Diphtheria Toxin−Epidermal Growth Factor Fusion Protein and

Seven of fourteen cell lines were highly sensitive to DAB389EGF alone, and six ..... positive cells (p < 0.0001) but not the mean EGFR density measure...
0 downloads 0 Views 625KB Size
Bioconjugate Chem. 2003, 14, 1107−1114

1107

Diphtheria Toxin-Epidermal Growth Factor Fusion Protein and Pseudomonas Exotoxin-Interleukin 13 Fusion Protein Exert Synergistic Toxicity against Human Glioblastoma Multiforme Cells Tie Fu Liu, Mark C. Willingham, Stephen B. Tatter, Kimberley A. Cohen, A. Corinne Lowe, Andrew Thorburn, and Arthur E. Frankel* Departments of Cancer Biology, Surgery, Pathology and Comparative Medicine, Wake Forest University School of Medicine, Winston-Salem, North Carolina 27157. Received July 1, 2003; Revised Manuscript Received October 7, 2003

The cytotoxicity of combinations of a diphtheria toxin-human epidermal growth factor fusion protein (DAB389EGF) and a Pseudomonas exotoxin-human interleukin 13 fusion protein (IL13PE38QQR) was tested against 14 human glioma cell lines. After cells were cultured for 48 h with various concentrations of the fusion proteins, the percentage reductions in thymidine incorporation were determined. Seven of fourteen cell lines were highly sensitive to DAB389EGF alone, and six cell lines were highly sensitive to IL13PE38QQR alone with IC90’s < 100 pM. When combined, synergistic cell killing was observed for seven of the cell lines based upon concave isobolograms and combination indices (CI’s) of 0.2 to 0.7. Supraadditive cytotoxicity was confirmed by measurements of induction of apoptosis. Receptor expression was assessed by flow cytometry and confocal microscopy. Marked heterogeneity of expression of EGFR and IL13RR2 was seen on all the glioma cell lines. This heterogeneity may contribute to incomplete cell killing with the individual fusion proteins and synergistic cell kill with the combination. These results suggest that both fusion proteins may yield antitumor effects in patients with recurrent gliomas and that combination fusion protein intracranial therapy of malignant gliomas may yield an improved therapeutic index.

INTRODUCTION

There are 17 000 cases of primary brain tumors per year in the United States (1). Over 80% of these tumors are infiltrative astrocytomas/GBM1 which arises from malignant transformation of astrocytes (2). In most cases, treatment is unsatisfactory with a median survival of one year for newly diagnosed patients and six months for patients with recurrent disease (3). The two-year survival rate for GBM patients is less than 20%. Both local spread of tumor cells beyond sites of surgery and radiation and chemotherapy- and radiation-resistant tumor cells contribute to the development of refractory disease (4, 5). Novel agents with different mechanisms of action are needed. One such class of therapeutics is fusion proteins consisting of protein synthesis-inactivating peptide toxins fused to brain tumor-selective ligands. The ligand directs the molecule to the glioma cell surface; after internalization and translocation to the cytosol, the peptide toxin moiety catalytically inactivates protein synthesis leading to cell death. Fusion proteins targeting brain tumor Tf receptors, IL13 receptors, IL4 receptors, and EGFR have been prepared and administered by CED directly into the brain tumor interstitium of patients with refractory GBM (6-9). CED creates a bulk flow that supplements diffusion and leads to achievement of drug concentrations * To whom correspondence should be addressed. Phone: (336) 716-3313; fax: (336) 716-0255; e-mail: [email protected]. 1 Abbreviations: GBM, glioblastoma multiforme; IL, interleukin; Tf, transferring; CED, convection-enhanced delivery; EGFR, epidermal growth factor receptor; PE, Pseudomonas exotoxin; DT, diphtheria toxin; MTD, maximum tolerated dose; DLT, dose-limiting toxicity.

hundreds-fold greater than by systemic infusion for large areas of the brain parenchyma (10). Fusion protein CED therapy has yielded remissions lasting years in a significant fraction of patients. However, not every patient responds and toxicities to normal brain have occurred. Thus, there is a need to improve the technology. Since most brain tumors are heterogeneous in their expression of receptors and even glioma cell lines display cell-to-cell variations in receptor density, we reasoned that combinations of fusion proteins targeting different receptors would enhance glioma cell kill and permit lower doses of each fusion protein to be usedsreducing normal brain toxicity. In this paper, we tested the cytotoxicity of combinations of DAB389EGF and IL13PE38QQR fusion proteins targeting the EGFR and IL13R. Fourteen human glioma cell lines were evaluated. Combination indices and isobolograms were used to assess the presence and degree of synergy (11). In addition, we measured the expression of each of the receptors and estimated the level of receptor heterogeneity. MATERIALS AND METHODS

Fusion Proteins and Antibodies. DAB389EGF was a gift of Ligand Pharmaceuticals (San Diego, CA). The fusion protein was synthesized and partially purified as previously described (12). Vials contained sterile filtered and lyophilized 500 µg of DAB389EGF, phosphate-buffered saline pH 7.2, 1% mannitol, and 50 µM EDTA. Vials were stored at -80 °C. Fusion protein was prepared for experiments by adding 2 mL of sterile water and gently agitating. Dissolved fusion protein was stored at 4 °C. Lot number was 3E13EB2. IL13PE38QQR was a gift of NeoPharm (Lake Forest, IL). The fusion protein was synthesized and partially

10.1021/bc034111+ CCC: $25.00 © 2003 American Chemical Society Published on Web 10/29/2003

1108 Bioconjugate Chem., Vol. 14, No. 6, 2003

purified as described (13). Vials contained sterile filtered IL13PE38QQR at 200 µg/mL in normal saline with 0.2% human serum albumin. Material was stored in aliquots at -80°C until used. Murine monoclonal antibody anti-IL13RR2 and Rphycoerythrin (PE)-conjugated murine monoclonal antibody anti-IL13RR2 was purchased from Diaclone SAS (Besancon, France). Mouse anti-EGFR antibody conjugated to fluorescein isothiocyanate (FITC), mouse isotype control antibody IgG2bκ conjugated to FITC, and PEconjugated mouse isotype control antibody IgG1 were obtained from BD Biosciences (Mountain View, CA). Mouse IgG and rhodamine-conjugated goat anti-mouse IgG were purchased from Jackson ImmunoResearch (West Grove, PA). Cell Lines. Cell lines U373MG and U138MG were obtained from the Wake Forest University Tissue Culture Core Laboratory. Cell lines A172, DBTRG05MG, T98G, U87MG, and U118MG were purchased from the American Type Culture Collection (Rockville, MD). Cell lines LN405, GAMG, DKMG, GMS10, 42MGBA, SNB19, and 8MGBA were purchased from the Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH (braunschweig, Germany). Cells were cultured at 37 °C in a humidified atmosphere containing 5% CO2 in air and maintained as described previously (14). Cell Cytotoxicity Assays. Aliquots of 5000 cells were incubated in triplicate in 100 µL of medium (same as that used to grow the cells) in Costar 96-well flat-bottomed plates. After 24 h, DAB389EGF, IL13PEQQR, or both in medium were added to each well to yield concentrations ranging from 0 to 1000 pM, and the cells were incubated at 37 °C in 5% CO2 for another 48 h. After the incubation, 0.5 µCi of [3H]thymidine (NEN DuPont, Boston, MA) in 50 µL of medium was added to each well, and incubation was continued for an additional 16 h at 37 °C in 5% CO2. Plates were frozen at -80 °C and thawed; cell lysates were harvested using a Skatron Cell Harvestor (Skatron Instruments, Lier, Norway) on to glass fiber mats; and the cpm of incorporated radiolabel were counted using an LKB liquid scintillation counter gated for 3H. The IC90 was defined as the concentration of toxin that inhibited thymidine incorporation by 90% compared to control wells. The percentage of maximum [3H]thymidine incorporation was plotted versus the log of the toxin concentration, and nonlinear regression with a variable-slope sigmoidal dose-response curve was generated along with IC90 with use of GraphPad Prism software (GraphPad Software, San Diego, CA). All assays were performed in triplicate. Multiple drug-effect analysis was done using CalcuSyn software (BioSoft, Ferguson, MO). Both the IC90 isobologram and the IC90 combination index (CI) were derived using the software and the equation of T. C. Chou (11). The CI ) A/Ao + B/Bo. Ao and Bo are individual drug concentrations producing the effect; A and B are the drug concentrations in the combination yielding the same effect. Isobolograms are graphs of A versus B with individual points (A, B) noted yielding the desired effect. The graphs of (A, B) points for a given effect yield concave, straight lines or convex curves for synergistic, additive or antagonistic interactions, respectively. For analysis of nuclear morphology, cells were treated with the drugs for 48 h then stained with Hoescht 3258 as previously described (15). Cells were deposited onto slides using a cytospinner and scored for apoptotic nuclei by fluorescence microscopy. Flow Cytometry. Aliquots of 1 000 000 cells were pelleted at 600g and resuspended in 80 µL of phosphate

Liu et al.

buffered saline (PBS) with 1% bovine serum albumin (BSA). Different aliquots were incubated with 10 µL of PE-conjugated mouse IgG1 anti-IL13RR2 antibody or PEconjugated mouse IgG1 isotype control antibody. After 45 min on ice, cells were washed with PBS/BSA and incubated with 10 µL of FITC conjugated mouse antiEGFR antibody or FITC conjugated mouse IgG2bκ isotype control antibody. After 30 min on ice, cells were washed with PBS/BSA and resuspended in 1 mL of PBS/ BSA. A volume of 250 µL of 3.7% formaldehyde was added to each tube, and the cells were assayed on an EPICS-XL flow cytometer (Coulter, Hialeah, FL) with filters set for simultaneous PE and FITC fluorescence detection. Gates were based on isotype control antibody staining such that there were 6000

>6000 >6000 >6000 13 8 56 >6000 85 15 >6000 1899 6.2 6824 >6000

ratio IC90’s

CI (90% cell kill)