Mouse liver glutathione S-transferase isoenzyme activity toward aflatoxin B1-8,9-epoxide and benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide

As part of the studies of the biochemical basis for species differences in biotransformation of the carcinogen aflatoxin B1 (AFB1) and its modulation by phenolic antioxidants, we have investigated the role of mouse liver glutathione S-transferase (GST) isoenzymes in the conjugation of AFB1-8,9-epoxide. Isoenzymes of GST were purified to electrophoretic homogeneity from Swiss-Webster mouse liver cytosol by affinity chromatography and chromatofocusing. The isoenzyme fractions were characterized in terms of activity toward surrogate substrates and immunologic cross-reactivity with antisera to rat GSTs. The major isoenzymes were identified as SW 4-4, SW 3-3, and SW 1-1. The specific activity of SW 4-4 toward AFB1-8,9-epoxide was at least 50- and 150-fold greater than that of SW 3-3 and SW 1-1, respectively. Relatively high activity toward another epoxide carcinogen, benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide, was observed with both SW 4-4 and SW 3-3. SW 1-1 had the highest activity toward 1-chloro-2,4-dinitrobenzene (CDNB) whereas SW 4-4 had relatively low CDNB activity. Following pretreatment with 0.75% butylated hydroxyanisole in the diet, the fraction of total GST contributed by SW 1-1 appeared to increase dramatically, whereas in control mice SW 3-3 constituted the predominant isoenzyme. The high GST activity of mouse liver cytosol toward AFB1-8,9-epoxide is apparently due to an isoenzyme that contributes little to the overall cytosolic CDNB activity.     

Comparative activities of glutathione-S-transferase and dialdehyde reductase toward aflatoxin B1 in livers of experimental and farm animals.

In order to gain a better understanding of the relative activities of glutathione-S-transferase (GST) and aldehyde reductase toward aflatoxin B1 (AFB1) in relation to the variation of species susceptibilities, we studied the in vitro cytosolic GST and reductase activities in liver tissues from male Fischer rats, ICR mice and golden hamsters, adult male rainbow trouts and female piglets. The GST activity was determined by incubating the liver cytosol with glutathione (GSH) and AFB1 in the presence of the hamster liver microsomes to metabolize AFB1 to AFB1-8, 9-epoxide. The reaction product, AFB1 and GSH conjugate (AFB1-GSH), was quantified with HPLC. The reductase activity was determined by incubating liver cytosol with AFB1-dialdehyde, followed by the quantification of the metabolic product, AFB1-dialcohol, with HPLC. All the animal species possessed the GST activities, and AFB1-GSH formed increasingly with the increase of the AFB1 concentration according to the model of first-order enzyme reaction kinetics. The V(max) and K(m) values of the GST activities in rodent species were higher and lower, respectively, than those in the trout and pig, being consistent with the relative susceptibilities to AFB1 of these animal species. However, no relationship was noted between the reductase activity and species susceptibility. Thus, the result of this study shows that GST toward AFB1, but not aldehyde reductase, is a determinant of the variation of species susceptibilities.     

Protective activity of different hepatic cytosolic glutathione S-transferases against DNA-binding metabolites of aflatoxin B1

To evaluate the role of glutathione S-transferase (GST) isoenzymes in induced resistance of hepatocytes to aflatoxin B1 (AFB1), we compared DNA protective activities of different hepatic cytosol preparations and purified GSTs from normal rats, rats exposed to different polychlorinated biphenyls (PCBs), and rats with carcinogen-induced hepatocellular neoplasms, with cytosols or purified GSTs from mouse, rainbow trout, and human livers. These comparisons were performed in an in vitro assay for [3H]AFB1-DNA binding after activation by rat liver microsomes. Cytosol and S-hexylglutathione-affinity-purified GST preparations from livers of mice consistently had strong protective activity against AFB1-DNA binding. The majority of this activity was dependent on the presence of reduced glutathione (GSH) but some GSH-independent protection was observed in mouse hepatic cytosol, but not in purified GST preparations. We found that all of the GSH-dependent DNA-protective activity in mouse liver eluted as a single GST isoenzyme by hydroxyapatite chromatography. Preparations of cytosol and purified GSTs from normal rat liver, rainbow trout liver, and human liver had much less AFB1-specific DNA protective activity than GSTs found in mouse liver preparations. Cytosol from rats with carcinogen-generated liver neoplasms and livers induced with 3,3',4,4'-tetrachlorobiphenyl and 2,2',4,4',5,5'-hexachlorobiphenyl had more GST activity toward CDNB than cytosol from normal rat liver. When equivalent units of GST activity (CDNB) were compared, there was little difference observed between the DNA-protective activities of PCB-induced and normal rat liver cytosols, yet cytosol from rat liver neoplasms was more protective. Purified GST-P (7-7), the GST isoenzyme most induced in carcinogen-generated rat liver neoplasms, was not protective when added at protein concentrations found to be protective for total GSTs isolated from these neoplasms. These studies demonstrate that the resistance of mouse liver to AFB1 can be explained primarily by a single constitutive GST isoenzyme (YaYa or 4-4) with a relatively high activity toward DNA-binding metabolites of AFB1. GST isoenzymes with such high specific DNA protective activity against AFB1 metabolites were not evident in human, rat, or rainbow trout liver or in PCB-induced or neoplastic rat liver preparations.     

Mouse strain differences in glutathione S-transferase activity and aflatoxin B1 biotransformation

Previous studies have suggested that mice are resistant to the carcinogenic effects of aflatoxin B1 (AFB1) and that this resistance is largely the result of expression of an isoenzyme of glutathione S-transferase (GST) with high activity toward AFB1-8,9-epoxide. Significant interstrain differences in cytosolic GST activities toward a variety of substrates have been reported in mice. If such differences exist for the conjugation of AFB1-8,9-epoxide, then there may be significant mouse strain differences in susceptibility to AFB1-induced hepatocarcinogenicity. The hepatic microsomal and cytosolic biotransformation of AFB1 was studied in 8 different strains of mice fed a purified diet. GST-mediated conjugation of AFB1-8,9-epoxide with glutathione and GST activity toward 1-chloro-2,4-dinitrobenzene (CDNB), 1,2-dichloro-4-nitrobenzene (DCNB), ethacrynic acid (ECA) and cumene hydroperoxide (CHP) were determined with cytosolic fractions from 8-10 pooled livers. Specific activities of cytochrome-P-450-mediated oxidation of AFB1 to aflatoxin Q1 (AFQ1), aflatoxin M1 (AFM1), and aflatoxin P1 (AFP1), as well as the reactive intermediate AFB1-8,9-epoxide, were determined with hepatic microsomal fractions from each mouse strain. No striking differences in specific activity between mouse strains were observed for any of the P-450- or GST-mediated enzymatic pathways measured, although some statistically significant differences were found. GST specific activities toward AFB1-8,9-epoxide, CDNB, DCNB, ECA and CHP ranged from 1.5-2.1, 2,830-5,370, 81-144, 38-69 and 32-73 nmol/mg protein/min, respectively. The rate of formation of AFB1-8,9-epoxide ranged from 208 to 465 pmol/mg protein/min. The specific activities of AFQ1,AFM1, and AFP1 formation by microsomes ranged from 36-70, 161-326, and 252-426 pmol/mg protein/min, respectively. Mice fed a standard rodent chow diet showed evidence of microsomal and cytosolic enzyme induction when compared to mice fed a purified diet. The lack of substantial differences in enzyme specific activities between mouse strains suggests that interstrain variations in the hepatocarcinogenic effects of AFB1 in mice should not be large.     

Complementary DNA cloning, protein expression, and characterization of alpha-class GSTs from Macaca fascicularis liver.

Large species differences exist in sensitivity to aflatoxin B(1) (AFB(1))-induced liver cancer. Mice are resistant to AFB(1)-induced liver cancer because they express an alpha-class GST (mGSTA3-3) that has high activity toward the reactive intermediate aflatoxin B(1)-8,9-epoxide (AFBO). Rats constitutively express only small amounts of a GST with high AFBO activity (rGSTA5-5) and thus are sensitive to AFB(1)-induced hepatocarcinogenesis, although induction of rGSTA5-5 can confer resistance in rats. In contrast to rodents, constitutively expressed human hepatic alpha-class GSTs have little or no AFBO detoxifying activity. Recently, we found that the nonhuman primate, Macaca fascicularis (Mf), has significant constitutive hepatic GST activity toward AFBO and most of this activity belongs to mu-class GSTs. To determine if any alpha-class GSTs in Mf liver have AFBO activity, a cDNA library from a male Mf liver was constructed and screened using the human alpha-class GstA1 cDNA as a probe. Three different cDNA clones with full-length open reading frames were identified from the Mf hepatic cDNA library. Analyses of the cDNA deduced protein sequences indicated that these three alpha-class cDNA clones were 97-98% homologous with each other, and shared 93, 95, and 95% identity with human GSTA1, and were named mfaGSTA1, mfaGSTA2, and mfaGSTA3, respectively. Bacterially expressed mfaGSTA1-1 recombinant protein had similar activities toward classic GST substrates such as DCNB, CHP, and ECA, but slightly lower CDNB conjugating activity relative to human GSTA1-1. However, similar to hGSTA1-1, mfaGSTA1-1 had no AFBO conjugating activity. In addition, similar to human GSTA1 gene, cDNA-derived amino acid sequence analyses demonstrated that all of these Mf alpha-class GSTs genes (mfaGSTA1, mfaGSTA2, and mfaGSTA3) had none of the six critical residues that were identified previously to confer high AFBO activity in mouse alpha-class GSTA3-3. Thus, in contrast to rodents but similar to humans, alpha-class GSTs from the nonhuman primate, Mf, have little conjugating activity toward AFBO.     

Metabolism of aflatoxins: key enzymes and interindividual as well as interspecies differences

Aflatoxins are potent hepatocarcinogen in animal models and suspected carcinogen in humans. The most important aflatoxin in terms of toxic potency and occurrence is aflatoxin B1 (AFB1). In this review, we mainly summarized the key metabolizing enzymes of AFB1 in animals and humans. Moreover, the interindividual and the interspecies differences in AFB1 metabolism are highly concerned. In human liver, CYP3A4 plays an important role in biotransforming AFB1 to the toxic product AFB1-8,9-epoxide. In human lung, CYP2A13 has a significant activity in metabolizing AFB1 to AFB1-8,9-epoxide and AFM1-8,9-epoxide. The epoxide of AFB1-8,9-epoxide could conjugate with glutathione to reduce the toxicity by glutathione-S-transferase (GST). In poultry species, CYP2A6, CYP3A37, CYP1A5, and CYP1A1 are responsible for bioactivation of AFB1. There are interindividual variations in the rate of activation of aflatoxins in various species, and there are also differences between children and adults. The age and living regions are important factors affecting resistance of species to AFB1. The rate of AFB1-8,9-epoxide formation and its conjugation with glutathione are key parameters in interspecies and interindividual differences in sensitivity to the toxic effect of AFB1. This review provides an important information for key metabolizing enzymes and the global metabolism of aflatoxins in different species.     

Effects of modulation of hepatic glutathione on biotransformation and covalent binding of aflatoxin B1 to DNA in the mouse

Relative to the rat and most other species tested, the mouse is resistant to the carcinogenic effects of aflatoxin B1 (AFB). Previous investigations in our laboratory demonstrated that mouse liver cytosol has 52 times greater hepatic glutathione S-transferase (GST) activity toward the AFB-epoxide, compared with rat liver cytosol. To determine the importance of GST-mediated detoxification of the AFB-epoxide in the mouse in vivo, we examined the effects of glutathione (GSH) depletion on the covalent binding of AFB to hepatic DNA in control and 2(3)-butyl-4-hydroxyanisole (BHA)-treated mice. Male Swiss-Webster mice were fed control or 0.75% BHA diet for 14 days. Depletion of hepatic GSH was accomplished with D,L-buthionine-S-sulfoximine (BSO, 0.6 g/kg in saline) and diethyl maleate (DEM, 0.75 ml/kg), administered by ip injection at 2 and 1.5 hr, respectively, prior to administration of 3H-AFB (0.25 mg/kg, ip in DMSO). The combined BSO-DEM treatment depleted hepatic GSH by 97 and 70% in control and BHA-treated mice, respectively. In mice receiving the control diet, GSH depletion was associated with a 30-fold increase in the covalent binding of AFB to hepatic DNA. AFB-DNA binding in mice treated with dietary BHA alone was reduced to 54% of control. In BHA-treated mice, BSO-DEM treatment increased AFB-DNA binding by 62%. Dietary BHA increased hepatic S-9 mediated activation of AFB to the AFB-epoxide by eightfold in both control and BSO-DEM mice. BHA also increased GST activity toward the AFB-epoxide by 52 and 68% in control and BSO-DEM mice, respectively. The BSO-DEM treatment alone had no significant effect on the in vitro biotransformation of AFB. These results support the hypothesis that GST is the principal determinant of species differences in susceptibility to AFB-induced hepatocarcinogenicity. The results also support the hypothesis that BHA may protect against the toxic and carcinogenic effects of xenobiotics in part by preventing the depletion of hepatic GSH.     

Biochemical basis for the extreme sensitivity of turkeys to aflatoxin B(1)

Poultry are the most susceptible food animal species to the toxic effects of the mycotoxin aflatoxin B(1) (AFB(1)). Feed contaminated with even small amounts of AFB(1) results in significant adverse health effects in poultry. The purpose of this study was to explain the biochemical mechanism(s) for this extreme sensitivity. We measured microsomal activation of AFB(1) to the AFB(1)-8,9-epoxide (AFBO), the putative toxic intermediate, as well as cytosolic glutathione S-transferase (GST)-mediated detoxification of AFBO, in addition to other hepatic phase I and phase II enzyme activities, in 3-week-old male Oorlop strain turkeys. Liver microsomes prepared from these turkeys activated AFB(1) in vitro with an apparent K(m) of 109 microM and a V(max) of 1.25 nmol/mg/min. Preliminary evidence for the involvement of cytochromes P450 (CYP) 1A2 and, to a lesser extent, 3A4 for AFB(1) activation was assessed by the use of specific mammalian CYP inhibitors. The possible presence of avian orthologues of these CYPs was supported by activity toward ethoxyresorufin and nifedipine, as well as by Western immunoblotting using antibodies to human CYPs. Cytosol prepared from turkey livers exhibited GST-mediated conjugation of 1-chloro-2,4-dinitrobenzene (CDNB) and 3,4-dichloronitrobenzene (DCNB), but at a much lower rate than that observed in other species. Western immunoblotting indicated the presence of alpha and sigma class GSTs and another AFB(1)-detoxifying enzyme, AFB(1)-aldehyde reductase (AFAR). Turkey liver cytosol also had quinone oxidoreductase (QOR) activity. Importantly, cytosol exhibited no measurable GST-mediated detoxification of microsomally activated AFB(1), indicating that turkeys are deficient in the most crucial AFB(1)-detoxification pathway. In total, our data indicate that the extreme sensitivity of turkeys to AFB(1) may be attributed to a combination of efficient AFB(1) activation and deficient detoxification by phase II enzymes, such as GSTs.     

Conjugation of model substrates or microsomally-activated aflatoxin B1 with reduced glutathione, catalysed by cytosolic glutathione-S-transferases in livers of rats, mice and guinea pigs

Glutathione-S-transferase (GST) activity has been examined in liver cytosol fractions from guinea pigs, mice, control fed rats or rats with pre-neoplastic nodular liver lesions. The levels of activity in unfractionated cytosols have been assayed using the model substrates 1-chloro-2,4-dinitrobenzene (CDNB), 3,4-dichloronitrobenzene (DCNB) and monobromobimane (mBrB) with reduced glutathione (GSH). The order of activities in the various liver fractions using CDNB as substrate were: mouse greater than pre-neoplastic nodular rat greater than guinea pig greater than control rat and paralleled the capacities of the cytosols to catalyse the formation of aflatoxin B1-GSH from microsomally-activated aflatoxin B1 (AFB1) and GSH. Quantitative differences between the activities of the cytosols using the three model substrates were observed. In the mouse fractionation of GST activity by isoelectric focusing (I.E.F.) on preparative granular gels showed that the most basic component (isoelectric point pH 9.4) with the highest conjugating activity with respect to microsomally-activated AFB1 did not correspond with the peak of most activity for conjugating CDNB. In the pre-neoplastic nodular rat liver the CDNB conjugating activities of all fractions separated on granular I.E.F. gels, were higher than the corresponding fractions isolated from control rat liver, with particular enhancement of the peak containing the 3:3 isoenzyme. In contrast to control rat liver the 7:7 isoenzyme was detected in pre-neoplastic nodular liver preparations. These isoenzymes (3:3 and 7:7) did not contribute significantly to the enhanced level of AFB1-GSH formation catalysed by cytosol fractions prepared from pre-neoplastic nodular rat liver. The microsomally-activated AFB1-conjugating activity of unfractionated rat liver cytosols was increased to a relatively greater extent than CDNB conjugating activity during the induction of pre-neoplastic nodular liver lesions, and the elevated level of the activated AFB1-conjugating activity was found to be associated with the most basic fraction (isoelectric point pH 9.0). Analytical isoelectric focusing gels using mBrB as substrate demonstrated the presence of a basic GST isoenzyme in the pre-neoplastic nodular rat liver, not detected in preparations from the livers of control rats. The low level of activated AFB1-conjugating activity present in unfractionated guinea-pig cytosol was found to correspond with the fraction containing the peak of CDNB conjugating activity on preparative isoelectric focusing (isoelectric point pH 7.5). The lack of correlation between the conjugation of model substrates and the conjugation of xenobiotics could be of import     

Comparative effects of butylated hydroxyanisole on hepatic in vivo DNA binding and in vitro biotransformation of aflatoxin B1 in the rat and mouse

To determine the mechanisms which mediate species- and treatment-related differences in susceptibility to aflatoxin B1 (AFB), we conducted a comparative study of the effects of dietary butylated hydroxyanisole (BHA) on the hepatic in vivo DNA binding and in vitro biotransformation of AFB in the rat and mouse. Mice are resistant to the hepatocarcinogenic effects of AFB, and BHA pretreatment has been shown to inhibit the carcinogenic effects of AFB in the highly susceptible rat. Rats and mice were fed a control diet or an identical diet containing 0.75% BHA for 10 days. On the 11th day, one-half of the control and BHA animals were administered [3H]AFB (0.25 mg/kg in dimethyl sulfoxide) via intraperitoneal injection. Animals were killed 2 hr later and covalent binding of AFB to hepatic DNA was determined. The remaining animals were killed for preparation of hepatic subcellular fractions used in in vitro assays. BHA treatment resulted in a decrease in in vivo hepatic AFB-DNA adduct formation in mice to 68% of control, but, in rats, treatment decreased AFB-DNA binding to 18% of control. Furthermore, hepatic AFB-DNA binding in control mice was only 1.2% of that measured in control rats. The rate of in vitro activation of AFB to the epoxide was 3.4-fold greater in control mice relative to control rats. BHA pretreatment increased the activation of AFB in mice 3.3-fold, but had no effect on oxidative metabolism in rats. Control mice had 52 times greater glutathione S-transferase (GST) activity toward the AFB-epoxide, but only 2.6 times greater GST activity toward 1-chloro-2,4-dinitrobenzene (CDNB), compared to that of control rats. In mice, BHA did not significantly increase GST activity toward the AFB-epoxide, but increased GST activity toward CDNB 3.1-fold. In rats, BHA increased GST activity toward the AFB-epoxide and CDNB by 3.2- and 2.1-fold, respectively. Epoxide hydrolase activity toward p-nitrostyrene oxide in mice was only 52% of the activity in rats. BHA increased epoxide hydrolase activity 3.8- and 2.5-fold in mice and rats, respectively. These data indicate that mice have high levels of an AFB-epoxide-specific GST activity relative to that of the rat. The rate of formation of the AFB-epoxide and the activity of epoxide hydrolase appear to be relatively unimportant under conditions of high GST activity, whereas elevated GST activity, and thus inactivation of the AFB-epoxide, appears to be the critical component in species- and BHA-induced differences in AFB-DNA adduct formation and, presumably, AFB hepatocarcinogenicity.     

The effect of feeding piglets with the diet containing green tea extracts or coumarin on in vitro metabolism of aflatoxin B1 by their tissues.

To clarify whether enzymes involved in aflatoxin B1 (AFB1) metabolism in pigs respond to antioxidant agents, the effect of feeding piglets with diets containing green tea extracts (Sunphenon) and coumarin on in vitro AFB1 metabolism by their liver and intestinal tissues was studied. The results showed that coumarin reduced AFB1-DNA adduct formation by both liver and intestinal microsomes, while Sunphenon did not have any effects. Both coumarin and Sunphenon enhanced the glutathione S-transferase (GST) activity to conjugate AFB1 to glutathione GSH in the intestine, although no effects were noted in the liver. Changes of the expression of mRNA of GSTA2 and GSTO1 were not in parallel with the observed changes of GST activity, suggesting that other GST subtypes are involved in the GST activity toward AFB1. As for lipophilic-free AFB1 metabolites, coumarin reduced the liver microsomal conversion of AFB1 to aflatoxin M1 (AFM1) and aflatoxin Q1 (AFQ1), but Sunphenon exerted no effects. Both coumarin and Sunphenon enhanced the conversion of AFB1 to aflatoxicol in the liver. All the results suggest that feeding with a diet containing coumarin affects AFB1 metabolism to enhance AFB1 detoxification through the suppression of P450 enzyme activity in the liver and the enhancement of GST activity in the intestine. Feeding with a diet containing Sunphenon enhances AFB1 detoxification, but the effects are noted mainly in the intestine.     

Protective effects of coffee diterpenes against aflatoxin B1-induced genotoxicity: mechanisms in rat and human cells.

The coffee-specific diterpenes cafestol and kahweol (C + K) have been reported to be anticarcinogenic in several animal models. Proposed mechanisms involve a co-ordinated modulation of several enzymes responsible for carcinogen detoxification, thus preventing reactive agents interacting with critical target sites. To address the human relevance of the chemoprotective effects of C + K against aflatoxin B(1) (AFB1) genotoxicity observed in rat liver, and to compare the mechanisms of protection involved in both species, animal and human hepatic in vitro test systems were applied. In rat primary hepatocytes, C + K reduced the expression of cytochrome P450 CYP 2C11 and CYP 3A2, the key enzymes responsible for AFB1 activation to the genotoxic metabolite aflatoxin B1-8,9 epoxide (AFBO). In addition, these diterpenes induced significantly GST Yc2, the most efficient rat GST subunit involved in AFBO detoxification. These effects of C + K resulted in a marked dose-dependent inhibition of AFB1-DNA binding in this rat in vitro culture system. Their relevance in humans was addressed using liver epithelial cell lines (THLE) stably transfected to express AFB1 metabolising cytochrome P450s. In these cells, C + K also produced a significant inhibition of AFB1-DNA adducts formation linked with an induction of the human glutathione S-transferase GST-mu. Altogether, these results suggest that C + K may have chemoprotective activity against AFB1 genotoxicity in both rats and humans.     

Changes in hepatic cytosolic glutathione S-transferase activity and expression of its class-P during prenatal and postnatal period in rats treated with aflatoxin B1

The effect of aflatoxin B1 (AFB1) on the expression of glutathione S-transferase-P (GST-P) which is the major isoform of GST in developmental stages has been investigated in rat liver during prenatal and postnatal stages. Following administration of AFB1 (0, 0.5, 1.0, 2.0, 3.0 or 4.0 mg/kg bw) injected I.P on day 8.5 of gestation the number of dead or reabsorbed fetuses and malformed embryos were recorded. Then the fetal livers were processed for measurement of total GST and GST-P activities, using 1-chloro-2,4-dinitrobenzene (CDNB) and ethacrynic acid (ETA) as substrates respectively. RT-PCR using rat GST-P specific primers was performed on mRNA extracted from livers. Besides, the effects of AFB1 on hepatic GST and GST-P were assessed in groups of suckling rats directly injected with the toxin. The results show that a single dose of AFB1 (1.0 or 2.0 mg/kg bw) caused approximately 50–60% depletion in fetal liver GST towards CDNB. Postnatal experiments revealed that liver GST (using CDNB as substrate) was significantly induced (~40%) in suckling rats injected with a single dose of AFB1 (3.0 mg AFB1/kg) 24 h before killing. Liver GST-P expression was unaffected due to AFB1 exposures of rats before and after the birth. This finding was substantiated by western blotting and RT-PCR techniques. These data suggest that AFB1-related induction in rat liver total GST after birth may be implicated in protective mechanisms against AFB1. In contrast, inhibition of this enzyme in fetal liver following placental transfer of the carcinogen may explain high susceptibility of fetal cells to transplancental aflatoxins. Furthermore, lack of influence of AFB1 on GST-P expression in developmental stages can role out the involvement of this class of GST in AFB1 biotransformation.     

Effects of dietary butylated hydroxytoluene on aflatoxin B1-relevant metabolic enzymes in turkeys.

We have shown previously that the extreme sensitivity of turkeys to aflatoxin B(1) (AFB(1)) is due to a combination of efficient AFB(1) activation by cytochrome P450s (CYPs) 1A and deficient detoxification by glutathione S-transferases (GSTs). Phenolic antioxidants such as butylated hydroxytoluene (BHT) have been shown to be chemoprotective in some animal models due, in part, to modulation of AFB(1)-relevant phase I and/or phase II activities, and we wished to determine whether BHT has a similar effect in turkeys. Ten-day-old male turkeys were maintained on diets amended with 1000 or 4000 ppm of BHT for 10 days, then sampled. Hepatic microsomal CYP 1A activity as well as conversion of AFB(1) to the putative toxic metabolite, the exo-AFB(1)-8,9-epoxide (AFBO), were significantly lower compared with control. Conversely, dietary BHT significantly increased activities of several isoforms of hepatic cytosolic GST, as well quinone oxidoreductase (QOR). Western immunoblotting confirmed that dietary BHT increased expression of homologues to rodent GST isoforms Yc1, Yc2 and Ya. There was, however, no observable BHT-related increase in GST-mediated specific conjugation with microsomally-generated AFBO. In total, our data indicates that dietary BHT modulates a variety of AFB(1)-relevant phase I and phase II enzymes, while having no measurable effect towards specific AFB(1) detoxification by GST.     

Molecular cloning and expression of a novel cytochrome p450 from turkey liver with aflatoxin b1 oxidizing activity

Cytochromes P450 are members of a superfamily of oxidative hemoprotein enzymes that metabolize a variety of endogenous and exogenous compounds. Previous studies in our laboratory have shown that efficient P450-mediated activation underlies the extreme sensitivity of poultry, specifically turkeys, to the toxic effects of the mycotoxin aflatoxin B1 (AFB1). Using 3'- and 5'-rapid amplification of cDNA ends (RACE), we amplified from turkey liver RNA a full-length 1.73 kb cDNA predicted to be 528 amino acids with 94.7% sequence identity to a CYP1A5 from chicken liver. A truncated construct of the turkey CYP1A5 gene with 29 amino acids deleted from the hydrophobic NH2-terminal region was cloned and heterologously expressed in Escherichia coli. The expressed protein from E. coli membranes had a CO-binding spectrum typical of P450s, and it catalyzed the O-dealkylation of the CYP1A prototype substrates ethoxyresorufin and methoxyresorufin. CYP1A5-mediated O-dealkylation of methoxyresorufin was completely inhibited by alpha-naphthoflavone, a specific CYP1A inhibitor. Inhibitors to other mammalian P450s (3A4, 2D, 2E, and 3A1) either slightly inhibited this activity or not at all. CYP1A5 oxidized AFB1 to form two metabolites: the reactive intermediate, AFB1 -8,9-epoxide (AFBO), and aflatoxin M1 (AFM1). Because of the importance of AFBO and AFM1 in the toxicity of AFB1, we conclude that this P450 probably plays some role in the well-known hypersensitivity of turkeys to AFB1. To our knowledge, this is the first P450 cloned and sequenced from turkeys, the species in which the toxicity of AFB1 was first discovered.     

Species differences in the toxicity and cytochrome P450 IIIA-dependent metabolism of digitoxin

In rats, cytochrome P450 (P450) IIIA enzymes are an important determinant of digitoxin toxicity. Induction of these liver microsomal enzymes decreases the toxicity of digitoxin by increasing its oxidative cleavage to digitoxigenin bis- and monodigitoxoside (dt2 and dt1). The present study shows that the susceptibility of different mammalian species to digitoxin toxicity is inversely related to liver microsomal P450 IIIA activity (measured as testosterone 6 beta-hydroxylase activity). Based on this correlation, we correctly predicted that hamsters, which have the highest P450 IIIA activity, are extremely resistant to digitoxin toxicity. To further examine the relationship between digitoxin toxicity and P450 IIIA activity, the pathways of digitoxin metabolism catalyzed by liver microsomes from nine mammalian species were examined by high performance liquid chromatography. The overall rate of digitoxin metabolism varied approximately 90-fold and followed the rank order: hamster greater than rat greater than guinea pig greater than dog greater than mouse approximately monkey greater than rabbit approximately cat greater than human. The qualitative differences in digitoxin metabolism were as striking as the quantitative differences. Formation of 16- and/or 17-hydroxydigitoxin was the major pathway of digitoxin oxidation catalyzed by liver microsomes from hamster, guinea pig, rabbit, cat, dog, and cynomolgus monkey. Guinea pig and, to a lesser extent, hamster liver microsomes also converted digitoxin to an unknown metabolite, the formation of which was catalyzed by P450. None of the species examined catalyzed the 12-hydroxylation of digitoxin to digoxin at a high rate. Similarly, none of the species examined catalyzed a high rate of conversion of digitoxin to dt2, with the notable exception of the rat. However, dt2 formation was the major pathway of digitoxin metabolism catalyzed by human liver microsomes, although humans were much less active (approximately 2%) than rats in this regard. The rate of dt2 formation varied approximately 41-fold among 22 samples of human liver microsomes, which was highly correlated (r = 0.841) with the rate of testosterone 6 beta-hydroxylation. Antibody against rat P450 IIIA1 inhibited the high rate of dt2 formation by rat liver microsomes and the low rate catalyzed by mouse, guinea pig, dog, monkey, and human liver microsomes. In contrast, anti-P450 IIIA1 did not inhibit the 12-, 16-, or 17-hydroxylation of digitoxin (or the formation of the unknown metabolite), despite the fact that anti-P450 IIIA1 strongly inhibited (greater than 70%) the 6 beta-hydroxylation of testosterone by liver microsomes from each of the species examined (except rabbit liver microsomes, which were inhibited only approximately 30%).(ABSTRACT TRUNCATED AT 400 WORDS)     

Metabolism of aflatoxin B1 by normal human bronchial epithelial cells

Aflatoxin B, (AFB1) is a potent hepatocarcinogen in animal models and a suspected carcinogen in humans. High concentrations of AFB, have been found in respirable grain dusts, and may therefore be a risk factor for human lung cancer in certain occupations. To study the potential for AFB, activation in human lung, cytochrome P-450 (CYP)-mediated activation and glutathione S-transferase (GST)-mediated detoxification of AFB1 were examined in cultured normal human bronchial epithelial (NHBE) cells. Cells were exposed to 0. 15 microM or 1.5 microM AFB, for 48 h and media was collected for metabolite analysis by high-performance liquid chromatography (HPLC). At 0. 15 microM, AFB1 was metabolized only to the detoxified metabolite aflatoxin Q1 (AFQ1). At 1.5 microM AFB1, both aflatoxin M1 (AFM1), and AFQ1 were produced. Cells pretreated with 50 degrees M 3-methylcholanthrene (3MC), a CYP 1A inducer, for 72 h prior to 0.15 microM AFB1, produced the activated AFB1 8,9-epoxide (AFBO). Similarly, microsomes prepared from 3MC-pretreated cells formed AFBO, but microsomes from noninduced cells did not. While AFB1-DNA adducts were not detected at low AFB1 concentrations in untreated NHBE, 3MC induction caused the production of AFB1-DNA adducts at 0.015 and 0.15 microM AFB1. Western immunoblots showed that the primary CYP isoforms responsible for AFB1 activation in the liver, 1A and 3A4, to be constitutively expressed in NHBE cells. Expression of CYP 1A was significantly increased in 3MC-pretreated cells, while CYP 3A4 expression increased slightly, but not to the extent of the 1A isoforms. The principal AFBO detoxifying enzyme, glutathione S-transferase (GST), was constitutively expressed in NHBE cells, and was increased approximately twofold by 3MC pretreatment. Cytosolic fractions from neither control nor 3MC-induced NHBE had measurable AFBO conjugating activity, indicating that these cells may lack AFB1-relevant GST activity. From these data, it appears that NHBE cells activate AFB1 inefficiently, but possess CYPs reportedly responsible for metabolism of AFB1. These data support earlier findings showing modest CYP-mediated AFB1 activation in human airways, but indicate that exposure to polycyclic aromatic hydrocarbons (PAHs), such as 3MC, which induce CYP(s) that specifically activate AFB1 may increase the harmful effects of AFB1 exposures in human airways.     

METABOLISM AND CYTOTOXICITY OF AFLATOXIN B 1 IN CYTOCHROME P-450-EXPRESSING HUMAN LUNG CELLS

The mycotoxin aflatoxin B 1 (AFB 1 ) is a hepatocarcinogen in many animal models and probably a human carcinogen. Besides being a dietary carcinogen, AFB 1 has been detected in dusts generated in the processing and transportation of AFB 1 -contaminated products. Inhalation of grain dusts contaminated with AFB 1 may be a risk factor in human lung cancer. Aflatoxin B 1 requires cytochrome P-450 (CYP)-mediated activation to form cytotoxic and DNA-reactive intermediates, and this activation in human liver is mediated by the CYP 1A2 and 3A4 isoforms. Which isoforms are important in AFB 1 activation in human lung is not well understood. To investigate whether these CYPs can activate AFB 1 at low, environmentally relevant concentrations in human lung cells, SV40 immortalized human bronchial epithelial cells (BEAS-2B) that were transfected with cDNA for CYPs 3A4 (B3A4) or 1A2 (B-CMV1A2) were used. B-CMV1A2 cultured in 15 n M AFB 1 produced the AFB 1 -glutathione conjugate (AFB 1 -GSH) and aflatoxin M 1 (AFM 1 ), while B3A4 cells produced only aflatoxin Q 1 (AFQ 1 ) at 0.15 M AFB 1 . Nontransfected BEAS-2B cells produced no metabolites, even at 1.5 m M AFB 1 . Microsomes prepared from B-CMV1A2 and B3A4 cells activated AFB 1 to AFB 1 8,9-epoxide (AFBO), while those from BEAS-2B cells did not produce AFBO. Cytosol from all three cell types was ineffective at glutathione S -transferase (GST)-mediated trapping of enzymatically generated AFB 1 8,9-epoxide. B-CMV1A2 cells were 100-fold more sensitive to AFB 1 compared to B3A4 cells, and were 6000-fold more sensitive than control BEAS-2B cells. Western immunoblots confirmed that only B-CMV1A2 cells expressed CYP 1A2 protein, while CYP 3A4 was only in B3A4 cells. B-CMV1A2 cells were the most sensitive to AFB 1 , followed by B3A4 cells. CYP 3A4, which has been predicted to activate AFB 1 primarily at higher AFB 1 concentrations, was also responsible for significant AFB 1 toxicity at low concentrations. These data indicate that human lung cells expressing these CYP isoforms are capable of activating AFB 1 , even at environmentally relevant concentrations.     

Reduction of aflatoxin B1 dialdehyde by rat and human aldo-keto reductases.

Oxidation of the mycotoxin aflatoxin (AF) B1 yields the 8,9-epoxide, which nonenzymatically hydrolyzes rapidly to a dihydrodiol that in turn undergoes slow, base-catalyzed ring opening to a dialdehyde [Johnson, W. W., Harris, T. M., and Guengerich F. P. (1996) J. Am. Chem. Soc. 118, 8213-8220]. AFB1 dialdehyde does not bind to DNA but can react with protein lysine groups. One enzyme induced by cancer chemopreventive agents is AFB1 aldehyde reductase (AFAR), which catalyzes the NADPH-dependent reduction of the dialdehyde to a dialcohol. AFB1 dialdehyde is known to convert nonenzymatically to AFB1 dihydrodiol at neutral pH, and we reinvestigated the enzymatic reaction by preparing AFB1 dialdehyde at pH 10 and then used this to initiate reactions (at neutral pH) with rat and human AFAR isozymes. Two monoalcohols were identified as products, and their identities were established by NaB2H4 reduction, chemical cleavage, and mass spectrometry. The monoalcohol corresponding to reduction at C-8 formed first in reactions catalyzed by either the rat or the human AFAR. This C-8 monoalcohol was further reduced to AFB1 dialcohol by AFAR. The other monoalcohol (C-6a) was formed but not reduced to the dialcohol rapidly. Steady-state kinetic parameters were estimated for the reduction of AFB1 dialdehyde by rat and human AFAR to the monoalcohols. The apparent k(cat) and K(m) values were not adequate to rationalize the observed DeltaA(340) spectral changes in a kinetic model. Simulation fitting was done and yielded parameters indicative of greater enzyme efficiency. A survey of 12 human liver cytosol samples showed a variation of 2.3-fold in AFAR activity. Rats treated with AFB1 excreted the dialcohol and a monoalcohol in urine. The results of these studies are consistent with a role of (rat and human) AFAR in protection against AFB1 toxicity.