Table of contents

Dpp homologs and the neural crest

To define the timing of neural crest formation, the fate of presumptive neural crest cells was challenged by grafting notochords, Sonic Hedgehog- (Shh) or Noggin-secreting cells at different stages of neurulation in chick embryos. Notochords or Shh-secreting cells are able to prevent neural crest formation at open neural plate levels, as assayed by DiI-labeling and expression of the transcription factor Slug, suggesting that neural crest cells are not committed to their fate at this time. In contrast, the BMP signaling antagonist, Noggin, does not repress neural crest formation at the open neural plate stage, but does so if injected into the lumen of the closing neural tube. The period of Noggin sensitivity corresponds to the time when BMPs are expressed in the dorsal neural tube but are down-regulated in the non-neural ectoderm. To confirm the timing of neural crest formation, either Shh or Noggin were added to neural folds at defined times in culture. Shh inhibits neural crest production at early stages (0-5 hours in culture), whereas Noggin exerts an effect on neural crest production only later (5-10 hours in culture). These results suggest three phases of neurulation that relate to neural crest formation: (1) an initial BMP-independent phase that can be prevented by Shh-mediated signals from the notochord; (2) an intermediate BMP-dependent phase around the time of neural tube closure, when BMP-4 is expressed in the dorsal neural tube, and (3) a later pre-migratory phase, which is refractory to exogenous Shh and Noggin (Selleck, 1998).

The role of mesoderm in the induction of the neural crest in Xenopus was examined using expression of neural plate (Xsox-2) and neural crest (Xslug and ADAM) markers. Conjugation experiments using different kinds of mesoderm, together with embryonic dissection experiments suggest that the dorsolateral mesoderm is capable of specifically inducing neural crest cells. Neural crest markers can be induced in competent ectoderm at varying distances from the inducing mesoderm, with dorsal tissue inducing neural crest at a distance while dorsolateral tissue induces neural crest directly only in adjacent ectoderm. The results suggest that dorsal mesoderm has a high level of inducer, while dorsolateral mesoderm has a lower level, consistent with an inductive gradient. The possible role of BMP and noggin was examined in the generation of such a hypothetical gradient. Progressively higher levels of BMP activity are sufficient for the specification of neural plate, neural crest, and nonneural cells. Progressively higher levels of noggin are able to induce neural crest at greater distances from the source of inducer and modification of the levels of BMP activity causes induction of the neural crest in absence of neural plate, suggesting independent induction of these two tissues. A model is proposed in which a gradient of BMP activity is established in the ectoderm by interaction between BMP in the ectoderm and BMP inhibitors in the mesoderm. Neural crest is induced when a threshold level of BMP is attained in the ectoderm. The dorsolateral mesoderm produces either BMP inhibitors or a specific neural crest inducer, with low BMP activity inducing neural plate, while high BMP activity induces epidermis (Marchant, 1998).

Inductive interactions between the neural plate and epidermis can generate neural crest cells, since juxtaposition of these tissues at early stages results in the formation of neural crest cells at the interface. BMP4 and BMP7 are expressed in the epidermal ectoderm and both proteins mimic its inductive activity. BMP4 is subsequently expressed in neural cells. MSX, the vertebrate homolog of Drosophila Muscle segment homeobox is expressed in neural crest precursors and appears to be a target of BMP4. Sonic hedgehog signals from the notochord provides an opposing influence, repressing MSX in ventral neural plate, thus restricting MSX transcription to dorsal (neural crest progenitor) cells. Conversly, BMP-4 acts to suppress the differentiation of ventral cell types (Liem, 1995).

Much of the neural crest that forms the brachial arches arises from the hindbrain or rhombencephalon, which is segmented along the rostrocaudal axis into eight units, termed rhombomeres. Neural crest is responsible for craniofacial patterning. Head muscle connective tissues derived from a specific rhombomeric origin are always exclusively anchored to skeletal domains derived from the same rhombomeric origin. The rhombomeric origin of complex muscle-skeletal connections assures that muscles innervated by a fixed set of rhombomere pairs are always properly connected to the complicated composite skeletal elements. Although rhombomeres 1/2, 4 and 6 are foci of crest production, rhombomeres 3 and 5 are depleted in neural crest development. Lack of neural crest from r3 and r5 is due to apoptotic elimination of crest from the brachial arches. Expression of msx-2 (Drosophila homolog: Muscle segment homeobox) displays a tight spatial and temporal correspondence with neural crest apoptosis in rhombomeres r3 and r5. Bmp4 (Drosophila homolog: decapentaplegic), also has an expression pattern that is coincident with the pattern of neural crest apoptosis. Expression of this gene is downregulated when either r3 or r5 is freed from its cell death program, suggesting that Bmp4 is regulated by an interaction of odd numbered rhombomeres with adjacent even-numbered rhombomeres. When BMP4 protein is added to cultured r3 or r5 cells, msx-2 expression is maintained and the apoptotic program is executed. BMP signaling in the limb demonstrates an analgous function to signaling in the neural crest. A dominant negative BMP receptor construct, which blocks BMP signaling, results in a lack of interdigital cell death. Likewise msx-2 is implicated as a target of BMP in the interdigital mesenchyme (Graham, 1996 and references).

How growth factors influence the fate of multipotent progenitor cells is not well understood. Most hematopoietic growth factors act selectively as survival factors, rather than instructively as lineage determination signals. In the neural crest, neuregulin instructively promotes gliogenesis, but how alternative fates are determined is unclear. Bone morphogenic protein 2 (BMP2) induces the basic-helix-loop-helix protein MASH1 and neurogenesis in neural crest stem cells. In vivo, MASH1+ cells are located near sites of BMP2 mRNA expression. Some smooth muscle differentiation is also observed in BMP2. A related factor, transforming growth factor beta1 (TGFbeta1), exclusively promotes smooth muscle differentiation. Like neuregulin, BMP2 and TGFbeta1 act instructively rather than selectively. The neural crest and hematopoietic systems may therefore utilize growth factors in different ways to generate cellular diversity (Shah, 1996).

An investigated has been carried out of the genetic circuitry underlying the determination of neuronal identity, using mammalian peripheral autonomic neurons as a model system. Treatment of neural crest stem cells (NCSCs) with bone morphogenetic protein-2 (BMP-2) leads to an induction of MASH1 expression and consequent autonomic neuronal differentiation. BMP2 also induces expression of the paired homeodomain transcription factor Phox2a, and the GDNF/NTN signaling receptor tyrosine kinase c-RET. Constitutive expression of MASH1 in NCSCs from a retroviral vector, in the absence of exogenous BMP2, induces expression of both Phox2a and c-RET in a large fraction of infected colonies, and also promotes morphological neuronal differentiation and expression of pan-neuronal markers. In vivo, expression of Phox2a in autonomic ganglia is strongly reduced in Mash1 -/- embryos. These loss- and gain-of-function data suggest that MASH1 positively regulates expression of Phox2a, either directly or indirectly. Constitutive expression of Phox2a, in contrast to MASH1, fails to induce expression of neuronal markers or a neuronal morphology, but does induce expression of c-RET. These data suggest that MASH1 couples expression of pan-neuronal and subtype-specific components of autonomic neuronal identity, and support the general idea that identity is established by combining subprograms involving cascades of transcription factors, which specify distinct components of neuronal phenotype (Lo, 1998).

The specification of noradrenergic neurotransmitter identity in neural crest stem cells (NCSCs) has been investigated. Retroviral expression of both wild-type and dominant-negative forms of the paired homeodomain transcription factor Phox2a, related to Drosophila Aristalless, indicates a crucial and direct role for this protein (and/or the closely related Phox2b) in the regulation of endogenous tyrosine hydroxylase (TH) and dopamine-beta hydroxylase (DBH) gene expression in these cells. In collaboration with cAMP, Phox2a can induce expression of TH but not of DBH or of panneuronal genes. Phox2 proteins are, moreover, necessary for the induction of both TH and DBH by bone morphogenetic protein 2 (BMP2) (which induces Phox2a/b) and forskolin. Phox2 proteins are also necessary for neuronal differentiation. These data suggest that Phox2a/b coordinates the specification of neurotransmitter identity and neuronal fate by cooperating with environmental signals in sympathetic neuroblasts (Lo, 1999).

For neural crest cells to engage in migration, it is necessary that epithelial premigratory crest cells convert into mesenchyme. The mechanisms that trigger cell delamination from the dorsal neural tube remain poorly understood. In 15- to 40-somite-stage avian embryos, BMP4 mRNA is homogeneously distributed along the longitudinal extent of the dorsal neural tube, whereas its specific inhibitor Noggin exists in a gradient of expression that decreases caudorostrally. This rostralward reduction in signal intensity coincides with the onset of emigration of neural crest cells. Hence, it is hypothesized that an interplay between Noggin and BMP4 in the dorsal tube generates graded concentrations of the latter; BMP4, in turn, turn triggers the delamination of neural crest progenitors. Consistent with this suggestion, disruption of the gradient by grafting Noggin-producing cells dorsal to the neural tube at levels opposite the segmental plate or newly formed somites, inhibits emigration of HNK-1-positive crest cells, which instead accumulate within the dorsal tube. Similar results were obtained with explanted neural tubes from the same somitic levels exposed to Noggin. Exposure to Follistatin, however, has no effect. The Noggin-dependent inhibition is overcome by concomitant treatment with BMP4, which when added alone, also accelerates cell emigration, when compared to untreated controls. Furthermore, the observed inhibition of neural crest emigration in vivo is preceded by a partial or total reduction in the expression of cadherin-6B and rhoB but not in the expression of slug mRNA or protein. Altogether, these results suggest that a coordinated activity of Noggin and BMP4 in the dorsal neural tube triggers delamination of specified, slug-expressing neural crest cells. Thus, BMPs play multiple and discernible roles at sequential stages of neural crest ontogeny, from specification through delamination and later differentiation of specific neural crest derivatives (Sela-Donenfeld, 1999).

The dHAND basic helix-loop-helix transcription factor is expressed in neurons of sympathetic ganglia and has previously been shown to induce the differentiation of catecholaminergic neurons in avian neural crest cultures. dHAND expression is sufficient to elicit the generation of ectopic sympathetic neurons in vivo. The expression of the dHAND gene is controlled by bone morphogenetic proteins (BMPs), as suggested by BMP4 overexpression in vivo and in vitro, and by noggin- mediated inhibition of BMP function in vivo. The timing of dHAND expression in sympathetic ganglion primordia, together with the induction of dHAND expression in response to the paired homeodomain gene Phox2b implicates a role for dHAND as a transcriptional regulator downstream of Phox2b in BMP-induced sympathetic neuron differentiation (Howard, 2000).

The generation of sympathetic neurons is controlled by a network of transcriptional regulators, including the bHLH proteins Mash1 and Cash1, and the paired homeodomain proteins Phox2a and Phox2b. The expression of Cash1 and Phox2 genes and the further development of sympathetic precursors are dependent on extrinsic signals, i.e. BMPs released from the dorsal aorta in the immediate vicinity of sympathetic ganglion primordia. In vitro studies showing that the dHAND and eHAND proteins induce catecholaminergic differentiation in neural crest cells, have suggested a role for HAND genes in the sympathetic neuron development. dHAND expression is sufficient to elicit sympathetic neuron development in vivo. The expression of HAND genes is controlled in vivo and in vitro by BMPs, as suggested by gain-and loss-of-function experiments. The onset of dHAND expression, observed after Cash1 and Phox2b, but before TH and DBH expression suggests a sequential action of Phox2b and dHAND in the control of terminal differentiation genes in sympathetic neurons (Howard, 2000).

Recent studies show that specification of some neural crest lineages occurs prior to or at the time of migration from the neural tube. Signaling events establishing the melanocyte lineage, which has been shown to migrate from the trunk neural tube after the neuronal and glial lineages, have been examined. Using in situ hybridization, it has been found that, although Wnts are expressed in the dorsal neural tube throughout the time when neural crest cells are migrating, the Wnt inhibitor cfrzb-1 is expressed in the neuronal and glial precursors and not in melanoblasts. This expression pattern suggests that Wnt signaling may be involved in specifying the melanocyte lineage. Wnt-3a-conditioned medium dramatically increases the number of pigment cells in quail neural crest cultures while decreasing the number of neurons and glial cells, without affecting proliferation. Conversely, BMP-4 is expressed in the dorsal neural tube throughout the time when neural crest cells are migrating, but is decreased coincident with the timing of melanoblast migration. This expression pattern suggests that BMP signaling may be involved in neural and glial cell differentiation or repression of melanogenesis. Purified BMP-4 reduces the number of pigment cells in culture while increasing the number of neurons and glial cells, also without affecting proliferation. These data suggest that Wnt signaling specifies melanocytes at the expense of the neuronal and glial lineages, and further, that Wnt and BMP signaling have antagonistic functions in the specification of the trunk neural crest (Jin, 2001).

The neural crest is a unique cell population induced at the lateral border of the neural plate. Neural crest is not produced at the anterior border of the neural plate, which is fated to become forebrain. The roles of BMPs, FGFs, Wnts, and retinoic acid signaling in neural crest induction were analyzed by using an assay developed for investigating the posteriorization of the neural plate. Using specific markers for the anterior neural plate border and the neural crest, the posterior end of early neurula embryos, was shown to be able to transform the anterior neural plate border into neural crest cells. In addition, tissue expressing anterior neural plate markers, induced by an intermediate level of BMP activity, is transformed into neural crest by posteriorizing signals. This transformation is mimicked by bFGF, Wnt-8, or retinoic acid treatment and is also inhibited by expression of the dominant negative forms of the FGF receptor, the retinoic acid receptor, and Wnt signaling molecules. The transformation of the anterior neural plate border into neural crest cells is also achieved in whole embryos, by retinoic acid treatment or by use of a constitutively active form of the retinoic acid receptor. By analyzing the expression of mesodermal markers and various graft experiments, the expression of the mutant retinoic acid receptor has been shown to directly affect the ectoderm. A two-step model is proposed for neural crest induction. Initially, BMP levels intermediate to those required for neural plate and epidermal specification induce neural folds with an anterior character along the entire neural plate border. Subsequently, the most posterior region of this anterior neural plate border is transformed into the neural crest by the posteriorizing activity of FGFs, Wnts, and retinoic acid signals. A unifying model is discussed where lateralizing and posteriorizing signals are presented as two stages of the same inductive process required for neural crest induction (Villanueva, 2002).

It is suggested that at the early-gastrula stage, a gradient of BMP activity is established in the ectoderm, which specifies the neural plate, the neural plate border, and the epidermis at progressively higher concentrations of BMP. The neural plate border, induced at a precise location within the mediolateral axis of the ectoderm, has an anterior character. Later, between early and midgastrula stage, signals presumably originating from the ventrolateral mesoderm transform a region of the anterior neural plate border into prospective neural crest cells. A role for this mesoderm in neural crest induction has been shown. The spread of these molecules from the mesoderm into the ectoderm consequently locates them only in large animal caps, explaining why the neural crest was not induced when small animal caps were used. These signals could correspond to Wnt8 and eFGF, since it is known that they are expressed in the ventrolateral mesoderm, and could correspond to lateralizing signals. However, the neural crest is not specified at this stage; this does not occur until the end of gastrulation. Thus, additional signals are required for the final induction of the neural crest. Finally, as gastrulation proceeds, the ventrolateral mesoderm becomes localized to the posterior region of the embryo, where it continues to produce Wnt8, eFGF, and possibly retinoic acid, as well as another, as yet unknown, posteriorizing agent(s) that generates an anterior-posterior gradient of these morphogenes. This gradient would be required for the final specification of the neural crest in the most posterior region of the neural plate border. Thus, the lateral-posterior regions of the neural plate border receive the lateralizing/posteriorizing signals for an extended period of time, finally specifying them as neural crest. In contrast, the anterior neural plate border does not receive such signals or these are inhibited by other agents produced by the anterior regions of the embryo, such as cerberus or dkk1, two known Wnts inhibitors, and, as a consequence, this border region does not develop as neural crest cells. It is tempting to speculate that the anterior-posterior differences within the neural crest could be controlled by a similar mechanism (Villanueva, 2002).

Neural crest is induced at the junction of epidermal ectoderm and neural plate by the mutual interaction of these tissues. BMP4 has been shown to pattern the ectodermal tissues, and BMP4 can induce neural crest cells from the neural plate. Epidermally expressed Delta1, which encodes a Notch ligand, is required for the activation and/or maintenance of Bmp4 expression in this tissue, and is thus indirectly required for neural crest induction by BMP4 at the epidermis-neural plate boundary. Notch activation in the epidermis additionally inhibits neural crest formation in this tissue, so that neural crest generation by BMP4 is restricted to the junction (Endo, 2002).

Cleft palate, the most frequent congenital craniofacial birth defects in humans, arises from genetic or environmental perturbations in the multi-step process of palate development. Mutations in the MSX1 homeobox gene are associated with non-syndromic cleft palate and tooth agenesis in humans. Msx1-deficient mice have been used as a model system that exhibits severe craniofacial abnormalities, including cleft secondary palate and lack of teeth, to study the genetic regulation of mammalian palatogenesis. Msx1 expression is restricted to the anterior of the first upper molar site in the palatal mesenchyme and Msx1 is required for the expression of Bmp4 and Bmp2 in the mesenchyme and Shh in the medial edge epithelium (MEE) in the same region of developing palate. In vivo and in vitro analyses indicate that the cleft palate seen in Msx1 mutants results from a defect in cell proliferation in the anterior palatal mesenchyme rather than a failure in palatal fusion. Transgenic expression of human Bmp4 driven by the mouse Msx1 promoter in the Msx1–/– palatal mesenchyme rescues the cleft palate phenotype and neonatal lethality. Associated with the rescue of the cleft palate is a restoration of Shh and Bmp2 expression, as well as a return of cell proliferation to the normal levels. Ectopic Bmp4 appears to bypass the requirement for Msx1 and functions upstream of Shh and Bmp2 to support palatal development. Further in vitro assays indicate that Shh (normally expressed in the MEE) activates Bmp2 expression in the palatal mesenchyme, which in turn acts as a mitogen to stimulate cell division. Msx1 thus controls a genetic hierarchy involving BMP and Shh signals that regulates the growth of the anterior region of palate during mammalian palatogenesis. These findings provide insights into the cellular and molecular etiology of the non-syndromic clefting associated with Msx1 mutations (Zhang, 2002).

The generation of noradrenergic sympathetic neurons is controlled by BMPs and the downstream transcription factors Mash1, Phox2b, Phox2a and dHand. The role of these signals in developing cholinergic parasympathetic neurons has been examined. The expression of Mash1 (Cash1), Phox2b and Phox2a in the chick ciliary ganglion is followed by the sequential expression of panneuronal, noradrenergic and cholinergic marker genes. BMPs are expressed at the site where ciliary ganglia form and are essential and sufficient for ciliary neuron development. Unlike sympathetic neurons, ciliary neurons do not express dHand; noradrenergic gene expression is eventually lost but can be maintained by ectopic dHand expression. Chick sympathetic and ciliary neurons differ with respect to the developmental expression of the noradrenergic marker genes TH and DBH. The results suggest that this difference is due to the selective expression of the transcription factor dHand in sympathetic neurons, maintaining TH/DBH expression in this lineage. Thus, the generation of different neuronal subtypes, noradrenergic sympathetic and cholinergic parasympathetic neurons, in response to a BMP signal is explained by the differential expression of transcriptional control elements, including dHand. This scenario would implicate fate-determining differences between sympathetic and parasympathetic ciliary precursor cells at the time they respond to BMPs Together, these results demonstrate a common BMP dependence of sympathetic neurons and parasympathetic ciliary neurons and implicate dHand in the maintenance of noradrenergic gene expression in the autonomic nervous system (Müller, 2002).

The neural crest is a population of cells that originates at the interface between the neural plate and non-neural ectoderm. The role that Notch and the homeoprotein Xiro1 play in the specification of the neural crest has been anayzed. Xiro1, Notch and the Notch target gene Hairy2A are all expressed in the neural crest territory, whereas the Notch ligands Delta1 and Serrate are expressed in the cells that surround the prospective crest cells. Inducible dominant-negative and activator constructs of both Notch signaling components and Xiro1 were used to analyze the role of these factors in neural crest specification without interfering with mesodermal or neural plate development (Galvic, 2004).

Activation of Xiro1 or Notch signaling leads to an enlargement of the neural crest territory, whereas blocking their activity inhibits the expression of neural crest markers. It is known that BMPs are involved in the induction of the neural crest and, thus, whether these two elements might influence the expression of Bmp4 was assessed. Activation of Xiro1 and of Notch signaling upregulates Hairy2A and inhibits Bmp4 transcription during neural crest specification. These results, in conjunction with data from rescue experiments, allow a model to be proposed wherein Xiro1 lies upstream of the cascade regulating Delta1 transcription. At the early gastrula stage, the coordinated action of Xiro1, as a positive regulator, and Snail, as a repressor, restricts the expression of Delta1 at the border of the neural crest territory. At the late gastrula stage, Delta1 interacts with Notch to activate Hairy2A in the region of the neural fold. Subsequently, Hairy2A acts as a repressor of Bmp4 transcription, ensuring that levels of Bmp4 optimal for the specification of the neural plate border are attained in this region. Finally, the activity of additional signals (WNTs, FGF and retinoic acid) in this newly defined domain induces the production of neural crest cells. These data also highlight the different roles played by BMP in neural crest specification in chick and Xenopus or zebrafish embryos (Galvic, 2004).

In conclusion, Notch signaling activates the expression of Hairy2A in the region of the neural folds, and thereby represses Bmp4 transcription. This effect of Notch signaling is dependent on Xmsx1 activity, since the inhibition of Notch by Su(H)DBMGR can be reversed by Xmsx1, and the effects produced by activating Notch can be blocked by a dominant-negative Xmsx1 construct. These results also provide a possible explanation for the apparent discrepancy in the role played by BMP in chick and Xenopus or zebrafish neural crest induction. At the time of neural crest induction, the levels of BMP at the neural plate border are high in both Xenopus and zebrafish, and low in the chick. If it is assumed that an intermediate level is required to induce neural crest in all these vertebrates, then an increase in BMP levels in the chick would establish similar levels to those generated by a decrease in Xenopus and zebrafish. Thus, because of the initial differences in the levels of BMP in these two groups of organisms, the molecular machinery that induces neural crest formation (e.g. Notch/Delta, Xiro1) must adjust the specific levels of BMP by producing opposing effects on BMP expression. Thus, Notch/Delta signaling induces the neural crest by increasing BMP expression in the chick, and decreasing it in Xenopus (Galvic, 2004).

There is evidence in Xenopus and zebrafish embryos that the neural crest/neural folds are specified at the border of the neural plate by a precise threshold concentration of a Bmp gradient. In order to understand the molecular mechanism by which a gradient of Bmp is able to specify the neural crest, a study was carried out of how the expression of Bmp targets, the Msx genes, is regulated, and the role that Msx genes has in neural crest specification was examined. Since Msx genes are directly downstream of Bmp, Msx gene expression was analyzed after experimental modification in the level of Bmp activity (1) by grafting a bead soaked with noggin into Xenopus embryos, (2) by expressing in the ectoderm a dominant-negative Bmp4 or Bmp receptor in Xenopus and zebrafish embryos, and (3) by examining Bmp pathway component mutants in the zebrafish. All the results show that a reduction in the level of Bmp activity leads to an increase in the expression of Msx genes in the neural plate border. Interestingly, by reaching different levels of Bmp activity in animal cap ectoderm, it has been shown that a specific concentration of Bmp induces msx1 expression to a level similar to that required to induce neural crest. These results indicate that an intermediate level of Bmp activity specifies the expression of Msx genes in the neural fold region. In addition, the role that msx1 plays on neural crest specification was examined. Since msx1 has a role in dorsoventral pattering, conditional gain- and loss-of-function experiments were carried out using different msx1 constructs fused to a glucocorticoid receptor element to avoid an early effect of this factor. msx1 expression is able to induce all other early neural crest markers tested (snail, slug, foxd3) at the time of neural crest specification. Furthermore, the expression of a dominant negative of Msx genes leads to the inhibition of all the neural crest markers analyzed. snail is one of the earliest genes acting in the neural crest genetic cascade. In order to study the hierarchical relationship between msx1 and snail/slug several rescue experiments were performed using dominant negatives for these genes. The rescuing activity by snail and slug on neural crest development of the msx1 dominant negative, together with the inability of msx1 to rescue the dominant negatives of slug and snail strongly argue that msx1 is upstream of snail and slug in the genetic cascade that specifies the neural crest in the ectoderm. A model is proposed where a gradient of Bmp activity specifies the expression of Msx genes in the neural folds; it is proposed that this expression is essential for the early specification of the neural crest (Tríbulo, 2004).

Delamination of premigratory neural crest cells depends on a balance between BMP/noggin and on successful G1/S transition. BMP regulates G1/S transition and consequent crest delamination through canonical Wnt signaling. Noggin overexpression inhibits G1/S transition and blocking G1/S abrogates BMP-induced delamination; moreover, transcription of Wnt1 is stimulated by BMP and by the developing somites, which concomitantly inhibit noggin production. Interfering with ß-catenin and LEF/TCF inhibits G1/S transition, neural crest delamination and transcription of various BMP-dependent genes, including Cad6B, Pax3 and Msx1, but transcription of Slug, Sox9 or FoxD3. Hence, it is proposed that developing somites inhibit noggin transcription in the dorsal tube, resulting in activation of BMP and consequent Wnt1 production. Canonical Wnt signaling in turn stimulates G1/S transition and generation of neural crest cell motility independently of its proposed role in earlier neural crest specification (Burstyn-Cohen, 2004).

The enteric nervous system (ENS) is derived from neural crest cells that migrate along the gastrointestinal tract to form a network of neurons and glia that are essential for regulating intestinal motility. Despite the number of genes known to play essential roles in ENS development, the molecular etiology of congenital disorders affecting this process remains largely unknown. To determine the role of bone morphogenetic protein (BMP) signaling in ENS development, the expressions of bmp2, bmp4, and bmprII during hindgut development were examined; these were found to be strongly expressed in the ENS. Moreover, functional BMP signaling, demonstrated by the expression of phosphorylated Smad1/5/8, is present in the enteric ganglia. Inhibition of BMP activity by noggin misexpression within the developing gut, both in ovo and in vitro, inhibits normal migration of enteric neural crest cells. BMP inhibition also leads to hypoganglionosis and failure of enteric ganglion formation, with crest cells unable to cluster into aggregates. Abnormalities of migration and ganglion formation are the hallmarks of two human intestinal disorders, Hirschsprung's disease and intestinal neuronal dysplasia. The results support an essential role for BMP signaling in these aspects of ENS development and provide a basis for further investigation of these proteins in the etiology of neuro-intestinal disorders (Goldstein, 2005).

During neural crest ontogeny, an epithelial to mesenchymal transition is necessary for cell emigration from the dorsal neural tube. This process is likely to involve a network of gene activities, which remain largely unexplored. N-cadherin inhibits the onset of crest delamination both by a cell adhesion-dependent mechanism and by repressing canonical Wnt signaling found to be necessary for crest delamination by acting downstream of BMP4. Furthermore, N-cadherin protein, but not mRNA, is normally downregulated along the dorsal tube in association with the onset of crest delamination, and this process is triggered by BMP4. BMP4 stimulates cleavage of N-cadherin into a soluble cytoplasmic fragment via an ADAM10-dependent mechanism. Intriguingly, when overexpressed, the cytoplasmic N-cadherin fragment translocates into the nucleus, stimulates cyclin D1 transcription and crest delamination, while enhancing transcription of β-catenin. CTF2 also rescues the mesenchymal phenotype of crest cells in ADAM10-inhibited neural primordia. Hence, by promoting its cleavage, BMP4 converts N-cadherin inhibition into an activity that is likely to participate, along with canonical Wnt signaling, in the stimulation of neural crest emigration (Shoval, 2007).

The neural crest is induced by a combination of secreted signals. Although previous models of neural crest induction have proposed a step-wise activation of these signals, the actual spatial and temporal requirement has not been analysed. Through analysing the role of the mesoderm this study shows that specification of neural crest requires two temporally and chemically different steps: first, an induction at the gastrula stage dependent on signals arising from the dorsolateral mesoderm; and second, a maintenance step at the neurula stage dependent on signals from tissues adjacent to the neural crest. By performing tissue recombination experiments and using specific inhibitors of different inductive signals, it was shown that the first inductive step requires Wnt activation and BMP inhibition, whereas the later maintenance step requires activation of both pathways. This change in BMP necessity from BMP inhibition at gastrula to BMP activation at neurula stages is further supported by the dynamic expression of BMP4 and its antagonists, and is confirmed by direct measurements of BMP activity in the neural crest cells. The demonstration that Wnt signals are required for neural crest induction by mesoderm solves an long-standing controversy. Finally, the results emphasise the importance of considering the order of exposure to signals during an inductive event (Steventon, 2009).

The development of neural crest cells involves an epithelial-mesenchymal transition (EMT) associated with the restriction of cadherin 6B expression to the pre-migratory neural crest cells (PMNCCs), as well as a loss of N-cadherin expression. Cadherin 6B, which is highly expressed in PMNCCs, persists in early migrating neural crest cells and is required for their emigration from the neural tube. Cadherin 6B-expressing PMNCCs exhibit a general loss of epithelial junctional polarity and acquire motile properties before their delamination from the neuroepithelium. Cadherin 6B selectively induces the de-epithelialization of PMNCCs, which is mediated by stimulation of BMP signaling, whereas N-cadherin inhibits de-epithelialization and BMP signaling. As BMP signaling also induces cadherin 6B expression and represses N-cadherin, cadherin-regulated BMP signaling may create two opposing feedback loops. Thus, the overall EMT of neural crest cells occurs via two distinct steps: a cadherin 6B and BMP signaling-mediated de-epithelialization, and a subsequent delamination through the basement membrane (Par, 2010).

Ancient evolutionary origin of the neural crest gene regulatory network

The vertebrate neural crest migrates from its origin, the neural plate border, to form diverse derivatives. It has been hypothesized that a neural crest gene regulatory network (NC-GRN) guides neural crest formation. This study investigated when during evolution this hypothetical network emerged by analyzing neural crest formation in lamprey, a basal extant vertebrate. 50 NC-GRN homologs were identified and morpholinos were used to demonstrate a critical role for eight transcriptional regulators. The results reveal conservation in deployment of upstream factors, suggesting that proximal portions of the network arose early in vertebrate evolution and have been conserved for >500 million years. Biphasic expression was found of neural crest specifiers and differences in deployment of some specifiers and effectors expected to confer species-specific properties. By testing the collective expression and function of neural crest genes in a single, basal vertebrate, the ground state of the NC-GRN was revealed and ambiguities were resolved between model organisms (Sauka-Spengler, 2007).

A uniquely vertebrate innovation, the neural crest is defined by its origin at the neural plate border, migratory capability, multipotentiality, and combinatorial gene expression. As a basal jawless vertebrate, lamprey possesses neural crest cells that move along similar pathways and form many, but not all, neural-crest-derived structures found in jawed vertebrates. However, there is little or no information about early steps in neural crest specification in the lamprey. Analysis of a hypothetical NC-GRN in this basal vertebrate promises to inform on the general architecture and evolutionary history of an archetypical vertebrate gene regulatory network. As both a critical test of this putative network and a representation of its ground state, functional tests were performed involving multiple interactions within a single, basal vertebrate (Sauka-Spengler, 2007).

Fifty genes involved in neural crest formation in lamprey were identified. The findings are consistent with several features of a putative NC-GRN proposed to function in jawed vertebrates, particularly with respect to its proximal elements. Expression of signaling molecules and neural plate border specifiers is highly conserved, as are the functions of border specifiers tested in this study. BMP, Wnt, and Delta expression was found in similar patterns to those noted in frog and zebrafish, suggesting that signaling cues are present in lamprey at proper times and places to play analogous functions in neural crest specification to those in other vertebrates; e.g., Wnt8 is expressed in the nonneural ectoderm abutting the neural rod, much like chick Wnt6. Similarly, lamprey MsxA, ZicA, Dlx, and Pax3/7 are found within and adjacent to the neural plate border, implying that their combinatorial presence in the border is highly conserved across all vertebrate neurulae (Sauka-Spengler, 2007).

In contrast to these proximal steps, distal portions of the gene regulatory network exhibit both conserved and divergent features. The results suggest that neural crest specifiers are activated in two phases, with one set of transcription factors activated at the neural plate border of the early neurula and the other during a second later phase wherein the neural crest in the dorsal neural tube is forming. This differs from the previous formulation of the NC-GRN in which there was no discrimination in the timing of deployment of neural crest specifier genes into early (neural plate border) and late (bona fide neural crest precursor) categories. It is noted that the expression patterns and functions of late neural crest specifiers, like FoxD3 and SoxE family members, in lamprey resemble those observed in other vertebrates, whereas c-Myc, Id, AP2, and Snail are first deployed in the early neurula at the neural plate border rather than in nascent neural crest cells. These early-activated neural crest specifiers are expressed only slightly after the border specifiers, suggesting they may be their direct targets. Furthermore, these genes are involved in cell cycle control and therefore may play a role in maintaining multipotency of neural crest progenitors by acting as a cell cycle control switch between proliferation, cell death, and cell fate decisions (Sauka-Spengler, 2007).

The slow development of lamprey offers the advantage of allowing exquisite temporal resolution not possible in rapidly developing organisms like Xenopus and zebrafish. In jawed vertebrates, c-Myc and its direct target Id3 are expressed at the neural/nonneural ectoderm border prior to Snail1 and Sox8, but after expression of the border specifiers Msx1 and Pax3, whereas Snail2, Sox9, and FoxD3 are expressed by premigratory neural crest. However, the rapid development of Xenopus makes the exact timing of these expression patterns much more difficult to resolve. In amniotes like chick, Id family members are expressed at the neural plate border, together with proto-oncogenes c-Myc and n-Myc and bHLH transcription factor AP2a. In contrast, Sox9, FoxD3, and Snail2 are first expressed in the neural folds, while Sox10 is first expressed in delaminating neural crest. Thus, subdivision of lamprey neural crest specifiers into early- and late-acting categories may reflect either a lack of conservation or a previously unrecognized characteristic of the vertebrate neural crest network in general (Sauka-Spengler, 2007).

A difference in gene expression between lamprey and other species is that Snail is expressed earlier at the lamprey neural plate border, in contrast to its expression in premigratory neural crest in frogs, fish, and birds. Furthermore, the Snail homolog identified does not display a neural-crest-specific pattern at premigratory stages, but rather appears to be ubiquitous, similar to hagfish SnailA, and thus may represent an interesting regulatory difference between cyclostomes and gnathostomes. Similarly, the transcription factor Ets1 is expressed in premigratory, migrating, and postmigratory neural crest in Xenopus and chick and proposed to function in neural crest cell specification. In contrast, no lamprey Ets1 homologs are expressed in the neural crest cells during specification stages; rather, the first expression of both Ets1a and Ets1b is in populations of early differentiating neural crest within the branchial arches. In addition, Ets1b is expressed in hematopoietic and endothelial precursors, similar to its higher vertebrate ortholog implicated in hematopoiesis, vasculogenesis, and angiogenesis; this suggests that the lamprey gene functions hematopoetically while lacking early neural crest specifier function. Along the same lines, Twist is expressed in the premigratory crest in Xenopus, whereas lamprey Twist homologs appear to be expressed only in postmigratory crest cells lining branchial arches and persist in mesenchyme forming buccal cartilage. Extensive searches have yielded four Twist and two Ets1 homologs plus one Ets1-related factor. Given the current genome coverage (~95%), existence of another Twist or Ets1 homolog is unlikely, suggesting that the lamprey genes lack early neural crest specifier function. Intriguingly, these genes may have been co-opted to an earlier function in gnathostomes, lost early specification function in lampreys, or both. In contrast to this apparent lack of conservation, signaling receptors and adhesion and matrix molecules like Neuropilin2, Robo, and Col2a1 have similar expression patterns in gnathostomes and lamprey. An N-cadherin-like adhesion molecule, Cadherin IA, is expressed in neural tube and periocular region, but is absent from branchial arch neural crest population. A type II Cadherin homolog (Cad IIA), similar to Cad-6/7/10/11, shares similarities with all of its gnathostome counterparts and is found in premigratory (in the case of Cad-6) and early migrating (in the case of Cad-7 and Cad-11) neural crest, as well as in differentiating neurogenic derivatives (in the case of Cad-7 and Cad-10) (Sauka-Spengler, 2007).

The cumulative results suggest that lamprey possesses a NC-GRN which is a modified version of that hypothesized to function in gnathostomes (Sauka-Spengler, 2007).

Gain- and loss-of-function experiments performed in various jawed vertebrates give clues about the genetic interactions leading to neural crest specification; e.g., morpholino knockdown of neural plate border specifiers Msx1, Msx2, Pax3, and Zic1 in Xenopus, as well as Pax7 in chick, causes alterations in expression of neural crest specifiers Slug, FoxD3, Sox9, and Sox10. Concomitantly, inactivation of these neural plate border specifiers leads to the expansion of Pax3 and Zic1 and the neural marker Sox2. Inactivation of early (c-Myc, Ids, or AP2) and late (Sox8, Sox9, and Sox10) gnathostome neural crest specifiers affects expression of all neural crest specifiers. However, functional evidence for aspects of this putative network often conflicts between different jawed vertebrates; e.g, depletions of AP2, FoxD3, or Sox10 in Xenopus disrupts neural crest induction, while in zebrafish, their knockdown impinges on differentiation but has no apparent effect on induction. These differences may be due to the tetraploidy of zebrafish, compensation by redundant paralogs, or both. The emerging data suggest that the neural crest specifiers extensively cross-regulate to maintain their expression, though hierarchical relationships remain difficult to ascribe (Sauka-Spengler, 2007).

To better understand the network and obtain a more comprehensive picture of the relationships between its elements, the effects were examined of knockdown of three neural plate border and five neural crest specifier genes on more neural crest markers than has previously been done in any other vertebrate (SoxE1, SoxE2, FoxD-A, AP2, n-Myc, and Id. Although Snail is typically used as a crest marker in Xenopus, its quasi-ubiquitous presence during premigratory stages in lamprey obviated its usefulness in this study (Sauka-Spengler, 2007).

In comparing the current results with those previously described in gnathostomes, it was found that inactivation of border specifiers MsxA, Pax3/7, and ZicA results in depletion of neural crest specifier expression, consistent with observations in Xenopus. However, lamprey neural plate border specifiers do not appear to mutually coactivate. An expansion of the dorsal neural tube was observed and, correspondingly, of Pax3/7 expression therein, suggesting that inactivation of border specifiers may result in a fate conversion from neural crest to neural tube. Because many of the neural plate border specifiers are later expressed in the dorsal neural tube, they appear to have later and separate functions in the developing nervous system. These data show that inactivation of FoxD-A, n-Myc, or Id decreases expression of Pax3/7, ZicA, and SoxB1 in the roof plate, in agreement with findings in Xenopus that FoxD3 induces Zic1 and neural markers, whereas Sox9 is required for later expression of Pax3 and Msx1 (Sauka-Spengler, 2007).

Interestingly, rescue experiments using Xenopus Zic1, Msx1, AP2, and Sox9, as well as chick Pax3 mRNA, suggest that these heterospecific proteins can functionally compensate for the loss of their lamprey orthologs. These experiments imply that the protein structure of these transcription factors has been sufficiently conserved during vertebrate evolution to be interchangeable in the context of neural-crest-inducing function (Sauka-Spengler, 2007).

Traditionally the neural crest is considered an evolutionary innovation of vertebrates, since protochordates lack bona fide neural crest. In cephalochordates signaling molecules like BMP, Notch, and Wnt are expressed in a pattern closely resembling that of vertebrates, consistent with their conserved role in patterning the early ectoderm. Furthermore, in both Amphioxus and ascidians, homologs of neural plate border specifiers Msx, Zic, and Pax3/7 are present within the neural plate border territory in late gastrula/early neurula, suggesting that the initial steps of border patterning and specification are already present in protochordates. In contrast, no neural crest specifiers, with the exception of Snail, are deployed at the neural plate border. Recently, a large number of gene interactions were tested in the ascidian Ciona intestinalis using morpholino-mediated gene knockdown. While interactions related to later events in central nervous system formation appear to be conserved between urochordates and vertebrates, neural-crest-specific links are absent in Ciona, and only the activation of Snail by Zic is reminiscent of the vertebrate NC-GRN (Sauka-Spengler, 2007).

Evolution of neural crest was likely driven by changes at the gene-regulatory level that may include co-option of ancestral gene batteries to a new purpose, as well as recruitment of a supplementary transcription factor or factors into the regulatory cascade. While the proximal gene regulatory elements are highly conserved between lamprey and gnathostomes, the neural crest specifier portion can clearly be subdivided into two temporally separated subsets. More distal regulatory modules that involve deployment of intracellular and extracellular signaling cues and gene batteries responsible for migration and differentiation of neural crest cells are present, implying a high degree of evolutionary constraint. Differences between gnathostome models are likely to reflect lineage-specific alterations in expression of paralogous genes or slight alterations in degrees of cis-regulatory robustness (Sauka-Spengler, 2007).

This study shows that the molecular mechanisms guiding formation of neural crest are a vertebrate synapomorphy. As such, this conserved network fits the proposed criteria for defining gene regulatory networks functioning during development of animal body plans. The NC-GRN is composed of one or more 'kernels'. The neural plate border regulatory module is an evolutionarily inflexible unit that plays an essential upstream function in establishing the identity of the neural crest progenitor territory and is also found in protochordates that lack bona fide neural crest. It is likely that the incorporation of the neural crest specifier module into the network led to the vertebrate innovation of the neural crest kernel consisting of two interconnected parts -- the neural plate border and neural crest specifier modules. Other 'plug-ins' and 'switches' may have been co-opted into the circuit from existing developmental programs. Such plug-ins may provide signaling inputs (Wnts) or guidance cues (Npn/Sema ligand-receptor couple), whereas switches like Myc/Id, integrated at the specification level of the network, provide a mechanism of cell cycle control that alternates between neural crest cell proliferation and cell death (Sauka-Spengler, 2007).

Addition of neural crest modules to the network occurred prior to separation of jawed and jawless vertebrates, likely during the transition from protochordates to vertebrates. This reflects an ancient origin of the NC-GRN during the early Cambrian period within the estimated 200 million years between the divergence of cephalochordates and vertebrates. Furthermore, it is likely that 'differentiation' subcircuits may have been incorporated and co-opted to more proximal use in revising the NC-GRN from agnathans to gnathostomes. As an example, Ets1 and Twist are found to be deployed late in migratory and postmigratory lamprey neural crest, but exist more proximally in the gnathostome network. Thus, though a neural crest gene network was largely fixed at the base of vertebrates, there appears to be remodeling of individual subcircuits that may be responsible for species-specific traits. It is interesting to note that a recent paper reported the successful isolation of embryos from another agnathan, hagfish, for the first time after 100 years of known attempts throughout the literature. Intriguingly, the gene expression patterns for the neural crest markers reported in this study appear highly reminiscent of those of lamprey (Sauka-Spengler, 2007).

These findings are all the more significant when taking into account recent fossil finds suggesting that modern lampreys are 'living fossils,' with similar characteristics to the common ancestor with jawed vertebrates, thus reflecting the primitive vertebrate condition and occupying an important ancestral position. Prior to this study, only two genes were studied thoroughly in the context of early events in neural crest formation in the lamprey. By studying a large group of molecules, these observations couple the formation of the neural crest proper with the establishment of a NC-GRN at the dawn of vertebrates, pushing back the date that such a gene regulatory network was invented by at least 200 million years, and thus giving deep insight into the steps necessary for the creation of defining vertebrate features (Sauka-Spengler, 2007).

DPP homologs, ectodermal differentiation and ectodermal patterning

Bone morphogenetic proteins (Bmps) promote ventral specification in both the mesoderm and the ectoderm of vertebrate embryos. Zebrafish DeltaNp63, encoding an isoform of the p53-related protein p63, is identified as an ectoderm-specific direct transcriptional target of Bmp signaling. DeltaNp63 itself acts as a transcriptional repressor required for ventral specification in the ectoderm of gastrulating embryos. Loss of DeltaNp63 function leads to reduced nonneural ectoderm followed by defects in epidermal development during skin and fin bud formation. In contrast, forced DeltaNp63 expression blocks neural development and promotes nonneural development, even in the absence of Bmp signaling. Together, DeltaNp63 fulfills the criteria to be the neural repressor postulated by the 'neural default model' (Bakkers, 2002).

p63, initially isolated from mammals and also known as p51 or KET, is a homolog of the tumor suppressor and transcription factor p53. The p63 gene is transcribed from two different promoters, which in combination with alternative splicing gives rise to at least six isoforms. Use of the distal promoter generates TAp63 isoforms with the three domains also present in p53: an amino-terminal acidic transactivating domain (TAD), a central DNA binding domain (DBD), and an oligomerization domain (OD). However, use of the second transcriptional start site in intron 3 leads to the generation of N-terminally truncated DeltaNp63 isoforms, which lack the TA domain. Both the TAp63 and the DeltaNp63 transcripts can undergo differential splicing, resulting in proteins with different C-terminal regions. The longest isoforms (alphas) contain a fourth domain, the sterile alpha motif (SAM), also found in numerous other developmental regulators, while ß and gamma forms lack most or all of their SAM domains, respectively. All six proteins act as transcription factors, which can either activate or repress the expression of genes under the control of p53-responsive elements (Bakkers, 2002).

In contrast to p53 mutant mice, mice lacking the p63 gene have severe developmental defects. They lack all squamous epithelia and their derivatives, including skin, hair, whiskers, teeth, as well as mammary, lacrimal, and salivary glands, and they die shortly after birth due to dehydration. In addition, they fail to form limbs, probably as a result of the incapability to maintain the apical ectodermal ridge (AER), a structure required for limb outgrowth. Two human disorders have recently been shown to result from mutations in p63. Patients suffering from the ectodermal dysplasia, ectrodactyly, and cleft plate syndrome (EEC, OMIM 604292) have skin defects and severe limb and craniofacial abnormalities, while the ankyloblepharon-ectodermal dysplasia-clefting syndrome (AEC or Hay-Wells, OMIM 106260) is characterized by fused eyelids and severe scalp dermatitis, but normal limb formation. These phenotypes, together with the high expression rates of p63 in proliferating basal cells of the epidermis, have led to the proposal that p63 is involved in the regulation of proliferation and differentiation programs in epithelial tissues. Since differentiated cells can be detected in the epidermis of knockout mice, it has been further proposed that p63 might be required to maintain the regenerative character of epithelial stem cells, rather than for keratinocyte differentation. It has been, however, impossible to specify which of the different p63 isoforms is essential for these processes. Also, little is known about the regulation of p63 expression (Bakkers, 2002).

The isolation of three different DeltaNp63 isoforms from the zebrafish is described. By using antisense morpholino oligonucleotides directed against DeltaNp63, it has been shown that p63 lacking the transactivation domain is required for skin formation and AER maintenance in zebrafish pectoral fin buds. Analyses of earlier stages of morphant embryos and overexpression studies further reveal that DeltaNp63 acts as a transcriptional repressor with a much earlier role during DV patterning of the zebrafish ectoderm. The early expression of DeltaNp63 in the ventral ectoderm is directly activated by Smad4/5-mediated Bmp signaling and is sufficient to block anterior neural specification while promoting early steps of epidermal specification, even in embryos lacking Bmp signaling (Bakkers, 2002).

The development of the feather buds during avian embryogenesis is a classic example of a spacing pattern. The regular arrangement of feather buds is achieved by a process of lateral inhibition whereby one developing feather bud prevents the formation of similar buds in the immediate vicinity. Lateral inhibition during feather formation implicates a role of long range signaling during this process. BMPs are able to enforce lateral inhibition during feather bud formation. However these results do not explain how the feather bud escapes the inhibition itself. It has been shown that this could be achieved by the expression of the BMP antagonist, Follistatin. Furthermore local application of Follistatin leads to the development of ectopic feather buds. It is suggested that Follistatin locally antagonizes the action of the BMPs and so permits the cellular changes associated with feather placode formation. Evidence is provided for the role of short range signaling during feather formation. Changes in cellular morphology in feather placodes has been corrolated with the expression of the gene Eph-A4, which encodes a receptor tyrosine kinase that requires direct cell-cell contact for activation. The expression of this gene precedes the cellular reorganization required for feather bud formation (Patel. 1999).

It has been an intriguing problem to solve: do the polypeptide growth factors belonging to the transforming growth factor-beta (TGF-beta) superfamily function as direct and long-range signaling molecules in pattern formation of the early embryo? In this study, the mechanism of signal propagation of bone morphogenetic protein (BMP) was determined in the ectodermal patterning of zebrafish embryos, in which BMP functions as an epidermal inducer and a neural inhibitor. To estimate the effective range of zbmp-2, whole-mount in situ hybridization analysis was performed. The zbmp-2-expressing domain and the neuroectoderm, marked by otx-2 expression, are complementary, suggesting that BMP has a short-range effect in vivo. Moreover, mosaic experiments using a constitutively active form of a zebrafish BMP type I receptor (CA-BRIA) demonstrate that the cell-fate conversion, revealed by ectopic expression of gata-3 and repression of otx-2, occurs in a cell-autonomous manner, denying the involvement of the relay mechanism. zbmp-2 is induced cell autonomously within the transplanted cells in the host ectoderm, suggesting that BMP cannot influence even the neighboring cells. This result is consistent with the observation that there is no gap between the expression domains of zbmp-2 and otx-2. Taken together, it is proposed that, in ectodermal patterning, BMP exerts a direct and cell-autonomous effect on the fates of uncommitted ectodermal cells, making them become epidermis (Nikaido, 1999).

Spacing patterns are of fundamental importance in various repeated structures that develop at regular intervals such as feathers, teeth and insect ommatidia. The mouse tongue develops a regular papilla pattern and provides a good model to study pattern formation. The expression patterns of the signaling molecules, sonic hedgehog (Shh), bone morphogenetic proteins -2 and -4 (Bmp-2 and Bmp-4), and fibroblast growth factor-8 (Fgf-8) were studied in mouse embryos between E 10.5 and 15. All four genes are expressed uniformly in the tongue epithelium between E 10.5 and 11. At E 13, before morphologically detectable gustatory papillae initiation, Shh, Bmp-2 and Bmp-4 expression segregates into discrete spots, whereas, Fgf-8 is downregulated. At E 14, small eminences in the anterior part of the tongue are the first morphological indications of fungiform papillae, and they express Shh and Bmp-2, whereas, Bmp-4 is almost absent in the tongue. It is concluded that these conserved signaling molecules are associated with the initiation and early morphogenesis of the tongue papillae (Jung, 1999).

Growth and differentiation of postnatal hair follicles are controlled by reciprocal interactions between the dermal papilla and the surrounding epidermal hair precursors. The first visible sign of hair formation is the hair placode, a thickening of the embryonic ectoderm induced by the underlying mesoderm. A signal from the hair placode then causes the mesenchymal cells to condense. A second mesodermal signal in turn induces proliferation in the ectodermal placode, which starts to grow down into the mesenchyme. Eventually, the epithelial cells surround the mesodermal condensation, which forms the dermal papilla, a permanent structure at the base of the follicle thought to control growth and differentiation of the hair. Once the basic structure of the hair follicle is established, differentiation of the concentric layers of keratinocytes begins. The peripheral layer forms the outer root sheath (ORS), which is continuous with the basal layer of the epidermis. Within the hair follicle, two new layers, the inner root sheath (IRS) and the hair shaft, develop from proliferating precursor cells in the matrix region surrounding the dermal papilla. As cells move distally, they stop dividing and differentiate according to their medio-lateral position within the follicle. Cells positioned next to the ORS form the IRS while centrally located precursors give rise to the hair shaft, consisting of the outer cuticle, the cortex and the central medulla (Kulessa, 2000).

The molecular nature of these interactions is largely unknown, but they are likely to involve several families of signaling molecules, including Fgfs, Wnts and Bmps. To analyze the function of Bmp signaling in postnatal hair development, transgenic mice were generated expressing the Bmp inhibitor, Noggin, under the control of the proximal Msx2 promoter, which drives expression in proliferating hair matrix cells and differentiating hair precursor cells. Differentiation of the hair shaft but not the inner root sheath is severely impaired in Msx2-Noggin transgenic mice. In addition to hair keratins, the expression of several transcription factors implicated in hair development, including Foxn1 (HNF-3/forkhead homolog 11) and Hoxc13, is severely reduced in the transgenic hair follicles. Proliferating cells, which are normally restricted to the hair matrix surrounding the dermal papilla, are found in the precortex and hair shaft region. These results identify Bmps as key regulators of the genetic program controlling hair shaft differentiation in postnatal hair follicles (Kulessa, 2000).

The expression of four genes, Foxn1, Hoxc13, Msx1 and Msx2, is strongly reduced or absent in Msx2-Noggin transgenic hair follicles, suggesting that they lie in a genetic pathway directly controlled by Bmp signaling. Can the loss of any of these genes account for the phenotype of the Msx2-Noggin transgenic mice? Foxn1 mutant nude mice develop a normal number of hair follicles, but show incomplete differentiation of the hair shafts, which form a discernible hair cortex and medulla but rarely penetrate the skin. Although Foxn1, like Bmps, is thought to influence both the proliferation and differentiation of hair keratinocytes, the nude hair phenotype is much weaker than that of the Msx2-Noggin transgenic mice. The loss of Foxn1 expression in the Noggin transgenic hair follicles may therefore contribute to the observed phenotype, but does not fully explain it. Hoxc13 mutant mice develop hair follicles with hair shafts that do not protrude through the skin. The morphological abnormalities and the molecular nature of the defect in Hoxc13 mutant hair follicles have not been characterized extensively, so it is difficult to say how they compare to the Msx2-Noggin phenotype. Nevertheless, the co-expression of Bmp4 and Hoxc13 is very striking. It is not only seen in the hair follicle, but extends to other keratinized structures like the nails and the filiform papillae of the tongue, suggesting that Bmps and Hoxc13 form part of a more general genetic program directing ‘hard’ keratin expression. Msx1 and Msx2 have been proposed as Bmp target genes in a number of tissues, suggesting that they mediate a common rather than a tissue-specific function of Bmp signaling. Overexpression of Msx2 in the hair matrix reduces proliferation and induces premature differentiation, consistent with it mediating part of the growth regulatory functions of Bmps. Gene inactivation of Msx1 and 2 has shown that they function redundantly during hair development. The requirement of Msx function for hair follicle induction raises the possibility that Bmp activity is already necessary at early stages of hair follicle development to maintain expression of Msx1 and 2. None of the Bmp-regulated transcription factors identified here clearly mediates the entire function of Bmp signaling in the hair follicle revealed by this study. This implies that it is the coordinated regulation of multiple factors by Bmps that controls hair shaft differentiation rather than the activation of a single key regulator (Kulessa, 2000 and references therein).

The formation of periodic patterns is fundamental in biology. Theoretical models describing these phenomena have been proposed for feather patterning, however, no molecular candidates have been identified. The feather tract is initiated by a continuous stripe of Shh, Fgf-4, and Ptc expression in the epithelium, which then segregates into discrete feather primordia that are more strongly Shh and Fgf-4 positive. The primordia also become Bmp-2 and Bmp-4 positive. Bead-mediated delivery of BMPs inhibits local feather formation in contrast with the activators, Shh and Fgf-4, which induce feather formation. Both Fgf-4 and Shh induce local expression of Bmp-4, while Bmp-4 suppresses local expression of both. Fgf-4 also induces Shh. Based on these findings, a model is proposed that involves (1) homogeneously distributed global activators that define the field; (2) a position-dependent activator of competence that propagates across the field, and (3) local activators and inhibitors triggered in sites of individual primordia that act in a reaction-diffusion mechanism. A computer simulation model for feather pattern formation is also presented (Jung, 1998).

The spacing of feather buds in a tract is thought to arise from the interaction between an inducing signal from the dermis and an inhibitory signal generated in the nascent buds. Local BMP-2 expression in the ectoderm precedes the formation of the ectodermal placodes, which are the first morphological sign of bud differentiation. The activity of BMP-2 or BMP-4 was altered in the ectoderm of the feather field by ectopically expressing them or their inhibitor noggin using retroviral vectors. These experiments demonstrate that BMP-2 is necessary and sufficient to mediate the lateral inhibition that positions buds in a tract. After buds are initiated, BMP-2 and BMP-4 continue to inhibit the adoption of bud fates and help to specify the size and shape of the bud. They may do so in part by their regulation of Fgf receptor expression in both the ectoderm and dermis. Additional insights into pattern formation in the feather bud can be inferred from the effects of altered BMP activity on bud morphogenesis (Noramly, 1998).

To evaluate the role of Bmp2/4 in cranial neural crest (CNC) formation or differentiation after its migration into the branchial arches, Xnoggin was used to block Bmp2/4 activities in specific areas of the CNC in transgenic mice. This resulted in depletion of CNC cells from the targeted areas. As a consequence, the branchial arches normally populated by the affected neural crest cells are hypomorphic and their skeletal and neural derivatives fail to develop. In further analyses, Bmp2 has been identified as the factor required for production of migratory cranial neural crest. Its spatial and temporal expression patterns mirror CNC emergence and Bmp2 mutant embryos lack both branchial arches and detectable migratory CNC cells. These results provide functional evidence for an essential role of BMP signaling in CNC development (Kanzler, 2000).

The neural crest originates at the interface between the surface ectoderm and neural plate by inductive interactions between both tissues. The results presented here are consistent with BMP signaling being part of this process. According to this hypothesis, a BMP signal, likely Bmp2 secreted by the surface ectoderm, would act on the neuroectoderm to produce neural crest cells able to delaminate and migrate into the surrounding mesenchyme. In this scenario, expression of Xnoggin in the dorsal neural tube would interfere with this signal. However, since both BMPs and Xnoggin are secreted molecules, it is also possible that the role of BMP signaling in early CNC development is more indirect. For instance, Bmp2 could be required for proper development of the mesenchyme that has to support CNC cell migration. In this case, inhibition of this BMP signal could also result in absence of migratory CNC cells even without direct effects on the crest cells themselves (Kanzler, 2000 and references therein).

Bone morphogenetic protein (BMP) signaling is known to be involved in multiple inductive events during embryogenesis, including the development of amniote skin. Early application of BMP-2 to the lateral trunk of chick embryos induces the formation of dense dermis, which is competent to participate in feather development. BMPs induce the dermis markers Msx-1 and the Twist-related bHLH transcription factor cDermo-1 and lead to dermal proliferation, to expression of beta-catenin, and eventually to the formation of ectopic feather tracts in originally featherless regions of chick skin. A detailed analysis of cDermo-1 expression during early feather development is presented. The data implicate that cDermo-1 is located downstream of BMP in a signaling pathway that leads to condensation of dermal cells (Scaal, 2002).

Dpp homologs and cement gland formation

otx2 activates ectopic formation of the Xenopus cement gland only in ventrolateral ectoderm, defining a region of the embryo that is permissive for cement gland formation. The molecular identity of this permissive area has been explored. One candidate permissive factor is BMP4: putative graded inhibition of BMP4 by factors such as noggin has been proposed to activate both cement gland and neural fates. Several lines of evidence are presented to suggest that BMP signaling and otx2 work together to activate cement gland formation.(1) BMP4 is highly expressed in the cement gland primordium together with otx2; (2) cement gland formation in isolated ectoderm is always accompanied by coexpression of otx2 and BMP4 mRNA, whether cement gland is induced by otx2 or by the BMP protein inhibitor noggin; (3) BMP signaling can modulate otx2 activity, such that increasing BMP signaling preferentially inhibits neural induction by otx2, while decreasing BMP signaling prevents cement gland formation. A hormone-inducible otx2 activates both ectopic neural and cement gland formation within the cement gland permissive region, in a pattern reminiscent of that found in the embryo (Gammill, 2000).

Thus there is a tight correlation between otx2 and BMP4 RNA expression and endogenous and experimentally induced cement gland formation. It is curious that these two factors are coexpressed in the cement gland primordium, since BMP4 inhibits otx2 expression and high levels of otx2 inhibit BMP4 RNA expression, and may help clear BMP4 RNA from the neural plate. How then is the overlap of otx2 and BMP4 RNA expression in the cement gland primordium achieved? Noggin and chordin are likely to be endogenous factors that activate otx2. While at high levels these factors inhibit BMP4 RNA expression, intermediate noggin concentrations (and therefore intermediate BMP signaling levels) activate otx2 and permit BMP4 RNA expression, leading to overlap of BMP4 and otx2 RNA expression. Thus, BMP4 RNA may be expressed in the cement gland primordium simply because its expression is not inhibited there. Alternatively, BMP4 may be actively induced in this region, a possibility suggested by higher levels of BMP4 RNA in presumptive cement gland than in adjacent ventral ectoderm. Within the cement gland, it is possible that otx2 and BMP4 are expressed in the same cells at levels where each does not efficiently inhibit the other. Alternately, since BMP4 is secreted and can act non-cell autonomously, otx2 and BMP4 need only be expressed in neighboring cells in order to functionally interact (Gammill, 2000).

Table of contents

decapentaplegic: Biological Overview | Transcriptional regulation | Targets of activity | Protein Interactions | Post-transcriptional Regulation | Developmental Biology | Effect of mutation | References

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.