Radiation-induced versus endogenous DNA damage and assays that measure parameters reflecting DNA damage on cell by cell basis: comments on the article by Pollycove and Feinendegen.

The article by Pollycove and Feinendegen raises important issues regarding the relative contributions of endogenous and radiation-induced DNA damage to the overall DNA damage burden following low level radiation exposures. Clearly, resolution of the issues raised in their article will have important implications regarding regulatory philosophy. Dose-limiting studies of DNA damage measured on a cell-by-cell basis was used to analyze available data in the context of the proposed model. If one proposes that significant numbers of oxidative DNA lesions are present in cells at a steady state level at any give time, then such damage will be included in the background measure of any DNA damage dependent parameter that is sensitive to these classes of DNA damage. Then the expected number of lesions per cell was compared, prior to X- or gamma-ray exposure, at the dose that gives the minimum statistically significant difference from background, at the dose where the DNA damage dependent parameter is twice background (i.e., the doubling dose). The lesion frequencies predicted from the model by Pollycove and Feinendegen are reasonable for the micronucleus assay and the inhibition of DNA supercoil rewinding, but appear to be inconsistent with results from the comet assay. Possible explanations for the inconsistency between the comet assay dose-response data and the predicted levels of DNA damage predicted by the model are discussed, suggesting that the estimates of the radiation induced damage are too low and those for endogenous damage are too high. The goal in introducing these issues is not to be negative to the article but to present a basis for future discussions and more importantly future experimental work, by which the important issues raised can be resolved.