Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Executive Summary DNA (deoxyribonucleic acid), a complex substance composed of polymers of small molecules called nucleotides, is present in the nuclei of all cells and is the carrier of genetic information. Recent advances in methodology and instrumentation permit detection and measurement of alterations within the individual nucleotides of DNA. In one important kind of DNA alteration, exogenous and xenobiotic materials bind to nucleotides within DNA to form addition products, or adducts. Radiolabeling, immunochemical, and physical methods can detect adducts at concentrations as low as one in 109-10'° nonadducted nucleotides. DNA adducts can form in many tissues, but they are not necessarily stable. They can decompose spontaneously, and a number of them can be removed by enzymatic repair systems at varying rates. The toxicologic significance of the presence of background adducts or even of induced adducts at per- sistently high concentrations is unknown. There is reason to regard some DNA adducts as markers of exposure to specific toxicants. Research must distinguish those from "background" adducts, and their long-term impact on organisms needs to be assessed. In addition, protein adducts, such as those found in the hemoglobin of red blood cells and in sperm protamine, are apparently stable for the lifetime of the cell and are thus good indicators of recent exposure. Protein adducts should be considered in the estimation of genetic or carcinogenic risk whenever they can be correlated with DNA binding, even though they themselves may play no putative mechanistic role. Interpretation of the presence of DNA adducts caused by even a single known carcinogen is highly complex. For example, whether an adduct is crucial to tumor initiation depends heavily on the chemical and toxicologic 3
4 DRINKING WATER AND HEALTH properties of the compound of concern, on the metabolic system of the host organism, and on the site of the adduct within a specific nucleotide, gene, and target tissue. Current evidence suggests associations between the occurrence of DNA adducts formed by specific compounds and various types of toxic effects (usually tissue-specific), such as mutation, cancer, and developmental effects, although in no case has a specific DNA adduct been firmly established as the cause of a tumor or mutation in a mammalian system in vivo. The presence of DNA adducts in a tissue does not necessarily establish risk, but is often an indication that additional testing for toxicity should be done. The technology for detecting and measuring adducts is qualitatively useful in the toxicologic evaluation of specific chemical contaminants of drinking water, but the only quantitative application that has been validated is in the assessment of exposure. Future quantitative applications might be in dosimetry, pharmacokinetics, hazard identification, and risk assessment; DNA-adduct tech- nology might also be applied eventually to large-scale human monitoring studies. The technology and the database have deficiencies that must be resolved before DNA-adduct detection, identification, and measurement can be viewed as routine components of general toxicity testing. With the newest technology, many adducts of unidentified chemical nature or unknown stability can be detected and counted, but the biologic impact of their presence cannot be established without more information. For example, the relationship between adduct formation and tumorigenesis varies with the chemical compound, the specific target tissue, the organism's exposure history, the duration and time of exposure, and so on. Large-scale DNA-adduct dosimetry studies in humans are now becoming possible, but they will be subject to much more variability because of population heterogeneity and thus will have many more problems than corresponding dosimetry studies in animal models. Even with population studies to identify variations due to age, sex, race, and interference factors, it will be much more difficult to predict extent of exposure or cancer risk based on DNA-adduct formation in humans than in homogeneous laboratory animals. The real advantage of gaining information about DNA adducts lies in the addition of this knowledge to the comprehensive database needed for assessing the risk of exposure to chemicals with identified toxicity. This information could be used in individual risk assessment to confirm suspected exposures, improve estimates of target tissue dose, and reveal metabolic activation and detoxification rates for specific carcinogen-DNA-adduct formation. In general risk assessment, it could be extremely valuable in estimating dosimetry and systemic distribution and in establishing possible target tissues or organs and the potential for irre- versible toxicity, such as cancer, mutation, or developmental effects. The use of DNA adducts as molecular dosimeters might make possible the study of differences in absorption, distribution, biotransformation, cell proliferation, and DNA repair and detoxification between high- and low-dose exposures, between species, and even between tissues. New, ultrasensitive methods of detection
Executive Summary 5 make it possible to monitor DNA adducts in animals at exposures below those feasible in chronic bioassays and closer to those expected in the human pop- ulation. Mathematical models that use such biologic dosimeters might yield more accurate extrapolations and thus improve quantitative risk assessment. At the present time, few studies of DNA adducts have been carried out under conditions that are appropriate for use in risk assessment. Additional scientific information is needed to improve the extrapolation process, in- cluding data from tests using a wide range of doses to determine the saturation points of detoxification and repair, which will represent nonlinearities in the dose-response curves. The following are some of the subcommittee's conclusions and recom- mendations regarding the application of DNA-adduct and protein-adduct as- says to EPA's assessment of drinking water contaminants: · DNA-adduct detection methods (especially the radiophosphorus-postla- beling method) have demonstrated the presence of numerous persistent (mostly unknown) DNA lesions in various target cells of untreated animals. The toxicologic importance of those "background" lesions is unknown; they might reflect endogenous exposures to normal body constituents or processes or exposures to naturally occurring mutagens and carcinogens in the envi- ronment, such as ultraviolet radiation and mutagens and carcinogens in foods. · Investigations into the origin and importance of these natural background DNA adducts are needed. The resulting information will be relevant to the interpretation of the impact of DNA alterations induced by occupation, life- style, or environmental exposure. · The presence of DNA adducts or protein adducts might not in itself establish a risk, and tests for their presence should not be used in isolation for hazard or risk assessment. · If the goal is to assess exposure to a drinking water contaminant identified as genetically toxic, it might sometimes be more appropriate to use protein adducts as dosimeters, although the biologic significance of protein adducts could be quite different from that of DNA adducts. Gerrn cell studies have suggested that protamine adducts are relevant to genetic risk assessment. · Baseline data on chemicals in drinking water that are presumed to be genetically toxic should be established, with an eye to revealing qualitative and quantitative associations between DNA adducts or protein adducts and other components of hazard identification. · Differences between rates of in vivo DNA-adduct formation and repair in somatic and germ cells should be studied. Risk assessment that includes DNA-adduct measurements usually focuses on tumor formation, but other heritable effects of genetic toxicants should be considered a major burden for the human population also.