Binding mode of chemically activated semiquinone free radicals from quinone anticancer agents to DNA

Chemical reduction of the highly active quinone-containing antitumor drugs, adriamycin and daunorubicin formed the same partially reduced free radical previously reported [9] by microsomal activation. In vitro incubation of the chemically activated free radical intermediates with DNA resulted in covalent binding of these drugs to DNA. The adriamycin semiquinone radical has a greater affinity for DNA and covalent complexes up to one adriamycin per 12 nucleotides were obtained. The daunorubicin semiquinone radical, on the other hand, showed a lesser binding affinity and gave rise to complexes in which one drug molecule was covalently bound per 135 nucleotides. The stronger covalent binding of adriamycin to DNA may account for more severe DNA damage induced by this drug.     

Metabolic activation of acetylenes. Covalent binding of [1,2-14C]octyne to protein, DNA and haem in vitro and the protective effects of certain thiol compounds

[1,2-14C]Oct-l-yne was used to investigate metabolic activation of the ethynyl substituent in vitro. Activation of octyne by liver microsomal cytochrome P-450-dependent enzymes gave intermediate(s) that bound covalently to protein, DNA and to haem. The time course and extent of covalent binding of octyne to haem and to protein were similar. However, two different activating mechanisms are probably involved. Whereas covalent binding to protein or to DNA was inhibited by nucleophiles such as N-acetylcysteine, that to haem was little affected. When N-acetylcysteine was included in the reaction mixtures, two major octyne-N-acetylcysteine adducts were isolated and purified by high-pressure liquid chromatography. G.l.c.-mass spectrometry and n.m.r. suggest that these are the cis-trans isomers of S-3-oxo-oct-1-enyl-N-acetylcysteine. Oct-1-yn-3-one reacted non-enzymically with N-acetylcysteine at pH 7.4 and 37 degrees C with a t1/2 of about 6 s also to yield S-3-oxo-oct-l-enyl-N-acetylcysteine. The same product was formed when microsomal fractions were incubated with oct-1-yn-3-ol, N-acetylcysteine and NAD(P)+. Octyn-3-one did not appear to react with haem or protoporphyrin IX. 5. A mechanism for the metabolic activation of oct-1-yne is proposed, consisting in (a) microsomal hydroxylation of the carbon atom alpha to the acetylenic bond and (b) oxidation to yield octyn-3-one as the reactive species.     

Metabolism of benzo(a)pyrene to reactive intermediate(s) via prostaglandin biosynthesis.

We have examined the oxidation of benzo (a) pyrene (BP) to electrophilic metabolites during the formation of prostaglandins (PG) and thromboxanes (TX) from arachidonic acid (AA) by guinea pig lung microsomal protein. In the presence of NADPH or AA, electrophilic metabolites of [¹⁴C]-BP were generated which were non-extractable from microsomal protein and thus assumed to be covalently bound. The total amount of BP metabolized in the presence of NADPH was 2–2 $\text{1}\text{2}$ times the amount of BP metabolized in the presence of AA. Only 4–5% of BP metabolized by the NADPH mediated mixed-function oxidase system was covalently bound, whereas 12–15% of the BP metabolized in the presence of AA was covalently bound to tissue protein and DNA. Quinones were the major metabolites produced by the AA dependent system, while dihydrodiols were the major metabolites formed by the NADPH dependent system. 7, 12-Dimethyl-benzanthracene, and 7,8-BP-dihydrodiol, but not 3 hydroxy-BP were also oxidized by PG synthetase to reactive metabolites.     

Evidence that 1-naphthol is not an obligate intermediate in the covalent binding and the pulmonary bronchiolar necrosis by naphthalene

Recent studies of a number of volatile aromatic hydrocarbons have suggested that the formation of covalently bound metabolites arises solely through the intermediate formation of phenols. This study further examines the involvement of 1-naphthol in the in vivo and in vitro formation of covalently bound metabolites and pulmonary bronchiolar necrosis by naphthalene. Marked differences were observed in the rate of 1-naphthol formation in lung and liver microsomal incubations without correspondingly large differences between the rates of formation of covalently bound metabolites from naphthalene and 1-naphthol. Glutathione decreased covalent binding in hepatic microsomal incubations containing 14[C]1-naphthol but did not result in the formation of any of the glutathione adducts isolated from identical incubations containing 14[C]naphthalene. Tissue levels of covalently bound radioactivity in mice treated with 14[C]1-naphthol or 14[C]naphthalene were similar; however, in contrast to studies with naphthalene, 1-naphthol administration did not deplete tissue glutathione nor result in detectable tissue injury. These studies indicate that 1-naphthol is not an obligate intermediate in the formation of covalently bound metabolites from naphthalene nor does it appear to be a more proximate lung toxic metabolite.     

The relationship between ionization potential and prostaglandin H synthase-catalyzed binding of aromatic hydrocarbons to DNA

A series of polycyclic aromatic hydrocarbons (PAH) with ionization potentials (IP) ranging from 7.7 to 6.96 eV were examined for their ability to bind to DNA in vitro via the prostaglandin H synthase (PHS)-catalyzed co-oxidation of PAH. A relatively low IP was required for significant binding of PAH to DNA to occur. PAH with IP below approximately 7.35 eV were active in the binding reaction, while those with IP at or above 7.35 eV exhibited much lower levels of binding. Among the PAH which bound at high levels, there was no correlation between the extent of binding and IP. The results suggest that the prostaglandin H synthase-mediated binding of PAH to DNA occurs by one-electron oxidation, similarly to previously reported horseradish peroxidase (HRP)-catalyzed binding of PAH to DNA.     

The microsomal activation of aflatoxin B1 and 2-(N-ethylcarbamoyloxymethyl)furan in vitro using a novel diffusion apparatus

The activation by rat liver microsomal systems in vitro of a naturally occurring and a synthetic furan-containing toxin, aflatoxin B₁ and 2-(N-ethylcarbamoyloxymethyl)furan (CMF) has been examined. Both compounds are metabolised to form products which bind covalently to DNA and microsomal protein, Using a specially designed two-chamber diffusion apparatus it has been demonstrated that the active metabolite of CMF is able to bind covalently to DNA separated by a membrane barrier from the microsomal site of activation. In the case of aflatoxin B₁ the DNA must be in physical contact with the microsomal system for the active metabolite of aflatoxin B₁ to bind covalently. Differences between the activation of the two compounds have also been found with regard to their relative efficiencies in binding to DNA and also the effects of the nucleophile GSH. These results have suggested that if the molecular mechanisms of activation of the two compounds be similar, other factors, for example differences in lipid solubility, may play important roles in determining the relative biological activaties of the compounds. The results suggested that the subcellular site of activation of aflatoxin B₁, unlike that of CMF, may need to be adjacent to the target DNA. It is proposed that this site might be the outer nuclear membrane. Alternatively a carrier molecular might exist for the activated aflatoxin B₁ metabolite in vivo.     

The covalent binding of 3-methylindole metabolites to bovine tissue

Tritiated 3-methylindole (3MI) was administered intravenously to calves. Total and covalent bound radioactivity were measured in different tissues. Pulmonary tissue showed the highest concentration of covalent bound radioactivity. (G-³H) or (methyl-¹⁴C) 3MI became covalently bound to microsomal protein when incubated with bovine lung microsomes. This covalent binding was dependent on time, temperature, oxygen and NADPH and was inhibited by SKF-525A, cytochrome c, a carbon monoxide enriched atmosphere and cysteine. The microsomal enzyme system catalysing covalent binding of 3MI has the classical characteristics of a cytochrome P-450 dependent mixed function oxidase. Metabolic activation of 3MI to a highly electrophilic intermediate, may be fundamental in the pathogenesis of 3MI induced pulmonary damage.     

Inhibition of simian virus 40 DNA replication by specific modification of T-antigen with oxidized ATP

We covalently bound periodate-oxidized ATP (oATP) to purified simian virus 40 (SV40) large T-antigen and determined the effect of this modification on viral DNA replication and three other biochemical activities of T-antigen. The oATP bound specifically to T-antigen, inhibiting the ATPase activity and preventing T-antigen from activating SV40 DNA replication in vitro. In contrast, binding of oATP had no effect on the DNA-binding activity of T-antigen nor on its ability to form a complex with DNA polymerase alpha. These results provide direct biochemical evidence suggesting that the T-antigen ATPase activity is necessary for viral DNA replication.     

Enzymatic repair of O-alkylated thymidine residues in DNA: involvement of a O4-methylthymine-DNA methyltransferase and a O2-methylthymine DNA glycosylase

The distribution and intracellular translocation of AFB1 in various subcellular fractions was investigated in isolated hepatocytes by pulse-chase experiments. After labeling the hepatocytes with [3H]-AFB1 (14.5 nM) for 15 min, the highest concentration of [3H]-AFB1 was found in the cytosolic fraction where 66% was bound noncovalently and 1.5% covalently. The lowest concentration of [3H]-AFB1 was found in the nuclear fraction; 36% and 4.9% were bound noncovalently and covalently respectively. When the [3H]-AFB1 loaded cells were chased with unlabeled AFB1 (1 microM), the radioactivity of [3H]-AFB1 in the cell lysate and cytosolic fraction decreased in time with an apparent rate of elimination (t1/2) of 93 min and 66 min, respectively. The levels of covalently bound AFB1 increased with time and reached a maximum at 60 min in nuclei (270%), and at 120 min in mitochondria (220%) and cytosol (430%) as compared to the zero time. Only in the microsomal fraction was there no significant increase with time in covalently bound AFB1. These results suggest that the toxin after activation by the microsomal mixed function oxidases was either detoxified or transported to other cellular organelles where covalent binding of macromolecules occurred.     

Use of spin-labelled substrate analogs for a comparative study of microsomal cytochrome P-450 and P-448

The previously described, iodine-labeled alkylating stable nitroxyl radicals located at different distances between the N-O. group and the iodine atom were used for a comparative study of the structure of microsomal cytochromes P-450 and P-448 active centers. The radicals were shown to change the optical spectra of Fe3+ located in the active site of the enzyme that are similar to those induced by cytochrome P-450 substrates. Some differences in the type of the radicals binding to control, phenobarbital- and 3-methylcholanthrene-induced microsomes were revealed. The alkylating radical substrate analogs covalently bound to microsomal cytochrome P-450 in the vicinity of the active center, resulting in the inhibition of oxidation of type I and II substrates (e. g., aniline and naphthalene). The value of the spectral binding constant (Ks) for naphthalene in the presence of the radical covalently bound to the cytochrome P-450 active center showed a tendency to increase. Using the ESR technique, the interaction between Fe3+ and the radical localized in the active site of cytochrome P-450 was demonstrated. The contribution of Fe3+ to the relaxation of the radicals covalently bound to cytochrome P-450 was evaluated from the values of the spin label ESR spectra saturation curves at 77K. The distances between the N-O. group of these radicals and Fe3+ in the enzyme active center for the three types of microsomes were determined. The data obtained point to structural peculiarities of the active center of cytochrome P-450, depending on the microsomal type.     

Identification of somatogenic binding sites in liver microsomes from normal mice and transgenic mice expressing human growth hormone gene.

Somatogenic binding sites were detected and characterized in microsomal preparations from livers of normal mice and mice expressing metallothionein-I/hGH (mMT/hGH) hybrid gene, using 125I-labelled bovine or human GH, or a photoreactive derivative of hGH (125I-AP-hGH1). Specific binding of 125I-bGH was detected in liver microsomes from both normal and transgenic mice with an apparent Kd of 2 nM. 125I-hGH was partially displaced by bGH. 125I-AP-hGH1 was covalently bound to the microsomal preparations, and bGH prevented the formation of the 130 kDa species with no appreciable effect on 63 kDa and 70 kDa lactogenic complexes.     

The origin of the DNA induced circular dichroism of CC-1065 and analogs

The calf thymus DNA (CT-DNA) and poly(dI-dC).poly(dI-dC) binding properties of the natural antitumor antibiotic CC-1065 and selected analogs of CC-1065 were studied by circular dichroism (CD) and absorbance methods. The results indicate that the intense long wavelength DNA-induced CD band of these molecules originates from a chiral electronic transition which is delocalized over the whole molecule. Both the covalently bound species (N-3 adenine adduct) and the reversibly bound species exhibit the characteristic spectral behavior of an inherently dissymmetric chromophore when these agents bind within the minor groove of B-form DNA. This mechanism of optical activity accounts for why CC-1065 shows a weak CD in buffer but a very intense induced CD at long wavelength when bound to DNA, why the intensity of the induced CD of CC-1065 analogs depends upon how many fused ring systems the analog contains, and why covalently bound analogs having the mirror image configuration of the natural configuration also exhibit an intense positive induced CD band at long wavelength.     

Radical Cations in Aromatic Hydrocarbon Carcinogenesis

Most carcinogens, including polycyclic aromatic hydrocarbons (PAH), require metabolic activation to produce the ultimate electrophilic species that bind covalently with cellular macromolecules to trigger the cancer process. Metabolic activation of PAH can be understood in terms of two main pathways: one-electron oxidation to yield reactive intermediate radical cations and monooxygenation to produce bay-region diol epoxides. The reason we have postulated that one-electron oxidation plays an important role in the activation of PAH derives from certain common characteristics of the radical cation chemistry of the most potent carcinogenic PAH. Two main features common to these PAH are: 1) a relatively low ionization potential, which allows easy metabolic removal of one electron, and 2) charge localization in the PAH radical cation that renders this intermediate specifically and efficiently reactive toward nucleophiles. Equally important, cytochrome P-450 and mammalian peroxidases catalyze one-electron oxidation. This mechanism plays a role in the binding of PAH to DNA. Chemical, biochemical and biological evidence will be presented supporting the important role of one-electron oxidation in the activation of PAH leading to initiation of cancer.     

Radical Cations in Aromatic Hydrocarbon Carcinogenesis

Most carcinogens, including polycyclic aromatic hydrocarbons (PAH), require metabolic activation to produce the ultimate electrophilic species that bind covalently with cellular macromolecules to trigger the cancer process. Metabolic activation of PAH can be understood in terms of two main pathways: one-electron oxidation to yield reactive intermediate radical cations and monooxygenation to produce bay-region diol epoxides. The reason we have postulated that one-electron oxidation plays an important role in the activation of PAH derives from certain common characteristics of the radical cation chemistry of the most potent carcinogenic PAH. Two main features common to these PAH are: 1) a relatively low ionization potential, which allows easy metabolic removal of one electron, and 2) charge localization in the PAH radical cation that renders this intermediate specifically and efficiently reactive toward nucleophiles. Equally important, cytochrome P-450 and mammalian peroxidases catalyze one-electron oxidation. This mechanism plays a role in the binding of PAH to DNA. Chemical, biochemical and biological evidence will be presented supporting the important role of one-electron oxidation in the activation of PAH leading to initiation of cancer.     

Studies on the covalent binding of an intermediate(s) in prostaglandin biosynthesis to tissue macromolecules.

We investigated the covalent binding of intermediates in prostaglandin biosynthesis to tissue macromolecules. Following incubation of [1-14C]arachidonic acid with the microsomal fraction from guinea pig lung, ram or bovine seminal vesicle, human platelets, rabbit kidney, or rat stomach fundus, the amount of covalent binding of arachidonic acid metabolites expressed as percentage of total arachidonic acid metabolized varied from tissue to tissue ranging from 3% in human platelets to 18.2% in ram seminal vesicles. In general, the thromboxane synthesizing tissues had less covalently bound metabolites than the other tissues. The amount of covalently bound metabolites was increased in the guinea pig lung microsomes when the thromboxane synthetase inhibitor, N-0164, was added to the incubation mixture. The covalent binding of arachidonic acid metabolite(s) was greatly reduced by the addition of glutathione to the incubation mixture. In addition to the covalently bound metabolites, water-soluble metabolites derived from arachidonic acid metabolism were also observed. The amount of water-soluble metabolites was small in each tissue except for the rat stomach fundus. In the rat stomach fundus the water-soluble metabolites accounted for over 50% of the total metabolites. Conditions which would tend to increase or decrease the levels of free prostaglandin endoperoxides during the incubation of arachidonic acid with the microsomes gave increased or decreased levels of covalent binding. Our data suggest that the prostaglandin endoperoxides are responsible for the covalent binding observed during prostaglandin biosynthesis. This covalent binding to tissue macromolecules may be of physiological and pathological significance.     

The binding of radioactive label from labelled phenacetin and related compounds to rat tissues in vivo and to nucleic acids and bovine plasma albumin in vitro

1. acetyl-(3)H- and ethyl-(14)C-labelled derivatives of phenacetin and related compounds are described. 2. Radioactive label from the ethyl-(14)C-labelled derivatives of 4-nitrophenetole, 4-phenetidine and phenacetin binds in vitro to various extents to bovine plasma albumin, salmon sperm DNA and yeast RNA; the extent of binding is increased in the presence of a rat liver microsomal hydroxylating system and further increased when the microsomal enymes are induced by prior treatment of rats with 3-methylcholanthrene. 3. The ratios of the bound radioactive labels in vitro from [ethyl-(14)C]phenacetin, N-acetoxy[ethyl-(14)C]phenacetin, [acetyl-(3)H]phenacetin and [diacetyl-(3)H]N-acetoxyphenacetin per g-atom of DNA P, RNA P and per mol of protein in the absence of the microsomal system are approximately 1:60:11:863, 1:68:41:1835 and 1:88:713:2399 respectively. 4. Radioactive label from labelled phenacetin binds in vitro to all tissues examined, including the spleen, intestines, kidney and bladder; about 80% of the radioactivity bound to the liver is concentrated in the RNA and proteins. 5. Comparison of the relative extents of binding of radioactive label derived from equimolar amounts of labelled phenacetin, ethanol or acetate shows that the incorporation of labelled C(2) units into tissues and biological macromolecules in vivo and in vitro may account for only a part of the total bound radioactive label derived from phenacetin and not at all from the incorporation of radioactive acetate into nucleic acids. 6. Some implications of these findings are discussed.     

Hepatic macromolecular covalent binding of mononitrotoluenes in Fischer-344 rats

The mononitrotoluenes are important industrial chemicals which display isomeric specificity in their ability to induce hepatic DNA excision repair in Fischer-344 rats. Covalent binding of the structurally related hepatocarcinogen, 2,6-dinitrotoluene, to hepatic DNA is markedly decreased by prior administration of the sulfotransferase inhibitors pentachlorophenol (PCP) and 2,6-dichloro-4-nitrophenol (DCNP). The objectives of this study were to determine whether hepatic macromolecular covalent binding of the mononitrotoluene isomers differed and to determine whether covalent binding of the mononitrotoluenes to hepatic DNA in vivo was decreased by inhibitors of sulfotransferase. Male Fischer-344 rats were given a single oral dose of [ring-U-14C]-2-, 3- or 4-nitrotoluene (2-, 3- or 4-NT) and killed at various times thereafter. Livers were removed and analyzed for total and covalently bound radiolabel. Maximal concentrations of total radiolabel were observed between 3 and 12 h after the dose, and there were no large differences among the 3 isomers in peak concentrations achieved. Covalent binding to hepatic macromolecules was maximal 12 h after administration for all three isomers. Thereafter, concentrations of covalently bound 2-NT-derived material were always 2-6 times higher than those of 3- or 4-NT-derived material. When DNA was isolated from livers of rats given the mononitrotoluenes 12 h previously, only 2-NT was observed to covalently bind at concentrations above the limits of detection of the assay. The covalent binding of 2-NT, but not that of 3- or 4-NT, to both total hepatic macromolecules and DNA was markedly decreased by prior administration of either PCP or DCNP. Covalent binding to hepatic DNA was decreased by greater than 96%. The results of this study correlate well with studies which have demonstrated that 2-NT, but not 3- or 4-NT, induces DNA excision repair. Furthermore, they suggest that 2-NT, like the hepatocarcinogen 2,6-dinitrotoluene, requires the action of sulfotransferase for its conversion to a species capable of covalently binding to hepatic DNA.     

Drug residue formation from ronidazole, a 5-nitroimidazole. III. Studies on the mechanism of protein alkylation in vitro

Ronidazole (1-methyl-5-nitroimidazole-2-methanol carbamate) is reductively metabolized by liver microsomal and purified NADPH-cytochrome P-450 reductase preparations to reactive metabolites that covalently bind to tissue proteins. Kinetic experiments and studies employing immobilized cysteine or blocked cysteine thiols have shown that the principal targets of protein alkylation ara cysteine thiols. Furthermore, ronidazole specifically radiolabelled with 14C in the 4,5-ring, N-methyl or 2-methylene positions give rise to equivalent apparent covalent binding suggesting that the imidazole nucleus is retained in the bound residue. In contrast, the carbonyl-14C-labeled ronidazole gives approx. 6--15-fold less apparent covalent binding indicating that the carbamoyl group is lost during the reaction leading to the covalently bound metabolite. The conversion of ronidazole to reactive metabolite(s) is quantitative and reflects the amazing efficiency by which this compound is activated by microsomal enzymes. However, only about 5% of this metabolite can be accounted for as protein-bound products under the conditions employed in these studies. Consequently, approx. 95% of the reactive ronidazole metabolite(s) can react with other constituents in the reaction media such as other thiols or water. Based on these results, a mechanism is proposed for the metabolic activation of ronidazole.     

Protein adsorption and leakage in carrier-enzyme systems

The capability of binding enzymes adsorptively to unmodified and silanized silica and glass as well as modified polystyrene carriers was studied for alpha-amylase, beta-amylase, and alpha-chymotrypsin. In most cases a high percentage of protein was bound very firmly under considerable loss of activity. The leakage of protein from the carriers was studied by measuring the intrinsic protein fluorescence on beta-amylase adsorptively bound to aminopropyl silica, aminomethyl, and hexadecylaminomethyl polystyrene. It was compared with the leakage of beta-amylase covalently bound to the same carriers via glutaraldehyde, trichloro-triazine, or benzoquinone. In the absence and in the presence of substrate, at 25 and at 60 degrees C, the leakage rates of the adsorptively bound enzymes were not higher than in the covalently bound systems. The poorest binding stability was found in benzoquinone-coupled beta-amylase derivatives. It is even reduced at higher temperatures, whereas the temperature did not show any remarkable influence on the leakage of the other derivatives. In adsorptively as well as in all the covalently bound systems, the presence of substrate did not promote the protein leakage.     

Irreversible protein binding of norethisterone (norethindrone) epoxide.

14,15-3H-Norethisterone-4 beta, 5 beta-epoxide, a metabolite of norethisterone, was incubated with several proteins and nucleic acids. After 30 min incubation 0.19 nmol of the epoxide were irreversibly bound per mg albumin which contains free sulfhydryl groups; proteins without SH-groups, such as concanavalin A, gamma-globulin, DNA and RNA, did not irreversibly bind norethisterone epoxide. A superoxide (O2) generating enzyme system comprised of xanthine oxidase and hypoxanthine was capable of catalyzing the irreversible binding of the parent compound, norethisterone, to albumin, indicating that an oxidation product was formed which reacted with the protein. When norethisterone epoxide was incubated for 60 min with hepatic microsomes of rats in absence of NADPH, about 2.0 nmol of the epoxide were irreversibly incorporated per mg microsomal protein. This binding was increased to 5.2 nmol by addition of a NADPH regenerating system. Addition of glutathione and cytosol decreased only the NADPH-dependent protein binding; phenobarbital pretreatment of rats induced this NADPH-dependent binding of norethisterone epoxide to microsomal protein by a factor of 2. In presence of NADPH, binding of the epoxide to microsomal protein depended on substrate concentration used. The results indicate that norethisterone epoxide is able to chemically react with proteins. In addition, hepatic microsomal enzymes convert the epoxide to another metabolite which also can react with proteins.