The tyrosine phosphatase activity of the CD45 cell-surface molecule may have a vital role in the function of the T cell antigen receptor.

Sooner or later, most immunologists must deal with the phenomenon of cellular 'activation'. Lymphocytes nor- mally exist in a resting, or quiescent, state. On receipt of some external stimulus, the cells undergo a cascade of 'activation events' culminating in biological function. For T cells, occupancy of the T cell antigen-specific re- ceptor (TCR) by its ligand (the combination of a pep- tide antigen and a major histocompatibility complex- encoded protein) is the principal physiological stimulus. The TCR consists of two functionally distinct portions: a ligand-binding, structurally variable heterodimer, typi- cally composed of the disulfide-linked 0t and [3 chains, and five invariant chains that comprise the CD3 complex (y, 8 and e) and a ~--~ homodimer or a ~-q heterodimer [1]. Occupancy of the TCR, or its cross-linking by spe- cific antibody, triggers the activity of a phospholipase C, resulting in the hydrolysis of phosphatidylinositol 4,5 his- phosphate to diacylglycerol (activating protein kinase C) and inositol trisphosphate (causing an increase in intra- cellular Ca 2 +). In addition, one or more protein tyrosine kinases are activated, which phosphorylate, among other substmtes, the TCR itself (on the ~ chain) (Fig.l). Recent data suggest that phospholipase C is coupled to the TCR via a tyrosine kinase, as two independent tyrosine kinase inhibitors prevent activation-induced increases in inositol phosphates and intracellular Ca 2 + [2,3]. The physiolog- ical roles of these different signal transduction pathways are not totally clear, but either or both are likely to lead to late phenotypic cellular responses, such as the secretion of intedeukin-2 (IL-2). It is generally held that the signalling capacity of the TCR resides completely in the interaction of its own in- tegral components with more distal, membrane-associ- ated or cytosolic, effector molecules. Now, such mod- els must accommodate data that suggest an obligate role for the CD45 molecule in immediate post-receptor TCR function. CD45 (originally designated T200, and also known as B220, Ly-5 and leukoqete-common antigen (L-CA)) was first described as a cell-surface marker present on T- and B-, but not nonlymphoid, cells [4,5], and is now known to have a number of distinctive features [6]. First, CD45 exists in several isoforms with different molecu- lar weights (from 170kD to 220kD or higher) due to alternative splicing of three exons near the 5' end of the gene (the N-terminal and extracellular portion of the molecule), potentially yielding up to eight different amino acid sequences, and extensive N- and O-linked glycosy- lation. Second, different isoforms are found on cells of different lineages. For example, most T cells express the 180 kD species, whereas most B cells express a 220 kD species. Furthermore, various CD45 isoforms are found on cells of the same basic lineage but at different stages (or along different lines) of differentiation or cellular ac- tivation, such as helper T cells versus suppressor T cells or naive T cells versus memory T cells. Third, and per- haps most directly relevant to the topic of signal transduc- tion, CD45 is the most abundant tyrosine phosphatase expressed by T cells. Although earlier studies demonstrated that anti-CD45 antibodies could alter lymphocyte function, the first solid evidence that CD45 normally participates in cellular acti- vation came from studies using variants of an antigen- specific T-cell clone that synthesizes CD45 but does not transport it to the cell surface [7]. As one would ex- pect, the wild-type cell proliferated vigorously when stim- ulated with specific antigen or with immobilized anti-TCR antibodies. The surprising finding was that these stimuli failed to cause the CD45- variants to proliferate. A re- vertant cell that expressed CD45, responded normally to TCR-mediated stimuli. Several factors made the interpre- tation of these results difficult; the relatively poor pro- liferative response of the CD45- cells to the TCR-in- dependent stimulus, IL-2, and the fact that gene recon- stitution studies could not be performed to verify that it was only the lack of cell surface CD45 that led to the functional abnormalities. Nonetheless, these experi- ments established a provocative correlation between the presence of cell surface CD45 and TCR-mediated cellular activation. These observations have been confirmed and extended with CD45- T-cell lymphoma variants. In one study, a CD45- variant of the human T cell leukemic line HPB-ALL was isolated that did not produce CD45 mRNA or CD45 protein (although both could be induced by treatment with a phorbol ester) [8]. The CD45 + parental cells responded to TCR cross-linking with production of inositol phosphates and increases in intracellular Ca 2 +. Although the CD45- variant also generated these re- sponses, they were quantitatively far less than that of the wild type cell. Restoration of low levels of CD45 ex- pression by transfection with CD45 cDNA resulted in en- hancement of the TCR-mediated increases in Ca 2 +. In another study, a CD45- variant was derived from Ju- rkat, another human T cell leukemic line [9]. In this vari- ant, cross-linking the TCR, or an independent cell sur- face activating molecule that is functionally dependent on TCR expression, CD2, failed to induce inositol phos- phate generation, tyrosine kinase activity, or IL-2 secre- tion. In contrast, in the presence of phorbol ester and Volume 1 Number 2 1991 111