Intrastriatal Transplantation of Human Neural Stem Cells Restores the Impaired Subventricular Zone in Parkinsonian Mice


  • Fuxing Zuo,

    1. Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China
    Search for more papers by this author
  • Feng Xiong,

    1. State Key Laboratory of Medical Molecular Biology & Department of Immunology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, Beijing, China
    Search for more papers by this author
  • Xia Wang,

    1. State Key Laboratory of Medical Molecular Biology & Department of Immunology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, Beijing, China
    Search for more papers by this author
  • Xueyuan Li,

    1. Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China
    Search for more papers by this author
  • Renzhi Wang,

    Corresponding author
    1. Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China
    • Correspondence: Xinjie Bao, M.D., Ph.D., Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China. Telephone: +86-10-69152532; Fax: +86-10-69152532; E-mail:; or Wei Ge, Ph.D., State Key Laboratory of Medical Molecular Biology & Department of Immunology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, DongdanSantiao 5# Dongcheng District, Beijing 100005, China. Telephone: +86-10-69156470; Fax: +86-10-69156470; E-mail:; or Renzhi Wang, M.D., Ph.D., Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China. Telephone: +86-10-69152532. Fax: +86-10-69152532. E-mail:

    Search for more papers by this author
  • Wei Ge,

    Corresponding author
    1. State Key Laboratory of Medical Molecular Biology & Department of Immunology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, Beijing, China
    • Correspondence: Xinjie Bao, M.D., Ph.D., Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China. Telephone: +86-10-69152532; Fax: +86-10-69152532; E-mail:; or Wei Ge, Ph.D., State Key Laboratory of Medical Molecular Biology & Department of Immunology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, DongdanSantiao 5# Dongcheng District, Beijing 100005, China. Telephone: +86-10-69156470; Fax: +86-10-69156470; E-mail:; or Renzhi Wang, M.D., Ph.D., Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China. Telephone: +86-10-69152532. Fax: +86-10-69152532. E-mail:

    Search for more papers by this author
  • Xinjie Bao

    Corresponding author
    1. Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China
    • Correspondence: Xinjie Bao, M.D., Ph.D., Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China. Telephone: +86-10-69152532; Fax: +86-10-69152532; E-mail:; or Wei Ge, Ph.D., State Key Laboratory of Medical Molecular Biology & Department of Immunology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, DongdanSantiao 5# Dongcheng District, Beijing 100005, China. Telephone: +86-10-69156470; Fax: +86-10-69156470; E-mail:; or Renzhi Wang, M.D., Ph.D., Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China. Telephone: +86-10-69152532. Fax: +86-10-69152532. E-mail:

    Search for more papers by this author


Cell replacement therapy using neural stem cells (NSCs) transplantation has recently emerged as a promising method of Parkinson's disease (PD) treatment; however, the underlying mechanisms are not fully understood. To gain new insights into the mechanisms of 6-hydroxydopamine (6-OHDA)-induced lesioning and therapeutic efficacy of human NSCs (hNSCs) transplantation, the striatum (ST) of intrastriatal 6-OHDA-injected parkinsonian mice were unilaterally engrafted with undifferentiated hNSCs. A high-throughput quantitative proteomic approach was used to characterize the proteome profiles of PD-related brain regions such as the SN, ST, olfactory bulb, and subventricular zone (SVZ) in these mice. The abundance of more than 5,000 proteins in each region was determined with high confidence in this study, which is the most extensive proteomic study of PD mouse models to date. In addition to disruption of the DA system, the quantitative analysis demonstrated profound disturbance of the SVZ proteome after 6-OHDA insult. After hNSC engraftment, the SVZ proteome was restored and the astrocytes in the ST were greatly activated, accompanied by an increase in neurotrophic factors. Furthermore, bioinformatics analysis demonstrated that the changes in the proteome were not caused by the proliferation of hNSCs or their progeny, but rather by the reaction of endogenous stem cells. Overall, this study elucidates the unexpected role of SVZ cells in PD progress and treatment, thereby providing new therapeutic targets for PD. Stem Cells 2017

Significance Statement

The mechanisms underlying the beneficial effects of hNSCs transplantation on parkinsonian mouse are not well understood. Using high throughput proteomic approaches, Zuo et al. show that hNSCs functions through rescuing the function of SVZ and eliciting endogenous response rather than directly differentiating into dopaminergic neurons


Parkinson's disease (PD) is a progressive neurodegenerative disorder that occurs in approximately 1% of the population over the age of 65 [1]. A major hallmark of PD is the selective loss of dopamine (DA)-producing neurons in the substantia nigra (SN) and the consequent deficit in DA release in the striatum (ST) and other target areas [2]. As the etiology of PD remains uncertain, it is not yet possible to stop or slow its progress. Although behavioral recovery could be achieved by treatments with l-dopa or DA agonists, these are symptomatic treatments with considerable side effects and their effectiveness diminishes with time [3].

Recently, cell replacement therapy in PD by using human neural stem cells (hNSCs) has shown great promise in rodent models of PD. This therapy is based on the rationale that by transplantation, lost neurons can be replaced, disordered neurotransmission regulated, and function restored [4]. NSCs are self-renewing, multipotent cells with the capability to differentiate into any type of neural cell, including neurons and glial cells [5, 6]. Therefore, NSCs possess an intrinsic capacity to rescue dysfunctional neural pathways and are an attractive source for grafting and the development of novel therapies [7, 8].

The mechanisms underlying the effects of hNSC transplantation are yet to be elucidated. hNSCs can survive, proliferate, migrate in host brain, and produce functional effects but could barely differentiate to DAergic cells without genetic engineering. Engrafted hNSCs are more likely to exert beneficial effects by mediating the “niche” and producing neuroprotective, neurogenic, angiogenic, and anti-inflammatory factors in the brain [7, 9-12]. Thus, donor hNSCs may influence other brain regions beyond the injected locus through multiple pathways. However, the disadvantages associated with hNSC transplantation, including high variability and graft-induced dyskinesia, have limited its further development [13]. Better understanding the underlying mechanisms will pave the way for further development and clinical application of this cell-based PD therapy.

Global unbiased hypothesis-free “omic” approaches, such as proteomics and genomics, can comprehensively depict the status and dynamic changes in specific tissues or cell lines, enabling the identification of key molecules and pathways involved in diverse physiological and pathological processes [14]. Most biological processes are mediated by proteins; however, the levels and function of proteins are poorly predicted by genomic and transcriptomic analyses [15], particularly in the nervous system, where proteins can be transported along the axons over long distances, or accumulate under certain pathological circumstances [16]. Therefore, the direct study of the functional proteome has the most potential to provide information regarding cellular events. The development of high-throughput mass spectrometry and tandem mass tags (TMTs) has greatly improved the sensitivity and fidelity of quantitative proteomic analysis and has facilitated the application of proteomic approaches in neuroscience [17].

In this study, adult mice were subjected to 6-hydroxydopamine (6-OHDA) lesions followed by intrastriatal transplantation of hNSCs. The engrafted mice exhibited improved behavioral performance. The changes in the proteomes of different brain regions in engrafted and control mice were analyzed. We found that the protein profile of the subventricular zone (SVZ) was the most severely distorted by 6-OHDA treatment and was greatly normalized after hNSC transplantation. Bioinformatic and biochemical analysis revealed the pathways impaired by 6-OHDA as well as the responses to hNSC transplantation in the SVZ. Our results showed that hNSCs restore the constituents and functions of the SVZ and most likely exert beneficial effects by producing neurotrophic factors and stimulating endogenous reactions rather than directly replacing DAergic neurons.

Materials and Methods

6-Hydroxydopamine Lesioning

All experimental protocols were performed in accordance with guidelines issued by the committee on animal research of Peking Union Medical College Hospital and were approved by the institutional ethics committee. Adult C57BL/6J female mice (10-week-old) weighing 25–30 g were maintained in 12-hour light/dark cycles in cages and acclimated to the experimental environment for 1 week before modeling. The PD model was induced by 6-OHDA lesion [18-24]. In brief, mice were pretreated with desipramine (5 mg/kg; Sigma, St. Louis, MO, to block 6-OHDA uptake by noradrenergic terminals. Following adequate anesthesia, animals were secured onto a stereotactic frame (Stoelting, Wood Dale, IL). A solution of 6-OHDA (5 mg/ml in sterilized saline containing 0.02% ascorbic acid) was injected into the left ST via microliter syringe at an infusion rate of 0.5 µl/minute by micropump (Stoelting) for a total dose of 15 µg at coordinates anteroposterior (AP), +0.09 cm; mediolateral (ML), +0.22 cm; dorsoventral (DV), −0.25 cm relative to bregma. Following a wait of 2 minutes, the needle was withdrawn slowly.

hNSC Transplantation

Preparation of hNSCs

hNSCs were obtained from a NSC line derived from the human fetal brain (Angecon Biotech, Shanghai, China) and prepared for transplantation. All procedures were approved by the Human Experimentation and Ethics Committee. Briefly, primary dissociated single cell suspensions prepared from the cortex tissue of legally terminated human embryos (with the approval of the National Health and Family Planning Commission of the People's Republic of China) were incubated in serum-free NSC medium (Angecon Biotech) according to the manufacturer's instruction. A plating density of 100 cells per liter favored the establishment of neurospheres at 37°C with 5% CO2. Neurosphere cultures were digested with 0.05% trypsin-EDTA (Invitrogen, Carlsbad, CA, and subsequently treated with trypsin inhibitor (Roche, Mannheim, Baden-Württemberg, Germany, and were replated at 100 cells per liter every 7–10 days. On the day of transplantation, hNSCs were dissociated and rewashed before a final concentration adjustment.

Animal Sacrifice

At days 0, 7, 14, and 35 after 6-OHDA lesion and day 28 after cell engraftment, under deep anesthesia, animals were perfused intracardially with cold 0.1 M PBS followed by 4% paraformaldehyde in 0.1 M PBS (PH 7.4). Brains were carefully removed and postfixed at 4°C overnight, and cryoprotected in 30% sucrose for 3 days. A complete set of coronal sections were then cut through the ST as well as the SN at a thickness of 30 μm by using a freezing microtome (Leica, Nussloch, Baden-Württemberg, Germany, and stored at −20°C.

For enzyme-linked immunosorbent assay (ELISA) analysis, mice were sacrificed via cervical dislocation, and tissues of the ST and ventral mesencephalon were collected. To determine protein levels of basic fibroblast growth factor (bFGF) and glial cell line-derived neurotrophic factor (GDNF), homogenized tissues were tested using a mouse ELISA kit (Abcam, Cambridge, Cambridgeshire, U.K., Each assay was performed in accordance with the manufacturer's instructions. The concentrations of neurotrophic factors in the brain tissue were expressed as pg/ml total protein.

Cell Transplantation

Mice subjected to 6-OHDA treatment were randomly assigned to two groups: (a) the hNSCs-treated group (6-OHDA + hNSCs) and (b) the control group (6-OHDA + vehicle). Mice received unilateral intrastriatal transplantation of undifferentiated hNSCs or vehicle 7 days after 6-OHDA administration. Following anesthesia, animals were secured on a stereotactic frame (Stoelting) using an incisor clamp and two ear bars. A midline rostro-caudal incision was created, and one burr hole was formed unilaterally at the site based on coordinates relative to the bregma (AP, +0.06 cm; ML, +0.20 cm) by using a high-speed drill. Each animal in the treated group received 1 µl of cell suspension at the following coordinates: DV, −0.22 and −0.28 cm, relative to the dura. Approximately 5 × 104 hNSCs in 0.5 µl PBS or an equal volume of vehicle was transplanted at two sites by microsyringe at an infusion rate of 0.1 µl/minute for a total dose of 1 × 105 per microliter. The needle was withdrawn slowly following a wait time of 2 minutes. Animals were returned to a temperature-controlled blanket until they recovered from anesthesia.

Behavioral Tests


The Rotarod test was performed on days 14 and 28 after 6-OHDA injection. All mice were pretrained for about 1 week before the tests with rods of approximately 3 cm diameter. The training consisted of three consecutive runs at 26 rpm, until the mice were unable to maintain themselves for 300 seconds on the rotating rod, and lasted for 3 days. One day prior to 6-OHDA administration, a baseline measurement was obtained under the same conditions. Subsequently, test series started at a speed of 26 rpm. The time to fall of each mouse was recorded.

Apomorphine-Induced Rotational Asymmetry

Mice were tested for rotational behavior induced by apomorphine on days 3, 7, 14, 28, 42, and 56 after hNSC transplantation, by using an automated rotometry system (San Diego Instruments, San Diego, CA) as previously described [18, 19]. Apomorphine hydrochloride (5 mg/kg; Sigma) in saline was injected intraperitoneally, and results were expressed as total number of rotations away from the lesion side in 30 minutes.

Morris Water Maze

The modified Morris water maze was used to investigate spatial learning and memory and was performed in a circular tank (100 cm in diameter) filled with water (20 ± 1°C) made opaque by the addition of milk to obscure an escape platform [20, 21]. The tank was divided into four quadrants, and the visible escape platform was centered in one of these quadrants prior to hidden platform test. During training sessions, mice were allowed to swim to the visible escape platform. Subsequently, the platform was left in the same position and submerged 1 cm below the water surface, and each mouse was subjected to two trials per day for 5 consecutive days starting from 28 days after hNSC transplantation. The escape latencies required for each mouse to locate the goal platform were recorded and averaged. One day following the last trial of the hidden platform test, the probe test was performed in the same tank without the escape platform. Mice swam freely in search of the quadrant of the original platform within 1 minute. The time and swimming distance spent in the correct quadrant as well as the frequency of target platform crossings were recorded. Monitoring was performed using an overhead camera connected to a video tracking system (Muromachi Kikai, Chuo-ku, Tokyo, Japan).

Micro-Positron Emission Tomography and Magnetic Resonance Imaging Scans

To assess the glucose metabolic activity in the ST, mice were examined using a 18F-FDG micro-positron emission tomography (PET) scanner at 1 month after hNSC transplantation [22, 23]. Mice were deprived of food for 24 hours prior to Micro-PET scans to ensure the baseline of plasma glucose. In brief, 18F-FDG was produced with a RDS111 cyclotron and the corresponding radiochemical synthesis system (CTI Company, Knoxville, TN) in Peking Union Medical College Hospital. Mice were anesthetized and 18F-FDG (0.4 mCi in a maximum volume of 0.3 mL saline) was injected intraperitoneally. Each animal was scanned with an Inveon dedicated micro-PET system (Siemens Medical Solutions, Knoxville, TN) 30 minutes after the bolus injection of 18F-FDG. All static acquisition data were sorted into three-dimensional sinograms. For visualization and quantitative data analysis, images were reconstructed using a statistical maximum a posteriori probability algorithm and analyzed using ASIPRO VM software (Concorde Microsystems, Knoxville, TN). Subsequently, the brain structure of each animal was scanned using a 7.0 T magnetic resonance imaging (MRI) scanner (Siemens Medical Solutions, Knoxville, TN) [22]. The MRI sequences included coronal scout and T2-weighted imaging (T2WI). T2WI were obtained by a fast spin echo sequence with the following imaging parameters: repetition time = 5,000.0 milliseconds, effective echo time = 36.0 milliseconds, echo train length = 8. Then, each individual 18F-FDG PET image was coregistered to its corresponding T2WI using MATLAB (Mathworks, Sherborn, MA). PET tracer uptake in the region of interest in the lesion ST was evaluated as relative metabolic activity (ratio of the lesion side compared to the intact side) [23].

ELISA Analysis

Mice were sacrificed via cervical dislocation, and tissues from the ST and SN were collected. Homogenized samples with high protein content were subjected to ELISA using a mouse ELISA kit (Abcam) to determine protein concentrations of GDNF, bFGF, interleukin 2 (IL-2), IL-1β, IL-10, and tumor necrosis factor-α (TNF-α). Each assay was performed in accordance with the manufacturer's instructions. The concentrations in the brain tissues were expressed as pg/ml total protein.

Western Blot

Western blotting is a traditional technique used to identify and partially quantify target proteins. Firstly, we determined the protein concentration by using a Nanodrop 2000 spectrophotometer (Thermo Scientific) and loaded 20 μg of protein into 10% polyacrylamide gels for SDS-PAGE. Gels were run according to standard methods, and the target proteins were electrophoretically transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat dry milk in TBS-T (TBS plus 0.5% Tween) for 30 minutes. The membranes were then incubated with primary antibodies (1:10,000 for α-synuclein (610786, BD Transduction Laboratories) and Gapdh (M171-3, MBI); 1:1,000 for Drd1 (ab81296, Abcam), Dat (sc-14002, Santa Cruze), and Cnp (ab6319, Abcam) overnight at 4°C. Membranes were then washed and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies and ECL reagents.

Immunohistochemical Staining and Analyses

The standard two-step immunohistochemical (IHC) assay was performed on frozen sections using polink-2 plus polymer HRP detection kit. Tissue sections (10 µm thick) were stained with commercially available monoclonal antibodies against Gfap, Th, Psd95, Ki67, Olig2, NeuN, and Iba1. Sections were treated with 3% H2O2 to quench the endogenous peroxidase activity. Three sequential washes (5 minutes each) were performed following each incubation step. Specific polyclonal antibodies were diluted 1:400 and applied for overnight incubation at 4°C. After extensive washes, sections were incubated with polymer enzyme conjugate for 30 minutes at room temperature. Sections were incubated with HRP labeled conjugates for 30 minutes at room temperature. The peroxidase was then developed in 3,3'-diaminobendizine (DAB) with H2O2. The sections were counterstained with haematoxylin, dehydrated in graded ethanol, cleared in xylene, and photographed using a microscopic system (OlympusDP72). Statistical analysis was performed using Image ProPlus. A p-value of less than 0.05 (p < 0.05) was considered statistically significant.

Proteomic Analysis

Sample Preparation

The brain tissues were prepared using modified standard autopsy procedures and stored under −80°C cryopreservation prior to use. The brain tissues were homogenized with ice-cold homogenization buffer, composed of 8 M urea in PBS, pH 8.0, 1× cocktail, and 1 mM phenylmethanesulfonyl fluoride. The homogenate was centrifuged at 12,000 rpm at 4°C for 15 minutes to remove cell debris, and the clarified supernatant was then transferred into a new 1.5-ml tube. The protein concentration was measured using a Nanodrop 2000 (Thermo Scientific) according to the manufacturer's instructions. The protein concentration was used to determine the volume of supernatant required to obtain 25 μg of protein from each sample. Samples were grouped based on brain region and treatment (Intact, 6-OHDA, or hNSCs, as described in 6-hydroxydopamine lesioning section), and the samples in the same groups were mixed together to obtain three pooled extracts, each containing 100 µg of protein.

TMT Labeling

The extracts were treated with 10 mM dithiothreitol for 1 hour at 55°C and then with 25 mM iodoacetamide for 30 minutes in the dark at room temperature. They were then digested with 1:100 w/w endopeptidase Lys-C overnight at 37°C. The extracts were then diluted with PBS (pH 8.0) to a final urea concentration of 1.0 M and then digested with 1:50 w/w trypsin overnight. The next morning, the extracts were first acidified with 1% formic acid, and then desalted with a reverse-phase column (Oasis HLB; Waters, MC). The extracts were dried with a vacuum concentrator and finally dissolved in 200 mM triethylammonium bicarbonate buffer for labeling with TMT reagents. Two sets of TMT labels were used to label different groups: set1: TMT-126 for ST-Intact, TMT-127 for ST-6-OHDA, TMT-128 for ST-hNSCs, TMT-129 for SN-Intact, TMT-130 for SN-6-OHDA, TMT-131 for SN-hNSCs; set2: TMT-126 for OB-Intact, TMT-127 for OB-6-OHDA, TMT-128 for OB-hNSCs, TMT-129 for SVZ-Intact, TMT-130 for SVZ-6-OHDA, TMT-131 for SVZ-hNSCs. The prepared TMT reagents (0.8 mg TMT dissolved in 40 µl of 99.9% acetonitrile) were added to the corresponding solutions and the labeling reaction was allowed to progress at room temperature. The reactions were stopped 1 hour later by the addition of 5 µl of 5% hydroxylamine for 5 minutes. All of the labeled extracts from the three pools were mixed, desalted, dried as previously described, and dissolved in 100 µl of 0.1% formic acid.

High Performance Liquid Chromatography Separation

The labeled peptides were fractionated. The TMT-labeled peptides (100 µl in 0.1% formic acid) were first transferred to MS tubes for high performance liquid chromatography (HPLC) (UltiMate 3000 UHPLC, Thermo Scientific) analysis. Subsequently, an Xbridge BEH300 C18 column (4.6 mm × 250 mm, 2.5 µm, Waters) maintained at 45°C with a flow rate of 1.0 ml/minute was used for further fractionation. UV absorbance was detected at 214 nm. Fractions were collected every 1.5 minute into 50 tubes and were dissolved with 20 µl of 0.1% formic acid for liquid chromatography (LC)-MS/MS analysis.

Peptide Analysis by LC-MS/MS

LC-MS/MS analysis was conducted on a Q Exactive mass spectrometer (Thermo Scientific). The fractions were separated with a 120-minute gradient elution at a flow rate of 0.30 μl/minute. The UltiMate 3000 RSLCnano System (Thermo Scientific) was interfaced with the Thermo Q Exactive Benchtop mass spectrometer (Thermo Scientific). The column used for analysis was a silica capillary column (75 μm ID, 150 mm length; Upchurch, Oak Harbor, WA) packed with C18 resin (300 Å, 5 µm; Varian, Lexington, MA). To set the Q Exactive mass spectrometer in data-dependent acquisition mode, we used Xcalibur 2.1.2 software for manipulation. Using Orbitrap (400–1,800 m/z, 60,000 resolution), the single full-scan mass spectrum was conducted before 10 data-dependent MS/MS scans at 27% normalized collision energy (higher energy C-trap dissociation). Proteome Discoverer 2.1 software (Thermo Scientific) was used for searching MS/MS spectra against mouse or human proteome databases from UniProt (released February 29, 2016) with a false discovery rate (FDR) equals to 0.01 and for estimating the absolute and relative abundance of proteins. The criteria for searching was set according to the software recommendations with the following modifications: full tryptic specificity with no more than two missed cleavages permitted; set carbamidomethylation (C, +57.021 Da) and TMT plex (lysine, K and any N-terminal) as static modifications; and oxidation (methionine, M) as a dynamic modification. The fragment ion mass tolerance was corrected to 20 mmu (all MS2 spectra), and the precursor ion mass tolerances were set at 20 ppm (all MS in an Orbitrap mass analyzer). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD004152 [25].

Bioinformatics Analysis

Analysis of proteomic results based on R 3.2.4. Protein annotation was performed with the R package and package. GO term statistics were analyzed with the topGO R package according to the classic method [26]. Fisher's exact test was used to determine the p-value of enriched terms.

Statistical Analysis

All data were presented as the mean ± SEM and analyzed using GraphPad Prism 6.0. The difference between two groups was analyzed by two-tailed Student's t test, while two-way ANOVA followed by Tukey's post hoc analysis were conducted for multiple comparisons between two or more groups. A p value <.05 was defined as the threshold for statistical significance.


hNSC Transplantation Promoted Functional Recovery in the 6-OHDA-Treated Mice

To evaluate the potential effects of hNSCs on Parkinson's disease, hNSCs were transplanted into 6-OHDA-induced parkinsonian mice followed by multiple assessments at various time points. (Fig. 1A) The details of experiments performed are listed in Supporting Information Table 1. The results showed significant 6-OHDA-induced lesioning at day 7, which was aggravated at later time points (Supporting Information Fig 1A–1D). To determine whether hNSC engraftment attenuated 6-OHDA-induced motor deficits, apomorphine-induced rotation and Rotarod tests were conducted. No difference was observed before cell transplantation. At 28 days after hNSC transplantation, significant reduction in the number of rotations and increase in duration on Rotarod were observed in grafted mice compared with the vehicle administered control group (Fig. 1B, 1C). The effect of hNSCs on nonmotor symptoms was then assessed by using the Morris water maze test with hidden platform; grafted mice displayed significantly shorter escape latency than did the sham-treated animals for 5 consecutive days after day 28 (Fig. 1D). In the probe test, wherein the platform was removed, hNSCs-engrafted mice spent a higher percentage of time in the target quadrant, with longer distance and higher frequency of crossing the exact location of the platform than the controls (Fig. 1E–1G). These results indicated that spatial learning and memory deficit caused by 6-OHDA were ameliorated by hNSCs.

Figure 1.

hNSC transplantation promoted functional recovery in the 6-OHDA-treated mice. (A): Scheme of experimental design and workflow. In brief, 7 days after unilateral 6-OHDA lesion, hNSCs were transplanted into lesioned mice. The effect of transplantation was assessed by behavioral tasks at various time points and histological and proteomic analysis at day 28. (B): Significant attenuation of side-biased rotational behavior in the apomorphine-induced rotation test was observed in hNSCs-treated mice compared to the control animals from 28 days after cell engraftment. (C): In the rotarod test, mice that received hNSCs displayed significantly increased latency on the rotating rod at day 28 compared to the animals treated with vehicle. (D): In the hidden platform test, escape latency was dramatically shorter in grafted mice than in the control animals from day 28. (E–G): In the probe test at day 28, mice treated with hNSCs exhibited higher percentages of distance travelled (E) and time spent (F) in the correct quadrant, accompanied by higher frequency of crossing within the platform quadrant (G) than animals given vehicle. Data in (B–D) represent mean ± SEM (*, p < .05; two-way ANOVA). Data in (E–G) are expressed as mean ± SEM (*, p < .05; two-tailed Student's t test). Abbreviations: hNSCs, human neural stem cells; 6-OHDA, 6-hydroxydopamine.

Post-hNSC-transplantation, glucose metabolism was determined by 18F-fluorodeoxyglucose (18F-FDG) imaging using micro-PET, and the 18F-FDG summation images were coregistered to the mouse MR imaging atlas for better illustration (Fig. 2A). In the 6-OHDA-lesioned ST, 18F-FDG binding was significantly decreased compared with that in the intact hemisphere, indicating a pronounced reduction in glucose metabolic activity. However, 18F-FDG uptake was normalized to intact levels 1 month after engraftment, indicative of active glucose metabolism due to the transplanted hNSCs. Quantitative measurements of 18F-FDG uptake showed that the impaired metabolism induced by 6-OHDA was substantially upregulated by hNSCs to an extent comparable to that of the intact contralateral side (Fig. 2B).

Figure 2.

Transplantation of hNSCs enhanced glucose metabolic activity following 6-OHDA administration. (A): Coronal micro-PET scans for 18F-FDG projected on MRI (gray scale) through the striatum showed an increase in radioactivity concentrations on the lesion side of mice receiving hNSCs compared to those of animals given vehicle at day 28 after transplantation. Quantitative measurements of 18F-FDG uptake values (B) revealed significantly higher glucose metabolism in the striatum of hNSCs-treated mice than in the control animals. The results were expressed as ratio of the lesion side to the intact side. Data are expressed as mean ± SEM (*, p < .05; two-tailed Student's t test). 18FDG, 18F-fluorodeoxyglucose; hNSCs, human neural stem cells; 6-OHDA, 6-hydroxydopamine; MRI, magnetic resonance imaging; PET, positron emission tomography; T2WI, T2-weighted imaging.

Characterization of the Protein Profiles of Multiple Brain Region

To investigate the mechanisms underlying 6-OHDA-induced lesioning and the neuroprotective effect of hNSCs, we used a HPLC-coupled high-throughput mass spectrometry approach to quantitatively characterize the proteome in the brain. At day 28 after transplantation, four PD-relevant brain regions, the SN, ST, OB, and SVZ, were subjected to proteomic analysis. Simultaneous identification and quantification of proteomes of four brain regions from three mouse groups (Intact, 6-OHDA, hNSCs) were achieved by using two sets of six-plex TMT (TMT 6-plex). Proteins sharing the same set or subset of peptides were considered as a single entry. With a threshold of FDR <1% and confidence score more than 10, we identified 4,788, 4,794, 5,389, and 5,403 proteins in the ST, SN, OB, and SVZ, respectively. Among these proteins, 6,361 were found in at least one brain area and 4,637 were found in all brain areas (Fig. 3A). The most abundant proteins in each region are listed in Figure 3B.

Figure 3.

Characterization of the protein profiles of multiple brain region. (A): The number of proteins unique to each region or shared by different regions is shown in a Venn diagram. (B): The gene names of the 10 most abundant proteins in each region are listed. (C): GO analysis was performed to illustrate the total brain protein distribution pattern based on cellular component categories. Abbreviations: GO, gene ontology; OB, olfactory bulb; SN, substantia nigra; ST, striatum; SVZ, subventricular zone.

To assess the proteome composition of the identified proteins, we annotated all the proteins to cellular component (CC) gene ontology (GO) terms. The largest proportion of proteins (73.16%) was annotated as cytoplasmic with a slightly preferential distribution in “mitochondrion” (17.82%), “cytoskeleton” (14.13%), and “cytosol” (13.77%). Approximately 13.75% of annotated proteins corresponded to neuron-specific GO terms such as “synapse” (8.42%), “dendrite” (5.31%), and “neuronal cell body” (5.15%; Fig. 3C).

Proteins in the SVZ Were Most Affected in the 6-OHDA-Treated Mouse

After normalization of total peptide intensity of each TMT group, the relative intensity of peptides was integrated and calculated to determine the relative abundance of proteins. The most upregulated or downregulated proteins after 6-OHDA injection in each brain region are listed in Supporting Information Table 2 . As expected, dopamine transporter (DAT, Gene name Slc6a3) and tyrosine hydroxylase (Th) were the most downregulated proteins in the ST, indicating the success of the 6-OHDA-induced parkinsonian model. Of note, the reduction of DAT in the SVZ was larger than that in the ST, indicating that loss of DAergic neurons may have a greater impact on cells in the SVZ.

Indeed, the proteome profile of the SVZ exhibited more severe alteration than that of any other region after 6-OHDA lesion. The number of changed proteins was substantially higher in the SVZ than in other brain areas, suggesting the impairment of multiple molecular pathways rather than selective disruption of DAergic innervation (Fig. 4A). To investigate the influence of 6-OHDA on different regions, the change of 4,637 proteins that were identified in all regions were compared and clustered (Fig. 4B). The changes in proteome profile in each brain region did not show strong correlations with each other, indicating that the biological processes occurring in these four brain regions after 6-OHDA injection were different from each other. Notably, most of the proteins that were altered in the SVZ did not show a pronounced change in the ST, which is spatially adjacent to the SVZ. Thus, 6-OHDA elicited distinct responses in the SVZ and ST, and the possibility of contamination during sample preparation is very low (Fig. 4B).

Figure 4.

Proteins in the SVZ were most affected in the 6-hydroxydopamine (6-OHDA)-treated mouse. (A): The relative abundance of proteins in 6-OHDA-treated mice compared to that in control mice shown as a histogram. For better representation, logged ratios <–2 or >2 are not shown. (B): Changed proteins in the SVZ were different from those in other brain regions. 4,637 proteins that were detected in all four brain regions were analyzed and clustered according to the relative protein abundance after 6-OHDA treatment. (C, D): Enrichment of significantly changed proteins in most overrepresented GO CC, BP, and MF are shown. Abbreviations: BP, biological pathway; CC, cellular component; MF, molecular function; OB, olfactory bulb; SN, substantia nigra; ST, striatum; SVZ, subventricular zone.

Proteins that changed more than 20.5 and less than 2−0.5 fold due to 6-OHDA treatment in the SVZ were defined as upregulated and downregulated proteins, respectively, and were subjected to GO analysis. The overrepresented CC, BP (biological pathway), and MF (molecular function) GO categories of significantly changed (SC) SVZ proteins were assessed by comparison with the complete SVZ proteome annotation. The significance and the number of annotated proteins in the most overrepresented categories in each class are shown in Figure 4C. CC GO analysis demonstrated that the upregulated proteins were mostly enriched in the ribosomal subunits, especially the ribosomes in the mitochondria. Consistently, the MFs of these proteins were structural constituent of ribosome and nucleic acid binding. Enriched BP and GO categories included cellular amide metabolic processes and translation, which are all involved in amino acid and protein synthesis, and gene expression was the most enriched BP category among the upregulated proteins (Fig. 4C). In summary, the biosynthesis systems in SVZ cells, ranging from transcription, amino acid synthesis, and protein synthesis to mitochondrion replication, were generally enhanced.

The GO analysis of the downregulated proteins displayed significant enrichment in plasma membrane components; especially protein constituents of the endoplasmic reticulum and synapse (Fig. 4D). BP and MF GO analysis suggested that these proteins mainly function as transmembrane ion transporters and neurotransmitter receptors. These results clearly displayed the dramatic loss of synaptic structure and plasma membrane systems. It is therefore likely that the innervation and differentiation of NSCs in the SVZ was drastically reduced. Taken together, the results of analysis of significantly changed proteins in the SVZ demonstrated that 6-OHDA lesioning affected the cell fate decision of SVZ NSCs, which were prevented from differentiating into functional neurons.

To demonstrate alteration of SVZ histologically, we investigated the alteration by immunostaining using antibodies against various cell and structure markers (Supporting Information Fig. 3A, 3B). DAergic denervation in SVZ was evident at 7 days after 6-OHDA lesion. Surprisingly, the number of endogenous NSCs, which were marked by Gfap, increased after lesion, suggesting inhibition of differentiation.

The Protein Profile of the SVZ Was Almost Completely Restored by hNSC Transplantation

To demonstrate the effect of hNSC transplantation, the changes in the proteome profile of mouse brains with or without hNSC engraftment were compared. Considering that low-abundance proteins are more susceptible to change due to environmental stimulation, the relative abundance of proteins in the 6-OHDA and hNSCs groups was analyzed along with the absolute abundance of proteins in the intact brain. Proteins with relative abundances of more than 20.5 or less than 2−0.5 fold in the 6-OHDA group were defined as SC-proteins. The results showed that there were not only considerably more SC-proteins in the SVZ (>20% of total proteins) than in other brain regions (less than 5% of total proteins), but that proteins were also quite distinct in their expression level. In other brain regions, the majority of SC-proteins were low-abundance proteins such as enzymes, kinases, and signaling molecules involved in regulating specific cell functions. In contrast, the SC-proteins in SVZ covered the complete abundance range, and many of these proteins were important structural proteins, which are highly expressed in the central nervous system, including brain acid soluble protein 1, glial fibrillary acidic protein (GFAP), and 2',3'-cyclic-nucleotide 3'-phosphodiesterase (CNP) (Fig. 5A, left panel). Thus, the SVZ showed substantial cellular component changes rather than simple cell loss or lowered cell function.

Figure 5.

The protein profile of the SVZ was almost completely restored by hNSC transplantation. (A): Fold changes of proteins with or without hNSC treatment are depicted against the absolute abundance in intact group. Proteins with a fold change <2−0.5 or >20.5 in the 6-OHDA group were defined as significantly changed proteins (SC-protein) and are represented as red dots; other proteins were defined as nonsignificantly changed proteins (NC-protein) and are represented as grey dots. SC-proteins that were restored by hNSC treatment are represented as green dots in the right panel. The proportions of upregulated, unchanged, and downregulated proteins are shown at the top of each plot. (B): The level of representative PD-related proteins was analyzed, and Gapdh was used as a control. Mean ± SEM of all detected unique peptides associated with the indicated proteins were shown and compared. (*, p < .05; **, p < .05; two-tailed Student's t test). (C): The changes in PD-related proteins were confirmed by Western blot. The relative abundance of proteins was determined by band density. Abbreviations: hNSCs, human neural stem cells; 6-OHDA, 6-hydroxydopamine; OB, olfactory bulb; SN, substantia nigra; ST, striatum; SVZ, subventicular zone.

Strikingly, after hNSC transplantation, most SC-proteins in the SVZ were restored to levels comparable to those in intact tissue and only 3% of the total proteins were dysregulated. Moreover, this extensive restoration was confined to the SVZ, as SC-proteins in other brain region were barely restored and the proteome profiles changed even more (Fig. 5A, right panel). The abundance of some representative PD-associated proteins (Th, Drd1, Dat) and control proteins (Cnp and Gapdh) are shown in Figure 5B, which clearly demonstrated the effect of hNSC transplantation. To confirm our mass spectrometry results, samples used in mass spectrometry were also subjected to western blot analysis. With Gapdh as the background reference protein, the fluctuation of these proteins in different groups and brain regions was highly consistent with the proteomics data (Fig. 5C). In the SVZ, α-synuclein was significantly upregulated by 6-OHDA treatment and reduced by hNSC transplantation, whereas the myelin structural protein Cnp and the DAergic terminal proteins Drd1 and Dat1 were reduced by 6-OHDA treatment and rescued by hNSC transplantation. However, hNSC transplantation did not have much influence on the changed proteins in the ST or SN (Fig. 5B, 5C). Meanwhile, the density of TH neurons in SN was also not substantially restored, suggesting that the effects of hNSCs were not mainly caused by replacement of the lost DAergic neurons (Supporting Information Fig. 2).

hNSC Transplantation Elicited Endogenous Responses and Stimulated the Production of Neural Protective Factors

Because the SVZ is one of only two brain regions that NSCs reside in, we then sought to determine the type and state of cells in the SVZ. We focused on specific marker proteins to identify different cell types, including neurons, astrocytes, oligodendrocytes, and microglia. For each cell type, the relative abundance of three cell marker proteins were compared. In the SVZ, the makers for astrocytes (Gfap, Prdx6, and Pvalb), neurons (Neurofilament, Synaptophysin, and Spock1), and oligodendrocytes (Cnp, Mog, and Plp1) were decreased by 6-OHDA lesion and restored by hNSC engraftment. The markers for microglia were not affected. The consistency of changing patterns of marker proteins of the same cell types indicated that the number of cells rather than the expression level of proteins were regulated. However, no significant changes were observed in other brain regions, including the ST where the hNSCs were transplanted (Fig. 6A). The results suggested that astrocytes, which were thought to possess the features of stem cells, accumulated in the SVZ after 6-OHDA lesion, accompanied by a decrease of neurons and oligodendrocytes that might be caused by DAergic denervation or interruption of endogenous NSC differentiation. The changes exhibited in proteomic analysis were further confirmed by IHC using corresponding antibodies. The results clearly showed the accumulation of Gfap-expressing NSCs and DAergic denervation in the SVZ in 6-OHDA group and the restorative effect of hNSCs. (Fig. 6B; Supporting Information Fig. 3C) To determine whether the change of cell number directly resulted from engrafted hNSCs or was an endogenous response, we analyzed the origins of the cell marker proteins. As a result of the sequence similarity between donor and host proteins and the fact that donor cells only comprise a small fraction of the sampled tissues, the abundance proteins derived from hNSCs could not be determined by mass spectrometry directly by searching against human protein databases. However, hNSC transplantation would influence the determination of the abundance of endogenous proteins, as the sequences derived from hNSCs but shared by human and mouse would also be counted when searching against mouse protein databases. Thus, we speculated that, for any protein, peptides specific in mouse (specific peptides), and peptides shared by both human and mouse (shared peptides) would exhibit different expression patterns after transplantation, if the protein was also expressed by hNSCs or their progeny. Therefore, the peptides of these marker proteins were classified, and the total abundance ratio of each peptide and group was calculated separately (Fig. 6C). The pattern of specific peptides and shared peptides of all three marker proteins (Gfap for astrocytes, neurofilament for neurons, and Cnp for oligodendrocytes) were nearly parallel, suggesting that the restoration of the SVZ proteome by hNSC transplantation, although dramatic, was most likely due to the endogenous responses, and not the contents of the hNSCs or their progeny.

Figure 6.

hNSC transplantation altered cell constituent and stimulated the production of neural protective factors. (A): Characterization of cell marker proteins. For each cell type, three marker proteins were chosen as indicated at the right of each panel. The markers for astrocytes, neurons, and oligodendrocytes were decreased by 6-OHDA lesion and restored by hNSC transplantation. (B): At day 28 after hNSC transplantation, the cellular component changes in the SVZ were investigated by immunohistochemistry using the indicated antibodies. Quantifications and stereological analysis of signal intensities are shown on the right side of each panel. Representative figures of five slides are shown. Scale bar = 20 μm. (C): The pattern of peptides specific in mouse (specific peptides) and peptides shared by both human and mouse (shared peptides) of three marker proteins (Gfap for astrocytes, Nefm for neurons, and Cnp for oligodendrocytes). (D): The protein levels of bFGF and GDNF in the ST and SN after hNSC transplantation were determined by ELISA. Data are expressed as mean ± SEM of 5–9 mice per group (*, p < .05; two-tailed Student's t test). Abbreviations: bFGF, basic fibroblast growth factor; ELISA, enzyme-linked immunosorbent assay; GDNF, glial cell line-derived neurotrophic factor; hNSCs, human neural stem cells; 6-OHDA, 6-hydroxydopamine; OB, olfactory bulb; SN, substantia nigra; ST, striatum; SVZ, subventricular zone.

Based on these results, we studied the neurotrophic factors reported to suppress neuroinflammation and improve the resistance of DAergic neurons [27-29]. The protein and mRNA levels of bFGF and GDNF in the ST and SN were determined by ELISA and reverse transcription polymerase chain reaction, respectively. Transplantation of hNSCs induced a significant increase in the production of bFGF and GDNF in the ST and SN compared with vehicle-treated group at 14 days post-transplantation. At 28 days after transplantation, a decline in the level of bFGF was observed, while GDNF remained significantly higher than in the control group (Fig. 6D; Supporting Information Fig. 4). The level of proinflammatory cytokines such as IL-1β, IL-2, and TNF-α and anti-inflammatory cytokine IL-10 were also evaluated. After hNSC transplantation, IL-1β, IL-2, and TNF-α were downregulated in the ST and SN at day 28. The level of IL-10 increased in the ST from 14 days post-transplantation, but did not significantly increase until 28 days post-transplantation in the SN, suggesting that hNSC transplantation suppressed inflammation in the brain and may exert influence on different brain regions at different times (Supporting Information Fig. 5).


Cell-based therapies for PD have been studied for more than 30 years using various cell types. In previous studies, the extent of DAergic damage and beneficial effects of hNSC engraftment was usually determined by Th IHC, mainly focused on the SN and ST. However, the changes in other brain regions that also contribute to the progress of PD and their response to cell therapy were rarely addressed. In this study, by using quantitative mass spectrometry augmented by HPLC and TMT labeling techniques, we measured and compared the abundance of more than 5,000 proteins in multiple brain regions accurately and obtained a comprehensive proteomic analysis of the influence of hNSCs on a 6-OHDA-induced PD animal model.

The intrastriatal injection of 6-OHDA produces selective and severe degeneration of DA nerve terminals in the ST, as well as a loss of DA cell bodies in the SN and, to a lesser extent, in the ventral tegmental area [18, 30]. In this study, we unexpectedly found that, in contrast to the ST, SN, and OB, the proteome of the SVZ was greatly affected by 6-OHDA administration, whereby the abundance of nearly 20% of the total proteins was significantly changed. The SVZ lining the lateral ventricles is one of the primary sites of neurogenesis in the adult mammalian brain, in which slowly proliferating endogenous neural stem cells (B cells) produce transit-amplifying progenitor cells (C cells) [31, 32]. It is reported that the proliferative capacity of C cells is regulated by DAergic fibers originated from the SN via the D2-like receptors, and the number of proliferating neural precursors in the SVZ reduced by approximately 40% in a 6-OHDA model of PD because of the diminished DAergic innervation [33-35]. Consistent with these findings, in this study, 6-OHDA lesioning led to significant downregulation of proteins involved in synaptic transmission, including Th, synaptic proteins Psd95, and synaptophysin, and the dopamine receptors Drd1 and Drd2, which clearly demonstrated DAergic denervation. However, this is not likely to be associated with full-range suppression of proliferation as the levels of proliferation markers such as PCNA and Ki67 were not decreased by 6-OHDA lesioning. Moreover, epidermal growth factor receptor (EGFR), the marker for proliferating cells targeted by DAergic fibers, also remained stable in the 6-OHDA group, suggesting that the proliferation capacity of the SVZ is maintained in the parkinsonian brain [36, 37].

On the other hand, the markers for undifferentiated NSCs, which possess characteristics of astrocytes and express Gfap, were significantly upregulated in the 6-OHDA group indicating an increase in the number of neural stem cells in the SVZ [32, 38]. While some of the most abundant structural proteins of mature neurons and oligodendrocytes, such as neurofilament, Cnp, and Mog exhibited drastic downregulation following 6-OHDA insult. Thus, it is possible that neurogenesis in the SVZ was impaired after 6-OHDA-induced dopamine denervation.

After hNSC engraftment, most of the disturbed proteins, including the aforementioned proteins, were regulated to a level comparable to that in the intact control, suggesting that DAergic innervation was re-established and the function of the SVZ was restored. The peptide sequence analysis suggested that the recovery was not because donor hNSCs migrated into the SVZ but rather because endogenous NSCs were restored. Our results corroborate the finding that transplantation of stem cells can result in stimulation of endogenous NPCs, neurogenesis, and functional effects [39-43]. Moreover, the transplantation-induced neurogenesis is associated with nigrostriatal neuroprotection, and inhibition of neurogenesis would attenuate the effects of transplantation [44, 45].

6-OHDA-induced lesioning is progressive and normally requires 14 days to achieve stability. However, we found that the restorative effect of hNSCs in the SVZ was correlated with time after engraftment. Transplantation at 7 days was much more efficacious compared to that at 14 days after 6-OHDA lesion (data not shown). To better illustrate the mechanism of hNSC-induced restoration, we chose to graft the cells 7 days after 6-OHDA treatment. Furthermore, we determined the state of lesions at 7 days by using behavioral task analysis (apomorphine-induced rotation), and non-lesioned mice were removed from our study. Lesioning was confirmed by TH staining and other behavioral tasks; mice at 7 days after treatment exhibited significant deficits compared to the control group, although to a lesser degree compared with those at 14 days. In contrast to the observations in the SVZ, the proteomic profiles of the other regions did not show much alteration after hNSC treatment; the DAergic neurons and innervation were not stimulated and the level of Th was not up-regulated. Nevertheless, the behavioral tests indicated the therapeutic efficacy of intrastriatal hNSC injection, suggesting that hNSCs may exert beneficial effects on the ST by protecting and/or enhancing pre-existing DAergic terminals. We demonstrated that the numbers of major cell types in the ST were not significantly altered by hNSC transplantation, suggesting that the restorative effect of hNSCs may not result from direct differentiation and neuron replacement, but rather from their multiple homeostatic functions [8]. Furthermore, we did not observe direct differentiation of hNSCs into functional neurons by immunostaining (data not shown). Thus, it is likely that the engrafted hNSCs function by slowing the denervation progress or protecting the existing neurons. Unsurprisingly, after hNSC transplantation, GDNF and bFGF, which is known to augment and protect DA systems [46-50], were significantly induced while proinflammatory cytokines were decreased after transplantation, consistent with the aforementioned findings. Our results support the speculation that hNSCs promote the homeostatic adjustment of host nigral DAergic neurons and their nigrostriatal projections, which may contribute to the restoration of the SVZ and improvement in behavioral performance [8].


We confirmed the efficacy of hNSCs on intrastriatal 6-OHDA lesioned mice. By combining biochemical and high-throughput proteomic approaches, we characterized the response of different brain regions to 6-OHDA insult and subsequent hNSC transplantation on molecular level. We revealed the unexpected change of SVZ in PD progress and hNSC treatment, and demonstrated the restoration and activation of endogenous NSCs by hNSC transplantation. Our studies represent the most comprehensive proteomic analysis on PD model animals to date, and provide valuable insights into the future clinical application of hNSCs and development of potential PD therapies.

Author Contributions

F.Z.: conception and design, provision of study material, collection and assembly of data, and manuscript writing; F.X.: conception and design, data analysis and interpretation, manuscript writing, and final approval of manuscript; X.W.: conception and design, collection and assembly of data, data analysis and interpretation, and final approval of manuscript; X.L.: provision of study material; R.W., W.G., and X.B.: conception and design, financial support, and final approval of manuscript. F.Z., F.X., and X.W. contributed equally to this work.

Disclosure of Potential Conflicts of Interest

The authors indicate no potential conflicts of interests.


This work was supported by National High Technology Research and Development Program (“863” Program) of China (2013AA020106, 2014AA020513), the National Natural Science Foundation of China (81200916, 81373150, 91632113), CAMS Innovation Fund for Medical Sciences (CIFMS) (2016-I2M-1-004), and Shanghai Brain-Intelligence Project from STCSM (16JC1420500).