Figure 1.
Targeted disruption of RanBP9 gene by gene trap strategy.
(A), Schematic view of the exon structure and the restriction map of RanBP9 wild-type (WT) gene (top) and the domain organization of the predicted WT protein (bottom). (B) Schematic view of the βgeo cassette trapped between exon 1 and 2 of RanBP9 gene (top) and the Ran-βgeo mutant protein (bottom). (C), Western blot analysis of P1 brain lysates showing RanBP9 protein levels completely absent in Ran−/− and about 50% levels in Ran+/− mice compared to WT mice (top panel). Western blots detection of Ran-βgeo mutant protein by β-gal protein specific antibody (middle panel). PCR amplifications of βgeo gene by primers targeted against β-gal gene (bottom), confirming successful integration of the βgeo cassette within RanBP9 gene. (D), Immunohistochemical detection of the expression of Ran-βgeo mutant protein in the cortex (top) and the hippocampus (bottom).
Figure 2.
Behavioral as well as body and brain weight differences between WT, Ran+/− and Ran−/− mice.
Due to defects in suckling, Ran−/− mice fail to drink milk (arrow), whereas WT mice are filled with white milk. (A), A representative WT (+/+) and littermate Ran−/− (−/−) P1 mice are shown. The histogram on the right shows reduced body weights by 12% in Ran+/− mice and by 31% in Ran−/− mice compared to WT littermates. (B), Representative brain pictures are shown for WT and littermate Ran−/− P1 mice. Histograms on the right shows reduced brain weights by 6% in Ran+/− and by 21% in Ran−/− mice compared to littermate WT mice. Body weights are expressed in grams and brain weights in milligrams. (C) A picture of typical WT and one of very few Ran−/− mice that survived until three weeks of age. One-way analysis of variance (ANOVA) followed by post-hoc Tukey-Kramer multiple comparisons test revealed significant differences. In each group n = 6, ±SEM. **, p<0.01, ***, p<0.001 when WT was compared with Ran+/− or Ran−/− mice. $$$, p<0.001 when Ran+/− mice was compared with Ran−/− mice.
Figure 3.
DAPI-stained brain sections from WT (A) and Ran−/− (B) P1 mice showing gross anatomical differences.
Of particular note is the robustly enlarged lateral ventricular volume (arrow) in Ran−/− brain compared to WT brain which has negligible size of the lateral ventricle at this developmental stage. (C) Immunohistochemical staining of neonatal WT P0 brain with RanBP9-specific antibody shows RanBP9 immunoreactivity in all layers of the cortex as well as hippocampus, whereas in Ran−/− brains such immunoreactivity is completely absent (D). CX, cortex; LV, lateral ventricle; HI, hippocampus; TH, thalamus; IIIV, the third ventricle and AN, amygdaloid nuclei.
Figure 4.
Quantitation of length and total region areas of the DAPI-stained brain sections of P1 mice showed significant differences between WT and Ran−/− mice.
Brain sections from level 1 (L1) to level 7 (L7) corresponding to the sections in the Electronic Prenatal Mouse Brain Atlas (EPMBA) were analyzed. The most affected brain area was the volume of the lateral ventricle which was significantly increased in Ran−/− brains at L2 (46-fold), L3 (10-fold), L4 (19-fold) and L5 (17-fold). The cortical plate showed significant decrease in the length at L2 (31%), L3 (47%), L4 (38%) and L5 (46%). The length of total cortex (marginal zone+cortical plate+intermediate zone) was also significantly reduced in Ran−/− brains at L2 (29%), L3 (46%), L4 (36%) and L5 (44%). Marginal zone was decreased only at L1 (39%) and L3 (34%) and the hippocampus only at L4 (49%), but intermediate zone was not affected in Ran−/− brains. One-way ANOVA followed by post-hoc Bonferroni multiple comparisons test revealed significant differences. In each group n = 6, ±SEM. *, p<0.05, **, p<0.01, ***, p<0.001in Ran−/− brains compared to littermate WT controls.
Table 1.
Levels 1–7 defined based on specific coronal layers (sections) in the Electronic Prenatal Mouse Brain Atlas (EPMBA).
Figure 5.
Confocal images of neuronal nuclear antigen (NeuN)-stained brain sections of WT and Ran−/− P1 mice at 4.5× (A), 20× (B) and 40× (C) as indicated.
Note robustly increased lateral ventricular volume in Ran−/− brains at 4.5× (A), highly decreased thickness of cortical plate at 20× (B) and significantly reduced post-mitotic neurons at 40× (C) in Ran−/− brains compared to littermate WT (+/+) control brains. Scale bars are indicated for each magnification. CX, cortex, LV, lateral ventricle, HI, hippocampus, CP, cortical plate, SP, subplate, MZ, marginal zone, IZ, intermediate zone and VZ, ventricular zone.
Figure 6.
Relative fluorescence intensity of brain sections stained with NeuN and analyzed by Image-Pro Plus 3D Suite software in WT and Ran−/−P1 brains.
A line profile analysis was performed for each area of interest (AOI) which generated a plot of the average intensity values. (A) Relative fluorescence intensity along the length of the cortex. Note the length differences in WT (400 µm) versus Ran−/− (300 µm) brains. (B) Fluorescence intensity normalized to the length of the cortex. The decreased fluorescence intensity is seen mostly in the cortical plate (CP) area and less so in the intermediate zone (IZ) and marginal zone (MZ).
Figure 7.
Confocal images of proliferating cell nuclear antigen (PCNA)-stained brain sections of WT and Ran−/− P1 mice at 40× as indicated.
Note the number of cells positive for PCNA, however they were not different in Ran−/− brains compared to littermate WT (+/+) brains. Scale bars are indicated for magnification. CP, cortical plate, SP, subplate, IZ, intermediate zone.
Table 2.
Quantification of PCNA positive cells in brain sections of P1 mice corresponding to level 5 from wild-type and Ran−/− brains in a defined area at 40× magnification.
Figure 8.
Cell numbers positively stained for NeuN, DCX, PCNA and caspase-3 in the subgranular region of the dentate gyrus in WT and RanBP9−/− (KO) mice.
(A), DAPI-stained brain sections to show the highlighted subgranular zone within the dentate gyrus region of the hippocampus used for cell counts shown in B. (B), Representative brain sections stained with anti-NeuN, anti-DCX, anti-PCNA, anti-capsase-3 and counter stained with DAPI. Cell counting revealed significantly decreased numbers of NeuN positive cells in RanBP9−/− (KO) brains (22%) compared to WT controls. However DCX, PCNA and caspase positive cell numbers were not significantly altered. In each group, n = 3, data presented as mean± SEM. **, p<0.01 by Student’s t-test.
Figure 9.
Cells positively stained for PCNA in the whole dentate gyrus region of the hippocampus in WT and RanBP9−/− (KO) mice.
(A). Representative brain section to show whole of the dentate gyrus stained with PCNA (red) and DAPI (blue) in the WT (+/+) mice and the RanBP9 KO (−/−) mice. (B), Quantitation of PNCA-positive cells showed an average of 100 cells in the WT mice compared to only 59 in the RanBP9−/− mice which was statistically significant. In each group, n = 3, data presented as mean± SEM. **, p<0.01 by Student’s t-test.
Figure 10.
ERK1 protein levels are significantly reduced in Ran−/− brains from P1 mice.
(A), Brain lysates from WT and Ran−/−mice were subjected to SDS-PAGE electrophoresis and ERK1/ERK2 & MEK1 proteins were detected using their specific polyclonal antibodies. Actin blot is shown as loading control and RanBP9 blot shows complete absence of RanBP9 protein in Ran−/− pups. (B) Quantitation of signal intensities by imageJ revealed significantly reduced protein levels for ERK1 (47%). n = 5 (WT) and 4 (Ran−/−),±SEM. *, p<0.05 by t test. (C), Endogenous L1CAM was pulled down by RanBP9 antibody (upper panel) and in a reciprocal coimmunopreciptation experiments, RanBP9 was pulled down by L1CAM antibodies (lower panel) suggesting that both L1CAM and RanBP9 interact with each other. (D), A model to explain mechanism of brain growth defects in Ran−/− brains. In response to cell adhesion signals, RanBP9 scaffolds L1CAM and ERK1 thereby activating downstream effectors such as Rho GTPases. Rho GTPases are known to physically interact and modulate cytoskeletal elements, which in turn alter neuronal differentiation/migration. RanBP9 is also known to bind MET-LFA integrin and p75NTR suggesting that alternative pathways to modulate differentiation/migration are also possible.