Time-dependent fate of transplanted neural precursor cells in experimental autoimmune encephalomyelitis mice

Experimental Neurology 230 (2011) 16–26

Contents lists available at ScienceDirect

Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / yex n r

Time-dependent fate of transplanted neural precursor cells in experimental autoimmune encephalomyelitis mice Angeliki Giannakopoulou a,b, Nikolaos Grigoriadis a,⁎, Eleni Polyzoidou a, Athanasios Lourbopoulos a, Eleni Michaloudi b, Georgios C. Papadopoulos b a b

Second Department of Neurology, AHEPA University Hospital, Aristotle University of Thessaloniki, Macedonia, Greece Laboratory of Anatomy, Histology and Embryology, Veterinary School, Aristotle University of Thessaloniki, Macedonia, Greece

a r t i c l e

i n f o

Article history: Received 4 January 2010 Revised 23 March 2010 Accepted 12 April 2010 Available online 24 April 2010 Keywords: Neural stem cells EAE Multiple Sclerosis CNS Migration Inflammatory cytokines Transplantation TNFa TGFb INFγ

a b s t r a c t Transplanted Neural Precursor Cells (NPCs) are capable of long-distance migration inside the inflamed CNS, but exhibit limited myelinating capacities in animal models of Multiple Sclerosis (MS). Inflammation seems to be both beneficial for the recruitment and migration of NPCs and restrictive for their terminal differentiation. In the present study, a set of transplantation experiments was applied in order to investigate the migratory potential, the differentiation pattern and long-term survival of NPCs in Experimental Autoimmune Encephalomyelitis (EAE) mice, the animal model of MS. The in vitro differentiation potential of NPCs in the presence of either pro- (TNFa, INFγ) or anti- (TGFb) inflammatory cytokines was also analyzed. According to the in vivo results obtained, at the acute phase of EAE only a small fraction of transplanted NPCs succeed to differentiate, whereas at chronic phase most of them followed a differentiation process to glial cell lineage along white matter tracts. However, this differentiation was not fully completed, since 8 months after their transplantation a number of NPCs remained as pre-oligodendrocytes. Glial differentiation of NPCs was also found to be inhibited or promoted following their treatment with TNFa or TGFb respectively, in vitro. Our findings suggest that inflammation triggers migration whereas the anti-inflammatory component is a prerequisite for NPCs to follow glial differentiation thereby providing myelinating oligodendrocytes. It is speculated that the fine balance between the pro- and anti-inflammatory determinants in the CNS may be a key factor for transplanted NPCs to exhibit a better therapeutic effect in EAE and MS. This article is part of a Special Issue entitled "Interaction between repair, disease, & inflammation." © 2010 Elsevier Inc. All rights reserved.

Introduction Multiple Sclerosis (MS) is the most commonly diagnosed immune mediated demyelinating disease in young adults. Demyelination, axonal degeneration, neuronal dysfunction and apoptosis are key features in MS pathology and current evidence suggests that remyelination to some extent also occurs spontaneously (Lassmann et al., 1997; Patrikios et al., 2006). Adult brain contains Neural Precursor Cells (NPCs) in regional pools that are able to be engaged in endogenous repair (Galli et al., 2003). However, the size of the lesion may surpass the capacity of endogenous NPCs to repair the damage, or the same cells could become the targets of the pathological process with adverse repercussions on the innate ability of the CNS to selfrepair (Pluchino and Martino, 2007). Therefore, in case the aetiology of disease progression resides in the host genome, transplantation of exogenous NPCs from a non-diseased source could be an option. ⁎ Corresponding author. Second Department of Neurology, AHEPA University Hospital, Aristotle University of Thessaloniki, 1 Stilp. Kyriakidi str, 54636 Thessaloniki, Macedonia, Greece. E-mail address: [email protected] (N. Grigoriadis). 0014-4886/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2010.04.011

Many studies have already proved that transplanted NPCs are capable of long-distance migration inside an inflamed CNS. They also exert immunosuppressive, anti-inflammatory and neurotrophic functions following their interaction with an inflammatory environment, in diseases such as EAE and MS (Einstein et al., 2006b; Pluchino et al., 2003). However, their myelinating capacities are limited, since a small percentage of transplanted NPCs differentiate into mature oligodendrocytes. Although the inflammation seems to be important for the recruitment and migration of NPCs, (Picard-Riera et al., 2002) at the same it may be prohibitive for their terminal differentiation into cells of oligodendrocyte lineage. In animal models of MS, such as the Experimental Autoimmune Encephalomyelitis (EAE) transplantation experiments reveal that NPCs differentiate mainly into glia (Ben-Hur et al., 2003b, 2007; Einstein et al., 2003, 2006a,b; Pluchino et al., 2003). However, it is not known how the inflammatory environment dictates the differentiation fate of NPCs as disease develops and if there is any correlation between the differentiation of NPCs and the phase of disease or their localisation in the brain. Moreover, there is little knowledge about the long-term survival of transplanted NPCs after their integration into the host parenchyma as similar experiments were terminated maximum

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

17

1–3 months after the graft implantation (Franklin et al., 1997; Shields et al., 1999; Woodruff and Franklin, 1999). In the present study we studied i) the long-term survival, migration and differentiation of transplanted NPCs in the inflammatory and demyelinating environment of chronic EAE and ii) the in vitro differentiation potential of NPCs in the presence of either proinflammatory (TNFa, INFγ) or anti-inflammatory (TGFb) cytokines.

and anti-GFAP (Dako, 1:800), anti-NG2 (Chemicon, 1:200) or antiGalC (Chemicon, 1:100). Cells were incubated with primary antibodies overnight at 4 °C and exposed to the appropriate fluorochrome-labelled secondary antibodies for 1.5 h at RT (all from Jackson ImmunoResearch Laboratories). Counterstaining was done with Dapi. Labelled cells were examined and photographed at appropriate wavelengths using a Nikon fluorescent microscope.

Materials and methods

Counting and statistical analysis of in vitro immunostained NPCs

Induction of chronic EAE and clinical evaluation of the animals

Quantification was performed blinded from photomicrographs systematically acquired from three independent cultures. A total of 60 digital photomicrographs for each cytokine and control and for each pair of primary antibodies (20 of the first primary antibody, 20 of the second primary antibody, and 20 of the Dapi-stained nuclei), were taken at 20× magnification. Using Adobe Photoshop software, the images of immunoreactive cells were overlaid and quantified. The number of cells in which a nucleus was visible was counted in each photomicrograph. Percentage numbers of positive cells that express each antigenic marker or coexpress both were calculated for each photograph. All the percentages combined and averaged to give a total percentage number for the antigenic markers and expressed as Mean ± SE. Data was statistically evaluated using One Way ANOVA with Bonferroni multiple comparison post-hoc testing between individual pairs of groups and P values b0.05 were deemed significant.

Animal experimentation received the approval of Veterinary Directorate of Thessaloniki and was conducted under compliance with National Institutes of Health guidelines, Greek Government guidelines and the local ethics committee. All possible adequate measures were taken to minimize pain or discomfort of treated animals. The induction of chronic EAE was held in five to six-week-old C57Bl/6 female mice, immunized subcutaneously with MOG 35–55 as described previously by (Einstein et al., 2006b). Clinical signs of EAE appeared typically after 10–12 days, reaching the maximum clinical score after 18– 22 days (acute phase). Remission period began 25–30 days post induction (PI) of EAE with modest or little gradual clinical improvement thereafter up to the point where the animals entered a plateau (45–50 days PI, chronic phase). Animals were examined daily for neurological signs (score: 0: healthy or asymptomatic; 1: partial loss of tail tonicity; 2: tail paralysis, ataxia; 3: hind limb weakness and difficulty to roll over; 4: hind limb paralysis; 5: tetraparesis; 6: death. Severely ill animals were rehydrated by intraperitoneal injections of saline until neurological improvement that enabled them to feed themselves independently. Isolation of NPCs from GFP+ newborn mice and growth of neurospheres NPCs were isolated from the cerebral hemispheres of newborn GFP-expressing mice (“Green mouse FM131” C57BL/6 TgN(act-EGFP) OsbC14-Y01-FM131) (Okabe et al., 1997), courtesy of Dr. M. Okabe (Osaka University, Japan) as described previously (Einstein et al., 2006b). The cells were suspended in N2 medium, plated in T-75 uncoated flask (107 cells per flask) and supplemented with 10 ng/ml basic fibroblast growth factor (bFGF) (R&D) and 20 ng/ml epidermal growth factor (EGF) (R&D), added daily. Under these conditions approximately 0.2% of cells proliferated into clusters of small round cells that grew into free-floating neurospheres within 7 days and then collected and diluted in DMEM/F12 medium. A quantity of the sphere preparation was used for quantification of neurospheres and after mechanical dissociation the total cell number and cell viability using trypan blue exclusion were measured. The neurosphere preparation for transplantation experiments was diluted in DMEM/F12 to the concentration of 500 neurospheres/μl approximately. In vitro differentiation of mouse NPCs under inflammatory cytokines To analyze the in vitro differentiation potential of neurospheres in the presence of the inflammatory cytokines TNFa, INFγ, TGFb, approximately 300 floating spheres/150 μl N2 medium were adhered to 35 mm tissue culture dishes coated with 10 μg/ml poly-D-lysine and 10 μg/ml fibronectin (Sigma). Following washing with DMEM/ F12, culture dishes were incubated in 2 ml N2 medium without FGF2 and EGF for 2 and 5 days in 37 °C and 10% CO2. On the first day following the induction of differentiation, 500 IU/ml N2 medium of TNFa (recombinant mouse tumor necrosis alpha), INFγ (recombinant mouse interferon γ), or TGF-b1 (recombinant human transforming Growth Factor beta 1), all from R&D, was added separately in each culture dish. The differentiation potential of these cells was assessed by double immunocytochemistry with anti-nestin (Chemicon, 1:200)

Neurosphere transplantation The immunized C57Bl/6 mice on day 7 after the induction of EAE were used as recipients of GFP+ NPCs. Stereotactic surgery was performed under deep anesthesia with the combination of ketamine (100 mg/kg)/xylazin (10 mg/kg) and with the animals mounted in a miniaturized stereotactic frame. Each animal was injected with 5000 spheres in a volume of 10 μl in both lateral ventricles (Coordinates: Lateral to Bregma = 1 mm, Bregma zero, Ventral to Bregma = 2.5 mm) (Paxinos and Franklin, 2001). Mice with EAE that received the injection of the DMEM/F12 medium without cells in the lateral ventricles were used as controls. Experimental schedule and brain fixation Animals were divided into 2 groups: control and experimental, i.e. transplanted with NPCs. Each group was divided further into 3 subgroups, according to the time of their sacrifice PI of EAE. Each subgroup of control group was consisted of 10 animals, while 1st, 2nd and 3rd subgroups of experimental group were consisted of 20, 40 and 20 animals respectively. The 1st group was sacrificed at the peak of the disease (17 days PI), the 2nd group during the chronic phase and the 3rd one, 8 months PI (Supplementary Fig. 1). Animals under deep ether anesthesia received transcardial perfusion with 4% paraformaldehyde (PFA) solution in PBS pH 7.4. Brains were quickly removed, post fixed by immersion in the same fixative at 4 °C for 4 h. Afterwards tissues were placed in 30% sucrose solution in PBS, deep frozen in isopentane and 6 μm sagittal and coronal serial sections were collected. For each section collected the next section was thrown away. All sections were examined with a fluorescent microscope for the detection of GFP+ cells. Measurement of transplanted cells migration Spatial distribution of GFP+ cells was evaluated in a blinded manner in each brain by microscopic visualization of serial sagittal or coronal sections. The migratory potential of transplanted GFP+ cells was estimated by composites of sagittal or coronal photomicrographs (at ×20 or × 10 magnification) comprising the entire migratory

18

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

pathway followed by transplanted cells. The distance covered by the migrating transplanted cells was measured from the edge of the lateral ventricle with the use Image of Pro Plus software and was expressed as a mean ± SE for each region (e.g. corpus callosum, fimbria of hippocampus etc.) and for each animal group or phase of disease.

Immunofluorescent staining of NPCs in vivo To determine whether transplanted cells were capable of differentiating along neural lineages, immunofluorescent stainings were performed. Section sampling was performed after the determination of spatial distribution of transplanted cells and the definition of the serial section population. Moreover, the sagittal brain sectioning was chosen in order to include in one section all regions of interest (i.e. corpus callosum, fimbria of hippocampus, internal capsule, periventricular areas). In line with this concept, brain area containing at least 20 GFP+ cells per sagittal serial section were identified between L = 0.49 mm and L = 1.45 mm lateral to bregma. We used the principles of “systematic random sampling” based on which, serial sections (the first being randomly selected) were sampled in predetermined intervals depending on the total number of sections per animal. Consequently, samples of 12 sections per animal per immunophenotypic marker studied, were analyzed. Data were expressed as the average percentage of immunoreactive transplanted cells expressing a specific antigenic marker (per section) ± SEM in any region of interest for each animal group. Sections were incubated with the following primary antibodies: mouse IgG anti-Nestin (Chemicon, 1:200), mouse IgM anti-PSANCAM (Chemicon, 1:200), mouse IgG anti-NeuN (Chemicon, 1:500), mouse IgG anti-tubulin Beta III (Chemicon, 1:200), rabbit IgG AntiNG2 (Chemicon, 1:200), mouse IgM anti-O4 (Chemicon, 1:100), mouse IgG anti-GalC (Chemicon, 1:100), mouse IgG anti-CNPase (Chemicon, 1:100) and rabbit IgG anti-GFAP Polyclonal (Dako, dilution 1:500). Secondary antibodies (all from Jackson ImmunoResearch Laboratories) were Rhodamine RedTM-X-conjugated and used at a dilution of 1:200. Counterstaining was done with Dapi. Labelled cells were examined and triads of digital microphotographs (one with fluorescent GFP+ cells, one with TRITC labelled primary antibody, and one with Dapi-stained nuclei), were taken using a Nikon fluorescent microscope. Under strict criteria and rigorous attention to detail, immunolabelling for each antigenic marker used were determined at numerous triads of microphotographs. GFP immunohistochemistry was also performed in order to investigate if yellowish cell like structures were necrotic or apoptotic GFP+ cells. Sections were incubated with rabbit IgG anti-GFP Polyclonal (Abcam, dilution 1:4000) diluted in PBS containing 5% NGS and 2.5% BSA for 1 h at RT. After washes in PBS a labelled polymer-HRP anti-rabbit (Envision, dilution 1:200) was used for 30 min at RT and then DAB was added for 5 min to reveal GFP. After thorough wash with DW, hematoxylin (Merk) staining was performed for nuclear signalling.

Statistical analysis All quantitative evaluations were made in a uniform fashion by an observer who was blinded to the experimental groups. Quantitative data from several experiments were pooled to obtain the mean ± SE for each experimental group. Due to the restricted size of samples, Kruskal–Wallis test was implemented for each parameter. The Mann–Whitney U-test was used for comparisons between any two groups. All analyses were conducted using the statistical software program SPSS v17. Differences were deemed statistically significant at P b 0.05.

Results Transplantation of NPCs attenuated the severity of clinical disease The clinical score was evaluated daily in individual mice and the cumulative severity of disease “area under curve” was calculated at day 18 PI of EAE (peak of the acute phase), at day 50 PI (representing the chronic phase) and 8 months PI. There was no difference in time initiation of the disease onset between controls and NPCs transplanted animals. The decrease in disease severity by neurosphere transplantation was already noted at acute phase of disease, but the most significant attenuation of the clinical paralytic signs was observed at chronic phase (Supplementary Table 1). The spatial distribution and migratory potential of transplanted cells at different phases of EAE At the acute phase of the disease, many of the transplanted GFP+ cells remained clustered within the ventricular space, attached to the wall of LV or trapped into the choroid plexus. A minority of them was found in the periventricular parenchyma, up to a depth of several layers of cells. Extensive migration at this phase was limited toward inflammatory infiltrations adjacent to the LVs and in the border between thalamus and hippocampus (choroid fissure) (292.7±32.6 μm, Pb 0.001 compared to periventricular localization). Single GFP+ cells were also found along the needle insertion site and nearby subpial surface due to a back-leakage of suspension. At chronic phase, the migration of GFP+ cells was mainly observed in white matter tracks lacking a clear cellular inflammatory process and especially in corpus callosum, ventricular hippocampal commissure, fimbria of hippocampus and internal capsule. Over 20 GFP+ cells per section were detected between L = 0.49 mm and L = 1.45 mm lateral to bregma and B = −11.2 mm and B = 0.6 mm rostra-caudal to bregma. However a percentage of GFP+ cells exhibited limited migration, i.e. were found either to cover the LVs' wall or periventricular in proximity to the nuclei of septum and thalamus. Eight months post transplantation; there were no transplanted GFP+ cells in the form of neurospheres or aggregates inside the LVs. All GFP+ cells integrated by the host parenchyma and were located in the same sites as their counterparts of chronic phase (Fig. 1). Comparison of the distances of GFP+ cells inside the brain parenchyma at different time points indicates that there is statistical difference between acute and chronic phase (Kruskal–Wallis test, P b 0.001), but not between chronic phase and 8 months PI. At chronic phase of EAE the longest migratory pathway was inside the corpus callosum, where the greatest distance covered by GFP+ cells was found to be 1.4 mm at coronal sections (Table 1). The differentiation potential of transplanted cells The overall co-expression of the various markers used is illustrated in Fig 2. Triads of images were used to identify GFP+ cells, simultaneously immunoreactive for an individual marker and the respective Dapi-stained nucleus (Supplementary Figs. 1, 2, 3). Used sections were thin enough not to permit overlapping of two different cells. At the acute phase of EAE transplanted cells that remained in the form of aggregates of neurospheres expressed the antigenic markers of nestin (80.7±2.1%), glial fibrillary acid protein (GFAP, 76.2±3.9%) and the polysialylated form of the neural adhesion molecule (PSA-NCAM, 30.8± 7.5%). The average number of GFAP+/GFP+ and nestin+/GFP+ cells were not significantly different (P=0.380), indicating that these two antigenic markers were coexpressed. The GFP+ cells that had moved into the parenchyma in a depth of several cell layers acquired processes and morphology of differentiating cells and lost the expression of nestin and PSA-NCAM. The GFAP was the major marker expressed by GFP+

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

19

Fig. 1. Morphologies acquired by GFP+ cells at acute phase of EAE in the perivascular glia limitans (A), subpial surface of cortex (Cx), B) and needle insertion region (C). D. Migratory GFP+ cells at chronic phase of EAE in genu of corpus callosum (Gc) with elongated cell body (thin arrow), extending long processes arranged in parallel with axons (thick arrow). E. GFP+ cells inside the corpus callosum (Cc). In boxed area there are small cells that are connected with longitudinal processes arrayed in parallel with axons resembling to oligodendrocytes. F. Again GFP+ cells inside the Cc. In boxed area there is a glial cell bearing velate processes extended in all directions. G. Mature glial GFP+ cells with fine, long and highly branched processes detected in fimbria of hippocampus (LV: lateral ventrical). Thin arrows indicate their processes. Scale bars: 10 μm (A,G); 100 μm (B,C,D,E,F).

Table 1 Covered distances (μm) of GFP+ cells into the brain parenchyma at different time points. Acute phase

Perivascular infiltrations Periventricular Corpus callosum (sagittal sections) Corpus callosum (coronal sections) Fimbria of hippocampus (hip.) Rostral Migratory Stream (RMS) Total

Chronic phase

8 months PI

Mean ± S.E.

Median

Maximum

Mean ± S.E.

Median

Maximum

Mean ± S.E.

Median

Maximum

292.7 ± 32.6‡ 39.8 ± 3.1 n. d. n. d. n. d. n. d 137.5 ± 18.3

271 31.8 n. d. n. d. n. d. n. d 53.6

723.7 106.1 n. d. n. d. n. d. n. d 723.7

v.f. 21.0 ± 1.6# 218.6 ± 19.1b 671.0 ± 27.7 70.7 ± 5.9 n. d 329.4 ± 17.7†

v.f. 20.4 153.4 689.6 49.2 n. d 172.7

v.f. 41.64 991.5 1436.8 197.8 n. d 1436.8

n. d. 35.8 ± 2.4!! 137.9 ± 6.4§ n.a. 70.0 ± 4.9 80.2 ± 9.3 94.8 ± 3.7

n. d. 28.2 119.5 n.a. 45.9 37.3 56.8

n. d. 77.1 507.2 n.a. 298.0 475.0 507.2

S.E.: standard error of mean, n.a.: not analyzed, n.d., no GFP+ cells detected, v.f.: very few GFP+ cells detected. † P b 0.001 when compared with acute phase of EAE and 8 months PI. ‡ P b 0.001 when compared with periventricular localization at acute phase of EAE. b P b 0.001 when compared with periventricular and fimbria localization at chronic phase and corpus callosum localization at 8 months PI. #P b 0.001 when compared with fimbria localization at chronic phase and periventricular localization at acute phase of EAE and 8 months PI. §P b 0.001 when compared with periventricular, fimbria and RMS localization at 8 months PI. !!P = 0.003 when compared with fimbria localization and P = 0.004 when compared with RMS localization at 8 months PI.

20

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

Fig. 2. Merged photomicrographs of GFP+ cells (green) inside the EAE mice brain at different time points of the disease expressing antigenic markers (red) of glial cell lineage. Nuclei are labelled with Dapi. Note that in the acute phase the GFP+ cells are found as neurospheres coexpressing nestin and GFAP or as astrocytes in periventricular infiltrates, whereas in the chronic phase and 8 months PI GFP+ cells exhibit extensive migration in white matter and express antigenic markers of mature glial cells (LV: Lateral Ventricle, Cc: Corpus Callosum, Ic: Internal capsule). Arrows indicate immunoreactive GFP+ cells, whereas arrowheads immunonegative GFP+ cells. Scale bars represent 100 μm.

integrated cells (83.6±2.2%), followed by nestin (52.6±11.2%), PSANCAM (26.8±5.3%) and NG2 (9.1±2%) (Table 2, Fig 3). At the chronic phase of EAE the majority of GFP+ cells acquired complex morphology with simple or ramified processes and expressed the antigenic markers of pre- and mature glial cells (Fig. 5). No GFP+ cells expressing either neuronal or microglial antigens were found. GFAP was the main antigenic marker expressed at this phase (52.2 ± 1.9%) by transplanted cells, followed by the expression of NG2 (28.3 ± 5.0%) and GalC (15.2 ± 2.0%). GFAP+/GFP+ cells detected perivascularly at chronic phase of EAE were smaller, with tiny processes and faint expression of GFP, resembling resting astroglia. Kruskal–Wallis test revealed that the expression of GFAP from GFP+ cells was site specific (P = 0.001). The average percentage of GFAP+/GFP+ cells located in fimbria of hippocampus was significantly higher compared to those located periventricularly (P = 0.013) and in corpus callosum (P b 0.001). Migrating GFP+/NG2+ progenitors were also detected at chronic phase of EAE having covered the longest distances among the transplanted ones. According to statistical analysis the localization of GFP+ cells had no effect on the expression of NG2 (Kruskal–Wallis test, P = 0.264) or GalC (Table 2). Occasional GFP+ cells and fibers expressing CNPase were observed at corpus callosum and fimbria of hippocampus but the quantification of single cells expressing this marker was not possible, as immunoreactive cellular processes could not always be attributed to the corresponding cell bodies.

8 months PI of EAE almost all of the GFP+ cells were mature and fully differentiated glial cells as determined by their complex morphology, their long and ramified processes and the expression of mature glial cells' antigenic markers. Most of the GFP+ cells were found as resting astrocytes (GFP+/GFAP+, 40.1 ± 4.6%) and mature oligodendrocytes (GFP+/GalC+, 30.6 ± 4.3%). GFP+/CNPase+ cells were also abundant but their quantification wasn't performed for the reasons already mentioned. Interestingly, some of the transplanted cells expressed the antigen marker NG2 (22.3 ± 3.5%) and had morphologic features of oligodendrocyte progenitors (OP) (Fig. 2). This finding may explain in part the failure of endogenous counterparts to fully restore the oligodendrocyte damage in EAE model, as some transplanted OP were unable to fully differentiate. Statistical analysis revealed that the expression of GFAP by GFP+ cells is strongly influenced by the phase of the disease (P b 0.001) and the region detected inside brain (P b 0.001). The average percentage of GFAP+/GFP+ cells in acute phase is significantly higher than those in chronic phase (P b 0.001) and 8 months PI (P b 0.001). According to statistical analysis, the NG2+ expression by GFP+ cells wasn't influenced by the phase of the disease (P = 0.205) and the region detected inside brain (P = 0.475). It should be emphasized that at chronic phase the majority of NG2+/GFP+ cells were migrating OP, but 8 months PI matured to more complex pre-oligodendrocytes. At last the average percentage of GalC+/GFP+ cells at 8 months PI was

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

21

Table 2 Differentiation of GFP+ cells at different phases of EAE. Differentiation of GFP+ cells at acute phase of EAE GFAP

NG2

Nestin

PSA-NCAM

P*

Localization

Mean±S.E.

Median

Mean±S.E.

Median

Mean±S.E.

Median

Mean±S.E.

Median

Perivascular infiltrations (PI)

88.3 ± 2.5

88.4

2.9 ± 0.5

3.23

n.i.d

n.i.d

n.i.d

n.i.d

GFAP/NG2: b0.01

15.6

GFAP/NG2: b0.05, GFAP/PSA-NCAM: b 0.001, NG2/nestin: b 0.05, Nestin/PSA-NCAM: b0.005 GFAP/NG2: b0.05, GFAP/Nestin: b 0.05,

Aggregates of neurospheres (AN)

76.2 ± 3.9

75.0

16.8 ± 4.4

13.1

80.7 ± 2.1

80.0

30.8 ± 7.5

Needle insertion region (NIR)

82.2 ± 3.6

80.7

11.0 ± 1.4

11.0

17.5 ± 3.1

14.8

20.8 ± 7.1

17.5

Subpial surface (SS)

92.6 ± 4.4

93.5

n.i.d

n.i.d

n.i.d

n.i.d

n.i.d

n.i.d

GFAP/PSA-NCAM: b 0.05

Total

83.6 ± 2.2

83.6

P*

AN/SS: b0.05

9.1 ± 2

8.6

PI/AN b0.05, PI/NIR b 0.05

52.6 ± 11.2

75.0

26.8 ± 5.3

15.9

GFAP/NG2: b0.001, GFAP/PSA-NCAM: b 0.001, NG2/nestin: b 0.001, GFAP/Nestin: =0.01, NG2/PSA-NCAM: = 0.01, Nestin/PSA-NCAM: b0.05

AN/NIR b 0.05

Differentiation of GFP+ cells at chronic phase of EAE GFAP

NG2

GalC

P*

Localization

Mean ± S.E.

Median

Mean ± S.E.

Median

Mean ± S.E.

Median

Periventricularly (P)

52.8 ± 3.1

50.0

31.7 ± 9.3

25.0

9.8 ± 2.5

8.5

Covering the LVs' walls (CLV) Needle insertion region (NIR) Corpus callosum (CC) Fimbria of hippocampus (FI)

n.i.d 51.8 ± 3.6 46.3 ± 22 67.0 ± 3.0

n.i.d 50.5 45.5 70.4

48.4 ± 18.7 22.4 ± 6.9 31.2 ± 11.7 13.7 ± 4.0

39.6 20.0 25.0 11.9

n.i.d n.i.d 21.0 ± 4.0 13.3. ± 1.9

n.i.d n.i.d 22.5 15.0

Total

52.2 ± 1.9

50.0

28.3 ± 5.0

25.0

15.2 ± 2.0

15.0

P*

P/CC: b 0.001, P/FI: b0.05, NIR/FI: = 0.01,

n.s

n.s

NG2

GalC

GFAP/GalC: b 0.01, NG2/GalC: b 0.05 GFAP/NG2: b0.01 GFAP/GalC: b 0.001 GFAP/NG2: b0.01 GFAP/GalC: b 0.01 GFAP/NG2: b0.001, GFAP/GalC: b 0.001, NG2/GalC: b 0.05

Differentiation of GFP+ cells 8 months of EAE GFAP

P*

Localization

Mean ± S.E.

Median

Mean ± S.E.

Median

Mean ± S.E.

Median

Corpus callosum (CC) Fimbria of hippocampus (FI) Total P*

30.3 ± 27 49.9 ± 1.2 40.1 ± 4.6 CC/FI: = 0.05

30.0 50.0 41.4

20.7 ± 6.0 25.5 ± 3.1 22.3 ± 3.5 n.s.

18.9 25.9 25.0

35.9 ± 5.4 21.0. ± 1.5 30.6 ± 4.3 n.s.

39.2 20.0 24.0

GFAP/GalC: = 0.05 GFAP/NG2: b0.05, GFAP/NG2: b0.01

S.E.: standard error of mean. Values represent the percentage of respective immunopositive cells from total number of GFP+ cells at the specific region of interest (%). n.i.d: no immunoreactive GFP+ cells detected, n.s: no statistical significant difference between immunoreactive GFP+ cells of each region detected. *Only the statistical significant differences were noted.

significantly higher than the corresponding of chronic phase (P = 0.017) (Table 2, Fig 3). Differentiation pattern of NPCs in vitro in the presence of inflammatory cytokines At the day 2 after the induction of differentiation, the number of nestin positive cells was higher in dishes treated with pro-and antiinflammatory cytokines (TNFa 81.2 ± 4.0%; INFγ 84.1 ± 3.6%; TGFb 83.4 ± 4.9%) compared to controls (71.,2 ± 5.2%), but there were no statistical significant differences. GFAP+ cells were numerous in dishes treated with TNFa (14.8 ± 2.5% compared to controls 10.0 ± control dishes and even less at dishes treated with INFγ and TGFb. However statistical significant differences were evident between the different types of cytokines, but not between cytokines and controls. Most cells coexpressing the antigenic markers GFAP and nestin were

detected in dishes treated with TNFa (8.1 ± 1.5%; INFγ 5.2 ± 0.9%; TGFb 3.8 ± 0.8%; controls 4.8 ± 0.5%). NG2+ cells were more at dishes supplemented with the cytokines TNFa (23.2 ± 2.5%) and TGFb (23.0 ± 2.8%) compared to controls (17.5 ± 2.5%), even though not statistically significant. At day 5 after the induction of differentiation, treatment with the inflammatory cytokine TNFa (57.1 ± 5.0%) had as a result the inhibition of differentiation since there were significantly more nestin positive cells compared to dishes treated with TGFb (28.9 ± 4.8%; P = 0.001) and controls (35.8 ± 4.7%; P = 0.015). Dishes treated with TGFb had the lowest percentage of nestin positive cells indicating that this anti-inflammatory cytokine induce the differentiation of NPCs. The significant increase of GFAP+ cells at dishes treated with TGFb (56.7 ± 2.2%; TNFa 31.7 ± 4.1%; INFγ 31.8 ± 4.4%; controls 34.9 ±3.8%; P b 0.001 ), which most of them do not coexpress the antigenic marker nestin indicates that TGFb induce the differentiation of NPCs to the

22

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

glial cell lineage. Similarly GFAP+ cells were less at dishes treated with INFγ and TNFα and control dishes. At day 5 after the induction of differentiation, there was no statistically significant difference between NG2 positive cells in dishes supplemented with the cytokines TNFa, INFγ or TGFb compared to controls (TNFa 19.4 ± 2.8%; INFγ 25.7 ± 2.2%; TGFb 22.1 ± 2.8%; controls 23.4 ± 2.9%). However, it is noteworthy that NG2+ cells in dishes with cytokines had many ramified processes, smaller nucleus and lesser cytoplasm compared to controls. Cells coexpressing nestin and NG2 were significantly more at dishes treated with INFγ (18.8 ±

4.6%) compared to dishes treated with TGFb (5.9 ± 1.7%; P = 0.032) indicating the inhibition of oligodendrocyte differentiation by the former cytokine and the promotion of differentiation by the later cytokine. Moreover, differentiation (highly branched NG2+ cells) was most prominent at the sites where the cells were packed, indicating that the tissue support is critical for NPCs differentiation (Figs. 4 and 5).

Discussion Migratory potential of GFP+ transplanted cells at different time points of EAE Normal adult brain is unable to support the migration of transplanted NPCs and their incorporation into the parenchyma. Einstein et al. (2006a) provided evidence that transplanted NPCs can survive and maintain their stem cell characteristics inside the LV of normal adult brain mice in the form of neurospheres for prolonged time, though their extensive proliferation and migration are not feasible. On the contrary, destruction of different cell types due to injury, inflammation or demyelinated process is a prerequisite for extensive proliferation and migration of transplanted NPCs. Similarly, CNS damage-related inflammatory process and production of proinflammatory cytokines (Jin et al., 2001; Picard-Riera et al., 2002) induce proliferation and migration of endogenous NPCs residing in the SVZ and other regional pools. In vitro, the proinflammatory cytokines TNFα and IFNγ have been shown to trigger the migration of neurosphere-derived NPCs (Ben-Hur et al., 2003a) by increasing the attachment molecules not only on NPCs, but also on the migratory paths (Ben-Hur et al., 2006; Miller and Krangel, 1992; Sobel and Mitchell, 1989; Tourbah et al., 1997; Washington et al., 1994). Moreover, proinflammatory cytokines in vitro promote the expression of matrix metalloproteinases, (MMPs) which are important in cellular migration (Ben-Hur et al., 2006). Pluchino et al. (2003), who have studied the migratory behaviour of intraventricularly transplanted NPCs at chronic EAE, found out that the migratory potential of transplanted cells is maximized when NPCs undergo the stimulatory effect of proinflammatory cytokines and have plenty of time to penetrate inside the parenchyma (Pluchino et al., 2003). In the present study, transplantation of NPCs before the onset of EAE, revealed that the inflammatory environment is essential for the precursor's mobilisation, but the extensive migration of GFP+ cells probably needs the tissue environment present at chronic phase of EAE. It is pertinent to note that Th1 proinflammatory cytokines (TNF-α and INF-γ)-activated microglia fails to induce migration of progenitors cells (Compston and Coles, 2002; Miller and Shevach, 1998; Steinman, 2001a,b; Steinman and Conlon, 2001; Steinman et al., 2002), while on the contrary, anti-

Fig. 3. Graphs showing the average percentages of immunoreactive NPCs in vivo. Data in the graphs are expressed as Means ± SE. A. The bars express the average percentages of the respective subpopulation of GFP+ cells at different phases of EAE. Asterisks indicate significant differences between subpopulations at different phases of EAΕ. In the acute phase, the majority of transplanted NPCs coexpress the antigenic markers nestin and GFAP. Expression of GFAP by NPCs is strongly influenced by the phase of the disease (P b 0.001). The average percentage of GalC+/GFP+ cells at 8 months PI is significantly higher than the corresponding of chronic phase (P = 0.017), Cc: Corpus callosum. B. The bars express the average percentages of the respective subpopulations of GFP+ cells at the most frequent places of their localization at chronic phase. The expression of GFAP by GFP+ cells is site specific (P = 0.001). GFAP+/GFP+ cells located in fimbria of hippocampus are significantly more in relation to those located periventricularly (P = 0.013), in corpus callosum (P b 0.001) and in needle insertion region (P = 0.010). The localization of GFP+ has no effect on the expression of NG2 or GalC. C. The bars express the stacked percentages of the respective subpopulations of GFP+ cells at different time points of EAE. Note that GFAP expression by NPCs declines progressively as the disease develops, while the expression of antigenic markers of oligodendrocyte lineage is progressively increased. However, the expression of NG2 is not significantly different between the various phases (P = 0.205), but their morphology is different between chronic phase and 8 months PI.

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

23

Fig. 4. Merged photomicrographs acquired from NPCs cultures in the presence of the cytokines. TNFa, INFg, TGFb and controls after double immunohistochemistry for nestin and GFAP or NG2. Nuclei are labelled with Dapi. Note that most nestin+ cells are detected in dishes treated with TNFa and TGFb, whereas in dishes with TGFb the majority of cells differentiate into GFAP+ and NG2+ cells that do not coexpress nestin. The morphology of NG2+ is different in dishes treated with cytokines when compared to controls. Scale bars represent 100 μm.

inflammatory Th2 cytokines TGF-β stimulate the production of Hepatocyte Growth Factor (HGF) from microglia cells, which acquire a new phrenotype and become cells participating in repair promoting the migration of OPs (Lalive et al., 2005). Transplanted NPCs use the white matter tracts for their widespread migration into the brain parenchyma (Ader et al., 2001; Ben-Hur et al., 2003b; Einstein et al., 2006a,b; Spassky et al., 2002), and follow the same migratory pathways used by endogenous counterparts at inflammatory and demyelinated diseases (Nait-Oumesmar et al., 1999; Picard-Riera et al., 2002). Similarly, in our experiments at chronic phase of EAE the most common migratory pathways followed by NPCs were found to be the corpus callosum, the fornix, the fimbria of hippocampus, the rostral migratory stream (RMS) and the internal capsule. Undifferentiated GFP+ cells with bipolar morphology were found inside the white matter tracks, indicating that the migration of GFP+ cells was still in process, while fully differentiated GFP+ cells were located in less remote sites from LVs. Probably, the latter were not capable of covering greater distances because of their ramified processes and the connections they may have established with nearby host cells. Also, the covered distance by transplanted cells was not statistically different between chronic phase and 8 months PI, suggesting that extensive migration happened when inflammation was in process and in particular during remission period of EAE. Thus, the

termination of migration coincides with the resolution of inflammatory phenomena. NPCs that localized nearby the migratory pathways managed to penetrate and to cover long distance, whereas a proportion of them just spread upon LV wall, seeking probably a gate to a migratory pathway. Thus, the extent of NPCs migration inside brain parenchyma depends not only on the inflammatory environment but also on the site of initial attachment upon the LVs. A subpopulation of transplanted cells penetrated the LVs' wall and migrated to a depth of few cellular layers. These GFP+ cells were in locations underneath of which were grey matter's nuclei such as septal, thalamic and caudate/ putamen nuclei. Consequently, the intraventricular transplantation of NPCs does not appear to be a promising replacement therapy for affected grey matter in MS. The site-specific differentiation of transplanted NPCs cells Many studies reveal that transplanted NPCs populations can respond to persistent site-specific differentiation cues in the adult CNS (Herrera et al., 1999; Rubio et al., 2000). In our experiments, GFP+ cells located in close proximity to inflammatory infiltrates differentiated to type I astrocytes, those found covering the LV walls were identified as NG2+

24

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

ependymal cells in agreement with their morphological characteristics, and a large proportion of GFP+ cells located inside the white matter tracks were positive for the antigenic markers of OP and mature oligodendrocytes. In corpus callosum, the transplanted cells differentiated in astrocytes and oligodendrocytes, with the percentage of these two cell types having no statistical difference, while in the fimbria of hippocampus they preferred to follow an astrocyte fate. Finally, the failure of NPCs to neuronal differentiation should be attributed to their

inability to migrate into grey matter and neurogenic structures such as RMS, olfactory bulb (OB) and cortex. Differentiation arrest of transplanted NPCs at acute phase of EAE due to proinflammatory cytokines At the inflammatory environment of acute phase of EAE most of the transplanted NPCs remained undifferentiated and only a small fraction of them located in close proximity to inflammatory foci differentiated mainly into activated GFAP+/GFP+ astrocytes hedging against inflammatory infiltration. Recently, Pluchino and Martino (2006, 2007) proposed that the inflammatory environment prevailing in EAE and MS should be accounted for the inability of endogenous NPCs derived from SVZs to differentiate into myelinating oligodendrocytes. In an attempt to explain why most of the demyelinated lesions in MS are detected periventicularly, suggested that the inflammatory activity leads the NPCs of SVZ (concretely type B cells) to undergo symmetrical cellular divisions, thus preventing the production of glial precursors and their differentiation in myelinating oligodendrocytes (Martino and Pluchino, 2006, 2007; Pluchino and Martino, 2007; Pluchino et al., 2008). This idea comes in agreement also with our in vitro findings, where treatment of NPCs with the proinflammatory cytokines TNFa and INFγ results in the increase of nestin positive cells 5 days later compared to controls. Another likely explanation of differentiation arrest it could be that there weren't sufficient time for NPCs differentiation to oligodendrocyte lineage. From transplantation to sacrifice the time interval was hardly 11–14 days, while in similar experiments mature oligodendrocytes are detected 30 days after transplantation of neurospheres or OP in vivo (Vitry et al., 2001). Besides, transplanted cells at the acute phase of disease had not yet been migrated in the parenchyma to interact with micro-environment or demyelinated axons, so the initiative signalling for their differentiation into oligodendrocytes was absent. NPCs differentiation at chronic phase of EAE At chronic phase of EAE, the lack of transplanted GFP+/nestin+ cells contrasted the abundance of GFP+/NG2+ OP. This finding is in consistence with the notion that Th2 cytokines, such as TGFβ, IL-4 and IL-10, which are also expressed during the effector phase of chronic EAE, are upregulated and peak during remission, resulting in suppression of immune response (Issazadeh et al., 1998). In this line, the inflammatory cytokine TGFb has been proven here to promote in vitro the glial differentiation of NPCs. Th2 cytokines and neurotrophins, contrary to Th1 cytokines, are protective in experimental models of MS and induce the differentiation of NPCs and OP in vitro (Molina-Holgado et al., 2001). Chronic phase is also characterized by decreased proportion of GFAP+/GFP+ cells and increased proportion of NG2+/GFP+ cells. We believe that some Fig. 5. Graphs showing the average percentages of immunoreactive NPCs in vitro after treatment with pro- and anti-inflammatory cytokines at D2 and D5 PID. Data in the graphs are expressed as Means ± SE. A. The bars express the average percentage of nestin+ and GFAP+ NPCS at D2 PID. No statistical significant differences were detected in the expression of nestin by NPCs in different dishes. In terms of GFAP+ cells most of them were detected in dishes treated with TNFa. However, statistical significant differences were evident between the different types of cytokines, but not between cytokine-treated and controls. Most cells coexpressing the antigenic markers GFAP and nestin were detected in dishes treated with TNFa. B. The bars express the average percentages of nestin+ and GFAP+ NPCs after treatment at D5 PID. Asterisks indicate significant differences between subpopulations at dishes treated with different types of cytokines. There were significantly more nestin positive cells in dishes treated with TNFa compared to those treated with TGFb (P = 0.001) and controls (P = 0.015). The significant increase of GFAP+ cells in dishes treated with TGFb (P b 0.001 ), which most of them do not coexpress the antigenic marker nestin, indicates that TGFb induce the differentiation of NPCs to the glial cell lineage. C. The bars express the average percentages of NG2+ NPCs after treatment at D2 and D5 PID. No statistical significant differences were detected between cytokines treated and controls.

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

GFAP+/GFP+ cells detected at the acute phase were not actually differentiated astrocytes, but type B NPCs that later produce cells of the oligodendrocyte lineage (Doetsch et al., 1999; Menn et al., 2006). However, the fact that the majority of GFP+ cells in white matter tracks at chronic phase express the antigenic marker GFAP and possess fine radiating processes calls for another explanation. It is likely that these cells represent embryonic astrocytes or have radial glial features in order to guide the endogenous NPCs migration, or alternatively constitute ‘isomorphic’ reactive astrocytes which allow damage resolution (Williams et al., 2007). Blakemore et al. (2003) suggested that the influence of astrocytes on remyelination is dependent on their activation status and a distinction, therefore, has to be made between the role of astrocytes in acute and chronic areas of demyelination. The absence of intraparenchymal migratory GFP+ cells eight months after their transplantation, in conjunction with the presence of GFP+/ NG2+ cells showing morphological features of pre-oligodendrocytes could be attributed to the lack of either environmental cues calling for cellular differentiation or any more need for terminal differentiation to mature oligodendrocytes. In any way these GFP+/NG2+ cells, remaining as OP would probably be ready to form myelinating cells in a new demyelinating attack. Unfortunately, this would be a short-term response, as there wouldn't be new NPCs to resupply them. Therefore, it can be said that the inability of the transplanted (and probably of the endogenous) NPCs to follow a differentiation process which might provide adequate myelinating oligodendrocytes is the result of the persistent inflammatory environment prevailing in EAE and MS. Thus, the appropriate combination of immunoregulation and NPCs' transplantation might be considered as a therapeutic approach in MS. However, the possibility that anti-inflammatory or immunoregulatory agends, particularly those currently used in MS treatment, might interfere with the kinetics and/or the behavior of transplanted NPCs within CNS needs further clarification in future pharmacologicalcombined with cell-mediated- therapy studies. Acknowledgments We thank Ofira Einstein for experimental guidance, helpful consultations and constructive discussions. This work was supported by PENED 2001. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.expneurol.2010.04.011. References Ader, M., Schachner, M., Bartsch, U., 2001. Transplantation of neural precursor cells into the dysmyelinated CNS of mutant mice deficient in the myelin-associated glycoprotein and Fyn tyrosine kinase. Eur. J. NeuroSci. 14, 561–566. Ben-Hur, T., Ben-Menachem, O., Furer, V., Einstein, O., Mizrachi-Kol, R., Grigoriadis, N., 2003a. Effects of proinflammatory cytokines on the growth, fate, and motility of multipotential neural precursor cells. Mol. Cell. Neurosci. 24, 623–631. Ben-Hur, T., Ben-Yosef, Y., Mizrachi-Kol, R., Ben-Menachem, O., Miller, A., 2006. Cytokine-mediated modulation of MMPs and TIMPs in multipotential neural precursor cells. J. Neuroimmunol. 175, 12–18. Ben-Hur, T., Einstein, O., Mizrachi-Kol, R., Ben-Menachem, O., Reinhartz, E., Karussis, D., Abramsky, O., 2003b. Transplanted multipotential neural precursor cells migrate into the inflamed white matter in response to experimental autoimmune encephalomyelitis. Glia 41, 73–80. Ben-Hur, T., van Heeswijk, R.B., Einstein, O., Aharonowiz, M., Xue, R., Frost, E.E., Mori, S., Reubinoff, B.E., Bulte, J.W., 2007. Serial in vivo MR tracking of magnetically labeled neural spheres transplanted in chronic EAE mice. Magn. Reson. Med. 57, 164–171. Blakemore, W.F., Gilson, J.M., Crang, A.J., 2003. The presence of astrocytes in areas of demyelination influences remyelination following transplantation of oligodendrocyte progenitors. Exp. Neurol. 184, 955–963. Compston, A., Coles, A., 2002. Multiple sclerosis. Lancet 359, 1221–1231. Doetsch, F., Caille, I., Lim, D.A., Garcia-Verdugo, J.M., Alvarez-Buylla, A., 1999. Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell 97, 703–716.

25

Einstein, O., Ben-Menachem-Tzidon, O., Mizrachi-Kol, R., Reinhartz, E., Grigoriadis, N., Ben-Hur, T., 2006a. Survival of neural precursor cells in growth factor-poor environment: implications for transplantation in chronic disease. Glia 53, 449–455. Einstein, O., Grigoriadis, N., Mizrachi-Kol, R., Reinhartz, E., Polyzoidou, E., Lavon, I., Milonas, I., Karussis, D., Abramsky, O., Ben-Hur, T., 2006b. Transplanted neural precursor cells reduce brain inflammation to attenuate chronic experimental autoimmune encephalomyelitis. Exp. Neurol. 198, 275–284. Einstein, O., Karussis, D., Grigoriadis, N., Mizrachi-Kol, R., Reinhartz, E., Abramsky, O., Ben-Hur, T., 2003. Intraventricular transplantation of neural precursor cell spheres attenuates acute experimental allergic encephalomyelitis. Mol. Cell. Neurosci. 24, 1074–1082. Franklin, R.J., Gilson, J.M., Blakemore, W.F., 1997. Local recruitment of remyelinating cells in the repair of demyelination in the central nervous system. J. Neurosci. Res. 50, 337–344. Galli, R., Gritti, A., Bonfanti, L., Vescovi, A.L., 2003. Neural stem cells: an overview. Circ. Res. 92, 598–608. Herrera, D.G., Garcia-Verdugo, J.M., Alvarez-Buylla, A., 1999. Adult-derived neural precursors transplanted into multiple regions in the adult brain. Ann. Neurol. 46, 867–877. Issazadeh, S., Navikas, V., Schaub, M., Sayegh, M., Khoury, S., 1998. Kinetics of expression of costimulatory molecules and their ligands in murine relapsing experimental autoimmune encephalomyelitis in vivo. J. Immunol. 161, 1104–1112. Jin, K., Minami, M., Lan, J.Q., Mao, X.O., Batteur, S., Simon, R.P., Greenberg, D.A., 2001. Neurogenesis in dentate subgranular zone and rostral subventricular zone after focal cerebral ischemia in the rat. Proc. Natl. Acad. Sci. U. S. A. 98, 4710–4715. Lalive, P.H., Paglinawan, R., Biollaz, G., Kappos, E.A., Leone, D.P., Malipiero, U., Relvas, J.B., Moransard, M., Suter, T., Fontana, A., 2005. TGF-beta-treated microglia induce oligodendrocyte precursor cell chemotaxis through the HGF-c-Met pathway. Eur. J. Immunol. 35, 727–737. Lassmann, H., Bruck, W., Lucchinetti, C., Rodriguez, M., 1997. Remyelination in multiple sclerosis. Mult. Scler. 3, 133–136. Martino, G., Pluchino, S., 2006. The therapeutic potential of neural stem cells. Nat. Rev. Neurosci. 7, 395–406. Martino, G., Pluchino, S., 2007. Neural stem cells: guardians of the brain. Nat. Cell Biol. 9, 1031–1034. Menn, B., Garcia-Verdugo, J.M., Yaschine, C., Gonzalez-Perez, O., Rowitch, D., AlvarezBuylla, A., 2006. Origin of oligodendrocytes in the subventricular zone of the adult brain. J. Neurosci. 26, 7907–7918. Miller, M.D., Krangel, M.S., 1992. Biology and biochemistry of the chemokines: a family of chemotactic and inflammatory cytokines. Crit. Rev. Immunol. 12, 17–46. Miller, S.D., Shevach, E.M., 1998. Immunoregulation of experimental autoimmune encephalomyelitis: editorial overview. Res. Immunol. 149, 753–759. Molina-Holgado, E., Vela, J.M., Arevalo-Martin, A., Guaza, C., 2001. LPS/IFN-gamma cytotoxicity in oligodendroglial cells: role of nitric oxide and protection by the antiinflammatory cytokine IL-10. Eur. J. NeuroSci. 13, 493–502. Nait-Oumesmar, B., Decker, L., Lachapelle, F., Avellana-Adalid, V., Bachelin, C., Van Evercooren, A.B., 1999. Progenitor cells of the adult mouse subventricular zone proliferate, migrate and differentiate into oligodendrocytes after demyelination. Eur. J. NeuroSci. 11, 4357–4366. Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T., Nishimune, Y., 1997. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett. 407, 313–319. Patrikios, P., Stadelmann, C., Kutzelnigg, A., Rauschka, H., Schmidbauer, M., Laursen, H., Sorensen, P.S., Bruck, W., Lucchinetti, C., Lassmann, H., 2006. Remyelination is extensive in a subset of multiple sclerosis patients. Brain 129, 3165–3172. Paxinos, G., Franklin, K.B.J., 2001. The Mouse Brain in Stereotaxic Coordinates. Academic Press. Picard-Riera, N., Decker, L., Delarasse, C., Goude, K., Nait-Oumesmar, B., Liblau, R., PhamDinh, D., Evercooren, A.B., 2002. Experimental autoimmune encephalomyelitis mobilizes neural progenitors from the subventricular zone to undergo oligodendrogenesis in adult mice. Proc. Natl. Acad. Sci. U. S. A. 99, 13211–13216. Pluchino, S., Martino, G., 2007. The therapeutic plasticity of neural stem/precursor cells in multiple sclerosis. J. Neurol. Sci. Pluchino, S., Muzio, L., Imitola, J., Deleidi, M., Alfaro-Cervello, C., Salani, G., Porcheri, C., Brambilla, E., Cavasinni, F., Bergamaschi, A., Garcia-Verdugo, J.M., Comi, G., Khoury, S.J., Martino, G., 2008. Persistent inflammation alters the function of the endogenous brain stem cell compartment. Brain 131, 2564–2578. Pluchino, S., Quattrini, A., Brambilla, E., Gritti, A., Salani, G., Dina, G., Galli, R., Del Carro, U., Amadio, S., Bergami, A., Furlan, R., Comi, G., Vescovi, A.L., Martino, G., 2003. Injection of adult neurospheres induces recovery in a chronic model of multiple sclerosis. Nature 422, 688–694. Rubio, F.J., Bueno, C., Villa, A., Navarro, B., Martinez-Serrano, A., 2000. Genetically perpetuated human neural stem cells engraft and differentiate into the adult mammalian brain. Mol. Cell. Neurosci. 16, 1–13. Shields, S.A., Gilson, J.M., Blakemore, W.F., Franklin, R.J., 1999. Remyelination occurs as extensively but more slowly in old rats compared to young rats following gliotoxininduced CNS demyelination. Glia 28, 77–83. Sobel, R.A., Mitchell, M.E., 1989. Fibronectin in multiple sclerosis lesions. Am. J. Pathol. 135, 161–168. Spassky, N., de Castro, F., Le Bras, B., Heydon, K., Queraud-LeSaux, F., Bloch-Gallego, E., Chedotal, A., Zalc, B., Thomas, J.L., 2002. Directional guidance of oligodendroglial migration by class 3 semaphorins and netrin-1. J. Neurosci. 22, 5992–6004. Steinman, L., 2001a. Blockade of gamma interferon might be beneficial in MS. Mult Scler. 7, 275–276. Steinman, L., 2001b. Multiple sclerosis: a two-stage disease. Nat. Immunol. 2, 762–764. Steinman, L., Conlon, P., 2001. Antigen specific immunotherapy of multiple sclerosis. J. Clin. Immunol. 21, 93–98.

26

A. Giannakopoulou et al. / Experimental Neurology 230 (2011) 16–26

Steinman, L., Martin, R., Bernard, C., Conlon, P., Oksenberg, J.R., 2002. Multiple sclerosis: deeper understanding of its pathogenesis reveals new targets for therapy. Annu. Rev. Neurosci. 25, 491–505. Tourbah, A., Linnington, C., Bachelin, C., Avellana-Adalid, V., Wekerle, H., Baron-Van Evercooren, A., 1997. Inflammation promotes survival and migration of the CG4 oligodendrocyte progenitors transplanted in the spinal cord of both inflammatory and demyelinated EAE rats. J. Neurosci. Res. 50, 853–861. Vitry, S., Avellana-Adalid, V., Lachapelle, F., Evercooren, A.B., 2001. Migration and multipotentiality of PSA-NCAM+ neural precursors transplanted in the developing brain. Mol. Cell. Neurosci. 17, 983–1000.

Washington, R., Burton, J., Todd III, R.F., Newman, W., Dragovic, L., Dore-Duffy, P., 1994. Expression of immunologically relevant endothelial cell activation antigens on isolated central nervous system microvessels from patients with multiple sclerosis. Ann. Neurol. 35, 89–97. Williams, A., Piaton, G., Lubetzki, C., 2007. Astrocytes—friends or foes in multiple sclerosis? Glia 55, 1300–1312. Woodruff, R.H., Franklin, R.J., 1999. Demyelination and remyelination of the caudal cerebellar peduncle of adult rats following stereotaxic injections of lysolecithin, ethidium bromide, and complement/anti-galactocerebroside: a comparative study. Glia 25, 216–228.