Dab1 is required for glia-independent somal translocation of early-born neurons. (A) Illustration of the strategy to inactivate Dab1. Embryos are electroporated in utero with Dcx-CRE-iGFP at E12.5 to target early-born neurons migrating primarily by glia-independent translocation. PP, preplate; VZ, ventricular zone. (B) Positioning defects of early-born Dab1-deficient neurons. Coronal sections from heterozygous (left panel) and homozygous (right panel) Dab1^fl embryos electroporated at E12.5 and analyzed at E17.5. Electroporated neurons are in green, nuclei in blue. (C) Correct polarization and process extension by Dab1-deficient neurons. Coronal sections from heterozygous (left panels) and homozygous (right panels) Dab1^fl embryos electroporated at E12.5 and analyzed at E14.5. Electroporated neurons are in green, nuclei in blue. Arrowheads identify leading processes. Quantification is % of electroporated cells in the CP, ± SEM. *, P < 0.0001 by Student’s t-test. (D) Morphological defects of Dab1-deficient neurons. Coronal sections from heterozygous (top panels) and homozygous (bottom panels) Dab1^fl embryos electroporated at E12.5 and analyzed at E15.5. Migrating electroporated neurons are in green. Immunostaining for Map2 (red) identifies the borders of the CP and SP. CP, cortical plate; IZ, intermediate zone; MZ, marginal zone; SP, subplate; SVZ, subventricular zone. V, layer V; VI, layer VI. Scale bars: 50 µm (B) and 25 µm (C–D). See also Figure S1.

Morphology and polarity of migrating Dab1-deficient neurons is normal at early stages, but defective during the final phase of migration. (A–B) Morphological analysis of migrating neurons. Dcx-CRE-iGFP was electroporated into heterozygous (left panels) or homozygous (right panels) Dab1^fl embryos at E15.5, followed by 3D reconstruction of GFP-positive neurons at E18.5 (A) or P3 (B). (C–D) Immunohistochemical analysis of polarity of migrating neurons. Electroporations at E15.5 as in (A–B), followed by nuclear staining (ToPro3, blue) and immunostaining for the Golgi marker GM130 (red) at E18.5 (C) or P3 (D). (E) Control and Dab1-deficient neurons initially make contact with the MZ. Electroporations at E15.5 as in (A–B), followed by analysis at E18.5. Electroporated neurons are shown in green. ToPro3 stain reveals the borders between the CP and MZ. (F) Dab1-deficient neurons fail to maintain contact with the MZ. Quantification of the % of electroporated neurons contacting the MZ 3 days (E18.5) and 7 days (P3) after electroporation. The data represent mean ± SEM. *, P < 0.001 by Student’s t-test. Scale bars: 10 µm. See also Fig. S3 and Movies S1–4.

Rap1 is required for glia-independent translocation. (A) Illustration of signaling pathways downstream of reelin. Molecules highlighted in different colors represent effectors targeted in (C). (B) Experimental design for manipulating downstream effectors during glia-independent translocation. Mutant forms of the downstream effectors depicted in (A) were expressed from the Dcx-IRES-GFP vector by in utero elecroporation. (C–D) Perturbing Rap1, but not other reelin downstream effectors, disrupts glia-independent translocation. (C) Coronal sections from embryos electroporated with the indicated constructs at E12.5 and analyzed at E16.5. Electroporated neurons are shown in green, nuclei in blue. (D) Coronal sections from embryos electroporated at E12.5 with GFP and either non-silencing control or shRNA constructs targeting Rap1a. Analysis performed at E16.5. (E) Quantification of % neurons from (C) and (D) that entered the cortical plate. The data represent mean ± SEM. *, P < 0.0001 by Student’s t-test. (F–G) Rap1GAP does not affect initial polarization or process extension. (F) Polarity of neurons electroporated at E12.5 and analyzed at E14.5. Internal polarity is shown by immunostaining for the Golgi marker GM130 (red) and nuclear staining with ToPro3 (blue). (G) Morphological analysis of neurons electroporated as in (F). Arrows point to polarized processes extending to the cortical plate. Abbreviations as in Fig. 1. Scale bars: 50 µm (C, D), 20 µm (F–G). See also Fig. S4.

Similar neocortical lamination defects in Dab1-NESTINko and Dab1-NEXko mice. (A) Immunostaining for Dab1 at E16.5 shows that Dab1 is expressed throughout the cortical wall in Dab1^fl/fl (No CRE) embryos, but is lost throughout the cortex in Dab1-NESTINko and specifically from neurons in Dab1-NEXko embryos. Note Dab1 immunoreactivity in RGCs in the VZ in Dab1-NEXko. (B) ToPro3 nuclear stain at E16.5 demonstrates disrupted cytoarchitecture in Dab1 mutant embryos compared to control. Dotted lines identify the borders between the pia and CP. (C–D) Failure to split the preplate in Dab1 mutants. Immunostaining for Map2 (C) and CSPG (D) at E16.5 reveals the SP in control embryos and its absence in mutants. Sections are counterstained with ToPro3 (blue). White arrowheads frame the SP. (E) Disrupted lamination in adult mutants. Top panels: Nissl-stained neocortex from control and mutant mice at P30. Bottom panels: higher magnification images demonstrating loss of layer I in mutants. (F) Quantification of the indicated layer markers at P30. Graphs represent positive cells in each of 10 equal-size vertical bins expressed as % total positive cells, ± S.E.M. Approximate positions of cortical layers are identified by alternating gray and white shaded areas. Scale bars: 50 µm (A–D), 250 µm (E, top) and 50 µm (E, bottom). Neocortical layers are labeled I–VI. WM, white matter. See also Figure S6.

Dab1 is required for migration of late-born neurons into upper layers. (A) Illustration of the strategy to inactivate Dab1. Embryos are electroporated in utero with Dcx-CRE-iGFP at E15.5 to target late-born neurons that migrate first by multipolar migration, then by glia-guided locomotion and finally terminal translocation. Radial glia cells are shown in black and neurons are colored according to migration modes. MZ, marginal zone; UCP, upper cortical plate; LCP, lower cortical plate; IZ, intermediate zone; SVZ, subventricular zone; VZ, ventricular zone. (B–E) Defective translocation in late-born Dab1-deficient neurons. Coronal sections from heterozygous (left panels) and homozygous (right panels) Dab1^fl embryos electroporated at E15.5 and analyzed at E17.5 (B), P0 (C), P3 (D) or P8 (E). Electroporated neurons are in green, nuclei in blue. Quantification of cell positions shown at right. Graphs represent positive cells in each of 10 equal-size vertical bins expressed as % of total electroporated cells, ± S.E.M. (F) Cell-autonomous displacement of mutant upper layer neurons. Coronal sections from E15.5 electroporations performed as in (E) and analyzed at P8. Immunostaining for Cux1 (red) reveals upper-layer neurons located ectopically in deep layers after Dab1 deletion. Abbreviations as in Fig. 1. Cortical layers are labeled I–VI. Scale bars: 50 µm. See also Figure S2.

Dab1-deficient neurons fail to undergo terminal somal translocation. (A) Images from time-lapse experiments tracking electroporated neurons. Dab1^+/fl and Dab1^fl/fl embryos were electroporated with Dcx-CRE-iGFP at E15.5 and processed for slice cultures at E18.5. Migrating neurons approaching the MZ were imaged. Top panels show a control (Dab1^+/fl) neuron undergoing terminal translocation over an 18 hr period. Middle panels demonstrate a mutant (Dab1^fl/fl (1)) neuron failing to complete terminal translocation and instead retracting the leading process (at t=15 h) and extending new processes (at t=18 h). Bottom panels show a mutant (Dab1^fl/fl (2)) neuron that does not undergo terminal translocation, but whose leading process remains in the MZ. Arrows identify the leading processes and asterisks label the somas of the sampled neurons. To the right, traces represent migrating neurons at 0 h (red), 9 h (green) and 18 h (blue). (B–C) Distances from the MZ to the leading edge (B) or cell soma (C) of sampled neurons from (A) are plotted for each time point. The space occupied by the MZ is shaded gray. CP, cortical plate; MZ, marginal zone. Scale bar: 50 µm. See also Movies S5–8.

Cadherins are required downstream of Rap1 for glia-independent translocation. (A) Immunostaining with a pan-Cdh antibody shows cadherin expression throughout the cortex at E15.5. (B) Adherens-junction-like structures in the E15.5 marginal zone. Electron micrographs of coronal sections. Different cell types were distinguished by cytoskeletal specializations and are shaded in different colors (neuronal processes, blue; RGC endfeet, orange, other cells, green). Arrows point to junctional structures. (C) Positioning defects of neurons expressing dominant-negative cadherin. Coronal sections from embryos electroporated at E12.5 and analyzed at E16.5. Electroporated neurons are shown in green, nuclei in blue. Quantification of % neurons reaching the CP shown at right. The data represent mean ± SEM. *, P < 0.0001 by Student’s t-test. (D) Morphological analysis of neurons electroporated as in (C). Arrows point to processes extending to the cortical plate. Internal polarity is shown by immunostaining for the Golgi marker GM130 (red) and nuclear staining with ToPro3 (blue). (E) Coronal sections from embryos electroporated at E12.5 with non-silencing control or an shRNA construct targeting Cdh2. Analysis performed at E16.5. Quantification of % neurons reaching the CP shown at right. The data represent mean ± SEM. *, P < 0.001 by Student’s t-test. (F) Cdh2 overexpression rescues the migration defects of Rap1GAP-electroporated neurons. Coronal sections from E12.5 electroporations with Rap1GAP alone or together with Cdh2, and analyzed at E16.5. Electroporated neurons are green, nuclei blue. Quantification of % neurons reaching the CP is shown at right. The data represent mean ± SEM from 3 separate experiments. *, P < 0.0001 by Student’s t-test. Abbreviations as in Fig. 1. Scale bars: 10 µm (A, right panel; D, right panel), 20 µm (D, left panel), 50 µm (A, left panel; B, insets; C, E), 200 µm (B). See also Fig. S5.

Normal preplate splitting but disrupted neocortical lamination in Dab1-CUX2ko mice. (A–C) The preplate is split in Dab1-CUX2ko embryos. Immunostaining at E15.5 for Map2 (A) and CSPG (B) reveals the subplate region below the developing CP in control (Dab1^+/fl) and Dab1-CUX2ko mutant (Dab1^fl/fl) embryos. Nuclei are in blue. Dotted lines identify the borders between the MZ and CP (upper), and between the CP and SP (lower). (C) Merged images from (A–B). (D–I) Immunostaining for layer markers in P30 control and Dab1-CUX2ko mice demonstrates that deep layers form normally, but superficial layers are misplaced in mutants. (D) Immunostaining for Cux1 shows upper-layer neurons misplaced beneath deep-layer neurons in mutants. Sections counterstained with BoBo1. (E) Ctip2 immunostaining reveals layer V–VI neurons located 0–200 µm below the pia in the mutants. Sections counterstained with ToPro3. (F) Immunostaining for Tbr1 shows layer VI neurons positioned 100–200 µm below the mutant cortical surface. Arrowheads frame a population of Tbr1-positive axons coursing approximately 300 µm below the cortical surface in mutants. Sections counterstained with ToPro3. (G) Double-labeling with Tbr1 and Ctip2 in mutant sections demonstrates that Ctip2-positive layer V neurons (red) are positioned above double-positive (yellow) layer VI neurons. Arrows identify yellow double-stained nuclei. (H–I) Immunostaining for Tbr1 (H) and Smi32 (I) shows bundles of axons running transversely through the mutant neocortex. Dotted lines frame the axons. II/III, layers II–III; IV, layer IV; V, layer V; VI, layer VI; AX, axons; WM, white matter. Scale bars: 50 µm (A–C, G–I) and 100 µm (D–F). See also Figure S7.