Regulation of αIIbβ3 Function in Human B Lymphocytes*

We studied the function of the platelet integrin αIIbβ3 using a B lymphocyte model in which αIIbβ3 can be induced to interact with fibrinogen using phorbol myristate acetate (PMA). To determine whether a G protein-coupled receptor could also activate αIIbβ3 in lymphocytes, we coexpressed the human formyl peptide receptor (fPR) and αIIbβ3, finding that the fPR agonist formyl Met-Leu-Phe (fMLP)-stimulated lymphocyte adherence to immobilized fibrinogen and binding of soluble fibrinogen to the lymphocyte surface. The response to fMLP, but not PMA, was abrogated by pertussis toxin, indicating that the fPR was coupled to the G-protein Gαi, whereas the protein kinase C inhibitor bisindolylmaleimide I inhibited the response to both fMLP and PMA, indicating that signaling from the fPR included protein kinase C. On the other hand, the tyrosine kinase inhibitor genistein, the Syk inhibitor piceatannol, and the RhoA inhibitor C3 exoenzyme had no effect, implying that neither tyrosine phosphorylation nor the GTPase RhoA were involved. Furthermore, whereas micromolar concentrations of cytochalasin D inhibited the PMA-stimulated interaction of αIIbβ3 with fibrinogen, nanomolar concentrations actually induced fibrinogen binding to unstimulated cells. Our studies demonstrate that αIIbβ3 expressed in B lymphocytes can be activated by a physiologic agonist and outline an activating pathway that includes Gαi, protein kinase C, and the actin cytoskeleton.

Exposure of a binding site for ligands such as fibrinogen and von Willebrand factor on the platelet integrin ␣IIb␤3 is a prerequisite for platelet aggregation (1). The process by which this binding site is exposed has been termed "inside-out" signaling and is initiated when agonist-generated intraplatelet signals induce a conformational change in ␣IIb␤3 by interacting with its cytoplasmic tails (2). To study the process of ␣IIb␤3 activation in vitro, we have developed a model system in which wild-type and mutant ␣IIb␤3 expressed in GM1500 cells, an Epstein-Barr virus-transformed human B lymphocyte line, can be induced to interact with soluble and immobilized fibrinogen by the phorbol ester phorbol 12-myristate 13-acetate (PMA) 1 (3,4). Using this system, we observed that the cytoplasmic tail of ␣IIb is not required for ␣IIb␤3 function in lymphocytes, that the conserved GFFKR motif in the ␣IIb tail is required for ␣IIb to interact with ␤3, and that signals interacting with the ␤3 cytoplasmic tail are responsible for the ability of agonists to stimulate ␣IIb␤3 function (4).
Phorbol esters such as PMA activate the conventional and novel isoforms of protein kinase C (PKC) (5). While phorbol esters are a potent stimulus for ligand binding to ␣IIb␤3 in platelets (6), they bypass the more proximal signaling events that are initiated when agonists bind to their cognate platelet membrane receptors. Most of the known platelet receptors for agonists are seven-transmembrane domain proteins that are coupled to various G proteins (7). Stimulation of these receptors on platelets is known to activate phospholipase C␤, generate diacylglycerol and inositol triphosphate, increase cytosolic calcium, and activate several isoforms of PKC (7). We were interested in determining whether stimulation of a G proteincoupled receptor on B lymphocytes would also expose the ligand-binding domain of ␣IIb␤3. Honda et al. (8) have reported that stimulation of the human N-formyl peptide chemoattractant receptor (fPR) in murine B lymphocytes induces the ␣4␤1mediated adherence of these cells to VCAM-1. The fPR is a seven-transmembrane domain protein coupled to the G protein G␣ i in leukocytes (9). Accordingly, we coexpressed the human fPR and ␣IIb␤3 in GM1500 cells and tested the ability of the transfected cells to interact with immobilized and soluble fibrinogen. We found that like PMA, the chemoattractant peptide formyl Met-Leu-Phe (fMLP), an fPR agonist, increased the avidity of ␣IIb␤3 for immobilized fibrinogen and its affinity for soluble fibrinogen. Moreover, using a number of metabolic inhibitors, we outlined an ␣IIb␤3 activation pathway involving the G-protein G i , PKC, and the actin cytoskeleton.

EXPERIMENTAL PROCEDURES
Reagents and Materials-The anti-FLAG monoclonal antibody (mAb) M1 and the fluorescent Ca 2ϩ indicator fluo-3 AM were obtained from IBI-Kodak and Molecular Probes, respectively. Immulon 2 flat bottom microtiter plates were purchased from Dynatech Laboratories. PMA, fMLP, pertussis toxin (PTX), bovine serum albumin, ionomycin, Arg-Gly-Asp-Ser (RGDS), and cytochalasin D were purchased from Sigma. Human fibrinogen was obtained from Enzyme Research Labs. Bisindolylmaleimide I (BIM I), bisindolylmaleimide V (BIM V), and genistein were purchased from Calbiochem. Recombinant C3 exoenzyme was purchased from Upstate Biotechnology. Lipofectin reagent and Opti-MEM media were obtained from Life Technologies, Inc. G418 was purchased from Mediatech. Piceatannol and hygromycin were obtained from Boehringer Mannheim.
Coexpression of the Human fPR and ␣IIb␤3 in Human B Lymphocytes-Because GM1500 cells express ␤3, introduction of a cDNA for ␣IIb results in the expression of ␣IIb␤3 on the cell surface (4). Therefore, to coexpress the fPR and ␣IIb␤3 in GM1500 cells, a cDNA for ␣IIb in the plasmid pREP4 containing a gene for resistance to the antibiotic hygromycin (3) and a FLAG octapeptide-tagged cDNA for the human fPR in the plasmid pRc/CMV containing a gene for resistance to the antibiotic neomycin (a gift of Drs. James J. Campbell and Eugene Butcher, Stanford University) (10) were sequentially introduced into 7.5 ϫ 10 6 cells by electroporation (250 V and 960 microfarads). Stable co-transfectants were selected by growth in RPMI media containing 20% fetal calf serum and both G418 (750 g/ml) and hygromycin (200 g/ml). The simultaneous presence of ␣IIb␤3 and the fPR on the lymphocyte surface was confirmed by flow cytometry after staining the cells with either the ␣IIb␤3-specific mAb A2A9 (11) or the anti-FLAG mAb M1, followed by staining with fluorescein (FITC)-conjugated goat antimurine IgG. Flow cytometry was performed using a FACScan flow cytometer (Becton-Dickinson) as described previously (12).
To ensure that the expressed fPR was functional, 5 ϫ 10 7 cells were loaded with 5 M fluo-3 AM at room temperature for 30 min. The cells were then incubated with either 30 nM ionomycin or 100 nM fMLP for 30 s and Ca 2ϩ flux-induced fluorescence was measured with a FACScan flow cytometer as described previously (8).
Measurement of ␣IIb␤3 Function in Human B Lymphocytes-The ability of ␣IIb␤3 expressed by lymphocytes to interact with fibrinogen was tested by measuring agonist-stimulated lymphocyte adherence to immobilized fibrinogen (3) and agonist-stimulated binding of soluble FITC-fibrinogen using flow cytometry as described previously (4).
To measure lymphocyte adherence to fibrinogen, the wells of microtiter plates were coated with 10 g/ml purified human fibrinogen in 50 mM NaHCO 3 buffer, pH 8.0, containing 150 mM NaCl. Unoccupied protein-binding sites on the wells were blocked with 5 mg/ml bovine serum albumin dissolved in the same buffer. 1.5 ϫ 10 5 B lymphocytes, metabolically labeled overnight with [ 35 S]methionine, were suspended in 100 l of 50 mM Tris-HCl buffer, pH 7.4, containing 150 mM NaCl, 0.5 mM CaCl 2 , 0.1% glucose, and 1% bovine serum albumin, stimulated with either PMA or fMLP, and added to the protein-coated wells. Following an incubation for 30 min at 37°C without agitation, the plates were vigorously washed four times with the suspension buffer and adherent cells were dissolved using 2% SDS. The SDS solutions were counted for 35 S in a liquid scintillation counter.
To measure the binding of soluble FITC-fibrinogen to agonist-stimulated lymphocytes, purified human fibrinogen (13) was labeled with FITC using a Calbiochem-FITC Labeling Kit as described by the manufacturer. Fibrinogen labeled with FITC in this manner remained monomeric as assessed by gel-filtration chromatography, supported platelet aggregation as well as unlabeled fibrinogen, and was 95% clottable with thrombin (13). 1.5 ϫ 10 5 B lymphocytes were then suspended in 100 l of 10 mM sodium phosphate buffer, pH 7.4, containing 137 mM NaCl, 1 mM CaCl 2 , and 1% bovine serum albumin (suspension buffer) and incubated with 0.25 M FITC-fibrinogen in the presence or absence of PMA or fMLP for 30 min at room temperature. The cells were washed once with suspension buffer and resuspended in a fixation solution consisting of 10 mM sodium phosphate buffer, pH 7.4, containing 137 mM NaCl and 0.37% formalin. Following a 10-min incubation on ice, the cells were again washed once with the suspension buffer, and analyzed by flow cytometry as described previously (4).
Effect of Botulinum C3 Exoenzyme on ␣IIb␤3 Function in GM1500 Cells-Recombinant C3 exoenzyme from Clostridium botulinum was introduced in GM1500 cells using Lipofectin Reagent. 1 ϫ 10 7 GM1500 cells were suspended in 1 ml of Opti-MEM media containing 40 g of Lipofectin and 12 g of C3 exoenzyme and incubated for 2 h at 37°C. The cell suspension was then divided into two aliquots. One aliquot was resuspended in complete media for 1 h and the ability of these cells to adhere to fibrinogen was tested as described above. ADP-ribosylation of the small GTPase RhoA during the incubation was examined using the second aliquot as described previously (14). Briefly, the cells were washed with Opti-MEM, resuspended in 50 l of 20 mM Tris-HCl buffer, pH 7.5, containing 0.25 M sucrose, 5 mM MgCl 2 , 1 mM EDTA, 1 mM dithiothreitol, 2 mM benzamidine, and 0.5 mM phenylmethylsulfonyl fluoride, and homogenized by sonication. Following centrifugation of the homogenate at 1000 ϫ g for 5 min, 15 l of the supernatant were incubated with 50 ng of C3 exoenzyme and 10 M [ 32 P]NAD ϩ (Amersham) for 1 h at 30°C in 100 mM Tris-HCl buffer, pH 8.0, containing 20 mM nicotinamide, 10 mM thymidine, 10 mM dithiothreitol, and 5 mM MgCl 2 in a total reaction volume of 100 l. Sufficient SDS and dithiothreitol were added to make their final concentrations 3% and 200 mM, respectively, and the solution was heated at 100°C for 3 min. Following 0.1% SDS-10% polyacrylamide gel electrophoresis, ADP-ribosylated RhoA was visualized by autofluorography.

Introduction of a Functional fPR into GM1500 Cells-Previ-
ously, we demonstrated that the phorbol ester PMA induces the adherence of B lymphocytes expressing ␣IIb␤3 to immobilized fibrinogen (3). To determine whether adherence could also be induced by stimulating a receptor on the lymphocyte surface, we introduced plasmids containing a cDNA for the human fPR tagged at its 5Ј end with the FLAG epitope and a cDNA for ␣IIb into GM1500 B lymphocytes and selected for cells that stably expressed both proteins using the antibiotics G418 and hygromycin. As shown in Fig. 1A, the simultaneous presence of ␣IIb␤3 and the fPR on the lymphocyte surface was confirmed by flow cytometry after staining the cells with the ␣IIb␤3specific mAb A2A9 (11) and the anti-FLAG mAb M1. The cells also stained with a FITC-labeled derivative of fMLP (data not shown).
Measurement of Ca 2ϩ Flux in GM1500 Cells Expressing the fPR-fMLP induces a Ca 2ϩ flux in murine B lymphocytes expressing the human fPR (8). To verify that the fPR we expressed was functional, transfected GM1500 cells were loaded with the fluorescent Ca 2ϩ indicator fluo-3 AM and exposed to either 30 nM ionomycin or 100 nM fMLP. The resulting change in fluo-3 fluorescence was then measured by flow cytometry. As shown in Fig. 1B, ionomycin and fMLP induced comparable changes in fluo-3 fluorescence, confirming that fPR stimulation could induce a flux of Ca 2ϩ in the transfected cells (Fig. 1B). Moreover, the fMLP-stimulated Ca 2ϩ flux was inhibited by preincubating the cells for 2 h with 100 ng/ml PTX, indicating that the fPR in GM1500 cells is coupled to a PTX-sensitive G protein.
Comparison of PMA and fMLP-stimulated ␣IIb␤3 Function-Next, we compared the ability of fMLP and PMA to stimulate the adherence of lymphocytes expressing ␣IIb␤3 to immobilized fibrinogen. As shown in Fig. 2A, there was a concentration-dependent increase in lymphocyte adherence following exposure of the cells to fMLP. In multiple experiments, maximal adherence was observed at 300 nM fMLP, but was never more than one-half to two-thirds that induced by 200 ng/ml PMA. Moreover, no additive effect was seen when cells were stimulated simultaneously with fMLP and PMA, suggesting that these agonists were acting through the same signaling pathway.
The adherence of PMA-stimulated lymphocytes to immobilized fibrinogen is inhibited by soluble fibrinogen (4). As shown in Fig. 2B, the adherence of fMLP-stimulated cells also decreased as the concentration of soluble fibrinogen in the suspending buffer increased. Moreover, like cells stimulated by PMA (4), inhibition was maximal at a concentration of soluble fibrinogen of 40 -50 M.
Platelet stimulation induces a change in the affinity of ␣IIb␤3 for ligands and enables it to bind soluble fibrinogen (13). Previously, we found that PMA stimulation enabled transfected lymphocytes expressing ␣IIb␤3 to bind soluble fibrinogen (4). To determine if fPR stimulation would do likewise, we stimulated lymphocytes coexpressing ␣IIb␤3 and the fPR with either 200 ng/ml PMA or 300 nM fMLP and used flow cytometry to compare the binding of FITC-fibrinogen. As we had seen previously (4), the fluorescence histogram of PMA-stimulated lymphocytes was shifted substantially to the right of the histogram of unstimulated cells, indicating that fibrinogen was bound to the stimulated cells (Fig. 3A). The fluorescence histogram of fMLP-stimulated lymphocytes was also shifted to the right, although again fMLP was a less potent agonist than PMA (Fig. 3B). Preincubating both sets of stimulated cells with the mAb A2A9 prevented the shifts in fluorescence, confirming that the FITC-fibrinogen was bound to ␣IIb␤3. Moreover, like the binding of fibrinogen to stimulated platelets (13,15), fibrinogen binding to PMA and fMLP-stimulated lymphocytes reached saturation at a fibrinogen concentration of Ϸ200 g/ml and was abolished by either a 15-fold excess of unlabeled fibrinogen or 200 M RGDS (data not shown).
Effect of Signaling Pathway Inhibitors on ␣IIb␤3 Function in Lymphocytes-Because PTX inhibits fMLP-stimulated Ca 2ϩ flux in GM1500 cells expressing the fPR, we asked whether PTX would inhibit fMLP-stimulated ␣IIb␤3 function in these cells. As expected, preincubating the cells with PTX had no effect on PMA-stimulated adherence to fibrinogen (Fig. 4). However, PTX reduced fMLP-stimulated adherence to nearly baseline levels. On the other hand, no inhibition was observed when the cells were stimulated with both fMLP and PMA, indicating that PMA was able to bypass the PTX effect. Identical results were seen when FITC-fibrinogen binding, rather than cell adherence, was used as the indicator of ␣IIb␤3 function (data not shown).
The fMLP-and interleukin 8-stimulated adherence of murine B cells mediated by ␣4␤1 is unaffected by inhibiting PKC (16). To determine the effect of PKC inhibitors on fMLP-stimulated ␣IIb␤3 function in human B cells, we incubated transfected GM1500 cells overnight with nanomolar concentrations of either the high affinity PKC inhibitor BIM I (17) or the low affinity inhibitor BIM V (18) and measured agonist-stimulated lymphocyte adherence to fibrinogen. As expected, BIM I reduced PMA-stimulated adherence to baseline levels, whereas the same concentrations of BIM V had no effect (Fig. 5). However, to our surprise, BIM I, but not BIM V, also reduced fMLP-stimulated adherence to baseline levels. Identical results were seen when FITC-fibrinogen binding was measured instead of adherence (data not shown). Thus, these experiments suggest that fPR stimulation regulates ␣IIb␤3 function in human B cells via a signaling pathway that includes PKC.
Stimulation of integrin function is associated with the activation of a number of protein tyrosine kinases (19). To determine whether tyrosine phosphorylation regulates ␣IIb␤3 function in lymphocytes, we incubated transfected GM1500 cells overnight with micromolar concentrations of the tyrosine ki-nase inhibitor genistein (20) and measured agonist-stimulated lymphocyte adherence to fibrinogen. As shown in Fig. 6, genistein concentrations as high as 150 M had no effect on either PMA-or fMLP-stimulated lymphocyte adherence. A small degree of inhibition (29 and 16%, respectively) was observed at a genistein concentration of 300 M, a concentration at which genistein also affects the activity of serine/threonine kinases (20). Activation of the non-receptor tyrosine kinase Syk is an early event after platelet stimulation by agonists such as collagen and thrombin (21, 22), although in Epstein-Barr virustransformed lymphocytes, Syk activity is constitutively inhibited by the Epstein-Barr virus-encoded protein LMP2 (23). Nevertheless, to determine whether residual Syk activity in GM1500 cells could regulate ␣IIb␤3 function, we incubated transfected cells with the Syk inhibitor piceatannol (24) and measured both PMA and fMLP-stimulated lymphocyte adherence to fibrinogen. As shown in Fig. 6, 30 g/ml piceatannol, a concentration that completely inhibited collagen-induced platelet aggregation, had no effect on ␣IIb␤3 function in cells stimulated by either PMA or fMLP.
Role of the Actin Cytoskeleton in Regulating ␣IIb␤3 Function in Lymphocytes-Inhibitors of actin polymerization impair ␤1 and ␤2 integrin function in leukocytes, suggesting that the actin cytoskeleton regulates integrin function in these cells (14,16,25). Whether the actin cytoskeleton regulates ␣IIb␤3 function in platelets is less certain, although it was recently reported that the actin polymerization inhibitor cytochalasin E inhibits thrombin-stimulated ␣IIb␤3 function (26). To determine if the actin cytoskeleton regulates ␣IIb␤3 function in GM1500 cells, we measured the effect of cytochalasin D on unstimulated and PMA-stimulated lymphocyte adherence to fibrinogen. We found that increasing concentrations of cytocha- ␣IIb␤3 Function in B Lymphocytes lasin D inhibited PMA-stimulated adherence, with few adherent cells remaining at a cytochalasin D concentration of 10 M (Fig. 7). By contrast, submicromolar concentrations unexpectedly, but consistently, increased unstimulated adherence, such that adherence in the presence 0.01 M cytochalasin D was 2-3-fold greater than in its absence. We also examined the effect of cytochalasin D on FITC-fibrinogen binding to both PMA-stimulated and unstimulated GM1500 cells. As shown in Fig. 8A, 0.1 M cytochalasin D, but not 0.01 M cytochalasin D, completely inhibited FITC-fibrinogen binding to PMA-stimulated cells. Conversely, whereas cytochalasin D concentrations of 0.1 M or greater did not influence the interaction of FITCfibrinogen with unstimulated lymphocytes, 0.01 M cytochalasin D consistently induced FITC-fibrinogen binding to these cells, albeit to a limited degree (Fig. 8B).
The Rho family of small GTPases regulates a number of cellular functions such as shape, motility, and adhesion by reorganizing the actin cytoskeleton (27). In lymphocytes, inhibiting RhoA with C3 exoenzyme from C. botulinum impairs agonist-stimulated cell adhesion mediated by ␣L␤2 (14) and ␣4␤1 (16). C3 exoenzyme has also been reported to inhibit thrombin-induced platelet aggregation (28). To determine whether RhoA plays a role in the regulation of ␣IIb␤3 function in GM1500 cells, we introduced C3 exoenzyme into the cells using Lipofectin and measured its effect on PMA and fMLPstimulated adherence to fibrinogen. To verify that C3 exoenzyme had ADP-ribosylated RhoA in the Lipofectin-treated cells, an aliquot of these cells was homogenized and re-exposed to the enzyme in the presence of [ 32 P]NAD ϩ . As shown in Fig.  9A, there was a substantial reduction in the incorporation of 32 P into RhoA from cells that had been treated with Lipofectin in the presence of C3 exoenzyme compared with cells that had been treated with Lipofectin in its absence. This indicates that RhoA in the former cells had been ADP-ribosylated during the first incubation with C3 exoenzyme, rendering it resistant to ADP-ribosylation during the second. Nevertheless, as shown in Fig. 9B, inhibiting RhoA with C3 exoenzyme had essentially no effect on the ability of the cells to adherence to fibrinogen following stimulation with either low or high concentrations of PMA and fMLP. DISCUSSION To delineate signaling pathways that can convert ␣IIb␤3 from an inactive to a ligand binding conformation, we have expressed recombinant ␣IIb␤3 in human B lymphocytes. Initially, we found that the phorbol ester PMA not only induced the adherence of these cells to fibrinogen, but enabled ␣IIb␤3 to bind soluble fibrinogen (4). In the current work, we asked whether an extracellular agonist acting through its cognate membrane receptor could induce ␣IIb␤3 function in these cells. Stimulation of G protein-coupled receptors in platelets results in ␣IIb␤3 activation (7). Similarly, stimulation of these receptors leads to integrin activation in lymphocytes and other cells of the hematopoietic lineage (9). We found that by coexpressing ␣IIb␤3 with the human formyl peptide receptor, a seven-trans-membrane domain G protein-coupled receptor (9), we could induce B cell adherence to immobilized fibrinogen and soluble fibrinogen binding to ␣IIb␤3 using the chemoattractant peptide fMLP. Moreover, we found that the signaling pathway initiated by fPR stimulation involved PKC and similar to signaling initiated by PMA, resulted in a change in the actin cytoskeleton.
The human fPR is normally expressed in phagocytic cells where it transduces signals by activating the PTX-sensitive G proteins G␣ i2 or G␣ i3 (9). We found that PTX inhibited the fMLP-stimulated interaction of co-transfected GM1500 cells with both immobilized and soluble fibrinogen, implying that G␣ i activation can regulate ␣IIb␤3 function in these cells. Whether G␣ i can also regulate ␣IIb␤3 function in platelets is uncertain. However, PTX inhibits thrombin-stimulated phosphoinositide hydrolysis in saponin-permeabilized human platelets, suggesting that at least the thrombin receptor in human platelets can couple to G␣ i (29). On the other hand, platelets from G␣ q -deficient mice fail to aggregate in response to thrombin, ADP, collagen, arachidonic acid, and U46619, despite normal levels of G␣ i , indicating that PTX-insensitive G protein G␣ q couples agonist receptors to ␣IIb␤3 in murine platelets (30). Nonetheless, it is possible that the difference between human lymphocytes and murine platelets may simply reflect a difference in the types of signaling pathways that are present in human and murine cells, similar to a difference in the types

FIG. 4. Inhibition of PMA-and fMLP-stimulated lymphocyte adherence to fibrinogen by PTX.
To determine the effect of PTX on ␣IIb␤3-mediated lymphocyte adherence to fibrinogen, GM1500 cells coexpressing the fPR and ␣IIb␤3 were preincubated with either buffer or 100 ng/ml PTX for 2 h. Lymphocyte adherence to immobilized fibrinogen stimulated by 200 ng/ml PMA, 100 nM fMLP, or both agonists together was measured as described in the legend to Fig. 2 and under "Experimental Procedures." The data are expressed as the mean and S.E. of quadruplicate determinations.

FIG. 5. Inhibition of PMA-and fMLP-stimulated lymphocyte adherence to fibrinogen by the PKC inhibitors BIM I and BIM V.
The effect of BIM I (f) and BIM V (u) on ␣IIb␤3-mediated lymphocyte adherence to fibrinogen was determined by incubating GM1500 cells coexpressing the fPR and ␣IIb␤3 overnight with the indicated concentrations of the inhibitors. Lymphocyte adherence to immobilized fibrinogen was stimulated by either 200 ng/ml PMA or 100 nM fMLP and measured as described in the legend to Fig. 2 and under "Experimental Procedures." The data are expressed as the mean and S.E. of quadruplicate determinations.
FIG. 6. Effect of protein tyrosine kinase inhibitors on PMAand fMLP-stimulated lymphocyte adherence to fibrinogen. GM1500 cells coexpressing the fPR and ␣IIb␤3 were incubated overnight with either the protein tyrosine kinase inhibitor genistein at 150 M or the Syk inhibitor piceatannol at 30 g/ml. Lymphocyte adherence to immobilized fibrinogen was stimulated by either 200 ng/ml PMA (f) or 100 nM fMLP (u) and measured as described in the legend to of thrombin receptors expressed by human and murine platelets (31,32).
The ability of inhibitors of PKC (33) to inhibit platelet aggregation and/or ligand binding to ␣IIb␤3 implies that protein phosphorylation by PKC regulates ␣IIb␤3 activity. Moreover, the PKC activator PMA is a potent stimulus for ligand binding to ␣IIb␤3 on both human platelets (6) and G␣ q -deficient murine platelets (30). The identity of the proteins phosphorylated by PKC in platelets to regulate ␣IIb␤3 function is unknown. The ␤3 cytoplasmic tail has been found to contain phosphorylated ␣IIb␤3 Function in B Lymphocytes threonine residues after platelet stimulation by thrombin, PMA, or the thromboxane analogue U46619 (34). However, the fraction of ␤3 containing phosphorylated threonine in both resting and stimulated platelets was low and unlikely to affect the function of more than a few ␣IIb␤3 heterodimers. We found that the fMLP-stimulated interaction of ␣IIb␤3 with either immobilized or soluble fibrinogen in B cells was prevented by the specific PKC inhibitor BIM I. Thus, activation of PKC, either directly with PMA or via fPR activation of G␣ i , is sufficient to induce ␣IIb␤3 function in lymphocytes. On the other hand, Laudanna et al. (16) observed that inhibiting PKC had no effect on fMLP-stimulated ␣4␤1 function in murine B cells (16). Again, it is possible that this difference simply reflects differences between human and murine cells. It is also possible that at least two signaling pathways can be initiated by fPR stimulation and that these pathways can differentiate between ␤1 and ␤3 integrins. In support of this possibility, Weber et al. (35) found that signaling pathways arising from the receptors for the chemoattractants RANTES, MCP-3, and C5a in eosinophils can differentially regulate the function of ␣4␤1 and ␣L␤2. Whether PKC-independent pathways also exist in B cells that can regulate ␣IIb␤3 function remains to be determined.
Phosphorylation of tyrosines 747 and 759 in the ␤3 cytoplasmic tail has also been detected after thrombin-stimulated platelet aggregation (36). However, this phosphorylation was not observed after thrombin stimulation in the absence of aggregation, suggesting that it is a consequence of ligand binding to ␣IIb␤3 ("outside-in" signaling), rather than being part of the process of ␣IIb␤3 activation. Blystone et al. (37) also detected phosphorylation of ␤3 Tyr-747 after monocytes were exposed to Mn 2ϩ or platelets were treated with either Mn 2ϩ or thrombin. When they expressed ␣v␤3 heterologously in K562 cells, they found Tyr-747 phosphorylation necessary, but not sufficient, to support either PMA and thrombin-stimulated cell adhesion, and like Tyr-747 phosphorylation in platelets (36), required ligand binding to ␣v␤3. We found that the protein tyrosine kinase inhibitor genistein had no effect on the ability of ␣IIb␤3 in B lymphocytes to interact with fibrinogen. Similarly, we found that piceatannol, an inhibitor reportedly specific for the tyrosine kinase Syk found in lymphocytes and platelets (38), had no effect on either PMA-or fMLP-stimulated ␣IIb␤3 function in lymphocytes. Thus, our data, combined with the inability to detect significant amounts of phosphorylated serine or threonine on the ␤3 of stimulated platelets (34), suggest that phosphorylation of ␣IIb␤3 is not required to regulate its interaction with ligands.
The biochemical events that follow PKC activation in lymphocytes and platelets are uncertain. However, one consequence of PKC-mediated signaling is regulation of membranecytoskeletal interactions (39). For example, Kucik and coworkers (40) found that exposing human B lymphocytes to PMA increased the diffusion of ␣L␤2 in the plane of the lymphocyte membrane and augmented lymphocyte adherence to ICAM-1. Similar effects were observed following exposure of the lymphocytes to low concentrations of cytochalasin D. These data suggest that cytoskeletal constraints, released by either PKC activation or cytochalasin D, maintain ␣L␤2 in a low avidity state. Lub et al. (41) extended these observations by showing that maximum ␣L␤2-mediated lymphocyte adherence required both ␣L␤2 clustering and an increase in its affinity for ICAM-1. We found that micromolar concentrations of cytochalasin D inhibited the PMA-stimulated interaction of lymphocytes expressing ␣IIb␤3 with either immobilized or soluble fibrinogen. Conversely, we found that nanomolar concentrations of cytochalasin D actually induced fibrinogen binding to ␣IIb␤3 on unstimulated cells. Thus, our observations suggest that both PKC and the actin cytoskeleton play a role in regulating both the avidity and affinity of ␣IIb␤3 for ligands. How this might occur is uncertain. Platelet stimulation results in a conformational change in ␣IIb␤3 that increases its affinity for ligands (42,43). It is also associated with the disassembly of polymerized actin, followed by actin reassembly and a change in platelet morphology (44). It is conceivable that low concentrations of cytochalasin D can initiate actin disassembly, resulting in increases in both integrin mobility and affinity. Fox et al. (44) found that a variable amount of the ␣IIb␤3 in detergent lysates of unstimulated platelets was recovered with fragments of the membrane skeleton and was redistributed to a detergent-insoluble fraction containing a network of cytoplasmic actin filaments after ligand binding. They also observed that high concentrations of cytochalasin E inhibited the binding of the activation-dependent mAb PAC1 to ␣IIb␤3 on ADP and thrombin-stimulated platelets (26). Thus, it is possible that the membrane skeleton in platelets, or in our case in lymphocytes, interacts with the cytoplasmic tails of ␣IIb␤3 to constrain the integrin in a low affinity configuration. Relief of this constraint by agonists (or cytochalasins) could then be responsible for an augmented interaction of ␣IIb␤3 with immobilized fibrinogen and for its ability to bind soluble ligands.
Platelet stimulation is also associated with the formation of clusters of ligand-occupied ␣IIb␤3 on the platelet surface (26). However, ligand valency does not appear to be a factor in the ability of ␣IIb␤3 to recognize soluble ligands (45). Moreover, ␣IIb␤3 is a univalent receptor (11) and electron microscopy of fibrinogen bound to ␣IIb␤3 suggests that a fibrinogen molecule can only bind to one ␣IIb␤3 heterodimer on the surface of a given platelet (46). Hence, it is unlikely that clustering of ␣IIb␤3 alone can account for its ability to bind soluble fibrinogen. Nevertheless, it is possible, and even likely, that agonistinduced clustering of ␣IIb␤3 contributes to the augmented adherence of stimulated lymphocytes and platelets to immobilized fibrinogen.
Activity of members of the Ras family of small GTPases has major effects on cytoskeletal organization. For example, microinjection of activated forms of the small GTPases cdc42, Rac, and Rho into Swiss 3T3 fibroblasts results in the formation of filopodia, lamellopodia, and stress fibers, respectively (27). Similar effects have been reported in macrophages (47). Moreover, the small GTPase RhoA appears to play an important role in regulating integrin function in leukocytes, and perhaps in platelets as well. For example, inhibiting RhoA with C3 exoenzyme prevented PMA-stimulated ␣L␤2-mediated homotypic lymphocyte aggregation (14) and the fMLP-stimulated ␣4␤1mediated adherence of murine lymphocytes to VCAM-1 (16). Morii and co-workers (28) have reported that C3 exoenzyme completely inhibits thrombin and PMA-stimulated human platelet aggregation, although they found little correlation between the extent of RhoA inhibition by the C3 exoenzyme and the extent to which platelet function was impaired. On the other hand, in preliminary studies, Leng et al. (48) did not observe the reported effect of C3 exoenzyme on platelet aggregation. We found that C3 exoenzyme had no effect on either PMA and fMLP-stimulated lymphocyte adherence to fibrinogen, despite substantial ADP-ribosylation of RhoA. Thus, RhoA is unlikely to transduce the signals that regulate ␣IIb␤3 function, at least in our system. A more likely candidate is Rac1. Microinjection of activated Rac1 into Swiss 3T3 cells induces the submembranous accumulation of actin and the formation of integrin-containing focal complexes at the cell margin (49). Furthermore, activation of Rac1 by the exchange factor Tiam1 in T-lymphoma cells induces the formation of submembranous actin filaments, membrane ruffling, and an invasive phenotype (50). Thus, it is conceivable that a Rac-mediated reorganization of the membrane skeleton in lymphocytes, and by extrapolation in platelets, could be an intermediary step in the regulation of ␣IIb␤3 function by agonists.
In summary, we have shown that the ability of integrin ␣IIb␤3, expressed in B lymphocytes, to interact with immobilized and soluble fibrinogen can be stimulated by the G proteincoupled formyl peptide receptor. Moreover, our studies begin to define a signaling pathway that includes a PTX-inhibitable G-protein, PKC, and the actin cytoskeleton.