br Results br Discussion Direct reprogramming of human somat

Results
Discussion Direct reprogramming of human somatic harmine hydrochloride into functional neurons has been proposed as an alternative strategy to obtain patient-specific neurons for disease modeling, drug screening, and toxicity tests (Qiang et al., 2014). Indeed, this approach offers some advantages as compared with classic reprogramming to hiPSCs and redifferentiation, notably in time and experimental variability. However, available protocols for generating expandable neural populations rely on the use of mouse cells and do not support the generation of human neurons on a large scale (Cheng et al., 2014; Han et al., 2012; Lujan et al., 2012; Ring et al., 2012; Thier et al., 2012). In the present study, we report a rapid and efficient approach to obtain iNCs from human hematopoietic cells using SeV. Compared with retroviral and lentiviral-based approach, SeV is more efficient for gene delivery to hematopoietic cells and has been used to generate blood-derived, integration-free hiPSCs (Ban et al., 2011). Here, we successfully obtained iNCs from CD133-positive CB cells in less than 1 week by coexpression of SOX2 and c-MYC SeV. The highly proliferative nature of CD133-positive CB cells during the first week of the reprogramming process made difficult to accurately determine the efficiency of CB-iNCs. Nevertheless, the yield of CB-iNCs was really remarkable as infection of 50,000 CD133-positive CB cells resulted in over 20 million of N-CAM-positive cells 2 weeks later. Moreover, CB-iNCs could be cryopreserved, propagated, and further expanded for >60 passages without losing their functional properties. Upon differentiation, CB-iNCs expressed typical mature neuronal markers, such as MAP2, TAU, and NEUN, and displayed electrophysiological properties including spontaneous action potentials and modulation of their activity by glutamate and GABA neurotransmitters. Transient expression of the four Yamanaka factors has been used as alternative approach to generate iNPCs (Kim et al., 2011; Lu et al., 2013; Thier et al., 2012). However, the presence of pluripotent intermediate cell types could not be excluded in those studies and the protocols were laborious and time consuming. In this regard, our study represents a technical advance for the fast and efficient generation of human integration-free iNCs, without forcing the cells to first go back to an intermediate pluripotent stage (Mitchell et al., 2014). Noteworthy, iNPCs have been recently obtained from mouse fibroblasts using small molecules under hypoxia conditions without exogenous transcription factors (Cheng et al., 2014). While these data represent a safe and attractive strategy for the generation of neural progenitors, the effectiveness of this approach on human cells is still unclear. It would be interesting to assay whether this approach allows lineage conversion of human hematopoietic cells. Direct neural conversion from different cell types has been achieved using the proneurogenic factor ASCL1/MASH1 (Caiazzo et al., 2011; Karow et al., 2012; Marro et al., 2011; Pang et al., 2011; Vierbuchen et al., 2010; Wapinski et al., 2013), where it collaborates with BRN2, MYTL1, and SOX2. Most recently, Wernig’s group demonstrated that MASH1 alone is sufficient to convert fibroblasts into iNCs, suggesting that MASH1 is the most important transcription factor for neural conversion (Chanda et al., 2014). Intriguingly, we did not observe activation of MASH1, suggesting that SOX2 activates the neural program either downstream or independently of MASH1. SOX2 is expressed during development in self-renewing and multipotent neuroepithelial stem cells (Avilion et al., 2003; Ferri et al., 2004). SOX2 maintains neural progenitor identity and inhibits neuronal differentiation (Bylund et al., 2003; Graham et al., 2003). Moreover, in several organs, SOX2-positive cells display self-renewal and maintain tissue homeostasis (Arnold et al., 2011). Of note, Cheng et al. (2014) showed that conversion of mouse fibroblasts into neural progenitors by inhibition of histone deacetylase inhibitors, TGF-β, and GSK-3β is accompanied by the activation of endogenous SOX2 expression. Accordingly, we observed high and sustained levels of endogenous SOX2 from day 10 of induction and the expression of SOX2 was maintained during propagation of CB-iNCs. These data are in line with previous works showing that forced expression of SOX2 alone or in combination with other transcription factors can convert mouse and human fibroblasts into neural stem cells (Ring et al., 2012) or multipotent neural progenitors (Lujan et al., 2012). Noteworthy, SOX2 and BRN2 have been recently used to convert fibroblasts into NPCs (Lujan et al., 2012; Zou et al., 2014). Furthermore, SOX2 has been shown to co-occupy with BRN2 (a POU factor) a large set of distal enhancers in NPCs and regulate together a subset of genes important for neural fate (Lodato et al., 2013; Miyagi et al., 2006). Consistent with these data we observed the induction of endogenous BRN2 in our CB-iNCs during the conversion process. Based on these findings, it is tempting to speculate that SOX2 with c-MYC initiate the neural conversion of blood cells and then SOX2 and BRN2 cooperate to drive the CB-iNCs toward neuronal linages.