Search

Embryogenesis relies on the precise interplay of signaling cascades to activate tissue-specific differentiation programs. An important player in these morphogenetic processes is β-catenin, which is a central component of adherens junctions and canonical Wnt signaling. Lack of β-catenin is lethal before gastrulation, but mice heterozygous for β-catenin (Ctnnb1) develop as wild type. Here, we confine β-catenin amounts below the heterozygous expression level to study the functional consequences for development. We generate embryonic stem (ES) cells and embryos expressing β-catenin only from the ubiquitously active ROSA26 promoter and thereby limit β-catenin expression to ~12.5% (ROSA26β/+) or ~25% (ROSA26β/β) of wild-type levels. ROSA26β/+ is sufficient to maintain ES cell morphology and pluripotent characteristics, but is insufficient to activate canonical target genes upon Wnt stimulation. This Wnt signaling deficiency is incompletely restored in ROSA26β/β ES cells. We conclude that even very low β-catenin levels are able to sustain cell adhesion, but not Wnt signaling. During development, ROSA26β/β as well as ROSA26β/+ partially rescues the knockout phenotype, yet proper gastrulation is absent. These embryos differentiate according to the neural default hypothesis, indicating that gastrulation depends on high β-catenin levels. Strikingly, if ROSA26β/+ or ROSA26β/β is first activated after gastrulation, subsequent development correlates with the dosage of β-catenin. Moreover, molecular evidence indicates that the amount of β-catenin controls the induction of specific Wnt target genes. In conclusion, by restricting its expression we determine the level of β-catenin required for adhesion or pluripotency and during different morphogenetic events.

We conditionally substituted E-cadherin (E-cad; cadherin 1) with N-cadherin (N-cad; cadherin 2) during intestine development by generating mice in which an Ncad cDNA was knocked into the Ecad locus. Mutant mice were born, demonstrating that N-cad can structurally replace E-cad and establish proper organ architecture. After birth, mutant mice gradually developed a mutant phenotype in both the small and large intestine and died at ~2-3 weeks of age, probably due to malnutrition during the transition to solid food. Molecular analysis revealed an extended domain of cells from the crypt into the villus region, with nuclear localization of β-catenin (β-cat; Ctnnb1) and enhanced expression of several β-cat target genes. In addition, the BMP signaling pathway was suppressed in the intestinal epithelium of the villi, suggesting that N-cad might interfere with BMP signaling in the intestinal epithelial cell layer. Interestingly, mutant mice developed severe dysplasia and clusters of cells with neoplastic features scattered along the crypt-villus axis in the small and large intestine. Our experimental model indicates that, in the absence of E-cad, the sole expression of N-cad in an epithelial environment is sufficient to induce neoplastic transformations.

Differential gene regulation integrated in time and space drives developmental programs during embryogenesis. To understand how the program of gastrulation is regulated by Wnt/β-catenin signaling, we have used genome-wide expression profiling of conditional β-catenin mutant embryos. Known Wnt/β-catenin target genes, known components of other signaling pathways, as well as a number of uncharacterized genes were downregulated in these mutants. To further narrow down the set of differentially expressed genes, we used whole-mount in situ screening to associate gene expression with putative domains of Wnt activity. Several potential novel target genes were identified by this means and two, Grsf1 and Fragilis2, were functionally analyzed by RNA interference (RNAi) in completely embryonic stem (ES) cell-derived embryos. We show that the gene encoding the RNA-binding factor Grsf1 is important for axial elongation, mid/hindbrain development and axial mesoderm specification, and that Fragilis2, encoding a transmembrane protein, regulates epithelialization of the somites and paraxial mesoderm formation. Intriguingly, the knock-down phenotypes recapitulate several aspects of Wnt pathway mutants, suggesting that these genes are components of the downstream Wnt response. This functional genomic approach allows the rapid identification of functionally important components of embryonic development from large datasets of putative targets.

Cadherin-mediated cell-cell adhesion plays important roles in mouse embryonic development, and changes in cadherin expression are often linked to morphogenetic events. For proper embryonic development and organ formation,the expression of E-cadherin must be tightly regulated. Dysregulated expression during tumorigenesis confers invasiveness and metastasis. Except for the E-box motifs in the E-cadherin promoter, little is known about the existence and location of cis-regulatory elements controlling E-cadherin gene expression. We have examined putative cis-regulatory elements in the E-cadherin gene and we show a pivotal role for intron 2 in activating transcription. Upon deleting the genomic intron 2 entirely, the E-cadherin locus becomes completely inactive in embryonic stem cells and during early embryonic development. Later in development, from E11.5 onwards, the locus is activated only weakly in the absence of intron 2 sequences. We demonstrate that in differentiated epithelia, intron 2 sequences are required both to initiate transcriptional activation and additionally to maintain E-cadherin expression. Detailed analysis also revealed that expression in the yolk sac is intron 2 independent, whereas expression in the lens and the salivary glands absolutely relies on cis-regulatory sequences of intron 2. Taken together, our findings reveal a complex mechanism of gene regulation, with a vital role for the large intron 2.

Many components of the Wnt/beta-catenin signaling pathway are expressed during mouse pre-implantation embryo development, suggesting that this pathway may control cell proliferation and differentiation at this time. We find no evidence for a functional activity of this pathway in cleavage-stage embryos using the Wnt-reporter line, BAT-gal. To further probe the activity of this pathway, we activated beta-catenin signaling by mating a zona pellucida3-cre (Zp3-cre) transgenic mouse line with a mouse line containing an exon3-floxed beta-catenin allele. The result is expression of a stabilized form of beta-catenin, resistant to degradation by the GSK3beta-mediated proteasome pathway, expressed in the developing oocyte and in each cell of the resulting embryos. Nuclear localization and signaling function of beta-catenin were not observed in cleavage-stage embryos derived from these oocytes. These results indicate that in pre-implantation embryos, molecular mechanisms independent of the GSK3beta-mediated ubiquitination and proteasome degradation pathway inhibit the nuclear function of beta-catenin. Although the mutant blastocysts initially developed normally, they then exhibited a specific phenotype in the embryonic ectoderm layer of early post-implantation embryos. We show a nuclear function of beta-catenin in the mutant epiblast that leads to activation of Wnt/beta-catenin target genes. As a consequence, cells of the embryonic ectoderm change their fate, resulting in a premature epithelial-mesenchymal transition.

β-Catenin is a central component of both the cadherin- catenin cell adhesion complex and the Wnt signaling pathway. We have investigated the role of β-catenin during brain morphogenesis, by specifically inactivating the β- catenin gene in the region of Wnt1 expression. To achieve this, mice with a conditional (‘floxed’) allele of β-catenin with required exons flanked by loxP recombination sequences were intercrossed with transgenic mice that expressed Cre recombinase under control of Wnt1 regulatory sequences. β-catenin gene deletion resulted in dramatic brain malformation and failure of craniofacial development. Absence of part of the midbrain and all of the cerebellum is reminiscent of the conventional Wnt1 knockout (Wnt1−/−), suggesting that Wnt1 acts through β- catenin in controlling midbrain-hindbrain development. The craniofacial phenotype, not observed in embryos that lack Wnt1, indicates a role for β-catenin in the fate of neural crest cells. Analysis of neural tube explants shows that β-catenin is efficiently deleted in migrating neural crest cell precursors. This, together with an increased apoptosis in cells migrating to the cranial ganglia and in areas of prechondrogenic condensations, suggests that removal of β-catenin affects neural crest cell survival and/or differentiation. Our results demonstrate the pivotal role of β-catenin in morphogenetic processes during brain and craniofacial development.

Molecular analysis of the cadherin-catenin complex elucidated the central role of β-catenin in this adhesion complex, as it binds to the cytoplasmic domain of E-cadherin and to α-catenin. β-catenin may also function in signalling pathways, given its homology to the gene product of the Drosophila segment polarity gene armadillo, which is known to be involved in the wingless signalling cascade. To study the function of β-catenin during mouse development, gene knock-out experiments were performed in embryonic stem cells and transgenic mice were generated. β-catenin null-mutant embryos formed blastocysts, implanted and developed into egg-cylinder-stage embryos. At day 7 post coitum, the development of the embryonic ectoderm was affected in mutant embryos. Cells detached from the ectodermal cell layer and were dispersed into the proamniotic cavity. No mesoderm formation was observed in mutant embryos. The development of extraembryonic structures appeared less dramatically or not at all affected. Our results demonstrate that, although β-catenin is expressed rather ubiquitously, it is specifically required in the ectodermal cell layer.

Monoclonal antibodies (mAbs) against a preparation of intermediate filaments from trophoblastoma cells were studied for their reactivity pattern during embryonic development and on adult tissue cells. Up to day 12 of embryonic development, epithelial cells of the three germ layers reacted with these mAbs. Later during development and in adult tissues, positive reactions could be observed only with epithelial cells derived from mesoderm and endoderm. Because of their tissue distribution, the proteins reacting with these mAbs might belong to the keratin family of intermediate filaments or they might represent a new group of intermediate filaments.