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

doi: 10.1242/10.1242/dev.00318

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 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 Ghislain, J. Articles by Frain, M. Search for Related Content PubMed PubMed Citation Articles by Ghislain, J. Articles by Frain, M.

Neural crest patterning: autoregulatory and crest-specific elements co-operate for Krox20 transcriptional control

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 (69K): [in a new window]   Fig. 1. Autoregulation contributes to Krox20 expression in the neural crest. Heterozygous (A,C, Krox20+/lacZ) and compound mutant (B,D, Krox20Cre/lacZ) embryos were analysed for ß-galactosidase activity in toto at 9.0 (A,B) and 9.5 (C,D) dpc. Expression in the neural crest and rhombomeres 3 and 5 are indicated (arrowheads). Embryos are oriented with anterior to the right. NC, migrating neural crest.

View larger version (81K): [in a new window]   Fig. 2. Identification of a neural crest-specific, autoregulatory element controlling Krox20 expression. (A-D) The transgenic line -31/+7 Krox20/lacZ (see Fig. 3, construct #1) was analysed for ß-galactosidase activity in toto at 8.5 dpc (A), 9.0 dpc (C) and 9.5 dpc (D), or by transverse section through rhombomere 5 at 9.0 dpc (B, dorsal to the top). (E,F) The -31/+7 Krox20/lacZ transgene was analysed in a Krox20 mutant background (Krox20Cre/Cre) at 9.0 dpc (E) and 9.5 dpc (F). Embryos are oriented with anterior to the right (A, C-F). NC, migrating neural crest. wt, wild type. The scale bar in B represents 40 µm.

View larger version (21K): [in a new window]   Fig. 3. Genomic constructs analysed for NCE activity by transgenesis in the mouse and electroporation in the chick neural tube. The -31/+7 Krox20/lacZ transgenic construct containing a lacZ in-frame insertion in Krox20 is indicated at the top. Restriction enzymes used to clone subfragments are shown. Distances are in kb and indicate the position relative to the start site of transcription of Krox20. Scale bars relevant to each subregion are indicated on the left. With the exception of constructs #1 and #2 which contain the Krox20/lacZ reporter, all subfragments were cloned into the ß-globin/lacZ promoter/reporter vector. Fragment #6 was cloned in both the sense (S) and antisense orientations (AS). K1/K2, Krox20 binding sites (open circle, wild type; X, mutated). For details of the Krox20 binding sites, their mutation, the regions deleted in constructs #19-23 and the sequences multimerised in construct #24 see Fig. 5. In mouse transgenic experiments, embryos were analysed at 9.5 dpc for ß-galactosidase activity. Those carrying construct #1 were analysed at F1 and all others were analysed in `transient' transgenesis. Ratios correspond to the number of ß-galactosidase positive/total number of transgenic embryos. Unless indicated in brackets, mouse embryos were scored for expression in the r5 neural crest stream (NC). Embryos with expression in r5 were among those that showed expression in the neural crest. Chick embryos were scored at Hamburger and Hamilton stages 13-14 for ß-galactosidase activity in r3, r5 and the r5 neural crest stream (r3/r5+NC), unless indicated in brackets. NC, expression restricted to the r5 neural crest stream; CNC, expression throughout the cranial neural crest; +, constructs for which at least 60% of the electroporated embryos gave the indicated profile; -, those that were essentially negative in these tissues in at least two sets of five electroporated embryos; weak, when the levels of ß-galactosidase activity detected were considerably lower than for the controls; ND, not determined.

View larger version (82K): [in a new window]   Fig. 4. Localisation of the mouse and chick NCE. Genomic fragments were fused to the ß-globin/lacZ promoter/reporter and analysed for ß-galactosidase activity by electroporation in the chick neural tube (B,C,E,G,H) or by transgenesis in the mouse at 9.5 dpc (A,D,F). (A,B) Mouse sequences between -31 kb and -23.5 kb relative to Krox20 (Fig. 3, construct #6S). (C,D) Sequences between -26.5 kb and -25.5 kb (Fig. 3, construct #11). (E-G) A 247 bp subfragment of this region either alone (E,F; Fig. 3, construct #17) or co-electroporated with a Krox20 expression construct (G; Fig. 3, construct #17+). (H) A 1.7 kb chick sequence with homology to mouse fragment #17. All embryos are whole mounts with anterior to the right. r3, r5 and neural crest (NC) are indicated by arrowheads. *, occasional ectopic expression.

View larger version (19K): [in a new window]   Fig. 5. Mouse, human and chick interspecies comparisons reveal sequence conservation within the NCE. The nucleotide sequence of mouse fragment #17 (Fig. 3) is shown. Restriction sites used to generate internal deletion fragments #19 (BfaI and Bsu36I), #20 (SmaI and BfaI), #21 (SmaI and BanII) and #22 (BanII and BfaI) are underlined. , indicates the region deleted in construct #23. Sequences multimerised in fragment #24 are shown. The putative, head-to-head HMG box binding sites are indicated (HMG1/HMG2). Conserved sequences identified in the human genome and in the chick 1.7 kb NCE (Fig. 4) are aligned to the mouse sequence. Conserved residues are indicated as a dash in the human and chick sequences. Sequences satisfying the known Krox20 binding site matrix in all three species and footprinted by the protein in vitro (Fig. 6) are double underlined (K1 and K2). The G to C substitutions used to inactivate the Krox20 binding sites in constructs #18-23 are shown. Nucleotide numbering corresponds to the mouse sequence.

View larger version (77K): [in a new window]   Fig. 6. In vitro analyses of the NCE identify two Krox20 binding sites. (A) Both the upper (left) and the lower (right) strand of the 247 bp fragment #17 (Fig. 3) were analysed for Krox20 binding sites in DNase I footprinting assays using extracts from control or Krox20-expressing bacteria. Two doses of DNase I were used (0.2 and 0.5 units). The G+A reaction is included for both strands. The position of the two Krox20 footprints and the range of sequence analysed are indicated for both strands. (B) Control and Krox20-expressing bacterial extracts were analysed in bandshift assays using either the wild-type (upper) or Krox20 binding site mutant (lower) fragment #17 as a probe. The volume of bacterial extract was varied between 0.5-2.0 µl. To identify specific complexes, unlabelled competitor oligonucleotides corresponding to a high affinity Krox20 binding site (wt) or a mutant version (mt) were included in the binding reaction at 50-200-fold molar excess (Sham et al., 1993). Specific complexes are indicated with brackets. The different mobility protein-DNA complexes are likely a result of the presence of two Krox20 binding sites, one of which is asymetrically localised (Fig. 5). FP, free probe.

View larger version (92K): [in a new window]   Fig. 7. In vivo analysis of essential NCE elements. (A-C) The activity of the NCE carrying mutations in the Krox20 binding sites alone (A,B; Fig. 3, construct #18) or both the Krox20 binding site mutations and a 15 nt deletion to inactivate the putative HMG box binding sites (C; Fig. 3, construct #23), were analysed by chick electroporation (A,C) or by mouse transgenesis (B). ß-galactosidase activity was dramatically reduced (A,B) or eliminated (C) as compared to the wild-type construct (#11, see Fig. 4B-D). (D,E) A 41 bp sequence encompassing the putative HMG box binding sites (Fig. 5) was multimerised, linked to the reporter (Fig. 3, construct #24) and analysed by chick electroporation (D) and mouse transgenesis (E). Weak ß-galactosidase activity is observed throughout the cranial neural crest. Embryos are whole mounts with anterior to the right. NC, neural crest.

View larger version (69K): [in a new window]   Fig. 8. Maintained Krox20 expression in the migrating neural crest requires Sox10. (A,B) The expression of Sox10 was analysed by in situ hybridisation in mouse embryos at 8.5 dpc. (A) Whole mount of a wild-type embryo. The territory corresponding to r5 is indicated. (B) Transverse section caudal to r5 of a Krox20+/lacZ embryo stained for both ß-galactosidase activity (light blue) and Sox10 (purple) revealing their co-expression in the migrating neural crest. (C,D) Krox20 expression was analysed by whole-mount in situ hybridisation in Sox10 heterozygous (Sox10+/Dom) and homozygous (Sox10Dom/Dom) mutant embryos at 9.0 dpc (20 somites). Embryos are presented with anterior to the right. NC, neural crest; NT, neural tube.

View larger version (13K): [in a new window]   Fig. 9. Synergistic activation of the NCE by Krox20 and Sox10. Expression constructs encoding Krox20 (pKrox20) and Sox10 (pSox10) or the empty plasmid were transfected into cultured HeLa cells along with either the empty ß-globin/lacZ promoter/reporter plasmid or the wild-type NCE (Fig. 3, fragment #11) or its derivatives fused to the promoter/reporter as indicated. The putative HMG box and Krox20 binding sites are indicated with squares and circles, respectively. X, represents the mutation of these sites. A 7X multimer of a 41 nucleotide sequence spanning the putative HMG box binding sites (Fig. 5) is indicated (right). Expression plasmids were transfected at either 100 ng/plate (++) or 20 ng/plate (+). The data show the ß-galactosidase activities of one experiment performed in duplicate and is representative of two independent experiments. Values from transfections with the empty promoter/reporter and expression plasmids were arbitrarily set to one. Data from all other transfections are presented as the fold induction over this level. Error bars represent the standard error.

View larger version (16K): [in a new window]   Fig. 10. A schematic model for the regulation of Krox20 expression in r5-derived neural crest. We propose that the establishment of Krox20 expression in r5-derived neural crest involves two steps. Krox20 is first activated in the entire r5 territory under the control of an initiator cis-acting element distinct from the NCE (large grey arrow). In r5 premigratory NC cells, the Krox20 protein and crest-specific Sox proteins bind to sequences in the NCE and co-operate in the establishment of a positive, autoregulatory loop (large black arrow). This autoregulatory mechanism is then sufficient to maintain Krox20 expression in migratory NC cells. K1 and K2, Krox20 binding sites; HMG1/2, putative Sox protein binding sites.