HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 29, 22590-22596, July 21, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/29/22590    most recent M000633200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kinashi, T. Articles by Takatsu, K. Search for Related Content PubMed PubMed Citation Articles by Kinashi, T. Articles by Takatsu, K. Social Bookmarking                      What's this?

Distinct Mechanisms of ₅₁ Integrin Activation by Ha-Ras and R-Ras^*

Tatsuo Kinashi^§¶, Koko Katagiri^§, Shin-ichi Watanabe^§, Bart Vanhaesebroeck, Julian Downward^**, and Kiyoshi Takatsu

From the  Department of Immunology, Institute of Medical Science, University of Tokyo, Tokyo 108, ^§ Bayer-chair, Department of Molecular Immunology and Allergy, Graduate School of Medicine, Kyoto University, Kyoto 606 Japan, the  Cell Signalling Laboratory, Ludwig Institute for Cancer Research, London, W1P 8BT, United Kingdom, and the ^** Signal Transduction Laboratory, Imperial Cancer Research Fund, London, WC2A 3PX, United Kingdom

Received for publication, January 27, 2000, and in revised form, April 28, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

To investigate the possible roles of the Ras/Rho family members in the inside-out signals to activate integrins, we examined the ability of Ras/Rho small GTPases to stimulate avidity of ₅₁ (VLA-5) to fibronectin in bone marrow-derived mast cells. We found that both Ha-Ras^Val-12 and R-Ras^Val-38 had strong stimulatory effects on adhesion and ligand binding activity of VLA-5 to fibronectin. However, only Ha-Ras^Val-12-, but not R-Ras^Val-38-induced adhesion was inhibited by wortmannin, which suggests that Ha-Ras^Val-12 is dependent on phosphatidylinositol (PI) 3-kinase on adhesion whereas R-Ras^Val-38 has another PI 3-kinase independent pathway to induce adhesion. The effector loop mutant Ha-Ras^Val-12E37G, but not Y40C retained the ability to stimulate adhesion of mast cells to fibronectin. Consistently, PI 3-kinase p110, predominantly expressed in mast cells, interacted with Ha-Ras^Val-12 E37G, but not Y40C, which was also correlated with the levels of Akt phosphorylation in mast cells. Furthermore, marked adhesion was induced by a membrane-targeted version of p110. These results indicate that Ha-Ras^Val-12 activated VLA-5 through PI 3-kinase p110. The mutational effects of the R-Ras effector loop region on adhesion were not correlated with PI 3-kinase activities, consistent with our contention that R-Ras has a distinct pathway to modulate avidity of VLA-5.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Adhesion mediated by integrins controls cell migration and localization. Adhesive interactions through integrins are modulated by adhesiveness (avidity) as well as expressions of integrins (1, 2). Although both types of regulations are important, avidity modulation of integrins in particular plays a critical role in leukocyte migration and localization during inflammatory responses (3). Several external stimuli were reported to modulate avidity of integrins without changes of integrin expressions, such as antigen, chemokines, and cytokines (3-5). A rapid change of avidity of integrins occurs within minutes and is triggered by intracellular signaling pathways, which are referred to as inside-out signals (6).

Avidity modulation of integrins is regulated by increasing affinity to ligands, or by spatial redistribution of integrins on cell surface, which increases the number of integrins on the contact site (7-10). Which types of avidity regulations are utilized largely depends on stimuli that induce adhesion. PMA¹ enhanced adhesion without detectable change in ligand-binding affinity of integrins (11-13). Recent studies have shown that an increase in lateral diffusion and clustering of integrins by PMA or cytochalasin D at low doses facilitates adhesion (10, 14), suggesting that the adhesion is mediated by low affinity, but multivalent bindings of integrins. On the other hand, affinity modulation in integrins detected with soluble ligands or antibodies recognizing the high affinity state was reported for ₄₁, ₅₁, _L2, and _IIb3 integrins in cells stimulated with activating antibodies, manganese ions, or cross-linking of the T cell receptor (12, 13, 15-19). We previously demonstrated that PMA-stimulated mast cells adhered to fibronectin without accompanying affinity modulation of VLA-5, while FcRI cross-linked mast cells adhered to fibronectin by the high affinity state of VLA-5. Steel factor-induced adhesion was considered to be brought by both mechanisms (20). Changes in the affinity state of integrins influenced cell migratory speeds on substrates (21). However, the physiological significance of two modes of avidity modulation has not yet been demonstrated clearly.

The Ras/Rho family of small GTPases regulates the actin cytoskeleton and contributes to the formation of membrane ruffling and focal adhesion (22, 23). Cytoskeletal reorganization subsequent to attachment to substrate leads to marked cell shape changes and strengthens adhesive interactions. Several members of the Ras/Rho family have been reported to influence integrin-mediated adhesion. Ha-Ras was shown to suppress the active form of _IIb3 chimeras through the MAP kinase pathway (24). A constitutively active R-Ras was found to enhance cellular adhesion to fibronectin by enhancing ₁-integrin ligand-binding affinity (25). We have recently shown that Rap1 has a unique property that causes an increase of ligand binding affinity of the ₂ integrin LFA-1, and that Rap1 was critically involved in T-cell receptor-mediated LFA-1/ICAM-1 adhesion (26). However, there are few comprehensive studies that examine whether or not the Ras/Rho family of small GTPases can modulate avidity of ₁ integrins directly.

To gain a clearer understanding of avidity regulations of ₁ integrins by the Ras/Rho family of small GTPases, we employed bone marrow-derived mast cells as a model system to analyze their ability to modulate avidity of VLA-5, because mast cells have been shown to adhere to fibronectin through VLA-5 upon physiologically relevant stimulation such as steel factor (27), or antigen cross-linking of FcRI (20) and are considered to be suitable for activation signal-dependent adhesion. With mast cells, one can also examine the affinity state of VLA-5 as we demonstrated that with antigen cross-linking of FcRI (20). Here we report that the active mutants of Ha-Ras and R-Ras among the Ras/Rho family member lead to strong adhesion to fibronectin with the high affinity state of VLA-5. Furthermore, our study reveals distinct mechanisms of Ha-Ras and R-Ras in regulation of avidity of VLA-5 through analyses of effector mutants and constitutively active downstream signaling molecules: PI 3-kinase p110 is critical to avidity modulation by Ha-Ras, while R-Ras has other mechanisms to regulate avidity of VLA-5.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Lines, Antibodies, and Chemicals-- Primary bone marrow-derived mast cell cultures were carried out as described (27). Primary mast cells were used from 4 to 10 weeks of culture after establishment. Retrovirus-mediated transfection was employed to introduce cDNAs into mast cells as using GP+E86 packaging cells (28). The anti-mouse VLA-5 monoclonal antibody MFR-5 (5H10-27) (29) was purified by affinity chromatography on protein G-Sepharose (Amersham Pharmacia Biotech). Monoclonal anti-Myc epitope (9E10) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), anti-T7 epitope (Novagen, Madison, WI), rabbit polyclonal anti-Ha-Ras antibody and anti-RalA antibody (Transduction Laboratory, Lexington, KY), anti-HA 12CA5 (Roche Molecular Biochemicals, Indianapolis, IN) antibody, anti-PI 3-kinase p110, p110 (Santa Cruz Biotechnology), p110 (30), anti-PI 3-kinase p85 subunit (Upstate Biotechnology, Inc., Lake Placid, NY), anti-phospho-Akt antibody (Ser⁴⁷³) (New England Biolabs Inc., Beverly, MA), and peroxidase-linked anti-mouse or rabbit antibody (Amersham Pharmacia Biotech) were used for immunoprecipitation and immunoblotting as described below. Wortmannin (Wako Pure Chemical Ltd., Tokyo, Japan), phorbol 12-myristate 13-acetate (PMA), recombinant murine steel factor (Genzyme, Boston, MA), and -estradiol (Sigma) were purchased.

Plasmids-- Constitutively active Ha-Ras, R-Ras, Rap1A, RalA, Rac, and Rho mutants were produced from their cDNAs by a single point mutation; substitution of glycine with valine at position 12 (Ha-Ras^Val-12, Rac^Val-12, and Rap1^Val-12) (Rap1^Val-12 was a gift from Drs. M. Hattori and N. Minato, Kyoto University), position 38 (R-Ras^Val-38), and position 14 (Rho^Val-14, a gift from Dr. S. Narumiya, Kyoto University), substitution of glutamic acid for leucine at position 72 (RalA-E72L, a gift from Dr. H. Koide, Tokyo Institute of Technology). Epitope tags were attached at the amino-terminal end of the mutant small GTPases (Myc epitope tag for Rac^Val-12 and Rho^Val-14, T7 epitope tag for Rap1). Effector loop mutants of Ha-Ras^Val-12 (T35S, E37G, D38E, and Y40C) were described (31). To make effector loop mutants of R-Ras^Val-38, polymerase chain reaction was used with R-Ras^Val-38 as a template to introduce a point mutation; substitution of threonine with serine at position 61 (E61S), glutamic acid with glycine at position 63 (E63G), asparatic acid with glutamic acid at position 64 (D64E), and tyrosine with cysteine at position 66 (Y66C). p110-CAAX was made by attaching the 20 carboxyl-terminal amino acids (KMSKDGKKKKKKSKTKCVIM) of K-Ras as described for Raf-CAAX (32) to the 3' end of p110 by polymerase chain reaction. All cDNAs were verified by sequencing both strands, and subcloned into a retrovirus vector pMX-neo to introduce into mast cells, or pSG5 (Strategene, La Jolla, CA) for COS7 cells. cDNAs encoding for rafER (33), Rlf-CAAX (34), and Raf-CAAX (32) were also subcloned into pMX-neo.

Preparation of Fibronectin and the 80-kDa Fragment, and Integrin Affinity Measurement-- Fibronectin and its 80-kDa tryptic fragment that contains the RGD binding motif for VLA-5 were produced as described (35, 36). The 80-kDa fibronectin fragment was radioiodinated with a modified method using chloramine T (37). The typical specific activity of the labeled 80-kDa fragment used in our experiments was about 3.5 × 10⁸ dpm/nmol. Its binding to cells was measured as described (20). Briefly, mast cells were washed once with binding buffer containing RPMI 1640 (Sigma), 0.1% BSA (Life Technologies, Inc.), and 10 mM HEPES, pH 7.4 (Sigma), and suspended with the same buffer at 1 × 10⁷ cells/ml. In a typical binding assay, performed in a 1.5-ml microcentrifuge tube, 100 µl of cells (1 × 10⁶ cells per tube) were mixed with 100 µl of the radiolabeled 80-kDa fragment. For inhibition with antibodies or wortmannin, mast cells were preincubated with antibodies (20 µg/ml), or wortmannin for 15 min at 25 °C before assays. After incubation for 30 min at 37 °C, samples were oil-separated by centrifugation at 8000 rpm for 1 min. The tip of tubes was amputated from the body with a blade and applied to a -counter to measure radioactivity of the bound (the tip) and the unbound (the body). The nonspecific binding was determined at each data point in the presence of a 50-fold excess of the unlabeled 80-kDa fragment. The specific binding was calculated by subtracting the nonspecific binding from the total binding.

Flow Cytometric Analysis-- Cells (1 × 10⁶) were incubated on ice for 30 min with 50 µl of staining buffer (phosphate-buffered saline, 0.1% BSA, 0.05% sodium azide) containing 1 µg of monoclonal Rat anti-VLA-5 antibody (MFR-5H10) or isotype-matched rat antibody. After washing with staining buffer three times, cells were stained with fluorescein isothiocyanate-labeled anti-Rat IgG as above. The stained cells were analyzed by FACScan (Becton Dickinson, San Jose, CA).

Adhesion Assays-- Assays of adhesion to fibronectin were performed as described (27). Briefly, mast cells labeled with 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) in 96-well plates precoated with fibronectin (1 µg/well) or 1% bovine serum albumin (BSA) were incubated in triplicate at 37 °C for 30 min, or in the presence of PMA as indicated. After washing the plate four times, bound fluorescence was measured with a fluorescence concentration analyzer (IDEXX Laboratories, Westbrook, ME). The level of adhesion was calculated by dividing bound fluorescence by input fluorescence. For the assay with antibodies or wortmannin, labeled mast cells were preincubated at room temperature for 15 min with 20 µg/ml antibodies or wortmannin as indicated before assays.

Western Blot-- Mast cells were prepared for cell lysates as described (28). Equal amounts of protein were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Following SDS-PAGE, the separated proteins were electrophoretically transferred to a polyvinylidene difluoride membrane. After blocking with 5% BSA, the membrane was incubated with antibodies as indicated and detected with the appropriate secondary antibody conjugated with horseradish peroxidase. The bands were visualized using enhanced chemiluminescence (ECL; Amersham Pharmacia Biotech). Stripping and reprobing were performed according to the manufacturer's instructions.

For detection of Myc-tagged Raf-CAAX or, HA-tagged Rlf-CAAX, cell lysates (5 × 10⁶ cells) were immunoprecipitated with anti-Myc (9E10) or anti-HA (12CA5) antibodies and subjected to Western blotting detected with anti-Raf (Upstate Biotechnology, Inc.) or anti-HA antibodies. For detection of p110-CAAX, cells (5 × 10⁵ cells) were suspended with SDS sample buffer, and briefly sonicated and boiled before SDS-PAGE and Western blotting.

Interactions of Ha-Ras Mutants and p110-- Cos7 cells were transfected with Ha-Ras mutants (5 µg) and Myc-tagged p110 (5 µg). After 48 h, cells were harvested and lyzed with lysis buffer (1% Nonidet P-40, 150 mM NaCl, 25 mM Tris, pH 7.4, 10 mM MgCl₂, 15% glycerol, 1 mM phenylmethylsulfonyl fluoride, 0.1 µM aprotinin). Immunoprecipitation was performed using anti-Myc epitope antibody (9E10) or anti-Ras antibody (Y13-238, Oncogene Science, Uniondale NY). Immunocomplexes were collected with protein G-Sepharose (Amersham Pharmacia Biotech) and washed with lysis buffer three times. SDS-PAGE and Western blotting were performed as above. Blots were incubated with anti-p110 antibody (30) or anti-Ras antibody (Transduction Laboratories).

Assay for MAP Kinase Activity-- Mast cells (5 × 10⁶ cells) that introduced rafER were stimulated with estradiol (1 µM) or steel factor (10 units/ml) at 37 °C for 30 min. Cells were lyzed with lysis buffer (1% Triton X-100, 150 mM NaCl, 20 mM Tris, pH 7.6, 1 mM phenylmethylsulfonyl fluoride, 0.1 µM aprotinin). Cell lysates were incubated with 1 µg of anti-ERK2 antibody (Santa-Cruz). Immunocomplexes were collected with protein G-Sepharose. Immune complex kinase assays were performed with myelin basic protein (Sigma) as a substrate (33). The levels of phosphorylation of myelin basic protein were quantitated with a PhosphorImager (BAS1000, Fujifilm, Tokyo, Japan).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The Effects of the Ras/Rho Family of Small GTPases on Adhesion to Fibronectin through VLA-5-- To examine the ability of the Ras/Rho small GTPases to modulate avidity of VLA-5 to fibronectin, we established bone marrow-derived mast cells by culturing bone marrow cells with interleukin-3 for 4 weeks, and then introduced active forms of Ha-Ras (Ha-Ras^Val-12), Rap1 (T7-tagged Rap1^Val-12), RalA (RalA^Leu-72), R-Ras (Myc-tagged R-Ras^Val-38), Rac (Myc-tagged Rac^Val-12), Rho (Myc-tagged Rho^Val-14), or neomycin only by retrovirus. Following drug selection, we examined adhesion of these transfectants to fibronectin. Control mast cells (neo) and uninfected mast cells did not adhere to fibronectin significantly when compared with BSA (Fig. 1). However, cells expressing Ha-Ras^Val-12 and R-Ras^Val-38 adhered strongly to fibronectin without stimulation. In both cases, adhesion to fibronectin was blocked by an anti-VLA-5 antibody, 5H10, indicating that adhesion was mediated by VLA-5 (Fig. 1). Adhesion to fibronectin was also induced slightly in Rap1^Val-12 or Rac^Val-12 expressing cells, which was inhibited with anti-VLA-5 antibody, but RalA^Leu-72 or Rho^Val-14 transfectants did not change the level of adhesion significantly. Expressions of VLA-5 of transfectants were similar to control cells (Neo) (Fig. 2A), or uninfected cells (not shown). We demonstrated expression of introduced active forms of the Ras/Rho small GTPases in mast cells by Western blot analysis (Fig. 2B). These results indicated that adhesiveness of VLA-5 was increased in cells expressing Ha-Ras^Val-12 and R-Ras^Val-38. Transfectants expressing Rap1^Val-12, RalA^Leu-72, Rac^Val-12, or Rho^Val-14 adhered to fibronectin when stimulated with steel factor or PMA, indicating that these active forms of small GTPases did not exert inhibitory effects on activation-dependent adhesion of mast cells (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Adhesion of mast cell transfectants to fibronectin. The levels of adhesion of mast cells expressing genes as indicated were measured in BSA- (open bars), or fibronectin-coated (filled bars) 96-well plates. Hatched bars represent mast cells treated with anti-VLA-5 antibody. Adhesion assays were performed in triplicate as described under "Experimental Procedures." The average and standard error of the triplicate determinations are shown.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   A, expressions of VLA-5 in mast cell transfectants. Mast cells transfected with genes as indicated were stained with control (open) or anti-mouse monoclonal VLA-5 (filled) antibodies followed by fluorescein isothiocyanate-labeled secondary antibody and FACS analysis. B, expressions of active forms of small GTPases transfected into mast cells. Total lysates from cells transfected with the neomycin gene (left lane) and cDNA encoding active forms of small GTPases (right lane) were subjected to SDS-PAGE, transferred to polyvinylidene difluoride filters, and immunoblotted with anti-Ras antibody (Ha-RasVal-12, 21 kDa), anti-T7 antibody (Rap1Val-12, 27 kDa), anti-RalA antibody (RalALeu-72, 24 kDa), anti-R-Ras antibody (R-RasVal-38, 28 kDa), anti-Myc antibody for RacVal-12 (28 kDa), and RhoVal-14 (29 kDa).

Ha-Ras^Val-12 and R-Ras^Val-38 Increase Ligand Binding Activity of VLA-5-- R-Ras^Val-38 was previously shown to augment ligand binding activity to fibronectin (25). To examine whether ligand binding activities of VLA-5 are augmented in cells expressing Ha-Ras^Val-12 and R-Ras^Val-38, we measured ligand-binding affinity using a soluble 80-kDa fibronectin fragment (FN80) containing the RGD motif that was recognized by VLA-5. We previously demonstrated that ligand bindings of unstimulated mast cells was low, but increased by FcRI cross-linking, but not PMA stimulation (20). In mast cells expressing either Ha-Ras^Val-12 or R-Ras^Val-38, ligand bindings were augmented compared with those of control cells (Fig. 3). The level of ligand bindings of cells expressing Ha-Ras^Val-12 was higher than that in R-Ras^Val-38 expressing cells. The increased ligand bindings were inhibited by anti-VLA-5 antibody, indicating that ligand binding activity of VLA-5 was increased in these cells.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   The specific bindings of the 125I-labeled fibronectin 80 fragment. The ligand binding assay was performed in triplicate using 125I-labeled fibronectin 80 (0.2 µg) and mast cells expressing the neomycin gene (open bar), Ha-RasVal-12 (closed bars), or R-RasVal-38 (hatched bars). The specific bindings of mast cells pretreated with the anti-VLA-5 antibody (anti-5, 5H10), or wortmannin at concentrations indicated. The data shown are representative of several experiments with similar results, and the average and standard errors are shown.

To explore whether PI 3-kinase is involved in affinity modulation of VLA-5 by Ha-Ras^Val-12 and R-Ras^Val-38, as is the case in mast cells stimulated with FcRI cross-linking (20), ligand binding assays were performed in the presence of wortmannin. Bindings to FN80 in both Ha-Ras^Val-12 and R-Ras^Val-38 expressing cells were completely inhibited with low doses of wortmannin (Fig. 3) or LY294002 (data not shown), suggesting that PI 3-kinase is involved in affinity modulation by Ha-Ras^Val-12 and R-Ras^Val-38.

Differential Effects of Wortmannin on Adhesion to Fibronectin of Cells Expressing Ha-Ras^Val-12 and R-Ras^Val-38-- Wortmannin also inhibited adhesion of Ha-Ras^Val-12 expressing cells to fibronectin at a concentration similar to those abolished bindings to FN80 (Fig. 4). On the other hand, adhesion of R-Ras^Val-38 expressing cells was resistant to treatment of wortmannin even at 100 nM, the dose of which completely blocked the bindings to FN80 (Fig. 3). LY294002 also failed to inhibit adhesion to fibronectin (data not shown). Anti-VLA-5 inhibited adhesion of R-Ras^Val-38 expressing cells to fibronectin in the presence of wortmannin (Fig. 4). These results indicate that the high affinity state of VLA-5 by PI 3-kinase likely accounted for adhesion induced by Ha-Ras^Val-12, while R-Ras^Val-38-induced adhesions were mostly independent from PI 3-kinase activities, and did not require the high affinity state of VLA-5 for adhesion.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Effects of wortmannin on adhesion to fibronectin of mast cells expressing Ha-RasVal-12 (closed bars) or R-RasVal-38 (hatched bars). Left panel, mast cells were pretreated with dimethyl sulfoxide (0.1%) (DMSO) or indicated amounts of wortmannin (10, 50, and 100 nM) for 15 min before adhesion assays. Right panel, adhesion to fibronectin of R-RasVal-38 expressing mast cells with/without the anti-VLA-5 antibody (anti-5, 5H10). The data are shown as in Fig. 1.

The Effects of Mutations in the Effector Loop Region of Ha-Ras^Val-12 on Adhesion and Akt Phosphorylation-- Point mutations in the effector loop region of the active form of Ha-Ras was reported to selectively inhibit the interaction and activation of downstream signal molecules such as Raf, RalGDS, and PI 3-kinase (38). Raf was shown to interact only with Ha-Ras^Val-12 T35S and Ha-Ras^Val-12 D38E, and RalGDS and PI 3-kinase p110 interacted only with Ha-Ras^Val-12 E37G and Y40C, respectively. To confirm PI 3-kinase dependence of Ha-Ras for adhesion, mast cells expressing Ha-Ras^Val-12, T35S (Ser³⁵), E37G (Gly³⁷), D38E (Glu³⁸), or Y40C (Cys⁴⁰) were established (Fig. 5B). There were no significant changes in surface levels of VLA-5 in these transfectants (data not shown). Contrary to our expectation, cells expressing the Gly³⁷, but not Cys⁴⁰ mutant showed the levels of adhesion equivalent to, or more than, cells expressing Ha-Ras^Val-12, whereas those of adhesion of mast cells expressing the Ser³⁵, Glu³⁸, or Cys⁴⁰ mutants were reduced considerably (Fig. 5A). We also examined the phosphorylation of Akt in effector mutant expressing cells as its phosphorylation is dependent on activities of PI 3-kinase (39). The level of Akt phosphorylation was augmented only in mast cells expressing the Gly³⁷ mutant, the level of which was more than that of the Val¹² mutant (Fig. 5C). The Ser³⁵ slightly increased Akt phosphorylation compared with control (Fig. 5C). The phosphorylation of Akt was completely abolished with treatment of wortmannin, confirming the requirement of PI 3-kinase activity for Akt phosphorylation (data not shown). Thus the levels of Akt phosphorylation were in good correlation with those of adhesion to fibronectin (Fig. 5C), which is consistent with the notion that PI 3-kinase is critically involved downstream of Ha-Ras for adhesion, and suggest that the interaction of Ha-Ras and PI 3-kinase occurs in the Gly³⁷, but not Cys⁴⁰ mutants in mast cells.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Adhesion to fibronectin of mast cells expressing effector loop mutants of Ha-RasVal-12. A, adhesion to fibronectin. Mast cells transfected with the neomycin gene (Neo), Ha-RasVal-12 (V12), or effector loop mutants (Ser35, Gly37, Glu38, and Cys40) in the background of the Val-12 mutation were subjected to adhesion assays without (closed bars) or with 10 ng/ml PMA (hatched bars). The data are shown as in Fig. 1. B, expressions of Ha-Ras mutants in mast cells transfected with the neomycin gene, or Ha-Ras mutants as indicated. C, phosphorylation of Akt. Cell lysates of mast cells expressing the neomycin gene (Neo) or Ha-Ras mutants were analyzed by Western blotting for phosphorylation of Akt by the antibody specific for phosphorylation of serine 473 of Akt (upper panel). The membrane was stripped and reprobed with anti-Akt antibody recognizing both phosphorylated and unphosphorylated Akt (lower panel).

Associations of PI 3-Kinase p110 with Ha-Ras Effector Mutants-- To confirm the possibility that the Gly³⁷, but not Cys⁴⁰ mutant associates with PI 3-kinase in mast cells, we examined the isotypes of the p110 catalytic subunit that were expressed in mast cells. Mast cells expressed predominantly p110, while p110 and  were barely detected (Fig. 6A). The interactions of Ha-Ras effector mutants and Myc-tagged p110 were examined by co-transfection into COS cells and immunoprecipitation with either anti-Myc antibody (Fig. 6B) or anti-Ha-Ras antibody (Fig. 6C) for associations with Ha-Ras mutants or p110, respectively. In both cases, p110 was co-immunoprecipitated with the Gly³⁷ as efficiently as Ha-Ras^Val-12. This result is consistent with strong Akt phosphorylation in the Gly³⁷ expressing mast cells (Fig. 5C).

View larger version (38K): [in this window] [in a new window]   Fig. 6.   Interactions of p110 and Ha-Ras mutants. A, expressions of the catalytic subunits of PI 3-kinase in mast cells. Anti-p85 immunoprecipitates were subjected to Western blotting detected by anti-p110, anti-p110, or anti-p110 antibodies. B and C, co-immunoprecipitation of Ha-Ras mutants and Myc-tagged p110 in COS7 cells. Cell lysates from COS7 cells transfected with a vector only (lane 1), or Myc-tagged p110 together with Ha-RasVal-12 (lane 2), Ser35 (lane 3), Gly37 (lane 4), Glu38 (lane 5), or Cys40 (lane 6), were immunoprecipitated with anti-Myc antibody, 9E10 (B), and anti-Ras antibody (Y13-238) (C). Western blots were probed with polyclonal anti-Ras antibody or anti-p110 antibody (upper panels in B and C), followed by stripping and reprobing with anti-p110 antibody or polyclonal anti-Ras antibody (lower panels in B and C).

p110-CAAX Induces Adhesion to Fibronectin-- To directly demonstrate that p110 itself is sufficient to induce adhesion to fibronectin, we introduced an activated membrane-targeted version of p110, p110-CAAX, into mast cells. We also tested membrane-targeted versions of two known effector molecules that bind to Ha-Ras, Raf-CAAX and Rlf-CAAX (Fig. 7A). Mast cells expressing p110-CAAX strongly adhere to fibronectin without stimulation. In contrast, cells expressing Raf-CAAX or Rlf-CAAX failed to adhere to fibronectin while they responded well to PMA to adhere to fibronectin. As shown in Fig. 7B, mast cells expressing p110-CAAX showed marked cell attachment and spreading on fibronectin compared with control cells (neo). Cell attachment and spreading of p110-CAAX expressing mast cells were comparable to those of Ha-Ras^Val-12, while R-Ras^Val-38 expressing cells tended to spread more on fibronectin (Fig. 7B).

View larger version (50K): [in this window] [in a new window]   Fig. 7.   Effects of Raf-CAAX, Rlf-CAAX, and p110-CAAX on adhesion to fibronectin. A, upper panel, mast cells transfected with Myc-tagged Raf-CAAX, HA-tagged Rlf-CAAX, or p110-CAAX were analyzed for adhesion to fibronectin. Adhesion assays were performed without stimulation (closed bars) or with PMA (hatched bar). The results are shown as in Fig. 1. Lower panel, expressions of Raf-CAAX, Rlf-CAAX, and p110-CAAX (arrow). Lanes 1, 3, and 5, mast cells transfected with the neomycin gene. Lane 2, Raf-CAAX (72 kDa); lane 4, Rlf-CAAX (60 kDa); lane 6, p110-CAAX (110 kDa). B, appearance of adhesion of control mast cells (neo), or mast cells expressing p100-CAAX, Ha-RasVal-12, or R-RasVal-38. The original magnification is 100-fold.

We further examined the effect of activation of the Raf-MAP kinase pathway on adhesion by introducing a chimera of the Raf kinase domain and the hormone-binding domain of the estrogen receptor (rafER) (33). A conditional activation of the Raf kinase activity by estradiol increased the kinase activity of ERK2 more than that by steel factor (Fig. 8B). However, mast cells did not adhere to fibronectin by estradiol, while they responded to steel factor for adhesion to fibronectin (Fig. 8A). Stimulation with estradiol did not affect adhesion to fibronectin by steel factor and PMA (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 8.   Effects of activation of MAP kinase on adhesion to fibronectin. A, mast cells transfected with rafER were unstimulated (), or stimulated with steel factor (SLF, 10 units/ml) or estradiol (ES, 1 µM) for 30 min in adhesion assays. The results are shown as in Fig. 1. B, activation of ERK2. Mast cells unstimulated or stimulated as in A were immunoprecipitated with anti-ERK2 antibody. Activation of ERK2 was measured as described under "Experimental Procedures." Phosphorylation of myelin basic protein (MBP) shown in the middle panel was quantitated with a PhosphorImager (upper panel). The amounts of EKR2 in immunoprecipitates were shown in the lower panel (ERK2).

Mutational Analysis of the Effector Loop of R-Ras-- The differential sensitivity to wortmannin on adhesion induced by Ha-Ras^Val-12 and R-Ras^Val-38 suggests PI 3-kinase independent activation mechanisms of VLA-5 in R-Ras^Val-38-expressing cells. Since R-Ras has the identical amino acid sequences of the effector loop region with Ha-Ras (40), we compared effects of mutations in the Ha-Ras and R-Ras effector loop on adhesion to fibronectin. We introduced R-Ras^Val-38 effector mutants that carry the same replacement mutations as Ha-Ras at the corresponding sites in the background of the Val³⁸ mutation. Established mast cells expressed comparable amounts of R-Ras mutants (Fig. 9B). When they were subjected to adhesion assays, only the Glu⁶⁴ mutant of R-Ras^Val-38 lost the ability to stimulate adhesion to fibronectin (Fig. 9A), while the Ser⁶¹, Gly⁶³, or Cys⁶⁴ expressing cells still adhered to fibronectin. On the other hand, the levels of Akt phosphorylation were not in correlation with adhesion, and the mutations in the effector loop reduced Akt phosphorylation to the comparable degrees in all effector mutants (Fig. 9C). This result was in contrast to that of Ha-Ras^Val-12, in which the levels of adhesion paralleled with PI 3-kinase activities, and supports PI 3-kinase independent mechanisms of R-Ras^Val-38 in stimulating adhesion to fibronectin.

View larger version (24K): [in this window] [in a new window]   Fig. 9.   Adhesion to fibronectin of mast cells expressing effector loop mutants of R-RasVal-38. A, adhesion to fibronectin. Mast cells transfected with the neomycin gene (Neo), R-RasVal-38 (V38), or effector loop mutants (Ser61, Gly63, Glu64, and Cys66) in the background of the Val-38 mutation were subjected to adhesion assays without (closed bars) or with 10 ng/ml PMA (hatched bars). The data are shown as in Fig. 1. B, expressions of R-Ras mutants in mast cells transfected with the neomycin gene, or R-Ras mutants as indicated. C, phosphorylation of Akt. Cell lysates of mast cells expressing the neomycin gene (Neo) or R-Ras mutants were analyzed by Western blotting for phosphorylation of Akt by the antibody specific for phosphorylation of serine 473 of Akt (upper panel). The membrane was stripped and reprobed with anti-Akt antibody recognizing both phosphorylated and unphosphorylated Akt (lower panel).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this study, we examined the ability of a series of the Ras/Rho family of small GTPases to activate VLA-5 to adhere to fibronectin in bone marrow-derived mast cells. We found that the active forms of Ha-Ras (Ha-Ras^Val-12) and R-Ras (R-Ras^Val-38) were most potent in stimulating adhesion to fibronectin. Both Ha-Ras^Val-12 and R-Ras^Val-38 induced the high affinity state of VLA-5. However, PI 3-kinase inhibitors abrogated adhesion by Ha-Ras^Val-12, but not R-Ras^Val-38. Among effector loop mutants of Ha-Ras^Val-12, only the Gly³⁷ mutant retained the ability to stimulate adhesion and also the ability to associate with p110, which is consistent with strong phosphorylation of Akt in the Gly³⁷ mutant expressing cells. The membrane-targeted version of p110 was sufficient to stimulate adhesion to fibronectin. These results indicate that Ha-Ras^Val-12 depends on PI 3-kinase  to activate VLA-5 in mast cells.

We showed that wortmannin had the marginal effect on adhesion induced by R-Ras^Val-38, whereas it abolished the increase in ligand binding activity of VLA-5. This result suggests that R-Ras depends on PI 3-kinase on induction of the high affinity state of VLA-5, which only made a minor contribution to R-Ras^Val-38-induced adhesion. The effects on adhesion of mutations in the effector loop region of R-Ras^Val-38 were distinct from those of Ha-Ras^Val-12 and the decrease of phosphorylation of Akt did not result in loss of adhesion, which further support the PI 3-kinase independent mechanism of R-Ras^Val-38 to activate VLA-5. Although it is currently unclear about adhesion mechanisms of R-Ras, it is conceivable that R-Ras^Val-38 has other mechanisms that induce adhesion mediated through the low affinity state of VLA-5, possibly by modulating lateral diffusion/clustering on VLA-5 on cell surface.

We have recently shown that PI 3-kinase is an affinity modulator of VLA-5, which is critically involved in adhesion induced by FcRI (20). Our results that PI 3-kinase was involved in Ha-Ras^Val-12 and R-Ras^Val-38 induced the high affinity state of VLA-5 are in line with previous reports (38, 41) on activation of PI 3-kinase by these small GTPases, and further implicate physiological roles in Ha-Ras and R-Ras in FcRI-induced adhesion. In fact, activation of Ha-Ras, but not R-Ras was most seen in mast cells when stimulated with cross-linking of FcRI. The kinetics of Ha-Ras activation paralleled with that of adhesiveness of VLA-5 increased by FcRI.² Thus, Ha-Ras could contribute to the high affinity state of VLA-5 by FcRI. However, we could not demonstrate that Ha-Ras was responsible for the high affinity state of VLA-5 by FcRI, because a dominant negative Ha-Ras was not expressed in mast cells, possibly due to the inhibition of their growth.

Ha-Ras was reported to suppress the activation of integrins through the Raf/MAP kinase pathway. In this study, chimeras of the extracellular regions of _IIb3 and intracellular regions of ₅₁ or ₆₁ integrins were used in adherent Chinese hamster ovary cells (24). In the present study, we showed that Raf-CAAX did not affect adhesion induced by PMA (Fig. 7A). We also showed that a conditional activation of the Raf/MAP kinase pathway did not increase adhesion (Fig. 8), and had no effect on steel factor- and FcRI-induced adhesion (data not shown). In fact, it has not been demonstrated directly whether Ha-Ras plays an inhibitory role in integrin activation by inside-out signals. The reason of the discrepancy is not known at present, but it is likely due to the differences in experimental systems including integrin adhesive property and structure. The other investigators reported that Ha-Ras was involved in adhesion induced by interleukin-3 and the activated form of Ha-Ras induced adhesion to fibronectin in an interleukin-3-dependent cell line Ba/f3, which was blocked by an inhibitor of phospholipase C, U-73122, but not inhibitors for PI 3-kinase (42). At present, we cannot explain this discrepancy. It could be due to the difference in cell context, or the other effects of U-73122, which caused marked morphological changes leading to cytolysis at high doses (43, 44).

The activated Ha-Ras interacts with and activates the PI 3-kinase catalytic subunit, p110 (38). The association of Ha-Ras and p110 was further characterized with the effector loop mutations of Ha-Ras (31). Notably the Cys⁴⁰ mutant was shown to interact with p110. p110 belongs to class IA of PI-3 kinase (45, 46). It is specifically expressed in leukocytes (30). Ha-Ras was also shown to interact with p110. p110 has biochemical properties similar to p110, including the sensitivity of PI 3-kinase inhibitors (30). However, p110 was co-immunoprecipitated with the Gly³⁷ mutant, but poorly with the Cys⁴⁰ mutant, as we showed in this study. The amino acid sequence of p110 in the Ras-binding domain is considerably diverged from that of p110 and p110 (30, 47). The sequence divergence in this region among p110 subunits likely contributes to the difference in the specific interaction sites of the effector loop region of Ha-Ras. The Gly³⁷ mutant was previously shown to interact with RalGDS and Rlf, GTP exchange factors for Ral (31, 34). However, the experiments with active forms of Ral, Rlf, or RalGDS (data not shown) rule out their critical roles in activation of integrins. Instead, the fact that Akt phosphorylation was augmented in the Gly³⁷, but not Cys⁴⁰ mutant expressing mast cells indicates that the association and activation of p110 with the Gly³⁷ mutant occur in mast cells. Taken together, our results demonstrate that p110 is a critical effector molecule of Ha-Ras in activating integrins in mast cells.

R-Ras^Val-38 was reported to increase ligand-binding affinity to ₅₁ in an interleukin-3-dependent myeloid cell line, 32D cells (25). We showed here that both R-Ras^Val-38 and Ha-Ras^Val-12 induced the high affinity state of VLA-5 in mast cells. The ligand binding activity was higher in Ha-Ras^Val-12 transfectants than in R-Ras^Val-38 transfectants. Our preliminary experiments showed that the dissociation constant of VLA-5 in Ha-Ras^Val-12 transfectants was between 20 and 50 nM, which was higher than that reported in R-Ras^Val-38 expressing 32D cells (250 nM) and equivalent to that induced by FcRI cross-linking (20). Importantly, our study revealed that the high affinity state of VLA-5 in R-Ras^Val-38 expressing cells was dispensable for adhesion to fibronectin, since the treatment of wortmannin abolished the high affinity state with a small inhibitory effect on adhesion. This is in contrast with Ha-Ras^Val-12 expressing cells, in which wortmannin abolished both ligand binding activity and adhesion to fibronectin at the similar doses. The PI 3-kinase independent adhesion by R-Ras was also supported by the analysis using the effector loop mutants. All of the effector loop mutations at the homologous sites of Ha-Ras resulted in decrease of Akt phosphorylation at the similar degree, but did not parallel levels of adhesion. Recently it has been reported that the R-Ras Gly⁶³ mutants among other mutants interacted more with the Ras-binding domain of PI 3-kinase p110, and that adhesion by the Gly⁶³ mutant was partially inhibited by a dominant negative Rac or Ral (48). We failed to detect distinct sites of the R-Ras effector loop region to interact with p110 by co-immunoprecipitation.³ The difference could be due to low homologies in the Ras-binding domain between p110 and p110 as discussed above. In addition, Rac^Val-12 or RalA^Leu-72 by itself failed to induce adhesion in our study, ruling out their critical roles downstream of R-Ras in activating integrins in our case, although they might promote adhesion by modulating cytoskeletal organization such as cell spreading.

Adhesion through integrins is mediated through multiple steps initiated by integrin activation by inside-out signals, leading to cytoskeletal reorganization and firm attachment by outside-in signals upon adhesion. Here in this study, we examined the ability of a series of Ras/Rho family of small GTPase to activate integrins in search for the possible inside-out signals. Our study clearly demonstrated that distinct members of small GTPases had the ability to regulate adhesiveness of integrins, which gives important clues to dissect regulatory processes of adhesion through integrins.

	ACKNOWLEDGEMENTS

We thank Drs. M. Hattori and N. Minato for T7-tagged Rap1A, Dr. S. Narumiya for Rho^Val-14, Dr. H. Koide for RalA^Leu-72, and Dr. T. Kitamura for pMX-neo.

	FOOTNOTES

^* This work was supported in part by a grant-in-aid by the Ministry of Education, Science, Sport, and Culture of Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence may be addressed: Bayer-chair, Dept. of Molecular Immunology, Graduate School of Medicine, Kyoto University, Yoshida-Konoe, Sakyo-ku, Kyoto 606-8501, Japan. Tel.: 81-75-771-8159; Fax: 81-75-771-8184; E-mail: tkinashi@mfour.med.kyoto-u.ac.jp.

To whom correspondence may be addressed: Dept. of Immunology, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108, Japan. Tel.: 81-5449-5265; Fax: 81-5449-5407; E-mail: takatsuk@ims.u-tokyo.ac.jp.

Published, JBC Papers in Press, May 8, 2000, DOI 10.1074/jbc.M000633200

² T. Kinashi, unpublished data.

	ABBREVIATIONS

The abbreviations used are: PMA, phorbol 12-myristate 13-acetate; MAP, mitogen-activated protein kinase; PI 3-kinase, phosphatidylinositol 3-kinase; BSA, bovine serum albumin; PAGE, polyacrylamide gel electrophoresis.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. Czyzyk, H.-C. Chen, K. Bottomly, and R. A. Flavell p21 Ras/Impedes Mitogenic Signal Propagation Regulates Cytokine Production and Migration in CD4 T Cells J. Biol. Chem., August 22, 2008; 283(34): 23004 - 23015. [Abstract] [Full Text] [PDF]

				L. E. Goldfinger, C. Ptak, E. D. Jeffery, J. Shabanowitz, D. F. Hunt, and M. H. Ginsberg RLIP76 (RalBP1) is an R-Ras effector that mediates adhesion-dependent Rac activation and cell migration J. Cell Biol., September 11, 2006; 174(6): 877 - 888. [Abstract] [Full Text] [PDF]

				H.-C. Chang, C.-Y. Cho, and W.-C. Hung Silencing of the Metastasis Suppressor RECK by RAS Oncogene Is Mediated by DNA Methyltransferase 3b-Induced Promoter Methylation. Cancer Res., September 1, 2006; 66(17): 8413 - 8420. [Abstract] [Full Text] [PDF]

				I. Oinuma, H. Katoh, and M. Negishi Semaphorin 4D/Plexin-B1-mediated R-Ras GAP activity inhibits cell migration by regulating {beta}1 integrin activity J. Cell Biol., May 22, 2006; 173(4): 601 - 613. [Abstract] [Full Text] [PDF]

				M. Dail, M. Richter, P. Godement, and E. B. Pasquale Eph receptors inactivate R-Ras through different mechanisms to achieve cell repulsion J. Cell Sci., April 1, 2006; 119(7): 1244 - 1254. [Abstract] [Full Text] [PDF]

				A. Ortiz-Stern and C. Rosales Fc{gamma}RIIIB stimulation promotes {beta}1 integrin activation in human neutrophils J. Leukoc. Biol., May 1, 2005; 77(5): 787 - 799. [Abstract] [Full Text] [PDF]

S. P. Holly, M. K. Larson, and L. V. Parise The Unique N-Terminus of R-Ras Is Required for Rac Activation and Precise Regulation of Cell Migration Mol. Biol. Cell, May 1, 2005; 16(5): 2458 - 2469.