The link between cell-cycle dependent radiosensitivity and repair pathways: a model based on the local, sister-chromatid conformation dependent switch between NHEJ and HR.

The different DNA damage repair pathways like homologous recombination (HR) and non-homologous end joining (NHEJ) have been linked to the variation of radiosensitivity throughout the cell cycle. However, no attempts have been made to test the various hypotheses derived from these studies in a quantitative way e.g. by using modeling approaches. Here we present the first modeling approach that allows predicting the cell cycle dependent radiosensitivity of repair proficient as well as of repair deficient cell lines after photon irradiation based on a small set of parameters and assumptions. A key element of the model is the classification of DNA damage according to its complexity on the level of chromatin loops of about 2Mbp size. Isolated DSB (iDSB), characterized by a single DSB within a chromatin loop, are distinguished from clustered DSB (cDSB), characterized by two or more DSB within a chromatin loop. The class of iDSB is further subdivided into two sub-classes, characterized by the replication status of the corresponding chromatin loop. For iDSB in replicated loops that are in close contact, error-free homologous recombination is assumed to be effective; in unreplicated loops or in replicated loops that have already been separated, iDSB are assumed to be repaired by error-prone non-homologous end joining. cDSB are assumed not to be repairable effectively by neither HR nor NHEJ. Assigning empirically derived lethalities to these three damage classes and pathways, we demonstrate that the model is able to accurately reproduce cell cycle dependent survival probabilities. Notably, the relevant parameters are derived solely from two survival curves for normal, repair proficient cells in G1 and late-S phase. Based on a comparison of model predictions with a large data set reported in the literature, we show that the lethality values for wild type cells are simultaneously predictive for the cell cycle dependent variation of sensitivity observed for HR-deficient and NHEJ-deficient cells.