Oscillations of the snail genes in the presomitic mesoderm coordinate segmental patterning and morphogenesis in vertebrate somitogenesis

The segmented body plan of vertebrate embryos arises through segmentation of the paraxial mesoderm to form somites. The tight temporal and spatial control underlying this process of somitogenesis is regulated by the segmentation clock and the FGF signaling wavefront. Here, we report the cyclic mRNA expression of Snail 1 and Snail 2 in the mouse and chick presomitic mesoderm (PSM), respectively. Whereas Snail genes' oscillations are independent of NOTCH signaling, we show that they require WNT and FGF signaling. Overexpressing Snail 2 in the chick embryo prevents cyclic Lfng and Meso 1 expression in the PSM and disrupts somite formation. Moreover, cells mis-expressing Snail 2 fail to express Paraxis, remain mesenchymal, and are thereby inhibited from undergoing the epithelialization event that culminates in the formation of the epithelial somite. Thus, Snail genes define a class of cyclic genes that coordinate segmentation and PSM morphogenesis.     

Segmentation in vertebrates: a molecular clock linked to periodic somite formation]

In the vertebrate embryo, somites constitute the basis of the segmental body pattern. They give rise to the axial skeleton, the dermis of the back and all striated muscles of the body. In the chick embryo, a pair of somites buds off, in a highly coordinated fashion, every 90 minutes, from the cranial end of the presomitic mesoderm (PSM) while new mesenchymal cells enter the paraxial mesoderm as a consequence of gastrulation. The processes leading to the segmentation of the somite are not yet understood. We have identified and characterised c-hairy1, an avian homologue of the Drosophila segmentation gene, hairy. c-hairy1 is strongly expressed in the presomitic mesoderm where its mRNA exhibits a cyclic posterior-to-anterior wave of expression whose periodicity corresponds to the formation time of one somite (90 min). Fate mapping of the rostral half of the PSM using the quail-chick chimera technique supports a model of cryptic segmentation within the presomitic mesoderm, and indicates that c-hairy1 expression dynamics are not due to massive cell displacement. Analysis of in vitro cultures of isolated presomitic mesoderm demonstrates that rhythmic c-hairy1 mRNA production and degradation is an autonomous property of the paraxial mesoderm. Rather than resulting from the caudal-to-rostral propagation of an activating signal, it arises from pulses of c-hairy1 expression that are coordinated in time and space. Blocking protein synthesis does not alter the propagation of c-hairy1 expression, indicating that negative autoregulation of c-hairy1 expression is unlikely to control its periodic expression. Most of the segmentation models proposed for somite formation rely on the existence of an internal clock coordinating the cells to segment together to form a somite. These results provide the first molecular evidence of a developmental clock linked to segmentation and somitogenesis of the paraxial mesoderm, and support the possibility that segmentation mechanisms used by invertebrates and vertebrates have been conserved.     

Zic2 and Zic3 synergistically control neurulation and segmentation of paraxial mesoderm in mouse embryo

Zic family zinc-finger proteins play various roles in animal development. In mice, five Zic genes (Zic1-5) have been reported. Despite the partly overlapping expression profiles of these genes, mouse mutants for each Zic show distinct phenotypes. To uncover possible redundant roles, we characterized Zic2/Zic3 compound mutant mice. Zic2 and Zic3 are both expressed in presomitic mesoderm, forming and newly generated somites with differential spatiotemporal accentuation. Mice heterozygous for the hypomorphic Zic2 allele together with null Zic3 allele generally showed severe malformations of the axial skeleton, including asymmetric or rostro-caudally bridged vertebrae, and reduction of the number of caudal vertebral bones, that are not obvious in single mutants. These defects were preceded by perturbed somitic marker expression, and reduced paraxial mesoderm progenitors in the primitive streak. These results suggest that Zic2 and Zic3 cooperatively control the segmentation of paraxial mesoderm at multiple stages. In addition to the segmentation abnormality, the compound mutant also showed neural tube defects that ran the entire rostro-caudal extent (craniorachischisis), suggesting that neurulation is another developmental process where Zic2 and Zic3 have redundant functions.     

BMP signaling in the epiblast is required for proper recruitment of the prospective paraxial mesoderm and development of the somites

Bmpr1a encodes the BMP type IA receptor for bone morphogenetic proteins (BMPs), including 2 and 4. Here, we use mosaic inactivation of Bmpr1a in the epiblast of the mouse embryo (Bmpr-MORE embryos) to assess functions of this gene in mesoderm development. Unlike Bmpr1a-null embryos, which fail to gastrulate, Bmpr-MORE embryos initiate gastrulation, but the recruitment of prospective paraxial mesoderm cells to the primitive streak is delayed. This delay causes a more proximal distribution of cells with paraxial mesoderm character within the primitive streak, resulting in a lateral expansion of somitic mesoderm to form multiple columns. Inhibition of FGF signaling restores the normal timing of recruitment of prospective paraxial mesoderm and partially rescues the development of somites. This suggests that BMP and FGF signaling function antagonistically during paraxial mesoderm development.     

Balancing segmentation and laterality during vertebrate development

Somites are the mesodermal segments of vertebrate embryos that become the vertebral column, skeletal muscle and dermis. Somites arise within the paraxial mesoderm by the periodic, bilaterally symmetric process of somitogenesis. However, specification of left–right asymmetry occurs in close spatial and temporal proximity to somitogenesis and involves some of the same cell signaling pathways that govern segmentation. Here, we review recent evidence that identifies cross-talk between these processes and that demonstrates a role for retinoic acid in maintaining symmetrical somitogenesis by preventing impingement of left–right patterning signals upon the paraxial mesoderm.     

Fate maps of the primitive streak in chick and quail embryo: ingression timing of progenitor cells of each rostro-caudal axial level of somites

Developmental fates of cells emigrating from the primitive streak were traced by a fluorescent dye Dil both in chick and in quail embryos from the fully grown streak stage to 12-somite stage, focusing on the development of mesoderm and especially on the timing of ingression of each level of somitic mesoderm. The fate maps of the chick and quail streak were alike, although the chick streak was longer at all stages examined. The anterior part of the primitive streak predominantly produced somites. The thoracic and the lumbar somites were shown to begin to ingress at the 5 somite-stage and 10 somite-stage in a chick embryo, and 6 somite-stage and 9 somite-stage in a quail embryo, respectively. The posterior part of the streak served mainly as the origin of more lateral or extra embryonic mesoderm. As development proceeded, the fate of the posterior part of the streak changed from the lateral plate mesoderm to the tail bud mesoderm and then to extra embryonic, allantois mesoderm. The fate map of the primitive streak in chick and quail embryo presented here will serve as basic data for studies on mesoderm development with embryo manipulation, especially for transplantation experiments between chick and quail embryos.     

The formation of mesodermal tissues in the mouse embryo during gastrulation and early organogenesis

Orthotopic grafts of [3H]thymidine-labelled cells have been used to demonstrate differences in the normal fate of tissue located adjacent to and in different regions of the primitive streak of 8th day mouse embryos developing in vitro. The posterior streak produces predominantly extraembryonic mesoderm, while the middle portion gives rise to lateral mesoderm and the anterior region generates mostly paraxial mesoderm, gut and notochord. Embryonic ectoderm adjacent to the anterior part of the streak contributes mainly to paraxial mesoderm and neurectoderm. This pattern of colonization is similar to the fate map constructed in primitive-streak-stage chick embryos. Similar grafts between early-somite-stage (9th day) embryos have established that the older primitive streak continues to generate embryonic mesoderm and endoderm, but ceases to make a substantial contribution to extraembryonic mesoderm. Orthotopic grafts and specific labelling of ectodermal cells with wheat germ agglutinin conjugated to colloidal gold (WGA-Au) have been used to analyse the recruitment of cells into the paraxial mesoderm of 8th and 9th day embryos. The continuous addition of primitive-streak-derived cells to the paraxial mesoderm is confirmed and the distribution of labelled cells along the craniocaudal sequence of somites is consistent with some cell mixing occurring within the presomitic mesoderm.     

Interfering with Wnt signalling alters the periodicity of the segmentation clock

Somites are embryonic precursors of the ribs, vertebrae and certain dermis tissue. Somite formation is a periodic process regulated by a molecular clock which drives cyclic expression of a number of clock genes in the presomitic mesoderm. To date the mechanism regulating the period of clock gene oscillations is unknown. Here we show that chick homologues of the Wnt pathway genes that oscillate in mouse do not cycle across the chick presomitic mesoderm. Strikingly we find that modifying Wnt signalling changes the period of Notch driven oscillations in both mouse and chick but these oscillations continue. We propose that the Wnt pathway is a conserved mechanism that is involved in regulating the period of cyclic gene oscillations in the presomitic mesoderm.     

Metameric Pattern Development in the Embryonic Axis of the Mouse I

The overall pattern of the mesoderm in the embryonic axis of the cranial region of mouse embryos was examined with the scanning electron microscope (SEM). A segmental organization was observed first in the paraxial mesodermal wings and midline axis of embryos at the late primitive streak stage. Each segmental unit consists of a somitomere in the paraxial region on each side of an enclosed stretch of midline notochord. Somitomeres appear initially as circular domains of radially arranged cells that swirl about the core center of the unit and are quite similar morphologically to those described recently in chick embryos [12]. Lying in tandem sequence, the segments comprise the chordamesoderm that underlies the neural plate. As additional pairs of somitomeres are added from the primitive streak at the caudal end of the axis, those established in the cranial region remain contiguous and undergo morphogenesis that is coordinate with neurulation. We divide the development of the cranial axis into five phases and associate somitomeres in the mesoderm with neuromeric segmentation in the neural plate. It was found that the first pair of somitomeres comes to underlie the prosencephalon, the second and third pairs underlie the mesencephalon, while the fifth, sixth, and seventh pairs of somitomeres underlie neuromeres of the metencephalon. The eighth pair of somitomeres are the first to separate themselves from the first seven and form the first pair of somites visible at the light microscope level. This study suggests that the cranial axis of the mouse embryo is initially organized into segments like the rest of the body and that subsequent cranial morphology is a consequence of differential development of these segments.     

Vertebrate somitogenesis: a novel paradigm for animal segmentation?

In vertebrates, the primary segmented tissue of the body axis is the paraxial mesoderm, which lies bilaterally to the axial organs, neural tube and notochord. The segmental pattern of the paraxial mesoderm is established during embryogenesis through the production of the somites which are transient embryonic segments giving rise to the vertebrae, the skeletal muscles and the dorsal dermis. Somitogenesis can be subdivided into three major phases (see Fig. 1). First a growth phase during which new paraxial mesoderm cells are produced by a growth zone (epiblast and blastopore margin or primitive streak and later on tail bud) and become organized as two rods of mesenchymal tissue,forming the presomitic mesoderm. Second a patterning phase occuring in the PSM, during which the segmental pattern is established at the molecular level. Third, the somitic boundaries are formed during the morphological segmentation phase. In all vertebrates, all cells of the paraxial mesoderm, during their maturation in the PSM, go successively through these three phases, which are tightly regulated at the spatio-temporal level. The first phase of paraxial mesoderm production falls out of the scope of this review, as it essentially pertains to the gastrulation process. Here, I essentially discuss the segmental patterning phase in vertebrates. Recent data suggest that establishment of the segmental pattern relies on a clock and wavefront mechanism which has been conserved in vertebrates. Furthermore, conservation of this system could extend to invertebrates, suggesting that the clock and wavefront is an ancestral mechanism.     

Researches on the formation of axial organs in the chick embryo. X. Further investigations on the role of ecto- and endoderm in somitogenesis

In continuation of previous experiments, recent results concerning the determinism of somitogenesis are reported. By means of two experimental devices on explanted chick embryos the influence upon mesoderm segmentation of the early removal of ectoderm (neural plate)--before and during the formation of the first somite pairs--and removal of endo- and ectoderm after 36-40 hours of incubation was investigated. Up to date results attest that during the shaping of the first somites no essential epigenetic interrelations between overlying ectoderm (neural plate) and paraxial mesoderm are necessary for segmentation. Just before the onset of segmentation a labile determination seems to be present in the presumptive somitogenic mesoderm. As to the later role of endo- and ectoderm in segmentation, present results reveal a relative independence of the segmentation process. The absence of the above-mentioned layers does not prevent further segmentation but induces a progressive slowing down of the process.     

Evidence for medial/lateral specification and positional information within the presomitic mesoderm.

In the vertebrate embryo, segmentation is built on repetitive structures, named somites, which are formed progressively from the most rostral part of presomitic mesoderm, every 90 minutes in the avian embryo. The discovery of the cyclic expression of several genes, occurring every 90 minutes in each presomitic cell, has shown that there is a molecular clock linked to somitogenesis. We demonstrate that a dynamic expression pattern of the cycling genes is already evident at the level of the prospective presomitic territory. The analysis of this expression pattern, correlated with a quail/chick fate-map, identifies a 'wave' of expression travelling along the future medial/lateral presomitic axis. Further analysis also reveals the existence of a medial/lateral asynchrony of expression at the level of presomitic mesoderm. This work suggests that the molecular clock is providing cellular positional information not only along the anterior/posterior but also along the medial/lateral presomitic axis. Finally, by using an in vitro culture system, we show that the information for morphological somite formation and molecular segmentation is segregated within the medial/lateral presomitic axis. Medial presomitic cells are able to form somites and express segmentation markers in the absence of lateral presomitic cells. By contrast, and surprisingly, lateral presomitic cells that are deprived of their medial counterparts are not able to organise themselves into somites and lose the expression of genes known to be important for vertebrate segmentation, such as Delta-1, Notch-1, paraxis, hairy1, hairy2 and lunatic fringe.     

Proteolipid protein 2 mRNA is expressed in the rabbit embryo during gastrulation

The expression pattern of the receptor tyrosine kinase gene EphB3 was examined during the early stages of chick embryogenesis, and is described in this report. In the gastrula, EphB3 is expressed in epiblast cells adjacent to and entering the anterior portion of the primitive streak; expression is extinguished once cells have ingressed. At headfold stages, EphB3 is strongly transcribed in the floor of the foregut and in anterior lateral endoderm, and is expressed in the subjacent cardiogenic mesoderm. EphB3 is transiently expressed in the lateral ectoderm, neural tube, and neural crest during these stages. Later neural expression is localized to the mesencephalon. In the somitic mesoderm, EphB3 is initially expressed in the sclerotome, but later is expressed predominantly in the dermatome. Prominent expression is also detected in the developing heart, liver, posterior ventral limb bud mesenchyme, pharyngeal arches, and head mesenchyme.     

Chick Hairy1 protein interacts with Sap18, a component of the Sin3/HDAC transcriptional repressor complex

Background  

The vertebrate adult axial skeleton, trunk and limb skeletal muscles and dermis of the back all arise from early embryonic structures called somites. Somites are symmetrically positioned flanking the embryo axial structures (neural tube and notochord) and are periodically formed in a anterior-posterior direction from the presomitic mesoderm. The time required to form a somite pair is constant and species-specific. This extraordinary periodicity is proposed to depend on an underlying somitogenesis molecular clock, firstly evidenced by the cyclic expression of the chick hairy1 gene in the unsegmented presomitic mesoderm with a 90 min periodicity, corresponding to the time required to form a somite pair in the chick embryo. The number of hairy1 oscillations at any given moment is proposed to provide the cell with both temporal and positional information along the embryo's anterior-posterior axis. Nevertheless, how this is accomplished and what biological processes are involved is still unknown. Aiming at understanding the molecular events triggered by the somitogenesis clock Hairy1 protein, we have employed the yeast two-hybrid system to identify Hairy1 interaction partners.     

Specification and maintenance of the spinal cord stem zone

Epiblast cells adjacent to the regressing primitive streak behave as a stem zone that progressively generates the entire spinal cord and also contributes to paraxial mesoderm. Despite this fundamental task, this cell population is poorly characterised, and the tissue interactions and signalling pathways that specify this unique region are unknown. Fibroblast growth factor (FGF) is implicated but it is unclear whether it is sufficient and/or directly required for stem zone specification. It is also not understood how establishment of the stem zone relates to the acquisition of spinal cord identity as indicated by expression of caudal Hox genes. Here, we show that many cells in the chick stem zone express both early neural and mesodermal genes; however, stem zone-specific gene expression can be induced by signals from underlying paraxial mesoderm without concomitant induction of an ambivalent neural/mesodermal cell state. The stem zone is a site of FGF/MAPK signalling and we show that although FGF alone does not mimic paraxial mesoderm signals, it is directly required in epiblast cells for stem zone specification and maintenance. We further demonstrate that caudal Hox gene expression in the stem zone also depends on FGF and that neither stem zone specification nor caudal Hox gene onset requires retinoid signalling. These findings thus support a two step model for spinal cord generation - FGF-dependent establishment of the stem zone in which progressively more caudal Hox genes are expressed, followed by the retinoid-dependent assignment of spinal cord identity.     

Molecular clocks underlying vertebrate embryo segmentation: A 10-year-old hairy-go-round

Segmentation of the vertebrate embryo body is a fundamental developmental process that occurs with strict temporal precision. Temporal control of this process is achieved through molecular segmentation clocks, evidenced by oscillations of gene expression in the unsegmented presomitic mesoderm (PSM, precursor tissue of the axial skeleton) and in the distal limb mesenchyme (limb chondrogenic precursor cells). The first segmentation clock gene, hairy1, was identified in the chick embryo PSM in 1997. Ten years later, chick hairy2 expression unveils a molecular clock operating during limb development. This review revisits vertebrate embryo segmentation with special emphasis on the current knowledge on somitogenesis and limb molecular clocks. A compilation of human congenital disorders that may arise from deregulated embryo clock mechanisms is presented here, in an attempt to reconcile different sources of information regarding vertebrate embryo development. Challenging open questions concerning the somitogenesis clock are presented and discussed, such as When?, Where?, How?, and What for? Hopefully the next decade will be equally rich in answers. Birth Defects Research (Part C) 81:65–83, 2007. © 2007 Wiley‐Liss, Inc.