A to H. Image analysis for the invadopodia screen. An example of DAPI-stained nuclei (blue channel) is shown in B, while A shows the corresponding actin (red channel) image. The overlay of both channels is displayed in 1.3. The image background for the red channel was subtracted and a 3-by-3 median filter was applied to smooth the image removing fine actin fibers (D). Nuclei detection (Cytoshop) was applied (E) and nuclear parameters were extracted from the segmented DAPI image. From the nuclei masks (E) cellular regions were assigned to each nucleus using tessellation (G, Cytoshop). The red channel was segmented using Cytoshop’s fluorescent aggregate detection to find rosettes (F). Finally, the detected nuclei, assigned cellular regions and detected rosettes were merged (H) and the number of detected rosettes was calculated on a cell-by-cell basis. Representative rosettes of invadopodia are highlighted by white arrows in C.
I. Screening results. 1280 active compounds were screened in duplicate. Cytotoxic compounds were removed from the active compound list. Thresholds for inhibition and activation were applied. Visual confirmation led to a final list of 7 active compounds.
J. Active compound validation. Src-3T3 cells were plated on fluorescently labeled gelatin-coated coverslips. Representative images of gelatin degradation (lower row) and F-actin staining (upper row) are shown.

A. EC50 measurements for invadopodia and cell cycle inhibition. Src-3T3 cells were treated with several concentrations of purvalanol A (Purv A) then assayed for invadopodia rosette formation by phalloidin staining (F-actin) and cell cycle distribution by FACS analysis. The image analyses for two representative concentrations of purvalanol A (0.2µM, 5µM) are shown as insets.
B. Abundance of Cdk5 in Src-3T3 cells compared with the parental 3T3 cell line. All measurements were calculated as percent of control (3T3) and error bars calculated as propagated standard errors of the mean of triplicate measurements from each experiment; *, p<0.05.
C. Subcellular localization of Cdk5. Box indicates a rosette showing colocalization of F-actin and Cdk5. The boxed rosette is shown in the lower row in X–Y, Y–Z and X–Z planes as imaged by confocal fluorescence microscopy with Z-sectioning.
D. Cdk5 knockdown in Src-3T3 cells. Scrambled or Cdk5-specific siRNAs cells were assayed for invadopodia rosette formation and function.
E. Quantitation of the percentage of cells with invadopodia rosettes and gelatin degradation. Measurements were calculated as described in Figure 2.
F. The effect of overexpression of Cdk5/p35 on purvalanol A inhibition of invadopodia rosette formation. Empty vector or plasmids expressing human Cdk5/p35 were transfected into Src-3T3 cells.

A. Src-3T3 cells transfected with scrambled (scr), Cdk5- or Mek1-specific siRNAs were immunoblotted with phospho-527 specific (P-cald), total caldesmon (cald), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibodies. Band intensities were quantified with the use of ImageJ software version 1.34 (http://rsb.info.nih.gov/ij/).
B. Src-3T3 cells were treated with either vehicle or MG132 (40µM) for 4 hours (left lanes) or 24 hours (right lanes), then lysates immunoblotted for total caldesmon (cald) and GAPDH.
C. 3T3 cells co-transfected with wild-type (wt) caldesmon together with either an empty vector or Cdk5/p35, were treated with MG132 (40µM) or vehicle for 4h, then subjected to immunoblotting with caldesmon (cald) and GAPDH antibodies.
D. Src-3T3 cells transfected with SA or SD caldesmon constructs and treated with MG132 (40µM) or vehicle for 4h were subjected to immunoblotting with caldesmon (cald) and GAPDH antibodies.
E. Effect of downregulation of caldesmon on the EC₅₀ for purvalanol A-mediated invadopodia inhibition. Caldesmon was silenced in Src-3T3 cell using specific siRNA. Purvalanol A was added at the indicated concentrations and the cells were fixed, stained and analyzed as described in Materials and Methods.
F. Quantitation of the effect of transfection of caldesmon WT, S527A or S527D on invadopodia formation. Measurements were calculated as described in Figure 2.
G. Quantitation of FITC-gelatin degradation of SCC61 cells transfected with scrambled or caldesmon specific siRNAs and subsequently treated with MG132 for 4 hours.

A. SCC61 cells were grown on gelatin-coated coverslips, treated with DMSO or 2µM purvalanol A (PurvA), GCP74514A (GCP), paclitaxel and cantharidin (Canth), then assayed for invadopodia formation (F-actin) and gelatin degradation (FITC-gelatin). Top, 63X for F-actin staining; bottom, 20X for labeled gelatin and DAPI staining. Representative invadopodia are highlighted with white arrows in the upper left panel.
B. Quantitation of invadopodia formation (invadopodia-positive cells and number of invadopodia per cell) in SCC61 cells treated with the indicated compounds.
C. Quantitation of FITC-gelatin degradation (percent degraded area normalized to cell number) by SCC61 cells treated with the indicated compounds, as described in Materials and Methods.
D. TR-SKOV3 cells were grown on coverslips, treated with DMSO or compounds at 2µM, and then assayed for invadopodia formation using F-actin staining.
E. TR-SKOV3 cells were treated with 2µM paclitaxel and assayed for matrigel invasion as described in Materials and Methods.
All measurements were calculated as percent of control (DMSO) and error bars calculated as propagated standard errors of the mean of triplicate measurements from each experiment; *, p<0.05.

A. Cdk5 knockdown in SCC61 cells. Cells were infected with scrambled control (scr) or three different Cdk5 shRNA (1465, 1466, 1467) and assayed for invadopodia formation by phalloidin staining (F-actin) and gelatin degradation (FITC-gelatin) (top). Quantification of percent of invadopodia-positive cells, number of invadopodia per cell and gelatin degradation is shown on the bottom. Measurements were calculated as described in Figure 2.
B. Immunoblot showing the abundance of Cdk5 in transfected cells.