Chemiluminescence response of human natural killer cells. I. The relationship between target cell binding, chemiluminescence, and cytolysis

The binding of tumor cells or fetal fibroblasts to human natural killer (NK) cells led to a rapid chemiluminescence response within seconds of target-effector interaction. The degree of chemiluminescence was dependent on the concentration of NK-enriched lymphocytes or target cells, and plasma membrane vesicles from K562 also induced a chemiluminescence response. Mild glutaraldehyde treatment of effector cells abrogated their ability to generate chemiluminescence, whereas K562 target cells treated in the same way were almost fully able to induce a chemiluminescence response to NK-enriched lymphocytes. These results show a directionality of response with NK as the responders and tumor cells as the stimulators. A survey of eight different tumor cell lines and fetal fibroblast lines revealed a striking correlation (r greater than 0.93, P less than 0.001) between the ability of a given line to bind to NK-enriched lymphocytes, induce chemiluminescence, and to be lysed. Three differentiated sublines of K562 grown in butyrate and cloned induced little chemiluminescence compared with the K562 parent, and they were selectively resistant to NK-mediated binding and cytolysis. In addition, treatment of K562 cells with higher concentrations of glutaraldehyde for longer periods led to varying degrees of target antigen preservation, as measured in cold target competition assays and in conjugate formation. The degree of NK target antigen preservation correlated directly with the ability of the cells to induce chemiluminescence (r greater than 0.95). The degree of NK activation was also important because interferon-pretreated effectors generated more chemiluminescence upon stimulation with K562 or MeWo targets. Monocytes or granulocytes did not contribute to the chemiluminescence induced by NK-sensitive targets. Some NK-resistant tumor cell lines were sensitive to monocyte-mediated cytolysis and also induced chemiluminescence in monocytes but not NK cells. These results show that the target structures recognized by the NK cell may play a role in NK activation because the degree of chemiluminescence was directly proportional to the ability of a given target cell line to bind to the NK cell and to be lysed.