www.elsevier.com/locate/ymcne Mol. Cell. Neurosci. 32 (2006) 187 – 198

Glial conversion of SVZ-derived committed neuronal precursors after ectopic grafting into the adult brain Ralph Seidenfaden,a,* Ange´lique Desoeuvre,a Andreas Bosio,b Isabelle Virard,a and Harold Cremer a a b

Institut de Biologie du De´veloppement de Marseille, CNRS, Universite´ de la Me´diterane´e, Campus de Luminy-case 907, 13288 Marseille cedex 9, France Miltenyi Biotec, Stoeckheimer Weg 1, 50829 Cologne, Germany

Received 5 January 2006; revised 27 March 2006; accepted 6 April 2006

In the adult mouse forebrain, large numbers of neuronal precursors, destined to become GABA- and dopamine-producing interneurons of the olfactory bulb (OB), are generated in the subventricular zone (SVZ). Although this neurogenic system represents a potential reservoir of stem and progenitor cells for brain repair approaches, information about the survival and differentiation of SVZ-derived cells in ectopic brain regions is still fragmentary. We show here that ectopic grafting of SVZ tissue gave rise to two morphologically distinguishable cell types displaying oligodendrocytic or astrocytic characteristics. Since SVZ tissue contains neuronal and glial progenitors, we used magnetic cell sorting to deplete A2B5+ glial progenitors from the dissociated SVZ and to positively select cells that express PSA-NCAM. This procedure allowed the purification of neuronal precursors expressing TUJ1, DCX and GAD65/67. Transplantation of these cells led again to the generation of the same two glial cell types, showing that committed interneuron precursors undergo glial differentiation outside their normal environment. D 2006 Elsevier Inc. All rights reserved.

Introduction In the olfactory system of the postnatal and adult rodent forebrain, large numbers of new neuronal precursors are generated in the subventricular zone (SVZ), from where they migrate via the rostral migratory stream (RMS) to the olfactory bulb (OB). After radial migration to the periphery of the OB, these precursors finally become GABAergic and dopaminergic interneurons in the granular and glomerular layers of the OB (Luskin, 1993; Kosaka et al., 1995; Alvarez-Buylla and Garcia-Verdugo, 2000). This ongoing neurogenesis attracted considerable attention over the last years.

* Corresponding author. Fax: +33 4 91 26 9726. E-mail address: [email protected] (R. Seidenfaden). Available online on ScienceDirect (www.sciencedirect.com). 1044-7431/$ - see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.mcn.2006.04.003

First, the demonstration that new neurons can be generated and integrated into the existing circuitry fundamentally changed our view of brain development and function. Second, the adult SVZRMS-OB system became an important model to study the molecular and cellular mechanisms that underlie neurogenesis in general. Third, the discovery of comparable neural stem cells and progenitors in humans raised hope for the use of this adult neurogenesis for cell-based brain repair (Liu and Martin, 2003; Bedard and Parent, 2004; Lie et al., 2004; Sanai et al., 2004). The rodent SVZ consists of at least 6 principal cell types (Doetsch et al., 1997): committed neuronal precursor cells migrating to the OB (type A cells), glial cells (type B2 cells, representing probably the stem cells of the system), astrocytes (type B1 cells) forming tunnel-like structures around the migrating neuronal precursors, rapidly dividing progenitors (type C cells), tanycytes (type D cells), ependymal cells (type E cells) lining the ventricle and 2% of so far unidentified cells. At least at the neonatal stage the SVZ also contains glial progenitor cells that migrate into white matter, cortex and striatum to generate astrocytes and oligodendrocytes (Marshall et al., 2003). Among these different cell types, committed neuronal precursor cells are the best characterized subpopulation, accounting for 33% of SVZ cells (Doetsch et al., 1997) and probably representing the largest reservoir of immature but clearly committed neuronal precursor in the adult brain (Goldman and Luskin, 1998; Alvarez-Buylla and Garcia-Verdugo, 2000). The neuronal identity of these cells is defined by ultrastructure and antigenic markers like expression of the polysialylated form of the neural cell adhesion molecule (PSA-NCAM), neuronal class III h-tubulin (TUJ1, Doetsch et al., 1997), doublecortin (DCX, Nacher et al., 2001), GABA (Bolteus and Bordey, 2004), ER81 and DLX (Stenman et al., 2003). Furthermore, homotypic grafting of SVZ tissue or retroviral infection experiments led exclusively to the generation of interneurons in the granular and periglomerular layers of the OB (Luskin, 1993; Doetsch and Alvarez-Buylla, 1996; Coskun et al., 2001; Hack et al., 2002; Hack et al., 2005; Kohwi et al., 2005).

188

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

A variety of factors that regulate the proliferation and migration of neuronal SVZ precursors have been described over the past decade (Hagg, 2005). Thus, neuronal precursors from the SVZ allow studying the behavior of a well-defined committed neuronal cell population that is present in the adult brain. A particular characteristic of these precursors is a clear binary terminal differentiation behavior, either dopaminergic or GABAergic. This type of precursors might be suitable for cell therapy approaches requiring the reconstitution of GABAergic or dopaminergic transmission like Huntington’s or Parkinson’s Diseases (Cao et al., 2002; Lindvall et al., 2004). Despite the obvious potential of these cells for therapeutic applications, information about their survival and differentiation in heterotypic positions is still fragmentary and in part contradictory. It was described that early postnatal mouse SVZ tissue grafted into the adult striatum showed little or no migration (Lois and Alvarez-Buylla, 1994) and predominantly astrocytic differentiation (Herrera et al., 1999), while a study in rats using early postnatal SVZ cells suggested migration away from the injection site and the differentiation into a phenotype morphologically resembling OB interneurons (Zigova et al., 1998). In all these studies, a heterogeneous mixture of SVZ cells containing all of the above-described cell populations was transplanted. However, recent work demonstrated the presence of additional, non-neuronal types of precursors in postnatal SVZ tissue, for example an astrocyte precursor population (Staugaitis et al., 2001) and new multipotent type C-like cells (Aguirre et al., 2004). Although it appears that gliogenesis from the SVZ does not persists in the normal adult brain (Marshall et al., 2003),

oligodendrocyte recruitment from the SVZ was described after lysolecithin-induced demyelination of the corpus callosum (NaitOumesmar et al., 1999), suggesting that glial progenitors cells persist in the adult structure. In the above-mentioned grafting studies, the donor cells were discriminated either by using a neuron-specific enolase-lacZ reporter (Lois and Alvarez-Buylla, 1994; Herrera et al., 1999) or by prelabeling with BrdU (Zigova et al., 1998). These approaches limited the possibilities to determine the phenotype of the surviving cells by their morphology and immunocytofluorescence. In this work, we analyzed the differentiation and survival of ectopically grafted SVZ tissue and cells by using actin-EGFP transgenic donor mice (Hadjantonakis et al., 1998), which allows a high-resolution morphologic analysis with multicolor-immunofluorescence. Using the phenotype-independent EGFP-label, we tested a broad panel of differentiation markers to characterize the fate of surviving cells and describe two morphologically distinguishable cell types showing specific glial properties. By grafting purified subpopulations from dissociated SVZ using magnetic activated cell sorting (MACS), we identified the surviving cells as derivatives of committed neuronal precursors. These data reveal the gliogenic potential of SVZ interneuron precursors after grafting outside their normal neurogenic environment in the OB. Thus, the use of these cells to restore neuronal functions requires manipulating their differentiation prior to or during the grafting process, they might be useful in CNS disorders linked to astrocyte cell death (Broe et al., 2004; Dienel and Hertz, 2005) and their application in remyelination attempts is validated (Keirstead et al., 1999; Vitry et al., 2001; Cayre et al., 2006).

Fig. 1. Homotypic transplantation of SVZ-derived EGFP+ cells into the adult brain results in migration to the olfactory bulb and differentiation into interneurons. (a) Low magnification EGFP fluorescence of an adult forebrain 4 weeks after transplantation. EGFP-fluorescence allows the visualization of the host tissue and the graft-derived EGFP+ cells. EGFP+ SVZ-tissue has been grafted into the wild-type subventricular zone underlying the corpus callosum (CC). EGFP+ cells migrate from the injection site (IS) via the rostral migratory stream (RMS) into the olfactory bulb (OB). In the OB, the cells differentiate into interneurons of the granular cell layer (GrL) and the glomerular layer (GL), scale bar: 300 Am, the white square in the GL represents the position of the high magnifications shown in panels b – e. (b) EGFP fluorescence of graft-derived periglomerular cells; (c) NeuN fluorescence showing glomerular neurons; (d) Hoechst fluorescence; (e) merged EGFP/NeuN/Hoechst fluorescence showing neuronal differentiation of graft-derived cells in the GL, scale bar: 10 Am.

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

Results Survival and differentiation of transplanted SVZ tissue Grafting of SVZ tissue from actin-EGFP transgenic donor mice allowed visualization of EGFP+ cells in the wild-type host environment (Figs. 1 and 2). After homotypic transplantation, graft-derived cells underwent migration and differentiation like the endogenous precursor cells (Fig. 1, Lois and Alvarez-Buylla, 1994). During these processes, EGFP expression in the graftderived cells remained unchanged and after 4 weeks neuronal differentiation of grafted cells is indicated by expression of NeuN (Fig. 1). Thus, SVZ cells transplanted into homotypic positions generated neurons in the OB. Due to its importance as a target structure for cell therapy, the adult striatum was chosen as ectopic host environment. Transplanted SVZ cells were clearly identifiable in the host striatum by their EGFP-fluorescence 3 weeks (Fig. 2a) and 6 months after transplantation (Fig. 2b). A large portion of grafted SVZ tissue underwent cell death during the first 3 weeks after grafting (Figs. 2c and e; white arrowheads indicate strongly

189

fluorescent cellular debris; hatched arrowheads point to morphologically intact cells, n = 16 animals). However, surviving cells were still evident after 6 months (Figs. 2d and f, n = 3 animals for each donor tissue). The comparison of neonatal (P5) and adult (P75)-derived SVZ donor tissue showed the same dynamics of initial cell death and long-term survival (Figs. 2c – f). The host tissue at the graft site presented normal gross microanatomy already 3 weeks after transplantation (Fig. 2g) and grafted cells were individually integrated in the fiber tracts (striosomes) or the striatal matrix (Fig. 2h), not showing an evident preference for one of the compartments. Furthermore, the spatial extension of EGFP+ cells was restricted and no long distance migration occurred in either case. To compare the survival of SVZ cells from neonatal versus adult donor tissue, we determined the density of EGFP+ cells in the center of the graft site 3 weeks after transplantation. Quantitative analysis of confocal image stacks from neonatal tissue revealed 275 (T115 SD) cells per frame and animal, while P75 tissue resulted in 39 (T11 SD) cells per frame. Accordingly, using comparable amounts of freshly dissected SVZ tissue from neonatal or adult SVZ, after dissociation of the tissue but before

Fig. 2. Long-term survival, tissue integration and morphological phenotype of SVZ-derived EGFP+ cells after transplantation into the striatum. Normarski contrast and EGFP fluorescence (green) showing the site and extent of grafted tissue 3 weeks (a) or 6 months (b) after transplantation. CT: cerebral cortex, CC: corpus callosum, ST: striatum, LV: lateral ventricle, scale bar: 200 Am. Postnatal day 5 (P5: c, d) and postnatal day 75 (P75: e, f) SVZ tissue 3 weeks (c, e) or 6 months (d, f) after grafting. (c – f) Scale bar: 10 Am, z-stacks: 5.4 Am, EGFP (green), Hoechst 33258 (blue), arrowheads point towards intact cell (black) or cellular debris (white). (g – h) Integration of grafted cells in the matrix and striosome compartments of the striatum 3 weeks after transplantation: Normarski contrast (g); EGFP/Hoechst fluorescence (h), scale bar: 200 Am, z-stack: 59 Am. (i) High magnification from panel h showing the two distinct morphologies of the surviving cells: type 1 and 2, scale bar: 5 Am, z-stack: 15 Am.

190

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

transplantation, we found 8 times less cells for adult tissue (P5: 4  105 T 6  104 SD cells versus P75: 5  104 T 5  103 cells). Thus, it is likely that the lower number of cells that we found after transplantation of adult SVZ compared to neonatal corresponds to the lower number of initially injected cells per tissue volume. Morphological analysis revealed two clearly distinguishable cell types. First, a bipolar cell type with an elongated cell soma carrying large diameter processes (Fig. 2i, cell type 1). Second, an irregular spherical cell type with highly arborized low diameter processes (Fig. 2i, cell type 2). Both cell types could be identified at 3 weeks as well as at 6 months (Fig. 2f) postgrafting, regardless if neonatal of adult donor tissue was used. An unequivocal criterion to distinguish the two cell types was the mean process thickness at the soma border: 4.4 Am for cell type 1 (mean process length 126 Am T 28 Standard Deviation, SD) and 0.4 Am for cell type 2 (the mean process length was not measurable due to morphological complexity). The morphologies of the two cell types were clearly different from the phenotypes observed after homotypic grafting of SVZ cells, where cells display the typical morphology of either migrating neuroblasts (leading process 24 Am T 7 SD, trailing process 5 Am T 2 SD), differentiating neurons or mature interneurons (see Lois and Alvarez-Buylla, 1994; Hack et al., 2002; Hack et al., 2005). Each cell type represented about 50% of the surviving cells, independent of the postnatal stage of the donor mice (Fig. 3a, n = 3 animals for each donor tissue). To determine if the two cellular phenotypes occurred specifically in response to the striatum (ST) as a host environment, we also grafted SVZ tissue into the cerebral motor cortex (CT) and the lateral posterior thalamic nucleus (TH). Again, the two cell types shown in Fig. 2i in the ST could be morphologically distinguished in CT and TH (data not shown) and quantification revealed that each cell type again represented about 50% of the surviving cells in the different host structures (Fig. 3b). SVZ-derived cells in ectopic positions display glial properties In the next step, the properties of the two cell types were analyzed by immunocytofluorescence (Fig. 4) and the use of a broad panel of differentiation markers. This analysis revealed that

both cell types were characterized by expression of mainly glial markers (for specificity of the markers, see Satoh and Kim, 1995; Doetsch et al., 1997; Doetsch et al., 1999; Farrer and Quarles, 1999; Zhang, 2001; Hachem et al., 2005), while mitotic and early neural markers were not detectable in graft-derived cells (Table 1). Both cell types expressed S100h (Figs. 4i – l), a protein expressed in the entire glial lineage. Cell type 1 was also positive for GFAP (Figs. 4a – d) and Vimentin (Figs. 4q – t), indicative of the astrocyte lineage, while cell type 2 expressed Olig2 (Figs. 4e – h) and CNPase (Figs. 4m – p), markers of the oligodendrocyte lineage. Combination of the morphological properties with the immuncytocharacterization (Table 1) assigned cell type 2 to a non-mitotic, non-migratory oligodendrocyte phenotype. In contrast, cell type 1 displayed properties of non-mitotic mature astrocytes. This astrocytic phenotype is also underlined by the observation of cellular processes contacting blood vessels with an astrocyte-like endfoot (data not shown, see Simard et al., 2003). Surprisingly, cell type 1 also expressed glutamate decarboxylase (GAD65/67, Figs. 4u and w) and its product GABA (Table 1). In the SVZ, GABA represents a marker of neuronal differentiation expressed by the migrating neuronal precursor cells (Bolteus and Bordey, 2004) and the majority of terminal differentiated interneurons in the OB (Kosaka et al., 1995). However, GABA expression was also described in astrocytes (Lin et al., 1993; Ochi et al., 1993; Jow et al., 2004) and therefore could either be attributed to glial properties or represent a rudiment of the originally neuronal commitment of the grafted cells. Other cellular phenotypes occurred very rarely. That is, out of 378 analyzed immunoreactive cells, one expressed NeuN and one NG2 (data not shown). Purification and transplantation of committed neuronal precursor cells As the SVZ contains different neuronal and glial progenitor populations, we aimed at determining the origin of the generated two cell types. Therefore, we characterized the different cell populations and established a protocol to isolate the committed neuronal precursor subpopulation by using cell surface antigens in combination with MACS. These experiments were performed with

Fig. 3. Grafted neonatal or adult SVZ tissue shows the same morphological differentiation into distinct cell types 1 and 2, independent of the ectopic target structure in the host brain. Percentage of the two morphologically distinguishable cell types shown in Fig. 2i (a) after grafting of dissociated SVZ cells from neonatal (SVZ-P5) or adult (SVZ-P75) animals into the striatum and (b) after grafting into 3 different ectopic positions: striatum (ST), cerebral motor cortex (CT) or lateral posterior thalamic nucleus (TH, means T SD).

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

191

Fig. 4. Characterization of SVZ cells grafted into the striatum by immunocytofluorescence and neural differentiation markers 3 weeks after transplantation. Differential expression of astrocytic markers in cell type 1 and oligodendrocytic markers in cell type 2. Grafted EGFP+ cells (green) are morphologically determined as cell types (1) and (2) by their process diameters. The blue fluorescence corresponds to a nuclear counterstain with Hoechst 33258. (A) EGFPfluorescence (-fl) and double immunostaining against GFAP (red, upper row) and Olig2 (red lower row) show mutually exclusive expression of these markers: GFAP is expressed in cell type 1 (with thick processes, upper row) and the oligodendroglial marker Olig2 in the nucleus of cell type 2 (lower row). (c, d, f – h) High magnifications from panels a and e. (a) Hoechst-, GFAP-, EGFP-fl. (b) Normarski contrast and EGFP-fl showing a cell type 1 and 2 lining a striosome fiber tract. (c) EGFP- and GFAP-fl; (d) GFAP- and Hoechst-fl; (e) Hoechst-, Olig2-, EGFP-fl; (f) EGFP-fl; (g) EGFP- and Olig2-fl (h) Hoechst-, Olig2- fl. Scale bars: in panel a for panels a, b, e and in panel c for panels c – h: 10 Am, z-stack: 2 Am. (B – E) EGFP-fl and immunostainings (red) against S100h (B), CNPase (C), Vimentin (D), Glutamate decarboxylase 65/67 (E). (i) Cell type 1, high diameter processes and (j) high magnification from panel i. (k) Cell type 2 and (l) high magnification from panel k. (m) Cell type 2 and (n) high magnification from panel m. (o) Cell type 1 and (p) high magnification from panel p. (q) Cell type 1, high diameter processes and (r) high magnification from panel q. (s) Cell type 2 and high diameter process cell type 1. (t) High magnification from panel r. (u) Cell type 1 and 2. (v) High magnification cell type 2 from panel u and (w) high magnification of cell type 1 from panel u. Scale bars (i – w): 10 Am, z-stack: 2 Am.

192

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

Table 1 Morphological and immunocytochemical phenotype of SVZ cells 3 weeks to 6 months after grafting Cell

Type 1

Type 2

Morphology

54% elongated bipolar few bifurcations 4.4 – 0.3 Am

46% spherical irregular arborized

+ + +

+

Process diameter Immunofluorescence markers Early neural Glial

Neuronal

Ki-67, GD2, Nestin, PSA, CD24 S100h m/p GFAP Vimentin Olig-2 NG2 CNPase MBP TUJ1, Nf, MAP-2 Calretinin, Calb. NeuN Th ChAt Gad65/67, GABA

0.4 – 0.02 Am

+ +

+

m/p: mono/polyclonal; marker abbreviations, see Table 2.

neonatal SVZ, since this tissue produced the same cell phenotypes after grafting (Fig. 3a) but contained higher cell numbers when compared to P75 tissue (see above), thereby making it easier to obtain sufficient cells for the sorting procedure. Committed neuronal precursor in the SVZ has been shown to express neuronal class III h-tubulin (TUJ1) and PSA-NCAM (Doetsch et al., 1997). Accordingly, after 1 h in vitro, dissociated neonatal SVZ cells include a majority of TUJ1+ cells that also present the surface antigen PSA-NCAM (PSA, Figs. 5a and c, cell type i). Furthermore, the tissue contains oligodendrocyte progenitors expressing PSA and A2B5 (Ben-Hur et al., 1998; Farrer and Quarles, 1999, Figs. 5a and c, cell type ii) and glial or O2A oligodendrocyte progenitors that expressed only A2B5 (Figs. 5a – c, cell type iii, Dietrich et al., 2002). The identity of glial progenitors was further confirmed by expression of CD44 in A2B5+/PSA-NCAM cells (Fig. 5b, cell type iii, Bouvier-Labit et al., 2002; Liu et al., 2004). Based on this information, we devised a two-step protocol of MACS. In a first step, the SVZ cell suspension was incubated with A2B5 specific and magnetic bead-linked antibodies to remove the A2B5+ cells. The A2B5 depleted cell suspension was subsequently incubated with PSA-specific and magnetic bead linked antibodies to isolate PSA-NCAM+ cells. This protocol allowed us to isolate 2 subpopulations: the A2B5+ fraction (corresponding to enriched glial progenitors) and the A2B5 / PSA-NCAM+ fraction (corresponding to committed neuronal precursors). Using this immunopurification by MACS from 6  106 dissociated cells, an average of 1.1  106 A2B5+ cells and 1.4  106 A2B5 /PSA-NCAM+ cells were recovered. In vitro characterization of unsorted SVZ suspensions by immunocytofluorescence showed that 62% (T4.2 SD) of the cells were TUJ1+, and thus represented neuronal precursors (Figs. 6a and b). Of the remaining cells, 36% expressed A2B5 (Fig. 6a), a subpopulation was TUJ1+/A2B5+ (2% T 1.6), and 3% T 1.2 of the cells were

TUJ1 /PSA-NCAM /A2B5 (data not shown). In the A2B5+ selected fraction, a relatively high number of TUJ1+ cells were observed (39% T 3.7; Figs. 6a and b). Lastly, the positive selection for PSA-NCAM from the A2B5-depleted flow through resulted in a 98% (T1.6) pure population of TUJ1+ neuronal precursor cells, nearly all of them (99% T 1) co-expressed PSA-NCAM (Figs. 6a and b). The neuronal identity of this SVZ fraction was further confirmed by expression of Gad65/67 in 93% (T2.7) and DCX in 96% (T0.3) of the cells (Fig. 6a). Thus, the antigenic expression profile of this SVZ cell type in vitro indicated that they did not correspond to multipotent precursors (type C cells), but represent neuronal precursors corresponding to type A cells (Doetsch et al., 1997; Aguirre et al., 2004). We then compared the differentiation and survival of the unsorted, A2B5+ or A2B5 /PSA-NCAM+ SVZ-fractions after transplantation into the striatum. For each condition, 5  104 cells were injected into the host striatum. Three weeks after grafting, all three suspensions exclusively led to the generation of the two described glial cell types above (Fig. 6c). Moreover, the relative percentage of the two cell types was similar in control and sorted cells (Fig. 5d). Interestingly, the amount of these surviving glial cells was negatively correlated with the amount of injected A2B5+ cells (compare Figs. 6e and f), while there was a striking positive correlation with the number of injected TUJ1+ cells (compare Figs. 6b and f). This strongly indicates that the surviving cells displaying glial properties originate from the committed neuronal precursors.

Discussion The present study investigates the survival and differentiation of neonatal and adult SVZ tissue, after grafting into ectopic positions of the adult brain. Our study leads to three major conclusions. First, SVZ-derived cells are able to survive in ectopic brain regions for at least 6 months. Second, the SVZ-cells differentiate into two cell types displaying either oligodendrocytic or astrocytic characteristics. Third, these surviving glial cells originate from committed interneuron precursors. These results have a general significance for cell therapeutic approaches in the adult brain. First clinical studies indicated that transplantation of fetal neurons can abrogate the clinical manifestations of neurodegeneration in Huntington’s or Parkinson’s Diseases. Since there are a number of problems with the use of human fetal tissue in clinical trials, alternative sources of transplantable cells are required. Important candidates are in vitro differentiated embryonic stem cells and neural stem cells that have been expanded as neurospheres. Both approaches include the risks of inducing neoplasms through the grafting of immature cells. Furthermore, in general, impure cell populations with low numbers of the desired neuronal phenotype are used for transplantation (reviewed in Bjorklund and Lindvall, 2000; Cao et al., 2002; Lindvall et al., 2004). In principle, SVZ precursors represent an alternative source of transplantable cells both present in the adult and committed to neuronal differentiation. Although the comparability of human and mouse olfactory neurogenesis is still a matter of debate, there is in vitro (Kirschenbaum et al., 1994) and in vivo evidence for the presence of interneuron precursors in adult humans (Liu and Martin, 2003; Bedard and Parent, 2004; Sanai et al., 2004; Lie et al., 2004). In addition, it appears likely that precursors, like the

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

193

Fig. 5. Characterization of glial progenitor cells in dissociated cultures from the neonatal SVZ after 1 h in vitro. (a) Triple-immunofluorescence for A2B5, PSANCAM (PSA) and TUJ1 reveals three different cell types (i – iii) that can be classified as different progenitors as indicated in panel c. (i) Committed neuronal precursors, (ii) migrating oligodendrocyte preprogenitors and (iii) glial progenitors that can give rise to astrocytes or oligodendrocytes. (b) Staining for A2B5, PSA-NCAM and CD44 confirms the glial identity of A2B5+/PSA-NCAM cells. Scale bar: 10 Am.

ones present in mouse SVZ, represent differentiation intermediates in all neurogenetic events. In case such cells can be produced and purified in sufficient quantities in vitro, they might represent an interesting candidate for transplantation-based therapy in the nervous system. Only two previous studies used grafting approaches to investigate the differentiation of SVZ-derived tissue in ectopic brain positions. Herrera et al. (1999) used electron microscopy in association with grafts derived from mice carrying a neuronspecific enolase (NSE) promoter to express LacZ as a transgene. Their finding of predominantly astrocyte-like cells showing coexpression of NSE-LacZ, rather a neuronal marker, is in agreement with our observation that the astrocytic subpopulation co-expressed the neuronal marker GABA. Furthermore, in the absence of a pancellular marker like actin-EGFP, it is well possible that the immature oligodendrocyte-like type 2 cells that we describe here were not detected. Indeed our analyses showed that, unlike in astrocyte-like type 1 cells, all analyzed neuronal markers are absent from cell type 2. Thus, it appears probable that the NSE promotor was not active in type 2 cells, making their identification as graftderived cells impossible. A further discrepancy concerns the existence of cells with characteristics of chain migrating neuro-

blasts in the surviving graft, which have been described in the study by Herrera et al. (1999). In our experiments, neither the use of antigenic markers like PSA-NCAM or TUJ1, nor close morphological observation revealed the existence of type A cells, even after a short survival time of 3 weeks. A second publication by Zigova et al. (1998) also investigated the consequences of SVZ-tissue grafting. This work showed more striking contradictions with the outcome of our study. The authors describe the predominant generation and long-term survival of cells with characteristics of OB interneurons, a phenotype never observed in our experimental setups. However, several experimental differences might account for these differences. First, the authors performed 6-OHDA lesion of the dopaminergic neurons in the substantia nigra before the grafting of postnatal SVZ tissue into the striatum. Although Herrera et al. (1999) showed that the behavior of grafted cells was largely independent of precedent lesions in the target structure, this might explain the acquisition of a neuronal phenotype. Along this line, it was shown that precedent lesions can improve the survival of transplanted fetal striatum cells (Mundt-Petersen et al., 2000). Nevertheless, the inconsistencies might also be due to differences in the identification and characterization of grafted cells. While our study relies on

194

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

Fig. 6. Purification of SVZ subpopulations by magnetic-activated cell sorting (MACS) using the cell surface antigens A2B5 and PSA-NCAM (PSA). (a) In vitro analysis of dissociated cells from the neonatal SVZ after MACS. Immunofluorescence for TUJ1, GAD65/67 (GAD) or DCX (neuronal precursors, green) and A2B5 (glial progenitors, red) in the different fractions; control: unsorted SVZ cells, A2B5+: SVZ cells positively selected for A2B5 expression, A2B5 /PSA+: SVZ cells depleted of A2B5+ cells and positively selected for PSA-NCAM+ cells. Scale bar: 50 Am. (b) Quantification of neuronal precursors identified by TUJ1 expression in control (unsorted) and the different MACS-purified cell fractions A2B5+ and A2B5 /PSA+ after 1 h in culture, means T standard error of the mean (SEM). (c) Cell morphologies of A2B5 /PSA+-purified neuronal precursor cells from the SVZ 3 weeks after transplantation into the striatum reveal the same distinct cells types (type 1 and 2) as transplantation of the whole tissue (Fig. 2h and i). Scale bar: 10 Am, z-stacks: 3.6 Am. (d) Percentage of the two morphologically distinguishable cell types shown in Fig. 2i after grafting of the different MACS-purified cell fractions (unsorted, A2B5+ and A2B5 /PSA+, means T SEM). (e) Quantification of A2B5 expressing cells in control (unsorted) and the different MACS-purified cell fractions (A2B5+ and A2B5 /PSA+, means T SEM). (f) Number of surviving cells 3 weeks after transplantation of 50,000 EGFP+ control (un-sorted) or MACSpurified cells (means T SEM).

genetically labeled graft tissue in combination with multiple immunological labelings and confocal microscopy, Zigova et al. (1998) used mainly morphological criteria to identify long-term surviving cells. But since small neurons such as OB interneurons look similar in size, shape, and tinctorial properties to glia (Rakic, 2002), a doubtless distinction by cresyl violet staining and morphology is a difficult task. Both studies discussed above used whole SVZ tissue as grafting material. However, it was shown that, both in the neonatal and in the adult animal, the SVZ is a heterogeneous tissue containing a variety of differentiated as well as stem and progenitor cell types (Doetsch et al., 1997; Marshall et al., 2003). Many progenitor cell types have been characterized by differential expression of cell surface markers. The large population of committed neuronal precursors expresses the cell surface marker PSA-NCAM (Doetsch and Alvarez-Buylla, 1996). Oligodendrocyte preprogenitors are

also positive for PSA-NCAM (Ben-Hur et al., 1998) but co-express A2B5 (Farrer and Quarles, 1999). Glial progenitors express A2B5 (Dietrich et al., 2002) and CD44 (Bouvier-Labit et al., 2002; Liu et al., 2004), but are negative for PSA-NCAM (Mayer-Proschel et al., 1997). All the different types of mature astrocytes, oligodendrocytes, and ependymocytes lack PSA-NCAM expression (Doetsch et al., 1997). Based on these cell surface markers, we devised a MACS-based method that allows the purification of the committed neuronal precursor subpopulation from the heterogeneous SVZ. We depleted A2B5+ cells and positively selected PSA-NCAM+ cells from the remaining cell population. The A2B5 /PSANCAM+ population was characterized by homogenous expression of an antigen pattern demonstrating neuronal commitment (PSANCAM+/TUJ1+/DCX+/GAD65/67+), typical of adult neuroblasts in vivo (Doetsch et al., 1997; Nacher et al., 2001; Bolteus and Bordey, 2004). Surprisingly, grafting of this population did neither

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

lead to higher cell death nor to the survival of cells with neuronal phenotypes, but again gave rise to the two cell types with astrocytic or oligodendrocytic properties. The general survival rate of transplanted cells was low (2.5 – 4.3% of the injected cells) and it appears possible that the surviving cells derived from a specific subpopulation of the A2B5 /PSA-NCAM+ cells that are solely able to survive. In case of the existence of such a subpopulation, they are included in the PSA-NCAM+/TUJ1+/DCX+/GAD65/67+ cells, the SVZ fraction which correlated with the number of surviving cells. This raises further the question if such a population is homogenous or if it includes two different precursors, each one responsible for the generation of one of the two cell types. The existence of separate precursor populations would be a prerequisite to specifically isolate and use cell type 2 (displaying oligodendrocyte characteristics) for remyelination attempts (Keirstead et al., 1999; Vitry et al., 2001, Cayre et al., 2006) or cell type 1 (showing astrocytic properties) in CNS disorders due to loss of astrocytes (Broe et al., 2004; Dienel and Hertz, 2005). Adult neurogenesis is considered to be possible within a particular niche that allows neuronal differentiation from multipotent progenitors. Inhibition of gliogenic BMP signaling by noggin has been implicated in the generation of this niche (Alvarez-Buylla and Lim, 2004; Lie et al., 2004; Ma et al., 2005). The A2B5 /PSA-NCAM+ committed neuronal precursors have already received the instructive signals that determine their neuronal fate. Thus, after ectopic grafting into the striatum, these precursors receive glial inducing cues that are able to revert the neuronal commitment. This demonstrates the inhibition of neuronal differentiation in mature brain tissue outside the

195

neurogenic niches. A multitude of recent reports showed that under certain experimental conditions, compensatory neurogenesis and recruitment in response to neurodegeneration in non-neurogenic brain regions are possible (Fallon et al., 2000; Arvidsson et al., 2002, Lindvall et al., 2004; Goldman, 2005). This demonstrates that under specific conditions progenitor cells are able overcome the glia instructing signals of the mature brain. Discovery of such signals should allow the generation of tools that can be used for their inhibition. Such tools are crucially important for the use of SVZ precursors and also other stem and progenitor cells in therapeutic approaches for neuronal replacement. Experimental methods Mice Six to ten-week-old C57BL/6 mice were purchased from IffaCredo and used as host animals. Donor actin-EGFP transgenic mice (Hadjantonakis et al., 1998) on a C57BL/6 background were produced in our animal facilities. Animals were housed under standard conditions with ad libitum access to water and food on a normal 12 h light/dark cycle. Experiments were performed in accordance with the principles for laboratory animals published by the French Ethical Committee. Transplantation Postnatal day 5 (P5) mice were anesthetized by hypothermia and killed by rapid decapitation; P75 mice were killed by cervical dislocation. Brains were dissected out and the hindbrain was removed by cutting

Table 2 Primary antibodies used for immunohistofluorescence Antigen

Antiserum

Source

Dilution

A2B5 Calb. (Calbindin D28K) Calretinin CD24 CD44-FITC ChAt (choline acetyltransferase) CNPase (2V3V cyclic nucleotide 3Vphosphodiesterase) DCX (doublecortin) GD2 GAD65/67 (glutamate decarboxylase) GABA (gamma-aminobutyric acid) GFAP (glial fibrillary acidic protein)

Mouse IgM Mouse IgG1 Rabbit IgG Rat IgG Rat IgG Rabbit IgG Mouse IgG1 Guinea pig IgG Mouse Rabbit Rabbit IgG Mouse IgG1 Rabbit IgG Rabbit Mouse IgG1 Mouse IgG1 Mouse IgG1 Mouse IgG1 Mouse IgG1 Rabbit IgG Rabbit IgG Rabbit IgG Mouse IgG2a Mouse IgM Rabbit IgG Rabbit IgG Mouse IgG2a Rabbit IgG Mouse IgM

G. Rougon, Marseille, France Sigma, Saint Louis, MO Swant, Bellinzona, Switzerland G. Rougon, Marseille, France M. Kursar, Berlin, Germany Chemicon, Temecula, CA Sigma Chemicon Chemicon Sigma Sigma Sigma Sigma abcam, Cambridge, UK Neomarkers, Westinghouse, CA Chemicon Hybridoma Bank, Iowa, IA Chemicon Sigma Sigma Chemicon Dana-Farber Cancer Institute, Boston MA R. Gerardy-Schahn, Hannover, Germany G. Rougon, Marseille, France Dako Cytomation, Glostrup, DK Covance, Berkeley, CA Covance Chemicon Sigma

1:400 1:1000 1:800 1:250 1:200 1:200 1:100 1:500 1:500 1:400 1:200 1:1000 1:800 1 :400 1:200 1:200 1:400 1:200 1:200 1:400 1:200 1:20,000 1:400 1:1000 1:400 1:2000 1:500 1:400 1:200

Ki-67 MAP-2 MBP (Myelin Basic Protein) Nestin NeuN Neurofilament (Nf ) 68, 160, 200 kDa 200 kDa NG2 (chondroitinsulfate proteoglycan) Olig2 Polysialic acid (PSA) 735 MenB S100h TUJ1 (neuronal class III h-tubulin) Th (tyrosine hydroxylase) Vimentin

196

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

between the cerebral cortex and cerebellum. The forebrain was embedded in 4% agar – agar and cut into 400 Am coronal sections using a vibratome (Leica). The anterior part of the SVZ was dissected from the striatal wall in 20 mM HBSS medium (Gibco) using slices between the opening of the lateral ventricle and the beginning of the hippocampus. SVZ tissue was cut into explants 100 Am in diameter and stereotaxically grafted into the striatum, cortex or thalamus using a Kopf apparatus and a 2 Al Hamilton syringe at the following coordinates: striatum (1.5 mm anterior to bregma, 2.0 mm lateral and 3.0 mm deep), cerebral motor cortex (1.5 mm anterior to bregma, 2.0 mm lateral and 1.0 mm deep) and lateral posterior thalamic nucleus (1 mm posterior to bregma, 1.5 mm lateral and 2.0 mm deep). In the same way, we injected suspensions of SVZ-derived cells, either unsorted or purified by MACS (procedure: see below), containing 50,000 cells/Al. To assess cell viability and capability to generate neurons, in each series of grafts, some of the SVZ tissue or cell suspension was transplanted homotypically into the SVZ of the host animals (1 mm anterior to bregma, 1 mm lateral and 2.4 mm deep) as previously described (Lois and Alvarez-Buylla, 1994; Hack et al., 2002). Three weeks or 6 months after the grafting, the animals were perfused intracardially with a 3.8% paraformaldehyde solution in phosphate buffer (PB). Brains were dissected out, postfixed for 24 h and serial sagittal sections (35 Am) were cut with a vibratome. Grafted EGFP+ cells were detected with a laser scanning microscope and an argon laser at 488 nm (LSM 510 Meta, Zeiss) and analyzed using a software module allowing measurements of cell process length and diameter. To verify the localization of the graft, the host tissue was additionally visualized by Normarski differential interference contrast. For quantification of the total number of surviving grafted cells, the host tissue containing EGFP+ cells was imaged using a 20 magnification with a confocal z-stack depth corresponding to the section thickness. Hoechst/EGFP double-positive cell bodies were counted and morphological identification of different cell types was performed on 3 randomly selected z-stacks per animal. Immunohistofluorescence Fixed brains were cryoprotected and frozen in liquid nitrogen with 3 repeated cycles of thawing and freezing. Saggital sections (20 Am) were serially cut using a freezing sliding microtome (Leica). Immunofluores-

cence staining of free-floating sections was performed with primary antibodies as listed in Table 2. Sections were blocked for 1 h at room temperature (RT) in 3% BSA/PB and afterwards incubated for 48 h at 4-C in fresh blocking solution containing a mixture of two primary antibodies. The sections were then rinsed in 3% BSA/PB and incubated in blocking solution with species specific secondary antibodies for 2 h at RT. Subsequently, the sections were washed, transferred to slides and airdried. Fluorescence was detected using a laser scanning microscope (Zeiss) with a 405 nm laser diode for Hoechst and lasers for EGFP (Argon, 488 nm), Cy3 (Helium Neon, 543 nm) and Cy5 or Alexa 633 (Helium Neon, 633 nm). Dissociation and MACS purification of SVZ tissue SVZ tissue from 12 postnatal day 5 brains was minced and incubated with 0.5% trypsin (Gibco) at 37-C for 10 min, after 5 min 0.5 mg/ml DNAse I (Roche) and 12 mM MgCl2 were added (final volume: 2 ml). The suspension was gently triturated with fire polished pipettes with decreasing tip diameters and diluted in 8 ml DMEM containing 10% (v/v) FCS and 1% (v/v) penicillin/streptomycin (all from Gibco). The cells were centrifuged at 300g for 7 min and the pellet (about 6  106 cells) was resuspended in 200 Al HBSS-BSA (Miltenyi) containing 0.6 Al c-ganglioside-specific mouse IgM A2B5 ascites (ATCC CRL-1520, kindly provided by G. Rougon, Marseille) and incubated on a gently rocking rotator at 4-C for 30 min. Afterwards the cells were washed in 10 ml HBSS-BSA with subsequent centrifugation at 300g for 7 min. The same incubation and washing conditions were applied for all further antibodies. The pellet was resuspended in 20% (v/v) anti-mouse IgM MicroBeads (Miltenyi) in 200 Al and, after incubation and washing, resuspended in 1 ml HBSS-BSA. Cells were separated using an MS Column and a MiniMACSTM Separator (Miltenyi) according to the manufacturer’s instruction. The A2B5+ fraction retained in the column was eluted and kept on ice, while the flow-through was incubated with 5 Ag/ml PSA-specific mouse IgG2a 735 antibody (kindly provided by R. Gerardy-Schahn, Hannover). Using 20% (v/v) antimouse IgG MicroBeads in 100 Al HBSS-BSA as secondary antibody against 735, the marked cells were selected by retention in a column placed in a MiniMACSTM Separator. Based on the expression of respective marker molecules in vitro, 98% of this A2B5 /PSA-NCAM+ fraction corresponds to the committed neuronal precursor cells from the SVZ in vivo.

Table 3 Combinations of secondary antibodies for multiple immunofluorescence staining Combination of primary antibodies

Combination of antiserums

Source

Dilution

Mouse IgM Mouse IgG Rabbit IgG

Goat anti-mouse IgM Alexa 488 Goat anti-mouse IgG Fc Cy3 Goat anti-rabbit IgG Cy5

Molecular Probes/Invitrogen Cergy Pontoise, France Jackson ImmunoResearch Laboratories Cambridgeshire, UK

1:2000 1:1000 1:800

Mouse IgG Rabbit IgG

Donkey anti-mouse IgG Cy3 Goat anti-rabbit IgG Cy5

Jackson

1:400 1:800

Mouse IgG Rabbit IgG

Goat anti-mouse IgG Alexa 633 Donkey anti-rabbit Cy3

Molecular Probes Jackson

1:1000 1:400

Mouse IgM Mouse IgG

Goat anti-mouse IgM Alexa 633 Goat anti-mouse IgG Fc Cy3

Molecular Probes Jackson

1:800 1:1000

Mouse IgM Rabbit IgG

Goat anti-mouse IgM Cy3 Goat anti-rabbit IgG Cy5

Jackson

1:1000 1:800

Mouse IgG Rat IgG

Goat anti-mouse IgG Alexa 633 Goat anti-rat Alexa 546

Molecular Probes

1:1000 1:1000

Guinea pig IgG Mouse IgM

Donkey anti-guinea pig IgG Cy3 Goat anti-mouse IgM Alexa 488

Jackson Molecular Probes

1:400 1:2000

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198 In vitro characterization of SVZ progenitors Dissociated SVZ cells were cultured in DMEM: Ham’s F12 medium (3:1) containing 2 mM Glutamax, 1% (v/v) N2 supplement, 1% (v/v) B27, 1% penicillin/streptomycin (v/v, all from GIBCO).We seeded 50,000 cells per well on poly-l-lysine-coated (0.01%, Sigma) glass coverslips in 4-well plates in 500 Al culture medium. A2B5 ascitic fluid was added 1:400 to the living cells for 20 min. The cells were fixed after 1 h to avoid in vitro differentiation. After fixation in 3.8% PFA in PB for 20 min at RT, they were washed and 3% BSA/PB was added as blocking buffer before incubation for 30 min. Primary antibodies (Table 2) were diluted in blocking buffer and applied for 2 h at RT. After washing, the cells were incubated 1 h with secondary antibodies (Table 3). After application of the secondary antibodies, the cells were washed 2  10 min in PB, 1  10 min in distilled water, air-dried and mounted with Vectashield H-1000 (Abcys). Fluorescence microscopy was performed using a Zeiss Axioplan2 equipped with an ApoTome imaging module, an AxioCam MRc digital camera and the AxioVison 4.2 software (Zeiss).

Acknowledgments We thank Dr. M. Bancila for help with stereotaxic graftings, Drs. P. Durbec, R. Gerardy-Schahn, M. Kursar, S. Pennartz, G. Rougon, D. Rowitch and C. Stiles for antibodies and Dr. L. Aniksztejn for patch clamp of the grafted cells. We are grateful to Drs. P. Durbec, J. Falk and M.C. Tiveron for critically reading of the manuscript. This work was supported by grants to R.S. from the Deutsche Forschungsgemeinschaft and to H.C. from the Fe´de´ration pour la Recherche sur le Cerveau, the European Community Network of Excellence NeuroNE and the Association Franc¸aise contre les Myopathies.

References Aguirre, A.A., Chittajallu, R., Belachew, S., Gallo, V., 2004. NG2expressing cells in the subventricular zone are type C-like cells and contribute to interneuron generation in the postnatal hippocampus. J. Cell Biol. 165, 575 – 589. Alvarez-Buylla, A., Garcia-Verdugo, J.M., 2000. Neurogenesis in adult subventricular zone. J. Neurosci. 22, 629 – 634. Alvarez-Buylla, A., Lim, D.A., 2004. For the long run: maintaining germinal niches in the adult brain. Neuron 41, 683 – 686. Arvidsson, A., Collin, T., Kirik, D., Kokaia, Z., Lindvall, O., 2002. Neuronal replacement from endogenous precursors in the adult brain after stroke. Nat. Med. 8, 963 – 970. Bedard, A., Parent, A., 2004. Evidence of newly generated neurons in the human olfactory bulb. Brain Res. Dev. Brain Res. 19 (151), 159 – 168. Ben-Hur, T., Rogister, B., Murray, K., Rougon, G., Dubois-Dalcq, M., 1998. Growth and fate of PSA-NCAM+ precursors of the postnatal brain. J. Neurosci. 18, 5777 – 5788. Bjorklund, A., Lindvall, O., 2000. Cell replacement therapies for central nervous system disorders. Nat. Neurosci. 3, 537 – 544. Bolteus, A.J., Bordey, A., 2004. GABA release and uptake regulate neuronal precursor migration in the postnatal subventricular zone. J. Neurosci. 24, 7623 – 7631. Bouvier-Labit, C., Liprandi, A., Monti, G., Pellissier, J.F., FigarellaBranger, D., 2002. CD44H is expressed by cells of the oligodendrocyte lineage and by oligodendrogliomas in humans. J. Neuro-oncol. 60, 127 – 134. Broe, M., Kril, J., Halliday, G.M., 2004. Astrocytic degeneration relates to the severity of disease in frontotemporal dementia. Brain 127, 2214 – 2220.

197

Cao, Q., Benton, R.L., Whittemore, S.R., 2002. Stem cell repair of central nervous system injury. J. Neurosci. Res. 68, 501 – 510. Cayre, M., Bancila, M., Virard, I., Borges, A., Durbec, P., 2006. Migrating and myelinating potential of subventricular zone neural progenitor cells in white matter tracts of the adult rodent brain. Mol. Cell. Neurosci. 31 (4), 748 – 758. Coskun, V., Venkatraman, G., Yang, H., Rao, M.S., Luskin, M.B., 2001. Retroviral manipulation of the expression of bone morphogenetic protein receptor Ia by SVZa progenitor cells leads to changes in their p19(INK4d) expression but not in their neuronal commitment. Int. J. Dev. Neurosci. 19, 219 – 227. Dienel, G.A., Hertz, L., 2005. Astrocytic contributions to bioenergetics of cerebral ischemia. Glia 50, 362 – 388. Dietrich, J., Noble, M., Mayer-Proschel, M., 2002. Characterization of A2B5+ glial precursor cells from cryopreserved human fetal brain progenitor cells. Glia 40, 65 – 77. Doetsch, F., Alvarez-Buylla, A., 1996. Network of tangential pathways for neuronal migration in adult mammalian brain. Proc. Natl. Acad. Sci. U. S. A. 93, 14895 – 14900. Doetsch, F., Garcia-Verdugo, J.M., Alvarez-Buylla, A., 1997. Cellular composition and three-dimensional organization of the subventricular germinal zone in the adult mammalian brain. J. Neurosci. 17, 5046 – 5061. 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. Fallon, J., Reid, S., Kinyamu, R., Opole, I., Opole, R., Baratta, J., Korc, M., Endo, T.L., Duong, A., Nguyen, G., Karkehabadhi, M., Twardzik, D., Patel, S., Loughlin, S., 2000. In vivo induction of massive proliferation, directed migration, and differentiation of neural cells in the adult mammalian brain. Proc. Natl. Acad. Sci. U. S. A. 19 (97), 14686 – 14691. Farrer, R.G., Quarles, R.H., 1999. GT3 and its O-acetylated derivative are the principal A2B5-reactive gangliosides in cultured O2A lineage cells and are down-regulated along with O-acetyl GD3 during differentiation to oligodendrocytes. J. Neurosci. Res. 57, 371 – 380. Goldman, S., 2005. Stem and progenitor cell-based therapy of the human central nervous system. Nat. Biotechnol. 23, 862 – 871. Goldman, S.A., Luskin, M.B., 1998. Strategies utilized by migrating neurons of the postnatal vertebrate forebrain. Trends. Neurosci. 21, 107 – 114. Hachem, S., Aguirre, A., Vives, V., Marks, A., Gallo, V., Legraverend, C., 2005. Spatial and temporal expression of S100B in cells of oligodendrocyte lineage. Glia 51, 81 – 97. Hack, I., Bancilla, M., Loulier, K., Carroll, P., Cremer, H., 2002. Reelin is a detachment signal in tangential chain-migration during postnatal neurogenesis. Nat. Neurosci. 5, 939 – 945. Hack, M.A., Saghatelyan, A., de Chevigny, A., Pfeifer, A., Ashery-Padan, R., Lledo, P.M., Gotz, M., 2005. Neuronal fate determinants of adult olfactory bulb neurogenesis. Nat. Neurosci. 8, 865 – 872. Hadjantonakis, A.K., Gertsenstein, M., Ikawa, M., Okabe, M., Nagy, A., 1998. Generating green fluorescent mice by germline transmission of green fluorescent ES cells. Mech. Dev. 76, 79 – 90. Hagg, T., 2005. Molecular regulation of adult CNS neurogenesis: an integrated view. Trends Neurosci. 28, 589 – 595. Herrera, D.G., Garcia-Verdugo, J.M., Alvarez-Buylla, A., 1999. Adultderived neural precursors transplanted into multiple regions in the adult brain. Ann. Neurol. 46, 867 – 877. Jow, F., Chiu, D., Lim, H.K., Novak, T., Lin, S., 2004. Production of GABA by cultured hippocampal glial cells. Neurochem. Int. 45, 273 – 283. Keirstead, H.S., Ben-Hur, T., Rogister, B., O’Leary, M.T., Dubois-Dalcq, M., Blakemore, W.F., 1999. Polysialylated neural cell adhesion molecule-positive CNS precursors generate both oligodendrocytes and Schwann cells to remyelinate the CNS after transplantation. J. Neurosci. 19, 7529 – 7536. Kirschenbaum, B., Nedergaard, M., Preuss, A., Barami, K., Fraser, R.A., Goldman, S.A., 1994. In vitro neuronal production and differentiation

198

R. Seidenfaden et al. / Mol. Cell. Neurosci. 32 (2006) 187 – 198

by precursor cells derived from the adult human forebrain. Cereb. Cortex 4, 576 – 589. Kohwi, M., Osumi, N., Rubenstein, J.L., Alvarez-Buylla, A., 2005. Pax6 is required for making specific subpopulations of granule and periglomerular neurons in the olfactory bulb. J. Neurosci. 25, 6997 – 7003. Kosaka, K., Aika, Y., Toida, K., Heizmann, C.W., Hunziker, W., Jacobowitz, D.M., Nagatsu, I., Streit, P., Visser, T.J., Kosaka, T., 1995. Chemically defined neuron groups and their subpopulations in the glomerular layer of the rat main olfactory bulb. Neurosci. Res. 23, 73 – 88. Lie, D.C., Song, H., Colamarino, S.A., Ming, G.L., Gage, F.H., 2004. Neurogenesis in the adult brain: new strategies for central nervous system diseases. Annu. Rev. Pharmacol. Toxicol. 44, 399 – 421. Lin, R.C., Polsky, K., Matesic, D.F., 1993. Expression of gamma-aminobutyric acid immunoreactivity in reactive astrocytes after ischemiainduced injury in the adult forebrain. Brain Res. 600, 1 – 8. Lindvall, O., Kokaia, Z., Martinez-Serrano, A., 2004. Stem cell therapy for human neurodegenerative disorders—How to make it work. Nat. Med. 10, 42 – 50. Liu, Y., Han, S.S., Wu, Y., Tuohy, T.M., Xue, H., Cai, J., Back, S.A., Sherman, L.S., Fischer, I., Rao, M.S., 2004. CD44 expression identifies astrocyte-restricted precursor cells. Dev. Biol. 276, 31 – 46. Liu, Z., Martin, L.J., 2003. Olfactory bulb core is a rich source of neural progenitor and stem cells in adult rodent and human. J. Comp Neurol. 459, 368 – 391. Lois, C., Alvarez-Buylla, A., 1994. Long-distance neuronal migration in the adult mammalian brain. Science 264, 1145 – 1148. Luskin, M.B., 1993. Restricted proliferation and migration of postnatally generated neurons derived from the forebrain subventricular zone. Neuron 11, 173 – 189. Ma, D.K., Ming, G.L., Song, H., 2005. Glial influences on neural stem cell development: cellular niches for adult neurogenesis. Curr. Opin. Neurobiol. 15, 1 – 7. Marshall, C.A., Suzuki, S.O., Goldman, J.E., 2003. Gliogenic and neurogenic progenitors of the subventricular zone: who are they, where did they come from and where are they going? Glia 43, 52 – 61. Mayer-Proschel, M., Kalyani, A.J., Mujtaba, T., Rao, M.S., 1997. Isolation of lineage-restricted neuronal precursors from multipotent neuroepithelial stem cells. Neuron 19, 773 – 785. Mundt-Petersen, U., Petersen, A., Emgard, M., Dunnett, S.B., Brundin, P., 2000. Caspase inhibition increases embryonic striatal graft survival. Exp. Neurol. 164, 112 – 120.

Nacher, J., Crespo, C., McEwen, B.S., 2001. Doublecortin expression in the adult rat telencephalon. Eur. J. Neurosci. 14, 629 – 644. 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. Ochi, S., Lim, J.Y., Rand, M.N., During, M.J., Sakatani, K., Kocsis, J.D., 1993. Transient presence of GABA in astrocytes of the developing optic nerve. Glia 9, 188 – 198. Rakic, P., 2002. Neurogenesis in adult primate neocortex: an evaluation of the evidence. Nat. Rev., Neurosci. 3, 65 – 71. Sanai, N., Tramontin, A.D., Quinones-Hinojosa, A., Barbaro, N.M., Gupta, N., Kunwar, S., Lawton, M.T., McDermott, M.W., Parsa, A.T., ManuelGarcia, V.J., Berger, M.S., Alvarez-Buylla, A., 2004. Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain migration. Nature 19 (427), 740 – 744. Satoh, J., Kim, S.U., 1995. Ganglioside markers GD3, GD2, and A2B5 in fetal human neurons and glial cells in culture. Dev. Neurosci. 17, 137 – 148. Simard, M., Arcuino, G., Takano, T., Liu, Q.S., Nedergaard, M., 2003. Signaling at the gliovascular interface. J. Neurosci. 23, 9254 – 9262. Staugaitis, S.M., Zerlin, M., Hawkes, R., Levine, J.M., Goldman, J.E., 2001. Aldolase C/zebrin II expression in the neonatal rat forebrain reveals cellular heterogeneity within the subventricular zone and early astrocyte differentiation. J. Neurosci. 21, 6195 – 6205. Stenman, J., Toresson, H., Campbell, K., 2003. Identification of two distinct progenitor populations in the lateral ganglionic eminence: implications for striatal and olfactory bulb neurogenesis. J. Neurosci. 23, 167 – 174. Vitry, S., Avellana-Adalid, V., Lachapelle, F., Baron-van Evercooren, A., 2001. Migration and multipotentiality of PSA-NCAM+ neural precursors transplanted in the developing brain. Mol. Cell. Neurosci. 17, 983 – 1000. Zhang, S.C., 2001. Defining glial cells during CNS development. Nat. Rev., Neurosci. 2, 840 – 843. Zigova, T., Pencea, V., Betarbet, R., Wiegand, S.J., Alexander, C., Bakay, R.A., Luskin, M.B., 1998. Neuronal progenitor cells of the neonatal subventricular zone differentiate and disperse following transplantation into the adult rat striatum. Cell Transplant. 7, 137 – 156.