Inhibition of CYP1A2 and CYP3A4 by Oltipraz Results in Reduction of Aflatoxin B1 Metabolism in Human Hepatocytes in Primary Culture'

Home , CYP3A4

ICANCER RESEARCH55. 5574-5579. December I. 19951 Inhibition of CYP1A2 and CYP3A4 by Oltipraz Results in Reduction of Aflatoxin B1 Metabolism in Human Hepatocytes in Primary Culture'

Sophie LangouÃ«t,2Brian Coles,3 Fabrice Morel, Laurent Becquemont, Philippe Beaune, F. Peter Guengerich, Brian Ketterer, and AndrÃ©Guillouzo

INSERM U49, Centre Hospitalier Regional Universitaire Ponchaillou, 35033 Rennes Cedex. France IS. L. F. M.. A. G.J: Biochemistry and Molecular Biology Department. University College London. London WIP6DB, United Kingdom [B. C., B. K.J: INSERM U 75. Centre Hospitalier Universitaire Necker. 75015 Paris. France IL B., P. B.!: and Deparunent of Biochemistrs and Center in Molecular Toxicology, Vanderbilt University, Nashville. Tennessee 37232 (F. P. G.J

ABSTRACT AFB1-associated carcinogenesis (8, 9), apparently by inducing ex pression of GST 10 (10, 11), the rat homologue of mouse GST Yc. Dithiolethiones are thought to act as potent chemoprotective agents It has been assumed that FB1 is also a human hepatocarcinogen against aflatoxin B1 (AFB1)-induced hepatocarcinogenesis in the rat by because hepatocellular carcinoma is common in many communities inducing glutathione S-transferases (GSTs). To determine whether these antioxidants can be similarly effective in human beings, we have investi that consume AFB 1-contaminated diets (12). Recently, this associa gated metabolism of AFB1 in primary human hepatocytes with or without tion was clearly established by an epidemiological study showing that pretreatment by oltipraz (OPZ), a synthetic derivative of the natural AFB1 is not only a hepatocarcinogen per se in humans, but is also a 1,2-dithiole-3-thione. Aflatoxin M1 (AFM1), glutathione conjugates of powerful cocarcinogen in combination with hepatitis B virus, which is AFB1 oxides (AFBSGs), and unchanged AFB1 were quantitated in cul often endemic in communities exposed to AFB1 (13). tures derived from eight human liver donors. Parenchymal cells obtained The question arises whether inducers, so effective in the rat, might @ from the three GST Mi-positive livers metabolized AFB1to AFM1and to be of value for the chemoprevention of the -associated hepato AFBSGs derived from the isomeric exo-and endo-8,9-oxides, whereas no cellular carcinogenesis in humans. The potential for GST-dependent AFBSGs were formed in the GST Mi-null celis, Pretreatment of the cells detoxication in the human liver is not as great as in the rat liver with 3-methylcholanthrene or rifampicin, inducers of CYP1A2 and because human liver has about 1 order of magnitude less GST activity CYP3A4, respectively, caused a significant increase in AFB1 metabolism. Although OPZ induced GST A2, and to a lesser extent GST Al and GST toward exo AFBO (1 1) and no apparent homologue for rat GST Ml, it decreased formation ofAFM1 and AFBSG, which involves CYP1A2 10â€”10and mouse GST Yc-Yc. The most active GST toward exo and CYP3A4. Inhibition by OPZ of AFB1 metabolism by reducing AFBO is human GST Mlâ€”l (14), the benefit of which is limited by CYP1A2 and CYP3A4 was also demonstrated by decreased activity of a genetic polymorphism involving a null allele, so abundant that about their monooxygenase activities toward ethoxyresorufin and nifedipine, 50% of the human population has a GST M 1-null phenotype (15). respectively. The significant inhibition by OPZ of human recombinant Human hepatocytes in primary culture have proved to be useful in yeast CYP1A2 and CYP3A4 was also shown. These results demonstrate past studies of GST induction (16). In the present work, the use of that AFBSG can be formed by GST Mi-positive human hepatocytes only, hepatocytes in culture has been extended to studies of the effects of and suggest that chemoprotection with OPZ is due to an inhibition of inducers on the metabolism of FB1. Particular importance is at activation of AFB1, in addition to a GST-dependent inactivation of the tached to OPZ, a pharmaceutical derived from D3T, which has been carcinogenic exo-epoxide. reported to be a natural product occurring in cruciferous vegetables (17). In human hepatocytes, OPZ proves to be not only an inducer of INTRODUCTION GSTs but also an inhibitor of exo AFBO biosynthesis by CYPIA2 and CYP3A4. The mechanism of this effect has been investigated, and the @ I a product of the mold Aspergillus flavus, which infests significance of these findings for chemoprevention of FB1-depend grain and other food stuff stored under warm, moist conditions, is ent hepatocellular carcinoma in humans is discussed. metabolized by certain CYPs to a powerful genotoxin, exo AFBO (1â€”3).AFB1 is therefore a potential carcinogen; however, its carci nogenicity varies from one species to another (4). In the mouse, its MATERIALS AND METHODS hepatocarcinogenicity is negligible (5); this has been attributed to the particularly effective detoxication of exo AFBO by the GST Yc-Yc (6, Chemicals. Culture media and FCS were obtained from GIBCO (Paisley, 7). On the other hand, it is a powerful hepatocarcinogen in the rat Scotland). Collagenase, MC, RIF, bovine albumin, bovine insulin, AFB1,and apparently due to the absence, in any quantity, of a GST that utilizes AFB1 metabolites were products from Sigma Chemical Co. (St. Louis, MO). exo AFBO very effectively (7). However, certain inducers prevent D3T was synthesized by Dr T. W. Kensler, and OPZ was kindly supplied by Dr C. G. Caillard (RhÃ´ne-PoulencRorer,Anthony, France). Cell Isolation and Culture. Humanliver sampleswereobtainedin France Received 6/19/95: accepted 10/2/95. from eight patients undergoing liver resection for primary or secondary hep The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with atomas (Table 1). Hepatocytes were isolated by a two-step collagenase perfu 18 U.S.C. Section 1734 solely to indicate this fact. siom procedure, and the experimental procedures used were approved by the I This work was supported by the Institut National de Ia Sante et de Ia Recherche National Ethics Committee (18). Cell viability was 70â€”85%, as estimated by Medicale,the liguecontreIc cancercomitÃ©d'Illeet vilaine.the AssociationpourIa the trypan blue exclusion test. Cell yields varied with the size of the liver Recherche comtrele cancer. the BIOAVENIR program. the Cancer Research Campaign (United Kingdom). and USPHS Grants CA44353 and ES0O267. S. Langouet was a fragment, and the design of each experiment (time points and treatments) recipient of a fellowship from the MinistÃ¨re de Ia Recherche et de l'Espace and was depended on the number of viable cells obtained. granted a short-term visiting fellowship from the European Science Foundation to carry Liver parenchymal cells were seeded at a density of 106viable cells/35-cm2 out part of the work in the London group. dish in 2 ml of a nutrient medium consisting of 75% MEM and 25% Medium 2 To whom requests for reprints should be addressed.

3 Present address: National Center for Toxicological Research, Jefferson, AZ 199, supplemented with 10 @xgbovineinsulin/mI, 0.2% BSA, and 10% FCS. 72079-9502. This medium, supplemented with 0. 1 p.M dexamethasone but lacking serum, 4 The abbreviations used are: . aflatoxin : AFBO, aflatoxin 8,9 epoxide; was renewed daily. AFBSG. glutathiome conjugate of FBI oxide: CYP, cytochrome P450; GSH, glutathione; Inducers were added 36â€”48 h after cell seeding and then 24, 48, and 72 h GST, glutathiome S-transferase: MC, 3-methylcholanthrene; D3T, l,2-dithiole-3-thione; OPZ, oltipraz: RIF, rifampicim; HPLC, high-pressure liquid chromatography; EROD, later, with medium renewal. MC, D3T, OPZ, and RIF were dissolved in 7-ethoxyresorufin 0-deethylase. DMSO before addition to the culture medium to give final concentrations of 5 5574

Table I Characteristics of human liver samples and AFB1 biotransfonnation in corresponding prinror@ hepatocyte cultures mycotoxin.HumanI metabolism was quantified by HPLC in untreated cellsafter 2â€”5daysof culture and8 h of exposure Io the

liver of FB1 (%)HL-lsamplesPathologyGST MlDays cultureAFBSGExo/endo ratioAFM1@xM remainingMetabolism (69: M)â€•Carcinoid tumor of the small bowel+2 76.00HL-2 30.04 0.431.0 1.00.060 0.3601.17 1. 1776.00 (unknown)unknownNull2 3 â€” â€” 0.089 0.24 95.20 4 â€” â€” 0.094 1.37 72.44 67.95HL-3 5â€” â€”â€” â€”0.036 0.1941.70 1.6065.90 (69; M)Colon adenocarcimomaNull2 3 â€” â€” 0.064 1.55 69.00 4 â€” â€” 0.060 0.73 85.40 80.20HL-4 5â€” â€”â€” â€”0.250 0.1253.17 0.9936.60 (59; M)Colon ademocarcimomaNull2 3 â€” â€” 0.115 1.68 66.40 87.80HL-5 4â€” â€”â€” â€”0.065 0.0474.50 0.6110.00 metastasis+30.0171.00.0751.7864.50HL-6(68; F)Hepatic (I;M)None+30.021.00.1002.9241.60HL-7 areasNull2â€”â€”0.0123.9920.20HL-8(48; F)Necrotic (49: M)Colon adenocarcimomaNull2 3 â€” â€” 0.047 4.07 18.60 4â€” â€”â€” â€”0.040 0.0334.66 3.227.00 35.60 â€”:not detected a (Age in years; sex).

@.LMforMC and 50 ,xM for D3T, OPZ, and RIF (in 0.2% DMSO, v/v). Control method of Prough er a!. (26). Samples of yeast microsomes containing 50 pmol cultures received the same concentration of solvent. of CYP1A2 were incubated at 28Â°Cwith variable concentrations of After different times of treatment with the inducers, the cells were exposed ethoxyresorufin (0.2â€”1.6 @xM)added in 0.1 M potassium phosphate buffer. The to 5 p.M for 8 h. At the end of the incubation with the mycotoxin, both reaction was initiated by the addition of I25 MMNADPH. Fluorescence was media and cells were collected and stored at â€”80Â°Cuntil use. measured directly in the cuvettes after stirring for 2 mm. Calibration was made Analysis of AFB1 Metabolites. FB1 metabolites were separated by by addition of 53 pmol of resorufin to the reaction medium. HPLC by direct injection of cell culture medium or the hepatocyte-soluble The 6f3-hydroxylationof testosterone catalyzed by CYP3A4 was tested by supernatant fraction onto a Waters p@-bondapak 8 reverse-phase column with incubating microsomal samples from the yeast expression system (containing elution by a linear gradient of 10â€”75% methanol in water buffered with 10 mM 50 pmol of enzyme) in the presence of a NADPH-generatingsystem for 30 mm ammonium acetate (pH 6.5). To separate AFM1 from OPZ metabolite(s) that at 28Â°C.Testosterone was added at various concentrations (25â€”200@xM). eluted at the same time, we used the same gradient conditions with a Alltech Extraction was done with 2 volumes of ethyl acetate after addition of 10 xl of C18 column, [email protected] were identified and quantitated by 50% trifluoroaceticacid.6f3-Hydroxytestosteronewasquantitatedby HPLCon reference to authentic standards of known concentration using fluorescence a 5-@xm particle Nucleosil C18 column (150 X 4.6 mm). The mobile phase, at excitation with a narrow-pass filter (370 mm) and a wide-pass emission filter a flow rate of 1 mI/mm, consisted of water containing 0. 1% (v/v) glacial acetic (430 nm) with a low wavelength cutoff. Typically, 50â€”100 @xlofsolution were acid/acetonitrile (1:1) for 20 mm, followed by a linear gradient increasing to analyzed; the injected volume was adjusted to keep peak height within the 0.1% glacial acetic acid in water/acetonitrile (7:3) in S mm. The reaction was fluorescence calibration window. To identify some metabolites, HPLC was followed spectrophotometrically at 254 mm. Calibration curves were performed also run at pH 4.5 using an Econosphere C18 reverse-phase column. This last by HPLC analysis of increasing amounts of 6j3-hydroxytestosterone. column was also used to separate the two stereoisomeric GSH conjugates, In both experiments, OPZ (in 0.1% DMSO, v/v) was added to the assay essentially as described by Raney et a!. (3). mixture at selected concentrations in the range 0. 1â€”[email protected] incuba GST Subunit Analysis. The cells were sonicated, and a total GST fraction tions contained the same amount of DMSO (0.1 %, vlv). was obtained from the supernatant by GSH agarose affinity chromatography and analyzed by reverse-phase HPLC using a 5-@xm particle size column (Rainin Instrument Co., Inc., Woburn, MA) and a flow rate of I ml/min (19). RESULTS GST subunits were quantified from the absorbance at 214 mm. Cellular protein AFB1 Metabolism by Untreated Hepatocytes. The use of fluo content was estimated by the Bradford procedure (20). rescence analysis allowed determination of FB1, unconjugated CYP1A2 and CYP3A4 Activities in Human Hepatocytes. Activities were measured in cultured living hepatocytes immediately after treatment. AFM1, and AFP1, together with exo and endo AFBSG, at less than 1 EROD activity was estimated essentially as described by Burke and Mayer flM concentration in the presence of the many UV-absorbing constit (2 1); reaction rates were determined under linear conditions with regard to uents of the cell culture medium. AFP1 levels were generally below incubation time and protein concentration. Oxidation of mifedipine to its the level of quantitation, and AFQ1 was not sufficiently fluorescent to pyridine oxidation product, a reaction catalyzed mainly by CYP3A4 (22), was be detected at this low concentration. Several other fluorescent species measured in cultures refreshed with MEM without phenol red and containing were detected after long incubation times but were not identified as nifedipine at a concentration of 20 @xM.The pyridine was measured in the known AFBJ metabolites and could not, therefore, be quantitated. supernatant by HPLC according to Guengench et a!. (22). Eight hepatocyte populations were investigated, of which three CYP1A2 and CYP3A4 Activities in Recombinant Yeasts. CYP1A2 and were found to be GST Ml positive and five GST Ml negative (Table CYP3A4 cDNAs were expressed in yeasts as described by Renaud et a!. (23) 1). The cells were exposed to S p@M FB1 for 8 h after 3, 4, 5, or 6 days and Truan et a!. (24). Microsomal proteins were prepared, and protein contents were determined using the Bradford method (20). Total CYP concentrations of culture. Large quantitative variations in overall FB1 metabolism were determined by Fe2@-CO versus Fe2' difference spectrophotometry as and the metabolites formed were observed depending on the cell described by Omura and Sato (25) using a molecular extinction coefficient of population investigated and the time of culture (Table 1). The con 91 mMâ€•cm@. centration of AFBSG varied from 0.017 to 0.43 p.M.AFM, from 0.012 EROD activity was determined by spectrofluorimetry according to the to 0.36 p.M. and FB1 from 0.93 to 4.7 @.tM. 5575

500 483 â€¢HL-1 @HL-2 450 â€¢HL-3

400 @HL-4 â€¢HL-5 Fig. 1. Effect of OPZ on AFB, biotransforma 350 â€¢HL-6 tion and on formation of AFBSG and AFM1 in primary cultures of human hepatocytes derived :@300 @HL-7 from eight donors after 48 h of treatment. AFBSG, 0 â€¢HL-8 @ AFM1, and unmetabolized AFB@ levels in OPZ U 250 225 @@ treated cultures were calculated as percentages of 218 the values measured in control cultures. Control @â€˜200 values for AFB1epoxidation, AFM1formation, and AFB1 level in HL- I to HL-8 were given in Table I. 150 *, <5% of the control value. 11 100

50 29.4 13.3 @ 16.3@ â€¢ a * 21 @12.8 @ 0 Ilifil @t AFM1 unmetabollzed AFB1

Table 2 lnduction of the GST subunits by OPZ in human hepatocytes in primary lations used in the present study after a 48-h exposure to OPZ. As culture after 48 h of treatment shown in Table 2, OPZ increased GST Al subunit 2.0-, 1.8-, and Human hepatocytes were maintained in the absence (C) or presence of OPZ for 48 h 3.8-fold; GST A2 by 4.0-, 1.9-, and 2.1- fold; and GST Ml by 1.6-, as described in â€œMaterialandMethods.â€•Thesoluble fractions of cellular homogenates were obtained, and the GST subunits were quantitated by glutathione affmity chroma 1.4-, and 2.0-fold in liver samples HL-l, HL-5, and HL-6, respec tography, followed by reversed-phase HPLC. The fold of induction is indicated in parentheses.

GST subunits (pg/mg cytosolic protein) AFBSG A SampleTreatmentMIaAlA2HL-lC 2 300 2.40(4.00)HL-5COpZ0.38 0.80 (2.10)1.80 3.90 (2.17)0.52

4.20(1.90)HL-6COPZ0.84 1.38 ( 1.64)3.50 6.20 (1.77)2.20

OPZ0.23 0.33 (1.43)1.70 6.50 (3.82)0.70 1.50(2.14)

MC D3T OPZ RIF MC+OPZ RIF+OPZ The number of time points examined was determined by the size of the liver sample, smaller samples limiting the number. A 3-day culture period appeared to give a good representation of metabolism, and this was chosen for comparative studies. AFM1 B Effect of GST Ml Polymorphism. Three metabolites were iden @ rifled in the GST Ml-positive liver cells (i.e., AFM@ and the GSH 400 conjugates of endo and exo AFBO); these two isomers were produced S 300 @ in the same proportion (Table 1). In contrast, only AFM@was formed 200 in GST Ml-negative hepatocyte cultures. @ 100 Effect of OPZ on AFB1 Metabolism and GST Expression. In all cases, OPZ-treated hepatocytes had lower AFB1 metabolism than Mc D3T oPz RW MC+OPZ RIF+OPZ their untreated counterparts after either 3, 4, or S days of culture. Less AFB1 was metabolized, and less AFBSG and AFM@ were produced after 48 h of treatment. In the three GST Mi-positive samples, AFBSG was negligible in one case and reduced by 71 and 84% in the unmetabolizedAFB1 C others. The production of AFM1 was negligible in 4 cases and decreased by 82â€”97%in the others (Fig. 1). A similar decrease in @350 - @ AFB1 metabolism was observed when hepatocytes were exposed to 300 the parent D3T. To determine whether the decrease in AFB@biotrans 8 250 formation observed in the culture medium was not due to the retention @200 -. râ€”'--t . @@ of metabolites or unmetabolized AFB@in the hepatocytes, cell homo @,150 I i-@-@i

@ genates were also assayed. Only negligible amounts of AFB@and its @1:@ r:i H iii @: metabolites were detectable in either untreated or OPZ-exposed hepa tocytes. Neither FB1 nor its metabolites were adsorbed into the MC D3T OPZ RIF MC+OPZ RIF+OPZ plastic culture dishes used. Fig. 2. Effect of MC, D3T, OPZ, and RIF on the formation of AFBSG and AFM1and We have already shown that OPZ increased human GST Al, A2, AFB1 metabolism in primary cultures of human hepatocytes derived from eight donors. The amounts of unmetabolized AFB@, AFBSG, and AFM@ formed in the presence of and to a lesser extent, GST Ml transcripts (16). This was reevaluated modulators were calculated as a percentage of the values found under control conditions at the protein level for the three GST Mi-positive hepatocyte popu and are the mean Â±SD (bars) of 3 (A) or 8 (B and C) experiments. 5576

OLTIPRAZ EFFECFS ON AFB1 METABOLISM A hadagreateraffinityforandwasconsequentlyamorepotentcom EROD petitive inhibitor of CYPIA2 in comparison with CYP3A4. Preincubation of OPZ with microsomes in presence of a NADPH 500 generating system for S mm before adding the substrate in the same @.450 2 400 conditions as above enhanced inhibition. The percentage of inhibition g350 was 71 Â±6%, with I.6 @Methoxyresorufin and 10 p.MOPZ without @ 300 preincubation, whereas it reached 85 Â±7% after a 5-mm preincuba 0 250 @5.200 tion of the microsomes with OPZ (results not shown). In the presence 0 of 50 pmol expressed CYP3A4 microsomes/ml, 100 p@ testosterone, 0 @ 100 and 100 p.Moltipraz, the percentage of inhibition varied from 6 Â±1% 50 without preincubation to 47 Â±6% after a preincubation with OPZ. 0 MC D3T OPZ RU' MC+OPZ RIF+OPZ These results support the conclusion that OPZ is not only a compet itive inhibitor of CYP1A2 and CYP3A4 substrates but also acts partly as an irreversible inhibitor.

NIFEDIPINEOXIDATION B DISCUSSION

700 By using human hepatocytes in primary culture, a model that mimics the in vivo human liver, we show here that endo and exo .@ 600 AFBSGs are formed on exposure to FB1. However OPZ, although :E@ an inducer of GSTs, markedly decreased AFBSG formation and FB1 00 @: metabolism in general. This was shown to be the result of inhibition of CYP1A2 and CYP3A4, two major CYPs involved in the biotrans formation of FB1 in humans. I@::: When compared with other species, the activity of GSTs toward 100 AFBO in human liver cytosols is apparently so low that some other

0 â€” â€” â€” â€” â€” â€” workers have not been able to detect it (27). For the first time, we MC D3T OPZ IIRU' MC+OPZ RIF+OPZ show that human hepatocytes are able to produce both I exo - Fig. 3. EROD (A ) and mifedipime oxidation (B) activities in primary cultures of human and endo-oxide GSH conjugates in appreciable quantity. Values hepatocytes after exposure to OPZ. MC. and RIF or to OPZ plus MC or RIF. Columns, mean percentages of the values found in controls of 4 experiments; bars, SD. published for the specific activity of purified human hepatic GSTs in the conjugation of exo AFBO are GST Alâ€”l, 0.04 nmol

tively. In cultures from these three livers, GST Mlâ€”l, the most A effective enzyme with exo AFBO ( 14), was induced the least by OPZ. Effect of MC and RIF on AFB1 Metabolism in the Presence or @. Absence of OPZ. The decreased metabolism of I in OPZ-treated 0 5o@M culture suggested that OPZ had a direct effect on biotransformation by CYPIA2 and CYP3A4. To address this hypothesis, hepatocyte cul A 75@tM tures were exposed to MC and RIF, specific inducers ofCYP1A2 and CYP3A4, respectively. As shown in Fig. 2, MC treatment resulted in 0 i00@M @ a marked increase in both metabolism of FB1 and formation of 200@iM AFBSG and FM1 . RIF also increased AFBSG levels but did not _____ affect FM1 formation, and had only a moderate positive effect on ______overall AFB1 metabolism. When coincubated with MC, OPZ partly -200 .160 .120 -80 -40 0 40 80 120 suppressed formation of the two metabolites, whereas coincubation [OPZ], @sM with RIF caused almost complete inhibition of increase of FB1 B metabolism seen with RIF alone. 0 o.v@sM Effect of Exposure to OPZ on Cellular Activities of CYP1A2 80@ and CYP3A4. OPZ given to cells in primary culture inhibited EROD A 0.53@sM and nifedipine oxidation, two activities catalyzed by CYP1A2 and 0 1.6pM CYP3A4, respectively (Fig. 3), whereas 4-fold inductions of both EROD by MC and nifedipine oxidation by RIF were observed after 2 days. Simultaneous incubation with OPZ resulted in a marked decrease in both CYP-related enzyme activities. Effect of OPZ on the Enzyme Kinetics on Heterologously Ex -______

@ pressed CYP1A2 and CYP3A4. CYP1 A2 and CYP3A4 were ex -20 .10 1'O 20 30 40 50 60 pressed in yeasts and microsomal preparations from the recombinants [OPZ], aM used to study the inhibition kinetics of OPZ. Dixon plots were constructed that showed that during the 30-mm incubation period, Fig. 4. Dixon-plot analyses of 6(3-hydroxylation of testosterone (A ) and EROD (B) catalyzed by CYP3A4 and CYP1A2, respectively, expressed in yeasts in the presence of OPz behaved as a competitive inhibitor of both enzymes (Fig. 4). variable OPZ concentrations. The reciprocal of 6(3-hydroxylation of testosterone (A ) and Respective K1s,calculated from the reciprocal plots of velocity versus EROD (B) is plotted against OPZ concentrations for incubations containing testosterone at 50, 75, 100, and 200 @.rMandethoxyresorufin at 0.27, 0.53, and 1.6 saM,respectively. OPz concentration, were 10 and 80 p.Mfor CYPIA2 and CYP3A4, The approximate K for OPZ-induced inhibition in this experiments was 80 ,SM for respectively. The 8-fold lower Ks for CYPIA2 indicates that OW CYP3A4 and 10 .LMforCYPIA2. 5577

@ min I mg I GST A2â€”2:0.02 nmol min I mg I and and GST of AFM@and AFBSG in our cells, whereas by inducing CYP3A4, RIF Mlâ€”!, 0.6 nmol min â€˜mg @,whereas for the endo AFBO, the enhanced formation of AFBSG. values are nil for both GST Alâ€”i and GST A2â€”2,but 1.4 nmol The inhibitory effect of OW on both CYP1A2 and CYP3A4 was min â€˜mg â€˜forGST Mlâ€”l (14). GST Miâ€”i is more active than confirmed by two other approaches: (a) in hepatocytes, EROD and the others by at least 1 order of magnitude. Our results fit with this nifedipine oxidation, which are catalyzed by CYP1A2 and CYP3A4, difference because appreciable AFBSG formation occurred only in respectively, were decreased in OPZ-treated cells; and (b) experi hepatocytes expressing GST M 1. Because human hepatocytes in ments with recombinant enzymes expressed in yeasts show that this primary culture formed both endo and exo AFBSG, and the exo inhibition is partly due to a competition between the I and OW oxide has been shown to react DNA very efficiently and cause for CYP1A2 and CYP3A4, and also partly to an irreversible inhibition mutation (28), the lack of GST Ml activity in GST Mi-null by OW of these two CYPs. There may, however, be another element individuals may be of considerable consequence. Our observations in inhibition observed in hepatocytes, particularly with respect to also support and explain the study of Liu et a!. (29), who reported CYP3A4, because OW reduced its activity to very low levels whether that human liver cytosol obtained from GST Ml-deficient individ the substrate was AFB1 or nifedipine, and much lower than would be uals (GST Ml null) was less protective against AFB1-induced expected for a K1 as high as 80 ELM.The possibility that the hepato mutagenicity in the Ames Salmonella assay than liver cytosol from cytes metabolize OW to a more potent inhibitor is under study. These GST Ml-positive individuals. The major role played by OST Mlâ€”l effects have also been observed with D3T, a natural compound, which in conjugation of AFBO with GSH in humans and by GST 10â€”10 is present in extracts of cruciferous vegetables, indicating that various in the rat is another example of a major discrepancy in behavior dithiolethiones can have the same inhibitory effect. between laboratory animals and humans. No equivalent of the rat To our knowledge, this is the first demonstration of an inhibitory GST 10â€”10is induced de novo in human liver. effect of OW on human CYP activities. Such an effect had been Our study has focused on quantitation of AFBO conjugates and evoked by PuU et al. (34) when studying metabolism of FBJ in AFM1 , but these metabolites represent only about 10% of the total OW-treated rats. It could be concluded that any shortcomings that AFB1 biotransformation. Most of the metabolites appeared to be might arise from low activities of GSTs or, even in GST Ml-null excreted because after an 8-h incubation, no intracellular accumula individuals, the absence of any GST-dependent detoxication could be compensated for in terms of chemoprotection by inhibition of activa tion of I or FB1 metabolites could be demonstrated. Further tion by Phase I enzymes. It must be borne in mind, however, that studies are directed toward identification of other metabolites pro without Phase I enzymes, FB1 cannot undergo appreciable detoxi duced by human hepatocytes, including AFQ1. This metabolite, cation by CYP, although reduction to aflatoxicol followed by glucu formed by 3a-hydroxylation of FBJ by CYP3A4, is regarded as a ronidation is an alternative mechanism. detoxication product and is known to have low acute toxic and carcinogenic effects, if any (30). In recent years, considerable attention has been given to the devel REFERENCES opment of chemointervention strategies involving induction of GSTs, with activity toward exo AFBO and a possible decrease in the extent 1. Garner, R. C., Miller, E. C., and Miller, J. A. Liver microsomal metabolism of @ aflatoxin to a reactive derivative toxic to Salmonella tvphimurium TAI53O. Cancer of DNA damage in populations with high aflatoxin exposure. OW has Res., 32: 2058 â€”2062,1972. been proposed as a good candidate because of its protective effect 2. Busby, W. F., and Wogan, G. N. Aflatoxins. In: C. E. Searle (ed), Chemical Carcinogemesis, Vol. 2, Ed. 2, pp. 945â€”1093.Washington, DC: American Cancer against AFB 1-induced hepatocarcinogenesis in the rat (17). OW is Society Monographs, 1984. known to be an inducer of GSTs and of some other enzymes, such as 3. Raney, K. D., Coles, B., Guengerich, F. P., and Harris, T. H. The endo-8,9-epoxide quinone reductase (3 1). In a previous study, we demonstrated that this of aflatoxin B,: a new metabolite. Chem. Res. Toxicol., 5: 333â€”335,1992. 4. O'Brien, K., Moss, E., Judah, D., and Neal, G. Metabolic basis of species difference compound also induces GSTs in human hepatocytes in primary cul to aflatoxin B, induced hepatotoxicity. Biochem. Biophys. Res. Commun., 114: ture at the mRNA level. We confirm these observations with other 813â€”821,1983. 5. Monroe, D. H., and Eaton, D. L. Comparative effects of butylated hydroxyanisole on human cell populations at the protein level and the higher response of hepatic in vivo DNA binding and in vitro biotransformation of aflatoxin , in the rat GST-a compared to [email protected]. OPZ increased GST Al by 4-fold andmouse.Toxicol.Appl.Pharmacol.,90:401â€”409,1987. and GST A2 and GST Ml by 2-fold. This induction effect was not 6. Ramsdell, H. S., and Eaton, D. L. Mouse liver glutathiomeS-transferase isoenzyme activity toward aflatoxim B,-8,9-epoxide and benzo(a)pyrene-7,8-dihydrodiol-9.lO associated with an increased formation of AFBSG; this was overriden epoxide. Toxicol. AppI. Pharmacol., 105: 216â€”225,1990. by an inhibitory effect of OW on FB1 metabolism by reduction of 7. Hayes, J. D., Judah, D. J., McLellan, L. I., and Neal, G. E. Contribution of the glutathione S-transferases to the mechanisms of resistance to aflatoxin B ,. Pharmacol. CYP1A2 and CYP3A4 activities. It is not a general effect on CYPs Ther., 50: 443â€”472,1991. because some other CYPs, such as CYP2C8, were not affected by the 8. Kensler, T. W., Egnert, P. A., Trush, M. A., Bueding, E., and Groopman, J. D. antioxidant (data not shown). In addition, this inhibitory effect is not Modification of aflatoxin , binding to DNA in vivo in rats fed phenolic antioxidants, ethoxyquin and a dithiolethione. Carcimogenesis(Lond.), 6: 759â€”763,1985. species specific because similar CYP inhibition by OW occurred in 9. Kensler, T. W., Groopman, J. D., Eaton, D. L., Curphey, T. J., and Roebuck, B. D. rat hepatocytes.5 It should be emphasized that OW was used at a Potent inhibition of aflatoxim-induced hepatic tumorigenesis by the momofunc tional enzyme inducer I,2-dithiol-3-thiome. Carcinogenesis (Lond.), 13: 95â€”100, concentration comparable to that in plasma from humans treated with 1992. a single dose of 500 mg of OW (32), therefore validating the extrap 10. Hayes, J. D., Judah, D. J., McLelIan,L. I., Kerr, L. A., Peacock, S. D., and Neal, G. E. olation of these in vitro data to the in vivo situation. Ethoxyquin-induced resistance to aflatoxin B , in the rat is associated with the expression of a novel alpha class glutathione S-transferase subunit Yc2. which In GST MI-positive cells, AFBSG production, which depends on possesses high catalytic activity for aflatoxim B,-8,9-epoxide. Biochem. J., 279: both CYP1A2 and CYP3A4, was reduced by as much as 84% after 385â€”398,1991. 11. Meyer, D., Harris, J. M., Gilmore, K. S., Coles, B., Kensler, 1. W., and Ketterer, B. treatment with OW. AFM1 production, which depends solely on Quantitation of tissue- and sex-specific induction of rat glutathione transferase sub CYPIA2, was decreased by as much as 97%. The use of specific units by dietary l,2-dithiole-3-thiomes. Carcinogenesis (Land.), 14: 567â€”572,1993. inducers for these two CYPs confirmed previous results, showing the 12. Evaluation of Carcinogenic Risk to Humans. IARC Monogr. Suppl., Vol. 7, pp. 83â€”87.Lyon, France: IARC Press, 1987. implication of CYPIA2 and CYP3A4 in the biotransformation of 13. Qian, G. S., Ross, R. K., Yu, M. C., Yaun, J. M., Gao, Y. T., Henderson, B. E., AFB1 (33). Thus, by inducing CYP1A2, MC increased the formation Wogan, G. N., and Groopman, J. D. A follow-up study of urinary markers of aflatoxin exposure and liver cancer risk in Shanghai, People's Republic of China. Cancer Epidemol., Biomarkers & Prey., 3: 3â€”10,1994.

S Langouet et al., manuscript in preparation. 14. Raney, K. D., Meyer, D. J., Ketterer, B., Harris, T. M., and Guengerich, F. P. 5578

Glutathione conjugation of aflatoxin , exo- and endo- epoxide by rat and human 24. Truan, G., Cullin, C., Reisdorf, P., Urban, P., and Pompon, D. Enhanced in vivo glutathiome S-transferases. Chem. Res. Toxicol., 5: 470â€”478, 1992. monooxygenase activities of mammalian P-45Os in engeneered yeast cells producing 15. Board, P. G. Biochemical genetics of glutathione S-transferase in man. Am. Hum. high level of NADPH P450 reductase and cytb5. Gene (Amst.), 125: 49â€”55,1993. Gem., 33: 36â€”43, 1981. 25. Omura, T., and Sato, E. Carbon monoxide binding pigment of liver microsomes. 16. Morel, F., Fardel, 0., Meyer, D. J., Langouet, S., Gilmore, K. S., Tu, C. P. D., J. Biol. Chem., 239: 2370â€”2379,1964. Kensler, T. W., Ketterer, B., and Guillouzo, A. Preferential increase of glutathione 26. Prough, R. A., Burke, M. D., and Mayer, R. T. Direct fluorimetric methods for S-transferase class a transcripts in cultured human hepatocytes by phenobarbital, measuring mixed-function oxidases. In: S. Fleischer and L. Packer (eds.), Methods in 3-methylcholanthrene, and dithiolethiones. Cancer Res., 53: 231â€”234,1993. Enzymology, Vol. 52, pp. 372â€”377.London: Academic Press, Ltd., 1978. 17. Kensler, T. W., Egner, P. A., Dolan, P. M., Groopman, J. D., and Roebuck, B. D. 27. Moss, E. J., and Neal, G. E. The metabolism of aflatoxin , by human liver. Biochem. Mechanism of protection against aflatoxin tumorigenicity in rats fed 5-(2-pyrazimyl)- Pharmacol., 34: 3193â€”3197, 1985. 4-methyl-1,2-dithiol-3-thione (oltipraz) and related l,2-dithiol-3-thiones and 1,2- 28. Iyer, R., Coles, B., Raney, K. D., Thier, R., Guengerich, F. P., and Harris, T. M. DNA dithiol-3-ones. Cancer Res., 47: 4271â€”4277, 1987. adduction by the potent carcinogen aflatoxin , : mechanistic studies. J. Am. Chem. 18. Guguen-Guillouzo, C., Campion, J. P., Brissot, P., Glaise, D., Bourel, M., and Soc.,116:1603â€”1609,1994. Guillouzo, A. High yield preparation of isolated human adult hepatocytes by enzy 29. Liu, Y. H., Taylor, J., Linko, P., Lucier, G. W., and Thompson, C. L. Glutathione matic perfusion of the liver. Cell Biol. Int. Rep., 6: 625â€”628, 1982. S-transferase @.Linhuman lymphocytes and liver: role in modulating formation of 19. Kispert, A., Meyer, D. J., Lalor, E., Coles, B., and Ketterer, B. Purification and carcinogen-derived DNA adducts. Carcinogenesis (Land.), 12: 2269â€”2275, 1991. characterization of a labile rat glutathione transferase of the @aclass.Biochem. J., 260: 789â€”793, 1989. 30. Raney, K. D., Shimada, T., Kim, D. H., Groopman, J. D., Harris, T. M., and 20. Bradford, M. A rapid and sensitive method for the quantitation of microgram Guengerich, F. P. Oxidation of aflatoxins and sterigmatocystin by human liver quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem., microsomes: significance of aflatoxin Q1 as a detoxication product of a aflatoxin B,. 72: 248â€”254, 1976. Chem.Res.Toxicol.,5:202â€”210,1992. 21 . Burke, M. D., and Mayer, R. T. Differential effects of phenobarbitone and 3-meth 31. Ansher, S. S., Dolan, P., and Bueding, E. Biochemical effects of dithiolethiomes. Food ylcholanthrene induction on the hepatic microsomal metabolism and cytochrome Chem. Toxicol., 24: 405â€”415,1986. P-450-binding of phenoxazone and a homologous series of its N-alkylethers 32. Berezin, F., Ratain, M. J., and Benson, A. B., HI. Pharmacokinetic study of 4-methyl (alkoxyresorufins). Chem. Biol. Interact., 45: 243â€”248,1983. 5(2-pyrazinyl)-l,2-dithiole-3-thione (oltipraz) in normal volunteers. Proc. Am. Assoc. 22. Guengerich, F. P., Martin, M. V., Beaune, P. H., Kremers, P., Wolf, T., and Waxman, Cancer Res., 33: 1244, 1992. D. J. Characterization of rat and human liver microsomal cytochrome P-450 forms 33. Ueng, Y. F., Shimada, T., Yamazaki, H., and Guengerich, F. P. Oxidation of aflatoxin involved in nifedipine oxidation, a prototype for genetic polymorphism in oxidative I by bacterial recombinant human cytochrome P450 enzymes. Chem. Res. Toxicol., drug metabolism. J. Biol. Chem., 261: 5051â€”5060,1986. 8: 219â€”224,1995. 23. Renaud, J. P., Cullin, C., Pompon, D., Beaune, P., and Mansuy, D. Expression of 34. Putt, D. A., Kensler, T. W., and Hollenberg, P. F. Effects of three chemoprotective human liver P-450 IIIA4 in yeast. A functional model for the hepatic enzyme. Eur. J. antioxidants, ethoxyquin, oltipraz and l,2-dithiole-3-thiones, on cytochrome P-450 Biochem., 194: 889â€”896, 1990. levels and aflatoxin B, metabolism. FASEB J., 5: A1517, 1991.

5579

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. Inhibition of CYP1A2 and CYP3A4 by Oltipraz Results in Reduction of Aflatoxin B 1 Metabolism in Human Hepatocytes in Primary Culture

Sophie Langouët, Brian Coles, Fabrice Morel, et al.

Cancer Res 1995;55:5574-5579.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/55/23/5574

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/55/23/5574. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.