Metastasis is the final stage of tumor progression and is thought to be responsible for up to 90% of deaths associated with solid tumors. Caveolin-1 (CAV1) regulates multiple cancer-associated processes related to malignant tumor progression. In the present study, we tested the hypothesis that CAV1 modulates the metastatic ability of cells from the Ewing's sarcoma family of tumors (ESFT). First, we analyzed the expression of CAV1 by immunostaining a tissue microarray containing 43 paraffin-embedded ESFT tumors with known EWS translocations. Even though no evidence was found for a significant association between CAV1 expression and stage, size or tumor site, all metastatic samples (10 of 10) had significantly high CAV1 expression, suggesting that high CAV1 content could positively contribute to enhance ESFT metastasis. To determine the effect of CAV1 on the migratory and invasive capabilities of ESFT cells, we knocked down CAV1 expression in TC252 and A673 cells by stably transfecting a previously validated shRNA construct. In vitro, migration and invasion assays showed that for both cell lines, CAV1 knocked-down cells migrated and invaded significantly less (P ≤ 0.01) than control cells. Moreover, control A673 cells introduced into BALB/c nude mice by tail vein injection strongly colonized the lungs. In contrast, animals injected with CAV1 knocked-down cells showed either no incidence of metastasis or developed lung metastases after a significant delay (P < 0.0001). Finally, we show that the molecular mechanisms by which CAV1 carries out its key role in regulating ESFT metastasis involve matrix metalloproteinase production and activation as well as the control of the expression of SPARC, a known determinant of lung colonization. Mol Cancer Res; 8(11); 1489–500. ©2010 AACR.

This article is featured in Highlights of This Issue, p. 1437

Metastasis is the final stage in tumor progression and is thought to be responsible for up to 90% of deaths associated with solid tumors. This multifaceted process consists of a series of steps whereby cancer cells enter the circulation, disseminate to distal capillary beds, enter a parenchyma by extravasation, adapt to the new host microenvironment, and eventually grow into lethal tumor colonies in distal organs (1). Much has been learned about the processes that initiate and sustain general tumor growth; however, the mechanisms that enable metastasis remain largely unknown.

The Ewing's sarcoma family of tumors (ESFT) includes aggressive bone-associated malignancies that affect the pediatric population. ESFT are characterized by early metastases, and metastatic spread is commonly hematogenous. Nearly all patients with ESFT already have micrometastases at diagnosis, resulting in a >95% relapse rate when treated locally and a 40% relapse rate after systemic chemotherapy. Therefore, knowledge of key regulators of circulating or dormant metastatic tumor cells after traditional therapy is needed. Most ESFT harbor a reciprocal translocation, t(11;22)(q24;q12), which links a strong transcriptional activation domain from the EWS to the ETS DNA-binding domain of FLI-1 (2). The EWS/FLI-1 fusion is required for Ewing's sarcoma oncogenesis, as inhibition of its function results in the loss of transformation of ESFT cells (3-5). Thus, enhancing our current knowledge of EWS/FLI-1 function is critical to understanding ESFT development.

Caveolin-1 (CAV1) was previously identified as a metastasis-associated gene that is a transcriptional target of EWS/FLI-1 as well as an important determinant of ESFT malignant phenotype, tumorigenicity, and resistance to chemotherapy-induced apoptosis of ESFT cells (6, 7). CAV1 is a multifunctional scaffolding protein with multiple binding partners that associates with cell surface caveolae. CAV1 regulates multiple cancer-associated processes including cellular transformation, tumor growth, cell death and survival, multidrug resistance, angiogenesis, cell migration, and metastasis (8). Clinical studies have revealed that CAV1 upregulation is associated with poor prognosis and the occurrence of metastasis in several human cancers (9). Moreover, it was recently reported that CAV1 regulates cell migration and invasion in human breast cancer cells (10).

A better understanding of the complex biology of ESFT may lead to the successful development of biologically targeted therapies. As the regulatory pathways responsible for transformation, growth, and metastasis of ESFT become well defined, the efficacy of therapeutic intervention will improve. Therefore, because the possible involvement of CAV1 in the ability of ESFT to metastasize has not been established to date, we tested the hypothesis that CAV1 modulates the metastatic ability of ESFT cells. Our results show that CAV1 controls migration and invasion in ESFT cells in culture by mechanisms involving the production and activation of metalloproteinases as well as lung colonization in nude mice by regulating SPARC expression levels. These data conclusively show that CAV1 plays a key role in ESFT metastasis.

Tissue samples

Forty-three tumor samples were procured from the archives of the St. Jude Children's Research Hospital Pathology Department and referring institutions. A minimum of three representative cores from ESFT tumors were used to construct a tissue microarray (TMA) using a manual arrayer (Beecher Instruments). All tumor samples were fixed in formalin, embedded in paraffin, and had not undergone decalcification. Cores from various normal paraffin-embedded tissues were scattered in each of the TMA blocks as controls. Samples from two patients in the TMA were lost during immunohistochemical procedures; therefore, Table 1 only shows results from 41 patients.

FIGURE 1.

Example of negative and positive CAV1 staining of a human ESFT patient sample from the TMA (magnification, ×40; insert, ×400).

FIGURE 1.

Example of negative and positive CAV1 staining of a human ESFT patient sample from the TMA (magnification, ×40; insert, ×400).

Close modal

Immunohistochemistry

Expression of CAV1 in human Ewing's sarcoma and lung metastasis specimens was analyzed using immunohistochemical techniques done essentially as previously described (6, 7). CAV1 was detected with a 1:2,000 dilution of a rabbit polyclonal antibody (from BD Biosciences). Protein expression levels were coded as 0, 1 (low expression), or 2, and 3 (high expression) according to the level of staining (Fig. 1). Matrix metalloproteinase 9 (MMP9) was detected with a 1:100 dilution of a mouse monoclonal antibody (Abcam).

Cell culture and stable transfections

A673 and TC252 cell lines (gifts from Dr. Heinrik Kovar, Children's Cancer Research Institute, Kindrespitalgasse, Vienna, Austria) were cultured in RPMI 1640 (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen). All cell lines were incubated at 37°C in a humidified atmosphere of 5% CO2 in air. Exponentially growing cells within two sequential passages were used for all experiments. Cells were transfected using Fugene (Roche) following the protocols of the manufacturer. Transfected cells were selected with neomycin (0.4 mg/mL, ShCAV1; Invitrogen) and with puromycin (0.5 μg/mL, ShSPARC; Sigma; ref. 11) for 14 days, and antibiotic-resistant pools and individual colonies were isolated for further analysis and maintained in the presence of neomycin (0.2 mg/mL) and puromycin (0.1 μg/mL), respectively.

Reverse transcription-PCR

Total RNA (3 μg), extracted using the RNeasy Mini Kit (Qiagen), was used for cDNA synthesis with SuperScript II Reverse Transcriptase (Invitrogen). Amplifications of SPARC, MMP2, MMP9, TIMP2, and β-actin were carried out using specific primers (Supplementary Table) designed using the Oligo 6.0 software (National Bioscience). For each set of primers, the number of cycles was adjusted so that the reaction end points fell within the exponential phase of product amplification, thus providing a semiquantitative estimate of relative mRNA abundance. Reverse transcription-PCR determinations were carried out at least twice for each relevant transcript.

Western blot

ESFT cells were lysed with radioimmunoprecipitation assay buffer containing protease inhibitors (1 mmol/L phenylmethylsulfonyl fluoride, 10 mg/mL aprotinin, and 10 mg/mL leupeptin) and the lysates were centrifuged at 13,000 × g, at 4°C, for 30 minutes. The protein content of the supernatants was determined with the BCA assay system (Pierce). Lysate aliquots (50 μg) were resolved by 10% SDS-PAGE and transferred onto nitrocellulose membranes. After blocking with 5% skimmed milk in PBS containing 0.2% Tween 20 at room temperature for 1 hour, membranes were incubated overnight at 4°C with the appropriate primary antibody (CAV1 from BD, SPARC from Abcam, and FLI-1 from Santa Cruz Biotechnology). Blots were then incubated at room temperature for 1 hour with a horseradish peroxidase–conjugated secondary antibody (1:2,000) and the peroxidase activity was detected by enhanced chemiluminescence (Pierce) following the instructions of the manufacturer. Immunodetection of β-actin was used as a loading reference.

Migration assays

Wound-healing assay.

Cells cultured to 95% confluence were scratched with a 10-μL pipette tip. Wounds were photographed immediately and followed by imaging at later time points (the ensuing wound closure was photographed using an OLYMPUS Inverted Microscope). The nude area was measured and the percentage of wound closure was calculated.

Transwell migration assay.

Cells (2.0 × 105) in 100 μL of serum-free medium were added to the top chamber of 8-μm Costar polycarbonate transwells, while 500 μL of complete medium was added to the bottom chamber. After allowing migration for various times (37°C, 100% humidity, 5% CO2 in air), cells on the upper membrane surface were removed and migrant cells on the membrane underside were fixed using 70% ethanol, stained using 0.1% crystal violet solution (Invitrogen), and visualized and counted under the microscope. Data were presented as the average number of migrating cells in eight high-power fields (200×). Each experiment was done in triplicate, and then the data were averaged for statistical analysis.

Invasion assays

Matrigel-coated, sterile 8-μm polycarbonate filters (BD PharMingen) were rehydrated as described above. The lower chambers of the 24-well plate were filled with 0.5 mL of complete medium, then 0.5 mL of serum-free RPMI 1640 containing 2.0 × 105 tumor cells was added to the transwell chambers, and the plate was incubated at 37°C in 5% CO2 humidified atmosphere for 48 hours. Further staining and recording of results was done as mentioned above (see Migration assays). The specific MMP2/9 inhibitor [N-phenylsulfonyl-hydroxamic acid derivative (2R-[(4-biphenylsulfonyl)amino]-N-hydroxy-3-phenylpropinamide); 10 μmol/L] was purchased from Calbiochem.

Gelatin zymography

Metalloproteinase activity was analyzed by gelatin zymography. In brief, conditioned medium was concentrated to one-third of its original volume and electrophoresed on SDS polyacrylamide gels (10%) containing 120 μg/mL of gelatin. After electrophoresis, the gel was incubated for 60 minutes in 2.5% Triton X-100 and overnight at 37°C in 50 mmol/L of Tris (pH 8.0)/5 mmol/L CaCl2. Then, the gels were stained with 0.1% Coomassie brilliant blue and destained with 10% isopropanol in 10% acetic acid, and the gelatinolytic activity was identified as transparent bands in the Coomassie brilliant blue–stained background.

Immunofluorescence and confocal microscopy

Cells were cultured in sterile slides (Lab-Tek Chamber Slide System, NUNC) for 24 hours and fixed with 4% formaldehyde, washed thrice in Dulbecco's PBS (D-PBS), permeabilized in 0.1% Triton for 2 minutes, blocked for 1 hour in blocking buffer (10% goat serum in D-PBS), and incubated with primary antibodies overnight. Cells were then washed thrice in D-PBS for 5 minutes each followed by a 1-hour incubation with secondary antibodies (Alexa Fluor 488 goat anti-mouse and Alexa Fluor 594 goat anti-rabbit; Invitrogen). Then, cells were washed twice in D-PBS for 10 minutes and twice in distilled water for 10 minutes, and mounted in ProLong Gold antifade reagent with 4′,6-diamidino-2-phenylindole (Invitrogen). Photographs were taken with a Leica TCS SP5 spectral confocal microscope (argon, 405 diode and DPSS561 lasers) using a lambda blue 63× 1.35 numerical aperture oil objective. Images were analyzed with the MacBiophotonics ImageJ program.

Triton X-114 extraction

Extraction of membrane proteins was done essentially as described previously (12). An equal number of cells (∼1 × 106) from each cell line tested was cultured in 100-mm dishes for 3 days. Cells were then washed with TBS, incubated for 20 minutes on ice with 2.5% Triton X-114 detergent (Sigma) and scraped. Triton X-114 has the property of separating into aqueous and detergent (hydrophobic) phases at temperatures above 20°C, and amphiphilic integral membrane proteins remain in the detergent phase. Triton X-114 extracts were centrifuged (13,000 × g for 15 min at 4°C) to remove Triton X-114–insoluble material, incubated at 37°C for 5 minutes and spun at 3,000 × g for 2 minutes at room temperature to separate the aqueous and the hydrophobic phases. Aliquots (100 μL) of each hydrophobic phase were stored at −80°C.

Experimental metastasis assay

An in vivo experimental metastasis model was established by injection of 2 × 106 A673 cells suspended in 100 μL of saline into the tail vein of athymic nude mice (BALB/cnu/nu) from Harlan. Animal care and procedures were followed according to the Institutional Guidelines for the Care and Use of Laboratory Animals. Mice were fed under standard conditions with weight monitoring, sacrificed after manifestation of morbidity, and their lungs were removed for further analysis.

Statistical analysis

Data were analyzed for statistical significance using Student's t test and ANOVA. Survival curves were generated by Kaplan-Meier method and compared by using the log rank test. Unless otherwise stated, P ≤ 0.01 was regarded as significant.

First, we analyzed the expression of CAV1 by immunostaining a TMA containing 43 paraffin-embedded ESFT tumors with known EWS translocations (13). As expected, and in agreement with previous studies, ∼85% of the patients expressed CAV1. Clinicopathologic analyses revealed that CAV1 overexpression correlated significantly with metastasis and poor prognosis in several human cancers, including prostate, breast, and lung cancer (14). Although no evidence was found for a significant association between CAV1 expression and stage, size, or tumor site, all metastatic samples (10 of 10) had significantly high CAV1 expression (Table 1), suggesting that high CAV1 content could positively contribute to enhancing ESFT metastasis.

Table 1.

Summary of TMA data with regard to CAV1 expression (negative versus positive) and clinical features

StageTumor sizeTumor site
LocalizedMetastatic<8 cm>8 cmAxialExt.
Caveolin (n = 41) 
    Negative (0.1) 
    Positive (2-3) 25 10 12 23 24 11 
    Null/unsure 
StageTumor sizeTumor site
LocalizedMetastatic<8 cm>8 cmAxialExt.
Caveolin (n = 41) 
    Negative (0.1) 
    Positive (2-3) 25 10 12 23 24 11 
    Null/unsure 

To metastasize, tumor cells need to migrate from the primary site (15). To determine the effect of CAV1 on the migratory and invasive capabilities of ESFT cells, we knocked down CAV1 expression in TC252 and A673 cells (Fig. 2A) by stably transfecting a previously validated shRNA construct (6, 7). EWS/FLI-1 protein levels were unaffected by downregulation of CAV1 (Supplementary Fig. S1), further demonstrating the specificity of the shRNA construct. In vitro, wound-healing (Fig. 2B) and transwell migration assays (Fig. 2C) showed that in both cell lines, CAV1 knocked-down cells migrated significantly less (P ≤ 0.01) than mock- and vector-transfected cells. These results are consistent with previous reports that show the key role of CAV1 regulating the migratory capability of selected metastatic cell lines (10, 16).

FIGURE 2.

CAV1 downregulation reduces the migration of ESFT cells. A, immunoblot showing substantially reduced CAV1 levels in individual clones of A673 and TC252 cells stably expressing CAV1 shRNA compared with untransfected and vector-transfected cells. β-Actin was the loading control. B, wound-healing assay carried out with cells derived from the TC252 CAV1 silencing model. Phase contrast microscope images (×100) were taken at the indicated times after monolayer scratching (left); quantification of wound-healing assays with A673- and TC252-derived cells (right) with vertical bars representing mean values of gap closure from three independent experiments, each done in triplicate; bars, SD (*, P ≤ 0.01). C, photomicrographs (×40) showing differences in transwell migration among the cells indicated (left); quantification of transwell cell migration (right), with vertical bars representing mean percentages of the number of migrated cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.01).

FIGURE 2.

CAV1 downregulation reduces the migration of ESFT cells. A, immunoblot showing substantially reduced CAV1 levels in individual clones of A673 and TC252 cells stably expressing CAV1 shRNA compared with untransfected and vector-transfected cells. β-Actin was the loading control. B, wound-healing assay carried out with cells derived from the TC252 CAV1 silencing model. Phase contrast microscope images (×100) were taken at the indicated times after monolayer scratching (left); quantification of wound-healing assays with A673- and TC252-derived cells (right) with vertical bars representing mean values of gap closure from three independent experiments, each done in triplicate; bars, SD (*, P ≤ 0.01). C, photomicrographs (×40) showing differences in transwell migration among the cells indicated (left); quantification of transwell cell migration (right), with vertical bars representing mean percentages of the number of migrated cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.01).

Close modal

As the acquisition of an invasive phenotype by cancer cells is a critical step for tumor progression (15), we also examined the effects of CAV1 downregulation on the invasive ability of CAV1 knocked-down ESFT cells derived from both cell lines. Relative to control cells, as shown in Fig. 3A, downregulation of CAV1 correlated significantly (P ≤ 0.01) with a reduction in the invasiveness of A673 and TC252 cells. Early events in cancer cell invasion and metastasis involve the proteolytic degradation of extracellular matrix components. MMP are zinc-containing proteolytic enzymes that cleave extracellular matrix proteins, and play important roles in tumor cell migration and invasion. Along with other family members, MMP9 and MMP2 are able to degrade the main components of the basement membranes and play a key role in metastasis (17). To determine whether ESFT invasion was associated with MMP activation, conditioned media from both model cell cultures were analyzed by zymography with gelatin as the substrate (Fig. 3B). The proteolytic activities of MMP9 and MMP2 were present in conditioned medium from all cell lines. Interestingly, the band detected at 100 kDa (corresponding to the pro-MMP9 precursor form that became active during zymography), was found to be substantially decreased in all CAV1 knocked-down cell lines. Furthermore, a faster migrating gelatinolytic form of MMP2 was present only in the control cells, but was undetectable in CAV1 knocked-down cells, suggesting that CAV1 expression is required for MMP9 production and MMP2 activation. The involvement of MMPs on the invasive ability of Ewing's sarcoma cells was further confirmed by using a known MMP2/9 inhibitor [N-phenylsulfonyl-hydroxamic acid derivative (2R-[(4-biphenylsulfonyl)amino]-N-hydroxy-3-phenylpropinamide)], which induced a significant reduction in the invasiveness of A673 and TC252 cells (Supplementary Fig. S2). MMP9 production had been shown to be regulated at the transcriptional level in other tumor types (18). To determine whether the MMP9 decrease was associated with a downregulation of its mRNA levels, we compared the expression of MMP9, MMP2, and TIMP2 in cells derived from both CAV1 silencing models by reverse transcription-PCR. As shown in Fig. 3C, only MMP9 mRNA was downregulated as a consequence of CAV1 knockdown, suggesting that CAV1 regulates MMP9 expression differently from the way it affects other MMPs.

FIGURE 3.

CAV1 downregulation reduces the invasive ability of ESFT cells. A, invasion of cells through 8 μm Matrigel-coated polycarbonate pores was determined using Boyden chamber assays. Vertical bars represent mean percentages of invading cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.01). B, gelatin zymogram measuring activation of MMP9 and MMP2 related to CAV1 expression in the same cells in A. C, MMP9, MMP2, and TIMP2 mRNA levels were analyzed by semiquantitative reverse transcription-PCR (left). β-Actin was used as the loading reference. Histogram bars (right) represent mean densitometry values from two independent experiments; bars, SD (*, P ≤ 0.05).

FIGURE 3.

CAV1 downregulation reduces the invasive ability of ESFT cells. A, invasion of cells through 8 μm Matrigel-coated polycarbonate pores was determined using Boyden chamber assays. Vertical bars represent mean percentages of invading cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.01). B, gelatin zymogram measuring activation of MMP9 and MMP2 related to CAV1 expression in the same cells in A. C, MMP9, MMP2, and TIMP2 mRNA levels were analyzed by semiquantitative reverse transcription-PCR (left). β-Actin was used as the loading reference. Histogram bars (right) represent mean densitometry values from two independent experiments; bars, SD (*, P ≤ 0.05).

Close modal

Activation of the secreted pro-MMP2 is mediated by a cell surface complex that consists of a homodimer of membrane type 1-MMP (MT1-MMP) and a single molecule of tissue inhibitor of metalloproteinases 2 (TIMP2; a natural inhibitor of MMPs; refs. 19, 20). Moreover, CAV1 had been shown to be required for invadopodia formation and extracellular matrix degradation through regulation of MT1-MMP function in human breast cancer cells (21). Ewing's sarcoma cells had been shown to express MT1-MMP (22). Consequently, zymography from isolated membrane extracts (12) in the A673 model showed that membrane-bound MMP2 was mostly observed in control, but barely present in low CAV1 cells (Supplementary Fig. S3), suggestive of MT1-MMP activation in control cells but not in low CAV1 cells. Furthermore, confocal microscopy analysis showed that the decrease of MT1-MMP activity correlated with a diffuse redistribution of this protein in low CAV1 cells (Fig. 4). We were unable to observe wide colocalization between CAV1 and MT1-MMP. However, in control cells, these two proteins colocalized in some structures at the bottom of the cell, suggestive of invadopodia (Supplementary Fig. S4).

FIGURE 4.

Localization of CAV1 and MT1-MMP in cells derived from the A673 CAV1 silencing model. Exponentially growing, A673-derived cells were immunostained with anti-CAV1 antibody using an Alexa Fluor 488 goat anti-mouse secondary antibody (CAV1), anti–MT1-MMP antibody using an Alexa Fluor 594 goat anti-rabbit secondary antibody (MT1-MMP) or stained with 4′,6-diamidino-2-phenylindole to identify the nuclear compartment. Merging of the three images (Merge) showed the relocation of MT1-MMP from the cell membrane to a more diffusely staining compartment in CAV1 knocked-down cells (magnification, ×630).

FIGURE 4.

Localization of CAV1 and MT1-MMP in cells derived from the A673 CAV1 silencing model. Exponentially growing, A673-derived cells were immunostained with anti-CAV1 antibody using an Alexa Fluor 488 goat anti-mouse secondary antibody (CAV1), anti–MT1-MMP antibody using an Alexa Fluor 594 goat anti-rabbit secondary antibody (MT1-MMP) or stained with 4′,6-diamidino-2-phenylindole to identify the nuclear compartment. Merging of the three images (Merge) showed the relocation of MT1-MMP from the cell membrane to a more diffusely staining compartment in CAV1 knocked-down cells (magnification, ×630).

Close modal

Based on the in vitro results, we asked whether downregulation of CAV1 in ESFT cells might affect their metastatic potential in an experimental metastasis assay in vivo. The cell lines derived from the A673 CAV1 silencing model were injected into the tail vein of BALB/c nude mice. Animals were sacrificed after manifestation of morbidity. In agreement with a previous report using A673 cells as a model of metastasis (23), our parental and vector-transfected A673 cells strongly colonized the lungs. On the contrary, animals injected with CAV1 knocked-down cells either showed essentially no incidence of metastasis to any organ, or showed lung colonization only after a significant delay. Consequently, there were highly significant (P = 0.0001) differences in survival among the experimental groups of mice (Fig. 5A). Macroscopically, metastatic foci were quantified showing significant differences between control and low CAV1 cell–derived lungs (Fig. 5B and C). Histologic analysis of paraffin-embedded lungs clearly evidenced several lung metastases in mice injected with control cells in comparison with the lungs of animals injected with A/ShCAV1-5 and A/ShCAV1-9 cells (Fig. 5D). Although previous studies with A673 cells also reported metastasis to sites other than the lungs, only lung metastases developed in our experimental system. This may be explained by their use of a different immunodeficient mouse strain, which has recently been shown to allow ESFT cells to colonize other organs in addition to the lungs (24). Immunohistochemical analyses of paraffin-embedded metastatic lung tumors showed either diminished or no detectable CAV1 expression in A/ShCAV1-derived lung tumors from mice sacrificed throughout the experiment compared with the highly positive staining of parental or control-derived lung tumors from mice sacrificed during the 3rd week (Fig. 5E). In agreement with the in vitro results, MMP9 expression appeared mostly in controls. These results further showed the validity of the shRNA construct used in the present study and strongly support the hypothesis that CAV1 is necessary for ESFT cells to metastasize.

FIGURE 5.

CAV1 downregulation reduces metastasis in nude mice. A, survival of mice bearing ESFT cells using a mouse lung metastasis model. Kaplan-Meier survival curves of nude mice bearing tumors initiated by injection of mock-transfected A673 cells (A673 Mock), control vector transfected cells (A673/ShC), or CAV1 knocked-down (A/ShCAV1-5 and A/ShCAV1-9) cells (n = 10 for each cell line). B, quantification of lung metastases derived from control (A673 Mock and A673/ShC) and CAV1 knocked-down (ShCAV1) cells (A/ShCAV1-5 and A/ShCAV1-9) C, representative lungs excised from animals injected with A673-derived cells. Control mice were sacrificed during the 3rd wk whereas animals injected with cells expressing shCAV1s were sacrificed during the 8th week. D, H&E from corresponding paraffin-embedded lungs (×200). E, CAV1 and MMP9 staining of metastasis on lung sections from mice injected with control vector transfected A673 cells (A673/ShC) sacrificed on day 24 and ShCAV1 transfected cells (A/ShCAV1) sacrificed on day 60 (magnification, ×100).

FIGURE 5.

CAV1 downregulation reduces metastasis in nude mice. A, survival of mice bearing ESFT cells using a mouse lung metastasis model. Kaplan-Meier survival curves of nude mice bearing tumors initiated by injection of mock-transfected A673 cells (A673 Mock), control vector transfected cells (A673/ShC), or CAV1 knocked-down (A/ShCAV1-5 and A/ShCAV1-9) cells (n = 10 for each cell line). B, quantification of lung metastases derived from control (A673 Mock and A673/ShC) and CAV1 knocked-down (ShCAV1) cells (A/ShCAV1-5 and A/ShCAV1-9) C, representative lungs excised from animals injected with A673-derived cells. Control mice were sacrificed during the 3rd wk whereas animals injected with cells expressing shCAV1s were sacrificed during the 8th week. D, H&E from corresponding paraffin-embedded lungs (×200). E, CAV1 and MMP9 staining of metastasis on lung sections from mice injected with control vector transfected A673 cells (A673/ShC) sacrificed on day 24 and ShCAV1 transfected cells (A/ShCAV1) sacrificed on day 60 (magnification, ×100).

Close modal

The ability of ESFT cells to enter into the bloodstream (intravasation) and to exit from capillary beds into the parenchyma of an organ (extravasation) might be explained by the regulation that CAV1 exerts over MMPs. However, the ability of these cells to colonize the lungs may also be driven by the so-called metastasis virulence genes (25). Among a subset of these genes, we focused in SPARC (secreted protein acidic and rich in cysteine), a matricellular glycoprotein that mediates interactions between cells and their microenvironment as both CAV1 and SPARC have been found to be determinants of metastasis in basal-like breast carcinomas and melanomas (26, 27). In addition, neither the expression of SPARC in ESFT nor the mechanistic interaction between the SPARC and CAV1 proteins have been reported to date. Therefore, we tested whether SPARC mRNA and protein expression were affected by CAV1 knockdown. Results showed that SPARC mRNA and protein levels were indeed downregulated in CAV1 knockdown cells (Fig. 6A and B), suggesting that CAV1 regulates SPARC expression at the transcriptional level. SPARC has been found to induce migration and invasion in different types of cancer (28). Consequently, to explore whether SPARC collaborated with CAV1 inducing such processes, we knocked down SPARC in A673 cells by stably transfecting a validated (11) specific shRNA construct (Fig. 6C) without affecting CAV1 expression (Supplementary Fig. S5) and analyzed its effects on migration, invasion and lung colonization. Even though wound-healing (Fig. 6D), transwell migration (Fig. 6E), and invasion assays (Fig. 6F) showed that SPARC knocked-down cells migrated and invaded significantly less (P ≤ 0.05) than control cells, the effect of SPARC downregulation did not recapitulate the extent of the inhibitory effects induced by CAV1 knockdown. Conditioned medium from SPARC knocked-down cell cultures were analyzed by zymography (Fig. 6G). Similar to what happened in CAV1 knocked-down cells, MMP2 activation was undetectable in SPARC knocked-down cells. In contrast, SPARC downregulation did not provoke a decrease of MMP9, suggesting that SPARC knocked-down cells invaded better than CAV1 knocked-down cells in part because MMP9 was unaffected and thus remained active despite the low cellular levels of SPARC. Lastly, experimental metastasis assays showed that in contrast with control cells, SPARC knocked-down cells were unable to colonize the lungs (Fig. 6H), suggesting that SPARC is necessary for lung metastasis by Ewing's sarcoma cells.

FIGURE 6.

SPARC acts downstream of CAV1 to regulate ESFT cell invasiveness. A, SPARC mRNA levels were analyzed by semiquantitative reverse transcription-PCR in individual clones of A673 and TC252 cells stably expressing CAV1 shRNAs were compared with those in untransfected cells and in cells transfected with control vector DNA. B, SPARC protein levels in cells derived from the A673 and TC252 CAV1 silencing models. C, immunoblot showing substantially reduced SPARC levels in individual clones of A673 cells stably expressing SPARC shRNA compared with untransfected cells and cells transfected with control vector DNA. Wherever applicable, β-actin was used as the loading control. D, quantification of wound-healing assays, where vertical bars represent mean values of gap closure from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.05). E, quantification of transwell cell migration where vertical bars represent mean percentages of the number of migrated cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.05). F, invasion of cells through 8 μm Matrigel-coated polycarbonate pores was determined using Boyden chamber assays. Vertical bars represent mean percentages of invading cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.05). G, gelatin zymogram measuring activation of MMP9 and MMP2 related to SPARC expression in the same cells in D. H, representative lungs excised after 15 d from mice injected with either cells transfected with control vector and from ShSPARC-transfected cells (top). Quantification of lung weight (bottom) from mice injected with control or SPARC knocked-down cells (*, P ≤ 0.001).

FIGURE 6.

SPARC acts downstream of CAV1 to regulate ESFT cell invasiveness. A, SPARC mRNA levels were analyzed by semiquantitative reverse transcription-PCR in individual clones of A673 and TC252 cells stably expressing CAV1 shRNAs were compared with those in untransfected cells and in cells transfected with control vector DNA. B, SPARC protein levels in cells derived from the A673 and TC252 CAV1 silencing models. C, immunoblot showing substantially reduced SPARC levels in individual clones of A673 cells stably expressing SPARC shRNA compared with untransfected cells and cells transfected with control vector DNA. Wherever applicable, β-actin was used as the loading control. D, quantification of wound-healing assays, where vertical bars represent mean values of gap closure from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.05). E, quantification of transwell cell migration where vertical bars represent mean percentages of the number of migrated cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.05). F, invasion of cells through 8 μm Matrigel-coated polycarbonate pores was determined using Boyden chamber assays. Vertical bars represent mean percentages of invading cells from three independent experiments each done in triplicate; bars, SD (*, P ≤ 0.05). G, gelatin zymogram measuring activation of MMP9 and MMP2 related to SPARC expression in the same cells in D. H, representative lungs excised after 15 d from mice injected with either cells transfected with control vector and from ShSPARC-transfected cells (top). Quantification of lung weight (bottom) from mice injected with control or SPARC knocked-down cells (*, P ≤ 0.001).

Close modal

CAV1 regulates multiple cancer-associated cellular processes and its expression is a marker that predicts poor cancer patient prognosis (14). In our TMA, all metastatic samples had significantly high CAV1 expressions. This is in agreement with several publications reporting that certain genes expressed in primary tumors were also found in metastases from the same patient, suggesting that the metastatic potential of human tumors is encoded in the bulk of the primary tumors (29).

Migrating cells display sequential changes in morphology characterized by protrusion of lamellipodia and filopodia at the leading edge, attachment to the substratum, forward flow of cytosol, focal adhesion loosening, and retraction of the rear of the cell (30). CAV1 drives the formation of plasma membrane caveolae and anchors them to the actin cytoskeleton, modulates cell interaction with the extracellular matrix, pulls together and regulates signaling molecules, and transports cholesterol. Through these functions, CAV1 could play an important role in cell movement by controlling cell membrane composition and membrane surface expansion, polarization of signaling molecules, and/or cytoskeleton remodeling (31). Therefore, our laboratory is currently investigating whether CAV1 controls ESFT cell migration directly, acting on the migrating protrusions or indirectly behaving as functional platforms that regulate the organization, coordination, and activation status of signaling complexes (receptors, transducers, and effectors) involved in cell migration.

Our data represent the first description of CAV1 differently regulating MMP9 and MMP2 to promote invasion in ESFT. The regulation of MMP expression and activity occurs at many levels, including gene transcription, posttranscriptional processing, and proenzyme activation (32). Our results show that CAV1 regulates MMP9 by indirectly modulating its expression at the transcriptional level, while it regulates MMP2 by promoting the activation of its proenzyme form. Intriguingly, CAV1 knockdown induced the downregulation of MMP9 but not MMP2 or TIMP2. Multiple cytokines and growth factors are able to induce MMP expression, although the tissue specificity of the diverse family members is mainly achieved by the combination of different transcriptional control mechanisms. The integration of multiple signaling pathways facilitates the strict spatiotemporal control of MMP transcriptional activity (33). Therefore, in our model, CAV1 could be regulating specific signaling pathways that affect MMP9 expression independently of other MMPs.

For MMP2 activation, TIMP2 binds to the catalytic domain of one of the MT1-MMP molecules in the dimer and to the HPX domain of pro-MMP2, thereby facilitating cleavage and activation of pro-MMP2 by the second (TIMP2-free) MT1-MMP molecule of the dimer (20). Recently, it was shown that CAV1 is required for invadopodia formation and extracellular matrix degradation in human breast cancer cells, demonstrating that CAV1 is an essential regulator of MT1-MMP function (21). Accordingly, our results strongly suggest that CAV1 mediates MMP2 activation through regulation of MT1-MMP activity. However, and contrary to our results, CAV1 has been shown to inhibit invasion and MMP expression in pancreatic cancer (34). Also, in breast cancer cell lines, CAV1 has been shown to inhibit invasion, MMP activation, and metastasis in vivo (35). These opposite results could be explained because ESFT is not an epithelial cancer as is the case of pancreatic and mammary tumors, indicating that depending on the tissue type, CAV1 may exert opposite functions. In agreement with our results, in melanoma (another mesenchymal tumor), CAV1 promotes invasion by regulating MMP expression (36), underscoring the importance of the tissue of origin as a determinant of protein function.

As in many other tumors, metastatic disease is a poor prognostic factor for overall survival of patients with ESFT. Furthermore, this family of tumors has a great propensity to metastasize to the lungs (37). Accordingly, our experimental metastasis model showed preferential lung colonization of A673 cells in nude mice. Very recently, it has been suggested that the organ specificity of metastatic cells is determined by unique infiltrative and colonization functions which are dependent on specific gene sets (1). There is no data about genes that determine lung colonization in ESFT. Nevertheless, we focused on SPARC because it has been described as a determinant of lung colonization in metastatic breast tumors (11) and it has been shown to induce the production and activation of several MMPs (38).

Similar to MMP9, SPARC mRNA was affected by CAV1 knockdown, suggesting that in ESFT, expression of both genes might be regulated by CAV1. However, not all ESFT cells expressed SPARC (data not shown), this might be explained by the fact that SPARC expression in tumor cells is highly regulated by mechanisms other than transcription (38). Therefore, in some ESFT cells, SPARC expression could be regulated epigenetically, posttranscriptionally, or by miRNAs. In our model, SPARC knockdown resulted in a significant reduction in the migratory and invasive capabilities of A673 cells. However, the effects induced by SPARC downregulation never recapitulated the extent of the inhibitory actions provoked by CAV1 knockdown, suggesting that CAV1 mediates migration and invasion in part by indirectly regulating the expression of SPARC, but also by regulating other still unidentified proteins. These observations are in agreement with the situation reported in melanomas, in which both proteins have been shown to promote metastatic disease (27, 36). Zymography analysis showed that as in CAV1 knockdown cells, MMP2 activity was reduced as a consequence of SPARC knockdown, in contrast, MMP9 production was unaffected explaining why SPARC knockdown did not induce the same level of invasive inhibition than CAV1 knockdown. Similar to CAV1, the activity of SPARC is context- and cell type–dependent. In fact, SPARC has shown seemingly contradictory effects on tumor progression in both clinical correlative studies and in animal models (39). Therefore, the activity of SPARC related to invasion might be different in ESFT than in other tumor types. Interestingly, compared with control cells, none of SPARC knocked-down cells were able to colonize the lungs, suggesting that SPARC, as an adhesion protein, may indeed be necessary for ESFT cells to adhere to the lung epithelia. Alternatively, the hypothesis that SPARC plays an important role in the fusion of tumor cells with bone marrow–derived cells in the circulation has been proposed a new explanation for its role in metastasis (40). Nevertheless, in ESFT, whether SPARC promotes metastasis through one mechanism or the other requires further investigation.

Overall, our study shows that CAV1 plays a key role in regulating the ability of ESFT cells to migrate, invade the extracellular matrix, and colonize distant organs such as the lungs. This is the first study to evaluate the role of CAV1 on ESFT metastasis. Our results add relevance to the key roles that CAV1 plays in ESFT biology by controlling tumorigenicity (6), resistance to chemotherapy-induced apoptosis (7), and metastatic ability.

No potential conflicts of interest were disclosed.

We acknowledge the efforts of Blanca Luena for her assistance with the in vivo metastasis experiments, Catherine Billups in providing outstanding assistance with statistics, and Jemina Moreto, from Cancer Epigenetics and Biology Program (IDIBELL), for her support with confocal images acquisition and processing.

Grant Support: L. Lagares-Tena is funded by the Comissionat per a Universitats i Recerca (CUR) from Departament d'Innovació, Universitats i Empresa (DIUE) de la Generalitat de Catalunya i del Fons Social Europeu. This work was funded by the Fondo de Investigaciones Sanitarias-ISCIII (CP06/00151; PI080259).

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.

1
Nguyen
DX
,
Bos
PD
,
Massagué
J
. 
Metastasis: from dissemination to organ-specific colonization
.
Nat Rev Cancer
2009
;
9
:
274
84
.
2
Delattre
O
,
Zucman
J
,
Plougastel
B
, et al
. 
Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours
.
Nature
1992
;
359
:
162
5
.
3
Maksimenko
A
,
Malvy
A
. 
Oncogene-targeted antisense oligonucleotides for the treatment of Ewing sarcoma
.
Expert Opin Ther Targets
2005
;
9
:
825
30
.
4
Kovar
H
,
Ban
J
,
Pospisilova
S
. 
Potentials for RNAi in sarcoma research and therapy: Ewing's sarcoma as a model
.
Semin Cancer Biol
2003
;
13
:
275
81
.
5
Mateo-Lozano
S
,
Gokhale
PC
,
Soldatenkov
VA
,
Dritschilo
A
,
Tirado
OM
,
Notario
V
. 
Combined transcriptional and translational targeting of EWS/FLI-1 in Ewing's sarcoma
.
Clin Cancer Res
2006
;
12
:
6781
90
.
6
Tirado
OM
,
Mateo-Lozano
S
,
Villar
J
, et al
. 
Caveolin-1 (CAV1) is a target of EWS/FLI-1 and a key determinant of the oncogenic phenotype and tumorigenicity of Ewing's sarcoma cells
.
Cancer Res
2006
;
66
:
9937
47
.
7
Tirado
OM
,
MacCarthy
CM
,
Fatima
N
,
Villar
J
,
Mateo-Lozano
S
,
Notario
V
. 
Caveolin-1 promotes resistance to chemotherapy-induced apoptosis in Ewing's sarcoma cells by modulating PKCα phosphorylation
.
Int J Cancer
2010
;
126
:
426
36
.
8
Van Golen
KL
. 
Is caveolin-1 a viable therapeutic target to reduce cancer metastasis?
Expert Opin Ther Targets
2006
;
10
:
709
21
.
9
Burgermeister
E
,
Liscovitch
M
,
Rocken
C
,
Ebert
Schmid RM
. 
Caveats of caveolin-1 in cancer progression
.
Cancer Lett
2008
;
268
:
187
201
.
10
Joshi
B
,
Strugnell
SS
,
Goetz
JG
, et al
. 
Phosphorylated caveolin-1 regulates Rho/ROCK-dependent focal adhesion dynamics and tumor cell migration and invasion
.
Cancer Res
2008
;
68
:
8210
20
.
11
Minn
AJ
,
Gupta
GP
,
Siegel
PM
, et al
. 
Genes that mediate breast cancer metastasis to lung
.
Nature
2005
;
436
:
518
24
.
12
Azzam
HS
,
Arand
G
,
Lippman
ME
,
Thompson
EW
. 
Association of MMP-2 activation potential with metastatic progression in human breast cancer cell lines independent of MMP-2 production
.
J Natl Cancer Inst
1993
;
85
:
1758
64
.
13
Lai
R
,
Navid
F
,
Rodriguez-Galindo
C
, et al
. 
STAT3 is activated in a subset of the Ewing sarcoma family of tumours
.
J Pathol
2006
;
208
:
624
32
.
14
Goetz
JG
,
Lajoie
P
,
Wiseman
SM
,
Nabi
IR
. 
Caveolin-1 in tumor progression: the good, the bad and the ugly
.
Cancer Metastasis Rev
2008
;
27
:
715
35
.
15
Sahai
E
. 
Illuminating the metastatic process
.
Nat Rev Cancer
2007
;
7
:
737
49
.
16
Ravid
D
,
Chuderland
D
,
Landsman
L
,
Lavie
Y
,
Reich
R
,
Liscovitch
M
. 
Filamin A is a novel caveolin-1-dependent target in IGF-I-stimulated cancer cell migration
.
Exp Cell Res
2008
;
314
:
2762
73
.
17
Egeblad
M
,
Werb
Z
. 
New functions for the matrix metalloproteinases in cancer progression
.
Nat Rev Cancer
2002
;
2
:
161
74
.
18
St-Pierre
Y
,
Couillard
J
,
Van Themsche
C
. 
Regulation of MMP-9 gene expression for the development of novel molecular targets against cancer and inflammatory diseases
.
Expert Opin Ther Targets
2004
;
8
:
473
89
.
19
Itoh
Y
,
Takamura
A
,
Ito
N
, et al
. 
Homophilic complex formation of MT1-MMP facilitates proMMP-2 activation on the cell surface and promotes tumor cell invasion
.
EMBO J
2001
;
20
:
4782
93
.
20
Zucker
S
,
Hymowitz
M
,
Conner
C
,
DeClerck
Y
,
Cao
J
. 
TIMP-2 is released as an intact molecule following binding to MT1-MMP on the cell surface
.
Exp Cell Res
2004
;
293
:
164
74
.
21
Yamaguchi
H
,
Takeo
Y
,
Yoshida
S
,
Kouchi
Z
,
Nakamura
Y
,
Fukami
K
. 
Lipid rafts and caveolin-1 are required for invadopodia formation and extracellular matrix degradation by human breast cancer cells
.
Cancer Res
2009
;
69
:
8594
602
.
22
Yabe
H
,
Fukuma
M
,
Urano
F
, et al
.
Biochem Biophys Res Commun
2002
;
293
:
61
71
.
23
Richter
GH
,
Plehm
S
,
Fasan
A
, et al
. 
EZH2 is a mediator of EWS/FLI1 driven tumor growth and metastasis blocking endothelial and neuro-ectodermal differentiation
.
Proc Natl Acad Sci U S A
2009
;
106
:
5324
9
.
24
Nanni
P
,
Nicoletti
G
,
Landuzzi
L
, et al
. 
High metastatic efficiency of human sarcoma cells in Rag2/gc double knockout mice provides a powerful test for antimetastatic targeted therapy
.
Eur J Cancer
2010
;
46
:
659
68
.
25
Nguyen
DX
,
Massagué
J
. 
Genetic determinants of cancer metastasis
.
Nat Rev Genet
2007
;
8
:
341
52
.
26
Sarrió
D
,
Rodriguez-Pinilla
SM
,
Hardisson
D
,
Cano
A
,
Moreno-Bueno
G
,
Palacios
J
. 
Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype
.
Cancer Res
2008
;
68
:
989
97
.
27
Alonso
SR
,
Tracey
L
,
Ortiz
P
, et al
. 
A high-throughput study in melanoma identifies epithelial-mesenchymal transition as a major determinant of metastasis
.
Cancer Res
2007
;
67
:
3450
60
.
28
Tai
IT
,
Tang
MJ
. 
SPARC in cancer biology: its role in cancer progression and potential for therapy
.
Drug Resist Updat
2008
;
11
:
231
46
.
29
Ramaswamy
S
,
Ross
KN
,
Lander
ES
,
Golub
TR
. 
A molecular signature of metastasis in primary solid tumors
.
Nat Genet
2003
;
33
:
49
54
.
30
Lauffenburger
DA
,
Horwitz
AF
. 
Cell migration, a physically integrated molecular process
.
Cell
1996
;
84
:
359
69
.
31
Navarro
A
,
Anand-Apte
B
,
Parat
MO
. 
A role for caveolae in cell migration
.
FASEB J
2004
;
18
:
1801
11
.
32
Clark
IM
,
Swingler
TE
,
Sampieri
CL
,
Edwards
DR
. 
The regulation of matrix metalloproteinases and their inhibitors
.
Int J Biochem Cell Biol
2008
;
40
:
1362
78
.
33
Fanjul-Fernández
M
,
Folgueras
AR
,
Cabrera
S
,
López-Otín
C
. 
Matrix metalloproteinases: evolution, gene regulation and functional analysis in mouse models
.
Biochim Biophys Acta
2010
;
1803
:
3
19
.
34
Han
F
,
Zhu
HG
. 
Caveolin-1 regulating the invasion and expression of matrix metalloproteinase (MMPs) in pancreatic carcinoma cells
.
J Surg Res
2010
;
159
:
443
50
.
35
Williams
TM
,
Medina
F
,
Badano
I
, et al
. 
Caveolin-1 gene disruption promotes mammary tumorigenesis and dramatically enhances lung metastasis in vivo. Role of Cav-1 in cell invasiveness and matrix metalloproteinase (MMP-2/9) secretion
.
J Biol Chem
2004
;
279
:
51630
46
.
36
Felicetti
F
,
Parolini
I
,
Bottero
L
, et al
. 
Caveolin-1 tumor-promoting role in human melanoma
.
Int J Cancer
2009
;
125
:
1514
22
.
37
Balamuth
NJ
,
Womer
RB
. 
Ewing's sarcoma
.
Lancet Oncol
2010
;
11
:
184
92
.
38
Chlenski
A
,
Cohn
SL
. 
Modulation of matrix remodeling by SPARC in neoplastic progression
.
Semin Cell Dev Biol
2010
;
21
:
55
65
.
39
Arnold
SA
,
Brekken
RA
. 
SPARC: a matricellular regulator of tumorigenesis
.
J Cell Commun Signal
2009
;
3
:
255
73
.
40
Pawelek
JM
,
Chakraborty
AK
. 
Fusion of tumour cells with bone marrow-derived cells: a unifying explanation for metastasis
.
Nat Rev Cancer
2008
;
8
:
377
86
.