Corneal endothelial cells (EC) form a monolayer of hexagonal contiguous cells located at the inner surface of the corne and are responsible for maintenance of its transparency, and therefore essential for vision. Just before birth, they lose the proliferative capacity and remain blocked in the G1 phase of the cell cycle. The absence of cell replacement by mitosis induces is responsible for certain blinding diseases such as cornea gutatta and pseudophakic bullous keratopathy, two prototype endothelial diseases that are among the first indications of corneal graft. The molecular mechanisms involved is the non-proliferation of EC remain only partially explained. The first part of this thesis aimed to identify whether changes in transcriptional expression of cell cycle regulators genes occurred during organ culture (OC) and during in vitro culture, in order to better define the potential targets to inhibit or to overexpress necessary to trigger a controlled cell proliferation. Forthe first time we have highlighted variable transcriptional profiles depending on endothelial environment, with a glolal activation of gene expression in routine OC and in primary cultures, especially with an increased expression of gene involved in cell cycle arrest at different points like DIRAS3, GADD45, p15 p16, p18 and p19 or involved in cycle regulation as the ubiquitin/proteasome complex (Culines, APC ...), suggesting that the antiproliferative brakes are even more complex than previously reported. In a second part we developed an original method of pan-corneal endothelial viability assessment. This innovative measurement tool combines a triple Hoechst/Ethidium/Calcein-AM labeling with a part endothelial image analysis allowing to define the new concept of viable endothelial cell density, that represent the reaviable cell pool of a cornea. This technique can be applied to the analysis of any procedure (surgical or not), likely to directly or indirectly alter the corneal endothelium
