Dlx-5 in Limb Initiation in the Chick EmbryoDEBORAH FERRARI, AMANDA HARRINGTON, CAROLINE N. DEALY, AND ROBERT A. KOSHER*Department of Genetics and Developmental Biology, University of Connecticut Health Center, Farmington, Connecticut

ABSTRACT Dlx-5 is a vertebrate homolog ofthe Drosophila Distal-less gene, one of the firstgenetic signals for limb formation in the fly. Inthe present study we have explored the possiblerole of Dlx-5 in limb initiation in the chick em-bryo. At stage 14 which is well before the initialformation of limb buds Dlx-5 is highly and specifi-cally expressed in the ectoderm of the presump-tive wing and leg forming regions of the lateralplate, but not in the intervening non-limb form-ing prospective flank. Thus, Dlx-5 expression dis-tinguishes the limb-forming territories prior tolimb budding, and is one of the first molecularmarkers of vertebrate limb initiation. Further-more, Dlx-5 expression is induced in the non-limb-forming flank within 12 hours after implantationof an FGF2-soaked bead, a procedure that resultsin the induction of an ectopic limb. The rapidinduction of Dlx-5 expression in response to asignal which ultimately leads to supernumerarylimb formation is consistent with a role for Dlx-5in limb initiation. We have also examined theexpression of Dlx-5 in the limb buds of ameliclimbless mutant chick embryos, which undergonormal limb formation but do not form an AERand thus fail to undergo further outgrowth. Dlx-5is transiently expressed by the ectoderm of emer-gent limbless limb buds, consistent with a role forDlx-5 in limb initiation. Together, our resultssuggest that Dlx-5 may be involved in the specifi-cation of the limb territories of the lateral plate,and in the initial formation of the limb bud fromthese regions. Dev Dyn 1999;216:10–15.r 1999 Wiley-Liss, Inc.

Key words: Dlx-5; limb formation; FGF; AER

INTRODUCTION

Dlx-5 is a homeobox-containing gene that is a verte-brate homolog of the Drosophila Distal- less gene. InDrosophila, Distal-less expression distinguishes thepresumptive limb-forming imaginal disc cells from thecells of the surrounding body wall (Cohen et al., 1989),and thus is one of the earliest genetic markers of flylimb formation. Loss of Distal-less expression results inloss or abnormal morphogenesis of the limbs (Cohenand Jurgens, 1989), and ectopic Distal- less expressioncauses supernumerary limb formation (Cohen et al.,1991; Simcox et al., 1991; Gorfinkiel et al., 1997).Multiple vertebrate Distal-less homologs, the Dlx genes,have been isolated from human, mouse, chicken, ze-

brafish and axolotl which are expressed during embryo-genesis in regions characterized by epithelial-mesenchy-mal interactions including the developing limbs (Dolleet al., 1992; Bulfone et al., 1993; Akimenko et al., 1994;Robinson and Mahon, 1994; Simeone et al., 1994; Zhaoet al., 1994; Ferrari et al., 1995; Mullen et al., 1996;Nakamura et al., 1996; Ellies et al., 1997). On the basisof the conserved expression of Distal- less and Dlx genesin the developing appendages of animals from widelydivergent phylogenetic taxa (Panganiban, 1995), it hasbeen suggested that Distal-less/Dlx genes may playfundamental and universal roles in animal appendageformation (Panganiban et al., 1997).

The present study was undertaken to explore apossible role for Dlx-5 in initial limb formation in thechick embryo. Our results indicate that Dlx-5 is one ofthe earliest genetic markers of limb initiation in thechick embryo, as it is in the fly, and suggest that Dlx-5may be involved in the establishment of the limbforming territories and in the initial formation of thelimb bud from these regions.

RESULTSDlx-5 Expression Becomes Restricted to thePresumptive Limb Territories BeforeLimb Budding

To determine if Dlx-5 is expressed in a temporal andspatial fashion consistent with a role in limb initiation,we examined its expression in the early chick embryoduring the period in which the limb-forming territoriesare being specified prior to limb budding (Fig. 1).Between stages 10–12 Dlx-5 is expressed throughoutthe surface ectoderm of the trunk of the embryo,including that overlying the somitic, intermediate, andlateral plate mesoderm (Fig. 1A–C). In contrast, be-tween stages 13–15, Dlx-5 expression is upregulated inthe ectoderm overlying the prospective wing- and leg-forming regions of the lateral plate and attenuates inthe ectoderm overlying the non-limb forming prospec-tive flank (Fig. 1D–G). During this same period Dlx-5expression is downregulated in the ectoderm overlyingthe somitic mesoderm. Thus, Dlx-5 expression becomesprogressively restricted to the presumptive limb territo-

Grant sponsor: National Institutes of Health; Grant number:HD22610.

*Correspondence to: Robert A. Kosher, Department of Genetics andDevelopmental Biology, MC3405, University of Connecticut HealthCenter, Farmington, CT 06030. E-mail: kosher@sun.uchc.edu

Received 17 March 1999; Accepted 1 June 1999

DEVELOPMENTAL DYNAMICS 216:10–15 (1999)

r 1999 WILEY-LISS, INC.

ries of the lateral plate well before the initial formationof the limb buds which occurs at stage 16/17.

By stage 16, Dlx-5 expression is most abundant in aregion that encompasses the distal tip and ventralectoderm (Fig. 2). Less Dlx-5 is expressed by dorsalwing ectoderm, and little or no Dlx- 5 is expressed by

the non-limb ectoderm overlying the somites (Fig. 2).The domain of high Dlx- 5 expression at the distal tipcorresponds to the presumptive AER according to thequail-chick chimera fate mapping studies of Michaud etal. (1997), although the dye marking studies of Altabefet al. (1997) indicate that cells that will contribute to

Fig. 1. Expression of Dlx-5 in the trunk of chick embryos at stage 10(A–C) and stages 14 and 15 (D–G). (A–C) At stage 10, Dlx-5 is expressedthroughout the surface ectoderm of the trunk of the embryo, including thatoverlying the somitic (sm), intermediate (im), and lateral plate (lp)mesoderm. (D–G) At stage 14, Dlx-5 expression is upregulated in the

ectoderm overlying the prospective wing- (arrows in D; see also G) andleg- (arrows in D; see also G) forming regions of the lateral plate andattenuates in the ectoderm overlying the non-limb forming prospectiveflank (E; G). By stage 14 Dlx-5 expression is also downregulated in thenon-limb ectoderm overlying the somitic and intermediate mesoderm.

Fig. 2. Correlation of Dlx-5 expression with discrete regions of limbectoderm. (A) Dlx-5 expression in a section through wing-forming regionof stage 16 chick embryo. (B) Fate map of chick wing ectoderm at stage16 adapted from Michaud et al. (1997). Dlx-5 expression is most abundantin a region that encompasses the distal tip and the ventral wing ectodermLess Dlx-5 is expressed by dorsal wing ectoderm, and little or no Dlx-5 is

expressed by the non-limb ectoderm overlying the somites. Although thedomain of high Dlx-5 expression at the distal tip corresponds to thepresumptive AER according to the quail-chicken chimera fate mappingstudies of Michaud et al. (1997), the dye marking studies of Altabef et al.(1997) indicate that cells that will contribute to the AER are initially widelyscattered in the dorsal and ventral ectoderm, as well as at the distal tip.

11Dlx-5 IN LIMB INITIATION

the AER are initially widely scattered in the dorsal andventral ectoderm, as well as at the distal tip. Dlx-5 isnot expressed by limb bud mesoderm at stage 16 (Fig.2A), but is expressed by limb mesoderm at stage 17 andsubsequent early stages of limb development (see Fig.3D and Ferrari et al., 1995).

Dlx-5 Expression Is Rapidly Induced inResponse to a Stimulus Which Results in EctopicLimb Formation

The restriction of Dlx-5 expression to the presump-tive limb-forming regions is consistent with a role forDlx-5 in limb initiation. To explore this possibilityfurther, we examined Dlx-5 expression in response toFGF2-soaked beads implanted into the flank (Fig. 3), aprocedure that results in the formation of supernumer-ary limbs from the normally non-limb forming flank(Cohn et al., 1995). If Dlx-5 is involved in limb initia-tion, induction of Dlx-5 expression would be expected tobe an early response to a signal which leads to ectopiclimb formation.

Indeed, as early as12 hours following implantation ofan FGF2-soaked bead, Dlx-5 is expressed in the flankectoderm, and also in the lateral plate mesodermdirectly adjacent to the bead (Fig. 3B). In contrast, littleor no Dlx-5 is expressed by the lateral plate ectoderm ormesoderm on the unoperated contralateral side (Fig.3B) or by the flank lateral plate of embryos which didnot receive a bead (Fig. 3A). Within 16–18 hours afterbead implantation, Dlx-5 transcripts are abundantlyexpressed in the ectoderm and mesoderm of the ectopiclimb buds emerging from the flank (Fig. 3C).

Dlx-5 Is Transiently Expressed During the InitialFormation of the Limb Buds of Limbless MutantChick Embryos

To further explore the role of Dlx-5 in initial limbformation, we examined its expression during the devel-opment of the wing buds of limbless mutant chickembryos (Fig. 4). Limbless is an autosomal recessivemutation that directly affects limb ectoderm, not meso-derm (Carrington and Fallon, 1988). The limb buds ofamelic limbless mutant chick embryos initiate nor-mally but subsequently fail to form an AER and thusfail to undergo further outgrowth (Carrington andFallon, 1988). Thus, limbless provides an extremely

Fig. 3. Dlx-5 expression in the trunk of chick embryos after implanta-tion of FGF2-soaked beads into the flank at stage 14. (A) Little or no Dlx-5is expressed by the ectoderm of the non-limb forming flank lateral plate ofembryos which did not receive a bead (arrowheads) except at the lateralbody folds. (B) Within 12 hours after implantation of an FGF2-soakedbead (asterisk), Dlx-5 transcripts are detected in the adjacent flankmesoderm (m) and overlying flank ectoderm (arrow). Little or no Dlx-5 isexpressed by the contralateral flank (arrowhead), except in the ectodermof the lateral body folds. (C) Within 17 hours of bead (asterisk) implanta-tion, Dlx-5 transcripts are readily detected in both the mesoderm andectoderm (arrow) of the ectopic limb bud forming from the flank. Note theabsence of Dlx-5 expression in the contralateral flank (arrowhead). (D)Dlx-5 expression in the normal stage 18 forelimb of the embryo in (C).

12 FERRARI ET AL.

useful system for exploring the nature of molecules thatmay or may not be involved either in initial limbformation or in AER activity.

At stage 18, shortly following limb budding, Dlx-5 isexpressed throughout the AER and nonridge ectodermof normal limb buds, and also throughout the ectodermof limbless mutant limb buds (Fig. 4A,C), consistentwith a role for Dlx-5 in limb initiation. However, atstage 21, Dlx-5 expression is maintained in the ecto-derm of normal limb buds, but its expression attenu-ates dramatically in the limbless mutant limb budswhich fail to form an AER (Fig. 4B,D).

DISCUSSION

The results of the present study suggest that Dlx-5may be involved in the establishment of the limbterritories of the lateral plate of the chick embryo, andin the initial formation of the limb bud from theseregions. Firstly, well before the formation of limb budsDlx-5 is highly and specifically expressed in the ecto-derm of the presumptive limb territories of the lateralplate, but not in the presumptive flank. The expressionof Dlx-5 in the prospective limb regions at stage 14precedes that of Wnt-3a and FGF-8, other ectodermallyexpressed genes that have been implicated in limbformation (Kengaku et al., 1998; Crossley et al. 1996;Vogel et al., 1996). Thus, Dlx-5 expression in the chick

embryo, like Distal-less expression in the fly, is one ofthe earliest genetic markers of limb initiation.

Secondly, Dlx-5 expression is rapidly induced afterimplantation of an FGF2-soaked bead into the flank.Thus, as would be expected for a gene involved in limbinitiation, Dlx-5 expression is an early response to asignal that ultimately leads to supernumerary limbformation. It is noteworthy that Dlx-5 expression isinduced as early as 12 hours after FGF application tothe flank, whereas the expression of FGF-8, which hasbeen implicated in early limb formation, is not detect-able until 15–16 hours after FGF bead implantation(Crossley et al., 1996; Ohuchi et al., 1997). Dlx-5expression is induced within 12 hours after FGF beadimplantation not only in flank ectoderm, but also in thelateral plate mesoderm directly adjacent to the bead.Since Dlx-5 is not normally expressed in limb meso-derm until after budding, this ectopic activation ofDlx-5 expression in flank mesoderm likely illustratesthe acute sensitivity of this gene to a stimulus thatrespecifies the flank as limb-forming. It is also notewor-thy that the induction of Dlx-5 after FGF applicationoccurs at about the time that changes in the combinato-rial pattern of expression of Hox9 genes occur inresponse to FGF beads (Cohn et al., 1997). Interest-ingly, the localization of Distal-less expression to theappendage forming regions of the Drosophila embryo is

Fig. 4. Dlx-5 expression in the wing buds of normal (A,B) and limbless(C,D) chick embryos at stage 18 (A,C) and stage 21 (B,D). At stage 18Dlx-5 is expressed in the ectoderm of limbless wing buds (C) at levelscomparable to that in the ectoderm and AER (arrow) of normal wing buds

(A). At stage 21 Dlx-5 expression is maintained at high levels in theectoderm and AER (arrow) of normal wing buds (B), whereas itsexpression attenuates dramatically in limbless wing buds.

13Dlx-5 IN LIMB INITIATION

controlled at least in part by homeotic genes of thebithorax complex (Vachon et al., 1992). It is thustempting to speculate that the localization of Dlx-5 tolimb territories may be regulated by the patterns ofexpression of various Hox genes along the lateral plate.

Thirdly, as would be expected for a gene that plays arole in limb initiation, Dlx-5 is transiently expressed inthe ectoderm during the initial formation of the limbbuds of limbless mutant chick embryos, which initiatelimb formation normally, but fail to form an AER. Thiscontrasts, for example, to FGF8 which is not expressedin limbless mutant limb buds and thus does not play arequisite role in limb initiation.

It is not clear from our studies whether Dlx-5 directsthe establishment of the limb territories or is a rapidresponse to their specification. Empirical evidence todate suggests that the specification of the limb territo-ries may be a mesodermal property at these earlystages (Saunders and Reuss, 1974). In this regard it isnoteworthy that the pattern of expression of Dlx-5 inpresumptive limb ectoderm corresponds quite well withthat of FGF10 in the underlying mesoderm. Like Dlx-5expression in ectoderm, FGF10 is widely expressed inthe mesoderm of the early chick embryo and subse-quently becomes restricted to and upregulated in theprospective limb mesoderm (Ohuchi et al., 1997). Thesetemporal and spatial correlations suggest a possibleregulatory interaction between Dlx-5 and FGF10. In-deed we have found that Dlx-5 expression is induced inthe flank in response to implantation of an FGF-soakedbead. Similarly, FGF-soaked beads implanted into themesoderm of denervated axolotl blastemas induce Dlx-3expression by the regenerating limb ectoderm (Mullenet al., 1996). Like Dlx-5 and FGF10, IGF-I, which caninduce the formation of limb buds from the flank (Dealyand Kosher, 1996) and TGF-a are specifically expressedin presumptive limb regions before limb budding andare excluded from the non limb-forming flank (Streck etal., 1992; Dealy et al., 1998). Thus, limb initiation islikely to involve complex interactions among a varietyof signaling molecules and regulatory genes, and under-standing the roles and relationships among Dlx-5,FGF10, IGF-I, TGF-a, and undoubtedly other genes inlimb initiation will require further investigation.

Before its restriction to and upregulation in theectoderm of the presumptive limb regions, Dlx-5 isexpressed throughout the surface ectoderm of the trunkof the embryo including that overlying the somitic andintermediate mesoderm as well as that overlying thelateral plate. This widespread expression of Dlx-5 inearly embryonic surface ectoderm may reflect a role inthe initial determination of prospective limb ectoderm,since the recent fate mapping studies of Michaud et al.(1997) indicate that before stage 14 the presumptivelimb ectoderm is not strictly limited to the ectodermoverlying the presumptive limb territories, but ratherencompasses most of the surface ectoderm includingthat overlying the somitic and intermediate mesoderm.Consistent with the possibility that Dlx-5 may define or

specify a prospective limb ectoderm morphogeneticfield, before stage 14 the Dlx-5-expressing surfaceectoderm outside of the presumptive limb territoriespossesses the competence to form an AER in response tothe mesoderm of the limb territories (Michaud et al.,1997). The widespread expression of Dlx-5 in earlysurface ectoderm is also consistent with the suggestionthat it may play a more general role in specification ofthe epidermis (Pera et al., 1999). It is noteworthy,however, that between stage 13–15 Dlx-5 expression isdownregulated in the dorsal surface ectoderm overlyingthe somitic mesoderm and becomes restricted to theectoderm of the prospective limb forming regions of thelateral plate, closely coincident with the restriction ofpresumptive limb ectoderm to the limb territories(Michaud et al., 1997).

EXPERIMENTAL PROCEDURESIn Situ Hybridization

In situ hybridization was performed on serially sec-tioned embryos using a 33P-labeled Dlx-5-specific cDNAprobe as previously described (Ferrari et al., 1995).Whole-mount in situ hybridization was performed essen-tially as described by Wilkinson (1992) using a 617 bpdigoxygenin-labeled RNA probe corresponding to the 38end of chicken Dlx-5 cDNA. The probe was generatedfrom the full length Dlx-5 cDNA (Ferrari et al., 1995) inBluescript SK- by digesting with NcoI and transcribingwith T7 RNA polymerase. Some embryos that had beensubjected to whole-mount in situ hybridization weresubsequently serially sectioned after processing andembedding as described by Nieto et al. (1996).

Bead Implantation

Heparin-acrylic beads (Sigma, St. Louis, MO) weresoaked in a 2 mg/ml solution of human recombinantFGF2 (Gibco, Grand Island, NY) at 37°C for 2 hr, andthen implanted into the flank of stage 14 embryos asdescribed (Cohn et al., 1995).

Preparation of Limb Buds From LimblessMutant Chick Embryos

Windows were cut into eggs obtained from matingmembers of limbless heterozygous chickens maintainedat the Storrs Agricultural Experiment Station, Univer-sity of Connecticut, and the right forelimbs of embryoswere removed and individually processed (Coelho et al.,1991). The windows in the shells of the donor embryoswere sealed, and the donor embryos were reincubateduntil the phenotype of their remaining limb buds wasclearly evident. Individual limb buds to be used for insitu hybridization were fixed in Bouin’s solution, pro-cessed to 70% ethanol, and stored in 70% ethanol untiltheir phenotype was established.

ACKNOWLEDGMENTS

This work was supported by National Institutes ofHealth grant HD22610 to R.A.K. and C.N.D.

14 FERRARI ET AL.

REFERENCESAkimenko M-A, Ekker M, Wegner J, Lin W, Westerfield M. 1994.

Combinatorial expression of three zebrafish genes related to distal-less: part of a homeobox gene code for the head. J Neurosci14:3475–3486.

Altabef M, Clarke JDW, Tickle C. 1997. Dorso-ventral ectodermalcompartments and origin of apical ectodermal ridge in developingchick limb. Development 124:4547–4556.

Bulfone A, Kim HJ, Puelles L, Porteus MH, Grippo JF, Rubenstein JL.1993. The mouse Dlx-2 (Tes-1) gene is expressed in spatiallyrestricted domains of the forebrain, face and limbs in midgestationmouse embryos. Mech Dev 40:129–140.

Carrington JL, Fallon JF. 1988. Initial limb budding is independent ofapical ectodermal ridge activity: evidence from a limbless mutant.Development 104:361–367.

Coelho CND, Krabbenhoft K, Upholt WB, Fallon JF, Kosher RA. 1991.Altered expression of the chicken homeobox-containing genes Ghox-7and Ghox-8 in the limb buds of limbless mutant chick embryos.Development 113:1487–1493.

Cohen B, Wimmer EA, Cohen SM. 1991. Early development of leg andwing primordia in the Drosophila embryo. Mech Dev 33:229–240.

Cohen SM, Jurgens G. 1989. Proximal-distal pattern formation inDrosophila: cell autonomous requirement for Distal-less gene activ-ity in limb development. EMBO J. 8:2045–2055.

Cohen SM, Broenner G, Kuttner F, Jurgens G, Jackle H. 1989.Distal-less encodes a homeodomain protein required for limb devel-opment in Drosophila. Nature 338:432–434.

Cohn MJ, Izpisua-Belmonte JC, Abud H, Heath JK, Tickle C. 1995.Fibroblast growth factors induce additional limb development fromthe flank of chick embryos. Cell 80:739–746.

Cohn MJ, Patel K, Krumlauf R, Wilkinson DG, Clarke JDW, Tickle C.1997. Hox9 genes and vertebrate limb specification. Nature 387:97–101.

Crossley PH, Minowada G, MacArthur CA, Martin GR, 1996. Roles forFGF-8 in the induction, initiation, and maintenance of chick limbdevelopment. Cell 84:127–136.

Dealy CN, Kosher RA. 1996. IGF-I and insulin in the acquisition oflimb-forming ability by the embryonic lateral plate. Dev Biol177:291–299.

Dealy CN, Scranton V, Cheng H-C. 1998. Roles of transforming growthfactor-a and epidermal growth factor in chick limb development.Dev Biol 202:43–55.

Dolle P, Price M, Duboule D. 1992. Expression of the murine Dlx-1homeobox gene during facial, ocular and limb development. Differen-tiation 49:93–99.

Ellies DL, Stock DW, Hatch G, Giroux G, Weiss KM, Ekker M. 1997.Relationship between the genomic organization and the overlappingembryonic expression patterns of the zebrafish dlx genes. Genomics45:580–590.

Ferrari D, Sumoy L, Gannon J, Sun H, Brown AMC, Upholt WB,Kosher RA. 1995. The expression pattern of the distal-less homeobox-containing gene Dlx-5 in the developing chick limb bud suggests itsinvolvement in apical ectodermal ridge activity, pattern formation,and cartilage differentiation. Mech Dev 52:257–264.

Gorfinkiel N, Morata G, Guerrero I. 1997. The homeobox geneDistal-less induces ventral appendage development in Drosophila.Genes Dev 11:2259–2271.

Kengaku M, Capdevila J, Rodriguez-Esteban C, De La Pena J,Johnson RL, Belmonte JCI, Tabin CJ. 1998. Distinct WNT path-ways regulating AER formation and dorsoventral polarity in thechick limb bud. Science 280:1274–1277.

Michaud JL, Lapointe F, Le Douarin NM. 1997. The dorsoventralpolarity of the presumptive limb is determined by signals producedby the somites and by the lateral somatopleure. Development124:1453–1463.

Mullen LM, Bryant SV, Torok MA, Blumberg B, Gardiner DM. 1996.Nerve dependency of regeneration: the role of Distal-less and FGFsignaling in amphibian limb regeneration. Development 122:3487–3497.

Nakamura S, Stock DW, Wydner KL, Bollekens JA, Takeshita K,Nagai BM, Chiba S, Kitamura T, Freeland TM, Zhao Z, Minowada J,Lawrence JB, Weiss KM, Ruddle FH. 1996. Genomic analysis of anew mammalian distal-less gene, Dlx7. Genomics 38:314–324.

Nieto MA, Patell K, Wilkinson DG. 1996. In situ hybridizationanalysis of chick embryos in whole mount and tissue sections. In:Bronner-Fraser M, editor. Methods in Cell Biology, Vol. 5. San Diego:Academic Press, Inc. p 227–228

Ohuchi H, Nakagawa T, Yamamoto A, Araga A, Ohata T, Ishimaru Y,Yoshioka H, Kuwana T, Nohno T, Yamasaki M, Itoh N, Noji S. 1997.The mesenchymal factor, FGF10, initiates and maintains the out-growth of the chick limb bud through interaction with FGF8, anapical ectodermal factor. Development 124:2235–2244.

Panganiban G, Sebring A, Nagy L, Carroll S. 1995. The development ofcrustacean limbs and the evolution of arthropods. Science 270:1363–1366.

Panganiban G, Irvine SM, Lowe C, Roehl H, Corley LS, Sherbon B,Grenier JK, Fallon JF, Kimble J, Wlaker M, Wray GA, Swalla BJ,Martindale MQ, Carroll SB. 1997. The origin and evolution ofanimal appendages. Proc Natl Acad Sci USA 94:5162–5166.

Pera E, Stein S, Kessel M. 1999. Ectodermal patterning in the avianembryo: epidermis versus neural plate. Development 126:63–73.

Robinson GW, Mahon KA. 1994. Differential and overlapping expres-sion domains of Dlx-2 and Dlx-3 suggest distinct roles for Distal-lesshomeobox genes in craniofacial development. Mech Dev 48:199–215.

Saunders JW Jr, Reuss C. 1974. Inductive and axial properties ofprospective wing-bud mesoderm in the chick embryo. Dev Biol38:41–50.

Simcox AA, Hersperger E, Shearn A, Whitle JRS, Cohen SM. 1991.Establishment of imaginal discs and histoblast nests in Drosophila.Mech Dev 34:11–20.

Simeone A, Acampora D, Pannese M, D’Esposito M, Stornaiuolo A,Gulisano M, Mallamaci A, Kastury K, Druck T, Huebner K, Bon-cinelli E. 1994. Cloning and characterization of two members of thevertebrate DLX gene family. Proc Natl Acad Sci USA 91:2250–2254.

Streck RD, Wood TL, Hsu S, Pintar JE. 1992. Insulin-like growthfactor-I and -II and insulin-like growth factor binding protein-2 RNAare expressed in adjacent tissues within rat embryonic and fetallimbs. Dev Biol 151:586–596.

Vachon G, Cohen B, Pfeifle C, McGuffin ME, Botas J, Cohen SM. 1992.Homeotic genes of the bithorax complex repress limb developmentin the abdomen of the drosophila embryo through target genedistal-less. Cell 71:437–450.

Vogel A, Rodriguez C, Warnken W, Izpisua-Belmonte J-C. 1996.Involvement of FGF-8 in initiation, outgrowth and patterning of thevertebrate limb. Development 122:1737–1750.

Wilkinson DG. 1992. Whole mount in situ hybridization of vertebrateembryos. In: Wilkinson DG, editor. In situ hybridization: a practicalapproach. Oxford: IRL Press. p 75–83.

Zhao G-Q, Zhao S, Zhao X, Eberspaicher H, Solursh M, CrombruggheBD. 1994. rDlx, a novel distal-less-like homeoprotein is expressedin developing cartilages and discrete neuronal tissues. Dev Biol 164:37–51.

15Dlx-5 IN LIMB INITIATION