Critical role of type 2 ryanodine receptor in mediating activity-dependent neurogenesis from embryonic stem cells

Cell Calcium 43 (2008) 417–431

Critical role of type 2 ryanodine receptor in mediating activity-dependent neurogenesis from embryonic stem cells Hui-Mei Yu a,b , Jing Wen c , Rong Wang a,b , Wan-Hua Shen c , Shumin Duan c , Huang-Tian Yang a,b,∗ a

Key Laboratory of Stem Cell Biology, Institute of Health Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences & Shanghai Jiao Tong University School of Medicine (SJTUSM), Shanghai 200025, China b Shanghai Stem Cell Institute of SJTUSM, Shanghai 200025, China c Institute of Neuroscience, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China Received 15 May 2007; received in revised form 22 July 2007; accepted 23 July 2007 Available online 4 September 2007

Abstract Activity-induced neurogenesis via Ca2+ entry may be important for establishing Hebbian neural network. However, it remains unclear whether intracellular Ca2+ mobilization is required and which subtypes of Ca2+ release channels expressed in Ca2+ store organelles are involved in the activity-dependent neurogenesis. Here, we demonstrated that the activity of intracellular Ca2+ signaling, expression of neuronal transcription factor NeuroD, and the rate of neurogenesis were significantly inhibited in neuronal cells derived from embryonic stem (ES) cells deficient in the Ca2+ release channel type 2 ryanodine receptors (RyR2−/− ). In wild-type (RyR2+/+ ) but not in RyR2−/− ES cells, activation of L-type Ca2+ channels, GABAA receptors, or RyRs promoted neuronal differentiation, while inhibition of these channels/receptors had an opposite effect. Moreover, neuronal differentiation promoted by activation of GABAA receptors or L-type Ca2+ channels in RyR2+/+ cells was prevented by RyR inhibitors. No significant difference was detected in the expression level of GABAA receptors and L-type channels between neuronal cells derived from two types of ES cells. Thus, activity-induced Ca2+ influx through L-type Ca2+ channels alone is not sufficient in promoting neurogenesis. Instead, an intimate cooperation of L-type Ca2+ channels with RyR2 is crucial for the activity-dependent neurogenesis induced by paracrine and/or autocrine GABA signaling. © 2007 Elsevier Ltd. All rights reserved. Keywords: Neurogenesis; Ryanodine receptors; ES cells; L-type Ca2+ channels; GABAA receptors

1. Introduction Neurogenesis is regulated by a variety of environmental and intracellular signals. Several lines of evidence suggest that intracellular calcium concentration ([Ca2+ ]i ) regulated by Ca2+ entry plays important roles in mediating the neurogenesis and neuronal differentiation [9,15,20,23,26]. ∗ Corresponding author at: Key Laboratory of Stem Cell Biology, Institute of Health Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences & Shanghai Jiao Tong University School of Medicine (SJTUSM), Shanghai 200025, China. Tel.: +86 21 63852593; fax: +86 21 63852593. E-mail address: [email protected] (H.-T. Yang). URL: http://www.ihs.ac.cn/emain-pi.asp?id=69 (H.-T. Yang).

0143-4160/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.ceca.2007.07.006

Spontaneous Ca2+ transients can be detected in cells as early as a neuronal phenotype can be determined [9] and appear to regulate onset of GABAergic phenotype during neural precursor differentiation [12]. In murine hippocampal progenitor cells, voltage-dependent inward currents present and parallel the progressive expression of functional neuronal phenotypes in vitro [38]. Moreover, neurogenesis is potently enhanced by Ca2+ influx through L-type Ca2+ channels in adult mouse neural stem/progenitor cells [15]. However, it is unclear whether intracellular Ca2+ mobilization is required for the activityinduced neurogenesis and which pathways are involved in the L-type Ca2+ channel-regulated neurogenesis. The endoplasmic reticulum (ER) represents an important [Ca2+ ]i store in various tissues. Two types of Ca2+ -release

418

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

channels are expressed in ER: ryanodine receptors (RyRs) and inositol 1,4,5-triphosphate receptors (IP3 Rs), with each type of the channels having three different isoforms [4,5]. In mature neurons, RyRs contribute to the regulation of synaptic transmission and plasticity and learning and memory [4,16]. During murine neurogenesis, all the three isoform of RyRs can be detected in cerebral hemispheres as early as the onset of neurogenesis [17]. However, it is unknown whether intracellular Ca2+ signaling regulated by RyRs contributes to early neurogenesis. Activity-dependent neurogenesis may be important for establishing Hebbian neural network [14,15]. It has been shown that GABAergic activity induces intracellular Ca2+ elevation in neural progenitor cells [7]. The effect of GABA on neurogenesis, however, can be either stimulatory [14,19,22,47] or inhibitory [22,30,35], depending on the type of progenitors studied and preparations used. Whether Ca2+ -induced Ca2+ release (CICR) mediated by RyRs [8,42] plays important roles in such activity-dependent neurogenesis and neuronal differentiation and which subtypes of RyRs are involved remains largely unexplored, probably due to the facts that RyRs still lack subtype-specific inhibitors and animals deficient in RyR2 die in early embryonic stage [45]. Embryonic stem (ES) cells have a potential to generate unlimited numbers of cells of all three primary germ layers including neurons, one possible source of transplantable cells. Embryoid bodies (EBs) formed from ES cells resemble early post-implantation embryos and differentiation of ES cells into neuronal cells represent a viable model to study early neuronal developmental paradigms [6,21]. This system is especially useful for identifying gene functions related to early development due to its easy accessibility for genetic manipulation. Using RyR2 deficient (RyR2−/− ) ES cells combining with multiple approaches, we demonstrated in the present study that Ca2+ release through RyR2 plays a critical role in mediating activity-dependent neurogenesis via amplifying intracellular Ca2+ signaling triggered by activation of L-type Ca2+ channels due to membrane depolarization caused by autocrine and/or paracrine released GABA.

drop medium, containing DMEM with 20% FBS and supplements as described above in ES cell basal medium for 2 days. Then the EBs were incubated in suspension medium, containing the same components as the ES cell basal medium without LIF for 6 days. RA was added during the last 4 days of suspension culture (Fig. 1). After 8 days, EBs were dissociated into single cells and the cells were plated onto gelatin-coated tissue culture dishes or coverslips in neuronal basal medium, containing modified DMEM supplemented with 10% F12, 2% B27, 1% N2, and 100 mg/ml bovine serum albumin (Sigma) for a further 8-day culture. All cultivation medium and other substances for cell cultures were purchased from Gibco BRL except where indicated. 2.2. Immunostaining EBs were fixed for 45 min at room temperature in 4% paraformaldehyde in 0.1 M PBS, containing (in Mm) 137 NaCl, 2.7 KCl, 5.4 Na2 HPO4 , 0.56 KH2 PO4 , pH 7.4, then embedded in OCT (Tissue Tek) and rapidly frozen by CO2 . Ebs were taken at 20–25 ␮m thicknesses. Cells on coverslips were fixed for 15 min in 4% paraformaldehyde and washed three times in PBS [35,43]. Ebs, EB slices and cell cultures were permealized with 0.3% Triton X-100 for 15 min and treated with 10% BSA for 1 h. Each of them were incubated with monoclonal antibodies against Tuj1 (1:1000, Promega,), nestin (1:100, Chemicon), MAP2ab-2 (1:100, Chemicon), RyR2 (1:100, Affinity Bioreagents), ␣1 subunit of GABAA receptors (1:400, Alomone labs), or GFAP (1:1000, Chamicon) overnight at 4 ◦ C. After washing three times in PBS for 10 min, cells were stained for 60 min with the appropriate fluorophore-conjugated secondary antibodies at 1:200 or 1:400 dilution in PBS against specific light chains of mouse or rabbit IgG (Santa Cruz Biotechnology). Cell nuclei were stained with Hoechst No. 33342 (1 mg/Ml) or PI (10 mg/Ml, Sigma) for 10 min. Stained cells were photographed under a Zeiss Axiophot I microscope equipped with epifluorescence optics or a Zeiss LSM510 confocal laser scanning microscope using appropriate fluorescence filters. 2.3. BrdU immunostaining

2. Methods 2.1. ES cell culture and neuronal differentiation RyR2+/+ and RyR2−/− R1 ES cells [25,51] were cultivated and differentiated into neuronal cells as previously described [1]. Briefly, undifferentiated ES cells were grown on a feeder layer of mitomycin C-inactivated mouse embryonic fibroblasts in the presence of leukemia inhibitory factor (LIF) in ES cell basal medium containing Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 15% heat-inactivated fetal bovine serum (FBS, Hycone), 1% non-essential amino acids, 2 mM glutamine, 100 ␮M ␤-mercaptoethanol. Differentiation of these two type ES cells into neuronal cells was initiated by a hanging drop method to form EBs in hanging

Cell cultures on coverslips were incubated with BrdU (20 ␮M, Molecular Probes) for 18 h and then were fixed in 70% EtOH + 30% 50 mM glycine, pH 2.0 solution at −20 ◦ C for 20 min [12]. Cells were washed twice with PBS, then incubated with 50 ␮l of 1:20 diluted mouse anti-BrdU (Alexa Fluor 546 conjugated, Roche) in incubator kits for 30 min. 2.4. Ca2+ imaging Cells were loaded with 2 ␮M of Fluo4-AM (Molecular Probes) for 20 min at 37o C and then were washed with extracellular medium containing (in mM): 140 NaCl, 3 KCl, 1 MgCl2 , 2 CaCl2 , 10 HEPES, 10 glucose and incubated for a further 20 min to allow complete de-esterification of the

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

419

Fig. 1. Expression of RyR2 during neuronal differentiation of ES cells. (A) The protocol for inducing ES cell differentiation into neural cells. All-transretinoic acid (RA) and pharmacological interventions (Pha-I) including various agonists or antagonists were applied from 5 days to 8 days of suspension culture. (B) Representative immunostaining of RyR2 (middle panels, red) and neural precursor marker nestin or mature neuronal marker MAP2ab (top panels, green) in embryoid bodies (EBs) or cells derived from RyR2+/+ ES cells. Rightest panel shows example of evoked action potentials (AP) recorded from 8 + 8 days wild-type neuronal cells. (C) Representative immunostaining of RyR2 (middle panels, red) and nestin or MAP2ab (top panels, green) in EBs or cells derived from RyR2−/− ES cells. Rightest panel, example of evoked AP recorded from RyR2−/− neuronal cells. (D) RT-PCR analysis of transcripts for RyR1, RyR2, RyR3 and house keeping gene 28s in RyR2+/+ and RyR2−/− cells at various differentiation stages as well as in adult brain. Scale bars in (B) and (C), 20 ␮m.

420

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

dye [17]. Ca2+ imaging was performed with Zeiss LSM510 confocal laser scanning microscope using appropriate fluorescence filters at room temperature (20 ± 2 ◦ C). Results were expressed as F/Fo [(F = 100 × (F − Fo ) and Fo = basal fluorescence].

GAACTGG and R, TAGGGACAGTGGGAATCTCG. DNA was visualized on 1% agarose gel containing ethidium bromide.

2.5. Electrophysiology

Data are presented as mean ± S.E.M. Statistical significance of differences was estimated by one way ANOVA or by Student’s t test or a paired t test, when appropriate (StatSoft, Version 5.1, StatSoft, Tulsa, OK, USA). P < 0.05 was considered significant. All chemicals were from Sigma except BayK 8644 which was from Calbiochem.

Voltage-gated Ca2+ currents and GABA-induced currents were recorded using the whole-cell patch clamp technique with Multi-clamp 700A amplifier (Axon instruments Inc.) [29]. The resistance of the electrode filled with internal solution was 3–9 M. Signals were filtered at 2 kHz and sampled at 10 kHz. Data were analyzed using Clampfit 9.0 (Axon Instruments Inc.). Solutions were rapidly exchanged using a gravity-driven multi-barrel perfusion system. Voltage-gated Ca2+ currents were recorded using previously described solutions. For GABA currents, the intracellular solution contained (in mM): 120 K-Gluconate, 30 KCl, 10 HEPES, 10 Na2 -phosphocreatine, 4 ATP-Na2 , 0.3 GTP-Na2 , 10 EGTA (pH 7.3). The external solution contained (in mM): 148 NaCl, 3 KCl, 10 HEPES, 3 CaCl2 , 8 Glucose, 2 MgCl2 (pH 7.3). 2.6. Total RNA isolation and RT-PCR Undifferentiated ES cells, differentiated cells and adult mouse brains were used to isolate total RNA. Total RNA from each sample was converted to cDNA by using Superscript II reverse transcriptase (Life-Tech) and oligodT (T16, 500 ng) in a final volume of 20 ␮l, according to the manufacturer’s instruction, and 0.4 ␮l of this was used for each PCR reaction. Semiquantitative RT-PCR reactions were performed with Taq DNA polymerase (Promega). The primer pairs of RyR1: F, 5 CCACTTTGCCGAGTTTGC 3 and R, 5 GGTCCTGGTTGGAGCGTA 3 ; RyR2: F, 5 -GACGGCAGAAGCCACTCACCTGCG-3 and R, 5 -CCTGCAGAGAAACTGACAACTGG-3 ; and RyR3: F, 5 GGAACCCACCTCAGAAGC 3 and R, 5 TCCCGCAGACCTACATCA 3 ; Hes1: F, CGGCTTCAGCGAGTG CAT and R, CGGTGTTAACGCCCTCACA [15]; NeuroD: F, TTCACGATTAGAGGCACG and R, CCAAAGGCAGTAACGACA [15]. Housekeeping gene, 28s: F, AGCAGCCGACTTA-

2.7. Statistics

3. Results 3.1. Neurogenesis in RyR2+/+ and RyR2−/− ES cells Wild-type (RyR2+/+ ) and RyR2−/− R1 ES cells [25,51] were cultured and differentiated into neural lineages by using 1 ␮M of all-trans-retinoic acid (RA, see Section 2 and Fig. 1A). Both undifferentiated RyR2+/+ and RyR2−/− ES cells had normal growth characteristics as observed previously [25,51] and formed similar size of EBs during the 8 days of suspension culture before and after addition of RA. RA was added after the first 4 days and left for the last 4 days. The 8 days EBs were dissociated with trypsin and plated for a further 8 days culture (8 + 8 days). Neural precursors, identified by staining with an antibody to the intermediate filament protein nestin [28], were detectable on peripheral cell layers of both RyR2+/+ and RyR2−/− EBs at 2 days and increased significantly at 5 days (Fig. 1B and C, upper panel). The RyR2+/+ and RyR2−/− cells with neuronal morphology (Figs. 1A and 2A) were positive for MAP2ab, a specific maker for mature neuron [43], examined at 8 + 0 and 8 + 4 days. Typical action potentials could be detected at 8 + 8 days (Fig. 1B, rightest panel). The time course of the expression of three isoforms of RyR transcripts in ES cell-derived cells was examined. The transcript of RyR2 was detectable from 5 days RyR2+/+ cells and increased significantly with time of differentiation, but it was not detected in RyR2−/− cells, whereas weak transcript signals of RyR1 and RyR3 were detectable from 8 + 0

Fig. 2. Neurogenesis from RyR2+/+ and RyR2−/− ES cells. (A) Representative immunostaining of Tuj1 (green) and propidium iodine (PI, red) in cells derived from RyR2+/+ and RyR2−/− ES cell at different development stages. Low (10 nM, Rya (L)) or high concentration (3 ␮M, Rya (H)) of ryanodine was applied during day 5–8 suspension culture to stimulate or inhibit RyRs, respectively. PI is used for labeling cell nuclear. Scale bars, 50 ␮m. (B) Summarized data from experiments shown in (A). All data were from four to six independent experiments in parallel cultures and >500 cells were analyzed per experiment from at least 5 areas. (C) Example (left panels) and summarized (right panels) RT-PCR analysis of transcripts for NeuroD, Hes1 and 28s in RyR2+/+ and RyR2−/− cells at different differentiation stages. Data were from four independent experiments in parallel cultures and normalized with the level of 28s in each group. (D) Left panels, representative immunostaining of Tuj1 (green), GFAP (red), and hoechst (blue) in cells derived from RyR2+/+ and RyR2−/− ES cells at 8 + 8 days. Right panels, summarized data of the percentage of GFAP+ cells and Tuj1+ cells (upper panel) and the percentage of total neural cells (lower panel, GFAP+ cells plus Tuj1+ cells) in total cells (neural cells plus non-neural cells) derived from RyR2+/+ and RyR2−/− ES cells at 8 + 8 days. All data were from four to six independent experiments in parallel cultures and >500 cells were analyzed per experiment from at least 5 areas. * P < 0.05, ** P < 0.01, compared with control groups or with the group indicated.

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

in both types of ES cell cultures (Fig. 1D). Consistently, the expression of RyR2 protein was detected by staining with an antibody selective for RyR2 in RyR2+/+ cells at 5, 8 + 0, and 8 + 4 days, but not in RyR2−/− cells (Fig. 1B and C,

421

middle panels). Moreover, the expression of RyR2 colocalized with the expressions of nestin or MAP2ab in RyR2+/+ group during neuronal differentiation (Fig. 1B and C, bottom panels).

422

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

To examine whether RyR2 contributes to the neuronal differentiation, we compared the profiles of neuronal differentiation in cells derived from RyR2+/+ and RyR2−/− ES cells. Cells with oval neuronal morphology were positive for neuronal marker ␤III-tubulin (Tuj1+ , [13]). The percentage of Tuj1+ cells in RyR2+/+ cultures examined at 8 + 0 days was significant higher than that in RyR2−/− cultures (Fig. 2A and B). Furthermore, a significant increase in the proportion of Tuj1+ cells in RyR2+/+ cultures was observed when examined at 8 + 4 days or 8 + 8 days compared with that examined at 8 + 0 days (Fig. 2A and B, left panels), while the Tuj1+ cells in RyR2−/− cultures kept at a constant low level at all the three development stages examined (Fig. 2A and B, right panels). 3.2. Activation of RyRs increases neurogenesis in RyR2+/+ but not in RyR2−/− ES cells To further examine whether the increased neuronal differentiation in RyR2+/+ cells is directly related to RyR function, we analyzed neuronal differentiation by adding ryanodine into suspension medium from 5 days at a low (10 nM) or a high (3 ␮M) concentration, which is known to stimulate or inhibit the activity of RyRs, respectively [52]. We found that treatment of RyR2+/+ cultures with 10 nM ryanodine resulted in a significant increase in the proportion of Tuj1+ cells at all the three development stages examined, while 3 ␮M ryanodine decreased the proportion of Tuj1+ cells to a level similar to that found in RyR2−/− cultures. However, similar treatment of RyR2−/− cultures with either the low or the high concentration of ryanodine failed to affect the proportion of Tuj1+ cells at all the three development stages examined (Fig. 2A and B). Differentiation of neural precursors/stem cells is controlled, to a large degree, by transcription factors with basic helix–loop–helix (bHLH) motif, such as neuronal determination factors (NeuroD-related factors), which mediate terminal neurogenesis, and inhibitory bHLH proteins (Hes factors), which inhibit neurogenesis [37,40]. To determine whether the difference in neurogenesis between RyR2+/+ and RyR2−/− cells is correlated with the deferential levels of bHLH factors, we examined the expression patterns of NeuroD and Hes1 genes during the early neuronal differentiation from the two types of cells. With increased neuronal differentiation, the expression of NeuroD and Hes1 transcripts in both RyR2+/+ and RyR2−/− cells increased. Furthermore, the expression of NeuroD in RyR2−/− cells was significantly lower than that in RyR2+/+ cells after 8 days suspension culture, while the expression of Hes1 in RyR2−/− cells was significantly higher than that in RyR2+/+ cells after 5 days suspension culture (Fig. 2C). Since Hes1 plays important roles in gliogenesis, we compared neuron/glia ratio in RyR2+/+ and RyR2−/− ES cells. A significant increase in the proportion of GFAP (astrocyte marker [43]) positive cells, calculated either as percentage of neural cells (neurons plus astrocytes) or percentage of total

Table 1 Proliferation of neuronal cells in RyR2+/+ and RyR2−/− cultures at 8 + 0 days RyR2+/+ (nestin+ )

Percent of precursors Percent of BrdU+ cells in precursors Percent of BrdU+ cells in neurons (Tuj1+ ) Percent of neurons displaying pyknotic nuclei *

42.4 87.5 22.0 1.4

± ± ± ±

3 2 4 1

RyR2−/− 46.1 87.6 11.0 1.9

± ± ± ±

7 3 4* 1

P < 0.05 compared with RyR2 +/+ group.

cells (neural cells plus non-neural cells), was observed in RyR2−/− cells compared with that in RyR2+/+ cells (Fig. 2D). However, the ratio of neural cells as percentage of total cells in RyR2−/− cultures was not significantly different from that in RyR2+/+ cultures examined at 8 + 8 days (right lower panel in Fig. 2D). These results, together with the finding that RyR2 promoted the proliferation of Tuj1+ cells, but not that of Nestin+ cells (Table 1), suggest that RyR2-mediated [Ca2+ ]i elevation enhance neurogenesis by shifting the differentiation of neural precursors toward the neuronal fate, rather than by simply inhibiting gliogenesis. 3.3. Decreased neurogenesis in RyR2−/− cells at immature neuron stage but not at neural precursor stage To examine whether the decreased neurogenesis in RyR2−/− cultures results from decreased differentiation of ES cells into neural precursors or from decreased differentiation of neural precursors into neurons, we immunostained RyR2+/+ and RyR2−/− EBs with neural precursor marker nestin and immature neuron marker Tuj1 every 8 h after RA treatment. Within 40 h of RA treatment, only nestin+ cells, but not Tuj1+ cells, were detected and the percentage of nestin+ fraction was similar between RyR2+/+ and RyR2−/− EB cultures (data not shown). At 48 h of RA treatment, a few Tuj1+ cells (Fig. 3A and B, right panel) could be detected and the percentage of nestin+ fraction was similar in both RyR2+/+ and RyR2−/− EB cultures (Fig. 3A and B, left panel). After 56 h treatment with RA, a significant higher proportion of Tuj1+ cells (Fig. 3A and B, right panel), but not of nestin+ cells (Fig. 3A and B, left panel), was detected in RyR2+/+ than in RyR2−/− EB cultures. To further determine whether the reduced number of Tuj1+ cells in RyR2−/− cultures is due to the decreased neuronal differentiation or/and decreased cell viability, we examined the percent of nestin positive cells, the proportion of nestin+ or Tuj1+ cells that were incorporated with BrdU (5-bromo2 deoxy-uridine) added for 18 h, and the percent of cells with pyknotic nuclei identified by Hoechst 33342 staining, an index for cell death [12], in RyR2+/+ and RyR2−/− cultures at 8 + 0 days (Table 1). Neither the proportion of nestin+ cells, nor the BrdU incorporation rate of nestin+ cells had a significant difference between RyR2+/+ and RyR2−/− groups (Table 1), whereas the BrdU incorporation rate of Tuj1+ cells was significantly higher in RyR2+/+ than that in RyR2−/− group (Table 1). The cell viability analyzed by comparing the

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

423

Fig. 3. Comparison of genesis of neural precursors and neurons from RyR2+/+ and RyR2−/− cultures. (A) Representative co-immunostaining of PI (red) and nestin (left panels)/Tuj1 (right panels, green) in RyR2+/+ and RyR2−/− EBs at 48 h and 56 h after RA treatment. Scale bars, 50 ␮m. (B) Summarized data of experiments as shown in (A) for the fraction of nestin (left panel) or Tuj1 (right panel) fluorescence signal (green) to PI signal (red) in RyR2+/+ and RyR2−/− cultures. All data were from four independent experiments in parallel cultures and n = 17–26 imaging fields for each group. *** P < 0.001 compared with RyR2+/+ group.

ratio of Tuj1+ cells with pyknotic nuclei was similar between two groups (Table 1). Taken together, these results indicate that RyR2 promotes differentiation of neural precursors to neurons and regulates the proliferation of immature neurons, but does not affect the proliferation rate of neural precursors or neuron viability. RyRs are Ca2+ release channels on ER and contribute to the dynamics of Ca2+ signaling in neurons [4]. It is not clear whether RyR2 contributes to the Ca2+ signaling during early neuronal developmental stage and whether the Ca2+ signaling regulated by RyR2 is required for neurogenesis. We, therefore, measured spontaneous Ca2+ transients [20] in neuronal cells (neural precursors and immature neurons) derived from RyR2+/+ and RyR2−/− ES cells at 8 + 0 days. The neuronal cells, characterized by their oval cell bodies and neuritic outgrowths, are distinct from other cell types in morphology and were confirmed by immuostaining (data not shown). The frequency of spontaneous Ca2+ transients was significantly higher in RyR2+/+ cells than that in RyR2−/− cells (Fig. 4A). Moreover, the percentage of active cells exhibiting Ca2+ tran-

sients within 30-min imaging period was significantly higher in RyR2+/+ cells than that in RyR2−/− cells (Fig. 4B). Thus, RyR2 is functional in regulating spontaneous Ca2+ transients in early-stage of neuronal cells derived from ES cells. To examine whether RyR2 is involved in the stimulation-evoked intracellular Ca2+ activity, we compared the Ca2+ transients induced by ryanodine (100 nM) or caffeine (10 ␮M), an activator of RyRs, in RyR2+/+ and RyR2−/− neuronal cells. Ryanodine could induce Ca2+ transients in both RyR2+/+ and RyR2−/− cells at 8 + 0 and 8 + 4 days, but the amplitude of Ca2+ transients in RyR2+/+ cells was significantly higher than that in RyR2−/− cells (Fig. 5A and B). Furthermore, the rising velocity of Ca2+ transients induced by ryanodine was slower in RyR2−/− cells at both 8 + 0 and 8 + 4 days than that in RyR2+/+ cells (Fig. 5A and C). The amplitude and rising velocity of ryanodine-induced Ca2+ transients in RyR2+/+ cells examined at 8 + 4 days stage were higher than that examined at 8 + 0 days stage, a result not observed in RyR2−/− cells (Fig. 5B and C). Similar results were obtained in the caffeine-induced Ca2+ transients (Fig. 5D–F).

424

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

Fig. 4. Comparison of spontaneous Ca2+ transients in neuronal cells derived from RyR2+/+ and RyR2−/− ES cells. (A) Representative traces (upper panel) and frequency (lower panel) of spontaneous Ca2+ transients in RyR2+/+ (n = 13) and RyR2−/− (n = 18) neuronal cells examined at 8 + 0 days. (B) Percentage of neural cells with spontaneous Ca2+ transients in RyR2+/+ (n = 13) and RyR2−/− cells (n = 18). ** P < 0.01, *** P < 0.001 compared with RyR2+/+ group.

3.4. Activation of L-type Ca2+ channels increases neurogenesis via triggering Ca2+ release from RyR2 Ca2+ entry plays an important role in the activitydependent neurogenesis in neuronal stem cells [15]. Since RyRs are activated by Ca2+ influx through membrane Ca2+ channels in neurons [11], we examined whether the Ca2+ entry is involved in the neuronal differentiation of ES cells and whether the regulatory effect of Ca2+ influx on neuronal differentiation is functionally coupled to RyR2. To address these questions, we first investigated the impact of activation of voltage-dependent Ca2+ channels on neurogenesis in RyR2+/+ and RyR2−/− cells by applying modest depolarizing level of extracellular KCl (20 mM), together with RA at 5 days. KCl-induced Ca2+ transients in these two kinds of EBs (data not shown). The proportion of Tuj1+ cells in RyR2+/+ cells at 8 + 0 days and 8 + 4 days was significantly increased by treatment with 20 mM KCl (Fig. 6A and B), indicating that Ca2+ influx promotes neuronal differentiation at early developmental stage. The promoting effect of KCl on the early neuronal differentiation observed in RyR2+/+ cells was

not detected in RyR2−/− cells (Fig. 6A). The Tuj1+ cells in RyR2−/− cultures in the presence of 20 mM KCl (4.6 ± 0.5%, 5.2 ± 0.4%, and 7.1 ± 1.6% at 8 + 0, 8 + 4, and 8 + 8 days, respectively) was not significantly different from that without high KCl treatment (3.9 ± 0.9%, 3.5 ± 0.5%, and 5.4 ± 0.4% at 8 + 0, 8 + 4, and 8 + 8 days, respectively). Furthermore, the KCl-induced increase in Tuj1+ cells in RyR2+/+ cultures was prevented by treatment with high concentration of ryanodine (3 ␮M) that inhibits RyRs (Fig. 6A and B), indicating that RyR2 is critical for the promotion of neuronal differentiation induced by treatment with 20 mM KCl. We then directly analyzed the functional relationship of RyR2 and L-type Ca2+ channels by applying BayK 8644 (1 ␮M), an agonist of L-type Ca2+ channels. We found that in RyR2+/+ cells, BayK 8644 significantly increased Tuj1+ cells at 8 + 0 and 8 + 4 days, an effect prevented by treatment of cultures with high concentration of ryanodine (3 ␮M, Fig. 6C and D). In contrast, treatment of RyR2+/+ cells with 10 ␮M of nifedipine, an antagonist of L-type Ca2+ channels, significantly inhibited proportion of Tuj1+ cells, suggesting that tonic activity of L-type Ca2+ channels is important for the neuronal differentiation in ES cells. On the other hand, neither BayK 8644 nor nifedipine had effects on the neuronal differentiation in RyR2−/− cells (Fig. 6C). The Tuj1+ cells in RyR2−/− cultures in the presence of BayK 8644 (4.1 ± 0.6%, 4.5 ± 0.5%, and 6.9 ± 1% at 8 + 0, 8 + 4, and 8 + 8 days, respectively) or nifedipine (3.8 ± 0.4%, 3.7 ± 0.5%, and 5.7 ± 1% at 8 + 0, 8 + 4, and 8 + 8 days, respectively) was not significantly different from control RyR2−/− cultures (3.9 ± 0.9%, 3.5 ± 0.5%, and 5.4 ± 0.4% at 8 + 0, 8 + 4, and 8 + 8 days, respectively). These results demonstrate that activation of L-type Ca2+ channel promotes the neuronal differentiation through its collaboration with RyR2. Lack effect of KCl or BayK 8644 treatment on the proportion of Tuj1+ cells examined at 8 + 8 stage in RyR2+/+ cultures (Fig. 6B and D) suggests that the basal activity of L-type Ca2+ channels is already saturated at this stage in promoting neurogenesis. To directly estimate the contribution of RyR2 in the elevated [Ca2+ ]i induced by activation of L-type Ca2+ channels in differentiating neuronal cells, we compared BayK 8644induced Ca2+ transients in RyR2+/+ and RyR2−/− cultures. As shown in Fig. 7, the amplitude of Ca2+ transients induced by application of BayK 8644 was significantly higher in RyR2+/+ cells at 8 + 0 and 8 + 4 days than that in RyR2−/− cells (Fig. 7A and B). Consistently, the rising velocity of Ca2+ transients was significantly faster in RyR2+/+ cells than that in RyR2−/− cells either at 8 + 0 or at 8 + 4 days (Fig. 7A and C). The reduced elevation of [Ca2+ ]i in RyR2−/− cells may not due to the decreased expression or function of L-type Ca2+ channels, because the current density of L-type Ca2+ channels recorded at RyR2−/− cells at 8 + 0 days was not significantly different from that recorded at RyR2+/+ cells (Fig. 7D). Taken together, these data indicate that RyR2 plays important roles in neurogenesis via amplifying [Ca2+ ]i signaling triggered by activation of L-type Ca2+ channels.

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

425

Fig. 5. Dynamics of ryanodine- or caffeine-induced Ca2+ transients in RyR2+/+ and RyR2−/− neural cells. (A–C) Representative traces (A), averaged amplitude (B), and the rising velocity for 50% peak of Ca2+ transients (C, V for 50% of F/F0 ) induced by ryanodine (100 nM) in RyR2+/+ and RyR2−/− neuronal cells at 8 + 0 (n = 8, 9) and 8 + 4 days (n = 9, 12). (D–F) Representative traces of Ca2+ transients (D), averaged amplitude (E), and V for 50% peak of F/F0 (F) induced by caffeine (10 ␮M) in RyR2+/+ and RyR2−/− neuronal cells at 8 + 0 (n = 10, 11) and 8 + 4 days (n = 24, 30). * P < 0.05, ** P < 0.01, *** P < 0.001 compared with RyR2+/+ group or with the group indicated.

3.5. Activation of GABAA receptors promotes neurogenesis through L-type Ca2+ channel-RyR2 pathway Activation of GABAA receptors is involved in the regulation of proliferation and differentiation of adult neural stem cells [14,18,27,29,47]. We then examined whether such GABAergic regulation is involved in the neurogenesis from ES cells and in particular, if such regulation exists, whether RyR2 is required for this GABAergic activity-dependent regulation of neurogenesis. Immunostaining with anti-␣1 subunit of GABAA receptors showed that both neural precursors (nestin+ cells) and neurons (Tuj1+ cells) from either RyR2+/+ or RyR2−/− cells at 8 + 0 days expressed GABAA receptors (Fig. 8A and B). Electrophysiological recording from either RyR2+/+ or RyR2−/− cells detected apparent GABA-induced currents that could be blocked by GABAA

receptor antagonist bicuculline (Fig. 8C). Furthermore, statistical analysis did not show significant difference in the density of GABA-induced currents between RyR2+/+ and RyR2−/− cells (Fig. 8C), indicating that RyR2 does not affect the expression and channel activity of GABAA receptors. However, treatment of cultures with GABA (100 ␮M) significantly increased the proportion of Tuj1+ cells at 8 + 0 days and 8 + 4 days stages in RyR2+/+ (Fig. 9A left panel and B) but not in RyR2−/− cells (Fig. 9A, right panel; summarized data not shown), an effect that could be prevented in the presence of either bicuculline or nifedipine. Moreover, bicuculline or nifedipine inhibited the proportion of Tuj1+ cells in RyR2+/+ cultures (Fig. 9A left panel and B) to a level comparable to that seen in RyR2−/− cells (Fig. 9A right panel, summarized data not shown), indicating that GABAA receptor activation induced by endogenously released GABA significantly promotes the neurogenesis, an effect depending

426

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

Fig. 6. Activation of L-type Ca2+ channels promotes neuronal differentiation via RyR2. (A) Representative co-immunostaining of Tuj1 (green) and PI (red) in RyR2+/+ and RyR2−/− cells pretreated with KCl (20 mM) in the absence (middle panels) or presence (bottom panels) of ryanodine at high concentration (3 ␮M, Ray (H)), respectively, at 8 + 4 days. (B) Summarized data of (A) in RyR2+/+ cells at 8 + 0, 8 + 4, and 8 + 8 days. (C) Representative co-immunostaining of Tuj1 (green) and PI (red) in RyR2+/+ and RyR2−/− cells pretreated with L-type Ca2+ channel agonist BayK 8644 (1 ␮M) or antagonist nifedipine (10 ␮M) in the absence or presence of ryanodine (3 ␮M) at 8 + 4 days. (D) Summarized data of (C) in RyR2+/+ cells at 8 + 0, 8 + 4, and 8 + 8 days. All data were from four to six independent experiments in parallel cultures and >500 cells were analyzed per experiment from 6 to 10 areas. * P < 0.05; ** P < 0.01 compared with control group. Scale bars in (A) and (C), 50 ␮m.

on intact function of RyR2 as well as activation of L-type Ca2+ channels. Lack effect of GABA treatment on the proportion of Tuj1+ cells examined at 8 + 8 stage in RyR2+/+ cultures (Fig. 9A and B) suggests that the effect of endogenous GABA is already saturated at this stage in promoting neurogenesis. Activation of GABAA receptors on neurons in early developmental stage causes depolarization of membrane potential due to higher intracellular Cl− concentration. This depolarization may induce Ca2+ influx through voltage-dependent L-type Ca2+ channels. To further elucidate mechanisms underlying GABAA receptor-mediated promotion in neurogenesis from ES cells, we examined whether GABA-induced similar [Ca2+ ]i elevation in RyR2+/+ and RyR2−/− cells using Ca2+ imaging technique. As shown in Fig. 9C–F, although perfusion of GABA-induced apparent [Ca2+ ]i elevation in both RyR2+/+ and RyR2−/− cells at 8 + 0 days and 8 + 4 days stages, a significant higher [Ca2+ ]i elevation induced by GABA was observed in RyR2+/+ cells. The involvement of both GABAA receptors and L-type Ca2+ channels in this

GABA-induced [Ca2+ ]i elevation was confirmed by using their specific antagonists bicuculline (100 ␮M) and nifedipine (100 ␮M), respectively (Fig. 9C and D).

4. Discussion 4.1. Critical role of RyR2-dependent [Ca2+ ]i signaling in neurogenesis Ca2+ elevation in mature neurons can be mediated by multiple pathways, including extracellular Ca2+ influx though various ligand- and voltage-gated channels and intracellular Ca2+ release from Ca2+ stores through IP3 Rs or RyRs [4,5,49]. IP3 Rs are activated by the second messenger IP3 produced by activation of G-protein coupled receptors. RyRs may be coupled with L-type Ca2+ channels in two different ways: type 1 RyR (RyR1) in skeletal muscle is physically linked with L-type Ca2+ channels (dihydropyridine receptor, DHPR), which function as a voltage sense and induce

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

427

Fig. 7. Involvement of RyR2 in Ca2+ transients induced by activation of L-type Ca2+ channels. (A–C) Representative traces (A), averaged amplitude (B), and the rising velocity (V) for 50% peak (C) of Ca2+ transients increased by 1 ␮M of BayK 8644, an agonist of L-type Ca2+ channels, in RyR2+/+ (n = 14, 16) and RyR2−/− (n = 14, 20) neuronal cells at 8 + 0 and 8 + 4 days. (D) Current density of L-type Ca2+ channels in RyR2+/+ and RyR2−/− neural cells at 8 + 0 days (n = 15 for each group). * P < 0.05, ** P < 0.01 compared with RyR2+/+ group.

conformation change and opening of RyR1 in response to membrane depolarization [36,39,45], while RyR2 is activated by extracellular Ca2+ influx through the DHPR via CICR mechanism [8,16,42]. The RyRs are involved in the generation of intracellular Ca2+ oscillations required for various neuronal functions [2,7,17,34,37,40,46]. Our results indicate that RyR2 plays a key role in mediating spontaneous Ca2+ signal and in promoting neuronal differentiation from ES cells. RyR2 is intensively expressed in neural precursors from wild-type ES cells (Fig. 1) and both the frequency of spontaneous Ca2+ transients and percentage of cells exhibiting Ca2+ transients are significantly lower in RyR2−/− than that in wild-type neural precursors (Fig. 4). Most importantly, the neurogenesis from RyR2−/− cultures was significantly slower than that in wild-type cultures (Fig. 2). Furthermore, in wild-type but not in RyR2−/− cultures, activation of RyRs with caffeine or low concentration of ryanodine promoted neurogenesis (Fig. 2), while inhibition of RyRs with high

concentration of ryanodine had an opposite effect (Fig. 2). The result that the percentage of Tuj1-positive cells but not that of nestin-positive cells is significantly decreased in cultures from RyR2−/− ES cells (Figs. 2 and 3) suggests that RyR2-mediated Ca2+ signaling promote neurogenesis from neural precursors but not affect generation and/or proliferation of neural precursors from ES cells. Consistent with this notion, we found that the level of transcription factor NeuroD, which promotes neuronal differentiation and maturation [40], was significant decreased, while that of Hes1, which inhibits the neuronal phenotype [37], was significantly enhanced in RyR2−/− cells (Fig. 2). In addition, the decreased population of Tuj+ neuronal cells is associated with an increased population of GFAP+ glial cells without change in the total population of neural cells (Tuj+ cells plus GFAP+ cells) in RyR2−/− cultures (Fig. 2D). Because neurons and glial cells are generated from a common progenitor [6,21,37], our results suggest that RyR2-mediated [Ca2+ ]i elevation plays

428

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

Fig. 8. Expression of GABAA receptors in RyR2+/+ and RyR2−/− neural cells. (A) and (B), Representative co-immunostaining of ␣1 subunit of GABAA receptors (red) and nestin (green) (A) or Tuj1 (green) (B) in cells derived from RyR2+/+ and RyR2−/− ES cells at 8 + 0 days. (C) Example traces (top and middle panels) and the summarized density (bottom panel) of GABA-induced current in the presence or absence of bicuculline (Bic, 100 ␮M), an antagonist of GABAA receptors, in RyR2+/+ and RyR2−/− neural cells (n = 17, 18) at 8 + 0 days. Scale bars in (A) and (B), 50 ␮m.

critical role in activity-dependent neurogenesis by shifting the differentiation of neural precursors toward the neuronal fate. 4.2. Critical role of RyR2 in mediating activity-dependent neurogenesis induced by activation of GABAA receptors and L-type Ca2+ channels We found that nifedipine-sensitive Ca2+ current could be recorded at 8 + 0 cultures when neural precursors were dominant. Furthermore, application of nifedipine inhibited proportion of Tuj1+ cells in wild-type cultures (Fig. 6), suggesting that tonic activation of L-type Ca2+ channels contributes to the neurogenesis from ES cells. L-type Ca2+ channels distribute mainly on cell bodies and proximal

dendrites of neurons and may not play important role in neurotransmitter release [10]. However, Ca2+ entry through L-type Ca2+ channels is more efficient than that though other types of Ca2+ channels in regulating gene transcription and thus plays a critical role in mediating excitation-transcription coupling in neurons [10,13,33]. Consistently, the expression of transcription factor NeuroD and Hes1 important for neural differentiation are up and down regulated, respectively, by activation of L-type Ca2+ channels [15]. Our results that NeuroD and Hes1 were significant decreased and enhanced, respectively, in RyR2−/− cultures (Fig. 2) suggest that the activity-dependent regulation of these transcription factors depends on RyR2-mediated CICR. Activation of CICR requires a relatively high level of threshold [Ca2+ ]i and thus needs a long lasting depolarization

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

429

Fig. 9. Promotion of neuronal differentiation induced by activation of GABAA receptors involves CICR mediated by L-type Ca2+ channels and RyR2. (A) Representative co-immunostaining of Tuj1 (green) and PI (red) in RyR2+/+ and RyR2−/− cells at 8 + 4 days, pretreated with GABA (100 ␮M) in the presence or absence of nifedipine (10 ␮M), an antagonist of L-type Ca2+ channels, or bicuculline (Bic, 100 ␮M), an antagonist of GABAA receptors. Scale bar, 50 ␮m. (B) Summarized data showing percentage of neurons (Tuj1+ ) in RyR2+/+ cells at 8 + 0, 8 + 4, and 8 + 8 days, under different pretreatments as shown in (A). All data were from four to six independent experiments in parallel cultures and >500 cells were analyzed per experiment from at least 5 areas. (C and D) Representative traces of GABA-induced Ca2+ transients in the absence or presence of bicuculline (C) or nifedipine (D) in RyR2+/+ and RyR2−/− neural cells at 8 + 0 days. Averaged amplitude (E) and velocity (V) for 50% peak (F) of Ca2+ transients induced by GABA in RyR2+/+ and RyR2−/− neural cells at 8 + 0 (n = 17, 20) and 8 + 4 days (n = 15, 33). * P < 0.05, ** P < 0.01, *** P < 0.001 compared with RyR2+/+ group or with the group indicated.

of the cell [49]. Compared with other types of voltagegated Ca2+ channels, L-type Ca2+ channels are activated at a more negative potential and are relatively resistant to inactivation [48], making them well-suited to tonic signaling [31,41]. Since neuronal precursors tend to be more easily depolarized than mature neurons [50], L-type Ca2+ channels may be partially opened at rest. The depolarization of neuronal precursors/immature neurons may be induced by cellular milieu, such as elevated extracellular [K+ ] during

local neuronal activity or voltage changes in active neighboring neurons [24]. In addition, the neuronal precursors may be depolarized by activation of plasma membrane receptors [14,15,19,22,47]. Activation of GABAA receptors opens the ligand-gated chloride channels and causes Cl− influx in most mature neurons to induce membrane hyperpolarization. However, opening of the same chloride channel by GABA in immature neurons may cause Cl− efflux and membrane depolarization due to higher intracellular Cl− concentration in

430

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431

these neurons [3,30]. Like neural progenitors from embryonic brain [22,30,31] or early postnatal brain [19,35,44], neural precursors from ES cells also express functional GABAA receptors (Fig. 8) that, when activated, induced intracellular Ca2+ elevation and promoted neurogenesis through activation of L-type Ca2+ channels (Fig. 9). The result that application of GABA promoted neurogenesis and bicuculline inhibited neurogenesis (Fig. 9) suggests involvement of tonic activation of GABAA receptors. Thus, the excitation-neurogenesis coupling described in adult neural stem cells involving activation of GABAA /NMDA receptors and L-type Ca2+ channels [14,15,47] is also responsible for the GABA-induced neurogenesis from ES cells. This is consistent with the observation that direct effect of GABA on neural progenitors tends to increase neurogenesis [14,19,22], although some in vivo or organotypic culture studies showed opposite effect by using systemic pharmacological treatments [22,30,35]. The latter phenomenon may be caused by complicated inhibitory effects of GABA on surrounding mature neurons, which may have an indirect inhibition on neurogenesis from neural precursors due to the decreased excitatory input from these mature neurons [14]. Unlike adult neural stem cells or postnatal neural precursors receiving various inputs (either synaptic or non-synaptic) from surrounding mature neurons, activation of GABAA receptors in neural precursors derived from ES cells may be induced by non-synaptic paracrine and/or autocrine GABA released from precursors and/or immature neurons themselves, similar to that described previously in embryonic [32] and more recently in adult [29] neural precursors/immature neurons. The most interesting finding in the present study is that neurogenesis induced by activation of GABAA receptors and L-type Ca2+ channels depends critically on the functional RyR2. Of the three RyR isoforms expressed in adult brain, RyR2 is the most abundant [4]. Furthermore, the expression of RyR2 protein, which is detectable as early as the onset of neurogenesis, is earlier and higher than the other two isoforms of RyRs in the brain [17], suggesting important roles of RyR2 in the activity-dependent neurogenesis in the brain. Consistent with this hypothesis, we found that although caffeine and ryanodine still evoked low level Ca2+ activity in RyR2−/− cultures (Fig. 5), this low level Ca2+ activity mediated probably by other subtypes of RyRs may not involved in the activity-induced neurogenesis in ES cells. Thus, antagonist (nifedipine) and agonist (BayK 8644) of L-type Ca2+ channels were effective in inhibiting and promoting neurogenesis, respectively, only in wild-type, but not in RyR2−/− cultures (Fig. 6). Similarly, inhibition and promotion of neurogenesis induced by bicuculline and GABA only occurred in wild-type, but not in RyR2−/− cultures (Fig. 9). Furthermore, no significant difference was found between two types of cultures in the expression of GABAA receptors (Fig. 8A and B), the density of GABA-induced inward current (Fig. 8C), or the density of Ca2+ current (Fig. 7D). These results demonstrate that activity-induced Ca2+ influx through L-type Ca2+ channels, if not powered by CICR mechanism, is not sufficient in

promoting neurogenesis. Moreover, although multiple subtypes of IP3 receptors and RyRs may be expressed in neural precursors, it is RyR2 that plays a key role in mediating the activity-dependent neurogenesis induced by paracrine and/or autocrine GABA signaling.

Conflict of interest None.

Acknowledgements This study was supported in part by grants of the National Natural Science Foundation of China (30270656; 30671045; 30321002), The State Major Research Program from the Ministry of Sciences and Technology, China (2006CB0F0900; G200077800) and Programs (06dj14001) from Science & Technology Committee of Shanghai Municipality, China. We thank Dr. Kenneth R. Boheler (NIH/NIA) for providing us R1 mouse ES cells.

References [1] G. Bain, D. Kitchens, M. Yao, J.E. Huettner, D.I. Gottlieb, Embryonic stem cells express neuronal properties in vitro, Dev. Biol. 168 (1995) 342–357. [2] D. Balschun, D.P. Wolfer, F. Bertocchini, V. Barone, A. Conti, W. Zuschratter, L. Missiaen, H.P. Lipp, J.U. Frey, V. Sorrentino, Deletion of the ryanodine receptor type 3 (RyR3) impairs forms of synaptic plasticity and spatial learning, EMBO J. 18 (1999) 5264–5273. [3] Y. Ben-Ari, Excitatory actions of gaba during development: the nature of the nurture, Nat. Rev. Neurosci. 3 (2002) 728–739. [4] M.J. Berridge, Neuronal calcium signaling, Neuron 21 (1998) 13–26. [5] M.J. Berridge, P. Lipp, M.D. Bootman, The versatility and universality of calcium signaling, Nat. Rev. Mol. Cell Biol. 1 (2000) 11–21. [6] M. Bibel, J. Richter, K. Schrenk, K.L. Tucker, V. Staiger, M. Korte, M. Goetz, Y.A. Barde, Differentiation of mouse embryonic stem cells into a defined neuronal lineage, Nat. Neurosci. 7 (2004) 1003–1009. [7] O. Caillard, Y. Ben-Ari, J.L. Gaiarsa, Activation of presynaptic and postsynaptic ryanodine-sensitive calcium stores is required for the induction of long-term depression at GABAergic synapses in the neonatal rat hippocampus, J. Neurosci. 20 (2000) RC94. [8] M.B. Cannell, H. Cheng, W.J. Lederer, The control of calcium release in heart muscle, Science 268 (1995) 1045–1049. [9] M.B. Carey, S.G. Matsumoto, Spontaneous calcium transients are required for neuronal differentiation of murine neural crest, Dev. Biol. 215 (1999) 298–313. [10] W.A. Catterall, Structure and function of neuronal Ca2+ channels and their role in neurotransmitter release, Cell Calcium 24 (1998) 307–323. [11] P. Chavis, L. Fagni, J.B. Lansman, J. Bockaert, Functional coupling between ryanodine receptors and L-type calcium channels in neurons, Nature 382 (1996) 719–722. [12] F. Ciccolini, T.J. Collins, J. Sudhoelter, P. Lipp, M.J. Berridge, M.D. Bootman, Local and global spontaneous calcium events regulate neurite outgrowth and onset of GABAergic phenotype during neural precursor differentiation, J. Neurosci. 23 (2003) 103–111. [13] K. Deisseroth, E.K. Heist, R.W. Tsien, Translocation of calmodulin to the nucleus supports CREB phosphorylation in hippocampal neurons, Nature 392 (1998) 198–202.

H.-M. Yu et al. / Cell Calcium 43 (2008) 417–431 [14] K. Deisseroth, R.C. Malenka, GABA excitation in the adult brain: a mechanism for excitation-neurogenesis coupling, Neuron 47 (2005) 775–777. [15] K. Deisseroth, S. Singla, H. Toda, M. Monje, T.D. Palmer, R.C. Malenka, Excitation-neurogenesis coupling in adult neural stem/progenitor cells, Neuron 42 (2004) 535–552. [16] R.T. Dirksen, Bi-directional coupling between dihydropyridine receptors and ryanodine receptors, Front Biosci. 7 (2002) d659–d670. [17] A.V. Faure, D. Grunwald, M.J. Moutin, M. Hilly, J.P. Mauger, I. Marty, M. De Waard, M. Villaz, M. Albrieux, Developmental expression of the calcium release channels during early neurogenesis of the mouse cerebral cortex, Eur. J. Neurosci. 14 (2001) 1613–1622. [18] M.L. Fiszman, L.N. Borodinsky, J.H. Neale, GABA induces proliferation of immature cerebellar granule cells grown in vitro, Brain Res. Dev. Brain Res. 115 (1999) 1–8. [19] N. Gago, M. El-Etr, N. Sananes, F. Cadepond, D. Samuel, V. vellana-Adalid, E.A. Baron-Van, M. Schumacher, 3alpha, 5alphaTetrahydroprogesterone (allopregnanolone) and gamma-aminobutyric acid: autocrine/paracrine interactions in the control of neonatal PSANCAM+ progenitor proliferation, J. Neurosci. Res. 78 (2004) 770–783. [20] X. Gu, E.C. Olson, N.C. Spitzer, Spontaneous neuronal calcium spikes and waves during early differentiation, J. Neurosci. 14 (1994) 6325–6335. [21] K. Guan, H. Chang, A. Rolletschek, A.M. Wobus, Embryonic stem cellderived neurogenesis. Retinoic acid induction and lineage selection of neuronal cells, Cell Tissue Res. 305 (2001) 171–176. [22] T.F. Haydar, F. Wang, M.L. Schwartz, P. Rakic, Differential modulation of proliferation in the neocortical ventricular and subventricular zones, J. Neurosci. 20 (2000) 5764–5774. [23] J. Holliday, R.J. Adams, T.J. Sejnowski, N.C. Spitzer, Calcium-induced release of calcium regulates differentiation of cultured spinal neurons, Neuron 7 (1991) 787–796. [24] J.G. Jefferys, Nonsynaptic modulation of neuronal activity in the brain: electric currents and extracellular ions, Physiol. Rev. 75 (1995) 689–723. [25] J.-D. Fu, J. Li, D. Tweedie, H.-M. Yu, L. Chen, R. Wang, D.R. Riordon, S.A. Brugh, S.-Q. Wang, K.R. Boheler, H.-T. Yang, Crucial role of the sarcoplasmic reticulum in the developmental regulation of Ca2+ transients and contraction in cardiomyocytes derived from embryonic stem cells, FASEB J. 20 (2006) 181–183. [26] S.M. Jones, A.B. Ribera, Overexpression of a potassium channel gene perturbs neural differentiation, J. Neurosci. 14 (1994) 2789–2799. [27] A.R. Kriegstein, GABA puts the brake on stem cells, Nat. Neurosci. 8 (2005) 1132–1133. [28] U. Lendahl, L.B. Zimmerman, R.D. McKay, CNS stem cells express a new class of intermediate filament protein, Cell 60 (1990) 585–595. [29] X. Liu, Q. Wang, T.F. Haydar, A. Bordey, Nonsynaptic GABA signaling in postnatal subventricular zone controls proliferation of GFAP-expressing progenitors, Nat. Neurosci. 8 (2005) 1179–1187. [30] J.J. LoTurco, D.F. Owens, M.J. Heath, M.B. Davis, A.R. Kriegstein, GABA and glutamate depolarize cortical progenitor cells and inhibit DNA synthesis, Neuron 15 (1995) 1287–1298. [31] Z. Mao, A. Bonni, F. Xia, M. Nadal-Vicens, M.E. Greenberg, Neuronal activity-dependent cell survival mediated by transcription factor MEF2, Science 286 (1999) 785–790. [32] D. Maric, I. Maric, J.L. Barker, Developmental changes in cell calcium homeostasis during neurogenesis of the embryonic rat cerebral cortex, Cereb. Cortex 10 (2000) 561–573. [33] P.G. Mermelstein, H. Bito, K. Deisseroth, R.W. Tsien, Critical dependence of cAMP response element-binding protein phosphorylation on L-type calcium channels supports a selective response to EPSPs in preference to action potentials, J. Neurosci. 20 (2000) 266–273.

431

[34] K. Narita, T. Akita, J. Hachisuka, S. Huang, K. Ochi, K. Kuba, Functional coupling of Ca(2+) channels to ryanodine receptors at presynaptic terminals. Amplification of exocytosis and plasticity, J. Gen. Physiol. 115 (2000) 519–532. [35] L. Nguyen, B. Malgrange, I. Breuskin, L. Bettendorff, G. Moonen, S. Belachew, J.M. Rigo, Autocrine/paracrine activation of the GABA(A) receptor inhibits the proliferation of neurogenic polysialylated neural cell adhesion molecule-positive (PSA-NCAM+) precursor cells from postnatal striatum, J. Neurosci. 23 (2003) 3278–3294. [36] E. Rios, G. Pizarro, Voltage sensor of excitation-contraction coupling in skeletal muscle, Physiol. Rev. 71 (1991) 849–908. [37] S.E. Ross, M.E. Greenberg, C.D. Stiles, Basic helix–loop–helix factors in cortical development, Neuron 39 (2003) 13–25. [38] R. Rozental, M.F. Mehler, M. Morales, A.F. Andrade-Rozental, J.A. Kessler, D.C. Spray, Differentiation of hippocampal progenitor cells in vitro: temporal expression of intercellular coupling and voltage- and ligand-gated responses, Dev. Biol. 167 (1995) 350–362. [39] M.F. Schneider, Control of calcium release in functioning skeletal muscle fibers, Annu. Rev. Physiol. 56 (1994) 463–484. [40] M.H. Schwab, A. Bartholomae, B. Heimrich, D. Feldmeyer, S. Druffel-Augustin, S. Goebbels, F.J. Naya, S. Zhao, M. Frotscher, M.J. Tsai, K.A. Nave, Neuronal basic helix-loop-helix proteins (NEX and BETA2/Neuro D) regulate terminal granule cell differentiation in the hippocampus, J. Neurosci. 20 (2000) 3714–3724. [41] V. See, A.L. Boutillier, H. Bito, J.P. Loeffler, Calcium/calmodulindependent protein kinase type IV (CaMKIV) inhibits apoptosis induced by potassium deprivation in cerebellar granule neurons, FASEB J. 15 (2001) 134–144. [42] J.S. Sham, L. Cleemann, M. Morad, Functional coupling of Ca2+ channels and ryanodine receptors in cardiac myocytes, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 121–125. [43] H. Song, C.F. Stevens, F.H. Gage, Astroglia induce neurogenesis from adult neural stem cells, Nature 417 (2002) 39–44. [44] R.R. Stewart, G.J. Hoge, T. Zigova, M.B. Luskin, Neural progenitor cells of the neonatal rat anterior subventricular zone express functional GABA(A) receptors, J. Neurobiol. 50 (2002) 305–322. [45] H. Takeshima, S. Komazaki, K. Hirose, M. Nishi, T. Noda, M. Iino, Embryonic lethality and abnormal cardiac myocytes in mice lacking ryanodine receptor type 2, EMBO J. 17 (1998) 3309–3316. [46] E.C. Toescu, A. Verkhratsky, Neuronal ageing from an intraneuronal perspective: roles of endoplasmic reticulum and mitochondria, Cell Calcium 34 (2003) 311–323. [47] Y. Tozuka, S. Fukuda, T. Namba, T. Seki, T. Hisatsune, GABAergic excitation promotes neuronal differentiation in adult hippocampal progenitor cells, Neuron 47 (2005) 803–815. [48] R.W. Tsien, D. Lipscombe, D. Madison, K. Bley, A. Fox, Reflections on Ca(2+)-channel diversity, 1988–1994, Trends Neurosci. 18 (1995) 52–54. [49] A. Verkhratsky, The endoplasmic reticulum and neuronal calcium signalling, Cell Calcium 32 (2002) 393–404. [50] U. Westerlund, M.C. Moe, M. Varghese, J. Berg-Johnsen, M. Ohlsson, I.A. Langmoen, M. Svensson, Stem cells from the adult human brain develop into functional neurons in culture, Exp. Cell Res. 289 (2003) 378–383. [51] H.T. Yang, D. Tweedie, S. Wang, A. Guia, T. Vinogradova, K. Bogdanov, P.D. Allen, M.D. Stern, E.G. Lakatta, K.R. Boheler, The ryanodine receptor modulates the spontaneous beating rate of cardiomyocytes during development, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 9225–9230. [52] R. Zucchi, S. Ronca-Testoni, The sarcoplasmic reticulum Ca2+ channel/ryanodine receptor: modulation by endogenous effectors, drugs and disease states, Pharmacol. Rev. 49 (1997) 1–51.