Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Identification of Two Smad4 Proteins in Xenopus

Identification of Two Smad4 Proteins in Xenopus THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 17, Issue of April 23, pp. 12163–12170, 1999 © 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. THEIR COMMON AND DISTINCT PROPERTIES* (Received for publication, December 16, 1998, and in revised form, January 28, 1999) Norihisa Masuyama‡, Hiroshi Hanafusa‡, Morioh Kusakabe‡, Hiroshi Shibuya§, and Eisuke Nishida‡ From the ‡Department of Biophysics, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502 and the §Division of Morphogenesis, Department of Developmental Biology, National Institute for Basic Biology, Okazaki 444- 8585, Japan Smad family proteins have been identified as media- which is not a direct substrate of receptors but participates in tors of intracellular signal transduction by the trans- signaling by associating with pathway-restricted Smads. The forming growth factor-b (TGF-b) superfamily. Each third subtype consists of those Smads that inhibit the activa- member of the pathway-restricted, receptor-activated tion of pathway-restricted Smads and are referred to as anti- Smad family cooperates and synergizes with Smad4, Smads (6 – 8). called co-Smad, to transduce the signals. Only Smad4 Pathway-restricted Smads contain a consensus phosphoryl- has been shown able to function as a common partner of ation motif, SS(V/M)S, for the type I serine/threonine receptors the various pathway-restricted Smads in mammals. at their carboxyl termini. This class of Smads interacts tran- Here we have identified a novel Smad4-like molecule in siently with specific activated type I receptors and thus be- Xenopus (XSmad4b) as well as a Xenopus homolog of a comes phosphorylated following ligand stimulation. Smad2 well established Smad4 (XSmad4a). XSmad4b is 70% and Smad3 are specific mediators of TGF-b and activin signal- identical to XSmad4a in amino acid sequence. Both of ing, whereas Smad1, Smad5, and Smad8 are involved in the the Xenopus Smad4s can cooperate with Smad1 and bone morphogenetic protein (BMP) pathway (9 –22). Signaling Smad2, the pathway-restricted Smads specific for bone by pathway-restricted Smads requires an association with co- morphogenetic protein and TGF-b, respectively. How- Smad. Only Smad4 is known as co-Smad in mammals, where ever, they show distinct properties in terms of their developmental expression patterns, subcellular local- its structure is divergent from that of pathway-restricted izations, and phosphorylation states. Moreover, XS- Smads. Smad4 lacks a carboxyl-terminal phosphorylation mo- mad4b, but not XSmad4a, has the potent ability to in- tif and does not associate with the TGF-b receptors (14, 15, 23). duce ventralization when microinjected into the dorsal Each member of the various pathway-restricted Smads forms a marginal region of the 4-cell stage of the embryos. These complex with Smad4 upon ligand stimulation. Then, the het- results suggest that the two Xenopus Smad4s have over- eromeric complex is translocated to the nucleus where it par- lapping but distinct functions. ticipates in the transcriptional activation of specific target genes (9, 14, 15, 19, 20, 23–26). Smad4 was originally identified as the product of the Dpc4 Recent studies identified a family of proteins, termed Smads, (deleted in pancreatic cancer) tumor suppressor gene that is as essential components in intracellular signaling pathways mutated or deleted in a high proportion of pancreatic cancers downstream of serine/threonine kinase receptors for the and in a smaller proportion of other cancers (27). A general TGF-b superfamily (1–5). In vertebrates, at least nine differ- requirement of Smad4 is suggested not only in mammalian ent kinds of Smads have been identified. Smad proteins share cells but also in Xenopus embryos, as a dominant-negative a high degree of homology in their amino- and carboxyl-termi- Smad4 construct interferes with Smad1 and Smad2 signaling nal domains, the MH1 and MH2 domains, respectively, con- (15, 23, 28), although cDNA cloning of a full-length Xenopus nected with a divergent proline-rich linker region. Smad pro- Smad4 has not yet been performed. It has been reported that teins can be classified into three subtypes according to Smad4-deficient mice die early in embryogenesis (29, 30). structure and function (1–5). One subtype is the pathway- These mice exhibit severe defects in cellular proliferation, gas- restricted Smad, which is a direct substrate of type I receptors trulation, and mesoderm differentiation (29, 30). Thus, Smad4 for the TGF-b superfamily. The second subtype is the co-Smad, has been supposed to be a shared and obligate partner, partic- ipating in both TGF-b/activin and BMP signaling pathways. Here we report the identification, cDNA cloning, and char- * This work was supported by grants from the Ministry of Education, Science and Culture of Japan (to E. N.). The costs of publication of this acterization of two kinds of Smad4 in Xenopus. Both of the two article were defrayed in part by the payment of page charges. This Xenopus Smad4s, termed XSmad4a and XSmad4b, are shown article must therefore be hereby marked “advertisement” in accordance able to function as common partners of both Smad1 and Smad2 with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted in Xenopus embryos as well as in mammalian cells to transduce TM to the GenBank /EBI Data Bank with accession number(s) AB022721 respective signals from the TGF-b superfamily. Moreover, XS- and AB022722. mad4a and XSmad4b themselves form heteromeric as well as To whom correspondence should be addressed. Fax: 81-75-753- homomeric complexes. In addition to their amino acid sequence 4235; E-mail: [email protected]. The abbreviations used are: TGF-b, transforming growth factor-b; diversity, these two XSmad4s differ in their expression profiles BMP, bone morphogenetic protein; HA, hemagglutinin; RT-PCR, re- during early Xenopus development and in their subcellular verse transcription-coupled polymerase chain reaction; XSmad, Xeno- localization and phosphorylation states when expressed in cul- pus Smad; hSmad, human Smad; MH1, mad homology 1 (amino-termi- tured cells. Furthermore, microinjection of mRNA encoding nal domain); DAI, dorsoanterior index; MH2, mad homology 2 (carboxyl-terminal domain); DPC, deleted in pancreatic cancer. XSmad4b into dorsal cells of Xenopus embryos leads to strong This paper is available on line at http://www.jbc.org 12163 This is an Open Access article under the CC BY license. 12164 The Second Smad4 FIG.1. cDNA cloning of XSmad4a and XSmad4b. A, alignment of the de- duced amino acid sequences of XSmad4a, human Smad4/DPC4 (hSmad4), and XS- mad4b is shown. Residues that are iden- tical in three proteins are shaded black and those identical in two of three pro- teins are shaded gray. Dashes denote gaps in the alignments. The sequences were aligned using the ClustalW 1.6 pro- gram. B, the schematic diagram of overall structures of XSmad4a and XSmad4b is shown. reached stage 13. The primer pairs used here for reverse transcription- ventralization, whereas XSmad4a has little ventralizing activ- coupled polymerase chain reaction (RT-PCR) were reported elsewhere ity. These results suggest that the two Xenopus Smad4s may (23, 33, 34). Whole mount in situ hybridization was performed essen- function as co-Smads with distinct properties in transducing a tially as described (35). set of TGF-b superfamily signals. Cell Culture, Transfection, and Transcriptional Reporter Assay— C2C12 cells were maintained in Dulbecco’s modified Eagle’s medium EXPERIMENTAL PROCEDURES supplemented with 10% fetal calf serum. Cells were transfected at 24 h Molecular Cloning and Plasmid Construction—A Xenopus oocyte after seeding using the LipofectAMINE Plus reagent (Life Technolo- cDNA library (CLONTECH) was screened using the human Smad4 gies, Inc.). After 48 h, lysates were prepared, and the luciferase activity coding region as a probe. RNase protection assay was carried out using was determined with the luciferase assay system (Promega). Relative an Ambion HybSpeed RPA kit (Ambion) according to the manufactur- luciferase activities were normalized by co-expressed b-galactosidase er’s instruction. The entire coding regions of XSmad4a and XSmad4b, activities. the carboxyl-terminal truncated forms of XSmad4a and XSmad4b, and Immunoprecipitation, Immunoblotting, and Metabolic Labeling—Af- the carboxyl-terminal serine to alanine mutant of XSmad4b (AAVN) ter 12–15 h, cells were treated with or without 10 ng/ml of human were amplified by polymerase chain reaction, and the amplified nucle- TGF-b (purchased from Austral Biologicals) or 300 ng/ml BMP (Xeno- otide sequences were confirmed by DNA sequencing. For the carboxyl- pus BMP4 (36)) for 1 h, and subsequently lysates were prepared as terminal truncated forms of XSmad4a and XSmad4b, the nucleotides described (37). Immunoprecipitation was performed by incubation with corresponding to amino acid residues 1–508 and 1–519 were amplified, the 9E10 anti-Myc antibody (Santa Cruz Biotechnology) and protein respectively. XSmad4a cDNA and XSmad4b cDNA were ligated into G-Sepharose (Amersham Pharmacia Biotech). The immunoprecipitates pSP64T or into a Myc tag fused version of pSP64T plasmids to synthe- and the aliquots of total lysates were separated in SDS-polyacrylamide size mRNAs (31). In vitro synthesis of capped mRNA was performed gel electrophoresis and transferred to a polyvinylidene difluoride mem- using mMESSAGE mMACHINE (Ambion) according to the manufac- brane (Millipore). Membranes were incubated with antibodies against turer’s instruction. Other constructs for cell transfection were inserted Myc or HA (Santa Cruz Biotechnology) and subsequently with horse- into pcDL-SRa456, pSRa-HA, or pSRa-Myc. radish peroxidase-conjugated sheep anti-mouse antibody or donkey an- Xenopus Embryo Manipulation, In Situ Hybridization, and Animal ti-rabbit antibody (Amersham). Immunoreactive bands were detected Cap Assay—Xenopus embryos were obtained by in vitro fertilization of by the ECL Western blotting detection system (Amersham). eggs with testes homogenates. Embryos were staged according to Nieu- For metabolic labeling of C2C12 cells, 24 h post-transfection cells wkoop and Faber (32). The animal cap assay was performed as de- scribed elsewhere (33). Dorsal marginal zone explants were dissected at were incubated with [ P]orthophosphate for 3 h and treated with or the gastrula stage (stage 10) and were cultured until sibling embryos without 10 ng/ml of TGF-b and lysed in TNE buffer (10 mM Tris-HCl, 1 The Second Smad4 12165 FIG.2. The temporal and spatial expression of XSmad4a and XSmad4b. A, XSmad4a and XSmad4b transcripts are present during Xenopus early embryogenesis. Equivalent amounts of total RNA iso- lated from each stage of embryos were analyzed for the expression of XSmad4a and XSmad4b in an RNase protection assay. Numbers rep- resent the developmental stages (st.) (32): stages 1 and 6, maternal; stage 11, gastrula; stage 16, neurula; stage 23, tailbud; stages 28, 34, and 39, tadpole. The expression of ornithine decarboxylase (ODC) was also examined as a control for equal loading of RNA. B–G, the spatial expression patterns of XSmad4a and XSmad4b in developing Xenopus embryos are analyzed by whole mount in situ hybridization: B and E, animal view of early gastrula stage embryos (stage 10, ventral, at the top); C and F, dorsal view of neurula stage embryos (stage 20, anterior to the left); D and G, lateral view of tadpoles (stage 31, anterior to the left). pH 7.8, 150 mM NaCl, 1% Nonidet P-40, and 1 mM EDTA) with protease inhibitors. Myc-tagged Smads were immunoprecipitated with the 9E10 anti-Myc antibody (Santa Cruz Biotechnology) and protein G-Sepha- rose (Amersham Pharmacia Biotech). The precipitates and the aliquots of cell lysates were resolved by SDS-polyacrylamide gel electrophoresis and visualized by autoradiography. Immunofluorescence—TGF-b stimulation of C2C12 cells was pro- vided by co-transfecting the activated TGF-b type I receptor, TbR-I (T204D), and treated with 10 ng/ml TGF-b for 1 h (25). Cells were then fixed by formaldehyde. Immunostaining was performed by incubation with the 9E10 anti-Myc antibody (Santa Cruz Biotechnology) for 2 h followed by incubation with the fluorescein isothiocyanate-conjugated FIG.3. XSmad4a and XSmad4b synergize with XSmad1 and goat anti-mouse antibody (1:400) for 1 h. XSmad2 to induce expression of mesodermal marker genes. A, effect of co-expression of XSmad4a or XSmad4b on XSmad1- or XS- RESULTS mad2-induced expression of mesodermal maker genes in isolated ani- cDNA Cloning of Xenopus Smad4a and Xenopus Smad4b—By mal caps. Animal caps were dissected at the blastula stage from em- bryos that had been injected with XSmad4a or XSmad4b mRNA (0.2 screening a Xenopus oocyte cDNA library with human Smad4 ng) together with XSmad1 or XSmad2 mRNA (0.2 ng) at the 2-cell stage as a probe under low stringency, we isolated several positive and were cultured until sibling embryos reached stage 11 (upper panel) clones. Sequence analysis of these cDNA clones revealed that or 26 (lower panel). Expression of Xenopus brachyury (Xbra), goosecoid one of them has a very high homology (91% identity) to human (gsc), muscle-actin, and a-globin was analyzed by RT-PCR. Expression Smad4/DPC4 in the coding region and the other has a rela- of EF-1a was also analyzed as a loading control. No signal was observed in the absence of reverse transcription (2RT). B, effect of carboxyl- tively lower (71% identity), but still the highest, homology to terminal truncated mutants of XSmad4a and XSmad4b (XSmad4a human Smad4 among the other Smad family proteins reported DCand XSmad4bDC) on the mesodermal gene expression induced by to date. Therefore, we referred to the former as Xenopus XSmad1 or XSmad2 in animal caps. Animal caps were dissected at the Smad4a (XSmad4a) and the latter as Xenopus Smad4b (XS- blastula stage from embryos that had been injected with XSmad1 mRNA or XSmad2 mRNA (1 ng) together with XSmad4a DCmRNA or mad4b). We considered XSmad4a a Xenopus ortholog of mam- XSmad4b DCmRNA (1 ng) and were cultured until sibling embryos malian Smad4 and XSmad4b the second Smad4 (see below). reached stage 11 or 26. Expression of marker genes was analyzed by The nucleotide sequences were predicted to encode proteins of RT-PCR. 549 and 560 amino acids for XSmad4a and XSmad4b with calculated molecular masses of 60 and 61 kDa, respectively (data not shown). (Fig. 1A). It is unlikely that these two Smad4-related Xenopus The Smad proteins typically consist of three modules: highly cDNA clones were derived from pseudoalleles resulting from conserved amino-terminal (MH1) and carboxyl-terminal (MH2) the pseudotetraploid nature of Xenopus laevis genome, because domains and a poorly conserved linker region. The amino acid we obtained several cDNA clones encoding pseudogenes for sequence comparison between XSmad4a or XSmad4b and hu- both XSmad4a and XSmad4b from the same cDNA library man Smad4 (hSmad4) is shown in Fig. 1A. The MH1 and MH2 12166 The Second Smad4 FIG.4. XSmad4a and XSmad4b cooperate with XSmad1 and XSmad2 to induce the reporter gene expression under the BMP- and TGF-b-responsive promoters. C2C12 cells were transiently transfected with the Xvent2-Luc reporter plasmid and an expression vector encoding XSmad1 with or without either of the XSmad4s (A)or with the 3TP-Lux reporter plasmid and an expression vector encoding XSmad2 with or without either of the XSmad4s (B). Cells were har- vested 48 h after transfection and assayed for luciferase activity. These results are the averages of three separate experiments. WT, wild type; DC, carboxyl-terminal truncated mutant. domains of XSmad4a are 100% identical to those of hSmad4, except that the coding region of XSmad4a initiates at the position corresponding to the second methionine residue of hSmad4. The linker region shows 74% identity between XS- mad4a and hSmad4. In contrast, the identity between XS- mad4b and hSmad4 is about 90% in the MH1 and MH2 do- FIG.5. Association of XSmad4a and XSmad4b with XSmad1 mains and 34% in the linker region (Fig. 1B). Another striking and XSmad2. A, C2C12 cells were transfected with Myc-tagged wild difference between XSmad4a and XSmad4b is that the carbox- type (WT) or carboxyl-terminal truncated mutant (DC) of XSmad4a or yl-terminal sequence of XSmad4a is QPLD, like mammalian XSmad4b together with HA-tagged XSmad1 or XSmad2 and stimulated with 10 ng/ml TGF-b (T) or 300 ng/ml BMP (B) for 1 h. Complex Smad4, whereas that of XSmad4b is SSVN, which resembles formation was detected by immunoprecipitation (IP) with the anti-Myc the carboxyl-terminal SS(V/M)S phosphorylation motif of path- antibody followed by immunoblotting (IB) with the anti-HA antibody, way-restricted Smads (Fig. 1B). and the aliquots were also blotted with anti-Myc antibody to detect the To examine the temporal expression of XSmad4a and XS- expression of Myc-tagged XSmad4a and Myc-tagged XSmad4b. Ali- mad4b during early development, an RNase protection analy- quots of the cell lysates were directly analyzed by immunoblotting with anti-HA antibody. B, cells were transfected with Myc-XSmad4a or sis was performed with probes corresponding to each of the Myc-XSmad4b combined with HA-XSmad1 or HA-Smad2 and stimu- linker region of XSmad4s. Each probe was confirmed not to lated with BMP or TGF-b at indicated concentrations for 1 h. Oligomer- cross-hybridize to the other (data not shown). XSmad4a mRNA ization was detected by immunoprecipitation followed by immunoblot- was markedly increased by zygotic expression after the blas- ting. C, homomeric or heteromeric oligomer formation was detected by immunoprecipitation followed by immunoblotting from the lysates of tula stage, whereas XSmad4b mRNA was highly abundant in C2C12 cells transfected with HA- or Myc-tagged XSmad4a and eggs and was decreased during the gastrula stage, although XSmad4b. both transcripts were detected throughout early embryogenesis (Fig. 2A). Thus, XSmad4 a mRNA and XSmad4b mRNA are expressed differently during early development. The spatial expression of XSmad4a and XSmad4b was ex- The Second Smad4 12167 FIG.7. Phosphorylation of XSmad4b. C2C12 cells were trans- fected with Myc-tagged XSmad4a, XSmad4b, or the carboxyl-terminal mutated version of XSmad4b (AAVN). Cells were metabolically labeled with [ P]orthophosphate and further incubated with or without TGF-b (10 ng/ml) for 1 h. Phosphorylation of Myc-tagged XSmad4s was ana- lyzed by immunoprecipitation with anti-Myc (a-Myc) antibody followed by autoradiography. WT, wild type; IB, immunoblotting. genes was significantly suppressed (Fig. 3B). These results suggest that XSmad4a and XSmad4b can function as common partners for pathway-restricted Smads to induce expression of specific marker genes in Xenopus animal caps. To elucidate more directly the cooperativity of XSmad4a and XSmad4b with XSmad1 and XSmad2, we tested the ability of the Smads to induce the reporter gene expression under the BMP- and TGF-b-responsive promoters in cultured cells. FIG.6. Subcellular localization of XSmad4a and XSmad4b pro- Transfection of XSmad1 alone into C2C12 mouse myoblast cells teins. C2C12 cells were transfected with Myc-tagged XSmad4a or induced a low level of luciferase expression under the promoter XSmad4b together with or without XSmad2 and stimulated with TGF-b of Xvent-2, a BMP-inducible Xenopus homeobox gene (38, 39). by co-transfection with activated TGF-b type I receptor plus treatment with TGF-b (10 ng/ml) for 1 h. Then the cells were fixed and stained However, co-expression of XSmad4a and XSmad4b, but not with anti-Myc (aMyc) antibody and 49,6-diamidino-2-phenylindole di- XSmad4a DCor XSmad4bDC, together with XSmad1 induced a hydrochloride (DAPI). high level of the reporter gene expression (Fig. 4A). Similarly, both XSmad4a and XSmad4b synergized with XSmad2 to in- amined by whole mount in situ hybridization with the probes duce luciferase expression under the TGF-b-responsive 3TP (Fig. 2, B–G). Both XSmad4a and XSmad4b mRNAs are ubiq- reporter, whereas the carboxyl-terminal truncated constructs uitously expressed in the ectoderm at the early gastrula stage did not (Fig. 4B). These results demonstrate that XSmad4a and (Fig. 2, B and E) but become restricted to the neuroectoderm at XSmad4b act synergistically with XSmad1 and XSmad2 to the neurula stage (Fig. 2, C and F). In the tadpole stage, both induce gene expression in cultured cells. of the XSmad4s are expressed in the central nervous system, XSmad4a and XSmad4b Associate with XSmad1 and XS- eye, and otic vesicle (Fig. 2, D and G). XSmad4b is expressed mad2—As XSmad4a and XSmad4b functionally cooperated more anteriorly than XSmad4a in the neural tube. with XSmad1 and XSmad2, we would expect that XSmad4a XSmad4a and XSmad4b Synergize with XSmad1 and XS- and XSmad4b bind to XSmad1 and XSmad2 to form hetero- mad2—To examine whether XSmad4a and XSmad4b act as meric complexes. To examine this hypothesis, we co-expressed co-Smad to cooperate with Smad1 and Smad2 to mediate BMP- Myc-tagged XSmad4s with HA-tagged XSmad1 or XSmad2 in like and TGF-b/activin-like responses, respectively, we first C2C12 cells and subjected the obtained cell lysates to immu- analyzed the expression of several marker genes in animal caps noprecipitation with anti-Myc antibody followed by immuno- obtained from Xenopus embryos injected with synthetic blotting with anti-HA antibody. The results showed that in mRNAs encoding XSmad4a or XSmad4b combined with Xeno- response to BMP treatment, XSmad1 associates with either of pus Smad1 or Xenopus Smad2. Injection of a low dose of XS- the two XSmad4s (Fig. 5A, upper panel) and that in response to mad1, XSmad2, XSmad4a, or XSmad4b mRNA alone was not TGF-b stimulation, XSmad2 associates with either of the two sufficient to induce any noticeable expression of mesodermal XSmad4s (Fig. 5A, lower panel). Thus, both XSmad4a and genes (Fig. 3A). When XSmad1 mRNA was injected together XSmad4b are able to form heteromeric complexes with XS- with either XSmad4a or XSmad4b mRNA, however, strong mad1 or XSmad2 in response to specific stimulation. expression of a ventral mesodermal marker, a globin, was We next assessed whether the extent of the ligand-induced induced. Similarly, when XSmad2 mRNA was injected with association of XSmad4s with XSmad1 or XSmad2 would differ either of the XSmad4 mRNAs, dorsal mesodermal markers, between XSmad4a and XSmad4b. The association of XSmad2 goosecoid and muscle actin, and a pan-mesodermal marker, with XSmad4a occurred at the same level as with XSmad4b in Xenopus brachyury, were strongly induced (Fig. 3A). a manner dependent on the TGF-b concentration (Fig. 5B, Because the carboxyl-terminal truncated form of human lower panel). In contrast, the association of XSmad1 with XS- Smad4 has been shown to act as a dominant-negative inhibitor mad4b in response to BMP stimulation occurred more strongly for signal transductions (15, 28), we next examined the action than with XSmad4a (Fig. 5B, upper panel). This may suggest of carboxyl-terminal truncated constructs of XSmad4a and XS- that although both of the XSmad4s can function as common mad4b in the animal cap assay. Injection of a high dose of partners of XSmad1 and XSmad2, they do not have completely XSmad1 or XSmad2 mRNA was sufficient to induce expression redundant functions, and XSmad4b may play a major role in of ventral or dorsal mesodermal marker genes, respectively. BMP signaling. When either XSmad4a DC or XSmad4b DC mRNA was injected The crystallographic structure analysis of the MH2 domain along with XSmad1 or XSmad2, the expression of these marker of human Smad4 implied that Smad4 was able to form a ho- 12168 The Second Smad4 motrimer (40). To examine whether XSmad4a and XSmad4b form a heteromeric complex, we co-expressed both XSmad4a and XSmad4b tagged with different epitopes in C2C12 cells. Stimulation-independent hetero-oligomerization as well as ho- mo-oligomerization was observed (Fig. 5C). Therefore, it is possible that XSmad4a and XSmad4b form a heteromeric tri- mer. It may also be possible that after stimulation, XSmad4a, XSmad4b, and one of the pathway-restricted Smads form a heteromeric trimer. XSmad4b, but Not XSmad4a, Is Constitutively Nuclear— When expressed in C2C12 cells, XSmad4a and XSmad4b pro- teins showed different subcellular distribution. XSmad4a was present predominantly in the cytoplasm whereas XSmad4b was predominantly in the nucleus (Fig. 6). Because it has been demonstrated that human Smad4 and a Drosophila co-Smad, Medea, are present in the cytoplasm in the absence of stimu- lation (25, 41), the above result may again indicate that XS- mad4a is a Xenopus ortholog of a previously known Smad4. XSmad4a became localized to the nucleus in more than 60% of the cells after TGF-b stimulation when XSmad2 was co-ex- pressed (Fig. 6, top). Hence, the expression of XSmad2 is re- quired for XSmad4a to change its subcellular localization in response to TGF-b under the conditions. In contrast, subcellu- lar localization of XSmad4b did not change after TGF-b stim- ulation, irrespective of co-expression of XSmad2 (Fig. 6, bottom). XSmad4b, but Not XSmad4a, Is a Phosphoprotein—In addi- tion to the high degree of diversity in the linker region, another striking difference in the amino acid sequence between XS- mad4a and XSmad4b is found in their carboxyl termini. The SSVN sequence in XSmad4b (see Fig. 1B) is similar to the carboxyl-terminal phosphorylation motif (SS(V/M)S) of path- way-restricted Smads in which the last two serine residues undergo phosphorylation upon stimulation (42, 43). We then examined whether XSmad4b could be phosphorylated. Myc epitope-tagged XSmads were transfected into C2C12 cells, and the cells were labeled with [ P]orthophosphate and stimulated with TGF-b. An increase in XSmad2 phosphorylation after TGF-b treatment was observed, as determined by immunopre- cipitation with anti-Myc antibody followed by autoradiography (Fig. 7). No or little phosphorylation of XSmad4a was observed before or after TGF-b stimulation (Fig. 7). On the contrary, phosphorylation of XSmad4b was observed even before stimu- lation, and its level did not increase after TGF-b treatment (Fig. 7). To explore the possible involvement of the SSVN sequence of XSmad4b in its phosphorylation, we constructed a mutant XSmad4b having an AAVN sequence instead of SSVN at its carboxyl terminus. This mutant was still phosphorylated irrespective of TGF-b stimulation, and no decrease in the phos- phorylation level was observed by this mutation (Fig. 7). There- fore, the carboxyl-terminal SSVN sequence of XSmad4b is not phosphorylated, and XSmad4b may not be a direct substrate of completely and those with a DAI of 5 are normal. C, immunoblotting analysis of the exogenously expressed XSmad4a and XSmad4b.N- terminal Myc-tagged XSmad4a and XSmad4b mRNAs were injected at the 4-cell stage embryos at the indicated doses, and extracts were obtained at the blastula stage (stage 9). Expressed proteins were de- FIG.8. XSmad4b induces ventralization of embryos. A, tadpole tected by immunoblotting with anti-Myc antibody. Myc-tagged stage (stage 35) Xenopus embryos and sibling embryos that have been XSmad4s had essentially the same effect as nontagged constructs on injected dorsally with XSmad4a mRNA or XSmad4b mRNA at the 4-cell the phenotypes of embryos (data not shown). D, expression of marker stage at the indicated doses are shown (Dorsal). Embryos injected genes in dorsal marginal zone explants. Dorsal marginal zone explants ventrally with XSmad4a or XSmad4b mRNA (2 ng) are also shown were dissected at the early gastrula stage from embryos that had been (Ventral). B, semiquantification of ventralization of the embryos by injected dorsally with XSmad4a, XSmad4b, XSmad1, or XSmad2 XSmad4b. At the 4-cell stage, two dorsal blastomeres were injected mRNA (each at 2 ng) at the 4-cell stage and were cultured until sibling with XSmad4a mRNA or XSmad4b mRNA at the indicated doses. The embryos reached the midgastrula stage. Expression of indicated DAI of the embryos was scored after 2 days, and the average DAI for marker genes was analyzed by RT-PCR. Expression of EF-1a was also each sample is shown. Numbers of embryos examined are indicated analyzed as a loading control. No signal was observed in the absence of above the figure. Embryos with a DAI of 0 lack dorsal structures reverse transcription (2RT). gsc, goosecoid; Xbra, Xenopus brachyury. The Second Smad4 12169 the TGF-b receptor. family (45). Moreover, lack of Smad4 in colon cancer cells leads XSmad4b Induces Ventralization of Xenopus Embryos—To to an increase in metastasis and malignancy, which suggests find the functional difference between the two XSmad4s, we that Smad4 has a tumor-suppressing function (46). These ob- tested the effect of expression of XSmad4 in Xenopus embryos. servations suggest that if mammals have the second co-Smad, Injection of XSmad4b mRNA, but not XSmad4a mRNA, into its function may be distinct from that of Smad4/DPC4. two dorsal cells of the 4-cell stage embryos led to strong ven- The Xenopus system has been extensively used to elucidate tralization, as revealed by the defects in anterior structures at the cellular signaling mechanism of growth factors, including the late stage (Fig. 8A). Similar phenotypes were reported to be the TGF-b superfamily, that control cell differentiation and induced by expressing Smad1 or Smad5 (16, 22). Injection of pattern formation during early embryogenesis (47–52). It has either of the mRNAs into the ventral side had little or no effect been suggested that activin, nodal, and Vg1, members of the on the embryonic development (Fig. 8A). The ventralizing effect TGF-b superfamily, are involved in the differentiation of dorsal of XSmad4b, when expressed in the dorsal marginal region, mesodermal tissues, whereas the BMP family ligands regulate was dose-dependent. This was clearly demonstrated by semi- the ventral mesoderm differentiation in Xenopus early devel- quantification by scoring the dorsoanterior index (DAI) of the opment. Xenopus homologs of Smad1 and Smad2 have been embryos (44), where 5 represents a normal embryo and 0 indi- identified and shown to mediate activin and BMP signaling cates an embryo lacking axial structures (Fig. 8B). We con- pathways, respectively (10, 12, 16, 53). Interestingly, Candia et firmed that XSmad4a and XSmad4b were expressed at almost al. (39) observed that the expression of an excess amount of the same level in this series of experiments (Fig. 8C). In addi- human Smad4 in Xenopus embryos compromised the antago- tion to the defects in anterior structures, injection of XSmad4b nism between the activin/Vg1 and BMP pathways. They pro- mRNA into dorsal marginal cells induced an increase in the posed an attractive model, which supports the theory that the expression of a ventral mesodermal marker, Xvent-2, and a activin/Vg1 and BMP pathways modulate each other’s activity decrease in the expression of a dorsal mesodermal marker, by sequestering a limited pool of Smad4, which commonly goosecoid, more strongly than did XSmad4a mRNA, as ana- participates in both pathways by associating with Smad2 and lyzed in isolated dorsal marginal zone explants (Fig. 8D). These Smad1, respectively. As human Smad4 corresponds to XS- results indicate that the two XSmad4 proteins have distinct mad4a, XSmad4a may be commonly used as a co-Smad for both abilities to induce ventralization in Xenopus embryos. pathways, or a putative heteromeric complex between XS- mad4a and XSmad4b may be a common partner for both XS- DISCUSSION mad1 and XSmad2. These considerations may be consistent Smad4/DPC4 has been shown to be required as a common with our idea that XSmad4a and XSmad4b have overlapping partner for pathway-restricted Smads to propagate the TGF-b but somewhat distinct functions as co-Smads. Thus the obser- family signals from the cell surface receptors to the nucleus by vations of Candia et al. are not inconsistent with our idea that forming heteromeric complexes. Here we have identified a XSmad4b may be rather preferentially used for the BMP path- novel Smad4-like molecule, XSmad4b, as well as a Xenopus way, which was derived from our results showing that XS- homolog of mammalian Smad4 (XSmad4a), and shown that mad4b binds to XSmad1 more tightly than does XSmad4a and both of the XSmad4s are able to function as co-Smad to coop- that XSmad4b has the more potent ventralizing activity. erate and synergize with XSmad1 and XSmad2. However, they Although we still do not know the molecular mechanism that have distinct properties in terms of their temporal expression defines the distinct ability between XSmad4a and XSmad4b in during early embryogenesis, subcellular distribution, phospho- cooperation with XSmad1 and XSmad2, it is likely that some rylation state, and ventralizing activity in Xenopus embryos. factors interact specifically with each of the XSmad4s. It is also Therefore, it is likely that two co-Smads, XSmad4a and XS- possible that phosphorylation of XSmad4b might have some mad4b, play both overlapping and distinct roles in transducing role in regulating the function of XSmad4b. We are currently a set of TGF-b superfamily signals in Xenopus. investigating the mechanisms that underlie the difference be- We reasoned from their subcellular distribution as well as tween XSmad4a and XSmad4b in their cooperativity with their amino acid sequence similarity to human or mouse pathway-restricted Smads, as well as the molecular basis for Smad4 that XSmad4a was a Xenopus ortholog of mammalian determining their subcellular localization. Smad4 and XSmad4b was a novel homolog. XSmad4a localized Acknowledgments—We thank Dr. Douglas A. Melton for providing predominantly in the cytoplasm and translocated to the nu- Xenopus Smad1 and Smad2 plasmids and Dr. Ken W. Y. Cho for the cleus after stimulation when a pathway-restricted Smad was Xvent2-Luc plasmid. We also thank K. Kawachi and M. Watanabe for co-expressed (Fig. 6). This behavior is the same as that previ- technical assistance. ously reported for human Smad4 (25). In contrast, XSmad4b REFERENCES was present predominantly in the nucleus before or after stim- ulation. The subcellular distribution of either of the XSmad4s 1. Derynck, R., and Feng, X. H. (1997) Biochim. Biophys. Acta 1333, F105–F150 2. Heldin, C. H., Miyazono, K., and ten Dijke, P. (1997) Nature 390, 465– 471 was not affected when the other XSmad4 was co-expressed in 3. Attisano, L., and Wrana, J. L. (1998) Curr. Opin. Cell Biol. 10, 188 –194 cells (data not shown), although they could bind to each other 4. Massague, J. (1998) Annu. Rev. Biochem. 67, 753–791 5. Whitman, M. (1998) Genes Dev. 12, 2445–2462 to form the XSmad4a-XSmad4b complex in vitro (Fig. 5C). 6. Hayashi, H., Abdollah, S., Qiu, Y., Cai, J., Xu, Y., Grinnell, B. W., Richardson, Although we do not know the functional significance of their M. A., Topper, J. N., Gimbrone, M. A., Jr., Wrana, J. L., and Falb, D. (1997) different subcellular localizations, the possible formation of the Cell 89, 1165–1173 7. Imamura, T., Takase, M., Nishihara, A., Oeda, E., Hanai, J., Kawabata, M., heteromeric complex between XSmad4a and XSmad4b after and Miyazono, K. (1997) Nature 389, 622– 626 the nuclear entry of XSmad4a in response to stimulation might 8. Nakao, A., Afrakhte, M., Moren, A., Nakayama, T., Christian, J. L., Heuchel, R., Itoh, S., Kawabata, M., Heldin, N. E., Heldin, C. H., and ten Dijke, P. have some physiological significance. (1997) Nature 389, 631– 635 An intriguing question is whether or not the existence of the 9. Hoodless, P. A., Haerry, T., Abdollah, S., Stapleton, M., O’Connor, M. B., second co-Smad is specific to Xenopus. At present, there is no Attisano, L., and Wrana, J. L. (1996) Cell 85, 489 –500 10. Graff, J. M., Bansal, A., and Melton, D. A. (1996) Cell 85, 479 – 487 evidence for the existence of another Smad4-like gene in mam- 11. Eppert, K., Scherer, S. W., Ozcelik, H., Pirone, R., Hoodless, P., Kim, H., Tsui, mals. We could not find any related genes in the data base of L. C., Bapat, B., Gallinger, S., Andrulis, I. L., Thomsen, G. H., Wrana, J. L., and Attisano, L. (1996) Cell 86, 543–552 expressed sequence tags (dbEST). Smad4 knockout mice were 12. Baker, J. C., and Harland, R. M. (1996) Genes Dev. 10, 1880 –1889 shown to die early in their development (29, 30), and the 13. Liu, F., Hata, A., Baker, J. C., Doody, J., Carcamo, J., Harland, R. M., and Smad4-deficient cells lost responsiveness to the TGF-b super- Massague, J. (1996) Nature 381, 620 – 623 12170 The Second Smad4 14. Macias-Silva, M., Abdollah, S., Hoodless, P. A., Pirone, R., Attisano, L., and 34. Shibuya, H., Iwata, H., Masuyama, N., Gotoh, Y., Yamaguchi, K., Irie, K., Wrana, J. L. (1996) Cell 87, 1215–1224 Matsumoto, K., Nishida, E., and Ueno, N. (1998) EMBO J. 17, 1019 –1028 15. Zhang, Y., Feng, X., We, R., and Derynck, R. (1996) Nature 383, 168 –172 35. Harland, R. M. (1991) Methods Enzymol. 36, 675– 685 16. Thomsen, G. H. (1996) Development 122, 2359 –2366 36. Iemura, S., Yamamoto, T. S., Takagi, C., Uchiyama, H., Natsume, T., Shi- 17. Yingling, J. M., Das, P., Savage, C., Zhang, C., Padgett, R. W., and Wang, X.-F. masaki, S., Sugino, H., and Ueno, N. (1998) Proc. Natl. Acad. Sci. U. S. A. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 8940 – 8944 95, 9337–9342 18. Lechleider, R. J., de Caestecker, M. P., Dehejia, A., Polymeropoulos, M. H., and 37. Moriguchi, T., Kuroyanagi, N., Yamaguchi, K., Gotoh, Y., Irie, K., Kano, T., Roberts, A. B. (1996) J. Biol. Chem. 271, 17617–17620 Shirakabe, K., Muro, Y., Shibuya, H., Matsumoto, K., Nishida, E., and 19. Kretzschmar, M., Liu, F., Hata, A., Doody, J., and Massague, J. (1997) Genes Hagiwara, M. (1996) J. Biol. Chem. 271, 13675–13679 Dev. 11, 984 –995 38. Onichtchouk, D., Gawantka, V., Dosch, R., Delius, H., Hirschfeld, K., Blumen- 20. Nakao, A., Imamura, T., Souchelnytskyi, S., Kawabata, M., Ishisaki, A., Oeda, stock, C., and Niehrs, C. (1996) Development 122, 3045–3053 E., Tamaki, K., Hanai, J., Heldin, C. H., Miyazono, K., and ten Dijke, P. 39. Candia, A. F., Watabe, T., Hawley, S. H., Onichtchouk, D., Zhang, Y., Derynck, (1997) EMBO J. 16, 5353–5362 R., Niehrs, C., and Cho, K. W. (1997) Development 124, 4467– 4480 21. Chen, Y., Bhushan, A., and Vale, W. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 40. Shi, Y., Hata, A., Lo, R. S., Massague, J., and Pavletich, N. P. (1997) Nature 12938 –12943 388, 87–93 22. Suzuki, A., Chang, C., Yingling, J. M., Wang, X. F., and Hemmati-Brivanlou, 41. Wisotzkey, R. G., Mehra, A., Sutherland, D. J., Dobens, L. L., Liu, X., Dohr- A. (1997) Dev. Biol. 184, 402– 405 mann, C., Attisano, L., and Raftery, L. A. (1998) Development 125, 23. Lagna, G., Hata, A., Hemmati-Brivanlou, A., and Massague, J. (1996) Nature 1433–1445 383, 832– 836 42. Abdollah, S., Macias-Silva, M., Tsukazaki, T., Hayashi, H., Attisano, L., and 24. de Caestecker, M. P., Hemmati, P., Larisch-Bloch, S., Ajmera, R., Roberts, Wrana, J. L. (1997) J. Biol. Chem. 272, 27678 –27685 A. B., and Lechleider, R. J. (1997) J. Biol. Chem. 272, 13690 –13696 43. Souchelnytskyi, S., Tamaki, K., Engstrom, U., Wernstedt, C., ten Dijke, P., and 25. Liu, F., Pouponnot, C., and Massague, J. (1997) Genes Dev. 11, 3157–3167 Heldin, C. H. (1997) J. Biol. Chem. 272, 28107–28115 26. Yingling, J. M., Datto, M. B., Wong, C., Frederick, J. P., Liberati, N. T., and 44. Kao, K. R., and Elinson, R. P. (1988) Dev. Biol. 127, 64 –77 Wang, X. F. (1997) Mol. Cell. Biol. 17, 7019 –7028 45. Zhou, S., Buckhaults, P., Zawel, L., Bunz, F., Riggins, G., Le Dai, J., Kern, 27. Hahn, S. A., Schutte, M., Hoque, A. T., Moskaluk, C. A., da Costa, L. T., S. E., Kinzler, K. W., and Vogelstein, B. (1998) Proc. Natl. Acad. Sci. Rozenblum, E., Weinstein, C. L., Fischer, A., Yeo, C. J., Hruban, R. H., and U. S. A. 95, 2412–2416 Kern, S. E. (1996) Science 271, 350 –353 46. Takaku, K., Oshima, M., Miyoshi, H., Matsui, M., Seldin, M. F., and Taketo, 28. Zhang, Y., Musci, T., and Derynck, R. (1997) Curr. Biol. 7, 270 –276 M. M. (1998) Cell 92, 645– 656 29. Sirard, C., de la Pompa, J. L., Elia, A., Itie, A., Mirtsos, C., Cheung, A., Hahn, 47. Sive, H. L. (1993) Genes Dev. 7, 1–12 S., Wakeham, A., Schwartz, L., Kern, S. E., Rossant, J., and Mak, T. W. 48. Dawid, I. B. (1994) J. Biol. Chem. 269, 6259 – 6262 (1998) Genes Dev. 12, 107–119 49. Kessler, D. S., and Melton, D. A. (1994) Science 266, 596 – 604 30. Yang, X., Li, C., Xu, X., and Deng, C. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 50. Smith, J. C. (1995) Curr. Opin. Cell Biol. 7, 856 – 861 3667–3672 51. Heasman, J. (1997) Development 124, 4179 – 4191 31. Krieg, P. A., and Melton, D. A. (1984) Nucleic Acids Res. 12, 7057–7070 52. Sasai, Y., and De Robertis, E. M. (1997) Dev. Biol. 182, 5–20 32. Nieuwkoop, P. D., and Faber, J. (1967) Normal Table of Xenopus laevis (Dau- din), North Holland Publishing Co., Amsterdam 53. Meersseman, G., Verschueren, K., Nelles, L., Blumenstock, C., Kraft, H., Wuytens, G., Remacle, J., Kozak, C. A., Tylzanowski, P., Niehrs, C., and 33. Gotoh, Y., Masuyama, N., Suzuki, A., Ueno, N., and Nishida, E. (1995) EMBO J. 14, 2491–2498 Huylebroeck, D. (1997) Mech. Dev. 61, 127–140 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Identification of Two Smad4 Proteins in Xenopus

Masuyama, Norihisa; Hanafusa, Hiroshi; Kusakabe, Morioh; Shibuya, Hiroshi; Nishida, Eisuke
Journal of Biological Chemistry – Apr 1, 1999
Download PDF
8 pages

Loading next page...
 
/lp/unpaywall/identification-of-two-smad4-proteins-in-xenopus-G34u82DoRo

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.274.17.12163
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 17, Issue of April 23, pp. 12163–12170, 1999 © 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. THEIR COMMON AND DISTINCT PROPERTIES* (Received for publication, December 16, 1998, and in revised form, January 28, 1999) Norihisa Masuyama‡, Hiroshi Hanafusa‡, Morioh Kusakabe‡, Hiroshi Shibuya§, and Eisuke Nishida‡ From the ‡Department of Biophysics, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502 and the §Division of Morphogenesis, Department of Developmental Biology, National Institute for Basic Biology, Okazaki 444- 8585, Japan Smad family proteins have been identified as media- which is not a direct substrate of receptors but participates in tors of intracellular signal transduction by the trans- signaling by associating with pathway-restricted Smads. The forming growth factor-b (TGF-b) superfamily. Each third subtype consists of those Smads that inhibit the activa- member of the pathway-restricted, receptor-activated tion of pathway-restricted Smads and are referred to as anti- Smad family cooperates and synergizes with Smad4, Smads (6 – 8). called co-Smad, to transduce the signals. Only Smad4 Pathway-restricted Smads contain a consensus phosphoryl- has been shown able to function as a common partner of ation motif, SS(V/M)S, for the type I serine/threonine receptors the various pathway-restricted Smads in mammals. at their carboxyl termini. This class of Smads interacts tran- Here we have identified a novel Smad4-like molecule in siently with specific activated type I receptors and thus be- Xenopus (XSmad4b) as well as a Xenopus homolog of a comes phosphorylated following ligand stimulation. Smad2 well established Smad4 (XSmad4a). XSmad4b is 70% and Smad3 are specific mediators of TGF-b and activin signal- identical to XSmad4a in amino acid sequence. Both of ing, whereas Smad1, Smad5, and Smad8 are involved in the the Xenopus Smad4s can cooperate with Smad1 and bone morphogenetic protein (BMP) pathway (9 –22). Signaling Smad2, the pathway-restricted Smads specific for bone by pathway-restricted Smads requires an association with co- morphogenetic protein and TGF-b, respectively. How- Smad. Only Smad4 is known as co-Smad in mammals, where ever, they show distinct properties in terms of their developmental expression patterns, subcellular local- its structure is divergent from that of pathway-restricted izations, and phosphorylation states. Moreover, XS- Smads. Smad4 lacks a carboxyl-terminal phosphorylation mo- mad4b, but not XSmad4a, has the potent ability to in- tif and does not associate with the TGF-b receptors (14, 15, 23). duce ventralization when microinjected into the dorsal Each member of the various pathway-restricted Smads forms a marginal region of the 4-cell stage of the embryos. These complex with Smad4 upon ligand stimulation. Then, the het- results suggest that the two Xenopus Smad4s have over- eromeric complex is translocated to the nucleus where it par- lapping but distinct functions. ticipates in the transcriptional activation of specific target genes (9, 14, 15, 19, 20, 23–26). Smad4 was originally identified as the product of the Dpc4 Recent studies identified a family of proteins, termed Smads, (deleted in pancreatic cancer) tumor suppressor gene that is as essential components in intracellular signaling pathways mutated or deleted in a high proportion of pancreatic cancers downstream of serine/threonine kinase receptors for the and in a smaller proportion of other cancers (27). A general TGF-b superfamily (1–5). In vertebrates, at least nine differ- requirement of Smad4 is suggested not only in mammalian ent kinds of Smads have been identified. Smad proteins share cells but also in Xenopus embryos, as a dominant-negative a high degree of homology in their amino- and carboxyl-termi- Smad4 construct interferes with Smad1 and Smad2 signaling nal domains, the MH1 and MH2 domains, respectively, con- (15, 23, 28), although cDNA cloning of a full-length Xenopus nected with a divergent proline-rich linker region. Smad pro- Smad4 has not yet been performed. It has been reported that teins can be classified into three subtypes according to Smad4-deficient mice die early in embryogenesis (29, 30). structure and function (1–5). One subtype is the pathway- These mice exhibit severe defects in cellular proliferation, gas- restricted Smad, which is a direct substrate of type I receptors trulation, and mesoderm differentiation (29, 30). Thus, Smad4 for the TGF-b superfamily. The second subtype is the co-Smad, has been supposed to be a shared and obligate partner, partic- ipating in both TGF-b/activin and BMP signaling pathways. Here we report the identification, cDNA cloning, and char- * This work was supported by grants from the Ministry of Education, Science and Culture of Japan (to E. N.). The costs of publication of this acterization of two kinds of Smad4 in Xenopus. Both of the two article were defrayed in part by the payment of page charges. This Xenopus Smad4s, termed XSmad4a and XSmad4b, are shown article must therefore be hereby marked “advertisement” in accordance able to function as common partners of both Smad1 and Smad2 with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted in Xenopus embryos as well as in mammalian cells to transduce TM to the GenBank /EBI Data Bank with accession number(s) AB022721 respective signals from the TGF-b superfamily. Moreover, XS- and AB022722. mad4a and XSmad4b themselves form heteromeric as well as To whom correspondence should be addressed. Fax: 81-75-753- homomeric complexes. In addition to their amino acid sequence 4235; E-mail: [email protected]. The abbreviations used are: TGF-b, transforming growth factor-b; diversity, these two XSmad4s differ in their expression profiles BMP, bone morphogenetic protein; HA, hemagglutinin; RT-PCR, re- during early Xenopus development and in their subcellular verse transcription-coupled polymerase chain reaction; XSmad, Xeno- localization and phosphorylation states when expressed in cul- pus Smad; hSmad, human Smad; MH1, mad homology 1 (amino-termi- tured cells. Furthermore, microinjection of mRNA encoding nal domain); DAI, dorsoanterior index; MH2, mad homology 2 (carboxyl-terminal domain); DPC, deleted in pancreatic cancer. XSmad4b into dorsal cells of Xenopus embryos leads to strong This paper is available on line at http://www.jbc.org 12163 This is an Open Access article under the CC BY license. 12164 The Second Smad4 FIG.1. cDNA cloning of XSmad4a and XSmad4b. A, alignment of the de- duced amino acid sequences of XSmad4a, human Smad4/DPC4 (hSmad4), and XS- mad4b is shown. Residues that are iden- tical in three proteins are shaded black and those identical in two of three pro- teins are shaded gray. Dashes denote gaps in the alignments. The sequences were aligned using the ClustalW 1.6 pro- gram. B, the schematic diagram of overall structures of XSmad4a and XSmad4b is shown. reached stage 13. The primer pairs used here for reverse transcription- ventralization, whereas XSmad4a has little ventralizing activ- coupled polymerase chain reaction (RT-PCR) were reported elsewhere ity. These results suggest that the two Xenopus Smad4s may (23, 33, 34). Whole mount in situ hybridization was performed essen- function as co-Smads with distinct properties in transducing a tially as described (35). set of TGF-b superfamily signals. Cell Culture, Transfection, and Transcriptional Reporter Assay— C2C12 cells were maintained in Dulbecco’s modified Eagle’s medium EXPERIMENTAL PROCEDURES supplemented with 10% fetal calf serum. Cells were transfected at 24 h Molecular Cloning and Plasmid Construction—A Xenopus oocyte after seeding using the LipofectAMINE Plus reagent (Life Technolo- cDNA library (CLONTECH) was screened using the human Smad4 gies, Inc.). After 48 h, lysates were prepared, and the luciferase activity coding region as a probe. RNase protection assay was carried out using was determined with the luciferase assay system (Promega). Relative an Ambion HybSpeed RPA kit (Ambion) according to the manufactur- luciferase activities were normalized by co-expressed b-galactosidase er’s instruction. The entire coding regions of XSmad4a and XSmad4b, activities. the carboxyl-terminal truncated forms of XSmad4a and XSmad4b, and Immunoprecipitation, Immunoblotting, and Metabolic Labeling—Af- the carboxyl-terminal serine to alanine mutant of XSmad4b (AAVN) ter 12–15 h, cells were treated with or without 10 ng/ml of human were amplified by polymerase chain reaction, and the amplified nucle- TGF-b (purchased from Austral Biologicals) or 300 ng/ml BMP (Xeno- otide sequences were confirmed by DNA sequencing. For the carboxyl- pus BMP4 (36)) for 1 h, and subsequently lysates were prepared as terminal truncated forms of XSmad4a and XSmad4b, the nucleotides described (37). Immunoprecipitation was performed by incubation with corresponding to amino acid residues 1–508 and 1–519 were amplified, the 9E10 anti-Myc antibody (Santa Cruz Biotechnology) and protein respectively. XSmad4a cDNA and XSmad4b cDNA were ligated into G-Sepharose (Amersham Pharmacia Biotech). The immunoprecipitates pSP64T or into a Myc tag fused version of pSP64T plasmids to synthe- and the aliquots of total lysates were separated in SDS-polyacrylamide size mRNAs (31). In vitro synthesis of capped mRNA was performed gel electrophoresis and transferred to a polyvinylidene difluoride mem- using mMESSAGE mMACHINE (Ambion) according to the manufac- brane (Millipore). Membranes were incubated with antibodies against turer’s instruction. Other constructs for cell transfection were inserted Myc or HA (Santa Cruz Biotechnology) and subsequently with horse- into pcDL-SRa456, pSRa-HA, or pSRa-Myc. radish peroxidase-conjugated sheep anti-mouse antibody or donkey an- Xenopus Embryo Manipulation, In Situ Hybridization, and Animal ti-rabbit antibody (Amersham). Immunoreactive bands were detected Cap Assay—Xenopus embryos were obtained by in vitro fertilization of by the ECL Western blotting detection system (Amersham). eggs with testes homogenates. Embryos were staged according to Nieu- For metabolic labeling of C2C12 cells, 24 h post-transfection cells wkoop and Faber (32). The animal cap assay was performed as de- scribed elsewhere (33). Dorsal marginal zone explants were dissected at were incubated with [ P]orthophosphate for 3 h and treated with or the gastrula stage (stage 10) and were cultured until sibling embryos without 10 ng/ml of TGF-b and lysed in TNE buffer (10 mM Tris-HCl, 1 The Second Smad4 12165 FIG.2. The temporal and spatial expression of XSmad4a and XSmad4b. A, XSmad4a and XSmad4b transcripts are present during Xenopus early embryogenesis. Equivalent amounts of total RNA iso- lated from each stage of embryos were analyzed for the expression of XSmad4a and XSmad4b in an RNase protection assay. Numbers rep- resent the developmental stages (st.) (32): stages 1 and 6, maternal; stage 11, gastrula; stage 16, neurula; stage 23, tailbud; stages 28, 34, and 39, tadpole. The expression of ornithine decarboxylase (ODC) was also examined as a control for equal loading of RNA. B–G, the spatial expression patterns of XSmad4a and XSmad4b in developing Xenopus embryos are analyzed by whole mount in situ hybridization: B and E, animal view of early gastrula stage embryos (stage 10, ventral, at the top); C and F, dorsal view of neurula stage embryos (stage 20, anterior to the left); D and G, lateral view of tadpoles (stage 31, anterior to the left). pH 7.8, 150 mM NaCl, 1% Nonidet P-40, and 1 mM EDTA) with protease inhibitors. Myc-tagged Smads were immunoprecipitated with the 9E10 anti-Myc antibody (Santa Cruz Biotechnology) and protein G-Sepha- rose (Amersham Pharmacia Biotech). The precipitates and the aliquots of cell lysates were resolved by SDS-polyacrylamide gel electrophoresis and visualized by autoradiography. Immunofluorescence—TGF-b stimulation of C2C12 cells was pro- vided by co-transfecting the activated TGF-b type I receptor, TbR-I (T204D), and treated with 10 ng/ml TGF-b for 1 h (25). Cells were then fixed by formaldehyde. Immunostaining was performed by incubation with the 9E10 anti-Myc antibody (Santa Cruz Biotechnology) for 2 h followed by incubation with the fluorescein isothiocyanate-conjugated FIG.3. XSmad4a and XSmad4b synergize with XSmad1 and goat anti-mouse antibody (1:400) for 1 h. XSmad2 to induce expression of mesodermal marker genes. A, effect of co-expression of XSmad4a or XSmad4b on XSmad1- or XS- RESULTS mad2-induced expression of mesodermal maker genes in isolated ani- cDNA Cloning of Xenopus Smad4a and Xenopus Smad4b—By mal caps. Animal caps were dissected at the blastula stage from em- bryos that had been injected with XSmad4a or XSmad4b mRNA (0.2 screening a Xenopus oocyte cDNA library with human Smad4 ng) together with XSmad1 or XSmad2 mRNA (0.2 ng) at the 2-cell stage as a probe under low stringency, we isolated several positive and were cultured until sibling embryos reached stage 11 (upper panel) clones. Sequence analysis of these cDNA clones revealed that or 26 (lower panel). Expression of Xenopus brachyury (Xbra), goosecoid one of them has a very high homology (91% identity) to human (gsc), muscle-actin, and a-globin was analyzed by RT-PCR. Expression Smad4/DPC4 in the coding region and the other has a rela- of EF-1a was also analyzed as a loading control. No signal was observed in the absence of reverse transcription (2RT). B, effect of carboxyl- tively lower (71% identity), but still the highest, homology to terminal truncated mutants of XSmad4a and XSmad4b (XSmad4a human Smad4 among the other Smad family proteins reported DCand XSmad4bDC) on the mesodermal gene expression induced by to date. Therefore, we referred to the former as Xenopus XSmad1 or XSmad2 in animal caps. Animal caps were dissected at the Smad4a (XSmad4a) and the latter as Xenopus Smad4b (XS- blastula stage from embryos that had been injected with XSmad1 mRNA or XSmad2 mRNA (1 ng) together with XSmad4a DCmRNA or mad4b). We considered XSmad4a a Xenopus ortholog of mam- XSmad4b DCmRNA (1 ng) and were cultured until sibling embryos malian Smad4 and XSmad4b the second Smad4 (see below). reached stage 11 or 26. Expression of marker genes was analyzed by The nucleotide sequences were predicted to encode proteins of RT-PCR. 549 and 560 amino acids for XSmad4a and XSmad4b with calculated molecular masses of 60 and 61 kDa, respectively (data not shown). (Fig. 1A). It is unlikely that these two Smad4-related Xenopus The Smad proteins typically consist of three modules: highly cDNA clones were derived from pseudoalleles resulting from conserved amino-terminal (MH1) and carboxyl-terminal (MH2) the pseudotetraploid nature of Xenopus laevis genome, because domains and a poorly conserved linker region. The amino acid we obtained several cDNA clones encoding pseudogenes for sequence comparison between XSmad4a or XSmad4b and hu- both XSmad4a and XSmad4b from the same cDNA library man Smad4 (hSmad4) is shown in Fig. 1A. The MH1 and MH2 12166 The Second Smad4 FIG.4. XSmad4a and XSmad4b cooperate with XSmad1 and XSmad2 to induce the reporter gene expression under the BMP- and TGF-b-responsive promoters. C2C12 cells were transiently transfected with the Xvent2-Luc reporter plasmid and an expression vector encoding XSmad1 with or without either of the XSmad4s (A)or with the 3TP-Lux reporter plasmid and an expression vector encoding XSmad2 with or without either of the XSmad4s (B). Cells were har- vested 48 h after transfection and assayed for luciferase activity. These results are the averages of three separate experiments. WT, wild type; DC, carboxyl-terminal truncated mutant. domains of XSmad4a are 100% identical to those of hSmad4, except that the coding region of XSmad4a initiates at the position corresponding to the second methionine residue of hSmad4. The linker region shows 74% identity between XS- mad4a and hSmad4. In contrast, the identity between XS- mad4b and hSmad4 is about 90% in the MH1 and MH2 do- FIG.5. Association of XSmad4a and XSmad4b with XSmad1 mains and 34% in the linker region (Fig. 1B). Another striking and XSmad2. A, C2C12 cells were transfected with Myc-tagged wild difference between XSmad4a and XSmad4b is that the carbox- type (WT) or carboxyl-terminal truncated mutant (DC) of XSmad4a or yl-terminal sequence of XSmad4a is QPLD, like mammalian XSmad4b together with HA-tagged XSmad1 or XSmad2 and stimulated with 10 ng/ml TGF-b (T) or 300 ng/ml BMP (B) for 1 h. Complex Smad4, whereas that of XSmad4b is SSVN, which resembles formation was detected by immunoprecipitation (IP) with the anti-Myc the carboxyl-terminal SS(V/M)S phosphorylation motif of path- antibody followed by immunoblotting (IB) with the anti-HA antibody, way-restricted Smads (Fig. 1B). and the aliquots were also blotted with anti-Myc antibody to detect the To examine the temporal expression of XSmad4a and XS- expression of Myc-tagged XSmad4a and Myc-tagged XSmad4b. Ali- mad4b during early development, an RNase protection analy- quots of the cell lysates were directly analyzed by immunoblotting with anti-HA antibody. B, cells were transfected with Myc-XSmad4a or sis was performed with probes corresponding to each of the Myc-XSmad4b combined with HA-XSmad1 or HA-Smad2 and stimu- linker region of XSmad4s. Each probe was confirmed not to lated with BMP or TGF-b at indicated concentrations for 1 h. Oligomer- cross-hybridize to the other (data not shown). XSmad4a mRNA ization was detected by immunoprecipitation followed by immunoblot- was markedly increased by zygotic expression after the blas- ting. C, homomeric or heteromeric oligomer formation was detected by immunoprecipitation followed by immunoblotting from the lysates of tula stage, whereas XSmad4b mRNA was highly abundant in C2C12 cells transfected with HA- or Myc-tagged XSmad4a and eggs and was decreased during the gastrula stage, although XSmad4b. both transcripts were detected throughout early embryogenesis (Fig. 2A). Thus, XSmad4 a mRNA and XSmad4b mRNA are expressed differently during early development. The spatial expression of XSmad4a and XSmad4b was ex- The Second Smad4 12167 FIG.7. Phosphorylation of XSmad4b. C2C12 cells were trans- fected with Myc-tagged XSmad4a, XSmad4b, or the carboxyl-terminal mutated version of XSmad4b (AAVN). Cells were metabolically labeled with [ P]orthophosphate and further incubated with or without TGF-b (10 ng/ml) for 1 h. Phosphorylation of Myc-tagged XSmad4s was ana- lyzed by immunoprecipitation with anti-Myc (a-Myc) antibody followed by autoradiography. WT, wild type; IB, immunoblotting. genes was significantly suppressed (Fig. 3B). These results suggest that XSmad4a and XSmad4b can function as common partners for pathway-restricted Smads to induce expression of specific marker genes in Xenopus animal caps. To elucidate more directly the cooperativity of XSmad4a and XSmad4b with XSmad1 and XSmad2, we tested the ability of the Smads to induce the reporter gene expression under the BMP- and TGF-b-responsive promoters in cultured cells. FIG.6. Subcellular localization of XSmad4a and XSmad4b pro- Transfection of XSmad1 alone into C2C12 mouse myoblast cells teins. C2C12 cells were transfected with Myc-tagged XSmad4a or induced a low level of luciferase expression under the promoter XSmad4b together with or without XSmad2 and stimulated with TGF-b of Xvent-2, a BMP-inducible Xenopus homeobox gene (38, 39). by co-transfection with activated TGF-b type I receptor plus treatment with TGF-b (10 ng/ml) for 1 h. Then the cells were fixed and stained However, co-expression of XSmad4a and XSmad4b, but not with anti-Myc (aMyc) antibody and 49,6-diamidino-2-phenylindole di- XSmad4a DCor XSmad4bDC, together with XSmad1 induced a hydrochloride (DAPI). high level of the reporter gene expression (Fig. 4A). Similarly, both XSmad4a and XSmad4b synergized with XSmad2 to in- amined by whole mount in situ hybridization with the probes duce luciferase expression under the TGF-b-responsive 3TP (Fig. 2, B–G). Both XSmad4a and XSmad4b mRNAs are ubiq- reporter, whereas the carboxyl-terminal truncated constructs uitously expressed in the ectoderm at the early gastrula stage did not (Fig. 4B). These results demonstrate that XSmad4a and (Fig. 2, B and E) but become restricted to the neuroectoderm at XSmad4b act synergistically with XSmad1 and XSmad2 to the neurula stage (Fig. 2, C and F). In the tadpole stage, both induce gene expression in cultured cells. of the XSmad4s are expressed in the central nervous system, XSmad4a and XSmad4b Associate with XSmad1 and XS- eye, and otic vesicle (Fig. 2, D and G). XSmad4b is expressed mad2—As XSmad4a and XSmad4b functionally cooperated more anteriorly than XSmad4a in the neural tube. with XSmad1 and XSmad2, we would expect that XSmad4a XSmad4a and XSmad4b Synergize with XSmad1 and XS- and XSmad4b bind to XSmad1 and XSmad2 to form hetero- mad2—To examine whether XSmad4a and XSmad4b act as meric complexes. To examine this hypothesis, we co-expressed co-Smad to cooperate with Smad1 and Smad2 to mediate BMP- Myc-tagged XSmad4s with HA-tagged XSmad1 or XSmad2 in like and TGF-b/activin-like responses, respectively, we first C2C12 cells and subjected the obtained cell lysates to immu- analyzed the expression of several marker genes in animal caps noprecipitation with anti-Myc antibody followed by immuno- obtained from Xenopus embryos injected with synthetic blotting with anti-HA antibody. The results showed that in mRNAs encoding XSmad4a or XSmad4b combined with Xeno- response to BMP treatment, XSmad1 associates with either of pus Smad1 or Xenopus Smad2. Injection of a low dose of XS- the two XSmad4s (Fig. 5A, upper panel) and that in response to mad1, XSmad2, XSmad4a, or XSmad4b mRNA alone was not TGF-b stimulation, XSmad2 associates with either of the two sufficient to induce any noticeable expression of mesodermal XSmad4s (Fig. 5A, lower panel). Thus, both XSmad4a and genes (Fig. 3A). When XSmad1 mRNA was injected together XSmad4b are able to form heteromeric complexes with XS- with either XSmad4a or XSmad4b mRNA, however, strong mad1 or XSmad2 in response to specific stimulation. expression of a ventral mesodermal marker, a globin, was We next assessed whether the extent of the ligand-induced induced. Similarly, when XSmad2 mRNA was injected with association of XSmad4s with XSmad1 or XSmad2 would differ either of the XSmad4 mRNAs, dorsal mesodermal markers, between XSmad4a and XSmad4b. The association of XSmad2 goosecoid and muscle actin, and a pan-mesodermal marker, with XSmad4a occurred at the same level as with XSmad4b in Xenopus brachyury, were strongly induced (Fig. 3A). a manner dependent on the TGF-b concentration (Fig. 5B, Because the carboxyl-terminal truncated form of human lower panel). In contrast, the association of XSmad1 with XS- Smad4 has been shown to act as a dominant-negative inhibitor mad4b in response to BMP stimulation occurred more strongly for signal transductions (15, 28), we next examined the action than with XSmad4a (Fig. 5B, upper panel). This may suggest of carboxyl-terminal truncated constructs of XSmad4a and XS- that although both of the XSmad4s can function as common mad4b in the animal cap assay. Injection of a high dose of partners of XSmad1 and XSmad2, they do not have completely XSmad1 or XSmad2 mRNA was sufficient to induce expression redundant functions, and XSmad4b may play a major role in of ventral or dorsal mesodermal marker genes, respectively. BMP signaling. When either XSmad4a DC or XSmad4b DC mRNA was injected The crystallographic structure analysis of the MH2 domain along with XSmad1 or XSmad2, the expression of these marker of human Smad4 implied that Smad4 was able to form a ho- 12168 The Second Smad4 motrimer (40). To examine whether XSmad4a and XSmad4b form a heteromeric complex, we co-expressed both XSmad4a and XSmad4b tagged with different epitopes in C2C12 cells. Stimulation-independent hetero-oligomerization as well as ho- mo-oligomerization was observed (Fig. 5C). Therefore, it is possible that XSmad4a and XSmad4b form a heteromeric tri- mer. It may also be possible that after stimulation, XSmad4a, XSmad4b, and one of the pathway-restricted Smads form a heteromeric trimer. XSmad4b, but Not XSmad4a, Is Constitutively Nuclear— When expressed in C2C12 cells, XSmad4a and XSmad4b pro- teins showed different subcellular distribution. XSmad4a was present predominantly in the cytoplasm whereas XSmad4b was predominantly in the nucleus (Fig. 6). Because it has been demonstrated that human Smad4 and a Drosophila co-Smad, Medea, are present in the cytoplasm in the absence of stimu- lation (25, 41), the above result may again indicate that XS- mad4a is a Xenopus ortholog of a previously known Smad4. XSmad4a became localized to the nucleus in more than 60% of the cells after TGF-b stimulation when XSmad2 was co-ex- pressed (Fig. 6, top). Hence, the expression of XSmad2 is re- quired for XSmad4a to change its subcellular localization in response to TGF-b under the conditions. In contrast, subcellu- lar localization of XSmad4b did not change after TGF-b stim- ulation, irrespective of co-expression of XSmad2 (Fig. 6, bottom). XSmad4b, but Not XSmad4a, Is a Phosphoprotein—In addi- tion to the high degree of diversity in the linker region, another striking difference in the amino acid sequence between XS- mad4a and XSmad4b is found in their carboxyl termini. The SSVN sequence in XSmad4b (see Fig. 1B) is similar to the carboxyl-terminal phosphorylation motif (SS(V/M)S) of path- way-restricted Smads in which the last two serine residues undergo phosphorylation upon stimulation (42, 43). We then examined whether XSmad4b could be phosphorylated. Myc epitope-tagged XSmads were transfected into C2C12 cells, and the cells were labeled with [ P]orthophosphate and stimulated with TGF-b. An increase in XSmad2 phosphorylation after TGF-b treatment was observed, as determined by immunopre- cipitation with anti-Myc antibody followed by autoradiography (Fig. 7). No or little phosphorylation of XSmad4a was observed before or after TGF-b stimulation (Fig. 7). On the contrary, phosphorylation of XSmad4b was observed even before stimu- lation, and its level did not increase after TGF-b treatment (Fig. 7). To explore the possible involvement of the SSVN sequence of XSmad4b in its phosphorylation, we constructed a mutant XSmad4b having an AAVN sequence instead of SSVN at its carboxyl terminus. This mutant was still phosphorylated irrespective of TGF-b stimulation, and no decrease in the phos- phorylation level was observed by this mutation (Fig. 7). There- fore, the carboxyl-terminal SSVN sequence of XSmad4b is not phosphorylated, and XSmad4b may not be a direct substrate of completely and those with a DAI of 5 are normal. C, immunoblotting analysis of the exogenously expressed XSmad4a and XSmad4b.N- terminal Myc-tagged XSmad4a and XSmad4b mRNAs were injected at the 4-cell stage embryos at the indicated doses, and extracts were obtained at the blastula stage (stage 9). Expressed proteins were de- FIG.8. XSmad4b induces ventralization of embryos. A, tadpole tected by immunoblotting with anti-Myc antibody. Myc-tagged stage (stage 35) Xenopus embryos and sibling embryos that have been XSmad4s had essentially the same effect as nontagged constructs on injected dorsally with XSmad4a mRNA or XSmad4b mRNA at the 4-cell the phenotypes of embryos (data not shown). D, expression of marker stage at the indicated doses are shown (Dorsal). Embryos injected genes in dorsal marginal zone explants. Dorsal marginal zone explants ventrally with XSmad4a or XSmad4b mRNA (2 ng) are also shown were dissected at the early gastrula stage from embryos that had been (Ventral). B, semiquantification of ventralization of the embryos by injected dorsally with XSmad4a, XSmad4b, XSmad1, or XSmad2 XSmad4b. At the 4-cell stage, two dorsal blastomeres were injected mRNA (each at 2 ng) at the 4-cell stage and were cultured until sibling with XSmad4a mRNA or XSmad4b mRNA at the indicated doses. The embryos reached the midgastrula stage. Expression of indicated DAI of the embryos was scored after 2 days, and the average DAI for marker genes was analyzed by RT-PCR. Expression of EF-1a was also each sample is shown. Numbers of embryos examined are indicated analyzed as a loading control. No signal was observed in the absence of above the figure. Embryos with a DAI of 0 lack dorsal structures reverse transcription (2RT). gsc, goosecoid; Xbra, Xenopus brachyury. The Second Smad4 12169 the TGF-b receptor. family (45). Moreover, lack of Smad4 in colon cancer cells leads XSmad4b Induces Ventralization of Xenopus Embryos—To to an increase in metastasis and malignancy, which suggests find the functional difference between the two XSmad4s, we that Smad4 has a tumor-suppressing function (46). These ob- tested the effect of expression of XSmad4 in Xenopus embryos. servations suggest that if mammals have the second co-Smad, Injection of XSmad4b mRNA, but not XSmad4a mRNA, into its function may be distinct from that of Smad4/DPC4. two dorsal cells of the 4-cell stage embryos led to strong ven- The Xenopus system has been extensively used to elucidate tralization, as revealed by the defects in anterior structures at the cellular signaling mechanism of growth factors, including the late stage (Fig. 8A). Similar phenotypes were reported to be the TGF-b superfamily, that control cell differentiation and induced by expressing Smad1 or Smad5 (16, 22). Injection of pattern formation during early embryogenesis (47–52). It has either of the mRNAs into the ventral side had little or no effect been suggested that activin, nodal, and Vg1, members of the on the embryonic development (Fig. 8A). The ventralizing effect TGF-b superfamily, are involved in the differentiation of dorsal of XSmad4b, when expressed in the dorsal marginal region, mesodermal tissues, whereas the BMP family ligands regulate was dose-dependent. This was clearly demonstrated by semi- the ventral mesoderm differentiation in Xenopus early devel- quantification by scoring the dorsoanterior index (DAI) of the opment. Xenopus homologs of Smad1 and Smad2 have been embryos (44), where 5 represents a normal embryo and 0 indi- identified and shown to mediate activin and BMP signaling cates an embryo lacking axial structures (Fig. 8B). We con- pathways, respectively (10, 12, 16, 53). Interestingly, Candia et firmed that XSmad4a and XSmad4b were expressed at almost al. (39) observed that the expression of an excess amount of the same level in this series of experiments (Fig. 8C). In addi- human Smad4 in Xenopus embryos compromised the antago- tion to the defects in anterior structures, injection of XSmad4b nism between the activin/Vg1 and BMP pathways. They pro- mRNA into dorsal marginal cells induced an increase in the posed an attractive model, which supports the theory that the expression of a ventral mesodermal marker, Xvent-2, and a activin/Vg1 and BMP pathways modulate each other’s activity decrease in the expression of a dorsal mesodermal marker, by sequestering a limited pool of Smad4, which commonly goosecoid, more strongly than did XSmad4a mRNA, as ana- participates in both pathways by associating with Smad2 and lyzed in isolated dorsal marginal zone explants (Fig. 8D). These Smad1, respectively. As human Smad4 corresponds to XS- results indicate that the two XSmad4 proteins have distinct mad4a, XSmad4a may be commonly used as a co-Smad for both abilities to induce ventralization in Xenopus embryos. pathways, or a putative heteromeric complex between XS- mad4a and XSmad4b may be a common partner for both XS- DISCUSSION mad1 and XSmad2. These considerations may be consistent Smad4/DPC4 has been shown to be required as a common with our idea that XSmad4a and XSmad4b have overlapping partner for pathway-restricted Smads to propagate the TGF-b but somewhat distinct functions as co-Smads. Thus the obser- family signals from the cell surface receptors to the nucleus by vations of Candia et al. are not inconsistent with our idea that forming heteromeric complexes. Here we have identified a XSmad4b may be rather preferentially used for the BMP path- novel Smad4-like molecule, XSmad4b, as well as a Xenopus way, which was derived from our results showing that XS- homolog of mammalian Smad4 (XSmad4a), and shown that mad4b binds to XSmad1 more tightly than does XSmad4a and both of the XSmad4s are able to function as co-Smad to coop- that XSmad4b has the more potent ventralizing activity. erate and synergize with XSmad1 and XSmad2. However, they Although we still do not know the molecular mechanism that have distinct properties in terms of their temporal expression defines the distinct ability between XSmad4a and XSmad4b in during early embryogenesis, subcellular distribution, phospho- cooperation with XSmad1 and XSmad2, it is likely that some rylation state, and ventralizing activity in Xenopus embryos. factors interact specifically with each of the XSmad4s. It is also Therefore, it is likely that two co-Smads, XSmad4a and XS- possible that phosphorylation of XSmad4b might have some mad4b, play both overlapping and distinct roles in transducing role in regulating the function of XSmad4b. We are currently a set of TGF-b superfamily signals in Xenopus. investigating the mechanisms that underlie the difference be- We reasoned from their subcellular distribution as well as tween XSmad4a and XSmad4b in their cooperativity with their amino acid sequence similarity to human or mouse pathway-restricted Smads, as well as the molecular basis for Smad4 that XSmad4a was a Xenopus ortholog of mammalian determining their subcellular localization. Smad4 and XSmad4b was a novel homolog. XSmad4a localized Acknowledgments—We thank Dr. Douglas A. Melton for providing predominantly in the cytoplasm and translocated to the nu- Xenopus Smad1 and Smad2 plasmids and Dr. Ken W. Y. Cho for the cleus after stimulation when a pathway-restricted Smad was Xvent2-Luc plasmid. We also thank K. Kawachi and M. Watanabe for co-expressed (Fig. 6). This behavior is the same as that previ- technical assistance. ously reported for human Smad4 (25). In contrast, XSmad4b REFERENCES was present predominantly in the nucleus before or after stim- ulation. The subcellular distribution of either of the XSmad4s 1. Derynck, R., and Feng, X. H. (1997) Biochim. Biophys. Acta 1333, F105–F150 2. Heldin, C. H., Miyazono, K., and ten Dijke, P. (1997) Nature 390, 465– 471 was not affected when the other XSmad4 was co-expressed in 3. Attisano, L., and Wrana, J. L. (1998) Curr. Opin. Cell Biol. 10, 188 –194 cells (data not shown), although they could bind to each other 4. Massague, J. (1998) Annu. Rev. Biochem. 67, 753–791 5. Whitman, M. (1998) Genes Dev. 12, 2445–2462 to form the XSmad4a-XSmad4b complex in vitro (Fig. 5C). 6. Hayashi, H., Abdollah, S., Qiu, Y., Cai, J., Xu, Y., Grinnell, B. W., Richardson, Although we do not know the functional significance of their M. A., Topper, J. N., Gimbrone, M. A., Jr., Wrana, J. L., and Falb, D. (1997) different subcellular localizations, the possible formation of the Cell 89, 1165–1173 7. Imamura, T., Takase, M., Nishihara, A., Oeda, E., Hanai, J., Kawabata, M., heteromeric complex between XSmad4a and XSmad4b after and Miyazono, K. (1997) Nature 389, 622– 626 the nuclear entry of XSmad4a in response to stimulation might 8. Nakao, A., Afrakhte, M., Moren, A., Nakayama, T., Christian, J. L., Heuchel, R., Itoh, S., Kawabata, M., Heldin, N. E., Heldin, C. H., and ten Dijke, P. have some physiological significance. (1997) Nature 389, 631– 635 An intriguing question is whether or not the existence of the 9. Hoodless, P. A., Haerry, T., Abdollah, S., Stapleton, M., O’Connor, M. B., second co-Smad is specific to Xenopus. At present, there is no Attisano, L., and Wrana, J. L. (1996) Cell 85, 489 –500 10. Graff, J. M., Bansal, A., and Melton, D. A. (1996) Cell 85, 479 – 487 evidence for the existence of another Smad4-like gene in mam- 11. Eppert, K., Scherer, S. W., Ozcelik, H., Pirone, R., Hoodless, P., Kim, H., Tsui, mals. We could not find any related genes in the data base of L. C., Bapat, B., Gallinger, S., Andrulis, I. L., Thomsen, G. H., Wrana, J. L., and Attisano, L. (1996) Cell 86, 543–552 expressed sequence tags (dbEST). Smad4 knockout mice were 12. Baker, J. C., and Harland, R. M. (1996) Genes Dev. 10, 1880 –1889 shown to die early in their development (29, 30), and the 13. Liu, F., Hata, A., Baker, J. C., Doody, J., Carcamo, J., Harland, R. M., and Smad4-deficient cells lost responsiveness to the TGF-b super- Massague, J. (1996) Nature 381, 620 – 623 12170 The Second Smad4 14. Macias-Silva, M., Abdollah, S., Hoodless, P. A., Pirone, R., Attisano, L., and 34. Shibuya, H., Iwata, H., Masuyama, N., Gotoh, Y., Yamaguchi, K., Irie, K., Wrana, J. L. (1996) Cell 87, 1215–1224 Matsumoto, K., Nishida, E., and Ueno, N. (1998) EMBO J. 17, 1019 –1028 15. Zhang, Y., Feng, X., We, R., and Derynck, R. (1996) Nature 383, 168 –172 35. Harland, R. M. (1991) Methods Enzymol. 36, 675– 685 16. Thomsen, G. H. (1996) Development 122, 2359 –2366 36. Iemura, S., Yamamoto, T. S., Takagi, C., Uchiyama, H., Natsume, T., Shi- 17. Yingling, J. M., Das, P., Savage, C., Zhang, C., Padgett, R. W., and Wang, X.-F. masaki, S., Sugino, H., and Ueno, N. (1998) Proc. Natl. Acad. Sci. U. S. A. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 8940 – 8944 95, 9337–9342 18. Lechleider, R. J., de Caestecker, M. P., Dehejia, A., Polymeropoulos, M. H., and 37. Moriguchi, T., Kuroyanagi, N., Yamaguchi, K., Gotoh, Y., Irie, K., Kano, T., Roberts, A. B. (1996) J. Biol. Chem. 271, 17617–17620 Shirakabe, K., Muro, Y., Shibuya, H., Matsumoto, K., Nishida, E., and 19. Kretzschmar, M., Liu, F., Hata, A., Doody, J., and Massague, J. (1997) Genes Hagiwara, M. (1996) J. Biol. Chem. 271, 13675–13679 Dev. 11, 984 –995 38. Onichtchouk, D., Gawantka, V., Dosch, R., Delius, H., Hirschfeld, K., Blumen- 20. Nakao, A., Imamura, T., Souchelnytskyi, S., Kawabata, M., Ishisaki, A., Oeda, stock, C., and Niehrs, C. (1996) Development 122, 3045–3053 E., Tamaki, K., Hanai, J., Heldin, C. H., Miyazono, K., and ten Dijke, P. 39. Candia, A. F., Watabe, T., Hawley, S. H., Onichtchouk, D., Zhang, Y., Derynck, (1997) EMBO J. 16, 5353–5362 R., Niehrs, C., and Cho, K. W. (1997) Development 124, 4467– 4480 21. Chen, Y., Bhushan, A., and Vale, W. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 40. Shi, Y., Hata, A., Lo, R. S., Massague, J., and Pavletich, N. P. (1997) Nature 12938 –12943 388, 87–93 22. Suzuki, A., Chang, C., Yingling, J. M., Wang, X. F., and Hemmati-Brivanlou, 41. Wisotzkey, R. G., Mehra, A., Sutherland, D. J., Dobens, L. L., Liu, X., Dohr- A. (1997) Dev. Biol. 184, 402– 405 mann, C., Attisano, L., and Raftery, L. A. (1998) Development 125, 23. Lagna, G., Hata, A., Hemmati-Brivanlou, A., and Massague, J. (1996) Nature 1433–1445 383, 832– 836 42. Abdollah, S., Macias-Silva, M., Tsukazaki, T., Hayashi, H., Attisano, L., and 24. de Caestecker, M. P., Hemmati, P., Larisch-Bloch, S., Ajmera, R., Roberts, Wrana, J. L. (1997) J. Biol. Chem. 272, 27678 –27685 A. B., and Lechleider, R. J. (1997) J. Biol. Chem. 272, 13690 –13696 43. Souchelnytskyi, S., Tamaki, K., Engstrom, U., Wernstedt, C., ten Dijke, P., and 25. Liu, F., Pouponnot, C., and Massague, J. (1997) Genes Dev. 11, 3157–3167 Heldin, C. H. (1997) J. Biol. Chem. 272, 28107–28115 26. Yingling, J. M., Datto, M. B., Wong, C., Frederick, J. P., Liberati, N. T., and 44. Kao, K. R., and Elinson, R. P. (1988) Dev. Biol. 127, 64 –77 Wang, X. F. (1997) Mol. Cell. Biol. 17, 7019 –7028 45. Zhou, S., Buckhaults, P., Zawel, L., Bunz, F., Riggins, G., Le Dai, J., Kern, 27. Hahn, S. A., Schutte, M., Hoque, A. T., Moskaluk, C. A., da Costa, L. T., S. E., Kinzler, K. W., and Vogelstein, B. (1998) Proc. Natl. Acad. Sci. Rozenblum, E., Weinstein, C. L., Fischer, A., Yeo, C. J., Hruban, R. H., and U. S. A. 95, 2412–2416 Kern, S. E. (1996) Science 271, 350 –353 46. Takaku, K., Oshima, M., Miyoshi, H., Matsui, M., Seldin, M. F., and Taketo, 28. Zhang, Y., Musci, T., and Derynck, R. (1997) Curr. Biol. 7, 270 –276 M. M. (1998) Cell 92, 645– 656 29. Sirard, C., de la Pompa, J. L., Elia, A., Itie, A., Mirtsos, C., Cheung, A., Hahn, 47. Sive, H. L. (1993) Genes Dev. 7, 1–12 S., Wakeham, A., Schwartz, L., Kern, S. E., Rossant, J., and Mak, T. W. 48. Dawid, I. B. (1994) J. Biol. Chem. 269, 6259 – 6262 (1998) Genes Dev. 12, 107–119 49. Kessler, D. S., and Melton, D. A. (1994) Science 266, 596 – 604 30. Yang, X., Li, C., Xu, X., and Deng, C. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 50. Smith, J. C. (1995) Curr. Opin. Cell Biol. 7, 856 – 861 3667–3672 51. Heasman, J. (1997) Development 124, 4179 – 4191 31. Krieg, P. A., and Melton, D. A. (1984) Nucleic Acids Res. 12, 7057–7070 52. Sasai, Y., and De Robertis, E. M. (1997) Dev. Biol. 182, 5–20 32. Nieuwkoop, P. D., and Faber, J. (1967) Normal Table of Xenopus laevis (Dau- din), North Holland Publishing Co., Amsterdam 53. Meersseman, G., Verschueren, K., Nelles, L., Blumenstock, C., Kraft, H., Wuytens, G., Remacle, J., Kozak, C. A., Tylzanowski, P., Niehrs, C., and 33. Gotoh, Y., Masuyama, N., Suzuki, A., Ueno, N., and Nishida, E. (1995) EMBO J. 14, 2491–2498 Huylebroeck, D. (1997) Mech. Dev. 61, 127–140

Journal

Journal of Biological Chemistry – Unpaywall

Published: Apr 1, 1999

There are no references for this article.