Research into embryogenesis and nervous system development h
Research into embryogenesis and nervous system development has been instrumental to the identification of factors required for cell specification. Using information gleaned from these studies, many groups have developed induction protocols to instruct hESCs and iPSCs to become a variety of neural cell types (Zhang et al., 2001), including motor neurons, dopamine neurons, striatal neurons, and oligodendrocytes (Aubry et al., 2008; Delli Carri et al., 2013; Li et al., 2005; Nistor et al., 2005; Perrier et al., 2004). Some of these traditional differentiation protocols use embryoid body (EB) formation as the first step of lineage restriction to mimic early human embryogenesis (Zhang et al., 2001), which is then followed by manual selection of neuroepithelial precursors. Interestingly, the efficiency of EB formation and subsequent differentiation can vary among hESC and iPSC lines, and in some instances fail using the same culturing conditions (Boulting et al., 2011; Hu et al., 2010; Osafune et al., 2008). While the mechanisms underlying these differences remain to be determined, this observation suggests that progressing through an EB step may not always be optimal. Additionally, EBs cannot be robustly expanded, so one must start with a large number of undifferentiated hESCs or iPSCs to generate enough EBs to push through the various differentiation steps for each experimental or therapeutic use, therefore increasing batch-to-batch variations among differentiation procedures. A technique that efficiently expands neural stem LJI308 cost from hESCs or iPSCs and allows consistent differentiation of neural tissue is of great interest, and there are a number of published protocols that have been developed (Chaddah et al., 2012; Elkabetz et al., 2008; Koch et al., 2009; Nemati et al., 2011). Elkabetz et al. (2008) used an extended EB formation period and sorting methods to isolate rosette stage neuroepithelial cells that allowed them to generate a transient population of expandable neural stem cells that retain differentiation potential. In contrast to other reports (Falk et al., 2012; Koch et al., 2009), only when these cells were grown in the presence of signaling molecules (e.g. sonic hedgehog and notch) were they able to retain rosette formation and structure, induce proliferation, and subsequently differentiate into motor neurons, dopamine neurons, and neural crest progenitor cells (Elkabetz et al., 2008). However, further expansion in the presence of growth factors resulted in the loss of rosette formation and in vitro regionalization capacity and biased the culture toward gliogenic differentiation (Elkabetz et al., 2008). Also, Koch et al. (2009) described a protocol in which neuroepithelial stem cells were mechanically isolated following EB formation and expanded in the presence of EGF and FGF-2 to successfully generate a variety of neural subtypes. However, cells became regionally restricted after ~15 in vitro passages (Falk et al., 2012; Koch et al., 2009). In the current study we have devised a method that generates pre-rosette stem cells directly from hESCs and iPSCs in a free-floating aggregate system in the presence of EGF and FGF-2. Due to their ease of expansion and differentiation, we have termed these cultures “EZ spheres”. Using our previously described method of a mechanical, non-enzymatic chopping technique (Svendsen et al., 1998), EZ spheres can be expanded for at least 30 passages while maintaining chromosomal stability. Given the proper neural differentiation conditions, rosettes appear within whole spheres and upon plate-down indicating that EZ spheres retain rosette properties after long-term exposure to EGF and FGF-2. Longitudinal analysis of neural gene expression patterns in EGF and FGF-2 culture conditions showed consistent and sustained expression of nestin and SOX2 for all lines, with more varied expression of region specific markers including FOXG1, GBX2, PAX7, and OTX2. Nevertheless, EZ spheres could be taken at any passage and placed into appropriate differentiation conditions to generate specialized neuronal and glial subtypes, such as dopamine neurons, motor neurons, striatal neurons, peripheral sensory neurons, astrocytes, and oligodendrocytes, with similar efficiencies between hESCs and iPSCs. Importantly, EZ spheres do not acquire regionally restricted differentiation potential over successive passages. As a result, the EZ sphere method eliminates the need for EB formation and manual selection, allows for exponential expansion of pre-rosette multipotent neural stem cells, is amenable to healthy and disease-specific iPSCs, and increases versatility of lineage specification over other published techniques.