No abstract

In in vitro assays using methylated DNAs as substrates, human liver fractions were shown to be able to catalyze the removal of 06-methylguanine. The amount of removal was proportional to the amount of protein added, and the loss of O6-methylguanine occurred with stoichiometric formation of guanine in the DNA and S-methylcysteine in protein. This indicates that human liver contains a protein similar to that previously found in bacteria exposed to ailcylating agents. This protein acts as a transmethylase, transferring the intact methyl group from 06-methylguanine in DNA to a cysteine residue on that protein. A similar activity is present in rodent liver, butit was found that human liver was about 10 times more active in carrying out this reaction. In contrast, there was no difference between the human and rat liver extracts in catalyzing the loss of another methylation product, 7-methylguanine, from alkylated DNA. The liver is the organ most likely to be alkylated after exposure to exogenous potential alkylating agents such as dimethylnitrosamine. The present results show that human liver has a significant capacity to repair 06-methylguanine in DNA, which has been implicated as a critical product in carcinogenesis and mutagenesis. Dimethylnitrosamine (Me2NNO) is known to be carcinogenic in many animal species (1). This effect has been associated with the capacity of the target tissues to metabolize Me2NNO into a mutagenic intermediate that reacts with DNA at various sites (2-4). Of the various DNA adducts formed, 06-methylguanine (MeGua) has been implicated in both mutagenesis and carcinogenesis (4-10). The persistence of06-alkylguanine in the DNA of a tissue correlates with the probability that that tissue will develop tumors after administration of alkylating agents, indicating that this adduct is a critical determinant in the initiation of carcinogenesis by nitrosamines such as Me2NNO (4, 9, 10). MeGua can be removed from DNA in Escherichia coli by a DNA repair system which transfers the methyl groups to a cysteine residue in protein (11,12). There is evidence that this system also occurs in rodent tissues, but it has not been fully characterized (4,(13)(14)(15)(16)(17). Humans are known to be exposed to nitrosamines, including Me2NNO (18,19), and Me2NNO is metabolized by human liver to generate the reactive intermediate that results in DNA alkylation (20,21).In the work presented here, we used an in vitro assay to demonstrate that human liver fractions can catalyze the removal of MeGua from DNA; we found that this capacity is about 10 times greater than that with comparable rat liver fractions. We have also shown that this human liver system transfers the methyl group from the 06 position of guanine to eysteine residue, regenerating guanine directly in the DNA. MATERIALS AND METHODS Chemicals. N-[3H]Methyl-N-nitrosourea (1.6 Ci/mmol; I Ci = 3.7 X 1010 becquerels) was obtained from New England Nuclear. All other biochemical reagents were obtained from Sigma.Tissues. The 10 samples of human liver (six m...

The levels of 3 DNA repair enzymes involved in alkylation and oxidative DNA damage repair in human peripheral blood leukocytes were measured in 20 smokers and 17 non-smokers. No differences in O6-alkylguanine-DNA-alkyltransferase (AGT) activity were found between the 2 groups and the AGT distribution within the population appeared to be unimodal. In contrast, the mean activities of both the methylpurine (MeP)- and the 2-6-diamino-4-hydroxy-5N formamidopyrimidine (FaPy)-DNA glycosylases were higher in the smokers, although only the difference between the MeP-DNA glycosylase means was statistically significant. The standard deviations of these 2 enzymes were also higher in the smokers. The MeP-DNA glycosylase activity showed a bimodal distribution when all subjects were considered. This may in part be due to the smoking habit; 83% of the subjects with enzyme activities higher than 500 fmoles/mg protein were current smokers, whilst 85% of the non-smokers had lower enzyme activities. However, if the smokers were considered separately, a bimodal distribution of this enzyme activity could still be observed. No strong correlation was observed between enzyme activity and age, although the slopes of the regression lines of enzyme activity on age were all negative. The relationship between enzyme activities was studied by bivariate distribution and a strong correlation was only found between the MeP-DNA glycosylase and the FaPy-glycosylase, with the highest values of both enzyme activities being observed in the smokers and the lowest in the non-smokers. Our results suggest that the activity of certain DNA repair enzymes can be modulated by environmental exposure.