1.1 Introduction
The differentiation ability of human pluripotent stem cells (PSC) to multiple cell types makes them attractive candidates for potential cell therapies, drug screening, and development. Directing PSCs (human embryonic stem cells and human induced pluripotent stem cells) to specific cell lineages is the first critical step required for successful use of PSCs in translational medicine and cell therapy. In the past decade, quite a number of protocols have been developed by researchers for directed differentiation of PSCs into neural and glial lineages with the intention to use the derived neural lineages in models of neurodegeneration. Most of these methods have been developed from the historical research on rodent brain development and vary considerably in terms of starting cell types, such as human iPSC or ESC, cells on feeder layers, and cell purity (Cai and Grabel 2007).
1.2 NSC Derivation from Rosette Formation of Embryoid Bodies
In general, these methods involved human ESC or iPSCs clumps to be first dissociated into near single cells (or very small clumps of 3–5 cells) and later to be aggregated in to embryoid body (EB) suspension cultures, followed by EB attachment to specific matrices such as poly-D-lysine or/and laminin coated culture dish surfaces for the generation of “neural rosettes,” radially organized columnar epithelial cells (Perrier et al. 2004; Zhang et al. 2001). The neural rosettes comprise cells expressing early neuro-ectodermal stem cell (NSC) markers such as Pax6 and Sox1, and are capable of differentiating into various brain regional specific neural and glial cell types in response to appropriate developmental cues (Li et al. 2005; Perrier et al. 2004). The broad neural and glial differentiation potential is unique to early rosette stage cells but is lost upon further in vitro proliferation of NSCs. This observation is not surprising, and reflects mimicking the embryonic in vivo development NSCs and their differential ability to pattern and in specification signals at the neural plate stage, exhibiting broad patterning potential to begin with versus neural precursors emerging after neural tube closure (Jessell 2000). Large number of researchers followed these methods to generate rosette derived neural stem/precursor cells almost up until 2011 and then subsequently differentiate NSCs to specific neural and glial lineages. The reader is referred to an excellent review that depicts various methods of EB formation, neural rosette generation, isolation, and differentiation to down-stream neural subtypes using known conventional morphogens and growth factors such as FGF8, SHH, and Retinoic acid (Cai and Grabel 2007), while human LIF, FGF2, and EGP were used predominantly for proliferation of neural stem/progenitors. Studies performed up until now suggest that Wnts and BMPs inhibit pluripotent stem cell neural induction process by inhibiting the primitive neuro-ectoderm formation. In contrast, LIF supported primitive neuro-ectodermal formation from PSCs (Akamatsu et al. 2009). These early studies provided a good framework for deciphering the molecular chemokine and growth factor interplay in triggering select pathways to yield specific restricted lineage progenitors of nervous system. Though the system appears to be straightforward and simple, it is complicated, as the resulting cells are heterogeneous and the process is less efficient and variable. Therefore, several studies have examined and improvised the process to produce NSCs using relatively easier methods with higher efficiency and swift timelines to get the desired neural lineages.
1.3 Rosette Free NSC in a Monolayer Culture
In a search for better and efficient neural induction methods of pluripotent stem cells, a novel method was developed with an efficient way of neural conversions of PSC by dual inhibition of SMAD signaling pathway, by using Noggin and a small molecule TGF beta inhibitor, SB431542 (Chambers et al. 2009). The synergistic action of two inhibitors, Noggin and SB431542 of SMAD pathway was sufficient for inducing rapid and complete neural induction in a monolayer adherent culture, bypassing the need for the embryoid body and rosette formation. While Noggin is known for BMP pathway inhibition, SB431542 is shown to inhibit Lefty/Activin/TGFβ pathways by blocking the phosphorylation of ALK4, ALK5, and ALK7 receptors. This study showed it is important to inhibit both of these pathways, since treatment with only one molecule is not sufficient enough to achieve full neural conversion. Further, this study by Chambers et al. also proved that the NSC population derived using this method of rosette/EB free dual SMAD inhibition, resulted in epiblast state of primitive NSC that retained positional (anterior-posterior and dorso-ventral) identity with scope to generate neural subtypes from different brain regions. The process was fast and gets completed in 11 days, although critical neural stem cell marker expression can be achieved by day 5 in culture. The data suggested the presence of an intermediate cell type at day 5 of differentiation, as the cells were negative for OCT4 and PAX6 but showed peak expression of FGF5 along with OTX2, which are epiblast markers. Furthermore, in these cultures, an early expression of SOX1 was noticed, much earlier than other neuroepithelial markers such as PAX6 or ZIC1, or even the anterior CNS (FOXG1) and neural crest (p75) expression. Based on the expression of specific markers, it appears that SOX1 might be directly being modulated by SMAD signaling in monolayer cultures, unlike that of early PAX6 expression from Noggin/SB431542 treated cultures. This is the first report of highly efficient neural conversion of PSCs bypassing all the hurdles associated with EB/Rosette formation. In addition, this method also allowed further generation of neural subtypes from the NSCs derived using SMAD inhibition pathways, relatively in a much shorter time period (∼19 days) relative to 30–50 day protocols.
1.4 Pathways Involved in Neural Tube and Neural Crest Lineages
In a further development to the rosette-free derivation of NSC, a robust method to derive, expand NSC, and differentiate into all of the neuronal subtypes derived from neural tube, (such as midbrain dopaminergic neurons and hind brain motor neurons) and also neural crest lineages (such as peripheral neurons and mesenchymal cells), besides astrocytes and oligodendrocytes has been reported (Reinhardt et al. 2013). This procedure of deriving all neural subtypes really makes disease modeling more realistic, since all neural types can be generated, starting with a neural stem/progenitor population. In order to get such type of multipotent neural stem cells from pluripotent cells, the researchers initiated neural induction process by synergistic inhibition of BMP and TGFβ signaling using the small molecules Dorsomorphin (DM) and SB431542 (SB) respectively. In addition, they stimulated a canonical WNT signaling pathway, by using CHIR99021 (CHIR), a GSK3 inhibitor, and the SHH pathway by Purmorphamine (PMA). The neuro-epithelial cells derived by this method were expandable and can be respecified along the dorso-ventral and rostro-caudal axis, indicating that neural subtypes such as dopaminergic or motor neurons could be differentiated from these expandable neural epithelial cells. This study further strengthens and supports the use of such NSCs in disease modeling and phenotype screening as well as in early human brain development studies.
1.5 Differentiation and Gene Expression in Human Brain Development
Several gene expression studies of human fetal development of CNS, along with differentiation of human neural stem cells to generic CNS neurons, made it possible to come up with a potential panel of markers for each of the neural subtypes from different brain regions. In the following sections, some of the recent methods developed and used to generate brain regional specific neural subtypes from pluripotent cells are described.
1.5.1 Differentiation to Forebrain Neurons from Pluripotent Stem Cells
Forebrain cortical interneurons are involved in the disease etiology in several complex diseases such as schizophrenia, autism, epilepsy, Huntington's disease, and hence, are important in disease modeling and cell therapy approaches. Further, it has been challenging as these cells undergo a long latency and protracted periods of development in vivo. Recently, three different groups of researchers came up with a small molecule and growth factor-based strategy for efficient induction of human forebrain cortical interneurons. Lorenz Studer's group used an approach to inhibit the WNT signaling at pharmacological doses, but simultaneously stimulate the SHH pathway to effectively induce forebrain NKX2.1+ neural precursors (Maroof et al. 2013). They employed original dual SMAD inhibition protocol using Noggin and TGF Beta inhibitor SB431542 that robustly induced FOXG1/PAX6 precursors. Replacement of Noggin with ALK2/ALK3 inhibitor, LDN 193189 induced better PAX6 expression, but lowered the FOXG1 expression. To overcome this hurdle, the authors used recombinant DKK1 or the tankyrase inhibitor XAV939, inhibitors of canonical Wnt signaling, that enhanced FOXG1 expression in a consistent manner, enabling a rapid and robust induction of forebrain fate across multiple iPSC donor lines (Maroof et al. 2013). Possibly, this is one of the few early studies to show that with the use of mere three small molecules (XAV939, SB431542, LDN193189) it is possible to achieve robust forebrain lineages in a well-defined and cost effective culture system of neural differentiation.
In a similar approach but with emphasis on neural maturation, another group led by Kriegstein reported the direct differentiation of hPSC into media ganglionic eminence (MGE) like progenitors (NKX2–1+/FOXG1+) that were further taken through the process of maturation into forebrain interneurons (Nicholas et al. 2013) . Both in vitro and in vivo transplantation studies into rodent brains showed that transplanted cells develop into GABAergic interneurons, but took a maturation time of ∼7 months. These authors developed this method still using the tedious step of embryoid body generation and by treating embryoid bodies with a medium that contains B27 enriched with Y27632, a Rho associated kinase inhibitor; SB431542, a TGFβ inhibitor; BMPR1A, bone morphogenetic protein receptor 1a Fc chimera; DKK1, Dickkopf homolog 1; PM, Purmorphamine up until day 25 in cultures and further added BDNF, a brain derived neurotrophic factor, and DAPT, inhibitor of γ-secretase up to day 35. The cells at this stage were either transplanted into animal models or co-cultured with cortical glial cells. Both in in vitro cultures as well as in transplantation experiments, they achieved maturation of hPSC deri...