Search

Blood leukocyte DNA obtained from 11 Hodgkin's disease patients undergoing ABVD chemotherapy was analysed for the presence of the precarcinogenic adduct O6-methylguanine (O6-meG) at various times (1-2 h up to 49 h) after i.v. treatment with the methylating drug dacarbazine. Adduct formation was detected in all but one of the patients examined at levels ranging up to 0.45 fmol/micrograms DNA (7.2 x 10(-7) mol/mol guanine). The levels of the adduct decreased by approximately 30% over the 24 h following exposure and were usually not detectable 49 h after exposure. In five out of seven individuals examined after more than one treatment, consistent methylation responses were noted, while in the remaining two cases the responses were mixed. No correlation between the extent of adduct formation and lymphocyte levels of the repair enzyme O6-alkylguanine-DNA alkyltransferase was observed. The average extent of O6-meG formation 1 h after dacarbazine treatment was (4.3 +/- 3.1) x 10(-2) fmol/micrograms DNA per mg/kg dose [( 1.2 +/- 0.8) x 10(-3) fmol/micrograms DNA per mg/m2 dose)]. Following exposure of rats to similar doses of dacarbazine, the corresponding levels of adduct in blood leukocyte DNA were 1.1 x 10(-2) fmol/micrograms DNA per mg/kg dose (2.6 x 10(-3) fmol/micrograms DNA per mg/m2 dose).

Two assays suitable for monitoring human populations are presented--one for O6-alkylguanine-type adducts and one for the corresponding repair activity. The assay for adducts is based on competition for repair by O6-alkylguanine-DNA-alkyltransferase of lesions in the unknown DNA and in an oligonucleotide containing a single residue of O6-methylguanine (O6-meGua). The assay can reliably detect as little as 0.5 fmol O6-meGua or 3 fmol O6-ethylguanine in 10-15 micrograms DNA, and it has been used to measure O6-meGua (up to approximately 5 x 10(-7) mol/mol guanine) in lymphocyte DNA from individuals receiving procarbazine or dacarbazine. The assay for repair activity utilizes the same oligonucleotide as substrate for the enzyme, in combination with immunoprecipitation for convenient separation of repaired from unrepaired substrate. Examination of repair activity in biopsy specimens of gastric mucosa has revealed a correlation with the corresponding activity in lymphocytes (r = 0.7; p less than 0.01), indicating that lymphocytes could be used as surrogate markers for repair activity in gastric mucosa.

O6-Methylguanine was measured by a competitive repair assay in blood leukocyte DNA of seven patients with Hodgkin's or non-Hodgkin's lymphoma during therapeutic exposure to procarbazine involving three daily p.o. doses (50 mg each) for 10 days (corresponding to 2.1 mg/kg/day for a 70-kg human). Adduct accumulation was observed in all seven cases, reaching levels up to 0.28 fmol/microgram of DNA (0.45 mumol/mol of guanine). In one individual, maximal levels of adduct were reached after 7 days of exposure, followed by a steady decline, whereas in all other individuals continuous accumulation was observed throughout the exposure period. In four individuals for which data were available for Day 11 (12 to 16 h after the final intake of procarbazine), decreased amounts of O6-methylguanine were observed relative to the last previous measurements. The accumulation of O6-methylguanine was linearly correlated (P less than 0.01) with the cumulative dose of procarbazine, with a slope of 0.011 fmol of O6-methylguanine/microgram of DNA per mg/kg of body weight or 2.68 x 10(-4) fmol of O6 methylguanine DNA per mg/m2. (Two h after the administration of single p.o. doses of 1 to 10 mg/kg of procarbazine to rats, O6-methylguanine formation in leukocyte DNA was just under half that in liver DNA and showed a linear relationship with dose with a slope of 0.017 fmol/microgram of DNA per mg/kg of body weight or 5.67 x 10(-4) fmol of O6-methylguanine/microgram of DNA per mg/m2. A negative correlation (P less than 0.05) between the rate of accumulation of O6-methylguanine in different individuals and lymphocyte O6-alkylguanine-DNA alkyltransferase (AGT) was observed, demonstrating a probable protective effect of AGT against the accumulation of O6-methylguanine during exposure to methylating agents. This observation supports the suggestion of a possible role of procarbazine-induced O6-methylguanine in the pathogenesis of acute nonlymphocytic leukemia appearing after treatment with chemotherapeutic protocols which include procarbazine, based on the finding of low lymphocyte AGT levels in patients with such therapy-related neoplastic disease (Sagher et al., Cancer Res., 48: 3084-3089, 1988). Lymphocyte AGT levels were mainly in the range of 5 to 10 fmol/micrograms of DNA and showed no consistent variation during procarbazine exposure.

DNA extracted from biopsies of normal or atrophic gastric mucosa obtained from 20 individuals was analysed for the presence of the precarcinogenic alkylation lesion O6-methylguanine by a recently developed, highly sensitive assay based on repair by the Escherichia coli O6-alkylguanine-DNA alkyltransferase (AGT) enzyme in competition with a radiolabelled oligonucleotide containing O6-methylguanine (O6-meG). With a limit of detection of 0.5 fmol O6-meG in 10 micrograms DNA, only one DNA sample (derived from a region of the stomach with advanced chronic atrophic gastritis) was found marginally positive, containing 0.52 fmol/10 micrograms DNA (8.3 X 10(-8) mol O6-meG/mol guanine). Measurements of AGT in 49 biopsies of normal, atrophic, hyperplastic or dysplastic mucosa obtained from the gastric antrum or corpus of 18 individuals did not reveal any significant effects of mucosal histology on AGT. The average AGT value found was 6.9 +/- 3.5 (SD) fmol/micrograms DNA, which is lower than the values reported for a number of other human tissues (liver, small intestine and lung). Measurement of AGT levels in gastric mucosa and circulating lymphocytes of the same individuals revealed a positive correlation (P less than 0.005), suggesting that lymphocytes may serve as a useful surrogate marker for AGT activity in gastric mucosa in studies of the epidemiology of this important repair enzyme.

Organic extracts of chlorinated drinking water collected from the water distribution system of Athens, Greece, were tested in parallel for their ability to cause mutations in the Salmonella His+ reversion system and sister chromatid exchanges (SCE) and chromosomal aberrations in CHO cells. They were found to induce His+ revertants in frequencies similar to those detected previously with drinking water extracts from other areas around the world. We observe that organic samples from surface drinking water induce SCE and chromatid type abnormalities in CHO cell cultures while samples collected from ground drinking water did not. These findings confirm that chlorinated surface drinking water contains substances which are mutagenic and clastogenic and may be carcinogenic in humans.

Phosphate acetyltransferase (EC 2.3.1.8) is shown to be unusually sensitive to thiol-blocking reagents, and one or more thiol groups probably catalyse the surface reaction. A novel method for assessing protection by adsorbed species is described. The adsorbed substrate molecules acetyl-CoA and phosphate are found to provide substantial protection against thiol-blocking reagents.

Human gastric juice previously treated with nitrite was examined for its ability to cause O6-alkylguanine-type modifications to 2'-deoxyguanosine or DNA in vitro. Analysis by radioimmunoassay indicated that, in five out of ten cases, incubation with 5 mM 2'-deoxyguanosine resulted in the formation of 375-1350 fmol/ml O6-ethyl-2'-deoxyguanosine (O6-etdGuo) or, in one case, 110 pmol/ml O6-methyl-2'-deoxyguanosine O6-medGuo). When gastric juice-treated calf-thymus DNA was examined for its ability to consume (through suicide repair of O6-alkylguanine-type damage) O6-alkylguanine-DNA alkyltransferase (AAT) from rat liver, eight out of eight samples could not. However, in four out of eight cases, a reduction in the rate of removal of O6-[3H]methylguanine from a 3H-methylated DNA substrate was observed. This finding is compatible with the presence, in gastric juice-treated DNA, of damage capable of binding to, but not undergoing repair by, the AAT.

Fasting human gastric juice was treated in vitro, at pH 2-7 and 37 degrees C for 2 h, with 5-100 microM sodium nitrite. Under these conditions (which simulated those occurring in vivo in normal or hypochlorhydric individuals), the formation of total N-nitroso compounds had the following characteristics: (i) it increased greatly at pH less than 3; (ii) it showed first-order dependence on nitrite concentration; (iii) it was faster at pH 7 than at pH 5. These observations are compatible with the N-nitroso compounds formed by the interaction of nitrite with gastric juice being N-nitrosamides or related compounds. Furthermore, based on the results of this study, it is suggested that in order for hypochlorhydria to give rise to increased formation of N-nitroso compounds in the stomach, it would be necessary for it to be accompanied by a greater than 5- to 10-fold increase in gastric nitrite concentration relative to that found in the normal population, a condition which is not necessarily fulfilled in all hypochlorhydric individuals or populations. The implications of this conclusion for the assessment of the role on gastric N-nitroso compounds in the etiology of gastric cancer are discussed.

The concentrations of nitrate, nitrite, N-nitroso compounds and bacteria were measured in 96 samples of fasting gastric juice, pH 0.90-8.50, obtained from 56 individuals just before or at various times (8 days - 1 year) after gastric operation. The mean pH of the post-operative samples [4.66 +/- 0.39 (SEM)] was significantly higher than that of the pre-operative ones [3.29 +/- 0.33 (SEM)]. A positive correlation with pH was observed for the concentrations of total and nitrate-reducing bacteria (median values 5.0 X 10(5) organisms/ml and 9.2 X 10(4) organisms/ml, respectively, for samples with pH greater than or equal to 1.2 X 10(3) organisms/ml and 0 organisms/ml, respectively, for samples with pH less than or equal to 2.5) and nitrite [mean values 22.5 +/- 3.1 (SEM) microM and 3.20 +/- 0.5 (SEM) microM for samples with pH greater than or equal to 6.5 and pH less than or equal to 2.5, respectively]. No correlation with pH was seen for the concentrations of nitrate [mean value 0.48 +/- 0.06 (SEM) mM] or N-nitroso compounds [mean value 0.30 +/- 0.06 (SEM) microM]. The concentrations of bacteria and nitrite, although increased in hypochlorhydric individuals, were lower than those reported for corresponding individuals in other, primarily British, studies. It is suggested that the relatively low concentrations of nitrite observed in our hypochlorhydric population may account for the absence of elevated concentrations of N-nitroso compounds and that the latter phenomenon may be related to the relatively low frequency of gastric cancer in Greece.

Following i.p. injection of dimethylnitrosamine into male C57BL mice, synthesis of liver nuclear heterogeneous RNA was inhibited significantly, reaching approximately 20% of control within 2 hr of a dose of 40 mg/kg. Synthesis of nucleolar RNA was also inhibited, although to a smaller extent, reaching about 70% of control after the same treatment. These effects were observed both during RNA synthesis in vivo and during in vitro transcription with isolated nuclei and nucleoli. Examination of RNA polymerases I and II, isolated and partially purified by diethylaminoethyl Sephadex column chromatography, did not indicate any change either in their activities in the transcription of exogenous DNA or in their in vivo binding to chromatin. On the other hand, the activity of purified chromatin as a template for transcription by added, partially purified RNA polymerase II was significantly reduced, suggesting that carcinogen-induced damage to chromatin was the cause of the observed inhibition of heterogeneous RNA synthesis. When purified DNA was used in place of chromatin as a template for transcription by partially purified RNA polymerase II, no inhibition was observed. Dimethylnitrosamine treatment had a pronounced effect on the kinetics of appearance of the cytoplasmic RNA species. Four hr after a 40-mg/kg dose of dimethylnitrosamine, the rate of appearance in the cytoplasm of polyadenylate-containing RNA was inhibited by 50%, while that of 4S, 18S, and 28S ribosomal RNA was inhibited by over 80%.

The extent of formation and the rate of repair of O6-methylguanine in rat liver nucleolar DNA following administration of a single dose of dimethylnitrosamine have been studied by means of radioimmunoassay for O6-methyldeoxyguanosine. No difference has been found in either the extent of formation or the rate of repair of this lesion between nucleolar and whole nuclear DNA.

A sensitive radioimmunoassay for O6-methyldeoxyguanosine has been developed, permitting the analysis of microgram amounts of DNA. This technique has been used in the study of the formation and removal of O6-methylguanine in mouse liver satellite and main-band DNA. The results indicate a reduced extent of O6-methylguanine formation in satellite DNA but similar rates of its removal from both classes of DNA.

A novel assay for O6-alkylguanine-type adducts in DNA is reported. It is based on the use of the suicide repair enzyme O6-alkylguanine-DNA-alkyltransferase (AGT) to repair such adducts in DNA in competition with an oligonucleotide containing a single residue of O6-methylguanine, end labeled to high specific activity. The stoichiometric mode of action of AGT results in decreased amounts of oligonucleotide being repaired in the presence of increasing levels of adducts in the competing DNA. The extent of oligonucleotide repair is determined by immunoprecipitation of the unrepaired form with rabbit antiserum directed against O6-alkyldeoxyguanosine and radiocounting. The amount of O6-alkylguanine in competing DNA is calculated by reference to a standard curve constructed using DNA of known alkylation. In view of its relatively wide spectrum of alkyl group specificity, use of AGT from rat liver permits the determination of both O6-methyl- and O6-ethylguanine (detection limits, 0.8 fmol and 3 fmol, respectively). On the other hand, the restricted specificity of Escherichia coli AGT to repair of O6-methylguanine makes the assay based on it specific for this type of lesion (detection limit, 0.5 fmol). The maximum amount of DNA which can be included in the assay is 15 micrograms and 10 micrograms for the rat liver and E. coli AGT-based assays, respectively, leading to a limit of sensitivity of 8 x 10(-8) mol O6-methylguanine/mol guanine (50 fmol/mg DNA) (both enzymes) and 3 x 10(-7) mol O6-ethylguanine/mol guanine (200 fmol/mg DNA) (rat liver AGT-based assay) and making this one of the most sensitive assays for these important precarcinogenic adducts. The new assay has been validated by assaying DNA from rat liver methylated in vivo with dimethylnitrosamine to a known extent and has been found to give results in close agreement with those of radioimmunoassay. Six h after i.p. administration of dimethylnitrosamine (0.01-1 mg/kg) to rats, O6-methylguanine was detectable by the competitive-repair assay in liver or lymphocyte DNA at levels of 0.14-14.4 mumol/mol guanine.

The gastric formation of N-nitroso compounds probably constitutes a major source of human exposure to this important class of environmental carcinogens. Following reduction of nitrate to nitrite by oral or gastric bacteria, reaction with nitrogenous constituents of gastric juice can occur leading to the in situ formation of N-nitroso compounds, probably primarily derived from amides, ureas or aromatic amines. While gastric nitrite concentrations are raised in the achlorhydric relative to the normal stomach, the latter, owing to its acidity, offers a particularly favourable environment for the formation of N-nitroso compounds, as indicated by the finding of greatly increased gastric concentrations of N-nitroso compounds following an oral dose of nitrate. This illustrates the importance of the dynamic nature of the relationships between the various parameters involved in the formation of N-nitroso compounds. While in principle the same is true of the process of inhibition of nitrosation by reducing agents such as ascorbic acid (since, depending on the relative concentrations of reducing agent, nitrite and oxygen, inhibition or catalysis of nitrosation can occur), ingestion of 1 g ascorbic acid brings about a significant reduction in the gastric concentration of N-nitroso compounds.

A simplified and highly sensitive assay for the determination of O6-alkylguanine-DNA-alkyltransferase has been developed and validated by the analysis of extracts of human urinary bladder mucosa. The new assay involves the use of a synthetic dodecanucleotide containing a single O6-methylguanine residue as substrate for the enzyme. This substrate is 5'-end-labelled with [35S]PO3 and separation of repaired and unrepaired oligonucleotide is achived by immuno-precipitation with polyclonal antibodies specific for O6-methyldeoxyguanosine. Kinetic analysis of the repair of the oligonucleotide by extracts of Escherichia coli and rat liver showed that the reaction is first-order in substrate and enzyme and gave the molecular rate constants 7.5 x 10(6) mol-1 1-1 sec-1 and 8.0 x 10(6) mol-1 1-1 sec-1, respectively. The rate constants for the repair of the corresponding O6-ethylguanine-containing oligonucleotide were 3.0 x 10(5) mol-1 l-1 sec-1 and 3.6 x 10(6) mol-1 l-1 sec-1, respectively. Analysis of extracts of 48 samples of normal or neoplastic human urinary bladder mucosa obtained by transurethral biopsy or at surgery, by the new method or by a method involving use of [3H]-methylated DNA as substrate and HPLC, indicated excellent agreement between the two methods. The mean AGT content of normal urinary bladder mucosa obtained from individuals without diagnosed bladder cancer was 0.181 +/- 0.081 (mean +/- SD) fmol/microgram protein, that of neoplastic samples 0.323 +/- 0.177 fmol/microgram protein and that of normal tissue obtained from patients with bladder cancer 0.183 +/- 0.068 fmol/microgram protein. The new method is convenient, rapid and extremely sensitive (it can readily measure femtomole quantities of enzyme) and should prove useful for studies of the biochemical epidemiology of DNA repair.

The formation of carcinogenic N-nitrosamines in neutral and alkaline aqueous solutions (pH 6-14) at 25 degrees C is reported using dissolved N2O3 and N2O4 gases. These reactions are very much faster than those with acidified nitrite: typically, 2 X 10(-3) M amine gives ca. 10-50% N-nitrosamine in a few seconds with 5-20 fold excess of nitrogen oxide. The N-nitrosamine yield in 0.1 M sodium hydroxide is independent of amine basicity from pKA 11.2-0.99, but decreases with decreasing pH of the reaction solution for the more basic amines. Significantly, N-nitrosamine yields are not lowered with diluted nitrogen oxides (1000 ppm) and moderately basic amines (eg. N-methylpiperazine) react readily at physiological pH. The mechanism by which these reactions occur is discussed, with particular reference to the existence of two reactive tautomeric forms of N2O3 and N2O4. The formation of carcinogenic N-nitrosamines from NO in ethanol at 25 degrees C is also reported. These reactions are slow in the absence of air (oxygen), I2 or metal salts. Oxygen accelerates nitrosation by converting NO via NO2 to either N2O3 or N2O4, but both I2 and metal salts are effective under anaerobic conditions, where reaction rates are virtually independent of amine basicity but depend on the nature of the added reagent. The most effective substance is I2, which gives quantitative yields of N-nitrosamine in a few minutes at 25 degrees C by forming the reactive nitrosyl iodide (NOI) reagent. Acceleration in ethanol at 25 degrees C is also observed with AgI, CuI, CuII, ZnII, FeIII and CoII salts, among others, with substantial amounts of N-nitrosamine being produced in ca. 30-300 min. Metal iodides intervene by way of the NOI reagent, as for I2, but other salts require a mechanism involving reaction between a metal-amine complex and NO, itself. The results show that carcinogenic N-nitrosamines may form under a much wider range of experimental conditions than suspected hitherto. Their relevance to human exposure is discussed, with particular reference to urban pollution and the effect of dietary antioxidants.

N-Nitrosocimetidine (NC) induces significant numbers of sister chromatid exchanges (SCE) and chromosome aberrations in cultured Chinese hamster ovary (CHO) cells even at a concentration of 1.2 x 10(-7) M. Its effectiveness in SCE induction is about two thirds that of the gastric carcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG). These results constitute further evidence that NC possesses carcinogenic activity.

In view of the possible involvement of N-nitroso compounds in the etiology of gastric carcinoma in man, we have examined the abilities of extracts of human gastric mucosa for the removal from DNA of O6-methylguanine, the primary precarcinogenic lesion induced in DNA by methylating carcinogenic N-nitroso compounds. We find that all 20 specimens examined (including 14 non-neoplastic and 6 neoplastic) possess significant activity, removing on average 0.73 +/- 0.06 (s.e.m.) pmol O6-methylguanine per mg protein in the extract. This activity, which has the characteristics of an O6-methylguanine-DNA-transmethylase, while exhibiting wide interindividual variations, shows no significant differences between the groups of normal and of neoplastic tissues or between the groups of non-neoplastic tissues obtained from individuals with benign or with malignant gastric disease. It is estimated that the average O6-methylguanine-removing capacity of the cells of normal gastric mucosa (calculated as 166 000 +/- 26 000 (s.e.m.) molecules per cell) greatly exceeds the extent of DNA damage likely to be caused by gastric-juice N-nitroso compounds.

Cimetidine was administered by stomach tube to rats at 70 or 700 mg/kg, doses corresponding to 5 or 50 times, respectively, the typical daily dose of individuals on cimetidine treatment. In some cases, cimetidine (70 mg/kg) was administered in combination with a 2-fold molar excess of sodium nitrite. This treatment was carried out up to 6 times over a 3-day period, the pH of the rat stomach being maintained at 2.3 to 3.0 for about 1 hr after each treatment. The DNA of the stomach, liver, and intestines (large and small pooled together) was isolated 6 hr after cessation of treatment and analyzed for the presence of O6-methylguanine using a sensitive and specific radioimmunoassay. No evidence could be obtained for the presence of this methylated base in any of the DNA samples examined, the limit of detection being 3 mumol O6-methylguanine per mol guanine. We suggest that the observed lack of DNA methylation may be primarily due to the slow rate of nitrosation of cimetidine in combination with its rapid absorption into the blood stream.