XB-ART-53944Development June 15, 2016; 143 (12): 2121-34.
Cadherin 2/4 signaling via PTP1B and catenins is crucial for nucleokinesis during radial neuronal migration in the neocortex.
Cadherins are crucial for the radial migration of excitatory projection neurons into the developing neocortical wall. However, the specific cadherins and the signaling pathways that regulate radial migration are not well understood. Here, we show that cadherin 2 (CDH2) and CDH4 cooperate to regulate radial migration in mouse brain via the protein tyrosine phosphatase 1B (PTP1B) and α- and β-catenins. Surprisingly, perturbation of cadherin-mediated signaling does not affect the formation and extension of leading processes of migrating neocortical neurons. Instead, movement of the cell body and nucleus (nucleokinesis) is disrupted. This defect is partially rescued by overexpression of LIS1, a microtubule-associated protein that has previously been shown to regulate nucleokinesis. Taken together, our findings indicate that cadherin-mediated signaling to the cytoskeleton is crucial for nucleokinesis of neocortical projection neurons during their radial migration.
PubMed ID: 27151949
PMC ID: PMC4920171
Species referenced: Xenopus
Genes referenced: cdh2 cdh4 cdh6 chdh ctnnb1 dcx epha8 npat pafah1b1 pax6 slc7a5 ucp1
Article Images: [+] show captions
|Fig. 1. Expression patterns of classical cadherins in the developing mouse lateral neocortex. (A) In situ hybridization for β-catenin, Cdh2, Cdh4, Cdh6, Cdh11 and Cdh13 was carried out on coronal sections of E14.5 and E16.5 brains. Examples are shown from the developing somatosensory cortex. (B-D) Immunohistochemistry for CDH2 (green) and CDH4 (green; red in B, bottom panels) on coronal sections of the developing neocortex. The sections were co-stained with PAX6 (red) and TUJ1 (blue) antibodies (C,D). (B) CDH2 and CDH4 are expressed along the width of the cortical primordium. Both proteins are expressed in the ventricular zone, although CDH4 at lower levels than CDH2. (C) Both PAX6+ RGCs and TUJ1+ neurons express CDH2. (D) CDH4 is expressed by PAX6+ and TUJ1+ cells. Small panels in C and D are taken from different images that come from the same experiments. Arrows in C and D point to cells co-expressing the two markers. SP, subplate. Scale bars: 10 μm (C,D, small panels); 50 μm (all other panels).|
|Fig. 2. CDH2 and CDH4 are required for radial migration in mouse cortex. (A) Illustration of the strategy to inactivate Cdh2 and Cdh4 in migrating neurons. Embryos from floxed animals were electroporated in utero at E14.5 with DCX-Cre-i-EGFP or DCX-i-GFP. Position of the electroporated cells was analyzed at E18.5 in the developing somatosensory cortex. (B) Representative images of coronal sections of embryos electroporated as described in A. Electroporated neurons are shown in green and nuclei in blue (TO-PRO). (C) Quantification of the percentage of electroporated neurons that enter the boxed areas in B, representing the upper 25% of the CP. Four animals (Cdh2fl/fl), five animals (Cdh4fl/fl) and six animals (Cdh2/4fl/fl) from three separate experiments were analyzed for each condition. The data represent mean±s.e.m. n.s., non significant, *P<0.01, **P<0.0001, ***P<1×10−13 by Bonferroni post-hoc analysis after one-way ANOVA. MZ, marginal zone; SP, subplate. Scale bar: 100 μm.|
|Fig. 5. Inhibition of nuclear and centrosomal movement. (A) Diagram of experimental strategy to analyze neuronal morphology and nucleokinesis. DN-CDH and adhesion-deficient Cdh2 and Cdh4 constructs were co-electroporated with a fluorescence-tagged Golgi marker at E14.5. The morphology of electroporated neurons and the position of the Golgi apparatus were analyzed at E18.5, as shown in C. (B) Time-lapse imaging strategy to assess nuclear and centrosomal movement. Control or DN-CDH plasmids were co-electroporated with a fluorescence-tagged centrosomal marker at E14.5. Brains were processed for slice culture and live imaging at E17.5, as shown in D. (C) Neurons expressing DN-CDH or any of the adhesion-deficient cadherins extend long leading processes towards the CP (green). The Golgi apparatus (red) shows a polar localization in front of the nucleus in all experimental conditions. Nuclei are marked with asterisks, arrows point to the Golgi apparatus. Process length is quantified in E. Scale bar: 50 µm in C for C,D. (D) Time-lapse imaging of migrating neurons electroporated with a control plasmid (green, top panels) or DN-CDH (green, bottom panels). In control conditions, the centrosome (red) moves ahead of the nucleus into the swelling followed by nuclear movement. In DN-CDH-expressing cells, the centrosome stays close to the nucleus and neither organelle moves forward during the imaging period. The arrows indicate centrosomal position; nuclear position is marked by an asterisk. Data are representative of two independent experiments, each with two brains electroporated with control plasmid, two with DN-CDH and two with CDH2-D134A. Migration speed is quantified in F. (E) Quantification (mean±s.e.m.) of the process length from neurons expressing control plasmid, DN-CDH or adhesion-deficient cadherins, as in C. *P<0.001 by Bonferroni post-hoc analysis after one-way ANOVA. The number of processes measured for each condition was: 58 (EGFP, three brains), 40 (DN-CDH, three brains), 28 (CDH2-W2A, three brains), 42 (CDH4-W2A, four brains), 53 (CDH2-D134A, four brains) and 30 (CDH4-D134A, three brains). (F) Quantification of migration speed in neurons electroporated with control plasmid or DN-CDH. *P<1×10−6 by Student's t-test; 31 control cells (five brains) and 41 DN-CDH electroporated cells (six brains) from three different experiments were analyzed.|
|Fig. 6. Formation of lateral protrusions containing F-actin. (A) Actin-GFP and mCherry were co-electroporated at E14.5 with or without DCX-Cdh2-D134A. Brains were processed for slice culture and live imaging at E16.5. (B) Actin (green) in control electroporated neurons is confined in a single leading process that shows no branching. The length of the leading process remains constant as the nucleus (red) moves forward. Bottom panels show individual channels for EGFP and m-Cherry for the 1- and 2-h images. (C) In neurons expressing CDH2-D134A, actin dynamics appear increased, with many transient actin-rich side branches (green) being created along the leading process. Note that the nucleus does not translocate and the neurons do not move forward, resulting in longer and thinner leading processes. Bottom panels show individual channels for EGFP and m-Cherry for the 1- and 2-h images. Experiments were performed five times, for a total of 17 control and 21 DN-CDH electroporated brains.|
|Fig. 7. Cadherin adhesion provides traction for radial migration. (A) Illustration of the strategy to test whether cadherin is necessary for sustaining the traction forces that are required for nucleokinesis. Control or DN-CDH-expressing plasmids were electroporated at E14.5. Brains were processed 48-72 h later for slice culture and live imaging. Growth medium was replaced with medium containing 10 nM calyculin A about 30 min into the imaging session, and imaging was continued for 50 min. (B) Neurons electroporated with DCX-i-EGFP either do not respond to calyculin A, or tend to translocate their nucleus into the leading process (upper panels). The majority of neurons expressing DN-CDH retract their leading process in response to calyculin A (bottom panels). The arrows point to the tip of the leading process; nuclear position is indicated by an asterisk. This experiment was repeated five times. t=0 refers to the time when calyculin A was added. (C) Quantification of the response to calyculin A treatment. *P<0.01 for all three possible outcomes by Student's t-test (P<0.001 for no response; P<1×10−7 for retraction and P<0.01 for translocation). A total of 119 control and 195 DN-CDH electroporated neurons from three (control) and four (DN-CDH) brains were analyzed.|
|Fig. 8. Functional interactions between CDH2 and LIS1. (A) Stack of confocal images of neurons expressing CDH2 (green), LIS1 (red) and BFP. EGFP-tagged CDH2 was co-electroporated with HA-tagged LIS1 and BFP at E14.5. CDH2 was visualized at E17.5 by EGFP fluorescence and LIS1 by staining with HA antibodies. Panels on the right are single confocal sections of the areas marked as A′ and A″. CDH2-EGFP and HA-LIS1 colocalize in the leading process (A′) and cell soma (A″). Arrows point to the leading processes of several neurons. (B) LIS1 localization is altered in neurons expressing DN-CDH. HA-tagged LIS1 was co-electroporated with either a control plasmid or DN-CDH at E14.5 and brains were analyzed 3 days later. DN-CDH-expressing neurons show increased HA-LIS1 staining in the leading process. B′ and B″ panels are single and combined channel images of the boxed areas in the main image. Dotted lines indicate leading processes. (C) Quantification of the relative HA (red) average fluorescence intensity in the leading processes versus soma of HA-LIS1 and control or DN-CDH co-electroporated neurons. *P<0.001 by Student's t-test. Fluorescence intensity was measured in 53 (control, four different brains) and 59 (DN-CDH, four different brains) neurons. (D) Co-expression of DCX-Lis1 partially rescues the migration defect caused by expression of DN-CDH. DN-CDH was co-electroporated either with a control plasmid or with DCX-Lis1-i-EGFP at E14.5 and the position of the electroporated cells was assessed at E18.5. Electroporated neurons are shown in black. (E) Quantification (mean±s.e.m.) of the data in D. *P<0.01 by Student's t-test. Neurons were counted in three brain slices from each of four animals obtained from three independent electroporation experiments. (F) High magnifications of the neurons expressing DN-CDH+control plasmid or DN-CDH+LIS1. Note the difference in the length of the leading processes between the two conditions. (G) Quantification of process length of the neurons in F. *P<0.01 by Student's t-test. Process length was measured in 44 (DN-CDH+EGFP, four different brains) and 48 (DN-CDH+DCX-Lis1, five different brains) neurons. MZ, marginal zone. Scale bars: 20 μm (A,B); 10 μm (A′,A″); 100 μm (D); 50 μm (F).|
References [+] :
Abe, EPLIN mediates linkage of the cadherin catenin complex to F-actin and stabilizes the circumferential actin belt. 2008, Pubmed