Intermittent dosing with oltipraz: relationship between chemoprevention of aflatoxin-induced tumorigenesis and induction of glutathione S-transferases.

Oltipraz [5-(2-pyrazinyl)-4-methyl-1,2-dithiole-3-thione] protects against chemical carcinogenesis in several animal models and is currently under evaluation as a possible chemopreventive agent in humans. Ideally, clinical chemopreventive interventions use dosing regimens that maximize efficacy while minimizing toxicity. Toward this end, the chemopreventive efficacy achieved by administration of intermittent doses of oltipraz was evaluated in rats. F344 rats were treated with oltipraz (0.5 mmol/kg, p.o.) once weekly, twice weekly, or daily over a 5-week period. After the first week, all rats were gavaged with 20 micrograms/kg of aflatoxin B1 for 28 consecutive days. Livers were analyzed 2 months after the last aflatoxin B1 dose, and the volume of liver occupied by glutathione S-transferase (GST)-P positive foci, a presumptive marker of neoplasia, was observed to be decreased > 95%, > 97%, or > 99% in livers of rats receiving once-, twice-weekly or daily oltipraz treatments, respectively. The chemopreventive actions of oltipraz have been associated with increases in the levels of phase 2 detoxifying enzymes, such as the glutathione S-transferase isozymes. Accordingly, GST conjugation activity measured with 1-chloro-2,4-dinitrobenzene as substrate increased 1.5-, 1.8-, or 2.4-fold for the once-weekly, twice-weekly or daily treatments, respectively, throughout a 7-day period. Quantitative HPLC analyses of GST subunits 24 h after 2 or 7 daily administrations of oltipraz showed that the levels of subunits Yb1, Yp, Yc2, and Ya2 were increased with maximum elevations of 5.6-, 11.1-, 6.4-, and 10.4-fold, respectively. In comparison, levels of subunits Yb2 and Yc1 were modestly elevated 1.8- to 2.6-fold, respectively, whereas subunit Ya1 was not induced. Remarkably, the levels of subunit Yp and Ya2 remained elevated approximately 2.3-fold 7 days after a single dose of oltipraz. In contrast, the levels of subunits Yb1 and Yc2 diminished to approximate control levels within 7 days after a single dose of oltipraz. GST mRNA levels for Ya, Yb, and Yp were measured by Northern blot analysis and were found to be elevated maximally to 13.7-, 13.5-, and 3.9-fold, respectively, after two daily oltipraz doses. Interestingly, GST Ya and Yb mRNA diminished to constitutive levels after 7 daily doses of oltipraz, with no corresponding decreases in GST subunit or activity levels. The levels of GST Ya and Yb mRNA decreased to constitutive levels within 4 days after a single oltipraz administration, whereas GST Yp mRNA levels remained elevated throughout the 7-day follow-up period.(ABSTRACT TRUNCATED AT 400 WORDS)

Xenobiotic-modulated expression of hepatic glutathione S-transferase genes in primary rat hepatocyte culture.

CYP 2B1/B2 and 1A1 expression in primary rat hepatocytes plated on a substratum of Vitrogen using Chee's Essential Medium has been reported to be responsive to xenobiotic treatment (Jauregui, H.O., Ng, S.F., Gann, K.L. and Waxman, D.J. (1991) Xenobiotica 21, 1091-1106). Class alpha, mu and pi glutathione S-transferase (GST) gene expression in response to xenobiotic treatment using this primary hepatocyte culture system was examined and the results compared with those obtained for P4502B1/B2 and 1A1 expression. Cytosolic GST activity decreased approx. 75% during the first 48 h of culture relative to freshly isolated hepatocytes and subsequently, increased, attaining a level at 96 h that was 134% of the activity at 48 h post-plating. Treatment of the hepatocyte cultures with phenobarbital (2 mM) or 3-methylcholanthene (5 microM) for 24, 48, or 72 h, beginning 24 h after plating, resulted in significant increases in glutathione S-transferase activity relative to control, with maximal increases of 158 and 164% measured at 72 h following phenobarbital or 3-methylcholanthrene treatment, respectively. SDS-PAGE analysis of cytosolic proteins showed a substantial increase in the intensities of protein bands migrating in the region of the GSTs following phenobarbital, beta-naphthoflavone or 3-methylcholanthrene treatment. Immunoblot analysis of cytosolic fractions using affinity-purified class-specific GST IgGs confirmed that alpha, mu and pi-class GST isozymes were elevated approx. 1.5- to 2-fold following phenobarbital, or beta-naphthoflavone treatment; 3-methylcholanthrene was less effective in enhancing GST expression in cultured hepatocytes as compared to phenobarbital or beta-naphthoflavone. Although GST pi was below the limit of detection in freshly-isolated hepatocytes, enhanced expression of this form was observed in untreated hepatocytes cultured for longer than 72 h. Immunoblot analysis of microsomal fractions revealed that cytochrome P-4502B1/2B2 and 1A1 levels were increased significantly in hepatocyte cultures treated with phenobarbital or 3-methylcholanthrene, respectively, relative to the undetectable levels found in untreated controls. Northern blot analysis of poly(A)+ mRNA isolated from cultures that had been treated with phenobarbital or 3-methylcholanthrene showed an approx. 2- and 4-fold increase in the expression of alpha and pi class glutathione S-transferase mRNAs, respectively, as compared to untreated cells. The level of P-4501A1 or 2B1 mRNA was also markedly elevated following 3-methylcholanthrene or phenobarbital treatment, respectively. The results of this study demonor the first time, that expression of alpha, mu and pi-class glutathione S-transferase genes is effectively modulated in primary yet culture system by different classes of xenobiotics.

Glutathione S-transferases and gamma-glutamyl transpeptidase in the rat nervous systems: a basis for differential susceptibility to neurotoxicants.

Glutathione and its related enzymes play a major role in the detoxification of toxic chemicals. In rat brain the pattern of distribution of reduced glutathione exhibits cellular heterogeneity, suggesting also the possibility of cellular differences in glutathione conjugating capacity. To understand the potential role of GSH in detoxification of neurotoxicants, the distributions of the glutathione conjugating and metabolizing enzymes, glutathione S-transferase (GST; alpha-, mu- and pi-classes) and gamma-glutamyl transpeptidase (gamma-GT) were determined immunohistochemically in brain, lumbar spinal cord and dorsal root ganglia (DRG) of adult Sprague-Dawley rats using polyclonal antibodies. The influence of tissue fixation on apparent distribution was also examined. Glial cells and neurons throughout the nervous system were only weakly positive with alpha-GST in frozen sections. No immunoreactivity for the alpha-class GSTs was observed in any of the paraformaldehyde-fixed neural specimens examined. In microwave-fixed frozen sections, immunoreactivity to mu-GST was found in astrocytes and neurons throughout the brain and spinal cord, and in the neurons and satellite cells of the DRG. Immunoreactivity for pi-GST was seen in oligodendrocytes but not in astrocytes in any region of the CNS examined. Similarly, satellite cells of the DRG were positive for pi-GST. Neuronal perikarya of the entire neopallium, hippocampus, cerebellum, brainstem, spinal cord and DRG were also positively stained for pi-GST. The differential staining of astrocytes and oligodendrocytes with pi- and mu-GST was unaltered in paraformaldehyde fixed tissues, but the neuronal immunostaining was lost. The ependyma, pia and choroid plexus stained positively with all three GST antibodies regardless of fixation. Gamma-Glutamyl transpeptidase-like immunoreactivity was confined to non-neuronal elements of both central and peripheral nervous systems. Ependymal cells throughout the central nervous systems stained intensely with antibodies directed against gamma-GT. Satellite and Schwann cells of the DRG and glial cells of the spinal cord and brain exhibited moderate to intense immunoreactivity for gamma-GT. The heterogeneous cellular distribution of glutathione and its metabolizing enzymes may reflect cellular differences in capacity for metabolic processing of both endogenous compound and xenobiotics.

Redox regulation of genes that protect against carcinogens.

Most carcinogens require activation to electrophilic metabolites or species that generate reactive oxygen in order to initiate the tumorigenic process. These reactive intermediates can, in turn, be detoxified by endogenous enzyme systems that and in the protection of cells from either toxic or mutagenic product formation. The levels of many of these enzymes are elevated by numerous compounds found in the diet, or by antioxidants. Recent evidence describes the mechanism for this induction of carcinogen detoxication enzymes to be regulated at the transcriptional level. Nuclear transcription factors bound to sites common among these carcinogen detoxication genes are activated by as yet unknown signal transduction pathways. The activity of these nuclear transcription factors are modulated by pro- and antioxidant reagents, suggesting that a redox-sensitive component governs the induction of enzymes involved in carcinogen metabolism. In this review, evidence for the redox regulation of the genes encoding carcinogen detoxication enzymes is presented. Evidence is also presented suggesting the participation of nuclear factor kappa B (NF-kappa B), mitogen-activated protein (MAP) kinase, and basic leucine zipper (bZIP) proteins and their activation pathways in this induction.

Signal transduction events elicited by natural products: role of MAPK and caspase pathways in homeostatic response and induction of apoptosis.

Many natural products elicit diverse pharmacological effects. Using two classes of potential chemopreventive compounds, the phenolic compounds and the isothiocyanates, we review the potential utility of two signaling events, the mitogen-activated protein kinases (MAPKs) and the ICE/Ced-3 proteases (caspases) stimulated by these agents in mammalian cell lines. Studies with phenolic antioxidants (BHA, tBHQ), and natural products (flavonoids; EGCG, ECG, and isothiocyanates; PEITC, sulforaphane), provided important insights into the signaling pathways induced by these compounds. At low concentrations, these chemicals may activate the MAPK (ERK2, JNK1, p38) leading to gene expression of survival genes (c-Fos, c-Jun) and defensive genes (Phase II detoxifying enzymes; GST, QR) resulting in survival and protective mechanisms (homeostasis response). Increasing the concentrations of these compounds will additionally activate the caspase pathway, leading to apoptosis (potential cytotoxicity). Further increment to suprapharmacological concentrations will lead to nonspecific necrotic cell death. The wider and narrow concentration ranges between the activation of MAPK/gene induction and caspases/cell death exhibited by phenolic compounds and isothiocyanates, respectively, in mammalian cells, may reflect their respective therapeutic windows in vivo. Consequently, the studies of signaling pathways elicited by natural products will advance our understanding of their efficacy and safety, of which many may become important therapeutic drugs of the future.

Purification and characterization of class mu glutathione S-transferase isozymes from rabbit hepatic tissue.

Class mu glutathione S-transferases (GSTs) are important in the detoxication of epoxides generated by oxidative metabolism. Phenobarbital, 3-methylcholanthrene, and pyridine have failed to enhance the expression of class mu GST isozymes in rabbit hepatic tissue (T. Primiano, S. G. Kim, and R. F. Novak, Toxicol. Appl. Pharmacol., 113, 64-73, 1992). Two class mu GST isozymes have been isolated from rabbit hepatic cytosol and purified to homogeneity using S-hexylglutathione-agarose, CM-Sepharose, and PBE94 chromatofocusing chromatography. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblot analyses showed that both isozymes possessed M(r) values of approximately 25,500 and cross-reacted with class mu-specific GST IgG. Gel filtration analysis revealed that these isozymes were dimers with molecular weights of approximately 45 kDa. The class mu GST isozymes had pIs of 7.8 and 7.2 as determined by nonequilibrium pH gel electrophoresis. The class mu GST 7.8 and 7.2 isozymes exhibited different metabolic activities toward the substrates 1-chloro-2,4-dinitrobenzene, bromosulfophthalein, 1,2-epoxy-3-(p-nitrophenoxy)propane, trans-4-phenyl-3-buten-2-one, p-nitrobenzyl chloride, and 3,4-dichloronitrobenzene. Metabolic activity of the two GSTs toward the substrate 1-chloro-2,4-dinitrobenzene was inhibited by Cibacron blue, triethyltin bromide, S-hexylglutathione, bromosulfophthalein, and indomethacin. The amino acid composition of GST mu 7.8 and 7.2 was determined and found to be very similar to those of purified rat class mu GST isozymes. N-terminal analysis of the first 21 residues of the pI 7.8 class mu GST isozyme revealed that it had 71 and 81% sequence identity with the Yb1 and Yb2 subunits, respectively. Similarly, N-terminal analysis of the first 21 residues of the pI 7.2 class mu GST isozyme revealed a 75% sequence identity with either the rat Yb1 or Yb2 subunit. Examination of class mu GST expression in rabbit hepatic cytosol following treatment with a series of known inducers including phenobarbital, 3-methylcholanthrene, isosafrole, pyrazine, trans-stilbene oxide, butylated hydroxyanisole, and tert-butylhydroquinone was accomplished. The data show that these agents not only failed to enhance class mu GST expression, but that 3-methylcholanthrene and isosafrole caused suppression of class mu GSTs. These results provide evidence for the existence of two closely related class mu GST isozymes in rabbit hepatic tissue and suggest that the molecular mechanisms regulating GST expression differ between rat and rabbit in response to these xenobiotics.

Enhanced expression, purification, and characterization of a novel class alpha glutathione S-transferase isozyme appearing in rabbit hepatic cytosol following treatment with 4-picoline.

A novel class alpha glutathione S-transferase (GST) isozyme is expressed in the hepatic cytosol of rabbits treated with 4-picoline. SDS-PAGE analysis revealed the presence of a new 28-kDa band which cross-reacted with class alpha GST-specific IgG. This new GST isozyme was isolated from the hepatic cytosol of 4-picoline-treated rabbits and purified to homogeneity using S-hexylglutathione-agarose, CM-Sepharose, and PBE118 chromatofocusing chromatography. The isozyme was determined by SDS-PAGE and gel filtration analyses to be a homodimer of approximately 28 kDa with blocked N-terminus. A heterodimer consisting of 25 and 28 kDa subunits with activity toward the substrate 1-chloro-2,4-dinitrobenzene was also purified. Immunoblot analysis revealed that the 25, 26.5, and 28 kDa bands cross-reacted with class alpha GST-specific IgG and failed to react with either class mu or class pi GST-specific antibodies. The 28 kDa enzyme had a pI of 8.2 as determined by nonequilibrium pH gel electrophoresis. The purified 28 kDa enzyme exhibited activity toward 1-chloro-2,4-dinitrobenzene (Km = 1.60 mM and Vmax = 73.5 mumol/min/mg) and cumene hydroperoxide (Km = 1.02 mM and Vmax = 6.92 mumol/min/mg). Amino acid sequence analysis of several fragments resulting from cyanogen bromide cleavage of the 28 kDa GST isozyme revealed a class alpha GST consensus sequence. In addition, proteolytic digestion with alpha-chymotrypsin yielded peptide maps which showed distinct differences between the purified 28 kDa GST and another purified class alpha GST isozyme present in rabbit liver. These results provide evidence that class alpha GST isozymes containing a novel 28 kDa subunit are expressed following treatment with 4-picoline.

Differences between rats and rabbits in hepatic cytosolic glutathione S-transferase expression in response to nitrogen heterocycle and other inducers.

Glutathione S-transferase (GST) expression was examined in hepatic cytosol from rats and rabbits treated with 4-picoline, pyrrole, pyridine, pyrazine, imidazole, or piperidine using enzymatic activity, SDS-PAGE, and immunoblot analyses and the results were compared to those obtained with phenobarbital and 3-methylcholanthrene. SDS-PAGE and immunoblot analyses of hepatic cytosol prepared from rats treated with pyrazine revealed the induction of class alpha (Ya and Yc) and mu (Yb) bands with a corresponding 2.4-fold increase in metabolic activity using 1-chloro-2,4-dinitrobenzene as substrate. A new class alpha band migrating in the region of the Yc band was observed in the SDS-PAGE and detected in the immunoblot of cytosol from pyrrole-treated rats, whereas treatment with 4-picoline, imidazole, or piperidine failed to alter the expression of the major classes of GST isozymes in this species. SDS-PAGE and immunoblot analyses of rabbit hepatic cytosol revealed a unique species-dependent difference in the expression of GSTs. While phenobarbital and 3-methylcholanthrene induce class alpha and mu GST expression in rat hepatic cytosol, one of the most interesting observations was that neither of these agents stimulated GST expression in the rabbit. Immunoblot analysis of cytosol isolated from 4-picoline-treated rabbits using GST class alpha-specific IgG showed the appearance of a novel class alpha 28-kDa GST band and the concomitant disappearance of a class alpha 29-kDa GST band. In addition, SDS-PAGE and immunoblot analyses showed that treatment of rabbits with pyrrole, pyrazine, imidazole, or piperidine resulted in the disappearance of this class alpha 29-kDa GST band with no detectable expression of the class alpha 28-kDa GST band; the level of the class alpha 29-kDa band was unaffected by pyridine treatment. In contrast, immunoblot analyses of hepatic cytosol revealed that a 25.5-kDa class mu GST band disappeared following treatment with pyridine, but was unaffected by treatment with other nitrogen heterocycles. The Vmax of glutathione conjugation to the substrate 1-chloro-2,4-dinitrobenzene decreased by 52, 36, 59, 41, 37, and 32% in hepatic cytosol isolated from 4-picoline-, pyrrole-, pyridine-, pyrazine-, imidazole-, and piperidine-treated rabbits, respectively. The results suggest that nitrogen heterocycles differ in their ability to modulate glutathione S-transferase isozyme expression in rat and rabbit hepatic tissue and that rabbit hepatic GSTs are refractory to induction by agents such as pyrazine, phenobarbital, or 3-methylcholanthrene and hence these xenobiotics do not appear to be bifunctional inducers in this species.

Enhanced glutathione S-transferase 7-7 expression in rat hepatic cytosol following treatment with pyrrole.

Rat hepatic GST 7-7 expression in cytosol and in the S-hexylglutathione-agarose affinity purified fraction of cytosol from saline- (control) and pyrrole-treated animals was examined using metabolic activity, SDS-PAGE, immunoblot, and HPLC analyses. Metabolic activity of hepatic cytosol from pyrrole-treated animals was assayed using the substrates 1-chloro-2,4-dinitrobenzene, ethacrynic acid, and acrolein, and an approximately 1.5-, 1.2-, and 1.3-fold increase, respectively, was monitored in the rate of GST-catalyzed substrate conjugation to reduced glutathione. SDS-PAGE and immunoblot analysis using Class Pi GST-specific IgG confirmed that the GST 7 subunit was expressed in hepatic cytosol and in the affinity purified fractions from pyrrole-treated rats. In contrast, the GST 7 subunit was below the limit of detection in hepatic cytosol of saline-treated animals. HPLC analysis demonstrated the presence of the GST 7 subunit in the affinity purified fraction from pyrrole-treated rat hepatic tissue and showed that the level of this subunit was increased approximately 8-fold relative to the barely detectable level present in control tissue. N-terminal amino acid sequencing analysis confirmed the identity of the HPLC peak as GST 7-7. GST subunits 3, 1a, and 1b were elevated approximately 1.7-, 2.0-, and 2.4-fold respectively, in response to pyrrole treatment. These data provide evidence that pyrrole treatment results in de novo expression of GST 7-7 in rat hepatic tissue and suggest that pyrrole-containing compounds potentially generated during disease and/or altered heme metabolism, or those ingested in food products, may alter GST expression in hepatic tissue.

Identification of dithiolethione-inducible gene-1 as a leukotriene B4 12-hydroxydehydrogenase: implications for chemoprevention.

Cancer chemoprevention is inhibition of neoplastic disease by naturally occurring or synthetic chemical agents. Dithiolethiones inhibit production of experimentally produced tumors by elevating the expression of several genes that encode for known cytoprotective enzymes. In an effort to discover additional molecular mechanisms mediating chemoprevention, cDNA clones representing a gene that is transcriptionally activated by dithiolethiones, hence named dithiolethione-inducible gene-1 (DIG-1), were isolated from rat liver via differential hybridization screening. The deduced amino acid sequence of DIG-1 was found to have 80% identity with the human liver enzyme leukotriene B4 (LTB4) 12-hydroxydehydrogenase. DIG-1, purified >400-fold from the liver of rats dosed with 1,2-dithiole-3-dithiolethione, possessed an NADP+-dependent activity to convert LTB4 to 12-oxo-LTB4. Kinetic analysis of DIG-1 revealed apparent Km and Vmax values of 28 mM and 8.1 nmol 12-oxo-LTB4 formed/min/mg purified protein respectively. Since LTB4 is a potent chemotactic factor and stimulator of production of reactive oxygen species from neutrophils, the effects of DIG-1 on these LTB4-mediated processes were examined. Pre-incubation of LTB4 with purified rat hepatic DIG-1 greatly diminished LTB4-stimulated migration of neutrophils. In addition, pre-incubation of LTB4 with purified rat hepatic DIG-1 reduced LTB4-stimulated production of superoxide anions in neutrophils, as evidenced by decreased lucigenin-derived chemiluminescence. These results suggest that DIG-1-catalyzed dehydrogenation of LTB4 to 12-oxo-LTB4 inhibits the pro-inflammatory actions of LTB4. Consequently, elevation of LTB4 catabolism via enhanced DIG-1 activity may suppress inflammatory processes implicated in tumorigenesis.

Differential expression of glutathione S-transferase isoforms in compartments of the testis and segments of the epididymis of the rat.

Specific cell types of the mammalian testes demonstrate varying susceptibility to toxic insult by chemical agents. The mammalian testis is divided into two major compartments: seminiferous tubules, the site of spermatogenesis, and interstitium, which contains the Leydig cells. Glutathione S-transferase (GST) expression was examined in isolated compartments of the rat testis and in segments of the epididymis. Western blot analysis revealed the presence of GST class alpha, mu, and pi bands in each of the isolated compartments of the testis, and HPLC analysis of monomeric isoforms provided evidence for differential expression of multiple GST isoforms in testicular compartments. All major isoforms (e.g., forms 1, 2, 3, 4, 6, 7, 8, 9, and 11) were detected in the cytosol of whole testis. Isoform subunit 4 was the major form in the tubule, whereas isoform subunit 11 is the dominant form in the Leydig cells. Isoform subunits 3, 4, and 6 were enriched in the tubules as compared to interstitial or Leydig cells. The preferential action of reproductive toxicants at specific stages of aging may be due to an age-dependent expression of the activating or detoxifying enzymes in the reproductive tract. Therefore, the age-dependent expression of testicular GST isoforms was also examined. Expression of isoform subunits 2 and 4 displayed an age dependence, with the largest increase in these subunits occurring between ages 4 and 15 weeks. Isoform expression did not correlate with serum testosterone levels. HPLC analysis of the GST isoforms in the longitudinal segments of the epididymis and vas deferens revealed differential expression within these segments. Total GST protein and catalytic activity was highest in the caput epididymis and progressively decreased toward the vas deferens. Isoform subunit 2 was the major form expressed in the epididymis. The results of this study indicate that the GSTs are differentially expressed in testicular compartments and epididymal segments, and that this may contribute to susceptibility of different cell types to xenobiotic damage.

Induction of hepatic heme oxygenase-1 and ferritin in rats by cancer chemopreventive dithiolethiones.

Treatment of rats with the cancer chemopreventive agent 1,2-dithiole-3-thione (D3T) resulted in a significant increase in hepatic heme oxygenase (HO) activity, which corresponded to increased protein levels of HO-1. Upon further analysis of proteins related to heme metabolism, the level of ferritin, the major iron storage protein in liver, was also found to be elevated. Diminished levels of intracellular free iron were monitored by EPR spectroscopy at times after administration of D3T that suggested that increased ferritin content sequesters intracellular iron. The increased levels of protein were associated with increased levels of steady-state RNA of HO-1 and the light (FL) and heavy (FH) subunits of ferritin. A direct relationship between enhanced rates of gene transcription and elevated levels of HO-1 and ferritin RNA was found. The inductions of FL and FH, but not HO-1, were sensitive to cycloheximide, suggesting that in vivo these genes are regulated by distinct D3T-responsive transcriptional mechanisms. The known protective roles for induced HO-1 and ferritin in cellular stress have been suggested to include increased levels of the antioxidant bilirubin and enhanced sequestration of intracellular iron into ferritin, which can effectively reduce iron-mediated reactive oxygen generation. Thus, protective actions of D3T against the cytotoxic and carcinogenic consequences of chemicals that exert electrophilic or oxidative stresses may be mediated, in part, by the induction of HO-1, FL and FH.

Isolation of cDNAs representing dithiolethione-responsive genes.

Dithiolethiones inhibit tumorigenicity elicited by many structurally diverse carcinogens in numerous target tissues. These protective actions are associated with the induction of several carcinogen detoxification enzymes, some of which have only recently been discovered. In order to identify additional novel inducible detoxification response genes, a cDNA library was prepared from liver of rats treated with 1,2-dithiole-3-thione (D3T) and was screened by a differential hybridization method. Complementary DNA clones for several known D3T-inducible genes were isolated, such as epoxide hydrolase, aflatoxin B1-aldehyde reductase, quinone reductase and multiple subunits of glutathione S-transferase. Clones representing genes not previously associated with detoxification were isolated, including those for ferritin heavy and light subunits, ribosomal proteins L18a and S16 and two novel genes, termed dithiolethione-inducible genes (or DIG-1 and DIG-2). Levels of mRNA recognized by each clone were increased from 2- to 31-fold, with maximum induction between 6 and 30 h after treatment with D3T. Except for epoxide hydrolase, the kinetics of induction of each mRNA was coordinate with increased rates of gene transcription. However, based on the time of response to D3T, at least two sets of responsive genes were identified. One set of genes, including glutathione S-transferase Yp, aflatoxin B1-aldehyde reductase, quinone reductase and DIG-1, had low constitutive and highly inducible expression (approximately 20-fold) and the other, including glutathione S-transferase Ya and Yb, epoxide hydrolase, ferritin heavy and light subunits, ribosomal proteins L18a and S16 and DIG-2, had relatively high constitutive and modestly inducible expression (approximately 5-fold). The simplest explanation for this differential expression of D3T-inducible genes is that multiple regulatory mechanisms govern their response. The transcriptional activation of ferritin, ribosomal protein, DIG-1 and DIG-2 genes in conjunction with those of carcinogen detoxification enzymes suggests that they participate in the pleiotropic cellular defense response to dithiolethiones that inhibits chemically produced tumorigenesis.

cDNA cloning, expression and activity of a second human aflatoxin B1-metabolizing member of the aldo-keto reductase superfamily, AKR7A3.

The aflatoxin B1 (AFB1) aldehyde metabolite of AFB1 may contribute to the cytotoxicity of this hepatocarcinogen via protein adduction. Aflatoxin B1 aldehyde reductases, specifically the NADPH-dependent aldo-keto reductases of rat (AKR7A1) and human (AKR7A2), are known to metabolize the AFB1 dihydrodiol by forming AFB1 dialcohol. Using a rat AKR7A1 cDNA, we isolated and characterized a distinct aldo-keto reductase (AKR7A3) from an adult human liver cDNA library. The deduced amino acid sequence of AKR7A3 shares 80 and 88% identity with rat AKR7A1 and human AKR7A2, respectively. Recombinant rat AKR7A1 and human AKR7A3 were expressed and purified from Escherichia coli as hexa-histidine tagged fusion proteins. These proteins catalyzed the reduction of several model carbonyl-containing substrates. The NADPH-dependent formation of AFB1 dialcohol by recombinant human AKR7A3 was confirmed by liquid chromatography coupled to electrospray ionization mass spectrometry. Rabbit polyclonal antibodies produced using recombinant rat AKR7A1 protein were shown to detect nanogram amounts of rat and human AKR7A protein. The amount of AKR7A-related protein in hepatic cytosols of 1, 2-dithiole-3-thione-treated rats was 18-fold greater than in cytosols from untreated animals. These antibodies detected AKR7A-related protein in normal human liver samples ranging from 0.3 to 0.8 microg/mg cytosolic protein. Northern blot analysis showed varying levels of expression of AKR7A RNA in human liver and in several extrahepatic tissues, with relatively high levels in the stomach, pancreas, kidney and liver. Based on the kinetic parameters determined using recombinant human AKR7A3 and AFB1 dihydrodiol at pH 7.4, the catalytic efficiency of this reaction (k2/K, per M/s) equals or exceeds those reported for other enzymes, for example cytochrome P450s and glutathione S-transferases, known to metabolize AFB1 in vivo. These findings indicate that, depending on the extent of AFB1 dihydrodiol formation, AKR7A may contribute to the protection against AFB1-induced hepatotoxicity.

Changes in adult metabolism of aflatoxin B1 in rats neonatally exposed to diethylstilbestrol. Alterations in alpha-class glutathione S-transferases.

Neonatal exposure of rats to xenobiotics has been shown to produce long-term alterations in hepatic enzyme activities and in levels of DNA adducts following carcinogen exposure. We exposed newborn male rats to diethylstilbestrol (DES), pregnenolone-16 alpha-carbonitrile, 7,12-dimethylbenz[a]anthracene or phenobarbital on days 1, 3 and 5 of age. At five months of age, males were injected with 1 mg/kg of [3H]aflatoxin B1 (AFB1), killed after 2 h and examined for AF-DNA adduction in the liver. Males neonatally exposed to DES showed a 35% decrease in DNA adduction levels. Analysis of the adducted DNA bases failed to show any changes in relative proportions of individual adducts in the DES samples compared to controls. Hepatic glutathione concentrations were unchanged. However, Western blot analysis of alpha-class glutathione S-transferases (alpha GST), enzymes known to inactivate the toxic AFB1-8,9-epoxide, showed a 2-fold increase in subunit levels in the DES-treated males, suggesting that the detoxifying activity of the cytosol may have been increased. To confirm this, in vitro tests were undertaken using butylated hydroxyanisole (BHA) induced mouse microsomes to activate [3H]AFB1 in the presence of treated cytosol and GSH. Analysis of metabolites by HPLC showed that DES-treated males formed 245% of the AFB-SG conjugate relative to vehicle controls. These results indicate that neonatal DES treatment resulted in long-term changes in basal alpha GST levels and suggest that these changes were responsible for lower levels of DNA adduction following adult exposure to AFB1.