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

doi: 10.1242/10.1242/dev.00539

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 Scardigli, R. Articles by Le Roux, I. Search for Related Content PubMed PubMed Citation Articles by Scardigli, R. Articles by Le Roux, I.

Direct and concentration-dependent regulation of the proneural gene Neurogenin2 by Pax6

Raffaella Scardigli^1,*, Nicole Bäumer², Peter Gruss², François Guillemot^1,, and Isabelle Le Roux¹

¹ Institut de Génétique et de Biologie Moléculaire et Cellulaire, BP 163, 67404 Illkirch, Cedex, CU de Strasbourg, France
² Max Planck Institute of Biophysical Chemistry, Department of Molecular Cell Biology, Am Fassberg 11, D-37077 Göttingen, Germany
^* Present address: Institute of Cell Biology and Tissue Engineering, Via di Castel Romano 100/102, 00128 Rome, Italy
Present address: Division of Molecular Neurobiology, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, UK

View larger version (105K): [in a new window]   Fig. 1. Activity of the Ngn2 enhancer, E1, is restricted to domains of the embryonic spinal cord and telencephalon expressing high levels of Pax6 protein. Double immunocytochemistry with an -ß-galactosidase antibody (red) and an -Pax6 antibody (green), on transverse sections of spinal cord (A-C) and frontal sections of telencephalon (D-F) from an E10.5 mouse embryo transgenic for the E1hsplacZ construct. C and F show high magnifications of areas boxed in B and E, respectively, with merged -ß-gal and -Pax6 staining. Activity of the E1 element is restricted to a ventromedial domain in the spinal cord and a lateral domain in the cerebral cortex.

View larger version (94K): [in a new window]   Fig. 2. Pax6 is both necessary and sufficient to regulate endogenous Ngn2 expression and activate the E1 enhancer. (A-D) Double immunocytochemistry with an -ß-galactosidase antibody (green) and an -Nkx6.2 antibody (red, A,B), or an -Nkx6.1 antibody (red, C,D), on transverse sections of spinal cord from E10.5 embryos, heterozygous for the Ngn2KIlacZ allele, and either wild-type (A,C) or homozygous Sey mutants (B,D) at the Pax6 locus. ß-gal expression is down-regulated in Nkx6.2-expressing cells (p1 domain, arrowhead in B) and dorsal Nkx6.1-expressing cells (p2 domain, arrowhead in D). (E-J) Dorsal views of whole-mount chick neural tubes labelled for Ngn2 (E), Pax6 (F,J), ß-gal (G,I) and GFP (H). Embryos were harvested 6 hours after being electroporated with a CMVPax6 vector (E,F,I,J), an E1ßglobinlacZ vector (G-J) or a CMVGFP vector (H). The electroporated side of neural tubes is at the bottom of the panels. Only a few Ngn2-positive and ß-gal-positive cells are detectable at this stage (arrowhead in the unelectroporated side of the neural tube in E, and electroporated side in G, respectively), where endogenous levels of Pax6 are low (top in F). In the presence of high exogenous levels of Pax6 protein (bottom in F and J), the number of cells expressing endogenous Ngn2 (E) and activating the E1 element (I) is strongly increased. The inset in J shows two cells co-expressing ß-gal and high levels of Pax6. Dashed lines in left panels outline the neural tube.

View larger version (51K): [in a new window]   Fig. 3. Identification of Pax6 binding sites in Ngn2 enhancers. (A) Schematic representation of the Ngn2 locus showing the position of the E1 and E3 enhancers, the organization of the E1 enhancer, and the position of the E1.1 and E3.2 Pax6 binding sites, showing high similarity with a consensus binding sequence. The blue box represents a block of sequence in the E1 element showing high similarity (94%) between the murine and human Ngn2 genes. The top sequences are the published consensus Pax6 binding site (Epstein et al., 1994; Czerny and Busslinger, 1995), the middle sequences are the Pax6 binding sites found in the human gene and the bottom sequences are the the same sites in the mouse gene. Red letters indicate conserved nucleotides between sequences in the Ngn2 enhancers and the consensus binding sequence, and black letters indicate mismatches. (B, left panel) Electromobility shift assay performed with recombinant Pax6 protein and oligonucleotides containing the E1.1 and E3.2 sequences, the consensus Pax6 binding site as a positive control (cons), a mutated version of E1.1 as a negative control (mtE1.1), and an optimized version of E1.1 (consE1.1). Oligonucleotides corresponding to the sequences surrounding and including the E1.1 and E3.2 binding sites form a complex with Pax6 protein, but twice the amount of Pax6 protein was required to form a complex with E1.1 as compared to E3.2 or the consensus sequence. The smaller amount of complex formed with the E1.1-containing oligonucleotide suggests that the E1.1 sequence has a low affinity for Pax6. (B, right panel) The interaction of Pax6 and E1.1 is disrupted by incubation with an antibody to Pax6 but not to Pax2. Also Pax3 and Pax8 recombinant proteins do not form complexes with E1.1. The interaction of Pax6 with E1.1 is therefore specific. (C) Double labelling for ß-gal (red, left panels) and GFP (green, right panels) on chick neural tubes 6 hours after electroporation with the constructs 4xE1.1-ßglobinlacZ (a,b,e,f), 4xE3.2-ßglobinlacZ (c,d,g,h) and CMVPax6 (e-h). A CMVGFP vector was co-electroporated to control for transfection efficiency (b,d,f,h). The E3.2 concatemer efficiently drives ß-gal expression in the early neural tube (c), where endogenous levels of Pax6 protein are low (see Fig. 1), while the E1.1 concatemer does not (a). Activity of the E1.1 concatemer is significantly enhanced in the presence of exogenous Pax6 protein (e), while activity of the E3.2 concatemer is not further increased. Dashed lines outline the neural tube.

View larger version (75K): [in a new window]   Fig. 4. Disruption of the E1.1 Pax6 binding sequence leads to a severe reduction of E1 enhancer activity. (A,B) Whole-mount X-gal staining of E10.5 mouse embryos carrying the control E1hsplacZ (A) and the mutated mtE1hsplacZ (B) transgenes. (C,D) Transverse sections through the same embryos at brachial levels. The mutation of the E1.1 sequence (B,D) leads to a severe decrease of transgene activity in the spinal cord (sc) and a complete loss of activity in the telencephalon (tel) as compared with the control transgenic embryo (A,C).

View larger version (111K): [in a new window]   Fig. 5. Increasing Pax6 expression levels by electroporation induces ectopic activity of the E1 enhancer in the chick spinal cord. Labelling for ß-gal (A,C,E,F), GFP (B) and Pax6 (D,F) on transverse sections of spinal cord, 48 hours after the electroporation of the E1ßglobinlacZ construct (A-F), together with the tracer CMVGFP (A,B) or the CMVPax6 construct (C-F). The activity of E1ßglobinlacZ is restricted to the medioventral part of the spinal cord, in chick (A,B) as in mouse (Fig. 4C). High level of Pax6 protein delivered by electroporation (D) leads to ectopic activation of E1ßglobinlacZ (C). E shows a higher magnification of the boxed area in C. F shows merged Pax6 and ß-gal staining of the same enlarged area. Co-expression of ß-gal and Pax6 shows that ß-gal is induced in cells that belong to a domain of low Pax6 expression but that themselves express high Pax6 levels. Note that the -Pax6 antibody used recognises both endogenous and exogenous proteins.

View larger version (46K): [in a new window]   Fig. 6. The domain of E1 enhancer activity in the telencephalon is expanded when the dose of Pax6 is increased. (A,B) X-gal staining of E10.5 E1hsplacZ transgenic embryos in a wild-type background (A) and a PAX6YAC transgenic background (B). (C-F) X-gal staining of frontal sections of the telencephalon of E10.5 (C,D) and E12.5 (E,F) E1hsplacZ transgenic mice in the same genetic backgrounds. E1 activity is restricted to a lateral domain in the cerebral cortex (A,C), which extends ventrally up to the border with the lateral ganglionic eminence (E). In the presence of multiple copies of the human Pax6 gene in Pax6YAC mice (Schedl et al., 1995), the domain of E1 activity is expanded both ventrally and dorsally (D,F). cc, cerebral cortex; lge, lateral ganglionic eminence.

View larger version (76K): [in a new window]   Fig. 7. Activity of the E1 enhancer is increased in the chick spinal cord when the low affinity Pax6 binding sequence has been replaced with a consensus binding site. Labelling for ß-gal (A,C,E,H), GFP (B,D,F,I) and Pax6 (G,J), of chick neural tubes harvested 6 hours (A-D) or 48 hours (E-J) after electroporation with the constructs E1ßglobinlacZ (A,B,E-G), consE1ßglobinlacZ (C,D,H-J) and CMVGFP (B,D,F,I). In A-D, neural tubes are shown in dorsal views and the electroporated side is towards the bottom. Activity of the E1 element is low at this early stage (HH stage 13-15), and introducing a consensus Pax6 binding sequence at the E1.1 site significantly increases activity of the E1 element (C). The dashed lines outline the shape of the neural tube. In E-J, -ß-gal and -GFP stainings were performed on the same transverse sections of spinal cord, and -Pax6 staining on adjacent sections. Activity of the E1 element at this stage (HH stage 21-22) is confined to a medial domain of high Pax6 concentration (F,G), whereas the modified element consE1 is active in a broader domain that includes cells expressing low Pax levels (I,J).

View larger version (39K): [in a new window]   Fig. 8. Optimizing the sequence of the E1.1 binding site leads to an expansion of the domain of E1 activity in the cerebral cortex. X-gal staining of E11 transgenic embryos (A,B) and of frontal sections of the telencephalon of E11 (C,D) and E12.5 (E,F) embryos. Activity of the E1 element is restricted to the lateral cortex (A,C,E), and introduction of a consensus Pax6 binding sequence into the E1.1 site leads to an expansion of the activity of the element to a more dorsal domain (B,D,F). Arrowheads in C and D mark the dorsal and ventral limits of the E1 activity domain.