SEARCH

SEARCH BY CITATION

Keywords:

  • chicken embryo;
  • avian;
  • facial prominence;
  • Bambi;
  • BMP receptor;
  • Noggin;
  • micromass culture;
  • Q PCR;
  • craniofacial;
  • retinoids

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Here, we examine the expression and regulation of the gene BAMBI, a kinase-deficient decoy receptor capable of interacting with type I bone morphogenetic protein (BMP) receptors in avian embryos. Initially, expression was limited to the endoderm during neurula and pharyngula stages. From embryonic day 3.5 (stage 20) and onward, BAMBI expression almost perfectly overlapped with known expression patterns for BMP4, particularly in the face and limbs. We performed bead implant experiments in the face to see which signals could be repressing or promoting expression of BAMBI. Our data point to retinoids and BMPs as being major positive regulators of BAMBI expression; however, fibroblast growth factor 2 acts to repress BAMBI. Furthermore, retinoic acid is likely to act directly on BAMBI as induction occurs in the presence of cycloheximide. The data suggested that BAMBI could be used to regulate Bmp signaling during tissue interactions that are an integral part of facial morphogenesis. Developmental Dynamics 237:1500-1508, 2008. © 2008 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Bmp (bone morphogenetic protein) signaling controls many aspects of development including apoptosis cell proliferation and differentiation. Bmps act through dimers of the type I and type II serine–threonine kinase receptors that in turn phosphorylate receptor-Smads (1,5,8; Miyazono et al.,2000; Balemans and Van Hul,2002; Shi and Massague,2003; Chen et al.,2004; Gazzerro and Canalis,2006). The Smads enter the nucleus and modify transcription of targets such as Runx2, a transcription factor required for bone formation. Antagonists of the Bmp pathway are present at several levels: extracellular (Noggin, Gremlin, Chordin), cell surface (Bambi, a decoy receptor), and intracellular inhibitory Smads 6,7 (Gazzerro and Canalis,2006; Gazzerro and Minetti,2007). We have previously reported that one of these antagonists, Noggin, is expressed in a very restricted part of the facial epithelium and that down-regulation occurs just before fusion of the lip. Furthermore, we showed that there is a balance of Bmp activity required in the fusion zone to promote merging of the frontonasal mass with the cranial edge of the maxillary prominence (Ashique et al.,2002a).

Here, we examine the expression and regulation of another antagonist, BAMBI (BMP and activin membrane-bound inhibitor) in the chicken embryo. BAMBI has sequence similarity to type I Bmp receptors but lacks an intracellular kinase domain that phosphorylates downstream targets. Thus, BAMBI first dimerizes with other type I receptors and then forms a complex with the BmprII receptor. The Bambi dimer interferes with Bmp and activin-like transforming growth factor-beta (TGF-β) signaling in a ligand-independent manner (Onichtchouk et al.,1999). Orthologues of gallus BAMBI are found in Xenopus (Onichtchouk et al.,1999), zebrafish (Tsang et al.,2000; Schebesta et al.,2006), rat (Loveland et al.,2003), and mouse (Grotewold et al.,2001). One of the most interesting features of Bambi is that expression is very tightly correlated with sites of Bmp4 expression in xenopus, mouse, and chicken (Onichtchouk et al.,1999; Grotewold et al.,2001; Zuzarte-Luis et al.,2004). Moreover, a feedback loop exists such that Bmps induce expression of Bambi, which will then decrease the level of signaling (Onichtchouk et al.,1999).

The expression of Bambi has been reported in the neural crest of Xenopus (Onichtchouk et al.,1999), zebrafish embryos (Tsang et al.,2000), regenerating zebrafish fins (Schebesta et al.,2006), developing teeth (Knight et al.,2001), and mouse facial prominences (Grotewold et al.,2001), but a full developmental series of Bambi expression has not been published. Furthermore, very little is known about the expression in chicken with the exception of the interdigital region of the limb bud (Zuzarte-Luis et al.,2004). Therefore, our goal was to examine the expression over a series of stages in the chicken embryo and then to focus on regulation of Bambi in a region with temporal and spatially restricted transcripts, the head.

During early organogenesis, the amniote face is composed of prominences or swellings of mesenchyme covered in epithelium that surround the oral cavity (Francis-West et al.,2003; Richman and Lee,2003; Jiang et al.,2006). In avian embryos, the upper beak is derived from the frontonasal mass and lateral nasal and maxillary prominences. The lower beak is entirely formed by the mandibular prominences. The upper beak is distinguished from the lower beak by the presence of the keratinized egg tooth, which helps the chick to emerge from the egg. The egg tooth is first visible at stage 31 (Hamburger and Hamilton,1951). At first, the facial prominences are all separate, but then at stage 28 (Hamburger and Hamilton,1951) fusion begins to take place between the frontonasal mass and maxillary prominences. Outgrowth of prominences is dependent on molecular signaling between epithelium and mesenchyme (Richman and Tickle,1989,1992; MacDonald et al.,2004).

Avian facial morphogenesis is controlled by several signals including BMPs (Barlow and Francis-West,1997; Lee et al.,2001; Ashique et al.,2002a; Abzhanov et al.,2004; Wu et al.,2004,2006), retinoids (Song et al.,2004), Sonic Hedgehog (Hu and Helms,1999; Abzhanov and Tabin,2004; Brito et al.,2006), Wnts (Wingless-related MMTV integration site; Brugmann et al.,2007), and fibroblast growth factors (FGFs; Richman et al.,1997; Mina et al.,2002; Mina and Havens,2007). BAMBI gene expression is likely to be regulated by BMPs in the avian face based on data from other vertebrate model organisms (Onichtchouk et al.,1999; Tsang et al.,2000); however, no one had previously studied this relationship in amniotes. Retinoids are capable of inducing BMPs (Rodriguez-Leon et al.,1999; Paralkar et al.,2002); therefore, we wished to test whether retinoic acid (RA) could also positively regulate BAMBI expression and whether this effect was likely to be direct. Finally, FGFs are often found in reciprocal relationships with BMPs (Neubuser et al.,1997), and we hypothesized that FGFs could regulate BAMBI expression. To address these questions, we first investigate BAMBI expression in the embryo and then focus on how BAMBI is regulated by BMP, FGF, and retinoid signals during facial morphogenesis.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Expression of GallusBAMBI RNA During Development

Although there is some evidence for Bambi expression in Xenopus neural crest (Onichtchouk et al.,1999), we did not see such expression in the chicken. At the four-somite stage, BAMBI was expressed entirely in the endoderm and this continued to be the case at stage 10, when abundant cranial neural crest cell migration was under way (Fig. 1A–D,N). We could not detect expression within the embryo proper at stage 15 except in the anterior intestinal portal and other regions of the endoderm (Fig. 1E). The same result was obtained in tissue sections hybridized to BAMBI probe (data not shown). It was striking that BAMBI was not expressed in any of the regions that are known to express BMP4 in the neurula and pharyngula stage embryos, such as the presumptive neural ectoderm and dorsal edges of the neural tube (Liem et al.,1995; Sela-Donenfeld and Kalcheim,1999).

thumbnail image

Figure 1. Expression of BAMBI during chick development. Whole-mount in situ hybridizations with the BAMBI probe. Hamburger and Hamilton (HH) stages are indicated in the top right corner. A–D: Ventral views (A,C) and dorsal views (B,D). Dashed line in C is plane of section in N,O. E,F: Lateral views. Arrowhead in F indicates expression in the dorsal brain. Dashed line in F indicates plane of section for panel O. G: Frontal view of the head. H: Lateral view of the whole embryo. Arrowheads indicate anterior and posterior expression domains in the limbs. I: Arrow indicates the apical ectodermal ridge. J–M: Frontal views. With increasing age, lateral expression frontonasal mass decreases and is upregulated in the center where the egg tooth will subsequently form. Dashed line in M is plane of section for P. N: Paraffin sections in the sagittal plane. Boxed area is shown in higher power. Expression is seen only in the endoderm. O: Frontal section showing expression in the pharyngeal endoderm, ectoderm of the mandibular (white arrowhead), and maxillary prominences (black arrowhead). P: Mid-sagittal section through the egg tooth (black arrowhead). aer, apical ectodermal ridge; d, diencephalon; en, endoderm; et, egg tooth; fnm, frontonasal mass; ip, anterior intestinal portal; lb, lower beak; lnp, lateral nasal prominence; md, mandibular prominence; mx, maxillary prominence; np, nasal pit; p, pharynx; pr, prosencephalon; s, somite; ub, upper beak. Scale bars = 500 μm in A–M,O, 100 μm in N, 300 μ in N inset, 300 μm in P.

Download figure to PowerPoint

thumbnail image

Figure 2. Effects of Noggin, bone morphogenetic protein-7 (BMP7), and fibroblast growth factor-2 (FGF2) on expression of BAMBI. Frontal views of embryos treated at stage 24. Beads are indicated with asterisks. Increases in expression are shown with white arrowheads. A: Control embryos treated with Tris-soaked beads had no change in expression. B,C: Expression is completely ablated. D–F: Signal is increased in strength and expanded. G,H: Decreased expression near bead. Inset in H is a positive control showing MSX2 expression induced by FGF2 protein 16 hr after bead implantation. I: Vibratome frontal section of embryo in H showing decreased BAMBI expression in the epithelium (black arrowhead). Bead is out of the plane of section. Scale bars = 500 μm.

Download figure to PowerPoint

Expression of BAMBI in the embryo was first observed at stage 20, where transcripts were localized to the apical ectodermal ridge (AER; Fig. 1F) and dorsal neural tube (Fig. 1F and data not shown). In the head, BAMBI was expressed in the lateral edges of the frontonasal mass and lateral nasal and maxillary prominences of the face (Fig. 1G). There was also strong expression in the midline of the mandibular prominence. Abundant transcripts were found in the roof plate of the rhombencephalon (Fig. 1F). In sections of whole-mount hybridized embryos, it was clear that BAMBI was expressed strongly in the facial ectoderm and in the subadjacent mesenchyme as well as in the ventral side of the pharyngeal endoderm (Fig. 1O). Dorsal to the brain, in the region of the diencephalon, there were Bambi transcripts in the mesenchyme and head ectoderm (Fig. 1O). At stage 24, expression of BAMBI continued to be mainly restricted to the face and limbs (Fig. 1H–J). There was a very restricted pattern in the face with the highest expression in the corners of the frontonasal mass, cranial maxillary prominences, and lateral nasal prominence. All of these regions will fuse together to make the upper beak. In the limb, there was BAMBI expression in the AER (Fig. 1I), but in addition, there were domains along the anterior and posterior edges of the limb buds (Fig. 1H). All of these regions in the face and limb also express BMP4 (Francis-West et al.,1994; Francis et al.,1994). At stage 28, intensity of expression in the face was still strong (Fig. 1K) and by stage 29 and 30, the main region in frontonasal mass expressing BAMBI had shifted to the newly formed egg tooth (Fig. 1L,M,P). The expression in egg tooth is fairly unique; therefore, BAMBI will be a useful marker for this structure, especially when egg teeth are induced out of context (Lee et al.,2001).

BMPs and FGFs Regulate BAMBI Expression

There is evidence for a role for BMPs or TGFβ in regulating Bambi in developmental (Onichtchouk et al.,1999; Tsang et al.,2000; Karaulanov et al.,2004) or cancer (Sekiya et al.,2004a,b) contexts, but no one has studied how BAMBI interacts with other signals in the facial prominences. In addition, the highly restricted expression of BAMBI in avian embryos suggests that there are signals actively repressing expression in areas such as the center of the frontonasal mass or lateral edges of the mandibular prominence. To investigate this question, we treated embryos with exogenous signals or antagonists of growth factor signalling at stages when BAMBI was maximally expressed in the face, stage 24. In addition, beads were implanted directly into one of the areas with high BAMBI expression. The intention was to see whether any of the factors could either suppress expression that was already present or induce ectopic expression in adjacent tissues.

BMP4 is found primarily in the epithelium of the cranial maxillary prominences, corners of the frontonasal mass, and midline of the mandibular prominences (Francis-West et al.,1994; Ashique et al.,2002a). BMP2 binds to the same receptors as BMP4 (Shi and Massague,2003) and is found in the mesenchyme underlying areas of BMP4 expression. To test the role of endogenous Bmps in the control of BAMBI expression, we treated embryos with Noggin, an antagonist of Bmp2, 4, and 7 (Zimmerman et al.,1996). BAMBI expression was completely ablated both at 6 hr (6/6) and 16 hr (8/8) after treatment (Fig. 2B,C; Table 1), confirming that several endogenous Bmps are acting to maintain BAMBI expression.

Table 1. Gene Expression Changes Detected on the Treated Side of the Frontonasal Mass or Maxillary Prominences Compared to the Untreated Sidea
TreatmentTime post-treatment (hours)No changeIncreaseDecrease
  • a

    RA, retinoic acid; BMP, bone morphogenetic protein; FGF, fibroblast growth factor.

RA6 (n = 6)150
 16 (n = 5)140
BMP73, 6 (n = 8)170
 16 (n = 5)140
NOGGIN6 (n = 6)006
 16 (n = 8)008
FGF26 (n = 11)308
 16 (n = 6)204

It is known from work on Xenopus animal caps that Bmp4 activates Bambi expression; therefore, it was expected that the same result would be obtained in chicken embryos. No one had tested whether BMP7, which is also expressed in the facial prominences (Ashique et al.,2002a), can activate BAMBI. Furthermore, BMP7 binds to a different type of type I Bmp receptor than do BMP2 and 4 (Alk2 as opposed to Alk3 and 6) and the Alk2 receptor does not bind to Bambi (Onichtchouk et al.,1999). It is not known whether signaling by means of the Alk2 receptor could induce BAMBI expression. To test this possibility, we implanted BMP7-soaked beads into the face. We found that the expression of BAMBI transcripts was increased strongly and expanded into adjacent mesenchyme (Fig. 2D–F; Table 1). The facial mesenchyme was extremely sensitive to exogenous BMP7 such that up-regulation could be detected in as little as 3 hr (Fig. 2D, 2/3) and persisted for up to 16 hr (Fig. 2E,F; Table 1). Thus, our data in combination with the embryo work of others (Onichtchouk et al.,1999) shows that signaling by means of any of three Bmp type I receptors will initiate a transcriptional activation of BAMBI expression.

FGF signalling controls the growth of facial mesenchyme (Richman and Crosby,1990; Richman et al.,1997; Mina et al.,2002; Mina and Havens,2007). In addition, one member of the FGF family, FGF8 is expressed in complementary domains to BMP4 in the epithelium of the facial prominences. FGF2 binds to the same mesenchymal FGF receptors as does FGF8 (Zhang et al.,2006) and for reasons of increased bioactivity, we substituted FGF2 protein in our experiments. To see whether FGF2 would repress or stimulate expression of BAMBI, we placed FGF2 soaked beads at the corner of the frontonasal mass. We observed a decrease of Bambi expression at 6 hr and 16 hr (Fig. 2G,H; Table 1). The same protein induced a strong up-regulation of MSX2 (Fig. 2H inset), showing that the effect on BAMBI is opposite to that of other FGF2 targets. We also checked the tissue-specific effects of FGF2 on BAMBI expression by sectioning the embryos and found the decrease to be both in mesenchyme and epithelium (Fig. 2I). These data suggest that endogenous FGFs such as FGF8 could be repressing BAMBI expression in the caudal maxillary prominence, lateral mandibular prominence, and near the nasal slit in the frontonasal mass.

Bambi Is Up-regulated by RA

RA can up-regulate BMPs; therefore, it was possible that the BAMBI protein was a common mediator of BMP and RA signaling. We, therefore, implanted beads soaked in RA into the face. Expression was strongly unregulated at both early (6 hr; 5/6) and later (16 hr; 4/5) time points (Fig. 3B–E; Table 1). Furthermore, although the beads were implanted into the mesenchyme, expression was mostly unregulated in the epithelium (Fig. 3D′,E′). This finding suggested that RA may alter tissue interactions that in turn affected BAMBI expression in the epithelium.

thumbnail image

Figure 3. Qualitative and quantitative effects of retinoic acid (RA) on expression of BAMBI. All embryos were treated with RA beads soaked in 5 mg/ml RA at stage 24. Beads are indicated with asterisks. Arrowheads show increased expression. A: Control bead does not affect expression. B–E: Increased expression in the frontonasal mass. D′: Frontal Vibratome section of embryo in D. Arrowhead shows increased ectodermal expression. E: Extensive up-regulation of expression across the frontonasal mass ectoderm. E′: Sagittal Vibratome section showing ectopic ectodermal expression in the midline of the frontonasal mass. F: Quantitative polymerase chain reaction results showing significant up-regulation of BAMBI in cultured frontonasal mass mesenchyme with RA in the absence or presence of cycloheximide. Values were normalized relative to those for gallus β-ACTIN. Analysis of variance was used to compare relative expression values between the four treatment groups. Asterisks indicate significant differences in expression (P < 0.01). chx, cycloheximide; DMSO, dimethylsulfoxide; fnm, frontonasal mass. Scale bars = 500 μm.

Download figure to PowerPoint

Next, we tested whether RA is a direct regulator of BAMBI expression. The RA-receptor complex translocates to the nucleus where DNA binding occurs and regulation of downstream targets ensues (Underhill et al.,2001; Weston et al.,2003). To distinguish between direct or indirect induction of BAMBI by RA, we added cycloheximide to embryos implanted with RA-soaked beads. Unfortunately, this treatment in ovo led to a general decrease in BAMBI expression within 6 hr, making it difficult to see any changes in gene expression (data not shown). We then turned to a cell culture assay involving primary cultures of stage 24 frontonasal mass mesenchyme. Some aspects of normal cell differentiation are recapitulated in this system (Richman and Crosby,1990), and these cells have been shown to respond to RA (Wedden et al.,1987; Langille et al.,1989). As expected, RA induced a significant increase in the expression of BAMBI in frontonasal mass mesenchyme cultures compared with dimethyl sulfoxide (DMSO) cultures (Fig. 3F; 144.7%, P = 0.001). However, the cultures treated with RA and cycloheximide also continued to show significant up-regulation of BAMBI compared with controls (Fig. 3F; 165.3%, P = 0.0002 for DMSO, P = 0.0003 for cycloheximide). Moreover, there is no significant difference between the level of BAMBI expression in RA-treated vs. RA + cycloheximide cultures (P = 0.1). These quantitative experiments demonstrate that RA can induce BAMBI expression without protein synthesis, and thus, BAMBI is likely to be a direct target of RA.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Compared with other type I or II Bmp receptors we have examined, BAMBI has very restricted temporal and spatial expression patterns. In particular, the lack of expression in the early embryo suggests that BAMBI protein functions primarily in later organogenesis. We were intrigued by the highly restricted expression of BAMBI in the developing face and performed bead implant experiments to see which signals could be controlling the expression domain. Our results have shown that Fgfs repress BAMBI, whereas retinoids and Bmps are major positive regulators of BAMBI expression.

Type I Bmp Receptors Are Expressed in the Same Areas as Bambi

In order for BAMBI to function as a decoy receptor, it is necessary to first interact with the other type I Bmp receptors in the same area to bind to the type II Bmp receptor and repress signaling (Onichtchouk et al.,1999). However, unlike other antagonists such as Noggin and Chordin, Bambi does not need to interfere with ligand binding to block receptor signaling (Onichtchouk et al.,1999; Tsang et al.,2000). Reviewing the areas of highest BAMBI expression and comparing these to the locations of Bmp receptors reveals that there is overlap in many regions. The endoderm, the tissue with highest Bambi expression in neurula and pharyngula stage embryos, expresses Bmpr1A and this tissue is dependent on signaling from this receptor for development (Davis et al.,2004). We have previously mapped expression of Bmp receptors in the avian face and found that, at stage 28, there are high expression levels of the type IB and 1A receptors, with BMPRIB being mainly found in the condensing cartilage and BMPR1A being more ubiquitously expressed in the mesenchyme (Ashique et al.,2002b). Thus, in most places where BAMBI is expressed, there will be appropriate receptors with which to dimerize, and BAMBI is in a position to control the level of Bmp signaling in a tissue-specific manner.

Bambi Regulation Is Mainly by Means of Bmps and Retinoids

RA induced BAMBI in the embryo model as well as in micromass cultures. One possible mechanism is that RA induced BMPs, which in turn induced BAMBI. RA has previously been shown to increase BMP4 (Rodriguez-Leon et al.,1999) and BMP7 (Paralkar et al.,2002) expression. Epithelial–mesenchymal recombination experiments have shown that the target of RA in the face is the mesenchyme rather than epithelium (Wedden,1987); however, we observed an increase in BAMBI expression in the epithelium of RA-treated embryos. This epithelial response is likely to be indirect and is probably secondary to the induction of BMPs in the mesenchyme caused by the RA bead.

In addition to indirect effects, our cycloheximide data show that RA can directly regulate BAMBI expression, perhaps by means of retinoic acid response elements (RARE) in the gene. However, in one study on the XenopusBambi gene, there was no evidence of RARE in the 6 kb of the promoter that was analyzed (Karaulanov et al.,2004). These structural analyses do not rule out a RARE elsewhere in the Bambi gene. It is also possible that the chicken sequence is different than other species and contains a RARE. We conclude that RA acts on BAMBI by both indirect (by means of induction of BMP) and direct means.

The promoter of Xenopus Bambi has a functional Bmp response element, and expression of Bambi is increased even in the presence of cycloheximide (Karaulanov et al.,2004). These data are consistent with the very rapid response to application of BMP7 and to blocking with Noggin. That Bambi has a typical Bmp response element further explains why any of the BMP proteins tested so far will activate BAMBI expression.

The up-regulation of other antagonists such as NOGGIN is a characteristic effect of BMP proteins released from beads (Lee et al.,2001; Ashique et al.,2002a), and we have shown that BAMBI is similarly increased. Once the temporary effects of the BMP protein have dissipated, the translated products of BAMBI and NOGGIN will likely repress Bmp signaling thereby restoring homeostasis in the embryo.

FGF2 Repression of BAMBI May Be Mediated by Effects on BMP4 Expression

Members of the FGF family are involved in reciprocal relationships with the Bmps (Neubuser et al.,1997; Bei and Maas,1998; Furthauer et al.,2004). Bmps act upstream of Fgf8 in both mouse (Liu et al.,2005a) and chicken (Ashique et al.,2002a) such that loss of Bmp signaling leads to an expansion of Fgf8 expression. Conflicting data exist on whether Fgfs act upstream of Bmps. In conditional knockouts of Fgf8 in the mouse AER, there is a slight increase in Bmp expression; however, in double knockouts of Fgf4 and 8, there is decreased expression. Other conditional deletions of Fgf8 in the mouse face did not show a change in Bmp4 expression (Trumpp et al.,1999). In our study, we observed a decrease in BAMBI expression following FGF2 protein application. In addition, other studies in our lab show that BMP4 is decreased following FGF2 bead implants (Szabo-Rogers et al.,2008). Taken together, these bead implant data suggest two things: first that FGF2 effects on BAMBI could be mediated by changes in BMP4 expression, and second that lack of BAMBI expression in certain regions of the face is partly due to repressive effects of FGF signals.

Mouse Gene Targeting Reveals Redundancy of Bambi Function in Embryo Development

The balance of Bmp signaling is very important for embryonic development in general and in particular for facial morphogenesis (Barlow and Francis-West,1997; Ashique et al.,2002a; Liu et al.,2005a, b). Full knockouts of the Bmp antagonist Bambi in mice are predicted to lead to increased Bmp signaling and defective embryos. Surprising, homozygous null mice were born alive and the only phenotype was reduced postnatal survival in some females (Chen et al.,2007).

One possible reason why phenotypes were not observed could be due to the relatively late onset of expression of Bambi, peaking in organogenesis stages. A similar observation was made in the mouse embryo (Grotewold et al.,2001) where Bambi was expressed in the extraembryonic membranes (allantois and amnion) at embryonic day (E) 7.5–E8.0. There are no in situ data presented between E8.5 and 10.0, but by 10.5, there is localized expression in the same regions of the facial prominences as in the stage 20–28 chicken embryo. Because many aspects of embryo patterning are well established by the time expression is detected in the embryo, targeted deletion of this gene may be too late to disturb morphogenesis. At the later stages, there are many other Bmp antagonists expressed and any of these might regulate Bmp levels in the absence of Bambi. It would be necessary to cross the Bambi null mice into a mouse line deficient in another Bmp antagonist such as Noggin (Brunet et al.,1998). This strategy would be predicted to achieve a more pronounced increase in the level of Bmp signaling. Finally, species differences in the requirement for Bambi gene function could exist. The mouse data do not preclude a role for BAMBI in human craniofacial development. In human populations with isolated clefts, genetically susceptible individuals may be pushed across the threshold to form a cleft by the alterations in the level of growth factor signaling (Riley and Murray,2007). Isolated human cleft lip could arise in individuals who have altered BAMBI function or expression.

In summary, we have shown that restricted expression of BAMBI is controlled by at least three different signaling pathways, two of which are by direct means (BMPs, RA) and one that is most likely indirect (Fgfs). BAMBI could, therefore, provide the means to control BMP homeostasis in both the epithelium and mesenchyme during facial and limb morphogenesis.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Embryos and Bead Implantation

Fertilized White Leghorn eggs were obtained from the University of Alberta and incubated to the appropriate stage (Hamburger and Hamilton,1951). Embryos were treated with several different signaling molecules or proteins using microscopic beads. In all cases, the compounds were applied locally to the lateral corner of the frontonasal mass (globular process) at stage 24.

AG1X-2 beads (Bio-Rad, format form, 200-μm diameter) were soaked in 5 or 10 mg/ml of all-trans-RA (Sigma) dissolved in DMSO for 30 min at room temperature. Control beads were soaked in DMSO only. Affi-Gel blue agarose beads (Bio-Rad) were soaked in 0.1 mg/ml BMP7, 0.65 mg/ml Noggin (Regeneron), 1 mg/ml FGF2 (Peprotech) for a minimum of 1 hr at room temperature.

Whole-Mount In Situ Hybridization

Whole-mount in situ hybridization using digoxigenin-labeled RNA probes was performed as described (Shen et al.,1997). A partial BAMBI chicken cDNA was generously provided by J. Hurle (Zuzarte-Luis et al.,2004). After hybridization, embryos were post-fixed in 10% formaldehyde in phosphate buffered saline (PBS) and then either paraffin embedded or embedded in 3.5% agarose. Paraffin embedding was carried out by first dehydrating specimens through an isopropanol series. Seven-micrometer sections were made and counterstained with eosin. Agarose embedding required first rinsing off the fixative with PBS and then positioning in molten agarose. Vibratome sections were made at 50 μm, and slides were coverslipped with MOWIOL.

Micromass Culture of Facial Mesenchyme

High-density micromass cultures were created from chicken frontonasal prominences of stage 24 embryos (Richman and Crosby,1990; Weston et al.,2000). Culture medium consisting of 40% Dulbecco's modified Eagle's medium and 60% F12 supplemented with fetal bovine serum to 10% was used. One day after setting up the cultures, RA was added to complete culture medium at a concentration of 500 ng/ml, with or without cycloheximide (initially dissolved in PBS). The final concentration of cycloheximide was 10 μg/ml. Control cultures had either cycloheximide or DMSO added without RA. Following 6 hr of treatment, total RNA was collected from the cells according to manufacturer's directions (Qiagen, RNAeasy).

Real-time PCR

High-Capacity cDNA reverse Transcription Kits (Applied Biosystems, Foster City, CA) were used to make cDNA. Quantitative real-time PCR analysis was performed on a 7500 Fast Real-Time PCR System (Applied Biosystems, US). Gallus-specific primers and their corresponding fluorescence probes were designed to cross the intron–exon boundaries between exon 2 and 3 of gBAMBI (primer and probe sequences are shown in Supplementary Table S1, which can be viewed at http://www.interscience.wiley.com/jpages/1058-8388/suppmat). In all cases, both forward and reverse primers were used at a concentration of 300 nM, while the concentration of the probe was 100 nM. Real-time PCR was performed using TaqMan Fast Universal PCR Master Mix (Applied Biosystems, US). For the PCR reaction, 12 ng of random-primed cDNA template was incubated with TaqMan Fast Universal PCR Master Mix in a final volume of 15 μl. Cycling parameters were as follows: 95°C for 20 sec, followed by 40 cycles of denaturation at 95°C for 3 sec, and primer extension at 60°C for 30 sec. The gβ-ACTIN gene was chosen as a reference gene (Supplementary Table S1). One-way analysis of variance (with 3 degrees of freedom) followed by Tukey's post hoc test was used to determine significant differences between groups (Statistica, Version 6.0).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The authors thank Lyndsay Grant and Jeff Chen for their contributions to the early stages of this project and Kathy Fu for expert technical assistance. The authors thank T.M. Underhill for assistance with the Q-PCR analysis. The work was funded by CIHR grants to J.M.R. J.M.R. is a Michael Smith Distinguished Scholar.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The Supplementary Material referred to in this article can be found at http://www.interscience.wiley.com/jpages/1058-8388/suppmat

FilenameFormatSizeDescription
dvdy21529-supplementarytable1.doc25KSupporting Information file dvdy21529-supplementarytable1.doc

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.