Evidence for a similar or common mechanism for natural killer cell activity and resistance to hemopoietic grafts

Abstract Two types of host reactivities not requiring immunization in the mouse and not mediated by T lymphocytes were compared: resistance of irradiated and nonirradiated F 1 hybrids to accept parental grafts of normal or malignant hemopoietic cells (Hh system), and the natural killer cell activity against mouse lymphomas (NK system). The effects of six independent variables known to influence resistance to marrow grafts were investigated in the NK system using YAC‐1 lymphoma cells as targets. The following properties were shared: (a) maturation during the fourth week of life; (b) low sensitivity to acute total body irradiation; (c) dependence on the integrity of bone marrow as demonstrated by reduced reactivity in 89 Sr‐treated mice; (d) suppression by a single injection of rabbit anti‐mouse bone marrow serum; (e) suppression by a single injection of the anti‐macrophage agents silica and i‐carrageenan; and (f) suppression by multiple injections of parental spleen cells into F 1 mice. These positive correlations are particularly significant because most of the variables have either opposing or no effect on conventional immunity. F 1 mice rendered specifically unresponsive to parental marrow grafts, could retain NK cell activity, and genetically susceptible mice could be rendered hyporeactive in terms of NK cells, indicating that the specificities of YAC‐1 and Hh‐1 incompatible targets were different. It is extremely unlikely that this remarkable parallelism is fortuitous. These results indicate that either a very similar, or more likely a common mechanism is operative in the two cell‐mediated natural reactivities: effector cells in the NK and Hh systems do not bear B or T lymphocyte markers but are nevertheless endowed with “specificity”. They are dependent for generation in vivo (presumably by maturation or by recruitment) on the interaction with nonlymphoid accessory cells not endowed with specificity, capable of also interacting in vitro with Thy‐1‐positive F 1 hybrid prekiller cells specific for parental targets. Because of thymus independence in vivo and apparent restriction to target cells of the hemopoietic system, these reactivities should be effective in the regulation of hemopoiesis and surveillance over leukemogenesis.