Quantification of radiation induced DNA double-strand breaks in human fibroblasts by PFGE: testing the applicability of random breakage models

Purpose: To assess the applicability of methods of quantification of double-strand breaks (DSB) based on the random breakage paradigm, measuring yield and distribution of DSB induced by varying radiation quality. Material and methods: 240 kVp X-rays and Pu-238 alpha-particles were used to induce DSB in AG01522B primary human fibroblasts. DNA molecular weight distributions were resolved by means of three pulsed-field gel-electrophoresis (PFGE) protocols, which, when combined together, allowed separation and quantification of double-stranded fragments between 5.7 Mbp and 12 kbp. Several analytical methods quantified the DSB yields. Results : Data showed significant differences in the fragmentation patterns according to radiation quality. For both X-rays and alpha-particles, it was observed that the shape of the fragmentation profiles deviates from the prediction of a random breakage mechanism. This is in contrast to other studies where sparsely ionizing radiations appeared to distribute breaks uniformly throughout the genome. Deviations from random breakage were more evident after high linear energy transfer (LET) radiation, which showed an excess of breaks <1 Mbp and a deficit in the production of fragments >1 Mbp, a value that could be dose-dependent. Conclusions : Current methods of DNA fragmentation analysis after induction of DSB may lead to contradictory conclusions on both DSB yields and distributions. This study showed that the application of different DSB quantification methods, derived from random breakage or supported by its concepts, resulted in different radiation biological effectivenesses (RBE) for the induction of DSB, depending on how these methods were employed. To compare experimental results from different laboratories, care should be taken to provide as many details as possible about the application of methods of quantification of DNA damage. For all the methods used, total DSB yields resulted in RBE less than those for mutation induction or reproductive cell death, suggesting that total DSB yields only gave a limited indication of the severity of the inflicted damage. Production of correlated breaks on the chromatin loop structures by single particle-track traversals may explain the deviations observed between experimental data and the predictions of the random breakage paradigm.