|
Fig. 1. Notch restricts the expression of blastula chordin- and noggin-expressing centre (BCNE) markers to the dorsal region. Expression of chd (A-F) and Xnr3 (G-L) in stage 9 Xenopus embryos. (A,B,G,H) Embryos injected with 1 ng of nicd-mt mRNA into a dorsal cell at the 4-cell stage (stage 3; B,H) and control siblings (A,G). Brown dots correspond to horseradish peroxidase immunolocalisation of the c-myc epitope fused to the intracellular domain of Notch (NICD). (C,D,I,J) Embryos injected with 20 ng of control morpholino (Mo; C,I) or Notch Mo (D,J) into a dorsal cell at stage 3. The injected side was revealed by alkaline-phosphatase immunostaining of the fluorescein tag (BCIP, turquoise staining in C,D,J) or by immunofluorescence (inset in I). Red and yellow dotted lines depict the area of expression of the BCNE markers on the non-injected and on the injected side, respectively. (E,F,K,L) Hemisectioned embryos injected with 20 ng of control Mo (E,K) or Notch Mo (F,L) before the first cleavage. The dorsal side is oriented to the right, animal up. Insets in E, F, K and L show the corresponding whole embryos. Whole embryos are shown in dorsal views, animal up.
|
|
Fig. 2. Notch impairs the ability of β- and δβ-catenin to induce ectopic expression of the blastula chordin- and noggin-expressing centre (BCNE) marker chd at the ventral side. (A-D) Expression of the BCNE marker chd in whole Xenopus embryos at stage 9. Embryos were injected into one ventral blastomere at the 4-cell stage with the following mRNAs: 1 ng of β-catenin-GFP (A), 1 ng of β-catenin-GFP + 1 ng of nicd-mt (B), 1 ng of δβ-catenin-GFP (C) or 1 ng of δβ-catenin-GFP + 1 ng of nicd-mt (D). All embryos are shown in ventral views. The upper insets show the immunofluorescence revealing the DOG tracer at the ventral side. The lower insets show the corresponding dorsal views with the normal expression of chd in the BCNE. (E,F) Percentage of embryos expressing (blue bars) or not expressing (red bars) ectopic chd at the ventral side. (E) Comparison of β-catenin-GFP (left) and β-catenin-GFP + nicd-mt injections (right). (F) Comparison between δβ-catenin-GFP (left) and δβ-catenin-GFP + nicd-mt injections (right).
|
|
Fig. 3. Notch gain of function produces ventralised phenotypes. Xenopus embryos were injected with 1 ng of nicd-mt RNA before the first mitotic division and were allowed to develop until stage 28, when they were scored according to the dorsoanterior index (DAI) (Kao and Elinson, 1988). DAI 5, normal; DAI 4, reduced eyes and forehead; DAI 2, microcephalic; DAI 1, acephalic; DAI 0, bauchstück. The percentage corresponding to each phenotype is shown at the left (n=47). (A,Aâ²) Lateral (A) and frontal (Aâ²) views of a sibling control embryo hybridised with the cephalic marker otx2 (Pannese et al., 1995; Blitz and Cho, 1995) and the posterior marker hoxb7 (López and Carrasco, 1992; Godsave et al., 1994). (B-Dâ²) Lateral (B-D) and frontal (Bâ²-Dâ²) views of embryos injected with nicd-mt RNA and hybridised with otx2 and hoxb7. Note the loss of cephalic structures at different grades, as revealed by otx2 expression. (E) An extremely ventralised embryo injected with nicd-mt mRNA.
|
|
Fig. 4. Notch abolishes the dorsalising activity of β-catenin. (A,C,D) External phenotypes of Xenopus tadpoles at stage 35. A shows a control tadpole. Sibling embryos were injected into one ventral blastomere at the 4-cell stage with 1 ng of β-catenin-GFP mRNA (C) or with 1 ng of β-catenin-GFP mRNA + 1 ng of nicd-mt mRNA (D) and fixed at the tadpole stage (stage 35). Embryos were scored according to the dorsoanterior index (DAI) (Kao and Elinson, 1988). Dorsalised phenotypes also include double heads and incomplete secondary axis. Most embryos injected with β-catenin-GFP mRNA were dorsalised (82%, n=74; the percentage of each dorsalised phenotype is shown in orange; C). Most embryos injected with 1 ng of β-catenin-GFP mRNA + 1 ng of nicd-mt mRNA were indistinguishable from sibling controls (82%, n=92; percentage in orange in D). (B) Frequency distribution of normal, dorsalised or ventralised phenotypes, expressed as percentage of injected embryos for β-catenin-GFP mRNA alone (blue bars, n=74) or β-catenin-GFP mRNA + 1 ng of nicd-mt mRNA (red bars, n=92), as described for C and D.
|
|
Fig. 5. Notch decreases the steady state levels of exogenous β-catenin. (A-L) Xenopus embryos were injected with 1 ng of β-catenin-GFP mRNA (A,E,I), 1 ng of β-catenin-GFP mRNA + 1 ng of nicd-mt mRNA (B,C,F,G,J,K) or 1 ng of nicd-mt mRNA alone (D,H,L). Embryos were fixed at stage 6.5 (A-D), 7-8 (E-H) or 9 (I-L). They were processed for immunohistochemistry with antibodies against GFP (A,B,E,F,I,J) or c-myc (C,D,G,H,K,L) and revealed with HRP/DAB (A,B,D-F,H-J,L; brown) or alkaline phospatase/BCIP (C,G,K; turquoise). All embryos are shown in animal views. E, G, H, I, K and L are magnifications of the embryos shown in the insets. The arrowhead in D points to the nuclear localisation of NICD-MT. The dotted yellow line in G and K depicts the intercellular boundary around two neighbouring animal cells. Notice the turquoise foci revealing the localisation of NICD-MT at both sides of the boundary, in cell-cell junctions. (M) Percentage of embryos expressing the recombinant β-catenin-GFP protein (a-GFP+, blue bars) or not expressing it (a-GFPâ, red bars). Comparison of β-catenin-GFP (left) and β-catenin-GFP + nicd-mt injections (right). (N) Percentage of embryos with nuclear (left) or without nuclear expression of the recombinant NICD-MT protein, which is instead relocated to cell-cell junctions (right). Comparison of β-catenin-GFP + nicd-mt injections (blue bars) and nicd-mt injections (red bars).
|
|
Fig. 6. Endogenous Notch can decrease the steady state levels of β-catenin before midblastula transition (MBT) through a mechanism that does not require its phosphorylation by GSK3. (A-F) Immunodetection of the GFP tag of β-catenin-GFP (A,B) or the GFP tag of δβ-catenin-GFP (C-F). Xenopus embryos were injected before the first cleavage with 1 ng of β-catenin-GFP mRNA + 20 ng of control morpholino (Mo; A), 1 ng of β-catenin-GFP mRNA + 20 ng of Notch Mo (B), 1 ng of δβ-catenin-GFP mRNA (C), 1 ng of δβ-catenin-GFP mRNA + 1 ng of nicd-mt mRNA (D), 1 ng of δβ-catenin-GFP mRNA + 20 ng of control Mo (E) or 1 ng of δβ-catenin-GFP mRNA + 20 ng of Notch Mo (F). All embryos were co-injected with DOG as tracer and classified by fluorescence. Animal views of embryos fixed at early blastula (stage 6.5-7) were photographed in PBS.
|
|
Fig. 7. Notch modulates the levels and the intracellular distribution of endogenous β-catenin. (A-L) Immunolocalisation of endogenous β-catenin in whole Xenopus embryos. Embryos were injected before the first mitotic division with 1 ng of nicd-mt mRNA (B,D), 20 ng of control morpholino (Mo; E,F,I,J) or 20 ng of Notch Mo (G,H,K,L). Injected embryos and their corresponding uninjected siblings (A,C) were fixed at early blastula (A,B,E-H) and late blastula (C,D,I-L) and were processed for whole-mount immunohistochemistry with an antibody raised against the last C-terminal 14 amino acids of β-catenin (brown). All embryos are shown in animal views, with the dorsal side oriented to the right, except in B, where it was not possible to assign the dorsal-ventral (D-V) orientation. F, H, J and L are magnifications of the area enclosed by the rectangular frame in E, G, I and K, respectively. Embryos were photographed in PBS (A-D) or in 75% Murray's solution in methanol (E-L). Uninjected controls (A,C) or embryos injected with control Mo (E,I) show a D-V gradient of β-catenin immunoreactivity, with fainter levels ventrally (yellow asterisks) and maximum levels dorsally (red asterisks). Notch Mo increased the levels of β-catenin protein on the ventral side relative to control Mo-injected siblings (compare yellow asterisks in E,F and G,H for stage 6.5 and in I,J and K,L for stage 9). Black and white arrows indicate the distribution of β-catenin in the peripheral cytoplasm and in the nuclei, respectively.
|
|
Fig. 8. Attenuation of Notch from the beginning of development produces an expansion of the brain, whereas gain of function of Notch produces the opposite results. (A-L) Expression of anterior-posterior (A-P) neural markers at stage 14 (A,B,E,F,I,J) or stage 15 (C,D,G,H,K,L). Xenopus embryos were injected prior to the first mitotic cleavage in the animal hemisphere with 20 ng of control morpholino (Mo; A,E, left side of I), 20 ng of Notch Mo (B,F,J, right side of I) or 1 ng of nicd-mt RNA (D,H,L). Embryos injected with control Mo and Notch Mo are siblings. C, G and K are the corresponding uninjected sibling controls of D, H and L, respectively. Specimens were processed for whole-mount in situ hybridisation for the following markers: otx2 (A-D); Xanf1, en2 and hoxb7 (E-L). A-H show anterior views, dorsal up. I-L show dorsal views, anterior up. The inset in D shows the expression of otx2 in a more severely affected nicd-mt RNA-injected embryo. I is a photocomposition showing a control Mo (Co-Mo)-injected embryo (left) and a Notch Mo (N-Mo)-injected sibling (right). F, right side of I and J show Notch Mo-injected embryos displaying different grades of expansion of the Xanf1 territory, with gradual extinction of the en2 stripes and caudal shift of the hoxb7 domain. Asterisks in A-C highlight the expression hole encircled by otx2, coincident with the area of expression of Xanf1 (asterisks in E-G). Red arrowheads in H indicate the reduction of the Xanf1 domain in nicd-mt RNA-injected embryos relative to control siblings (G, green arrowheads). Red arrowheads in I and J indicate the posterior shift of the hoxb7 domain in Notch Mo-injected embryos relative to control Mo-injected siblings (I, green arrowhead). The red arrowhead in L indicates the anterior shift of the hoxb7 domain in nicd-mt RNA-injected embryos relative to control siblings (K, green arrowheads). shcg, stomodeal-hypophyseal and cement gland anlage; t, presumptive telencephalon; d, presumptive diencephalon; m, presumptive mesencephalon; m/h, presumptive boundary between midbrain and hindbrain; h, presumptive hindbrain; sc, presumptive spinal cord.
|
|
Fig. 9. Attenuation of Notch from the 1-cell stage onwards enhances dorso-anterior development. Sibling Xenopus embryos were injected prior to the first cleavage with 20 ng of control morpholino (Mo) or 20 ng of Notch Mo. (A) External morphology of stage 35 tadpoles injected with control Mo (upper embryo) or Notch Mo. Three examples of enhanced dorso-anterior development produced by Notch Mo are shown below the control Mo-injected tadpole. (B,C) Expression pattern of otx2 at stage 28 in a control Mo-injected tailbud (B) and a Notch Mo-injected sibling (C), shown in frontal views. The dotted red line in B depicts the contour of cephalic structures marked by otx2 in a control Mo-injected embryo, including the prospective telencephalic domain (t, encircled by a yellow dotted line), the prospective eye domains (e, encircled by yellow dotted lines) and the prospective mesencephalic domain (m). These domains were projected on the Notch Mo-injected sibling shown in C, taking the telencephalic otx2 domain as reference, to emphasise the expansion of the most cephalic structures. (D) Lateral view of a control Mo-injected tailbud (upper embryo) and a Notch Mo-injected sibling (lower embryo) at stage 28, showing the expression of hoxb7, aligned with respect to the anterior limit of hoxb7 (vertical yellow line). The left arrowheads point to the cement gland, and the right arrowheads to the tip of the tail. Head length (h, white line) was measured from the cement gland to the hoxb7 anterior limit. Trunk length (t, blue line) was measured from the hoxb7 anterior limit to the tip of the tail. (E) Notch Mo-injected tailbuds (red bar) display a significantly higher head/trunk length ratio than control Mo-injected siblings (blue bar). P<0.0001, two-tailed t-test.
|
|
Fig. 10. Model for Notch function during early development of Xenopus. (A-C) Normal development of a Xenopus embryo injected with control Mo, showing β-catenin dynamics at cleavage or early blastula stages (A), the blastula chordin- and noggin-expressing centre (BCNE) at late blastula (B) and the anterior-posterior (A-P) pattern of the neural plate at early neurula (C). During cleavage stages or early blastula, β-catenin is normally degraded at the ventral side but accumulates in the nuclei of the dorsal side, forming a dorsal-to-ventral gradient, as illustrated in green (A). For simplicity, only the GSK3 and Notch pathways are shown (in red and orange, respectively). GSK3 phosphorylates serine residues 33 and 37 of β-catenin. Phosphorylated β-catenin (β-P) is targeted for fast degradation by the ubiquitin-proteasome system (red dotted lines). Notch targets hypophosphorylated β-catenin (β) for degradation, probably by the endocytic-lysosomal pathway (orange dotted lines). Nuclear β-catenin activity in the dorsal animal/marginal cells of the blastula promotes the formation of the BCNE (purple area in B). The whole forebrain and most of the midbrain and the hindbrain derive from the BCNE (C). (D-F) Changes observed in β-catenin dynamics at cleavage or early blastula stages (D), in the extent of the BCNE at late blastula (E) and in the A-P pattern of the neural plate at early neurula (F) when Notch is attenuated by injection of Notch Mo prior to the first cleavage. When Notch is attenuated, hypophosphorylated β-catenin is further stabilised in more ventral locations (D). This results in a ventral expansion of the BCNE at late blastula (E) and BCNE derivatives are enlarged in the early neurula (F). C and F are diagrams of embryos in anterior-dorsal views, summarising the results shown in Fig. 8A-F. An, animal; Veg, vegetal; D, dorsal; V, ventral; shcg, stomodeal-hypophyseal and cement gland anlage; t, presumptive telencephalon; d, presumptive diencephalon; m, presumptive mesencephalon; m/h, presumptive boundary between midbrain and hindbrain; h, presumptive hindbrain; sc, presumptive spinal cord.
|
|
Fig. S1. Attenuation of Notch from stage 3 induces weak ectopic expression of blastula chordin- and noggin-expressing centre (BCNE) markers at the ventral side. (A-F) Xenopus embryos were injected with 20 ng of Notch morpholino (Mo) into one ventral cell at stage 3. Expression of chd (A-C) and Xnr3 (D-F) at stage 9 is shown in dorsal (A,D) and ventral (B,E) view. B and E are ventral views of the embryos shown in A and D, respectively, with ectopic chd (B) or Xnr3 (E) expression. These embryos represent the strongest phenotypes. Green fluorescence of the co-injected Dextran Oregon Green tracer (DOG; C,F) correspond to the same views shown in B and E.
|
|
Fig. S2. Attenuation of CSL-mediated Notch signalling does not result in an expansion of BCNE markers on the animal hemisphere. (A-L) Expression of chd (A,B,E,F,I,J) and Xnr3 (C,D,G,H,K,L) at stage 9 in Xenopus embryos injected with 2 ng of su(H)1DBM mRNA prior to the first cleavage (B,F,J,D,H,L) or uninjected sibling controls (A,E,I,C,G,K). A-D are dorsal views. E-H are animal views of the embryos shown in A-D, respectively. I-L are hemisections of the embryos shown in A-D, respectively. The dorsal side is to the right, animal up.
|
|
Fig. S3. Attenuation of CSL-mediated Notch signalling does not result in an expansion of the brain. (A-F) Expression of anterior-posterior (A-P) neural markers at stage 15 in control Xenopus embryos (A,C,E) and sibling embryos injected with 2 ng of su(H)1DBM mRNA (B,D,F). Embryos were injected prior to the first cleavage in the animal hemisphere and were processed for whole-mount in situ hybridisation for the following markers: otx2 (A,B); Xanf1, en2 and hoxb7 (C-F). A-D show anterior views, dorsal up. E and F are dorsal views of the embryos shown in C and D, respectively; anterior up. Asterisks in A and B highlight the expression hole encircled by otx2, which is conserved in su(H)1DBM-injected embryos. shcg, stomodeal-hypophyseal and cement gland anlage; t, presumptive telencephalon; d, presumptive diencephalon; m, presumptive mesencephalon; m/h, presumptive boundary between midbrain and hindbrain; h, presumptive hindbrain; sc, presumptive spinal cord.
|