Glaucoma, just as many other neurodegenerative diseases, triggers neuronal death and remained incurable, hence representing a heavy burden for the society. Therefore, there is a critical need for developing new therapeutic strategies to delay the progression of and, ultimately, cure neurological conditions. For decades, neuroscientists studying injuries and diseases of the CNS have largely focused on understanding the mechanisms of axon degeneration to identify new targets for axonal protection and regeneration. But recent data indicates that dendritic deficits represent an early feature of neurodegeneration, a phenomenon now called dendritic pathology and playing a key role in the pathogenesis of neurodegenerative diseases including glaucoma. Because dendrites are essential structures for neuronal communication and function, it is therefore crucial to protect or restore connectivity as well as axons of surviving neurons to improve patients’ condition. In spite of this, the ability of injured neurons to regenerate dendrites remains largely ignored. The central hypothesis of the thesis is that: i) adult CNS neurons can regrow their dendrites after axonal injury, and ii) the identification of underlying signalling pathways would offer new therapeutic avenues to slow or prevent retinal ganglion cell death during ocular neuropathies such as glaucoma. In the first part of my thesis, I demonstrated that mammalian neurons are endowed with the ability to restore their dendritic arbor and synaptic connectivity. Using adult transgenic mice subjected to optic nerve axotomy, we have shown that retinal ganglion cells (RGCs) rapidly undergo dendritic shrinkage before cell death or axonal damage become visible. We also demonstrated that daily insulin, administered topically (eye drops) or systemically (intraperitoneal) after dendritic arbour shrinkage and prior to neuronal loss results in a robust regeneration of dendrites and successful reconnection with presynaptic targets. Moreover, insulin-mediated restoration of dendritic arbors extended neuronal survival and rescued lighttriggered retinal responses. Targeted loss-of-function experiments using siRNAs revealed that insulin-dependent regeneration requires both the activity of both mTOR complexes, mTORC1 and mTORC2 which act synergistically, mTORC1 promoting new dendritic branching to restore arbor complexity, while mTORC2 drives dendritic process elongation. In the second study presented in my thesis, we showed for the first time that morgana, a chaperone protein downstream of mTORC2, is expressed by RGCs and severely downregulated soon after axonal injury. We also demonstrate that morgana is required for successful insulinmediated regeneration of RGC dendrites and neuroprotection. Morgana specific knockdown using siRNA designed against morgana resulted in substantial alterations of dendrite elongation, without changes in arbor complexity. Further, we showed that AAV-mediated rescue of morgana expression selectively in RGCs promoted striking regeneration of dendrites and synapses. Hence, our findings identified a new role for morgana in the regulation of dendritic arbor morphology in adult mammalian neurons Collectively, the findings presented in this thesis contribute to a better understanding of the pathological mechanisms underlying RGC dendritic pathology and identified promising targets for the development of novel neuroprotective treatments for neurodegenerative diseases such as glaucoma.