The apical complex couples cell fate and cell survival to cerebral cortical development

Abstract

Cortical development depends upon tightly controlled cell fate and cell survival decisions that generate a functional neuronal population, but the coordination of these two processes is poorly understood. Here we show that conditional removal of a key apical complex protein, Pals1, causes premature withdrawal from the cell cycle, inducing excessive generation of early-born postmitotic neurons followed by surprisingly massive and rapid cell death, leading to the abrogation of virtually the entire cortical structure. Pals1 loss shows exquisite dosage sensitivity, so that heterozygote mutants show an intermediate phenotype on cell fate and cell death. Loss of Pals1 blocks essential cell survival signals, including the mammalian target of rapamycin (mTOR) pathway, while mTORC1 activation partially rescues Pals1 deficiency. These data highlight unexpected roles of the apical complex protein Pals1 in cell survival through interactions with mTOR signaling.

Figure 1. Cerebral cortex-specific ablation of Pals1 causes severe cortical malformation

(A) Complete loss of medial cortex is seen after cortical-specific deletion of Pals1, using Emx1Cre (Figure S3). The heterozygote cortex is smaller, though lateral cortex is relatively spared. (B) Representative rostral, medial, and caudal 2T weighted MRI sections of wild type (WT) and Pals1 CKO adult mice. Arrows point to intact 3^rd and 4^th ventricles. Arrowheads point to a fluid-filled cystic space in Pals1 CKO mice that would normally be occupied by the cerebral cortex in WT mice. (C) X-gal labeling at E16 show that areas with strong Cre expression in normal brain are absent in Pals1 CKO; Rosa26 brain. (D) Histology of the whole brain at P4 shows a reduced brain size in both Pals1 Het and Pals1 CKO animals. The medial structures are completely absent in Pals1 CKO animals. The size bar shows 1mm. See also Figure S1.

Figure 2. Pals1 loss reveals dorsal cortex-dependent and -independent behaviors

(A) Body and brain weight of littermates at 4 weeks of age. The brain mass of Pals1 CKO and Pals1 Het mice are smaller than WT (t-test, p<0.001; n=3–5). (B) Mice tested on the accelerating rotarod showed no significant differences between Pals1 CKO, Pals1 Het, or control mice. (C) Similar footprint patterns of wild type (top), Pals1 Het (middle), and Pals1 CKO (bottom) mice. (D) Pals1 CKO mice had reduced latency to fall in the wire hang test versus controls and Pals1 Het mice (t-test: p<0.0005 and p<0.05, respectively, n=10). (E) Representative traces of swimming patterns of WT (left), Pals1 Het (middle), and Pals1 CKO (right) mice in a Morris water maze where mice swam towards a visible platform (P) located in quadrant 3. (F) Quantification of time spent per quadrant during swimming trial in (E). Pals1 CKO mice spent the least time in the quadrant with the platform versus wild type and Pals1 Het mice (t-test: p<0.005; n=10). (G) Representative traces of exploratory behavior of WT (left), Pals1 Het (middle), and Pals1 CKO (right) mice during a 10 minute open field trial. (H) Quantification of time spent in center quadrant during open field test (G). Pals1 Het and Pals1 CKO mice spent less time exploring the open field versus WT controls (t-test: p<0.05 and p<0.0005, respectively; n=10).

Figure 3. Pals1 deficiency depletes the neural progenitor pool by premature cell cycle exit

(A) The Pals1 CKO cortex is smaller than littermate controls at E12. (B) At E14, both Pals1 Het and Pals1 CKO cortices are extremely small compared to controls. (C) A schematic of interkinetic movement of neural progenitors. (D) S-Phase cells transiently labeled by BrdU, and M-phase cells labeled with anti-P-H3 are substantially reduced in the Pals1 CKO brain. Intermediate reduction is seen in Pals1 Het embryo, compared to WT at E12 and E14 (D, E, F, respectively)(t-test; BrdU at E12, p<0.001; BrdU at E14, p<0.005; P-H3 at E12, p<0.01; P-H3 at E14, p<0.01; n=3 and n=4). (G, H) Neural progenitors exiting the cell cycle after 24 hours are double-labeled by BrdU and Ki67. Cells exiting the cell cycle are positive for BrdU, but negative for Ki67, and are increased in the Pals1 CKO versus controls. See also Figure S3.

Figure 4. Pals1 is necessary and sufficient for self-renewal of neural progenitors

(A-E) Progenitors in mitosis (P-H3-positive staining cells) are decreased in the Pals1 silenced (Pals1 shRNA) cortical region, and dispersed from the apical margin. (A) Non-target shRNA vectors were used as controls. (B-E) The proportion of P-H3-positive cells in the Pals1 shRNA treated cells (GFP-positive) reduced to 17% as compared to control 21% (t-test; p<0.005; n=5). (G-J) Reduction and scattering of S-phase proliferative cells in Pals1 shRNA tissue was analyzed 1-hr after a BrdU pulse. Arrows (panels C, H) separate electroporated regions from adjacent nonelectroporated regions. (J) Quantification of BrdU-positive cells among shRNA treated cells reveals fewer S-phase proliferating cells upon Pals1 shRNA versus control (t-test; p<0.0001; n=5). (L-N) Pals1 downregulation increases Tuj1-positive staining neurons compared to control, indicating a premature cell-cycle exit of cortical progenitors. GFP and Tuj1 immunoreactivity largely overlap, indicating a substantial cell-autonomous impact of Pals1 silencing in promoting neuronal differentiation. (O, S) E13.5 embryos were electroporated with GFP (O, S) or pCAGGS-Pals1 (P, Q, R, T), and cortical tissue was analyzed at P3. (O, S) GFP electroporated neurons are clustered in the cortical plate. (P, T) Conversely, Pals1 overexpressing cells are scattered more widely in the cortical and subcortical regions, suggesting abnormalities in exiting the VZ, and potentially in neuronal migration. A small fraction of cells were still localized in the SVZ, suggesting failure to leave the proliferative regions (arrows). (Q) High magnification images show Pals1 overexpression. (R) Example of Pals1 overexpressing cells confined to the SVZ. These cells continue to express Ki67, indicating their persistent proliferative state. (S, T) Higher magnification images of the boxed areas in O and P respectively. (U, V) Two days after pCAGGS-Pals1 electroporation, Pals1 is highly expressed in the cytoplasm and in the apical junction in overexpressing cells. (W, X) High magnification pictures of boxed areas of V. See also Figure S4.

Figure 5. Pals1 deficiency causes massive and rapid cell death of abnormally generated postmitotic neurons

(A, B) More cells in the Pals1 CKO undergo apoptotic cell death than in Pals1 Het or WT littermates, as seen by TUNEL staining at E12. Dying cells are observed by E10 and peak at E11-E12 (ANOVA; WT vs. Het p<0.05; WT vs. CKO p<0.0005; Het vs. CKO p<0.05; n=3 and n=5). (C) The cells undergoing apoptotic cell death are primarily neuronal, as shown by double-labeling with progenitor markers, (apical progenitors: Pax6; basal progenitors: Tbr2) and the nuclear neuronal markers Hu, Tuj1 together with TUNEL staining. Hu and Tuj1 extensively overlap with TUNEL-stained dying cells while progenitor markers rarely overlap with TUNEL-positive staining cells. The size markers represent 75μm. (D) The cells in M phase or G2 phase, which are labeled by a 4.5 hour BrdU pulse, rarely overlap with cleaved caspase 3 (CC3)-positive staining cells. (E) Cells labeled by a 7.5 hour BrdU pulse are mainly in G1 or exiting the cell cycle. In the Pals1 CKO brain, most BrdU-positive cells double label with CC3 (E, arrow). (F) Pals1 knockdown cells acutely undergo apoptotic cell death since over 20% of Pals1 shRNA-treated cells are TUNEL-positive as early as 15 hrs after electroporation. See also Figure S5.

Figure 6. Pals1 loss disrupts apical complex proteins and adherens junctions

(A) The expression of Pals1 in the neuroepithelium at E12 is diminished in a medial to lateral gradient in Pals1 CKO mutant mice versus WT. (B) At E12, the apical complex proteins aPKCλ and Crb2 were reduced in an analogous manner to the Pals1 loss in the Pals1 CKO. (C) Defects of Crb2 protein localization are observed as early as 9 hrs after Pals1 shRNA electroporation. The Crb2 protein diffuses in the cytoplasm in the shRNA treated cells, whereas the cells treated with control shRNA show apical localization. (D) Adherens junctions are relatively intact at E12 in Pals1 CKO mutant mice. At E13, adherens junctions are either displaced basally, or absent in Pals1 CKO mice. At E14, adherens junctions are completely absent in medial cortex of mutant mice. (E) Top panels: Scanning EM reveals fewer primary cilia in the Pals1 CKO neuroepithelium compared to wild type (arrowheads). Middle panels: Pericentrin staining was reduced in Pals1 CKO mice, indicating a reduction in basal bodies (the base of cilia). Lower panels: Pals1 CKO mice have shorter cilia than WT controls. In the Pals1 CKO, the apical membrane shows large and irregularly shaped particles compared to controls, which have smaller and relatively uniform particles at E12 (arrows). See also Figure S2.

Figure 7. mTOR activity is attenuated in Pals1 CKO neurons

(A-C) Abnormally generated neurons marked by Tuj1 (A) and p27 (B, C) in the Pals1 CKO mice show reduced or absent pS6 expression, indicating attenuation of mTOR activity compared to WT littermates at E11. In control animals, 72% of p27-positive cells express pS6 while in the Pals1 Heterozygote and Pals1 CKO mice, 48% and 30% of p27-postive cells overlap with pS6 respectively (t-test, p<0.05; n= 3 for WT n=3 for Het and n=5 for CKO). (D, E) The activation of mTOR by elimination of a negative regulator, Tsc2, genetically restores pS6 staining in postmitotic neurons marked by Tuj1 in the double mutants (Pals1 CKO; Tsc2 CKO; Emx1Cre) at E12 compared to single mutant (Pals1 CKO). (E) At P21, the medial cortex of the Pals1; Tsc2 double mutant is partially restored compared to the Pals1 CKO single mutant in which the medial cortex is almost absent.

Figure 8. GSK-3β activity has a synergistic effect on neuronal cell death together with Pals1 down-regulation

(A-F) Neuronal cell death induced by Pals1 down regulation (A-C) is efficiently rescued by simultaneous downregulation of GSK-3β activity obtained by expression of a kinase-dead GSK-3β expression vector (D-F). (G-L) GSK-3β overexpression elicits only a minor effect on cell death in control cortical tissue, while when combined with Pals1 down-regulation exerts a dramatic increase in neuronal cell death (J-L). DNA constructs are electroporated in E13.5 mouse cortices and results analyzed 48hr later. C, F, I, L show merged images of CC3-positive (red, apoptotic cells) and GFP electroporated cells (green). (M) Quantification of the electroporated (GFP+, green) and apoptotic (CC3+, red) cells per field for each different condition analyzed in (A-L)(ANOVA; p<0.0001; n=3). (N) Percentage of CC3-positive apoptotic figures within the total electroporated GFP-positive each different condition analyzed (ANOVA; p<0.0001; n=3).