Downregulation of functional Reelin receptors in projection neurons implies that primary Reelin action occurs at early/premigratory stages

Abstract

Reelin signaling is essential for correct development of the mammalian brain. Reelin binds to apolipoprotein E receptor 2 and very low-density lipoprotein receptor and induces phosphorylation of Dab1. However, when and where these reactions occur is essentially unknown, and the primary function(s) of Reelin remain unclear. Here, we used alkaline phosphatase fusion of the receptor-binding region of Reelin to quantitatively investigate the localization of functional Reelin receptors (i.e., those on the plasma membrane as mature forms) in the developing brain. In the wild-type cerebral cortex, they are mainly present in the intermediate and subventricular zones, as well as in radial fibers, but much less in the cell bodies of the cortical plate. Functional Reelin receptors are much more abundant in the Reelin-deficient cortical plate, indicating that Reelin induces their downregulation and that it begins before the neurons migrate out of the intermediate zone. In the wild-type cerebellum, functional Reelin receptors are mainly present in the cerebellar ventricular zone but scarcely expressed by Purkinje cells that have migrated out of it. It is thus strongly suggested that Reelin exerts critical actions on migrating projection neurons at their early/premigratory stages en route to their final destinations, in the developing cerebral cortex and cerebellum.

Figure 1.

Construction of AP–RR36 and its ability to specifically detect FRRs. A, Schematic diagram of Reelin protein (top), control AP (middle), and AP–RR36 (bottom). Two sites that are cleaved in vivo by an unknown metalloprotease and minimum receptor-binding unit are shown in the top. B, Western blotting (WB) analysis. Culture supernatants containing either AP (lane 1) or AP–RR36 (lane 2) were separated by SDS-PAGE and proteins were transferred to a polyvinylidene difluoride membrane. The membrane was then incubated with anti-AP antibody, followed by incubation with HRP-conjugated secondary antibody and detection with a chemiluminescence kit. Positions of molecular weight markers are shown on the right in kilodaltons. C–N, AP–RR36 specifically binds to Reelin receptors. COS-7 cells were transfected with control vector (C, F, I, L), ApoER2 expression vector (D, G, J, M), or VLDLR expression vector (E, H, K, N). Two days later, the cells were incubated with control AP (C–E), AP–RR36 (F–H), AP–RR36 mixed with GST protein (I–K), or AP–RR36 mixed with GST–RAP (L–N). Cells expressing ApoER2 or VLDLR were stained with AP–RR36 (G, H, respectively). This staining was not affected by addition of GST protein (J, K) but was virtually abolished by coincubation with GST–RAP (M, N). O, Estimation of Reelin concentration. AP–RR36 (lane 1, 1.6 nm as calculated from AP-activity assay) (Flanagan et al., 2000) and Reelin (lane 2) were analyzed by Western blot using R5A antibody as described above. P, COS-7 cells expressing ApoER2 (left) or VLDLR (right) were incubated with culture medium from mock-transfected (white bars) or Reelin-transfected (gray bars) cells in the presence of sodium azide for 20 min. They were then incubated with AP–RR36, and the amount of its binding was quantitated as described in Materials and Methods (n = 3). Q, AP–RR36 induces phosphorylation of Dab1. Primary cortical neurons were incubated with the samples indicated above the lanes for 20 min at 37°C, and Dab1 phosphorylation was measured as described previously (Nakano et al., 2007). PY, Phosphotyrosine. Scale bar (in C): C–N, 100 μm.

Figure 2.

Localization of FRRs in the developing mouse cerebral cortex. A, B, E15.5 mouse brains were cut coronally and stained with AP–RR36. A magnified view of the area surrounded by a broken line in A is shown in B. Red and white bars in B indicate the region corresponding to the CP and IZ/SVZ, respectively. C, The brain slice from the same mouse as in A was stained by control AP. D, The staining by AP–RR36 is abolished in the presence of GST–RAP. Slices from an E15.5 mouse brain were stained with AP–RR36 in the presence of control GST (left) or GST–RAP (right). E–J, Localization of mRNA in the E15.5 cerebral cortex. Coronal (E–G) or sagittal (H–J) slices were hybridized with antisense probes for Reelin (E, H), ApoER2 (F, I), or VLDLR (G, J). Note the diffuse Reelin mRNA signal around the SVZ and IZ (arrowheads in E, H). Most of the ApoER2 mRNA is localized in the VZ/SVZ/IZ (F, I), whereas the VLDLR mRNA is more abundantly expressed in the CP (G, J). Scale bars: A–D, E (for E–G), and H (for H–J), 200 μm.

Figure 3.

Comparison of the localization of FRRs with those of MAP2 and Dab1. A, E13.5 mouse brains were cut coronally and stained with AP–RR36. B, C, E15.5 mouse brain slices were stained with AP–RR36 (B) or with anti-MAP2 (C). D, E, Frozen brain sections (14 μm) were immunostained with anti-Dab1 (D) or with control rabbit antibodies (E). F, G, E17.5 mouse brain slices were stained with AP–RR36 (F) or with anti-MAP2 (G). Scale bars: A, B (for B–E), and F (for F, G), 100 μm.

Figure 4.

FRRs are increased in the cerebral cortex of reeler mouse but less in that of the yotari mouse. Brain slices from E15.5 (A–D), E17.5 (E–H), and P0 (I–L) reeler mice (rl/rl; B, F, J) or heterozygote littermates (rl/+; A, E, I), and yotari mice (yot/yot; D, H, L) or heterozygote littermates (C, G, K) were stained with AP–RR36. Scale bars: A (for A–D), E (for E–H), and I (for I–L), 100 μm.

Figure 5.

Localization of FRRs in the developing cerebellum of control and reeler mice. Sagittal slices from rl/+ (A, C, G) or reeler (B, D, H) mice at E15.5 (A, B), E17.5 (C, D), or P0 (G, H) were stained with AP–RR36. The arrowheads and asterisk indicate the CVZ and DCN, respectively. n, r, and egl indicate NTZ, RLS, and EGL, respectively. E, The slice shown in C was stained with Hoechst33342. The dotted line in C and E indicates the position of Purkinje cells. F, The slice shown in H was stained with anti-calbindin; arrows in F and H indicate clusters with high amounts of FRRs. Scale bars: in A (for A, B), C (for C–E), and G (for F–H), 200 μm.

Figure 6.

Localization of FRR in the postnatal cerebellum. A, B, P3 cerebellar slice from rl/+ mouse stained with AP–RR36. C, D, RNA in situ hybridization for ApoER2. E, F, RNA in situ hybridization for VLDLR. B, D, and F are magnified views of A, C, and E, respectively. G–J, P3 cerebellar slices from reeler mouse stained with AP-RR36 (G), anti-calretinin (H), or anti-calbindin (I). J, A merged image of H, I, and Hoechst33342 staining (blue). Scale bars: A (for A, G–J), B, C (for C, E), and D (for D, F), 200 μm.

Figure 7.

Proposed model of Reelin actions. For details, see Discussion. MZ, Marginal zone; CP, cortical plate; SP, subplate; IZ, intermediate zone; SVZ, subventricular zone; VZ, ventricular zone; PC, Purkinje cell; CVZ, cerebellar ventricular zone; EGL, external granular layer; RL, rhombic lip.