Synthesis of depurinating DNA adducts formed by one-electron oxidation of 7H-dibenzo[c,g]carbazole and identification of these adducts after activation with rat liver microsomes

7H-Dibenzo[c,g]carbazole: Metabolic pathways and toxicity

7H-Dibenzo[c,g]carbazole (DBC), a local and systemic carcinogen in animal studies, is a common environmental pollutant. It generally co-occurs in a variety of organic complex mixtures derived from incomplete combustion of organic matter. Despite high lipophilicity, DBC is more water-soluble and faster metabolized than the homocyclic aromatics. Moreover, greater polarity, high bioaccumulation potential, and persistence in the environment may imply DBC's higher biological significance and impact on human health, even at lower concentrations. The biotransformation pathways of DBC are incompletely known and the ultimate carcinogenic metabolite(s) are not clearly identified as yet. Structure-biological studies suggest two ways of activation: at the ring carbon atoms and at the pyrrole nitrogen. It is supposed that the particular pathway of biotransformation might be connected with the tissue/organ specificity of DBC. Cytochrome P450 (CYP) family of enzymes plays a pivotal role in the metabolism of DBC; though, the one-electron activation and the aldo-keto reductase-catalyzed oxidation are also involved in metabolic activation. Additionally, DBC can be photoactivated even at physiologically relevant doses of UVA light due to the extended aromatic ring system resulting in strong genotoxicity and oxidative stress. The goal of this review is to summarize current knowledge on mechanisms of DBC activation and possible implications for toxicity, genotoxicity, and carcinogenicity.

Ultraviolet A radiation potentiates the cytotoxic and genotoxic effects of 7 H-dibenzo[c,g]carbazole and its methyl derivatives

7H-Dibenzo[c,g]carbazole (DBC) is a heterocyclic aromatic hydrocarbon that is carcinogenic in many species and tissues. DBC is a common environmental pollutant, and is therefore constantly exposed to sunlight. However, there are limited data exploring the toxicity of DBC photoexcitation products. Here, we investigated the impact of ultraviolet (UV) A radiation on the biological activity of DBC and its methyl derivatives, 5,9-dibenzo[c,g]carbazole and N-methyl dibenzo[c,g]carbazole, on human skin HaCaT keratinocytes. Co-exposure of HaCaT cells to UVA and DBC derivatives resulted in a sharp dose-dependent decrease in cell survival and apparent changes in cell morphology. Under the same treatment conditions, significant increases in DNA strand breaks, intracellular reactive oxygen species, and oxidative damage to DNA were observed in HaCaT cells. Consistent with these results, an apparent inhibition in superoxide dismutase, but not glutathione peroxidase activity, was detected in cells treated with DBC and its derivatives under UVA irradiation. The photoactivation-induced toxicity of individual DBC derivatives correlated with the electron excitation energies approximately expressed as the energy difference between the highest occupied and the lowest vacant molecular orbital. Our data provide the first evidence that UVA can enhance the toxicity of DBC and its derivatives. Photoactivation-induced conversion of harmless chemical compounds to toxic photoproducts associated with reactive oxygen species generation may substantially amplify the adverse health effects of UVA radiation and contribute to increased incidence of skin cancer.

Genotoxicity of 7H-dibenzo[c,g]carbazole and its tissue-specific derivatives in human hepatoma HepG2 cells is related to CYP1A1/1A2 expression

The goal of this study was to investigate the genotoxicity of 7H-dibenzo[c,g]carbazole (DBC), a ubiquitous environmental pollutant, and its methyl derivatives, 5,9-dimethylDBC (DiMeDBC), a strict hepatocarcinogen, and N-methylDBC (N-MeDBC), a specific sarcomagen in human hepatoma HepG2 cells, and to infer potential mechanisms underlying the biological activity of particular carcinogen. All dibenzocarbazoles, regardless the tissue specificity, induced significant DNA strand break levels and micronuclei in HepG2 cells; though a mitotic spindle dysfunction rather than a chromosome breakage was implicated in N-MeDBC-mediated micronucleus formation. While DBC and N-MeDBC produced stable DNA adducts followed with p53 protein phosphorylation at Ser-15, DiMeDBC failed. A significant increase in DNA strand breaks following incubation of exposed cells with a repair-specific endonuclease (Fpg protein) suggested that either oxidative DNA damage or unstable DNA-adducts might underlie DiMeDBC genotoxicity in human hepatoma cells. DiMeDBC and N-MeDBC increased substantially also the amount of CYP1A1/2 expression in HepG2 cells. Pretreatment of cells with substances affecting AhR-mediated CYP1A family of enzymes expression; however, diminished DiMeDBC and N-MeDBC genotoxicity. Our data clearly demonstrated differences in the mechanisms involved in the biological activity of DiMeDBC and N-MeDBC in human hepatoma cells; the genotoxicity of these DBC derivatives is closely related to CYP1A1/2 expression.

Differences in DNA damage and repair produced by systemic, hepatocarcinogenic and sarcomagenic dibenzocarbazole derivatives in a model of rat liver progenitor cells

Liver progenitor (oval) cells are a potential target cell population for hepatocarcinogens. Our recent study showed that the liver carcinogens 7H-dibenzo[c,g]carbazole (DBC) and 5,9-dimethyldibenzo[c,g]carbazole (DiMeDBC), but not the sarcomagen N-methyldibenzo[c,g]carbazole (N-MeDBC), induced several cellular events associated with tumor promotion in WB-F344 cells, an in vitro model of liver oval cells [J. Vondracek, L. Svihalkova-Sindlerova, K. Pencikova, P. Krcmar, Z. Andrysik, K. Chramostova, S. Marvanova, Z. Valovicova, A. Kozubik, A. Gabelova, M. Machala, 7H-Dibenzo[c,g]carbazole and 5,9-dimethyldibenzo[c,g]carbazole exert multiple toxic events contributing to tumor promotion in rat liver epithelial 'stem-like' cells, Mutat. Res. Fundam. Mol. Mech. Mutagen. 596 (2006) 43-56]. In this study, we focused on the genotoxic effects generated by these dibenzocarbazoles in WB-F344 cells to better understand the cellular and molecular mechanisms involved in hepatocarcinogenesis. Lower IC(50) values determined for DBC and DiMeDBC, as compared with N-MeDBC, indicated a higher sensitivity of WB-F344 cells towards hepatocarcinogens. Accordingly, DBC produced a dose-dependent DNA-adduct formation resulting in substantial inhibition of DNA replication and transcription. In contrast, DNA-adduct number detected in DiMeDBC-exposed cells was almost negligible, whereas N-MeDBC produced a low level of DNA adducts. Although all dibenzocarbazoles significantly increased the level of strand breaks (p<0.05) and micronuclei (p<0.001) after 2-h treatment, differences in the kinetics of strand break rejoining were found. The strand break level in DiMeDBC- and N-MeDBC-exposed cells returned to near the background level within 24h after treatment, whereas a relatively high DNA damage level was detected in DBC-treated cells up to 48h after exposure. Additional breaks detected after incubation of DiMeDBC-exposed WB-F344 cells with a repair-specific endonuclease, along with a nearly 3-fold higher level of reactive oxygen species found in these cells as compared with control, suggest a possible role of oxidative stress in DiMeDBC genotoxicity. We demonstrated qualitative differences in the DNA damage profiles produced by hepatocarcinogens DBC and DiMeDBC in WB-F344 cells. Different lesions may trigger distinct cellular pathways involved in hepatocarcinogenesis. The low amount of DNA damage, together with an efficient repair, may explain the lack of hepatocarcinogenicity of N-MeDBC.

Metabolic activation of polycyclic and heterocyclic aromatic hydrocarbons and DNA damage: a review

Polycyclic aromatic hydrocarbons (PAHs) and heterocyclic aromatic compounds (HACs) constitute a major class of chemical carcinogens present in the environment. These compounds require activation to electrophilic metabolites to exert their mutagenic or carcinogenic effects. There are three principal pathways currently proposed for metabolic activation of PAH and HAC: the pathway via bay region dihydrodiol epoxide by cytochrome P450 enzymes (CYPs), the pathway via radical cation by one-electron oxidation, and the ortho-quinone pathway by dihydrodiol dehydrogenase (DD). In addition to these major pathways, a brief description of a minor metabolic activation pathway, sulfonation, for PAHs that contain a primary benzylic alcoholic group or secondary hydroxyl group(s) is included in this review. The DNA damages caused through the reactive metabolites of PAH/HAC are described involving the DNA covalent binding to form stable or depurinating adducts, the formation of apurinic sites, and the oxidative damage. The review emphasizes the chemical/biochemical reactions involved in the metabolic processes and the chemical structures of metabolites and DNA adducts.

Benzo[a]pyrene diol epoxide forms covalent adducts with deoxycytidylic acid by alkylation at both exocyclic amino N(4) and ring imino N-3 positions

The carcinogen 7r,8t-dihydroxy-9t,10t-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (anti-BPDE) alkylates DNA at dGuo, dAdo, and dCyd. dCyd adducts, formed in small amounts, elute near the more abundant dGuo adducts. We isolated the dCyd adducts formed with dCMP. Each BPDE enantiomer forms three major adducts with dCMP, two cis and one trans. The trans adduct and one of the cis adducts form by alkylation at exocyclic N(4), while the second cis adduct is a dUrd adduct formed by alkylation at ring N-3 followed by deamination. Epoxide ring-opening geometries were assigned on the basis of halide and temperature effects on adduct yield, the sign of the major CD band, and benzo ring proton NMR coupling constants. One of each set of cis adducts is fluorescent (FL), and the other is nonfluorescent (NF). The trans and FL cis adducts have fluorescence quantum yields 40-50% of that of the BPDE hydrolysis product. The long wavelength UV maxima of the FL and NF cis adducts are red-shifted 1 and 3 nm relative to the trans adduct. (1)H NMR deuterium exchange experiments indicate that in the trans and FL cis adducts N(4)-H is coupled to C10-H. Adduct formation experiments with methyl-protected Cyd derivatives show that NF cis adducts result from alkylation at N-3. MS results, pK(a) measurements, and dUrd alkylation experiments indicate that the N-3 dCyd adducts spontaneously deaminate to dUrd adducts. NMR coupling constants show that in the NF cis adduct the C7 and C8 substituents are quasi equatorial and the C9 substituent is quasi axial, unlike in other cis BPDE adducts. (1)H NOESY spectra of the (-)-BPDE NF cis adduct reveal that it exists in two conformers. Molecular modeling shows that the conformers result from two low-energy conformations of very similar energies with the pyrimidine in opposite orientations, separated by significant barriers to rotation of the uracil moiety.

DNA adduct formation by 7H-dibenzo[c,g]carbazole and its tissue- and organ-specific derivatives in Chinese hamster V79 cell lines stably expressing cytochrome P450 enzymes

The cytochrome P4501A subfamily (CYP1A) is involved in the metabolic activation of 7H-dibenzo[c,g]carbazole (DBC) and its tissue- and organ-specific derivatives, N-methyldibenzo[c,g]carbazole (MeDBC)and 5,9-dimethyldibenzo[c,g]carbazole (diMeDBC). In this study, we have evaluated the relationship between the tissue specificity and (32)P-postlabeled adduct patterns produced by these compounds by using a panel of Chinese hamster V79 cell lines stably expressing human CYP1A1 and CYP1A2 and/or N-acetyltransferase. Treatment of the parental cell lines V79MZ and V79NH, which are devoid of any CYP activity, with DBC and its derivatives did not result in detectable adducts. The highest DNA adduct levels were found in CYP1A1-expressing V79MZh1A1 cells after DBC and MeDBC treatment (24.5 +/- 7.2 and 16.2 +/- 3.6 adducts/10(8) nucleotides, respectively). Exposure of this cell line to DBC resulted in five distinct spots, while six spots with different chromatographic mobilities were detected in MeDBC-treated cells. DiMeDBC produced only very low levels of DNA adducts in V79MZh1A1 cells. DBC and MeDBC formed relatively low levels of DNA adducts in CYP1A2-expressing V79MZh1A2 cells (0.7 +/- 0.2 and 2.1 +/- 1.2 adducts/10(8) nucleotides, respectively). DBC formed three weak spots and MeDBC five spots in V79MZh1A2 cells, and all the spots had different chromatographic mobilities. In contrast, diMeDBC did not induce any DNA adducts in these cells, although diMeDBC induced a significant dose-dependent increase in micronucleus frequency under similar treatment conditions (r = 0.76; P < 0.001). The significant increase in DNA damage in the Comet assay following incubation of exposed cells with a repair-specific endonuclease (Fpg protein) suggests that base modifications such as 8-oxodG or Fapy-adducts might be responsible for the genotoxicity of diMeDBC in V79MZh1A2 cells. The similarities between the DNA adduct patterns produced by DBC and MeDBC in V79MZh1A1 and V79MZh1A2 cells suggest that biotransformation mediated via CYP1A1 and CYP1A2 might depend on a PAH-type pathway involving the aromatic ring system.

A metabolic activation mechanism of 7H-dibenzo[c,g]carbazole via o-quinone. Part 2: covalent adducts of 7H-dibenzo[c,g]carbazole-3,4-dione with nucleic acid bases and nucleosides

7H-dibenzo[c,g]carbazole (DBC) is a potent multispecies, multisite carcinogen present in the environment. The metabolic activation pathways of DBC are not completely known. It is hypothesized that DBC may be metabolically activated by oxidation to the reactive Michael acceptor o-quinones, which can form stable and depurinating DNA adducts. The synthesis of DBC-3,4-dione has been previously reported by this research group. In the present article, we describe the synthesis and chemical structural elucidation of nine DBC-nucleic acid adducts produced from reactions of DBC-3,4-dione with Ade, Cyt, 2'-deoxyguanosine (dGuo), 2'-deoxycytidine (dCyd), and Guo. Adducts were isolated from reaction mixtures by HPLC and analyzed using MS including elemental compositions and collision-activated dissociation (CAD), (1)H NMR, and two-dimensional chemical shift correlation spectroscopy (COSY) NMR. The adducts, 7-[3,4-dione-DBC-1-yl]-Ade, N4-[3,4-dione-DBC-1-yl]-Cyt, 5-[3,4-dione-DBC-1-yl]-Cyt, two conformational isomers of N2-[3,4-dihydroxy-DBC-1-yl]-dGuo, and two conformational isomers of N2-[3,4-dihydroxy-DBC-1-yl]-Guo, were characterized. Two adducts from reactions of DBC-3,4-dione with dCyd were identified by MS but not fully characterized by NMR due to instability of the adducts. Under similar conditions, the reactions of DBC-3,4-dione with Gua and 2'-deoxyadenosine (dAdo) did not result in an identifiable adduct. Liver DNA adducts from mice treated topically with DBC-3,4-dione (100 microg) in dimethyl sulfoxide/acetone (15/85, 100 microL) were identified with 32P-postlabeling. The major adduct chromatographically matched one of the adducts formed from livers of DBC-treated mouse (adduct 3) using identical conditions.

In vitro toxicity of 7H-dibenzo[c,g]carbazole in human liver cell lines

7H-Dibenzo[c,g]carbazole (DBC) is a model N-heterocyclic aromatic compound (NHA) which is both a hepatotoxin and hepatocarcinogen in rodents. The focus of this investigation was to determine whether human liver cell lines display differential sensitivities to DBC-induced toxicity. Treatment of cell lines with increasing DBC concentrations produced apoptosis only in HepG2 cells. Although DBC inhibited the clonogenic growth of all cell lines at high concentrations, only the survival of HepG2 cells was reduced at lower concentrations. DBC inhibited DNA synthesis in two (HepG2, HLF) of the three cell lines at lower concentrations and was effective only at a high concentration in Mahlavu cells. Differences in DBC uptake were not observed in any of the cell lines, suggesting that bioavailability was not a limiting factor. DBC-DNA adducts were not detected in HLF or Mahlavu cells at either low or high concentrations of DBC. Consistent with the DNA adduct data, RP-HPLC analysis indicated that DBC was metabolized to a lesser degree in the HLF and Mahlavu cells. These results suggest that human liver cell lines differ markedly in the ability to metabolize DBC to toxic species and that DBC-induced apoptosis is only observed in cells that produce detectable metabolites and DBC-DNA adducts.

Role of cytochrome P4501A1 in biotransformation of a tissue specific sarcomagen N-methyldibenzo[c,g]carbazole

7H-dibenzo[c,g]carbazole (DBC) is a potent liver and skin carcinogen, while its synthetic methyl derivative N-methyldibenzo[c,g]carbazole (MeDBC) is tissue specific sarcomagen. It is supposed that sarcomagenic activity of DBC depends on biotransformation at ring-carbon atoms, as with PAH, whereas the heterocyclic nitrogen plays an important role in liver carcinogenicity. The objective of this study was to elucidate the role of cytochrome P4501A1 in metabolic activation of sarcomagenic derivatives of DBC and to characterize the DNA damage profiles induced by DBC and MeDBC in relation to the mode of metabolic activation. The genetically engineered V79MZh1A1 cell line with stable expression of cDNA of human cytochrome P4501A1, the parental V79MZ cell line lacking any cytochrome P450 activity and human hepatocarcinoma Hep G2 cells were used as a model cells. Dose-dependent decrease in colony forming ability (CFA) was found in the V79MZh1A1 cell line after treatment of cells with DBC and MeDBC; however, no change in CFA was induced in parental V79MZ cells. These results were in a good correlation with DNA damaging effects of these two derivatives measured by the alkaline DNA unwinding (ADU) and the modified single cell gel electrophoresis (SCGE) techniques. Differences in DNA damage profiles induced by DBC and MeDBC were found in V79MZh1A1 and Hep G2 cells. These differences were probably the result of different reactive metabolite formation depending on chemical structure of the molecule and ways of biotransformation. This study showed that the cytochrome P4501A1 took part in activation of sarcomagenic DBC derivatives. Moreover, V79 cell lines with stable expression of different cytochromes P450 in combination with DNA repair endonucleases should be a useful tool for characterization of the role of individual cytochromes in metabolic activation pathways of DBC and MeDBC.

Synthesis and structure determination of 6-methylbenzo[a]pyrene-deoxyribonucleoside adducts and their identification and quantitation in vitro and in mouse skin

Activation of the moderate carcinogen 6-methylbenzo[a]pyrene (6-CH(3)BP) by one-electron oxidation to form DNA adducts was studied. Iodine oxidation of 6-CH(3)BP in the presence of dGuo produces BP-6-CH(2)-N(2)dGuo, BP-6-CH(2)-N7Gua and a mixture of 6-CH(3)BP-(1&3)-N7Gua, whereas in the presence of Ade the adducts BP-6-CH(2)-N1Ade, BP-6-CH(2)-N3Ade, BP-6-CH(2)-N7Ade and 6-CH(3)BP-(1&3)-N1Ade are obtained. Furthermore, for the first time an aromatic hydrocarbon radical cation afforded an adduct with dThd, the stable adduct BP-6-CH(2)-N3dThd. Formation of these adducts indicates that the 6-CH(3)BP radical cation has charge localized at the 6, 1 and 3 position. When 6-CH(3)BP was activated by horseradish peroxidase in the presence of DNA, two depurinating adducts were identified, BP-6-CH(2)-N7Gua (48%) and 6-CH(3)BP-(1&3)-N7Gua (23%), with 29% unidentified stable adducts. In the binding of 6-CH(3)BP catalyzed by rat liver microsomes, the same two depurinating adducts, BP-6-CH(2)-N7Gua (22%) and 6-CH(3)BP-(1&3)-N7Gua (10%), were identified, with 68% unidentified stable adducts. In 6-CH(3)BP-treated mouse skin, the two depurinating adducts, BP-6-CH(2)-N7Gua and 6-CH(3)BP-(1&3)-N7Gua, were identified. Although quantitation of these two adducts was not possible due to coelution of metabolites on HPLC, they appeared to be the major adducts found in mouse skin. These results show that 6-CH(3)BP forms depurinating adducts only with the guanine base of DNA, both in vitro and in mouse skin. The weaker reactivity of 6-CH(3)BP radical cation vs. BP radical cation could account for the weaker tumor-initiating activity of 6-CH(3)BP in comparison to that of BP.

A comparison of apoptosis and necrosis induced by hepatotoxins in HepG2 cells

7H-Dibenzo[c,g]carbazole (DBC), an N-heterocyclic aromatic hydrocarbon, is cytotoxic and carcinogenic in rodent liver. While DBC leads to necrotic lesions in the liver, the induction of apoptosis by DBC has not been investigated. The focus of this study was to determine the degree to which apoptosis and necrosis contributed to DBC cytotoxicity in a human hepatoma cell line (HepG2). To determine if these effects were unique to DBC, the results were compared to another hepatotoxin, aflatoxin B(1) (AFB(1)). DBC produced a distinct biphasic LDH release curve within 24 h of exposure. During the same time period lower concentrations of DBC (<10 microM) induced the formation of DBC-DNA adducts and increased p53 protein levels followed by apoptotic cell death. However, increasing the concentration of DBC to 80 microM led to lower DNA adduct and p53 protein levels. At this concentration, intracellular ATP levels were rapidly depleted followed by cell swelling and loss of membrane integrity consistent with necrotic cell death. In contrast to DBC, a biphasic LDH release curve was not observed for AFB(1). Instead, AFB(1) induced a concentration-dependent increase in apoptosis that reached two- to threefold higher levels than DBC. These results suggest that differences exist in the extent and type of cell death induced by DBC and AFB(1) at equimolar concentrations. Apoptosis and necrosis result from low and high concentrations of DBC, respectively, and may be dependent upon intracellular ATP levels.

Ionization potentials and metabolic activations of carbazole and acridine derivatives

7H-Dibenzo[c,g]carbazole (DBC) and dibenz[a,j]acridine (DBA) representing environmental nitrogen-heterocyclic aromatic (NHA) genotoxicants undergo differing metabolism and exhibit differing DNA binding patterns and carcinogenic activities. Two chemical oxidation-related parameters, anodic peak potentials (E(pa)) and maximum absorption energies (E(CT)) of the charge-transfer complexes, were measured for a series of 18 derivatives of carbazole and acridine. On the basis of the E(pa) and E(CT) values and the ionization potential (IP) data of the parent carbazole and acridine that are available in the literature, with linear regression analyses, IP values of the 18 carbazoles and acridines were reported for the first time. The two sets of IP values determined from either E(pa) or E(CT) agreed with one another for most of the compounds. Carbazoles possessed IP values (ranging from 7.2 to 7.6 eV) that are lower than those of acridines (i.e., 7.8-8.1 eV). These data are consistent with the potential activation of carbazole and/or acridine derivatives. For DBC having an IP of approximately 7.3 eV, both one-electron oxidation and monooxygenation pathways are involved in the metabolic activation. In contrast, DBA with a high IP of approximately 8.0 eV is activated through the monooxygenation pathway only. Therefore, just as it is known for carcinogenic PAHs, IP appears to be an important parameter in predicting the metabolic activation for genotoxic NHA in the environment.

Estrogen-nucleic acid adducts: dissection of the reaction of 3, 4-estrone quinone and its radical anion and radical cation with deoxynucleosides and DNA

Previous studies from our laboratory have shown that 3,4-estrone quinone (3,4-EQ) can redox-cycle and is capable of inducing single-strand DNA breaks in MCF-7 breast cancer cells, as well as reacting with various deoxynucleosides to give several estrogen-nucleic acid adducts. While reactions of 3,4-EQ with all the deoxynucleosides under acidic conditions gave only the N7-Gua adduct, which could proceed by Michael addition, reactions of 3,4-EQ under reductive conditions gave several adducts, including the N7-Gua, C8-Ade, C8-Gua, N3-Thy, and N4-Cyt adducts, suggesting the involvement of a 3,4-EQ radical species. The question as to which of the reactive species, the estrogen quinone or the estrogen semiquinone, that is responsible for estrogen's genotoxic activity has been the subject of recent investigations in several laboratories. To explore this in more detail, we carried out studies on the reactivity of 3,4-EQ, the 3,4-EQ radical anion, and the 3, 4-EQ radical cation with both deoxynucleosides and calf thymus DNA under different pH conditions. Both stable and unstable adducts with guanine and thymine were observed from reactions with DNA. Although adduct levels were somewhat different, the adduct profiles obtained from reactions of 3,4-EQ and its radical anion with both DNA and deoxynucleosides were quite similar and were found to be significantly different from product profiles obtained from reactions with the 3,4-EQ radical cation. Studies conducted with the human breast tumor cell line MCF-7 demonstrated the formation of the N7- and C8-Gua adducts in which the profiles were similar to those obtained from reactions of 3,4-EQ with DNA. These results suggest that the reactive species that is responsible for adduct formation under physiological conditions is most likely to be the 3,4-EQ radical anion.

Synthesis and structure determination of the adducts formed by electrochemical oxidation of Dibenzo[a,l]pyrene in the presence of adenine

Because the radical cations of polycyclic aromatic hydrocarbons (PAH) are involved in tumor initiation, determination of the structures of biologically formed PAH-DNA adducts is important and relies on comparison of their properties with those of synthesized adducts. One of the possible sites of adduct formation is the N-3 position of Ade, but this depurinating adduct is not obtained by one-electron oxidation of dibenzo[a,l]pyrene (DB[a,l]P) in the presence of deoxyadenosine. Therefore, we turned to electrochemical oxidation of DB[a,l]P in the presence of Ade in dimethylformamide and produced the following adducts: DB[a,l]P-10-N1Ade (47%), DB[a, l]P-10-N3Ade (5%), DB[a,l]P-10-N7Ade (2%), and DB[a,l]P-10-N(6)Ade (6%). In Me(2)SO, this reaction afforded the same four adducts, but in slightly different yields: DB[a,l]P-10-N1Ade (44%), DB[a, l]P-10-N3Ade (9%), DB[a,l]P-10-N7Ade (1%), and DB[a,l]P-10-N(6)Ade (3%). These adducts were purified by reverse-phase HPLC, and the subtle differences between the isomers were revealed by NMR, tandem mass spectrometry, and fluorescence line-narrowing spectroscopy. The relative yields of the N1Ade, N3Ade, and N7Ade adducts reflect the nucleophilicity and steric accessibility of these three nitrogen atoms in Ade.

An unusual DNA adduct derived from the powerfully mutagenic environmental contaminant 3-nitrobenzanthrone

The covalent binding of an N-hydroxy metabolite of the powerfully mutagenic 3-nitrobenzanthrone (NBA) to 2'-deoxyguanosine (dG) and calf thymus DNA has been investigated in vitro. The major adduct obtained from the reaction of the N-acetoxy-N-acetyl derivative (N-Aco-N-Ac-ABA) of 3-aminobenzanthrone (ABA) and dG was identified as N-acetyl-3-amino-2-(2'-deoxyguanosin-8-yl)benzanthrone (dG-N-Ac-ABA) by 1H NMR and mass spectroscopies as well as by the reaction of N-Aco-N-Ac-ABA with the double-stranded calf thymus DNA. The coupling with the dG moiety occurred exclusively at C-2 of benzanthrone (BA), suggesting a significant contribution of a resonance-stabilized arenium ion intermediate derived from BA to the production of this new type of adduct. The preferred conformation of the adduct has been shown to be syn by 1H and 13C NMR.

Synthesis of adducts formed by iodine oxidation of aromatic hydrocarbons in the presence of deoxyribonucleosides and nucleobases

Polycyclic aromatic hydrocarbons (PAH) undergo two main pathways of metabolic activation related to the initiation of tumors: one-electron oxidation to give radical cations and monooxygenation to yield bay-region diol epoxides. Synthesis of standard adducts is essential for identifying biologically formed adducts. Until recently, radical cation adducts were synthesized by oxidation of the PAH in an electrochemical apparatus, not readily available in many organic chemistry laboratories. We have developed a convenient and efficient method for synthesizing PAH-nucleoside adducts by using I2 as the oxidant. Adducts of benzo[a]pyrene (BP), dibenzo[a, l]pyrene (DB[a,l]P), and 7,12-dimethylbenz[a]anthracene were synthesized with deoxyguanosine (dG), deoxyadenosine, guanine (Gua), or adenine in either Me2SO or dimethylformamide (DMF) with or without AgClO4. When, for example, the potent carcinogen BP was dissolved in DMF in the presence of 3 equiv of I2, 5 equiv of dG, and 1 equiv of AgClO4, 45% of the BP was converted to BP-6-N7Gua. When BP was placed under the same reaction conditions in the absence of AgClO4, the extent of formation of BP-6-N7Gua decreased to 30%. When the potent carcinogen DB[a,l]P was dissolved in DMF in the presence of 3 equiv of I2, 5 equiv of dG, and 1 equiv of AgClO4, 43% of the DB[a,l]P was converted to DB[a,l]P-10-N7Gua. In the more polar solvent Me2SO under the same reaction conditions, however, the yield of DB[a,l]P-10-N7Gua was only 20%. Synthesis of adducts with the oxidant I2 is more convenient and, in some cases, more efficient than synthesis by electrochemical oxidation. This method simplifies the synthesis of PAH-nucleoside and nucleobase adducts that are essential for studying biologically formed PAH-DNA adducts.