Physiology of receptor capping induced by erythrocyte surface antigens

The capping of antigen-binding cell receptors by bound sheep erythrocytes (SRC) demonstrates that antigen mounted on a cell surface can generate a signal leading to the capping reaction. SRC-induced capping of ABC is: (a) highly dependent on both aerobic and anaerobic glycolysis, (b) unaffected by agents altering intracellular cyclic nucleotide concentrations, (c) slightly more vigorous in strain A than in CBA mice, (d) inhibited by calcium ionophore, (e) inhibited by the local anesthetic dibucaine and the tranquillizer chlorpromazine, (f) dependent on cytoskeletal activity (i.e., inhibited by the simultaneous presence of colchicine and cytochalasin B), (g) not dependent on the membrane ATPases inhibited by ouabain, (h) not dependent on motility, in that agents which inhibit motility (cytochalasin B alone) or stimulate motility (carbachol) do not alter capping behavior. These properties represent similarities between the capping of surface Ig by the cellular antigens on SRC and by proteins such as anti-Ig. SRC-induced capping is much slower than anti-Ig-induced capping, and only engages 30–40% of ABC, indicating that the nature of the crosslinking agent can influence the kinetics and extent of capping. But SRC cap with the rapid kinetics typical of anti-Ig-induced capping if the surface membrane of the ABC is first cleared of other glycoproteins with trypsin. The removal of negatively charged sialic acid residues by neuraminidase has no such effect. It is probable that the compression of bound SRC into a small area of the membrane requires more energy than does the capping of protein ligands, and that some cells cannot muster enough energy to achieve it.     

Two distinct mechanisms for redistribution of lymphocyte surface macromolecules. II. Contrasting effects of local anesthetics and a calcium ionophore

In the previous study, lymphocyte surface molecules were separated into two subsets depending on whether capping was associated was associated with redistribution of cytoplasmic myosin. In the present study, the effects of the local anesthetic chlorpromazine and of the Ca2+ ionophore {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 were compared. Both drugs affected the surface redistribution of immunoglobulin (Ig), Fc receptors, and the TL antigen- -molecules that appear to cap by association with microfilaments--but had no effect on the Thy.1 (theta) and H2 antigens--molecules that cap slowly, apparently unlinked to microfilament function. The capping of Ig, Fc receptor, and TL was inhibited while that of H2 and theta was not. Both drugs reversed the Ig Fc receptor, and TL caps but not the H2 and theta caps. In the former group, the reversal of caps was accompanied by a parallel reversal of the myosin segregated to the cap area. The appearance of myosin after drug treatment varied: chlorpromazine resulted in a diffuse pattern similar to that of normal lymphocytes, whereas {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187 produced an array of aggregates and coarse filaments. The results are compatible with the view that two mechanisms for capping exist in the lymphocyte. The Ca2+ ionophore may affect capping of microfilament-dependent caps by producing a systemic activation of contractile proteins while chlorpromazine may act by disrupting a Ca2+-dependent link between surface complexes and the contractile proteins.     

A comparison of ligand-induced redistribution of surface immunoglobulins, alloantigens, and concanavalin A receptors on mouse lymphoid cells

Redistribution of surface immunoglobulins (Ig), H-2^b, Thy-1.2 and TL. 1,2,3 alloantigens, and concanavalin A (Con A) receptors on mouse thymus, lymph node and spleen cells into “caps” induced by bivalent antibodies or ligands was compared by immunofluorescence. Surface Ig was capped rapidly following attachment of anti-Ig antibody at 37°. Capping of alloantigens and Con A receptors occurred very slowly following attachment of alloantibody or Con A, but much more rapidly after addition of a secondary bivalent antibody. An inverse relationship between the number of surface component sites per cell and the extent of capping of that component was observed. Capping of alloantigens sparsely represented on the cell surface was not inhibited by high concentrations of alloantibody, in contrast to capping of alloantigens present in greater quantities. These results suggest that factors in addition to molecular cross-linking may be involved in ligand-induced redistribution of cell surface components.     

The Timing and Magnitude of CaSignaling by CD32 Depends on Its Redistribution on the Cell Surface

Ca²⁺signaling was correlated with microaggregation and capping of CD32 molecules on the myeloid cell line, U937. The cytosolic free Ca²⁺signal was related to the extent of CD32 cross-linking and arose asymmetrically within individual cells. Both the magnitude and the delay before Ca²⁺signaling via CD32 on U937 cells was dependent on the extent of CD32 cross-linking. The delay time was extended in cells in which lateral diffusion in the membrane was reduced by covalently cross-linking of surface proteins. Under these conditions, capping but not surface microaggregation of CD32 molecules was prevented. The delay time before Ca²⁺signaling but not the magnitude was also affected. At a higher density of covalent cross-linking of surface proteins, the magnitude of the Ca²⁺signal by CD32 was also reduced and could be completely inhibited. This evidence therefore shows that the formation of a CD32 “cap” was not required for Ca²⁺signaling by this route. However, the signaling delay time was a consequence of lateral diffusion of CD32 molecules in the membrane to form signaling-competent microaggregates, and the redistribution of CD32 molecules on the cell surface was required for Ca²⁺signal generation.     

B cell activation. VIII. Membrane immunoglobulins transduce signals via activation of phosphatidylinositol hydrolysis

Considerable evidence indicates that cross-linking of B cell surface Ig results in a "first signal" in B cell activation. We have shown that transduction of this signal is manifest by changes in plasma membrane potential leading to increased expression of surface I-A antigen. In previous studies, we have provided evidence that suggests that this signal is transduced via phosphatidylinositol (PI) hydrolysis liberating diacylglycerol (DAG), which subsequently activates protein kinase C. These biochemical events are aspects of a transmembrane signal transduction mechanism that is common in nature and utilizes the PI metabolic cycle for generation of "second messenger" diacylglycerol. Here we report direct evidence that treatment of B cells with various antibodies to surface Ig results in activation of the PI cycle. Results suggest that the increased phospholipid metabolism that occurs in B cells in response to anti-Ig involves only those phospholipids in the PI cycle and is a consequence of turnover of existing lipid rather than de novo synthesis. Furthermore, we show that PI cycle activation requires cross-linking of membrane Ig and is inhibitable by increased intracellular cyclic AMP. These findings are particularly important in view of previous studies that have shown identical requirements for and inhibitability of induction of B cell membrane depolarization and increased I-A expression. Thus, these results are consistent with our previous hypothesis that early B cell activation events initiated by receptor Ig occupancy are mediated via PI hydrolysis, diacylglycerol generation, and protein kinase C activation.     

Lymphocyte mechanical response triggered by cross-linking surface receptors

Using a recently developed method (Petersen, N. O., W. B. McConnaughey, and E. L. Elson, 1982, Proc. Natl. Acad. Sci. USA., 79:5327-5331), we have measured changes in the deformability of lymphocytes triggered by cross-linking cell surface proteins. Our study was motivated by two previously demonstrated phenomena: the redistribution ("capping") of cross-linked surface immunoglobulin (sIg) on B lymphocytes and the inhibition of capping and lateral diffusion ("anchorage modulation") of sIg by the tetravalent lectin Concanavalin A (Con A). Both capping and anchorage modulation are initiated by cross-linking cell surface proteins and both require participation of the cytoskeleton. We have shown that the resistance of lymphocytes to deformation strongly increased when sIg or Con A acceptors were cross-linked. We have measured changes in deformability in terms of an empirical "stiffness" parameter, defined as the rate at which the force of cellular compression increases with the extent of compression. For untreated cells the stiffness was approximately 0.15 mdyn/micron; for cells treated with antibodies against sIg or with Con A the stiffness increased to approximately 0.6 or 0.4 mdyn/micron, respectively. The stiffness decreased after completion of the capping of sIg. The increases in stiffness could be reversed to various extents by cytochalasin D and by colchicine. The need for cross-linking was demonstrated by the failure both of monovalent Fab' fragments of the antibodies against sIg and of succinylated Con A (a poor cross-linker) to cause an increase in stiffness. We conclude that capping and anchorage modulation involve changes in the lymphocyte cytoskeleton and possibly other cytoplasmic properties, which increase the cellular viscoelastic resistance to deformation. Similar increases in cell stiffness could be produced by exposing cells to hypertonic medium, azide ions, and to a calcium ionophore in the presence of calcium ions. These results shed new light on the capabilities of the lymphocyte cytoskeleton and its role in capping and anchorage modulation. They also demonstrate that measurements of cellular deformability can characterize changes in cytoskeletal functions initiated by signals originating at the cell surface.     

Translocation of protein kinase C during membrane immunoglobulin-mediated transmembrane signaling in B lymphocytes

Previous studies have implicated a role for protein kinase C (PKC) in transmembrane signal transduction by B cell surface immunoglobulin (Ig). Specifically, the pharmacologic PKC activator phorbol myristate acetate mimics the biologic effects of mIg cross-linking ligands, and cross-linking of membrane Ig (mIg) induces polyphosphoinositide hydrolysis generating diacylglycerol, a potent activator of PKC. Studies described here additionally implicate PKC in mIg-mediated signaling by demonstrating rapid translocation of activatable PKC (PKCa) from cytosol to Triton-soluble membrane fractions after cross-linking of cell surface IgM or IgD. This response, which is also induced by phorbol myristate acetate and lipolysaccharide, is detectable within 1 min of mIg cross-linking and is followed within 4 min by additional translocation of PKCa to a Triton-insoluble particulate compartment. The ability of dbcAMP plus theophylline to inhibit polyphosphoinositide hydrolysis, PKCa translocation, and the B cell's subsequent biological response suggests that these events may be causally related.     

Lymphoma models for B-cell activation and tolerance. VII. Pathways in anti-Ig-mediated growth inhibition and its reversal

WEHI-231, CH33, and CH31 are B-cell lymphomas that are inhibited in their growth by crosslinking of surface Ig receptors during early G1. This "negative signaling" process can be prevented or reversed under certain conditions. In the present paper, we use a cell synchronization procedure to demonstrate that activation of protein kinase C (PKC) is not involved in the negative signal for growth, but rather that PKC activation prevents growth inhibition when present early in the cell cycle. Indeed, the prevention of negative signaling is only accomplished by active phorbol esters. Moreover, phorbol esters and a calcium ionophore fail to deliver a negative signal under conditions in which anti-Ig can significantly prevent cell cycle progression into S phase, thereby ruling out synergy between PKC and calcium in growth inhibition. Whether phorbol esters reverse negative signaling by preventing internalization of the immune complex or phosphorylation of a critical intracellular protein is discussed.     

Calcium ionophore enhances the electron spin resonance signal from isolated skeletal muscle

Electron spin resonance studies of the free radical signal from isolated skeletal muscle during experimental damage have shown that elevation of muscle intracellular calcium with the calcium ionophore A23187 induced an average 61% increase in the amplitude of the signal of g value 2.0047 compared to paired, untreated control muscles, accompanied by a large efflux of intracellular creatine kinase to the external medium. Inhibitors of the calcium-induced loss of cell viability leading to enzyme efflux, i.e., chlorpromazine, phenidone and nordihydroguaiaretic acid had variable effects on the signal, suggesting that the free radical signal obtained from skeletal muscle by electron spin resonance techniques is stimulated by intracellular calcium overload, but is not directly related to the mechanisms by which calcium overload leads to a loss of cell viability leading to intracellular enzyme efflux.     

Concanavalin A receptors, immunoglobulins, and theta antigen of the lymphocyte surface. Interactions with concanavalin A and with Cytoplasmic structures

The effect of concanavalin A (Con A) on the capping of mouse lymphocyte surface immunoglobulin (surface Ig), cross-linked by rabbit anti-mouse Ig antibody, and on the capping of mouse thymocyte theta antigen, cross- linked by anti-theta alloantibody and rabbit anti-mouse Ig antibody, has been studied by immunofluorescence, using fluorescein conjugated Con A and rhodamine-conjugated anti-mouse Ig antibody, and by electron microscopy, using native or fluorescein-conjugated Con A and ferritin- conjugated anti-mouse Ig antibody. Prior incubation of the cells with Con A inhibited only partially capping os surface Ig, whereas it blocked almost completely capping of theta antigens. Both on cells with rings and on cells with caps the staining for surface Ig or theta antigen was superimposed to the staining for Con A. When Con A receptors on spleen cells were capped by Con A at concentrations of 10 mug/ml or higher, and the distribution of surface Ig was examined under noncapping conditions, all detectable surface Ig were found in the caps. As shown by electron microscopy, surface Ig remained dispersed in a layer of Con A. The ability of Con A to cap surface Ig was not altered by the presence of cohchicine or vinblastine. These results suggest that surface Ig are cross-linked by Con A to other Con A receptors. In these conditions surface Ig behave essentially as Con A receptors, as for example, in their sensitivity to cytochalasin B during inhibition or reversal of capping induced by this drug. The behavior of surface Ig parallels that of Con A receptors also in the presence of vinblastine. It is concluded that in the presence of Con A, antimitotic drugs do not modify directly the interaction between Con A receptors and surface Ig, but probably influence the capping ability of the Con A receptors or, more in general, affect the ability to elicit movements over the cell surface. The role in capping of cytochalasin- sensitive and vinblastine-sensitive structures is discussed. Both types of structures appear to play an active role in the formation of a cap, although the former probably corresponds to the main mechanical system responsible for the active displacement of cytoplasmic and surface material.     

Biogenesis, transit, and functional properties of the insulin proreceptor and modified insulin receptors in 3T3-L1 adipocytes. Use of monensin to probe proreceptor cleavage and generate altered receptor subunits

The biogenesis, intracellular transport, and functional properties of the insulin proreceptor and modified insulin receptors were studied in hormone-responsive 3T3-L1 adipocytes. After control cells were labeled with [35S]Met for 7 min, the principal polypeptide that was precipitated by anti-insulin receptor antibodies had a molecular weight (Mr) of 180 000. This initial precursor was rapidly converted (t1/2 = 35 min) to a 200-kilodalton (kDa) polypeptide, designated the insulin proreceptor, by the apparent posttranslational addition of N-linked, high mannose core oligosaccharide units. Mature alpha (Mr 130 000) and beta (Mr 90 000) subunits were derived from sequences within the proreceptor by proteolytic cleavage and late processing steps, and these subunits appeared on the cell surface 2-3 h after synthesis of the 180-kDa precursor. The cation ionophore monensin was used in combination with metabolic labeling, affinity cross-linking, and external proteolysis to probe aspects of proreceptor function, transit, and the development of insulin sensitivity at the target cell surface. At 5 micrograms/mL, monensin potently inhibited the proteolytic cleavage step, and the 200-kDa polypeptide accumulated. Lower concentrations of the ionophore selectively blocked late processing steps in 3T3-L1 adipocytes so that apparently smaller alpha' (Mr 120 000) and beta' (Mr 85 000) subunits were produced. Proreceptor and alpha' and beta' subunits were translocated to the cell surface, indicating that the signal for intracellular transit occurs in the 200-kDa polypeptide and is independent of the posttranslational proteolysis and late processing steps. The alpha' subunit bound insulin both at the surface of intact cells and after solubilization with Triton X-100; the beta' subunit was phosphorylated in an insulin-stimulated manner. The detergent-solubilized 200-kDa proreceptor also exhibited both functional properties. However, the proreceptor that was transported to and exposed on the cell surface was incapable of binding insulin in intact adipocytes. Thus, late processing is not essential for the expression of functions associated with mature alpha and beta subunits. In contrast, it appears that the proteolytic generation of subunits is required for the correct orientation of the hormone binding site in the plasma membrane bilayer and the development of insulin responsiveness in 3T3-L1 adipocytes.     

Pathogenesis of passive Heymann glomerulonephritis: chlorpromazine inhibits antibody-mediated redistribution of cell surface antigens and prevents development of the disease

Two hypotheses were tested: first, that in LEW rats the interaction of sheep (or rabbit) anti-brush border antibodies with antigens (Heymann antigens) expressed on the plasma membrane of glomerular visceral epithelial cells is characterized by initial redistribution of immune complexes on the cell surface and by subsequent shedding of immune complexes in the subepithelial part of the capillary wall; and secondly, that this interaction is inhibited by chlorpromazine, a drug that displaces calcium ions from binding sites linking the plasma membrane to the cytoskeleton, and which blocks the redistribution of IgG on the surface of B lymphocytes exposed to anti-IgG antibodies. The studies were performed in vitro on cultured LEW glomerular epithelial cells and in vivo in LEW rats. In cultured glomerular epithelial cells exposed at 37 degrees C to anti-brush border IgG, chlorpromazine prevented, in a dose-dependent manner, the redistribution ("capping") of Heymann antigens and the fixation of complement. The renal glomeruli of chlorpromazine-treated LEW rats examined 6 and 48 hr after transfer of anti-brush border antibodies had punctate and, later, punctate and diffuse deposits of sheep (or rabbit) IgG on glomerular epithelial cells, but not similar deposits of rat C3. Moreover, granular subepithelial deposits of sheep (or rabbit) IgG and rat C3, characteristic of passive Heymann glomerulonephritis, did not develop, although deposits of sheep IgG were detected by immunoelectron microscopy on the microvilli of glomerular epithelial cells. Comparative studies on rats with similar reductions in glomerular filtration rates, produced by high doses of chlorpromazine or with renal artery stenosis, showed that the findings were not the consequence of insufficient delivery of antibody to glomerular epithelial cells. The results are consistent with the interpretation that Heymann glomerulonephritis is induced by mechanisms of redistribution of cell surface antigens comparable to those that govern the interaction of surface antigens (or receptors) with appropriate ligands in B lymphocytes and other classical in vitro systems.     

Effects of linoleic acid on capping,lectin mediated mitogenesis,surface antigen expression,and fluorescent polarization in lymphocytes and BHK cells

The incubation of linoleic acid with cells causes profound effects on membrane associated phenomenon. Using the fluorescent probe diphenyl hexatriene (DPH) to monitor lipid changes in the microenvironment of the cell surface, we find that linoleic acid reduces the polarization values (P) in mouse lymphocytes and BHK cells. Measurements on lipids extracted from the cells grown in linoleic acid produce similar results. We also find in the mouse lymphocyte that capping of Ig is inhibited and con A stimulated mitogenesis is unaffected. In contrast to the latter effect, LPS and PHA stimulated mitogenesis is inhibited and in the rat lymph node, con A stimulated mitogenesis, greatly enhanced. We also show that linoleic acid alters the binding of antibodies to the cell surface of EL-4 lymphoma cells. These observations suggest that linoleic acid alters cellular function by interfering with protein/lipid interactions within the surface membrane.     

Release of leukotrienes C4 and B4 and prostaglandin E2 from human monocytes stimulated with aggregated IgG, IgA, and IgE

Purified human peripheral blood monocytes were stimulated with aggregated human myeloma proteins of different classes or the calcium ionophore A23187 and the release of leukotrienes C4 and B4 (LTC4, LTB4), and prostaglandin E2 (PGE2) into the supernatant was determined. The ionophore induced release of 10 +/- 5 ng LTC4/10(6) cells and 25 +/- 8 ng LTB4/10(6) cells. Aggregated IgG, IgA, and IgE, but not IgM or monomeric immunoglobulins (Ig), induced release of LTC4 and LTB4 that was approximately 10 to 20% of that induced by ionophore. In addition, IgG, IgA, and IgE, but not IgM, induced release of PGE2 (range 0.015 to 0.22 ng/10(6) cells). Aggregated Ig induced LTC4, LTB4, and PGE2 release in a dose-dependent manner; maximal leukotriene (LT) release was observed by 30 min, in contrast to PG release, which continued to increase up to 2.5 hr. Both ionophore- and Ig-induced LTC4 and LTB4 release were completely inhibited by removal of calcium from the media and by preincubation of cells with nordihydroguaiaretic acid. Indomethacin inhibited Ig-induced PGE2 release by 80%. Phagocytosis of the Ig aggregates was not required for LT or PGE2 release, since release was not inhibited by cytochalasin B. Release of LTC4, LTB4, and PGE2 induced by IgG, IgA, and IgE, but not IgM, correlated with the presence or absence of monocyte Fc receptors (FcR) as determined by rosette assays. The data suggest that IgG, IgA, and IgE immune complexes mostly likely induce monocyte arachidonic acid metabolism via cross-linking of FcR. The ability of monocytes to release eicosanoids in the absence of phagocytosis suggests that interaction of monocytes with immobilized immune complexes, such as those deposited in blood vessel walls or glomerular basement membranes, could initiate metabolism of arachidonic acid by monocytes. Such a mechanism could contribute to inflammatory reactions characterized by mononuclear cell infiltrates.     

Regulation of adhesion of mouse bone marrow-derived mast cells to laminin

We have reported that mast cells adhere to laminin after activation with PMA. In this study, we demonstrate that the cross-linking of cell surface high-affinity IgE-R on mast cells derived from mouse bone marrow cultured for 3 wk in the presence of WEHI-3-conditioned media acts as a highly sensitive physiologic stimulus for this attachment and that receptor activation is also induced by calcium ionophore A23187. Adherence occurred at threefold log concentrations less of A23187 and Ag than required for histamine release in a selective subpopulation comprising 20 to 30% of the total cells. At higher concentrations of agonist that permitted histamine release, the time course for degranulation was shown to be more rapid than that of adherence. Adherence was inhibited by antibodies to laminin and laminin receptor and was calcium ion and temperature dependent. Treatment of cells with dibutyryl cAMP, which activates protein kinase A, inhibited both adherence and histamine release induced by Ag or calcium ionophore. Treatment of cells with staurosporin, which inhibits protein kinase C, also inhibited adherence and histamine release induced by calcium ionophore, but was not significantly active against either adherence or histamine release induced by Ag. It thus appears that agents which modulate intracellular signaling mechanisms are equally effective toward histamine release and adherence, suggesting these two events are intimately linked in stimulus secretion coupling. Specific cytokines stimulating mast cell adhesion to laminin could not be found; however, culture of mast cells with TGF-beta 1 was determined to enhance IgE-mediated adherence to laminin. Hence, the high-affinity IgE-R on the mast cell functions not only in exocytosis but also facilitates the process of mast cell adherence to laminin.     

The role of tyrosine phosphorylation in signal transduction through surface Ig in human B cells. Inhibition of tyrosine phosphorylation prevents intracellular calcium release.

Cross-linking surface Ig on human B cells, or the TCR complex on T cells leads to the rapid appearance of newly tyrosine phosphorylated proteins. This is associated with inositol phospholipid turnover and a rise in intracellular calcium. Incubation of human B or T lymphocytes with the tyrosine kinase inhibitors, herbimycin and genistein, inhibits new tyrosine phosphorylation after receptor-linked activation. This is associated with complete abrogation of the increase in intracellular calcium in these lymphocytes and inhibition of inositol phospholipid turnover. Herbimycin- and genistein-treated lymphocytes are nevertheless still capable of responding to aluminum fluoride with a rise in intracellular calcium. These data support the contention that a B cell-associated protein tyrosine kinase regulates signal transduction via phospholipase C. CD45, the membrane associated protein tyrosine phosphatase, and PMA that activates protein kinase C, both inhibit the calcium response in B lymphocytes induced by receptor cross-linking. PMA and cross-linking CD45 both induced the appearance of tyrosine phosphorylated proteins in human B cells, although the pattern is quite distinct from that seen when surface lg is cross-linked. However, the induction of new tyrosine phosphorylation by anti-mu does not appear to be affected by these reagents. Although this may reflect an insensitivity of the tyrosine phosphorylation assay, it could indicate that regulation of the calcium response and regulation of the tyrosine kinase can be independent processes.     

Tegumental Ca-stimulated adenosine triphosphatase activity in adult Schistosoma mansoni worms

A Ca-stimulated ATPase activity (pH 9.5) associated with the tegumental membrane enriched (TME) fraction of Schistosoma mansoni adults was partially inhibited by NAP-taurine or by increasing concentrations of chlorpromazine; endogenous calmodulin was found associated with the TME fraction. A similar activity (pH 8.6) was histochemically visualized within the tegument of fixed worms on the cytoplasmic leaflet of both the double surface membrane and the basement membrane; this reaction was inhibited by 1 microM chlorpromazine and it was also observed on the inner side of double membrane vesicles present in the TME fraction. No ATPase activity could be seen at alkaline pH with added Mg or Na/K ions. Without ATP, the addition of external Ca to the fixed worms induced the appearance of lead precipitates on the tegumental discoid bodies; this reaction was inhibited by molybdate and not by chlorpromazine. The intrategumentary regulation of calcium by the systems described and the possible use of phenothiazines against schistosomes are discussed.     

Role of phosphatidylinositol-anchored proteins in T cell activation

A novel class of cell surface proteins are attached to the plasma membrane via a phosphatidylinositol (PI)-glycan anchoring structure, and these proteins can be selectively removed from the cell surface by the enzyme PI-specific phospholipase C (PI-PLC). Enzyme treatment led to a prolonged reduction in cell surface expression of several PI-anchored proteins. Activation of T cells led to a marked decrease in the ability of PI-PLC to remove PI-anchored surface proteins from the activated T cells. This decrease in PI-PLC sensitivity may reflect an alteration in the PI-glycan anchoring structures, or in a general membrane property, which renders the PI-anchored proteins inaccessible to the enzyme. When murine T lymphocytes were treated with PI-PLC and then stimulated with either Con A, the calcium ionophore A23187 and PMA, or an anti-CD3 mAb, the response to Con A stimulation was inhibited by 90%, whereas the responses to ionophore and PMA or anti-CD3 were not affected. Removal of PI-anchored proteins inhibited an early event in the activation process in response to Con A because both IL-2 production and IL-2R expression were inhibited by the PI-PLC treatment. Inhibition of the Con A response was secondary to removal of a PI-linked protein from the responder T cell population because PI-PLC treatment of T-depleted spleen cells did not alter their ability to act as a source of accessory cells. It is unlikely that removal of the known PI-linked proteins on murine T cells, Thy-1 and Ly-6, can fully account for the inhibition of Con A response because the cell line M2B3, that lacks these surface proteins, responded normally to Con A stimulation. These studies demonstrate that one or more PI-anchored T cell proteins play an important role in an early step of Con A activation, perhaps involving T cell-accessory cell interactions. In contrast, the ability to stimulate T cells by direct cross-linking of TCR/CD3 complex is not dependent on the presence of these PI-anchored proteins.