|Home | About | Journals | Submit | Contact Us | Français|
Caveolin-1 (CAV1) has been implicated in the regulation of several signaling pathways and in oncogenesis. Previously, we identified CAV1 as a key determinant of the oncogenic phenotype and tumorigenic activity of cells from tumors of the Ewing’s Sarcoma Family (ESFT). However, the possible CAV1 involvement in the chemotherapy resistance commonly presented by an ESFT subset has not been established to date. This report shows that CAV1 expression determines the sensitivity of ESFT cells to clinically relevant chemotherapeutic agents. Analyses of endogenous CAV1 levels in several ESFT cells and ectopic CAV1 expression into ESFT cells expressing low endogenous CAV1 showed that the higher the CAV1 levels, the greater their resistance to drug treatment. Moreover, results from antisense- and shRNA-mediated gene expression knockdown and protein re-expression experiments demonstrated that CAV1 increases the resistance of ESFT cells to doxorubicin (Dox)- and cisplatin (Cp)-induced apoptosis by a mechanism involving the activating phosphorylation of PKCα. CAV1 knockdown in ESFT cells led to decreased phospho(Thr638)-PKCα levels and a concomitant sensitization to apoptosis, which were reversed by CAV1 re-expression. These results were recapitulated by PKCα knockdown and re-expression in ESFT cells in which CAV1 was previously knocked down, thus demonstrating that phospho(Thr638)-PKCα acts downstream of CAV1 to determine the sensitivity of ESFT cells to chemotherapeutic drugs. These data, along with the finding that CAV1 and phospho(Thr638)-PKCα are co-expressed in ~45% of ESFT specimens tested, imply that targeting CAV1 and/or PKCα may allow the development of new molecular therapeutic strategies to improve the treatment outcome for ESFT patients.
Tumors of the Ewing’s sarcoma family (ESFT), such as Ewing’s sarcoma (EWS) and primitive neuroectodermal tumors (PNET), are highly aggressive malignancies predominantly affecting children and young adults, with EWS being the second most common bone tumor in such patient populations1. Most ESFT (over 90% in EWS cases) express chimeric proteins encoded by hybrid genes generated by translocations involving the EWS gene and several genes belonging to the ETS family of transcription factors, most commonly the FLI1 gene. These hybrid EWS/FLI1 genes are heterogeneous with regard to the location of the translocation junction, resulting in different breakpoints2 but, regardless of the fusion type, all EWS/FLI1 proteins act as aberrant transcription factors that are responsible for the malignant properties of ESFT.3
Multimodal therapeutic regimens combining more intensive systemic treatment with chemotherapy, enhanced surgical procedures, and advanced radiotherapy have improved the overall survival of ESFT patients.4 Nevertheless, survival rates for most patients presenting metastasis at diagnosis are only about 20%.2 ESFT metastatic cells are highly resistant to microenviromental5 and chemotherapy-induced death.6 Therefore, identification of metastasis markers that may also play a role in determining the sensitivity or resistance of ESFT cells to chemotherapeutic treatment is urgently needed to improve ESFT treatment and overall patient survival. Using gene expression arrays, we previously identified caveolin-1 (CAV1) as a metastasis-associated gene that is a transcriptional target of EWS/FLI1 as well as an important determinant of ESFT malignant phenotype and the tumorigenic activity of ESFT cells.7 However, the possible involvement of CAV1 in the response of ESFT to chemotherapy has not been established to date.
CAV1 is the main component of specialized invaginated plasma membrane microdomains, termed caveolae, which are found in most mammalian cells including adipocytes, endothelial cells, fibroblasts, myoepithelial cells, and some epithelial cells.8 Owing to its subcellular localization and rather ubiquitous distribution, CAV1 has been reported to play major roles in lipid transport, membrane trafficking, gene regulation, and signal transduction.9 However, the role that CAV1 may play in the response to chemotherapy remains unclear, with several studies reporting that CAV1 expression was involved in resistance to chemotherapy-induced apoptosis,10,11 while others suggested that elevated CAV1 expression correlated with drug sensitivity.12 Nevertheless, overall, the evidence provided by existing literature tends to favor an association of CAV1 expression with drug resistance rather than sensitization to treatment. In this regard, CAV1 over-expression has been associated with poor therapeutic response in patients with meningioma13 and kidney,14 lung15 and prostate cancers.16 Moreover, CAV1 has been found differentially over-expressed in multidrug resistant colon cancer cells, adriamycin-resistant breast cancer cells, and taxol- and gemcitabine-resistant lung cancer cells.17–19
Despite being considered as generally chemosensitive, first line therapeutic failure in ESFT patients with primary refractory, recurrent or metastatic disease is associated with very poor prognosis.20 Consequently, identification of chemotherapy-resistance markers in ESFT is essential to develop improved therapies to overcome the drug-resistant phenotype. The present study identifies a role for CAV1 expression in rendering ESFT cells resistant to chemotherapy-induced apoptosis through a mechanism that involves the α isoform of protein kinase C (PKCα) as a key downstream effector of CAV1. Results support the notion that targeting CAV1 and/or PKCα may be a valuable alternative for developing new therapeutic strategies to improve the treatment outcome for ESFT patients.
EWS (TC-71, SK-ES-1 and A4573) and PNET (TC-32 and SK-N-MC) cell lines were cultured as previously described.21 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. Cell survival was determined by the trypan blue exclusion assay as described.7,21,22 Doxorubicin (Dox), cisplatin (Cp) and vincristine were purchased from Sigma-Aldrich (St. Louis, MO). The Gö6976 PKCα inhibitor and the PKCβ inhibitor were purchased from Calbiochem/EMD Biosciences, Inc. (San Diego, CA). Antibodies specific to cleaved caspase-3, PKCα, phospho(Thr638) PKCα, and XIAP were from Cell Signaling Technology (Beverly, MA), and anti-CAV1 was from BD Biosciences (San Jose, CA). Anti-β-actin (Abcam, Cambridge, England) was used for loading normalization. All other general reagents were from Sigma-Aldrich.
ESFT cells were lysed with RIPA buffer containing protease inhibitors (1 mM PMSF, 10 mg/ml aprotinin and 10 mg/ml leupeptin) and the lysates were centrifuged at 13,000 g, at 4°C, for 30 min. The protein contents of the supernatants were determined with the BCA assay system (Pierce, Rockford, IL). Lysate aliquots (50 µg) were resolved by 10% SDS-PAGE and transferred onto nitrocellulose membranes. After blocking with 5% skim milk in PBS containing 0.2% Tween-20, at room temperature, for 1 h, membranes were incubated overnight at 4°C with the appropriate primary antibody. Blots were then incubated at room temperature for 1 h with a HRP-conjugated secondary antibody (1/2000) and the peroxidase activity was detected by chemiluminescence (ECL, Pierce) following manufacturer's instructions. Immunodetection of β-actin was used as loading reference.
Expression of CAV1 and phospho(Thr638) PKCα in human Ewing’s sarcoma specimens was analyzed using immunohistochemical techniques performed essentially as previously described.7,22 CAV1 was detected with a 1:2000 dilution of a rabbit polyclonal antibody (#610059, from BD Biosciences), and phospho(Thr638) PKCα was analyzed using a 1:250 dilution of a rabbit monoclonal antibody (E195 clone, from Epitomics Inc., Burlingame, CA) raised against a synthetic phosphopeptide corresponding to amino acid residues surrounding the Thr638 in the primary sequence of human PKCα.
Antisense (pBK-ASCAV1) as well as CAV1 (pShCav1-1) and control (pShCont) shRNA vectors were generated as described.7 A4573 cells were transfected using Lipofectamine (Invitrogen, Carlsbad, CA), selected with neomycin (0.8 mg/ml) for 14 days, and antibiotic-resistant pools and individual colonies were isolated for further analysis, being maintained in the presence of neomycin (0.2 mg/ml). The pBK-CMV-CAV1 construct, including the entire CAV1 open reading frame (ORF), but lacking the 3’-untranslated region, was used to re-express CAV1 in A4573 cells expressing a shRNA targeted to the CAV1 3’- sequences (pShCav1-1+CAV1). The pBK-CMV-CAV1 vector and pCHC6-hygro, a plasmid providing hygromycin resistance, were co-transfected (50:1 molar ratio) into A4574/pShCav1-1 cells using Lipofectamine, selected for 2 weeks with hygromycin (100 µg/ml) in medium containing the neomycin maintenance dose (0.2 mg/ml), and resistant cellular pools and individual colonies were isolated for further analysis, being maintained in the presence of neomycin (0.2 mg/ml) and hygromycin (50 µg/ml).
Cells stably expressing shRNA against PKCα were generated essentially as described,23 using PKCα-specific HuSH shRNA constructs and a sequence-scrambled control (from Origene, Rockville, MD). A4573 cells were transfected with the various shRNAs using Lipofectamine (Invitrogen), selected in media with 100 ng/ml puromycin for two weeks, and resistant cellular pools and individual colonies were isolated for further analysis, being maintained in the presence of puromycin (50 ng/ml). A pCMV6-XL4-PKCα construct, including the entire PKCα ORF (Origene), but lacking the 3’-untranslated region, was used to re-express PKCα in either A4573 cells in which CAV1 had been previously knocked down (A4573/pShCav1-1) or in A4573 cells expressing a shRNA targeted to the 3’-sequences of PKCα. The pCMV6-XL4-PKCα construct was transferred into A4573/pShCav1-1 cells by co-transfection with the pCHC6-hygro, as described above. After transfection using Lipofectamine, cells were selected for two weeks with puromycin (100 ng/ml), neomycin (0.2 mg/ml) and hygromycin (50 µg/ml) in the case of A4573/pShCAV1-1, or with puromycin (100 ng/ml) and neomycin (0.8 mg/ml) in the case of A4573/pShPKCα, and resistant cellular pools and individual colonies were isolated for further analysis, being cultured in the presence of maintenance concentrations of the appropriate antibiotics.
Unless otherwise indicated, western blot analyses and cell death determinations were repeated at least three times. For statistical analysis, ANOVA or student-t tests were used to assess the significance of differences between groups or individual variables, respectively. P ≤ 0.01 was regarded as significant.
On the basis of reports showing a correlation between CAV1 expression and drug resistance in human non-small cell lung carcinoma19 and in oral squamous cell carcinoma24, we explored the possibility that CAV1 expression correlated with resistance to chemotherapy-induced apoptosis in ESFT by testing the response of EWS and PNET cell lines expressing various CAV1 levels to drug treatment. Cells were treated for up to 72 h with several doses of drugs currently used in ESFT treatment such as Dox, vincristine and Cp.4,25,26 Determinations of the IC50 values for the various cell lines tested showed that ESFT cells with higher CAV1 expression levels (A4573 and TC-32) were clearly more resistant to Dox-and Cp-induced cell death than cells expressing lower CAV1 levels (Fig. 1a). The specificity of the CAV1 effect on the response to these drugs was supported by the fact that the response of the same cell lines to vincristine treatment was independent of their CAV1 content (data not shown). To corroborate these data, we transfected low CAV1-expressing TC-71 cells with either control pBK-CMV vector DNA or with a pBK-CMV-derived construct expressing the full-length CAV1 ORF, and selected stably transfected cellular pools and individual clones with significantly increased CAV1 expression levels (Fig. 1b). Results (Fig. 1c) from treatments of the CAV1 over-expressing TC-71 cells with Dox or Cp for up to 72 h showed that, relative to untransfected and empty vector-transfected cells, the increased expression of CAV1 significantly enhanced the resistance of TC-71 cells to both Dox (~3.8-fold) and Cp (~3.2-fold). These results agreed with the increased IC50 values determined for TC-71 cells over-expressing CAV1 for Dox (49.7±2.3 ng/ml) and Cp (9.6±0.8 µg/ml) relative to those of untransfected TC-71 cells (Fig. 1a), and provided strong support for the notion that CAV1 may be a key determinant of the drug response in ESFT cells.
The reverse experiment was also performed. High CAV1-expressing A4573 cells were transfected with either control vectors (pBK-CMV and pShCont), a cDNA construct expressing antisense (AS) CAV1 (AS/CAV1),7or a shRNA vector targeting CAV1 (pShCav1-1),7 and then treated with Cp or Dox. Both the AS and the shRNA approaches efficiently down-regulated CAV1 in A4573 cells (Fig. 2a). Regardless of the knockdown method used, cells with over 60% reduced CAV1 expression showed greater sensitivity to both drugs (by about 2.3-fold) than cells from vector-transfected control cultures (Fig. 2b). The fact that cells in which CAV1 was down-regulated were slightly more prone to cell death in the absence of drug treatment might be due to the enrichment of particular cellular sub-populations somewhat more unstable than control cells during the selection process to generate the AS- and shCav-expressing cells. Overall, these results agreed with the decreased IC50 values of AS/CAV1-expressing cells (18±3.3 ng/ml for Dox; 2.3±0.9 µg/ml for Cp) and of cells expressing pShCav1-1 (33±2.5 ng/ml for Dox; 3.5±1.1 µg/ml for Cp) relative to the IC50 values of cells transfected with the corresponding controls pBK-CMV (59±3.8 ng/ml for Dox; 8.03±2.6 µg/ml for Cp) and the pSh-Control (65±1.3 ng/ml for Dox; 8.9±1.8 µg/ml for Cp). More importantly, CAV1 re-expression into pShCav1-1 cells (pShCav1-1+CAV1) resulted in a significant recovery of the resistance to both Dox and Cp (Fig. 2b), which was also reflected in an increase in their IC50 values (71±3.4 ng/ml for Dox; 10.2±2.7 µg/ml for Cp) relative to those indicated above for pSh-Cav1-1-expressing cells. Western immunoblot analyses showed that both Cp and Dox did trigger an enhanced apoptotic response in cells expressing lower CAV1 levels, as evidenced by the parallel increase in the cleavage of both caspase-3 and XIAP (Fig. 2c), and that the apoptotic process was partially reverted by CAV1 reintroduction into the CAV1 knocked-down cells (Fig. 2c). These results provided strong evidence indicating that CAV1 is indeed an important determinant of the resistance of ESFT cells to the induction of apoptosis by Dox and Cp.
CAV1 is frequently up-regulated in multiple drug resistant (MDR) cancer cells, and it has been proposed that the association between CAV1 and resistance to chemotherapy-induced apoptosis is related to the MDR phenotype.27 Consequently, we first explored the possibility of a correlation between the expression levels of CAV1 and P-glycoprotein (Pgp) in the ESFT cell lines tested. The fact that no significant differences in Pgp expression were observed (data not shown) suggested that Pgp was not a major player in the drug resistance mechanism promoted in ESFT cells by CAV1. On the basis of previous reports showing that (a) PKCα is a key regulator of the MDR phenotype,28–30 (b) PKCα associates with, and phosphorylates CAV1 in non-ESFT cells,31,32 and (c) PKCα and CAV1 may be co-regulated,33 we hypothesized that the role of CAV1 in chemotherapy resistance in ESFT cells may involve the regulation of PKCα. To explore such possibility, we first determined the levels of PKCα and phosphorylated, active PKCα in ESFT cells. Given that the functional activation of PKCα, similar to other PKC isoforms, is regulated by serine/threonine trans- and auto-phosphorylation reactions,34 we performed immunoblot analyses of extracts from ESFT cells to evaluate the phosphorylation status of the threonine residue at position 638 (Thr638), which is an autophosphorylation site indicative of functional PKCα activation.35,36 Results (Fig. 3a) showed that, although there were no remarkable differences in total PKCα protein, the ESFT cell lines tested expressed different levels of Thr638-phosphorylated PKCα. When the relative levels of PKCα, Thr638-phosphorylated PKCα and CAV1 in the cell lines tested were graphically represented, it became obvious that the levels of CAV1 did not correlate with those of PKCα, but correlate well with the levels of Thr638-phosphorylated PKCα (Fig. 3b). To corroborate these data, we compared the expression levels of both PKCα and Thr638-phosphorylated PKCα in low-CAV1 expressing TC-71 cells with the corresponding levels in CAV1 over-expressing TC-71 cells. Results (Fig. 3c) confirmed that increased CAV1 expression did not appreciably change the levels of PKCα, but substantially increased the levels of Thr638-phosphorylated PKCα. These results supported the notion that CAV1 is involved in regulating the phosphorylation status and activity of PKCα in ESFT cells, and suggested that PKCα phosphorylation, rather than the total PKCα content, may mediate the role of CAV1 in the response of ESFT cells to chemotherapeutic agents.
We then tested whether CAV1 down-regulation in high CAV1-expressing A4573 cells had any effect on the level of PKCα protein expression and/or its phosphorylation status and activity, and whether disruption of PKCα signaling activity had any repercussions on chemotherapy sensitization. To test this possibility, we initially investigated whether the inhibition of PKCα activity could mediate the increased drug sensitivity of CAV1-knockdown cells. To that end, A4573 cells were treated with a PKCα-selective inhibitor (Gö6976), known to sensitize other tumor cells to different chemotherapeutic agents,37 prior to Cp and Dox exposure. In vector-transfected (pBK-CMV) cells and in CAV1 knockdown cells in which CAV1 was reintroduced (pShCav1-1+CAV1), PKCα inhibition recapitulated the sensitization to Cp and Dox promoted by CAV1 down-regulation in A4573/pShCav1-1 and in A4573/AS10/CAV-1 cells (Fig. 2b), with regard to both the extent of cell death (Fig. 4a) and the expression of cell death-related markers (Fig. 4b). Although Gö6976 has a greater selectivity for PKCα, it may also inhibit PKCβ. Therefore, we performed experiments identical to those carried out with the Gö6976 inhibitor, but including a potent PKCβ inhibitor with 60-fold greater selectivity for PKCβ isozymes (IC50 ≈ 5–20 nM) than for other PKC isoforms (IC50 in the µM range). Interestingly, this staurosporin-related, aniline-monoindolylmaleimide PKCβ inhibitor38 did not alter the apoptotic response of A4573 cells to Cp or to Dox (Fig. 4c), thus strongly suggesting a high degree of specificity for the involvement of PKCα in the drug response mechanism as an important mediator of the pro-survival function performed by CAV1 in ESFT cells.
Because the data obtained by chemical inhibition of PKCα strongly suggested that CAV1 may regulate PKCα activation, and taking into consideration that Gö6976 could also provoke a variety of nonspecific effects,39 it became important to test whether CAV1 expression maintained elevated levels of phosphorylated, active PKCα in ESFT cells. Accordingly, we performed immunoblot analyses of extracts from A4573 cells engineered to express different levels of CAV1 to evaluate the phosphorylation status of PKCα at the Thr638 amino acid position. Results from western immunoblot analyses (Fig. 5a) showed that A4573 cells transfected with the pShCont or pBK-CMV control vectors expressed high levels of total and Thr638-phosphorylated PKCα, and that CAV1 down-regulation substantially decreased the levels of Thr638-phosphorylated PKCα in A4573/pShCav1-1 and A4573/AS10/CAV-1 cells, while having relatively minor effect on the total PKCα protein levels. Moreover, re-introduction of CAV1, which restored the resistance of CAV1-knockdown cells to Dox and Cp (Fig. 2b), also restored the levels of Thr638-phosphorylated PKCα (Fig. 5a), strongly indicating that CAV1 levels contribute to regulate the activating phosphorylation of PKCα in ESFT cells.
To conclusively establish the involvement of PKCα in CAV1-induced drug resistance in ESFT cells, PKCα was efficiently knockdown by shRNA-mediated transfection into A4573 cells, resulting in a marked decrease in the level of Thr638-phosphorylated PKCα (Fig. 5b). PKCα-knockdown cells (A4573/pShPKCα) were exposed to Cp (5 µg/ml) for up to 72 h. Results (Fig. 5c) showed that, relative to cells transfected with control plasmid DNA (A4573/pShCont), PKCα down-regulation in A4573 cells provoked a significant increase in Cp-induced cell death and, most importantly, that the Cp-sensitization effect was efficiently reversed, concomitantly with a remarkable increase of the levels of Thr638-phosphorylated PKCα above those in control cells (Fig. 5b), by forced re-expression of PKCα (A4573/pShPKCα+PKCα) in cells in which it had been previously knocked down (A4573/pShPKCα). Furthermore, re-expression of PKCα in A4573 cells in which CAV1 had been previously knocked down (A4573/pShCav1-1+PKCα) also caused a significant reversion of the Cp-sensitization effect caused by CAV1 down-regulation, and this Cp-sensitization effect was paralleled by an increase in the levels of Thr638-phosphorylated PKCα. These results demonstrated that Thr638-phosphorylated PKCα acts downstream of CAV1 in ESFT cells to mediate their resistance to chemotherapeutic drugs.
In order to explore whether the [CAV1–phospho(Thr638)PKCα] functional interaction observed to play a role in determining the drug response of ESFT cells in culture may be also involved in the pathobiology and drug response of tumors of the Ewing’s sarcoma family, we evaluated the status of the expression of CAV1 and phospho(Thr638) PKCα in human tumors by performing immunohistochemical analyses of a number of ESFT specimens. Results showed a direct correlation between the levels of CAV1 and those of phospho(Thr638) PKCα in about 45% (8 of 18) of the specimens analyzed (Fig. 6). These results strongly suggested a straight link of the functional interaction between CAV1 and phospho(Thr638) PKCα observed in cultured ESFT cells with the clinical situation of a substantial proportion of human tumor samples, and lent additional support and greater significance to the in vitro findings.
The study described here represents the first report on the involvement of CAV1 in determining the chemoresistant response of ESFT cells to DNA damaging agents such as Dox and Cp.40,41 Using shRNA-mediated gene expression knockdown, ectopic protein expression and re-expression studies, our results conclusively show that the drug sensitivity of ESFT cells can be substantially modified by modulating the cellular content of CAV1. A ~2.9-fold increase in CAV1 expression in TC-71 cells (Fig. 1b) enhanced their Cp resistance ~3.2-fold and their Dox resistance ~3.8-fold (Fig. 1c). Conversely, CAV1 down-regulation by over 60% with AS or shRNA in A4573 cells (Fig. 2a) increased their sensitivity to Cp and Dox ~2.2-fold and ~2.3- fold (Fig. 2b), respectively. Our data also indicate that increased cellular levels of CAV1 efficiently antagonize Dox- and Cp-triggered apoptotic processes, ultimately rendering ESFT cells resistant to apoptotic cell death, thus favoring tumor cell survival and further progression of their malignant phenotype. Results reported here identify a functional CAV1–PKCα interaction that, by enhancing the activating phosphorylation of PKCα at Thr638, plays a fundamental role in the acquisition of chemoresistance by ESFT cells during tumor progression.
It must be pointed out that, although Dox and Cp are DNA-damaging agents, it is possible that CAV1 and PKCα may participate in promoting apoptosis resistance via alternative mechanisms of action. In fact, it has been reported that the induction of apoptosis in ESFT cells by Dox is not necessarily mediated by its DNA-damaging activity,42 but may involve Fas/FasL-mediated mechanisms,43 or insulin growth factor-I (IGF-I)-triggered pathways involving phosphatidylinositol 3-kinase (PI3K) and/or AKT.44 However, because these studies42,43 were performed using the Dox-sensitive, SK-N-MC PNET cell line, which our study identifies as expressing low CAV1 levels (Fig. 1a), it is possible that any role of CAV1 or PKCα in the process may have gone undetected. Interestingly, the fact that experiments on the interaction between IGF-I signaling and the response to Dox44 were carried out using TC-71 (low CAV1-expressing cells) and TC-32 (high CAV1-expressing cells), along with the observation that the IGF-I effect on the Dox-response was similar in both cell lines, indicate that the survival function supported by IGF-I in ESFT cells is CAV1 independent, and confirm that ESFT cells may activate different survival pathways in response to diverse microenvironmental conditions.
A survival function for CAV1 was proposed previously on the basis that CAV1 is up-regulated in various cancers, such as human prostate malignancies and multidrug resistant colon tumors, at late, advanced progression stages of the disease, when metastatic and drug resistant phenotypes are prevalent.45 This seems to be also the case in ESFT, in which the involvement of CAV1 in malignant progression and tumorigenic activity was first identified using metastasis-associated gene expression arrays.7 In this regard, our findings agree with previous reports about other tumor types describing that CAV1 over-expression increased the cellular resistance to Dox10 and other drugs,11 but contrast with other reports indicating that CAV1 expression causes drug sensitization.12 It appears highly likely that the different functions attributed to CAV1 in the cellular drug response may depend on the type of cancer, the tumor grade and the progression stage modeled by the experimental cellular system of choice.
Although elevated PKCα expression has been generally associated with increased resistance to Dox and Cp in several tumor types,46 information on the function of PKCs in general, and of PKCα in particular, in the pathophysiology and drug response of ESFT is extremely limited. Our data represent the first description of a role for PKCα in the resistance of ESFT cells to chemotherapeutic drugs. Results clearly indicate that what mediates the pro-survival function of CAV1 in ESFT cells is the increased activating phosphorylation of PKCα at Thr638 that correlates with higher levels of CAV1 (Fig. 3a,b), rather than the total cellular content of PKCα, which showed only minor differences among ESFT cells either endogenously expressing or engineered to express diverse levels of CAV1. Our mechanistic interpretation of these findings is depicted in Figure 7. In ESFT cells with high CAV1 content, CAV1 association with discrete cellular membrane microdomains results in the formation of caveolae,7 where the high number of CAV1 molecules facilitates the binding, through a specific CAV-binding domain,28,29 of PKCα molecules which are either efficiently activated by phosphorylation at Thr638 upon binding, and/or remain there more stably in the active state, if phosphorylated prior to binding.47 In cells with low CAV1 levels, as shown previously in CAV1-knocked down cells,7 caveolae formation is defective or absent and CAV1 associates more sparsely with membrane microdomains, thus making the binding, activating phosphorylation and/or the stable maintenance of the phosphorylated status of PKCα less probable, thereby limiting its contribution to drug resistance. This mechanism is supported by our data (Fig. 5b) showing that (a) re-expression of PKCα in PKCα knocked down A4573 (pShPKCα) cells, which have high CAV1 levels, resulted in a ~4-fold increase in total PKCα and a ~3.91-fold increase in Thr638-phosphorylated PKCα in the (pShPKCα+PKCα) cells, that is, a nearly 1:1 ratio of total vs. phosphorylated PKCα; and that (b) PKCα re-expression in CAV1-knocked down A4573 (pShCav1-1) cells, which express only about 39% CAV1 relative to the controls, resulted in ratio close to 2:1 (2.89-fold vs. 1.61-fold) of total vs. phosphorylated PKCα. These results and those from the response of PKCα-knockdown and re-constituted cells in the presence of high and low CAV1 levels (Fig 5c) clearly show the role of the PKCα-activating phosphorylation at Thr638 as a CAV1 downstream effector in the promotion of drug resistance.
In this regard, the role of Thr638-phosphorylated PKCα in the process of CAV1-promoted drug resistance, of Cp resistance in particular, is reminiscent of the mechanism of Cp resistance promoted by expression of the mt-PCPH oncogene in human prostate carcinoma cells, which also express high CAV1 levels.48 In that system,20 Cp-resistance is also mediated by activation of PKCα through phosphorylation at Thr638, an event that in turn increased the phosphorylation of the Bcl-2 protein and rendered it resistant to proteasome-mediated degradation in prostate cancer cells. Whether a similar downstream effect of PKCα on Bcl-2 is also operative in ESFT cells remains to be elucidated. However, there is evidence suggesting that the anti-apoptotic function of PKCα may not be attributable to its Bcl-2 phosphorylating activity in ESFT cells, taking into consideration the fact that the endogenous levels of Bcl-2 expression in the ESFT cell lines used in our study do not correlate with their CAV1 content or with their response to Dox and Cp treatment (e.g. A4573 cells, the most drug resistant ESFT cell line tested expresses nearly undetectable levels of the Bcl-2 protein49). Further experimentation is obviously required to ascertain the mechanisms that, acting downstream of Thr638-phosphorylated PKCα, contribute to determine the response of ESFT cells to chemotherapeutic treatment.
Overall, this study demonstrates that CAV1 is an important determinant of the resistance of ESFT cells to chemotherapy-induced apoptosis and that PKCα is a critical mediator of the pro-survival functions of CAV1. The significance of these findings is further supported by the co-expression at correlative levels of CAV1 and Thr638-phosphorylated PKCα (Fig. 6) in about 45% of the ESFT specimens analyzed. Although a larger tumor sample needs to be evaluated, the data presented here strongly support the notion that the development of new CAV1- and/or PKCα-targeting strategies may enhance the efficacy of current ESFT therapies and improve the treatment outcome for ESFT patients, particularly those presenting with advanced, metastatic or recurrent disease.
This work was supported by the U.S. National Cancer Institute, NIH/USPHS grant RO1-CA134727, and by the Tissue Culture Shared Resource of the Lombardi Comprehensive Cancer Center funded through USPHS grant P30-CA-CA51008.
We thank Dr. Soledad Gallego (Unitat d’Oncologia Pediatrica, Hospital Universitari Vall d’Hebron, Barcelona, Spain) and Dr. Enrique de Alava (Centro de Investigación del Cáncer, Salamanca, Spain) for contributing ESFT samples to this investigation.