WAS it suicide or murder, and how was the deed done? These are questions not only for the police but for immunologists trying to find out how T-lineage cells stimulated through their antigen receptors (TCR) meet their end. Three papers in this issue1-3-on pages 438, 441 and 444-provide evidence that activation of the cell-surface receptor known as Fas, AP0-1 or CD95 is the weapon involved and that the process, at least as it happens in vitro, is suicidal.

In the thymus, deletion of immature T cells that happen to express self-antigenreactive receptors is a safeguard against autoimmunity (central tolerance). In peripheral lymphoid tissues, the physiological death of TCR-activated, mature T cells probably serves to delete autoreactive T cells with specificity for antigens that are not presented in the thymus (peripheral tolerance), and to terminate an immune response after the battle against a pathogen has been won. Activated T cells are a potential hazard. They produce copious amounts of inflammatory cytokines, and contribute to infection-induced hyperplasia in lymph nodes (potentially a forerunner of lymphoid malignancy); it is therefore safer to remove them when their job has been done and they are no longer needed. Activation-induced death was first observed in T-cell hybridomas4. These cells undergo growth arrest, functional activation-for example secretion of interleukin-2 (IL-2)- and apoptosis upon stimulation with antigens, mitogens or antibodies specific for TCR/CD3, Thy-1 or Ly-6. Signal transduction initiated by all these stimuli requires a functional TCR/CD3 complex and de novo macro-