De Novo Expression of the Integrin α5β1 Regulates αvβ3-mediated Adhesion and Migration on Fibrinogen
2003; Elsevier BV; Volume: 278; Issue: 24 Linguagem: Inglês
10.1074/jbc.m212538200
ISSN1083-351X
AutoresDaphne P. Ly, Kathleen Zazzali, Siobhan Corbett,
Tópico(s)Blood properties and coagulation
ResumoRecent evidence demonstrates that interactions between different integrins that are present on the cell surface can strongly influence the adhesive function of individual receptors. In this report, we show that Chinese hamster ovary cells that express the integrin αvβ3 in the absence of α5β1 demonstrate increased adhesion and migration on fibrinogen. Furthermore, αvβ3-mediated adhesion to fibrinogen is not augmented by the soluble agonist, MnCl2, suggesting that αvβ3 exists in a higher affinity state in these cells. De novo expression of wild-type α5β1 negatively regulates αvβ3-mediated adhesion and migration. This effect is not seen with expression of a chimeric α5β1 integrin in which the cytoplasmic portion of the α5 integrin subunit is replaced by the cytoplasmic portion of the α4 integrin. In addition, it does not require ligation of α5β1 by fibronectin. Cells that express a constitutively active β3 integrin that contains a point mutation in the conserved membrane proximal region of the cytoplasmic tail, D723R, are resistant to the effect of α5β1 expression. These data provide additional evidence of "cross-talk" between the integrins α5β1 and αvβ3, and support the idea that α5β1 regulates αvβ3-mediated ligand binding. This provides a relevant biological mechanism whereby variations in α5β1 expression in vivo may modulate activation of αvβ3 to influence its adhesive function. Recent evidence demonstrates that interactions between different integrins that are present on the cell surface can strongly influence the adhesive function of individual receptors. In this report, we show that Chinese hamster ovary cells that express the integrin αvβ3 in the absence of α5β1 demonstrate increased adhesion and migration on fibrinogen. Furthermore, αvβ3-mediated adhesion to fibrinogen is not augmented by the soluble agonist, MnCl2, suggesting that αvβ3 exists in a higher affinity state in these cells. De novo expression of wild-type α5β1 negatively regulates αvβ3-mediated adhesion and migration. This effect is not seen with expression of a chimeric α5β1 integrin in which the cytoplasmic portion of the α5 integrin subunit is replaced by the cytoplasmic portion of the α4 integrin. In addition, it does not require ligation of α5β1 by fibronectin. Cells that express a constitutively active β3 integrin that contains a point mutation in the conserved membrane proximal region of the cytoplasmic tail, D723R, are resistant to the effect of α5β1 expression. These data provide additional evidence of "cross-talk" between the integrins α5β1 and αvβ3, and support the idea that α5β1 regulates αvβ3-mediated ligand binding. This provides a relevant biological mechanism whereby variations in α5β1 expression in vivo may modulate activation of αvβ3 to influence its adhesive function. Integrins are transmembrane glycoproteins that are the principle mediators of cell interactions with the extracellular matrix (ECM) 1The abbreviations used are: ECM, extracellular matrix; CHO, Chinese hamster ovary; FBG, fibrinogen; FN, fibronectin; TSP, thrombospondin; PMA, phorbol 12-myristate 13-acetate; PKA, protein kinase A; PBS, phosphate-buffered saline; FACS, fluorescence-activated cell sorter. (1Berman A.E. Kozlova N.I. Membr. Cell Biol. 2000; 13: 207-244PubMed Google Scholar, 2Hynes R.O. Cell. 1992; 69: 11-25Abstract Full Text PDF PubMed Scopus (9026) Google Scholar, 3Sonnenberg A. Curr. Top. Microbiol. Immunol. 1993; 184: 7-35Crossref PubMed Scopus (177) Google Scholar). They are composed of noncovalently associated α and β subunits, which together determine the ligand-binding specificity of the receptor. Integrin-ECM interactions are essential for a diverse variety of important biological processes, including embryogenesis, cell survival, and wound healing (4Almeida E.A. Ilic D. Han Q. Hauck C.R. Jin F. Kawakatsu H. Schlaepfer D.D. Damsky C.H. J. Cell Biol. 2000; 149: 741-754Crossref PubMed Scopus (337) Google Scholar, 5Clark R.A. J. Invest. Dermatol. 1990; 94 (S): 128S-1134Abstract Full Text PDF PubMed Scopus (252) Google Scholar, 6Martin-Bermudo M.D. Dunin-Borkowski O.M. Brown N.H. J. Cell Biol. 1998; 141: 1073-1081Crossref PubMed Scopus (51) Google Scholar). During wound healing, alterations in integrin expression coincide with deposition of a newly formed provisional ECM whose primary structural components include fibrinogen (FBG) and fibronectin (FN) (5Clark R.A. J. Invest. Dermatol. 1990; 94 (S): 128S-1134Abstract Full Text PDF PubMed Scopus (252) Google Scholar, 7Clark R.A. Lanigan J.M. DellaPelle P. Manseau E. Dvorak H.F. Colvin R.B. J. Invest. Dermatol. 1982; 79: 264-269Abstract Full Text PDF PubMed Scopus (523) Google Scholar). Members of the β3 integrin family are the primary receptors that mediate cell interactions with FBG (8Corbett S.A. Schwarzbauer J.E. J. Surg. Res. 1999; 83: 27-31Abstract Full Text PDF PubMed Scopus (7) Google Scholar, 9Bonnefoy A. Liu Q. Legrand C. Frojmovic M.M. Biophys. J. 2000; 78: 2834-2843Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar, 10Parise L.V. Steiner B. Nannizzi L. Criss A.B. Phillips D.R. Biochem. J. 1993; 289: 445-451Crossref PubMed Scopus (24) Google Scholar). In platelets, the interaction of αIIbβ3 with FBG is dependent upon agonist-stimulated signaling events that alter the ligand-binding function of the αIIbβ3 receptor, a process termed "integrin activation" (11Shattil S.J. Kashiwagi H. Pampori N. Blood. 1998; 91: 2645-2657Crossref PubMed Google Scholar). The adhesive function of αvβ3 also seems to be regulated by intracellular events as recent studies indicate that the basal affinity of this receptor varies both by intracellular location and among cell types (12Byzova T.V. Kim W. Midura R.J. Plow E.F. Exp. Cell Res. 2000; 254: 299-308Crossref PubMed Scopus (106) Google Scholar, 13Kiosses W.B. Shattil S.J. Pampori N. Schwartz M.A. Nat. Cell Biol. 2001; 3: 316-320Crossref PubMed Scopus (233) Google Scholar, 14Felding-Habermann B. O'Toole T.E. Smith J.W. Fransvea E. Ruggeri Z.M. Ginsberg M.H. Hughes P.E. Pampori N. Shattil S.J. Saven A. Mueller B.M. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 1853-1858Crossref PubMed Scopus (486) Google Scholar). Activation of αvβ3 results in different functional states of the receptor that influence important integrin-dependent events, including cell migration, angiogenesis, and metastatic activity (14Felding-Habermann B. O'Toole T.E. Smith J.W. Fransvea E. Ruggeri Z.M. Ginsberg M.H. Hughes P.E. Pampori N. Shattil S.J. Saven A. Mueller B.M. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 1853-1858Crossref PubMed Scopus (486) Google Scholar, 15Byzova T.V. Goldman C.K. Pampori N. Thomas K.A. Bett A. Shattil S.J. Plow E.F. Mol. Cell. 2000; 6: 851-860Abstract Full Text Full Text PDF PubMed Google Scholar). Although rapid regulation of αvβ3 function may represent a common mechanism for the modulation of cell-ECM interactions, little is known about the molecular events that affect receptor affinity. Recent evidence demonstrates that interactions between different integrins that are present on the cell surface can strongly influence the adhesive function of individual receptors. This effect, referred to as integrin "cross-talk," has been demonstrated in a number of systems (16Diaz-Gonzalez F. Forsyth J. Steiner B. Ginsberg M.H. Mol. Biol. Cell. 1996; 7: 1939-1951Crossref PubMed Scopus (155) Google Scholar). For example, ligation of α4β1 inhibits α5β1-dependent expression of metalloproteases (17Huhtala P. Humphries M.J. McCarthy J.B. Tremble P.M. Werb Z. Damsky C.H. J. Cell Biol. 1995; 129: 867-879Crossref PubMed Scopus (369) Google Scholar). Interactions between β3 integrins and α5β1 have also been described. Ligation of αIIbβ3 by an integrin-specific FBG ligand suppresses the adhesive function of both α5β1 and α2β1 (16Diaz-Gonzalez F. Forsyth J. Steiner B. Ginsberg M.H. Mol. Biol. Cell. 1996; 7: 1939-1951Crossref PubMed Scopus (155) Google Scholar). Similarly, anti-αvβ3 antibodies block both α5β1-mediated phagocytosis of FN-coated beads and α5β1-mediated migration toward FN (18Blystone S.D. Graham I.L. Lindberg F.P. Brown E.J. J. Cell Biol. 1994; 127: 1129-1137Crossref PubMed Scopus (222) Google Scholar). The reverse effect also appears to be true as anti-α5β1 antibodies inhibit αvβ3-mediated cell migration, without influencing cell adhesion (19Kim S. Harris M. Varner J.A. J. Biol. Chem. 2000; 275: 33920-33928Abstract Full Text Full Text PDF PubMed Scopus (190) Google Scholar). Interestingly, both integrins are up-regulated on migrating cells post-injury where they are important receptors for provisional ECM proteins (5Clark R.A. J. Invest. Dermatol. 1990; 94 (S): 128S-1134Abstract Full Text PDF PubMed Scopus (252) Google Scholar, 20Stouffer G.A. Hu Z. Sajid M. Li H. Jin G. Nakada M.T. Hanson S.R. Runge M.S. Circulation. 1998; 97: 907-915Crossref PubMed Scopus (133) Google Scholar, 21Gailit J. Clarke C. Newman D. Tonnesen M.G. Mosesson M.W. Clark R.A. Exp. Cell Res. 1997; 232: 118-126Crossref PubMed Scopus (122) Google Scholar). In this paper, we present further evidence of cross-talk between the integrins α5β1 and αvβ3. We report that in the absence of α5β1, αvβ3 exists in an "activated" state demonstrating high affinity for FBG that is not augmented by soluble agonists. De novo expression of α5β1, however, suppresses αvβ3-mediated adhesive functions. The α5β1-mediated modulation of αvβ3 occurs through a mechanism that is dependent on the cytoplasmic tail of the α integrin subunit. Furthermore, cells that express a constitutively active β3 integrin are resistant to the effects of α5β1, supporting evidence that the expression of α5β1 regulates αvβ3 affinity for its ligands. These data provide a relevant biological mechanism whereby variations in α5β1 expression in vivo may modulate activation of αvβ3 to influence its adhesive function. Construction of Chimeric Integrin cDNA—A 1.8-kb BamHI-XhoI fragment encoding the amino-terminal portion of the human α5 cDNA in pLJ (a gift of Dr. Jean Schwarzbauer, Princeton University, Princeton, NJ) was cloned into pcDNA 3.1(+) (pcDNA 3.1 α5-N, Invitrogen). A 2.5-kb fragment containing the carboxyl terminus of the human α5 cDNA was then ligated into a XhoI digest of pcDNA 3.1 α5-N to yield a complete human α5 cDNA in pcDNA 3.1 (X5C5). The X4C4 cDNA, encoding the wild-type human α4 integrin subunit (a gift of Dr. Patricia Keely, University of Wisconsin Medical School, Madison WI), was used to create the chimeric X5C4 cDNA. Briefly, a 1.3-kb KpnI-XbaI fragment of X4C4, containing the cytoplasmic portion of the receptor was cloned into pcDNA 3.1 to generate pcDNA 3.1 C4. A 3.1-kb HindIII digest of X5C5 was used to generate the amino-terminal portion of α5 that encoded the extracellular and transmembrane regions of the receptor. This HindIII fragment was cloned into a HindIII-digested pcDNA 3.1 C4 to yield X5C4. The cytoplasmic portions of X5C5 and X5C4 were sequenced at the Robert Wood Johnson Medical School DNA Core Facility to confirm the sequence of each chimeric construct. Cells, Cell Culture, and Transfection—The α5-negative, β3-negative Chinese hamster ovary (CHO) B2 cells were a kind gift of Dr. Jean Schwarzbauer. CHO B2 αvβ3 (D723R) cells were a kind gift of Dr. Mark Ginsberg (Scripps Research Institute, La Jolla, CA) and were designated CHO B3(D-R). CHO B2 cells were transfected with a wild-type human β3 cDNA by electroporation. Stable transfectants, designated CHO B3, were established by selection in 500 μg/ml Zeocin (Invitrogen) and were analyzed for αvβ3 expression as described below. CHO B3 cells were then transfected with either the X5C5 cDNA (CHO B3C5) or the chimeric X5C4 cDNA (CHO B3C4)), and selected in Zeocin and 200 μg/ml Geneticin (G418, Invitrogen). CHO B3(D-R) cells were transfected with the X5C5 cDNA (CHO B3(D-R)C5). Sterile fluorescence-activated cell sorting (FACS), performed on an EPIC Altra high speed cell sorter (Beckman-Spinco, Miami, FL), was used to establish cell populations that expressed similar levels of αvβ3 and α5β1. CHO X5C5 (αvβ3–/α5β1+) and X5C4 cells (αvβ3–/α5mutβ1+) were established by transfection of CHO B2 cells with the X5C5 or X5C4 cDNAs, respectively and selection as described above. Cell surface expression of both αvβ3 and α5β1 were assessed and monitored by FACS weekly to ensure stable integrin expression. All cell lines were maintained in Dulbecco's modified Eagle's medium (Invitrogen) containing 10% fetal calf serum (HyClone Laboratories, Logan, UT), 2 mm glutamine (Invitrogen), 1% non-essential amino acids (Invitrogen), 1% sodium pyruvate (Invitrogen), 1% antibiotics/antimycotics (Invitrogen), and Zeocin or G418 as indicated. Chimeric Integrin Expression—For additional verification of the wild-type and mutant α5 integrins, cells were grown in 10-cm tissue culture plates until confluent. Cells were washed once with ice-cold PBS and lysed. Equal amounts of total cell lysate were incubated with 2 μg of a monoclonal antibody specific for the α5 integrin extracellular domain (Chemicon International Co., Temecula, CA) at 4 °C overnight. Immunocomplexes were recovered by incubating with Protein A-Protein G-Sepharose (Pierce) for 2 h. The immunoprecipitates were analyzed by SDS-PAGE and transferred to nitrocellulose (Fisher Scientific). The membranes were blocked with 5% nonfat dry milk at 4 °C overnight. Immunodetection was performed using a polyclonal antibody specific for the α5 integrin extracellular domain (Chemicon) or monoclonal antibodies specific for the α5 integrin cytoplasmic domain or the α4 integrin cytoplasmic domain (Chemicon and Santa Cruz Biotechnology, respectively) followed by horseradish peroxidase-conjugated goat anti-rabbit (α5) or bovine anti-goat (α4) secondary antibody (Pierce). The specific proteins were detected with enhanced chemiluminescence (ECL, Pierce) according to the manufacturer's instructions and exposed to film (X-Omat, Eastman Kodak). Determination of Cell Surface Integrin Expression—Cells were trypsinized, washed with incubation buffer (1× PBS and 2% fetal calf serum) and resuspended at a concentration of 1 × 107 cells/ml in incubation buffer. Cells were incubated on ice with monoclonal antibody specific for either αvβ3 (LM609; 1 μg/1 × 106 cells; Chemicon) or α5β1 (VC5; 1 μg/1 × 106 cells; BD Pharmingen) for 30 min, washed once, and then incubated on ice with an Alexa Fluor 488 goat anti-mouse IgG (Molecular Probes, Eugene, OR). Following additional washes, cells were pelleted, resuspended in incubation buffer, and analyzed by FACS on the EPIC Altra cell sorter. Preparation of FN-depleted FBG—Human FBG (Calbiochem/Novabiochem, San Diego, CA) was brought into solution at a concentration of 100 mg/15 ml according to the manufacturer's instructions. FN was depleted using gelatin-Sepharose beads (Amersham Biosciences). Briefly, 7.5 ml of FBG solution was combined with 2 ml of packed gelatin-Sepharose beads. The mixture was rotated at room temperature for 1 h. After pelleting the beads, the supernatant was transferred to a second gelatin-Sepharose containing tube and the process was repeated. Beads were then pelleted, and the supernatant containing FN-depleted FBG was collected. Protein concentration was determined using BCA protein assay reagent (Pierce) and adjusted to 5–6 mg/ml. FN-depleted FBG was then aliquoted and stored at –80 °C. FBG was analyzed by SDS-polyacrylamide gel electrophoresis followed by Coomassie stain to assess FN depletion. Cell Adhesion Assays—96-Well dishes were coated overnight at 4 °C with increasing concentrations of FBG, thrombospondin (TSP), or FN as indicated. Control wells were incubated with PBS. For each experiment, wells were formed in triplicate for each condition. Cell adhesion assay was performed as previously described (8Corbett S.A. Schwarzbauer J.E. J. Surg. Res. 1999; 83: 27-31Abstract Full Text PDF PubMed Scopus (7) Google Scholar). Briefly, 8 × 104 cells were added to each well, and incubated at 37 °C for 1 h. Adherent cells were fixed and stained for 25 min in a mixture of 2% EtOH, 100 mm sodium borate, and 0.5% crystal violet. Dye was then eluted with 10% acetic acid. The optical density (OD) at 570 nm was measured for each well using a 96-well microtiter plate reader (Bio-Tek Instruments, Winooski, VT). To eliminate the contribution of cell spreading to adhesion, short time course experiments were also performed. In these experiments, 24-well dishes were coated with either 50 μg/ml FBG or 10 μg/ml TSP. 2 × 105 cells were added to each well and allowed to adhere for 15 min. The wells were then washed gently to remove non-adherent cells. Adherent cells were then fixed, stained, and counted as described below. Each well was formed in duplicate. In certain experiments, cell adhesion was performed in the presence of function blocking antibodies to either αvβ3 (LM609) or α5β1 (P1D6; 25 μg/ml; Oncogene Research Products, La Jolla, CA). For integrin activation assays, cell adhesion was performed in the presence or absence of either 100 μm MnCl2 or 200 nm PMA. To quantify cell adhesion, cells were fixed with methanol for 15 min, and stained with modified Giemsa stain (Sigma) for1hat room temperature. Cells from 3 random high power fields were counted by light microscopy and averaged. For each experiment, the mean of the samples was obtained and normalized relative to the adhesion of B3 cell. The data are expressed as the mean ± S.E. of three separate experiments. A one-way analysis of variance followed by Newman-Keul's test was used for statistical analysis of the data. Cell Spreading Assays—24-Well tissue culture dishes were coated overnight with 50 μg/ml FBG at 4 °C. Cells in serum-free medium were placed in FBG-coated wells (2.5 × 105 cells/well) and incubated for increasing time as indicated. At each time point, cells were fixed with methanol and stained with modified Giemsa. Stained cells were viewed under high power using inverted bright-field optics. Photographic images were captured using a Spot color camera (Diagnostic Instruments, Sterling Heights, MI) connected to a MacIntosh G4 computer equipped with IPLab image analysis software. Cell Migration Assays—Costar transwells (8.0 μm pores, 6.5 mm diameter (Fisher Scientific)) were coated on both sides or on the under-surface only with 50 μg/ml FBG or 10 μg/ml FN overnight at 4 °C, washed with PBS, and blocked with 1% bovine serum albumin prior to usage. 2.5 × 105 cells in serum-free medium with 1% bovine serum albumin were placed into the upper chamber of coated transwells and incubated under tissue culture conditions for 5 h. Non-migrated cells were removed by wiping the upper side of the membrane with a cotton swab. The transwells were washed 3 times with PBS, fixed with methanol, and stained with modified Giemsa. The transwells were then rinsed with Milli-Q water. The stained cells in three random high powered fields were counted by light microscopy. Each experiment was repeated three times. Data are expressed either as the mean cell count ± S.E. or as the percent of cells migrated relative to B3 cells. Statistical analysis of the data was performed using one-way analysis of variance followed by Newman-Keul's test. Clot Retraction Assays—12-Well non-tissue culture dishes were blocked with 1% bovine serum albumin at 4 °C overnight. Wells were washed 3 times in PBS the next day and left to dry. Cells were detached as described above and resuspended in 25 mm Hepes-buffered saline. For integrin activation assays, cells were pretreated with 100 μm MnCl2 for 15 min. To prepare fibrin clots for retraction assays, the clotting components as listed were mixed in a volume of 1.5 ml at room temperature: FN-depleted FBG (720 μg/ml), 1 mm CaCl2, 20 μg/ml aprotinin (Sigma), 2 μg/ml human plasma coagulation factor XIII (Calbiochem), 25 mm Hepes saline, ±MnCl2, 1.0 units/ml thrombin (Sigma). Cells (3 × 106 cells/clot) were added to the clot components followed quickly by thrombin. After the addition of thrombin, the mixture was rapidly pipetted into 24-well tissue culture dishes and incubated at 37 °Cfor1hina tissue culture incubator. Serum-free medium was then added to the wells and the clots were carefully detached from the dishes using a dissecting microscope. For integrin activation assays, clot retraction was performed in the presence of 100 μm MnCl2. Clot diameter was measured under a dissecting microscope at T0 and at increasing times as indicated. Clot retraction was calculated by subtracting the diameter recorded at each time point from the starting diameter (T0). The data is the mean of six separate experiments and is expressed as clot retraction (mm) ± S.E. Statistical analysis of the data was performed using two-way analysis of variance followed by Newman-Keul's test. Determination of Integrin Expression in CHO B2-transfected Cells—The CHO B2 cell line (22Schreiner C.L. Bauer J.S. Danilov Y.N. Hussein S. Sczekan M.M. Juliano R.L. J. Cell Biol. 1989; 109: 3157-3167Crossref PubMed Scopus (113) Google Scholar), which does not express endogenous hamster α5β1, was transfected with a cDNA encoding the human β3 integrin subunit. Stable transfectants expressing αvβ3 were selected (CHO B3) and surface expression of αvβ3 was determined by flow cytometry using a monoclonal antibody specific for αvβ3. B3 cells were then transfected with either a cDNA for the wild-type human α5 integrin subunit (X5C5) or a mutant α5 cDNA in which the cytoplasmic domain of the α5 integrin was replaced by the cytoplasmic domain of the α4 integrin (X5C4). Stable cell populations expressing αvβ3 and either X5C5 or X5C4 were established by FACS using an anti-α5 monoclonal antibody to establish the B3C5 and B3C4 cell lines, respectively. Fig. 1A demonstrates that B3 and B3C5 cells express similar levels of αvβ3. B3C5 cells, but not B3 cells, express α5β1. B3C5 and B3C4 cells expressed similar levels of both α5β1 and αvβ3 (Fig. 1A). Immunoblot analysis of B3C5 and B3C4 cell lysates with antibodies specific to either the α5 or α4 cytoplasmic tails confirm the expression of the chimeric receptor (Fig. 1B). α5β1Regulates αvβ3-mediated Adhesive Events—FBG, a key component of the provisional ECM that is deposited in early wounds, is a ligand for αvβ3 (21Gailit J. Clarke C. Newman D. Tonnesen M.G. Mosesson M.W. Clark R.A. Exp. Cell Res. 1997; 232: 118-126Crossref PubMed Scopus (122) Google Scholar). To determine the effect of α5β1 on αvβ3-mediated cell adhesion to FBG, CHO cells expressing αvβ3 alone (B3), α5β1 alone (X5C5), or both αvβ3 and α5β1 (B3C5) were examined. Cells transfected with an empty vector served as a negative control (P3). B3 cells adhere to FBG in a concentration dependent fashion (Fig. 2A). B3 cell adhesion to FBG is completely inhibited by LM609, a function-blocking antibody specific for αvβ3 (data not shown). Surprisingly, B3C5 cells that express both αvβ3 and α5β1 demonstrate significantly decreased adhesion to FBG when compared with B3 cells despite expression of comparable levels of the αvβ3 integrin (p = 0.0002 at 100 μg/ml FBG). Control cells and cells expressing α5β1 alone (X5C5) do not adhere to FBG. B3, B3C5, and X5C5 cell adhesion to FN was also examined. No discernible difference was measured, supporting previous work that demonstrates that individually both αvβ3 and α5β1 support adhesion to FN (data not shown). To determine whether the effect of α5β1 on αvβ3-mediated adhesion was specific for FBG, adhesion of B3 and B3C5 cells to TSP was also examined. B3C5 cell adhesion to TSP was decreased by 50% when compared with B3 cells, an effect comparable with that seen for FBG (Fig. 2C). Taken together, these data suggest that co-expression of α5β1 with αvβ3 negatively regulates αvβ3 interaction with its ligands. At 1 h after plating on FBG, both B3 and B3C5 cells are partially spread (Fig. 3A). To eliminate the contribution of cell spreading to cell adhesion, short time course adhesion assays were also performed. At 15 min following adhesion to FBG, B3 cells remain round. By 30 min, however, a significant number of cells are partially spread (data not shown). Therefore, adhesion assays were performed at the 15-min time point. B3C5 cells demonstrate a significant decrease in cell adhesion to both FBG and TSP at 15 min (Fig. 2, B and D). This data confirms that co-expression of α5β1 with αvβ3, significantly inhibits αvβ3-mediated cell adhesion to its ligands, independent of cell spreading. The cell adhesion experiments were performed in the absence of exogenous FN. However, CHO B2 cells do produce a small amount of cellular FN. Therefore, the inhibitory effect of α5β1 on αvβ3-mediated adhesion could require the ligation of α5β1 by endogenous FN. To test this hypothesis, B3C5 cells were allowed to adhere to FBG in the presence of the α5β1 function-blocking antibody. As demonstrated in Fig. 2, E and F, incubation of B3C5 cells with P1D6 did not significantly affect αvβ3-mediated adhesion to FBG at either the 15- or 60-min time points (p = 0.24). This suggests that the inhibitory effect of α5β1 on αvβ3 does not require the binding of FN to α5β1. Cell spreading follows cell attachment to adhesive substrates. Whereas B3C5 cell adhesion to FBG is significantly decreased, some cells do adhere. When B3C5 cell spreading on FBG was examined, however, cells show decreased spreading at all time points when compared with B3 cells (Fig. 3A). α5β1Blocks αvβ3-mediated Migration on FBG—Integrin αvβ3 has an important role in cell migration, particularly during wound healing. Therefore, we sought to determine the effect of α5β1 expression on αvβ3-mediated cell migration on FBG. Random migration on FBG was determined using a Boyden chamber assay in which transwells were coated on both sides with FBG. As demonstrated in Fig. 3B, B3 cells migrate efficiently on FBG. Expression of integrin α5β1, however, inhibited greater than 90% of αvβ3-mediated cell migration (p = 0.0001). To determine whether this reflected a global decrease in migratory function of B3C5 cells, cell migration on FN was also examined. When B3 and B3C5 cell migration on FN were compared, no discernible difference was measured (Fig. 3B). Cells expressing α5β1 alone do not migrate on FBG (Fig. 4A). However, these cells can migrate efficiently on FN (Fig. 4B). These data indicate that expression of the integrin α5β1 suppresses αvβ3-mediated cell migration Inhibition of αvβ3by α5β1Is Dependent on the Cytoplasmic Tail of the α5Integrin Subunit—One possible explanation for the effect of α5β1 expression on αvβ3-mediated adhesive events may be that the presence of α5β1 sterically hinders the interaction of αvβ3 with its ligand. To test this hypothesis, B3 cells were transfected with a chimeric α5 integrin cDNA, in which the cytoplasmic domain of the α5 integrin was replaced by the cytoplasmic domain of the α4 integrin. A stable cell line expressing the mutant α5 was established as described above (B3C4). B3C5 and B3C4 cells express similar levels of cell surface α5β1 (Fig. 1A). As previously demonstrated, the expression of wild-type α5β1 significantly inhibits αvβ3-mediated migration on FBG (Fig. 4A). When the B3C4 cells expressing the chimeric α5 integrin are examined, however, their migration on FBG is significantly greater than the B3C5 cells and is comparable with the untransfected B3 cells. To confirm that the chimeric α5 integrin is functional, CHO B2 cells were transfected with wild-type (X5C5) or mutant (X5C4) α5β1 cDNA and stable cell lines were established by FACS. There was no difference between the X5C5 and X5C4 cell migrations on FN (Fig. 4B). Taken together, these data suggest that α5β1 acts in an "inside-out" fashion to negatively regulate αvβ3-mediated migration and that this effect is dependent on the α5 integrin cytoplasmic domain. Mn2+Promotes B3C5 Cell Adhesion to FBG—Intracellular signaling events can modulate integrin ligand binding affinity, a process termed activation or inside-out signaling (23Bazzoni G. Hemler M.E. Trends Biochem. Sci. 1998; 23: 30-34Abstract Full Text PDF PubMed Scopus (229) Google Scholar, 24Liddington R.C. Ginsberg M.H. J. Cell Biol. 2002; 158: 833-839Crossref PubMed Scopus (255) Google Scholar). This process has been well described for members of the β3 integrin family (12Byzova T.V. Kim W. Midura R.J. Plow E.F. Exp. Cell Res. 2000; 254: 299-308Crossref PubMed Scopus (106) Google Scholar, 25Byzova T.V. Plow E.F. J. Cell Biol. 1998; 143: 2081-2092Crossref PubMed Scopus (98) Google Scholar). We hypothesized that co-expression of the integrin α5β1 leads to an alteration in the affinity state of αvβ3 resulting in decreased ligand binding. To test this hypothesis, we used Mn2+, a divalent cation known to stimulate integrin activity and interactions with ligand (26Bazzoni G. Shih D.T. Buck C.A. Hemler M.E. J. Biol. Chem. 1995; 270: 25570-25577Abstract Full Text Full Text PDF PubMed Scopus (252) Google Scholar). B3, B3C5, or X5C5 cells were allowed to adhere to FBG in the presence or absence of 100 μm Mn2+ for 1 h (Fig. 5A). Treatment with Mn2+ does not significantly improve B3 cell adhesion to FBG, suggesting that αvβ3 exists in a higher affinity state in these cells. When Mn2+ is added to B3C5 cells, however, there is a significant increase in adhesion to FBG to a level that is comparable with the B3 cells (p = 0.41). X5C5 cells show no increased affinity for FBG in the presence of MnCl2. Similar results were obtained when adhesion was performed at 15 min (data not shown). Interestingly, enhanced adhesion of B3C5 cells to FBG is not seen in the presence of PMA (Fig. 5B). These data demonstrate that αvβ3 exists in a low affinity state when co-expressed with α5β1. Furthermore, the decreased αvβ3-mediated ligand binding can be rescued by the addition of MnCl2 but not by PMA. α5β1Regulates αvβ3-mediated Retraction of Fibrin Matrices—The retraction of fibrin matrices by cells that express the integrin αvβ
Referência(s)