SEARCH

SEARCH BY CITATION

Keywords:

  • En-1;
  • dorsoventral axis;
  • Lmx1;
  • keratin;
  • scales;
  • Shh;
  • skin;
  • chick;
  • Wnt-7a

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The dorsal and ventral scales of the chick foot can be distinguished morphologically and molecularly: the dorsal oblong overlapping scuta expressing both α and β keratins, and the ventral roundish nonprotruding reticula expressing only α keratins. The question arises how En-1 and Lmx1, whose role in dorsoventral limb patterning has been well established, can affect skin morphogenesis, which occurs 8 to 12 days later. Forced expression of En-1 or of Lmx1 in the hindlimb have, respectively, as expected, a ventralizing or a dorsalizing effect on skin, leading to the formation of either reticula-type or scuta-type scales on both faces. In both cases, however, the scales are abnormal and even glabrous skin without any scales at all may form. The normal inductive interactions between dermis and epidermis are disturbed after En-1 or Lmx1 misexpression. Effectively, while Lmx1 endows the dermal precursors of the ventral region with scuta inducing ability, En-1 blocks the competence of the dorsal epidermis to build scuta. Developmental Dynamics 229:564–578, 2004. © 2004 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In chick, while most of the body is covered with feathers, the feet bear scales. Two main types of scales can be distinguished (Lucas and Stettenheim, 1972). Large, asymmetrical, distally overlapping scales (scuta) are arranged in two longitudinal rows on the dorsal side of the tarsometatarsus and in one row on the upper face of each toe. Small nonoverlapping, symmetrical tubercular scales (reticula) are arranged in a tight hexagonal pattern and cover the plantar surface. Not only are the shape and distribution of these two types of scales different, but the outline of their dermal–epidermal junction (DEJ), the thickness of the epidermis, as well as the type of keratins, which are expressed, i.e., α keratins by the reticula, both alpha- and beta-keratins by the scuta (O'Guin and Sawyer, 1982), also differ. The question arises how particular genes, already known to be involved in limb dorsoventral patterning, might contribute to define those different skin characteristics.

During limb bud outgrowth three different axes: proximodistal, anteroposterior, and dorsoventral, are established (reviewed by Johnson and Tabin, 1997; Tickle, 1999; Capdevila and Izpisua Belmonte, 2001). These axes are established early, by means of interacting signalling systems. Here, we concentrate on dorsoventral patterning, which is controlled by signals from the ectoderm (review in Irvine and Vogt, 1997; Chen and Johnson, 1999). Evidence for the importance of ectodermal signalling in dorsoventral patterning comes from experiments in which the ectodermal jacket of a chick limb bud was rotated 180 degrees, such that dorsal ectoderm contacts ventral mesenchyme. The results showed that from stage Hamburger and Hamilton (HH) 15/16 to 25 the ectoderm imposes dorsoventral patterning on the underlying limb mesoderm (MacCabe et al., 1973, 1974; Pautou and Kieny, 1973; Pautou, 1977; confirmed by Geduspan and MacCabe, 1987, 1989; Akita, 1996). This ectodermal influence acts on cartilage and muscle patterns as well as on skin morphogenesis. From stage HH 26/27, the results are inverted, and the type of scale depends on the mesoderm orientation (Pautou, 1977). Recent recombination studies (Piedra et al., 2000) show that the recombinant ectoderm maintains previously established domains of gene expression but reorganizes dorsoventral patterning in the progress zone. Several days later, when the skin morphogenesis occurs, heterotopic dermal–epidermal recombinants showed that its regional diversity depends both on dermal inductive properties and on epidermal competence (reviewed by Dhouailly, 1977; Dhouailly and Sengel, 1983; Sawyer, 1983). Particularly, the dorsal tarsometatarsal epidermis is competent to form feathers (Rawles, 1963), scuta, or reticula (Linsenmayer, 1972), while the plantar epidermis shows a restricted ability to differentiate into scales only (Linsenmayer, 1972; Kanzler et al., 1997).

More recent studies have identified several molecules involved in early dorsoventral limb patterning. Wnt-7a, a member of the Wnt family of secreted proteins, and the LIM-homeodomain transcription factor Lmx1, have been shown to be expressed in the dorsal ectoderm and dorsal mesoderm, respectively, during mouse (Gavin et al., 1990; Parr et al., 1993) and chick (Dealy et al., 1993) limb bud development, while the homeobox-containing gene Engrailed-1 (En-1) was shown to be expressed in the ventral ectoderm of the developing limb bud in mouse (Davis et al., 1991; Wurst et al., 1994) and chick (Davis et al., 1991; Gardner and Barald, 1992). In mice, loss of Wnt-7a function (Parr and McMahon, 1995) results in the transformation of dorsal limb structures to a more ventral phenotype, whereas loss of En-1 function (Loomis et al., 1996) transforms ventral limb structures to a more dorsal phenotype. Such genetic analyses in mutant mice (Cygan et al., 1997), as well as misexpression studies in chick, suggest that the dorsalizing activity of Wnt-7a in the mesenchyme is mediated through the regulation of Lmx1 (Riddle et al., 1995; Vogel et al., 1995) and that En-1 represses Wnt-7a–mediated dorsal differentiation by limiting the expression of Wnt-7a to the dorsal ectoderm (Logan et al., 1997). Although misexpression of mouse En-1 in the early chick limb bud clearly leads to a perturbation of dorsoventral patterning and a loss of posterior skeletal elements (Logan et al., 1997), the infected limbs were analysed too early to be able to determine their skin phenotype.

In Wnt-7a mutant mice (Parr and McMahon, 1995), dorsal skin structures, such as hair follicles and nails, are reduced, whereas the ventral ones, such as footpads and thick epidermis with parallel wrinkles, develop in the dorsal face, which also has less fur. Likewise, in En-1 mutant mice (Loomis et al., 1996), ventral skin adopts a dorsal pattern with development of hair follicles, cylindrical nails, and loss of distal pads and eccrine glands. In chick, Lmx1 misexpression results in digits with large scales and the development of feathers on the ventral and/or the dorsal side of the foot (Vogel et al., 1995). Conversely, down-regulation of Lmx1 activity results in digits that lack scuta and bear reticula-like scales on their dorsal surface (Rodriguez-Esteban et al., 1998). Thus, although it has been established that the perturbation of dorsoventral signalling at the limb bud stage acts on cartilage, muscle patterning, and on tegument regional specification, is it not known how skin morphogenesis, which is a late event, is redefined. It would be interesting to identify what skin characteristics might be changed and how exactly the dermal inductive properties and the epidermal competence are affected.

Here, we approach this question by examining the development of chick foot skin after misexpression of Lmx1 and En-1. We first follow the expression of Wnt-7a, Lmx1, and En-1 from stage 21 to stage 36/38 in normal development, as well as after En-1 and Lmx1 overexpression. By identifying not only the external skin phenotype but also the histology and molecular characteristics, i.e., the keratin type of cutaneous appendages of infected feet, from stages HH 36 to 44, we confirm that, after En-1 misexpression, there is a reversal from dorsal to ventral skin phenotype and, after Lmx1 misexpression, from ventral to dorsal. However, in both cases there is not a complete reversal, the scale type is abnormal and glabrous skin areas form, revealing discordance between the inductive properties of the dermis and the competence of the epidermis. This conflict was confirmed in the case of En-1 misexpression by the results of dermal/epidermal recombination experiments, showing that, in this case, only the dorsal epidermal properties are affected, while the dorsal dermis keeps its dorsal-type inductive properties.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Scutate and Reticulate Scale Differences in Control Hindlimbs

We studied this only in the case of RCAS-mEn-1 infection. Except in a few cases in which the infection spreads to encompass both feet, skin development of the contralateral (i.e., control) foot occurs normally. This development was described previously (Sawyer, 1972; Sawyer and Craig, 1977; O'Guin and Sawyer, 1982; Knapp et al., 1993). In the following, we concentrate on the differences between scuta and reticula. The first sign of scuta formation is the appearance of three epidermal placodes on the dorsal side of the shank at stage HH 35–36 (Fig. 1A). On the ventral side, reticulae appear 2 days later (stage HH 37–38) on the metatarsal and digital pads, as small dome-shaped elevations without any placode formation (Fig. 1B). During subsequent development, the scutate placodes elongate distally to become asymmetric, and are associated with a dermal condensation. From stage HH 42, the outer epidermal surface of each scuta is approximately twice as thick as its inner surface (Fig. 1C,E), but thinner than the reticulate epidermis (Fig. 1D,F). Reticulae arise as symmetrical anlagen, with epidermis of uniform thickness, and remain radially symmetrical without any dermal condensation, throughout their development. The DEJ is also significantly different between the two types of scales: the DEJ of scuta is linear (Fig. 1E), while the DEJ of reticula is papillomatous (Fig. 1F). Moreover, the transition in cell shape between the basal and the intermedium stratum is abrupt in the epidermis of the scuta (Fig. 1E), while it is gradual in the reticula (Fig. 1F). Another noteworthy difference between mature scuta and reticula is their keratin distribution. In particular, both α and β keratins are expressed in the stratum intermedium of scuta (Fig. 1G,I), while only α keratins are expressed in the stratum intermedium of reticula (Fig. 1H,J).

thumbnail image

Figure 1. Skin histogenesis and keratin expression in contralateral legs (controls). A,B: Longitudinal sections of asymmetrical scutate (A) and symmetrical reticulate (B) scale anlagen at stage Hamburger and Hamilton (HH) 39. Note the beginning of a thinner epidermis (e) in the scuta hinge region, whereas the thickness of the reticula epidermis is constant. C–F: At stage HH 44, the differences are more obvious. C: The developing scuta points toward the distal end (arrow) of the foot and involves an inner thin (is) and an outer thickened (os) epidermis. E: The later is lying on a straight dermal/epidermal junction (arrowhead), and composed of a basal stratum (sb), a stratum intermedium (si) and a well individualized stratum corneum (sc). D,F: The reticula is symmetrical (D), its epidermis (F) is papillomatous (arrowhead), and there is a gradual transition from the cells of the basal stratum to the thin stratum corneum. G–J: The outer epidermal surface of the scuta synthesizes both α (G) and β (I) keratins, while in the reticula, the epidermis synthesizes only α keratins (H,J). d, dermis. The arrows point to the distal end of the foot. A–F: Hematoxylin/Biebrich scarlet staining. G–J: Indirect immunofluorescence with polyclonal antibodies against α (G,H) and β (I–J) keratins. Scale bars = 50 μm in A,B, 100 μm in C,D, 10 μm in E,F, 100 μm in G–J.

Download figure to PowerPoint

cWnt-7a,cLmx1, cEn-1, and cShh Expression During Hindlimb Development and Morphogenesis of Its Integument

We compared expression of cWnt-7a and cLmx1 at different stages in chick leg development. cWnt-7a is expressed in dorsal ectoderm in the early limb bud, but, as it grows out, expression is reduced proximally. At stage HH 27, when the first cartilage ray of the foot is developing, cWnt-7a is expressed in distal dorsal ectoderm (Fig. 2A,B), while cLmx1 is expressed throughout the dorsal mesenchyme, with a slight reduction near the tip and the anterior margin of the digital plate (Fig. 2C). Sections show that cLmx1 transcripts are distributed in a thick layer of dorsal mesenchyme (Fig. 2D). From stage 30, cWnt-7a (Fig. 2E,F) and subsequently cLmx1 (Fig. 2G,H) expression, in dorsal ectoderm and mesenchyme, respectively, decrease. At stage HH 34, cWnt-a expression reappears in the distal tarsometatarsal ectoderm, at the base of digits 3 and 4 (Fig. 2I), in exactly the same place where the first two groups of scuta placodes will form at stage HH 36 (Sawyer, 1972; Dhouailly et al., 1980). From stage HH 37, the expression of cWnt-7a in the epidermis is restricted to the distal part of the differentiating scutate placodes, which become the inner epidermal surface of the scuta, as shown here in the control foot (Fig. 6G). By contrast, from stage HH 35/36, cLmx1 is no longer expressed in subectodermal mesenchymal cells, but is found deeper, around the developing tendon (Fig. 2J).

thumbnail image

Figure 2. cWnt-7a and cLmx1 expression during normal foot skin morphogenesis. A,B: At stage Hamburger and Hamilton (HH) 27, dorsal view of the hindlimb (A) and transversal section of the autopod (B). cWnt-7a expression is generally strong distally and weak proximally (arrow). In region of developing digit IV (black arrowheads), cWnt-7a expression is reduced, in contrast to interdigital region (white arrowhead). C: At the same stage, cLmx1 expression is strong throughout the dorsal surface of the leg, except anteriorly and distally (arrowhead), where expression is weak. D: A transverse section shows this expression throughout a thick layer of dorsal mesenchyme (arrowhead). E,F: At stage HH 30, The cWnt-7a expression is low proximally (white arrow), as well as over digits, while it is strong over interdigital regions (back arrows); dorsal view (E) and transverse section at the level of the digits (F). G,H: At the same stage, cLmx1 expression is reduced over central regions of digits III and IV; dorsal view (G) and transverse section at the level of digit III (H). I: At stage HH 34, cWnt-7a expression reappears at the base of digit 3 and 4, where the first scutate placodes will form. J: At stage HH 36, cLmx1 expression is reduced and even undetectable in the dorsal tarsometatarsal skin where the two rows of scuta will form. Scale bars = 340 μm in A,C, 500 μm in E,G, 3 mm in I, 3.7 mm in J.

Download figure to PowerPoint

thumbnail image

Figure 6. Morphology and alteration in mEn-1, cShh, and cWnt-7a dorsal expression patterns in RCAS-mEn-1–infected right foot, compared with control left foot. A,B: At stage Hamburger and Hamilton (HH) 39, the control (A) shows narrow (arrowhead) mEn-1 expressing region, by contrast with the entirely expressing dorsal region of the right foot (B). C: The control leg bears rows of scuta (s) on the dorsal skin. D: In contrast, abnormal scales (abs) or reticula-like buds (rl), form on dorsal infected leg. E,F:cShh is expressed throughout scuta placodes (s) of control foot (E), whereas the infected foot shows spots of cShh-expressing cells (F), which correspond to forming reticula-like (rl) structures. G,H:cWnt-7a is expressed on the distal edge of the scuta (arrowhead) in control foot, whereas it is absent and/or displays an irregular pattern over a small area (arrowhead in H) of the infected foot. I: At stage HH 44, in the control leg (I), scutate scales (s) cover the dorsal surface. Some abnormally wrinkled scuta (abs) may result from a weak mEn-1 expression (compare with A). J: The dorsal skin of infected foot shows three types of abnormalities: reticula-like structures (rl), abnormal misshaped scuta (abs), and glabrous skin (g). Whole-mount in situ hybridization analysis (A,B,E–H); morphology (C,D,I,J) after Bouin Hollande fixation; I–IV, digit number; f, feather; abf, abnormal feather. Scale bars = 0.9 mm in A–F, 0.7 mm in G,H, 1.7 mm in I,J).

Download figure to PowerPoint

Previous studies have shown that during the initial stage of limb bud outgrowth, cEn-1 is expressed uniformly throughout the ventral limb ectoderm and extends into the ventral half of the apical ectodermal ridge (Fig. 3A; Davis et al., 1991; Gardner and Barald, 1992; Logan et al., 1997). To examine the putative role of En-1 during skin morphogenesis, its spatial and temporal pattern of expression was studied at later stages (i.e., HH 30 to HH 40). Little or no expression was detected at stage HH 30 (Fig. 3B). At stage HH 36, cEn-1 is strongly expressed ventrally at the tip of each digit where claws are forming (data not shown). By stage HH 37, when the first reticula begin to form right in the middle of the plantar region (Dhouailly et al., 1980), the epidermis of the central footpad, as well as that of the digital pads, express cEn-1 (Fig. 3C,D). Subsequently, the expression of cEn-1 becomes restricted to the reticula epidermis (Fig. 3E,F). cEn-1 expression was never detected in the developing scuta of the dorsal integument. In contrast, the expression of cShh occurs both in ventral (Fig. 3G) and dorsal (Fig. 3H) epidermis by stage HH 39 and 38, respectively, in distinct patterns that correspond to reticula and scuta buds.

thumbnail image

Figure 3. cEn-1 and cShh expressions during normal foot skin morphogenesis. A–H: Whole-mount in situ hybridization analysis of endogenous cEn-1 mRNA expression (blue) at stage Hamburger and Hamilton (HH) 21 (A), 30 (B), 37 (C,D), 40 (E,F), and of cShh at stage HH 39 (G) and 38 (H). A: Endogenous cEn-1 is uniformly expressed in the ventral ectoderm (ve) of the hindlimb during early bud outgrowth. B: Little or no expression is detected at stage HH 30. C,D: At stage HH 37 a new wave of expression precedes the formation of reticula throughout the entire pads in the plantar and digital regions (arrowheads, C), and is localized in the epidermis (D). E,F: Later, this expression becomes restricted to the developing reticula (r). Note that the scuta (s) do not express cEn-1. G,H: At stage HH 39 and 38, respectively, cShh is expressed (white arrowheads) in the forming reticula (G) and forming scuta (H). de, dorsal epidermis; s, abnormal scales; cb, caudal bud; so, somites. Scale bars = 160 μm in A, 0.5 mm in B–D, 1.5 mm in E, 160 μm in F, 400 μm in G, 600 μm in H.

Download figure to PowerPoint

cLmx1 Infection Disrupts Skin Morphogenesis

RCAS-cLmx1 was injected into hindlimb mesoderm at stage HH 10/11. Seven embryos were recovered at stage HH 24/25, more than 3 days after infection, and showed a complete infection in dorsal and ventral mesenchyme (data not shown). Three embryos were recovered at stage HH 40. The infected leg of one embryo had hyperextended toes that stood up almost vertically (data not shown) similar to earlier reports (Riddle et al., 1995; Vogel et al., 1995). In the other two embryos, some toes curved ventrally and had double nails (Fig. 4A), which were shorter than the single nails on toes of the same foot. The ventral epidermis of infected foot either had large scutate-like scales (Fig. 4B,C), or remains glabrous with no evidence of any scales, or had small reticulate-like scales. In one instance, all three features were seen in the same toe (Fig. 4D): at its distal end, scuta-like scales, an intermediate smooth region and proximally tubercular reticula-like scales. Thus, dorsalization appears more marked distally than proximally. Moreover, toes that had double nails had glabrous skin throughout.

thumbnail image

Figure 4. Alteration of the ventral integument of the digits from RCAS-cLmx1-infected right foot. A: At stage Hamburger and Hamilton (HH) 40, the ventral view of the distal region, shows a double nail on digit 2 (small arrows). The digital pads (strong arrows) are glabrous. B,C: In another case, dorsal (B) and ventral (C) views of digit 3 show the formation of scuta on both surfaces. D: Ventral view of another digit 3 shows scuta-type scales distally (arrows), a smooth area (asterisk) and reticula scales proximally (arrowheads). Scale bars = 500 μm in A, 700 μm in B–D.

Download figure to PowerPoint

RCAS-mEn-1 Infection Alters the Distribution of cShh and cWnt-7a Expression and Disrupts Skin Morphogenesis

Abnormal dorsal tarsometatarsal skin development was observed in 78% of the right hindlimbs infected before stage HH 9 (n = 9), 54% of those infected between stages HH 10 and 15 (n = 177), and 40% of those infected between stage HH 16 and 23 (n = 17; Table 1). Additionally, appendage formation is generally delayed in RCAS-mEn-1–infected legs.

Table 1. Different Types of Dorsal Tarsometatarsal Skin Morphology Obtained After RCAS En-1 Infection of the Hindlimb
Stage of RCAS En-1 infectionTotal number of recovered caseNormal phenotype (scuta)Abnormal dorsal skin phenotype
Reticula likeDisorganized scuta likeGlabrous%
  1. Note that one, two, or three phenotypes can be present on the dorsal region of a same infected leg. HH, Hamburger and Hamilton.

HH ≤ 9 (7 somites or less)9234278
HH ≤ 15 (8 to 26 somites)1778118653754
HH 16 to 2317935340

At stages HH 37–39, ectopic mEn-1 expression can be distinguished in the dorsal epidermis of the right limb, and in several cases does not reach the ventral side (Fig. 5A,B). It should be noted that, after ectoderm infection, the misexpression was restricted to the epidermis in all analysed cases, as already noted in previous publications (e.g., Riddle et al., 1995; Morgan and Fekete, 1996). At stage HH 39, in some cases, faint mEn-1 expression can be observed on the left foot (Fig. 6A). At the same stage, in the dorsal skin of the right limb, mEn-1 is expressed in the periphery of abnormal scale primordia or in uniform patches of epidermis that remain glabrous (Fig. 6B). Two distinct rows of large scuta primordia cover the tarsometatarsus of the uninfected limb (Fig. 6C), whereas on the dorsal tarsometatarsal skin of the RCAS-mEn1–infected limb, small dispersed reticula-like primordia appear more often (Fig. 6D). At the same stage, the control left foot displays homogeneous expression of cShh in the epidermis of the oblong-shaped scutate scale buds (Fig. 6E). By contrast, the RCAS-mEn1–infected right foot shows either a punctuate, irregular distribution of cShh on its dorsal surface (Fig. 6F) or, in some regions, a complete lack of cShh expression (data not shown). This punctuate expression of cShh was comparable to that in plantar and digital pads of the controls (Fig. 3G). cWnt-7a expression, which is present in the distal epidermis of the control scutate scale buds (Fig. 6G), is either irregular or absent from the dorsal epidermis in the case of RCAS-mEn-1 misexpression (Fig. 6H).

thumbnail image

Figure 5. mEn-1 expression is restricted to the dorsal epidermis of infected foot. A,B: After infection at stage Hamburger and Hamilton (HH) 10, whole-mount in situ hybridization analysis at stage HH 37 and corresponding section of digit 3 (B) show that the misexpression is restricted to the dorsal surface of the foot (A), more precisely (B) to the dorsal epidermis (de). ve, ventral epidermis. Scale bars = 400 μm in A,300 μm in B.

Download figure to PowerPoint

When recovered at stages HH 40 to 44, the control leg shows well-defined scuta on its dorsal surface, albeit with abnormal wrinkles, which are likely due to minor mEn-1 expression (compare Fig. 6A and I). In contrast, the skin morphologic appearance of the contralateral infected leg is severely disrupted. In some cases reticula-like structures are formed or, more often, glabrous skin and/or disorganized scuta-like structures are seen (Fig. 6J). In the latter case, the phenotype varies from concave non-overlapping scuta, which include a reticula-like center (Figs. 7A, 9A,C) to convoluted elevations with an unrecognizable pattern (Fig. 7B). In fact, three kinds of abnormal dorsal skin development can be distinguished on the infected feet: (1) dome shaped structures resembling reticulate scales, (2) glabrous skin, and (3) disorganized structures resembling scutate scales. In some cases, all three phenotypes are present on different dorsal regions of the same leg. To follow the possible phenotypical evolution from dorsal reticula-like buds at stage HH 38 to disorganized scuta at stage HH 44, skin from infected hindlimbs, which displayed reticula-like primordia on their entire surface at stage HH 38 was dissected (Fig. 7C), and grafted onto chick chorioallantoic membrane. After 6 days, the skin grafts were covered with disorganized structures (Fig. 7D) resembling the abnormal scuta present on some infected legs at stage HH 44 (Fig. 7B). To better characterize those abnormally shaped appendages, it was relevant to check on their keratinocyte differentiation.

thumbnail image

Figure 7. Variability in abnormal morphogenesis after RCAS-mEn-1 infection (right foot shown). A: At stage Hamburger and Hamilton (HH) 44, an infected case displays concave nonoverlapping scuta and roundish reticula-like structure in the center of the abnormal scuta (abs). B: Another example shows convoluted abnormal scuta. C,D: When dorsal infected tarsometatarsal skin with reticula-like (rl) primordia is dissected at stage HH 39 (C), then grafted on chorioallantoic membrane, and allowed to develop for 6 additional days, the reticula-like structures subsequently developed into convoluted abnormal scuta (D). Morphology after Bouin Hollande fixation; I–IV: digit number; gf: localisation of dissected skin for grafting; f, feather. Scale bars = 0.9 mm in A, 0.7 mm in B, 1.4 mm in C, 1.1 mm in D.

Download figure to PowerPoint

thumbnail image

Figure 9. The dorsal integument of RCAS-mEn-1–infected right foot does not synthesise β keratin at stage Hamburger and Hamilton (HH) 44, but in some abnormal scuta. A,C: In the case of abnormal scales, the reticula-type central structure is made of only α keratins, while some β keratins are irregularly present in the surrounding scuta area. B,D: When glabrous skin formation follows mEn-1 misexpression, the dorsal epidermis (e) expresses only α keratins. Note: compare Figure 9A,C with Figure 7A. d, dermis. Indirect immunofluorescence with polyclonal antibodies against α and β keratins. Scale bars = 100 μm in A–D.

Download figure to PowerPoint

RCAS-mEn-1–Infected Leg Dorsal Epidermis Is Histologically and Biochemically Similar to the Plantar Epidermis

At stage HH 38, histologic analysis revealed that, unlike true reticula primordia, the reticula-like structures present on the dorsal surface of RCAS-mEn-1–infected legs are not strictly symmetrical (Fig. 8A). However, the thickness of the infected dorsal epidermis ranged from three to eight cell layers (Fig. 8B) and more closely resembled that of the normal plantar epidermis (Fig. 1B). This dorsal epidermal thickening after RCAS-mEn-1 infection is present both on reticula-like structures and between them.

thumbnail image

Figure 8. RCAS-mEn-1–infected right foot dorsal epidermis is histologically similar to plantar epidermis. A: At stage HH 39, the dorsal tarsometatarsal skin develops reticula-like primordia. B: However, as shown at a higher magnification, these primordia involve an abnormal thickening (arrowheads) of the epidermis (e). C,E,G: At stage HH 44, glabrous skin (C), reticula-like structures (E), or abnormal scuta (G) are present on the dorsal tarsometatarsal skin. D,F,H: Higher magnification shows that, in all these cases, the epidermis differentiates with a ventral phenotype, even when a scuta shape is recognizable (G,H). Note the papillomatous dermal–epidermal junction (arrowheads), as well as the gradual transition from the cells of the basal stratum (sb) to the thin stratum corneum (sc). D, dermis. Hematoxylin/Biebrich scarlet staining. Scale bars = 100 μm in A,C,E,G, 10 μm in B,D,F,H.

Download figure to PowerPoint

At stage HH 44, three different dorsal tarsometatarsal skin phenotypes were well established: glabrous skin (Fig. 8C), abnormal reticula (Fig. 8E), or abnormal scuta (Fig. 8G). In all cases, however, the dorsal tarsometatarsal skin presented a papillomatous DEJ, and displayed an overall thickening of the epidermis (Fig. 8D,F,H), two characteristics typical of reticula. Indeed, more than 15 living cell layers can be identified in the dorsal epidermis of RCAS-mEn-1–infected feet, in contrast to the outer surface of normal scutate scales that have only 5 to 7 living cell layers (compare Fig. 8D,F,H with Fig. 1E,F). Moreover, unlike the outer epidermal surface of normal scutate scales, the transition from the stratum basal to the stratum corneum is gradual and a thick stratum corneum was never observed. Immunologic analyses using polyclonal antibodies specific for α or β keratins show that RCAS-mEn-1–infected right legs produced only α keratins in reticula-like structures (Fig. 9A,C) or in glabrous areas (Fig. 9B,D). Moreover, β keratins were rarely present, except in structures morphologically resembling scuta, which overlie the formation of previous reticula (compare Fig. 9C with Fig. 7A). By contrast, the dorsal tarsometatarsal scales of left control legs expresses both α and β keratins (Fig. 1G,I). RCAS-mEn-1 infection, thus, changes the dorsal morphogenesis of the epidermis into a ventral one, even if in some cases the overall shape of scales resembles scuta. We then investigated the potential modifications of the inductive properties and competence of the two skin components.

RCAS-mEn-1 Infection Permanently Changes the Developmental Potential of Leg Dorsal Epidermis, but Does Not Directly Affect Leg Dermal Properties

To follow the changes in the tissue developmental potential which occur after mEn-1 misexpression, dermal–epidermal recombinants were performed between skin tissues from stage HH 38 infected leg and normal skin tissues from different regions of noninfectable legs (Table 2). It should be kept in mind that, in infected foot, the dorsal epidermis expresses mEn-1 but the dermis does not (Fig. 5B). Recombinants were grafted onto chick chorioallantoic membrane and allowed to develop for 8 days before analysis. Two pieces of skin were dissected from the dorsal tarsometatarsal region of each infected leg. One piece (1) serves as a phenotypic control after grafting. The other skin piece was separated into epidermis and dermis. The epidermis (2) from infected foot was re-associated with either a normal dorsal tarsometatarsal dermis or a normal back dermis. The dermis from infected foot (3) was re-associated either with a normal dorsal tarsometatarsal or a normal midventral epidermis. For each set of three grafts (1, 2, plus 3), seven to nine cases were recovered 8 days after grafting. Six trios were also performed between tarsometatarsal skin dissected at stage HH 39 from infected and normal feet. The results were unambiguous. Skin samples involving infected epidermis, controls as well as recombinants (Fig. 10A–C and D–F) give rise to glabrous explants, or to explants covered with irregular deformed scales, or simply to a wrinkled epidermis. In all cases, the DEJ is papillomatous (Fig. 10C,F). Identical results were obtained by using a normal back feather-forming dermis, instead of the normal foot dorsal dermis (data not shown). The epidermis of these heterotypic grafts and their controls synthesized only α keratins (Fig. 10B,C,E,F). By contrast, the recombinants involving dorsal tarsometatarsal dermis from infected foot form oblong and even sometimes overlapping scuta (Fig. 10G), the epidermis of which synthesizes both α and β keratins (Fig. 10H,I). These apparently normal scuta were formed both in the recombinants using a tarsometatarsal epidermis as well as a midventral apterium epidermis (data not shown).

Table 2. Phenotypic Outcomes of Recombinants Involving Tissues from RCAS-mEn-1–infected Foot and from Normal Embryos
DermisEpidermis
10-11d tmt epidermis from infected foot10-11d tmt epidermis from normal foot10-11d mva epidermis from normal embryo
  1. tmt, tarsometatarsal; mva, midventral apterium

10-11d tmt dermis from infected footReticula-like/abnormal scuta/glabrous skin (8 cases)Scuta (8 cases)Scuta (7 cases)
10-11d tmt dermis from normal footReticula-like/abnormal scuta/glabrous skin (9 cases)Scuta (8 cases)Scuta (Cadi et al., 1983)
7-7.5d back dermis from normal embryoGlabrous skin/arrested feathers (8 cases)Feathers (Rawles, 1963)Feathers (Sengel et al., 1969)
thumbnail image

Figure 10. RCAS-mEn-1 infection, which affects only the epidermis, does not change the dorsal dermal properties. Skin recombinations performed at stage 38 between tissues of RCAS-mEn-1–infected foot and tissues from a normal foot of a noninfectable species, developed for 6 days on the chick chorioallantoic membrane. A–C: Phenotypic control of infected dorsal tarsometatarsal skin. Formation of irregular wrinkles (A), the epidermis (e) of which (B,C) synthesizes only α keratins. D–F: Recombinant of normal dorsal tarsometatarsal dermis with an infected dorsal tarsometatarsal epidermis. The results are the same as those obtained in the control infected grafted skin. G–I: Recombinant of normal tarsometatarsal epidermis and dorsal tarsometatarsal dermis from an infected foot. Formation of overlapping scuta (G), which synthesize both α and β keratins (H,I). Indirect immunofluorescence with polyclonal antibodies against α and β keratins. d, dermis. Scale bars = 0.4 μm in A,D,G, 100 μm in B,C,E,F,H,I.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Previous studies have shown that the dermis is able to induce different types of cutaneous appendages, depending not only on its own origin but also on the competence of its overlying epidermis (Dhouailly, 1977, 1984). At the time of skin morphogenesis, the epidermis appears already subdivided into three categories: feather-forming, scale-forming, and neutral.

In general, the epidermis of feather fields (pterylae) is biased toward feather formation, while the epidermis of the midventral apterium, like the anterior tarsometatarsal epidermis, can form either feathers or scales, depending on the origin of the dermis which they are associated with (Sengel et al., 1969; Cadi et al., 1983). The plantar epidermis, on the other hand, is biased toward scale formation. When recombined with an anterior tarsometatarsal dermis, the plantar epidermis forms scutate scales (Linsenmayer, 1972). However, these oblong scales are not overlapping and have a very low expression of β keratins in all the cells of their stratum intermedium (Dhouailly, unpublished data). When recombined with a plantar or back dermis, the plantar epidermis forms normal or abnormal reticula, respectively, whereas the association of plantar dermis and apteric epidermis leads to the formation of abnormally short, but well-differentiated feathers expressing β keratins (Kanzler et al., 1997). Hence, we have previously advanced the hypothesis that reticula are not proper scales, but growth-arrested feathers (Kanzler et al., 1997). The question thus arises how and when are these different properties of the foot epidermis and dermis acquired? The results presented here shed light on the molecular mechanisms underlying skin foot morphogenesis, in particular the specific roles of the epidermis and dermis.

We show that expression of cShh and cWnt-7a is repressed after RCAS-mEn-1 misexpression. During normal development, cShh is expressed in the entire oblong dorsal scale bud, whereas in infected dorsal epidermis, it may be expressed in spots, a pattern similar to the normal distribution of Shh expression in plantar epidermis, or more or less totally absent. The latter case might correspond to the formation of a glabrous skin. The alteration in cShh and cWnt-7a expression could be an indirect consequence of the early ventral specification of the dorsal ectoderm by ectopic mEn-1 expression, resulting later in expression of cEn-1 during scale bud morphogenesis, or it could be a direct effect. It should be noted that retinoic acid treatment might bypass the inhibitory effect of En-1 in plantar epidermis. Indeed this treatment that affects only the chick epidermal properties (Cadi et al., 1983) is able to entail both the scuta and reticula epidermis to feather formation (Dhouailly et al., 1980; Fisher et al., 1988). This change correlates with increasing epidermal signaling as cShh (Prin and Dhouailly, unpublished data), which has been shown previously to be involved in the outgrowth of skin appendages (Ting-Berreth and Chuong, 1996; Morgan et al., 1998; Widelitz et al., 1999).

It is well known that Lmx1 expression is induced in the limb bud mesoderm in response to Wnt-7a (Riddle et al., 1995; Vogel et al., 1995). Here, we show that, during chick leg development, cWnt-7a, as well as cLmx1 expression are progressively and successively reduced, beginning proximally. Particularly, scutate scales are induced by the dermis after the Lmx1 “dorsalizing gene” is no longer expressed. Thus, the dorsal mesoderm might be patterned as dorsal to begin with, and this dorsal quality might be relayed, proximally first, by some other genes that allow the maintenance of the dorsal-inducing properties of the tarsometatarsal dorsal dermal progenitors. After cLmx1 misexpression, dorsoventral alterations occur preferentially distally in the digits. In cLmx1-infected limbs, the suspected relay of dorsalizing genes might not occur, and the effect on skin consequently reversed proximally first. Another explanation could be that the limb mesenchyme is specified in a progressive proximodistal manner, and the experimental conditions affected late properties of the integument.

It is well known that cEn-1 is expressed early throughout the chick ventral limb bud ectoderm (Davis et al., 1991; Gardner and Barald, 1992; Logan et al., 1997). Here, we show that this expression disappears later during chick limb outgrowth and then reappears uniformly in the epidermis concomitant with the formation of plantar foot pads. This expression becomes subsequently punctuate, corresponding to individualization of reticula. This later cEn-1 expression in the foot skin is specific to the plantar region as no signal was detected in the developing scuta. It should be noted that, in adult mice, cEn-1 expression is also restricted to the plantar foot pads (Mainguy et al., 1999). These two stages of expression suggest that cEn-1 plays a role at two different developmental stages. When expressed in the ventral ectoderm at the limb bud stage, it allows the formation of a dorsoventral limb, by preventing the expression of cWnt-7a ventrally. Its later expression throughout all the ventral foot epidermis during skin morphogenesis might be implicated in the restricted competence of the plantar epidermis. The dorsal late expression of cWnt-7a, that does not induce cLmx1 activation, appears first in the place of the future scutate placodes, then at the distal/inner face of the scutate scale buds and, thus, might be involved in their determination as well as their proximodistal orientation and growth, plus the subsequent expression of β keratins. This late expression of cWnt-7a is similar to that observed first throughout the feather placode, and then in the posterior part of the feather bud epidermis (Chuong et al., 1996; Widelitz et al., 1999). It is important to note that cWnt-7a expression was not observed at any stage during reticula development but that cEn-1 expression was seen throughout the developing bud. Given that ectopic RCAS-mEn-1 expression in the dorsal ectoderm is sufficient to repress earlier endogenous cWnt-7a expression (Logan et al., 1997), it is likely that, during normal development, the later phase of cWnt-7a expression in the epidermis is similarly repressed by En-1 expression.

After RCAS-mEn-1 infection in the presumptive hindlimb dorsal territory, the infected dorsal tarsometatarsal and digit epidermis forms first mainly ventral reticula-type primordia at stages HH 37–39. In contrast to the partial transformation of the underlying dorsal mesoderm of the limb (Logan et al., 1997), abnormal skin phenotypes can occur on the entire dorsal surface of tarsometatarsus and digits. Recombination experiments, involving infected epidermis and normal dermis from different regions, clearly demonstrate that the tarsometatarsal dorsal epidermis has lost its normal competence to form scuta or feathers as well as to synthesize β keratins. Forced expression of mEn-1 in the dorsal limb bud, therefore, leads to an irreversible ventral-type differentiation of the dorsal epidermis by modifying its capacity to interact with its underlying dermis. Of interest, the scaleless mutation, known to affect the epidermis (Sengel and Abbott, 1963; Song et al., 1996; Viallet et al., 1998), modifies its capacity to interact with its underlying dermis (Dhouailly and Sawyer, 1984; Viallet et al., 1998).The scaleless mutant, characterized by the absence both of scuta and of most of the feathers, presents a ventral foot epidermis similar histologically and biochemically to the reticula (Sawyer, 1979; O'Guin and Sawyer, 1982), and, moreover, morphologically similar to the dorsal integument of mEn-1–infected feet. It should be noted that the scaleless dorsal foot dermis retains its ability to induce scuta and β keratin expression in a normal epidermis (Dhouailly and Sawyer, 1984). Likewise, the tarsometatarsal dorsal dermis from a RCAS-mEn-1–infected foot retains its ability to induce the formation of both scuta shape and β keratin expression by normal epidermal cells. The misexpression of mEn-1 in the epidermis, thus, is not sufficient to offset the dorsal properties of dermal cells and/or change them to plantar properties. It has already been noted (Logan et al., 1997) that mEn-1 expression in the limb bud perturbs dorsal patterning but does not result in a complete transformation to a ventral phenotype. Analyses of En-1/Wnt-7a −/− mutant mice (Cygan et al., 1997) show that mesenchyme can induce ventral specific structures in the overlying ectoderm in the absence of En-1 function but that this skin morphogenesis is correlated with the absence of Lmx1 expression in the mesenchyme at early limb bud stage.

When recovered at stage HH 44, RCAS-mEn-1–infected dorsal tarsometatarsal epidermis displays abnormal cutaneous appendages: (1) reticula-type structures, (2) glabrous skin, and (3) convoluted scuta. However, these three different structures have similar epidermal architecture: the number of cell layers, dermal/epidermal junction, and type of keratins (α) are those of a plantar-type epidermis. Grafts of stage HH 38 infected skin onto chick chorioallantoic membrane show that convoluted scales may develop from reticula-like primordia. They might result from a later dorsal-inducing activity from the dermis, which, however, cannot change the histologic and biochemical differentiation of the epidermis. It has been known for many years that the dermis is responsible for the acquisition of the general shape of a cutaneous appendage, while the epidermis is responsible for its own cell number, shape, and keratinisation (Dhouailly, 1977; Dhouailly et al., 1998). Thus, in infected limbs, the dorsal epidermis starts to differentiate into roundish ventral-type reticula primordia, and then the dorsal scutate oblong shape is overlayed by the dermal induction. Due to the intermediate shape of the scales or even the formation of glabrous skin, we suggest that there is a conflict between the dermal induction and the competence of the epidermis. In all cases, the RCAS-mEn-1–infected dorsal epidermis displays histologic and biochemical ventral characteristics.

In conclusion, these results show that the morphogenesis of the dorsal and ventral chick scales relies directly on cEn-1 and indirectly on cLmx1 expressions. cEn-1 expression in the ventral epidermis suppresses the epidermal competence for cutaneous appendage outgrowth in the plantar skin. By contrast, cLmx1 expression in the hindlimb dorsal mesenchyme is succeeded by still undetermined and epidermis-independent gene expression in the dermis, responsible for the acquisition of the oblong shape of dorsal scales.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Chick Embryos

Embryos were staged according to Hamburger and Hamilton (1951). Line O eggs from BBSRC Institute for Animal Health (Berks, UK) were used to produce primary cultures of chick embryo fibroblasts (CEFs). White Leghorn germ-free embryos from the Haas poultry farm (Strasbourg, France) or from Poydon farm Waltham Cross, Hertfordshire (UK) were used for retroviral infection. Warren embryos from the Cerveloup poultry farm (Moirans, France) were used for grafting on the chorioallantoic membrane. The noninfectable JA-957 strain (SFPA, France) was used as a source of noninfected tissues for the recombination experiments.

Retroviral Construction and Infection

RCAS-cLmx1 contains a chick Lmx1 cDNA spanning the entire coding region. The cDNA was cloned into the shuttle vector SLAX-13 and then subcloned into retroviral vector RCAS(BP)A (Riddle et al., 1995). RCAS-cLmx1 virus at titre of 108 was injected into the presumptive hindlimb at stage 10/11.

The RCAS vector RCASBP(A) encoding mouse En-1 (Logan et al., 1997) and the control vector RCASBP-AP (A) (Fekete and Cepko, 1993), were used to transfect CEFs using transfectam (SEPRACOR), according to the manufacturer's instructions. After 5 days of culture, the infection rate reached 100% as assessed using an anti-Gag antibody (Potts et al., 1987). To graft infected cells, confluent cultures were transferred to untreated plastic dishes. After 24 hr, the cells formed compact aggregates. Such aggregates were used to infect stage HH 10–23 germ-free chick embryos by implanting cells into a space made by lifting the ectoderm over the right presumptive dorsal hindlimb region (Altabef et al., 1997; Michaud et al., 1997). Alternatively, concentrated viral stocks (107 to 108 infectious units/ml), prepared as described by Fekete and Cepko (1993), were used to infect the same presumptive region at stages HH 4–23, by placing on the ectoderm. The stage at RCAS-mEn-1 infection and the number of recovered cases are outlined in Table 1.

Recovery of Infected Feet and Recombination Experiments

After RCAS-cLmx1 infection, 10 embryos were recovered at stage HH 24/25 and were used for Lmx1 hybridization, and 3 at stage HH 40 for skin morphogenesis. After RCAS-m-En-1 infection, embryos were allowed to develop to stage HH 36 to 44. Approximately half of recovered infected and contralateral control hindlimbs was processed for in situ hybridization or embedded in OCT compound and subsequently processed for immunocytochemistry. The second half was fixed in Bouin-Holland fluid for phenotypic analyses. To follow phenotypic changes from stage HH 38 to 44, tarsometatarsal dorsal skin fragments were dissected from some stage HH 38 infected hindlimbs, grafted onto chick chorioallantoic membrane, and allowed to develop for 6 days. To perform skin recombinants, skin was dissected from the dorsal tarsometatarsal region (stages HH 36 and 38) of RCAS-mEn-1–infected embryos, and from the back (stage HH 32), the midventral apterium (stage HH 36), and the dorsal tarsometatarsal region (stage HH 36 and 38) of noninfectable JA-957 strain embryos. Recombinants were performed as previously described (Kanzler et al., 1997). Briefly, dermis and epidermis were separated after incubation in 1% trypsin/2% pancreatin, and re-associated as outlined in Table 2. Recombinants were transferred to the chorioallantoic membrane of stage HH 36 embryos and cultured for 8 days. Half of the recombinants was fixed in Bouin's fluid and photographed, the other half was embedded in OCT compound and processed for immunocytochemical analysis.

In Situ Hybridization and Probes

Probes to detect specifically mEn-transcripts, as well as endogenous chick Wnt-7a and En-1 were made as previously described (Logan et al., 1997). The cShh probe was as described in Riddle et al. (1993). The cLmx1 probe was a full-length 3.2 kb. Nonradioactive whole-mount in situ hybridizations were done essentially as described by Wilkinson and Nieto (1993).

Immunologic and Histologic Analyses

The skin phenotypes of the infected right foot, grafted infected skin, or dermoepidermal recombinants, were analysed by macroscopy and histology by staining for hematoxylin/Biebrich scarlet or with Mallory's trichrome. For immunofluorescent staining of keratin, we used polyclonal antibodies that specifically recognize α or β keratins (Dhouailly and Sawyer, 1984).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The authors thank Dr. C. Tickle and Dr. A. Lumsden for their critical advice, Dr. R.H. Sawyer for help with the anti-α and β keratin polyclonal antibodies, and Dr. D.J. Pearton for critical reading of the manuscript. They also thank Mrs. Genevieve Chevalier for histologic work, Mrs. Lulu Liu for help with whole-mount in situ hybridization, and Mrs. Brigitte Peyrusse for the iconography. D.D. was funded by a grant from the CNRS and University Joseph Fourier, A.L. received a grant from the Wellcome Trust and the Howard Hughes Medical Institute, and C.L. received an AHFMR Establishment and MRC operating grant. F.P. was the recipient of a doctoral fellowship from the French Ministère de l'Education Nationale, de la Recherche et de la Technologie. D.D. and M.E. were funded by a grant from the Welcome Trust to Cheryll Tickle.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Akita K. 1996. The effect of the ectoderm on the dorsoventral pattern of epidermis, muscles and joints in the developing chick leg: a new model. Anat Embryol (Berl) 193: 377386.
  • Altabef M, Clarke JD, Tickle C. 1997. Dorsoventral ectodermal compartments and origin of apical ectodermal ridge in developing chick limb. Development 124: 45474556.
  • Altabef M, Logan C, Tickle C, Lumsden A. 2000. Engrailed-1 misexpression in chick embryos prevents apical ridge formation but preserves segregation of dorsal and ventral ectodermal compartments. Dev Biol 222: 307316.
  • Capdevila J, Izpisua Belmonte JC. 2001. Patterning mechanisms controlling vertebrate limb development. Annu Rev Cell Dev Biol 17: 87132.
  • Cadi R, Dhouailly D, Sengel P. 1983. Use of retinoic acid for the analysis of dermal-epidermal interactions in the tarsometatarsal skin of the chick embryo. Dev Biol 100: 489495.
  • Chen X, Johnson RL. 1999. Dorsoventral patterning of the vertebrate limb: a process governed by multiple events. Cell Tissue Res 296: 6773.
  • Chuong CM, Widelitz RB, Ting-Berreth S, Jiang TX. 1996. Early events during avian skin appendage regeneration: dependence on epithelial-mesenchymal interaction and order of molecular reappearance. J Invest Dermatol 107: 639646.
  • Cygan JA, Johnson RL, McMahon AP. 1997. Novel regulatory interactions revealed by studies of murine limb pattern in Wnt-7a and En-1 mutants. Development 124: 50215032.
  • Davis CA, Holmyard DP, Millen KJ, Joyner AL. 1991. Examining pattern formation in mouse, chicken and frog embryos with an En- specific antiserum. Development 111: 287298.
  • Dealy CN, Roth A, Ferrari D, Brown AM, Kosher RA. 1993. Wnt-5a and Wnt-7a are expressed in the developing chick limb bud in a manner suggesting roles in pattern formation along the proximodistal and dorsoventral axes. Mech Dev 43: 175186.
  • Dhouailly D. 1977. Dermo-epidermal interactions during morphogenesis of cutaneous appendages in amniotes. In: RobertL, editor. Frontier matrix biology. Vol 4. Paris: Creteil. p 86121.
  • Dhouailly D. 1984. Specification of feather and scale patterns. In: RdacinskiGM, BryantSN, editors. Pattern formation. New York: MacMillan.
  • Dhouailly D, Sawyer RH. 1984. Avian scale development. XI. Initial appearance of the dermal defect in scaleless skin. Dev Biol 105: 343350.
  • Dhouailly D, Sengel P. 1983. Feather forming properties of the foot integument in avian embryos. In: SawyerRH, FallonJF, editors. Epithelial-mesenchymal interactions in development. New-York: Praeger Press. p 147161.
  • Dhouailly D, Hardy MH, Sengel P. 1980. Formation of feathers on chick foot scales, a stage dependent morphogenetic response to retinoic acid. J Embryol Exp Morphol 58: 6378.
  • Dhouailly D, Prin F, Kanzler B, Viallet JP. 1998. Variation of cutaneous appendages: regional specification and cross-species signals. In: ChuongCM, editor. Molecular basis of epithelial appendage morphogenesis. Vol. 1. Georgetown, TX: R.G. Landes Company. p 4556.
  • Fekete DM, Cepko CL. 1993. Replication-competent retroviral vectors encoding alkaline phosphatase reveal spatial restriction of viral gene expression/transduction in the chick embryo. Mol Cell Biol 13: 26042613.
  • Fisher CJ, Knapp LW, Sawyer RH. 1988. Retinoic acid induction of feather like structures from reticulate scales. Teratology 38: 321328.
  • Gardner CA, Barald KF. 1992. Expression patterns of engrailed-like proteins in the chick embryo. Dev Dyn 193: 370388.
  • Gavin BJ, McMahon JA, McMahon AP. 1990. Expression of multiple novel Wnt-1/int-1-related genes during fetal and adult mouse development. Genes Dev 4: 23192332.
  • Geduspan JS, MacCabe JA. 1987. The ectodermal control of mesodermal patterns of differentiation in the developing chick wing. Dev Biol 124: 398408.
  • Geduspan JS, MacCabe JA. 1989. Transfer of dorsoventral information from mesoderm to ectoderm at the onset of limb development. Anat Rec 224: 7987.
  • Hamburger V, Hamilton HL. 1951. A series of normal stages in the development of the chick embryo. J Morphol 88: 4492.
  • Irvine KD, Vogt TF. 1997. Dorsal-ventral signaling in limb development. Curr Opin Cell Biol 9: 867876.
  • Johnson RL, Tabin CJ. 1997. Molecular models for vertebrate limb development. Cell 90: 979990.
  • Kanzler B, Prin F, Thelu J, Dhouailly D. 1997. CHOXC-8 and CHOXD-13 expression in embryonic chick skin and cutaneous appendage specification. Dev Dyn 210: 274287.
  • Knapp LW, Shames RB, Barnes GL, Sawyer RH. 1993. Region-specific patterns of beta keratin expression during avian skin development. Dev Dyn 196: 283290.
  • Linsenmayer TF. 1972. Control of integumentary patterns in the chick. Dev Biol 27: 244271.
  • Logan C, Hornbruch A, Campbell I, Lumsden A. 1997. The role of Engrailed in establishing the dorsoventral axis of the chick limb. Development 124: 23172324.
  • Loomis CA, Harris E, Michaud J, Wurst W, Hanks M, Joyner AL. 1996. The mouse Engrailed-1 gene and ventral limb patterning. Nature 382: 360363.
  • Lucas AM, Stettenheim PR. 1972. Avian anatomy integument. In: Avian anatomy handbook. Vol. 362. Part 2. Washington DC: US Government Printing Office.
  • MacCabe JA, Saunders JW Jr, Pickett M. 1973. The control of the anteroposterior and dorsoventral axes in embryonic chick limbs constructed of dissociated and reaggregated limb-bud mesoderm. Dev Biol 31: 323335.
  • MacCabe JA, Errick J, Saunders JW Jr. 1974. Ectodermal control of the dorsoventral axis in the leg bud of the chick embryo. Dev Biol 39: 6982.
  • Mainguy G, Erno H, Montesinos ML, Lesaffre B, Wurst W, Volovitch M, Prochiantz A. 1999. Regulation of epidermal Bullous Pemphigoid antigen 1 (BPAG1) synthesis by homeoprotein transcription factors. J Invest Dermatol 113: 643650.
  • Michaud JL, Lapointe F, Le Douarin NM. 1997. The dorsoventral polarity of the presumptive limb is determined by signals produced by the somites and by the lateral somatopleure. Development 124: 14531463.
  • Morgan BA, Fekete DM. 1996. Manipulating gene expression with replication-competent retroviruses. Methods Cell Biol 51: 185218.
  • Morgan BA, Orkin RW, Noramly S, Perez A. 1998. Stage-specific effects of sonic hedgehog expression in the epidermis. Dev Biol 201: 112.
  • O'Guin WM, Sawyer RH. 1982. Avian scale development. VII. Relationships between morphogenetic and biosynthetic differentiation. Dev Biol 89: 485492.
  • Parr BA, McMahon AP. 1995. Dorsalizing signal Wnt-7a required for normal polarity of D-V and A-P axes of mouse limb. Nature 374: 350353.
  • Parr BA, Shea MJ, Vassileva G, McMahon AP. 1993. Mouse Wnt genes exhibit discrete domains of expression in the early embryonic CNS and limb buds. Development 119: 247261.
  • Pautou MP. 1977. Etablissement de l'axe dorso-ventral dans le pied de l'embryon de poulet. J Embryol Exp Morphol 42: 177194.
  • Pautou MP, Kieny M. 1973. Interaction ecto-mesodermique dans l'etablissement de la polarite dorsoventrale du pied de l'embryon de poulet. C R Acad Sci Ser D 227: 12251228.
  • Piedra ME, Rivero FB, Fernandez-Teran M, Ros MA. 2000. Pattern formation and regulation of gene expressions in chick recombinant limbs. Mech Dev 90: 167179.
  • Potts WM, Olsen M, Boettiger D, Vogt VM. 1987. Epitope mapping of monoclonal antibodies to gag protein p19 of avian sarcoma and leukaemia viruses. J Gen Virol 68(Pt 12): 31773182.
  • Rawles ME. 1963. Tissue interactions in the scale and feather development as studied in dermal-epidermal recombinations. J Exp Zool 135: 503527.
  • Riddle RD, Johnson RL, Laufer E, Tabin C. 1993. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75: 14011416.
  • Riddle RD, Ensini M, Nelson C, Tsuchida T, Jessell TM, Tabin C. 1995. Induction of the LIM homeobox gene Lmx1 by WNT7a establishes dorsoventral pattern in the vertebrate limb. Cell 83: 631640.
  • Rodriguez-Esteban C, Schwabe JW, Pena JD, Rincon-Limas DE, Magallon J, Botas J, Belmonte JC. 1998. Lhx2, a vertebrate homologue of apterous, regulates vertebrate limb outgrowth. Development 125: 39253934.
  • Sawyer RS. 1972. Avian scale development I. Histogenesis and morphogenesis of the epidermis and dermis during formation of the scale ridge. J Exp Zool 181: 365384.
  • Sawyer RH. 1979. Avian scale development. Effect of the scaleless gene on morphogenesis and histogenesis. Dev Biol 68: 115.
  • Sawyer RH. 1983. The role of epithelial-mesenchymal interactions in regulating gene expression during avian scale morphogenesis. In: SawyerRH, FallonJF, editors. Epithelial-mesenchymal interactions in development. New-York: Praeger Press. p 115146.
  • Sawyer RH, Craig KF. 1977. Avian scale development. Absence of an epidermal placode in reticulate scale morphogenesis. J Morphol 154: 8393.
  • Sengel P, Abbott UK. 1963. In vitro studies with the scaleless mutant: interaction during feather and scale differentiation. J Hered 54: 254262.
  • Sengel P, Dhouailly D, Kieny M. 1969. Aptitude of the skin constituents of the mid-ventral apterium of the chicken for forming feathers. Dev Biol 19: 436446.
  • Song HK, Wang Y, Goetinck PF. 1996. Fibroblast growth factor 2 can replace ectodermal signaling for feather development. Proc Natl Acad Sci U S A 93: 1024610249.
  • Tickle C. 1999. Morphogen gradients in vertebrate limb development. Semin Cell Dev Biol 3: 345351.
  • Ting-Berreth SA, Chuong CM. 1996. Sonic Hedgehog in feather morphogenesis: induction of mesenchymal condensation and association with cell death. Dev Dyn 207: 157170.
  • Viallet JP, Prin F, Olivera-Martinez I, Hirsinger E, Pourquié O, Dhouailly D. 1998. Chick delta-1 gene expression and the formation of the feather primordia. Mech Dev 72: 159168.
  • Vogel A, Rodriguez C, Warnken W, Izpisua Belmonte JC. 1995. Dorsal cell fate specified by chick Lmx1 during vertebrate limb development [published erratum appears in Nature 1996;379:848]. Nature 378: 716720.
  • Widelitz RB, Jiang TX, Chen CW, Stott NS, Chuong CM. 1999. Wnt-7a in feather morphogenesis: involvement of anterior-posterior asymmetry and proximal-distal elongation demonstrated with an in vitro reconstitution model. Development 126: 25772587.
  • Wilkinson DG, Nieto MA. 1993. Detection of messenger RNA by in situ hybridization to tissue sections and whole mounts. Methods Enzymol 225: 361373.
  • Wurst W, Auerbach AB, Joyner AL. 1994. Multiple developmental defects in Engrailed-1 mutant mice: an early mid-hindbrain deletion and patterning defects in forelimbs and sternum. Development 120: 20652075.