[CANCER RESEARCH 47, 673-679, February 1, 1987] Characterization of the Indirect Antitumor Effect of 7-Interferon Using Ascites-associated Macrophages in a Human Tumor Clonogenic Assay Toshiaki Saito, Michael E. Berens, and E. Charles Welander Section on Gynecologic Oncology, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, North Carolina 27103 ABSTRACT The indirect antitumor effects of recombinant 7-interferon (IKN-T) were investigated using an in vitro tumor clonogenic assay modified to include ascites-associated macrophages (AAM). Untreated AAM stim ulated tumor colony growth; conversely, AAM treated with II' N-T at clinically achievable doses demonstrate a significant growth-inhibiting effect. The indirect ant ¡proliferativi- activity was dependent on the density of AAM. Supernatant* from IFN-7-pretreated AAM cultures derived from 11 different ovarian cancer patients significantly inhibited the colony growth of ovarian cancer cell line BG-1, as well as five of six other cell lines. Physicochemical characteristics of the supernatant indicated that a significant part of the antiproliferotive activity is heat sensitive, de stroyed by proteolytic enzymes, and is dependent on RINA and protein synthesis for production. Neutralizing antiserum against tumor necrosis factor significantly reduced the antiproliferative activity of the supema- tants. Production of this factor by AAM was induced by exposure to 1000 units/ml of ll-'V-y for 15 min, although activity in the supernatants was not detected until 8 h after exposure to IKN-f. Potency of the supernatants reached a peak 12 h after priming and ceased by 22 h. Production of antiproliferative activity was maintained over 5 days by intermittent treatment of AAM with IFN-7. Combinations of IFN-7 and supernatant from IFN-ir-treated AAM showed potentiated antiprolifer ative activity against BG-1 in an additive to synergistic manner. Antitu- mor effects of IF,\ -7 may be dependent on tumor-associated macrophages and treatment scheduling. INTRODUCTION In a previous report we described direct and indirect antipro liferative effects of IFN-71 against human tumor clonogenic cells (1). High concentrations of IFN-7 in the absence of accessory cells demonstrate a reduction in tumor colony for mation. Addition of adherent cells from malignant effusions (>90% macrophages) allows the assessment of IFN-7 effects on intermediary cells, which demonstrate a significant increase in antiproliferative response. This enhancement of antiprolif erative activity was strongly suggested to be attributed to an antiproliferative factor secreted from macrophages. Activation by IFN-7 of cells from monocyte and macrophage lineage for antitumor activity has been documented using cy- tocidal or cytostatic assays (2-6). These antitumor effects have been reported to be attributed to cytotoxic factors secreted by macrophages, such as some enzymes (7, 8) or oxygen metabo lites (9, 10). More recently, several macromolecules with cyto toxic or cytostatic effects have been reported to mediate human monocyte antitumor activity (11-15). Peri et al. (16) report the induction of cytotoxic activity of tumor-associated macro phages by IFN, other lymphokines, or endotoxin. The purpose of this investigation was 2-fold: (a) to biologi cally characterize the antiproliferative supernatant from 11N Received 7/10/86; revised 10/9/86; accepted 10/15/86. 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. 1The abbreviations used are: IFN, interferon; AAM, ascites-associated mac rophages; HTCA, human tumor clonogenic assay; PBS, phosphate-buffered sa line; TAME, SDiliuni p tusyl-i -arginine methyl ester; 11,CK, sodium-/)-tosyl i.- lysine chloromethyl ketone; TNF, tumor necrosis factor; Hu-TNF, human tumor necrosis factor; HMT, human monocyte toxin; LPS, lipopolysaccharide; NK, natural killer, FCS, fetal calf serum. -,-treated macrophages; and (h) to explore the kinetics of pro duction of this substance. In order to investigate these objec tives, this study used a clonogenic assay as a means to repro- ducibly measure the biological end point of tumor colony growth. Antiproliferative effects of an agent which renders cells nondividing are measured as a reduction in tumor colony for mation. MATERIALS AND METHODS Human Tumor Clonogenic Assay The method originally described by Hamburger and Salmon (17) was used in this study with slight modification. Briefly, lower layers were prepared in 35-mm wells of tissue culture plates (Linbro; Flow Lab, McLean, VA) using RPMI-1640 culture medium with 20% FCS and 1% penicillin/streptomycin (Grand Island Biological Company, Grand Island, NY) in 0.8% agarose (FMC Corp., Rockland, ME). Target cells in the upper layer were suspended in the same medium as the lower layer, with 0.3% agarose. The plating density of each cell line was between 1 x 10*and 1 x 10*cells per well, selected in order to achieve at least 200 tumor colonies per control well as well as being a cloning density located midway on the linear portion of the dose-response curve. This allows assessment of colony growth stimulation as well as inhibition. Single cell suspensions were ensured at the time of plating by visual inspection using inverted microscopy. Agarose cultures of tumor cells were incubated at 37*C in a hu midified atmosphere with 7.5% CO2 for 7 to 10 days. At the end of the incubation, colonies were counted with a computerized feature analysis system (Omnicon FAS II; Bausch & Lomb, Rochester, NY) (18). Colonies were defined as cell aggregates at least 60 ^ni in diameter. Previous experience with cytotoxic agents as well as cytokines indicates that interpretation of drug effects on cloning is not significantly influ enced by selection of colony sizes between 40 and 70 urn, as long as the cell density resides on the linear portion of the dose-response curve. The plating efficiency of each cell line was consistent over the course of serial passages as previously reported (1). Human Tumor Cell Lines Seven cell lines were used in this study as target cells in the colony- forming assay. They are derived from two ovarian carcinomas (BG-1) (19) and (BG-3),2 an endometrial carcinoma (HEC-1A) (20), a mela noma (SK-MEL 28) (21), a cervical carcinoma (ME-180) (22), a breast carcinoma (MCF-7) (23), and a myeloma (RPMI-8226) (24). A mac- rophage-like cell line (U-937) (25) was used as feeder cells in the lower layer of the colony-forming assay. U-937 and RPMI-8226 were serially maintained as suspension cultures in RPMI-1640 medium supple mented with 20% heat-inactivated FCS (Gibco, Chagrin Falls, OH). The other 6 cell lines were maintained as monolayer cultures in Eagle's minimal essential medium with 10% FCS and 0.1% insulin and pas saged after mild trypsinization. Feeder Cells for Clonogenic Assay Cultures As previously described in detail (1), macrophages were isolated from malignant ascites obtained from patients having histologically proven ovarian cancer. Briefly, mononuclear cells were separated from ascitic fluid using gradient centrifugal ¡onon Ficoll-Paque (Pharmacia, Pis cataway, NJ). After washing with PBS, the cells were resuspended in RPMI-1640 medium supplemented with 20% FCS and decanted into ' Unpublished data. 673 Research. on February 12, 2016. © 1987 American Association for Cancer cancerres.aacrjournals.org Downloaded from