Role of aromatic amine acetyltransferases, NAT1 and NAT2, in carcinogen-DNA adduct formation in the human urinary bladder.

The metabolic activation and detoxification pathways associated with the carcinogenic aromatic amines provide an extraordinary model of polymorphisms that can modulate human urinary bladder carcinogenesis. In this study, the metabolic N-acetylation of p-aminobenzoic acid (PABA) to N-acetyl-PABA (NAT1 activity) and of sulfamethazine (SMZ) to N-acetyl-SMZ (NAT2 activity), as well as the O-acetylation of N-hydroxy-4-aminobiphenyl (OAT activity; catalyzed by NAT1 and NAT2), were measured in tissue cytosols prepared from 26 different human bladder samples; then DNA was isolated for determination of NAT1 and NAT2 genotype and for analyses of carcinogen-DNA adducts. Both PABA and OAT activities were detected, with mean activities +/- SD of 2.9 +/- 2.3 nmol/min/mg protein and 1.4 +/- 0.7 pmol bound/mg DNA/min/mg protein, respectively. However, SMZ activities were below the assay limits of detection (< 10 pmol/min/mg protein). The levels of putative carcinogen-DNA adducts were quantified by 32P-postlabeling and averaged 2.34 +/- 2.09 adducts/10(8) deoxyribonucleotide phosphate (dNp). Moreover, the DNA adduct levels in these tissues correlated with their NAT1-dependent PABA activities (r = 0.52; P < 0.01) but not with their OAT activities. Statistical and probit analyses indicated that this NAT1 activity was not normally distributed and appeared bimodal. Applying the NAT1:OAT activity ratios (N:O ratio) allowed arbitrary designation of rapid and slow NAT1 phenotypes, with a cutpoint near the median value. Within each of these subgroups, NAT1 correlated with OAT (P < 0.05); DNA adduct levels were elevated 2-fold in individuals with the rapid NAT1 or NAT1/OAT phenotype. Examination of DNA sequence polymorphisms in the NAT1 gene by PCR have demonstrated that an NAT1 polyadenylation polymorphism is associated with differences in tissue NAT1 enzyme activity; accordingly, NAT1 activity in the bladder of individuals with the heterozygous NAT1*10 allele was 2-fold higher than in subjects homozygous for the putative wild-type NAT1*4 allele. Likewise, DNA adduct levels in the mucosa of the urinary bladder were found to be 2-fold (P < 0.05) higher in individuals with the heterozygous NAT1*10 allele (3.5 +/- 2.1 adducts/10(8) dNp) as compared to NAT1*4 homozygous (1.8 +/- 1.9 adducts/10(8) dNp). Thus, these data provide strong support for the hypothesis that NAT1 activity in the urinary bladder mucosa represents a major bioactivation step that converts urinary N-hydroxy arylamines to reactive N-acetoxy esters that form covalent DNA adducts.(ABSTRACT TRUNCATED AT 400 WORDS)

Polymorphism in the N-acetyltransferase 1 (NAT1) polyadenylation signal: association of NAT1*10 allele with higher N-acetylation activity in bladder and colon tissue.

Exposures to carcinogens present in the diet, in cigarette smoke, or in the environment have been associated with increased risk of bladder and colorectal cancer. The aromatic amines and their metabolites, a class of carcinogen implicated in these exposures, can be N- or O-acetylated by the NAT1 and NAT2 enzymes. Acetylation may result in activation to DNA-reactive metabolites or, in some cases, detoxification. Many studies have focused on genetic variation in NAT2 and its potential as a risk factor in bladder and colorectal cancer; however, NAT1 activity is higher in bladder and colonic mucosa than NAT2, and the NAT1 enzyme also exhibits phenotypic variation among human tissue samples. We hypothesized that specific genetic variants in the polyadenylation signal of the NAT1 gene would alter tissue levels of NAT1 enzyme activity and used a PCR-based method to distinguish polymorphic NAT1 alleles in samples obtained from 45 individuals. When the NAT1 genotype was compared with the NAT1 phenotype in bladder and colon tissue samples (p-aminobenzoic acid activity), we observed a approximately 2-fold higher NAT1 enzyme activity in samples from individuals who inherited a variant polyadenylation signal (NAT1*10 allele). This is the first observation relating a genetic polymorphism in NAT1 to a rapid/slow NAT1 phenotype in humans.

O6-alkylguanine-DNA alkyltransferase activity in schistosomiasis-associated human bladder cancer.

O6-Alkylguanine-DNA-alkyltransferase (ATase) activity was measured in extracts of 55 bladder tissue samples (46 tumour and nine uninvolved mucosal tissue) from Egyptian patients with schistosome-associated bladder carcinoma. Activity varied from 2.0 to 16.2 fmole ATase/microgram DNA (mean +/- S.D.; 5.6 +/- 4.0) or from 28 to 351 fmole ATase/mg (117 +/- 71). ATase levels in schistosome-associated bladder cancer tissues (5.6 +/- 4.0 fmole ATase/microgram DNA) tended to be lower than those observed in normal human bladder mucosal tissue (8.5 +/- 4.4 fmole ATase/microgram DNA). In a previous study (Badawi et al., Carcinogenesis, 1992, 13, 877-881) DNA-alkylation damage (O6-methyldeoxyguanosine) was found in 44/46 of these schistosome-associated bladder cancer samples at levels ranging from 0.012 to 0.485 mumole O6-MedG/mole deoxyguanosine. We now report an inverse correlation between the levels of methylation damage and ATase activity (r = -0.67; P < 0.001). These observations encourage further investigations of the possible role of environmental alkylating agents in the aetiology of early bladder cancer associated with schistosomiasis.

Molecular and genetic events in schistosomiasis-associated human bladder cancer: role of oncogenes and tumor suppressor genes.

Carcinoma of the urinary bladder is the most common malignancy in many tropical and subtropical countries and is mainly due to endemic schistosomal infection. Schistosomiasis-associated bladder cancer defines a characteristic pathology and cellular and molecular biology that differs from urothelial carcinoma of non-schistosomal origin. N-Nitroso compounds are suspected etiologic agents in the process of bladder cancer induction during schistosomiasis. Elevated levels of DNA alkylation damage have been detected in schistosome-infected bladders and are accompanied by an inefficient capacity of DNA repair mechanisms. Consequently, high frequency of G --> A transition mutations were observed in the H-ras gene and at the CpG sequences of the p53 tumor suppressor gene. Genetic changes have also been detected in the c-erbB-1 and c-erbB-2 oncogenes and in the cdkn2 and Rb tumor suppressor genes. The potential application of these mutational patterns in providing a biological marker suitable for the biomonitoring and early detection of this neoplasm could indicate new avenues of approach that might alleviate the problem in the future. It can also assist in elucidating the mechanisms by which schistosomiasis augments human bladder cancers.

Involvement of alkylating agents in schistosome-associated bladder cancer: the possible basic mechanisms of induction.

Carcinoma of the urinary bladder is a common malignancy in many tropical and subtropical countries. There is a well documented sequela of chronic urinary schistosomal infection and bladder cancer associated with schistosomiasis is a major cause of morbidity and mortality in the endemic areas. Experimental bladder cancer can be induced in schistosome-infected animals. Multiple factors have been suggested as causative agents in schistosome-associated bladder carcinogenesis and the N-nitroso compounds appear to be of particular importance. These agents have long been suspected to play a major role in the aetiology of a variety of human cancers. A model for the induction of bladder cancer associated with schistosomiasis is proposed which takes into account the interrelationships between different factors resulting from the infection, especially the role of alkylating agents that can contribute to the induction of this neoplasm.

Formation of promutagenic methylation damage in tissue-DNA of mice treated with antischistosomal agents.

The existence of the promutagenic methylation damage O6-MedG has been measured at various time intervals in different tissue DNAs of mice received a single therapeutic dose of various antischistosomal agents (hycanthone, oxaminiquine and metrifonate). Liver-DNA exhibited the highest levels of O6-MedG in all treated animals while, spleen DNA contained the lowest. The three antischistosomal agents tested seemed to exert the peak concentrations of their alkylating metabolites over a period of several hours following the administration. In mice which had received hycanthone, liver-DNA contained readily detectable amounts of O6-MedG by 6 h post-treatment (0.089 mol O6-MedG/mol dG) and by the end of 48 h, this was decreased by about 3-fold to reach a level of 0.026 mumol/mol dG. In intestinal-DNA, however, O6-MedG was formed more slowly and contained about half the level of that found in the liver-DNA. In the tissue-DNA of animals which had received oxaminiquine, the highest level of O6-MedG was observed at 6 h after administration and at a 24-h time point, the adduct dramatically decreased in the liver and intestine-DNA to undetectable values. In neither tissues was there any evidence for O6-MedG accumulation in the DNA at the end of a 48-h post-treatment. A pattern of O6-MedG, almost similar to that of oxaminiquine, was also observed in tissue-DNA of mice pretreated with metrifonate. These results demonstrate that treatment with antischistosomal agents leads to the formation of highly promutagenic alkylated lesions in the tissue-DNA. The implication of such existence for antischistosomal-induced toxicity and carcinogenicity are discussed.

Salivary nitrate, nitrite and nitrate reductase activity in relation to risk of oral cancer in Egypt.

It has been suggested that nitrate and nitrite may play a role in the etiology of human oral cancer. We investigated whether salivary nitrate and nitrite and the activity of nitrate reductase (NRase) may affect the risk of oral cancer in Egypt, an area with high levels of environmental nitrosating agents. Levels of salivary nitrite (8.3 +/- 1.0 micrograms/ml) and nitrate (44 +/- 3.7 micrograms/ml) and activity of NRase (74 +/- 10 nmol/ml/min) were significantly (P < 0.05) higher in oral cancer patients (n = 42) compared to control Egyptian healthy individuals (n = 40, nitrite = 5.3 +/- 0.3 micrograms/ml, nitrate = 27 +/- 1.2 micrograms/ml, and NRase activity = 46 +/- 4 nmol/ml/min). The adjusted odds ratio (OR) and the 95% confidence intervals (C.I.) for risk of oral cancer, categorized by the levels of salivary nitrate and nitrite and NRase activity, showed a higher cancer risk associated with nitrite > 7.5 micrograms/ml (OR: 3.0, C.I.: 1.0-9.3), nitrite > 40 micrograms/ml (OR: 4.3, C.I.: 1.4-13.3) and NRase activity > 50 nmol/ml/min (OR: 2.9, C.I.: 1.1-7.4). Our findings suggest that increased consumption of dietary nitrate and nitrite is associated with elevated levels of salivary nitrite. Together with the increased activity of salivary NRase, these observations may explain, at least in part, the role of nitrate and nitrite in the development of oral cancer in individuals from an area with a high burden of N-nitroso precursors.

Role of human cytochrome P450 1A1, 1A2, 1B1, and 3A4 in the 2-, 4-, and 16alpha-hydroxylation of 17beta-estradiol.

The steady-state kinetics and specific activity of 2-, 4-, and 16alpha-hydroxylation of 17beta-estradiol (E(2)) were evaluated for human cytochrome P450 (CYP) 1A1, 1A2, 1B1, and 3A4 enzymes, using complementary DNA-expressed CYP isoforms. CYP1A2 showed the highest 2-hydroxylation activity, followed by CYP1A1, 1B1, and 3A4. CYP1B1 had the highest 4-hydroxylation activity, followed by CYP1A2, 1A1, and 3A4. The 16alpha-hydroxylation reaction was catalyzed mainly by CYP1A2 and, to a similar, slightly lower extent, CYP3A4 and 1A1, with a lesser contribution by CYP1B1. The E(2) 2-, 4-, and 16alpha-hydroxylation activities of human liver microsomes were 1.3 +/- 0.3, 0.5 +/- 0.06, and 0.3 +/- 0.05 nmol metabolite/min/nmol P450, respectively. The contribution of CYP1A1 and 1B1 (mainly extrahepatic) to the E(2) hydroxylation reactions, relative to CYP1A2 and 3A4 (predominantly hepatic), may be relevant to understanding the process of hormonal carcinogenesis both in liver and in extrahepatic tissues.

Role of schistosomiasis in human bladder cancer: evidence of association, aetiological factors, and basic mechanisms of carcinogenesis.

Several epidemiological, clinical and experimental studies have been carried out to determine whether there is an aetiological role for schistosomiasis in the multi-stage process of bladder carcinogenesis. Lines of evidence supporting the association between bladder cancer and schistosomiasis include indications from the geographical correlation between the two conditions, the distinctive patterns of gender and age at diagnosis, the clinicopathological identity of schistosome-associated bladder cancer and the extensive experimental evidence in infected laboratory animals. Although the causative role of schistosomiasis is now accepted, various associated factors have been proposed in the induction of this particular type of cancer. While all may contribute to the carcinogenic process taking place in the infected bladder, none of these has yet been confirmed. Most attention has been directed at theories proposing possible roles for urinary chemical carcinogens, particularly tryptophan metabolites, N-nitroso compounds and of beta-glucuronidase, as factors that are primarily involved in the initiation of bladder carcinogenesis in areas endemic for schistosomiasis.

Effect of hormonal status on the expression of the cyclooxygenase 1 and 2 genes and prostaglandin synthesis in rat mammary glands.

Hormonal effects on mammary carcinogenesis have been linked to prostaglandin (PG) synthesis. The purpose of the present study was to examine the expression of the cyclooxygenase (COX) 1 and 2 genes and levels of PG synthesis in the mammary glands of rats that have different levels of susceptibility to mammary gland carcinogenesis associated with pregnancy, lactation, post-lactation involution, and ovariectomy. The expression of COX-1 mRNA, measured by Northern blot analysis, was similar in virgin, lactating, pregnant, and post-lactational animals of the same age. Ovariectomized animals exhibited significantly lower levels of COX-1 mRNA (approximately 40%) compared to the sham-operated controls or the ovariectomized animals treated with estradiol and progesterone. COX-2 mRNA, measured by RT-PCR, was detectable only in the mammary glands of lactating animals and ovariectomized animals administered estradiol and progesterone. Induction on COX-2 expression occurred in both stromal and epithelial cells in lactating rat mammary glands. COX enzymatic activities, determined by measuring the conversion rate of [1-14 C]-arachadonic acid to prostanoids, showed that lactating animals had a significantly higher activity compared to virgin (approximately 40%), pregnant (approximately 30%), or postlactational animals (approximately 40%). Ovariectomized animals had significantly lower COX enzymatic activity compared to the sham operated animals. Significant induction of COX activity, however, was observed in ovariectomized animals administered estradiol and progesterone. These changes in COX enzymatic activity were paralleled by similar changes in the mammary gland PGE2 content, measured by enzyme immunoassay. Our results suggest that the effect of hormones on the genesis of mammary cancer in the rat may be mediated, at least in part, by their effects on COX-2 expression and PG synthesis.

Modifications in the carcinogen-metabolizing capacity of mouse liver treated with N-nitroso compounds.

One tenth of the LD50 as a single dose of various N-nitroso compounds (N-nitrosodimethylamine; NDMA, N-nitrosodiethylamine; NDEA, N-nitrosoethylpropylamine; NEPA, N-nitrosodipropylamine; NDPA, N-nitrosomethylethylamine; NMEA, N-nitroso-methylbutylamine; NMBA and N-nitrosoethylbutylamine; NEBA) was administrated into male mice. This dose markedly increased the hepatic contents of cytochrome P450 and cytochrome b5 and activities of NADPH-cytochrome c reductase and aryl hydrocarbon hydroxylase (AHH). The highest increase in the activity of cytochrome P450 (+142% relative to the control value) was shown in animals treated with either N-nitrosoethylpropylamine or N-nitrosodiethylamine. On the other hand, the lowest increase in the activity (+16%) was revealed in animals treated with N-nitrosodimethylamine (not significant compared to the control value). Cytochrome b5 content was increased by 190% of the control value in mice treated with N-nitrosomethylbutylamine, while N-nitrosodibutylamine induced the lowest increase (+20%). The maximum increase (+182%) in the activity of aryl hydrocarbon hydroxylase was shown in animals which received N-nitrosomethylbutylamine, while the lowest increase (+23%) in animals which received N-nitrosodiethylamine. The activity of hepatic AHH was also increased above the control value in animals treated with NDMA, NEBA NDPA, NMEA and NDBA by 138, 98, 90, 89 and 69%, respectively. Identically, NADPH-cytochrome c reductase activity was increased in animals which received NEPA, NMBA, NDMA, NMEA, NDPA, NEBA and NDEA by 202, 150, 110, 95, 94, 77 and 37%, respectively.

Risk factors for hepatocellular carcinoma in Egypt: the role of hepatitis-B viral infection and schistosomiasis.

Hepatitis-B viral (HBV) infection and schistosomiasis are among the most common causes of liver cancer (hepatocellular carcinoma; HCC) in Egypt. The present study investigates the effects of both infectious diseases and other demographical and environmental factors on the risk of HCC among a representative group of Egyptian patients with HCC (n = 102) and controls with no signs of hepatopathology (n = 96). Factors associated with an increased risk of HCC in Egypt were age over 60 yrs-old, farming, cigarette smoking and occupational exposure to chemicals such as pesticides. However, schistosomiasis (relative risk, RR: 5.22; 95% confidence intervals, C.I.: 2.93-9.31) and HBV infection (RR: 12.51; 95% C.I.: 6.11-25.59) were the major risk factors in the development of HCC. Schistosomiasis increased the severity of HBV infection and elevated the risk of HCC over that associated with the HBV infection alone. Understanding these relationships may enable us to determine the susceptibility to HCC among high risk groups and to provide these individuals with effective measures for early prevention or intervention.

O6-methylguanine and O6-methylguanine-DNA [corrected] methyltransferase activity in tissues of BDF-1 mice treated with antiparasitic drugs.

Levels of the DNA promutagenic methylation damage, O6-methylguanine (O6-MeG) and the activity of the O6-methylguanine-DNA methyltransferase (MGMT), the enzyme responsible for repairing O6-MeG, were measured at various time intervals in tissues of BDF-I mice administered a single therapeutic dose of the antischistosomal agents hycanthone, oxaminiquine and metrifonate. Hycanthone increased O6-MeG in the liver-DNA after 6 h, then decreased by 3-fold after 48 h. Lower levels of the adduct and a slower rate of formation were found in the intestine and bladder. MGMT activities were significantly lower in the liver (74%) and bladder (25%) compared to control animals after 6 h, then restored by 48 h. Oxaminiquine increased O6-MeG in all tissues, but spleen, after 6 h and persisted only in the bladder after 48 h. Liver and bladder tissues of these animals exhibited a pattern of alteration in the MGMT activity similar to that observed for hycanthone. Metrifonate induced a profile of O6-MeG comparable to that of oxaminiquine but the levels of the adduct were about 2-fold lower. Hepatic MGMT in these animals was significantly lower (approximately 38%) than the control values after 6 h, then restored by 48 h. A significant negative correlation was obtained between O6-MeG and MGMT activity in the liver (r=- 0.85), intestine (r=- 0.62) and bladder (r=- 0.59). These results demonstrate that treatment with antischistosomal agents may lead to the formation of promutagenic alkylation damage in the tissue DNA and alterations in the DNA repair capacity.

Genetic susceptibility and carcinogen-DNA adduct formation in human urinary bladder carcinogenesis.

Differences in human urinary bladder cancer susceptibility have often been attributed to genetic polymorphisms in carcinogen-metabolizing enzymes, especially those involved in the biotransformation of aromatic amines (AAs) and polycyclic aromatic hydrocarbons (PAHs). Metabolic activation generally involves an initial cytochrome P450-dependent oxidation to form N-hydroxy, phenol, or dihydrodiol intermediates that undergo further conjugation or oxidation to form DNA adducts. The acetyltransferases, NAT1 and NAT2, can participate in these pathways by catalyzing detoxification (by AA N-acetylation) or further activation (by N-OH-AA O-acetylation) reactions. NAT2 polymorphisms, which are due to point mutations in the structural gene, have long been associated with higher risk for bladder cancer. In collaborative studies, we now have found that NAT1 is also expressed polymorphically in human bladder due to mutations in the NAT1 polyadenylation signal, which has recently been associated with increased bladder cancer risk. Moreover, we have found that the bladder NAT1*10 genotype and phenotype are correlated with significantly higher levels of putative AA-DNA adducts in human bladder as measured by 32P-postlabelling. Preliminary data have also suggested that putative PAH-DNA adducts in human bladder are correlated with a polymorphism in the total metabolism of benzo[a]pyrene (BP) by bladder microsomes and especially with the formation of BP-7,8-diol. Since each of these correlations was observed without adjusting for carcinogen intake, it would appear that, with ubiquitous human exposure to AAs and PAHs, the expression of carcinogen-metabolizing enzymes may be a more critical determinant of carcinogen-DNA adduct formation and of individual cancer susceptibility.

Host determinants of DNA alkylation and DNA repair activity in human colorectal tissue: O(6)-methylguanine levels are associated with GSTT1 genotype and O(6)-alkylguanine-DNA alkyltransferase activity with CYP2D6 genotype.

There is increasing evidence that alkylating agent exposure may increase large bowel cancer risk and factors which either alter such exposure or its effects may modify risk. Hence, in a cross-sectional study of 78 patients with colorectal disease, we have examined whether (i) metabolic genotypes (GSTT1, GSTM1, CYP2D6, CYP2E1) are associated with O(6)-methyldeoxyguanosine (O(6)-MedG) levels, O(6)-alkylguanine-DNA alkyltransferase (ATase) activity or K-ras mutations, and (ii) there was an association between ATase activity and O(6)-MedG levels. Patients with colon tumours and who were homozygous GSTT1(*)2 genotype carriers were more likely than patients who expressed GSTT1 to have their DNA alkylated (83 versus 32%, P=0.03) and to have higher O(6)-MedG levels (0.178+/-0.374 versus 0.016+/-0.023 micromol O(6)-MedG/mol dG, P=0.04) in normal, but not tumour, DNA. No such association was observed between the GSTT1 genotype and the frequency of DNA alkylation or O(6)-MedG levels in patients with benign colon disease or rectal tumours. Patients with colon tumours or benign colon disease who were CYP2D6-poor metabolisers had higher ATase activity in normal tissue than patients who were CYP2D6 extensive metabolisers or CYP2D6 heterozygotes. Patients with the CYP2E1 Dra cd genotype were less likely to have a K-ras mutation: of 55 patients with the wild-type CYP2E1 genotype (dd), 23 had K-ras mutations, whereas none of the 7 individuals with cd genotype had a K-ras mutation (P=0.04). No other associations were observed between GSTT1, GSTM1, CYP2D6 and CYP2E1 Pst genotypes and adduct levels, ATase activity or mutational status. O(6)-MedG levels were not associated with ATase activity in either normal or tumour tissue. However, in 15 patients for whom both normal and tumour DNA contained detectable O(6)-MedG levels, there was a strong positive association between the normal DNA/tumour DNA adduct ratio and the normal tissue/tumour tissue ATase ratio (r(2)=0.66, P=0.001). These results indicate that host factors can affect levels both of the biologically effective dose arising from methylating agent exposure and of a susceptibility factor, the DNA repair phenotype.

Non-steroidal anti-inflammatory drugs in chemoprevention of breast and prostate cancer.

Despite convincing evidence from animal experiments, epidemiological studies linking the use of non-steroidal anti-inflammatory drugs (NSAIDs) with lower risk of breast and prostate cancer have been equivocal. One explanation for the inconsistencies among epidemiological studies may relate to individual differences in NSAID metabolism due to genetic polymorphisms in enzymes such as N -acetyltransferases and cytochrome P4502C9, which are known to be involved in the metabolic biotransformation of NSAIDs. The exclusion of these molecular biomarkers of individual susceptibility may have contributed to the inconsistent findings on the effects of NSAIDs in breast and prostate cancer.

Possible mechanisms of alteration in the capacities of carcinogen metabolizing enzymes during schistosomiasis and their role in bladder cancer induction.

Carcinoma of the urinary bladder is the most common malignancy in many tropical and subtropical countries. There is a well documented association with chronic urinary schistosomal infection, and bladder cancer associated with schistosomiasis is a major cause of morbidity and mortality in the endemic areas. Many factors have been suggested as possible causative agents in schistosome-associated bladder carcinogenesis but theories concerning the possible role of schistosomal infection in altering host metabolism of chemical carcinogens have received most attention. In experimental schistosomiasis there is a common pattern of changes in the activities of several hepatic Phase I and Phase II enzymes. Phase I enzymes show increased activities in the early stages of infection but these activities are reduced to below their preinfection levels in the intermediate and late chronic stages of the disease. The activities of Phase II enzymes are altered in favour of the deconjugation pathways in the later stages of the disease. The possible basic mechanisms that might be involved in such changes during parasitism and their potential role in the induction of bladder neoplasia are discussed.

Detection of O6-methylguanine in human DNA.

Using very specific antibodies in sensitive radioimmunoassays for O6-methyldeoxyguanosine (O6-medGuo), we have been able to detect the presence of this modification in human DNA. Since O6-medGuo is not likely to be a normal component of DNA, its presence must be due to exposure to environmental alkylating agents. We have studied two groups of samples, one of which appears to have received exposure to an alkylating agent (so that most members of the group show a detectable level of O6-medGuo). Although some individuals in the other group showed detectable O6-medGuo (sometimes at levels exceeding those observed in the first group), the majority showed undetectable levels. This may indicate that they had much less exposure to environmental alkylating agents than the first group and that some individuals may have received additional exposures due to other factors such as life style or drugs that they may have been given.