XB-ART-39166Development. February 1, 2009; 136 (4): 575-84.
A role for Syndecan-4 in neural induction involving ERK- and PKC-dependent pathways.
Syndecan-4 (Syn4) is a heparan sulphate proteoglycan that is able to bind to some growth factors, including FGF, and can control cell migration. Here we describe a new role for Syn4 in neural induction in Xenopus. Syn4 is expressed in dorsal ectoderm and becomes restricted to the neural plate. Knockdown with antisense morpholino oligonucleotides reveals that Syn4 is required for the expression of neural markers in the neural plate and in neuralised animal caps. Injection of Syn4 mRNA induces the cell-autonomous expression of neural, but not mesodermal, markers. We show that two parallel pathways are involved in the neuralising activity of Syn4: FGF/ERK, which is sensitive to dominant-negative FGF receptor and to the inhibitors SU5402 and U0126, and a PKC pathway, which is dependent on the intracellular domain of Syn4. Neural induction by Syn4 through the PKC pathway requires inhibition of PKCdelta and activation of PKCalpha. We show that PKCalpha inhibits Rac GTPase and that c-Jun is a target of Rac. These findings might account for previous reports implicating PKC in neural induction and allow us to propose a link between FGF and PKC signalling pathways during neural induction.
PubMed ID: 19144724
PMC ID: PMC2685954
Article link: Development.
Grant support: Biotechnology and Biological Sciences Research Council , Medical Research Council , G0801145 Medical Research Council , MRC_G0801145 Medical Research Council
Genes referenced: cdk20 chrd.1 fgf2 fos gal.2 jun mapk1 myod1 ncam1 nrp1 prkca prkcd rac1 sdc4 sox2 sox3 tbxt xk81a1
Morpholinos referenced: sdc4 MO1 sdc4 MO2
Article Images: [+] show captions
|Fig. 1. Dynamic expression of Syn4 in the neural plate region. Whole-mount in situ hybridisation analysis of Syn4 and Sox expression. (A) Lateral view of a stage 10.5 Xenopus embryo showing Syn4 expression in the dorsal marginal zone. Dorsal (d) to the right; ventral (v), left; animal pole to the top. (B) Fate map of a stage 10.5 embryo, shown in the same orientation as in A. NP, prospective neural plate; m, prospective mesoderm. (C) At stage 12, Syn4 expression (arrowheads) is restricted to the dorsal region of the embryo (orientation as in A). (D) Fate map of a stage 12 embryo, shown in the same orientation as in C. NP, neural plate. (E) At stage 14, Syn4 expression is seen in the neural plate, but is absent from the dorsal midline (arrowhead). Dashed line indicates the plane of the section in G. (F) Stage 14 embryo showing Sox2 expression. The expression pattern is similar to that of Syn4 in E. Dashed line indicates the plane of the section in H. Arrowhead, dorsal midline. (G) Section of a stage 14 embryo, showing Syn4 expression. No expression is observed at the midline or in mesoderm. (H) Section of a stage 14 embryo, showing Sox2 expression (I) At stage 16, Syn4 expression is seen in the neural plate. (J) Sox2 expression at stage 16.|
|Fig. 2. Syn4 is required for neural induction.Sox2 and Nrp1 expression was analysed by whole-mount in situ hybridisation. (A-H) MO-injected samples analysed at stage 14. The lineage tracer in shown in the insets. (A) Syn4 MO injected in one animal blastomere of an 8-cell stage Xenopus embryo. Note the inhibition of Sox2 expression on the injected (right-hand) side (60%, n=45). (B) Similar to A, but injection with control MO. No effect on Sox2 expression is observed (0%, n=57). (C) Syn4 MO was injected into the dorsal animal blastomere (A1) at the 32-cell stage. Note the inhibition of Sox2 expression in the injected region (arrowhead) (55%, n=50). (D) Syn4 mRNA mutated in the MO sequence region was co-injected with Syn4 MO into the A1 blastomere of a 32-cell stage embryo. Note the rescue of the expression of Sox2 (8%, n=66). (E) Syn4 MO injected into one animal blastomere of an 8-cell stage embryo. Note the inhibition of Nrp1 expression on the injected side (95%, n=42). (F) No inhibition is observed with the control MO (0%, n=21). (G) Control-morpholino-injected ectoderm was grafted into an uninjected embryo. Sox2 expression is normal (100%, n=10). Black dashed line indicates the anterior border of Sox2 expression. White dashed line outlines the graft. Inset shows the position of the graft by its fluorescence. (H) Syn4 MO-injected ectoderm was grafted into an uninjected embryo. Sox2 expression is absent from the grafted area (asterisk; 70%, n=10). Inset shows the position of the graft by its fluorescence. (I-N) Analysis of Sox2 expression at stage 11. (I) Expression of Sox2 is initially observed in the dorsal region (arrowhead). (J) Expansion of Sox2 expression by BMP inhibition [BMP antagonists: truncated BMP receptor (tBR) and chordin (chd) mRNA]. (K) Early induction of Sox2 was eliminated by co-injection of Syn4 MO (69%, n=50). (L) Control animal caps, showing no expression of Sox2. (M) Animal caps taken from embryos injected as in J, showing weak upregulation of Sox2. (N) Animal caps taken from embryos as in K. Co-injection of Syn4 MO blocks Sox2 induction.|
|Fig. 3. Syn4 overexpression neuralises ectoderm. (A-D) Anterior view, dorsal to the top (A,B); ventral view (C,D). (A,C) 500 pg of NLS-β-gal mRNA was injected into the A4 blastomere of a 32-cell stage Xenopus embryo, and the expression of Sox2 (A) and epidermal keratin (Epk) (C) was analysed by in situ hybridisation. (B,D) 1 ng of Syn4 mRNA was injected into the A4 blastomere of a 32-cell stage embryo, and the expression of Sox2 (C; 96% of induction, n=87) and Epk (D; 80% of inhibition, n=70) was analysed. (E,F) Higher magnification of embryos injected as in B,D. Arrows indicate nuclei stained with X-Gal (blue). (G-J) Dorsal (G,I) and ventral (H,J) views of embryos injected as in B with Syn4 mRNA. No ectopic expression of MyoD (0%) or Xbra (0%) was observed. Inset in H shows injected fluorescein. (K-N) Animal cap (AC) assay for Sox2 (K,L) or Epk (M,N) expression. (K,M) Control caps express only Epk. (L,N) Syn4 mRNA-injected animal caps express Sox2 and downregulate Epk.|
|Fig. 4. The FGF/MAPK pathway is required for neural induction by Syn4. (A-F) Sox2 expression analysed by whole-mount in situ hybridisation in Xenopus animal caps. (A) Control caps. No Sox2 expression. (B) Syn4-injected caps express Sox2. (C) Co-injection of Syn4 and a dominant-negative form of FGF receptor 1 (XFD-1) inhibits Sox2 expression. (D) Syn4-injected caps treated with 40μ M SU5402 (in DMSO) do not show Sox2 expression. (E) Syn4-injected caps treated with 80 μM U0126 do not express Sox2. (F) Syn4-injected caps treated with 80 μM U0124 show expression of Sox2.(G) Animal caps were injected as indicated (Cont, control) and samples taken for western blot analysis of MAPK phosphorylation at the equivalent of stage 12. Antibodies against MAPK or phosphorylated MAPK (p-MAPK) can recognise both p42 and p44 as a single band. Each experiment was repeated three times with at least 50 animal caps.|
|Fig. 5. Membrane translocation of PKCδ by FGF is inhibited by Syn4. Xenopus animal caps analysed by confocal microscopy after injection/treatment as indicated. (A-C) Control animal cap shows cytoplasmic localisation of PKCδ. (D-F) Animal caps injected with Syn4 mRNA. PKCδ shows cytoplasmic distribution. (G-L) Phorbol ester (PMA; G-I) or FGF2 (J-L) triggers the translocation of PKCδ into the membrane. (M-O) Syn4 inhibits the translocation of PKCδ activated by FGF2.|
|Fig. 6. Inhibition of PKCδ or activation of PKCα is essential for neural induction by Syn4. (A-H) Sox2 expression in Xenopus embryos injected, as indicated, into A4 blastomeres at the 32-cell stage. (A-D) Ventral view, dorsal to the top. Insets on left show a dorsal view. (A) Dominant-negative Xenopus PKCδ (DN-PKCδ). Note the ectopic Sox2 induction (48%, n=71). (B) Co-injection of Syn4 and Xenopus full-length PKCδ mRNAs represses the ectopic expression of Sox2 (18%, n=91). Inset on the right shows the ventral ectopic Sox2 expression induced by Syn4 mRNA (95%, n=85). (C) Human (h) PKCα mRNA can induce ectopic Sox2 expression (68%, n=92). (D) Co-injection of hPKCα and PKCδ mRNAs shows inhibition of neural induction (20%, n=124). (E) Co-injection of hPKCα with control MO (CMO). Arrowhead indicates the ectopic expression of Sox2 (70%, n=68). (F) Co-injection of hPKCα mRNA with Syn4 MO. Sox2 induction is not observed in the injected cells (arrowhead) (23%, n=51). (G) Co-injection of Syn4 and a dominant-negative form of PKCα (DN-PKCα-EGFP) inhibits Sox2 expression (15%, n=51). (H) Injection of PKCα-EGFP mRNA. Ectopic induction of Sox2 is similar to that upon PKCα injection, showing that EGFP does not affect the activity of the fusion protein. (I-N) Confocal images of animal caps injected as indicated. PKCα-EGFP mRNA was injected into both blastomeres at the 2-cell stage. At the 16-cell stage, Syn4 or control MO and membrane Cherry mRNA were injected into one blastomere. (I-K) PKCα-EGFP spontaneously localises at the membrane, colocalising with membrane Cherry. (L-N) The distribution of Syn4 MO can be identified by the fluorescence of mCherry. Note that cells with a high level of Syn4 MO (asterisks) exhibit a low level of PKCα-EGFP in the membrane.|
|Fig. S1. Syn4/PKC/Rac pathways involved in neural induction. Embryos were injected as indicated into the animal blastomeres of an 8-cell stage embryo or into A1 or A4 of a 32-cell embryo, and the expression of neural or epidermal markers analyzed at stage 19. (A) Control MO injected at the 8-cell stage. Analysis of Nrp1 (0% inhibition, n=18). (B) Syn4 MO injected at the 8-cell stage. Analysis of Nrp1 (96% inhibition, n=27). (C) PKCδ mRNA injected at the 8-cell stage. Analysis of Nrp1 (46% inhibition, n=23). (D) Constitutively active Rac mRNA injected at the 8-cell stage. Analysis of Nrp1 (39% inhibition, n=42). (E) PKCδ mRNA injected at the 32-cell stage. Analysis of NCAM (43% inhibition, n=7). (F) Constitutively active Rac mRNA injected at the 32-cell stage. Analysis of NCAM (32% inhibition, n=25). (G) Sox2 expression of embryo injected with PKCα and a dominant-negative form of FGF receptor (XFD-1). Note that the ectopic induction is not inhibited (58% induction, n=31). (H) Epidermal keratin (Epk) expression after injection of PKCα mRNA into A4 blastomeres. Note the inhibition of epidermis (63% inhibition, n=22).|
|Fig. S2. Syn4 has a neuralising activity that does not depend on BMP inhibition. (A,B) Formation of secondary axis. (A) Embryos were injected in ventral blastomeres with Syn4 (A) or chordin (B) mRNA and the lineage tracer β-galactosidase. Note the secondary axis induced by chordin injection; but no secondary axis was observed in Syn4-injected embryos. (C,D) Sox3 expression in stage 14 embryos, dorsal to the top. Inset shows the dorsal side. (C) Control expression of Sox3. (D) Embryo was injected with Syn4 mRNA into A4 blastomeres and the expression of Sox3 analyzed. Note the ectopic induction of Sox3 in ventral epidermis (89% induction, n=49). (E-G) Neuralisation by Syn4 does not require chordin. (E) Ventral view of an embryo injected with Syn4 mRNA. Note the ectopic Sox2 expression (91%, n=12). (F) Ventral view of embryo co-injected with Syn4 mRNA and chordin MO. There was no effect on the ectopic induction of Sox2 (89%, n=37). (G) Embryo injected with chordin MO to show the expected ventralised phenotype (100%, n=9). (H,I) Sox2 expression analysed in stage 20 embryos. Anterior view, dorsal to the top. (H) Control. (I) Embryo injected in A4 with Syn4 mRNA. Note the strong ectopic induction of Sox2. (J,K) Animal caps showing induction of Otx2 by Syn4 expression (K); no induction was observed in control animal caps (J).|
|Fig. S3. Analysis of neuralising activity of Syn4. (A) Deletion constructs of Syn4. 1, wild-type Syn4. 2, Syn4ΔGAG, with deletion of the glycosaminoglycan domain of Syn4. 3, Syn4ΔCyt-Cherry, with deletion of the intracellular domain of Syn4, which was replaced by the fluorescent protein Cherry. (B) Embryos were injected in A4 with the indicated mRNAs and the expression of Sox2 analyzed. The percentage of embryos showing normal (blue bars) or ectopic (red bars) Sox2 expression. Different concentrations of each construct were tested.|
|Fig. 7. The PKC-dependent pathway of neural induction is mediated by the Rac/JNK/AP-1 pathway. (A) RT-PCR of the indicated genes using RNA from Xenopus whole embryos (WE, lane 1), animal caps (AC, lane 2), animal caps expressing PKCα (PKCα, lane 3) or animal caps expressing PKCα and treated with the inhibitor U0126 (+U0126, lane 4). Note that Sox2 is induced by PKCα even when MAPK is inhibited (lane 4). (B) Western blot of control- or PKCα-injected animal caps, detecting c-Fos, phosphorylated MAPK (p-MAPK) and MAPK. Neuralisation of the animal caps correlates with an increase in c-Fos levels, whereas p-MAPK is unchanged. (C) Rac activation assay. (Left) Rac1-His recombinant protein (24 kDa; 20 ng) was loaded and detected by anti-Rac antibody as a positive control. (Right) The same volume of reaction mix was loaded for each condition. Lane 1 (AC+GDP), negative control; lane 2 (AC+GTP), positive control that shows the total amount of Rac protein in the animal cap samples; lane 3 (AC), endogenous active Rac present in animal caps at stage 11.5; lane 4 (Syn4), endogenous active Rac present in animal caps at stage 11.5 injected with 500 pg of Syn4 mRNA (note that Syn4 abolishes endogenous Rac activity); lane 5 (Syn4+GTP), total amount of Rac protein after Syn4 injection. The experiment was repeated three times. (D-I) Whole-mount in situ hybridisation analysis of embryos injected, as indicated, into the A4 blastomere of 32-cell stage embryos. (D) Human (h) PKCα mRNA induces ectopic Sox2 expression (68%, n=92). (E) Co-injection of constitutively active Rac and hPKCα inhibits ectopic Sox2 expression (21% of induction, n=34). (F) Injection of constitutively active Rac does not induce Sox2 expression (0%, n=32). (G) Dominant-negative Rac1 mRNA induces ectopic ventral Sox2 expression (90%, n=40). (H,I) Ventral view of embryos injected with c-Fos-GR mRNA and activated with dexamethasone (DEX) at stage 10.5. Ectopic Sox2 expression is observed only after DEX treatment (I), being absent when no DEX is added (H; inset shows dorsal side). (J-O) In situ hybridisation for Sox2 in animals caps. (J) Control animal caps show no Sox2 expression. (K) Animal caps from embryos injected with chordin (chd) mRNA show Sox2 expression. (L) Animal caps from embryos injected with chordin, constitutive Rac and c-Fos-GR mRNA but without adding DEX. Activation of Rac leads to inhibition of Sox2. (M) Similar to L, but c-Fos-GR is activated by DEX treatment. Rescue of Sox2 expression is observed. (N) Injection of c-Fos-GR mRNA, but without adding DEX. (O) c-Fos-GR-injected animal caps activated with DEX show an increase in Sox2 expression.|