, alanine aminotrans-ferase; Aro, Aroclor 1254; BHT, butylated hydroxytoluene; 0NF, 0-naph-thoflavone; CoPr, cobalt protoporphyrin IX; CYP, cytochromes P450 (thenomenclature used for CYP is that recommended by Nelson et al., 1996);EROD, ethoxyresorufin-O-dealkylase; GGT, -^glutamyltransferase; H&E he-matoxylin & eosin; MP, methapyrilene; Mthz, methimazole; PB, sodiumphenobarbitone; Pip.b, piperonyl butoxide; PROD, pentoxyresorufin-0-deal-kylase

5 1096-6080/98 $23.00Copyright © 1998 by the Society of Toxicology.All rights of reproduction in any form reserved.

Downloaded from https://academic.oup.com/toxsci/article-abstract/46/1/185/1684118by gueston 11 April 2018

186 RATRA, COTTRELL, AND POWELL

water ad libitum. Animals were randomly allocated to treatment groups of fiveor six and allowed to acclimatize for at least 5 days.

Methapyrilene-HCl (+99%), B-naphthoflavone Q3NF, 95%), butylated hy-droxytoluene (BHT, +99%), methimazole (Mthz, —98%), cobalt protopor-phyrin IX (CoPr), and piperonyl butoxide (Pip.b, 90%) were from Sigma-Aldrich Chemical Co. Ltd. (Poole, Dorset, UK), phenobarbitone (PB, +98%)was from BDH (Poole), food-grade corn oil was from Safeways Supermarket,and Aroclor 1254 (Aro) was a gift from Dr. Nigel Lawrence (Robens Instituteof Health and Safety, University of Surrey, Guildford, UK). [4-14C]Androst-4-ene-3,17-dione (sp act 57 mCi/mmol) was from Amersham International(Amersham, UK). All other chemicals were of the highest available purity.

Groups of animals received PB (80 mg/kg body wt, dissolved in sterilewater), /3NF (80 mg/kg body wt, dissolved in com oil), Aro (125 mg/kg bodywt, dissolved in com oil), or respective vehicle alone, by ip injection on 3consecutive days, or BHT in the diet (1000 ppm) for 7 days, followed byadministration of either methapyrilene hydrochloride (150 mg/kg body wt,dissolved in sterile water) or vehicle alone, by oral gavage, for a further 3consecutive days. Further groups of animals received a single dose, sc, of CoPr(60 mg/kg body wt, dissolved in saline) 72 h prior to the 3 days of metha-pyrilene treatment, or Mthz (100 mg/kg body wt, dissolved in sterile water), orPip.b (400 mg/kg body wt, dissolved in corn oil), by ip injection, 30 min pnorto each methapyrilene dose. All dosing took place between 10 and 11 AM.Animals were killed 24 h after receiving the final dose by carbon dioxideasphyxiation. Blood was withdrawn from the inferior vena cava, placed inheparinized tubes, and mixed immediately by repeated inversion and rollermixing. The tubes were centrifuged at 3000g for 15 min at 4°C and separatedplasma was stored at —70°C.

At autopsy, a full macroscopic examination was performed and samples ofeach major lobe of the liver were fixed in 10% neutral buffered Formalin.Tissues were processed through paraffin wax for histological examination andstained with haematoxylin and eosin (H&E).

Plasma aspartate aminotransferase (ALT), alkaline phosphatase (ALP), andy-glutamyltransferase (GGT) were measured as indices of hepatic damage, inonce-thawed plasma samples on a COBAS BIO centrifugal analyzer, using com-mercial kits (Roche Diagnostics Ltd., Welwyn Garden City, Hertfordshire, UK).

Liver microsomes were prepared from each liver according to the method ofMann et al. (1985). Total microsomal cytochrome P450 (CYP) was determinedby the spectrophotometric method of Omura and Sato (1964), androst-4-ene-3,17-dione 16a, 68, and la hydroxylation activities were assayed radiomet-rically according to Jiang et al. (1994), and pentoxyresorufin-O-dealkylase(PROD) and ethoxyresorufin-O-demethylase (EROD) activities were assayedby the fluorometnc method of Burke and Mayer (1983). Protein was deter-mined by the method of Lowry et al. (1951) using bovine serum albumin(fraction V) as standard.

Statistical analysis of data were by ANOVA and Student's t test, and in thecase of plasma enzyme levels, data were log transformed prior to analysis.

RESULTS

Methapyrilene treatment resulted in a 10-fold elevation ofplasma ALT, which was significantly diminished when ani-mals were pretreated with any of the liver enzyme modulatingagents (Fig. 1A). This demonstrated that prior treatment withany of these compounds afforded protection against metha-pyrilene-induced hepatotoxicity. However, with the exceptionof the group receiving cobalt protoporphyrin LX, plasma ALTlevels from all methapyrilene rats remained significantlyhigher, relative to animals receiving respective pretreatmentsalone, indicating that the protection was not always absolute.Methapyrilene treatment also increased plasma GGT 6-fold(Fig. IB.), indicating biliary damage or dysfunction. Only /3NF

or piperonyl butoxide pretreatments resulted in significantlylowering methapyrilene-induced GGT, which in the latter in-stance actually fell below the limits of detection. Plasma alka-line phosphatase, another, although less specific, marker ofbiliary damage, was also increased after methapyrilene treat-ment (2-fold, results not shown). Elevations were generallyreduced following any of the pretreatments, except methima-zole. Taken together, these data suggest that the protectionafforded by these pretreatment regimes was primarily directedtoward the parenchymal cells, although /3NF and piperonylbutoxide were also effective at protecting against biliary dam-age. Although the pretreatments alone were not generally dam-aging to the liver, administration of Aroclor 1254 caused asmall but significant increase in ALP.

Histological assessment of representative sections of livershowed that methapyrilene treatment resulted in periportal hepaticdamage, characterized by an increased inflammatory cell infil-trate, hydropic degeneration and necrosis of hepatocytes, and anincrease in mitotic figures (Fig. 2b). All pretreatments resulted ina reduction in severity of methapyrilene-induced hepatocellulardamage, the magnitude of which varied between treatment groupsand correlated well with the clinical chemistry data: /3NF or BHTpretreatment resulted in the greatest protection against degenera-tive hepatic damage and this correlated with the largest reductionin methapyrilene-induced increase in plasma ALT (Figs. 2c and2d). In the livers of piperonyl butoxide- or phenobarbitone-pre-treated animals, only the hepatocytes immediately adjacent to thetracts had undergone hydropic degeneration. The numbers ofnecrotic hepatocytes were markedly reduced, and only a relativelymild inflammatory portal infiltration was observed (Figs. 2e and2f). Methimazole or Aroclor 1254 pretreatment resulted in theleast protection, with marked inflammatory cell infiltration aroundthe portal tracts, moderate hydropic degeneration of hepatocytestwo to three cells out from the tracts, along with necrosis ofindividual cells. The hepatocytes in this region demonstrated amore eosinophilic cytoplasm, suggestive of some early degener-ative changes (Figs. 2g and 2h). Methapyrilene-induced damageto bile duct epithelial cells in /3NF or piperonyl butoxide-pre-treated animals was negligible and correlated well with the ab-sence of any increase in plasma GGT. No hepatocellular damagewas observed in livers of animals treated with methapyrilene aftercobalt protoporphyrin LX administration (Fig. 2i). Modest dilata-tion of sinusoids was observed, but this also was a feature of thepretreatment alone (not shown). /3NF treatment alone showed amild serosal irritant reaction (not shown), which has been reportedpreviously (Cottrell et aL, 19%).

With the exceptions of phenobarbitone and cobalt protopor-phyrin IX, each of the pretreatment regimes or methapyrilenealone significantly increased relative liver weight (Fig. 3).However, except after the administration of methimazole orBHT, further administration of methapyrilene did not promptfurther increase, suggesting that most pretreatments blockedany hepatic mitogenic or hypertrophic response to metha-pyrilene.


CYTOCHROME P450 AND METHAPYRILENE 187

Pretreatment

• Water • Pip.b H RNF 13 Aro

Pretreatment + MP

PB BHT S Mthz I CoPr

FIG. 1. (A) Alanine aminotransferase (ALT) and (B) f-glutamyltransferase (GGT) levels in plasma samples from methapyrilene (MP)-treated male HanWistar rats pretreated with piperonyl butoxide (Pip.b), /3-naphthoflavone (0NF), Aroclor 1254 (Aro), phenobarbitone (PB), butylated hydroxytoluene (BHT),methimazole (Mthz), or cobalt protoporphyrin IX (CoPr). Control animals received either sterile water or pretreatment alone, with sterile water also administeredin the place of methapyrilene. Enzyme levels are given as mean values ± SEM, n = 5 or 6, except for groups pretreated with Pip.B, where n = 4, or PB, wheren = 3, where one and two premature deaths occurred following a single dose of MP and a single dose of PB, respectively. Significant differences are indicated$ and " , from matched pretreatment, and + , + + , and + + + , from methapyrilene treatment alone (p < 0.05, p < 0.01, and p < 0.001, respectively). Note, MPalone, or following Pip.b, Aroclor 1254, PB, BHT, methimazole, or CoPr IX pretreatment resulted in significant elevations (/> < 0.05, p < 0.01, orp < 0.001)in enzyme levels when compared with water only.

Methapyrilene treatment alone resulted in a 25% reductionof total hepatic CYP (Fig. 4). This was in contrast to theincreases in CYP following treatments with piperonyl butox-ide, /3NF, Aroclor 1254, or BHT alone. Phenobarbitone, con-trary to reports elsewhere, failed to elevate total CYP (Sipesand Gandolfi, 1986; Waxman and Azaroff, 1992), and methim-azole, a flavin monooxygenase inhibitor, was also withouteffect. Further administration of methapyrilene tended to de-crease any pretreatment-induced elevations in CYP by ~25%and, with the exception of phenobarbitone or Aroclor 1254,these decreases were significant. However, with the exceptionof BHT, levels of CYP were consistently greater than withmethapyrilene alone. Cobalt protoporphyrin DC administrationresulted in a marked depletion of CYP to 10% of control, asreported previously (Drummond and Kappas, 1982; Spaethe

and Jollow, 1989). Further administration of methapyrilene didnot result in a significant additional loss.

No significant changes in PROD or EROD activities, markersfor CYP2B and CYP1A activities, respectively, were observed inresponse to methapyrilene alone (Fig. 5). Piperonyl butoxide,Aroclor 1254, and phenobarbitone pretreatments resulted in ~30-fold increases in PROD activity (p < 0.001), while a relativelymodest 3-fold increase was seen in response to BHT treatment(p < 0.001). Interestingly, the phenobarbitone-induced elevationin PROD activity occurred in the absence of any increase in CYPor liver weight, both features usually associated with this treat-ment (Sipes and Gandolfi, 1986; Waxman and Azaroff, 1992).Significant 2-, 25-, and 30-fold increases in EROD activities wereobserved after methimazole, £NF, and Aroclor 1254 treatments,respectively (p < 0.001).



FIG. 2. Livers from (a) control and (b) methapyrilene(MP)-treated male Han Wistar rats and (c-i) methapyrilene-treated rats pretreated with (c)0-naphthoflavone, (d) butylated hydroxytoluene, (e) piperonyl butoxide, (0 phenobarbitone, (g) methimazole, (h) Aroclor 1254, and (i) cobalt protoporphynnIX. Experimental details were as described in Fig. 1. Arrows indicate portal damage. V, central vein; P, portal tract Original magnification X32, H&E staining.

Increases in PROD activity (following piperonyl butoxide,phenobarbitone, or BHT), EROD activity (/3NF or methima-zole), or both (Aroclor 1254), or inhibition of both activities, asseen after cobalt protoporphyrin DC pretreatment, were there-fore coincident with protection against methapyrilene hepato-toxicity, the latter likely a reflection of this compound's abilityto deplete CYP. Further methapyrilene-related increases inEROD or PROD activities following either Aroclor 1254 orBHT pretreatments, respectively, and reduction in EROD ac-tivity by methapyrilene following methimazole pretreatmentdemonstrated the ability of this hepatotoxin to selectively mod-ulate inducible, but not constitutively expressed, 0-dealkylaseactivities.

In contrast to the effects on PROD and EROD activities,methapyrilene administration alone caused a significant reduc-tion in androstenedione hydroxylation, such that 16a, 60, andla hydroxylations were decreased to 55, 40, and 65% ofcontrol, respectively (Fig. 6). Of the enzyme-inducing pretreat-ments, only /3NF significantly modulated androstenedione 16ahydroxylation activity, representative of CYP2C11 activity,decreasing it by 50%. However, subsequent administration ofmethapyrilene did not cause any further decrease in andro-stenedione 16a hydroxylation and actually resulted in a small,

but significant, increase in activity. The slight, but statisticallysignificant, increase in androstenedione 16a hydroxylation thatoccurred when methapyrilene was administered after Aroclor1254 may be largely attributed to one individual, suggestingthat this increase may not be biologically significant. All theCYP induction regimes associated with protection againstmethapyrilene hepatotoxicity therefore prevented the metha-pyrilene-invoked loss in androstenedione 16a hydroxylationactivity. However, the flavin monooxygenase inhibitor me-thimazole, which is only a poor CYP inducer, failed to protectagainst androstenedione 16a hydroxylation activity loss.

Androstenedione 6/3 hydroxylation, representative of CYP3Aactivity, was significantly induced by 50, 100, or 150% by pip-eronyl butoxide, BHT, or Aroclor 1254 pretreatments, respec-tively. However, following methapyrilene administration, Aroclor1254- and BHT-induced 6/3 hydroxylation activities were signif-icantly decreased by 25%, compared with the respective pretreat-ment alone. Piperonyl butoxide-induced 60 hydroxylation wasalso decreased 25% by subsequent methapyrilene treatment, butin this case not significantly so. This suggests that methapyrilenehad selectively suppressed chemically induced CYP3A activity,although not to as great an extent as when administered alone.Androstenedione la hydroxylation, representative of CYP2A,



P

P

FIG. 2—Continued



V

FIG. 2—Continued


CYTOCHROME P45O AND METHAPYRILENE 191

I

P

FIG. 2—Continued



V

m

t • . • * , - •

FIG. 2—Continued



10

a

i?Im

0 >-Pretreatment Pretreatment + MP

D Non* H Plpjj E3 BNF B Are B PB Y. BHT K Uthz CoPr

FIG. 3. Livenbody weight ratios and from methapyrilene(MP)-treatedmale Han Wistar rats pretreated with Pip.b, /3NF, Aro, PB, BHT, Mthz, orCoPr. Experimental details were as described in Fig. I. Results are given asmean values ± SEM (n = 3-6). Significant differences are indicated. $ and w ,from matched pretreatment alone, and + + , from methapynlene alone (p <0.05 and p < 0.01, respectively). Note, all pretreatments alone except PB orCoPr IX significantly increased livenbody weight ratio; Mp alone or followingall pretreatments except CoPr IX significantly increased livenbody weightratio {p < 0.05, p < 0.01, orp < 0.001).

activity was mildly elevated by 30-50% in response to piperonylbutoxide, /3NF, and Aroclor 1254 pretreatments only. Furtheradministration of methapyrilene resulted in no additional changein the activity of this isoform. As expected, cobalt protoporphyrinEX pretreatment resulted in very low androstenedione 16a, 6)3, or

la hydroxylation, and subsequent methapyrilene administrationhad little or no effect.

DISCUSSION

The aim of this study was to establish the role of biotrans-formation in the activation and detoxication of this periportaltoxin. Clinical chemistry and histopathological assessments ofmethapyrilene-induced hepatotoxicity correlated well anddemonstrated that all the pretreatments afforded some degreeof protection. All the liver enzyme-inducing agents werelargely more proficient in preventing the elevation of plasmaalanine aminotransferase levels than they were -y-glutamyltransferase, suggesting that their protective role was moredirected toward the parenchyma rather than the bile duct epi-thelial cells, both of which are targeted by methapyrilene.Exceptions to this were /3NF and piperonyl butoxide whichwere effective in protecting both cellular structures.

Cobalt protoporphyrin LX is reported to specifically cause amarked and lasting depletion in CYP (Drummond and Kappas,1982; Spaethe and Jollow, 1989). Therefore, the check onplasma ALT elevation and preservation of liver architecture inmethapyrilene-treated animals pretreated with cobalt protopor-phyrin IX provided strong evidence that CYP-mediated bioac-tivation was a prerequisite for methapyrilene hepatotoxicity.However, the nondiscriminant nature of the CYP depletionprecluded the identification of any specific isoform(s) as beingnecessarily involved in methapyrilene bioactivation.

In an earlier study, Lampe and Kammerer (1987a) also

Water Pip.B BNF Aroclor BHT Methimazole CoPr

FIG. 4. Total hepatic microsomal cytochrome P450 from male Han Wistar rats pretreated with Pip.b, 0NF, Aroclor 1254, PB, BHT, methimazole, or CoPrwith (hatched bars) or without (closed bars) methapyrilene. Experimental details were as described in Fig. 1. Results are given as mean values ± SEM (n = 3-6).Significant differences are indicated. $, " , and ***, from matched pretreatment, and + + + , from methapyrilene treatment alone (p < 0.05, p < 0.01, and p <0.001). Note that MP alone or CoPr IX pretreatment significantly reduced (p < 0.001) total CYP content of the liver, whereas Pip.b, 0NF, Aroclor 1254, or BHTsignificantly increased (p < 0.05 orp < 0.01) total CYP content.



1400

o> 1200

Water PipB BNF Aro

Water PipB BNF Aro BHT Mthz CoPr

FIG. 5. (A) Microsomal pentoxy- and (B) ethoxyresorufin-O-dealkylaseactivities from Pip.B, j3NF, Aro, PB, BHT, Mthz, or CoPr pretreated male HanWistar rats with (hatched bars) or without (closed bars) methapyrilene treat-ment. Experimental details were as described in Fig. I. Results are given asmean values ± SEM (n = 3-6). Significant differences are indicated s and $$,from matched pretreatment, and + from methapyrilene treatment alone (p <0.05, p < 0.01). Note that all pretreatments alone except Pip.b or phenobar-bitone resulted in significant differences in EROD values compared withcontrol (p < 0.001); all pretreatments alone except /3NF or methimazole aloneresulted in significant differences in PROD values compared with control (p <0.001).

claimed evidence for CYP involvement in the formation of"toxic" methapyrilene metabolites, by demonstrating theNADPH dependence of covalent binding to calf thymus DNAin vitro and its inhibition when CYP inhibitors were includedwith rat liver microsomes. However, while this may have beenconsistent with CYP-dependent metabolism it clearly did notdemonstrate that the metabolites generated were necessarilythe cause of hepatotoxicity.

In contrast to cobalt protoporphyrin IX, all the other protec-tive agents induced one or more isoforms of CYP, suggestingthat, while oxidative metabolism was required for hepatotox-icity, it was also important for the detoxication of metha-pyrilene. The extent of protection given by any one pretreat-ment was in the following order of decreasing efficacy, as

determined by clinical chemistry and histopathology: /3NF,BHT, PB, piperonyl butoxide, methimazole, and Aroclor 1254.Since these compounds modulated different isoforms of CYP,their mechanisms of protection against methapyrilene hepato-toxicity were unlikely to be the same in each case.

/3NF treatment increased CYPlA-associated activity, andsince this corresponded with the greatest degree of protection,it is plausible that biotransformation by this isoform is adetoxication route for methapyrilene. Phenobarbitone treat-ment is normally associated with both CYP2B and, to a lesserdegree, CYP3A induction (Waxman and Azaroff, 1992, andreferences therein). However, in this study, CYP2B- associated

l l$

Jsii Hi

T 1Water PipB BNF Aro PB BHT Mthz CoPr

B

Water PipB BNF Aro PB BHT Mthz CoPr

C - 0.6

I 0.4

I 0.35| 0.2JZ

i- 0.1

Wltsr DmC (1K|E »rn Dd RUT l | i h , CoPr

FIG. 6. (A) Microsomal androstenedione 16a- (B) 6/3-, and (C) la hy-droxylation activities from pretreated male Han Wistar rats Pip.B, /3NF, Aro,PB, BHT, Mthz, or CoPr with (hatched bars) or without (closed bars) metha-pyrilene treatment. Experimental details were as described in Fig. 1. Resultsare given as mean values ± SEM (n = 3-6). Significant differences areindicated. $ and s$, from matched pretreatment, + + and + + + , from metha-pyrilene treatment alone (p < 0.05, p < 0.01, and p < 0.001 respectively).Note that methapyrilene alone resulted in significant decrease in all hydroxy-lation activities, compared with control (p < 0.01).

1" HI 111


CYTOCHROME P450 AND METHAPYRTLENE 195

activity was significantly elevated in the absence of an in-creased CYP3A-associated activity. This suggested that, fol-lowing pretreatment with phenobarbitone, CYP2B was in-volved in the detoxication of methapyrilene and protectionagainst hepatic damage. This does not exclude the possibilitythat some other isoform(s) of CYP may participate in thedetoxication of methapyrilene.

In substantiation of this, Lampe and Kammerer (1987a)found that phenobarbitone pretreatment in vivo failed to in-crease formation of a covalent binding species, when metha-pyrilene was incubated with rat liver microsomes in the pres-ence of NADPH. However, it did increase the formation ofnonbinding metabolites; normethapyrilene, hydroxylpyridylmethapyrilene, and methapyrilene amide. It is possible thatphenobarbitone pretreatment protected against methapyrilenehepatotoxicity by increasing the formation of these metabolitesin vivo, thereby reducing the amount of substrate available forbioactivation.

Aroclor 1254 pretreatment afforded the least protectionagainst methapyrilene toxicity, despite inducing two CYP iso-forms already suggested to be involved in detoxication path-ways, CYP1A and CYP2B. Aroclor 1254 also elevatedCYP3A-associated activity, implicating this isoform in thebioactivation of methapyrilene. However, Aroclor 1254 is avery broad spectrum inducer and thus the bioactivation ofmethapyrilene by one or more induced CYP activities notdetermined here cannot be excluded. Furthermore, piperonylbutoxide and BHT pretreatments induced CYP3A, yet pro-tected to a greater extent than did Aroclor 1254.

Piperonyl butoxide and BHT induction of CYP3A andCYP2B activities, respectively, were in agreement with thefindings of Sipes and Gandolfi (1986) and Dalvi and Dalvi(1991). The protection afforded by these pretreatments may bepartly due to the induction of CYP2B, similar to that seenfollowing phenobarbitone pretreatment, but CYP3A cannot yetbe excluded from being involved in methapyrilene detoxica-tion.

BHT is also an effective inducer of epoxide hydrolase (Pow-ell et al, 1986), an enzyme involved in the detoxication ofhighly reactive, often short-lived, epoxides. Although metha-pyrilene epoxide products have yet to be isolated, a thiophenering substituent of the molecule, which is a candidate toxicmetabolite (Singer et al, 1987; Kammerer et al, 1988; Wrigh-ton et al, 1991), may, like other thiophene rings (Bray et al,1971; Dansette et al, 1990, 1992), undergo epoxidation. Con-sequently BHT induction of epoxide hydrolase could contrib-ute to the deactivation of such a species and provide a protec-tive role. Furthermore, methapyrilene itself is reported toelevate epoxide hydrolase following the same dose regimeused here (Graichen et al., 1985). This might suggest autoin-duction of a detoxication pathway in response to production ofreactive species. However, the antioxidant property of BHTcannot be excluded from contributing to its protective roleagainst methapyrilene hepatotoxicity.

Methapyrilene-related reduction in total CYP has been re-ported previously (Graichen et al, 1985; Wrighton et al,1991). Furthermore, Lampe and Kammerer (1987i>) demon-strated that microsomes prepared from the livers of rats pre-treated in vivo with methapyrilene had a decreased ability tometabolize methapyrilene in vitro. This suggested the forma-tion of metabolites which either inhibited or inactivated theenzyme(s) involved. Investigations here have demonstratedthat methapyrilene specifically reduced the catalytic activitiesof CYP isoforms 2C11, 3A, and 2A but not 1A and 2B. Lossof CYP2C11- and CYP3A-related activity was qualitatively,if not quantitatively, in agreement with the findings of Wrigh-ton et al. (1991). On the basis of substantial (80%) and specifictreatment-related decreases in CYP2C11 activity, theseworkerssuggested that this isoform was responsible for the formationof reactive intermediates which bound to this same isoform anddecreased its catalytic activity in a suicide substrate manner.Suicide substrate activation of methapyrilene would be consis-tent with results here, but suggests that CYP3A and CYP2Aboth may be inactivated in a similar fashion. The prevention ofany methapyrilene-related decrease in CYP2C11-associatedactivity on administration of any pretreatment was consistentwith a diversion of metabolism away from suicide substratebioactivation and further implicated CY2C11 as being respon-sible for the metabolism of methapyrilene to the toxic species.

In contrast to findings here, Wrighton et al (1991) alsodemonstrated a methapyrilene-related increase in CYPlA-re-lated activity, a discrepancy which may be attributable to thelonger duration of treatment in their study.

The overall toxicologic outcome of methapyrilene adminis-tration must be dependent on the relative balance betweenpathways of bioactivation and detoxication and the relativedifferences in affinities of the enzymes involved for the sub-strate. CYP2C11 is the predominant CYP isoform in the malerat, accounting for 50% of the total CYP (Morgan et al, 1985),while CYP3A accounts for 25%. It is possible that both ofthese isoforms are high-capacity-low-affinity enzymes, whichmetabolize methapyrilene to reactive intermediates. Subse-quently, when CYP1A or CYP2B isoform activities are ele-vated, these may provide alternative higher affinity pathwaysfor methapyrilene metabolism and detoxication which did notpreviously exist, due to their low constitutive expression in theliver. This "diverted" metabolism could provide an explanationfor the minimal loss in CYP2C11 observed when metha-pyrilene is administered following pretreatments.

In summary, this study provides persuasive evidence thatmethapyrilene hepatotoxicity is dependent on bioactivation byCYP and that pretreatment-related decreases in methapyrilenehepatotoxicity are related to the modulation of CYP isoforms.Furthermore, methapyrilene itself selectively decreases en-zyme activities associated with CYP2C11 and CYP3A, possi-bly by suicide substrate activation. Induction of the 1A and 2Bisoforms of CYP are consistent with protection against hepa-totoxicity and prevention of CYP2C11 destruction, implicating



this isoform in the bioactivation of methapyrilene. However,none of the pretreatment regimes employed are completelyspecific in their actions and contributions to their protectiverole from, e.g., elevation in GSH, induction of phase II en-zymes, and increased bile flow, cannot be excluded.

ACKNOWLEDGMENTS

The authors thank Mr. I. Garrod, Mrs. J. Wilson, and Miss C. Noel for theirexpert technical assistance.

REFERENCES

Bray, H. G., Carpanini, F. M. B., and Waters, B. D. (1971). The metabolism

of thiophene in the rabbit and rat Xenobiotica 1, 157-168.

Brennan, L. M., and Creasia, D. A. (1982). The effects of methapyrilenehydrochloride on hepatocarcinogenicity and pentobarbitaJ-induced sleeping

time in rats and mice. Toxicol. Appl. Pharmacol. 66, 252-258.

Burke, M. D., and Mayer, R. T. (1983). Differential effects of phenobarbitoneand 3-methylcholanthrene induction on the hepatic microsomal metabolismand cytochrome P-450-binding of phenoxanzone and a homologous series of

n-octyl ethers (alkoxyresorufins). Chem. Biol. Interact. 45, 243-258.

Cottrell, S., Oliver, K., Lake, B. G., and Powell, C. J. (1996). Strain-specificenhancement or inhibition of coumarin hepatotoxicity in mice followingpretreatment with two different liver enzyme-inducing agents. Fundam,

Appl. Toxicol. 34, 47-55.

Dalvi, R. R., and Dalvi, P. S. (1991). Differences in the effects of piperine andpiperonyl butoxide on hepatic drug-metabolising system in rats. Drug.

Chem. Toxicol. 14, 219-229.

Dansette, P. M., Amar, C , Smith, C , Pons, C., and Mansuy, D. (1990).Oxidative activation of the thiophene ring by hepatic enzymes. Biochem.

Pharmacol. 39, 911-918.

Dansette, P. M., Thang, D. C , El Amri, H., and Mansuy, D. (1992). Evidencefor thiophene-S-oxide as a primary reactive metabolite of thiophene in vivo:Formation of a dihydrothiophene sulfoxide mercapturic acid. Biochem.

Biophys. Res. Commun. 186, 1624-1630.

Drummond, G. S., and Kappas, A. (1982). The cytochrome P450 depletedanimal: An experimental model for in vivo studies in chemical biology.

Proc. Natl. Acad. Sci. USA 79, 2384-2388.

Graichen, M. E., Neptun, D. A., Dent, J. G., Popp, J. A., and Leonard, T. B.(1985). Effects of methapyrilene on rat hepauc xenobioUc enzymes and liver

morphology. Fundam. Appl. Toxicol. 5, 165-174.

Jiang, X-M., Cantrill, E., Farrell, G. C , and Murray, M. (1994). Pretransla-tional down-regulation of male specific hepatic P450s after portal bypass.Biochem. Pharmacol. 48, 701-708.

Kammerer, R. C., and Schmitz, D. A. (1987). Species differences in the in vitrometabolism of methapyrilene. Xenobiotica 17, 1121-1130.

Kammerer, R. C , and Schmitz, D. A. (1986). Metabolism of methapyrilene byrat-liver homogenate. Xenobiotica 16, 671-680.

Kammerer, R. C , Schmitz, D. A., Lampe, M. A., and Kloc, K. (1988). The invivo metabolism of methapyrilene, a hepatocarcinogen, in the rat. Xenobi-otica 18, 869-881.

Kelly, D. W., Holder, C. L., Korfmacher, W. A., Getek, T. A., Lay, J. O. Jr.,Casciano, D. A., Shaddock, J. G., Duhart, H. M., and Slikker Jr., W. (1992).Metabolism of methapyrilene by Fischer-344 rat and B6C3F, mouse hepa-tocytes. Xenobiotica 22, 1367-1381.

Lampe, M. A., and Kammerer, R. C. (1990). Species differences in themetabolism and macromolecular binding of methapyrilene: A comparison ofrat, mouse and hamster. Xenobiotica 20, 1269-1280.

Lampe, M. A., and Kammerer, R. C. (1987a). Cytochrome P-450 dependentbinding of methapyrilene to DNA in vitro. Carcinogenesis 8, 1525-1529.

Lampe, M. A., and Kammerer, R. C. (1987ft). The effect of chronic metha-pyrilene treatment on methapyrilene metabolism in vitro. Carcinogenesis 8,221-226.

Lijinsky, W., Knutsen, G., and Reuber, M. D. (1983). Failure of methapyrileneto induce tumours in hamsters or guinea pigs. J. Toxicol. Environ. Health 12,653-657.

Lijinsky, W., Reuber, M. D., and Blackwell, B.-N. (1980). Liver tumoursinduced in rats by oral administration of the antihistaminic methapyrilenehydrochlonde. Science 209, 817-819.

Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951). Proteinmeasurement with the folin phenol reagent. J. Biol. Chem. 193, 265—275.

Mann, A. H , Price, S. C, Mitchell, F. E., Grasso, P., Hinton, R. H., andBridges, J. W. (1985). Comparison of the short-term effects of di-(2-ethylhexyl) phthalate, di(n-hexyl) phthalate and di-(n-octyl) phthalate inrats. Toxicol. Appl. Pharmacol. 77, 116-132.

Morgan, E. T., MacGeoch, C , and Gustafsson, J.-A. (1985). Sexual differen-tiation of cytochrome P450 in rat liver. Evidence for a constitutive isozymeas the male-specific 16a-hydroxylase. Mol. Pharmacol 27, 471-479.

Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R.,Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W.,Gunsalus, I. C , and Nebert, D. W. (1996). P450 superfamily: Update onnew sequences, gene mapping, accession numbers and nomenclature. Phar-macogenetics 6, 1-42.

Omura, T., and Sato, R. (1964). The carbon-monoxide-binding pigment ofliver microsomes. J. Biol. Chem. 239, 2379-2384.

Powell, C. J., Connelly, J. C, Jones, S. M., Grasso, P., and Bridges, J. W.(1986). Hepatic response to the administration of high doses of BHT to therat: Their relevance to hepatocarcinogenicity. Food Chem. Toxicol. 24,1131-1143.

Schenkman, J. B., Thummel, K. L., and Favreau, L. V. (1991). Physiologicaland pathophysiological alterations in rat hepatic cytochrome P450. In Mo-lecular Aspects of Monooxygenases and Bioactivation of Toxic Compounds(E. Arinc, J. B. Schenkman, and E. Hodgson, Eds.), pp. 233-254. Plenum,New York.

Singer, S. S., Lijinsky, W., Kratz, L. E., Castagnoli, N. Jr., and Rose, J. E.(1987). A comparison of in vivo and in vitro metabolites of the Hl-antagonist N,W-dimethyl-A''-2-pyridyl-A''-(2-thienylmethyl)-1,2-ethanedia-mine (methapyrilene) in the rat. Xenobiotica 17, 1279-1291.

Sipes, I. G., and Gandolfi, A. J. (1986). Biotransformation of Toxicants. InCassarett and Doull's Toxicology—The Basic Science of Poisons (M. O.Amdur, J. Doull, and C. D. Klaassen, Eds.), 3rd ed., pp. 65-98. PergamonPress, New York.

Spaethe, J. M., and Jollow, D. M. (1989). Effects of cobalt protoporphyrin ondrug metabolising enzymes. Biochem. Pharmacol. 38, 2027-2038.

Waxman, D. J., and Azaroff, L. (1992). Phenobarbital induction of cytochromeP-450 gene expression. Biochem, J. 281, 577-592.

Wrighton, S. A., Vandenbranden, M., Brown, T. J., Van Pelt, C. S., Thomas,P. E., and Shipley, L. A. (1991). Modulation of rat hepatic cytochromesP450 by chronic methapyrilene treatment. Biochem, Pharmacol. 42, 1093-1097.


effects of induction and inhibition of cytochromes p450 on the

Documents