α-Synuclein Alters Notch-1 Expression and Neurogenesis in Mouse Embryonic Stem Cells and in the Hippocampus of Transgenic Mice

Altered expression and mutations in α-synuclein (α-syn) have been linked to Parkinson's disease (PD) and related disorders. The neurological alterations in PD patients have been associated with degeneration of dopaminergic cells and other neuronal populations. Moreover, recent studies in murine models have shown that alterations in neurogenesis might also contribute to the neurodegenerative phenotype. However, the mechanisms involved and the effects of α-syn expression on neurogenesis are not yet clear. To this end, murine embryonic stem (mES) cells were infected with lentiviral (LV) vectors expressing wild-type (WT) and mutant α-syn. Compared with mES cells infected with LV–green fluorescent protein (GFP), cells expressing WT and mutant α-syn showed reduced proliferation as indicated by lower 5-bromo-2′-deoxyuridine uptake, increased apoptosis, and reduced expression of neuronal markers such as neuron specific enolase and β-III tubulin. The alterations in neurogenesis in α-syn-expressing mES cells were accompanied by a reduction in Notch-1 and Hairy and enhancer of split-5 (Hes-5) mRNA and protein levels. Moreover, levels of total Notch-1 and Notch intracellular domain (NICD) were lower in mES cells expressing WT and mutant α-syn compared with GFP controls. The reduced survival of α-syn-expressing mES cells was reverted by overexpressing constitutively active NICD. Similarly, in α-syn transgenic mice, the alterations in neurogenesis in the hippocampal subgranular zone were accompanied by decreased Notch-1, NICD, and Hes-5 expression. Together, these results suggest that accumulation of α-syn might impair survival of NPCs by interfering with the Notch signaling pathway. Similar mechanisms could be at play in PD and Lewy body disease.


Introduction
Progressive accumulation of misfolded ␣-synuclein (␣-syn) in cortical and subcortical brain regions has been proposed to play a central role in the pathogenesis of Parkinson's disease (PD) and Lewy body disease (LBD) . Supporting a role for the presynaptic protein ␣-syn in the pathogenesis of this heterogeneous group of movement disorders with dementia (Iwai et al., 1994), previous studies have shown that (1) this molecule accumulates in Lewy bodies (Spillantini et al., 1997;Wakabayashi et al., 1997;Takeda et al., 1998), (2) mutations and multiplication in the ␣-syn gene are associated with rare familial forms of parkinsonism (Polymeropoulos et al., 1997;Kruger et al., 1998;Singleton et al., 2003), and (3) its expression in transgenic (tg) mice (Masliah et al., 2000;Lee et al., 2002Lee et al., , 2004 and Drosophila (Feany and Bender, 2000) mimics several aspects of PD.
The mechanisms through which ␣-syn aggregates might lead to neurodegeneration are not entirely clear. Recent evidence indicates that abnormal accumulation of misfolded ␣-syn in the synaptic terminals and axons plays an important role (Iwatsubo et al., 1996;Hashimoto and Masliah, 1999;Lansbury, 1999;Hashimoto et al., 2004;Kramer and Schulz-Schaeffer, 2007). Neurodegeneration in these disorders affects primarily the striatonigral system, but, in cases with cognitive impairment, there is more widespread degeneration including neuronal populations in the striatum, hippocampus, and neocortex (Harding et al., 2002;McKeith et al., 2004). Moreover, recent evidence suggests that accumulation of ␣-syn in the limbic system might contribute to the neurodegenerative phenotype by interfering with adult neurogenesis (Winner et al., 2004(Winner et al., , 2007. Neurogenesis in the mature CNS occurs in the olfactory bulb, the hippocampus, and the subventricular zone (Gage et al., 1998). In the hippocampus, neural precursor cells (NPCs) in the subgranular zone (SGZ) contribute to the maturation of neuronal populations in the dentate gyrus (DG) in which they might play a role in learning and memory (Gage et al., 1998;van Praag et al., 1999van Praag et al., , 2002Brown et al., 2003).
We have shown previously that accumulation of wild-type (WT) and mutant (mut) ␣-syn in the CNS of tg mice results in reduced neurogenesis in the olfactory bulb and hippocampus (Winner et al., 2004(Winner et al., , 2007. Moreover, recent studies have shown that overexpression of ␣-syn in human and mouse embryonic stem (mES) cells results in defective neurogenesis and cell death and might recapitulate some aspects of the neurodegenerative process in PD (Yamashita et al., 2006;Schneider et al., 2007). However, the mechanisms through which accumulation of ␣-syn might lead to reduced neurogenesis are not completely clear. For this purpose, we investigated the effects of abnormal ␣-syn accumulation in pathways involved in neurogenesis in mES cells that were infected with lentivirus (LV) expressing ␣-syn and in the DG of ␣-syn tg mice. Our studies suggest that ␣-syn accumulation contributes to alterations in neurogenesis by reducing the survival of NPCs via downregulation of Notch-1 expression.

Materials and Methods
ES cell culture. The D3 mES cell line (purchased from American Type Culture Collection, Manassas, VA) was cultured on tissue culture plates coated with 0.1% (v/v) porcine gelatin (Sigma, St. Louis, MO). Cells were cultured in ES medium, consisting of knock-out DMEM (Invitrogen, Carlsbad, CA), 15% knock-out serum replacement (Invitrogen), 55 M mercaptoethanol, 1 mM sodium pyruvate, 2 mM glutamine, 0.1 mM nonessential amino acids, and 1000 U/ml murine leukemic inhibitory factor (LIF) (Millipore Bioscience Research Reagents, Temecula, CA), in 5% CO 2 , 95% air at 37°C. Cells were trypsinized and replated or replaced with fresh ES medium every other day.
Neuronal differentiation of mES cells. Neuronal differentiation of ES cells was based on a modification of previous published methods (Bibel et al., 2004). Briefly, for embryoid body formation, 3 ϫ 10 6 mES cells were plated onto nonadherent Petri dishes in EB medium (ES medium without LIF and only 10% serum replacement) and incubated for 8 d. Medium was changed every 2 d, and 5 M retinoic acid (Sigma) was added after 4 d. Embryoid bodies were then dissociated with 0.05% trypsin in 0.04% EDTA/PBS (Invitrogen), and the cells were plated on L-ornithine and laminin-coated plates in Neurobasal medium (Invitrogen) in the presence of 1 mM L-glutamine and N2 supplement (Invitrogen) (N2 medium). The following day, the N2 medium was changed, and after 2 d, the medium was replaced by Neurobasal medium in the presence of 1 mM L-glutamine and B27 supplement (Invitrogen) (B27 medium).
After washing, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. Detection was performed using the VersaDoc Imaging System (Bio-Rad).
Quantitative real-time PCR. RNA was purified from cultured cells or the hippocampus from control and tg mice using the RNeasy mini kit (Qiagen, Valencia, CA). Total RNA was reverse transcribed using iScript cDNA Synthesis kit (Bio-Rad) with 1 g of total RNA in accordance with the instructions of the manufacturer. Quantitative PCR (qPCR) was performed with primers specific for Notch-1, Jagged, Delta, Notch-4, Hes-1, Hes-5, and ␤-actin (Table 1) using the iCycler iQ Real-Time PCR Detection System (Bio-Rad). Reactions were performed in a volume of 25 l using the iQ SYBR Green Supermix (Bio-Rad) according to the instructions of the manufacturer.
Production of lentivirus vector. Vector plasmids were constructed essentially as described previously (Marr et al., 2003;Singer et al., 2005) for the production of third-generation LV vectors that expressed WT ␣-syn, A53T mut ␣-syn, or green fluorescent protein (GFP). The human cytomegalovirus promoter was used to drive expression of the transgenes. Lentiviral vectors were generated using a four-plasmid transfection with Lipofectamine transfection reagent (Invitrogen). Briefly, as described previously (Naldini et al., 1996a,b), 293T cells were transfected with vector and packaging plasmids (pMDL g/p.RRE, pMD.G, and pRSV-Rev) and incubated for 3 d. The supernatants including vectors were collected, and cell debris was eliminated by centrifugation.
Lentivirus and adenovirus-mediated infection of mES cells. For delivery of ␣-syn, the LV system was used. The day before transduction, 3 ϫ 10 6 mES cells were plated on 0.1% gelatin-coated 10 cm dishes in mES medium. After 1d, the mES culture medium was changed, and LV-GFP, LV-WT ␣-syn, or LV-mut ␣-syn was added to the dishes. The mES cells were incubated for 24 -72 h in a 5% CO 2 , 95% air at 37°C. All LV experiments were performed in cells infected with both vector and GFP controls, but most of the data included in the results used vector-infected cells to allow for double-labeling immunocytochemical analysis. An adenoviral vector system was used for intracellular delivery of NICD or LacZ control. For transduction, the cells were incubated with adenovirus for 24 h in 5% CO 2 , 95% air at 37°C.
In vitro cell proliferation and viability assays. Cell proliferation rates were determined by the amount of 5-bromo-2Ј-deoxyuridine (BrdU) incorporation using a labeling and detection kit (Roche Diagnostics) in accordance with the instructions of the manufacturer. Briefly, dissociated embryoid bodies were cultured in 24-well plates (2 ϫ 10 5 cells) and differentiated into neurons according to the protocol described above. After 9 d, BrdU (10 M final concentration) was added, and the cells were incubated for 24 h. The cells were fixed and incubated with anti-BrdUperoxidase (20 U/ml) for 30 min at 37°C. The color was developed by the addition of peroxidase substrate, 2,2Ј-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid), and measured at 405 nm (reference at 490 nm). Cell death by apoptosis was evaluated in in vitro cultures by modified terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay (ApopTag In Situ Apoptosis Detection kit; Millipore Bioscience Research Reagents) (Biebl et al., 2005) using fixed cells prepared on glass coverslips.
␥-Secretase activity assay. ␥-Secretase activity was measured using the ␥-secretase activity kit (R & D Systems, Minneapolis, MN) according to the instructions of the manufacturer.
Transgenic mice. Transgenic mice expressing human WT or A53T mut ␣-syn under the control of the platelet-derived growth factor-␤ promoter were used. Mice were generated and bred as described previously (Masliah et al., 2000;Hashimoto et al., 2003). For these experiments, a total of 15 mice (n ϭ 5 per group; 5 months old; C57BL6/DBA) were used.
All mice were killed following National Institutes of Health guidelines for the humane treatment of animals. The hippocampus from the right hemibrain was dissected and used for RNA analysis. The left hemibrain was fixed in 4% paraformaldehyde/PBS for 24 h and placed in 30% sucrose/PBS solution. The brains were cut sagittally into 25 m sections using a sliding microtome on dry ice. The sections were stored in cryoprotectant (ethylene glycol, glycerol, 0.1 M phosphate buffer pH 7.4, 1:1:2 by volume) at Ϫ20°C until processed further for immunocytochemical analysis of neurogenesis and Notch expression.
Analysis of apoptosis. Histological detection of cell death in vibratome sections was performed using the TUNEL assay (ApopTag In Situ Apoptosis Detection kit; Millipore Bioscience Research Reagents) in a modified procedure for free-floating sections as described previously (Biebl et al., 2005).
Image analysis and stereology. For quantification of neurogenic cells, a systematic random counting procedure, similar to the optical disector (Gundersen et al., 1988a,b), was used as described by Williams and Rakic (1988). To determine the number of DCX-, PCNA-or TUNEL-positive (TUNEL ϩ ) cells, every sixth section (160 m interval) of the left hemisphere was selected from each animal and processed for immunohistochemistry. Sections were analyzed for DCX-, PCNA-, nestin-, NeuN-, GFAP, or TUNEL-positive cells in the SGZ of the DG. As described previously (Winner et al., 2004(Winner et al., , 2007, BrdU-positive (newborn cells) cells were distinguished from BrdU/NeuN double-positive cells (newborn neurons) and BrdU/GFAP double-positive cells (newborn astrocytes), and ratios of these cell types were determined to assess the number of newborn cells that ultimately achieve neuronal or astrocytic fates. The reference volume was determined by tracing the areas using a semiautomatic stereology system (Stereoinvestigator; MicroBrightField, Colchester, VT). Positive cells were counted within a 30 ϫ 30 m counting frame, which was spaced in a 300 ϫ 300 m counting grid. Positive profiles that intersected the uppermost focal plane (exclusion plane) or the lateral exclusion boundaries of the counting frame were not counted. The total counts of positive profiles were multiplied by the ratio of reference volume to sampling volume to obtain the estimated number of positive cells for each structure.
Statistical analysis. The data are expressed as mean values Ϯ SEM. Statistical analysis was performed using one-way ANOVA followed by the Bonferroni's multiple comparison test (Prism Graph Pad Software; GraphPad Software, San Diego, CA). The significance level was set at p Ͻ 0.05.

␣-syn accumulation interferes with survival of mES cellderived neural progeny
To investigate the mechanisms through which ␣-syn might impair the development and maturation of NPCs, the mES cell system was used as a model to recapitulate the effects of ␣-syn on neurogenesis. The neuronal differentiation protocol used in this study has been described previously (Bibel et al., 2004) and includes five stages (A through E) ( Fig. 1 A). After withdrawal of LIF and treatment with retinoic acid, the mES cells ( Fig. 1 B) (stage A) formed embryoid bodies (Fig. 1C) (stages B and C). After 4 d, embryoid bodies were dissociated into the putative NPCs ( Fig.  1 D) (stage D). The EB media was then replaced with B27 media, and cells began to differentiate into a neuronal phenotype ( Fig.  1 E, F ) (stage E). At early stages of the differentiation protocol (stages A and B), the precursor cells expressed nestin (supplemental Fig. 1 A-C, available at www.jneurosci.org as supplemental material) and later DCX (stage C-D) (supplemental Fig.  1 D-F, available at www.jneurosci.org as supplemental material). Over the course of neuronal differentiation (8 d) (Fig. 1 E, F, J, K ), low levels of ␤-tubulin and NSE expression were first noted at stage D, with a progressive increase during stage E, reaching their peak at day 8 (E8) (Fig. 1G-K ) (supplemental Fig. 1G-I, available at www.jneurosci.org as supplemental material). This was confirmed by Western blot analysis (supplemental Fig. 1 J, available at www.jneurosci.org as supplemental material). By the end of stage E, ϳ85% of the cells expressed neuronal markers, including ␤-tubulin, MAP2 (data not shown), and NSE and displayed polarization with formation of elongated neuritic processes (supplemental Fig. 1 I, available at www.jneurosci.org as supplemental material). Approximately 5% of the cells expressed GFAP, whereas the remaining cells did not express neuronal or astroglial markers (data not shown). These heterogeneous cultures were composed primarily of cells expressing immature and mature neuronal markers, with a minority of cells expressing astroglial markers or markers of proliferating NPCs; therefore, we defined these cultures as "mES cell-derived neural progeny." To address the effects of ␣-syn on neuronal differentiation, at stage A, mES cells were infected with LV vectors encoding GFP ( Fig. 1 L), WT ␣-syn, or mut ␣-syn. Approximately 48 h after infection, Ͼ95% of the mES cells expressed GFP (Fig. 1 M). Expression of this protein was observed at the various stages of the differentiation process and was abundantly produced by cells displaying neuronal morphology at the end of the procedure (Fig.  1 L-P). This indicates that the LV strategy used was effective at driving the transgene expression in mES cells, viability was not compromised, and levels of transgene expression were maintained throughout the differentiation process.
Similarly, mES cells infected with control LV vector demonstrated formation of embryoid bodies and neuronal differentia-tion (supplemental Fig. 2 A-L, available at www.jneurosci.org as supplemental material). mES cells infected with LV-␣-syn displayed sustained levels of the transgene product through the various time points of the protocol (supplemental Fig. 2 M-X, available at www.jneurosci.org as supplemental material) and was observed to accumulate in differentiated neurons 4 d into the final phase (supplemental Fig. 2V-X, available at www.jneurosci.org as supplemental material). Although high levels of ␣-syn were observed at all stages, there was no apparent disruption of the differentiation process from embryoid bodies to progenitor cells to neurons (supplemental Fig. 2 M-X, available at www.jneurosci.org as supplemental material). However, compared with cells infected with an LV control (Fig. 2 A-C) or GFP (data not shown), mES cells expressing WT and mut ␣-syn displayed alterations in their neuritic processes (Fig. 2 D-I ) with decreased branching, length (Fig. 2 J), and expression of ␤-tubulin (Fig.  2 K). Similar to the alterations in ␤-tubulin expression, the mature neuronal markers NSE and MAP2 were also reduced at stage E8 (data not shown). In these cells, ␣-syn accumulated in the cytoplasm, forming irregular aggregates (Fig. 2 D-I ). Immunoblot analysis showed that both WT and mut ␣-syn were expressed at high and comparable levels at the various stages of differentiation (Fig. 2 L). Immunoblot analysis also confirmed that, com- pared with vector control, mES cells infected with WT ␣-syn or mut ␣-syn displayed a significant decrease in levels of ␤-tubulin expression (Fig. 2 L, M ).
The reduced expression of neuronal markers in the differentiating mES cells suggests that ␣-syn interferes with neuronal differentiation and the maturation of newly developing neurons or increases cell death of newborn neurons. To further characterize the effects of ␣-syn in this process, BrdU and TUNEL assays were performed. These analyses showed that, compared with LV control, in WT and mut ␣-syn-expressing cells, BrdU uptake was reduced by ϳ30% (Fig. 2 N), whereas the number of TUNEL-positive cells was increased approximately twofold to threefold (Fig. 2O) by the end of the neuronal differentiation protocol. Together, these results suggest that the reduction in neuronal markers in differentiating mES cells may primarily be an effect of reduced cell survival mediated by the accumulation of ␣-syn in NPCs.

Overexpression of ␣-syn in mES cells results in reduced Notch-1 expression in neural progeny
Because previous reports indicate that Notch signaling is required for the survival of NPCs (Mason et al., 2006) and neuronal differentiation Mason et al., 2005), we investigated the possibility that ␣-syn might disturb the process of neurogenesis by interfering with the expression of Notch-1 and related genes. Immunoblot analysis showed that levels of full-length (FL) and cleaved (NCID) Notch-1 were dramatically reduced ( Fig.  3 A, B). Immunocytochemical studies confirmed that levels of FL-Notch-1 protein and active Notch-1 that translocated into the nucleus were reduced in ␣-synexpressing neural progeny derived from mES cells (Fig. 3C-F ). Consistent with the results obtained by immunoblot, by qPCR, only levels of Notch-1 mRNA (Fig. 3G), but not the Notch ligands Jagged or Delta (Fig. 3H ), were significantly decreased in mES cells infected with LV-WT ␣-syn and mut ␣-syn at E8 of the neuronal differentiation protocol. To further investigate the effects of ␣-syn expression on Notch-1 signaling, levels of downstream targets of the Notch signaling pathway were analyzed by qPCR. These studies demonstrated a reduction in the levels of Hes-5, but levels of Hes-1 appeared unchanged (Fig. 3I ). Moreover, analysis of an additional member of the Notch receptor family, Notch-4, also showed no significant changes in expression levels (Fig. 3G), suggesting that the effects of ␣-syn are specific to the Notch-1 signaling cascade and downregulate signaling through this pathway primarily via Hes-5. Notch-1 activation is known to be regulated by ␥-secretase, so ␥-secretase activity was then examined in dif-ferentiated cells at the E8 time point. Although cleavage of Notch-1 was decreased in LV-␣-syn-infected cells, there were no significant differences in ␥-secretase activity (Fig. 3J ), suggesting that Notch-1 inactivation in ␣-syn-overexpressing cells is likely a result of decreased Notch-1 expression.

Adenovirus-driven NICD expression reverses the defects in neurogenesis by increasing survival of ␣-syn-expressing mEScell derived neural progeny
To further investigate the possibility that Notch-1 might be involved in the alterations in NPC survival observed in differenti- ated LV-␣-syn-expressing mES cell cultures, we transduced NICD into mES cell-derived neural progeny using an adenoviral vector. The efficiency of infection by adenovirus encoding lacZ was determined by ␤-galactosidase staining of cells infected at different multiplicities of infection (MOI): 3, 10, or 30. Even the lowest MOI showed high transduction efficiency. NICD expression was also confirmed by immunoblotting with adenoviralinfected cell lysates (Fig. 4 A). To analyze the effect of NICD transduction on cell proliferation, we performed immunoblotting and cell proliferation (BrdU uptake) assays with the mES cell-derived neural progeny infected with adenoviral vectors encoding lacZ or NICD. Notch pathway activation by NICDtransduction reversed the observed decrease of ␤-tubulin expression in ␣-syn-infected cells (Fig. 4 A, B), increased cell proliferation to levels comparable with the control samples (Fig.  4C), and reduced the amount of cells undergoing apoptosis in both WT ␣-syn and mut ␣-syn-infected cells (Fig. 4 D).

Alterations in neurogenesis in ␣-syn tg mice are accompanied by reduced Notch-1 expression
Previous studies have shown that neurogenesis is disturbed in Parkinson's disease (Hoglinger et al., 2004) as well as in animal models (Winner et al., 2004(Winner et al., , 2007. Therefore, to further investigate in vivo the involvement of Notch-1 in the ␣-syn-dependent alterations, neurogenesis was analyzed in the hippocampus of young ␣-syn tg mice. Compared with non-tg controls, in the SGZ of the DG of the ␣-syn tg mice, there was a significant (40 -50%) decrease in the number of surviving BrdU-positive NPCs (Fig.  5A-C). This was accompanied by a reduction in the numbers of DCX ϩ neuroblasts (Fig. 5D-F ) that was more severe in the mut ␣-syn tg mice. To better understand whether the alterations in neurogenesis were related to disturbances in differentiation or increased cell death, PCNA, TUNEL, and double-labeling studies were performed. This analysis showed comparable numbers of proliferating PCNA ϩ (Fig. 5G) NPCs in the non-tg and WT Figure 3. Effects of ␣-syn on Notch-1 expression in mES cells. A, Immunoblot analysis of neuronal differentiated cells harvested at stage E8. Proteins in total cell lysates were separated by SDS-PAGE and immunoblotted with anti-␤-tubulin, anti-Notch-1 (full-length), anti-NICD, or anti-␤-actin. B, Quantification of Notch-1 expression (normalized to ␤-actin levels). C-E, Doubleimmunolabeling analysis of differentiated neuronal cells at stage E8 with antibodies against Notch-1 and ␤-tubulin. Scale bar, 30 m. F, Levels of Notch-1 immunoreactivity in stage E8 differentiated neuronal cells infected with LV-vector, WT ␣-syn, or A53T mut ␣-syn. G, mRNA expression of Notch-1 and Notch-4 as measured by qPCR in stage E8 neuronal differentiated mES cells infected with LV-vector, WT ␣-syn, or A53T mut ␣-syn. H, mRNA expression of Jagged and Delta as measured by qPCR in stage E8 neuronal differentiated mES cells infected with LV-vector, WT ␣-syn, or A53T mut ␣-syn. I, mRNA expression of Hes-1 and Hes-5 as measured by qPCR in stage E8 neuronal differentiated mES cells infected with LV-vector, WT ␣-syn, or A53T mut ␣-syn. J, ␥-Secretase activity in stage E8 differentiated mES cells infected with LV-vector, WT ␣-syn, or A53T mut ␣-syn. *p Ͻ 0.05 compared with vector-infected control cells (one-way ANOVA with post hoc Dunnett's test).
␣-syn tg mice but a significant decrease in the mut ␣-syn tg mice. Moreover, there was an increase in the numbers of TUNEL ϩ NPCs (Fig. 5H ), consistent with previous studies in aged ␣-syn tg mice (Winner et al., 2004(Winner et al., , 2007. This decreased survival of NPCs in the DG was more severe in the mut ␣-syn tg mice than in the WT ␣-syn tg animals (Fig. 5H ). In contrast, numbers of nestinpositive progenitor cells (Fig. 5I ) were unchanged. In addition, the proportion of BrdU-positive NPCs converting into NeuNpositive mature neurons or GFAP-positive mature astrocytes were unchanged (Fig. 5J ). Together, these results suggest that the reduced neurogenesis in the ␣-syn tg mice might be related to decreased survival of NPCs.
Because Notch-1 signaling has been shown to play an important role in neurogenesis and neuronal survival Mason et al., 2006), double-immunolabeling studies were performed to confirm that Notch-1-expressing neuroblasts also express ␣-syn and markers of neurogenesis. This analysis showed that neuroblasts in the hippocampus of ␣-syn tg mice coexpressed Notch-1 and ␣-syn (Fig. 6 A-C), Notch-1 and DCX (Fig.  6 D-F ), and ␣-syn and DCX (Fig. 6G-I ). To further investigate the role of Notch-1 in neurogenesis, levels of components of this pathway were analyzed by immunocytochemistry, immunoblot, and qPCR. Consistent with previous in situ hybridization studies (Stump et al., 2002), with the antibody against Notch-1, granular cells in the DG of the non-tg mice showed punctate cytoplasmic immunostaining, and NPCs in the SGZ displayed primarily a diffuse cytoplasmic pattern of immunoreactivity (Fig. 7A). The neuropil in the molecular layer also showed punctate immunolabeling. Compared with control mice, and consistent with the results in mES cells, the levels of Notch-1 immunoreactivity were significantly reduced in the DG of ␣-syn tg mice (Fig. 7A-C). Consistent with the immunocytochemical studies, Western blot ( Fig. 7D-F ) and qPCR (Fig. 7G-I ) analyses confirmed that the expression of Notch-1 (Fig. 7 D, E,G) and Hes-5 (Fig. 7 D, F,I ) were reduced in the ␣-syn tg mice compared with non-tg controls, whereas expression of Notch-4 (Fig. 7 D, E,G), Hes-1 (Fig.   7 D, F,I ), and Jagged and Delta (Fig. 7H ) were not affected. Together, these in vivo data support the notion that ␣-syn might interfere with neurogenesis by reducing NPC survival via downregulation of Notch-1 expression and signaling.

Discussion
The present study showed that overexpression of ␣-syn in mES cells reduced NPC survival, leading to a disruption of neurogenesis. These results are consistent with recent studies in mouse (Yamashita et al., 2006) and human (Schneider et al., 2007) ES cells, showing that ␣-syn compromises the viability of NPCs. In the present study, we report that LV-driven expression of WT and mut ␣-syn in mES cells resulted in reduced BrdU uptake and ␤-tubulin expression, accompanied by an increase in the numbers of mES cell-derived neural progeny undergoing apoptosis. The reduced viability observed in the ␣-syn-expressing neural progeny is in agreement with a recent report by Yamashita et al. (2006). In their model, mES cells transfected to express fulllength human WT ␣-syn or the A30P and A53T mutated forms were differentiated into dopaminergic neurons. The resulting neurons showed increased susceptibility to cellular stress triggered by oxidative agents and proteasome and mitochondrial inhibition, factors likely to be involved in PD pathogenesis (Hattori and Mizuno, 2004). Notably, when the same ES cell lines were differentiated into oligodendrocytes, no differences in cell viability were noted, suggesting that ␣-syn-induced cell death might be neuron specific (Yamashita et al., 2006). In another study, it has been reported that, in human NPCs derived from ES cells, overexpression of mut ␣-syn (A53T) also compromised the viability and maturation of neurons (Schneider et al., 2007). Moreover, WT and A53T mut ␣-syn progressively impaired the innate ability of progenitors to switch toward gliogenesis at later passages (Schneider et al., 2007).
ES cell models have been incorporated recently into PD research as an alternative tool for modeling the human disease. Although most studies focus on dopaminergic neurons, a growing number of reports have shown that other subcortical neuronal populations are affected in PD as well, such as pyramidal neurons in the hippocampus and neocortex of the brains of patients with LBD (Harding et al., 2002;McKeith et al., 2004). In addition, ␣-syn accumulation in the limbic system has been shown to interfere with neurogenesis in the hippocampus and olfactory bulb in aged adult ␣-syn tg mice (Winner et al., 2004(Winner et al., , 2007. Consistent with these latter observations, in young-adult ␣-syn tg mice, we observed decreased neurogenesis in the SGZ of the DG. The alterations in neurogenesis in the WT and mut ␣-syn tg mice were primarily related to decreased survival of NPCs (as reflected by increased TUNEL staining and decreased BrdU uptake after 30 d) rather than to reduced neuronal differentiation. In addition, in the mut ␣-syn tg mice, there was reduced proliferation of NPCs. Together, the studies in the mES cells as well as in tg mice suggest that the alterations in neurogenesis are related to reduced survival of NPCs.
The mechanisms through which accumulation of ␣-syn in precursor cells might impair neurogenesis are not completely clear. Our studies in ␣-syn-overexpressing mES cells and in young-adult ␣-syn tg mice offer a clue suggesting that accumulation of ␣-syn might impair neurogenesis by reducing NPC survival via downregulation of Notch-1 expression. Notch-1 is traditionally understood to be required for neuronal and glial differentiation (Louvi and Artavanis-Tsakonas, 2006), but previous studies also implicate Notch-1 in the survival of NPCs (Oishi et al., 2004;Mason et al., 2006). Moreover, an elegant recent report showed that Notch-1 regulates the fate and dendritic morphology of newborn neurons in the postnatal DG (Breunig et al., 2007). In support of these dual roles for Notch-1, studies using conditional gene ablation have shown that loss of Notch-1 results in incomplete differentiation and elimination by apoptosis , suggesting that Notch activity supports the survival of both progenitors and newly differentiating cells in the developing nervous system (Mason et al., 2006). Therefore, restoration of Notch signaling might allow the completion of neuronal differentiation by promoting the survival of NPCs. In support of this possibility, we showed that gene transfer of NICD recovers the expression of neuronal markers in ␣-synoverexpressing mES cell-derived neural progeny, and this is accompanied by increased BrdU uptake and decreased numbers of cells displaying staining by TUNEL assay. This might suggest that the downregulation of Notch-1 by ␣-syn interferes with neurogenesis by triggering elimination of these neuronal precursors via apoptosis. The mechanisms through which Notch-1 might regulate survival of NPCs may involve pathways distinct from those targeted in differentiation (Oishi et al., 2004). Future studies will be necessary to better understand the precise connection between Notch, ␣-syn, neurogenesis, and programmed cell death.
Although multiple pathways may be involved in Notch-1-mediated regulation of NPC survival, the canonical Notch-1 signaling cascade targets the transcriptional regulators Hes-1 and Hes-5. Because in NPCs accumulation of ␣-syn interferes with neurogenesis by downregulating Notch-1, it was predicted that downstream signaling molecules in this pathway might be similarly affected. In agreement with this possibility, we found that levels of Hes-5 were significantly reduced in mES cells and tg mice overexpressing ␣-syn.
Accumulation of ␣-syn in NPCs might lead to reduced Notch-1 via transcriptional or posttranscriptional effects. Because in our in vitro model we did not detect a decrease in ␥-secretase activity, and Jagged and Delta ligands were not affected in neurons derived from LV-␣-syninfected mES cells, the observed effects on Notch-1 expression are most likely the result of alterations in transcription. The mechanisms through which accumulation of ␣-syn might disrupt Notch-1 gene expression are unclear. One possibility might be interference with transcriptional machinery. In support of this, ␣-syn has been found to be associated with histone proteins (Goers et al., 2003) and affects both their acetylation and expression levels (Goers et al., 2003;Duce et al., 2006;Kontopoulos et al., 2006;Vartiainen et al., 2006). Although ␣-syn is almost exclusively present in nerve terminals during adulthood (Iwai et al., 1994), it has been found in the perikarya during development (Hsu et al., 1998;Galvin et al., 2001;Raghavan et al., 2004). This change in subcellular localization indicates that ␣-syn function may evolve as neural differentiation and maturation progresses (Winner et al., 2004). NPCs express endogenous ␣-syn, although at a low level. When this protein is overexpressed, it accumulates in both cytoplasmic microvesicles and the cell nucleus (Schneider et al., 2007). In this context, it is possible that increased accumulation of ␣-syn in NPCs might result in nuclear translocation and interactions with the DNA matrix, as suggested by its association with histone proteins (Goers et al., 2003). This in turn might influence the transcription patterns of genes involved in neurogenesis such as Notch-1. Verification of this hypothesis and additional studies on other genes affected by ␣-syn awaits further investigation.
Accumulation of ␣-syn is central to the pathogenesis of disorders with parkinsonism . Abnormal accumulation is considered to be the result of alterations in the balance between ␣-syn production, aggregation, and clearance (Hashimoto et al., 2004). Although accumulation of ␣-syn in mature neurons might result in synaptic dysfunction and eventually in cell death, in NPCs ␣-syn interferes with neurogenesis, more specifically with the survival of neural progeny in the terminal phase of neuronal differentiation. This is supported by our studies in mES cells demonstrating that the alterations in the expression of neuronal markers occurred in the latest stages of neuronal differentiation (E8). Moreover, in the tg mice, abnormal accumulation of ␣-syn in NPCs reduced the number of DCX-positive neuroblasts but did not alter the number of progenitor cells in the hippocampal DG. Neurogenesis is a multistep process that includes proliferation of precursor cells, migration of the newborn cells, differentiation, integration into neuronal circuits, and survival (Scholzke and Schwaninger, 2007). Alterations in neurogenesis are important not only for familial but also for sporadic forms of LBD, because recent studies have shown that, in the adult CNS, this process plays an important role in learning and memory (van Praag et al., 1999(van Praag et al., , 2002Brown et al., 2003). Our results reveal that the alterations in Notch-1 expression associated with ␣-syn accumulation plays a detrimental role in the later stages of neurogenesis by interfering with the survival of neural progeny, resulting in the elimination of these cells via apoptosis.
In summary, these results suggest that accumulation of ␣-syn might contribute to the neurodegenerative process by impairing neurogenesis and survival of differentiating neural progeny through alterations in Notch-1 expression. The subsequent disruption of the Notch signaling pathway might result in a decrease in NPC proliferation and survival. Together, these results suggest that accumulation of ␣-syn plays a predominantly detrimental role in the survival of NPCs, triggering their elimination by apoptosis. Similar mechanisms could be at play in PD and LBD and might contribute to the neurodegenerative phenotype in the limbic system.  . Immunocytochemical, immunoblot, and qPCR analyses of the effects of ␣-syn on Notch expression in the hippocampus of tg mice. Images are from the DG, including the molecular layer (ML), DG, and SGZ of 5-month-old mice (n ϭ 5). A, Patterns of Notch-1 expression in the SGZ of a non-tg mouse. B, Reduced Notch-1 expression in an A53T mut ␣-syn tg mice. Scale bar, 50 m. C, Analysis of levels of Notch-1 expression (immunoreactivity) in the DG. D, Proteins in brain homogenates from non-tg, WT ␣-syn tg, and A53T mut ␣-syn tg mice were separated by SDS-PAGE and immunoblotted with antibodies against Notch-1, Notch-4, Hes-1, Hes-5, or ␤-actin, as indicated. E, Quantitative analysis of immunoblot levels of Notch-1 and Notch-4. F, Quantitative analysis of immunoblot levels of Hes-1 and Hes-5. G, Comparison of levels of Notch-1 and Notch-4 mRNA in the hippocampus by qPCR. H, Comparison of levels of Jagged and Delta mRNA in the hippocampus by qPCR. I, Comparison of levels of Hes-1 and Hes-5 mRNA in the hippocampus by qPCR. *p Ͻ 0.05 compared with non-tg control (one-way ANOVA with post hoc Dunnett's test).