October 1, 2001;
Siamois functions in the early blastula to induce Spemann''s organiser.
In Xenopus, the Spemann organiser is defined as a dorsal territory in the early gastrula
that initiates development of the embryonic axis. It has been shown that the early zygotic transcription factor Siamois
is essential for Spemann''s organiser formation. By the onset of gastrulation, the organiser is patterned into a vegetal head
organiser, which induces anterior
structures upon transplantation, and a more animal trunk
organiser, which induces a posterior
neuraxis. However, it is unclear when these distinct organiser domains are initially specified. To shed light on this question, we analysed the temporal activity of Siamois
, as this factor induces both head
development, when ectopically expressed via mRNA injection. In this study, we expressed Siamois
ectopically at different time points and analysed the extent of axial development. Using a hormone-inducible version of Siamois
, we found evidence for a tight window of competence during which ventral
cells can respond to Siamois
by commencing both the head
and the trunk
genetic programmes. The competence to form Spemann''s organiser was lost 2 h before gastrulation, although partial axis formation could still occur following delayed activation of Siamois
. We demonstrate that this late response to Siamois
involves a new role for this gene, which can indirectly repress ventral
gene expression, in the absence of known organiser genes.
[+] show captions
Fig. 3. Organiser gene expression in GR-Sia injected embryos induced at various stages. Embryos were injected ventrally with 50 pg/bl GR-Sia mRNA and 400 pg/bl LacZ RNA at the four-cell stage and DEX (10 μM) was either omitted (−DEX), or added 15 min before stage 9 (9−), or at stage 9.5. β-Galactosidase activity was revealed in red, and embryos were then processed for in situ hybridisation. For dkk1 and frzb, embryos were manually bisected prior to hybridisation. In all panels, the natural dorsal side is right. For otx2, the animal view (otx2/an) reveals the extent of the anterior neural plate. All dorsal genes examined, belonging to head and trunk classes, were robustly induced when DEX was added at stage 9−, while treatment at stage 9.5 provoked no or very weak induction.
Fig. 4. Delayed activation of Siamois leads to repression of ventral gene expression. Embryos were injected ventrally with 50 pg/bl GR-Sia mRNA and 400 pg/bl LacZ mRNA at the four-cell stage and DEX (10 μM) was either omitted (−DEX), or added 15 min before stage 9 (9−), or at stage 9.5. β-Galactosidase activity was revealed in red, and embryos were then processed for in situ hybridisation. In all panels, dorsal is up. Note that the ventral genes PV.1 and Xwnt8 were repressed by GR-Sia activated at stage 9− or at stage 9.5. However, PV.1 expression was repressed several cell diameters away from the injected cells upon early DEX treatment (bracket), whereas it was found adjacent to the injected cells upon late DEX treatment.
Fig. 1. Late blastula expression of Siamois is not sufficient to promote anterior development. (A,B) Embryos were injected at the four-cell stage with 20 pg/blastomere (bl) Siamois mRNA. (C,D) Embryos were injected at the four-cell stage with 100 pg/bl CSKA-Siamois plasmid DNA. In (B,D), cross-sections were counterstained with hematoxylin/eosin. Note that embryos injected with CSKA-Siamois plasmid DNA did not develop a head (100% of negative cases) but occasionally developed a notochord (50% of positive cases). (E) RNAse protection analysis of animal explants injected with the indicated amount of RNAs or 100 pg/bl CSKA-Siamois plasmid DNA at the two-cell stage. Animal caps (20 per probe) were dissected at stage 9 and incubated up to stage 28. The absence of mesoderm in the explants was assessed by the lack of muscle actin specific signal. FGFR1 expression served as a loading control. Siamois RNA but not DNA induced the anterior markers XCG1 (cement gland) and otx2 (forebrain/midbrain) as well as the neural cell adhesion marker N-CAM in a manner comparable to the neural inducer noggin (Nog). Siamois did not induce the en2 (midbrain/hindbrain boundary) and Hoxb9 (spinal cord) markers.
Fig. 2. Time-window of competence to respond to Siamois. (A) Schematic representation of the GR-Sia expression construct. (B) Embryos were injected ventrally with 50 pg/bl GR-Sia mRNA at the four-cell stage, and DEX (10 mM) was added at stage 6 (32-cell) or stage 9.25, and embryos allowed to develop to tadpoles (left). Note that induction of Siamois activity from stage 9.25 was not sufficient to allow for head formation. Hematoxylin–eosin staining on cross-sections revealed the formation of a full set of duplicated dorsal structures including a notochord in embryos treated with DEX at stage 6 (6/6), while embryos treated at stage 9.25 typically lacked a notochord in the secondary axis (11/12) (right). (C) Bar graph showing a compilation of three independent experiments where the activity of GR-Sia is represented as a function of time of addition of DEX. Duplicated axes were scored complete when they showed a cement gland and at least one eye. Note that head induction did not occur past stage 9 of the blastula, while trunk duplication was still pronounced 2-h later, at the onset of gastrulation (stage 10). n = total number of injected embryos.
ventx1.1 (VENT homeobox 1, gene 1) gene expression in Xenopus laevis embryos, NF stage 11, as assayed by in situ hybridization. vegetal view: dorsal up.