Cell death is now a very lively field. The study of active cellular self-destruction through apoptosis (Wyllie et al., 1980) has become very much part of the mainstream in cell biology, particularly in immunology, developmental biology, and oncology. The significance of the process is increasingly recognized both in physiological regulation and in pathological situations (reviewed by Williams et al., 1992). As the subject has been largely neglected for a long time, particularly at the molecular level, there is a lot of catching up to do. There are few, if any, other generally important aspects of cell behavior where there is so much fundamental work still to be done, which makes this a very interesting field to watch. It is now widely accepted that apoptosis is a genedirected process and can be seen, alongside more familiar gene-directed processes like differentiation, as part of the repertoire available to the cell to respond to external and internal stimuli (Figure 1). Until recently, most of the information available on the genetics of programed cell death was derived from studies on the nematode Caenorhabditis elegans (reviewed by Ellis et al., 1991) but now important information about the intracellular molecular signals involved in stimulation and suppression of apoptosis in mammalian cells is also emerging. Intracellular Inducers of Apoptosis The external signals that lead to apoptosis are probably as varied as those that lead to differentiation and proliferation (reviewed by Williams et al., 1992) and can include the withdrawal of extracellular signals as well as their appearance. Indeed, dependenceon essential survival factors for suppression of apoptosis appears to be very widespread (reviewed by Raff, 1992). In hemopoietic stem cells, the primary function of colony-stimulating factors appears to be suppression of apoptosis (Williams et al., 1990) and this effect allows an intrinsically determined pathway of differentiation to be followed (Fairbairn et al., 1993). In contrast, the stimulation of some cell surface molecules, such as the tumor necrosis factor receptor and the APO-l antigen (now known to be identical to the Fas antigen), can often induce cell death by apoptosis, although this is not always the outcome (Mapara et al., 1993). Several different second messenger systems have been associated with induction of apoptosis, and the final response varies with the cell type and the other signals being received. Consequently, in addition to genes exclusively involved in active cell death, there are likely to be some Minireview

The mammalian immune system is essential for surviving challenge infections with a great range of potential pathogens. The protective effect produced is dependent on many different types of cells which require flexible and independent production and regulation. In particular, many important responses are carried out by lymphocytes, which recognise foreign antigen through exquisitely specific receptors: i.e. surface immunoglobulin (sIg) on B lymphocytes and the T cell receptor (TCR) on T lymphocytes. Each lymphocyte displays receptors with a single specificity, allowing cells with particular specificities to be regulated independently. Since millions of different Igs and TCRs are expressed, the precise selection and regulation of each T and B cell population to produce a useful self-tolerant repertoire is a very complex process. Control of cell populations can, in theory, be exercised at a number of levels, including modulation of active cell death by apoptosis. Recent research has demonstrated that regulation of apoptosis is indeed a crucial element in the control of the immune system in general, and in the development of the TCR and Ig repertoires in particular. The molecular analysis of apoptosis now takes a high priority and the proto-oncogene bcl-2 appears to be responsible for specific suppression of apoptosis in several important situations. It is also clear that malfunctions affecting apoptosis, and in particular bcl-2, can result in significant progression towards malignancy.

TNFR1 and TNFR2, the genes encoding the two forms of the human tumor necrosis factor receptor, were localized to normal human chromosomes by in situ hybridization and Southern blot analysis of a series of human x mouse hybrid cell lines. TNFR1 maps to 12p13 and TNFR2 maps to 1p36.

The survival, differentiation, proliferation and development of haemopoietic precursor cells and the functional activity of mature blood cells are all influenced by colony stimulating factors (CSFs). As haemopoietic cells rapidly die in the absence of appropriate CSF, the promotion of cell survival mediated by CSFs, or growth factors, is fundamental to all the other effects exerted by these factors. This enhancement of cell survival is distinct from the stimulation of proliferation. Here we show that the death of haemopoietic precursor cells on withdrawal of the relevant CSF. is due to active cell death, or apoptosis, indicating that CSFs promote cell survival by suppression of the process of apoptosis. The existence of a positive control mechanism regulating precursor cell survival has important implications both for the regulation of normal haemopoiesis and for tumorigenesis.

The high affinity of antibodies produced during responses to T-cell-dependent antigens is associated with somatic mutation in the variable region of the immunoglobulin. Indirect evidence indicates that: (1) this arises by a process of hypermutation, acting selectively on rearranged immunoglobulin variable-region genes, which is activated in centroblasts within germinal centres; and (2) centrocytes, the progeny of centroblasts, undergo selection on the basis of their ability to receive a positive signal from antigen. We have now performed experiments analysing this selection process, and found that, on culture, centrocytes isolated from human tonsil kill themselves within a few hours by apoptosis. This is not a feature of other tonsillar B cells. Centrocytes can be prevented from entering apoptosis if they are activated both through their receptors for antigen and a surface glycoprotein recognized by CD40 antibodies.

The receptors found on most T lymphocytes bind to antigen presented on major histocompatibility complex proteins and consist of dimers of alpha- and beta-polypeptides associated with the invariant CD3 complex. A fully competent immune system requires a diverse array of T-cell antigen receptors (TCRs) with different specificities. This diversity is generated by rearrangement of TCR alpha- and beta-chain gene segments within the thymus where the receptors are first expressed. Any cells carrying self-reactive receptors must be eliminated, suppressed or inactivated so that destructive autoimmunity is avoided. Recently, compelling evidence has shown that one process involved in producing such self-tolerance is clonal deletion of autoreactive cells within the thymus by an as-yet-undefined mechanism. Here we show that engaging the CD3/TCR complex of immature mouse thymocytes with anti-CD3 antibodies produces DNA degradation and cell death through the endogenous pathway of apoptosis. Activation of this process in immature T cells by the binding of the TCR to self-antigens may therefore be the mechanism which produces clonal deletion and consequently self-tolerance.