June 1, 2001;
Difference in XTcf-3 dependency accounts for change in response to beta-catenin-mediated Wnt signalling in Xenopus blastula.
Wnt signalling functions in many tissues and during different stages of animal development to produce very specific responses. In early Xenopus embryos there is a dramatic change in response to Wnt signalling within only a few hours of development. Wnt signalling in very early embryos leads to a dorsalising response, which establishes the endogenous dorsal axis. Only a few hours later in development, almost the opposite happens: Xwnt-8
functions to pattern the embryonic mesoderm
by promoting ventral
and lateral mesoderm
. The specificity of the response could conceivably be carried out by differential use of different signal transduction pathways, many of which have recently been described. We have found, however, that this dramatic shift in response to Wnt signalling in early Xenopus is not brought about by differential use of distinct signal transduction pathways. In fact beta-catenin, a downstream component of the canonical Wnt signal transduction pathway, functions not only in the early dorsalising response but also in the later ventrolateral-promoting response. Interaction of beta-catenin with the XTcf-3
transcription factor is required for the early dorsalising activity. In contrast, our experiments suggest that late Wnt signalling in the ventrolateral mesoderm
does not require a similar dependency of beta-catenin function on XTcf-3
. Our results highlight the potential versatility of the canonical Wnt pathway to interact with tissue
-specific factors downstream of beta-catenin, in order to achieve tissue
[+] show captions
Fig. 2. Lithium-mediated inhibition of GSK3β in late blastula shows involvement in ventrolateral-promoting Wnt signalling. Xenopus embryos analysed for marker gene expression with whole-mount RNA in situ hybridisation carried out at stage 11. Control embryos (A-D) were treated with NaCl and experimental embryos (E-H) were treated with LiCl at stage 9.5. In all embryos dorsal is towards the top. Control embryos (A-D) show normal molecular marker expression. The molecular markers used were chordin (Chd) and Xnot, which are dorsal specific, and XmyoD and Xpo, which mark ventral and lateral. Chordin acts as a control as it has previously been shown that it is unaffected by late blastula Wnt signalling. The expression of chordin therefore remains unchanged in the dorsal side of the embryo by lithium treatment (compare E with A). The expression of the dorsal marker Xnot is greatly reduced after inhibition of GSK3β by lithium (F) when compared with the control embryo (B). Expression of the ventral and lateral markers XmyoD (G) and Xpo (H) are expanded into the dorsal midline of the embryo after lithium treatment whereas in control embryos (C,D) no expression is seen in the dorsal midline.
Fig. 3. Effects of lithium treatment carried out at a series of successive stages in Xenopus embryonic development. Stages indicate the stage at which lithium treatment was carried out (stage 6.5-10). Whole-mount RNA in situ hybridisation was carried out on all embryos at gastrulation stage 11. In all embryos dorsal is towards the top. The molecular markers used were chordin (Chd) and Xnot, which are dorsal specific, and XmyoD and Xpo, which mark ventral and lateral. Treatment with lithium at early blastula stages (A,G,M,S) results in dorsalisation of the embryo with expansion of Chd and Xnot into the ventral and lateral sides of the embryo and reduction or loss, respectively of the ventral and lateral markers Xpo and XmyoD. Treatment with lithium at early gastrula stage (F,L,R,X) results in ventralisation of the embryo with loss or reduction of the dorsal marker Xnot and expansion of the ventral and lateral markers XmyoD and Xpo into the dorsal midline. Chordin acts as a control during lithium treatment of late blastula stages as it has previously been shown that chordin is unaffected by late blastula Wnt signalling (see also Fig. 2E). Therefore the expression of chordin remains unchanged in the dorsal side of the embryo. Analysis of molecular markers shows the change of response to Wnt signalling from early dorsal-promoting to later ventrolateral-promoting is different for different Wnt responsive genes. The dorsal markers Chd and Xnot show ectopic expression in early to mid-blastula stages of development (A-C,G-I) with normal expression when lithium treated at stage 9.5 for Chd (E) and reduced expression at stage 10 for Xnot (L). Lithium treatment at stage 7.5 causes a reduction in the expression of XmyoD dorsolaterally (N). Treatment at stages 9-10 results in ventrolateral promoting response with ectopic expression in the dorsal side of the embryo (P-R). Treatment with lithium, even at the very early stages of development, results in the expansion of expression of the ventrolateral marker Xpo into the dorsal midline (T-X).
Fig. 4. Overexpression of β-catenin shows involvement in late blastula Wnt signalling. GFP, β-catenin and marker gene expression in gastrulation stage Xenopus embryos. Using a double construct (hsp::β-catenin::hsp::GFP) to generate transgenic embryos, both β-catenin and GFP can be overexpressed to high levels (see Materials and Methods). This increase in levels can be seen in the western blot carried out with stage 10 embryos (A). GFP positive embryos (1-4) also have high levels of exogenous β-catenin protein. In non-transgenic embryos (5-8) only low levels of GFP and exogenous β-catenin protein is detectable, which is probably due to cytoplasmic expression from unincorporated plasmid DNA. (B-K) Detection of ubiquitous GFP expression under UV light (G) therefore allows transgenic embryos overexpressing β-catenin (G-K) to be selected from control embryos (B-F). (G) GFP expressing embryo under UV-light (compare with non-GFP-expressing embryo in B). (H) Morphological phenotype of an embryo overexpressing β-catenin (compare with control embryo in C and to Xwnt-8 overexpressing embryo in Fig. 7G). The embryos in D-F show normal molecular marker expression after whole-mount RNA in situ hybridisation is carried out at stage 11 and embryos I-K show molecular marker expression in β-catenin overexpressing embryos. In all embryos dorsal is at the top. Expression of Xnot, a dorsal molecular marker (D) is repressed in β-catenin overexpressing embryos (I), whereas the ventral and lateral mesodermal markers XmyoD and Xpo, are ectopically expressed in the dorsal midline of these β-catenin overexpressing embryos (compare J and K with E and F). (L-Q) As a control, transgenic embryos using a single construct (hsp::GFP) to overexpress only GFP (O,Q) develop a normal morphology (P), as compared to non-transgenic siblings (L-N).
Fig. 5. N-Tcf-3 inhibits dorsalising but not ventrolateral-promoting Wnt signalling when expressed in Xenopus embryos. (A) N-Tcf-3 construct used in this experiment. The morphology and molecular marker expression of uninjected control embryos are shown (B-F). N-Tcf-3 RNA was injected into the dorsal marginal zone (DMZ, G-K) or the ventral marginal zone (VMZ, L-P) at the two-cell stage and embryos were analysed by morphology at stage 31 (G,L) and molecular marker expression at stage 11 (H-K,M-P). Dorsally injected embryos have a strong ventralised phenotype with reduced dorsal structures (G). Analysis of molecular markers show that there is a repression of dorsal markers (chordin, H) and Xnot (I) (compared with C,D) and ectopic expression of ventral and lateral markers XmyoD (J) and Xpo (K) in the dorsal midline (compared with E,F). Ventrally injected embryos are unaffected by injection of N-Tcf-3 RNA and show the same morphology and molecular marker expression as the control embryos (compare L-P with B-F). (Q) Western blot detecting HA-tagged N-XTcf-3 protein at stages 6 and 10 of development after injection of the RNA at the four-cell stage into the dorsal marginal zone (DMZ) or the ventral marginal zone (VMZ). As a control, uninjected embryos show no N-XTcf-3 protein present; however, injected embryos show comparable levels of protein at stages 6 and 10 with both ventral and dorsal injection of N-XTcf-3 RNA.
Fig. 6. N-XTcf-3 inhibits dorsalising and ventrolateral-promoting Wnt signalling when expressed in Xenopus embryos. (A) N-XTcf-3 construct used in this experiment. The morphology and molecular marker expression of uninjected control embryos are shown (B-F). N-XTcf-3 RNA was injected into the dorsal marginal zone (DMZ; G-K) or the ventral marginal zone (VMZ; L-P) at the two-cell stage and embryos were analysed by morphology at stage 31 (G and L) and molecular marker expression at stage 11 (H-K,M-P). Dorsally injected embryos have a ventralised phenotype with reduced dorsal mesodermal structures (G). Analysis of molecular markers show that there is a repression of the dorsal marker (chordin, H; compare with C) and a reduction of Xnot (I; compare with D). Ectopic expression of the ventral and lateral markers XmyoD (J) and Xpo (K) in the dorsal midline is also seen (compared with E,F). Ventrally injected embryos are affected by injection of N-XTcf-3 with normal dorsal mesodermal structures but reduced ventral structures. This can be seen both morphologically (L compared with B) and with molecular markers (M-P compared with C-F). The expression of the dorsal marker Chd is comparable with the control expression (M,C) and Xnot, a notochord specific dorsal marker shows a slight expansion (N) into the ventral cells, which express N-XTcf-3 (compare with normal expression pattern, D). Reduced ventral development is also shown by the reduction of ventrolateral gene expression of XmyoD and Xpo in the ventral cells which express N-XTcf-3 (O,P compared with E,F). (Q) Western blot detecting HA-tagged N-XTcf-3 protein at stages 6 and 10 of development after injection of the RNA at the four-cell stage into the DMZ or the VMZ. As a control, uninjected embryos show no N-XTcf-3 protein present, however, injected embryos show comparable levels of protein at stages 6 and 10 with both ventral and dorsal injection of N-XTcf-3 RNA.