Fig 2: Comparison of proteins extracted from five different tissues with EXB and EXB Plus, respectively. A Using EXB Plus considerably higher lysate concentrations could be obtained for each of the tissue types. B Western blot analysis of E-Cadherin, Erk and ß-actin in five different tissues. For both buffers we obtained clear bands in the western blot, confirming that we were able to extract non-degraded, full-length, and immunoreactive proteins. In addition it is distinguishable that the extracted protein amount from all five tissues of all three analysed proteins is higher using EXB Plus.

Fig 3: Actin bundling by Eps8 promotes cortex tension and intracellular pressure to drive leader bleb formation.(A) Schematic representation of wild-type (WT) mouse Eps8 (mEps8) mutant constructs. Top: wild-type Eps8. ‘*’ indicates one part of the split PH domain, ‘dSH2’ indicates degenerate SH2, ‘capping’ indicates domain required for actin capping and which is subject to negative regulation by the Raf/MEK/Erk pathway, ‘bundling’ indicates the domain required for actin bundling. ‘?bund’ indicates the bundling defective double point mutant L757A/K759A (red bars), ‘?cap’ indicates the capping defective double point mutant V689D/L693D (red bars), ‘SATA’ indicates double alanine point mutation of Erk phosphorylation sites S624A/T628A (red bars). (B–I) Images and analyses of A375 cells plated on glass and confined under an agar slab. (C, D, G–I) Tukey box plots in which ‘+’ and line denote the mean and median, respectively. (C, D) Quantification of leader bleb area expressed as a % of cell body area for (C) cells treated with non-targeting or Eps8 siRNAs, with or without the additional expression of Emerald-mEps8, Emerald-mEps8?bund or EGFP human a-actinin or (D) cells over-expressing (OE) Emerald-mEps8, Emerald-mEps8?bund or EGFP human a-actinin. Data for mEps8 are re-displayed from Figure 2 for comparison. (E) Quantitation of the percent of cells that migrate from time-lapse phase contrast videos, treatments as in (C). (F) Confocal images through the ventral Z-plane of a cell depleted of Eps8 and co-expressing Emerald-mEps8?bund and either FusionRed-F-tractin (top) or FusionRed-myosin II regulatory light chain (bottom, RLC). (G) Analysis of cortical actin density (see Materials and methods) in the cell body from images of phalloidin, treatments as in (D), normalized to the mean value of over-expression of soluble EGFP (EGFP alone). (H, I) Cortex tension (H) and intracellular pressure (H) determined from AFM analysis of cells under the conditions described in (D). *p = 0.05, **p = 0.01, ***p = 0.001, ****p = 0.0001, NS: p > 0.05. See also Video 4.DOI: http://dx.doi.org/10.7554/eLife.08314.014

Fig 4: Inhibition of Eps8 capping activity is not sufficient to maintain leader blebs in the presence of Erk inhibitor.(A) Time-lapse confocal image series of an A375 cells plated on glass and confined under agarose cell that was treated with siRNA directed towards human Eps8 (hEps8 siRNA) and additionally expressing Emerald-tagged mouse Eps8 bearing double point mutations V689D/L693D that inactivate its actin capping activity (green, ‘mEps8 ?cap’). Time indicates minutes relative to perfusion with 10 µM U0126 to inhibit Erk.DOI: http://dx.doi.org/10.7554/eLife.08314.026

Fig 5: MEK/Erk-mediated phosphorylation of S624 and T628 coordinates Eps8 capping and bundling activities to mediate leader bleb-based migration.(A–J) Images and analyses of A375 cells plated on glass and confined under an agar slab. (A) Color encoded time-overlay of confocal images through the central Z-plane of a cell depleted of Eps8 by siRNAs targeting human Eps8 (hEps8 siRNA) and expressing Emerald-mEps8 bearing double alanine point mutation of Erk phosphorylation sites S624A/T628A (Emerald-mEps8-SATA). (B, C, F, I, J) Tukey box plots in which ‘+’ and line denote the mean and median, respectively. (B–C) Quantification of leader bleb area expressed as a % of cell body area for (B) cells treated with hEps8 siRNAs and additionally expressing wild type (WT) Emerald-mEps8 or Emerald-mEps8-SATA or (C) cells over-expressing (OE) Emerald-mEps8-WT or Emerald-mEps8-SATA. (D) Quantitation of the percent of cells that migrate from time-lapse phase contrast videos, treatments as in (B). (E) Confocal images through the central Z-plane of cells depleted of Eps8 and co-expressing Emerald-mEps8-SATA and either FusionRed-F-tractin (top) or FusionRed-myosin II regulatory light chain (bottom, RLC). (F) Analysis of cortical actin density (see Materials and methods) in the cell body from images of phalloidin, treatments as in (C), normalized to the mean value from over-expression of Emerald-mEps8 WT. (G) Confocal images through the central Z-plane of cells treated with 10 µM U0126 and over-expressing either Emerald-mEps8 WT or Emerald-mEps8?bund (see Figure 4A) and actin stained with fluorescent phalloidin. (H) Quantitation of the percent of cells containing central actin bundles from confocal images of fluorescent phalloidin-stained cells. ‘F-tractin’ indicates over-expression of FusionRed-F-tractin, ‘UO’ indicates treatment with 10 µM U0126, ‘mEps8,’ ‘mEps8?cap,’ ‘mEps8?bund,’ ‘mEps8-SATA,’ indicate over-expression of EGFP-tagged versions of the respective constructs (see Figure 4A). (I, J) Cortex tension (I) and intracellular pressure (J) determined from AFM analyses of cells under the conditions described in (C). *p = 0.05, **p = 0.01, ***p = 0.001, ****p = 0.0001, NS: p > 0.05. See also Video 7.DOI: http://dx.doi.org/10.7554/eLife.08314.024