XB-ART-10392Dev Growth Differ August 1, 2000; 42 (4): 347-57.
The maternal Xenopus beta-catenin signaling pathway, activated by frizzled homologs, induces goosecoid in a cell non-autonomous manner.
In spite of abundant evidence that Wnts play essential roles in embryonic induction and patterning, little is known about the expression or activities of Wnt receptors during embryogenesis. The isolation and expression of two maternal Xenopus frizzled genes, Xfrizzled-1 and Xfrizzled-7, is described. It is also demonstrated that both can activate the Wnt/beta-catenin signaling pathway as monitored by the induction of specific target genes. Activation of the beta-Catenin pathway has previously been shown to be necessary and sufficient for specifying the dorsal axis of Xenopus. beta-Catenin is thought to work through the cell-autonomous induction of the homeobox genes siamois and twin, that in turn bind to and activate the promoter of another homeobox gene, goosecoid. However, it was found that the beta-catenin pathway regulated the expression of both endogenous goosecoid, and a goosecoid promoter construct, in a cell non-autonomous manner. These data demonstrate that maternal Frizzleds can activate the Wnt/beta-catenin pathway in Xenopus embryos, and that induction of a known downstream gene can occur in a cell non-autonomous manner.
PubMed ID: 10969734
Article link: Dev Growth Differ
Genes referenced: ctnnb1 dvl2 eef1a2 en2 fzd1 fzd7 gsc h4c4 myc nodal3.1 nog otx2 sia1 sia2 snai2 tbx2 twist1 wnt5a wnt8a
Article Images: [+] show captions
|Fig. 1. Dendrogram and sequence alignment of Xfrizzled-1 and Xfrizzled-7. (A) Conceptual translations of the maternal Xenopus frizzled coding regions were compared to published Frizzled proteins using a ClustalW alignment (MacVector 6.5; R, Rattus; M, murine; H, human; X, Xenopus; G, Gallus) and found to cluster with Frizzled-1 and Frizzled-7, respectively. (B) Xfrizzled-1 and Xfrizzled-7 proteins share 76% identity and 85% similarity.|
|Fig. 2. Developmental expression of Xfrizzled-1 and Xfrizzled-7. A reverse transcription–polymerase chain reaction (RT-PCR) analysis was used to assay relative levels of Xfrizzled-1 and Xfrizzled-7 expression at various stages of embryogenesis. (A) Xfrizzled-1 is abundant in unfertilized eggs (UF) but is expressed at very low levels during gastrulation (10, 12). Zygotic Xfrizzled-1 expression is detected at neurula stages and later in embryogenesis. (B) Xfrizzled-7 is expressed throughout embryogenesis. Two Xfrizzled-7 polymerase chain reaction (PCR) primer pairs using a common forward primer and distinct reverse primers yielded similar results. Histone H4 serves as a control for cDNA synthesis and gel loading. The -RT control demonstrates that contaminating genomic DNA is not present in the samples analyzed.|
|Fig. 3. Xfrizzled-1 and Xfrizzled-7 are expressed in overlapping and distinct regions of Xenopus embryos. All embryos are oriented with anterior to the left and dorsal to the top. Embryos in (A,G) were cleared in Murray’s clearing agent. (A) Xfrizzled-7 transcripts are broadly expressed in late gastrulae (stage 13), and are intensely expressed in migrating cranial neural crest of neurula-stage embryos (B, arrow). (C) A control tadpole treated with a sense probe exhibits no staining. (D) Xfrizzled-7 is expressed in the brain (arrowhead), neural crest (nc), ventral otic vesicle (ov), and pronephros (P) of tailbud-stage embryos. (E) Xfrizzled-7 transcripts are present in the brain (arrowheads), heart (h), and pronephros (P) of tadpoles. Xfrizzled-1 transcripts are first detected in neurula stages in the brain (data not shown), and brain expression domains are maintained in later embryos (F,G, arrowheads). (F,G) Xfrizzled-1 is expressed in the otic vesicle (ov, unlabeled arrow in (G)), but in a pattern complementary to Xfrizzled-7. Like Xfrizzled-7, Xfrizzled-1 is expressed in the pronephros (P) and heart (h) of tailbud (F) and tadpolestage (G) embryos, as well as the proctodeum (pd) of tadpoles (F).|
|Fig. 4. Xfrizzled-1 and Xfrizzled-7 activate the Wnt/b-catenin pathway. (A,B) Overexpression of Xfrizzled-1 or Xfrizzled-7 in animal cap explants can induce expression of the Wnt/bcatenin- responsive genes Xnr-3 and siamois. Whole-embryo total ribonucleic acid (RNA) serves as a positive control in each experiment (left lane in each panel). Animal caps from uninjected embryos express neither of the b-catenin-responsive genes (center lanes), but overexpression of either Xfrizzled-1 (A, right lane) or Xfrizzled-7 (B, right lane) results in expression of both Xnr-3 and siamois. Histone H4 and elongation factor (EF)1a serve as loading controls. (C). Xfrizzled-7 induces neural crest markers in neuralized animal caps. Embryos were injected with RNA encoding Noggin and a control (green fluorescent protein) RNA or Xfrizzled-7 RNA. Late blastula (stage 9) animal caps were cultured until sibling embryos reached stage 18–20. Total RNA was prepared and was subjected to reverse transcription–polymerase chain reaction (RT-PCR) analysis. While embryos express all markers used, uninjected control explants express only the ubiquitous EF1a. Noggin-expressing explants express the anterior neural marker otx-2, while coexpression of Xfrizzled-7 results in the induction of a more posterior marker, engrailed-2. These caps also express the neural crest markers slug and twist.|
|Fig. 5. Xfrizzled-1 and Xfrizzled-7 recruit Xwnt-8 and Dishevelled but not Xwnt-5a to the plasma membrane. The indicated ribonucleic acid (RNA) was overexpressed in animal caps and the ectopic Wnt and Dishevelled proteins were analyzed by confocal microscopy. (A,D,F) Ectopic Wnt proteins are detected at the plasma membrane and in the secretory apparatus in animal caps. (B,C) Coexpression of Xfrizzled-1 or Xfrizzled-7 results in a recruitment of Xwnt-8 to the plasma membrane. (E) Xwnt-5A is not recruited by Xfrizzled-1. (G) In an independent experiment, Xfrizzled-7 does not recruit Xwnt-5A from the secretory apparatus to the plasma membrane. (H) Overexpressed Dishevelled-GFP localizes to punctate structures throughout the cytoplasm but is recruited to the plasma membrane by Xfrizzled-1 (I) and Xfrizzled-7 (J).|
|Fig. 6. Cell non-autonomous induction of endogenous goosecoid by b-catenin. Ribonucleic acid was injected into ventral vegetal cells or regions at the 4–32-cell stages, encoding b-catenin mutated to accumulate in nuclei and tagged with a c-myc epitope. Embryos at the gastrula stage were then processed for whole-mount in situ hybridization to localize goosecoid transcripts (blue–purple) and the expression of the c-myc-tagged b-catenin was visualized by immunostaining (brown). (A–C) The in situ hybridizations reveal a normal endogenous domain of goosecoid expression (E) as well as an ectopic site (arrow). This ectopic goosecoid expression extends several cell diameters away from the nearest cells expressing b-catenin (b), indicating that b-catenin can induce goosecoid in a cell non-autonomous manner. (D) A fixed embryo was cut in half prior to analysis, revealing that expression of b-catenin deep within the embryo (b) induces goosecoid (arrow) in the overlaying cells in a cell non-autonomous manner.|
|Fig. 7. Cell non-autonomous induction of a goosecoid reporter construct by b-catenin signaling. (A) The goosecoid-1500gsc/LUC Luciferase construct (Watabe et al. 1995) was injected into the marginal zone of a single ventral (V) or dorsal (D) cell at the 4-cell stage. b-Catenin was then injected into the same cell (Vb) or the neighboring lateral cell in some of the ventral-injected embryos (V/b) and Luciferase activity was determined at the gastrula stage. (B) The goosecoid-1500gsc/LUC Luciferase construct was injected into a ventral vegetal (V) or dorsal (D) cell at the 16–32-cell stage, and where indicated (V//b) embryos injected ventrally with the reporter construct were then injected with b-catenin two cells away laterally. This left one uninjected cell between the two injected cells. (C) The siamois S01234 Luciferase construct (Brannon et al. 1997) was injected into ventral (V) or dorsal (D) cells as in (A). Where indicated ventral injection of the siamois reporter construct was followed by injection of b-catenin into the adjacent ventral cell (V/b). This level of b-catenin is sufficient to induce this promoter when coexpressed in the same cell (data not shown). Relative light units are normalized to that of the dorsal side for each experiment, and two to three sets of embryos were used for each condition in each experiment. Data are also representative of multiple independent experiments.|