Maturation of Synaptic Contacts in ... - Wiley Online Library

TISSUE-SPECIFIC STEM CELLS Maturation of Synaptic Contacts in Differentiating Neural Stem Cells STEFAN LIEBAU,a BIANCA VAIDA,a ALEXANDER STORCH,b TOBIAS M. BOECKERSa Institute of Anatomy and Cell Biology, Ulm University, Ulm, Germany; bDepartment of Neurology, Technical University of Dresden, Dresden, Germany

a

Key Words. Synaptogenesis • Neural stem cells • Synaptophysin • Bassoon • ProSAP • Shank

ABSTRACT NSCs are found in the developing brain, as well as in the adult brain. They are self-renewing cells that maintain the capacity to differentiate into all major brain-specific cell types, such as glial cells and neurons. However, it is still unclear whether these cells are capable of gaining full functionality, which is one of the major prerequisites for NSCbased cell replacement strategies of neurological diseases. The ability to establish and maintain polarized excitatory synaptic contacts would be one of the basic requirements for intercellular communication and functional integration into existing neuronal networks. In primary cultures of hippocampal neurons, it has already been shown that synaptogenesis is characterized by a well-ordered, time-dependent targeting and recruitment of pre- and postsynaptic proteins. In this study, we investigated the expression and localization

of important pre- and postsynaptic proteins, including Bassoon and synaptophysin, as well as proteins of the ProSAP/ Shank family, in differentiating rat fetal mesencephalic NSCs. Moreover, we analyzed the ultrastructural features of neuronal cell-cell contacts during synaptogenesis. We show that NSCs express and localize cytoskeletal and scaffolding molecules of the pre- and postsynaptic specializations in a well-defined temporal order, leading to mature synaptic contacts after 14 days of differentiation. The temporal and spatial pattern of synaptic maturation is comparable to synaptogenesis of hippocampal neurons grown in primary culture. Therefore, with respect to the general ability to create mature synaptic contacts, NSCs seem to be well equipped to potentially compensate for lost or injured brain tissue. STEM CELLS 2007;25:1720 –1729

Disclosure of potential conflicts of interest is found at the end of this article.

INTRODUCTION NSCs are self-renewing cells that maintain the capacity to differentiate into brain-specific cell types and may also replace or repair diseased brain tissue [1]. It has been consistently demonstrated by several groups [1] that NSCs have the potential to generate mature cells of all neural lineages, including astroglia, oligodendroglia, and neurons. It has already been shown that hippocampus-derived NSCs differentiate into excitatory and inhibitory neurons expressing the appropriate neurotransmitters [2] and/or form functional synapses with cocultured neurons or within the mouse brain [3, 4]. Nevertheless, it is still unclear how these synaptic contacts emerge, especially with respect to the expression and targeting of key molecules of the pre- and postsynaptic specializations. Moreover, the exact pattern of steps finally leading to the formation of structurally mature excitatory synapses still needs to be elucidated. Chemical synapses are highly specialized cell-cell contacts that mediate efficient communication between nerve cells. Ultrastructurally, especially in glutamatergic synapses, distinct pre- and postsynaptic specializations mark the sites of neurotransmitter release and reception [5, 6]. Presynaptically, the regulated release of neurotransmitter-filled vesicles into the synaptic cleft is restricted to the active zone, which is characterized by a cy-

toskeletal meshwork. This cytomatrix at the active zone (CAZ) is exactly aligned with an electron-dense cytoskeletal matrix underneath the postsynaptic membrane known as the postsynaptic density (PSD) [7, 8]. The mature CAZ is defined by a set of multidomain proteins that harbor several protein-protein or protein-lipid interaction domains [9]. The complete protein composition of the CAZ remains elusive, but it includes the proteins Munc13-1 [10], RIMs [11, 12], ERC/CAST [13, 14], Piccolo/Aczonin, and Bassoon [15–17]. The PSD is defined by the accumulation of several hundred different proteins, including membrane-bound receptors, cell adhesion proteins, proteins of several signaling cascades, and cytoskeletal components [18]. Moreover, the PSD is characterized by specific scaffolding molecules of the MaGuK family, including SAP90/PSD-95 [19], that directly interact with and cluster, for example, Nmethyl-D-aspartate (NMDA) receptors. In deeper layers of the PSD, master scaffolding proteins of the ProSAP/Shank family are localized [20, 21]. ProSAP/Shank proteins are multidomain proteins and are able to mediate a physical connection between glutamate receptor complexes or components of several signaling cascades at the postsynaptic membrane and the actin-based cytoskeleton [22]. ProSAP1/Shank2 and, especially, ProSAP2/Shank3 are early components of the maturing synapse and have been found to be involved in the shaping and remodeling of synaptic con-

Correspondence: Tobias M. Boeckers, M.D., Ulm University, Institute of Anatomy and Cell Biology, Albert Einstein Allee 11, 89081 Ulm, Germany. Telephone: 49-731-5023220; Fax: 49-731-5023217; e-mail: [email protected] Received January 8, 2007; accepted for publication March 14, 2007; first published online in STEM CELLS EXPRESS March 22, 2007. ©AlphaMed Press 1066-5099/2007/$30.00/0 doi: 10.1634/stemcells.2006-0823

STEM CELLS 2007;25:1720 –1729 www.StemCells.com

Liebau, Vaida, Storch et al.

1721

tacts, which is thought to be the morphological basis for synaptic plasticity [20, 23]. In addition, recent data indicate that ProSAP2/Shank3 can associate via a C-terminal SAM domain to build huge platforms within the PSD [24], which seem to be a core component of the postsynaptic specialization. In this respect, the recent detection of well-defined repetitive fiber-like structures within the PSD has to be reconsidered [25]. Synaptogenesis requires the specific assembly of protein clusters at both sides of the synaptic contact by mechanisms that are barely understood. In particular, the exact events that induce the formation of a synapse and the pattern of localization with respect to targeting and recruitment of pre- and postsynaptic molecules have not yet been completely resolved [26]. There is, however, increasing evidence that the general mechanisms of protein targeting and assembly during synaptogenesis are different at the presynaptic compared with the postsynaptic microcompartments [27, 28]. It is believed that the presynaptic specialization is mainly built by preformed dense core vesicles that are transported along the axonal processes and fuse with the presynaptic membrane during synaptogenesis. These dense core vesicles are also called Piccolo transport vesicles, since they can be stained and tracked within the neuron with antibodies directed against the huge CAZ protein Piccolo. In contrast, molecules of the PSD seem to be individually targeted or recruited to synaptic sites, where they integrate into a macromolecular complex [27]. Recently, however, it has been demonstrated that a trimeric complex including Shank1, Neuroligin, and GKAP could be a prebuilt dynamic postsynaptic precursor complex [29]. Here, we studied the synaptic targeting of the described prominent multidomain proteins of both the presynapse and the PSD, namely Bassoon, ProSAP1/Shank2, ProSAP2/Shank3, and synaptophysin. We describe the differentiation of clonogenic fetal NSCs into functional neurons showing a time-dependent course of synaptic protein expres-

sion, transport, and localization that is comparable to synaptogenesis observed in primary hippocampal neurons [30, 31].

MATERIALS

AND

METHODS

Expansion of Mesencephalic NSCs Mesencephalic NSCs from rat embryonic brain were prepared as previously described [32, 33]. In brief, ventral mesencephalic tissue was harvested from embryonic day 14.5 rat embryos. For expansion of neurospheres, tissue samples were incubated in 0.1% trypsin (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) for 10 minutes at room temperature, incubated in DNase I (40 mg/ml; Sigma-Aldrich) for 10 minutes at 37°C, and homogenized to a quasi-single-cell suspension by gentle triturating. The cells were added to 25-cm2 flasks (2–3 ⫻ 106 viable cells per flask) in serum-free medium containing 63% Dulbecco’s modified Eagle’s medium (DMEM) high glucose, 32% Ham’s F-12 medium, 1% glutamate, 2% B27 supplement, 1% penicillin/streptomycin, and 1% nonessential amino acids (all from Gibco, Tulsa, OK, http:// www.invitrogen.com) supplemented with 20 ng/ml of the mitogen epidermal growth factor (Sigma-Aldrich). Cultures were placed in a humidified incubator at 37°C and 5% CO2, 95% air (21% O2). After 4 –7 days, sphere formation was observed. The medium was changed once a week, and growth factor was added twice a week. The neurospheres were expanded for an additional 3–10 weeks in suspension. As shown previously, these progenitor cells can be expanded as clonal cultures that maintain the capacity to differentiate into neurons, astroglia, and oligodendroglia, confirming their pluripotent stem cell fate [32].

Differentiation of Mesencephalic NSCs NSCs were differentiated by plating them onto poly-L-lysine-coated 24-well plates at 50,000 or 150,000 cells per cm2 in medium consisting of a DMEM (high glucose)/Ham’s F-12 medium mixture (1:1), 1% glutamate, 2% B27 supplement, 100 units of penicillin/

Table 1. Primers for quantitative real-time reverse transcription-polymerase chain reaction Gene (protein)

Primer sequence (forward, reverse)

Product length (bp)

ProSAP1

5⬘-GGT CGC CCT TCA CTC CTG-3⬘ 5⬘-GCC GAT GCT CAG AAC TTT G-3⬘

185

ProSAP2

5⬘-CGC TCA ACT ACG GGC TAT TC-3⬘ 5⬘-CGA AAC TCC AGA TAG GGC AG-3⬘

124

BSN (Bassoon)

Rn_Bsn_1_SG QuantiTect primer assay (NM_019146) QT00392350

118

SYP (synaptophysin)

5⬘-GCT GAG CGT GGA GTG TGC-3⬘ 5⬘-GAC GAG GAG TAG TCC CCA AC-3⬘

156

HMBS (hydroxymethylbilane synthase)

5⬘-CGA CAC TGT GGT AGC GAT GC-3⬘ 5⬘-CCT TGG TAA ACA GGC TCT TCT CTC-3⬘

134

a

The x-axis represents temperature (from 68°C to 92°C in 4°C steps), and the y-axis indicates fluorescence ⫺d(F1)dT.

www.StemCells.com

Melting point analysisa

1722

Synaptogenesis from Neural Stem Cells

Figure 2. Time course of mRNA levels of pre- and postsynaptic proteins during NSC differentiation. Semiquantitative real-time one-step reverse transcription-polymerase chain reaction with total RNA extracted from NSCs during the expansion phase (neurospheres) and at different time points of differentiation (days 1–21). Expression levels are shown relative to the housekeeping gene HMBS (mean ⫾ SEM; n ⫽ 4). Primers and melting curve analysis are given in Table 1. All mRNAs investigated showed increasing expression levels within the first week of differentiation. ProSAP1/Shank2 mRNA levels showed the highest increase at days 8 and 12 compared with the other synaptic proteins. ProSAP2/Shank3 mRNA levels were already relatively high in the neurospheres, peaked in the first week of differentiation, and dropped sharply down in the second week. Abbreviations: BSN, Bassoon; HMBS, hydroxymethylbilane synthase; SYP, synaptophysin.

pass [BP] 450 – 490, FT 510; emission, BP 515–565), Alexa Fluor 568 (red; filter set, excitation, BP 534 –558 nm, FT 560; emission, BP 575– 640), and Alexa Fluor 647 (magenta; filter set, excitation, BP 610 – 670 nm, Fourier transformation [FT] 660; emission, BP 640 –740) (Invitrogen), all diluted 1:500. Images were captured using an upright fluorescence microscope (Axioskop 2; Carl Zeiss, Oberkochen, Germany, http://www.zeiss.com) equipped with a CCD camera (16 bits; 1,280 ⫻ 1,024 pixels per image) and analyzed using Axiovision software (Carl Zeiss).

Figure 1. Time course of differentiation of NSCs. (A): Undifferentiated adherent neural progenitor cell expressing the stem cell marker nestin (NES, red), an intermediate filament protein that is mainly expressed in neuroepithelial stem cells. The nucleus is counterstained with DAPI. (B–G): Morphologies of developing neurons at different stages of differentiation. Note that the cells expressed ␤-tubulin III (Tuj1, green), an early, neuron-specific protein of the cytoskeleton. From d 2 to d 5, most emerging neurons showed a bipolar morphology. The outgrowth of several neurites can be observed from d 5 onward, leading to a complex dendritic tree after 2 weeks of differentiation. The nuclei were counterstained with DAPI (blue). Scale bars ⫽ 10 ␮m. Abbreviations: d, day; DAPI, 4,6-diamidino-2-phenylindole; NES, nestin.

100 mg streptomycin per milliliter, and 1% nonessential amino acids. To induce differentiation, medium was supplemented with 5% serum replacement (Gibco) and by plating cells on poly-L-lysine-treated coverslips. The cells were allowed to differentiate for up to 35 days at 37°C in a humidified atmosphere of 5% CO2 in air (21% oxygen). Serum replacement was removed on the second day by replacing medium. Every third day, half of the medium was changed.

Immunocytochemistry Cells were fixed at different time points after differentiation using 4% paraformaldehyde in phosphate-buffered saline. Immunocytochemistry was carried out using standard protocols. Cell nuclei were counterstained with 4,6-diamidino-2-phenylindole (DAPI). Antibodies and dilutions were as follows: primary antibodies, ␤-Tubulin III (Tuj1), chicken, 1:1,000 (Invitrogen); Nestin, mouse, 1:500 (BD Biosciences, San Diego, http://www.bdbiosciences.com); synaptophysin, mouse, 1:800 (Sigma-Aldrich); Bassoon, mouse (Stressgen, Ann Arbor, MI, http://www.stressgenbioreagents.com); ProSAP1/ Shank2, ProSAP2/Shank3 [20, 23]. Fluorescence-labeled secondary antibodies were Alexa Fluor 488 (green; filter set, excitation, band

Semiquantitative Real-Time Polymerase Chain Reaction Total cellular RNA was extracted from NSCs and 14-day differentiated NSCs using the RNeasy total RNA purification kit followed by treatment with RNase-free DNase (Qiagen, Hilden, Germany, http://www1.qiagen.com). Quantitative real-time onestep reverse transcription-polymerase chain reaction (RT-PCR) was carried out using the LightCycler system (Roche, Mannheim, Germany, http://www.roche.com), and amplification was monitored and analyzed by measuring the binding of the fluorescence dye SYBR Green I to double-stranded DNA. One microliter (50 ng) of total RNA was reverse-transcribed and subsequently amplified using Quanti-Tect SYBR Green RT-PCR Master mix (Qiagen) and a 0.5 ␮mol/l concentration of both sense and antisense primers. Tenfold dilutions of total RNA were used as external standards. Standards and samples were simultaneously amplified. After amplification, melting curves of the RT-PCR products were acquired to demonstrate product specificity. The results are expressed relative to the housekeeping gene hydroxymethylbilane synthase (HMBS). Primer sequences, lengths of the amplified products, and melting point analyses are summarized in Table 1. Both primers for Bassoon were purchased as a ready-to-use validated primer pair (Qiagen). Data are presented as mean ⫾ SEM of at least three independent experiments, and significant differences were evaluated by the analysis of variance (ANOVA) model for multigroup comparisons (using the StatView 5 [Cary, NC, http://www.statview.com] software package).

Transmission Electron Microscopy Cells were fixed in 0.1 M phosphate buffer, pH 7.3, containing 2.5% glutaraldehyde and 1% sucrose and were osmicated for 1 hour in 2% OsO4. Then, they were dehydrated in graded series of ethanol, contrasted in 2% uranyl acetate, and embedded in epoxy


1723

Figure 3. Time course of protein expression and localization of the presynaptic marker protein SYP in developing mesencephalic NSCs. (A): Already, at d 5, the antibody detected SYP-positive regions in a distinct perinuclear region. (B, C): Two to 3 days later, the antigen localized within a single neurite. At this time point, SYP staining identified the axon of the developing neuron. (D–F): Interestingly, at later stages, SYP was targeted to small axonal protrusions that can be observed as filopodia-like structures. (G–I): From d 12 onward, the protein was transported and/or localized in small vesicle-like structures that were finally seen along the entire axon. This pattern of SYP expression and localization was seen in all NSCs investigated and seems to be independent of cell-to-cell contacts but solely dependent upon age and stage of differentiation. In all panels, SYP staining is shown in green. In (C, F, G–I), the cells were counterstained with Tuj1 (red). Nuclei were labeled with DAPI (blue). Scale bars ⫽ 10 ␮m. Also shown is the time course of expression and localization of the cytomatrix active zone protein BSN in differentiating NSCs. (J, K): At early stages of development, BSN was localized as clusters in the perinuclear area. (L–P): From d 4 onward, relatively large vesicles could be detected in only one neurite of the developing neuron, most likely representing the axon. The BSN-positive vesicles are known as active zone precursor vesicles (dense core vesicles), and they contain a preformed complex of several presynaptic proteins. The number of these BSN-positive vesicles increased within the first week according to the stage of differentiation and length of the axon. (Q, R): Finally, more irregularly shaped presynaptic specializations were seen at d 10 and d 12, when synaptic contacts were built in larger numbers. BSN staining is shown in red; double staining was performed with Tuj1 antibodies (green). Nuclei were counterstained with DAPI (blue). Scale bars ⫽ 10 ␮m. Abbreviations: BSN, Bassoon; d, day; DAPI, 4,6-diamidino-2-phenylindole; SYP, synaptophysin.

resin (Sigma-Aldrich) at 60°C. Thin sections of 70 – 80 nm were cut with a diamond knife on a Reichert (Nussloch, Germany, http://www.leica-microsystems.com) ultramicrotome and collected on 300-mesh grids. The sections were contrasted with 0.3% lead citrate for 1 minute and analyzed on an EM 10 transmission electron microscope (Carl Zeiss) at 80 kV.

RESULTS Time Course of Neuronal Maturation of Rat Mesencephalic NSCs For our studies on synaptogenesis from NSCs, we used the previously characterized multipotent midbrain-derived rat NSCs [32, 33], expressing several NSC marker proteins [34 –36], including the intermediate filament protein nestin [37, 38] (Fig. 1A). As reported earlier, for rodent and human www.StemCells.com

mesencephalic precursors [32, 33], after removal of growth factors or mitogens, NSCs differentiate spontaneously into nerve cells with extensive arborizations (Fig. 1B–1G). From day 2 after differentiation, a positive staining for the neuronal marker Tuj1 could be observed (Fig. 1B–1G). Over a time period of approximately 14 days, NSCs were then able to differentiate into fully developed neurons. They became positive for MAP2ab and even tyrosine hydroxylase, GABA, or glutamate (data not shown) [32–34]. The growth in size of the individual neuron, as well as the formation of dendrites, occurred in a gradual manner (Fig. 1) and was quite comparable to the differentiation steps of cultured primary hippocampal neurons [39, 40]. With respect to morphological aspects, differentiated NSCs maturate into different types of neurons (e.g., bipolar and multipolar), whereas primary hippocampal cultures mostly consist of multipolar excitatory pyramidal neurons.

1724


Figure 4. Developmental maturation of synaptic contacts as revealed by electron microscopy. At early stages of development, contacts between axo-axonal (A) or axo-dendritic compartments (B) could be observed. Arrowheads indicate the presence of dense core vesicles that might represent CAZ precursor vesicles. (C, D): At d 11 and d 12, axo-dendritic contacts were often observed; these contacts were characterized by several clear vesicles at the presynaptic compartment (and some dense core vesicles). At this time point, the postsynaptic membrane was seldom thickened. (E): At d 15, excitatory synapses could be observed that displayed a typical polarized morphology with a readily detectable active zone (arrow) and a postsynaptic density (arrowhead). Scale bars ⫽ 0.5 ␮m. Abbreviation: d, day.

Time Course of mRNA Expression of Pre- and Postsynaptic Proteins During NSC Differentiation The mRNAs encoding the presynaptic proteins synaptophysin (SYP), Bassoon (BSN), and the postsynaptic proteins ProSAP1 and ProSAP2 could already be detected at low levels in proliferating NSCs within the neurospheres. They showed significantly increasing expression levels within the first week of differentiation (Fig. 2). Interestingly, however, over a period of 3 weeks, all four transcripts displayed individual patterns of expression. SYP mRNA levels showed a continuous increase of expression, with the highest level at day 19 after initiation of differentiation (F value ⫽ 115.2; p ⬍ .0001; ANOVA), whereas BSN mRNA showed a fast increase within the first week of differentiation, with peak levels at day 8 (F value ⫽ 5.0; p ⫽

.035; Fig. 2). ProSAP1/Shank2 mRNA levels were quite stable over the observation period (F value ⫽ 1.3; p ⬍ .26). However, a steep increase could be detected on day 12. In contrast, ProSAP2/SHANK3 mRNA levels sharply increased within the first week, followed by a 10-fold decrease of mRNA amount (F value ⫽ 10.2; p ⫽ .004).

Development of Synaptic Structures in NSC Cultures Presynaptic Development. After day 5 of differentiation (Fig. 3A), synaptophysin was found juxtaposed to the nucleus, indicating processing of the protein within the rough endoplasmic reticulum/Golgi complex. Two days later, vesicular synaptophy-


1725

Figure 5. Time course of protein expression and localization of the postsynaptic protein ProSAP1/Shank2 during development of mesencephalic NSCs. Stainings with antibodies directed against BSN/SYP (green) and ProSAP1/Shank2 (red) resulted in the localization of the proteins in already clearly defined axonal and dendritic compartments of the neuron. (A, B): BSN-positive transport vesicles, as well as SYP, were found in the axons, whereas ProSAP1/Shank2 protein was targeted to dendrites of the neurons. (C): After d 14, ProSAP1/Shank2 was found in clusters, and dendrites were decorated with several ProSAP1/Shank2-positive postsynaptic complexes. (D): In contrast, the postsynaptic scaffolding molecule ProSAP2/ Shank3 was still more homogenously distributed within the cytoplasm and neurites of differentiating neurons at d 11. (E): Three days later, ProSAP2/Shank3 started to be clustered opposite SYP-positive presynaptic structures. (F): At d 21, ProSAP2/Shank3 clustered within the dendritic compartment. Double staining of Tuj1 (red) and ProSAP2/Shank3 (green) shows neurons with several postsynaptic specializations as revealed by ProSAP2/Shank3 antibodies. Nuclei were counterstained with DAPI (blue). Scale bars ⫽ 10 ␮m. Abbreviations: BSN, Bassoon; d, day; DAPI, 4,6-diamidino-2-phenylindole; SYP, synaptophysin.

sin-positive structures could easily be detected along the early presumed axon (Fig. 3B). As the axon was elongated rapidly, synaptophysin was expressed in high concentrations and was transported along the axonal protrusion (Fig. 3C). Even if the cells did not have contact with other neurons at that time, synaptophysin filled these axonal structures more and more over the following days (Fig. 3D). After day 10, synaptophysin was already abundant in early filopodia-like structures (Fig. 3E), and first contacts with other neurons led to a specific localization of synaptophysin in the presynaptic compartment (Fig. 3F). During the following days, numerous synaptic contacts were established (Fig. 3G), until the cell was covered with synaptophysin-positive cell-cell contacts (Fig. 3H). Networks of neurons with www.StemCells.com

synaptophysin-positive cell-cell contacts were readily observed from day 12 onward (Fig. 3I). The presynaptic CAZ protein Bassoon (Fig. 3J–3R) showed a comparable timeframe of synaptic targeting. As has already been shown for hippocampal cultures [26], Bassoon can be detected at early developmental stages in large vesicles (presumably dense core vesicles; Fig. 4). Similar to synaptophysin, the translation and transport of Bassoon seems to be independent of intercellular contacts. Postsynaptic Development. In accordance with the data we obtained at early stages of NSC differentiation with respect to presynaptic marker molecules, both postsynaptic scaffolding molecules, ProSAP1/SHANK2 and ProSAP2/SHANK3, were also cytoplasmic until day 7. At later stages, from day 9 onward,

1726


Figure 6. Combined expression of the pre- and postsynaptic protein SYP/BSN and ProSAP1/2 within the developing synapse. NSCs were stained after 14, 17, and 23 days of differentiation for the synaptic marker proteins; Tuj1 was used as a neuronal marker. (A, C): Stainings of the presynaptic marker proteins SYP/BSN, together with the postsynaptic protein ProSAP1/Shank2, clearly demonstrated the colocalization of both types of proteins within synaptic contacts. (B, D): At d 23 of differentiation, the postsynaptic molecule ProSAP2/Shank3 was also targeted to synaptic contacts, as revealed by costainings with BSN or SYP antibodies. Nuclei were counterstained with DAPI (blue). Scale bars ⫽ 10 ␮m. Abbreviations: BSN, Bassoon; d, day; DAPI, 4,6-diamidino-2-phenylindole; SYP, synaptophysin.

they were found to be highly enriched in the dendritic compartment and localized to synaptic structures (Fig. 5). Between day 14 and day 21 (Fig. 6), triple stainings clearly indicated that the pre- and postsynaptic molecules investigated were always colocalizing at synaptic cell-cell contacts. At day 23, we often found differentiated NSCs that perfectly mirrored hippocampal pyramidal neurons in culture (Fig. 6D) [26, 30].

Ultrastructure of Developing Synapses Ultrastructural investigations using electron microscopy of cellcell contact development during neuronal differentiation revealed that outgrowing processes of NSCs were found in close proximity from day 4 onward. In early stages, many contacts were seen, characterized by clear vesicles on both sites of the contact site, quite similar to axo-axonal synapses (Fig. 4A). Moreover, the dense core vesicles, most likely representing CAZ precursors [9, 41, 42], were already detected, intermingled within the vesicle pool. Often, they were located close to the membrane. The number of dense core vesicles steadily increased, and the highest numbers were found between day 10 and day 11 (Fig. 4B, 4C). Clearly polarized synaptic contacts between axonal and dendritic neuronal specializations were observed from day 12 onward, when densely packed clear vesicles were found directly at the presynaptic membrane facing toward a synaptic cleft and a slightly thickened postsynaptic membrane (Fig. 4D). Morphologically mature synapses could be observed after 14 days of differentiation. Here, all morphological features of excitatory synapses were developed with docked and partially fused vesicles at the presynapse and a

submembranous, electron-dense web (PSD) underneath the postsynaptic membrane (Fig. 4E).

DISCUSSION NSCs existing in the developing and adult brain retain the capacity to self-renew and to produce the major cell types of the brain. This opens new avenues for restorative therapy of neuropsychiatric disorders. Isolated from fetal central nervous system, these cells can also be maintained in vitro and retain the potential to differentiate into nervous tissue. When allowed to differentiate spontaneously, after removal of growth factors or mitogens in serum-free media [43, 44], NSCs generate oligodendrocytes, neurons, and astrocytes in a ratio of approximately 1:5:25, respectively. Several studies have shown that neurons derived in vitro from expanded stem cells are to a large degree GABAergic [45]; there are, however, numerous neurons that are immunopositive for other neurotransmitters, such as dopamine, glutamate, serotonin, or acetylcholine. Under certain conditions and with the addition of specific extrinsic factors, the ratio of neurons using different neurotransmitters can be altered [32, 46]. As primarily shown in primary cultured hippocampal neurons, synapses are known to be initially established by the attachment of distinct areas of an axonal and a dendritic membrane (initial contact) via certain cell adhesion molecules, including molecules of the integrin and cadherin families [26]. This step is followed by an inductive step, finally leading to several waves of different proteins being added and incorpo-

Liebau, Vaida, Storch et al. rated to the pre- and postsynaptic complex. The molecular setup of such an elaborated contact zone determines the morphology and function of the newly built neuronal cell-cell contact. In our study, we investigated the generation of specific cell-cell-contacts in differentiating mesencephalic NSCs. To that end, we stained our cultured cells at different time points, with several combinations of primary antibodies directed to synaptic proteins, which have already been shown to play a pivotal role in the stability, plasticity, and function of neuronal synapses. Synaptophysin is an integral Ca2⫹-binding synaptic vesicle membrane protein [47, 48], highly abundant in synapses of the vertebrate brain [49], being present in virtually all synaptic terminals. Synaptophysin, which is often used as a general marker protein of presynaptic nerve endings [50], interacts with other proteins involved in exocytosis and is required for controlled vesicle fusion and neurotransmitter release [51]. Bassoon is among the first CAZ proteins to appear at nascent synapses in cultured neurons and is a component of the active zone precursor vesicles [41, 52]. It is a very large multidomain protein that is intimately anchored to the cortical actin/spectrin cytoskeleton and is present at both excitatory and inhibitory synapses in the brain [17, 42, 53, 54]. The early appearance at synapses and the molecular scaffolding properties make Bassoon a prime candidate for the assembly of functional active zones. ProSAPs/Shanks, on the other hand, are postsynaptic multidomain proteins that interact directly and indirectly with a large number of synaptic proteins [22, 55]. In particular, ProSAPs/Shanks interact with NMDA-type glutamate receptors (via GKAP/SAPAP and SAP90/PSD-95) [20, 21, 23] and with metabotrophic glutamate receptors (via Homer/VESL) [56]. In addition, ProSAPs/Shanks are very early components of postsynaptic specializations and interact with the actin-based cytoskeleton via cortactin [57] and ␣-fodrin [20]. Thus, ProSAPs/Shanks have been considered to constitute “master scaffold” PSD molecules that are able to anchor and cluster different types of glutamate receptor complexes to cytoskeletal elements [22, 55], for example, during the maturation of synaptic contacts. In our study, we focused on the recruitment of synaptophysin/Bassoon to the CAZ and ProSAP1 (Shank2) and ProSAP2 (Shank3) to the PSD of nascent synapses built by NSCs. As a central finding, we could demonstrate that not only are differentiating NSCs able to give rise to neurons but these neurons are capable of creating synaptic contact sites with respect to morphology and the expression and localization of synaptic key molecules. This ability was preserved even after long-term proliferation of NSCs in vitro. Interestingly, the time frame of synaptic maturation was quite comparable to that of cultured primary hippocampal neurons [26]. Therefore, one might assume that once the development of NSCs is committed toward a neuronal phenotype, the “program” of synaptogenesis is initiated, leading to the expression and localization of synaptic proteins at membranous attachment sites, which maturate into synapses after 14 days. These synaptic contacts are indistinguishable from those created by hippocampal neurons [30]. With respect to mRNA levels, the upregulation of the synaptic proteins during the first days of differentiation, confirms these data. ProSAP2/ Shank3, however, displays a slightly different expression pattern, with relatively high levels of mRNA already in expanding cells. This might point to a functional role of this molecule not only in neurons but also in other cells derived from different ectodermal precursors [22]. Therefore, it might well be that ProSAP2/Shank3 also organizes subcellular compartments within undifferentiated NSCs. In line with several investigations providing evidence that the axonal specialization is easily to be defined by the selective localization of presynaptic proteins [28, 52], we found www.StemCells.com

1727

synaptophysin, as well as Bassoon, in early axonal structures from day 7 onward. At early developmental stages, synaptophysin could be detected homogenously in axons and was found to be localized in tiny axonal protrusions. This distribution pattern could well be explained by groups of larger clear vesicles, which were often found on the ultrastructural level at identical time points. Later (from day 11 onward), immunoreactivity was concentrated in dot-like structures, which could be identified as presynaptic specializations. Bassoon, which seems to be concentrated in the Golgi apparatus at very early stages, was always found to be clustered. Initially, these relatively large clusters could be identified in the axon (day 6 – 8), most likely reflecting the described axonal active zone transport vesicles [52, 58]. Accordingly, the analysis of the NSC cultures by electron microscopy clearly showed the large number of dense core vesicles in the axonal/presynaptic compartment at identical time points (Fig. 4). For the extensively investigated presynaptic marker molecule synaptophysin, as well as for Bassoon, a similar time frame of mRNA/protein expressions was found by several groups for embryonic stem cell-derived neurons and neurons from hippocampal precursors [2, 59 – 62]. At the postsynaptic site, a more homogenous staining of dendrites could be observed and clustering of ProSAP/Shank molecules was found, especially at the end of the second week of differentiation. At day 14, when mature synaptic contacts were detected at the ultrastructural level, pre- and postsynaptic molecules aligned perfectly when double and/or triple immunocytochemical staining methods were used. The number of synapses per neuron was comparable to that in hippocampal neurons [30]. The ultrastructural analyses showed that the neuronal contact sites, after 14 days of differentiation, displayed all features of functional excitatory synapses, with groups of clear vesicles localized at or attached to the presynaptic membrane, a synaptic cleft, and a juxtaposed postsynaptic specialization (PSD) extending in register to the presynaptic vesicle fusion site. Taken together, the data give clear evidence that NSCs have the potential to create neuronal networks, including contacts that display important molecular and morphological features of functional synapses.

SUMMARY We demonstrate here that NSCs are able to build cell-to-cell contacts with all major morphological components of synapses, similar to primary hippocampal neurons. Our study was further aimed at investigating whether key proteins of the presynaptic compartment and postsynaptic density are expressed in differentiating neural stem cells and are transported to maturing synapses in a time-dependent manner. Moreover, we used electron microscopy to characterize neuronal contact sites. Here, we show that rat mesencephalic fetal NSCs are able to express and transport synaptic components specifically into the axonal and dendritic compartment, finally leading to synaptic contacts displaying all molecular and morphological features of functional synapses. The time frame of synaptic targeting and the mechanism of transport were quite similar to those of hippocampal neurons in culture. These data indicate that NSCs are readily able to differentiate into neurons that are able to create a neuronal network by interconnecting via mature synaptic contacts during the first 2 weeks of the differentiation.


1728

ACKNOWLEDGMENTS We thank Nicola Martin for excellent technical assistance and Eberhard Schmid from the Central Electron Microscopy Unit, University of Ulm, for help with embedding and processing. This study was supported by Ulm University (to S.L.), by the Land Baden-Wu¨rttemberg (1423/74 to S.L. and

T.M.B.), and by the Deutsche Forschungsgemeinschaft (SFB497/B8 to T.M.B.).

DISCLOSURE

25

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

21 22

23

24

Storch A, Schwarz J. Neural stem cells and neurodegeneration. Curr Opin Investig Drugs 2002;3:774 –781. Vicario-Abejo´n C, Collin C, Tsoulfas P et al. Hippocampal stem cells differentiate into excitatory and inhibitory neurons. Eur J Neurosci 2000;12:677– 688. Ge S, Goh EL, Sailor KA et al. GABA regulates synaptic integration of newly generated neurons in the adult brain. Nature 2006;439:589 –593. Toda H, Takahashi J, Mizoguchi A et al. Neurons generated from adult rat hippocampal stem cells form functional glutamatergic and GABAergic synapses in vitro. Exp Neurol 2000;165:66 –76. Kim E, Ko J. Molecular organization and assembly of the postsynaptic density of excitatory brain synapses. Results Probl Cell Differ 2006;43: 1–23. Wenzel J, Bogolepov NN. [Electronmicroscopical and morphometrical study of rat hippocampal synapses]. J Hirnforsch 1976;17:399 – 448. Dresbach T, Qualmann B, Kessels MM et al. The presynaptic cytomatrix of brain synapses. Cell Mol Life Sci 2001;58:94 –116. Ziff EB. Enlightening the postsynaptic density. Neuron 1997;19: 1163–1174. Schoch S, Gundelfinger ED. Molecular organization of the presynaptic active zone. Cell Tissue Res 2006;326:379 –391. Brose N, Hofmann K, Hata Y et al. Mammalian homologues of Caenorhabditis elegans unc-13 gene define novel family of C2-domain proteins. J Biol Chem 1995;270:25273–25280. Wang Y, Okamoto M, Schmitz F et al. Rim is a putative Rab3 effector in regulating synaptic-vesicle fusion. Nature 1997;388:593–598. Wang Y, Sugita S, Sudhof TC. The RIM/NIM family of neuronal C2 domain proteins. Interactions with Rab3 and a new class of Src homology 3 domain proteins. J Biol Chem 2000;275:20033–20044. Ohtsuka T, Takao-Rikitsu E, Inoue E et al. Cast: A novel protein of the cytomatrix at the active zone of synapses that forms a ternary complex with RIM1 and munc13–1. J Cell Biol 2002;158:577–590. Wang Y, Liu X, Biederer T et al. A family of RIM-binding proteins regulated by alternative splicing: Implications for the genesis of synaptic active zones. Proc Natl Acad Sci U S A 2002;99:14464 –14469. Cases-Langhoff C, Voss B, Garner AM et al. Piccolo, a novel 420 kDa protein associated with the presynaptic cytomatrix. Eur J Cell Biol 1996;69:214 –223. tom Dieck S, Sanmarti-Vila L, Langnaese K et al. Bassoon, a novel zincfinger CAG/glutamine-repeat protein selectively localized at the active zone of presynaptic nerve terminals. J Cell Biol 1998;142:499–509. Wang X, Kibschull M, Laue MM et al. Aczonin, a 550-kD putative scaffolding protein of presynaptic active zones, shares homology regions with Rim and Bassoon and binds profilin. J Cell Biol 1999;147:151–162. Boeckers TM. The postsynaptic density. Cell Tissue Res 2006;326: 409 – 422. Sampedro MN, Bussineau CM, Cotman CW. Postsynaptic density antigens: Preparation and characterization of an antiserum against postsynaptic densities. J Cell Biol 1981;90:675– 686. Boeckers TM, Kreutz MR, Winter C et al. Proline-rich synapse-associated protein-1/cortactin binding protein 1 (ProSAP1/CortBP1) is a PDZdomain protein highly enriched in the postsynaptic density. J Neurosci 1999;19:6506 – 6518. Naisbitt S, Kim E, Tu JC et al. Shank, a novel family of postsynaptic density proteins that binds to the NMDA receptor/PSD-95/GKAP complex and cortactin. Neuron 1999;23:569 –582. Boeckers TM, Bockmann J, Kreutz MR et al. ProSAP/Shank proteins: A family of higher order organizing molecules of the postsynaptic density with an emerging role in human neurological disease. J Neurochem 2002;81:903–910. Boeckers TM, Winter C, Smalla KH et al. Proline-rich synapse-associated proteins ProSAP1 and ProSAP2 interact with synaptic proteins of the SAPAP/GKAP family. Biochem Biophys Res Commun 1999;264: 247–252. Baron MK, Boeckers TM, Vaida B et al. An architectural framework

CONFLICTS

The authors indicate no potential conflicts of interest.

REFERENCES 1

OF POTENTIAL OF INTEREST

26 27 28 29 30 31 32

33 34 35 36 37 38 39 40 41 42 43 44 45 46

47 48 49 50

that may lie at the core of the postsynaptic density. Science 2006; 311:531–535. Zuber B, Nikonenko I, Klauser P et al. The mammalian central nervous synaptic cleft contains a high density of periodically organized complexes. Proc Natl Acad Sci U S A 2005;102:19192–19197. Garner CC, Waites CL, Ziv NE. Synapse development: Still looking for the forest, still lost in the trees. Cell Tissue Res 2006;326:249 –262. Bresler T, Shapira M, Boeckers T et al. Postsynaptic density assembly is fundamentally different from presynaptic active zone assembly. J Neurosci 2004;24:1507–1520. Ziv NE, Garner CC. Cellular and molecular mechanisms of presynaptic assembly. Nat Rev Neurosci 2004;5:385–399. Gerrow K, Romorini S, Nabi SM et al. A preformed complex of postsynaptic proteins is involved in excitatory synapse development. Neuron 2006;49:547–562. Verderio C, Coco S, Pravettoni E et al. Synaptogenesis in hippocampal cultures. Cell Mol Life Sci 1999;55:1448 –1462. Boeckers TM, Liedtke T, Spilker C et al. C-terminal synaptic targeting elements for postsynaptic density proteins ProSAP1/Shank2 and ProSAP2/Shank3. J Neurochem 2005;92:519 –524. Carvey PM, Ling ZD, Sortwell CE et al. A clonal line of mesencephalic progenitor cells converted to dopamine neurons by hematopoietic cytokines: A source of cells for transplantation in Parkinson’s disease. Exp Neurol 2001;171:98 –108. Storch A, Lester HA, Boehm BO et al. Functional characterization of dopaminergic neurons derived from rodent mesencephalic progenitor cells. J Chem Neuroanat 2003;26:133–142. Ling ZD, Potter ED, Lipton JW et al. Differentiation of mesencephalic progenitor cells into dopaminergic neurons by cytokines. Exp Neurol 1998;149:411– 423. Milosevic J, Storch A, Schwarz J. Spontaneous apoptosis in murine free-floating neurospheres. Exp Cell Res 2004;294:9 –17. Storch A, Paul G, Csete M et al. Long-term proliferation and dopaminergic differentiation of human mesencephalic neural precursor cells. Exp Neurol 2001;170:317–325. Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell 1990;60:585–595. Gage FH. Mammalian neural stem cells. Science 2000;287:1433–1438. Bradke F, Dotti CG. Establishment of neuronal polarity: Lessons from cultured hippocampal neurons. Curr Opin Neurobiol 2000;10:574 –581. Dotti CG, Sullivan CA, Banker GA. The establishment of polarity by hippocampal neurons in culture. J Neurosci 1988;8:1454 –1468. Shapira M, Zhai RG, Dresbach T et al. Unitary assembly of presynaptic active zones from Piccolo-Bassoon transport vesicles. Neuron 2003;38: 237–252. Richter K, Langnaese K, Kreutz MR et al. Presynaptic cytomatrix protein bassoon is localized at both excitatory and inhibitory synapses of rat brain. J Comp Neurol 1999;408:437– 448. Gritti A, Parati EA, Cova L et al. Multipotential stem cells from the adult mouse brain proliferate and self-renew in response to basic fibroblast growth factor. J Neurosci 1996;16:1091–1100. Johe KK, Hazel TG, Muller T et al. Single factors direct the differentiation of stem cells from the fetal and adult central nervous system. Genes Dev 1996;10:3129 –3140. Jain M, Armstrong RJ, Tyers P et al. GABAergic immunoreactivity is predominant in neurons derived from expanded human neural precursor cells in vitro. Exp Neurol 2003;182:113–123. Milosevic J, Brandt A, Roemuss U et al. Uracil nucleotides stimulate human neural precursor cell proliferation and dopaminergic differentiation: Involvement of MEK/ERK signalling. J Neurochem 2006;99: 913–923. Knaus P, Betz H, Rehm H. Expression of synaptophysin during postnatal development of the mouse brain. J Neurochem 1986;47:1302–1304. Rehm H, Wiedenmann B, Betz H. Molecular characterization of synaptophysin, a major calcium-binding protein of the synaptic vesicle membrane. EMBO J 1986;5:535–541. Wiedenmann B, Franke WW. Identification and localization of synaptophysin, an integral membrane glycoprotein of Mr 38,000 characteristic of presynaptic vesicles. Cell 1985;41:1017–1028. Masliah E, Honer WG, Mallory M et al. Topographical distribution of


51 52 53 54 55 56 57

synaptic-associated proteins in the neuritic plaques of Alzheimer’s disease hippocampus. Acta Neuropathol (Berl) 1994;87:135–142. Rubenstein JL, Greengard P, Czernik AJ. Calcium-dependent serine phosphorylation of synaptophysin. Synapse 1993;13:161–172. Zhai RG, Vardinon-Friedman H, Cases-Langhoff C et al. Assembling the presynaptic active zone: A characterization of an active one precursor vesicle. Neuron 2001;29:131–143. Fenster SD, Chung WJ, Zhai R et al. Piccolo, a presynaptic zinc finger protein structurally related to bassoon. Neuron 2000;25:203–214. Fenster SD, Garner CC. Gene structure and genetic localization of the PCLO gene encoding the presynaptic active zone protein Piccolo. Int J Dev Neurosci 2002;20:161–171. Sheng M, Kim E. The Shank family of scaffold proteins. J Cell Sci 2000;113:1851–1856. Tu JC, Xiao B, Naisbitt S et al. Coupling of mGluR/Homer and PSD-95 complexes by the Shank family of postsynaptic density proteins. Neuron 1999;23:583–592. Du Y, Weed SA, Xiong WC et al. Identification of a novel cortactin SH3

www.StemCells.com

1729

58

59 60 61 62

domain-binding protein and its localization to growth cones of cultured neurons. Mol Cell Biol 1998;18:5838 –5851. Dresbach T, Torres V, Wittenmayer N et al. Assembly of active zone precursor vesicles: Obligatory trafficking of presynaptic cytomatrix proteins Bassoon and Piccolo via a trans-Golgi compartment. J Biol Chem 2006;281:6038 – 6047. Guan K, Chang H, Rolletschek A et al. Embryonic stem cell-derived neurogenesis. Retinoic acid induction and lineage selection of neuronal cells. Cell Tissue Res 2001;305:171–176. Rohwedel J, Kleppisch T, Pich U et al. Formation of postsynaptic-like membranes during differentiation of embryonic stem cells in vitro. Exp Cell Res 1998;239:214 –225. Song HJ, Stevens CF, Gage FH. Neural stem cells from adult hippocampus develop essential properties of functional CNS neurons. Nat Neurosci 2002;5:438 – 445. Bresler T, Ramati Y, Zamorano PL et al. The dynamics of SAP90/ PSD-95 recruitment to new synaptic junctions. Mol Cell Neurosci 2001; 18:149 –167.