C8-Guanine modifications: effect on Z-DNA formation and its role in cancer.

Base modifications are known to affect the structure and function of DNA. C8-guanine adducts from various carcinogenic compounds have been shown to be potent Z-DNA inducers. Hence, it has been hypothesized that Z-DNA plays a role in cancer and other genetic diseases. In this comprehensive review, Z-DNA and the effect of prevalent C8-guanine adducts on the B-Z transition are addressed. The discoveries of Z-DNA binding proteins including ADAR1, E3L, DLM1, and PKZ have suggested the relevance of Z-DNA in living systems. In addition, increasing evidence on the Z-DNA connection to gene transcription and inhibition reveals potential biological functions of the left-handed DNA. Finally, C8-guanine adducts that promote Z-DNA formation can be used as a tool to explore the Z-DNA function and its role in carcinogenesis.

In vitro reaction of barbiturates with formaldehyde.

Barbiturates are widely used as sedatives, hypnotics, and antiepileptics, and, when coupled with their narrow therapeutic index, the probability that their use will result in accidental or intentional death is significant. When barbiturates are implicated in a murder or suicide, analysis for their presence is often required. Under certain conditions, barbiturates are quite stable, but conditions found in vivo immediately after death or after embalming may promote barbiturate decomposition. If extensive decomposition occurs, analysis for them may be difficult or impossible. Here, the stability of three representative barbiturates, under conditions that model those likely to prevail in vivo shortly after death and after embalming, have been studied. Solutions of phenobarbital were found to slowly decompose in water over the pH range of approximately 3.5 to 9.5. More rapid decomposition occurred at higher pH, and 2-phenylbutyric acid was the main decomposition product. Formaldehyde (5-20%) accelerated the decomposition rate 3-10-fold such that phenobarbital decomposition could be complete after 30 days. In contrast, pentobarbital decomposed roughly 10 times more slowly and secobarbital did not detectably decompose under any of the conditions studied. Thus, certain barbiturates may partially or completely decompose in vivo after death, especially after embalming, and thus analysis for them may lead to false negatives. However, this work shows that analysis for the parent barbiturate or its predicted decomposition product may provide data that will reduce the likelihood of false negatives.

Stability of benzodiazepines in formaldehyde solutions.

Benzodiazepine-type drugs are used in the treatment of a number of pathologic disorders, but they may be implicated in forensic toxicology cases because of their abuse potential. Occasionally, it becomes necessary to measure drug levels following exposure to formaldehyde (postembalming or after tissue storage) if drug involvement was not previously suspected. Virtually no information exists on the decomposition of benzodiazepines in the presence of formaldehyde (the active ingredient in many embalming fluids), yet formaldehyde is known to be highly reactive, particularly with nitrogen-containing compounds. In order to evaluate the effects of formaldehyde on benzodiazepines, 10 benzodiazepine drugs were exposed to various concentrations of formaldehyde and various pH conditions (to simulate potential postembalming conditions), and the decomposition of each drug was measured by high-performance liquid chromatography over a 30-day period. The decomposition rates of all but one of the benzodiazepines were accelerated (to differing degrees) by formaldehyde as compared to controls, and this decomposition was in several cases both pH and formaldehyde concentration dependent. Thus, forensic examiners must be particularly cautious when attempting to determine benzodiazepine concentrations postembalming because the compound may have reacted with formaldehyde to form other products not inherently obvious analytically. Determination of these reaction products will serve to provide alternate analytes, allowing for establishment of accurate conclusions during forensic analyses.

In vitro reaction of formaldehyde with fenfluramine: conversion to N-methyl fenfluramine.

Embalming is common, and it can create problems for the forensic scientist if a drug has been the cause death and this drug is also reactive toward the embalming fluid. Previous studies have focused on the tricyclic amines nortriptyline and desipramine. In the presence of formaldehyde, a typical component of embalming fluid, either of these two compounds can be rapidly converted to their methylated derivatives amitriptyline and imipramine, respectively. We have begun a larger project designed to determine the reactivity and reactions of a wide range of drugs with formaldehyde. We report here our results from fenfluramine, which, like the tricyclic amines, is reactive towards formaldehyde and is converted into its N-methyl derivative. The rate of conversion is dependent upon pH and formaldehyde concentration. Up to 100% conversion in 24 h was observed. In addition, we have also devised a simplified procedure for monitoring this process that may be useful for others working in this area. Finally, we note that the reactions of fenfluramine studied here and of amines in general with formaldehyde need to be considered when performing postmortem/postembalming forensic analysis.

Ethanol-mediated CYP1A1/2 induction in rat skeletal muscle tissue.

The causes of non-trauma-mediated rhabdomyolysis are not well understood. It has been speculated that ethanol-associated rhabdomyolysis may be attributed to ethanol induction of skeletal muscle cytochrome P450(s), causing drugs such as acetaminophen or cocaine to be metabolized to myotoxic compounds. To examine this possibility, the hypothesis that feeding ethanol induces cytochrome P450 in skeletal muscle was tested. To this end, rats were fed an ethanol-containing diet and skeletal muscle tissue was assessed for induction of CYP2E1 and CYP1A1/2 by immunohistochemical procedures; liver was examined as a positive control tissue. Enzymatic assays and Western blot analyses were also performed on these tissues. In one feeding system, ethanol-containing diets induced CYP1A1/2 in soleus, plantaris, and diaphragm muscles, with immunohistochemical staining predominantly localized to capillaries surrounding myofibers. Antibodies to CYP2E1 did not react with skeletal muscle tissue from animals receiving a control or ethanol-containing diet. However, neither skeletal muscle CYP1A1/2 nor CYP2E1 was induced when ethanol diets were administered by a different feeding system. Ethanol consumption can induce some cytochrome P450 isoforms in skeletal muscle tissue; however, the mechanism of CYP induction is apparently complex and appears to involve factors in addition to ethanol, per se.

Activation of AP-1 through the MAP kinase pathway: a potential mechanism of the carcinogenic effect of arenediazonium ions.

Arenediazonium ions such as those found in the common mushroom Agaricus bisporus have been convincingly demonstrated to be tumorigenic. The specific mechanism of their tumorigenicity remains unclear. It has been shown that arenediazonium ions can be metabolized to aryl radicals, and that reaction of these aryl radicals with DNA produces aryl adducts. These metabolic processes also produce the reactive oxygen species superoxide and hydroxyl radicals which have been implicated in AP-1 activation. To further investigate the mechanism of tumorigenesis by arenediazonium ions, we studied the effect of a representative arenediazonium ion on AP-1 activation and phosphorylation of the signal transduction proteins ERK1, ERK2, JNK, and p38 kinase, both in vitro and in vivo. We also identified the specific radicals produced by spin trapping and ESR analysis. Here, it was found that p-methylbenzenediazonium ion (2a) induced a 16-fold increase in the extent of AP-1 activation at micromolar concentrations, and that this increase coincided with phosphorylation of the signaling kinases ERK1 and -2 and p38 kinase, but not JNK, in JB6 mouse epithelial cells. In vivo studies using AP-1 luciferase reporter-bearing transgenic mice supported the increase in the extent of AP-1 activation in 2a-treated mice over controls, and showed that this effect was different in different tissue types. The antioxidant N-acetylcysteine (NAC), a general antioxidant, showed an inhibitory effect on 2a-mediated AP-1 induction, while aspirin, a hydroxyl radical scavenger, had no effect. Spin trapping studies showed that while NAC suppressed radical formation from 2a, aspirin did not alter radical production from 2a. It appears that 3a, a carbon-centered radical formed from 2a, is responsible for AP-1-induced activation, and therefore, radical species that are not oxygen-centered are also capable of inducing AP-1. These results represent a step toward understanding the mechanism of tumorigenicity of arenediazonium ions.

Reactive oxygen species in the aerobic decomposition of sodium hydroxymethanesulfinate.

Sodium hydroxymethanesulfinate, (HOCH2SO2Na, HMS) is relatively stable in aqueous alkaline environments, but rapidly decomposes in acidic medium to give a variety of products that include sulfur dioxide. A detailed kinetic and mechanistic study of the decomposition of HMS in slightly acidic medium has shown a process that produces dithionite, S2O2-4, which is preceded by an induction period which persists for as long as molecular oxygen is present in the reaction solution. The complete consumption of molecular oxygen is a prerequisite for the formation of S2O2-4. Among some of the intermediates detected in the decomposition of HMS is the sulfite radical, SO-3. Comparisons are made between the decomposition mechanisms of thiourea dioxide (aminoiminomethanesulfinic acid) and HMS.

C8-Arylguanine and C8-aryladenine formation in calf thymus DNA from arenediazonium ions.

Arylhydrazides, arylhydrazines, and N-alkyl-N-arylnitrosamines are metabolized to arenediazonium ions which yield C8-arylpurine adducts in calf thymus and cellular DNA. The mechanism of adduct formation has not been fully elucidated. C8-Arylguanine adducts likely form from direct aryl radical (Ar*) addition to the C8 position of guanine. However, the amounts of C8-aryladenine adducts measured here are inconsistent with direct radical attack at the C8 position of adenine. An intermediate product, an aryltriazene, is likely formed which then decomposes to the C8-aryladenine adduct. We have demonstrated that N1-aryl-N3-purinyltriazene adducts are formed from a variety of para-substituted arenediazonium ions with adenine. Decomposition of the N1-aryl-N3-purinyltriazene, at high pH and elevated temperatures, has been shown to give C8-aryladenine derivatives, and a free radical mechanism for this process has been proposed. Here we show that this process can occur under physiological conditions and that the C8-aryladenine adduct can be quantitated by HPLC. ESR studies, in which DMPO was used as a spin trap, have been used to demonstrate the intermediacy of aryl radicals during the decomposition of the N1-aryl-N3-purinyltriazenes and to demonstrate that this process also occurs in calf thymus (ct) DNA treated with arenediazonium ions. These results suggest the involvement of an aryl radical in the formation of the observed DNA adducts. Finally, we have found that the treatment of ct DNA with arenediazonium ions produces a significant amount of depurination. Both the formation of C8-arylguanine and C8-aryladenine adducts and the generation of apurinic sites may contribute to the genotoxicity of arylhydrazides, arylhydrazines, N-alkyl-N-arylnitrosamines, and arenediazonium ions.

Cobalt-mediated generation of reactive oxygen species and its possible mechanism.

Electron spin resonance spin trapping was utilized to investigate free radical generation from cobalt (Co) mediated reactions using 5,5-dimethyl-1-pyrroline (DMPO) as a spin trap. A mixture of Co with water in the presence of DMPO generated 5,5-dimethylpyrroline-(2)-oxy(1) DMPOX, indicating the production of strong oxidants. Addition of superoxide dismutase (SOD) to the mixture produced hydroxyl radical (.OH). Catalase eliminated the generation of this radical and metal chelators, such as desferoxamine, diethylenetriaminepentaacetic acid or 1,10-phenanthroline, decreased it. Addition of Fe(II) resulted in a several fold increase in the .OH generation. UV and O2 consumption measurements showed that the reaction of Co with water consumed molecular oxygen and generated Co(II). Since reaction of Co(II) with H2O2 did not generate any significant amount of .OH radicals, a Co(I) mediated Fenton-like reaction [Co(I) + H2O2-->Co(II) + .OH + OH-] seems responsible for .OH generation. H2O2 is produced from O2.- via dismutation, O2.- is produced by one-electron reduction of molecular oxygen catalyzed by Co. Chelation of Co(II) by biological chelators, such as glutathione or beta-ananyl-3-methyl-L-histidine alters, its oxidation-reduction potential and makes Co(II) capable of generating .OH via a Co(II)-mediated Fenton-like reaction [Co(II) + H2O2-->Co(III) + .OH + OH-]. Thus, the reaction of Co with water, especially in the presence of biological chelators, glutathione, glycylglycylhistidine and beta-ananyl-3-methyl-L-histidine, is capable of generating a whole spectrum of reactive oxygen species, which may be responsible for Co-induced cell injury.

Aryl radical formation during the metabolism of arylhydrazines by microsomes.

Many arylhydrazines are genotoxins, although the mechanism of their genotoxicity is unknown. Previous studies have shown that arylhydrazines are metabolized to arenediazonium ions, which produce C8-arylguanine adducts in DNA suggesting the intermediacy of an aryl radical. Here we have looked for the formation of aryl radicals from arylhydrazines and microsomes by ESR spin trapping. Only hydroxyl radicals are trapped upon incubation of p-methylphenylhydrazine with rat liver microsomes and 5,5-dimethyl-1-pyrroline N-oxide (DMPO). However, hydroxyl and aryl radicals were trapped upon incubation of p-(methoxymethyl)phenylhydrazine with rat liver microsomes. Evidence for hydroperoxyl radical formation was also obtained. In contrast, when either of these substrates was incubated with microsomes from C5O cells, aryl and hydroxyl radicals were trapped. The ESR signal intensity of the spin-trapped aryl radicals parallels the extent of C8-arylguanine formation in DNA, and therefore, the aryl radical is likely the intermediate responsible for C8-arylguanine adduct formation. Aryl radicals and C8-arylguanine adducts may be related to the genotoxicity of arylhydrazines and related compounds that are oxidatively metabolized to arenediazonium ions, the precursor to aryl radicals, including arylalkyl nitrosamines, arylazo compounds, and triazenes.

8-Arylguanine adducts from arenediazonium ions and DNA.

Arenediazonium ions (ArN2+) are genotoxic though the source of their genotoxicity is unknown. The present studies were undertaken to determine if reductive decomposition of ArN2+ to aryl radicals (Ar) in the presence of calf thymus DNA (ctDNA) or in cells results in the formation of DNA adducts. We found that when arenediazonium ions of the general structure p-X-ArN2+ (X = CH3, CH2OCH3, CH2OH) are allowed to react with ctDNA or incubated with cells under conditions that produce p-X-Ar, DNA adducts are formed with guanine. The structure of the adduct is the C8-substitution product derived from guanine and p-X-Ar. Formation of p-X-Ar was determined by ESR spin-trapping with 5,5-dimethyl-1-pyrroline N-oxide (DMPO). The extent of C8-arylguanine adduction was measured by high performance liquid chromatography (HPLC) analysis of the DNA hydrosylate and comparison with authentic synthetic standards. The C8-arylguanine adducts observed to form may be important in regard to the genotoxicity of ArN2+, though other DNA adducts such as the N6-triazene of adenine or C8-aryladenine adducts can form. Finally, though the formation of C8-arylguanine adducts from arenediazonium ions has been proposed, this is the first report demonstrating their formation in DNA.

Vitamin B6 and cancer: synthesis and occurrence of adenosine-N6-diethylthioether-N-pyridoximine-5'-phosphate, a circulating human tumor marker.

In the course of studies aimed at deciphering the metabolic transformations of [3,4-14C] and [3H]C6-pyridoxine hydrochloride by tumor-bearing rats and tumor cells in culture, biosynthesis of a novel labeled product was observed. Its production began with the onset of tumor growth and increased as cell proliferation increased. Chemical, enzymatic, precursor labeling, and analytical tests on the isolated product indicated this product as adenosine-N6-diethylthioether-N-pyridoximine-5'-phosphate (compound 1). In confirmation, the chemical synthesis and characterization of compound 1 are presented in this study. In addition, blood samples from 28 normal subjects, 28 cancer patients with different malignancies, and 39 patients with a variety of other than cancer ailments were screened for compound 1 on a blind basis using reverse phase ion-paired high-performance liquid chromatography. The results show that the level of the vitamin B6 conjugate in the circulation of control subjects, cancer patients in remission, and patients with other diseases was only minimal. Cancer patients with active disease had 3-4-fold higher levels (P < 0.00001). Our results also confirm previous findings regarding the structure of compound 1 and show its potential value as a circulating human tumor marker that could be successfully used for cancer detection.

Reaction of Cr(VI) with ascorbate and hydrogen peroxide generates hydroxyl radicals and causes DNA damage: role of a Cr(IV)-mediated Fenton-like reaction.

Incubation of Cr(VI) with ascorbate generated Cr(V), Cr(IV) and ascorbate-derived carbon-centered alkyl radicals, as well as formyl radicals. H2O2 caused generation of hydroxyl radicals (OH) and much higher levels of Cr(V), showing that .OH can be generated via a Cr(IV)-mediated Fenton-like reaction (Cr(IV) + H2O2-->Cr(V) + .OH + OH-). 1,10-Phenanthroline and deferoxamine inhibited the formation of both .OH and Cr(V) from the reaction of Cr(VI) with ascorbate in the presence of H2O2. Electrophoretic assays showed that ascorbate-derived free radicals caused DNA double-strand breaks. .OH radicals generated by Cr(V)- and Cr(IV)-mediated Fenton-like reactions also caused DNA double-strand breaks. HPLC measurements showed that .OH radicals generated by Cr(IV) and Cr(V) from H2O2 caused 2'-deoxyguanine hydroxylation to form 8-hydroxy-2'-deoxyguanine.

One-electron reduction of vanadate by ascorbate and related free radical generation at physiological pH.

The one-electron reduction of vanadate (vanadium(V)) by ascorbate and related free radical generation at physiological pH was investigated by ESR and ESR spin trapping. The spin trap used was 5,5-dimethyl-1-pyrroline N-oxide (DMPO). Incubation of vanadium(V) with ascorbate generated significant amounts of vanadium(IV) in phosphate buffer (pH 7.4) but not in sodium cacodylate buffer (pH 7.4) nor in water. The vanadium(IV) yield increased with increasing ascorbate concentration, reaching a maximum at a vanadium(V): ascorbate ratio of 2:1. Addition of formate to the incubation mixture containing vanadium(V), ascorbate, and phosphate generated carboxylate radical (.COO-), indicating the formation of reactive species in the vanadium(V) reduction mechanism. In the presence of H2O2 a mixture of vanadium(V), ascorbate, and phosphate buffer generated hydroxyl radical (.OH) via a Fenton-like reaction (vanadium(IV)+H2O2-->vanadium(V)+.OH+OH-). The .OH yield was favored at relatively low ascorbate concentrations. Omission of phosphate sharply reduced the .OH yield. The vanadium(IV) generated by ascorbate reduction of vanadium(V) in the presence of phosphate was also capable of generating lipid hydroperoxide-derived free radicals from cumene hydroperoxide, a model lipid hydroperoxide. Because of the ubiquitous presence of ascorbate in cellular system at relatively high concentrations, one-electron reduction of vanadium(V) by ascorbate together with phosphate may represent an important vanadium(V) reduction pathway in vivo. The resulting reactive species generated by vanadium(IV) from H2O2 and lipid hydroperoxide via a Fenton-like reaction may play a significant role in the mechanism of vanadium(V)-induced cellular injury.

Chromate-mediated free radical generation from cysteine, penicillamine, hydrogen peroxide, and lipid hydroperoxides.

The Cr(VI)-mediated free radical generation from cysteine, penicillamine, hydrogen peroxide, and model lipid hydroperoxides was investigated utilizing the electron spin resonance (ESR) spin trapping technique. Incubation of Cr(VI) with cysteine (Cys) generated cysteinyl radical. Radical yield depended on the relative concentrations of Cr(VI) and Cys. The radical generation became detectable at a cysteine:Cr(VI) ratio of about 5, reached its highest level at a ratio of 30, and declined thereafter. Cr(VI) or Cys alone did not generate a detectable amount of free radicals. Similar results were obtained with penicillamine. Incubation of Cr(VI), Cys or penicillamine and H2O2 led to hydroxyl (.OH) radical generation, which was verified by quantitative competition experiments utilizing ethanol. The mechanism for .OH radical generation is considered to be a Cr(VI)-mediated Fenton-like reaction. When model lipid hydroperoxides such as t-butyl hydroperoxide and cumene hydroperoxide were used in place of H2O2, hydroperoxide-derived free radicals were produced. Since thiols, such as Cys, exist in cellular systems at relatively high concentrations, Cr(VI)-mediated free radical generation in the presence of thiols may participate in the mechanisms of Cr(VI)-induced toxicity and carcinogenesis.

Base pairing of 8-oxoguanosine and 8-oxo-2'-deoxyguanosine with 2'-deoxyadenosine, 2'-deoxycytosine, 2'-deoxyguanosine, and thymidine.

Base pair formation between O-silyl derivatives of 8-oxoguanosine (8-OxoG) and 8-oxo-2'-deoxyguanosine (8-OxodG) with 2'-deoxyadenosine (dA), 2'-deoxycytosine (dC), 2'-deoxyguanosine (dG), and thymidine (dTHd) have been examined by NMR methods in chloroform. 8-OxoG and 8-OxodG form base pairs with all four nucleosides, suggesting that they can mimic the base pairing properties of any DNA base. In 8-OxodG.dA or 8-OxoG.dA base pairs both bases assume Hoogsteen geometry with respect to both bases. Hoogsteen and Watson-Crick base pairs are formed between 8-OxodG or 8-OxoG and either dThd or dG. Only Watson-Crick geometry is seen for 8-OxodG.dC and 8-OxoG.dC base pairs. In all base pairs the glycosidic bond of 8-OxodG or 8-OxoG is syn and anti for the base pair partner. Additional base pairing studies with the modified nucleosides 1-methyl-8-oxoguanosine (1-Me-8-oxoG), 7-methyl-8-oxoguanosine (7-Me-8-oxoG), and 8-methoxyguanosine (8-OMeG), modified nucleosides structurally related to 8-OxoG and 8-OxodG, support the proposed base pair structures and aided in the determination of base pair geometry. Association constants were determined for 8-OxoG.dN and 8-OxodG.dN (N = A, C, G, Thd) base pairs as well as the normal dTHd-A and dG.dC base pairs. The magnitude of the association constants allows the relative stabilities of the base pairs to be determined: 8-OxodGWC.dC (8-OxoGWC.dC) approximately dGWC.dC > 8-OxodGH.dG (8-OxoGH.dG) > 8-OxodGH.dA (8-OxoGH.dA) > 8-OxodGH.dThd (8-oxoGH.dThd) approximately dThd.dA. The formation of these base pairs is consistent with the known mutagenic nature of 8-oxo-2'-deoxyguanosine and with NMR studies of oligonucleotides containing the 8-oxoG modification.

Deferoxamine inhibition of Cr(V)-mediated radical generation and deoxyguanine hydroxylation: ESR and HPLC evidence.

Electron spin resonance (ESR) and high-performance liquid chromatography (HPLC) techniques were utilized to investigate the effect of deferoxamine on free radical generation in the reaction of Cr(V) with H2O2 and organic hydroperoxides. ESR measurements demonstrated that deferoxamine can efficiently reduce the concentration of the Cr(V) intermediate as formed in the reduction of Cr(VI) by NAD(P)H or a flavoenzyme glutathione reductase/NADH. ESR spin trapping studies showed that deferoxamine also inhibits Cr(V)-mediated .OH radical generation from H2O2, as well as Cr(V)-mediated alkyl and alkoxy radical formation from t-butyl hydroperoxide and cumene hydroperoxide. HPLC measurements showed that .OH radicals generated by the Cr(VI)/flavoenzyme/NAD(P)H enzymatic system react with 2'-deoxyguanine to form 8-hydroxy-2'-deoxyguanine (8-OHdG), a DNA damage marker. Deferoxamine effectly inhibited the formation of 8-OHdG also.

Chemical oxidation and metabolism of N-methyl-N-formylhydrazine. Evidence for diazenium and radical intermediates.

N-Methyl N-formlhydrazine (1), a component of the mushroom Gyromitra esculenta, is a carcinogen. Its mode of action, however, is poorly understood. To determine the intermediates that may form during the metabolism of 1, we examined its oxidative chemistry, identified the products and inferred the intermediates on the basis of these products. The incubation of 1 with rat liver microsomes was also studied and the metabolites determined and quantified. Both the chemical and the microsome-mediated oxidation of 1 yielded formaldehyde and acetaldehyde. The formation of acetaldehyde requires (i) the oxidation of 1 to a diazenium ion (I) or diazene (II) and (ii) fragmentation of I/II to formyl and methyl radicals. It is suggested that these radical intermediates may be important in understanding and elucidating carcinogenesis by 1.

8-Hydroxy-2'-deoxyguanosine formation during the catalytic oxidation of hydrazines in the presence of 2'-deoxyguanosine.

The reaction of 2'-deoxyguanosine (1), substituted hydrazines (Hy) and oxygen, in the presence of Cu(II) as catalyst, yields 8-hydroxy-2'-deoxyguanosine (2). The rate of formation of 2 was found to be dependent upon the oxidation potential of the Hy and on structural factors of the Hy. The conversion of 1 to 2 under these conditions suggested that a Haber-Weiss/Fenton type of process was involved with Hy acting as reductant. However, the dependence of the rate of the conversion of 1 to 2 upon the structure of Hy suggested that the Hy substrates studied may be more directly involved in the process than that of a reducing agent. Additional studies of this reaction system, including the oxygen consumption, radical trapping studies and direct ESR measurements suggest that the conversion of 1 to 2, under the conditions used, involves the intermediacy of complex composed of the metal catalyst, 1, Hy and oxygen. The rate data for the conversion of 1 to 2 appear to be correlated with the carcinogenicity of Hy and therefore 2 may be an important DNA adduct in the carcinogenesis of hydrazines.