Role of p190RhoGAP in β2 Integrin Regulation of RhoA in Human Neutrophils Karim Dib, Fredrik Melander and Tommy Andersson This information is current as J Immunol 2001; 166:6311-6322; ; of September 26, 2021. doi: 10.4049/jimmunol.166.10.6311 http://www.jimmunol.org/content/166/10/6311 Downloaded from References This article cites 53 articles, 34 of which you can access for free at: http://www.jimmunol.org/content/166/10/6311.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/ • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 26, 2021 Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. ␤ Role of p190RhoGAP in 2 Integrin Regulation of RhoA in Human Neutrophils1

Karim Dib,2 Fredrik Melander, and Tommy Andersson ␤ We found that engagement of 2 integrins on human neutrophils induced activation of RhoA, as indicated by the increased ratio ␤ ؉ of GTP:GTP GDP recovered on RhoA and translocation of RhoA to a membrane fraction. The clustering of 2 integrins also ␤ induced a time-dependent increase in GDP bound to RhoA, which correlated with 2 integrin-induced activation of p190RhoGAP. The activation of p190RhoGAP was completely blocked by [4-amino-5-(4-methylphenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine] ␤ (PP1), a selective inhibitor of Src family tyrosine kinases. However, clustering of 2 integrins did not increase the basal tyrosine ␤ phosphorylation of p190RhoGAP, nor did it affect the amount of p120RasGAP bound to p190RhoGAP. Instead, the 2 integrin- induced activation of p190RhoGAP was accompanied by increased tyrosine phosphorylation of a p190RhoGAP-associated protein, ␤ p120RasGAP, and accumulation of both p120RasGAP and p190RhoGAP in a membrane fraction. PP1 blocked the 2 integrin- Downloaded from induced phosphorylation of p120RasGAP, as well as the translocation of p190RhoGAP and p120RasGAP, but it did not affect the accumulation of RhoA in the membrane fraction. In agreement with the mentioned findings, PP1 also increased the GTP:GTP ؉ ␤ ␤ GDP ratio recovered on RhoA immunoprecipitated from 2 integrin-stimulated cells. Thus, in neutrophils, 2 integrin-induced activation of p190RhoGAP requires a signal from a Src family tyrosine kinase, but it does not occur via the signaling pathway responsible for activation of RhoA. The Journal of Immunology, 2001, 166: 6311–6322. http://www.jimmunol.org/ ␤ embers of the 2 integrin family are the dominating long to the Rho subfamily, members of the Ras superfamily of integrins expressed on neutrophils, and they play a small GTPases, have been strongly implicated (11). These small M major role in cell-cell and cell-matrix adhesion of neu- GTP-binding proteins cycle between a GDP-bound inactive form ␤ trophils. 2 integrins are noncovalently associated heterodimers to a GTP-bound active form (12). Guanine nucleotide exchange composed of a common ␤-chain, CD18, and one of four unique factors (GEFs),3 activated by extracellular stimuli, are responsible ␣-chains; CD11a, CD11b, CD11c, or CD11d, with CD11b/CD18 for the GDP-GTP exchange, a transition that allows translocation being the most prominent on neutrophils (1, 2). The signal trans- of the monomeric G proteins to the plasma membrane (13, 14). In duction properties of these integrins are vital for the migration of their GTP-bound state, these proteins interact with specific effec- neutrophils, but to date have only been partly elucidated. tors to initiate downstream signals and functions. The subsequent by guest on September 26, 2021 ␤ Engagement of 2 integrins on neutrophils has been accom- hydrolysis of bound GTP to GDP is mediated via the intrinsic plished in experimental models by plating the cells on surface- GTPase activity of these GTP-binding proteins, which, however, is bound anti-␤ integrin Abs or on ICAM-1 or other types of coated 2 so low that the proteins are totally dependent on GTPase-activating surfaces. The results show that ␤ integrin engagement triggers 2 proteins (GAPs) to achieve adequate hydrolysis of GTP (12). various signal transduction events, including tyrosine phosphory- The best-characterized Rho-specific GAP is p190RhoGAP, lation of several different proteins (3–8), activation of the Src fam- which was originally reported to be a tyrosyl-phosphorylated pro- ily tyrosine kinases p59/61hck and p58c-fgr (3, 9), and activation of tein that coprecipitated with p120RasGAP (15). Molecular cloning p21ras (7). These signals are associated with a profound alteration of the cognate cDNA of p190RhoGAP has revealed an NH -ter- in cell morphology that leads to cell spreading and locomotion. It 2 has been reported that these morphological changes are driven by minal region that contains several sequence motifs that are shared ␤ by all known GTP-binding proteins, and a C-terminal region with dynamic 2 integrin-induced rearrangement of the actin cytoskel- eton (10), but little is known about the nature of the signal(s) a GAP domain specific for the GTP-binding proteins Rho and Rac ␤ (16). Several lines of evidence support the idea that p190RhoGAP generated by 2 integrins regulating cytoskeletal rearrangements. However, in other cell types, small GTP-binding proteins that be- somehow participates in rearrangement of actin cytoskeleton. For example, microinjection of the C-terminal region of p190RhoGAP into Swiss 3T3 cells has been found to block serum-induced stress Division of Experimental Pathology, Department of Laboratory Medicine, Lund Uni- fiber formation (17). McGlade et al. (18) reported that constitutive versity, Malmo¨University Hospital, Malmo¨, Sweden association of GAP-N (i.e., p120RasGAP lacking the C-terminal Received for publication May 5, 2000. Accepted for publication March 7, 2001. Ras-binding domain) with p190RhoGAP increased the GAP ac- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance tivity of this complex, caused a subsequent cellular loss of actin with 18 U.S.C. Section 1734 solely to indicate this fact. stress fibers, and reduced cellular adhesion. In adherent melanoma 1 ␤ This work was supported by the Swedish Cancer Foundation, the Network for In- cells, Nakahara et al. (19) found that engagement of 1 integrins flammation Research (funded by the Swedish Foundation for Strategic Research), the Universitetssjukhust-Malmo¨ Alma¨nna Sjukhus Research Foundations, the Medical Faculty of Lund University, The King Gustaf V Memorial Foundation, The Crafoord Foundation, The O¨ sterlund Foundation, and The Koch Foundation. 2 Address correspondence and reprint requests to Dr. Karim Dib, Department of Lab- 3 Abbreviations used in this paper: GEF, guanine nucleotide exchange factor; FN, oratory Medicine, Division of Experimental Pathology, Lund University, Malmo¨Uni- fibrinogen; GAP, GTPase-activating protein; PP1, [4-amino-5-(4-methylphenyl)-7-(t- versity Hospital, Entrance 78, SE-205 02 Malmo¨, Sweden. E-mail address: butyl)pyrazolo[3,4-d]pyrimidine]; RAM, rabbit anti-mouse; EGF, epidermal growth [email protected] factor; PVDF, polyvinylidene difluoride.

Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00 ␤ 6312 2 INTEGRIN REGULATION OF p190RhoGAP

induced increased phosphorylation of p190RhoGAP, which en- were washed twice with PBS and then incubated with casein (2.0 mg/ml) abled association with filamentous actin. In addition, it was dem- for 2 h. Thereafter, the plates were washed twice with PBS and anti-CD18 ␮ onstrated (20) that increased tyrosine phosphorylation of (IB4) Ab (10 g/ml) or the isotype-matched negative control IgG2a Ab (10 ␮g/ml) was added, and the plates were incubated overnight. Before use, the p190RhoGAP correlated with rapid and exaggerated disassembly plates were washed and incubated with 10% heat-inactivated FCS for 2 h of actin stress fibers. The cited authors suggested that the intrinsic and then washed extensively (three times with PBS and twice with the ␤ RhoGAP activity of p190RhoGAP might be augmented by phos- calcium-containing medium). 2 integrins on neutrophils were engaged by phorylation, but they have not tested that possibility. adding the cells to the coated dishes at 37°C and incubating for different periods of time. Alternatively, neutrophils were plated on a surface coated In this paper, we investigated the signal transduction pathways with FN (20 ␮g/ml) in the presence of TNF-␣ (20 ng/ml), a procedure that involved in ␤ integrin-mediated regulation of RhoA activity in ␤ ␤ 2 specifically activates 2 integrins on neutrophils (24). As a control for 2 neutrophils. In particular, we researched whether a relationship integrin-independent spreading of neutrophils, we also plated these cells on exists between activation of p190RhoGAP and RhoA. anti-CD59 Ab-coated surfaces (the method used was identical with that ␤ used for anti- 2 integrin Ab coating). The reactions were terminated by placing the plates on ice and adding a lysis buffer (100 mM Tris-HCl (pH Materials and Methods 7.5), 1% Nonidet P-40, 5 mM EDTA, 5 mM EGTA, 50 mM NaCl, 5 mM ␮ ␮ Antibodies NaF,1mMNa3VO4, and protease inhibitors (20 g/ml aprotinin; 1 g/ml of each pepstatin, leupeptin, and antipain; 2.5 mM benzamidine; and 2 mM The Abs and their sources were as follows: mAb IB4 (mouse anti-human Pefabloc)). CD18, IgG2a isotype) was obtained from S. Wright (Rockefeller Univer- ␤ Engagements of 2 integrins on nonadherent neutrophils were per- sity, New York, NY) (21); mAb 4G10 (mouse anti-phosphotyrosine) and formed by incubating the cells in calcium-containing medium at 37°C for mouse anti-p190RhoGAP mAb used for immunoprecipitation were ob- 20 min with either the IB4 Ab (10 ␮g/ml) or the isotype-matched IgG2a tained from Upstate Biotechnology (Lake Placid, NY); mAb PY20 (mouse ␮ ␤ Downloaded from mAb (10 g/ml). Thereafter, the cells were washed, and 2 integrins were anti-phosphotyrosine), the anti-p190RhoGAP mAb, and the polyclonal anti- engaged by adding RAM Igs (1/50) for 30 s (8). The reaction was termi- p120RasGAP antiserum were purchased from Transduction Laboratories nated by placing the cells on ice and adding lysis buffer (see above). In (Lexington, KY); the anti-CD59 mAb was purchased from PharMingen experiments using PP1, the tyrosine kinase inhibitor was added along with Europe (Heidelberg, Germany); rabbit anti-RhoA antiserum was obtained the IB4 or isotype-matched Abs and also during the subsequent stimulation from Santa Cruz Biotechnology (Santa Cruz, CA); and the negative control periods. monoclonal IgG2a Ab (directed against Asperigillus niger glucose oxi- dase), peroxidase-conjugated Igs, and rabbit anti-mouse (RAM) Igs were

purchased from Dakopatts (Glostrup, Denmark). Immunoprecipitation and Western blotting http://www.jimmunol.org/ Chemicals Cell lysates were clarified by centrifugation, and the supernatants were immunoprecipitated. This was performed by exposure to the anti- The chemicals and their sources were as follows: protein A-Sepharose, phosphotyrosine Ab (3 ␮g/ml) or anti-p120RasGAP antiserum (3 ␮g/ml) dextran, and Ficoll-Hypaque were purchased from Pharmacia (Uppsala, for 1 h, and then to 40 ␮l of a 50% slurry of protein A-Sepharose for 45 Sweden); protein G plus agarose and recombinant human RhoA were ob- min. The beads were subsequently collected by centrifugation and washed tained from Calbiochem (Cambridge, MA); the protease inhibitors Pefab- three times in a wash buffer (50 mM HEPES (pH 7.4), 1% Triton X-100, loc, pepstatin, leupeptin, aprotinin, and antipain were obtained from Roche 0.1%, SDS, 150 mM NaCl, and 1 mM Na3VO4). The beads were then (Mannheim, Germany); benzamidine, casein, fibrinogen (FN), soluble pro- resuspended in 2ϫ concentrated Laemmli sample buffer and boiled under tein A, and p-coumaric acid were obtained from Sigma-Aldrich (Stock- reducing conditions for 5 min. The immunoprecipitated proteins were sub-

holm, Sweden); polylysine and luminol were obtained from ICN Pharma- jected to electrophoresis on 7.5% SDS-PAGE and transferred to polyscreen by guest on September 26, 2021 ceuticals (Costa Mesa, CA); the tyrosine kinase inhibitor [4-amino-5-(4- polyvinylidene difluoride (PVDF) transfer membranes. The membranes methylphenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine] (PP1) was purchased were blocked in PBS supplemented with 0.2% Tween 20 and 3% BSA, from Alexis Biochemicals (La¨ufelfingen, Switzerland); all electrophoresis incubated for 1 h with a primary Ab (1/5,000 dilution of the anti- reagents were obtained from Bio-Rad (Richmond, CA); and [␥-32P]GTP phosphotyrosine Ab 4G10; 1/750 dilution of the anti-p190RhoGAP Ab or and [32P]orthophosphate were purchased from Amersham (Aylesbury, anti-p120RasGAP Ab), and thereafter washed three times for 15 min in U.K.). TNF-␣ was obtained from R&D Systems (Minneapolis, MN). All PBS supplemented with 0.2% Tween 20. The membranes were subse- other chemicals were of analytical grade and purchased from quently incubated for 1 h with peroxidase-conjugated anti-mouse IgGs (1/ Sigma-Aldrich. 10,000) in PBS supplemented with 0.2% Tween 20 and 3% BSA. The blots were washed, and Ab binding was visualized by ECL. Isolation of human neutrophils Blood from healthy donors was collected and isolated under endotoxin-free Measurement of guanyl nucleotides bound to RhoA conditions as previously described (22). The neutrophils were isolated by Neutrophils were incubated for2hinacalcium- and phosphate-free me- dextran sedimentation followed by a brief hypotonic lysis of erythrocytes. dium (136 mM NaCl, 5.9 mM KCl, 1.2 mM MgSO , 5.0 mM NaHCO , 5.5 The lysis was stopped by adding 3 ml of buffer A (565 mM NaCl, 2.7 mM 4 3 mM glucose, and 20 mM HEPES, pH 7.4) supplemented with KCl, 6.7 mM Na HPO 2H O, and 1.5 mM KH PO , pH 7.3) and 3 ml of 2 4 2 2 4 [32P]orthophosphate (2 mCi/ml). Clearly, this incubation period is not Ringer’s modified phosphate buffer (120 mM NaCl, 4.9 mM KCl, 1.7 mM enough to obtain a complete labeling equilibrium of the nucleotide pools. KH PO , 1.2 mM MgSO 7HO, 8.3 mM Na HPO 2H O, and 10 mM 2 4 4 2 2 4 2 However, due to the limited life span of human neutrophils this is the glucose, pH 7.3). The cell suspension was then centrifuged on Ficoll- optimal labeling period that can be used for such experiments. The cells Hypaque (15 ml) and washed twice with Ringer’s modified phosphate were then washed in the calcium- and phosphate-free medium, resuspended buffer. Finally, the cells were resuspended in a calcium-containing medium in the calcium-containing medium, and finally plated on a surface coated (136 mM NaCl, 4.7 mM KCl, 1.2 mM KH PO , 1.2 mM MgSO , 5.0 mM 2 4 4 with the anti-CD18 Ab IB4 for different periods of time. The cells were NaHCO , 1.1 mM CaCl , 0.1 mM EGTA, 5.5 mM glucose, and 20 mM 3 2 then lysed in an ice-cold lysis buffer (50 mM HEPES (pH 7.4), 1% Triton HEPES, pH 7.4). The cell suspension consisted of ϳ97% neutrophils. Al- X-100, 0.5% deoxycholate, 0.05% SDS, 500 mM NaCl, 15 mM MgCl ,1 ternatively, for in vivo labeling with [32P]orthophosphate, cells were 2 mM EGTA, 10 mM benzamidine, 1 mM Na VO ,10␮g/ml leupeptin, 10 washed twice in calcium- and phosphate-free medium (136 mM NaCl, 5.9 3 4 ␮g/ml aprotinin, 1 mM Pefabloc, and 1 ␮g/ml of pepstatin and antipain mM KCl, 1.2 mM MgSO , 5.0 mM NaHCO , 5.5 mM glucose, and 20 mM 4 3 containing the anti-RhoA antiserum (4 ␮g/ml)). The lysates were clarified HEPES, pH 7.4) and then resuspended in the same medium. by centrifugation, after which the tubes containing the supernatants were ␤ integrin engagement rotated in a cold room for 1 h. Thereafter, 40 ␮l of a 50% slurry of protein 2 A-agarose was added, and the incubation was continued for an additional ␤ Direct engagement of 2 integrins was accomplished by plating the cells on 45 min. The collected immune complexes were washed three times in ␤ a surface coated with anti- 2 integrin Abs (7, 23). Briefly, petri dishes were ice-cold wash buffer (50 mM HEPES (pH 7.5), 500 mM M NaCl, 0.1% incubated with poly-L-lysine (0.1 mg/ml in PBS) for 30 min and then Triton X-100, 0.005% SDS, and 5 mM MgCl2). The nucleotides bound to washed twice with PBS. Thereafter, glutaraldehyde (2.5% in PBS) was these immune complexes were eluted at 68°C for 20 min using a medium added, and the plates were allowed to stand for 15 min and then washed consisting of 5 mM DTT, 5 mM EDTA, 0.2% SDS, 0.5 mM GTP, and 0.5 twice with PBS. All subsequent steps were performed at 4°C. Soluble mM GDP. The radiolabeled nucleotides bound to RhoA were resolved by protein A dissolved in PBS (0.1 mg/ml) was added, and 5 h later the plates TLC using 78 g/L ammonium formate and 9.6% (v/v) concentrated HCl The Journal of Immunology 6313

(25) and then were quantified with a PhosphorImager (Molecular Dynam- 1.26 Ϯ 0.09, 1.48 Ϯ 0.19, and 1.86 Ϯ 0.26 in cells plated for 10, ics, Sunnyvale, CA) as previously described (7, 26). 20, and 30 min, respectively (n ϭ 4). We refrained from examining 32 Measurement of RhoGAP activity accumulation of P-labeled guanine nucleotides bound to RhoA for longer than 30 min due to dilution of the labeled GDP/GTP Neutrophils were plated on a surface coated with the anti-CD18 Ab IB4 for pool. We calculated for each time point of the kinetic the ratio of different periods of time and then lysed as described above. The ϩ p190RhoGAP molecules in each sample were immunoprecipitated (see GTP:GTP GDP recovered on RhoA, an index of RhoA activa- above) with an anti-p190RhoGAP Ab (3 ␮g/ml) for 2 h at 4°C. The im- tion. As shown in Fig. 1B, there was a 2.3-fold increase in the mune complexes were captured on protein G-agarose beads and washed GTP:GTP ϩ GDP ratio recovered on RhoA in cells that were three times with ice-cold wash buffer (50 mM Tris-HCl (pH 7.5), 1 mg/ml plated for 10 min on an IB4-coated surface compared with non- BSA, 10 mM MgCl2, and 1 mM DTT). Simultaneously, purified RhoA (30 ng) was loaded with [␥-32P]GTP by incubation in a buffer (50 mM Tris- adherent control cells. However, plating neutrophils on this surface HCl (pH 7.5), 50 mM NaCl, 5 mM EDTA, 1 mg/ml BSA, 1 mM DTT, and for Ͼ10 min did not induce any further significant increase in the 1 ␮M GTP (10 ␮Ci of [␥-32P]GTP)) for 10 min at 37°C. The immuno- GTP:GTP ϩ GDP ratio recovered on RhoA. In contrast, plating precipitates were then resuspended in 100 ␮l of wash buffer and incubated neutrophils on a surface coated with a control IgG2a Ab did not for 10 min at 25°C with 30 ng of RhoA protein (preloaded with result in any increase in the accumulation of either [32P]GTP or [␥-32P]GTP). This reaction was stopped by adding 0.4 ml of ice-cold stop [32P]GTP-bound RhoA compared with nonadherent control cells buffer (500 mM Tris-HCl (pH 7.5), 10 mM MgCl2, and 1 mM DTT). The samples were centrifuged (2000 ϫ g, 10 s), and aliquots (0.25 ml) of the (Fig. 1A). In addition, there was a significant increase in the GTP: supernatant were filtered through nitrocellulose filters (0.45 ␮m). The fil- GTP ϩ GDP ratio recovered on RhoA in neutrophils in suspension ters were subsequently washed five times with the stop buffer described that had their ␤ integrins engaged (0.086) compared with control above. The retained radioactivity was determined in a liquid scintillation 2 counter. cells in suspension (0.018; duplicate experiment). We also evalu- Downloaded from ␤ ated RhoA activity by determining the effect of 2 integrin en- Determination of the translocation of RhoA, p190RhoGAP, and gagement on the translocation of RhoA to a membrane-enriched p120RasGAP fraction (Fig. 1, C and D). These experiments were based on the Neutrophils were stimulated as described in the figure legends and then well-known fact that translocation of RhoA and other small GTP- directly disrupted by nitrogen cavitation. Briefly, cells were suspended (in binding proteins to the plasma membrane is related to the activa- 100 mM Tris-HCl (pH 7.5); 5 mM EDTA; 5 mM EGTA; 50 mM NaCl; 5 tion and function of these molecules (13, 14). Plating neutrophils http://www.jimmunol.org/ mM NaF; 1 mM Na VO ;20␮g/ml aprotinin; 1 ␮g/ml each of pepstatin, 3 4 for 30 min on a surface coated with the IB4 Ab resulted in a 2-fold leupeptin, and antipain; 2.5 mM benzamidine; and 2 mM Pefabloc), placed in a cell disruption bomb at 4°C (27), equilibrated at 1000 lb/in.2 for 10 increase in accumulation of RhoA in the membrane fraction (Fig. ␤ min, and then rapidly released. The suspension was collected, and nuclei, 1C), and a similar increase was seen upon engagement of the 2 heavy membrane fractions, and undisrupted cells were sedimented by cen- integrins on nonadherent neutrophils (Fig. 1D). In parallel, there ϫ trifugation at 10,000 g for 10 min. The resulting supernatant was cen- was a slight decrease in RhoA in the cytosol (data not shown). trifuged at 100,000 ϫ g for 1 h. The obtained pellet contained a crude membrane fraction, which was resuspended in the disruption buffer de- scribed above. The protein content was determined (28), and aliquots of the ␤ suspension were mixed with the Laemmli buffer supplemented with 50 mM 2 integrin-induced modulation of the RhoGAP activity of

DTT and boiled. The proteins were separated on either 7.5 or 12% SDS- p190RhoGAP by guest on September 26, 2021 PAGE, and immunoblot analysis was performed as described above, using ␮ The rapid hydrolysis of GTP-bound RhoA that occurs following an anti-RhoA antiserum (1 g/ml), anti-p190RhoGAP mAb (1/750), or ␤ anti-p120RasGAP mAb (1 ␮g/ml). engagement of neutrophil 2 integrins may be due to an increase in overall RhoGAP activity in the cells. To explore that hypothesis, Measurement of the amount of the membrane marker HLA Cl-1 anti-p190RhoGAP immunoprecipitates were tested for their ability protein to enhance the GTPase activity of RhoA in vitro. Neutrophils were The amount of HLA Cl-1 was determined in membrane fractions using the plated for 0–40 min on a surface coated with the IB4 Ab. There- mixed ELISA for class I Ag as previously described (29). after, the cells were lysed, and anti-p190RhoGAP immunoprecipi- tates were obtained from the different samples. The ability of the Results immunoprecipitates to promote GTP hydrolysis was determined ␤ 2 integrin engagement induced activation of RhoA by adding purified, bacterially expressed RhoA preloaded with ␤ ␥ 32 Engagement of leukocyte 2 integrins is known to be associated [ - P]GTP. In this assay an increase in RhoGAP activity is de- with profound alteration of the cytoskeleton. Moreover, small GT- tected as a decrease in ␥-32P label bound to RhoA, and we ex- Pases have been strongly implicated in cytoskeletal changes pressed RhoGAP activity of anti-p190RhoGAP immunoprecipi- brought about by integrins (30). Therefore, we examined the pos- tates as a percentage of the total [␥-32P]GTP hydrolyzed during the ␤ sibility that clustering of neutrophil 2 integrins regulates the ac- 10-min incubation period. We found a time-dependent increase in tivity of the small GTPase RhoA (the dominating Rho protein in RhoGAP activity in the immunoprecipitates when neutrophils neutrophils). We evaluated RhoA activity in two ways. First, we were plated on an IB4-coated surface. The maximum RhoGAP measured accumulation of radiolabeled guanine nucleotides bound activity was found in anti-p190RhoGAP immunoprecipitates from to RhoA in neutrophils that had been labeled with 32Pi and sub- cells that were plated for 30–40 min (ϳ2.5-fold increase over ␤ jected to engagement of 2 integrins (Fig. 1). We found both nonadherent control cells; Fig. 2A). As indicated in Fig. 3, plating [32P]GTP and [32P]GDP bound to RhoA in nonadherent control neutrophils on a surface coated with anti-CD18 Ab did not affect cells. There was a 20-fold greater amount of [32P]GDP bound to the amount of p190RhoGAP immunoprecipitated. RhoA vs [32P]GTP bound to RhoA in resting cells (Fig. 1A). En- In Fig. 2B, we tested how pretreatment of neutrophils with PP1, ␤ gagement of 2 integrins, achieved by plating neutrophils on a a selective inhibitor of Src family tyrosine kinases (31), would ␤ surface coated with anti-CD18 Abs, induced a time-dependent in- affect 2 integrin-induced activation of RhoGAP activity of crease in both [32P]GDP and [32P]GTP bound to RhoA (Fig. 1A). p190RhoGAP. We found a significant (1.9-fold) increase in The fold increases in [32P]GTP bound to RhoA vs nonadherent RhoGAP activity in the immunoprecipitates from neutrophils Ϯ Ϯ Ϯ ␤ control cells were 4.3 1.0, 5.0 1.8 and 7.9 3.9 in cells plated plated for 30 min on a surface coated with an anti- 2 integrin Ab, for 10, 20, and 30 min, respectively (n ϭ 4). The fold increases in but no rise in activity in immunoprecipitates from control neutro- [32P]GDP bound to RhoA vs nonadherent control cells were phils plated on a surface coated with an IgG2a Ab (Fig. 2B, lower ␤ 6314 2 INTEGRIN REGULATION OF p190RhoGAP Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

␤ 32 32 FIGURE 1. 2 integrin-induced effects on the accumulation of [ P]GTP and [ P]GDP bound to RhoA and on the translocation of RhoA to a membrane fraction. Suspended neutrophils were incubated for 2 h with [32P]orthophosphate and then plated on surfaces coated with an anti-CD18 Ab for the indicated periods of time (A and B). Neutrophils were lysed, and clarified lysates were used for immunoprecipitation with a polyclonal anti-RhoA Ab. Guanine nucleotides bound to RhoA were eluted and separated by TLC, and TLC plates were subjected to PhosphorImager analysis. A, A representative experiment. The positions of GTP and GDP standards are indicated to the left. The amounts of RhoA immunoprecipitated were analyzed by densitometry for each time point. The results were (in arbitrary units): 23.3 Ϯ 5.4 for 0 min, 25.2 Ϯ 2.1 for 10 min, 28.1 Ϯ 4.2 for 20 min, and 23.3 Ϯ 2.5 for 30 min (n ϭ 3). B, Densitometric analysis of the time kinetics of the GTP:GTP ϩ GDP ratio recovered on RhoA. The data are the mean Ϯ SEM of four separate experiments ␤ ␤ (each performed in duplicate). C and D, RhoA translocation following 2 integrin engagement was measured. Engagement of 2 integrins was achieved by plating the cells on a surface coated with the anti-human CD18 Ab (C) or by incubating suspended neutrophils with mouse anti-human CD18 Ab and subsequently adding RAM IgGs (D). The neutrophils were subsequently disrupted by nitrogen cavitation, and membrane fractions were prepared. The membrane proteins were resolved on 12% SDS-PAGE, transferred to PVDF membranes, and immunoblotted with an anti-RhoA Ab. The top panels in C and D depict representative Western blots; the arrow indicates the position of RhoA. The lanes in the blot are in the same order as the bars in the lower panel. The lower parts of C and D show the results of the same type of experiments after densitometric analysis of each individual blot. Amounts of RhoA are expressed as a percentage of that in unstimulated control cells and represent the mean Ϯ SEM of 14 (C)or20(D) separate experiments. Statistical .(p Ͻ 0.001 (by paired Student’s t test ,ءءء ;p Ͻ 0.01 ,ءء ;p Ͻ 0.05 ,ء :significance vs control

␤ ␤ panel). Furthermore, the 2 integrin-induced increase in RhoGAP on FN, which has been shown to be a natural ligand for 2 inte- activity was abolished in cells pretreated with PP1 (Fig. 2B, lower grins in the presence of TNF-␣ (24) (Fig. 2C). Next, we tested panel). Control experiments revealed that the [␥-32P]GTP hydro- whether activation of p190RhoGAP was a direct consequence of ␤ lyzed in nonspecific immunoprecipitates represented only 2.1% of 2 integrin signaling or was due to cell spreading. To this end, the total amount of RhoA-associated [␥-32P]GTP (data not shown). neutrophils were plated for 30 min on a surface coated with anti- As shown in Fig. 2B (top panel), the amount of p190RhoGAP CD59 Abs, after which RhoGAP activity in anti-p190RhoGAP protein that was immunoprecipitated was not affected by the dif- immunoprecipitates was measured as described above. We found ␤ ferent experimental conditions. We also confirmed the ability of 2 indications that more p190RhoGAP was immunoprecipitated from integrins to activate p190RhoGAP in neutrophils by plating them cells plated on an anti-CD59 Ab-coated surface. Therefore, we The Journal of Immunology 6315

have normalized the RhoGAP activity of p190RhoGAP immuno- precipitates in Fig. 2C to the amount of p190RhoGAP measured by Western blot analysis. It is clear that plating neutrophils on an FN-coated, but not on an anti-CD59-coated, surface induced acti- vation of p190RhoGAP. ␤ 2 integrin engagement did not induce tyrosine phosphorylation of p190RhoGAP ␤ After discovering that PP1 blocked the 2 integrin-induced in- crease in p190RhoGAP activity, we addressed the question of ␤ whether engagement of 2 integrins affects the tyrosine phosphor- ylation status of p190RhoGAP. Neutrophils were plated on an anti- CD18-coated surface for different periods of time and then lysed, and p190RhoGAP was subsequently immunoprecipitated with a specific Ab. Anti-phosphotyrosine immunoblotting of these immu- noprecipitates indicated that the basal level of tyrosine phosphor- ␤ ylation of p190RhoGAP was not altered by clustering of 2 inte- grins, and it was also clear that the basal phosphorylation level of

p190RhoGAP was not affected by PP1 (Fig. 3A, top panel). In Downloaded from parallel, control experiments revealed that plating neutrophils on an anti-CD18-coated surface resulted in a time-dependent increase in the overall tyrosine phosphorylation of proteins, an effect that was totally blunted by pretreatment with PP1 (Fig. 3A, bottom panel).

To complement the results presented in Fig. 3A, we investigated http://www.jimmunol.org/ tyrosine phosphorylation of p190RhoGAP following Ab-induced ␤ engagement of 2 integrins on neutrophils in suspension (a situa- tion mimicking adhesion-induced integrin clustering) and again found no increase in the indicated phosphorylation (Fig. 3B, top panel). In parallel, control experiments revealed that engagement ␤ of 2 integrins on suspended cells resulted in an increase in the overall tyrosine phosphorylation of proteins, an effect that was totally blocked by pretreatment with PP1 (Fig. 3B, bottom panel). by guest on September 26, 2021 ␤ 2 integrin engagement induced tyrosine phosphorylation of p190RhoGAP-associated proteins It is known that p190RhoGAP can associate with phosphoproteins such as p120RasGAP (18). In fact, p190RhoGAP was originally reported to be a tyrosyl-phosphorylated protein that coprecipitated with p120RasGAP (15). Accordingly, we performed experiments to determine whether increased tyrosine phosphorylation of a

Thereafter, the amount of radioactivity bound to RhoA was determined by filtration and liquid scintillation counting. The data are expressed as a percentage of [32P]GTP hydrolyzed for each anti-p190RhoGAP immuno- precipitates (A). The data are the mean Ϯ SEM of three separate experi- ments (each performed in duplicate). B, Neutrophils were plated for 30 min on surfaces coated with either an anti-CD18 Ab (IB4) or a negative control IgG2a Ab. These incubations were performed in the absence or the pres- ence of the specific Src family tyrosine kinase inhibitor PP1 (3 ␮M). There- after, RhoGAP activity associated with anti-p190RhoGAP immunoprecipi- tates was measured as described in A. The data are the mean Ϯ SEM of four or five separate experiments (each performed in triplicate). The in- serted blot shows that equal amounts of p190RhoGAP are immunoprecipi- tated under the different conditions. The lanes in the blot are in the same order as the bars in the lower panel. C, Neutrophils were plated for 30 min on surfaces coated with either FN or an anti-CD59 Ab. When plated on a FN-coated surface, TNF-␣ was added (20 ng/ml). Thereafter, the relative ␤ FIGURE 2. 2 integrin-induced modifications of the GAP activity of RhoGAP activity in anti-p190RhoGAP immunoprecipitates was measured p190RhoGAP for RhoA. Neutrophils were incubated for the indicated as follows. RhoGAP activity in anti-p190RhoGAP immunoprecipitates periods of time on surfaces coated with an anti-CD18 Ab (IB4). Neutro- was determined as described in A and normalized to the amount of phils were lysed, and clarified lysates were used for immunoprecipitation p190RhoGAP recovered in each immunoprecipitate. The data are the mean Ϯ with an anti-p190RhoGAP Ab. The immunoprecipitates were incubated for SEM of three separate experiments (each performed in triplicate). Statistical .(p Ͻ 0.01 (by unpaired Student’s t test ,ءء ;p Ͻ 0.05 ,ء :min with purified RhoA protein that was preloaded with [␥-32P]GTP. significance vs control 10 ␤ 6316 2 INTEGRIN REGULATION OF p190RhoGAP

p190RhoGAP-associated protein could explain the effect of PP1 ␤ on 2 integrin-induced activation of p190RhoGAP. This was ac- complished by analyzing anti-phosphotyrosine immunoprecipi- ␤ tates of neutrophils that had had their 2 integrins engaged for ␤ different periods of time by plating on an anti- 2-integrin Ab- coated surface (Fig. 4) or by exposure to the Ab in suspension (Fig. 5). Immunoblotting of the anti-phosphotyrosine immunoprecipi- tates with an anti-p190RhoGAP Ab showed a single band at 190 kDa (Figs. 4A and 5A). A representative time course of the effects ␤ of 2 integrins on the presence of p190RhoGAP in anti- phosphotyrosine immunoprecipitates is depicted in Fig. 4B. Den- sitometric analysis of such blots from several separate experiments revealed that the amount of p190RhoGAP in these immunopre- cipitates reached a maximum after 30 min (3.3-fold increase over controls) and returned back to control levels after 40 min (Fig. 4B). Pretreatment of neutrophils with PP1 abolished the presence of p190RhoGAP in all immunoprecipitates (Fig. 4C). Similar results ␤ were found after clustering 2 integrins on nonadherent neutro-

phils (Fig. 5). Downloaded from

␤ 2 integrin engagement induced tyrosine phosphorylation of p120RasGAP It is not known whether a p120RasGAP/p190RhoGAP complex (15, 32, 33) exists in neutrophils. To examine that issue, we analyzed anti-p120RasGAP immunoprecipitates of control cells http://www.jimmunol.org/ ␤ and cells with their 2 integrins engaged in the absence or the presence of PP1. Western blotting of these immunoprecipitates with an anti-p190RhoGAP Ab demonstrated that the basal level of p190RhoGAP associated with p120RasGAP was not affected ␤ by either engagement of 2 integrins or pretreatment with PP1 (Fig. 6A, top panel). To ensure that equal amounts of p120RasGAP had been immunoprecipitated, the blot was also

analyzed with an anti-p120RasGAP Ab (Fig. 6A, middle panel). by guest on September 26, 2021 Reprobing and analysis of the immunoprecipitates with an anti- ␤ phosphotyrosine Ab showed that engagement of 2 integrins sig- nificantly increased the low basal level of tyrosine phosphorylation of p120RasGAP and that this phosphorylation was totally blocked by PP1 (Fig. 6A, lower panel). Similarly, plating neutrophils on a surface coated with anti-CD18 Abs resulted in an identical in- crease in p120RasGAP tyrosine phosphorylation, which was to- tally blocked by a pretreatment with PP1 (Fig. 6B, lower panel).

PP1 (3 ␮M). Thereafter, neutrophils were lysed, and clarified lysates were used for immunoprecipitation with an anti-p190RhoGAP Ab (upper and middle panels in A and B). The immunoprecipitates were subjected to 7.5% SDS-PAGE, transferred to a PVDF membrane, and then immunoblotted with an anti-phosphotyrosine Ab (indicated as blot PY) as described in Materials and Methods. The arrow indicates the position of p190RhoGAP. After stripping the blots, the membranes were reblotted with an anti- p190RhoGAP Ab (indicated as blot p190RhoGAP) to assess that equal amounts of p190RhoGAP were immunoprecipitated in the different con- ditions (middle panels in A and B). These blots are representative of eight (A)or10(B) experiments performed. Following densitometric analysis of all blots, the ratio of p190RhoGAP tyrosine phosphorylation vs that of total p190RhoGAP immunoprecipitated was calculated for each time point. A, The values are 0.9 for 0 min, 1.1 for 10 min, 1.0 for 20 min, 0.9 for 30 min, 1.0 for 40 min, and 1.0 for 30 min with PP1. B, The values are 1.2 for ␤ FIGURE 3. Lack of tyrosine phosphorylation of p190RhoGAP follow- controls, 0.9 for cells in which 2 integrins were engaged, and 0.9 for cells ␤ ing 2 integrin engagement. Neutrophils were incubated for the indicated pretreated with PP1 before engaging the integrins. Whole lysate extracts periods of time on surfaces coated with an anti-CD18 Ab (IB4; A), or were also subjected to 7.5% SDS-PAGE, transferred to a PVDF membrane, ␤ engagement of 2 integrins was achieved by incubating suspended neutro- and then immunoblotted with an anti-phosphotyrosine Ab as described phils with mouse anti-human CD18 Ab and subsequently adding RAM IgG above to assess overall tyrosine phosphorylation of proteins (bottom panels

(B). These incubations were performed in the absence or the presence of in A and B). The Mr (in kilodaltons) of proteins is indicated on the left. The Journal of Immunology 6317 Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

␤ FIGURE 4. 2 integrin engagement by adhesion triggers tyrosine phosphorylation of p190RhoGAP-associated proteins. Neutrophils were incubated for the indicated periods of time on surfaces coated with an anti-CD18 Ab (IB4). Incubations were performed in the absence (A and B) or the presence (C) of the specific Src family tyrosine kinase inhibitor PP1 (3 ␮M). Neutrophils were lysed, and clarified lysates were used for immunoprecipitation with an anti-phosphotyrosine Ab. The immunoprecipitates were subjected to 7.5% SDS-PAGE, transferred to a PVDF membrane, and then immunoblotted with an anti-p190RhoGAP Ab as described in Materials and Methods. Densitometric analysis of the time course of tyrosine phosphorylation of p190RhoGAP- associated proteins is illustrated in B. The data are expressed as the mean Ϯ SEM of three to eight separate experiments. Statistical significance vs control: .(p Ͻ 0.001 (by unpaired Student’s t test ,ءءء ;p Ͻ 0.01 ,ءء

␤ 2 integrin engagement induced relocalization of p190RhoGAP an impact on the amount of the membrane marker HLA Cl-1 in and p120RasGAP to a membrane fraction the analyzed membrane fractions, which agrees well with pre- It has been reported that modification of the activity of GAP pro- viously published data (35). teins is associated with relocalization of these molecules (20, 34). ␤ Therefore, we investigated the effects of engagement of neu- Effect of PP1 on 2 integrin-induced activation of RhoA in vivo ␤ trophil 2 integrins on accumulation of p120RasGAP and Additional experiments were performed to determine whether the ␤ p190RhoGAP in a membrane fraction. We found that engage- 2 integrin-induced effects on p190RhoGAP and p120RasGAP ment of the integrins led to an ϳ1.7-fold increase in were related to regulation of RhoA activity in vivo. To this end, p190RhoGAP in the membrane fraction, and this effect was neutrophils were or were not pretreated with PP1 and then plated blocked by PP1 (Fig. 7A); the same was seen for p120RasGAP for 30 min on a surface coated with the IB4 Ab. Thereafter, we (Fig. 7B). Similar results were obtained by plating the neutro- measured the level of radiolabeled guanine nucleotides bound to phils on an anti-CD18 Ab-coated surface (data not shown). In RhoA. As shown in Fig. 8, in the presence of PP1 the ratio of ␤ ϩ Ͻ contrast, PP1 did not affect the 2 integrin-induced transloca- GTP:GTP GDP on RhoA was increased by 28% ( p 0.01); tion of RhoA to the membrane fraction (Fig. 7C). As shown in albeit small, this increase is in the same range as that found by ␤ Fig. 7D, neither 2 integrin engagement nor PP1 treatment had other investigators (36). Thus, activation and translocation of ␤ 6318 2 INTEGRIN REGULATION OF p190RhoGAP

We measured the amounts of 32P-labeled guanine nucleotides associated with immunoprecipitated RhoA, and our results dem- onstrate that both [32P]GDP and [32P]GTP were bound to RhoA in resting suspended neutrophils, with RhoA-[32P]GDP being the ␤ predominant guanine nucleotide-bound form. Engagement of 2 integrins on neutrophils induced a time-dependent increase in both [32P]GTP and [32P]GDP bound to RhoA. The increase in RhoA- 32 ␤ associated [ P]GTP agrees with our previous finding that 2 in- tegrin engagement causes phosphorylation of the exchange factor Vav (7) and also with the recent conclusion that Vav-2 is a GEF ␤ for the RhoA subfamily (37). Engagement of 2 integrins induced a rapid activation of RhoA as reflected by the 2.5-fold increase in the GTP:GTP ϩ GDP ratio recovered on RhoA in neutrophils that were plated for 10–30 min on an IB4-coated surface. The GTP: GTP ϩ GDP ratio was maximum in neutrophils that had been plated for 10 min on an anti-CD18 Ab surface. Despite the fact that there was a gradual increase in GTP-bound RhoA over the time kinetic, we could not observe any further increase in this ratio

when the neutrophils were plated for longer periods of time on an Downloaded from anti-CD18 Ab-coated surface. In a resting cell most members of the Rho family are maintained as cytosolic complexes with Rho-GDP dissociation inhibitor pro- tein (38, 39). Stimulation with an agonist causes these complexes to dissolve, whereupon Rho proteins are released and translocated

to their site of action at the plasma membrane (13, 14, 40, 41). http://www.jimmunol.org/ Linking of these molecules to the membrane is associated with an exchange of GDP for GTP; hence the Rho-membrane interaction may be controlled by membrane-associated GEFs (13, 14). It is also possible that recruitment of Rho to the plasma membrane involves other targets for activated Rho (42–46). If such move- ment of Rho does occur, it is considered to reflect activation of this GTPase regardless of what mechanism(s) is involved in the trans- location of Rho-GTP-binding proteins to the membrane and de- spite the finding that such translocation is probably not required for by guest on September 26, 2021 formation of stress fibers (41, 47). In support of our previously ␤ ␤ FIGURE 5. Engagement of 2 integrins on suspended neutrophils trig- mentioned data in this study we found that engagement of 2 in- gers tyrosine phosphorylation of p190RhoGAP-associated proteins. En- tegrins in both adherent and nonadherent neutrophils led to in- ␤ gagement of 2 integrins was accomplished by incubating suspended neu- creased accumulation of RhoA in the membrane fraction. ␮ trophils with mouse anti-human CD18 Ab (IB4; 10 g/ml) and In subsequent experiments, we investigated the mechanisms by subsequently adding RAM IgGs as described in Materials and Methods. ␤ which clustering of 2 integrins could give rise to a rapid accu- This was done in the absence or the presence of the Src family specific 32 tyrosine kinase inhibitor PP1 (3 ␮M). The neutrophils were lysed, and mulation of RhoA-bound [ P]GDP. We (present study) and others clarified lysates were used for immunoprecipitation with an anti- (48, 49) propose that this can be explained by the existence of an phosphotyrosine Ab. The immunoprecipitated proteins were subjected to inducible intrinsic GTPase activity in leukocytes, whereby the ex- 7.5% SDS-PAGE, transferred to a PVDF membrane, and then immuno- change of GDP for GTP on RhoA is immediately counteracted by blotted with an anti-p190RhoGAP Ab (A) as described in the legend to Fig. conversion of bound GTP to GDP. Accordingly, we were eager to 4. A, A representative experiment. B, Densitometric analysis of the level of determine whether a relationship exists between accumulation of tyrosine phosphorylation of p190RhoGAP-associated proteins (percentage [32P]GDP-bound RhoA (which reflects a rapid turnover of guanine Ϯ of unstimulated cells). The data in this panel are expressed as the mean nucleotides on RhoA) and activation of p190RhoGAP, a GAP pro- ءء SEM of seven separate experiments. Statistical significance vs control: , tein for RhoA (16, 17). We found that ␤ integrin engagement p Ͻ 0.001 (by unpaired Student’s t test). 2 ,ءءء ;p Ͻ 0.01 induced a rise in the RhoGAP activity of anti-p190RhoGAP im- munoprecipitates, an effect that was totally blocked by pretreating p190RhoGAP to the membrane must contribute at least in part to the cells with the selective Src family tyrosine kinase inhibitor ␤ 2 integrin-induced deactivation of RhoA. PP1. In accordance with this, it has been shown that Src family tyrosine kinases, such as p59/61hck, p58c-fgr, and p53/56lyn, play ␤ Discussion essential roles in 2 integrin signaling and regulation of neutrophil The extensive changes in morphology exhibited by neutrophils adherence and spreading (6, 9). Consequently, these tyrosine ki- ␤ ␤ during spreading and locomotion are associated with 2 integrin- nases are conceivable candidates for mediating the 2 integrin- induced intracellular signals that, among other things, cause rear- induced up-regulation of p190RhoGAP activity. However, in neu- rangement of the actin-based cytoskeleton (10). In many cell types trophils plated on an anti-CD18 Ab surface, activation of the small GTPases of the Rho subfamily have been implicated in p190RhoGAP was slow compared with the overall tyrosine phos- cytoskeletal modulations generated by growth factors (11) and in- phorylation of proteins. This raises the possibility that up-regula- ␤ ␤ tegrins (30). However, it is not yet known whether 2 integrins tion of RhoGAP activity is not a direct consequence of 2 integrin actually participate in activation of the Rho family GTPases in signaling, but perhaps is initiated by cytoskeletal rearrangements neutrophils or how such involvement might occur. and cell spreading. However, we can rule out this possibility, since The Journal of Immunology 6319

␤ FIGURE 6. Engagement of 2 integrins on neutro- phils triggers tyrosine phosphorylation of p120RasGAP. ␤ A, Engagement of 2 integrins was accomplished in suspended neutrophils as described in the legend to Fig. 5. B, Neutrophils were plated on a surface coated with anti-CD18 Abs as described in the legend to Fig. 4. The neutrophils were lysed, and clarified lysates were used for immunoprecipitation with an anti-p120RasGAP Ab. The immunoprecipitated proteins were subjected to 7.5% SDS-PAGE, transferred to a PVDF membrane, and then immunoblotted with an anti-p190RhoGAP Ab (upper panels) or an anti-p120RasGAP Ab (middle pan- els) as described in Materials and Methods. After strip- ping the blots, the membranes were reblotted with an

anti-phosphotyrosine Ab (lower panels). The arrows on Downloaded from the right indicate the position of p120RasGAP or p190RhoGAP. http://www.jimmunol.org/

there was no increase in RhoGAP activity in anti-p190RhoGAP (52). Furthermore, p120RasGAP can form a complex with p62 immunoprecipitates from neutrophils spread on an anti-CD59 Ab- (15) and SHC proteins (53). In fibroblasts, p120RasGAP has been ␤ coated surface. Thus, a specific signal emanating from 2 integrins shown to be tyrosine phosphorylated upon EGF stimulation and by is responsible for the activation of p190RhoGAP. overexpression of v-Src (15). Therefore, we investigated the pos- ␤ Various reports have described increased tyrosine phosphoryla- sibility that 2 integrin-induced modification of p190RhoGAP tion of p190RhoGAP in different types of cells stimulated with a activity is due to increased tyrosine phosphorylation of by guest on September 26, 2021 variety of agonists (19, 20, 50, 51). Chang et al. (20) noticed a p120RasGAP. We immunoprecipitated tyrosine-phosphorylated correlation between the cellular level of tyrosine-phosphorylated proteins from a lysate extract using an anti-phosphotyrosine Ab p190RhoGAP and actin stress fiber disassembly in Kϩ cells over- and then blotted the immunoprecipitated fraction with an anti- expressing pp60c-Src, and these investigators suggested that a co- p190RhoGAP Ab. This approach should reflect tyrosine phosphor- valent modification of p190RhoGAP may regulate the GAP activ- ylation of p190RhoGAP-associated proteins, because in our pre- ␤ ity of that protein toward Rho/Rac proteins. In contrast, although vious experiments it was obvious that clustering of 2 integrins did we did detect basal tyrosine phosphorylation of p190RhoGAP in not cause increased tyrosine phosphorylation of p190RhoGAP. We ␤ neutrophils, we found no increase in p190RhoGAP phosphoryla- found that engagement of 2 integrins transiently increased ty- ␤ tion following engagement of 2 integrins, and that was confirmed rosine phosphorylation of a p190RhoGAP-associated protein(s), by measurements of p120RasGAP bound to p190RhoGAP. In- an effect that was totally blocked by PP1. In parallel, clustering of deed, formation of this complex is regulated predominantly by the integrins induced relocalization of p190RhoGAP to a mem- interaction between the phosphorylated tyrosine residue of brane fraction. Interestingly, although the integrin clustering did p190RhoGAP and the SH2 domains of p120RasGAP (32, 33); thus not increase tyrosine phosphorylation of p190RhoGAP, transloca- the amount of the complex indirectly reflects the level of tyrosine tion of this RhoGAP to the membranes was blocked by PP1. Con- ␤ ␤ phosphorylation of p190RhoGAP. We found that 2 integrin en- sequently, it is likely that the 2 integrin-induced relocalization of ␤ gagement did not affect the amount of p120RasGAP associated p190RhoGAP to the membrane is due to 2 integrin-induced ty- with p190RhoGAP in resting neutrophils. Accordingly, our results rosine phosphorylation of a p190RhoGAP-associated protein(s). indicate that tyrosine phosphorylation of p190RhoGAP per se does Simultaneous redistribution of p190RhoGAP and p120RasGAP ␤ not play a role in regulating 2 integrin-dependent up-regulation of into perinuclear concentric arcs has been described in C3H10T1/2 p190RhoGAP activity. In agreement, Roof and coworkers (32) murine fibroblasts overexpressing c-Src after being stimulated detected epidermal growth factor (EGF)-induced phosphorylation with EGF (20) and in mouse fibroblasts during integrin-mediated ␤ of p190RhoGAP only in 10T1/2 cells overexpressing EGF recep- interaction with a substrate (34). We found that engagement of 2 tor and c-Src but not in 10T1/2 Neo control cells. The contrasting integrins induced tyrosine phosphorylation and translocation of results of Schieffer and colleagues (50), who reported angiotensin p120RasGAP to a membrane-enriched fraction. Clearly, it is the II-induced phosphorylation of p190RhoGAP, is most likely ex- phosphorylation of p120RasGAP that regulates translocation of plained by their approach to only isolate anti-phosphotyrosine im- the protein to the membrane, since both these events were com- munoprecipitates to evaluate the phosphorylation statues of pletely blocked by PP1. This finding agrees with the work of Park p190RhoGAP. and Jove (54) showing that one role of tyrosine phosphorylation of As mentioned previously, p190RhoGAP is found associated p120RasGAP is to increase the association of this RasGAP with ␤ with p120RasGAP, and this complex is predominantly cytosolic the membrane. Thus, together these results suggest that the 2 ␤ 6320 2 INTEGRIN REGULATION OF p190RhoGAP Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

␤ FIGURE 7. Engagement of 2 integrins on suspended neutrophils triggers relocalization of p120RasGAP and p190RhoGAP to the membrane. Engage- ␤ ment of 2 integrins was accomplished in suspended neutrophils as described in the legend to Fig. 5. The neutrophils were subsequently disrupted by nitrogen cavitation, and membrane fractions were prepared. The membrane proteins were resolved on either 7.5 or 12% SDS-PAGE, transferred to PVDF membranes, and immunoblotted with anti-p190RhoGAP Ab (A), anti-p120RasGAP Ab (B), or anti-RhoA Ab (C). The top inserts in A–C depict repre- sentative Western blots; the arrows indicate the positions of p190RhoGAP (A), p120RasGAP (B), and RhoA (C). The lanes in the blots are in the same order as the bars in the corresponding panel. The bars show the results of densitometric analysis of each individual blot. Amounts of p120RasGAP, p190RhoGAP, and RhoA are expressed as a percentage of unstimulated control values and are the mean Ϯ SEM of six (A), three (B), or 10 (C) experiments. D, The amount of a membrane marker HLA Cl-1 was measured by a mixed ELISA method (see Materials and Methods). The amounts of HLA Cl-1 protein in membrane fractions are expressed as a percentage of that in control cells and are the mean Ϯ SEM of four separate experiments. Statistical significance .(p Ͻ 0.001 (by unpaired Student’s t test ,ءءء ;p Ͻ 0.01 ,ءء :vs control

integrin-induced tyrosine phosphorylation of p120RasGAP is, di- murine fibroblasts participates in the regulation of Rho activity, ␤ rectly or indirectly, responsible for the 2 integrin-induced activa- since translocation of those proteins to the membrane was accom- tion and translocation of p190RhoGAP to a membrane fraction. panied by EGF-induced disassembly and reassembly of actin stress ␤ In addition to blocking 2 integrin-induced activation and trans- fibers (20). Another important observation is that PP1 did not ␤ location of p190RhoGAP to the membrane, PP1 also increased the block 2 integrin-induced relocalization of RhoA to the mem- ϩ ␤ ratio of GTP to GDP GTP on RhoA from neutrophils with stim- brane, which explains why the 2 integrin-induced accumulation ␤ ulated 2 integrins. Thus, p190RhoGAP appears to be involved in of Rho-bound GTP was not blunted by PP1. The latter finding ␤ ␤ the 2 integrin-induced regulation of RhoA activity in neutrophils suggests that the 2 integrin-induced activation of RhoA guanine in vivo. In support of that, it has been suggested that EGF-induced nucleotide exchange activity is regulated by a signaling pathway relocalization of p120RasGAP and p190RhoGAP in C3H10T1/2 that does not depend on a Src family tyrosine kinase. The Journal of Immunology 6321

13. Bokoch, G. M., B. P. Bohl, and T.-H. Chuang. 1994. Guanine nucleotide ex- change regulates membrane translocation of Rac/Rho GTP-binding proteins. J. Biol. Chem. 269:31674. 14. Boukharov, A. A., and C. M. Cohen. 1998. Guanine nucleotide-dependent trans- location of RhoA from cytosol to high affinity membrane binding sites in human erythrocytes. Biochem. J. 330:1391. 15. Ellis, C., M. Moran, F. McCormick, and T. Pawson. 1990. Phosphorylation of GAP and GAP-associated proteins by transforming and mitogenic tyrosine ki- nases. Nature 343:377. 16. Settleman, J., V. Narasimhan, L. C. Foster, and R. A. Weinberg. 1992. Molecular cloning of cDNAs encoding the GAP-associated protein p190: implications for a signaling pathway from Ras to the nucleus. Cell 69:539. 17. Ridley, A. J., A. F. Self, F. Kasmi, H. F. Paterson, A. Hall, A, C. J. Marshall, and C. Ellis. 1993. Rho family GTPase activating proteins p190, bcr and rhoGAP show distinct specificities in vitro and in vivo. EMBO J. 12:5151. 18. McGlade, J., B. Brunkhorst, D. Anderson, G. Mbamalu, J. Settleman, S. Dedhar, M. Rozakis-Adcock, L. B. Chen, and T. Pawson. 1993. The N-terminal region of GAP regulates cytoskeleton structure and cell adhesion. EMBO J. 12:3073. 19. Nakahara, H., S. C. Mueller, M. Nomizu, Y. Yamada, Y. Yeh, and W.-T. Chen. ␤ 1998. Activation of 1 integrin signaling stimulates tyrosine phosphorylation of p190RhoGAP and membrane-protrusive activities at invadopodia. J. Biol. Chem. 273:9. 20. Chang, J. H., S. Gill, J. Settleman, and S. J. Parsons. 1995. c-Src regulates the simultaneous rearrangement of actin cytoskeleton, p190RhoGAP, and p120RasGAP following epidermal growth factor stimulation. J. Cell Biol. 130: Downloaded from FIGURE 8. Effect of the tyrosine kinase inhibitor PP1 on the ratio of 355. ϩ ␤ 21. Wright, S. D., P. E. Rao, W. C. van Voorhiss, L. S. Craigmyle, K. Iida, GTP:GDP GTP recovered on RhoA following 2 integrin engagement. M. A. Talle, E. F. Westberg, G. Goldstein, and S. C. Silverstein. 1983. Identifi- Suspended neutrophils were incubated for 2 h with [32P]orthophosphate cation of the C3bi receptor of human monocytes and macrophages by using and then pretreated, or not, with PP1 (3 ␮M) for 10 min. Thereafter, cells monoclonal antibodies. Proc. Natl. Acad. Sci. USA 18:5699. were plated for 30 min on surfaces coated with an anti-CD18 Ab (IB4). 22. Bo¨yum, A. 1968. Isolation of leucocytes from human blood: further observations: methylcellulose, dextran, and Ficoll as erythrocyte aggregating agents. Scand. Neutrophils were lysed, and clarified lysates were used for immunopre- J. Clin. Lab. Invest. 97(Suppl.):31. cipitation with an anti-RhoA antiserum. Guanine nucleotides bound to 23. Berton, G., C. Laudanna, C. Sorio, and F. Rossi. 1992. Generation of signals http://www.jimmunol.org/ RhoA were eluted and separated by TLC. The TLC plates were subjected activating neutrophil functions by leukocyte integrins: LFA-1 and gp150/95, but to PhosphorImager analysis. The amounts of GTP and GDP were quanti- not CR3, are able to stimulate the respiratory burst of human neutrophils. J. Cell ϩ Biol. 4:1007. fied as described in the legend to Fig. 1. The ratio of GTP:GTP GDP 24. Fuortes, M., W.-W. Jin, and C. Nathan. 1993. Adhesion-dependent protein ty- recovered on immunoprecipitated RhoA was calculated. The data are the rosine phosphorylation in neutrophils treated with tumor necrosis factor. J. Cell mean Ϯ SEM of six separate experiments performed in duplicate. Statis- Biol. 120:777. p Ͻ 0.01 (by paired Student’s t test). 25. Hallberg, B., S. I. Rayter, and J. Downward. 1993. Interaction of Ras and Raf in ,ءء :tical significance vs control intact mammalian cells upon extracellular stimulation. J. Biol. Chem. 269:3913. 26. Zheng, L., J. Eckerdal, I. Dimitrijevic, and T. Andersson. 1997. Chemotactic peptide-induced activation of Ras in human neutrophils is associated with inhi- bition of p120-GAP activity. J. Biol. Chem. 272:23448. Acknowledgments 27. Klempner, M. S., R. B. Mikkelsen, D. H. Corfman, and J. Andre´-Schwartz. 1980. by guest on September 26, 2021 We are grateful to Dr. Bengt Hallberg for very helpful suggestions, to Dr. Neutrophil plasma membranes. I. High-yield purification of human neutrophil Niels Borregaard and Charlotte Horn for quantification of HLA Cl-1 pro- plasma membrane vesicles by nitrogen cavitation and differential centrifugation. tein, and to Patty O¨ dman for linguistic revision of this manuscript. J. Cell Biol. 86:21. 28. Schaffner, W., and C. Weissman. 1973. A rapid, sensitive, and specific method for the determination of protein in dilute solution. Anal. Biochem. 56:502. References 29. Bjerrum, O. W., and N. Borregaard. 1990. Mixed enzyme-linked immunosorbent 1. Sanchez-Madrid, F., J. A. Nagy, E. Robbins, P. Simon, and T. A. Springer. 1983. assay (MELISA) for HLA class I antigen: a plasma membrane marker. Scand. A human leukocyte differentiation antigen family with distinct ␣-subunits and a J. Immunol. 31:305. common ␤-subunit: the lymphocyte function-associated antigen (LFA-1), the 30. Clark, E. A., W. G. King, J. S. Brugge, M. Symons, and R. O. Hynes. 1998. C3bi complement receptor (OKM1/Mac-1), and the p150,95 molecule. J. Exp. Integrin-mediated signals regulated by members of the Rho family of GTPases. Med. 158:1785. J. Cell Biol. 142:573. 2. Gahmberg, C. G., M. Tolvanen, and P. Kotovuori. 1997. Leukocyte adhesion. 31. Hanke, J. H., J. P. Garden, R. L. Dow, P. S. Changelian, W. H. Brissette, Eur. J. Biochem. 245:215. E. J. Weringer, B. A. Pollok, and P. A. Connelly. 1996. Discovery of a novel, ␤ potent, and Src family-selective tyrosine kinase inhibitor. J. Biol. Chem. 271:695. 3. Berton, G., L. Fumagalli, C. Laudanna, and C. Sorio. 1994. 2 integrin-dependent protein tyrosine phosphorylation and activation of the FGR protein tyrosine ki- 32. Roof, R. W., M. D. Haskell, B. D. Dukes, N. Sherman, M. Kinter, and nase in human neutrophils. J. Cell Biol. 126:1111. S. J. Parsons. 1998. Phosphorylation (p-Tyr)-dependent and -independent mech- 4. Fuortes, M., W. W. Jin, and C. Nathan. 1993. Adhesion-dependent protein ty- anisms of p190 RhoGAP-p120 rasGAP interaction: tyr 1105 of p190, a substrate rosine phosphorylation in neutrophils treated with tumor necrosis factor. J. Cell for c-Src, is the sole p-Tyr mediator of complex formation. Mol. Cell. Biol. Biol. 120:777. 18:7052. 5. Graham, I. L., D. C. Anderson, V. M. Holers, and E. J. Brown. 1994. Comple- 33. Bryant, S. S., S. Briggs, T. E. Smiyhgall, G. A. Martin, F. McCormick, J.-H. ␣ ␤ Chang, S. J. Parsons, and R. Jove. 1995. Two SH2 domains of p120 Ras GTPase- ment receptor 3 (CR3, Mac-1, integrin M 2, CD11b/CD18) is required for ty- rosine phosphorylation of paxillin in adherent and non-adherent neutrophils. activating protein bind synergistically to tyrosine phosphorylated p190 Rho GT- J. Cell Biol. 127:1139. Pase-activating protein. J. Biol. Chem. 270:17947. ␤ 6. Dib, K., and T. Andersson. 2000. 2 integrin signaling in leukocytes. Front. 34. Sharma, S. V. 1998. Rapid recruitment of p120RasGAP and its associated pro- Biosci. v5, d438. tein, p190RhoGAP, to the cytoskeleton during integrin-mediated cell-substrate 7. Zheng, L., A. Sjo¨lander, J. Eckerdal, and T. Andersson. 1996. Antibody-induced interaction. Oncogene 17:271. ␤ engagement of 2 integrins on adherent human neutrophils triggers activation of 35. Sengelov, H., L. Kjeldsen, M. S. Diamond, T. A. Springer, and N. Borregaard. ras ␣ ␤ p21 through tyrosine phosphorylation of the protooncogene product Vav. Proc. 1993. Subcellular localization and dynamics of Mac-1 ( m 2) in human neutro- Natl. Acad. Sci. USA 93:8431. phils. J. Clin. Invest. 92:1467. 8. Hellberg, C., L. Molony, L. Zheng, and T. Andersson. 1996. Ca2ϩ signalling 36. Gohla, A., R. Harhammer, and G. Schultz. 1998. The G-protein G13 but not G12 ␤ ␥ mechanisms of the 2 integrin on neutrophils: involvement of phospholipase C 2 mediates signaling from lysophosphatidic acid receptor via epidermal growth and Ins(1,4,5)P3. Biochem. J. 317:403. factor receptor to Rho. J. Biol. Chem. 273:4653. 9. Lowell, C. A., L. Fumagalli, and G. Berton. 1996. Deficiency of Src family 37. Schuebel, K. E., N. Movilla, J. L. Rosa, and X. R. Bustelo. 1998. Phosphoryla- kinases p59/61hck and p58c-fgr results in defective adhesion-dependent neutrophil tion-dependent and constitutive activation of Rho proteins by wild-type and on- functions. J. Cell Biol. 133:895. cogenic Vav-2. EMBO J. 17:6608. ␤ 10. Lo¨fgren, R., J. Ng-Sikorski, A. Sjo¨lander, and T. Andersson. 1993. 2 integrin 38. Ueda, T., A. Kikuchi, N. Ohga, J. Yamamoto, and Y. Takai. 1990. Purification engagement triggers actin polymerization and phosphatidylinositol trisphosphate and characterization from bovine brain cytosol of a novel regulatory protein formation in non-adherent human neutrophils. J. Cell Biol. 123:1597. inhibiting the dissociation of GDP from and the subsequent binding of GTP to 11. Hall, A. 1998. Rho GTPases and the actin cytoskeleton. Science 279:509. Rho p20, a Ras p21-like GTP-binding protein. J. Biol. Chem. 265:9373. 12. Boguski, M. S., and F. McCormick. 1993. Proteins regulating Ras and its rela- 39. Fukumoto, Y., K. Kaibuchi, Y. Hori, H. Fujioka, S. Araki, T. Ueda, A. Kikuchi, tives. Nature 366:643. and Y. Takai. 1990. Molecular cloning and characterization of a novel type of ␤ 6322 2 INTEGRIN REGULATION OF p190RhoGAP

regulatory protein (GDI) for the Rho proteins, Ras p21-like small GTP-binding 47. Ridley, A. J., and A. Hall. 1992. The small GTP-binding protein Rho regulates proteins. Oncogene 5:1321. the assembly of focal adhesions and actin stress fibers in response to growth 40. Takaishi, K., T. Sasaki, T. Kameyana, S. Tsukita, S. Tsukita and Y. Takai. 1995. factors. Cell 70:389. Translocation of activated Rho from the cytoplasm to membrane ruffling area, 48. Laudanna, C., J. J. Campbell, and E. C. Butcher. 1996. Role of Rho in chemoat- cell-cell adhesions and cleavage furrows. Oncogene 11:39. tractant-activated leukocyte adhesion through integrins. Science 271:981. 41. Fleming, I. N., C. M. Elliot, and J. H. Exton. 1996. Differential translocation of 49. Laudanna, C., J. J. Campbell, and E. Butcher. 1997. Elevation of intracellular Rho family GTPases by lysophosphatidic acid, endothelin-1, and platelet-derived growth factor. J. Biol. Chem. 271:33067. cAMP inhibits RhoA activation and integrin-dependent leukocyte adhesion in- 42. Leung, T., E. Manser, L. Tan, and L. Lim. 1995. A novel serine/threonine kinase duced by chemoattractants. J. Biol. Chem. 272:24141. binding the Ras-related RhoA GTPase which translocates the kinase to peripheral 50. Schieffer, B., W. G. Paxton, Q. Chai, M. B. Marrero, and K. E. Bernstein. 1996. membranes. J. Biol. Chem. 270:29051. Angiotensin II controls p21ras activity via pp60c-Src. J. Biol. Chem. 271:10329. 43. Matsui, T., M. Amano, T. Yamamoto, K. Chihara, M. Nakafuku, M. Ito, 51. Auricchio, A., M. di Domenico, G. Castoria, A. Bilancio, and A. Migliaccio. T. Nakano, K. Okawa, A. Iwamatsu, and K. Kaibuchi. 1996. Rho-associated 1994. Epidermal growth factor induces protein tyrosine phosphorylation and as- kinase, a novel serine/threonine kinase, as a putative target for small GTP-binding sociation of p190 with Ras-GTPase activating protein in caco-2 cells. FEBS Lett. protein Rho. EMBO J. 15:2208. 353:16. 44. Amano, M., H. Mukai, Y. Ono, Y., K. Chihara, T. Matsui, Y. Hamajima, 52. Gibbs, J. B., M. D. Schaber, W. J. Allard, L. S. Sigal, and E. M. Scolnick. 1988. K. Okawa, A. Iwamatsu, and K. Kaibuchi. 1996. Identification of a putative target Purification of Ras GTPase activating protein from bovine brain. Proc. Natl. for Rho as the serine-threonine kinase protein kinase N. Science 271:648. Acad. Sci. USA 85:5026. 45. Ren, X. D., G. M. Bokoch, A. Traynor-Kaplan, G. H. Jenkins, R. A. Anderson, and M. A. Schwartz. 1996. Physical association of the small GTPase Rho with a 53. Dupont, H., and M. Blancq. 1999. Formation of complexes involving RasGAP 68-kDa phosphatidylinositol 4-phosphate 5-kinase in Swiss 3T3 cells. Mol. Biol. and p190 RhoGAP during morphogenetic events of the gastrulation in Xenopus. Cell 7:435. Eur. J. Biochem. 265:530. 46. Bowman, E. P., D. J. Uhlinger, and J. D. Lambeth. 1993. Neutrophils phospho- 54. Park, S., and R. Jove. 1993. Tyrosine phosphorylation of Ras GTPase-activating lipase D is activated by a membrane-associated Rho family small molecular protein stabilizes its association with p62 at membranes of V-Src transformed weight GTP-binding protein. J. Biol. Chem. 268:21509. cells. J. Biol. Chem. 268:25728. Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021