Oral administration of naturally occurring coumarins leads to altered phase I and II enzyme activities and reduced DNA adduct formation by polycyclic aromatic hydrocarbons in various tissues of SENCAR mice.

Several naturally occurring coumarins, to which humans are routinely exposed in the diet, were previously found to inhibit P450-mediated metabolism of benzo[a]pyrene (B[a]P) and 7,12-dimethylbenz[a]anthracene (DMBA) in vitro, block DNA adduct formation in mouse epidermis and inhibit skin tumor initiation by B[a]P and/or DMBA when applied topically to mice. The present study was designed to investigate the effects of two of these compounds, of the linear furanocoumarin type, when given orally (70 mg/kg per os, four successive daily doses), on P450 and glutathione S-transferase (GST) activities and DNA adduct formation by B[a]P and DMBA in various mouse tissues. Imperatorin and isopimpinellin significantly blocked ethoxyresorufin O-deethylase (EROD) and pentoxyresorufin O:-dealkylase (PROD) activities in epidermis at 1 and 24 h after oral dosing. Imperatorin and isopimpinellin modestly inhibited EROD activities in lung and forestomach at 1 h and significantly inhibited PROD activities in lung and forestomach at 1 h after the final oral dose. Twenty-four hours after the final oral dose of imperatorin or isopimpinellin EROD and PROD activities remained inhibited in epidermis and lung. However, forestomach P450 activity had returned to control levels. Interestingly, imperatorin and isopimpinellin treatment inhibited liver EROD activity at 1 h, had no effect on PROD activity at this time point, but elevated both these enzyme activities at 24 h. Elevated EROD and PROD activities coincided with elevated hepatic P450 content. Imperatorin and isopimpinellin treatment also increased liver cytosolic GST activity at both 1 and 24 h after the final oral dose by 1.6-fold compared with corn oil controls. Oral administration of imperatorin and isopimpinellin also had a protective effect against DNA adduct formation by B[a]P and DMBA. Imperatorin pretreatment decreased formation of DNA adducts by DMBA in forestomach. Pretreatment with isopimpinellin led to reduced DNA adduct levels in liver (B[a]P), lung (B[a]P) and mammary epithelial cells (DMBA). These results suggest that imperatorin and isopimpinellin may have potential chemopreventive effects when administered in the diet.

Both (+/-)syn- and (+/-)anti-7,12-dimethylbenz[a]anthracene-3,4-diol-1,2-epoxides initiate tumors in mouse skin that possess -CAA- to -CTA- mutations at Codon 61 of c-H-ras.

We have determined the tumor-initiating activity of (+/-)syn- and (+/-)anti-7,12-dimethylbenz[a]anthracene-3,4-diol-1,2-epoxide (syn- and anti-DMBADE), the two metabolically formed bay-region diol epoxides of DMBA, and we have also analyzed mutations in the H-ras gene from tumors induced by these compounds. Using a two-stage, initiation-promotion protocol for tumorigenesis in mouse skin, we have found that both syn- and anti-DMBADE are active tumor initiators, and that the occurrence of papillomas is carcinogen dose dependent. All of the papillomas induced by syn-DMBADE (a total of 40 mice), 96% of those induced by anti-DMBADE (a total of 25 mice), and 94% of those induced by DMBA (a total of 16 mice) possessed a -CAA- to -CTA- mutation at codon 61 of H-ras. No mutations in codons 12 or 13 were detected in any tumor. Topical application of syn- and anti-DMBADE produced stable adducts in mouse epidermal DNA, most of which comigrated with stable DNA adducts formed after topical application of DMBA. Further analysis of the data showed that levels of the major syn- and anti-DMBADE-deoxyadenosine adducts formed after topical application of DMBA are sufficient to account for the tumor-initiating activity of this carcinogen on mouse skin. Previously, we showed that both the syn- and anti-DMBADE bind to the adenine (A182) at codon 61 of H-ras. Collectively, these results indicate that the adenine adducts induced by both bay-region diol epoxides of DMBA lead to the mutation at codon 61 of H-ras and, consequently, initiate tumorigenesis in mouse skin.

Characterization of the major DNA adducts in the liver of rats chronically exposed to tamoxifen for 18 months.

Our previous study has shown that chronic exposure to tamoxifen (TAM) induced formation of high levels of DNA adducts in the liver, the target tissue of TAM-induced carcinogenesis in rats. One of the major DNA adducts (spot 1), as detected by 32P-postlabeling, accounted for 53% of the total adducts. To characterize this major adduct, the current study has compared spot 1 with two previously identified TAM-DNA adducts, i.e. alpha-TAM-N2-deoxyguanine (alpha-TAM-N2-dG) and alpha-N-desmethyl TAM-N2-deoxyguanine (alpha-N-dmTAM-N2-dG) by various rechromatography methods. It was found that spot 1 was further resolved into two fractions during rechromatography analysis, one fraction co-migrated with the alpha-TAM-N2-dG and the other fraction co-migrated with the alpha-N-dmTAM-N2-dG. These findings have demonstrated that chronic exposure to tamoxifen induced the same major DNA adducts, i.e. alpha-TAM-N2-dG and alpha-N-dmTAM-N2-dG as those detected in acutely exposed rats.

Analysis of aromatic DNA adducts and 7,8-dihydro-8-oxo- 2'-deoxyguanosine in lymphocyte DNA from a case-control study of lung cancer involving minority populations.

The purpose of this study was to examine the level of smoking-related aromatic DNA adducts and oxidative DNA damage in current smokers from a lung cancer case-control study in African Americans and Mexican Americans. In addition, mutagen sensitivity (bleomycin-induced chromatid breaks), a marker of genetic susceptibility, was assessed in these patients and correlated with the level of DNA damage. Lymphocyte DNA from cases and age-, sex-, and ethnicity-matched controls was analyzed for aromatic DNA adducts (43 cases and 47 controls) and the level of 7, 8-dihydro-8-oxo-2'-deoxyguanosine (8-oxo-dG) was determined in 46 cases and 48 controls using (32)P-postlabeling. Overall, lung cancer cases had significantly (P < 0.05) higher levels of aromatic DNA adducts and 8-oxo-dG (mean+/-SEM; 6.03+/-1.16/10(8) nucleotides and 5.82+/-0.77/10(5) nucleotides, respectively) compared to the controls (2.80+/-0.36/10(8) nucleotides and 3.65+/-0.56/10(5) nucleotides, respectively). The case-control differences for these two biomarkers were especially evident in current smokers. Both male and female lung cancer cases had higher levels of aromatic DNA adducts compared to the corresponding controls but only in men was the difference statistically significant (P=0.002). Cases who started smoking at earliest age had highest levels of aromatic DNA adducts and 8-oxo-dG. The level of aromatic DNA adducts in lung cancer cases, but not controls, was positively correlated with bleomycin-induced chromatid breaks (P=0.011). In contrast, the level of 8-oxo-dG was not correlated with mutagen sensitivity in either cases or controls or with the level of aromatic DNA adducts. The data suggest that levels of both aromatic DNA adducts and 8-oxo-dG may be useful in predicting risk of lung cancer in these minority populations. The correlation between aromatic DNA adducts and mutagen sensitivity in lung cancer cases and the trend for higher levels of DNA damage in cancer cases who started smoking earliest are particularly interesting and merit further study.

Analysis of highly polar DNA adducts formed in SENCAR mouse epidermis following topical application of dibenz[a,j]anthracene.

The formation of DNA adducts in mouse epidermis has been examined following topical application of dibenz[a,j]anthracene (DB[a,j]A) and its metabolites, i.e., DB[a,j]A-3,4-diol, DB[a,j]A-3,4-10, 11-bis-diol, DB[a,j]A-3,4-8,9-bis-diol, 10-OH-DB[a,j]A-3,4-diol, or 11-OH-DB[a,j]A-3,4-diol, using a 32P-postlabeling assay. At initiating doses (400-1600 nmol), DB[a,j]A produced at least 23 DNA adduct spots, including four less polar (derived from the bay-region syn- and anti-diol-epoxides) and 19 highly polar DNA adducts. DB[a, j]A-3,4-diol produced 13 DNA adduct spots, four less polar and nine highly polar DNA adducts, and DB[a,j]A-3,4-10,11-bis-diol produced nine highly polar DNA adducts. Eight and seven of the highly polar DNA adducts generated by DB[a,j]A-3,4-diol and DB[a,j]A-3,4-10, 11-bis-diol, respectively, migrated in the chromatography system like the highly polar DNA adducts produced by the parent compound. Sufficient amounts of radioactivity were associated with highly polar adduct spots 11, 13, and 22 to confirm their chromatogaphic identity in DNA samples from DB[a,j]A-, DB[a,j]A-3,4-diol-, and DB[a, j]A-3,4-10,11-bis-diol-treated mice. 10-OH-DB[a,j]A-3,4-diol and 11-OH-DB[a,j]A-3,4-diol did not produce any highly polar DNA adducts that could be detected under our experimental conditions. At an initiating dose of 400 nmol, DB[a,j]A, DB[a,j]A-3,4-diol, and DB[a, j]A-3,4-10,11-bis-diol produced 22.4 +/- 13.0, 15.6 +/- 10.1, and 5. 5 +/- 0.3 (mean +/- SD) adducts/10(9) nucleotides, of which 77, 65, and 100%, respectively, represented highly polar DNA adducts. At the same dose of 400 nmol per mouse, DB[a,j]A and its 3,4-diol were able to initiate papillomas in SENCAR mouse skin (3.08 +/- 1.89 and 3.48 +/- 2.72 papillomas per mouse, respectively, after 16 weeks of promotion with 12-O-tetradecanoyl phorbol 13-acetate), while the 3, 4-10,11-bis-diol of DB[a,j]A was inactive as a tumor initiator. A quantitative correlation (r = 0.935; p = 0.0196) between levels of less polar DNA adducts and tumor-initiating activity of DB[a,j]A, DB[a,j]A-3,4-diol, and anti-DB[a,j]ADE was observed. This study demonstrates that the highly polar DNA adducts formed from DB[a,j]A in mouse epidermis arise primarily from the DB[a,j]A-3,4-10, 11-bis-diol. However, the contribution of this metabolite to the tumor-initiating activity of DB[a,j]A appears to be small.

High levels of endogenous DNA adducts (I-compounds) in pig liver. Modulation by high cholesterol/high fat diet.

I (indigenous)-compounds are bulky endogenous DNA adducts which are detected by 32P-postlabeling in unexposed animals. I-compound levels in rodents depend on age, species, strain, gender, tissue, diet, and chemical exposure. There are two classes of I-compounds, type I and type II. While many type I I-compounds may not reflect DNA damage, type II I-compounds have been identified as oxidative DNA lesions some of which can be produced in vitro under Fenton reaction conditions. In rats, caloric restriction (CR) increases the levels of many type I I-compounds compared with ad libitum fed animals, while high fat diet has the opposite effect. Here, we have tested whether hepatic DNA of a non-rodent mammal, the pig, contains I-compounds and whether feeding a high cholesterol/high fat (HC/HF) diet modulates their levels, assuming this would affect the formation of lipid-related precursors and cause oxidative stress. Male Yorkshire pigs aged 2 months old, were fed either control or HC/HF diet (control diet supplemented with 2% cholesterol and 19% lard) for 2 months. Pig liver DNA contained at least 19 type I and five type II I-compounds. Among the former, only five matched corresponding spots in rat liver DNA, while all the latter DNA lesions were detected in both species. The levels of both types of DNA modifications were six to eight-fold higher in pig DNA. HC/HF diet reduced levels of many type I I-compounds up to several fold but had little effect on the oxidative lesions. Several type I I-compounds showed negative linear correlations with serum cholesterol levels, while this association was positive for total type II I-compounds. The substantially elevated steady-state levels of bulky endogenous DNA adducts in the species with the longer life expectancy were surprising. Thus, for the first time, an intimate link between nutritional status and endogenous DNA modifications has been established in a non-rodent system. We propose that in order to explain our observations, differences in diet composition, antioxidant defenses, and DNA repair, as well as cytochrome P450 modulation of precursor levels and hormonal effects need to be considered.

Partial characterization of two major liver I-compounds as unstable adducts which are readily hydrolyzed to unmodified guanine nucleotides.

I-compounds are endogenous bulky DNA modifications which are detected by nuclease P1-enhanced 32P-post-labeling in tissue DNA of animals not knowingly exposed to carcinogens. Their profiles and levels depend inter alia on animal age, species, strain, tissue, gender, diet and exposure to chemicals such as cytochrome P450 inducers and carcinogens. Due to lack of sufficient material obtainable from in vivo sources, chemical structures of I-compounds and their parent normal bases have not yet been identified. In this report we provide 32P-post-labeling and chromatographic evidence that two prominent I-compounds, herein called C1 and C2, which occur at relatively high levels in pig liver DNA are guanine derivatives. This result was obtained by showing that both compounds, isolated from 32P-post-labeling thin-layer maps, were chemically unstable, i.e. they could be readily hydrolyzed to 32P-post-labeled deoxyguanosine 3',5'-bisphosphate by heating in water. C1 appeared particularly labile, undergoing hydrolysis during thin-layer chromatography at pH 3.3 without heating. Several other I-compounds and adducts, as well as the four normal DNA nucleotides, were, however, highly resistant to hydrolysis under the conditions used here. The possible significance of these findings will be briefly discussed.

Analysis of 7-methylbenz[a]anthracene-DNA adducts formed in SENCAR mouse epidermis by 32P-postlabeling.

The present study has analysed the DNA adducts formed in SENCAR mouse epidermis following topical application of 7-methylbenz[a]anthracene (7-MBA). Mice were treated with 400 nmol of 7-MBA, which represents an initiating dose of this hydrocarbon for SENCAR mice. DNA adducts were analysed 24 h after topical application of the hydrocarbon by 32P-postlabeling coupled with either HPLC analysis or an improved TLC procedure giving better resolution of DNA adducts through the use of a D6 solvent [isopropanol:4N NH4OH (1:1)] following D5. Twenty-four hours after topical application of 400 nmol 7-MBA, the level of total covalent binding was 0.37 +/- 0.07 pmol/mg DNA as determined by 32P-postlabeling. This level of binding correlated well with the relative tumor initiating activity of this hydrocarbon compared to 7,12-dimethylbenz[a]anthracene (6.4 +/- 0.01 pmol/mg DNA) and dibenz[a,j]anthracene (0.03 +/- 0.01 pmol/mg DNA). Analysis of the 32P-labeled 3',5'-diphosphodeoxyribonucleosides by HPLC and TLC revealed the presence of deoxyguanosine (dGuo) and deoxyadenosine (dAdo) adducts formed from both the anti- and syn-bay-region diol-epoxides of 7-MBA (anti- and syn-7-MBADEs). The major DNA adduct derived from 7-MBA in mouse epidermis was tentatively identified as (+) anti-7-MBADE-trans-N2-dGuo. In addition, a minor dGuo adduct derived from the bay-region syn-diol-epoxide of 7-MBA was detected as well as a minor dAdo adduct from this diol-epoxide. Another minor dAdo adduct was also detectably present which arose from either the anti- or syn-diol epoxide. Furthermore, several unidentified DNA adducts were present in both HPLC and TLC chromatograms of DNA samples from 7-MBA-treated mice. These results are discussed in terms of the role of specific 7-MBA-DNA adducts in tumor initiation by this hydrocarbon.

Rapid decreases in indigenous covalent DNA modifications (I-compounds) of male Fischer-344 rat liver DNA by diquat treatment.

I-compounds are indigenously appearing covalent DNA modifications that can be detected by 32P-postlabeling assay in tissues of normal animals without known exposure to any carcinogens or toxins. Although these compounds have not been structurally identified, indirect evidence from earlier work suggested the possibility of involvement of molecular fragments derived from lipid peroxides. Diquat is a herbicide that stimulates lipid peroxidation and massive intrahepatic oxidant stress through redox cycling-mediated generation of reactive oxygen species. In the present study, we examined the effects of diquat on hepatic I-compounds of male Fischer-344 rats. Two groups of rats, approximately 14 weeks and 8 weeks old, were given a hepatotoxic dose (0.1 mmol/kg) of diquat or equal volumes of saline, i.p. Two and 6 h later plasma alanine aminotransferase (ALT) activities were measured and hepatic DNA I-compound levels were examined by nuclease P1-enhanced 32P-postlabeling. Elevated ALT activities were observed in some animals in both groups, at both time points, but considerable inter-animal variation was seen. A total of 15-16 I-compound fractions were measured in control and in diquat-treated animals, but no extra spots indicative of treatment-induced adducts were detected. Despite the qualitative similarities, the quantities of individual I-compounds were markedly decreased at 2 h in diquat-treated animals of both age groups. In 14 week old rats the hepatic I-compound contents were decreased at 2 h by 22-59%, which was statistically significant (ANOVA, P < 0.05) for all of the 9 polar I-compound fractions and none of the non-polar fractions. Eleven I-spots from this group showed significant negative linear correlations (P < 0.05) with ALT values. In 8 week old rats treated with diquat a 22-43% depletion in I-compound contents was statistically significant for 4 of the 7 nonpolar and 2 of the 8 polar adduct fractions, but there was no significant correlation of I-compound contents with ALT values at the 2 h time point. By 6 h most of the I-spot levels had returned to normal or above normal values in both groups of animals. While most I-spots from 14 week old rats did not correlate with ALT levels at 6 h, two I-spots displayed positive correlations in the 8 week group. Overall, the susceptibility to diquat-associated DNA alterations appeared to differ with age.(ABSTRACT TRUNCATED AT 400 WORDS)

32P-postlabeling of bile components: bulky adduct-like behavior in polyethyleneimine-cellulose thin layer chromatography.

The 32P-postlabeling assay has been used widely in carcinogen-DNA adduct analysis because of its sensitivity and reproducibility. Cloned T4 polynucleotide kinase (PNK), routinely used in this assay, phosphorylates the 5'-OH groups of adducted nucleotides in the presence of [gamma-32P]ATP. However, as an exception to this property, PNK has been reported to phosphorylate non-adducted carcinogen metabolites, such as tetrol derivatives of benzo[a]pyrene and chrysene. Also, PNK phosphorylates both 5'-OH and 3'-OH groups of safrole-adducted deoxydinucleoside monophosphates having an unmodified purine in the 3'-position. In the present study we show that T4 PNK catalyzed the transfer of [32P]phosphate from [gamma-32P]ATP to rat bile components or purified bile acids (derivatives of 3 alpha-hydroxy-5 beta-cholanic acid) in the absence of nucleic acids or nucleases. However, labeling of the bile acids appeared over 100,000-fold less efficient than labeling of 2'-deoxyadenosine-5'-monophosphate. There was no reaction in the absence of bile components or PNK. Dehydrocholic acid, which lacks hydroxyl groups, was resistant to phosphorylation. On polyethyleneimine-cellulose TLC maps, 32P-labeled rat bile extract gave an array of non-polar radioactive spots which resembled carcinogen-DNA adducts, while 32P-labeled purified bile acids each gave a single spot. These 32P-labeled products liberated 32Pi upon incubation with prostatic acid phosphatases. Two of the radioactive spots obtained from rat bile were identified as phosphorylated taurocholic and taurodeoxycholic acids by co-chromatography with 32P-labeled standards. These findings demonstrate for the first time that PNK is able to phosphorylate natural products other than nucleotides and further emphasize the need to rule out contamination with bile acids and possibly other bulky/hydrophobic alcohols when analyzing DNA samples by 32P-postlabeling.

Rapid purification of tRNA(Lys) from rat liver.

Fast protein liquid chromatography (FPLC) system using Mono Q (HR 5/5) anion-exchange column chromatography followed by highly cross-linked urea-polyacrylamide gel electrophoresis (urea-PAGE) was used for the purification of lysine-specific tRNA (tRNA(Lys)) from rat liver. Crude tRNA from rat liver was fractionated with a linear gradient of NaCl (0.3-0.8 M) in triethanolamine-HCl buffer, pH 4.5, and the activity of tRNA(Lys) was found to elute between 0.51 and 0.57 M NaCl. Using this concentration range of NaCl, tRNA(Lys) was refractionated on the same column with a shallow gradient, where a single peak of tRNA(Lys) activity was obtained. tRNA(Lys)-rich fractions recovered from the second run were electrophoretically separated on 16% polyacrylamide-7 M urea gel into one major band and three minor bands. The major band showed a specific activity of 997 pmols/A260 U for tRNALys with a 43-fold purification and approximately 17% recovery. The minor bands displayed negligible or no activity for lysine. tRNA(Lys) obtained by this method was found to be homogeneous by competitive aminoacylation. The advantages of FPLC followed by urea-PAGE in the purification of an amino acid-specific tRNA over conventional column chromatography are discussed.

5-Fluorouracil induces structural changes in rat liver tRNA.

5-fluorouracil (FUra) has been shown to modulate the aminoacylation function of rat liver tRNA. The present study was aimed at studying the structure-function relationship of FUra-substituted tRNA. Male Wistar rats (2-3 month old) were given a single i.p. injection of FUra at 50, 250, or 500 mg/kg body wt. and FUra-substituted total liver tRNA, i.e. tRNA(FUra50, 250, and 500, respectively, were isolated 3 h later. Normal tRNA (tRNA(N)) was isolated from saline-treated control rats. Thermal denaturation studies showed higher melting temperatures for tRNA(FUra) compared to tRNA(N). Heat denaturation followed by renaturation of total tRNA did not affect the activity of tRNA(N) and tRNA(FUra50), where as tRNA(FUra250 and 500) lost 35% and 72% of activity, respectively, compared to the corresponding group of non-denatured tRNA. Antibodies specific to rat liver tRNA recognized normal and FUra-substituted tRNA in the order of tRNA(N) > tRNA(FUra50) > or = tRNA(FUra250) > tRNA(FUra500) in an avidin-biotin micro-enzyme linked immunosorbant assay. tRNA(N) or tRNA(FUra50) preincubated with tRNA antiserum showed 74% and 59% of aminoacylation activity, respectively, compared to that of corresponding tRNA preincubated with normal rabbit IgG. However, activities of similarly treated tRNA(FUra250 and 500) were not affected. The observations of possible changes in the secondary structure of rat liver tRNA upon incorporation of FUra are discussed.

5-Fluorouracil modulates aminoacylation of rat liver tRNA.

Among the different classes of RNA, the effect of 5-fluorouracil (FUra), an anticancer drug, has been studied extensively on tRNA function in E. coli, but to a limited extent in eukaryotes, with specific reference to the mammalian system. Here we compared the aminoacylation function of rat liver tRNA substituted with FUra. Three hours after a single i.p. injection of 50, 250 or 500 mg/kg body wt. of FUra, total tRNAFUra50, tRNAFUra250, and tRNAFUra500, respectively were isolated from the livers of 2-3 month old male Wistar rats. The activity of tRNAFUra was compared with normal tRNA (tRNAN) isolated from saline-treated controls. tRNAFUra50 accepted [14C]-labeled total and five individual amino acids (lysine, aspartic acid, methionine, tryptophan, and serine) at a significantly higher rate compared to tRNAN. On contrary, tRNAFUra250 & 500 displayed a dose-dependent inhibition in aminoacylation with total amino acids, lysine, and methionine. Acceptance of leucine was inhibited by tRNAFUra in a dose-dependent way. Overall, the amino acid acceptance was variable among the three populations of rat liver tRNAFUra isolated with varying doses of FUra and the possible reasons for the altered function of tRNA are discussed.

Circadian rhythm of covalent modifications in liver DNA.

32P-postlabeling analysis recently revealed that in addition to 5-methylcytosine, mammalian DNA contains covalently modified nucleotides of unknown structures and functions termed I-compounds whose levels increase with age. I-compound levels, in addition, depend on species, strain, sex, tissue, and diet and are generally lowered by carcinogen exposure. As shown here, levels of several non-polar I-compounds in liver DNA of untreated male C3H mice were elevated 2 to 8.5 times at 1800 h and 2400 h as compared to 0600 h and 1200 h, while polar I-compounds and persistent carcinogen-DNA adducts induced by safrole were unaffected by time of day. In liver DNA of male F-344 rats 4 non-polar I-compounds and 4 polar I-compounds showed significant circadian rhythm at 2000 h compared to 0800 h. This novel circadian variation of DNA structure implies mechanisms precisely regulating I-compound levels in vivo and may conceivably be linked to diurnal differences of DNA synthesis and gene expression.