Carcinoma cell motility and invasion are prerequisites for tumor cell metastasis, which requires regulation of the actin cytoskeleton. Cortactin is an actin-related protein 2/3 (Arp2/3) complex–activating and filamentous (F)-actin–binding protein that is implicated in tumor cell motility and metastasis, partially by its ability to become tyrosine phosphorylated. Cortactin is encoded by the CTTN gene and maps to chromosome 11q13, a region amplified in many carcinomas, including head and neck squamous cell carcinoma (HNSCC). CTTN gene amplification is associated with lymph node metastasis and poor patient outcome, and cortactin overexpression enhances motility in tumor cells lacking 11q13 amplification. However, a direct link between increased motility and invasion has not been reported in tumor cells with chromosome 11q13 amplification and cortactin overexpression. In this study, we have examined the relationship between CTTN amplification and tumor cell motility in HNSCC. In 11 of 39 (28%) HNSCC cases, cortactin overexpression determined by immunohistochemistry correlates with lymph node metastasis and CTTN gene amplification. HNSCC cells containing cortactin gene amplification and protein overexpression display increased binding and activation of Arp2/3 complex, and were more motile and invasive than HNSCC cells lacking CTTN amplification. Down-regulation of cortactin expression in CTTN-amplified HNSCC cells by small interfering RNA impairs HNSCC motility and invasion. Treatment of HNSCC cells with the epidermal growth factor receptor inhibitor gefitinib inhibits HNSCC motility and down-regulates cortactin tyrosine phosphorylation. These data suggest that cortactin may be a valid prognostic and therapeutic marker for invasive and metastatic HNSCC and other carcinomas with 11q13 amplification. (Cancer Res 2006; 66(16): 8017-25)

Tumor cell motility is required for local-regional invasion and distant metastasis. Motility is initiated by protrusion of the leading edge of the cell, resulting in the production of polarized lamellipodia that are oriented toward the direction of movement. Production and extension of lamellipodia occurs in response to a wide variety of extracellular ligands that signal through transmembrane receptors (1). Lamellipodia serve to orient tumor cells toward the vasculature in response to chemotatic agents and are required for tumor cell invasion (2).

Chemotatic receptor activation results in increased Rho GTPase and Src family kinase activity, which in turn leads to the activation of Wiskott-Aldrich syndrome proteins (WASp) that drive lamellipodia protrusion (3). Activation of WASps by GTPase activation or tyrosine phosphorylation exposes a cryptic carboxyl-terminal domain containing an acidic region that binds and activates the actin-related protein 2/3 (Arp2/3) complex. Arp2/3 complex activity results in nascent polymerization of actin filaments at the cell periphery, providing the protrusive force responsible for lamellipodia extension at the leading edge (4). Regulation of the actin cytoskeleton by Arp2/3 complex activity has been proposed as a mechanism for controlling tumor cell migration, invasion, and metastasis (5).

In addition to WASps, another protein that binds and activates Arp2/3 complex is cortactin, a Src kinase substrate and filamentous (F)-actin–binding protein that is enriched in lamellipodia (6, 7). Cortactin consists of an amino-terminal acidic domain followed by a repeat region, a proline-rich region containing sites of tyrosine and serine phosphorylation, and a distal Src homology 3 (SH3) domain that interacts with a variety of proline-rich proteins (6, 7). Cortactin directly activates Arp2/3 complex actin nucleation activity through the binding of the amino-terminal acidic domain region to Arp2/3 complex and the repeat region to F-actin (7). Nucleation activity and lamellipodia protrusion is further enhanced by the binding of the SH3 domain to dynamin 2 or WASp-interacting protein (8, 9). The cortactin SH3 domain also binds and activates neuronal (N)-WASp, which in turn promotes Arp2/3 actin nucleation and cell motility (10, 11). Additionally, cortactin modulates cell movement by the phosphorylation of three specific tyrosine residues (421, 470, and 486 in humans) by Src and other oncogenic tyrosine kinases (12). Tyrosine-phosphorylated cortactin is found in lamellipodia (13) and correlates with enhanced cell motility and tumor cell metastasis in nude mice (14). Other functions of cortactin include stabilization of Arp2/3–F-actin networks and participation in receptor-mediated endocytosis (15, 16).

Genomic instability is a hallmark of cancer progression and results in the accumulation of specific genetic events. Amplification of chromosome 11q13 is found in several tumors and is one of the most prevalent genetic alterations in head and neck squamous cell carcinoma (HNSCC), occurring in approximately one third of all cases (17). Amplification of 11q13 correlates with several important clinical variables, including increased invasive disease, regional metastasis, more frequent recurrence, and increased disease-specific mortality (18). The human cortactin locus CTTN (formerly designated EMS1) is located within the 11q13 region and cortactin overexpression resultant from gene amplification is found in HNSCC and other carcinomas (19, 20). CTTN amplification and overexpression directly correlates with poor outcome in HNSCC patients (21). Based on these data, it has been proposed that cortactin is the critical gene in the amplified chromosome 11q13 region responsible for enhancing carcinoma motility and invasion (22).

Indirectly supporting this, several recombinant overexpression studies in cells lacking 11q13 amplification have established cortactin as key regulator in migration, invasion, and metastasis (12, 23, 24). To determine the role cortactin plays in motility and invasion in the context of tumors with 11q13 amplification, we have investigated the effect of cortactin overexpression in HNSCC cells with and without 11q13 amplification and have determined that cortactin overexpression derived from 11q13 amplification results in increased Arp2/3 complex activity, tumor cell motility, and invasion. We also show that cortactin tyrosine phosphorylation is required for efficient HNSCC migration, and that inhibition of cortactin tyrosine phosphorylation with the targeted therapeutic drug gefitinib reduces HNSCC motility and corresponds with decreased cortactin tyrosine phosphorylation. These results indicate that cortactin directly contributes to elevated motility and invasion in HNSCC tumors with 11q13 amplification by increasing Arp2/3 actin nucleation activity, and that cortactin tyrosine phosphorylation may serve as a potential diagnostic marker for validation of targeted therapeutic agents against tyrosine kinase pathways involved in tumor cell migration and invasion.

Tumor samples. Formalin-fixed, paraffin-embedded tissue blocks from 10 cases of HNSCC were retrieved from the University of Colorado Health Sciences Center Department of Pathology archives. All tissue and case report information was used under approval from the Colorado Multiple Institutional Review Board. An additional 29 HNSCC cases were analyzed from a HNSCC tissue microarray (YTMA7) constructed at Yale University Department of Pathology.

Immunohistochemistry. Five-micrometer sections cut from tissue blocks and the HNSCC microarray were processed for immunohistochemical labeling as described (25), except that sections were incubated with 0.4% pepsin for 10 minutes at 37°C for antigen retrieval. The pan-specific cortactin monoclonal antibody 4F11 and avian-specific control cortactin antibody 1H3 (26) were used at 1 μg/mL. Samples were visualized with an Olympus AX70 microscope and images were acquired using the MicroBrightField system (Williston, VT).

Fluorescence in situ hybridization. Dual-color fluorescence in situ hybridization (FISH) analysis was conducted to determine CTTN copy numbers in tumor specimens and in HNSCC cell lines. Two P1 genomic cortactin clones (20) were used in combination with a control probe for the centromeric region of chromosome 11 (CEP 11 SpectrumGreen, Vysis/Abbott Molecular, Des Plaines, IL). The CTTN P1 clones were labeled by nick translation with dUTP-conjugated SpectrumRed, combined with the SpectrumGreen CEP 11, and hybridized as described (27).

Tumor specimens were quantitated for CTTN gene amplification by epifluorescence, with cases containing a ratio of >2 CTTN/CEP 11 copies per nucleus considered amplified. For determining cortactin gene amplification in HNSCC cell lines, mitotic arrest was achieved by treating the cells with colcemid (0.05 μg/mL) for 2 hours and chromosome elongation by treatment with ethidium bromide (5 μg/mL) for 1 hour. Cells were trypsinized, lysed with 75 mmol/L potassium chloride, and fixed with 3:1 (v/v) mixture of methanol and acetic acid. Cells were dropped on glass slides and probes were hybridized and processed as described (28). The number of CTTN and CEP 11 probe signals were determined for a minimum of 100 nuclei in each cell line by epifluorescence microscopy.

Cell lines, transfection, and Western blotting. HNSCC lines UMSCC2, UMSCC8, UMSCC10A, UMSCC19, and HN4 were maintained in DMEM (Mediatech, Herndon, VA) containing 10% fetal bovine serum (FBS; HyClone, Logan, UT), 50 μg/mL penicillin-streptomycin, and nonessential amino acids. The MSK921 and 584 lines were cultured in l-glutamine–free RPMI (Mediatech) supplemented with 10% FBS and 50 μg/mL penicillin-streptomycin. For transfection, 106 cells were incubated with 2 μg FITC-labeled control small interfering RNA (siRNA); siRNA targeted against human cortactin (Qiagen, Valencia, CA); pEGFPN1 green fluorescent protein (GFP) vector (BD Clontech, Palo Alto, CA); GFP-cortactin (9); FLAG-wild type or FLAG-Y421F, FLAG-Y466F, FLAG-Y482F [triple point mutant (TPM)] cortactin (13) in solution V (Amaxa Biosystems, Berlin, Germany); and transfected with the Nucleofector I device (Amaxa Biosystems) according to the instructions of the manufacturer. Cells were plated and allowed to recover for 24 to 48 hours before use in motility, invasion, or Western blot experiments. Cell lysis and Western blot analysis was done as previously described (13). The following antibodies and final concentrations were used: 4F11 (0.5 μg/mL), anti-ARPC2 (p34 subunit of Arp2/3 complex, 1:1,000; Upstate, Waltham, MA), anti-β-actin (1:5,000; Calbiochem, La Jolla, CA), Living Colors anti-GFP clone JL-8 (1:1,000; BD Clontech), and antiphosphotyrosine 4G10 (1:1,000; Upstate). Scanned blots were quantified using ImageQuant TL software (Amersham Biosciences, Piscataway, NJ). Blots were done in triplicate.

Arp2/3 pulldown and actin incorporation assays. The recombinant glutathione-S-transferase (GST)-verprolin cofilin homology acidic (VCA) domain of N-WASp was expressed and purified using standard techniques. Cell lysates (1 mg) were incubated with 50 μg GST-VCA prebound to 40 μL glutathione-agarose (50% slurry; Pierce, Rockford, IL) for 3 hours at 4°C, washed, and processed for Western blotting with 4F11 and anti-ARPC2 antibodies. Assays were done in triplicate. Globular (G)-actin incorporation (barbed end) assays on serum-starved cells were done essentially as described (29), except Alexa-488–G-actin (Invitrogen, Carlsbad, CA) was used and cells were stimulated for 10 minutes with 100 ng/mL epidermal growth factor (EGF; Upstate). Fixed cells were imaged with a Nikon TE2000 epifluoresence microscope equipped with a Cool-Snap f/x charge-coupled device camera (Roper Scientific, Tucson, AZ). Quantification of actin incorporation was done using a script written in the Metamoph v6.0 software suite (Molecular Devices, Downingtown, PA). The script identifies peripheral areas around the cell with pixel intensities above background and creates a 3-pixel-thick ring over the regions of circumferential fluorescent actin incorporation. Border widths were adjusted manually as required to compensate for varying degrees of cortical actin thickness. Quantification of G-actin incorporation was taken as the total fluorescence intensity divided by the surface area of the measured region. Fifty EGF-stimulated and nonstimulated cells for each line were analyzed in three independent experiments.

Cell motility and invasion assays. Transwell cell migration and Matrigel-coated invasion assays were conducted using 8 μm pore size chambers (Fisher Scientific, Pittsburgh, PA and BD Transduction, San Jose, CA) essentially as described (30). Cells (105) were added to each upper chamber and cells were allowed to migrate or invade for 24 hours. In some experiments, 1 μmol/L gefitinib (Iressa, AstraZeneca, Macclesfield, Cheshire, United Kingdom) was added to both chambers at the beginning of the experiment. Fixed cells were stained with 3 μg/mL 4′,6-diamidino-2-phenylindole (DAPI) and motility was assessed by counting nuclei in four random microscopic fields through the use of an automated Metamorph script that accurately determines cell count by delineating cell shapes and dividing by the average nuclear surface area. For siRNA rescue experiments, cell motility was measured using electric substrate impedance sensing (Applied BioPhysics, Troy, NY), as described (31), using 40 kHz voltage pulse for initial cell ablation and 10 ng/mL EGF as the motile stimulus. Impedance was measured for 24 hours. A minimum of three independent experiments were done in all assays.

Statistical analysis.P values were determined by the unpaired, two-tailed, Student's t test.

Cortactin overexpression in invasive HNSCC. Although cortactin gene amplification and transcription has been documented with PCR-based methods in HNSCC and correlates with poor prognosis (21), cortactin localization and CTTN amplification status in HNSCC have not been visualized and correlated to invasive capacity. To address this, we analyzed cortactin expression in 39 cases of primary HNSCC by immunohistochemistry and CTTN amplification by FISH analysis from archived tissue sections and a HNSCC tissue microarray (Fig. 1). In 11 cases (28%), cortactin overexpression was detected in the tumor cell cytoplasm, with marked enrichment of the protein at the cell periphery and at the leading edge of cells in tumors that displayed local-regional invasion and cervical lymph node metastasis (Fig. 1A). Serial sections treated with a monoclonal antibody specific to avian cortactin as a control were not significantly stained (Fig. 1A). All tumors that overexpressed cortactin also contained CTTN amplification as determined by FISH (Fig. 1B). Noninvasive, node-negative tumors had little cortactin staining and did not contain CTTN amplification (Fig. 1). These data indicate that cortactin overexpression and gene amplification in HNSCC correlates with an invasive phenotype.

Figure 1.

Immunohistochemical and FISH analysis of cortactin overexpression and CTTN amplification in invasive HNSCC. A, serial sections of tumor samples from two patients (invasive-1 and invasive-2) with node-positive, invasive HNSCC, and sections from one patient (non-invasive) with node-negative, noninvasive HNSCC were processed for immunohistochemistry with pan-species reactive (cort IHC) or chicken-specific (ctl) cortactin monoclonal antibodies and counterstained with hematoxylin. Arrowheads, areas of cortical cortactin enrichment. Bar, 100 μm. B, analysis of cortactin overexpression and CTTN copy levels in a HNSCC tissue microarray. Top, cortactin overexpression was determined by immunohistochemistry as in (A, cort IHC), with overexpression shown for two representative cases (amp-1 and amp-2) and normal expression shown in a third case (disomic). Arrowheads, cortical localization. Bottom, determination of CTTN levels by FISH (cort FISH). A serial section of the HNSCC microarray was hybridized with probes against CTTN (red) and the centromeric region of chromosome 11 (green). Blue, nuclei visualized by DAPI staining. Arrows, regions of clustered amplification for the cortactin gene. Bar for immunohistochemistry images, 100 μm. Bar for FISH images, 20 μm.

Figure 1.

Immunohistochemical and FISH analysis of cortactin overexpression and CTTN amplification in invasive HNSCC. A, serial sections of tumor samples from two patients (invasive-1 and invasive-2) with node-positive, invasive HNSCC, and sections from one patient (non-invasive) with node-negative, noninvasive HNSCC were processed for immunohistochemistry with pan-species reactive (cort IHC) or chicken-specific (ctl) cortactin monoclonal antibodies and counterstained with hematoxylin. Arrowheads, areas of cortical cortactin enrichment. Bar, 100 μm. B, analysis of cortactin overexpression and CTTN copy levels in a HNSCC tissue microarray. Top, cortactin overexpression was determined by immunohistochemistry as in (A, cort IHC), with overexpression shown for two representative cases (amp-1 and amp-2) and normal expression shown in a third case (disomic). Arrowheads, cortical localization. Bottom, determination of CTTN levels by FISH (cort FISH). A serial section of the HNSCC microarray was hybridized with probes against CTTN (red) and the centromeric region of chromosome 11 (green). Blue, nuclei visualized by DAPI staining. Arrows, regions of clustered amplification for the cortactin gene. Bar for immunohistochemistry images, 100 μm. Bar for FISH images, 20 μm.

Close modal

Cortactin overexpression and gene amplification correlate in HNSCC cells. To determine the relationship between cortactin overexpression and gene amplification in HNSCC cells, we quantified cortactin protein and gene levels in several HNSCC cell lines where cortactin expression and/or chromosome 11q13 amplification status had been partially defined (Fig. 2; refs. 3234). Cortactin protein and gene levels were standardized against UMSCC10A, a line normal for cortactin expression compared with nontransformed epithelial cells (not shown) and disomic for CTTN. Western blot analysis indicated a 6- to 12-fold increase in cortactin expression for lines UMSCC19, UMSCC2, and MSK921 (Fig. 2A and B). The HN4 line had a 2-fold increase in cortactin expression, whereas UMSCC8 and UMSCC584 contained normal cortactin levels (Fig. 2A and B). The amount of Arp2/3 complex for each line was also determined with an antibody against the ARPC2 complex subunit. In all lines, no significant differences in Arp2/3 expression were detected (Fig. 2A). Cytogenetic analysis in metaphase spreads indicated that these lines were hypo-diploid to near-triploid for CTTN. Quantification of the CTTN locus by FISH indicated that UMSCC19, UMSCC2, and MSK921 contained clustered CTTN amplification, with mean copy numbers ranging from 8 to over 20 per cell (Fig. 2C). All other lines had on average two to four CTTN copies per cell. With the exception of the HN4 line, the level of cortactin protein expression directly correlates with CTTN copy number in each line (Fig. 2B and C). These data indicate that cortactin expression and gene levels are tightly coupled, and that amplification of CTTN correlates with cortactin overexpression in HNSCC.

Figure 2.

Determination of cortactin protein and CTTN levels in HNSCC cell lines. A, Western blot analysis of cortactin and Arp2/3 complex expression. Equal amounts of cell extracts from HNSCC cell lines (top) were subjected to SDS-PAGE and Western blotted with antibodies against cortactin, against the ARPC2 subunit of Arp2/3 complex, and against β-actin as a loading control. B, quantification of protein levels in HNSCC cell lines. Western blots described in (A) were scanned and quantified with the cortactin/β-actin ratio for each cell line determined for three independent experiments. Columns, averaged values normalized to the UMSCC10A cortactin/β-actin values (designated as 1.0) for control purposes; bars, SE. *, P < 0.05. C, CTTN levels in HNSCC cell lines. Fixed cells were analyzed by FISH using a combination of two independent probes containing genomic CTTN sequences conjugated to SpectrumRed. The chromosome 11 centromere was visualized using a specific centromeric probe conjugated to SpectrumGreen. Columns, average number of CTTN genes per cell, for lines with normal or near-normal ploidy levels (UMSCC10A, UMSCC8, UMSCC584, and HN4). For cell lines with clustered cortactin CTTN amplification (UMSCC2, UMSCC19, and MSK921), the lower limit of each average value is shown. Bars, SE. *, P < 0.001.

Figure 2.

Determination of cortactin protein and CTTN levels in HNSCC cell lines. A, Western blot analysis of cortactin and Arp2/3 complex expression. Equal amounts of cell extracts from HNSCC cell lines (top) were subjected to SDS-PAGE and Western blotted with antibodies against cortactin, against the ARPC2 subunit of Arp2/3 complex, and against β-actin as a loading control. B, quantification of protein levels in HNSCC cell lines. Western blots described in (A) were scanned and quantified with the cortactin/β-actin ratio for each cell line determined for three independent experiments. Columns, averaged values normalized to the UMSCC10A cortactin/β-actin values (designated as 1.0) for control purposes; bars, SE. *, P < 0.05. C, CTTN levels in HNSCC cell lines. Fixed cells were analyzed by FISH using a combination of two independent probes containing genomic CTTN sequences conjugated to SpectrumRed. The chromosome 11 centromere was visualized using a specific centromeric probe conjugated to SpectrumGreen. Columns, average number of CTTN genes per cell, for lines with normal or near-normal ploidy levels (UMSCC10A, UMSCC8, UMSCC584, and HN4). For cell lines with clustered cortactin CTTN amplification (UMSCC2, UMSCC19, and MSK921), the lower limit of each average value is shown. Bars, SE. *, P < 0.001.

Close modal

HNSCC cells containing cortactin overexpression have increased Arp2/3 binding and activation. Our findings that cortactin is enriched at the cell periphery when overexpressed in invasive tumors and that Arp2/3 levels are unchanged suggests that cortactin overexpression resulting from chromosome 11q13 amplification may increase the amount of cortactin associated with Arp2/3 complex, resulting in enhanced Arp2/3 actin nucleation activity. To test this possibility, we conducted Arp2/3 pulldown experiments from HNSCC cells with and without cortactin overexpression using a N-WASp recombinant GST-VCA domain, a region of N-WASP that simultaneously binds with cortactin to different subunits of Arp2/3 (35). A 1.5- to 2-fold increase in cortactin binding to Arp2/3 complex was observed in cell lines UMSCC2 and UMSCC19 that overexpress cortactin compared with UMSCC10A and UMSCC8 (Fig. 3A). A similar increase in binding was observed in the overexpressing line MSK921 (data not shown). These results suggested that the increased association between Arp2/3 and cortactin may result in increased cortactin-mediated Arp2/3 actin nucleation activity. To test this, serum-starved HNSCC cells were permeabilized with saponin in the presence of Alexa-488–G-actin and then stimulated with EGF to allow the incorporation of labeled actin into nascent filament ends (Fig. 3B and C). This method allows for determining the degree of Arp2/3 actin-nucleation activity within newly forming lamellipodia downstream of EGF receptor (EGFR) activation (36). In all HNSCC lines examined, maximal incorporation of Alexa-488–G-actin into the peripheral cell edge occurred 10 minutes after EGF stimulation, resulting in circumferential rings of varying fluorescence intensities (Fig. 3B). Levels of EGFR were similar in all lines examined (data not shown). Quantitative microscopic analysis of this region for the MSK921 and UMSCC19 lines showed a nearly 2-fold increase in cortical actin polymerization compared with the UMSCC8 and 584 lines (Fig. 3C). Similar results were observed in UMSCC2 cells, although precise localization of labeled actin within lamellipodia was difficult to discern in this cell line (data not shown). These data indicate that HNSCC cells with cortactin overexpression are enhanced in their ability to bind and activate Arp2/3 complex at the cell periphery.

Figure 3.

Cortactin overexpression in HNSCC enhances Arp2/3 actin nucleation activity. A, enhanced association of Arp2/3 complex in tumor cells overexpressing cortactin. Top, 1 mg total cell extracts from the indicated cell lines was incubated with GST-N-Wasp VCA domain prebound to glutathione agarose beads, pelleted, and washed. Bound proteins were detected by Western blotting with antibodies against cortactin and the ARPC2 subunit of Arp2/3 complex. Bottom, quantification of cortactin associated with Arp2/3 complex. Western blots were quantified and the cortactin/ARPC2 ratio was determined for each cell line for three independent experiments. Columns, averaged values normalized to the UMSCC10A/ARPC2 values (designated as 1.0 for control purposes); bars, SE. *, P < 0.05. B, increased EGF-mediated actin polymerization in HNSCC cells overexpressing cortactin. Cell lines containing normal cortactin levels (UMSCC8 and UMSCC584) or cortactin overexpression (MSK921 and UMSC19) were plated onto fibronectin-coated coverslips, serum starved, and either mock-treated (−EGF) or stimulated with EGF (+EGF) in the presence of saponin and Alexa 488–conjugated G-actin. Cells were fixed and visualized for actin incorporation (Actin). Images were evaluated for cortical actin incorporation by the application of a 3-pixel-wide border programmed to cover peripheral areas of actin incorporation and calculate the average pixel intensity of the entire border (orange outline, Actin+Periphery). Representative images for lines UMSCC 8 and UMSCC 19. C, quantification of peripheral actin incorporation. Columns, average pixel intensity for the cortical regions, quantified as described in Materials and Methods; bars, SE, *, P < 0.02.

Figure 3.

Cortactin overexpression in HNSCC enhances Arp2/3 actin nucleation activity. A, enhanced association of Arp2/3 complex in tumor cells overexpressing cortactin. Top, 1 mg total cell extracts from the indicated cell lines was incubated with GST-N-Wasp VCA domain prebound to glutathione agarose beads, pelleted, and washed. Bound proteins were detected by Western blotting with antibodies against cortactin and the ARPC2 subunit of Arp2/3 complex. Bottom, quantification of cortactin associated with Arp2/3 complex. Western blots were quantified and the cortactin/ARPC2 ratio was determined for each cell line for three independent experiments. Columns, averaged values normalized to the UMSCC10A/ARPC2 values (designated as 1.0 for control purposes); bars, SE. *, P < 0.05. B, increased EGF-mediated actin polymerization in HNSCC cells overexpressing cortactin. Cell lines containing normal cortactin levels (UMSCC8 and UMSCC584) or cortactin overexpression (MSK921 and UMSC19) were plated onto fibronectin-coated coverslips, serum starved, and either mock-treated (−EGF) or stimulated with EGF (+EGF) in the presence of saponin and Alexa 488–conjugated G-actin. Cells were fixed and visualized for actin incorporation (Actin). Images were evaluated for cortical actin incorporation by the application of a 3-pixel-wide border programmed to cover peripheral areas of actin incorporation and calculate the average pixel intensity of the entire border (orange outline, Actin+Periphery). Representative images for lines UMSCC 8 and UMSCC 19. C, quantification of peripheral actin incorporation. Columns, average pixel intensity for the cortical regions, quantified as described in Materials and Methods; bars, SE, *, P < 0.02.

Close modal

HNSCC cells with 11q13 amplification display increased cell motility and invasion. To determine if HNSCC cells with 11q13 amplification and cortactin overexpression display increased cell motility and invasion, transwell migration and Matrigel invasion assays were conducted (Fig. 4). No significant changes in cell division rates were observed between all lines during the assay, allowing for direct comparisons to be made. Two different chemoattractants, EGF and FBS, were evaluated because all HNSCC lines used overexpress EGFR (except UMSCC10A) and EGFR overexpression is clinically known to correlate with increased HNSCC motility and invasion (37). Increased migration and invasion was observed with UMSCC19, UMSCC2, and MSK921 cells toward 10 ng/mL EGF compared with all other lines tested (Fig. 4). Similar results were obtained when 10% FBS was used as the chemoattractant, with some variability in the motility and/or invasive response observed in the UMSCC8 and MSK921 lines when compared with EGF (Fig. 4). Collectively, these data suggest that 11q13 amplification and cortactin overexpression correlates with increased HNSCC motility and invasion.

Figure 4.

Increased motility and invasion in HNSCC cells with chromosome 11q3 amplification and cortactin overexpression. HNSCC cells were plated in uncoated transwells (top) or Matrigel-coated wells (bottom) and induced to migrate with either 10 ng/mL EGF or 10% FBS for 24 hours. Columns, number of cells that migrated (top) or invaded (bottom) into the lower chamber; bars, SE for a minimum of three independent experiments. P < 0.02 for motility and P < 0.05 for invasion.

Figure 4.

Increased motility and invasion in HNSCC cells with chromosome 11q3 amplification and cortactin overexpression. HNSCC cells were plated in uncoated transwells (top) or Matrigel-coated wells (bottom) and induced to migrate with either 10 ng/mL EGF or 10% FBS for 24 hours. Columns, number of cells that migrated (top) or invaded (bottom) into the lower chamber; bars, SE for a minimum of three independent experiments. P < 0.02 for motility and P < 0.05 for invasion.

Close modal

Cortactin directly modulates HNSCC motility and invasion. The 11q13 locus is a gene-rich region that upon amplification may express proteins besides cortactin that can contribute to cell motility and invasion in HNSCC. To determine the contribution of cortactin in HNSCC motility and invasion, cortactin expression was down-regulated in HNSCC cells with 11q13 amplification by transfection with a cortactin-specific siRNA. The degree of cortactin knockdown varied between the different cell lines tested, ranging to near-complete knockdown for UMSCC2 (97%), 83% for UMSCC19, and 40% for MSK921 (Fig. 5A). This variation is reflective of differing transfection efficiencies among the tested lines as determined by control transfections with a plasmid encoding GFP (data not shown). UMSCC2 cells were inhibited in motility 47% toward EGF and 43% serum when compared with internal, cell line–specific controls (Fig. 5B). The corresponding values for UMSCC19 were 51% and 52%; for MSK921, the values were 34% and 9%, respectively. Invasion was also inhibited in cells with cortactin knockdown, with a 29% reduction for UMSCC2, 21% for UMSCC19, and 10% for MSK921 with EGF and the motile stimulus. These values were 43%, 65%, and 14%, respectively, when FBS was used (Fig. 5B). To further confirm that cortactin expression levels influence HNSCC motility and invasion, UMSCC8 and HN4 cells (lines without 11q13 amplification) were transfected with a construct expressing either GFP or GFP-cortactin. Expression of GFP-cortactin in these cells raised the total cortactin levels to 125% in UMSCC8 and 140% in HN4 (Fig. 5C). Motility and invasion for both lines expressing GFP-cortactin was increased 15% to 63% above the values for internal, GFP-expressing controls (Fig. 5D). These values are similar to what has been reported in other cell systems with similar methodology (23). Taken together, these data indicate that cortactin overexpression directly results in enhanced HNSCC motility and invasion in tumor cells with 11q13 amplification.

Figure 5.

Modulation of cortactin expression in HNSCC cells directly affects motility and invasion. A, down-regulation of cortactin in overexpressing HNSCC cells. Top, cells were transfected without RNA (mock), with cortactin siRNA (siRNA), or with a control siRNA conjugated to FITC (FITC-ctl). After 48 hours, cells were lysed and 50 mg total cell protein were assayed by Western blot analysis for knockdown of cortactin expression with anticortactin antibodies. Blots were stripped and reprobed with antibodies against β-actin as a loading control. Bottom, degree of cortactin knockdown in siRNA-treated cells was quantified by densitometry and expressed relative to cortactin levels in cells transfected with FITC-conjugated control siRNA. B, cortactin knockdown in overexpressing HNSCC cells impairs cell motility and invasion. HNSCC cells 48 hours after transfection with cortactin siRNA were assayed for transwell motility (top) and invasion with 10 ng/mL EGF or 10% FBS as mitogenic stimuli (bottom). Columns, percentage motility or invasion compared with each corresponding line transfected with FITC-conjugated control siRNA (control, 100%); bars, SE for a minimum of three independent experiments, *, P < 0.05. C, ectopic cortactin overexpression in HNSCC cells. HNSCC cells with normal cortactin expression (UMSCC8 and HN4) were transfected with GFP-cortactin or GFP control constructs for 24 hours and cell lysates were analyzed by Western blot analysis with an anti-GFP monoclonal antibody (top). Blots were stripped and subsequently probed with cortactin and β-actin antibodies. The positions of GFP-cortactin, GFP, and endogenous cortactin are indicated (right). Bottom, scanned blots were quantified for GFP-cortactin and endogenous cortactin expression, and the degree of total cortactin overexpression is shown relative to GFP transfected controls. D, ectopic cortactin overexpression increases HNSCC motility and invasion. UMSCC8 and HN4 cells expressing GFP or GFP-cortactin were analyzed for motility and invasion as in (B). Columns, percentage motility or invasion compared with each corresponding line transfected with GFP (control, 100%); bars, SE for a minimum of three independent experiments.

Figure 5.

Modulation of cortactin expression in HNSCC cells directly affects motility and invasion. A, down-regulation of cortactin in overexpressing HNSCC cells. Top, cells were transfected without RNA (mock), with cortactin siRNA (siRNA), or with a control siRNA conjugated to FITC (FITC-ctl). After 48 hours, cells were lysed and 50 mg total cell protein were assayed by Western blot analysis for knockdown of cortactin expression with anticortactin antibodies. Blots were stripped and reprobed with antibodies against β-actin as a loading control. Bottom, degree of cortactin knockdown in siRNA-treated cells was quantified by densitometry and expressed relative to cortactin levels in cells transfected with FITC-conjugated control siRNA. B, cortactin knockdown in overexpressing HNSCC cells impairs cell motility and invasion. HNSCC cells 48 hours after transfection with cortactin siRNA were assayed for transwell motility (top) and invasion with 10 ng/mL EGF or 10% FBS as mitogenic stimuli (bottom). Columns, percentage motility or invasion compared with each corresponding line transfected with FITC-conjugated control siRNA (control, 100%); bars, SE for a minimum of three independent experiments, *, P < 0.05. C, ectopic cortactin overexpression in HNSCC cells. HNSCC cells with normal cortactin expression (UMSCC8 and HN4) were transfected with GFP-cortactin or GFP control constructs for 24 hours and cell lysates were analyzed by Western blot analysis with an anti-GFP monoclonal antibody (top). Blots were stripped and subsequently probed with cortactin and β-actin antibodies. The positions of GFP-cortactin, GFP, and endogenous cortactin are indicated (right). Bottom, scanned blots were quantified for GFP-cortactin and endogenous cortactin expression, and the degree of total cortactin overexpression is shown relative to GFP transfected controls. D, ectopic cortactin overexpression increases HNSCC motility and invasion. UMSCC8 and HN4 cells expressing GFP or GFP-cortactin were analyzed for motility and invasion as in (B). Columns, percentage motility or invasion compared with each corresponding line transfected with GFP (control, 100%); bars, SE for a minimum of three independent experiments.

Close modal

Gefitinib inhibits HNSCC motility and down-regulates cortactin tyrosine phosphorylation. Cortactin tyrosine phosphorylation is required for efficient motility and metastasis (12, 14). EGFR and subsequent Src activation leads to phosphorylation of cortactin by Src (13). Because the EGFR-Src pathway is involved in HNSCC motility and invasion (38), we sought to determine if down-regulation of EGFR activation impairs HNSCC motility and if this correlated with reduced cortactin tyrosine phosphorylation. HNSCC cells treated with 1 μmol/L of the EGFR small-molecule inhibitor gefitinib were reduced in EGF-stimulated motility by over 80% (excluding the EGFR-null line UMSCC10A; Fig. 6A). Gefitinib inhibited motility in a similar but more variable manner when 10% serum was the chemoattractant, with UMSCC10a displaying 6% inhibition, UMSCC 93%, UMSCC2 77%, UMSCC19 61%, and MSK921 32% (Fig. 6B). Interestingly, the degree of motility inhibition by gefitinib in response to EGF and serum is inversely proportional to the level of cortactin expression in HNSCC (Fig. 6A,, B versus Fig. 2B). Cortactin immunoprecipitates from HNSCC lines treated with gefitinib and stimulated with 10% serum showed a marked reduction in tyrosine phosphorylation in all cases (Fig. 6C), indicating that cortactin tyrosine phosphorylation levels are modulated by EGFR activation in HNSCC, and that gefitinib-mediated inhibition of HNSCC cell migration correlates with down-regulated cortactin tyrosine phosphorylation.

Figure 6.

Gefitinib inhibits HNSCC motility and cortactin tyrosine phosphorylation. HNSCC cells were assayed for transwell motility toward 10 ng/mL EGF (A) or 10% FBS (B) in the absence or presence of 1 mmol/L gefitinib for 24 hours (top). Bars, SE for a minimum of three independent experiments. P < 0.05. The percentage motility for gefitinib-treated cells is shown relative to untreated control cells (bottom). C, gefitinib impairs cortactin tyrosine phosphorylation in HNSCC cells. Cortactin immunoprecipitated from HNSCC cells treated without or with 1 mmol/L gefitinib for 4 hours was assayed for tyrosine phosphorylation (pTyr, top) and cortactin expression (bottom) by Western blotting. D, cortactin tyrosine phosphorylation is required for efficient HNSCC migration. UMSCC2 cells were assayed for motility by electric substrate impedance sensing in the presence of 10 ng/mL EGF for 24 hours following transfection with control siRNA conjugated to FITC (FITC Control), cortactin siRNA (siRNA), murine FLAG–wild-type cortactin (Cortactin o/x), murine FLAG–wild-type cortactin with cortactin siRNA (Cortactin o/x + siRNA), and murine FLAG-TPM cortactin with cortactin siRNA (Cortactin TPM + siRNA). Columns, percentage motility compared with FITC control-transfected siRNA cells (100%); bars, SE for three independent experiments. *, P < 0.05 compared with FITC control cells.

Figure 6.

Gefitinib inhibits HNSCC motility and cortactin tyrosine phosphorylation. HNSCC cells were assayed for transwell motility toward 10 ng/mL EGF (A) or 10% FBS (B) in the absence or presence of 1 mmol/L gefitinib for 24 hours (top). Bars, SE for a minimum of three independent experiments. P < 0.05. The percentage motility for gefitinib-treated cells is shown relative to untreated control cells (bottom). C, gefitinib impairs cortactin tyrosine phosphorylation in HNSCC cells. Cortactin immunoprecipitated from HNSCC cells treated without or with 1 mmol/L gefitinib for 4 hours was assayed for tyrosine phosphorylation (pTyr, top) and cortactin expression (bottom) by Western blotting. D, cortactin tyrosine phosphorylation is required for efficient HNSCC migration. UMSCC2 cells were assayed for motility by electric substrate impedance sensing in the presence of 10 ng/mL EGF for 24 hours following transfection with control siRNA conjugated to FITC (FITC Control), cortactin siRNA (siRNA), murine FLAG–wild-type cortactin (Cortactin o/x), murine FLAG–wild-type cortactin with cortactin siRNA (Cortactin o/x + siRNA), and murine FLAG-TPM cortactin with cortactin siRNA (Cortactin TPM + siRNA). Columns, percentage motility compared with FITC control-transfected siRNA cells (100%); bars, SE for three independent experiments. *, P < 0.05 compared with FITC control cells.

Close modal

The ability of gefinitib to suppress HNSCC motility and cortactin tyrosine phosphorylation suggested that cortactin tyrosine phosphorylation contributes to HNSCC motility. To directly assess this, cortactin expression in UMSCC2 cells was reduced by siRNA and expression was restored by transfection of either wild-type FLAG-tagged murine cortactin or with a FLAG-cortactin construct containing phenylalanine to tyrosine point mutations at the Src-targeted codons 421, 466, and 482 (FLAG-TPM cortactin), previously shown to suppress cell motility (12). Overexpression of FLAG-wild type cortactin in UMSCC2 cells without cortactin knockdown led to a 260% increase in cell migration compared with control-transfected cells (Fig. 6D). This increase in motility is likely due to the additional high amount of recombinant murine cortactin expression driven by the cytomegalovirus promoter present in the recombinant cortactin constructs (data not shown). Using these cells as the control for recombinant murine cortactin expression, expression of FLAG-wild type murine cortactin following siRNA-mediated knockdown of endogenous cortactin expression resulted in a 140% reduction in cell migration compared with FLAG-wild type cortactin–overexpressing cells, reflecting the contribution of endogenous cortactin expression on UMSCC2 migration. Expression of FLAG-TPM cortactin in UMSCC2 cells following cortactin knockdown reduced migration 261% compared with the overexpression control, 121% compared with cells with FLAG-wild-type cortactin overexpression in the presence of endogenous cortactin knockdown, and 99% of the transfection control (Fig. 6D). Direct comparison between FLAG-wild type and FLAG-TPM cortactin motility in cells with endogenous cortactin knockdown indicates that FLAG-TPM cells migrate at 45% of the level of FLAG-wild type cortactin–expressing cells. These data indicate that cortactin phosphorylation at tyrosines 421, 466, and 482 is responsible for 55% of the motility ascribed to cortactin in UMSCC2 cells and confirms that cortactin tyrosine phosphorylation is an essential requirement for effective HNSCC motility.

Studies to date have shown that cortactin overexpression resulting from 11q13 amplification in HNSCC and other carcinomas corresponds with poor patient outcome, and that ectopic overexpression of cortactin in cancer cells lacking 11q13 amplification increases tumor cell motility, invasion, and metastasis. Our results directly show that cortactin overexpression resulting from 11q13 and corresponding CTTN amplification in carcinomas leads to increased tumor cell motility and invasion. In addition, our data indicate that these effects are likely due to increased binding between cortactin and Arp2/3 complex, potentially serving to enhance Arp2/3 actin nucleation activity within tumor cell lamellipodia. The resulting enhancement of lamellipodia protrusion would allow tumor cells to become more motile, increasing their ability to undergo local-regional invasion, intravasation and extravasation, and distant metastasis. The corresponding ability of the EGFR inhibitor gefitinib to suppress HNSCC motility and cortactin tyrosine phosphorylation indicates that tyrosine phosphorylation of cortactin may also serve as a valid therapeutic end point marker for metastatic efficacy in the treatment of HNSCC patients with inhibitory compounds targeting the EGFR pathway.

In agreement with previous work (19, 21), our results show that chromosome 11q13 amplification corresponds with cortactin overexpression in all examined HNSCC cases, correlating with tumor invasion and lymph node metastasis. The enrichment of cortactin within lamellipodia in invasive HNSCC cases is likely reflective of a role in enhancing tumor cell motility, as evidenced by the increased motility and invasion of HNSCC lines with cortactin overexpression. Moreover, in HNSCC cells, the degree of cortactin overexpression corresponds with the level of CTTN amplification, suggesting a direct dose dependency between gene amplification and protein expression. The ability of tumor cells to amplify and tolerate high cortactin levels is to date unique among Arp2/3-activating proteins, possibly due to the lower ability of cortactin to activate Arp2/3 complex compared with N-WASp at similar amounts (15). Slower activation of Arp2/3 by cortactin, even at high cellular concentrations, could still permit the polymerization/depolymerization–based recycling of the lamellipodial actin pool, which is required for proper lamellipodia dynamics and motility (4). In contrast, cells overexpressing activated WASp VCA domains mislocalize Arp2/3 complex and do not form lamellipodia (39), suggesting that overly high WASp protein or activation levels lead to constitutive Arp2/3 activation, which in turn could disrupt lamellipodia actin dynamics to a degree incompatible for promoting cell movement.

How cortactin regulates cell motility is becoming increasingly clearer and more complex. Recent work has shown a role for the cortactin amino-terminal acidic domain and actin-binding domains in enhancing and stabilizing protrusive lamellipodia in HT1080 fibrosarcoma cells (24), the same regions responsible for activation of Arp2/3 complex and stabilization of Arp2/3/F-actin networks (15). The same study also showed decreased G-actin incorporation in the lamellipodia of cells with suppressed cortactin expression, consistent with our finding of elevated actin incorporation (via the production of free barbed ends) in HNSCC cells with CTTN amplicafication and cortactin overexpression. Collectively, these results support a role for cortactin in promoting cell motility through regulating Arp2/3-dependent actin assembly and/or stabilizing the resulting actin network, resulting increased lamellipodia persistence. Overexpression of cortactin resulting from CTTN amplification would likely enhance these effects, as well as carboxyl-terminal SH3 domain–activated cell motility attributed to increased N-WASp activation (11).

Concurrent with its role in cell migration, evidence to date indicates that cortactin also regulates tumor cell invasion. In many carcinomas, cortactin is enriched within invadopodia, protrusive membranous structures in tumor cells that emanate from the ventral cell surface that regulate extracellular matrix degradation and protrusion (40). Recent studies have shown that cortactin is required for invadopodia formation and is essential for the recruitment of membrane type 1 matrix metalloproteinase, the main enzyme required for invadopodia gelatinase activity, to invadopodia (41). The ability of cortactin to regulate N-WASp and Arp2/3 activity is likely responsible in part for invadopodia formation because invadopodia require the activity of both of these molecules to form and extend (42). In addition to regulating invadopodia protrusion, cortactin may also play an important role in the endocytosis of proteolyzed matrix components as evidenced by the requirement of cortactin SH3 domain ligand dynamin 2 for invadopodia-induced matrix degradation (43).

Although our data indicates that down-regulation of cortactin expression with siRNA in tumor cells with CTTN amplification significantly decreases carcinoma motility and invasion, it is possible that other genes within the 11q13 amplicon are overexpressed that may also play a role in regulating cell migration and invasion. For example, the locus for the serine/threonine kinase Pak1 maps to the 11q13 region, and Pak1 overexpression has been documented in invasive HNSCC and breast carcinoma (44). Pak1 phosphorylates cortactin and is associated with lamellipodia formation in platelets (45). Pak1 also phosphorylates and activates Arp2/3 complex, inducing lamellipodia extension and cell motility (46). Pak1 may therefore serve to regulate carcinoma motility and invasion directly through Arp2/3 complex and indirectly through cortactin. Additionally, other genes yet to be characterized within the 11q13 amplicon may also contribute to cell motility. Whereas CTTN may be the predominant 11q13 gene involved in regulating carcinoma invasion, it is likely that 11q13 amplification exerts a pleiotropic effect on tumor cell motility and invasion through the overexpression of several gene products.

Consistent with previous work (44), gefitinib inhibits motility in all HNSCC lines analyzed in this study that overexpress EGFR. Moreover, our data indicate that cortactin tyrosine phosphorylation is down-regulated in HNSCC cells treated with gefitinib, indicating that the EGFR-Src signaling pathway is used for Src activation and subsequent cortactin tyrosine phosphorylation in HNSCC. Direct evaluation of cortactin tyrosine phosphorylation on HNSCC motility by expression of recombinant phosphorylation–deficient cortactin in cells lacking endogenous cortactin reduced HNSCC migration levels by 55%, clearly indicating that tyrosine phosphorylation of cortactin provides a substantial contribution to HNSCC motility. Whereas gefitinib largely inhibited motility (>90%) in response to EGF regardless of 11q13 amplification and cortactin expression status, the ability of gefitinib to inhibit cell migration in response to serum, a more physiologically relevant stimulus, was reduced in cells with 11q13 amplification and cortactin overexpression. The HNSCC cell lines used in this study show an inverse dose-response relationship between cortactin expression levels and cell motility in the presence of gefitinib. One explanation for this can be derived from a recent study that showed reduced ligand-induced EGFR internalization, ubiquitinlyation, and receptor degradation in HNSCC cells containing cortactin overexpression (47). The subsequent persistence of EGFR in internalized endosomes in HNSCC cells with cortactin overexpression may provide a sustained counteractive effect to gefitinib, especially in HNSCC cases with EGFR overexpression, because endosomal EGFR is still capable of activating Src (48) and therefore would promote cortactin phosphorylation. Alternatively, multiple pathways besides EGFR induce cortactin tyrosine phosphorylation (6), potentially allowing HNSCC cells to circumvent gefitinib-blocked EGFR and lead to cortactin tyrosine phosphorylation. The fact that gefitinib does not completely abolish cortactin tyrosine phosphorylation in HNSCC (Fig. 6C) supports these hypotheses.

The resistance of serum-stimulated HNSCC cells with cortactin overexpression to the antimotility effects of gefitinib is consistent with similar findings in HNSCC cells where cyclin D1–overexpressing lines were resistant to the antiproliferative effects of gefitinib (49). Because cyclin D1 and cortactin are often coamplified and overexpressed in HNSCC (18), it is possible that the two effects are linked, and that the overexpression of genes resulting from chromosome 11q13 amplification in HNSCC and other carcinomas may have additive effects that lead to gefitinib resistance. This might partially explain the modest success observed in phase II clinical trials with gefitinib in HSNCC (50). Such results may support validation of cortactin or other genes overexpressed from the 11q13 amplicon as potential prognostic or therapeutic biomarkers for HNSCC and other carcinomas as predictors of tumor cell growth, motility, and invasion.

Note: B.L. Rothschild and A.H. Shim contributed equally to this work.

Grant support: NIH grants DE014364 and DE014578 (S.A. Weed).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank R. Greer (Department of Diagnostic and Biological Sciences, University of Colorado at Denver and Health Sciences Center, Aurora, CO), D. Hines, M. Said, and K. Shroyer (Department of Pathology, University of Colorado at Denver and Health Sciences Center, Aurora, CO) for tissue archive and pathologic assistance; D. Rimm (Department of Pathology, Yale University, New Haven, CT) for the HNSCC tissue microarray; N. Popescu (Center for Cancer Research, National Cancer Institute, Bethesda, MD) for cortactin P1 genomic probes; A. Weaver (Department of Cancer Biology, Vanderbilt University, Nashville, TN) for the N-WASp GST-VCA domain; T. Carey (Department of Otolaryngology, University of Michigan, Ann Arbor, MI) for UMSCC2, UMSCC8, UMSCC10A, and UMSCC19; and A. Senderowicz (Oral and Pharyngeal Cancer Branch, National Institute of Dental and Craniofacial Research, Bethesda, MD) for HN4 cell lines; the University of Colorado Cancer Center Cytogenetics Facility for related assistance; A. Kapus and C. Di Ciano-Oliveira (Department of Surgery, Toronto General Hospital, Toronto, ON, Canada) for help with the actin incorporation assays; M. Auble (Department of Craniofacial Biology, University of Colorado at Denver and Health Sciences Center, Aurora, CO), T. Cooke (A.G. Heinze Inc., Lake Forest, CA), S. Huang (Department of Human Oncology, University of Wisconsin, Madison, WI), J. Altemus (Department of Neurobiology and Anatomy, West Virginia University, Morgantown, WV), and S. Frisch (Mary Babb Randolph Cancer Center, West Virginia University, Morgantown, WV) for technical assistance; and AstraZeneca for gefitinib.

1
Small JV, Stradal T, Vignal E, Rottner K. The lamellipodium: where motility begins.
Trends Cell Biol
2002
;
12
:
112
–20.
2
Condeelis JS, Wyckoff JB, Bailly M, et al. Lamellipodia in invasion.
Semin Cancer Biol
2001
;
11
:
119
–28.
3
Millard TH, Sharp SJ, Machesky LM. Signalling to actin assembly via the WASP (Wiskott-Aldrich syndrome protein)-family proteins and the Arp2/3 complex.
Biochem J
2004
;
380
:
1
–17.
4
Pollard TD, Borisy GG. Cellular motility driven by assembly and disassembly of actin filaments.
Cell
2003
;
112
:
453
–65.
5
Yamazaki D, Kurisu S, Takenawa T. Regulation of cancer cell motility through actin reorganization.
Cancer Sci
2005
;
96
:
379
–86.
6
Weed SA, Parsons JT. Cortactin: coupling membrane dynamics to cortical actin assembly.
Oncogene
2001
;
20
:
6418
–34.
7
Daly RJ. Cortactin signalling and dynamic actin networks.
Biochem J
2004
;
382
:
13
–25.
8
Schafer DA, Weed SA, Binns D, Karginov AV, Parsons JT, Cooper JA. Dynamin2 and cortactin regulate actin assembly and filament organization.
Curr Biol
2002
;
12
:
1852
–7.
9
Kinley AW, Weed SA, Weaver AM, et al. Cortactin interacts with WIP in regulating Arp2/3 activation and membrane protrusion.
Curr Biol
2003
;
13
:
384
–93.
10
Martinez-Quiles N, Ho HY, Kirschner MW, Ramesh N, Geha RS. Erk/Src phosphorylation of cortactin acts as a switch on-switch off mechanism that controls its ability to activate N-WASP.
Mol Cell Biol
2004
;
24
:
5269
–80.
11
Kowalski JR, Egile C, Gil S, Snapper SB, Li R, Thomas SM. Cortactin regulates cell migration through activation of N-WASP.
J Cell Sci
2005
;
118
:
79
–87.
12
Huang C, Liu J, Haudenschild CC, Zhan X. The role of tyrosine phosphorylation of cortactin in the locomotion of endothelial cells.
J Biol Chem
1998
;
273
:
25770
–6.
13
Head JA, Jiang D, Li M, et al. Cortactin tyrosine phosphorylation requires Rac1 activity and association with the cortical actin cytoskeleton.
Mol Biol Cell
2003
;
14
:
3216
–29.
14
Li Y, Tondravi M, Liu J, et al. Cortactin potentiates bone metastasis of breast cancer cells.
Cancer Res
2001
;
61
:
6906
–11.
15
Weaver AM, Karginov AV, Kinley AW, et al. Cortactin promotes and stabilizes Arp2/3-induced actin filament network formation.
Curr Biol
2001
;
11
:
370
–4.
16
Cao H, Orth JD, Chen J, Weller SG, Heuser JE, McNiven MA. Cortactin is a component of clathrin-coated pits and participates in receptor-mediated endocytosis.
Mol Cell Biol
2003
;
23
:
2162
–70.
17
Williams ME, Gaffey MJ, Weiss LM, Wilczynski SP, Schuuring E, Levine PA. Chromosome 11Q13 amplification in head and neck squamous cell carcinoma.
Arch Otolaryngol Head Neck Surg
1993
;
119
:
1238
–43.
18
Schuuring E. The involvement of the chromosome 11q13 region in human malignancies: cyclin D1 and EMS1 are two new candidate oncogenes—a review.
Gene
1995
;
159
:
83
–96.
19
Schuuring E, Verhoeven E, Mooi WJ, Michalides RJ. Identification and cloning of two overexpressed genes, U21B31/PRAD1 and EMS1, within the amplified chromosome 11q13 region in human carcinomas.
Oncogene
1992
;
7
:
355
–61.
20
Yuan BZ, Zhou X, Zimonjic DB, Durkin ME, Popescu NC. Amplification and overexpression of the EMS 1 oncogene, a possible prognostic marker, in human hepatocellular carcinoma.
J Mol Diagn
2003
;
5
:
48
–53.
21
Rodrigo JP, Garcia LA, Ramos S, Lazo PS, Suarez C. EMS1 gene amplification correlates with poor prognosis in squamous cell carcinomas of the head and neck.
Clin Cancer Res
2000
;
6
:
3177
–82.
22
van Rossum AG, Gibcus J, van der WJ, Schuuring E. Cortactin overexpression results in sustained epidermal growth factor receptor signaling by preventing ligand-induced receptor degradation in human carcinoma cells.
Breast Cancer Res
2005
;
7
:
235
–7.
23
Patel AS, Schechter GL, Wasilenko WJ, Somers KD. Overexpression of EMS1/cortactin in NIH3T3 fibroblasts causes increased cell motility and invasion in vitro.
Oncogene
1998
;
16
:
3227
–32.
24
Bryce NS, Clark ES, Leysath JL, Currie JD, Webb DJ, Weaver AM. Cortactin promotes cell motility by enhancing lamellipodial persistence.
Curr Biol
2005
;
15
:
1276
–85.
25
Tringler B, Gup CJ, Singh M, et al. Evaluation of p16INK4a and pRb expression in cervical squamous and glandular neoplasia.
Hum Pathol
2004
;
35
:
689
–96.
26
Wu H, Parsons JT. Cortactin, an 80/85-kilodalton pp60src substrate, is a filamentous actin-binding protein enriched in the cell cortex.
J Cell Biol
1993
;
120
:
1417
–26.
27
Sunpaweravong P, Sunpaweravong S, Puttawibul P, et al. Epidermal growth factor receptor and cyclin D1 are independently amplified and overexpressed in esophageal squamous cell carcinoma.
J Cancer Res Clin Oncol
2005
;
131
:
111
–9.
28
Bunn PA, Jr., Helfrich B, Soriano AF, et al. Expression of Her-2/neu in human lung cancer cell lines by immunohistochemistry and fluorescence in situ hybridization and its relationship to in vitro cytotoxicity by trastuzumab and chemotherapeutic agents.
Clin Cancer Res
2001
;
7
:
3239
–50.
29
Di Ciano C, Nie Z, Szaszi K, et al. Osmotic stress-induced remodeling of the cortical cytoskeleton.
Am J Physiol Cell Physiol
2002
;
283
:
C850
–65.
30
Huang SM, Li J, Harari PM. Molecular inhibition of angiogenesis and metastatic potential in human squamous cell carcinomas after epidermal growth factor receptor blockade.
Mol Cancer Ther
2002
;
1
:
507
–14.
31
Helfer B, Boswell BC, Finlay D, et al. Caspase-8 promotes cell motility and calpain activity under nonapoptotic conditions.
Cancer Res
2006
;
66
:
4273
–8.
32
Patel AM, Incognito LS, Schechter GL, Wasilenko WJ, Somers KD. Amplification and expression of EMS-1 (cortactin) in head and neck squamous cell carcinoma cell lines.
Oncogene
1996
;
12
:
31
–5.
33
van Damme H, Brok H, Schuuring-Scholtes E, Schuuring E. The redistribution of cortactin into cell-matrix contact sites in human carcinoma cells with 11q13 amplification is associated with both overexpression and post-translational modification.
J Biol Chem
1997
;
272
:
7374
–80.
34
Singh B, Gogineni SK, Sacks PG, et al. Molecular cytogenetic characterization of head and neck squamous cell carcinoma and refinement of 3q amplification.
Cancer Res
2001
;
61
:
4506
–13.
35
Weaver AM, Heuser JE, Karginov AV, Lee WL, Parsons JT, Cooper JA. Interaction of cortactin and N-WASp with Arp2/3 complex.
Curr Biol
2002
;
12
:
1270
–8.
36
Chan AY, Raft S, Bailly M, Wyckoff JB, Segall JE, Condeelis JS. EGF stimulates an increase in actin nucleation and filament number at the leading edge of the lamellipod in mammary adenocarcinoma cells.
J Cell Sci
1998
;
111
:
199
–211.
37
Howell GM, Grandis JR. Molecular mediators of metastasis in head and neck squamous cell carcinoma.
Head Neck
2005
;
27
:
710
–7.
38
Zhang Q, Thomas SM, Xi S, et al. SRC family kinases mediate epidermal growth factor receptor ligand cleavage, proliferation, and invasion of head and neck cancer cells.
Cancer Res
2004
;
64
:
6166
–73.
39
Machesky LM, Insall RH. Scar1 and the related Wiskott-Aldrich syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex.
Curr Biol
1998
;
8
:
1347
–56.
40
Buccione R, Orth JD, McNiven MA. Foot and mouth: podosomes, invadopodia and circular dorsal ruffles.
Nat Rev Mol Cell Biol
2004
;
5
:
647
–57.
41
Artym VV, Zhang Y, Seillier-Moiseiwitsch F, Yamada KM, Mueller SC. Dynamic interactions of cortactin and membrane type 1 matrix metalloproteinase at invadopodia: defining the stages of invadopodia formation and function.
Cancer Res
2006
;
66
:
3034
–43.
42
Yamaguchi H, Lorenz M, Kempiak S, et al. Molecular mechanisms of invadopodium formation: the role of the N-WASP-Arp2/3 complex pathway and cofilin.
J Cell Biol
2005
;
168
:
441
–52.
43
Baldassarre M, Pompeo A, Beznoussenko G, et al. Dynamin participates in focal extracellular matrix degradation by invasive cells.
Mol Biol Cell
2003
;
14
:
1074
–84.
44
Yang Z, Bagheri-Yarmand R, Wang RA, et al. The epidermal growth factor receptor tyrosine kinase inhibitor ZD1839 (Iressa) suppresses c-Src and Pak1 pathways and invasiveness of human cancer cells.
Clin Cancer Res
2004
;
10
:
658
–67.
45
Vidal C, Geny B, Melle J, Jandrot-Perrus M, Fontenay-Roupie M. Cdc42/Rac1-dependent activation of the p21-activated kinase (PAK) regulates human platelet lamellipodia spreading: implication of the cortical-actin binding protein cortactin.
Blood
2002
;
100
:
4462
–9.
46
Vadlamudi RK, Li F, Barnes CJ, Bagheri-Yarmand R, Kumar R. p41-Arc subunit of human Arp2/3 complex is a p21-activated kinase-1-interacting substrate.
EMBO Rep
2004
;
5
:
154
–60.
47
Timpson P, Lynch DK, Schramek D, Walker F, Daly RJ. Cortactin overexpression inhibits ligand-induced down-regulation of the epidermal growth factor receptor.
Cancer Res
2005
;
65
:
3273
–80.
48
Belsches AP, Haskell MD, Parsons SJ. Role of c-Src tyrosine kinase in EGF-induced mitogenesis.
Front Biosci
1997
;
2
:
d501
–18.
49
Kalish LH, Kwong RA, Cole IE, Gallagher RM, Sutherland RL, Musgrove EA. Deregulated cyclin D1 expression is associated with decreased efficacy of the selective epidermal growth factor receptor tyrosine kinase inhibitor gefitinib in head and neck squamous cell carcinoma cell lines.
Clin Cancer Res
2004
;
10
:
7764
–74.
50
Cohen EE, Rosen F, Stadler WM, et al. Phase II trial of ZD1839 in recurrent or metastatic squamous cell carcinoma of the head and neck.
J Clin Oncol
2003
;
21
:
1980
–7.