- Email: [email protected]
β-catenin signaling controls Mespo expression to regulate segmentation during Xenopus somitogenesis
Developmental Biology 304 (2007) 836 – 847 www.elsevier.com/locate/ydbio
Genomes & Developmental Control
Wnt/β-catenin signaling controls Mespo expression to regulate segmentation during Xenopus somitogenesis Jinhu Wang 1 , Shangwei Li 1 , Yuelei Chen, Xiaoyan Ding ⁎ Laboratory of Molecular Cell Biology, Key Laboratory of Stem Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue Yang Road, Shanghai 200031, China Graduate School of Chinese Academy of Sciences, 320 Yue Yang Road, Shanghai 200031, China Received for publication 12 May 2006; revised 17 November 2006; accepted 14 December 2006 Available online 21 December 2006
Abstract The vertebral column is derived from somites, which are transient segments of the paraxial mesoderm that are present in developing vertebrates. The strict spatial and temporal regulation of somitogenesis is of crucial developmental importance. Signals such as Wnt and FGF play roles in somitogenesis, but details regarding how Wnt signaling functions in this process remain unclear. In this study, we report that Wnt/β-catenin signaling regulates the expression of Mespo, a basic–helix–loop–helix (bHLH) gene critical for segmental patterning in Xenopus somitogenesis. Transgenic analysis of the Mespo promoter identifies Mespo as a direct downstream target of Wnt/β-catenin signaling pathway. We also demonstrate that activity of Wnt/β-catenin signaling in somitogenesis can be enhanced by the PI3-K/AKT pathway. Our results illustrate that Wnt/ β-catenin signaling in conjunction with PI3-K/AKT pathway plays a key role in controlling development of the paraxial mesoderm. © 2007 Published by Elsevier Inc. Keywords: AKT; Paraxial mesoderm; PI3-K; Wnt; Xenopus laevis
Introduction During vertebrate embryogenesis, the paraxial mesoderm separates into somites, a metameric series of homologous subunits, which bud off sequentially from the rostral end of the presomitic mesoderm (PSM) at regular intervals along the anterior–posterior (AP) axis. The process in which the segmental or metameric pattern is established accompanying the constant posterior elongation of the embryo body axis is termed somitogenesis (Weisblat et al., 1994). Somitogenesis occurs in a strict spatial and temporal progression (Dubrulle et al., 2001; Pourquie, 2003). The molecular nature of the signals responsible for controlling somitogenesis remains elusive. One potential candidate for regulating this process is Wnt/β-catenin signaling (Aulehla et al., 2003; Yamaguchi et al., 1999). Previous studies have shown that Wnt/β-catenin signaling is implicated in ⁎ Corresponding author. Fax: +86 21 34230165. E-mail address: [email protected] (X. Ding). 1 These authors contributed equally to this work. 0012-1606/$ - see front matter © 2007 Published by Elsevier Inc. doi:10.1016/j.ydbio.2006.12.034
controlling the development of paraxial mesoderm (Aulehla et al., 2003; Galceran et al., 2004; Hofmann et al., 2004; Pourquie, 2001). In mouse embryos, Wnt3a is expressed in the primitive streak during gastrulation and in the tailbud during the later stages of mouse development. Animals homozygous for null alleles of Wnt3a lack somites posterior to the forelimbs (Ikeya and Takada, 2001; Takada et al., 1994; Wilson et al., 1993). The double mutant harboring mutations in lef1 and tcf1, identified as nuclear mediators of Wnt/β-catenin signaling that activate Wnt-responsive genes by associating with β-catenin, results in lack of paraxial mesoderm resembling the Wnt3a null mutation (Galceran et al., 1999). In chick embryos, it has been shown that altering Wnt/β-catenin signaling activity in the PSM causes defects in somite formation (Aulehla et al., 2003). However, the exact molecular nature of Wnt/β-catenin signaling in somitogenesis is not fully understood, and whether Wnt/ β-catenin signaling is involved in somitogenesis of the lower vertebrates such as Xenopus is still questionable. The FGF signaling pathway also participates in somitogenesis (Delfini et al., 2005; Dubrulle et al., 2001; Moreno and Kintner, 2004). Fgf8 mRNA forms a gradient along the PSM in mouse
J. Wang et al. / Developmental Biology 304 (2007) 836–847
embryos (Delfini et al., 2005; Dubrulle et al., 2001). Altering the activity of FGF signaling in the PSM resulted in segmental defects (Dubrulle et al., 2001). These data demonstrate that FGF signaling is required for somitogenesis. It has been proposed that FGF signaling may act through the PI3-K/AKT pathway in the PSM to participate in somitogenesis (Dubrulle and Pourquie, 2004). Since PI3-K/AKT signaling can enhance accumulation of β-catenin by inhibiting GSK-3β during cardiomyocyte growth (Haq et al., 2003), it may regulate Wnt/β-catenin activity in the segmentation process by a similar mechanism. Somite formation and patterning are thought to be controlled by Mespo, a member of the basic–helix–loop–helix (bHLH) transcription factor family (Moreno and Kintner, 2004; Yoon et al., 2000; Yoon and Wold, 2000). Mespo expression encompasses the entire caudal PSM of Xenopus embryos, and its expression is reduced when the PSM cells enter the segmentation process (Joseph and Cassetta, 1999; Kim et al., 2000; Moreno and Kintner, 2004). Gain of function experiments showed that Mespo can induce expression of paraxial mesoderm marker genes (Yoon et al., 2000). Mespo null mouse embryos exhibited no identifiable somites or segmental patterning in the trunk posterior to forelimbs, and severe disruption of gene expressions in posterior paraxial mesoderm (Yoon and Wold, 2000). These results suggest that Mespo plays essential roles in somitogenesis. The phenotypic resemblance between Mespo and Wnt3a mutant mouse embryos provides a strong hint that Wnt/β-catenin signaling regulates the expression of Mespo during paraxial mesoderm development. However, whether Mespo expression is regulated by Wnt/β-catenin signaling in the PSM remains unclear. Here, we report that Wnt/β-catenin signaling directly regulates the expression of Mespo, and this regulation is enhanced by PI3-K/AKT signaling. These results demonstrate that Wnt/β-catenin signaling plays a key role in controlling somitogenesis by regulating Mespo gene expression, and that PI3-K/AKT signaling is also involved in this process. Materials and methods Embryo manipulations, drug treatment Preparation and injection of Xenopus embryos were carried out as previously described (Ding et al., 1998). Embryos were staged according to Nieuwkoop and Faber (1967). LY294002 (Calbiochem) was used at 100 μM. Chemicals were dissolved in DMSO and then diluted into 0.1× MMR. In each case, about 20 embryos were examined per condition per probe. Each experiment was repeated three times independently.
Morpholino oligomer and in vitro translation The 25-bp morpholino antisense oligomer for Xenopus Mespo was obtained from Gene Tools and consisted of the following sequence: MespoMO: GGGATGGTGCAGAGTCTCCATCAGT Morpholino was dissolved in water to a concentration of 1 mM, which was then further diluted to a give a working solution of 0.3 mM. For the rescue assay, we generated a mutated Mespo cDNA (mMespo) differed in seven bases (underlined) from Mespo: 5′-CTACTATGGAGACTCTGCACCATCCCCT.
837
The in vitro transcription/translation of Mespo was performed according to the protocol of TNT coupled reticulocyte lysate system (Promega, L4610).
In situ hybridization, immunocytochemistry Whole-mount in situ hybridization was performed as previously described (Harland, 1991) using DIG-labeled probes. For bleaching of wild-type embryos, hybridized embryos were treated with bleaching solution (0.5× SSC with 1% hydrogen peroxide and 5% formamide) under a fluorescent light. The following plasmid templates were linearized, and DIG-labeled antisense RNA probes were made using either T7 or SP6 RNA polymerase: Thylacine1 (Sparrow et al., 1998); Mespo (Joseph and Cassetta, 1999); GFP (Geng et al., 2003); Xwnt8 (Kazanskaya et al., 2004) and Dkk1 (Glinka et al., 1998); Paraxis (a PCRamplified full-length cDNA of Paraxis was cloned into pCS2+); AKT (a PCRamplified coding region of Xenopus AKT was cloned into pCS2+); PAPC (Kim et al., 2000); XBra (Smith et al., 1991). To visualize somite size, embryos were fixed in 2% paraformaldehyde, 2% glutaraldehyde in PBS, and stained with the muscle-specific monoclonal antibody 12/101 (Kintner and Brockes, 1985). Then embryos were post-fixed in MEMFA (0.1 M MOPS pH 7.4, 2 mM EGTA, 1 mM MgSO4, 3.7% formaldehyde), embedded in paraffin, sectioned at 10 μm, and stained with DAPI.
Mutagenesis and transgenesis The Exsite™ PCR-based site-directed mutagenesis system (Stratagene, cat. 200502-5) was used to generate the mutants of site L, p-4317mLuc and p-511m. The primers are shown below (nucleotides altered shown in bold): 5′AGGAGACAGCAAGGTGTTAACATG-3′ (reverse) and 5′-CTAGACTCCTCCATTAAAACGCCCACT-3′ (forward). Transgenic Xenopus embryos were generated as described previously (Kroll and Amaya, 1996). Plasmids used for transgenesis assays were linearized by NotI digestion. Three independent rounds of transgenesis were performed for each plasmid.
Western blot For measuring phospho-Akt levels, the caudal region of Xenopus embryos at stage 23 were divided into three equal parts corresponding to the caudal domain, the rostral domain of the PSM and a region including the last formed somites. In these experiments, 100 embryos were used. The samples were lysed in EDTAfree RIPA (0.1% NP40, 20 mM Tris–HCl pH 8, 10% glycerol) with cocktail protease inhibitors (Bio Basics) and Na3VO4. For measuring accumulation of β-catenin in the nucleus and cytosol, the caudal domain of the PSM was dissected and cultured in 1× MMR containing LY294002 at 100 μM or DMSO for 2 h. Then β-catenin in either nucleus or cytosol was extracted according to PIERCE protocol (NE-PER™ nuclear and cytoplasmic extraction reagents, 78833). All the samples were quantitated by Bradford assay and adjusted to equal concentration before loading. Western blot was performed as described (He et al., 2005). Polyclonal antibody P14L was used to visualize β-catenin (dilution 1:2000), and anti-phospho-Akt polyclonal antibodies (dilution 1:2000; Cell Signaling Technology) were used to detect endogenous Akt and phospho-Akt. For a loading control, equal volume of proteins were separated by SDS–PAGE and visualized with Coomassie Blue.
Chromatin immunoprecipitation (ChIP) Chromatin immunoprecipitation (ChIP) assay was performed, with modifications, according to the method described previously (Sachs and Shi, 2000). 100 Xenopus embryos were selected at stage 13 from embryos injected with appropriate plasmids, and crosslinked with 1% formaldehyde at room temperature for 20 min. Following treatment with 0.125 M glycine, embryos were washed and resuspended in 200 μl ChIP lysis buffer. The lysates were sonicated on ice to obtain DNA fragments on an average around 500 bp and immunoprecipitated with anti-HA monoclonal antibody (dilution 1:3000; Sigma). DNA fragments were washed, eluted and recovered from the immunoprecipitated complex. PCR was performed under an optimized condition: 94 °C
838
J. Wang et al. / Developmental Biology 304 (2007) 836–847
for 30 s, 58 °C for 30 s, and 72 °C for 30 s, for 30–35 cycles. Primer sequences are as follows: Mespo promoter upstream 5′-TGAAAGCCAATTTAGCTGAAGG3′, downstream 5′-GCTGGCAAAGTTCCATGTGAG-3′; ODC upstream 5′GATCATGCACATGTCAAGCC-3′, downstream 5′-CAGGGAGAATGCCATGTTCT-3′; pGL3-Basic upstream 5′-CAGGGGATAACGCAGGAAA-3′, downstream 5′-AAGGGAGAAAGGCGGACA G-3′.
Results MO-based Mespo depletion caused defects of somite formation Previous analysis of the morphological defects in Mespo knockout mouse suggests that Mespo plays a role in somitogenesis (Yoon and Wold, 2000). In Xenopus, Mespo expression is firstly detected in ventrolateral mesoderm but is absent from dorsal mesoderm in gastrulae, and is localized to the most caudal (unsegmented) paraxial mesoderm in neurulae
and tailbud stage (Joseph and Cassetta, 1999; Yoon et al., 2000). To test whether Mespo contributes to Xenopus somitogenesis, we used antisense morpholino oligonucleotides (MO) to deplete Mespo. We designed a Mespo MO that spans the ATG initiation codon (Fig. 1A). This MO inhibits translation of recombinant Mespo RNA in an in vitro transcription/translation reaction (Fig. 1B). To disrupt endogenous Mespo expression we injected Mespo MO into the ventral marginal zone (VMZ) of one blastomere at four-cell stage, with the uninjected side serving as a control. The somite morphology of the injected embryos was then examined at tadpole stage by whole-mount immunostaining with 12/101 (a monoclonal antibody specifically recognizes differentiated muscle cells). While control MO-injected embryos exhibit no morphological defects in somite formation (Fig. 1C), embryos injected with Mespo MO show remarkable malformation of
Fig. 1. Antisense MO against Mespo depletes Mespo protein and causes somite formation defects. (A) Sequence of Mespo MO aligned with the 5′-untranslated region (UTR) of Mespo RNA. (B) In vitro transcription/translation of Mespo is inhibited by Mespo MO. Arrow indicates the band corresponding to Mespo protein. (C, D) Whole mount immunostaining with 12/101, a muscle-specific antibody, shows that the segmental pattern of somites in Mespo MO-injected embryos (D) is disrupted and the somite size is irregular (compared with control MO-injected embryo in panel C). Bracket in panel D marks the malformed somites. (E–H) Embryos injected with Mespo MO (F, H) or control MO (E, G) stained with 12/101, then sectioned longitudinally and labeled with DAPI. Panels E, G and Panels F, H are the same view of one embryo, respectively. Embryos injected with Mespo MO panel F, H show the disorganization of segmental pattern, based on myotomal morphology (panel F, compared with panel E) and the arrangement of somatic nuclei panel H, compared with G). Bracket in panel H marks the malformed somites.
J. Wang et al. / Developmental Biology 304 (2007) 836–847
somites. The defects include disrupted somite boundaries and irregular somite size (79%, N = 100, Fig. 1D). The malformation of somites in Mespo MO-injected embryos is shown more clearly on frontal sections stained with 12/101 or with DAPI staining of nuclei. Compared with the uninjected side, somites in Mespo MO-injected side are poorly defined, with no metameric shape typical for this tissue. Also, somite nuclei are abnormally aligned, indicating that the patterning of somite is severely disrupted (67%, N = 30, Figs. 1F, H). As a control, the somite morphology in control MO-injected embryos remains identical to that in the uninjected side, with regularly arranged allied cells (Figs. 1E, G). Aiming to verify that the phenotype described above is due to specific depletion of Mespo, we carried out rescue experiments. We prepared a mutated Mespo (mMespo) cDNA, which codes for the same polypeptide as the Mespo protein but differs from Mespo cDNA by seven bases. Injection of the resulting pCS– mMespo plasmid effectively rescues the phenotypic defects caused by Mespo MO in 85% (N = 100) of the injected embryos (Supplemental Fig. 1). We also injected Mespo MO into DMZ of four-cell-stage embryos, where Mespo is not expressed, and found no effect on embryo development (data not shown). These results indicate that the effect of Mespo MO is specific. Taken together, our results indicate that depletion of Mespo caused the morphological defects of somite patterning. Depleting Mespo disrupts expression of genes in the PSM The results described above indicate that lacking of Mespo in embryos causes segmentation defects. In Xenopus, several molecules have been demonstrated to play roles in segmentation process, including PAPC (Kim et al., 2000), Thylacine1 (Thy) (Sparrow et al., 1998) and XDelta-1 (Pourquie, 2003). To evaluate whether depleting Mespo interferes the expression of segmental marker genes, we examined the expression of genes such as PAPC, Thy and XDelta-1 in embryos injected with Mespo MO. Compared with the uninjected side, the expression of PAPC is lost in Mespo MO-injected side (85%, N = 78, Fig. 2B). The loss of PAPC expression in embryos injected with Mespo MO is specific since it could be rescued by coinjecting with pCS–mMespo (70%, N = 60, Fig. 2C). The segmental expression of Thy in somitomeres is severely disrupted (87%, N = 65, Fig. 2E), and it was observed that the somitomeric expression pattern of XDelta-1 is also perturbed (80%, N = 81, Fig. 2G). As a control, injection of a mixture of pCS–LacZ and control MO resulted in no difference in the expression level and pattern of PAPC, Thy or XDelta-1 in both injected and uninjected sides (Figs. 2A, D, F). These results indicate that depletion of Mespo profoundly changes the expression of PAPC, Thy and XDelta-1 in the PSM, suggesting that Mespo acts in the PSM to control segmental patterning. Since XDelta-1 is a component of the Notch/Delta pathway, these results suggest that Mespo controls segmental patterning by regulating Notch/Delta signaling (Pourquie, 2003). To gain further insight into the cause of somite defects in embryos lacking Mespo function, we examined the expression of XBra that plays major roles in development of paraxial
839
mesoderm (Conlon and Smith, 1999). We found that in embryos depleted for Mespo, the expression of XBra is increased in the Mespo MO-injected side (67%, N = 90, Fig. 2I), as compared to the control side, indicating that XBra expression is negatively regulated by Mespo, a phenomenon also observed in Mespo null mutant mouse (Yoon and Wold, 2000). XBra expression is not changed in control MO-injected embryos (Fig. 2H). Interestingly, in embryos injected with Mespo MO, Mespo mRNA is expressed at levels higher than the control side (78%, N = 78, Fig. 2K). Expression of Mespo is not changed in the control MO-injected embryos (Fig. 2J). This result implies that Mespo has a negative regulatory effect on its own mRNA expression in the PSM. Taken together, we provide evidence here that Mespo is responsible not only for segmental specific gene expression, but is also involved in regulating Xbra and Mespo expression. Thus we conclude that Mespo is essential for somitogenesis. Wnt/β-catenin signaling in the PSM Observations from studies on mouse embryos have demonstrated that Wnt/β-catenin signaling is involved in somitogenesis (Aulehla et al., 2003). Regional similarity between expression patterns of Wnts and Mespo in mouse embryos, as well as the phenotypic similarities of Wnt and Mespo null mutant mouse (Takada et al., 1994; Yoon and Wold, 2000), raises a possibility that Wnt signaling regulates Mespo expression in somitogenesis. However, the existence of Wnt signaling in Xenopus PSM has not been documented. We performed whole-mount in situ hybridization to analyze Wnt gene expression in the PSM. Like in other vertebrates, Xenopus PSM is subdivided into two distinct regions featured by expression pattern of the bHLH family genes (Fig. 3K) (Kim et al., 2000; Pourquie, 2003). In the caudal domain of the PSM, Mespo is expressed in the entire region referred as the tailbud domain (TBD) (Figs. 3I, J). Thy transcripts are restricted in the rostral region of the PSM and localized to the anterior half of somitomeres (Figs. 3G, H, I, J). At the neurula stage, XWnt8 is expressed at high level in the TBD (Figs. 3A, B, E, F, G, H), whereas Wnt antagonist Dkk1 is uniformly expressed in somitomeres (Figs. 3C, D, E, F). The expression patterns of these genes indicate that Wnt signals exist in the TBD and support that Wnt/β-catenin signaling is involved in Xenopus somitogenesis. Altering Wnt/β-catenin signaling activity affects expression of Mespo To test whether Wnt/β-catenin signaling regulates Mespo expression, pCS–Lef1–VP16 (a dominant active form of Lef1) DNA was injected into VMZ of one cell of four-cell-stage embryos and Mespo expression was examined at stage 19. Mespo expression in the injected side is increased compared with the control side (75%, N = 70, Fig. 4B). When embryos were injected with pCS–β-catenin DNA, Mespo expression is also increased (Supplemental Fig. 2B). In contrast, injection of pCS–Lef1–EnR (a dominant negative form of Lef1) DNA
840
J. Wang et al. / Developmental Biology 304 (2007) 836–847
Fig. 2. Expression patterns of genes responsible for segmental patterning are altered in embryos lacking Mespo function. For all embryos, anterior is to the left. (A–K) One VMZ cell of four-cell-stage embryos injected with control MO (A, D, F, H and J), or Mespo MO (B, E, G, I and K) or Mespo MO plus 35 pg pCS–mMespo (C) and the embryos were fixed at neurula stage for in situ hybridization to detect the PSM marker gene expression: (A–C) PAPC, (D, E) Thy, (F, G) XDelta-1, (H, I) XBra, (J, K) Mespo. Bracket in panels B and E indicates the disruption of segmental expression of PAPC and Thy, respectively. Bracket in panel C indicates the rescue of PAPC expression by coinjecting Mespo MO with pCS–mMespo. Arrowhead in panel G indicates the disruption of segmental expression of XDelta-1. Arrowhead in panel I indicates the increase of XBra expression. Arrowhead in panel K indicates the increase of Mespo expression.
results in the opposite effect where Mespo expression is markedly reduced (56%, N = 65, Fig. 4C). We also observed that Mespo expression is decreased at stage 10 when injecting
embryos with lef1–EnR mRNA at four-cell stage (Supplemental Fig. 2D). These changes are tightly associated with the changing of Wnt/β-catenin signaling activity, because the
J. Wang et al. / Developmental Biology 304 (2007) 836–847
841
Mespo is a direct downstream target of Wnt/β-catenin signaling cascade
Fig. 3. Wnt/β-catenin signaling in the PSM. (A–J) The neurula stage embryos were stained by whole-mount in situ hybridization for Xwnt8 (A, B, arrowhead), Dkk1 (C, D, arrow), Xwnt8 and Dkk1 (E, F, arrowhead indicates XWnt8 and arrow indicates DKK1), Xwnt8 and Thy (G, H, arrowhead indicates XWn8 and arrow indicates Thy), or Mespo and Thy (I, J, arrowhead indicates Mespo and arrow indicates Thy). Embryos in panels A, C, E, G and I are oriented with anterior to the left and shown in lateral views. Embryos in panels B, D, F, H and J show posterior–dorsal view. (K) Schematic diagram (adapted from Kim et al., 2000) illustrates gene expression patterns related to somite formation in the PSM of Xenopus embryos.
expression patterns of Mespo or Paraxis are un-changed in control embryos or control side (Figs. 4A, D, E, F). These results indicate that the activity of Wnt/β-catenin signaling is responsible for Mespo expression in the PSM.
To experimentally determine how Wnt/β-catenin signaling regulates Mespo expression, we identified a 4.3 kb Mespo promoter that is able to recapitulate the endogenous Mespo expression from gastrulae to tailbud stage (Wang and Ding, 2006) and used it for further analysis. We first tested whether pCS–Lef1–VP16 could induce the expression of a luciferase reporter controlled by the 4.3 kb Mespo promoter (p-4317Luc) in animal caps. A 10-fold induction of the reporter activity is observed (Fig. 5B). To investigate whether Mespo is a direct target of Wnt/ β-catenin signaling, we examined the Mespo promoter for LEF/TCF binding sites. A perfectly conserved LEF/TCF binding site (site L) was found (van de Wetering et al., 1991), at position − 73 to − 67 relative to the transcription start. In a search of GenBank database for the 5′ regulatory regions of homologous Mespo promoters in zebrafish and chick, we found that this LEF/TCF binding site is conserved in zMespo (the zebrafish Mespo homolog) and cMespo-1 (the chick Mespo homolog) (Fig. 5A). The conservation within the regulatory regions among these vertebrates suggests that this LEF/TCF binding site plays an important role in the regulation of Mespo expression. We then generated a mutation in site L and examined the activity of the mutant promoter in animal cap assay. The results show that the mutant promoter construct (p-4317mLuc) has similar activity as the p-4317Luc (Fig. 5C). However, when coinjected with pCS–Lef1–VP16, the luciferase activity of the p-4317mLuc plasmid is significantly reduced as compared to that of the p-4317Luc plasmid (Fig. 5D). Based on the results obtained from animal cap assays, we injected reporter plasmids into the VMZ of both cells of embryos at four-cell stage to examine the function of the LEF/ TCF binding site (site L) in vivo. The results show that the luciferase activity of the p-4317mLuc plasmid is reduced to half of the p-4317Luc plasmid (Fig. 5E). The functional importance of site L was further confirmed by generating transgenic embryos with the 4.3 kb GFP construct carrying the site L mutation. When site L was mutated, GFP expression in the PSM is abolished in approximately 90% transgenic embryos (N = 200, Fig. 5H) while the expression of GFP mRNA in approximately 51% p-4317GFP transgenic embryos (N = 180, Fig. 5G) is detected in the PSM (Fig. 5I), strongly suggesting that the site L is indeed essential for Mespo expression in the PSM. LEF1 binds site L in vivo To examine whether Lef1 binds to site L, chromatin immunoprecipitation (ChIP) assay was performed as previously described (Sachs and Shi, 2000). HA-tagged Lef1 was coinjected with a truncated form of Mespo promoter, p-511, which comprised a proximal 511-bp region of the promoter, followed by precipitation of plasmid DNA bound to Lef1 with an antiHA antibody. PCR amplification using specific primers flanking site L revealed that binding of Lef1 to wild type site
842
J. Wang et al. / Developmental Biology 304 (2007) 836–847
Fig. 4. Wnt/β-catenin signaling regulates Mespo expression. For all embryos, anterior is to the left. Four-cell-stage embryos were injected in VMZ of one blastomere with lineage tracer pCS–LacZ as control (A, D), or together with pCS–Lef1–VP16 (B, E) or together with pCS–Lef1–EnR (C, F). At stage 19, whole mount in situ hybridization was performed for Mespo (A–C) or Paraxis (D–F), as indicated. Arrowhead in panel B shows the increase of Mespo expression. Arrowhead in panel C shows the decrease of Mespo expression.
L is significantly stronger than that to the mutated one (Fig. 6A). We also examined whether the binding of Lef1 to the Mespo regulatory region can be detected in vivo. Indeed, when injecting HA-tagged Lef1 into both VMZ cells at four-cell stage, the binding of Lef1 to site L is detected (Fig. 6B). As a control, no binding of Lef1 is detected in the exon region of Mespo (Fig. 6B). These results indicate that Mespo is a direct downstream target of Wnt/β-catenin signaling in the PSM. PI3-K/AKT signaling regulates Mespo expression Recent work indicates that FGF signaling is involved in controlling somitogenesis through MKK1/ERK signaling pathway by regulating Mespo expression (Delfini et al., 2005; Moreno and Kintner, 2004). However, FGF signaling has other downstream effectors like PI3-K/AKT in the PSM (Dubrulle and Pourquie, 2004). This raises a question of whether PI3-K/ AKT-mediated FGF signaling contributes to the regulation of Mespo expression. In order to answer this question, we checked whether Akt is present in the PSM as the phosphorylated form. Results from Western blot show that both Akt and its phosphorylated form are detected in the caudal PSM (Fig. 7B). To evaluate whether PI3-K/AKT signaling contributes to the expression of Mespo, we treated embryos with LY294002 which is a specific PI3 kinase inhibitor (Knight et al., 2006; Walker et al., 2000) and then dissected caudal PSM for analysis. Treating embryos with LY294002 causes a reduction of Akt phosphorylation (Fig. 7C), and leads to a decreased Mespo expression (67%, N = 55, Fig. 7E) as compared with the control (Fig. 7D). LY294002 or DMSO treatment causes no change in Paraxis expression (Figs. 7F, G). The decrease of Mespo expression was further
confirmed by RT–PCR. As shown in Fig. 7H, Mespo mRNA is dramatically downregulated when the caudal PSM cells were treated with LY294002 (Fig. 7H). These results indicate that PI3-K/AKT signaling contributes to Mespo expression. Interaction of PI3-K/AKT and Wnt signaling on Mespo expression Since Wnt/β-catenin signaling directly regulates Mespo expression and blocking PI3-K/AKT pathway decreases Mespo expression, we reasoned that Wnt/β-catenin and PI3-K/AKT pathway might cooperate to regulate Mespo expression. To test this possibility, we examined whether the activity of Wnt/β-catenin signaling was affected when PI3-K/ AKT signaling was blocked. A sharp decrease of β-catenin protein in the nuclei of the caudal PSM cells was detected after a 2-h treatment with LY294002, and a decrease in cytosolic β-catenin was also observed (Fig. 8A). To test what led to the loss of β-catenin, we took advantage of a specific antibody, which only recognizes the phosphor-form of βcatenin that is phosphorylated by GSK-3β. Western blot using this antibody shows that when treating the PSM cells with LY294002 the phosphorylated form of β-catenin is significantly increased within 1.5 h, and decreased after 2.5 h (Fig. 8B). This result indicates that the blocking of PI3-K/AKT pathway causes the phosphorylation and degradation of βcatenin via an Axin/GSK-3β-dependent mechanism (Aberle et al., 1997; Polakis, 2001). Therefore, PI3-K/AKT pathway affects the stability of β-catenin. On the other hand, when injecting embryos with pCS–Lef1–VP16 DNA to activate Wnt/β-catenin signaling, the injected side shows increased Mespo expression (79%, N = 70, Fig. 8D). However, LY294002 treatment could not cause significant reduction of
J. Wang et al. / Developmental Biology 304 (2007) 836–847
843
Fig. 5. The LEF/TCF binding site is responsible for the expression of Mespo. (A) Schematic representation of Mespo promoters from various species with a conserved LEF/TCF binding site (red). The position of potential LEF/TCF binding site (box) in Xenopus Mespo 5′-upstream sequence is indicated. (B–D) Fold induction of pCS–Lef1–VP16 in pGL3basic and p-4317Luc reporter plasmids (B), luciferase activities of p-4317Luc and p-4317mLuc reporter plasmids (C) or fold induction of pCS–Lef1–VP16 in p-4317Luc and p-4317mLuc reporter plasmids (D) in animal cap assays. Plasmids were injected into the animal pole at the four-cell stage. Animal caps were dissected at stage 8.5 and luciferase activities were measured at stage 12.5. RLU, relative light units. (E) Luciferase activities of the indicated plasmids injected into both VMZ cells at four-cell stage and measured at stage 19. RLU, relative light units. (F) Expression of endogenous Mespo mRNA at the tailbud stage. (G–I) Expression of GFP mRNA in p-4317GFP transgenic embryos (G) or in p-4317mGFP transgenic embryos (H) at the tailbud stage. The percentages of transgenic embryos that show the PSM expression (purple), non-specific expression (brown), and no detected signal (white) are shown in panel I.
the increased Mespo expression induced by injecting pCS– Lef1–VP16, while Mespo expression is clearly reduced in the uninjected side (61%, N = 90, Fig. 8E). Thus injecting pCS– Lef1–VP16 expressing the constitutively active Lef1–VP16 minimizes the effect of blocking the PI3-K/AKT pathway. Taken together, these results indicate that the PI3-K/AKT pathway participates in regulating Wnt/β-catenin signaling mediated Mespo expression, through the regulation of βcatenin stability.
Discussion Mespo is involved in segmental patterning during Xenopus somitogenesis Mespo knock out mouse shows severe defects in somitogenesis, including perturbation of segmental patterning. The mutant mice have no identifiable somites, and the expression of genes associated with somitogenesis like Mesp2 (the ortholog of
844
J. Wang et al. / Developmental Biology 304 (2007) 836–847
and found that XWnt3a expression is restricted in neural crest cells, while XWnt8 expression is detected in the caudal PSM (Figs. 3A, B, E–H). Furthermore, the Wnt antagonist Dkk1 is expressed in somitomeres (Figs. 3C–F). These observations suggested that the activity of Wnt/β-catenin signaling exists in
Fig. 6. Lef1 binds to site L in the Mespo promoter in vivo. (A) ChIP assay of embryos injected with pCS–Lef1–HA and p-511, or pCS–Lef1–HA and p511m in both VMZ cells at four-cell stage to detect the binding of Lef1 to site L. (B) ChIP assay detecting the binding of Lef1 to endogenous Mespo regulatory sequence. pCS–Lef1–HA was injected into both VMZ cells of embryos at fourcell stage.
Xenopus Thy), PAPC and Dll-3 (the ortholog of XDelta-1) is abolished in absence of Mespo (Yoon and Wold, 2000). This phenotype implies that Mespo is involved in establishing segmental patterning in somitogenesis. Here, we report that in Xenopus embryos depleting Mespo by Mespo Mo causes disorder of somite cell arrangement. During Xenopus somitogenesis, the rostral PSM cells compact to form cell aggregates named somitomeres, which later separate from the PSM to form somites. The nuclei of somite cells line up in the middle of each somite, representing the well-organized segmental pattern. There is no such distinct morphological feature in Mespodepleting Xenopus embryos. Instead, nuclei are randomly distributed, indicating the disorganization of somite cells (Fig. 1H). The expression patterns of genes associated with segmental patterning are also perturbed (Fig. 2). Thus, our data indicate that Mespo functions in establishing the segmental pattern by means of regulating the genes involved in Xenopus somitogenesis. Wnt/β-catenin signaling controls gene expressions in somitogenesis Previous studies have demonstrated that Wnt3a is involved in somitogenesis (Aulehla et al., 2003; Galceran et al., 2004; Hofmann et al., 2004; Ikeya and Takada, 2001). In chick embryos, increasing Wnt/β-catenin signaling activity by implanting beads covered with Wnt3a expressing cells induced smaller somites (Aulehla et al., 2003). In mouse embryos, it has been proposed that Wnt3a is involved in somitogenesis via Axin2 (Aulehla et al., 2003) and Wnt/β-catenin signaling controlled segmentation by directly regulating Dll1 expression (Galceran et al., 2004; Hofmann et al., 2004). However, the effect of Wnt/β-catenin signaling in somitogenesis has not been fully understood. Our study provides new information on the role of Wnt/β-catenin signaling in somitogenesis. We checked XWnt3a and XWnt8 expression in Xenopus embryos
Fig. 7. PI3-K/AKT signaling regulates Mespo expression. (A) Hatched lines indicate the three caudal regions (C, R, and S) used for western blot of Akt and phosphorylated Akt. (B) Western blot of Akt and phosphorylated Akt in different regions indicated in panel A. (C) Western blot of phosphorylated Akt in the caudal PSM equivalent to region C (illustrated in panel A) treated with either LY294002 or DMSO. (D–G) The neurula stage embryos were treated with either DMSO (D, F), or LY294002 (E, G) for 2 h, and then stained for Mespo and Paraxis as indicated. (H) RT–PCR of Mespo mRNA in the caudal PSM equivalent to region C (illustrated in panel A) treated with either DMSO or LY294002. The caudal PSM were dissected at stage 23 and treated with drugs for 2 h, then harvested for RT–PCR.
J. Wang et al. / Developmental Biology 304 (2007) 836–847
845
Fig. 8. Interaction between PI3-K/AKT and Wnt/β-catenin signaling in PSM. (A) Western blot of β-catenin in the nucleus and cytosol of the caudal PSM cells from embryos treated with either DMSO or LY294002 for 2 h. (B) Western blot of β-catenin, phosphorylated β-catenin, Akt and phosphorylated Akt in the caudal PSM cells treated with DMSO for 2.5 h, or LY294002 for 1.5 h or 2.5 h. (C–E) Four-cell-stage embryos were injected in VMZ of one cell with pCS–LacZ as control (C) or together with pCS–Lef1–VP16 (D, E). At the neurula stage, injected embryos were treated with DMSO (C, D) or LY294002 (E) for 2 h, then fixed, stained with X-gal to reveal the tracer (light blue) and analyzed by whole-mount in situ hybridization for Mespo. Arrow in panel D indicates the increase of Mespo expression in the injected side and arrowhead indicates the expression of Mespo in the uninjected side. Arrow in panel E indicates the increased Mespo expression in the injected side and arrowhead indicates the decrease of Mespo expression in the uninjected side.
Xenopus PSM and is restricted in the caudal PSM. This Wnt/ β-catenin signaling activity is essential for Mespo expression. Increasing Wnt/β-catenin signaling activity by injecting pCS– Lef1–VP16 enhances Mespo expression in the PSM, and decreasing Wnt/β-catenin signaling activity by injecting pCS– Lef1–EnR downregulates Mespo expression (Fig. 4). We also found that Wnt/β-catenin signaling, acting upstream of Mespo, directly controls Mespo expression, based on the recruitment of Lef1 to the site in the Mespo 5′-regulatory region (Fig. 6B). Since Mespo expression is essential for segmental patterning, we concluded that Wnt/β-catenin signaling controls segmentation process by regulating Mespo expression.
Regulation of Mespo expression in the PSM The bHLH family member Mespo is a key regulator of genes associated with proper segmentation and successful somitogenesis (Yoon and Wold, 2000). FGF and RA signaling were proposed to act upstream of Mespo. It has been shown that blocking FGF signaling by SU5402 or increasing RA signaling by treating with RA caused a decrease of Mespo expression (Moreno and Kintner, 2004). However, the molecular nature of how Mespo expression is regulated remains unclear. We show here that Wnt/β-catenin signaling is required for Mespo expression. Increasing Wnt/β-catenin signaling activity by overexpressing β-catenin or Lef1–VP16 increases the
846
J. Wang et al. / Developmental Biology 304 (2007) 836–847
expression of Mespo (Fig. 4B and Supplemental Fig. 2B), and interfering Wnt/β-catenin signaling by overexpressing Lef1– EnR resulted in the opposite effect (Fig. 4C and Supplemental Fig. 2D). Our results also show that the PI3-K/AKT pathway participates in regulating Mespo expression (Fig. 7E). Since it has been proposed that MKK1/ERK signaling is involved in regulating Mespo expression (Moreno and Kintner, 2004), it seems that FGF signaling contributes to the expression of Mespo through PI3-K/AKT and MKK1/ERK pathway together. Thus, multiple signaling pathways regulate expression of Mespo that is a key regulator of somitogenesis. Interaction of signals in somitogenesis Previous studies have shown that FGF, Wnt and RA signaling are all involved in somitogenesis (Aulehla et al., 2003; Dubrulle et al., 2001; Galceran et al., 2004; Hofmann et al., 2004; Vermot and Pourquie, 2005). However, whether or how these signals interact to precisely control somite formation is poorly understood. In this report, we have provided evidence that Wnt/β-catenin signaling is involved in controlling somitogenesis by interacting with FGF signaling. FGF signaling downstream effector PI3-K/AKT regulated the stability of β-catenin in the nucleus of the PSM cells (Fig. 8A). Several lines have also shown that RA and FGF signaling interact to regulate somitogenesis. RA signaling may inhibit the MAPK/ERK pathway by activating the expression of MKP3, and FGF signaling seems to regulate RA signaling activity by regulating the expression of enzymes for RA production and degradation (Moreno and Kintner, 2004). Further experiments are needed to explore whether Wnt and RA signaling interact in somitogenesis. Acknowledgments The authors are grateful to Drs. D. Sparrow, E. Joseph, C. Niehrs and J.C. Smith for sharing plasmids, Dr. Schneider (Max-Planck-Institut fur Entwicklungsbiologie) for β-catenin antibody. We thank Drs. W. Wu and Q. Tao for discussion of the project. This work was supported by the National Natural Science Foundation of China (90408005, 30270650) and the National Key Project for Basic Science Research of China (2001CB509901, 2005CB522704) to X.D. We thank Dr. Dangsheng Li for his professional polish of the MS. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ydbio.2006.12.034. References Aberle, H., Bauer, A., Stappert, J., Kispert, A., Kemler, R., 1997. beta-Catenin is a target for the ubiquitin–proteasome pathway. EMBO J. 16, 3797–3804. Aulehla, A., Wehrle, C., Brand-Saberi, B., Kemler, R., Gossler, A., Kanzler, B., Herrmann, B.G., 2003. Wnt3a plays a major role in the segmentation clock controlling somitogenesis. Dev. Cell 4, 395–406.
Conlon, F.L., Smith, J.C., 1999. Interference with brachyury function inhibits convergent extension, causes apoptosis, and reveals separate requirements in the FGF and activin signalling pathways. Dev. Biol. 213, 85–100. Delfini, M.C., Dubrulle, J., Malapert, P., Chal, J., Pourquie, O., 2005. Control of the segmentation process by graded MAPK/ERK activation in the chick embryo. Proc. Natl. Acad. Sci. U. S. A. 102, 11343–11348. Ding, X., Hausen, P., Steinbeisser, H., 1998. Pre-MBT patterning of early gene regulation in Xenopus: the role of the cortical rotation and mesoderm induction. Mech. Dev. 70, 15–24. Dubrulle, J., Pourquie, O., 2004. fgf8 mRNA decay establishes a gradient that couples axial elongation to patterning in the vertebrate embryo. Nature 427, 419–422. Dubrulle, J., McGrew, M.J., Pourquie, O., 2001. FGF signaling controls somite boundary position and regulates segmentation clock control of spatiotemporal Hox gene activation. Cell 106, 219–232. Galceran, J., Farinas, I., Depew, M.J., Clevers, H., Grosschedl, R., 1999. Wnt3a−/−-like phenotype and limb deficiency in Lef1(−/−)Tcf1(−/−) mice. Genes Dev. 13, 709–717. Galceran, J., Sustmann, C., Hsu, S.C., Folberth, S., Grosschedl, R., 2004. LEF1mediated regulation of Delta-like1 links Wnt and Notch signaling in somitogenesis. Genes Dev. 18, 2718–2723. Geng, X., Xiao, L., Lin, G.F., Hu, R., Wang, J.H., Rupp, R.A., Ding, X., 2003. Lef/Tcf-dependent Wnt/beta-catenin signaling during Xenopus axis specification. FEBS Lett. 547, 1–6. Glinka, A., Wu, W., Delius, H., Monaghan, A.P., Blumenstock, C., Niehrs, C., 1998. Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391, 357–362. Haq, S., Michael, A., Andreucci, M., Bhattacharya, K., Dotto, P., Walters, B., Woodgett, J., Kilter, H., Force, T., 2003. Stabilization of beta-catenin by a Wnt-independent mechanism regulates cardiomyocyte growth. Proc. Natl. Acad. Sci. U. S. A. 100, 4610–4615. Harland, R.M., 1991. In situ hybridization: an improved whole-mount method for Xenopus embryos. Methods Cell Biol. 36, 685–695. He, Z., Li, J., Zhen, C., Feng, L., Ding, X., 2005. Knockdown of p53 by RNAi in ES cells facilitates RA-induced differentiation into muscle cells. Biochem. Biophys. Res. Commun. 335, 676–683. Hofmann, M., Schuster-Gossler, K., Watabe-Rudolph, M., Aulehla, A., Herrmann, B.G., Gossler, A., 2004. WNT signaling, in synergy with T/ TBX6, controls Notch signaling by regulating Dll1 expression in the presomitic mesoderm of mouse embryos. Genes Dev. 18, 2712–2717. Ikeya, M., Takada, S., 2001. Wnt-3a is required for somite specification along the anteroposterior axis of the mouse embryo and for regulation of cdx-1 expression. Mech. Dev. 103, 27–33. Joseph, E.M., Cassetta, L.A., 1999. Mespo: a novel basic helix–loop–helix gene expressed in the presomitic mesoderm and posterior tailbud of Xenopus embryos. Mech. Dev. 82, 191–194. Kazanskaya, O., Glinka, A., del Barco Barrantes, I., Stannek, P., Niehrs, C., Wu, W., 2004. R-Spondin2 is a secreted activator of Wnt/beta-catenin signaling and is required for Xenopus myogenesis. Dev. Cell 7, 525–534. Kim, S.H., Jen, W.C., De Robertis, E.M., Kintner, C., 2000. The protocadherin PAPC establishes segmental boundaries during somitogenesis in Xenopus embryos. Curr. Biol. 10, 821–830. Kintner, C.R., Brockes, J.P., 1985. Monoclonal antibodies to the cells of a regenerating limb. J. Embryol. Exp. Morphol. 89, 37–55. Knight, Z.A., Gonzalez, B., Feldman, M.E., Zunder, E.R., Goldenberg, D.D., Williams, O., Loewith, R., Stokoe, D., Balla, A., Toth, B., Balla, T., Weiss, W. A., Williams, R.L., Shokat, K.M., 2006. A pharmacological map of the PI3-K family defines a role for p110alpha in insulin signaling. Cell 125, 733–747. Kroll, K.L., Amaya, E., 1996. Transgenic Xenopus embryos from sperm nuclear transplantations reveal FGF signaling requirements during gastrulation. Development 122, 3173–3183. Moreno, T.A., Kintner, C., 2004. Regulation of segmental patterning by retinoic acid signaling during Xenopus somitogenesis. Dev. Cell 6, 205–218. Nieuwkoop, P.D., Faber, J., 1967. Normal Table of Xenopus laevis. North Holland, Amsterdam. Polakis, P., 2001. More than one way to skin a catenin. Cell 105, 563–566. Pourquie, O., 2001. Vertebrate somitogenesis. Annu. Rev. Cell Dev. Biol. 17, 311–350.
J. Wang et al. / Developmental Biology 304 (2007) 836–847 Pourquie, O., 2003. The segmentation clock: converting embryonic time into spatial pattern. Science 301, 328–330. Sachs, L.M., Shi, Y.B., 2000. Targeted chromatin binding and histone acetylation in vivo by thyroid hormone receptor during amphibian development. Proc. Natl. Acad. Sci. U. S. A. 97, 13138–13143. Smith, J.C., Price, B.M., Green, J.B., Weigel, D., Herrmann, B.G., 1991. Expression of a Xenopus homolog of brachyury (T) is an immediate-early response to mesoderm induction. Cell 67, 79–87. Sparrow, D.B., Jen, W.C., Kotecha, S., Towers, N., Kintner, C., Mohun, T.J., 1998. Thylacine 1 is expressed segmentally within the paraxial mesoderm of the Xenopus embryo and interacts with the Notch pathway. Development 125, 2041–2051. Takada, S., Stark, K.L., Shea, M.J., Vassileva, G., McMahon, J.A., McMahon, A.P., 1994. Wnt-3a regulates somite and tailbud formation in the mouse embryo. Genes Dev. 8, 174–189. van de Wetering, M., Oosterwegel, M., Dooijes, D., Clevers, H., 1991. Identification and cloning of TCF-1, a T lymphocyte-specific transcription factor containing a sequence-specific HMG box. EMBO J. 10, 123–132. Vermot, J., Pourquie, O., 2005. Retinoic acid coordinates somitogenesis and left–right patterning in vertebrate embryos. Nature 435, 215–220.
847
Walker, E.H., Pacold, M.E., Perisic, O., Stephens, L., Hawkins, P.T., Wymann, M.P., Williams, R.L., 2000. Structural determinants of phosphoinositide 3-kinase inhibition by wortmannin, LY294002, quercetin, myricetin, and staurosporine. Mol. Cell 6, 909–919. Wang, J., Ding, X., 2006. Cloning and analyzing of Xenopus Mespo promoter in retinoic acid regulated Mespo expression. Acta Biochim. Biophys. Sin. 38 (11), 759–764 (Shanghai). Weisblat, D.A., Wedeen, C.J., Kostriken, R.G., 1994. Evolution of developmental mechanisms: spatial and temporal modes of rostrocaudal patterning. Curr. Top. Dev. Biol. 29, 101–134. Wilson, V., Rashbass, P., Beddington, R.S., 1993. Chimeric analysis of T (Brachyury) gene function. Development 117, 1321–1331. Yamaguchi, T.P., Takada, S., Yoshikawa, Y., Wu, N., McMahon, A.P., 1999. T (Brachyury) is a direct target of Wnt3a during paraxial mesoderm specification. Genes Dev. 13, 3185–3190. Yoon, J.K., Wold, B., 2000. The bHLH regulator pMesogenin1 is required for maturation and segmentation of paraxial mesoderm. Genes Dev. 14, 3204–3214. Yoon, J.K., Moon, R.T., Wold, B., 2000. The bHLH class protein pMesogenin1 can specify paraxial mesoderm phenotypes. Dev. Biol. 222, 376–391.
Genomes & Developmental Control
Wnt/β-catenin signaling controls Mespo expression to regulate segmentation during Xenopus somitogenesis Jinhu Wang 1 , Shangwei Li 1 , Yuelei Chen, Xiaoyan Ding ⁎ Laboratory of Molecular Cell Biology, Key Laboratory of Stem Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue Yang Road, Shanghai 200031, China Graduate School of Chinese Academy of Sciences, 320 Yue Yang Road, Shanghai 200031, China Received for publication 12 May 2006; revised 17 November 2006; accepted 14 December 2006 Available online 21 December 2006
Abstract The vertebral column is derived from somites, which are transient segments of the paraxial mesoderm that are present in developing vertebrates. The strict spatial and temporal regulation of somitogenesis is of crucial developmental importance. Signals such as Wnt and FGF play roles in somitogenesis, but details regarding how Wnt signaling functions in this process remain unclear. In this study, we report that Wnt/β-catenin signaling regulates the expression of Mespo, a basic–helix–loop–helix (bHLH) gene critical for segmental patterning in Xenopus somitogenesis. Transgenic analysis of the Mespo promoter identifies Mespo as a direct downstream target of Wnt/β-catenin signaling pathway. We also demonstrate that activity of Wnt/β-catenin signaling in somitogenesis can be enhanced by the PI3-K/AKT pathway. Our results illustrate that Wnt/ β-catenin signaling in conjunction with PI3-K/AKT pathway plays a key role in controlling development of the paraxial mesoderm. © 2007 Published by Elsevier Inc. Keywords: AKT; Paraxial mesoderm; PI3-K; Wnt; Xenopus laevis
Introduction During vertebrate embryogenesis, the paraxial mesoderm separates into somites, a metameric series of homologous subunits, which bud off sequentially from the rostral end of the presomitic mesoderm (PSM) at regular intervals along the anterior–posterior (AP) axis. The process in which the segmental or metameric pattern is established accompanying the constant posterior elongation of the embryo body axis is termed somitogenesis (Weisblat et al., 1994). Somitogenesis occurs in a strict spatial and temporal progression (Dubrulle et al., 2001; Pourquie, 2003). The molecular nature of the signals responsible for controlling somitogenesis remains elusive. One potential candidate for regulating this process is Wnt/β-catenin signaling (Aulehla et al., 2003; Yamaguchi et al., 1999). Previous studies have shown that Wnt/β-catenin signaling is implicated in ⁎ Corresponding author. Fax: +86 21 34230165. E-mail address: [email protected] (X. Ding). 1 These authors contributed equally to this work. 0012-1606/$ - see front matter © 2007 Published by Elsevier Inc. doi:10.1016/j.ydbio.2006.12.034
controlling the development of paraxial mesoderm (Aulehla et al., 2003; Galceran et al., 2004; Hofmann et al., 2004; Pourquie, 2001). In mouse embryos, Wnt3a is expressed in the primitive streak during gastrulation and in the tailbud during the later stages of mouse development. Animals homozygous for null alleles of Wnt3a lack somites posterior to the forelimbs (Ikeya and Takada, 2001; Takada et al., 1994; Wilson et al., 1993). The double mutant harboring mutations in lef1 and tcf1, identified as nuclear mediators of Wnt/β-catenin signaling that activate Wnt-responsive genes by associating with β-catenin, results in lack of paraxial mesoderm resembling the Wnt3a null mutation (Galceran et al., 1999). In chick embryos, it has been shown that altering Wnt/β-catenin signaling activity in the PSM causes defects in somite formation (Aulehla et al., 2003). However, the exact molecular nature of Wnt/β-catenin signaling in somitogenesis is not fully understood, and whether Wnt/ β-catenin signaling is involved in somitogenesis of the lower vertebrates such as Xenopus is still questionable. The FGF signaling pathway also participates in somitogenesis (Delfini et al., 2005; Dubrulle et al., 2001; Moreno and Kintner, 2004). Fgf8 mRNA forms a gradient along the PSM in mouse
J. Wang et al. / Developmental Biology 304 (2007) 836–847
embryos (Delfini et al., 2005; Dubrulle et al., 2001). Altering the activity of FGF signaling in the PSM resulted in segmental defects (Dubrulle et al., 2001). These data demonstrate that FGF signaling is required for somitogenesis. It has been proposed that FGF signaling may act through the PI3-K/AKT pathway in the PSM to participate in somitogenesis (Dubrulle and Pourquie, 2004). Since PI3-K/AKT signaling can enhance accumulation of β-catenin by inhibiting GSK-3β during cardiomyocyte growth (Haq et al., 2003), it may regulate Wnt/β-catenin activity in the segmentation process by a similar mechanism. Somite formation and patterning are thought to be controlled by Mespo, a member of the basic–helix–loop–helix (bHLH) transcription factor family (Moreno and Kintner, 2004; Yoon et al., 2000; Yoon and Wold, 2000). Mespo expression encompasses the entire caudal PSM of Xenopus embryos, and its expression is reduced when the PSM cells enter the segmentation process (Joseph and Cassetta, 1999; Kim et al., 2000; Moreno and Kintner, 2004). Gain of function experiments showed that Mespo can induce expression of paraxial mesoderm marker genes (Yoon et al., 2000). Mespo null mouse embryos exhibited no identifiable somites or segmental patterning in the trunk posterior to forelimbs, and severe disruption of gene expressions in posterior paraxial mesoderm (Yoon and Wold, 2000). These results suggest that Mespo plays essential roles in somitogenesis. The phenotypic resemblance between Mespo and Wnt3a mutant mouse embryos provides a strong hint that Wnt/β-catenin signaling regulates the expression of Mespo during paraxial mesoderm development. However, whether Mespo expression is regulated by Wnt/β-catenin signaling in the PSM remains unclear. Here, we report that Wnt/β-catenin signaling directly regulates the expression of Mespo, and this regulation is enhanced by PI3-K/AKT signaling. These results demonstrate that Wnt/β-catenin signaling plays a key role in controlling somitogenesis by regulating Mespo gene expression, and that PI3-K/AKT signaling is also involved in this process. Materials and methods Embryo manipulations, drug treatment Preparation and injection of Xenopus embryos were carried out as previously described (Ding et al., 1998). Embryos were staged according to Nieuwkoop and Faber (1967). LY294002 (Calbiochem) was used at 100 μM. Chemicals were dissolved in DMSO and then diluted into 0.1× MMR. In each case, about 20 embryos were examined per condition per probe. Each experiment was repeated three times independently.
Morpholino oligomer and in vitro translation The 25-bp morpholino antisense oligomer for Xenopus Mespo was obtained from Gene Tools and consisted of the following sequence: MespoMO: GGGATGGTGCAGAGTCTCCATCAGT Morpholino was dissolved in water to a concentration of 1 mM, which was then further diluted to a give a working solution of 0.3 mM. For the rescue assay, we generated a mutated Mespo cDNA (mMespo) differed in seven bases (underlined) from Mespo: 5′-CTACTATGGAGACTCTGCACCATCCCCT.
837
The in vitro transcription/translation of Mespo was performed according to the protocol of TNT coupled reticulocyte lysate system (Promega, L4610).
In situ hybridization, immunocytochemistry Whole-mount in situ hybridization was performed as previously described (Harland, 1991) using DIG-labeled probes. For bleaching of wild-type embryos, hybridized embryos were treated with bleaching solution (0.5× SSC with 1% hydrogen peroxide and 5% formamide) under a fluorescent light. The following plasmid templates were linearized, and DIG-labeled antisense RNA probes were made using either T7 or SP6 RNA polymerase: Thylacine1 (Sparrow et al., 1998); Mespo (Joseph and Cassetta, 1999); GFP (Geng et al., 2003); Xwnt8 (Kazanskaya et al., 2004) and Dkk1 (Glinka et al., 1998); Paraxis (a PCRamplified full-length cDNA of Paraxis was cloned into pCS2+); AKT (a PCRamplified coding region of Xenopus AKT was cloned into pCS2+); PAPC (Kim et al., 2000); XBra (Smith et al., 1991). To visualize somite size, embryos were fixed in 2% paraformaldehyde, 2% glutaraldehyde in PBS, and stained with the muscle-specific monoclonal antibody 12/101 (Kintner and Brockes, 1985). Then embryos were post-fixed in MEMFA (0.1 M MOPS pH 7.4, 2 mM EGTA, 1 mM MgSO4, 3.7% formaldehyde), embedded in paraffin, sectioned at 10 μm, and stained with DAPI.
Mutagenesis and transgenesis The Exsite™ PCR-based site-directed mutagenesis system (Stratagene, cat. 200502-5) was used to generate the mutants of site L, p-4317mLuc and p-511m. The primers are shown below (nucleotides altered shown in bold): 5′AGGAGACAGCAAGGTGTTAACATG-3′ (reverse) and 5′-CTAGACTCCTCCATTAAAACGCCCACT-3′ (forward). Transgenic Xenopus embryos were generated as described previously (Kroll and Amaya, 1996). Plasmids used for transgenesis assays were linearized by NotI digestion. Three independent rounds of transgenesis were performed for each plasmid.
Western blot For measuring phospho-Akt levels, the caudal region of Xenopus embryos at stage 23 were divided into three equal parts corresponding to the caudal domain, the rostral domain of the PSM and a region including the last formed somites. In these experiments, 100 embryos were used. The samples were lysed in EDTAfree RIPA (0.1% NP40, 20 mM Tris–HCl pH 8, 10% glycerol) with cocktail protease inhibitors (Bio Basics) and Na3VO4. For measuring accumulation of β-catenin in the nucleus and cytosol, the caudal domain of the PSM was dissected and cultured in 1× MMR containing LY294002 at 100 μM or DMSO for 2 h. Then β-catenin in either nucleus or cytosol was extracted according to PIERCE protocol (NE-PER™ nuclear and cytoplasmic extraction reagents, 78833). All the samples were quantitated by Bradford assay and adjusted to equal concentration before loading. Western blot was performed as described (He et al., 2005). Polyclonal antibody P14L was used to visualize β-catenin (dilution 1:2000), and anti-phospho-Akt polyclonal antibodies (dilution 1:2000; Cell Signaling Technology) were used to detect endogenous Akt and phospho-Akt. For a loading control, equal volume of proteins were separated by SDS–PAGE and visualized with Coomassie Blue.
Chromatin immunoprecipitation (ChIP) Chromatin immunoprecipitation (ChIP) assay was performed, with modifications, according to the method described previously (Sachs and Shi, 2000). 100 Xenopus embryos were selected at stage 13 from embryos injected with appropriate plasmids, and crosslinked with 1% formaldehyde at room temperature for 20 min. Following treatment with 0.125 M glycine, embryos were washed and resuspended in 200 μl ChIP lysis buffer. The lysates were sonicated on ice to obtain DNA fragments on an average around 500 bp and immunoprecipitated with anti-HA monoclonal antibody (dilution 1:3000; Sigma). DNA fragments were washed, eluted and recovered from the immunoprecipitated complex. PCR was performed under an optimized condition: 94 °C
838
J. Wang et al. / Developmental Biology 304 (2007) 836–847
for 30 s, 58 °C for 30 s, and 72 °C for 30 s, for 30–35 cycles. Primer sequences are as follows: Mespo promoter upstream 5′-TGAAAGCCAATTTAGCTGAAGG3′, downstream 5′-GCTGGCAAAGTTCCATGTGAG-3′; ODC upstream 5′GATCATGCACATGTCAAGCC-3′, downstream 5′-CAGGGAGAATGCCATGTTCT-3′; pGL3-Basic upstream 5′-CAGGGGATAACGCAGGAAA-3′, downstream 5′-AAGGGAGAAAGGCGGACA G-3′.
Results MO-based Mespo depletion caused defects of somite formation Previous analysis of the morphological defects in Mespo knockout mouse suggests that Mespo plays a role in somitogenesis (Yoon and Wold, 2000). In Xenopus, Mespo expression is firstly detected in ventrolateral mesoderm but is absent from dorsal mesoderm in gastrulae, and is localized to the most caudal (unsegmented) paraxial mesoderm in neurulae
and tailbud stage (Joseph and Cassetta, 1999; Yoon et al., 2000). To test whether Mespo contributes to Xenopus somitogenesis, we used antisense morpholino oligonucleotides (MO) to deplete Mespo. We designed a Mespo MO that spans the ATG initiation codon (Fig. 1A). This MO inhibits translation of recombinant Mespo RNA in an in vitro transcription/translation reaction (Fig. 1B). To disrupt endogenous Mespo expression we injected Mespo MO into the ventral marginal zone (VMZ) of one blastomere at four-cell stage, with the uninjected side serving as a control. The somite morphology of the injected embryos was then examined at tadpole stage by whole-mount immunostaining with 12/101 (a monoclonal antibody specifically recognizes differentiated muscle cells). While control MO-injected embryos exhibit no morphological defects in somite formation (Fig. 1C), embryos injected with Mespo MO show remarkable malformation of
Fig. 1. Antisense MO against Mespo depletes Mespo protein and causes somite formation defects. (A) Sequence of Mespo MO aligned with the 5′-untranslated region (UTR) of Mespo RNA. (B) In vitro transcription/translation of Mespo is inhibited by Mespo MO. Arrow indicates the band corresponding to Mespo protein. (C, D) Whole mount immunostaining with 12/101, a muscle-specific antibody, shows that the segmental pattern of somites in Mespo MO-injected embryos (D) is disrupted and the somite size is irregular (compared with control MO-injected embryo in panel C). Bracket in panel D marks the malformed somites. (E–H) Embryos injected with Mespo MO (F, H) or control MO (E, G) stained with 12/101, then sectioned longitudinally and labeled with DAPI. Panels E, G and Panels F, H are the same view of one embryo, respectively. Embryos injected with Mespo MO panel F, H show the disorganization of segmental pattern, based on myotomal morphology (panel F, compared with panel E) and the arrangement of somatic nuclei panel H, compared with G). Bracket in panel H marks the malformed somites.
J. Wang et al. / Developmental Biology 304 (2007) 836–847
somites. The defects include disrupted somite boundaries and irregular somite size (79%, N = 100, Fig. 1D). The malformation of somites in Mespo MO-injected embryos is shown more clearly on frontal sections stained with 12/101 or with DAPI staining of nuclei. Compared with the uninjected side, somites in Mespo MO-injected side are poorly defined, with no metameric shape typical for this tissue. Also, somite nuclei are abnormally aligned, indicating that the patterning of somite is severely disrupted (67%, N = 30, Figs. 1F, H). As a control, the somite morphology in control MO-injected embryos remains identical to that in the uninjected side, with regularly arranged allied cells (Figs. 1E, G). Aiming to verify that the phenotype described above is due to specific depletion of Mespo, we carried out rescue experiments. We prepared a mutated Mespo (mMespo) cDNA, which codes for the same polypeptide as the Mespo protein but differs from Mespo cDNA by seven bases. Injection of the resulting pCS– mMespo plasmid effectively rescues the phenotypic defects caused by Mespo MO in 85% (N = 100) of the injected embryos (Supplemental Fig. 1). We also injected Mespo MO into DMZ of four-cell-stage embryos, where Mespo is not expressed, and found no effect on embryo development (data not shown). These results indicate that the effect of Mespo MO is specific. Taken together, our results indicate that depletion of Mespo caused the morphological defects of somite patterning. Depleting Mespo disrupts expression of genes in the PSM The results described above indicate that lacking of Mespo in embryos causes segmentation defects. In Xenopus, several molecules have been demonstrated to play roles in segmentation process, including PAPC (Kim et al., 2000), Thylacine1 (Thy) (Sparrow et al., 1998) and XDelta-1 (Pourquie, 2003). To evaluate whether depleting Mespo interferes the expression of segmental marker genes, we examined the expression of genes such as PAPC, Thy and XDelta-1 in embryos injected with Mespo MO. Compared with the uninjected side, the expression of PAPC is lost in Mespo MO-injected side (85%, N = 78, Fig. 2B). The loss of PAPC expression in embryos injected with Mespo MO is specific since it could be rescued by coinjecting with pCS–mMespo (70%, N = 60, Fig. 2C). The segmental expression of Thy in somitomeres is severely disrupted (87%, N = 65, Fig. 2E), and it was observed that the somitomeric expression pattern of XDelta-1 is also perturbed (80%, N = 81, Fig. 2G). As a control, injection of a mixture of pCS–LacZ and control MO resulted in no difference in the expression level and pattern of PAPC, Thy or XDelta-1 in both injected and uninjected sides (Figs. 2A, D, F). These results indicate that depletion of Mespo profoundly changes the expression of PAPC, Thy and XDelta-1 in the PSM, suggesting that Mespo acts in the PSM to control segmental patterning. Since XDelta-1 is a component of the Notch/Delta pathway, these results suggest that Mespo controls segmental patterning by regulating Notch/Delta signaling (Pourquie, 2003). To gain further insight into the cause of somite defects in embryos lacking Mespo function, we examined the expression of XBra that plays major roles in development of paraxial
839
mesoderm (Conlon and Smith, 1999). We found that in embryos depleted for Mespo, the expression of XBra is increased in the Mespo MO-injected side (67%, N = 90, Fig. 2I), as compared to the control side, indicating that XBra expression is negatively regulated by Mespo, a phenomenon also observed in Mespo null mutant mouse (Yoon and Wold, 2000). XBra expression is not changed in control MO-injected embryos (Fig. 2H). Interestingly, in embryos injected with Mespo MO, Mespo mRNA is expressed at levels higher than the control side (78%, N = 78, Fig. 2K). Expression of Mespo is not changed in the control MO-injected embryos (Fig. 2J). This result implies that Mespo has a negative regulatory effect on its own mRNA expression in the PSM. Taken together, we provide evidence here that Mespo is responsible not only for segmental specific gene expression, but is also involved in regulating Xbra and Mespo expression. Thus we conclude that Mespo is essential for somitogenesis. Wnt/β-catenin signaling in the PSM Observations from studies on mouse embryos have demonstrated that Wnt/β-catenin signaling is involved in somitogenesis (Aulehla et al., 2003). Regional similarity between expression patterns of Wnts and Mespo in mouse embryos, as well as the phenotypic similarities of Wnt and Mespo null mutant mouse (Takada et al., 1994; Yoon and Wold, 2000), raises a possibility that Wnt signaling regulates Mespo expression in somitogenesis. However, the existence of Wnt signaling in Xenopus PSM has not been documented. We performed whole-mount in situ hybridization to analyze Wnt gene expression in the PSM. Like in other vertebrates, Xenopus PSM is subdivided into two distinct regions featured by expression pattern of the bHLH family genes (Fig. 3K) (Kim et al., 2000; Pourquie, 2003). In the caudal domain of the PSM, Mespo is expressed in the entire region referred as the tailbud domain (TBD) (Figs. 3I, J). Thy transcripts are restricted in the rostral region of the PSM and localized to the anterior half of somitomeres (Figs. 3G, H, I, J). At the neurula stage, XWnt8 is expressed at high level in the TBD (Figs. 3A, B, E, F, G, H), whereas Wnt antagonist Dkk1 is uniformly expressed in somitomeres (Figs. 3C, D, E, F). The expression patterns of these genes indicate that Wnt signals exist in the TBD and support that Wnt/β-catenin signaling is involved in Xenopus somitogenesis. Altering Wnt/β-catenin signaling activity affects expression of Mespo To test whether Wnt/β-catenin signaling regulates Mespo expression, pCS–Lef1–VP16 (a dominant active form of Lef1) DNA was injected into VMZ of one cell of four-cell-stage embryos and Mespo expression was examined at stage 19. Mespo expression in the injected side is increased compared with the control side (75%, N = 70, Fig. 4B). When embryos were injected with pCS–β-catenin DNA, Mespo expression is also increased (Supplemental Fig. 2B). In contrast, injection of pCS–Lef1–EnR (a dominant negative form of Lef1) DNA
840
J. Wang et al. / Developmental Biology 304 (2007) 836–847
Fig. 2. Expression patterns of genes responsible for segmental patterning are altered in embryos lacking Mespo function. For all embryos, anterior is to the left. (A–K) One VMZ cell of four-cell-stage embryos injected with control MO (A, D, F, H and J), or Mespo MO (B, E, G, I and K) or Mespo MO plus 35 pg pCS–mMespo (C) and the embryos were fixed at neurula stage for in situ hybridization to detect the PSM marker gene expression: (A–C) PAPC, (D, E) Thy, (F, G) XDelta-1, (H, I) XBra, (J, K) Mespo. Bracket in panels B and E indicates the disruption of segmental expression of PAPC and Thy, respectively. Bracket in panel C indicates the rescue of PAPC expression by coinjecting Mespo MO with pCS–mMespo. Arrowhead in panel G indicates the disruption of segmental expression of XDelta-1. Arrowhead in panel I indicates the increase of XBra expression. Arrowhead in panel K indicates the increase of Mespo expression.
results in the opposite effect where Mespo expression is markedly reduced (56%, N = 65, Fig. 4C). We also observed that Mespo expression is decreased at stage 10 when injecting
embryos with lef1–EnR mRNA at four-cell stage (Supplemental Fig. 2D). These changes are tightly associated with the changing of Wnt/β-catenin signaling activity, because the
J. Wang et al. / Developmental Biology 304 (2007) 836–847
841
Mespo is a direct downstream target of Wnt/β-catenin signaling cascade
Fig. 3. Wnt/β-catenin signaling in the PSM. (A–J) The neurula stage embryos were stained by whole-mount in situ hybridization for Xwnt8 (A, B, arrowhead), Dkk1 (C, D, arrow), Xwnt8 and Dkk1 (E, F, arrowhead indicates XWnt8 and arrow indicates DKK1), Xwnt8 and Thy (G, H, arrowhead indicates XWn8 and arrow indicates Thy), or Mespo and Thy (I, J, arrowhead indicates Mespo and arrow indicates Thy). Embryos in panels A, C, E, G and I are oriented with anterior to the left and shown in lateral views. Embryos in panels B, D, F, H and J show posterior–dorsal view. (K) Schematic diagram (adapted from Kim et al., 2000) illustrates gene expression patterns related to somite formation in the PSM of Xenopus embryos.
expression patterns of Mespo or Paraxis are un-changed in control embryos or control side (Figs. 4A, D, E, F). These results indicate that the activity of Wnt/β-catenin signaling is responsible for Mespo expression in the PSM.
To experimentally determine how Wnt/β-catenin signaling regulates Mespo expression, we identified a 4.3 kb Mespo promoter that is able to recapitulate the endogenous Mespo expression from gastrulae to tailbud stage (Wang and Ding, 2006) and used it for further analysis. We first tested whether pCS–Lef1–VP16 could induce the expression of a luciferase reporter controlled by the 4.3 kb Mespo promoter (p-4317Luc) in animal caps. A 10-fold induction of the reporter activity is observed (Fig. 5B). To investigate whether Mespo is a direct target of Wnt/ β-catenin signaling, we examined the Mespo promoter for LEF/TCF binding sites. A perfectly conserved LEF/TCF binding site (site L) was found (van de Wetering et al., 1991), at position − 73 to − 67 relative to the transcription start. In a search of GenBank database for the 5′ regulatory regions of homologous Mespo promoters in zebrafish and chick, we found that this LEF/TCF binding site is conserved in zMespo (the zebrafish Mespo homolog) and cMespo-1 (the chick Mespo homolog) (Fig. 5A). The conservation within the regulatory regions among these vertebrates suggests that this LEF/TCF binding site plays an important role in the regulation of Mespo expression. We then generated a mutation in site L and examined the activity of the mutant promoter in animal cap assay. The results show that the mutant promoter construct (p-4317mLuc) has similar activity as the p-4317Luc (Fig. 5C). However, when coinjected with pCS–Lef1–VP16, the luciferase activity of the p-4317mLuc plasmid is significantly reduced as compared to that of the p-4317Luc plasmid (Fig. 5D). Based on the results obtained from animal cap assays, we injected reporter plasmids into the VMZ of both cells of embryos at four-cell stage to examine the function of the LEF/ TCF binding site (site L) in vivo. The results show that the luciferase activity of the p-4317mLuc plasmid is reduced to half of the p-4317Luc plasmid (Fig. 5E). The functional importance of site L was further confirmed by generating transgenic embryos with the 4.3 kb GFP construct carrying the site L mutation. When site L was mutated, GFP expression in the PSM is abolished in approximately 90% transgenic embryos (N = 200, Fig. 5H) while the expression of GFP mRNA in approximately 51% p-4317GFP transgenic embryos (N = 180, Fig. 5G) is detected in the PSM (Fig. 5I), strongly suggesting that the site L is indeed essential for Mespo expression in the PSM. LEF1 binds site L in vivo To examine whether Lef1 binds to site L, chromatin immunoprecipitation (ChIP) assay was performed as previously described (Sachs and Shi, 2000). HA-tagged Lef1 was coinjected with a truncated form of Mespo promoter, p-511, which comprised a proximal 511-bp region of the promoter, followed by precipitation of plasmid DNA bound to Lef1 with an antiHA antibody. PCR amplification using specific primers flanking site L revealed that binding of Lef1 to wild type site
842
J. Wang et al. / Developmental Biology 304 (2007) 836–847
Fig. 4. Wnt/β-catenin signaling regulates Mespo expression. For all embryos, anterior is to the left. Four-cell-stage embryos were injected in VMZ of one blastomere with lineage tracer pCS–LacZ as control (A, D), or together with pCS–Lef1–VP16 (B, E) or together with pCS–Lef1–EnR (C, F). At stage 19, whole mount in situ hybridization was performed for Mespo (A–C) or Paraxis (D–F), as indicated. Arrowhead in panel B shows the increase of Mespo expression. Arrowhead in panel C shows the decrease of Mespo expression.
L is significantly stronger than that to the mutated one (Fig. 6A). We also examined whether the binding of Lef1 to the Mespo regulatory region can be detected in vivo. Indeed, when injecting HA-tagged Lef1 into both VMZ cells at four-cell stage, the binding of Lef1 to site L is detected (Fig. 6B). As a control, no binding of Lef1 is detected in the exon region of Mespo (Fig. 6B). These results indicate that Mespo is a direct downstream target of Wnt/β-catenin signaling in the PSM. PI3-K/AKT signaling regulates Mespo expression Recent work indicates that FGF signaling is involved in controlling somitogenesis through MKK1/ERK signaling pathway by regulating Mespo expression (Delfini et al., 2005; Moreno and Kintner, 2004). However, FGF signaling has other downstream effectors like PI3-K/AKT in the PSM (Dubrulle and Pourquie, 2004). This raises a question of whether PI3-K/ AKT-mediated FGF signaling contributes to the regulation of Mespo expression. In order to answer this question, we checked whether Akt is present in the PSM as the phosphorylated form. Results from Western blot show that both Akt and its phosphorylated form are detected in the caudal PSM (Fig. 7B). To evaluate whether PI3-K/AKT signaling contributes to the expression of Mespo, we treated embryos with LY294002 which is a specific PI3 kinase inhibitor (Knight et al., 2006; Walker et al., 2000) and then dissected caudal PSM for analysis. Treating embryos with LY294002 causes a reduction of Akt phosphorylation (Fig. 7C), and leads to a decreased Mespo expression (67%, N = 55, Fig. 7E) as compared with the control (Fig. 7D). LY294002 or DMSO treatment causes no change in Paraxis expression (Figs. 7F, G). The decrease of Mespo expression was further
confirmed by RT–PCR. As shown in Fig. 7H, Mespo mRNA is dramatically downregulated when the caudal PSM cells were treated with LY294002 (Fig. 7H). These results indicate that PI3-K/AKT signaling contributes to Mespo expression. Interaction of PI3-K/AKT and Wnt signaling on Mespo expression Since Wnt/β-catenin signaling directly regulates Mespo expression and blocking PI3-K/AKT pathway decreases Mespo expression, we reasoned that Wnt/β-catenin and PI3-K/AKT pathway might cooperate to regulate Mespo expression. To test this possibility, we examined whether the activity of Wnt/β-catenin signaling was affected when PI3-K/ AKT signaling was blocked. A sharp decrease of β-catenin protein in the nuclei of the caudal PSM cells was detected after a 2-h treatment with LY294002, and a decrease in cytosolic β-catenin was also observed (Fig. 8A). To test what led to the loss of β-catenin, we took advantage of a specific antibody, which only recognizes the phosphor-form of βcatenin that is phosphorylated by GSK-3β. Western blot using this antibody shows that when treating the PSM cells with LY294002 the phosphorylated form of β-catenin is significantly increased within 1.5 h, and decreased after 2.5 h (Fig. 8B). This result indicates that the blocking of PI3-K/AKT pathway causes the phosphorylation and degradation of βcatenin via an Axin/GSK-3β-dependent mechanism (Aberle et al., 1997; Polakis, 2001). Therefore, PI3-K/AKT pathway affects the stability of β-catenin. On the other hand, when injecting embryos with pCS–Lef1–VP16 DNA to activate Wnt/β-catenin signaling, the injected side shows increased Mespo expression (79%, N = 70, Fig. 8D). However, LY294002 treatment could not cause significant reduction of
J. Wang et al. / Developmental Biology 304 (2007) 836–847
843
Fig. 5. The LEF/TCF binding site is responsible for the expression of Mespo. (A) Schematic representation of Mespo promoters from various species with a conserved LEF/TCF binding site (red). The position of potential LEF/TCF binding site (box) in Xenopus Mespo 5′-upstream sequence is indicated. (B–D) Fold induction of pCS–Lef1–VP16 in pGL3basic and p-4317Luc reporter plasmids (B), luciferase activities of p-4317Luc and p-4317mLuc reporter plasmids (C) or fold induction of pCS–Lef1–VP16 in p-4317Luc and p-4317mLuc reporter plasmids (D) in animal cap assays. Plasmids were injected into the animal pole at the four-cell stage. Animal caps were dissected at stage 8.5 and luciferase activities were measured at stage 12.5. RLU, relative light units. (E) Luciferase activities of the indicated plasmids injected into both VMZ cells at four-cell stage and measured at stage 19. RLU, relative light units. (F) Expression of endogenous Mespo mRNA at the tailbud stage. (G–I) Expression of GFP mRNA in p-4317GFP transgenic embryos (G) or in p-4317mGFP transgenic embryos (H) at the tailbud stage. The percentages of transgenic embryos that show the PSM expression (purple), non-specific expression (brown), and no detected signal (white) are shown in panel I.
the increased Mespo expression induced by injecting pCS– Lef1–VP16, while Mespo expression is clearly reduced in the uninjected side (61%, N = 90, Fig. 8E). Thus injecting pCS– Lef1–VP16 expressing the constitutively active Lef1–VP16 minimizes the effect of blocking the PI3-K/AKT pathway. Taken together, these results indicate that the PI3-K/AKT pathway participates in regulating Wnt/β-catenin signaling mediated Mespo expression, through the regulation of βcatenin stability.
Discussion Mespo is involved in segmental patterning during Xenopus somitogenesis Mespo knock out mouse shows severe defects in somitogenesis, including perturbation of segmental patterning. The mutant mice have no identifiable somites, and the expression of genes associated with somitogenesis like Mesp2 (the ortholog of
844
J. Wang et al. / Developmental Biology 304 (2007) 836–847
and found that XWnt3a expression is restricted in neural crest cells, while XWnt8 expression is detected in the caudal PSM (Figs. 3A, B, E–H). Furthermore, the Wnt antagonist Dkk1 is expressed in somitomeres (Figs. 3C–F). These observations suggested that the activity of Wnt/β-catenin signaling exists in
Fig. 6. Lef1 binds to site L in the Mespo promoter in vivo. (A) ChIP assay of embryos injected with pCS–Lef1–HA and p-511, or pCS–Lef1–HA and p511m in both VMZ cells at four-cell stage to detect the binding of Lef1 to site L. (B) ChIP assay detecting the binding of Lef1 to endogenous Mespo regulatory sequence. pCS–Lef1–HA was injected into both VMZ cells of embryos at fourcell stage.
Xenopus Thy), PAPC and Dll-3 (the ortholog of XDelta-1) is abolished in absence of Mespo (Yoon and Wold, 2000). This phenotype implies that Mespo is involved in establishing segmental patterning in somitogenesis. Here, we report that in Xenopus embryos depleting Mespo by Mespo Mo causes disorder of somite cell arrangement. During Xenopus somitogenesis, the rostral PSM cells compact to form cell aggregates named somitomeres, which later separate from the PSM to form somites. The nuclei of somite cells line up in the middle of each somite, representing the well-organized segmental pattern. There is no such distinct morphological feature in Mespodepleting Xenopus embryos. Instead, nuclei are randomly distributed, indicating the disorganization of somite cells (Fig. 1H). The expression patterns of genes associated with segmental patterning are also perturbed (Fig. 2). Thus, our data indicate that Mespo functions in establishing the segmental pattern by means of regulating the genes involved in Xenopus somitogenesis. Wnt/β-catenin signaling controls gene expressions in somitogenesis Previous studies have demonstrated that Wnt3a is involved in somitogenesis (Aulehla et al., 2003; Galceran et al., 2004; Hofmann et al., 2004; Ikeya and Takada, 2001). In chick embryos, increasing Wnt/β-catenin signaling activity by implanting beads covered with Wnt3a expressing cells induced smaller somites (Aulehla et al., 2003). In mouse embryos, it has been proposed that Wnt3a is involved in somitogenesis via Axin2 (Aulehla et al., 2003) and Wnt/β-catenin signaling controlled segmentation by directly regulating Dll1 expression (Galceran et al., 2004; Hofmann et al., 2004). However, the effect of Wnt/β-catenin signaling in somitogenesis has not been fully understood. Our study provides new information on the role of Wnt/β-catenin signaling in somitogenesis. We checked XWnt3a and XWnt8 expression in Xenopus embryos
Fig. 7. PI3-K/AKT signaling regulates Mespo expression. (A) Hatched lines indicate the three caudal regions (C, R, and S) used for western blot of Akt and phosphorylated Akt. (B) Western blot of Akt and phosphorylated Akt in different regions indicated in panel A. (C) Western blot of phosphorylated Akt in the caudal PSM equivalent to region C (illustrated in panel A) treated with either LY294002 or DMSO. (D–G) The neurula stage embryos were treated with either DMSO (D, F), or LY294002 (E, G) for 2 h, and then stained for Mespo and Paraxis as indicated. (H) RT–PCR of Mespo mRNA in the caudal PSM equivalent to region C (illustrated in panel A) treated with either DMSO or LY294002. The caudal PSM were dissected at stage 23 and treated with drugs for 2 h, then harvested for RT–PCR.
J. Wang et al. / Developmental Biology 304 (2007) 836–847
845
Fig. 8. Interaction between PI3-K/AKT and Wnt/β-catenin signaling in PSM. (A) Western blot of β-catenin in the nucleus and cytosol of the caudal PSM cells from embryos treated with either DMSO or LY294002 for 2 h. (B) Western blot of β-catenin, phosphorylated β-catenin, Akt and phosphorylated Akt in the caudal PSM cells treated with DMSO for 2.5 h, or LY294002 for 1.5 h or 2.5 h. (C–E) Four-cell-stage embryos were injected in VMZ of one cell with pCS–LacZ as control (C) or together with pCS–Lef1–VP16 (D, E). At the neurula stage, injected embryos were treated with DMSO (C, D) or LY294002 (E) for 2 h, then fixed, stained with X-gal to reveal the tracer (light blue) and analyzed by whole-mount in situ hybridization for Mespo. Arrow in panel D indicates the increase of Mespo expression in the injected side and arrowhead indicates the expression of Mespo in the uninjected side. Arrow in panel E indicates the increased Mespo expression in the injected side and arrowhead indicates the decrease of Mespo expression in the uninjected side.
Xenopus PSM and is restricted in the caudal PSM. This Wnt/ β-catenin signaling activity is essential for Mespo expression. Increasing Wnt/β-catenin signaling activity by injecting pCS– Lef1–VP16 enhances Mespo expression in the PSM, and decreasing Wnt/β-catenin signaling activity by injecting pCS– Lef1–EnR downregulates Mespo expression (Fig. 4). We also found that Wnt/β-catenin signaling, acting upstream of Mespo, directly controls Mespo expression, based on the recruitment of Lef1 to the site in the Mespo 5′-regulatory region (Fig. 6B). Since Mespo expression is essential for segmental patterning, we concluded that Wnt/β-catenin signaling controls segmentation process by regulating Mespo expression.
Regulation of Mespo expression in the PSM The bHLH family member Mespo is a key regulator of genes associated with proper segmentation and successful somitogenesis (Yoon and Wold, 2000). FGF and RA signaling were proposed to act upstream of Mespo. It has been shown that blocking FGF signaling by SU5402 or increasing RA signaling by treating with RA caused a decrease of Mespo expression (Moreno and Kintner, 2004). However, the molecular nature of how Mespo expression is regulated remains unclear. We show here that Wnt/β-catenin signaling is required for Mespo expression. Increasing Wnt/β-catenin signaling activity by overexpressing β-catenin or Lef1–VP16 increases the
846
J. Wang et al. / Developmental Biology 304 (2007) 836–847
expression of Mespo (Fig. 4B and Supplemental Fig. 2B), and interfering Wnt/β-catenin signaling by overexpressing Lef1– EnR resulted in the opposite effect (Fig. 4C and Supplemental Fig. 2D). Our results also show that the PI3-K/AKT pathway participates in regulating Mespo expression (Fig. 7E). Since it has been proposed that MKK1/ERK signaling is involved in regulating Mespo expression (Moreno and Kintner, 2004), it seems that FGF signaling contributes to the expression of Mespo through PI3-K/AKT and MKK1/ERK pathway together. Thus, multiple signaling pathways regulate expression of Mespo that is a key regulator of somitogenesis. Interaction of signals in somitogenesis Previous studies have shown that FGF, Wnt and RA signaling are all involved in somitogenesis (Aulehla et al., 2003; Dubrulle et al., 2001; Galceran et al., 2004; Hofmann et al., 2004; Vermot and Pourquie, 2005). However, whether or how these signals interact to precisely control somite formation is poorly understood. In this report, we have provided evidence that Wnt/β-catenin signaling is involved in controlling somitogenesis by interacting with FGF signaling. FGF signaling downstream effector PI3-K/AKT regulated the stability of β-catenin in the nucleus of the PSM cells (Fig. 8A). Several lines have also shown that RA and FGF signaling interact to regulate somitogenesis. RA signaling may inhibit the MAPK/ERK pathway by activating the expression of MKP3, and FGF signaling seems to regulate RA signaling activity by regulating the expression of enzymes for RA production and degradation (Moreno and Kintner, 2004). Further experiments are needed to explore whether Wnt and RA signaling interact in somitogenesis. Acknowledgments The authors are grateful to Drs. D. Sparrow, E. Joseph, C. Niehrs and J.C. Smith for sharing plasmids, Dr. Schneider (Max-Planck-Institut fur Entwicklungsbiologie) for β-catenin antibody. We thank Drs. W. Wu and Q. Tao for discussion of the project. This work was supported by the National Natural Science Foundation of China (90408005, 30270650) and the National Key Project for Basic Science Research of China (2001CB509901, 2005CB522704) to X.D. We thank Dr. Dangsheng Li for his professional polish of the MS. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ydbio.2006.12.034. References Aberle, H., Bauer, A., Stappert, J., Kispert, A., Kemler, R., 1997. beta-Catenin is a target for the ubiquitin–proteasome pathway. EMBO J. 16, 3797–3804. Aulehla, A., Wehrle, C., Brand-Saberi, B., Kemler, R., Gossler, A., Kanzler, B., Herrmann, B.G., 2003. Wnt3a plays a major role in the segmentation clock controlling somitogenesis. Dev. Cell 4, 395–406.
Conlon, F.L., Smith, J.C., 1999. Interference with brachyury function inhibits convergent extension, causes apoptosis, and reveals separate requirements in the FGF and activin signalling pathways. Dev. Biol. 213, 85–100. Delfini, M.C., Dubrulle, J., Malapert, P., Chal, J., Pourquie, O., 2005. Control of the segmentation process by graded MAPK/ERK activation in the chick embryo. Proc. Natl. Acad. Sci. U. S. A. 102, 11343–11348. Ding, X., Hausen, P., Steinbeisser, H., 1998. Pre-MBT patterning of early gene regulation in Xenopus: the role of the cortical rotation and mesoderm induction. Mech. Dev. 70, 15–24. Dubrulle, J., Pourquie, O., 2004. fgf8 mRNA decay establishes a gradient that couples axial elongation to patterning in the vertebrate embryo. Nature 427, 419–422. Dubrulle, J., McGrew, M.J., Pourquie, O., 2001. FGF signaling controls somite boundary position and regulates segmentation clock control of spatiotemporal Hox gene activation. Cell 106, 219–232. Galceran, J., Farinas, I., Depew, M.J., Clevers, H., Grosschedl, R., 1999. Wnt3a−/−-like phenotype and limb deficiency in Lef1(−/−)Tcf1(−/−) mice. Genes Dev. 13, 709–717. Galceran, J., Sustmann, C., Hsu, S.C., Folberth, S., Grosschedl, R., 2004. LEF1mediated regulation of Delta-like1 links Wnt and Notch signaling in somitogenesis. Genes Dev. 18, 2718–2723. Geng, X., Xiao, L., Lin, G.F., Hu, R., Wang, J.H., Rupp, R.A., Ding, X., 2003. Lef/Tcf-dependent Wnt/beta-catenin signaling during Xenopus axis specification. FEBS Lett. 547, 1–6. Glinka, A., Wu, W., Delius, H., Monaghan, A.P., Blumenstock, C., Niehrs, C., 1998. Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391, 357–362. Haq, S., Michael, A., Andreucci, M., Bhattacharya, K., Dotto, P., Walters, B., Woodgett, J., Kilter, H., Force, T., 2003. Stabilization of beta-catenin by a Wnt-independent mechanism regulates cardiomyocyte growth. Proc. Natl. Acad. Sci. U. S. A. 100, 4610–4615. Harland, R.M., 1991. In situ hybridization: an improved whole-mount method for Xenopus embryos. Methods Cell Biol. 36, 685–695. He, Z., Li, J., Zhen, C., Feng, L., Ding, X., 2005. Knockdown of p53 by RNAi in ES cells facilitates RA-induced differentiation into muscle cells. Biochem. Biophys. Res. Commun. 335, 676–683. Hofmann, M., Schuster-Gossler, K., Watabe-Rudolph, M., Aulehla, A., Herrmann, B.G., Gossler, A., 2004. WNT signaling, in synergy with T/ TBX6, controls Notch signaling by regulating Dll1 expression in the presomitic mesoderm of mouse embryos. Genes Dev. 18, 2712–2717. Ikeya, M., Takada, S., 2001. Wnt-3a is required for somite specification along the anteroposterior axis of the mouse embryo and for regulation of cdx-1 expression. Mech. Dev. 103, 27–33. Joseph, E.M., Cassetta, L.A., 1999. Mespo: a novel basic helix–loop–helix gene expressed in the presomitic mesoderm and posterior tailbud of Xenopus embryos. Mech. Dev. 82, 191–194. Kazanskaya, O., Glinka, A., del Barco Barrantes, I., Stannek, P., Niehrs, C., Wu, W., 2004. R-Spondin2 is a secreted activator of Wnt/beta-catenin signaling and is required for Xenopus myogenesis. Dev. Cell 7, 525–534. Kim, S.H., Jen, W.C., De Robertis, E.M., Kintner, C., 2000. The protocadherin PAPC establishes segmental boundaries during somitogenesis in Xenopus embryos. Curr. Biol. 10, 821–830. Kintner, C.R., Brockes, J.P., 1985. Monoclonal antibodies to the cells of a regenerating limb. J. Embryol. Exp. Morphol. 89, 37–55. Knight, Z.A., Gonzalez, B., Feldman, M.E., Zunder, E.R., Goldenberg, D.D., Williams, O., Loewith, R., Stokoe, D., Balla, A., Toth, B., Balla, T., Weiss, W. A., Williams, R.L., Shokat, K.M., 2006. A pharmacological map of the PI3-K family defines a role for p110alpha in insulin signaling. Cell 125, 733–747. Kroll, K.L., Amaya, E., 1996. Transgenic Xenopus embryos from sperm nuclear transplantations reveal FGF signaling requirements during gastrulation. Development 122, 3173–3183. Moreno, T.A., Kintner, C., 2004. Regulation of segmental patterning by retinoic acid signaling during Xenopus somitogenesis. Dev. Cell 6, 205–218. Nieuwkoop, P.D., Faber, J., 1967. Normal Table of Xenopus laevis. North Holland, Amsterdam. Polakis, P., 2001. More than one way to skin a catenin. Cell 105, 563–566. Pourquie, O., 2001. Vertebrate somitogenesis. Annu. Rev. Cell Dev. Biol. 17, 311–350.
J. Wang et al. / Developmental Biology 304 (2007) 836–847 Pourquie, O., 2003. The segmentation clock: converting embryonic time into spatial pattern. Science 301, 328–330. Sachs, L.M., Shi, Y.B., 2000. Targeted chromatin binding and histone acetylation in vivo by thyroid hormone receptor during amphibian development. Proc. Natl. Acad. Sci. U. S. A. 97, 13138–13143. Smith, J.C., Price, B.M., Green, J.B., Weigel, D., Herrmann, B.G., 1991. Expression of a Xenopus homolog of brachyury (T) is an immediate-early response to mesoderm induction. Cell 67, 79–87. Sparrow, D.B., Jen, W.C., Kotecha, S., Towers, N., Kintner, C., Mohun, T.J., 1998. Thylacine 1 is expressed segmentally within the paraxial mesoderm of the Xenopus embryo and interacts with the Notch pathway. Development 125, 2041–2051. Takada, S., Stark, K.L., Shea, M.J., Vassileva, G., McMahon, J.A., McMahon, A.P., 1994. Wnt-3a regulates somite and tailbud formation in the mouse embryo. Genes Dev. 8, 174–189. van de Wetering, M., Oosterwegel, M., Dooijes, D., Clevers, H., 1991. Identification and cloning of TCF-1, a T lymphocyte-specific transcription factor containing a sequence-specific HMG box. EMBO J. 10, 123–132. Vermot, J., Pourquie, O., 2005. Retinoic acid coordinates somitogenesis and left–right patterning in vertebrate embryos. Nature 435, 215–220.
847
Walker, E.H., Pacold, M.E., Perisic, O., Stephens, L., Hawkins, P.T., Wymann, M.P., Williams, R.L., 2000. Structural determinants of phosphoinositide 3-kinase inhibition by wortmannin, LY294002, quercetin, myricetin, and staurosporine. Mol. Cell 6, 909–919. Wang, J., Ding, X., 2006. Cloning and analyzing of Xenopus Mespo promoter in retinoic acid regulated Mespo expression. Acta Biochim. Biophys. Sin. 38 (11), 759–764 (Shanghai). Weisblat, D.A., Wedeen, C.J., Kostriken, R.G., 1994. Evolution of developmental mechanisms: spatial and temporal modes of rostrocaudal patterning. Curr. Top. Dev. Biol. 29, 101–134. Wilson, V., Rashbass, P., Beddington, R.S., 1993. Chimeric analysis of T (Brachyury) gene function. Development 117, 1321–1331. Yamaguchi, T.P., Takada, S., Yoshikawa, Y., Wu, N., McMahon, A.P., 1999. T (Brachyury) is a direct target of Wnt3a during paraxial mesoderm specification. Genes Dev. 13, 3185–3190. Yoon, J.K., Wold, B., 2000. The bHLH regulator pMesogenin1 is required for maturation and segmentation of paraxial mesoderm. Genes Dev. 14, 3204–3214. Yoon, J.K., Moon, R.T., Wold, B., 2000. The bHLH class protein pMesogenin1 can specify paraxial mesoderm phenotypes. Dev. Biol. 222, 376–391.