Search

Every vertebrate species has its own unique morphology adapted to a particular lifestyle and habitat. Limbs and fins are strikingly diversified in size, shape, and position along the body axis. This diversity in morphology suggests the existence of a variety of embryonic developmental programs. However, comparisons of various embryos suggest common mechanisms underlying limb/fin formation. Here, we report the existence of continuous stripes of competency for appendage formation along the dorsal midline and the lateral trunk of all of the major jawed vertebrate (gnathostome) groups. We also show that the developing fin buds of cartilaginous fish share a mechanism of anterior-posterior axis formation as well as an shh (sonic hedgehog) expression domain in the posterior bud. We hypothesize a continuous distribution of competent stripes that represents the common developmental program at the root of appendage formation in gnathostomes. This schema would have permitted subsequent divergence into various levels of limbs/fins in each animal group.

The autopod, including the mesopodium and the acropodium, is the most distal part of the tetrapod limb, and developmental mechanisms of autopod formation serve as a model system of pattern formation during development. Cartilage rudiments of the autopod develop after proximal elements have differentiated. The autopod region is marked by a change in the expression of two homeobox genes: future autopod cells are first Hoxa11/Hoxa13-double-positive and then Hoxa13-single-positive. The change in expression of these Hox genes is controlled by upstream mechanisms, including the retinoic acid pathway, and the expression of Hoxa13 is connected to downstream mechanisms, including the autopod-specific cell surface property mediated by molecules, including cadherins and ephrins/Ephs, for cell-to-cell communication and recognition. Comparative analyses of the expression of Hox genes in fish fins and tetrapod limb buds support the notion on the origin of the autopod in vertebrates. This review will focus on the cellular and molecular regulation of the formation of the autopod during development and evolutionary developmental aspects of the origin of the autopod.

The Xenopus adult limb has very limited regeneration ability, and only a simple cartilaginous spike structure without digits is formed after limb amputation. We found that expression of Shh and its downstream genes is absent from the regenerating blastema of the Xenopus froglet limb. Moreover, we found that a limb enhancer region of the Shh gene is highly methylated in the froglet, although the sequence is hypomethylated in the Xenopus tadpole, which has complete limb regeneration ability. These findings, together with the fact that the promoter region of Shh is hardly methylated in Xenopus, suggest that regenerative failure (deficiency in repatterning) in the Xenopus adult limb is associated with methylation status of the enhancer region of Shh and that a target-specific epigenetic regulation is involved in gene re-activation for repatterning during the Xenopus limb regeneration process. Because the methylation level of the enhancer region was low in other amphibians that have Shh expression in the blastemas, a low methylation status may be the basic condition under which transcriptional regulation of Shh expression can progress during the limb regeneration process. These findings provide the first evidence for a relationship between epigenetic regulation and pattern formation during organ regeneration in vertebrates.

Understanding the mechanisms that control amphibian limb regeneration should allow us to decipher the critical differences between amphibians and humans, which have the limited ability of organ regeneration. However, many issues at the cellular and molecular levels still remain unresolved. We have generated a transgenic Xenopus laevis line that expresses green fluorescent protein (GFP) under the control of mouse prx1 limb enhancer, which directs reporter gene expression in limb mesenchyme in mice, and found that GFP accumulated in blastemal mesenchymal cells of the transgenic froglets after limb amputation. Thus, this transgenic line should provide a new approach to gain insights into the cellular dynamics and signaling pathways involved in limb blastema formation. We have also developed a culture system for forelimb explants of froglets and found that treatment with inhibitors of mitogen-activated protein kinase (MAPK)/extracellular signal-regulated protein kinase (ERK) kinase 1/2 (MEK1/2) and phosphatidylinositol 3-kinase (PI3K) repressed GFP expression. These effects were partially reversible, and down-regulation of GFP was associated with inhibition of cell-cycle progression and induction of ectopic apoptosis. In addition, we found that ERK1/2 and AKT, downstream mediators of MEK1/2 and PI3K pathways, were activated in amputated forelimb stumps. These results demonstrate that MEK/ERK and PI3K/AKT pathways regulate limb blastema formation in the X. laevis froglet.

Mammalians have a low potency for limb regeneration compared to that of amphibians. One explanation for the low potency is the deficiency of cells for regenerating amputated limbs in mammals. Amphibians can form a blastema with dedifferentiated cells, but mammals have few such cells. In this paper, we report limb formation, especially bone/cartilage formation in amputated limbs, because bone/cartilage formation is a basic step in limb pattern regeneration. After the amputation of limbs of a neonatal mouse, hypertrophy of the stump bone was observed at the amputation site, which was preceded by cell proliferation and cartilage formation. However, no new elements of bone/cartilage were formed. Thus, we grafted limb buds of mouse embryo into amputated limbs of neonatal mice. When the intact limb bud of a transgenic green fluorescent protein (GFP) mouse was grafted to the limb stump after amputation at the digit joint level, the grafted limb bud grew and differentiated into bone, cartilage and soft tissues, and it formed a segmented pattern that was constituted by bone and cartilage. The skeletal pattern was more complicated when limb buds at advanced stages were used. To examine if the grafted limb bud autonomously develops a limb or interacts with stump tissue to form a limb, the limb bud was dissociated into single cells and reaggregated before grafting. The reaggregated limb bud cells formed similar digit-like bone/cartilage structures. The reaggregated grafts also formed segmented cartilage. When the reaggregates of bone marrow mesenchymal cells were grafted into the stump, these cells formed cartilage, as do limb bud cells. Finally, to examine the potency of new bone formation in the stump tissue without exogenously supplied cells, we grafted gelatin gel containing BMP-7. BMP induced formation of several new bone elements, which was preceded by cartilage formation. The results suggest that the environmental tissues of the stump allow the formation of cartilage and bone at least partially, and that limb formation will be possible by supplying competent cells endogenously or exogenously in the future.

Limb regeneration in amphibians is a representative process of epimorphosis. This type of organ regeneration, in which a mass of undifferentiated cells referred to as the "blastema" proliferate to restore the lost part of the amputated organ, is distinct from morphallaxis as observed, for instance, in Hydra, in which rearrangement of pre-existing cells and tissues mainly contribute to regeneration. In contrast to complete limb regeneration in urodele amphibians, limb regeneration in Xenopus, an anuran amphibian, is restricted. In this review of some aspects regarding adult limb regeneration in Xenopus laevis, we suggest that limb regeneration in adult Xenopus, which is pattern/tissue deficient, also represents epimorphosis.

Xenopus laevis larvae can regenerate an exact replica of the missing part of a limb after amputation at an early limb bud stage. However, this regenerative capacity gradually decreases during metamorphosis, and a froglet is only able to regenerate hypomorphic cartilage, resulting in a spike-like structure (spike). It has been reported that the spike has tissue deformities, e.g., a muscleless structure. However, our previous study demonstrated that the muscleless feature of the spike can be improved. The existence of other kinds of tissue, such as tendon, has not been clarified. In this study, we focused on the tendon and dermis, and we isolated the scleraxis and dermo-1 genes, which are known to be marker genes for the tendon and dermis, respectively. The expressions of these genes were investigated in both the developmental and regenerating processes of a Xenopus limb. Although muscle was needed to maintain scleraxis expression, scleraxis transcription was detectable in the muscleless spike. Additionally, although grafting of matured skin, including dermal tissue, inhibited limb regeneration, the expression of dermo-1, a dermal marker gene, was detected from the early stage of the froglet blastema. These results indicate that tendon precursor cells and dermal cells exist in the regenerating froglet blastema. Our results support the idea that spike formation in postmetamorphic Xenopus limbs is epimorphic regeneration., ((c) 2006 Wiley-Liss, Inc.)

In Xenopus laevis, amputation of the adult limb results in the formation of a simple (hypomorphic) spike-like structure without joints, although tadpole limb bud regenerates complete limb pattern. The expression of some joint marker genes was examined in limb development and regeneration. Bmp-4 and gdf-5 were expressed and sox-9 expression was decreased in the joint region. Although developing cartilages were well-organized and had bmp-4 expressing perichondrocytes, the spike cartilage did not have such a structure, but only showed sparse bmp-4 expression. Application of BMP4-soaked beads to the spike led to the induction of a joint-like structure. These results suggest that the lack of joints in the spike is due to the deficiency of the accumulation of the cells that express bmp-4. Improvement of regeneration in the Xenopus adult limb that we report here for the first time will give us important insights into epimorphic regeneration., ((c) 2005 Wiley-Liss, Inc.)

Positioning of the limb is one of the important events for limb development. In the early stage of embryogenesis, the lateral plate mesoderm splits into two layers and the dorsal layer (the somatic mesoderm) gives rise to a series of distinct structures along the rostrocaudal axis, including the forelimb bud, flank body wall, and hindlimb bud. To determine whether positional information in the somatic mesoderm for regionalization along the rostrocaudal axis is also inherited by the ventral layer of the lateral plate mesoderm (the splanchnic mesoderm), experiments in which the splanchnic mesoderm was transplanted under the ectoderm in an in ovo chick system were carried out. Transplantation of the wing-, flank-, and leg-level splanchnic mesoderm resulted in the formation of wings, nothing, and legs, respectively. These results suggest that the splanchnic mesoderm possesses the ability to form limbs and that the ability differs according to the position along the rostrocaudal axis. The position-specific ability to form limbs suggests that there are some domains involved in the formation of position-specific structures in the digestive tract derived from the splanchnic mesoderm, and results of cell fate tracing supported this possibility. In contrast, analysis of shh expression suggested that the anteroposterior polarity in the limb region seems not to be inherited by the splanchnic mesoderm. We propose that the positioning of limb buds is specified and determined in the very early stage of development of the lateral plate mesoderm before splitting and that the polarity in a limb bud is established after the splitting of the mesoderm., (Copyright 2005 Wiley-Liss, Inc)

We previously showed that recombinant limb buds with dissociated and reaggregated mesenchyme develop more than 30 digits in Xenopus laevis, which exhibits different capacities for limb regeneration at different developmental stages (Yokoyama et al. [1998] Dev Biol 196:1-10). Cell-cell contact among anterior- and posterior-derived mesenchymal cells is required for anteroposterior (AP) axis formation of recombinant limbs in an intercalary manner. However, whether one-way induction from posterior cells to anterior cells as proposed by the polarizing zone model or interactions between anterior and posterior cells evoke the AP axis formation in recombinant limbs remains unclear. In this study, we found, by a combination of X-ray irradiation and a recombinant limb technique, that not one-way induction but interactions between anterior and posterior cells accompanied by cell contribution are indispensable for AP axis formation in recombinant limbs. Shh was expressed in posterior-derived not anterior-derived cells. We propose that the recombinant limb is an excellent model for examining the axis formation mechanism in regenerating limbs because, as in recombinant limbs, cell-cell contact among cells derived from different positions of an amputation plane occurs in the blastema of regenerating limbs., (Copyright 2002 Wiley-Liss, Inc.)

Stage-dependent cell sorting in vitro is an intriguing property that mesenchymal cells of a chick limb bud have. We previously proposed that N-cadherin, a cell adhesion molecule, is involved in the sorting process and is likely to be a component of the mechanism of proximal-distal patterning in the developing limb (Yajima et al., (1999) Dev. Dynam. 216:274-284). Here, we present more direct evidence that N-cadherin is one of the molecules responsible for regulation of stage-dependent cell sorting in vitro. Our results suggest that N-cadherin, which accumulates in the distal region of the chick limb bud as limb development proceeds, is related to the positional identity that gives rise to the different shapes and numbers of cartilaginous elements along the proximal-distal axis. In this article we also give insights into positional identity which is mediated by Hoxgenes and cell surface property during limb development.

Limb-type-specific expression of Tbx5/Tbx4 plays a key role in drawing distinction between a forelimb and a hindlimb. Here, we show insights into specification and determination during commitment of limb-type identity, in particular that median tissues regulate Tbx expressions. By using the RT-PCR technique on chick embryos, the onset of specific Tbx5/Tbx4 expression in the wing/leg region was estimated to be stage 13. Specification of the limb-type identity is thought to occur before stage 9, since all explants from stage 9 through 14 expressed the intrinsic Tbx gene autonomously in a simple culture medium. The results of transplantation experiments revealed that axial structures medial to the lateral plate mesoderm at the level of the wing region are capable of transforming leg identity to wing identity, suggesting that a factor(s) from the median tissues is involved in the limb-type determination. Nevertheless, the transplanted wing region was not converted to leg identity. The results of the transplantation experiments also suggested that wing-type identity is determined much earlier than is leg-type identity. Finally, we also found that inhibitory effects of median tissues mediate the specific expression of Tbx5/Tbx4 in the presumptive wing/leg region. We propose a model for limb-type identification in which inhibitory regulation is involved in restricting one Tbx gene expression by masking the other Tbx expression there.

Emerging developmental studies contribute to our understanding of vertebrate evolution because changes in the developmental process and the genes responsible for such changes provide a unique way for evaluating the evolution of morphology. Endoskeletal limbs, the locomotor organs that are unique to vertebrates, are a popular model system in the fields of palaeontology and phylogeny because their structure is highly visible and their bony pattern is easily preserved in the fossil records. Similarly, limb development has long served as an excellent model system for studying vertebrate pattern formation. In this review, the evolution of vertebrate limb development is examined in the light of the latest knowledge, viewpoints and hypotheses.

To clarify the roles of fibroblast growth factors (FGF) in limb cartilage pattern formation, the effects of various FGF on recombinant limbs that were composed of dissociated and reaggregated mesoderm and ectodermal jackets were examined. Fibroblast growth factor-soaked beads were inserted just under the apical ectodermal ridge (AER) of recombinant limbs and the recombinant limbs were grafted and allowed to develop. Control recombinant limbs without FGF beads formed one or two cartilage elements. Recombinants with FGF-4 beads formed up to five cartilage elements, which were aligned along the anteroposterior (AP) axis. Each cartilage element showed digit-like segmentation. In contrast, recombinants with FGF-2 beads showed formation of multiple thick and unsegmented cartilage rods, which elongated inside and outside the AP plane from the distal end of the recombinants. Recombinants with FGF-8 beads formed a truncated cartilage pattern and recombinants with FGF-10 beads formed a cartilage pattern similar to that of the control recombinants. The expression of the Fgf-8, Msx-1 and Hoxa-13 genes in the developing recombinant limbs were examined. FGF-4 induced extension of the length of the Fgf-8-positive epidermis, or AER, along the AP axis 5 days after grafting, at which time the digits are specified. FGF-2 induced expansion of the Msx-1-positive area, first in the proximal direction and then along the dorsoventral axis. The functions of these FGF in recombinant and normal limb patterning are discussed in this paper.

During vertebrate limb development, the apical ectodermal ridge (AER) plays a vital role in both limb initiation and distal outgrowth of the limb bud. In the early chick embryo the prelimb bud mesoderm induces the AER in the overlying ectoderm. However, the direct inducer of the AER remains unknown. Here we report that FGF7 and FGF10, members of the fibroblast growth factor family, are the best candidates for the direct inducer of the AER. FGF7 induces an ectopic AER in the flank ectoderm of the chick embryo in a different manner from FGF1, -2, and -4 and activates the expression of Fgf8, an AER marker gene, in a cultured flank ectoderm without the mesoderm. Remarkably, FGF7 and FGF10 applied in the back induced an ectopic AER in the dorsal median ectoderm. Our results suggest that FGF7 and FGF10 directly induce the AER in the ectoderm both of the flank and of the dorsal midline and that these two regions have the competence for AER induction. Formation of the AER of the dorsal median ectoderm in the chick embryo is likely to appear as a vestige of the dorsal fin of the ancestors., (Copyright 1999 Academic Press.)

Although regional differences in mesenchymal cell affinity in the limb bud represent positional identity, the molecular basis for cell affinity is poorly understood. We found that treatment of the cell surface with bacterial phosphatidylinositol-specific phospholipase C (PI-PLC) could change cell affinity in culture. When PI-PLC was added to the culture medium, segregation of the progress zone (PZ) cells from different stage limb buds was inhibited. Similarly, sorting out of the cells from different positions along the proximodistal (PD) axis of the same stage limb buds was disturbed. Since PI-PLC can remove glycosylphosphatidylinositol (GPI)-anchored membrane bound proteins from the cell surface, the GPI-anchored cell surface proteins may be involved in sorting out. To define the GPI-anchored molecules that determine the segregation of limb mesenchymal cells, we examined the effect of neutralizing antibody on the EphA4 receptor that binds to GPI-anchored cell surface ligands, called ephrin-A. Sorting out of the PZ cells at different stages could be inhibited by the neutralizing antibody to EphA4. These results suggest that EphA4 and its GPI-anchored ligands are, at least in part, involved in sorting out of limb mesenchymal cells with different proximal-distal positional values, and that GPI-anchored cell surface proteins play important roles in determining cell affinity in the limb bud., (Copyright 1998 Academic Press.)

We prepared recombinant limb buds of Xenopus tadpoles by grafting a mesenchyme mass of the hindlimb bud. The Xenopus recombinant limb buds with dissociated and reaggregated mesenchyme developed more than 30 digits with cartilage segmentation, while those with undissociated mesenchyme developed a limb with normal cartilage pattern. Before the formation of multiple digits, a patchy expression pattern of fgf-8, an AER marker, was observed in the distal region of recombinant limb buds. shh, a ZPA (zone of polarizing activity) marker, was expressed broadly in the distal region of recombinants. Recombinant limb buds with the reaggregated mesenchyme of anterior halves formed anterior digits with claws, and those with the mesenchyme of posterior halves formed posterior digits without claws. The temporal and spatial changes in the potency of multiple digit formation are discussed with reference to the regenerative capacity of Xenopus limb buds., (Copyright 1998 Academic Press.)

The apical ectodermal ridge (AER) is a specialized thickening of the distal limb ectoderm, and its signals are known to support limb morphogenesis. The expression of a homeobox gene, Msx1, in the distal limb mesoderm depends on signals from the AER. In the present paper it is reported that Msx1 expression in the distal mesoderm is necessary for the transfer of AER signals in chick limb buds. Interruption of AER-mesoderm interaction by insertion of a thick filter led to the inhibition of pattern specification in the mesoderm just under the filter. In such cases, the expression of Msx1 disappeared in the mesoderm under the filter, suggesting that AER is able to signal over short ranges. In advanced limb buds, Msx1 is also expressed in the proximal mesoderm under the anterior ectoderm. However, it was found that a grafted antero-proximal mesoderm shows no inhibitory effects on pattern specification of the host mesoderm, as is the case with the distal mesoderm. On the other hand, grafted mesoderms without potent Msx1 re-expression, even underneath AER, disturbed normal limb development. In such cases, the expression of Msx1 disappeared in the mesoderm under the grafts, whereas Fgf-8 expression was maintained in the AER above the graft. These results indicate that the expression of Msx1 in the mesoderm is important for the transfer of AER signals.

Apical ectodermal ridge (AER)-mesoderm interaction is important for morphogenesis in the developing chick limb bud. Genes whose expression is dependent upon the presence of AER, are likely to play important roles in the AER-mesoderm interaction. We report here the gene expression pattern of the chick homolog of the 67 kDa laminin binding protein (LBP), which is a non-integrin laminin receptor whose function relates to cell attachment, spreading, and polarization. Northern analysis showed that a single 1.4 kb transcript exists in stage 20 limb buds and which is dramatically reduced 24 hr after removal of AER. In situ hybridization analysis revealed that the chick 67 kDa laminin binding protein gene (cLbp) was expressed in the mesodermal region overlapping the Msx1-expressing domain and in the AER in early stage limb buds. Expression in the mesoderm was gradually restricted to the distal region underneath the AER as development proceeds. The expression in the limb mesoderm could be induced by local application of FGF-2 which could thus mimic the AER functions. These results indicated that the expression of cLbp depends on AER signals and that the 67 kDa non-integrin receptor binding to laminin plays a role in the AER-mesoderm interaction.

The zone of polarizing activity (ZPA) is a specialized region involved in the antero-posterior (A-P) axis formation in chick and mouse limb buds. The existence of ZPA in the posterior margin is suggested in Xenopus hindlimb buds because 180 degree rotation of the distal limb tip induces the supernumerary limb. In this study, we investigated the expression of Sonic hedgehog (shh), a molecular marker for ZPA, in Xenopus developing limb buds and regenerating blastemas by whole-mount in situ hybridization. Although shh was expressed in the posterior margin of the limb bud like in chicks, its expression domain did not correspond to the ZPA map of Xenopus hindlimb buds. shh expression was distant from the ZPA at stage 53 in particular. To clarify the difference between the shh expression domain and ZPA, we examined shh expression in 180 degree rotated limb buds. As a result, ectopic shh expression was newly induced in the proximal region to its original expression domain. These results suggest that ZPA is accompanied by shh expression as in chick limb buds. Furthermore we examined shh expression in regenerating blastemas. shh was reexpressed in the posterior margin of the blastema. This result supports the possibility that ZPA also exists in the regenerating blastema.

We have previously shown that MRC-5 cells induce the duplication of the chick limb bud following the implantation into the anterior limb bud only during pre-limb-bud stages. We now report the process of duplicated pattern formation caused by MRC-5 cells. The duplicated patterns are also formed following the implantation into the center of the limb bud and an excess apical ectodermal ridge (AER) with Msx2 expression is induced prior to these duplicated pattern formulations. Only after the implantation into the anterior leg bud, the shh gene is expressed additionally in the anterior leg bud and the mirror-symmetric duplication along the anteroposterior (A-P) axis is formed. The map of the polarizing activity in stage 21 embryo suggests that the high polarizing activity of the normal flank region is responsible for the changes in the A-P polarity when MRC-5 cells are grafted into the anterior leg bud. These results indicate that MRC-5 cells induce the AER and that the excess AER produces the duplicated cartilage pattern of the limb bud.

The anteroposterior (A-P) axis pattern of the chick limb is likely to be controlled by a small region of mesenchyme cells at the posterior margin of the limb bud (ZPA, zone of polarizing activity). In this study, we found that MRC-5 fibroblast cells had the capacity for duplicated-pattern formation of the chick limb along the A-P axis when grafted to the anterior region of the limb bud. MRC-5 cells were effective only during pre-limb bud stages, and the leg bud was more responsive than the wing bud. Grafted cells remained at the base of the limb bud when limb development proceeded. These results suggest that the products of MRC-5 cells are involved in three possible processes of duplicated-pattern formation: induction of the polarizing activity; maintenance of this activity, which is present weakly at pre-limb bud stages; and determination of A-P axis as the ZPA factor(s). By allowing the use of a small number of cells of embryonic tissues with polarizing activity, the analysis of duplicate formation with the MRC-5 cell line provides a powerful tool for elucidating the molecular nature of the ZPA.

The apical region of chick limb buds (progress zone, PZ) is known to be essential for limb pattern formation. In the PZ, the cells are in an uncommitted state and change their positional values under the influence of the apical ectodermal ridge (AER). Using a culture system of chick PZ cells, we have examined the factor(s) which helps to maintain some of the characteristics of PZ cells. It was found that basic FGF (bFGF) stimulated the expression of AV-1 and Msx 1 mRNA, which are rich in PZ cells at early stages, in the cultured PZ cells. These two molecules were rapidly lost in the control culture without the bFGF. Further, bFGF promoted growth and inhibited chondrogenesis of the cultured PZ cells. When quail PZ cells were cultured in FGF medium, harvested, and grafted to the apical region of chick limb buds, in about half of the grafts, the cells distributed to form a long band along the proximodistal axis and participated in normal host cartilage pattern formation as did intact PZ tissues. In contrast, those cultured in the medium without bFGF distributed as a cluster and did not participate in host cartilage pattern formation. The culture with conditioned medium of BRL cells increased the percentage of the grafts participating in normal pattern formation. These results indicate that bFGF helps to maintain some of the characteristics of PZ cells in culture and suggest that bFGF may be a PZ maintenance factor from the AER.

Retinoic acid (RA) is known to mimic the action of the zone of polarizing activity (ZPA) in inducing the formation of anteroposterior duplicates in the chick limb bud. Although RA had been thought to be a morphogen produced by the ZPA, recently we (Noji et al., 1991) and Wanek et al. (1991) independently concluded that RA induces polarizing activity in cells at the anterior margin of the chick limb bud. In this study, we examined the distribution of RA-induced polarizing activity in the limb bud. At first, RA-containing beads were implanted into various regions of stages 20-23 limb buds. After 24 hr, we grafted tissue fragments from sites adjacent to the RA beads into the anterior margin of host limb buds. High polarizing activity was induced in anterior-distal and apical mesoderm regions immediately under the apical ectodermal ridge (AER) when beads presoaked in 1 mg/ml RA were grafted to anterior sites at stages 19-20. However, in mesoderms distant from the AER, even if adjacent to the implanted RA bead, only low activity was induced. Although implantation of beads soaked in a high concentration (10 mg/ml) of RA did not result in duplications but rather in truncations of the limbs, the apical mesodermal cells in these limbs were converted to ZPA cells, indicating that induction of polarizing activity by RA can occur independently of duplicate formation. The ability to respond to RA treatment was rapidly lost in the cells along the anterior margin of stage 22 limb buds. When the anterior AER was removed, the induction of polarizing activity in the anterior margin of limb buds was markedly reduced. Thus, the AER seems to be necessary for the RA-directed induction of polarizing activity.

Local application of retinoic acid (RA) to the anterior-distal margin of chick limb buds causes digit pattern duplications identical to those that result from zone of polarizing activity (ZPA) grafts. The AV-1 antigen is a cell membrane glycoprotein which is expressed during chick limb bud development in a stage- and position-specific manner that correlates with cartilage pattern formation. In order to further understand the limb patterning mechanisms affected by RA and by ZPA grafts, we first examined the effects of RA treatment on AV-1 expression. We then compared the effect of RA with those of ZPA grafts. When beads soaked in varying concentrations of RA were applied to the anterior-distal margin of wing buds, changes in the pattern of AV-1 antigen expression were observed. AV-1 expression appears to correlate with regions that will give rise to a digit 3. These results suggest that the AV-1 protein is one of the molecules which may be involved in the process of cartilage pattern formation during limb development. ZPA grafts produced similar alterations in AV-1 expression, but the initial changes were observed earlier with ZPA grafts than with RA beads. These results support the idea that RA changes anterior limb bud cells into ZPA cells.

When a mouse zone of polarizing activity (ZPA) at the posterior margin of the limb bud was grafted at the anterior margin of the chick limb bud, we found that expression of the chick homeobox genes, Chox-4.7 and -4.8, was induced prior to the formation of chick extra digits. The induction was observed in the restricted domain close to both grafted mouse ZPA and the chick apical ectodermal ridge (AER). When the posterior half of the AER was removed, the normal expression was diminished in the posterodistal region. Thus, it is likely that at least two distinct factors, one from the ZPA and the other from the AER, provide positional information to induce cooperatively the sequential expression of the Chox-4 genes.

Retinoic acid is a putative morphogen in limb formation in the chick and other vertebrates. In chick limb formation, it is thought that retinoic acid is released from the zone of polarizing activity (ZPA) and the concentration gradient of retinoic acid formed from the posterior to the anterior provides positional cues for digit formation. Implantation of a bead containing retinoic acid at the anterior margin of the limb bud induces a mirror-image symmetrical duplication of the digit pattern similar to that observed when the ZPA is grafted into the anterior margin of the host limb bud. Also, the level of endogenous retinoic acid (25 nM on average) is higher in the posterior one third of the limb bud. We found that when the bead containing either retinoic acid or an analogue but not the ZPA, was implanted in the anterior margin of the chick limb bud, expression of the retinoic acid receptor type-beta gene was induced around the bead within 4 h. These results indicate that exogenous retinoic acid is not identical with the ZPA morphogen. As the anterior tissue exposed to retinoic acid has polarizing activity, we conclude that the primary function of exogenous retinoic acid is to induce polarizing activity in the limb bud.

Retinoic acid induces anteroposterior duplicate formation in developing chick limb bud, and it may be a natural morphogen involved in limb pattern formation. Retinoic acid is produced from retinol locally in the limb bud via retinal, and thus, to elucidate the distribution of these retinoids in the limb bud seems to be important for the understanding of the morphogen formation. We produced a monoclonal antibody against the retinoids with BSA-RA (bovine serum albumin-retinoic acid) conjugate for antigen, and investigated the distribution of retinoids in the chick limb bud. The antibody predominantly bound to retinoic acid, but weakly to retinol and retinal. Retinoids appeared in the limb bud at stage 18 and were distributed through stages 20-24, when the pattern formation in distal mesoderm was in progress. Initially they were found evenly in the whole mesoderm, but disappeared gradually from core mesoderm and remained only in the region of peripheral mesoderm at stage 24. At stage 26, retinoids were detected only in ectoderm. These results support the idea that the retinoids actually play roles in limb pattern formation and suggest that the retinoids in the peripheral mesoderm are important for pattern formation. Further, the role of retinoids in epidermis development at later limb bud stages is also suggested.

Limb bud mesoderm of stage 22-23 embryos was dissected into four pieces along the anteroposterior axis and dissociated cells of each region were separately cultured under various conditions. When the cells were cultured in medium containing 0.1% fetal calf serum (serum-poor medium) only a slight increase in cell number occurred in the cultures of all four regions. However, when the cells were cultured in medium containing 10% FCS, only cells of two central regions proliferated rapidly, and no growth promotion was observed in cells in the most anterior and posterior regions. Using the serum-poor medium, we examined the growth-promoting effects of cocultured limb bud fragments and of some growth factors on the cells of four regions. Anterior, distal, and proximal fragments promoted cell proliferation and their promotive effect on the cells of each region was equal. On the other hand, posterior fragments (containing ZPA) showed stronger promotive effects on preaxial cells than on postaxial cells. For comparison with the growth-promotive effect of the posterior fragment, fibroblast growth factor (FGF), epidermal growth factor (EGF), insulin, and retinoic acid were tested in cell culture. FGF showed position-dependent growth promotion, while EGF and insulin promoted growth in the cells of all four regions to a similar degree. Retinoic acid showed no effect on cell growth at low concentrations, and was rather toxic at high concentrations. These results suggest that the cells of the posterior region secrete an FGF-like growth factor(s), which controls normal limb development and experimental duplicate formation.

Distal and proximal mesoderm of chick limb bud was respectively dissociated and cultured in the medium containing various concentrations of retinoic acid (RA). At low concentrations (5-50 ng/ml), RA promoted proliferation and chondrogenesis in the distal mesodermal cells. The distal cells of stage 20-24 limb buds were responsive to RA, although those of stages 25-27 were unresponsive. Both the cells of anterior and posterior regions of the distal mesoderm were responsive to RA, while the cells of proximal mesoderm were unresponsive. At higher concentrations, the growth-promoting effect of RA was reduced and chondrogenesis in the distal cells was rather inhibited. These results were discussed in relation to the role of RA as the morphogen in normal limb development and experimental duplicate formation.

To analyze the molecular mechanism of the limb pattern formation, we have tried to make monoclonal antibodies against antigens from chick limb buds. We obtained one antibody named AV-1 which recognized a specific region of chick limb buds. AV-1 reacted with the distal portion of the anteroventral mesoderm of only developmentally early chick limb buds. Grafts of ZPA region tissue to an anterior site in an embryonic chick wing bud resulted in mirror-image dupliction of the AV-1 antigen region. These data show the possibility that this antigen plays some role in the limb pattern formation. This is the first evidence that a position specific substance really exists in developmentally early limb buds in which the pattern has been considered to be unspecified.

We have previously succeeded in culturing whole leg bud from stage 21-23 chick embryos and observed a leg structure with typical cartilage pattern in vitro. In the present study, we have attempted the organ culture of the fragmented leg bud and investigated its capacity to form cartilage. Leg buds from stages 17-21 chick embryos were dissected into four pieces in the anteroposterior sequence (named 1, 2, 3, and 4, respectively) and cultured on a membrane filter in a medium consisting of Ham's F-12, chick serum, and chick embryo extract. After 6 days in culture, two central fragments (2 and 3) developed into large cartilaginous masses, while anterior (1) and posterior (4) fragments formed few or small cartilaginous masses. In addition, when these less chondrogenic fragments were combined, pinned together, and cultured, large cartilaginous masses were formed from 1 + 4 combinations but not from 1 + 1 or 4 + 4 combinations. These observations were analyzed quantitatively by measurement of 35SO4 incorporation into the sulfated glycosaminoglycan (S-GAG) and of final DNA content per explant, and by histological reconstruction of the chick-quail chimera explant. The results showed that (a) the 1 + 4 combination resulted in higher S-GAG synthesis and final DNA content than the 1 + 1 or 4 + 4 combinations in stage 18 and 21 leg buds (P less than 5%); (b) removal of ectoderm from the leg bud inhibited the increase observed for the 1 + 4 combination; c) in chick-quail chimera explants the cartilage formed from the 1 + 4 combination was largely of fragment 1 origin. These results demonstrate, first, the presence of a difference in chondrogenic capacity along the anteroposterior axis in the leg bud and, second, the occurrence of an interaction between anterior and posterior fragments which mimics the effects of grafting a zone of polarizing activity (ZPA). The mechanism of ZPA function is still unknown but the ectoderm may play some role. Some roles for ectoderm in ZPA function and differences in mesodermal responsiveness to ZPA factor(s) are suggested.

In a previous study, we demonstrated the presence of a position-specific antigen (AV-1) in chick limb buds at an early developmental stage. Here, we reported the temporal and spatial expressions and the biochemical characterization of the AV-1 antigen. Indirect immunofluorescence staining and immunoblot analysis clearly showed that the AV-1 antigen is a glycoprotein that is localized on the plasma membrane and that it is expressed from stage 19 and highly expressed at stages 22-26 in some middle-distal to anterior-distal region of limb buds. In the wing bud, at stage 28, the AV-1 antigen was faintly detected in the restricted space between the precartilaginous regions of the radius and the ulna, and those of the metacarpals 2 and 3, but not those of the metacarpals 3 and 4. Such stage-specific and "position-specific" expressions of the AV-1 antigen in limb buds strongly suggest that the AV-1 antigen or cells containing it are involved in determination of the limb pattern formation.