METHYLATION AND BREAKDOWN OF MICROSOMAL AND SOLUBLE RIBONUCLEIC ACID FROM RAT LIVER BY DIAZOMETHANE

N-(2-chloroethyl)-N-nitrosoureas covalently bound to nonionic and monocationic lexitropsin dipeptides. Synthesis, DNA affinity binding characteristics, and reactions with 32P-end-labeled DNA

The synthesis and characterization of a series of compounds that contain an N-alkyl-N-nitrosourea functionality linked to DNA minor groove binding bi- and tripeptides (lexitropsins or information-reading peptides) based on methylpyrrole-2-carboxamide subunits are described. The lexitropsins (lex) synthesized have either a 3-(dimethylamino)propyl or propyl substituent on the carboxyl terminus. The preferred DNA affinity binding sequences of these compounds were footprinted in 32P-end-labeled restriction fragments with methidiumpropyl-EDTA.Fe(II), and in common with other structural analogues, e.g., distamycin and netropsin, these nitrosoureas recognize A-T-rich runs. The affinity binding of the compound with the dimethylamino terminus, which is ionized at near-neutral pH, appeared stronger than that observed for the neutral dipeptide. The sequence specificity for DNA alkylation by (2-chloroethyl)nitrosourea-lex dipeptides (Cl-ENU-lex), with neutral and charged carboxyl termini, using 32P-end-labeled restriction fragments, was determined by the conversion of the adducted sites into single-strand breaks by sequential heating at neutral pH and exposure to base. The DNA cleavage sites were visualized by polyacrylamide gel electrophoresis and autoradiography. The alkylation of DNA by Cl-ENU-lex was compared to that by N-(2-chloroethyl)-N'-cyclohexyl-N-nitrosourea (CCNU), which has no DNA affinity binding properties. While all the Cl-ENU compounds generate DNA breaks as a consequence of the formation of N7-alkyl-guanine, the Cl-ENU-lex compounds induced, in a time- and dose-dependent fashion, intense DNA cleavage bands at adenine, cytosine, and thymine residues associated with affinity binding sites. These non-G cleavages induced by Cl-ENU-lex were inhibited by the coaddition of distamycin at concentrations that did not affect G alkylation break sites. CCNU, even at much higher concentrations, does not generate any similar detectable lesions at non-G sites. Therefore, linking the Cl-ENU moiety to minor groove binders is a viable strategy to qualitatively and quantitatively control the delivery and release of the ultimate DNA alkylating agent in a sequence-dependent fashion.

Studies on the stability of trialkyl phosphates and di-(2'deoxythymidine) phosphotriesters in alkaline and neutral solution. A model study for hydrolysis of phosphotriesters in DNA and on the influence of a beta hydroxyethyl ester group

Various trialkyl phosphates were investigated as model compounds for DNA-phosphotriesters for their stability in neutral or alkaline conditions. The results show that phosphotriesters were highly stable even at strongly alkaline pH, with the exception of diethyl 2-hydroxyethyl phosphate (DHP). The extreme instability of the latter was found to be due to the 2-hydroxy function. In accordance with earlier interpretations the 2-hydroxyethyl group is proposed to participate in the formation of a highly reactive dioxaphospholane ring intermediate which decays rapidly by hydrolysis. Alkylation of 3'- and 5'-deoxythymidine monophosphates with methyl- or hydroxyethylnitrosourea (MNU, HENU) results in practically exclusive phosphate alkylation. In analogy with the model phosphotriesters, di(2'-deoxythymidine) phosphotriesters generated after reaction with MNU or HENU showed extreme dependence of their stabilities on the nature of the alkyl group transferred to phosphate. Whereas the methyl phosphotriester was highly stable, the corresponding hydroxyethyl analogue showed half lives of decay of less than 1 min (pH 12.5), 27 min (pH 9.1) and 60 min (pH 7). Thus the introduction of a 2-hydroxyethyl function into phosphate strongly decreases the stability of the phosphate link of DNA, resulting in DNA single strand breaks, in analogy to RNA phosphotriesters which have been found earlier to be highly unstable because of the presence of the ribose 2'-OH-group.

1,7-Diethylguanosine formation in tRNA chemical ethylation by ethionine and ethylnitrosourea

The mechanism of ethionine carcinogenesis and more generally the relationship between alkylation of nucleic acids by chemical carcinogens and oncogenesis still remain obscure. In the present study the rat liver tRNA ethylation by L-[ethyl-1-3H]ethionine was reinvestigated by examining in particular the highly radioactive 'pyrimidine-nucleotide-like' fraction found earlier in acid hydrolysates of hepatic tRNA from ethionine-treated rats. The following results were obtained: (1) ultraviolet-spectral and chromatographic analyses showed the presence of 1,7-diethylguanosine in this 'pyrimidine-nucleotide-like' fraction; (2) the dialkyl compound was recovered exclusively in the form of imidazole-ring-opened derivatives. When [1-14C]ethylnitrosourea was used as alkylating agent, the in vivo ethylation pattern of tRNA from various organs of rat showed an analogous radioactive 'pyrimidine-nucleotide-like' fraction as main radioactive product. On the contrary, tRNA ethylation pattern after in vitro reaction with [1-14C]ethylnitrosourea exhibited a main radioactivity peak (85% of the total radioactivity recovered) in coincidence of the chromatographic area of 1,7-diethylguanine. The 1,7-diethylguanosine moieties of tRNA were extremely labile both under physiological and alkaline conditions. The 1,7-diethylguanine-associated radioactivity was completely lost from [14C]ethyl-tRNA after only 7 h incubation at 37 degrees C and pH 7.3, while at pH 11.4 this process was preceded by the conversion of the 1,7-diethylguanosine residues into imidazole-ring-opened derivatives.

Origin and formation of 1,7-dimethylguanosine in tRNA chemical and enzymatic methylation

tRNA chemical methylation: 1. 1,7-Dimethylguanosine was found in in vivo methylated tRNA from liver and kidney of rat after exposure to a low dose of dimethylnitrosamine (4 mg/kg body weight). 2. At 4 h after dimethylnitrosamine administration, the 1,7-dimethylguanosine: 7-methylguanine ratio (product ratio) for liver and kidney tRNA was 0.017 and 0.091, respectively. At 24 h after dimethylnitrosamine administration, the product ratio was lower in both hepatic and renal tRNA. 3. When dimethylnitrosamine was given in four separate daily injections, the product ratio in hepatic tRNA 4 h after the last dose was the same as for the same total dose given by a single injection, but in renal tRNA it was lower. No dialkyl compound was found in liver and kidney tRNA 24 h after the last multiple injection. tRNA enzymatic methylation: 1. Base analyses of Escherichia coli B tRNA methylated in vitro, by using S-adenosylmethionine as physiological methyl donor and enzyme preparations from liver and kidney of normal rat, indicated that 1,7-dimethylguanosine was also a product of enzymatic methylation. 2. The amount of 1,7-dimethylguanosine formed by kidney enzyme preparation was 3-times that produced by the liver extract. 3. A second type of enzymatic methylation assay where chemically methylated tRNA was used as substrate indicated that the 7-methylguanosine residues in the nucleic acid are not the substrate of the methylase activity forming the 1,7-dimethylguanosine moieties. Analogous data were obtained for the origin of 1,7-dimethylguanosine residues in tRNA chemical methylation by dimethyl sulphate.

tRNA chemical methylation. In vitro and in vivo formation of 1,7-dimethylguanosine at high concentrations of methylating agents

The methylation patterns produced in Escherichia coli B tRNA by a range of concentrations of the weak carcinogen dimethyl sulphate were examined with the following results: 1. 1,7-Dimethylguanosine was found to be formed in high amounts in the tRNA methylation reaction at high concentrations of methylating agent. 2. The dialkylated compound was recovered mainly in the form of derivatives, the spectral and chromatographic behaviour of which varied according to the procedures used for their isolation. Similar results were obtained for the in vivo methylation of rat-liver tRNA: after administration of a very high dose of the powerful carcinogen dimethylnitrosamine, 1,7-dimethylguanosine was found in rat-liver tRNA. Moreover, the analysis of the time-course of nucleic acid methylation indicated that this dialkylated product was still present in rat-liver tRNA when the major product of alkylation, 7-methylguanine, had almost completely disappeared.

Mechanism of benzo[a]pyrene diol epoxide induced deoxyribonucleic acid strand scission

Approximately 1% of (+/-)-7 beta, 8 alpha-dihydroxy-9 alpha, 10 alpha-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BaP-diol epoxide) DNA alkylation sites rearrange with strand scission at neutral pH. Phosphotriester hydrolysis and depurination/depyrimidination strand scission were critically examined as possible mechanisms for this phenomenon. The catalysis of nicking by alkali and the inhibition of nicking by counterions were consistent with either mechanism. The kinetics of nicking, however, were characteristic of a multistep reaction such as depurination/depyrimidination strand scission and the detection of apurinic sites in BaP-diol epoxide alkylated DNA strongly supported this mechanism. The number of such sites, especially at lower reaction levels, was probably sufficient to account for strand scission. No direct evidence was obtained for nicking occurring through phosphotriester hydrolysis. Studies with model substrates, including dibutyl phosphate, DNA homopolymers, and TMV RNA, indicated that if BaP-diol epoxide forms phosphotriesters in DNA or RNA, they do not hydrolyze with strand scission. Besides apurinic/apyrimidinic sites, a second alkali-sensitive rearrangement product was present in BaP-diol epoxide modified DNA. These latter sites accumulated with time and after 24 h accounted for as much as 4% of the initial alkylation events. Although relatively stable at neutrality, they spontaneously nicked the DNA backbone at high pH. It is possible that these sites represent a rearrangement of the major N2 guanine adduct.

Influence of hydrogen bonding in DNA and polynucleotides on reaction of nitrogens and oxygens toward ethylnitrosourea

The reactivity of ethylnitrosourea toward hydrogen-bonded sites in double-stranded DNA or oly(rA).poly(rU) was compared with those sites in single-stranded DNA, RNA, or poly(rA). Alkylation of the N-1 of A in poly(rA).poly(rU) was almost suppressed at 5 degrees C but could be markedly increased by raining the reaction temperature to 25 degrees C, well below the Tm of 56 degrees C. In contrast, the N-7 and N-6 of A, which are not hydrogen bonded, reacted to the same extent at temperatures ranging from 5 to 65 degrees C. The extent of reaction at the N-3 of A varied inversely with the reactivity of the N-1 of A, indicating that of these two nitrogens the N-1 of A is the most reactive. The proportion of reaction at the various nitrogens in poly(rA) was not affected by temperature. Hydrogen-bonded oxygens in double-stranded DNA are the O-6 of G, the O-4 of T, and the O2 of C. All are equally reactive at 5, 25, and 51 degrees C. It is concluded that the observed temperature independence is due to these oxygens having an electron pair not involved in hydrogen bonding and, thus, available for reaction. In contrast, the electron pair of the N-1 of A (or the N-3 of C) is involved in hydrogen bonding, and the extent of their reactivity is dependent on thermal fluctuation providing transiently open base pairs at temperatures far below the Tm.

Acute biochemical and morphological effects of N-nitrosomorpholine in comparison to dimethyl- and diethylnitrosamine

Some biochemical and ultrastructural changes induced in the livers of rats treated with N-nitrosomorpholine are described and compared with parallel observations in rats given dimethyl- or diethylnitrosamine. Hepatotoxic doses of the nitrosamines caused inhibition of incorporation of [14C]leucine into hepatic proteins, accompanied by progressive disaggregation of polysomes which paralleled the known time course of metabolism of each compound. Dimethylnitrosamine (DMN) and N-nitrosomorpholine (NM) inhibited incorporation of [14C]orotate into liver RNA but diethylnitrosamine (DEN) caused a slight stimulation of orotate incorporation. Electron microscopy revealed similar hepatic cytoplasmic changed induced by each nitrosamine, including dilation and degranulation of the rough surfaced endoplasmic reticulum and subsequent increase of the smooth endoplasmic reticulum. Nuclear changes differed with each compound, N-nitrosomorpholine having more marked effects than either dialkyl compound. The results are discussed with particular reference to the metabolism of N-nitrosomorpholine in the liver.

Inhibition of protein synthesis by N-methyl-N-nitrosourea in vivo

1. The intraperitoneal injection of N-methyl-N-nitrosourea (100mg/kg) caused a partial inhibition of protein synthesis in several organs of the rat, the maximum effect occurring after 2-3h. 2. In the liver the inhibition of protein synthesis was paralleled by a marked disaggregation of polyribosomes and an increase in ribosome monomers and ribosomal subunits. No significant breakdown of polyribosomes was found in adult rat brains although N-methyl-N-nitrosourea inhibited cerebral and hepatic protein synthesis to a similar extent. In weanling rats N-methyl-N-nitrosourea caused a shift in the cerebral polyribosome profile similar to but less marked than that in rat liver. 3. Reaction of polyribosomal RNA with N-[(14)C]methyl-N-nitrosourea in vitro did not lead to a disaggregation of polyribosomes although the amounts of 7-methylguanine produced were up to twenty times higher than those found after administration of sublethal doses in vivo. 4. It was concluded that changes in the polyribosome profile induced by N-methyl-N-nitrosourea may reflect the mechanism of inhibition of protein synthesis rather than being a direct consequence of the methylation of polyribosomal mRNA.

Methylation of ribonucleic acid by the carcinogens dimethyl sulphate, N-methyl-N-nitrosourea and N-methyl-N'-nitro-N-nitrosoguanidine. Comparisons of chemical analyses at the nucleoside and base levels

1. The following methods for hydrolysis of methyl-(14)C-labelled RNA, and for chromatographic isolation and determination of the products, were investigated: enzymic digestion to nucleosides at pH6 or 8; alkaline hydrolysis and conversion into nucleosides; hydrolysis by acid to pyrimidine nucleotides and purine bases, or completely to bases; chromatography on Dowex 50 (NH(4) (+) form) at pH6 or 8.9, or on Dowex 50 (H(+) form), or on Sephadex G-10. 2. The suitability of the various methods for determination of methylation products was assessed. The principal product, 7-methylguanosine, was unstable under the conditions used for determinations of nucleosides. 3- and 7-Methyladenine and 3- and 7-methylguanine are best determined as bases; 1-methyladenine and 3-methylcytosine can be isolated as either nucleosides or bases; O(6)-methylguanine is unstable under the acid hydrolysis conditions used and can be determined as the nucleoside; 3-methyluracil was detected, but may be derived from methylation of the ionized form of uracil. 3. Differences between the patterns of methylation of RNA and homopolyribonucleotides by the N-methyl-N-nitroso compounds and dimethyl sulphate were found: the nitroso compounds were able to methylate O-6 of guanine, were relatively more reactive at N-7 of adenine and probably at N-3 of guanine, but less reactive at N-1 of adenine, N-3 of cytosine and probably at N-3 of uridine. They probably reacted more with the ribose-phosphate chain, but no products from this were identified. 4. The possible influences of these differences on biological action of the methylating agents is discussed. Nitroso compounds may differ principally in their ability to induce miscoding in the Watson-Crick sense by reaction at O-6 of guanine. Both types of agent may induce miscoding to a lesser extent through methylation at N-3 of guanine; both can methylate N atoms, presumably preventing Watson-Crick hydrogen-bonding. N-Methyl-N-nitrosourea can degrade RNA, possibly through phosphotriester formation, but this mechanism is not proven.

The action of mono- and di-functional sulphur mustards on the ribonucleic acid-containing bacteriophage mu2

Bacteriophage mu2 is inactivated by both mono- and di-functional sulphur mustards at relatively low extents of alkylation. No degradation of alkylated RNA was detected. Cross-linking of RNA to protein was observed with the difunctional agent, but this reaction was only a minor contribution to the inactivation. Analyses of the reaction products in bacteriophage RNA showed that, at the mean lethal doses, more than one mono-alkylation of guanine had occurred but the sum total of other types of RNA alkylation was close to a single event. The results therefore suggest that inactivation results from the mono-alkylation of adenine or cytosine. In experiments with the difunctional agent cross-linking of RNA bases or of RNA to protein also prevented replication, the existence of these reactions accounting for the greater sensitivity of the bacteriophage to this agent.

Occurrence of 7-methylguanine in nucleic acids of rat liver

1. Microsomal and soluble RNA of rat liver have been studied by column and paper chromatography after administration of [Me-(14)C]methionine; evidence was obtained for the occurrence of 7-methylguanine, the methyl group being derived from methionine. 2. No evidence was obtained for the occurrence of 7-methylguanine in DNA.

A possible mechanism of inhibition of protein synthesis by dimethylnitrosamine

1. The incorporation of [(14)C]leucine into liver proteins of rats was measured in vivo at various times after treatment of the animals with dimethylnitrosamine and was correlated with the state of the liver ribosomal aggregates. Inhibition of incorporation ran parallel with breakdown of the aggregates. 2. Inhibition of leucine incorporation into protein and breakdown of ribosomal aggregates were not preceded by inhibition of incorporation of [(14)C]orotate into nuclear RNA of the liver. 3. Evidence was obtained of methylation of nuclear RNA in the livers of rats treated with [(14)C]dimethylnitrosamine. 4. Zonal centrifugation analysis of radioactive, nuclear, ribosomal and transfer RNA from livers of rats treated with [(14)C]dimethylnitrosamine revealed labelling of all centrifugal fractions to about the same extent. 5. It is suggested that methylation of messenger RNA might occur in the livers of dimethylnitrosamine-treated rats and the possible relation of this to inhibition of hepatic protein synthesis is discussed.

Alterations in microsomal electron transport, oxidative N-demethylation and azo-dye cleavage in carbon tetrachloride and dimethylnitrosamine-induced liver injury

The effect of administration of carbon tetrachloride and dimethylnitrosamine in vivo on hepatic microsomal function related to drug metabolism was measured. It was found that the capacity of isolated microsomes to demethylate dimethylaniline was diminished during the first hour after carbon tetrachloride poisoning and during the second hour after dimethylnitrosamine poisoning. Thereafter the microsomes from carbon tetrachloride-poisoned livers showed a continuous decline in activity so that at 24hr. there was little residual capacity to undertake demethylation. Microsomes from dimethylnitrosamine-poisoned animals were not different from controls at 24hr. During the first 3hr. there was a transient rise in the accumulation of the N-oxide intermediate in carbon tetrachloride-poisoned livers, with a subsequent fall to below control values. In dimethylnitrosamine poisoning there was a parallel decrease in N-oxide accumulation with decreased demethylation. In the latter part of the first 24hr. the ratio of N-oxide accumulation to demethylation was increased in both instances. At 2hr. after poisoning with either compound there was no evidence of altered NADPH(2)-dependent neotetrazolium reduction or lipid peroxidation. NADPH(2)-dependent azo-dye cleavage was decreased. There was no difference in microsomal cytochrome b(5) content, but there was a decrease in the amount of cytochrome P-450. This latter change was correlated with the decreased capacity for NADPH(2)-dependent oxidative demethylation. It is suggested that dimethylnitrosamine is associated with a defect in microsomal NADPH(2)-dependent electron transport at the level of cytochrome P-450. In addition to affecting cytochrome P-450, carbon tetrachloride is associated with a second severe block involving the release of formaldehyde from the N-oxide intermediate.