Somite formation and patterning

Estelle!-&singer,Caroline Jouve, Julien Dubrulle, and OlivierPourquie Laboiatoire de GBnCtique et de Physiologie du D&eloppement (LGPD), Developmental Biology Institute of Marseille (IBDM), CNRS-INSERMUniversiti: de la Mkditerranke-AP de Marseille, Campus de Luminy, Case 907, 13288 Marseille Cedex 09, France

As a consequenceof their segmented arrangement and the diversityof their tissue derivatives,somites are key elements in the establishmentof the metameric body plan in vertebrates.This article aims to largely reviewwhat is known about somite development, from the initial stages of somite formation through the processof somite regionalization along the three major body axes. The role of both cell intrinsicmechanismsand environmentalcues are evaluated. The periodicand bilaterallysynchronousnature of somite formation is proposed to rely on the existence of a developmentalclock. Molecular mechanismsunderlyingthese events are reported. The importance of an antero-posterior somiticpolarity with respect to somite formation on one hand and body segmentation on the other hand is discussed.Finally,the mechanismsleading to the regionalizationof somites along the dorso-ventraland medio-lateralaxes are reviewed.This somitic compartmentalizationis believed to underlie the segregation of dermis, skeleton, and dorsal and appendicular musculature. KEY WORDS: Somite, Chick embryo, Muscle,Vertebra, Dermis, Segmentation, Somitogenesis,Signaling: 0 2000AcademicPRESS.

I. introduction

In vertebrates, all skeletal muscles of the trunk arise from the transient embryonic somites. In amniotes, these structures first appear as epithelial spheres, which provide the early framework on which the segmental pattern of the body is established. Somite formation occurs sequentially and synchronously on both sides of the neural tube with each epithelial sphere pinching off from the anterior extremity of the caudal unsegmented paraxial mesoderm (Fig. 1). Somites form with a constant periodicity until the International Review of Cytology, Vol. I98 cKt74.7696/00 $35.00

1

Copyright 0 2000 lay Academic Press AU rights of reproduction in any form resewed.

HIRSINGER

‘\

‘.

ET AL,

Presomitic mescderm

FIG. 1 Schematic representation of paraxial mesoderm differentiation. Semitic material is produced by ingression of material from the epiblast through the anterior primitive streak and later on from the tail bud. These cells become organized as two stripes of mesenchymal PSM which subsequently form epithelial spheres called somites. This process called somite segmentation occurs in a coordinated fashion on both sides of the midline organs (neural tube and notochord). Somites become subsequently polarized into a ventral mesenchymal compartment called the sclerotome which will give rise to the axial skeleton and a dorsal epithelial dermomyotome. This latter compartment will yield all the skeletal muscles of the body and the dermis of the back. It will later subdivide into a myotome which comes to lie

SOMITOGENESIS

3

definitive number, characteristic of the species, is reached (Richardson et al., 1998). In amniotes, somites give rise to the vertebral column, ribs, all the skeletal muscles of the trunk and limbs, and the dermis of the back; they also contribute endothelial cells, smooth muscles, and connective tissues such as the meninges. In this review, we will summarize our current knowledge of the development of somites in the chick ‘embryo and compare this to what is known in the mouse embryo. Due to the major advantages of using avian embryos as a model system to study embryonic development, such as their accessibility at early stages, much of our understanding of somitogenesis has come from analyses of this organism. Genetic studies in mice greatly contributed to the analysis of this mechanism at the molecular level. We will discuss our current comprehension of somitogenesis from the establishment of the somitic territory during gastrulation to the specification of the various somitic lineages. Somite regionalization along the antero-posterior body axis and the role of the homeotic genes in that process is not discussed here; refer to Gossler and Hrabe de Angelis (1998) for a recent review on the subject. In the chick embryo, the somitic material becomes specified during gastrulation when a lateral territory of the epiblast ingresses into the primitive streak. This invaginated material forms the presumptive territory of the paraxial mesoderm, which thereafter resides in the rostra1 primitive streak. During regression of the primitive streak and tail bud elongation, paraxial mesoderm migrates through the node and laterally and comes to lie on both sides of the neuraxis. The first somite forms from the anterior end of the presomitic mesoderm (PSM) after 24 hr of incubation at stage I-III 7 (Hamburger, 1992). During chick somitogenesis, which lasts up to embryonic day 5, approximately 52 somites are formed. Somites follow a rostrocaudal gradient of differentiation, the anterior-most somites being the oldest ones. Therefore, at different embryonic ages, the differentiation stage of a somite can be measured with respect to the time elapsed since its segmentation. This led Ordahl(l993) to propose a somite staging system reflecting this intrinsic state of differentiation, independent of the embryonic stage. According to this system, the most recently formed somite is called somite I, the second most recently formed is somite II, and so on. This nomenclature will be extensively used in this review.

between the sclerotome and the dermatome. The myotome provides all the epaxial muscles. dm, dermomyotome; m, myotome; nc, notochord; np, neural plate; nt: neural tube; psm, PSM, SC,somitocoele; scl, sclerotome; wd, wolffian duct. The level of somite I and V (SI and SV) is indicated.

4

HIRSINGERET AL.

The primary segmentation of the vertebrate embryo displayed by somitic organization underlies much of the segmental organization of the body. The primary segments, the somites, can be further subdivided into anterior and posterior compartments, which exhibit different properties with respect to neural crest cell and motor axon migration through the somite. This somitic subdivision is thus responsible for peripheral nervous system segmentation. The embryonic vasculatory system initiates development during sornitogenesis. While the dorsal aorta elongates caudally, branches arise laterally and bifurcate dorsally each passing through an intersomitic cleft. Thus, the segmental pattern of blood vessels is also largely imposed by the somites (Christ et al., 1998). Recently, new insights have been gained about the molecular mechanisms involved in somite segmentation and the formation of rostra1 and caudal somite compartments. Implications of the Notch signaling pathway in these processes and the role of a newly identified molecular clock linked to segmentation will be discussed. In the chick and mouse embryo, the first somite lies immediately caudal to the otic vesicle (Huang et al., 1997) and participates in the formation of the occipital bone of the skull. Anterior to this somite, the paraxial mesoderm is referred to as head or cephalic mesoderm. This anterior tissue will contribute to skeletal muscles and bones of the head. The issue of whether this tissue is segmented is still a matter of controversy. In this review, we will focus strictly on somites. Somites bud off from the anterior segmental plate as epithelial balls surrounding a cavity, the somitocoele, containing mesenchymal cells. By somite stage III-IV, the dorsal portion of the somite remains epithelial and constitutes the dermomyotome while its ventral moiety undergoes an epithelio-mesenchymal transition leading to the formation of the sclerotome. The sclerotome, which also contains the somitocoele cells, gives rise to the skeletal elements. By stage X, these cells migrate either ventrally toward the notochord to form vertebral bodies and intervertebral discs, or dorsally to constitute neural arches while at the thoracic level they also form the proximal ribs. The distal ribs have been shown to derive from the dermomyotome (Kato and Aoyama, 1998). The dermomyotome predominantly contains dermal precursors, which contribute to the dermis of the back, and the skeletal muscle precursors. By stage VII, the myotome, from which paraspinal muscles arise, starts to migrate from the dermomyotomal medial edge and comes to lie between the dermomyotome and the sclerotome. Laterally at the interlimb level, a lateral myotome forms in mirror image to the medial edge and gives rise to abdominal and intercostal muscles. At the wing level from a stage VII somite and at the leg level from a stage I somite, cells disperse from the dermomyotome lateral edge, migrate through lateral mesoderm and settle in limb mesenchyme where they coalesce to form premuscular masses and

SOlvllTOGENESlS

5

further differentiate as appendicular muscles. Connective tissues of the trunk muscles also derive from somites, whereas those of the limb muscles along with their skeletal elements arise from the lateral plate. Endothelial cells arise from all the compartments of the somite to contribute to vessel formation throughout the trunk region (Wilting et al., 1995). By stage XV; the dermomyotome undergoes a global epithelio-mesenchymal transition, releasing muscle precursors ventrally and precursors of the back dermis dorsally, which migrate farther dorsally to reach their final position under the epidermis. The other regions of dermis derive from lateral mesoderm. Thus, the epithelial somite, which initially appears as a homogenous structure, undergoes regionalized morphogenetic transformations leading to the formation of a great variety of tissues. How is this diversity of ceili types generated within the epithelial somite? After discussing the means by which paraxial mesoderm is specified and subsequently segments into somites, we will review the studies of how the somites are molecularly patterned along the three classical axes (i.e., the antero-posterior, the dorsoventral, and the medio-lateral axes). The intersections of these axes detine compartments from which different cell types arise. Cell autonomous mechanisms controlling cell fate decisions will also be discussed.

II. Origin

and Commitment

A. Origin of Semitic

of the Semitic

Cells in the Gastrulating

Cells Embryo

Various lineage tracing techniques have been used to follow cell movement and to fate map presumptive territories in the avian blastoderm (Nicolet, 1971; Ooi et al, 1986; Psychoyos and Stern, 1996a; Sprat& 1955). These include labeling cells with vital dyes or with fluorescent or histochemical markers or constructing quail-chick chimeras. Using such techniques in the chick embryo, the progenitors of the paraxial mesoderm have been localized in mid and definitive primitive streak stages (stages 2-4 HH) to the epiblast lateral to the midline and subsequently to the anterior primitive streak. When the primitive streak regresses, it lays down in its wake, on both sides of the axis, a stream of mesenchymal PSM cells which will subsequently segment into somites (Hatada and Stern, 1994; Psychoyos and Stern, 1996a; Schoenwolf et aZ., 1992a; Selleck and Stern, 1991). When the primitive streak has totally regressed, its remnant occupies a small zone at the caudal end of the embryo which forms the tail bud. This structure will provide the remainder of the paraxial mesoderm cells (Catala et al, 1995). A similar distribution of the somite presumptive territory is also found in the mouse embryo where progenitors of the somites have been

6

HIRSINGER

ET AL.

localized to the rostra1 primitive streak and then to the tail bud (Tam and Beddington, 1987, 1992). After complete ingression of the epiblastic material around stage 4 in chick, the presumptive cells of the somites are found in the anterior part of the primitive streak (Psychoyos and Stern, 1996a; Schoenwolf et al., 1992a; Selleck and Stern, 1991). They reside in this structure and later in the tail bud as a population of stem cells giving rise to the progenitor cells from which all the somites arise (Nicolas and Bonnerot, 1988; Nicolas et aZ., 1996; Stem, 1992). The antero-posterior organization of the embryonic mesodermal territories of the primitive streak corresponds to the future medio-lateral organization of the embryo. Thus the most rostra1 territory corresponds to the notochord, followed more caudally by the somites and then by the intermediate mesoderm, lateral plate and extra-embryonic mesoderm (Psychoyos and Stern, 1996a; Schoenwolf et al., 1992b). This order also seems to apply within a defined territory because, in the case of somites, the most rostra1 part of the streak contains the medial somitic progenitors whereas lateral somitic cells are localized in more caudal aspects of the rostral streak (Selleck and Stern, 1991). In the primitive streak, there is no evident prepattern reflecting the future distribution of the somitic cells along the antero-posterior axis, at any defined time point. However, comparing cells at equivalent levels of the primitive streak at different ages shows that they are fated to populate progressively more caudal somites as the age of the embryo increases. It is therefore unlikely that the antero-posterior distribution of the somites reflects by an early spatial arrangement of progenitor cells in the primitive streak.

B. Specification of Precursors Paraxial Mesoderm

Cells to Form

Paraxial mesoderm precursors in the epiblast and the primitive streak show extensive developmental plasticity and are not stably committed to a definitive fate. In chick, the prospective paraxial mesoderm territory in the epiblast can differentiate into neural tissue when ectopically grafted into the prospective neurectoderm (Garcia-Martinez et al., 1997). Similarly, when anterior streak tissue is transplanted to the posterior streak level, it differentiates into extra-embryonic or lateral plate mesoderm indicating that somitic precursor cells in the anterior primitive streak are also not committed (Beddington, 1982; Garcia-Martinez and Schoenwolf, 1992). These anterior primitive streak cells can also contribute to the neural plate when transplanted to the prospective neuro-ectoderm region of the epiblast (GarciaMartinez et al., 1997) and will differentiate into notochord when grafted

SOMlTOGENESlS

7

into Hensen’s node of the mid-streak state embryo (Selleck and Stern, 1992). Even more surprisingly, complete ablation of Hensen’s node and rostra1 primitive streak including the whole presumptive somitic territory leads to the development of a normal embryo when performed at stage 3+/4- in the chick embryo (Psychoyos and Stern, 1996b). These experiments demonstrate the lack of commitment of both the epiblast and rostra1 primitive streak to form paraxial mesoderm. Even in the PSM and caudal somites, cells are not stably committed to a somitic fate. Quail caudal somites, when grafted under the ectoderm of a primitive streak stage embryo, form mainly mesoderm but also, to a lesser extent, endoderm, lateral plate, and endothelium (Veini and Bellairs, 1991). In addition, cell lineage studies using injection of a fluorescent tracer into cells of the caudal part of the PSM show that these cells can contribute both to somites and to intermediate or lateral mesoderm (Stern et al., 1988). It has also been shown that PSM can be converted to lateral plate by treatment with BMP4 (Tonegawa et al., 1997; Tonegawa and Takahashi, 1998). At the PSM level, paraxial mesoderm identity seems to correspond to an unstable state which is maintained by a balance between opposing signals such as BMP4 and its antagonists Noggin or Chordin, expressed on either side of the PSM. Whereas the fate of cells ingressing at different levels of the primitive streak are well described in mouse and chick embryos, the molecular mechanisms that specify these fates, and in particular those responsible for the commitment of these cells to form paraxial mesoderm, are poorly understood. In chick, Hensen’s node appears to be dispensable for paraxial mesoderm specification because its ablation, at least up to stage 4-, does not affect somitogenesis (Psychoyos and Stern, 1996b). In addition, mouse embryos lacking the node and notochord do form somites (Ang and Rossant, 1994; Weinstein et al., 1994). Mutational analyses in mice have begun to elucidate the role of signalling molecules during paraxial mesoderm specification. The phenotype of FGFRl-deficient mice indicates a potential role for FGF signaling in specifying paraxial mesoderm cell fates. FGFRl-deficient embryos exhibit an expansion of the axial mesoderm, primitive streak defects, and an absence of somites (Yamaguchi et aZ., 1994). The loss of somites seems to be a secondary consequence of the expansion of the axial mesoderm at tbe expense of paraxial mesoderm. In embryos lacking another signaling molecule, Wnt3a, somitogenesis is disrupted after the formation of the first 5-6 somites (Takada et al., 1994). A similar phenotype has been observed in mutant embryos lacking Brachyury (Wilson ef al., 1995). In the Tbx6 mutant, trunk somites are transformed into ectopic neural tubes flanking the axis, resulting in a mouse with three neural tubes (Chapman and Papaioannou, 1998). These molecules therefore appear to be involved in segregating

8

HIRSINGERET AL.

the presumptive embryonic territories corresponding to axial mesoderm, paraxial mesoderm, and neural plate. In all these mutants, it seems that the first few somites that are formed are spared from the effects of the mutation. A similar observation has also been made in most of the mutants of the Notch-delta signaling pathway in which only the first few somites form. This suggests that the activity of these genes is required specifically for the formation and segmentation of the more posterior somitic mesoderm, indicating that formation of the anterior somites could have different requirements. Although a few genes have been characterized as potential regulators of paraxial mesoderm identity, thus far, no gene has been found to be required specifically for the specification of the paraxial mesoderm along the entire body axis.

III. Somite

Formation

A. A Prepattern of Segmentation Paraxial Mesoderm?

in the

Once paraxial mesoderm has been formed, the next issue is to understand how the mesenchymal PSM tissue becomes transformed into epithelial spheres that form iteratively and synchronously on both sides of the embryo. Meier has proposed that the PSM displays a metameric arrangement of groups of cells called somitomeres. These transient structures are visible only by stereo scanning electron microscopy (Meier, 1984), and their existence remains controversial (Wachtler et al., 1982b). According to Meier and colleagues, these groups of mesenchymal cells are organized around a central point and are formed in a strictly anterior to posterior order. The number of somitomeres in the PSM varies between species but remains constant within a given species (12 in the chicken and 6 in the mouse), even though the length of the PSM changes during development (Tam and Beddington, 1986). The number of somitomeres was shown to correspond to the number of prospective somites of the PSM, thus reflecting an early spatial allocation of these cells along the antero-posterior axis (Jacobson, 1988; Tam and Beddington, 1986). Lineage analysis of chick or mouse PSM cells suggests that somitomeres are not acting as clonal compartments like the rhombomeres (Fraser et al, 1990). Clonal descendants of single cells or groups of cells are often found to cross the boundary between somitomeres (Bagnall, 1992; Bagnall et al., 1992; Stern et aZ., 1988; Tam, 1988; Tam and Beddington, 1987). Furthermore, when a portion of chick segmental plate is replaced by a piece of quail segmental plate of an equivalent size, which is known to contain more

SOMITOGENESIS

9

prospective somites, the number of somites formed by the quail transplant is equal to the number of chick prospective somites that have been removed and not the same as the number of quail prospective somites that have been transplanted (Packard et aZ., 1993). Therefore, if a prepattern of the PSM exists, it is not entirely fixed and is subject to regulating processes until it undergoes segmentation. Moreover, genetic expression patterns offer little support for the somitomere concept since no gene has been observed to be expressed in a somitomeric fashion except in the most rostra1 PSM. Therefore it seems that apart from the original arguments based on scanning electron microscope images, little evidence has been provided to confirm the existence of somitomeres in the PSM. Nonetheless, because little cell movement is observed in the PSM, the distribution of cells in this tissue, to some extent, reflects a prepattern corresponding to their future relative position in the somites along the antero-posterior axis.

B. Tissue Interactions

Involved in Somite Formation

Interactions between the paraxial mesoderm and surrounding tissues, in particular the axial structures, were postulated to be linked directly to the process of somite formation (Fraser, 1960; Keynes and Stern, 1988). It was observed that the PSM can segment in isolation of the neural tube, the notochord, or the endoderm (Packard and Jacobson, 1976; Sandor and Fazakas-Todea, 1980; Tam and Beddington, 1986). However, these explants failed to form somites in the absence of ectoderm (Ladher et aZ., 1996; Lash and Yamada, 1986; Palmeirim et al, 1998), although surprisingly, they displayed segmental properties as evidenced by delta1 mRNA expression (Palmeirim et aZ., 1998). In addition, ablation of the ectoderm in vivo prevents somite formation in the operated region (Sosie et al., 1997). Therefore, establishment of the segmental pattern in the PSM appears to be an intrinsic property of this tissue, although formation of the epithelial somite requires signaling from the ectoderm.

C. Cell-Cell and Cell-Matrix During Somitogenesis

Interactions

Cells destined to form somites undergo profound changes in their adhesive properties during somitogenesis. These cell-cell and cell-matrix interactions are essential to somite formation, as was demonstrated by disrupting these interactions. Specific cell-cell adhesion is dependent not only on the presence of tissue-specific cell adhesion molecules (CAMS) but also on the

IO

HIRSINGERET AL.

spatio-temporal regulation of CAM types. In the chick paraxial mesoderm, N-cadherin, cadherin-11, and NCAM expression are up-regulated during differentiation (Duband et&, 1987; Kimura et aZ.,1995). Enhanced cadherin expression may be related to the increase in cell adhesivity observed during PSM maturation and somite formation (Bellairs et al., 1978). Accordingly, the epithelial structure of the newly formed somites can be disrupted by in vitro treatment with antibodies against N-cadherin (Duband et al., 1987). A null mutation of N-cadherin led to small and irregular somites whose epithelial structure was partially disrupted (Radice et aZ., 1997). In contrast, the loss of function mutation in the NCAM gene had no effect on somite formation (Cremer et aZ., 1994; Duband et aZ., 1987). In mouse and chick embryos, differentiation of the PSM is accompanied by the elaboration of matrix materials (Duband et aZ., 1987; Newgreen, 1984; Ostrovsky et aZ., 1988). In the mouse, the amount of fibronectin in the mesenchyme and at the interface with adjacent epithelia increases during the maturation of the PSM. Interestingly, somites do not form in mouse embryos mutant for the fibronectin gene (George et aZ., 1993). Similar phenotypes are observed in mice mutant for focal adhesion kinase (FAK), a nonreceptor tyrosine kinase implicated in transducing signals generated by cell-matrix interactions (Furuta et al., 1995). Thus, the dynamic expression patterns of adhesion molecules, along with experiments disrupting the interactions they mediate, suggest that cell-cell and cellmatrix interactions play an important role during somitogenesis.

D. Molecular 1. A Molecular

Aspects

of Somitogenesis

Clock Linked to Somitogenesis

Several theoretical models have been proposed to account for somitogenesis, a process characterized by its rythmicity and bilateral synchrony. Some of these models are based on cell communication and implicate adhesion molecules or Notch-delta mediated lateral inhibition as the driving force of somite formation (Conlon et aZ., 1995). But these models do not account for the periodicity of this process. Others such as the “clock and wave front” model (Cooke and Zeeman, 1976) Meinhardt’s model (Meinhardt, 1986) or the cell cycle model (Stern et al., 1988) proposed the existence of an oscillator or clock in the presomitic cells. The purpose of such an oscillator was to generate a temporal periodicity, which would be translated into the spatial periodicity of the somites. Recently, the identification of c-hairyl, an avian homolog of the fly pair rule gene, has provided molecular support for the existence of such a clock linked to segmentation (Palmeirim et aZ., 1997). This gene is strongly expressed in the PSM where its mRNA

Ii

SOMITOGENESIS

exhibits cyclic waves of expression whose temporal periodicity corresponds to the time required for the formation of one somite (Palmeirim et al., 1997). In vitro studies have demonstrated that these waves result from an intrinsic property of the PSM and do not rely on cell migration or extrinsic signal propagation. Prospective somitic cells express pulses of c-haluyl mRNA as soon as they leave the site of gastrulation and enter the PSM. Because the PSM contains 12 prospective somites in the chick, cells will undergo 12 c-hairy1 expression cycles before their incorporation into a somite. In zebrafish, a homolog of hairy, herl, has been identified (Muller et al., 1996). This gene presents no dynamic PSM expression as observed in chick, but it is expressed in alternate primordia of the presumptive somites and thus follows a pair-rule expression in the PSM. her-l expression constitutes the only evidence for a pair-rule type of mechanism in vertebrates. This result raises the possibility that part of the machinery involved in the segmentation process has been conserved between Drosophila and zebrafish. Interestingly, lunatic fringe has been recently identified in the chick and mouse as another gene expressed in a cyclical fashion in the PSM (Forsberg et al., 1998; McGrew et al., 1998). lunatic fringe mRNA is expressed in a rhythmic fashion in the PSM, with a periodicity corresponding ta the formation of one somite. When protein synthesis is blocked, the lunatic fvinge wave is abolished, in the contrast to c-hairyl, suggesting that lunatic fringe could act dowstream of c-hairyl. It appears that c-hairy1 and lunatic fringe are two genes regulated by the clock linked to somitogenesis. The role of the lunatic fringe gene has been investigated in the mouse by mutational analysis. Mutant embryos exhibit defects in somite formation, demonstrating a crucial role for this gene during somitogenesis (Evrard et al., 1998; Zhang and Gridley, 1998). In Drosqphila, fringe has been shown to play a role in modulating Notch-delta signaling, resulting in the establishment of the dorso-ventral boundary of the wing margin (Fleming et al., 1997; Panin et al., 1997). As described later, the Notch-delta signaling pathway is indeed involved in defining somite boundaries. 2. Role of the Notch Signal@

Pathway

Some major contributions to the understanding of the molecular mechanisms underlying somitogenesis have come from the use of gene knockout experiments in the mouse. Surprisingly enough, the study of the homologs of fly neurogenic genes in vertebrates revealed that homologs of these genes implicated in the Notch-delta pathway are expressed in the PSM, including the receptors Notch1 (Conlon et al., 1995; Swiatek et al., 1994) and Notch2 (Lindsell et al., 1996; Weinmaster et al., 1992); the ligands

12

HIRSINGER

ET AL.

Delta1 (Bettenhausen et aZ., 1995) Delta3 (Dunwoodie et al., 1997), and Serrate/Jagged1 (Lindsell et al., 1996; Mitsiadis et aZ., 1998); downstream effecters such as the suppressor of hairless homolog RBP-jK (Oka et aZ., 1995) and the enhancer of split homolog HESS (de la Pompa et aZ., 1997) as well as other genes involved in this pathway such as Presenilinl (Wong et aZ., 1997) and lunatic $kinge (Evrard et aZ., 1998; Forsberg et al, 1998; Johnston et aZ., 1997; McGrew et aZ., 1998; Zhang and Gridley, 1998). Inactivation of most of these genes leads to a strong disruption of somitic segmentation. In Notch1 mutant mice, somite formation occurs but its bilateral coordination is affected (Conlon et al., 1995). A similar phenotype is observed in mutants for Delta1 and Delta3, where somites are present but not fully epithelialized (Hrabe de Angelis et aZ., 1997; Kusumi et al., 1998). Mutation of the mouse homologs of other genes involved in Notch signaling such as RBP-jK or Presenilinl also generates somitogenesis defects (Oka et aZ., 1995; Wong et aZ., 1997). RBP-jK is the vertebrate homolog of the fly suppressor of hairless gene which codes for a transcription factor acting downstream of Notch (Fortini and Artavanis-Tsakonas, 1994; Jarriault et al., 1995). Mice embryos mutant for RBP-jK have irregularly shaped somites, as in the Notch1 and Delta1 mutants, however the phenotype is more severe than in the Notch1 mutant, suggesting some degree of Notch redundancy during embryogenesis. This might be expected because other Notch genes are similarly expressed in the PSM. The Presenilinl gene codes for a multipass transmembrane protein whose C. elegans homolog has been shown to act in the Notch-delta pathway (Rohan de Silva and Patel, 1997). Mice mutant for Presenilinl exhibit somitogenesis defects similar to those observed for the other members of the Notch pathway (Wong et aZ., 1997). The importance of the Notch-delta pathway during somitogenesis is further supported by the analysis of Xenopus X-Delta2. Injection of RNA coding for dominant negative form of X-Delta2, as well as ectopic expression of X-Delta2 in Xenopus embryos leads to perturbations of segmentation and, in the more severe cases, abolishes the segmental pattern (Jen et aZ., 1997). It is apparent that any disruption of the activity of the Notch signaling pathway leads to abnormal somitogenesis without the loss of segments. Therefore, the Notch-delta pathway does not seem to play a crucial role in the establishment of the basic metameric pattern, but rather seems to coordinate and fine-tune somite formation. In Drosophila, this pathway is implicated, along with other systems, in the process of lateral inhibition, which controls cell fate choice in the nervous system and in the process of boundary definition in the establishment of the wing margin (Blair, 1997; Kimble and Simpson, 1997). In vertebrates the role of this pathway during somitogenesis might be more related to this latter function, as evidenced by the phenotype of lunatic fringe mutant mice (Evrard et al., 1998; Zhang and Gridley, 1998). In homozygous null mice, somites form but exhibit

SOMITOGENESIS

13

irregularities in size and shape. Loss of lur&c fringe function also leads, in the rostra1 PSM, to loss of the sharply demarcated domains of expression of several genes encoding members of the Notch-Delta signaling. Thus, a possible explanation for the lunatic fvl’nge phenotype is that the loss of Notch-delta signaling may alter the positioning of the somite boundaries and the allocation of cells to the nascent somite. From these observations, it is likely that the Notch-delta signaling pathway acts downstream of the clock linked to somitogenesis and that ItlnaGc fringe is the molecular link between this clock and the Notch-delta pathway. Thus, clock control of the local modulation of the Notch signaling pathway could confer the periodic arrangement of boundaries that underly the segmental body plan.

3. Other Players in Somitogenesis

A significant role for bHLH transcription factors other than hairy during somitogenesis has been demonstrated. The related transcription factors Mespl and 2 in the mouse, Thylacinel in the frog, and c-mesol in the chick are expressed in the PSM as a stripe in its rostra1 part (Buchberger et al., 1998; Saga et al., 1996,1997; Sparrow et al., 1998). A mutation in the Mesp2 gene leads to defects in somitogenesis, particularly at the level of the caudal somites (Saga et al, 1997). In these mutants, expression of Notch1 and 2 as well as FGFRl is strongly down-regulated in the PSM, suggesting that Mesp2 acts upstream of these factors. Accordingly, overexpression studies of the frog Mesp homolog, Thylacinel, indicate that it is part of the Notch signaling cascade important for somite segmentation. Another bHLH transcription factor, paraxis, is expressed in the rostra1 PSM and in the newly formed epithelial somite (Blanar et aZ., 1995; Burgess et dl., 1995). Mutation of this gene in the mouse leads to a loss of somite formation, whereas the segmented arrangement of the paraxial mesoderm derivatives is maintained (Burgess et al., 1996). Furthermore, blocking expression of p,araxis, using antisense oligonucleotides in the chick, leads to an impairment of son-rite epithelialization (Barnes et aZ., 1997). In the chick embryo, paraxis expression correlates with the epithelialization of the PSM, and it was shown that its expression is under the control of factors derived from the ectoderm, which is necessary for somite epithelialization (Gallera, 1966; Palmeirim et al., 1998; Sosic et al., 1997). However,pam.xis expression is not sufficient for epithelization because its expression is maintained in cultured isolated PSM explants, which fail to form epithelial somites (Psimeirim et al, 1998). These results show that the segmentation process of mesodermal derivatives can be dissociated from the formation of the epithelial somite.

14 IV. Antero-posterior of the Somites

HIRSINGER

ET AL.

Compartmentalization

As discussed in the next section, somite formation might also necessitate the juxtaposition of anterior and posterior compartments within the future somite. Later on, this compartmentalization along the antero-posterior axis is clearly visible in the sclerotome whose anterior and posterior halves exhibit distinct cellular and molecular properties.

A. Anatomical and Biochemical the Sclerotome

Compartmentalization

of

During somite maturation, the ventral part of the somite de-epithelializes to form the sclerotome, a mesenchymal tissue, which contains the precursors of the axial skeleton. The polarization of the sclerotome into rostra1 and caudal compartments, which differ in their permissivity to neural crest cell and motor axon migration, has been well studied and has given rise to the notion of antero-posterior (AP) compartmentalization of the somite. Histological analyses have shown that the two sclerotome halves are morphologically separated by a cleft called the Von Ebner’s fissure. This fissure separates the rostra1 from the caudal sclerotome. Both compartments differ in their cell density which is higher in the caudal half (Stern and Keynes, 1987; Von Ebner, 1888). Thus, the paraxial mesoderm becomes histologically segmented by two kinds of boundaries-the intersomitic clefts and the intrasomitic clefts. Rostra1 and caudal sclerotome halves also differ in the biochemical composition of their extracellular matrix. The caudal moiety contains versican, a chondroitin sulfate proteoglycan (Landolt et al., 199.5) as well as collagen IX (Ring et aZ., 1996), and reacts with fluorescent Peanut agglutinin (PNA), whereas the rostra1 moiety is devoid of these molecules (Bagnall and Sanders, 1989). Several cell adhesion molecules are also differentially expressed in the sclerotome. For example, T-cadherin is specifically present in the caudal half of the somite (Ranscht and Bronner-Fraser, 1991). Thus, these morphological, biochemical, and molecular differences between the two parts of the sclerotome are thought to play an important role in the control of migration of both neural crest cells and motor axons. In vitro studies have implicated adhesion molecules, such as T-cadherin or collagen IX, in inhibiting both axon and neural crest cell migration (Fredette et al., 1996; Ring et al., 1996). More recently, Eph family ligands and receptors have been shown to display differential AP sclerotomal expression and are implicated in the

15

SOMITOGENESIS

control of axon guidance in the paraxial mesoderm. EphB3, an Eph receptor tyrosine kinase, localizes to the rostra1 half of the sclerotome and is also expressed by neural crest cells migrating through this territory, whereas the Iigand ephrin Bl is restricted to the caudal half (Krull et aZ., 1997) where neural crest cells do not migrate. Similarly, Wang and Anderson (1997) have demonstrated the presence of the Eph-related receptors TUJl in both migrating neural crest cells and motor axons, whereas Lerk-2, an Eph Iigand, is detected in regions devoid of peripheral nervous system derivatives such as the caudal half of the sclerotome and the dermomyotome. Neural crest cells are unable to migrate on substrates containing Eph family Iigands such as HtkL or Lerk-2, suggesting that these molecules provide repulsive signals to the migrating neural crest cells (Krull et al., 1997; Wang and Anderson, 1997). Thus, neural crest cell migration and motor axon outgrowth in the rostra1 part of the sclerotome is likely to be due to a repulsive activity of molecules specifically expressed in the caudal half of the sclerotome. These Eph-related molecules are also likely to play a role in defming or maintaining the integrity of compartments or compartment boundaries. In the hindbrain, cek8, an Eph-related receptor, is expressed in rhombomeres 3 and 5. In retinoic acid-treated embryos, the posterior morphological boundaries of rhombomeres are lost, and this phenomenon is preceded by the down-regulation of cek8 in rhombomere 5 (Nittenberg et a& 1997). A similar role for the Eph-related family of molecules in the segregation of the rostraI and caudal somitic compartments and in maintainance of the intrasomitic boundary was recently demonstrated in zebrafish (Durbin et al., 1998).

6. Sclerotomal

Fate and the Resegmentation

Theory

Subdivision of the sclerotome into rostra1 and caudal compartments preempts the future contribution of the sclerotomal cells to the vertebral column (see for a review Christ and Wilting, 1992). Cell lineage analyses of various parts of the somite, including the caudal and rostra1 halves have been performed using the quail-chick chimaera system. By orthotopical replacement of one cervical somite, it has been p’ossible to follow the contribution of a single embryonic somite to the vertebral column (BagnaIl et al., 1998). Cells derived from a single somite colonize a delimited region which comprises one half of each of two adjacent vertebrae as well as the intervening disc (Bagnall et al, 1988; Huang et al., ,1996). Moreover, grafting trains of multiple caudal or rostra1 halves has shown that each part, of the sclerotome contributes to defined vertebral structures. In multiple caudal half-son&e grafts, the pedicle of the vertebral arch is abnormally extended

16

HIRSINGER

ET AL.

and the intervertebral disc almost disappears. In contrast, in multiple rostra1 half-somite replacements, the pedicle is almost absent and the intervertebral disc is present (Goldstein and Kalcheim, 1992). These results suggest that a vertebra is formed from the caudal parts of one somite pair and the rostra1 parts of the next pair. This phenomenon, which was first described by Remak (1850) and later reinforced by von Ebner (1888) during the last century, led to the theory of resegmentation: the boundaries of the definitive adult segments do not correspond to the embryonic metameric units because they are shifted by half a segment. Thus, the initial segments, the somites could represent “parasegments,” a situation which is similar to that described in the fly for body segments (Lawrence, 1992). This concept is however still controversial because conflicting results have been obtained (Baur, 1969; Blechschmidt, 1957; Christ et al., 1979; Tajbakhsh and Sporle, 1998; Verbout, 1985).

C. Establishment 1. Intrinsic

of Antero-posterior

Determination:

Compartmentalization

Cellular and Molecular

Aspects

In contrast to the dorso-ventral or medio-lateral polarity of the somite, AP polarity is already established when the somite forms (Aoyama and Asamoto, 1988). Experimental evidence has demonstrated that this polarity is not determined by surrounding cues provided by the environment of the somite. For instance, inversion of the AP axis of the PSM leads to an inversion of the progression of segmentation from caudal to rostral. The somites formed from the inverted PSM present a reversed AP polarity as evidenced by neural crest cells which now migrate through the caudal part of the somite (former rostral) or by the inversion of the c-delta1 expression pattern (Bronner-Fraser and Stern, 1991; Palmeirim et al., 1998). All these data indicate that AP compartmentalization is determined within the PSM. Several genes have been identified; they are restricted to either the caudal or rostra1 part of the newly formed somite (i.e., before the sclerotome has formed, such that this rostrocaudal compartmentalization is already apparent by the epithelial stage). Most of the genes belonging to the Notchdelta pathway are expressed in such a restricted pattern (Bettenhausen et al., 1995; Dunwoodie et al., 1997; Henrique et al., 1995; Reaume et al., 1992). In the mouse, both Notch1 and Notch2, as well as Delta1 and DeZta3, are expressed in the PSM and subsequently in either the rostra1 or the caudal compartments of the newly formed somites (Bettenhausen et al, 1995; de1 Amo et al., 1992; Dunwoodie et al., 1997; Reaume et al., 1992; Saga et al., 1996; Williams et aZ., 1995). In the Delta1 mutant, no epithelial somite forms in the caudal region of the embryo. The caudal identity of

I?

SOMITOGENESIS

the sclerotome is lost, but the primary metameric pattern is established as indicated by the segmental arrangement of myotomes. A similar phenotype altering rostro-caudal compartmentalization is also observed in the mouse lunatic fringe mutant (Evrard et al., 1998; Zhang and Gridley, 1998). AI1 these data suggest that the Notch-delta pathway is involved in specifying the rostro-caudal identity of the somites, as well as in maintaining segments borders, as mentioned earlier (Hrabe de Angelis et al., 1997). Similar to the mouse, chick c-d&z1 is expressed in the PSM and is then restricted to the caudal part of the epithelial somite. PSM cultured in v&o without any surrounding tissue has been shown to display a segmental expression of this gene (Palmeirim et al., 1998). Alternate stripes of Cdelta1 appear sequentially whereas no epithelialization is observed. Thus, epithelialization and segmental patterning defined as the rostro-caudal compartmentalization of gene expression in the rostra1 PSM are two independent events that can be uncoupled in vitro.

2. Relationship Between Antero-posterior and Somite Formation

Polarity

The following results suggest that somitic AP compartmentalization might be a prerequisite for somite formation. When multiple caudal or rostra1 halves of epithelial somites are placed adjacent to each other, scleratomal cells derived from like half-somites mix, whereas juxtaposition of cells from unlike half-somites leads to a strict segregation of each population (i.e., rostra1 vs caudal) (Stern and Keynes, 1987). This suggests that formation and maintenance of segment borders could depend on the juxtaposition of two cell-states, such as A and P as proposed in the Meinhardt model (Meinhardt, 1986). In this model, cells of the PSM would oscillate between two different states and would eventually undergo stable acquisition of one of these states. Juxtaposition of cells in one state with cells in the other creates a border, similar to the mechanism for formation of parasegmental boundaries in Drosophila. However, a system of two-segment periodicity must be postulated and superimposed on this two-state model to enable border formation only between P and A of different somites and not between A and P of the same somite. Such a system could require the activity of genes similar to the pair-rule genes in Drosophila. In summary, the role of AP compartmentalization of the somites in the establishment of segmental pattern within the body, such as for the peripheral nervous system and vertebrae, is beginning to be well understood. Within somites, it is still unclear whether this compartmentalization is restricted to the sclerotome or whether it also impinges upon the development of the dermomyotome.

18

HIRSINGER

ETA.

At the molecular level, Eph molecules have been shown to have a restricted expression in the paraxial mesoderm and are thought to play a major role in maintaining the boundaries of each half somite domain. Cell-cell interactions mediated by the Notch pathway are implicated in the specification of rostro-caudal polarity of the somite as well as in the establishment of segment borders. Nevertheless, further experiments should be designed to address the molecular and cellular events that mediate alternation of anterior and posterior compartments in the par-axial mesoderm. Besides, the exact relationships between the existence of these compartments and somite formation are still unclear.

V. Dorso-ventral

Patterning

A. Three Different Lineages Dorso-ventral Axis

of the Avian Segregate

Somite

along the

1. Somites Give Rise to Skeleton, Muscle, and Dermis

Once formed, the epithelial somite then undergoes regionalized morphogenetic transformations along a second axis, the dorso-ventral axis. This process leads to the segregation of different somitic lineages, dermis, muscle, and skeleton. The somitic contribution to the formation of the axial skeleton has been mapped using quail-chick chimeras (Christ and Wilting, 1992). Although the six most rostra1 somites give rise to the basi-occipital bone of the skull (Couly et aZ., 1993), all the axial body skeleton derives from the remaining somites (Christ and Ordahl, 1995). The vertebrae and intervertebral disks, the ribs, and part of the scapula derive from the somites, whereas the limb and girdle skeleton derive from the lateral plate. It has now been well established that the vertebral body arises from cells of the ventro-medial sclerotome which migrate ventrally toward the notochord (Christ and Wilting, 1992; Ordahl and Le Douarin, 1992). The vertebral pedicles of the neural arches have been shown to derive from the caudal part of the somitic compartment, whereas the rostra1 somitic half yields the intervertebral disk (Goldstein and Kalcheim, 1992). The dorsal spinous process of the vertebra also derives from the somite, and its formation requires a specific set of interactions with the surrounding structures like the ectoderm and neural tube (Monsoro-Burq et al, 1994; Takahashi et al, 1992). The origin of the ribs is more complex because their vertebral part derives from the somitocoele and caudal somite cells (Huang et al., 1994) whereas the sternal part derives from the lateral dermomyotome (Kato and Aoyama, 1998). Therefore, although the axial skeleton arises primarily

SOMITOGENESIS

19

from the sclerotome, the dermomyotome is also able to give rise to some skeletal derivatives. The anatomical origin of the myotame has been and still is a subject of controversy. Three major models have been proposed. Williams, among others, proposed in 1910 that the myotome originates from the medial and, to a lesser extent, from the lateral edges of the dermomyotome. Later on, Langman and Nelson (1968) concluded, from labeling experiments with tritiated thymidine, that cells from the whole surface of the dermomyotome delaminate and form the underlying myotome. Later still, Mestres and Hinrichsen (1976) were led to believe that sclerotomal cells in contact with the overlying dermatome reaggregate to form the myotome. These different models have been put to the test using progressively more refined techniques. Christ and colleagues (1978), using quail-chick grafting techniques, demonstrated that the myotome derives exclusively from the dermomyotome, refuting Mestres and Hinrichsen’s model. By studying the onset of desmin expression in the somite, Kaehn and colfaborators (1988) reached the conclusion that myogenic precursors arise strictly from the crania-medial corner of the dermomyotome, contradicting the results obtained by Langmann and Nelson. More recently, DiI injection mapping of the dermomyotome (Denetclaw et aZ., 1997) demonstrated that myogenic precursors emerge not only from the crania-medial edge of the dermomyotome but also from the entire extent of the medial lip. The existence of a subset of myotomal cells which become postmitotic very early at the epithelial somite stage, has been reported in the chick by Kahane and co-workers (199&b). These very early muscle precursors were termed pioneer cells and arise from the dorso-medial epithelial somite. They are the first cells to migrate beneath the dermomyotome and elongate along the antero-posterior axis to form the primary myotome. By analogy with Drosophila and zebrafish muscle pioneers (Ho et al., 1983; see Jellies, 1990 for a review), this primary myotome composed of pioneer cells could serve as a longitudinal scaffold, providing guidance cues for the migration of a secondary wave of myotamal precursors, which arise from the anterior and posterior lips of the dermomyotome (Kahane et aZ., 1998a). Due to the lack of morphological criteria to follow the differentiation of dermal cells, dermis differentiation is far less well understood than myotome or sclerotome differentiation. Only the dermis of the back arises from the somite (Christ and Ordahl, 19953. The dermis of the limbs is derived from the somatopleura, and the dermis of the face arises from the neural crest (Couly and Le Douarin, 1988). It is’ generally assumed that dermis arises from the dermatome (i.e., the epithelial plate located above the myotome). Beginning at E3 in the chick embryo, the dermatomes lose their epithelial structures, and their cells migrate beneath the epidermis (Brill et al, 1995).

20

HIRSINGERET AL.

2. External Cues Are Involved in the Segregation of the Three Semitic Lineages How is the dorso-ventral axis established and thereby how are the different lineages specified? Two major mechanisms are classically considered to regulate cell diversification-either cell lineage relationships or environmental cues. Cell lineage analyses have been conducted in the PSM (Stern et al., 1988). When a single cell is labeled at any level of the unsegmented mesoderm, its progeny is not restricted to dermatome, myotome, or sclerotome. Progeny of the injected cell can be found in two or all three of these derivatives. Therefore it appears that, in the PSM at least, the segregation of the three fates does not rely on cell lineage mechanisms. To ask whether the establishment of the dorso-ventral axis is controlled by external cues, dorso-ventral inversion of the last three somites (Aoyama and Asamoto, 1988), or heterotopic grafts of dorsal or ventral somite pieces (Aoyama, 1993; Christ et al., 1992) were performed. The manipulated somites I and II differentiate according to their new local environment (i.e., the former dorsal half forms the sclerotome, whereas the former ventral side generates the dermomyotome). In contrast, somite III exhibits an inverted polarity suggesting that, from this level forward, the dorso-ventral axis may be irreversibly determined (Aoyama and Asamoto, 1988). Along the same lines, when quail ventral somitic halves are grafted in place of chick dorsal halves at the level of the two last formed somites, the quail cells form dermomyotome and subsequently differentiate as dermis and muscle (Christ et al., 1992). The lability of the dorso-ventral axis is progressively lost as one moves toward the anterior extremity. Taken together these results suggest that the establishment of dorsoventral polarity is under the control of environmental cues. What is their nature? Are they inductive, in the sense that they instruct a naive tissue to adopt particular cell fates, or are they permissive, in that they promote an intrinsic and preexisting potential of differentiation?

B. Self Differentiation

or Induction?

1. Historical Aspects Whether somitic differentiation is an example of self-differentiation or whether it is dependent upon extrinsic cues has been a long-standing debate. Theoretical models have varied along with experimental strategies. Since the beginning of the century, numerous studies focused on chondrogenesis, mainly because manifestations of on-going chondrocyte differentiation were readily observable under the microscope. Based on the observations that isolated chick somites grafted on the chorio-allantoidian membrane

SOMITOGENESIS

21

underwent chondrogenesis and that this phenomenon was amplified if axial structures were present, the concept of self differentiation prevailed until the early 1940s (Hall, 1977). Extirpation experiments performed in the 1940s in amphibian and avian embryos established that, in VIVO,the notochord and neural tube influence cartilage differentiation (Holtzer and Detwiler, 1953; Horstadius, 1944; Strudel, 1955; Watterson et aZ., 1954). In the 195Os, I:12 vitro analyses of mice explant combinations clarified that cartilage forms from somitic mesoderm only if an inducer is present (Grobstein and Holtzer, 1955). Finally in the 1960s attention focused on the somite as a responding tissue, and induction was consequently considered as enhancing a preexisting potential rather than as imposing new potential on a naive tissue (Hall, 1977). 2. The Commitment State of Semitic Cells Evolves along Temporal and Spatial Axes With the characterization of molecular markers for sclerotogenic and myogenie lineages in the 198Os, the controversy of “self-differentiation versus induction” is now being revisited in detail with respect to chondrogenesis and myogenesis. In vivo, the paraxial mesoderm at the unsegmented level does not exhibit molecular differentiation along a particular lineage. It is only in the segmented region that somites start to express differentiation markers in a rostro-caudal gradient. To assay the state of commitment of paraxial mesoderm, this tissue was explanted and cultured in isolation. Observations of in vitro cultures of chick or mice explants led to the condusion that, at a given stage, PSM and the last two or three formed somites are uncommited toward any lineage. In contrast, more rostra1 somites differentiate autonomously, suggesting that somites progress along an anteroposterior gradient of commitment (Buffinger and Stockdale, 1994; Cossu et al., 1996; Ebensperger et aZ., 1995; Fan and Tessier-Lavigne, 1994; Kos et aZ., 1998; Mtiller et aZ., 1996; Mtinsterberg and Lassar, 1995; Reshef et aZ., 1998; Rong et aZ., 1992; Spence et aZ., 1996). In addition to the antero-posterior variation of the commitment state, somites of a given level explanted at different embryonic ages exhibit increasing myogenic capacities with age (Kenny-Mobbs and Thorogood, 1987; Rong et al., 1992; Stern and Hauschka, 1995; Vivarelli and Cossu, 1986). This occurs because the rate of myogenic onset is approximately 1.5 times faster than that of somite formation (Borman and Yorde, 1994a, 1994b). These data suggest that myogenesis is controlled by both somitic competence and developmentally regulated differentiation signals and not solely by the temporal sequence of somite formation (Borycki et aZ., 1997; Rong et aZ., 1992). In contrast, activation of sclerotome markers always occurs at the same’somitic level, suggesting ‘that it is strictly controlled by

22

HIRSINGER

ET AL.

somite formation and differentiation competence, rather than by timing of the differentiation signals (Borycki et aZ., 1997). Consequently, it appears that the state of somite commitment develops along both temporal and spatial axes and that specification of muscle and skeleton lineages exhibit different modalities. Another means by which one can determine when cells are stably committed is to place them in a challenging environment, that is an environment known to drive cells towards an alternative fate. By grafting tissue fragments into the limb bud, different groups have shown that myogenic potential arises in the primitive streak during gastrulation (Chen and Solursh, 1991; Krenn et aZ., 1988; von Kirschhofer et al., 1994) and that muscular and chondrogenic determination occurs as early as the 8-somite stage (Wachtler et al., 1982a). Williams and Ordahl (1997) used a different challenge assay in which the somitic dorso-medial quadrant was exposed in viva to supernumerary notochords that act as sclerotome inducers. They demonstrated that a number of muscle cells manage to differentiate in this ventralizing environment, indicating their irreversible commitment to the myogenic lineage. This number increases with the age of the donor somite. In a parallel study, Dokter and Ordahl (1998) demonstrated that it is not until somite stage XII that the first cartilage progenitors cells are determined and not until E4 that the entire sclerotome is stably committed to cartilage differentiation. Therefore myogenic and sclerotogenic determination occur progressively and asynchronously during development. This suggests that, at a given time point, the somite is composed of a mosaic of cellular territories, each having reached a different state of commitment toward a specific lineage. 3. Environmental Signals Are Required Somite Differentiation

for

As previously mentioned, culturing PSM or epithelial somites in isolation or in association with adjacent structures revealed that this explanted tissue does not undergo differentiation unless an adjacent structure is present in the culture (Buffinger and Stockdale, 1994; Cossu et al., 1996; Ebensperger et al., 199.5; Fan and Tessier-Lavigne, 1994; Kos et aZ., 1998; Mtiller et al., 1996; Mtinsterberg and Lassar, 1995; Reshef et aZ., 1998; Rong et aZ., 1992; Spence et aZ., 1996). These data demonstrate that adjacent structures are necessary for correct development of the somite, but they do not determine the exact nature of their influences. In addition to controlling the competence of the paraxial mesoderm, which will be discussed later, intercellular communications within paraxial mesoderm have been shown to play a role in a “community effect.” First described in Xenopus mesoderm induction (Gurdon, 1988) the ability of a cell to respond to inducing signals is enhanced when it is surrounded by

23

SOMITOGENESIS

neighboring cells following the same differentiation pathway at the same time. Indeed, such a mechanism is observed in mammals (Cossu et al., 1995) and in cultures of chick dissociated epiblasts, in which myogenesis occurs to a greater extent when cells are plated at high density than at low density (George-Weinstein et al., 1996). This “community effect” might be mediated by homophilic cell-cell adhesion molecules, the cadherins, as evidenced by their expression patterns and blocking experiments (Duband et al,, 1987; George-Weinstein et al., 1997; Hatta and Takeichi, 1986; Holt et al., 1994). 4. But Cell-Autonomous

Mechanisms May Also Function

Despite the data described earlier, one cannot rule out the possibility of a cell autonomous component contributing to cell fate when one considers in vivo or tissue dissociation studies such as those described later. When the neural tube/notochord complex, which is thought to be essential for myogenic and sclerotogenic differentiation, is extirpated in vivo, a few myogenic or sclerogenic cells nevertheless differentiate from somites which have not been exposed to axial organ influence after segmentation (Bober et aZ., 1994; Teillet et al., 1998). Similar results have been obtained when the presumptive territory of notochord and floor plate is ablated (Catala et aL, 1996). Moreover, fragments or dissociated cultures of segmental plate or of epithelial somites unlike intact explanted tissues, can undergo myogenesis and chondrogenesis (Game1 et al., 1995; George-Weinstein et al., 1994). One interpretation of these data is that cells of the PSM are already endowed with myogenic potential and that cellular interactions, possibly mediated by the Notch signaling pathway, prevent precocious expression of this potential. Signals derived from surrounding structures might therefore be permissive rather than instructive thereby imposing temporal control on the onset of chondrogenic and myogenic ‘differentiation in the paraxial mesoderm. George-Weinstein and co-workers (1998) proposed an explanation to reconcile the autonomous versus inductive views. Populations of stably committed cells are randomly distributed in the mesoderm and cohabit with uncommitted cells. Differentiation of the latter depends upon communication between these two populations, regulated both by cell-cell interactions within the mesoderm and by influences from surrounding tissues. C. Combinatorial and Antagonistic Signals Emanating from the Overlying Ectoderm, the Neural Tube, and the Notochord Establish the Dorso-ventral Polarity

Somites are surrounded by the ectoderm dorsally, the lateral plate laterally, the neural tube and notochord medially, and the endoderm ventrally. Ex-

24

HIRSINGERET AL.

cept for the endoderm, which as yet has not been studied in this context in great detail, the putative influence of each of these structures ‘has been examined. 1. The Notochord and Floor Plate Exert a Ventralizing Influence on the Somite Following on from the pioneering work of many laboratories (Hall, 1977), several groups have provided evidence that the notochord and floor plate are structures which are both necessary and sufficient for directing the differentiation of the adjacent paraxial mesoderm tissue toward the sclerotomal lineage. The patterning activities of these two structures on the ventral neural tube are strikingly similar. Notochord has been shown to specify the ventral part of the neural tube and more precisely to promote directly the differentiation of the floor plate and the bilaterally positioned motoneurons. The floor plate subsequently exhibits a patterning activity similar to that of the notochord on motoneurons (Placzek, 1995). In the case of the paraxial mesoderm, when a supernumerary notochord is grafted dorsal to the PSM, the dorsal compartment appears to be transformed into a ventral one: dorsal cells undergo an epithelio-mesenchymal transition such that no dermomyotome is formed (Pourquie et al, 1993); the expression domains of sclerotomal markers such as Pax-1 and Pax-9 are dorsally expanded, whereas dorsal markers (Pax-3 and Pax-7) are downregulated (Brand-Saberi et aZ., 1993; Dietrich et al., 1998, 1997; Goulding et aZ., 1994). Subsequently, extra cartilage is produced at the expense of axial musculature and dermis (Goulding et aZ., 1994; Pourquie et aZ., 1993). A floor plate can also convert the dermomyotome towards a sclerotomal fate (Brand-Saberi et aZ., 1993; Pourquie et aZ., 1993). In the loop-tail mouse mutant, as a consequence of overdifferentiation of the notochord and floor plate, the neural tube does not close dorsally and somites appear ventralized as evidenced by up-regulation of Pax-l expression and down-regulation of Pax-3. This somite phenotype is likely to be due to increased signaling from the expanded ventralizing structures (Greene et aZ., 1998). In vitro experiments in which naive presomitic cells were cocultured with notochord or floor plate, led to similar conclusions about the ventralizing activities of these structures (Ebensperger et al., 1995; Fan and Tessier-Lavigne, 1994; Mtiller et aZ., 1996). Conversely, early ablation of the notochord along its whole length, such that floor plate is not induced, or ablation of both floor plate and notochord leads to an absence of Pax-l, Pax-9, and MyoD and to a ventral extension of the Pax-3 and Pax-7 expression domains in most of the region of the ablation (Dietrich et al., 1997; Goulding et al., 1994; Pownall et aZ., 1996; Teillet et aZ., 1998). If these embryos are allowed to develop further, no

SOMITOGENESIS

25

ventral derivatives like cartilage are observed, whereas dorsal ones like muscle and dermis are present (Van Straaten and Hekking, 1991). Analyses of mouse mutants in which notochord and floor plate development are affected support these results. In the truncate and Brachyury curtailed mutant embryos, the notochord does not develop: Pax-l expression is never activated, whereas the whole somite expresses Pax-3 (Dietrich et aZ., 1993). In the Dan@rth’s short-tail and pintail mutant embryos, the notochord degenerates secondarily: Pax-l expression is induced but is subsequently lost, whereas Pax-3 expression invades the whole somite (Asakura and Tapscott, 1998; Dietrich et aZ., 1993; Koseki et al., 1993). These data therefore suggest that notochord signaling is required for sclerotome development. If the ablation is performed after induction of the floor plate, then somites develop normally, indicating that the notochord and floor plate have redundant patterning activities (Dietrich et al, 1997; Ebensperger et al., 1995; see also Watterson et al., 1954). Thus, as in the case of dorso-ventral patterning of the neural tube, these two structures seem to exhibit strikingly similar dorso-ventral patterning activities on the somite. It is not yet clear whether they act synergisticahy or in a cascade in which the floor plate acts as a functional relay of the notochord. These observations have led to the proposal that dorsal differentiation in the somite reflects the existence of a default pathway for the somitic lineages (Pourquie et al, 1993). This hypothesis provides an alternative to the myogenic induction models, which will be discussed later. In the absence of notochord and floor plate signaling, somitic cells would differentiate toward the muscle and dermal lineages. This hypothesis implies that dorsal lineages require only permissive signals for their differentiation, whereas ventral lineages require instructive signaling provided by the floor plate and notochord. This default model is further supported by results from ipt vitro culture of dissociated chick epiblast in which 99% of the cells undergo myogenesis after 2 hr, suggesting that muscle differentiation could occur autonomously as early as epiblast stages before gastrulation. Such a precocious ‘event would be prevented in vivo by tissular integrity and presumably cell communication (George-Weinstein et aL, 1996). 2. The Dorsal Ectodem and Neural Tube Deliver Dorsalizing Signals That Antagonize the NotochoRUFloor Plate Influence Whether they are of an inductive or of a permissive nature, signals emanating from surrounding structures play a role in defining a ‘dorsal identity. In vivo manipulations consisting of ablation or ectopic grafts and in vitro experiments using chick embryos have shown that the ectoderm and/or the dorsal neural tube are necessary and sufficient for the epithelialization of the

26

HIRSINGERET AL.

somite and subsequently of the dermomyotome (Gallera, 1966; Palmeirim et al., 1998; Sosic et al., 1997; Spence et al., 1996). The individual roles of these structures have been investigated in vitro. When naive, non-dorso-ventrally determined murine presomitic explants are cultured with notochord or floor plate, these cells express a sclerotomal marker (Pax-l). When similar explants are cultured with dorsal ectoderm or dorsal neural tube, they express dermomyotomal markers such as Pax-3, Pax-7 and Sim-1 (Fan and Tessier-Lavigne, 1994; Maroto et al., 1997; Reshef et aZ., 1998). Moreover, if such coculture is performed with more mature somites that already have sclerotogenic properties, cartilage differentiation is strongly impeded (Kenny-Mobbs and Thorogood, 1987). liz vivo experiments have aided our understanding of the complex network of interactions between the different structures. Independent ablation of ectoderm or dorsal neural tube does not affect the dorso-ventral polarity of somites (Dietrich et al., 1998, 1997; Hirano et aZ., 1995; Kuratani et al., 1994). In contrast, ectopic grafts of these tissues individually leads to ectopic activation of the dorsal marker Pax-3 and down-regulation of the ventral marker Pax-l (Dietrich et al, 1997). Both structures have to be removed at the same time in order to obtain fully ventralized somites (Dietrich et al., 1997). In the open brain mutant mice, the neural tube lacks its dorsal domain leading to abnormal expression of Pax-3 and dermomyotome defects. Whether the ectoderm is affected in these mice is not known (Sporle et a& 1996). Taken together, these results indicate that ectoderm and dorsal neural tube have necessary and redundant dorsalising activities that antagonize the ventralizing signals from notochord and floor plate. However, these dorsal signals can be overridden by ventral signals. For example, the graft of an ectopic notochord dorsal to the PSM leads to complete absence of muscles and dermis on the grafted side (Pourquie et aZ., 1993). Accordingly, the dorso-ventral inversion of both neural tube-notochord complex leads to a concomitant dorso-ventral inversion of this axis in the juxtaposed somites (Dietrich et al, 1997; Spence et aL, 1996). A role for the dorsal neural tube in patterning the dorsal sclerotome, fated to form the neural arches, is also suggested by the analysis of the open brain mouse mutant (Sporle et al., 1996). Furthermore, the notochord mutant, Danforth’s short-tail, exhibits defects solely in the ventral sclerotome resulting in abnormal vertebrae formation (Koseki eta& 1993). Therefore, antagonistic influences from the dorsal neural tube and notochord/ floor plate might also act to pattern the sclerotome itself along the dorsoventral axis. As differentiation proceeds, the ectoderm is also necessary for dermatome-specific gene expression such as gMHox (Kuratani et aZ., 1994). The dorsal neural tube is required for a subsequent differentiation step of this compartment, the epithelial-mesenchymal conversion of the dermatome precursors (Brill et al., 1995).

SOMITOGENESIS

27

In summary, it seems that notochord/floor plate, dorsal ectoderm, and neural tube exert antagonistic influences to pattern the somite along the dorso-ventral axis. It is also noteworthy that correct dorso-ventral patterning of the neural tube is necessary for proper dorso-ventral patterning of somites. Interestingly, the nature of the notochord, floor plate and neural tube signals seem to be diffusible, whereas the ectoderm effect requires cell-cell contact (Cooper, 1965; Fan and Tessier-Lavigne, 1994; Flower and Grobstein, 1967; Lash et aZ., 1957). 3. The Neural Tube/Ectaderm and Notochord/Floor Involved in Myogenic Differentiation

Plate Are

The precise means by which myogenesis is initiated and is then maintained in the somite has been and still is highly controversial, although it is now widely accepted that the neural tube, surface ectoderm, lateral plate, notochord, and floor plate are all involved to some extent in these processes. The discrepancies, which exist between the various studies addressing this issue, might be accounted for by the different strategies used and by variability in the age of the embryos and in the time of operation. The observation that myogenesis in the embryo is first initiated in the somitic territory apposed to the dorsal neural tube prompted several laboratories to test the myogenic-inducing role of the neural tube. Using in vitro approaches, it was demonstrated that the paraxial mesoderm and the newly formed somites require axiaJ signaling to differentiate into muscle (KennyMobbs and Thorogood, 1987; Packard and Jacobson, 1976; Rong et GE:, 1992; Vivarelli and Gossu, 1986). More recently, this question was revisited by a number of labs and yielded a series of conflicting results. Buffinger and Stockdale (1994) reported that the neural tube but not the notochord could elicit myogenesis from segmental’ plate explants. In contrast, Stern and collaborators (1995), showed that, if the complex neural tube/notochord induces a 100% myogenic response in the paraxial mesoderm, the dorsal neural tube induces an 80% response, whereas the ventral neural tube with or without notochord triggers a 30% and a 10% myogenic response, respectively. A role for the dorsal neural tube is further confirmed by the phenotype of the open brain mutant mice in which’muscle differentiation is impaired. These mice exhibit altered expression patterns of Pax-3 and myf-5 followed by severe defects in muscle formation (Sporle et al., 1996). Miinsterberg and Lassar (1995) for their part have shown that, in vitro, myogenic induction ‘from segmental plate requires coculture with both notochord and neural tube, whereas more rostra1 somites require only neural tube signaling to differentiate. They1 thus propose a model involving a two step mechanism whereby notochord signaling would confer upon the somite the competence to respond to myogenic signals from the neural

28

HIRSINGERET AL.

tube. In viva manipulations have provided further evidence in favor of such a model. Caudal ablation of the notochord leads to failure of MyoD induction, but when the notochord is removed more anteriorly, the presence of the neural tube and ectoderm are sufficient to maintain MyoD expression (Pownall et al, 1996). A parallel situation is found in the case of the Danforth’s short-tail mutant mice: in anterior regions, a notochord develops but degenerates secondarily, whereas the somites at this level exhibit normal myogenesis raising the possibility that initial exposure to the notochord in the presence of the neural tube/floor plate and the ectoderm was sufficient to initiate myogenesis (Asakura and Tapscott, 1998). Conversely, in neural tube-ablated embryos, myogenic markers are induced but not maintained (Bober et al., 1994). The role of the notochord and floor plate in myogenic specification has been further addressed in vivo by performing ectopic grafts of these structures. In contrast to sclerotome induction by the notochord, no tissue adjacent to the paraxial mesoderm has been shown to exhibit such a clear inducing capacity in the recruitment of paraxial mesodermal cells toward the myogenic pathway. However, ectopic notochord grafts have been reported to induce short-term ectopic MyoD activation in the region of the graft in in vitro cultures of quail embryos (Pownall et al., 1996). In the majority of cases, notochord implantation leads to an inhibition of the myogenic program in nearby cells, although it promotes the process in cells located at a distance from the graft (Bober et al., 1994; Brand-Saberi et al, 1993; Dietrich et aZ., 1997; Pourquie et aZ., 1993). In most of the cases, these grafting experiments result in abnormal MyoD expression in the dermomyotome. This could result either from premature activation of the myogenic factors in territories which are normally fated to give rise to muscle, like the lateral somite, or from disrupted architecture of the myotome itself resulting in local concentrations of MyoD expressing cells. The only evidence for instructive signaling in myogenesis comes from dorsoventral inversion of the neural tube which can elicit formation of an ectopic dermomyotome from the ventral somite (Dietrich et aZ., 1997; Spence et al., 1996). In conclusion, both axial structures seem important for myogenesis, but the permissive or inductive nature of the signal emanating from these axial structures is still unclear. In viva, if the notochord and neural tube are both removed, no epaxial myogenesis occurs because all medial somitic cells die within 24 hours (Teillet and Le Douarin, 1983). These structures thus provide at least a trophic support to somitic cells. Backgraft of either the neural tube or the notochord alone rescues myogenesis (Rong et al, 1992). Backgraft of the dorsal neural tube is far less efficient in rescuing MyoD expression indicating that most of this trophic (i.e., permissive) effect can be accounted for by the notochord and, to a lesser extent, by the ventral neural tube (Teillet et al., 1998). The role of the notochord in initializing the myogenic program

SOMITOGENESIS

29

is further questioned by the fact that ablation of the presumptive territory of the notochord and floor plate or simulatneous ablation of the neural tube and notochord does not prevent the subsequent differentiation of a few muscle cells (Teillet et al., 1998; Bober el al., 1994; Catala et a& 19%). Moreover, in the caudal region of a Danforth’s short-tail mutant, where neither notochord nor floor plate develop, weak myf-5 is nevertheless expressed in somites (Asakura and Tapscott, 1998). In summary, axial structures play an important role in myogenic differentiation, but this role seems to be more of a permissive (or trophic) nature. A third tissue, the dorsal ectoderm, may also regulate myogenesis. Culture of isolated lateral PSM (Cossu et aZ., 1995) or caudal somites (Reshef et al., 1998) with the overlying ectoderm leads to the activation of MyoL? expression. Moreover, ablation and ectapic graft experiments led Dietrich and collabarators (1998; 1997; see aIso Hirano et al., 1995) to conclude that, as for sclerotome induction, ectoderm and dorsal neural tubes have redundant myogenic activities (see also Spence et al., 1996). Controversy also exists concerning the mechanisms involved in this process. Bnffinger and Stockdale (1995), using transhlter cultures, proposed that the signals emanating from both the neural tube and the notachord are diffusible, whereas other groups have shown that neural tube and ectoderm require close contact with the somitic tissue to exert their effects (Cossu et al., 1996; Stern and Hauschka, 1995). To conclude, these conflicting results reflect the complexity of the regulation of the onset of myogenesis. Developmental time windows seem to be a crucial parameter; various surrounding structures send combinatorial andlor antagonizing signals to the paraxial mesoderm, which subsequently interprets these signals according to its own state of maturation. In addition, because some of these signals are diffusible, it is possible that different signaling threshold levels elicit different cellular responses. This morphogen-type response of somitic cells could account for the observation that the myotome forms at an intermediate position relative to dorsalizing (the ectoderm and the dorsal neural tube) and ventralizing signals (the notochord and the floor plate) (Dietrich et aZ., 1997). Alternatively, myogenesis could reflect a default pathway of the somitic cells and the role crf these signals would then be to provide traphic support to somitic cells and to control precisely the timing of the onset of terminal differentiation (Pourquie et al., 1995, 1993; Teillet et al., 1998).

D. Shh and Wnt Proteins Are Key Flayers in the Segregation of the Different Semitic Lineages

The previously described complexity of surrounding tissue interactions that pattern the son-rite is also reflected at the molecular level. The predominant

30

HIRSINGER

ET AL.

role of two genes, both encoding secreted molecules, has been extensively examined in vivo and in vitro. Sonic hedgehog (Shh) is expressed in the notochord and floor plate along their entire length (Echelard et aZ., 1993; Krauss et aZ., 1993; Marti et aZ., 1995b; Riddle et aZ., 1993) whereas several Wnt proteins are expressed either in the dorsal neural tube (Wntl, 3a, 4) or in the dorsal ectoderm (Wnt4,6,7a) (Hollyday et al., 1995; Marcelle et aZ., 1997; McMahon and Bradley, 1990; Parr et al, 1993; Tajbakhsh et aZ., 1998). Moreover, these factors have been shown to exhibit patterning activities in diverse systems (reviewed in Cadigan and Nusse, 1997; Tanabe and Jessell, 1996). Therefore, because these molecules are present at the right time and place, they appear to be good candidates for mediating somite patterning. 1. Shh Promotes Ventral Cell Fates whereas Wnt Proteins Specify Dorsal Lineages In vitro and in vivo experiments have provided evidence of a role for Shh in promoting sclerotome differentiation. Coculture of uncommitted paraxial mesoderm explants either with cells secreting Shh (Fan and TessierLavigne, 1994) or with purified Shh protein (Fan et al., 1995; Maroto et al., 1997; Mtinsterberg et aZ., 1995; Reshef et al., 1998) leads to activation of Pax-l expression, as was obtained in similar cocultures with notochord. Accordingly, replacement of notochord and floor plate by a Shh secreting bead or by secreting cells or, conversely, antisense inhibition of quail Shh expression together demonstrated, in vivo, that Shh is important for maintaining and/or inducing Pax-l expression (Borycki et al., 1998; Teillet et al., 1998). Furthermore, in an in vitro assay, Shh protein was shown to be capable of antagonizing the induction of dermomyotomal markers elicited by dorsal neural tube (Fan and Tessier-Lavigne, 1994; Maroto et aZ., 1997). In accordance with these studies, ectopic expression of Shh in the chick paraxial mesoderm using a retroviral expression system, revealed that Pax-1 expression expanded dorsally while Pax-3 was down-regulated and the dermatome was anatomically perturbed (Johnson et al., 1994). An equivalent situation is found in the loop-tail mutant embryo in which Shh is expressed in an abnormally broad expression domain. The Pax-l expression domain is expanded dorsally while Pax-3 expression diminishes (Greene et al., 1998). These experiments suggest that exposure to Shh can convert the dorsal somitic compartment toward ventral cell fates, and therefore can mimick the effect of the graft of a supernumary notochord in a dorsal position. However, several lines of evidence indicate that Shh does not account for all the ventralizing properties of the notochord. First, in the Shh retroviral overexpression studies in the chick, a strong MyoD up-regulation is ob-

SOMITOGENESIS

31

served in the overexpression domain. In contrast, notochord grafts lead to MyoD down-regulation in the close vicinity. Second, such a direct role for Shh in sclerotome induction appears unlikely because of the phenotype of Shh homozygous null mice. Sclerotome derivatives are, as expected, sever@ reduced, and the Pax-3 expression domain extends ventrally. Surprisingly, however, Pax-l expression is transiently induced although, in agreement with the results obtained in vitro and in vivo, this expression is not maintained (Chiang et aZ., 1996). A similar situation is found in the case of the Danforth’s short-tail mutant (Asakura and Tapscott, 1998) or after ablation of the notochord/neural tube complex, both scenarios being that the embryos lose the source of Shh protein (Teillet et al., 1998): the somites which develop in the absence of Shh signaling activity exhibit weak Pax-l expression. These data suggest that Shh is involved in the maintenance of Pax-1 expression rather than in its induction in the epithelial somite. In a reciprocal manner, retroviral ectopic expression af Writ-1 in somites elicits the opposite response: Pax-l and Pax-9 expression are lost and at ElO, the axial skeleton is almost completely absent, suggesting that the development of the sclerotome has been severely impaired. Nevertheless, unlike the capacity of Shh to extend Pax-l expression dorsally, Wnt-1 overexpression does not expand the Pax-3 expression domain ventrally (Capdevila et al., 1998). In vitro experiments support the model that different Wnt proteins mediate the dorsalizing role of the neural tube and ectoderm (Fan et aZ., 1997; Maroto et al., 1997). Uncommitted mesoderm cultured with cells expressing Wnt-1, 3a, 4, or 6 expresses dermomyotomal markers. Interposition of a nucleopore filter leads to a decrease in the effect of Wnt-4 and 6 activities, whereas the signaling potential of Wnt-1 and 3a is unaffected (Fab et al., 1997). Considering their exfiression patterns and their diffusibility, Wnt-1 and 3a appear to be good candidates for the secreted dorsalizing factor derived from rhe neural tube, whereas Wnt-4 and 6 meet all the necessary characteristics of the contact-dependent dorsalizing derived factor from the ectoderm. The phenotype of the compound mutant Wnt-l/Wnt3a highlights the redundant influences of these two sets of Wnt genes in this process: the dorso-ventral pblarity of their somites is normal, the Wnt proteins from the ectoderm possibly compensating for the loss of those from the neural tube (Ikeya and Takada, 1998). The antagonistic activities of Shh and Wnts has been demonstrated in sandwich cultures where a naive mesoderm explant is positioned between Shh-expressing cells on one side and Wnt-l-expressing cells on the other. This results in the’ induction of Pax-l and Pax-3 expression in domains close to the ventralizing and dorsalizing signal sources, respectively. When Shh signaling is increased, the Pax-l expression domain extends toward

32

HIRSINGERET AL.

the Wnt-1 protein source whereas the Pax-3 expression domain is reduced (Fan et al., 1997). These data indicate that Writ-1 and Shh signaling pattern the paraxial mesoderm in an antagonistic manner suggestive of a concentrationdependent action. Such a concentration-dependent action has been previously documented for patterning of the neural tube (Roelink et aZ., 1995). However, a strong caveat to this model arises from experiments performed by Teillet et al. (1998). When they replaced the neural tube/notochord complex with Shh expressing cells, a fairly normal dorso-ventral patterning of the somites was observed. Accordingly, surprisingly normal vertebrallike structures and associated muscles formed. These experiments suggest that Shh essentially rescues the death of somitic cells which normally occurs after ablation of the neural tube and notochord (Teillet and Le Douarin, 1983). If the choice to differentiate into muscle or into cartilage results from exposure to different levels of Shh and Wnt signals, then it seems very unlikely that grafting a pellet of cells producing an uncontrolled quantity of Shh would rescue an exact pattern of cartilage and muscles. Moreover in the rescue experiments, two different cell lines producing Shh were used-a stable transfectant and a cell line infected by a retrovirus-producing Shh. It is very unlikely that both lines produce the same amount of Shh, yet they elicit the same phenotype. These experiments therefore strongly argue in favor of a prepattern of the paraxial mesoderm along the dorso-ventral axis. The role of Shh and Wnt in this case would thus be permissive rather than instructive. Furthermore, studies aimed at addressing the diffusibility of these molecules demonstrate the complexity of the biochemistry of the two molecules. In vivo, Shh is autocatalytically processed into a soluble. C-terminus form and an N-terminus form (Shh-N), which becomes anchored in the membrane via a cholesterol bond (Lee et aZ., 1994). However, the N-terminus form alone accounts for all the signaling activity of the molecule both in patterning the mesoderm and the neural tube (Fan et al., 1995; Marti et al, 1995a; Mtinsterberg et al, 1995; Roelink et aZ., 1995). If the effects of Shh-N are direct, then the protein needs to be released from the membrane and to subsequently diffuse from the inducing tissue. Surprisingly however, cells transfected with full-length Shh exhibit a Pax-l-inducing activity, whereas the N-terminal form is not detected in the culture medium. In Drosophila, a gene necessary for Hedgehog diffusion has been characterized: Tout-velu is a fly homolog of a human tumor suppressor gene, Ext-1, that is also implicated in Hedgehog protein diffusion, thereby suggesting the existence of a conserved mechanism implicated in this process (Bellaiche et al., 1998). On the other hand, analysis of Shh signaling in limb development favors the idea that its long-range signaling activity is indirect

SOMTOGENESIS

33

and mediated by the release of a diffusible secondary signal (Yang el al, 1997). Similar to Shh, Wnt proteins have been shown to be poorly soluble and to bind tightly to proteoglycans at the cell surface (Smolich et al., 1993). In Drosophila, the Wg long-range activity appears to be mediated by active intracellular transport of the protein from one cell to the next (Gumbiner, 1998). Thus, whether Wnt-1 and Shh proteins act directly by diffusion or cellular transport or whether their range of activity is mediated in a relay fashion by triggering a second diffusible messenger is still unclear. 2. Combinatorial SignaIling by Shh and Wnt Proteins Promote Myogenesis Although these molecules seem to have antagonistic effects on the specification of the dorsal-most and ventral-most derivatives of somites, they seem to specify cooperatively the intermediate compartment, the myotome. Coculture of unsegmented mesoderm or caudal somites in the presence of Shh alone does not result in muscle marker activation (Fan and TessierLavigne, 1994; Kos et al, 1998; Mtinsterberg et al., 1995; Tajbakhsh et al., 1998). In contrast, PSM culture in the presence of Wnt proteins (Wnt-1 or 4 or Sa or 6 or 7a) results in a low-level induction of myogenic precursors (Stern et aZ., 1995; Tajbakhsh et aZ., 1998). Therefore, it appears that Wnt proteins and Shh, when individually applied to naive mesoderm, are not fully sufficient to promote muscle development fully. Furthermore, in the case of the Shh knockout (Chiang et aZ., 1996) or as seen in the caudal region of the Danforth’s short-tail embryos (Asakura and Tapscott, 1998), expression of myf-5 is activated despite the lack of Shh protein. The phenotype of embryos lacking both Wnt-1 and Wnt3a genes suggests that these genes are essential for the early myf-5 expression, (Ikeya and Takada, 1998). Taken together, these results suggest that, as for the specification of the dorsal somitic compartment, the Wnt proteins produced by dorsal neural tube (Wnt-1, 3a, 4) and dorsal ectoderm (Wnt-4, 6, 7a) are both involved in myogenic specification. The situation is different when both signaling pathways are activated simultaneously. When PSM is cocultured in the presence of both Shh and Wnt proteins, MyoD and myf-5 expression are robustly up-regulated (Maroto et al., 1,997; Mtinsterberg et aZ., 1995; Reshef ef aZ., 1998; Tajbakhsh et al, 1998). When caudal~somites or PSM from oIder ‘embryos that have had a longer exposure to the influence of the notochord ipl situ are cocultured with Wnt proteins alone, myogenic precursors differentiate. Conversely, in experiments using antisense inhibition of Shh expression, lMyoD is no longer induced in caudal somites, (Borycki et aZ., 19%) There appearsto be a difference in the requirement of continued Shh signaling in viva and in

34

HIRSINGERET AL.

vitro because the continuous presence of Shh in vitro is not required for the maintenance of MyoD expression in explants (Mtinsterberg et al., 1995), whereas antisense inhibition experiments suggest that in vivo maintenance of previously induced MyoD expression requires continuous Shh signaling (Borycki et al., 1998). The results of these experiments suggest that Shh and Wnt fulfill the criteria for the combinatorial signals from the notochord and the dorsal neural tube of the model proposed by Munsterberg and Lassar (1995). They also provide molecular evidence in favor of the model described earlier in which notochord signaling, mediated by Shh, and neural tube signaling, mediated by Wnt proteins, acts synergistically to promote somitic myogenesis. 3. How Is the Response to Axial Signaling Spatially and Temporally Regulated? Considering the previous data, two major comments can be made. First, it appears that, although Wnt and Shh signals are permanently present along the whole length of the antero-posterior axis, sclerotome formation at all embryonic stages is initiated only after somite segmentation; myogenesis is activated still later in somites from stage 11 HH onward (Borycki et al., 1998; Buffinger and Stockdale, 1994; Dietrich et al., 1997). Secondly, Shh signaling can induce both myotomal and sclerotomal lineages. So one might wonder how cells can interpret and respond to the same signal in such distinct ways. With respect to the regulation of the initation of myogenesis and chondrogenesis, different mechanisms may be involved: a. The Notch Signaling Pathway The Notch signaling pathway, which is well established as a regulator of cell competence (Artavanis-Tsakonas et al, 1999), is a candidate for regulating the state of responsiveness of the paraxial mesoderm. The Notch1 and Notch2 receptors are expressed in the mouse PSM and are down-regulated, along with their ligands Dlll and D113, as somites form (Bettenhausen et aZ., 1995; de1 Amo et aZ., 1992; Dunwoodie et al., 1997; Hayashi et aZ., 1996; Henrique et al., 1995; Myat et al., 1996; Palmeirim et aZ., 1998; Reaume et aZ., 1992; Swiatek et al., 1994; Williams et al, 1995). Besides its role in the segmentation process discussed previously, forced activation of Notch signaling in Xenopus embryos or in mammalian cultures leads to a blockade of myogenesis (Jarriault et aZ., 1998; Kato et aZ., 1997; Kopan et aZ,, 1994; Lindsell et aZ., 1995; Luo et aZ., 1997; Nye et aZ., 1994; Shawber et al., 1996), possibly by directly interfering with the activity of MyoD (Kopan et aZ., 1994; Nye et al., 1994). Accordingly, in Drosophila, Notch has been implicated in regulation of myogenic differentiation (Baylies et aZ., 1998). As proposed by George-Weinstein et al.,

SOMITOGENESIS

35

(1998), Natch signaling pathway could therefore prevent precocious expression of the myogenic potential present from these early stages in the paraxial mesoderm. b. Controlling Expression of the Shh Receptor, Patched In avian embryos, regulation of the expression of Patched, the Shh receptor, could provide another molecular mechanism for temporally restricting PSM responsiveness. Patched is not expressed in the PSM, which may explain why neither sclerotome nor myotome specification occurs at this level despite the close vicinity of Shh-expressing tissues. Furthermore, in avian embryos younger than stage HHll, Patched is expressed exclusively in ventral somites, thereby explaining why myogenesis, unlike sclerotome formation, is not induced before stage 11 HH (Borycki et al., 6998). Similar mechanisms may occur in mammals because the two mouse Patched genes isolated so far exhibit expression patterns similar to their avian homologs during early development (Goodrich et al., 1996; Hahn et al., 1996; Motoyama et ai., 1998). c. Presence of Inhibitors of the Writ Pathway A third mechanism possibly involved in the regulation of mesodermal competence is based on Wnt signaling inhibitors. In Xenopus, one such molecule, Frzb, which is a soluble form of the Wnt receptor, Frizzled, has been shown to bind some Wnt proteins in vitro thereby inhibiting their activities (Leyns et al., 1997; Wang et al., 1997a, 1997b). Murine homologs of this gene have been cloned, and these are expressed in the PSM and the somites (Hoang et al., 1998). Cerberus, a member of the DAN family, has also been shown, in Xenopus, to inhibit Wnt signaling (Bouwmeester et al., 1996; Glinka et aZ., 1997). Its murine homolog is expressed in the PSM and the two last formed somites (Belo et al., 1997; Biben et al., 1998; Shawlot et al., 1998). The presence of these inhibitors of the Wnt pathway in the unsegmented mesoderm might provide a molecular mechanism to prevent precocious myogenic induction. Therefore activation of Notch signaling, restricted expression of the Shh receptor, or expression of Wnt signaling inhibitors, in the PSM, could account for the restricted responsiveness of the paraxial mesoderm to differentiation signals. With respect to Shh possibly eliciting myogenic or sclerotogenic differential responses, two mechanisms can be considered: controling expression of the effecters genes of the Shh pathway and c®ulation of the myogenic factors activity. d. Controling Expression of the Effecters Genes of the Shh Patlzway Gli and Gli2/4 are the vertebrate homologs of the fly transcription factor Cubitus interruptus (Ci) which is known to mediate the nuclear transcriptional

36

HIRSINGER

ET AL.

activity of Hedgehog signaling in Drosophila. The Gli genes are differentially expressed in somites (Borycki et aZ., 1998) and, as was described earlier for Patched, the time of their activation is correlated with somite segmentation. The Gli expression domain is confined to the ventro-medial domain of somites and its activation is regulated by the notochord and Shh itself. Gli2/4 is initially expressed throughout the whole somite but is later restricted to its medio-dorsal domain from where MyoD expressing cells arise; its activation is regulated by the surface ectoderm (Borycki et al., 1998). The differential expression patterns of genes downstream of the Shh pathway might thus provide a mechanism by which Shh signaling may be differentially interpreted by different parts of the somite: Gli would mediate sclerotome specification while Gli2/4 may transduce the myogenic effect of Shh signaling. Obviously, to validate such a model, one needs to show that the different Gli genes transactivate different target genes and thereby evoke different cellular responses. Similar mechanisms may occur in mammals because the three Gli genes isolated so far exhibit expression patterns similar to their quail homologs during early development. Moreover, simple or double mutations of these genes lead to abnormal skeletal differentiation in the mutant mice (Hui et al., 1994; MO et aZ., 1997; Platt et aZ., 1997). e. Cis-Regulation of the Myogenic Factors Activity The onset of myogenesis is controled by the bHLH transcription factors, MyoD and myf-5, which, at the trunk level, act downstream of the Pax-3 transcription factor (Maroto et al., 1997; Tajbakhsh et aZ., 1997). Both MyoD and myf5 bind to the ubiquitous bHLH El2 protein, and subsequently this heterodimer transactivates muscle specific markers (reviewed in Weintraub et aZ., 1991). Two related murine bHLH genes, MTwist and Dermo-1, along with members of the Id family, encode HLH proteins that lack the basic domain essential for specific DNA binding (Benezra et al., 1990; Christy et aZ., 1991; Li et al., 1995; Riechmann et al., 1994; Sun et al., 1991; Wolf et al., 1991). These proteins exhibit similar but distinct HLH sequences, expression patterns, and activities. Unlike other bHLH proteins, they seem to inhibit myogenesis in a competitive manner by binding to El2 and/or the DNA binding site of myogenic factors but without transactivating muscle gene transcription (Benezra et al., 1990; Hebrok et al., 1994; Jen et al., 1992; Li et aZ., 1995; Melnikova and Christy, 1996; Spicer et al., 1996). Their transcripts are detected in the PSM, epithelial somites, and are later excluded from the myotome, but they persist in the sclerotome and dermomyotome (Evans and O’Brien, 1993; Fuchtbauer, 1995; Li et aZ., 1995; Stoetzel et aZ., 1995; Wang et al., 1992; Wolf et aZ., 1991). Therefore, they appear to be good candidates, first, to block precocious myogenic induction in the PSM and, second, to restrict myogenic induction to the myotome. Nevertheless,

37

SOMITOGENESIS

targeted disruption of the Mtwist gene does not lead to ectopic myogenesis (Chen and Behringer, 1995), although this may be due to functional redun-. dancy. Another controversial aspect of a putative early inhibitory role for Mtwist is that, although its transcripts are present in the PSM and in epithelial somites, no protein could be detected in these tissues until somites differentiate (Gitelman, 1997). Another protein that is likely to play a role in the spatial restriction of myogenesis to the myotome compartment is I-mf, a mouse pratein of unknown structure which is strongly expressed in the sclerotome (Chen et aZ., 1996). I-mf is capable of repressing myogenesis. This is likely to be due either to retaining the MyoD family members in the cytoplasm by binding to them and masking their nuclear localization signal or by interfering with their DNA binding activities (Chen et al, 1996). The differential expression domains of Dermo-1, Mtwist, Id proteins, and I-mf in the sclerotome and/ or dermomyotome may explain in part how somitic compartments respond differently to axial signaling mediated by Shh and Wm. In summary, extrinsic cues emanating from structures surrounding the somites pattern paraxial mesoderm along the dorso-ventral axis, thereby segregating the different somitic lineages. Establishment of the mediolateral axis results in creating, within each of the three lineages, medial and lateral compartments, which also differ in terms of differentiation modalities and cell fate.

VI. Mediodateral

Polarity

of the Atian

Somite

The third axis set up in the developing somite is the medio-lateral axis. It is discussed here in a separate section although it is’apparent that the establishment of the dorso-ventral and medio-lateral axes are interconnected because these processes implicate similar molecular pathways and tissular interactions. Whereas the medio-Iateral polarity of the dermomyotome compartments has been extensively examined, data are not yet clear about medio-lateral polarity within the sclerotome compartment.

A. Two Different

Myogenic

Lineages

Arise from Somites

1. Two Myogenic Lineages Emerge from Different Semitic Compartments

In 1895, Fischel described that muscles of the back arise from the myotome, and that the ventral edge of the dermomyotome generates abdominal mus-

38

HIRSINGER

ET AL.

cles while migrating cells from the lateral somite colonize the limb bud to form appendicular muscles. This has been demonstrated by replacing medial chick halves with their quail counterparts and vice versa (Ordahl and Le Douarin, 1992; Ordahl and Williams, 1998). The axial skeleton and epaxial myogenic lineage, which will give rise to the paravertebral musculature, arise from the medial half of the somite, whereas the hypaxial myogenic lineage, which will generate muscles of the body wall (abdominal and intercostal muscles) and limbs, emerges from the lateral somitic moiety.

2. Muscular Development Is Different in the Epaxial and Hypaxial Lineages at the Bra&al Level

Cells of these two lineages undergo muscular differentiation according to different modalities. Epaxial precursors start differentiating in situ in the somite and activate myogenic genes as early as myotome formation (Buckingham, 1992; Pownall and Emerson, 1992). In contrast, differentiation of hypaxial precursors is delayed until they have migrated from the somite to their final location and activation of myogenic genes occurs 48 hours later than in the epaxial lineage, when muscle masses start to condense. Moreover, the epaxial muscles initially differentiate as mononuclear myocytes (Keynes and Stern, 1988) unlike the multinucleate hypaxial muscles (Rutz et al, 1982). Last, each somitic half seems to be differentially dependent upon trophic factors released from axial structures. Neural tube and notochord ablation are followed by necrosis within the myotome and sclerotome and the subsequent absence of axial muscles and skeleton. In contrast, the hypaxial musculature develops normally and therefore appears to be trophically independent of the axial structures (Asakura and Tapscott, 1998; Rong et aZ., 1992; Teillet et al, 1998; Teillet and Le Douarin, 1983). Another important difference between the two lineages is that epaxial muscles are innervated by the dorsal ramus of the spinal nerve, whereas the hypaxial muscles are innervated by the ventral ramus (Ordahl, 1993). Because hypaxial cells differentiate 48 hr later than their epaxial neighbors and only after having reached their final destination, the cellular readout of a lateral identity might consist of delaying overt differentiation of committed cells (Cossu et al., 1996; Pourquie et al., 1995, 1996) until they have reached their final destination. While these different developmental modalities occur at the limb level, the same is not strictly true at the thoracic and abdominal levels. In these regions, the hypaxial musculature forms from a ventral myotome, this being a mirror-image of the dorsal myotome (Christ et al, 1983) and differentiation modalities are similar, although delayed, to that of the epaxial lineage (Christ and Ordahl, 1995). Therefore the distinction between epaxial and hypaxial lineages is much less obvious at the thoracic

SOMiTOGEMESlS

39

and abdominal levels compared to the brachial ones (Ordahl and Williams, 1998). In conclusion, the two somitic lineages exhibit a different developmental history in terms of responsivity, origin, and fate. One might wonder whether these two compartments are specified differently because the cells that populate each of them are initially different or because an initially homogenous paraxial mesoderm becomes patterned along the medio-lateral axis by extrinsic cues. 3. Epaxial and Hypaxial Muscle Cells Have a Distinct Origin during Gastrulation To explore the first hypothesis, cell lineage analyses have been conducted. Marking cells in the Hensen’s node region with various techniques first led Pasteels (1937) and then Selleck and Stern (1991) to the conclusion that medial and lateral compartments in the somite originate from distinct populations of gastrulating cells. The medial compartment has been shown to derive from cells ingressing in the node itself while the progeny of cells ingressing at the level of the anterior primitive streak, just caudal to the node, populate the lateral compartment. This view was challenged by Schoenwolf and collaborators (1992b) who found that somitic cells derive exclusively from progenitors ingressing at the level of the primitive streak. Nevertheless, a more recent study confirms that medial and lateral compartments derive from cell populations emerging from different regions of the primitive streak which subsequently migrate along distinct routes (Psycboyos and Stern, 1996a). 4. Semitic Medio-lateral External Cues

Patterning

Is Under the Control of

To investigate somite plasticity along the medio-lateral axis, switch-graft experiments have been conducted in which medial halves of the last-formed somites are replaced with lateral halves and vice versa (Ordahl and Le Douarin, 1992). The manipulated somites subsequently exhibit normal development. Even more dramatically, when PSM is grafted inside the lateral plate, it adopts a lateral plate fate (Tonegawa et al., 1997). Therefore, similar to what has been shown for the establishment of dorso-ventral polarity, external cues are involved in the establishment of somitic mediolateral polarity. Likewise, this polarity is not yet determined at the level of the PSM or the last-formed somites. Nevertheless, unlike the dorsoventral polarity, both intrinsic and extrinsic mechanisms seem to be involved in establishing the medio-lateral axis. Indeed, even though cells of the PSM are not yet allocated to a dorsal or a ventral compartment, they are already

48

HIRSINGER

ET AL.

leading to the definition of eight major somitic compartments (Fig. 2). Each compartment gives rise to distinct sublineages such as epaxial and hypaxial muscles, dermis, anterior and posterior part of the vertebra, and so on. Therefore, a cell reads its position by assessing and integrating the different patterning signals. Interpreting these external cues in the context of its intrinsic characteristics, this cell will adopt a particular pathway of differentiation. Surrounding structures and their signals are crucial for somitic differentiation. However, the exact cellular mechanisms employed are still controversial: are these signals inductive or only permissive, allowing the expansion of previously committed cells as suggested by George-Weinstein and collaborators (1998)? Alternatively, both types of mechanisms may be in operation because they are not theoretically mutually exclusive.

FIG. 2 Schematic representation of the interactions between signaling molecules involved in somite medio-lateral and dorso-ventral patterning. Because the antero-posterior axis is not taken into account on this scheme, only four of the eight somitic compartments are shown. See text. EC, ectoderm; En, endoderm; LP, lateral plate; LS, lateral somite; MS, medial somite.

SOMITOGENESIS

41

axial mesoderm is exposed to only one of these structures, the developing somites are either totally medial or totally lateral in character, depending on the identity of this structure. 2. Ectoderm Plays a Pivotal Role in Liiing and Media-lateral Axes

Dorso-ventral

The role of the ectoderm in the specification of the somite medio-lateral axis has been studied in vitro in the mouse embryo. In mouse explant cultures, the ectoderm can promote myogenesis in the lateral somite by first inducing expression of the MyoD gene. This induction appears to involve the Wnt7b gene (Tajbakhsh et al, 1998). In the medial domain, myogenesis requires the neural tube and involves myf-5 activation first probably in response to Wntl signalling (Cossu et aZ., 1996; Tajbakhsh et al., 1998). In the chick embryo, ectoderm ablation in a lateral region leads to conversion of the dorso-lateral quadrant toward a ventral fate, unlike the medial compartment, which maintains its dorsal fate. This is most likely due to the presence of the neural tube. Conversely, ventral grafts of ectoderm induce ectopic dorsal markers in both cases. Moreover, these markers are of a lateral nature if the lateral ectoderm is placed ventro-laterally (Dietrich et al., 1998). These results highlight the fact that ectoderm plays a pivotal role linking the dorso-ventral and medio-lateral axes: by itself, it specifies the dorsal compartment and, in combination with lateral plate, the dorsolateral identity. Consequently, because ablation of the notochord/floor plate complex does not lead to the ventral expansion of dorso-lateral markers (Dietrich et al., 1998) one might wonder what structure specifies the ventral aspect of the lateral quadrant. Endoderm would be a suitable candidate although such an issue is difficult to address because of the lack of specific ventro-lateral markers. Further experiments in which permeable or impermeable obstacles were implanted between the somite and either of the adjacent structures demonstrated the diffusible nature of the signals emitted by both the neural tube and the lateral plate (Pourquie et al, 1996). In addition, in viva and in vitro studies (Miiller et aZ., 1996; Tonegawa et al, 1997) suggest that the lateralizing signal acts in a concentration-dependent manner. In contrast, the influence of the ectoderm seems to require cell-cell contact (Dietrich et al., 1998). 3. Specification

of Trunk Versus Limb Hypaxial

Lineages

As mentioned earlier, in addition to medio-lateral’patterning, the lateral somite is also regionalized in terms of thoracic and abdominal levels versus

42

HIRSINGER

ET AL.

brachial regions. Depending on the axial level, the hypaxial lineage exhibits various differentiation modalities and cell fates. At the thoracic and abdominal levels, hypaxial cells forming the trunk (abdominal and intercostal) muscles undergo an “epaxial-like” differentiation program unlike those at the limb level. Characterization of the chick homeobox gene Lbxl provided a specific marker for the migrating hypaxial lineage (Dietrich et al., 1998; Mennerich et al, 1998) (i.e., for the cells that disperse and migrate out from somites to form the limb musculature). The limb-level identity of the hypaxial cells has been shown to be conferred by the lateral plate (Kieny, 1960). Indeed, grafts of limb-level lateral plate to another axial level result in the outgrowth of an extra-limb at this level (Kieny, 1960) and to the ectopic induction of Lbxl (Dietrich et al., 1998). Whether Lbxl behaves as a marker or as a key regulator in specifying this limb subpopulation has still to be determined. In mice, an analogous mechanism might operate. It has been proposed from analyses of homozygous null mutant mice that the myogenic factors lMyoD and myf-5 differentially regulate the development of limb versus trunk skeletal muscle respectively (Kablar et al., 1997; Ordahl and Williams, 1998). Cells that first activate myf-5, whose expression depends upon neural tube signals, would differentiate into “epaxial-like” muscles (paraspinal, abdominal, and intercostal muscles), whereas those that first activate MyoD expression, which is regulated by the ectoderm, form limb muscles (Cossu et al, 1996; Kablar et aZ., 1997; Ordahl and Williams, 1998). From these data, it appears that three muscle populations arise from the combination of medio-lateral polarity and antero-posterior regionalization: the epaxial paraspinal muscles, the abdominal and intercostal muscles that have a hypaxial origin but an epaxial-like development and the hypaxial limb muscles.

C. BMPs, Wnt, Shh, and Noggin Interact Somites along the Medio-lateral Axis

to Pattern

1. BMP4 Mediates the Lateral Plate Effect Via a Concentration-Dependent Mechanism

The first molecule shown to play a role in medio-lateral patterning was BMP4, a member of the TGF-fi family (Pourquie et al., 1996). This secreted molecule is highly expressed in the lateral plate and has been previously shown to play crucial roles in inductive interactions during embryogenesis in both vertebrates and invertebrates (Hogan, 1996). It therefore exhibits all the characteristics expected of a candidate molecule that would mediate lateral plate signaling. To test its role, BMP4-expressing cells were grafted

43

SOMITOGENESIS

medially between the neural tube and the somite. Expression of the dorsomedial marker MyoD and the ventro-medial marker Pax-1 were lost, whereas that of the lateral marker Sim-1 was extended medially as if the whole somite was lateralized (Dietrich et al., 1998; Hirsinger et al., 1997; Pourquie et al., 1996). Furthermore, in vitro experiments show that BMP4 blocks the medializing signal delivered by axial structures (Reshef et al., 1998). Unfortunately, the study of the BMP4 null mouse is not very informative because most mutant mice arrest development at the egg cyliader stage and therefore show little or no mesoderm formation (Winnier et nZ., 1995). Therefore BMP4 activity seems to account for most of the lateral plate effects described earlier. Whereas BMP4 can induce Siml expression in the lateral somite, it is unable to mimick the Lbxl gene inducing activity of the limb-level lateral plate (Dietrich et aZ., 1998). This is, however, not surprising considering its homogenous expression pattern along the rostro-caudal axis. Candidate molecules such as the SGF/HGF also failed to induce ectopic Lbxl expression in the paraxial mesoderm (Mennerich et al., 1998). To define further the mechanistic details of lateralization, a quantitative analysis of BMP4 effects has been conducted. When COS cells, strongly expressing BMP4, are grafted into the unsegmented mesoderm, this tissue adopts a lateral plate identity (i.e., Pax-3 and Sim-l expression are lost) and cytokeratin expression, a marker of the lateral plate, is activated; when BMP4 cells are more and more diluted with control cells, the paraxial mesoderm is less and less lateralized. The minimal perturbation corresponds to a somite having two lateral halves (Tonegawa et aZ,, 1997). These results show that BMB4 acts in a concentration-dependent manner. To decide whether BMP4 acts as a true morphogeu, it must in addition exhibit a direct 1ongArange effect. Xenopus studies suggest that BMP4 acts directly but diffuses poorly (Dosch et d., 1997; Jones et al, !996,199&). Its Dvosophila countepart, Decapentaplegic, is however highly diffusible (Lecuit et al., 1996; Nellen et al, 1996). This issue is therefore still open to debate. 2. Noggin Antagonizes

the BMP4 Lateralizing

Effect

In addition to a BMP4 concentration-dependent effect, previous studies have shown the necessity of a medializing signal, antagonizing BMP4 lateralizing signal, in order to specify a medial and a lateral identity within the somite. A molecular candidate to mediate this medializing signal is Noggin. Noggin is a secreted protein (Smith and Harland, 1992), which has been shown to bind to BMP4 and thereby prevent itfrom binding to its receptor (Zimmerman et al., 1996). In this way, it acts as a BMP4 antagonist. Among other sites, it is dynamically expressed in the paraxial mesoderm and its presumptive territory in locations compatible with a putative role as a

44

HIRSINGERET AL.

BMP4 antagonist (Capdevila and Johnson, 1998; Hirsinger et al., 1997; Marcelle et al, 1997; Tonegawa and Takahashi, 1998). To examine whether Noggin blocks the BMP4 lateralizing signal in somites, the protein has been ectopically expressed in the par-axial mesoderm either by grafting Noggin expressing cells (Hirsinger et aZ., 1997; Marcelle et al., 1997; Reshef et al., 1998; Tonegawa and Takahashi, 1998) or by retroviral-mediated overexpression (Capdevila and Johnson, 1998). In all cases, ectopic Noggin expression converts the lateral compartment into a medial one, as evidenced by the down-regulation of Sim-1 and the lateral expansion of lMyoD and Pax-l expression domains. At an earlier developmental stage, when Noggin-expressing cells are implanted in the presumptive region of the lateral plate, ectopic somites are formed, suggesting that the lateral plate fate has been converted to a paraxial mesoderm fate by blocking the BMPClateralizing signal (Tonegawa and Takahashi, 1998). Conversely, when Noggin expression is abolished in homozygous null mice, the somitic phenotype shows a medial extension of the Sim-1 expression domain (McMahon et al., 1998). These data indicate that the whole somite becomes lateralized. These results taken together with in vitro studies (Reshef et d., 1998) are consistent with the hypothesis that Noggin acts as an antagonist of BMP4 signaling in somite differentiation. Initially, the antagonistic interactions between Noggin and BMP4 lead to the definition of the boundary between paraxial and lateral mesoderm; as development proceeds, this mechanism is progressively involved in specifying lateral and medial compartments within the somite. Unlike the case with BMP4, it is well-established that in Xenopus Noggin diffusion is long range and its effect is direct (Dosch et aZ., 1997; Jones and Smith, 1998). Therefore the BMP4 activity gradient is more likely to be achieved by the long range effects of its antagonists than by diffusion of the BMP4 protein itself (Jones and Smith, 1998). To date, it is unclear whether Noggin has activities independent of this antagonistic role. No Noggin receptor has been identified so far. Results obtained in vitro are contradictory since, Reshef and colleagues (1998) find that naive somites cultured with Noggin are not driven toward any particular differentiation pathways, whereas McMahon and collaborators (1998) argue that Noggin can induce Pax-l expression in presomitic explants. Different experimental procedures could account for these discrepancies. As described earlier, PSM is temporally delayed in its response to the prior presence of differentiation signals. As previously mentioned, activation of Notch signaling, regulation of the Shh pathway by its downstream genes, or presence of Wnt antagonists are strong candidates for the possible mechanisms that could account for this phenomenon. Blockade of BMP4 signal by Noggin might be another factor. Indeed, at the level of unseg-

45

SOMITOGENESIS

mented mesoderm, Noggin is expressed in the lateral edge of the paraxi mesoderm, in apposition to the BMP4 expression domain (Hirsinger et al., unpublished data; Capdevila and Johnson, 1998; Marcelle et al., 1997; Tonegawa and Takahashi, 1998). Therefore, Noggin activity in this location could prevent the PSM from responding to signals emanating from the lateral plate and additionally act to maintain the boundary between paraxial mesoderm and lateral mesoderm (Tonegawa and Takahashi, 1998). In addition, it has been shown in Xenopus and mouse that the DAN family member Cerberus, in addition to antagonising Wnt signaling, blocks BMP4 activity, by binding to it (Biben et al., 1998; Hsu et al., 1998). Murine cerberus is found in the PSM and the two last formed somites (Bela et al., 1997; Biben et al., 1998; Shawlot et al., 1998). Therefore, presence of Noggin and Cerberus in these locations could provide an additional molecular mechanism for the delayed response of the PSM toward environmental signals. 3. Wnt and Shh Pathways Antagonize Via Noggin

BMP4 Signaling

The previous results suggest that Noggin plays a role in differentiation events, such as myogenesis and sclerogenesis, that are controlled by the Wnt and Shh pathways. Therefore the possibility that these two factors might also antagonize BMP4 has been investigated. Furthermore, these genes are medially expressed, in the dorsal neural tubelectoderm and in the notochordlfloor plate, respectively. Retroviral-mediated ectopic expression of Shh (Capdevila et al., 1998; Johnson et al., 1994) and Wnt-1 (Capdevila et al., 1998; Johnson er al., 1994) or grafts of Shh or Wnt-1 expressing cells (Hirsinger et al., 1997; Marcelle et al., 1997) alter the normal patterning along the medio-lateral axis. In all cases, medial markers (MyoD, Wnt-11) are laterally expanded while the lateral marker Sim-1 is lost. Accordingly, the phenotypes of embryos mutant either for Shh or for Wnt-lAVnt3a support these data (Chiang et al., 1996; Ikeya and Takada, 1998): the lateral somitic compartment is expanded at the expense of the medial compartment. Therefore, Shh and Wnt-1 appear to exhibit the same antagonistic activities as the structures in which they are expressed and as Noggin itself, As a consequence, one might wonder whether these factors exert their effect independently of Noggin or if they are part of the same pathway. Grafts of Shh expressing celis was shown to induce Noggin expression ectopically whereas cells producing Wnt-1 mimick the influence of the dorsal neural tube on Noggin expression (Hirsinger ef aZ., 1997). Furthermore, Wnt-1 protein is capable of inducing Noggin expression in naive somites in vitro (Reshef et AZ.,1998). Accordingly, the major site of Noggin

46

HIPSINGERET AL.

expression in the somite is the medial dermomyotome, in close vicinity to the Writ-1 expressing domain in the dorsal neural tube; Noggin is also transiently found colocalized with Shh in the notochord and floor plate. Therefore Noggin may act downstream of both the Wnt and Shh pathways where it functions to antagonize the BMP4 lateralizing signal. In parallel, other lines of evidence suggest that Noggin and Shh might also act synergistically in the specification of the sclerotomal lineage (McMahon et al., 1998). Shh is able to activate Pax-l expression in cultured PSM; however, in conjunction with Noggin, lower levels of Shh are required to obtain the same effect. Some aspects of the Noggin mutant mouse phenotype provide further support for this hypothesis (McMahon et aZ., 1998). In conclusion, Noggin, present in the medial dermomyotome, seems to be downstream of both the axial Shh and Wnt-1 pathways and appears to act synergistically with Shh to block the effect of lateral plate-derived BMP4 signaling. However, Noggin is not the only BMP4 antagonist expressed in somites. A Follistatin-related gene, Flik, has been cloned in chick (Pate1 et al., 1996) and is expressed in the medial dermomyotome (Amthor et al., 1996). Similar to Noggin, Flik expression is regulated by the neural tube (Amthor et cd., 1996). Because Follistatin has been shown to bind to BMP4 and thereby block its activity (Fainsod et al., 1997), a similar role might be expected for Flik. Another likely candidate is the secreted protein Chordin, which is expressed in the notochord (Sasai et al, 1994) and has been shown to diffuse long distances (Jones and Smith, 1998). Like Noggin and Follistatin, it acts to block BMP4 activity by binding to it (Piccolo et aZ., 1996). These secreted molecules thus exhibit features compatible with a role in defining a medial somitic identity by blocking a BMP4 lateralizing signal. 4. Dorsal Neural Tube-Derived Medial Semitic Identity

BMP4 Can Indirectly

Promote

BMP4 expression in the dorsal neural tube has been shown to play a role in the dorso-ventral patterning of the neural tube (Liem et aZ., 1993, but one might question how the presence of BMP4 in this location can be reconciled with its lateralizing activity on the somite. Firstly, the somite and the neural tube are each surrounded by a heparin-rich basal lamina, which is likely to inhibit movement of heparin binding factors like BMP4. Secondly, Noggin is expressed in the adjacent medial dermomyotome, adjacent to the dorsal neural tube, and is thus correctly spatially located for it to block any BMP4 effect on the paraxial mesoderm. Third, Marcelle and collaborators (1997) have shown in vivo that injecting cells, that express BMP4 into the neural tube, can activate the ectopic expression of Wnt-1 in the dorsal neural tube. This activation then positively regulates the

SOMITOGENESIS

47

expression of a medial dermomyotomal marker, Writ-11 (Marcelle et aZ., 1997) and also promotes Noggin expression in that region (Hirsinger et al., 1997). Thus, in an indirect manner, BMP4 couldpromote the specification of a medial compartment while simultaneously being prevented from directly lateralizing the somite by the expression of Noggin in this region. 5. The Origin of Ribs and the Lateral Sclerotome It is noteworthy that the medio-lateral polarity of the dermomyotome has been extensively studied’but little is known about the existence of such a regionalization within the sclerotome. For example, the existence of a lateral sclerotome is not yet well established, although this compartment appears to be regionalized in terms of developmental fate and gene expression patterns. Its medial region, expressing high levels of Pax-l, gives rise to the vertebral body and intervertebral discs, whereas the neural arches and proximal ribs probably derive from the lateral sclerotome, which expresses Sim-1 and high levels of Pax-9 (Balling et aZ., 1996; Pourquie et al., 1996). However, whereas Sim-1 expression is promoted by lateralizing cues and is down-regulated by medializing signals (Pourquie et al, 1996) as would be expected for a marker of the lateral compartment, both Pax genes are positively regulated by medializing signals and negatively regulated by lateralizing ones (Hirsinger et aZ., 1997; Mtiller et aZ., 1996), thereby suggesting that they both are markers of the medial compartment. Fate mapping of half somite derivatives in the chick have led to the suggestion that ribs are derived from the lateral somitic half (Christ and Ordahl, 1995). Surprisingly, recent fate mapping experiments have demonstrated that in chick the distal rib originates in the’ dermomyotome (Kato and Aoyama, 1998). Ablation of progressively more lateral regions of the dermomyotome affects progressively more distal parts of the ribs. Therefore ribs and intercostal muscles that interact in the adult animal derive from the same embryological &ucture, the’:dermomyotome, whereas the rest of the axial skeleton derives from the sclerotome. Accordingly, ~2~$-5 knockout mice exhibit defects both in distal rib formation and in the development of intercostal muscles (Braun et al., 1992) suggesting that,‘in addition to their closely related anatomical origin, their subsequent development might be regulated by the same molecular cascades.

VII.

Conclusion

Taking into account all these results, it appears that tissues surrounding the somites build a complex network of synergistic and antagonistic interactions

48

HIRSINGER

ET AL.

leading to the definition of eight major somitic compartments (Fig. 2). Each compartment gives rise to distinct sublineages such as epaxial and hypaxial muscles, dermis, anterior and posterior part of the vertebra, and so on. Therefore, a cell reads its position by assessing and integrating the different patterning signals. Interpreting these external cues in the context of its intrinsic characteristics, this cell will adopt a particular pathway of differentiation. Surrounding structures and their signals are crucial for somitic differentiation. However, the exact cellular mechanisms employed are still controversial: are these signals inductive or only permissive, allowing the expansion of previously committed cells as suggested by George-Weinstein and collaborators (1998)? Alternatively, both types of mechanisms may be in operation because they are not theoretically mutually exclusive.

FIG. 2 Schematic representation of the interactions between signaling molecules involved in somite medio-lateral and dorso-ventral patterning. Because the antero-posterior axis is not taken into account on this scheme, only four of the eight somitic compartments are shown. See text. EC, ectoderm; En, endoderm; LP, lateral plate; LS, lateral somite; MS, medial somite.

SOMITOGENESIS

49

One form of permissive signal may be, for example, to provide trophic support. It is known that the neural tubelnotochord complex delivers trophic signals necessary for the survival of the medial somite and for the maintenance and amplification of myotomal and sclerotomal cells (Asakura and Tapscott, 1998; Rong et al, 1992; Strudel, 1955; Teillet et aL, 1998; Teillet and Le Douarin, 1983). When the complex is ablated, Pax-1 and MyoD expression seems nevertheless transiently induced, questioning any role for these structures in the initiation of differentiation markers (Teillet et at., 1998). Ectoderm can compensate for the loss of axial tissue in the induction of n/lyoD (Dietrich et aZ., 1997), but the case of Pax-I remains unresolved. It has recently been shown that the neural tube/notochord trophic effect is mediated by Shh (Teillet et al., 1998). In addition, Shh has been shown to induce cell proliferation in the retina and lung, in the PSM, and in committed skeletal muscles (Bellusci et al., 1997; Duprez et al., 1998; Fan et al., 1995; Jensen and Wallace, 1997). However, the Shh null mice which show transient expression of Pax-l suggest that this signal, while mediating some notochordal properties such as trophic and maintenance activities, is not able to substitute for all the activities. The question is now to decipher whether the activity of this molecule in the notochord and the Aoor plate can be separated from that of other signals in these tissues and whether it can exert, in addition, any specific inductive properties.

Acknowledgments We thank Drs Kim Dale, Mike McGrew, Miguel Maroto, and Pr. Monte Westerfield for their helpful comments and their critical reading of the manuscript. Financiai support for our own work on somitogenesis was provided by the Centre National de la Recherche Scienthique (CNRS), the Association Fraqaise contre les myopathies (AFM), the Association pour la Recherche contre le Cancer (ARC), and the Fondation pour la Recherche Medicale (FRM).

References Amthor, H., Connolly, D., Patel, K., Brand-Saber?, B., Wiikinson, D. G., Cooke, J., and Christ, B. (1996). The expression and regulation of follistatin and a foilistatin-like gene during avian somite compartmentalization and myogenesis. Dev Bid. 178,343-362. Ang, S. L., and Rossant, J. (1994). HNF-3 beta is essential for node and notochord formation in mouse development. Cell 78,561-574. Aoyama, H. (1993). Developmental plasticity of the prospective dermatome and the prospective sclerotome region of an aviau somite. Dev. Growth. Differ. 35, 507-519. Aoyama, H., and Asamoto, K. (1988). Determination of somite cells: Independence of cell differentiation and morphogenesis. DeveZopment 104, 15-28. Artavanis-Tsakonas, S., Rand, M. D., and Lake, R. J. (1999). Notch signaling: Cell fate control and signal integration in development. Science ZS4,770-776.

50

HIRSINGER

ET AL.

Asakura, A., and Tapscott, S. J. (1998). Apoptosis of epaxial myotome in Danforth’s short-tail (Sd) mice in somites that form following notochord degeneration. Dev. BioZ. 203,276-289. Bagnall, K. M. (1992). The migration and distribution of somite cells after labelling with the carbocyanine dye, Dil: The relationship of this distribution to segmentation in the vertebrate body. Anat. Embryol. (Berl.) 185, 317-324. Bagnall, K. M., and Sanders, E. J. (1989). The binding pattern of peanut lectin associated with sclerotome migration and the formation of the vertebral axis in the chick embryo. Anat. Embryol. (Berl.) 180, 505-513. Bagnall, K. M., Higgins, S. J., and Sanders, E. J. (1988). The contribution made by a single somite to the vertebral column: Experimental evidence in support of resegmentation using the chick-quail chimaera model. Development 103, 69-85. Bagnall, K. M., Sanders, E. J., and Berdan, R. C. (1992). Communication compartments in the axial mesoderm of the chick embryo. Anat. Embryol. (Berl.) 186, 195-204. Balling, R., Helwig, U., Nadeau, J., Neubiiser, A., Schmahl, W., and Imai, K. (1996). Pax genes and skeletal development. Ann. N.Y. Acad. Sci. 78.5, 27-33. Barnes, G. L., Alexander, P. G., Hsu, C. W., Mariani, B. D., and Tuan, R. S. (1997). Cloning and characterization of chicken Paraxis: A regulator of paraxial mesoderm development and somite formation. Dev. Biol. 189, 95-111. Baur, R. (1969). Zum problem der neugliederung der wirbelsaule. Acta Amt. 72, 321-356. Baylies, M. K., Bate, M., and Ruiz Gomez, M. (1998). Myogenesis: A view from Drosophila. Cell 93, 921-927.

Beddington, R. S. (1982). An autoradiographic analysis of tissue potency in different regions of the embryonic ectoderm during gastrulation in the mouse. J. Embryol. Exp. Morphol. 69,265-285.

Bellaiche, Y., The, I., and Perrimon, N. (1998). Tout-velu is a Drosophila homologue of the putative tumour suppressor EXT-1 and is needed for Hh diffusion [see comments]. Nature 394,85-88.

Bellairs, R., Curtis, A. S., and Sanders, E. J. (1978). Cell adhesiveness and embryonic differentiation. J. Embryol. Exp. Morphol. 46, 207-213. Bellusci, S., Furuta, Y., Rush, M. G., Henderson, R., Winnier, G., and Hogan, B. L. (1997). Involvement of Sonic hedgehog (Shh) in mouse embryonic lung growth and morphogenesis. Development

K&53-63.

Belo, J. A., Bouwmeester, T., Leyns, L., Kertesz, N., Gallo, M., Follettie, M., and De Robertis, E. M. (1997). Cerberus-like is a secreted factor with neutralizing activity expressed in the anterior primitive endoderm of the mouse gastrula. Mech. Dev. 68,45-57. Benezra, R., Davis, R. L., Lockshon, D., Turner, D. L., and Weintraub, H. (1990). The protein Id: A negative regulator of helix-loop-helix DNA binding proteins. Cell 61,49-59. Bettenhausen, B., Hrabe de Angelis, M., Simon, D., Guenet, J. L., and Gossler, A. (1995). Transient and restricted expression during mouse embryogenesis of DID, a mu&e gene closely related to Drosophila Delta. Development 121, 2407-2418. Biben, C., Stanley, E., Fabri, L., Kotecha, S., Rhinn, M., Drinkwater, C., Lah, M., Wang, C. C., Nash, A., Hilton, D., Ang, S. L., Mohun, T., and Harvey, R. P. (1998). Murine cerberus homologue mCer-1: A candidate anterior patterning molecule [In Process Citation]. Dev. BioZ. 194,135-151. Blair, S. S. (1997). Limb Development: Marginal fringe benefits. Curr. Biol. 7, 686-690. Blanar, M. A., Crossley, P. H., Peters, K. G., Steingrimsson, E., Copeland, N. G., Jenkins, N. A., Martin, G. R., and Rutter, W. J. (1995). Meso 1, a basic-helix-loop-helix protein involved in mammalian presomitic mesoderm development. Proc. Nutl. Acad. Sci. U.S.A. 92,5870-5874.

Blechschmidt, E. (1957). Die entwicklungsbewegungen der somiten und ihre bedeuntung ftir die gliederung der wirbelsaule. 2. Anat. Entwicklungsgesch. 3.20, 150-172.

SClMITOGEiESIS

51

Bober, E., Brand-Saber& B., Ebensperger, C., Wilting, J,, Balling, R., Paterson, B. M., Arnold, H. H., and Christ, B. (1994). Initial steps of myogenesis in somites are independent of influence from axial structures. Development X&3073-3082. Borman, W. H., and Yorde, D. E. (1994a). Analysis of chick somite myogenesis by in situ confocal microscopy of de&n expression. J. Histochem. Cytochem. 42,265-272. Borman, W. H., and Yorde, D. E. (1994b). Barrier inhibition of a temporal neuraxial influence on early chick somitic myogenesis. Dev. Dyn. 200, 68-78. Borycki, A. G., Mendham, L., and Emerson, C. (1998). Control of somite patterning by Sonic hedgehog and its downstream signal response genes. Development I&777-790. Borycki, A. G., Strunk, K. E., Savary, R., and Emersan, C. P. J. (1997). Distinct signal/ response mechanisms regulate pax1 and QmyoD activation in sclerotomal and myotomal lineages of quail somites. Dev. Biol. 185,185-200. Bouwmeester, T., Kim, S., Sasai, Y., Lu, B., and De Robertis, E. M. (1996). Cerberus is a head-inducing secreted factor expressed in the anterior endoderm of Spemann’s organizer. Nature

382,595-601.

Brand-Saberi, B., Ebensperger, C., Wilting, J., Balling, R., and Christ, B. (1993). The ventralizing effect of the notochord on somite differentiation in chick embryos. Anat. EmbryoE.

(Bed.) 188, 239-245. Braun, T., Rudnicki, M. A., Arnold, H. H., and Jaenisch, R. (1992). Targeted inactivation of the muscle regulatory gene Myf-5 results in abnormal rib development and perinatal death. Cell 71, 369-382.

Brill, G.,‘Kahane, N., Carmeli, C., von Schack, D., Barde, Y. A., and Kalcheim, C. (1995). Epithelial-mesenchymal conversion of dermatome progenitors requires neural tube-derived signals: Characterization of the role of Neurotrophin-3. Develapment l.Zl, 2583-2594. Bronner-Fraser, M., and Stern, C. (1991). Effects of mesoderma1 tissues on avian neural crest cell migration. Dev. Biol. 143, 213-217. Buchberger, A., Seidl, K., Klein, C., Eberhardt, H., and Arnold, H. H. (1998). cMeso-1, a novel bHLH transcription factor, is involved in somite formation in chicken embryos. Dev. Biol.

199,201-215.

Buckingham, M. (1992). Making muscle in mammals. Trends Genet. 8, 144-148. Buffinger, N., and Stockdale, F. E. (1994). Myogenic specification in somites: induction by axial structures. Development l20,1443-1452. Buffinger, N., and Stockdale, F. E. (1995). Myogenic specification of somites is mediated by diffusible factors. Dev. Biol. 169, 96-108. Burgess, pi., Cserjesi, P., Ligon, K. L., and Olson, E. N. (1995). Paraxis: A basic helixloop-helix protein expressed in paraxial mesoderm and developing somites. Dev. Biol.

168,296-306. Burgess, R., Rawls, A., Brown, D., Bradley, A., and Olson, E. N. (1996). Requirement of the paraxis gene for somite formation and musculoskeletal patterning. Nature 384, 570-573. Cadigan, K. M., and Nusse, R. (1997). Wnt signaling: a common theme in animal development. Genes

Dev.

l&3286-3305.

Capdeviia, J., and Johnson, R. L. (1998). Endogenous and ectopic expression of Noggin suggests a conserved mechanism for regulation of BMP function during limb and somite patterning. Dev. Biol. 197,205-217. Capdevila, J., Tabin, C., and Johnson, R. L. (1998). Control of dorsoventral somite patterning by Wnt-1 and beta-catenin. Dev. Biol. 193, 182-194. Catala, M., Teillet, M. A., De Robertis, E. M., and Le Douarin, M. L. (1996). A spinal cord fate map in the avian embryo: While regressing, Hensen’s node lays down the notochord and floor plate thus joining the spinal cord lateral walls. Development X22,2599-2610. Catala, M., Teillet, M. A,, and Le Douarin, N. M. (1995). Organization and development of the tail bud analyzed with the quail-chick chimaera system. Me&. Dev. 51, 51-65.

52

HIRSINGERET AL.

Chapman, D. L., and Papaioannou, V. E. (1998). Three neural tubes in mouse embryos with mutations in the T-box gene Tbx6. Nature 391, 695-697. Chen, C. M., Kraut, N., Groudine, M., and Weintraub, H. (1996). I-mf, a novel myogenic repressor, interacts with members of the MyoD family. Cell 86, 731-741. Chen, Y., and Solursh, M. (1991). The determination of myogenic and cartilage cells in the early chick embryo and the modifying effects of retinoic acid. Roux’s Arch. Dev. Biol. 200, 162-171. Chen, Z. F., and Behringer, R. R. (1995). Twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev. 9,686-699. Chiang, C., Litingtung, Y., Lee, E., Young, K. E., Corder-r, J. L., Westphal, H., and Beachy, P. A. (1996). Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature 383, 407-413. Christ, B., and Ordahl, C. P. (1995). Early stages of chick somite development. Anat. Embryol. (BerL) 191, 381-396. Christ, B., and Wilting, J. (1992). From somites to vertebral column. Anat. Anz. 174, 23-32. Christ, B., Brand-Saberi, B., Grim, M., and Wilting, J. (1992). Local signaIling in dermomyotomal cell type specification. Anat. Embryol. (Berl.) 186, 505-510. Christ, B., Jacob, H. J., and Jacob, M. (1978). On the formation of the myotomes in avian embryos. An experimental and scanning electron microscope study. Experientia 34,514-516. Christ, B., Jacob, H. J., and Jacob, M. (1979). ijber gestaltungsfunktionen der somiten bei der entwicklung der korperwand von htihnerembryonen. Verb. Anat. Ges. 73,509-518. Christ, B., Jacob, M., and Jacob, H. J. (1983). On the origin and development of the ventrolatera1 abdominal muscles in the avian embryo. An experimental and ultrastructural study. Anat. EmbryoZ. (Berl.) 166,87-101. Christ, B., Schmidt, C., Huang, R., Wilting, J., and Brand-Saberi, B. (1998). Segmentation of the vertebrate body. Anat. Embryol. (Berl.) 197, l-8. Christy, B. A., Sanders, L. K., Lau, L. F., Copeland, N. G., Jenkins, N. A., and Nathans, D. (1991). An Id-related helix-loop-helix protein encoded by a growth factor-inducible gene. Proc. Natl. Acad. Sci. U.S.A. 88, 1815-1819. Conlon, R. A., Resume, A. G., and Rossant, J. (1995). Notch1 is required for the coordinate segmentation of somites. Development l21,1533-1545. Cooke, J., and Zeeman, E. C. (1976). A clock and wavefront model for control of the number of repeated structures during animal morphogenesis. .I. Theor. BioZ. 58,455-476. Cooper, G. W. (1965). Induction of somite chondrogenesis by cartilage and notochord: A correlation between inductive activity and specific stages of cytodifferentiation. Dev. BioZ. 12, X35-212. Cossu, G., Kelly, R., Di Donna, S., Vivarelli, E., and Buckingham, M. (1995). Myoblast differentiation during mammalian somitogenesis is dependent upon a community effect. Proc.

Natl.

Acad.

Sci. U.S.A.

92, 2254-2258.

Cossu, G., Kelly, R., Tajbakhsh, S., Di Donna, S., Vivarelli, E., and Buckingham, M. (1996). Activation of different myogenic pathways: myf-5 is induced by the neural tube and MyoD by the dorsal ectoderm in mouse paraxial mesoderm Development 122,429-437. Couly, G., and Le Douarin, N. M. (1988). The fate map of the cephalic neural primordium at the presomitic to the 3-somite stage in the avian embryo. Development 103 Suppl, 101-113. Couly, G., Coltey, P., and Le Douarin, N. M. (1993). The triple origin of skull in higher vertebrates: A study in quail-chick chimeras. Dev. Suppl. 117,409-429. Cremer, H., Lange, R., Christoph, A., Plomann, M., Vopper, G., Roes, J., Brown, R., Baldwin, S., Kraemer, P., Scheff, S., and et al. (1994). Inactivation of the N-CAM gene in mice results in size reduction of the olfactory bulb and deficits in spatial learning. Nature 367,455-459. de la Pompa, J. L., Wakeham, A., Correia, K. M., Samper, E., Brown, S., Aguilera, R. J., Nakano, T., Honjo, T., Mak, T. W., Rossant, J., and Conlon, R. A. (1997). Conservation of the Notch signalling pathway in mammalian neurogenesis. Development l24,1139-1148.

53

,SOMITOGF;NESIS

de1 Amo, F. F., Smith, D. E., Swiatek, P. J., Gendron-Maguire, M., Greenspan, R. J., McMahon, A. P., and Gridley, T. (1992). Expression pattern of Match, a mouse homolog of Drosophila Notch, suggests an important role in early postimplantation mouse development. Development 115,131-744. Denetclaw, W. F. J., Christ, B., and Ordahl, C. P. (1997). Location and growth of epaxial myotome precursor cells. Development lZ& 1601-1610. Dietrich, S., Schubert, F. R., and Gruss, P. (1993). Altered Pax gene expression in murine notochord mutants: The notochord is required to initiate and maintain ventral identity in the somite. Mech. Dev. 44,189-207. Dietrich, S., Schubert, F. R., and Lumsden, A. (1997). Control of dorsoventral pattern in the chick par-axial mesoderm. Development K&$3895-3908. Dietrich, S., Schubert, F. R., Healy, C., Sharpe, P. T., and Lumsden, A. (1998). Specification of the hypaxial musculature. Development 1252235-2249. Dockter, J. L,, and Ordahl, C. P. (1998). Determination of sclerotome to the cartilage fate. Development

X&2113-2124.

Dosch, R., Gawantka, V., Delius, H., Blumenstock, C., and Niehrs, C. (1997). Bmp-4 acts as a morphogen in dorsoventral mesoderm patterning in Xenopus. Development l24,2325-2334. Duband, J. L., Dufour, S., Hatta, K., Takeichi, M., Edelman, G. M., and Thiery, J. P. (1987). Adhesion molecules during somitogenesis in the avian embryo. J. Cell Biol. 104,1361-1374. Dunwoodie, S. L., Hemique, D., Harrison, S. M., and Beddington, R. S. P. (1997). Mouse D113: A novel divergent Delta gene which may complement the function of other Delta homologues during early pattern formation in the mouse embryo. DeveZopment 124,30653076. Duprez, D., Fournier-Thibault, C., and Le Douarin, N. (1998). Sonic Hedgehog induces proliferation of committed skeletal muscle cells in the chick limb. Development 125,495~505. Durbin, L., Brennan, C., Shiomi, K., Cooke, J., Barrios, A., Shanmugalingam, S., Guthrie, B., Lindberg, R., and Holder, N. (1998). Eph signaling is required for segmentation and differentiation of the somites. Genes Dev. 32, 3096-3109. Ebensperger, C., Wilting, J., Brand-Saber?, B., Mizutani, Y., Christ, B., Balling, R., and Koseki, H. (1995). Pax-l, a regulator of sclerotome development is induced by notochord and floor plate signals in avian embryos. Anat. Embryol. (Bed.) 191, 297-310. Echelard, Y., Epstein, D. J., St-Jacques, B., Shen, L., Mohler, J., McMahon, J. A., and McMahon, A. P. (1993). Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. CeEZ75,1417-1430, Evans, S. M., and O’Brien, T. X. (1993). Expression of the helix-loop-helix factor Id during mouse embryonic development. Dev. Biol. l59,485-499. Evrard, Y. A., Lun, Y., Aulehla, A., Gan, L., and Johnson, R. L. (1998). lunatic fringe is an essential mediator of somite segmentation and patterning. Nature 394, 377-381. Fainsod, A., Deissler, K., Yelm, R., Marom, K., Epstein, M., Pillemer, G., Steinbeisser, H., and Blum, M. (1997). The dorsalizing and neural inducing gene follistatin is an antagonist of BMP-4. Mech. Dev. 63, 39-50. Fan, C. M., and Tessier-Lavigne, M. (1994). Patterning of mammalian somites by surface ectoderm and notochord: Evidence for sclerotome induction by a hedgehog homolog. Cetl 79,1175-1186.

Fan, C. M., Lee, C. S., and Tessier-Lavigne, M. (1997). A role for WNT proteins in induction of dermomyotome. Dev. Bid. B&160-165. Fan, C. M., Porter, J. A., Chiang, C., Chang, D. T., Beachy, P. A., and Tessier-Lavigne, M. (1995). Long-range sclerotome induction by sonic hedgehog: direct role of the aminoterminal cleavage product and modulation by the cyclic AMP signaling pathway. Cell

f&457-4$5. Fiscbel, A. (1895). Zur entwicklung der ventralen rumpf- und extremitatenmuskulatur Vogel und saugetiere. Morph. Jb. 23, 544-561.

der

54

HIRSING@3

ET AL.

Fleming, R. J., Gu, Y., and Hukriede, N. A. (1997). Serrate-mediated activation of Notch is specifically blocked by the product of the gene fringe in the dorsal compartment of the Drosophila wing imaginal disc. Development l24, 2973-2981. Flower, M., and Grobstein, C. (1967). Interconvertibility of induced morphogenetic responses of mouse embryonic somites to notochord and ventral spinal cord. Dev. Biol. 15,193-205. Forsberg, H., Crozet, F., and Brown, N. A. (1998). Waves of mouse Lunatic fringe expression, in four-hour cycles at two-hour intervals, precede somite boundary formation. Curr. Biol. S, 1027-1030. Fortini, M. E., and Artavanis-Tsakonas, S. (1994). The suppressor of hairless protein participates in notch receptor signaling. Cell 79, 273-282. Fraser, R. C. (1960). Somite genesis in the chick. III. The role of induction. J. Exp. Zool. 145,151-167. Fraser, S., Keynes, R., and Lumsden, A. (1990). Segmentation in the chick embryo hindbrain is defined by cell lineage restrictions. Nature 344, 431-435. Fredette, B. J., Miller, J., and Ranscht, B. (1996). Inhibition of motor axon growth by Tcadherin substrata. Development lZ2,3163-3171. Fuchtbauer, E. M. (1995). Expression of M-twist during postimplantation development of the mouse. Dev. Dyn. 204,316-322. Furuta, Y., Ilic, D., Kanazawa, S., Takeda, N., Yamamoto, T., and Aizawa, S. (1995). Mesoderma1 defect in late phase of gastrulation by a targeted mutation of focal adhesion kinase, FAK. Oncogene 11,1989-1995. Gallera, J. (1966). [Evidence for the role of the ectoblast in the differentiation of somites in birds] Mise en evidence du role de l’ectoblaste dam la differentiation des somites chez les oiseaux. Rev. Suisse. Zool. 73, 492-503. Game& A. J., Brand-Saberi, B., and Christ, B. (1995). Halves of epithelial somites and segmental plate show distinct muscle differentiation behavior in vitro compared to entire somites and segmental plate. Dev. Biol. 172, 625-639. Garcia-Martinez, V., and Schoenwolf, G. C. (1992). Positional control of mesoderm movement and fate during avian gastrulation and neurulation. Dev. Dyn. 193, 249-256. Garcia-Martinez, V., Darnell, D. K., Lopez-Sanchez, C., Sosic, D., Olson, E. N., and Schoenwolf, G. C. (1997). State of commitment of prospective neural plate and prospective mesoderm in late gastrula/early neurula stages of avian embryos. Dev. Biol. 181, 102-11.5. George, E. L., Georges-Labouesse, E. N., Patel-King, R. S., Rayburn, H., and Hynes, R. 0. (1993). Defects in mesoderm, neural tube and vascular development in mouse embryos lacking fibronectin. Development 119, 1079-1091. George-Weinstein, M., Gerhart, J., Blitz, J., Simak, E., and Knudsen, K. A. (1997). N-cadherin promotes the commitment and differentiation of skeletal muscle precursor cells. Dev. Biol. l&5,14-24. George-Weinstein, M., Gerhart, J. V., Foti, G. J., and Lash, J. W. (1994). Maturation of myogenic and chondrogenic cells in the presomitic mesoderm of the chick embryo. Exp. Cell Res. 211, 263-274. George-Weinstein, M., Gerhart, J., Reed, R., Flynn, J., Callihan, B., Mattiacci, M., Miehle, C., Foti, G., Lash, J. W., and Weintraub, H. (1996). Skeletal myogenesis: the preferred pathway of chick embryo epiblast cells in vitro. Dev. Biol. 173, 279-291. George-Weinstein, M., Gerhart, J., Mattiacci-Paessler, M., Simak, E., Blitz, J., Reed, R., and Knudsen, K. (1998). The role of stably committed and uncommitted cells in establishing tissues of the somite. Ann. N.Y. Acad. Sci. 842, 16-27. Gitehnan, I. (1997). Twist protein in mouse embryogenesis. Dev. Biol. 189,205-214. Glinka, A., Wu, W., Onichtchouk, D., Blumenstock, C., and Niehrs, C. (1997). Head induction by simultaneous repression of Bmp and Wnt signalling in Xenopus. Nature 389,517-519. Goldstein, R. S., and Kalcheim, C. (1992). Determination of epithelial half-somites in skeletal morphogenesis. Development 116, 441-445.

55

SOMiTOG$NESIS

Goodrich, L. V., Johnson, R. L., Milenkovic, L., McMahon, J. A., and Scott, M. I?. (1996). Conservation of the hedgehog/patched signaling pathway from flies to mice: Induction of a mouse patched gene by Hedgehog. Genes Dev. l&301-312. Gossler, A., andHrabe de Angelis, M. (1998). Somitogenesis. Curr. Top. Dev. Viol. 38,22X%7. Goulding, M., Lumsden, A., and Paquette, A. J. (1994). Regulation of Pax-3 expression in the dermomyotome and its role in muscle development. DcveEopm&zl l%J, 957-971. Greene, N. D., Gerrelli, D., Van Straaten, H. W., and Copp, A. .I. (1998). Abnormalities of floor plate, notochord and somite differentiation in the loop-tail (Lp) mouse: A model of severe neural tube defects. Me&. Dev. 73,59-72. Grobstein, C., and Holtzer, H. (1955). In vitro studies of cartilage induction in mouse somite mesoderm. J. Exp. Zool. 128, 333-351. Gumbiner, B. M. (1998). Propagation and localization of Wnt signaling. Curr. @in. Genet. Dev.

8,430-435.

Gurdon, J. B. (1988). A community effect in animal development. Nature 336, 772-774. Hahn, H., Christiansen, J., Wicking, C., Zaphiropoulos, P. G., Chidambaram, A., Cerrard, B., Vorechovsky, I., Bale, A. E., Toftgard, R., Dean, M., and Wainwright, B. (1996). A mammalian patched homolog is expressed in target tissues of sonic hedgehog and maps to a region associated with developmental abnormalities. J. Biol. Chem. 271,12125-12128. Hall, B. K. (1977). Chondrogenesis of the somitic mesoderm. Adv. Anat. Embryol. Cell Biol. 53, 3-41. Hamburger, V. (1992). The stage series of the chick embryo [comment]. Da, Dyes.I%, 273-275. Hatada, Y., and Stern, C. D. (1994). A fate map of the epiblast of the early chick embryo. Development

X20,2879-2889.

Hatta, K., and Takeichi, M. (1986). Expression of N-cadherin adhesion molecules associated with early morphogenetic events in chick development. Nature 320,441~449. Hayashi, II., Mochii, M., Kodama, R., Hamada, Y., Miiuno, N., Eguchi, G., and Tachi, 63. (1996). Isolation of a novel chick homolog of Serrate and its coexpression with C-Notch& 1 in chick development. Znt. .I Dev. Biol. a@, 1089-1096. Hebrok, M., Wertz, K., and Fuchtbauer, E. M. (1994). M-twist is an inhibitor of muscle differentiation. Dev. Biol. 165, 537-544. Henrique, D., Adam, .I., Myat, A., Chitnis, A., Lewis, J., and Ish-Horowicz, D. (1995). Expression of a Delta homologue in prospective neurons in the chick. Nature 375,787-790. Hirano, S,, Hirako, R., Kajita, N., and No&a, M. (1995). Morphological analysis of the role of the neural tube and notochord in the development of somites. Anat. EMwyoZ. (BerL)

192,445-457. Hirsinger, E., Duprez, D., Jouve, C., Malapert, P., Cooke, J., and Pourquie, 0. (19Q7). Noggin acts downstream of Wnt and Sonic Hedgehog to antagonize BMP4 in avian somite patterning. Development X&&4605-4614. Ho, R. K., Ball, E. E., and Goodman, C. S. (1983). Muscle pioneers: large mesodermal cells that erect a scaffold for daveloping muscles and motoneurones in grasshopper embryos. Nature 301, 66-69. Hoang, B. H., Thomas, J. T., Abdul-Karim, F. W., Correia, K. M., Conlon, R. A., Luyten, F. P., and Ballock, R. T. (1998). Expression pattern of two Frizzled-related genes, l?r&l and Sfrp-1, during mouse embryogenesis suggests a role for modulating action of Wnt family members. Dev. Dyn. 212, 364-372. Hogan, B. L. (1996). Bone morphogenetic proteins: multifunctional regulators of vertebrate development. Genes Dev. lO, l%O-1594. Hollyday, M., McMahon, J. A., and McMahon, A. P. (1995). Wnt expression patterns in chick embryo nervous system. Me&. Dev. 52,9-25. Holt, C. E., Lemaire, P., and Gurdon, J. B. (1994). Cadherin-mediated cell interactions are necessary for the activation of MyoD in Xenopus mesoderm. Proc. Natl. Acad. Sci, U.S.A. 91,10844-10848.

56

HIRSING~R

ET AL.

Holtzer, H., and Detwiler, S. R. (1953). An experimental analysis of the development of the ’ spinal column. III. Induction of skeletogenous cells. J. Exp. 2001. 123, 335-369. Horstadius, S. (1944). Ueber die folgen von chordaexstirpation an spaeten gastrulae turd neurulae von Amblystoma punctatum. Acta 2002. 25, 75-87. Hrabe de Angelis, M., McIntyre, J., and Gossler, A. (1997). Maintenance of somite borders in mice requires the Delta homologue DIIl. Nature 386, 717-721. Hsu, D. R., Economides, A. N., Wang, X., Eimon, P. M., and Harland, R. M. (1998). The Xenopus dorsalizing factor Gremlin identifies a novel family of secreted proteins that antagonize BMP activities. Mol. Cell 1, 673-683. Huang, R., Zhi, Q., Neubtiser, A., Mtiller, T. S., Brand-Saberi, B., Christ, B., and Wilting, J. (1996). Function of somite and somitocoele cells in the formation of the vertebral motion segment in avian embryos. Actu Anut. (Bawl) 155,231-241. Huang, R., Zhi, Q., Ordahl, C. P., and Christ, B. (1997). The fate of the first avian somite. Anat.

Embryol.

(Berl.)

195, 435-449.

Huang, R., Zhi, Q., Wilting, J., and Christ, B. (1994). The fate of somitocoele cells in avian embryos. Anat. Embryol. (BerZ.) 190,243-250. Hui, C. C., Slusarski, D., Platt, K. A., Holmgren, R., and Joyner, A. L. (1994). Expression of three mouse homologs of the Drosophila segment polarity gene cubitus interruptus, Gli, Gli-2, and Gli-3, in ectoderm- and mesoderm-derived tissues suggests multiple roles during postimplantation development. Dev. Biol. 163402-413. Ikeya, M., and Takada, S. (1998). Wnt signaling from the dorsal neural tube is required for the formation of the medial dermomyotome. Development 125, 4969-4976. Jacobson, A. G. (1988). Somitomeres: Mesodermal segments of vertebrate embryos. Development

104 Suppl,209-220.

Jarriault, S., Brou, C., Logeat, F., Schroeter, E. H., Kopan, R., and Israel, A. (1995). Signalling downstream of activated mammalian Notch. Nature 377, 35.5-358. Jarriault, S., Le Bail, O., Hirsinger, E., Pourquie, O., Logeat, F., Strong, C. F., Brou, C., Seidah, N. G., and Israel, A. (1998). Delta-l activation of notch-l signaling results in HES1 transactivation. Mol. Cell Biol. 18, 7423-7431. Jellies, J. (1990). Muscle assembly in simple systems. Trends Neurosci. 13, 126-131. Jen, W. C., Wettstein, D., Chitnis, D., and Kintner, C. (1997). The Notch ligand, X-Delta2, mediates segmentation of the paraxial mesoderm in Xenopus embryos. Dev. Suppl. 124 1169-1178. Jen, Y., Weintraub, H., and Benezra, R. (1992). Overexpression of Id protein inhibits the muscle differentiation program: In vivo association of Id with E2A proteins. Genes Dev. 6,1466-1479.

Jensen, A. M., and Wallace, V. A. (1997). Expression of Sonic hedgehog and its putative role as a precursor cell mitogen in the developing mouse retina. Development W&363-371. Johnson, R. L., Laufer, E., Riddle, R. D., and Tabin, C. (1994). Ectopic expression of Sonic hedgehog alters dorsal-ventral patterning of somites. Cell 79, 1165-1173. Johnston, S. H., Rauskolb, C., Wilson, R., Prabhakaran, B., Irvine, K. D., and Vogt, T. F. (1997). A family of mammalian Fringe genes implicated in boundary determination and the Notch pathway. Development 124, 2245-2254. Jones, C. M., and Smith, J. C. (1998). Establishment of a BMP-4 morphogen gradient by long-range inhibition. Dev. Biol. 194, 12-17. Jones, C. M., Armes, N., and Smith, J. C. (1996). Signalling by TGF-beta family members: Short-range effects of Xnr-2 and BMP-4 contrast with the long-range effects of activin. Curr. Biol. 6, 1468-1475. Kablar, B., Krastel, K., Ying, C., Asakura, A., Tapscott, S. J., and Rudnicki, M. A. (1997). MyoD and Myf-5 differentially regulate the development of limb versus trunk skeletal muscle. Development 124, 4729-4738.

57

lSQMITQGE$lESIS

Kaehn, K., Jacob, II. J., Christ, B., Hinrichsen, K., and Poelmann, R. E. (1988). The onset of myotome formation in the chick, Anat. Embryol. (Bk%.) 177,191-201. Kahane, N., Cinnamon, Y., and Kalcheim, C. (1998a). The cellular mechanism by which the dermomyotome contributes to the second wave of myotome development. Development

l25,4259-4271. Kahane, N., Cinnamon, Y ., and Kalcheim, C. (1998b). The origin and fate of pioneer myotomal cells in the avian embryo. Me&. Dev. 74,59-73. Kate, N., and Aoyama, H. (1998). Dermomyotomal origin of the ribs as revealed by extirpation and transplantation experiments in chick and quail embryos. Development X%,3437-3443. Kate, H., Taniguchi, Y., Kurooka, H., Minoguchi, S., Sakai, T., Nomura-Okazaki, S., Tamura, K., and Honjo, T. (1997). Involvement of RBP-J in biological functions of mouse Notch1 and its derivatives. Development lZ& 4133-4141. Kenny-Mobbs, T., and Thorogood, P. (1987). Autonomy of differentiation in avian branchial somites and the influence of adjacent tissues. Development 100,449-462. Keynes, R. J., and Stern, C. D. (1988). Mechanisms of vertebrate segmentation. Development

103,413-429. Kieny, M. (1960). Role inducteur du mesoderme dam la differentiation precoce du bourgeon de membre chez l’embryon de Poulet, J. Embryol. Exp. Morphol. 8,457-467. Kimble, J., and Simpson, P. (1997). The LIN-1ZNotch signaling pathway and its regulation. Ann.

Rev.

Cell Dev.

Biol.

l3, 333-361.

Kimura, Y., Matsunami, H., Inoue, T., Shimamura, K., Uchida, N., Ueno, T., Miyazaki, T., and Takeichi, M. (1995). Cadherin-11 expressed in association with mesenchymal morphogenesis in the head, somite, and limb bud of early mouse embryos. Dev. Biol. 169,347-358. Kopan, R., Nye, J. S., and Weintraub, H. (1994). The intracellular domain of mouse Notch: A constitutively activated repressor of myogenesis directed at the basic helix-loop-helix region of MyoD. Development lZO,2385-2396. Kos, L., Chiang, C., and Mahon, K. A. (1998). Mediolateral patterning of somites: Multiple axial signals, including Sonic hedgehog, regulate Nkx-3.1 expression. Mech. Dev. 70,25-34. Koseki, H., Wallin, J., Wilting, J., Mizutani, Y., K&pert, A., Ebensperger, C., Hermann. B. G., Christ, B., and Balling, R. (1993). A role for Pax-l as a mediator of notochordal signals during the dorsoventral specification of vertebrae. Development W&649-660. Krauss, S., Concordet, J. P., and Ingham, P. W. (1993). A functionally conserved homolog of the Drosophila segment polarity gene hh is expressed in tissues with polarizing activity in zebra&h embryos. Cell 75,1431-1444. Krenn, V., Gorka, P., Wachtler, F., Christ, B., and Jacob, H. J. (1988). On the origin of cells determined to form skeletal muscle in avian embryos. Anat. Embryol. (Bed) 179, 49-54. Krull, C. E., Lansford, R., Gale, N. W., Collazo, A., Marcelle, C., Yancopoulos, G. D., Fraser, S. E., and Bronner-Fraser, M. (1997). Interactions of Eph-related receptors and ligands confer rostrocaudal pattern to trunk neural crest migration. Curr. Bid. 7, 571-580. Kuratani, S., Martin, J. F., Wawersik, S., Lilly, B., Eichele, G., and Olson, E. N. (1994). The expression pattern of the chick homeobox gene gMHox suggests a role in patterning of the limbs and face and in compartmentalization of somites., Dev. BioE. 161, 357-369. Kusumi, K., Sun, E. S., Kerrebrock, A. W., Bronson, R. T., Chi, D. C., Bulotsky, M. S., Spencer, J. B., Birren, B. W., Frankel, W. N., and Lander, E. S. (1998). The mouse pudgy mutation disrupts Delta homologoue D113 and initiation of early somite boundaries Nut. Genet.

19,274-278.

Ladher, R., Mohun, T. J., Smith, J. C., and Snape, A. M. (1996). Xom: A Xenopus homeobox gene that mediates the early effects of BMP-4. Development K&Z, 2385-2394. Landolt, R. M., Vaughan, L., Winter-halter, K. H., and Zimmermaun, D. R. (1995). Versican is selectively expressed in embryonic tissues that act as barriers to neural crest cell migration and axon outgrowth. Development X&2303-2312.

58

HIRSINGCP

ET

4L. 1

Langman, J., and Nelson, G. R. (1968). A radioautographic study of the development of the somite in the chick embryo. J. Embryol. Exp. Morphol. 19, 217-226. Lash, J., Holtzer, S., and Holtzer, H. (1957). An experimental analysis of the development of the spinal column. VI. Aspects of cartilage induction. Exp. Cell Rex 13, 292-303. Lash, J, W., and Yamada, K. (1986). The adhesion signal of fibronectin: A possible trigger mechanism for compaction during somitogenesis. In “Somites in Developing Embryos” (R. Bellairs, D. Ede, J. Lash, Eds.), pp. 201-208. Plenum Press, New York and London. Lawrence, P. A. (1992). “The Making of a Fly.” Blackwell Science, Oxford. Lecuit, T., Brook, W. J,, Ng, M., Calleja, M., Sun, H., and Cohen, S. M. (1996). Two distinct mechanisms for long-range patterning by Decapentaplegic in the Drosophila wing. Nature 381,387-393.

Lee, J. J., Ekker, S. C., von Kessler, D. P., Porter, J. A., Sun, B. I., and Beachy, P. A. (1994). Autoproteolysis in hedgehog protein biogenesis [see comments]. Science 266, 1528-1537. Leyns, L., Bouwmeester, T., Kim, S. H., Piccolo, S., and De Robertis, E. M. (1997). Frzb1 is a secreted antagonist of Wnt signaling expressed in the Spemann organizer. CeZl S&747-756.

Li, L., Cserjesi, P., and Olson, E. N. (1995). Dermo-1: a novel twist-related bHLH protein expressed in the developing dermis. Dev. Biol. 172, 280-292. Liem, K. F. J., Tremml, G., Roelink, H., and Jessell, T. M. (1995). Dorsal differentiation of neural plate cells induced by BMP-mediated signals from epidermal ectoderm. Cell 82,969-979.

Lindsell, C. E., Boulter, J., diSibio, G., Gossler, A., and Weinmaster, G. (1996). Expression patterns of Jagged, Deltal, Notchl, Notch2, and Notch3 genes identify ligand-receptor pairs that may function in neural development. Mol. Cell Neurosci. 8, 14-27. Lindsell, C. E., Shawber, C. J., Boulter, J., and Weinmaster, G. (1995). Jagged: A mammalian ligand that activates Notchl. Cell 80, 909-917. Luo, B., Aster, J. C., Hasserjian, R. P., Kuo, F., and Sklar, J. (1997). Isolation and functional analysis of a cDNA for human Jagged2, a gene encoding a ligand for the Notch1 receptor. Mol. CellBiol. 17, 6057-6067. Marcelle, C., Stark, M. R., and Bronner-Fraser, M. (1997). Coordinate actions of BMPs, Writs, Shh and Noggin mediate patterning of the dorsal somite. Development 124, 3955-3963. Maroto, M., Reshef, R., Mtinsterberg, A. E., Koester, S., Goulding, M., and Lassar, A. B. (1997). Ectopic Pax-3 activates MyoD and Myf-5 expression in embryonic mesoderm and neural tissue. Cell 89, 139-148. Marti, E., Bumcrot, D. A., Takada, R., and McMahon, A. P. (1995a). Requirement of 19K form of Sonic hedgehog for induction of distinct ventral cell types in CNS explants [see comments]. Nature 375, 322-325. Marti, E., Takada, R., Bumcrot, D. A., Sasaki, H., and McMahon, A. P. (1995b). Distribution of Sonic hedgehog peptides in the developing chick and mouse embryo. Development X21,2537-2547.

McGrew, M. J., Dale, J. K., Fraboulet, S., and Pourquie, 0. (1998). The Lunatic Fringe gene is a target of the molecular clock linked to segmentation iu avian embryos. Curr. Biol. 8,979-982. McMahon, A. P., and Bradley, A. (1990). The Wnt-1 (int-1) proto-oncogene is required for development of a large region of the mouse brain. Cell 62,1073-1085. McMahon, J. A., Takada, S., Zimmerman, L. B., Fan, C. M., Harland, R. M., and McMahon, A. P. (1998). Noggin-mediated antagonism of BMP signaling is required for growth and patterning of the neural tube and somite. Genes Dev. Z&1438-1452. Meier, S. (1984). Somite formation and its relationship to metameric patterning of the mesoderm. Cell Differ. 14,235-243. Meinhardt, H. (1986). Models of segmentation. “Somites in Developing Embryos” (R. Bellairs, D. Ede, J. Lash, Eds.), pp. 179-191. Plenum Press, New York and London.

59

SOMITDGENESIS

Melnikova, I. N., and Christy, B. A. (1996). Muscle cell differentiation is inhibited by the helix-loop-helix protein Id3. CeZZGrowth. Differ. 7,1067-1079. Mennerich, D., Schafer, K., and Braun, T. (1998). Pax-3 is necessary but not sufficient for lbxl expression in myogenic precursor cells of the limb. n/lech. Dev. 73,147-158. Mestres, P., and Hinrichsen, K. (1976). Zur histogenese der somiten beim Htihnchen. .I. Embryol.

Exp.

Morph.

36, 669-693.

Mitsiadis, T. A., Hirsinger, E., Lendahl, U., and Goridis, C. (1998). Delta-notch signaling in odontogenesis: Correlation with cytodiiferentiation and evidence for feedback regulation [In Process Citation]. Dev. Biol. 204,420-431. MO, R., Freer, A. M., Zinyk, D. L., Crackower, M. A., Michaud, J., Heng, H. H., Chik, K. W., Shi, X. M., Tsui, L. C., Cheng, S. H., Joyner, A. L., and Hui, C. (1997). Specific and redundant functions of Gli2 and GIi3 zinc finger genes in skeletal patterning and development. Development l2q 113-123. Monsoro-Burq, A. H., Bontoux, M., Teillet, M. A., and Le Douarin, N. M. (1994). Heterogeneity in the development of the vertebra. Proc. Natl. Acad. Sci. U.S.A. 9,10435-10439. Motoyama, J., Takabatake, T., Takeshima, K., and Hui, C. (1998). Ptch2, a second mouse Patched gene is co-expressed with Sonic hedgehog [letter]. Nat. Genet. 18, 104-106. Muller, M., v Weizsacker, E., and Campos-Ortega, .I. A. (1996). Expression domains of a zebra&h hamologue of the Drosophila pair-rule gene hairy correspond to primordia of alternating somites. Development K&2,2071-2078. Mtiller, T. S., Ebensperger, C., Neubtiser, A., Koseki, H., Balling, R., Christ, B., and Wilting, J. (1996). Expression of avian Pax1 and Pax9 is intrinsically regulated in the pharyngeal endoderm, but depends on environmental influences in the pat-axial mesoderm. Dev. Biol. 178,403-417. Miinsterberg, A. E., and Lassar, A. B. (1995). Combinatorial signals from the neural tube, floor plate and notochord induce myogenic bHLH gene expression in the somite. Development

E&1,651-660.

Mtinsterberg, A. E., Kitajewski, J., Bumcrot, D. A., McMahon, A. P., and Lassar, A. B. (1995). Combinatorial signaling by Sonic hedgehog and Wnt family members induces myogenic bHLH gene expression in the somite. Genes Dev. 9,2911-2922. Myat, A,, Henrique, D., Ish-Horowitz, D., and Lewis, J. (1996). A chick homologue of Serrate and its relationship with Notch and Delta homologues during central nemogenesis. Dev. Biol.

174,233-247.

Nellen, D., Burke, R., Struhl, G., and Basler, K. (1996). Direct and long-range action of a DPP morphogen gradient. Cell 85,357-368. Newgreen, D. (1984). Spreading of explants of embryonic chick mesenchymes and epithelia on fibronectin and laminin. Cell, Tissue Res. 236, 26.5-277. Nicolas, J. F., and Bonnerot, C. (1988). Recombinant retrovirus, cell lineage and gene expression in the mouse embryo. Prog. Clin. Biol. Res. 284, 125-145. Nicolas, J. F., Mathis, L., Bonnerot, C., and Saurin, W. (1996). Evidence in the mouse for self-renewing stem cells in the formation of a segmented longitudinal structure, the myotome. Development

a&2933-2946.

Nicolet, G. (1971). Avian gastrulation. Adv. Morphog. 9, 23-262. Nittenberg, R., Patel, K., Joshi, Y., Krumlauf, R., Wilkinson, D. G., Brickell, P. M., TicMe, C., and Clarke, J. D. (1997). Cell movements, neuronal organisation and gene expression in hindbrains lacking morphological boundaries. Development X24,2297-2306. Nye, J. S., Kopaq, R., and Axel, R. (1994). An activated Notch suppresses neurogenesis and myogenesis but not gliogenesis in mammalian cells. Development X20,2421-2430. Oka, C., Nakano, ‘I., Wakeham, A., de la Pompa, J. L., Mori, C., Sakai, T., Okazaki, S., Kawaichi, M., Shiota, K., Mak, T. W., and Honjo, T. (1995). Disruption of the mouse RBPJ kappa gene results in early embryonic death. Development E&3291-3301.

60

HIRSINGERET AL,

Ooi, V. E., Sanders, E. J., and Bellairs, R. (1986). The contribution of the primitive streak to the somites in the avian embryo. J. EmbryoI. Exp. Morphol. 92,193-206. Ordahl, C. P. (1993). Myogenic lineages within the developing somite. “Molecular Basis of Morphogenesis” (M. Bernfield, Ed.), pp. 165-176. John Wiley and Sons, New York. Ordahl, C. P., and Le Douarin, N. M. (1992). Two myogenic lineages within the developing somite. Development 114339-353. Ordahl, C. P., and Williams, B. A. (1998). Knowing chops from chuck: roasting myoD redundancy. Bioessays 20,357-362. Ostrovsky, D., Sanger, J. W., and Lash, J. W. (1988). Somitogenesis in the mouse embryo. Cell Differ. 23, 17-25. Packard, D. S. J., and Jacobson, A. (1976). The influence of axial structures on chick somite formation. Dev. Biol. 53, 36-48. Packard, D. S. J., Zheng, R. Z., and Turner, D. C. (1993). Somite pattern regulation in the avian segmental plate mesodenn. Development 117,779-791. Palmeirim, I., Dubrulle, J., Henrique, D., Ish-Horowitz, D., and PourquiC, 0. (1998). Uncoupling segmentation and somitogenesis in the chick presomitic mesoderm. Dev. Genet. 23, 77-85. Palmeirim, I., Henrique, D., Ish-Horowitz, D., and Pourquie, 0. (1997). Avian hairy gene expression identifies a molecular clock linked to vertebrate segmentation and somitogenesiss. Cell 91, 639-648. Panin, V. M., Papayannopoulos, V., Wilson, R., and Irvine, K. D. (1997). Fringe modulates Notch-ligand interactions. Nature 387, 908-912. Parr, B. A., Shea, M. J., Vassileva, G., and McMahon, A. P. (1993). Mouse Wnt genes exhibit discrete domains of expression in the early embryonic CNS and limb buds. Development 119,%7-261. Pasteels, J. (1937). Etudes sur la gastrulation des Vertebres meroblastiques. III. Oiseaux. IV. Conclusions generales. Arch. Biol. 48,381-488. Patel, K., Connolly, D. J., Amthor, H., Nose, K., and Cooke, J. (1996). Cloning and early dorsal axial expression of Flik, a chick follistatin-related gene: evidence for involvement in dorsalizationmeural induction. Dev. Biol. 178, 327-342. Piccolo, S., Sasai, Y., Lu, B., and De Robertis, E. M. (1996). Dorsoventral patterning in Xenopus: Inhibition of ventral signals by direct binding of chordin to BMP-4. Cell 86, 589-598. Placzek, M. (1995). The role of the notochord and floor plate in inductive interactions. Curr. Opin. Genet. Dev. 5, 499-506. Platt, K. A., Michaud, J., and Joyner, A. L. (1997). Expression of the mouse Gli and Ptc genes is adjacent to embryonic sources of hedgehog signals suggesting a conservation of pathways between flies and mice. Mech. Dev. 62, 121-135. Pourquie, O., Coltey, M., BrCant, C., and Le Douarin, N. M. (1995). Control of somite patterning by signals from the lateral plate. Proc. Natl. Acad. Sci. U.S.A. 92, 3219-3223. Pourquie, O., Coltey, M., Teillet, M. A., Ordahl, C., and Le Douarin, N. M. (1993). Control of dorsoventral patterning of somitic derivatives by notochord and floor plate. Proc. Natl. Acad. Sci. U.S.A. 90, 5242-5246. Pourquit, O., Fan, C. M., Coltey, M., Hirsinger, E., Watanabe, Y., Breant, C., Francis-West, P., Brickell, P., Tessier-Lavigne, M., and Le Douarin, N. M. (1996). Lateral and axial signals involved in avian somite patterning: A role for BMP4. Cell 84, 461-471. Pownall, M. E., and Emerson, C. P. J. (1992). Sequential activation of three myogenic regulatory genes during somite morphogenesis in quail embryos. Dev. Biol. 151,67-79. Pownall, M. E., Strunk, K. E., and Emerson, C. P. J. (1996). Notochord signals control the transcriptional cascade of myogenic bHLH genes in somites of quail embryos. Development 122,1475-1488.

61

SOMI~OGENESIS

Psychoyos, D., and Stem, C. D. (1996a). Fates and migratory routes of primitive streak cells in the chick embryo. Development l22,1523-1534. Psychoyos, D., and Stern, C. D. (1996b). Restoration of the organizer after radical ablation of Hensen’s node and the anterior primitive streak in the chick embryo. Development X22,3263-3273.

Radice, G. L., Rayburn, H., Matsunami, H., Knudsen, K. A., Takeichi, M., and Hynes, R. 0. (1997). Development defects in mouse embryos lacking N-cadherin. Dev. Biol. l&64-78. Ranscht, B., and Bronner-Fraser, M. (1991). T-cadherin expression alternates with migrating neural crest cells in the trunk of the avian embryo. Development HI, U-22. Resume, A. G., Conlon, R. A., Zirngibl, R., Yamaguchi, T. P., and Rossant, J. (1992). Expression analysis of a Notch homologue in the mouse embryo. Dev. Biol. X54,377-387. Remak, R. (1850). “Untersuchungen tiber die entwicklung der Wirbeltiere.” Reimer, Berlin. Reshef, R., Maroto, M., and Lassar, A. B. (1998). Regulation of dorsal somitic cell fates: BMPs and Noggin control the timing and pattern of myogenic regulator expression Genes Dev.

C&290-303.

Richardson, M. K., Allen, S. P., Wright, G. M., Raynaud, A., and Hanken, J. (1998). Somite number and vertebrate evolution. Development 125, 151-160. Riddle, R. D., Johnson, R. L., Laufer, E., and Tabin, C. (1993). Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 751401-1416. Riechmann, V., van Cruchten, I., and Sablitzky, F. (1994). The expression pattern of Id4, a novel dominant negative helix-loop-helix protein, is distinct from Idl, Id2 and Id3. Nucleic Acids

Res. 22,149-155.

Ring, C., Hassell, J., and Halfter, W. (1996). Expression pattern of collagen IX and potential role in the segmentation of the peripheral nervous system. Dev. Biol. 180,41-53. Roelink, H., Porter, J. A., Chiang, C., Tanabe, Y., Chang, D. T., Beachy, P. A., and Jessell, T. M. (1995). Floor plate and motor neuron induction by different concentrations of the amino-terminal cleavage product of sonic hedgehog autoproteolysis. Cell 81, 445-4.55. Rohan de Silva, H. A., and Patel, A. J. (1997). Presenilins and early-onset familial Alzheimer’s disease. Neuroreport 8, i-xii. Rong, P. M., Teillet, M. A., Ziller, C., and Le Douarin, N. M. (1992). The neural tube/ notochord complex is necessary for vertebral but not limb and body wall striated muscle differentiation. Development 115, 651-612. Rutz, R., Haney, C., and Hauschka, S. (1982). Spatial analysis of limb bud myogenesis: A proximodistal gradient of muscle colony-forming cells in chick embryos leg buds. Dev. Biol. 90,399-411. Saga, Y., Hata, N., Kobayashi, S., Magnuson, T., Seldin, M. F., and Taketo, M. M. (1996). MesPl: A novel basic helix-loop-helix protein expressed in the nascent mesodermal cells during mouse gastrulation. Development 122, 2769-2778. Saga, Y., Hata, N., Koseki, H., and Taketo, M. M. (1997). Mesp2: A novel mouse gene expressed in the presegmented mesoderm and essential for segmentation initiation. Genes Dev. l&1827-1839. Sandor, S., and Fazakas-Todea, I. (1980). Researches on the formation of axial organs in the chick embryo. X. Further investigations on the role of ecto- and endoderm in somitogenesis. Morphol.

Embryol.

(Bucw.)

26,29-32.

Sasai, Y., Lu, B., Steinbeisser, H., Geissert, D., Gont, L. K., and De Robertis, E. M. (1994). Xenopus chordin: A novel dorsalizing factor activated by organizer-specific homeobox genes. Cell 79,779-790. Schoenwolf, G. C., Garcia-Martinez, V., and Dias, M. S. (1992a). Mesoderm movement and fate during avian gastrulation and neurulation. Dev. Dyn. 193, 235-248. Schoenwolf, G. C., Garcia-Martinez, V., and Dias, M. S. (1992b). Mesoderm movement and fate during avian gastrulation and neurnlation. Dev. Dyn. 193,235-248.

62

HIRSINGER

ET AL.

Selleck, M. A., and Stern, C. D. (1991). Fate mapping and cell lineage analysis of Hensen’s node in the chick embryo. Development ll2, 615-626. Selleck, M. A., and Stern, C. D. (1992). Commitment of mesoderm cells in Hensen’s node of the chick embryo to notocord and somite. Development 114,403-415. Shawber, C., Nofziger, D., Hsieh, J. J., Lindsell, C., Bogler, O., Hayward, D., and Weinmaster, G. (1996). Notch signaling inhibits muscle cell differentiation through a CBFl-independent pathway. Development I22,3165-3773. Shawlot, W., Deng, J. M., and Behringer, R. R. (1998). Expression of the mouse cerberusrelated gene, Cerrl, suggests a role in anterior neural induction and somitogenesis. Proc. Natl. Acad. Sci. U.S.A. 9.5, 6198-6203. Smith, W. C., and Harland, R. M. (1992). Expression cloning of Noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70, 829-840. Smolich, B. D., McMahon, J. A., McMahon, A. P., and Papkoff, J. (1993). Wnt family proteins are secreted and associated with the cell surface. Mol. Biol. Cell 4, 1267-1275. Sosic, D., Brand-Saber-i, B., Schmidt, C., Christ, B., and Olson, E. N. (1997). Regulation of paraxis expression and somite formation by ectoderm- and neural tube-derived signals, Dev. Biol. 185,229-243. Sparrow, D. B., Jen, W. C., Kotecha, S., Towers, N., Kintner, C., and Mohun, T. J. (1998). Thylacine 1 is expressed segmentally within the paraxial mesoderm of the Xenopus embryo and interacts with the Notch pathway. Development 12.5,2041-2051. Spence, M. S., Yip, J., and Erickson, C. A. (1996). The dorsal neural tube organizes the dermamyotome and induces axial myocytes in the avian embryo. Development l22,231-241. Spicer, D. B., Rhee, J., Cheung, W. L., and Lassar, A. B. (1996). Inhibition of myogenic bHLH and MEF2 transcription factors by the bHLH protein Twist [see comments]. Science 272,1476-1480.

SpSrle, R., Gunther, T., Struwe, M., and Schughart, K. (1996). Severe defects in the formation of epaxial musculature in open brain (opb) mutant mouse embryos. Development X22,79-86. Spratt, N. T. (1955). Analysis of the organizer center in the early chick embryo. I. Localisation of prospective notochord and somite cells. J. Exp. Zool. 128, 121-162. Stern, C. D. (1992). Mesoderm induction and development of the embryonic axis in amniotes. Trends

A Genet.

8, 153-163.

Stern, H. M., and Hauschka, S. D. (1995). Neural tube and notochord promote in vitro myogenesis in single somite explants. Dev. Biol. 167, 87-103. Stern, C. D., and Keynes, R. J. (1987). Interactions between somite cells: The formation and maintenance of segment boundaries in the chick embryo. Development 99,261-272. Stern, H. M., Brown, A. M., and Hauschka, S. D. (1995). Myogenesis in paraxial mesoderm: Preferential induction by dorsal neural tube and by cells expressing Wnt-1. Development l21,3675-3686. Stern, C. D., Fraser, S. E., Keynes, R. J., and Primmett, D. R. (1988). A cell lineage analysis of segmentation in the chick embryo. Development 104,231-244. Stoetzel, C., Weber, B., Bourgeois, P., Bolcato-Bellemin, A. L., and Perrin-Schmitt, F. (1995). Dorso-ventral and rostro-caudal sequential expression of M-twist in the postimplantation murine embryo. Me&. Dev. 51,251-263. Strudel, G. (1955). L’action morphogene du tube nerveux et de la corde sur la differentiation des Verttbres et des muscles vertebraux chez l’embryon de Poulet. Arch. Anat. Microsc. Morph.

Exp.

44(3),

209-235.

Sun, X. H., Copeland, N. G., Jenkins, N. A., and Baltimore, D. (1991). Id proteins Id1 and Id2 selectively inhibit DNA binding by one class of helix-loop-helix proteins. Mol. Cell BioL l&5603-5611. Swiatek, P. J., Lindsell, C. E., de1 Amo, F. F., Weinmaster, G., and Gridley, T. (1994). Notch1 is essential for postimplantation development in mice. Genes Dev. 8, 707-719.

SOMIT&XENESIS

63

Tajbakhsh, S., and Sporle, R. (1998). Somite development: constructing the vertebrate body. Cell 92, 9-16.

Tajbakhsh, S., Borello, U., Vivarelli, E., Kelly, R., Papkoff, J., Duprez, D., Buckingham, M., and Cossu, G. (1998). Differential activation of Myf5 and MyoD by different Wnts in explants of mouse paraxial mesoderm and the later activation of myogenesis in the absence of Myf5.

Development

l25,4155-4162.

Tajbakhsh, S., Rocancourt, D., Cossu, G., and Buckingham, M. (1997). Redefining the genetic hierarchies controlling skeletal myogenesis: Pax- 3 and Myf-5 act upstream of MyoD. Cell S9,127-138. Takada, S., Stark, K. L., Shea, M. J., Vassileva, G., McMahon, J. A., and McMahon, A. P. (1994). Wnt-3a regulates somite and tailbud formation in the mouse embryo. Gerzes Dev. 8,174-189. Takahashi, Y., Monsoro-Burq, A. H., Bontoux, M., and Le Douarin, N. M. (1992). A role for Quox-8 in the establishment of the dorsoventral pattern during vertebrate development. Proc.

Natl.

Acad.

Sci. U.S.A.

89,10237-10241.

Tam, P. P. (1988). The allocation of cells in the presomitic mesoderm during somite segmentation in the1mouse embryo. Development 103, 379-390. Tam, P. P., and Beddington, R. S. (1987). The formation of mesodermal tissues in the mouse embryo during gastrulation and early organogenesis. Development 99,109-126. Tam, P. P., and Beddington, R. S. (1992). Establishment and organization of germ layers in the gastrulating mouse embryo. Ciba Found. Symp. l&5,27-41. Tam, P. P. L., and Beddington, R. S. P. (1986). The metameric organization of the presomitic mesoderm and somite specification in the mouse embryo. “Somites in Developing Embryos” (R. Bellairs, D. Ede, J. Lash, eds.), pp. 17-36. Plenum Press, New York and London. Tanabe, Y., and Jessell, T. M. (1996). Diversity and pattern in the developing spinal cord lpubiished erratum appears in Science 1997 Apr 4;276(5309):21]. Science 274, 1115-1123. Teillet, M. A., and Le Douarin, N. M. (1983). Consequences of neural tube and notochord excision on the development of the peripheral nervous system in the chick embryo. Dev. Biol. 98,192-211. Teillet, M., Watanabe, Y., Jeffs, P., Duprez, D., Lapointe, F., and Le Douarin, N. M. (1998). Sonic hedgehog is required for survival of both myogenic and chondrogenic somitic lineages. Development 125,2019-2030. Tonegawa, A., and Takahashi, Y. (1998). Somitogenesis controlled by Noggin. Dev. Biol. 202,172-182. Tonegawa, A., Funayama, N., Ueno, N., and Takahashi, Y. (1997). Mesodermal subdivision along the mediolateral axis in chicken controlled by different concentrations of BMP-4. Development

x24,1975-1984.

Van Straaten, H. W., and Hekking, J. W. (1991). Development of floor plate, neurons and axonal outgrowth pattern in the early spinal cord of the notochord-deficient chick embryo. Anat. Embxyol. (Berl.) 184, 55-63. Veini, M., and Bellairs, R. (1991). Early mesoderm differentiation in the chick embryo. Amt. Embryol. (BerZ.) 183, 143-149. Verbout, A. J. (1985). The development of the vertebral column. Adv. Anat. Embryol. Cell Biol.

90, 1-122.

Vivarelli, E., and Cossu, G. (1986). Neural control of early myogenic differentiation in cultures of mouse somites. Dev. Biol. 117,319-325. Von Ebner, V. (1888). Unvirbel und Neugliederung der WirbelSaule. Sitzungsbev. Akad. Wiss.

Wien

IIu97,

194-206.

von Kirschhofer, K., Grim, M., Christ, B., and Wachtler, F. (1994). Emergence of myogenic and endothehal cell lineages in avian embryos. Dev. BioZ. 163,270-278. Wachtler, F., Christ, B., and Jacob, H. J. (1982a). Grafting experiments on determination and migratory bahaviour of presomitic, somitic and somatopleural cells in avian embryos. Anat. Embryol. (Berl.) 164,369-378.

64

HIRSINGER

fl

AL.

Wachtler, F., Christ, B., and Jacob, H. J. (1982b). Grafting experiments on determination and migratory behaviour of presomitic, somitic and somatopleural cells in avian embryos. Anat. Embryol. 164 369-378. Wang, H. U., and Anderson, D. J. (1997). Eph family transmembrane ligands can mediate repulsive guidance of trunk neural crest migration and motor axon outgrowth. Neuron l&383-396. Wang, Y., Benezra, R., and Sassoon, D. A. (1992). Id expression during mouse development: A role in morphogenesis. Dev. Dyn. 194,222~230. Wang, S., Krinks, M., Lin, K., Luyten, F. P., and Moos, M. J. (1997a). Frzb, a secreted protein expressed in the Spemann organizer, binds and inhibits Wnt-8. Cell g&757-766. Wang, S., Ktinks, M., and Moos, M., Jr. (1997b). Frzb-1, an antagonist of Wnt-1 and Wnt8, does not block signaling by Wnts3A, -5A, or -11. Biochem. Biophys. Rex Commun. 236,502~504. Watterson, R., Fowler, I., and Fowler, B. J. (1954). The role of the neural tube and notochord in development of the axial skeleton of the chick. Am. J. Anat. 95,337-399. Weinmaster, G., Roberts, V. J., and Lemke, G. (1992). Notch2: A second mammalian Notch gene. Development 116, 931-941. Weinstein, D. C., Ruiz i Altaba, A., Chen, W. S., Hoodless, P., Prezioso, V. R., Jessell, T. M., and Dame& J. E., Jr. (1994). The winged-helix transcription factor HNF-3 beta is required for notochord development in the mouse embryo. Cell 78,575-588. Weintraub, H., Davis, R., Tapscott, S., Thayer, M., Krause, M., Benezra, R., Blackwell, T. K., Turner, D., Rupp, R., and Hollenberg, S. (1991). The myoD gene family: nodal point during specification of the muscle cell lineage. Science 251,761-766. Williams, B. A., and Ordahl, C. P. (1997). Emergence of determined myotome precursor cells in the somite. Development 124, 4983-4997. Williams, L. W. (1910). The somites of the chick. Am. J. Anat. 11,55-100. Williams, R., Lendahl, U., and Lardelli, M. (1995). Complementary and combinatorial patterns of Notch gene family expression during early mouse development. Me&. Dev. 53,357-368. Wilson, V., Manson, L., Skarnes, W. C., and Beddington, R. S. (1995). The T gene is necessary for normal mesodermal morphogenetic cell movements during gastrulation. Development 121,877~886. Wilting, J., Brand-Saberi, B., Huang, R., Zhi, Q., Kontges, G., Ordahl, C. P., and Christ, B. (1995). Angiogenic potential of the avian somite. Dev. Dyn. 202, 165-171. Winnier, G., Blessing, M., Labosky, P. A., and Hogan, B. L. (1995). Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 9,2105-2116. Wolf, C., Thisse, C., Stoetzel, C., Thisse, B., Gerlinger, P., and Perrin-Schmitt, F. (1991). The M-twist gene of Mus is expressed in subsets of mesodermal cells and is closely related to the Xenopus X-twi and the Drosophila twist genes. Dev. Bid. 143,363-373. Wong, P. C., Zheng, H., Chen, H., Becher, M. W., Sirinathsinghji, D. J., Trumbauer, M. E., Chen, H. Y., Price, D. L., Van der Ploeg, L. H., and Sisodia, S. S. (1997). Presenilin 1 is required for Notch1 and DIIl expression in the paraxial mesoderm. Nature 387,288-292. Yamaguchi, T. P., Harpal, K., Henkemeyer, M., and Rossant, J. (1994). fgfr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev. 8, 3032-3044. Yang, Y., Drossopoulou, G., Chuang, P. T., Duprez, D., Marti, E., Bumcrot, D., Vargesson, N., Clarke, J., Niswander, L., McMahon, A., and Tickle, C. (1997). Relationship between dose, distance and time in Sonic Hedgehog-mediated regulation of anteroposterior polarity in the chick limb. Development 1244393-4404.

s

SOMITQGENESIS

65

Zhang, N., and Gridley, T. (1998). Defects in somite formation in lunatic fringe-deficient mice. Nature 394, 374-377. Zimmerman, L. B., De Jesus-Escobar, J. M., and Garland, R. M. (1996). The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. CeZZ86, 599406.