Comparative characterization of human fetal neural stem cells and induced neural stem cells from peripheral blood mononuclear cells

Human-induced neural stem cells (iNSCs) transplantation is a potential treatment of neurodegeneration diseases. However, whether the reprogrammed cells have the same characterizations as human fetal neural stem cells needs further exploration. Here we isolated human fetal neural stem cells from aborted 12-week fetal brains and compared with iNSCs reprogrammed from human peripheral blood mononuclear cells in gene expression, proliferation ability, differentiation capacity, and the responses to tumor necrosis factor-α. We found that iNSCs and NSCs both expressed neural stem cell markers Nestin, SOX1, and SOX2. However, only iNSCs can be patterned into dopaminergic neurons and motor neurons. Furthermore, both iNSCs and NSCs can differentiate into oligodendrocyte progenitor cells. In addition, a low dose of tumor necrosis factor-α did not inhibit the proliferation and differentiation of iNSCs and NSCs. In conclusion, iNSCs have properties similar to, and even better than, fetal neural stem cells and may be suitable for disease modeling and transplantation. BIOCELL 2020 44(1): 13-18 Doi: 10.32604/biocell.2020.07593 www.techscience.com/journal/biocell Introduction Neurodegenerative diseases are characterized by the loss of neurons in Alzheimer’s disease (AD), loss of dopamine (DA) neurons in Parkinson’s disease (PD), and loss of motor neurons in amyotrophic lateral sclerosis (ALS) (Muñoz, 2018). Transplanted DA neurons could restore striatal DA release and relieve symptoms in PD patients (Kordower et al., 2008; Li et al ., 2008). The patientspecific DA neuroblasts derived from induced pluripotent stem cells (iPSCs) and embryonic stem cells (ESCs) could eliminate ethical concerns and avoid immune reactions (Elkabetz et al., 2008; Frontini-López et al., 2018). In addition, the induction of neural stem cells (iNSCs) from somatic cells reduces the risk of tumor development associated with stem cells (Thier et al., 2012; Lu et al., 2013). However, whether the reprogrammed NSCs have the same features as human fetal neural stem cells or not still needs further exploration. In this study, we compared the iNSCs derived from human peripheral blood mononuclear cells (PB-MNCs) with NSCs derived from human fetal brain for gene expression, proliferation ability, differentiation capacity, and response to tumor necrosis factor-α (TNF-α). Our results showed that iNSCs *Address correspondence to: Shengyong Lan, gxlansyu@163.com; Yimin Fan, fanyimin2012@126.com derived from PB-MNCs may be suitable for disease modeling and transplantation for neurodegenerative disease therapy. Materials and Methods Isolation of fetal NSCs and PB-MNCs Fetal brain NSCs were isolated from the women seeking abortion as described previously (Wang et al., 2010). PBMNCs were isolated as described previously (Tang et al., 2016). Briefly, 8 mL venous blood samples were collected from the donor who signed informed consent, and PBMNCs were isolated and cultured in the medium containing 100 ng/ mL SCF, 40 ng/mL insulin-like growth factor 1 (IGF-1), 10 ng/mL interleukin 3 (IL-3), 2 U/mL erythropoietin (EPO) (all from Life Technologies, Carlsbad, USA), and 100 μg/mL transferrin and 1 μM dexamethasone (both from Sigma). For reprogramming of PB-MNCs, PBMNCs were transfected with episomal vectors as described previously (Yu et al., 2009). Briefly, the cDNAs for human OCT4, SOX2, NANOG, LIN28, c-Myc, and KLF4 genes were subcloned into oriP/ EBNA1-based pCEP4 episomal vector (Invitrogen). PBMNCs were transfected with the episomal vectors and then directly plated to irradiated mouse embryonic fibroblasts (MEFs) seeded dishes. The culture medium was exchanged every other day. On day 3 post-transfection, the culture medium was replaced with iNSCs medium. Colonies with morphology similar to iNSCs were visible on day 10 post-transfection.


Introduction
Neurodegenerative diseases are characterized by the loss of neurons in Alzheimer's disease (AD), loss of dopamine (DA) neurons in Parkinson's disease (PD), and loss of motor neurons in amyotrophic lateral sclerosis (ALS) (Muñoz, 2018). Transplanted DA neurons could restore striatal DA release and relieve symptoms in PD patients (Kordower et al., 2008;Li et al., 2008). The patientspecific DA neuroblasts derived from induced pluripotent stem cells (iPSCs) and embryonic stem cells (ESCs) could eliminate ethical concerns and avoid immune reactions (Elkabetz et al., 2008;Frontini-López et al., 2018). In addition, the induction of neural stem cells (iNSCs) from somatic cells reduces the risk of tumor development associated with stem cells (Thier et al., 2012;Lu et al., 2013). However, whether the reprogrammed NSCs have the same features as human fetal neural stem cells or not still needs further exploration.
In this study, we compared the iNSCs derived from human peripheral blood mononuclear cells (PB-MNCs) with NSCs derived from human fetal brain for gene expression, proliferation ability, differentiation capacity, and response to tumor necrosis factor-α (TNF-α). Our results showed that iNSCs *Address correspondence to: Shengyong Lan, gxlansyu@163.com; Yimin Fan, fanyimin2012@126.com derived from PB-MNCs may be suitable for disease modeling and transplantation for neurodegenerative disease therapy.

Isolation of fetal NSCs and PB-MNCs
Fetal brain NSCs were isolated from the women seeking abortion as described previously (Wang et al., 2010). PBMNCs were isolated as described previously (Tang et al., 2016). Briefly, 8 mL venous blood samples were collected from the donor who signed informed consent, and PBMNCs were isolated and cultured in the medium containing 100 ng/ mL SCF, 40 ng/mL insulin-like growth factor 1 (IGF-1), 10 ng/mL interleukin 3 (IL-3), 2 U/mL erythropoietin (EPO) (all from Life Technologies, Carlsbad, USA), and 100 μg/mL transferrin and 1 μM dexamethasone (both from Sigma). For reprogramming of PB-MNCs, PBMNCs were transfected with episomal vectors as described previously (Yu et al., 2009). Briefly, the cDNAs for human OCT4, SOX2, NANOG, LIN28, c-Myc, and KLF4 genes were subcloned into oriP/ EBNA1-based pCEP4 episomal vector (Invitrogen). PBMNCs were transfected with the episomal vectors and then directly plated to irradiated mouse embryonic fibroblasts (MEFs) seeded dishes. The culture medium was exchanged every other day. On day 3 post-transfection, the culture medium was replaced with iNSCs medium. Colonies with morphology similar to iNSCs were visible on day 10 post-transfection.

Statistical analysis
Data were shown as mean ± SEM and analyzed by GraphPad Prism 5 (GraphPad, La Jolla, CA, USA). Comparison was made by two-way ANOVA with Dunnett's test. p < 0.05 was considered to be significant.

Only iNSCs generated dopaminergic neurons and motor neurons
To assess for differentiation capacity of NSCs and iNSCs, they were cultured in a medium that promoted specific neural lineage differentiation. The majority of NSCs and iNSCs were differentiated into MAP2+ neurons, and only minor portion differentiated into GFAP+ astrocytes (Figs. 2(A) and 2(B)). For differentiation into dopaminergic neurons, 9.05% ± 1.89 % of iNSCs became TH + dopaminergic neurons, but there were no TH + positive cells in NSCs (Figs. 2(C) and 2(D)). For differentiation into motor neurons, 9.85% ± 1.95% of iNSCs expressed HB9, a marker of motor neurons, while no HB9 positive cells were found in NSCs (Figs. 2(E) and 2(F)). Therefore, iNSCs have a better differentiation capacity than NSCs.
Both iNSCs and NSCs were patterned to oligodendrocyte progenitor cells Oligodendrocyte plays an important role in the generation of the myelin sheath around the axon. Thus, we explored whether NSCs and iNSCs can differentiate into oligodendrocytes. We cultured iNSCs and NSCs in special medium (Fig. 3(A)), and both iNSCs and NSCs appeared to be bipolar cells (Figs. 3(B) and 3(C)). Following further differentiation, both iNSCs and NSCs expressed OLIG2 NKX2.2 and O1, the markers of oligodendrocyte progenitors (Fig. 3(D)). The differentiation of iNSCs and NSCs into oligodendrocyte progenitors showed no significant difference (Fig. 3(E)).

Low dose of TNF-α did not inhibit iNSCs and NSCs proliferation and differentiation
Elevated TNF-α levels have been reported in many nervous diseases, and TNF-α had various effects on the proliferation and differentiation of NSCs (Walter et al., 2011;Ye et al., 2013;Ben-Hur et al., 2003). We treated iNSCs and NSCs with TNF-α, and immunostaining of proliferation marker Ki-67 at 0, 24, 48, and 72 h time points showed that TNF-α did not affect the proliferation of iNSCs and NSCs at 1 ng/mL and 10 ng/mL (Figs. 4(A) and 4(B)). In addition, TNF-α did not affect the differentiation of iNSCs and NSCs into MAP2positive neuronal cells at 10 ng/mL (Figs. 4(C) and 4(D)).

Discussion
Transplantation of patient-specific iPSCs has become a new approach in regenerative medicine (Peng et al., 2018;Tang et al., 2016;Thier et al., 2012). Currently, a variety of methods have been developed to directly reprogram somatic cells to iNSCs, which greatly promote the application of iNSCs in the clinical (Shahbazi et al., 2018). For example, a recent study reported that iNSCs reprogrammed from PBMNCs could be differentiated into specialized DA neurons, which can then be transplanted into a PD mouse model to evaluate the efficacy for PD treatment (Zheng and Chen, 2019).
In this study, we compared iNSCs derived from PBMNCs and human fetal NSCs. The results showed that iNSCs express higher levels of midbrain and hindbrain genes such as En1, Hoxb2, Hoxb8, while NSCs express higher levels of anterior brain genes. These findings suggest that iNSCs can be at an early stage of neural system development and have the potential of plasticity, consistent with previous reports (Koch et al., 2009;Wang et al., 2019). In contrast, human fetal NSCs could not respond to patterning cue to instruct spinal motor neuron or midbrain dopamine neuron fate. RA induced iNSCs to differentiate into oligodendrocytes (Okada et al., 2008). The activation of the SHH pathway can promote iNSCs to a ventral and posterior progenitor cell fate. In our study, we used SAG1 (SHH agonist) and FGF8 to promote iNSCs to differentiate into TH positive midbrain neurons and used RA and SAG1 to pattern iNSCs to HB9-positive motor neurons, but NSCs could not be instructed into midbrain neurons or motor neurons. This may be due to the fact that human NSCs have regional specificity (Horiguchi et al., 2004;Ono et al., 2007). For example, TH positive neurons are mainly generated from the precursor cells from diencephalon and mesencephalon, while HB9 positive motor neurons are mainly from spinal progenitor cells (Li et al., 2000;Yan et al., 2007).
TNF-α is a key factor in the regulation of pathological processes such as inflammation, autoimmunity (Tchelingerian et al., 1993). TNF-α has been reported to have diverse effects on the proliferation of human NSCs (Widera et al., 2006;Liu et al., 2005). TNF-α could promote the survival of NSCs by activating nuclear factor kappa B pathway (Shih et al., 2015). In this study, we found that TNF-α at low dose had no significant effects on the proliferation and differentiation of NSCs and iNSCs. Further studies are needed to verify the in vivo effects of TNF-α on NSCs in order to facilitate the transplantation of NSCs into diseased condition to promote the functional recovery and evaluate the risk of cancer development following the transplantation. In addition, novel components of medicinal plants could be utilized to promote the proliferation and differentiation of iNSCs with an impact on regenerative medicine (Mahmoudi et al., 2019;Montes et al., 2019;Villa-Hernández et al., 2018).
In conclusion, iNSCs have properties similar to and even better than fetal NSCs and may be suitable for disease modeling and transplantation.