Investigation of double strand breaks induced by alpha particle irradiation using C.N.B.G. microbeam in human keratinocytes

To understand the mechanisms of interaction of ionizing radiation with living tissues exposed to low and protracted doses remains a major issue for risk evaluation. The response cannot be found in epidemiological studies because the only available data concern accidental exposures to high doses of radiation. The natural exposure represents the main source of exposure in the daily life, just before the medical sources (radiology, radiotherapy). In addition, this kind of exposure is very difficult to reproduce in vitro by irradiating cell lines. The method per preference is based on random irradiation of cell populations. The mean number of particles having traversed cells is then calculated on the basis of Poisson statistics. In addition to inevitable multiple impacts, the numerous potential intracellular targets (nuclei, cytoplasm), the indirect effects induced by the impact of particles on neighbouring cells or simply the extracellular targets, constitute phenomena that make more complex the interpretation of experimental data. A charged particle microbeam was developed at C.E.N.B.G. to perform the targeted irradiation of individual cells with a targeting precision of a few microns. It is possible to deliver a counted number of alpha particles down to the ultimate dose of one alpha per cell, to target predetermined cells and then to observe the response of the neighbouring cells. This facility has been validated during this work on human keratinocyte cells expressing a recombinant nuclear fluorescent protein (histone H2B-GFP). The combination of ion micro-beams with confocal microscopy and numeric quantitative analysis allowed the measurement of DNA double strand breaks via the phosphorylation of the histone H2A.X in individual cells. The mechanisms of DNA reparation and apoptosis induction were also in the scope of those studies. The experimental results obtained during this thesis validate the methodology we have developed by demonstrating the targeting reproducibility and the dose control. A dose-effect relationship has been also investigated as a function of time. (author)