316

Binchimiea et Biaphysica Acta. 1073(1991)316-323 © 1991ElsevierSciencePublishers B.V.030,~-4165/91/$03.50 ADONIS 030~16591000970

increased catalytic activity of cytochrome P-450IIE1 in pericentral hepatocytes compared to periportal hepatocytes isolated from pyrazole-treated rats E l i s a D i c k e r , T e r e s a M c H u g h a n d A r t h u r I. C e d e r b a u m Department of Biochemistry, Mount Sinai School of Medicine, New York, NY (USA)

(Received 12 June 1990) Key words: Cytochrome P-4501Iel; Livermetabolism; Pyrazole;iRat hepatocyte) Cytochrome P-450IIEI is induced by a variety of agents, including acetone, ethanol and pyrazole. Recent studies employing iramlmohistochemleal methods have shown that P-45011EI was expressed primarily in the pericentral zone of the liver. In order to evaluate whether catalytic activity of P-45011EI is preferentially localized in the pericentral zone of the liver acinus, the oxidation of aniline and p-nitrephenol, two ¢~fective substrates for P-45011EI, by periportal and pericentral hepafneyles isolated from pyrazole-treated rats was determined. Peripertal and pericentral hepatocytes were prepared by a digitonin-collagonase procedure; the marker enzymes glutumine synthetase and y-glutamyl transpepddase indicated reasonable separation of the two cell populations. Viability, yield and total cytochrome P-450 content were similar for the periportal and perleentral hepatocytes. Pmlcentral hepatocytes oxidized aniline and p-nilrophenol at rates that were 2-4-fold greater than periportal hepatocytes under a variety of conditions. Carbon monoxide inhibited the oxidation of the substrates with both preparations and abolished the increased oxidation found with the pericentral hcpatocytes. Pyrazole or 4-methylpyrazole, added in vitro, effectively inhibited the oxidatien el aniline and p-nitrophenol and prevented the augmented rate of oxidation by the pericentral hepatocyfes. Western blots carded out using isolated mtcrosomes revealed a more than ?.-fold increase in immunochemical staining with microsomes isolated from the pericentral hepatocytes, which correlated to the 2-4-fold increase in the rate o! oxidation of aniline er p-nitrophenol by the pericontral hepatocytes. These results suggest that functional catalytic activity el eytochrome P-4f~OIIEI is preferendally localized in the pevicentral zone el the liver ~einos, and that most el the Induction by pyrazole at P-45011EI appears to occur within the pefieentral zone.

Cytoehrome P-45011E1 is induced by a variety of agents including ethanol [1-31, acetone [4,5], isoniazid [6,7], pyridiue [8], and the potent inhibitors of alcohol dehydrogenasc, pyrazole and 4-methylpyrazole [9-13]. A~sociated with the induction of cytochrome P-45011E1 by these agents there is an increase in the oxidation of a variety of substrates, including aniline [1,6.14], pnitrophenol [15.16], nitrosamines [7.9,17], alcohols [2,3,18], acetone [4], and pyrazole [19]. Cytochrome P4501IE1 appears to play an important role in the metabolism of carcinogens such as various nitrosamines [17,20] and toxins such as carbon tetracbloride, acetaminephen and benzene [21-23].

Correspondence: A.t. C~derbaum, Departm©nt of Biochemistry, Box 1020. Mount Sinai Sch~l of Medicine.One Gustave L. Levy Place~ Ne~ York,NY 10029,U.S,A.

The enzyme profile and associated metabolic processes of the Uver may be heterogeneously distributed with the liver acinus [24-27]. The aeinar zone I closely surrounds the hepatic terminal portal venule, zone IlI surrounds the hepatic vein, and zone 11 is intermediate between the pcriponal and pcriccntral regions [28]. Among the functions proposed to be preferentially localized in the [~riporlal zone arc oxidative energy metabolism, amino acid catabolism, r-oxidation, and gluconeogenesis, whereas glycolysis, lipogenesis, urea synthesis and certain hiofransformation reactions appear to be enriched in the pericontral zone [24-27,29]. In general, the content of cytochrome / - 4 5 0 and the activity of NADPH-cytochrome P--450 reductase predominate in the periccntral zone [30-36]. The phenobarbital-inducible /:'-450 isozyme, the 3-methylcbo!anthrene-inducible P-450 isozyrne, aryl hydrocarbon hydroxylase activity, and ethoxycoumarin-o-deethylase activity are p~ferentially localized in the pericentral zone [33-42]. Phenobarbital and 3-methyleho]anthrene ap-

pear to induce the respective P-450 isozymes in both perieentral and periportal zones of the liver acinus [33,35,3~]. With respect to P-450IIE1, two recent reports which employed irnmunohistochemistry and immunoanalysis (Western blots) of microsomes isolated from periportal and periceutral zones of the rat livers indicated that P-45011E1 was expressed aimost exclusively in the pericentral zone [43,44]. Induction of P-45011E1 by ethanol occurred primarily in the pericentral zone, although some induction also occurred in the periportal and mid-zone regions of the liver aainus [43,4zt]. Functional, catalytic activity of P-45011EI in intact periportal or pericentral hepaloeytes isolated from control rats or rats treated with inducers of P-4501IEI has not been described. Gasenn-Barre et a!. [45] recently reported that aniline, an effective substrate for oxidation by P45011E1, was oxidized at only slightly higher (non-significant) rates by perieentral hepatocytes compared to periportal hepatocytes. The hepatocytes were isolated from fed, control rats, a nutritional condition in which P-45011E1 levels ate very low [46,47]. The experiments described in the present report were conducted in order to evaluate whether functional activity of P-45011EI is preferentially localized in the pericentrai zone of the liver acinus. The oxidation of aniline and p-nitrophenol in periportal and pericentral hepatocytes isolated from rats treated with pyrazole as the inducer of P-4501IE1 was determined as a reflection of the P-450IIEI catalytic activity. Materials and Metbeds Methods Male, Sprague Dawley rats weighing 150-175 g were injected intraperitoneally with pyraznie (150 m g / k g body weight) once a day for 2 days and fasted overnight prior to the day of the experiment. Hepatoeytes derived mainly from either the periportal or pericentral regions of the liver were prepared by a digiioniu-collager~z~ perfuzioa procedtwe, according to Lindros and Penttila [48] with ",linor modifications. After sodium pentobatb~ad (~o m s / k g body weight) anesthesia, the liver was perfused with pH 7.4 Hunk's buffer containing Ca 2+ (but not Mg 2+) via the portal vein at 370C and at a rate of 40 ml/mln. When the liver was completely blanched, the rate of perfusion was decreased to 20 ml/min and a second cannula was introduced through the superior vena cava. The inferior vena eava was then figured above the renal vessels. To prepare pe.riporlal cells, a pulse of digitonin (7.0 raM) was Jnfu,sed through the superior vena cava at a rate of 10 ml/mln for 15-20 s accordin 8 to the size of the liver. The digitouin-digested ceils w e ~ immediately flushed with Ca 2+ free Hank's buffer perfused through the portal vein at 20 m l / m i n for 10 rain to remove any

remaining digllonin. Periportal cells were isolated by perfusion with Hank's buffer plus calcium for a few rain, followed by perfusion with 2O0 ml of the same buffer containing collagenase (40 U/rol). Pericentral cells were prepared by reversing the directions of perfusion. Digitonin was added through the portal vein and eollagenase was added through the superior vena earn. After the collagenase digestion, cells were suspended and washed twice by centrifugation at 35 × g for 2 rain in Kz-ebs-Hepes buffer (pH 7.4) supplemented with 1% defatted bovine serum albumin (BSA). All buffers were saturated with a mixture of 95% CO2/5% 02. Viability was determined by Trypan blue exclusion. Verification of enriched periportal or pericentral fractions was determined in the sonicated hepatocytes by assaying for ginranune symherase [49] and ¥-gintamyl transpeptidase. The tatter was assayed with the use of Sigma kit 545 with the following modification: before step 5 of the procedure the incubation was terminated by addition of 0,3 ml of 20% trlehloroacetie acid and aher eentrifugation the pellet was discarded and the snpernatant used for the assay. The metabolism of p-nittophenol to 4-nitrocatechol or aniline to p-aminophenol was determined using about 10 mg liver cell protein, varying concentrations of p,'tittophenol or aniline and Krebs-Hepes*BSA buffer containing 20 mM xylitul in a final volume of 1 rid. Reactions were conducted in 25 m| polyearbonate flasks at 3 7 ° C and were initiated by the addition of aniline or p-nitrophenol_ Reactions were terminated at various time points by the addition of trichinroaeetic acid (TCA) to a final concentration of 5% (w/v). The samples were centrifuged and aliquots of the supernatant were assayed, for 4-nlrrocatechol by adding 0.1 mi of 10 M NaOH per nd supernarant and immediately determining the absorbance at 546 rim, or assayed for paminophenol hy adding 0.2 nd of 2.5 M sodium carbonate per ml of supernatant followed by 0.1 ml of a 5% phenol in 2.5 M NaOH solution. After 30 rain of color development, the absorbancc at 630 mn was determined. Concentrations of 4-nitroeatechol or paminophenol were calculated using extinction coefficients of 10.28 or 15 mM - t . em -z, respectively [15,50]. All values were corrected for zero-time controls in which the TCA was added prior to initiating the reaction with substzare. The content of cytochrome P-150 was determined by the method of Omura and Sato [51] after resuspending isolated c,~lls in a sodium phosphate buffer (pH 7.4) containing 0.7'~ Er~u~gcn 91i, 0.55 sodium cholate, 0.1 mM EDTA and 20% glycerol, and sonieating for 15 s. Microsomes were isolated from the pcriportal and pericentral hepatocytes by differential centrifugation and a quantity corresponding to approx. 0.01 nmol total cytochrome P-450 was subjected to SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes as previously described

318 [13,52]. The membrane was incubated with anti-pyra7.o1¢cyto~hrome P-450 IgG and subsequently with goat anti-rabbit IgG conjugated with horseradish peroxidase, and the blots stained with HRP colour developing reagent (4-ehioro-l-naphthol plus H202). Immunoreactive proteins were quantitated by scanning the blots with a LKB UltroScan XL dcnsitometer; all gels cont~ned cytcehrome P-45011E1, purified from pyrazol0-treated rats [13] as a standard. The IgG was raised in rabbits against the purified pyrazole-indaced cytochrome P450lIE1 isozyme as previously demr~bed [13,52]. Materials

Rats were purchased from Charles River Laboratoties; collagenase type 2 was from Worthington Biochemicals, Freehold, NJ; mgitomn was from BD H Chc,rl--':ca1~rdal!erd-Seb-lesinger; Sigma Kit 545, defatted serum albumin, p-nitrophenol and aniline were from Sigma Chemicals, St. Lords, MO; Pyrazole was from Aldrich Chemicals, Milwaukee, WI; goat anti-rabbit IgG conjugated to horseradish peroxidase was from Boehringer Manaheim, Indianapolis, IN; and HRP colour development reagent and chemicals for the gels were from Bio-Rad, Richmond, CA. Results Hepatoeytes were isolatigi from the periportul and perieentral regions of the liver acinus using the methods proposed by Lindros and Pantilla [48] and Quistorff [53]. Pyrazolo was used to induce cytochrome P-450IIE1 as induction by this agent in rats has been well established and characterized [9-13]. Ghitamine synthetase has hcen shown to have a preferential localization in the pericentral zone, while y.giutm'nyl transpeptidase appears to be a good marker for the periportal region [54]. The activities of these two enzymes were determined in digitonin eluates (lysates) and in sonicated homogenates prepared from the isolated kepatoeytes. The periceotral eiuates showed a 40-fold higher activity of glutamine

synthetase as compared to the periportal eluates (Table l). In contrast, the a~tivity of y-giuw.myl transpeptidase was 20-fold higher in the periportal eluates and the ratio of ghitamine synthetase to -/-ghitamyl transpeptidase was markedly higher with the pericentral ehiatas. Similar results were found with the isolated hepatocytes, although the differences were not as marked as those found with the ehiates. Glutaminc synthetase activity was 9-fold higher with the pericentral heparocytes, whereas y-glutamyl transpoptidase activity was 5-fold higher with the periportal hepatocytes, resnldng in a 50-fnld difference in the ratio of activities of glmamine synthetase to y-glutamyl transpeptidase (Table I). These results suggested that there was reasonably effective enrichment of the hepatocyte preparations with either cells from the periportal or pericentral zones of the liver acinus. Viability of the hepatenytas from the two zones was similar and generally ranged between 85-90%. The yield of periportal or pericentral hepatocytes per liver was also similar: about 500 mg liver cell protein (equivalent to .~bont 220 million cells) was obtained per liver for rats weighing about 175 g. The content of oytochromo P-450 was also similar for the two types of hepatoqtte preparations (Table 1). Pyrezole treatment has previously been shown not to increase significantly the total content of P-450 in rat liver mierosomes [18], although the isozyme composition is altered. To assess P-450 function in the isolated periportal and pericentral hepatocytes, the oxidation of aniline and p-nitroplienol was determined. These agents are metabolized only by P-450-depend0nt reactions and they ase effective 0ubstrates for eytcehrom¢ P-4501IEI [1,6,14-x6]. We have found that these two substrates are oxidized by intact hepasoc3/tes isolated from rats treated with pyrazole at rates which were about 3-4.-fnld higher than saline controls [fulal, indicating the effectiveness of these two substratas in demonstrating induction by pyrazole of P-45011E1 in a mixed liver oeli population. Therefore, it was considered that these two

TABLE I Marker enzyme aaicities and ¢ontent of cyt~throme P.450 in perlponal and pericentrat preparation~

The activities of glutamine synthetase and y.glulamyl transpeptidase (expressed as units per g Uver cell ~eteini and the content of total cytochroroe P-450(nmol per mg liver cell protein) were determined in either the eluate fra¢ilonsor the hepatccyt~ isolated from peliportni or pcrleentral regions of the liversof pyr~ole-tr©ated rats. Values for the ©luatefractions ate from two to tour preparations, whereas t~sults for the hepat~ytes ~ from eight to 16 preparations. Results are expressed as the mean:kstandard ert~r. Enzyme

Eluale pefiport~

Olutaminesynthetes¢ y-Glm~myltranspeptldase Glutamiassynthetase/-~-glut~mayl transpeplidasc Cylochrolne P-450

9.05±1.S2 3J~ 3.02

Hepalo~yte

p~ricentral 382±14.5 0.1S 2547

~ricennni/ lamportal 42.2 0.05

p~fipona]

~fic~nu~d

3.42 ±0,8 3,03 ±1.11

~1.5 ±4.'/3 0.50 ±0~7

pelk:cntrld/ p~portal 9.2 0.19

1,I3 0.139±0.014

53.3 0.133±0,019

0.96

.....1

1

o 1/./ .....1 / / " Time(mln) Fig. l, Time-tour. for the oxidation of 1 mM aniline (A) or 0.35m i p-nitroplieno! (B) by petlpottul (~) . d pericentral 0l) hepatocyt~ i~lated from pyrazole-treated rats. Experimentswere performed as described under Materialsand Methods. Results &refrom five to nin2 preparations of hepatoeylesand refer to the mean+ S.E. Wherenot indicated th.*5.E. was within the area of the symbol, Agall thr:epoints, the rate of oxidation of aniline or p-nitrophenul was sig~ificantlygreaterby penc~ntrM,comp~¢d to penIx)rtal, he]~atocytm (P < 0.01).

substrates might prove effective in demonstrating possible zonation in the catalytic function 0f P-450IIE1 after pyrazole treatment. A time-coarse for the oxidation of 0.35 m M p-introphenol and 1 m M aniline by hepmo¢ytes isolated from periportal and pericentral zones of the liver is shown in Fig. 1. Activities were linear for about 10 rain (p-nitrophenol) or 15 rain (aniline), and the rates of p-nitrocatechol or p-aminophenol formation were more than 2-fold greater with the pericentral hepato~tes (Fig. 1). Substrate concentration curves for the oxidation of aniline and p*nitrophanol are shown in Fig. 2A and B, respectively. Oxidation of aniline was maximal at about 5 m M with the isolated hepatocytes, whereas t.be oxidation of p-nitsophenol increased up to a maximal value at 0.20 m M p-nitrophenol, and then levelled off or began to decline. Substrate inhibition of p-nitrophanol

....~ - "

o.,/.~i,~'~; ; £ ........ "....

oxidation has been observed by othc~ {15,16]. At all concentrations of aniline or p-nitropheaol utilized, r a t ~ of oxidation weae significantly greater with the I~ncentral hepatoeytes (Fig. 2), Maximal differences between the pericentlal and perlported hepatecytes were found at tlie lower concentrations of mfiline (0.5-1 mM, Fig. 2A) or p-nltrophenol (0.05 raM, F i g 2B), suggesting preferential localization of low K= r~tabolizing enzymes in the pericentral hepatneytes, To demonstrate a role for cyt~hrome P-450 in the oxidation of aniline and p-nitrophanol by the periporzal and pericentral hepatocytes and in the increased rates found with the periuentral hepatocytes, the effect of carbon monoxide as an inhibitor of P-450-eatalyzed reactions was evaluated. The effect of a 5 0 ~ carbon monoxide/50~ air mixture was compared to that of a 50~ a r g o n / 5 0 ~ air mixture. "Vile latter produced a

° °~o2"

£~-'~i'. . ,

o~a

oxidation of 025, 0,50, I, 2, $ ~at] 10 r,tM aniline {A) o¢ 0,05, O,lO, 0,20, 0.35 ~ad 0.5 mM p-nitropheaol (B) was ~sayed as desclibed under Matefia~ and Methods. Resultsare fron five to nine pz,7~,ratiol~sof hepatocyl~ and ~f~ to the mean:kS.E. At ul] con~aztrations, the rate of oxidation of aniline or n.~Jtrophenol was slgalficantly gz~a~erby ~ e e n w a L ~mpa.~'~ te ~,~*.~- =~.-~-~m-t~3a.m( P ,:.~,,-.~ ~ from < 00~ to < o,ool).

320 TABLE n Effect o/carbon monoxide on the oxidation of anil:ne and p.nitrophenol b)"perlportal and per, cen,~l neparocytes Tile oxidation of aniline or p-nitrophenol by periportal or pcficentral h e p a t ~ ) t ~ was datelined under air or in the presence of gas rmxtur~ ~nmining 50% air plus 50% argon or 50% air plus 50~ c~bon monoxide. The cff~t of CO is expressed reiatiue to the argon values, to a~ount rot any effect due to the lowered ~n~mratiun of oxygen. Gas mixture

Substrate

100~ air 50% argon 50% CO 1005 air 50% argon so~ c o

anihne

Rate of oxidation periportal pcrieentral (mul/zrfin/mg av~ ~ll protein) 0.046 0.231 0.030 0A70 0.010 0.029 0.060 0.290 0.050 0.210 o.oo0 0.020

p-nitrophenol

2 0 - 3 0 % i n h i b i t i o n of the o x i d a t i o n o f a n i l i n e a n d p nitrophenol by both the periportal and pericentral hepatocytes as c o m p a r e d to 100% air, s u g g e s t i n g s o m e a n a e r o b i c effect (Table II). C a r b o n m o n o x i d e p r o v e d t o b e a p o t e n t i n h i b i t o r o f the o x i d a t i o n o f b o t h substratez b y t h e p e r i p o r t a l a n d perieentral hepatocytes. M o s t of the increase in t h e o x i d a t i o n of a n i l i n e o r p - n i t r o p h e n o l b y the pericentral h e p a t o c y t e s w a s sensitive to c a r b o n m o n o x i d e , suggesting t h a t this i n c r e a s e d o x i d a t i o n w a s d u e to a c y t o c h r o m e P - 4 5 0 - d e p e n d e n t r e a c t i o n (Table II).

Effect of CO periportai

pericentrai

-67

-83

- 04

- 9O

P y r a z o l c a n d 4 - m e t h y l p y r a z o l e react in vitro w i t h c y t o c h r o m e P-45011E1 to p r o d u c e s u b s t r a t a b i n d i n g spectra [55,56] a n d c a n i n h i b i t P-450IIEl-eata12¢zed o x i d a t i o n o f o t h e r substrates, e.g., e t h a n o l [57,581. Pyrazole a n d 4 - m e t h y l p y r a z o l e w e r e s t r o n g i n h i b i t o r s of the o x i d a t i o n of aniline, a n d especially p - n i t r o p b e n o l b y periporta| and peficentral hepatoeytes (Table lll). With e i t h e r substrata, b o t h h e p a t o c y t e p r e p a r a t i o n s w e r e equally affected b y p y r a z o l e a n d 4 - m a t h y l p y r ~ o ! e . I n a g r e e m e n t w i t h t h e results o b s e r v e d w i t h isolated m i c r o s o m c s [57,58], 4 - m c t h y l p y r a z o l e w a s m o r e t h a n a n o r d e r

TABLE Ill Inhibition of the oxidation of p-nitrophenol or aniline by 4.metbylpymzole or py,azole in periporlal and pericentral hepatocytes The oxidation of 0.35 raM p-nitrophenol or 2 mM aniline by periporlal or pericentral hepatocyles was determined in the presence of the indicated concentrations of 4-methylpyrazolc or pyrazole. Results are from two preparations.

p-nitrophenol

Bone 4-methylpyrazole

pyrite

Ani0ne

none

4-rnethylpyraznie

Con~ntration of inhibitor (raM)

Rate of oxidation Effect of 4-methylpyrazole periportal pericentral or pyrazole (nmol/min per mg liver cell protein) peripor tal perieentrai {%)

o o.oi

0.150

0.470

-

-

0.048

0.122

- 68

- 74

0.02

0.019

0.0S8

- 87

- SS

0.05

O.Ol1 0.O06

0.062

- 93

- 87

0.034

- 96

- 93

0.052 0.032

0,184 0.112

-60

o.o13

0.038

-61 -70 - 88

0.076 0.0S0 0.040

0.220

0.25 0.2 0.5 2,5 o 0.02 o,os

o.lo 0.25 pyrazole

0.50 0.2 ~5 2.5

-70 - 91

0.020

0.132 0.088 0.000 0.048

- 34 -47 - g0 - 74

0,008 o.oss

0.034 0.2oo

0.050 0.057

0.160 0.160

- 11 -22 - 25

-40 60 - 61 - 78 -85 - 9 -27 - 27

- 89

0.042

0.116

-45

-47

o o31

O.3OOL a

/

i

"

/t= 1

/: 0.150

oooo

:~;

o.ooo

" 2.000

o ooo J / / ~ ; ~ " . , o.ooo 2.000 4.000 Concentrationof aniline(raM)

~ 4.000

Concentrofionof on[line(ink4)

•

,

Fis. 3. Substrate concentration curve for the inbJEtion of mailineoyddation by 4-methylpyrazol¢in pefiportul (AI or pcncenw,d IB) hClralc~yt~. rhe oxidation of 0 5, l, 2 and 5 mM aniline was determined in the ab~nce (periportal It. pericentralI1) or p r ~ of ¢ith~ 0,05 mM (,~) or 0.25 mM (v) 4-methylpyrazole. Results are from two p~'eparations. With periportal hepatocytes, the pelccnt inhibition by 0.05 and 0.25 mM 4-melhylpyr~-ole was 81 and 91.70 and 85. 47 and 74 and 13 and 26% at aniline concentrations of 0.5. 1.2 and S raM, respectively.With pericemral hepatocyl¢~the percent inhibition by 0.05 and 025 mM 4-methylpyrazobiw~ 82 and 94, 71 and 86. 60 and 73 and 27 and 51% at aniline concenlrationsof 0.5.12 and 5 mM, r~pectlvely. Of magnitude more effective than pyrazole as an inhibitor in the isolated hepatocytes (Table Ill). Most of the increare, in the rate of p-nitrophenol or aniline oxidation by the perieentral hepatoeytes was sensitive to inhibition by pyra2ole and 4-methylpyrarole. To evaluate the type of inhibition by 4-methylpyrazole, the rate of oxidation of various concentrations of aniline in the presence of a freed concentration of the inhibitor was determined. Results in Fig. 3 show that as the concentration of aniline was elevated over the range of 0.5 to 5.0 raM, the extent of the inhibition of aniline oxidation by 4-methylpyrazole in the periportal as well

as the pedcentral hepatocylcs was decreased, suggesting a competitive componem to the inhibition by 4-methylpyrazole. Irtgelman-Sundberg et al. 143] reported that the intensity of staining of P-45011EI after ethanol treatment was 2.g-fold greater with perieentral hepatocytes as compared to penportal hepatocytes. Western blots were performed with microsomes prepared from the periportal and pericentral hepatocytes using an anti-pyrarole P-450 I,gG (Fig. 4). There was an increase in irnraunochemical staining with mierosomes isolated from the pericerttral hepatocytes (lanes 1, 2, 3, 7, 8 and 9) as compared to mierosomes isolated from the periportal hepatocytes (lane: 4, 5 and 6, Fig. 4). Densitomerry measurements indicated that there was a 2.4-fold increase in intensity in the tmmunoblots of microsomes from the pericentrul hepatocytes, confwnung that the increase in p-4501lE1 apoprotein in peficentral microseines found after ethanol treatme-t is also observed after pyrazule treatment. Discassiea

1

2

3

4

5

6

7

8

9

10

Fi~. 4. Westernblot for the immunochemicaldetectioll of P-4~IIEI in mistletoes isolated from petiportal and pericentralhepatocyt~ of pyrazole-tt~ated rots. Approx. 10 pmol total cytochl~me p-150 w~ applied pet lane to lanes 1-9, while 1.3 pmol of P-4501IEIpulified from pyrazolc-treatedrain wa~ applied to lane lO. Lanes 1, 2, 3, 7, 8 and 9 c~n~aed n~croson'~ from penceatral hq,atoy==, whereas la/tcs 4, 5 and 6 contained ~ from petlporlal hepatocybis. Densitomctl~ ~ t s iatdicaled that lanes ~ 6 avmaged 132 arbitrary irons of P450tIEl. while lanes 1, ~ 3, 7, S and 9 averaged 318arbitrary units of paSOIIE1.

The preferential localization of T-giutamyl transpeptidase in pmiportal regions of the liver acinus [59,60] and of giutamlne synthetase in the pericantr~ region [61,62] have been shown by histochemical and immunohistochemical methods. As a consequence, the ratio of these two ~ has been suggested to serve as a powerful masker of the asinar origin of isolated hepatocyt¢ populations [54]. As shown in Table I, the carlo of ~utamine synthetase activity in pexicentral to periportal hepatoeytes was 9, this value falls within the range reported by others: 10 by Lindros et aL [54] and 4 by Gascon-Barre et al. [45]. The ratio of ¥-glutamyl

transpeptidase in pericentral to periportal hepatoeytes was about 0.2, which is similar to the values of 0.29 and 0.35 reported by Lindros et al. [54] and Suolinna et at. [631, respectively. These results confirm the utility of glutamine synthetase and y-glntamyl tmnspeptidase as biochemical markers for the reparation of periportal and pericentral hepatocytes [54| and suggest that adequate separation of the hepatocytes was accomplished under our conditions. However, it should be pointed out that the enrichment of glutamine synthetase in pertcentral hepatoeytes O-fold) or of 7-glutamyl transpeptidase in periportal hepatocytes (5-fold) was considerably less than the enrichment of these enzymes in cellular lysates (Table I). Therefore, even though the individual preparations contain more of the hepatocytes from one zone as compared to the other, they are not completely 'pure' preparations of periportal or perlcentral hepatocytes. Since pyrazole treatment appears to induce only P450IIE1 [12] and aniline and p-nltrophenol are effective substrates for this isozyme, it was considered that rates of oxidation of aniline and p-nitrophenol could be taken as a reasonable reflection of the catalytic activity of P-.450IIE1 in the intact hepatoeytes. Hepatocytas isolated from the perleentral region of the liver Iobule of rats treated in vivo with pyrazole oxidized aniline and p-nitrophanol at an elevated rate compared to pertportal hepatoeytes under a variety of conditions. The increased rate of oxidation was sensitive to carbon monoxide, suggesting the requirement for P-450, and to pyrazole and 4-methylpyrazol¢, added as competing sabstrates. The recent immunoloEical results of Ingelman-Sundherg et al. [43] and Tsutsmid et at. [44] indicated that P-450IIE1 displays a preferential localization in the pericentral region of the liver. The increased oxidation of aniline and p.nitrophenol by pericentral hepatocytes confirms this preferential localization, and also indicates that functional activity of P-45011E1 is greater in pericentral hepat~cytes. MoT~over, the 2-4fold increase in oxidation of aniline and p-nitrophenol by the pericentral hepatocytes is identical to the increase in immunoblot staining for P-4501lE1 epoprota]n as observed with mi,~osomes isolated from pericentral hepatocytes from pyrazole-treated rats (Fig. 4) and from ethanol-treated rats [43]. A recent communication by Gascon-Butte et at. 145] suggested no specific zonation for aniline oxidation in hepetocytes isolated from fed, non-induced rats. Rates of aniline oxidation were reported to be about 0.28 and 0.35 nmol per rain per rtmol total P-450 for petiportal and perieentral hepatocytes, respeetivel3, [45]. After pyrazule treatment, rates o1 oxidation of 1 mM aniline by periportal or perieentral hepatoeytes were 0.034 and 0.140 nmol per rain per nag liver cell protein (Fig. 2). Utilizing the P-450 values shown in Table I (about 0.13 nmol Po450 per rng protein), these rates of aniline

oxidation would correspond to values of about 0.26 and 1.1 nmol per rain per nmol total P-450 for periportal and perieentral hepatocytes, respectively. Thus, rates of aniline oxidation by periportal hepatocytes front "ontrol rats or pyrazole-treated rats appear to be similar, whereas there is a 3-fold increase in the rate of aniline oxidation by pericentral hepatocytes from pyrazoletreated rats compared to control rats. These results would suggest that most of the induction of aniline oxidation by the pyrnzole treatment occurs in the peficentral zone. Under certain conditions, oxidation of drug substrafes by intact cells (as opposed to microsomal systems) may be limited by the availability of the N A D P H cofactor [42,64|, and this may affect efforts to link the increased oxidation of the drug substrata to the induction of the appropriate P-450 isozyme. Rats were starved for 24 h to deplete glycogen and to lessen the possibility that varying rates of aniline or p-nltrophenol oxidation between peticentral and periportal hepatocytes could be due to different amounts of glycogen in the hepatocytes; in the fed state, NADPH is largely derived from glucose metabolism via the p~ntose phosphate cycle. Xylitol was added to both hepatocyte preparations because this substrata has been shown to be an effective generator of N A D P H and to support P ~ 5 0 function in intact liver preparations [64-86]. Belinsky et al. [42] have shown that the N A D P H / N A D P redox ratio was similar for perieentral and periportal zones of the liver Iobule and that xyfitol increased this ratio and stimulated ¢thoxycoumarin metabolism comparably in both zones of the liver. In summary, perlcentral hepatocytes isolated from pyrazole-treated rats oxidized effective substmtes for P-450IIE1 at elevated rates compared to periportal hepatocytes, suggesting that catalytic activity of P45011E1 is preferentially localized in the pericentral zone of the liver. The enhanced rate of oxidation of aniline or p-nitropheunl by the peficentral hepatocytes appears to correlate with the increased content of P45011E1 detected immunochemiceUy in the peficentral zone of the liver acinus. Since toxins such as carbon tetraehinride or acetaminophen or ethanol, which are oxidized by P-45011E1 cause selective destruction of the centrilobular zone of the fiver, it would appear reasonable to specuh~te that increased oxidation of these substrates due to the elevated catalytic activity of P-45011EI in the pericentral zone may play a role in the toxicoingicat actions of these agents. Acknowledgements These studies were supported by USPHS Grant AA06610 from the National Institute on Alcohol Abuse and Alcoholism. We thank Ms. Lucy Martthcz for typing the manuscript, We thank DI. Kai O. Lindros, Alko,

Helsinld, F i n l a n d , for visiting a n d t e a c h i n g u s t h e m e t h o d o l o g y for p r e p a r i n g p e r i p o r t a | a n d p e r i c e n t r a l hcpatocytes. References I

Morgan, EmT.,K ~ p , D.K and C~r~ MJ. (1982) 1. Biol. Chem. 257.13951-13957. 2 Koop. D.R., Morgan, E.T.. Tart, G.E. and Coon. M3. (19821 J. Biol. Chem. 257, 8472-8480. 3 Ryask D.E. Koop. D.IL, Thomas, P.E.. Coon. MJ. and Levm. W. (1986) Arch. Bio(h*m. Biophys. 246. 633-644. 4 Koop. D.R. and Casa72~ J.P. (1985) J. Biol. Chem. 260.136O713612. $ Patten, CJ., Nine, KM., Lu. A.Y.H. ~Lndy ~ g , C.S. (1986) Arch. Biochem. B~ophys. 251, 629-638. 6 Ryan, D.E., Ramanalhan, L. lida, S., Thomas. P.B., Haldu, M., S~ivf]y, J.E., Li(ber, C.S. ant] l.,e~TJn,W. (1983) J. Biol. Chem. 26O, 6385-6393. 7 Thomas, P.B. Bandier& $,, Maines, S,L. Ryan, D.E. and Levin, W. (1987) B~ccherohtry 26, 2280-2289. 8 Kim, S,D., Williams. D,P-, Schuet~ E.D. Guz¢lian, P,J, and No~aR, R.F. (1988) J. Pharma~L Exp. Ther. 246,1175-1182. 9 TB, Y.Y.. Sonmmbe~g. J. Lewis, K.F. and Yah& CmS. (1981) B~c~hem, Biophys, Res, Commtm~ 103, ~5-912, l0 IngehnarPSundb~g, M. and JornvaB, H. (19M) Biochero_ Bi~ phys. Res. Coraraon. 124. 375-382. 11 Koop, D.R~ Cntmp, B.L, NordblonL G.D. and Coon, MJ. (1985) Proc.Natl, Acad. Sc~ USA 82, 4066-4O69. 12 Son& R.J., C-¢lboln, H., PaxL SmS~Yaag, C.S. and Gonzale~ F, (1950 £ B~oLChem. 261,16689-16697. 13 Palakodety, R.B., Clejan, L.A.. Klik-un, G.. Feierman, D.E. and Ce~baurrk A.1. (1988) J, Biol. Che1~. 263, 878-$84. 14 M s , G. and GraBt. L, (1976) mochem, pharmacoI. 25. 21e~72201. 15 R ~ k e , L.A, and Moye¢, MJ. 0985) Drug Mctab. Dispos, 13, 543-552. 16 Koop, D.B. (1986) Molec. PharmacoL 29, 399-404. 17 Yah& C.S.. Keep, DJK, Wang, T. a~l Coon, MJ. (1985) Bicchel~ Sic~y~ Res. Commun. 128, I007-1013. 18 Igaikml. O. attd Cederbamzh A.I. (1984) Biochlm. Bioph3 s. Acta 801.131-137. 19 Feierman, D.E, and Ced~baum, A.I. (1987) Drug Mclab. Dis[ms. 13, 634-639, 20 IO~,h~ds, DraM, goop, D.R., ScarpeBi, D.D., Coon, MJ, and HoHenber~ PmF.(1~5) Calx~ R~,. 45, 619-624. 21 Morgan, F_T.. Koop. DmR. and Cool~ MJ. (1983) Biochem. Biophys. R~. Coramun. 112, 8-13. 22 .Iohanssorh I. and ln~¢lman-Sundber 8, M, (1985) FEBS L¢tt. 183, 263-269. 23 Koop, D,R., Umfllcm, C.I. and ,~bnJ~, O.G. 0989) Tox~col. Appl. Phannaool.9~. ~78-288. i i 3tmserman~ K. 0986) Enzyme 35.161-180. J ~ IL and Katz,N, (1982) Hepatology 2, 385-395, Thurman, R.D, and Kauffman, F.C, (1985) Hcpatolosy 5,144151. 27 Gmmur.i~ JJ. (1989) Hepatology 9,154-160. 28 Rappaport, A.M. (1958) Anal Rec. 130, 637-686. 29 "galxman, I~G., Kaoffman, F.C. aud J ~ K. (1986) ReKoLttion of Hepatic MetaboUsm: L~t~aaud Intercenular C~'npartmem~tion, Pie~em Press, New YotL 30 G o o ~ P.F-, C h a ~ l , J., Saw,/eL B, and Slater, T.F. (19T~) Chenx-mol. interact. 20, 299-310. 31 ~ Y., Gn~B~B, Pro.K~ke, G.F., Red.k, £Am ~ d B a r ~ J. (1980) MoL PharmacY. 18, 304-312. . 32 Sl~th, M,T. and Wills, E.D. (1981) FEBS Letc, 127, 33-36.

33 Baron, J., Redlck, J.A. aad Gucngcfich. F.P. i1981) 1. Biol. (hem, 256. ~031-5937. 34 Baron, J., Ral~k, £A. ~,~ Gucn~¢h, F.P. ¢1982l J. reel. C h ~ , 257,933-9S7. 35 ~ e n g t ~ n . G., Julklmen. A., Pentlila, K.~. and Lindrc~ K.O., (1987) J. PharmacoL Exp. Tker. 240, 663-667. 36 Vp.anancn, H. (1986) 1. Hepatolo~y 2,174~13L 37 OhnishL K., NUchlma. A. and Okuda. K. (19~12)Hcpat~og, y 2, ~49 SSS. 38 Moody. D.F-, Taylor, L.A., SrnukleL F~A., Levm, W, ~mdThom,~, P.E, (1983) DTUB Nietab. Dispo~ 11, 339-343. 39 Smith. M.T.. Redlck, 1~. and Baron, J. (1983} J, Histochem. C~ochcm. 31,1183-1189. 40 WatTenberg, L.W. and Lcon$, J.L. (1962) J. Histcchcm. Cyt~ chem. 10, 412~420. 41 T ~ w 3 . J., Eadou~ H , Sate, A., Hasunlura. Y. and Takeuchi, J., (1988) Biochem. InL 16.1127-1135. 42 lkUr~ky, B.A., L'.a~Umaa, F.C., Ji, S., l.¢mastc, x J.L. ~ d Th~r43 IngelmawSuadber& M., Johamsua, I,, Femtila, K,E., Glaumem~ H. and Lindl~. K.O. ( 1 ~ ) l~ochel,. Bioph~s. R ~ Commen, 137, S5-60. 44 TxmsmnL M.. LaskeL J.M.. ~ M., Rosman. A.S. and Lieber, C.K 0989) Hepatotogy 10, 437-446. 45 Gascon.Bah'e, M., Benlzrahink N, and Tremblay, C. (1959) Can. J. PhysioL P ~ l . 67,1015-102Z 46 Miller, K.W. and YanL C.S. (19~4) Arch. Biochem. Biophys. 229. ~3-690. 47 Hong. J., pan. $,, G o r ~ FJ., Gelboir~ H.V. and Yans, C.S(1987) Bio(hem. Biophys. Re*.. Co~mu~. 142.1077-1083. 48 Lindms, K.O. al~d PezttiJa, K.E. (1995) Biochem. L 228, 757-76O. 49 Wellner, V.P. and Mei~e~, A, 0 % 6 ) B ~ I P y 5, 8"12-879. 50 Micyal, 1,1. and B I ~ , J.L. (1@6) J. Biol. Chem. 251. 34423446. 51 Omura, T. and Sato. R. 0964) J. gioL Chem. 239, 2370-2378. 32 Clejan, I,A., Koop, D.R. and C ~ a u . m , AmI. ~1999) Drug McUIb, Dispo~ 17, 694-698. 53 QuistorfL EL (19~S) Biocllem. $. 229, 221-226. 54 Lindt~. K.O., PeBt~a, K.F-, Janz~. $,G, Mec~16~, A.F.M., ,~y, H. and Israel, Y. (1989) J. Hepatology $, 338-343. 543 Dicker, E., M~Hugh. T. and C e d ~ t l m , A.I. (199D) Bloc.him. Biophys, a¢~a 1035. 249-256. 55 Ruhin, E. Gang. H. and Liebex, C.S. (1971) Biochem. Btoph~ Re*. Commea. 42,1-S. 56 Feierman, D.E. and C~k=b~m. A.I. (1985) piochem. Biophy~ Rc~. Comrmm. 126,1076-1~I. 57 Feterman, D,E. a~d Ccdcdo~mlk A.I, (19~) Biochem, J, 239, 671-677. ~s Fe~rman. D.E ~nd Ced=baura. A.I. (19¢O ~,xhem. e h a n ~ ¢.o1. 36. 32"/7-3283. 59 Ideo. G., T c ~ d . S., eegob=om~ A.. Bdla~ld, O. and RoncM, O., (19eo) J. ~ Ch