The nucleo-shuttling of the ATM protein as a basis for a novel theory of radiation response: resolution of the linear-quadratic model

Larry Bodgi,Nicolas Foray
DOI: https://doi.org/10.3109/09553002.2016.1135260
Abstract:Purpose: For 50 years, cellular radiosensitivity has been defined in vitro as the lack of clonogenic capacity of irradiated cells and its mathematical link with dose has been described by the target theory. Among the numerous formulas provided from the target theory, the linear-quadratic (LQ) model empirically describes cell survival as a negative exponential of a second degree polynomial dose-function in which αD is the linear component and βD(2) is the quadratic one. The LQ model is extensively used in radiobiology (to describe survival curves) and in radiotherapy (the α/β ratio indicates whether tissue reactions can occur early or late after the treatment). However, no biological interpretation of the LQ parameters was proposed to explain together the radiation response in a wide dose range, the radiosensitivity of some genetic syndromes caused by the mutation of cytoplasmic proteins and the hyper-radiosensitivity phenomenon specific to low-dose. The model: From a solid amount of experimental data, we hypothesized that the major forms of ataxia telangiectasia mutated (ATM) are cytoplasmic dimers and that ionizing radiation induce ATM monomerization. The resulting ATM monomers diffuse into nucleus to facilitate double-strand-breaks (DSB) recognition and repair. Such hypotheses lead to a coherent molecular interpretation of the LQ model by considering the yield of recognized but unrepaired (α-type) DSB and the non-recognized (β-type) DSB. The notion of cell tolerance to unrepaired DSB was introduced by considering that not all DSB are lethal. Cell survival and DSB repair and signaling immunofluorescence data from 42 normal skin fibroblast and 18 tumor human cell lines were used to verify the validity of this biomathematical model proposed. Results: Our model is validated at different levels by one of the widest spectrum of radiosensitivity. That mathematical developments of the present model imply that β is a Lorentzian function of α was confirmed experimentally. Our model is also relevant to describe the hypersensitivity to low-dose phenomenon. Conclusions: Our model provides a very general picture of human radiosensitivity, independently of the dose, the cell type and the genetic status.
What problem does this paper attempt to address?