Search

Acrylamide is formed during the heating of food and is also found in cigarette smoke. It is classified by the International Agency for Research on Cancer as a probable human carcinogen (Group 2A). Glycidamide, an epoxide metabolite of acrylamide, is implicated in the mechanism of acrylamide carcinogenicity. Acrylamide causes oxidative DNA damage in target organs. We sought to clarify the mechanism of acrylamide-induced oxidative DNA damage by investigating site-specific DNA damage and reactive oxygen species (ROS) generation by a putative metabolite of acrylamide, acrylohydroxamic acid (AA). Our results, using 32P-5’-end-labeled DNA fragments, indicated that, although AA alone did not damage DNA, AA treated with amidase induced DNA damage in the presence of Cu(II). DNA cleavage occurred preferentially at T and C, and particularly at T in 5’-TG-3’ sequences, and the DNA cleavage pattern was similar to that of hydroxylamine. The DNA damage was inhibited by methional, catalase, and Cu(I)-chelator bathocuproine, suggesting that H2O2 and Cu(I) are involved in the mechanism of DNA damage induced by AA treated with amidase. In addition, amidase-treated AA increased 8-oxo-7,8-dihydro-2’-deoxyguanosine formation in calf thymus DNA, an indicator of oxidative DNA damage, in a dose-dependent manner. In conclusion, hydroxylamine, possibly produced from AA treated with amidase, was autoxidized via the Cu(II)/Cu(I) redox cycle and H2O2 generation, suggesting that oxidative DNA damage induced by ROS plays an important role in acrylamide-related carcinogenesis.

Naphthalene is a carcinogenic polycyclic aromatic hydrocarbon, to which humans are exposed as an air pollutant. Naphthalene is metabolized in humans to reactive intermediates such as 1,2-hydroxynaphthalene (1,2-NQH2), 1,4-NQH2, 1,2-naphthoquinone (1,2-NQ), and 1,4-NQ. We examined oxidative DNA damage by these naphthalene metabolites using 32P-labeled DNA fragments from human cancer-relevant genes. 1,2-NQH2 and 1,4-NQH2 induced DNA damage in the presence of Cu(II). The DNA-damaging activity of 1,2-NQH2 was significantly increased in the presence of the reduced form of nicotinamide adenine dinucleotide (NADH), whereas that of 1,4-NQH2 was not. In the presence of NADH, 1,2-NQ induced Cu(II)-dependent DNA damage, whereas 1,4-NQ did not. The calculated energy of the lowest unoccupied molecular orbital (LUMO), which corresponds to the reduction potential, was estimated to be −0.67 eV for 1,2-NQ and −0.75 eV for 1,4-NQ. These results suggest that 1,2-NQ was reduced more easily than 1,4-NQ. Furthermore, 1,2-NQH2, 1,4-NQH2, and 1,2-NQ plus NADH formed 8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodG) as an oxidative DNA marker. Catalase and bathocuproine inhibited DNA damage, suggesting that H2O2 and Cu(I) were involved. These results indicate that NQH2s are oxidized to the corresponding NQs via semiquinone radicals, and that H2O2 and Cu(I) are generated during oxidation. 1,2-NQ is reduced by NADH to form the redox cycle, resulting in enhanced DNA damage. The formation of the corresponding semiquinone radicals was supported by an electron paramagnetic resonance (EPR) study. In conclusion, the redox cycle of 1,2-NQ/1,2-NQH2 may play a more important role in the carcinogenicity of naphthalene than that of 1,4-NQ/1,4-NQH2.

Morin is a potential inhibitor of amyloid β-peptide aggregation. This aggregation is involved in the pathogenesis of Alzheimer's disease. Meanwhile, morin has been found to be mutagenic and exhibits peroxidation of membrane lipids concurrent with DNA strand breaks in the presence of metal ions. To clarify a molecular mechanism of morin-induced DNA damage, we examined the DNA damage and its site specificity on 32P-5'-end-labeled human DNA fragments treated with morin plus Cu(II). The formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), an indicator of oxidative DNA damage, was also determined in calf thymus DNA treated with morin plus Cu(II). Morin-induced DNA strand breaks and base modification in the presence of Cu(II) were dose dependent. Morin plus Cu(II) caused piperidine-labile lesions preferentially at thymine and guanine residues. The DNA damage was inhibited by methional, catalase and Cu(I)-chelator bathocuproine. The typical •OH scavengers ethanol, mannitol and sodium formate showed no inhibitory effect on DNA damage induced by morin plus Cu(II). When superoxide dismutase was added to the solution, DNA damage was not inhibited. In addition, morin plus Cu(II) increased 8-oxodG formation in calf thymus DNA fragments. We conclude that morin undergoes autoxidation in the presence of Cu(II) via a Cu(I)/Cu(II) redox cycle and H2O2 generation to produce Cu(I)-hydroperoxide, which causes oxidative DNA damage.

Modification of DNA is believed to be a key step in mutagenesis and carcinogenesis. Reactive oxygen species (ROS) generated by environmental factors can cause oxidative DNA damage. This review focused on the role of oxidative DNA damage in mutagenesis and carcinogenesis mediated by environmental factors. This research investigated the mechanism of DNA damage induced by environmental chemicals and dietary factors using 32P-labeled DNA fragments obtained from the c-Ha-RAS-1 protooncogene and the p16 and p53 tumor suppressor genes. In addition, the content of 8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodG) was measured by using a high performance liquid chromatograph equipped with an electrochemical detector. 8-oxodG is probably one of the most abundant DNA lesion formed during oxidative stress. Antioxidants are considered as the most promising chemopreventive agents against various human cancers. However, some antioxidants play paradoxical roles acting as “double-edged sword”. The present research also investigated the mechanism of DNA damage induced by antioxidants using human cultured cell lines and 32P-labeled DNA fragments. This review shows recent experimental results and discusses the mechanisms of oxidative DNA damage in relation to carcinogenesis.

6-Hydroxydopamine (6-OHDA) is a neurotoxin to produce an animal model of Parkinson's disease. 6-OHDA increased the formation of 8-oxo-7, 8-dihydro-2'-deoxyguanosine (8-oxodG), a biomarker of oxidatively damaged DNA, and induced apoptosis in human neuroblastoma SH-SY5Y cells. Iron or copper chelators inhibited 6-OHDA-induced 8-oxodG formation and apoptosis. Thus, iron and copper are involved in the intracellular oxidatively generated damage to DNA, a stimulus for initiating apoptosis. This study examined DNA damage caused by 6-OHDA plus metal ions using (32)P-5'-end-labelled DNA fragments. 6-OHDA increased levels of oxidatively damaged DNA in the presence of Fe(III)EDTA or Cu(II). Cu(II)-mediated DNA damage was stronger than Fe(III)-mediated DNA damage. The spectrophotometric detection of p-quinone and the scopoletin method showed that Cu(II) more effectively accelerated the 6-OHDA auto-oxidation and H(2)O(2) generation than Fe(III)EDTA. This study suggests that copper, as well as iron, may play an important role in 6-OHDA-induced neuronal cell death.

Chloramphenicol (CAP) was an old antimicrobial agent. However, the use of CAP is limited because of its harmful side effects, such as leukemia. The molecular mechanism through which CAP has been strongly correlated with leukemogenesis is still unclear. To elucidate the mechanism of genotoxicity, we examined DNA damage by CAP and its metabolites, nitroso-CAP (CAP-NO), N-hydroxy-CAP (CAP-NHOH), using isolated DNA. CAP-NHOH have the ability of DNA damage including 8-oxo-7,8-dihydro-2'-deoxyguanosine formation in the presence of Cu(II), which was greatly enhanced by the addition of an endogenous reductant NADH. CAP-NO caused DNA damage in the presence of Cu(II), only when reduced by NADH. NADH can non-enzymatically reduce the nitroso form to hydronitroxide radicals, resulting in enhanced generation of reactive oxygen species followed by DNA damage through the redox cycle. Furthermore, we also studied the site specificity of base lesions in DNA treated with piperidine or formamidopyrimidine-DNA glycosylase, using (32)P-5'-end-labeled DNA fragments obtained from the human tumor suppressor gene. CAP metabolites preferentially caused double base lesion, the G and C of the ACG sequence complementary to codon 273 of the p53 gene, in the presence of NADH and Cu(II). Therefore, we conclude that oxidative double base lesion may play a role in carcinogenicity of CAP.

4-Hydrazinobenzoic acid, an ingredient of mushroom Agaricus bisporus, is carcinogenic to rodents. To clarify the mechanism of carcinogenesis, we investigated DNA damage by 4-hydrazinobenzoic acid using (32)P-labeled DNA fragments obtained from the human p53 and p16 tumor suppressor genes. 4-Hydrazinobenzoic acid induced Cu(II)-dependent DNA damage especially piperidine-labile formation at thymine and cytosine residues. Typical hydroxyl radical scavengers showed no inhibitory effects on Cu(II)-mediated DNA damage by 4-hydrazinobenzoic acid. Bathocuproine and catalase inhibited the DNA damage, indicating the participation of Cu(I) and H(2)O(2) in the DNA damage. These findings suggest that H(2)O(2) generated by the autoxidation of 4-hydrazinobenzoic acid reacts with Cu(I) to form reactive oxygen species, capable of causing DNA damage. Interestingly, catalase did not completely inhibit DNA damage caused by a high concentration of 4-hydrazinobenzoic acid (over 50 microM) in the presence of Cu(II). 4-Hydrazinobenzoic acid induced piperidine-labile sites frequently at adenine and guanine residues in the presence of catalase. 4-Hydrazinobenzoic acid increased formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), a characteristic oxidative DNA lesion, in calf thymus DNA, whereas 4-hydrazinobenzoic acid did not increase the formation of 8-oxodG in the presence of catalase. ESR spin-trapping experiments showed that the phenyl radical was formed during the reaction of 4-hydrazinobenzoic acid in the presence of Cu(II) and catalase. Matrix-assisted laser desorption/ionization time-of-flight mass (MALDI-TOF/mass) spectrometry analysis showed that phenyl radical formed adduct with adenosine and guanosine. These results suggested that 4-hydrazinobenzoic acid induced DNA damage via not only H(2)O(2) production but also phenyl radical production. This study suggests that both oxidative DNA damage and DNA adduct formation play important roles in the expression of carcinogenesis of 4-hydrazinobenzoic acid.

Antioxidants are considered as the most promising chemopreventive agents against various human cancers. However, some antioxidants play paradoxical roles, acting as "double-edged sword." A primary property of effective and acceptable chemopreventive agents should be freedom from toxic effects in healthy population. Miscarriage of the intervention by beta-carotene made us realize the necessity for evaluation of safety before recommending use of antioxidant supplements for chemoprevention. We have evaluated the safety of antioxidants on the basis of reactivity with DNA. Our results revealed that phytic acid, luteolin, and retinoic acid did not cause DNA damage under the experimental condition. Furthermore, phytic acid inhibited the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine, an indicator of oxidative DNA damage, in cultured cells treated with a H(2)O(2)-generating system. Thus, it is expected that these chemopreventive agents can safely protect humans against cancer. On the other hand, some chemopreventive agents with prooxidant properties (alpha-tocopherol, quercetin, catechins, isothiocyanates, N-acetylcysteine) caused DNA damage via generation of reactive oxygen species in the presence of metal ions and endogenous reductants under some circumstances. Furthermore, other chemopreventive agents (beta-carotene, genistein, daidzein, propyl gallate, curcumin) exerted prooxidant properties after metabolic activation. Therefore, further studies on safety should be required when antioxidants are used for cancer prevention.

Eugenol used as a flavor has potential carcinogenicity. DNA adduct formation via 2,3-epoxidation pathway has been thought to be a major mechanism of DNA damage by carcinogenic allylbenzene analogs including eugenol. We examined whether eugenol can induce oxidative DNA damage in the presence of cytochrome P450 using [ 32 P]-5′-end-labeled DNA fragments obtained from human genes relevant to cancer. Eugenol induced Cu(II)-mediated DNA damage in the presence of cytochrome P450 (CYP)1A1, 1A2, 2C9, 2D6, or 2E1. CYP2D6 mediated eugenol-dependent DNA damage most efficiently. Piperidine and formamidopyrimidine-DNA glycosylase treatment induced cleavage sites mainly at T and G residues of the 5′-TG-3′ sequence, respectively. Interestingly, CYP2D6-treated eugenol strongly damaged C and G of the 5′-ACG-3′ sequence complementary to codon 273 of the p53 gene. These results suggest that CYP2D6-treated eugenol can cause double base lesions. DNA damage was inhibited by both catalase and bathocuproine, suggesting that H 2 O 2 and Cu(I) are involved. These results suggest that Cu(I)–hydroperoxo complex is primary reactive species causing DNA damage. Formation of 8-oxo-7,8-dihydro-2′-deoxyguanosine was significantly increased by CYP2D6-treated eugenol in the presence of Cu(II). Time-of-flight-mass spectrometry demonstrated that CYP2D6 catalyzed O -demethylation of eugenol to produce hydroxychavicol, capable of causing DNA damage. Therefore, it is concluded that eugenol may express carcinogenicity through oxidative DNA damage by its metabolite.

Interstrand DNA cross-linking has been considered to be the primary action mechanism of cyclophosphamide (CP) and its hydroperoxide derivative, 4-hydroperoxycyclophosphamide (4-HC). To clarify the mechanism of anti-tumor effects by 4-HC, we investigated DNA damage in a human leukemia cell line, HL-60, and its H(2)O(2)-resistant clone HP100. Apoptosis DNA ladder formation was detected in HL-60 cells treated with 4-HC, whereas it was not observed in HP100 cells. 4-HC significantly increased 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG) formation, a marker of oxidative DNA damage, in HL-60 cells. On the other hand, CP did not significantly induce 8-oxodG formation and apoptosis in HL-60 cells under the same conditions as did 4-HC. Using (32)P-labeled DNA fragments from the human p53 tumor suppressor gene, 4-HC was found to cause Cu(II)-mediated oxidative DNA damage, but CP did not. Catalase inhibited 4-HC-induced DNA damage, including 8-oxodG formation, suggesting the involvement of H(2)O(2). The generation of H(2)O(2) during 4-HC degradation was ascertained by procedures using scopoletin and potassium iodide. We conclude that, in addition to DNA cross-linking, oxidative DNA damage through H(2)O(2) generation may participate in the anti-tumor effects of 4-HC.

In this study, we compared the DNA-damaging ability of bifunctional busulfan and monofunctional ethyl methanesulfonate (EMS, CH3SO2OC2H5) and their cytotoxicity towards human promyelocytic leukemia HL-60 cells. We also investigated the site specificity of busulfan-induced DNA damage using 32 P-5′end-labeled DNA fragments obtained from the human p16 tumor suppressor gene. Furthermore, time-of-flight mass spectrometry (TOF-MS) was performed to identify the DNA crosslinks formed by busulfan. Materials and Methods Materials. Restriction enzymes (EcoRI and BssHII) and proteinase K were from Boehringer Mannheim GmbH (Mannheim, Germany). T4 polynucleotide kinase was obtained from New England Biolabs (Beverly, MA). [γ- 32 P]ATP (222 TBq/mmol) was from New England Nuclear (Boston, MA). Diethylenetriamine-N,N,N′,N′′,N′′-pentaacetic acid (DTPA) was from Dojin Chemical Co. (Kumamoto). Busulfan, calf thymus DNA, 3-hydroxypicolinic acid and citric acid were from Sigma Chemical Co. (St. Louis, MO). Acrylamide, dimethylsulfoxide, bisacrylamide and piperidine were from Wako Pure Chemical Industries (Osaka). EMS and ethanol were from Nacalai Tesque, Inc. (Kyoto). Mouse 3-methyladenine DNA glycosylase (Aag) was obtained from Trevigen Inc. (Gaithersburg, MD). Growth-inhibitory and cytotoxic effects of busulfan and EMS on cultured cells. HL-60 cells (2.5×10 5 cells/ml) were incubated with 100 or 200 µM busulfan or 200 µM EMS in 2 ml of RPMI 1640 (Gibco Laboratories, NY) supplemented with 6% fetal calf serum (FCS, Whittaker Bioproducts) for 24 h at 37°C. Cell viability was determined by trypan blue exclusion and counting in a hemocytometer. For statistical analysis of the data, an independent t test was used, and the criterion of significance was set at P

Green tea catechins, especially (-)-epigallocatechin gallate (EGCG), are believed to mediate much of the cancer chemopreventive effects of tea. However, it was reported that green tea catechins enhanced colon carcinogenesis in rats. Experiments using 32P-labeled DNA fragments obtained from human cancer-related genes showed that catechins induced DNA damage in the presence of metals such as Cu(II) and Fe(III) complexes. In the presence of Fe(III)EDTA, the order of DNA damaging ability was EGCG approximately (-)-epigallocatechin>(-)-epicatechin gallate>>catechin. Catechins plus Fe(III)EDTA caused DNA damage at every nucleotide, most likely due to *OH generation from H(2)O(2). In the presence of Cu(II), the order was (-)-epigallocatechin>catechin>EGCG>(-)-epicatechin gallate. Cu(II)-mediated DNA damage by EGCG occurred most frequently at T and G residues, especially of 5'-TG-3' and GG sequences. Catalase and bathocuproine inhibited the Cu(II)-mediated DNA damage, suggesting the involvement of H(2)O(2) and Cu(I). In the presence of metal ions, increased amounts of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG) were found in DNA treated with EGCG. Furthermore, EGCG increased amounts of 8-oxodG in HL-60 cells, but not in the H(2)O(2)-resistant clone HP100. When GSH was reduced by L-buthionine-[S, R]-sulfoximine, a low concentration of EGCG increased amounts of 8-oxodG in HL-60 cells, further supporting the involvement of H(2)O(2) in cellular DNA damage. It is concluded that EGCG can induce H(2)O(2) generation and subsequent damage to isolated and cellular DNA, and that oxidative DNA damage may mediate the potential carcinogenicity of EGCG.

3′-Azido-3′-deoxythymidine (AZT) is carcinogenic to experimental animals and can cause the formation of 8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodG) in humans and animals. To clarify the mechanism of carcinogenesis by AZT, we investigated DNA damage induced by its photodegradation products, using 32P-5′-end-labeled DNA fragments obtained from human genes. Following exposure to UVB, AZT induced DNA damage in the presence of Cu(II). Catalase inhibited DNA damage, indicating the involvement of H2O2. UVB-exposed AZT plus Cu(II) induced 8-oxodG formation in a dose-dependent manner. Mass spectrum of UVB-exposed AZT demonstrated the generation of a hydroxylamine derivative. The colorimetric determination suggested that AZT was converted into the hydroxylamine derivative depending on UVB doses. UVB-exposed AZT induced double base damage at the 5′-A CG -3′ sequence, complementary to a hot spot of the p53 gene. The basic compound, hydroxylamine, showed similar site specificity. The hydroxylamine derivative produced by photodegradation and/or possible metabolism of AZT induces oxidative DNA damage, which may participate in carcinogenesis.

Carcinogenic urethane (ethyl carbamate) forms DNA adduct via epoxide, whereas carcinogenic methyl carbamate can not. To clarify a mechanism independent of DNA adduct formation, we examined DNA damage induced by N-hydroxyurethane, a urethane metabolite, using 32P-5'-end-labeled DNA fragments. N-hydroxyurethane induced Cu(II)-mediated DNA damage especially at thymine and cytosine residues. DNA damage was inhibited by both catalase and bathocuproine, suggesting a role for H(2)O(2) and Cu(I) in DNA damage. Free (*) OH scavengers did not inhibit the DNA damage, although methional did inhibit it. These results suggest that reactive species, such as the Cu(I)-hydroperoxo complex, cause DNA damage. Formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG) was increased by N-hydroxyurethane in the presence of Cu(II). When treated with esterase, N-hydroxyurethane induced 8-oxodG formation to a similar extent as that induced by hydroxylamine. Enhancement of DNA cleavages by endonuclease IV suggests that hydroxylamine induced depurination. Furthermore, hydroxylamine induced a significant increase in 8-oxodG formation in HL-60 cells but not in its H(2)O(2)-resistant clone HP 100 cells. o-Phenanthroline significantly inhibited the 8-oxodG formation in HL-60 cells, confirming the involvement of metal ions in the 8-oxodG formation by hydroxylamine. Electron spin resonance spectroscopy, utilizing Fe[N-(dithiocarboxy)sarcosine](3), demonstrated that nitric oxide (NO) was generated from hydroxylamine and esterase-treated N-hydroxyurethane. It is concluded that urethane may induce carcinogenesis through oxidation and, to a lesser extent, depurination of DNA by its metabolites.

The mechanism of DNA damage photosensitized by Nile blue (NB) was studied using (32)P-5'-end-labeled DNA fragments. NB bound to the DNA strand was possibly intercalated through an electrostatic interaction. Photoirradiated NB caused DNA cleavage at guanine residues when the DNA fragments were treated with piperidine. Consecutive guanines, the underlined G in 5'-GG and 5'-GGG, were selectively damaged through photoinduced electron transfer. The fluorescence lifetime of NB was decreased by guanine-containing DNA sequence, supporting this mechanism. Single guanines were also slightly damaged by photoexcited NB, and DNA photodamage by NB was slightly enhanced in D2O. These results suggest that the singlet oxygen mechanism also partly contributes to DNA photodamage by NB. DNA damage photosensitized by NB via electron transfer may be an important mechanism in medicinal applications of photosensitizers, such as photodynamic therapy in low oxygen.

We examined the redox properties of the "carcinogenic" catechol and the "noncarcinogenic" hydroquinone in relation to different DNA damaging activities and carcinogenicity using 32P-labeled DNA fragments obtained from the human genes. In the presence of endogenous NADH and Cu2+, catechol induces stronger DNA damage than hydroquinone, although the magnitudes of their DNA damaging activities were reversed in the absence of NADH. In both cases, DNA damage resulted from base modification at guanine and thymine residues in addition to strand breakage induced by Cu+ and H2O2, generated during the oxidation of catechol and hydroquinone into 1,2-benzoquinone and 1,4-benzoquinone, respectively. EPR and 1H NMR studies indicated that 1,2-benzoquinone is converted directly into catechol through a nonenzymatic two-electron reduction by NADH whereas 1,4-benzoquinone is reduced into hydroquinone through a semiquinone radical intermediate through two cycles of one-electron reduction. The reduction of 1,2-benzoquinone by NADH proceeds more rapidly than that of 1,4-benzoquinone. This study demonstrates that the rapid 1,2-benzoquinone two-electron reduction accelerates the redox reaction turnover between catechol and 1,2-benzoquinone, resulting in the enhancement of DNA damage. These results suggest that the differences in NADH-mediated redox properties of catechol and hydroquinone contribute to their different carcinogenicities.

Reactive species generated by chemicals and UV radiation can cause sequence-specific DNA damage and play important roles in mutagenesis, carcinogenesis and aging. We have investigated sequence specificity of oxidative stress-mediated DNA damage by using 32P-labeled DNA fragments obtained from the human c-Ha-ras-1 and p53 genes. Free hydroxyl radical causes DNA damage with no marked site specificity. Reactive nitrogen species, sulfate radicals, nitrogen-centered radicals, benzoyloxyl radical and alkoxyl radical show different sequence specificity. Benzoyloxyl radical specifically causes damage to the 5'-G in GG sequence. UVA radiation also causes DNA damage at this site through electron transfer in the presence of certain photosensitizers. The 5'-G in GG sequence is easily oxidized because a large part of the highest occupied molecular orbital is distributed on this site. On the basis of these findings, the sequence specificity of DNA damage is presumably determined by (a) redox potential of reactive species; (b) ionization potential of DNA bases; and (c) site-specific binding of metal ion to DNA. Here we discuss the mechanisms of sequence-specific DNA damage in relation to carcinogenesis and aging.

We investigated the amplification of bleomycin-induced DNA cleavage by synthetic triamides containing N-methylpyrrole (Py) and/or N-methylimidazole (Im), PyPyPy, PyPyIm, PyImPy, and PyImIm, using 32P-labeled DNA fragments obtained from the human c-Ha-ras-1 and p53 genes. Peplomycin, a bleomycin analog, plus Fe(II) caused DNA cleavage at the 5'-GC-3' and 5'-GT-3' sequences (damaged bases are underlined). The addition of PyPyPy dramatically enhanced the cleavage, particularly at cytosine residues 3' to consecutive guanines. Alteration in the site specificity was not observed with other triamides (PyPyIm, PyImPy, and PyImIm). DNase I footprinting revealed that PyPyPy bound to the sites adjacent to the sites where DNA cleavage was enhanced by PyPyPy, and that PyPyPy enhanced DNase I-induced cleavage in GC-rich regions. These findings suggest that binding of PyPyPy to the DNA minor groove changes the DNA conformation to allow peplomycin to cleave DNA more efficiently at GC-rich sequences, resulting in intensive site-specific DNA cleavage particularly at cytosines at the 3'-side of polyguanines. The present study on amplifiers of antitumor drugs would appear to offer a novel approach to the establishment of more effective chemotherapy.

Hydroxyurea is a chemotherapeutic agent used for the treatment of myeloproliferative disorders (MPD) and solid tumors. The mutagenic and carcinogenic potential of hydroxyurea has not been established, although hydroxyurea has been associated with an increased risk of leukemia in MPD patients. To clarify whether hydroxyurea has potential carcinogenicity, we examined site-specific DNA damage induced by hydroxyurea using (32)P-5'-end-labeled DNA fragments obtained from the human p53 and p16 tumor suppressor genes and the c-Ha-ras-1 protooncogene. Hydroxyurea caused Cu(II)-mediated DNA damage especially at thymine and cytosine residues. NADH efficiently enhanced hydroxyurea-induced DNA damage. The DNA damage was almost entirely inhibited by catalase and bathocuproine, a Cu(I)-specific chelator, suggesting the involvement of hydrogen peroxide (H(2)O(2)) and Cu(I). Typical free hydroxyl radical scavengers did not inhibit DNA damage by hydroxyurea, but methional did. These results suggest that crypto-hydroxyl radicals such as Cu(I)-hydroperoxo complex (Cu(I)-OOH) cause DNA damage. Formation of 8-hydroxy-2'-deoxyguanosine (8-OHdG) was induced by hydroxyurea in the presence of Cu(II). An electron spin resonance spectroscopic study using N-(dithiocarboxy)sarcosine as a nitric oxide (NO)-trapping reagent demonstrated that NO was generated from hydroxyurea in the presence and absence of catalase. In addition, the generation of formamide was detected by both gas chromatography-mass spectrometry (GC-MS) and time-of-flight-mass spectrometry (TOF-MS). A high concentration of hydroxyurea induced depurination at DNA bases in an H(2)O(2)-independent manner, and endonuclease IV treatment led to chain cleavages. These results suggest that hydroxyurea could induce base oxidation as the major pathway of DNA modification and depurination as a minor pathway. Therefore, it is considered that DNA damage by hydroxyurea participates in not only anti-cancer activity, but also carcinogenesis.

An important approach to improve chemotherapy of members of the bleomycin (BLM) family of antibiotics is to find compounds (amplifiers) that enhance the activity of BLM-mediated DNA cleavage and apoptosis. Using a DNA-sequencing technique and pulsed field gel electrophoresis, we have investigated whether BLM-mediated cleavage of isolated and cellular DNA is amplifed by three compounds (RW-12, LS-20, 1S-5Me) which have a conformationally flexible, unfused polyaromatic system and cationic side chain in the molecules. RW-12 enhanced most effectively both pepleomycin (PEM)-induced cytotoxicity and apoptosis. The order of the maximum enhancing effect of amplifiers on PEM-mediated DNA damage is RW-12 > LS-20 > 1S-5Me. RW-12 amplified PEM-mediated DNA cleavage most effectively not only in vitro but also in cultured cells. We have reported that the order of the DNA binding constants of these compounds is RW-12 > LS-20 > 1S-5Me. In this study, we found a good correlation between PEM-mediated cleavage of isolated DNA and cellular DNA. These results suggest that BLM amplifiers bind to DNA and by doing so enhance drug-mediated DNA degradation, ultimately leading to apoptosis. The present study on amplifiers of anticancer agents shows a novel approach to the potentially effective anticancer therapy.

Nitroxyl anion (NO−), the one-electron reduction product of nitric oxide (NO⋅), is formed under various physiological conditions. We have used four different assays (DNA strand breakage, 8-oxo-deoxyguanosine formation in calf thymus DNA, malondialdehyde generation from 2′-deoxyribose, and analysis of site-specific DNA damage using32P-5′-end-labeled DNA fragments of the human p53tumor suppressor gene and the c-Ha-ras-1 protooncogene) to study the effects of NO− generated from Angeli's salt on DNA damage. It was found that strong oxidants are generated from NO−, especially in the presence of H2O2 plus Fe(III)-EDTA or Cu(II). NO⋅released from diethylamine-NONOate had no such effect. Distinct effects of hydroxyl radical (HO⋅) scavengers and patterns of site-specific DNA cleavage caused by Angeli's salt alone or by Angeli's salt, H2O2 plus metal ion suggest that NO− acts as a reductant to catalyze the formation of the HO⋅ from H2O2 plus Fe(III) and formation of Cu(I)-peroxide complexes with a reactivity similar to HO⋅ from H2O2 and Cu(II). Angeli's salt and H2O2 exerted synergistically cytotoxic effects to MCF-7 cells, determined by lactate dehydrogenase release assay. Thus NO− may play an important role in the etiology of various pathophysiological conditions such as inflammation and neurodegenerative diseases, especially when H2O2 and transition metallic ions are present.

Pulsed field gel electrophoresis showed that the initiation time of DNA breakage induced by the DNA alkylating agent duocarmycin A, which is not a redox-cycling agent, was almost the same in the human leukemia cell line HL-60 and its H2O2-resistant clone HP100. Catalase activity of HP100 cells was much higher than that of HL-60 cells. Duocarmycin A-mediated DNA ladder formation in HP100 cells was delayed compared with that in HL-60 cells, suggesting the involvement of H2O2 in duocarmycin A-induced apoptosis. Flow cytometry demonstrated that peroxide formation preceded loss of mitochondrial membrane potential (delta psi m) in cells treated with duocarmycin A. Then, caspase-3 was activated, followed by DNA ladder formation. These findings suggest that DNA damage by duocarmycin A induces H2O2 generation, which causes delta psi m loss and subsequently caspase-3 activation, resulting in apoptosis.

DNA damage by metabolites of a food additive, butylated hydroxytoluene (BHT), was investigated as a potential mechanism of carcinogenicity. The mechanism of DNA damage by 2,6-di-tert-butyl-p-benzoquinone (BHT-quinone), 2,6-di-tert-butyl-4-hydroperoxyl-4-methyl-2,5-cyclohexadienone (BHT-OOH), and 3,5-di-tert-butyl-4-hydroxybenzaldehyde (BHT-CHO) in the presence of metal ions was investigated by using 32P-labeled DNA fragments obtained from the c-Ha-ras-1 proto-oncogene and the p53 tumor suppressor gene. BHT-OOH caused DNA damage in the presence of Cu(II), whereas BHT-quinone and BHT-CHO did not. However, BHT-quinone did induce DNA damage in the presence of NADH and Cu(II). Bathocuproine inhibited Cu(II)-mediated DNA damage, indicating the participation of Cu(I) in the process. Catalase also inhibited DNA damage induced by BHT-quinone, but not that induced by BHT-OOH. The DNA cleavage pattern observed with BHT-quinone plus NADH was different from that seen with BHT-OOH. With BHT-quinone plus NADH, piperidine-labile sites could be generated at nucleotides other than adenine residue. BHT-OOH caused cleavage specifically at guanine residues. Pulsed field gel electrophoresis showed that BHT-OOH and BHT-quinone induced DNA strand breaks in cultured cells, whereas BHT-CHO did not. Both BHT-quinone and BHT-OOH induced internucleosomal DNA fragmentation, which is the characteristic of apoptosis. Furthermore, flow cytometry analysis revealed an increase of peroxides in cultured cells treated with BHT-OOH or BHT-quinone. These results suggest that BHT-OOH participates in oxidative DNA damage directly, whereas BHT-quinone causes DNA damage through H2O2 generation, which leads to internucleosomal DNA fragmentation.

The ability of Cu(II) and Fe(III) to promote site-specific DNA damage in the presence of endogenous reductants was investigated by using 32P-5′-end-labeled DNA fragments obtained from the human p53 tumor suppressor gene and the c-Ha-ras-1 protooncogene. Ascorbate induced metal-dependent DNA damage most efficiently (ascorbate>GSH>NADH). Cu(II) induced endogenous reductants-dependent DNA damage more efficiently than Fe(III). Endogenous reductants plus Fe(III) caused DNA cleavage at every nucleotide, without marked site preference. DNA damage by Fe(III) was inhibited by hydroxyl free radical (⋅OH) scavengers and catalase. These results suggest that endogenous reductants plus Fe(III) generate free or extremely near free ⋅OH via H2O2 formation, and that ⋅OH causes DNA damage. In the presence of 50 μM Cu(II) in bicarbonate buffer, ascorbate caused DNA cleavage frequently at sites of two or more adjacent guanine residues. In contrast, in the presence of 20 μM Cu(II), ascorbate caused DNA cleavage frequently at thymine residues. Catalase and a Cu(I)-specific chelator inhibited DNA damage by Cu(II), whereas ⋅OH scavengers did not. Fe(III)-dependent 8-oxo-7,8-dihydro-2′-deoxyguanosine formation was inhibited by ⋅OH scavengers, whereas no inhibition by ⋅OH scavengers was observed with Cu(II). These results suggest that ⋅OH is the main active species formed with Fe(III), whereas copper-peroxide complexes with a reactivity similar to ⋅OH participate in Cu(II)-dependent DNA damage. The polyguanosine sequence specificity of DNA damage in the presence of high concentrations of Cu(II) can be explained by the preferential binding of Cu(II) to guanine residues.

It was recently reported that p-aminophenol (p-AP) induces DNA cleavage in mouse lymphoma cells, CHO cells and human lymphoblastoid cells. The mutagenicity of p-AP has not, however, been detected by reverse mutation assays. The purpose of this study was to assess the mutagenicity of p-AP by reverse mutation assay using Escherichia coli WP2uvrA/pKM101, which has a spectrum for detecting mutations different from those of other strains in the family with an AT base pair at the mutation site and has higher sensitivity to certain oxidative mutagens as compared to other strains. We found that p-AP was mutagenic to E. coli WP2uvrA/pKM101. The mutagenic activity of this compound was suppressed with the addition of dimethylsulfoxide or catalase, suggesting the involvement of active oxygen species in the mutagenic process induced by p-AP. To further elucidate the underlying mechanism, we used isolated DNA for the following experiments. It was revealed, by gel electrophoretic analysis, that p-AP induced DNA cleavage in the presence of Fe(III). However, p-AP alone did not induce this cleavage. Formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine by p-AP in calf thymus DNA was also detected in the presence of Fe(III) by HPLC with an electrochemical detector. ESR-spin trapping experiments using DMPO detected the production of hydroxyl radical (.OH) in the solution of p-AP with Fe(III). Both p-AP mediated DNA damages and .OH production by p-AP in the presence of Fe(III) were completely inhibited by .OH scavengers (ethanol, mannitol, sodium formate, dimethylsulfoxide) and catalase. These results suggest that .OH derived from the reaction between H2O2 and Fe(III) (Fenton reaction) participates in the oxidative DNA damage. Accordingly, the same mechanism might be working in E. coli WP2uvrA/pKM101 during induction of the mutation by p-AP.

Two hair dye components, carcinogenic 4-nitro-2-aminophenol and 5-nitro-2-aminophenol, induced Cu(II)-dependent DNA cleavage frequently at thymine and guanine residues in DNA fragments obtained from the c-Ha-ras-1 protooncogene. When the p53 tumor suppressor gene was used, 4-nitro-2-aminophenol caused Cu(II)-dependent piperidine-labile sites at poly G sequences. In the presence of Cu(II), both components increased 8-oxo-7,8-dihydro-2′-deoxyguanosine formation in DNA. The inhibitory effects of catalase and bathocuproine on DNA damage suggest the involvement of H2O2 and Cu(I). It is speculated that nitro-2-aminophenols undergo Cu(II)-mediated autoxidation to generate active oxygen species causing DNA damage which leads to their carcinogenesis.

The sequence of six amino acid residues -Ser-Cys-Cys-Ser-Cys-Cys- is present in all mammalian metallothionein sequences and has been highly conserved during evolution, although the metallothioneins have divergent primary sequences. To determine whether two serines in the sequence play a crucial role in metal-binding of metallothioneins, a mutant metallothionein with these two serines replaced by leucines was obtained using an Escherichia coli expression system. The expressed protein was analyzed for its chemical and spectroscopic properties. It was confirmed that the mutant metallothionein (MT) bound cadmium through a metal-thiolate complex and that there was no strong difference between the mutant and the wild-type MTs in retaining the metal-binding cluster. However, the metal-binding cluster of the mutant metallothionein was more unstable than that of the wild-type metallothionein. The two conservative serines could play a role in the stability of metal-binding ligands.

DNA cleavage induced by metallothionein (MT) containing copper was investigated by a DNA sequencing technique. Reconstituted Cd7-MT showed no ability to cause DNA cleavage. Commercially available rabbit MT I caused DNA cleavage, suggesting that DNA cleavage is due to the metal contained in commercial Mt. Cu2Cd5-MT and Cu12-MT were prepared by the treatment of commercial rabbit MT I with [Cu(CH3CN)4]CIO4. Cu12-MT frequently induced an alteration of thymine residues, especially in the 5'-GTC-3' sequence, and piperidine treatment led to chain cleavage at the thymine residues. The site specificity was similar to that obtained with Cu(I) plus H2O2. H2O2 enhanced DNA cleavage induced by Cu12-MT. Catalase and a Cu(I)-specific chelating agent, bathocuproine, inhibited DNA cleavage. These results suggest that Cu(I) and H2O2 have important roles in the production of active species causing DNA cleavage. Commercial MT and Cu2Cd5-MT induced DNA cleavage much less than Cu12-MT, but gave particularly specific DNA cleavage. Cu2Cd5-MT induced cleavage specifically at the central guanine residue of the 5'-GGT-3' sequence. A similar cleavage pattern was obtained with commercial MT. No effect of piperidine treatment suggests that the DNA cleavage might not be due to base damage and/or liberation. The DNA cleavage was inhibited efficiently by EDTA, but not by bathocuproine and catalase. Experiments with DNA ligands, albumin, and denatured DNA suggest that commercial MT and Cu2Cd5-MT induce nonoxidative cleavage of the deoxyribose phosphate backbone through its DNA recognition. These two types of cleavage mechanisms are discussed in relation to the possible role of Cu-MT in carcinogenesis.

The tryptophan metabolites 3-hydroxyanthranilic acid (3-HAA) and 3-hydroxykynurenine (3-HKyn) are carcinogens. DNA damage by 3-HAA and 3-HKyn in the presence of metal ions was investigated as a potential mechanism of their carcinogenicity. Pulsed field gel electrophoresis showed that in the presence of Mn(II), 3-HAA and 3-HKyn induced DNA double-strand breaks in cultured human cells. DNA single-strand breaks were observed with alkali treatment. The enhancing effect of catalase inhibitor and the inhibitory effect of o-phenanthroline on the strand breakage indicated the involvement of H 2 O 2 and endogenous transition metal ion. Damage to DNA fragments obtained from c-Ha-ras-1 protooncogene was investigated by a DNA sequencing technique. 3-HAA and 3-HKyn induced piperidine-labile sites frequently at thymine and guanine residues in the presence of Cu(II). The inhibitory effects of bathocuproine and catalase on Cu(II)-mediated DNA damage suggest that Cu(I) and H 2 O 2 have important roles in the production of active species causing DNA damage. The Cu(II)-mediated DNA damage was enhanced by preincubation of 3-HAA with Mn(II). UV-visible spectroscopy showed that Mn(II) and Cu(II) enhanced the rate of autoxidation of 3-HAA in different ways. These results suggest that in the presence of Mn(II) or Cu(II), these tryptophan metabolites produce H 2 O 2 , which is activated by transition metal ion to cause damage to DNA both in the case of isolated DNA and cultured cells

The U.S. National Toxicology Program has shown clear evidence of carcinogenicity of nickel subsulfide (Ni(3)S(2)) and some evidence of carcinogenicity of NiO (green) in rats. In the present study, DNA damage in cultured cells and in lungs of rats induced by nickel compounds was investigated to clarify the mechanism of nickel carcinogenesis. In cultured HeLa cells, Ni(3)S(2) induced a significant increase in 8-hydroxydeoxyguanosine (8-OH-dG) formation, whereas NiO (black), NiO (green), and NiSO(4) did not. On the other hand, in rats, intratracheal instillation of all these nickel compounds significantly increased 8-OH-dG content in the lungs. The disparities in DNA damage between cultured cells and animals could be accounted for by two different mechanisms for nickel-induced oxidative DNA damage in lungs of rats. One is direct oxidative DNA damage: Ni(II) enters the cells and then reacts with endogenous and/or nickel sulfide-produced hydrogen peroxide (H(2)O(2)) to give reactive oxygen species that cause DNA damage. This mechanism is supported by oxidative damage to isolated DNA treated with Ni(II) and H(2)O(2). The other mechanism is indirect oxidative DNA damage due to inflammation. This double mechanism for DNA damage may explain the relatively high carcinogenic risk associated with Ni(3)S(2).

Amitrole (3-amino-1,2,4-triazole) is a widely used herbicide. Amitrole induces thyroid and liver tumors in rodents. However, the mechanism of carcinogenesis by amitrole remains to be clarified. To clarify the mechanism of carcinogenesis induced by amitrole, we investigated the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), a characteristic of oxidatively generated DNA damage, by an amitrole metabolite, 3-amino-5-mercapto-1,2,4-triazole (AMT), in the presence of Cu(II). The amount of 8-oxodG was increased by AMT in the presence of Cu(II). AMT-induced 8-oxodG formation was enhanced in deuterium oxide (D₂O), which prolongs the half life of singlet oxygen (¹O₂), more than that in H₂O. Sodium azide and 1,4-diazabicyclo[2,2,2]-octane (DABCO), potent and relatively specific scavengers of ¹O₂, inhibited AMT-mediated 8-oxodG formation. Bathocuproine, a Cu(I) chelator, also inhibited the 8-oxodG formation. On the other hand, typical OH scavengers did not inhibit the generation of 8-oxodG. AMT plus Cu(II) also induced piperidine-labile DNA lesions frequently at every guanine residue. These results suggest that ¹O₂ and Cu(I) play an important role in DNA damage induced by AMT. It is concluded that oxidatively generated DNA damage induced by AMT via the generation of ¹O₂ may contribute to carcinogenicity of amitrole.

Cancer photodynamic therapy (PDT) requires photosensitizers that efficiently and selectively destroy tumor cells. We investigated 5,10,15,20-tetrakis (N-methyl-4-pyridyl)-21H,23H-porphyrin (TMPyP) as a potential cancer treatment. Confocal fluorescence microscopy showed that TMPyP was localized in the nuclei, whereas 5-aminolevulinic acid (ALA)-derived protoporphyrin IX (PPIX) was localized diffusely in the cytoplasm of human leukemia (HL-60) cells. In HL-60 cells under UVA irradiation, TMPyP effectively induced apoptosis. Moreover, 8-oxo-7,8-dihydro-2'-deoxyguanosine, an oxidative product of 2'-deoxyguanosine, was accumulated in the DNA of cells treated with photoirradiated TMPyP, whereas only small amounts were observed in ALA-treated cells in the presence of UVA light. TMPyP and UVA caused extensive damage at every guanine residue in DNA fragments obtained from the human p53 tumor suppressor gene and the c-Ha-ras-1 proto-oncogene, whereas PPIX induced little DNA damage under these conditions. Electron spin resonance spectroscopy using a singlet oxygen (1O2) probe and D2O showed that photoexcited TMPyP generated 1O2. These results suggest that photoexcited TMPyP reacts with oxygen to generate 1O2, which in turn, oxidizes guanine residues. Taken together, the results demonstrated that TMPyP was localized in the nucleus where it was photosensitized to induce DNA damage, suggesting that TMPyP may have clinical utility as a nucleus-targeted PDT.

Aspirin has been proposed as a possible chemopreventive agent. On the other hand, a recent cohort study showed that aspirin may increase the risk for pancreatic cancer. To clarify whether aspirin is potentially carcinogenic, we investigated the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), which is correlated with the incidence of cancer, in cultured cells treated with 2,3-dihydroxybenzoic acid (2,3-DHBA), a metabolite of aspirin. 2,3-DHBA induced 8-oxodG formation in the PANC-1 human pancreatic cancer cell line. 2,3-DHBA-induced DNA single-strand breaks were also revealed by comet assay using PANC-1 cells. Flow cytometric analyses showed that 2,3-DHBA increased the levels of intracellular reactive oxygen species (ROS) in PANC-1 cells. The 8-oxodG formation and ROS generation were also observed in the HL-60 leukemia cell line, but not in the hydrogen peroxide (H(2)O(2))-resistant clone HP100 cells, suggesting the involvement of H(2)O(2). In addition, an hprt mutation assay supported the mutagenicity of 2,3-DHBA. We investigated the mechanism underlying the 2,3-DHBA-induced DNA damage using (32)P-labeled DNA fragments of human tumor suppressor genes. 2,3-DHBA induced DNA damage in the presence of Cu(II) and NADH. DNA damage induced by 2,3-DHBA was enhanced by the addition of histone peptide-6 [AKRHRK]. Interestingly, 2,3-DHBA and histone peptide-6 caused base damage in the 5'-ACG-3' and 5'-CCG-3' sequences, hotspots of the p53 gene. Bathocuproine, a Cu(I) chelator, and catalase inhibited the DNA damage. Typical hydroxyl radical scavengers did not inhibit the DNA damage. These results suggest that ROS derived from the reaction of H(2)O(2) with Cu(I) participate in the DNA damage. In conclusion, 2,3-DHBA induces oxidative DNA damage and mutations, which may result in carcinogenesis.

Anthraquinones are typical photosensitizers used in photodynamic therapy (PDT). However, systemic toxicity is a major problem for anthraquinones due to their ability not only to bind DNA but also to cause oxidative stress even without photoirradiation. To avoid such disadvantages in cancer therapy, we designed and synthesized a novel 9-nitroanthracene derivative (1) as a precursor of anthraquinone. Under photoirradiation, 1 is converted into anthraquinone via generation of nitric oxide as confirmed by ESR. Strong DNA cleavage specifically at guanine under photoirradiation was also observed, characteristic of DNA-cleaving reactions by photoirradiated anthraquinones. We propose development of 1 as an alternative approach toward PDT that reduces the systemic toxicity of anthraquinone.

Although capsaicin exhibits antitumor activity, carcinogenic potential has also been reported. To clarify the mechanism for expression of potential carcinogenicity of capsaicin, we examined DNA damage induced by capsaicin in the presence of metal ion and various kinds of cytochrome P450 (CYP) using 32P-5'-end-labeled DNA fragments. Capsaicin induced Cu(II)-mediated DNA damage efficiently in the presence of CYP1A2 and partially in the presence of 2D6. CYP1A2-treated capsaicin caused double-base lesions at 5'-TG-3', 5'-GC-3' and CG of the 5'-ACG-3' sequence complementary to codon 273, a hotspot of p53 gene. DNA damage was inhibited by catalase and bathocuproine, a Cu(I) chelator, suggesting that reactive species derived from the reaction of H2O2 with Cu(I) participate in DNA damage. Formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine was significantly increased by CYP1A2-treated capsaicin in the presence of Cu(II). Therefore, we conclude that Cu(II)-mediated oxidative DNA damage by CYP-treated capsaicin seems to be relevant for the expression of its carcinogenicity.

Tetrahydropapaveroline (THP), a metabolite of dopamine, has been suspected to be associated with dopaminergic neurotoxicity of L-DOPA. THP induced apoptosis in human leukemia cell line HL-60 cells, but did not in its hydrogen peroxide (H(2)O(2))-resistant clone HP100. THP-induced DNA ladder formation in HL-60 cells was inhibited by a metal chelator. THP induced damage to (32)P-labeled DNA fragments in the presence of metals. In the presence of Fe(III)EDTA, THP caused DNA damage at every nucleotide. The DNA damage was inhibited by free hydroxy radical ((.)OH) scavengers and catalase, suggesting that the Fe(III)EDTA-mediated DNA damage is mainly due to (.)OH generation. In the presence of Cu(II), THP caused DNA damage mainly at T and G of 5'-TG-3' sequence. The inhibitive effect of catalase and bathocuproine on Cu(II)-mediated DNA damage suggested that H(2)O(2) and Cu(I) participate in the DNA damage. This study demonstrated that THP-induced apoptosis via reactive oxygen species generated from reaction of H(2)O(2) and metals plays an important role in cytotoxicity of L-DOPA.

In γ-irradiation, •OH is directly generated from water and causes DNA damage leading to carcinogenesis. Exposure of proteins to γ-irradiation, in the presence of oxygen, gives high yields of hydroperoxides. To clarify whether these hydroperoxides, particularly those formed on DNA-binding histone proteins, participate in γ-irradiation-induced carcinogenesis, experiments using 32P-labelled DNA fragments obtained from human cancer-related genes were undertaken. Histone protein-hydroperoxides induced significant DNA damage in the presence of Cu(I). Histone H1- and H3-hydroperoxides showed stronger DNA damage compared with histone H2A- and H4-hydroperoxides at 0.7 μM. Histone H1-hydroperoxides caused Cu(I)-dependent DNA damage predominantly at guanine residues, especially at 5′-GGC-3′, 5′-GGA-3′, 5′-GGT-3′ and single G bases. In contrast, histone H3-hydroperoxides/Cu(I) induced DNA damage at 5′-G in GG sequences; this sequence specificity is identical with that generated by 2,2′-azobis (2-amidinopropane) dihydrochloride, which is known to produce peroxyl radicals (RO2•). The difference in site specificity of DNA damage induced by histone H1- and H3-hydroperoxides may arise from their amino acid composition or their mode of binding to DNA. The histone H1-hydroperoxides/Cu(I) system also induced 8-oxo-7,8-dihydro-2′-deoxyguanosine formation in calf thymus DNA. It is concluded that histone protein-hydroperoxides can induce guanine-specific DNA damage, which may contribute to γ-irradiation-induced carcinogenesis.

Photosensitized DNA damage participates in solar-UV carcinogenesis, photogenotoxicity and phototoxicity. A chemoprevention of photosensitized DNA damage is one of the most important methods for the above phototoxic effects. In this study, the chemopreventive action of xanthone (XAN) derivatives (bellidifolin [BEL], gentiacaulein [GEN], norswertianin [NOR] and swerchirin [SWE]) on DNA damage photosensitized by riboflavin was demonstrated using [32P]-5'-end-labeled DNA fragments obtained from genes relevant to human cancer. GEN and NOR effectively inhibited the formation of piperidine-labile products at consecutive G residues by photoexcited riboflavin, whereas BEL and SWE did not show significant inhibition of DNA damage. The four XAN derivatives decrease the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodGuo), an oxidative product of G, by photoexcited riboflavin. The preventive action for the 8-oxodGuo formation of these XAN derivatives increased in the following order: GENNORBELSWE. A fluorescence spectroscopic study and ab initio molecular orbital calculations suggested that the prevention of DNA photodamage is because of the quenching of the triplet excited state of riboflavin by XAN derivatives through electron transfer. This chemoprevention is based on neither antioxidation nor a physical sunscreen effect; rather, it is based on the quenching of a photosensitizer. In conclusion, XAN derivatives, especially GEN, may act as novel chemopreventive agents by the quenching mechanism of an excited photosensitizer.

Reactive oxygen species (ROS) generated by environmental chemicals can cause sequence-specific DNA damage, which may lead to carcinogenesis and aging. We investigated the mechanism of DNA damage by environmental chemicals (catechol, propyl gallate and bisphenol-A), homocysteine and UVA radiation using human cultured cell lines and(32)P-labeled DNA fragments. Carcinogenic catechol induced piperidine-labile sites frequently at thymine residues in the presence of Cu(II) and NADH. Furthermore, catechol increased the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), a characteristic oxidative DNA lesion, in human leukemia cell line HL-60, but not in HP100, a hydrogen peroxide (H(2)O(2))-resistant cell line derived from HL-60. Thus, it is concluded that oxidative DNA damage through generation of H(2)O(2) plays an important role in the carcinogenic process of catechol. In addition, an environmental factor, bisphenol-A, and a dietary factor, propyl, gallate, also induced sequence-specific DNA damage via ROS generation.UVA, as well as UVB, contributes to photoaging. In humans, telomere shortening is believed to be associated with cell senescence. In this study, we investigated the shortening rate of telomeres in human WI-38 fibroblasts exposed to UVA irradiation. The telomere length (as measured by terminal restriction fragment length) in WI-38 fibroblasts irradiated with UVA decreased with increasing the irradiation dose. UVA irradiation with riboflavin caused damage specifically at the GGG sequence in the DNA fragments containing telomere sequence (TTAGGG)(4). We concluded that the GGG-specific damage in telomere sequence induced by UVA irradiation participates in the increase of the telomere shortening rate.In this report, we show our experimental results and discuss the mechanisms of sequence-specific DNA damage in relation to carcinogenesis and aging.

Acetamide is carcinogenic in rats and mice. To clarify the mechanism of carcinogenesis by acetamide, we investigated DNA damage by an acetamide metabolite, acetohydroxamic acid (AHA), using 32P-5′-end-labeled DNA fragments. AHA treated with amidase induced DNA damage in the presence of Cu(II) and displayed a similar DNA cleavage pattern of hydroxylamine. DNA damage was inhibited by both catalase and bathocuproine, suggesting that H2O2 and Cu(I) are involved. Carboxy-PTIO, a specific scavenger of nitric oxide (NO), partially inhibited DNA damage. The amount of 8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodG) by amidase-treated AHA was similar to that by hydroxylamine. ESR spectrometry revealed that amidase-treated AHA as well as hydroxylamine generated NO in the presence of Cu(II). From these results, it has been suggested that AHA might be converted into hydroxylamine by amidase. These results suggest that metal-mediated DNA damage mediated by amidase-catalyzed hydroxylamine generation plays an important role in the carcinogenicity of acetamide.

We investigated whether oxidative stress, which contributes to aging, accelerates the telomere shortening in human cultured cells. The terminal restriction fragment (TRF) from WI-38 fibroblasts irradiated with UVA (365-nm light) decreased with increasing of the irradiation dose. Furthermore, UVA irradiation dose-dependently increased the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG) in both WI-38 fibroblasts and HL-60 cells. In order to clarify the mechanism of the acceleration of telomere shortening, we investigated site-specific DNA damage induced by UVA irradiation in the presence of endogenous photosensitizers using (32)P 5' end-labeled DNA fragments containing telomeric oligonucleotide (TTAGGG)(4). UVA irradiation with riboflavin induced 8-oxodG formation in the DNA fragments containing telomeric sequence, and Fpg protein treatment led to chain cleavages at the central guanine of 5'-GGG-3' in telomere sequence. Human 8-oxodG-DNA glycosylase introduces a chain break in a double-stranded oligonucleotide specifically at an 8-oxodG residue. The amount of 8-oxodG formation in DNA fragment containing telomere sequence [5'-CGC(TTAGGG)(7)CGC-3'] was approximately five times more than that in the DNA fragment containing nontelomere sequence [5'-CGC(TGTGAG)(7)CGC-3']. Furthermore, H(2)O(2) plus Cu(II) caused DNA damage, including 8-oxodG formation, specifically at the GGG sequence in the telomere sequence (5'-TTAGGG-3'). It is concluded that the formation of 8-oxodG at the GGG triplet in telomere sequence induced by oxidative stress could participate in acceleration of telomere shortening.

Melatonin, an indolic pineal hormone, is produced primarily at night in mammals and is important in controlling biological rhythms. Although melatonin is known to be effective as a free radical scavenger and has an anti-cancer effect, carcinogenic properties have also been reported. In relation to its carcinogenic potential, we have examined whether 6-hydroxymelatonin, a major melatonin metabolite, can induce DNA damage in the presence of metal ion using [ 32 P]-5′-end-labeled DNA fragments obtained from genes relevant to human cancer. 6-Hydroxymelatonin induced site-specific DNA damage in the presence of Cu(II). Formamidopyrimidine-DNA glycosylase treatment induced cleavage sites mainly at G residues of the 5′-TG-3′ sequence, whereas piperidine treatment induced cleavage sites at T mainly of 5′-TG-3′. Interestingly, 6-hydroxymelatonin strongly damaged G and C of the 5′-ACG-3′ sequence complementary to codon 273 of the p53 gene. These results suggest that 6-hydroxymelatonin can cause double-base lesions. DNA damage was inhibited by both catalase and bathocuproine, Cu(I)-specific stabilizer, suggesting that reactive species derived from the reaction of H 2 O 2 with Cu(I) participate in DNA damage. Cytochrome P450 reductase efficiently enhanced 6-hydroxymelatonin-induced oxidative DNA damage and oxygen consumption, suggesting the formation of redox cycle. It is noteworthy that 6-hydroxymelatonin can efficiently induce DNA damage via non- o -quinone type of redox cycle. Formation of 8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodG), a characteristic oxidative DNA lesion, in calf thymus DNA was significantly increased by 6-hydroxymelatonin in the presence of Cu(II). Furthermore, 6-hydroxymelatonin significantly increased the formation of 8-oxodG in human leukemia cell line HL-60 but not in HP100, a hydrogen peroxide (H 2 O 2 )-resistant cell line derived from HL-60. The 6-hydroxymelatonin-induced 8-oxodG formation in HL-60 cells significantly decreased by the addition of bathocuproine or o -phenanthroline. Therefore, it is concluded that melatonin may exhibit carcinogenic potential through oxidative DNA damage by its metabolite.

Green tea catechins have antimutagenic and anticarcinogenic activities. On the other hand, several epidemiological studies have indicated significant positive relationship between green tea consumption and cancer. Catechins enhance colon carcinogenesis in rats initiated with chemical carcinogen. To clarify the mechanism underlying the potential carcinogenicity, we investigated the DNA-damaging ability of catechins in human cultured cells. Catechin increased the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), a characteristic oxidative DNA lesion, in human leukemia cell line HL-60 but not in HP100, a hydrogen peroxide (H2O2)-resistant cell line derived from HL-60. The catechin-induced formation of 8-oxodG in HL-60 cells significantly decreased by bathocuproine. Furthermore, we investigated DNA damage and its site-specificity induced by catechins, using 32P-labeled DNA fragments. Catechin and epicatechin induced extensive DNA damage in the presence of Cu(II). Catechin caused piperidine-labile sites at thymine and cytosine residues in the presence of Cu(II). Catalase and bathocuproine inhibited the DNA damage, indicating the involvement of H2O2 and Cu(I). NADH enhanced catechins plus Cu(II)-induced 8-oxodG formation in calf thymus DNA, suggesting the redox cycle between catechins and their corresponding quinones, the oxidized forms of catechins. The DNA-damaging ability of epicatechin is stronger than that of catechin, possibly due to the greater turnover frequency of the redox cycle. The difference in their redox properties could be explained by their redox potentials estimated form an ab initio molecular orbital calculation. The present study demonstrated that catechins could induce metal-dependent H2O2 generation during the redox reactions and subsequently damage to cellular and isolated DNA. Therefore, it is reasonably considered that green tea catechins may have the dual function of anticarcinogenic and carcinogenic potentials.

Procarbazine [N-isopropyl-alpha-(2-methylhydrazino)-p-toluamide], a hydrazine derivative, which has been shown to have effective antineoplastic activity, induces cancer in some experimental animals and humans. To clarify a new mechanism for its carcinogenic effect, we examined DNA damage induced by procarbazine in the presence of metal ion, using 32P-5'-end-labeled DNA fragments obtained from the human p53 tumor suppressor gene and the c-Ha-ras-1 protooncogene. Procarbazine plus Cu(II) induced piperidine-labile and formamidopyrimidine-DNA glycosylase-sensitive lesions at the 5'-ACG-3' sequence, complementary to a hotspot of the p53 gene, and the 5'-TG-3' sequence. Catalase partially inhibited DNA damage, suggesting that not only H(2)O(2) but also other reactive species are involved. Procarbazine plus Cu(II) significantly increased the formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine, which was completely inhibited by calatase. Electron spin resonance spin-trapping experiments revealed that methyl radicals were generated from procarbazine and Cu(II). On the basis of these findings, it is considered that procarbazine causes DNA damage through non-enzymatic formation of the Cu(I)-hydroperoxo complex and methyl radicals. In conclusion, in addition to alkylation, oxidative DNA damage may play important roles in not only antitumor effects but also mutagenesis and carcinogenesis induced by procarbazine.

Propyl gallate (PG), widely used as an antioxidant in foods, is carcinogenic to mice and rats. PG increased the amount of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG), a characteristic oxidative DNA lesion, in human leukemia cell line HL-60, but not in HP100, which is hydrogen peroxide (H2O2)-resistant cell line derived from HL-60. Although PG induced no or little damage to 32P-5'-end-labeled DNA fragments obtained from genes that are relevant to human cancer, DNA damage was observed with treatment of esterase. HPLC analysis of the products generated from PG incubated with esterase revealed that PG converted into gallic acid (GA). GA induced DNA damage in a dose-dependent manner in the presence of Fe(III)EDTA or Cu(II). In the presence of Fe(III) complex such as Fe(III)EDTA or Fe(III)ADP, GA caused DNA damage at every nucleotide. Fe(III) complex-mediated DNA damage by GA was inhibited by free hydroxy radical (*OH) scavengers, catalase and an iron chelating agent. These results suggested that the Fe(III) complex-mediated DNA damage caused by GA is mainly due to *OH generated via the Fenton reaction. In the presence of Cu(II), DNA damage induced by GA occurred at thymine and cytosine. Although *OH scavengers did not prevent the DNA damage, methional inhibited the DNA damage. Cu(II)-mediated DNA damage was inhibited by catalase and a Cu(I) chelator. These results indicated that reactive oxygen species formed by the interaction of Cu(I) and H2O2 participates in the DNA damage. GA increased 8-oxodG content in calf thymus DNA in the presence of Cu(II), Fe(III)EDTA or Fe(III)ADP. This study suggested that metal-mediated DNA damage caused by GA plays an important role in the carcinogenicity of PG.

The anticancer mechanism of doxorubicin (DOX), an anthracycline antibiotic, is believed to involve DNA damage through topoisomerase II inhibition and free radical generation. The free radical generation may also participate in genotoxicity, as well as cardiotoxicity, in normal human cells. The present study showed that DOX generates 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxo-dG), an indicator of oxidative DNA damage, in HL-60 cells, but not in H(2)O(2)-resistant HP100 cells, suggesting the involvement of H(2)O(2) in cellular DNA damage. Since DOX has both p-quinone and p-hydroquinone residues, free radical generation can be initiated by either reduction or oxidation of DOX. To clarify whether the oxidized or reduced form is more important for DOX-induced H(2)O(2) generation, we investigated the site-specific DNA damage induced by DOX in the presence of Cu(II), in comparison with that in the presence of cytochrome P450 reductase, using (32)P-labeled DNA fragments. DOX caused DNA damage in the presence of Cu(II) or cytochrome P450 reductase. The degree of Cu(II)-mediated DNA damage, including 8-oxodG formation, was much greater than that of cytochrome P450 reductase-mediated DNA damage. DOX plus Cu(II) caused DNA damage specifically at guanine, thymine and cytosine residues, particularly at 5'-GG-3', 5'-GT-3' and 5'-TG-3' sequences. Scavenger experiments suggested the involvement of reactive species generated from H(2)O(2) and Cu(I). When cytochrome P450 reductase and NADPH were used instead of Cu(II), every nucleotide was uniformly damaged, suggesting the participation of.OH. We conclude that DOX may induce carcinostatic and genotoxic effects through oxidation of its p-hydroquinone moiety by metal ion rather than through p-quinone reduction by cytochrome P450 reductase.

UVA contributes to skin cancer by solar UV light. Photosensitizers are believed to play an important role in UVA carcinogenesis. We investigated the mechanism of DNA damage induced by photoexcited xanthone (XAN) analogues (XAN, thioxanthone [TXAN] and acridone [ACR]), exogenous photosensitizers, and the relationship between the DNA-damaging abilities and their highest occupied molecular orbital (HOMO) energies. DNA damage by these photosensitizers was examined using 32P-labeled DNA fragments obtained from the p53 tumor suppressor gene. Photoexcited XAN caused DNA cleavage specifically at 5'-G of the GG sequence in the double-stranded DNA only when the DNA fragments were treated with piperidine, suggesting that DNA cleavage is due to base modification with little or no strand breakage. With denatured single-stranded DNA, the extent of XAN-sensitized photodamage was decreased. An oxidative product of G, 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxo-dGuo), was formed by photoexcited XAN, and the 8-oxo-dGuo formation was decreased in single-stranded DNA. TXAN and ACR induced DNA photodamage as did XAN, although the order of DNA-damaging ability was XANTXANACR. These findings suggest that photoexcited XAN analogues induce nucleobase oxidation at 5'-G of GG sequence in double-stranded DNA through electron transfer. The HOMO energies of these photosensitizers, estimated from ab initio molecular orbital (MO) calculation, decreased in the following order: XANTXANACR. Extents of DNA damage increased exponentially with the HOMO energies of XAN analogues. This study suggests that DNA-damaging abilities of photosensitizers can be estimated from their HOMO energies.

Semicarbazide, a hydrazine derivative, is carcinogenic to mice but shows no or little mutagenicity in the Salmonella-microsome test. To clarify whether or not the genotoxic mechanism contributes to the non-mutagenic carcinogenicity of semicarbazide, we investigated DNA damage induced by semicarbazide using 32P-5'-end-labeled DNA fragments obtained from the c-Ha-ras-1 protooncogene and the p53 tumor suppressor gene. Semicarbazide caused DNA damage frequently at the thymine and cytosine residues in the presence of Cu(II). Catalase and bathocuproine partially inhibited DNA damage, suggesting that hydrogen peroxide plus Cu(I) participates in DNA damage. When a high concentration of semicarbazide was used in the presence of catalase, DNA damage was induced, especially at G in 5'-AG and slightly at 5'-G in GG and GGG sequences. An electron paramagnetic resonance (EPR) spectroscopic study has confirmed that the reaction of semicarbazide with Cu(II) produces carbamoyl radicals (z.rad;CONH(2)), possibly generated via the nitrogen-centered radicals of semicarbazide. Azodicarbonamide also produced carbamoyl radicals and induced DNA damage frequently at 5'-G in GG and GGG sequences, suggesting that carbamoyl radicals participate in this sequence-specific DNA damage by semicarbazide. On the basis of our previous reports, we consider that the sequence-specific DNA damage at G in 5'-AG in the present study is due to the nitrogen-centered radicals. This study has shown that semicarbazide induces DNA damage in the presence of Cu(II) through the formation of hydrogen peroxide and Cu(I). In addition, semicarbazide-derived free radicals participate in DNA damage. DNA damage induced by these reactive species may be relevant to the carcinogenicity of semicarbazide.

DNA damage mediated by photosensitizers participates in solar carcinogenesis. Fluorescence measurement and high-performance liquid chromatography analysis demonstrated that photoirradiated folic acid, one of the photosensitizers in cells, generates pterine-6-carboxylic acid (PCA). Experiments using 32 P -labeled DNA fragments obtained from a human gene showed that ultraviolet A-irradiated folic acid or PCA caused DNA cleavage specifically at consecutive G residues in double-stranded DNA after Escherichia coli formamidopyrimidine–DNA glycosylase or piperidine treatment. The amount of 8-oxo-7,8-dihydro-2′-deoxyguanosine formed through this DNA photoreaction in double-stranded DNA exceeded that in single-stranded DNA. Kinetic studies suggested that DNA damage is caused mainly by photoexcited PCA generated from folic acid rather than by folic acid itself. In conclusion, photoirradiated folic acid generates PCA, which induces DNA photooxidation specifically at consecutive G residues through electron transfer. Excess intake of folic acid supplements may increase a risk of skin cancer by solar ultraviolet light.

We examined the mechanism of DNA damage induced by carcinogenic Ni(II) in the presence of SH compounds. In the presence of model endogenous SH compounds, dithiothreitol (DTT), 1,4-dithio-L-threitol, and dithioerythritol, Ni(II) induced damage to 3 2 P-5'-end-labeled DNA fragments obtained from the human c-Ha-ras-1 protooncogene and the p53 tumor suppressor gene. The intensity of Ni(II)-mediated DNA damage induced by DTT was stronger than that by other model endogenous SH compounds, 1,4-dithio-L-threitol and dithioerythritol. DNA damage induced by Ni(II) plus DTT was observed only when the DNA was treated with piperidine, suggesting that Ni(II) plus DTT caused only base damage. Formamidopyrimidine-DNA glycosylase, which is known to recognize 8-oxodG as well as Fapy residues, treatment induced cleavage sites, mainly guanine residues, particularly at the 5'-GG-3', 5'-GGG-3', and 5'-GGGG-3' sequences, in DNA incubated with Ni(II) in the presence of DTT. SOD and catalase inhibited the DNA damage, suggesting that DNA damage involved superoxide anion and hydrogen peroxide. Sodium azide, a potent and relatively specific scavenger of 1 O 2 , inhibited DNA damage by Ni(II) in the presence of DTT, whereas the sequence specificity of DNA damage was different from that obtained by 1 O 2 generating agent. The formation of 8-oxodG in calf thymus DNA by Ni(II) was observed with the physiological thiols, dihydrolipoic acid and mercaptopyruvate, as well as with DTT. These results suggest that Ni(II) and DTT form a reactive species, which may be responsible for causing guanine-specific DNA damage. Endogenous SH compounds, which have similar chemical structures to DTT, would participate in nickel carcinogenesis through causing oxidative DNA damage.