Chemotactic Peptide-induced Activation of Ras in Human Neutrophils Is Associated with Inhibition of p120-GAP Activity*

The monomeric G-protein Ras is now considered to function as an initial regulator of multiple signaling pathways in both normal and transformed cell types. Adhesion and chemoattractant receptors are known to trigger activation of Ras in human neutrophils, but the signaling mechanism that activates Ras has only been partially elucidated. The present results show that in neutrophils, a time- and dose-dependent f-Met-Leu-Phe (FMLP)-induced activation of Ras is mediated by Gi2-proteins, because such activation is inhibited by pertussis toxin and because direct stimulation of heterotrimeric G-proteins with AlF4 − is sufficient to activate Ras. Pretreatment of neutrophils with tyrosine kinase inhibitors,i.e. genistein or erbstatin that completely block FMLP-stimulated protein tyrosine phosphorylations, did not affect the FMLP-induced activation of Ras. Moreover, FMLP did not induce any detectable translocation of Grb2 and Sos to the plasma membrane of neutrophils. Other signaling molecules, such as protein kinase C, phosphatidylinositol 3-kinase and Ca2+, do not appear to be involved in the FMLP-induced Ras activation. Instead, stimulation of neutrophils with FMLP or C5a, the latter of which also activates Gi2-proteins, resulted in transient inhibition of the activity of Ras GTPase-activating proteins (GAP) with kinetics that correlated well with the kinetics of Ras activation. Moreover, decreased Ras-GAP activity was found in p120-GAP but not in neurofibromin immunoprecipitates of FMLP-stimulated cells. These results suggest that tyrosine kinase-dependent Ras exchange factors do not contribute to the FMLP-induced activation of Ras but that such activation is mediated via inhibition of p120-GAP in neutrophils.

Recruitment of human neutrophils (PMN) 1 to sites of infection is a critical step in host defense against invading microorganisms. This recruitment involves several distinct phases, for example adhesion to the vessel wall and subsequent transmigration and locomotion of PMN in a chemotactic gradient(s) formed at the site of infection (1). The mechanisms by which PMN respond to chemotactic stimuli, such as f-Met-Leu-Phe (FMLP) and C5a, include their interaction with seven transmembrane-spanning receptors coupled to heterotrimeric G i2proteins (2,3). That interaction triggers G i2 -proteins to dissociate into a GTP-bound G␣ subunit and a ␤␥ complex, which in turn activate effectors, e.g. various phospholipases, the Srcrelated tyrosine kinase Lyn, and phosphatidylinositol 3-kinase (PI3K; Refs. [2][3][4][5]. In addition, recent studies have shown that stimulation of adhesion and chemoattractant receptors leads to activation of the small G-protein Ras in PMN (6 -10). Considering these findings and the growing list of Ras targets (e.g. Rac, Rho, and PI3K; all of which are key proteins in the control of the actin cytoskeleton; Refs. [11][12][13][14], make it crucial to reveal the mechanisms involved in the regulation of Ras. Despite the undoubted relevance of such pathways for the regulation of PMN locomotion, the signaling events mediating chemoattractant-induced activation of Ras has only been partially elucidated. Ras was originally identified as the key signaling protein involved in the regulation of cell growth. The activity of Ras is controlled by guanine nucleotide exchange factors (GEF), which exchange GDP for GTP on Ras, and by GTPase-activating proteins (GAP), which accelerate the hydrolysis of GTP to GDP (15,16). Two putative GEF, Vav and Sos, have been identified in human leukocytes (15). The activity of Vav is regulated by direct tyrosine phosphorylation (17)(18)(19)(20) and possibly by the direct association of this exchange factor with Ras (6,20). Sos, on the other hand, regulates Ras when it is recruited to the cell membrane by the adaptor protein Grb2, which binds to phosphotyrosine residues on Shc proteins and/or receptor tyrosine kinases (15,21). The mechanism by which a receptor might regulate Ras-GAP activity is not clear, although inhibition of GAP activities has been shown to activate Ras in T-cells stimulated via antigen-receptor and in HEL cells stimulated with erythropoietin (22)(23)(24).
Ras is now considered to be an initial regulator of multiple signaling pathways that may bifurcate further downstream or at the level of Ras itself (15,25). We have recently shown that engagement of ␤ 2 integrins activates Ras through tyrosine phosphorylation of Vav and the subsequent association of Vav with Ras (6). This finding, together with several reports on the ability of different chemoattractants to activate Ras (7)(8)(9)(10)26), implies that Ras may play an important role in the coordination of signals from adhesion and chemoattractant receptors needed for an efficient locomotory response of PMN. Studies on how such coordination occurs are hampered by the fact that the mechanisms by which Ras are activated in chemoattractantstimulated PMN have not been fully elucidated. However, indirect studies on the activation of Raf/MAPK, the best defined downstream effectors of Ras, suggest a possible role of tyrosine kinase and/or PI3K in Ras activation (7,9,27,28). The aim of the present study was to increase our knowledge about the mechanism(s) by which Ras is activated by the chemotactic peptide FMLP in human PMN by direct analysis of the ratio of Ras-bound GTP/GDP.
Preparation of Human PMN-PMN were isolated from whole blood (drawn from healthy volunteers) by dextran sedimentation followed by a brief hypotonic lysis and centrifugation on Ficoll-Hypaque (Pharmacia, Biotech Inc.) as described previously (30,31).
Ras Activation Assay-Activation of Ras in PMN was determined by analyzing the ratio between GTP and GDP bound to immunoprecipitated Ras as described previously (6,32). PMN were washed twice and then labeled at 37°C for 2 h with 2 mCi 32 P i /ml in a Ca 2ϩ /phosphatefree medium (33) supplemented with 0.1% bovine serum albumin. After washing, the cells were suspended at a concentration of 2 ϫ 10 7 /ml in Ca 2ϩ medium (31) and 0.1% bovine serum albumin. Portions of the cell suspension (500 l) were prewarmed at 37°C for 5 min and then stimulated with FMLP, C5a, or PMA for the indicated periods of time.
In some experiments, the cells were pretreated with various inhibitors and then stimulated with FMLP in the presence of the same inhibitor as specified in the figure legends. The reaction was terminated by adding 125 l of ice-cold lysis buffer A (150 mM NaCl, 5% Nonidet P-40, 50 mM MgCl 2 , 12 g mAb Y13-259/ml, 5 mM phenylmethylsulfonyl fluoride, 5 mM Na 3 VO 4 , 25 g of leupeptin, and aprotinin/ml, in 100 mM Tris/HCl, pH 7.5), and the samples were then incubated on ice for 30 min. After centrifugation (15,000 ϫ g for 10 min), the cell lysates were immediately adjusted to 0.5 M NaCl/0.05% SDS/0.5% sodium deoxycholate, and the obtained immunocomplexes were collected with protein G-plus agarose. The radiolabeled guanine nucleotides bound to Ras were then resolved by thin layer chromatography and quantified with a Phosphor-Imager (Fuji BAS 1000) as described previously (6). The specificity of this assay was confirmed by omitting the Ras mAb during the immunoprecipitation step (6).
Western Blot Analysis-PMN were pretreated with genistein or medium alone and then stimulated with FMLP. The reaction was stopped by a quick spin, and total cellular extracts were prepared as described by Torres et al. (34). Thereafter, proteins (30 g/lane) were subjected to SDS-PAGE, electrotransferred to nitrocellulose membranes, and analyzed by immunoblotting with mAb 4G10 and a commercial ECL kit.
Membrane Association of Grb2 and Sos-After stimulation with FMLP for different periods of time, 10 7 PMN were rapidly pelleted, resuspended in 1 ml of lysis buffer B (2 mM EDTA, 0.5 mM EGTA, 2 mM MgCl 2 , 20 mM Tris, pH 7.4, and protease inhibitors, i.e. 0.3 mM phenylmethylsulfonyl fluoride, 1 mM Na 3 VO 4 , and 5 g of leupeptin and aprotinin/ml) and then sonicated (35). After an initial centrifugation (10,000 ϫ g for 5 min), the obtained supernatants were further centrifuged at 200,000 ϫ g for 30 min to obtain cytosolic and membrane fractions. The membrane fractions were washed once with lysis buffer B and then recentrifuged (200,000 ϫ g for 30 min) and thereafter resuspended in Laemmli sample buffer (36). Proteins were subjected to SDS-PAGE, electrotransferred to nitrocellulose membranes, and analyzed by immunoblotting with anti-Grb2 or anti-Sos Ab and a commercial ECL kit.
Assay for Ras-GAP Activity-The Ras-GAP activity was detected by incubating cell lysates with [␥-32 P]GTP-loaded Ras proteins and then quantifying the decrease in radiolabeled Ras-GTP upon nitrocellulose filtration as described previously (24,37). In short, after stimulation, PMN (7 ϫ 10 7 cells/ml, 200 l/sample) were lysed by adding 50 l of ice-cold 50 mM Hepes buffer, pH 7.4, containing 25 mM MgCl 2 , 5% Triton X-100, 0.5 M okadaic acid, and protease inhibitors as in lysis buffer A, and the nuclei were then removed by centrifugation. Ras was loaded with [␥-32 P]GTP (final volume, 30 l) at 30°C for 10 min in 20 mM Tris buffer, pH 7.5, containing 100 mM NaCl, 0.5 mM EDTA, 0.5 mg bovine serum albumin/ml, 0.005% sodium cholate, 0.5 mM dithiothreitol, 300 ng of bacterially expressed wild-type human p21 H-ras and 30 Ci of [␥-32 P]GTP (specific activity, 5000 Ci/mmol). Aliquots of the [␥-32 P]GTP-loaded Ras (20 ng) were mixed with 20 l of cell lysate and 60 l of assay buffer (24), and these mixtures were incubated for 10 min at 30°C. The incubation period was reduced to 5 min when measuring the Ras-GAP activity in PMA-stimulated cells, because it has been observed by us (present study) and by others (37) that the activity was no longer linear with time if more than 50 -60% Ras-GTP was hydrolyzed. Reactions were stopped by adding 1 ml of ice-cold 25 mM Tris, pH 7.5, containing 0.1 M NaCl and 5 mM MgCl 2 , and the mixtures were then rapidly vacuum filtered through nitrocellulose. The filters were washed three times with 5 ml of the same buffer, and the radioactivity was quantified by liquid scintillation counting. Activities were expressed as the percentage of Ras-bound [␥-32 P]GTP that was hydrolyzed, as compared with control buffer (24,37).
To measure the GAP activity in the immunoprecipitates, clarified lysates of 2 ϫ 10 7 cells prepared as above were incubated with 3 g of nonimmune rabbit IgG, anti-p120-GAP, or anti-NF1 Ab for 1 h and then with protein G-plus Sepharose beads for 1.5 h at 4°C. The immunoprecipitates were washed three times with a Tris buffer (25 mM, pH 7.5) containing 0.1% Triton X-100, 0.5 M NaCl, and 5 mM MgCl 2 , and once with assay buffer. Thereafter, the precipitates were resuspended in 80 l of assay buffer and incubated with 4 ng of [␥-32 P]GTP-loaded Ras for 15 min at 30°C. The reaction was terminated, and the GAP activity was determined as described above. Activities were expressed as the percentage of Ras-bound [␥-32 P]GTP that was hydrolyzed, as compared with lysate-unexposed beads.
Statistical Analysis-All data are given as the means Ϯ S.D. The significance of the differences was analyzed by Student's t test (paired).

FMLP-stimulated Activation of Ras-
The nucleotide pools of PMN were labeled with 32 P i , and the ratio between the GTP and the GDP bound to immunoprecipitated Ras was determined as an index of activation. Fig. 1 (A and B) shows the time course of Ras activation after stimulation with 10 nM FMLP. The activation was rapid, the ratio of GTP-bound Ras (i.e. the ratio between GTP and GTP plus GDP bound to Ras) reached a maximum after stimulation for 1 min. Thereafter, the ratio gradually decreased, although a 2-fold increase as compared with the initial level was still found after 10 min of stimulation ( Fig. 1, A and B). In dose dependence experiments, the FMLPinduced activation of Ras occurred at a concentration of only 1 nM, and the maximal effect was observed at 10 -100 nM (Fig. 1,  C and D).
The Role of Tyrosine Kinase in FMLP-induced Ras Activation-Engagement of FMLP receptors on PMN has been shown to activate Lyn and cause a subsequent tyrosine phosphorylation of Shc (4). This is interesting, because in other systems, phosphorylation of Shc appears to activate Ras via Shc-Grb2-Sos complexes (21,38,39). To investigate whether a tyrosine kinase(s) is involved in the FMLP-induced activation of Ras, PMN were pretreated with 100 M genistein (40) or 150 M erbstatin (41) for 30 min at 37°C and then stimulated with FMLP in the presence of the same inhibitor. In agreement with previous reports (34,42), we found that a variety of cellular proteins were tyrosine phosphorylated in PMN after stimulation with FMLP alone (i.e. without an inhibitor, Fig. 2). Pretreatment of cells with genistein abolished this FMLP-induced protein tyrosine phosphorylation (Fig. 2). Similar results were obtained in cells pretreated with erbstatin (data not shown). However, when using an identical protocol, the two inhibitors had no effect on FMLP-stimulated activation of Ras (Fig. 3). It also should be noted that in a previous study (6), we found that the same concentrations of genistein and erbstatin completely blocked ␤ 2 integrin-induced activation of Ras in PMN.
To further exclude the involvement of a tyrosine kinase in the FMLP-induced activation of Ras, we examined membrane accumulation of Grb2 and Sos as an index of their activation (15). A detectable level of both Grb2 and Sos were found in membrane fractions from resting PMN (Fig. 4). More importantly, however, stimulation of PMN with FMLP for various periods of time failed to induce any detectable increase of these proteins in the membrane fraction (Fig. 4). Neither genistein nor erbstatin had any effect on the content of Grb2 and Sos in membrane fractions from resting or FMLP-stimulated PMN (data not shown).  (45), did not affect the FMLP-induced activation of Ras (Fig. 3). Furthermore, a combination of genistein and wortmannin caused only a marginal decrease in the FMLP-induced binding of GTP to Ras (Fig. 3). The possible involvement of cytosolic free Ca 2ϩ and PKC in this Ras activation was also investigated by pretreating PMN with 1 M U73122 or 1 M staurosporine for 10 min at 37°C before exposure to FMLP; U73122 is an inhibitor of FMLP-induced Ca 2ϩ signaling (46), and staurosporine is a potent protein kinase inhibitor (47). Neither of the inhibitors had any effect on the FMLP-stimulated activation of Ras. The ratio of GTPbound Ras in control cells stimulated with 10 nM FMLP for 1 min was 56.1 Ϯ 6.4, and the ratio was 55.4 Ϯ 6.7 and 58.8 Ϯ 6.8 for cells pretreated with U73122 and staurosporine, respectively (n ϭ 3). The Role of G i2 Proteins in FMLP-induced Activation of Ras-Many of the responses induced in PMN by exposure to FMLP are mediated by G i2 proteins and are therefore sensitive to PT (2, 3). To determine whether activation of G i2 proteins is involved in FMLP-stimulated activation of Ras, PMN were pretreated with 1 g of PT/ml for 2 h before stimulation with FMLP. Such treatment inhibited FMLP-induced activation of Ras by about 80% (Fig. 5), suggesting that the activation of Ras occurs downstream of G i2 protein activation. To further elucidate the role of a heterotrimeric G-protein in the FMLP-induced activation of Ras, we studied the effects of AlF 4 Ϫ on the activity of Ras; it is known that AlF 4 Ϫ binds directly to and activates heterotrimeric but not small G-proteins (48). As can be seen in Fig. 6, exposure to AlF 4 Ϫ caused a significant increase in the ratio of GTP-bound Ras in PMN.
The Role of Ras-GAP in Activation of Ras in PMN-Theoretically, activation of Ras can be accomplished by inhibition of Ras-GAP or by stimulation of GEF. In light of the abovementioned results, we investigated the possibility that FMLP regulates Ras by inhibiting its GAP. Equal amounts of proteins from unstimulated PMN and PMN stimulated with FMLP for different periods of time were tested for their ability to hydrolyze [␥-32 P]GTP bound to exogenous Ras. As shown in Fig. 7A, about 60% of Ras-bound [␥-32 P]GTP was hydrolyzed within 10 min by a lysate of unstimulated PMN. The percentage of hydrolysis decreased dramatically to 25% when testing a lysate of PMN stimulated with FMLP for 30 s. Thereafter, the GAP activity was gradually recovered with increasing time of exposure to FMLP. For comparison, the FMLP-induced activation of Ras (Fig. 1) is also outlined in Fig. 7A.
To further elucidate the role of Ras-GAP in Ras activation mediated by G i2 -coupled receptors, we tested C5a (which also activates G i2 -protein) regarding its effect on activation of Ras and Ras-GAP activity. We found that C5a induced a rapid activation of Ras in PMN, and this showed similar kinetics but PMN were incubated with 32 P i in the presence or the absence of 1 g of PT/ml at 37°C for 2 h and then stimulated with 10 nM FMLP for various periods of time. Thereafter, the cells were lysed, and precipitation of Ras and determination of the ratios of Ras-bound GTP/(GTP ϩ GDP) were performed as described in Fig. 1. A shows representative TLC results. B illustrates the ratios of Ras-bound radioactive GTP/(GTP ϩ GDP), which were determined as an index of activation. The compiled results from several TLC plates are given as the means Ϯ S.D. of four separate experiments. a lower magnitude than that triggered by FMLP (Fig. 7B). A rapid but transient inhibition of Ras-GAP activity was observed in the cells stimulated with C5a, with a maximal inhibition (about 48%) at 1 min. Consequently, the kinetics of inhibition of Ras-GAP in PMN stimulated with FMLP or C5a is closely correlated with the kinetics of Ras activation (Fig. 7, A  and B).
PMA is a nonreceptor agonist that acts independently of G i2 -proteins and is known to activate Sos and Vav (49,50). Hence, as expected, stimulation of neutrophils with PMA activated Ras but did so less effectively (Fig. 7C) than exposure to FMLP or C5a (Fig. 7, A and B). In contrast to G i2 -coupled receptors, PMA stimulates the Ras-GAP activity in PMN (Fig. 7C).
The most well known Ras-GAP in human cells are p120-GAP and NF1, and in our initial experiments, Western blot analysis revealed that both of these are expressed in PMN (data not shown). To determine which of them is involved in the FMLPmediated inhibition of Ras-GAP activity, we measured the Ras-GAP activity in the immunoprecipitates of those proteins. Both p120-GAP and NF1 precipitates from resting PMN exhibited Ras-GAP activity (Fig. 8), whereas antibody control immunoprecipitates did not (data not shown). Stimulation of PMN with FMLP caused about 60% inhibition of the GAP activity in p120-GAP precipitates but had no effect on the GAP activity in NF1 precipitates (Fig. 8). DISCUSSION By direct analysis of the ratio between GTP and GDP bound to Ras, we here show that stimulation of PMN with FMLP results in a pertussis toxin-sensitive activation of Ras. Apparently, this activation does not require the participation of the two known Ras-GEF expressed in PMN, i.e. Sos and Vav, both of which require the activation of a tyrosine kinase (15). Instead, FMLP-induced activation of Ras is correlated with a distinct inhibition of Ras-GAP activity in PMN.
Agonist activation of G i -coupled receptors leads to the formation of activated G␣-GTP and the release of G␤␥ subunits. In many cell types, the G␤␥ subunit has been identified as the primary mediator of Ras activation induced by stimulation of G i -coupled receptors (2,3,51). In this context, it has been suggested that G␤␥ subunits activate a Src-related tyrosine kinase (e.g. Lyn in PMN) that phosphorylates Shc, which allows the formation of Shc-Grb2-Sos complexes and the subse-quent activation of Ras (3,51). However, in our experiments, neither genistein nor erbstatin affected FMLP-induced activation of Ras, which indicates that a signaling event other than the activation of a tyrosine kinase mediates the G i2 -coupled receptor-induced activation of Ras. This conclusion is supported by our observations that both inhibitors effectively suppressed all detectable FMLP-induced tyrosine phosphorylation events and also abolished the ␤ 2 integrin-mediated activation of Ras in PMN (6). The present results are substantiated by the recent report that in transfected fibroblasts FMLP can activate MAPK in the absence of Lyn and tyrosine phosphorylation of Shc (26). In other words, the earlier findings together with our present results indicate that activation of a tyrosine kinase is not involved in the regulation of Ras by G i2 -protein-coupled receptors in PMN, although it might well participate in the further downstream regulation of the MAPK cascade (7,34).
Theoretically, Ras is stimulated by the activation of GEF and/or the inhibition of GAP. At present, two putative GEF, i.e. Sos and Vav, have been identified in human leukocytes (15). The activities of these GEF are regulated by tyrosine phosphorylation (Vav) or by the formation of a complex composed of Sos and its adaptor proteins Grb2-Shc (3,15,36). The part these GEF play in FMLP-stimulated activation of Ras is now being questioned, because such activation is not affected by tyrosine kinase inhibitors. Furthermore, in our study, stimulation with FMLP did not induce any detectable translocation of either Sos or Grb2 to a membrane fraction of PMN. Thus, it is unlikely that Sos participates in the chemotactic peptide-induced activation of Ras.
In addition to tyrosine phosphorylation, activation of PI3K or PKC and an increase in [Ca 2ϩ ] i have also been implicated in the regulation of Ras activity (23,39,44,52). It was recently reported that PI3K is an early intermediator in the G␤␥-induced activation of Ras in COS cells (43). The fact that G␤␥sensitive PI3K is present in PMN (53) and that FMLP triggers the activation of PI3K (5,33) indicates that this enzyme may regulate the FMLP-induced activation of Ras. However, the possibility of such a regulatory mechanism is refuted by the fact that wortmannin, which completely inhibits PI3K activity in PMN (45), failed to inhibit FMLP-induced activation of Ras in our experiments. Moreover, genistein and wortmannin combined had only a marginal inhibitory effect, which further supports our suggestion that the activation of a tyrosine kinase and/or PI3K is not required for the FMLP-induced activation of Ras in PMN. In agreement with our results, it was recently observed that wortmannin inhibits the activation of Raf/MAPK but not that of Ras in PMN stimulated with C5a and interleukin-8 (9). These findings imply that activation of Ras and Raf/MAPK activation can be controlled by different regulatory mechanisms. Considering the other possible signaling events mentioned above, our results show that neither an increase in [Ca 2ϩ ] i nor an activation of PKC is involved in FMLP-induced activation of Ras in PMN.
Most studies on the regulation of Ras have been focused on receptor-mediated control of GEF, and little is known about the role of GAP in the activation of Ras. In the present study, we found that stimulation of PMN with either FMLP or C5a, both of which activate G i2 -proteins, induced a rapid but transient inhibition of Ras-GAP activity, and the kinetics correlated well with their activation of Ras. Because Ras-GAP is a negative regulator of Ras, strong inhibition of its activity in PMN shortly after stimulation with chemoattractants would allow progressive accumulation of the Ras-GTP complex. The decrease in GTP-bound Ras observed in PMN after longer stimulation (5 min) was probably due to the recovery of Ras-GAP activity. These results indicate that the inhibition of Ras-GAP activity is an important regulator of the chemotactic peptide-induced activation of Ras in PMN, although we cannot of course exclude the possibility that the chemotactic peptide activates Ras in part through novel/unknown GEF and/or a novel regulatory mechanism for those GEF.
Although both p120-GAP and NF1 are expressed in PMN, the decreased Ras-GAP activity in FMLP-stimulated cells was most likely due to the inhibition of p120-GAP. The precise modulatory mechanisms underlying the inhibition of Ras-GAP observed in our study are at present unclear. Although FMLP can induce tyrosine phosphorylation of p120-GAP (54), it has been shown that such phosphorylation per se is not sufficient to inhibit the activity of p120-GAP (32). That finding is compatible with our results showing that genistein abolished the FMLP-induced tyrosine phosphorylation of pp120 but had no effect on the activation of Ras. Recent studies in vitro have shown that G␤␥ subunits can reversibly bind to the pleckstrin homology domain of several proteins, such as Ras-GAP, and overexpression of pleckstrin homology domain peptides in vivo has been found to inhibit the G i -coupled receptor-mediated activation of Ras (55,56). Therefore, such a mechanism might be involved in the inhibition of Ras-GAP activity seen in our experiments. This hypothesis is in agreement with a recent report that FMLP triggers the translocation of p120-GAP to the membrane of PMN as early as 20 s after stimulation (54).
PMN are nonmitotic cells and their main function is to migrate to sites of infection; hence the effects of Ras on cell growth and differentiation are irrelevant in PMN. Marshall (57) recently proposed that the outcome of Ras/MAPK signaling is largely dependent on the duration of its activation. If that is the case, rapid activation of Ras in PMN may represent a distinct category of signaling. Accumulating evidence indicates that Ras is an initial regulator of multiple-signaling pathways, including those comprising key proteins involved in the regulation of the actin cytoskeleton (i.e. Rac, Rho, and PI3K; Ref. 11). Transfection of the active form of Ras into granulocytic HL-60 cells is associated with an increased level of F-actin and a tyrosine phosphorylation of the cytoskeletal protein paxillin (58). Recently, Joneson et al. (25) reported that regulation of the MAPK pathway and the pathway controlling cell morphology bifurcates at the level of Ras itself, lending further support for a role of Ras in the regulation of the actin cytoskeleton. ␤ 2 integrins, which are essential for PMN locomotion, were recently found to induce activation of Ras via tyrosine phosphorylation of the Ras-GEF Vav (6). This is in contrast to FMLPinduced activation of Ras, which is most likely controlled by an inhibition of Ras-GAP activity (present study). Therefore, it is tempting to suggest that the different signaling mechanisms used by receptors for adherence and chemotaxis to activate Ras might lead to a better coordination in the rearrangement of the actin cytoskeleton and thereby a more efficient locomotory response of PMN. FIG. 8. Effect of FMLP on the activity of Ras-GAP in p120-GAP and NF1 immunoprecipitates. PMN (2 ϫ 10 7 /sample) were exposed to 10 nM FMLP at 37°C for 0 (open bars) or 30 (closed bars) s and then lysed. The lysates were immunoprecipitated (IP) with 3 g of anti-p120-GAP or anti-NF1 Ab. Precipitates were incubated with [␥-32 P]GTPloaded Ras at 30°C for 15 min, and the GAP activity was measured as described under "Materials and Methods." The results are the means Ϯ S.D. of three separate experiments, each in duplicate.