|
FIG. 1.XHex and cerberus are early markers of the anterior endomesoderm. In situ hybridisation to adjacent serial midsagittal sections of a stage 10ï°‚ gastrula embryo. One embryo probed with (A) XHex, cerberus, XSox17b, and Xbra and another embryo probed with (B) XHex, cerberus, chordin, and goosecoid. The dorsal lip is oriented to the bottom left in each. XSox17b marks the future endoderm, Xbra marks the future mesoderm, with chordin and goosecoid marking both prechordal mesoderm and some endodermal cells. A group of cells in the deep organiser expressing XHex, cerberus, goosecoid, and chordin but not XSox17b or Xbra is indicated by the white arrowhead. cerberus-expressing cells at the ventral mesendoderm boundary are indicated by the black arrowhead. The approximate endoderm/ mesoderm boundary based on XSox17ï°€ expression is indicated by the red line. Notice XHex and cerberus are expressed in deep endoderm cells. (C) Cleared whole-mount in situ hybridisation of XHex and cerberus; both embryos are in vegetal view with the dorsal lip uppermost. Note that cerberus is expressed in a broader domain than XHex. (D) In situ hybridisation of XHex and cerberus to sectioned stage 9 blastulae shows that these genes are asymmetrically expressed in the vegetal-equatorial region on one side of the embryo prior to gastrulation.
|
|
FIG. 6. Dorsal Beta-catenin, TGF-Beta, and BMP-antagonist signalling pathways are required for endogenous XHex and cerberus expression. We tested whether the following signalling pathways are required for XHex and cerberus expression: (1) the maternal Wnt/beta-catenin pathway, (2) the dorsalising TGF-Beta signals, and (3) the dorsalising BMP antagonists noggin and chordin. Each of the pathways was blocked by the overexpression of dominant-inhibitory components of the respective pathways. Synthetic mRNA encoding deltaN-XTcf-3, 0.5 ng; XTcf-3, 0.5 ng; dnActRIIB, 1 ng; ActRBII, 1 ng; or BMP4, 1 ng was microinjected subequatorially into the two dorsal blastomeres at the 4-cell stage. At stage 10 0.5 embryos were subjected to whole-mount in situ hybridisation with XHex, cerberus, Xbra, and chordin probes. Representative embryos are shown in each case. The mRNA injected is indicated along the left and the in situ probe is indicated on the top. Inhibition of Beta-catenin or TGF-Beta signalling as well as over-expression of BMP4 blocked endogenous XHex and cerberus expression. Control injections of ActRIIB and XTcf-3 had no effect.
|
|
cer1 (cerberus) gene expression in bisected Xenopus laevis embryo, mid-sagittal section, assayed via in situ hybridization, NF stage 10.25, dorsal lip bottom left. (Black arrow indicates the ventral mesendoderm boundary)
|
|
FIG 2. Anterior endomesoderm is specified by blastula stage. (A) Schematic diagram showing the experimental design. At the 4-cell stage, the dorsal blastomeres, which are lighter and smaller, are marked with Nile blue vital dye. At stage 7–8 the dorsal-vegetal tissues are explanted, free of mesoderm, and cultured until stage 10.5 when they are subjected to RT-PCR. (B) RT-PCR analysis of XHex, cerberus, XSox17Beta, gsc, Xbra, and Ef-1alpha expression in whole embryos stage 8 (WE st8) and stage 10.5 (WE st10.5), dorsal endoderm explants at stage 10.5 (DE), and a whole embryo stage 10.5 control cDNA with no reverse transcriptase (WE -RT). Mesoderm-free dorsal-vegetal explants that do not express Xbra do express XHex or cerberus.
|
|
FIG. 3. Cell– cell signalling is required for anterior endomesoderm formation. (A) 4- to 8-cell-stage embryos were dissociated in Ca/Mg-free medium. Dissociated cells were maintained in a dispersed state by frequent agitation until sibling embryos reached stage 10.5 when the cells were harvested and subjected to RT-PCR. (B) A representative RT-PCR analysis is shown. XHex mRNA is not detected in dispersed cells (DissE), while cerberus and XSox17Beta expression is dramatically down-regulated. As control markers, Xbra expression has been shown to require cell contact, while Xwnt8 and gsc are expressed in a cell-autonomous manner. Stage 10.5 whole-embryo (WE) controls of cDNA with or without reverse transcriptase (+RT, -RT) are shown. Development of the AE requires communication between cells.
|
|
FIG. 4. Maternal Wnt/B-catenin and TGF-B pathways can induce XHex and cerberus expression. Candidate dorsalising molecules were tested for the ability to induce ectopic XHex and cerberus expression. (A) At the 4-cell stage, each of the two ventral blastomeres was injected with the following synthetic mRNA: B-catenin, 200 pg; siamois, 5 pg; bVg1, 100 pg; tBR, 250 pg; noggin, 100 pg; dnXwnt8, 250 pg; tBR+dnXwnt8, 250 pg each; Xnr1, 50 pg; Xnr2, 50 pg; Xnr3, 200 pg. Ventral blastomeres were marked with Nile blue vital dye to identify the injected tissue. At stage 10–10.5 the ventral endoderm tissue was explanted and (B) assayed by RT-PCR for XHex, cerberus, XSox17Beta, siamois, chordin, gsc, vent-1, and Ef-1alpha. Ectopic expression of XHex and cerberus was induced by Beta-catenin, siamois, bVg1, Xnr1, and Xnr2. A slight up-regulation of cerberus, but not XHex, was observed upon tBR and noggin injection. Stage 10.5 whole embryo (WE) controls of cDNA with or without reverse transcriptase (+RT, -RT) are shown
|
|
FIG. 5. Induction of XHex and cerberus in animal cap cells. (A) Synthetic mRNA was injected into the animal pole of 2-cell embryos. At stage 8–9 animal cap tissue was explanted and cultured until sibling embryos reached stage 10.5, when they were subjected to RT-PCR. (B) Embryos were injected with increasing doses of bVg1 mRNA (2, 20, 200 pg), siamois mRNA (10, 30, 100 pg), or noggin mRNA (10, 50, 200 pg). bVg1 induced XSox17ô°€, XHex and cerberus, while siamois induced only XHex expression. (C) Animal caps injected with XSox17ô°€ mRNA (500 pg) express cerberus but not XHex. (D) Animal caps injected with VegT/apod mRNA (500 pg) express XSox17ô°€ and cerberus, but not XHex. Stage 10.5 whole embryo (WE) controls of cDNA with or without reverse transcriptase (ô°‚RT, ô°ƒRT) are shown.
|
|
FIG 7. XHex can induce the expression of cerberus. The interre- lationship between XHex and cerberus expression was assayed by the injection of synthetic mRNA encoding either XHex (10, 100, 250 pg) or Cerberus (10, 100, 250 pg) into each ventral blastomere at the 4-cell stage, similar to the experimental design of Fig. 4. At stage 10.5, ventral endoderm tissue was explanted and assayed by RT-PCR for the expression of the other gene. XHex can induce cerberus, but Cerberus cannot induce XHex at the doses tested. Stage 10.5 whole-embryo (WE) controls of cDNA with or without reverse transcriptase (+RT, -RT) are shown.
|
|
FIG. 8. Model of anterior endomesoderm development. Based on our results we propose the following model of anterior endomesoderm development. (A) Spatial arrangement of signalling domains in the embryo (V, ventral, and D, dorsal). (B) Signalling molecules involved. The model is presented in three temporal phases indicated along the bottom: (1) maternally acting factors, (2) zygotic factors acting after the midblastula transition (MBT) during the blastula stage, and (3) signals acting during the gastrula stage. The combination of the maternal Wnt/ô°€-catenin (dorsal, black) and vegetal/endodermal TGF-ô°€ (grey) signals initially determines the site of the future AE, where their activities overlap (chequered). After the MBT, siamois mediates the ô°€-catenin signal, and factors such as Xnr1-2, XSox17ô°/ô°€, and Mixer (and possibly the continued activity of maternal Vg1; dashed arrow) mediate the endoderm-specific signals. As a result XHex and cerberus transcription is activated in the AE. Within the AE, the activity of the XHex protein promotes cerberus expression. Finally, at the gastrula stage, noggin and chordin secreted from the organiser (stippled, dorsal) promote the expression of XHex and cerberus by inhibiting the effects of ventral BMP4 signals (stippled, ventral).
|
|
cer1 (cerberus) gene expression in bisected Xenopus laevis embryo, assayed via in situ hybridization, NF stage 10.25, vegetal view, dorsal up.
|
