Abstract

Cerebral cortical neurons are known to be produced from both apical progenitors in the ventricular zone (VZ) and basal (intermediate) progenitors in the subventricular zone (SVZ). On the other hand, we have shown that many SVZ cells assume multipolar morphology and show a characteristic movement termed “multipolar migration.” The relationship between multipolar cells and basal progenitors in the SVZ has yet to be investigated. Herein, we followed postmitotic cells generated in the VZ and found that they stayed for more than 10 h in the VZ after becoming postmitotic and then accumulated in the lower part of the SVZ (multipolar cell accumulation zone: MAZ) as multipolar cells (slowly exiting population: SEP), whereas basal progenitors rapidly migrated into the SVZ or intermediate zone (IZ) (rapidly exiting population: REP) with somal translocation morphology. Although REP reached the SVZ/IZ earlier than the SEP, REP stayed within in the SVZ/IZ, whereas SEP moved steadily and entered the CP prior to the REP. We also observed SEP to eventually differentiate into pyramidal neurons in layers II/III. This study provides in vivo evidence of differences in migration modes between postmitotic cells generated from apical progenitors and basal progenitors.

Introduction

During the development of the mammalian cerebral cortex, the majority of projection neurons are thought to be born in the ventricular zone (VZ) and then to migrate radially toward the pial surface through the subventricular zone (SVZ), intermediate zone (IZ), and the developing cortical plate (CP). Previous histological analyses revealed that the major population of the SVZ is multipolar cells (Stensaas and Stensaas 1968; Shoukimas and Hinds 1978) and that they differ morphologically from the cells in two known migration modes, locomotion and somal translocation (STL) (Rakic 1972; Miyata et al. 2001; Nadarajah et al. 2001; Tamamaki et al. 2001). The visualization of multipolar cells with GFP using the in utero electroporation system, which we established (Tabata and Nakajima 2001), revealed multipolar cells to show characteristic movements, in which they extended and retracted multiple processes dynamically while advancing slowly toward the CP. We named this migration mode “multipolar migration” (Tabata and Nakajima 2003).

Recently, time-lapse observations and marker analyses have revealed that the mitotic cells within the SVZ or the lower IZ, which are called basal (intermediate) progenitors, secondary proliferative population, or nonsurface dividing cells, also produce neurons (Takahashi et al. 1995; Haubensak et al. 2004; Miyata et al. 2004; Noctor et al. 2004; Englund et al. 2005). In fact, it has been shown that this population, during late cortical development, gives rise to pyramidal neurons in layers II/III in vivo (Wu et al. 2005). Interestingly, several SVZ markers in late cortical development such as svet1 and Cux-1/2 are expressed in layers II/III after birth. Moreover, Pax6-deficient mice, in which svet1- or Cux-1/2–positive SVZ is greatly reduced, show marked reductions of neurons in layers II/III (Tarabykin et al. 2001; Nieto et al. 2004; Zimmer et al. 2004). Brn1/2 double knockout mice and Tlx knockout mice exhibit severely reduced cell proliferation in the SVZ during the late stages of cortical development, and also show reductions of superficial layers in the perinatal stage (Sugitani et al. 2002; Land and Monaghan 2003; Roy et al. 2004). These marker studies and genetic analyses suggest that basal progenitors in the SVZ might be the main source of superficial pyramidal neurons, although they also appear in the early stages of cortical development and contribute to the deep layers (Haubensak et al. 2004; Noctor et al. 2008).

Despite the above-mentioned general view of the SVZ as the secondary proliferating zone, we have seldom found cell divisions during the course of detailed observations of multipolar migrating cells within the SVZ. Elucidating the relationships between multipolar migrating cells and basal progenitors required that the migratory behaviors of SVZ cells and their mitotic activity be examined in detail.

In this study, we labeled VZ cells with GFP using electroporation and further applied thymidine analogues to distinguish the postmitotic cells among all VZ-derived GFP-positive cells. In the initial phase of migration, the direct progenies of VZ cells stayed in the VZ for more than 10 h and accumulated rather specifically in the lower part of the SVZ as multipolar migrating cells. To indicate this zone where postmitotic multipolar migrating cells accumulate, we dubbed this zone the “multipolar cell accumulation zone (MAZ).” The proliferative cells, on the other hand, rapidly exited the VZ assuming STL morphology and distributed within the SVZ or IZ (SVZ/IZ). Because they differ from each other in the timing of exiting the VZ, we termed them the “slowly exiting population (SEP)” and the “rapidly exiting population (REP),” respectively. We found that, whereas the REP entered the SVZ/IZ earlier than the SEP, SEP transformed into locomotion cells and entered the CP earlier than the REP. The SEP eventually differentiated into non–g-aminobutyric acid (GABA)ergic pyramidal neurons in layers II/III.

Materials and Methods

Animals

Pregnant ICR mice were purchased from Japan SLC (Shizuoka, Japan). The day of vaginal plug detection was considered to be embryonic day (E) 0. All animal experiments were performed according to the Guidelines for the Care and Use of Laboratory Animals of Keio University School of Medicine.

In Utero Electroporation

Pregnant mice were deeply anesthetized, and their intrauterine embryos were surgically manipulated as described previously (Nakajima et al. 1997). In utero electroporation was carried out as described previously (Tabata and Nakajima 2001). In brief, pCAGGS vector (Niwa et al. 1991) carrying the enhanced GFP (Clontech, Palo Alto, CA) cDNA was purified using a Qiagen plasmid maxi kit (Hilden, Germany). The concentration of the plasmid was adjusted to 5 mg/mL with HEPES-buffered saline (HBS). The plasmid containing 0.01% Fast Green solution was injected into the lateral ventricle of intrauterine embryos and electronic pulses (30 V, 50 ms, 4 times) were then applied using an electroporator (CUY-21; Nepa gene, Chiba, Japan) with a forceps-type electrode (CUY650P5). The embryos were allowed to live within the uterine horn until the desired time of observation.

Virus Infection and Culture

The replication-incompetent enhanced GFP expressing retrovirus, used for in utero injection, was produced from the vector pMX-GFP (gift from T. Kitamura). The virus production was performed as described previously (Morita et al. 2000; Kitamura et al. 1995). Briefly, we transfected Plat-E cells (gift from T. Kitamura [Morita et al. 2000]) with pMX-GFP using FuGENE6 (Roche Diagnostics, Mannheim, Germany), and collected the medium two days later. The virus was concentrated by centrifugation at 8000 g for 16 h, and the titer was adjusted to 1 × 108 pfu/ml. The virus concentrated stock was complemented with 8 μg/mL polybrene and 0.01% Fast Green solution and injected into the lateral ventricles of E15.5 mouse embryos in utero. To culture the infected cells in vitro, we dissected out the SVZ/IZ above the MAZ, as well as the cortical wall without the CP, followed by dissociation into single cells with 0.25% trypsin and 0.5 mg/mL DNase I. The cells were then cultured on PLL coated cover slips at a density of 5 × 105 cells per well on a 6-well plate the Neurobasal medium + B27 supplement (Invitrogen, Carlsbad, CA). The cultures were maintained with support by coculturing E16.5 brain slices on a multipore filter (Millicell CM, Millipore, Bedford, MA). After 5 days of culture, the cover slips were fixed with 4% PFA and stained with TuJ1 antibody. The proportions of TuJ1+ cell containing colonies were evaluated as the average ± standard deviation among independent cultures. The significance of difference among proportion was determined using the student t test.

Time-Lapse Analyses

Time-lapse observations were performed using a previously described method (Tabata and Nakajima 2003).

Tissue Preparation and Observation

Tissue samples were prepared as described previously (Tabata and Nakajima 2001). The mouse monoclonal antibodies used in this study were anti-HuC/D (1:100; Molecular Probes, Inc., Eugene, OR), and anti-proliferative cell nuclear antigen (PCNA) (PC10, 1:100; Santa Cruz Biotechnology, Santa Cruz, CA). The rabbit polyclonal antibodies used in this study were anti-Iba1 (1:300; Wako Pure Chemical Industry, Osaka, Japan), anti-GFP (1:1000; MBL International, Watertown, MA), anti-NG2 (1:200; Chemicon, Temecula, CA), and anti-phosphohistone H3 (PH3) (1:200; Upstate, Spartanburg, SC). The goat polyclonal antibodies used in this study were anti-NeuroD (N-19, 1:500; Santa Cruz, CA). Olig2 was detected using biotinylated goat anti-Olig2 antibody (1:50; R&D Systems, Minneapolis, MN). To detect PCNA and NeuroD, sections were incubated at 85 °C in 0.01 M Citrate buffer (pH6.0) for 10 min prior to treatment with primary antibody. To detect NeuroD and Olig2, we used a tyramide signal amplification (TSA) fluorescence system (NEN Life Science Products, Boston, MA). For nuclear staining, 0.1 μM quinolinium, 4-[3-(3-methyl-2(3H)-benzothiazolylidene)-1-propenyl]-1-[3-(trimethylammonio)propyl]-, diiodide (TO-PRO-31; Molecular Probes, Eugene, OR) was used. Images were acquired using confocal microscopes (Zeiss LSM510, Jena, Germany, and Olympus FV300, Tokyo, Japan). The SVZ was determined following the definition of the Boulder Committee (Boulder Committee 1970), that is, the region containing proliferative cells not attached to the ventricular surface and not showing interkinetic nuclear movement during the mitotic cycle. Using this definition, the upper SVZ border was difficult to identify. Therefore, we applied the term “SVZ/IZ” to indicate the region between the VZ and CP. The proportions of PCNA-positive cells and of each marker positive cells among the GFP-positive cells were represented as averages ± standard deviations.

Labeling with Iododeoxyuridine (IdU) and Bromodeoxyuridine (BrdU) and Detection

IdU (Sigma, St Louis, MO) solution (50 mg/mL in dimethyl sulfoxide) was diluted with distilled water by 1/5 to make a 10 mg/mL solution. BrdU (Sigma) was dissolved in phosphate-buffered saline (PBS) at 10 mg/mL. IdU or BrdU solution was injected into the abdominal cavities of pregnant mice at 50 μg/g body weight. For immunostaining of BrdU, cryosections were incubated in PBS with 0.01% Triton X100 for 30 min, and then 2 N HCl at 37 °C for 30 min. After washing with PBS 3 times, the sections were incubated with mouse monoclonal anti-BrdU antibody (1:50; BD Biosciences, Mountain View, CA). For double labeling of IdU and BrdU, we used mouse anti-BrdU (BD Biosciences) antibody to detect both IdU and BrdU, and rat anti-BrdU (ICR1, 1:200; Abcam, Cambridge, UK) antibody to detect BrdU specifically. For NeuN labeling after IdU/BrdU staining, sections were incubated with anti-NeuN antibody (1:100, Chemicon), which had been biotinylated using a ProtOn biotin labeling kit (Vector Laboratories, Burlingame, CA), after acquiring the fluorescent images. The biotinylated antibody was detected using a Vectastain ABC kit (Vector Laboratories) with diaminobentidine. To carry out the triple staining for GFP, BrdU, and doublecortin shown in Figure 4C, or for GFP, BrdU, and Tbr2 in Figure 4D, we first stained the sections treated with HCl as described above with a rabbit anti-doublecortin (1:500, Abcam) or with a rabbit anti-Tbr2 (1:2000, gift from R. Hevner [Englund et al. 2005]) antiserum, followed by staining with a chick anti-GFP (1:500, Abcam) antiserum and a rat anti-BrdU antibody.

Quantitative Analyses of the Distribution of BrdU-Negative and BrdU-Positive Cells

The distributions of BrdU-negative and BrdU-positive cells were analyzed on coronal sections at the level of the anterior commissure. At least four sections from three independent embryos were analyzed. For the brains electroporated at E15, the position of each cell relative to the entire distance from the bottom of SVZ to the pial surface was measured using ImageJ software (National Institutes of Health shareware program), followed by sorting into 10 bins. For the dorsomedial region of E16 electroporated brains, the absolute distance was measured and fractioned into 10-μm bins. The statistical data are represented as averages ± standard deviations.

Results

Morphological Changes in Migrating Cells Depend on the Time after Electroporation

To visualize the morphology of migrating neurons, we introduced the GFP-expression vector into the VZ cells of the dorsomedial cerebral wall using the in utero electroporation system (Tabata and Nakajima 2001, 2002). When we introduced the GFP vector on embryonic day 16 (E16) and fixed the specimens 12 h later, the major population of GFP-positive cells was in the VZ (Fig. 1A,C). Besides this population, some GFP-positive cells were also found in the SVZ or IZ at various distances from the VZ (Fig. 1A, arrows). These cells in the SVZ/IZ frequently had a long ascending process and a retraction bulb, representing STL morphology (Fig. 1D) (Nadarajah et al. 2001). Thirty-six hours after electroporation, most GFP-positive cells were found just above the VZ (Fig. 1B). These cells tended to be aligned tangentially and assumed typical multipolar cell morphology with tangentially oriented thin multiple processes (Fig. 1F). The zone where multipolar cells accumulate (Fig. 1E, “*”, right) is characterized by two features: high cell-density (Fig. 1E, left) and positive staining for a neuronal marker, HuC/D (Fig. 1E, middle). Besides this multipolar cell population, GFP-positive cells were scattered throughout the SVZ/IZ, whereas only a very few were seen in the CP (Fig. 1B). Thus, morphological appearances changed depending on the time after electroporation. Cells showing STL cell morphology were observed in the SVZ/IZ 12 h after electroporation, and multipolar cells were abundant just above the VZ 36 h after electroporation.

Figure 1.

Morphological and positional differences of cells from the cortical VZ 12 or 36 h after labeling. (A-E) VZ cells of the cerebral wall of the mouse embryos were transfected with GFP-expression vector at E16 and observed the GFP expression 12 h (A, C, D) and 36 h (B, E) later. At the 12-h time point, GFP expressing cells were mainly found in the VZ (A) but some were found in the IZ (arrows in A). High magnification view revealed that they took radial glial or pin-like morphology in the VZ (C), and STL morphology having an ascending process (arrowheads in D) and a retraction bulb (arrow in D) in the SVZ or IZ. At the 36-h time point, GFP-positive cells accumulated just above the VZ (B), where the cell density is high (showing nuclear staining using TO-PRO-3I) and HuC/D, a neuronal marker, is positive (“*” in the left E). (F) The high magnification image of GFP-positive cells in this region in the brains electroporated at E14.5 and fixed at E16 showed that they were multipolar cells extending the multiple processes tangentially. The arrow indicates the radially oriented cell having tangentially extended axon. The broken lines in (A), (B), (E) indicate the margin of tissues or borders between the VZ and SVZ, or between the IZ and CP. Because the upper border of the SVZ is not clear, it is shown with a broken vertical line. Scale bars: 50 μm in (A), (B), (D), (E) and 20 μm in (C), (F).

Figure 1.

Morphological and positional differences of cells from the cortical VZ 12 or 36 h after labeling. (A-E) VZ cells of the cerebral wall of the mouse embryos were transfected with GFP-expression vector at E16 and observed the GFP expression 12 h (A, C, D) and 36 h (B, E) later. At the 12-h time point, GFP expressing cells were mainly found in the VZ (A) but some were found in the IZ (arrows in A). High magnification view revealed that they took radial glial or pin-like morphology in the VZ (C), and STL morphology having an ascending process (arrowheads in D) and a retraction bulb (arrow in D) in the SVZ or IZ. At the 36-h time point, GFP-positive cells accumulated just above the VZ (B), where the cell density is high (showing nuclear staining using TO-PRO-3I) and HuC/D, a neuronal marker, is positive (“*” in the left E). (F) The high magnification image of GFP-positive cells in this region in the brains electroporated at E14.5 and fixed at E16 showed that they were multipolar cells extending the multiple processes tangentially. The arrow indicates the radially oriented cell having tangentially extended axon. The broken lines in (A), (B), (E) indicate the margin of tissues or borders between the VZ and SVZ, or between the IZ and CP. Because the upper border of the SVZ is not clear, it is shown with a broken vertical line. Scale bars: 50 μm in (A), (B), (D), (E) and 20 μm in (C), (F).

Majority of Multipolar Cells Accumulating in the SVZ are Postmitotic

The place where tangentially oriented multipolar cells accumulate is within the SVZ, which is known to harbor basal progenitors as well. To clarify the relationship between these two populations in the SVZ, we first followed the fate of the postmitotic cells generated from VZ cells during a certain specific period. First, we applied iododeoxyuridine (IdU), a thymidine analogue, at E15, and then carried out electroporation with a GFP-expression vector two hours later. We further applied bromodeoxyuridine (BrdU) sequentially every 3 h from 12 to 36 h after electroporation (Fig. 2A). In this experiment, GFP+/IdU+/BrdU cells were thought to be postmitotic cells that had undergone the last S phase at E15 in the VZ, for the following reasons. First, when the plasmid vectors are introduced using in utero electroporation techniques, they are incorporated only into the cells that have direct contact with the ventricular surface (Hatanaka and Murakami 2002). Hence, GFP+ cells were thought to have been derived from VZ cells in the E15 cortex. Second, at this stage, the S phase is about 4 h, which is longer than the interval of BrdU application, and the total length of the cell cycle is 17.5 h (Takahashi et al. 1996b), which is shorter than the period of BrdU application. Even though there may be some cells that have a longer cell cycle than 17.5 h (Calegari et al. 2005), we assume that it would be unlikely for IdU+ cells, which had been in the S phase 2 h before electroporation, to re-enter the S phase after the 36-h time point. Hence, IdU+/BrdU cells were thought to be postmitotic cells that had finished the last S phase by the 12-h time point. To detect BrdU and IdU, we used an antibody that recognizes both BrdU and IdU, and a specific antibody for BrdU. Therefore, the GFP+ cells that had migrated out of the VZ were identified as either GFP+/IdU+/BrdU or GFP+/BrdU+. As a result, GFP+/IdU+/BrdU cells were mainly located just above the VZ as multipolar cells 36 h after electroporation (Fig. 2B, arrows). To quantify the distribution of the GFP+ cells, we divided the thickness of the cerebral wall into 10 bins from the upper (= basal) border of the VZ to the pial surface. GFP+/IdU+/BrdU cells were mainly found in bin#1, which accounted for 76.2 ± 9.6% and 91.6 ± 3.5% of all GFP+/IdU+/BrdU cells in the dorsomedial and lateral cortices, respectively (Fig. 2D,E), whereas 51.1 ± 11.9% and 47.5 ± 1.8% of all counted cells, which were either GFP+/IdU+/BrdU or GFP+/BrdU+, were found in bin#1, respectively. As to the dorsomedial cortex, the basal border of the zone where multipolar cells accumulated was located within bin#2. Therefore, the actual proportion of cells within this zone of all GFP+/IdU+/BrdU is thought to be higher than 76.2 ± 9.6%. In the same type of experiment using E16 embryos, 90.5% of GFP+/BrdU cells were found within 40 μm from the basal border of the VZ, which corresponds to the zone showing multipolar cell accumulation (Supplemental Fig. 1). In contrast, GFP+/BrdU+ cells were distributed throughout the SVZ/IZ (Fig. 2B, arrowheads; Fig. 2D,E) corresponding to the scattered distribution of the GFP+ cells within the SVZ/IZ 36 h after the electroporation (Fig. 1B,F). Although the GFP+/BrdU+ cells were distributed within the SVZ/IZ, only a very few were found in the CP. These observations are unlikely to be the result of toxicity due to repeated applications of BrdU, because there were barely any BrdU+ cells in the CP either, even when only a single shot of BrdU had been applied on E15.5 followed by fixation 24 h later (Supplemental Fig. 2). To rule out the possibility that the accumulation of the GFP+ cells above the VZ at 36 h was an adverse effect of electroporation, we prepared a control group from the same litter in which electroporation was not performed (Supplemental Fig. 3). In these control embryos, IdU and BrdU had been applied in the same schedule as for the electroporated embryos. The staining of IdU and BrdU on these brains revealed a band of IdU+/BrdU cells just above the VZ in both the lateral and the dorsomedial cortices, whereas the BrdU-positive cells were observed in the SVZ/IZ. These observations indicate that the major population finishing their final cell division in the VZ settles just above the VZ as postmitotic multipolar cells by 36 h after electroporation in vivo. We named this region the MAZ. Because the SVZ is defined as the region containing proliferative cells that are not attached to the ventricular surface and do not show interkinetic nuclear movement during the mitotic cycle (Boulder Committee 1970; Bystron et al. 2008), the MAZ overlaps with the lower part of the SVZ.

Figure 2.

Postmitotic cells generated from VZ cells accumulated just above the VZ (=MAZ) as multipolar cells. (A) Experimental procedure to identify postmitotic cells generated from VZ cells in the specific period. (B, C) The brains after the experiment described in A were fixed at 36-h (B) and 60-h (C) time point, and then BrdU (magenta) and IdU (blue) were detected. Arrows and arrowheads indicate IdU+/GFP+/BrdU cells (=the direct progeny of the VZ cells) and GFP+/BrdU+ cells, respectively. (D, E) The histograms of the distributions in dorsomedial (D) and lateral cerebral walls (E). SVZ/IZ, SVZ or IZ. Because the upper border of the SVZ is not clear, it is shown with a broken vertical line. Scale bar; (B) and (C), 50 μm.

Figure 2.

Postmitotic cells generated from VZ cells accumulated just above the VZ (=MAZ) as multipolar cells. (A) Experimental procedure to identify postmitotic cells generated from VZ cells in the specific period. (B, C) The brains after the experiment described in A were fixed at 36-h (B) and 60-h (C) time point, and then BrdU (magenta) and IdU (blue) were detected. Arrows and arrowheads indicate IdU+/GFP+/BrdU cells (=the direct progeny of the VZ cells) and GFP+/BrdU+ cells, respectively. (D, E) The histograms of the distributions in dorsomedial (D) and lateral cerebral walls (E). SVZ/IZ, SVZ or IZ. Because the upper border of the SVZ is not clear, it is shown with a broken vertical line. Scale bar; (B) and (C), 50 μm.

Cells entering the SVZ/IZ Faster than Postmitotic Multipolar Cells are Mitotically Active

Our birth-dating experiment revealed that, 36 h after electroporation, GFP+/BrdU+ cells were distributed throughout the SVZ/IZ, whereas GFP+/IdU+/BrdU cells had accumulated in the MAZ (Fig. 2B,D,E). One possible explanation for this is that the GFP+/BrdU+ cells were basal progenitors and had migrated into the SVZ/IZ faster than the postmitotic multipolar cells and then incorporated BrdU within the SVZ/IZ. Another possible interpretation is that the cells became postmitotic in the VZ and quickly migrated into the SVZ/IZ as postmitotic cells. To address this issue, we examined their immunoreactivity for the anti-PCNA antibody, a marker for mitotically active cells in all phases of the cell cycle (Bravo et al. 1987). Within the SVZ/IZ other than the MAZ, high proportions of GFP-positive cells were PCNA positive, accounting for 62.3 ± 8.4% (n = 261, 3 brains), whereas only 23.9 ± 6.3% (n = 449, 3 brains) of the GFP-positive cells in the MAZ were PCNA positive (Fig. 3A, arrows). We also examined PCNA expression in brains after the experiment described in Figure 2A (Supplemental Fig. 4A,B). At 36 h after electroporation, the brains were stained with anti-GFP, PCNA and BrdU. A major proportion of the GFP+/BrdU+ cells in the SVZ/IZ were PCNA positive, whereas GFP+/BrdU cells in the MAZ were basically PCNA-negative (Supplemental Fig. 4). These observations further confirmed that the majority of BrdU+ cells in the SVZ/IZ were proliferative, although some might already have become postmitotic in the SVZ/IZ, thereby ruling out the possibility that GFP+/BrdU cells in the MAZ might still be proliferative and re-enter the S phase after the 36-h time point. At the 12-h time point, we observed a scattered distribution of GFP-positive cells in the SVZ/IZ, that did not appear to stop in the vicinity of the MAZ (just above the VZ) (Fig. 1A). This observation raised the possibility that these GFP-positive cells in the SVZ/IZ at the 12-h time point may become the GFP+/BrdU+ SVZ/IZ cells observed at the 36-h time point. We therefore examined the proliferative activity of GFP-positive cells in the SVZ/IZ at the 12-h time point. We applied a single shot of BrdU 12 h after electroporation and fixed the embryos 30 min later. As mentioned above, GFP-positive cells at this time-point frequently assumed STL morphology, and we found a high proportion to be BrdU positive (58%, n = 151) (Fig. 3B, arrows). In the E16 brain, PH3-positive M phase cells were distributed throughout the SVZ/IZ, and did not show specific accumulation in the MAZ (Supplemental Fig. 5). These observations supported the idea that the GFP+ S phase cells found in the SVZ/IZ at the 12-h time point would undergo M phase in the SVZ/IZ. To visualize the proliferative activity of these cells directly, we then performed time-lapse observations. We introduced CAG-DsRedExpress vector at E14.5 using electroporation and prepared slices at E15 (Fig. 3C, Supplemental Movie 1). Cells a and b both divided into two cells, after losing their ascending processes, and then began dynamic process movement, similar to that of multipolar cells in the MAZ. Taken together, these observations indicate that the majority of GFP-positive cells in the SVZ/IZ at the 12-h time point after electroporation undergo cell divisions. The resulting progeny cells are likely to be located in the IZ at the 36-h time point.

Figure 3.

The REP has proliferative activity. (A) The immunostaining of PCNA (magenta) in the brains that were electroporated at E16 and fixed 36 h later. The left panel shows PCNA staining, and the right panel shows the merged image with GFP. The arrows indicate the GFP+/PCNA+ cells. The GFP-positive cells above the MAZ were frequently PCNA positive. (B) BrdU incorporation of GFP+ cells migrated out of the VZ by 12 h after electroporation. Embryos were electroporated at E16 and fixed 12 h later. BrdU was injected 30 min before fixation. The arrows indicate GFP+/BrdU+ cells in the SVZ/IZ. (C) The time-lapse observation of the SVZ/IZ cells on the brain electroporated at E14.5 and sliced 12 h after the electroporation. Cell “a” having STL morphology divided into cell “a1” and “a2,” and lost its ascending fiber. Cell “b” having no ascending process also divided into cell “b1” and “b2”. Scale bars; (A), (B); 50 μm, (C); 20 μm.

Figure 3.

The REP has proliferative activity. (A) The immunostaining of PCNA (magenta) in the brains that were electroporated at E16 and fixed 36 h later. The left panel shows PCNA staining, and the right panel shows the merged image with GFP. The arrows indicate the GFP+/PCNA+ cells. The GFP-positive cells above the MAZ were frequently PCNA positive. (B) BrdU incorporation of GFP+ cells migrated out of the VZ by 12 h after electroporation. Embryos were electroporated at E16 and fixed 12 h later. BrdU was injected 30 min before fixation. The arrows indicate GFP+/BrdU+ cells in the SVZ/IZ. (C) The time-lapse observation of the SVZ/IZ cells on the brain electroporated at E14.5 and sliced 12 h after the electroporation. Cell “a” having STL morphology divided into cell “a1” and “a2,” and lost its ascending fiber. Cell “b” having no ascending process also divided into cell “b1” and “b2”. Scale bars; (A), (B); 50 μm, (C); 20 μm.

Thus, we herein identified two distinct populations migrating from the VZ in the late stages of cortical development. One population finishes the last S phase in the VZ and migrates into the MAZ to become postmitotic multipolar cells. These multipolar cells accumulate specifically in the MAZ. The other population migrates into the SVZ/IZ quickly and undergoes further cell divisions to become basal progenitor cells. We refer to these cells as the SEP and the REP, respectively.

Characterization of the REP and SEP

The proliferative activity of the REP raised the possibility that these cells might include glial cells. However, the GFP-positive SVZ/IZ cells were principally not labeled with the markers for astrocytes (glial fibrillaryacidic protein [GFAP] [Bignami and Dahl 1974], 8/172 cells in 10 brains), oligodendrocytes (NG2 [Nishiyama et al. 1996], 0/159 cells in 10 brains; platelet-derived growth factor receptor α [PDGFRα] [Ellison and de Vellis 1994], 0/245 cells in 10 brains), or microglia (Iba1 [Ito et al. 1998], 0/254 cells in 10 brains) at the 36-h time point. Nevertheless, there is a possibility that they were as yet too immature to express these markers at this time point. Olig2 is a marker for glial progenitor cells (Marshall et al. 2005), and it has been shown that, during the late stage of cortical development, Olig2-expressing cells in the cerebral cortex mainly produce astrocytes and oligodendrocytes (Ono et al. 2008). We found a small number of GFP-positive SVZ/IZ cells located above the MAZ at the 36-h time point to be positive for Olig2 (Fig. 4A; 7.6%, 184 cells/5 brains), suggesting that the REP includes glial progenitors as the minor population. We next examined an early neuronal marker, NeuroD (Sommer et al. 1996; Ge et al. 2006), and revealed that the majority of GFP-positive SVZ/IZ cells located above the MAZ at the 36-h time point were positive for NeuroD (Fig. 4B; 65 ± 17%, 179 cells/9 brains), although the proportion was smaller than that in the MAZ (96 ± 4%, 334 cells/9 brains). To more directly demonstrate whether the REP indeed produces neurons, we injected GFP-retrovirus into the lateral ventricle on E15.5 in vivo and dissected out two types of tissue fragments under a dissecting microscope 18–22 h after the infection, that is, the upper (=basal) part of the SVZ/IZ above the MAZ and the part including the VZ and whole SVZ/IZ. These tissue fragments were dissociated and cultured for 5 days. We observed that the former produced more colonies containing TuJ1-positive/GFP-positive cells than the latter (83.4 ± 4.2% vs. 63.5 ± 6.1%, P < 0.01, 3 independent cultures). Therefore, we concluded that GFP-positive SVZ/IZ cells above the MAZ at 36 h, which corresponded to the REP or its progeny, contained more neuronal progenitors than the underlying VZ cells.

In human brains, GABAergic neurons are generated from the cortical VZ in addition to the basal telencephalon (Letinic et al. 2002). To determine whether or not the REP produces GABAergic neurons, SVZ/IZ cells of the E16 cortex, which had been infected with a GFP-retrovirus on E14.5, were cultured for 5 days and stained with an antibody for GABA. As a result, GFP-positive cells were negative for GABA (0/111 cells), demonstrating that the REP principally produces non-GABAergic cells, at least at this stage.

As to the SEP, we cannot completely exclude the possibility that the GFP+/IdU+/BrdU cell population in the MAZ might include some “mitotic” cells with a very long cell cycle. However, it is unlikely that the majority of SEP is dividing, because we found that most of the GFP-positive cells in the MAZ at 36-h time point was not only NeuroD-positive (Fig. 4B), but also positive for TuJ1, a marker for differentiated neurons (71.8%, 326 cells in 5 brains, Supplemental Fig. 6). To characterize the SEP further, we examined the expression of doublecortin (DCX), another early neuronal marker, on the brain sections that had been injected with a GFP-expression vector and thymidine analogues as shown in Figure 2A and had been fixed 36 h after electroporation. As the results, most of the GFP+/BrdU cells (=SEP cells) expressed DCX (Fig. 4C). The immunohistochemical analysis for Tbr2, which is a good marker for the basal progenitors (Englund et al. 2005), on the brain sections fixed at 36-h time point in the experiment shown in Figure 2A revealed that SEP cells were negative or, if any, only faintly positive for Tbr2, whereas the cells just beneath the MAZ and a subset of REP were strongly positive (Fig. 4D). These marker analyses supported the hypothesis that SEP is postmitotic neurons and REP includes basal neuronal progenitors.

VZ Cells or their Immediate Progeny can Transform into Multipolar Cells Without Mitosis

Sequential BrdU injections demonstrated that multipolar cells located within the MAZ 36 h after electroporation had finished their final S phase of the cell cycle by 12 h after electroporation (Fig. 2B,D,E). At 12 h after electroporation, the major population of GFP+ cells was still located in the VZ and assumed a radially oriented bipolar configuration extending a thin ascending process toward the pial surface and a thick foot process attached to the ventricular surface (Fig. 1A,C). Some cells had a foot process and a short or no ascending process, representing pin-like morphology (Gal et al. 2006; Ochiai et al. 2007) (Fig. 1C). Ochiai et al. demonstrated that pin-like VZ cells labeled with DiI transformed into multipolar cells without mitosis. To determine how frequently this transformation occurs to generate multipolar cells, time-lapse observations of VZ cells were conducted. In this experiment, we used GFP-retrovirus to sparsely label VZ cells. The retrovirus was injected at E15.5 into the lateral ventricle and the slices were prepared 18–24 h later. The example in Figure 5A shows a VZ cell which divided into two cells (Fig. 5A, 0.5 h). At least one of the daughter cells possessed the foot process after the cell division to maintain the contact with the ventricular surface. After staying within the VZ for 15 h, the cells retracted their foot processes and transformed into multipolar cells within the lower part of the SVZ/IZ (Fig. 4A, Supplemental Movie 2). In another case shown in Figure 5B, we observed two radial cells that shared the ventricular attachment with their foot processes, suggesting that they might have been derived from a common progenitor (Fig. 5B, 0 h). One of them showed the radial glial morphology, the other pin-like morphology (indicated by an arrow). The pin-like cell maintained the foot process for the first 6 h of observation. This cell then retracted the foot process and transformed into a multipolar cell within the lower SVZ/IZ (Fig. 5B, Supplemental Movie 3). The patterns of multipolar cell generation are classified in Figure 5C. We observed 38 instances of the production of multipolar cells during the first 24 h of the time-lapse analyses in five independent cultures. Because it was technically difficult in many cases to clearly distinguish whether the resulting multipolar cells are SEP or REP, we did not distinguish them in this classification. At first, we classified them according to whether or not the cells divided during the observations (Fig. 5C). When cell division was not observed, the multipolar cells were derived from either VZ or STL (in Fig. 5C) cells. VZ cells here indicate those that had contact with the ventricular surface including radial glia and short neural precursors (Gal et al. 2006) or pin-like cells. On the other hand, STL cells are those that moved away from the ventricular surface with bipolar morphology. In the event of cell division, multipolar cells were also derived from either VZ or STL cells. The divisions of VZ and STL cells took place at the ventricular surface and away from the ventricular surface, respectively. According to this classification, the events illustrated in Figure 5A,B were classified as the division of VZ cells to generate two multipolar cells (VZ cell → 2xMP cell), and the transformation of a VZ cell to one multipolar cell (VZ cell → MP cell), respectively. Thus, the direct transformation of VZ cells into multipolar cells was the most common mode of multipolar cell generation, accounting for 53%. Even in the case of the VZ cell-division to generate two multipolar cells, the progenies frequently stayed in the VZ with their foot processes for more than 10 h. Taken together, these observations indicate that cells becoming postmitotic in the VZ (=SEP) initially assume bipolar-cell morphology such as a pin-like structure and require at least 10 h to transform into multipolar cells without mitosis.

Figure 4.

Characterization of the REP and SEP. (A, B) The brains electroporated at E15 and fixed 36 h later were stained with anti-Olig2 (A, A1, 2, magenta) and anti-NeuroD1 (B, B1, 2, magenta), respectively. The boxed regions of (A) and (B) are shown in A2 and B2. A1 and B1 show the images without GFP signals. A few GFP-positive cells in the IZ were positive for Olig2 (arrows in A1, 2), but the major population was positive for NeuroD (arrows in B1, 2). (C, D) The brains fixed at the 36-h time point after the experiment described in Figure 2A were subjected to immunostaining for doublecortin (DCX, C, red) or Tbr2 (D, red) together with GFP (green) and BrdU (blue). The boxed region of C is shown in C14 in different channel combinations. The expression of DCX in the GFP-positive/BrdU-negative cells (= SEP) was detected (arrows). The morphological features of cells that were strongly positive for DCX but negative for GFP in the VZ or MAZ suggest that they are tangentially migrating interneurons (arrowheads). (D) The merged image is shown in the left panel and the image without GFP signals in the same field is shown in the right panel. The GFP+/BrdU- cells (=SEP, arrowheads) were negative or, if any, only faintly positive for Tbr2, whereas a subset of GFP+/BrdU+ cells in the IZ (=REP, arrows) was strongly positive. Because the upper border of the SVZ is not clear, it is shown with a broken vertical line. Scale bars; A, B, C; 50 μm, C14; 10 μm, A1, A2, B1, B2, D; 20 μm.

Figure 4.

Characterization of the REP and SEP. (A, B) The brains electroporated at E15 and fixed 36 h later were stained with anti-Olig2 (A, A1, 2, magenta) and anti-NeuroD1 (B, B1, 2, magenta), respectively. The boxed regions of (A) and (B) are shown in A2 and B2. A1 and B1 show the images without GFP signals. A few GFP-positive cells in the IZ were positive for Olig2 (arrows in A1, 2), but the major population was positive for NeuroD (arrows in B1, 2). (C, D) The brains fixed at the 36-h time point after the experiment described in Figure 2A were subjected to immunostaining for doublecortin (DCX, C, red) or Tbr2 (D, red) together with GFP (green) and BrdU (blue). The boxed region of C is shown in C14 in different channel combinations. The expression of DCX in the GFP-positive/BrdU-negative cells (= SEP) was detected (arrows). The morphological features of cells that were strongly positive for DCX but negative for GFP in the VZ or MAZ suggest that they are tangentially migrating interneurons (arrowheads). (D) The merged image is shown in the left panel and the image without GFP signals in the same field is shown in the right panel. The GFP+/BrdU- cells (=SEP, arrowheads) were negative or, if any, only faintly positive for Tbr2, whereas a subset of GFP+/BrdU+ cells in the IZ (=REP, arrows) was strongly positive. Because the upper border of the SVZ is not clear, it is shown with a broken vertical line. Scale bars; A, B, C; 50 μm, C14; 10 μm, A1, A2, B1, B2, D; 20 μm.

Figure 5.

Transformation of the postmitotic cells in the VZ, which is identified as the SEP, into the multipolar cells. (A, B) Time-lapse analyses showing the generation of multipolar cells from the radial VZ cells without further mitosis. GFP-retrovirus was injected at E15.5 and slices were prepared 18 h later. A pin-like cell is pointed by arrows (B). The short horizontal lines represent the luminal surface of the VZ. (C) Classification of the multipolar cell formation. MP, multipolar. Scale bars; 50 μm.

Figure 5.

Transformation of the postmitotic cells in the VZ, which is identified as the SEP, into the multipolar cells. (A, B) Time-lapse analyses showing the generation of multipolar cells from the radial VZ cells without further mitosis. GFP-retrovirus was injected at E15.5 and slices were prepared 18 h later. A pin-like cell is pointed by arrows (B). The short horizontal lines represent the luminal surface of the VZ. (C) Classification of the multipolar cell formation. MP, multipolar. Scale bars; 50 μm.

Migratory Profile of SEP In Vitro and In Vivo

Based on the observations of cell morphology in vivo, we previously proposed that multipolar cells would be transformed into locomotion cells before entering the CP (Tabata and Nakajima 2003). This was later directly demonstrated by Noctor et al. using retrovirus labeling in a long-term slice culture system (Noctor et al. 2004). To show that SEP does actually transform into locomotion cells and for clarification of the time course of this transformation with a high time resolution, we conducted automatic confocal time-lapse observations at 30-min intervals. When the GFP-expression vector was introduced into the dorsomedial cortex at E14.5 and the slices were prepared at E15.5, labeled cells with multipolar morphology accumulated in the MAZ at the beginning of the observation (Fig. 6, 0 h), They continued to be densely packed in the MAZ through approximately 20 h of culture. Dynamic extension and retraction of multiple processes were observed for these cells, whereas their somas moved very slowly, indicating these cells to be the multipolar migrating cells (Supplemental Movie 4). These SEP then gradually changed in morphology to become locomotion cells at around 12–18 h before entering the CP.

Figure 6.

Transformation of multipolar cells in the MAZ into locomotion cells. The brains electroporated at E14.5 was sliced at E15.5 and observed using a confocal microscope. The broken lines indicate the margin of the tissue. Scale bar; 50 μm.

Figure 6.

Transformation of multipolar cells in the MAZ into locomotion cells. The brains electroporated at E14.5 was sliced at E15.5 and observed using a confocal microscope. The broken lines indicate the margin of the tissue. Scale bar; 50 μm.

Next, we investigated how common this transformation from multipolar cells (SEP) into locomotion cells is in vivo. The SEP was identified as GFP+/IdU+/BrdU cells using the method described in Figure 2A. When the embryos were fixed 60 h after the electroporation, some GFP-positive cells found in the CP had typical locomotion cell morphology (Fig. 2C, arrows). Triple staining revealed that 86.0 ± 12.1% of the GFP-positive locomotion cells in the CP were IdU+/BrdU in the dorsomedial cerebral wall. Considering that most of the GFP+/IdU+/BrdU cells had been located in the MAZ 36 h after electroporation, the major population of the GFP-positive locomotion cells in the CP at 60 h after electroporation was thought to have been derived from the multipolar cells observed in the MAZ at 36 h after electroporation. In the lateral cortex, there was also a superficial shift of SEP comparable to that in the dorsomedial region (Fig. 2E, right), although only a few cells were found in the CP at the 60-h time point. The difference in the timing of entering the CP may have resulted from the difference in the required migration distances to reach the CP between the dorsomedial and lateral cortices. Although the REP moved into the SVZ/IZ faster than the SEP, this cell population did not show an obvious positional shift toward the CP by the 60-h time point in the lateral cortex.

The Final Fate of SEP

To determine the final fate of the SEP, we investigated the GFP+/IdU+/BrdU cells on postnatal day 17 (P17) in the dorsomedial cortex (Fig. 7A,B, arrowheads). Most of these cells were located in layers II/III with pyramidal cell morphology and were positive for a neuronal marker, NeuN (Fig. 7B; 96.8%, n = 124), although some cells were also found in the white matter and deep cortex as a minor population (data not shown). These observations indicate that the major population of postmitotic neurons found in the MAZ at 36 h after electroporation differentiates into the pyramidal neurons of layers II/III.

Figure 7.

Final fate of postmitotic cells generated within the VZ (SEP). (A) SEP-derived cells identified as GFP+/IdU+/BrdU using the method described in Figure 2A were mainly found in layers II/III. (B) NeuN expression in SEP-derived cells. The boxed region in (A) is shown in the left and middle (B). Arrowheads indicate SEP-derived cells. After took fluorescent images of GFP, IdU, and BrdU, the sections were further stained with anti-NeuN, a marker for mature neurons, using peroxidase and diaminobentidine, and the same visual field was observed (right panel). Scale bars: 50 μm.

Figure 7.

Final fate of postmitotic cells generated within the VZ (SEP). (A) SEP-derived cells identified as GFP+/IdU+/BrdU using the method described in Figure 2A were mainly found in layers II/III. (B) NeuN expression in SEP-derived cells. The boxed region in (A) is shown in the left and middle (B). Arrowheads indicate SEP-derived cells. After took fluorescent images of GFP, IdU, and BrdU, the sections were further stained with anti-NeuN, a marker for mature neurons, using peroxidase and diaminobentidine, and the same visual field was observed (right panel). Scale bars: 50 μm.

Discussion

The Model of Late Cortical Development

We examined the fate of cortical VZ-derived cells generated at specific periods by combining the methods of in utero electroporation and the application of thymidine analogues, BrdU and IdU, in this study. We observed two distinct populations in terms of migration patterns from the VZ (Fig. 8). One population, SEP, finishes the final cell division within the VZ and migrates slowly into the MAZ, where the cells assume typical multipolar cell morphology and accumulate. The other population, REP, exits the VZ rapidly via STL and the majority, at least, undergoes further cell division within the SVZ/IZ. SEP-derived multipolar cells transform into locomotion cells to migrate into the CP, and eventually become non-GABAergic neurons that constitute layers II/III.

Figure 8.

Model of the migratory behavior of REP and SEP cells. The cell bodies of GFP-positive cells are shown in gray, and the BrdU-positive nuclei are shown in black.

Figure 8.

Model of the migratory behavior of REP and SEP cells. The cell bodies of GFP-positive cells are shown in gray, and the BrdU-positive nuclei are shown in black.

The BrdU incorporation rate of GFP-positive cells in the SVZ/IZ 12 h after electroporation was extremely higher (58%) than expected. Because the total cell cycle length and S phase length at E15 are about 17.5 and 4 h (Takahashi et al. 1996b), respectively, the content of the S phase cells would be expected to be around 23%. Considering that the cell cycle length of neurogenic progenitors is longer than that of proliferative progenitors (neural stem cells), the incorporation rate is expected to be less than this estimation (Calegari et al. 2005). This fact suggests that REP would leave the VZ at a particular phase of cell cycle. If they leave the VZ at late G1 phase or during the S phase, the high incorporation rate of BrdU would be explainable. Recent study has reported that the calcium waves mediated by P2Y1 ATP receptor are involved in the translocation of basal progenitors into the SVZ (Liu et al. 2008). This same machinery is known to initiate the DNA synthesis in VZ cells (Weissman et al. 2004). Hence, there is a possibility that the ATP signaling stimulates the transition from G1 phase to S phase in the basal progenitors (=REP), and as a result, it initiates the translocation of basal progenitors into the SVZ.

In this paper, we were not able to demonstrate whether the REP enters the CP in later stages, because this cell population was indistinguishable from the “BrdU-positive SEP” that had passed the last S-phase, later than the 12-h time point after electroporation. However, we believe that REP would eventually migrate into the CP later than SEP for the following reasons. First, our NeuroD staining of the fixed sections and the primary culture of the isolated SVZ/IZ cells indicated that at least the major population of REP would differentiate into neurons. Second, previous studies have shown that the progenies of basal progenitors migrate into the CP (Noctor et al. 2004) and they can indeed differentiate into pyramidal neurons (Wu et al. 2005). Finally, we observed a group of GFP-“negative”/IdU-positive/BrdU-negative cells around the lower border of the CP at the 36-h time point (Fig. 2B). Because they were GFP-negative, this population was not thought to have been in contact with the ventricular surface at the time of electroporation. Hence, these might be postmitotic cells derived from basal progenitors that had been located in the SVZ/IZ at the time of electroporation, and were located above the BrdU+ basal progenitors in the SVZ/IZ at the 36-h time point. These facts, therefore, strongly suggest that REP progenies also migrate into the CP after finishing their final cell divisions in the SVZ/IZ.

Previous studies have revealed the existence of two distinct populations of postmitotic cells by means of a single injection of [3H]Thymidine, or with sequential injections of BrdU during the initial phase of radial migration (Altman and Bayer 1990; Takahashi et al. 1996a). One population was distributed just above the VZ and the other was found in the IZ 12–24 h after the first injection of [3H]Thymidine. The former was designated as the inferior band (ib) or Qs, and the latter was called the superior band (sb) or Qr. There was a possibility that the sb or Qr might have been generated from VZ cells similar to the ib or Qs, but then migrated and reached the IZ faster than the ib or Qs. However, by utilizing the in utero electroporation system in addition to the thymidine analogues to mark VZ-derived cells, we found that the GFP+/IdU+/BrdU cells were mainly located in the MAZ, and that IdU+/BrdU cells around the lower border of the CP were GFP negative. These observations indicated that the Qr is derived from basal progenitors rather than VZ cells.

The differences in migration modes between the SEP and REP suggest that their fates may be determined within the VZ. We observed REP to possess a long ascending process and to assume STL morphology in the SVZ/IZ in vivo (Fig. 1A,D). Previous time-lapse analyses demonstrated that the ascending process of basal progenitor cells was an inheritance from the parental radial glia (Miyata et al. 2004; Noctor et al. 2004), indicating that the basal progenitor or REP is the basal progeny of asymmetric cell division of the radial glia in the VZ. On the other hand, we observed that the major population of multipolar cells generated through transformation without cell division was derived from VZ cells, which were frequently recognized as the pin-like cells on time-lapse analyses. Consequently, SEP is thought to be the direct apical progeny of asymmetric cell division. These observations raised the possibility that the difference in fate between REP and SEP might be decided through asymmetric cell division in the VZ.

The Majority of the Embryonic SVZ Cells are the Postmitotic Multipolar Cells

In this study, we have proposed that postmitotic multipolar cells temporarily localize to the high cell-density zone just above the VZ, which we named “MAZ.” This region in fact overlaps with the SVZ, because the SVZ has been “histologically” recognized as a high cell-density zone with irregular cell-orientation adjacent to the VZ (Bayer and Altman 1991). This is somewhat controversial, as it contradicts the general concept that the SVZ is the place where basal progenitor cells accumulate. This may arise from ambiguity in the definition of the SVZ. The SVZ concept was defined by the Boulder Committee (Boulder Committee 1970) as the zone adjacent to the VZ with proliferative activity but without interkinetic nuclear movement during the mitotic cycle in the mid and late stages of cortical development. Based on this “functional” definition, the SVZ may partially overlap with the VZ and IZ, because basal progenitors were widely distributed from the upper border of the VZ throughout the IZ (Figs 2B,D,E and 3A,B) (Takahashi et al. 1995; Wu et al. 2005). To avoid confusion, we propose the term MAZ, which must be distinguished from but is histologically included in the “functionally defined” SVZ, to indicate the multipolar-cell–accumulating zone above the VZ. Several genes are known to be “SVZ”-specific markers based on the “histological” definition of the SVZ, and have been shown to be expressed in basal progenitor cells (Tarabykin et al. 2001; Nieto et al. 2004; Zimmer et al. 2004). However, our observations may raise a need to investigate whether these markers are expressed specifically in basal progenitor cells alone, or also in postmitotic multipolar cells in the MAZ. In fact, we recently showed that the Svet1 RNA, a widely used SVZ marker, is part of the sequence of the primary transcript of Unc5d and that the unc5D/Svet1 RNA and the Unc5D protein are mainly expressed in the postmitotic multipolar cells in the MAZ, although it is also expressed in the basal progenitors (Sasaki et al. 2008).

The Contributions of SEP and REP to the Total Number of Layers II/III Neurons

The expression patterns of SVZ markers such as Svet1 and Cux-1/2 during cortical development, or phenotypes of mutant mice lacking Pax6 and Brn1/2 allow us to hypothesize that VZ cells directly produce deep layer neurons, whereas basal progenitors produce the upper (II/III) layer neurons. It would therefore be important to estimate the contributions of REP and SEP to cortical development. We observed that the SEP proportion was predominant (58.6 ± 8.5%) over that of REP (41.4 ± 8.5%) in the dorsomedial cortex at the 36-h time point. The REP population is comprised of heterogeneous cell types, including Olig2-positive cells, as well as neurons. NeuroD-positive cells in the REP population accounted for 65 ± 17%, whereas 96 ± 4% were positive for NeuroD in the MAZ. Therefore, the “maximal” contribution rate of the SEP to neuronal production at this stage is roughly estimated to be 0.586 × 0.96/(0.586 × 0.96 + 0.414 × 0.65) = 67.6%. This is likely to be an overestimate, because the REP may increase through cell divisions and each division of REP cells in the later stages may give rise to 2 neurons. However, the number of cell divisions of basal progenitors is expected to be one or two (Noctor et al. 2004), and the BrdU incorporation rate in the GFP+ SVZ/IZ cells above the MAZ at the 36-h time point was decreased to 24.3% (data not shown) from 58% at the 12-h time point (Fig. 3B, arrows). We therefore assume that the REP would not expand substantially after the 36-h time point. Even if all the NeuroD-negative REP cells might divide and produce 2 neurons at each division in the later stages, the estimation of the contribution of SEP to the total neuronal output would be roughly 0.586 × 0.96/{0.586 × 0.96 + 0.414 × 0.65 + 0.414 × (1 − 0.65) × 2} = 50.2%. Thus, the SEP would account for a significant proportion of layer II/III neurons. Indeed, we found SEP-derived pyramidal neurons in layers II/III of P17 brains (Fig. 7). Therefore, layer II/III neurons are derived from both the SEP and the REP.

Importantly, the proportional balance between the SEP and the REP differs among cortical areas and may also change as development proceeds. In the lateral cortex, REP (BrdU-positive cells) accounted for 64.4 ± 5.6%, which is higher than the proportion in the dorsomedial cortex (41.4 ± 8.5%). In the lateral cerebral wall, basal or apical progenitors may have to produce more neurons than those in the medial wall to cover a broader region of the developing cortex. Thus, there is a possibility that the balance between the SEP and the REP is regulated by unknown mechanisms designed to meet the demand for cortical neurons.

Preserving Inside-Out Configuration of the CP

We observed GFP+/BrdU+ cells (REP) to be localized to the more superficial side than the GFP+/IdU+/BrdU cells (SEP) at the 36-h time point. Thus, in this situation, the recently generated cells were located more superficially than the older cells. If the superficially located cells (REP-derived neurons, relatively younger neurons) move toward the brain surface and reach a site beneath the MZ earlier than the deeper positioned cells (SEP, relatively older neurons), the birth date-dependent “inside-out” pattern of neuronal alignment in the CP would be disrupted. However, we observed that the SEP moved toward the CP prior to the REP, and, at least, in the dorsomedial cortex, the SEP entered the CP before the REP by the 60-h time point. This must contribute to preserving the birth date-dependent “inside-out pattern” of neocortical layer formation. These observations are consistent with those of a previous study using [3H]Thymidine and BrdU, which showed Qs and Qr to merge under the CP and enter the developing CP together (Takahashi et al. 1996a). As mentioned above, Qs and Qr are thought to be the SEP itself and the postmitotic cells generated from the REP, respectively. Hence, the SEP would presumably enter the CP with the progeny of an earlier cohort of the REP, which migrated into the SVZ/IZ at an earlier stage. These facts suggested that the timing of entry into the CP would be determined by the time after the final cell division but not by the time after the start of migration from the VZ.

Considering the fast migration rate of the STL mode (Nadarajah et al. 2001), it is reasonable to speculate that the REP assumes the STL morphology and enters the SVZ/IZ faster than the SEP (Fig. 1A,D). However, this cell population does not enter the CP directly (Fig. 2C,D,E), and the cells lose their ascending processes during the mitosis within the SVZ/IZ (Fig. 3C). Reflecting this event, the distribution of mitotic cells decreases sharply in the CP (Fig. 3B). The mechanisms to prevent direct entry of these cells and cues to resume migration toward the CP must be clarified in the future.

The Biological Meaning of the Accumulation of Postmitotic Multipolar Cells in the MAZ

SEP settled in the MAZ before they entered the CP. The biological meaning of this phenomenon is not yet clear. The fact that the REP is able to pass through the MAZ suggests that the accumulation of SEP in the MAZ is an active step, rather than simply reflecting a disturbance in migration due to physical obstacles. In addition, soma and processes of multipolar cells are not randomly oriented but rather tend to orient tangentially in parallel to the axons in the IZ (Tabata and Nakajima 2003). Several histological analyses have shown migrating neurons to have axons in the early step of the migration (Fig. 1F, arrow) (Stensaas and Stensaas 1968; Shoukimas and Hinds 1978; Ozaki and Wahlsten 1998). In this study, we observed that, when SEP cells accumulated in the MAZ, they already had tangentially extended axon-like structures within the MAZ (Fig. 1F). These observations indicate that the phase of positioning in the MAZ corresponds to the timing of directed axonal extension. The exploratory behavior of multipolar cells might reflect searching for directional cues for axonal protrusion. The late-born neurons give rise to layer II/III neurons, which send out callosal axons running in the lower IZ. Thus, the accumulation of SEP cells within the MAZ, at a high density, might contribute to the fasciculation of callosal axons.

Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/

Funding

Japan Society for the Promotion of Science; the Ministry of Education, Culture, Sports; and Science and Technology of Japan; the Ichiro Kanehara Foundation; the Tokyo Biochemical Research Foundation; the Sumitomo Foundation; the Uehara Memorial Foundation; the Takeda Science Foundation; and the Brain Science Foundation.

We thank T. Kitamura for the pMX-GFP and Plat-E cells, J. Miyazaki for the CAG promoter, R. Hevner for anti-Tbr2 antiserum, and members of Nakajima laboratory for valuable suggestions and encouragement. Conflict of Interest: None declared.

References

Altman
J
Bayer
SA
Horizontal compartmentation in the germinal matrices and intermediate zone of the embryonic rat cerebral cortex
Exp Neurol.
 , 
1990
, vol. 
107
 (pg. 
36
-
47
)
Bayer
SA
Altman
J
Neocortical development
 , 
1991
New York
Raven Press
Bignami
A
Dahl
D
Astrocyte-specific protein and neuroglial differentiation. An immunofluorescence study with antibodies to the glial fibrillary acidic protein
J Comp Neurol.
 , 
1974
, vol. 
153
 (pg. 
27
-
38
)
Boulder Committee
Embryonic vertebrate central nervous system: revised terminology
Anat Rec.
 , 
1970
, vol. 
166
 (pg. 
257
-
261
)
Bravo
R
Frank
R
Blundell
PA
Macdonald-Bravo
H
Cyclin/PCNA is the auxiliary protein of DNA polymerase-delta
Nature.
 , 
1987
, vol. 
326
 (pg. 
515
-
517
)
Bystron
I
Blakemore
C
Rakic
P
Development of the human cerebral cortex: Boulder Committee revisited
Nat Rev Neurosci.
 , 
2008
, vol. 
9
 (pg. 
110
-
122
)
Calegari
F
Haubensak
W
Haffner
C
Huttner
WB
Selective lengthening of the cell cycle in the neurogenic subpopulation of neural progenitor cells during mouse brain development
J Neurosci.
 , 
2005
, vol. 
25
 (pg. 
6533
-
6538
)
Ellison
JA
de Vellis
J
Platelet-derived growth factor receptor is expressed by cells in the early oligodendrocyte lineage
J Neurosci Res.
 , 
1994
, vol. 
37
 (pg. 
116
-
128
)
Englund
C
Fink
A
Lau
C
Pham
D
Daza
RA
Bulfone
A
Kowalczyk
T
Hevner
RF
Pax6, Tbr2, and Tbr1 are expressed sequentially by radial glia, intermediate progenitor cells, and postmitotic neurons in developing neocortex
J Neurosci.
 , 
2005
, vol. 
25
 (pg. 
247
-
251
)
Gal
JS
Morozov
YM
Ayoub
AE
Chatterjee
M
Rakic
P
Haydar
TF
Molecular and morphological heterogeneity of neural precursors in the mouse neocortical proliferative zones
J Neurosci.
 , 
2006
, vol. 
26
 (pg. 
1045
-
1056
)
Ge
W
He
F
Kim
KJ
Blanchi
B
Coskun
V
Nguyen
L
Wu
X
Zhao
J
Heng
JI
Martinowich
K
, et al.  . 
Coupling of cell migration with neurogenesis by proneural bHLH factors
Proc Natl Acad Sci USA.
 , 
2006
, vol. 
103
 (pg. 
1319
-
1324
)
Hatanaka
Y
Murakami
F
In vitro analysis of the origin, migratory behavior, and maturation of cortical pyramidal cells
J Comp Neurol.
 , 
2002
, vol. 
454
 (pg. 
1
-
14
)
Haubensak
W
Attardo
A
Denk
W
Huttner
WB
Neurons arise in the basal neuroepithelium of the early mammalian telencephalon: a major site of neurogenesis
Proc Natl Acad Sci USA.
 , 
2004
, vol. 
101
 (pg. 
3196
-
3201
)
Ito
D
Imai
Y
Ohsawa
K
Nakajima
K
Fukuuchi
Y
Kohsaka
S
Microglia-specific localisation of a novel calcium binding protein, Iba1
Brain Res Mol Brain Res.
 , 
1998
, vol. 
57
 (pg. 
1
-
9
)
Kitamura
T
Onishi
M
Kinoshita
S
Shibuya
A
Miyajima
A
Nolan
GP
Efficient screening of retroviral cDNA expression libraries
Proc Natl Acad Sci USA.
 , 
1995
, vol. 
92
 (pg. 
9146
-
9150
)
Land
PW
Monaghan
AP
Expression of the transcription factor, tailless, is required for formation of superficial cortical layers
Cereb Cortex.
 , 
2003
, vol. 
13
 (pg. 
921
-
931
)
Letinic
K
Zoncu
R
Rakic
P
Origin of GABAergic neurons in the human neocortex
Nature.
 , 
2002
, vol. 
417
 (pg. 
645
-
649
)
Liu
X
Hashimoto-Torii
K
Torii
M
Haydar
TF
Rakic
P
The role of ATP signaling in the migration of intermediate neuronal progenitors to the neocortical subventricular zone
Proc Natl Acad Sci USA.
 , 
2008
, vol. 
105
 (pg. 
11802
-
11807
)
Marshall
CA
Novitch
BG
Goldman
JE
Olig2 directs astrocyte and oligodendrocyte formation in postnatal subventricular zone cells
J Neurosci.
 , 
2005
, vol. 
25
 (pg. 
7289
-
7298
)
Miyata
T
Kawaguchi
A
Okano
H
Ogawa
M
Asymmetric inheritance of radial glial fibers by cortical neurons
Neuron.
 , 
2001
, vol. 
31
 (pg. 
727
-
741
)
Miyata
T
Kawaguchi
A
Saito
K
Kawano
M
Muto
T
Ogawa
M
Asymmetric production of surface-dividing and non-surface-dividing cortical progenitor cells
Development.
 , 
2004
, vol. 
131
 (pg. 
3133
-
3145
)
Morita
S
Kojima
T
Kitamura
T
Plat-E: an efficient and stable system for transient packaging of retroviruses
Gene Ther.
 , 
2000
, vol. 
7
 (pg. 
1063
-
1066
)
Nadarajah
B
Brunstrom
JE
Grutzendler
J
Wong
RO
Pearlman
AL
Two modes of radial migration in early development of the cerebral cortex
Nat Neurosci.
 , 
2001
, vol. 
4
 (pg. 
143
-
150
)
Nakajima
K
Mikoshiba
K
Miyata
T
Kudo
C
Ogawa
M
Disruption of hippocampal development in vivo by CR-50 mAb against reelin
Proc Natl Acad Sci USA.
 , 
1997
, vol. 
94
 (pg. 
8196
-
8201
)
Nieto
M
Monuki
ES
Tang
H
Imitola
J
Haubst
N
Khoury
SJ
Cunningham
J
Gotz
M
Walsh
CA
Expression of Cux-1 and Cux-2 in the subventricular zone and upper layers II-IV of the cerebral cortex
J Comp Neurol.
 , 
2004
, vol. 
479
 (pg. 
168
-
180
)
Nishiyama
A
Lin
XH
Giese
N
Heldin
CH
Stallcup
WB
Co-localization of NG2 proteoglycan and PDGF alpha-receptor on O2A progenitor cells in the developing rat brain
J Neurosci Res.
 , 
1996
, vol. 
43
 (pg. 
299
-
314
)
Niwa
H
Yamamura
K
Miyazaki
J
Efficient selection for high-expression transfectants with a novel eukaryotic vector
Gene.
 , 
1991
, vol. 
108
 (pg. 
193
-
199
)
Noctor
SC
Martinez-Cerdeno
V
Ivic
L
Kriegstein
AR
Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases
Nat Neurosci.
 , 
2004
, vol. 
7
 (pg. 
136
-
144
)
Noctor
SC
Martinez-Cerdeno
V
Kriegstein
AR
Distinct behaviors of neural stem and progenitor cells underlie cortical neurogenesis
J Comp Neurol.
 , 
2008
, vol. 
508
 (pg. 
28
-
44
)
Ochiai
W
Minobe
S
Ogawa
M
Miyata
T
Transformation of pin-like ventricular zone cells into cortical neurons
Neurosci Res.
 , 
2007
, vol. 
57
 (pg. 
326
-
329
)
Ono
K
Takebayashi
H
Ikeda
K
Furusho
M
Nishizawa
T
Watanabe
K
Ikenaka
K
Regional- and temporal-dependent changes in the differentiation of Olig2 progenitors in the forebrain, and the impact on astrocyte development in the dorsal pallium
Dev Biol.
 , 
2008
, vol. 
320
 (pg. 
456
-
468
)
Ozaki
HS
Wahlsten
D
Timing and origin of the first cortical axons to project through the corpus callosum and the subsequent emergence of callosal projection cells in mouse
J Comp Neurol.
 , 
1998
, vol. 
400
 (pg. 
197
-
206
)
Rakic
P
Mode of cell migration to the superficial layers of fetal monkey neocortex
J Comp Neurol.
 , 
1972
, vol. 
145
 (pg. 
61
-
83
)
Roy
K
Kuznicki
K
Wu
Q
Sun
Z
Bock
D
Schutz
G
Vranich
N
Monaghan
AP
The Tlx gene regulates the timing of neurogenesis in the cortex
J Neurosci.
 , 
2004
, vol. 
24
 (pg. 
8333
-
8345
)
Sasaki
S
Tabata
H
Tachikawa
K
Nakajima
K
The cortical subventricular zone-specific molecule Svet1 is part of the nuclear RNA coded by the putative Netrin receptor gene Unc5d and is expressed in multipolar migrating cells
Mol Cell Neurosci.
 , 
2008
, vol. 
38
 (pg. 
474
-
483
)
Shoukimas
GM
Hinds
JW
The development of the cerebral cortex in the embryonic mouse: an electron microscopic serial section analysis
J Comp Neurol.
 , 
1978
, vol. 
179
 (pg. 
795
-
830
)
Sommer
L
Ma
Q
Anderson
DJ
Neurogenins, a novel family of atonal-related bHLH transcription factors, are putative mammalian neuronal determination genes that reveal progenitor cell heterogeneity in the developing CNS and PNS
Mol Cell Neurosci.
 , 
1996
, vol. 
8
 (pg. 
221
-
241
)
Stensaas
LJ
Stensaas
SS
An electron microscope study of cells in the matrix and intermediate laminae of the cerebral hemisphere of the 45 mm rabbit embryo
Z Zellforsch Mikrosk Anat.
 , 
1968
, vol. 
91
 (pg. 
341
-
365
)
Sugitani
Y
Nakai
S
Minowa
O
Nishi
M
Jishage
K
Kawano
H
Mori
K
Ogawa
M
Noda
T
Brn-1 and Brn-2 share crucial roles in the production and positioning of mouse neocortical neurons
Genes Dev.
 , 
2002
, vol. 
16
 (pg. 
1760
-
1765
)
Tabata
H
Nakajima
K
Efficient in utero gene transfer system to the developing mouse brain using electroporation: visualization of neuronal migration in the developing cortex
Neuroscience.
 , 
2001
, vol. 
103
 (pg. 
865
-
872
)
Tabata
H
Nakajima
K
Neurons tend to stop migration and differentiate along the cortical internal plexiform zones in the Reelin signal-deficient mice
J Neurosci Res.
 , 
2002
, vol. 
69
 (pg. 
723
-
730
)
Tabata
H
Nakajima
K
Multipolar migration: the third mode of radial neuronal migration in the developing cerebral cortex
J Neurosci.
 , 
2003
, vol. 
23
 (pg. 
9996
-
10001
)
Takahashi
T
Nowakowski
RS
Caviness
VS
Jr
Early ontogeny of the secondary proliferative population of the embryonic murine cerebral wall
J Neurosci.
 , 
1995
, vol. 
15
 (pg. 
6058
-
6068
)
Takahashi
T
Nowakowski
RS
Caviness
VS
Jr
Interkinetic and migratory behavior of a cohort of neocortical neurons arising in the early embryonic murine cerebral wall
J Neurosci.
 , 
1996
, vol. 
16
 (pg. 
5762
-
5776
)
Takahashi
T
Nowakowski
RS
Caviness
VS
Jr
The leaving or Q fraction of the murine cerebral proliferative epithelium: a general model of neocortical neuronogenesis
J Neurosci.
 , 
1996
, vol. 
16
 (pg. 
6183
-
6196
)
Tamamaki
N
Nakamura
K
Okamoto
K
Kaneko
T
Radial glia is a progenitor of neocortical neurons in the developing cerebral cortex
Neurosci Res.
 , 
2001
, vol. 
41
 (pg. 
51
-
60
)
Tarabykin
V
Stoykova
A
Usman
N
Gruss
P
Cortical upper layer neurons derive from the subventricular zone as indicated by Svet1 gene expression
Development.
 , 
2001
, vol. 
128
 (pg. 
1983
-
1993
)
Weissman
TA
Riquelme
PA
Ivic
L
Flint
AC
Kriegstein
AR
Calcium waves propagate through radial glial cells and modulate proliferation in the developing neocortex
Neuron.
 , 
2004
, vol. 
43
 (pg. 
647
-
661
)
Wu
SX
Goebbels
S
Nakamura
K
Kometani
K
Minato
N
Kaneko
T
Nave
KA
Tamamaki
N
Pyramidal neurons of upper cortical layers generated by NEX-positive progenitor cells in the subventricular zone
Proc Natl Acad Sci USA.
 , 
2005
, vol. 
102
 (pg. 
17172
-
17177
)
Zimmer
C
Tiveron
MC
Bodmer
R
Cremer
H
Dynamics of Cux2 expression suggests that an early pool of SVZ precursors is fated to become upper cortical layer neurons
Cereb Cortex.
 , 
2004
, vol. 
14
 (pg. 
1408
-
1420
)