Fig 2: Mutant p53 is associated with high Rac1 activity in human tumor samples. (A) p53 mutations are associated with high Rac1 activity in human colorectal tumors. The levels of p-PAK1/2 were analyzed by IHC staining using an anti-p-PAK1/2 antibody. (Top panel) Representative images of p-PAK1/2 IHC staining in human colorectal tumor tissues. (High staining) More than 30% of cells stained with p-PAK1/2; (low staining) 10%–30% of cells stained with p-PAK1/2; (negative staining) <10% of cells stained with p-PAK1/2. Bar, 100 µm. (Bottom panel) Correlation between p53 mutations and Rac1 activity in human colorectal tumors. n = 98. P = 0.0001, ?2 test. (B) The high levels of p-PAK2Ser141 in metastatic castration-resistant prostate cancer lesions obtained at rapid autopsy (n = 8) with missense p53 point mutations as compared with tumors containing wild-type p53 or p53 deletions (n = 6). P = 0.02. Each sample was evaluated via the mass spectrometer with technical replicates, and both data points are displayed. Unpaired t-test with Welch's correction was used to compare the two groups.

Fig 3: Mimicking phosphorylation of Cortactin restores invadopodia dynamics in p27-/- cells.(A–E) p27-/- E6 MEFs were infected with empty vector or with vectors encoding wild type Cortactin (WT), S113A-Cortactin (S113A) or S113D-Cortactin (S113D). (A) Cortactin levels after retroviral infection were determined by immunoblot with rabbit anti-Cortactin (H-191) antibodies. ß-actin was used as loading control. (B–E) Cells were seeded on gelatin-A488 for 48 hr. After Tks5 staining, cells forming invadopodia (B–C), or the area of degraded gelatin, expressed in fold-change compared to WT Cortactin transfected cells (D–E), were quantified in at least ten fields per condition in each experiment, representing a minimum of 179 cells per genotype. The graphs show the means of at least three independent experiments. (F–J) p27-/- E6 MEFs were infected with empty vector or with vectors encoding WT Cortactin, Cortactin TA (S113A/S150A/S282A) or Cortactin TD (S113D/S150D/S282D). (J) Cortactin levels after retroviral infection were determined as in (A). (G–J) Cells were processed as in (B–E) to quantify cells forming invadopodia (G–H), or the area of degraded gelatin (I–J), with a minimum of 222 cells counted per genotype per experiment. The graphs show the means of 3 independent experiments. (K) Schematic representation of the Rac1/PAK1/phospho-Cortactin pathway involved in invadopodia turnover and matrix degradation and its proposed regulation by p27. ****p<0.0001; ***p<0.001; **p<0.01; *p<0.05.DOI: http://dx.doi.org/10.7554/eLife.22207.03210.7554/eLife.22207.033Figure 8—source data 1.Quantification of cells forming invadopodia (Figure 8B–C) and degraded gelatin area (Figure 8D–E) after infection with S113 phospho-mutants of Cortactin; quantification of cells forming invadopodia (Figure 8G–H) and degraded gelatin area (Figure 8I–J) after infection with triple phospho-mutants of Cortactin; quantification of P-Ser Cortactin/Cortactin ratio (Figure 8—figure supplement 1B).DOI: http://dx.doi.org/10.7554/eLife.22207.03310.7554/eLife.22207.034Figure 8—source data 2.Statistical analyses for Figure 8.DOI: http://dx.doi.org/10.7554/eLife.22207.03410.7554/eLife.22207.035Figure 8—source data 3.Statistical analyses for Figure 8—figure supplement 1B.DOI: http://dx.doi.org/10.7554/eLife.22207.03510.7554/eLife.22207.036Figure 8—source data 4.Mascot search results for Cortactin for Figure 8—figure supplement 2.DOI: http://dx.doi.org/10.7554/eLife.22207.036

Fig 4: Active PAK is essential for bacterial attachment and pedestal formation. (A) Immunoblot of the level of total PAK and active (i.e., phosphorylated on serine 144) PAK (PAK-P) in Hap1 cells infected with the indicated strains of EPEC. (B) Attachment of WT EPEC to Hap1 cells pretreated with DMSO (control) or the PAK inhibitor G5555, NVS-PAK-1-1, or IPA3. Values are relative to attachment to control Hap1 cells (typically 6 to 7 per cell). Each bar represents the average of results from 3 separate experiments (300 to 500 cells for each experiment). Error bars indicate SD. (C) Attachment of WT EPEC to ?PAK1 and ?PAK2 Hap1 cells, with and without additional knockdown of the remaining PAK isoform using siRNA. Values are relative to attachment in WT Hap1 cells (there are typically 6 to 7 per cell). Each bar represents the average of results from 3 separate experiments (300 to 500 cells for each experiment). Error bars indicate SD. (D) Fluorescence microscopy images of EPEC pedestals on cells infected as described for panel C. Actin is stained with Alexa Fluor 488-phalloidin (green), and bacteria are stained with an anti-intimin antibody (red). Scale bar, 1 µm. ***, P < 0.001; **, P < 0.01 (one-way ANOVA followed by a post hoc Dunnett comparison) relative to control (B) or WT Hap1 (C) cells.

Fig 5: Shikonin inhibits PAK1-related signaling pathways in pancreatic cancer cells. (A) BxPC-3 and PANC-1 cells were treated with shikonin for indicated concentrations, and Western blot assay was conducted using indicated antibodies. (B) The expression of p-PAK1 and PAK1 was quantified by densitometry analysis and normalized against GAPDH. (C) BxPC-3 and PANC-1 cells were treated with shikonin for indicated time points, and Western blot assay was conducted using indicated antibodies. (D) The expression of p-PAK1 and PAK1 was quantified by densitometry analysis and normalized against GAPDH. The experiments were conducted at least twice.