Neural stem cells reversed disease symptoms in an experimental autoimmune encephalomyelitis mouse model of multiple sclerosis

Multiple sclerosis (MS) is an autoimmune inammatory disease of the central nervous system (CNS). MS affects millions of people and causes a great economic and societal burden. Currently used treatment drugs have side effects and only address the symptoms but not the causes of MS. In this study, a novel approach of transplanting neural stem cells (NSCs) derived from human primitive mesenchymal stem cells (MSCs) was investigated in an experimental autoimmune encephalomyelitis (EAE) mouse model of MS.


Introduction
Multiple sclerosis (MS) is a chronic autoimmune in ammatory disease of the central nervous system (CNS) (1). In MS, in ammatory responses are elicited by the intrinsic immune system as well as by parenchymal glial cells, which contribute to oligodendrocyte damage and demyelination, axonal damage and impaired or reduced neuronal signaling (2). MS affects approximately 400,000 individuals in the United States and 2.5 million people worldwide and in icts substantial economic and societal burden due to the early age of disease onset and recurrent relapses (3). Multiple factors including genetics and environmental agents seem to be involved with the onset of MS (2). Although, drug treatments help in reducing disease progression or minimizing disability in patients, they cause severe side effects and do not reverse the symptoms of MS (4). MS drugs pose serious side effects and only help in reducing severity of symptoms or slowing progression rather than treating the cause of disease (5). Therefore, it is imperative to design therapeutic strategies that not only stop disease progression, but also repair the damage to the CNS without severe side effects to MS patients. With the advent of recent progress in stem cell research, cell therapy could provide a promising alternative to drug therapy (6). Mesenchymal stem cells (MSCs) have shown paracrine effects by secreting cytokines and trophic factors involved in immunomodulation (7). MSCs are known to home towards the site of injury and participate in healing and repair processes throughout life (8). Therefore, there is tremendous interest in MSCs due to their medicinal properties. However, MSCs isolated from adult sources have poor growth, limited self-renewal and differentiation potential (8). Moreover, MSCs isolated from adult sources have poor growth, limited self-renewal and differentiation potential as well as pose signi cant risk of graft verse host disease (GvHD). On the other hand, MSCs isolated from perinatal sources are more naïve, have a low risk of host immune response due to the lack of expression of the major histocompatibility complex (MHC) class II antigens and thus do not cause GvHD (8). Recently, we have isolated highly proliferative primitive MSCs, which can be rapidly ampli ed and differentiated into functional derivatives for large-scale clinical studies (8)(9)(10). Although MSCs are known to home towards the site of injury or damage as well as secrete cytokines and growth factors that may indirectly promote endogenous tissue regeneration, transplanted MSCs and their derivatives crossing the blood brain barrier (BBB) to repair the damaged CNS is not well established (11). We sought to determine whether CNS damage due to MS could be repaired by augmentation of neural stem cells (NSCs). Several reports described the ability of NSCs migrate to the area of pathology in experimental models of CNS disease (12)(13)(14). Therefore, we applied a novel approach whereby primitive MSCs were differentiated into NSCs to investigate their effect on the disease process in treating an experimental autoimmune encephalitis (EAE) mouse model, which is known to share similar characteristics pathological features with MS (15). Our results demonstrate reversal of disease symptoms and signi cant improvement in the pathology of EAE animals transplanted with NSCs. This is the rst report examining the anti-in ammatory responses, neuroprotection and endogenous regeneration of perinatal MSC derived NSCs.

Materials And Methods
Maintenance and culture of primitive MSCs Previously isolated and well characterized primitive MSCs (10), were maintained using growth medium (GM) containing DMEM nutrient mix F12 medium (DMEM/F12; Life Technologies, Carlsbad, CA, USA), supplemented with 10% fetal bovine serum (FBS; VWR, Radnor, PA, USA), and 5.6% of antibiotic solution (0.1% gentamicin, 0.2% streptomycin, and 0.12% penicillin); (Sigma, St Louis, MO, USA) and incubated at 37 o C in an atmosphere of 5% CO 2 in a humidi ed incubator. Cells were routinely analyzed using uorescence-activated cell sorting (FACS) as previously described in our published results (10).

Differentiation of primitive MSCs into NSCs
Primitive MSCs were induced towards the neural lineage by culturing the cells in induction media containing 10 µg epidermal growth factor (EGF; PeproTech, Rocky Hill, NJ, USA) in DMEM/F12 basal medium supplemented with 5.6% antibiotic solution for 3 days. Then replaced with neural media containing 20 µg EGF, 20 ng basic broblast growth factor (bFGF), 2 mM Glutamine (Sigma), 1x B27 supplement (Thermo Fisher Scienti c, Waltham, MA, USA), in neurobasal medium for 2 weeks. Characterization of NSCs for the expression of neural markers was performed by immunocytochemical staining, FACS, and quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) analysis.
Immunocytochemical analysis: Cells were xed with 4% paraformaldehyde for 10 minutes at room temperature, permeabilized with 0.5% Triton X-100 (Sigma) and blocked with 2% bovine serum albumin (Sigma) for 1 hour. Cells were then stained with primary antibodies at 1:100 dilution: NESTIN, TUJ1, VIMENTIN, and PAX6 at 4 o C overnight, followed by staining of secondary antibodies at 1:200 dilution at room temperature for 2 hours. Cells were counterstained with DAPI at 1:100 dilution for 5 minutes at room temperature. Fluorescent images were captured using a confocal microscope (NIKON Instruments Inc, Melville, NY, USA). Fluorescent intensity and percentage of positive cells were calculated using ImageJ software (NIH, Bethesda, MD, USA).
Western blot analysis of cells: MSCs and NSCs were lysed for protein using RIPA buffer (Sigma) and quanti ed using the Pierce 660 nm protein assay on the NanoDrop 1000 spectrophotometer (Fisher Scienti c). The lysate (30 µg of total proteins) was resolved using SDS-PAGE with 12% resolving gel and 6% stacking gel and transferred to a nitrocellulose membrane (BioRad) at a continuous current of 100V for 90 minutes. For antibody staining, the membrane was blocked with 5% nonfat dry milk dissolved in TBS 1X containing 0.1% Tween-20 for 30 minutes and incubated with primary antibodies at a 1:500 dilution in the blocking solution overnight at 4° C. The membrane was then washed with TBST and incubated with the secondary antibody conjugated with HRP at a 1:10000 dilution in the blocking solution for 2 hours at room temperature. After washing with TBST, the blot was stained with BioRad chemiluminescence for 5 minutes and bands were visualized using a chemidoc (BioRad). Band intensities were quanti ed using ImageJ software (NIH, Bethesda, MD, USA) and normalized to GAPDH.
Experimental animals and EAE induction: All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC), Oakland University, Rochester, Michigan (IACUC #18081) and the Institutional Biosafety Committee (IBC), Oakland University, Rochester, Michigan (IBC #2858). A total of 25 C57BL/6J mice from Jackson Laboratory were used for this study. EAE was induced in 6 weeks old C57Bl/6 mice by subcutaneous immunization with 200 µg of MOG 33 − 35 peptide in complete Freud's adjuvant containing 2-5 mg killed mycobacterium tuberculosis H37Ra/mL emulsion (Hooke Laboratories, Lawrence, MA, USA). Mice were also injected intraperitoneally (i.p.) with pertussis toxin (100 ng, Hooke Laboratories) at the day of immunization and after 24 hours following the published protocol (16). Clinical score was blindly registered according to the following scale: 0: no clinical signs, 0.5: partial limp tail, 1: limp tail, 1.5: mild impaired righting, 2: severe impaired righting, 2.5: mild paresis of one hind limb, 3: severe paresis of one hind limb, 3.5: severe paresis of one hind limb and mild paresis of second hind limb, 4: paresis of two hind limbs, 4.5: paresis of both hind limb and partial fore limb, and 5: severe paralysis or death. Non-injected C57BL/6J mice were used as controls. All mice were maintained at the Oakland University Animal Facility under a 12/12-hour light and dark cycle.

Cell transplantation in an EAE mouse model
Cells were labeled with cell membrane labeling dye PKH26 (Sigma) following manufacturer's instructions (17). Confocal microscopy and FACS analysis were used to con rm the e ciency of the uorescently labeled cells prior to transplantation. There were eight treatments (n = 6 each), 1. Untreated healthy, 2, healthy + MSCs, 3. healthy + NSCs, 4. Untreated EAE, 5. EAE stage 1 + MSCs, 6. EAE stage 1 + NSCs, 7. EAE stage 2 + MSCs, and 8. EAE stage 2 + NSCs. The experiments were performed in triplicates. The mice were restrained in a rodent restrainer for cell transplantation. 10 6 primitive MSCs or NSCs were injected intravenously via the tail vein at either clinical scores of 1 or 2. The mice were monitored for 15 days following cell transplantation. Mice were euthanized humanly and CNS, lymphoid tissues (spleen, liver, brain, and spinal cord), and blood were extracted.

Tracking of transplanted cells
Whole tissue and blood samples was place in red blood cell lysis buffer for 5 minutes and then homogenized in a centrifuge tube. Culturing medium was then added to the tube, inverted, and centrifuged at 1300 rpm for 5 minutes at 4 o C. The supernatant was discarded, and cells were resuspended in PBS 1x. The cells were then ltered through a cell strainer. Labeled cells in each sample were determined using a FACS Canto II (Becton Dickinson) and Diva Software (Becton Dickinson).
Histological analysis: For para n embedding, brain and spinal cord tissue samples were xed with 4% paraformaldehyde for 24 hours. They were then dehydrated in 70% ethanol for 1 hour, 95% ethanol for 1 hour, 100% ethanol for 4 hours, xylene for 2 hours, and para n for 3 days. Tissue samples were then imbedded in para n and sectioned (5-10 µm thick) using a microtome. The para n embedded samples were stained with hematoxylin and eosin (H&E; Thermo Fisher Scienti c) to evaluate the cellular structure of the tissue and Luxol fast blue (LFB; Sigma) to examine myelination. To detect the protein levels of the human neural and in ammatory markers expressed by mouse brain and spinal cord tissue, the sections were stained with primary antibodies 1:100 dilution: TUJ1, NESTIN, MOG, MBP, GFAP, CD45, CD68, and CD3E, respectively, at 4 o C overnight followed by staining of secondary antibodies at 1:200 dilution at room temperature for 2 hours. They were counterstained with DAPI at 1:100 dilution for 5 minutes at room temperature. Fluorescent images of the stained sections were captured using a confocal microscope (NIKON Instruments Inc.).

Quanti cation of Treg and Th17 cells in the blood, spleen, and CNS tissues
For FACS analysis, whole tissue and blood samples were placed in red blood cell lysis buffer for 5 minutes and then homogenized. Culture medium was then added to the tube, inverted, and centrifuged at 1300 rpm for 5 minutes at 4 o C. The supernatant was discarded, and cells were resuspended in PBS 1x.
qRT-PCR analysis of cells and mouse tissue: Isolation of the total cellular mRNA was performed using the GeneJET RNA puri cation kit (Thermo Fisher Scienti c) following manufacturer's instructions. Total RNA was puri ed with DNase and incubated at 37 o C for 30 minutes using a thermocycler (Bio-Rad, Hercules, CA, USA). cDNA was synthesized using iScript kit (Bio-Rad). qRT-PCR was performed by using Sso-Advanced Universal SYBR Green Supermix Kit (Bio-Rad) on CFX96 Real-Time System (Bio-Rad). A 10 µL reaction was used which included 5 µL Sybr green, 3 µL of distilled water, 0.5 µL of forward primer, 0.5 µL of reverse primer, and 1 µL of 1:10 diluted cDNA. Each reaction was exposed to the following conditions: 98°C for 10 minutes, followed by 30 seconds of 98°C, 20 seconds of 60°C, and 30 seconds of 72°C for 44 cycles in 96-well optical reaction plates (Bio-Rad). Species speci c reference genes were used to normalize the targeted genes. Primer sequences are listed in Additional le 1.

Experimental design and statistical analysis
Data are presented as the mean ± standard error of the mean (SEM) of triplicates per analysis. Results with **p ≤ 0.01 were considered statistically signi cant. All analyses were performed using SPSS version 26 (SPSS Inc. USA). Statistical analysis were carried out by using one-way analysis of variance (ANOVA) test for multiple comparisons followed by post hoc tests and ANOVA for repeated measures followed by Tukey's post-hoc analysis.
Effect of cell transplantation in EAE mice. We then investigated the effects of MSCs and NSCs on the disease process in EAE mice. EAE was induced by myelin oligodendrocyte glycoprotein (MOG) peptide immunization in animals using an established protocol (16) and disease progression was followed by neurobehavioral analysis ( Fig. 2A) (19). The rst signs of EAE induction in animals were noticed 11-12 days after MOG injections. Then EAE stage 1 and 2 animals were intravenously injected with PKH26 labeled 10 6 cells and animals were evaluated for neurobehavioral changes twice a day until they were sacri ced for post-transplantation analysis. Neurobehavioral analyses results showed that average clinical score of the EAE control (4.5 ± 0.5) was signi cantly higher than the average score of mice transplanted with MSCs (1.7 ± 0.3) and NSCs (0.8 ± 0.1) at stage 1 and MSCs (2.4 ± 0.2) and NSCs (2.2 ± 0.1) at stage 2 ( Fig. 2B and Additional le 3). In addition, the average weight of the untreated EAE control (13g ± 0.6g) was signi cantly lower than the average weights of the mice transplanted with MSCs (15.7g ± 0. 7g) and NSCs (16g ± 0.4g) at stage 1 and MSCs (14g ± 0.7g) and NSCs (14.6g ± 0.7g) at stage 2 ( Fig. 2C and Additional le 3). When MSCs were transplanted at EAE stage 1, they slowed and halted the disease progression. However, NSCs not only slowed but also reversed the disease process, and animal conditions were improved to near normal levels. Overall, NSCs showed greater e cacy than MSCs when transplanted at EAE stage 1 than stage 2. Animals were sacri ced two weeks post-transplantation, and blood, lungs, lymphatic and CNS tissues were collected for histological, biochemical, immunological and molecular analyses.
Homing of transplanted cells to the blood, lymphatic, and CNS systems. The labeled transplanted cells homed to the blood, spleen, brain, and spinal cord (Fig. 3A). Only an insigni cant number of transplanted cells were found in the lungs of all animals, suggesting minimal lung trapping. Interestingly, signi cantly more transplanted MSCs (10.0%) than NSCs (6.7%) were observed in the blood of healthy animals, suggesting greater survivability of MSCs in the vascular system. In the blood of EAE stage 1 animals, the number of transplanted MSCs and NSCs was 9.3% and 10.3%, respectively. Strikingly, a very low number of MSCs and NSCs (1.2% and 2.0%, respectively) were found in the blood of EAE stage 2 animals. Nearly half of transplanted cells were found in the spleen in all animals. Importantly, a signi cant number of transplanted cells were present in the CNS of EAE animals but not in healthy controls. In the case of EAE stage 1 animals, a similar number of transplanted MSCs and NSCs were found in the brain (7.0% and 7.5%, respectively) but signi cantly more NSCs than MSCs (10.5% and 12.3%, respectively) were found in spinal cord. Overall, the number of transplanted labeled MSCs and NSCs found in the CNS of EAE stage 1 animals were more than Stage 2. Notably, NSCs were signi cantly more in the spinal cord than MSCs. Overall, NSCs exhibited greater survivability and/or proliferation in EAE stage 1 animals than MSCs. Improvement in the histopathology of the brain and spinal cord of EAE mice by transplanted cells. EAE induction mediates T cell activation resulting in increased in ltrates and demyelination as well as permeation of the BBB (20). The brain and spinal cord (Fig. 3B and 3D) showed presence of cell in ltrates in EAE animals but not in heathy controls. The number of in ltrates in both brain and spinal cord were signi cantly decreased in EAE stage 1 compared to stage 2 animals treated with MSCs and NSCs.
Furthermore, the LFB staining of the brain and spinal cord (Fig. 3C and 3E) was greatly improved in NSC treated animals compared to those treated with MSCs. The stain intensity was signi cantly increased in the cortex and spinal cord of EAE stage 1 (less in stage 2) animals treated with the cells suggesting endogenous remyelination. Taken together, these results suggest that transplanted NSCs not only reduced cell in ltrates, but also showed increased remyelination in the CNS at EAE stage 1 compared to the stage 2.
Modulation of the immune response by transplanted cells. The main effector cells of cell mediated immune response in EAE and MS are Treg (CD4 + and CD25+) and Th17 (CD4 + and IL17A+) (20). We investigated the level of these cells in the blood, spleen, and CNS by ow cytometry. Treg cells were signi cantly reduced to 8.6% and 7.3% in EAE animals from 35.6% and 43.0% in healthy controls in both the blood and spleen (Fig. 4A). In contrast, Treg cells were substantially increased in EAE animals transplanted with cells in the blood and spleen. In EAE stage 1 animals treated with MSCs, Treg cells were 27.4% and 41.1% in the blood and spleen, respectively. Whereas in EAE stage 2 animals treated with MSCs, Treg cells were 15.8% and 37.5% in the blood and spleen, respectively. On the other hand, in EAE stage 1 animals treated with NSCs, Treg cells were 39.3% and 42.0% in the blood and spleen, respectively. In EAE stage 2 animals treated with NSCs, Treg cells were 22.3% and 42.6% in the blood and spleen, respectively. In conclusion, Treg cell numbers increased in animals treated with both MSCs and NSCs; however, the improvements were more signi cant in the case of NSCs and their e ciency of restoring Treg levels was greater in EAE stage 1 than stage 2 animals. In the case of the CNS, Treg levels were 14.6% and 22.6%, respectively, in EAE animals. While cell transplantation did not affect the in ltrated Treg cells in the brain, Treg cells were signi cantly reduced to 14.8% and 10.0% in the spinal cord of EAE stage 1 animals treated with MSCs and NSCs, respectively. Cell transplantation in EAE stage 2 animals was less effective, as Treg cells reduced only to 17.0% in both MSCs and NSCs. Taken together, NSCs showed greater e cacy than MSCs in reducing Treg cells in the spinal cord in EAE stage 1 animals. It is known that Treg cells increase during the chronic late stages of EAE (21). So, reduced Treg levels in animals treated with cells could contribute to improved disease symptoms.
As expected, Th17 cells were signi cantly increased from 10.3% and 6.9% in the blood and spleen of healthy controls to 15.3% and 13.7% in EAE animals, respectively. When treated with MSCs, Th17 cell levels were not signi cantly affected in the blood (14.9%) but were reduced to normal levels (10.1%) in the spleen in EAE stage 1 animals; however, no effect was noticed in the case of EAE stage 2 (Fig. 4A). Importantly, NSC transplantation reduced Th17 in EAE stage 1 and stage 2 animals to levels similar to the healthy control animals. Both the brain and spinal cord of EAE animals had signi cantly high levels of Th17 cells (22.5% and 30.5%, respectively) as compared healthy control animals (1.6% and 0.9%, respectively). Upon MSC transplantation, Th17 cells were signi cantly reduced in the brain and spinal cord of EAE stage 1 animals (12.6% and 20.3%, respectively) but not in EAE stage 2 animals. On the other hand, NSC transplantation had a more prominent effect on reducing Th17 cells in both the brain and spinal cord in EAE stage 1 (11.5% and 13.7%, respectively) and in stage 2 (19.4% and 22.2%, respectively) animals. However, the NSC effect was less signi cant in EAE stage 2 animals.
Effect of transplanted cells on immune in ltrates in the CNS. To investigate the effect of transplanted cells on the in ammatory response in EAE mice, we analyzed the expression of CD45, CD68, and CD3E markers, representing leukocytes, macrophages, and T cells, respectively. The results depicted in Fig. 4B to 4D show high expression of the markers in the brain and spinal cord of EAE animals but not in healthy controls. Importantly, MSCs and NSCs were more effective in reducing the expression of all three markers in the spinal cord of EAE stage 1 but not stage 2 animals. Clearly, NSCs exert a greater anti-in ammatory response, particularly in EAE stage 1 animals potentially by reducing the pro-in ammatory cells, macrophages, leukocytes, and T cells.
Restoration of myelination and inhibition of astrogliosis by MSCs and NSCs. We next examined the effect of transplanted cells on the proteins associated with myelination and neural regeneration. We observed a signi cant decrease in the expression of MOG, MBP, TUJ1, and NESTIN in the brain of EAE animals ( Fig. 5A and 5C). The expression of MOG and MBP was signi cantly increased in the brain upon transplantation of NSCs in both stages of EAE. However, MSCs had a signi cant increase in EAE stage 1 animals only. Similarly, the expression of TUJ1 was signi cantly increased in the brain of both EAE stage 1 and stage 2 animals transplanted with NSCs, but this was only seen in MSC treated stage 1 animals. Upon induction of EAE, the expression of GFAP increased ( Fig. 5A and 5C), due to increased astrogliosis in the brain of EAE animals (22). However, GFAP expression was reduced in both EAE stage 1 and stage 2 animals transplanted with cells. The trend in the expression of MOG, MBP, TUJ1, and NESTIN in the spinal cord of EAE animals treated with cells was similar to the brain. However, NESTIN was signi cantly increased in EAE stage 1 animals treated with NSCs and the decrease of GFAP in the spinal cord of transplanted animals was not signi cant ( Fig. 5B and 5D). Overall, NSCs promoted greater remyelination and neural recovery as well as reduction of astrogliosis in the EAE stage 1 animals.
Effect of cells on the transcription for in ammatory, astrogliosis, neuroprotection, and myelination genes.
To determine the effects of transplanted cells on the expression of in ammatory, neuroprotection, and myelination genes, transcriptional analysis of the spleen, brain, and spinal cord tissues was performed.
As expected, higher expression of the in ammatory markers, Cd3e, Il-17a, Il-2, Il1β, Il-6, Tnfα, and Ifnγ was observed in the brain, spinal cord, and spleen, respectively, in EAE animals compared to the healthy controls ( Fig. 6A and 6B, Additional le 4, and Additional le 5). Expression of these markers was signi cantly reduced in the brain of MSC and NSC treated EAE stage 1 and stage 2 animals, except Il-17a in MSC treated Stage 2 animals. In contrast, expression of the in ammatory markers were signi cantly reduced in the spinal cord when NSCs and MSCs were transplanted at EAE stage 1 but not stage 2.
Similar to the brain, expression of these markers was signi cantly reduced in the spleen of animals transplanted with both types of cells at EAE stage 1 and stage 2, except for the Cd3e, Il-17a, Il-2, and Il-1β markers, in MSCs transplanted in EAE stage 2 animals.
Cytokine storm is elicited in MS. Although many cytokines take part in the "cytokine storm", IL-6 plays a crucial role. Dysregulated continual release of IL-6 has pleiotropic effects on chronic in ammation and autoimmunity, and also lead to the onset or development of various diseases including MS, acute respiratory distress syndrome, and multiple organ dysfunction syndrome (23). In EAE, IL-6 aggravates clinical symptoms and spinal cord pathology mainly by promoting pathogenic Th17 cells, which initiate and perpetuate in ammation and demyelination (24)(25)(26). IL-6 also affects autoreactive effector T cells in MS. IL-6 signaling supports T effector cell resistance to regulation by Treg cells, which may contribute to disease aggravation (27). Our results show that cell transplantation was effective at countering the induction of Il-6 gene expression in spleen and CNS tissues. In addition, NSCs showed more e cacy in regulating Treg cells when compared to MSCs, suggesting that NSCs are more potent in eliciting antiin ammatory effects than MSCs in the CNS.
As anticipated, the expression of neural markers, Tuj1 and Nestin, was downregulated in the CNS in EAE animals but was signi cantly increased in transplanted animals ( Fig. 6A and 6B and Additional le 5). Studies have shown that upregulation of Gfap and Ccnd1 is associated with reactivation of astrocytes (astrogliosis) (22,28). Overall, EAE stage 1 animals transplanted with NSCs displayed reduction of astrogliosis genes in the brain and spinal cord. Our results also suggest that expression of Gfap and Ccdn1 is likely to be regulated by the JAK/STAT pathway and promote the transcription of astrogliosis genes such as Gfap, Cntfrα, and N -b in EAE animals ( Fig. 6A and 6B and Additional le 5). Our results demonstrated that there was an upregulation of these astrogliosis genes via activation of the JAK/STAT pathway leading to an increase of astrogliosis in EAE mice. However, in EAE stage 1 animals treated with NSCs, these genes were signi cantly downregulated causing a decrease in astrogliosis in the CNS.
We determined that several paracrine neurotrophic factors were signi cantly upregulated in NSCs versus MSCs in vitro (Fig. 1E). Predictably, these factors contributed to the greater e cacy of NSCs for neuroprotection and the reversal of EAE symptoms. The CNS also showed higher expression of several neurotrophic factors such as Igf, Bdnf, and Trkb, which are pertinent to neuroprotection (Fig. 6A and 6B and Additional le 5). The fact that the expression of these factors is improved more signi cantly in EAE animals transplanted with NSCs, and that the animals recovered to near normal provide a 'proof of concept' that the use of NSC therapy in MS patients would more effective than using MSCs.
Factors controlling myelination have been reported to synergistically induce the expression of Krox-20, which is involved in activation of myelin genes such as Mpz and Mbp (29,30). Our results showed that several genes, Erk2, Krox-20, Oct-6, Mpz, Mbp, and Mog, associated with myelination were downregulated in EAE animals. Expression of these genes was signi cantly increased in the brain and spinal cord of EAE stage 1 animals injected with both types of cells, but NSCs provided a greater improvement compared to EAE stage 2 ( Fig. 6A and 6B and Additional le 5).
Mechanism of neural protection and regeneration. Our results showed downregulation in EAE animals but increased expression of Oct-6 and Krox-20 in transplanted animals, which suggests that the MAPK pathway may be involved in the regulation of these genes, because inhibition of MAPK signaling has shown to upregulate myelination associated genes (31). Chronic demyelination favors neurodegeneration of denuded axons, which causes irreversible neuronal de cits and disability in MS patients (32).
Furthermore, transplanted cells promoted neural survival and regeneration responsible for increased LFB stain intensity, indicative of an increase in remyelination of denuded axons. Taken together, results of this investigation suggest that NSCs in uence not only MAPK but also several other signaling pathways, particularly MAPK, JAK/STAT, and PI3K/AKT as illustrated in Fig. 6C.

Discussion
In the current study, we used highly proliferative MSCs isolated from a novel region of the umbilical cord tissue (9, 10), and differentiated them into a large amount of NSCs required for animal studies. The results of behavioral analysis showed that NSCs were more e cacious than MSCs when injected into animals with an early stage of EAE. While MSCs only halted the progression of disease symptoms, NSCs not only slowed but also reversed the disease process. Reversal of disease progress was parallel with weight gain that is typically lost due to EAE.
Although the PKH26 is a lipophilic tracer, previous studies have shown that the label has been traceable for more than 8 weeks following injection (17,33). In the current study, post-transplantation analysis showed that majority of the PKH26 labeled transplanted cells were found in the blood, spleen, and CNS in EAE stage 1 animals. However, there was a signi cant decrease of the number of cells found in the blood in EAE stage 2 animals transplanted with cells, possibly due to autophagy. It has been previously reported that regulation of autophagy plays an important role in providing neuroprotection via modulation of in ammatory and immune reaction in the CNS (34)(35)(36). Conceivably, cells transplanted at EAE stage 1 evaded autophagy induction, homed to the CNS and exhibited anti-in ammatory and neuroprotective characteristics.
In EAE mice, T cell activated cells travel through the BBB into the CNS leading to an increase of cell in ltrates and demyelination of CNS neurons (20). We also observed a signi cant increase of cell in ltrates in the brain and spinal cord of EAE animals as compared to the healthy control. However, when cells were transplanted at stage 1, there was a signi cant decrease of in ltrate cells. This could be due to the anti-in ammatory properties of MSCs and NSCs leading to a signi cant reduction of in ltrates (37). However, NSCs displayed a greater effect in reducing the number of in ltrates in the CNS. In addition, EAE mice transplanted with cells at stage 1 also had a greater effect on remyelination of the neurons in the CNS compared to stage 2 mice. These results are in agreement with previous studies where pharmacological drugs promoted remyelination of spinal cord neurons (38). Our approach signi cantly restored the lost myelin to near healthy control levels in the CNS, speci cally those transplanted with NSCs at stage 1.
Studies have shown that Treg and Th17 play an important role in the development of EAE and the immune response correlated with the disease (39)(40)(41). Treg maintains cell immune tolerance, whereas Th17 mediates the in ammatory response under abnormal conditions such as during an injury (40). We observed an imbalance between Treg and Th17 in EAE animals similar to previous reports (40). At the onset of EAE, the levels of Treg decreased and Th17 increased in the blood and spleen. However, after NSC transplantation at EAE stage 1, the imbalance was corrected and the levels of Treg and Th17 resembled those found in the blood and spleen of the healthy controls. Similarly, levels of Treg and Th17 were reduced in the CNS of animals transplanted with NSCs. The underlying mechanism for the restoration of the Treg/Th17 levels could be due to the transplanted cells modulating the function of T cells by inhibiting the production of TNFα and IFNγ, as reported previously (42)(43)(44). Therefore, we investigated the transcription of TNFα and IFNγ in EAE mice transplanted with cells. As expected, we found a reduction of these pro-in ammatory markers in the spleen and CNS tissues of animals transplanted with cells, further con rming the regulation of Treg and Th17 by transplanted cells. More importantly, the anti-in ammatory effect was found to be more prominent in animals transplanted with cells at EAE stage 1. Furthermore, the activity of leukocytes, macrophages, and T cells in the CNS were also decreased in animals transplanted with cells at EAE stage 1. Again, NSCs provided a greater affect in the reduction of these in ltrates in the spinal cord.
In order to elucidate the mechanism that might explain the therapeutic effects of MSCs and NSCs in EAE mice, we investigated astrogliosis, neuroprotection, and remyelination in the CNS. It is known that in chronic EAE, lesions become inactive resulting in accumulation of reactive astrocyte (GFAP+) laments in the CNS (45). We found that astrogliosis genes, Gfap and Ccnd1, were upregulated in EAE mice as well as GFAP + mouse cells, which were integrated into the brain and grey matter of the spinal cord. Upon cell transplantation, a signi cant decrease in the expression of astrogliosis genes in the CNS as well as a recession of GFAP + cells was found in the grey matter. These results agreed with previous reports showing the potential of MSCs and NSCs to reduce astrogliosis leading to an increase in neuroprotection and remyelination of neurons in the CNS (42,46,47).
MSCs and NSCs have the capability of promoting neuroprotection by secreting paracrine factors (BDNF) that can trigger the repair pathways (i.e. PI3K/AKT) (48)(49)(50). Our results showed an upregulation of the neurotrophic factors in EAE mice when transplanted with MSCs and NSCs at EAE stage 1. However, at EAE stage 2, despite insigni cant difference in the expression of neurotrophic factors, NSCs appeared to be more effective in providing neuroprotection than MSCs. These results correspond with the improvement of neurobehavior observed in the transplanted animals.
In MS, the demyelination is caused by myelin-speci c T cells that target the myelin on neurons in the CNS (51).
After entering the CNS, these cells are reactivated by local and in ltrating activated antigen-presenting cells, resulting in subsequent in ammation, then demyelination and axonal damage.
Our results also demonstrated that NSCs promoted remyelination in the CNS as indicated by LFB staining as reported in previous studies (52,53). Furthermore, upregulation of the genes involved in myelination, suggested that transplanted cells promoted remyelination in the CNS. Our results also suggest predictive mechanisms involved in the neuroprotection and remyelination of neuron via the activation of JAK/STAT, SMAD, MAPK, and PI3K pathways as depicted in Fig. 6C.

Conclusion
In brief, this study demonstrated the potential of NSCs to induce anti-in ammatory responses and provide neuroprotection as well as suggest promotion of endogenous neuroregeneration and remyelination of neurons denuded in EAE animals. Although, a number of clinical trials utilizing autologous bone marrow MSCs and umbilical cord derived MSCs, the results were not always conclusive.
In some cases, there was improvement in the quality of life measures one month following treatment; however, the participants reported similar levels of pain one year after treatment (54,55). Furthermore, most of these studies were limited to phase I and II trials due to limited availability of cells. Four clinical trials utilized bone marrow MSC derived NSCs, showing improvement of neurological disability (56). Two clinical trials used human fetal-derived NSCs (57); however, results of these phase I/II studies have not been reported. Again, these NSCs were limited in numbers their isolation from fetal tissue and use could pose ethical and moral concerns (58). While our results are highly promising, further studies should be pursued to determine if higher or multiple doses of cells injected at advanced stages of EAE could reverse the disease process. It would also be interesting to know if the cell treatment at the time of MOG injection could block induction of EAE in mice. Nevertheless, our study provided proof of the concept and basis for investigating the e cacy of NSCs derived from primitive MSCs to treat MS in clinical trials.

Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information les.
Gene expression was normalized to Gapdh and Actin and error bars represent the standard error of the