The network of signals arising from transmembrane receptor engagement converge on p190-A. (A) Summary of the connections from α₅β₁ integrin and syndecan-4 during the initiation of focal adhesion formation. (B) The engagement of receptors, including integrins and cadherins, results in tyrosine phosphorylation of p190-A (gray box). The primed p190-A molecule is itself a target for serine and threonine phosphorylation events, tyrosine dephosphorylation events downstream of syndecan-4 and growth factor receptors, and trafficking molecules such as p120RasGAP. Although there is extensive crosstalk between these pathways, p190-A stands out as a nexus for the multiple signaling events during RhoA regulation.

Engagement of α₅β₁ integrin and syndecan-4 each contribute to efficient suppression of RhoA during spreading on fibronectin. (A) Rho-A activity of MEFs spreading on fibronectin (broken line) or 50K (solid line) measured by ELISA. (B) RhoA activity of syndecan-4–null MEFs spreading on fibronectin. (C) Hypothetical RhoA activity calculated by adding the percentage suppression caused by H/0 on prespread cells to the suppression caused during spreading on 50K. Fourth-order polynomial curves were fitted to each dataset. Experiments were repeated on at six separate occasions. Equal loading between time points was confirmed by a total protein assay. Error bars represent standard error. *, significant change from initial activity (P < 0.05).

p190-A is recruited to lamellae in response to engagement of syndecan-4. Fibroblasts prespread on a recombinant integrin ligand (50K) were stimulated with heparin-binding fragments of fibronectin (H/0) or a heparin-binding mutant (H/0-GAG) before fixation and staining for p190-A (green) and the lipid raft marker GM-1 (red). (A) Immunostaining of p190-A and GM-1, including masks used to measure the pixel intensity of the lamellar border. Bar, 5 μm. (B–E) Pixel intensity of p190-A (B and D) and GM-1 (C and E) within a 0.5-μm-thick border around the edge of each lamella after stimulation with H/0 (B and C) or H/0-GAG (D and E). The pixel intensity of 20 different cells was measured for each time point. The experiment was repeated on six separate occasions. Error bars represent standard error. *, significant change (P < 10⁻⁸).

α₅β₁ integrin and syndecan-4 regulate p190-A by distinct mechanisms. (A) Phosphotyrosine blots of p190-A immunoprecipitated from high-detergent lysates of fibroblasts prespread on 50K and stimulated with H/0. (B) TrX-soluble p190-A blots of fibroblasts prespread on 50K, inhibited with 10 μM PP2, and stimulated with H/0. (C) Phosphotyrosine blots of p190-A immunoprecipitated from the cytosolic and membrane-bound fractions of fibroblasts prespread on 50K, stimulated with H/0 for 10 min, and lysed by sonication. Loading was adjusted to analyze equal amounts of p190-A from each fraction. Phosphotyrosine blots of p190-A immunoprecipitated from high-detergent lysates of fibroblasts fully spread on fibronectin and 50K (D), spreading on fibronectin (E), spreading on 50K (F), or syndecan-4–null MEFs spreading on fibronectin (G). (H) Cytosolic and membrane-bound p190-A in sonication lysates of fibroblasts spreading on 50K. Equal loading between time points was confirmed by blotting for HSP70 in redistribution experiments and by reprobing blots for p190-A in tyrosine phosphorylation experiments. Error bars represent standard error. *, significant change (P < 0.05) from experiments repeated on at least four separate occasions.

p190-A mediates RhoA regulation in response to engagement of syndecan-4. MEFs prespread on a recombinant integrin ligand (50K) were stimulated with heparin-binding fragments of fibronectin (H/0) before measuring the RhoA activity of lysates by ELISA. Wild-type MEFs (A), p190-A–null MEFs (B), p190-A–null MEFs reexpressing p190-A (C), MEFs transfected with a nontargeting siRNA (E), and MEFs transfected with an siRNA targeted against p190-A (F) are shown. (G) Direct comparison of RhoA activity in cells fully spread on 50K after transfection with a nontargeting or p190-A–targeted siRNA. (D and H) Expression levels of p190-A in null and reexpressing MEFs (D) or MEFs transfected with siRNAs (H); lysates were also probed for the related p190-B isoform to eliminate the possibility of molecular compensation or off-target knockdown. Equal loading between time points was confirmed by a total protein assay. Error bars represent standard error. *, significant change (P < 0.05) from experiments repeated on at least four separate occasions.

PKCα mediates syndecan-4–induced redistribution of p190-A. MEFs prespread on a recombinant integrin ligand (50K) were stimulated with heparin-binding fragments of fibronectin (H/0) before preparing lysates and then measuring levels of soluble p190-A in TrX lysates by Western blotting. Wild-type (A) and syndecan-4–null (B) MEFs reexpressing a Y188L mutant syndecan-4 defective in PKCα binding to endogenous levels, treated with 200 nM of the pharmacological PKC inhibitor BIM-1 (C), transfected with a nontargeting siRNA (D), and transfected with an siRNA targeted against PKCα (E). (F) Expression of PKCα after transfection with control or targeting siRNA. (G) Phosphotyrosine blots of p190-A immunoprecipitated from high-detergent lysates of MEFs expressing wild-type or Y188L-syndecan-4 spread on 50K and stimulated with H/0 for 10 min. (H) Phosphorimager scan and blot of p190-A immunoprecipitated from high-detergent lysates of MEFs expressing wild-type or Y188L–syndecan-4, metabolically labeled with [³²P]orthophosphate spread on 50K, and stimulated with H/0 for 10 min. RhoA activity ELISAs of MEFs expressing wild-type (I) or Y188L–syndecan-4 (J) prespread on 50K and stimulated with H/0. Equal loading between time points was confirmed by blotting for HSP70 or a total protein assay. Error bars represent standard error. *, significant change (P < 0.05) from experiments repeated on at least five separate occasions.

Engagement of syndecan-4 regulates membrane association of p190-A. Fibroblasts prespread on a recombinant integrin ligand (50K) were stimulated with heparin-binding fragments of fibronectin (H/0) before preparing lysates and measuring redistribution of proteins between soluble and membrane-bound fractions or RhoA activity by Western blotting. (A and B) Soluble, tyrosine-phosphorylated protein in lysates prepared in 0.5% (wt/vol) TrX or 1% (wt/vol) TrX, 0.5% (wt/vol) deoxycholate, and 0.1% (wt/vol) SDS. Arrowheads indicate a 190-kD band that varied in abundance between conditions. (C and D) TrX-soluble p190-A in lysates after stimulation with H/0 or a heparin-binding mutant (H/0-GAG). (E) Cytosolic and membrane-bound p190-A in sonication lysates. (F) Relative distribution of p190-A and cytosolic and membrane-bound markers, tailless complex polypeptide 1α and transferrin receptor, between membrane and cytosolic fractions of sonication lysates prepared from equal numbers of cells spread on 50K and stimulated with H/0 for 10 min. (G) Correlation between RhoA activity (dashed line) and membrane-associated p190-A (solid line). Equal loading between time points was confirmed by blotting for vinculin or HSP70. Error bars represent standard error. *, significant change (P < 0.05) from experiments repeated on at least four separate occasions.

Expression of p190-A is necessary for focal adhesion formation in response to syndecan-4 engagement. (A) Wild-type, syndecan-4–null, siRNA-transfected, p190-A–null, or p190-A reexpressing MEFs prespread on a recombinant integrin ligand (50K) were stimulated with heparin-binding fragments of fibronectin (H/0) for 60 min before fixing and staining for vinculin (green) and/or actin (red). Whole cell images are shown in Fig. S2 A (available at http://www.jcb.org/cgi/content/full/jcb.200711129/DC1). Bar, 10 μm. (B) Focal adhesion area per cell was quantified digitally, after background subtraction, by measuring the area above an empirically determined threshold of fluorescence intensity within a single experiment. (C) Total area of individual cells. White bars represent cells spread on 50K and black bars represent spread cells after stimulation with H/0. (D and E) Correlation between membrane-bound p190-A (circles), RhoA activity (squares), total focal adhesion area per cell (crosses), and mean focal adhesion length (triangles) in human fibroblasts (D) and MEFs (E). (F) Percentage of human fibroblasts spread on 50K after 2 h in the presence of inhibitory antibodies against α₅ integrin (mAb16), α_v integrin (17E6), or nonimmune IgG. Experiments were repeated on four separate occasions. Error bars represent standard error. *, significant change (P < 0.005) from 20 cells per condition.