Cadherin adhesion, tissue tension, and noncanonical Wnt signaling regulate fibronectin matrix organization - PubMed (original) (raw)

Cadherin adhesion, tissue tension, and noncanonical Wnt signaling regulate fibronectin matrix organization

Bette J Dzamba et al. Dev Cell. 2009 Mar.

Abstract

In this study we demonstrate that planar cell polarity signaling regulates morphogenesis in Xenopus embryos in part through the assembly of the fibronectin (FN) matrix. We outline a regulatory pathway that includes cadherin adhesion and signaling through Rac and Pak, culminating in actin reorganization, myosin contractility, and tissue tension, which, in turn, directs the correct spatiotemporal localization of FN into a fibrillar matrix. Increased mechanical tension promotes FN fibril assembly in the blastocoel roof (BCR), while reduced BCR tension inhibits matrix assembly. These data support a model for matrix assembly in tissues where cell-cell adhesions play an analogous role to the focal adhesions of cultured cells by transferring to integrins the tension required to direct FN fibril formation at cell surfaces.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Non-canonical Wnt signaling regulates FN assembly and cadherin adhesion

(A,C,E,G,I, K) FN fibril assembly in dorsal marginal zones from stage 11 embryos was analyzed by immunofluoresence microscopy using anti-FN pAb 32. (A) FN forms a network of fibrils in control dorsal marginal zones by this stage. (C) dnWnt11 inhibits fibril assembly but FN is deposited at cell boundaries. (E) FN fibril assembly is rescued by Dsh lacking the DIX domain. (G) Dsh lacking the DEP domain is unable to rescue fibril assembly. (I) Constitutively active RhoV14 also rescues FN assembly. (K) dn integrin HAβ1 blocks RhoV14 rescue of fibril assembly. Scale bar in K=50 μm. (B,D,F,H,J,L) Animal cap extension assays plus (+) or minus (-) activin. (B) Control activin-treated animal caps extend. (D) Caps expressing dnWnt11 do not extend. (F) Dsh construct lacking the DIX domain rescues cap extension. (H) Dsh lacking the DEP domain does not rescue cap extension. (J) Constitutively active RhoV14 rescues extension. (L) RhoV14 is unable to rescue extension when co-expressed with HAβ1. (M) Representative traction stress maps for BCR cells plated on substrates coated with C-Cad.fc. Colors express the magnitude of traction stress (× 10 4 dyn/cm2) mapped to cell regions. (N) Average traction stress per cell (+/- S.D.) for control, blebbistatin treated and dnWnt11 expressing cells. (*) p < 0.001. (ns) not significant. (O-P) Distribution of traction stress for control and dnWnt11 expressing cells on (O) FN and (P) C-cad.fc substrates. Vertical axis values represent percentage of total pixels with a specific stress value. For panels (N-P) averages were calculated from 10-15 individual cell measurements. Frequencies of low (open arrow) and high (closed arrow) stress attributed to dnWnt11 and control cells, respectively, are indicated. (Q) Western blot to compare cell surface levels of C-cad and β1 integrins in intact biotinylated BCRs, (+) or (-) dnWnt 11.

Figure 2

Figure 2. Cadherin adhesion regulates FN fibril assembly

Confocal projections (15 μM) of BCRs from embryos fixed at (A-D) early (st. 10) or (E-H) late (st. 11.5) gastrula stage. BCRs were stained with antibodies against FN (green), C-cad (blue), and phalloidin to visualize filamentous actin (red). (A,E,I) Composite of all 3 fluorochromes. At stage 10, C-cad is noted at cell edges (B) and in linear arrays perpendicular to the cell edge (B, arrows), little cortical actin is present (C) and FN is arranged in punctae at cell surfaces (D). Cells in the BCR are rounded at stage 10. By stage 11, cells are polygonal in shape (compare B and F), actin staining is predominantly cortical (G) and FN is arranged in a fibrillar network (H). (I-L) Embryos expressing N-cad at stage 10 are similar to control embryos at stage 11.5; cells are polygonal (J), actin is organized at the cell cortex (K), and FN fibrils are assembled at the cell surface (L). Scale bar=25 μm in (L). (M,N). Saggital confocal sections of stage 10 embryos injected in one blastomere at the two cell stage with transcripts encoding a myc-epitope tagged N-cad (M), compared to an uninjected sibling control (N). Embryos were immunostained with antisera to FN (green) and mAb 9E10 to detect N-Cad-myc (red). (M, arrows) FN assembles in ectopic locations at the borders between non-expressing cells and cells expressing N-Cad. (N, arrows) In normal embryos FN is restricted to the surface facing the blastocoel cavity. Scale bar =50 μm in (N). (O, P). BCRs dissected from stage 11 embryos injected in one blastomere at the two cell stage with mRNA encoding (O) a dominant negative cadherin construct consisting of the IL2 receptor extracellular domain fused to the C-cad transmembrane and cytoplasmic domains or, (P) uninjected sibling control embryos. Isolated BCRs were immunostained with pAb32 to FN (green) and a mAb to the IL2 receptor (red). (O) FN matrix is absent from cells (red) expressing the dominant negative cadherin while, (O, P) FN fibril assembly is underway in control cells at this stage. Scale bar =50μm in (P). (Q) Western blot of cell surface levels of C-cad and β1 integrins in intact BCRs over expressing C-cad or dn C-cad.

Figure 3

Figure 3. Traction force mapping of single vs. clustered cells on FN

(A-H) BCR cells on FN. (A-B, E-F) control cells and (C-D, G-H) cells overexpressing C-Cad. (A,C,E,G) Traction stress maps and (B,D,F,H) phase contrast images of same field. Colors express the magnitude of traction stress (× 10 4 dyn/cm2) mapped to cell regions. Note that the range of stresses represented by the color scale in (A-D) is different from scale in (E-H). Scale bars = 25 μm. (I) Cell-cell contact increases the average traction stress on FN exerted by both control cells and cells over expressing C-cad. Average traction stress per cell (+/-S.D.). (*) denotes statistical significance (p < 0.001). (ns) not significant. (J) Distribution of traction stress comparing single and clustered cells over expressing C-cad on FN substrates. Vertical axis represents percentage of total pixels with a certain stress value. Data derived from a total of 30 individual cell measurements (groups of 4-9 cells/cluster) per condition. Single cell data in (I) derived from 9 cells (3 separate experiments).

Figure 4

Figure 4. Rac activity is required for cadherin to promote FN assembly

Embryos injected with mRNA encoding N-cadherin and either (A-C) dominant negative RacN17 or (D-F) RhoN19 were fixed at stage 10. BCRs immunostained with (A,D,C,F, green) anti-FN pAb 32 or, (B,E,C,F, red) mAb 9E10 to detect areas of myc epitope-tagged N-cad. Co-expression of dominant negative Rac but not dominant negative Rho inhibited the fibril promoting effects of N-cad expression (compare A-C to D-F). Bar=25μm. (G) Western blot analysis of expression levels of myc epitope-tagged RacN17 Rac and RhoN19. β-actin used as loading control.

Figure 5

Figure 5. Rac and Pak activity are required for FN fibril assembly

(A-D) Confocal images of BCRs from stage 11.5 embryos expressing dominant negative Rac N17 and immunolabeled with (A green, B) pAb32 to FN, (A blue, D) mAb 9E10 to detect cells expressing myc epitope tagged RacN17 and, (A red, C) phalloidin. Cells expressing RacN17 are unable to assemble FN fibrils. Actin filaments are detected throughout the cytoplasm and are no longer enriched at the periphery of cells. (E-H) Control or (I-L) inhibitory Pak peptides were delivered at blastula stage 8 by intrablastocoelic injection and the embryos cultured to late gastrula stage. (E-G, I-K) Confocal images of stage 11.5 BCRs immunostained as in (A-C). Inhibition of both FN fibril assembly and dramatic reorganization of the actin cytoskeleton was observed in embryos injected with the Pak inhibitory peptide (compare F,G to J,K). (H) Phospho-MLC (pMLC) was detected at cell peripheries in stage-matched sibling embryos injected with control peptide. (L) pMLC was greatly reduced in Pak peptide injected BCR cells. (M) Western blot of embryo lysates from control (-) and Pak peptide (+) injected embryos. β-actin included as loading control. (N) Embryos injected with control peptide develop normally while (O) embryos injected with the PAK inhibitory peptide were shortened along the anterior posterior axis. (P-S) Confocal images of BCRs from stage 11.5 embryos immunostained for (P-green, Q) FN, and (P-red, S) phospho-Pak and, (R) F-actin with phalloidin. (P) Phosphorylated Pak detected as punctae that extensively colocalize with FN fibrils. Scale bar in (A) and (P) =25μm. Scale bar in (N) =1 mm.

Figure 6

Figure 6. Tissue tension regulates FN assembly

(A) Embryos were punctured with a fine glass pipette open at both ends prior to the onset of gastrulation (stage 10), and left in place until stage 11.5 with exogenous FN present in the culture medium. (B) Sibling control embryos were cultured under the same conditions without puncturing. BCRs were fixed and immunostained (C,D) using mAb 4H2, and (E,F) pAb 881 to detect FN and α5 integrin, respectively. (C) FN is detected at cell edges in punctured embryos but (D) arranged in fibrils in controls. (E) α5 integrins localized to cell membranes in punctured embryos as in controls (F), but control cells are arranged in a more polygonal pattern. Scale bar in (F) =50 μm. (G-J) Increased tension on the BCR promotes assembly. Individual mid-blastula (st. 8) embryos were (G) forced into a nitex grid or (H) placed on top of the grid and cultured until stage 10 and fixed. (I) FN fibrils are precociously assembled on the BCR in a pattern that corresponds to the perimeter of the nitex grid where the embryo is compressed. (J) pre-fibrillar FN is localized to cell edges in controls, typical of normal embryos at this stage. Scale bar in (J) =75 μm.

Figure 7

Figure 7. Model for regulation of FN assembly in tissues

The current view of FN matrix assembly is based on observations of cells (e.g., fibroblasts) in culture. The cell is anchored to the actin cytoskeleton via integrins in focal adhesions (light blue). An increase in local mechanical tension is transmitted to α5β1 integrins (blue) with bound dimeric FNs (green) and as integrin/FN complexes move centripetally (arrows) from focal adhesions, cryptic sites in FN are exposed that promote FN-FN assembly. On the BCR and in epithelia, the analogous role of the focal adhesion in culture is supplanted by associations of the actin cytoskeleton with cadherins (red) at adherens junctions. Local mechanical tension is transmitted to α5β1 with bound FN dimers. Cryptic sites in FN are exposed and promote FN-FN assembly as integrin/FN complexes move centripetally (arrows) from sites of cell-cell contact and across the free surfaces of adjacent cells in a Rac/Pak dependent process. It is possible that a single FN dimer could be bound to integrins on adjacent cells moving in opposite directions or to two integrins within a single cell, the latter of which is more similar to movement out of the focal contact in cultured cells.

Comment in

Similar articles

Cited by

References

    1. Adams CL, Chen YT, Smith SJ, Nelson WJ. Mechanisms of epithelial cell-cell adhesion and cell compaction revealed by high-resolution tracking of E-cadherin-green fluorescent protein. J Cell Biol. 1998;142:1105–1119. - PMC - PubMed
    1. Adams CL, Nelson WJ, Smith SJ. Quantitative analysis of cadherin-catenin-actin reorganization during development of cell-cell adhesion. J Cell Biol. 1996;135:1899–1911. - PMC - PubMed
    1. Arthur WT, Noren NK, Burridge K. Regulation of Rho family GTPases by cell-cell and cell-matrix adhesion. Biol Res. 2002;35:239–246. - PubMed
    1. Baneyx G, Baugh L, Vogel V. Fibronectin extension and unfolding within cell matrix fibrils controlled by cytoskeletal tension. Proc Natl Acad Sci U S A. 2002;99:5139–5143. - PMC - PubMed
    1. Bershadsky A. Magic Touch: how does cell-cell adhesion trigger actin assembly? Trends in Cell Biology. 2004;14:589–593. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources