Src activity regulates invadopodia formation in HNSCC lines. (A) Protein levels of active c-Src (Src-pY⁴¹⁸), total c-Src (Src) and β-actin (loading control) in HNSCC lines. For quantification of active Src and total Src, expression levels were normalized to MSK921 cells, a line with low Src activity that does not form invadopodia. The multiple bands in the Src-pY⁴¹⁸ panel presumably represent additional Src Family kinases; see Fig. 3A. (B) HNSCC cell lines with or without Src-527F were incubated on FITC-gelatin (pseudocolored white) coverslips for 12 hours and labeled with TRITC–phalloidin (red) and anti-cortactin (green). Invadopodia are identified by the yellow aggregates in the merged images of actin and cortactin (white arrows) that localize with the dark holes in the FITC-gelatin (black arrows). Scale bars: 10 μm.

Inhibition of endogenous Src expression decreases matrix degradation independently of changes in invadopodia number or increased Src activity. (A) UMSCC1 cells transfected with siRNA directed against Src (SrcSi) or a non-targeting siRNA (Ctl) were evaluated for Src knockdown 2 and 3 days after transfection by anti-Src western blotting. β-actin blotting was used as a loading control. (B) Cell lysates from UMSCC1 cells transfected with non-targeting siRNA (Ctl) or SRC siRNA (SrcSi) alone, or in combination with cerulean-tagged SrcWT (WT) or Src-527F (527) were evaluated by immunoblotting. Lysates were probed with anti-Src and anti-β-actin antibodies. Filled arrowhead indicates exogenously expressed Src (WT or 527F), open arrowhead denotes endogenous Src. (C) Representative confocal images of UMSCC1 cells transfected with non-targeting siRNA (Ctl) or SrcSi alone, or in combination with cerulean-tagged SrcWT (WT) or Src-527F (527). Cells were plated on FITC-gelatin-coated (pseudocolored white) coverslips for 10 hours and immunolabeled with TRITC–phalloidin (red) and cortactin (green). Scale bars: 10 μm. (D) Percentage of cells displaying invadopodia (top), the number of invadopodia per cell (middle) and the amount of matrix degradation per cell (bottom) were examined for each line evaluated in B. Data are presented as mean ± s.e.m.; ^#, statistically different from control (P≤0.01); groups under the bar are statistically different (P<0.01).

Invadopodia produced by constitutively active Src in SYF fibroblasts fail to degrade ECM. (A) Src family kinase expression in HNSCC and fibroblast cell lines. Clarified cell lysates from SYF, SYF^+/+, UMSCC1 (treated with control and SrcSi), and OSC19 (expressing control vector and SRC shRNA) cells were resolved by SDS-PAGE and immunoblotted with anti-Src, anti-Yes, anti-Fyn and anti-β-actin antibodies. (B) Time course of tsLa29 v-Src activation and resulting cortactin phosphorylation. Cells transfected with tsLa29 were incubated at 35°C (permissive temperature) for the indicated times and were lysed or returned to 41°C (non-permissive temperature) for 15 or 30 minutes and analyzed for Src-pY⁴¹⁸, GFP, cortactin, and cortactin-pY⁴²¹. (C) Invadopodia formation in cells expressing v-Src. SYF cells were transfected with tsLa29–GFP (pseudocolored light blue) and incubated at 41°C or 35°C and labeled with TRITC–phalloidin (red) and cortactin (green). Cells were visualized by confocal microscopy and z-stack sectioning. Invadopodia are visible in z-stack images as actin- and cortactin-rich puncta that are several micrometers in length (white arrows). Scale bars: 10 μm, 5 μm (z-stacks). (D) Defective ECM degradation in SYF cells expressing v-Src. SYF cells transfected with tsLa29-v-Src and non-transfected UMSCC1 cells (positive control) were plated directly onto FITC-gelatin-coated coverslips. After 24 hours, cells were labeled with TRITC–phalloidin and anti-cortactin antibodies. White arrows indicate invadopodia and black arrows, areas of degraded matrix in C and D. Scale bars: 10 μm.

WT Src rescues invadopodia maturation in cells expressing constitutively active Src. (A) SYF, SYF^+/+ and UMSCC1 cells that were non-transfected (NT), transfected with Src-527F or ts-v-Src, were lysed, resolved by SDS-PAGE and immunoblotted with anti-Src-pY418, anti-Src clone EC10 (only recognizes avian Src), anti-cortactin-pY421, anti-cortactin (4F11), and anti-β-actin antibodies. (B) SYF, SYF^+/+ and UMSCC1 cells that were non-transfected (NT), transfected with SRC-527F, or tsLa29-v-Src were plated onto FITC-conjugated gelatin coverslips and evaluated by confocal microscopy. Cells were labeled with TRITC–phalloidin (red), cortactin (green), and anti-phosphotyrosine (light blue) antibodies. Invadopodia are identified by the yellow aggregates in the merged image of actin and cortactin in cells containing phosphotyrosine. In addition, degrading (mature) invadopodia localize with the dark holes (arrows) in the FITC-gelatin (white). (C) Quantification of the percentage of cells forming invadopodia structures (actin and cortactin aggregates, left), and the percentage of invadopodia-forming cells that contain matrix degradation (mature invadopodia, right). Data are represented as mean ± s.d.; *P≤0.05.

Expression of regulated Src is necessary for ECM degradation at invadopodia. (A) Confocal imaging of SYF cells coexpressing Src–Cerulean and Src527F–mCherry, or Src-295M–Cerulean and Src527F–mCherry. Cells were immunolabeled with anti-cortactin and Alexa-Flour-647-labeled phalloidin. Dashed boxes represent enlarged image regions shown below. Arrows denote invadopodia. (B) Representative fields of SYF and SYF^+/+ cells expressing Src527F–mCherry, SYF cells coexpressing Src527F–mCherry and Src–Cerulean, or SYF cells coexpressing Src527F–mCherry and Src295M–Cerulean. Cells were plated on FITC-gelatin-coated coverslips for 24 hours and immunolabeled with anti-cortactin. Areas of gelatin degradation and clearing appear black against the pseudocolored white background. (C) Percentage of cells with forming invadopodia (actin and cortactin aggregates, left), and percentage of invadopodia-forming cells with degraded matrix (mature invadopodia, right) from the experimental conditions shown in B. Data are represented as mean ± s.d.; *P≤0.05. (D) Confocal imaging (top) or swept-field imaging (bottom) of SYF cells coexpressing c-Src527F and Src–Cerulean. Cells were plated on FITC-gelatin-coated coverslips and immunolabeled with anti-cerulean (dark blue) and anti-cortactin (yellow) (colocalization appears white in merged image; top panel) or directly imaged for Src527F–mCherry (red) and Src–Cerulean (dark blue) (bottom panels). Arrow indicates colocalization of Src 527F–mCherry and WT Src–Cerulean at sites of gelatin degradation. (E) Confocal imaging of Src activity in SYF and SYF^+/+ cells expressing Src527F. Cells were incubated on FITC-gelatin-coated coverslips for 24 hours, fixed and immunolabeled with TRITC–phalloidin (red), anti-pY418 Src (light blue) and anti-Src (green) antibodies. White arrows denote invadopodia; black arrows, matrix degradation. Scale bars: 20 μm (B), 10 μm (A,D,E).

WT Src regulates cortactin phosphorylation during invadopodia maturation. (A) SYF and SYF^+/+ cells expressing Src527F were incubated on FITC-gelatin-coated coverslips for 24 hours, fixed and immunolabeled with TRITC–phalloidin (red), human anti-pY421 cortactin (light blue) and anti-cortactin (green) antibodies. Note that cortactin is phosphorylated in pre-invadopodia and in mature invadopodia, as defined by the absence or presence of matrix degradation. (B) Validation of GFP-tagged cortactin expression in stable cell lines. Cellular extracts from SYF or SYF^+/+ cells stably expressing GFP–CortWT or GFP–CortTYM were resolved by SDS-PAGE and immunoblotted with an anti-cortactin (4F11) antibody. *, endogenous cortactin; **, GFP-tagged cortactin (the slight mobility disparity is due to the size difference in AcGFP and EGFP tags); ratios of exogenous (WT and TYM) to endogenous cortactin is depicted under the blot. (C) Clarified lysates from SYF^+/+ GFP–CortWT and SYF^+/+ GFP–CortTYM cells were transfected with Cttn siRNA alone or in combination with Src527F. Cortactin was immunoprecipitated with the anti-cortactin (4F11) antibody, immunocomplexes resolved by SDS-PAGE and immunoblotted with human anti-pY421 cortactin and anti-cortactin (4F11) antibodies. Total cell lysates were immunoblotted with anti-β-actin as a loading control. (D) SYF^+/+ cells stably expressing human GFP–CortWT or GFP–Cort TYM were transfected with murine cortactin-targeted siRNA to eliminate endogenous cortactin expression. Two days later cells were transfected with Src527F and plated on FITC-gelatin-coated coverslips for 24 hours to promote gelatin degradation. (E) Cells were assessed for the percentage of cells forming invadopodia (actin and cortactin aggregates) and the percentage of invadopodia-forming cells with matrix degradation. Data are represented as mean ± s.d., *P<0.05. Scale bars: 10 μm (A), 20 μm (B).

Enhanced binding of SH2 domains binding in SYF^+/+ cells expressing Src527F. In vitro SH2 and PTB domain-binding profiles for SYF, SYF^+/+ cells, and SYF and SYF^+/+ cells expressing Src527F. Binding of SH2 or PTB domains to cell lysates are shown as a heat map where intensity of red indicates strength of binding. Domain names are on the top row, with a rank order of binding to SYF^+/+ Src527F to denote potential importance in invadopodia maturation. Sample labels are on the left side. Difference=Signal[SYF^+/+ Src527F]–Signal[SYF Src527F].