Regulation of Paraxis Expression and Somite Formation by Ectoderm- and Neural Tube-Derived Signals

Home , Ectoderm, Somite, Surface ectoderm

DEVELOPMENTAL BIOLOGY 185, 229–243 (1997) ARTICLE NO. DB978561 Regulation of paraxis Expression and Somite Formation by Ectoderm- and Neural Tube-Derived Signals

Drazˇen Sˇosˇic´,* Beate Brand-Saberi,† Corina Schmidt,† Bodo Christ,† and Eric N. Olson*,1 *Hamon Center for Basic Cancer Research, The University of Texas, Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, Dallas, Texas 75235-9148; and †Institute of Anatomy, University of Freiburg, P.O. Box 111, 79001 Freiburg i. Brsg., Germany

During vertebrate embryogenesis, the paraxial mesoderm becomes segmented into somites, which form as paired epithelial spheres with a periodicity that reflects the segmental organization of the embryo. As a somite matures, the ventral region gives rise to a mesenchymal cell population, the sclerotome, that forms the axial skeleton. The dorsal region of the somite remains epithelial and is called dermomyotome. The dermomyotome gives rise to the trunk and limb muscle and to the dermis of the back. Epaxial and hypaxial muscle precursors can be attributed to distinct somitic compartments which are laid down prior to overt somite differentiation. Inductive signals from the neural tube, notochord, and overlying ectoderm have been shown to be required for patterning of the somites into these different compartments. Paraxis is a basic helix– loop–helix transcription factor expressed in the unsegmented paraxial mesoderm and throughout epithelial somites before becoming restricted to epithelial cells of the dermomyotome. To determine whether paraxis might be a target for inductive signals that influence somite patterning, we examined the influence of axial structures and surface ectoderm on paraxis expression by performing microsurgical operations on chick embryos. These studies revealed two distinct phases of paraxis expression, an early phase in the paraxial mesoderm that is dependent on signals from the ectoderm and independent of the neural tube, and a later phase that is supported by redundant signals from the ectoderm and neural tube. Under experimental conditions in which paraxis failed to be expressed, cells from the paraxial mesoderm failed to epithelialize and somites were not formed. We also performed an RT-PCR analysis of combined tissue explants in vitro and confirmed that surface ectoderm is sufficient to induce paraxis expression in segmental plate mesoderm. These results demonstrate that somite formation requires signals from adjacent cell types and that the paraxis gene is a target for the signal transduc- tion pathways that regulate somitogenesis. ᭧ 1997 Academic Press

INTRODUCTION mation have been documented in detail, little is known of the transcriptional mechanisms that control somite forma- Somites are segmental units of the paraxial mesoderm tion and patterning. that form in a rostral-to-caudal progression during verte- Immediately after they bud off from the segmental plate, brate embryogenesis. The reiterative arrangement of so- somites appear as paired epithelial spheres that surround a mites along the rostrocaudal axis reﬂects the segmental or- loose aggregate of mesenchymal cells. As somites mature, ganization of the vertebrate embryo and imposes segmental they undergo a series of morphological and molecular patterning on the axial skeleton, the intrinsic back muscu- changes that culminate with the formation of three somitic lature, the peripheral nervous system, and the vasculature. compartments, the dermatome, myotome, and sclerotome While the morphological events associated with somite for- (reviewed in Keynes and Stern, 1988; Christ and Ordahl, 1995; Brand-Saberi et al., 1996). Initially, the ventral region Sequence data from this article have been deposited with the of the newly formed somite undergoes an epithelial-to-mes- EMBL/GenBank Data Libraries under Accession No. U76665. enchymal transition, giving rise to the sclerotome, the ori- 1 To whom correspondence should be addressed. Fax: (214)648- gin of the axial skeleton. The remaining epithelial sheet in 1196. E-mail: [email protected]. the dorsal somite, referred to as the dermomyotome and

0012-1606/97 $25.00 Copyright ᭧ 1997 by Academic Press All rights of reproduction in any form reserved. 229

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba 230 Sˇ osˇic´ et al.

FIG. 1. Deduced primary amino acid sequence of chicken paraxis protein and its homology with mouse paraxis. (A) Nucleotide sequence of chicken paraxis cDNA and deduced amino acid sequence of paraxis protein. The bHLH region is indicated in bold. (B) Homology between chicken and mouse paraxis proteins. The bHLH region is underlined. later the dermotome, gives rise to the myotome at its crani- the ventrolateral body wall (Christ et al., 1974, 1977, 1983; omedial edge (Kaehn et al., 1988; Christ and Ordahl, 1995). Chevallier et al., 1977; Ordahl and Le Douarin, 1992). The myotome consists of a layer of postmitotic, differenti- Cells surrounding the somites play important roles in ated skeletal muscle cells that form the back muscles (for speciﬁying and patterning of different somitic cell lineages. a review, see Christ et al., 1990). The lateral part of the Sclerotome differentiation occurs in response to the se- dermomyotome gives rise to the muscles of the limb and creted morphogen sonic hedgehog, which is produced by

FIG. 2. Expression of paraxis transcripts during chick embryogenesis. Paraxis mRNA was detected by whole-mount in situ hybridization. (A) HH stage 5 chick gastrula. Paraxis mRNA is present in prospective segmental plate mesoderm (arrows). (B) Stage 7 embryo. Paraxis is expressed in the rostral end of the segmental plate and in newly formed somites. (C) HH stage 11 embryo (12 somites), ventral view. Paraxis transcripts are found in the segmental plate and somites. (D) Higher magniﬁcation of the embryo shown in (C). The rostrocaudal gradient of paraxis expression can be observed with highest expression in the segmental plate and lowest in the rostral-most somites. (E) HH stage 19 chick embryo. Paraxis is expressed in all somites and in the segmental plate. (F) Higher magniﬁcation of the embryo shown in (E). The most recently formed somites show uniform expression of paraxis. More mature somites show highest expression in the mediocaudal quadrant, while the rostral-most somites express paraxis predominantly at their rostral and caudal edges. Arrows in (F) indicate paraxis expression at the rostral and caudal edges of the somite. The arrowhead in (F) points to precursors of tongue muscle. Staining in the head in (E) and (F) is background. Abbreviations: ps, primitive streak; s, somite; sp, segmental plate; nt, neural tube.

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 231

AID DB 8561 / 6x23$$8561 05-07-97 11:44:50 dba 232 Sˇ osˇic´ et al. hybridization. (A) Schematic in situ Transverse sections of HH stages 11 and 16 chick embryos showing paraxis expression detected by FIG. 3. illustration of cross-sections throughshown HH in stage (B), 11 (C), andparaxial and 16 mesoderm. (D), embryos. (C) respectively. As The (B)an lines somites Section anterior designated mature, through paraxis the B, somite rostral expression Cmyotome. that end Abbreviations: is and has nt, of lost D neural differentiated the in in tube; segmental themyotome; into this no, s, sclerotome, plate. panel notochord; dermatome, Paraxis sclerotome; but i, indicate d, myotome, intermediate mRNA persists the is mesoderm; dermatome; in and planes so, found m, the sclerotome. of somatic throughout myotome. dermomyotome. mesoderm; section the Paraxis (D) sp, entire Section expression splanchnic mesoderm; through is dm, dermo- restricted to the dermatome and

AID DB 8561 / 6x23$$8561 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 233 the notochord and ventral neural tube (Brand-Saberi et al., expression of paraxis during chick embryogenesis. We show 1993; Pourquie et al., 1993; Fan and Tessier-Lavigne, 1994; that there are two distinct phases of paraxis expression in Johnson et al., 1994; Fan et al., 1995; Ebensperger et al., paraxial mesoderm and developing somites. The initial ex- 1995). Induction of muscle gene expression in the myotome pression of paraxis in unsegmented paraxial mesoderm is has also been shown to be influenced by signals from the dependent on the surface ectoderm and independent of the notochord and the dorsal neural tube (Rong et al., 1990, neural tube, whereas the later expression of paraxis in the 1992; Buffinger and Stockdale, 1994, 1995; Mu¨nsterberg and dermomyotome appears to be dependent on redundant sig- Lassar, 1995; Stern and Hauschka, 1995; Pownall et al., nals from the neural tube and surface ectoderm. In regions 1996; Spense et al., 1996; Amthor et al., 1996). Members of of manipulated chick embryos in which paraxis was not the wingless family of growth factors (Wnts), which are expressed, the paraxial mesoderm failed to epithelialize, re- secreted by the dorsal neural tube, and sonic hedgehog ap- sulting in the absence of epithelial somites and the dermo- pear to mediate the effects of axial organs on myotome myotome. These results demonstrate the importance of cel- development (Stern et al., 1995; Mu¨ nsterberg et al., 1995). lular signaling for somitogenesis and are consistent with Myogenic cells of the ventrolateral portion of the dermomy- the notion that the paraxis gene is a target for redundant otome, which form the lateral muscle cell lineage, appear signaling pathways from the ectoderm and neural tube that to be influenced by different signals than the medial lineage control somite formation and epithelialization of the para- (Gamel et al., 1995; Pourquie et al., 1995). Bone morphogen- xial mesoderm. etic protein-4 (BMP-4), which is expressed in the lateral mesoderm, inhibits differentiation of the lateral myogenic lineage (Pourquie et al., 1996). The ectoderm has also been MATERIAL AND METHODS shown to play an important role in patterning of the somites and in activation of lineage-specific genes. Culture of bra- Embryos chial somites with surface epithelium promotes myogen- Fertilized White Leghorn chicken eggs (Gallus Domesticus) were esis and decreases chondrogenesis within the somites obtained from Texas A&M University and incubated at 38.5ЊCin (Kenny-Mobbs and Thorogood, 1987; Cossu et al., 1996). a humidified incubator. Embryos were staged according to Ham- Gene expression in the dermomyotome is also dependent burger and Hamilton (1951). on signals from the ectoderm. Expression of the dermomyotomal markers Pax3, Sim1, and Pax7 is induced in explants Library Screening of somites cultured in close contact with surface ectoderm (Fan and Tessier-Lavigne, 1994). Conversely, removal of the A stage 14–17 (Hamburger and Hamilton, 1951) chick cDNA surface ectoderm leads to cessation of expression of another library (generously provided by Gregor Eichele, Baylor College of 32 dermomyotomal marker, MHox, in the underlying somites Medicine, Houston, TX) was screened using P-labeled full-length (Kuratani et al., 1994). paraxis cDNA as a probe (Burgess et al., 1995), under low stringency While the importance of inductive signals in specification conditions (Edmondson and Olson, 1989). Positive clones were pu- rified, converted to the plasmid form, and sequenced using auto- and patterning of somitic cell lineages has been well docu- mated DNA sequencing. mented, little is known of the potential role of cell–cell signaling in the earlier steps of somite formation. Previous studies have demonstrated that somites can form from iso- In Situ Hybridization and Histology lated paraxial mesoderm in vitro in the absence of neural Embryos were removed from the yolk and washed briefly in PBS tube and notochord (Bellairs, 1963; Packard and Jacobson, before overnight fixation at room temperature in 4% paraformalde- 1976). These studies led to the conclusion that the ability hyde/PBS. Embryos were stored in 70% ethanol at 020ЊC until to form somites is an intrinsic property of the paraxial meso- needed. Probes were prepared according to Genius protocols (Boeh- derm that is independent of extrinsic influences. However, ringer-Mannheim), and whole-mount in situ hybridization was per- in those earlier studies of somitogenesis in vitro, ectoderm formed as described (Barth and Ivarie, 1994), with the only excep- was included with the paraxial mesoderm, raising the possi- tion that we omitted polyvinyl alcohol from the color reaction bility that it might provide a signal for somite formation. mixture. After the color reaction was considered complete (4–24 Paraxis (Burgess et al., 1995), also known as bHLH-EC-2 hr), embryos were washed in methanol for 3 hr and postfixed with (Quertermous et al., 1994) and Meso-1 (Blanar et al., 1995), 4% paraformaldehyde/0.1% glutaraldehyde. Selected embryos were processed for paraffin sectioning as described (Lyons et al., 1991). is a basic helix–loop–helix (bHLH) transcription factor that Staining with hematoxylin and eosin was performed as described is expressed in the paraxial mesoderm immediately prior to (Martin et al., 1995). Whole embryos and sections were photo- somite formation and in newly formed epithelial somites. graphed using a Nikon zoom stereo microscope with Kodak Ekta- As somites mature, paraxis expression becomes localized chrome 64T film. to the epithelial cells of the dermomyotome. Recent gene knockout experiments in mice have shown that paraxis is required for the formation of epithelial somites (Burgess et Embryo Surgery al., 1996). Here, we investigated the potential influences of Chick embryos undergoing surgery ranged in stages from 10 to axial organs and surface ectoderm on somite formation and 14 Hamilton–Hamburger (HH). In all cases, they were incubated

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba 234 Sˇ osˇic´ et al.

at 38ЊC and 78% humidity. In a first series of experiments, quail embryos (Coturnix coturnix japonica) served as donors of the notochords and White Leghorn chick embryos as hosts. Notochords from quail embryos ranging from HH stages 11–14 were prepared as follows. The quail donor was dissected from the egg and transferred to a petri dish containing 0.4 % trypsin in Locke’s solution. The endoderm was removed from underneath the notochord at the level of the segmental plate after approximately 1 min by tungsten needles. After removal of the paraxial and lateral plate mesoderm the remaining neural tube–ectoderm–notochord complex from the caudal region of the embryo was thoroughly rinsed in horse serum to stop the enzymatic digestion. The dissection of the notochord was then carried out in Locke’s solution containing penicillin. The graft was transferred to the chick host by means of a Spemann pipet. The notochordal graft was inserted into a channel prepared by a glass needle between surface ectoderm and intermediate mesoderm. The grafts differed in length and were estimated to be 300– 700 mm long. The surviving hosts were sacrificed 15 hr to 2 days after the operation and fixed in 4% paraformaldehyde/PBS for in situ hybridization. In a second series, both neural tube and notochord were removed at the level of caudal segmental plate. Neural tube alone was then grafted back into its original position and hosts were sacrificed about 20 hr later and fixed. In a third series, the neural tube was removed at the level of the caudal unsegmented paraxial mesoderm. In a fourth series, a cut was made between the paraxial organs on the one hand and the caudal portion of the segmental plate on the other hand and a 1-mm-thick gold foil was inserted into the slit. In a fifth series, surface ectoderm overlying the entire segmental plate and the three caudal-most somites was removed from one side of the embryo. In a sixth series, ectoderm was removed and gold foil was inserted between the neural tube and paraxial mesoderm at the level of the caudal segmental plate.

Culturing of Embryo Explants and RT-PCR

Mesoderm was isolated from the caudal region of the segmental plate. Ectoderm, endoderm, notochord, neural tube, and lateral plate mesoderm surrounding the rostral portion of the segmental plate were isolated (see Fig. 8A). Mesoderm from the caudal segmental plate was cultured alone or in the presence of one of the above mentioned tissues. All explants were cultured on collagen gels for 1 day, as described previously (Mu¨ nsterberg et al., 1995). The methods for RNA isolation from explants and RT-PCR were described previously (Mu¨ nsterberg et al., 1995). All RT-PCR results were conﬁrmed in at least three independent experiments. The

the region in which notochord was deleted from separate embryos is in blue. The dashed line indicates the approximate level of sections in C and D. (B). Whole-mount in situ hybridization of an embryo containing an ectopic notochord on the right side using paraxis as a probe. There is no change in paraxis expression upon grafting of the notochord. Arrowheads mark the approximate level FIG. 4. Notochord does not affect paraxis expression. (A) Sche- of notochord graft. (C) Transverse section of the embryo in (B) at matic illustration of operations performed on embryos. Notochord the level of the graft. The ectopic notochord is indicated no*. The was grafted to HH stage 10 chick embryos at the level of the caudal paraxis expression domain is not affected by notochord graft. (D) three to eight somites and embryos were ﬁxed at HH stage 15. In Transverse section of the embryo in which the notochord was ab- a second operation, the notochord was removed at the level of the lated. An asterisk designates the position where the notchord caudal segmental plate of HH stage 13 chick embryo. The embryo would normally be located. Abbreviations: nt, neural tube; no, no- was ﬁxed 17 hr later. The ectopic notochord is indicated in red and tochord.

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 235

FIG. 5. Disruption of neural tube signaling affects only the late phase of paraxis expression in somites. (A) shows a schematic diagram of the caudal somites and segmental plate and the region of gold foil insertion. B and C are whole-mounts of the caudal ends of embryos and D and E are transverse sections. (B) Gold foil was inserted between the neural tube/notochord and segmental plate on the right side of a HH stage 13 embryo and 6 hr later the embryo was fixed (gold foil was discarded during fixation) and paraxis transcripts were detected by whole-mount in situ hybridization. There is no change in the early paraxis-expression domain in the caudal end of the embryo. An asterisk indicates the position of the gold foil that was discarded during fixation. (C) Gold foil was inserted between the neural tube/notochord and segmental plate of a HH stage 13 embryo and 24 hr later the embryo was fixed and hybridized with paraxis probe. Paraxis is expressed only at the medial and lateral edges of the paraxial mesoderm in the region deprived of contact with the neural tube/notochord upon insertion of the gold foil (indicated by an arrow). (D) Transverse section through the operated region of the embryo shown in (C). The operation reduces paraxis expression to two regions located in the medial and dorsal part of the paraxial mesoderm (indicated by arrows), whereas paraxis is expressed throughout the dermomyotome on the unoperated side. (E) Transverse section through the operated region of a chick embryo from which the neural tube at the segmental plate level was excised at HH stage 12. The embryo was fixed 18 hr after the operation. Paraxis is weakly expressed in the dorsomedial and dorsolateral regions of the paraxial mesoderm (arrows). Abbreviations: nt, neural tube; no, notochord; sp, segmental plate. glyceraldehyde 3-dehydrogenase (GAPDH) and paraxis reaction a chicken paraxis cDNA encoding the entire open reading products were amplified for 23 and 30 cycles, respectively. Primers frame (Figs. 1A and 1B, see Materials and Methods). The bHLH used for paraxis PCR amplification were 5؅-AAGGTGCCCAGG- region of this protein had 96% amino acid identity with the .(AAGACGGGG-3؅ (nucleotides 623–643) and 5؅-TCATCTCCG- corresponding region of mouse paraxis (Burgess et al., 1995 TGCCACTCGCAG-3؅ (nucleotides 1009–1030). These primers Outside the bHLH region, the chick and mouse proteins shared yielded a PCR product of 410 bp. The primers used for GAPDH 66% amino acid identity and 85% homology. This chicken PCR amplification were described previously (Mu¨ nsterberg et al., 1995) and yielded a product of 330 bp. The above numbers of cycles paraxis protein also showed high homology to the human of PCR amplification were within the linear range. PCR amplifica- (Quatermous et al., 1994), hamster (Blanar et al., 1995), and tion was also shown to be dependent on reverse transcriptase and frog (K. Ligon and E. Olson, unpublished) proteins (not shown). both primers. Five microliters of each PCR reaction was separated The expression pattern of paraxis was analyzed by whole- on a 6% acrylamide gel, which was dried and exposed to X-ray film mount in situ hybridization to embryos from gastrulation to for 12 hr. In all cases, PCR products were of the predicted sizes. the onset of organogenesis (stages 3 to 19; Hamburger and The PCR primers used to detect paraxis transcripts corresponded Hamilton, 1951). Paraxis mRNA expression was first observed to sequences in different exons of the gene, which made it possible at HH stage 5, when it marked a subpopulation of cells lateral to confirm that the PCR products were derived from RNA and not to the primitive streak (Fig. 2A). Although it is difficult to genomic DNA contamination. Identities of PCR products were also clearly distinguish between various prospective regions of the confirmed by restriction enzyme digestion. chick gastrula at this stage of development, the domain of paraxis expression appeared to correspond with the area of prospective segmental plate mesoderm and somites (Christ et RESULTS al., 1972; Schoenwolf and Watterson, 1989; Selleck and Stern, 1991). Later in development, paraxis mRNA was expressed Cloning and Expression of Chick paraxis throughout the segmental plate and newly formed somites To investigate the potential influence of axial organs on (Figs. 2B– 2F); its expression was restricted to the paraxial expression of paraxis during chick embryogenesis, we isolated mesoderm and was not detected in intermediate or lateral

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba 236 Sˇ osˇic´ et al. in situ hybridization in the region of in situ hybridization. (B) The expression of paraxis transcripts by whole-mount in situ hybridization. (F) The expression of paraxis transcripts by whole-mount in situ Paraxis expression is influenced by the overlying ectoderm. (A) A schematic diagram of the surgical manipulations that were performed the surgery. (G) Aside transverse where section paraxis through would the normallythe region be control of expressed. side the (H) surgery. of Theplate; The the control no, arrow embryo. notochord; side indicates of Surface nt, the the neural ectoderm region embryo tube; is of with sp, the indicated a paraxial segmental in normal plate. mesoderm blue dermomyotome. on Paraxis in the is A operated expressed and only E. on Abbreviations: ec, ectoderm; dm, dermomyotome; np, neural hybridization at the caudalrespectively. (E) end A of schematic the diagramend of embryo. of the the The surgical embryo lines manipulationstranscripts and were designated that a were detected C gold performed by and foil in barrier D F –H. was in Surface additionally ectoderm this inserted was panel on removed the denote from left the the side. caudal level After of incubation sections for shown 20 hr, in the (C) embryo and was (D), fixed and paraxis FIG. 6. in B–D. (A –D)was Surface fixed ectoderm and was removed paraxis from transcripts the caudal were end detected of by the embryo, only on the left side. After incubation for 6 hr, the embryo

AID DB 8561 / 6x23$$8561 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 237 Dependance of somite epithelialization on signals from the neural tube and surface ectoderm. Surface ectoderm was removed and a gold FIG. 7. foil barrier was insertedembryos between the were neural ﬁxed, tubecontralateral sectioned, and operated paraxial side. and C mesoderm stained andof at D with the more show mature caudal H&E. sagittal end sections somites.points A of at Note to the and the the the level embryo, C normal complete of as location the show absence shown ﬁrst of in of sections epithelial surface somite. Fig. epithelial ectoderm, from A 6E. somites which and the Twenty was (D) B hr removed. unoperated and show later, transverse a side. sections dermomyotome at B (B) the and level on D the show operated side. sections The from arrow the in (B)

AID DB 8561 / 6x23$$8561 05-07-97 11:44:50 dba 238 Sˇ osˇic´ et al. mesoderm (Fig. 3B). Within the somites, there were spatial gold foil insertion between the neural tube and segmental differences in paraxis expression along the rostrocaudal axis of plate (Figs. 5C and 5D). Whereas normal somites at this the embryo. Whole mounts showed that soon after somite stage of development express paraxis throughout the dermo- formation, paraxis expression declined in the lateral region of myotome, paraxial mesoderm that was separated from the the somite (Figs. 2D–2F). Later in development, paraxis tran- neural tube by gold foil expressed paraxis in only two re- scripts were expressed at the highest levels at the cranial and gions, one located medially and the other dorsally (Fig. 5D). caudal edges of the more mature somites (Fig. 2F). We also excised the neural tube at the segmental plate As the dermomyotome and sclerotome formed, paraxis level and examined the effect on paraxis expression. Under transcripts disappeared from the ventral regions of the somite, these conditions, paraxis was expressed in a pattern similar but they persisted in the dermomyotome (Fig. 3C). When the to that observed in paraxial mesoderm separated from the dermomyotome developed further into the dermatome and neural tube by the gold foil barrier (Fig. 5E). Together, these myotome, paraxis transcripts were detected in both compart- results demonstrate that the initial phase of paraxis expres- ments, although they were more prominent in the dermatome sion in the segmental plate does not require persistant sig- (Fig. 3D). The presence of paraxis transcripts in the newly nals from the neural tube. However, the neural tube appears formed myotome may reflect the fact that myotomal cells are to play a role in determining the spatial pattern of paraxis derived from the dermomyotome. At later stages, paraxis was expression in the developing somite. not expressed in the differentiated myotome. Beginning at HH stage 18, paraxis expression was also detected in migrating Expression of Paraxis Is Dependent on the precursors of tongue muscle (Fig. 2F) (see van Bemmelen, Overlying Ectoderm 1889; Schemainda, 1979). Paraxis expression was also ob- We also examined whether surface ectoderm might influ- served in the limb buds at this stage, which is likely to reflect ence paraxis expression. Indeed, when the surface ectoderm muscle precursor cells emigrating from the dermomyotome over the segmental plate was removed on one side of the em- where paraxis is expressed. bryo, the initial expression of paraxis in the segmental plate was delayed (Figs. 6B and 6C). These results suggested that the Effects of Notochord and Neural Tube on Paraxis ectoderm was required for the early phase of paraxis expression Expression in the adjacent segmental plate mesoderm. However, as the segmental plate matured, paraxis expression appeared to be- To begin to investigate whether paraxis expression in the come independent of the overlying ectoderm (Figs. 6B and 6D). paraxial mesoderm was influenced by neighboring cell We hypothesized that the neural tube might provide a types, we grafted a segment of a quail notochord between secondary signal that could substitute for ectodermally-de- the intermediate mesoderm and ectoderm at the level of rived signals to maintain the later phase of paraxis expres- the caudal three to eight somites of stage 10 chick host sion. To test this, we removed the overlying ectoderm and embryos (Figs. 4A–4C). After incubation for 21 hr, embryos inserted gold foil between the neural tube and the segmental at HH stage 15 were isolated and expression of paraxis was plate (Fig. 6E). Under these conditions, paraxis failed to be examined by in situ hybridization. An ectopic notochord expressed in the region of the operation (Fig. 6F). did not affect paraxis expression (Figs. 4B and 4C). Thin sections of the above embryos showed no evidence We also excised the notochord from the region of the for epithelial somites or for the dermomyotome under con- segmental plate in which paraxis was not expressed (Fig. ditions in which paraxis was not expressed (Fig. 6G). How- 4A). Operations were performed on stage 13 chick embryos ever, on the contralateral unoperated side, somite formation and expression of paraxis transcripts was then examined 17 and compartmentalization proceded normally (Fig. 6H). Sec- hr later. As shown in Fig. 4D, activation of paraxis expres- tions of these embryos were stained with H&E. Whereas sion in the segmental plate mesoderm was unaffected by on the control side, epithelial somites and dermomyotomes notochord ablation. Thus, there was no evidence for a role were apparent, the operated side showed only mesenchymal of the notochord in paraxis regulation. cells within the paraxial mesoderm (Fig. 7). These results We next examined the potential role of the neural tube suggest that the surface ectoderm provides an initial signal in the regulation of paraxis expression by inserting an im- that supports paraxis expression and that a later signal from permeable gold foil barrier between the neural tube and the the neural tube can substitute at least partially for this sig- caudal segmental plate beginning at a level in which paraxis nal. The absence of epithelial cells within the paraxial transcripts were not detected by in situ hybridization and mesoderm under conditions in which paraxis failed to be extending to the first caudal somite, where paraxis is ex- expressed is consistent with the notion that paraxis is re- pressed (Fig. 5A). Expression of paraxis was then examined quired for this step in somitogenesis (Burgess et al., 1996). by in situ hybridization at various times after the operation. Six hours after insertion of gold foil between the neural tube and the segmental plate, there was no change in paraxis Surface Ectoderm Is Sufficient to Induce Paraxis expression in adjacent somites and segmental plate (Fig. 5B). Expression in Vitro However, the paraxis expression domain was dramatically To further test the responsiveness of paraxis expression to reduced in somites at the level of the operation, 24 hr after ectoderm-derived signals, we examined the effect of surface

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 239

FIG. 8. Induction of paraxis mRNA in cultured explants of segmental plate mesoderm by surface ectoderm. (A) A schematic diagram of the surgical manipulations and assays that were performed. The caudal segmental plate was dissected from HH stage 11 chick embryos. This region does not express paraxis as judged by in situ hybridization (Fig. 2C) or by RT-PCR (t Å 0). After 1 day of incubation, RNA was isolated from the cultures and transcripts for paraxis and GAPDH were assayed by RT-PCR. Similar results were obtained in three independent experiments. (B) RT-PCR products were analyzed by separation on a 6% acrylamide gel. Paraxis products were detected only in the presence of ectoderm plus segmental plate, whereas GAPDH products were present in all samples. Explants were cultured alone or in the presence of neural tube, notochord, endoderm, ectoderm, or lateral mesoderm for 1 day as indicated.

ectoderm on expression of paraxis mRNA in isolated segmen- DISCUSSION tal plate mesoderm in vitro (Fig. 8A). Paraxis mRNA was undetectable after 30 cycles of RT-PCR in explants of caudal The results of this study demonstrate that surface ecto- segmental plate isolated from stage 11 chick embryos (Fig. derm and neural tube are the sources of inductive signals 8B). As expected, paraxis mRNA was undetectable in isolated required for paraxis expression and somite formation. In surface ectoderm from the same region of the embryo (Fig. experimentally manipulated embryos in which paraxis 8B). However, when the segmental plate and surface ectoderm was not expressed in the paraxial mesoderm, there was were cocultured for 1 day, paraxis transcripts were readily no evidence for the formation of epithelial somites or a detected. In contrast, coculture of paraxial mesoderm with dermomyotome. These findings are consistent with previ- neural tube, notochord, endoderm, or lateral mesoderm failed ous studies of paraxis-null mice, which lack somites, and to induce paraxis expression. Surface ectoderm was also un- suggest that paraxis mediates the effects of surface ecto- able to induce paraxis expression in lateral mesoderm, indicat- derm and axial organs on somitogenesis. In this process, ing that there is specificity in the ability of different mesoder- two morphogenetic events should be distinguished: one is mal cell types to respond to the ectodermal signal. Transcripts epithelialization and the other is metamerization. While for GAPDH, which is expressed constitutively, were present epithelialization seems to depend on contact with adja- at comparable levels under all conditions. These results con- cent tissues, metamerization occurs independently. firm the results of surgical manipulations and demonstrate Primitive metamers in vertebrates are reflected by somi- that surface ectoderm alone can induce paraxis expression in tomeres (Meier, 1979; Jacobson, 1988). Metamerization adjacent paraxial mesoderm. seems to be an intrinsic property of the paraxial meso-

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba 240 Sˇ osˇic´ et al. derm that is independent of inducing signals (Bellairs, 1963; Christ et al., 1972). This phenomenon also explains the ﬁnding that in paraxis-null mice mesenchymal metamers can be seen, despite the absence of epithelial somites (Burgess et al., 1996).

Paraxis as a Mediator of Inductive Signals Regulating Somitogenesis In the chick embryo, paraxial mesoderm includes two morphologically distinct portions; an unsegmented part at the posterior end of the embryo, also called the segmental plate; and segmented units, called somites. All of the somites form from the segmental plate, at an approximate rate of one pair every 100 min. During this dynamic process, FIG. 9. A model for the role of paraxis in the control of somite the size of the segmental plate is kept relatively constant epithelialization and responsiveness to inductive signals. Activa- by mitosis within the plate and by recruitment of the cells tion of paraxis expression occurs in response to signals from the from the primitive streak and tail bud, which are positioned ectoderm. Maintenance of paraxis expression is influenced both by ectoderm- and neural tube-derived signals. more posteriorly. Since paraxis is expressed only in the anterior part of the segmental plate, we reasoned that cells from the posterior segmental plate, upon their posterioanterior translocation must come in contact with certain signal(s) cSim1 expression (Pourquie et al., 1996). The restricted ex- that would activate paraxis gene expression. Indeed, our pression of paraxis to two localized domains in the absence results reveal that this signal(s) is provided by neighboring of neural tube signal is likely to reflect combinations of tissues. Our results reveal two distinct phases of paraxis positive and negative signals that are revealed when the expression, activation and maintenance. Activation of par- neural tube signal is removed. Expression of the homeobox axis expression appears to be dependent on the presence of gene MHox and Pax-3 in the dermomyotome has also been the surface ectoderm and independent of the neural tube shown to be dependent on surface ectoderm (Kuratani et (Figs. 5B and 6B). On the unoperated side of the embryo one al., 1994; Fan and Tessier-Lavigne, 1994). The fact that Pax- can observe the boundary between paraxis nonexpressing 3 is expressed in paraxis-null mice demonstrates that Pax- and -expressing domains of the segmental plate (Figs. 5B and 3 and paraxis are regulated independently. 6B) Presumably at this boundary, cells from the posterior Consistent with the conclusion that ectoderm has par- segmental plate receive signals that enable them to express axis-inducing properties, coculture of segmental plate paraxis. While this process is undisturbed after neural tube mesoderm with surface ectoderm resulted in induction of separation (Fig. 5B), ectoderm removal appears to abolish, paraxis expression. We conclude from these results that sur- or at least delay it (Fig. 6B). Further evidence that neural face ectoderm is necessary and sufficient to induce paraxis tube does not participate in paraxis gene activation comes expression in the paraxial mesoderm in the absence of other from explant experiments, which showed that isolated neu- cell types. Fan and Tessier-Lavigne (1994) also used an in ral tube could not induce paraxis expression in an explant vitro explant assay to show that nonneural ectoderm can of the segmental plate. Following paraxis activation, its ex- induce the expression of several dermomyotomal markers pression becomes supported by redundant signals emanat- in segmental plate mesoderm. We found no evidence for a ing from the ectoderm and neural tube. In the absence of role of the notochord in the regulation of paraxis expression, neural tube, paraxis expression was reduced and restricted although it is known that such grafts affect the expression to two regions of the dorsal somite, one located medially of many other dermomyotome markers (Brand-Saberi et al., and the other laterally (Figs. 5D and 5E). This residual par- 1993; Goulding et al., 1994; Bober et al., 1994). axis expression was eliminated when the surface ectoderm We do not yet know the identity of the factor(s) produced was removed (Fig. 6G). The localized expression of paraxis by the neural tube or ectoderm that influence paraxis ex- in these two domains when signaling from the neural tube pression or whether these two sources produce the same or is blocked suggests that there is heterogeneity among cells different inducing factors. There is evidence to suggest that from the dorsal paraxial mesoderm or that signaling from the nonneural ectoderm provides a signal (or signals) that the ectoderm is nonuniform. specifies dorsal cell types within the neural tube and that Previous studies have shown that patterning of the so- the response is dependent on the competence of the neural mites and expression of dermomyotomal markers are con- tissue (Dickinson et al., 1995). Several members of the Wnt trolled by antagonistic signals from surrounding cell types. family are expressed in the dorsal neural tube (Roelink and BMP-4, produced by the lateral mesoderm, induces expres- Nusse, 1991; Parr et al., 1993). BMP-4 and -7 are also ex- sion of the bHLH gene cSim1 in the lateral dermomyotome, pressed in the surface ectoderm and dorsal neural tube, for example, whereas signals from the neural tube repress where they have been implicated in dorsoventral patterning

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 241 of the neural tube by antagonizing the ventralizing activity al., 1978) and upregulation of several extracellular matrix of sonic hedgehog (Liem et al., 1995). molecules and of cell adhesion molecules such as N-cad- herin and N-CAM (Duband et al., 1987). Fibronectin and laminin have also been implicated in epithelialization of Paraxis Regulates Epithelialization in the Paraxial the segmental plate mesoderm (Revel et al., 1973; Lipton Mesoderm and Jacobson, 1974; Lash et al., 1984; Jacob et al., 1991). It Paraxis expression is upregulated immediately prior to is tempting to speculate that paraxis relays inductive signals the formation of epithelial somites and is maintained in the from the ectoderm and neural tube to the genes encoding epithelial dermomyotome. In regions of surgically manipu- cell adhesion, cytoskeletal, or matrix molecules which are lated embryos in which paraxis transcripts were not ex- required for the epithelialization steps in somitogenesis. pressed, the paraxial mesoderm failed to epithelialize, Ongoing studies are addressing this possibility. which is consistent with the notion that paraxis is required for these cells to assume an epithelial phenotype. In this regard, recent gene knockout experiments in mice indicate ACKNOWLEDGMENTS that paraxis is required for the formation of epithelial somites and the epithelial dermomyotome (Burgess et al., We thank C. M. Fan for reagents and helpful experimental advice. 1996). These results suggest that paraxis mediates the ef- We thank J. Burlefinger and A. Tizenor for the drawings. The work fects of inductive signals on somite formation and pat- was supported by a grant (Ch 44/12-2) to B.C.S. and B.B.-S. from terning (Fig. 9). the Deutsche Forschungsgemeinschaft and by grants from The Na- In paraxis-null mouse embryos, there is no morphological tional Institutes of Health, The Muscular Dystrophy Association, evidence for the dermomyotome, but Pax-3, which is a NIH Program Project Grant AR41940, and The Robert A. Welch marker for the dermomyotome, is expressed in the correct Foundation to E.N.O. region of the presumptive somite (Burgess et al., 1996). Sim- ilarly, the myotome and sclerotome are specified in paraxis mutant embryos. We believe, therefore, that paraxis con- REFERENCES trols epithelialization of the paraxial mesoderm, but it is not involved in specifying the fates of somitic cell lineages. Amthor, H., Connolly, D., Patel, K., Brand-Saberi, B., Wilkinson, Previous studies showed that cells from the paraxial meso- D. G., Cooke, J., and Christ, B. (1996). The expression and regula- derm become competent to respond to notochord signals tion of Follistatin and Follistatin-like gene during avian somite concomitant with epithelialization into somites (Pownall compartmentalization and myogenesis. Dev. Biol. 178, 343–362. et al., 1996) and that epithelial cells from newly formed Barth, J., and Ivarie, R. (1994). Polyvinyl alcohol enhances detection of low abundance transcripts in early stage quail embryos in somites but not from the segmental plate were responsive to a nonradioactive whole mount in situ hybridization technique. notochord induction of myotome differentiation (Buffinger Biotechniques 17, 324–327. and Stockdale, 1994). However, the presence of myogenic Bellairs, R. (1963). The development of the somites in the chick cells in the paraxial mesoderm of paraxis mutant embryos embryo. J. Embryol. Exp. Morphol. 11, 697–714. demonstrates that activation of the myotomal program does Bellairs, R., Curtis, A. S., and Sanders, E. J. (1978). Cell adhesiveness not require an epithelial intermediate cell type. and embryonic differentiation. J. Embryol. Exp. Morphol. 46, 207–213. Blanar, M. A., Crossley, P. H., Peters, K. G., Steingrimsson, E., Inductive Signals Regulating Somite Formation Copeland, N. G., Jenkins, N. A., Martin, G. R., and Rutter, W. J. Previous studies have demonstrated that somites can (1995). Meso1, a basic helix-loop-helix protein involved in mam- malian presomitic mesoderm development. Proc. Natl. Acad. form autonomously from paraxial mesoderm in vitro in the Sci. USA 92, 5870–5874. absence of axial organs (Bellairs, 1963; Christ et al., 1972, Bober, E., Brand-Saberi, B., Ebensperger, C., Wilting, J., Balling, R., 1992; Packard and Jacobson, 1976). These studies led to the Paterson, B. M., Arnold, H. H., and Christ, B. (1994). Initial steps conclusion that the paraxial mesoderm possessed the intrin- of myogenesis in somites are independent of influence from axial sic ability to form somites without a requirement for signals structures. Development 120, 3073–3082. from other cell types. However, in those earlier in vitro Brand-Saberi, B., Ebensperger, C., Wilting, J., Balling, R., and Christ, assays, surface ectoderm was present along with segmental B. (1993). The ventralizing effect of the notochord on somite plate. When segmental plates were cultivated without ecto- differentiation in chick embryos. Anat. Embryol. 188, 239–245. derm, only the most cranial portion which was already epi- Brand-Saberi, B., Wilting, J., Ebensperger, C., and Christ, B. (1996). thelialized formed somites (Lash et al., 1984). Our results The formation of somite compartments in the avian embryo. Int. J. Dev. Biol. 40(1), 411–420. suggest that the ectoderm may have been required for so- Buffinger, N., and Stockdale, F. E. (1994). Myogenic specification in mite formation in these assays. somites: Induction by axial structures. Development 120, 1443– Epithelialization of the paraxial mesoderm during somite 1452. formation is dependent on cell–cell and cell–matrix inter- Buffinger, N., and Stockdale, F. E. (1995). Myogenic specification actions and has been shown to be accompanied by an in- of somites is mediated by diffusible factors. Dev. Biol. 169, 96– crease in cell adhesiveness (Christ et al., 1972; Bellairs et 108.

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba 242 Sˇ osˇic´ et al.

Burgess, R., Cserjesi, P., Ligon, K. L., and Olson, E. N. (1995). Par- entiation behavior in vitro compared to entire somites and seg- axis: A basic helix–loop–helix protein expressed in paraxial mental plate. Dev. Biol. 172, 625 –639. mesoderm and developing somites. Dev. Biol. 168, 296–306. Goulding, M. D., Lumsden, A., and Paquette, A. J. (1994). Regula- Burgess, R., Rawls, A., Brown, D., Bradley, A., and Olson, E. N. tion of Pax-3 expression in the dermomyotome and its role in (1996). Requirement of the paraxis gene for somite formation muscle development. Development 120, 957–971. and musculoskeletal patterning. Nature 384, 570–573. Hamburger, V., and Hamilton, H. L. (1951). A series of normal Chevallier, A., Kieny, M., and Mauger, A. (1977). Limb-somite rela- stages in the development of the chick embryo. J. Morphol. 88, tionship: Origin of the limb musculature. J. Embryol. Exp. Mor- 49–92. phol. 41, 245–258. Jacob, M., Christ, B., Jacob, H., and Poelmann, R. E. (1991). The Christ, B., Brand-Saberi, B., Grim, M., and Wilting, J. (1992). Local role of fibronectin and laminin in development and migration of signalling in dermomyotomal cell type specification. Anat. the avian Wolffian duct with reference to somitogenesis. Anat. Embryol. 186, 505–510. Embryol. 183, 385–395. Christ, B., Brand-Saberi, B., Jacob, H. J., Jacob, M., and Seifert, R. Jacobson, A. G. (1988). Somitomeres: Mesodermal segments of ver- (1990). Principles of early muscle development. In ‘‘The Avian tebrate embryos. Development 104, 209–220. Model in Developmental Biology: From Organism to Genes’’ (N. Johnson, R. L., Laufer, E., Riddle, R. D., and Tabin, C. (1994). Ec- Le Douarin, F. Dieterlen-Lie`vre, and J. Smith, Eds.), pp. 139 –151. topic expression of sonic hedgehog alters dorsal–ventral pat- CNRS, Paris. terning of somites. Cell 79, 1165–1173. Christ, B., Jacob, H. J., and Jacob, M. (1972). Experimentelle Unter- Kaehn, K., Jacob, H. J., Christ, B., Hinrichsen, K., and Poelmann, suchungen zur Somitenentstehung beim Hu¨ hnerembryo. Z. R. E. (1988). The onset of myotome formation in the chick. Anat. Anat. Entwickl-Gesch 138, 82–97. Embryol. 177, 191–201. Christ, B., Jacob, H. J., and Jacob, M. (1974). Die Somitogenese beim Kenny-Mobbs, T., and Thorogood, P. (1987). Autonomy of differen- Hu¨ hnerembryo. Zur Determination der Segmentierungsrich- tiation in avian branchial somites and the influence of adjacent tung. Verh. Anat. Ges. 68, 573–579. tissues. Development 100, 449–462. Christ, B., Jacob, H. J., and Jacob, M. (1977). Experimental analysis Keynes, R. J., and Stern, C. D. (1988). Mechanisms of vertebrate of the origin of the wing musculature in avian embryos. Anat. segmentation. Development 103, 413–429. Embryol. 150, 171–186. Kuratani, S., Martin, J. F., Wawersik, S., Lilly, B., Eichele, G., and Christ, B., Jacob, M., and Jacob, H. J. (1983). On the origin and Olson, E. N. (1994). The expression pattern of the chick homeo- development of the ventro-lateral trunk musculature in the avian box gene gMHox suggests a role in patterning of the limbs and embryo. An experimental and ultrastructural study. Anat. face and in compartmentalization of the somites. Dev. Biol. 161, Embryol. 166, 87 –101. 357–369. Christ, B., and Ordahl, C. P. (1995). Early stages of chick somite Lash, J. W., Seitz, A. W., Cheney, C. M., and Ostrovsky, D. (1984). development. Anat. Embryol. 191, 381–396. On the role of fibronectin during the compaction stage of somito- Cossu, G., Kelly, R., Tajbakhsh, S., Di Donna, S., Vivarelli, E., genesis in the chick embryo. J. Exp. Zool. 232, 197–206. and Buckingham, M. (1996). Activation of different myogenic Liem, K. F., Tremml, G., Roelink, H., and Jessell, T. M. (1995). pathways: myf-5 is induced by the neural tube and MyoD by the Dorsal differentiation of the neural plate cells induced by BMP- dorsal ectoderm in mouse paraxial mesoderm. Development 122, mediated signals from epidermal ectoderm. Cell 82, 969–979. 429–437. Lipton, B. H., and Jacobson, A. G. (1974). Analysis of normal somite Dickinson, M. E., Selleck, M. A., McMahon, A. P., and Bronner- development. Dev. Biol. 38, 73 –90. Fraser, M. (1995). Dorsalization of the neural tube by the non- Lyons, G. E., Buckingham, M. E., Tweedie, S., and Edwards, Y. H. neural ectoderm. Development 121, 2099–106. (1991). Carbonic anhydrase III, an early mesodermal marker, is Duband, J. L., Dufour, S., Hatta, K., Takeichi, M., Edelmann, G. M., expressed in embryonic mouse skeletal muscle and notochord. and Thiery, J. P. (1987). Adhesion molecules during somitogen- Development 111, 233 –244. esis in the avian embryo. J. Cell Biol. 104, 1361–1374. Martin, J. F., Bradley, A., and Olson, E. N. (1995). The paired-like Edmondson, D. G., and Olson, E. N. (1989). A gene with homology homeobox gene MHox is required for early events of skeletogen- to the myc similarity region of MyoD1 is expressed during myo- esis in multiple lineages. Genes Dev. 9, 1237–1249. genesis and is sufficient to activate the muscle differentiation Meier, S. (1979). Development of the chick embryo mesoblast. For- program. Genes Dev. 3, 628–640. mation of the embryonic axis and establishment of the meta- Ebensperger, C., Wilting, J., Brand-Saberi, B., Mizutani, Y., Christ, meric pattern. Dev. Biol. 73, 24–45. B., Balling, R., and Koseki, H. (1995). Pax-1, a regulator of sclero- Mu¨ nsterberg, A. E., and Lassar, A. B. (1995). Combinatorial signals tome development is induced by notochord and floor plate signals from the neural tube, floor plate and notochord induce myogenic in avian embryos. Anat. Embryol. 191, 297 –310. bHLH gene expression in the somite. Development 121, 651– Fan, C. M., and Tessier-Lavigne, M. (1994). Patterning of mamma- 660. lian somites by surface ectoderm and notochord: Evidence for Mu¨ nsterberg, A. E., Kitajewski, J., Bumcrot, D. A., McMahon, A. P., sclerotome induction by a hedgehog homolog. Cell 79, 1175– and Lassar, A. B. (1995). Combinatorial signaling by sonic hedge- 1186. hog and Wnt family members induces myogenic bHLH gene ex- Fan, C. M., Porter, J. A., Chiang, C., Chang, D. T., Beachy, P. A., pression in the somite. Genes Dev. 9, 2911–2922. and Tessier-Lavigne, M. (1995). Long-range sclerotome induction Ordahl, C. P., and Le Douarin, N. M. (1992). Two myogenic lineages by sonic hedgehog: Direct role of the amino-terminal cleavage within the developing somite. Development 114, 339–353. product and modulation by the cyclic AMP signalling pathway. Packard, D. S., Jr., and Jacobson, A. G. (1976). The influence of axial Cell 81, 457–465. structures on chick somite formation. Dev. Biol. 53, 36–48. Gamel, A. J., Brand-Saberi, B., and Christ, B. (1995). Halves of epi- Parr, B. A., Shea, M. J., Vassileva, G., and McMahon, A. P. (1993). thelial somites and segmental plate show distinct muscle differ- Mouse Wnt genes exhibit discrete domains of expression in the

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba Regulation of paraxis Expression and Somite Formation 243

early embryonic CNS and limb buds. Development 119, 247– Rong, P. M., Teillet, M. A., and Ziller, C. (1990). Differential inﬂu- 261. ence of the neural primordium on muscle development in myo- Pourquie, O., Coltey, M., Teillet, M.-A., Ordahl, C., and Le Dou- tome and limb. In ‘‘The Avian Model in Developmental Biology: arin, N. M. (1993). Control of dorsoventral patterning of somite From Organism to Genes,’’ pp. 153 –164. CNRS. derivatives by notochord and ﬂoor plate. Proc. Natl. Acad. Sci. Rong, P. M., Teillet, M. A., Ziller, C., and Le Douarin, N. M. (1992). USA 90, 5242–5246. The neural tube/notochord complex is necessary for vertebral Pourquie, O., Coltey, M., Breant, C., and Le Douarin, N. M. (1995). but not limb and body wall striated muscle differentiation. De- Control of somite patterning by signals from the lateral plate. velopment 115, 657–672. Proc. Natl. Acad. Sci. USA 92, 3219–3223. Schemainda, H. (1979). U¨ ber die Entwicklung der occipitalen So- Pourquie, O., Fan, C. M., Coltey, M., Hirsinger, E., Watanabe, Y., miten. Experimentelle Untersuchungen an Wachtel- und Hu¨ hner- Breant, C., Francis-West, P., Brickell, P., Tessier-Lavigne, M., and embryonen. Verh. Anat. Ges. 73, 527 –532. Le Douarin, N. M. (1996). Lateral and axial signals involved in Schoenwolf, G. C., and Watterson, R. L. (1989). ‘‘Laboratory Studies avian somite patterning: A role for BMP-4. Cell 84, 461–471. of Chick, Pig and Frog Embryos.’’ Macmillan, New York. Pownall, M. E., Strunk, K. E., and Emerson, C. P., Jr. (1996). Noto- Selleck, M. A., and Stern, C. D. (1991). Fate mapping and cell lin- chord signals control the transcriptional cascade of myogenic eage analysis of Hensen’s node in the chick embryo. Develop- bHLH genes in somites of quail embryos. Development 122, ment 112, 615 –626. 1475–1488. Spense, M. S., Yip, J., and Erickson, C. A. (1996). The dorsal neural Quertermous, E. E., Hidai, H., Blanar, M. A., and Quertermous, T. tube organizes the dermamyotome and induces axial myocytes (1994). Cloning and characterization of a basic helix–loop–helix in the avian embryo. Development 122, 231–241. protein expressed in early mesoderm and the developing somites. Stern, H. M., and Hauschka, S. D. (1995). Neural tube and noto- Proc. Natl. Acad. Sci. USA 91, 7066–7070. chord promote in vitro myogenesis in single somite explants. Revel, J. P., Yip, P., and Chang, L. L. (1973). Cell junctions in the Dev. Biol. 167, 87–103. early chick embryo. Dev. Biol. 35, 302–317. van Bemmelen, J. F. (1889). U¨ ber die Herkunft der Extremita¨ten- Roelink, H., and Nusse, R. (1991). Expression of two members of und Zungenmuskulatur bei Eidechsen. Anat. Anz. 4, 240–255. the Wnt family during mouse development restricted temporal and spatial patterns in the developing neural tube. Genes Dev. Received for publication October 21, 1996 5, 381–388. Accepted February 28, 1997

AID DB 8561 / 6x23$$$181 05-07-97 11:44:50 dba