May 1, 2005;
Notch signaling modulates the nuclear localization of carboxy-terminal-phosphorylated smad2 and controls the competence of ectodermal cells for activin A.
Loss of mesodermal competence (LMC) during Xenopus development is a well known but little understood phenomenon that prospective ectodermal cells (animal caps) lose their competence for inductive signals, such as activin A
, to induce mesodermal genes and tissues after the start of gastrulation. Notch
signaling can delay the onset of LMC for activin A
in animal caps [Coffman, C.R., Skoglund, P., Harris, W.A., Kintner, C.R., 1993. Expression of an extracellular deletion of Xotch
diverts cell fate in Xenopus embryos. Cell 73, 659-671], although the mechanism by which this modulation occurs remains unknown. Here, we show that Notch
signaling also delays the onset of LMC in whole embryos, as it did in animal caps. To better understand this effect and the mechanism of LMC itself, we investigated at which step of activin signal transduction pathway the Notch
signaling act to affect the timing of the LMC. In our system, ALK4
(activin type I receptor) maintained the ability to phosphorylate the C-terminal region of smad2
upon activin A
stimulus after the onset of LMC in both control- and Notch
-activated animal caps. However, C-terminal-phosphorylated smad2
could bind to smad4
and accumulate in the nucleus
only in Notch
-activated animal caps. We conclude that LMC was induced because C-terminal-phosphorylated smad2
lost its ability to bind to smad4
, and consequently could not accumulate in the nucleus
signal activation restored the ability of C-terminal-phosphorylated smad2
to bind to smad4
, resulting in a delay in the onset of LMC.
[+] show captions
Fig. 1. The effect of Notch signaling on activin-induced Xbra and goosecoid expression. Animal caps injected with 1 ng of Notch ΔE were dissected at stage 9, 10.5, and 11.5, and then treated with 10 ng/ml of activin A, followed by incubation for 5 h. The relative Xbra (A) and goosecoid (B) expression levels were calculated using real-time RT-PCR. (C) Animal caps injected with 1 ng Notch ΔE alone or co-injected with 1 ng Notch ΔE and 1 ng Su(H) DBM were dissected at stage 11.5, and then treated with 10 ng/ml activin A, followed by a 5-h incubation. The relative Xbra expression level was calculated using real-time RT-PCR. The efficiency of cDNA synthesis was assessed on the basis of real-time RT-PCR for ODC. The results represent the mean from three or four independent experiments and error bars indicate the SEM. Black column, Notch ΔE-injected animal caps; White column, uninjected control animal caps, Gray column, Notch ΔE and Su(H) DBM-injected animal caps.
Fig. 2. LMC in vivo. (A) Schematic diagram of activin bead implantation. An activin bead loaded with 50 ng/ml activin A was implanted into the ventral margin of blastocoels of stage-9 or 11.5-embryos. v, ventral; d, dorsal. (B) The activin bead was implanted into stage-9 embryos (B), stage-11.5 embryos (C), Notch ΔE-injected stage-11.5 embryos (D), stage-12 embryos (E), and Notch ΔE-injected stage-12 embryos. Red arrowhead indicates induced protrusions (F). A control bead was transplanted into control embryos (G). (H) Notch ΔE-injected embryo without activin bead. (I) Transverse section of a stage-11.5 embryo injected with Notch ΔE and an implanted activin bead shown in panel D. (I) Section stained with hematoxylin and eosin. (J) Higher magnification of panel I. Red dotted line represents the border between the host endoderm and the secondary trunk-tail structure. (K) The section was stained with the muscle-specific 12/101 antibody. Black arrow, ectopic muscle. (L) The section was stained with the neuron-specific Neu-1 antibody. Black arrow, ectopic neuron. (M) The section was stained with the notochord-specific Tor 70 antibody. (Bars=100 μm) (N,O) The expression pattern of Xbra at stage 12 was examined by whole-mount in situ hybridization. (N) LacZ-injected control embryos. (O, O′) Embryos were injected with 1.0 ng Notch ΔE together with 100 pg of lacZ mRNA as lineage tracer into one blastomere at the 2-cell stage. Notch ΔE injection reduced Notochordal Xbra expression (blue arrowhead), while it enhanced Xbra expression in involuting mesoderm (red arrow).
Fig. 3. The effect of Notch signaling on ALK4 expression level and C-terminal-phosphorylation of smad2. ALK4 maintained the ability to phosphorylate the C-terminal region of smad2 upon activin A stimulus even after the onset of LMC. (A) The relative expression levels of ALK4 were calculated using real-time RT-PCR. Animal caps injected with 1 ng of Notch ΔE were dissected at stages 9, 10.5, and 11.5. The efficiency of cDNA synthesis was assessed on the basis of the real-time RT-PCR for ODC. The results represent the mean from three or four independent experiments and error bars indicate the SEM. Black column, Notch ΔE-injected animal caps; White column, uninjected control animal caps. (B) The amount of C-terminal-phosphorylated smad2 was examined by western blotting using anti-phosphorylated-smad2C antibody (αp-smad2C), which recognized the phosphorylated C-terminal region of smad2. Animal caps injected with 1 ng of Notch ΔE or non-injected controls were dissected at stage 9 and stage 11, and treated with 50 ng/ml of activin A. Actin levels served as the loading control.
Fig. 4. The effect of Notch signaling on formation of a complex formation between C-terminal-phosphorylated smad2 and smad4. Animal caps injected with 1 ng of Notch ΔE (lanes 3 and 5) or control animal caps (lanes 1, 2 and 4) were dissected at stage 9 (lanes 1, 2 and 3) or stage 11 (lanes 4 and 5), and treated with 50 ng/ml of activin A. Lysates were immunoprecipitated with anti-smad4 antibody (IP) and western blotted with αp-smad2C antibody. The level of smad2 C-terminal phosphorylation was quantitated and the values plotted.
Fig. 5. The effect of Notch signaling on smad2 nuclear accumulation. Animal caps were injected with 500 pg of GFP-smad2 alone or co-injected with 500 pg of GFP-smad2 and 1 ng of Notch ΔE, and dissected at stage 9 (A) or stage 11 (B), and then dissociated in dissociation medium. Dissociated animal cap cells were treated with 5 ng/ml activin A for 30 min and then loaded onto a fibronectin-coated slide. The localization of GFP-smad2 was analyzed by confocal microscopy.
Fig. 6. The effect of Notch signaling on linker-region-phosphorylation of smad2. (A) Diagram of GFP-smad2 and GFP-smad2-L3SA in which the three linker region serine residues were substituted by alanine residues. αp-smad2L selectively recognized smad2 phosphorylation at the linker region. (B) Two-cell-stage embryos were injected with GFP-smad2 or GFP-smad2-L3SA. Embryos were harvested at stage 12 and smad2 linker-region phosphorylation levels were examined by western blotting using αp-smad2L. (C) Animal caps injected with 1 ng of Notch ΔE or non-injected controls were dissected at stage 11, and smad2 linker-region phosphorylation levels were examined by western blotting using αp-smad2L. Smad2 levels served as the loading control.