Correlation of DNA adduct formation and riddelliine-induced liver tumorigenesis in F344 rats and B6C3F1 mice [Cancer Lett. 193 (2003) 119-125

Riddelliine is a naturally occurring pyrrolizidine alkaloid that induces liver hemangiosarcomas in male and female F344 rats and male B6C3F1 mice. We previously reported that eight dehydroretronecine (DHR)-derived DNA adducts were formed in liver DNA of rats treated with riddelliine. In order to examine the relationship between DNA adduct levels and the incidence of hemangiosarcomas, we have measured DHR-derived DNA adduct levels in purified rat and mouse liver endothelial cells, the cells of origin for the hemangiosarcomas. F344 rats and B6C3F1 mice were treated by gavage 5 days per week for 2 weeks with riddelliine at 1.0 mg/kg for rats and 3.0 mg/kg for mice. One, 3, 7, and 28 days after the last dose, liver parenchymal and endothelial cell fractions were isolated, and the quantities of DHR-derived DNA adducts were determined by 32P-postlabeling/HPLC. The DHR-derived DNA adduct levels in the endothelial cells were significantly greater than in the parenchymal cells. The DNA adduct levels in rat endothelial cells were greater than in the mouse endothelial cells. These results indicate that the levels of riddelliine-induced DNA adducts in specific populations of liver cells correlate with the preferential induction of liver hemangiosarcomas by riddelliine.     

Differential mutagenicity of riddelliine in liver endothelial and parenchymal cells of transgenic big blue rats

Riddelliine is a naturally occurring pyrrolizidine alkaloid that induces liver hemangiosarcomas in rats and mice. We previously reported higher levels of DNA adducts in liver endothelial cells than in liver parenchymal cells of riddelliine-treated mice and rats [Cancer Lett. 193 (2003) 119], suggesting that the tumor specificity is due to higher levels of DNA damage in the cells that form hemangosarcomas. In the present study, we evaluated the cell-specificity of riddelliine mutagenicity in rat liver. Female transgenic Big Blue rats were treated by gavage with 0.3 mg riddelliine per kg body weight, 5 days a week for 12 weeks. One day after the last treatment, the rats were sacrificed and liver parenchymal and endothelial cell fractions were isolated and purified. DNA was extracted from the cell fractions and used to assay for mutant frequency (MF) in the cII transgene. While there was no difference in the cII MFs of liver parenchymal cells in control and riddelliine-treated rats, the cII MF of liver endothelial cells from treated rats was significantly greater than the cII MF of endothelial cells from control rats. Molecular analysis of the mutants in liver endothelial cells indicated that G:C-->T:A transversion, a mutation that is characteristically induced by riddelliine, accounted for only 9% of all mutations in control rats, but made up 17% of mutations in treated rats. In contrast, G:C-->A:T transition, the major mutation in control rats where it made up 54% of all mutations, was reduced to 40% of mutations in riddelliine-treated rats. These results suggest that the relatively high mutagenicity of riddelliine in rat liver endothelial cells may be partially responsible for the tumorigenic specificity of this agent.     

Genotoxic potential of tamoxifen and analogues in female Fischer F344/n rats, DBA/2 and C57BL/6 mice and in human MCL-5 cells.

Chronic administration of tamoxifen to female rats causes hepatocellular carcinomas. We have investigated damage to liver DNA caused by the administration of tamoxifen to female Fischer F344/N rats or C57B1/6 or DBA/2 mice using 32P-postlabelling. Following the administration of tamoxifen for 7 days (45 mg/kg/day) and extraction of hepatic DNA, up to 7 radiolabelled adduct spots could be detected after PEI-cellulose chromatography of the 32P-labelled DNA digests. Tamoxifen caused a time-dependent increase in the level of adduct detected up to a value of at least 1 adduct/10(6) nucleotides after 7 days dosing. A dose response relationship was demonstrated over the range of 5-45 mg/kg/day (0.013-0.12 mmol/kg/day). On cessation of dosing there was a loss of adducts from the liver DNA. These adducts were not detected in DNA from vehicle-dosed controls or in DNA from kidney, lung, spleen, uterus or peripheral lymphocytes. Pyrrolidinotamoxifen caused a similar level of adduct formation as tamoxifen. In contrast, no significant adduct formation could be detected in liver DNA from rats given droloxifene or toremifene. Mice given tamoxifen (45 mg/kg/day for 4 days) showed levels of adducts in the liver which were 30-40% of those present in rats. Exposure of rat hepatocytes to tamoxifen in vitro, resulted in induction of unscheduled DNA synthesis, when preparations from rats which had been pretreated with tamoxifen in vivo were used. No such increase could be detected in hepatocytes from control rats, suggesting tamoxifen may induce enzymes responsible for its own activation. Tamoxifen induced a significant increase in micronucleus formation in a dose dependent manner in cultures of MCL-5 cells, a human cell line that expresses 5 different human cytochrome P450 isoenzymes, as well as epoxide hydrolase.     

Formation of DNA adducts of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in male Fischer-344 rats.

2-Amino-1-methyl-6-phenylimidazo[4,5-b]-pyridine (PhIP) is known to induce colon tumors in male Fischer-344 rats. Using 32P-postlabeling assays, we have examined PhIP-DNA adduct formation in various organs and white blood cells (WBCs) of the male Fischer-344 rat 24 h after a single oral dose of 0, 0.5, 5 or 50 mg PhIP/kg. Three PhIP-DNA adducts were detected in WBCs and in all organs, except in the liver and stomach which had only two adducts. The extent of adduct formation was dose-related, but at 0.5 mg/kg no adducts could be detected in any of the organs. At 50 mg/kg, adduct levels, expressed as relative adduct labeling values (RAL x 10(7), or adducts per 10(7) nucleotides assuming complete labeling) were highest in the large intestine (5.66), followed by WBCs (5.04), stomach (1.44), small intestine (1.32), kidney (1.16), liver (0.67) and lungs (0.52). It is concluded that orally administered PhIP forms high levels of specific DNA adducts in the large intestine, the target organ in PhIP carcinogenesis in the male Fischer-344 rat, and that the high level of adducts in WBCs indicates that significant amounts of the ultimate carcinogenic form of PhIP are present in the circulation.     

Association between acetylator genotype and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) DNA adduct formation in colon and prostate of inbred Fischer 344 and Wistar Kyoto rats

2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), a heterocyclic amine (HCA) found in cooked meats, causes colon and prostate tumors in male rats. Polymorphic N-acetyltransferase metabolizes N-hydroxy-PhIP to a DNA-reactive form. Liver, colon, and prostate PhIP-DNA adduct levels were compared in male rapid-acetylator Fischer 344 (F344) and slow-acetylator Wistar-Kyoto (WKY) rats fed 0.01 or 0.04% PhIP. Liver PhIP-DNA adduct levels at both PhIP doses, and colon PhIP-DNA adduct levels at the 0.01% PhIP dose were unaffected by acetylator genotype. However, in rats fed 0.04% PhIP, colon PhIP-DNA adduct levels were higher in rapid acetylator F344 rats (P < 0.05). Similarly, prostate PhIP-DNA adduct levels were higher in rapid acetylator F344 rats at both PhIP doses (P < 0.05). The combination of the high-PhIP dose and rapid-acetylator genotype resulted in the highest level of PhIP-DNA adducts in rat colon and prostate.     

NTP technical report on the toxicity and metabolism studies of chloral hydrate (CAS No. 302-17-0). Administered by gavage to F344/N rats and B6C3F1 mice

Chloral hydrate is widely used as a sedative and a hypnotic in pediatric medicine. It is also a byproduct of water chlorination. Chloral hydrate has been shown to be genotoxic in numerous prokaryotic and eukaryotic assay systems including human lymphocytes in vitro. One of its metabolites, trichloroacetic acid, has demonstrated hepatocarcinogenic activity in mice. Trichloroethylene and perchloroethylene, both of which are metabolized to chloral hydrate, have been shown to be carcinogenic in rats and/or mice. Because of this evidence of carcinogenicity and because of the wide-spread use of chloral hydrate, 16- or 17-day range-finding toxicity studies and separate 16- or 17-day metabolism studies were performed in F344/N rats and B6C3F1 mice in preparation for further long-term rodent studies. In addition, in vitro studies of the metabolism and DNA-binding capacity of chloral hydrate and its metabolites were performed. Genetic toxicity studies were conducted in Salmonella typhimurium, cultured Chinese hamster ovary cells, Drosophila melanogaster, and mouse bone marrow cells. For the range-finding studies, groups of eight male and eight female F344/N Nctr BR rats and B6C3F1/Nctr BR (C57BL/6N x C3H/HeN MTV-) mice were administered 0, 50, 100, 200, 400, or 800 mg chloral hydrate per kg body weight in water by gavage 5 days per week for 17 days (rats) or 16 days (mice) for a total of 12 doses. One male rat receiving 800 mg/kg died after five doses. Two 800 mg/kg female rats died after dosing ended but before study termination. One male mouse in each group except the 400 mg/kg group died before the end of the study. Two 800 mg/kg female mice also died before the end of the study. The final mean body weight of 800 mg/kg male rats and the mean body weight gains of 400 and 800 mg/kg males were significantly less than those of the vehicle controls. The mean body weight gains of all groups of dosed male mice were significantly greater than that of the vehicle control group. The only clinical finding in rats and mice attributed to chloral hydrate treatment was light sedation in the 400 mg/kg groups and heavy sedation in the 800 mg/kg groups; sedation subsided within 30 minutes or 3 hours, respectively. The liver weights of 400 mg/kg male mice and 800 mg/kg male and female mice were significantly greater than those of the vehicle control groups. No chemical-related lesions were observed in rats or mice. Male and female rats and mice were administered a single dose of 50 or 200 mg chloral hydrate per kg body weight in water by gavage, or 12 doses of 50 or 200 mg/kg over 17 days (rats) or 16 days (mice). Plasma concentrations of chloral hydrate and its metabolites were determined 15 minutes, 1, 3, 6, and 24 hours, and 2, 4, 8, and 16 days after receiving 1 or 12 doses. Maximum concentrations of chloral hydrate were observed at the initial sampling point of 15 minutes. By 1 hour, the concentrations had dropped substantially, and by 3 hours, chloral hydrate could not be detected in rats or mice. Trichloroacetic acid was the major metabolite detected in the plasma. In rats, the concentrations rose slowly, with the peaks occurring between 1 and 6 hours after treatment. In mice, the peak concentrations were found 1 hour after dosing. The concentrations then slowly decreased such that by 2 days the metabolite could no longer be detected in rats or mice. Trichloroethanol was assayed both as the free alcohol and its glucuronide. In rats, the maximum concentrations of free trichloroethanol occurred at 15 minutes, while the peak concentrations of trichloroethanol glucuronide were found at 1 hour; by 3 hours, concentrations of both metabolites approached background levels. In mice, the maximum concentrations of both metabolites occurred at 15 minutes, and by 1 to 3 hours concentrations approached background levels. The plasma concentrations of chloral hydrate and its metabolites were dose dependent in rats and mice. In mice, plasma concentrations of trichloroacetic acid were significantly higher after a single dose than after 12 doses. None of the metabolic parameters appears to account for species differences that may exist in hepatocarcinogenicity. The data from the study of metabolism and DNA adduct formation indicated that in vitro metabolism of 200 microM to 5 mM chloral hydrate by male B6C3F1 mouse liver microsomes (control microsomes) generated free radical intermediates that resulted in endogenous lipid peroxidation, forming malondialdehyde, formaldehyde, acetaldehyde, acetone, and propionaldehyde. Similar concentrations of trichloroacetic acid and trichloroethanol, the primary metabolites of chloral hydrate, also generated free radicals and induced lipid peroxidation. Lipid peroxidation induced by trichloroacetic acid nearly equaled that induced by chloral hydrate, while that from trichloroethanol was three- to fourfold less. Metabolism of 200 microM to 5 mM chloral hydrate, trichloroacetic acid, and trichloroethanol by liver microsomes of B6C3F1 mice pretreated with pyrazole (pyrazole-induced microsomes) yielded lipid peroxidation products at concentrations two- to threefold greater than those from liver microsomes of untreated mice. Additionally, chloral hydrate-induced lipid peroxidation catalyzed by control and pyrazole-induced microsomes was reduced significantly by 2,4-dichloro-6-phenylphenoxyethylamine, a general cytochrome P450 inhibitor. Human lymphoblastoid transgenic cells expressing cytochrome P(450)2E1 metabolized 200 to 5,000 micrograms/mL chloral hydrate to reactants inducing mutations, whereas the parental cell line was inactive. The malondialdehyde-modified DNA adduct, 3-(2-deoxy-beta-D-erythro-pentofuranosyl)pyrimido[1,2 alpha]purin-10(3H)-one (MDA-MG-1), formed from the metabolism of 1 mM chloral hydrate, trichloroacetic acid, and trichloroethanol by control B6C3F1 mouse liver microsomes, mouse pyrazole-induced microsomes, male F344/N rat liver microsomes, and human liver microsomes in the presence and absence of calf thymus DNA was also determined. When incubated in the absence of calf thymus DNA, the amount of malondialdehyde formed from metabolism by pyrazole-induced mouse microsomes was twice that from rat or human liver microsomes. Amounts of chloral hydrate-induced and trichloroacetic acid-induced lipid peroxidation products formed from metabolism by rat and human liver microsomes were similar, and these quantities were about twice those formed from the metabolism of trichloroethanol. The quantity of MDA-MG-1 formed from the metabolism of chloral hydrate, trichloroacetic acid, and trichloroethanol by mouse, rat, and human liver microsomes exhibited a linear correlation with the quantity of malondialdehyde formed under incubation conditions in the absence of calf thymus DNA. Chloral hydrate was shown to be mutagenic in vitro and in vivo. At doses from 1,000 to 10,000 micrograms/plate, it induced mutations in S. typhimurium strain TA100, with and without S9 activation; an equivocal response was obtained in S. typhimurium strain TA98 in the absence of S9, and no mutagenicity was detected with strain TA1535 or TA1537. Chloral hydrate at doses from 1,700 to 5,000 micrograms/mL induced sister chromatid exchanges; at doses from 1,000 to 3,000 micrograms/mL, chromosomal aberrations were induced in cultured Chinese hamster ovary cells, with and without S9. Results of a sex-linked recessive lethal test in D. melanogaster were unclear; administration of chloral hydrate by feeding produced an inconclusive increase in recessive lethal mutations, results of the injection experiment were negative. An in vivo mouse bone marrow micronucleus test with chloral hydrate at doses from 125 to 500 mg/kg gave a positive dose trend. In summary, due to the absence of chloral hydrate-induced histopathologic lesions in rats and mice, no-observed-adverse-effect levels (NOAELs) were based on body weights of rats and liver weights of mice. The NOAELs for rats and mice were 200 mg/kg. Chloral hydrate was rapidly metabolized by rats and mice, with trichloroacetic acid occurring as the major metabolite. Peak concentrations of trichloroacetic acid occurred more quickly in mice. Plasma concentrations of chloral hydrate were dose dependent, but metabolic rates were unaffected by dose or sex. Chloral hydrate was mutagenic in vitro and in vivo. Metabolism of chloral hydrate and its metabolites produced free radicals that resulted in lipid peroxidation in liver microsomes of mice, rats, and humans. Induction of cytochrome P(450)2E1 by pyrazole increased the concentrations of lipid peroxidation products; inhibition of cytochrome P(450)2E1 by 2,4-dinitrophenylhydrazine reduced these concentrations. Metabolism of chloral hydrate and its metabolites by mouse, rat, and human liver microsomes formed malondialdehyde, and in the presence of calf thymus DNA formed the DNA adduct MDA-MG-1.     

Accumulation and persistence of DNA adducts of the synthetic steroid cyproterone acetate in rat liver

Cyproterone acetate (CPA) is a synthetic steroid which is widelyused in antlandrogenic and gestagenic drugs. We have recentlyshown that CPA induces DNA adducts in cultured rat hepatocytesand in rat liver (1). In the present investigation, we studiedthe persistence and accumulation of CPA-derived DNA adductsin the liver of rats using the 32 technique. To study the persistenceof CPA-DNA adducts, rats were treated with a single oral doseof 10 (female rats) or 100 mg CPA/kg body wt (male rats). FourDNA adducts were detected in the liver of both gender. In femalerats, maximal total DNA adduct levels of 3.40±0.04 adducts/10⁶nucleotides were observed after 1 week. Eleven weeks later,40% of the adducts determined after 1 week were still detectable.In male rats, maximal hepatic DNA adduct levels of     

Effects of bile acids on 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine-induced aberrant crypt foci and DNA adduct formation in the rat colon

The effects of deoxycholic acid (DCA) and ursodeoxycholic acid (UDCA) on 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP)-induced aberrant crypt foci (ACF) in the rat colon were examined. The effect of these bile acids on DNA adduct formation by PhIP in the colon was then analyzed, since the main action of PhIP is the formation of DNA adducts and subsequent gene mutations. For the ACF study, male F344 rats were administered PhIP-HCl (75 mg/kg, 10 doses) by gavage, and a diet containing bile acid (0.4% DCA or UDCA) was provided from 3 days before the first dose of PhIP for 8 weeks. The mean number of ACF per colon of DCA, UDCA and controls were 9.9, 2.4 and 5.5, respectively. The ACF number was significantly increased by DCA and decreased by UDCA (P<0.001). To examine the effect of bile acids on DNA adduct formation, male F344 rats were fed a diet supplemented with bile acids (0.1 or 0.4% of DCA and UDCA) 7 days prior to the PhIP administration. All rats were administered a single dose of PhIP-HCl (50 mg/kg) by gavage and sacrificed 48 hours later. DNA adduct levels of the 0.1% UDCA, 0.1% DCA and controls were 2.93 (adducts/10(7) nucleotides), 2.65 and 1.10, respectively. Those of 0.4% UDCA, 0.4% DCA and controls were 1.64, 1.30 and 1.00, respectively. The PhIP-DNA adduct level was significantly increased by administration of 0.1% UDCA, 0.1% DCA (P<0.05) and 0.4% UDCA (P<0.01). The increasing effect of both DCA and UDCA on PhIP-induced DNA adduct formation was unexpected, and was not directly associated with ACF formation.     

DNA adduct formation in B6C3F1 mice and Fischer-344 rats exposed to 1, 2, 3-trichloropropane

1, 2, 3-Trichloropropane (TCP) is a multispecies, multisitecarcinogen which has been found to be an environmental contaminantIn this study, we have characterized and measured DNA adductsformed in vivo following exposure to TCP. [¹⁴C]TCP was administeredto male B6C3F1 mice and Fischer-344 rats by gavage at dosesused in the NTP carcinogenesis bioassay. Both target and nontargetorgans were examined for the formation of DNA adducts. Adductswere hydrolyzed from DNA by neutral thermal or mild acid hydrolysis,isolated by HPLC, and detected and quanti-tated by measurementof radioactivity. The HPLC elution profile of radioactivitysuggested that one major DNA adduct was formed. To characterizethis adduct, larger yields were induced in rats by intraperitonealadministration of TCP (300 mg/kg). The DNA adduct was isolatedby HPLC based on coelution with the radiolabeled adduct, andcompared to previously identified adducts. The isolated adductcoeluted with S-[1-(hydroxymethyl)-2-(N⁷-guanyl)-ethyljglutathione,an adduct derived from the structurally related carcinogen 1,2-dibromo-3-chloropropane (DBCP). Analysis by electrospray massspectrometry suggested that the TCP-induced adduct and the DBCP-derivedadduct were identical. The ¹⁴C-labeled DNA adduct was distributedwidely among the organs examined. Adduct levels varied dependingon species, organ, and dose. In rat organs, adduct concentrationsfor the low dose ranged from 0.8 to 6.6 µmol per mol guanineand from 7.1 to 47.6 µmol per mol guanine for the highdose. In the mouse, adduct yields ranged from 0.32 to 28.1 µmolper mol guanine for the low dose and from 12.2 to 208.1 µmolper mol guanine for the high dose. The relationship betweenDNA adduct formation and organ-specific tumorigenesis was unclear.Although relatively high concentrations of DNA adducts weredetected in target organs, several nontarget sites also containedhigh adduct levels. Our data suggest that factors in additionto adduct formation may be important in TCP-induced carcinogenesis.     

Removal of DNA adducts of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in the male Fischer-344 rat

The food mutagen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyrldine(PhIP) is carcinogenic in the male Fischer-344 rat, affectingprincipally the colon. PhIP-DNA adducts may play a role in theinitiation of the carcinogenic process. We have evaluated theformation and persistence of PhIP-DNA adducts in the colon,circulating white blood cells (WBC) and several other non-targetorgans of the male Flscher-344 rat. Young adult male animalswere given a single dose of PhIP (50 mg/kg) by gavage. Animalswere killed 1, 2, 6, 12, 16 or 20 days after dosing (4 anlmals/timepoint) and their liver, lungs, stomach, small intestine, cecum,colon, kidneys, WBC, heart and spleen were removed for isolationof DNA and assay of PhIP-DNA adducts by ³²P-postlabeling. Forinterorgan comparisons of cell turnover, rats were given a singlei.p. dose of [methyl-³H], after which DNA was isolated at thesame time intervals as for adduct analysis and its sp. act.(d.p.m. ³ H/100µg) was determined. In all organs up tothree adducts could be isolated and the adduct pattern was thesame in each case. On day 1, total adduct levels were highestin the colon (the target organ), followed by the spleen, cecum,small intestine, stomach, liver, kidneys, lungs, WBC and heart.Rates of adduct removal were similar in the colon, spleen, cecum,liver, lungs, stomach and small intestine, with day 16 and day20 levels falling to <16% of those on day 1; rates of removalwere slower in the heart and kidneys (52.0 and 30.3% of day1 values remaining on day 16 respectively). Adducts in WBC increasedat first (day 2) and decreased thereafter to virtually non-detectablelevels on days 16 and 20. Heart adducts on days 2–12 increasedslightly or remained as high as those on day 1, then decreasedto lower levels on days 16 and 20(53.0 and 28.7% of day 1 levelsrespectively). There was no preferential removal or persistenceof any individual adduct in WBC or in any of the organs. Ondays 1 and 2, the sp. act. of intestinal DNA (small intestine,cecum and colon) was >30-fold higher than that in severalother organs, including the liver. These sp. act. decreasedto the low sp. act. of the liver on day 20. It is concludedthat the rates of adduct removal from the intestines are morelikely to be related to cell turnover of epithelial cells thanto enzymatic repair. Factors other than rate of adduct repairmay contribute to the striking single organ target of PhIP inthe male Flscher-344 rat.     

Similar patterns of DNA adduct formation of 2-amino-3-methylimidazo [4,5-f]quinoline in the Fischer 344 rat, CDF1 mouse, cynomolgus monkey and Salmonella typhimurium

2-Amino-3-methylimidazo [4,5-f]quinoline (IQ) is a known liver carcinogen in the Fischer 344 rat, the CDF1 mouse and in the cynomolgus monkey. Using 32P-postlabeling assays, we compared IQ-DNA adduct formation in the liver of IQ-treated Fischer 344 rats, CDF1 mice and cynomolgus monkeys with that in Salmonella typhimurium (strain TA98) incubated with IQ (in the presence of a liver S9 activating system) or N-hydroxy-IQ. Up to five adducts could be detected, the pattern of which was identical in all cases. The major adduct co-chromatographed with standard N-(deoxyguanosin-8-yl)-IQ in all cases and comprised 54.7-82.8% of the total. The four minor adducts were not identified. It is concluded that N-(deoxyguanosin-8-yl)-IQ is the major IQ-DNA adduct under all experimental conditions and that the pattern of N-hydroxy-IQ-DNA adducts is identical to that found in the liver of animals exposed to IQ, and to that found after reacting IQ with DNA in the presence of a liver S9 activating system. Thus, N-hydroxylation of IQ is a critical step in the formation of IQ-DNA adducts.     

In vivo formation of aflatoxin B1-DNA adducts in parenchymal and non-parenchymal cells of rat liver.

The induction of hepatocellular carcinoma from liver parenchymal cells in laboratory animals by aflatoxin B1 (AFB1) is well documented. In contrast no tumours arising from the sinusoidal cell population have been reported after exposure to AFB1. The apparent resistance of the latter cell type was investigated at the level of DNA adduct formation in vivo in male Sprague-Dawley rats. Liver parenchymal and non-parenchymal cell populations were isolated from rats at 20 min and 1, 24 and 72 h after administration of 240 microCi (0.6 mg) [G-3H]AFB1/kg. AFB1-DNA binding was observed in both liver cell subpopulations and was 3- to 5-fold higher in parenchymal cells than in non-parenchymal cells. The major DNA adduct found in parenchymal cells at 1 h after AFB1 administration was 8,9-dihydro-8-(N7-guanyl)-9-hydroxyaflatoxin B1 (AFB1-gua), whereas at later time points the persistent secondary adduct, AFB1-formamidopyrimidine, predominated. In contrast, AFB1-gua was not observed at any time in DNA from non-parenchymal cells and the secondary adducts predominated throughout. These observations are discussed with reference to the susceptibility of different liver cell types to AFB1-carcinogenesis and the possible roles of the major AFB1-DNA adduct species.     

Formation of 7-(4-oxobutyl)guanine in hepatic DNA of rats treated with N-nitrosopyrrolidine.

We have reported previously the formation of two structurally distinct exocyclic guanine adducts (adducts 1 and 6) in liver DNA of F344 rats treated with N-nitrosopyrrolidine (NPYR). In this study, we detected and characterized a previously unidentified guanine adduct in liver DNA of NPYR-treated rats. The structure of this adduct was established as 7-(4-oxobutyl)guanine (adduct 2) by comparison with the synthetic standard and confirmed by NaBH4 reduction to 7-(4-hydroxybutyl)guanine. The level of adduct 2 in liver DNA of F344 rats treated with 450 mg/kg of NPYR by i.p. administration was 643 +/- 9 mumol/mol guanine, approximately one-third of the level of adduct 1. This study is the first to demonstrate the in vivo formation of a formylalkyl-substituted guanine adduct by a nitrosamine.     

Formation of the DNA adduct S-[2-(N7-guanyl)ethyl]glutathione from ethylene dibromide: effects of modulation of glutathione and glutathione S-transferase levels and lack of a role for sulfation

Hepatic S-[2-(N7-guanyl)ethyl]glutathione DNA adducts were determined in several strains of rats and mice after i.p. injection of a dose of 37 mg ethylene dibromide/kg body wt. More adducts were formed in rats than in mice, while no difference was noted among strains within each species. Removal of adducts in liver DNA was relatively slow in all animals tested. On the contrary, in vitro incubation of calf thymus DNA with ethylene dibromide and either rat cytosol or mouse cytosol gave rise to similar amounts of adduct, yet mouse cytosol showed much higher glutathione (GSH) S-transferase activity toward 1-chloro-2,4-dinitrobenzene. Human cytosol also activated ethylene dibromide, with the extent of conjugation being approximately half that of rat cytosol. Pretreatment of rats with phenobarbital or beta-naphthoflavone induced GSH S-transferases but did not increase the in vivo formation of DNA adducts, suggesting that concomitant induction of cytochrome P450 might abolish the effect of induction of GSH S-transferase by increasing the oxidation of ethylene dibromide. Butylated hydroxytoluene induced GSH S-transferase and also markedly increased DNA adduct levels. Disulfiram, a known cytochrome P450 inhibitor, significantly increased the formation of DNA adducts whereas it did not affect GSH S-transferase activity. Depletion of GSH by pretreatment of rats with diethylmaleate or buthionine sulfoximine resulted in decreased in vivo DNA adduct levels and the degree of reduction was well correlated with the extent of GSH depletion. In vitro incubation of tritiated S-(2-hydroxyethyl)GSH with calf thymus DNA in the presence of 3'-phosphoadenosine-5'-phosphosulfate and rat liver cytosol did not result in significant binding to DNA, suggesting that sulfation of the alcohol does not readily occur to add a leaving group and regenerate an episulfonium ion. These results suggest that induction of the Phase II enzyme GSH S-transferase can be detrimental in the case of ethylene dibromide and that decreases in GSH levels reduce DNA alkylation in rats.     

DNA adduct formation, cell proliferation and aberrant crypt focus formation induced by PhIP in male and female rat colon with relevance to carcinogenesis

2-Amino-1-methyl-6-phenylimidazo[4, 5-b]pyridine (PhIP) inducescolon tumors in male, but not female, F344 rats, We investigatedthe mechanisms leading to this difference by measuring the levelof PhlP-DNA adducts, the enhancement of cell proliferation andaberrant crypt focus (ACF) formation in colon mucosa. PhIP wasadministered in the diet at a level of 0.04% to both male andfemale F344 rats for 1–8 weeks. The level of DNA adductsin the colon mucosa was measured using the ³²P-postlabelingmethod. Four major PhlP-DNA adducts were detected in fairlyconstant proportions in all the animals examined. The levelof PhlP-DNA adducts in male and female rats was the same, indicatingno direct correlation between adduct levels and carcinogenesis.Labeling indices (LIs) were determined by measuring BrdU incorporationin rats after feeding with a PhIP diet for 4, 8 and 12 weeks.After 8 weeks administration the LI had increased 1.5-fold inthe colon of the male rats, but no increase was observed inthe female rats. ACF formation was examined after feeding witha PhIP diet for 14 weeks. The number of aberrant crypt fociwas 6.6 ± 1.5 per rat in males and 1.9 ± 0.5 perrat in females. Thus differences in colon tumor developmentin male and female rats takes place at an early stage(s). Ourresults suggest that, in addition to DNA adduct formation, enhancedproliferation contribites to the formation of ACFs, which arepremalignant lesions of the colon.     

Species- and length of exposure-dependent differences in the benzo(a)pyrene:DNA adducts formed in embryo cell cultures from mice, rats, and hamsters

The activation of benzo(a)pyrene (BaP) to DNA-binding metabolites in early-passage embryo cell cultures prepared from various species of rodents was investigated by exposing cells from mice (BALB/c and Sencar), rats (Wistar and Fischer 344), and Syrian hamsters to [3H]BaP for various lengths of time. The BaP:DNA adducts containing cis-vicinal hydroxyl groups such as those formed from 7 beta,8 alpha-dihydroxy-9 alpha,10 alpha-epoxy-7,8,9,10-tetrahydrobenzo(a)pyrene (anti-BaPDE) were separated from the other types of BaP:DNA adducts by immobilized boronate chromatography, and the individual adducts were analyzed by high-performance liquid chromatography. A number of BaP:DNA adducts were present in the DNA from the cultures from all three species after 5 h of BaP treatment. After a 24-h exposure to BaP, the mouse and hamster embryo cell DNA contained a large amount of the adduct formed by reaction of (+)-anti-BaPDE with the 2-amino group of deoxyguanosine (dGuo) and a small amount of a 7 beta,8 alpha-dihydroxy-9 beta,10 beta-epoxy-7,8,9,10-tetrahydrobenzo(a)pyrene:dGuo adduct. A large number of BaP:DNA adducts derived from 7 beta, 8 alpha-dihydroxy-9 beta,10 beta-epoxy-7,8,9,10-tetrahydrobenzo(a)pyrene and other unidentified BaP metabolites were present in rat embryo cell cultures at all times. Neither the Fischer 344 nor the Wistar rat embryo cell cultures had a significant amount of (+)-anti-BaPDE:dGuo adduct after 5 h of BaP treatment, and in the Wistar rat cells larger amounts of other adducts were present even after a 96-h exposure to BaP. In cell cultures from all three species the proportion of (+)-anti-BaPDE:dGuo adduct increased as the length of time of exposure to BaP increased. There are major differences in the metabolic activation of BaP to DNA binding metabolites in embryo cells from various species of rodents. However, the variations between cell cultures from different strains of rats or mice are not as great as the variations between cell cultures from different species. The time-dependent alterations in the BaP:DNA adducts indicate that analysis after various lengths of time of exposure to BaP is essential to characterize accurately the pathways of metabolic activation of BaP in cells from various species and tissues.     

7-cysteine-pyrrole conjugate: A new potential DNA reactive metabolite of pyrrolizidine alkaloids

Pyrrolizidine alkaloids (PAs) require metabolic activation to exert cytotoxicity, genotoxicity, and tumorigenicity. We previously reported that (±)-6,7-dihydro-7-hydroxy-1-hydroxymethyl-5H-pyrrolizine (DHP)-derived DNA adducts are responsible for PA-induced liver tumor formation in rats. In this study, we determined that metabolism of riddelliine and monocrotaline by human or rat liver microsomes produced 7-cysteine-DHP and DHP. The metabolism of 7-glutathionyl-DHP by human and rat liver microsomes also generated 7-cysteine-DHP. Further, reaction of 7-cysteine-DHP with calf thymus DNA in aqueous solution yielded the described DHP-derived DNA adducts. This study represents the first report that 7-cysteine-DHP is a new PA metabolite that can lead to DNA adduct formation.     

Elevated mitochondrial cisplatin-DNA adduct levels in rat tissues after transplacental cisplatin exposure

Although there is evidence that the toxic effects of cis- diamminedichloroplatinum(II) (cisplatin) include morphologically abnormal mitochondria, direct demonstrations of mitochondrial DNA damage by this chemotherapeutic agent have rarely been reported. Here we show that, in rats exposed to a single dose of cisplatin during gestation, cisplatin-DNA binding levels in both maternal and fetal liver and brain mitochondrial DNA are higher than those observed in genomic DNA. Pregnant F344/NCr rats were injected i.p. with either 5 or 15 mg cisplatin/kg body wt at 18 days of gestation and killed 24 h later. Cisplatin-DNA adducts were determined by dissociation-enhanced lanthanide fluoroimmunoassay using a cisplatin-DNA standard modified in the same range as the biological samples. Values for genomic cisplatin- DNA adducts in multiple maternal and fetal tissues have been presented elsewhere. Here, genomic DNA adduct levels for liver, brain, kidney and placenta are reported again for comparison with mitochondrial DNA adduct levels in the same tissues. In maternal and fetal brain, mitochondrial DNA adduct levels were approximately 7- to 50-fold higher than genomic DNA adduct levels, and in fetal liver they were approximately 2- to 16-fold higher than genomic DNA adduct levels. These studies demonstrate extensive cisplatin-DNA adduct formation in brain and liver mitochondria of fetal rats exposed transplacentally and suggest that mitochondrial DNA in some organs may be a particular target for cisplatin genotoxicity.      

Covalent binding of styrene to DNA in rat and mouse

Covalent binding of (7-³H)styrene (S) to DNA in vivo was measuredand evaluated in a quantitative manner in order to investigatewhether DNA adduct formation could form a mechanistic basisfor tumor induction in a carcinogenicity bioassay. [7-³H]S wasadministered by inhalation in a closed chamber to male and femaleCD rats and B6C3F1 mice. After 4.5–6 h (rats) and 6–9h (pools of four mice), S doses of 23–39 and 85–110mg/kg respectively had been metabolized. DNA was purified toconstant specific radioactivity which was measurable in allsamples. DNA was enzymatically degraded to the 3'-nucleotideswhich were separated by HPLC for the detection of radiolabelednucleotide-S adducts. The fractions with the normal nucleotidescontained most of the radioactivity. In mouse liver DNA, a minutebut significant level of adduct radioactivity was also detected.In the units of the Covalent Binding Index CBI = (µmoladduct/mol DNA nucleotide)/(mmol chemical/kg body wt), valuesof 0.05–0.09 and 0.07–0.18 were calculated for malesand females respectively. In the rat, no DNA adducts were detectablein the liver at a limit of detection of 0.1 CBI units. Two ofthe four lung samples of the female rats showed adduct-relatedradioactivity corresponding to 0.07 CBI units. The CBI valuesare compatible with styrene 7,8-oxide as the reactive intermediate.The data are compared with CBI values and carcinogenic potenciesof established genotoxic carcinogens. It is concluded that theDNA-binding potency of S is so low that significant tumor inductionin a standard bioassay for carcinogenicity is unlikely to bedue to DNA adduct formation alone. Consequences for a humanrisk estimation are discussed.     

DNA adduct levels in congenic rapid and slow acetylator mouse strains following chronic administration of 4-aminobiphenyl.

4-Aminobiphenyl (4-ABP) is a human and mouse bladder carcinogen. Epidemiological studies have shown that individuals with a slow acetylator phenotype, especially those exposed to high levels of carcinogenic aromatic amines, show an increased susceptibility to bladder cancer. In order to determine if a slow acetylator phenotype results in increased DNA damage, congenic mouse strains C57BL/6J and B6.A-Nat(s), which differ genetically at the acetyltransferase (EC 2.3.1.5) locus as homozygous rapid (Natr/Natr) and homozygous slow (Nat(s)/Nat(s)) acetylators respectively, were continuously administered 4-ABP.HCl (55-300 p.p.m.) in their drinking water for 28 days. The levels of covalently bound N-(deoxyguanosin-8-yl)-4-ABP-DNA adducts, which are believed to be critical for the initiation of tumors, were quantitated in the liver and bladder by 32P-postlabeling analysis. The levels of the hepatic DNA adduct increased with dose in both sexes, but were independent of the mouse acetylator genotype. At comparable doses, however, the levels of DNA adducts were 2-fold higher in the liver of the female as compared to the male animals. The DNA adducts also increased with dose in bladder of the male mice, but in contrast to the liver, the adduct levels were approximately 2-fold lower in the bladder DNA of the female mice. Also in contrast to the liver, the levels of bladder DNA adducts were significantly higher (P < or = 0.03) in the phenotypic rapid acetylator females compared to the slow acetylators at both 75 and 150 p.p.m. doses; the median levels of adducts were 10-20% higher in the phenotypic slow acetylator male bladders compared to their rapid acetylator counterparts. The results of these studies are consistent with the increased carcinogenicity of 4-ABP to the liver of female mice and the bladder of male mice. They further suggest that factors other than acetylator phenotype limit the extent of DNA adduct formation from 4-ABP in these mice.