Buisson N et al. (2014), An adhesome comprising laminin, dystroglycan an...

XB-ART-49633

Development 2014 Dec 01;14123:4569-79. doi: 10.1242/dev.116103.

Show Gene links Show Anatomy links

An adhesome comprising laminin, dystroglycan and myosin IIA is required during notochord development in Xenopus laevis.

Buisson N , Sirour C , Moreau N , Denker E , Le Bouffant R , Goullancourt A , Darribère T , Bello V .

Abstract
Dystroglycan (Dg) is a transmembrane receptor for laminin that must be expressed at the right time and place in order to be involved in notochord morphogenesis. The function of Dg was examined in Xenopus laevis embryos by knockdown of Dg and overexpression and replacement of the endogenous Dg with a mutated form of the protein. This analysis revealed that Dg is required for correct laminin assembly, for cell polarization during mediolateral intercalation and for proper differentiation of vacuoles. Using mutations in the cytoplasmic domain, we identified two sites that are involved in cell polarization and are required for mediolateral cell intercalation, and a site that is required for vacuolation. Furthermore, using a proteomic analysis, the cytoskeletal non-muscle myosin IIA has been identified for the first time as a molecular link between the Dg-cytoplasmic domain and cortical actin. The data allowed us to identify the adhesome laminin-Dg-myosin IIA as being required to maintain the cortical actin cytoskeleton network during vacuolation, which is crucial to maintain the shape of notochordal cells.

PubMed ID: 25359726
Article link: Development

Species referenced: Xenopus laevis
Genes referenced: acta1 chrd dag1 eng fbn1 fbn2 fn1 hopx lama1 mtor myh10 myh9 tuba4b
Antibodies: Acta1 Ab4 Dag1 Ab2 Dag1 Ab4 Fbn2 Ab1 Fn1 Ab1 GFP Ab15 GFP Ab17 Lama1 Ab1 Myh9 Ab1 Notochord Ab1 Tuba4b Ab2
Morpholinos: dag1 MO1 dag1 MO2 myh9 MO1

Article Images: [+] show captions

	Fig. 1. Dystroglycan depletion affects notochord morphogenesis. (A) Immunodetection of Dg on transverse cryosections from stage 13 to 37 (indicated in the top right of the images). The arrowhead in c indicates a vacuole. (B) Immunoblot of protein extracts from wild-type (WT) embryos or embryos injected with a 5-mispair control morpholino (C-MO) or Dg-MO. (C) Cryosections of embryos co-injected with ras-GFP mRNAs and C-MO (a) or with Dg-MO (b) and then immunostained with antibodies against Dg. In Dg morphants, the Dg staining is lost. (D) The top panels show the embryo phenotypes observed with C-MO or an increasing amount of Dg-MO at stage 32. (a-d) Cryosections treated with Tor70 antibodies showing a dose-dependent effect on the notochord. The arrowhead in c indicates the discontinuity in the notochord sheath. A cell containing Dg-MO (labeled with GFP, arrow and inset in c) escaped from the notochord. (d) At a high amount of Dg-MO, the sheath disappears and the characteristic structure of the notochord is lost. Dotted lines outline the neural tube (NT). (E) Notochord morphogenesis is rescued in Dg morphants through injections of mRNA encoding Dg (Dg-FL mRNA). Dotted lines outline the NT. NC, notochord; S, somite. Scale bars: 25 μm in A,C,Da-d,E; 500 μm in upper panel in D; 10 μm in Dc inset.
	Fig. 2. Dystroglycan depletion affects laminin distribution and notochord morphogenesis. (A) Embryos were co-injected with ras-GFP mRNAs and C-MO or Dg-MO. Immunodetection of laminin or using the Tor70 antibody at stage 13 through to stage 25 (indicated at the top right of the images). (a-h) In Dg morphants, laminin is absent around the notochord (NC). Dotted lines outline the notochord. (g′,h′) In situ hybridization for chordin and immunodetection using Tor70 were combined at stage 25. Laminin remains absent, and the notochord shape is affected. Arrowheads in h and h′ point to two chordin-positive cells (blue in h′) that were depleted of Dg (green in h) and had not integrated in the notochord. (B,C) Embryos were co-injected with ras-GFP mRNAs with C-MO or Dg-MO. Immunodetection of fibrillin (B) and fibronectin (C) at stages 13, 15 and 25. These components of the extracellular matrix are detected in Dg morphants. (D) Immunoblot of protein extracts of wild-type (WT) or Dg-MO embryos (stage 25). In Dg-MO protein extract, Dg is undetected and the amount of laminin is comparable to that of WT. NC, notochord; NP, neural plate; S, somite; R, ratios of laminin and β-Dg to α-tubulin normalized against that of WT are shown under each lane. Scale bars: 25 μm.
	Fig. 3. Dystroglycan-depleted cells failed to integrate into the notochord. At the four-cell stage, the two left blastomeres were injected with C-MO or Dg-MO (33 ng) and ras-GFP mRNAs, the two right blastomeres were injected with RLDx as internal controls. Transverse sections at stages 15 (A-F) and 32 (G-L) are shown. (A-C,G-I) Controls. Notochordal cells containing C-MO (green) or RLDx (red) are implicated equally in notochord formation. The axis of symmetry is indicated by the dotted lines. (D-F,J-L) Dg-MO cells (green) are not integrated into the notochord. NC, notochord; NT, neural tube; S, somite. Scale bars: 25 μm.
	Fig. 4. Dystroglycan mutants lacking the cytoplasmic domain (Dg-δcyto) and the tyrosine residue of the PPPY motif affect vacuolation. (A) Whole embryos are shown for wild-type (WT) and Dg-δcyto mutants at stage 32. (a,b) Stage 15 and (c,d) stage 32. Immunodetection of laminin or the Tor70 antibody in controls (a,c), in Dg-δcyto mutants (b) and in replacement of Dg with Dg-δcyto mutants (d). In mutants, cells are round instead of exhibiting a fusiform phenotype (b, arrowheads). Laminin is detected. Tor70 staining reveals a discontinuous sheath (d, arrowhead). (B) Schematic representation of Dg cytoplasmic mutants. Dg-δcyto encodes a protein that does not possess the C-terminus tail. Four amino acids after the transmembrane domain are conserved to ensure that the Dg-δcyto integrates into the membrane. Dg-FL PPPY/LALA encodes a protein in which the PPPY motif is mutated to LALA. Dg-FL Glu and Dg-FL Phe encode proteins in which the tyrosine residue is mutated to glutamate and phenylalanine, respectively. The Dg-FL KKR/ENG encodes a protein in which the KKR motif is mutated to ENG. The Dg-FL PPEY/AAAA mutant encodes a protein in which the PPEY motif is mutated to AAAA. Underlining indicates consensus binding sequences. (C) Dg-MO was co-injected with mRNAs encoding the Dg mutants. At stage 32, protein lysates were subjected to western blotting using antibodies against β-Dg and α-tubulin. Equivalent amounts of propeptides, endogenous Dg (WT), full-length Dg (Dg-FL) and the mutated β-Dg proteins are observed. R, ratios of Dg-propeptide and β-Dg to α-tubulin normalized to WT. (D) Whole embryos are shown for point mutants at stage 32 in upper panels. Overexpression of Dg-FL PPPY/LALA (a), Dg-FL Glu (b) and Dg-FL Phe (c) proteins on sections at stage 15. Replacement of Dg with Dg-FL PPPY/LALA (d), Dg-FL Glu (e) and Dg-FL Phe (f) on sections at stage 32. Immunodetection of laminin (a-c) and the Tor70 antibody (d-f). The arrowhead in e points to a vacuole. Scale bars: 500 μm in A upper panels, D upper panels; 25 μm in Aa-d,Da-f.
	Fig. 5. Mutations in the KKR or PPEY motif result in an enlarged notochord with many vacuoles. (A) Whole embryos are shown for the point mutants at stage 32. Immunodetection of laminin (a-c) and with the Tor70 antibody (g-i) at the stages indicated in the top right. The arrowhead in b points the laminin (red) present around the cell. Arrowheads in h and i indicate numerous vacuoles. (d-f) In situ hybridization for the chordin gene. (B) Histogram showing the number of Hoechst-stained nuclei in ten transverse sections in the truncal part of wild-type and mutant embryos at stages 15 and 32. Results are means from five independent experiments. (C) Frontal sections of mutants at the level of the notochord. Four-cell-stage embryos were injected with ras-GFP mRNA alone (control; green), or with ras-GFP mRNA and mutant mRNAs into the right blastomeres. RLDx (red) was injected into the left blastomeres. In controls (a,c), red and green cells periodically intercalate and extend along the notochord width (arrowheads in a,c). In mutants, the regular succession of ras-GFP cells (green, arrowheads in b,d) and RLDx cells (red) is lost. Dashed lines outline the notochord. Scale bars: 500 μm in A upper panels; 25 µm in Aa-i,Ca-d.
	Fig. 6. Myosin IIA is a Dg partner in notochord. (A) Immunoprecipitation (IPP) using an antibody against β-Dg. The precipitate and supernatant of wild-type (WT) and Dg-MO-injected notochords were subjected to western blotting with antibodies against myosin IIA and β-Dg. (B) Identification of Dg and myosin IIA localization using immunofluorescence confocal microscopy. Colocalization of Dg and myosin IIA is observed at cell cortices at stage 28 in WT embryos. (C) Immunodetection of myosin IIA (a,b) and Tor70 proteins (c,d) in C-MO and Dg-MO notochords. Labeling of myosin IIA was lost in Dg morphants at stage 32. Nuclei are stained with Hoechst (blue). (D) Protein extracts of WT, Dg-MO and myosin IIA-MO embryos were subjected to western blotting with antibodies against myosin IIA, actin and α-tubulin (6% SDS-PAGE) and with an antibody against β-Dg (12% SDS-PAGE). In Dg-MO extracts, Dg expression was strongly decreased, whereas the amount of myosin IIA was comparable to that of controls. Note that the decrease in Dg and myosin is not accompanied by a decrease in α-tubulin (α-Tub) or actin. R, ratios of myosin IIA, β-Dg and actin to α-tubulin normalized to WT. Scale bars: 25 µm in B,C.
	Fig. 7. Myosin IIA is required for cytoskeletal integrity. (A) Protein extracts of wild-type (WT) or myosin IIA-MO embryos were subjected to western blotting with antibodies against myosin IIA and α-tubulin, which showed the efficiency of the morpholinos. R, ratios of myosin IIA to α-tubulin normalized to WT. (B) Immunodetection of myosin IIA in C-MO and myosin IIA-MO notochords. Labeling of myosin IIA was lost in morphants at stage 28. Dotted lines outline the notochord (NC). (C) Phenotypes of control and myosin IIA morphants at stage 32. (a-d) Immunodetection of Dg and laminin in C-MO embryos and morphants. The notochord diameter and the vacuole size increased. Myosin IIA depletion does not affect Dg and laminin expression, but they do not localize normally. (e,f) Tor70 labeling confirms the increase in the size of the notochord and vacuoles. (D) F-actin and α-tubulin localization, by using confocal microscopy, in control and myosin IIA- and Dg-depleted notochord at stage 32. (a-f) F-actin is revealed by staining with phalloidin. (d-f) Magnification at the cell cortex showing the loss of labeling in morphants; the area corresponds to that highlighted by the white box in the panel above. (g-l) Tubulin is revealed by using an antibody against α-tubulin. (j-l) Magnification at the cell cortex showing the network disruption of α-tubulin; the area corresponds to that highlighted by the white box in the panel above. Dotted lines outline the notochord. S, somite. Scale bars: 25 μm in B,Ca-f,Da-c,Dg-i; 500 μm in C upper panels; 10 μm in Dd-f,Dj-l.
	Fig. 8. The Dg adhesome is involved throughout notochord morphogenesis. The data have established that, when Dg binds to laminin, the RKKRK and PPEY sites of the cytoplasmic domain are mobilized to ensure cell polarity and mediolateral intercalation. Laminin-Dg interactions lead to phosphorylation of the tyrosine residue in the PPPYVPP site to generate a signaling pathway that allows vacuole differentiation. Finally, an adhesome comprising laminin-Dg-myosin-IIA-actin is established to maintain cortical integrity during vacuolation. The sheath is shown in yellow. (Adapted from Keller et al., 1989.)
	Supplementary Figure S2. Chordin is present in controls and Dg-morphants during gastrulation. In toto hybridizations were performed on C-Mo and Dg-Mo embryos during gastrulation at stages 10 (a, b), 10.5 (c, d), 11.5 (e, f) and 13 (g, h). The spatiotemporal distribution of chordin mRNA was similar in controls and Dg-morphants. Scale bars: 500 µm.
	Supplementary Figure S3. Chordin is present in control and Dg-morphant notochords during neurulation. In situ hybridizations were performed on C-Mo and Dg-Mo embryos during neurulation at stages 15 (a, b), and 18 (c, d). On cryosections, a mediolateral enlargement (stage 15) is observed. In b, dotted lines delimit somites. Scale bars: 25 µm.
	Supplementary Figure S6. Activin induced animal cap to obtain notochord differentiated explants in vitro. (a-c) Two-cell stage embryos were injected with C-Mo, Dg-Mo or Dg-Mo + Dg-FL mRNAs and animal caps were explanted. In toto analyses showed that the elongation observed in CMo activin-treated explants is strongly inhibited by Dg depletion (Dg-Mo, 33 ng) and importantly, is rescued to a large extent by the co-injection of 100 pg of Dg-FL mRNAs. Scale bars: 50 µm. (d-f) Animal caps stage 28 equivalent were sectioned and Tor-70 protein was immunodetected. Well-differentiated notochords are clearly observable in control (d) and rescued (f) animal caps. In e, notochord structure is indistinguishable. Scale bars: 500 µm. (g, h) The expression of the chordin gene was detected by whole-mount in situ hybridization on control (g) and Dg-Mo explants (h). Detection of the ras-GFP, which was used as tracer of CMo and Dg-Mo is shown in panel h. The expression of chordin is detected in control (g) and Dg-depleted explants stage 18 equivalent (h). Scale bars: 50 µm.