Table of contents

DPP homologs and dorsoventral patterning: Frogs - part 1

Epidermal fate in Xenopus ectoderm has been shown to be induced by a secreted growth factor, bone morphogenetic protein 4 (BMP4). However, the molecular mechanism mediating this response is poorly understood. The expression of the homeobox gene msx1 (homolog of Drosophila Muscle segment homeobox) is an immediate early response to BMP4 in Xenopus embryos. The timing of expression and embryonic distribution of msx1 parallels the timing and distribution described for BMP4. Overexpression of msx1 in early Xenopus embryos leads to their ventralization, as described for BMP4. Consistent with mediating a BMP type of signaling, overexpression of msx1 is sufficient to induce epidermis in dissociated ectoderm cells, which would otherwise form neural tissue. msx1 can also rescue neuralization imposed by a dominant negative BMP receptor (tBR) in ectodermal explants. It is proposed that Xenopus msx1 acts as a mediator of BMP signaling in epidermal induction and inhibition of neural differentiation (Suzuki, 1997a).

The specification of neural fate in Xenopus embryos is regulated by negative factors. One of these factors is the secreted protein bone morphogenetic protein-4 (BMP4): in the early gastrula ectoderm, BMP4 can both inhibit neural fate and induce epidermis. Two other Xenopus BMP genes, BMP2 and BMP7, carry out the same types of activities. Expression of a dominant negative form of the BMP2 ligand, which blocks normal processing of the wild-type ligand, causes neuralization of Xenopus ectoderm. The Xenopus BMP2/7 receptor (XALK2) was isolated and a constitutively active mutant was generated that signals in a ligand-independent manner. Signals from the activated BMP2/7 receptor also inhibit neuralization and induce epidermis in dissociated ectoderm cells. Consistent with both findings, secreted BMP2 and BMP7 ligands can also mediate neural inhibition and epidermal induction. These results suggest that both BMP2 and BMP7 may be involved independently or together with BMP4 in the inhibition of the neural fate and the onset of the epidermal induction pathway in vivo. This further supports the idea that epidermal induction is due to the effects of multiple signals from heterogeneous BMP genes (Suzuki, 1997b).

Chordin (a Xenopus homolog of Short-gastrulation) antagonizes signaling by mature bone morphogenetic proteins (BMPs) by blocking binding to their receptors. Chordin carries out this function by binding directly to BMPs. The neural induction activity of Chordin can mimic Spemann's organizer signals at concentrations close to physiological levels. Chordin induces Neural cell adhesion molecule in Xenopus animal cap explants. Addition of equimolar amounts of BMP-4 Decapentaplegic antagonizes neural induction by Chordin. Chordin also induces mesoderm dorsalization. Induction of dorsal mesoderm is evidenced by activation of muscle actin mRNA, activation of N-CAM, and elongation of explants and formation of somites. BMP-R reverses the dorsalized phenotype caused by Chordin (Piccolo, 1996).

Not surprisingly, given the number of proteins associated with the organizer, Noggin also binds to and inactivates BMP-4. This is in addition to the known binding and inactivation of activin (a potent mesoderm inducer and homolog of TGFß) by follistatin. BMP-4 and BMP-2 effectively compete for binding to Noggin, while BMP-7 binds less tightly and TGFß not as all. Noggin binding prevents BMP-4 from binding to its receptor (Zimmerman, 1996 and references).

Noggin, follistatin and chordin, three molecules secreted by the organizer, dorsalize mesoderm and promote neural induction. Follistatin was initially identified as an activin-binding protein that prevents the interaction of this growth factor with its receptor. In this paper, follistatin is shown to bind directly to BMP-4. Follistatin antagonizes the activities of BMP-4 in frog embryos and mouse teratocarcinoma cells. In Xenopus embryos follistatin blocks the ventralizing effect of BMP-4. In mouse P19 cells follistatin promotes neural differentiation. BMP-4 antagonizes the action of follistatin and prevents neural differentiation. Thus follistatin is a direct neural inducer functioning in the gastrula organizer as an antagonist to BMP-4. The three proteins that directly interact with BMP-4, chordin (Drosophila homolog: Short gastrulation), follistatin and noggin perform the main activities attributed to the organizer and its derivatives. As BMP proteins are capable of forming heterodimers, some of the activities of the BMP binding proteins might not be directed solely against BMP-4 but against other BMPs as well (Fainsod, 1997). Xmsx-1, the Xenopus homolog of Drosophila Muscle segment homeobox, is a target gene specifically regulated by BMP-4 signaling. Injection of BMP-4 messenger RNA, but not activin mRNA, induces Xmsx-1 expression in the dorsal marginal zone at the early gastrula stage, and introduction of a dominant negative form of BMP-4 receptor RNA suppresses Xmsx-1 expression in animal cap and ventral marginal zone explants at stage 14. Embryos injected with Xmsx-1 mRNA in dorsal blastomeres at the 4-cell stage exhibit a ventralized phenotype, with microcephaly and swollen abdomen. Histological observation and immunostaining reveal that these embryos have a large block of muscle tissue in the dorsal mesodermal area in place of what should be the notochord. BMP-4 is ventralizing while Xmsx-1 induces dorsolateral tissue. These results indicate that Xmsx-1 is a target gene of BMP-4 signaling, but it carries out a distinct function in relation to dorsal-ventral patterning of mesodermal tissues (Maeda, 1997).

Patterning of the embryonic ectoderm is dependent upon the action of negative (antineural) and positive (neurogenic) transcriptional regulators. Msx1 and Dlx3 are two antineural genes for which the anterior epidermal-neural boundaries of expression differ, probably due to differential sensitivity to BMP signaling in the ectoderm. In the extreme anterior neural plate, Dlx3 is strongly expressed while Msx1 is silent. While both of these factors prevent the activation of genes specific to the nascent central nervous system, Msx1 inhibits anterior markers, including Otx2 and cement gland-specific genes. Dlx3 has little, if any, effect on these anterior neural plate genes, instead providing a permissive environment for their expression while repressing more panneural markers, including prepattern genes belonging to the Zic family and BF-1. Zic3 is activated by chordin and suppressed by BMP4; overexpression of this factor results in conversion of ectoderm to anterior neural tissue. The finding that Dlx3 is able to suppress the activation of Zic3 suggests that a Dlx3-mediated regulatory step might exist between the initial disruption of BMP signaling and activation of this gene. To test this hypothesis, truncated BMP-4 receptor, Dlx3 and Zic3 RNAs were injected in combinations, followed by animal cap excision, culture and RNA isolation for Northern blot analysis. Dlx3 blocks the activation of the panneural marker Nrp1 by truncated BMP-4 receptor. Addition of Zic3 RNA to the injection mixture restores Nrp1 expression to levels comparable to those of truncated BMP-4 induced caps. Based on these results, it is concluded that the inductive effects of Zic3 function downstream of the antineurogenic stem mediated by Dlx3. These properties define a molecular mechanism for translating the organizer-dependent morphogenic gradient of BMP activity into spatially restricted gene expression in the prospective anterior neural plate (Feledy, 1999).

The ectoderm of the pre-gastrula Xenopus embryo is at least partially patterned along the dorsal-ventral axis. The early expression of the anti-neural homeodomain gene Dlx3 is localized to the ventral ectoderm by a mechanism that occurs prior to gastrulation and is independent of the Spemann organizer. The repression of Dlx3 is mediated by signaling through beta-catenin, but is probably not dependent on the induction of the Xnr3 or chordin genes by beta-catenin. It is proposed that the establishment of the dorso-ventral axis in Xenopus, which occurs during the first cell cycle and requires an enrichment of beta-catenin in prospective dorsal cells, leads to the repression of Dlx3 in the most dorsal ectoderm, prior to the formation of the Spemann organizer. This inhibition could be mediated by repression of BMP-4 expression, but could also be direct. Injection experiments in which beta-catenin and BMP-4 are co-injected indicate that the repression of Dlx3 by beta-catenin is at least partially dependent on inhibition of BMP-4 expression. Since Dlx3 is an inhibitor of neural gene expression, this repression could account for the propensity of dorsal ectoderm to respond to neural inducers, and the tendency of ventral ectoderm to express epidermal markers. This model is compatible with the observation that ventral ectoderm can be induced to become neural by organizer transplants, or equivalent procedures which would repress expression of Dlx3. This model also predicts that neural induction will take place more readily with dorsal versus ventral ectoderm, which is what has been observed (Beanan, 2000).

In early development of Xenopus laevis, it is known that activities of polypeptide growth factors are negatively regulated by their binding proteins. In this study, follistatin, originally known as an activin-binding protein, was shown to inhibit all aspects of bone morphogenetic protein (BMP) activity in early Xenopus embryos. Furthermore, using a surface plasmon resonance biosensor, it was demonstrated that follistatin can directly interact with multiple BMPs at significantly high affinities. Interestingly, follistatin is noncompetitive with the BMP receptor for ligand binding and forms a trimeric complex with BMP and its receptor. The results suggest that follistatin acts as an organizer factor in early amphibian embryogenesis by inhibiting BMP activities by a different mechanism from that used by chordin and noggin (Iemura, 1998).

Transforming growth factor-beta (TGF-beta) superfamily members elicit signals through stimulation of serine/threonine kinase receptors. Recent studies of this signaling pathway have identified two types of novel mediating molecules, the Smads and TGF-beta activated kinase 1 (TAK1: Drosophila homolog TGF-ß activated kinase 1). Smads mimic the effects of bone morphogenetic protein (BMP), activin and TGF-beta. TAK1 and TAB1 have been identified as a MAPKKK and its activator, respectively, which might be involved in the up-regulation of TGF-beta superfamily-induced gene expression, but their biological role is poorly understood. The roles of TAK1 and TAB1 have been examined in the dorsoventral patterning of early Xenopus embryos. Ectopic expression of Xenopus TAK1 (xTAK1) in early embryos induces cell death. Interestingly, however, concomitant overexpression of bcl-2 with the activated form of xTAK1 or both xTAK1 and xTAB1 in dorsal blastomeres not only rescues the cells but also causes the ventralization of the embryos. In addition, a kinase-negative form of xTAK1 (xTAK1KN), which is known to inhibit endogenous signaling, can partially rescue phenotypes generated by the expression of a constitutively active BMP-2/4 type IA receptor (BMPR-IA). xTAK1KN can block the expression of ventral mesoderm marker genes induced by Smad1 or 5. These results suggest that xTAK1 and xTAB1 function in the BMP signal transduction pathway in Xenopus embryos in a cooperative manner (Shibuya, 1998).

Signaling by members of the TGFbeta superfamily is thought to be transduced by Smad proteins. This paper describes a zebrafish mutant in smad5 designated somitabun (sbn). The dominant maternal and zygotic effect of the sbntc24 mutation is caused by a change in a single amino acid in the L3 loop of Smad5 protein, which transforms Smad5 into an antimorphic version, inhibiting wild-type Smad5 and related Smad proteins. sbn mutant embryos are strongly dorsalized, similar to mutants in Bmp2b, its putative upstream signal. Double mutant analyses and RNA injection experiments show that sbn and bmp2b interact and that sbn acts downstream of Bmp2b signaling to mediate Bmp2b autoregulation during early dorsoventral (D-V) pattern formation. A comparison among early marker gene expression patterns, chimera analyses and rescue experiments involving temporally controlled misexpression of bmp or smad in mutant embryos revealed three phases of D-V patterning: an early sbn- and bmp2b-independent phase, when a coarse initial D-V pattern is set up; an intermediate sbn- and bmp2b-dependent phase, during which the putative morphogenetic Bmp2/4 gradient is established, and a later sbn-independent phase during gastrulation, when the Bmp2/4 gradient is interpreted and cell fates are specified (Hild, 1999).

Classical experiments performed in the amphibian embryo established that the vertebrate nervous system arises, during gastrulation, from inductive interactions between the dorsal mesoderm and the overlying ectoderm. Ectopic expression of a truncated activin type II receptor, which interferes with the signaling pathway of several members of the TGF-beta family, induces the formation of neural tissue in Xenopus explants. Since interference with type II activin receptor signaling in these animal caps unveils a neural fate, it was proposed that the formation of neural tissue in vertebrate embryos is under inhibitory control. The current study shows that in Xenopus embryos, a truncated type II activin receptor (Delta1XAR1), capable of blocking signals by several transforming growth factor (TGF)-beta family members, can induce neural tissue, suggesting neural fate is under inhibitory control. Activin and bone morphogenetic protein 4 (BMP4) can act as neural inhibitors but only BMP4 can induce epidermis in Xenopus ectodermal cells. The pluripotent mouse embryonal carcinoma cell line P19 was used to examine whether the mechanisms of ectodermal cell fate decisions are conserved among vertebrates. A P19 cell line expressing Delta1XAR1 differentiates into neurons. In addition, BMP4 inhibits retinoic acid (RA)-induced neural differentiation of P19 cells and induces keratin expression. These results suggest that in mammals, as in amphibians, neural fate is under inhibitory control and BMP4 can alter ectodermal differentiation (Hoodless, 1997).

Smads are proteins that transduce signals on behalf of members of the TGF beta superfamily of growth factors. Recently, three inhibitory Smads (Smad6, Smad7, and Dad) were isolated from human, mouse, and fly respectively. These anti-Smads were shown to inhibit TGF beta signaling by stably associating to TGF beta type I receptors or, as it was shown for Smad6, by binding to receptor-activated Smad1. Xenopus Smad7 (XSmad7) inhibits signaling from the activin and BMP pathways in animal explants, although at different thresholds. When expressed in the embryo, low concentrations of XSmad7 dorsalize the ventral mesoderm, thus inducing a secondary axis. At higher concentrations however, XSmad7 inhibits both mesoderm induction and primary axis specification. In addition, XSmad7 acts as a direct neural inducer both in the context of ectodermal explants and in vivo. It is suggested that XSmad7 affects distinct TGFbeta pathways at different thresholds: at low doses, it selectively blocks the BMP pathway, whereas at higher concentrations, it is additionally capable of inhibiting the activin/TGFbeta-like pathways. XSmad7 is present maternally and maintains ubiquitous expression at least until the onset of gastrulation when the mesoderm, ectoderm, and endodem are specified (Casellas, 1998).

Recent studies on Xenopus development have revealed an increasingly complex array of inductive, prepatterning, and competence signals that are necessary for proper mesoderm formation. Fibroblast growth factor (FGF) signals through mitogen-activated protein kinase kinase (MAPKK) to induce mesodermal gene expression. A partially activated form of MAPKK restores expression of the mesodermal genes Xcad-3 and Xbra, eliminated by the dominant-negative FGF receptor (delta FGFR). Expression of a dominant-negative form of MAPKK (MAPKKD) preferentially eliminates the dorsal expression of Xcad-3 and Xbra. Does the regional localization of bone morphogenetic protein-4 (BMP-4) explain why both MAPKKD and delta FGFR eliminate the dorsal but not the ventral expression of Xcad-3 and Xbra? Ectopic expression of BMP-4 is sufficient to maintain the dorsal expression of Xcad-3 and Xbra in embryos containing delta FGFR, and expression of a dominant-negative BMP receptor reduces the dorsal-ventral differences in delta FGFR embryos. These results indicate that regional localization of BMP-4 is responsible for the dorsal-ventral asymmetry in FGF/MAPKK-mediated mesoderm induction (Northrop, 1995).

The activity of bone morphogenetic protein (BMP) heterodimers has been shown to be more potent than that of homodimers in a number of contexts, including mesoderm induction. Although BMP-2/7 and -4/7 hetero-dimers are potent inducers of ventral mesoderm in ectodermal explants, they are not a necessary component of the primary mesoderm-inducing signal in intact Xenopus embryos. The secreted BMP antagonists noggin and gremlin both efficiently block mesoderm induction by BMP homo- and heterodimers in animal caps. When these antagonists are ectopically expressed in the ventral marginal zone of early embryos, the initial formation of mesoderm as indicated by panmesodermal markers remains unaffected. Only the subsequent dorsal/ventral patterning of this mesoderm appears to be altered, with expression of a number of organizer-specific transcripts observed in the marginal zone where BMP signaling has been abolished. Thus, it is concluded that BMPs do not contribute an essential signal to mesodermal induction or patterning until gastrulation. The activities of noggin and gremlin are strikingly different from the activity of the multifunctional antagonist cerberus, which completely abolishes mesoderm induction when misexpressed during early development (Eimon, 1999).

Analysis of whole embryos injected ventrally with noggin or gremlin mRNA reveals no disruption of the panmesodermal markers Xbra, Eomes, or VegT. These findings provide compelling evidence that BMPs are not a necessary component of the general mesoderm inducing signal in Xenopus embryos. In contrast, injection of cerberus completely abolishes expression of all three mesodermal markers. The ability of cerberus to inhibit mesoderm induction can likely be attributed to the broad range of growth factors it antagonizes. Previous studies have shown that cerberus can completely inhibit the induction of Xbra in ectodermal explants by Xnr-2 and partially block mesoderm induction by activin. This latter activity may be limited to animal caps exposed to low levels of activin. Under these conditions, expression of Xbra is thought to be mediated in part through a relay, in which Xnr-1 and -2 are induced by activin; they then go on to activate Xbra transcription. Cerberus binds Xnr-1 and Xwnt-8 directly, in addition to BMPs. The observation that cerberus mRNA injected into the ventral marginal zone of a four-cell embryo efficiently blocks general mesoderm induction, provides an important positive control for noggin- and gremlin-treated embryos. It demonstrates that a growth factor antagonist can be translated from ectopically expressed mRNAs in time to completely inhibit primary mesoderm induction provided the antagonist has the proper activity. It is concluded that, although BMP heterodimers are capable of forming ventral types of mesoderm in animal cap assays, they do not constitute an essential component of the primary mesoderm-inducing signal in Xenopus. This does not preclude the possibility that, at least in the case of ventral mesoderm, there are two redundant pathways for induction, one through BMPs and the other through growth factors such as Xnrs, which are not inhibited by the BMP antagonists noggin and gremlin (Eimon, 1999).

To address the patterning function of the Bmp2, Bmp4 and Bmp7 growth factors, antisense morpholino oligomers (MO) were designed that block Bmp activity in Xenopus laevis. Bmp4 knockdown is sufficient to rescue the ventralizing effects caused by loss of Chordin activity. Double Bmp4 and Bmp7 knockdown inhibits tail development. Triple Bmp2/Bmp4/Bmp7 depletion further compromises trunk development but does not eliminate dorsoventral patterning. Unexpectedly, blocking Spemann organizer formation by UV treatment or beta-Catenin depletion causes BMP inhibition to have much more potent effects, abolishing all ventral development and resulting in embryos having radial central nervous system (CNS) structures. Surprisingly, dorsal signaling molecules such as Chordin, Noggin, Xnr6 and Cerberus are not re-expressed in these embryos. It is concluded that BMP inhibition is sufficient for neural induction in vivo, and that in the absence of ventral BMPs, Spemann organizer signals are not required for brain formation (Reversade, 2005).

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.