DPP homologs and morphogenesis: limb patterning

The signaling molecule encoded by Sonic hedgehog participates in the patterning of several embryonic structures including limbs. During early fin development in zebrafish, a subset of cells in the posterior margin of pectoral fin buds express shh. Regulation of shh in pectoral fin buds is consistent with a role in mediating the activity of a structure analogous to the zone of polarizing activity (ZPA). During growth of the bony rays of both paired and unpaired fins, and during fin regeneration, there does not seem to be a region equivalent to the ZPA and one would predict that shh would play a different role, if any, during these processes specific to fish fins. The expression of shh was examined in the developing fins of 4-week old larvae and in regenerating fins of adults. A subset of cells in the basal layer of the epidermis in close proximity to the newly formed dermal bone structures of the fin rays, the lepidotrichia, express both shh and ptc1 (which is thought to encode the receptor of the SHH signal). The expression domain of ptc1 is broader than that of shh; adjacent blastemal cells releasing the dermal bone matrix also express ptc1. Further observations indicate that the bmp2 gene, in addition to being expressed in the same cells of the basal layer of the epidermis as shh, is also expressed in a subset of the ptc1-expressing cells of the blastema. Amputations of caudal fins immediately after the first branching point of the lepidotrichia, and global administration of all-trans-retinoic acid, two procedures known to cause fusion of adjacent rays, result in a transient decrease in the expression of shh, ptc1 and bmp2. The effects of retinoic acid on shh expression occur within minutes after the onset of treatment, suggesting direct regulation of shh by retinoic acid. These observations suggest a role for shh, ptc1 and bmp2 in the patterning of the dermoskeleton of developing and regenerating teleost fins (Laforest, 1998).

Pax1 expression in vertebrate limb buds is confined to cells in a discrete anterior proximal domain. In dorsoventral patterning of Drosophila, expression of pox meso, an insect gene with high sequence similarity to Pax1, is repressed by decapentaplegic (dpp) in dorsal mesoderm and, thus, is restricted to a discrete ventral domain. In the chick wing, cells expressing a vertebrate homolog of dpp, bone morphogenetic protein 4 (Bmp4), abut the Pax1 domain, suggesting a similar relationship between homologous genes in both vertebrates and invertebrates. Two BMPs (BMP4, and BMP2, also highly related to dpp) can repress Pax1 in the developing chick wing. Chick wing bud cells expressing Pax1 give rise to the shoulder girdle. Cells in an equivalent position in the mouse forelimb also express Pax1, and Pax1 mutant mice display shoulder girdle defects. Similarly, in chick embryos, girdle defects are produced by treatments with signaling molecules that lead to the expression of BMPs, which subsequently reduce Pax1 expression in the limb bud. BMP4 inhibits Pax1 expression in the developing trunk and Pax9 expression in developing teeth. Thus, a property of BMPs appears to be to regulate pox meso homologs negatively and, thus, limit their expression domains (Hofmann, 1998).

During limb development, the mesenchymal cells in restricted areas of the limb bud, the anterior necrotic zone, the posterior necrotic zone, the opaque zone and the interdigital necrotic zones, are all eliminated by programmed cell death. The transcripts of bone morphogenetic protein Bmp-2 and Bmp-4 are first detected in the areas where cell death is observed, then they show overlapping expression with the programmed cell death zones, except for the opaque zone. To investigate the function of BMP-2 and BMP-4 during limb pattern formation, a dominant negative form of BMP receptor (dnBMPR-1a) was overexpressed in chick leg bud via a replication-competent retrovirus to block the endogenous BMP-2/-4 signaling pathway. This results in excess web formation at the anterior and posterior regions of limb buds in addition to marked suppression of the regression of webbing at the interdigital regions. Significant reductions in the number of apoptotic cells in these three necrotic zones are found in the limb buds that receive the virus carrying dominant negative BMP receptor. This indicates that extra tissue formation is due to suppression of programmed cell death in the three necrotic zones. Moreover, BMP-2/-4 protein induces apoptosis of mesenchymal cells isolated from the interdigital region in vitro. Other TGFbeta family proteins as TGFbeta1 and Activin did not show this effect. These results suggest that BMP-2 and BMP-4 are the apoptotic signal molecules of the programmed cell death process in the chick limb buds (Yokouchi, 1996).

Mice lacking the RARgamma gene and one or both alleles of the RARbeta gene (i.e., RARbeta+/-/RARgamma-/- and RARbeta-/-/RARgamma-/- mutants) display a severe and fully penetrant interdigital webbing (soft tissue syndactyly), caused by the persistence of the fetal interdigital mesenchyme. These compound mutants were used to investigate the cellular and molecular mechanisms involved in retinoic acid (RA)-dependent formation of the interdigital necrotic zones (INZs). The mutant INZs show a marked decrease in the number of apoptotic cells accompanied by an increase of cell proliferation. This marked decrease is not paralleled by a reduction of the number of macrophages, indicating that the chemotactic cues that normally attract these cells into the INZs are not affected. The expression of a number of genes known to be involved in the establishment of the INZs, the patterning of the autopod, and/or the initiation of apoptosis is also unaffected. These genes included BMP-2, BMP-4, Msx-1, Msx-2, 5' members of Hox complexes, Bcl2, Bax, and p53. In contrast, the mutant INZs display a specific, graded, down-regulation of tissue transglutaminase (tTG) promoter activity and of stromelysin-3 expression upon the removal of one or both alleles of the RARbeta gene from the RARgamma null genetic background. Because retinoic acid response elements are present in the promoter regions of both tTG and stromelysin-3 genes, it is proposed that RA might increase the amount of cell death in the INZs through a direct modulation of tTG expression and that it also contributes to the process of tissue remodeling, which accompanies cell death, through an up-regulation of stromelysin-3 expression in the INZs. Expression of tTG is known to be a precocious feature of cells that are committed to apoptosis: tTG catalyzes the irreversible cross-linking of intracellular proteins. Stromolysin-3 is a matrix metalloproteinase whose expression has been associated with tissue remodeling processes characterized by extensive extracellular matrix turnover during embryonic development, wound healing, or tumor invasion. Approximately 10% of the RARbeta-/-/RARgamma-/- mutants display a supernumerary preaxial digit on hindfeet, which is also a feature of the BMP-7 null phenotype. BMP-7 is globally down-regulated at an early stage in the autopods of these RAR double null mutants, prior to the appearance of the digital rays. Therefore, RA may exert some of its effects on anteroposterior autopod patterning through controlling BMP-7 expression (Dupe, 1999).

In an effort to define the roles of bone morphogenic proteins (BMPs) and fibroblast growth factors (FGFs) during chick limb development more closely, beads impregnated with these growth factors were implanted into chick limb buds between stages 20 and 26. Embryos were sacrificed at the time the bone chondrocyte condensations first appear (stages 27-28). Implantation of beads containing BMPs at the earlier stages (20-22) causes apoptosis to occur, in the most severe cases leading to complete limb degeneration. Application of FGF4, either in the same, or in different beads, prevents the BMP-induced apoptosis. It is argued that the apoptosis observed on removal of the AER prior to stage 23 of development is brought about by BMPs. The action of epithelial FGF in preventing BMP-mediated apoptosis in the mesenchyme would define a novel aspect of epithelial-mesenchymal interactions. Implanting the BMP4 beads into the core of the limb bud a day later (stages 25-26) causes intense chondrogenesis rather than apoptosis. FGF4 can nullify this effect and by itself causes a reduction in bone size. This is the reverse of the functional relationship these growth factors have in mouse tooth specification (where it is BMP4 that inhibits the FGF8 function), and suggests that the balance between the effects of FGFs and BMPs controls the size of the chondrocyte precursor cell pool. In this way members of these two growth factor families control the size of appendages when they are initially formed (Buckland, 1998).

In the final stages of limb morphogenesis, the autopodial cells, whose fate is to form the digits, differentiate as they leave the progress zone: they will become cartilage or undergo apoptotic cell death, depending on whether they are incorporated into the digital rays or interdigital spaces. Most evidence indicates that these two opposite fates of the autopodial mesoderm are controlled by BMP signaling. However, some controversy concerns the molecular basis for these two distinct actions of BMPs, including the receptors involved in the process. This question was addressed by exploring the presence in the developing autopod of diffusible signals able to modulate BMP function and by analyzing the effects of their exogenous administration on the pattern of expression of BMP receptor genes. Tgfbeta2 and noggin genes are expressed in the condensing region of the developing digital rays in addition to the well-known distribution in the autopodial tissues of FGFs (apical ectodermal ridge, AER) and BMPs (AER, progress zone mesoderm, and interdigital regions). Exogenous administration of all the factors causes changes in the expression of the bmpR-1b gene, which is followed by parallel alterations of the skeletal phenotype: FGFs inhibit the expression of bmpR-1b compatible with their function in the maintenance of the progress zone mesoderm in an undifferentiated state; TGFbetas induce the expression of bmpR-1b and promote ectopic chondrogenesis, compatible with a function in the establishment of the position of the digital rays. Evidence is provided for the occurrence of an interactive loop between BMPs and noggin, accounting for the spatial distribution of bmpR-1b, which may control the size and shape of the skeletal pieces. In contrast to the bmpR-1b gene, the bmpR-1a gene is expressed at low levels in the autopodial mesoderm and its expression is not modified by any of the tested factors regardless of their effects on chondrogenesis or cell death. The role of BMPs in programmed cell death is confirmed here by the intense inhibitory effect of noggin on apoptosis, but the lack of correlation between changes in the pattern of cell death induced by treatment with the studied factors and the expression of either bmpR-1a or bmpR-1b genes suggest that a still-unidentified BMP receptor may account for this BMP function (Merino, 1998).

The expression and function of Gremlin in the developing avian limb has been examined. Gremlin is a member of the DAN family of BMP antagonists; it appears to be highly conserved through evolution and is able to bind and block BMP2, BMP4 and BMP7. At early stages of development, gremlin is expressed in the dorsal and ventral mesoderm in a pattern complementary to that of bmp2, bmp4 and bmp7. The maintenance of gremlin expression at these stages is under the control of the AER, ZPA, and BMPs. Exogenous administration of recombinant Gremlin indicates that this protein is involved in the control of limb outgrowth. This function appears to be mediated by the neutralization of BMP function to maintain an active AER, to restrict the extension of the areas of programmed cell death and to confine chondrogenesis to the central core mesenchyme of the bud. During the stages of digit formation, gremlin is expressed in the proximal boundary of the interdigital mesoderm of the chick autopod. The anti-apoptotic influence of exogenous Gremlin, which results in the formation of soft tissue syndactyly in the chick, together with the expression of gremlin in the duck interdigital webs, indicates that Gremlin regulates the regression of the interdigital tissue. At later stages of limb development, gremlin is expressed in association with the differentiating skeletal pieces, muscles and the feather buds. The different expression of Gremlin in relation with other BMP antagonists present in the limb bud, such as Noggin, Chordin and Follistatin, indicates that the functions of BMPs are regulated specifically by the different BMP antagonists, acting in a complementary fashion rather than being redundant signals (Merino, 1999).

The gene noggin, originally cloned in Xenopus, encodes a secreted factor expressed in the Spemann organizer, where it functions to oppose the ventralizing influence of bone morphogenetic proteins (BMPs). Noggin protein acts by binding directly to BMPs, thereby preventing them from interacting with their receptors. The pattern of expression is described for the chicken noggin gene during somite and limb development, two tissues in which BMPs have been postulated to play essential patterning roles. noggin is expressed in dynamic restricted patterns consistent with an important role in the modulation of BMP signaling. Using a replication competent retrovirus, noggin was ectopically expressed in developing somitic and limb bud mesoderm; phenotypes were observed consistent with the complete block of BMP activity. This includes suppression of lateral somite differentiation and, in the limb, complete inhibition of chondrogenesis and local suppression of programmed cell death. In addition, ectopic noggin expression in the limb has no effect on anteroposterior limb pattern, suggesting that BMPs are unlikely to play a significant role in this process. Taken together, these results indicate that Noggin is a key regulator of vertebrate limb and somite patterning and suggest that the antagonistic Noggin-BMP interaction is a widely used mechanism to modulate BMP signaling during multiple inductive events in vertebrate embryogenesis (Capdevila, 1998b).

During early stages of chick limb development, the homeobox-containing gene Msx-2 is expressed in the mesoderm at the anterior margin of the limb bud and in a discrete group of mesodermal cells at the midproximal posterior margin. These domains of Msx-2 expression roughly demarcate the anterior and posterior boundaries of the progress zone, the highly proliferating posterior mesodermal cells underneath the apical ectodermal ridge (AER) that give rise to the skeletal elements of the limb and associated structures. Later in development, as the AER loses its activity, Msx-2 expression expands into the distal mesoderm and subsequently into the interdigital mesenchyme, which demarcates the developing digits. The domains of Msx-2 expression exhibit considerably less proliferation than the cells of the progress zone and also encompass several regions of programmed cell death including the anterior and posterior necrotic zones and interdigital mesenchyme. Therefore, it has been suggested that Msx-2 may be in a regulatory network that delimits the progress zone by suppressing the morphogenesis of the regions of the limb mesoderm in which it is highly expressed. In the present study, ectopic expression of Msx-2 via a retroviral expression vector in the posterior mesoderm of the progress zone severely impairs limb morphogenesis from the time of the initial formation of the limb bud. Msx-2-infected limbs are typically very narrow along the anteroposterior axis, are occasionally truncated, and exhibit alterations in the pattern of formation of skeletal elements, indicating that as a consequence of ectopic Msx-2 expression the morphogenesis of large portions of the posterior mesoderm has been suppressed. Msx-2 also impairs limb morphogenesis by reducing cell proliferation and promoting apoptosis in the regions of the posterior mesoderm in which it is ectopically expressed. The domains of ectopic Msx-2 expression in the posterior mesoderm also exhibit ectopic expression of BMP-4, a secreted signaling molecule that is coexpressed with Msx-2 during normal limb development in the anterior limb mesoderm, the posterior necrotic zone, and interdigital mesenchyme. This indicates that Msx-2 regulates BMP-4 expression and that the suppressive effects of Msx-2 on limb morphogenesis might be mediated in part by BMP-4. These studies indicate that during normal limb development, Msx-2 is a key component of a regulatory network that delimits the boundaries of the progress zone by suppressing the morphogenesis of the regions of the limb mesoderm in which it is highly expressed, thus restricting the outgrowth and formation of skeletal elements and associated structures to the progress zone. Rather large numbers of apoptotic cells, as well as proliferating cells, were found to be present throughout the AER during all stages of normal limb development examined, indicating that many of the cells of the AER are continuously undergoing programmed cell death at the same time that new AER cells are being generated by cell proliferation. Thus, a balance between cell proliferation and programmed cell death may play a very important role in maintaining the activity of the AER (Ferrari, 1998).

Most mouse embryos homozygous for the Bmp4(tm1blh) null allele die around the time of gastrulation, with little or no mesoderm. Two independently derived Bmp4(tm1) mutations were backcrossed onto the C57BL/6 genetic background. Several independently expressed, incompletely penetrant abnormalities are observed in heterozygotes, including cystic kidney, craniofacial malformations, microphthalmia, and preaxial polydactyly of the right hindlimb. In addition, heterozygotes are consistently underrepresented at weaning. These results indicate that Bmp4 gene dosage is essential for the normal development of a variety of organs and for neonatal viability. Two mutations that enhance the penetrance and expressivity of the polydactylous phenotype were identified: Gli3(XtJ), a deletion mutation involving a gene encoding a zinc-finger protein related to Drosophila cubitus interruptus, and Alx4(tm1rwm), a targeted null mutation in a gene encoding a paired class homeoprotein related to Drosophila aristaless. All double Bmp4(tm1)/Gli3(XtJ) heterozygotes have extensive anterior digit abnormalities of both fore- and hindlimbs, while all double Bmp4(tm1)/Alx4(tm1) heterozygotes display ectopic anterior digits only on the hindlimbs. These genetic interactions suggest a model for the multigenic control of anterior digit patterning during vertebrate limb development. It is likely that both Alx4 and Gli3 are repressors of anterior Shh expression; loss-of-function mutations in either gene would lead to ectopic anterior Shh expression and to overproliferation of anterior mesenchyme in homozygous limbs. The addition of a weak Shh mitogenic signal to the reduced cell death resulting from decreased BMP4 levels would promote the formation of anterior supernumerary digits (Dunn, 1997).

It has been proposed that digit identity in chick limb bud is specified in a dose-dependent fashion by a long-range morphogen, produced by the polarizing region. One candidate is Sonic hedgehog (Shh) protein, but it is not clear whether Shh acts long or short range or via Bmps. The relationship between Shh and Bmp signaling is dissected in this study. Shh is necessary not only for initiating bmp2 expression but also for sustaining its expression during the period when additional digits are being specified. Much of the effect of Shh during this period can be reproduced by applying only Bmp2. It has been demonstrated, by transiently adding Noggin or Bmp antibodies to limbs treated with Shh, that Bmps are responsible for digit specification. In such limbs, multiple additional digits still form but they all have the same identity. Time dependency and range of Shh signaling is explored by examining ptc expression. High-level ptc expression is induced rapidly when either Shh beads or polarizing regions are grafted to a host limb. Furthermore, high-level ptc expression is first widespread but later more restricted. All these data lead to the proposal of a new model for digit patterning. It is suggested that Shh initially acts long range to prime the region of the limb competent to form digits and thus control digit number. Then later, Shh acts short range to induce expression of Bmps, whose morphogenetic action specifies digit identity (Drossopoulou, 2000).

These findings suggest that Shh acts in two steps, first, long range, and then, short range, to control digit pattern. Whether Shh acts short range or long range will be determined by whether or not high-level expression of binding molecules such as ptc has been induced in responding cells. It is proposed that, in the first step, Shh acts long range to 'prime' limb mesenchyme cells and thus make them competent to form digits. This in essence will determine the number of digits that can form and will be related to the length of the apical ridge. In the second step, Shh acts short range to induce and maintain bmp2 expression. Bmp2 then acts in a dose-dependent fashion on the competent limb mesenchyme cells to specify digit identity. Thus, in terms of classical models for polarizing region signaling, it is proposed that Bmps act as polarizing morphogens and progressively diffuse into adjacent mesenchyme. This establishes a concentration gradient and cells that are first specified as anterior digits are later promoted to form posterior digits. This model accounts for the truncations found in the shh knockout, because, in the absence of Shh, no cells will be competent to form digits. The model also seems to be able to account for all previous results on the polarizing region and Shh induction of digits. For example, the dose-dependent effects of Shh on digit pattern are mediated by dose-and time-dependent induction of bmp2 expression, just as the dose-dependent effects of retinoic acid are mediated by dose-dependent induction of shh expression. The only exception is the finding that shh expressed tethered to the membrane protein CD4 appears to be able to induce full duplications. However, it is still possible that there is some kind of active diffusion of this Shh::CD4 involving cleavage and/or intracellular transport (Drossopoulou, 2000 and references therein).

This model could be widely applicable. The two phases of Sonic Hedgehog signaling are reminiscent of the way in which Hedgehog patterns the insect eye and sequential signaling, short range by Hh, followed by Dpp long range, is a well-established model in Drosophila wing patterning. In vertebrates, it has been shown that Shh signaling acts in two different phases to specify motor neuron identity during neural tube development; furthermore, sequential signaling by Shh and Bmps has been proposed to be involved in sclerotome specification in somites (Drossopoulou, 2000 and references therein).

Analysis of the skeletal phenotypes caused by the genetic inactivation of individual Bmps, along with the study of their expression patterns, suggests a possible functional redundancy for these molecules. To investigate the effect on skeleton development of the combined absence of some Bmp genes expressed in the same areas, heterozygous Bmp7 mice were intercrossed with Bmp2 +/-, Bmp4 +/-, or Bmp5 +/- animals. Bmp2/7 and Bmp5/7 double heterozygous animals do not exhibit any abnormalities. In contrast, Bmp4/7 double heterozygotes develop minor defects in two restricted areas of the skeleton: the rib cage, and the distal part of the limbs. In the ribs, Bmp4 and Bmp7 seem to act in the same pathway to assure proper guidance of mesenchymal condensations of the ribs extending toward the sternum. In the limbs, these molecules appear to play a similar role in controlling digit number, possibly through induction of apoptosis in the interdigital and anterior mesenchyme (Katagiri, 1998).

The mouse brachypodism locus encodes a bone morphogenetic protein (BMP)-like molecule called growth/differentiation factor 5 (GDF5). Gdf5 transcripts are expressed in a striking pattern of transverse stripes within many skeletal precursors in the developing limb. The number, location and time of appearance of these stripes corresponds to the sites where joints will later form between skeletal elements. Null mutations in Gdf5 disrupt the formation of more than 30% of the synovial joints in the limb, leading to complete or partial fusions between particular skeletal elements, and changes in the patterns of repeating structures in the digits, wrists and ankles. Mice carrying null mutations in both Gdf5 and another BMP family member, Bmp5, show additional abnormalities not observed in either of the single mutants. These defects include disruption of the sternebrae within the sternum and abnormal formation of the fibrocartilaginous joints between the sternebrae and ribs. Previous studies have shown that members of the BMP family are required for normal development of cartilage and bone. The current study suggests that particular BMP family members may also play an essential role in the segmentation process that cleaves skeletal precursors into separate elements. This process helps determine the number of elements in repeating series in both limbs and sternum, and is required for normal generation of the functional articulations between many adjacent structures in the vertebrate skeleton (Storm 1996).

Bone Morphogenetic Protein 2 (BMP-2) and Osteogenic Protein 1 (OP-1, also termed BMP-7) are both members of the transforming growth factor beta superfamily. Using heparin beads as carriers and administering locally, at different stages and locations of the chick limb bud, it was demonstrated that these BMPs provide potent apoptotic signals for the undifferentiated limb mesoderm but not for the ectoderm or the differentiating chondrogenic cells. The apoptotic regions are found in the anterior and posterior necrotic zones that remove the anterior and posterior marginal mesoderm of the early limb bud. The BMPs promote intense radial growth of the differentiating cartilages and disturb the formation of joints accompanied by alterations in the expression patterns of Indian hedgehog and ck-erg (an ETS transcription factor). Note: the Pointed and Yan sites provide additional information on ETS receptors (Macias, 1997).

Interestingly, the effects of these two BMPs on joint formation differ: the predominant effect of BMP-2 is alteration in joint shape, while OP-1 is a potent inhibitory factor for joint formation. In situ hybridizations to check whether this finding is indicative of specific roles for these BMPs in the formation of joints reveals a distinct and complementary pattern of expression for these genes during the formation of the skeleton of the digits. While Op-1 exhibits an intense expression in the perichondrium of the developing cartilages, with characteristic interruptions in the zones of joint formation, Bmp-2 expression is a positive marker for the articular interspaces. These data suggest that in addition to the proposed role for BMP-2 and OP-1 in the establishment of the anteroposterior axis of the limb, they may also play at least two direct roles in limb morphogenesis: (1) regulating the amount and spatial distribution of the undifferentiated prechondrogenic mesenchyme and (2) controlling the location of the joints and the diaphyses of the cartilaginous primordia of the long bones, once the chondrogenic aggregates are established (Macias, 1997).

During limb development, the mesenchymal cells in restricted areas of the limb bud, the anterior necrotic zone, the posterior necrotic zone, the opaque zone and the interdigital necrotic zones, are all eliminated by programmed cell death. The transcripts of bone morphogenetic protein Bmp-2 and Bmp-4 are first detected in the areas where cell death is observed, then they show overlapping expression with the programmed cell death zones, except for the opaque zone. To investigate the function of BMP-2 and BMP-4 during limb pattern formation, a dominant negative form of BMP receptor (dnBMPR-1a) was overexpressed in chick leg bud via a replication-competent retrovirus to block the endogenous BMP-2/-4 signaling pathway. This results in excess web formation at the anterior and posterior regions of limb buds in addition to marked suppression of the regression of webbing at the interdigital regions. Significant reductions in the number of apoptotic cells in these three necrotic zones are found in the limb buds that receive the virus carrying dominant negative BMP receptor. This indicates that extra tissue formation is due to suppression of programmed cell death in the three necrotic zones. Moreover, BMP-2/-4 protein induces apoptosis of mesenchymal cells isolated from the interdigital region in vitro. Other TGFß family proteins as TGFß1 and Activin did not show this effect. These results suggest that BMP-2 and BMP-4 are the apoptotic signal molecules of the programmed cell death process in the chick limb buds (Yokouchi, 1996).

The mechanisms controlling growth and patterning along the proximal-distal axis of the vertebrate limb are yet to be understood. Restriction of expression of the homeobox gene Meis2 to proximal regions of the limb bud is essential for limb development, since ectopic Meis2 severely disrupts limb outgrowth. An antagonistic relationship between the secreted factors Gremlin and BMPs required to maintain the Shh/FGF loop that regulates distal outgrowth has been uncovered. These proximal and distal factors have coordinated activities: Meis2 can repress distal genes, and Bmps and Hoxd genes restrict Meis2 expression to the proximal limb bud. Moreover, combinations of BMPs and AER factors are sufficient to distalize proximal limb cells. These results unveil a novel set of proximal-distal regulatory interactions that establish and maintain outgrowth of the vertebrate limb (Capdevila, 1999).

Signaling via the bone morphogenetic protein receptor IA (BMPR-IA) is required to establish two of the three cardinal axes of the limb: the proximal-distal axis and the dorsal-ventral axis. A conditional knockout of the gene encoding BMPR-IA (Bmpr) was generated that disrupted BMP signaling in the limb ectoderm. In the most severely affected embryos, this conditional mutation resulted in gross malformations of the limbs with complete agenesis of the hindlimbs. The proximal-distal axis is specified by the apical ectodermal ridge (AER), which forms from limb ectoderm at the distal tip of the embryonic limb bud. Analyses of the expression of molecular markers, such as Fgf8, demonstrate that formation of the AER is disrupted in the Bmpr mutants. Along the dorsal/ventral axis, loss of engrailed 1 (En1) expression in the non-ridge ectoderm of the mutants results in a dorsal transformation of the ventral limb structures. The expression pattern of Bmp4 and Bmp7 suggest that these growth factors play an instructive role in specifying dorsoventral pattern in the limb. This study demonstrates that BMPR-IA signaling plays a crucial role in AER formation and in the establishment of the dorsal/ventral patterning during limb development (Ahn, 2001).

It is hypothesized that BMP signaling pathway mediates the initial induction of DV pattern in the presumptive limb ectoderm before limb bud formation. Bmp4 and Bmp7 are expressed in the lateral mesoderm and the overlying ectoderm just prior to the induction of limb bud formation, and therefore are correctly positioned in both time and space to induce ventral limb ectoderm. Furthermore, BMPR-IA signaling is required for the induction of En1, and subsequently the specification of ventral limb identity. Finally, the dorsalizing effect of somites, is hypothesized to be mediated by the expression of the BMP antagonist, noggin, in the myotomal compartment of the somite. Noggin expression in the somites could explain the ability of somites to dorsalize the ectoderm in transplantation experiments because of their ability to inhibit BMP signals from the lateral mesoderm. Noggin expressed in the somites could neutralize the effects of BMPs in ectoderm that is in close apposition to the somites. Alternatively, the somitic mesoderm could induce the expression of an unidentified BMP antagonist in dorsal ectoderm, which suppresses the BMP signaling within the dorsal ectoderm itself. Finally, the mesoderm-derived inductive signal would have to be downregulated as the lateral mesoderm loses its ability to induce the overlying ectoderm and as the dorsally fated ectoderm moves over the limb mesenchyme. The expression of Bmp4 and Bmp7 are, in fact, rapidly downregulated in most of the lateral mesenchyme as the limb bud is formed, although Bmp4 expression is maintained in the distal limb where BMP signaling is required for AER formation (Ahn, 2001).

The data demonstrate a crucial role for BMPR-IA signaling in the formation of the AER. It is conceivable that the AER defects are secondary to DV patterning defects. However, this seems unlikely because the penetrance of the DV patterning defect is complete, whereas the AER defect is quite variable. This difference in penetrance argues that AER formation and DV patterning are two independent processes. This argument is further bolstered by analyses of the eudiplopodia chick mutant that suggest that AER formation is not strictly dependent on establishment of a DV border at the distal tip of the limb. Classical studies have shown that lateral mesoderm induces the AER. As Bmp4 and Bmp7 are expressed in lateral mesoderm, they are candidates for the initial inductive event required for AER formation (Ahn, 2001).

In summary, the conditional knockout of the gene for the most widely expressed type I BMP receptor, BMPR-IA, results in limb malformations that are due to the disruption of AER formation and loss of DV patterning. This is the first demonstration that BMPR-IA signaling is essential for these early events in limb morphogenesis (Ahn, 2001).

Dorsoventral (DV) patterning of the vertebrate limb requires the function of the transcription factor Engrailed 1 (EN1) in the ventral ectoderm. EN1 restricts, to the dorsal half of the limb, the expression of the two genes known to specify dorsal pattern. Limb growth along the proximodistal (PD) axis is controlled by the apical ectodermal ridge (AER), a specialized epithelium that forms at the distal junction between dorsal and ventral ectoderm. Using retroviral-mediated misexpression of the bone morphogenetic protein (BMP) antagonist Noggin or an activated form of the BMP receptor in the chick limb, it has been demonstrated that BMP plays a key role in both DV patterning and AER induction. Thus, the DV and PD axes are linked by a common signal. Loss and gain of BMP function experiments show that BMP signaling is both necessary and sufficient to regulate EN1 expression, and consequently DV patterning. These results also indicate that BMPs are required during induction of the AER. Manipulation of BMP signaling results in either disruptions in the endogenous AER, leading to absent or severely truncated limbs or the formation of ectopic AERs that can direct outgrowth. Moreover, BMP controls the expression of the MSX transcription factors, and the results suggest that MSX acts downstream of BMP in AER induction. It is proposed that the BMP signal bifurcates at the level of EN1 and MSX to mediate differentially DV patterning and AER induction, respectively (Pizette, 2001).

Dickkopf-1 (Dkk-1) is a potent inhibitor of Wnt/ß-catenin signaling. Expression of Dkk-1 overlaps significantly with the sites of programmed cell death in normal as well as mutant vertebrate limb development. Several of Dkk-1's upstream regulators, one of which is Bmp-4, have been identified. Interestingly, Bmp-4 activates Dkk-1 only when it concomitantly induces apoptosis. Moreover, Dkk-1 is heavily up-regulated by UV irradiation and several other genotoxic stimuli. Normal expression of Dkk-1 is dependent on the Ap-1 family member c-Jun and overexpression of Dkk-1 enhances Bmp-triggered apoptosis in the vertebrate limb. Taken together, these results provide evidence for an important role of Dkk-1-mediated inhibition of Wnt/ß-catenin signaling in response to different stress signals that all converge on the activation of c-Jun in vivo (Gotewold, 2002).

It is propose that the c-Jun-mediated activation of Dkk-1 is fundamental for Bmp-induced apoptosis. The cytoplasmic kinase TAK-1 has been reported to be essential for Bmp-2-induced apoptosis, and Bmp-4 can also directly activate this kinase. There are several further implications for TAK-1 in apoptosis. Importantly, overexpression of TAK-1 in the Drosophila visual system leads to ectopically induced apoptosis mediated by JNK. Enhanced apoptosis has also been observed in transgenic frogs and mice overexpressing TAK-1. TAK-1 activates Jnk signaling, which in turn activates c-Jun. Transcription of c-Jun is then autoregulated by the c-Jun protein, the overexpression of which is sufficient to induce apoptosis. This cascade might provide the link between Bmp and the induction of c-Jun reported in this study. It is suggested that the predominant activation of a particular intracellular signaling cascade downstream of the Bmp receptor also contributes to the different effects that Bmps have on limb mesodermal cells. According to this model, Bmp would induce apoptosis when the Bmp/Jnk pathway dominates the Bmp/Smad pathway to activate certain genes, as is the case for Dkk-1. Further support for this model comes from a study showing that the distortion of positional information determined by dpp and wg signaling gradients leads to the activation of the Drosophila JNK apoptotic pathway, which subsequently induces cell death in the Drosophila wing. This pathway is likely to be latent in normal wing development, but is activated upon abnormal dpp signaling to maintain proper development. Thus, the possibility that this pathway might only be used upon inappropriate signaling cannot be ruled out (Gotewold, 2002).

To determine the role of Bone morphogenetic protein (BMP) signaling in murine limb development in vivo, the keratin 14 promoter was used to drive expression of the BMP antagonist Noggin in transgenic mice. Phosphorylation and nuclear translocation of Smad1/5 were dramatically reduced in limbs of the transgenic animals, confirming the inhibition of BMP signaling. These mice develop extensive limb soft tissue syndactyly and postaxial polydactyly. Apoptosis in the developing limb necrotic zones is reduced with incomplete regression of the interdigital tissue. The postaxial extra digit is also consistent with a role for BMPs in regulating apoptosis. Furthermore, there is persistent expression of Fgf8, suggesting a delay in the regression of the AER. However, Msx1 and Msx2 expression is unchanged in these transgenic mice, implying that induction of these genes is not essential for mediating BMP-induced interdigital apoptosis in mice. These abnormalities are rescued by coexpressing BMP4 under the same promoter in double transgenic mice, suggesting that the limb abnormalities are a direct effect of inhibiting BMP signaling (Guha, 2002).

Tbx3, a T-box gene family member related to the Drosophila gene optomotor blind (omb) and encoding a transcription factor, is expressed in anterior and posterior stripes in developing chick limb buds. Tbx3 haploinsufficiency has been linked with the human condition ulnar-mammary syndrome, in which predominantly posterior defects occur in the upper limb. Omb is expressed in Drosophila wing development in response to a signalling cascade involving Hedgehog and Dpp. Homologous vertebrate signals Sonic hedgehog (Shh) and Bone morphogenetic protein 2 (Bmp2) are associated in chick limbs with signalling of the polarizing region, which controls anteroposterior pattern. Tissue transplantations and grafting with beads soaked in Shh, Bmps, and Noggin have been carried out in chick limb buds, and Tbx3 expression has been analyzed. Tbx3 expression was also analyzed in limb buds of chicken and mouse mutants and retinoid-deficient quail in which anteroposterior patterning is abnormal. Tbx3 expression in anterior and posterior stripes is regulated differently. Posterior Tbx3 expression is stable and depends on the signalling cascade centered on the polarising region involving Shh and Bmps, while anterior Tbx3 expression is labile and depends on the balance between positive Bmp signals, produced anteriorly, and negative Shh signals, produced posteriorly. These results are consistent with the idea that posterior Tbx3 expression is involved in specifying digit pattern and thus provides an explanation for the posterior defects in human patients. Anterior Tbx3 expression appears to be related to the width of limb bud, which determines digit number (Tümpel, 2002).

Despite extensive studies on the anterior-posterior (AP) axis formation of limb buds, mechanisms that specify digit identities along the AP axis remain obscure. Using the four-digit chick leg as a model, Tbx2 and Tbx3 are shown to specify the digit identities of digits IV and III, respectively. Misexpression of Tbx2 and Tbx3 induced posterior homeotic transformation of digit III to digit IV and digit II to digit III, respectively. Conversely, misexpression of their mutants VP16ΔTbx2 and VP16ΔTbx3 induced anterior transformation. In both cases, alterations in the expression of several markers (e.g., BMP2, Shh, and HoxD genes) were observed. In addition, Tbx2 and Tbx3 rescued Noggin-mediated inhibition of interdigital BMP signaling, signaling which is pivotal in establishing digit identities. Hence, it is concluded that Tbx3 specifies digit III, and the combination of Tbx2 and Tbx3 specifies digit IV, acting together with the interdigital BMP signaling cascade (Suzuki, 2004).

Thus chick Tbx3 and Tbx2 specify posterior digit identities by regulating interdigital BMP signaling. Misexpression of Tbx3 and Tbx2 induced posterior homeotic transformation of digit II to III and digit III to IV, respectively. In contrast, misexpression of VP16ΔTbx3 and VP16ΔTbx2 induced anterior transformation, thereby converting digit III to II and digit IV to I or II. In some cases, truncation of the posterior digits was observed, indicating that Tbx3 and Tbx2 also control the development of the posterior digits. Tbx2 and Tbx3 are known to have specific expression patterns in the interdigital autopod regions; namely, chick Tbx3 is expressed in ID3 and 4, and Tbx2 in ID4. Since the interdigit BMP level regulates its anterior digit identity, these expression patterns suggest that Tbx2 and Tbx3 might be direct regulators of the posterior digit identities. More specially, Tbx2 acts upstream of Shh and BMP2, and Tbx3 regulates BMP2. Conversely, Shh and BMP4 upregulate the posterior expression of Tbx2 and Tbx3. These lines of evidence suggest that the feedback and feedforward regulation between Tbx2/3 and the Shh and BMP signaling cascades is pivotal for the specification of posterior digit identities (Suzuki, 2004).

In an attempt to identify new genes implicated in the control of programmed cell death during limb development, a cDNA library was generated from the regressing interdigital tissue of chicken embryos. 804 sequences were analzyed from this library and 23 genes were identified involved in apoptosis in different models. One of the genes that came up in the screening was the Bone Morphogenetic Protein family member Bmp5 that had not been previously found to be involved in the control of apoptosis during limb development. In agreement with a possible role in the control of cell death, Bmp5 exhibited a regulated pattern of expression in the interdigital tissue. Transcripts of Bmp5 and BMP5 protein were abundant within the cytoplasm of the fragmenting apoptotic interdigital cells in a way suggesting that delivery of BMPs into the tissue is potentiated during apoptosis. Gain-of-function experiments have demonstrated that BMP5 has the same effect as other interdigital BMPs inducing apoptosis in the undifferentiated mesoderm and growth in the prechondrogenic mesenchyme. Both Smad proteins and MAPK p38 have been characterized as intracellular effectors for the action of BMPs in the developing limb autopod. Activation of Smad signaling involves the receptor-regulated genes Smad1 and -8, and the inhibitory Smad6, and results in both the upregulation of gene transcription and protein phosphorylation with subsequent nuclear translocation. MAPK p38 is also quickly phosphorylated after BMP stimulation in the limb mesoderm. Treatment with the inhibitor of p38, SB203580, revealed that there are interdigital genes induced by BMPs in a p38-dependent manner (DKK, Snail and FGFr3), and genes induced in a p38-independent manner (BAMBI, Msx2 and Smads). Together, these results suggest that Smad and MAPK pathways act synergistically in the BMP pathway controlling limb development (Zuzarte-Luísa, 2004).

In humans and mice, loss of HOXA13 function causes defects in the growth and patterning of the digits and interdigital tissues. Analysis of Hoxa13 expression reveals a pattern of localization overlapping with sites of reduced Bmp2 and Bmp7 expression in Hoxa13 mutant limbs. Biochemical analyses identified a novel series of Bmp2 and Bmp7 enhancer regions that directly interact with the HOXA13 DNA-binding domain and activate gene expression in the presence of HOXA13. Immunoprecipitation of HOXA13-Bmp2 and HOXA13-Bmp7 enhancer complexes from the developing autopod confirm that endogenous HOXA13 associates with these regions. Exogenous application of BMP2 or BMP7 partially rescues the Hoxa13 mutant limb phenotype, suggesting that decreased BMP signaling contributes to the malformations present in these tissues. Together, these results provide conclusive evidence that HOXA13 regulates Bmp2 and Bmp7 expression, providing a mechanistic link between HOXA13, its target genes and the specific developmental processes affected by loss of HOXA13 function (Knosp, 2004).

Normal patterning of tissues and organs requires the tight restriction of signaling molecules to well-defined organizing centers. In the limb bud, one of the main signaling centers is the zone of polarizing activity (ZPA) that controls growth and patterning through the production of sonic hedgehog (SHH). The appropriate temporal and spatial expression of Shh is crucial for normal limb bud patterning, because modifications, even if subtle, have important phenotypic consequences. However, although there is a lot of information about the factors that activate and maintain Shh expression, much less is known about the mechanisms that restrict its expression to the ZPA. This study shows that BMP activity negatively regulates Shh transcription and that a BMP-Shh negative-feedback loop serves to confine Shh expression. BMP-dependent downregulation of Shh is achieved by interfering with the FGF and Wnt signaling activities that maintain Shh expression. FGF induction of Shh requires protein synthesis and is mediated by the ERK1/2 MAPK transduction pathway. BMP gene expression in the posterior limb bud mesoderm is positively regulated by FGF signaling and finely regulated by an auto-regulatory loop. These study emphasizes the intricacy of the crosstalk between the major signaling pathways in the posterior limb bud (Bastida, 2009).

DPP homologs and regeneration

The regeneration of digit tips in mammals, including humans and rodents, represents a model for organ regeneration in higher vertebrates. Digit tip regeneration during fetal and neonatal stages of digit formation has been characterized in the mouse; regenerative capability correlates with the expression domain of the Msx1 gene. Using the stage 11 (E14.5) digit, digit tip regeneration is shown to occur in organ culture and Msx1, but not Msx2, mutant mice display a regeneration defect. Associated with this phenotype, it has been found that Bmp4 expression is downregulated in the Msx1 mutant digit and that mutant digit regeneration can be rescued in a dose-dependent manner by treatment with exogenous BMP4. Studies with the BMP-binding protein noggin show that wild-type digit regeneration is inhibited without inhibiting the expression of Msx1, Msx2 or Bmp4. These data identify a signaling pathway essential for digit regeneration, in which Msx1 functions to regulate BMP4 production. Evidence is provided that endogenous Bmp4 expression is regulated by the combined activity of Msx1 and Msx2 in the forming digit tip; however, a compensatory Msx2 response has been discovered that involves an expansion into the wild-type Msx1 domain. Thus, although both Msx1 and Msx2 function to regulate Bmp4 expression in the digit tip, the data are not consistent with a model in which Msx1 and Msx2 serve completely redundant functions in the regeneration response. These studies provide the first functional analysis of mammalian fetal digit regeneration and identify a new function for Msx1 and BMP4 as regulators of the regenerative response (Han, 2003).

DPP homologs and external genitalia

Coordinated growth and differentiation of the genital tubercle (GT), an embryonic anlage of external genitalia, generates the proximodistally elongated structure suitable for copulation, erection, uresis and ejaculation. Despite recent progress in molecular embryology, few attempts have been made to elucidate the molecular developmental processes of external genitalia formation. Bone morphogenetic protein genes (Bmp genes) and their antagonists are spatiotemporally expressed during GT development. Exogenously applied BMP increases apoptosis of GT and inhibits its outgrowth. The distal urethral epithelium (DUE), distal epithelia marked by the Fgf8 expression, may control the initial GT outgrowth. Exogenously applied BMP4 downregulates the expression of Fgf8 and Wnt5a, concomitant with increased apoptosis and decreases cell proliferation of the GT mesenchyme. Furthermore, noggin mutants and Bmpr1a conditional mutant mice display hypoplasia and hyperplasia of the external genitalia respectively. noggin mutant mice exhibited downregulation of Wnt5a and Fgf8 expression with decreased cell proliferation. Consistent with such findings, Wnt5a mutant mice display GT agenesis with decreased cell proliferation. By contrast, Bmpr1a mutant mice display decreased apoptosis and augmented Fgf8 expression in the DUE associated with GT hyperplasia. These results suggest that some of the Bmp genes could negatively affect proximodistally oriented outgrowth of GT with regulatory functions on cell proliferation and apoptosis. The DUE region can be marked only until 14.0 dpc (days post coitum) in mouse development, while GT outgrowth continues thereafter. Possible signaling crosstalk among the whole distal GT regions were also investigated (Suzuki, 2003).

Table of contents

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.