Current evidence suggests that T-cell receptor (TCR) recognition of antigen bound to the major histocompatibility complex (Ag-MHL),. insufficient to lead to T-cell proliferation or effector function. For a I: elper T cell to produce sufficient interleukin 2 (IL-2) to allow autocrine-driven clonal expansion, there is a requirement for so-called'co-stimulatory'or'accessory'signals in addition to TCR ligation by Ag-MHC. The interaction of the CD28 receptor on T cells with B7 on antigen-presenting cells (APCs) supplies one such co-stimulatory signal. However, the recent discovery that CD28 and B7 are each members of larger gene families suggests that the regulation of co-stimulation i~ more complex than previously imagi; Ted. Here, Carl June and colleagues highlight recent adwmces in the understanding of the CD28 and B 7 receptor families.

A two-signal model of lymphocyte activation has been developed and refined by a number of investigators (reviewed in Ref. 1). The model was originally proposed to address the problem of self-nonself discrimination and, in the case of T cells, is particularly relevant to antigens that are not expressed or encountered in the thymus. The key observation of Lafferty and coworkers in the 1970s was that foreign antigens on tissues of non-hematopoietic origin were tmable to induce a productive cellular immune response in vivo unless viable hematopoietic cells were also provided as a source of'co-stimulatory activity'. In its simplest form, as originally proposed by Bretcher and Cohn, the model describes how an individual antigen-presenting cell (APC) can independently display either the antigen bound to the major histocompatibility complex (Ag-MHC) and/or the co-stimulatory ligand. Given the binary nature of this model, it is possible to predict three distinct outcomes consequent to the interaction between the APC and the T cell (Fig. 1). Activation of the T-cell receptor (TCR) in the presence of co-stimulatory signals results in T-cell clonal expansion and the induction of effector functions such as the production of lymphokines. By contrast, the interaction of T cells with cognate Ag in the absence of costimulatory ligand is not a neutral event but, rather, leads to induced unresponsiveness or to death by apoptosis. Finally, when resting T cells encounter co-stimulatory ligand in the absence of cognate antigen, no effects are predicted. Jenkins, Schwartz 2nd others suggested in the 1980s that the co-stimulatory signal was fundamentally different from the signals delivered by the TCR (Ref. 2). This led to a search for T-cell surface molecules present on resting T cells that had signal-transducing properties distinct from the TCR.