Cell Mol Neurobiol DOI 10.1007/s10571-017-0513-1

ORIGINAL RESEARCH

Human Umbilical Cord Mesenchymal Stem Cells Protect Against SCA3 by Modulating the Level of 70 kD Heat Shock Protein Tan Li1 • Yi Liu1 • Linjie Yu1 • Jiamin Lao1 • Meijuan Zhang1,2 • Jiali Jin1,2 Zhengjuan Lu1,2 • Zhuo Liu1,2 • Yun Xu1,2

•

Received: 12 March 2017 / Accepted: 17 June 2017 Ó Springer Science+Business Media, LLC 2017

Abstract Spinocerebellar ataxia 3 (SCA3), which is a progressive neurodegenerative disease, is currently incurable. Emerging studies have reported that human umbilical cord mesenchymal stem cells (HUC-MSCs) transplantation could be a promising therapeutic strategy for cerebellar ataxias. However, few studies have evaluated the effects of HUC-MSCs on SCA3 transgenic mouse. Thus, we investigated the effects of HUC-MSCs on SCA3 mice and the underlying mechanisms in this study. SCA3 transgenic mice received systematic administration of 2 9 106 HUCMSCs once per week for 12 continuous weeks. Motor coordination was measured blindly by open field tests and footprint tests. Immunohistochemistry and Nissl staining were applied to detect neuropathological alternations. Neurotrophic factors in the cerebellum were assessed by ELISA. We used western blotting to detect the alternations of heat shock protein 70 (HSP70), IGF-1, mutant ataxin-3, Tan Li and Yi Liu have contributed equally to this work.

Electronic supplementary material The online version of this article (doi:10.1007/s10571-017-0513-1) contains supplementary material, which is available to authorized users.

and apoptosis-associated proteins. Tunel staining was also used to detect apoptosis of affected cells. The distribution and differentiation of HUC-MSCs were determined by immunofluorescence. Our results exhibited that HUCMSCs transplantation significantly alleviated motor impairments, corresponding to a reduction of cerebellar atrophy, preservation of neurons, decreased expression of mutant ataxin-3, and increased expression of HSP70. Implanted HUC-MSCs were mainly distributed in the cerebellum and pons with no obvious differentiation, and the expressions of IGF-1, VEGF, and NGF in the cerebellum were significantly elevated. Furthermore, with the use of HSP70 analogy quercetin injection, it demonstrated that HSP70 is involved in mutant ataxin-3 reduction. These results showed that HUC-MSCs implantation is a potential treatment for SCA3, likely through upregulating the IGF-1/ HSP70 pathway and subsequently inhibiting mutant ataxin3 toxicity. Keywords Ataxia
Human umbilical cord mesenchymal stem cells
HSP70
Neurotrophins
Purkinje cells
Spinocerebellar ataxia 3

& Zhuo Liu [email protected]

Jiali Jin [email protected]

Tan Li [email protected]

Zhengjuan Lu [email protected]

Yi Liu [email protected]

Yun Xu [email protected]

Linjie Yu [email protected]

1

Department of Neurology, Affiliated Drum Tower Hospital of Nanjing University Medical School, 321 ZhongShan Road, Nanjing City 210008, Jiangsu Province, People’s Republic of China

2

Department of Neurology, Drum Tower Hospital of Nanjing Medical University, Nanjing, People’s Republic of China

Jiamin Lao [email protected] Meijuan Zhang [email protected]

123

Cell Mol Neurobiol

Abbreviations DMEM Dulbecco’s modified Eagle’s medium HLA Human leukocyte antigen HSP70 Heat protein 70 HUC-MSCs Human umbilical cord mesenchymal stem cells IGL Internal granular layer ML Molecular layer MSCs Mesenchymal stem cells PCL Purkinje cell layer PFA Paraformaldehyde P3 Passage three SCA3 Spinocerebellar ataxia 3 WT Wild type

Introduction Spinocerebellar ataxia 3 (SCA3) is a devastating progressive neurodegenerative disorder that often begins with ataxia and is related with other neurological symptoms, such as cognitive impairment and pyramidal signs (Harding 1983). It is caused by the expansion of a CAG trinucleotide repeat in the coding region of the causative gene (Cendelin 2016). To date, no effective therapies have been developed. Currently, stem cell therapy has been tested using animal models and may be a prospective treatment for neurodegenerative diseases. Human umbilical cord mesenchymal stem cells (HUC-MSCs) are defined as multipotent progenitor cells that have the potential to differentiate into many lineage cells, such as neural cells, osteoblasts, hepatocytes, and adipocytes (Lagasse et al. 2000; Li et al. 2015; Pittenger et al. 1999; Troyer and Weiss 2008). Compared with other stem cells, HUC-MSCs hold several merits, such as their non-invasive harvest procedure, good tolerance by the immune system, easy in vitro expansion, and acquisition from an uncontroversial source (Li et al. 2015). These attractive advantages make HUC-MSCs a promising candidate for cell-based therapy. In our previous study, we proved that HUC-MSCs treatment is capable of alleviating the motor impairments and cerebellar atrophy in an ataxic mouse model induced by Ara-C (Zhang et al. 2011). Furthermore, we also confirmed the safety, feasibility, and efficacy of HUC-MSCs therapy in 16 SCA patients with 12 month of follow-up (Jin et al. 2013). Another study using Lurcher mice by Jones et al. (2010) described that intracerebellar transplantation of bone marrow mesenchymal stem cells can significantly improve motor performance and they observed that grafted cells were located adjacent to Purkinje cells and produced neurotrophic factors, which may contribute to the survival of Purkinje cells (Jones et al. 2010). However, the inherent molecular mechanisms are still unclear.

123

We performed this study to investigate the utility and related mechanisms of HUC-MSCs for treating SCA3 mice, with the goal of developing a new effective therapy for SCA3 patients.

Materials and Methods Animals SCA3 transgenic mice (C57BL/6 background) of either sex which had been acquired from the Jackson Laboratory were used in our experiments. These SCA3 transgenic mice harbor a YAC transgene that expresses a human ataxin-3 gene modified with an expanded 84 CAG repeat motif that is associated with Machado-Joseph disease in humans. We used the method of quantitative real-time PCR to determine homozygosity or hemizygosity for the transgene. Homozygous SCA3 mice (SCA3 tg/tg) were selected. When the mice were at the age of 12 weeks and showed no obvious symptoms of dyskinesia or neuropathological changes in the cerebellum and brainstem, HUC-MSCs were intravenously injected into them. Within 12 weeks, about 2 9 106 HUCMSCs were injected per SCA3-Tg mouse each consecutive week. The mice were divided to three groups: wild-type C57BL/6 mice, SCA3 tg/tg mice injected with normal saline and HUC-MSC-treated SCA3 tg/tg mice. Quercetin (50 mg/ kg, Sigma, USA), an inhibitor of HSP70, was given intraperitoneally 30 min before, during, and the 1st, 2nd, 3rd, 4th, 5th, 6th day after HUC-MSCs transplantation to determine the effects of HSP70 on the mutant ataxin-3. This protocol has been reported to effectively suppress the expression of HSP70 (Cho et al. 2006; Maliutina Ia et al. 2001). JB1 (18 ng/g; Bachem), an inhibitor of IGF-1, was injected intravenously in a week after HUC-MSCs transplantation to determine the effects of IGF-1 on HSP70. This protocol has been reported to effectively suppress the expression of IGF-1. All animal experiments were approved by the National Regulations of Experimental Animal Administration and in accordance with the Committee of Experimental Animal Administration of Nanjing University. Isolation and Culture of HUC-MSCs Fresh human umbilical cord tissues were collected from fullterm neonates after cesarean delivery in the Affiliated Drum Tower Hospital of Nanjing University. We obtained consent from the donors to ensure that they were informed of the use of their umbilical cords for medical research. After the removement of the blood, blood clots, umbilical arteries, and veins, the umbilical cord tissues were cut into tiny pieces (approximately 1 mm3) and put into 10 cm plastic dishes disposed with poly-D lysine (Sigma, America) beforehand.

Cell Mol Neurobiol

The tissues were incubated in the complete Dulbecco’s modified Eagle’s medium/F12 (DMEM/F12) (Thermo, USA) supplemented with 10% fetal bovine serum (BI, Isracl) at 37 °C in 5% CO2 and 95% humidity. Primary cell adherence (passage zero) was observed at about 1 month, and the umbilical cord tissues were removed. The media was changed every 3 days, and the cells were trypsinized and subcultured every 5 days. The P3 subcultured cells were used for experiments. We analyzed the cell phenotype via flow cytometry (BD Pharmingen, USA). Additionally, common microorganisms, such as hepatic virus, HIV virus, and bacterium, were tested to further ensure these cells were free from known infectious diseases.

perfusion with 4% paraformaldehyde (PFA). The removed brains were kept in 4% PFA for over 24 h and then dehydrated gradually in 15% and 30% (w/v) sucrose in 0.01 M PBS overnight. After dehydration, the fixed brains were embedded in paraffin and sectioned at a thickness of 4 lm for further experiments. Immunostaining

The open field test was used to assess mice locomotor function and explorative behavior. Mice were placed in a isolated space (50 cm 9 50 cm 9 50 cm). Their movement activities, such as location, distance, frequency of standing, duration of travel bouts, and occurrence of stereotypical behaviors, were recorded for 5 min using an Acti-Track System (Panlab, Barcelona, Spain). Mice were tested at the age of 12 weeks, and the test was carried out weekly before their sacrifice.

The coronal sections were incubated in xylene and graded ethanol solutions (100, 95, 70, and 50% (v/v)), respectively, and then washed with 0.3% H2O2. Heat-induced antigen retrieval was performed to repair denatured proteins. Subsequently, the samples were washed three times with 0.25% Triton X-100 (15 min) and incubated with following primary antibodies at 4 °C overnight: calbindin (Cell Signaling, 1:200), GFAP (Abcam, 1:200), HSP70 (1:50, Abcam), MAB1287 (Millipore, 1:30), NeuN (Millipore, 1:200), and 1C2 (Millipore, 1:1000) (detecting the mutant ataxin-3) overnight at 4 °C. The sections treated with secondary antibodies, goat anti-rabbit and/or mouse, conjugated to Alexa 488 or 594 (Invitrogen) for 2 h at room temperature for immunofluorescence. For immunohistochemistry, anti-rabbit and/or mouse second antibodies were used. A DAB kit (Vector Laboratories) was used for microscopic analysis, and hematoxylin was used for nuclear counterstaining.

Footprint Test

Nissl Staining

As described previously (Bichelmeier et al. 2007; Nobrega et al. 2013), the footprint test was applied for gait analysis. The hind- and forefeet of the mice were colored with black and red nontoxic paints, respectively. Mice were placed over a piece of white paper and allowed to walk along a 100-cmlong, 5-cm-wide runaway (with 5-cm high walls). Two indexes were used to assay the footprint patterns: (1) foot length, the forward movement distance of each stride, and (2) footprint overlap, corresponding to distance of the left or right front footprint/hind footprint overlap. The distance between the center of the front footprint and the rear hind footprint was recorded. Only including the middle of a sequence of six steps, the distance between the center of the front and the hind footprint was recorded. For statistical analysis, three mice per group were included and one independent operator performed all footprints tests in a blind fashion.

The secretion function of Purkinje cells was assessed by Nissl staining, as previously described (Peng et al. 2015). Coronal brain sections were stained with a solution containing 0.01% cresyl violet (Sigma, America), 0.01% methylene blue (Amresco, America), 0.01% toluidine blue (Amresco, America), and 0.01% thionin (Regal, China) for 24 h at room temperature. The brain sections were then soaked in 95% ethyl alcohol for 5–30 min. Check microscopically for staining after rinsing for 5–10 min in double distilled water. Lastly, the sections were dehydrated in ethanol, cleared using xylene and mounted in neutral gum. The Nissl bodies were observed and photographed under a high-magnification (9400) light microscope (Olympus, Japan).

Histological Processing

Brain damage was assessed by terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL) assay (Roche, In Situ Cell Death Detection kit, AP, Mannheim, Germany). Brain sections were incubated with a blocking solution for 10 mins at room temperature and then incubated in phosphate-buffered saline for an

Behavior Test Open Field Test

Tissue Preparation 12 weeks after HUC-MSCs transplantation, the mice were anesthetized with 1% carbrital and fixed via transcardial

TUNEL Assay

123

Cell Mol Neurobiol

additional 5 mins three times, and then treated with a terminal deoxynucleotidyl transferase reaction mixture of DNA strand breaks for 60 mins at 37 °C in a dark humidified atmosphere. The sections were then incubated with alkaline phosphatase (AP) for 30 mins at 37 °C, followed by nitroblue tetrazolium substrate to develop a blueviolet color. TUNEL-positive cells were examined using a microscope. Cells in each microscopic field were determined using Image-Pro Plus 6.0 (Media Cybernetics, Bethesda, MD, USA). Blueviolet cells were counted as positive cells. Three mice per group were used in this test. Western Blotting Protein was collected from the cerebellums and brainstems with protein extraction kits (Beyotime Biotechnology, China) according to the manufacturer’s instruction. We measured the concentrations of protein using BCA protein assay kits (Bioworld, USA). The proteins extracts were isolated by SDS-PAGE and transferred electrophoretically to PVDF membranes. Membranes were blocked with 5% powdered skim milk for 1 h and then incubated overnight at 4 °C with first antibody: anti-1C2 (1:1000, Millipore, USA), anti-Bcl-2 (1:1000, Bioworld, USA), anti-Bax (1:1000, Cell Signaling, USA), anti-caspase-3 (1:500, Cell Signaling, USA), anti-caspase-9 (1:1000, Cell Signaling, USA), anti-HSP70 (1:500, Abcam, Britain), anti-IGF-1 (1:500, Millipore, USA), anti-b-actin (1:5000, Bioworld, USA), or anti-GAPDH (1:5000, Bioworld, USA). b-actin and GAPDH were used as internal controls. The membranes were incubated with secondary antibodies and visualized using an ECL kit (Bioworld, USA). The intensities of bands were quantified using Quantity One Software (Bio-Rad, Hercules, USA).

and treated with an enzyme-linked second primary antibody solution for another 2 h. Then the wells were rinsed three times with a substrate solution added to the wells and incubated in the dark for 30 min. We measured the intensity of color at a wavelength of 450 nm. The concentrations of the neurotrophins were normalized per total milligram of protein assayed. Purkinje Cell Counts For Purkinje cell counts, every sixth coronal section (10 lm) was collected on a microslide to ensure the tissue sections in each group were at the same level. Purkinje cell counts were acquired by determining the total number of neurons in the region of the folium within 50 lm of the primary fissure using three sections per mouse (n = 5). Weight Assessment The mouse brains were dissected from the skull and rolled on tissue paper. Then the cerebellum was dissected from the brain with a small knife inserted parallel to the mouse dorsal hindbrain. The tissues were weighed to the nearest 0.1 mg. All of these procedures were performed at room temperature. Statistical Analysis The results were reported as the mean ± SEM. Group differences were analyzed by one-way ANOVA with Bonferroni’s post hoc tests for multiple groups as the sources of variation (SPSS 13.0 system, Chicago, IL, USA). The criterion for statistical significance was set to 0.05.

ELISA

Results The tissues were dissected from cerebellum and pons and homogenized in 0.1 mM PBS lysis buffer containing protease inhibitors (sigma). The supernatant was collected via centrifugation at 10,000 rpm for 15 min to remove the cell debris and stored at -80 °C immediately for further use. The concentration of neurotrophic factor was assayed with IGF-1 ELISA kits (Cusabio, China), NGF ELISA kits (Cusabio, China), VEGF ELISA kits (4Abio., China), GDNF ELISA kits (Cusabio, China), and BDNF ELISA kits (Cusabio, China) according to the manufacturer’s instructions. Briefly, for the standard curve, standard control samples of the protein were diluted serially (1:2) with their calibrator diluent and plated to two columns of wells. The frozen tissue supernatants were thawed on ice, and every sample plated in triplicate and incubated for 2 h at room temperature. The wells were rinsed with wash buffer

123

In Vitro Cell Characterization of HUC-MSCs All of the cells in the cultures were fibroblast-like and spirally arranged (Fig. 1a, b). The flow cytometric analysis revealed that HUC-MSCs were positive for CD29, CD44, CD73, CD90, and CD105, and negative for CD34, CD45, and HLA-DR (Fig. 1c), confirming that the cell population was mainly composed of mesenchymal stem cells. Characterization of SCA3-Tg Mice The SCA3-Tg mice were characterized as smaller and with decreased weights compared with their littermate controls. They presented hunchback, mild tremors, moderate hypoactivity, and gait disturbances. We chose the

Cell Mol Neurobiol

homozygous mice (SCA3 tg/tg) that exhibited earlier onset and increased severity of symptoms due to the doubled expression of pathological ataxin-3 as the experimental animals. The mice exhibited motor deficits at 10 weeks, degeneration of the cerebellar molecular layer, and a loss of Purkinje cells that were statistically significant at 20 weeks, as initially reported (Shakkottai et al. 2011). SCA3 tg/tg mice usually died prematurely at the age of 40 weeks.

vs. 124.21 ± 18.35 cm, p \ 0.05). The improvement lasted for the following 4 weeks. The improvement in motor function was further investigated by the footprint test (Fig. 2b). The mice treated with HUC-MSCs exhibited a statistically significant smaller overlap (from 8 weeks posttreatment; Fig. 2c) and a consistently larger stride length (from 6 weeks post-treatment; Fig. 2d) compared to control group that was administered normal saline (n = 10 for each group).

HUC-MSCs Transplantation Improved the Motor Behaviors of SCA3-Tg Mice

HUC-MSCs Alleviated Cerebellar Atrophy and Neuropathological Changes in the Cerebellum and Pons

The open field test was performed to evaluate the exploratory activities and locomotor abilities of the SCA3Tg mice. HUC-MSCs-treated group (n = 10) exhibited increased movement compared to the vehicle control group (n = 10; Fig. 2a). They traveled longer distances in five minutes and traveled more in the center of the cage. Slightly better behavior performance appeared at 4 weeks after HUC-MSCs treatment, and statistical significance emerged at 8 weeks post-transplant (177.43 ± 11.37 cm

The impact of HUC-MSCs on SCA3-typical neuropathology was assessed 12 weeks after administration. Cerebellar atrophy is one of the apparent pathological hallmarks observed in the SCA3 model. It was observed that HUCMSCs transplantation alleviated cerebellar atrophy in macroscopic changes. In the HUC-MSCs group, the cerebellum was observably larger than in normal saline-injected mice (Fig. 3). The weights of the mouse whole

Fig. 1 The properties of HUC-MSCs in vitro. a The HUC-MSCs in culture were spirally arranged. Scale bar: 200 lm. b The cells were fibroblast-like when cultured to passage 3. Scale bar: 100 lm. c The

cells were positive for CD29, CD44, CD73, CD90, and CD105, and negative for HLA-DR, CD34, and CD45 based on the flow cytometric analysis. HUC-MSCs human umbilical cord mesenchymal stem cells

123

Cell Mol Neurobiol

Fig. 2 Effects of HUC-MSCs transplantation on motor behavior. a Open field test (n = 10 for each group); b footprint test (n = 10 for each group); c the statistical analysis of foot length. *p \ 0.05, **p \ 0.01 versus WT group; #p \ 0.05, ##p \ 0.01 versus control group

brains and cerebellums were also calculated, and the results were in consistent with the macroscopic appearance (36.1 ± 3.04 mg in control group vs. 49.8 ± 3.31 mg in HUC-MSCs-treated group; p \ 0.01) (Table 1). Purkinje cells and neurons from the pons are the most vulnerable in patients suffering from SCA3 and in the transgenic animal models used in this study. Therefore, we investigated if HUC-MSCs could preserve Purkinje cells by calbindin staining and Nissl staining to test the morphology and function of the cells. HUC-MSCs-treated mice exhibited an increased number of calbindin-positive surviving Purkinje cells (Fig. 4a, b; 22 ± 2.17 cells per field in the control group vs. 30 ± 2.03 cells in the HUC-MSCs-treated group; p \ 0.05) and enriched Nissle bodies (Fig. 4c) compared to vehicle control mice. Thus, HUC-MSCs transplantation could preserve both the number and physiological function of Purkinje cells. Mutant ataxin-3 is another important neuropathological hallmark of SCA3. We then detected mutant ataxin-3 expression and location by western blotting and immunohistochemistry. Our results revealed that the mutant ataxin-3 was significantly reduced following HUC-MSCs treatment [Fig. 5a–d; 53% decrease in HUCMSCs-treated group (from 1 to 0.745–0.145); p \ 0.01] (n = 5 for each group) (Fig. 5a, b). In conclusion, HUCMSCs transplantation promoted preservation of the

123

cerebellum, including Purkinje cells and neurons from the pons, and inhibited the conformation of mutant ataxin-3. Implantation and Differentiation of HUC-MSCs We traced the survival and distribution of HUC-MSCs by immunostaining with MAB1287. The grafted cells that were positive for MAB1287 could be observed in the molecular layer and the Purkinje layer of the cerebellum and the pons (Fig. 6a–d). The number of the stem cells was few. The differentiation of HUC-MSCs into brain parenchyma cells was analyzed by double staining of cerebellum and pons sections with MAB1287 and anti-NeuN (neurons marker), anti-GFAP (astrocytes marker), or anti-calbindin (Purkinje cells marker). The results showed that little to no co-localization between MAB1287 and NeuN, GFAP, or calbindin was detected (Fig. 6b–d), indicating that few of the implanted HUC-MSCs differentiated following transplantation, as our group previously reported (Zhang et al. 2011). HUC-MSCs Enhanced the Expression of Neurotrophic Factors in Cerebellar Tissue Bystander capacities, including the modulation of neuroinflammation and the production of neurotrophic factors,

Cell Mol Neurobiol Fig. 3 The appearance of the cerebellum in WT, control, and HUC-MSCs-treated groups

Table 1 Effects of HUC-MSCs on cerebellar atrophy in SCA3 transgenic mice Group number

Group name

Mouse genotype

No. of mice

Body weight (g)

Whole brain weight (mg)

Cerebellum weight (mg)

CW/WBW

1

WT

WT

6

24.19 ± 1.06

488.79 ± 12.82

58.32 ± 3.05

0.12

2

Control

SCA3-YAC-84Q

6

23.07 ± 1.27

429.84 ± 16.55**

3

HUC-MSCs

SCA3-YAC-84Q

6

23.64 ± 1.55

464.86 ± 12.32

36.1 ± 3.04

0.08

49.84 ± 3.31##

0.11

WBW whole brain weight, CW cerebellar weight **p \ 0.01 versus normal group ## p \ 0.01 versus control group

Fig. 4 HUC-MSCs transplantation alleviated pathological alterations of Purkinje cells in SCA3 transgenic mice. a A loss of Purkinje cells and a thinner ML were observed in the control group. More Purkinje cells were preserved in the HUC-MSCs-treated group. Scale bar: 100 lm (n = 5 for each group). b Quantification of the number of

Purkinje cells of WT, control and HUC-MSCs-treated mice in the images at 9200 microscopic magnification. **p \ 0.01 versus WT, #p \ 0.05 versus control group. c The secretion function of Purkinje cells was assessed by Nissl staining. Scale bar: 10 lm (n = 5 for each group)

are among the leading roles of HUC-MSCs for exerting protective functions. Thus, the protein levels of certain neurotrophic factors in the cerebellum and pons were assayed by ELISA. The results showed that IGF-1, VEGF,

and NGF protein levels were higher in the treated group (56.25 ± 1.92; 29.09 ± 1.26; 19.46 ± 1.14 pg/mg) compared to the control group (26.73 ± 1.26; 16.7 ± 1.45; 12.2 ± 1.09 pg/mg), whereas no significant differences

123

Cell Mol Neurobiol

Fig. 5 HUC-MSCs transplantation inhibited the expression of the mutant ataxin-3 in SCA3-Tg mice. The mutant ataxin-3 was found in Purkinje cells of the cerebellum (a) and in neurons of the pons (b). Scale bar: 20 lm. c Representative image of western blotting of

mutant ataxin-3 in the cerebellum and pons. d Quantitative analysis of the protein level of mutant ataxin-3 in the WT, control, and HUCMSCs-treated mice group. **p \ 0.01 (n = 5 for each group)

were shown for BDNF and GDNF expression levels in our study (Fig. 7a–e).

western blotting, the results showed that, in the quercetin group, the expression of mutant ataxin-3 protein rose to 2.06-fold (p \ 0.05) compared to the HUC-MSCs-treated group (Fig. 8c, d), suggesting that HSP70 was closely associated with the deposition and degradation of mutant ataxin-3.

HSP70 Was Involved in the Reduction of Mutant Ataxin-3 The elevated neurotrophin IGF-1 could modulate the expression of HSP70, as determined by western blotting with the IGF-1 analog JB1, as previously reported (BatistaNascimento et al. 2011) (Fig. 8a). Given the role of HSP70 as molecular chaperone in mutant protein recognition and protein conformation suppression, we examined the expression of HSP70 in sections of the cerebellum and pons. Upon immunostaining, the results showed that the HSP70 positive cells co-localized with the mutant ataxin-3 (Fig. 8b). To confirm the effects of HSP70 on the decreased conformation of mutant ataxin-3, we used quercetin, an HSP70 inhibitor, for further verification. By

123

HUC-MSCs Transplantation Inhibited the Caspase3-Mediated Apoptosis Signaling Pathway To assess the apoptosis signaling pathway change after HUC-MSCs administration, the expressions of Bax, Bcl-2, and caspase9 and the subsequent activation of caspase3 were measured via western blotting. The results demonstrated that HUC-MSCs treatment significantly decreased the levels of Bax (p \ 0.05) and cleaved-caspase9 (p \ 0.01) together with an elevated expression of Bcl-2

Cell Mol Neurobiol

Fig. 6 The implantation and differentiation of HUC-MSCs after systematic administration via the tail vein. a Disseminated HUCMSCs could be detected at 3 months after treatment, as determined by immunohistochemical staining. Scale bar: 20 lm (n = 3 for each group). b–d The transplanted HUC-MSCs could be observed in the

pons, PCL, and ML of the cerebellum. The immunostaining of MAB1287 (red) did not co-localize with that of anti-NeuN (green), anti-GFAP (green), or anti-calbindin (green). Scale bar: 50 lm (n = 3 for each group)

Fig. 7 HUC-MSCs transplantation elevated the levels of some neurotrophic factors in the cerebellum and pons of SCA3 transgenic mice. a–e The protein levels of IGF-1, NGF, VEGF, BDNF, and

GDNF in the cerebellum and pons were determined by ELISA. *p \ 0.05 versus control group, **p \ 0.01 versus control group, **p \ 0.01 versus control group. n = 5 per group

(p \ 0.05) (Fig. 9a, b). Furthermore, HUC-MSCs could inhibit the activation of caspase3 (p \ 0.01) (Fig. 9a, b) in SCA3-tg mice. In conclusion, HUC-MSCs treatment could decrease the apoptosis of affected cells by modulation of

the caspase3-related cell apoptosis signaling pathway. To further verify that HUC-MSCs transplantation could protect apoptotic purkinje cells, we use TUNEL staining (Fig. 10). As expected, there were few TUNEL-positive

123

Cell Mol Neurobiol

cells in the WT brain. Additionally, TUNEL-positive cells in the HUC-MSCs-treated group were reduced as compared with that in the controlled group (8.11 ± 0.68 vs. 17.39 ± 1.81%, p \ 0.05).

Discussion In the present study, we showed for the first time that HUCMSCs transplantation significantly inhibited the expression of mutant ataxin-3 in SCA3 animal models, which might protect SCA3-Tg mice. Furthermore, upregulation of

Fig. 8 Effects of IGF-1-regulated HSP70 on the mutant ataxin-3 protein after HUS-MSCs administration. a IGF-1 increased the protein level of HSP70, as determined by western blotting. b Immunostaining of anti-mutant ataxin-3 (red) co-localized with antiHSP70 (green). Scale bar: 20 lm. c HSP70 inhibited the expression

123

HSP70 seems to be involved in the decline of mutant ataxin-3, which is associated with elevated IGF-1 induced by HUC-MSCs treatment. Thus, our results suggest that HUC-MSCs represent a promising strategy for SCA3 treatment, likely through upregulating the IGF-1/HSP70 pathway and subsequently inhibiting ataxin-3 toxicity. Mesenchymal stem cells (MSCs) can be obtained from multiple organs and tissues, such as bone marrow (Friedenstein et al. 1966), adipose tissue (Zuk et al. 2001), cord blood (Erices et al. 2000), umbilical cord (Ishige et al. 2009), skin (Lai et al. 2014), placenta (In ‘t Anker et al. 2004), tonsils (Ryu et al. 2014), dental pulp (Ponnaiyan

of mutant ataxin-3, as determined by western blotting. d Quantitative analysis of panel C (n = 5 for each group). **p \ 0.01 versus control group; #p \ 0.05 versus HUC-MSCs-treated group, ##p \ 0.01 versus HUC-MSCs-treated group

Cell Mol Neurobiol

et al. 2012), and even fetal lung and liver (in ‘t Anker et al. 2003; Joshi et al. 2012). HUC-MSCs have several superiorities compared with other tissue-derived cells, including higher proliferation potential, more constant doubling time (Lu et al. 2006) and reduced immunogenicity (Cho et al. 2008; Deuse et al. 2011), which guarantee the success of transplantations. In the present study, the number of the implanted cells that could be observed in the damaged site was small, as a large proportion of the stem cells migrated to and became distributed in the capillaries within the lung, liver, and spleen, as previously reported (Phinney and Prockop 2007; Zhang et al. 2011). In addition, the HUCMSCs that crossed the blood-brain barrier were mainly located in the cerebellum and pons. In the current stage, the therapeutic mechanisms of MSCs are still unclear. Host cell replacement, paracrine actions, and cell–cell contact are the three leading theories (Li et al. 2015). However, in our study, we failed to observe the differentiation of HUCMSCs into endogenous cell lineages of brain tissue upon double staining with MAB1287 and anti-NeuN, anti-calbindin, or anti-GFAP. In fact, it is unlikely that the benefits of stem cell transplantation mainly depend on the cell replacement. First, the number of the cells distributed in the injured site is small after intravenous infusion and therefore is far a sufficient amount to rebuilt damaged tissue (Phinney and Prockop 2007). Second, functional cell replacement demands the establishment of physiologically relevant connections by integration of stem cell-derived neurons into a neuronal network (Chopp and Li 2002; Mendonca et al. 2015), which is difficult to accomplish. Thus, the paracrine actions of HUC-MSCs have become more popular. HUC-MSCs can release an extensive variety of bioactive factors, which may help to preserve a friendly micro-environment for endogenous restorative processes

(Arno et al. 2014) via apoptosis prevention (Zhang et al. 2011), angiogenesis (Barlow et al. 2008), and immunoregulation (Kawabori et al. 2013; Ooboshi et al. 2005). The neuroprotective functions of HUC-MSCs were further detected in our study. The data demonstrated that HUC-MSCs infusion elevated IGF-1, VEGF, and NGF protein levels in brain tissue, in accordance with results described previously (Zhang et al. 2004, 2006). The MSCs behave as a factory of multiple of bioactive factors: they not only produce many biochemical and molecular factors themselves (Chen et al. 2002a, b), but also induce remodeling of the central nervous system by contacting with the parenchymal tissue and activating intrinsic restorative progress, although the precise mechanisms directly relating HUC-MSCs interaction with host tissue to stimulate the expression of neurotrophins are not known (Li et al. 2002; Zhang et al. 2005). The mouse ELISA kits which have no significant crossreactivity according to their manufacturer’s instructions demonstrated elevated neurotrophin. Therefore, we believed that large increase of the neurotrophin should be mainly generated by mouse tissue. See the values of body weight, whole brain weight and cerebellum weight in Table 1, it is evident that the use of HUC-MSCs induces only a partial recovery in SCA3YAC-84Q mice compared to WT mice, which is consistent to our pathological result that mutant ataxin-3 is not complete removal in HUC-MSCs treated group, we speculate that there may other more pathogenic mechanisms relevant, in addition, transplantation frequency and quantity of cells may be related to the efficacy. HSP70 is an important member of the heat shock protein family. It serves as a molecular chaperon to aid in the exact folding of nascent polypeptides or machinery clearance for misfolded proteins (Chaudhury et al. 2006; Luo et al.

Fig. 9 HUC-MSCs inhibited the caspase3-mediated apoptosis signaling pathway in the SCA3 transgenic mice. a The protein levels of Bax, Bcl-2, caspase9, and caspase3 in the cerebellum and pons of mice were determined by western blotting. b Quantitative analysis of

the protein levels in panel A. **p \ 0.01 versus WT group; #p \ 0.05 versus control group, ##p \ 0.01 versus control group; &p \ 0.05 versus HUC-MSCs-treated group, &&p \ 0.01 versus HUC-MSCstreated group

123

Cell Mol Neurobiol

Fig. 10 Apoptosis determined by TUNEL. TUNEL-positive cells significantly increased in the SCA3 transgenic mice versus that in WT group, and HUC-MSCs treatment could reverse it (a–b). The percent of TUNEL-positive was 17.39 ± 1.81% in the control group and

decreased to 8.11 ± 0.68% after treatment with HUC-MSCs (c). Six representative microscopic fields were analyzed for each group; *p \ 0.05 versus WT group; #p \ 0.05 versus control group

2010). Several lines of evidence have indicated that the heat shock protein family is closely associated with mutant proteins in polyglutamine diseases. The mutant proteins in transfected cells could co-localize with endogenous HSP, and overexpression of HSP decreased the frequency of mutant ataxin-1 expression (Cummings et al. 1998; Muchowski et al. 2000). Additionally, the affected cells decreased significantly in the brain when SCA1 mice were crossed with HSP70-overexpressing mice (Cummings et al. 2001). Consistent with these results, in our study, we use an inhibitor of HSP70, Quercetin, to suppressed HSP70 expression, and observed a significant increase in the expression of mutant ataxin-3. The pathogeny of SCA3 is the conformation and deposition of mutant ataxin-3 in the affected cell nucleus. Some studies have showed that the conformation of mutant protein is multistage. The polyQ expanded exon 1 of mutant protein are first amyloid-like.

They are soluble and deposited in the cytosol (Scherzinger et al. 1999). Then large insoluble protein form when a critical concentration is reached (Muchowski et al. 2000). The HSP70 chaperones may work during the first stage to help the clearance of soluble protein by proteasome machinery. Hence, in the present study, HSP70 chaperones delayed disease initiation and slowed disease progression in unaffected SCA3 transgenic mouse. Furthermore, it is well reported that alternations in IGF1 signaling have an impact on HSP expression and activation (Batista-Nascimento et al. 2011; Urban et al. 2012). For instance, IGF-1 signaling helps to reduce multimeric HSP27 into functional oligomers or monomers via protein kinase C and Akt activation (Abisambra et al. 2010). Thus, the higher level of HSP70 observed in our study in the HUC-MSCs-treated group may be attributed to the elevated IGF-1 level induced by HUC-MSCs. We also use the

123

Cell Mol Neurobiol

IGF-1 analog JB1 intravenous injection to confirm the effects of elevated IGF-1 on HSP70 with western blotting. Apoptosis is a common physiopathologic alternation in many neurodegenerative diseases, including SCA3. Some studies have strongly indicated that mutant ataxin-3 could lead to apoptosis of pontine nuclei by Bax upregulation and caspase3 activation (Chou et al. 2011). Additionally, IGF1, VEGF and NGF not only exert their neurotrophic effects but have been well documented for their anti-apoptosis functions (Hao et al. 2011; Hu et al. 2014; Mnich et al. 2014; Sheng et al. 2013). Therefore, along with mutant ataxin-3 conformation clearance through HSP70 chaperones, our study also indicated that HUC-MSCs transplantation could alleviate polyglutamine-induced symptoms by suppressing apoptosis of affected neurons via upregulating Bcl-2 expression, downregulating Bax expression, and eventually inhibiting the activation of caspase9- and caspase3-mediated apoptotic pathways. In summary, the present study showed that HUC-MSCs intravenous infusion is effective in alleviating symptoms of SCA3 by IGF-1/HSP70 pathway and pathway-dependent mutant ataxin-3 clearance. Moreover, anti-apoptosis is another underlying avenue by which HUC-MSCs exert their neuroprotective functions in SCA3.

Conclusion These results showed that HUC-MSCs implantation may be applicable for SCA3 likely via promoting neurotrophic factor secretion and modulating HSP70 levels by elevating IGF-1. Acknowledgements We express our sincere appreciation to the patients who provided umbilical cord in the research. We thank the department of Obstetrics and Gynecology of Drum Tower Hospital of Nanjing University Medical School for material support. Funding This work was funded by the National Nature Science Foundation of China (81200876, 81230026, 81300988 and 81501028) and the Science Foundation from Jiangsu Provincial Commission of Health and Family Planning (H201538). Authors contribution TL designed experiments, contributed ideas, and revised manuscript. YL, LY, and JL contributed to the experiments performance and the collection/assembly of data. MZ, YX, JJ, and ZL contributed to data analysis, results interpretation, and manuscript editing. ZL conceived the experiments, developed the project, contributed ideas, and wrote the manuscript. All authors read and approved the final manuscript. Compliance with Ethical Standards Conflict of interest The authors declare that they have no conflict of interest. Informed Consent The consent is obtained from the all the participants.

Statement of Human Rights All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.’’ Human tissues were obtained under approval of the Ethical Committee of Health Sciences Faculty of the Nanjing University (China) (Project Number 2013-081-01). Statement on the Welfare of Animals All procedures performed in studies involving animals were in accordance with the ethical standards of the National Regulations of Experimental Animal Administration and the Committee of Experimental Animal Administration of Nanjing University.

References Abisambra JF et al (2010) Phosphorylation dynamics regulate Hsp27mediated rescue of neuronal plasticity deficits in tau transgenic mice. J Neurosci 30:15374–15382. doi:10.1523/JNEUROSCI. 3155-10.2010 Arno AI et al (2014) Human Wharton’s jelly mesenchymal stem cells promote skin wound healing through paracrine signaling. Stem Cell Res Ther 5:28. doi:10.1186/scrt417 Barlow S et al (2008) Comparison of human placenta- and bone marrow-derived multipotent mesenchymal stem cells. Stem Cells Dev 17:1095–1107. doi:10.1089/scd.2007.0154 Batista-Nascimento L, Neef DW, Liu PC, Rodrigues-Pousada C, Thiele DJ (2011) Deciphering human heat shock transcription factor 1 regulation via post-translational modification in yeast. PLoS ONE 6:e15976. doi:10.1371/journal.pone.0015976 Bichelmeier U et al (2007) Nuclear localization of ataxin-3 is required for the manifestation of symptoms in SCA3: in vivo evidence. J Neurosci 27:7418–7428. doi:10.1523/JNEUROSCI.4540-06. 2007 Cendelin J (2016) Transplantation and stem cell therapy for cerebellar degenerations. Cerebellum 15:48–50. doi:10.1007/s12311-0150697-1 Chaudhury S, Welch TR, Blagg BS (2006) Hsp90 as a target for drug development. ChemMedChem 1:1331–1340. doi:10.1002/cmdc. 200600112 Chen X et al (2002a) Human bone marrow stromal cell cultures conditioned by traumatic brain tissue extracts: growth factor production. J Neurosci Res 69:687–691. doi:10.1002/jnr.10334 Chen X et al (2002b) Ischemic rat brain extracts induce human marrow stromal cell growth factor production. Neuropathology 22:275–279 Cho JY, Kim IS, Jang YH, Kim AR, Lee SR (2006) Protective effect of quercetin, a natural flavonoid against neuronal damage after transient global cerebral ischemia. Neurosci Lett 404:330–335. doi:10.1016/j.neulet.2006.06.010 Cho PS et al (2008) Immunogenicity of umbilical cord tissue derived cells. Blood 111:430–438. doi:10.1182/blood-2007-03-078774 Chopp M, Li Y (2002) Treatment of neural injury with marrow stromal cells. Lancet Neurol 1:92–100 Chou AH, Lin AC, Hong KY, Hu SH, Chen YL, Chen JY, Wang HL (2011) p53 activation mediates polyglutamine-expanded ataxin3 upregulation of Bax expression in cerebellar and pontine nuclei neurons. Neurochem Int 58:145–152. doi:10.1016/j.neuint.2010. 11.005 Cummings CJ, Mancini MA, Antalffy B, DeFranco DB, Orr HT, Zoghbi HY (1998) Chaperone suppression of aggregation and altered subcellular proteasome localization imply protein misfolding in SCA1. Nat Genet 19:148–154. doi:10.1038/502

123

Cell Mol Neurobiol Cummings CJ et al (2001) Over-expression of inducible HSP70 chaperone suppresses neuropathology and improves motor function in SCA1 mice. Hum Mol Genet 10:1511–1518 Deuse T et al (2011) Immunogenicity and immunomodulatory properties of umbilical cord lining mesenchymal stem cells. Cell Transpl 20:655–667. doi:10.3727/096368910X536473 Erices A, Conget P, Minguell JJ (2000) Mesenchymal progenitor cells in human umbilical cord blood. Br J Haematol 109:235–242 Friedenstein AJ, Piatetzky S II, Petrakova KV (1966) Osteogenesis in transplants of bone marrow cells. J Embryol Exp Morphol 16:381–390 Hao CN, Geng YJ, Li F, Yang T, Su DF, Duan JL, Li Y (2011) Insulin-like growth factor-1 receptor activation prevents hydrogen peroxide-induced oxidative stress, mitochondrial dysfunction and apoptosis. Apoptosis 16:1118–1127. doi:10.1007/ s10495-011-0634-9 Harding AE (1983) Classification of the hereditary ataxias and paraplegias. Lancet 1:1151–1155 Hu H et al (2014) Targeted NGF siRNA delivery attenuates sympathetic nerve sprouting and deteriorates cardiac dysfunction in rats with myocardial infarction. PLoS ONE 9:e95106. doi:10. 1371/journal.pone.0095106 In ‘t Anker PS, Scherjon SA, Kleijburg-van der Keur C, de GrootSwings GM, Claas FH, Fibbe WE, Kanhai HH (2004) Isolation of mesenchymal stem cells of fetal or maternal origin from human placenta. Stem Cells 22:1338–1345. doi:10.1634/stem cells.2004-0058 in ‘t Anker PS et al (2003) Mesenchymal stem cells in human secondtrimester bone marrow, liver, lung, and spleen exhibit a similar immunophenotype but a heterogeneous multilineage differentiation potential. Haematologica 88:845–852 Ishige I et al (2009) Comparison of mesenchymal stem cells derived from arterial, venous, and Wharton’s jelly explants of human umbilical cord. Int J Hematol 90:261–269. doi:10.1007/s12185009-0377-3 Jin JL et al (2013) Safety and efficacy of umbilical cord mesenchymal stem cell therapy in hereditary spinocerebellar ataxia. Curr Neurovasc Res 10:11–20 Jones J, Jaramillo-Merchan J, Bueno C, Pastor D, Viso-Leon M, Martinez S (2010) Mesenchymal stem cells rescue Purkinje cells and improve motor functions in a mouse model of cerebellar ataxia. Neurobiol Dis 40:415–423. doi:10.1016/j.nbd.2010.07.001 Joshi M, Joshi M, Patil P, He Z, Holgersson J, Olausson M, SumitranHolgersson S (2012) Fetal liver-derived mesenchymal stromal cells augment engraftment of transplanted hepatocytes. Cytotherapy 14:657–669. doi:10.3109/14653249.2012.663526 Kawabori M, Kuroda S, Ito M, Shichinohe H, Houkin K, Kuge Y, Tamaki N (2013) Timing and cell dose determine therapeutic effects of bone marrow stromal cell transplantation in rat model of cerebral infarct. Neuropathology 33:140–148. doi:10.1111/j. 1440-1789.2012.01335.x Lagasse E et al (2000) Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat Med 6:1229–1234. doi:10.1038/81326 Lai D, Wang F, Dong Z, Zhang Q (2014) Skin-derived mesenchymal stem cells help restore function to ovaries in a premature ovarian failure mouse model. PLoS ONE 9:e98749. doi:10.1371/journal. pone.0098749 Li Y et al (2002) Human marrow stromal cell therapy for stroke in rat: neurotrophins and functional recovery. Neurology 59:514–523 Li T, Xia M, Gao Y, Chen Y, Xu Y (2015) Human umbilical cord mesenchymal stem cells: an overview of their potential in cellbased therapy. Expert Opin Biol Ther 15:1293–1306. doi:10. 1517/14712598.2015.1051528 Lu LL et al (2006) Isolation and characterization of human umbilical cord mesenchymal stem cells with hematopoiesis-supportive function and other potentials. Haematologica 91:1017–1026

123

Luo W, Sun W, Taldone T, Rodina A, Chiosis G (2010) Heat shock protein 90 in neurodegenerative diseases. Mol Neurodegener 5:24. doi:10.1186/1750-1326-5-24 Maliutina Ia V, Semina OV, Semenets TN, Budagova KR, Shinkarkina AP, Kabakov AE, Poverennyi AM (2001) The protective effect of thermal treatment before irradiation on CFUs from bone marrow of mice: a potential involvement of heat shock proteins. Radiats Biol Radioecol 41:153–156 Mendonca LS, Nobrega C, Hirai H, Kaspar BK, Pereira de Almeida L (2015) Transplantation of cerebellar neural stem cells improves motor coordination and neuropathology in Machado-Joseph disease mice. Brain 138:320–335. doi:10.1093/brain/awu352 Mnich K, Carleton LA, Kavanagh ET, Doyle KM, Samali A, Gorman AM (2014) Nerve growth factor-mediated inhibition of apoptosis post-caspase activation is due to removal of active caspase-3 in a lysosome-dependent manner. Cell Death Dis 5:e1202. doi:10. 1038/cddis.2014.173 Muchowski PJ, Schaffar G, Sittler A, Wanker EE, Hayer-Hartl MK, Hartl FU (2000) Hsp70 and hsp40 chaperones can inhibit selfassembly of polyglutamine proteins into amyloid-like fibrils. Proc Natl Acad Sci USA 97:7841–7846. doi:10.1073/pnas. 140202897 Nobrega C, Nascimento-Ferreira I, Onofre I, Albuquerque D, Hirai H, Deglon N, de Almeida LP (2013) Silencing mutant ataxin-3 rescues motor deficits and neuropathology in Machado-Joseph disease transgenic mice. PLoS ONE 8:e52396. doi:10.1371/ journal.pone.0052396 Ooboshi H et al (2005) Postischemic gene transfer of interleukin-10 protects against both focal and global brain ischemia. Circulation 111:913–919. doi:10.1161/01.CIR.0000155622.68580.DC Peng XM, Gao L, Huo SX, Liu XM, Yan M (2015) The mechanism of memory enhancement of Acteoside (Verbascoside) in the senescent mouse model induced by a combination of D-gal and AlCl3. Phytother Res 29:1137–1144. doi:10.1002/ptr.5358 Phinney DG, Prockop DJ (2007) Concise review: mesenchymal stem/multipotent stromal cells: the state of transdifferentiation and modes of tissue repair–current views. Stem Cells 25:2896–2902. doi:10.1634/stemcells.2007-0637 Pittenger MF et al (1999) Multilineage potential of adult human mesenchymal stem cells. Science 284:143–147 Ponnaiyan D, Bhat KM, Bhat GS (2012) Comparison of immunophenotypes of stem cells from human dental pulp and periodontal ligament. Int J Immunopathol Pharmacol 25:127–134. doi:10. 1177/039463201202500115 Ryu KH et al (2014) Tonsil-derived mesenchymal stem cells alleviate concanavalin A-induced acute liver injury. Exp Cell Res 326:143–154. doi:10.1016/j.yexcr.2014.06.007 Scherzinger E et al (1999) Self-assembly of polyglutamine-containing huntingtin fragments into amyloid-like fibrils: implications for Huntington’s disease pathology. Proc Natl Acad Sci USA 96:4604–4609 Shakkottai VG, do Carmo Costa M, Dell’Orco JM, Sankaranarayanan A, Wulff H, Paulson HL (2011) Early changes in cerebellar physiology accompany motor dysfunction in the polyglutamine disease spinocerebellar ataxia type 3. J Neurosci 31:13002–13014. doi:10.1523/JNEUROSCI.2789-11.2011 Sheng Z, Yao Y, Li Y, Yan F, Huang J, Ma G (2013) Bradykinin preconditioning improves therapeutic potential of human endothelial progenitor cells in infarcted myocardium. PLoS ONE 8:e81505. doi:10.1371/journal.pone.0081505 Troyer DL, Weiss ML (2008) Wharton’s jelly-derived cells are a primitive stromal cell population. Stem Cells 26:591–599. doi:10.1634/stemcells.2007-0439 Urban MJ, Dobrowsky RT, Blagg BS (2012) Heat shock response and insulin-associated neurodegeneration. Trends Pharmacol Sci 33:129–137. doi:10.1016/j.tips.2011.11.001

Cell Mol Neurobiol Zhang J, Li Y, Chen J, Yang M, Katakowski M, Lu M, Chopp M (2004) Expression of insulin-like growth factor 1 and receptor in ischemic rats treated with human marrow stromal cells. Brain Res 1030:19–27. doi:10.1016/j.brainres.2004.09.061 Zhang J et al (2005) Human bone marrow stromal cell treatment improves neurological functional recovery in EAE mice. Exp Neurol 195:16–26. doi:10.1016/j.expneurol.2005.03.018 Zhang J et al (2006) Bone marrow stromal cells reduce axonal loss in experimental autoimmune encephalomyelitis mice. J Neurosci Res 84:587–595. doi:10.1002/jnr.20962

Zhang MJ et al (2011) Human umbilical mesenchymal stem cells enhance the expression of neurotrophic factors and protect ataxic mice. Brain Res 1402:122–131. doi:10.1016/j.brainres.2011.05. 055 Zuk PA et al (2001) Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng 7:211–228. doi:10.1089/107632701300062859

123