December 1, 2012;
Dishevelled limits Notch signalling through inhibition of CSL.
and Wnt are highly conserved signalling pathways that are used repeatedly throughout animal development to generate a diverse array of cell types. However, they often have opposing effects on cell-fate decisions with each pathway promoting an alternate outcome. Commonly, a cell receiving both signals exhibits only Wnt pathway activity. This suggests that Wnt inhibits Notch
activity to promote a Wnt-ON/Notch
-OFF output; but what might underpin this Notch
regulation is not understood. Here, we show that Wnt acts via Dishevelled
to inhibit Notch
signalling, and that this crosstalk regulates cell-fate specification in vivo during Xenopus development. Mechanistically, Dishevelled
binds and directly inhibits CSL
transcription factors downstream of Notch
receptors, reducing their activity. Furthermore, our data suggest that this crosstalk mechanism is conserved between vertebrate and invertebrate homologues. Thus, we identify a dual function for Dishevelled
as an inhibitor of Notch
signalling and an activator of the Wnt pathway that sharpens the distinction between opposing Wnt and Notch
responses, allowing for robust cell-fate decisions.
[+] show captions
Fig. 2. Dishevelled regulates Notch-dependent cell fate decisions in vivo. (A-C) XDvl2 is required to regulate Notch signalling during development. Xenopus tropicalis embryos were injected at the one-cell stage with control morpholino (MOC) or one targeting Xdvl2 (MODvl2). Ciliated cell precursors were detected by α-tubulin expression (purple staining). (D) Precursors were counted within a box of standard area drawn over the centre of each embryo (see inset). Data are presented as mean number of precursors counted (±s.e.m.). Uninjected (UI) and MOC embryos were indistinguishable (A,B), whereas MODvl2 embryos exhibited a significant reduction in precursor number (C) (one-way ANOVA with Tukey's post-hoc tests). (E) qRT-PCR analysis of α-tubulin and esr1 expression in MOC and MODvl2 embryos. Expression was normalised to rpl8. Data are presented as mean fold change (±s.e.m.) in normalised expression values relative to uninjected embryos. esr1 expression was significantly increased in the MODvl2 embryos (two-tailed t-test, n=3). (F-I′) XDvl2 expression inhibits endogenous Notch signalling and rescues the NICD gain-of-function phenotype. Xenopus laevis embryos were injected in one blastomere of the two-cell embryo with mRNA encoding β-gal (F,F′), β-gal and XDvl2 (G,G′), β-gal and XNICD (H,H′), or β-gal, XNICD and XDvl2 (I,I′). (J) Ciliated cell precursors were quantified as above (see D). X-Gal staining was performed to distinguish the injected side (pale red). Images of the uninjected and injected sides of the same embryo are shown. XDvl2 promoted (G′) and XNICD inhibited (H′) ciliated cell precursor specification (two-tailed paired t-test). XDvl2 completely rescued the XNICD phenotype (I′) and there was no significant difference between the uninjected sides of any condition, and the β-gal or XNICD + XDvl2 expressing sides (two-way ANOVA and Bonferroni's post-hoc test). (K) qRT-PCR analysis of α-tubulin and esr1 expression in embryos expressing GFP or XDvl2 conducted as above (see E). esr1 expression was significantly decreased in XDvl2-expressing embryos (two-tailed t-test, n=3). Scale bars: 500 μm in A-C,F-I′. **P<0.01; ***P<0.001; ns, P>0.05.
Fig. 3. Dishevelled does not inhibit Notch cleavage or the nuclear translocation of NICD. (A) mDvl2 inhibits all four human Notch paralogues. CHO-K1 cells were transfected with RBPJκ-Luc and pRL-CMV. Notch signalling was initiated by expressing active forms of each human Notch paralogue (δN-hN1-4) in the presence or absence of mDvl2. Data are presented as mean fold change (±s.e.m.) in RLU compared with each Notch construct alone. Dvl2 inhibited each Notch paralogue (***P<0.001, one-way ANOVA and Tukey's post-hoc test, n=3). (B) mDvl2 does not inhibit Notch cleavage. Cells transfected with RBPJκ-lacZ and pRL-CMV were also transfected with vectors encoding δN-mN1 or mDvl2. Lysates were analysed by immunoblotting to examine β-gal reporter gene activity and δN-mN1 cleavage (NICD Val1744). R. luciferase is a loading control. (C,D) NICD released from δN-mN1 translocates to the nucleus even in the presence of mDvl2. (C) Cells expressing δN-mN1-GFP and mDvl2-V5, as indicated, were fixed and immunostained for GFP (green) and V5 (red) epitopes. DAPT treatment to inhibit γ-secretase function prevented nuclear translocation of NICD. (D) CHO-K1 cells expressing δN-mN1 and mDvl2, as indicated, were fractionated and nuclear accumulation of NICD was analysed by immunoblotting the nuclear fraction (Nuc) and total lysates (Total). LaminB1 is a loading control. Positions of molecular weight markers (in kDa) are indicated.
Fig. 6. Dishevelled DIX and PDZ domains are required for inhibition of RBPJκ. (A) Schematic of the structure of Dishevelled and the deletion constructs used. (B) The C terminus of mDvl2 is not required for inhibition of RBPJκ activity. CHO-K1 cells were transfected with RBPJκ-Luc and pRL-CMV, and vectors encoding δN-mN1 and the mDvl2 constructs illustrated. Data are presented as mean fold change (±s.e.m.) in RLU relative to δN-mN1 alone. Both δC-mDvl2 and mDvl2, but not δN-mDvl2, inhibit Notch signalling (one-way ANOVA and Tukey's post-hoc test, n≥3). (C-E) XDvl2 but not Ds1 inhibits endogenous Notch signalling in vivo. Xenopus laevis embryos were injected in one blastomere of a two-cell embryo with mRNA encoding β-gal and XDvl2 (C,C′) or β-gal and Ds1 (D,D′). Ciliated cell precursors were detected by α-tubulin expression and the injected side was determined by X-Gal staining. Images of the uninjected (UI) and injected sides of the same embryo are shown. (E) Ciliated cell precursors were quantified as in Fig. 2D. Data are presented as the mean number of precursors counted (±s.e.m.). XDvl2 significantly increased ciliated cell precursor specification but Ds1 did not (two-tailed paired t-test). XDvl2 also increased precursor specification compared with Ds1 (two-way ANOVA and Bonferroni's post-hoc test). (F) Ds1 does not inhibit XSu(H) activity. CHO-K1 cells were transfected with NRE Su(H)-Luc and pRL-CMV, and vectors encoding XSu(H)-ANK, XDvl2 or Ds1. Data are presented as mean fold change (±s.e.m.) in RLU relative to XSu(H)-ANK alone (one-way ANOVA and Tukey's post-hoc test, n=3). (G) Ds1 shows greatly reduced XSu(H) binding. CHO-K1 cells expressing XSu(H)-ANK-myc and XDvl2-GFP or Ds1-GFP were subjected to immunoprecipitation using GFP-Trap beads. Immunoprecipitation samples were analysed by immunoblotting for myc and GFP, alongside total lysates. Positions of molecular weight markers (in kDa) are shown. Scale bar: 500 μm. **P<0.01; ***P<0.001; ns, P>0.05.