cMeso-1, a Novel bHLH Transcription Factor, Is Involved in Somite Formation in Chicken Embryos

DEVELOPMENTAL BIOLOGY 199, 201–215 (1998) ARTICLE NO. DB988919

cMeso-1, a Novel bHLH Transcription Factor, Is Involved in Somite Formation in Chicken Embryos Astrid Buchberger,1 Klaus Seidl, Christian Klein, Heike Eberhardt, and Hans-Henning Arnold Department of Cell and Molecular Biology, Technical University of Braunschweig, Spielmann Strasse 7, 38106 Braunschweig, Germany

The segmentation of somites from the paraxial mesoderm is a crucial event in vertebrate embryonic development; however, the mechanisms underlying this process are not well understood. In a yeast two-hybrid screen we have identified the novel basic-helix-loop-helix (bHLH) protein cMeso-1 which is expressed in the presomitic mesoderm of early chicken embryos. Initially the gene is activated in the epiblast and transcripts concentrate later in and around the primitive streak. When the segmental plate is laid down the cMeso-1 expression domain successively retracts toward the caudal end but a second domain appears in bilateral stripes in the anterior paraxial mesoderm. This highly dynamic domain of cMeso-1 transcripts demarcates the area immediately posterior to the next prospective pair of somites in cyclic waves which apparently correspond to the formation of new somites. Loss of cMeso-1 function by antisense RNA or oligonucleotides results in severe attenuation of somitogenesis suggesting that it plays an important role in setting up the segmentation process. The dynamic and periodically reiterated expression of cMeso-1 along the anteroposterior axis is not dependent on anterior structures or the propagation of a signal along the anteroposterior axis but seems to follow an intrinsic patterning program which is already set up in the segmental plate. © 1998 Academic Press Key Words: bHLH transcription factor; mesoderm development; somitogenesis.

INTRODUCTION One of the crucial events during vertebrate development is the segmentation of paraxial 1mesoderm which gives rise to somites, the first observable metameric units in the embryo. Paraxial mesoderm develops on both sides of the notochord and neural tube in a rostrocaudal direction when Hensen’s node and the primitive streak retract caudally (Slack, 1991). Somitogenesis starts in the cranial region of the embryo and is accompanied by considerable morphogenetic changes of the tissue architecture within the mesoderm. The initially mesenchymal cells of the paraxial mesoderm undergo compaction and transition from mesenchyme to a spherical epithelium which surrounds the somitocoel that is filled with mesenchymal cells. These epithelial somites are subsequently subdivided into the ventral mesenchymal sclerotome and the dorsal epithelial dermomyotome. The sclerotome will develop into the axial skeleton, while the dermomyotome further subdivides into 1 To whom correspondence should be addressed. Fax: 495313918178. E-mail: [email protected].

0012-1606/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

the myotome and dermatome giving rise to epaxial musculature and connective tissue of the dermis, respectively. The broad contribution of somitic cells to multiple tissues implies that mutations in genes which are expressed in somites and are functionally involved in the formation of the skeleton and musculature will have severe effects on body morphology (Tam and Trainor, 1994). Genes and molecular mechanisms underlying cell type specification of somitic derivatives, such as the sclerotome and dermomyotome (Bober et al., 1994; Fan and Tessier-Lavigne, 1994; Fan et al., 1995; Olson and Klein, 1994; Burgess et al., 1995; Cserjesi et al., 1995; Pourquie et al., 1996; Winnier et al., 1997), have been studied in great detail and their actions are particularly well understood for the development of skeletal muscles (reviewed by Arnold and Braun, 1996). In contrast, molecular information on the process of somitogenesis itself and the maintenance of segment borders is rather limited (Yamaguchi, 1997). Although the presomitic mesoderm appears unsegmented under the light microscope, several lines of evidence suggest that the paraxial mesoderm is prepatterned in so-called somitomeres (Packard and Jacobson, 1976; Pack-

201

202

Buchberger et al.

FIG. 1A. Nucleotide and deduced amino acid sequence of cMeso-1 cDNA. The bHLH region is marked by a gray box and the polyadenylation signal is underlined.

ard, 1978; Meier, 1979; Jacobson and Meier, 1986). Thus, the segmentation program may be established in mesoderm prior to overt morphological appearance of segment borders. It has been demonstrated recently that the vertebrate members of the Notch/Delta gene families may play a role in the segmentation process. Delta and Notch proteins interact with each other and are thought to be required for cell fate decisions (Corbin et al., 1991) as well as epithelial– mesenchymal transitions (Hartenstein et al., 1992; Tepass and Hartenstein, 1995). The mouse genes Notch-1 and Notch-2 are expressed in the paraxial mesoderm (Del Amo et al., 1992; Lardelli and Lendahl, 1993; Reaume et al., 1992) and mice mutants lacking Notch-1 exert disorganized and delayed somitogenesis (Conlon et al., 1995). The mouse Delta-like gene Dll1 is expressed in the presomitic mesoderm and in the posterior half of somites (Bettenhausen et al., 1995). Dll1 mutant embryos loose the craniocaudal polarity of somites and the ability for epithelialization, although the primary metameric pattern of the mesoderm seems to be maintained (Hrabe de Angelis et al., 1997). Paraxis, a basic helix-loop-helix protein, is also expressed in paraxial mesoderm and somites (Burgess et al., 1995). Mice carrying a homozygous null mutation of the paraxis gene are unable to form epithelial somites and show disruption of correct patterning of tissues derived from the paraxial mesoderm (Burgess et al., 1996). In an attempt to identify additional genes which may play a role in mesoderm segmentation we isolated a novel chicken basic helix-loop-helix (bHLH) protein, referred to as cMeso-1. This gene is first expressed in the epiblast of Hamburger–Hamilton (HH) stage 5 chicken embryos and subsequently concentrates in and around the primitive streak. During further development cMeso-1 expression retracts caudally and an additional dynamic expression domain arises as a bilateral stripe in the anterior presomitic mesoderm. This cMeso-1 domain seems to oscillate with the formation of new somites and always precedes the actual segmentation process. Application of antisense RNA or antisense oligonucleotides to chicken embryos severely

FIG. 1B. Comparison of the cMeso-1 bHLH domain with that of other bHLH proteins. Amino acid identities are indicated by dashes. The compared sequences are MesP2 by Saga et al. (1997); MesP1 by Saga et al. (1996); paraxis by Barnes et al. (1997); M-twist by Blanar et al. (1995); and dHand by Srivastava et al. (1995).

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

203

cMeso1 in Somitogenesis

FIG. 2. Northern blot analysis of cMeso-1 mRNA in embryonic chicken tissues. Twenty-five micrograms of total RNA from indicated tissues was electrophoresed on gel, transferred to a membrane, and hybridized with a radiolabeled cMeso-1-specific probe (EcoRI/BamHI cDNA fragment). Transcripts of approximately 1.2 kb were detected exclusively in the presomitic mesoderm of stage 12 embryos (top). For the RNA loading control the blot was hybridized with a chicken GAPDH probe (bottom).

attenuates somitogenesis suggesting that cMeso-1 plays an important role in the segmentation process.

MATERIALS AND METHODS Yeast two-hybrid screening and isolation of cMeso-1 cDNA. cDNA of HH stage 8 chicken embryos was cloned into the HybriZAP vector (Stratagene, Heidelberg, Germany) to generate a primary lambda library. The library was amplified and converted by in vivo excision to a pAD-GAL4 plasmid library for expression in yeast. The bait vector Gal4-BD-E12 was obtained by ligation of the PCR-generated EcoRI/SalI fragment encoding the bHLH dimerization domain of E12 (aa 508 to 654) into the vector which already contained the GAL4 DNA-binding domain. Yeast strain YRG2 was

transformed with the Gal4-BD-E12 vector and grown under Leu selection. Resulting yeast transformants were transformed with 100 mg of the embryonic chicken library and incubated for 4 to 7 days under Leu/Trp/His selection. Growing colonies were transferred to filters (Whatman No. 1, Grade 1), permeabilized in liquid nitrogen, and assayed for expression of b-galactosidase using X-gal substrate. Positive colonies appeared within 0.5 to 12 h. All procedures of the yeast two-hybrid screening were performed as recommended by the manufacturer of the cloning kit. Plasmid isolation from b-galactosidase-positive yeast colonies was performed according to the method described by Hoffman and Winston (1987). The DNA sequence of isolated plasmids was determined by the dideoxynucleotide method on both strands with the appropriate oligonucleotide primers. cDNA inserts were released from the pADGAL4 plasmid with EcoRI and XhoI. To obtain full-length cMeso-1 cDNA the chicken stage 8 cDNA lambda library was rescreened using the EcoRI/XhoI fragment obtained in the yeast screen as probe. Yeast strain YRG-2 has the following genotype: Mata ura3-52 his3-200 ade2-101 lys2-801 trp1-901 leu2-3 112 gal4-542 gal80-538 LYS::UAS GAL1 -TATA GAL1 -HIS3 URA3::UAS GAL4 17MERS(X3) TATACYC1-LACZ. Northern blot analysis. Total RNA from chicken embryonic tissues was isolated by the guanidinium thiocyanate method (Auffray and Rougeon, 1980). The RNA was glyoxylated, separated on a 1% agarose gel, and transferred to a Nylon membrane (PALL Biodyne A). Hybridization was carried out in 50% formamide, 53 Denhardt’s, 0.1% SDS, 50 mM sodium phosphate, 53 SSPE, and 50 mg/ml denaturated salmon sperm DNA at 42°C. A 798-base-pair EcoRI/BamHI fragment was 32P labeled by random priming to a specific activity of 1 3 108 dpm/mg. A GAPDH probe served as RNA loading control. Whole-mount in situ hybridizations. Antisense and sense RNA probes for cMeso-1 and cFKH-1 were labeled with digoxygenin-11-UTP using T7 polymerase and the pAD-GAL4-cMeso-1 plasmid linearized with EcoRI. Double in situ hybridizations were performed with dll and paraxis probes labeled with fluorescein and detected with anti-fluorescein antibody using Fast Red as substrate. Chick embryos of various stages were fixed with 4% paraformaldehyde/PBS at 4°C overnight. Whole-mount in situ hybridiza-

FIG. 3. Whole-mount in situ hybridizations of chicken embryos between HH stages 5 and 8 with a cMeso-1 probe. Dorsal views of stages 5 (A), 6 (B), 7 (C), and 8 (D) embryos are shown. First transcripts are detected in and around the primitive streak. Initially faint staining is also observed in bilateral stripes at the level of Hensen’s node (A) which later increases in intensity in the anterior presomitic mesoderm (B–D). The posterior domain regresses caudally in parallel to the retracting primitive streak (B–D). Accumulation of cMeso-1 transcripts in the anterior domain precedes somite formation and becomes undetectable in epithelial somites. Transverse vibratome sections (30 mm) through the posterior expression domain of stage 5 (E and F) and 6 (H) embryos indicate that cMeso-1 transcripts first accumulate in the epiblast (E) and subsequently in the mesoderm (F–H). A section through the anterior domain shows the cMeso-1 transcript in the mesoderm (G). The level of sections in indicated. hn, Hensen’s node; ps, primitive streak; hf, head fold; pso, prospective somite; m, mesoderm; en, endoderm; e, ectoderm; ep, epiblast; pf, primitive fold. FIG. 4. cMeso-1 expression in stage 9 to 15 embryos shown by whole-mount in situ hybridization. Dorsal views of embryos indicate that cMeso-1 transcripts retract to the caudal end of the primitive streak, while bilateral stripes of cMeso-1 positive cells appear in the anterior paraxial mesoderm. These stripes of cMeso-1 expression move in an anteroposterior direction in front of the newly forming somites and almost reach the tailbud in a stage 15 embryo (A–D). Higher power magnification of C is shown in D to demonstrate the position of cMeso-1 positive cells in yet an unsegmented mesoderm posterior to the region which will form the next somites. A shows an embryo in which one stripe has shifted toward the anterior end of the presomitic mesoderm, while more caudally the next stripe has already activated cMeso-1 expression. Vibratome sections at the levels indicated in B and C confirm that transcripts are limited to the presomitic mesoderm with higher expression in cells of the periphery which are destined to become the epithelium (E and F). nc, notochord; nt, neural tube; pm, presomitic mesoderm; pso, prospective somite; ps, primitive streak.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

204

Buchberger et al.

cMeso1 in Somitogenesis

205

206

Buchberger et al.

tions and vibratome sections were performed as described previously (Buchberger et al., 1996). Embryo culture and treatment with antisense oligonucleotides or RNA. Fertilized eggs from white Leghorn chickens were incubated at 38.5°C for about 45 h when embryos were removed from the eggs, carefully cleaned from yolk in Tyrode’s solution, and staged by counting somites as described by Hamburger and Hamilton (1951). Embryos with 7 to 14 somites were used for treatment with antisense nucleic acids. For cultivation in vitro embryos were placed on agar dishes (0.6% Bacto agar, 0.7% NaCl, 50% egg albumin) with the dorsal side up and incubated at 38.5°C. Antisense oligonucleotides were synthesized as phosphorothioate derivates, purified by HPLC, and dissolved in sterile PBS to a final concentration of 80 mM. Two microliters of sense or nonsense oligonucleotides was added directly to the embryo at the start of the experiment and every 2 h thereafter. The treatment was terminated after 6 h by fixation of embryos with 4% paraformaldehyde in PBS overnight prior to whole-mount in situ hybridization. A control group of animals was treated with PBS alone. Numbers of somites were counted at the start and the end of the experiments. Normal development of head and heart was monitored to exclude general toxicity of the oligonucleotides and adverse effects of in vitro cultivation. Treatment with antisense RNA in ovo was performed on embryos which were contrasted by injection of India-Ink solution (1:5 dilution in PBS). In vitro transcribed antisense or sense RNA (5 mg) was incubated for 15 min with 300 ml Superfect transfection reagent (Qiagen, Hilden, Germany) and 3–5 ml was injected through the vitelline membrane directly onto the embryos. A control group of embryos received Superfect transfection reagent without RNA. The eggs were sealed with polythene foil and incubated for a further 6 h. Pairs of somites and normal development were scored as described above. The following oligonucleotides were used: cMeso-1 antisense oligo, 59-GTGGCAGGAACCCCGTGCAAGCTCC-39; cMeso-1 sense oligo, 59-GGAGCTGCACGGGGTTCCTGCCAC-39; and cMeso-1 nonsense oligo, 59-GTCCGGACCGGTAGCCTCGTACGG-39. All selected oligonucleotides did not show significant homology with other gene sequences contained in GenBank databases. For statistical evaluation all data were subjected to Student’s t test and expressed as means 6 SEM. Surgical manipulation of embryos. Stage 9 to 12 chicken embryos were removed from the eggs, cleaned from yolk in PBS, and transversely dissected with tungsten needles at various axial levels as illustrated in Fig. 7. Anterior and posterior parts of the embryos were cultivated separately on agar dishes (0.6% Bacto agar, 0.7% NaCl, 50% egg albumin) for 8 to 20 h during which somitogenesis was monitored. The explants were fixed in 4% paraformaldehyde/PBS for in situ hybridization with cMeso-1, cFKH-1, or paraxis probes.

RESULTS Isolation of the chicken cDNA cMeso-1 encoding a novel bHLH transcription factor. To identify novel bHLH proteins expressed during early chick development we employed the yeast two-hybrid system using the conserved bHLH dimerization domain of the ubiquitously expressed transcription factor E12 as bait. A cDNA library of HH stage 8 chicken embryos was screened for E12 dimerization partners. From a total of 1 3 106 independent transformants

we obtained 23 clones which grew under Leu/Trp/His selection and also expressed b-galactosidase. These clones were sequenced and the results subjected to a Blast search of the GenBank database. Among several known bHLH proteins, such as paraxis, eHand, and dHand, one partial cDNA clone of 0.9 kb encoded a novel bHLH domain. To isolate the full-length cDNA the HH stage 8 embryonic chick library was rescreened. The longest clone contained an insert of 1082 nucleotides with an open reading frame of 828 nucleotides encoding 287 amino acids (Fig. 1A). This clone was designated cMeso-1 for its mesodermal expression (see below). A typical polyadenylation signal (AATAAA) was located 13 nucleotides upstream of the poly(A) stretch. The first methionine codon was preceded by the Kozak consensus sequence (Kozak, 1997) suggesting that it represented the translation initiation site. Comparison of the cMeso-1 bHLH domain with those of known proteins revealed 76.4% and 74.5% identity with the recently isolated mouse genes MesP2 and MesP1 (Saga et al., 1996, 1997) and homology to paraxis (Barnes et al., 1997), M-twist (Blanar et al., 1995), and dHand (Srivastava et al., 1995) of 49, 47, and 45%, respectively (Fig. 1B). Essentially no homology to these or any other proteins was found outside of the bHLH domain at the nucleotide or protein level. Based on this limited sequence similarity it seems questionable whether cMeso-1 represents the chicken orthologue of mouse MesP1 or MesP2, as bHLH domains of true orthologues of chicken and mouse bHLH proteins exhibit homologies between 94 and 100% throughout the entire coding region as recently exemplified for paraxis (Sosic et al., 1997). Thus, cMeso-1 may be a new member of the MesP-related family of transcription factors. cMeso-1 transcripts accumulate in the presomitic mesoderm and seem to demarcate the area of the next to form somites during early chick development. First evidence that embryonic cMeso-1 expression may be restricted to paraxial mesoderm came from Northern blot analysis of embryonic chicken tissues which revealed cMeso-1 transcripts of about 1.2 kb in the presomitic mesoderm but in no other tissue (Fig. 2). To determine the spatiotemporal expression pattern in more detail whole-mount in situ hybridizations were performed on chicken embryos with a digoxigenin-labeled cMeso-1 antisense RNA probe. First transcripts were detected in cells surrounding the primitive streak of stage 5 embryos (Fig. 3). This expression domain was limited to the medial and posterior portion of the embryo and appeared much wider at its anterior margin. During subsequent development the cMeso-1 expression domain became more concentrated to the medial mesoderm and gradually retracted toward the caudal end in parallel to the regression of the primitive streak (Figs. 3A–3D). A second, initially weak cMeso-1 expression domain became visible in two bilateral stripes at the anterior end of the arising paraxial mesoderm approximately at the level of Hensen’s node (Fig. 3). With ongoing somitogenesis this domain increased in intensity and moved in a rostrocaudal direction remaining at an approximately equal dis-

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

207

cMeso1 in Somitogenesis

tance posterior to the newly formed somites (Figs. 3B–3D). Transversal vibratome sections through the anterior and posterior domain in early stage embryos demonstrated that cMeso-1 transcripts were initially present in the epiblast and were later limited to the arising mesoderm (Figs. 3E–3H). The highly dynamic mode of cMeso-1 expression in the paraxial mesoderm became even more apparent when older embryos were analyzed. At stages 9 to 15 the posterior expression domain approached the tailbud, while the anterior stripes continued to reappear in consistent waves throughout the somite formation period (Figs. 4A– 4D). No cMeso-1 expression was found so far in later tissues. Interestingly, in most embryos analyzed the anterior cMeso-1 domain appeared as one pair of bilateral stripes caudal to the last formed somites leaving a gap of approximately one somite diameter (Figs. 4B– 4D). In a fraction of embryos two cMeso-1 positive stripes were observed, one immediately next to the last formed somites and one at a distance of approximately one prospective somite (Fig. 4A). Embryos without any cMeso-1 positive stripes were never observed. To better define the precise location of cMeso-1 transcripts within the presomitic mesoderm double in situ hybridizations were performed with the cMeso-1 probe together with either a dll (Delta-like)-specific or paraxis-specific probe (Fig. 5). Both of these genes are expressed in distinct patterns in the paraxial mesoderm and in somites (Bettenhausen et al., 1995; Burgess et al., 1995). Whole-mount hybridizations and transversal vibratome sections confirmed that cMeso-1 transcripts accumulate in register with prospective somites, initially posterior to the next somite forming region whose anterior border is defined by dll expression (Fig. 5A). Thus, dll expression overlaps with cMeso-1 but apparently precedes it. Similarly, cMeso-1 expression also overlaps with paraxis expression but the paraxis domain extends more caudally, suggesting that it also precedes cMeso-1 activation (Figs. 5C and 5D). Significantly, however, the anterior and posterior borders of cMeso-1 positive stripes in relation to dll and paraxis expression, respectively, were somewhat variable indicating the dynamic nature of the cMeso-1 domain during the somite formation cycle (Figs. 5A–5D). Moreover, in some embryos two cMeso-1 stripes were observed with the anterior stripe appearing less then one somite in length (Figs. 5E and 5F). This observation suggests that cMeso-1 transcripts disappear from the posterior of the next-to-form somite prior to the anterior portion, while the next caudal expression domain has already begun to accumulate cMeso-1 transcripts. It should be noted that we never observed cMeso-1 expression within already segmented somites. Transverse sections through the overlapping expression domains revealed that cMeso-1 mRNA primarily accumulates at the dorsal and ventral sites of the presomitic mesoderm unlike the transcripts for dll and paraxis which appear more evenly distributed throughout the medial mesoderm. In summary, cMeso-1 transcription appears to be initiated in the paraxial mesoderm that is in preparation

of somite formation but the gene is active only until somites actually begin to form. cMeso-1 antisense RNA or antisense oligonucleotides attenuate somitogenesis in chicken embryos. As an attempt to explore the possible importance of cMeso-1 for somitogenesis we took a loss of function approach by applying antisense RNA or antisense oligonucleotides to chicken embryos. In a series of experiments antisense RNA dissolved in SuperFect reagent was injected through the vitelline membrane directly on stage 10 embryos in ovo and incubation was continued for an additional 6 h. The number of somites was counted at the time of RNA injection and 6 h later. Under this treatment we observed no abnormal development of head and heart structures suggesting that the reagents were not generally toxic. Control animals injected with SuperFect reagent alone or with sense RNA developed on average 3.4 (60.22) somites within 6 h (Fig. 6) similar to embryos that were not treated at all (data not shown). In contrast, chicken embryos injected with cMeso-1 antisense RNA formed only one additional pair of somites during the same period. This significant difference strongly suggested that cMeso-1 is necessary for the formation of somites. To avoid unspecific effects which may possibly be exerted by antisense RNA a second series of experiments was performed using antisense oligonucleotides on chicken embryos which were cultivated on agar dishes. Oligonucleotides dissolved in PBS were applied to the embryos every 2 h for a total of 6 h. The number of newly formed somites was scored as described above. We also checked for normal development of head and heart. As shown in Fig. 6B, approximately one pair of somites developed every 2 h in the control groups that had received either PBS alone or PBS plus randomized nonsense oligonucleotides. The numbers of 3.5 6 0.34 and 3.00 6 0.25 newly formed somites in embryos treated with PBS or nonsense oligonucleotides were in good agreement with those observed in the in ovo experiments. Application of cMeso-1 antisense oligonucleotides to the embryos resulted in significant attenuation of somitogenesis with only one additional pair of somites formed in 4 h and almost complete arrest of somitogenesis thereafter (Fig. 6B). Identical results were obtained with oligonucleotide injections in ovo (data not shown). In addition to the suppression of somitogenesis, both antisense RNA and antisense oligonucleotides also led to posterior truncations of the body axis in the majority of embryos (Fig. 6H, compare embryos treated with antisense and nonsense oligonucleotides). To control the effect of antisense oligonucleotides on the level of cMeso-1 transcripts whole-mount in situ hybridizations were performed on embryos fixed 1, 3, or 6 h following treatment. Severe reduction of cMeso-1 staining was already observed 1 (data not shown) and 3 h after the addition of antisense oligonucleotides (Fig. 6D) with essentially no transcripts remaining after 6 h (Fig. 6E). Randomized nonsense oligonucleotides had no effect on the level of cMeso-1 transcripts within 6 h (Fig. 6C). The specificity of oligonucleotide treatment was further controlled by hybridization with the cFKH1 probe

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

FIG. 5. Double in situ hybridizations of stage 11 to 15 chick embryos with cMeso-1 probe (blue) and either dll (delta-like) probe (red in A, B, and E) or paraxis probe (red in C, D, and F). Dll staining in the presomitic mesoderm defines the anterior border of the segmental plate and illustrates the dynamic position of cMeso-1 expression relative to this border in different embryos (A, B, and E). Arrows indicate the positions of the last formed somites; arrowheads mark the anterior end of the unsegmented paraxial mesoderm. The relationship of paraxis and cMeso-1 expression domains is shown in C, D, and F. Transverse sections were taken through the region of overlapping expression on a vibratome.

208 Buchberger et al.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

209

cMeso1 in Somitogenesis

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

210

Buchberger et al.

which detects a transcript of the forkhead 1 gene in the presomitic mesoderm and somites (Buchberger et al., 1998). As shown in Fig. 6F, cFKH1 transcript accumulation was unaffected by the 6-h treatment with antisense oligonucleotides. The same treatment also did not affect the normal accumulation of paraxis transcripts in the presomitic mesoderm and in somites (Fig. 6G). Taken together these results provide evidence that antisense oligonucleotides specifically inhibit the expression of cMeso-1 in the embryo. The lack of cMeso-1 then leads to impairment of somite formation. Whether under these conditions somitogenesis was completely blocked or only severely delayed cannot be decided in our experimental setup, as the observation of developmental defects was not extended considerably beyond 6 h to avoid possible general retardation. cMeso-1 gene activation appears independent of anterior structures. Somitogenesis occurs in isolated segmental plates which have been removed from their natural environment. Somites also form in the discontinuous paraxial mesoderm that has been disrupted by microsurgery (Sandor and Amels, 1970; Packard and Jacobson, 1976; Bellairs and Veini, 1980; Sandor and Fazakas-Todea, 1980). These observations have been interpreted as an indication for an intrinsic program which directs somite formation within the segmental plate. To investigate whether cMeso-1 expression follows a similar program anterior and posterior pieces of embryos generated by transversal sections at different axial levels were cultivated separately on agar dishes up to 24 h. Somitogenesis in explants was monitored and cMeso-1 expression was determined by whole-mount in situ hybridization. As outlined schematically in Figs. 7A and 7B, operations were performed to obtain unsegmented mesoderm still attached to already formed somites or mesoderm free of all somites but including the cMeso-1 domain. We also dissected the paraxial mesoderm far caudal to the cMeso-1 domain (Fig. 7C) and asymmetrically to retain somites in the posterior piece on one side and the unsegmented mesoderm on the other side (Fig. 7D) which essentially combines the situations depicted in Figs. 7A and 7C in one embryo. Regardless at which level of the A–P axis the mesoderm was dissected somites always continued to form preceded by the expression of cMeso-1 in the expected area of the presomitic

mesoderm (Fig. 7). These data indicate that neither preformed somites nor the region of the paraxial mesoderm which transiently expressed cMeso-1 is necessary to instruct the posterior segmental plate to perpetuate somite formation and to activate the cMeso-1 gene. Interestingly, in the asymmetrically cut embryos somite formation and cMeso-1 expression first proceeded unilaterally in the more anterior mesoderm on the left side and resumed bilaterally in register after the appropriate time period (Fig. 7D). This suggests that an intrinsic timing program in the paraxial mesoderm may direct cMeso-1 expression and somite formation in a coordinate fashion on both sides of the neural tube along the anterioposterior axis. Thus, cMeso-1 expression like somitogenesis does not seem to depend on any kind of A–P signaling but rather follows a preestablished program to ensure the correct spatiotemporal and segmentspecific pattern.

DISCUSSION We have isolated a chicken cDNA which encodes a novel bHLH transcription factor, referred to as cMeso-1. The gene is initially expressed in the epiblast and in the gastrulating mesoderm. During somitogenesis a prominent expression domain arises in the anterior portion of the segmental plate demarcating a limited region caudal to the next pair of prospective somites. This expression pattern is dynamic and appears to be established in time and space by a program that is intrinsic to the segmental plate. Loss of cMeso-1 function by an antisense approach results in severe impairment of somite formation. cMeso-1 is a protein of 287 amino acids related to the recently isolated mouse MesP1 and MesP2 genes (Saga et al., 1996, 1997). Although MesP2 has a similar expression pattern and also seems to be involved in somitogenesis, it probably is not the orthologue of cMeso-1. This is inferred from the limited sequence similarity within the bHLH domain only, while true chicken and mouse orthologues usually are extensively homologues throughout the coding region (Barnes et al., 1997). In addition, the expression patterns of cMeso-1 in chicken and MesP2 in mouse appear similar but not identical, because the mouse gene appar-

FIG. 6. cMeso-1 antisense RNA or oligonucleotides attenuate somitogenesis in chicken embryos. A shows numbers of somites formed within 6 h after the application of RNA dissolved in polycationic transfection reagent directly onto stage 10 embryos in ovo (for details see Materials and Methods). The open bar represents injections of SuperFect alone, dark gray columns stand for sense RNA, and light gray columns indicate antisense RNA. Figures within the bars give the number of tested embryos. B shows a time course of somite formation in explanted embryos treated with PBS alone (open bars), randomized nonsense oligonucleotides (dark gray bars), and antisense oligonucleotides (light gray bars) 2, 4, and 6 h after the start of the experiment. The accumulation of cMeso-1 transcripts after 6 h of treatment with randomized nonsense oligonucleotides is shown in C and compared to embryos treated for 3 (D) and 6 h (E) with antisense oligonucleotides. The accumulation of unrelated cFKH-1 (F) and paraxis (G) transcripts in the paraxial mesoderm and somites is unaffected by a 6-h treatment with antisense oligonucleotides. H illustrates the caudal truncation in an embryo treated with antisense oligonucleotide (left) compared to an embryo treated with nonsense oligonucleotide (right). All embryos shown are explant cultures. The significance of reduction of newly formed somites under treatment was at a P value ,0.01.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

211

cMeso1 in Somitogenesis

ently lacks the early expression in the epiblast and in the posterior mesoderm (Saga et al., 1997). First cMeso-1 positive mesodermal cells migrate toward the streak and consequently accumulate medially in the paraxial mesoderm. Fates and migratory routes of primitive streak cells has been mapped extensively (Rosenquist, 1966; Schoenwolf et al., 1992; Schoenwolf and Sheard, 1990; Garcia-Martinez and Schoenwolf, 1993). According to fate maps performed by Psychoyos and Sterm (1996) cells of the anterior primitive streak give rise to the somitic mesoderm. The regions of primitive streak which express cMeso-1 may contribute to the presomitic mesoderm and possibly to the lateral plate mesoderm, as c-Meso-1 initially is also expressed in the posterior part of the embryo. In agreement with a contribution of cMeso-1 positive cells to the paraxial mesoderm, we find cMeso-1 expression shifting from an initially more posterior position toward anterior at the late primitive streak stage when the primitive streak regresses. Interestingly, we frequently observed caudal truncations in embryos treated with antisense cMeso-1 RNA or antisense oligonucleotides. This may reflect the contribution of early cMeso-1 positive cells to the growing paraxial mesoderm. The disturbance of longitudinal growth and the formation of paraxial mesoderm in the presence of antisense oligonucleotides may well contribute to the problem of somitogenesis. However, because posterior truncations occurred to variable degrees and not with complete penetrance in antisense oligonucleotide treated embryos, while the somite segmentation defects were always observed with amazingly little variation, we believe that truncation of the body axis is a separate event and not the sole reason for the defect in somite formation. In this view, the functions of cMeso-1 expressed early in the posterior regressing domain and later in the anterior presomitic mesoderm would be distinct. However, we cannot entirely exclude overlapping cMeso-1 functions in both domains, because our antisense approach does not allow to separate cleanly both aspects. The second expression domain of cMeso-1 is located in the anterior presomitic mesoderm initially in a stripe immediately posterior to the next prospective somites. Transcripts in this stripe seem to persist until the somites actually start to pinch off the mesenchyme. Thus, depending on the phase of the somite formation cycle in which the

embryo has been fixed for analysis, one finds one or two stripes at the anterior end of the presomitic mesoderm. This cMeso-1 expression appears functionally associated with somite formation, because antisense oligonucleotides attenuate or even prevent somitogenesis. Somite formation is accompanied by cell compaction (Primmett et al., 1989), changes of the cytoskeletal organization and extracellular matrix composition (Meier, 1979; Ostrovski et al., 1988), and transformation of mesenchyme to epithelium (Ostrovski et al., 1988). It has been shown that increased cadherin and NCAM expression correlates with the increase in cell adhesiveness during maturation of the segmental plate (Bellairs et al., 1978; Duband et al., 1987). Therefore, it is conceivable that cell adhesion molecules may represent downstream target genes for cMeso-1. Fibroblast growth factor receptor 1 (FGFR-1) is upregulated in the presomitic mesoderm during maturation of somitomeres and later becomes restricted to the rostral parts of newly formed somites (Orr-Urtreger et al., 1991; Yamaguchi et al., 1992). FGFs are known to play a role in morphogenetic cell movements during gastrulation (Smith and Howard, 1992), specification of anteroposterior body axis (Slack and Tannahill, 1992) as well as cell proliferation and tissue differentiation. They may also be involved in regulating cMeso-1 gene activity or be subject to regulation by this transcription factor. Vertebrate homologues of the Drosophila Notch/Delta gene family have been identified in Xenopus, mouse, and chick (Lardelli et al., 1994; Chitnis et al., 1995; Henrique et al., 1995; Uyttendaele et al., 1996; Jen et al., 1997) and constitute important signals for specification of cell fates. They are expressed in the neuroectoderm and in the presomitic mesoderm. Dll1 null mutant mice (Hrabe de Angelis et al., 1997) exert defects in somite formation, such as loss of craniocaudal polarity and epithelialization, which indicates that Notch/Delta signaling is essential for somitogenesis. Dll1 transcripts accumulate in the presomitic mesoderm immediately caudal to the newly formed somites overlapping with cMeso-1 expression. Preliminary results from this laboratory suggest that Delta expression is unaltered in embryos treated with antisense cMeso-1 oligonucleotides which would be consistent with cMeso-1 acting downstream of dll. The recently described MesP2 null mutant mice also show normal Delta expression but lack

FIG. 7. cMeso-1 expression is not influenced by anterior structures. Stage 9 –10 embryos were transversely dissected at different axial levels as depicted schematically in A–D (top level). Anterior (a) and posterior (p, contrasted by blue background) pieces were cultivated separately and newly formed somites (indicated schematically in dark blue) were counted. Already formed somites and the region of the next-to-form somites at the time of the operation are shown in yellow by solid and dotted lines, respectively. The expression domain of cMeso-1 is shown in red. Middle panels schematically illustrate the formation of somites and cMeso-1 expression after the incubation period. The bottom panels show anterior and posterior pieces following separate cultivation and the expression of cMeso-1 in the posterior pieces by whole-mount in situ hybridization. Note that both somitogenesis and cMeso-1 expression proceed regardless of the presence of preformed somites (A and B) or the anterior domain of cMeso-1 in the segmental plate (C). In asymmetrically dissected embryos somitogenesis proceeds unilaterally until the same axial level is reached on both sides and somite formation and cMeso-1 expression continue in register (D). The images are representative examples of at least six embryos operated in each group of dissections.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

212

Buchberger et al.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

213

cMeso1 in Somitogenesis

expression of Notch-1 and -2 as well as FGFR1 implying that these genes function downstream of MesP1 (Saga et al., 1997). Further studies will clarify the epstatic relationship of Notch/Delta and cMeso-1. Another gene with similar expression to cMeso-1 is the vertebrate fringe gene which has been isolated from mouse (Johnston et al., 1997) and chicken (Sakamoto et al., 1997). In Drosophila fringe acts upstream of Notch and is responsible for the determination of segment boundaries (Irvine and Wieschaus, 1994). The mouse fringe gene expressed in flies is functional which argues for a conserved function of the gene and possibly common pathways in both organisms (Johnston et al., 1997). It will be interesting to test the role of fringe for the expression of cMeso-1 in chicken. First evidence for the importance of cMeso-1 in somitogenesis is provided by the loss of function experiments shown in this paper. Both antisense RNA and oligonucleotides in ovo and on explanted embryos result in specific inhibition of cMeso-1 transcripts and lead to the arrest or marked delay of somitogenesis. Thus, our results strongly support the functional involvement of cMeso-1 in somitogenesis, in particular in the segmentation process. Recently, the expression of the bHLH protein paraxis in chicken embryos has been abolished by antisense oligonucleotides resulting in a similar disruption of somite formation (Barnes et al., 1997) which is in full agreement with the phenotype of paraxis null mutant mice (Burgess et al., 1996). Since paraxis expression in the paraxial mesoderm overlaps with cMeso-1 and loss of function generates a similar phenotype, one may suspect that both genes function in one pathway or may even collaborate. The dynamic expression of cMeso-1 in a limited domain of the anterior presomitic mesoderm raises the interesting question of how this pattern may be established. The simplest assumption would be that anterior signals emanating from the last formed somites would instruct the posteriorly adjacent mesoderm to express the gene. On the other hand, it has been shown that somitogenesis does not depend on anteroposterior signaling, as isolated segmental plate explants are fully capable of somite formation (Sandor and Amels, 1970; Packard and Jacobson, 1976; Bellairs and Veini, 1980; Sandor and Fazakas-Todea, 1980). This suggests that somites segregate according to an intrinsic mesodermal space–time program. We have tested whether cMeso-1 expression depends on the presence of anterior somites or the anterior cMeso-1 domain itself. Clearly, cMeso-1 was activated correctly in explants consisting of the posterior segmental plate without any anterior somites and without the cMeso-1 domain. The time and space of cMeso-1 gene activation appeared to be predetermined within the presomitic mesoderm together with somite formation itself (Bellairs, 1963; Packard and Jacobson 1976). Interestingly, c-hairy 1, another avian bHLH gene, has recently been shown to be expressed in the presomitic mesoderm in cyclic waves which correspond to the formation time of new somites (Palmeirim et al., 1997). Like for cMeso-1, the rhythmic expression of c-hairy1 is an autono-

mous property of the presomitic mesoderm and not driven by the propagation of an activating signal along the anteroposterior axis. In contrast to cMeso-1 however, c-hairy1 exhibits dynamic expression which extends to the posterior end of the presomitic mesoderm in reiterated waves corresponding to each formation cycle of new somites. Thus, c-hairy1 may be part of a counting mechanism which defines the precise time point at which somitogenesis should start. Since cMeso-1 expression is not reiterated in the same cells but seems to correlate in its oscillation pattern with c-hairy1, one may speculate that it constitutes part of the link between the molecular clock and segmentation and somitogenesis. Determination of the precise epistatic relationship of both genes will help to clarify this proposition.

ACKNOWLEDGMENTS We thank E. N. Olson and M. Hrabe de Angelis for the paraxis and dll probes, respectively. This work was supported by the Deutsche Forschungsgemeinschaft SFB 271 TP1 and the Fond der Chemischen Industrie.

REFERENCES Arnold, H. H., and Braun, T. (1996). Targeted inactivation of myogenic factor genes reveals their role during mouse myogenesis: A review. Int. J. Dev. Biol. 40, 27– 47. Auffray, C., and Rougeon, F. (1980). Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur. J. Biochem. 107, 303–314. Barnes, G. L., Alexander, P. G., Hsu, C. W., Mariani, B. D., and Tuan, R. S. (1997). Cloning and characterization of chicken paraxis: A regulatory of paraxial mesoderm development and somite formation. Dev. Biol. 189, 95–111. Bellairs, R. (1963). The development of the somites in the chick embryo. J. Embryol. Exp. Morphol. 11, 697–714. Bellairs, R., and Veini, M. (1980). An experimental analysis of somite segmentation in the chick embryo. J. Embryol. Exp. Morphol. 55, 93–108. Bellairs, R., Curtis, A. S. G., and Sanders, E. J. (1978). Cell adhesiveness and embryonic differentiation. J. Embryol. Exp. Morphol. 46, 207–213. Bettenhausen, B., Hrabe de Angelis, M., Simon, D., Guenet, J. L., and Gossler, A. (1995). Transient and restricted expression during mouse development of Dll1, a murine gene closely related to Drosophila Delta. Development 121, 2407–2418. Blanar, M. A., Crossley, P. H., Peters, K. G., Steingrimasson, 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. Natl. Acad. Sci. USA 92, 5870 –5874. Bober, E., Brand-Saberi, B., Ebernsperger, C., Wilting, J., Balling, R., Paterson, R. B., Arnold, H. H., and Christ, B. (1994). Initial steps of myogenesis in somites are independent of influence from axial structures. Development 120, 3073–3082. Buchberger, A., Pabst, O., Brandt, T., Seidl, K., and Arnold, H. H. (1996). Chick Nkx-2.3 represents a novel family member of vertebrate homologues to the Drosophila homeobox gene tin-

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

214

Buchberger et al.

man: Differential expression of cNKx-2.3 and cNKx-2.5 during heart and gut development. Mech. Dev. 56, 151–163. Buchberger, A., Schwarzer, M., Brand, T., Pabst, O., Seidl, K., and Arnold, H. H. (1998). The chicken winged-helix transcription factor cFKH-1 prefigures axial and appendicular skeletal structures during chicken embryogenesis. Dev. Dyn., in press. Burgess, R. P., Cserjesi, P., Ligon, K. L., and Olson, E. N. (1995). Paraxis: A basic helix-loop-helix protein expressed in paraxial mesoderm and developing somites. Dev. Biol. 168, 296 –306. Burgess, R. P., 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. Chitnis, A., Henrique, D., Lewis, J., Ish-Horowicz, D., and Kintner, C. (1995). Primary neurogenesis in Xenopus regulated by a homologue of the Drosophila neurogenic gene Delta. Nature 375, 761–766. Conlon, R. A., Reaume, A. G., and Roussant, J. (1995). Notch 1 is required for the coordinate segmentation of somites. Development 121, 1533–1545. Corbin, V., Michelson, A. M., Abmayr, S. M., Neel, V., Alcamo, E., Maniatis, T., and Young, M. W. (1991). A role for the Drosophila neurogenic genes in mesoderm differentiation. Cell 67, 311–323. Cserjesi, P., Brown, D., Ligon, K. L., Lyons, G. E., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., and Olson, E. N. (1995). Scleraxis: A basic helix-loop-helix protein that prefigures skeletal formation during mouse embryogenesis. Development 121, 1099 –1110. Del Amo, F., Smith, D. E., Swiatek, P. J., Gendron Maguire, M., Greenspan, R. J., McMahon, A. P., and Gridley, T. (1992). Expression pattern of Motch, a mouse homolog of Drosophila Notch, suggests an important role in early postimplantation mouse development. Development 115, 737–744. 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. 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. Cell 79, 1175– 1186. 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 amino-terminal cleavage product and modulation by the cyclic AMP signaling pathway. Cell 81, 457– 465. Garcia-Martinez, V., and Schoenwolf, G. C. (1993). Primitive streak origin of the cardiovascular system in avian embryos. J. Morphol. 61, 257–284. Hamburger, V., and Hamilton, H. (1951). A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49 –92. Hartenstein, A. Y., Rugendorff, A., Tepass, U., and Hartenstein, V. (1992). The function of the neurogenic genes during epithelial development in the Drosophila embryo. Development 116, 1203–1220. Henrique, D., Adam, J., 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. Hoffman, C. S., and Winston, F. (1987). A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli. Gene 57, 267–272. Hrabe de Angelis, M., McIntyre, J., II, and Gossler, A. (1997). Maintenance of somite borders in mice requires the Delta homologue Dll1. Nature 386, 717–721.

Irvine, K. D., and Wieschaus, E. (1994). fringe, a boundary-specific signaling molecule, mediates interactions between dorsal and ventral cells during Drosophila wing development. Cell 79, 595– 606. Jacobson, A., and Meier, S. (1986). Somitomeres: The primordial body segments. In ‘‘Somites in Developing Embryos’’ (R. Bellairs, D. A. Ede, and J. W. Lash, Eds.), pp. 1–16. Plenum Press, New York and London. Jen, W.-C., Wettstein, D., Turner, D., Chitnis, A., and Kintner, C. (1997). The Notch ligand, X-Delta-2, mediates segmentation of the paraxial mesoderm in Xenopus embryos. Development 124, 1169 –1178. 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. Kozak, M. (1997). Recognition of AUG and alternative initiator codons is augmented by G in position 14 but is not generally affected by the nucleotides in position 15 and 16. EMBO J. 16, 2482–2492. Lardelli, M., and Lendahl, U. (1993). Motch A and Motch B—Two mouse Notch homologues coexpressed in a wide variety of tissues. Exp. Cell Res. 204, 364 –372. Lardelli, M., Dahlstrand, J., and Lendahl, U. (1994). The novel Notch homologue mouse Notch 3 lacks specific EGF-repeats and is expressed in proliferating neuroepithelium. Mech. Dev. 46, 123–136. Meier, S. (1979). Development of the chick embryo mesoblast. Formation of the embryonic axis and establishment of the metameric pattern. Dev. Biol. 73, 25– 45. Olson, E. N., and Klein, W. H. (1994). bHLH factors in muscle development: Dead lines and commitments, what to leave in and what to leave out. Genes Dev. 8, 1– 8. Orr-Urtreger, A., Givol, D., Yayon, A., Yarden, Y., and Lonai, P. (1991). Developmental expression of two murine fibroblast growth factor receptors, flg and bek. Development 113, 1419 – 1434. Ostrovsky, D., Sanger, J., and Lash, J. W. (1988). Somitogenesis in the mouse embryo. Cell Differ. 23, 17–26. Packard, D. S., Jr. (1978). Chick somite determination: The role of factors in young somites and the segmental plate. J. Exp. Zool. 203, 295–306. Packard, D. S., Jr., and Jacobson, A. G. (1976). The influence of axial structures on chick somite formation. Dev. Biol. 53, 36 – 48. Palmeirim, I., Henrique, D., Ish-Horowicz, D., and Pourquie, O. (1997). Avian hairy gene expression identifies a molecular clock linked to vertebrate segmentation and somitogenesis. Cell 91, 639 – 648. Pourquie, O., Fan, C.-M., Coltey, M., Hirsinger, E., Wantanabe, 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 BMP-4. Cell 84, 461– 471. Primmett, D. R. N., Norris, W. E., Carlson, G., Keynes, R. J., and Stern, C. D. (1989). Periodic segmental anomalies induced by heat shock in the chick embryo are associated with the cell cycle. Development 105, 119 –130. Psychoyos, D., and Stern, C. D. (1996). Fates and migratory routes of primitive streak cells in the chick embryo. Development 122, 1523–1534. Reaume, 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. 154, 377–387.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

215

cMeso1 in Somitogenesis

Rosenquist, G. C. (1966). A radioautographic study of labeled grafts in the chick blastoderm. Development from primitive-streak stages to stage 12. Contrib. Embryol. Carnegie Inst. Wash. 38, 71–110. Saga, Y., Hata, N., Kobayashi, S., Magnuson, T., Seldin, M. F., and Taketo, M. M. (1996). MesP1: 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. 11, 1827–1839. Sakamoto, K., Yan, L., Imai, H., Takagi, M., Nabeshima, Y., Takeda, S., and Katsube, K. (1997). Identification of a chick homologue of Fringe and C-Fringe 1: Involvement in the neurogenesis and the somitogenesis. Biochem. Biophys. Res. Commun. 234, 754 –759. Sandor, S., and Amels, D. (1970). Researches on the development of axial organs. VI. The influence of the neural tube on somitogenesis. Rev. Roum. Embryol. 7, 49 –57. 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. Rev. Roum. Morphol. Embryol. Physiol. Morphol–Embryol. 26, 29 –32. Schoenwolf, G. C., and Sheard, P. (1990). Fate mapping the avian epiblast with focal injections of a fluorescent-histochemical marker: Ectodermal derivatives. J. Exp. Zool. 255, 323–329. Schoenwolf, G. C., Garcia-Martinez, V., and Dias, M. S. (1992). Mesoderm movement and fate during avian gastrulation and neurolation. Dev. Dyn. 193, 235–248. Slack, J. M. W. (1991). ‘‘From Egg to Embryo,’’ 2nd ed. Cambridge Univ. Press, Cambridge. Slack, J. M. W., and Tannahill, D. (1992). Mechanism of anterior posterior axis specification in vertebrates. Lessons from the amphibians. Development 114, 285–302.

Smith, J. C., and Howard, J. E. (1992). Mesoderm induced factors and the control of gastrulation. Development 1992(Suppl.), 127– 136. Sosic, D., Brand-Saberi, 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. Srivastava, D., Cserjesi, P., and Olson, E. N. (1995). A subclass of bHLH proteins required for cardiac morphogenesis. Science 270, 1995–1999. Tam, P. P. L., and Trainor, P. A. (1994). Specification and segmentation of the paraxial mesoderm. Anat. Embryol. 189, 275–305. Tepass, U., and Hartenstein, V. (1995). Neurogenic and proneural genes control cell fate specification in the Drosophila endoderm. Development 121, 393– 405. Uyttendaele, H., Marazzi, G., Wu, G., Yan, Q., Sassoon, D., and Kitajewski, J. (1996). Notch4/int-3, a mammary proto-oncogene, is an endothelial cell-specific mammalian Notch gene. Development 122, 2251–2259. Winnier, G. E., Hargett, L., and Hogan, B. L. M. (1997). The winged helix transcription factor MFH1 is required for proliferation and patterning of paraxial mesoderm in the mouse embryo. Genes Dev. 11, 926 –940. Yamaguchi, T. P., Conlon, R. A., and Rossant, J. (1992). Expression of the fibroblast growth factor receptor FGFR-1/flg during gastrulation and segmentation in the mouse embryo. Dev. Biol. 152, 75– 88. Yamaguchi, T. P. (1997). New insights into segmentation and patterning during vertebrate somitogenesis. Curr. Opin. Genet. Dev. 7, 513–518. Received for publication December 22, 1997 Revised April 7, 1998 Accepted April 7, 1998

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.