SEARCH

SEARCH BY CITATION

Keywords:

  • Frodo;
  • mouse;
  • in situ hybridization;
  • migrating cells;
  • Wnt;
  • Dapper;
  • somites;
  • primitive streak;
  • neural crest

Abstract

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

Frodo has been identified as a protein interacting with Dishevelled, an essential mediator of the Wnt signaling pathway, critical for the determination of cell fate and polarity in embryonic development. In this study, we use specific gene probes to characterize stage- and tissue-specific expression patterns of the mouse Frodo homologue and compare them with Frodo expression patterns in Xenopus embryos. In situ hybridization analysis of mouse Frodo transcripts demonstrates that, similar to Xenopus Frodo, mouse Frodo is expressed in primitive streak mesoderm, neuroectoderm, neural crest, presomitic mesoderm, and somites. In many cases, Frodo expression is confined to tissues undergoing extensive morphogenesis, suggesting that Frodo may be involved in the regulation of cell shape and motility. Highly conserved dynamic expression patterns of Frodo homologues indicate a similar function for these proteins in different vertebrates. Developmental Dynamics 235:279–284, 2006. © 2005 Wiley-Liss, Inc.


INTRODUCTION

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

Cell fates commonly are specified during embryonic development through a cascade of inductive signaling events that occur between the three germ layers: ectoderm, mesoderm, and endoderm. Signaling from the dorsal inducing center (the Spemann organizer, in amphibians, or the primitive streak, in amniotes) is critical for specification of the central nervous system (CNS) and the somites (Lemaire and Kodjabachian, 1996; Beddington and Robertson, 1998; Smith and Schoenwolf, 1998; Weng and Stemple, 2003). Several stage- and tissue-specific genes participating in these signaling events have been identified in the past several years, and their roles in several major pathways have been uncovered. Products encoded by these genes usually represent secreted ligands, their receptors, or transcription factors associated with particular cell fates. In contrast, common cytoplasmic or nuclear mediators of signal transduction are often ubiquitous and not limiting for signaling. However, a scaffold or an “adaptor” protein, essential for a specific signaling event, may itself serve as a key regulatory point in a pathway. For example, mitogen-activated protein kinase signaling requires specific scaffolding proteins (Morrison and Davis, 2003). Thus, the knowledge of spatial and temporal expression patterns for these gene products is critical for the analysis of the corresponding signaling pathways.

Frodo (Gloy et al., 2002) and a highly related protein, Dapper (Cheyette et al., 2002), are novel mediators of cell signaling that were shown to contain an N-terminal leucine-zipper domain, a highly conserved middle domain, and a C-terminal PDZ-binding domain. None of these three domains are present in other proteins. Frodo was identified by virtue of its physical and functional interactions with Dishevelled, an essential mediator of Wnt signaling (Wharton, 2003; Brott and Sokol, 2005). Our recent study demonstrated that Frodo associates with a T-cell factor (TCF), a transcriptional regulator of Wnt target genes, and modulates a reporter gene containing TCF-binding sites (Hikasa and Sokol, 2004). Using morpholino (MO) antisense oligonucleotides (Summerton, 1999; Heasman et al., 2000), in vivo depletion of Frodo resulted in selective down-regulation of dorsal mesoderm genes and neural tissue markers, suggesting that Frodo plays a key role in organizer formation and neural tissue specification during Xenopus development (Hikasa and Sokol, 2004). Moreover, Frodo appeared to be necessary for proper morphogenetic movements during gastrulation and neurulation (Hikasa and Sokol, 2004). A zebrafish homologue of Frodo (Frodo/dapper 1, Frd1; Gillhouse et al., 2004) was proposed to stimulate the canonical Wnt signaling during anteroposterior CNS patterning (Waxman et al., 2004). Whereas these findings suggest that the Frd gene family carries essential functions in mesoderm and neural tissue development and has a role in Wnt signaling pathway in lower vertebrates (Brott and Sokol, 2005), the roles of Frodo homologues in mammalian embryogenesis have not been investigated.

To initiate the study of Frd homologues in mammals, we obtained a mouse Frodo-specific probe and compared spatial and temporal patterns of expression using early Xenopus and mouse embryos. In mouse embryos, we observed highly dynamic patterns of expression in the primitive streak and adjacent mesoderm, neuroectoderm, neural crest, presomitic mesoderm, and somites, similar to the expression domain of Xenopus Frodo. These findings reflect the conservation of inductive signaling in lower and higher vertebrates and suggest that Frodo is likely to play an important role in mammalian development.

RESULTS AND DISCUSSION

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

Frodo Expression After Xenopus Gastrulation and Neurulation

As the initial characterization of Frodo expression has been limited (Gloy et al., 2002), a more thorough analysis of the spatial distribution of Frodo transcripts was carried out by whole-mount in situ hybridization. At the beginning of gastrulation, Frodo transcripts were present throughout the whole animal pole hemisphere, including presumptive neuroectoderm. Weak expression was detectable at the dorsal blastopore lip, corresponding to involuting dorsal mesendoderm (Fig. 1A). At stage 10.5, Frodo was strongly expressed in posterior dorsal mesoderm, with lower transcript levels present in neuroectoderm and ventral mesoderm (Fig. 1B). By the onset of neurulation, the Frodo expression domain in neuroectoderm became gradually restricted to the edge and midline of neural plate (Fig. 1C,D). Two bands visible in the anterior neural plate (Fig. 1C,D, arrowheads) may correspond to the bands of Frodo expression in the posterior diencephalon of zebrafish embryos (Gillhouse et al., 2004). At late neurula and early tail bud stages, Frodo transcripts were enriched in the neural crest and the eyes (Fig. 1D,E,H). At later stages, the most significant domains of Frodo expression were eyes, and branchial arches (Fig. 1E–G). The other major expression domains included the presomitic mesoderm at the tip of the tail and the somites (Fig. 1E–G).

thumbnail image

Figure 1. Localization of Frodo transcripts during Xenopus embryogenesis. A,B: Lateral views of stage 10 (A) and stage 10.5 (B) embryos (animal is up; dorsal is to the right), showing Frodo expression in the dorsal mesoderm (DM) and neuroectoderm (NE). Embryos were bisected sagittally before whole-mount in situ hybridization. Black arrowheads point to the dorsal lip (DL). C,D: Dorsal views of stage 14 (C) and stage 18 (D) embryos (anterior is up). Frodo expression is detectable predominantly at the edge and the midline (ML) of the neural plate (NP), a part of the diencephalon (arrowheads), in early neurulae (C) and in neural crest cells (NC) at the late neurula stage (D). A–D: Right panels contain schematic diagrams for each embryo shown. E,F: Lateral views of stage 23 (E) and stage 28 (F) embryos (anterior is to the left). Frodo is expressed in the eyes, branchial arches (BA), and presomitic mesoderm (PSM, arrowhead) at the early tail bud stage (E) and is also detectable in the somites (S, arrows) at the late tail bud (F) stage. G: Lateral view of stage 33 embryo (anterior is to the left), showing Frodo expression in branchial arches (BA), somites (S, arrows), and the tip of the tail (arrowhead). H: A transverse section of stage 18 embryo, counterstained in eosin, shows enriched expression of Frodo in neural crest (NC).

Download figure to PowerPoint

Expression of Frodo in Mouse Embryogenesis

In situ hybridization analysis was used to examine the expression of the mouse homologue of Frodo in mouse early embryos. Sectioned 7.0–8.0 days post coitum (dpc) embryos revealed high levels of mFrodo transcripts in the primitive streak and mesoderm adjacent to the primitive streak (Fig. 2A–C,E). mFrodo is also expressed in the splanchnic mesoderm and faintly in differentiating myocardial plate (Fig. 2F). Frodo expression was not detectable in the epiblast or head folds at these stages.

thumbnail image

Figure 2. Localization of mFrodo transcripts during mouse development in 7.0–8.0 days post coitum (dpc) embryos. AC: Diagram of an 7.0–7.5 dpc mouse embryo (after Kaufman, 1992); lines b and c in A correspond to cross-sections in B and C. Embryonic mesoderm (EM), red arrowheads. Primitive streak (PS), black arrowhead. D: Diagram of an 8.0 dpc mouse embryo; lines e–h in D correspond to cross-sections in E–H. E: mFrodo expression is detected in caudal neuroectoderm (NE, filled black arrows). F: Foregut diverticulum (FD, red arrow). mFrodo expression is detected in splanchnic mesoderm (SM, red line). Low levels of mFrodo expression are seen in differentiating myocardial plate (M, black line). G: Coelomic channel (CC, open black arrow). H: mFrodo transcripts are detected in caudal neuroectoderm and a subset of rostral neuroectoderm adjacent to the head folds (filled black arrow). CM, Cephalic mesenchyme.

Download figure to PowerPoint

Whole-mount and section staining of 8.5 dpc and 9.5 dpc embryos revealed new expression domains of mFrodo in the somites, presomitic mesoderm, and septum transversum (including the proepicardial organ and hepatic diverticulum) (Fig. 3A–K). mFrodo expression can also be detected in migratory neural crest/cranial mesenchymal cell populations, with low levels in premigratory neural crest (Fig. 3D,F,H,H′, and not shown). In the developing somites, the staining was distributed in a graded manner, being stronger caudally and being mainly restricted to the ventral–posterior region of each somite, known to contribute to the sclerotome (Fig. 3J,K). The gap of expression between presomitic mesoderm and the first somite suggests that Frodo may be involved in the control of somite periodicity (Fig. 3A,J,K).

thumbnail image

Figure 3. Localization of mFrodo transcripts in 8.5–9.5 days post coitum (dpc) mouse embryos detected by in situ hybridization. A–D: Whole-mount staining of 8.5 dpc embryos. A: Side view of an embryo showing mFrodo expression in somites (S, black lines), and presomitic mesoderm (PSM, black arrowhead). B: Dorsal view, somites (S, black lines). C: Ventral view, somites (S, black lines) and foregut diverticulum (FD, red arrow). D: Dorsal/lateral view of the head region. Frodo expression is detected in neural crest and cephalic mesenchyme (NC/CM, black arrows). E: A diagram of an 8.5 dpc embryo, in which lines f–i correspond to transverse sections through embryos shown in the corresponding F–I. F: Neural crest and cephalic mesenchyme (NC/CM; black arrows). G: mFrodo expression in presomitic mesoderm (PSM, arrowheads) and neuroectoderm (NE, red arrowheads). H: mFrodo expression in neuroectoderm (NE, red arrowheads), somites (S, black line), and migrating neural crest and cephalic mesenchyme (NC/CM, black arrow). H′: Higher magnification of the mFrodo neural crest expression domain shown in H. I: mFrodo transcripts are abundant in the septum transversum (ST, red line), which includes the proepicardial organ and hepatic diverticulum. J,K: Whole-mount staining of 9.5 dpc embryos. mFrodo expression in neural crest/cephalic mesenchyme (NC/CM, black arrow), septum transversum (ST, red lines), somites (S, black lines), and presomitic mesoderm (PSM, black arrowheads).

Download figure to PowerPoint

In conclusion, the mouse Frodo gene is mainly expressed in mesodermal tissues, including the primitive streak and mesoderm adjacent to the primitive streak, somites, and presomitic mesoderm. Additional expression of mFrodo in neuroectoderm derivatives, such as the neural crest, and caudal neural ectoderm, agrees with the observation that human Frodo mRNA is enriched in the embryonic brain (data not shown). This pattern is similar to the pattern observed in Xenopus embryos (Fig. 1), reflecting evolutionary conservation of transcriptional regulatory mechanisms that control Frodo expression. Moreover, Wnt signaling pathways that were proposed to involve Frodo are known to be active in the same embryonic tissues (Pinson et al., 2000; Yamaguchi, 2001). Together with the evidence that Frd2/dpr2, a Frodo-related gene, regulates Nodal signaling in zebrafish (Zhang et al., 2004), it is possible that Frodo is involved in more than one signaling pathway.

Our data suggest that vertebrate Frd homologues may function during mesoderm specification, somitogenesis, neural tissue, and neural crest development. This suggestion is consistent with our previous study, which demonstrated an essential role for Frodo in dorsal mesoderm formation and neural tissue specification during Xenopus development (Hikasa and Sokol, 2004). A common feature of mFrodo expression domains is that they correspond to the area of active cell migration and/or mesenchymal/epithelial transformation. This observation is supported by the demonstration that Frodo is required for normal morphogenetic processes in Xenopus embryos (Hikasa and Sokol, 2004). Thus, both mammalian and frog Frodo homologues may function in migrating mesodermal and neuroectodermal tissues to control cell shape/motility as well as cell fate during development. The experimental proof for this view in mammals should be provided by the genetic ablation of the mouse Frodo gene.

EXPERIMENTAL PROCEDURES

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

Plasmids and Probe Preparation

Partial cDNA encoding mouse Frodo (pT7T3Pac-mFrodo) was obtained from the IMAGE consortium. The mouse expressed sequence tag clone was sequenced, and the sequence was submitted to NCBI (accession no. AY208970). pT7T3Pac-mFrodo was linearized with EcoRI and was transcribed with T3 polymerase in the presence of digoxigenin (DIG) -11-UTP for in situ hybridization probe. Similar results were obtained for a different mFrodo antisense probe, derived from a genomic DNA clone, confirming the conclusions reported in this study (data not shown).

Xenopus Embryos and Whole-Mount In Situ Hybridization

In vitro fertilization and embryo culture in 0.1 × Marc's modified Ringer's solution (MMR) were carried out as described (Peng, 1991). Staging was according to Nieuwkoop and Faber (1967). Whole-mount in situ hybridization was carried out according to (Harland, 1991) with slight modifications as described previously (Hikasa and Sokol, 2004). Embryo sections were carried out in Paraplast and counterstained with eosin as described (Gloy et al., 2002). For hemisections, the embryos stored in ethanol after fixation were rehydrated in 1 × phosphate buffered saline (PBS), 0.1% Tween 20 and bisected with a razor blade before hybridization. DIG-labeled antisense RNA probes were synthesized pBSSK-Frodo (Gloy et al., 2002) using the DIG-labeling mixture (Boehringer Mannheim). The 5-bromo-4-chloro-3-indolyl phosphate (BCIP; Sigma) and nitro blue tetrazolium (NBT; Sigma) were used for chromogenic reactions.

Mouse Embryos and In Situ Hybridization

ICR mice were used in this study. The morning of vaginal plug was defined as 0.5 dpc. Staging was according to Kaufman (1992). Whole-mount in situ hybridization protocol has been derived from previously published methods (Nieto et al., 1992; Bao and Cepko, 1997; Dymecki et al., 2002). Embryos were fixed in 4% paraformaldehyde in PBS for 6 hr at 4°C, incubated in 6% hydrogen peroxide in PBS for 1 hr, rinsed, and incubated with 10 μg/ml proteinase K in PBS for 6–9 min. Embryos were then refixed; prehybridized in 50% formamide, 5× standard saline citrate, 1% sodium dodecyl sulfate, 100 μg/ml yeast tRNA, and 50 μg/ml heparin for 2 hr at 70°C; and hybridized in the same buffer, containing 1 μg/ml DIG-labeled probe, overnight at 70°C. Posthybridization washes were as described (Nieto et al., 1992; Bao and Cepko, 1997; Dymecki et al., 2002). After washing, embryos were incubated in 1% blocking reagent (Roche) in TBST with 2 mM levamisole for 1 hr, after which they were incubated overnight at 4°C in fresh blocking solution containing anti-DIG antibody coupled to alkaline phosphatase (Roche, 1:1,000 dilution). Embryos were washed and stained in NTMT (100 mM NaCl, 100 mM Tris-HCl pH 9.5, 50 mM MgCl2, 0.1% Tween 20), containing 4.5 μl/ml NBT and 3.5 μl/ml BCIP.

For embryo sections, embryos were fixed in 4% paraformaldehyde for 3 hr, equilibrated in 30% sucrose in PBS at 4°C, frozen in O.C.T. (Tissue-Tek), and sectioned at 30 μm. Sections were air-dried on Superfrost Plus slides (VWR), re-fixed in 4% paraformaldehyde in PBS for 10 min, and incubated in the following solutions: PBS for 5 min, 1 μg/ml proteinase K in PBS for 5 min, 2 mg/ml glycine in PBS for 5 min, PBS for 5 min, 4% paraformaldehyde in PBS for 10 min, 0.25% acetic anhydride in 0.1 M triethanolamine for 10 min, and PBS for 5 min. Hybridization, washes, and color development were performed essentially as described above.

Acknowledgements

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

We thank Tara Huber and Miyuki Hikasa for helpful comments on this manuscript. This study was supported by NIH grants to S.S. and S.D.

REFERENCES

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