QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 22 October 2003
doi: 10.1242/dev.00840

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Garcia-Dominguez, M. Articles by Charnay, P. Search for Related Content PubMed PubMed Citation Articles by Garcia-Dominguez, M. Articles by Charnay, P.

Ebf gene function is required for coupling neuronal differentiation and cell cycle exit

Unité 368 de l'Institut National de la Santé et de la Recherche Médicale, Ecole Normale Supérieure, 46 rue d'Ulm, 75230 Paris Cedex 05, France

View larger version (98K): [in a new window]   Fig. 1. Expression of Ebf1 and Ebf3 in the chick neural tube. In situ hybridisation was performed as indicated (stages and probes). (A,B) Flat mounts of the hindbrain region showing the generalised expression of Ebf1 and Ebf3, with the exclusion of rhombomeres 3 and 5, coincident with neurogenesis at stage HH14. (C-F) Transverse sections through the spinal cord at stages HH17 and HH20 showing high expression of Ebf1 and Ebf3 in the mantle layer. r, rhombomere.

View larger version (144K): [in a new window]   Fig. 2. Induction of Ebf gene expression by Ngn2 and NeuroM. Stage HH10 chick embryos were co-electroporated at the level of the anterior spinal cord with the GFP expression vector and Ngn2 (A,D), NeuroM (B,E) or NSCL1 (C,F) expression plasmids. Embryos were collected 24 hours later and processed for in situ hybridisation with Ebf1 (A-C) and Ebf3 (D-F) probes. The neural tubes were then flat-mounted. Ebf1 and Ebf3 were induced by Ngn2 and NeuroM, but not by NSCL1. This conclusion reflects the behaviour of more than 90% of the embryos, from three independent experiments, each involving at least eight embryos per probe. Electroporation was on the right side.

View larger version (93K): [in a new window]   Fig. 3. Ebf1 misexpression promotes neuronal differentiation. (A-F) Stage HH15 chick embryos were either electroporated with a lacZ expression plasmid, or co-electroporated with lacZ and Ebf1 expression plasmids, incubated for the indicated periods, processed for X-gal staining and transversally sectioned. Each pattern shown reflects the situation in more than 90% of the embryos, from eight independent experiments, each involving six embryos per condition. (G-I) Stage HH15 embryos were electroporated with HA-tagged Ebf1 or R409W mutant Krox20 (K20m-HA, encoding an inactive transcription factor) and then subjected to 2-hour BrdU pulse-labelling immediately before collection at the indicated time following electroporation. Vibratome sections were then analysed by immunofluorescence with antibodies directed against HA (red) and BrdU (green). The dashed line in I indicates the separation between the mantle layer (ML) and the ventricular zone (VZ). (J) Quantification of the data obtained from experiments presented G-I. The bars represent the percentage of electroporated cells (HA-Ebf1- or Krox20m-HA-positive, red or yellow) that are located in the ventricular zone (black bars), the percentage of electroporated cells that are BrdU-positive (yellow, white bars), and the percentage of electroporated cells within the ventricular zone, which are BrdU-negative (red, grey bars). The cell counts correspond to the analysis of six to nine sections from at least three independently processed embryos for each condition. The data represent mean±s.e.m. (K,L) Flat-mounted hindbrains from embryos that were either not electroporated or co-electroporated with Ebf1 and GFP expression vectors at stage HH10, then stained for neurofilaments by immunochemistry 24 hours later. (M-O) Stage HH10 embryos were electroporated with HA-tagged Ebf, sectioned 24 hours later at the level of r6 and analysed by immunofluorescence with antibodies directed against the HA epitope (red) and neurofilaments (green, N), or the HA epitope and Tuj1 (green, O). The arrowheads point to cells co-expressing the two markers. M shows the part of the sections presented in N,O. Each analysis was performed on three independent series of six embryos. Tuj1 and neurofilaments were detected in approximately 100% and 80% of the HA-Ebf1-positive cells, respectively. (P-R) Flat-mounted hindbrains from embryos co-electroporated with Ebf1 and GFP expression vectors at stage HH10, and processed 24 hours later for in situ hybridisation with N-cadherin (N-cad) and R-cadherin (R-cad) probes. In R, double in situ hybridisation was performed (N-cad, purple; R-cad, red). In K,L,P-R the patterns shown were observed in more than 90% of the embryos, from four independent experiments, each involving at least six embryos per condition. (S,T) Stage HH15 embryos were co-electroporated with GFP and chicken Id2 (S), or GFP, Id2 and Ebf1 expression vectors (T). They were then subjected to 2-hour BrdU pulse-labelling immediately before collection, 30 hours after electroporation. Vibratome sections were then analysed by immunofluorescence with antibodies directed against BrdU (green), GFP (red) and Tuj1 (blue). In S and T, 60±5% and 55±3% of transfected cells were BrdU+, respectively, and 98±0.6% and 88±4% were located in the VZ, respectively. The data represent mean±s.e.m. and correspond to the analysis of seven sections from three independently processed embryos for each condition. Electroporation was on the right side. Electroporated constructs are indicated at the top of each panel and immunolabelling or in situ hybridisation probes at the bottom. h, hours.

View larger version (134K): [in a new window]   Fig. 4. Transient activation of proneural genes following Ebf1 misexpression. Stage HH10-HH12 chick embryos were co-electroporated on the right side with Ebf1 and GFP expression plasmids, collected 10 hours (A-C) or 20 hours (D-F) later, and processed for in situ hybridisation with the indicated probes. The pictures show flat mounts of the anterior region of the spinal cord. The cases shown are representative of more than 90% of the embryos, from three independent experiments, each involving at least eight embryos per probe.

View larger version (78K): [in a new window]   Fig. 5. Ebf1 misexpression affects neuronal subtype specification. (A-E) Stage HH15 chick embryos were co-electroporated with Ebf1 and GFP expression vectors, incubated for 30 hours, sectioned and processed for in situ hybridisation with CRABPI (A) and Lim1 (B) probes, or for double in situ hybridisation with CRAPBI and Islet1 (C), CRABPI and Islet2 (D), or Lim1 and Islet1 (E) probes. In double-labelling experiments, the colour code is shown underneath. Note the presence of cells ectopically expressing CRABPI or Lim1 within the ventricular zone (arrows), and at the level of the motor column on the electroporated side (arrowheads). These latter cells do not express Islet1 (C,E). The cases shown are representative of more than 80% of the embryos, from six independent experiments, each involving at least six embryos per probe. (F-H) Embryos were electroporated with HA-tagged Ebf1 at stage HH13 (F), HH15 (G) and HH17 (H), and collected 30 hours later. Vibratome sections were processed for immunofluorescence analysis with antibodies directed against the HA epitope (red) and Islet1/Islet2 (green). (I) Quantification of the data obtained from the experiments presented in F-H. The bars represent the number of Islet-positive cells per section on the control side (green, black bars), the number of HA-Ebf1-positive cells per section within the estimated motor column on the electroporated side (red, grey bars) and the number of Islet-positive cells per section on the electroporated side (green, white bars). The data represent mean±s.e.m. of cell counts and correspond to the analysis of six to nine sections of at least three independently processed embryos for each condition. Only sections containing at least 10 HA-Ebf1-positive cells within the motor column have been taken into account. P<0.001 for the differences observed in the number of Islet-positive cells between control and electroporated sides. Electroporation was on the right side. Electroporated constructs are indicated at the top of each panel, and immunolabelling or in situ hybridisation probes at the bottom. MC, motor column.

View larger version (110K): [in a new window]   Fig. 6. Ebf acts as a dominant-negative mutant and impairs neuronal differentiation. (A-D) Flat-mounted hindbrains from chick embryos that were not electroporated (A) or were co-electroporated at stage HH10 with the indicated constructs (B-D) and the GFP expression vector, then collected 24 hours later and processed for neurofilaments immunochemistry. The examples shown are representative of more than 90% of the embryos, from six independent experiments, each involving at least eight embryos. (E-H) Transverse sections from stage HH15 embryos that were co-electroporated with Ebf and GFP expression vectors, collected 30 hours later and processed for whole-mount in situ hybridisation with CRABPI (E), Islet1 (F), Ngn2 (G) and NeuroM (H) probes. The cases shown are representative of more than 80% of the embryos, from four independent experiments, each involving at least eight embryos per probe. Electroporation was on the right side. Note that Ebf blocks neurofilament induction promoted by Ebf1 and reduces the level of late (CRABPI, Islet), but not early (Ngn2, NeuroM), neurogenesis markers.

View larger version (88K): [in a new window]   Fig. 7. Ebf uncouples cell cycle exit from neuronal differentiation. (A-C) Stage HH15 chick embryos were co-electroporated with GFP and mutant Krox20 (K20m-HA, Kr), Ebf (Ebf, E) or Ebf1 (HA-Ebf1, E1) expressing constructs as indicated, subjected to 2-hour BrdU pulse-labelling 28 hours later, and vibratome sectioned for direct detection of GFP fluorescence (shown in red) and immunofluorescence analysis of incorporated BrdU (green). The mutant Krox20 protein is inactive and was used as a control. The dashed line in A indicates the separation between the mantle layer (ML) and the ventricular zone (VZ). (D) Quantification of the data obtained from the experiments presented in A-C. The bars represent the percentage of electroporated cells (GFP-positive, red or yellow) that are located in the ventricular zone (black bars), the percentage of electroporated cells that are BrdU-positive (yellow, white bars), and the percentage of electroporated cells within the ventricular zone that are BrdU-negative (red, grey bars). (E-G) Stage HH15 chick embryos were electroporated with Flag-Ngn2 (n2) alone or together with Ebf or Ebf1 expression constructs as indicated, subjected to 2-hour BrdU pulse-labelling 28 hours later, and vibratome sectioned for immunofluorescence analysis of incorporated BrdU (green), Flag (blue) and HA (red) epitopes, marking Ngn2 and Ebf1, respectively. (H) Quantification of the data obtained from the experiments presented in E-G. The bars represent the percentage of electroporated cells that are Ngn2-positive (blue, purple or white), which are located in the ventricular zone (black bars), and the percentage of electroporated cells that are BrdU-positive (white, white bars). In D and H the data represent mean±s.e.m. and correspond to the analysis of six to nine sections from at least three independently processed embryos for each condition. (I-K) Stage HH15 chick embryos were co-electroporated with GFP and Ebf, Flag-Ngn2 alone, and Flag-Ngn2 and Ebf expressing constructs as indicated, incubated for 30 hours and then vibratome sectioned for immunofluorescence analysis of Tuj1 (green) and Flag epitope (Blue), or direct detection of GFP fluorescence (red). (L,M) Stage HH15 chick embryos were co-electroporated with lacZ and Krox20m or Ebf expression plasmids, respectively, incubated for 40 hours and then vibratome sectioned for immunofluorescence analysis of Tuj1 (blue), and ß-galactosidase (ß-gal, red) and TUNEL analysis (green). In sections electroporated to the same extent, as judged by lacZ expression, apoptosis is significantly increased (3.5±1-fold) in Ebf-versus Krox20m-electroporated embryos. More than half of the TUNEL-positive cells in M are also lacZ-positive, although only weakly. Cell counts were performed on eight sections from three independently processed embryos. Electroporation was on the right side.

View larger version (19K): [in a new window]   Fig. 8. Schematic representation of epistatic relationships in the CNS neurogenic pathway (see text for detailed explanation). Cellular stages are indicated on the left. Plain arrows indicate established regulatory links between genes. Dashed arrows correspond to links participating in putative positive-feedback loops involved in stable commitment. The grey arrow represents data derived from the work of Perron and collaborators (Perron et al., 1999).