br Experimental Procedures br Author Contributions br Acknow

Experimental Procedures
Author Contributions
Acknowledgments
Introduction The mammalian adult cerebral cortex has a limited capacity to regenerate lost neural tissue after distributor damage, due to the limited ability of the neural precursor cell (NPC) population residing in the two neurogenic niches to migrate and fully differentiate in response to injury-derived signals. To overcome this restriction, efforts have focused on the reprogramming of resident astroglial cells, initially in vitro and more recently in vivo, toward neurogenesis and the formation of functional synapse-forming neurons (Guo et al., 2014; Heinrich et al., 2010). Indeed, reactive astroglial cells isolated from non-neurogenic regions of the adult brain after local injury share hallmarks with NPCs and developmental radial glia (Sofroniew and Vinters, 2010), having the potential to be reprogrammed into induced neurons (Heinrich et al., 2010, 2011). Thus, forced expression of transcription factors are known to instruct neurogenesis in embryonic development, among which the basic-helix-loop-helix (bHLH) gene Neurogenin-2 (NEUROG2) or a combination of ASCL1, LMX1B, and NURR1 transcription factors direct reactive postnatal astrocytes in vitro toward generation of functional glutamatergic (Heinrich et al., 2011) or dopaminergic neurons (Addis et al., 2011), respectively. Furthermore, recent evidence demonstrates that the neuron-forming capacity of astrocytes is also active in vivo following neurodegeneration of the cortex and striatum (Guo et al., 2014; Magnusson et al., 2014; Niu et al., 2013; Torper et al., 2013), highlighting the existence of an endogenous cell source capable to restore connectivity and function following brain trauma. Similarly, it has been shown that not only astrocytes but also somatic cells more distant in lineage to the CNS, such as fibroblasts, can be reprogrammed either into NPCs or directly into various types of neurons, including glutamatergic, dopaminergic, and spinal motor neurons, by different cocktails of transcription factors and neuron-specific microRNAs (Caiazzo et al., 2011; Vierbuchen et al., 2010; Wernig et al., 2008; Yoo et al., 2011) and may thus be used as an exogenous cell source to restore damage following neuronal loss. The aim of this study is to explore the reprogramming potential of the neurogenic factor CEND1 in inducing the reprogramming of astrocytes and embryonic fibroblasts. CEND1 directs neural stem and precursor cells to acquire a neuronal phenotype both in vitro and in vivo (Georgopoulou et al., 2006; Katsimpardi et al., 2008; Politis et al., 2007). It forms part of the neuronal differentiation pathway(s) activated by proneural genes and is directly activated by the proneural genes Neurogenin-1 and 2 (Katsimpardi et al., 2008; Papadodima et al., 2005). Here, we report a synergistic action of CEND1 and NEUROG2 in reprogramming of mouse postnatal cortical astrocytes and embryonic fibroblasts toward acquisition of a neuronal precursor and/or differentiated neuron phenotype. Additionally, we show that CEND1 expression is critical for the NEUROG2-driven reprogramming of astrocytes, suggesting the existence of a reciprocal feedback loop leading to neurogenesis.
Results
Discussion In this study we addressed the question of whether lineage-distant somatic cell types residing inside and outside the CNS can be turned into NPCs and neurons upon forced expression of factors that induce neurogenesis during development. We demonstrated that (1) the neurogenic molecule CEND1 can reprogram mouse astrocytes toward GABA+ neurons and acts synergistically with NEUROG2 to trans-differentiate them toward proliferating NPCs and subtype-specific neurons, (2) CEND1 and NEUROG2 possess a broader neurogenic potential, inducing neuronal reprogramming of MEFs, (3) CEND1 is a key downstream player in NEUROG2-induced astrocytic reprogramming participating in a positive feedback loop leading to neurogenesis, and (3) double overexpression of CEND1 and NEUROG2 results in activation of the Wnt/β-catenin signaling pathway driving astrocytes to acquire an NSC multipotent character. Taken together, our results demonstrate that CEND1 is not only a potent inducer of neuronal differentiation in neural precursor cells as previously shown (Katsimpardi et al., 2008; Politis et al., 2007), but it also has a neurogenic capacity in different cell contexts.