A Role for the Interleukin 1 Receptor in the Synergistic Antitumor Effects of Human Interleukin la and Etoposide against Human Melanoma Cells

To investigate the possibility that anticancer drugs combined with cytokines may show increased activity, human tumor cells were treated with combinations of human recombinant interleukin la (rIL-la) and etoposide (VP-16). The cytotoxicity of these combinations was evaluated by the 3-(4,S-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide as say using rIL-la-sensitive A375-C6 melanoma cells and A37S-C5 cells, a clonal variant line that is resistant to IL-la. Data were analyzed for synergism by the median effect principle of T-C. Chou and P. Talalay (J. Biol. Chem., 252:6438-6442,1977). At a dose ratio of VP-16 to rIL-la of 12 UM:I unit/ml in either simultaneous or sequential exposure (VP-16 first), the calculated combination index values indicated synergistic cy totoxicity toward both A375-C6 cells and A375-C5 cells. IL-la treatment 24 h prior to VP-16 exposure had no advantage over simultaneous treatment. Surface IL-la receptors on both A375-C6 and A375-C5 cells were measured using I25l-radiolabeled rIL-la binding; A375-C6 cells had 701 ±128 (SD) receptor molecules/cell and A375-C5 cells only had 58 ±33 receptor molecules/cell. The dissociation constants for IL-la were similar in both cell types (19 ±6 pM for A375-C6 and 17 ±2 pM for A375-CS). The specific binding of rIL-la to the surface IL-la receptors of both sensitive and resistant cells was significantly increased in a dosedependent fashion by the prior treatment with VP-16 (1.75-fold on A375C6 cells and 3.5-fold on A375-C5 cells). VP-16 also enhanced the internalization of receptor-bound rIL-la, suggesting that a possible mech anism of the synergistic cytotoxicity of rIL-la and VP-16 might be related to the modulation of rIL-la receptors by VP-16, resulting in increased internalization of rIL-la. INTRODUCTION IL-12, a cytokine produced predominantly by monocytes and macrophages, has many diverse activities in vivo and in vitro including antitumor effects (1). Direct antitumor activity of IL1 has been reported against malignant melanoma /'/;vitro, breast cancer cells in vitro, and murine pancreatic cancer in vivo (28). However, the efficacy of IL-1 alone against various cancers is limited because of its severe host toxicity in the clinic. Although the mechanism(s) by which IL-1 induces cell killing is not clear yet, the binding of IL-1 to its membrane receptors and subsequent internalization are essential for cytotoxicity. VP-16, a synthetic derivative of podophyllotoxin, is clinically active against several human cancers (9). The mechanism of the antitumor effects of VP-16 has been suggested to result from its binding and stabilization of the cleavable DNA-topoisomerase II complex, resulting in both DNA singleand doublestrand breaks. In addition, metabolic activation of VP-16 by cytochrome P-450 and prostaglandin synthase to reactive spe cies has also been indicated to contribute significantly in tumor cell kill (10-13). Received 8/24/90; accepted 11/6/90. 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. 1To whom requests for reprints should be addressed at Building 10, Room 6N-119, National Cancer Institute, NIH, Bethesda. MD 20892. 2The abbreviations used are: IL-1, interleukin 1; VP-16, etoposide; rIL-la, recombinant human inlerleukin la; TNF, tumor necrosis factor; IL-1R, interleu kin-1 receptor. Recently, combinations of cytotoxic agents and biological response modifiers have evolved as an important modality for treating cancers, because it is assumed that these agents are different from chemotherapeutic agents (14, 15). IL-I showed synergistic antitumor effects in combination with another cy tokine, tumor necrosis factor-o, toward malignant melanoma (16) and chondrosarcoma (17). Recently, studies from our laboratory have shown that syn ergistic antitumor effects of rIL-la combined with VP-16 for A375-C6 melanoma cells, which are sensitive to IL-1, and A375-C5 cells, a variant line 10-fold resistant to IL-1 (18). In this study, we have investigated the interactions between rlLla and VP-16 by comparing several drug administration sched ules and dose ratios of the 2 agents. Additionally, we examined the effects of each agent on the cellular accumulation of the other. A possible mechanism of the synergism between rIL-la and VP-16 may be related to the increased membrane receptor binding and internalization of rIL-1«following VP-16 treat ment. MATERIALS AND METHODS Drugs and Reagents. rIL-la (specific activity, 2.26 x 10' units/mg protein; A/r, 18,063) was generously provided by Dainippon Pharma ceutical Co. (Osaka, Japan). VP-16 was a gift from Bristol Myers Co. (Syracuse, NY). 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide, glycine, and dimethyl sulfoxide were purchased from Sigma Chemical Co. (St. Louis, MO). RPMI1640, trypsin-EDTA, fetal bovine serum, antibiotic mixture, and phosphate buffered saline were pur chased from Gibco Laboratories (Grand Island, NY). '"I-labeled rlL1«(specific activity, 2000 Ci/mmol; M,, 18,200) was purchased from Amersham (Arlington Heights, IL). Cells. Human melanoma A375-C6 (IL-la-sensitive clone) cells and A375-C5 (IL-la-resistant clone) cells, kindly provided by Dr. Kouji Matsushima, National Cancer Institute, were maintained in RPMI 1640 supplemented with 10% fetal bovine serum and antibiotic mixture under standard culture conditions at 37°Cin a humidified 5% COi atmosphere. Cytotoxicity Assay. The antitumor activity of drugs was measured by the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide as say (19). For the sequential administration, the first drug was removed and the cells were washed once by drug-free medium before adding the second drug. The evaluation for the synergy of the combinations of rlLla and VP-16 was carried out by the analysis developed by Chou and Talalay (20-22). l25I-IL-la Receptor Binding Assay. The cells (1.2 x 106/well in 12well tissue culture plates) were incubated with a range of concentrations of '"I-labeled rIL-la (specific activity, 2000 Ci/mmol) in 0.5 ml of cold RPMI 1640 supplemented with 10% fetal bovine serum in the absence or presence of a 1000-fold excess of unlabeled rIL-la for 2 h at 4°C. After the cells were washed with cold complete medium, 0.5 ml of l N NaOH were added to dissolve the cells and release the rIL-la. Subsequently, the contents of the wells were transferred into scintilla tion vials, the wells were rinsed with 0.5 ml of l N HC1, and 10 ml of Hydrofluor (Manville, NJ) were added to each vial for liquid scintilla tion counting of the bound '"I-labeled rIL-la. To investigate the effects of VP-16 on the IL-la receptor, the cells