Digital development and morphogenesis - Wiley Online Library

J. Anat. (2003) 202, pp51–58

Digital development and morphogenesis Blackwell Science, Ltd

J. J. Sanz-Ezquerro1 and C. Tickle2 1

Departamento de Inmunología y Oncología, Centro Nacional de Biotecnología, Cantoblanco UAM, 28049 Madrid, Spain Division of Cell & Developmental Biology, School of Life Sciences, MSI / WTB Complex, University of Dundee, Dow Street, Dundee DD1 5EH, UK 2

Abstract Signalling interactions between the polarizing region, which produces SHH, and the apical ectodermal ridge, which produces FGFs, are essential for outgrowth and patterning of vertebrate limbs. However, mechanisms that mediate translation of early positional information of cells into anatomy remain largely unknown. In particular, the molecular and cellular basis of digit morphogenesis are not fully understood, either in terms of the formation of the different digits along the antero-posterior axis or in the way digits stop growing once pattern formation has been completed. Here we will review recent data about digit development. Manipulation of morphogenetic signals during digit formation, including application of SHH interdigitally, has shown that digit primordia possess a certain plasticity, and that digit anatomy becomes irreversibly fixed during morphogenesis. The process of generation of joints and thus segmentation and formation of digit tips is also discussed. Key words chick; embryo; joint; limb; spacing mechanisms.

Introduction One of the major challenges in developmental biology is to understand how detailed anatomy is generated. The limb contains more than 50 named muscles, precisely shaped individual skeletal elements and tendons with specific attachment sites to link muscles and skeleton. The limb is also supplied with blood vessels and nerves and these elements have a stereotypical pattern. Here we will review recent work on morphogenesis of the digits, which also involves such fundamental processes as spacing and segmentation.

Stages in digit morphogenesis The different parts of the limb along the proximo-distal axis are laid down in sequence as the limb bud grows out. Bud outgrowth is mediated by signals from the apical ectodermal ridge, the thickened epithelium that rims the tip of the limb bud and expresses Fgf

Correspondence Dr C. Tickle, Division of Cell & Developmental Biology, School of Life Sciences, MSI/WTB Complex, University of Dundee, Dow Street, Dundee DD1 5EH, UK. E-mail: [email protected] Accepted for publication 6 November 2002

© Anatomical Society of Great Britain and Ireland 2003

genes (Saunders, 1948; Summerbell, 1974; Martin, 1998; see also Lonai, this issue). Sequential formation of proximo-distal structures is particularly clearly seen with respect to the development of the limb skeleton which is initially laid down in cartilage. Chondrogenesis can be monitored by deposition of matrix which can be seen using alcian blue staining or incorporation of S35 (Hinchliffe, 1977). A single chondrogenic element forms in the proximal part of the limb which will develop into humerus. Slightly later, a Y-shaped condensation appears – the arms of the Y represent forerunners of radius/ulna or tibia/fibula. Then wrist/ankle elements can be seen, and finally digits. In the chick wing, separate condensations that will form each of the three digits emerge between stages 27 and 28 (Hamilton–Hamburger stages; 5–6 days of development) in a posterior to anterior sequence (Hinchliffe, 1977; Fig. 1). These condensations initially form a series of rays within the hand plate and are joined by soft tissue. In the chick leg, condensations that will form each of the four toes arise at around stage 27, again in a posterior to anterior sequence (Fig. 1). Digital rays start off as continuous rods of cartilage that elongate and periodically segment to form interphalangeal joints and thus generate a precise number of phalanges (Fig. 2).

52 Digit morphogenesis, J. J. Sanz-Ezquerro and C. Tickle

Fig. 1 Alcian Blue-stained chick limb buds at early stages in digit formation. First metacarpo/metatarso-phalangeal joints (asterisks) and digit / toe condensations (arrows) appear in the most posterior rays. R, radius; U, ulna; mc, metacarpal bone; dc, digital condensation; T, tibia; F, fibula; mt, metatarsal bone; tc, toe condensation.

Establishing digit number and identity Both digit number and digit identity (thumb vs. little finger/big toe vs. little toe) are controlled by signalling from the polarizing region (see Panman and Zeller, this issue). The polarizing region is a small group of

mesenchyme cells at the posterior margin of the bud. A mirror image duplication of digits can be induced by grafting a polarizing region from one chick limb bud to the anterior margin of another early bud (Saunders & Gasseling, 1968). In the wing, this manipulation produces a complete set of additional digits (423 in mirror image with the normal set, 234) when carried out at stage 22 or earlier (Summerbell, 1974). Given that a polarizing graft takes about 18 h to produce full duplications in the wing (data for stage 20; Smith, 1980), this suggests that the number of digital rays must be set around 18 h after stage 22, i.e. around stage 25. Stage 25 is the latest stage at which posterior wing tissue can elicit full digit duplications when grafted to the anterior margin of an early wing bud (Honig & Summerbell, 1985). When the graft is made later, branching of digits is produced (Summerbell, 1974). Digit identity depends on distance from the polarizing region, the most posterior digit forming next to the polarizing region; the most anterior furthest away. Reducing polarizing strength (for example, by irradiating the polarizing region before grafting or grafting a smaller number of polarizing region cells) has the same effect as reducing length of exposure to the signal (for example, removing the polarizing region graft after a short length of time); only additional anterior digits are specified (Smith et al. 1978; Smith, 1980; Tickle, 1981). This can be understood in terms of a model in which signalling by the polarizing region first induces cells next to the graft to form an anterior digit and then promotes this anterior digit into a posterior digit and

Fig. 2 Series of Alcian Blue-stained chick leg buds showing sequential appearance of joints and phalanges in toes. Continuous distal cartilage condensations become segmented by generation of a joint. From stage 29 to stage 34 represents about 3 days of development. © Anatomical Society of Great Britain and Ireland 2003

Digit morphogenesis, J. J. Sanz-Ezquerro and C. Tickle 53

specifies a new anterior digit further away (Tickle, 1995). This could be envisaged as being due to a local increase in signal strength together with further spread of the signal away from the graft. Digit number is related to the width of the bud and this depends on the length of the apical ectodermal ridge (Lee & Tickle, 1985; Brickell & Tickle, 1989). After polarizing region grafts, the limb bud widens and the apical ridge is maintained over the anterior part of the wing bud. In normal chick limb development, cell marking experiments have shown that cells in the anterior part of the apical ridge leave the ridge and form non-ridge ectoderm as the bud grows out (Vargesson et al. 1997). Cells in the polarizing region express Shh and there is evidence that both digit number and identity depend on SHH signalling. In SHH–/– mutant mouse embryos, only a single rudimentary digit forms and it has been suggested that this is an ‘anterior’ digit (Chiang et al. 2001; Kraus et al. 2001). This fits with the idea that development of a proper hand plate with a series of digits and progressive posteriorization of digit identity depends on SHH. The effects of SHH seem to be mediated by downstream signals such as BMPs and also Gremlin, a BMP antagonist (reviewed Sanz-Ezquerro & Tickle, 2001; see also Panman & Zeller, this issue). Very recently, it has been shown that an important function of SHH signalling is to prevent processing of Gli3 protein to the repressor form in the posterior part of the limb bud where digits will form. Gli3 is expressed widely in the limb bud, particularly in the anterior region, and in the absence of SHH, the repressive form of Gli3 protein predominates and further distal development of almost the entire limb bud is shut down (Litingtung et al. 2002; Welscher et al. 2002). Gremlin appears to be necessary to allow expression of Fgf4 in the posterior region of the apical ridge (Zuniga et al. 1999). There is evidence for a positive feedback loop involving FGF4 (and probably FGF9, FGF17) that ensures continued expression of Shh in the polarizing region, and SHH, in turn ensures continued expression of Fgf4 and these other Fgf genes in posterior ridge (Laufer et al. 1994; Niswander et al. 1994). Thus Gremlin appears to be the apical ridge maintenance factor that was hypothesized to be produced by the polarizing region, in addition to the patterning signal (reviewed in Saunders, 1977). Fate mapping experiments have shown that the digits come from the posterior part of the limb bud under the region of the ridge expressing Fgf4 (Vargesson et al. © Anatomical Society of Great Britain and Ireland 2003

1997). Fgf4 expression in the chick wing bud persists at least until around stage 26 (Duprez et al. 1996), just after the stage at which the number of digital rays appears to be fixed. Shh expression seems to switch off at about the same time (stage 26–27, Dahn & Fallon, 2000; J.J.S.E., unpublished observations), although some reports have suggested that expression persists until later stages (stage 29, Riddle et al. 1993). The signalling molecules that pattern the chick wing and leg are the same. A major step forward is the identification of Tbx genes, Tbx5 and Tbx4, which are expressed specifically in upper and lower limbs, respectively. How exactly Tbx gene expression contributes to the patterning process is not understood (reviewed by Niswander, 1999).

Digit morphogenesis Recent work has shown that, surprisingly, development of the digital rays is relatively plastic (Dahn & Fallon, 2000). According to the model for polarizing region signalling outlined earlier, each of the digital rays will develop from cells with a particular antero-posterior identity and this identity should then determine the subsequent morphogenesis of that particular ray, for example number, relative length and shape of phalanges. In a series of experiments using barriers to bisect digital primordia, removing interdigital mesenchyme and grafting digital primordia to heterotopic interdigital host environments, Dahn and Fallon showed that morphogenesis of rays can be modified by adjacent interdigital mesenchyme and that the rays develop in accordance with the most posterior interdigital cues received. In the chick leg, each toe has a different number of phalanges (going from anterior to posterior, toe 1, 2 phalanges; toe 2, 3; toe 3, 4; toe 4, 5; see Fig. 3). Digital primordia transplantations can lead both to gain and loss of phalanges in digital ray development, and thus, as Dahn and Fallon point out, digital rays can be anteriorized as well as posteriorized. This contrasts with the earlier limb bud antero-posterior patterning process in which specification always seems to proceed from anterior to posterior, i.e. posteriorization. Toes with extra phalanges can also be produced by grafting beads soaked in SHH to interdigital tissue (Dahn & Fallon, 2000). This was also discovered independently in a series of experiments designed to examine the effects of SHH on interdigital cell death (Sanz-Ezquerro & Tickle, 2000). Figure 4 shows an example in which toe


with an endogenous signalling system that controls morphogenesis of each individual digital ray, possibly based on Ihh. The finding that mutations in Ihh are present in patients with brachydactyly type A-1 (characterized by shortening or absence of middle phalanges) supports this idea (Gao et al. 2001; see also Wilkie, this issue).

Separation and spacing of digits

Fig. 3 Stage 36 (10 days incubation) chick embryo leg stained for cartilage with Alcian Blue. Completed toes can be seen with phalanges, joints and tips, which will bear claws. Numbers refer to identity of toes, from anterior (1 = big toe) to posterior (4). Note increasing number of phalanges in more posterior toes.

Fig. 4 Application of beads soaked in SHH to the interdigital spaces leads to generation of longer digits with extra phalanges (arrow) and joint (arrowhead). Asterisk marks position of SHH-soaked bead.

2 developed an additional phalange. It should be noted that, at the time when these experiments are carried out, stage 28, Shh is no longer expressed. One interpretation of these effects of SHH is that this is interfering

The mechanisms involved in setting up digital vs. interdigital areas and thus spacing the digits are not understood. The initial divergence between digital and interdigital regions determines the fate of cells in an alternating fashion, and different programmes of cell differentiation (chondrogenesis or apoptosis, respectively) will be activated. Members of the TGFβ superfamily of signalling molecules have been implicated as the signals responsible for executing the two different programmes – TGFβs as chondrogenic signals and BMPs as apoptotic signals (Ganan, 1996). However, interdigital tissue has a high chondrogenic potential, both in vitro, when assayed in micromass or in organ explant cultures, and even in vivo. Removal of dorsal ectoderm or interdigital apical ectodermal ridge leads to chondrogenesis, with sometimes the development of a reasonably formed extra digit (Hurle & Ganan, 1986). This can also be produced by application to interdigital mesenchyme of several signalling molecules, including TGFβ and with low frequency, SHH. Figure 5 shows such an extra digit with phalanges, joint and tip. This has led to the proposal that chondrogenic potential of interdigital cells before the onset of cell death must be actively repressed. Negative ectodermal influences through FGF-FGFR2 signalling (Moftah et al. 2002) and of neighbouring digits, perhaps also involving retinoic acid signalling (Lee et al. 1994), have been proposed. This potential of interdigital tissue to form cartilage could account for the reported observation that interdigital cells contribute to digit condensations by cell migration (Omi et al. 2000). Some kind of lateral inhibition mechanism could also be operating during digital condensation, and expression of members of the Notch signalling pathway in the edges of digital rays from the beginning of condensation is suggestive of a role in such a process (Vargesson et al. 1998). An autoregulatory loop of chondrogenic activators and diffusible inhibitors, on the lines of diffusion–reaction models, © Anatomical Society of Great Britain and Ireland 2003


(Chautan et al. 1999) and could account for at least part of interdigital death. Recently, a direct regulation by a homeobox transcription factor of genes involved in activating cell death in a morphogenetic context (head formation in Drosophila) has been reported (Lohmann et al. 2002). It will be of interest to investigate the possibility that a similar process operates in vertebrate embryos during interdigital cell death, and the fact that Hoxa13 mutant mice have syndactyly is very suggestive in this context (Stadler et al. 2001). In contrast, however, in Hypodactyly mouse mutants in which Hoxa13 mutations have a dominant negative effect (Post et al. 2000), there is extensive cell death in the digital plate and only one digit forms (Robertson et al. 1996).

Segment and joint formation

Fig. 5 Ectopic extra digit (arrow) induced in the interdigital space of a chick embryo leg bud by application of sn SHHsoaked bead (asterisk). Note that extra digit contains a distal phalange and tip.

has also been proposed but, to date, not directly demonstrated to occur in vivo. Another important process during digit morphogenesis is cell death, which helps sculpt the autopod by freeing digits. Interdigital cell death has been shown to occur mainly by caspase-dependent apoptosis. TUNELpositive cells can be readily detected during interdigital death and mice null for molecules involved in the execution (Apaf-1 –/–) (Cecconi et al. 1998) or regulation (double bax –/–, bak –/–) (Lindsten et al. 2000) of apoptosis have soft-tissue syndactyly due to lack of interdigital cell death. However, a necrotic caspaseindependent mechanism has also been reported © Anatomical Society of Great Britain and Ireland 2003

Morphological observations support the model in which an initially continuous cartilage condensation becomes divided successively into distinct segments as a result of joint formation (Fig. 2; reviewed in Francis-West et al. 1999b). This happens through the appearance first of the interzone, a region of higher cell density where chondrogenesis is repressed. These interzone cells stop expressing characteristic cartilage markers, such as type II collagen, and this region will eventually cavitate to form the joint cavity and surrounding capsule. However, the mechanisms involved both in specifying a joint area and in spacing of joints are not understood. Several genes are expressed in joints during digit morphogenesis, including those encoding signalling molecules such as BMP2, their secreted antagonists such as Chordin or transcription factors such as Cux1 (Lizarraga et al. 2002). However, two genes have been more directly implicated in joint formation. Gdf5, a gene member of the TBFβ superfamily, is specifically expressed in developing joints (Fig. 6), and null or lossof-function mutations in this gene result in the absence or aberrant formation of joints both in mice and in humans (Polinkovsky et al. 1997; Storm & Kingsley, 1999). However, the effect of ectopic application of GDF5 is not the induction of a joint, but rather inhibition of chondrogenesis (Francis-West et al. 1999a; Merino et al. 1999; Storm & Kingsley, 1999), likely placing the function of this gene downstream of joint specification. So far, the only gene suggested to have an inductive role in joint formation is the gene encoding WNT-14, another secreted molecule but belonging to


Fig. 6 Stage 35 chick leg bud showing expression (arrow) of Gdf5 by wholemount in situ hybridization. Note intense staining of all interphalangeal joints as well as metacarpo-phalangeal joints.

the WNT family. Wnt-14 is expressed at sites of joint formation (Hartmann & Tabin, 2001), and when overexpressed is able to inhibit chondrogenesis and induce changes, both morphological (gaps in chondrogenic condensation) and molecular (activation of joint markers including Gdf5), characteristic of joint formation. Also in the context of these experiments, the observation that such ectopic joint-like regions are able to suppress formation of adjacent endogenous joints has led to the proposal that an auto-inhibition process could account for spacing of joints. Once a joint has been specified, among its products would be a secreted inhibitor that would prevent, above a certain threshold, the induction of a new interzone in the vicinity. Growth of the condensation would lead to loss of the negative influence and then positive signals for joint induction could operate again to form a new joint. Positive influences of BMPs from the interdigital areas, and negative influences of the BMP inhibitor Chordin from joints have also been suggested to play a role.

Another aspect of joint development is cell death, which occurs during later differentiation of joints. Although the occurrence of apoptosis has been reported in connection with cavitation, its functional significance has yet to be investigated.

Tip formation After pattern formation has been completed, limb buds stop growing out. However, the molecular mechanisms involved in the cessation of outgrowth and tip formation are not known. Heterochronic grafts between mesenchyme and ectodermal hulls have shown that the signal that primarily maintains outgrowth resides in the mesenchyme (Rubin & Saunders, 1972). Since it is now known that apical ridge signalling is mediated by FGFs, and mesenchyme acts by maintaining apical ridge signalling, the problem then is to control how long expression of FGFs is maintained in the ridge. This process may be related to earlier antero-posterior patterning, © Anatomical Society of Great Britain and Ireland 2003


in that disappearance of Fgf8 expression from the apical ridge in chick embryo leg buds occurs at different times in different digital rays. Fgf8 switches off first from anterior digit tips and later from more posterior digits (Merino et al. 1998). Interestingly, the time at which Fgf8 expression switches off correlates not with absolute digit length but rather with the number of phalanges that will form, and thus is related to digit identity. Digit tips have special regenerative powers. Higher vertebrates have pretty much lost the regenerative potential characteristic of lower animals, such as urodeles, and, in the case of limbs, is restricted to digit tips. This regenerative potential has been correlated with the expression domain of Msx1 gene in that region (Reginelli et al. 1995). Understanding the process of tip determination-formation could lead to improvements in regenerative ability. Finally, the control of digit growth and tip formation could represent a mechanism leading to evolutionary morphological diversity. A delay in timing the end of digit formation could allow for generation of longer digits, by maintaining the process of digit growth and segmentation. This could have been the case for the hyperphalangic flippers of dolphins and whales (Richardson & Oelschläger, in press).

Acknowledgments Cheryll Tickle is a Royal Society Professor and the authors’ research on limb development is supported by the MRC.

References Brickell PM, Tickle C (1989) Morphogens in chick limb development. Bioessays 11, 145 –149. Cecconi F, Alvarez-Bolado G, Meyer BI, Roth KA, Gruss P (1998) Apaf1 (CED-4 homolog) regulates programmed cell death in mammalian development. Cell 94, 727– 737. Chautan M, Chazal G, Cecconi F, Gruss P, Golstein P (1999) Interdigital cell death can occur through a necrotic and caspase-independent pathway. Curr. Biol. 9, 967– 970. Chiang C, Litingtung Y, Harris MP, Simandl BK, Li Y, Beachy PA, et al. (2001) Manifestation of the limb prepattern: limb development in the absence of sonic hedgehog function. Dev. Biol. 236, 421– 435. Dahn RD, Fallon JF (2000) Interdigital regulation of digit identity and homeotic transformation by modulated BMP signaling. Science 289, 438 – 441. Duprez D, Kostakopoulou K, Francis-West PH, Tickle C, Brickell PM (1996) Activation of Fgf-4 and HoxD gene expression by © Anatomical Society of Great Britain and Ireland 2003

BMP-2 expressing cells in the developing chick limb. Development 122, 1821–1828. Francis-West PH, Abdelfattah A, Chen P, Allen C, Parish J, Ladher R, et al. (1999a) Mechanisms of GDF-5 action during skeletal development. Development 126, 1305 –1315. Francis-West PH, Parish J, Lee K, Archer CW (1999b) BMP/GDF– signalling interactions during synovial joint development. Cell Tissue Res. 296, 111–119. Ganan Y (1996) Role of TGF betas and BMPs as signals controlling the position of the digits and the areas of interdigital cell death in the developing chick limb autopod. Development 122, 2349 – 2357. Gao B, Guo J, She C, Shu A, Yang M, Tan Z, et al. (2001) Mutations in IHH, encoding Indian hedgehog, cause brachydactyly type A-1. Nat. Genet. 28, 386 – 388. Hartmann C, Tabin CJ (2001) Wnt-14 plays a pivotal role in inducing synovial joint formation in the developing appendicular skeleton. Cell 104, 341– 351. Hinchliffe J (1977) The Chrondogenic Pattern in Chick limb morphogenesis: a problem of development and evolution. In The Third Symposium of the British Society for Developmental Biology (eds Ede D, Hinchcliffe JR, Balls M), pp. 293– 310. Cambridge: Cambridge University Press. Honig L, Summerbell D (1985) Maps of strength of positional signalling activity in the developing chick wing bud. J. Embryol. Exp. Morph. 87, 163 –174. Hurle JM, Ganan Y (1986) Interdigital tissue chondrogenesis induced by surgical removal of the ectoderm in the embryonic chick leg bud. J. Embryol. Exp. Morph. 94, 231– 244. Kraus P, Fraidenraich D, Loomis CA (2001) Some distal limb structures develop in mice lacking Sonic hedgehog signaling. Mech. Dev. 100, 45 – 58. Laufer E, Nelson CE, Johnson RL, Morgan BA, Tabin C (1994) Sonic hedgehog and Fgf-4 act through a signaling cascade and feedback loop to integrate growth and patterning of the developing limb bud. Cell 79, 993 –1003. Lee KK, Li FC, Yung WT, King JL, Ng JL, Cheah CS (1994) Influence of digits, ectoderm and retinoic acid on chondrogenesis by mouse interdigital mesoderm in culture. Dev. Dyn. 201, 297– 309. Lee J, Tickle C (1985) Retinoic acid and pattern formation in the developing chick wing. SEM and quantitative studies of early effects on the apical ectodermal ridge and bud outgrowth. J. Embryol. Exp. Morph. 90, 139 –169. Lindsten T, Ross AJ, King A, Zong WX, Rathmell JC, Shiels HA, et al. (2000) The combined functions of proapoptic Bc1–2 family members bak and bax are essential for normal development of multiple tissues. Mol. Cell 6, 1389 –1399. Litingtung Y, Dahn RD, Li Y, Fallon JF, Chiang C (2002) Shh and Gli3 are dispensable for limb skeleton formation but regulate digit number and identity. Nature 418, 979 – 983. Lizarraga G, Lichtler A, Upholt WB, Kosher RA (2002) Studies on the role of Cux1 in regulation of the onset of joint formation in the developing limb. Dev. Biol. 243, 44 –54. Lohmann I, McGinnis N, Bodmer M, McGinnis W (2002) The Drosophila Hox gene deformed sculpts head morphology via direct regulation of the apoptosis activator reaper. Cell 110, 457– 466. Martin GR (1998) The roles of FGFs in early development of vertebrate limbs. Genes Dev. 12, 1571–1586.


Merino R, Ganan Y, Macias D, Economides AN, Sampath KT, Hurle JM (1998) Morphogenesis of digits in the avian limb is controlled by FGFs, TGFbetas, and noggin through BMP signaling. Dev. Biol. 200, 35 – 45. Merino R, Macias D, Ganan Y, Economides AN, Wang X, Wu Q, et al. (1999) Expression and function of Gdf-5 during digit skeletogenesis in the embryonic chick leg bud. Dev. Biol. 206, 33–45. Moftah M, Downie S, Bronstein N, Mezentseva N, Pu J, Maher P, et al. (2002) Ectodermal FGFs induce perinodular inhibition of limb chondrogenesis in vitro and in vivo via FGF Receptor 2. Dev. Biol. 249, 270. Niswander L, Jeffrey S, Martin G, Tickle C (1994) A positive feedback loop coordinates growth and patterning in the limb. Nature 371, 609 – 612. Niswander L (1999) Developmental biology. Legs to wings and back again. Nature 398, 751– 752. Omi M, Sato-Maeda M, Ide H (2000) Role of chondrogenic tissue in programmed cell death and BMP expression in chick limb buds. Int. J. Dev. Biol. 44, 381– 388. Polinkovsky A, Robin NH, Thomas JT, Irons M, Lynn A, Goodman FR, et al. (1997) Mutations in CDMP1 cause autosomal dominant brachydactyly type C. Nat. Genet. 17, 18–19. Post JC, Margulies EH, Kuo A, Innis JW (2000) Severe limb defects in Hypodactyly mice result from the expression of a novel, mutant HOXA13 protein. Dev. Biol. 217, 290 – 300. Reginelli AD, Wang YQ, Sassoon D, Muneoka K (1995) Digit tip regeneration correlates with regions of Msx1 (Hox 7) expression in fetal and newborn mice. Development 121, 1065–1076. Richardson MK, Oelschläger HHA (in press) Time, pattern and heterochrony: a stud of hyperphalangy in the dolphin embryo flipper. Evol. Dev. in press. Riddle RD, Johnson RL, Laufer E, Tabin C (1993) Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75, 1401– 1416. Robertson KE, Chapman MH, Adams A, Tickle C, Darling SM (1996) Cellular analysis of limb development in the mouse mutant hypodactyly. Dev. Genet. 19, 9 – 25. Rubin L, Saunders JW Jr (1972) Ectodermal–mesodermal interactions in the growth of limb buds in the chick embryo: constancy and temporal limits of the ectodermal induction. Dev. Biol. 28, 94 –112. Sanz-Ezquerro JJ, Tickle C (2000) Autoregulation of Shh expression and Shh induction of cell death suggest a mechanism for modulating polarising activity during chick limb development. Development 127, 4811– 4823.

Sanz-Ezquerro JJ, Tickle C (2001) ‘Fingering’ the vertebrate limb. Differentiation 69, 91– 99. Saunders JW (1948) The proximo-distal sequence opf origin of the parts of the chick wing and the role of the ectoderm. J. Exp. Zool. 108, 363 – 404. Saunders JW (1977) The experimental analysis of chick limb bud development. In Vertebrate Limb and Somite Morphogenesis (eds Ede DA, Balls, MJ), pp. 1–24. Cambridge: Cambridge University Press. Saunders JW, Gasseling MT (1968) Ectodermal–mesenchymal interactions in the origin of limb symmetry. In Epithelial– Mesenchymal Interactions (eds Fleischmeyer R, Billingham RE), pp. 78 – 97. Baltimore: Williams & Wilkins. Smith JC, Tickle C, Wolpert L (1978) Attenuation of positional signalling in the chicklimb by high doses of gamma radiation. Nature 272, 612– 613. Smith J (1980) The time required for positional signalling in the chick wing bud. J. Embryol. Exp. Morph. 60, 321–328. Stadler H, Higgins KM, Capecchi MR (2001) Loss of Ephreceptor expression correlates with loss of cell adhesion and chondrogenic capacity in Hoxa13 mutant limbs. Development 128, 4177– 4188. Storm EE, Kingsley DM (1999) GDF5 coordinates bone and joint formation during digit development. Dev. Biol. 209, 11– 27. Summerbell D (1974) Interaction between the proximo-distal and antero-posterior coordinates of positional value during the specification of positional information in the early development of the chick limb bud. J. Embryol. Exp. Morph. 32, 227– 237. Tickle C (1981) The number of polarizing region cells required to specify additional digits in the developing chick wing. Nature 289, 295 – 298. Tickle C (1995) Vertebrate limb development. Curr. Opin. Genet. Dev. 5, 478 – 484. Vargesson N, Clarke JDW, Vincent K, Coles C, Wolpert L, Tickle C (1997) Cell fate in the chick limb bud and relationship to gene expression. Development 124, 1909 –1918. Vargesson N, Patel K, Lewis J, Tickle C (1998) Expression patterns of Notch1, Serrate1, Serrate2 and Delta1 in tissues of the developing chick limb. Mech. Dev. 77, 197–199. te Welscher P, Zuniga A, Kuijper S, Drenth T, Goedemans HJ, Meijlink E (2002) Progression of vertebrate limb development through SHH-mediated counteraction of GL13. Science 298, 827– 830. Zuniga A, Haramis AP, McMahon AP, Zeller R (1999) Signal relay by BMP antagonism controls the SHH / FGF4 feedback loop in vertebrate limb buds. Nature 401, 598 – 602.

© Anatomical Society of Great Britain and Ireland 2003