Cep120 and TACCs control interkinetic nuclear migration and the neural progenitor pool

Abstract

Centrosome- and microtubule-associated proteins have been shown to be important for maintaining the neural progenitor pool during neocortical development by regulating the mitotic spindle. It remains unclear whether these proteins may control neurogenesis by regulating other microtubule-dependent processes such as nuclear migration. Here, we identify Cep120, a centrosomal protein preferentially expressed in neural progenitors during neocortical development. We demonstrate that silencing Cep120 in the developing neocortex impairs both interkinetic nuclear migration (INM), a characteristic pattern of nuclear movement in neural progenitors, and neural progenitor self-renewal. Furthermore, we show that Cep120 interacts with transforming acidic coiled-coil proteins (TACCs) and that silencing TACCs also causes defects in INM and neural progenitor self-renewal. Our data suggest a critical role for Cep120 and TACCs in both INM and neurogenesis. We propose that sustaining INM may be a mechanism by which microtubule-regulating proteins maintain the neural progenitor pool during neocortical development.

Figure 1

Expression and localization of Cep120. (A–C) Lysates prepared in RIPA buffer were resolved by SDS-PAGE and subjected to immunoblot analysis. (A) The Cep120 antiserum recognizes exogenous Cep120 expressed in COS7 cells and endogenous Cep120 in brain lysates. The arrow indicates the Cep120 protein. (B) Cep120 protein levels in E17.5 mouse tissues. (C) Cep120 protein levels in the mouse brain at different developmental stages. Each lane was loaded with 20 μg total proteins in (B) and (C). (D and E) Cep120 is a centrosomal protein. (D) EGFP-Cep120 colocalizes with pericentrin in cultured neocortical cells. (E) Affinity-purified Cep120 antibody labels the centrosome in cultured neocortical cells. (F and G) Expression of Cep120 in the mouse neocortex at E12.5. The Cep120 antibody prominently labels centrosomes (indicated by pericentrin co-staining) along the ventricular surface (bottom). The boxed area in (F) is shown in (G) at higher magnification to reveal the puncta labeled by both Cep120 and pericentrin antibodies along the ventricular surface. Scale bars: 10 μm in (D) and (E), 20 μm in (F) and (G).

Figure 2

Cep120 is important for INM. (A–C) Mouse embryos were electroporated at E11.5 and sacrificed at E14.5. (A) A transfected cell with an apical process contacting the ventricular surface (outlined by dashed lines) expresses Nestin, a neural progenitor marker. (B and C) The nuclei of Cep120 i2968-expressing neural progenitors are less elongated along the apical-basal axis compared to the control or rescue (Cep120 i2968 + Cep120 i2968 resistant Cep120 plasmid). (B) Representative neural progenitors. The nuclei and the ventricular surface are outlined by red and white dashed lines, respectively. (C) Quantification of the ratio of nuclear length to width. ** P<0.01, one-way ANOVA. (D and E) Ectopic PH3-positive nuclei in Cep120 i2968-expressing neural progenitors. Mouse embryos were electroporated at E11.5 and sacrificed at E13.5. (D) Representative transfected PH3-positive cells (indicated by arrows). The ventricular surface is indicated by white dashed lines. (E) Quantification of the distance between PH3-positive nuclei and the ventricular surface. * P<0.05, one-way ANOVA. (F–H) Silencing Cep120 impairs INM in brain slice preparations. Mouse embryos were electroporated at E11.5 and sacrificed at E13.5. Acute brain slices of 150-μm were then prepared and imaged for 4 h. (F) A representative imaging area at the VZ (boxed region, transfected cells are green). (G) Examples of transfected cells in which the nuclei (outlined by red dashed lines) moved toward the ventricular surface (bottom). The control and the rescue cell divided before the end of the imaging period. (H) Quantification of the total distance the nuclei traveled during the 4-h imaging period. ** P<0.01, *** P<0.001, one-way ANOVA. Scale bars: 10 μm in (A), (B) and (D), 20 μm in (G).

Figure 3

Cep120 is required for maintaining the neural progenitor pool. (A–E) Progenitors expressing Cep120 siRNA prematurely exit the cell cycle. Mouse embryos were electroporated at E11.5, exposed to BrdU at E13.5, and sacrificed at E14.5. The proliferation of progenitors in the neocortex was analyzed by immunostaining using antibodies against PH3, BrdU, and Ki67. (A) Representative images showing PH3-positive in the VZ of electroporated embryos. White arrows indicate transfected PH3-positive cells in the control group. (B) Quantification of the mitotic index, i.e. the percentage of transfected cells in the VZ that are PH3-positive. (C) Representative images showing transfected cells in neocortices co-stained with BrdU and Ki67 antibodies. White arrows indicate transfected cells that are positive for both BrdU and Ki67. White arrowheads indicate transfected cells that are positive for BrdU but negative for Ki67. (D) Quantification of the BrdU labeling index, i.e. the percentage of transfected cells that are BrdU-positive. (E) Quantification of the cell cycle exit index, i.e. the percentage of transfected BrdU-positive cells that do not express Ki67. Compared to the control, Cep120 i2968 and Cep120 i1265 groups exhibit a significant decrease in the mitotic index and BrdU labeling index and a significant increase in the cell cycle exit index. ** P<0.01, *** P<0.001, t-test. (F–H) Cep120 siRNA causes increased neuronal production. Mouse embryos were electroporated at E11.5 and sacrificed at E14.5. (F) Representative images showing the distribution of transfected cells across neocortical layers. (G) Quantification of the distribution of transfected cells. A higher percentage of transfected cells in Cep120 siRNA groups are located in the IZ or CP compared to the control or rescue group. * P<0.05, ** P<0.01, one-way ANOVA. (H) Quantification of the percentage of transfected cells that express Tuj1. The percentage is significantly higher in Cep120 siRNA groups compared to the control or rescue group. *** P<0.001, one-way ANOVA. CP, cortical plate; IZ, intermediate zone; SVZ/VZ, subventricular zone/ventricular zone. Scale bars: 20 μm in (A) and (C), 50 μm in (F).

Figure 4

Cep120 interacts with TACCs. (A) β-Gal assay was performed using yeast cells expressing the coiled-coil domain of Cep120 and/or the C-terminus of TACCs. Positive results (blue color) were obtained only when Cep120 and TACCs were co-expressed. (B) Lysates from COS7 cells expressing exogenous Cep120 and either HA-TACC2B or HA-TACC3 were subjected to anti-HA immunoprecipitation. A fraction of Cep120 was co-immunoprecipitated. The asterisk indicates that a significant amount of HA-TACC3 was pulled down nonspecifically by sepharose beads. (C and D) Cep120 and TACCs operate together to increase the size of the microtubule aster emanating from the centrosome in transfected COS7 cells. (C) Examples of transfected cells immunostained with an α-tubulin antibody. (D) Measurement of the size of the microtubule aster. * P<0.05, *** P<0.001, one-way ANOVA. Scale bars: 10 μm.

Figure 5

TACC siRNA impairs INM and neural progenitor self-renewal. (A–C) TACC siRNA impairs INM. (A) The nuclei of TACC siRNA-expressing neural progenitors are less elongated along the apical-basal axis compared to the control. Mouse embryos were electroporated at E11.5 and sacrificed at E14.5, and the ratio of nuclear length to width in transfected neural progenitors was quantified. ** P<0.01, t-test. (B) Ectopic PH3-positive nuclei in TACC siRNA-expressing neural progenitors. Mouse embryos were electroporated at E11.5 and sacrificed at E13.5, and the distance between PH3-positive nuclei and the ventricular surface was quantified. * P<0.05, t-test. (C) TACC siRNA impairs INM in brain slice preparations. Mouse embryos were electroporated at E11.5 and acute brain slices of 150-μm were prepared at E13.5 for imaging. The total distance the nuclei traveled during the 4-h imaging period was quantified. * P<0.05, t-test. (D–H) Progenitors expressing TACC siRNA prematurely exit the cell cycle. Mouse embryos were electroporated at E11.5 and exposed to BrdU at E13.5, and PH3-, BrdU-, and Ki67-immunostaining was performed at E14.5. (D) Representative images showing PH3-positive cells in the VZ of electroporated embryos. White arrows indicate transfected PH3-positive cells in the control group. White dashed lines outline the ventricular surface. (E) Quantification of the mitotic index. (F) Representative images showing transfected cells in neocortices co-stained with BrdU and Ki67 antibodies. White arrows indicate transfected cells labeled by both BrdU and Ki67. White arrowheads indicate transfected cells labeled by BrdU but not Ki67. (G) Quantification of the BrdU labeling index. (H) Quantification of the cell cycle exit index. TACC siRNA caused a significant decrease in the mitotic index and BrdU labeling index, and a significant increase in the cell cycle exit index. * P<0.05, ** P<0.01, t-test. (I–K) TACC siRNA causes increased neuronal production. Mouse embryos were electroporated at E11.5 and sacrificed at E14.5. (I) Representative images showing transfected cells in the neocortex immunostained with a Tuj1 antibody. (J) Quantification of the distribution of transfected cells. Compared to the control, TACC siRNA increased the percentage of transfected cells in the CP or IZ and decreased the percentage of transfected cells in the SVZ or VZ. *** P<0.001, t-test. (K) Quantification of the percentage of transfected cells that are Tuj1-positive. The percentage is significantly higher in the TACC siRNA group. *** P<0.001, t-test. Scale bars: 20 μm in (D) and (F), 50 μm in (I).

Figure 6

An important role for Cep120 and TACCs in the integrity of microtubules coupling the centrosome and the nucleus and for Cep120 in the centrosomal localization of TACC3. (A–C) Cep120 and TACCs regulate microtubules coupling the centrosome and the nucleus. Neocortical cells lining the lateral ventricle of mouse embryos were co-transfected with Venus and siRNA plasmids via electroporation and dissociated for α-tubulin-immunostaining analysis (see Experimental Procedures). (A) A microtubule “fork”-like structure (indicated by α-tubulin immunostaining) associated with both the centrosome (indicated by Cep120 immunostaining) and the nucleus (indicated by Hoechst labeling) in a non-transfected cell. Several Z-series images were merged to show the entire “fork”-like structure (outlined by white dashed lines). (B) Representative images showing the microtubule “fork”-like structure (outlined by white dashed lines) in transfected cells. Several Z-series images were merged to show the entire “fork”-like structure. The nuclei are outlined by red dashed lines. The microtubule “fork”-like structure is smaller in cells expressing Cep120 i2968 or TACC siRNA compared to the control. (C) Quantification of the area occupied by the largest microtubule bundle of the “fork”-like structure. ** P<0.01, *** P<0.001, one-way ANOVA. (D and E) Cep120 i2968 and TACC siRNA increase nucleus-centrosome distance in neural progenitors in vivo. Mouse embryos were electroporated with Venus and siRNA plasmids at E11.5 and cryosections were prepared at E14.5. Transfected cells were identified by immunostaining using a chicken anti-GFP antibody. Centrosomes were visualized by Cdk5rap2 immunostaining. (D) Examples of transfected neural progenitors in which centrosomes (indicated by white arrows) were visualized by Cdk5rap2 labeling. The ventricular surface is outlined by white dashed lines. (E) Quantification showing that the nucleus-centrosome distance is significantly increased in neural progenitors expressing Cep120 i2968 or TACC siRNA. * P<0.05, ** P<0.01, one-way ANOVA. (F and G) Cep120 regulates centrosomal localization of EGFP-TACC3 in neural progenitors in vivo. Mouse embryos were co-electroporated with Cep120 i2968, EGFP-TACC3, and RFP-CETN2 at E12.5. The localization of EGFP-TACC3 in transfected neural progenitors was examined about 30h after electroporation by anti-GFP immunostaining. (F) Representative images showing the enrichment of EGFP-TACC3 at and around the centrosome in the control cell (upper panels) but not in the Cep120 i2968-expressing cell (lower panels). Only the apical processes of transfected progenitors were shown. RFP-CETN2 labels the centrosome (indicated by white arrows). White and red dashed lines outline the ventricular surface and apical processes of transfected progenitors, respectively. (G) Quantification of the effect of Cep120 i2968 on centrosomal localization of EGFP-TACC3. The relative intensity of EGFP-TACC3 fluorescence at and around the centrosome (i.e. the ratio of the mean intensity of EGFP-TACC3 fluorescence at and around the centrosome to that in a nearby segment of the apical process) was significantly lower in progenitors expressing Cep120 i2968 compared to the control. ** P<0.01, t-test. Scale bars: 5 μm in (A), (B), and (F), 10 μm in (D).

Figure 7

A tentative model for INM. The centrosome is localized at the tip of the apical process adjacent to the ventricular surface. Cep120 recruits TACCs to the centrosome and together with TACCs promotes the assembly of long microtubules from the centrosome. The nucleus may be transported toward the centrosome along these microtubules by dynein or be pulled toward the centrosome due to the shortening of these microtubules. Lis1 may sustain the migration of the nucleus by regulating the activity of dynein. Upon depletion of Cep120 or TACCs, microtubules emanating from the centrosome are too short to reach the nucleus when the nucleus is located deep within the ventricular zone. Consequently, the transport or pulling of the nucleus toward the ventricular surface is impaired.