DNA Adducts as Biomarkers for Monitoring Cancer
August 1, 1999
DNA Adducts as Biomarkers for Monitoring Cancer
By Misun Park, PhD, and Gerald N. Wogan, PhD
It is well known that exposure to certain environmental or occupational toxic chemicals can induce or enhance neoplastic disease, including both benign and malignant tumors, in the process of chemical carcinogenesis. Cumulative knowledge about mechanisms of carcinogenesis has led to the generalization that most carcinogens are really procarcinogens and must be activated by metabolism in vivo to become ultimate carcinogens. This process frequently occurs via reactive intermediates called proximate carcinogens. Ultimate carcinogens are strongly electrophilic reactants that can bind covalently to cellular macromolecules producing DNA or protein adducts.1
Many types of DNA adducts are also formed spontaneously from endogenous pathways such as alkylation, oxidation, and lipid peroxidation. Interpretation of the health significance of DNA adduct level is based largely on experimental data from animal studies in which adduct formation was studied in relation to tumor development. Carcinogenic potencies of various types of DNA adducts have been estimated through quantitative relationships between adduct level and tumor incidence.2
In rat liver, calculated adduct concentration of several liver carcinogens, including aflatoxin B1, tamoxifen, and dimethylnitrosamine, were responsible for a 50% liver tumor incidence ranges from 53 to 2083 adducts per 108 normal nucleotides. While such direct correlation has not been exhibited in humans due to the complexity of environmental exposures, DNA adducts are clearly elevated in cells and tissues of populations at high risk for cancer development. For example, polycyclic aromatic hydrocarbon (PAH)-DNA adducts are elevated in blood cells of foundry and coke oven workers, individuals exposed to high levels of air pollution, and smokers. The DNA adduct level in the target tissue or a surrogate represents a biologically effective dose of the corresponding carcinogen, taking into account individual differences in absorption and metabolism of the chemical as well as repair of DNA damage. Thus, quantification of DNA adduct level has been recognized as a promising method to assess individual cancer risk resulting from chemical exposure and genetic susceptibility.3 The information obtained from these studies would lead to development and evaluation of cancer intervention strategies that may be applied when exposures cannot be reduced. Additionally, the effectiveness and toxicological features of chemotherapeutic agents can be evaluated by biological monitoring of their DNA adducts.
Measurement of DNA Adducts
Measurement of carcinogen-DNA adducts is an important component of molecular epidemiology of cancer associated with exposure to carcinogens of both endogenous and exogenous origins, which can provide biologically effective doses of the exposure and the resulting cancer risks. A variety of methods have been developed for the measurement of DNA adducts, including 32P-postlabeling, immunoassays, gas or liquid chromatography/mass spectrometry (GCMS or LCMS), high performance liquid chromatography (HPLC) with electrochemical detection (ECD) or fluorescence detection, and other physicochemical techniques.4-6 Each analytical method has specific advantages, but accompanying limitations may introduce significant uncertainty in interpretation of qualitative and quantitative aspects of DNA adduct data in humans.
32P-postlabeling has been the most widely used method for DNA adduct detection in tissue samples and has high sensitivity, allowing detection of 1 adduct in 108-10 normal nucleotides. DNA adducts are enzymatically labeled with [g]32P-ATP and detected by autoradiography after thin layer chromatographic (TLC) separation. This assay has been used to detect DNA adducts of carcinogens representing many chemical classes, including arylamines, nitrosamines, PAHs, mycotoxins, alkylating agents, and oxygen radical generators. However, lack of accuracy and reproducibility, substrate-variable labeling efficiency, and labor-intensive procedure have been pointed out as the drawbacks of this method.
Combination of HPLC and 32P-postlabeling has alleviated the difficulties to some extent, but the intrinsic problem associated with the use of enzymatic reaction remains. Immunoassays, including enzyme-linked immunosorbent assay (ELISA) and immunoslot blot assay (ISB), have a sensitivity of about 1 adduct per 108 nucleotides and are useful in analyzing large numbers of samples. Adduct-specific antibodies or antisera are used to detect the corresponding adducts. ELISA has been commonly used for the analysis of DNA adducts of PAH, 4-aminobiphenyl, aflatoxin, and oxidative damage in humans, exhibiting good correlation with the results from other independent analyses. ISB has been recently adapted in measuring the major DNA adduct of malondialdehyde, a carcinogenic product of lipid peroxidation and prostaglandin biosynthesis. The highly fluorescent adduct, pyrimidopurinone, was quantitated by using 1 mcg of DNA from white blood cells and gastric biopsy sample, with a detection limit of 2.5 adducts per 108 bases.
The disadvantages of these assays are dependency on availability of antibodies to the specific adduct and cross-reactivity to structurally similar compounds that creates problems in quantitation. Mass spectrometric measurements of DNA adducts followed by gas or liquid chromatographic separation in general are very accurate and specific. GCMS has been applied to the detection of modified DNA bases from cells exposed to nitric oxide and DNA samples from smokers after volatile derivatization. LCMS with electrospray ionization (LC/ESMS) is advantageous in determining adducted DNA or protein of high molecular weight. However the low sensitivity of these analyses requires large quantities of DNA, which precludes many investigations on clinical samples. Although the sensitivity has been somewhat improved by advanced techniques, including liquid chromatography/tandem mass spectrometry (LC/MS/MS) and LC/ESMS with selected ion monitoring (SIM) mode quantitation, the high cost of these instruments limits wide use of the method.
HPLC with fluorescence or electrochemical detection has been used in the cases of PAH-DNA adducts, aflatoxin B1-albumin adduct, and oxidative DNA damage. This method is relatively straightforward and sensitive but is confined to specific adducts that possess the requisite material properties. The analytical techniques for biomonitoring are often used in combination or in parallel to obtain more reliable information. For instance, platinum-DNA adducts have been measured in peripheral blood leukocytes by means of atomic absorption spectroscopy and ELISA, to correlate adduct formation with clinical response and toxicity of platinum-based chemotherapy.
Promising Analytical Techniques
Other promising analytical techniques recently developed or modified include capillary electrophoresis/laser-induced fluorescence spectroscopy (CE/LIF), ligation-mediated polymerase chain reaction (LMPCR), accelerator mass spectrometry (AMS), and 35S-postlabeling (ADAM). Laser-induced fluorescence (LIF) detection of dansylated nucleotides following capillary electrophoretic separation (CE) has been applied to the DNA adducts of cisplatin and carboplatin.7 The fluorescent dansyl group was introduced to the platinated cross-link adducts by a chemical reaction, posing the possibility of its application to other types of adducts. Yet the reported sensitivity of CE/LIF, which is about 1 adduct per 107 nucleotides after enrichment, may not be adequate for analysis of biological samples. Ligation-mediated polymerase chain reaction (LMPCR) represents a biological approach to adduct analysis.8 It is a PCR-based method capable of detecting DNA adducts at individual nucleotide positions in mammalian genes. Strand breaks are induced by using specific enzymes or chemicals at adducted sites in DNA, and the positions of these cleavages are detected by LMPCR. This technique has been employed primarily to map the distribution of UV-induced DNA lesions and PAH adducts. The results of these analyses suggest correlation between DNA adduct distribution in critical genes, such as p53 and their mutational spectra in cancers, which may help to implicate causative carcinogens.
Accelerator mass spectrometry (AMS) is a technique for quantifying rare isotopes with high sensitivity and precision by measuring the isotope ratios.9 Radiolabeled DNA or protein adducts, with either 3H or 14C, in tissues can be detected with a sensitivity of low attomole level (10-18 mol), following the administration of the labeled parent compound. Although it requires the use of radioactive material, this technique could provide critical information on metabolism and adduct formation of carcinogens as well as repair of adducts at low levels. The ADAM method (adduct detection by acylation with 35S-methionine or 35S-postlabeling) developed in our laboratory utilizes chemical derivatization in radiolabeling DNA nucleoside adducts to overcome the difficulties observed in 32P-postlabeling and to achieve high sensitivity and reproducibility.10 Using this method, acylation of the hydroxyl groups of nucleosides with 35S-methionine derivative was accomplished under mild reaction conditions to give a single product. The feasibility of this method for biomonitoring studies was demonstrated by detection of 4-aminobiphenyl adduct in animal tissues with a sensitivity of low femtomole (10-15 mol). This type of approach is capable of monitoring various kinds of nucleoside adducts from biological samples.
Conclusion
The information summarized above demonstrates that existing methods for DNA adduct analysis are capable of detecting the presence of many specific types of adducts in surrogate cells as well as tissues at putative carcinogenic risk. Although much development effort has been devoted to this area, few of the methods have been completely validated by careful corroborative studies. The chemical form of postlabeling, ADAM, followed by HPLC with radioactivity detection fulfilled several important methodological requirements concerning accuracy, sensitivity, and reproducibility. However, problems with stability, purity, and availability of the commercial 35S reagent severely curtailed future validation and application of this method.
In our continuous effort to develop a strategy for chemical derivatization, a method utilizing fluorescence postlabeling combined with laser-induced fluorescence (LIF) detection has been devised.11 Acylation of carcinogen-deoxynucleoside adducts by a fluorescent dye, BODIPY FL (4,4-difluoro-5,7-dimethyl-4-bora-3a, 4a-diaza-s-indacene-3-propionic acid), and subsequent quantification by HPLC/LIF afforded a detection limit at the attomole level. DNA adducts of 4-aminobiphenyl, as well as adducts from methylation and oxidative damage, were effectively quantified by this procedure. Careful optimization and validation should precede the analysis of biological samples for each carcinogen-DNA adduct, as is the case for other methods. The method retains the advantages of chemical labeling exhibited in ADAM with enhanced sensitivity and safety, and promises to have broad applicability in cancer biomonitoring studies.
The goals of biomonitoring and molecular epidemiological studies involve assessment of cancer risk associated with endogenous and exogenous toxic exposure, elucidation of carcinogenesis mechanisms, and development of cancer prevention strategies. Among many biological markers for cancer from toxic exposure to clinical outcome, DNA adducts serve as biomarkers for both genotoxic exposure and initiation of carcinogenesis. During the last generation, the methodologies for DNA adduct detection have become highly sophisticated and efficient. Substantial amounts of meaningful data have been produced through DNA adduct biomonitoring in conjunction with other studies, including chemical and biological investigation of cancer. We hope that the knowledge obtained from these studies will contribute significantly to the overall body of knowledge of cancer biology. (Dr. Wogan is Professor of Toxicology and Chemistry and Dr. Park is Post-Doctoral Associate at Division of Bioengineering and Environmental Health, Massachusetts Institute of Technology, Cambridge, MA.)
References
1. Miller EC, Miller JA. Cancer 1981;47:1055-1064.
2. Ottender M, Lutz WK. Mutation Res 1999;424:237-247.
3. Wogan GN. Environ Health Perspect 1989;81:9-17.
4. Warren AJ, Shields PG. Proc Soc Exp Biol Med 1997;216:172-180.
5. Dale CM, Garner RC. Food Chem Toxicol 1996;34:905-919.
6. Leuratti C, Singh R, Lagneau C, et al. Carcinogenesis 1998;19:1919-1924.
7. Sharma M, Jain R, Ionescu E, et al. Anal Biochem 1995;228:307-311.
8. Pfeifer GP, Denissenko MF, Tang MS. Toxicol Lett 1998;102-103:447-451.
9. Dingley KH, Roberts ML, Velsko CA, et al. Chem Res Toxicol 1998;11:1217-1222.
10. Sheabar FZ, Morningstar ML, Wogan GN. Proc Natl Acad Sci U S A 1994;91:1696-1700.
11. Park M, Ozbal CC, Skipper PL, et al. Proc Natl Acad Sci U S A 1999;in press.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content