Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
???displayArticle.abstract???
One of the earliest lineage restriction events in embryogenesis is the specification of the primary germ layers: ectoderm, mesoderm and endoderm. In Xenopus, germ layer specification occurs prior to gastrulation and requires the transcription factor VegT both for the cell-autonomous specification of endoderm and the generation of mesoderm-inducing signals. In the absence of VegT, ectoderm is formed in all regions of the embryo. In this work, we show that VegT-depleted vegetal cells (prospective endoderm) behave like animal cells in sorting assays and ectopically express early markers of ectoderm. To gain insight into how ectoderm is specified, we looked for candidate ectoderm-specific genes that are ectopically expressed in VegT-depleted embryos, and examined the role of one of these, the LIM homeobox gene Xlim5, in ectoderm development. We show that overexpression of Xlim5 in prospective endoderm cells is sufficient to impair sorting of animal cells from vegetal cells but is not sufficient (at similar doses) to induce ectoderm-specific genes. In whole embryos, Xlim5 causes vegetal cells to segregate inappropriately to other germ layers and express late differentiation markers of that germ layer. Inhibition of Xlim5 function using an Engrailed repressor construct or a morpholino oligonucleotide causes loss of animal cell adhesion or delay in neural fold morphogenesis, respectively, without significantly affecting early ectoderm gene expression. Taken together, our results provide evidence that a primary role for Xlim5 is to specifically regulate differential cell adhesion behaviour of the ectoderm.
Fig. 2. Expression of the LIM-homeobox gene Xlim5 is induced in VegT-depleted embryos and repressed by Xnr2. (A) Vegetal mass explants (Vg exp.) were dissected from stage 9 embryos and cultured to either stage 10 (top) or stage 12 (bottom) along with sibling whole embryos (WE) prior to processing for real-time RT-PCR. Relative expression levels for each gene were determined by comparison to a standard curve generated by serial dilution (100%-10%) of uninjected stage 12 controls. Expression levels of all genes were normalized to the level of ornithine decarboxylase (ODC) prior to quantitation (not shown). Xlim5 expression is increased in both VegT-depleted whole embryos and vegetal explants. Known ectoderm markers Epidermal keratin and E-cadherin are included as controls and are also upregulated. (B) Xlim5 expression is inhibited by Xnr2 overexpression. VegT-depleted embryos and VegT-depleted embryos rescued with 60-600 pg of Xnr2 RNA were analysed by real-time RT-PCR for Xlim5 expression at stage 12. This cDNA was previously generated for Kofron et al. (Kofron et al., 1999).
Fig. 1. VegT-depleted vegetal cells fail to sort from control animal cap and control vegetal cells. Control vegetal cells injected with RLDX were dissected from stage 9 embryos, dissociated and mixed with unlabelled dissociated animal cap cells (A) or dissociated vegetal cells (B). Aggregates were fixed after 4 hours in culture and viewed by confocal microscopy. Control vegetal cells sort out from animal cells (A) but not from other vegetal cells (B). (C,C′,D,D′) Dissociated, RLDX-labelled VegT-depleted vegetal cells from stage 9 embryos were mixed with unlabelled control animal cells (C,C′) or unlabelled control vegetal cells (D,D′). VegT-depleted vegetal cells fail to sort out in either case and remain randomly distributed.
Fig. 3. Xlim5 impairs the sorting of vegetal cells from animal cells. Dissociated animal cap cells injected with RLDX were mixed with uninjected (A-A′) or Xlim5-injected (1 ng) (B-B′) vegetal cells and allowed to reaggregate. Aggregates were cleared and viewed by confocal microscopy and scored for sorting. A summary of two experiments is shown in C. Dark grey, positive sorting; white, no sorting. (D,E) Timelapse movies of control animal and vegetal cells in sorting assays. (D,D′) Animal cells labelled with TRITC were aggregated with unlabelled vegetal cells and filmed by timelapse video microscopy. Two animal cells (arrows and arrowheads) are shown moving through the aggregate from a time point ∼30 minutes into filming (D) to a time point 3 minutes later (D′). TRITC-labelled vegetal cells aggregated with unlabelled animal cells and filmed over a similar time course do not show any translocation through the aggregate (E,E′).
Fig. 4. Xlim5 does not change germ layer-specific gene expression in vegetal endoderm explants. Vegetal masses were dissected from stage 9 embryos either uninjected (Un Vg) or injected with 1 ng of Xlim5 RNA (Xlim5 Vg) and cultured until stage 22. Explants were assayed by real-time RT-PCR as described above. Note that ectoderm (NCAM, Epidermal keratin and E-cadherin) and mesoderm (Muscle actin) are not induced in Xlim5-injected explants. Endoderm markers [Endodermin (Edd) and Xsox17a] are not significantly affected. Relative expression was determined versus stage 22 whole embryos.
Fig. 5. (A,B) Overexpression of Xlim5 causes vegetal cells to enter other germ layers. (A) Uninjected embryos and (B) embryos injected vegetally with 4 ng Xlim5 RNA. Note ventral patches of pigmented animal cap cells. (C-H) X-gal staining of tailbud stage embryos injected with Xlim5 in a vegetal blastomere at the 32-cell stage. Arrows show ectopic locations of Xlim5-injected cells in D, F and H. Control embryos injected withβ -gal alone are shown in C, E and G. Lineage labelling results are summarized in Table 2. (I-N) X-gal staining of late gastrula (I,J) and early neurula stage (L-N) embryos injected with Xlim5 in a vegetal blastomere at the 32-cell stage. Arrows show ectopic locations of Xlim5-injected cells in (L,N). Control embryos injected with β-gal alone are shown in I, K and M. Arrowhead in H indicates staining in the ectoderm.
Fig. 6. Ectopic Xlim5-expressing cells express a marker of mature somites. Embryos injected with either β-gal RNA alone (A) orβ -gal + Xlim5 RNA (B) vegetally at the 32-cell stage were fixed at the tailbud stage and immunostained after cryosectioning. Sections were stained with mAb 12/101 (somite, red) along with an antibody against β-galactosidase (green). Embryos injected withβ -gal alone show non-overlapping staining of 12/101 withβ -gal (arrowhead) while a population of Xlim5-co-injected cells shows colocalization of the two antibodies (arrow, yellow). The slight green staining in the epidermis is due to background staining.
Fig. 7. Inhibition of Xlim5 interferes with normal ectoderm development. (A-C) Injection of Xlim5-EnR causes cell dissociation during gastrulation. (A) Uninjected stage 10.5 embryos. (B) Sibling embryos injected with 1 ng Xlim5-EnR RNA. Note the number of embryos with non-uniform pigment and white patches indicating dissociated cells. (C) Rescued embryos co-injected with 1 ng Xlim5-EnR + 1 ng Xlim5. Embryos have recovered normal adhesion. (D,E) Injection of Xlim5-EnR into animal blastomeres inhibits ectoderm adhesion. (D) Injection of β-gal into a ventral-animal blastomere at the 32-cell stage. Labelled cells populate the epidermis (scattered blue cells). (E) Co-injection of β-gal with 100 pg Xlim5-EnR. Cells are lost from the epidermis and form clumps in the pharynx and gut (arrows). (F) Uninjected (left, blue) and Xlim5-EnR-injected embryos (right, red) obtained by the host transfer technique showing dissociation of animal and equatorial cells but not vegetal cells. (G,H) Control (G) or Xlim5-EnR-injected (H) animal cap cells were dissociated and reaggregated in OCM. Large aggregates are formed in the controls but are absent in the Xlim5-EnR-injected cells. (I-K) Sytox Green staining of dissociated cells. Uninjected (I) and Xlim5-EnR-injected cells (J) do not stain with Sytox Green, whereas positive control dead cells (K) stain brightly. (L) Depletion of Xlim5 with an antisense MO delays neural fold morphogenesis. (Top) Uninjected control embryos at stage 18. (Middle) Sibling embryos injected with 40 ng Xlim5-MO. Notice open anterior neural folds (arrow). (Bottom) MO-injected embryos injected with 1 ng Xlim5 RNA. Neural fold closure is rescued in these embryos. (M) Ectoderm marker gene expression in Xlim5-MO injected embryos assayed by real-time RT-PCR. UN, uninjected stage 13 embryos; MO, 40 ng Xlim5-MO; MOR, 40 ng Xlim5-MO + 1 ng Xlim5 RNA.
