INTRODUCTION

Skeletal muscles of the trunk develop exclusively from somites(Christ et al., 1986). Within the somites two differentmyogenic cell lineages can be distinguished: (1) the so-calledmyotome, which gives rise to axial skeletal muscles (i.e. thedeep musculature of the back) and (2) cells that form themuscles in limbs and body wall (Christ et al., 1986; Wachtlerand Christ, 1992). Precursor cells for these latter musclesmigrate from the ventrolateral dermomyotome of early somitesinto the limb buds, prior to the formation of the myotome(Solursh et al., 1987; Christ et al., 1977; Ordahl and LeDouarin, 1992).

No genuine molecular markers for migrating myogenicprecursors have been described, making it difficult to followthis population of cells in situ. In contrast, myotomalmyoblasts and muscle cells in the developing limbs at laterstages can be identified readily, due to the expression ofmyogenic determination genes of the bHLH family (reviewedby Arnold and Braun, 1993). Transcripts of

Myf-5, theearliest myogenic regulatory factor, appear around day 8 p.c.(post coitum) during mouse development in undifferentiated,still epithelial somites (Ott et al., 1991). Expression ofmyogenin, Myf-6 and MyoD follows in cells of the myotome

in this sequence (reviewed by Buckingham, 1992; Arnold andBraun, 1993).

It has been shown recently that the formation of limbmuscles is strongly impaired in the homozygous splotchmutant mouse, while development of back and ventral bodywall musculature appears to be normal except for somereduction of dorsal muscles at the level of the neural tubedefect (Franz et al., 1993; Franz, 1993).

The murine splotch mutant is a well-defined model forneural crest and neural tube defects. Homozygous splotchanimals typically display neural tube defects and deficienciesin neural crest-derived tissues, such as dorsal root ganglia,sympathetic trunk, the septum of the truncus arteriosus, andSchwann and pigment cells (Auerbach, 1954; Moase andTrasler, 1989; Franz, 1989, 1990; Grim et al., 1992). Thevarious splotch alleles (Splotch, Sp; Sp-delayed, Spd; Sp-retarded, Spr; Sp1H; Sp2H) display a semi-dominant phenotypein structures closely associated with Pax-3 expression. Boththe Sp, Spr, Sp1H, Sp2H and Sp4H alleles have been character-ized at the molecular level and have been shown to correspondto mutations in the Pax-3 gene (Epstein et al., 1991, 1993,Vogan et al., 1993, Goulding et al., 1993a). Pax-3, a memberof the murine paired-box-containing gene family encodes atranscription factor with 479 amino acids in length containing

603Development 120, 603-612 (1994)Printed in Great Britain © The Company of Biologists Limited 1994

Limb muscles in vertebrates originate from dermomy-otomal cells, which during early development migrate fromthe ventrolateral region of somites into the limb buds.These progenitor cells do not express any muscle-specificmarker genes or myogenic transcription factors until theyreach their destination in the limbs. Here, we demonstrateby in situ hybridization that myogenic cells in somites anda population of presumably migratory muscle precursorcells in somatopleural tissue as well as myoblasts in thedeveloping limbs express

Pax-3. Significantly, in homozy-gous splotch mutant mice, which synthesize altered Pax-3mRNA but make no normal protein, no cells positive forPax-3 transcripts can be detected in the region of migratinglimb muscle precursors or in the limb itself. In contrast,

myotomal precursor cells and axial skeletal musclescontain Pax-3 transcripts also in the mutant. Interestingly,these animals fail to develop limb musculature as demon-strated by the lack of hybridization with various probes formyogenic transcription factors (Myf-5, myogenin, MyoD)but make apparently normal axial muscles. These obser-vations suggest that Pax-3 is necessary for the formation oflimb muscles, affecting either the generation of myogenicprecursors in the somitic dermomyotome or the migrationof these cells into the limb field.

Key words: myogenic factors, limb muscle, Pax-3, splotch, somites,cell migration, myoblasts, dermomyotome, mouse

SUMMARY

Pax-3 is required for the development of limb muscles: a possible role for the

migration of dermomyotomal muscle progenitor cells

Eva Bober1, Thomas Franz2, Hans-Henning Arnold1, Peter Gruss3 and Patrick Tremblay3,*1Department of Cell and Molecular Biology, Technical University of Braunschweig, Spielmannstr. 7, 38106 Braunschweig,Germany2Institute of Anatomy, Department of Neuroanatomy, University Hospital Eppendorf, Martinistr. 52, 20246 Hamburg, Germany3Department of Molecular Cell Biology, Max Planck Institute for Biophysical Chemistry, Am Fassberg, 37077 Göttingen, Germany

*Author for correspondence

604

two conserved DNA-binding motifs: the 128 amino acid paireddomain and a 60 amino acid paired-type homeodomain(Goulding et al., 1991). Pax-3 expression has been detectedduring brain development, in the alar and roof plates of thedeveloping neural tube, in various neural crest derivatives (i.e.Schwann cells and dorsal root ganglia) and within somites(Goulding et al., 1991).

Whereas the expression of Pax-3 in neural structures andmost aspects associated with it in splotch mutants have beendocumented extensively, the role of Pax-3 for the developmentof paraxial mesoderm remained unclear. The specific skeletalmuscle phenotype recently observed in the splotch mutantindicates that Pax-3 may be essential for normal developmentof limb muscles that are known to be derived from the lateralregion of somites. In order to gain some insight into thepossible role of Pax-3 in developing somites and to understandbetter the splotch-associated defect in limb muscle formation,we analysed the distribution of cells containing Pax-3 andmyogenic factor transcripts in normal and homozygous splotchmutant mice. We demonstrate that the Pax-3 probe marks apopulation of cells that migrates from the lateral dermomy-otome into the limbs. We furthermore show that this cell pop-ulation is missing in the splotch mutant. We also show thattranscripts for Pax-3 and myogenic factor genes that appearcoexpressed in normal limbs are absent in mutant embryos.From these results, we conclude that Pax-3 may be function-ally involved in the formation of limb muscle originating fromearly somitic dermomytome.

MATERIALS AND METHODS

MiceSp, Sp1H and Sp2H mouse lines were obtained from Jackson Labora-tories (Bar Harbour) and MRC Radiobiology Unit (Harwell, Didcot,England). They were bred with C57BL/6 (Sp) and C3H (Sp1H andSp2H) strains. Embryos were staged, counting the appearance of thevaginal plug as day 0.5 p.c.

Whole-mount hybridizationExpression of the Pax-3 gene was analysed by in situ hybridizationusing a 520 bp HindIII/PstI fragment of the cDNA. It encodes the 3

′part of the paired-type homeodomain and most of the carboxytermi-nus (Goulding et al., 1991).

Mouse embryos were collected and treated as described byWilkinson (1992). Briefly, embryos were fixed in 50 mM phosphate-buffered saline pH 7.2 (PBS) containing 4% paraformaldehyde for 16hours followed by two washing steps in PBT (PBS and 0.1% Tween-20) and dehydration with 25%, 50%, 75% and100% methanol. Fixedand dehydrated embryos were stored at −20°C.

For whole-mount in situ hybridization, embryos were rehydratedthrough the methanol series, washed twice in PBT and bleached inPBT containing 6% hydrogen peroxide for 1 hour. Embryos were thensubjected to digestion with 10 µg/ml of proteinase K for 7-15 minutesfollowed by washing in glycine solution (2 mg/ml). These embryoswere refixed in gluteraldehyde/paraformaldehyde (0.2%/4%) for 20minutes, washed twice in PBT and transferred to hybridization buffer(50% formamide, 5× SSC pH 4.5, 50 µg/ml yeast RNA, 1% SDS, 50µg/ml heparin) for prehybridization at 70°C for at least 1 hour.Synthesis of sense and antisense digoxigenin-labelled RNA probeswas performed using a DIG RNA Labeling Kit (BoehringerMannheim) according to the manufacturer´s instructions. Hybridiz-ation was performed in the same buffer as used for prehybridization

containing 1 µg/ml of digoxigenin-labelled RNA probe at 70°Covernight. Following hybridization, embryos were washed twice for30 minutes in 50% formamide, 5× SSC pH 4.5, 1% SDS at 70°C,followed by two washing steps in 0.5 M NaCl, 10 mM Tris-HCl pH7.5, 0.1% Tween-20 with 100 µg/ml RNase A at 37°C and two stepsin 50% formamide, 2× SSC pH 4.5. Embryos were then rinsed severaltimes in TBST buffer (0.14 M NaCl, 25 mM KCl, 25 mM Tris-HClpH 7.5, 1% Tween-20, 2 mM levamisole) and incubated overnight at4°C with sheep polyclonal alkaline phosphatase-conjugated anti-digoxigenin antibody (Boehringer Mannheim) preabsorbed withmouse embryo powder. Next day, embryos were washed first withTBST, then in NTMT buffer (100 mM NaCl, 100 mM Tris-HCl pH9.5, 50 mM MgCl2, 1% Tween-20, 2 mM levamisole) and the phos-phatase reaction was performed in the presence of nitroblue-tetra-zolium chloride (0.34 mg/ml) and 5-bromo-4-chloro-3-indolyl-phosphate (0.18 mg/ml) in NTMT buffer. The reaction was stoppedin PBT.

Embryos were either cleared in glycerol PBT solution or embeddedin a mixture of gelatin, albumen and sucrose, and sectioned at 50 µmusing the Pelco 101 vibratome. Sections were mounted on gelatin-subbed slides and photographed with a Zeiss Axiophot microscopeusing bright-field illumination. Cleared embryos were photographedunder a Zeiss SV11 stereomicroscope using a DIA-duplicator(Elinchrom Co.) to provide a flashlight.

In situ hybridizationIn situ hybridization was performed on 7 µm thick paraffin tissuesections as described previously (Sassoon et al., 1988; Bober et al.,1991). The 35S-labelled probes for myogenin and MyoD wereprepared as described by Sassoon et al. (1988), and for Myf-5 asdescribed by Ott et al. (1991). The Pax-3 probe was used as describedfor whole-mount in situ hybridization.

RESULTS

Pax-3 transcripts in developing somites of normaland splotch mutant embryosIn order to determine Pax-3 expression in somites of wild-typeand splotch mutant embryos, a series of whole-mount in situhybridizations were performed between days 9 and 12 p.c. Assomites develop in a rostrocaudal direction, newly formedundifferentiated somites and already fully differentiatedsomites can be observed in the same embryo in caudal andcranial regions respectively (Fig. 1). We found that Pax-3expression was evenly distributed over newly formed somites(Fig. 1A: day 9.5 p.c., caudal part), whereas at later stages ofsomitic development the Pax-3 signal was confined to the most

E. Bober and others

Fig. 1. Expression of the Pax-3 gene in normal and homozygous Spembryos visualized by whole-mount in situ hybridization. Lateralviews. (A-E) Control embryos; (F-J) homozygous Sp embryos. Atday 9.5 p.c. (A,F) arrowheads point at the beginning redistribution ofthe homogeneous Pax-3 signal along the caudal and ventrolateralsomitic edges in normal (A) but not in mutant embryos (F). At day10-10.5 p.c. (B,C,G,H), arrowheads mark the signal concentrated atthe ventrolateral bud of normal (C) and mutant embryos (H). C andH are enlargments of B and G, respectively. At day 11-11.5 p.c.(D,E,I,J) arrowheads demarcate the compact Pax-3 signal of normalembryos; compare with homozygous mutants where the signal ishomogenously distributed and disorganized at its ventral tip. Arrow:in Sp embryos, the Pax-3 domain is significantly shifted laterally,away from the neural tube. Abbreviations: fl, forelimb; hl, hind limb;nt, neural tube; sb, spina bifida.

605

Pax-3 is an early marker for myogenic precursors

caudal and ventrolateral edge in eachsomite (Fig. 1A-C, arrowhead, see alsoFig. 2C,D). The restricted localization ofthe Pax-3 signal was most pronouncedaround day 11 p.c., when it wasclustered in the ventrolateral budappearing in individual somites of thetrunk (Fig. 1D,E, arrowhead). Inaddition a clearly weaker expressiondomain was observed more dorsomedi-ally in two parallel stripes along theanterior and posterior borders of eachsomite (Fig. 1E, arrow). This distribu-tion of Pax-3 transcripts may correlate toyet unidentified functional domainswithin developing somites.

In homozygous splotch embryos,distinct differences in the distribution ofPax-3 transcripts were observed (Fig.1F-J). While the Pax-3 signal in earlysomites appeared similar to wild-type,no local redistribution to the ventrolat-eral part occurred in the more advancedsomites (day 9.5 p.c.; Fig. 1F,arrowhead). Later, (day 10.5-11.5 p.c.)when the somitic domain labelled byPax-3 normally elongated further ven-trolaterally, this area appeared muchshorter and disorganized in mutantembryos (Figs 1G-J, 2E,F). The edges ofsomites delineated by the Pax-3 signalalso appeared irregular (Fig. 1H,J,arrowhead). Furthermore, the Pax-3expression domain was slightly shiftedventrally, away from the neural tube(Fig. 1J, arrow).

Similar evidence for the disorganiza-tion of somites in homozygous splotchembryos (days 10.5 and 11.5 p.c.) wasalso observed when the myotomalmarker Myf-5 was used as a probe (datanot shown). Thus, the structural organ-ization of developing somite seemed tobe altered in splotch animals.

Migration of Pax-3-expressingcells from somiticdermomyotome into the limbbudsIn addition to somites, Pax-3 was alsofound to be expressed in the developinglimbs (Figs 1, 2). Pax-3-positive cellswere present in forelimbs and also inhind limbs of control embryos betweenday 9.5 and 10.5 p.c. (Fig. 1A-C). Inmarked contrast, no positive cells wereobserved in limbs of homozygoussplotch animals at the equivalent stages(Fig. 1F-H).

Fig. 2 illustrates a time course of Pax-3 transcript accumulation in the devel-Fig. 1

606

oping forelimbs of control (Fig. 2A-D) and homozygoussplotch embryos (Fig. 2E,F), as seen in a dorsal view. Pax-3-expressing cells appeared to emigrate from 5 to 6 adjacentsomites toward the limb field (Figs 1A, 2). The progression ofPax-3 signals from the somites into the forelimbs was firstobserved at day 9.5 p.c., increasing in a proximodistal directionbetween days 10.0 and 11.0 p.c., and decreasing thereafter. Theearly distribution of Pax-3 signal appeared to be compact (Fig.2A,B), while it displayed a more scattered distribution of Pax-3-expressing cells at later stages (Fig. 2C,D). A similar colo-nization of the hind limbs by Pax-3-expressing cells was alsoobserved lagging about 0.5 day behind the forelimbs (data notshown and Figs 3, 4). Fig. 3 illustrates the progressive proxi-modistal invasion of forelimbs and hind limbs by Pax-3-expressing cells originating from the ventrolateral somiticregion. The early expression seemed to be confined to the der-momyotomal cap of the somites (Fig. 3A). As this structureelongated ventrolaterally (Fig. 3B), cells appeared to detachfrom the somitic domain, accumulate at the body-limb junctionand then further invade the limbs (Fig. 3C,D). Approximatelyat day 10.5 p.c., the Pax-3-expressing cell population segre-gated into dorsally and ventrally located domains (Fig. 3E,F).Significantly, no Pax-3-expressing cells that would leave thesomites or populate the limbs between days 9.5 and 11.5 p.c.were observed in homozygous splotch embryos, although Pax-3 expression was very strong in the dermomyotome and theneural tube of these animals (Figs 1-3). These findings stronglysuggest that dermomytomal myogenic precursors either do notexist or are unable to migrate into the limbs in the homozy-gous splotch mutant.

Pax-3 expression domains overlap with theexpression of the myogenic regulatory factor Myf-5in somites and the developing limbs The spatiotemporal expression of Pax-3 in paraxial mesodermand regions of the limb buds suggests a potential role in thedevelopment of skeletal muscles. In order to correlate Pax-3

expression domains to early muscle lineage markers, in situhybridizations using Pax-3 and Myf-5 probes were performedon serial sections of mouse embryos. It has been shown previ-ously that Myf-5 constitutes the first myogenic regulatory genethat is expressed in somites and muscle cells in the developinglimbs (Ott et al., 1991).

Fig. 4 shows in situ hybridizations on transverse sections ofa day 10.5 p.c. mouse embryo at the forelimb and hind limblevels. Both Pax-3 and Myf-5 transcripts were detected insomites and limb buds in overlapping but not completelyidentical areas. While Myf-5 transcripts were confined to themyotomal layer of somites, Pax-3 expression was clearly widerincluding both myotome and dermatome (Fig. 4A-D,K,L).

Similar expression domains for Pax-3 and Myf-5 wereobserved in two locations of forelimbs at the ventral and dorsalsides, which later give rise to the flexor and extensor muscles(Fig. 4I,J). At the same stage, Pax-3 transcripts have alreadyaccumulated in the hind limb, whereas Myf-5 was not yetexpressed (Fig. 4K,L). This indicates the later onset of Myf-5gene activity compared to Pax-3, which was also observed inforelimbs of younger embryos (data not shown). Interestingly,Pax-3 transcripts were also detected in the region betweentrunk and limbs (Fig. 4A,C), an area that presumably containsmigrating muscle precursor cells as shown by chicken/quailsomite grafting experiments (Christ et al., 1977; Ordahl and LeDouarin, 1992). These cells can not be recognized with theMyf-5 probe as they express no myogenic bHLH factors (Fig.4B; Ott et al., 1991; Bober et al., 1991).

Later during development, Pax-3 expression graduallydecreased to almost undetectable levels at day 12.5 p.c. (datanot shown). In addition to mesodermal cells, Pax-3 transcriptswere also seen in the dorsal region of the neural tube asdescribed earlier (Goulding et al., 1991).

From these results, we conclude that Pax-3 is expressed inmesodermal cells that give rise to axial muscles and thepremuscle cells in the limbs. Its expression starts clearly priorto Myf-5 in somites and overlaps with Myf-5 expression in the

E. Bober and others

Fig. 2. Dorsal view of control and homozygous Sp embryos at thefore limb level: Expression of Pax-3. (A,D) Control animalsbetween days 9.5 and 10.5 p.c. Note the progressive migration ofPax-3-positive cells into the limbs delineated by the broken line.(E) and (F) demonstrate that no Pax-3-positive cells colonize thefore limbs in Sp mutants. The number of somites in each studiedembryo is written in the lower right corner.

607Pax-3 is an early marker for myogenic precursors

myotome. Unlike Myf-5, it is also expressed in the lateraldermatome, which represents cells that migrate to the limbfield where they form muscle.

Myogenic determination genes are not expressed inlimbs of splotch embryosBased on morphology and the expression of the muscle marker

desmin, it has been suggested previously that the splotchmutation affects normal development of skeletal musculaturein the limbs (Franz et al., 1993). In order to assess at whichstage limb muscle development might be impaired by the Pax-3 mutation, we performed in situ hybridizations using themyogenic bHLH genes Myf-5, MyoD and myogenin as probes.As these genes encode cell-type-specific transcription factors

Fig. 3. Expression of the Pax-3 gene in limbs of control and homozygous Sp embryos. Vibratome sections obtained from whole-mount in situembryos. (A-C) Sections at the hind limb level of control embryos; (D-F) sections at the fore limb level of control embryos; (G,H) sections ofhomozygous Sp embryos at the fore limb level. Note that the progressive migration of Pax-3-positive cells in control embryos is completelyabsent in the mutant. The corresponding number of somites is written in the lower right corner.

608

that are essential for the establishment of the skeletal musclelineage, they constitute highly specific and early markers(Arnold and Braun, 1993).

Comparison of myogenin and MyoD expression in day 11.5p.c. embryos of wild-type and homozygous splotch mutantmice revealed that both myogenic regulatory genes wereexpressed at similar levels in axial and body wall muscles ofcontrol and mutant animals (Fig. 5 and data not shown).

However, no myogenin and MyoD expression was found inlimbs of mutant embryos, although control animals of the samestage highly expressed these factors in limb musculature (Fig.5 and data not shown).

Similar results were obtained in day 9.5 and 10.5 p.c.embryos using Myf-5 as an additional probe (data not shown).This analysis was extended to day 13.5 p.c. embryos, the latestviable stage of the splotch mutant. As shown in Fig. 6, no

E. Bober and others

Fig. 4. Comparison of Myf-5 and Pax-3 expression in trunk somites and developing limbs of a day 10.5 p.c. wild-type embryo. Adjacenttransverse sections were hybridized with Pax-3 (A,I,C,K) and Myf-5 (B,J,D,L) probes. The left panel shows dark-field illuminations underlower magnification (×250). The middle and right panels present higher magnification (×450) of the limb area under bright- and dark-fieldoptics, respectively. The two upper rows demonstrate forelimbs, the lower ones hind limbs. Abbreviations: d and v indicate dorsal and ventralpremuscular masses in the forelimb; nt, neural tube; s, somites; arrows point at the Pax-3-positive cells at the body-limb junction.

609Pax-3 is an early marker for myogenic precursors

muscle could be labelled in limbs of the splotch mutant usingMyoD as a hybridization probe, whereas all muscle groups incontrol animals were highly positive for this marker. Incontrast, no major difference for the expression of MyoD wasobserved in axial and body wall muscles of wild-type andmutant embryos. Occasionally,a slight reduction of deep backmuscles was seen in mutants atthe level of the neural tubedefect (data not shown).

Taken together, these obser-vations suggest that a mutationin the Pax-3 gene causes asevere defect in the formation ofthe limb muscles, whereas axialmuscles are less or not affected.The defect appears not to be dueto a delay of limb muscleformation but rather to theinability of dermomyotomalmuscle precursor cells to reachthe limb bud.

DISCUSSION

Somitic Pax-3 expressionincludes myotomal anddermomyotomal cellsWhole-mount in situ hybridiza-tions on early mouse embryosindicate that the Pax-3 gene isexpressed over the entire area of

newly formed somites. During somite maturation, this Pax-3-positive domain undergoes a remarkable redistributionbetween days 9 and 9.5 p.c. of development. At this stage, Pax-3 transcripts become confined to the caudal region and the ven-trolateral aspects of each somite. As somites further mature

Fig. 5. Absence of myogenic cells in developing limbs of a day 11.5 p.c. splotch embryo. Hybridization with myogenin probe on sections at theforelimb level (×65) are shown for control (A,C) and homozygous Sp1H embryos (B,D). Higher magnifications (×250) of sections at the samelevel from a splotch embryo hybridized with myogenin (E,G) or MyoD (F,H) are presented.

Fig. 6. MyoD expression in developing limb and axial muscles of day 13.5 p.c. wild-type and Sp1H

embryos. Dark-field pictures of transverse sections at the hind limb level (×30) hybridized with MyoDare shown. (A) Day 13.5 p.c. of control embryo; (B) age-matched Sp1H embryo.

610

(between day 10 and 10.5 p.c.) and dermomyotomal cells beginto invade adjacent limbs and ventral body wall, the ventrolat-eral part of somites retain high levels of Pax-3 expression. Atthe trunk level, epithelial-like Pax-3-positive cells progres-sively invade the lateral somatopleure. Based on this distribu-tion, we believe that these Pax-3-expressing cells correspondto actively proliferating precursor cells emanating from thedermomyotome (Kaehn et al., 1988).

In contrast to Pax-3-expressing cells in the myotome, whichalso express the muscle-specific transcription factor Myf-5,lateral dermomyotomal cells are devoid of any myogenicbHLH proteins. Therefore, Pax-3 constitutes a convenientmarker to label this cell population.

Pax-3 transcripts are present in migratory cells, theprospective myoblasts of the limbsApproximately between days 9 and 9.5 p.c., Pax-3-positivecells can be observed emerging from the dermomyotome at thelevel of the forelimb bud. About half a day later, similar cellscan be seen next to the hind limb bud. These cells appeardispersed in small clusters rather than as sheets of epithelialcells. At later stages, the Pax-3 signals become organized in twodomains in the limbs, one located dorsally and one ventrally inthe position of the limb premuscle masses. These Pax-3-positive cells colocalize with Myf-5-expressing cells. As theother myogenic factors begin to be expressed (MyoD,myogenin) and muscle differentiation starts, the Pax-3 signaldisappears gradually from this region (days 11-12 p.c.). Takentogether, this evidence suggests to us that Pax-3-expressingcells invade the limbs and correspond to migratory limb muscleprecursors. The spatiotemporal distribution of Pax-3-positivecells, their progressive proximodistal extension into the limbs,and the subsequent organization into dorsal and ventral cellmasses that also express myogenic factors support this notion.The fact that Pax-3-positive cells colonizing the limbs originatefrom 5-6 adjacent somites is in agreement with previous reportson the origin of limb myoblasts in chicken (Christ et al., 1997;Hayashi and Ozawa, 1991; Beresford, 1983).

Since migratory somitic cells have been shown to give riseonly to skeletal muscle, while mesenchymal components of thelimb originate from the somatopleure (Chevallier et al., 1977;Christ et al., 1977; Wachtler and Christ, 1992; Ordahl and LeDouarin, 1992), it appears unlikely that these Pax-3-express-ing cells contribute to other structures than myogenic limbprecursor cells. However, formal proof of this point willrequire lineage analysis or functional assays. Pax-3 also labelscells in the ventrolateral bud of somites, which migrate into thelateral somatopleure and give rise to body wall muscles (Christet al., 1983). Again, Pax-3 expression appears to label earlymigrating myogenic precursors at the trunk level.

Until now, no molecular marker for these migrating cells hasbeen available and they have only be followed by interspeciestransplantation (chicken-quail) or by injection of vitallipophilic dye into the somitic cavity (Christ et al., 1977;Ordahl and Le Douarin, 1992; Hayashi and Ozawa, 1991).Pax-3 appears to be a suitable marker to trace these earlymyogenic precursors both at the limb and trunk levels.

A functional Pax-3 protein is required for theestablishment of muscle progenitor cells in the limbThe recent demonstration of a dramatic loss of limb muscula-

ture in homozygous Sp1H and Spd mice (Franz et al., 1993)underlines the importance of Pax-3 for the development oflimb muscles. The experiments presented here demonstratethat, in these mice, Pax-3-expressing cells can be detected inthe dermomyotome but not in the lateral region betweensomites and limb buds or in the limbs itself. While otherdomains of Pax-3 expression such as in the dorsal neural tubeand the somitic myotome are preserved in splotch, the limbexpression domain is specifically missing in homozygous Sp,Sp1H and Sp2H mice. In addition, no myoblasts expressingmyogenic factors are detected in the limbs at any stage. Takentogether, these results strongly suggest that Pax-3 protein isrequired either for the generation of early limb muscle precur-sors within the somites or for the migration of these progeni-tors from the dermomyotome into the limbs. Alternatively, itremains formally possible that the population of myogeniccells shuts off Pax-3 expression in the mutant thereby escapingdetection, although they still migrate from the somites. If thiswould be the case, their myogenic potential should bedependent on a functional Pax-3 protein and differentiationwould be blocked in its absence.

Homozygous splotch embryos (Sp1H) appeared to havenormal facial muscles but some reduction of the tonguemuscles was observed in our analysis. However, whether thisrepresents a muscle deficiency or a delay in muscle develop-ment will require further investigation. Interestingly, thetongue muscles originate from occipital somites, from whichprecursors disssociate before migrating anteriorly(Schemainda 1979; Noden 1983; Couly et al., 1993). In thisregard, muscle development in the tongue and the limbs issimilar, as in both cases precursors leave the somites and reachtheir final destination before they differentiate. Thus, it remainslikely that Pax-3 mutations predominantly affect migratoryprecursors. Since axial, facial and body wall muscles, incontrast to limb muscles, develop almost normally in thesplotch mutant (Franz et al., 1993; Franz, 1993), Pax-3 appearsnot to be required in general for muscle differentiation.However, as some reduction of the axial muscle mass at thelevel of the neural tube defect has also been observed in splotchmice (Franz et al., 1993), a role for Pax-3 in the control of earlymyoblast proliferation cannot be excluded. Two other lines ofevidence may support such a role for Pax-3. Firstly, aberrantPAX3 transcripts are found in a human alveolar rhab-domyosarcoma associated with a frequent chromosomaltranslocation. This abnormal PAX3 expression may favor pro-liferation of myogenic precursors thereby inhibiting theirdifferentiation (Barr, 1993). Secondly, Pax genes have beenshown to be able to transform NIH 3T3 and 208 fibroblasts invitro (Maulbecker and Gruss, 1993).

Because Pax-3 expression precedes the establishment ofdorsoventral polarity in somites, it could conceivably beinvolved in a cascade leading to the specification of the der-momyotome. The expression of Pax-3 in the neural tube hasalready been shown to respond to ventralizing signals from thenotochord (Goulding et al., 1993b). In fact, a ventralizingfunction of the notochord and floor plate can be assumed forthe paraxial mesoderm. Transplantation of these structures tothe dorsolateral part of the neural tube repatterns adjacentsomites, thereby preventing muscle formation and stimulatingsclerotomal differentiation (Pourquie et al., 1993). Furtherexperiments will be required to determine whether dorsal

E. Bober and others

611Pax-3 is an early marker for myogenic precursors

somitic structures develop in a default pathway in the absenceof ventralizing signals or whether a dorsalizing influence isnecessary for the establishment of Pax-3 expression and theformation of dermomyotome.

Thus, in addition to being an important factor whose inacti-vation can lead to neural crest and neural tube defects, Pax-3is essential for the specification, proliferation and/or migrationof myoblast stem cells destined to colonize the limbs and giverise to the different muscles of the limbs.

Interestingly, the various neural crest deficiencies observedin splotch appear to result from a migration defect since neuralcrest cells isolated in vitro migrate away from the neural tubein a delayed fashion (Moase and Trasler, 1990). Although themolecular basis for this aberrant cellular behavior is unknown,we assume that, for instance, some cell adhesion molecules ortheir cellular receptors normally involved in guidance ofmyoblasts and neural crest cells, may be deficient in splotchanimals. At the limb level, fibronectin, hyaluronic acid andother mucopolysaccharides are possibly involved in thisprocess (for a review see Wachtler and Christ, 1992). Thegeneral disorganization of the dermomyotome at the trunklevel in the splotch embryos may support the hypothesis thatthe morphological integrity of tissues depends on the respec-tive adhesive properties of their cellular components. It hasalready been postulated that homeobox-containing proteinsmay be directly involved in the transcriptional control of suchmolecules (Edelman and Jones, 1992). Recently, binding toand activation of the NCAM gene promoter by the Hox B8 andB9 proteins has been demonstrated in tissue culture experi-ments (Jones et al., 1992). Pax proteins may participate in sucha process as well.

Various lesions in PAX3, which represents the humanhomologue of the murine Pax-3 gene, have been associatedwith the human Waardenburg syndromes (Hoth et al., 1993).These syndromes (WS type 1, type 2 and type 3), generallyinherited in an autosomal dominant trait, constitute a pheno-typically and genetically heterogeneous family of disorders.However, the major clinical manifestations of these syndromesinvolve a combination of defects of neural crest origin: dystopiacanthorum, (lateral displacement of the inner corner of the eye),deafness and pigmentation deficiencies (white forelock andeyelashes, hypopigmented iris, premature greying, hypopig-mented skin lesions). Interestingly, WS3 patients suffer fromlimb deformities resulting from muscle and skeletal hypoplasia(McKusick, 1992). The data presented here for mouse, demon-strating that Pax-3 is essential for the proper specification, pro-liferation and/or migration of myoblast precursors destined tocolonize the limb, provide a possible explanation for thephenotype observed in patients with WS3. However, one mustbear in mind that the phenotype observed in WS3 patients isdominant, observed in individuals with one normal PAX3 allele.It remains to be seen whether quantitative differences in themigration of myoblast precursors can also be observed in miceheterozygous for the Sp mutation.

We thank Christiane Müller for excellent technical assistance. Weare particularly grateful to Thomas Braun for helpful discussions andFabienne Pituello, Luc St-Onge and Michael Kessel for criticalcomments on this manuscript. P. T. is the recipient of a fellowshipfrom the Medical Research Council of Canada. This work wassupported by grants of the Deutsche Forschungsgemeinschaft and theEuropean Community to H. H. A. and by the Max Planck Society.

REFERENCES

Arnold, H.-H. and Braun, T. (1993). Myogenic control genes in vertebrates.In Advances in Developmental Biochemistry vol. 2 (ed. P. M. Wassarman),pp. 111-158. London: JAI Press.

Auerbach, R. (1954). Analysis of the developmental effects of a lethalmutation in the house mouse. J. Exp. Zool. 127, 305-329.

Barr, F.G. (1993). Rearrangement of the PAX3 paired box gene in thepediatric solid tumor alveolar rhabdomyosarcoma. Nature Genet. 3, 113-117.

Beresford, B. (1983). Brachial muscles in the chick embryo: the fate ofindividual somites. J. Embryol. Exp. Morph. 77, 99-116.

Bober, E., Lyons, G. E., Braun, T., Cossu, G., Buckingham, M. and Arnold,H-H. (1991). The muscle regulatory gene, myf-6, has a biphasic patternof expression during early mouse development. J. Cell Biol. 113, 1255-1265.

Buckingham, M. (1992). Making muscle in mammals. Trends Gen. 8, 144-149.

Chevallier, A., Kieny, M., Mauger, A. and Sengel, P. (1977). Developmentalfate of the somitic mesoderm in the chick embryo. In Vertebrate Limb andSomite Morphogenesis (ed. D.A. Ede, J. R. Hinchliffe and J. Balls), pp. 421-432. London: Cambridge University Press.

Christ, B., Jacob, H. and Jacob, M. (1977). Experimental analysis of theorigin of the wing musculature in avian embryos. Anat. Embryol. 150, 171-186.

Christ, B, Jacob, M. and Jacob, H. J. (1983). On the origin and developmentof the ventrolateral abdominal muscles of the avian embryo. Anat. Embryol.166, 87-101.

Christ, B., Jacob, M., Jacob, H. J., Brand, B. and Wachtler, F. (1986). InSomites and Developing Embryos, pp. 261-276. New York: Plenum Press

Couly, G. F., Coltey, P. M. and LeDouarin, N. M. (1993). The triple origin ofskull in higher vertebrates: a study in quail-chick chimeras. Development117, 409-429.

Edelman, G. M. and Jones, F. S. (1992). Cytotactin: a morphoregulatorymolecule and a target for regulation by homeobox gene products. TrendsBiochem. 17, 228-232.

Epstein, D. J., Vekemans, M. and Gros, P. (1991). Splotch (Sp2H), a mutationaffecting development of the mouse neural tube, shows a deletion within apaired homeodomain of Pax-3. Cell 67, 767- 774.

Epstein, D. J., Vogan, K. J., Trasler, D. G. and Gros, P. (1993). A mutationwithin intron 3 of the Pax-3 gene produces aberrantly spliced mRNAtranscripts in the splotch (Sp) mouse mutant. Proc. Natl. Acad. Sci. USA 90,532-536.

Franz, T. (1989). Persistent truncus arteriosus in the Splotch mutant mouse.Anat. Embryol. 180, 457-464.

Franz, T. (1990). Defective ensheathment of motoric nerves in the Splotchmutant mouse. Acta Anat. 138, 246-253.

Franz, T. (1993). The Splotch (Sp1H) and Splotch-delayed (Spd) alleles:differential phenotypic effects on neural crest and limb musculature. Anat.Embryol. 187, 371-377.

Franz, T., Kothary, R., Surani, M. A. H., Halata, Z. and Grim, M. (1993).The Splotch mutation interferes with muscle development in the limbs. Anat.Embryol. 187, 153-160.

Goulding, M. D., Chalepakis, G., Deutsch, U., Erselius, J. R. and Gruss, P.(1991). Pax-3, a novel murine DNA-binding protein expressed during earlyneurogenesis. EMBO J. 10, 1135-1147.

Goulding, M., Sterrer, S., Fleming, J., Balling, R., Nadeau, J., Moore, D.,Brown, S. D. M., Steel, K. P. and Gruss, P. (1993). Analysis of the Pax-3gene in the mouse mutant splotch. Genomics 17, 355-363.

Goulding, M. D., Lumsden, A. and Gruss, P. (1993). Signals from thenotochord and floor plate regulate the region-specific expression of two Paxgenes in the developing spinal cord. Development 117, 1001-1016.

Grim, M., Halata, Z. and Franz, T. (1992). Schwann cells are not required forguidance of motor nerves in the hind limb in Splotch mutant mouse embryos.Anat. Embryol. 186, 311-318.

Hayashi, K. and Ozawa, E. (1991). Vital labelling of somite-derivedmyogenic cells in the chicken limb bud. Roux’s Arch. Dev. Biol. 200, 188-192.

Hoth, C. F., Milunski, A., Lipsky, N., Sheffer, R., Clarren, S. K. andBaldwin, C. T. (1993). Mutations in the paired domain of the human PAX3gene cause Klein-Waardenburg syndrome (WS-III) as well as Waardenburgsyndrome type I (WS-I). Am. J. Hum. Genet. 52, 455-462.

Jones, F. S., Prediger, E. A., Bittner, D. A, De Robertis, E. M., Edelman, G.R. (1992). Cell adhesion molecules as targets for Hox genes: neural cell

612

adhesion molecule promoter activity is modulated by cotransfection withHox-2.5 and -2.4. Proc. Natl. Acad. Sci. USA 89, 2086-2090.

Kaehn, K., Jacob, H. J., Christ, B., Hinrichsen, K. and Poelmann, R. E.(1988). The onset of myotome formation in the chick. Anat. Embryol. 177,191-201.

McKusick, V. A. (1992). Mendelian inheritance in Man. Baltimore: JohnsHopkins University Press.

Maulbecker, C. C. and Gruss, P. (1993). The oncogenic potential of Paxgenes. EMBO J. 12, 2361-2367.

Moase, C. E. and Trasler, D. G. (1989). Spinal ganglia reduction in theSplotch-delayed mouse neural tube defect mutant. Teratology 40, 67-75.

Moase, C. E. and Trasler, D. G. (1990). Delayed neural crest cell emigrationfrom Sp and Spd mouse neural tube explants. Teratology 42, 171-182.

Noden, D. M.(1983). The embryonic origins of avian cephalic and cervicalmuscles and associated connective tissues. Am. J. Anat. 168, 257-276.

Ordahl, C. P. and Le Douarin, N. M. (1992). Two myogenic lineages withinthe developing somite. Development 114, 339-353.

Ott, M.-O., Bober, E., Lyons, G., Arnold, H. and Buckingham, M. (1991).Early expression of the myogenic regulatory gene, myf-5, in precursor cellsof skeletal muscle in the mouse embryo. Development 111, 1097-1107.

Pourquie, O., Coltey, M., Teillet, M.-A., Ordahl, C. and Le Douarin, N. M.(1993). Control of dorsoventral patterning of somitic derivatives bynotochord and floor plate. Proc. Natl. Acad. Sci. USA 90, 5242-5246.

Sasoon, D., Garner, I. and Buckingham, M. E. (1988). Transcripts for a-cardiac and a-skeletal actins are early markers for myogenesis in the mouseembryo. Development 104, 155-164.

Schemainda, H. (1979). Über die Entwicklung der occipitalen Somiten.Experimentelle Untersuchungen an Wachtel- un Hühnerembryonen. Verh.Anat. Ges. 73, 527-532.

Solursh, M., Drake, C. and Meier, S. (1987). The migration of myogenic cellsfrom the somites at the wing level in avian embryos. Dev. Biol. 121, 389-396.

Vogan, K. J., Epstein, D. J., Trasler, D. G. and Gros, P. (1993). The splotch-delayed (Spd) mouse mutant carries a point mutation within the paired box ofthe Pax-3 gene. Genomics 17, 364-369.

Wachtler, F. and Christ, B. (1992). The basic embryology of skeletal muscleformation in vertebrates: the avian model. Seminars in Dev. Biol. 3, 217-227.

Wilkinson, D. G. (1992). In In situ hybridization: A Practical Approach. (ed.D. G. Wilkinson), pp. 75-82, Oxford: IRL Press.

(Accepted 30 November 1993)

E. Bober and others