Erik Kriek

Polycyclic aromatic hydrocarbon-DNA adducts in humans: relevance as biomarkers for exposure and cancer risk

The methodology applied for DNA adducts in humans has become more reliable in recent years, allowing to detect even background carcinogenic adduct levels in environmentally exposed persons. Particularly, combinations of the various methods now allow the elucidation of specific adduct structures with detection limits of 1 adduct in 108 unmodified nucleotides or even lower. The quantification of polycyclic aromatic hydrocarbon-DNA (PAH-DNA) adducts in human tissues and cells has been achieved with a number of highly sensitive techniques: immunoassays and immunocytochemistry using polyclonal or monoclonal antisera specific for DNA adducts or modified DNA, the assay, and adduct identification using physicochemical instrumentation. The results summarized in this review show that PAH-DNA adducts have been detected in a variety of human tissues, including target organs of PAH- and tobacco-associated cancers. Although dosimetry has not always been precise, a large number of data now clearly show that lowering exposure to carcinogenic PAH results in decreasing PAH-DNA adduct levels. In most studies, however, bulk DNA of a certain tissue or cell type has been examined, and there were relatively few studies in which mutations as a consequence of DNA damage at specific genes have been investigated. Promising as these biomarker studies seem for epidemiology and health surveillance, future biomonitoring and molecular epidemiological studies should be directed to combine several endpoint measurements: i.e., adduct formation (preferably at specific sites), mutational spectra in cancer-relevant genes, and genetic markers of (cancer) susceptibility in a number of cancer-predisposing genes.

32P-Postlabelling of aromatic DNA adducts in white blood cells and alveolar macrophages of smokers : saturation at high exposures

DNA adducts may serve as a molecular dosimeter of exposure to cigarette smoke-associated carcinogens such as polycyclic aromatic hydrocarbons (PAH). Target tissues for cigarette smoke-induced carcinogenesis are rarely accessible; therefore, peripheral blood cells or cells obtained by bronchoalveolar lavage (BAL) may be used as surrogate sources of exposed DNA. However, the relationship between cigarette smoke exposure and aromatic-DNA adducts in white blood cells and BAL cells is still unclear. In this study, we examined DNA adduct formation in lymphocytes and BAL cells in several populations of smoking individuals by means of 32P-postlabelling. Significant correlations between the amount of cigarettes smoked per day and the level of aromatic-DNA adducts were found in lymphocytes. In BAL cells, DNA adduct levels were associated with age (p = 0.05) and gender (p = 0.10) after adjustment for smoking behaviour. Adduct formation levelled off at higher exposure levels, suggesting less efficient adduct formation; decreases in the formation of adducts per unit of exposure were found in lymphocytes (r(s) = -0.80, p < 0.001) and BAL cells (r(s) = -0.72, p < 0.001). To assess intra-individual variation in adduct levels at constant smoking behaviour, sampling was repeated after a period of 2 and 6 months. In lymphocytes, repeated measurements with an interval of 2 months were highly correlated (r = 0.84, p = 0.009, n = 8), whereas repeated measurements with an interval of 6 months showed no correlation (r = 0.30, p = 0.27, n = 16). Repeated measurements in BAL cells showed a significant correlation after 6 months (r = 0.68, p = 0.03, n = 10). Furthermore, in a group of occupationally exposed aluminium workers, adduct levels in total white blood cells were correlated with the average concentrations of PAH in the ambient air of workers who smoked cigarettes, whereas in non-smokers, no such relationship was found. We conclude that cigarette smoking may directly or indirectly influence DNA adduct levels and saturation of DNA adduct formation may occur, leading to non-linear dose-response relationships.

Methylation of deoxyribonucleic acid by diazomethane

Abstract 1. 1. Aqueous solutions of salmon-sperm DNA were treated with diazomethane and diazoethane dissolved in ether. In both cases 7-alkylguanines and 3-alkyladinines were formed in a ratio of 3–4:1. The yields of the ethylated purines were much smaller than those of the methylated purines. The ethylating capacity appeared to be inversely related to the rate of protonation of diazoethane. Per 100 deoxynucleotides about 16 phosphate groups were esterified and 6 purines were methylated by diazomethane. 2. 2. Incubation of DNA with methylamine plus nitrous acid at pH 4 led also to the formation of 7-methylguanine. 3. 3. The hyperchromicity was fully retained in methylated DNA, but the slope of the absorbancy-temperature curve suggested that the change in the secondary structure (unfolding) was preceded by one in the primary structure of methylated DNA during heating. The latter change (main chain fission) may be triggered off by the loss of methylated purines, 3-methyladenine being released 4–5 times faster than 7-methylguanine at pH 7.4 and 37°. 4. 4. The differences in reaction between DNA and RNA with diazomethane are described.

On the mechanism of action of carcinogenic aromatic amines: II. Binding of N-hydroxy-N-acetyl-4-aminobiphenyl to rat-liver nucleic acids in vivo

Abstract The sulfuric acid ester of N-hydroxy-N-4-acetylaminobiphenyl reacted with guanosine at neutrality in a manner similar to N-acetoxy-N-2-acetylaminofluorene yielding N-(guanosin-8-yl)-4-acetylaminobiphenyl. It had properties identical with those of the corresponding fluorene derivative, e.g. it was hydrolysed by 0.01 N NaOH at 100° to N-(guanosin-8-yl)-4-aminobiphenyl and by 1 N HCl at 100° to N-(guanin-8-yl)-4-aminobiphenyl. The latter was prepared by acid hydrolysis of the condensation product of 4-aminobiphenyl and 8-bromoguanosine 2′,3′,5′-triacetate. The binding of AABP and ABP to rat-liver ribosomal RNA and DNA in vivo was studied 24 h after intraperitoneal injection of a single dose of N-hydroxy-4-acetylamino-[3H]BP and of N- hydroxy -N-[2′- 3 H ] acetyl-ABP . The hepatic nucleic acids were hydrolyzed at pH 9 with a mixture of venom phosphodiesterase and alkaline phosphatase from Escherichia coli. The enzymatic digests were analyzed by Sephadex LH-20 column chromatography. In contrast to the values previously obtained for N-hydroxy-AAF in male rats, only 12% of the radioactivity associated with ribosomal RNA was identified as N-(guanosin-8-yl)-4-acetylamino-[3H]BP, while 88% of the RNA-bound material was found to be deacetylated. In DNA 22% of the bound carcinogen had retained the N-acetyl group, a value which has been found also for N-hydroxy-AAF. The binding of N-hydroxy-[9-14C]AF, N-hydroxylamino-[3H]BP and [2′-3H]-acetyl-ABP-N-sulfate (potassium salt) to ribosomal RNA have been compared in aqueous solutions of pH 6 and pH 7. The results obtained confirmed the large difference in reactivity between N-hydroxy-AF and N-hydroxy-ABP as well as the pH-dependence of the reactions.

Fifty years of research onN-acetyl-2-aminofluorene, one of the most versatile compounds in experimental cancer research

SummaryIt is just about 50 years since the publication of the report on the toxicity and carcinogenicity of the potent carcinogenN-acetyl-2-aminofluorene (AAF). In 1940 very few reports on the carcinogenic activity of chemical compounds in experimental animals were available. The discovery of pure chemicals as carcinogens, such as AAF, azo dyes and benzo[a]pyrene, provided cancer researchers with a number of tools whereby the progressive changes involved in the induction of cancer could be studied in experimental systems. Contrary to the results with other carcinogens then known, AAF induced numerous types of tumors, but not at the site of application. This finding stimulated a great deal of interest in its use as an experimental carcinogen to study its metabolic fate and mechanism of action. During the following years an ever increasing number of reports appeared on the carcinogenicity of AAF in various species, on its metabolic fate, on the interaction of reactive metabolites with nucleic acids and proteins, and on its mutagenic activity. Particularly studies on the metabolism of AAF and the interaction with nucleic acids have contributed appreciably to our understanding of the mechanism of action of aromatic amines and also of other chemical carcinogens. It can be expected that AAF and its derivatives will continue to be used for specific applications in experimental cancer research. One of the most recent achievements is the preparation of site-specific AAF- and aminofluorene-modified DNA sequences for mutagenesis studies.

Detection of benzo[a]pyrene-diol-epoxide-DNA adducts in white blood cells of psoriatic patients treated with coal tar

An enzyme-linked immunosorbent assay (ELISA) was used to detect BPDE-DNA adducts in white blood cells of 23 psoriatic patients undergoing clinical coal tar therapy. Ten of these patients were reanalyzed 2-5 months after the end of the coal tar treatments. The results show that the mean adduct level during the treatment period was 0.26 +/- 0.16 fmole BPDE/micrograms DNA (7.7 +/- 4.9 adducts/10(8) nucleotides), while 2-5 months later the mean adduct level had decreased significantly (P less than 0.005) to 0.11 +/- 0.08 fmole BPDE/micrograms DNA (3.3 +/- 2.4 adducts/10(8) nucleotides). No relationship could be ascertained between the level of exposure and the amount of BPDE-DNA adducts. In addition, no difference in the level of DNA adducts was found between smoking and non-smoking patients.

Comparison of 32P-postlabeling and HPLC-FD analysis of DNA adducts in rats acutely exposed to benzo(a)pyrene

DNA adduct analysis is often used for biomonitoring individuals exposed to polycyclic aromatic hydrocarbons (PAH). The 32P-postlabeling assay is routinely applied to study the formation of aromatic bulky adducts, but cannot positively identify individual adduct types. Recently, an HPLC assay with fluorescence detection (HPLC-FD) was developed which was sufficiently sensitive to detect adducts formed by benzo[a]pyrene (B[a]P) diolepoxide isomers [(+/-)anti- and (+/-)syn-BPDE] in occupationally exposed subjects (Rojas et al. Carcinogenesis, 16 (1995) 1373-1376). In this study, we compared both techniques using DNA samples of rats which were treated i.p. with B[a]P (10 mg/kg bw). The internal dose was assessed by measuring 3-OH-B[a]P excretion in urine. The detection limit of the HPLC-FD assay varied from 0.5 to 7.4 adducts per 10(8) nucleotides, while the detection limit of the 32P-postlabeling assay was around 1 adduct per 10(9) nucleotides. HPLC-FD analysis showed that BPDE-DNA adduct levels were highest in the heart, lung and liver respectively. The most predominant B[a]P-tetrol was the I-1 isomer, which derives from hydrolysis of the major reaction product of DNA and (+)-anti-BPDE. 32P-postlabeling analysis revealed an adduct spot that comigrated with a [3H]BPDE-DNA standard. The putative BPDE-DNA adduct levels were highest in heart followed by lung and liver and correlated significantly with tetrol I-1 levels determined by HPLC-FD (r = 0.72, P = 0.006). In samples in which both tetrol I-1 and II-2 were detected by means of HPLC-FD, this correlation was even better (r = 0.95, P = 0.01). Estimated half-lives of BPDE-DNA adducts were in the ranking order; heart, lung and liver for both techniques. By 32P-postlabeling, adducts other than BPDE-DNA were also found, resulting in highest total DNA adduct levels in the liver, heart and lung respectively. Furthermore, mean 24 h urinary excretion of 3-OH-B[a]P was related to BPDE-DNA adduct levels in lung, liver and heart. The 32P-postlabeling assay is sensitive and capable of detecting exposures to complex mixtures, whereas the HPLC-FD assay can be used to identify BPDE-isomers and might therefore be of value in risk assessment of individuals exposed to PAH.

The differential reactivity of the oxidation products of o-aminophenols towards protein and nucleic acid

Abstract 1. 1. The reaction of the oxidation products of o -aminophenols with protein and nucleic acids was sstudied. 2. 2. Changes in the spectrum of bovine serum albumin following its exposure to the oxidation products of 2-aminophenol and 3-hydroxyanthranilic acid indicated that these compounds had combined with protein. 3. 3. Treatment of DNA with the air oxidation product of 2-amino-1-naphthol failed to alter its spectrum, hyperchromicity, T m , sedimentation behavior or base composition. 4. 4. The ferricyanide oxidation products of 2-aminophenol, 3-hydroxyanthranilic acid, 2-amino-1-naphthol and 2-amino-1-fluorenol and the air oxidation product of 2-amino-i-naphthol were shown to be without effect on the spectral properties of s-RNA. The treatment of s-RNA with the ferricyanide oxidation product of 2-amino-1-fluorenol was without effect on its sedimentation behavior. 5. 5. This work supports the conclusion that the oxidation products of o -aminophenols combine with protein in vitro but not with nucleic acids. These results suggest that it is improbable that the oxidation products of o -aminophenols react with nucleic acids in vivo .

Biosynthesis and reactivity of the glucuronide of N-hydroxy-4-acetylaminobiphenyl

Abstract Adult rats of both sexes excreted approx. 30% of a single dose (50 mg/kg) of N -hydroxy-4-acetylaminobiphenyl ( N -hydroxy-AABP) in the bile in 24 h as the glucuronide conjugate of N -hydroxy-AABP ( N -Gl O -AABP). N -Gl O -[G- 3 H]AABP was isolated from the bile of rats given N -hydroxy-[G- 3 H]AABP and was purified by chromatography on DEAE-Sephadex A-25 and Sephadex G-10. This glucuronide was much less reactive with tRNA and guanosine than was the glucuronide of N -hydroxy-2-acetylaminofluorene ( N -Gl O -AAF). Incubation of tRNA with N -Gl O -[G- 3 H]AABP led to the binding of approx. 1 nmole of aminobiphenyl derivatives per mg of RNA whereas, under comparable conditions of reaction with N -Gl O -[9- 14 C]AAF 27 nmoles of aminofluorene derivatives were bound per mg of tRNA. The lower reactivity of N -Gl O -AABP compared to N -Gl O -AAF extended over a range of glucuronide concentrations and was not significantly influenced by increasing the pH of the reaction mixture. The low levels of radioactivity associated with tRNA after reaction with N -Gl O -[G- 3 H]AABP were mainly due to the binding of deacetylated derivatives (-aminobiphenyl residues) to guanine of the RNA.

Epidemiologic studies of passive smoking and lung cancer: A possible role for Benzo(a)pyrene‐DNA adduct measurements

The question as to whether passive smoking increases lung cancer risk continues to be a controversial issue. A brief critical review is presented of the available epidemio‐logical studies. The discrepancies between studies have been largely attributed to the widely varying ways exposure to passive smoke has been assessed in different studies. So far, all studies have estimated exposure from interview data only. Since passive smoking may occur in many settings and may result from many sources, and since the intensity of exposure is very difficult to assess by interviewing techniques, it has been found virtually impossible to quantify exposure in the studies available now. It is argued that incorporation of biochemical assessments of tobacco smoke constituents and their metabolites can greatly improve future studies by providing a more accurate measure of exposure. The requirements that such laboratory methods should meet in order to become a useful tool in epidemiological investigations are discussed.