Induction of c-jun and TGF-beta 1 in Fischer 344 rats during amiodarone-induced pulmonary fibrosis.

Amiodarone (AM) is an antidysrhythmic agent with a propensity to cause pulmonary toxicity, including potentially fatal fibrosis. In the present study, the potential roles of c-Jun and transforming growth factor (TGF)-beta 1 in AM-induced inflammation and fibrogenesis were examined after intratracheal administration of AM (1.83 micromol/day on days 0 and 2) or an equivalent volume (0.4 ml) of distilled water to male Fischer 344 rats. Northern and immunoblot analyses demonstrated that lung TGF-beta 1 (mRNA and protein) expression was increased 1.5- to 1.8-fold relative to control during the early inflammation period and 1 day, 1 wk, and 2 wk post-AM treatment. Lung c-Jun protein expression was increased concomitantly with evidence of AM-induced fibrosis; at 5 wk post-AM treatment, c-Jun protein was increased 3.3-fold relative to control. The results indicate a role for induction of c-jun and TGF-beta 1 expression in the development of AM-induced pulmonary fibrosis in the Fischer 344 rat and provide potential targets for therapeutic intervention.

Effects of vitamin E on cytotoxicity of amiodarone and N-desethylamiodarone in isolated hamster lung cells.

Amiodarone (AM) is a potent and efficacious antidysrhythmic agent that can cause potentially life-threatening pulmonary fibrosis. Vitamin E has been demonstrated to decrease AM-induced pulmonary fibrosis in vivo in hamsters. In the present in vitro study, we investigated the effects of vitamin E on cell death induced by AM and its primary metabolite, N-desethylamiodarone (DEA), in freshly isolated hamster lung cells. Following incubation for 24 or 36 h, 300 microM vitamin E decreased (P<0.05) 100 microM AM-induced cytotoxicity (0.5% trypan blue uptake) in alveolar macrophages by 11.7+/-3% or 21.4+/-12%, respectively, but did not decrease cytotoxicity in fractions enriched with alveolar type II cells or non-ciliated bronchiolar epithelial (Clara cells) or in isolated unseparated cells (cell digest). Vitamin E had no effect on 50 microM DEA-induced cytotoxicity. Vitamin E did not alter cellular levels of AM or DEA in any cell fraction. Lipid peroxidation (assessed by isoprostane formation) was increased (P<0.05) in cell digest, alveolar type II cell and Clara cell enriched fractions incubated with 500 microM carbon tetrachloride (CCl(4)) for 4 h but not in enriched fractions of cells exposed to 100 microM AM or 50 microM DEA. No AM-induced loss of viability was observed at this time point, but DEA decreased (P<0.05) Clara cell viability by approximately 25%. These results demonstrate cell type selective protection against AM-induced cytotoxicity by vitamin E, and suggest that lipid peroxidation does not initiate AM- or DEA-induced cytotoxicity in isolated hamster lung cells.

Human lung microsomal cytochrome P4501A1 (CYP1A1) activities: impact of smoking status and CYP1A1, aryl hydrocarbon receptor, and glutathione S-transferase M1 genetic polymorphisms.

There are numerous conflicting epidemiological studies addressing correlations between cytochrome P450 1A1 (CYP1A1) genetic polymorphisms and lung cancer susceptibility, with associations plausibly linked to alterations in carcinogen bioactivation. Similarly, correlations between aryl hydrocarbon receptor gene (AHR) codon 554 genotype and CYP1A1 inducibility are controversial. The objective of this study was to determine whether smoking status, and CYP1A1, AHR, and glutathione S-transferase M1 gene (GSTM1) polymorphisms correlate with altered CYP1A1 activities. Lung microsomal CYP1A1-catalyzed 7-ethoxyresorufin O-dealkylation (EROD) activities were much higher in tissues from current smokers (n = 46) than in those from non-/former smokers (n = 24; 12.11 +/- 13.46 and 0.77 +/- 1.74 pmol/min/mg protein, respectively, mean +/- SD; P < 0.05). However, EROD activities in lung microsomes from current smokers CYP1A1*1/1 (n = 33) and heterozygous MspI variant CYP1A1*1/2A (n = 10) were not significantly different (12.23 +/- 13.48 and 8.23 +/- 9.76 pmol/min/mg protein, respectively, P > 0.05). Three current smokers were heterozygous variant CYP1A1*1/2B (possessing both *2A and *2C alleles), and exhibited activities similar to individuals CYP1A*1/1. One current smoker was heterozygous variant CYP1A1*4 and exhibited activities comparable with individuals CYP1A1*1/1 at that locus. EROD activities in microsomes from current smokers AHR(554)Arg/Arg (n = 41) and heterozygous variant AHR(554)Arg/Lys (n = 5) were not significantly different (12.13 +/- 13.56 and 12.01 +/- 14.23 pmol/min/mg protein, respectively; P > 0.05). Furthermore, microsomal EROD activities from current smokers with the GSTM1-null genotype (n = 28) were not significantly different from those (n = 18) carrying at least one copy of GSTM1 (12.61 +/- 14.24 and 11.34 +/- 12.53 pmol/min/mg protein, respectively; P > 0.05). Additionally, when genotypic combinations of CYP1A1, AHR, and GSTM1 were assessed, there were no significant effects on EROD activity. On the basis of microsomal enzyme activities from heterozygotes, CYP1A1*1/2A, CYP1A1*1/2B, CYP1A1*1/4, and AHR(554) Arg/Lys variants do not appear to significantly affect CYP1A1 activities in human lung, and we observed no association between CYP1A1 activity and the GSTM1-null polymorphism.

Disruption of mitochondrial function and cellular ATP levels by amiodarone and N-desethylamiodarone in initiation of amiodarone-induced pulmonary cytotoxicity.

Amiodarone (AM), a potent antidysrhythmic agent, can cause potentially life-threatening pulmonary fibrosis. In the present investigation of mechanisms of initiation of AM lung toxicity, we found that 100 microM AM decreased mitochondrial membrane potential in intact hamster lung alveolar macrophages and preparations enriched in isolated alveolar type II cells and nonciliated bronchiolar epithelial (Clara) cells, following 2 h of incubation. This was followed by a drop in cellular ATP content (by 32--77%) at 4 to 6 h, and 30 to 55% loss of viability at 24 h. Supplementation of incubation media with 5.0 mM glucose or 2.0 mM niacin did not reduce AM-induced ATP depletion or cell death in macrophages, and the mitochondrial permeability transition inhibitor cyclosporin A (1.0 microM) did not affect AM cytotoxicity. At 50 microM, the AM metabolite N-desethylamiodarone (DEA) produced effects similar to those of AM, but more rapidly and extensively, with the Clara cell-enriched preparation being particularly susceptible. In isolated whole lung mitochondria, DEA was accumulated to a greater extent than AM. Both AM and DEA inhibited complex I- and complex II-supported respiration, but DEA inhibited complex II to a greater degree than AM. These results demonstrate that AM and DEA disrupt mitochondrial membrane potential prior to ATP depletion and subsequent lung cell death, that DEA is more potent than AM, and that the mitochondrial permeability transition is not involved in mitochondrial perturbation by AM. This suggests that AM- and DEA-induced perturbations of mitochondrial function may initiate AM-induced pulmonary toxicity.

Mechanisms of aflatoxin B1 lung tumorigenesis.

Although aflatoxin B1 (AFB1) is best known as a hepatocarcinogen, the respiratory system can also be a target of this mycotoxin. In isolated lung cells from rabbits and mice, AFB1 is bioactivated by cytochromes P450, primarily in nonciliated bronchiolar epithelial (Clara) cells. However, mutagenesis experiments suggest that the DNA-binding AFB1 epoxide metabolite can leave the cells of origin, and potentially interact with other cell types. Consistent with DNA adduct studies, AFB1-induced AC3F1 mouse lung tumors contain point mutations at guanine residues in K-ras, with the anticipated bias for the A/J allele. Furthermore, following AFB1 treatment but prior to tumor development, K-ras mutations occur preferentially in mouse Clara cells. However, in contrast to findings with other carcinogens, AFB1-induced mouse lung tumors demonstrate frequent, but heterogeneously distributed, overexpression of p53 protein as well as p53 point mutations, suggesting a carcinogen-specific response. Unlike lung tissue from mice and rabbits, human peripheral lung bioactivates AFB1 primarily by prostaglandin H synthase--and/or lipoxygenase-catalyzed cooxidation, with activity concentrated in macrophages. In addition, although glutathione S-transferase M1-1 has high specific activity for AFB1 epoxide conjugation, lung tissues from GSTM1-null individuals do not demonstrate diminished rates of conjugation, compared to tissues from GSTM1-positive individuals. In summary, AFB1 tumorigenesis in mice demonstrates unique properties, and processes of bioactivation show significant species differences.

Glutathione S-transferase-catalyzed conjugation of bioactivated aflatoxin B(1) in human lung: differential cellular distribution and lack of significance of the GSTM1 genetic polymorphism.

Epidemiological studies suggest that aflatoxin B(1) (AFB(1)), a mycotoxin produced by certain Aspergillus species, may play a role in human respiratory cancers in occupationally-exposed individuals. AFB(1) requires bioactivation to the corresponding exo-8,9-epoxide for carcinogenicity, and glutathione S-transferase (GST)-catalyzed conjugation of the epoxide with glutathione (GSH) is a critical determinant of susceptibility to AFB(1). Of the purified human GST enzymes studied, the polymorphic hGSTM1-1 has the highest activity towards AFB(1) exo-epoxide. The influence of the GSTM1 polymorphism on AFB(1)-GSH formation, as well as the abilities of cytosols from preparations enriched in different isolated lung cell types to conjugate AFB(1)-epoxides, were examined. In whole-lung cytosols from patients undergoing clinically indicated lobectomy, GSTM1 genotype correlated with GSTM1 phenotype as determined by [(3)H]trans-stilbene oxide conjugation: GSTM1-positive = 295 +/- 31 pmol/mg/h (n = 6); GSTM1-negative = 92.8 +/- 23.3 pmol/mg/h (n = 4) (P < 0.05). In contrast, conjugation of microsome-generated [(3)H]AFB(1)-epoxides with GSH was low and variable between patients, and did not correlate with GSTM1 genotype: GSTM1-positive = 11.9 +/- 8.1, 111 +/- 66 and 510 +/- 248 fmol/mg/h (n = 6); GSTM1-negative = 15.3 +/- 16.7, 167 +/- 225 and 540 +/- 618 fmol/mg/h (n = 4) (for 1, 10 and 100 microM [(3)H]AFB(1), respectively). GSH conjugates of AFB(1) exo-epoxide and the much less mutagenic stereoisomer AFB(1) endo-epoxide were produced in a ratio of approximately 1:1 in cytosols from both whole lung and isolated cells. Total cytosolic AFB(1)-epoxide conjugation was significantly higher in fractions enriched in alveolar type II cells (3.07 +/- 1.61 pmol/mg/h) than in unseparated lung cells (0.143 +/- 0.055 pmol/mg/h) or fractions enriched in alveolar macrophages (0. 904 +/- 0.319 pmol/mg/h; n = 4) (P < 0.05). Furthermore, AFB(1)-GSH formation and percentage of alveolar type II cells in different cell fractions were correlated (r = 0.78, P < 0.05). These results demonstrate that human lung GSTs exhibit very low conjugation activity for both AFB(1)-8,9-epoxide stereoisomers, and that this activity is heterogeneously distributed among cell types, with alveolar type II cells exhibiting relatively high activity. Of the GSTs present in human peripheral lung which contribute to AFB(1) exo- and endo-epoxide detoxification, hGSTM1-1 appears to play at most only a minor role.

Biotransformation of the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in freshly isolated human lung cells.

Metabolism of the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) was characterized in human lung cells isolated from peripheral lung specimens obtained from 12 subjects during clinically indicated lobectomy. NNK biotransformation was assessed in preparations of isolated unseparated cells (cell digest), as well as in preparations enriched in alveolar type II cells, and alveolar macrophages. Metabolite formation was expressed as a percentage of the total recovered radioactivity from [5-(3)H]NNK and its metabolites per 10(6) cells per 24 h. 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) was the major metabolite formed in all lung cell preparations examined, and its formation ranged from 0.50 to 13%/10(6) cells/24 h. Formation of alpha-carbon hydroxylation end-point metabolites (bioactivation) and pyridine N-oxidation metabolites (detoxification), ranged from non-detectable to 0.60% and from non-detectable to 1.5%/10(6) cells/24 h, respectively, reflecting a large degree of intercellular and inter-individual variability in NNK metabolism. Formation of the alpha-hydroxylation end-point metabolite 4-hydroxy-1-(3-pyridyl)-1-butanol (diol) was consistently higher in alveolar type II cells than in cell digest or alveolar macrophages (0.0146 +/- 0.0152, 0.0027 +/- 0.0037 and 0.0047 +/- 0.0063%/10(6) cells/24 h, respectively; n = 12; P < 0.05). SKF-525A was used to examine cytochrome P450 contributions to the biotransformation of NNK. SKF-525A inhibited keto reduction of NNK to NNAL by 85, 86 and 74% in cell digest, type II cells, and macrophages, respectively (means of 11 subjects, P < 0.05). Type II cell incubates treated with SKF-525A formed significantly lower amounts of total alpha-hydroxylation metabolites compared with type II cells without SKF-525A (0.0776 +/- 0.0841 versus 0.1694 +/- 0. 2148%/10(6) cells/24 h, respectively; n = 11; P < 0.05). The results of this first study examining NNK biotransformation in freshly isolated human lung cells indicate that NNK metabolism is subject to a large degree of inter-individual and intercellular variability, and suggest a role for P450s in human lung cell NNK metabolism. Both alveolar type II cells and alveolar macrophages may be potential target cells for NNK toxicity based on their alpha-carbon hydroxylation capabilities. In addition, carbonyl reduction of NNK to NNAL is SKF-525A sensitive in human lung cells.

Ki-ras activation in lung cells isolated from AC3F1 (A/J x C3H/HeJ) mice after treatment with aflatoxin B1.

Lung cells isolated from AC3F1 (A/J x C3H/HeJ) mice 7 wk after treatment with the carcinogenic mycotoxin aflatoxin B1 (AFB1) were examined for point mutations in the Ki-ras oncogene. Ki-ras mutant allele frequencies in fractions enriched with nonciliated bronchiolar epithelial (Clara) cells were consistently higher than in alveolar type II cell fractions. Mutant alleles were undetectable or minimal in macrophage- and polymorphonuclear leukocyte-enriched fractions. In cells from vehicle (dimethyl sulfoxide)-treated mice, small proportions of mutant Ki-ras alleles were found in the Clara cell-enriched fraction but not in other cell fractions. The results indicated that Clara cells are particularly susceptible to AFB1-induced Ki-ras mutation, an early event in AFB1-induced mouse lung tumorigenesis.

High frequency and heterogeneous distribution of p53 mutations in aflatoxin B1-induced mouse lung tumors.

Inactivation of the p53 tumor suppressor gene is one of the most frequent genetic alterations observed in human lung cancers. However, p53 mutations are more rarely detected in chemically induced mouse lung tumors. In this study, 62 female AC3F1 (A/J x C3H/HeJ) mice were treated with aflatoxin B1 (AFB1; 150 mg/kg i.p. divided into 24 doses over 8 weeks). At 6-14 months after dosing, mice were killed, and tumors were collected. A total of 71 AFB1-induced lung tumors were examined for overexpression of p53 protein by immunohistochemical staining. Positive nuclear p53 staining was observed in 79% of the AFB1-induced tumors, but the pattern was highly heterogeneous. In approximately 73% of the positively stained tumors, fewer than 5% of cells demonstrated positive staining; in the other 27%, between 10% and 60% of the cells stained positively, with staining localized to the periphery of the tumors in many cases. Single-strand conformational polymorphism analysis of the evolutionarily conserved regions of the p53 gene (exons 5-8) from AFB1-induced whole lung tumor DNA revealed banding patterns consistent with point mutations in 20 of 76 (26%) tumors, with 85% of the mutations in exon 7 and 15% of the mutations in exon 6. Identification of point mutations could not be confirmed by direct sequence analysis because bands representing putative mutations appeared only weakly on autoradiograms. This was presumably due to the heterogeneous nature of the DNA analyzed. Single-strand conformational polymorphism analysis of DNA from laser capture microdissected cells of paraffin-embedded AFB1-induced tumor tissue sections stained for p53 produced banding patterns consistent with point mutations in 18 of 30 (60%) DNA samples. Direct sequencing of the microdissected samples revealed mutations at numerous different codons in exons 5, 6, and 7. Of 26 mutations found in microdissected regions from adenomas and carcinomas, 9 were G:C-->A:T transitions, 11 were A:T-->G:C transitions, and 5 were transversions (2 G:C-->T:A, 2 T:A-->A:T, and 1 A:T-->C:G), whereas 1 deletion mutation was identified. The concordance between immunostaining and molecular detection of p53 alterations was 72% when laser capture microdissection was used versus 17% based on whole tumor analysis. The high mutation frequency and heterogeneous staining pattern suggest that p53 mutations occur relatively late in AFB1-induced mouse lung tumorigenesis and emphasize the value of analyzing different staining regions from paraffin-embedded mouse lung tumors.

Effects of dietary vitamin E supplementation on pulmonary morphology and collagen deposition in amiodarone- and vehicle-treated hamsters.

Amiodarone (AM) is a potent antidysrhythmic agent that is limited in clinical use by its adverse effects, including potentially life-threatening AM-induced pulmonary toxicity (AIPT). The present study tested the ability of dietary supplementation with vitamin E (500 IU d,1-alpha-tocopherol acetate/kg chow) to protect against pulmonary damage following intratracheal administration of AM (1.83 micromol) to the male golden Syrian hamster. At 21 days post-dosing, animals treated with AM had increased lung hydroxyproline content and histological disease index values compared to control (P < 0.05), which were indicative of fibrosis. Dietary vitamin E supplementation for 6 weeks resulted in a 234% increase in lung vitamin E content at the time of AM dosing, and maintenance on the diet prevented AM-induced elevation of hydroxyproline content and disease index 21 days post-dosing. Dietary vitamin E supplementation also decreased hydroxyproline content and disease index values in hamsters treated intratracheally with distilled water, the AM vehicle. These results demonstrate a protective role for vitamin E in an in vivo model of AIPT, and suggest that this antioxidant may have non-specific antifibrotic effects in the lung.

Amiodarone-induced disruption of hamster lung and liver mitochondrial function: lack of association with thiobarbituric acid-reactive substance production.

Amiodarone (AM) is an efficacious antidysrhythmic agent that is limited clinically by numerous adverse effects. Of greatest concern is AM-induced pulmonary toxicity (AIPT) due to the potential for mortality. Mitochondrial alterations and free radicals have been implicated in the etiology of AM-induced toxicities, including AIPT. Isolated hamster lung and liver mitochondria were assessed for AM-induced effects on respiration, membrane potential, and lipid peroxidation. AM (50-400 microM) stimulated state 4 (resting) respiration at complexes I and II of tightly coupled lung mitochondria, with higher concentrations (200 and 400 microM) resulting in a subsequent inhibition. This biphasic effect of AM (200 microM) was also observed with isolated liver mitochondria. Only inhibition of respiration was observed with AM (50-400 microM) in less tightly coupled lung mitochondria. Based on safranine fluorescence, 200 microM AM decreased lung mitochondrial membrane potential (p < 0.05), while a concentration-dependent (50-200 microM) decrease of membrane potential was observed with liver mitochondria exposed to AM (p < 0.05). Formation of thiobarbituric acid-reactive substances (TBARS) was not altered by AM (50-400 microM) in incubations lasting up to 1 h. These results indicate that lipid peroxidation, as indicated by levels of TBARS, does not play a role in AM-induced alterations in mitochondrial respiration and membrane potential.

Human glutathione S-transferase P1 polymorphisms: relationship to lung tissue enzyme activity and population frequency distribution.

The association between glutathione S-transferase (GST) activity as measured by 1-chloro-2,4-dinitrobenzene (CDNB) conjugation and genotype at exon 5 and exon 6 of the human GSTP1 gene was investigated in normal lung tissue obtained from 34 surgical patients. These samples were genotyped for previously identified polymorphisms in exon 5 (Ile105Val) and exon 6 (Ala114Val) by PCR-RFLP and direct sequencing. GST enzyme activity was significantly lower among individuals with the 105 Val allele. Homozygous Ile/Ile samples (n = 18) had a mean cytosolic CDNB conjugating activity of 74.9 +/- 3.8 nmol/mg per min; heterozygotes (n = 13) had a mean specific activity of 62.1 +/- 4.2 nmol/mg per min and homozygous Val/Val (n = 3) had a mean specific activity of 52.5 +/- 4.5 nmol/mg per min. The CDNB conjugating activity measured for the Ile/Ile genotype group was significantly different from that observed in the Ile/Val group (P = 0.03), and from Ile/Val and Val/Val genotypes combined (P = 0.009). Mean GST activity values were consistently lower in individuals with genotypes containing the 105 valine allele, regardless of smoking exposure. Genotypes at codon 114 were also assessed but the mean GST activity was not significantly lower in individuals with the 114 valine allele. A new haplotype, present in two samples who were homozygous 105Ile and had a 114Val, was identified and proposed as GSTP1*D. Frequencies of the exon 5 and exon 6 polymorphisms were determined in samples obtained from European-Americans, African-Americans and Taiwanese. The differences observed were highly significant suggesting the possibility of GSTP1 genotype-associated, ethnic differences in cancer susceptibility and chemotherapeutic response.

Differential susceptibilities of isolated hamster lung cell types to amiodarone toxicity.

Treatment of cardiac dysrhythmias with the iodinated benzofuran derivative amiodarone (AM) is limited by pulmonary toxicity. The susceptibilities of different lung cell types of male Golden Syrian hamsters to AM-induced cytotoxicity were investigated in vitro. Bronchoalveolar lavage and protease digestion to release cells, followed by centrifugal elutriation and density gradient centrifugation, resulted in preparations enriched with alveolar macrophages (98%), alveolar type II cells (75-85%), and nonciliated bronchiolar epithelial (Clara) cells (35-50%). Alveolar type II cell and Clara cell preparations demonstrated decreased viability (by 0.5% trypan blue dye exclusion) when incubated with 50 microM AM for 36 h, and all AM-treated cell preparations demonstrated decreased viability when incubated with 100 or 200 microM AM. Based on a viability index ((viability of AM-treated cells/viability of controls) x 100%), the Clara cell fraction was significantly (p<0.05) more susceptible than all of the other cell types to 50 microM AM. However, AM cytotoxicity was greatest (p<0.05) in alveolar macrophages following incubation with 100 or 200 microM AM. There was no difference between any of the enriched cell preparations in the amount of drug accumulated following 24 h of incubation with 50 microM AM, whereas alveolar macrophages accumulated the most drug during incubation with 100 microM AM. Thus, the most susceptible cell type was dependent on AM concentration. AM-induced cytotoxicity in specific cell types may initiate processes leading to inflammation and pulmonary fibrosis.

ATP-dependent transport of aflatoxin B1 and its glutathione conjugates by the product of the multidrug resistance protein (MRP) gene.

Glutathione-S-transferase-catalyzed conjugation of glutathione (GSH) to aflatoxin B1-8,9-epoxide plays an important role in preventing binding of this ultimate carcinogen to target macromolecules. Once formed, the aflatoxin B1-epoxide-GSH conjugates are actively extruded from the cell by an unidentified ATP-dependent export pump or pumps. Two possible candidates for this GSH conjugate pump are the 190-kDa multidrug resistance protein (MRP) and the 170-kDa P-glycoprotein. Both proteins belong to the ATP-binding cassette superfamily of transmembrane transport proteins and confer resistance to a similar spectrum of natural-product drugs. Using membrane vesicles from MRP-transfected cells, we found that MRP transports GSH conjugates of both the endo-isomers and exo-isomers of aflatoxin B1-8,9-epoxide in an ATP-dependent, osmotically sensitive manner (V(max) = 180 pmol/mg/min, K(m) = 189 nM). Membrane vesicles from P-glycoprotein-overexpressing cells showed very low levels of transport. MRP-mediated transport was inhibited by an MRP-specific monoclonal antibody and by a variety of GSH derivatives and cholestatic steroid glucuronides. ATP-dependent transport of unmodified aflatoxin B1 by MRP-enriched membrane vesicles was low but markedly enhanced in the presence of 5 mM GSH, even though GSH conjugates of aflatoxin B1 were not formed by the vesicles. These data demonstrate that MRP is capable of energy-dependent transport of aflatoxin B1 and its GSH conjugates and suggest a potential protective role for MRP in mammalian chemical carcinogenesis.

Biotransformation of aflatoxin B1 in human lung.

In addition to being a potent hepatocarcinogen, aflatoxin B1 (AFB1) is a pulmonary carcinogen in experimental animals, and epidemiological studies have shown an association between AFB1 exposure and lung cancer in humans. This study investigated AFB1 bioactivation and detoxification in human lung tissue obtained from patients undergoing clinically indicated lobectomy. [3H]AFB1 was bioactivated to a DNA binding metabolite by human whole lung cytosols in a time-, protein concentration-, and AFB1 concentration-dependent manner. Cytosolic activation of [3H]AFB1 correlated with lipoxygenase (LOX) activity and was inhibited by the LOX inhibitor nordihydroguaiaretic acid (NDGA; 100 microM), indicating that LOXs were largely responsible for the observed cytosolic activation of AFB1. In whole lung microsomes, low levels of indomethacin inhibitable prostaglandin H synthase (PHS)-mediated [3H]AFB1-DNA binding and cytochrome P-450 (P450)-mediated [3H]AFB1-DNA binding were observed. Cytosolic glutathione S-transferase (GST)-catalyzed detoxification of AFB1-8,9-epoxide, produced by rabbit liver microsomes, was minimal at 1 and 10 microM [3H]AFB1. With 100 microM [3H]AFB1, [3H]AFB1-8,9-epoxide conjugation with reduced glutathione was 0.34 +/- 0.26 pmol/mg/h (n = 10). In intact, isolated human lung cells, [3H]AFB1 binding to cellular DNA was higher in cell fractions enriched in macrophages than in either type II cell-enriched fractions or fractions containing unseparated cell types. Indomethacin produced a 63-100% decrease in [3H]AFB1-DNA binding in macrophages from five of seven patients, while NDGA inhibited [3H]AFB1-DNA adduct formation by 19, 40 and 56% in macrophages from three of seven patients. In alveolar type II cells, NDGA decreased [3H]AFB1-DNA binding by 30-100% in cells from three patients and indomethacin had little effect. SKF525A, an isozyme non-selective P450 inhibitor, enhanced [3H]AFB1 binding to cellular DNA in unseparated cells, macrophages, and type II cells, suggesting that P450-mediated bioactivation of AFB1 is not a major pathway by which AFB1-8,9-epoxide is formed in human lung cells. Overall, these studies suggest that P450 has a minor role in the bioactivation of AFB1 in human lung. Rather, LOXs and PHS appear to be important bioactivation enzymes. Co-oxidative bioactivation of AFB1, in combination with the low conjugating activity displayed by human lung cytosolic GSTs, likely contributes to human pulmonary susceptibility to AFB1.

Glutathione S-transferase-catalyzed conjugation of bioactivated aflatoxin B1 in rabbit lung and liver.

Aflatoxin B1 (AFB1) requires bioactivation to AFB1-8,9-epoxide for carcinogenicity, and glutathione S-transferase (GST)-catalyzed conjugation of activated AFB1 with glutathione (GSH) is a critical determinant of susceptibility to the mycotoxin. Incubations containing [3H]AFB1, rabbit liver microsomes, an NADPH-generating system, 1 mM GSH, and GST-containing lung or liver cytosol were performed to assess the abilities of lung and liver GSTs to conjugate AFB1-8,9-epoxide. [3H]AFB1-GSH was isolated by isocratic reverse-phase high-performance liquid chromatography (HPLC) and quantitated by liquid scintillation spectroscopy. Maximal [3H]AFB1-GSH formation rates were significantly lower for lung than for liver (0.3 +/- 0.1 and 1.7 +/- 0.4 nmol/mg/hr, respectively). Immunoprecipitation of rabbit pulmonary cytosolic GSTs with anti-alpha or anti-mu GST antisera decreased [3H]AFB1-GSH production by approximately 45 and 51%, respectively, indicating that alpha-class and mu-class GSTs are of similar importance in catalyzing this reaction in the lung. Because mu-class GSTs comprise only a small proportion of total lung GST content, these enzymes have high specific activity toward AFB1-8,9-epoxide. In contrast, the pi-class GST appeared to play a negligible role. Using a rat liver microsomal system to generate both AFB1 exo- and endoepoxide isomers, and analysis based on chiral HPLC, we found that rabbit liver cytosolic GSTs catalyzed formation of both AFB1 exo- and endo-epoxide-GSH conjugates, whereas pulmonary cytosolic GSTs catalyzed formation of only the exo stereoisomer at detectable levels. Despite a preference for conjugating the more mutagenic AFB1 exo-epoxide isomer, the relatively low capacity for GST-catalyzed detoxification of bioactivated AFB1 in lung may be an important factor in the susceptibility of the lung to AFB1 toxicity.

Activation of K-ras in aflatoxin B1-induced lung tumors from AC3F1 (A/J x C3H/HeJ) mice.

In addition to being a potent hepatocarcinogen, aflatoxin B1 (AFB1) is a pulmonary carcinogen in experimental animals and epidemiological studies have shown an association between AFB1 exposure and lung cancer in humans. Since point mutations at codons 12, 13 and 61 of the K-ras protooncogene are often implicated in chemically induced mouse lung tumors and in human lung adenocarcinomas, we undertook an investigation of the role of K-ras activation in AFB1-induced pulmonary carcinogenesis. Female AC3F1 (A/J x C3H/HeJ) mice were treated with AFB1 (150 mg/kg i.p., divided into 24 doses over 8 weeks), and 6-14 months after the completion of dosing mice were killed and pulmonary adenomas and carcinomas removed. Of the 76 AFB1-induced lung tumors analyzed by single strand conformation polymorphism (SSCP) and direct sequencing, 75 possessed K-ras codon 12 mutations (46 GTT, 14 GAT, 13 TGT and 2 TTT; normal, GGT) and one had a GGC-->CGC mutation in codon 13. The observation that K-ras mutations occurred only at G:C base pairs is in agreement with N7-guanine being the primary site of AFB1-DNA adduct formation and with guanine residues being targets for AFB1-induced oxidative DNA damage via formation of 8-hydroxydeoxyguanosine (8-OHdG). The AFB1-specific nature of the observed K-ras mutation spectrum and the fact that 100% of the tumor samples examined contained K-ras mutations is consistent with K-ras activation being an early, critical event in AFB1-induced pulmonary carcinogenesis in AC3F1 mice. The parental origin of the observed K-ras mutations was determined by allele-specific PCR amplification of AFB1-induced lung tumor DNA followed by SSCP analysis. In the vast majority of tumors (73/76), the mutated K-ras allele was derived from the lung tumor susceptible A/J parent. This finding supports the existence of a link between K-ras and differences in mouse lung tumor susceptibility.

The 1995 Pharmacological Society of Canada Merck Frosst Award. Cellular and molecular targets in pulmonary chemical carcinogenesis: studies with aflatoxin B1.

Although most notorious as a liver carcinogen, the mycotoxin aflatoxin B1 targets other tissues as well, including those of the respiratory system. Because the biotransformation of aflatoxin B1 to toxic and nontoxic metabolites has been fairly well characterized, it serves as a useful and relevant model carcinogen for studying the biochemical (i.e., balance between bioactivation and detoxification) and molecular (i.e., mutations in target genes) mechanisms of pulmonary chemical carcinogenesis. Because of the cellular diversity of the lung, it is of particular interest to assess these processes in different lung cell types, if we are to identify the target cells for carcinogen action. This review summarizes studies that have been aimed at identifying the basis for susceptibility of the lung to aflatoxin B1.

Modulation of aflatoxin B1 biotransformation by beta-naphthoflavone in isolated rabbit lung cells.

The cytotoxic and carcinogenic mycotoxin aflatoxin (AF) B1 (AFB1) is biotransformed by the cytochrome P450 monooxygenases (CYP) to a number of relatively nontoxic metabolites, as well as to the ultimate toxic metabolite, AFB1-8,9-epoxide. In a number of tissues and species, AFB1 hydroxylation to the relatively nontoxic metabolite, AFM1, is induced by beta-naphthoflavone (BNF) treatment. Although the liver is the principle target organ for AFB1 toxicity, the mycotoxin is also toxic and carcinogenic to respiratory tissues. To determine if BNF treatment alters the extent of pulmonary AFB1 bioactivation by enhancing detoxification and thereby decreasing epoxidation, the effects of BNF on pulmonary AFB1 metabolism were examined. Rabbit lung cells, isolated by protease digestion and centrifugal elutriation, were incubated with [3H]AFB1. In nonciliated bronchiolar epithelial (Clara) cell-enriched (45-50%) fractions, [3H]AFM1 production (pmol/mg DNA per 2 h) was increased by prior treatment of rabbits with BNF (80 mg/kg per day, 3 and 2 days before cell isolation) as follows: with 1.0 microM [3H]AFB1; control, 10.6 +/- 2.3; BNF, 30.0 +/- 6.4; with 0.10 microM [3H]AFB1; control, 9.4 +/- 4.7; BNF, 20.6 +/- 5.9. With 1.0 microM [3H]AFB1, prior treatment of animals with BNF abolished formation of [3H]aflatoxicol (AFL) but not [3H]AFQ1. The activation (epoxidation) of [3H]AFB1 was measured indirectly as covalent binding to endogenous DNA. With 1.0 microM [3H]AFB1, treatment of rabbits with BNF did not alter DNA binding (pmol/mg DNA per 2 h) in the Clara cell-enriched fraction: control, 103 +/- 41; BNF, 114 +/- 49. However, with 0.10 microM [3H]AFB1, DNA binding in the same fraction was 47% lower in cells from BNF treated animals: control, 17.4 +/- 4.2; BNF, 9.3 +/- 3.9. Formation of 8,9-dihydro-8,9-dihydroxy-AFB1, and the glutathione conjugate of the aflatoxin epoxide (AFB1-GSH) were not detectable at the AFB1 concentration and time point studied, in cells from either BNF-treated or control rabbits. Incubation of isolated, unseparated lung cells from untreated rabbits with 5.0 to 50 microM BNF decreased [3H]AFB1-DNA binding in the presence of 0.1 microM [3H]AFB1 by 35 to 77%, while lower BNF concentrations did not alter DNA binding. In lung cells isolated from BNF treated rabbits, BNF was not detectable (i.e. < 0.5 microM detection limit). Therefore, the amount of BNF present in isolated rabbit lung cells following in vivo treatment with BNF was below that required to directly inhibit AFB1-DNA adduct formation. The decrease in AFB1-DNA binding from rabbits treated with BNF is apparently due to the selective induction of CYP isozymes and related increases in AFM1 formation, and not to direct inhibition of epoxidation or enhanced conjugation of AFB1-8,9-epoxide with glutathione.

Evaluation of reactive oxygen species involvement in amiodarone pulmonary toxicity in vivo and in vitro.

Amiodarone (AM) is an effective antidysrhythmic agent, restricted in use by the development of adverse effects, including potentially fatal AM-induced pulmonary toxicity (AIPT). Although the pathogenesis of AIPT is unknown, an oxidant mechanism has been proposed. The present study evaluated the role of reactive oxygen species (ROS) in AM-induced toxicity. The effect of inhibiting lung antioxidant defense on in vivo development of AIPT was evaluated in hamsters. Lung glutathione reductase activity was inhibited by 66%, 6 hours following administration of 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) (20 mg/kg i.p.). When AM (1.83 mumol) was administered intratracheally 6 hours after BCNU, toxicity was enhanced, as indicated by lung hydroxyproline content and histological evaluation 21 days later. However, BCNU treatment did not affect AM-induced alterations in lung glutathione, suggesting that the increased toxicity was not due to decreased antioxidant capacity following BCNU. The effect of BCNU on AM cytotoxicity in vitro was evaluated using rabbit lung alveolar macrophages. Incubation with 5 microM BCNU for 2 hours caused greater than 95% inhibition of glutathione reductase activity. However, BCNU treatment had no effect on 146 microM AM-induced cytotoxicity, as assessed by lactate dehydrogenase latency following 12 hours of incubation. Rabbit macrophages loaded with 2',7'-dichlorofluorescein, which is oxidized by ROS to fluorescent 2',7'-dichlorofluorescein (DCF), were used to evaluate ROS generation by AM. Incubation of macrophages with AM (73 or 146 microM) for 1 hour, with or without the catalase inhibitor sodium azide (1 mM), did not result in DCF formation. Overall, these results do not support the hypothesis that AIPT is due to ROS action.