TY - JOUR
AU - Xiaoqun, Wang,
AB - Abstract Cajal-Retzius (CR) cells are one of the earliest populations of neurons in the cerebral cortex of rodents and primates, and they play a critical role in corticogenesis and cortical lamination during neocortical development. However, a comprehensive morphological and physiological profile of CR cells in the mouse neocortex has not yet been established. Here, we systematically investigated the dynamic development of CR cells in Tg(Ebf2-EGFP)58Gsat/Mmcd mice. The morphological complexity, membrane activities and presynaptic inputs of CR cells coordinately increase and reach a plateau at P5–P9 before regressing. Using 3D reconstruction, we delineated a parallel-stratification pattern of the axonal extension of CR cells. Furthermore, we found that the morphological structure and presynaptic inputs of CR cells were disturbed in Reelin-deficient mice. These findings confirm that CR cells undergo a transient maturation process in layer 1 before disappearing. Importantly, Reelin deficiency impairs the formation of synaptic connections onto CR cells. In conclusion, our results provide insights into the rapid maturation and axonal stratification of CR cells in layer 1. These findings suggest that both the electrophysiological activities and the morphology of CR cells provide vital guidance for the modulation of early circuits, in a Reelin-dependent manner. 3D reconstruction, axonal stratification, Cajal-Retzius cell, development, Reelin deficiency Introduction Cajal-Retzius (CR) cells were discovered by Ramón y Cajal (1891) and Retzius (1893, 1894) and are one of the earliest neuronal cell types generated during neocortical development. Mouse CR cells are derived from 3 locations, namely the cortical hem (Takiguchi-Hayashi et al. 2004; Garcia-Moreno et al. 2007), the septum and the ventral pallidum (Bielle et al. 2005), at approximately embryonic days 10–12. These cells migrate tangentially into the early marginal zone (MZ) and cover the entire surface of the neocortex. Interestingly, the tangential migration pattern of CR cells was recently reported to be absent in chick (Nakajima et al. 2014). As a consequence of cell death, combined with dilution caused by cortical expansion and the migration to deeper layers, CR cells gradually disappear from layer 1 during postnatal neocortical development (Derer and Derer 1990; Soda et al. 2003; Chowdhury et al. 2010). Earlier studies identified CR cells based on their distinctive morphological features: an oval soma, a prominent, tapered, mainly horizontally oriented stem dendrite originating from a pole of the soma, and a slender axon emerging from the side opposite the dendrite (Derer 1987; Imamoto et al. 1994; Hestrin and Armstrong 1996; Radnikow et al. 2002; Ma et al. 2014). In more recent reports, a considerable number of specific and highly expressed genes were identified in both embryonic and postnatal CR cells (Gong et al. 2003; Yamazaki et al. 2004; Tissir et al. 2009). Using these specific genetic markers, transgenic mouse lines have been generated in which the CR cells are identifiable on the basis of their fluorescence (del Rio and DeFelipe 1997; Meskenaite 1997; Soda et al. 2003; Chowdhury et al. 2010; Chuang et al. 2011; Anstotz et al. 2014; Dzaja et al. 2014; Barber et al. 2015). One particular protein, early B-cell factor 2 (Ebf2) has been identified in a screen for CR neuron-specific markers (Yamazaki et al. 2004; Ochsner 2008) and is now commonly used to identify CR cells in mice (Hanashima et al. 2007; Chowdhury et al. 2010; Chuang et al. 2011). Early Golgi-impregnation studies identified morphological attributes of neurons in layer 1 of the neocortex (Cajal 1891; Retzius 1893, 1894). Subsequent studies provided extended and refined descriptions of CR cell morphology (Marin-Padilla 1998, 2015; Meyer et al. 1999; Fairen et al. 2002; Rakic and Zecevic 2003; Soriano and Del Rio 2005; Anstotz et al. 2014; Martinez-Cerdeno and Noctor 2014), and further studies have identified CR cells based on these morphological properties. Recent neuroanatomical studies of CR cells mainly focus on their location, axonal/dendritic length, and complexity on the basis of previous results (Radnikow et al. 2002; Portera-Cailliau et al. 2005; Anstotz et al. 2014; Ma et al. 2014; Gabbott 2016). These studies reveal that functional development and synaptic connectivity of CR cells are highly consistent with their morphological properties. However, the three-dimensional axonal distribution pattern and the quantitative morphological changes in CR cells located in the neocortex remain unresolved. Additional studies have recently characterized the physiological parameters of CR cells (Hestrin and Armstrong 1996; Radnikow et al. 2002; Soda et al. 2003; Anstotz et al. 2014). CR cells show specific patterns of action potentials (APs) and intrinsic membrane properties (Zhou and Hablitz 1996; Bystron et al. 2008) and are integrated into the neural networks of the immature neocortex (Mienville 1998; Kilb and Luhmann 2001; Radnikow et al. 2002; Soda et al. 2003). However, comprehensive dynamic changes in the membrane properties and excitability of CR cells during early development have not been addressed. Functional imaging studies suggest that CR cells participate in embryonic and postnatal cortical spontaneous activities and that they form a critical part of the neural networks in the MZ (Aguilo et al. 1999). Using viral tracing, previous studies revealed that CR cells receive synaptic inputs from interneurons in layer 1, pyramidal neurons in layer 5 and Reelin-positive neurons in layer 6 (Cocas et al. 2016). To characterize these synaptic inputs, electron microscopic studies were performed in both mice and rats; these studies showed that both GABAergic and non-GABAergic synapses are present in the somatic region and dendrite of CR cells (Radnikow et al. 2002; Molnar et al. 2003; Anstotz et al. 2014). Using receptor antagonists, CR cells in both mice and rats were shown to receive GABAergic inputs through the GABAA receptor (Radnikow et al. 2002; Chan and Yeh 2003; Soda et al. 2003). In contrast, NMDA receptor-mediated synaptic transmission was found in CR cells in rats and ICR mice but not in C57BL/6J mice (Radnikow et al. 2002; Chan and Yeh 2003). These results suggest that glutamatergic inputs to CR cells were strain-specific (Chan and Yeh 2003). However, the spatial and temporal pattern of presynaptic inputs to CR cells during the neonatal period also remains to be elucidated. Reelin is a glycoprotein mainly secreted by CR cells into the MZ during cerebral development (D’Arcangelo et al. 1995; Hirotsune et al. 1995; D’Arcangelo et al. 1997). Reelin is crucial for both neural migration and neocortical lamination (D’Arcangelo and Curran 1998; Frotscher 1998; Marin-Padilla 1998; Higashi et al. 2005; Hirota and Nakajima 2017), and it regulates the inside first-outside last layer patterning and coordination of early-generated principal neurons (Luskin and Shatz 1985; Rakic and Caviness 1995; Marin-Padilla 1998; Noctor et al. 2001). In Reelin-deficient mice, also called reeler mice, the cortical neurons failed to form an inside-out pattern (Caviness 1982). In addition, these mice exhibited changes in synaptic density, distribution, and topology in various brain regions, such as the cerebellum, hippocampus (Mariani et al. 1977; Borrell et al. 1999) and thalamus (Garcia-Moreno et al. 2018), but the characteristics of CR cells in Reelin-deficient mice have not been reported. Here, we set out to study the development of functional and morphological properties of CR cells in Ebf2-EGFP and Reelin-deficient mice. We found that from P1 to P14, CR cells undergo distinct changes in their characteristics of membrane excitability, morphological complexity, and presynaptic inputs. Moreover, we found that CR cells maintain a specific stratification during neocortical development. Last, we found that inverted cortical lamination leads to abnormal axonal lamination and to impaired presynaptic inputs to CR cells. Materials and Methods Mice All experimental procedures were conformed to the Laboratory Animal Care Center at the Institute of Biophysics, Chinese Academy of Sciences. Mice were maintained on a 12 h (7 a.m.–7 p.m.) light and dark cycle with standard mouse food and water ad libitum. Strain B6C3Fea/a-Relnrl/Nju (reeler) mice (NJU stock No. N000174) were purchased from Nanjing Biomedical Research Institute of Nanjing University. Strain Tg(Ebf2-EGFP)58Gsat/Mmcd mice were obtained from Qin Shen lab, Tsing Hua University. Histology and Immunohistochemistry For immunohistochemistry analysis after electrophysiological recording, acute slices harvested from experiments were fixed with 4% paraformaldehyde (PFA) in PBS (pH 7.4). Other histology and immunohistochemistry were performed to confirm EGFP expression and location of injected CTB, mice were perfused with PBS containing 4% paraformaldehyde (PFA). Brains were then sectioned using a vibratome after 24 h fixation in PBS containing 4% paraformaldehyde. For immunohistochemistry, brain slices were incubated in 0.1% Triton-X PBS with 10% donkey serum (DS) for 2 h, then incubated in PBS (10% DS, 0.1% triton-X) with primary antibodies overnight at 4 °C. Primary antibodies: GFP (Aves Lab, GFP-1010, 1:500), Reelin (Milipore, MAB5366, 1:500), Calretinin (Milipore, AB1550, 1:300, Swant, 6B3, 1:500), CXCR4 (Abcam, ab124824, 1:300), Caspase-3 (Cell Signaling Technology, 9661 S, 1:500). Then rinsed with PBS for 10 min 3 times, after rinsing brain slices were incubated with secondary antibodies: Texas-Red Streptavidin (Vector laboratories, SA-5006, 1:500), Goat anti-chicken 488 (Life techonologies, A-11039, 1:500), Donkey anti-mouse 647 (Life techonologies, A-31571, 1:500), Donkey anti-goat 594 (Life techonologies, A-11058, 1:500). Nucleus was then stained by DAPI (Thermo Fisher, D1306, 1:1000). Fluorescence images were scanned by confocal microscopy (Olympus FV1000) using 10×, 20×, 30× and 40× objectives. Z-series images were taken at 1μm steps at 20× or higher and 3μm steps at 10× or lower and analyzed using FLuoView (Olympus), ImageJ (National Institutes of Health) and Photoshop (Adobe Systems). Patch Clamp Recording Mice (P1–P14) were anaesthetized and decapitated. The brain was then quickly removed and submerged in oxygenated (95% O2 and 5% CO2) ice-cold sucrose-based artificial cerebrospinal fluid (sucrose-based aCSF) containing (in mM): 234 sucrose, 2.5 KCl, 26 NaHCO3, 1.25 NaH2PO4, 11 d-glucose, 0.5 CaCl2 and 10 MgSO4. Coronal slices (300 μm) were prepared using a vibratome (VT1200S, Leica) and kept in an incubating chamber filled with oxygenated aCSF (in mM): 126 NaCl, 3 KCl, 26 NaHCO3, 1.2 NaH2PO4, 10 d-glucose, 2.4 CaCl2 and 1.3 MgCl2 at 34 °C. After a recovery period at least 60 min, an individual slice was transferred to a recording chamber and was continuously superfused with oxygenated aCSF at a rate of 3–5 mL per minute at 30 ± 1 °C. An inverted microscope (Olympus) equipped with epifluorescence and infrared-differential interference contrast (DIC) illumination, a camera, and one air immersion lens (4×) and one water immersion lens (40×) were used to visualize and target electrodes, pipette and cells. Whole-cell patch clamp recording was performed on cells from the neocortex of mice. Patch pipettes had a 5–7 MΩ resistance when filled with intracellular solution (in mM): 130 potassium gluconates, 16 KCl, 2 MgCl2, 10 HEPES, 0.2 EGTA, 4 Na2-ATP, 0.4 Na3-GTP, 0.1% Texas Red and 0.5% neurobiotin (Vector laboratories, SP-1125), pH = 7.25, adjusted with KOH. Spontaneous EPSCs were recorded by holding the membrane potential at −70 mV. Inward currents were recorded in voltage-clamp mode with a basal holding potential of −60 mV followed by stimulating pulses from −80 mV to 60 mV with a step size of 10 mV. Evoked action potentials were recorded in current-clamp mode using a series of depolarizing currents ranged from −100 pA to 280 pA in increments of 20 pA. Evoked field EPSCs (fEPSC) were recorded using record glass pipettes as above 5–7 MΩ and were evoked using a concentric bipolar electrode (WPI; FHC, CBBEB75) which was used to stimulate neurons with a brief current pulse (50 ms), and the current pulse was delivered by a stimulus isolation unit (ISO-Flex, A.M.P.I., Jerusalem, Israel). Signals were filtered at 2 kHz and digitized at 100 kHz using Digidata 1440 A (Molecular Devices). Data acquisition and slope measurement were carried out using pClamp 10.2 (Molecular Devices). Pulse generation was achieved using a Master 8 stimulator (A.M.P.I., Jerusalem, Israel). Pharmacology After whole cell recording on a neuron, pharmacological experiment was performed to confirm receptor type of synaptic of neurons. While adding drugs, CNQX (Sigma, C127, 20 μM)/APV (Sigma, A8054, 100 μM)/BMI (Sigma, 14343, 20 μM) were added by bath application in a very slow speed. The slices were rinsed for 30 min with ACSF after drug treatments. l-Glutamamine (Sigma, G8540, 100 μM) was added by drug-filled glass pipette (1–5 MΩ), this experiment used Mg2+ free aCSF (in mM): 126 NaCl, 3 KCl, 26 NaHCO3, 1.2 NaH2PO4, 10 d-glucose, 2.4 CaCl2 and 0.4 sucrose to avoid inhibition from Mg2+. Morphology Tracing and 3D Reconstruction CR cells in layer 1 were randomly recognized from its EGFP, patched and filled with Neurobiotin through the recoding pipette for at least 10 min. After 5 min additional diffusion, slices were fixed in 4% PFA in 0.1 M PBS overnight. Labeled cells were visualized by incubation in 1:200 Texas-red streptavidin (Vector Laboratories) for at least 2 h. Morphology was viewed and recorded using a confocal laser scanning microscope (Olympus). Full morphological data 3D reconstructions were obtained using Vaa3d (Allen institutes) and decorated by IMARIS (Bitplane), corrected by manual works and custom-written MATLAB (Mathworks, Natick, MA) program. Dendrites were identified by CR cell dendrites typical characteristics. The rest of them were recognized as axons. Data were analyzed by custom-written Matlab-based program. All images were processed by Photoshop (Adobe Systems) and ImageJ. CTB Injection and Surgical Preparation All mice (P3-P7) were anaesthetized with ice. And then rapidly moved to heating pad at 37 °C. Volume of Cholera Toxin B Subunit (CTB) (100 nL, 1 mg/mL in PBS, Thermo Fisher C34777) injected in each animal were kept constant for each age group. Location of injections sites was confirmed by checking CTB 594 fluorescence on each brain. We used glass microelectrode pipette (Drummond #5-000-1001-X10) with a tip diameter of ~20 μm for injection. Pipette was carefully pierced into skull and pial surface under microscope (Olympus). Mice were then housed in their home cages for at least 2 days before experiments. Then we harvested coronal slices (300 μm) at P7-P9. Fluorescence images were scanned by confocal microscopy (Olympus FV1000) using 10× and 40× objectives. Z-series images were taken at 1μm steps and analyzed using FLuoView, ImageJ, and Photoshop. MNI-Glutamate Uncaging with the Two-photon Method Target cells were imaged with a two-photon microscope (Scientific Inc.) equipped with a mode-locked Ti:sapphire laser (MaiTai, spectra-physics) and a water immersion objective lens (Apo40×W/NIR, NA 0.8; Nikon). The emitted photons are split into two channels and detected by photomultiplier tubes (PMT) (Hamamatsu, protected GaAsP H10770PA-40), i.e., into a green channel for Ebf2-EGFP and a red channel for Texas Red fluorescence. Multiphoton filter set (525/50 nm and 620/60 nm emission filters with 565LP dichroic) was applied to separate green and red emitting dyes. The frame rate was 30 frames/s. The compound MNI-caged-l-glutamate was purchased from Tocris (Tocris 1490). 6 mM MNI-Glutamate was photolyzed by a 720-nm beam at 3 and 16 mW from a Ti:sapphire laser (Coherent, 690–1080 nm). The stimulus duration was 100 ms and the photolysis procedure was controlled using a custom written LabVIEW (National Instruments) program. We used ImageJ (National Institutes of Health), FluoView (Olympus) and MATLAB (Mathworks, Natick, MA) software for image analysis. Statistical Analysis All data are presented as the mean ± SEM. Statistical significance was determined using paired or unpaired t tests and data with P values <0.05, <0.01 and <0.001 are indicated by asterisks *, ** and ***, respectively. All analyses were performed using Origin 8.0 (OriginLab) and GraphPad Prism software (Graphpad). 3D morphological data analysis was performed by using custom-written Matlab-based program. Analysis of electrophysiological recording data was performed using Clampfit 10.3. Results Immunohistochemical Characterization of EGFP-Positive Neurons in Layer 1 In our study, we used Tg(Ebf2-EGFP)58Gsat/Mmcd transgenic mice designed to express EGFP in CR cells (Yamazaki et al. 2004; Hanashima et al. 2007; Chowdhury et al. 2010) (Fig. S1A, Fig. S1B). To test whether the EGFP signal was strong enough for identifying CR cells in neocortex, we triple-immunolabeled mouse brain slices with antibodies targeting Reelin, GFP and Calretinin. We were able to accurately identify the dynamic populations of EGFP-positive cells and Reelin-positive cells during the first 2 postnatal weeks (Fig. 1). Our triple-immunolabeling method specifically detected CR cells, as identified by the presence of oval somata and horizontal stem dendrites on all EGFP-positive cells (Fig. 1A). Considering the colocalization of signals, almost all EGFP-positive cells detected at P1 were also positive for Reelin and/or Calretinin expression (Reelin+EGFP+ for 26.1 ± 3.0 × 103 per mm3, Calretinin+EGFP+ for 9.1 ± 1.4 × 103 per mm3, Reelin+Calretinin+EGFP+ for 7.1 ± 1.0 × 103 per mm3) (Fig. 1C, Fig. S1E–H), 2 representative markers of CR cells (Yamazaki et al. 2004). The observed colocalization remained at high for a further 2 weeks. To compare Ebf2 with another specific CR cell marker, CXCR4 (Anstotz et al. 2014), we used a CXCR4 antibody to perform immunostaining on tissue blocks from Ebf2-EGFP mice (Fig. S2A). CXCR4-expressing CR cells also experienced this same process of population transition as observed for Ebf2-expressing CR cells (Fig. S2). These 2 markers had almost 60% colocalization (Fig. S2B middle), indicating that these 2 subgroups of CR cells showed similar characteristics. Together, these results suggested that Ebf2-EGFP-positive cells represent CR cells located in the neocortex, and they were used as such for further experiments. Figure 1. View largeDownload slide Dynamic changes in the population of Cajal-Retzius cells in the neocortex during early development. (A) Frozen coronal sections of Ebf2-EGFP transgenic mice from postnatal day 1 (P1) to P14 were stained with antibodies targeting GFP, Reelin and Calretinin. EGFP signals revealed the population of Cajal-Retzius cells. Reelin and EGFP were highly colocalized (arrows). Calretinin-positive cells also almost entirely expressed EGFP signal (arrows). The right column shows enlargements of the white dotted boxes in the second column. (B) Whole-mount tissue blocks from somatosensory cortex of Ebf2-EGFP transgenic mice were immunostained with antibodies targeting EGFP and Reelin. EGFP-positive cells were evenly distributed in the neocortex at P1, P7, and P14. EGFP-positive cells were highly colocalized with Reelin. (C) The density of Cajal-Retzius and Reelin-positive neurons in somatosensory cortex from P1 to P14. (D) The percentages of EGFP and Reelin double-positive neurons within the overall EGFP-positive cells at P1, P7, and P14 in somatosensory cortex. (E) The percentages of cells showing EGFP and Reelin colocalization within the populations of EGFP-positive cells at P1, P7, and P14 in somatosensory cortex. DAPI: 4’,6-diamidino-2-phenylindole. All scale bars: 50 μm. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Figure 1. View largeDownload slide Dynamic changes in the population of Cajal-Retzius cells in the neocortex during early development. (A) Frozen coronal sections of Ebf2-EGFP transgenic mice from postnatal day 1 (P1) to P14 were stained with antibodies targeting GFP, Reelin and Calretinin. EGFP signals revealed the population of Cajal-Retzius cells. Reelin and EGFP were highly colocalized (arrows). Calretinin-positive cells also almost entirely expressed EGFP signal (arrows). The right column shows enlargements of the white dotted boxes in the second column. (B) Whole-mount tissue blocks from somatosensory cortex of Ebf2-EGFP transgenic mice were immunostained with antibodies targeting EGFP and Reelin. EGFP-positive cells were evenly distributed in the neocortex at P1, P7, and P14. EGFP-positive cells were highly colocalized with Reelin. (C) The density of Cajal-Retzius and Reelin-positive neurons in somatosensory cortex from P1 to P14. (D) The percentages of EGFP and Reelin double-positive neurons within the overall EGFP-positive cells at P1, P7, and P14 in somatosensory cortex. (E) The percentages of cells showing EGFP and Reelin colocalization within the populations of EGFP-positive cells at P1, P7, and P14 in somatosensory cortex. DAPI: 4’,6-diamidino-2-phenylindole. All scale bars: 50 μm. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Imaging the whole area of EGFP expression (Fig. S1B, 1C and 1D), we chose somatosensory cortex (S1, central area in Fig. S1) for investigating the molecular attributes of CR cells. To accurately characterize the dynamic population of CR cells from P1 to P14, we harvested whole-mount tissue blocks (Fig. 1B). CR cells constitute the major population in layer 1 at P1 (Fig. 1A) (85.0% ± 4.1%) (Fig. 1B). Both somata and dendrites showed strong positive signals for Reelin (Fig. 1B). The neocortex rapidly expands from P1 to P7 (Marin-Padilla 1990; Chowdhury et al. 2010). Therefore, both the density of EGFP (P1, 36.0 ± 0.2 × 103 per mm3, n = 4 and P7, 12.2 ± 0.2 × 103 per mm3, n = 3) and the Reelin signal were significantly reduced (Fig. 1C). From P7 to P14, the density of Reelin was found to be stable, while the detected EGFP signal steadily decreased (Reelin at P7, 17.7 ± 0.01 × 103 per mm3, n = 3 and at P14, 14.6 ± 0.4 × 103 per mm3, n = 4; EGFP at P7, 12.2 ± 0.2 × 103 per mm3, n = 3 and at P14, 4.3 ± 0.7 × 103 per mm3, n = 4). These results are consistent with previous reports that suggested that more non-CR cells secreting Reelin emerged during the first week after birth (Soda et al. 2003). Furthermore, Caspase3 signals emerged at P7 in Ebf2-EGFP-positive cells and became fairly strong by P14 (Fig. S2C and D). That suggested that apoptotic cell death induced numerous CR cells to disappear during the second postnatal week in mice (Chowdhury et al. 2010). Together, these results strongly suggest that CR cells are a neuronal population that is transiently present in layer 1 (Kerr et al. 1972; Chowdhury et al. 2010). To further verify these results, we calculated the proportions of EGFP, Reelin and Calretinin in S1. We found that the percentage of EGFP-positive cells expressing Reelin was stable through the first 2 postnatal weeks (P1, 79.8% ± 10.5%, n = 4; P7, 81.3% ± 1.9%, n = 3; and P14, 69.6% ± 9.9%, n = 3) (Fig. 1D). However, the proportion of Reelin-positive cells expressing EGFP decreased continuously (P1, 67.9% ± 9.0%, n = 4; P7, 52.1% ± 11.6%, n = 3; and P14, 27.3% ± 7.9%, n = 3) (Fig. 1E). Together, these results confirmed that CR cells continue to secrete Reelin during P1–P14, though the persisting Reelin-positive cells include a considerable number of non-CR cells. In summary, these findings demonstrate that Ebf2-EGFP is an accurate marker for tracing various properties of CR cells at each stage of neocortical development. Developmental Changes in the Morphological Parameters of CR Cells To characterize the dynamic morphological properties of CR cells in Ebf2-EGFP mice during neocortical development, we evaluated cell morphologies at different postnatal days by injecting neurobiotin into the CR cells (Fig. S3B). To fully acquire each morphological property of CR cells, acute brain slices and whole-mount tissue blocks of Ebf2-EGFP mice were harvested at different postnatal days. We found that the CR cells showed a typical appearance, an oval soma and a horizontally oriented stem dendrite, both in whole-mount preparation (Fig. 2A, Fig. S3C) and in coronal sections (Fig. 2C, Fig. S3A). To observe the axonal distribution of CR cells in the neocortex, we reconstructed the morphology of CR cells in whole-mount tissue blocks. These observations revealed the extended, branched axonal structure of CR cells (Fig. 2B). Axons of CR cells extended parallel with the pial surface and covered a large proportion of the neocortex at early neonatal days (Fig. 2B). CR cells maintained their complex axonal structures for several days and then started to lose their branches and reduce their coverage area (Fig. 2B). In addition, in order to verify the dendritic and axonal growth direction of CR cells, we delineated CR cells in coronal acute slices (Fig. 2D). We found that the branches of dendrites and axons of CR cells elongated mainly towards the pial surface. Many specific vertically oriented structures were present on both dendrites and axons of CR cells during postnatal development (Fig. 2D). The presence of these morphological features suggests that CR cells exhibit dynamic morphological changes during their neocortical development. Further, we found that a large percentage of branches of axons and dendrites grew vertically upwards to the pial surface, enabling the identification of CR cells by specific orthogonal axonal structures absent in other cells in the MZ. Figure 2. View largeDownload slide Multiple perspectives of Cajal-Retzius cell morphology and 3D reconstruction for detailed morphological analysis throughout the first 2 postnatal weeks. (A) Images of neurobiotin-injected Cajal-Retzius cells on tissue blocks. These injected cells were stained with Texas Red Streptavidin secondary antibody. Scale bars: 20 μm. (B) Representative reconstructed axons (red) and dendrites (blue) of Cajal-Retzius cells at different stages. Scale bars: 50 μm. (C) Coronal acute slices of Ebf2-EGFP transgenic mice with neurobiotin-filled EGFP-positive cells. Scale bars: 10 μm. (D) Morphological changes of Cajal-Retzius cells on coronal slices during early neocortical development. Numbers of vertical segments of both axons (red arrows) and dendrites (blue arrows) were observed from P3 to P9; few of these were observed at P1, P11, or P14. High-resolution micrographs of these typical vertical structures are shown in black boxes and come from the dotted black boxed images at P5 and P7. The upper border in these images represents the pial surface. Scale bars: 40 μm. (E) Photographic montage of a whole neurobiotin-filled Cajal-Retzius cell at P7. Blue arrow, dendrite; red arrow, axon. Scale bars: 50 μm. (F) 3D reconstruction of this representative cell (E). Axons (red arrow) and dendrites (blue arrow) are shown. (G–J) Lengths of total axons (G) and dendrites (I) and axonal (H) and dendritic (J) branch points from P1 to P14 were analyzed based on 3D reconstruction data. All coordinate systems are right-handed coordinate systems. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Figure 2. View largeDownload slide Multiple perspectives of Cajal-Retzius cell morphology and 3D reconstruction for detailed morphological analysis throughout the first 2 postnatal weeks. (A) Images of neurobiotin-injected Cajal-Retzius cells on tissue blocks. These injected cells were stained with Texas Red Streptavidin secondary antibody. Scale bars: 20 μm. (B) Representative reconstructed axons (red) and dendrites (blue) of Cajal-Retzius cells at different stages. Scale bars: 50 μm. (C) Coronal acute slices of Ebf2-EGFP transgenic mice with neurobiotin-filled EGFP-positive cells. Scale bars: 10 μm. (D) Morphological changes of Cajal-Retzius cells on coronal slices during early neocortical development. Numbers of vertical segments of both axons (red arrows) and dendrites (blue arrows) were observed from P3 to P9; few of these were observed at P1, P11, or P14. High-resolution micrographs of these typical vertical structures are shown in black boxes and come from the dotted black boxed images at P5 and P7. The upper border in these images represents the pial surface. Scale bars: 40 μm. (E) Photographic montage of a whole neurobiotin-filled Cajal-Retzius cell at P7. Blue arrow, dendrite; red arrow, axon. Scale bars: 50 μm. (F) 3D reconstruction of this representative cell (E). Axons (red arrow) and dendrites (blue arrow) are shown. (G–J) Lengths of total axons (G) and dendrites (I) and axonal (H) and dendritic (J) branch points from P1 to P14 were analyzed based on 3D reconstruction data. All coordinate systems are right-handed coordinate systems. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. To more precisely characterize the morphological traits of CR cells, we reconstructed images of CR cells into three-dimensional structures (Fig. 2E, F, Fig. S4A) and measured morphological traits on the basis of these 3D models (Fig. 2G–J, Fig. S4B). The maximum axonal length (P1, 7.3 ± 1.2 mm, n = 4; P5, 11.8 ± 1.3 mm, n = 8) and number of branch points (P1, 12.0 ± 4.2, n = 4; P5, 23.4 ± 5.6, n = 8) appeared at P5 (Fig. 2G and H). CR cells’ axons had fewer branches at P9, and those branches eventually disappeared completely. At P14, only a short primary axon containing few branches remained. The dendrite length increased from P1 to P5 (P1, 0.5 ± 0.04 mm, n = 4; P5, 1.2 ± 0.1 mm, n = 8) and decreased after P9 (P9, 1.1 ± 0.1 mm, n = 4; P14, 0.7 ± 0.1 mm, n = 3) (Fig. 2I). Most of the dendritic branches were found at P5 (Fig. 2J). Dendrites of CR cells also become unbranched at P14. Together, these results indicate that both the dendrites and axons of CR cells become more complex during the first postnatal week in mice. The lengths and numbers of branches of axons and dendrites peaked at P5 and reduced from P9 onward. At P14, the CR cells displayed dramatically fewer dendritic and axonal branches. Number and length of these branches represent ability of CR cells to become a component in developing circuits. The morphological degeneration and functional decline in CR cells began at the same time (Fig. 4), suggesting that the dynamic morphological changes occur because CR cells function only as a transient component in early networks. Axonal Stratification Pattern of CR Cells To assess whether CR cells have a specific axonal structure or specific stratification, we compared 3D reconstructions of CR cells from P1 to P14. During that time period, many vertical segments were scattered over the axons of CR cells (Fig. 3A), connecting the separate axonal sublayers of each CR cell (Fig. 3B–D). The complexity of these segments seemed to change during the development of CR cells (Fig. 3A and B). We therefore counted the number and length of these vertical segments. The number of segments increased in the first postnatal week but decreased in the second postnatal week (P1, 2.0 ± 0.6, n = 4; P7, 6.8 ± 1.0, n = 8; P14, 1.7 ± 0.3, n = 3) (Fig. 3E). In addition, these structures reached their maximal length at P7 and gradually became shorter during the second week (P1, 13.1 ± 1.3 μm, n = 4; P7, 31.6 ± 1.1 μm, n = 8; P14, 15.8 ± 2.0 μm, n = 3) (Fig. 3F). Together, these results indicate that CR cells have a specific stratification during neocortical development. Changes in CR cell complexity are consistent with the dynamic population and length of these vertical segments. Figure 3. View largeDownload slide Stable axonal and dendritic stratification structures of Cajal-Retzius cells during early neocortical development. (A) 3D reconstructed Cajal-Retzius cells at different stages had similar stratification structures. Scale bars: 50 μm. (B) 3D reconstructed Cajal-Retzius cell at P7 to show axonal (red arrow) and dendritic (blue arrow) vertical structures. Dotted black boxes 1, 2, 3, and 4 are shown at higher magnification in black boxes 1, 2, 3, and 4, displaying fine structures of these vertical segments dividing the complete axons and dendrites into two laminae. (C) Drawing highlighting the dendritic vertical structures in (B). Images are rotated around the red axis. (D) Stratification structure of Cajal-Retzius cell in (B). Dotted red lines showed the upper and lower bounds of the Cajal-Retzius cell. (E, F) Population (E) and average length (F) of axonal vertical structures in Cajal-Retzius cells from P1 to P14. (G) 3D reconstruction of two overlapping Cajal-Retzius cells. Black arrows indicate cell bodies. (H) Two overlapping Cajal-Retzius cells distributed in a narrow space, with dotted black lines indicating their boundaries. (I) Two overlapping Cajal-Retzius cells share a common convex surface. (J) Magnification of separate axons from black boxes 1 and 2 in (G), revealing the accuracy of our 3D reconstruction. Directions of cells in (A), (D), (H), (I) are shown with the bars (V for ventral and D for dorsal). All coordinate systems are right-handed coordinate systems. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Figure 3. View largeDownload slide Stable axonal and dendritic stratification structures of Cajal-Retzius cells during early neocortical development. (A) 3D reconstructed Cajal-Retzius cells at different stages had similar stratification structures. Scale bars: 50 μm. (B) 3D reconstructed Cajal-Retzius cell at P7 to show axonal (red arrow) and dendritic (blue arrow) vertical structures. Dotted black boxes 1, 2, 3, and 4 are shown at higher magnification in black boxes 1, 2, 3, and 4, displaying fine structures of these vertical segments dividing the complete axons and dendrites into two laminae. (C) Drawing highlighting the dendritic vertical structures in (B). Images are rotated around the red axis. (D) Stratification structure of Cajal-Retzius cell in (B). Dotted red lines showed the upper and lower bounds of the Cajal-Retzius cell. (E, F) Population (E) and average length (F) of axonal vertical structures in Cajal-Retzius cells from P1 to P14. (G) 3D reconstruction of two overlapping Cajal-Retzius cells. Black arrows indicate cell bodies. (H) Two overlapping Cajal-Retzius cells distributed in a narrow space, with dotted black lines indicating their boundaries. (I) Two overlapping Cajal-Retzius cells share a common convex surface. (J) Magnification of separate axons from black boxes 1 and 2 in (G), revealing the accuracy of our 3D reconstruction. Directions of cells in (A), (D), (H), (I) are shown with the bars (V for ventral and D for dorsal). All coordinate systems are right-handed coordinate systems. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. We then wanted to ascertain whether sublayers of different CR cells coordinate at the same surface, we reconstructed several tangled CR cells in a 3D model (Fig. 3G and H). Axons from different CR cells occupied the same lamination and seldom contact with each other (Fig. 3I and J). These results suggest that axons of different cells are arranged in parallel with 3D reconstruction (Fig. 3J). These observations indicate that the axons of different CR cells are distributed in a usually narrow region of layer 1 (Fig. 3A and Fig. S4C). In conclusion, our results indicate that the axons and dendrites of CR cells maintain a specific stratification and distribution pattern so that CR cells are able to connect a group of neurons in a specific sublayer. Membrane Properties of CR Cells During Early Postnatal Development To verify whether these morphological and functional changes occur simultaneously, whole-cell patch-clamp recording was performed at different postnatal stages. We first studied the passive electrical properties of the CR cells, such as the resting membrane potential (RMP, Fig. 4A) and input resistance (Rin, Fig. 4B). Previous studies showed that the RMP of CR cells undergoes a hyperpolarization process after birth. Our results showed that the CR cells were significantly depolarized at P1 compared with the cells at P9 (Fig. 4A). The Rin also gradually decreased after P1 and reached its lowest value by P5 (Fig. 4B). Figure 4. View largeDownload slide Quantitative and dynamic electrophysiological properties of layer 1 Cajal-Retzius cells at different developmental stages. (A) Resting membrane potential of the Cajal-Retzius cell somata. (B) Input resistance of Cajal-Retzius cell membranes. (C) Voltage responses elicited by a series of current injections. (D) Firing rates of action potentials at different developmental stages. (E) Firing rates of action potentials evoked by an inward current of 100 pA. (F–H) Threshold (F), amplitude (G), and half-width (H) of evoked action potentials. (I) Representative current responses evoked by a series of voltage steps. (J, K) The amplitudes of outward (J) and inward (K) currents at different postnatal days. (L) The amplitude of outward current evoked by a voltage step from −60 mv to 60 mV. (M) The amplitude of inward current evoked by a voltage step from −60 mV to 20 mV. Green bars represent the extremes for each category of data. RMP: resting membrane potential; AP: action potential. Corresponding P-value heat maps (on the right side of each figure) reveal differences in the relevant data according to the Two-sample T-test. The color-coded bar represents –ln(P values) *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Figure 4. View largeDownload slide Quantitative and dynamic electrophysiological properties of layer 1 Cajal-Retzius cells at different developmental stages. (A) Resting membrane potential of the Cajal-Retzius cell somata. (B) Input resistance of Cajal-Retzius cell membranes. (C) Voltage responses elicited by a series of current injections. (D) Firing rates of action potentials at different developmental stages. (E) Firing rates of action potentials evoked by an inward current of 100 pA. (F–H) Threshold (F), amplitude (G), and half-width (H) of evoked action potentials. (I) Representative current responses evoked by a series of voltage steps. (J, K) The amplitudes of outward (J) and inward (K) currents at different postnatal days. (L) The amplitude of outward current evoked by a voltage step from −60 mv to 60 mV. (M) The amplitude of inward current evoked by a voltage step from −60 mV to 20 mV. Green bars represent the extremes for each category of data. RMP: resting membrane potential; AP: action potential. Corresponding P-value heat maps (on the right side of each figure) reveal differences in the relevant data according to the Two-sample T-test. The color-coded bar represents –ln(P values) *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. To evaluate the developmental tuning in the excitability of CR cells, we measured the threshold, the amplitude, the half-width and the firing rate of APs from P1 to P14. We elicited AP trains by injecting a series of currents into the somata of CR cells (Fig. 4C). Our results showed that with increasing current injection, the firing rate increased in a linear manner, reaching its maximum value at stimulation with 100 pA. Beyond 100 pA, the firing rate was slightly reduced (Fig. 4D). From P1 to P7, the ability of CR cells to fire APs increased, with the firing frequency remaining steady for 2–4 days before gradually regressing (Fig. 4E). For calculations, we defined the AP threshold as the membrane potential when the rising rate of voltage exceeds 10 V/s (Valiullina et al. 2016). The AP threshold at P5 was significantly lower than on the previous days, and it reached its minimum value at P9. Subsequently, the AP threshold value then returned to a depolarized state (Fig. 4F). Next, we measured the amplitude of APs from the onset of the AP to the peak. AP amplitude peaked at P9 and then declined rapidly (Fig. 4G). The half-width of APs also indicates that membrane properties of CR cells became more mature at P7 (Fig. 4H). Together, our results demonstrate that early development of CR cells occurs in 2 stages. In the first stage, from P1 to P5, CR cells display a higher AP threshold, a smaller AP amplitude and slower AP kinetics and are unable to maintain the firing of AP trains. These properties become more mature at P7 and reach their peak values from P7 to P9. In the second stage, from P9 to P14, excitability of CR cells shows a remarkable decline. These results suggest that CR cells undergo a transient maturation before undergoing selective cell death (Fig. S2C and 2D) (Anstotz et al. 2014). Excitability of CR cells was then investigated by measuring the current response evoked by a series of voltage pulses from P1 to P14 (Fig. 4I). Our results showed that amplitude of inward currents increased from P1 to P9 and began to decrease after P9 (Fig. 4K and M), whereas the amplitude of outward currents peaked at P11 before declining (Fig. 4J and L). These changes were consistent with the developmental changes in the intrinsic properties of CR cells. The specific developmental pattern of physiological properties of CR cells implies that CR cells play a transient but functional role in the early neocortical circuits. Ebf2-EGFP positive cells mainly derived from ventral pallium (Yamazaki et al. 2004; Hanashima et al. 2007; Chowdhury et al. 2010). Recently works reported that Ebf2-EGFP cells also migrated from septum and cortical hem (Chiara et al. 2012). We thus tried to figure out whether the physiological and morphological properties of Ebf2-EGFP cells originating from different originating areas were consistent. Using Calretinin (Calr) staining approach, previous studies were able to distinguish septum-derived (Calr–) CR cells from ventral pallium- and cortical hem-derived (Calr+) CR cells (Bielle et al. 2005; Chiara et al. 2012). We performed patch clamp recording on these 2 subgroups of Ebf2-EGFP+ CR cells (Fig. S6A) and found the membrane capacity (Cm), input resistance (Rin) and resting membrane potential (RMP) were identical (Fig. S6B). The firing rate, kinetics of action potential, evoked inward and outward current from septum-derived Ebf2-EGFP+ cells were same as ventral pallium- and cortical hem-derived Ebf2-EGFP+ cells at P1, P7, and P14 (Fig. S6 C–F). The perimeter of soma and the length of stem dendrite were consistent between these 2 subgroups (Fig. S7). These results demonstrated that physiological and morphological properties were consistent between Ebf2-EGFP+ CR cells from different origins, implying the process of CR cells’ maturation could occur after migration into the neocortex and is independent of CR cells’ origins. Characterization of Presynaptic Input to the CR Cells Since we observed dye coupling between CR cells and other cells during patch-clamp recording (Fig. S5), we next wondered whether connections of CR cells are developmentally regulated in neocortex. To address this question, we injected CTB-594 into the layer 1 at P5 to reveal axon terminals of other neurons around CR cells (Fig. 5A, Fig. S8A and 8B). We found many retrogradely labeled neurons in layer 1 and layer 2/3 (Fig. 5A). Based on our previous results, CR cells are a major population in layer 1 during P1-P7. Thus, it appears that these CTB-marked neurons in layer 1 and layer 2/3 were mainly located upstream of CR cells, and we therefore placed our stimulating electrodes in layer 1 or layer 2/3 to elicit presynaptic inputs onto CR cells (Fig. 5B and C), and recorded the excitatory postsynaptic currents (EPSCs) in CR cells. After choosing an appropriate stimulus intensity (Fig. S8C), we measured the EPSCs elicited by presynaptic stimulation. The EPSC amplitudes at P7 were notably larger than those at P1 and P14 (Layer 1, P1, 39.9 ± 5.4 pA, n = 11; P7, 82.6 ± 10.1 pA, n = 14; P14, 37.2 ± 7.4 pA, n = 8; layer 2/3, P1, 23.8 ± 4.3 pA, n = 6; P7, 55.7 ± 10.1 pA, n = 6; P14, 31.9 ± 5.9 pA, n = 5) (Fig. 5D, E and Fig. S8C). These data suggest that the amplitude peak of synaptic input occurs at P7, coinciding with the time of maximal excitability (Fig. 4). Figure 5. View largeDownload slide Synaptic properties and types of Ebf2-EGFP-positive cells at various developmental stages. (A) CTB-594 was injected into layer 1 of Ebf2-EGFP transgenic mice at P5, and brain sections were harvested at P7. Neurons in layer 1 (most of which are Cajal-Retzius cells (Fig. 1)) were wide-ranging, as shown by retrograde tracing. Cells upstream of these neurons were revealed by red signals in the images. Scale bars: 50 μm. (B) After acquiring the approximate upstream inputs to CR cells as in (A), stimulation and recording pipettes were set in the locations indicated in the graph. The stimulation and patch-clamp recording pipettes were separated by a constant distance of 200 μm. (C) Schematic chart describing the electrical stimulation experiment and cell distribution. (D) Representative fEPSC traces at P1, P7, and P14. The representative black traces are the averages of correlated individual experiments (gray traces) for each age group. (E) Amplitude of fEPSCs in (D) at P1, P7 and P14 separately in layer 1 and layer 2/3. (F, H) Addition of a mixture of the glutamatergic receptor antagonists CNQX and APV (red traces) did not produce a significant effect, whereas the GABAA receptor antagonist BMI (blue traces) impaired the increase in the fEPSC amplitude of Cajal-Retzius cells (black traces). (G, I) Amplitudes of fEPSCs in (F) and (H). (J) Local glutamate uncaging for Ebf2-EGFP-negative cells (J) and Ebf2-EGFP-positive cells (CR cells) (M). Scale bar is 20 μm. In both (J) and (M), Texas Red and neurobiotin were mixed in the intracellular solution to reveal the dendrites of patch-clamped cells. (K, N) Color-coded boxes in (J) and (M) indicate the stimulated areas. MNI-caged-l-glutamate evoked EPSC traces shown in (K) and (N). (L, O) Current responses invoked by stimuli of −40 pA and 80 pA indicated the firing patterns of patch-clamped cells. (P) GFP and Neurobiotin immunostaining of the cell in (M). Scale bar is 20 μm. (Q) Amplitudes of EPSCs elicited by different stimulus intensities in the cyan-outlined ROIs in Ebf2-negative (J) and Ebf2-positive (M) cells. Images are shown at magnifications of 5× and 7×. (R) Amplitudes of EPSCs at 5× magnification and 3 mW stimulation in Ebf2-negative and Ebf2-positive cells. CNQX: 6-cyano-7-nitroquinoxaline-2,3-dione; APV: (2R)-amino-5-phosphonopentanoatic acid; BMI: 1(S),9(R) - (−)-Bicuculline methiodide. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Figure 5. View largeDownload slide Synaptic properties and types of Ebf2-EGFP-positive cells at various developmental stages. (A) CTB-594 was injected into layer 1 of Ebf2-EGFP transgenic mice at P5, and brain sections were harvested at P7. Neurons in layer 1 (most of which are Cajal-Retzius cells (Fig. 1)) were wide-ranging, as shown by retrograde tracing. Cells upstream of these neurons were revealed by red signals in the images. Scale bars: 50 μm. (B) After acquiring the approximate upstream inputs to CR cells as in (A), stimulation and recording pipettes were set in the locations indicated in the graph. The stimulation and patch-clamp recording pipettes were separated by a constant distance of 200 μm. (C) Schematic chart describing the electrical stimulation experiment and cell distribution. (D) Representative fEPSC traces at P1, P7, and P14. The representative black traces are the averages of correlated individual experiments (gray traces) for each age group. (E) Amplitude of fEPSCs in (D) at P1, P7 and P14 separately in layer 1 and layer 2/3. (F, H) Addition of a mixture of the glutamatergic receptor antagonists CNQX and APV (red traces) did not produce a significant effect, whereas the GABAA receptor antagonist BMI (blue traces) impaired the increase in the fEPSC amplitude of Cajal-Retzius cells (black traces). (G, I) Amplitudes of fEPSCs in (F) and (H). (J) Local glutamate uncaging for Ebf2-EGFP-negative cells (J) and Ebf2-EGFP-positive cells (CR cells) (M). Scale bar is 20 μm. In both (J) and (M), Texas Red and neurobiotin were mixed in the intracellular solution to reveal the dendrites of patch-clamped cells. (K, N) Color-coded boxes in (J) and (M) indicate the stimulated areas. MNI-caged-l-glutamate evoked EPSC traces shown in (K) and (N). (L, O) Current responses invoked by stimuli of −40 pA and 80 pA indicated the firing patterns of patch-clamped cells. (P) GFP and Neurobiotin immunostaining of the cell in (M). Scale bar is 20 μm. (Q) Amplitudes of EPSCs elicited by different stimulus intensities in the cyan-outlined ROIs in Ebf2-negative (J) and Ebf2-positive (M) cells. Images are shown at magnifications of 5× and 7×. (R) Amplitudes of EPSCs at 5× magnification and 3 mW stimulation in Ebf2-negative and Ebf2-positive cells. CNQX: 6-cyano-7-nitroquinoxaline-2,3-dione; APV: (2R)-amino-5-phosphonopentanoatic acid; BMI: 1(S),9(R) - (−)-Bicuculline methiodide. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. To test the specificity of the synaptic receptors of CR cells, we used the antagonists of glutamatergic and GABAergic receptors. The evoked responses displayed little sensitivity to the combination of CNQX and APV (Fig. 5F and 5G), specific antagonists for AMPA and NMDA receptors, respectively. In contrast, these responses were notably subdued by bicuculline methiodide (BMI) (Fig. 5H, I and Fig. S8D), a specific antagonist of GABAA receptors. To confirm these results, we puffed l-glutamine around CR cells to elicit glutamatergic responses (Fig. S9A). Although no response was detected when l-glutamine was puffed on somata or proximal or distal dendrites of CR cells (Fig. S9B and 9C), we recorded responses from non-CR cells (Fig. S9D). Considering how widely puffed l-glutamine can spread, we performed two-photon photolysis of MNI-caged-l-glutamate (4-methoxy-7-nitroindolinyl-caged-l-glutamate) in parts of dendrites of individual Ebf2-EGFP-negative and Ebf2-EGFP-positive cells (Fig. 5J–P). Using a separate stimulus ROI and a different intensity (Fig. 5J and M), strong current responses were recorded from non-CR cells (Fig. 5K). However, only a few CR cells were capable of showing EPSCs (Fig. 5Q), and the amplitude of these CR cell EPSCs remained extremely small (intensity of stimulus at 16 mW, CR cells, 2.6167 ± 2.0700 pA, n = 13 cells, non-CR cells, 86.7367 ± 18.5870 pA, n = 10 cells) (Fig. 5Q and R). These uncaging results indicate that only a minimal fraction of Ebf2-EGFP-positive CR cells respond to glutamatergic inputs and that these responses are significantly lower in intensity than those of non-CR cells. In summary, our results showed that the synaptic strength of CR cells increases from P1 to P7 and subsequently decreases without regaining its original strength. The peak of synaptic strength appeared at approximately P7, coinciding with the peaks of membrane excitability and morphological complexity of CR cells. GABAergic presynaptic inputs were detected in all CR cells. In comparison, only weak glutamatergic presynaptic inputs of a few CR cells were recorded in Ebf2-EGFP mice. This is consistent with previous studies showing that CR cells display strain-specific complements of receptors. CR cells in Reelin-Deficient Mice To identify the features of CR cells in Reelin-deficient mice, we generated a double-transgenic mouse line by crossing two existing lines: Tg(Ebf2-EGFP)58Gsat/Mmcd and B6C3Fea/a-Relnrl/Nju. We observed a substantial number of EGFP-positive but Reelin-negative neurons intermingled with other neurons in the superficial region of neocortex in a disordered arrangement (Fig. 6A). In the reeler mice, the somata of EGFP-positive cells were found to be confined to a thin part of the outermost layer of the neocortex, in contrast to a more regular distribution of CR cells in the control Ebf2-EGFP mice (Fig. 6B). We reconstructed the morphology of these CR cells in reeler mice and control mice in a 3D model. Furthermore, we found that these EGFP-positive cells in reeler mice also displayed extensive and branched axons (Fig. 6C and Fig. S10A). When we measured the axonal lengths in reeler and control mice, we found no significant difference (reeler, 7.6 ± 0.9 mm, n = 5; control, 10.3 ± 0.7 mm, n = 8) (Fig. 6D). However, we found that the numbers of axonal branch points were significantly reduced in reeler mice (reeler, 10.0 ± 1.7, n = 5; control, 22.8 ± 2.8, n = 8) (Fig. 6E). We also found that the specific vertical-oriented structures observed in control mice (Fig. 3) disappeared in reeler mice (Fig. 6F, G, and Fig. S10B). Furthermore, few extended inverted structures were found in reeler mice (Fig. 6F). These morphological disruptions indicate that the Reelin deficiency disturbs the synaptic wiring of CR cells. Figure 6. View largeDownload slide Membrane properties, morphology and synaptic connectivity of Cajal-Retzius cells in Reelin-deficient mice. (A) Whole-mount tissue blocks from Ebf2-EGFP transgenic mice and reeler mice at P7 were stained with antibodies for GFP and Reelin. EGFP signals revealed the populations of Cajal-Retzius cells. (B) Coronal acute sections of Ebf2-EGFP and reeler mouse brains at P7 were stained with GFP and DAPI. Dotted white lines indicated the superficial area of the mouse brain. (C) Representative morphology of Ebf2-EGFP-positive cells in control (red) and reeler mice (cyan). (D, E) Lengths (D) and branch points (E) of total axons in Cajal-Retzius cells. (F, G) The dotted black boxed regions in (C) are shown at higher magnification in (F) and (G). Red and blue arrows indicate axonal and dendritic vertical structures, respectively, in Cajal-Retzius cells. The upper bounds of these images are the pial surface. (H) Cajal-Retzius cells in reeler mice were injected with neurobiotin for morphological tracing. Slices after injection were stained with Texas Red Streptavidin, anti-Reelin and DAPI. White dotted lines noted the superficial surface of neocortical. (I) Voltage responses elicited by a series of injected current of Cajal-Retzius cells in control (red) and reeler (cyan) mice. (J) Firing rates of action potentials of control (red) and Reeler (control) mice. (K, L) Input resistance of Cajal-Retzius cell membranes (K) and resting membrane potentials of the Cajal-Retzius cell somata (L) in control (red) and reeler (cyan) mice. (M) Traces of representative evoked EPSCs from CR cells in control (red) and reeler (cyan) mice. (N) Statistics of amplitude in (M) at P7. All coordinate systems are right-handed coordinate systems. All scale bars: 50 μm. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. Figure 6. View largeDownload slide Membrane properties, morphology and synaptic connectivity of Cajal-Retzius cells in Reelin-deficient mice. (A) Whole-mount tissue blocks from Ebf2-EGFP transgenic mice and reeler mice at P7 were stained with antibodies for GFP and Reelin. EGFP signals revealed the populations of Cajal-Retzius cells. (B) Coronal acute sections of Ebf2-EGFP and reeler mouse brains at P7 were stained with GFP and DAPI. Dotted white lines indicated the superficial area of the mouse brain. (C) Representative morphology of Ebf2-EGFP-positive cells in control (red) and reeler mice (cyan). (D, E) Lengths (D) and branch points (E) of total axons in Cajal-Retzius cells. (F, G) The dotted black boxed regions in (C) are shown at higher magnification in (F) and (G). Red and blue arrows indicate axonal and dendritic vertical structures, respectively, in Cajal-Retzius cells. The upper bounds of these images are the pial surface. (H) Cajal-Retzius cells in reeler mice were injected with neurobiotin for morphological tracing. Slices after injection were stained with Texas Red Streptavidin, anti-Reelin and DAPI. White dotted lines noted the superficial surface of neocortical. (I) Voltage responses elicited by a series of injected current of Cajal-Retzius cells in control (red) and reeler (cyan) mice. (J) Firing rates of action potentials of control (red) and Reeler (control) mice. (K, L) Input resistance of Cajal-Retzius cell membranes (K) and resting membrane potentials of the Cajal-Retzius cell somata (L) in control (red) and reeler (cyan) mice. (M) Traces of representative evoked EPSCs from CR cells in control (red) and reeler (cyan) mice. (N) Statistics of amplitude in (M) at P7. All coordinate systems are right-handed coordinate systems. All scale bars: 50 μm. Two-sample T-test *P < 0.05; **P < 0.01; ***P < 0.001. Data are shown as the mean ± SEM. To confirm the existence of functional changes caused by miswiring, we employed patch-clamp recordings of EGFP-positive cells in reeler mice (Fig. 6H, Fig. S10E and 8F). We found that the RMP (Fig. 6L), input resistance (Fig. 6K) and firing pattern of APs (Fig. 6I and J) of the EGFP-positive cells in reeler mice were equivalent to those of their counterparts in control mice at P7. These observations suggested that the disorganized wiring in reeler mice does not affect the membrane excitability of CR cells. Furthermore, we estimated the presynaptic inputs onto CR cells in reeler mice (Fig. S10C). We observed that the amplitudes of current responses to the local electrical stimulation were significantly decreased in reeler mice compared with those in control mice (reeler, 45.4 ± 7.4 pA, n = 11; control, 82.6 ± 10.1 pA, n = 14) (Fig. 6M, N and Fig. S10D). In conclusion, we generated a new double-transgenic mouse line to identify CR cells in Reelin-deficient mice, and we found that the membrane excitability in Reelin-deficient mice was consistent with that in control mice. In contrast, the morphological and synaptic properties of CR cells were disturbed in Reelin-deficient mice. Discussion Our investigation revealed that development of membrane activities, axons and connectivity of CR cells almost synchronously reach the peak at P7. We observed that CR cells maintain a unique stratification pattern during early development, which suggested that CR cells would contact with a specific group of neurons. In Reelin-deficient mice, all of these specific axonal stratifications and synaptic connections of CR cells were disrupted, but the innate development was not influenced. Previous reports showed that CR cells have primary dendrites with vertical side branches and large-scale spread horizontal axonal projections that are involved in the complex circuits of the early neocortex (Borrell et al. 1999). Our results based on 3D reconstruction analysis indicate that dendritic length and number of branches of CR cells show bell-shaped developmental profiles, which is consistent with the process by which CR cell excitability is modulated. These results suggest that CR cells are important active elements in the early neocortical network. Although the somata of CR cells are located at various depths in layer 1 (Ma et al. 2014), we observed that their axonal projections were stratified in a single narrow band. Our data suggest that axonal projections of various CR cells form a dense mesh structure that parallels the pial surface. Axons of CR cells extended to another sublayer via specific orthogonal segments to form a relatively sparse network. This layer-restricted axonal stratification suggests that CR cells produce synaptic inputs to specific neurons whose dendrites elongate into this lamination. Furthermore, these results suggest that through these specific synaptic connections, dense axonal networks of CR cells modulate the activities of their postsynaptic targets to produce the discrete functional radial circuits. Previous studies showed that neocortical neurons are organized into functional columns with highly specific connections (Hubel and Wiesel 1962; Yoshimura et al. 2005). These exquisite microcircuits arise from radial clones produced by the same mother cell during cortical development (Yu et al. 2009). The pyramidal neurons in the column form synaptic connections in the MZ (Marin-Padilla 1990, 1998), and the spontaneous activity in layer 1 may serve as a scaffold for the development of intracortical connections (Aguilo et al. 1999). A previous report revealed that CR cells migrate tangentially and participate in the correlated network in layer 1 (Aguilo et al. 1999). The tangential migration of CR cells is observed to be an innovation of mammals (Metin et al. 2007; Garcia-Moreno et al. 2018), and it might be a substantial condition of forming a complex neocortex. Our results from patch-clamp recording indicate that CR cells had high input resistances and low firing rates. These properties of CR cells indicate that CR cells serve as components that convert the spreading of spontaneous activity in layer 1 into discrete spikes. Overall, these results suggest that the spatial and temporal patterns of outputs from CR cells become cues to help their prospective synaptic targets to form a fine-scale functional columnar architecture. During the first postnatal week, CR cells gradually achieved neuronal activity patterns resembling those of mature cells and maintained this state for several days (P5–9). This period coincides with the time windows of the emergence of many functional radial structures, such as the appearance of barrel rings (Wu et al. 2011), and with the switch in connectivity from gap junctions to chemical synapses in ontogenetic columns (Yu et al. 2009, 2012). During the second postnatal week, CR cells exhibited decreased excitability and underwent an apoptotic process that was promoted by GABAergic signaling (Blanquie et al. 2017). Together, our results demonstrated that the developmental process of electrophysiological activities of CR cells is consistent in period with structure and molecule developmental events in neocortex. On the basis of these findings, we propose that the excitability of CR cells strengthens the formation of the functional structure of the neocortex. Previously published imaging results showed that CR cells participated in the spontaneous electrical activity in layer 1 (Aguilo et al. 1999). Our electrical stimulation experiments provide strong evidence that the presynaptic inputs into CR cells are already present at P1. Delivering electrical stimulation in different layers, we provide further proof that the inputs into CR cells originate from layers 1–3 of neocortex, consistent with previous reports (Cocas et al. 2016). Our results indicate that the developmental processes for the amplitudes of presynaptic inputs, membrane activity, and morphology are the same. We also noticed that presynaptic inputs of CR cells were mainly mediated by GABAA receptors. However, we identified tiny glutamatergic inputs into a small fraction of CR cells. These results are confirmation of earlier studies reporting strain-specific receptors on CR cells (Chan and Yeh 2003). Together, our results indicate that CR cells function as components of early neocortical circuits and integrate GABAergic signals not only from layer 1 but also from deeper layers and modulate their postsynaptic targets during the functional development of the neocortex. CR cells can be distinguished from other layer 1 neurons based on several criteria, such as Reelin expression and morphological properties. However, layer 1 is not discernible, and the positioning of cells comprising the other layers is inverted in Reelin-deficient mice. In these mice, morphological criteria for identifying CR cells are absent, as many abnormal non-CR cells also show a horizontally extended primary dendrite beneath the surface of the neocortex (Terashima et al. 1983). Therefore, little is known about the developmental profile of CR cells in Reelin-deficient mice. We identified CR cells in Reelin-deficient mice by their expression of EGFP. Our results showed that CR cells were pushed to the outermost region of cortex in the mutant. The vertical segments of their dendrites and axons, which were observed in Ebf2-EGFP mice, were absent in the Reelin-deficient mice. These results in Reelin-deficient mice demonstrate a disruption of presynaptic wiring of CR cells in the absence of Reelin, also confirmed that Reelin deficiency had little effect on axon elongation but clearly affected synaptic density, distribution, and topology (Mariani et al. 1977; Borrell et al. 1999). These results indicate that CR cells are integral components in the electrical activities of early neural networks, but this function was impaired in Reelin-deficient mice. Together, our findings highlight a developmental window, P5–P9, in studies of the development of CR cells. Further studies must characterize the molecular signals resulting in the functional maturation of CR cells during this window. Our results about specific axonal stratification and presynaptic inputs of CR cells will help to identify specific upstream and downstream neurons of CR cells by using viral tracing. The mechanism by which Reelin deficiency disrupts CR cells is still unknown. Therefore, further studies are needed to better understand the molecular mechanisms of Reelin activity in CR cells themselves and to understand the specific roles of CR cells in the establishment of early circuits. Funding This work was supported by the National Basic Research Program of China (2014CB964600) to X.W., the National Natural Science Foundation of China (NSFC) (31400939) to L.S., the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA16020601)to X. W., Newton Advanced Fellowship (NA140416) to X.W., Youth Innovation Promotion Association of the Chinese Academy of Sciences to Q.W. and National Natural Science Foundation of China (31171039) to Q.S. Notes We thank Dr. Na Pan for comments on the manuscript and technique support, and members of the Wang laboratory for discussions. X.W., L.S., and R.C. designed the experiment, L.S., R.C. and Y.B. conducted the experiments, L.S. and R.C. performed the patch clamp recording and analyzed the data. L.S. injected neurobiotin and CTB. R.C. performed analysis of mice brain tissue. J.L. crossed the mice and confirmed mouse genotype. X.W. conceived and supervised the project. X.W., L.S., R.C., Q. S. and Q. W. wrote the manuscript. Conflict of Interest: The authors declare no competing financial interests. References Aguilo A , Schwartz TH , Kumar VS , Peterlin ZA , Tsiola A , Soriano E , Yuste R . 1999 . Involvement of Cajal-Retzius neurons in spontaneous correlated activity of embryonic and postnatal layer 1 from wild-type and reeler mice . J Neurosci . 19 : 10856 – 10868 . Google Scholar Crossref Search ADS PubMed WorldCat Anstotz M , Cosgrove KE , Hack I , Mugnaini E , Maccaferri G , Lubke JH . 2014 . Morphology, input-output relations and synaptic connectivity of Cajal-Retzius cells in layer 1 of the developing neocortex of CXCR4-EGFP mice . Brain Struct Funct . 219 : 2119 – 2139 . Google Scholar Crossref Search ADS PubMed WorldCat Barber M , Arai Y , Morishita Y , Vigier L , Causeret F , Borello U , Ledonne F , Coppola E , Contremoulins V , Pfrieger FW , et al. . 2015 . Migration speed of cajal-retzius cells modulated by vesicular trafficking controls the size of higher-order cortical areas . Curr Biol . 25 : 2466 – 2478 . Google Scholar Crossref Search ADS PubMed WorldCat Bielle F , Griveau A , Narboux-Neme N , Vigneau S , Sigrist M , Arber S , Wassef M , Pierani A . 2005 . Multiple origins of Cajal-Retzius cells at the borders of the developing pallium . Nat Neurosci . 8 : 1002 – 1012 . Google Scholar Crossref Search ADS PubMed WorldCat Blanquie O , Liebmann L , Hubner CA , Luhmann HJ , Sinning A . 2017 . NKCC1-mediated GABAergic signaling promotes postnatal cell death in neocortical Cajal-Retzius cells . Cereb Cortex . 27 : 1644 – 1659 . Google Scholar PubMed WorldCat Borrell V , Del Rio JA , Alcantara S , Derer M , Martinez A , D’Arcangelo G , Nakajima K , Mikoshiba K , Derer P , Curran T , et al. . 1999 . Reelin regulates the development and synaptogenesis of the layer-specific entorhino-hippocampal connections . J Neurosci . 19 : 1345 – 1358 . Google Scholar Crossref Search ADS PubMed WorldCat Bystron I , Blakemore C , Rakic P . 2008 . Development of the human cerebral cortex: Boulder Committee revisited . Nat Rev Neurosci . 9 : 110 – 122 . Google Scholar Crossref Search ADS PubMed WorldCat Cajal SRY . 1891 . Sur la structure de l’e´corce ce´re´brale de quelques mammife’res . Cellule . 7 : 125 – 176 . WorldCat Caviness VS Jr. 1982 . Development of neocortical afferent systems: studies in the reeler mouse . Neurosci Res Program Bull . 20 : 560 – 569 . Google Scholar PubMed WorldCat Chan CH , Yeh HH . 2003 . Enhanced GABA(A) receptor-mediated activity following activation of NMDA receptors in Cajal-Retzius cells in the developing mouse neocortex . J Physiol . 550 : 103 – 111 . Google Scholar Crossref Search ADS PubMed WorldCat Chiara F , Badaloni A , Croci L , Yeh ML , Cariboni A , Hoerder-Suabedissen A , Consalez GG , Eickholt B , Shimogori T , Parnavelas JG , et al. . 2012 . Early B-cell factors 2 and 3 (EBF2/3) regulate early migration of Cajal-Retzius cells from the cortical hem . Dev Biol . 365 : 277 – 289 . Google Scholar Crossref Search ADS PubMed WorldCat Chowdhury TG , Jimenez JC , Bomar JM , Cruz-Martin A , Cantle JP , Portera-Cailliau C . 2010 . Fate of cajal-retzius neurons in the postnatal mouse neocortex . Front Neuroanat . 4 : 10 . Google Scholar PubMed WorldCat Chuang SM , Wang Y , Wang Q , Liu KM , Shen Q . 2011 . Ebf2 marks early cortical neurogenesis and regulates the generation of cajal-retzius neurons in the developing cerebral cortex . Dev Neurosci . 33 : 479 – 493 . Google Scholar Crossref Search ADS PubMed WorldCat Cocas LA , Fernandez G , Barch M , Doll J , Zamora Diaz I , Pleasure SJ . 2016 . Cell type-specific circuit mapping reveals the presynaptic connectivity of developing cortical circuits . J Neurosci . 36 : 3378 – 3390 . Google Scholar Crossref Search ADS PubMed WorldCat del Rio MR , DeFelipe J . 1997 . Double bouquet cell axons in the human temporal neocortex: relationship to bundles of myelinated axons and colocalization of calretinin and calbindin D-28k immunoreactivities . J Chem Neuroanat . 13 : 243 – 251 . Google Scholar Crossref Search ADS PubMed WorldCat Derer P . 1987 . [Demonstration of Cajal-Retzius cells and their processes in the neocortex of newborn mice using horseradish peroxidase] . C R Acad Sci III . 304 : 61 – 66 . Google Scholar PubMed WorldCat Derer P , Derer M . 1990 . Cajal-Retzius cell ontogenesis and death in mouse brain visualized with horseradish peroxidase and electron microscopy . Neuroscience . 36 : 839 – 856 . Google Scholar Crossref Search ADS PubMed WorldCat Dzaja D , Hladnik A , Bicanic I , Bakovic M , Petanjek Z . 2014 . Neocortical calretinin neurons in primates: increase in proportion and microcircuitry structure . Front Neuroanat . 8 : 103 . Google Scholar PubMed WorldCat D’Arcangelo G , Curran T . 1998 . Reeler: new tales on an old mutant mouse . Bioessays . 20 : 235 – 244 . Google Scholar Crossref Search ADS PubMed WorldCat D’Arcangelo G , Miao GG , Chen SC , Soares HD , Morgan JI , Curran T . 1995 . A protein related to extracellular matrix proteins deleted in the mouse mutant reeler . Nature . 374 : 719 – 723 . Google Scholar Crossref Search ADS PubMed WorldCat D’Arcangelo G , Nakajima K , Miyata T , Ogawa M , Mikoshiba K , Curran T . 1997 . Reelin is a secreted glycoprotein recognized by the CR-50 monoclonal antibody . J Neurosci . 17 : 23 – 31 . Google Scholar Crossref Search ADS PubMed WorldCat Fairen A , Morante-Oria J , Frassoni C . 2002 . The surface of the developing cerebral cortex: still special cells one century later . Prog Brain Res . 136 : 281 – 291 . Google Scholar Crossref Search ADS PubMed WorldCat Frotscher M . 1998 . Cajal-Retzius cells, Reelin, and the formation of layers . Curr Opin Neurobiol . 8 : 570 – 575 . Google Scholar Crossref Search ADS PubMed WorldCat Gabbott PL . 2016 . “Subpial Fan Cell” – a class of Calretinin neuron in layer 1 of adult monkey prefrontal cortex . Front Neuroanat . 10 : 28 . Google Scholar Crossref Search ADS PubMed WorldCat Garcia-Moreno F , Anderton E , Jankowska M , Begbie J , Encinas JM , Irimia M , Molnar Z . 2018 . Absence of tangentially migrating glutamatergic neurons in the developing avian brain . Cell Rep . 22 : 96 – 109 . Google Scholar Crossref Search ADS PubMed WorldCat Garcia-Moreno F , Lopez-Mascaraque L , De Carlos JA . 2007 . Origins and migratory routes of murine Cajal-Retzius cells . J Comp Neurol . 500 : 419 – 432 . Google Scholar Crossref Search ADS PubMed WorldCat Gong S , Zheng C , Doughty ML , Losos K , Didkovsky N , Schambra UB , Nowak NJ , Joyner A , Leblanc G , Hatten ME , et al. . 2003 . A gene expression atlas of the central nervous system based on bacterial artificial chromosomes . Nature . 425 : 917 – 925 . Google Scholar Crossref Search ADS PubMed WorldCat Hanashima C , Fernandes M , Hebert JM , Fishell G . 2007 . The role of Foxg1 and dorsal midline signaling in the generation of Cajal-Retzius subtypes . J Neurosci . 27 : 11103 – 11111 . Google Scholar Crossref Search ADS PubMed WorldCat Hestrin S , Armstrong WE . 1996 . Morphology and physiology of cortical neurons in layer I . J Neurosci . 16 : 5290 – 5300 . Google Scholar Crossref Search ADS PubMed WorldCat Higashi S , Hioki K , Kurotani T , Kasim N , Molnar Z . 2005 . Functional thalamocortical synapse reorganization from subplate to layer IV during postnatal development in the reeler-like mutant rat (shaking rat Kawasaki) . J Neurosci . 25 : 1395 – 1406 . Google Scholar Crossref Search ADS PubMed WorldCat Hirota Y , Nakajima K . 2017 . Control of neuronal migration and aggregation by Reelin signaling in the developing cerebral cortex . Front Cell Dev Biol . 5 : 40 . Google Scholar Crossref Search ADS PubMed WorldCat Hirotsune S , Takahara T , Sasaki N , Hirose K , Yoshiki A , Ohashi T , Kusakabe M , Murakami Y , Muramatsu M , Watanabe S , et al. . 1995 . The reeler gene encodes a protein with an EGF-like motif expressed by pioneer neurons . Nat Genet . 10 : 77 – 83 . Google Scholar Crossref Search ADS PubMed WorldCat Hubel DH , Wiesel TN . 1962 . Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex . J Physiol . 160 : 106 – 154 . Google Scholar Crossref Search ADS PubMed WorldCat Imamoto K , Karasawa N , Isomura G , Nagatsu I . 1994 . Cajal-Retzius neurons identified by GABA immunohistochemistry in layer I of the rat cerebral cortex . Neurosci Res . 20 : 101 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Kerr JF , Wyllie AH , Currie AR . 1972 . Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics . Br J Cancer . 26 : 239 – 257 . Google Scholar Crossref Search ADS PubMed WorldCat Kilb W , Luhmann HJ . 2001 . Spontaneous GABAergic postsynaptic currents in Cajal-Retzius cells in neonatal rat cerebral cortex . Eur J Neurosci . 13 : 1387 – 1390 . Google Scholar Crossref Search ADS PubMed WorldCat Luskin MB , Shatz CJ . 1985 . Studies of the earliest generated cells of the cat’s visual cortex: cogeneration of subplate and marginal zones . J Neurosci . 5 : 1062 – 1075 . Google Scholar Crossref Search ADS PubMed WorldCat Ma J , Yao XH , Fu Y , Yu YC . 2014 . Development of layer 1 neurons in the mouse neocortex . Cereb Cortex . 24 : 2604 – 2618 . Google Scholar Crossref Search ADS PubMed WorldCat Mariani J , Crepel F , Mikoshiba K , Changeux JP , Sotelo C . 1977 . Anatomical, physiological and biochemical studies of the cerebellum from Reeler mutant mouse . Philos Trans R Soc Lond B Biol Sci . 281 : 1 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat Marin-Padilla M . 1990 . Three-dimensional structural organization of layer I of the human cerebral cortex: a Golgi study . J Comp Neurol . 299 : 89 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Marin-Padilla M . 1998 . Cajal-Retzius cells and the development of the neocortex . Trends Neurosci . 21 : 64 – 71 . Google Scholar Crossref Search ADS PubMed WorldCat Marin-Padilla M . 2015 . Human cerebral cortex Cajal-Retzius neuron: development, structure and function. A Golgi study . Front Neuroanat . 9 : 21 . Google Scholar PubMed WorldCat Martinez-Cerdeno V , Noctor SC . 2014 . Cajal, Retzius, and Cajal-Retzius cells . Front Neuroanat . 8 : 48 . Google Scholar PubMed WorldCat Meskenaite V . 1997 . Calretinin-immunoreactive local circuit neurons in area 17 of the cynomolgus monkey, Macaca fascicularis . J Comp Neurol . 379 : 113 – 132 . Google Scholar Crossref Search ADS PubMed WorldCat Metin C , Alvarez C , Moudoux D , Vitalis T , Pieau C , Molnar Z . 2007 . Conserved pattern of tangential neuronal migration during forebrain development . Development . 134 : 2815 – 2827 . Google Scholar Crossref Search ADS PubMed WorldCat Meyer JH , Lee S , Wittenberg GF , Randall RD , Gruol DL . 1999 . Neurosteroid regulation of inhibitory synaptic transmission in the rat hippocampus in vitro . Neuroscience . 90 : 1177 – 1183 . Google Scholar Crossref Search ADS PubMed WorldCat Mienville JM . 1998 . Persistent depolarizing action of GABA in rat Cajal-Retzius cells . J Physiol . 512 ( Pt 3 ): 809 – 817 . Google Scholar Crossref Search ADS PubMed WorldCat Molnar Z , Kurotani T , Higashi S , Yamamoto N , Toyama K . 2003 . Development of functional thalamocortical synapses studied with current source-density analysis in whole forebrain slices in the rat . Brain Res Bull . 60 : 355 – 371 . Google Scholar Crossref Search ADS PubMed WorldCat Nakajima M , Gorlich A , Heintz N . 2014 . Oxytocin modulates female sociosexual behavior through a specific class of prefrontal cortical interneurons . Cell . 159 : 295 – 305 . Google Scholar Crossref Search ADS PubMed WorldCat Noctor SC , Flint AC , Weissman TA , Dammerman RS , Kriegstein AR . 2001 . Neurons derived from radial glial cells establish radial units in neocortex . Nature . 409 : 714 – 720 . Google Scholar Crossref Search ADS PubMed WorldCat Ochsner KN . 2008 . The social-emotional processing stream: five core constructs and their translational potential for schizophrenia and beyond . Biol Psychiatry . 64 : 48 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat Portera-Cailliau C , Weimer RM , De Paola V , Caroni P , Svoboda K . 2005 . Diverse modes of axon elaboration in the developing neocortex . PLoS Biol . 3 : e272 . Google Scholar Crossref Search ADS PubMed WorldCat Radnikow G , Feldmeyer D , Lubke J . 2002 . Axonal projection, input and output synapses, and synaptic physiology of Cajal-Retzius cells in the developing rat neocortex . J Neurosci . 22 : 6908 – 6919 . Google Scholar Crossref Search ADS PubMed WorldCat Rakic P , Caviness VS Jr. 1995 . Cortical development: view from neurological mutants two decades later . Neuron . 14 : 1101 – 1104 . Google Scholar Crossref Search ADS PubMed WorldCat Rakic S , Zecevic N . 2003 . Emerging complexity of layer I in human cerebral cortex . Cereb Cortex . 13 : 1072 – 1083 . Google Scholar Crossref Search ADS PubMed WorldCat Retzius G . 1893 . Die Cajalschen Zellen der Grosshirnrinde beim Menschen und bei Säugetieren . Biol Unters . 5 : 1 – 9 . WorldCat Retzius G . 1894 . Weitere Beiträge zur Kenntnis der Cajalschen Zellen der Grosshirnrinde des Menschen . Biol Unters . 6 : 29 – 34 . WorldCat Soda T , Nakashima R , Watanabe D , Nakajima K , Pastan I , Nakanishi S . 2003 . Segregation and coactivation of developing neocortical layer 1 neurons . J Neurosci . 23 : 6272 – 6279 . Google Scholar Crossref Search ADS PubMed WorldCat Soriano E , Del Rio JA . 2005 . The cells of cajal-retzius: still a mystery one century after . Neuron . 46 : 389 – 394 . Google Scholar Crossref Search ADS PubMed WorldCat Takiguchi-Hayashi K , Sekiguchi M , Ashigaki S , Takamatsu M , Hasegawa H , Suzuki-Migishima R , Yokoyama M , Nakanishi S , Tanabe Y . 2004 . Generation of reelin-positive marginal zone cells from the caudomedial wall of telencephalic vesicles . J Neurosci . 24 : 2286 – 2295 . Google Scholar Crossref Search ADS PubMed WorldCat Terashima T , Inoue K , Inoue Y , Mikoshiba K , Tsukada Y . 1983 . Distribution and morphology of corticospinal tract neurons in reeler mouse cortex by the retrograde HRP method . J Comp Neurol . 218 : 314 – 326 . Google Scholar Crossref Search ADS PubMed WorldCat Tissir F , Ravni A , Achouri Y , Riethmacher D , Meyer G , Goffinet AM . 2009 . DeltaNp73 regulates neuronal survival in vivo . Proc Natl Acad Sci U S A . 106 : 16871 – 16876 . Google Scholar Crossref Search ADS PubMed WorldCat Valiullina F , Akhmetshina D , Nasretdinov A , Mukhtarov M , Valeeva G , Khazipov R , Rozov A . 2016 . Developmental changes in electrophysiological properties and a transition from electrical to chemical coupling between excitatory layer 4 neurons in the rat barrel cortex . Front Neural Circuits . 10 : 1 . Google Scholar Crossref Search ADS PubMed WorldCat Wu CS , Ballester Rosado CJ , Lu HC . 2011 . What can we get from ‘barrels’: the rodent barrel cortex as a model for studying the establishment of neural circuits . Eur J Neurosci . 34 : 1663 – 1676 . Google Scholar Crossref Search ADS PubMed WorldCat Yamazaki H , Sekiguchi M , Takamatsu M , Tanabe Y , Nakanishi S . 2004 . Distinct ontogenic and regional expressions of newly identified Cajal-Retzius cell-specific genes during neocorticogenesis . Proc Natl Acad Sci U S A . 101 : 14509 – 14514 . Google Scholar Crossref Search ADS PubMed WorldCat Yoshimura Y , Dantzker JL , Callaway EM . 2005 . Excitatory cortical neurons form fine-scale functional networks . Nature . 433 : 868 – 873 . Google Scholar Crossref Search ADS PubMed WorldCat Yu YC , Bultje RS , Wang X , Shi SH . 2009 . Specific synapses develop preferentially among sister excitatory neurons in the neocortex . Nature . 458 : 501 – 504 . Google Scholar Crossref Search ADS PubMed WorldCat Yu YC , He S , Chen S , Fu Y , Brown KN , Yao XH , Ma J , Gao KP , Sosinsky GE , Huang K , et al. . 2012 . Preferential electrical coupling regulates neocortical lineage-dependent microcircuit assembly . Nature . 486 : 113 – 117 . Google Scholar Crossref Search ADS PubMed WorldCat Zhou FM , Hablitz JJ . 1996 . Postnatal development of membrane properties of layer I neurons in rat neocortex . J Neurosci . 16 : 1131 – 1139 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes Le Suna and Ruiguo Chen contributed equally to this work. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Morphological and Physiological Characteristics of Ebf2-EGFP-Expressing Cajal-Retzius Cells in Developing Mouse Neocortex
JF - Cerebral Cortex
DO - 10.1093/cercor/bhy265
DA - 2019-08-14
UR - https://www.deepdyve.com/lp/oxford-university-press/morphological-and-physiological-characteristics-of-ebf2-egfp-aObdJSaY0n
SP - 3864
VL - 29
IS - 9
DP - DeepDyve
ER -