Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Developmental Biology 288 (2005) 113 – 125 www.elsevier.com/locate/ydbio Bmp signaling promotes intermediate mesoderm gene expression in a dose-dependent, cell-autonomous and translation-dependent manner Richard G. James, Thomas M. Schultheiss * Molecular and Vascular Medicine Beth Israel Deaconess Medical Center, Boston, MA 02215, USA Harvard Medical School, Boston, MA 02215, USA Received for publication 16 June 2005, revised 8 September 2005, accepted 8 September 2005 Available online 21 October 2005 Abstract The intermediate mesoderm lies between the somites and the lateral plate and is the source of all kidney tissue in the developing vertebrate embryo. While bone morphogenetic protein (Bmp) signaling is known to regulate mesodermal cell type determination along the medio-lateral axis, its role in intermediate mesoderm formation has not been well characterized. The current study finds that low and high levels of Bmp ligand are both necessary and sufficient to activate intermediate and lateral mesodermal gene expression, respectively, both in vivo and in vitro. Dosedependent activation of intermediate and lateral mesodermal genes by Bmp signaling is cell-autonomous, as demonstrated by electroporation of the avian embryo with constitutively active Bmp receptors driven by promoters of varying strengths. In explant cultures, Bmp activation of Oddskipped related 1 (Odd-1), the earliest known gene expressed in the intermediate mesoderm, is blocked by cyclohexamide, indicating that the activation of Odd-1 by Bmp signaling is translation-dependent. The data from this study are integrated with that of other studies to generate a model for the role of Bmp signaling in trunk mesodermal patterning in which low levels of Bmp activate intermediate mesoderm gene expression by inhibition of repressors present in medial mesoderm, whereas high levels of Bmp repress both medial and intermediate mesoderm gene expression and activate lateral plate genes. D 2005 Elsevier Inc. All rights reserved. Keywords: Kidney; Chick embryo; Bmp; Embryonic patterning; Mesoderm Introduction Secreted growth factors of the bone morphogenic protein (Bmp) family establish concentration gradients that contribute to patterning along the embryonic medio-lateral axis (in other embryos, such as Drosophila and Xenopus, this axis is ‘‘dorsoventral’’). In several contexts in both Drosophila (Holley and Ferguson, 1997; Podos and Ferguson, 1999; Rusch and Levine, 1996) and vertebrates (Harland and Gerhart, 1997; Hogan, 1996; Niehrs et al., 2000), the level of Bmp signaling along the medio-lateral axis correlates with cell fate. In vertebrate, mesoderm high levels of Bmp signal have been found to promote formation of lateral structures such as blood; intermediate levels to promote intermediate structures such as * Corresponding author. Molecular and Vascular Medicine Beth Israel Deaconess Medical Center, 330 Brookline Avenue, RW-663, Boston, MA 02215, USA. Fax: +1 617 667 2913. E-mail address: [email protected] (T.M. Schultheiss). 0012-1606/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2005.09.025 kidney; and lower levels are associated with muscle and notochord development (Dosch et al., 1997; Jones et al., 1996). Two specific mechanisms have been proposed to explain dose-dependent activation of target genes by Bmp. One proposed mechanism is that components of the Bmp pathway interact with regulatory regions of Bmp-responsive genes in a concentration-dependent manner. In the case of Drosophila Decapentaplegic (Dpp, the Drosophila Bmp-2/4 homolog), enhancer elements can have differential binding affinity for the Dpp transducer Mothers against dpp (Mad) (Wharton et al., 2004): high-dose responders have low affinity Mad binding sites and are transcribed only in response to high concentrations of Mad, whereas low-dose responders have high affinity Mad binding sites and are transcribed in response to both low and high concentrations of Mad. Alternatively, in the Drosophila wing and ectoderm, autonomous expression of some low-dose Bmp responders does not require Mad. Instead, in the absence of Dpp signal, the protein Brinker represses transcription of these genes (Campbell and Tomlinson, 1999; Jazwinska 114 R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 et al., 1999a; Minami et al., 1999). Low doses of Dpp facilitate Smad-dependent silencing of Brinker (Jazwinska et al., 1999a,b; Pyrowolakis et al., 2004) and indirect activation of the low-dose responders. In vertebrates, Bmp responders such as Msx-1/2, Id-1/2/3 and others can be directly activated by Bmp signaling in vitro (Hollnagel et al., 1999), and Smad binding sites in the Msx-2 enhancer are necessary for its expression (Brugger et al., 2004). However, it is unclear how differential doses of Bmp signaling activate tissue-specific genes during vertebrate mesodermal patterning in vivo. We sought to address this issue by studying the formation of the avian intermediate mesoderm (IM). The IM is a strip of tissue, located between the developing somites and lateral plate, and which is the source of all kidney tissue in the body (Sainio and Raatikainen-Ahokas, 1999). Shortly after gastrulation, the transcription factors Odd-1 (previously known as Osr-1) (So and Danielian, 1999) (Fig. 1A), Pax-2 (Dressler et al., 1990) (Fig. 1B) and Lim-1 (Fujii et al., 1994) (Fig. 1C) are expressed in the developing IM. Reports in Xenopus and zebrafish have shown that Bmp signaling is necessary for the expression of Lim-1 and Pax-2 (Kishimoto et al., 1997; Mullins et al., 1996), and the presence of nuclear localized phosphorylated Smad-1 protein in the IM demonstrates that it is an area of active Bmp signaling (Faure et al., 2002). In this report, we demonstrate that low, but not high doses of Bmp signal activate transcription of Lim-1, Pax-2 and Odd-1 in a cell-autonomous, but translation-dependent manner. Combining our findings with those of others, we propose a two-part model to explain how low-dose Bmp response genes are activated specifically in the intermediate mesoderm. First, lowdoses of Bmp signal promote transcription of Odd-1, Lim-1 and Pax-2 in the IM by inhibiting a repressive activity that is present in somitic mesoderm. Second, high doses of Bmp signal activate additional inhibitors in the lateral plate, which restrict the expression of Odd-1, Pax-2 and Lim-1 to the intermediate mesoderm. Materials and methods Cloning of chick Odd-1 A portion of the mouse Odd-skipped related 1 gene (mOdd-1) (So and Danielian, 1999) corresponding to the coding part of the three zinc finger motifs was used to probe 5 105 plaques from an HH stage 11 – 14 chick embryo lambda ZapII Phage library (Nieto et al., 1994). Three plaques were identified, one of which contained a full-length coding sequence for a gene 81% identical to mOdd-1 at the amino acid level. This clone, which is named chick Odd-1 (cOdd-1), will be described in detail in a separate publication (R.G.J. and T.M.S., in preparation). Fig. 1. Exogenous Bmp protein can induce IM gene expression in the paraxial mesoderm. (A – C) Expression of Odd-1 (A), Pax-2 (B) and Lim-1 (C) in stages 10 – 11 control embryos. (D – F) Effects of a heparin-acrylic bead soaked in recombinant human Bmp-2 (b) placed in the embryo at stages 5 – 6 and cultured until stages 10 – 11. A control bead (c) was placed on the opposite side. On the Bmp-treated side, expressions of Odd-1 (D), Pax-2 (E) and Lim-1 (F) are all moved towards the midline (arrows, D, E, F, G, K), into the region where paraxial genes are normally expressed. Note that in some regions expression of IM genes is also decreased on the treated side. (G – I) Bmp-2 affects the identity of paraxial mesoderm cells and not their migration. After labeling prospective paraxial mesoderm with DiI at HH stage 5, when it resided in the primitive streak (H), a Bmp-2 bead was placed in lateral plate, and embryos were cultured through stage 10 (I) when they were processed for expression of Odd-1 by in situ hybridization (G). Comparison of treated and control sides at stage 10 reveals that Odd-1 was expressed more medially on the Bmp-treated side (G), while migration patterns of the prospective paraxial mesoderm were unchanged (H, I). (J – K) Sections of Bmp-treated embryos. In Bmp-treated embryos, lateral plate (J, Cytokeratin) and intermediate mesoderm (K, Lim-1) gene expression moved towards the midline. np, neural plate; s, somite. Expression plasmids In situ hybridization Constitutively active Alk3 and Alk6 (caALK3 and caAlk6) were obtained from L. Niswander (Zou et al., 1997) and subcloned into pCIG (Faure et al., 2002), which drives gene expression from a combination of chick beta actin enhancer and CMV promoter and which contains an Internal Ribosomal Entry Sequence (IRES) driving nuclear Green Fluorescence Protein. They were also subcloned into pCS2+, which drives gene expression from a CMV promoter/enhancer. pCAGGSdsRed was obtained from C. Cepko (Matsuda and Cepko, 2004). RNA probes were generated for chicken Odd-1 (this manuscript), Pax-2 (Burrill et al., 1997; Herbrand et al., 1998), Lim-1 (Tsuchida et al., 1994), Paraxis (Barnes et al., 1997), Tbx-6L (Knezevic et al., 1997) and cytokeratin (Tonegawa et al., 1997) using standard methods, as previously described (Schultheiss et al., 1995). Whole mount in situ hybridization was performed as previously described (Schultheiss et al., 1995). R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 Immunohistochemistry A peptide consisting of the predicted amino acids 125 – 139 of the chicken ODD-1 protein was synthesized, linked to KLH (Pierce) and injected into guinea pigs (Covance). Bleeds were tested by immunofluorescence microscopy for their ability to detect a nuclear signal specifically in cells where Odd-1 message is expressed (in the IM and medial part of the lateral plate). High titer bleeds were affinity-purified against the immunizing peptide using a Sulfolink Kit (Pierce). The following primary antibodies were used: guinea pig – anti-Odd-1 (1:750), mouse – anti-Lim-1/2 (1:10) (Developmental Studies Hybridoma Bank), rabbit – anti-Pax-2 (1:250) BAbCo, mouse – anti-Pax7 (1:10) DSHB, rabbit – anti-Scl/Tal (1:50) (Drake et al., 1997), anti-Quail nuclear antigen QCPN (1:20) DSHB and mouse and rabbit anti-GFP (Molecular Probes). Affinity-purified secondary antibodies were purchased from Jackson Immunoresearch and used at 1:250. DAPI (Sigma) was included in the penultimate wash at 1 Ag/ml to label DNA. Embryo culture and electroporation Embryos were grown in modified New culture as previously described (James and Schultheiss, 2003). Heparin acrylamide beads were soaked in 33 ng/Al human recombinant Bmp-2 (R&D) and implanted into embryos as previously described (Schultheiss et al., 1997). Noggin-expressing fibroblasts (Smith and Harland, 1992) were cultured as previously described (Schultheiss et al., 1997), aggregated by brief centrifugation followed by 1 h at 37- in a humidified incubator, following which pellets of aggregated cells were cut and placed in the embryo using a sharpened tungsten microneedle. Electroporation was carried out as previously described (Wilm et al., 2004). In some embryos, DiI was injected into the primitive streak at the same time as electroporation, as previously described (James and Schultheiss, 2003). For some experiments, electroporation was performed in quail embryos, using similar methods, and electroporated portions of the primitive streak were transplanted into chicken host embryos, as previously described (James and Schultheiss, 2003). Explant culture Regions of the primitive streak or mesoderm were dissected and cultured in collagen gels as previously described (James and Schultheiss, 2003). In some cases, recombinant Bmp-2 or noggin (R&D Systems) was added to the medium. To block protein synthesis, cyclohexamide (Sigma) was added to the medium at a concentration of 10 Ag/ml. At this concentration, greater than 95% of protein synthesis is blocked as measured by 35S-methionine incorporation into proteins (T.M.S., unpublished data). Explants were processed for in situ hybridization in the collagen gels, as previously described (James and Schultheiss, 2003). 115 al., 2002). We hypothesize that Bmp signaling promotes IM gene expression as this tissue differentiates. This paper is an investigation of that general hypothesis and the specific mechanism by which Bmp signaling activates dose-responsive gene expression in the avian IM. Exogenous Bmp-2 promotes lateral and intermediate mesoderm fates at the expense of paraxial fates To gain initial insight into the in vivo role of Bmp signaling on IM patterning, we investigated the effects of manipulating Bmp levels during the time period when IM was being determined. Bmp-2-soaked heparin-coated beads were implanted into the lateral plate of HH4 – 8 embryos, cultured for 15– 20 h and then assayed for IM gene expression. We observed that the lateral plate was expanded (see Cytokeratin in Fig. 1J) while somites were reduced in size or absent (arrows, Figs. 1D –F, G, K). The IM was shifted medially and was often compressed (Odd-1: Figs. 1A, D; Pax-2: Figs. 1B, E; and Lim-1: Figs. 1C, F). Note that the medial shift of IM markers is actually comprised of two effects: repression of IM genes in the normal IM position accompanied by ectopic expression of IM genes in a portion of the paraxial region. The ectopic expression of IM genes in more medial positions could be explained by two mechanisms: (1) re-patterning of the trunk resulting in a compressed somite/IM compartment or (2) death of somite cells and migration of the endogenous IM into a paraxial position. To differentiate between these two possibilities, the migration of primitive streak cells was mapped in the presence of Bmp-2 beads (Figs. 1G – I). Lipophilic fluorescent dye was injected into the primitive streak of stages 4– 8 embryos after implantation of the beads and was tracked for 15 – 20 h. In all cases (10/10), the migration pattern of the labeled cells was normal and consistent with previous fate maps (Garcia-Martinez and Schoenwolf, 1992; James and Schultheiss, 2003; Psychoyos and Stern, 1996), while expression of Odd-1 was shifted medially (Fig. 1G). This result implies that the somite and IM compartments are being re-patterned and compressed upon treatment with exogenous Bmp-2. Results Exogenous Bmp-2 promotes dose-dependent activation of IM gene expression in somitic explants The precursors of the anterior IM migrate through the primitive streak starting at Hamburger Hamilton stage (HH) 4 (Hamburger and Hamilton, 1951) and reach their final destination lateral to the somites at HH9 (Garcia-Martinez and Schoenwolf, 1992; James and Schultheiss, 2003; Psychoyos and Stern, 1996). Previously, we have shown that during this time period the fates of neither IM nor other trunk mesodermal tissues are determined and that co-culture of somitic mesoderm with lateral plate causes activation of Lim-1 and Pax-2 within the somitic tissue (James and Schultheiss, 2003). Several Bmp family members are expressed in the lateral plate (Schultheiss et al., 1997), and phosphorylated Smad1, an indicator of active Bmp signaling, is detectable in IM precursor cells between HH5 and HH8 but not in somites or their precursors (Faure et The previous results suggest that Bmp-2 can re-pattern the somite region of the embryo and cause it to express IM markers. This effect could be the result of the action of Bmp itself or of interaction between Bmp-2 and signals derived from other tissues in the embryo such as Hensen’s node, notochord or neural tube. Additionally, the previous experiments did not examine the dose dependence of the Bmp response. In order to investigate whether specific levels of Bmp signaling can repattern somite tissue to IM in the absence of embryonic signaling centers, paraxial mesoderm or its precursors were dissected from chick embryos and cultured in serum-free medium containing Bmp-2 at concentrations ranging from 10 7 to 5 10 7 g/ml. Explants were cultured for 15– 20 h and then processed for in situ hybridization. 116 R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 In the absence of Bmp-2, explant cultures of HH5 anterior primitive streak express the somite marker Paraxis (Fig. 2A, top panel), but not the lateral plate marker Cytokeratin (Fig. 2D, top panel; the small foci of expression seen in these explants are in the ectoderm included in the explant) or the IM markers Odd-1 (Fig. 2B, top panel) or Lim-1 (Fig. 2E, top panel). Paraxis is downregulated (Fig. 2A, middle and bottom panels) and Cytokeratin is upregulated (Fig. 2D, middle and bottom panels) in anterior primitive streak cultured with low or high concentrations of Bmp-2. Intriguingly, Odd-1 (Fig. 2B, compare middle and bottom panels) and Lim-1 (Fig. 2E, compare middle and bottom panels) are expressed more efficiently in anterior primitive streaks cultured with low concentrations of Bmp-2. These results indicate that the anterior primitive streak is malleable with respect to its response to Bmp signaling and that IM genes are preferentially activated by low concentrations of Bmp. Note that the normal pattern of Odd-1 expression is broader than that of Lim-1 (compare Figs. 1A, C). Consistent with this, Odd-1 is capable of being expressed in a broader range of Bmp concentrations, as evidenced by the expression of Odd-1 in some explants treated with high doses of Bmp (Fig. 2B, bottom panel). Similar effects were seen when older HH8 paraxial mesoderm was treated with Bmp-2 (Figs. 2C, F), indicating that low doses of Bmp-2 can re-pattern the somite to express IM markers several stages after the IM is initially specified. By HH10, somites do not express IM markers in response to Bmp2 treatment (data not shown), indicating that they are determined with respect to Bmp signals at this time. Expression of high levels of activated Alk3 and Alk6 inhibits differentiation of somite and promotes lateral plate fates The previous experiments demonstrate that Bmp signaling of low and high levels is sufficient to promote IM and lateral plate formation, respectively, in somitic tissue. One important Fig. 2. Dose-dependent activation of intermediate mesoderm genes by Bmp-2 in explant culture. Treatment of anterior primitive streak with recombinant human Bmp-2 resulted in downregulation of the paraxial marker Paraxis (A) and upregulation of the IM markers Odd-1 (B) and Lim-1 (E) and the lateral plate marker cytokeratin (D). The IM markers Odd-1 and Lim-1 were more strongly activated at lower doses of Bmp-2 (B, E), while cytokeratin was also activated strongly at higher doses (D). Stage 8 somite tissue shows the same dose-dependent activation of IM genes in response to Bmp-2 (C, F). Doses of Bmp-2 are given in g/ml. R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 question that has not been addressed in this or earlier research is whether the Bmp signals that pattern the IM are received and translated by individual cells. To begin to address this question, plasmids were constructed that express constitutively active forms of the Bmp-2/4 type I receptors Alk3 and Alk6 (caAlk3 and caAlk6; also called caBmpRIa and caBmpRIb) (Zou et al., 1997) under the control of a strong avian promoter, CAGGS (Niwa et al., 1991, see also Supplementary Fig. 1). Both plasmids contain an IRESNuclear-GFP cassette so that GFP can be visualized in the nucleus of every cell that expresses the activated Alk gene. Chick embryos were transfected by electroporation into the primitive streak at HH stages 4 – 6 and assayed for gene expression after culture for 15 –20 h. Embryos in which most of the trunk cells were transformed lost somite morphology and contained patches of paraxial cells which did not express Paraxis (Fig. 3A, section Fig. 3G) or the presomitic mesoderm marker Tbx-6L (Fig. 3B, section Fig. 3H). Similar to what was observed in the Bmp bead studies, the lateral plate marker Cytokeratin is upregulated in electroporated embryos (arrows, Fig. 3C). In sections, Cytokeratin expression can be seen adjacent to the notochord in clumps of cells that are not associated with the other somite cells (Figs. 3F, I). Similar results were obtained with both caAlk3 and caAlk6 constructs, while these effects were not seen with control empty CAGGS electroporations. 117 In order to characterize the effects of high levels of activated Alk3/6 on trunk gene expression in individual cells, the electroporation experiments were analyzed by immunofluorescence. Several insights emerged from these studies. First, GFP cells never co-stained with the somite marker Pax-7, even if they were isolated within an aggregate of cells that looked superficially like a somite (‘‘x’’ in Figs. 4A, D, G, J). Second, GFP-positive cells were rarely, if at all, co-stained with IM markers. If a GFP-positive cell was isolated in the IM (arrowhead, Fig. 4D, G, and insets), it did not express Pax-2 (arrowhead, Fig. 4E and inset) or Odd-1 (arrowhead, Fig. 4H and inset). On occasion, ectopic Pax-2 (star, Fig. 4E) and Odd-1 (star, Fig. 4H)-positive cells were found in or near the somite. In most of these cells, GFP could not be detected, while in others GFP expression was barely detectable (see star, Figs. 4D,G). Either these cells express low levels of activated Alk3 or they are induced to express Pax-2 and Odd-1 in a non-cell-autonomous manner (see Discussion). Lastly, GFP cells often formed clumps adjacent to the notochord (‘‘x’’ in Figs. 4A, D, G, J). Previous studies have shown that Bmp-4 promotes vasculogenesis in paraxial mesoderm (Nimmagadda et al., 2005) and that, in the absence of Noggin, putative vascular cells inappropriately migrate to the midline and form accumulations (Reese et al., 2004), similar to what we observed in cells expressing activated Alk3. In order to test whether the GFPpositive cells that were observed near the midline are Fig. 3. Effects on mesodermal gene expression of constitutively active Alk3 (BmpRIa) expressed under control of a strong promoter. Images of embryos electroporated with pCAGGS-caAlk3-IRES-GFP (A – C, G – I) or control embryos (D – F) stained by in situ hybridization for Paraxis (A, D, G), Tbx-6L (B, E, H) or Cytokeratin (C, F, I). In regions where many cells were electroporated, somite morphology is interrupted (arrows in A, B, G, H, I). caAlk3 leads to downregulation of the paraxial markers Paraxis (A, D, G) and Tbx-6 (B, E, H) and ectopic expression of cytokeratin in paraxial mesoderm (C, F, I). Arrows in panels G – I indicate affected areas. N, notochord; np, neural plate; s, somite. 118 R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 Fig. 4. Examination of the effects of high level expression of caAlk3 at single cell resolution. Sections of embryos electroporated with pCAGGS-caAlk3 (A, B, D, E, G, H, J, K) or control embryos (C, F, I, L). pCAGGS-caAlk3 generates clumps of cells which accumulate adjacent to the notochord (x in A, D, G, J). These clumps express the blood/vascular marker Scl/Tal (J, K) but do not express Pax-7 (A, B), Pax-2 (D, E) or Odd-1 (G, H). caAlk3-expressing cells (as indicated by GFP expression) are in general not seen in well-formed somites, although they can be seen in poorly formed paraxial regions (D). GFP-expressing cells in the endogenous IM typically do not express Odd-1 or Pax-2 (arrows in D, E and G, H, higher magnification shown in inset). A small number of ectopic Pax-2 and Odd-1-expressing cells are seen (asterisks in E, H), which typically do not express detectable levels of GFP (D, G). im, intermediate mesoderm; lp, lateral plate; n, notochord; np, neural plate; s, somite. The vertical line marks the border between the somite and the intermediate mesoderm. vasculogenic, electroporated embryos were stained for the vascular/hematopoietic marker SCL/TAL (Drake et al., 1997). Strikingly, more than 50% (counted in 11/11 embryos) of all GFP-expressing cells located in the somitic or IM regions expressed SCL/TAL (arrowhead, Figs. 4J, K). All ectopic SCL/ TAL cells observed in the paraxial region also express GFP. In summary, the most significant effect of expressing high levels of activated Alk-3/6 constructs was to convert somite and IM cells into lateral plate and vascular/hematopoietic cells. Expression of low levels of activated Alk3 in somites promotes an IM fate cell-autonomously In order to examine the cell-autonomous effects of low levels of Bmp signaling, we generated plasmids that express activated Alk3 or Alk6 driven off the CS2 promoter (CS2caAlk3 or CS2-caAlk6) and transformed them into chicken embryos. The CS2 promoter expresses activated Alk constructs at much lower levels than CAGGS in chicken embryos (Zou et al., 1997, see Supplementary Fig. 1). Unlike the effect seen with CAGGS-caAlk3/6, electroporation of CS2-caAlk3 or CS2-caAlk6 did not noticeably affect the morphology of the embryos. Upon sectioning and staining for Pax-7, most of the GFP-expressing cells in the somite were Pax-7-positive (Figs. 5A, B). Ectopic Odd-1 expression was widespread in these embryos in GFP-positive cells (arrowhead Figs. 5E, F), and it is likely that Odd-1 and Pax-7 are coexpressed in many somitic cells. Pax-2 expression was observed in the somites of every embryo that was electroporated with CS2-caAlk3 (11/11), although Pax-2 was activated in a much lower percentage of GFP-expressing cells than Odd-1. Ectopic Pax-2-positive cells were all GFP-positive and occasionally appeared to be migrating dorsally with a morphology resembling the nephric duct (arrowhead, Figs. 5C, D). Lastly, SCL/TAL expression was occasionally seen in some extremely bright GFP somitic cells (arrowhead, Figs. 5G, H), consistent with it requiring high levels of Bmp signal to be expressed. These data demonstrate that low levels of Bmp signaling are sufficient to induce IM gene expression in a cell-autonomous manner, in at least a subset of paraxial cells. R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 Fig. 5. Effects of low-level expression of caAlk3 on mesodermal gene expression. Alk3, expressed under the CMV promoter, was electroporated into chick embryos together with a GFP reporter plasmid. Expression of caAlk3 did not inhibit paraxial Pax-7 expression (A, B). Many caAlk3-expressing cells in the somite expressed ectopic Odd-1 (E, F), and some expressed ectopic Pax-2 (C, D arrow). The ectopic Pax-2-positive cells tended to express GFP at higher levels and were often seen appearing to be separating from the somite dorsally. Most GFP-expressing cells in the somite did not express Scl/Tal, although some cells outside of the somite coexpressed GFP and Scl/Tal (G, H, arrow). Abbreviations as in Fig. 4. Translation is required for Bmp-dependent activation of Odd-1 The above findings indicate that Bmp signals are capable of inhibiting somite differentiation and promoting expression of Odd-1, Lim-1 and Pax-2 (Figs. 1 –5) in paraxial tissue. This effect is cell-autonomous (Figs. 4, 5) and dosedependent (Figs. 2, 5). To further examine the response of mesodermal genes to Bmp, the exact time course of transcription factor response to Bmp-2 was detailed. Presomitic mesoderm from Hamburger Hamilton stage 8 embryos was cultured in media alone or in media supplemented with Bmp-2 and harvested after 1.5, 3 or 5 h. We found that expression of the somite markers Paraxis (Fig. 6A, compare 1st and 2nd panel) and Foxc2 (data not shown) were not decreased in response to Bmp-2 within 5 h of culture (contrast with Fig. 2A, bottom panel). In contrast, Tbx-6L and Odd-1 responded quickly: after 3 h of culture with Bmp-2, 119 Tbx-6 expression decreased (Fig. 6C, compare 1st and 2nd panels) and Odd-1 transcription was activated (Fig. 6D, compare 1st and 2nd panel). Interestingly, Odd-1 was also upregulated in explants treated with high concentrations of Bmp-2 starting at 1.5 h (Fig. 6B, 1st panel), and expression was maintained for at least 5 h in culture (Fig. 6B, 2nd panel). In contrast, Odd-1 expression was not detectable after 15 h of culture with high concentrations of Bmp-2 (Fig. 2C, bottom panel). Combined, these results suggest that Bmp-2 signals activate expression of Odd-1 in somitic tissue, but at high concentrations they also activate a slower acting repressor of Odd-1 transcription (see Discussion). To test whether the modifications of Tbx-6L or Odd-1 transcription due to Bmp-2 were direct responses to activation of the signaling pathway, we tested if the changes still occurred in the presence of the translation inhibitor cyclohexamide. In each case, presomitic mesoderm was cultured for 3 h under four conditions: (1) in media alone, (2) in media containing 5 10 7 g/ml Bmp-2, (3) in media containing cyclohexamide and (4) in media containing 5 10 7 g/ml Bmp-2 and cyclohexamide. First, we found that the repression of Tbx-6L expression mediated by Bmp-2 was translation-dependent. Although Bmp-2 alone potently downregulates Tbx-6L (Fig. 6C, compare 1st and 2nd panels), it did not do so in the presence of cyclohexamide (Fig. 6C, compare 3rd and 4th panels), indicating that the Tbx-6L downregulation is likely indirect. Interestingly, Odd-1 expression was somewhat upregulated in explants cultured in cyclohexamide alone (Fig. 6D, compare 1st and 3rd panels). This observation suggests that Odd-1 transcription may be repressed by a factor present in the developing somite that requires active translation. Expression of Odd-1 in explants treated with both Bmp-2 and cyclohexamide was not appreciably greater than in explants treated with cyclohexamide alone (Fig. 6D, compare 3rd and 4th panels), implying that Bmp-mediated activation of Odd-1 expression is also translation-dependent. As a positive control for the cyclohexamide experiments, we also assayed explants for expression of the direct Bmp effectors Msx-1 and Msx-2 (Hollnagel et al., 1999). As expected, the expression of both was upregulated in the presence of Bmp-2 with or without cyclohexamide (Msx-2 is shown in Fig. 6E). Thus, transcription of the earliest known gene to be expressed in the IM – Odd1– does not appear to be activated directly by transducers of the Bmp signaling pathway. Bmp signals are required during IM specification The previous set of experiments examined the ability of Bmp signaling to regulate IM gene expression. In order to investigate whether Bmp signaling is required for IM differentiation, loss of function experiments were performed in which explanted tissue was cultured in the presence of the extracellular Bmp inhibitor Noggin. We first examined whether Noggin could re-pattern the posterior primitive streak, which is a zone of high Bmp signaling that normally gives rise to lateral plate tissue (data not shown). When grown in the highest 120 R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 Fig. 6. Does regulation of IM and somite genes by Bmp require protein synthesis? Paraxial tissue from stage 8 embryos was cultured either alone or in the presence of 10 7 g/ml (A) or 5 10 7 g/ml (B – E) of Bmp-2. (A) Paraxial explants did not downregulate Paraxis after 5 h of exposure to Bmp-2 (compare with Fig. 2A, where explants were cultured at the same dose for 16 h). (B) Explants upregulated Odd-1 within 1.5 h after being exposed to Bmp-2. (C – E). Paraxial explants were cultured for 3 h either untreated (top panels), in the presence of Bmp-2 (second panels), in the presence of cyclohexamide (third panels) or in the presence of both Bmp-2 and cyclohexamide (bottom panels). In the presence of cyclohexamide, Bmp does not upregulate Odd-1 beyond the mild upregulation seen with cyclohexamide alone (D), indicating that Odd-1 is not an immediate response gene to Bmp signaling. Bmp-induced repression of Tbx-6L expression is blocked by cyclohexamide (C), indicating that downregulation of Tbx-6L expression is also not an immediate response gene to Bmp signaling. As a positive control, Msx-2 was activated by Bmp-2 even in the presence of cyclohexamide (E), indicating that it is an immediate response gene to Bmp signaling. concentration of Noggin, posterior streak explants upregulated expression of Paraxis (Fig. 7A, 3/3) and lost expression of Cytokeratin (Fig. 7B, 0/5). Weak expression of Lim-1 was observed at intermediate (Fig. 7C, middle panel, 3/4), but not at the highest concentration (Fig. 7C, bottom panel, 0/4) of Noggin. These results are consistent with a model in which low levels of Bmp signaling promote IM gene expression. To test whether Bmp is necessary for IM gene expression after stage 5, middle primitive streak, which is specified to express IM factors (Figs. 7D, E, top panel), was explanted and cultured in Noggin-containing medium. Lim-1 was potently downregulated by Noggin at 10 6 g/ml (Fig. 7E, bottom panel, 0/6) and mildly downregulated at 10 7 g/ml (Fig. 7E, middle panel, 3/5) relative to control explants (Fig. 7E, top panel, 6/6). In contrast, Odd-1 expression in developing middle primitive streak explants was unaffected by Noggin concentration (Fig. 7D), which implies that Bmp signals are not required for Odd-1 expression after HH5. HH8 prospective IM explanted in media containing Noggin maintained expression of Lim-1 even at the highest Noggin concentration (Fig. 7F). These results imply that Odd-1 expression does not require Bmp signals after HH5, whereas Lim-1 expression requires Bmp signaling between HH5 and HH8. To investigate whether Bmp signaling is necessary in vivo for IM formation, pelleted rat fibroblast cells stably transfected with Noggin cDNA (Smith et al., 1993) were implanted into the presumptive lateral plate or IM of HH4 – 7 embryos. These were grown for 15– 20 h and analyzed for morphological changes and modification of kidney gene expression. When noggin-secreting cells were implanted prior to HH6, ectopic somites were often formed adjacent to the endogenous somitic mesoderm (data not shown), consistent with previous results (Tonegawa et al., 1997). Two different results were obtained upon visualization of IM markers. First, the domain of Odd-1 (Fig. 7G) was expanded into the lateral plate mesoderm near the graft (star, Fig. 7G) but inhibited in tissue immediately adjacent to the graft. This result is consistent with the hypothesis that Odd-1 is activated in response to low doses of Bmp signaling. In contrast, Pax-2 (Fig. 7H) and Lim-1 (Fig. 7I) gene expression was inhibited in response to the noggin grafts. The observed inhibition occurs both near the graft (carat, Figs. 7H, I) and at some distance away (star, Figs. 7H, I). Neither Pax-2 nor Lim-1 was expressed in secondary locations in the lateral plate as would be expected if their expression was simply responding to Bmp levels. Coupled with the fact that Pax-2 is activated at a much lower rate than Odd-1 by CS2-caAlk3/6 (Fig. 5), this finding suggests that, within the context of the embryo in vivo, expression of Pax-2 and Lim-1 may require other signaling in addition to that from Bmps. Discussion Bmp signaling promotes IM gene expression in a dose-dependent and cell-autonomous manner Genetic manipulation of Bmp pathway components in Xenopus, zebrafish and mouse embryos has demonstrated that R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 121 Fig. 7. Effects of Bmp inhibition on mesodermal gene expression. Explants from the indicated region of the primitive streak (A – E) or mesoderm (F) were cultured for 16 h in the presence of the indicated concentration of the Bmp antagonist noggin (g/ml). Posterior primitive streak activates Paraxis expression and represses Cytokeratin expression at the highest dose (A, B), while Lim-1 is expressed most strongly at intermediate levels (C). Treatment of mid-streak cultures with noggin results in repression of Lim-1 expression (E), but Odd-1 expression is not significantly affected (D). By stage 8 (F), Lim-1 expression in the IM can no longer be repressed by noggin. (G – I) Pellets of rat fibroblasts expressing noggin (circled regions) were implanted in stages 5 – 6 embryos, which were cultured until stages 11 – 12. Control fibroblasts were implanted on the opposite side. Noggin caused a broadening of Odd-1 expression into more lateral regions of the embryo (*) but also a repression of Odd-1 expression in the area immediately surrounding the cell pellet (G). In contrast, Pax-2 (H) and Lim-1 (I) were repressed by the noggin cells (* and <). 122 R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 Bmp signals can act in a dose-responsive manner in the patterning of vertebrate trunk mesoderm (reviewed in Harland and Gerhart, 1997; Hogan, 1996; Munoz-Sanjuan and Brivanlou, 2001; Niehrs et al., 2000; Schier, 2001; Sive, 1993). In most of these studies, Bmp signaling was manipulated at very early stages in development. Since early alterations in Bmp signaling could have multiple complex effects on embryonic development, one could not determine from these experiments the time during embryogenesis at which Bmp signaling was actually acting to pattern the mesoderm. In addition, most of these studies altered Bmp signaling throughout the embryo. Thus, the question of whether individual cells can modulate mesodermal gene expression in response to variations in Bmp signaling strength could not be directly addressed. In the present study, we have taken advantage of the experimental strengths of the avian embryo model system to manipulate Bmp signaling levels in a time-specific manner in tissues and individual cells, in vivo and in vitro, in order to study the mechanism of Bmp-dependent gene activation in the trunk mesoderm. Timing: Bmp signaling patterns trunk mesoderm during and after gastrulation Treatment of prospective somites explanted from HH5 or HH8 embryos with low and high doses of Bmp promotes IM and lateral plate fate, respectively (Fig. 2), while at stage 10 somitic mesoderm is determined relative to Bmp-2 levels. In addition, Lim-1 expression is inhibited by treatment of prospective IM with the Bmp antagonist noggin at HH stage 5 but not at HH stage 8, while Odd-1 expression was not inhibited by noggin even at HH stage 5. These results imply that Bmp patterning of the IM is already underway by midgastrulation (HH5) and continues through early somite stages (HH8) but is completed by HH10, by which time a battery of IM transcription factors has already begun to be expressed (James and Schultheiss, 2003; Mauch et al., 2000; ObaraIshihara et al., 1999). Cells fated to develop into anterior paraxial, intermediate and lateral plate mesoderm initially gastrulate through the primitive streak at stages 4 and 5 (Garcia-Martinez and Schoenwolf, 1992; James and Schultheiss, 2003; Psychoyos and Stern, 1996). At these stages, the anterior (300 – 600 Am posterior to Henson’s node), middle (600 – 900 Am) and posterior (900 –1200 Am) regions within the primitive streak are already specified to differentiate as paraxial, intermediate and lateral plate mesoderm (Figs. 2, 7). The spatio-temporal localization of active Bmp signaling in the mesoderm has been mapped in chick embryos by distribution of phosphorylated Smad-1/5/8 (Faure et al., 2002). At stages 4 and 5, phosphorylated Smad-1/5/8 is localized in the middle and posterior, but not anterior primitive streak, while at stage 8 phosphorylated Smad-1/5/8 is detectable within the intermediate and lateral plate mesoderm and not in presomitic mesoderm. Thus, the pattern and timing of Bmp signaling in the embryo, as revealed by phosphorylated Smad-1/5/8 activity, are consistent with the current finding that low doses of Bmp signaling pattern the IM between stages 5 and 8. Bmp signals act cell-autonomously Expression of constitutively activated Alk-3 or Alk-6 under the control of weak and strong promoters cell-autonomously activates expression of IM and lateral plate genes, respectively (Figs. 3 –5). These results indicate that, during mesodermal patterning, individual mesodermal cells can respond to local Bmp levels and activate mesodermal genes in a dose-specific cell-autonomous manner. Several elegant reports of the Drosophila wing and ectoderm have also identified genes that respond cell-autonomously to low or high doses of Bmp signals (Lecuit et al., 1996; Nellen et al., 1996). While the vast majority of cells expressing ectopic IM markers in response to intracellular expression of CaAlk3/6 also expressed GFP (see Figs. 3– 5), there were a few cells in which this was not clear. To examine more rigorously the possibility of non-cell-autonomous activation of IM target genes by Bmp, we transplanted quail primitive streak cells that were electroporated with activated Alk3 into chicken anterior primitive streak (Supplementary Fig. 2). Pax-2 and Odd-1 were occasionally activated ectopically in isolated chicken cells that were immediately adjacent to quail grafts. These isolated cells represent a small minority of the ectopic Pax-2 and Odd-1 gene expression as most of these cells co-stained with the quail epitope. However, because they exist, we cannot rule out the possibility that some indirect activation of Pax-2 and Odd-1 is caused by other signals activated by Bmp. This result is only observed in cells adjacent to those electroporated with activated Alk3 under the control of the strong CAGGS promoter. Additionally, non-quail ectopic Pax-2-positive cells were observed only within one cell diameter of a quail cell, indicating that any non-cell-autonomous effect is very short range. In the zebrafish neural tube, misexpression of activated Alk3 was shown to activate epidermal gene expression strictly in cells that expressed a co-transfected tracer dye (Nikaido et al., 1999). Taken together, our data suggest that signals from ectopic Bmp receptors activate transcription of mesodermal target genes in a dose-dependent and largely cell-autonomous manner. The role of Bmp signaling in a mesodermal transcription factor network The findings in this report suggest that Bmp signals activate transcription of intermediate mesodermal markers in a dosedependent, cell-autonomous and translation-dependent manner. We have combined these findings with those of previous studies to generate a model (Fig. 8) in which Bmp signals regulate IM gene expression through a three-fold mechanism: (1) cell-autonomous factors expressed in the paraxial mesoderm negatively regulate IM gene expression; (2) Bmp signaling (at high or low levels) inhibits paraxial gene expression, thereby de-repressing IM gene expression; and (3) high levels of Bmp signaling repress IM gene expression, thereby preventing IM genes from being expressed in the lateral plate. Through this mechanism, IM gene expression is restricted to a strip of tissue between the somites and the lateral plate. The components of the model are discussed below. R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 123 and expanded lateral plate mesoderm. The data presented in this study indicate that activation of Bmp signaling can inhibit somite differentiation in a cell-autonomous manner (Figs. 4, 5). The repression of Tbx-6 (Fig. 6C) was dependent on protein translation, suggesting that Bmp-mediated repression of Tbx-6 is indirect. The immediate Bmp responder Msx-1 has been shown to physically associate with Histone H1b, specifically repress transcription of MyoD and prevent muscle differentiation (Lee et al., 2004). One possibility is that Msx-1, or other transcription factors that are directly activated by Bmp signaling, represses transcription of somitic genes such as Tbx-6, Paraxis and Foxc1/2. Fig. 8. Model of the regulation of trunk mesodermal gene expression by Bmp signaling. In paraxial (somite) regions, Bmp signaling is very low to absent. Paraxial transcription factors, such as Foxc1/2, activate somite genes and repress IM gene expression. In the IM, low levels of Bmp signaling repress somite gene expression and as a result de-repress IM gene expression, leading to activation of IM genes. In the lateral plate, high levels of Bmp signaling activate genes such as cytokeratin and repress IM genes. See text for details. IM gene expression is under negative regulation by cell-autonomous factors expressed in the paraxial mesoderm The findings in Fig. 6D (compare 1st and 3rd panels) demonstrate that Odd-1 transcription can be activated in presomitic mesoderm upon treatment with cyclohexamide alone. This suggests that some molecule(s) present in the somite may be necessary to actively inhibit Odd-1 transcription. In the absence of translation of this/these molecule(s), all signals necessary for Odd-1 transcription are present in somitic mesectoderm. Previous studies have shown that the forkhead transcription factors Foxc1 and Foxc2 are necessary and sufficient to inhibit the transcription of IM factors (Wilm et al., 2004). Mice and zebrafish lacking Foxc1 and Foxc2 do not develop somites (Topczewska et al., 2001; Wilm et al., 2004). Instead, expression of Odd-1 and Lim-1 is expanded medially so that they are present in paraxial and intermediate mesoderm (Wilm et al., 2004). Additionally, when Foxc2 or Foxc1 is transiently transfected into the IM of chick embryos, expression of Pax-2 and Lim-1 is repressed in the treated cells (Wilm et al., 2004). It is unclear if Foxc1/2 directly repress transcription of IM genes or if they act indirectly via other molecules. Bmp signaling represses paraxial gene expression Previous studies have found that Bmp pathway activation inhibits formation of somitic mesoderm and promotes more lateral fates (reviewed in Dale and Jones, 1999; Harland and Gerhart, 1997; Hogan, 1996; Schier, 2001). Exposure of chick presomitic mesoderm to high levels of ectopic Bmp ligand either in culture (Fig. 2; Reshef et al., 1998) or in vivo (Fig. 1; Pourquie et al., 1996; Tonegawa et al., 1997) leads to inhibition of somite differentiation and activation of lateral plate fate. Consistent with this, loss of function of the Bmp antagonists Noggin (McMahon et al., 1998) and Chordin (Dick et al., 2000; Wagner and Mullins, 2002) results in reduced paraxial An activity within the lateral plate inhibits IM differentiation Several reports indicate that there is an activity within the lateral plate that can inhibit IM differentiation. If specified IM (HH8 or younger) is transplanted into the lateral plate mesoderm (James and Schultheiss, 2003) or cultured in vitro in the presence of lateral plate (Mauch et al., 2000), it will not express Lim-1 or Pax-2 as it would if cultured alone (Fig. 7, James and Schultheiss, 2003). Additionally, experimental expansion of the neural plate leads to expansion of somitic mesoderm into the lateral plate (Mariani et al., 2001). Rather than causing lateral displacement of the IM, this manipulation results in an absence of IM gene expression, indicating that the lateral plate environment is not permissive for IM differentiation. In the current study, culture of presomitic mesoderm treated with high concentrations of Bmp-2 for 15 h results in upregulation of the lateral plate marker Cytokeratin and absence of Odd-1, Lim-1 and Pax-2 (Fig. 2). Conversely, if the same tissue is cultured for only 5 h, Odd-1 transcription is upregulated (Fig. 6). These results indicate that Odd-1 is initially activated by the presence of Bmp but subsequently inhibited. This is likely due to the indirect activation of an inhibitor that responds to high concentrations of Bmp signaling. An alternative to the model presented here is that low doses of Bmp signaling activate transcriptional activators present in the intermediate mesoderm. These activators could in turn promote transcription of Odd-1, Pax-2 and Lim-1. One or all of these could then actively repress expression of somite transcription factors such as Foxc genes. However, mice mutant for both Foxc1 and Foxc2 exhibit medial expansion of IM expression in the absence of changes in the distribution of phosphorylated Smad-1 (Wilm et al., 2004). Thus, IM gene expression can be activated by the inhibition of paraxial gene expression, in the absence of alterations in Bmp signaling. These data are consistent with a model in which Bmp signaling activates IM genes by negatively regulating inhibitors of IM gene expression such as Foxc1 and Foxc2 (as proposed in Fig. 8), rather than by inducing activators of intermediate mesodermal gene transcription. Acknowledgments The authors would like to thank the following for providing probes: P. Danielian (mouse Odd-1), R. Maas and E. Matsunaga (chick Pax-2), T. Jessell (chick Lim-1), R. Tuan 124 R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 (chick Paraxis), S. Mackem (chick Tbx-6L) and Y. Takahashi (chick cytokeratin). L. Niswander and M. Whitman kindly provided Alk3 and Alk6 expression constructs. S. Brandt generously provided the anti-chicken Scl/Tal antibody. The authors would especially like to thank Mozhgan Afrakhte and Devin Powell for helping with the cloning of the chick Odd-1 cDNA and to all members of the Schultheiss laboratory for very helpful discussion and comments. R.G.J. was supported by a predoctoral grant from the Howard Hughes Medical Institute. This work was supported by grant R01-DK59980 to T.M.S. Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ydbio.2005.09.025. References Barnes, G.L., Alexander, P.G., Hsu, C.W., Mariani, B.D., Tuan, R.S., 1997. Cloning and characterization of chicken Paraxis: a regulator of paraxial mesoderm development and somite formation. Dev. Biol. 189, 95 – 111. Brugger, S.M., Merrill, A.E., Torres-Vazquez, J., Wu, N., Ting, M.C., Cho, J.Y., Dobias, S.L., Yi, S.E., Lyons, K., Bell, J.R., Arora, K., Warrior, R., Maxson, R., 2004. A phylogenetically conserved cis-regulatory module in the Msx2 promoter is sufficient for BMP-dependent transcription in murine and Drosophila embryos. Development 131, 5153 – 5165. Burrill, J.D., Moran, L., Goulding, M.D., Saueressig, H., 1997. PAX2 is expressed in multiple spinal cord interneurons, including a population of EN1+ interneurons that require PAX6 for their development. Development 124, 4493 – 4503. Campbell, G., Tomlinson, A., 1999. Transducing the Dpp morphogen gradient in the wing of Drosophila: regulation of Dpp targets by brinker. Cell 96, 553 – 562. Dale, L., Jones, C.M., 1999. BMP signalling in early Xenopus development. Bioessays 21, 751 – 760. Dick, A., Hild, M., Bauer, H., Imai, Y., Maifeld, H., Schier, A.F., Talbot, W.S., Bouwmeester, T., Hammerschmidt, M., 2000. Essential role of Bmp7 (snailhouse) and its prodomain in dorsoventral patterning of the zebrafish embryo. Development 127, 343 – 354. Dosch, R., Gawantka, V., Delius, H., Blumenstock, C., Niehrs, C., 1997. Bmp-4 acts as a morphogen in dorsoventral mesoderm patterning in Xenopus. Development 124, 2325 – 2334. Drake, C.J., Brandt, S.J., Trusk, T.C., Little, C.D., 1997. TAL1/SCL is expressed in endothelial progenitor cells/angioblasts and defines a dorsalto-ventral gradient of vasculogenesis. Dev. Biol. 192, 17 – 30. Dressler, G.R., Deutsch, U., Chowdhury, K., Nornes, H.O., Gruss, P., 1990. Pax2, a new murine paired-box-containing gene and its expression in the developing excretory system. Development 109, 787 – 795. Faure, S., de Santa Barbara, P., Roberts, D.J., Whitman, M., 2002. Endogenous patterns of BMP signaling during early chick development. Dev. Biol. 244, 44 – 65. Fujii, T., Pichel, J.G., Taira, M., Toyama, R., Dawid, I.B., Westphal, H., 1994. Expression patterns of the murine LIM class homeobox gene lim1 in the developing brain and excretory system. Dev. Dyn. 199, 73 – 83. Garcia-Martinez, V., Schoenwolf, G.C., 1992. Positional control of mesoderm movement and fate during avian gastrulation and neurulation. Dev. Dyn. 193, 249 – 256. Hamburger, V., Hamilton, H.L., 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49 – 92. Harland, R., Gerhart, J., 1997. Formation and function of Spemann’s organizer. Annu. Rev. Cell Dev. Biol. 13, 611 – 667. Herbrand, H., Guthrie, S., Hadrys, T., Hoffmann, S., Arnold, H.H., RinkwitzBrandt, S., Bober, E., 1998. Two regulatory genes, cNkx5-1 and cPax2, show different responses to local signals during otic placode and vesicle formation in the chick embryo. Development 125, 645 – 654. Hogan, B.L., 1996. Bone morphogenetic proteins in development. Curr. Opin. Genet. Dev. 6, 432 – 438. Holley, S.A., Ferguson, E.L., 1997. Fish are like flies are like frogs: conservation of dorsal – ventral patterning mechanisms. Bioessays 19, 281 – 284. Hollnagel, A., Oehlmann, V., Heymer, J., Ruther, U., Nordheim, A., 1999. Id genes are direct targets of bone morphogenetic protein induction in embryonic stem cells. J. Biol. Chem. 274, 19838 – 19845. James, R.G., Schultheiss, T.M., 2003. Patterning of the avian intermediate mesoderm by lateral plate and axial tissues. Dev. Biol. 253, 109 – 124. Jazwinska, A., Kirov, N., Wieschaus, E., Roth, S., Rushlow, C., 1999a. The Drosophila gene brinker reveals a novel mechanism of Dpp target gene regulation. Cell 96, 563 – 573. Jazwinska, A., Rushlow, C., Roth, S., 1999b. The role of brinker in mediating the graded response to Dpp in early Drosophila embryos. Development 126, 3323 – 3334. Jones, C.M., Dale, L., Hogan, B.L., Wright, C.V., Smith, J.C., 1996. Bone morphogenetic protein-4 (BMP-4) acts during gastrula stages to cause ventralization of Xenopus embryos. Development 122, 1545 – 1554. Kishimoto, Y., Lee, K.H., Zon, L., Hammerschmidt, M., Schulte-Merker, S., 1997. The molecular nature of zebrafish swirl: BMP2 function is essential during early dorsoventral patterning. Development 124, 4457 – 4466. Knezevic, V., De Santo, R., Mackem, S., 1997. Two novel chick T-box genes related to mouse Brachyury are expressed in different, nonoverlapping mesodermal domains during gastrulation. Development 124, 411 – 419. Lecuit, T., Brook, W.J., Ng, M., Calleja, M., Sun, H., Cohen, S.M., 1996. Two distinct mechanisms for long-range patterning by Decapentaplegic in the Drosophila wing. Nature 381, 387 – 393. Lee, H., Habas, R., Abate-Shen, C., 2004. MSX1 cooperates with histone H1b for inhibition of transcription and myogenesis. Science 304, 1675 – 1678. Mariani, F.V., Choi, G.B., Harland, R.M., 2001. The neural plate specifies somite size in the Xenopus laevis gastrula. Dev. Cell 1, 115 – 126. Matsuda, T., Cepko, C.L., 2004. Electroporation and RNA interference in the rodent retina in vivo and in vitro. Proc. Natl. Acad. Sci. U. S. A. 101, 16 – 22. Mauch, T.J., Yang, G., Wright, M., Smith, D., Schoenwolf, G.C., 2000. Signals from trunk paraxial mesoderm induce pronephros formation in chick intermediate mesoderm. Dev. Biol. 220, 62 – 75. McMahon, J.A., Takada, S., Zimmerman, L.B., Fan, C.M., Harland, R.M., McMahon, A.P., 1998. Noggin-mediated antagonism of BMP signaling is required for growth and patterning of the neural tube and somite. Genes Dev. 12, 1438 – 1452. Minami, M., Kinoshita, N., Kamoshida, Y., Tanimoto, H., Tabata, T., 1999. brinker is a target of Dpp in Drosophila that negatively regulates Dppdependent genes. Nature 398, 242 – 246. Mullins, M.C., Hammerschmidt, M., Kane, D.A., Odenthal, J., Brand, M., van Eeden, F.J., Furutani-Seiki, M., Granato, M., Haffter, P., Heisenberg, C.P., Jiang, Y.J., Kelsh, R.N., Nusslein-Volhard, C., 1996. Genes establishing dorsoventral pattern formation in the zebrafish embryo: the ventral specifying genes. Development 123, 81 – 93. Munoz-Sanjuan, I., Brivanlou, A., 2001. Early posterior/ventral fate specification in the vertebrate embryo. Dev. Biol. 237, 1 – 17. Nellen, D., Burke, R., Struhl, G., Basler, K., 1996. Direct and long-range action of a DPP morphogen gradient. Cell 85, 357 – 368. Niehrs, C., Dosch, R., Onichtchouk, D., 2000. Embryonic patterning of Xenopus mesoderm by Bmp-4. Ernst Schering Res. Found., 165 – 190. Nieto, M.A., Sargent, M.G., Wilkinson, D.G., Cooke, J., 1994. Control of cell behavior during vertebrate development by Slug, a zinc finger gene. Science 264, 835 – 839. Nikaido, M., Tada, M., Takeda, H., Kuroiwa, A., Ueno, N., 1999. In vivo analysis using variants of zebrafish BMPR-IA: range of action and involvement of BMP in ectoderm patterning. Development 126, 181 – 190. Nimmagadda, S., Geetha Loganathan, P., Huang, R., Scaal, M., Schmidt, C., R.G. James, T.M. Schultheiss / Developmental Biology 288 (2005) 113 – 125 Christ, B., 2005. BMP4 and noggin control embryonic blood vessel formation by antagonistic regulation of VEGFR-2 (Quek1) expression. Dev. Biol. 280, 100 – 110. Niwa, H., Yamamura, K., Miyazaki, J., 1991. Efficient selection for highexpression transfectants with a novel eukaryotic vector. Gene 108, 193 – 199. Obara-Ishihara, T., Kuhlman, J., Niswander, L., Herzlinger, D., 1999. The surface ectoderm is essential for nephric duct formation in intermediate mesoderm. Development 126, 1103 – 1108. Podos, S.D., Ferguson, E.L., 1999. Morphogen gradients: new insights from DPP. Trends Genet. 15, 396 – 402. Pourquie, O., Fan, C.M., Coltey, M., Hirsinger, E., Watanabe, Y., Breant, C., Francis-West, P., Brickell, P., Tessier-Lavigne, M., Le Douarin, N.M., 1996. Lateral and axial signals involved in avian somite patterning: a role for BMP4. Cell 84, 461 – 471. Psychoyos, D., Stern, C.D., 1996. Fates and migratory routes of primitive streak cells in the chick embryo. Development 122, 1523 – 1534. Pyrowolakis, G., Hartmann, B., Muller, B., Basler, K., Affolter, M., 2004. A simple molecular complex mediates widespread BMP-induced repression during Drosophila development. Dev. Cell 7, 229 – 240. Reese, D.E., Hall, C.E., Mikawa, T., 2004. Negative regulation of midline vascular development by the notochord. Dev. Cell 6, 699 – 708. Reshef, R., Maroto, M., Lassar, A.B., 1998. Regulation of dorsal somitic cell fates: BMPs and Noggin control the timing and pattern of myogenic regulator expression. Genes Dev. 12, 290 – 303. Rusch, J., Levine, M., 1996. Threshold responses to the dorsal regulatory gradient and the subdivision of primary tissue territories in the Drosophila embryo. Curr. Opin. Genet. Dev. 6, 416 – 423. Sainio, K., Raatikainen-Ahokas, A., 1999. Mesonephric kidney—A stem cell factory? Int. J. Dev. Biol. 43, 435 – 439. Schier, A.F., 2001. Axis formation and patterning in zebrafish. Curr. Opin. Genet. Dev. 11, 393 – 404. Schultheiss, T.M., Xydas, S., Lassar, A.B., 1995. Induction of avian cardiac myogenesis by anterior endoderm. Development 121, 4203 – 4214. Schultheiss, T.M., Burch, J.B.E., Lassar, A.B., 1997. A role for bone 125 morphogenetic proteins in the induction of cardiac myogenesis. Genes Dev. 11, 451 – 462. Sive, H.L., 1993. The frog princess: a molecular formula for dorsoventral patterning in Xenopus. Genes Dev. 7, 1 – 12. Smith, W.C., Harland, R.M., 1992. Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70, 829 – 840. Smith, W.C., Knecht, A.K., Wu, M., Harland, R.M., 1993. Secreted noggin protein mimics the Spemann organizer in dorsalizing Xenopus mesoderm. Nature 361, 547 – 549. So, P.L., Danielian, P.S., 1999. Cloning and expression analysis of a mouse gene related to Drosophila odd-skipped. Mech. Dev. 84, 157 – 160. Tonegawa, A., Funayama, N., Ueno, N., Takahashi, Y., 1997. Mesodermal subdivision along the mediolateral axis in chicken controlled by different concentrations of BMP-4. Development 124, 1975 – 1984. Topczewska, J.M., Topczewski, J., Shostak, A., Kume, T., Solnica-Krezel, L., Hogan, B.L., 2001. The winged helix transcription factor Foxc1a is essential for somitogenesis in zebrafish. Genes Dev. 15, 2483 – 2493. Tsuchida, T., Ensini, M., Morton, S.B., Baldassare, M., Edlund, T., Jessell, T.M., Pfaff, S.L., 1994. Topographic organization of embryonic motor neurons defined by expression of LIM homeobox genes. Cell 79, 957 – 970. Wagner, D.S., Mullins, M.C., 2002. Modulation of BMP activity in dorsal – ventral pattern formation by the chordin and ogon antagonists. Dev. Biol. 245, 109 – 123. Wharton, S.J., Basu, S.P., Ashe, H.L., 2004. Smad affinity can direct distinct readouts of the embryonic extracellular Dpp gradient in Drosophila. Curr. Biol. 14, 1550 – 1558. Wilm, B., James, R.G., Schultheiss, T.M., Hogan, B.L.M., 2004. The forkhead genes, Foxc1 and Foxc2, regulate paraxial versus intermediate mesoderm cell fate. Dev. Biol. 271, 176 – 189. Zou, H., Wieser, R., Massague, J., Niswander, L., 1997. Distinct roles of type I bone morphogenetic protein receptors in the formation and differentiation of cartilage. Genes Dev. 11, 2191 – 2203.