Chemoprevention of Aflatoxin B1 ... - Cancer Research

[CANCER RESEARCH 60, 957–969, February 15, 2000]

Chemoprevention of Aflatoxin B1 Hepatocarcinogenesis by Coumarin, a Natural Benzopyrone That Is a Potent Inducer of Aflatoxin B1-Aldehyde Reductase, the Glutathione S-Transferase A5 and P1 Subunits, and NAD(P)H:Quinone Oxidoreductase in Rat Liver1 Vincent P. Kelly,2 Elizabeth M. Ellis,2,3 Margaret M. Manson, Simon A. Chanas, Graeme J. Moffat, Ronald McLeod, David J. Judah, Gordon E. Neal, and John D. Hayes4 Biomedical Research Centre, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, Scotland [V. P. K., E. M. E., S. A. C., R. M., J. D. H.]; Medical Research Council Toxicology Unit, Hodgkin Building, University of Leicester, Leicester LE1 9HN, England [M M. M., D. J. J., G. E. N.]; Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, England [G. J. M.], United Kingdom

ABSTRACT

INTRODUCTION

Structurally diverse compounds can confer resistance to aflatoxin B1 (AFB1) hepatocarcinogenesis in the rat. Treatment with either phytochemicals [benzyl isothiocyanate, coumarin (CMRN), or indole-3-carbinol] or synthetic antioxidants and other drugs (butylated hydroxyanisole, diethyl maleate, ethoxyquin, !-naphthoflavone, oltipraz, phenobarbital, or trans-stilbene oxide) has been found to increase hepatic aldo-keto reductase activity toward AFB1-dialdehyde and glutathione S-transferase (GST) activity toward AFB1-8,9-epoxide in both male and female rats. Under the conditions used, the natural benzopyrone CMRN was a major inducer of the AFB1 aldehyde reductase (AFAR) and the aflatoxin-conjugating class-" GST A5 subunit in rat liver, causing elevations of between 25- and 35-fold in hepatic levels of these proteins. Induction was not limited to AFAR and GSTA5: treatment with CMRN caused similar increases in the amount of the class-# GST P1 subunit and NAD(P)H: quinone oxidoreductase in rat liver. Immunohistochemistry demonstrated that the overexpression of AFAR, GSTA5, GSTP1, and NAD(P)H:quinone oxidoreductase affected by CMRN is restricted to the centrilobular (periacinar) zone of the lobule, sometimes extending almost as far as the portal tract. This pattern of induction was also observed with ethoxyquin, oltipraz, and trans-stilbene oxide. By contrast, induction of these proteins by !-naphthoflavone and diethyl maleate was predominantly periportal. Northern blotting showed that induction of these phase II drug-metabolizing enzymes by CMRN was accompanied by similar increases in the levels of their mRNAs. To assess the biological significance of enzyme induction by dietary CMRN, two intervention studies were performed in which the ability of the benzopyrone to inhibit either AFB1-initiated preneoplastic nodules (at 13 weeks) or AFB1-initiated liver tumors (at 50 weeks) was investigated. Animals pretreated with CMRN for 2 weeks prior to administration of AFB1, and with continued treatment during exposure to the carcinogen for a further 11 weeks, were protected completely from development of hepatic preneoplastic lesions by 13 weeks. In the longer-term dietary intervention, treatment with CMRN before and during exposure to AFB1 for a total of 24 weeks was found to significantly inhibit the number and size of tumors that subsequently developed by 50 weeks. These data suggest that consumption of a CMRN-containing diet provides substantial protection against the initiation of AFB1 hepatocarcinogenesis in the rat.

AFB15 is a potent hepatocarcinogen produced by Aspergillus flavus, a mold that frequently contaminates rice and cereal crops in humid areas of the world (1). In combination with hepatitis B, AFB1 is thought to be largely responsible for the high incidence of hepatocellular carcinoma in southeast China and southern Africa (2). Like most chemical carcinogens, the mycotoxin requires bioactivation to exert its carcinogenic effects. The ultimate carcinogen of AFB1 is the exo-8,9-epoxide, and once generated by the actions of CYP (3, 4), it readily forms adducts with DNA (1). In humans, hepatocellular carcinoma resulting from exposure to AFB1 is associated with mutations in codon 249 of the p53 tumor suppressor gene (5), whereas in the rat it is associated with mutations in codons 12 and 13 of ras oncogenes (6). Although primates and rats are sensitive to AFB1, the mouse can tolerate high levels of the mycotoxin without showing signs of acute liver damage or of developing liver cancer. (See Refs. 7–10 for further details about the selective toxicity of AFB1.) Investigations into the metabolic basis for the natural variation in sensitivity of laboratory animals to the mycotoxin suggest that it is the high basal expression of GST, which inactivates the exo-8,9-epoxide, rather than either high levels of epoxide hydrolase, which might detoxify epoxidated AFB1, or low levels of CYP, which activate AFB1, that plays a pivotal role in protection against carcinogenesis (7). Specifically, liver cytosols prepared from adult male mice normally possess !50-fold higher levels of GST activity toward epoxidated AFB1 than do hepatic cytosols from male rats or humans (7, 11). Depletion of GSH can result in a 25-fold increase in the sensitivity of murine liver to form AFB1-DNA adducts (8). The enhanced capacity of mouse liver to detoxify AFB1-8,9-epoxide is attributable to a class-! transferase (9, 10); for reviews of GSH and GST in drug resistance, see (12, 13). Consistent with the hypothesis that the level of GST influences sensitivity to carcinogenesis, Townsend et al. (14) have reported that stable transfection of the murine class-! transferase A3 subunit [also called Yc (9, 10)] into a hamster V79 cell line confers a 5-fold increase in resistance to the cytotoxic effects of AFB1 and a 3.3-fold reduction in the amount of DNA adducts formed following exposure to the mycotoxin.

Received 8/23/99; accepted 12/16/99. 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 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by Grant G9322073PA from the Medical Research Council and contract FS1709 from Ministry of Agriculture, Fisheries and Food. E. M. E. is a Beit Memorial Research Fellow. S. A. C. holds a CASE Medical Research Council Ph.D. studentship. 2 These authors contributed equally to this work. 3 Present address: Department of Bioscience and Biotechnology and Pharmaceutical Sciences, University of Strathclyde, Glasgow G1 1XW, Scotland, United Kingdom. 4 To whom requests for reprints should be addressed, at the Biomedical Research Centre, Ninewells Hospital and Medical School, Level 5, University of Dundee, Dundee DD1 9SY, Scotland, United Kingdom. E-mail: [email protected].

5 The abbreviations used are: AFB1, aflatoxin B1; CYP, cytochrome P450; GST, glutathione S-transferase; GSH, reduced glutathione; AKR, aldo-keto reductase; AFAR, aflatoxin B1 aldehyde reductase; OPZ, (oltipraz) 4-methyl-5-pyrazinyl-3H-1,2-dithiole-3thione; BHA, butylated hydroxyanisole; EQ, (ethoxyquin) 6-ethoxy-1,2-dihydro-2,2,4trimethylquinoline; PB, phenobarbital; "-NF, "-naphthoflavone; I3C, indole-3-carbinol; CMRN, coumarin; BITC, benzyl isothiocyanate; NQO, NAD(P)H:quinone oxidoreductase; ARE, antioxidant responsive element; XRE, xenobiotic responsive element; DEM, diethyl maleate; t-SO, trans-stilbene oxide; CDNB, 1-chloro-2,4-dinitrobenzene; 2-CBA, 2-carboxybenzaldehyde; HPLC, high-pressure liquid chromatography; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PAH, polycyclic aromatic hydrocarbon; DMBA, dimethylbenz(a)anthracene; GGT, #-glutamyl transpeptidase; GPEI, glutathione S-transferase P enhancer I; t-BHQ, tert-butylhydroquinone.

957

INDUCTION OF AFAR, GST, AND NQO BY CMRN

control because it is inducible by a wide range of chemopreventive agents (34, 35). Significantly, the promoter of the rat NQO1 gene contains both an ARE and an XRE enhancer (36) that together are responsive to metabolizable antioxidants, Michael reaction acceptors, and planar aromatic compounds (12, 35, 36). To establish whether the pattern of gene expression increased by these phytochemicals is distinct from that achieved by synthetic drugs, their effects on the levels of AFB1-detoxication enzymes in rat liver was compared with the effects of BHA, EQ, and OPZ, as well as with model inducers of detoxication enzymes.

In addition to GST-catalyzed conjugation of AFB1 with GSH, other detoxification mechanisms exist that probably contribute to resistance to the mycotoxin. Following oxidation by CYP, the resulting epoxidated AFB1 can hydrolyze to AFB1-8,9-dihydrodiol (15); this may be catalyzed by epoxide hydrolase or occur spontaneously. At physiological conditions, the latter metabolite can rearrange to generate a reactive dialdehyde that modifies lysyl residues in proteins through Schiff bases (15, 16). A member of the AKR superfamily has been isolated that catalyzes the reduction of AFB1-dialdehyde to AFB1dialcohol (17). This aflatoxin aldehyde reductase, designated AFAR, is only distantly related to other mammalian AKRs (18) and has subsequently been placed in the AKR7 family (19). It has been proposed that AFAR reduces the cytotoxicity of AFB1 by preventing the binding of the dialdehydic form of the mycotoxin to primary amine groups in intracellular proteins (17). Provocatively, in humans, it has recently been shown that the AFAR gene is located on chromosome 1 in a region that frequently is deleted in sporadic colorectal and liver cancer, suggesting that the protein may function in humans to inhibit tumorigenesis (20). The ability of chemical agents to prevent the development of cancer has provoked much interest as a means of reducing the incidence of neoplastic disease in human populations (21). Because it is highly improbable that AFB1-producing molds can be eradicated from the environment, chemoprevention is an attractive strategy to protect individuals from the risk of liver cancer caused by exposure to the mycotoxin. At present, the antischistosomal drug OPZ is being investigated as a chemopreventive agent against AFB1 hepatocarcinogenesis in humans (22), but it is desirable to develop alternative therapies for this purpose. Using the rat as an experimental model, the phenolic antioxidants BHA and EQ, and the dithiolethione OPZ have been reported to inhibit AFB1 hepatocarcinogenesis (reviewed in Ref. 23). Similarly, PB and "-NF, synthetic model inducers of drug-metabolizing enzymes, as well as I3C, a breakdown product of glucosinolates generated during consumption of cruciferous vegetables, have also been reported to inhibit either AFB1-initiated liver cancer or the in vivo binding of AFB1 to hepatic DNA in the rat (23–26). Identification of the genes that are induced by chemopreventive agents has focused primarily on synthetic antioxidants and dithiolethiones (26 –29). Some of these agents induce GSTA5, a rat class-! transferase subunit [originally called Yc2 or 10 (12)] that possesses high glutathione conjugating activity toward AFB1-8,9-epoxide (27, 30). Although AFAR has been shown to be highly inducible by BHA, butylated hydroxytoluene, EQ, and 1,2-dithiole-3-thione (17, 28, 31, 32), and modestly inducible by I3C, OPZ, and phenethyl isothiocyanate (24, 31, 32), little is known about its regulation by other chemopreventive agents. Relatively few studies have documented the ability of naturally occurring chemicals to induce AKR or GST isoenzymes. In the present report, the roles of CMRN, BITC, and I3C in regulating the levels of these two enzyme systems in rat liver of both sexes were explored because they are present in vegetable-enriched diets that are believed to protect against malignant disease (33). Previous studies in male rats have shown that feeding them diets containing I3C diminishes events associated with tumorigenesis (24, 25). This indole has not, however, been studied in the female rat. The possible chemopreventive effects of CMRN and BITC against AFB1 have not been investigated in either sex of the rat. Experiments are now described to determine whether CMRN, BITC, and I3C enhance detoxification of AFB1 by inducing AFAR and/or GSTA5. The regulation of these detoxication enzymes has been studied in both sexes because the expression of at least one of the proteins, GSTA5, is sexually dimorphic in the rat (27). In this study NQO has been used as a positive

MATERIALS AND METHODS Chemicals The sources of all chemicals and reagents have been reported previously (27, 31, 37). Animals Test Feeding with Chemopreventive Agents to Identify Potent Enzyme Inducers. Ten-week-old male and female Fischer 344 rats, obtained from Harlan Olac Ltd. (Bicester, Oxon, United Kingdom), were housed in Moredun isolators (Moredun Animal Health Ltd., Edinburgh, Scotland) under negative pressure with 12-h light and dark cycles at a temperature range of 19 –23°C and a humidity of 40 – 60%. Animals were allowed to acclimatize for 2 weeks before being fed either a normal diet (powdered RM1 diet supplemented with 2% arachis oil) or normal diet containing the inducing agent of interest. Each test compound was administered to three male and three female Fischer rats, and during the experiment they were given free access to food and water. The chemopreventive agents were provided in the animal food for 2 weeks in the following amounts: BHA, 0.75% (w/w); BITC, 0.5% (w/w); CMRN, 0.5% (w/w); EQ, 0.5% (w/w); I3C, 0.5% (w/w); and OPZ, 0.075% (w/w). DEM [at 0.5% (w/w)] was administered for 5 days in the food. The t-SO (at 400 mg/kg) was dissolved in 0.5 ml of arachis oil, before daily i.p. administration on 3 consecutive days, and "-NF (at 200 mg/kg) was dissolved in PBS before daily i.p. administration for 7 consecutive days. PB was added to the drinking water at a concentration of 0.1% (w/v) for 7 days. Short-Term Intervention with CMRN to Study Development of Preneoplastic Foci. Six groups of eight 12-week-old male F344 rats were administered one of the following experimental diets for 13 weeks: group 1, RM1 control maintenance diet throughout; group 2, 0.05% (w/w) CMRN in RM1 diet throughout; group 3, 2 ppm AFB1 in RM1 diet for 6 weeks, followed by RM1 control diet for 7 weeks; group 4, 2 ppm AFB1 in RM1 throughout; group 5, 0.05% (w/w) CMRN in RM1 diet for 2 weeks followed by 2 ppm AFB1 in RM1 diet containing 0.05% (w/w) CMRN for 11 weeks; group 6, 2 ppm AFB1 in RM1 diet for 6 weeks, followed by AFB1 in RM1 diet containing 0.05% (w/w) CMRN for 7 weeks. In these experiments, AFB1 was dissolved in arachis oil and mixed into powdered RM1 diet to give a final concentration of 2 ppm AFB1 and 2% (w/w) arachis oil. CMRN was similarly dissolved in arachis oil to give a final concentration of 0.05% (w/w) in the RM1 diet; a lower concentration of CMRN was used in the intervention study than in the short-term feeding study because of possible hepatotoxicity effects (38). Long-Term Intervention with CMRN to Study Tumor Formation. Six groups of eight 12-week-old male F344 rats were administered the same diets described above. However, in this long-term study, the rats were placed on CMRN- and AFB1-containing diets for 24 weeks before being transferred onto a control diet from week 25 until termination of the experiment at week 50. Tissue Preparation. Animals were culled using CO2, and tissues were removed immediately. Microsomal and cytosolic fractions were prepared from fresh liver or from samples snap-frozen in liquid N2 and stored at "70°C. Tissue slices were taken into ice-cold acetone for immunohistochemistry and into buffered formalin for H&E histology. Enzyme and Protein Assays

958

Rat liver extracts were routinely prepared in 20 mM sodium phosphate buffer (pH 7.0) that contained 1 mM DTT, and 100,000 # g supernatants (cytosols) were prepared by ultracentrifugation. GST activity toward CDNB


was assayed using a Cobas Fara centrifugal analyzer (Hoffmann-La Roche Ltd, Basel, Switzerland; Ref. 27). AKR activity toward 2-CBA and 9,10-phenanthrenequinone was measured at pH 7.0 and 25°C by following the oxidation of NADPH at 340 nm (31, 39). GST activity toward AFB1-8,9-epoxide and AKR activity toward AFB1-dialdehyde were measured simultaneously using a HPLC-based assay (17). Protein concentrations were determined by the Coomassie dye-binding method with BSA as a calibration standard (31).

Western Blotting

Source and Production of Antibodies

Resolution of AKR Isoenzymes

In this study, rabbit polyclonal and mouse monoclonal antibodies against AKR, GST, and NQO isoenzymes were used, both to monitor protein expression and to identify reductases eluted from column fractions. Antiserum against the inducible rat AFAR is highly specific (17). However, when excessive amounts of the antibody are used in incubations, cross-reactivity is observed toward a related electrophoretically distinct constitutively expressed rat AFAR2 protein; this rat AFAR2 polypeptide cross-reacts much more strongly with antisera raised against human AFAR1;6 the characterization of human AFAR1 and the specificity of antibodies raised against the protein have been described previously (40). The antibodies used against GST are also highly specific (13, 37, 41). Antibodies against rat NQO1 were prepared during the present study using bacterially expressed protein. To this end, the cDNA for NQO1 was cloned into the pET15b vector and transformed into Escherichia coli BL21 pLysS to allow isopropyl "-D-thiogalactoside-inducible expression of polyhistidinetagged oxidoreductase. Two PCR steps were used to produce a full-length rat NQO1 cDNA for ligation into pET15b: the first involved generation of cDNA comprising the entire coding sequence of NQO1 from pDTD55 (42), a clone that represents codons 13–274 of rat NQO1 (43); the second step involved addition of an oligonucleotide to the 5$-end of the cDNA, which provided EcoRI and NdeI restriction sites immediately adjacent to the ATG initiation codon. In the first reaction, pDTD55 (42) was used as the template using the forward 54-nt primer, 5$-ATGGCGGTGAGAAGAGCCCTGATTGTATTGGCCCACGCAGAGAGGACATCATTC-3$, corresponding to codons 1–18 of the native protein (43), and the reverse 28-nt primer, 5$-CGCGGATCCGTCTAACTACATGGTATGG-3$, which contained 19 nucleotides corresponding to the distal 3$-untranslated region of the cDNA, as well as an engineered BamHI site. PCR was performed in a total volume of 100 $l of Pfu reaction buffer, containing 40 ng of pDTD55, 10 $g of each of the two oligonucleotides, 2 units of cloned Pfu DNA polymerase, 5% (v/v) DMSO, and each of the four dNTPs at a concentration of 200 $M. Amplification was carried out for 30 cycles, each of which entailed denaturation of the template DNA at 94°C for 2 min, annealing of primers at 56°C for 2 min, and extension at 72°C for 3 min. Finally, the amplified product was purified using QIAquick PCR purification kit (QIAGEN Ltd, Crawley, United Kingdom) before an aliquot was used as template for the second PCR step. During the second step, the forward 28-nt primer, 5$-CCGGAATTCATATGGCGGTGAGAAGAGC3$, containing the first 17 coding nucleotides of NQO1, along with the same reverse primer used in the first PCR step, was used in amplification under conditions identical to those described above. The resulting product was purified, digested with EcoRI and BamHI, and ligated into pBluescript SK% that had been similarly treated. This construct (pBS-NQO1) was transformed into E. coli NM522 cells, and the DNA insert was sequenced over 200 bp from both 5$ and 3$ ends to establish that the primers had been incorporated into the cDNA satisfactorily. Lastly, the insert in pBS-NQO1 was removed by digestion with NdeI and BamHI and ligated into pET15b that had been restricted with the same enzymes. The resulting plasmid (pET-NQO1) was transformed into E. coli strain BL21 pLysS and used to produce polyhistidine-tagged NQO1 that was purified on a HiTrap Chelating column as described for human AKR (39). From 1.2 liters of bacterial culture, !1 mg of homogeneous rat NQO1 was obtained. Aliquots (100 $g of protein) of purified NQO1 in complete Freund’s adjuvant were used as primary immunogen in female New Zealand White rabbits, whereas NQO1 (100 $g of protein) in Freund’s incomplete adjuvant was used in each of three booster immunizations on days 28, 42, and 56 after the initial immunization; bleeds were collected on day 1 (pre-immune sera) and day 70 (working antibodies).

The AKR isoenzymes were resolved by anion exchange chromatography on Q-Sepharose (31). Hepatic cytosols (100,000 # g supernatant) from rats fed control diet and diet containing a chemopreventive agent were analyzed in parallel. For practical reasons, only two samples were processed at a time, and livers from male rats were analyzed separately from those of females. Cytosols were prepared from !5 g of liver in 20 mM Tris-HCl (pH 8.2; the pH was adjusted with buffer at room temperature prior to transfer to 4°C) containing 1 mM DTT (buffer A), and were dialyzed for 18 h against two changes, each of 2 liters, of buffer A before being subjected to Q-Sepharose chromatography. To ensure comparability, the control liver cytosol and the chemopreventive agent-treated cytosol were each applied to columns of identical size (1.6 # 40.0 cm) that were equilibrated and eluted (31.5 ml/h) with buffer A and, following application of the samples to the columns, were developed in parallel with 0 –120 mM NaCl in buffer A from the same gradient-forming reservoir. After the gradient was complete, the remaining AKRs were eluted with 250 mM NaCl in buffer A. The volume of column fractions collected was kept constant at 6.3 ml throughout the study.

6

Western blotting was performed as described previously (31, 37). Prior to blotting, an SDS-PAGE gel was stained with Coomassie Blue to provide visual confirmation that equivalent amounts of the various samples were being analyzed. After the electrophoretically resolved proteins were blotted onto Immobilon-P, the membranes were stained with Ponceau S to verify that transfer of all samples was satisfactory.

Quantitation of GST Subunits by Reversed-Phase HPLC Soluble GSTs were purified by affinity chromatography on glutathioneagarose (27). Cytosols from 2 g of liver, prepared in 50 ml of ice-cold 20 mM sodium phosphate buffer (pH 7.0) containing 1 mM DTT (buffer B), were applied at 25 ml/h to 1.6 # 10.0 cm columns of glutathione-agarose that were eluted at 4°C with buffer B. The columns were washed extensively until the absorbance of the eluate at 280 nm was &0.05 units. Under these conditions, it was found that typically at least 90% of the GST activity toward CDNB was retained by glutathione-agarose. The class-!, -$, and -% GST isoenzymes were eluted from the affinity column as a single pool with 10 mM GSH in 200 mM Tris-HCl buffer (pH 9.0). The affinity-purified material was dialyzed against two 1-liter changes of buffer B before individual GST subunits were resolved by reversed-phase HPLC using a 0.46 # 25.0 cm Brownlee C18 column (7 $m particle size; 300 Å pore size). The chromatography conditions used were similar to those described elsewhere (37); a gradient of 35–56% acetonitrile formed in aqueous 0.1% (v/v) trifluoroacetic acid was formed over 60 min using Waters 510 HPLC pumps to deliver solvent at 1.0 ml/min. Aliquots of affinity-purified GST protein (40 –100 $g) were injected for HPLC analysis; the sample applied to the HPLC column was normalized on the basis of the amount of total liver cytosolic protein loaded onto the glutathione-agarose affinity column. The amount of each GST subunit was estimated from the peak area at 214 nm, using the following molar absorption coefficients (&214): GSTA1, 29.5 # 104; GSTA2, 29.5 # 104; GSTA3, 31.1 # 104; GSTA4, 30.5 # 104; GSTA5, 31.1 # 104; GSTM1, 43.1 # 104; GSTM2, 44.1 # 104; and GSTP1, 30.2 # 104. Northern Blotting Total RNA was isolated from rat livers using the method of Chomczynski and Sacchi (44) and was size fractionated by denaturing electrophoresis in formaldehyde-agarose gels. The RNA was transferred to nylon membranes and probed with radioactively labeled cDNA: for AFAR, the probe comprised the entire coding region of the reductase in pEE60 (18); for GSTA5, the probe was the NdeI-BamHI restriction fragment obtained from pET-rGSTYc2-k (27); for GSTP1, the probe was the EcoRI-SalI restriction fragment from pGP5 (45); for NQO1, the probe was the NdeI-BamHI insert from pBS-NQO1 (above). As a loading control, a GAPDH probe was used (31). Immunohistochemistry This was carried out using acetone-fixed paraffin-embedded tissue by methods described previously (46). The polyclonal antibodies against AFAR,

V. P. Kelly, L. S. Ireland, E. M. Ellis, and J. D. Hayes, unpublished results.

959


GSTP1, and NQO1 were used at a dilution of 1:200. The monoclonal antibodies against GSTA5 were used without prior dilution. Statistical Analyses

Table 2 Effect of cancer chemopreventive agents on liver GST activities toward AFB18,9-epoxide and the model substrate CDNB GST activity was measured as described in the text. In each group, soluble 100,000 # g supernatants were from the livers of three animals (n ' 3). GST activitya

Unless otherwise stated, statistical significance was determined by one-way ANOVA by Dunnett’s test, with a 5% critical value.

AFB1-8,9-epoxide Treatment

RESULTS As a first step to determine whether the phytochemicals chosen for study might be effective at preventing AFB1 hepatocarcinogenesis, a 2-week feeding trial was instigated to establish whether they increase detoxification activity toward reactive metabolites of the mycotoxin. In the initial study, rat were fed ad libitum with the naturally occurring chemicals BITC, CMRN, and I3C, all at 0.5% (w/w) in RM1 diet. The inducing capabilities of these compounds were compared with EQ, BHA, OPZ, PB, and "-NF, all of which have been reported to prevent AFB1 hepatocarcinogenesis; these latter compounds were administered at doses taken from the literature that were known to confer protection (23). Consumption of Phytochemicals Causes Enhanced Detoxification of AFB1-Dialdehyde in Livers of Male and Female Rats. Table 1 shows that AFAR activity is increased in the livers of rats fed diets containing BITC, CMRN, and I3C. CMRN produced a dramatically greater increase in hepatic reductase activity than did either BITC or I3C. In males, the increase in catalytic activity toward AFB1-dialdehyde was estimated to be !400-fold and in females, it was found to be !125-fold. By contrast, the increased activity in livers from rats of either sex fed diets containing BITC or I3C was between 5- and 7-fold. The synthetic antioxidants EQ and BHA, which were supplemented in the diet at amounts similar to the phytochemicals, appeared to be less potent inducers of AFAR activity than CMRN; treatment with antioxidants caused increases in reductase activity of between 10- and 110-fold in male rats and between 5and 20-fold in female rats. DEM, a GSH-depleting agent, was a more effective inducer of AFAR activity in male rats than was BITC or I3C, but it was not as effective as CMRN. The only compound that appeared to be as effective as CMRN at inducing AFAR activity in male rats was t-SO, a model inducer of phase II drug-metabolizing enzymes; the dosing schedule for this drug differed significantly from that used for the phytochemicals, and this must be recognized when making comparisons. Table 1 Hepatic AKR activities toward aflatoxin B1 dialdehyde and the model substrate 2-CBA in rats treated with chemopreventive agents Specific activities were measured in cytosols prepared from the livers of three animals in each group (n ' 3), using the substrates indicated. The cofactor for both substrates was NADPH. The experiments for each of the compounds tested were performed on at least two separate occasions, and the results shown represent typical observations. AKR activitya AFB1 (CHO)2 Treatment Control EQ OPZ CMRN I3C BHA BITC PB DEM t-SO "-NF a

2-CBA

Males

Females

Males

Females

0.4 ( 0.03 44.8 ( 1.3b 5.5 ( 0.25 155.8 ( 6.1b 2.9 ( 0.10 4.3 ( 0.13 2.1 ( 0.06 2.7 ( 0.10 11.9 ( 0.45b 176.4 ( 5.1b 3.9 ( 0.11

0.1 ( 0.01 2.1 ( 0.17b 0.3 ( 0.01 12.5 ( 0.35b 0.5 ( 0.01b 0.5 ( 0.02b 0.7 ( 0.02b 0.4 ( 0.02 0.5 ( 0.01b 5.5 ( 0.28b 0.8 ( 0.04b

2.4 ( 0.9 21.6 ( 0.2b 5.4 ( 0.1b 27.1 ( 0.3b 5.2 ( 0.2b 7.3 ( 0.5b 15.9 ( 0.1b 6.3 ( 0.3b 13.2 ( 0.6b 39.4 ( 0.4b 6.2 ( 0.1b

1.7 ( 0.2 16.4 ( 0.1b 2.8 ( 0.1b 11.1 ( 0.5b 3.5 ( 0.2b 3.8 ( 0.1b 7.3 ( 0.1b 4.5 ( 0.2b 5.2 ( 0.6b 18.9 ( 0.1b 4.7 ( 0.2b

Activity in pmol/min/mg for AFB1(CHO)2 and nmol/min/mg for 2-CBA. Significantly different from control animals (P & 0.05) using Dunnett’s multiple comparison test. b

Control EQ OPZ CMRN I3C BHA BITC PB DEM t-SO "-NF

CDNB

Males

Females

Males

Females

4.0 ( 0.3 166.0 ( 5.3b 49.0 ( 1.7b 271.0 ( 7.9b 42.0 ( 1.0b 58.0 ( 1.2b 61.0 ( 1.4b 80.0 ( 2.6b 66.0 ( 1.3b 159.0 ( 3.8b 62.0 ( 2.0b

45.0 ( 3.1 75.5 ( 4.5b 59.0 ( 2.0b 119.0 ( 5.5b 81.0 ( 3.9b 69.0 ( 2.3b 80.0 ( 2.3b 107.0 ( 5.3b 68.0 ( 2.9b 61.0 ( 2.4b 61.0 ( 2.4b

1.06 ( 0.08 3.42 ( 0.01b 2.24 ( 0.05b 3.84 ( 0.15b 2.37 ( 0.09b 3.02 ( 0.05b 1.88 ( 0.02b 3.30 ( 0.02b 0.83 ( 0.16 3.50 ( 0.10b 2.84 ( 0.07b

0.74 ( 0.04 2.65 ( 0.05b 1.12 ( 0.01b 2.07 ( 0.45b 2.05 ( 0.07b 1.40 ( 0.02b 1.57 ( 0.10b 2.01 ( 0.03b 0.80 ( 0.10 2.22 ( 0.06b 2.13 ( 0.05b

a

Activity in pmol/min/mg for AFB1-8,9-epoxide and $mol/min/mg for CDNB. Significantly different from control animals (P & 0.05) using Dunnett’s multiple comparison test. b

The increase in AFAR activity brought about by the various treatments appeared to be more pronounced in male rats (Table 1). The possibility that this sex difference might represent an in vitro artifact of the assay system is unlikely because the same trend was apparent when reductase activity toward 2-CBA was measured; this latter substrate has been shown to be relatively specific for AFAR in the livers of male rats that are selenium deficient (37). Thus, CMRN produced a 11.0- and 6.5-fold increase in activity toward 2-CBA in male and female rats, respectively. The relative increase in hepatic 2-CBA reductase activity produced by CMRN was greater than those observed with synthetic antioxidants. Male rats treated with BHA, EQ, or OPZ showed increases in 2-CBA reductase activity in the liver of 3.0-, 9.0-, and 2.3-fold, respectively, whereas female rats similarly treated showed increases of 2.2-, 9.6-, and 1.6-fold. The increase in reductase activity toward 2-CBA produced by CMRN was, however, comparable to those produced by the model inducers of drug-metabolizing enzymes DEM and t-SO; both of these increased 2-CBA reductase activity in the livers of male rats, but only t-SO caused a major increase in female rat liver. The Ability of Phytochemical-containing Diets to Enhance Hepatic Transferase Activity toward AFB1-8,9-Epoxide Is More Marked in Male than in Female Rats. Feeding rats diets that contained BITC, CMRN, or I3C resulted in a significant increase in hepatic GST activity (Table 2). In males, the increase varied between !10- and 65-fold, and although the increase was less marked in female rats, this is in part due to the sexually dimorphic expression of GSTA5, the major AFB1-metabolizing rat transferase subunit. It had been established previously that the female rat possesses higher constitutive hepatic levels of transferase activity toward epoxidated AFB1 than the male (27). Consistent with differences in the basal levels of the GSTA5 subunit in male and female rat liver, an !11-fold difference in AFB1-GSH-conjugating activity was observed between the two sexes on control diet (Table 2). Despite the relatively elevated basal GST activity toward AFB1 in female rat liver, BITC, CMRN, and I3C further increased this activity between 1.8- and 2.6-fold. CMRN was a strong inducer of AFB1-GSH-conjugating activity. Among the antioxidants and model inducing agents studied, EQ and t-SO produced the largest increases in this activity in male rat liver, but were less effective than CMRN. In addition, EQ and t-SO were found to be not particularly effective at increasing AFB1-GSH-conjugating activity in the female. The fact that both xenobiotics increased general transferase activity toward CDNB indicates that the

960


relative inability to enhance conjugation of GSH with AFB1 is probably due to EQ and t-SO serving as better inducers of class-$ GST than GSTA5 in the female. It is noteworthy that PB proved to be good at increasing hepatic GST activity in male rats, but was only a modest inducer of hepatic AKR activity. Regulation of AFAR and GSTA5 in Rat Liver by Phytochemicals and Other Xenobiotics. Western blotting was carried out to determine whether the increases in hepatic reductase and transferase activities toward AFB1 in the treated rats were due to elevation of AFAR and GSTA5 protein. Western blotting showed that the control livers from male and female rats contained levels of AFAR that were barely detectable. However, large amounts of AFAR were observed in samples from all of the animals treated with cancer chemopreventive agents (Fig. 1). Consistent with previous studies, the amount of AFAR was increased !15-fold, as estimated by phosphorimaging, in male rats fed EQ-containing diets (31). The phytochemicals BITC, CMRN, and I3C were all found to increase the level of AFAR protein in liver extracts. As might have been predicted from the data in Table 1, the CMRN-containing diet was shown to produce a dramatic 40-fold increase in AFAR. By comparison, the BITC- and I3C-containing diets produced just modest increases in the protein concentration. Increases of a magnitude similar to those obtained with CMRN were observed in t-SO-treated rats, whereas "-NF treatment also increased the content of AFAR significantly in male and female livers. In agreement with the AKR activities presented in Table 1, OPZ was found to be a significantly better inducer of AFAR in livers of male rats than in livers of female rats. Immunoblotting with a monoclonal antibody against GSTA5 showed that the livers of male rats fed a control diet essentially lack this transferase subunit, whereas it is clearly detectable in female control liver (Fig. 1). The GSTA5 subunit was found to be significantly elevated in the livers of male and female rats fed BITC-, CMRN-, or I3C-containing diets, the response observed with CMRN being the most marked; EQ elicited a increase in GSTA5 similar to that induced by CMRN. Interestingly, the blots showed OPZ to be a good inducer of GSTA5 in female rat liver, a result that contrasts with its effect on AFAR. Pleotropic Effects of Phytochemicals on Phase II Drug-metabolizing Enzymes in Male and Female Rats. To ascertain that the effects of BITC, CMRN, and I3C on drug metabolism were not restricted to AFAR and GSTA5, Western blotting experiments were performed using antibodies against other members of the AKR and GST superfamilies, as well as those against NQO1. Antibodies against rat aldehyde reductase, aldose reductase, 3!-hydroxysteroid dehydro-

genase, and )4-3-ketosteroid-5" reductase failed to provide any evidence that these proteins are inducible in rat liver by BITC, CMRN, or I3C (data not shown). However, probing the blots with a range of antibodies against class-!, -$, and -% GST showed that many of the rat transferase subunits are responsive to the dietary additives used in this study. Among the phytochemicals studied, CMRN proved a potent inducer of the class-% GSTP1 subunit (Fig. 1). The relative increases in the levels of the A5 and P1 polypeptides following treatment with different xenobiotics suggest that these enzymes are co-induced by at least some of the compounds investigated, such as EQ, t-SO, and "-NF, but not by others, such as I3C and PB. Immunoblotting with antibodies against NQO1 showed that this protein is present at higher levels in female rat livers than male rat livers (Fig. 1). The amount of NQO1 was found to be increased significantly in the livers of male and female rats fed diets containing either CMRN or I3C. Little induction of NQO1 was apparent in hepatic cytosols of either male or female rats fed BITC-containing diets. The synthetic chemopreventive agents EQ and OPZ induced NQO1 in both sexes, whereas induction by BHA appeared to be less pronounced in the livers of male rats than in female rats. Similarly, induction of NQO1 by "-NF is less pronounced in the livers of male rats than in female rats. Inducible Expression of AFAR, GST, and NQO in Different Zones of the Liver. In untreated animals of both sexes, the GSTA5 subunit and NQO1 were localized to hepatocytes in the centrilobular areas, whereas GST P1-1 was confined to the biliary epithelium. Staining for AFAR was indistinct but appeared in centrilobular areas in both sexes, with some periportal staining in females. Immunohistochemistry demonstrated that hepatocytes in the centrilobular, midzonal, and periportal regions, as well as biliary epithelial cells, all have the capacity to respond to chemopreventive agents and model inducers. The regions where induction occurred were found to vary with the agent under investigation, and in some cases with the enzyme being induced (Table 3). For most of the compounds examined, the pattern of induction was similar in both sexes, albeit with some variation in intensity. All four proteins were induced in the centrilobular zones by CMRN, EQ, and OPZ. By contrast, these proteins were induced primarily in the periportal zone by DEM and "-NF. The GST P1-1 isoenzyme was not induced in hepatocytes by I3C, BITC, or PB, but the two phytochemicals appeared to increase the intensity of staining in the biliary epithelial cells. Increased nuclear localization of the GSTA5 and GSTP1 subunits and NQO1 was apparent with some treatments. OPZ, EQ, DEM, and

Fig. 1. Increase in the levels of AFAR, GST, and NQO enzymes in livers of male and female rats treated with cancer chemopreventive agents. Portions (2 $g of protein) of soluble liver extract from control and treated male (m) and female (f) rats, along with appropriate protein standards, were subjected to SDS-PAGE and transferred to immobilon P. In all of the immunoblots shown, the samples were loaded in an identical fashion, as indicated above the lanes in the top blot. The immobilized polypeptides were probed with rabbit polyclonal antibodies against AFAR (1:3000 dilution), mouse monoclonal antibodies against GSTA5 (undiluted), rabbit polyclonal antibodies against GSTP1 (1:1000 dilution), or rabbit polyclonal antibodies against NQO1 (1:2000 dilution). Cross-reacting polypeptides were visualized by enhanced chemiluminescence. The data presented are typical examples of blots performed on at least four occasions using freshly prepared material from separate animals. Std, standard; Con, control.

961


Table 3 Induction of AFAR, GST subunits and NQO in rat liver by chemopreventive agents Livers were obtained from animals fed diets containing cancer chemopreventive agents or model enzyme inducers. Samples of liver were fixed in acetone, and sections were probed with antibodies and stained as described in “Materials and Methods.” Patterns of expression of inducible detoxication proteins Treatment Control EQ OPZ CMRN I3C BHA BITC PB DEM t-SO

"-NF

Sex

AFAR

GSTA5

GSTP1

NQO1

Male Female Male Female Male Female Male Female Male Female Male Female Male Female Male Female Male Female Male Female Male Female

Centrilobular, biliary Periportal, biliary %% throughout liver %% periportal %% centrilobular % centrilobular %% centrilobular & midzonal %% centrilobular & midzonal %% periportal (patchy) % periportal % throughout (less in midzone) % periportal % centrilobular "ve % centrilobular % centrilobular %% periportal % periportal %% throughout liver %% throughout liver %% periportal %% periportal

Centrilobular, biliary Centrilobular, biliary %% centrilobular, not periportal %% centrilobular, not periportal %% centrilobular %% centrilobular %% centrilobular & midzonal %% centrilobular & midzonal % patchy throughout liver % patchy throughout liver % throughout (less in midzone) % throughout (less in midzone) % mostly centrilobular % throughout (less periportal) % centrilobular % centrilobular % throughout (less in midzone) % throughout (less in midzone) %% mostly centrilobular %% mostly centrilobular % periportal % periportal

Biliary Biliary, few % hepatocytesa %% throughout liver %% throughout liver % centrilobular % centrilobular %% throughout liver %% centrilobular & midzone % biliary % biliary % periportal % periportal %% biliary %% biliary "ve "ve % periportal % periportal % centrilobular %% throughout liver % periportal % periportal (diffuse)

Centrilobular Centrilobular %% centrilobular & biliary %% centrilobular & biliary %% centrilobular & biliary % centrilobular %% centrilobular & midzonal %% centrilobular & midzonal %% centrilobular & biliary %% centrilobular, periportal & biliary % throughout (less in midzone) % throughout (less in midzone) % centrilobular & biliary % throughout (less in midzone) % centrilobular (very restricted) % centrilobular (very restricted) % throughout % periportal & biliary % centrilobular (weak) % centrilobular (weak) %% throughout liver %% throughout (less in midzone)

a The following symbols provide a semiquantitative index of enzyme induction in hepatocytes in different regions of the liver and in biliary epithelium: "ve, no induction; %, modest increase in staining; %%, strong increase in staining.

"-NF were capable of inducing nuclear localization of all three proteins, whereas CMRN and t-SO affected only GST P1-1 and NQO1. Induction by CMRN of Hepatic Carbonyl-reducing Enzymes Is Limited to AFAR. In view of the dramatic induction of hepatic AFAR protein by CMRN, it was of interest to determine whether other carbonyl-reducing enzymes might be inducible by this phytochemical. Hepatic cytosolic AKR isoenzymes from male and female rats fed the control diet were subjected to anion exchange chromatography, and both yielded a single peak of 2-CBA reductase activity that was eluted by the final 250 mM NaCl step (Fig. 2). In cytosol from rats fed the CMRN-containing diet, this highly anionic peak was essentially unchanged in size. However, a major additional peak of 2-CBA activity that eluted from Q-Sepharose at !65 mM NaCl (between fractions 60 and 80) was found in the livers from these treated rats. This additional peak coeluted with the 2-CBA reductase activity from rats fed EQ-containing diet, indicating that it is chromatographically indistinguishable from authentic AFAR protein (data not shown). Antibodies raised against rat AFAR cross-reacted strongly with the additional peak of 2-CBA reductase activity (fractions 60 – 80), and during SDS-PAGE, the immunoreactive polypeptide was electrophoretically identical to the immunogen. The anti-AFAR sera showed no cross-reactivity with any of the other AKR peaks eluted by the linear 0 –120 mM NaCl gradient (data not shown). The second, relatively acidic peak of 2-CBA reductase activity eluted by 250 mM NaCl (at approximately fraction 130) showed a little cross-reactivity with antibodies against rat AFAR when the blot was exposed for extended periods. Interestingly, this protein cross-reacted strongly with antibodies raised against human AFAR1 (data not shown). In this case, the immunoreactive polypeptide was distinguishable from that in the first peak because it had a faster electrophoretic mobility. CMRN Induction of GST Subunits in Rat Liver. The immunoblotting described above showed that GSTA5 and GSTP1 are highly inducible by CMRN. Examination of the Coomassie Blue-stained SDS-PAGE gel suggested that the hepatic content of other transferases might also be increased by this compound. HPLC was therefore used to resolve a total of nine GST subunits (A1, A2, A3, A4, A5,

M1, M2, M3, and P1) from rat liver to determine whether these were inducible. As shown in Fig. 3, both male and female rats fed CMRNcontaining diets showed relative increases in A2, A3, A5, M1, and P1; in males the increases in A2, A3, A5, M1, and P1 were 4.5-, 1.7-, 20-, 1.5-, and *60-fold, respectively, whereas in females the increases in A2, A3, A5, M1, and P1 were 4.3-, 1.3-, 6.0-, 4.4-, and *80-fold. In these experiments, no obvious increases in A1, A4, M2, or M3 were observed (Table 4). CMRN Increases Hepatic mRNA Levels of AFAR, GST, and NQO. Northern blotting experiments established that elevated AFAR, GSTA5, GSTP1, and NQO1 protein levels in the livers of male and female rats treated with CMRN were due to increased steady-state mRNA levels (Fig. 4). By contrast, the level of GAPDH mRNA was identical between samples. The higher constitutive levels of GSTA5 and NQO1 in females that were noted at the protein level (Figs. 1 and 2) were also reflected at the mRNA level. CMRN Can Protect against AFB1-initiated Preneoplastic Hepatic Foci. The 2-week feeding study established that CMRN can markedly increase the expression of proteins in rat liver that are believed to be responsible for detoxification of AFB1. Experiments were therefore undertaken to determine whether consumption of this phytochemical could confer resistance against the mycotoxin. An initial 13-week feeding program was instigated that involved dietary intervention with CMRN at several time points. The concentration of CMRN used in this intervention study was reduced from 0.5 to 0.05% (w/w) because of concerns of long-term toxicity (47). Assays were performed to confirm that the lower dose of CMRN was effective at enhancing AFB1 detoxification capacity in rat liver. Western blotting showed that in the rats given 0.05% (w/w) CMRN alone for the entire 13-week duration of the experiment (Table 5, group 2), there were 1.3- and 3.7-fold increases in AFAR and GSTA5 polypeptides, respectively. Enzyme assays showed that these samples possessed a 3.1-fold increase in capacity to reduce AFB1-dialdehyde and a 2.9fold increase in ability to conjugate AFB1-8,9-epoxide with GSH. In the animals given 0.05% (w/w) CMRN alone for 2 weeks followed by 0.05% (w/w) CMRN with 2 ppm AFB1 for the following 11 weeks (Table 5, group 5) there was a 1.8-fold increase in AFAR and a 9.6-fold increase in GSTA5; hepatic cytosols from this group of rats

962


Fig. 2. AFAR is the principal CMRN-inducible AKR in male and female rat liver. Resolution of rat hepatic AKR was achieved by anion exchange chromatography on Q-Sepharose in 20 mM Tris-HCl buffer (pH 8.2) that contained 1 mM DTT (buffer A). Dialyzed liver cytosols from male rats fed either control (A) or CMRN-containing (B) diets were subjected simultaneously to chromatography on two 1.6 # 30.0 cm columns of Q-Sepharose that were developed in parallel from the same buffer reservoir. On a separate occasion, liver cytosols from female rats that had been fed a control diet (C) or a CMRN-containing diet (D) were similarly chromatographed on columns of Q-Sepharose. The columns were eluted at 31.5 ml/h, and all were developed with linear 0 –120 mM NaCl gradients formed in 600 ml of buffer A between fractions 30 and 125. Fractions of 6.3 ml were collected, and AKR activity toward 2-CBA (Œ) and 9,10-phenanthrenequinone (9,10-PQ; E) was measured. The final peak of reductase activity was eluted (around fraction 130) by the stepwise addition of 250 mM NaCl in buffer A. The identity of the reductases in the various peaks was determined using a panel of antibodies against rat AKR, and the solid horizontal line shows the fractions that contained AFAR1 immunoreactivity.

possessed 5.1- and 5.0-fold elevations in reductase and GST activities toward AFB1. In the rats given 2 ppm AFB1 alone for the first 6 weeks, followed by 2 ppm AFB1 with 0.05% (w/w) CMRN for the last 7 weeks of the experiment (Table 5, group 6), a 1.5-fold increase in AFAR and a 15.8-fold increase in GSTA5 were observed; liver cytosols from this group exhibited 6.0- and 4.9-fold increases in reductase and GST activities toward AFB1. The development of liver tumors as a consequence of exposure to AFB1 was monitored by the production of altered hepatic foci that stained positively either in immunohistochemistry for the GSTP1 subunit or in histochemistry for GGT. Treatment with CMRN alone produced no foci. However, significant numbers of foci positive for GSTP1 or GGT were apparent after 6 weeks of AFB1 treatment, and the numbers were substantially greater after 13 weeks of treatment (Table 5, groups 3 and 4). Pretreatment with CMRN before exposure to AFB1 almost completely prevented formation of foci (Table 5, group 5). However, when CMRN was given after AFB1, the numbers of foci were similar to those seen with 6 weeks of exposure to AFB1 alone (Table 5, compare groups 3 and 6). These data suggest that CMRN is effective at preventing initiation of carcinogenesis by AFB1, but that when administered after exposure to the mycotoxin, it does not reverse the changes that have already occurred.

CMRN Inhibits AFB1 Tumorigenesis. The prediction made from the short-term intervention study of preneoplastic foci, that CMRN would block AFB1 hepatocarcinogenesis, was in part also borne out in the long-term study. Although all animals exposed to AFB1 for 24 weeks developed multiple liver tumors, those that were pretreated with CMRN were partly protected against tumor formation. However, the livers of seven of the eight animals pretreated with CMRN were found to have some macroscopic lesions (Table 6, group 11). Although this finding was disappointing, the liver tumors in the rats pretreated with CMRN were generally fewer in number and smaller than in the rats that received no CMRN or received CMRN only subsequent to AFB1 exposure.

DISCUSSION The use of chemical agents to prevent liver cancer caused by exposure to AFB1 represents a valuable approach to improve public health in regions of the world where the mycotoxin-producing A. flavus mold is found. Treatment of individuals exposed chronically to AFB1 with the antischistosomal drug OPZ has been shown to inhibit the development of carcinogenesis-associated biomarkers (22). The

963


investigated as examples of phytochemicals, present in garden cress, tonka beans, and Brussels sprouts, respectively, that have been reported to increase GST activity toward CDNB (12). As shown in Tables 1 and 2, all of these compounds when administered at 0.5% in RM1 diet increased the catalytic capacity of rat liver to detoxify AFB1. Among these compounds, CMRN was found to be particularly effective at inducing AFAR and GSTA5, with the relative level of overexpression being at least equal to that observed with the synthetic antioxidants EQ and BHA. The enzyme induction results suggested that feeding rats diets containing CMRN should inhibit AFB1-initiated liver cancer. Furthermore, the facts that CMRN is efficiently absorbed by the liver and causes enzyme induction in several regions of the liver (Table 3) suggest that this phytochemical should be effective at conferring resistance to AFB1. This hypothesis was tested using both preneoplastic foci and tumors as biological end points. Evidence suggests that the incidence of cholangiofibroma, cholangiocarcinoma, and parenchymal liver cell tumors may be increased in rats receiving 0.5% CMRN but not in rats receiving 0.2% CMRN (47). In the intervention studies, it was therefore decided to supplement the diet with 0.05% CMRN rather than the higher dose used in the 2-week screening experiment. Rats fed diets containing 0.05% CMRN were found to possess between 3.0- and 6.0-fold increases in hepatic AKR and GST activity toward AFB1. When a 13-week feeding protocol was used, CMRN treatment prior to exposure to AFB1 prevented the formation of hepatic preneoplastic nodules (Table 5). In addition, when included in a 24-week feeding protocol, CMRN significantly reduced the final liver tumor burden in both number and size at 50 weeks (Table 6). Both the 13- and 24-week feeding studies demonstrated that addition of CMRN to the diet after a 6-week exposure to AFB1 was much less effective at inhibiting tumorigenesis than when CMRN was used prior to exposure to AFB1. Induction of Multiple Phase II Drug-metabolizing Enzymes by CMRN. Although the present study represents the first evidence that CMRN can prevent AFB1 hepatocarcinogenesis, this phytochemical has been shown previously to protect rodents against PAHs. It has, for example, been reported that administration of CMRN can prevent DMBA-initiated mammary carcinoma in the rat (48). In addition, Wattenberg et al. (49) have shown that CMRN can prevent benzo(a)pyrene from causing tumors of the forestomach in the mouse. In view of the ability of CMRN to confer resistance against multiple chemical carcinogens, it is possibly not surprising that consumption of diets containing this compound results in overexpression of a spectrum of detoxication systems. The ability of CMRN to induce hepatic detoxication proteins is not restricted to the AFB1-metabolizing enzymes AFAR and those GSTs containing the A5 subunit. In rat liver, CMRN was also found to markedly increase the level of the GSTP1 polypeptide and NQO1. Induction of the GST A2, A3, and M1 subunits was also observed (Fig. 3). The effect of CMRN on expression of drug-metabolizing enzymes in rat mammary gland was not explored in the present study. Because GST P1-1 efficiently detoxifies a number of epoxides formed from PAHs (12), presumably overexpression of this transferase accounts for the chemopreventive effects

Fig. 3. CMRN is a major inducer of class-!, -$, and -% GST subunits in rat liver. The transferases in rat liver were purified by glutathione-affinity chromatography, and the individual subunits were resolved by reversed-phase HPLC on a Brownlee 4.6 # 250 mm (7-$m) column as described in “Materials and Methods.” The absorbance of column eluate was monitored at 214 nm. Identification of each of the GST subunit-containing peaks eluted from the HPLC column has been reported previously (27). Panels represent the GST elution profiles obtained from the livers of male control rats (A), male CMRNtreated rats (B), female control rats (C), and female CMRN-treated rats (D).

risk of developing liver cancer from AFB1 exposure is highest in economically developing countries (2). Because financial constraints are likely to limit the use of prophylactic drugs in such regions, it would be helpful to identify naturally occurring foodstuffs that can prevent AFB1 hepatocarcinogenesis because they could be included in the human diet at minimal cost. Prevention of AFB1 Hepatocarcinogenesis by CMRN. On the basis of the hypothesis that induction of phase II drug-metabolizing enzymes confers resistance to AFB1, a 2-week test feeding study was undertaken to establish which phytochemicals are capable of inducing rat AFAR and GSTA5. In this study, BITC, CMRN, and I3C were

Table 4 Quantitation of hepatic GST subunits in rats fed CMRN-containing diets Values shown are the means ( SE (n ' 3). The GST subunits are listed in the order of their elution from the reversed-phase HPLC. GST subunit (mg/g protein) Sex

Diet

GSTM1

GSTM2

GSTP1

GSTA3

GSTA5

GSTM3

GSTA1

GSTA2

GSTA4

Male Male Female Female

Control CMRN Control CMRN

3.9 ( 0.1 5.9 ( 0.5 1.1 ( 0.1 4.9 ( 0.1

7.0 ( 0.1 7.3 ( 0.6 2.2 ( 0.1 5.9 ( 0.1

&0.1 5.9 ( 0.4 &0.1 7.9 ( 0.4

4.3 ( 0.1 7.23 ( 0.4 6.8 ( 0.4 8.5 ( 0.1

0.15 ( 0.01 3.3 ( 0.5 0.59 ( 0.03 3.8 ( 0.2

&0.1 &0.1 &0.1 &0.1

3.4 ( 0.1 3.5 ( 1.1 1.1 ( 0.1 3.8 ( 0.5

2.9 ( 0.1 13.1 ( 1.3 4.1 ( 0.9 17.6 ( 0.2

0.22 ( 0.02 0.4 ( 0.04 0.22 ( 0.05 0.4 ( 0.04

964


Fig. 4. CMRN-containing diet causes marked coordinate increases in hepatic mRNAs for AFAR, GSTA5, GSTP1, and NQO1 in male and female rats. Total mRNA from male and female rats fed either control or CMRN-containing diets was separated by denaturing electrophoresis before being transferred to a nylon membrane. The identities of the samples are at the top. The blot was probed with radioactively labeled cDNAs encoding AFAR (A), GSTA5 (B), NQO1 (C), and GSTP1 (D). As a loading control, the blot was also probed with cDNA for GAPDH (E). F, agarose gel. The agarose gel was stained with ethidium bromide and visualized by being placed on a UV light box and illuminated at 365 nm.

of CMRN against DMBA-initiated mammary carcinoma in the rat. Furthermore, the GSTP1 subunit has been shown to inhibit c-junNH2-kinase (50), and the overexpression of this class-% transferase may also influence signaling pathways in the mammary gland that

influence cell proliferation or apoptosis. Induction of NQO1 is likely to protect against the cytotoxic effects of PAHs by inhibition of redox cycling (51); although overexpression of NQO1 is not the principal mechanism of CMRN-induced resistance against DMBA, it is clear from gene knockout experiments that this reductase protects against the redox-cycling agent menadione (52). It is therefore concluded that CMRN not only increases the ability of the rat to detoxify AFB1, but through GSTP1 overexpression, it increases resistance to PAHs. Mechanism of Enzyme Overexpression Affected by CMRN. Northern blotting showed that the steady-state levels of the mRNAs for AFAR, GSTA5, GSTP1, and NQO1 are elevated in CMRNtreated liver, indicating that overexpression is regulated primarily at the nucleic acid level rather than at the protein level. It remains to be established whether the increases in these mRNAs involve transcriptional activation of the respective genes or stabilization of their message. The rat AFAR gene has not been described to date, and therefore the molecular mechanisms involved in its regulation are not known. However, the promoters of GSTP1 and NQO1 contain enhancers, designated GPEI (53) and ARE (36), that have been demonstrated to respond to the BHA metabolite t-BHQ (54). The 5$-flanking region of GSTA5 also contains a putative ARE (55). The presence of these enhancers suggests that the GSTA5, GSTP1, and NQO1 genes are probably regulated at the transcriptional level rather than by mRNA stabilization. It should be noted that the 5$-flanking region of rat NQO1 contains an XRE (36), and therefore this enhancer may contribute to induction of the quinone oxidoreductase by some of the compounds, such as "-NF and I3C (12). The upstream region of GSTP1 does not, however, contain an XRE, and therefore the coinduction of GSTP1, NQO1, GSTA5 and AFAR proteins by CMRN is most likely to occur through an ARE/GPEI (56). Further work is required to resolve this issue and to determine which transcription factors mediate induction by CMRN. It is probable that CMRN has to be metabolized for it to act as an enzyme inducer, although it is unclear what modification is required to achieve maximal induction. CMRN is subject to species-specific oxidation by CYP, a process that results in variable amounts of 7-hydroxycoumarin, 3-hydroxycoumarin, and o-hydroxyphenylpropionic acid being produced by the rat and mouse (57). Despite these species-specific differences in the biotransformation of CMRN, the benzopyrone is an excellent inducer of hepatic GST activity in both the rat and mouse (12). Dinkova-Kostova et al. (58) have found that CMRN and its 7-hydroxylated metabolite are poor inducers of NQO activity in the murine hepatoma Hepa 1c1c7 cell line. By contrast, these workers have shown that 3-hydroxycoumarin is a potent inducer of NQO activity in Hepa 1c1c7 cells (58). It was suggested that the ability of 3-hydroxycoumarin to undergo keto-enol tautomerism might be the chemical feature responsible for it acting as a highly effective inducing agent. Alternatively, once hydroxylated at the 3 position, the pyrone ring structure of CMRN can open to yield

Table 5 Effect of dietary CMRN on formation of AFB1-initiated hepatic preneoplastic foci Twelve-week-old male Fischer 344 rats were placed on specified diets for 13 weeks before examination of the liver for preneoplastic foci using an antibody raised against GSTP1-1 or by histochemical assay for GGT using methods described previously (24, 46). Each experimental group included eight animals. The diets were control (RM1), CMRN (0.05% CMRN in RM1), AFB1 (2 ppm AFB1 in control RM1 diet), and AFB1/CMRN (2 ppm AFB1 in RM1 diet fortified with 0.05% CMRN). Diet administered

a

Group

Diet

1. 2. 3. 4. 5. 6.

Control CMRN AFB1 (6 weeks) AFB1 (13 weeks) CMRN followed by AFB1 with CMRN AFB1 followed by AFB1 with CMRN

Weeks 3–6

Weeks 7–13

GST P1-1 %vea

GGT %ve

Control CMRN AFB1 AFB1 AFB1/CMRN AFB1

Control CMRN Control AFB1 AFB1/CMRN AFB1/CMRN

0 &1.0 65.4 ( 28.5 203.4 ( 117.1 &1.0 62.3 ( 28.5

0 0 7.4 ( 3.3 35.3 ( 29.4 &1.0 17.8 ( 8.4

Weeks 1–2 Control CMRN AFB1 AFB1 CMRN AFB1

Number of foci

%ve, the foci were scored positive if they were found to contain at least 10 cells that stained for either class % GST or GGT.

965


Table 6 Effect of dietary CMRN on formation of AFB1-initiated liver tumors Twelve-week-old male Fischer 344 rats were placed on specified diets for 24 weeks, followed by return to control diet for 26 weeks, before examination of the liver for hepatoma at the end of week 50. The diets were control (RM1), CMRN (0.05% CMRN in RM1), AFB1 (2 ppm AFB1 in control RM1 diet), and AFB1/CMRN (2 ppm AFB1 in RM1 diet fortified with 0.05% CMRN). Diet administered

Group

Diet

Weeks 1–2

7. 8. 9. 10. 11. 12.

Control CMRN AFB1 (6 weeks) AFB1 (24 weeks) CMRN followed by AFB1 with CMRN AFB1 followed by AFB1 with CMRN

Control CMRN AFB1 AFB1 CMRN AFB1

Characteristics of tumors

Weeks 3–6

Weeks 7–24

Weeks 25–50

No. of rats per group with tumors

No. of tumors per animal (range)

Tumor size (mm)

No. of rats with tumor *5 mm

Control CMRN AFB1 AFB1 AFB1/CMRN AFB1

Control CMRN Control AFB1 AFB1/CMRN AFB1/CMRN

Control Control Control Control Control Control

1/8 1/8 5/8 7/7 7/8 7/7

0.01 (0–1) 0.25 (0–2) 0.63 (0–2) 12.86 (10–20) 5.5 (0–5) 7.71 (1–15)

2.5 2 2–10 2–22 2–18 2–45

0/8 0/8 2/8 7/7 2/8 5/7

o-hydroxyphenylacetaldehyde, which in turn is converted in vivo to o-hydroxyphenylacetic acid (59). At present, it is not known whether it is 3-hydroxycoumarin that serves as the ultimate inducing agent, or whether this is achieved by “downstream” ring-opened metabolites. The data in Figs. 1– 4 demonstrate that CMRN is an excellent inducer of AFAR, GSTA2, GSTA5, GSTP1, and NQO1. As discussed above, evidence suggests that a metabolite of CMRN probably transcriptionally activates gene expression through the ARE and the related GPEI enhancer. Should this hypothesis be correct, it is anticipated that the extraordinary potency of CMRN as an inducing agent will be explained either because its metabolism by CYP generates free radicals or because it is converted into a strong Michael reaction acceptor (34, 35, 56). It has been pointed out that among phenylpropenoids that are potential chemopreventive agents, the better enzyme inducers possess o-hydroxyl groups on the aromatic ring (58, 60). This chemical signature is present on ring-opened metabolites formed from 3-hydroxycoumarin (59), and it is possible that o-hydroxyphenylacetaldehyde and/or o-hydroxyphenylacetic acid serve as effective inducing agents because of this feature. Specificity in Protein Induction Caused by Different Chemopreventive Agents. In addition to induction of AFAR, GSTA5, GSTP1, and NQO1 by CMRN, treatment with EQ, BHA, and "-NF also resulted in overexpression of all four proteins. By contrast, other xenobiotics did not cause coordinated overexpression of these proteins. For example, both BITC and I3C induced AFAR and GSTA5, but neither phytochemical induced GSTP1 in hepatocytes. BITC proved to be a weak inducer of NQO1 in livers of male rats and failed to induce NQO1 in livers of female rats. I3C induced NQO1 in both male and female rat livers. The drug OPZ induced AFAR and GSTA5, but it only modestly increased the amounts of GSTP1 and NQO1. The model inducer t-SO also exhibited unexpected specificity in that it was a potent inducer of AFAR but was relatively ineffective at inducing NQO1. At present it is unclear why these xenobiotics display selective induction of genes that are widely regarded as being coordinately regulated. It appears likely that the sequence context of different AREs affects their function. (See Ref. 61 for further discussion about this point.) Certainly, the data shown in Fig. 1 indicate that in vivo the GPEI enhancer in GSTP1 and the ARE enhancer in NQO1 are functionally distinct. An aspect of chemoprevention that has attracted little attention is that of variability in the site of action of different agents. The immunohistochemistry shown in Fig. 5 (and summarized in Table 3) revealed heterogeneity within the liver where enzyme induction occurs. Many of the agents have been found to act primarily in the centrilobular zone, with the more potent inducers, such as CMRN, causing increased protein expression in the midzonal and periportal regions of the liver. Several of the inducing agents, however, affected increased protein expression primarily in the periportal zone of the liver, with less induction being observed in the centrilobular zone. Thus, dietary

treatment with OPZ caused induction of AFAR, GSTA5, GSTP1, and NQO1 almost exclusively in the centrilobular region of the liver (Table 3). Treatment with CMRN also resulted in increased expression of these proteins in the centrilobular region, but induction was found to extend throughout the liver. By contrast, induction of AFAR, GSTA5, GSTP1, and NQO1 by "-NF and DEM was found to be mostly limited to the periportal zone of the liver. In addition to the zonal responsiveness of phase II drug-metabolizing enzymes to inducing agents, immunohistochemistry also showed that certain of the drug treatments could result in nuclear localization of the proteins. This was most apparent with GSTP1 and NQO1, where significant nuclear staining was apparent following treatment with CMRN, EQ, OPZ, t-SO, DEM, and "-NF. Interestingly, nuclear localization of inducible protein was not observed with AFAR. Sex-specific Effects of Chemopreventive Agents. Over the past 10 years, a large number of studies into the mechanisms of cancer chemoprevention have been reported. Many of the investigations that have used rodents as experimental models have focused on the male rat or male mouse. Few studies have considered that the effects of chemopreventive agents might be sexually dimorphic. Clearly, this is an important point that frequently is overlooked. In the rat, the GSTA5 subunit is known to be expressed constitutively in the livers of females at significantly higher levels than in males (27). In the rat, GSTA5 is regulated by growth hormone and testosterone (62), and the 5$-flanking region of the gene contains a putative estrogen-responsive half-site (55). The present report provides evidence that certain of the agents studied induce detoxication enzymes in a sex-specific fashion. The GSTA5 subunit was generally found to be more inducible in the livers of male rats than in female rats. In addition, it was noted that OPZ, I3C, BHA, BITC, PB, and DEM were significantly better at inducing AFAR in male rats than in female rats. Conversely, induction of NQO1 by CMRN, I3C, BHA, and "-NF was found to be more marked in livers of female rats than male rats. Induction of Human Phase II Drug-metabolizing Enzymes. An increasing body of literature indicates that human detoxication genes are inducible (13, 56). However, relatively little is known about the regulation of AFAR, GST, and NQO isoenzymes in humans by phytochemicals. Two human AFAR cDNAs have been cloned (40, 63), and although the expression of one of these varies significantly in livers from different individuals (63), it has not been established whether either is inducible. In this context, it should be noted that among human AKRs, dihydrodiol dehydrogenase has been found to be inducible in HepG2 hepatoma cells by "-NF, ethacrynic acid, and t-BHQ (64) and also in HT29 colon carcinoma cells by ethacrynic acid, t-BHQ, and DEM (65). Induction of human AKRs by phytochemicals has not been reported. By contrast with AKR, more is known about regulation of GST. In primary human hepatocytes, class-! and class-$ GST have been

966


Fig. 5. Overexpression of AFAR, GST, and NQO caused by dietary administration of CMRN is restricted primarily to centrilobular hepatocytes. Comparison between the immunohistochemical staining of liver sections from male rats fed control diets (A, C, and E) with those fed CMRN-containing diets (B, D, and F) shows that the increases in the levels of AFAR (A versus B), GSTP1 (C versus D), and NQO1 (E versus F) are most apparent in the centrilobular part of the liver. p, portal tract; c, central vein.

phase II drug-metabolizing enzymes in the colonic mucosa of patients who were placed on various doses of OPZ. The cohort of patients investigated all had an increased risk of colorectal cancer. Increases of up to 4-fold were noted in the expression of NQO1 in these individuals. Furthermore, an increase of 5.5-fold was noted in the level of #-glutamylcysteine synthase in these patients. Together, these results suggest that enhancement of detoxification and antioxidant capacity by chemopreventive agents is feasible in the clinical setting. Certainly, induction of these cytoprotective systems by chemopreventive

shown to be inducible by 3-methylcholanthrene, PB, OPZ, and 1,2dithiole-3-thione (66). The first three of these inducing agents are synthetic compounds, and although 1,2-dithiole-3-thione has been reported to be present in cruciferous vegetables, this finding has been disputed (67). Induction of these isoenzymes by BITC, CMRN, or I3C has not been described, but the level of class-% GST has been shown to be increased !30% in the colonic epithelium of volunteers who consumed 300 g of Brussels sprouts daily for 7 days (68). O’Dwyer et al. (69) reported a significant increase in expression of 967


“blocking” agents is not a phenomenon unique to rodents. The clinical value of such prophylactic strategies remains to be established. Concluding Comments. This report describes the identification of phytochemicals that are effective inducers of AFB1 detoxication enzymes. The study showed that CMRN is highly effective at inducing not only AFAR and GSTA5, but also certain other drug-metabolizing enzymes. On the basis of this information, the hypothesis that enzyme induction by CMRN would confer resistance to AFB1 tumorigenesis was tested in the rat. The results from dietary intervention showed that CMRN consumption does indeed provide protection against initiation of AFB1 hepatocarcinogenesis. The data presented in this report also reveal the ability of different phytochemicals and synthetic drugs to induce different enzymes in the liver in zone- and sex-specific fashions. Furthermore, certain inducing agents possess the ability to cause nuclear translocation of drug-metabolizing enzymes, thereby emphasizing the complexity of gene-environment interactions. ACKNOWLEDGMENTS We thank Helen W. L. Ball for expert help with the immunohistochemistry.

REFERENCES 1. Eaton, D. L., and Gallagher, E. P. Mechanisms of aflatoxin carcinogenesis. Annu. Rev. Pharmacol. Toxicol., 34: 135–172, 1994. 2. Groopman, J. D., Cain, L. G., and Kensler, T. W. Aflatoxin exposure in human populations: measurements and relationship to cancer. CRC Crit. Rev. Toxicol., 19: 113–145, 1988. 3. Iyer, R. S., Coles, F. F., Raney, K. D., Thier, R., Guengerich, F. P., and Harris, T. M. DNA adduction by the potent carcinogen aflatoxin B1: mechanistic studies. J. Am. Chem. Soc., 116: 1603–1609, 1994. 4. Gallaher, E. P., Kunze, K. L., Stapleton, P. L., and Eaton, D. L. The kinetics of aflatoxin B1 oxidation by human cDNA-expressed and human liver microsomal cytochromes P450 1A2 and 3A4. Toxicol. Appl. Pharmacol., 141: 595– 606, 1996. 5. Aguilar, F., Hussain, S. P., and Cerutti, P. Aflatoxin B1 induces the transversion of G3 T in codon 249 of the p53 tumor suppressor gene in human hepatocytes. Proc. Natl. Acad. Sci. USA, 90: 8586 – 8590, 1993. 6. McMahon, G., Davis, E. F., Huber, L. J., Kim, Y., and Wogan, G. N. Characterization of c-Ki-ras and N-ras oncogenes in aflatoxin B1-induced rat liver tumors. Proc. Natl. Acad. Sci. USA, 87: 1104 –1108, 1990. 7. Monroe, D. H., and Eaton, D. L. Comparative effects of butylated hydroxyanisole on hepatic in vivo DNA binding and in vitro biotransformation of aflatoxin B1 in the rat and mouse. Toxicol. Appl. Pharmacol., 90: 401– 409, 1987. 8. Monroe, D. H., and Eaton, D. L. Effects of modulation of hepatic glutathione on biotransformation and covalent binding of aflatoxin B1 to DNA in the mouse. Toxicol. Appl. Pharmacol., 94: 118 –127, 1988. 9. Buetler, T. M., Slone, D., and Eaton, D. L. Comparison of the aflatoxin B1-8,9epoxide conjugating activities of two bacterially expressed ! class glutathione Stransferase isozymes from mouse and rat. Biochem. Biophys. Res. Commun., 188: 597– 603, 1992. 10. Hayes, J. D., Judah, D. J., Neal, G. E., and Nguyen, T. Molecular cloning and heterologous expression of a cDNA encoding a mouse glutathione S-transferase Yc subunit possessing high catalytic activity for aflatoxin B1-8,9-epoxide. Biochem. J., 285: 173–180, 1992. 11. Moss, E. J., and Neal, G. E. The metabolism of aflatoxin B1 by human liver. Biochem. Pharmacol., 34: 3193–3197, 1985. 12. Hayes, J. D., and Pulford, D. J. The glutathione S-transferase supergene family: regulation of GST and contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit. Rev. Biochem. Mol. Biol., 30: 445– 600, 1995. 13. Hayes, J. D., and McLellan, L. I. Glutathione and glutathione-dependent enzymes represent a co-ordinately regulated defence against oxidative stress. Free Rad. Res., 31: 273–300, 1999. 14. Townsend, A. J., Fields, W. R., Haynes, R. L., Doss, A. J., Li, Y., Doehmer, J., and Morrow, C. S. Chemoprotective functions of glutathione S-transferases in cell lines induced to express specific isoenzymes by stable transfection. Chem.-Biol. Interact., 111–112: 389 – 407, 1998. 15. Neal, G. E., and Colley, P. J. The formation of 2,3-dihydro-2,3-dihydroxy aflatoxin B1 by the metabolism of aflatoxin B1 in vitro by rat liver microsomes. FEBS Lett., 101: 382–386, 1979. 16. Sabbioni, G., Skipper, P. L., Buchi, G., and Tannenbaum, S. R. Isolation and characterization of the major serum albumin adduct formed by aflatoxin B1 in vivo in rats. Carcinogenesis (Lond.), 8: 819 – 824, 1987. 17. Hayes, J. D., Judah, D. J., and Neal, G. E. Resistance to aflatoxin B1 is associated with the expression of a novel aldo-keto reductase which has catalytic activity towards a cytotoxic aldehyde-containing metabolite of the toxin. Cancer Res., 53: 3887–3894, 1993. 18. Ellis, E. M., Judah, D. J., Neal, G. E., and Hayes, J. D. An ethoxyquin-inducible aldehyde reductase from rat liver that metabolizes aflatoxin B1 defines a sub-family of aldo-keto reductases. Proc. Natl. Acad. Sci. USA, 90: 10350 –10354, 1993.

19. Jez, J. M., Bennett, M. J., Schlegel, B. P., Lewis, M., and Penning, T. M. Comparative anatomy of the aldo-keto reductase superfamily. Biochem. J., 326: 625– 636, 1997. 20. Praml, C., Savelyeva, L., Perri P., and Schwab, M. Cloning of the human aflatoxin B1-aldehyde reductase gene at 1p35-1p36.1 in a region frequently altered in human tumor cells. Cancer Res., 58: 5014 –5018, 1998. 21. Wattenberg, L., Lipkin, M., Boone, C. W., and Kelloff, G. J. (eds.), Cancer Chemoprevention. Boca Raton, FL: CRC Press, Inc., 1992. 22. Wang, J-S., Shen, X., He, X., Zhu, Y-R., Zhang, B-C., Wang, J-B., Qian, G-S., Kuang, S-Y., Zarba, A., Egner, P. A., Jacobson, L. P., Mun˜oz, A., Helzlsouer, K. J., Groopman, J. D., and Kensler, T. W. Protective alterations in phase 1 and 2 metabolism of aflatoxin B1 by oltipraz in residents of Qidong, People’s Republic of China. J. Natl. Cancer Inst., 91: 347–354, 1999. 23. Hayes, J. D., McLeod, R., Ellis, E. M., Pulford, D. J., Ireland, L. S., McLellan, L. I., Judah, D., J., Manson, M. M., and Neal, G. E. Regulation of glutathione S-transferases and aldehyde reductase by chemoprotectors: studies of mechanisms responsible for inducible resistance to aflatoxin B1. In: B. W. Stewart, D. McGregor, and P. Kleihues (eds.), Principles of Chemoprevention, pp. 175–187. IARC Scientific Publ. No. 139. Lyon, France: IARC, 1996. 24. Manson, M. M., Hudson, E. A., Ball, H. W. L., Barrett, M. C., Clark, H. L., Judah, D. J., Verschoyle, R. D., and Neal, G. E. Chemoprevention of aflatoxin B1-induced carcinogenesis by indole-3-carbinol in rat liver—predicting the outcome using early biomarkers. Carcinogenesis (Lond.), 19: 1829 –1836, 1998. 25. Stresser, D. M., Williams, D. E., McLellan, L. I., Harris, T. M., and Bailey, G. S. Indole-3-carbinol induces a rat liver glutathione transferase subunit (Yc2) with high activity towards AFB1 exo-epoxide. Association with reduced levels of hepatic aflatoxin-DNA adducts in vivo. Drug Metab. Dispos., 22: 392–399, 1994. 26. Kensler, T. W., Groopman, J. D., Eaton, D. L., Curphey, T. J., and Roebuck, B. D. Potent inhibition of aflatoxin-induced hepatic tumorigenesis by the monofunctional enzyme inducer 1,2-dithiole-3-thione. Carcinogenesis (Lond.), 13: 95–100, 1992. 27. Hayes, J. D., Nguyen, T., Judah, D. J., Petersson, D. G., and Neal, G. E. Cloning of cDNAs from fetal rat liver encoding glutathione S-transferase Yc polypeptides: the Yc2 subunit is expressed in adult rat liver resistant to the hepatocarcinogen aflatoxin B1. J. Biol. Chem., 269: 20707–20717, 1994. 28. Primiano, T., Gastel, J. A., Kensler, T. W., and Sutter, T. R. Isolation of cDNAs representing dithiolethione-responsive genes. Carcinogenesis (Lond.), 17: 2297–2303, 1996. 29. Primiano, T., Li, Y., Kensler, T. W., Trush, M. A., and Sutter, T. R. Identification of dithiolethione-inducible gene-1 as a leukotriene B4 12-hydroxydehydrogenase—implications for chemoprevention. Carcinogenesis (Lond.), 19: 999 –1005, 1998. 30. Johnson, W. W., Ueng, Y-F., Widersten, M., Mannervik, B., Hayes, J. D., Sherratt, P. J., Ketterer, B., and Guengerich, F. P. Conjugation of highly reactive aflatoxin B1 exo-8,9-epoxide catalysed by rat and human glutathione transferases: estimation of kinetic parameters. Biochemistry, 36: 3056 –3060, 1997. 31. Ellis, E. M., Judah, D. J., Neal, G. E., O’Connor, T., and Hayes, J. D. Regulation of carbonyl-reducing enzymes in rat liver by chemoprotectors. Cancer Res., 56: 2758 – 2766, 1996. 32. Manson, M. M., Ball, H. W. L., Barrett, M. C., Clark, H. L., Judah, D. J., Williamson, G., and Neal, G. E. Mechanism of action of dietary chemoprotective agents in rat liver: induction of phase I and II drug metabolizing enzymes and aflatoxin B1 metabolism. Carcinogenesis (Lond.), 18: 1729 –1738, 1997. 33. Potter, J. D., and Steinmetz, K. Vegetables, fruit, and phytoestrogens as preventive agents. In: B. W. Stewart, D. McGregor, and P. Kleihues (eds.), Principles of Chemoprevention, pp. 61–90. IARC Scientific Publ. No. 139. Lyon, France: IARC, 1996. 34. Prochaska, H. J., and Talalay, P. Regulatory mechanisms of monofunctional and bifunctional anticarcinogenic enzyme inducers in murine liver. Cancer Res., 48: 4776 – 4782, 1988. 35. Prestera, T., Holtzclaw, W. D., Zhang, Y., and Talalay, P. Chemical and molecular regulation of enzymes that detoxify carcinogens. Proc. Natl. Acad. Sci. USA, 90: 2965–2969, 1993. 36. Favreau, L. V., and Pickett, C. B. Transcriptional regulation of the rat NAD(P)H: quinone reductase gene. Identification of regulatory elements controlling basal level expression and inducible expression by planar aromatic compounds and phenolic antioxidants. J. Biol. Chem., 266: 4556 – 4561, 1991. 37. McLeod, R., Ellis, E. M., Arthur, J. R., Neal, G. E., Judah, D. J., Manson, M. M., and Hayes, J. D. Protection conferred by selenium deficiency against aflatoxin B1 in the rat is associated with the hepatic expression of an aldo-keto reductase and a glutathione S-transferase subunit that metabolize the mycotoxin. Cancer Res., 57: 4257– 4266, 1997. 38. Carlton, B. D., Aubrun, J-C., and Simon, G. S. Effects of coumarin following perinatal and chronic exposure in Sprague-Dawley rats and CD-1 mice. Fundam. Appl. Toxicol., 30: 145–151, 1996. 39. O’Connor, T., Ireland, L. S., Harrison, D. J., and Hayes, J. D. Major differences exist in the function and tissue-specific expression of human aflatoxin B1 aldehyde reductase and the principal human aldo-keto reductase AKR1 family members. Biochem. J., 343: 487–504, 1999. 40. Ireland, L. S., Harrison, D. J., Neal, G. E., and Hayes, J. D. Molecular cloning, expression, and catalytic activity of a human AKR7 member of the aldo-keto reductase superfamily: evidence that the major 2-carboxybenzaldehyde reductase from human liver is a homologue of rat aflatoxin B1-aldehyde reductase. Biochem. J., 332: 21–34, 1998. 41. Hayes, J. D., and Mantle, T. J. Use of immuno-blot techniques to discriminate between the glutathione S-transferase Yf, Yk, Ya, Yn/Yb and Yc subunits and to study their distribution in extrahepatic tissues. Evidence for three immunochemically distinct groups of transferase in the rat. Biochem. J., 233: 779 –788, 1986.

968


42. Bayney, R. M., Rodkey, J. A., Bennett, C. D. Lu, A. Y. H., and Pickett, C. B. Rat liver NAD(P)H:quinone reductase nucleotide sequence analysis of a quinone reductase cDNA clone and prediction of the amino acid sequence of the corresponding protein. J. Biol. Chem., 262: 572–575, 1987. 43. Bayney, R. M., Morton, M. R., Favreau, L. V., and Pickett, C. B. Rat liver NAD(P)H: quinone reductase. Regulation of quinone reductase gene expression by planar aromatic compounds and determination of the exon structure of the quinone reductase structural gene. J. Biol. Chem., 264: 21793–21797, 1989. 44. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156 – 159, 1987. 45. Suguoka, Y., Kano, T., Okuda, A., Sakai, M., Kitagawa, T., and Muramatsu, M. Cloning and the nucleotide sequence of rat glutathione S-transferase P cDNA. Nucleic Acids Res., 13: 6049 – 6057, 1985. 46. Green, J. A., Carthew, P., Heuillet, E., Simpson, J. L., and Manson, M. M. Cytokeratin expression during AFB1-induced carcinogenesis. Carcinogenesis (Lond.), 11: 1175–1182, 1990. 47. National Toxicology Program. U.S. Department of Health and Human Services, National Institutes of Health. NTP Technical Report on the Toxicology and Carcinogenesis Studies of Coumarin in F344/N Rats and B6C3F1 Mice. NIH Publication No. 93-3153. Bethesda, MD: NIH, 1993. 48. Feuer, G., Kellen, J. A., and Kovacs, K. Suppression of 7,12-dimethylbenz(a)anthracene-induced breast carcinoma by coumarin in the rat. Oncology, 33: 35–39, 1976. 49. Wattenberg, L. W., Lam, L. K. T., and Fladmoe, A. V. Inhibition of chemical carcinogen-induced neoplasia by coumarins and !-angelicalactone. Cancer Res., 39: 1651–1654, 1979. 50. Adler, V., Yin, Z., Fuchs, S. Y., Benezra, M., Rosario, L., Tew, K. D., Pincus, M. R., Sardana, M., Henderson, C. J., Wolf, C. R., Davis, R. J., and Ronai, Z. Regulation of JNK signaling by GSTp. EMBO J., 18: 1321–1334, 1999. 51. Prochaska, H. J., and Talalay, P. The role of NAD(P)H:quinone reductase in protection against the toxicity of quinones and related agents. In: H. Sies (ed.), Oxidants and Antioxidants, pp. 195–211. London: Academic Press, 1992. 52. Radjendirane, V., Joseph, P., Lee, Y-H., Kimura, S., Klein-Szanto, A. J. P., Gonzalez, F. J., and Jaiswal, A. K. Disruption of the DT diaphorase (NQO1) gene in mice leads to increased menadione toxicity. J. Biol. Chem., 273: 7382–7389, 1998. 53. Okuda, A., Imagawa, M., Maeda, Y., Sakai, M., and Muramatsu, M. Structural and functional analysis of an enhancer GPEI having a phorbol 12-O-tetradecanoate 13-acetate responsive element-like sequence found in the rat glutathione transferase P gene. J. Biol. Chem., 264: 16919 –16926, 1989. 54. Favreau, L. V., and Pickett, C. B. The rat quinone reductase antioxidant response element. Identification of the nucleotide sequence required for basal and inducible activity and detection of antioxidant response element-binding proteins in hepatoma and non-hepatoma cell lines. J. Biol. Chem., 270: 24468 –24474, 1995. 55. Pulford, D. J., and Hayes, J. D. Characterization of the rat glutathione S-transferase Yc2 subunit gene, GSTA5: identification of a putative antioxidant responsive element in the 5$-flanking region of rat GSTA5 that may mediate chemoprotection against aflatoxin B1. Biochem. J., 318: 75– 84, 1996. 56. Hayes, J. D., Ellis, E. M., Neal, G. E., Harrison, D. J., and Manson, M. M. Cellular response to cancer chemopreventive agents: contribution of the antioxidant responsive element to the adaptive response to oxidative and chemical stress. In: C. P. Downes, C. R. Wolf, and D. P. Lane (eds.), Cellular Responses to Stress, Biochem. Soc. Symp. No. 64, pp. 141–168. London: Portland Press, 1999.

57. Fentem, J. H., and Fry, J. R. Species differences in the metabolism and hepatoxicity of coumarin. Comp. Biochem. Physiol., 104C: 1– 8, 1993. 58. Dinkova-Kostova, A. T., Abeygunawardana, C., and Talalay, P. Chemoprotective properties of phenylpropenoids, bis(benzylidene)cycloalkanones, and related Michael reaction acceptors: correlation of potencies as phase 2 enzyme inducers and radical scavengers. J. Med. Chem., 41: 5287–5296, 1998. 59. Pelkonen, O., Raunio, H., Rautio, A., Pasanen, M., and Lang, M. A. The metabolism of coumarin. In: R. O’Kennedy, and R. D. Thornes (eds.), Coumarins. Biology, Applications and Mode of Action, pp. 67–92. Chichester: John Wiley & Sons, Ltd, 1997. 60. Dinkova-Kostova, A. T., and Talalay, P. Relation of structure of curcumin analogs to their potencies as inducers of phase 2 detoxification enzymes. Carcinogenesis (Lond.), 20: 911–914, 1999. 61. Hayes, J. D., Chanas, S. A., Henderson, C. J., McMahon, M., Sun, C., Moffat, G. J., Wolf, C. R., and Yamamoto, M. The Nrf2 transcription factor contributes both to the basal expression of glutathione S-transferases in mouse liver and to their induction by the chemopreventive synthetic antioxidants, butylated hydroxyanisole, and ethoxyquin. Biochem. Soc. Trans., 28: 33– 41, 2000. 62. Staffas, L., Ellis, E. M., Hayes, J. D., Lundgren, B., DePierre, J. W., and Mankowitz, L. Growth hormone- and testosterone-dependent regulation of glutathione transferase subunit A5 in rat liver. Biochem. J., 332: 763–768, 1998. 63. Knight, L. P., Primiano, T., Groopman, J. D., Kensler, T. W., and Sutter, T. R. cDNA cloning, expression and activity of a second human aflatoxin B1-metabolizing member of the aldo-keto reductase superfamily, AKR7A3. Carcinogenesis (Lond.), 20: 1215–1223, 1999. 64. Burczynski, M. E., Lin, H-K., and Penning, T. M. Isoform-specific induction of a human aldo-keto reductase by polycyclic aromatic hydrocarbons (PAHs), electrophiles, and oxidative stress: implications for the alternative pathway of PAH activation catalyzed by human dihydrodiol dehydrogenase. Cancer Res., 59: 607– 614, 1999. 65. Ciaccio, P. J., Jaiswal, A. K., and Tew, K. D. Regulation of human dihydrodiol dehydrogenase by Michael acceptor xenobiotics. J. Biol. Chem., 269: 15558 –15562, 1994. 66. Morel, F., Fardel, O., Meyer, D. J., Lanouet, S., Gilmore, K. S., Meunier, B., Tu, C-P. D., Kensler, T. W., Ketterer, B., and Guillouzo, A. Preferential increase of glutathione S-transferase class transcripts in cultured human hepatocytes by phenobarbital, 3-methylcholanthrene, and dithiolethiones. Cancer Res., 53: 231–234, 1993. 67. Kensler, T. W., Groopman, J. D., Roebuck, B. D., and Curphey, T. J. Chemoprevention by 1,2-dithiole-3-thiones. In: M-T. Huang, T. Osawa, C-T. Ho, and R. T. Rosen (eds.), Food Phytochemicals for Cancer Prevention I. Fruits and Vegetables ACS Symposium Series, Vol. 546, pp. 154 –163. Washington, DC: American Chemical Society, 1994. 68. Nijhoff, W. A., Grubben, M. J. A. L., Nagengast, F. M., Jansen, J. B. M. J., Verhagen, H., van Poppel, G., and Peters, W. H. M. Effects of consumption of Brussels sprouts on intestinal and lymphocytic glutathione S-transferases in humans. Carcinogenesis (Lond.), 16: 2125–2128, 1995. 69. O’Dwyer, P. J., Szarka, C. E., Yao, K-S., Halbherr, T. C., Pfeiffer, G. R., Green, F., Gallo, J. M., Brennan, J., Frucht, H., Goosenberg, E. B., Hamilton, T. C., Litwin, S., Balshem, A. M., Engstrom, P. F., and Clapper, M. L. Modulation of gene expression in subjects at risk for colorectal cancer by the chemopreventive dithiolethione Oltipraz. J. Clin. Investig., 98: 1210 –1217, 1996.

969