Human embryonic stem cells (hESCs), derived from inner cell mass (ICM) of human blastocyst embryos, can proliferate indefinitely in undifferentiated state and differentiate into most somatic cell types of three germ layers under the specific culture conditions. The remarkable differentiation potential and proliferative capability of hESCs promise an unlimited supply of donor cells for cell therapy of neurodegenerative disorders such as Parkinson’s disease (PD) and spinal cord injury. The differentiation of large and defined populations of the specific types of functional neurons from hESCs is prerequisite to replace lost cells in neurodegenerative diseases. Therefore, neural differentiation protocol of hESCs for generating neural progenitors (NPs), capable of retaining proliferation capability and differentiation potentials into various types of neurons for a long time, should be established.
In this study, differentiation protocol of hESCs into neurons was divided into two different stages, including selection and expansion of NPs and neuronal differentiation of NPs. In stage I for differentiation into NPs, hESCs were detached from feeder layers and spontaneously differentiated through embryoid body (EB) formation in a suspension culture. Neural cells were selected and expanded under defined serum and feeder free conditions in an adherent culture. The expanded NPs with neural rosettes and/or neural tube-like structures in the presence of basic fibroblast growth factor (bFGF) were isolated mechanically and isolated neural clumps formed spherical neural masses (SNMs) similar to neurospheres. By removing non-neural structures of SNMs along with mechanical passage, the homogeneity of SNMs could be increased and homogeneous SNMs were expanded for a long time under defined culture condition. The homogeneous SNMs robustly expressed the markers for NPs such as nestin, musashi, sox2 and pax6. Therefore, it is suggested that SNMs may be supplied as a cell source of generating neurons for cell based therapy of various neurodegenerative diseases.
In stage II for neuronal differentiation of SNMs, small fragments of SNMs were attached onto the matrigel coated culture dishes and differentiated into neurons by the withdrawal of bFGF. For the induction and maturation of dopamine (DA) neurons, Sonic hedgehog (SHH), fibroblast growth factor 8 (FGF8), glial cell derived neurotrophic factor (GDNF) and ascorbic acid (AA) were treated in stage II for 10 days. hESC-derived TH+ neurons expressed midbrain DA markers, including En1, aromatic amino acid decarboxylase (AADC), Nurr1 and Pitx3, but DBH, PNMT and GABA were hardly detectable. These results supported that most of the TH+ neurons from hESCs are midbrain DA neurons. When transplanted into a PD rat model, hESC-derived DA neurons survived and induced behavioral recovery, suggesting therapeutic potential for PD.
In conclusion, SNMs differentiated from hESCs could be expanded for several months and cryopreserved, retaining differentiation potential into neurons. These SNMs efficiently differentiated into midbrain DA neurons by adding instructive factors. Transplanted SNM-derived DA neurons in PD rats survived and elicited clear behavioral recovery of animal model. Based on these results, it is suggested that neural differentiation protocol developed in this study may be essentially useful for the large-scale generation of neurons for cell replacement therapy in neurodegenerative disorders.