Cellular interpretation of multiple TGF-beta signals: intracellular antagonism between activin/BVg1 and BMP-2/4 signaling mediated by Smads

TGF-beta signaling from receptors to the nucleus

In the past three years, a novel signal transduction pathway downstream of the transforming growth factor-beta (TGF-beta) superfamily receptor serine-threonine kinases has been shown to be mediated by a family of latent transcription factors called 'Smads'. These proteins mediate a short-circuited pathway in which a set of receptor-activated Smads are phosphorylated directly by the receptor kinase and then translocate to the nucleus complexed to the common mediator, Smad4, to participate in transcriptional complexes. Smads 2 and 3 mediate signals predominantly from the TGF-beta receptors. Of these, specific roles have been ascribed to Smad3 in control of chemotaxis of neutrophils and macrophages and the inhibition of Smad3 activity by the oncogene Evi-1 suggests that it may play a role in leukemogenesis. Other data, such as the induction by the inflammatory cytokine interferon-gamma of an inhibitory Smad, Smad7, which blocks the actions of Smad3, suggest that identification of the specific gene targets of Smad proteins in immune cells will provide new insight into the mechanisms of TGF-beta action on these cells.

Involvement of the small GTPases XRhoA and XRnd1 in cell adhesion and head formation in early Xenopus development

The Rho family of small GTPases regulates a variety of cellular functions, including the dynamics of the actin cytoskeleton, cell adhesion, transcription, cell growth and membrane trafficking. We have isolated the first Xenopus homologs of the Rho-like GTPases RhoA and Rnd1 and examined their potential roles in early Xenopus development. We found that Xenopus Rnd1 (XRnd1) is expressed in tissues undergoing extensive morphogenetic changes, such as marginal zone cells involuting through the blastopore, somitogenic mesoderm during somite formation and neural crest cells. XRnd1 also causes a severe loss of cell adhesion in overexpression experiments. These data and the expression pattern suggest that XRnd1 regulates morphogenetic movements by modulating cell adhesion in early embryos. Xenopus RhoA (XRhoA) is a potential XRnd1 antagonist, since overexpression of XRhoA increases cell adhesion in the embryo and reverses the disruption of cell adhesion caused by XRnd1. In addition to the potential roles of XRnd1 and XRhoA in the regulation of cell adhesion, we find a role for XRhoA in axis formation. When coinjected with dominant-negative BMP receptor (tBR) in the ventral side of the embryo, XRhoA causes the formation of head structures resembling the phenotype seen after coinjection of wnt inhibitors with dominant-negative BMP receptor. Since dominant-negative XRhoA is able to reduce the formation of head structures, we propose that XRhoA activity is essential for head formation. Thus, XRhoA may have a dual role in the embryo by regulating cell adhesion properties and pattern formation.

Cardiac looping and the vertebrate left-right axis: antagonism of left-sided Vg1 activity by a right-sided ALK2-dependent BMP pathway

The rightward looping of the primary heart tube is dependent upon upstream patterning events that establish the vertebrate left-right axis. In Xenopus, a left-sided Vg1 signaling pathway has been implicated in instructing cells to adopt a ‘left-sided identity’; however, it is not known whether ‘right-sided identity’ is acquired by a default pathway or by antagonism of Vg1 signaling. Here, we propose that an antagonistic, BMP/ALK2/Smad-mediated signaling pathway is active on the right side of the Xenopus embryo. Truncated ALK2 receptor expression on the right side of the blastula elicits heart reversals and altered nodal expression. Consistent with these findings, constitutively active ALK2 (CA-ALK2) receptor expression on the left side of the blastula also elicits heart reversals and altered nodal expression. Coexpression of CA-ALK2 with mature Vg1 ligand results in predominantly left-sided nodal expression patterns and normal heart looping, demonstrating that the ALK2 pathway can ‘rescue’ left-right reversals that otherwise occur following right-sided misexpression of mature Vg1 ligand alone. Results with chimeric precursor proteins indicate that the mature domain of BMP ligands can mimic the ability of the ALK2 signaling pathway to antagonize the Vg1 pathway. Consistent with the observed antagonism between BMP and Vg1 ligands, left-sided ectopic expression of Xolloid results in heart reversals. Moreover, ectopic expression of Smad1 or Smad7 identified two downstream modulators of the BMP/ALK2 signaling pathway that also can regulate cardiac orientation. Collectively, these results define a BMP/ALK2-mediated pathway on the right side of the Xenopus embryo and, moreover, suggest that left-right patterning preceding cardiac morphogenesis involves the activation of two distinct and antagonistic, left- and right-sided TGF(beta)-related signaling pathways.

Smad8B, a Smad8 splice variant lacking the SSXS site that inhibits Smad8-mediated signalling

BACKGROUND

Members of the TGF-beta superfamily of ligands bind to and activate surface serine/threonine-kinase receptors. Transduction of these signals requires the Smad proteins, which transiently interact with the activated receptor complex and are phosphorylated on their C terminus, SSXS site, by the type I receptor. Smad8 is a downstream signalling mediator of ALK2/ActRIA.

RESULTS

We have cloned a splice variant of Smad8, designated Smad8B. The Smad8 and Smad8B cDNAs are identical in sequence, except that Smad8B lacks a portion encoding 47 amino acids, including the SSXS phosphorylation site, in the C-terminal MH2 region. Both Smad8 and Smad8B were expressed in many of the same cell types. Smad8B was capable of specific complex formation with either Smad8 or Smad4 in mammalian cells. In cells expressing constitutively activated ALK2, Smad8B was localized to the cytoplasmic region, whereas Smad8 was translocated into the nucleus. In mammalian cells, Smad8B acted as a dominant inhibitor of BMP signalling.

CONCLUSIONS

Smad8B, a splice variant of Smad8, was isolated and found to specifically associate with both Smad8 and Smad4. Smad8B inhibited BMP signalling. Smad8 and Smad8B thus represent novel signal transduction proteins that may regulate the BMP signalling pathway.

Characterization of zebrafish smad1, smad2 and smad5: the amino-terminus of smad1 and smad5 is required for specific function in the embryo

Members of the TGFbeta superfamily of signalling molecules play important roles in mesendoderm induction and dorsoventral patterning of the vertebrate embryo. We cloned three intracellular mediators of TGFbeta signalling, smad1, 2 and 5, from the zebrafish. The three smad genes are expressed ubiquitously at the onset of gastrulation. The pattern of expression becomes progressively restricted during somitogenesis suggesting that at later stages not only the distribution of the TGFbeta signal but also that of the intracellular smad signal transducer determine the regionally restricted effects of TGFbeta signalling. Forced expression of smad1 leads to an expansion of blood cells resembling the phenotype of moderately ventralized zebrafish mutants. In contrast to Smad1, neither Smad2 nor Smad5 caused a detectable effect when expressed as full-length molecules suggesting that these latter two Smads are more dependent on activation by the cognate TGFbeta ligands. N-terminal truncated Smad2 dorsalized embryos, in agreement with a role downstream of dorsalizing TGFbeta members such as Nodals. In contrast to the C-terminal MH2 domain of Smad2, the C-terminal region of Smad1 and Smad5 lead to pleiotropic effects in embryos giving rize to both dorsalized and ventralized characteristics in injected embryos. Analysis of truncated zebrafish Smad1 in Xenopus embryos supports the notion that the C-terminal domain of smad1 is both a hypomorph and antimorph which can act as activator or inhibitor depending on the region of expression in the embryo. These results indicate a specific function of the MH1 domain of Smad1 and 5 for activity of the molecules.

Silencing of TGF-beta signalling by the pseudoreceptor BAMBI

Members of the transforming growth factor-beta (TGF-beta) superfamily, including TGF-beta, bone morphogenetic proteins (BMPs), activins and nodals, are vital for regulating growth and differentiation. These growth factors transduce their signals through pairs of transmembrane type I and type II receptor kinases. Here, we have cloned a transmembrane protein, BAMBI, which is related to TGF-beta-family type I receptors but lacks an intracellular kinase domain. We show that BAMBI is co-expressed with the ventralizing morphogen BMP4 (refs 5, 6) during Xenopus embryogenesis and that it requires BMP signalling for its expression. The protein stably associates with TGF-beta-family receptors and inhibits BMP and activin as well as TGF-beta signalling. Finally, we provide evidence that BAMBI's inhibitory effects are mediated by its intracellular domain, which resembles the homodimerization interface of a type I receptor and prevents the formation of receptor complexes. The results indicate that BAMBI negatively regulates TGF-beta-family signalling by a regulatory mechanism involving the interaction of signalling receptors with a pseudoreceptor.

A SMAD ubiquitin ligase targets the BMP pathway and affects embryonic pattern formation

The TGF-beta superfamily of proteins regulates many different biological processes, including cell growth, differentiation and embryonic pattern formation. TGF-beta-like factors signal across cell membranes through complexes of transmembrane receptors known as type I and type II serine/threonine-kinase receptors, which in turn activate the SMAD signalling pathway. On the inside of the cell membrane, a receptor-regulated class of SMADs are phosphorylated by the type-I-receptor kinase. In this way, receptors for different factors are able to pass on specific signals along the pathway: for example, receptors for bone morphogenetic protein (BMP) target SMADs 1, 5 and 8, whereas receptors for activin and TGF-beta target SMADs 2 and 3. Phosphorylation of receptor-regulated SMADs induces their association with Smad4, the 'common-partner' SMAD, and stimulates accumulation of this complex in the nucleus, where it regulates transcriptional responses. Here we describe Smurf1, a new member of the Hect family of E3 ubiquitin ligases. Smurf1 selectively interacts with receptor-regulated SMADs specific for the BMP pathway in order to trigger their ubiquitination and degradation, and hence their inactivation. In the amphibian Xenopus laevis, Smurf1 messenger RNA is localized to the animal pole of the egg; in Xenopus embryos, ectopic Smurf1 inhibits the transmission of BMP signals and thereby affects pattern formation. Smurf1 also enhances cellular responsiveness to the Smad2 (activin/TGF-beta) pathway. Thus, targeted ubiquitination of SMADs may serve to control both embryonic development and a wide variety of cellular responses to TGF-beta signals.

Functional antagonism between activin and osteogenic protein-1 in human embryonal carcinoma cells

Activin A and osteogenic protein-1 (OP-1) exerted antagonistic effects on each other's responses on the human Tera-2 embryonal carcinoma cell line. OP-1 dose dependently inhibited activin A-induced activation of p3TP-Lux transcriptional reporter, containing part of the human plasminogen activator inhibitor-1 (PAI-1) promoter, while activin A inhibited OP-1-mediated alkaline phosphatase induction. Approximately equimolar concentrations of both growth factors resulted in 50% inhibition of the respective biological responses. Affinity cross-linking studies using 125I-activin A or 125I-OP-1 followed by receptor-immunoprecipitations revealed that both ligands bound to the activin type II receptor (ActR-II), but recruited different type I receptors. In addition, OP-1 competed with binding of 125I-activin A, and activin A competed with binding of 125I-OP-1 to ActR-II. Transient transfection studies showed that competition between activin A and OP-1 also occurred at the type I receptor (ActR-1) level; constitutively active (CA)-ActR-I inhibited CA-ActR-IB-mediated p3TP-Lux reporter induction. There was no competition between activin A and OP-1 for availability of Smad4, indicating that the concentration of this common signal transducer is not limiting for generating the observed biological responses. Overexpression of ActR-II abolished the inhibitory effect of OP-1 on activin A-induced p3TP-Lux activation and, surprisingly, led to OP-1-induced transcriptional reporter activity. Whereas the exact mechanism of competition is unclear, the role of ActR-II in the competition between activin A and OP-1 is discussed in light of the observed interference in downstream signaling by CA-ActR-I and CA-ActR-IB.

Establishment of anterior-posterior polarity in avian embryos

In pre-streak chick embryos, the extraembryonic posterior marginal zone is able to induce an embryonic axis at an ectopic site without contributing cells to the induced primitive streak. This region expresses mesoderm-inducing factors that are capable of inducing an ectopic streak. Downstream of these events, chordin and bone morphogenetic protein acting within the central disc may play mutually opposing roles influencing streak formation. Although extraembryonic regions are important in establishing the embryonic axis, there does not appear to be an anterior region with head-inducing activity similar to that of the anterior visceral endoderm of the mammalian embryo.

Interaction of Smad complexes with tripartite DNA-binding sites

The Smad family of transcription factors function as effectors of transforming growth factor-beta signaling pathways. Smads form heteromultimers capable of contacting DNA through the amino-terminal MH1 domain. The MH1 domains of Smad3 and Smad4 have been shown to bind to the sequence 5'-GTCT-3'. Here we show that Smad3 and Smad4 complexes can contact three abutting GTCT sequences and that arrays of such sites elevate reporter expression relative to arrays of binding sites containing only two GTCTs. Smad3/4 complexes bound synergistically to probes containing two of the four possible arrangements of three GTCT sequences and showed a correlated ability to synergistically activate transcription through these sites. Purified Smad3 and Smad4 were both able to contact three abutting GTCT sequences and reporter experiments indicated that either protein could mediate contact with all three GTCTs. In contrast, the Smad4 MH1 domain was essential for reporter activation in combination with Smad1. Together, these results show that Smad complexes are flexible in their ability to interact with abutting GTCT triplets. In contrast, Smads have high affinity for only one orientation of abutting GTCT pairs. Functional Smad-binding sites within several native response elements contain degenerate GTCT triplets, suggesting that trimeric Smad-DNA interaction may be relevant in vivo.

DPC4/SMAD4 mediated tumor suppression of colon carcinoma cells is associated with reduced urokinase expression

We recently identified DPC4/Smad4 as a candidate tumor suppressor gene mutated or lost in one half of pancreatic carcinomas and in a subset of colon and biliary tract carcinomas. DPC4 plays a key role in signal transduction of the TGF-beta superfamily of molecules and inactivation of TGF-beta mediated growth inhibition is supposed to be the driving force for DPC4 inactivation in human tumors. However, DPC4 mediated tumor suppression by reconstitution of defective cells has not yet been reported. Here we show suppression of tumorigenicity in nude mice by stable reexpression of DPC4 in SW480 colon carcinoma cells. In vitro growth of DPC4-transfected cells was not affected and resistance towards TGF-beta mediated growth inhibition was retained. Instead, cells exhibited morphological alterations and adhesion and spreading were accelerated. These phenotypic changes were associated with reduced expression levels of the endogenous urokinase-type plasminogen activator (uPA) and plasminogen-activator-inhibitor-1 (PAI-1) genes, the products of which are implicated in the control of cell adhesion and invasion. In patients, high expression levels of uPA and PAI-1 correlate with poor prognosis. Thus, reduced expression of uPA and PAI-1 is consistent with suppression of tumorigenicity in DPC4 reconstituted cells. These results demonstrate DPC4's tumor suppressive function and suggest a potential role for DPC4 as a modulator of cell adhesion and invasion.

Identification of two Smad4 proteins in Xenopus. Their common and distinct properties

Smad family proteins have been identified as mediators of intracellular signal transduction by the transforming growth factor-beta (TGF-beta) superfamily. Each member of the pathway-restricted, receptor-activated Smad family cooperates and synergizes with Smad4, called co-Smad, to transduce the signals. Only Smad4 has been shown able to function as a common partner of the various pathway-restricted Smads in mammals. Here we have identified a novel Smad4-like molecule in Xenopus (XSmad4beta) as well as a Xenopus homolog of a well established Smad4 (XSmad4alpha). XSmad4beta is 70% identical to XSmad4alpha in amino acid sequence. Both of the Xenopus Smad4s can cooperate with Smad1 and Smad2, the pathway-restricted Smads specific for bone morphogenetic protein and TGF-beta, respectively. However, they show distinct properties in terms of their developmental expression patterns, subcellular localizations, and phosphorylation states. Moreover, XSmad4beta, but not XSmad4alpha, has the potent ability to induce ventralization when microinjected into the dorsal marginal region of the 4-cell stage of the embryos. These results suggest that the two Xenopus Smad4s have overlapping but distinct functions.

A short loop on the ALK-2 and ALK-4 activin receptors regulates signaling specificity but cannot account for all their effects on early Xenopus development

Activin, a member of the transforming growth factor beta (TGF-beta) superfamily, signals through a heteromeric complex of type I and type II serine-threonine kinase receptors. The two activin type I receptors previously identified, ALK-2 (ActR-I) and ALK-4 (ActR-IB), have distinct effects on gene expression, differentiation and morphogenesis in the Xenopus animal cap assay. ALK-4 reproduces the effects of activin treatment including the dose-dependent induction of progressively more dorso-anterior mesodermal and endodermal markers, whereas ALK-2 induces only ventral mesodermal markers and counteracts the effects of ALK-4. To identify regions of the receptors that determine signaling specificity we have generated chimeras of the constitutively active ALK-2 and ALK-4 receptors (termed ALK-2* and ALK-4*). The effects of these chimeric receptors on gene expression and morphogenetic movements implicate the loop between kinase subdomains IV and V in mediating the strong dorsal gene-inducing properties of ALK-4*; when the seven amino acids comprising this loop are transferred from ALK-4* to ALK-2*, the resulting chimeric receptor is capable of inducing the expression of dorsal-specific genes. In contrast, when the equivalent region of ALK-2* is transferred to the ALK-4* backbone it cannot effectively counteract the dorsalizing effects of ALK-4*, suggesting that other regions of type I receptors are also involved in determining signal specificity.

Transcriptional regulation of Xvent homeobox genes

The Xvent homeobox multigene family is essential for the patterning of the ventral mesoderm in Xenopus embryos. We have identified two novel members of this family, Xvent-1B and Xvent-2B, and have characterized their genomic structures. These two genes show a clustered organization and have probably arisen by gene duplication with subsequent inversion. Cis-regulatory elements within the promoters of both genes have been identified which contribute to their spatial activation. Xvent-2B is activated by BMP-2/4 in the absence of de novo protein synthesis, suggesting that this gene is a direct target of BMP-signalling. In contrast, Xvent-1B does not directly respond to BMP-2/4, but is activated by Xvent-2B. This activation is documented by Xvent-1B promoter/reporter studies, Xvent-2B overexpression and loss-of-function analysis using a dominant-negative Xvent-2 mutant. However, cycloheximide experiments reveal that Xvent-2B by itself is not sufficient to activate transcription of the Xvent-1B gene, but that there is a requirement for additional factor(s) being synthesized after midblastula transition.

Bone morphogenetic protein antagonism of Spemann's organizer is independent of Wnt signaling

The Xenopus homeobox gene twin is involved in the Wnt-mediated induction of Spemann's organizer. Additionally, several lines of evidence indicate that bone morphogenetic proteins (BMPs) play a role in repressing the formation of the organizer by antagonizing the expression of genes involved in organizer establishment. In order to determine at what level BMPs exert their effect, we measured the activity of different genes expressed within the organizer region. We report that BMP signaling can antagonize the induction of the dorsal-specific gene goosecoid but is unable to affect Wnt signaling at the level of twin. These results suggest that the antagonistic activities of BMPs in organizer formation occur postzygotically, independent of twin regulation, and that Wnt-like dorsal determinant signaling pathways do not crosstalk with BMPs.

Smad1 recognition and activation by the ALK1 group of transforming growth factor-beta family receptors

Two structural elements, the L45 loop on the kinase domain of the transforming growth factor-beta (TGF-beta) family type I receptors and the L3 loop on the MH2 domain of Smad proteins, determine the specificity of the interactions between these receptors and Smad proteins. The L45 sequence of the TGF-beta type I receptor (TbetaR-I) specifies Smad2 interaction, whereas the related L45 sequence of the bone morphogenetic protein (BMP) type I receptor (BMPR-I) specifies Smad1 interactions. Here we report that members of a third receptor group, which includes ALK1 and ALK2 from vertebrates and Saxophone from Drosophila, specifically phosphorylate and activate Smad1 even though the L45 sequence of this group is very divergent from that of BMPR-I. We investigated the structural elements that determine the specific recognition of Smad1 by ALK1 and ALK2. In addition to the receptor L45 loop and the Smad1 L3 loop, the specificity of this recognition requires the alpha-helix 1 of Smad1. The alpha-helix 1 is a conserved structural element located in the vicinity of the L3 loop on the surface of the Smad MH2 domain. Thus, Smad1 recognizes two distinct groups of receptors, the BMPR-I group and the ALK1 group, through different L45 sequences on the receptor kinase domain and a differential use of two surface structures on the Smad1 MH2 domain.

Spemann organizer activity of Smad10

The Spemann organizer induces neural tissue, dorsalizes mesoderm and generates a second dorsal axis. We report the isolation and characterization of Smad10, which has all three of these Spemann activities. Smad10 is expressed at the appropriate time to transduce Spemann signals endogenously. Like the organizer, Smad10 generates anterior and posterior neural tissues. Smad10 appears to function downstream of the Spemann organizer, consistent with a role in mediating organizer-derived signals. Interestingly, Smad10, unlike previously characterized mediators of Spemann activity, does not appear to block BMP signals. This finding, coupled with the functional activity and expression profile, suggests that Smad10 mediates Spemann action in a novel manner.

Regulation of avian cardiac myogenesis by activin/TGFbeta and bone morphogenetic proteins

Previous studies have identified two signaling interactions regulating cardiac myogenesis in avians, a hypoblast-derived signal acting on epiblast and mediated by activin or a related molecule and an endoderm-derived signal acting on mesoderm and involving BMP-2. In this study, experiments were designed to investigate the temporal relationship between these signaling events and the potential role of other TGFbeta superfamily members in regulating early steps of heart muscle development. We find that while activin or TGFbeta can potently induce cardiac myogenesis in pregastrula epiblast, they show no capacity to convert noncardiogenic mesoderm toward a myocardial phenotype. Conversely, BMP-2 or BMP-4, in combination with FGF-4, can readily induce cardiac myocyte formation in posterior mesoderm, but shows no capacity to induce cardiac myogenesis in epiblast cells. Activin/TGFbeta and BMP-2/BMP-4 therefore have distinct and reciprocal cardiac-inducing capacities that mimic the tissues in which they are expressed, the pregastrula hypoblast and anterior lateral endoderm, respectively. Experiments with noggin and follistatin provide additional evidence indicating that BMP signaling lies downstream of an activin/TGFbeta signal in the cardiac myogenesis pathway. In contrast to the cardiogenic-inducing capacities of BMP-2/BMP-4 in mesoderm, however, we find that BMP-2 or BMP-4 inhibits cardiac myogenesis prior to stage 3, demonstrating multiple roles for BMPs in mesoderm induction. These and other published studies suggest a signaling cascade in which a hypoblast-derived activin/TGFbeta signal is required prior to and during early stages of gastrulation, regulated both spatially and temporally by an interplay between BMPs and their antagonists. Later cardiogenic signals arising from endoderm, and perhaps transiently from ectoderm, and mediated in part by BMPs, act on emerging mesoderm within cardiogenic regions to activate or enhance expression of cardiogenic genes such as GATA and cNkx family members, leading to cardiac myocyte differentiation.

SMAD3/4-dependent transcriptional activation of the human type VII collagen gene (COL7A1) promoter by transforming growth factor beta

The human type VII collagen gene (COL7A1) recently has been identified as an immediate-early response gene for transforming growth factor beta (TGF-beta)/SMAD signaling pathway. In this study, by using MDA-MB-468 SMAD4-/- breast carcinoma cells, we demonstrate that expression of SMAD4 is an absolute requirement for SMAD-mediated promoter activity. We also demonstrate that the SMAD binding sequence (SBS) representing the TGF-beta response element in the region -496/-444 of the COL7A1 promoter functions as an enhancer in the context of a heterologous promoter. Electrophoretic mobility-shift assays with nuclear extracts from COS-1 cells transfected with expression vectors for SMADs 1-5 indicate that SMAD3 forms a complex with a migration similar to that of the endogenous TGF-beta-specific complex observed in fibroblast extracts. Electrophoretic mobility-shift assays using recombinant glutathione S-transferase-SMAD fusion proteins indicate that both SMAD4 and C-terminally truncated SMAD3, but not SMAD2, can bind the COL7A1 SBS. Coexpression of SMAD3 and SMAD4 in COS-1 cells leads to the formation of two complexes: a DNA/protein complex containing SMAD3 alone and another slower-migrating complex containing both SMAD3 and SMAD4, the latter complex not being detected in fibroblasts. Maximal transactivation of COL7A1 SBS-driven promoters in either MDA-MB-468 carcinoma cells or fibroblasts requires concomitant overexpression of SMAD3 and SMAD4. These data may represent the first identification of a functional homomeric SMAD3 complex regulating a human gene.

Suppression of GATA factor activity causes axis duplication in Xenopus

In Xenopus, the dorsoventral axis is patterned by the interplay between active signalling in ventral territories, and secreted antagonists from Spemann's organiser. Two signals are important in ventral cells, bone morphogenetic protein-4 (BMP-4) and Wnt-8. BMP-4 plays a conserved role in patterning the vertebrate dorsoventral axis, whilst the precise role of Wnt-8 and its relationship with BMP-4, are still unclear. Here we have investigated the role played by the GATA family of transcription factors, which are expressed in ventral mesendoderm during gastrulation and are required for the differentiation of blood and endodermal tissues. Injection ventrally of a dominant-interfering GATA factor (called G2en) induced the formation of secondary axes that phenocopy those induced by the dominant-negative BMP receptor. However, unlike inhibiting BMP signalling, inhibiting GATA activity in the ectoderm does not lead to neuralisation. In addition, analysis of gene expression in G2en injected embryos reveals that at least one known target gene for BMP-4, the homeobox gene Vent-2, is unaffected. In contrast, the expression of Wnt-8 and the homeobox gene Vent-1 is suppressed by G2en, whilst the organiser-secreted BMP antagonist chordin becomes ectopically expressed. These data therefore suggest that GATA activity is essential for ventral cell fate and that subsets of ventralising and dorsalising genes require GATA activity for their expression and suppression, respectively. Finally, using G2en, we show that suppression of Wnt-8 expression, in conjunction with blocked BMP signalling, does not lead to head formation, suggesting that the head-suppressing Wnt signal may not be Wnt-8.

Physical and functional interaction of murine and Xenopus Smad7 with bone morphogenetic protein receptors and transforming growth factor-beta receptors

Members of the transforming growth factor-beta (TGF-beta) family transmit signals from membrane to nucleus via intracellular proteins known as Smads. A subclass of Smad proteins has recently been identified that antagonize, rather than transduce, TGF-beta family signals. Smad7, for example, binds to and inhibits signaling downstream of TGF-beta receptors. Here we report that the C-terminal MAD homology domain of murine Smad7 (mSmad7) is sufficient for both of these activities. In addition, we show that mSmad7 interacts with activated bone morphogenetic protein (BMP) type I receptors (BMPR-Is), inhibits BMPR-I-mediated Smad phosphorylation, and phenocopies the effect of known BMP antagonists when overexpressed in ventral cells of Xenopus embryos. Xenopus Smad7 (XSmad7, previously termed Smad8) and mSmad7 are nearly identical within their bioactive C-domain, but have quite distinct N-domains. We found that XSmad7, similar to mSmad7, interacted with BMP and TGF-beta type I receptors and inhibited receptor-mediated phosphorylation of downstream signal-transducing Smads. However, XSmad7 is a less efficient inhibitor of TbetaR-I-mediated responses in mammalian cells than is mSmad7. Furthermore, overexpression of XSmad7 in Xenopus embryos produces patterning defects that are not observed following overexpression of mSmad7, suggesting that mSmad7 and XSmad7 may preferentially target distinct signaling pathways. Our results are consistent with the possibility that the C-domain of antagonistic Smads is an effector domain whereas the N-domain may confer specificity for distinct signaling pathways.

Smad3 mutant mice develop metastatic colorectal cancer

TGFbeta-related growth factors have been implicated in a variety of developmental and physiological processes in organisms ranging from nematodes to mammals. TGFbeta transduces its signal to the interior of the cell via Smad2, Smad3, and Smad4. We report the cloning and targeted disruption of the mouse Smad3 gene. Smad3 mutant mice are viable and fertile. Between 4 and 6 months of age, the Smad3 mutant mice become moribund with colorectal adenocarcinomas. The neoplasms penetrate through the intestinal wall and metastasize to lymph nodes. These results directly implicate TGFbeta signaling in the pathogenesis of colorectal cancer and provide a compelling animal model for the study of human colorectal cancer.

Transcriptional regulation of BMP-4 in the Xenopus embryo: analysis of genomic BMP-4 and its promoter

Recent experiments in the Xenopus embryo suggest that proper regulation of BMP-4 signaling is critical to the dorsal ventral specification of both mesoderm and ectoderm. Regulation of BMP-4 signaling is known to occur extracellularly by direct binding with chordin, noggin, and follistatin, and intracellularly through the antagonistic signal interaction with dorsalizing TGF-beta family member activin. However, tight repressional regulation of BMP transcription may also be required to sustain the dorsal and neural status of the induced cells. Here we demonstrate that the dominant negative mutant of the BMP receptor (DN-BR) or the BMP-4 antagonizers, chordin and noggin, negatively regulate BMP-4 transcription in animal cap explants. We suggest that repression of BMP-4 transcription is important in the maintenance of dorsal fate and that continuous input of BMP-4 signaling is required to sustain the expression of BMP-4 transcription in the maintenance of epidermal/ventral fate. Consistent with this postulation, we found that the promoter region of the isolated BMP-4 genomic DNA includes several consensus binding sites for transcriptional regulators functioning under BMP-4 signaling such as GATA binding and ventralizing homeobox genes. In a functional assay we found that the GATA binding and ventral homeobox proteins can positively modulate BMP-4 promoter activity. We also observed that DN-BR decreases BMP-4 promoter activity. This was likely due to a repression of the above-mentioned transcription factors. The significance of these observations to embryonic patterning is discussed.

BMPs, Smads and metalloproteases: extracellular and intracellular modes of negative regulation

Bone morphogenetic (BMPs), members of the TGF-beta superfamily, have critical functions in many biological contexts. Recent findings in Drosophila and vertebrates suggest that BMP signaling can be modulated extracellularly and intracellularly by the availability of BMP inhibitors and Smads, respectively.

The tumor suppressor Smad4/DPC4 and transcriptional adaptor CBP/p300 are coactivators for smad3 in TGF-beta-induced transcriptional activation

Smads regulate transcription of defined genes in response to TGF-beta receptor activation, although the mechanisms of Smad-mediated transcription are not well understood. We demonstrate that the TGF-beta-inducible Smad3 uses the tumor suppressor Smad4/DPC4 and CBP/p300 as transcriptional coactivators, which associate with Smad3 in response to TGF-beta. The association of CBP with Smad3 was localized to the carboxyl terminus of Smad3, which is required for transcriptional activation, and a defined segment in CBP. Furthermore, CBP/p300 stimulated both TGF-beta- and Smad-induced transcription in a Smad4/DPC4-dependent fashion. Smad3 transactivation and TGF-beta-induced transcription were inhibited by expressing E1A, which interferes with CBP functions. The coactivator functions and physical interactions of Smad4 and CBP/p300 with Smad3 allow a model for the induction of gene expression in response to TGF-beta.

Determinants of specificity in TGF-beta signal transduction

Signal transduction by the TGF-beta family involves sets of receptor serine/threonine kinases, Smad proteins that act as receptor substrates, and Smad-associated transcription factors that target specific genes. We have identified discrete structural elements that dictate the selective interactions between receptors and Smads and between Smads and transcription factors in the TGF-beta and BMP pathways. A cluster of four residues in the L45 loop of the type I receptor kinase domain, and a matching set of two residues in the L3 loop of the Smad carboxy-terminal domain establish the specificity of receptor-Smad interactions. A cluster of residues in the highly exposed alpha-helix 2 of the Smad carboxy-terminal domain specify the interaction with the DNA-binding factor Fast1 and, as a result, the gene responses mediated by the pathway. By establishing specific interactions, these determinants keep the TGF-beta and BMP pathways segregated from each other.

Smad2 and Smad3 positively and negatively regulate TGF beta-dependent transcription through the forkhead DNA-binding protein FAST2

We identify a mammalian forkhead domain protein, FAST2, that is required for induction of the goosecoid (gsc) promoter by TGF beta or activin signaling. FAST2 binds to a sequence in the gsc promoter, but efficient transcriptional activation and assembly of a DNA-binding complex of FAST2, Smad2, and Smad4 requires an adjacent Smad4 site. Smad3 is closely related to Smad2 but suppresses activation of the gsc promoter. Inhibitory activity is conferred by the MH1 domain, which unlike that of Smad2, binds to the Smad4 site. Through competition for this shared site, Smad3 may prevent transcription by altering the conformation of the DNA-binding complex. Thus, we describe a mechanism whereby Smad2 and Smad3 positively and negatively regulate a TGF beta/activin target gene.

Mads and Smads in TGF beta signalling

The discovery of the Mothers against dpp (Mad) gene in Drosophila has opened a window on an entirely unique signalling pathway that functions to mediate responses to the tumour growth factor beta (TGF beta) superfamily. This pathway, which is comprised of a family of proteins related to Mad, acts to convey signals directly from TGF beta receptors to the nucleus and is implicated in the pathogenesis of human diseases.

Identification of receptors and Smad proteins involved in activin signalling in a human epidermal keratinocyte cell line

BACKGROUND

Activin A is a multifunctional protein, which is a member of the transforming growth factor-beta (TGF-beta) superfamily. Smad proteins have recently been shown to transduce signals for the TGF-beta superfamily of proteins, and Smad2 was implicated in activin signalling in Xenopus embryos.

RESULTS

We identified the receptors and Smad proteins activated by activin A in a human epidermal keratinocyte cell line, HaCaT. The major activin receptors expressed on HaCaT cells were activin type II receptor (ActR-II) and activin type IB receptor (ActR-IB). We have also shown that in HaCaT cells, activin A induced the phosphorylation of Smad3 and, to a lesser extent, of Smad2. On the other hand, TGF-beta induced an efficient phosphorylation of both Smad2 and Smad3. Activin A preferentially induced the nuclear translocation of Smad3 in HaCaT cells, whereas TGF-beta strongly induced the nuclear translocation of Smad2, as well as other Smads. Moreover, a constitutively active form of ActR-IB efficiently stimulated the formation of a heteromeric complex between Smad3 and Smad4 in COS cells transfected with Smad cDNAs.

CONCLUSIONS

These results suggest that activin A binds to a receptor complex of ActR-II and ActR-IB, and preferentially activates Smad3 in HaCaT human keratinocytes.