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

First published online 5 January 2005
doi: 10.1242/dev.01593

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 Serpente, P. Articles by Gould, A. P. Search for Related Content PubMed PubMed Citation Articles by Serpente, P. Articles by Gould, A. P.

Direct crossregulation between retinoic acid receptor ß and Hox genes during hindbrain segmentation

Patricia Serpente¹, Stefan Tümpel², Norbert B. Ghyselinck³, Karen Niederreither^3,*, Leanne M. Wiedemann^2,4, Pascal Dollé³, Pierre Chambon³, Robb Krumlauf^2,5 and Alex P. Gould^1,

¹ Medical Research Council, National Institute for Medical Research, The Ridgeway, Mill Hill, London, NW7 1AA, UK
² Stowers Institute for Medical Research, 1000 East 50th Street, Kansas City, MI 64110, USA
³ Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC), CNRS/INSERM/ULP/Collège de France, BP10142, 67404 ILLKIRCH Cedex, France
⁴ Department of Pathology and Laboratory Medicine, Kansas University Medical Center, Kansas City, KS 66160, USA
⁵ Department of Anatomy and Cell Biology, Kansas University Medical Center, Kansas City, KS 66160, USA

View larger version (56K): [in a new window]   Fig. 1. Rarb is expressed more anteriorly than Hoxb4 in the presegmented hindbrain. The anterior neural expression borders of Hoxb4 protein (A,C,E) and Rarb mRNA (B,D,F) are compared at E8.5 and E9.5. (A,B) At E9.5, neural expression of both genes extends up to a straight border at the r6/r7 boundary (arrowhead), lying just caudal to the otic vesicle. (C,D) At E8.5, using the otic sulcus (Os) as a landmark for presumptive r5/r6, it can be seen that neural Rarb expression extends more anteriorly than that of Hoxb4. (E,F) At E8.5, with Krox20 expression marking presumptive r5, it can be observed that Rarb is expressed in all of presumptive r6 but Hoxb4 is not. At this stage, the anterior expression border of both genes is diffuse and, for Rarb, weak expression may also extend into the posterior part of presumptive r5. Anterior is towards the top in this and all subsequent figures.

View larger version (52K): [in a new window]   Fig. 2. Early hindbrain expression of Hoxb4 and Rarb requires Raldh2 activity. The effects of loss of Raldh2 activity on the early neural expression of Hoxb4 mRNA (A,B), Hoxb4 protein (C,D) and Rarb gene (E,F) are shown. (A,B) Dorsal views of E8.5 embryos showing the Hoxb4 mRNA distribution in Raldh2+/+ (A) and Raldh2-/- (B) genetic backgrounds. Loss of Raldh2 function is associated with absence of Hoxb4 mRNA within the hindbrain but expression within the developing spinal cord (sc) remains unaffected. (C,D) Dorsal views of E8.5 embryos showing Hoxb4 protein expression in Raldh2+/- (C) and Raldh2-/- (D) embryos. Removal of Raldh2 activity leads to loss of most Hoxb4 protein within both the developing hindbrain (hb) and spinal cord (sc). However, weak expression is still observed within a few cells at caudal hindbrain and anterior spinal levels. (E,F) Rarb mRNA induction within the hindbrain requires Raldh2 activity. Lateral views of E8.5 embryos of the genotypes Raldh2+/- (E) or Raldh2-/- (F). Loss of Raldh2 activity is associated with the absence of all detectable Rarb expression within the developing hindbrain and spinal cord. Os and Ov indicate the otic sulcus and otic vesicle, lying adjacent to r5 and r6 in this and subsequent figures, and the approximate position of the paraxial mesodermal limit of Hoxb4 expression at the somite 6/7 border is also shown (horizontal line in A-D).

View larger version (54K): [in a new window]   Fig. 3. Hoxb4 and Hoxd4 regulate the hindbrain expression of Rarb at late but not at early stages. All panels show the expression of Rarb mRNA in dorsal views of E9.0-E10.5 embryos lacking from 0 to 4 wild-type alleles of Hoxb4 and Hoxd4. The wild-type position of the neural Rarb expression border at r6/r7 (arrowhead) is indicated. (A) E9.0 embryos with Hoxb4+/-; Hoxd4-/- (left) and Hoxb4-/-; Hoxd4-/- (right) genotypes showing wild-type patterns of Rarb expression. At this intermediate stage, although the Rarb neural border lies in the vicinity of the forming r6/r7 junction, it is not as sharp as at E9.5. (B) E10.5 Hoxb4+/+; Hoxd4-/- embryo with a normal neural border of Rarb expression at r6/r7. (C,D) E9.5 embryos that are wild type (C) or Hoxb4-/-; Hoxd4-/- (D) showing abnormal regression of the Rarb border associated with loss of function of both Hox paralogues. (E,F) E10.5 embryos that are wild type (E) or Hoxb4-/-; Hoxd4-/- (F), showing that, by this late stage, removing the activity of both Hox paralogues leads to loss of most Rarb expression from the hindbrain.

View larger version (43K): [in a new window]   Fig. 4. The Rarb Proximal enhancer recapitulates early neural expression and responds transiently to RA. All panels show dorsal views of Rarb2lacZ transgenic embryos at E8-E10.25, histochemically stained for ß-galactosidase activity. (A,B) E8-E8.5 transgenic embryos at the two-somite (A) and seven-somite (B) stages. At both time points, strong reporter expression is present within the neural plate with weaker staining also visible in paraxial and lateral mesoderm. By the seven-somite stage, a diffuse anterior border maps to the vicinity of the Os. (C,D) E9.5 embryos of the genotype Rarb2lacZ/+; Hoxb4+/+; Hoxd4+/- (C) and Rarb2lacZ/+; Hoxb4-/-; Hoxd4-/- (D) showing that, at this late stage, the neural activity of the proximal enhancer is largely confined to the developing spinal cord with only residual low levels remaining within the caudal hindbrain. This late pattern is unaffected when the functions of both Hoxb4 and Hoxd4 are removed. (E,F) E9.25 transgenic embryos treated with an exogenous pulse of RA, showing a clear ectopic response within r4-r6 at 4 hours (E), but not 24 hours (F), after treatment. The expression pattern in F is similar to that seen in C.

View larger version (36K): [in a new window]   Fig. 5. The Rarb Distal enhancer specifies the late r6/r7 border and requires inputs from Hoxb4 and Hoxd4. Panels show dorsal views (A) or lateral views (B-H) of Rarb1lacZ transgenic embryos at E8.5-E10.5, histochemically stained for ß-galactosidase activity. (A,B) An E8.5 transgenic embryo (A) and a E9.5 transgenic embryo (B), indicating that distal enhancer activity is initiated after E8.5 and by E9.5 is restricted to the dorsal neural tube with a sharp border at r6/r7. (C,D) Transgenic embryos treated with a single dose of RA at E9.25 and allowed to develop in utero for either 4 hours (C) or 24 hours (D). No ectopic response to RA is detected in either case. (E-H) E10.5 transgenic embryos carrying various combinations of Hoxb4 and Hoxd4 loss-of-function alleles. Strong distal enhancer activity is seen in Rarb1lacZ/+; Hoxb4+/+; Hoxd4+/- (E), weak distal enhancer activity in Rarb1lacZ/+; Hoxb4+/-; Hoxd4+/- (F) and Rarb1lacZ/+; Hoxb4+/-; Hoxd4-/- (G) and no detectable distal enhancer activity in Rarb1lacZ/+; Hoxb4-/-; Hoxd4-/- embryos (H).

View larger version (53K): [in a new window]   Fig. 6. A conserved Hoxb4-binding site is required for Rarb distal enhancer activity in chick embryos. (A) Map of the 5' end of the Rarb gene. Transcripts for the RARß1 and RARß3 isoforms initiate at the P1 promoter and those of the RARß2 and RARß4 isoforms at the P2 promoter. The positions of the RARE (blue oval; proximal enhancer) and the Hox/Pbx (HP) element (green oval, distal enhancer) are shown. White and grey boxes indicate transcribed non-coding and coding sequences, respectively. The positions of the 2.3 kb distal enhancer and the 3.8 kb proximal enhancer are also shown. Exon E1', which is equivalent to E4 in previous publications, is relabelled here to emphasize that it is the first exon of Rarb2/4 transcripts and not contiguous with Rarb1/3 transcripts. (B) Alignment of the Hox/Pbx (HP) element and flanking sequences in five mammalian species. The HP site is boxed in green and indicated below is the region multimerized for the 3xHP construct used in F. (C,D) Electrophoretic mobility shift assays showing Hoxb4 binding to the HP site. Labelled oligonucleotides containing the HS1+HS2 site (Gould et al., 1997) or the HP site (from Rarb) were incubated with increasing amounts of Hoxb4 protein (C) or with a constant amount of Hoxb4 protein in combination with Hoxb4 antibody (Hoxb4) or unlabelled competitor oligonucleotides as indicated (D). (E-G) Dorsal views of chick embryos electroporated with lacZ reporter constructs containing the mouse 2.3 kb distal enhancer (E), 3xHP oligonucleotide (F) or 2.3 kb distal enhancer with mutated HP site (G). lacZ expression with an r6/r7 boundary can be observed on the electroporated (right) side of the neural tube, except when the distal enhancer containing the mutated HP site (G) was used. The positions of the otic vesicle (OV) and the r6/7 boundary (arrowhead) are shown. Constructs are diagrammed below each panel, the circle indicating intact HP site (green fill) or HP MUT site (cross).

View larger version (24K): [in a new window]   Fig. 7. The Hox-RAR feedback circuit. Cartoon shows the Hox-responsive enhancers (HOX, green) and RA-responsive enhancers (RARE, blue) from Rarb (top), Hoxb4 (middle) and Hoxd4 (bottom). Enhancer-promoter interactions (black arrows), transcription and translation (unfilled arrows), and transcriptional regulation by Hoxb4 and Hoxd4 proteins (green arrows) and RARs (blue arrows) are shown. The external RA signal that initiates this neural genetic circuit (red arrow) is synthesized by Raldh2 in the adjacent mesoderm. A functional equivalent of the LNE/distal enhancer (HOX element with broken outline) has, thus far, not been identified for Hoxd4 (Nolte et al., 2003). The correct positioning of the single or multiple transcription start sites of each gene has been omitted for clarity. The central importance of the HoxRARß interactions described in this study are emphasized with thick green arrows. This figure also summarizes genetic interactions identified in several other published studies (see text for details).