May 1, 2000;
A screen for targets of the Xenopus T-box gene Xbra.
(T), a member of the T-box gene family, is essential for the formation of posterior mesoderm
in vertebrate development. Expression of the Xenopus homologue of Brachyury
, causes ectopic ventral
and lateral mesoderm
formation in animal cap explants and co-expression of Xbra
, a forkhead/HNF3beta-related transcription factor, induces notochord
. Although eFGF
and the Bix genes are thought to be direct targets of Xbra
, no other target genes have been identified. Here, we describe the use of hormone-inducible versions of Xbra
to construct cDNA libraries enriched for targets of these transcription factors. Five putative targets were isolated: Xwnt11
, the homeobox gene Bix1
, the zinc-finger transcription factor Xegr-1, a putative homologue of the antiproliferative gene BTG1
, and BIG3/1A11, a gene of unknown function. Expression of Xegr-1 and Xbtg1
is controlled by Pintallavis
alone as well as by a combination of Xbra
. Overexpression of Xbtg1
perturbed gastrulation and caused defects in posterior
tissues and in notochord
formation, a phenotype reminiscent of that observed with a dominant-negative version of Pintallavis
-En(R). The Brachyury
-inducible genes we have isolated shed light on the mechanism of Brachyury
function during mesoderm
formation. Specification of mesodermal cells is regulated by targets including Bix1
-4 and eFGF
, while gastrulation movements and perhaps cell division are regulated by Xwnt11
[+] show captions
Fig. 2. Expression patterns of Xbra (A), Xwnt11 (B) and BIG3 (C) at the mid-gastrula stage.
1a11 (71 kDa protein) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 10.5, vegetal view dorsal up.
wnt11b (wingless-type MMTV integration site family, member 11B) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 10.5, vegetal view, dorsal up.
Fig. 1. Hormone-dependent induction of notochord in animal cap tissues by co-expression of Xbra-GR and Pintallavis-GR. (A) Schematic illustration of hormone-inducible Xbra and Pintallavis. The hormone-binding domain of the human glucocorticoid receptor (GR) was fused to the carboxy-terminus of full length Xbra and Pintallavis. Stop codons of both genes were replaced with two amino acids (D and L) creating a BglII site (AGATCT) (see Tada et al., 1997). (B) RNAase protection assay of Bix1, a direct target of Xbra, and Xnot in animal caps expressing either Xbra-GR or Pintallavis-GR alone or a combination of the two. Pintallavis-GR alone could not induce either Bix1 or Xnot. Bix1 and Xnot were induced by Xbra-GR alone in a hormone (DEX)-dependent manner. Bix1 expression was induced by co-expression of Xbra-GR and Pintallavis-GR but at a slightly reduced level, while Xnot expression was enhanced. (C) DEX-dependent notochord formation in animal caps co-expressing Xbra-GR and Pintallavis-GR. Animal caps were fixed at stage 41 and stained with the notochord-specific antibody MZ15. Neither Xbra-GR nor Pintallavis-GR alone could induce ectopic notochord formation in animal caps (data not shown).
Fig. 3. Xwnt11 is a direct target of Xbra. Animal caps expressing Xbra-GR were dissected at stage 8 and were either left untreated or were treated with DEX in the presence or absence of cycloheximide (CHX) from stage 8 or 10 for 3 h. ODC serves as a loading control. Expression of Xwnt11 is induced by DEX even in the presence of cycloheximide.
Fig. 5. Expression patterns of Xegr-1 (A,D,G), Xbtg1 (B,E,H) and Pintallavis (C,F,I), at early- (A–C), mid- (D–F) and late- (G–I) gastrula stages. Note the similar expression patterns of these three genes from early to mid-gastrula stages.
Fig. 6. RNAase protection analysis of Xegr-1 and BIG3. (A) Expression of Xegr-1 is induced in animal caps by either Xbra or Pintallavis alone, but maximal activation requires both gene products. (B) Xbra-GR and Pintallavis-GR were expressed either alone or together in animal cap explants and the effects of cycloheximide (CHX) on DEX-dependent Xegr-1 and BIG3 induction were assessed. See text for details.
Fig. 7. Pintallavis is capable of binding to several target sites present in the promoter and in the first intron of the Xegr-1 gene. (A) Gel-shift assays showing that Pintallavis protein binds in vitro to a DNA recognition sequence (XFD-1 in C) (Kaufmann et al., 1995). 32P-Labelled XFD-1 probe was incubated with either reticulocyte lysate alone (lane 2) or with Pintallavis protein synthesized in vitro by reticulocyte lysate (lane 3). Competition was carried out by incubating the Pintallavis protein, prior to the addition of the 32P probe, with a 100-fold excess of unlabelled XFD-1 (lane 4) or a mutated version of XFD-1 (mutXFD-1; lane 5) which has a single base substitution (GTCAACA to TTCAACA in the 7 base pair core sequence of the consensus). The retarded band was competed out by the XFD-1 (lane 4) but not by mutXFD-1 (lane 5), showing that the DNA-protein complex formation is specific. (B) Competition experiments using the double-stranded oligonucleotides listed in (C) as cold competitors. In vitro synthesized Pintallavis protein was incubated either without competitor (lane 3) or with 10- (even-numbered lanes) or 50-fold excess (odd-numbered lanes) of unlabelled competitors indicated above (lanes 4–21), prior to the addition of 32P-labelled XFD-1 probe. (C) The sequences of double-stranded oligonucleotides used in the gel retardation assay. The central 18 base pair of all sequences are flanked by CAGT at the 5′ end and ACGT at the 3′ end for 32P-labelling, which are indicated by italics. The sequence of XFD-1 is adopted from one of the three DNA recognition sites selected by PCR-based binding site selection (Kaufmann et al., 1995). The consensus sequence is compiled from the results of the binding-site selection (Kaufmann et al., 1995). The central 18 base pair sequence of #1 to #4 can be found in the 5′ upstream region and those of #5 to #8 in the first intron of Xegr-1 (Panitz et al., 1998). Nucleotide positions are relative to the transcription start site. Sequences which match the consensus are underlined. The orientation of the sequences relative to the direction of transcription are also indicated. The binding strength determined by the competition assays are indicated on the right most column.
Fig. 8. Xbtg1 is a target of both Xbra and Pintallavis and its overexpression blocks gastrulation movements. (A) Construction of Pintallavis-EnR. The repressor domain of the Drosophila Engrailed protein was fused to the carboxy terminus of the DNA-binding domain of Pintallavis (amino acids 85–241). (B,C) Overexpression of Xbtg1 inhibits gastrulation movements. (B) Section of control stage 12 embryo. (C) Sibling embryo injected with 1 ng Xbtg1 RNA. Gastrulation movements are inhibited. Sections in (B,C) are stained by the Feulgen technique to visualize nuclei (purple) and with an antibody specific for phosphorylated histone H3 to reveal mitotic nuclei (brown). (D–O) Phenotypes of control embryos or embryos injected with Pintallavis-EnR or Xbtg1. Embryos were fixed at stage 35/36 and stained either with notochord-specific antibody MZ15 (E,F,I,J,M,N) or muscle-specific antibody 12/101 (G,K,O). Otic vesicles are indicated by arrowheads in (E,F). Embryos injected with Pintallavis-EnR or Xbtg1 frequently had enlarged notochords (compare (F), a control uninjected embryo, with (J), an embryo injected with Pintallavis-EnR, and (N), an embryo injected with Xbtg1; all three are dorsal views at the same magnification). (P–R) Suppression of Xbtg1 expression by Xbra-EnR and Pintallavis-EnR. A single cell of embryos at the 8- or 32-cell stage was injected with Xbra-EnR (P), Pintallavis-EnR (Q) or both (R), together with lineage tracer Fluorescein-dextran (red). The injected embryos were fixed at stage 10.5 and were hybridized with Xbtg1 probe (blue).