Risk assessment of DNA-reactive carcinogens in food.

Risk assessment of DNA-reactive carcinogens in food requires knowledge of the extent of DNA damage in the target organ which results from the competition between DNA adduct formation and repair. Estimates of DNA adduct levels can be made by direct measurement or indirectly as a consequence of their presence, for example, by tumor formation in animal models or exposed populations epidemiologically. Food-borne DNA-reactive carcinogens are present from a variety of sources. They are generally not intrinsically DNA-reactive but require bioactivation to DNA-reactive metabolites a process which may be modulated by the compound itself or the presence of other xenobiotics. A single DNA reactant may form several distinct DNA adducts each undergoing different rates of repair. Some DNA reactants may be photochemically activated or produce reactive oxygen species and thus indirect oxidative DNA damage. The levels of DNA adducts arising from exposures influenced by variations in the doses, the frequency with which an individual is exposed, and rates of DNA repair for specific adducts. Each adduct has a characteristic efficiency with which it induces mutations. Based on experience with the well-studied DNA-reactive food carcinogen aflatoxin B(1) (AFB(1)), a limit of 20 ppb or approximately 30 microg/day has been set and is considered a tolerable daily intake (TDI). Since AFB(1) is considered a potent carcinogen, doses of <1.5 microg of unknown compounds are considered TDIs. Most DNA-reactants, including acrylamide, heterocyclic amines, and alpha,beta-unsaturated carbonyl are below this value. Above that value, measurement of actual DNA adducts levels in either experimental animals with a risk assessment, or, when this occurs, exposed humans are needed. A number of approaches to undertake this are described including immunological, mass spectrometric and (32)P-postlabeling or the use of surrogates such as hemoglobin adducts, together with approaches to evaluate the results. A discussion of approaches to estimating possible threshold effects for DNA-reactive carcinogens is made.