E1A Sensitizes HER2/neu-overexpressing Ewing’s Sarcoma Cells to Topoisomerase II-targeting Anticancer Drugs

INTRODUCTION

Ewing’s sarcoma is the second most common malignant bone tumor in children (1) . The majority of patients have microscopic metastases at diagnosis; the lung is the most common metastatic site (2) . These patients initially respond to systemic chemotherapy plus surgery and/or radiation. However, at 2 years, 40–60% will, depending on the prognostic indicators, have a recurrence. Adverse prognostic indicators include large tumor volume or tumor length, tumor location, elevated serum lactic dehydrogenase, and metastatic spread to bone, lung, or bone marrow. A number of different chemotherapy regimens, including those comprising total-body irradiation and autologous bone marrow transplant, have been unsuccessful in increasing the survival rates at 2 and 3 years (3, 4, 5, 6, 7) . Furthermore, relapsed patients are particularly resistant to second-line chemotherapy, leaving few treatment options.

The HER2/neu oncogene encodes a M_r 185,000 human epidermal growth factor receptor-2 transmembrane protein. Overexpression of HER2/neu has been found in ∼30% of human breast and ovarian cancers (8 , 9) . Enhanced expression of HER2/neu increased the tumorigenicity and metastatic potential of human cancer cells (10 , 11) . Clinical studies indicate that overexpression of HER2/neu correlates with poor prognosis, shorter patient survival in osteosarcoma (12 , 13) . Because of the close association reported between overexpression of HER2/neu and resistance to chemotherapeutic agents (14 , 15) and because of the poor disease-free survival rates associated with multi-agent chemotherapy in Ewing’s sarcoma, we investigated HER2/neu expression in several different Ewing’s sarcoma cell lines. HER2/neu expression was high in all three of the Ewing’s sarcoma cell lines examined.

Down-regulation of HER2/neu may be an important approach for cancer therapy (16 , 17) . The E1A 3 gene has been shown to be a tumor suppressor for cells that overexpress HER2/neu (18 , 19) . Transfer of the E1A gene results in transcriptional repression of HER2/neu via an effect on the HER2/neu promoter. In vivo, E1A gene transfer suppressed the dissemination of ovarian cancer cells that overexpressed HER2/neu and enhanced the in vivo sensitivity of the cells to paclitaxel (20 , 21) . In this study, we found that E1A down-regulated HER2/neu expression, inhibited tumor cell growth, and increased cellular sensitivity to topoisomerase II-directed anticancer drugs. These results suggest that E1A gene therapy aimed at down-regulation of HER2/neu may provide a new approach for the therapy of Ewing’s sarcoma.

MATERIALS AND METHODS

Cell Lines.

TC71 human Ewing’s sarcoma cells, kindly provided by Dr. P. Pepe (University of Southern California, Los Angeles, CA), were cultured in Eagle’s modified essential medium with 10% fetal bovine serum. SKRB-3 human breast cancer cells and RD human Ewing’s sarcoma cells were obtained from American Type Culture Collection (Manassas, VA). A4573 human Ewing’s sarcoma cells were a generous gift from Dr. Soldatenkov (Georgetown University Medical Center, Washington, DC) and were cultured in RPMI 1640 with 15% fetal bovine serum. Normal human osteoblast cells were purchased from Clonetics, Inc. (San Diego, CA) and were maintained in the special medium provided by Clonetics. All cells were free of Mycoplasma as screened by Gen-Probe (Gen-Probe, San Diego, CA).

Recombinant Adenovirus.

Ad-E1A is an adenovirus type 5 that contains E1A but lacks E1B and E3 (18) . The control adenovirus, Ad-β-gal, is an adenovirus type 5-based vector that lacks E1A, E1B, and E3 but contains β-gal. Both of these recombinant replication-deficient adenoviral vectors were propagated in 293 cells as described previously (22) . The viruses were purified twice using cesium chloride-gradient ultracentrifugation and then dialyzed and titrated by the standard method. Cells were infected with adenovirus at 10 plaque-forming units/cell for 48 h and then treated as indicated for various experiments.

Cytostasis Assay.

Cells were seeded into 96-well cell culture plates (5000/well) and allowed to adhere overnight before being infected with Ad-E1A or Ad-β-gal. Various concentrations of VP-16, Adriamycin, or cisplatin were then added in triplicate. The antiproliferative activity was determined 48 h later by the MTT assay as described previously (23) .

Western Blot.

Cells were set in 100-mm dishes (2 × 10⁶ per dish) 1 day before treatment and then were infected with Ad-E1A or Ad-β-gal for 48 h. Cells were washed with cold PBS buffer and lysed in RIPA buffer (1% NP40, 0.5% sodium deoxycholate, and 0.1% SDS in PBS) containing the fresh protease inhibitor aprotinin (2 μg/ml), leupeptin (2 μg/ml), pepstatin A (1 μg/ml), and phenylmethylsulfonyl fluoride (100 μg/ml; Sigma Chemical Co., St. Louis, MO). The cells were then scraped and passed through a 21-gauge needle. Protein concentrations were determined using the Bio-Rad protein assay kit (Bio-Rad Laboratories, Hercules, CA). After 50 μg of the protein were solubilized in sample buffer, the solution was boiled for 5 min before loading onto a 7.5% SDS-polyacrylamide gel and then transferred to a nitrocellulose membrane. The specific protein bands were detected with a human Her2/c-neu (Ab-3) monoclonal antibody (Calbiochem, San Diego, CA), E1A (adenovirus type 5) antibody, PARP antibody (BD PharMingen, San Diego, CA), human topoIIα antibody (TopoGEN, Columbus, OH), and β-actin monoclonal antibody (Sigma) using the ECL Western blotting analysis system (Amersham) according to the manufacturers’ instructions. Densitometric analysis was performed, and values were normalized with β-actin loading control.

Flow Cytometry Analysis.

Apoptotic cells were detected by flow cytometry as described previously (24) . TC71 cells were treated as indicated and then were washed with PBS buffer twice; 10⁵ cells were incubated with propidium iodide solution (50 μg/ml propidium iodide in 0.1% sodium citrate plus 0.1% Triton X-100) for 2 h. Cells were analyzed with a FACScan (Becton Dickinson Co., Mountain View, CA).

RESULTS

High levels of HER2/neu protein were detected in three different Ewing’s sarcoma cell lines (TC7, RD, and A4573; Fig. 1 ⇓ ) compared with the level in normal human osteoblast cells (NH Osteoblast). SKBR-3, a human breast cancer cell line well known to overexpress HER2/neu, was used as the positive control. Using HER2/neu expression in the normal human osteoblast cells as the baseline, the HER2/neu expression was 20.5-fold higher in TC71 cells, 11.2-fold higher in RD cells, and 13.9-fold higher in A4573 cells. We concluded that all three Ewing’s sarcoma cell lines overexpress HER2/neu.

Effect of E1A on HER2/neu Expression.

To determine whether E1A could down-regulate HER2/neu overexpression, TC71 cells were infected with Ad-E1A or control Ad-β-gal. HER2/neu and E1A expression were determined 48 h later (Fig. 2) ⇓ . TC71 cells infected with Ad-E1A expressed very high levels of E1A and had significantly reduced HER2/neu expression. The densitometric analysis suggested that HER2/neu expression was 80% lower in the E1A-infected TC71 cells than in the TC71 control cells. In contrast, TC71 cells infected with Ad-β-gal did not express E1A and had no significant change in the expression of HER2/neu.

Effect of HER2/neu on Cell Growth and Sensitivity to Chemotherapeutic Agents.

Infection of TC71 cells with Ad-E1A inhibited cell growth by 25% (Fig. 3) ⇓ . By contrast, infection with Ad-E1A did not increase cytostasis in normal HER2/neu-expressing human osteoblast cells compared with those infected with Ad-β-gal. Down-regulation of HER2/neu overexpression by E1A resulted in inhibition of TC71 tumor cell growth. E1A also induced an 18-fold increase in the sensitivity of TC71 cells to VP-16 (Fig. 4) ⇓ . After transduction with the E1A gene, the IC₅₀ was reduced from 1500 to 80 nm. Ad-β-gal did not significantly change the sensitivity of the cells to VP-16. The sensitivity of TC71 cells to Adriamycin was also increased, with a reduction in the IC₅₀ from 90 to 18 nm (Table 1) ⇓ . By contrast, the sensitivity of TC71 cells to cisplatin did not significantly change after infection with Ad-E1A.

We next quantified the apoptosis of TC71 cells, both control and E1A-transduced, after exposure to VP-16 (Fig. 5) ⇓ . The percentage of sub-G₁ phase cells was only 2.4% in the control cells, 9.5% in cells treated with Ad-E1A, and 7.1% in cells treated with low-dose VP-16 (0.3 μm). The percentage of apoptotic cells in the cells treated with the same dose of VP-16 (0.3 μm) after infection with Ad-E1A was significantly increased to 29.3%. By contrast, Ad-β-gal did not significantly increase the rate of apoptosis induced by VP-16. These results were supported by Western blot analysis of PARP cleavage. PARP is a M_r 116,000 nuclear chromatin-associated enzyme. During cell apoptosis, PARP is cleaved from the M_r 116,000 intact form into the M_r 85,000 fragment. This cleaved band is a marker of apoptosis. As shown in Fig. 6 ⇓ , the M_r 85,000 fragment was significantly more prominent in TC71 cells treated with Ad-E1A and VP-16 than in TC71 control cells, cells treated with VP-16 alone, or cells treated with Ad-β-gal and VP-16.

Effect of E1A Gene Transfer on topoIIα.

topoIIα is the enzyme target for VP-16 and one of the targets for Adriamycin. By contrast, cisplatin is not considered to be a topoII-reactive drug. Because E1A gene transfer increased both VP-16 and Adriamycin sensitivity but had no significant effect on cisplatin sensitivity, we hypothesized that E1A gene transfer may affect not only HER2/neu but also topoIIα. As shown in Fig. 7 ⇓ , topoIIα protein levels were significantly increased in TC71 cells after exposure to Ad-E1A but not in cells exposed to Ad-β-gal.

DISCUSSION

Overexpression of the HER2/neu gene is correlated with enhanced tumorigenicity, enhanced metastatic potential, poor patient prognosis, and decreased chemosensitivity in many types of cancer, including breast, ovarian, lung, and stomach carcinomas (25 , 26) . This gene was also reported recently to be a potential prognostic indicator for osteosarcoma (13) . The study presented here demonstrated that the HER2/neu oncogene is overexpressed in three different Ewing’s sarcoma cell lines. E1A gene transfer using an adenoviral vector resulted in down-regulation of HER2/neu, decreased cell growth, and increased cellular sensitivity to VP-16 and Adriamycin but not cisplatin. Down-regulation of HER2/neu has been shown to increase apoptosis in tumor cells that overexpress the oncogene. We demonstrated here a similar increase in cell apoptosis and cytostasis after HER2/neu down-regulation.

Increased cell sensitivity to chemotherapy after E1A gene transfer has been demonstrated in several different tumor cell lines (27) . However, the mechanism by which E1A enhances drug activity is not clear. E1A indirectly suppresses the HER2/neu promoter through the co-activator p300 (28) . How this function then impacts on drug sensitivity and why the activity of certain drugs is enhanced whereas the activity of others is not are poorly defined. In this study, we show that, in addition to down-regulating HER2/neu, E1A transduction also significantly increases topoIIα protein expression. Using a specific human topoIIα antibody, we were able to detect a significant increase in the topoIIα protein levels in cells transduced with the E1A gene but not in those transduced with the β-gal gene. Sensitivity to VP-16 and Adriamycin has been shown to be directly linked to the amount of topoIIα protein in the cell. Indeed, we demonstrated recently that cellular resistance to VP-16 and doxorubicin could be circumvented after human topoIIα gene transfer (29) . topoIIα gene transfer into MDA-VP etoposide-resistant human breast cancer cells using an adenoviral vector increased the level of topoIIα protein, and the sensitivity of these cells to VP-16 and doxorubicin was enhanced. topoIIα is the sole enzyme target of VP-16 and one of the targets of Adriamycin. In contrast, the mechanism by which cisplatin kills tumor cells has nothing to do with topoIIα. Thus, our data showing that E1A gene transfer enhanced the activity of VP-16 and Adriamycin, but not that of cisplatin, support our hypothesis that the mechanism by which E1A enhanced the chemosensitivity of Ewing’s sarcoma cells was an increase in topoIIα.

Ewing’s sarcoma is a relatively rare disease with limited therapeutic options. The majority of patients are initially responsive to chemotherapy with vincristine, Adriamycin, and cyclophosphamide. However, relapsed disease is usually explosive and extremely difficult to treat because of its resistance to chemotherapy. The role of oncogenes and growth factors as well as the etiology of chemoresistance in this disease is poorly understood.

We have demonstrated that HER2/neu is overexpressed in three different human Ewing’s sarcoma cell lines. E1A down-regulated HER2/neu expression, inhibited tumor cell growth, and increased cellular sensitivity to two topoII-directed anticancer drugs. Interestingly, E1A gene transfer had no affect on the growth rate of normal osteoblast cells or on their susceptibility to chemotherapy (Fig. 3 ⇓ ; other data not shown). These data indicate tumor-specific treatment effect, an important consideration for therapeutic potential. These results suggest that E1A gene therapy may provide a new approach for treatment of Ewing’s sarcoma. Clinical trials with E1A gene therapy are being conducted in patients with metastatic breast cancer and ovarian cancer (30) . Because there is a dearth of therapeutic options for patients with relapsed Ewing’s sarcoma, our data indicate that E1A gene therapy may have potential in the treatment of this disease.