Gut microbes define liver cancer risk in mice exposed to chemical and viral transgenic hepatocarcinogens.

BACKGROUND AND AIMS: Hepatocellular carcinoma (HCC) frequently results from synergism between chemical and infectious liver carcinogens. Worldwide, the highest incidence of HCC is in regions endemic for the foodborne contaminant aflatoxin B1 (AFB1) and hepatitis B virus (HBV) infection. Recently, gut microbes have been implicated in multisystemic diseases including obesity and diabetes. Here, the hypothesis that specific intestinal bacteria promote liver cancer was tested in chemical and viral transgenic mouse models. METHODS: Helicobacter-free C3H/HeN mice were inoculated with AFB1 and/or Helicobacter hepaticus. The incidence, multiplicity and surface area of liver tumours were quantitated at 40 weeks. Molecular pathways involved in tumourigenesis were analysed by microarray, quantitative real-time PCR, liquid chromatography/mass spectrometry, ELISA, western blot and immunohistochemistry. In a separate experiment, C57BL/6 FL-N/35 mice harbouring a full-length hepatitis C virus (HCV) transgene were crossed with C3H/HeN mice and cancer rates compared between offspring with and without H hepaticus. RESULTS: Intestinal colonisation by H hepaticus was sufficient to promote aflatoxin- and HCV transgene-induced HCC. Neither bacterial translocation to the liver nor induction of hepatitis was necessary. From its preferred niche in the intestinal mucus layer, H hepaticus activated nuclear factor-kappaB (NF-kappaB)-regulated networks associated with innate and T helper 1 (Th1)-type adaptive immunity both in the lower bowel and liver. Biomarkers indicative of tumour progression included hepatocyte turnover, Wnt/beta-catenin activation and oxidative injury with decreased phagocytic clearance of damaged cells. CONCLUSIONS: Enteric microbiota define HCC risk in mice exposed to carcinogenic chemicals or hepatitis virus transgenes. These results have implications for human liver cancer risk assessment and prevention.

Formaldehyde cross-linking and immunoprecipitation demonstrate developmental changes in H1 association with transcriptionally active genes.

The in vivo association of histone H1 with specific genes in Tetrahymena thermophila was studied by using a simplified cross-linking and immunoprecipitation technique. Four genes were analyzed whose activities vary in three different developmental states (logarithmic growth, starvation, and conjugation). Hybridization of the immunoprecipitated DNA to cloned probes showed an inverse correlation between the level of immunoprecipitation with H1 antiserum and transcriptional activity. This represents the first demonstration of an alteration in histone H1-DNA interaction associated with developmental changes in transcriptional activity.

Effect of diethyldithiocarbamate on cis-diamminedichloroplatinum(II)-induced cytotoxicity, DNA cross-linking, and gamma-glutamyl transpeptidase inhibition.

Diethyldithiocarbamate (DDTC) has been shown to protect against the toxicity of cis-diamminedichloroplatinum(II) (DDP) without inhibition of antitumor effect. We report here that DDTC is unreactive toward DDP complexes in which both chlorides have been replaced by guanine residues but removes platinum from a variety of other ligands, and that this difference in reactivity may provide the basis for the selective protection observed with DDTC. Platinum-DNA complexes were unreactive toward DDTC (10 mM, greater than 4 h) when the platinum:base ratio r less than 0.02. DDTC did not react with the 1:2 complex of DDP:guanosine but reacted rapidly with the 1:1 complex and with the 1:2 complexes of DDP:adenosine. Reaction of DDP with DDTC was second order with a rate constant k = 4.4 M-1 min-1 at 37 degrees C, corresponding to a t 1/2 = 150 min at [DDTC] = 1 mM. Treatment of L1210 cells with DDTC (0.5-1 mM) after exposure to DDP indicated that DDTC had no effect on cell kill if DDTC treatment was delayed for 1 h after DDP. The effect of DDTC on DDP-induced DNA interstrand cross-links was also examined in L1210 cells. Interstrand cross-links were decreased by approximately 50% when cells were treated with DDTC immediately after DDP; no change in DNA interstrand cross-links was observed when DDTC treatment occurred 3 h after DDP. A modified alkaline elution procedure was used to evaluate the effects of high concentrations of DDTC, thiourea, and cyanide on platinum:DNA cross-links from L1210 DNA. Exposure to DDTC (0.5 M, 4 h) did not alter interstrand cross-links, but both thiourea and cyanide caused extensive reversal of cross-links at concentrations as low as 10 and 1 mM, respectively. Both commercial and rat kidney brush border preparations of gamma-glutamyl transpeptidase were inhibited by exposure to 2 mM DDP; exposure of the inhibited enzyme to DDTC (1 or 10 mM) resulted in significant restoration of enzyme activity. These data indicate that DDTC has unique chemical specificity in its reactions with platinum complexes and that this specificity is ideal for application as a chemoprotective drug against cis-platinum toxicity.

Selective protection against cis-diamminedichloroplatinum(II)-induced toxicity in kidney, gut, and bone marrow by diethyldithiocarbamate.

Diethyldithiocarbamate (DDTC) has been shown to inhibit nephrotoxicity induced by cis-platinum (DDP) without inhibition of tumor response in the rat. We report here that DDTC at doses of 25-300 mg/kg inhibits DDP-induced nephrotoxicity and bone marrow toxicity in C57BL/6 X DBA/2F1 (hereafter called B6D2F1) mice, F344 rats, and beagle dogs and is also antiemetic in the dog. DDTC doses which afford excellent protection do not decrease median survival time following DDP treatment in L1210 and P388 leukemias, B16 melanoma, and Lewis lung and colon 26 carcinomas in B6D2F1 mice when DDTC is given 2 h after DDP. Preliminary experiments indicate that DDTC does not alter median survival time after treatment of P388 leukemia with the platinum analogues diammine(1,1-cyclobutanedicarboxylato)platinum(II) and cis-diisopropylamine-cis-dichloro-trans-dihydroxyplatinum(IV ). Maximum blood urea nitrogen levels after DDP treatment are reduced significantly by DDTC in all species; blood urea nitrogen elevation, total kidney platinum, weight loss, and leukopenia correlate with DDP-DDTC interval in the rat and indicate optimum protection at 2 h, the shortest interval examined. Bone marrow toxicity was assessed by peripheral white blood cell counts in all species and by marrow cellularity in the mouse. White blood cell nadirs were higher and bone marrow recovered more rapidly after DDTC compared with DDP given alone. DDP reduced marrow cellularity 50-60% in the mouse; administration of DDTC 2 h after DDP afforded no protection to the lymphocytes in the marrow but maintained the granulocyte + precursor population near control levels. DDTC plasma pharmacokinetic values have been determined after s.c., i.p., and i.v. administration in the mouse, rat, and dog. Peak plasma levels of 0.3-1.2 mM are observed after a 250-mg/kg dose, with a plasma half-life of 10-20 min. Our data indicate that DDTC may provide protection against most clinically significant toxicities arising from cis-platinum at doses which do not inhibit tumor response.

Characterization of the reactions of platinum antitumor agents with biologic and nonbiologic sulfur-containing nucleophiles.

Substitution reactions with biologic nucleophiles appear to govern the antitumor and toxic properties of platinum complexes. In this paper we have characterized the reactions of several platinum antitumor agents with sulfur-containing amino acids, peptides, proteins, and nonbiologic nucleophiles. The rate constants for the reactions of trans-diamminedichloroplatinum(II) (trans-DDP), cis-diamminedichloroplatinum(II) (DDP), diammine (1,1-cyclobutanedicarboxylato)platinum(II) (CBDCA) and cis-diisopropylamine-cis-dichloro-trans-dihydroxy platinum(IV) (CHIP) with cysteine (Cys), methionine (Met), and glutathione (GSH) were determined at 37 degrees. A reactivity ratio of 1:1.5:22:6500 was determined for the reaction of GSH with CHIP, CBDCA, DDP, and trans-DDP respectively. The rate constant for the binding of DDP to DNA, 7.4 X 10(-5) sec-1, decreased to 5.9 X 10(-5) sec-1 and 1.7 X 10(-5) sec-1 in the presence of 0.5 and 5 mM GSH respectively. The products formed in the reaction of GSH with trans-DDP, DDP, and CBDCA were also examined. Under conditions of high platinum concentration (2-3 mM), CBDCA and DDP form large molecular weight species with GSH as indicated by 1H-NMR and ultrafiltration experiments. The complex [Pt(GSH)2 X 3H2O]n was isolated from the reaction of 3 mM DDP with 6 mM GSH. The product formed in the reaction of 3 mM trans-DDP with 6 mM GSH was not macromolecular in nature, and 1H-NMR spectra revealed that platinum was bound to the Cys sulfhydryl group. Rate constants were determined for the reactions of these platinum complexes with diethyldithiocarbamate (DDTC) and thiosulfate, two agents known to reduce platinum-mediated nephrotoxicity. DDTC, but not thiosulfate, was shown to rapidly chelate platinum from [Pt(GSH)2 X 3H2O]n. The effects of DDP, CBDCA, and CHIP on the sulfhydryl-dependent rat renal proximal tubule membrane enzymes alkaline phosphatase (AP), gamma-glutamyltranspeptidase (GGTP), leucine aminopeptidase (LAP), and the Na+/K+- and Mg2+-adenosine-5'-triphosphatases (ATPases) were also investigated in vitro. The ability of platinum complexes to inhibit these enzymes parallels their reactivity with other nucleophiles. DDTC and thiourea were shown to restore activity to platinum-inhibited enzymes. Chloride ion was found to reduce platinum-mediated enzyme inhibition in an unpredictable manner, the greatest effect being observed with LAP and GGTP and the least with the ATPases. None of these renal enzymes was directly inhibited by DDP in vivo.

Removal by human apurinic/apyrimidinic endonuclease 1 (Ape 1) and Escherichia coli exonuclease III of 3'-phosphoglycolates from DNA treated with neocarzinostatin, calicheamicin, and gamma-radiation.

DNA strand breaks with terminal 3'-phosphoglycolate groups are produced by agents that can abstract the hydrogen atom from the 4'-carbon of DNA deoxyribose groups. Included among these agents are gamma-radiation (via the OH radical) and enediyne compounds, such as neocarzinostatin and calicheamicin. However, while the majority of radiation-induced phosphoglycolates are found at single-strand breaks, most of the phosphoglycolates generated by these two enediynes are found at bistranded lesions, including double-strand breaks. Using a 32P-post-labelling assay, we have compared the enzyme-catalyzed removal of phosphoglycolates induced by each of these agents. Both human apurinic/apyrimidinic endonuclease 1 (Ape 1) and its Escherichia coli homolog exonuclease III rapidly removed over 80% of phosphoglycolates from gamma-irradiated DNA, although there appeared to be a small resistant subpopulation. The neocarzinostatin-induced phosphoglycolates were removed more slowly, though not to completion, while the calicheamicin-induced phosphoglycolates were extremely refractory to both enzymes. These data suggest that unless other enzymes are capable of acting upon the phosphoglycolate termini at enediyne-induced double-strand breaks, such termini will be resistant to end rejoining repair pathways.

An improved method for the rapid assessment of DNA bending by small molecules.

Assessing the effects of non-covalently bound chemicals on DNA structure is particularly challenging. Traditional methods require the use of cumbersome electrophoretic techniques or that the compounds bind DNA with an extremely high affinity. Here we demonstrate that, by extending the use of the exonuclease Bal 31, we can rely on a standard cyclization assay technique and one dimensional gel electrophoresis to identify and quantitate chemical induced DNA bending. An important application of this method is to the study of small molecules that bind to DNA non-covalently and we illustrate the method with the antitumor antibiotic calicheamicin. Our results suggest that the distribution of circles resulting from the polymerization of a 21 bp DNA construct reflects the kinetics of the competing cyclization and dimerization reactions and provides a method for rapidly screening compounds for DNA bending.

Peroxynitrite-induced DNA damage in the supF gene: correlation with the mutational spectrum.

Tissue inflammation and chronic infection lead to the overproduction of nitric oxide and superoxide. These two species rapidly combine to yield peroxynitrite (ONOO(-)), a powerful oxidizing and nitrating agent that is thought be involved in both cell death and an increased cancer risk observed for inflamed tissues. ONOO(-) has been shown to induce single-strand breaks and base damage in DNA and is mutagenic in the supF gene, inducing primarily G to T transversions clustered at the 5' end of the gene. The mutagenicity of ONOO(-) is believed to result from chemical modifications at guanine nucleobases leading to miscoding DNA lesions. In the present work, we applied a combination of molecular and analytical techniques in an attempt to identify biologically important DNA modifications induced by ONOO(-). pUC19 plasmid treated with ONOO(-) contained single-strand breaks resulting from direct sugar damage at the DNA backbone, as well as abasic sites and nucleobase modifications repaired by Fpg glycosylase. The presence of carbon dioxide in the reaction mixture shifted the ONOO(-) reactivity towards reactions at nucleobases, while suppressing the oxidation of deoxyribose. To further study the chemistry of the ONOO(-) interactions with DNA, synthetic oligonucleotides representing the mutation-prone region of the supF gene were treated with ONOO(-), and the products were analyzed by liquid chromatography-negative ion electrospray ionization mass spectrometry (LC-ESI(-) MS) and tandem mass spectrometry. 8-Nitroguanine (8-nitro-G) was formed in ONOO(-)-treated oligonucleotides in a dose-dependent manner with a maximum at a ratio of [ONOO(-)]: [DNA]=10 and a decline at higher ONOO(-) concentrations, suggesting further reactions of 8-nitro-G with ONOO(-). 8-Nitro-G was spontaneously released from oligonucleotides (t(1/2)=1 h at 37 degrees C) and, when present in DNA, was not recognized by Fpg glycosylase. To obtain more detailed information on ONOO(-)-induced DNA damage, a restriction fragment from the pSP189 plasmid containing the supF gene (135 base pairs) was [32P]-end-labeled and treated with ONOO(-). PAGE analysis of the products revealed sequence-specific lesions at guanine nucleobases, including the sites of mutational "hotspots." These lesions were repaired by Fpg glycosylase and cleaved by hot piperidine treatment, but they were resistant to depurination at 90 degrees C. Since 8-nitro-G is subject to spontaneous depurination, and 8-oxo-guanine is not efficiently cleaved by piperidine, these results suggest that alternative DNA lesion(s) contribute to ONOO(-) mutagenicity. Further investigation of the identities of DNA modifications responsible for the adverse biological effects of ONOO(-) is underway in our laboratory.

Chemistry meets biology in colitis-associated carcinogenesis.

The intestine comprises an exceptional venue for a dynamic and complex interplay of numerous chemical and biological processes. Here, multiple chemical and biological systems, including the intestinal tissue itself, its associated immune system, the gut microbiota, xenobiotics, and metabolites meet and interact to form a sophisticated and tightly regulated state of tissue homoeostasis. Disturbance of this homeostasis can cause inflammatory bowel disease (IBD)-a chronic disease of multifactorial etiology that is strongly associated with increased risk for cancer development. This review addresses recent developments in research into chemical and biological mechanisms underlying the etiology of inflammation-induced colon cancer. Beginning with a general overview of reactive chemical species generated during colonic inflammation, the mechanistic interplay between chemical and biological mediators of inflammation, the role of genetic toxicology, and microbial pathogenesis in disease development are discussed. When possible, we systematically compare evidence from studies utilizing human IBD patients with experimental investigations in mice. The comparison reveals that many strong pathological and mechanistic correlates exist between mouse models of colitis-associated cancer, and the clinically relevant situation in humans. We also summarize several emerging issues in the field, such as the carcinogenic potential of novel inflammation-related DNA adducts and genotoxic microbial factors, the systemic dimension of inflammation-induced genotoxicity, and the complex role of genome maintenance mechanisms during these processes. Taken together, current evidence points to the induction of genetic and epigenetic alterations by chemical and biological inflammatory stimuli ultimately leading to cancer formation.

Determination of binding mode: intercalation.

A small molecule can be assumed to bind to DNA by intercalation between base pairs if it causes lengthening and unwinding of the DNA helix and undergoes changes in its spectral properties, such as DNA-induced hypochromism and quenching of its UV absorbance. DNA lengthening and unwinding can be determined from the change in viscosity of a solution of linear or plasmid DNA, respectively. Intercalation of a ligand can also be seen as a reduction in the UV/visible absorbance of the intercalator, as well as a shift in the absorbance maximum.

Determination of binding thermodynamics.

This unit serves as a starting point for exploring the thermodynamic properties of interactions between small molecules and DNA. It covers the determination of simple, apparent association/dissociation constants. The concentration of DNA-bound ligand and free ligand are determined and a binding constant is extracted from these data. Data gathering and curve fitting are discussed.

Influence of thiol structure on neocarzinostatin activation and expression of DNA damage.

Neocarzinostatin (NCS) is an enediyne antitumor antibiotic that cleaves DNA following a thiol-induced electronic rearrangement to a diradical form. Structure-function studies with 11 thiol-containing compounds were undertaken to clarify the role of the thiol in NCS-mediated DNA damage. The rates of activation of NCS in the presence of DNA with the various thiols approximated a Brønsted relation (beta = 0.43, r2 = 0.86), which suggests that the basicity/nucleophilicity of the thiol is important to NCS activation. However, an additional contribution to NCS activation may arise from the affinity of the thiol for DNA, since there is a correlation between the concentration of thiol producing maximal DNA damage, assessed by quantitating the topologic forms of plasmid pBR322 following treatment with NCS, and the apparent ability of the thiol to bind to DNA by hydrophobic or electrostatic interactions. The overall second-order rate constants for the activation of NCS were found to be inversely correlated with the thiol optima; a plot of the former versus the reciprocal of the optimal thiol concentration revealed a first-order rate constant of activation of 0.013 s-1 in the presence of DNA. This indicates that maximal DNA damage occurs when NCS is activated with a half-life of 52 s, a relatively slow rate of activation that suggests that NCS binds to DNA before undergoing activation by thiol. Finally, an analysis of strand breaks in pBR322 shows that thiols possessing a carboxylate moiety produce larger quantities of bistranded DNA lesions than their esterified or non-carboxylate-containing counterparts.

Analysis of oxidized DNA fragments by gel electrophoresis.

Polyacrylamide gel electrophoresis is used to define and quantify products of deoxyribose oxidation in DNA, based on the unique electrophoretic mobility of DNA fragments possessing deoxyribose oxidation products on their termini. This approach allows initial estimation of the chemistry. Once the chemical identity of damage products has been confirmed, this technique allows sensitive quantitation of the various damage products.

Cu(II)/H2O2-induced DNA damage is enhanced by packaging of DNA as a nucleosome.

Copper is a physiologically important, redox-active metal that may be involved in endogenous DNA damage and mutagenesis. To understand the factors that affect the location and quantity of copper-induced oxidative DNA damage in cells, we used the 5S rDNA nucleosome as a model to assess the effect of chromatin structure on DNA damage produced by Cu(II)/H2O2. Packaging of DNA into a nucleosome increased the extent of Cu(II)/H2O2-induced strand breaks by a factor of 2, while the extent of base lesions sensitive to Fpg and endo III glycosylases increased 8-fold. We also observed that Cu(II)/H2O2 caused slightly more strand breaks than base lesions in isolated 5S rDNA (ratio of base lesions to strand breaks of approximately 0.6), while base lesions outnumbered strand breaks by a factor of 3-4 when the DNA was incorporated into a nucleosome. Apart from several sites of enhanced or diminished DNA damage, there were no major changes in the sequence selectivity of Cu(II)/H2O2, and there was no apparent footprinting effect associated with nucleosome structure, such as that observed with the Fe(II)-EDTA complex. Possible mechanisms for explaining these observations include (1) an increase in Cu(II) concentration in the vicinity of nucleosomal DNA caused by binding of Cu to histone proteins or (2) increased reactivity or accessibility of nucleobases caused by DNA conformational changes associated with nucleosome structure. The enhancement of Cu(II)/H2O2-induced DNA damage in nucleosomes stands in contrast to the protective effect afforded DNA by proteins in chromatin against radiation-induced DNA damage.

Sequence-specific double-strand breakage of DNA by neocarzinostatin involves different chemical mechanisms within a staggered cleavage site.

Direct double-strand breaks in DNA have been implicated in cellular lethality of the antitumor antibiotic neocarzinostatin, but the mechanism of their formation has not been elucidated. Evidence is presented that neocarzinostatin causes sequence-specific direct double-strand breaks whose formation is strongly influenced by the activating thiol. Seven-fold more double-strand breaks result when glutathione rather than 2-mercaptoethanol is used to activate the drug to its putative diradical form, while the sequence specificity of cleavage remains the same. These data explain earlier inconsistencies in the ratios of double-strand to single-strand breaks obtained from in vitro and in vivo studies. Double-strand cleavage sites, occurring predominantly at GT steps, especially AGT.ACT, consist of trinucleotide sequences with a two-nucleotide 3'-stagger of the cleaved residues. The chemical structures of the cleavage sites suggest a model in which a neocarzinostatin-induced double-strand break results from abstraction of a C5' hydrogen atom from the T of ACT and the C4' hydrogen atom of the T of AGT by a single molecule of the diradical form of the drug. Single-strand breaks at these sites occur as separate events with attack at the C5' hydrogens. These findings permit the generalization that single-strand breaks produced by neocarzinostatin show a base preference but no clear sequence specificity, while bistranded lesions are sequence-specific in nature.

DNA bending is a determinant of calicheamicin target recognition.

Calicheamicin is a hydrophobic enediyne antibiotic that binds noncovalently to DNA and causes sequence-selective oxidation of deoxyribose. While the drug makes several base contacts along the minor groove, the diversity of binding-site sequences and the effects of DNA conformation on calicheamicin-induced DNA cleavage suggest that sequence recognition per se is not the primary determinant of target selection. We now present evidence that calicheamicin bends its DNA targets. Using a gel mobility assay, we observed that polymers of oligonucleotide constructs containing AGGA and ACAA binding sites for calicheamicin did not possess intrinsic curvature. Binding of calicheamicin epsilon, the aromatized form of the parent calicheamicin gamma(1)(I), to oligonucleotide constructs containing binding sites in phase with the helical repeat caused a shift to smaller circle sizes in T4 ligase-mediated circle formation assays, with a much smaller shift observed with constructs containing out-of-phase binding sites. It was also observed that binding of calicheamicin epsilon to a 273 bp construct with phased binding sites caused an increase in the molar cyclization factor, J, from 8 x 10(-8) to 9 x 10(-6) M. These results are consistent with DNA bending as part of an induced-fit mechanism of DNA target recognition and with the hypothesis that the preferred targets of calicheamicin, the 3' ends of oligopurine tracts, are characterized by unique conformational properties.

Formation of the 1,N2-glyoxal adduct of deoxyguanosine by phosphoglycolaldehyde, a product of 3'-deoxyribose oxidation in DNA.

Oxidation of deoxyribose in DNA results in the formation of a variety of electrophilic products that have the potential to react with nucleobases to form adducts. We now report that 2-phosphoglycolaldehyde, a model for the 3'-phosphoglycolaldehyde residue generated by 3'-oxidation of deoxyribose in DNA, reacts with dG and DNA to form the diastereomeric 1,N2-glyoxal adducts of dG, 3-(2-deoxy-beta-D-erythro-pentofuransyl)-6,7-dihydro-6,7-dihydroxyimidazo[1,2-a]purine-9(3H)-one. The glyoxal adducts were the predominant species formed under biological conditions (pH 7.4 and 37 degrees C), with several minor fluorescent adducts, including 1,N6-ethenoadenine. The adducts were fully characterized by HPLC, mass spectrometry, and UV and NMR spectroscopy. The reaction of 2-phosphoglycolaldehyde with dG occurred with a rate constant of 10(-6) M(-1) s(-1) compared to the rate constants of 0.08 and approximately 10(-9) M(-1) s(-1) for the reactions of glyoxal and glycolaldehyde with dG, respectively. The kinetic results rule out contamination of 2-phosphoglycolaldehyde preparations with glyoxal as the basis for our observations. The rate constant for the formation of glyoxal from 2-phosphoglycolaldehyde (10(-8) s(-1)) is consistent with glyoxal generation being the rate-limiting step in the formation of dG adducts in reactions with 2-phosphoglycolaldehyde. Mechanistic studies were also undertaken to define the basis for the different oxidation states of glyoxal and 2-phosphoglycolaldehyde. Although 2-phosphoglycolaldehyde produced a weak ESR signal consistent with generation of hydroxyl radicals and it caused DNA strand breaks at high concentrations, the formation of the glyoxal adducts of dG was insensitive to radical quenchers (e.g., sorbitol) and independent of molecular oxygen. In contrast, the formation of glyoxal-dG adducts with glycolaldehyde was dependent on molecular oxygen and quenched by sorbitol, and the glycolaldehyde-glyoxal rearrangement produced a strong ESR signal characteristic of alkyl radicals. These observations are consistent with a model in which glyoxal is generated from 2-phosphoglycolaldehyde by a nonradical, oxygen-independent mechanism that is currently under investigation. Our results provide a mechanistic basis for the observation by Murata-Kamiya et al. [(1995) Carcinogenesis 16, 2251-2253] that oxidation of DNA with the Fe(II)-EDTA complex results in the formation of the glyoxal adducts of dG.

Neocarzinostatin-mediated DNA damage in a model AGT.ACT site: mechanistic studies of thiol-sensitive partitioning of C4' DNA damage products.

Double-strand (DS) DNA damage caused by neocarzinostatin (NCS) has been studied in the trinucleotide AGT-ACT sequence in an AP-1 transcription factor binding site. There are strong similarities between bistranded lesions produced at AGT.ACT and AGC-GCT, including the fact that DS lesions outnumber SS lesions on the AGT and AGC strands, while SS exceed DS on the ACT and GCT strands. Structure-function studies revealed that a variety of different thiols produced bistranded lesions in this model by predominantly C4'-hydrogen atom abstraction (84-93%) at the T of AGT and C5'-hydrogen atom abstraction (87-91%) at the T of ACT. Single-strand (SS) lesions were found to represent a variable mixture of C4' and C5' chemistry. The C4'-hydroxylated abasic site occurred in both SS and DS lesions at both sites and accounted for most of the DS damage at AGT (60-83%); the remaining damage consisted of 3'-phosphoglycolate- and 3'-phosphate-ended fragments. The nature of the thiol was found to affect the partitioning of the breakdown products arising from C4' and, to a lesser extent, C5' hydrogen atom abstraction. Production of 3'-phosphoglycolate residues, restricted mainly to the T of AGT in bistranded lesions, correlated with the incidence of direct DS breaks in the AGT.ACT model and in plasmid DNA and appeared to be influenced by the reducing power of the thiol activator. Furthermore, hydrazine and sodium borohydride both inhibited the formation of glycolate, an effect that was exploited to determine the rate constant for 3'-phosphoglycolate formation: 0.06 min-1 at 0 degree C, pH 7.4. Under anaerobic conditions, the nitroaromatic radiation sensitizer misonidazole caused a large increase in glycolate production in both SS and DS lesions formed by NCS, which suggests that the formation of 3'-phosphoglycolate, like 3'-formylphosphate generated by C5' chemistry, involves an oxyradical intermediate. The pathways for DNA damage involving C4' and C5' hydrogen atom abstraction thus share many common features, several of which are consistent with a mechanism for the production of NCS-mediated bistranded lesions at AGT.ACT that involves a tetraoxide bridge joining the lesions on opposite strands of DNA.

Enediyne-mediated DNA damage in nuclei is modulated at the level of the nucleosome.

DNA damage in HeLa nuclei and isolated nucleosome core particles has been examined for several members of the enediyne family of antitumor antibiotics: calicheamicin gamma 1I (CAL), esperamicin A1 (ESP A1), esperamicin C (ESP C), and neocarzinostatin (NCS). In nuclei, both NCS and ESP A1 produced DNA damage limited to the linker region of the nucleosome, while CAL and ESP C, an analog of ESP A1 missing the deoxyfucose-anthranilate moiety, damaged both the core and linker DNA. DNA fragments produced by CAL and ESP C in the nucleosome core occurred with a 10-11-nucleotide periodicity similar to that produced by DNase I, while damage produced by NCS and ESP A1 appeared to be limited to the terminal portions of the core DNA. The damage in nuclei is shown to be caused directly by the drugs with little contribution from endogenous factors, such as nucleases and topoisomerases. Features of drug structure that may limit damage to the nucleosome core include the presence of substituents on both sides of the CAL/ESP-type core, and the presence of an intercalating moiety, such as the naphthoate of NCS and possibly the anthranilate of ESP A1.