Reelin differentially shapes dendrite morphology of medial entorhinal cortical ocean and island cells

Abstract

The function of medial entorhinal cortex layer II (MECII) excitatory neurons has been recently explored. MECII dysfunction underlies deficits in spatial navigation and working memory. MECII neurons comprise two major excitatory neuronal populations, pyramidal island and stellate ocean cells, in addition to the inhibitory interneurons. Ocean cells express reelin and surround clusters of island cells that lack reelin expression. The influence of reelin expression by ocean cells and interneurons on their own morphological differentiation and that of MECII island cells has remained unknown. To address this, we used a conditional reelin knockout (RelncKO) mouse to induce reelin deficiency postnatally in vitro and in vivo. Reelin deficiency caused dendritic hypertrophy of ocean cells, interneurons and only proximal dendritic compartments of island cells. Ca2+ recording showed that both cell types exhibited an elevation of calcium frequencies in RelncKO, indicating that the hypertrophic effect is related to excessive Ca2+ signalling. Moreover, pharmacological receptor blockade in RelncKO mouse revealed malfunctioning of GABAB, NMDA and AMPA receptors. Collectively, this study emphasizes the significance of reelin in neuronal growth, and its absence results in dendrite hypertrophy of MECII neurons.

Keywords: Dendritic growth; Island cells; Medial entorhinal cortex and postnatal development; Ocean cells; Reelin.

Fig. 1.

Immunolabelling of ocean cells and island cells from mouse brain. (A-D) Confocal images (taken at 4× magnification) from wild-type mouse P10 parasagittal brain section. (A,E) Staining with the nuclear marker TO-PRO-3. (B) Immunostaining of island cells expressing Wfs1. The islands are indicated with white arrows. (C) Immunostaining of the ocean cells expressing reelin. (D) Overlay of A-C. (F-I) Higher magnification images of the area outlined in B at 40× magnification. Scale bars: 300 μm in A-D; 30 μm in E-I. MEC, medial entorhinal cortex; Vis, visual cortex; Cp, caudate putamen; Sub, subiculum; Hp, hippocampus.

Fig. 2.

Verification of reelin elimination in the Reln^cKO mice. (A-E) Confocal merge images of reelin immunofluorescence (red) and the nuclear marker TO-PRO-3 (blue) after tamoxifen injection. Reelin immunofluorescence is shown after 2 days (A), 3 days (B), 4 days (C) and 5 days (D). The negative control (no primary antibody against reelin) is shown in E. Scale bars: 50 μm. (F-O) Expression of Dab1 protein in wild-type and Reln^cKO mice. (F) Confocal merge image from P10 wild type of MECII at 10× magnification show Wfs1 immunofluorescence (green), TO-PRO-3 (blue) and Dab1 immunofluorescence (red). (G-J) Confocal images taken at 60× magnification of the area outlined in F. The overlay is shown in J. Scale bars: 200 μm for F; 20 μm for G-J. (K) Confocal merge images from P10 Reln^cKO of MECII at 10× magnification show Wfs1 immunofluorescence (green), TO-PRO-3 (blue) and Dab1 immunofluorescence (red). (L-O) Confocal images taken at 60× magnification of the area outlined in K. The overlay is shown in O. Scale bars: 200 μm for K; 20 μm for L-O. (P) Box and whisker plot of mean fluorescence intensity of Dab1 immunoreactivity in wild-type and Reln^cKO mice. The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. The number of the analysed MEC slices are indicated above the box plots. The data were obtained from two independent preparations. The Mann–Whitney U-test revealed a statistically significant difference between the two groups (***P<0.001).

Fig. 3.

Effect of reelin on ocean cell dendritic growth in OTCs and in vivo. OTCs were transfected with EGFP-N1 at 5 DIV, and fixed and stained at 10 DIV for morphometrical quantifications. (A) A flow chart of the experimental procedures. (B) Box and whisker plot shows that mean dendritic length of Reln^cKO ocean cells is significantly higher when compared with wild type. (C) Box and whisker plot shows that the mean number of dendritic segments is significantly higher in the Reln^cKO group in comparison with the wild type. (D) Box and whisker plot shows that the number of primary dendrites remained unaltered. The number of cells reconstructed per group is given above the box plot in B. (E,F) Representative images at 40× magnification of a wild-type (E) and a Reln^cKO (F) ocean cell. The respective traces are provided next to the images. Scale bars: 50 μm. The data were obtained from two independent OTC preparations. Mann–Whitney U-test (***P<0.001). (G-L) Effect of reelin on ocean cells dendritic growth in vivo. (G) A flow chart of the experimental procedures. (H) Box and whisker plot shows that the mean dendritic length of Reln^cKO ocean cells is significantly higher when compared with wild type. (I) Box and whisker plot shows that the mean number of dendritic segments is significantly higher in the Reln^cKO group in comparison with the wild-type group. (J) Box and whisker plot shows that the number of primary dendrites remained unaltered. The number of cells reconstructed per group is given above the box plot in H. The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. (K,L) Representative images at 40× magnification of a wild-type (K) and a Reln^cKO (L) ocean cell. The respective traces are provided next to the images. Scale bars: 50 μm. The data were obtained from two independent OTC preparations. Mann–Whitney U-test (***P<0.001).

Fig. 4.

Effect of reelin on island cell dendritic growth in OTCs and in vivo. OTCs were transfected with EGFP-N1 at 5 DIV, fixed and stained at 10 DIV for morphometrical quantifications. (A) Box and whisker plot shows that the mean apical dendritic length is significantly higher in Reln^cKO in comparison with wild type. (B) Box and whisker plot shows that the mean number of apical dendritic segments is significantly higher in the Reln^cKO group in comparison with the wild-type group. (C,D) Box and whisker plots show that the mean basal dendritic length (C) and mean basal dendritic segments (D) were increased in the Reln^cKO island cells in comparison with wild type. The number of cells reconstructed per group is given above the box plot in A. The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. Data were obtained from three independent OTC preparations. Mann–Whitney U-test (***P<0.001). (E,F) Example images at 40× magnification of a wild-type island cell (E) and a Reln^cKO island cell (F). The respective traces are shown below the images. Scale bars: 50 μm. (G-L) Effect of reelin on island cell dendritic growth in vivo. (G) Box and whisker plot shows that the mean apical dendritic length is significantly higher in Reln^cKO in comparison with wild type. (H) Box and whisker plot shows that the mean number of apical dendritic segments is significantly higher in the Reln^cKO group in comparison with the wild-type group. (I,J) Box and whisker plots show that the mean basal dendritic length (I) and mean basal dendritic segments (J) were increased in the Reln^cKO island cells in comparison with wild type. The number of cells reconstructed per group is given above the box plot in G. The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. Data were obtained from three independent OTC preparations. Mann–Whitney U-test (***P<0.001). (K,L) Example image at 40× magnification of a wild-type island cell (K) and a Reln^cKO island cell (L). The respective traces are shown below the images. Scale bars: 50 μm.

Fig. 5.

Cell compartment-specific regulation of dendritic complexity by reelin. The 3D reconstructed cells in Fig. 3 (ocean cells) and Fig. 4 (island cells) were subjected to Sholl analyses. (A) Sholl analyses show that the number of apical proximal dendritic intersections was significantly increased in Reln^cKO in comparison with wild-type island cells. (B) The total number of apical dendritic intersections was significantly increased in Reln^cKO island cells. (C) Sholl analyses show that the number of basal proximal dendritic intersections was significantly increased in Reln^cKO in comparison with wild-type island cells. (D) The total number of basal dendritic intersections was significantly increased in Reln^cKO island cells. (E) Sholl analyses show that ocean cell dendritic intersections were significantly increased in Reln^cKO in comparison with wild type. The increase in complexity is not cell compartment specific. (F) The total number of ocean cell dendritic intersections was significantly increased in Reln^cKO. Data are mean±s.e.m. in A,C,E. (B,D,F) The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. The total number of cells analysed for island cells apical dendrites is shown (A), basal dendrites (C) and ocean cells (E). Mann–Whitney U-test (***P<0.001, **P<0.01).

Fig. 6.

Reelin rescues dendritic hypertrophy in Reln^cKO island cells. (A) OTCs were grouped as follows: wt+wt, Reln^cKO+Reln^cKO or wt+Reln^cKO. (B) Box and whisker plot shows the mean values of apical dendritic length of island cells. (C) Box and whisker plot shows the mean values of apical dendritic segments of island cells. (D) Box and whisker plot shows the mean values of basal dendritic length of island cells. (E) Box and whisker plot shows the mean values of basal dendritic segments of island cells. The number of cells reconstructed per group is given below the box plot in B. The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample (from three independent OTC preparations). Mann–Whitney U-test (***P<0.001).

Fig. 7.

Ca²⁺ recording of island cells and ocean cells in Reln^cKO and wild type. To assess neuronal activity in ocean cells and island cells, OTCs were transfected at 5 DIV with a GCaMP6s construct to visualize the morphology of the cell type, and record Ca²⁺ activity at 10 DIV. (A,B) Confocal images of a GCaMP6s transfected island cell from wild-type OTC at 10 DIV, during Ca²⁺ imaging at resting fluorescence (A; F⁰) and at maximal amplitude peak (B). (C,D) Typical examples of a Ca²⁺ transient recorded from a wild-type island cell (C) and from a Reln^cKO island cell (D) showing resting fluorescence (F⁰) and the maximum amplitude peak. (E,F) Box and whisker plots show no change in maximal Ca²⁺ amplitude signals between control wild-type and Reln^cKO island cells (E), but a significant increase in Ca²⁺ frequency (F). (G,H) Confocal images of a GCaMP6s transfected ocean cell from a wild-type OTC at 10 DIV, during Ca²⁺ imaging at resting fluorescence (G; F⁰) and at maximal amplitude peak (H). (I,J) Typical examples of Ca²⁺ transients recorded from a wild-type ocean cell (I) and a Reln^cKO ocean cell (J) showing resting fluorescence (F⁰) and the maximum amplitude peak. (K,L) The box plots show no change in maximal Ca²⁺ amplitude signals between control wild-type and Reln^cKO island cells (K) but a significant increase in Ca²⁺ frequency (L). One-way-ANOVA and a Holm-Sidak multiple comparison test (***P<0.001). The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. The number of recorded cells is indicated above the box plot (E, island cells; K, ocean cells). Data were obtained from three independent OTC preparations. Scale bars: 20 μm.

Fig. 8.

Effect of GABA_BR blockade by CGP on island cell dendritic growth. OTCs were transfected with an EGFP-construct at 5 DIV, and at 10 DIV they were fixed and stained for morphometrical quantification. (A) Box and whisker plot shows that CGP increased the mean dendritic length of wild-type but not of Reln^cKO island cells. (B) Box and whisker plot shows that CGP increased the number of apical dendritic segments in wild-type but not in Reln^cKO island cells. (C) Box and whisker plot shows that CGP increased the basal dendritic length of wild-type but not of Reln^cKO island cells. (D) Box and whisker plot shows that CGP increased the number of basal dendritic segments in wild-type but not in Reln^cKO island cells. The number of cells reconstructed per group is given above the box plot in A. One-way-ANOVA on Ranks followed by Tukey's multiple comparison test (***P<0.001). The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. Data were obtained from two independent OTC preparations. (E-H) Example images at 40× magnification of a wild-type island cell (E), a wild-type island cell treated with CGP (F), a Reln^cKO island cell (G) and a Reln^cKO island cell treated with CGP (H). The corresponding traces are shown below the images. Scale bars: 50 μm.

Fig. 9.

Effect of reelin on MECII interneuron dendritic growth in vivo. (A) Box and whisker plot shows that the mean dendritic length of GAD67-Reln^cKO interneurons is significantly higher in comparison with the GAD67-wt interneurons. (B) Box and whisker plot shows that the mean number of dendritic segments is significantly higher in the GAD67-Reln^cKO interneurons in comparison with the GAD67-wt interneurons group. (C) Box and whisker plot shows that the number of primary dendrites remained unaltered. The number of cells reconstructed per group is given above the box plot in A. The horizontal lines in the box plots represent the median; whiskers indicate variability outside the upper and lower quartiles; the box indicates the middle half of the sample. The data were obtained from two independent OTC preparations. Mann–Whitney U-test (***P<0.001). (D,E) Representative Golgi staining images of GAD67-wt interneurons (40× magnification) (D) and GAD67-Reln^cKO interneurons (40× magnification) (E). The respective traces are provided below the images. Scale bars: 50 μm.