Ectopic expression of T in the paraxial mesoderm disrupts somite maturation in the mouse.

T is the founding member of the T-box family of transcription factors; family members are critical for cell fate decisions and tissue morphogenesis throughout the animal kingdom. T is expressed in the primitive streak and notochord with mouse mutant studies revealing its critical role in mesoderm formation in the primitive streak and notochord integrity. We previously demonstrated that misexpression of Tbx6 in the paraxial and lateral plate mesoderm results in embryos resembling Tbx15 and Tbx18 nulls. This, together with results from in vitro transcriptional assays, suggested that ectopically expressed Tbx6 can compete with endogenously expressed Tbx15 and Tbx18 at the binding sites of target genes. Since T-box proteins share a similar DNA binding domain, we hypothesized that misexpressing T in the paraxial and lateral plate mesoderm would also interfere with the endogenous Tbx15 and Tbx18, causing embryonic phenotypes resembling those seen upon Tbx6 expression in the somites and limbs. Interestingly, ectopic T expression led to distinct embryonic phenotypes, specifically, reduced-sized somites in embryos expressing the highest levels of T, which ultimately affects axis length and neural tube morphogenesis. We further demonstrate that ectopic T leads to ectopic expression of Tbx6 and Mesogenin 1, known targets of T. These results suggests that ectopic T expression contributes to the phenotype by activating its own targets rather than via a straight competition with endogenous T-box factors.

Functionally distinct roles for T and Tbx6 during mouse development.

The mouse T-box transcription factors T and Tbx6 are co-expressed in the primitive streak and have unique domains of expression; T is expressed in the notochord, while Tbx6 is expressed in the presomitic mesoderm. T-box factors are related through a shared DNA binding domain, the T-domain, and can therefore bind to similar DNA sequences at least in vitro We investigated the functional similarities and differences of T and Tbx6 DNA binding and transcriptional activity in vitro and their interaction genetically in vivo We show that at one target, Dll1, the T-domains of T and Tbx6 have different affinities for the binding sites present in the mesoderm enhancer. We further show using in vitro assays that T and Tbx6 differentially affect transcription with Tbx6 activating expression tenfold higher than T, that T and Tbx6 can compete at target gene enhancers, and that this competition requires a functional DNA binding domain. Next, we addressed whether T and Tbx6 can compete in vivo First, we generated embryos that express Tbx6 at greater than wild-type levels embryos and show that these embryos have short tails, resembling the T heterozygous phenotype. Next, using the dominant-negative TWis allele, we show that Tbx6+/- TWis/+ embryos share similarities with embryos homozygous for the Tbx6 hypomorphic allele rib-vertebrae, specifically fusions of several ribs and malformation of some vertebrae. Finally, we tested whether Tbx6 can functionally replace T using a knockin approach, which resulted in severe T null-like phenotypes in chimeric embryos generated with ES cells heterozygous for a Tbx6 knockin at the T locus. Altogether, our results of differences in affinity for DNA binding sites and transcriptional activity for T and Tbx6 provide a potential mechanism for the failure of Tbx6 to functionally replace T and possible competition phenotypes in vivo.

Impaired intermediate formation in mouse embryos expressing reduced levels of Tbx6.

Intermediate mesoderm (IM) is the strip of tissue lying between the paraxial mesoderm (PAM) and the lateral plate mesoderm that gives rise to the kidneys and gonads. Chick fate mapping studies suggest that IM is specified shortly after cells leave the primitive streak and that these cells do not require external signals to express IM-specific genes. Surgical manipulations of the chick embryo, however, revealed that PAM-specific signals are required for IM differentiation into pronephros-the first kidney. Here, we use a genetic approach in mice to examine the dependency of IM on proper PAM formation. In Tbx6 null mutant embryos, which form 7-9 improperly patterned anterior somites, IM formation is severely compromised, while in Tbx6 hypomorphic embryos, where somites form but are improperly patterned along the axis, the impact to IM formation is lessened. These results suggest that IM and its derivatives, the kidneys and the gonads, are directly or indirectly dependent on proper PAM formation. This has implications for humans harboring Tbx6 mutations which are known to have somite-derived defects including congenital scoliosis.

TET-mediated DNA demethylation controls gastrulation by regulating Lefty-Nodal signalling.

Mammalian genomes undergo epigenetic modifications, including cytosine methylation by DNA methyltransferases (DNMTs). Oxidation of 5-methylcytosine by the Ten-eleven translocation (TET) family of dioxygenases can lead to demethylation. Although cytosine methylation has key roles in several processes such as genomic imprinting and X-chromosome inactivation, the functional significance of cytosine methylation and demethylation in mouse embryogenesis remains to be fully determined. Here we show that inactivation of all three Tet genes in mice leads to gastrulation phenotypes, including primitive streak patterning defects in association with impaired maturation of axial mesoderm and failed specification of paraxial mesoderm, mimicking phenotypes in embryos with gain-of-function Nodal signalling. Introduction of a single mutant allele of Nodal in the Tet mutant background partially restored patterning, suggesting that hyperactive Nodal signalling contributes to the gastrulation failure of Tet mutants. Increased Nodal signalling is probably due to diminished expression of the Lefty1 and Lefty2 genes, which encode inhibitors of Nodal signalling. Moreover, reduction in Lefty gene expression is linked to elevated DNA methylation, as both Lefty-Nodal signalling and normal morphogenesis are largely restored in Tet-deficient embryos when the Dnmt3a and Dnmt3b genes are disrupted. Additionally, a point mutation in Tet that specifically abolishes the dioxygenase activity causes similar morphological and molecular abnormalities as the null mutation. Taken together, our results show that TET-mediated oxidation of 5-methylcytosine modulates Lefty-Nodal signalling by promoting demethylation in opposition to methylation by DNMT3A and DNMT3B. These findings reveal a fundamental epigenetic mechanism featuring dynamic DNA methylation and demethylation crucial to regulation of key signalling pathways in early body plan formation.

Autosomal dominant spondylocostal dysostosis is caused by mutation in TBX6.

In humans, congenital spinal defects occur with an incidence of 0.5-1 per 1000 live births. One of the most severe syndromes with such defects is spondylocostal dysostosis (SCD). Over the past decade, the genetic basis of several forms of autosomal recessive SCD cases has been solved with the identification of four causative genes (DLL3, MESP2, LFNG and HES7). Autosomal dominant forms of SCD have also been reported, but to date no genetic etiology has been described for these. Here, we have used exome capture and next-generation sequencing to identify a stoploss mutation in TBX6 that segregates with disease in two generations of one family. We show that this mutation has a deleterious effect on the transcriptional activation activity of the TBX6 protein, likely due to haploinsufficiency. In mouse, Tbx6 is essential for the patterning of the vertebral precursor tissues, somites; thus, mutation of TBX6 is likely to be causative of SCD in this family. This is the first identification of the genetic cause of an autosomal dominant form of SCD, and also demonstrates the potential of exome sequencing to identify genetic causes of dominant diseases even in small families with few affected individuals.

The Wnt3a/ß-catenin target gene Mesogenin1 controls the segmentation clock by activating a Notch signalling program.

Segmentation is an organizing principle of body plans. The segmentation clock, a molecular oscillator best illustrated by the cyclic expression of Notch signalling genes, controls the periodic cleavage of somites from unsegmented presomitic mesoderm during vertebrate segmentation. Wnt3a controls the spatiotemporal expression of cyclic Notch genes; however, the underlying mechanisms remain obscure. Here we show by transcriptional profiling of Wnt3a (-/-) embryos that the bHLH transcription factor, Mesogenin1 (Msgn1), is a direct target gene of Wnt3a. To identify Msgn1 targets, we conducted genome-wide studies of Msgn1 activity in embryonic stem cells. We show that Msgn1 is a major transcriptional activator of a Notch signalling program and synergizes with Notch to trigger clock gene expression. Msgn1 also indirectly regulates cyclic genes in the Fgf and Wnt pathways. Thus, Msgn1 is a central component of a transcriptional cascade that translates a spatial Wnt3a gradient into a temporal pattern of clock gene expression.

Tbx18 and Tbx15 null-like phenotypes in mouse embryos expressing Tbx6 in somitic and lateral plate mesoderm.

Members of the T-box family of transcription factors play essential roles in cell type specification, differentiation, and proliferation during embryonic development. All T-box family members share a common DNA binding domain - the T-domain - and can therefore recognize similar sequences. Consequently, T-box proteins that are co-expressed during development have the potential to compete for binding at downstream targets. In the mouse, Tbx6 is expressed in the primitive streak and presomitic mesoderm, and is sharply down-regulated upon segmentation of the paraxial mesoderm. We sought to determine the phenotypic and molecular consequences of ectopically expressing Tbx6 within the segmented paraxial mesoderm and its derivatives using a 3-component transgenic system. The vertebral column, ribs, and appendicular skeleton were all affected in these embryos, which resembled Tbx18 and Tbx15 null embryos. We hypothesize that these phenotypes result from competition between the ectopically expressed Tbx6 and the endogenously expressed Tbx18 and Tbx15 at the binding sites of target genes. In vitro luciferase transcriptional assays provide further support for this hypothesis.

Kinked tail mutation results in notochord defects in heterozygotes and distal visceral endoderm defects in homozygotes.

Proper formation of the anterior-posterior (AP) axis in the developing embryo is critical for the correct patterning and often survival of the organism. In the mouse, an initial step in axis establishment is the specification and migration of the distal visceral endoderm (DVE). We have identified a semi-dominant spontaneous mutation in mouse, named kinked tail (knk), which when heterozygous results in a kinky tail phenotype due to fusions and dysmorphology of the tail vertebrae. Vertebral fusions appear to be a secondary effect of notochord thickening and branching in the tail region. Homozygosity for knk results in early embryonic lethality by embryonic day 8.5 due to improper timing of DVE specification and migration, and subsequent failure to establish the AP axis.

Generation of transgenic mice expressing Cre recombinase under the control of the Dll1 mesoderm enhancer element.

To study paraxial mesoderm formation in the mouse, transgenic lines that can be used to either selectively delete or express genes of interest in the paraxial mesoderm are required. We have generated a transgenic mouse line that expresses Cre recombinase in the paraxial mesoderm (PAM) beginning at e7.5. A lacZ Cre recombinase reporter line showed that in addition to PAM and its derivatives, lateral plate and intermediate mesoderm derivatives were also exposed to Cre activity, while the node, notochord, and cardiac mesoderm were not. We further demonstrate that 70-75% of the fibroblasts generated from Dll1-msd Cre, ROSA26-rtTA embryos possess Cre recombinase activity. These mice can therefore be used in combination with tet-responsive transgenic lines to generate mesoderm-derived embryonic fibroblasts that inducibly express a gene of interest.

Mesp2 and Tbx6 cooperatively create periodic patterns coupled with the clock machinery during mouse somitogenesis.

The metameric structures in vertebrates are based on the periodicity of the somites that are formed one by one from the anterior end of the presomitic mesoderm (PSM). The timing and spacing of somitogenesis are regulated by the segmentation clock, which is characterized by the oscillation of several signaling pathways in mice. The temporal information needs to be translated into a spatial pattern in the so-called determination front, at which cells become responsive to the clock signal. The transcription factor Mesp2 plays a crucial role in this process, regulating segmental border formation and rostro-caudal patterning. However, the mechanisms regulating the spatially restricted and periodic expression of Mesp2 have remained elusive. Using high-resolution fluorescent in situ hybridization in conjunction with immunohistochemical analyses, we have found a clear link between Mesp2 transcription and the periodic waves of Notch activity. We also find that Mesp2 transcription is spatially defined by Tbx6: Mesp2 transcription and Tbx6 protein initially share an identical anterior border in the PSM, but once translated, Mesp2 protein leads to the suppression of Tbx6 protein expression post-translationally via rapid degradation mediated by the ubiquitin-proteasome pathway. This reciprocal regulation is the spatial mechanism that successively defines the position of the next anterior border of Mesp2. We further show that FGF signaling provides a spatial cue to position the expression domain of Mesp2. Taken together, we conclude that Mesp2 is the final output signal by which the temporal information from the segmentation clock is translated into segmental patterning during mouse somitogenesis.

Dll1 is a downstream target of Tbx6 in the paraxial mesoderm.

Tbx6 is a member of the T-box family of transcription factors. In the mouse, Tbx6 is expressed in the primitive streak, tail bud, and presomitic mesoderm and is essential for the specification of posterior paraxial mesoderm; in its absence, posterior somites are replaced by ectopic neural tubes. Analysis of embryos expressing reduced levels of Tbx6 also revealed that it is required for the correct patterning of the somites as well as their initial specification. As a first step toward identifying downstream targets of Tbx6, we examined the DNA binding properties of Tbx6 and identified a Tbx6 consensus binding site. Previously, we have shown that expression of Dll1, which encodes a Notch ligand, is lost in the Tbx6 mutant and that Tbx6 and Dll1 genetically interact, indicating that Dll1 may be a direct target of Tbx6 in the paraxial mesoderm. We uncovered four putative Tbx6 binding sites within a Dll1 paraxial mesoderm enhancer and show that Tbx6 can bind two of these sites in vitro. Altogether, these results lend further support for Dll1 being a direct target of Tbx6 in the presomitic mesoderm.

Regulation of Tbx6 expression by Notch signaling.

Somites are the first overt sign of segmentation in the vertebrate embryo and form from bilateral strips of paraxial mesoderm. Paraxial mesoderm arises from the primitive streak; it then migrates laterally and comes to lie on both sides of the neural tube. In the mouse, the T-box transcription factor Tbx6 is required for both somite formation and patterning. Tbx6 expression corresponds both temporally and spatially to somite formation, with expression in the primitive streak and presomitic mesoderm. Its expression in the latter could simply be explained by maintenance following its initial activation in the primitive streak. Alternatively, its expression in the presomitic mesoderm may be contributed by separate regulatory elements possibly under the control of different signals. We have begun to investigate how Tbx6 expression is controlled during development using a transgenic approach to identify the cis-acting regulatory regions. We show that it is possible to separate an element required for presomitic mesoderm expression from that driving expression in the primitive streak. Further, we show that a binding site for the Notch transcription factor RBP-Jkappa is necessary for Tbx6 presomitic mesoderm enhancer activity, indicating that Notch signaling is upstream of Tbx6 in the pathway directing somite formation and patterning.

The neural tube patterns vessels developmentally using the VEGF signaling pathway.

Embryonic blood vessels form in a reproducible pattern that interfaces with other embryonic structures and tissues, but the sources and identities of signals that pattern vessels are not well characterized. We hypothesized that the neural tube provides vascular patterning signal(s) that direct formation of the perineural vascular plexus (PNVP) that encompasses the neural tube at mid-gestation. Both surgically placed ectopic neural tubes and ectopic neural tubes engineered genetically were able to recruit a vascular plexus, showing that the neural tube is the source of a vascular patterning signal. In mouse-quail chimeras with the graft separated from the neural tube by a buffer of host cells, graft-derived vascular cells contributed to the PNVP, indicating that the neural tube signal(s) can act at a distance. Murine neural tube vascular endothelial growth factor A (VEGFA) expression was temporally and spatially correlated with PNVP formation, suggesting it is a component of the neural tube signal. A collagen explant model was developed in which presomitic mesoderm explants formed a vascular plexus in the presence of added VEGFA. Co-cultures between presomitic mesoderm and neural tube also supported vascular plexus formation, indicating that the neural tube could replace the requirement for VEGFA. Moreover, a combination of pharmacological and genetic perturbations showed that VEGFA signaling through FLK1 is a required component of the neural tube vascular patterning signal. Thus, the neural tube is the first structure identified as a midline signaling center for embryonic vascular pattern formation in higher vertebrates, and VEGFA is a necessary component of the neural tube vascular patterning signal. These data suggest a model whereby embryonic structures with little or no capacity for angioblast generation act as a nexus for vessel patterning.

Critical role for Tbx6 in mesoderm specification in the mouse embryo.

Tbx6 is a member of the T-box family of transcription factor genes. Two mutant alleles of this gene establish that Tbx6 is involved in both the specification and patterning of the somites along the entire length of the embryo. The null allele, Tbx6(tm1Pa), causes abnormal patterning of the cervical somites and improper specification of more posterior paraxial mesoderm, such that it forms ectopic neural tubes. In this study, we use this allele to further investigate the mechanism of action of the Tbx6 gene and investigate possible genetic interactions. We have tested the developmental and differentiation potential of Tbx6(tm1Pa)/Tbx6(tm1Pa) cells in ectopic sites, in vitro, and in chimeras in vivo. We have also documented cell proliferation and cell death in mutant tail buds in an attempt to explain the mechanism of tail bud enlargement in the Tbx6 mutant embryos. Our results indicate specific developmental restrictions on the differentiation of posterior cells lacking Tbx6, once they have traversed the primitive streak, but no restrictions in differentiation of anterior somites, or of Tbx6 null embryonic stem (ES) cells. We further demonstrate that Tbx6 null ES cells fail to populate posterior somites in chimeric embryos. To discover whether different T-box proteins interact on the same down stream targets in areas of expression overlap, we have explored potential interactions between Tbx6 and T (Brachyury) in genetic crosses. Our results reveal that the T(Wis) mutation is epistatic to the Tbx6(tm1Pa) mutation and that there is no apparent genetic interaction. However, homozygosity for Tbx6(tm1Pa) and heterozygosity for T(Wis) mutation shows a combinatorial interaction at the phenotypic level.

Defective somite patterning in mouse embryos with reduced levels of Tbx6.

During vertebrate embryogenesis, paraxial mesoderm gives rise to somites, which subsequently develop into the dermis, skeletal muscle, ribs and vertebrae of the adult. Mutations that disrupt the patterning of individual somites have dramatic effects on these tissues, including fusions of the ribs and vertebrae. The T-box transcription factor, Tbx6, is expressed in the paraxial mesoderm but is downregulated as somites develop. It is essential for the formation of posterior somites, which are replaced with ectopic neural tubes in Tbx6-null mutant embryos. We show that partial restoration of Tbx6 expression in null mutants rescues somite development, but that rostrocaudal patterning within them is defective, ultimately resulting in rib and vertebral fusions, demonstrating that Tbx6 activity in the paraxial mesoderm is required not simply for somite specification but also for their normal patterning. Somite patterning is dependent upon Notch signaling and we show that Tbx6 genetically interacts with the Notch ligand, delta-like 1 (Dll1). Dll1 expression, which is absent in the Tbx6-null mutant, is restored at reduced levels in the partially rescued mutants, suggesting that Dll1 is a target of Tbx6. We also identify the spontaneous mutation rib-vertebrae as a hypomorphic mutation in Tbx6. The similarity in the phenotypes we describe here and that of some human birth defects, such as spondylocostal dysostosis, raises the possibility that mutations in Tbx6 or components of this pathway may be responsible for these defects.

Mammalian vestigial-like 2, a cofactor of TEF-1 and MEF2 transcription factors that promotes skeletal muscle differentiation.

Expression of many skeletal muscle-specific genes depends on TEF-1 (transcription enhancer factor-1) and MEF2 transcription factors. In Drosophila, the TEF-1 homolog Scalloped interacts with the cofactor Vestigial to drive differentiation of the wing and indirect flight muscles. Here, we identify three mammalian vestigial-like genes, Vgl-1, Vgl-2, and Vgl-3, that share homology in a TEF-1 interaction domain. Vgl-1 and Vgl-3 transcripts are enriched in the placenta, whereas Vgl-2 is expressed in the differentiating somites and branchial arches during embryogenesis and is skeletal muscle-specific in the adult. During muscle differentiation, Vgl-2 mRNA levels increase and Vgl-2 protein translocates from the cytoplasm to the nucleus. In situ hybridization revealed co-expression of Vgl-2 with myogenin in the differentiating muscle of embryonic myotomes but not in newly formed somites prior to muscle differentiation. Like Vgl-1, Vgl-2 interacts with TEF-1. In addition, we show that Vgl-2 interacts with MEF2 in a mammalian two-hybrid assay and that Vgl-2 selectively binds to MEF2 in vitro. Co-expression of Vgl-2 with MEF2 markedly co-activates an MEF2-dependent promoter through its MEF2 element. Overexpression of Vgl-2 in MyoD-transfected 10T(1/2) cells markedly increased myosin heavy chain expression, a marker of terminal muscle differentiation. These results identify Vgl-2 as an important new component of the myogenic program.