September 1, 1999;
Patterning of the mesoderm involves several threshold responses to BMP-4 and Xwnt-8.
Two secreted signaling molecules, Xwnt-8
, play an essential role in the dorso-ventral
patterning of the mesoderm
in Xenopus. Here we investigate how the Wnt-8
and the BMP-4
pathways are connected and how they regulate target genes in the lateral
and ventral marginal zone
regulates the transcription of Xwnt-8
in a threshold dependent manner. High levels of BMP-4
induce the expression of the Wnt antagonist sizzled
in the ventral marginal zone
, independent of Xwnt-8
induces the early muscle
in the lateral marginal zone
in a BMP independent manner. The expression of the homeobox gene Xvent-1
can be modulated through both the BMP-4
and the Xwnt-8
pathways. The spatial distribution and the level of BMP-4
activity in the lateral
and ventral marginal zone
is reflected in the dynamic expression pattern of Xwnt-8
. The data support the view that Xwnt-8
is involved in the specification of lateral
in the specification of ventral mesoderm
[+] show captions
Fig. 1. Activation/repression of Xwnt-8 expression as a function of BMP-4 RNA dose. The effect of changes in the level of Xwnt-8 expression as a function of the level of BMP-4 in LiCl-dorsalized embryos. The expression level of Xwnt-8 was analyzed by RT-PCR and the phenotype of LiCl-treated embryos was determined at stage 34/35. (A) RT-PCR analysis of Xwnt-8 expression in LiCl-dorsalized embryos injected with increasing amounts of BMP-4 mRNA, histone H4 served as a loading control. (B) Control untreated embryos. (C) Phenotype of embryos dorsalized by partial LiCl treatment. (D) Embryos dorsalized by partial LiCl treatment and injected with 0.1 ng of BMP-4 mRNA appear almost normal. (E) Injection of 0.5 ng of BMP-4 capped RNA into embryos treated with partial LiCl results in reversal of the LiCl effect to a ventral phenotype. (F) Further increase in the amount of BMP-4 mRNA injected to 1.0 ng into partial LiCl dorsalized embryos results in strong ventralization.
Fig. 2. Repression of Xwnt-8 by high levels of BMP-4. Embryos were injected dorsally with increasing amounts of BMP-4 mRNA. At stage 10.5 the dorsal part of the marginal zone was cut including the dorsolateral region and expression of Xwnt-8 in this region was studied by RT-PCR. Sibling embryos were allowed to develop to stage 40/41 for phenotypic analysis of the injected RNAs. (A) Schematic description of the experiment. (B) RT-PCR analysis of Xwnt-8 expression in embryos injected with 0.2 or 1 ng of BMP-4 mRNA, histone H4 served as a loading control. (C) Control untreated embryo. (D) Embryo injected with 0.2 ng of BMP-4 RNA exhibiting intermediate ventralization. (E) Injection of 1 ng of BMP-4 mRNA resulted in strong ventralization.
Fig. 3. Spatial response of Xwnt-8 to changes in the levels of BMP-4. Embryos co-injected with LacZ RNA for lineage tracing and increasing amounts of BMP-4 mRNA and dorsalized by full LiCl treatment. At stage 11 the embryos were stained for b-galactosidase activity (red) and the expression of Xwnt-8 determined by in situ hybridization (blue). (A) Control embryo injected only with LacZ mRNA showing the loss of Xwnt-8 as a result of the dorsalization. (B) Injection of 120 pg of BMP-4 RNA results in very low levels of expression of Xwnt-8 at the site of injection. (C) BMP-4 mRNA injected at intermediate levels (240 pg) activates expression of Xwnt-8 at the site of injection. (D) High levels of BMP-4 RNA injected (400 pg) result in the inhibition of Xwnt-8 expression at the site of injection while Xwnt-8 is activated in bordering regions. The red arrowhead marks the site of injection tracer b-galactosidase activity; the black arrowhead points to the Xwnt-8 expression signal.
Fig. 4. Specific regulation of downstream genes by BMP-4 and Xwnt-8. Embryos were injected with Xwnt-8, BMP-4, dominant negative Xwnt-8 and dominant negative BMP receptor RNAs singly or in combination to achieve BMP-4 overexpression when Xwnt-8 signaling was blocked or Xwnt-8 overexpression when BMP signaling was blocked. Under these conditions the effect on myf-5, Xvent-1 and sizzled expression was determined by RT-PCR. The histone H4 gene served as a loading control.
Fig. 5. Expression of sizzled is regulated by BMP-4. Embryos were injected with dominant negative BMP receptor capped RNA or different amounts of BMP-4 mRNA. Expression of the sizzled gene was studied in the treated embryos at stage 11.5 by in situ hybridization. (A) Control embryo exhibiting the normal sizzled expression pattern. (B) Overexpression of the dominant negative BMP receptor which blocks BMP signaling results in a reduction in sizzled expression. (C) Injection of BMP-4 mRNA (0.8 ng) results in the dorsal expansion of sizzled expression. (D) Injection of higher amounts of BMP-4 RNA results in a wide expansion of sizzled expression along the marginal zone.
Fig. 6. Regulation of myf-5 by Xwnt-8 and BMP-4 during late gastrulation. Embryos were injected with mRNAs to manipulate the Xwnt-8 and the BMP-4 signaling pathways and the effects of these manipulations on myf-5 expression were studied. All embryos were co-injected with LacZ RNA to mark the site of injection (red). Following injection, the embryos were dorsalized by LiCl treatment. Injected embryos were fixed at stage 12 and studied by in situ hybridization (blue). (A) Control embryo exhibiting the normal myf-5 pattern of expression. (B) LiCl dorsalization suppresses the expression of myf-5. (C) High levels of BMP-4 overexpression can activate myf-5 expression but the activation is in regions bordering the site of injection. (D) Co-injection of BMP-4 and dominant negative Xwnt-8 encoding mRNAs show that the activation of myf-5 by BMP-4 requires active Xwnt-8 signaling. (E) Injection of 60 ng of an Xwnt-8 expression plasmid resulted in the activation of myf-5 expression at the site of injection. (F) Higher amounts of injected Xwnt-8 expression plasmid (160ng) result in the repression of myf-5 at the site of injection but expression can be observed in neighboring regions. (G,H) Co-injection of the Xwnt-8 expression plasmid together with antisense BMP-4 RNA did not change the effect on myf-5 expression. The red arrowhead marks the site of injection and the black arrowhead marks the in situ hybridization signal.
Fig. 7. BMP-4 functions in the refinement of the Xwnt-8 expression pattern during gastrulation. Embryos were injected with BMP-4 mRNA or RNA encoding the dominant negative BMP receptor. (A) Control embryo injected with LacZ RNA exhibiting the normal Xwnt-8 expression pattern at stage 11. (B) Stage 12.5 embryo exhibiting the late gastrula pattern of Xwnt-8 expression. (C) Ventral overexpression of BMP-4 at stage 11 reduces the Xwnt-8 expression along the ventral marginal zone. This modifie0d pattern resembles the late gastrula Xwnt-8 pattern. (D) Blocking BMP-4 signaling with the dominant negative BMP receptor at stage 12.5 allows Xwnt-8 to expand ventrally to resemble the stage 11 pattern. The red and green lines mark the regions along the marginal zone where the Xwnt-8 expression pattern is modified.