Rapid killing of actinomycin D-treated tumor cells by human mononuclear cells. I. Effectors belong to the monocyte-macrophage lineage

Preincubation of tumor cells with actinomycin D (Act D) rendered various murine and human lines susceptible to killing by unstimulated human peripheral blood mononuclear cells (PBM) in a 6-hr ⁵¹Cr-release assay. The murine WEHI 164 sarcoma was selected for analysis of drug-dependent cellular cytotoxicity (DDCC) because high levels of killing were detected with this tumor, and it was considerably resistant to natural killer (NK) cell activity. Optimal conditions for induction of susceptibility to lysis included a 3-hr preincubation with 1 μg/ml Act. D. Effector cells of cytotoxicity against Act D-treated WEHI 164 cells were plastic adherent (> 85% monocytes). Cells nonadherent to plastic and nylon wool (≤ 1% monocytes) had no appreciable DDCC activity. In contrast NK activity against K562 cells was mediated by nonadherent cells. When PBM were fractionated on a one step discontinuous gradient of Percoll designated to enrich for monocytes (> 90% pure), DDCC activity was found in the monocyte fraction, and the lymphoid cell-enriched fraction had no cytotoxicity against Act D-treated WEHI 164 cells. In contrast, NK activity against K562 was recovered with lymphoid cells, and monocytes had no NK cytotoxicity. Upon fractionation on a six step Percoll gradient designed to enrich for large granular lymphocytes (LGL), the denser lymphocytes (fraction 4-6) and the less dense LGL with NK activity (fraction 2-3) had no cytotoxicity against Act D-treated WEHI 164 sarcoma cells. DDCC activity sedimented in fraction 1 in association with monocytes. PBM were fractionated according to monoclonal antibody-defined surface markers by using a fluorescence-activated cell sorter. Effector cells of DDCC were positive for monocyte markers (Mo2, UCHM1) and were negative for NK cell (B73.1, HNK1), T cell (T11), and B cell (Leu-10) markers. Macrophages obtained by culturing blood monocytes in vitro for 5 to 10 days had DDCC activity. Similarly, peritoneal and bronchoalveolar macrophages had considerable cytotoxicity against Act D-treated target cells, whereas minimal or no NK activity was found at the same anatomic sites. Cells of human or murine origin, preincubated with Act D for 3 hr, were heterogeneous in their susceptibility to monocyte killing in a 6-hr ⁵¹Cr-release assay. High levels of cytotoxicity were observed with the murine WEHI 164 sarcoma and 3T3 'fibroblast' line and with the human CEM leukemia. Monocytes were weakly (but significantly) cytotoxic against the ALAB breast carcinoma (human) and the 8387 sarcoma (human). The YAC-1 lymphoma, TU5 transformed kidney, SW 626 ovarian carcinoma, the K562 leukemia and Raji lymphoma, two fibroblast preparations (human and murine), human lymphocytes, and lymphoblasts were not appreciably killed after preincubation with Act D under these experimental conditions. Cytotoxicity on Act D-treated tumor cells was not the expression of rapid elimination of effete tumor cells, already incapable of self renewal, because monocyte-mediated DDCC was also detected in a colony assay. These data are consistent with the possibility that unstimulated effector cells involved in rapid killing of tumor cells pretreated with the chemotherapeutic drug Act D belong to the monocyte-macrophage lineage. DDCC may provide a measure of human monocyte-macrophage cytotoxic potential in a short term ⁵¹Cr-release assay and may be a pharmacologic tool to investigate the regulation of target cell susceptibility in macrophage-tumor cell interactions.