SEARCH

SEARCH BY CITATION

Keywords:

  • chick embryo;
  • somite;
  • Wnt reporter;
  • GFP

Abstract

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

A critical mediator of cell–cell signaling events during embryogenesis is the highly conserved Wnt family of secreted proteins. Reporter constructs containing multimerized TCF DNA binding sites have been used to detect Wnt β-catenin dependent activity during animal development. In this report, we have constructed and compared several TCF green fluorescent protein (GFP) reporter constructs. They contained 3, 8, or 12 TCF binding sites upstream of a minimal promoter driving native or destabilized enhanced GFP (EGFP). We have used the electroporation of somites in the chick embryo as a paradigm to test them in vivo. We have verified that they all respond to Wnt signaling in vivo. We have then assessed their efficiency at reflecting the activity of the Wnt pathway. Using destabilized EGFP reporter constructs, we show that somite cells dynamically regulate Wnt/β-catenin–dependent signaling, a finding that was confirmed by performing time-lapse video confocal observation of electroporated embryos. Developmental Dynamics 239:346–353, 2010. © 2009 Wiley-Liss, Inc.


INTRODUCTION

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

The Wnt family of signaling proteins participates in multiple developmental events during embryogenesis and has also been implicated in adult tissue homeostasis. Wnt signals are pleiotropic, with effects that include mitogenic stimulation, cell fate specification, and differentiation. There are 19 known Wnts in mouse and human (18 in chick) that cluster into 13 subfamilies, 7 of which are present in Drosophila (see http://www.stanford.edu/∼rnusse/wntwindow.html; http://compare.ibdml. univ-mrs.fr/; http://www.treefam.org/; Gordon and Nusse, 2006, Logan and Nusse, 2004; Clevers, 2006; Klaus and Birchmeier, 2008, and references therein).

Wnts mediate their activity through the binding to transmembranal receptors, named Frizzled (Fz). There are 10 distinct Fz in mammals. Their structure encompasses an extracellular cysteine-rich domain that is sufficient to bind Wnt proteins. However, it is likely that Fz mediates Wnt signal together with members of the LRP family of single pass transmembrane proteins that, therefore, act as co-receptors for Wnts.

Intracellularly, Wnt activates multiple signaling pathways. The first characterized (and thus named “canonical” pathway) leads to the cytoplasmic accumulation of β-catenin and to its translocation into the nucleus. There, β-catenin partners with members of the TCT/LEF family of transcription factors to activate the transcription of Wnt target genes. In the absence of nuclear β-catenin, TCF associates with the transcriptional repressor Groucho and histone deacetylases to form a repressive complex that blocks transcription of Wnt target genes. When β-catenin enters the nucleus, it replaces Groucho from its binding to TCF and converts the complex to a transcriptional activator, thereby triggering the transcription of Wnt target genes.

Alternate, β-catenin–independent Wnt signaling has also been characterized. In vertebrates, noncanonical Wnt pathway controls aspects of gastrulation movements, cochlear hair cell morphology, neural tube closure, muscle fiber orientation, heart patterning, etc. In Drosophila, this pathway (named Planar Cell Polarity or PCP) controls the orientation of bristles on the body and wings of the fly, as well as the orientation of the ommatidia in the eye. Intracellular effectors of noncanonical signaling include calcium flux, JNK, and small heterotrimeric G proteins (reviewed in Klein and Mlodzik, 2005; Wang and Nathans, 2007; Gordon and Nusse, 2006).

To track Wnt β-catenin–dependent activity during animal development and in cell lines, several laboratories have built reporter constructs containing multimerized TCF DNA binding sites. The first Wnt reporters (Korinek et al., 1997) were constructed using three copies of the optimal TCF motif CCTTTGATC, upstream of a minimal c-Fos promoter driving luciferase expression (TOPFLASH) or upstream of a minimal herpes virus thymidine kinase (TK) promoter driving chloramphenicol acetyltransferase (TOPCAT). Three copies of the mutant TCF motif CCTTTGGCC upstream of the same constructs served as negative controls of Wnt activity (FOPFLASH and FOPCAT, respectively). These constructs were transfected in a murine B cell line to test for Wnt activity. A little later, the first active transgenic Wnt reporter mouse line was created: the TOPGAL mouse (DasGupta and Fuchs, 1999). The TOPGAL transgene was based on the reporter TOPFLASH containing 3 TCF binding sites, but with lacZ replacing luciferase as the reporter gene. A TCF reporter derived from the same TOPFLASH construct was also used in zebrafish, TOPd2EGFP, that used a destabilized form of green fluorescent protein (GFP) as a reporter gene (Dorsky et al., 2002). More recently, a TCF-LacZ transgenic mouse reporter, called BAT-gal, that contains seven TCF sites upstream of the minimal promoter-TATA box of the gene siamois has been described (Maretto et al., 2003). Eight (Veeman et al., 2003), 12, or 16 (DasGupta et al., 2005) TCF-binding sites upstream of minimal promoters driving the luciferase reporter gene have also been successfully used in mammalian and Drosophila cell line assays (Barolo, 2006, for review). Axin2/conductin is a feedback inhibitor of Wnt β-catenin–dependent signaling activated in many, and perhaps all, sites of Wnt signaling (Jho et al., 2002 and references therein). Thus, mouse line carrying a LacZ knock-in in the Axin2/conductin locus have also been used to monitor Wnt β-catenin–dependent signaling during development (Jho et al., 2002; Lustig et al., 2002).

The role of Wnt signaling in early myogenesis has been extensively studied, often using cultured mesoderm explants (Fan and Tessier-Lavigne, 1994; Münsterberg et al., 1995; Münsterberg and Lassar, 1995; Stern et al., 1995; Maroto et al., 1997; Tajbakhsh et al., 1998; Reshef et al., 1998; Borello et al., 2006). The technique of in vivo electroporation of somites (which are the embryonic structures from which body and limb skeletal muscles derive) is particularly suitable to characterize the role of Wnt signaling during embryonic development. It allows investigators to specifically target the expression of cDNA constructs of interest in distinct cell populations that constitute the somite. Using this technology, it was shown that the Wnt/β-catenin–dependent pathway is necessary for muscle fate acquisition, as determined by the activation of the Muscle Regulatory transcription Factor, Myf5 (Gros et al., 2009), thereby confirming similar analyses made in mouse (Borello et al., 2006). Wnt/β-catenin–dependent signaling is also required for the maintenance of the epithelial organization of the dermomyotome, that constitutes the dorsal-most compartment of somites (Schmidt et al., 2004; Linker et al., 2005). In contrast, noncanonical PCP signaling regulates the oriented elongation of muscle fibers in the anteroposterior axis of the embryo (Gros et al., 2009).

Combining the imaging technologies to observe cell behavior in live embryos with the electroporation of reporter genes, we are now for the first time in a position to monitor the activities of signaling pathways as they operate in an amniote embryo. Such an approach should allow the identification of cells that activate specific pathways and to correlate this activity with unique cell behavior (shape changes, migration, cell fate acquisition, etc.) during early development. For this, tools that faithfully reflect the activity of the signaling pathways of interest are required. In this report, we have compared multiple TCF GFP reporter constructs that we have made in our laboratory, using somite electroporation as an experimental paradigm. They contained 3, 8, or 12 TCF binding sites upstream of a TK minimal promoter driving native or destabilized enhanced GFP (EGFP). We have tested that they all respond to Wnt/β-catenin–dependent signaling in vivo. Expectedly, increasing the number of TCF binding sites considerably increases the number of cells expressing the GFP reporter, likely as a result of the enhanced sensitivity of the constructs to Wnt signaling. By comparing the expression of the protein and the mRNA of the reporter gene GFP, we show that the activity of the Wnt/β-catenin dependent pathway is more faithfully reflected by the use of a destabilized (d2, i.e., 2 hr half-life) than by stable GFP. Surprisingly, this analysis also showed that, at a given time, only a fraction of the electroporated somite cells express the reporter, although they are all competent to do so when an activated form of β-catenin is co-electroporated with the reporter. This suggests that somite cells temporarily regulate Wnt activity. We confirmed this by performing a time-lapse video confocal observation of electroporated embryos, which demonstrated the dynamic activation of Wnt β-catenin activity in somite cells.

RESULTS AND DISCUSSION

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

We have used four Wnt reporter constructs in this study, named 3TOPFLASH-EGFP, 8TOPFLASH-d2EGFP, 12TOPFLASH-EGFP, and 12TOPFLASH-d2EGFP that differ by the number of TCF binding sites (3, 8, and 12, respectively), and by the stability of the GFP reporter inserted downstream of the thymidine kinase minimal promoter (see the Experimental Procedures section). While EGFP is reported to have a very long half-life of 24–26 hr (Corish and Tyler-Smith, 1999; Yen et al., 2008), d2EGFP has a half-life of approximately 2 hr (Li et al., 1998). A 2-hr half-life is within the range of the half-life of most proteins coded in the genome that display a 30 min to 2 hr half-life (Yen et al., 2008).

TCF Reporter Constructs Respond to Wnt β-Catenin–Dependent Signaling in Chick Somites

Here, we have used somite electroporation as a paradigm to test the efficiency of the reporter constructs. We have first compared the GFP reporter activities of 3TOPFLASH-EGFP (3 TCF binding sites) and of 12TOPFLASH-EGFP (12 TCF binding sites) in the chick embryo. Embryos were electroporated as described (see the Experimental Procedures section) in the medial portion of newly formed epithelial somites. After incubation, this results in the expression of the constructs by cells located in the dorsomedial lip of the dermomyotome (DML), in the transition zone (TZ) and in the myotome (see Fig. 1A, and Supp. Fig. S1A,B, which is available online). A plasmid containing a fusion protein of the histone H2B and the fluorophore RFP (labeling the nuclei, in red) was co-electroporated with the TOPFLASH reporter genes, and served as internal control of electroporation. Embryos were examined 24 hr after electroporation of the constructs. Increasing the number of TCF binding sites has been used to increase the sensitivity of reporter constructs to β-catenin–dependent signaling (Barolo, 2006, for review). Accordingly, considerably more cells expressed GFP with 12 TCF binding sites (Fig. 1C, in green) than with 3 TCF binding sites (Fig. 1B, in green). A quantification was made by comparing the total number of electroporated cells (indicated by H2B-RFP staining) with those co-expressing the TOPFLASH reporters and showed that the proportion of cells that express 12 TCF binding sites (82%) is much higher than with 3 TCF binding sites (9%; Fig. 1E). It was important to show that the TCF reporters respond to Wnt signaling in somites. To test this, a constitutively active form of β-catenin was co-electroporated with 3TOPFLASH-EGFP. Analysis of these embryos showed that the expression of GFP was observed in nearly all electroporated cells (88%; Fig. 1D,E compared with 9% without the active form of β-catenin, Fig. 1B,E; see also Fig. 2K–M), indicating a strong activation of the promoter/enhancer activity. No GFP expression was observed when a vector containing variant, inactive TCF binding sites (8XFOPFLASH) was electroporated in somites (not shown). These data suggest that multimerized (3 or 12) TCF binding sites act as valid tools to analyze Wnt/β-catenin–dependent signaling in somites.

thumbnail image

Figure 1. A: Schematics of the different regions of the somite. DML, dorsomedial lip; TZ, transition zone; NT, neural tube; Sm, somite. B–D: Series of confocal views (image stacks) of chick embryo somites electroporated in the DML, incubated for 24 hr and immunostained for green fluorescent protein (GFP; in green) and red fluorescent protein (RFP; in red). Somites are shown in ventral view. B: Somite electroporated with 3TOPFLASH-EGFP (n = 15 embryos; 45 somites). C: Electroporation of 12TOPFLASH-EGFP (n = 14 embryos; 53 somites). D: Co-electroporation of 3TOPFLASH-EGFP with a constitutively active form of β-catenin (n = 8 embryos; 18 somites). E: Bars represent the proportion of cells that were GFP- (green), H2BRFP-positive (red), or positive for both fluorochromes (yellow). Scale bars = 50 μm.

Download figure to PowerPoint

thumbnail image

Figure 2. A–L: Series of confocal views (image stacks) of chick embryo somites electroporated in the dorsomedial lip of the dermomyotome (DML), incubated for 24 hr and immunostained for green fluorescent protein (GFP; in green) or stained for red fluorescent protein (RFP) mRNA (in red). Somites are shown in ventral view. A–C: Electroporation of 3TOPFLASH-EGFP (n = 20 embryos; 58 somites). D–F: Electroporation of 12TOPFLASH-EGFP (n = 20 embryos; 43 somites). G–I: Electroporation of pCAGGS-EGFP (n = 12 embryos; 32 somites). J–L: Co-electroporation of 3TOPFLASH-EGFP with a constitutively active form of β-catenin (n = 15 embryos; 56 somites). M: Bars represent the proportion of cells that were stained for the GFP protein (green), for the GFP mRNA (red) or positive for both the protein and the mRNA (yellow). Scale bars = 50 μm.

Download figure to PowerPoint

A Comparison of Stable and Destabilized GFP Reporters to Monitor Wnt Signaling In Vivo

A question arose as to whether all GFP-positive cells observed in these experiments are actually activating the signaling pathway. Given the high stability of the GFP protein, it was possible that cells would remain positive for the protein after they have shut down Wnt/β-catenin–dependent signaling, that would result in a halt in the transcription of the GFP mRNA. We performed double stainings for the GFP protein and for its mRNA. When the 3TOPFLASH-EGFP and the 12TOPFLASH-EGFP electroporated embryos were double-stained, only 46% or 40%, respectively, of somite cells co-expressed the protein and the mRNA for GFP, while 53% or 59%, respectively, expressed the protein only; 1–2% of cells were positive only for the mRNA (Fig. 2A–F,M). This result shows that less than half of the somite cells positive for GFP are actively transcribing this gene. We verified whether the discrepancy between protein and mRNA staining is not due to a higher sensitivity of the protein compared with mRNA detection. We performed double staining for GFP protein and mRNA on embryos electroporated with a GFP reporter under the control of a strong, ubiquitous promoter enhancer (pCLGFPA-GFP, that contains a CMV/Chick beta actin). In this case, 95% of cells were positive for the protein and the mRNA, while 3% were positive for the protein only and 2% for the mRNA (Fig. 2G–I,M). This indicates that the sensitivity of detection of protein (by immunohistochemistry) and mRNA (by fluorescent in situ hybridization, FISH) is comparable. We then verified whether all somite cells are competent to activate Wnt signaling. The co-electroporation of an activated form of β-catenin with the 3TOPFLASH-EGFP resulted in nearly all cells (90%) co-expressing the EGFP mRNA and protein (Fig. 2J–L), confirming at an mRNA level what was observed at a protein level described above (Fig. 1D) and indicating that all electroporated cells are competent to activate this pathway. Altogether, our results indicate that the half-life of stable EGFP is too long to observe the dynamic activation of the Wnt β-catenin–dependent pathway in somites.

We determined which of the available destabilized GFPs would be more suitable to faithfully detect cells that activate the pathway. Destabilized d1-, d2-, and d4-EGFP, that are reported to display a half life of 1, 2, or 4 hr, respectively, are commercially available (Clontech). A hyper-destabilized EGFP was also recently developed (LuVeLu vector; Aulehla et al., 2008) that contains a Venus-PEST fusion destabilized GFP protein, linked to a Lunatic Fringe (Lfng) 3′ untranslated region that further destabilizes the mRNA. The signals detected after electroporation of a d1-EGFP or the Venus-PEST-3′Lfng, both under the control of 12 TCF binding sites, were too low to be used in our assays (not shown). In contrast, when d2-EGFP (2 hr half-life) was placed under the control of 8 (8TOPFLASH-d2EGFP) or 12 (12TOPFLASH-d2EGFP) TCF binding sites, 5 or 26% of electroporated cells, respectively, were GFP positive (Fig. 3A–C). This experiment allows one to evaluate the effect of replacing native GFP by an unstable d2-EGFP, in the same molecular context, i.e., 12 TCF-binding sites. This leads to a significant decrease of GFP-positive cells, from 40 to 26%, respectively (compare Fig. 1C and Fig. 3B). To answer whether the expression of d2-EGFP reflects the transcription of the GFP mRNA, we performed double stainings for the GFP protein and for its mRNA. After electroporation of 8TOPFLASH-d2EGFP and 12TOPFLASH-d2EGFP, in both cases 83% of electroporated cells co-expressed the mRNA and the protein (Fig. 3D–J), indicating that the expression of the d2-EGFP protein product faithfully (but not totally) mimics that of its mRNA. Surprisingly, 8 and 16%, respectively, of the cells expressed the mRNA only. It is unclear at present why this may be: it is possible that the d2-EGFP protein is being degraded faster than its mRNA. Altogether, these data suggest that constructs that contain destabilized EGFP with a half-life of 2 hr, downstream of TCF binding sites reflect the activity of Wnt/β-catenin dependent signaling.

thumbnail image

Figure 3. Series of confocal views (image stacks) of chick embryo somites electroporated in the dorsomedial lip of the dermomyotome (DML), and incubated for 24 hr. A,B: Co-electroporation of 8TOPFLASH-d2EGFP (A; n = 12 embryos; 32 somites) or 12TOPFLASH-d2EGFP (B; n = 15 embryos; 55 somites) with CAGGS-H2BRFP, immunostained for green fluorescent protein (GFP; in green) and red fluorescent protein (RFP; in red). GFP-positive cells are observed in the DML and in the TZ. C: Bars represent the proportion of cells that were GFP- (green), H2BRFP-positive (red), or positive for both fluorochromes (in yellow). D–I: Electroporation of 8TOPFLASH-d2EGFP (D–F; n = 16 embryos; 56 somites) or 12TOPFLASH-d2EGFP (G–I; n = 21 embryos; 45 somites) immunostained for GFP (in green) or stained for GFP mRNA (in red). J: Bars represent the proportion of cells that were stained for the GFP protein (green), for the GFP mRNA (red), or positive for both the protein and the mRNA (yellow). Scale bars = 50 μm.

Download figure to PowerPoint

Using these two reporters, we could determine the somitic regions where the Wnt/β-catenin–dependent pathway is active. GFP-positive cells were observed in the DML, and in the TZ (Fig. 3A,B), that are morphologically recognizable on transverse optical sections, and are immunologically distinguished by their specific labeling with Myf5 or MF20 (Supp. Fig. S1). This indicates that the pathway is active in both regions of the somite, an observation that was confirmed by FISH staining with both vectors (Fig. 3D–I). Wnt/β-catenin– dependent signaling maintains the epithelial structure of the dermomyotome (Schmidt et al., 2004; Linker et al., 2005) and is required for the activation of the myogenic program in the medial somite (Fan and Tessier-Lavigne, 1994; Münsterberg et al., 1995; Münsterberg and Lassar, 1995; Stern et al., 1995; Maroto et al., 1997; Tajbakhsh et al., 1998; Reshef et al., 1998; Borello et al., 2006; Manceau et al., 2008; Gros et al., 2009). Further analyses will be needed to test the function of Wnt signaling in different parts of the somite.

These analyses also uncovered a surprising feature of Wnt signaling in somites. The comparison of the proportion of electroporated cells versus those expressing the destabilized GFP showed that at a given time, only a fraction of the electroporated cells expressed the reporter (Fig. 3A–C), although, as shown above, they are all competent to do so (Fig. 2J–L). This suggests that somite cells temporarily regulate Wnt activity. To verify this, we performed a confocal time-lapse videomicroscopy (as described in Gros et al., 2005) of embryos electroporated with 12TOPFLASH-d2EGFP (Fig. 4A–G and Supp. Movie S1). We observed epithelial DML cells activating d2-EGFP expression over the time period of the observation, thereby demonstrating that Wnt signaling is dynamically regulated within somites. Importantly, this finding could not be easily predicted from the previous analyses made with stable reporters (EGFP or LacZ) of Wnt activity and this underlines the crucial need to develop molecular tools that allow the observation of dynamic processes in live tissues. Although this observation may be of crucial biological importance to understand the role of this pathway in somite differentiation, at present, we do not know the mechanisms that regulate the dynamic activation of Wnt signaling in somites.

thumbnail image

Figure 4. A–G: Confocal views (image stacks) of a chick embryo somite co-electroporated in the dorsomedial lip of the dermomyotome (DML) of somites with the 12TOPFLASH-d2EGFP (in green) and the CAGGS-H2B-RFP (in red). The embryo was first incubated for 8 hr in ovo, at which time a slice (200–250 mm thick) of the trunk region was examined overnight with a spinning disk confocal microscope. Image stacks were taken every 11′. A is a snapshot of the stack at the beginning of the acquisition. Dotted lines indicate the position of the DML, the transition zone (TZ) and the myotome (My). B–G are snapshots of the movie taken at indicated times (see Supp. Materials). The position of B–G is delineated by a yellow box in A.

Download figure to PowerPoint

Combining emerging technologies of in vivo imaging to observe various morphogenetic processes in developing chick embryos with the electroporation technique of reporter genes allows for the first time to monitor the activities of signaling pathways as they operate in the embryo. In this study, we have designed a molecular tool, the 12TOPFLASH-d2EGFP vector, that is appropriate to observe the activity of the Wnt β-catenin–dependent signaling pathway in vivo. Such a tool will enable the correlation of the activation of Wnt activity with unique cell behavior (shape changes, migration, cell fate acquisition, etc.) during early development.

EXPERIMENTAL PROCEDURES

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

Electroporation Vectors, Somite Electroporation

Four Wnt reporter constructs were used in this study, named 3TOPFLASH-EGFP; 8TOPFLASH-d2EGFP; 12TOPFLASH-EGFP; 12TOPFLASH-d2EGFP.

8TOPFLASH-d2EGFP is derived from SUPER8TOPFLASH (Veeman et al., 2003, kindly provided by R. Moon). The eight TCF-binding elements and the TK minimal promoter of SUPER8TOPFLASH were cloned upstream of a destabilized (2 hr half-life) d2EGFP-N1 (Clontech) in a pBluescript-SK (Stratagene) backbone vector. The d2EGFP variant is a fusion of the EGFP with a fragment of the mouse ornithine decarboxylase which contains a PEST amino acid sequence that targets the fused protein for degradation.

Three TCF-binding sites were placed upstream of the TK minimal promoter followed by EGFP (from the ptkEGFP plasmid, kindly provided by M. Uchikawa, Uchikawa et al., 2003) to construct the 3TOPFLASH-EGFP. We replaced the 3 TCF binding sites by 12 TCF-binding sites, derived from dTF12 (DasGupta et al., 2005, kindly provided by N. Perrimon), to give the 12TOPFLASH-EGFP. Finally, the EGFP of 12TOPFLASH-EGFP was replaced by d2EGFP-N1 (Clontech) to give the 12TOPFLASH-d2EGFP.

A constitutively activate form of β-catenin was constructed from a Xenopus β-catenin cDNA clone by removing the sequences coding for the N terminal 47 amino acids (as described in Yost et al., 1996). This was cloned into the pCAGG vector.

The pCLGFPA-GFP vector has been described in Scaal et al., 2004.

A pCAGGS-H2B-mRFP containing a fusion protein of the histone H2B and the fluorophore RFP, kindly provided by S. Tajbakhsh.

Electroporation was performed as described previously (Scaal et al., 2004; Gros et al., 2004, 2009). Newly formed interlimb (somites 22–28) somites of stage 15–16 HH (Hamburger and Hamilton, 1992) chick embryos were electroporated in all experiments. The positive and negative electrodes were positioned so that the medial region of the epithelial somite would be electroporated. A few hours later, the somites differentiate into a dorsal epithelial compartment (the dermomyotome) and a ventral mesenchyme (the sclerotome). We have shown earlier (Gros et al., 2004) that this electroporation technique results in the targeting of the plasmids in the cells of DML of the dermomyotome (i.e., the medial-most portion of the dermomyotome, directly apposed to the neural tube). While a variable fraction (20–60%) of the DML cells are electroporated using this technique, the co-electroporation of the H2B-mRFP served as an internal control to compare experiments. Embryos were reincubated 17–24 hr before analysis. After incubation, the progeny of the electroporated cells were located in the DML, in the TZ and in the myotome (Gros et al., 2004, 2009). The allocation of electroporated cells to the three somite regions was determined by shape, position within the stack, and immunostainings of the cells. Cytoplasmic GFP allowed us to determine the shape of the cells expressing the Wnt reporter genes, which is different in the three regions (pseudo-stratified epithelium in the DML, mesenchymal-like in the TZ, and elongated in the myotome). Immunostainings of electroporated embryos with a marker of the myotome (Myosin Heavy Chain, MyHc), in addition to antibodies directed against GFP and RFP, was included in most experiments (see Supp. Fig. S1), thus allowing to determine unequivocally the position of electroporated cells in relation to the myotome (see Supp. Fig. S1). Position of the cells within confocal image series was determined by scanning through the stacks of images and by examining transverse optical sections of the stacks. The figures that are presented throughout the study are projections (i.e., superpositions) of confocal image series (typically 20–40 images, 1 μm apart) of somites immunostained for GFP and red fluorescent protein (RFP; or immunostained for GFP and stained by ISH for its mRNA), and examined in ventral view (Figs. 1, 3, 4). A transverse optical section of a confocal stack immunostained for GFP, RFP, and MyHc is presented in Supp. Fig. S1 to better visualize the position of electroporated cells in a typical experiment. The combination of these approaches showed that no electroporated cells were located in the sclerotome.

Cell countings were performed by scanning and manually counting through the confocal image series and not on the projections presented here. For RFP and GFP immunostainings, around 10,000 cells per point were counted; when GFP protein and mRNA were analyzed, around 1,000 cells were counted per point.

Immunostaining, FISH, Time-Lapse Confocal Videomicroscopy, Imaging

Electroporated embryos were fixed in 4% formaldehyde, immunostained, cleared in 90% glycerol/H2O and examined with a Zeiss LSM 510 confocal microscope. Antibodies used are a chicken anti-GFP (Abcam), a rabbit anti-RFP (Abcam), and a mouse anti-Myosin Heavy Chain MF20 (Hybridoma Bank). Fluorescent in situ hybridizations were performed as described (Denkers et al., 2004).

We observed the dynamic activation of Wnt signaling in somite cells, using an ex vivo embryo slice culture system combined with confocal time-lapse videomicroscopy as described (Gros et al., 2005). Briefly, embryo slices (200–250 μm thick) of the trunk region were placed in a 3-cm Petri dish where the bottom was replaced by a glass coverslip. The sample was placed in a heated chamber and examined overnight on an inverted microscope (Zeiss Axiovert) to which a Nipkow spinning disk confocal head (Perkin Elmer Ultraview) was attached.

Image stacks (50–100 μm thick) were treated with Volocity (Improvision) and Imaris softwares for image visualization and three-dimensional reconstruction. Movies were visualized with Imaris softwares and improved with ImageJ software suites.

Acknowledgements

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

We thank Dr. Xavier Cousin for the 8TOPFLASH-d2EGFP and Aurélie Jory and Dr. Sharaghim Tajbakhsh for the CAGGS-H2B-RFP. The help of Pascal Weber from the Institute's Imaging Facility is acknowledged. A.R. is supported by a grant from the Agence Nationale de la Recherche (ANR); this study is funded by the EEU 6th Framework Programme Network of Excellence MYORES.

REFERENCES

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

Supporting Information

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

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
DVDY_22174_sm_suppinfoFigS1.tif1028KThis figure (derived from Figure 1C) is a projection (i.e., superposition) of confocal images of a somite co-electroporated in the DML with 12TOPFLASH-EGFP (in green) and of H2B-RFP (in red). Twenty-four hours after electroporation, it was immunostained for GFP and RFP and for MyHC (MF20, in blue), to show the position of the nascent myotome. This is a ventral view of the image stack. B: is a transversal view of the same stack, that allows to see the position of the cells within the somite. The shape of the DML (and of the dermomyotome), of the transition zone (TZ) and of the Myotome are delineated by dotted lines.
DVDY_22174_sm_suppinfomovieS1.mp4563KSupp. Movie S1. This movie shows the activation of Wnt beta catenin dependent signaling in a dorsomedial lip (DML) cell of a chick somite. In red are the nuclei of the electroporated cells (shown by the H2B-RFP fluorescence); in green are the cells that activate the Wnt pathway (shown by the 12TOPFLASH-d2EGFP). One epithelial DML cell is seen to strongly activate the pathway over the time of the observation.

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.