The role of ATP signaling in the migration of intermediate neuronal progenitors to the neocortical subventricular zone

Abstract

Most neurons of the cerebral cortex are generated in the germinal zones near the embryonic cerebral ventricle and migrate radially to the overlying cortical plate. Initially, all dividing cells are attached to the surface of the embryonic ventricle (ventricular zone) until a subset of dividing cells (basal or intermediate neuronal progenitors, INPs), recognized by their immunoreactivity to Tbr2, detach from the ventricular surface and migrate a short distance to establish a secondary proliferative compartment (the subventricular zone). The mechanism that regulates migration of the Tbr2⁺ INPs from the ventricular to the subventricular zones is unknown. Here, we show that INPs, unlike the postmitotic neurons that tend to lose the ATP response, continue to express the purinergic P2Y1 receptor. Furthermore, blocking ATP signaling by the P2Y1 blockers, MRS2176, suramin, and apyrase, reduces Ca²⁺ transients and retards INP migration to the subventricular zone. In addition, genetic knockdown of the P2Y1 receptor by in vivo application of short hairpin RNA selectively impairs the migration of INPs to the subventricular zone. Together, these results suggest that intercellular ATP signaling is essential for the migration of INPs and the proper formation of the subventricular zone. Interference of ATP signaling or abnormal Ca²⁺ fluctuations in INPs may play a significant role in variety of genetic or acquired cortical malformations.

For almost four decades we have known that neurons destined for the cerebral cortex in all mammalian species, including the human, are generated in the primary and secondary proliferative layers situated near the embryonic cerebral ventricle, respectively called the ventricular (VZ) and subventricular zones (SVZ) (1, 2). It is also generally recognized that the SVZ is formed by a subclass of neuronal progenitors that lose their attachment to the ventricular surface and translocate their nucleus and surrounding somatic cytoplasm to the SVZ, where they continue to divide and produce neurons destined predominantly for the upper cortical layers (3–5). The use of in vitro imaging has revealed that the SVZ progenitors (recently renamed the intermediate neuronal progenitors or INPs, to emphasize their continuous proliferation), originate in the VZ by asymmetric mitotic division of the neuroepithelial cells (6). After translocation of their soma to the SVZ, INPs amplify production of cortical neurons via asymmetrical and symmetric modes of mitotic divisions (6–9). The relative large size of the SVZ in primates, particularly in humans, has inspired suggestions that cells in this zone contribute to the expansion and elaboration of the neocortex during evolution (2, 10, 11). Furthermore, genetic and environmental disturbances of neuronal production in the SVZ affect the development and function of the cerebral cortex (5, 12–14).

Considering the developmental and clinical significance of the SVZ, surprisingly little is known about its origin and how INPs are distinguished from other proliferative neural precursors and postmitotic migrating neurons. Similar to the postmitotic neurons, INPs leave the VZ by extending their basal processes toward the cortical plate (CP) along radial glia (RG) shafts and translocating their nuclei within the growing leading processes (1, 6). However, they stop migration after reaching the level of the SVZ, where they re-enter the cell cycle. It has been found that genes, such as p27, Ngn2, FLNa, and MEKK4 are involved in the final mitotic division and onset of migration from the VZ via modulation of the actin cytoskeleton (15–17). Reelin (18) and the neurotransmitters GABA and glutamate and their receptors might also contribute to this process (19, 20). However, the nature of the complex intercellular signaling involved in the migration and settlement of INPs in the SVZ remains unexplored.

We started with the assumption that Ca²⁺, a ubiquitous secondary messenger, and ATP released via gap junctions and hemichannels in the VZ might be involved in the cell-cell communication and the initiation of INP migration. Our hypothesis was based on the previous findings that Ca²⁺ signaling influences the mitosis of neuronal progenitors and their nuclear interkinetic translocation in the embryonic retina (21), and that [Ca²⁺] fluctuations also affect cellular migration in cerebellar granular cells (22–24), cultured neuronal progenitors (25), and postmitotic cortical neurons (26). Spontaneous Ca²⁺ transients or Ca²⁺ waves are visible in both the VZ and SVZ and increase in intensity during corticogenesis (27, 28), suggesting their involvement in neuronal proliferation and migration. It has also been shown that embryonic VZ cells express connexins within radial clusters containing RG and neuronal precursors that are interconnected via gap junctions (29, 30). Finally it has been reported that gap junction and hemichannel-dependent ATP signaling is related to the Ca²⁺ waves and that disrupting Ca²⁺ waves prohibits the cells' entry into S-phase in the embryonic VZ (28). In the present study we tested the hypothesis that intercellular ATP signaling plays a role in the migration of INPs and the formation of the SVZ in the embryonic cerebral wall.

Results

The SVZ Cells Originate and Migrate from the VZ.

Although it has been well established that the SVZ in primates produces neocortical neurons that migrate radially to the overlying cortical plate (5, 10, 14), recent evidence in E13 to E14 mouse embryos suggests that this secondary proliferative zone also contributes neurons to the cortical plate (9). In the present study, we confirmed this finding and, in addition, found that the SVZ cells synthesize DNA and undergo mitotic divisions even at the later stages of corticoneurogenesis (E15–E16) (Fig. 1 A and B), suggesting that this zone continues to contribute neurons to the cerebral cortex throughout the entire period of neurogenesis. We used in utero electroporation to transfect cells near the cerebral ventricle with plasmids expressing EGFP to demonstrate that SVZ cells originate from the VZ (see Fig. 1B). Previous observations in the cultured embryonic cerebrum have shown that the newly generated neurons in the SVZ usually display multiple processes and undergo retrograde movements toward the ventricle before starting radial migration to the CP (6). By using time-lapse live imaging in situ, we have observed the whole process of cell division in the VZ and migration to the SVZ followed by additional division before they begin to migrate toward the CP [illustration diagram in Fig. 1C; for the time-lapse movie see the supporting information (SI)]. Distinguished from the presumably postmitotic neurons, which usually wait for a while before their migration, the INPs migrate out of the VZ directly after their division. Our observations revealed that INPs display distinct behavior during their radial migration to the SVZ, which might be modulated by local intercellular signaling.

Fig. 1.

The proliferating SVZ cells originate from the VZ. (A) Two hours after BrdU injection (i.p. 50 mg/kg), BrdU-labeled cells (green) in the embryonic mouse neocortex locate at both the VZ and the SVZ (E16), which suggest these cells were in the S-phase. The contour of the neocortex is shown by Tuj-1 staining (red). Scale bar: 15 μm. (B) The plasmid expressing EGFP (pCAG-EGFP) was electroporated into the VZ cells at the ventricular surface at E14. Immunostaing for M phase marker phosphor-Histone-H3 (pH3) was performed at E15. pH3⁺ cells are observed in the SVZ and are colocalized with EGFP (arrows), which suggests these proliferating SVZ cells originate from the VZ. Scale bar: 10 μm. (C) Schematic diagram shows the behavior of INPs in the SV/SVZ. The M phase cells at the ventricular surface generate INPs via asymmetric division. The INP migrates to the SVZ to generate two postmitotic neurons (PNs) via symmetric division. ATP released in the VZ/SVZ activates the P2Y1 receptor expressed in VZ/SVZ cells, including the RGs and INPs. Activation of P2Y1 receptor in INPs will induce Ca²⁺ transients and modulate their migration from the VZ to the SVZ.

P2Y1 Receptor Is Expressed in the Migrating INPs.

To demonstrate that ATP signaling is involved in the neuronal migration from the VZ to SVZ, we first determined whether the ATP receptor is expressed in the mouse embryonic VZ/SVZ at the appropriate embryonic ages (Fig. 2A), as has been reported in rats (28). Indeed, intensive purinergic P2Y1 receptor expression was observed at the interface of the VZ and SVZ at E14 to E16 while the ventricular surface showed a weaker expression. Likewise, the expression of P2Y1 receptor was reduced in the most superficial portion of the SVZ and further diminished toward the pial surface. Both INPs and postmitotic neurons are generated by neural precursors at the ventricular surface via asymmetric divisions and then migrate out of the VZ (6, 8). To distinguish the INPs, we immunostained for Ngn2, a specific marker for INPs and newborn postmitotic neurons in the VZ (15). As shown in Fig. 2B, Ngn2⁺ cells located at the upper portion of the VZ expressed the P2Y1 receptor. Few Tuj1⁺ cells appeared in the VZ/SVZ and most of them are P2Y1-negative, suggesting that the expression of the P2Y1 receptor is down-regulated in the postmitotic neurons of the VZ/SVZ as they commence their journey to the cortex.

Fig. 2.

The P2Y1 receptor is expressed in the migrating INPs in the VZ/SVZ. Immunostainings were performed in embryonic mouse neocortex (E16). (A) Immunostaining for P2Y1 and Tuj-1. P2Y1 expression (green) mainly locates at the interface of the VZ/SVZ. The upper portion of the VZ shows more intense expression than the lower portion. The P2Y1 expression in the SVZ is gradually reduced toward the upper portion. Tuj-1 (red) is mainly expressed in the CP and IZ. Some cells in the SVZ but very few in the VZ express Tuj-1. Scale bar: 30 μm. IZ, intermedial zone; MZ, marginal zone. (B) Immunostaining for P2Y1 and Ngn2. Ngn2⁺ cells (red) mainly locate at the upper portion of the VZ and in the SVZ, and coexpress P2Y1 (green) as indicated by the arrows. Scale bar: 10 μm.

Because the P2Y1 receptor is highly expressed in the VZ/SVZ cells, we decided to explore whether ATP can induce Ca²⁺ responses in these cells. We found that application of ATP (10 μM) induced strong Ca²⁺ responses in the VZ/SVZ cells (Fig. S1A) while the ATP receptor blocker suramin (100 μM) totally blocks these responses (Fig. S1B). Furthermore, these Ca²⁺ responses were also blocked by the P2Y1 receptor blocker MRS2179 (data not shown). ATP (1–10 μM) induced Ca²⁺ responses were repeatable with little sign of desensitization, but at higher concentration ATP (>100 μM) usually desensitized the Ca²⁺ response as previously reported (28). The responsive cells were located in both the VZ and SVZ and accounted for ≈40 to 60% of the total cell populations in these proliferative zones.

ATP Receptors Are Involved in Ca²⁺ Signaling in the VZ/SVZ Cells.

We recorded spontaneous Ca²⁺ transients in 60 to 80% of the cells in the VZ/SVZ in situ (Fig. 3A) similar to, but with higher frequency than that reported in rats (28). The Ca²⁺ transients showed a mean frequency of 8.0 peaks/10 min with a mean duration of 12 s (4–20 s). To determine whether ATP receptors are involved in the spontaneous Ca²⁺ transients in the VZ/SVZ cells, we measured Ca²⁺ transients before and during the application of ATP receptor antagonists. We found that suramin application reduced the frequency of the spontaneous Ca²⁺ transients in the VZ/SVZ cells, but could not completely abolish them (see Fig. 3A). Cells with remaining Ca²⁺ transients usually displayed increased amplitude and prolonged duration. The mean values of the frequency and amplitude were changed to 58.0 ± 10.6% (P < 0.01, n = 5, slices) and 99.2 ± 14.9% (n = 5), respectively, after suramin application (Fig. 3C). Similarly, the specific P2Y1 receptor blocker MRS2179 (50 μM) reduced the frequency to 70.9 ± 6.6% (P < 0.05, n = 4) (see Fig. 3C).

Fig. 3.

ATP receptor is involved in the Ca²⁺ signaling in VZ/SVZ cells. (A) Traces show representative Ca²⁺ transients in 10 cells before and during suramin application. The frequency of Ca²⁺ transients in VZ/SVZ cells was reduced and some cells exhibited transients with increased amplitude and longer duration. (B) Forcal stimulation evoked Ca²⁺ wave propagation in the VZ/SVZ in controls and in the presence of MRS2179. The arrows indicate the sites of stimulation. Scale bars: 20 μm. (C) Changes of the mean frequency and amplitude of Ca²⁺ transients in VZ/SVZ cells before and during the application of suramin (100 μM; **, P < 0.01, n = 5) and P2Y1 receptor antagonist MRS2179 (50 μM; *,P < 0.05, n = 4). (D) Effect of ATP receptor antagonists on Ca²⁺ waves evoked by stimulation in the embryonic mouse neocortex. Propagated distances of Ca²⁺ wave within the VZ/SVZ before and during the application of suramin (100 μM; **, P < 0.01, n = 4) and MRS2179 (50 μM; **, P < 0.01, n = 8).

Spontaneous Ca²⁺ waves have been observed in the embryonic cerebral cortex in vivo (28). To demonstrate whether the P2Y1 receptor is involved in the propagation of Ca²⁺ waves in the VZ/SVZ cells, we used focal stimulation-evoked Ca²⁺ waves as previously reported (28). Focal stimulation at the ventricular surface induced Ca²⁺ wave propagation both radially and tangentially in the embryonic mouse VZ/SVZ (Fig. 3B). The tangentially propagating Ca²⁺ waves showed an average distance of 99.2 ± 10.4 μm (n = 7) with an average velocity of 9.0 ± 0.9 μm/s (n = 7). Blocking P2Y1 receptor with MRS2179 (50 μM) abolished the Ca²⁺ wave propagation and significantly reduced the propagating distance to 12.4 ± 2.0 μm (n = 8, P < 0.01) (see Fig. 3 B, and D). Similar results were observed with suramin (100 μM), in which the propagating distance was reduced to 6.8 ± 3.4 μm (n = 6, P < 0.01) (see Fig. 3D).

Blocking ATP Signaling Impairs the Cellular Migration in the VZ/SVZ.

We next tested whether ATP signaling is involved in the cellular migration from the VZ to SVZ. The cells in S-phase were labeled by short-term (2 h) BrdU incorporation (E16, 50 mg/kg, i.p.) and BrdU⁺ cells were detected throughout the VZ/SVZ (see Fig. 1A). BrdU-labeled brain slices were then cultured in medium with or without ATP receptor antagonists, suramin (100 μM), MRS2179 (50 μM), or IP3 receptor blocker 2-APB (100 μM). After 24 h, most BrdU⁺ cells left the ventricular surface and migrated into the SVZ in control slices, and some BrdU⁺ cells appeared in the IZ (Fig. 4A). In contrast, BrdU⁺ cells were mainly located within the VZ in the suramin- or MRS2179-treated slices and many of them remained at the ventricular surface (see Fig. 4A). Similar results were observed in the 2-APB treated slices. Fig. 4B shows the distribution of BrdU⁺ cells according to their migrated distance relative to the ventricular surface (from VZ to IZ).

Fig. 4.

Blocking ATP signaling impaired the cellular migration in the VZ/SVZ. (A) BrdU-labeled brain slices were cultured for 24 h with vehicle (control), ATP receptor, or IP3 receptor antagonists. In the control slice most BrdU⁺ cells (green) left the ventricular surface and migrated into the SVZ. Some BrdU⁺ cells entered into the IZ. In suramin (100 μM), MRS2179 (50 μM), or 2-APB (100 μM) treated slices, BrdU⁺ cells mainly located within the VZ and many of them remained on the ventricular surface. Scale bars: 20 μm. (B) The distribution of BrdU⁺ cells according to their migrated distances from the ventricular surface (from the VZ to the IZ) in control (n = 4 slices), suramin (n = 5), MRS2179 (n = 4), and 2-APB (n = 4) treated slices. *, P < 0.05; **, P < 0.01 versus control. (C) BrdU-labeled brain slices were cultured for 48 h with or without apyrase (80 u/ml). In the control, many BrdU⁺ cells migrated into the IZ and few BrdU⁺ cells remained in the VZ. In the apyrase incubated slice BrdU⁺ cells mainly remained in the VZ/SVZ and many BrdU⁺ cells are still detected in the lower portion of the VZ. The ventricular surface locates at the bottom of the images. Scale bars: 15 μm. (D) The distribution of BrdU⁺ cells according to their migrated distances from the ventricular surface (from the VZ to the pial surface) in control (n = 13) and apyrase (n = 6) treated slices. *, P < 0.05; **, P < 0.01 versus control.

We further tested if scavenging the extracellular ATP with apyrase, which has been used to block ATP signaling in retinal slices (31), could affect the cell migration in the VZ/SVZ. BrdU-labeled brain slices (E15) were incubated with and without apyrase (80 U/ml) for 48 h. In control slices, many BrdU⁺ cells migrated into the IZ, and few BrdU⁺ cells remained in the VZ (Fig. 4C). In contrast, the BrdU⁺ cells mainly remained in the VZ/SVZ and many of them were found in the lower portion of the VZ in the apyrase treated slices (see Fig. 4C). The distribution of the BrdU⁺ cells according to their migrated distance across the whole neocortex (from the VZ to the pial surface) is shown in Fig. 4D.

Both postmitotic neurons and INPs were produced via asymmetric division at the ventricular surface of the VZ and it has been observed that blocking the ATP signaling prolonged the mitosis in retinal neuronal progenitors (21). As a first step we tested whether the BrdU⁺ cells retained in the VZ are precursors that failed to pass the M phase. Shortly after BrdU incorporation (2 h in vivo), all BrdU⁺ cells in the VZ were colabeled with Ki67, a cell proliferation marker (Fig. S2A). After incubation for 24 h, many BrdU⁺ cells remaining in the VZ became Ki67⁻ in MRS2179 treated slices (Fig. S2B), which suggests these BrdU⁺ cells had passed the M phase. The percentage of Ki67⁻ cells in BrdU⁺ cells in the VZ was 65% in both suramin and MRS2179 treated slices. We next tested whether the BrdU⁺ cells remaining in the VZ are INPs. As shown in Fig. S2C, many BrdU⁺ cells in the VZ were Tbr2⁺ in slices treated with apyrase (48 h), which accounts for 36% of the BrdU⁺ cells in the VZ. These results suggest the retarded cells in the VZ included the INPs.

To exclude the migration defects were due to cell apoptosis, TUNEL staining was performed (Fig. S2D). The numbers of TUNEL⁺ cells in control and in slices treated with suramin and 2-APB (0.97 ± 0.14 and 0.97 ± 0.12 per 10,000 μm², respectively) were in the range of that reported in cultured organotypic slices (32). Furthermore, RC2 immunostaining showed that there was no obvious defect in RG shafts in the slices incubated with ATP or IP3 receptor antagonists (Fig. S2E).

P2Y1 Receptor Knock Down Interferes with the Migration of INPs from the VZ to the SVZ.

P2Y1 receptor was expressed in the INPs of the VZ/SVZ and sharply diminished toward the IZ and CP (see Fig. 2A). It has previously been reported that interfering with ATP signaling has no obvious effects on neuronal migration into the CP (33). Here, we tested whether P2Y1 receptor is involved in the migration of INPs within the VZ/SVZ. The plasmid expressing short hairpin RNA (shRNA) for P2Y1 receptor or control scrambled shRNA was electroporated with an EGFP expression plasmid (pCAG-EGFP) into the embryonic cerebrum (E14) and the EGFP expression was detected at E15.5. As shown in Fig. 5A, most EGFP⁺ cells migrated into the upper portion of the SVZ and some of them entered the IZ in the control scrambled shRNA electroporated cerebrum. In contrast, in the P2Y1 shRNA electroporated cerebrum most EGFP⁺ cells were located at the ventricular portion of the SVZ and there were more EGFP⁺ cells remaining in the VZ. The distribution of EGFP⁺ cells according to their migrated distance from the VZ to the IZ is shown in Fig. 5B.

Fig. 5.

P2Y1 receptor is involved in the migration of INPs in the VZ/SVZ. The plasmids expressing P2Y1 shRNA or scrambled control shRNA were electroporated with the pCAG-EGFP into the embryonic neocortex at E14 and the cell migration was analyzed at E15.5. (A) In control shRNA electroporated neocortex, most EGFP⁺ cells (green) migrated to the SVZ and some were found in the IZ. Tbr2 immunostaining (red) shows that Tbr2⁺/EGFP⁺ cells were around the border of the SVZ and IZ. In P2Y1 shRNA electroporated neocortex, more EGFP⁺ cells remained in the lower portion of the SVZ and some of them stayed in the VZ. Many Tbr2⁺/EGFP⁺ INPs are stacked in the lower portion of the SVZ. Scale bars: 20 μm. (B) The distribution of EGFP⁺ cells according to their positioning from the ventricular surface (from the VZ to the IZ) in scrambled control and P2Y1 shRNA electroporated neocortex. **, P < 0.01, n = 8, 7 respectively, versus scrambled control. (C) P2Y1 expression in control transfected and P2Y1 shRNA (2 mg/ml) transfected GL261 cells. (D) Normalized P2Y1 expression in control, 05.mg/ml and 2 mg/ml P2Y1 shRNA transfected groups. *, P < 0.05; **, P < 0.01 versus control, n = 6, 4, 5, respectively.

To confirm the nature of progenitors under study, we used Tbr2 as a cell class-specific marker for the INPs in the SVZ (7). Compared to the control scrambled shRNA electroporated neocortex or the contralateral hemisphere, we found that in the P2Y1 shRNA electroporated hemisphere, Tbr2⁺ cells were accumulated in the lower portion of the SVZ (see Fig. 5A, Fig. S3 A and B). Because the distribution of Tbr2⁺ cells might vary at different regions of the neocortex, we normalized the Tbr2⁺ cell distribution to that of the corresponding region of the contralateral hemisphere and then compared the difference between the P2Y1 shRNA and control shRNA electroporated brain. As shown in Fig. S3C, the relative indexes of Tbr2⁺ cells were around 1.0 in all bins in control electroporated brain, while it was increased at the upper portion of the VZ and decreased at the upper portion of the SVZ in the P2Y1 shRNA electroporated brain. These results suggest that P2Y1 receptor-mediated ATP signaling is necessary for the proper migration of INPs in the VZ/SVZ. To further confirm shRNA-dependent knockdown of P2Y1, the GL261 cells were transfected with plasmid expressing P2Y1 shRNA. Western blot analysis showed that P2Y1 expression was significantly reduced (Fig. 5 C and D). Finally, we demonstrated that the migration defect in P2Y1 shRNA electroporated brain was caused by neither increased apoptosis nor abnormal RG scaffolds (Fig. S3 D and E).

Discussion

Intracellular Ca²⁺ signaling has been implicated in many aspects of nervous system development, including cell proliferation (28, 31), differentiation (20), migration (23, 24, 34), neurite outgrowth and growth cone behavior, as well as synaptic genesis and plasticity. In the present study we focused on spontaneous Ca²⁺ transients in the INPs that form the SVZ in situ, and found that they are associated with intercellular ATP signaling. Because the Ca²⁺ fluctuations might result from both Ca²⁺ influx or release from the endoplasmic reticulum (23, 27), our results cannot exclude the possibility that GABA, glutamate, glycin, and other signaling molecules may also be involved. However, it is clear that the Ca²⁺ wave propagation, which spontaneously occurs in the embryonic cerebral cortex in vivo (28), depends largely on the ATP signaling. Migrating INPs are known to dramatically alter their morphology shortly after exit from the asymmetric division (1, 6, 35), which requires substantial changes in gene transcription/translation. For example, the bHLH transcription factors are up-regulated in migrating neuronal progenitors/postmitotic neurons (15). Thus, the intercellular ATP signaling might alter expression levels of these genes by modulating the intracellular Ca²⁺ signaling.

Cells in the embryonic VZ are extensively coupled in clusters (90–115 cells) via gap junctions (30, 36), which provide direct passage for metabolic and electrical communication among coupled cells that form radial columns (1, 14). Thus, the coupled cells in the VZ can act as a syncytium and synchronize their behaviors. The coupled cells can also communicate with surrounding cells by releasing bioactive molecules, such as ATP and glutamate, via the undocked connexons (hemichannels) (31). Previous studies have shown that proliferative cells in the VZ/SVZ express connexins (29, 37). Furthermore, the proliferative cells in the S and G2 phases are gap junction coupled, while the M phase cells and the migrating postmitotic neurons become uncoupled (38). Therefore, ATP released from the neural precursors via gap junctions/hemichannels might activate the P2Y1 receptor and induce Ca²⁺ fluctuations in neuronal progenitors at the time they leave the VZ and begin migrating. Precise control over the number of proliferating neural precursors in the VZ and the number of cells exiting the cell cycle during neurogenesis is essential for proper neocortical growth (39), and it has been observed that gap junction/hemichannel-dependent Ca²⁺ waves are necessary for the proliferation of neural precursors (28). In this study we extended the influence of ATP signaling to the migration of INPs in the VZ/SVZ and propose that paracrine intercellular ATP signaling might be required for the formation of the SVZ, the second neurogenesis center in the forebrain.

Knock down P2Y1 receptor with shRNA altered the distribution of INPs in the SVZ, but showed less migration defect in the VZ, which might be because of the time lag of suppressing P2Y1 expression after electroporation (8). In addition, functional redundancy and compensation of other members of the ATP receptor family might also contribute to this alteration. The expression of the P2Y1 receptors is down-regulated toward the pial surface and ATP application seldom induced Ca²⁺ responses in postmitotic neurons labeled by Dsred expression (use Tα-1 as promoter, X.L. and K.H.-T., unpublished data). Previous work has demonstrated that ATP signaling showed no effect in the migration of the postmitotic neurons in the cerebrum (33). However, it did not give careful attention to the migration within the VZ and the SVZ. Because Ngn2 is expressed in both the INPs and the new generated postmitotic neurons (15), and there is no specific marker to distinguish the INPs and the early stage postmitotic neurons, our results could not exclude the possibility that the BrdU⁺ cells in the VZ included the new generated postmitotic neurons. In addition, recent study has shown that ATP signaling might play a role in the neuronal differentiation and process outgrowth (40), both of which are essential for the initiation of radial migration of the postmitotic neurons.

The proliferative SVZ in the forebrain has a uniquely important role during the second half of intrauterine development in the human brain (5, 10). Compared to the VZ neurogensis component, the SVZ in humans displays more cellular diversities, histological complexity, and protracted period of cell proliferation (5, 10), which makes it vulnerable and susceptible to biogenetic or environmental insult factors. For instance, abnormal myelination in the forebrain observed in periventricular leukomalacia may be caused by selective targeting of oligodendrocyte progenitors that also originate at later stages from the SVZ (41). Considering the essential role of SVZ for the evolutionary expansion of the human cerebral cortex (2), even subtle damage to the SVZ during development may go unrecognized morphologically, but might contribute to complex psychiatric disorders such as schizophrenia, autism, or bipolar disorder (42). Our study indicates that such changes can be induced by genetic or environmental factors, such as ischemia, infections, and drugs that might affect the Ca²⁺ fluctuations at selective developmental stages.

Materials and Methods

Solution and Pharmacological Agents.

Artificial cerebrospinal fluid (ACSF) solution contained (in mM) 125 NaCl, 2.5 KCl, 1.25 NaH₂PO4, 1 MgCl₂, 2 CaCl₂, 25 NaHCO₃, and 10 Glucose. The osmolarity is around 310 mOsm/L and pH is 7.4 when bubbled with 95% O₂/5% CO₂. All experiments were performed at room temperature (≈25°C). The following drugs were used: ATP (0.1–100 μM), suramin (100 μM), MRS 2179 (100 μM), 2-APB (50 μM), apyrase (80U/ml). All are Sigma products.

Tissue Preparation and Slice Culture.

All experimental procedures were in accordance with the animal welfare guidelines of Yale University on the ethical use of animals for experimentation. For the details regarding slice preparation and culture and cell migration assay see SI Text.

Time-Lapse Live Imaging, Calcium Imaging, and Data Analysis.

Time-lapse live imaging was performed as previously reported (43). Calcium imaging was performed as previously described (28, 31). For the details regarding the methods and data analysis, see SI Text.

Plasmid Constructs and in Utero Electroporation.

We used the shRNA Target finder at Genscript Corp to design shRNA sequences against the mouse P2Y1, and the shRNA fragment was inserted into the pRNAT-U6.2/Lenti vector. The targeting sequences are as follows;

• P2Y1 5′-GAGAGATAAAGAACGAGTTTA-3′

• The scrambled shRNA control sequence is 5′-GGATCCCGTACGTTCTGTTGTCTAATCATTTGATATCCGATGATTAGACAACAGAACGTA

• TTTTTTCCAACTCGAG -3′.

The P2Y1 knockdown effect of shRNA was confirmed in GL261 cells by using Western blot after transfection. To analyze the effect of P2Y1 knockdown in vivo, we used in utero electroporation. For more detail see SI Text.

Immunohistochemistry and Immuonocytochemistry.

For immunohistochemistry and immunocytochemistry on postnatal brain slices and cultured embryonic neocortical slices see SI Text.

Statistical Analysis.

Statistical analysis was performed using student's t test or ANOVA test as stated in appropriate experiments. P < 0.05 was considered as significant. Error bars are the standard error of the mean.