AF6 Negatively Regulates Rap1-induced Cell Adhesion
2005; Elsevier BV; Volume: 280; Issue: 39 Linguagem: Inglês
10.1074/jbc.m505057200
ISSN1083-351X
AutoresZhongchun Zhang, Holger Rehmann, Leo Price, Jürgen Riedl, Johannes L. Bos,
Tópico(s)Protein Kinase Regulation and GTPase Signaling
ResumoAF6 is involved in the connection of membrane-associated proteins to the actin cytoskeleton. It binds to Ras-like small GTPases and is suggested to be an effector of both Ras and Rap. Here we show that knockdown of AF6 in T cells by RNA interference enhanced Rap1-induced integrin-mediated cell adhesion, whereas overexpression of AF6 had the opposite effect. Interestingly, AF6-induced inhibition of cell adhesion correlated with an increase in RapGTP levels. Like AF6, protein KIAA1849 contains a Ras association domain and interacted with Rap1. However, KIAA1849 did not inhibit Rap1-induced cell adhesion. We concluded that AF6 is a negative regulator of Rap-induced cell adhesion. We proposed that AF6 inhibits Rap-mediated cell adhesion by sequestering RapGTP in an unproductive complex and thus prevents the interaction of Rap1 not only with effectors that mediate adhesion but also with Rap GTPase-activating proteins. Thus, AF6 may buffer RapGTP in resting T cells and maintain them in a non-adherent state. AF6 is involved in the connection of membrane-associated proteins to the actin cytoskeleton. It binds to Ras-like small GTPases and is suggested to be an effector of both Ras and Rap. Here we show that knockdown of AF6 in T cells by RNA interference enhanced Rap1-induced integrin-mediated cell adhesion, whereas overexpression of AF6 had the opposite effect. Interestingly, AF6-induced inhibition of cell adhesion correlated with an increase in RapGTP levels. Like AF6, protein KIAA1849 contains a Ras association domain and interacted with Rap1. However, KIAA1849 did not inhibit Rap1-induced cell adhesion. We concluded that AF6 is a negative regulator of Rap-induced cell adhesion. We proposed that AF6 inhibits Rap-mediated cell adhesion by sequestering RapGTP in an unproductive complex and thus prevents the interaction of Rap1 not only with effectors that mediate adhesion but also with Rap GTPase-activating proteins. Thus, AF6 may buffer RapGTP in resting T cells and maintain them in a non-adherent state. Rap proteins (Rap1a, -1b, -2a, and -2b) are small GTPases closely related to Ras. They are activated by a variety of extracellular signals through the regulation of specific guanine nucleotide exchange factors and GTPase-activating proteins (GAP 2The abbreviations used are: GAP, GTPase-activating protein; RA, Ras association; RapL, regulator of adhesion and polarization enriched in lymphoid tissues; Riam, Rap1-GTP-interacting adaptor molecule; 007, 8-(4-chlorophenylthio)-2′-O-methyladenosine-3′,5′-cyclic monophosphate; HA, hemagglutinin; GST, glutathione S-transferase; mGppNHp, 2′-/3′-O-(N′-methylanthraniloyl)-guanyl-5′-yl-imidodiphosphate; shRNA, short hairpin RNA; AF6L, AF6-like; TK, thymidine kinase.2The abbreviations used are: GAP, GTPase-activating protein; RA, Ras association; RapL, regulator of adhesion and polarization enriched in lymphoid tissues; Riam, Rap1-GTP-interacting adaptor molecule; 007, 8-(4-chlorophenylthio)-2′-O-methyladenosine-3′,5′-cyclic monophosphate; HA, hemagglutinin; GST, glutathione S-transferase; mGppNHp, 2′-/3′-O-(N′-methylanthraniloyl)-guanyl-5′-yl-imidodiphosphate; shRNA, short hairpin RNA; AF6L, AF6-like; TK, thymidine kinase.(s)) (1Bos J.L. Nat Rev. Mol. Cell. Biol. 2003; 4: 733-738Crossref PubMed Scopus (412) Google Scholar). Rap1 is involved in various cellular processes, most notably, the regulation of integrin-mediated cell adhesion and cadherin-mediated cell junction formation (2Knox A.L. Brown N.H. Science. 2002; 295: 1285-1288Crossref PubMed Scopus (201) Google Scholar, 3Hogan C. Serpente N. Cogram P. Hosking C.R. Bialucha C.U. Feller S.M. Braga V.M. Birchmeier W. Fujita Y. Mol. Cell. Biol. 2004; 24: 6690-6700Crossref PubMed Scopus (218) Google Scholar, 4Price L.S. Hajdo-Milasinovic A. Zhao J. Zwartkruis F.J. Collard J.G. Bos J.L. J. Biol. Chem. 2004; 279: 35127-35132Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar). A variety of effectors have been identified that mediate Rap1 function (5Bos J.L. Curr. Opin. Cell. Biol. 2005; 17: 123-128Crossref PubMed Scopus (392) Google Scholar). These effectors include RapL and Riam (6Lafuente E.M. van Puijenbroek A.A. Krause M. Carman C.V. Freeman G.J. Berezovskaya A. Constantine E. Springer T.A. Gertler F.B. Boussiotis V.A. Dev. Cell. 2004; 7: 585-595Abstract Full Text Full Text PDF PubMed Scopus (317) Google Scholar, 7Katagiri K. Ohnishi N. Kabashima K. Iyoda T. Takeda N. Shinkai Y. Inaba K. Kinashi T. Nat. Immunol. 2004; 5: 1045-1051Crossref PubMed Scopus (161) Google Scholar, 8Katagiri K. Maeda A. Shimonaka M. Kinashi T. Nat. Immunol. 2003; 4: 741-748Crossref PubMed Scopus (396) Google Scholar). Both proteins interact with Rap1 through a Ras association (RA) domain and mediate Rap-induced integrin-dependent cell adhesion. RapL may function by direct binding to both Rap1 and integrins, whereas Riam may function through an interaction with the actin-regulatory proteins profilin and Ena/Vasp. Arap3 is another RA domain containing protein that interacts with Rap1. This protein is an Arf- and RhoGAP and mediates Rap1-induced inactivation of Rho (9Krugmann S. Williams R. Stephens L. Hawkins P.T. Curr. Biol. 2004; 14: 1380-1384Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). However, not all effectors of Rap1 have a RA domain. For instance, Vav2, a guanine nucleotide exchange factor for the small GTPase Rac, binds to Rap1 through its PH domain and mediates Rap-induced cell spreading (10Arthur W.T. Quilliam L.A. Cooper J.A. J. Cell Biol. 2004; 167: 111-122Crossref PubMed Scopus (211) Google Scholar).AF6 (also called afadin) has a N-terminal region containing two RAs, one of which interacts with Ras-like small GTPases, including Ras and Rap (11Boettner B. Govek E.E. Cross J. Van Aelst L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 9064-9069Crossref PubMed Scopus (169) Google Scholar, 12Kuriyama M. Harada N. Kuroda S. Yamamoto T. Nakafuku M. Iwamatsu A. Yamamoto D. Prasad R. Croce C. Canaani E. Kaibuchi K. J. Biol. Chem. 1996; 271: 607-610Abstract Full Text Full Text PDF PubMed Scopus (187) Google Scholar, 13Linnemann T. Geyer M. Jaitner B.K. Block C. Kalbitzer H.R. Wittinghofer A. Herrmann C. J. Biol. Chem. 1999; 274: 13556-13562Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar). This protein was first identified as the fusion partner of ALL-1 protein in human acute myeloid leukemia (14Prasad R. Gu Y. Alder H. Nakamura T. Canaani O. Saito H. Huebner K. Gale R.P. Nowell P.C. Kuriyama K. Cancer Res. 1993; 53: 5624-5628PubMed Google Scholar). AF6 is a multidomain actin-binding protein that serves as a scaffold protein between cell membrane-associated proteins and the actin cytoskeleton (11Boettner B. Govek E.E. Cross J. Van Aelst L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 9064-9069Crossref PubMed Scopus (169) Google Scholar). Among the proteins that interact with AF6 are the tight junction protein ZO-1, the cell-cell adherence junction molecule nectin, various Eph receptors and the actin-regulatory protein profilin (11Boettner B. Govek E.E. Cross J. Van Aelst L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 9064-9069Crossref PubMed Scopus (169) Google Scholar, 15Yamamoto T. Harada N. Kano K. Taya S. Canaani E. Matsuura Y. Mizoguchi A. Ide C. Kaibuchi K. J. Cell Biol. 1997; 139: 785-795Crossref PubMed Scopus (283) Google Scholar, 16Buchert M. Schneider S. Meskenaite V. Adams M.T. Canaani E. Baechi T. Moelling K. Hovens C.M. J. Cell Biol. 1999; 144: 361-371Crossref PubMed Scopus (163) Google Scholar, 17Takahashi K. Nakanishi H. Miyahara M. Mandai K. Satoh K. Satoh A. Nishioka H. Aoki J. Nomoto A. Mizoguchi A. Takai Y. J. Cell Biol. 1999; 145: 539-549Crossref PubMed Scopus (438) Google Scholar). AF6 was found to be an effector for Ras in the control of cell junction formation via direct interaction with ZO-1 (15Yamamoto T. Harada N. Kano K. Taya S. Canaani E. Matsuura Y. Mizoguchi A. Ide C. Kaibuchi K. J. Cell Biol. 1997; 139: 785-795Crossref PubMed Scopus (283) Google Scholar). In addition, in Drosophila, the AF-6 homolog Canoe is an effector of Rap in the regulation of dorsal closure (18Boettner B. Harjes P. Ishimaru S. Heke M. Fan H.Q. Qin Y. Van Aelst L. Gaul U. Genetics. 2003; 165: 159-169PubMed Google Scholar). Recently it was shown that AF6 can interact with Rap GAPs, such as Rap1GAP and SpaI through its PDZ domain (19Su L. Hattori M. Moriyama M. Murata N. Harazaki M. Kaibuchi K. Minato N. J. Biol. Chem. 2003; 278: 15232-15238Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). This interaction is mediated by a conserved internal β-turn in the Rap GAPs. AF6 recruits Rap GAPs to negatively regulate the level of Rap1GTP and consequently Rap1-induced integrin-mediated cell adhesion. Because these functional studies were largely performed in overexpression systems, we addressed the question whether AF6 is involved in the control of cell adhesion induced by activation of endogenous Rap and depletion of endogenous AF6. Specific activation of endogenous Rap can be achieved using the cAMP analogue 8-pCPT-2′-O-Me-cAMP (007) that specifically activates Epac (exchange protein directly activated by cAMP) (20Enserink J.M. Christensen A.E. de Rooij J. van Triest M. Schwede F. Genieser H.G. Døskeland S.O. Blank J.L. Bos J.L. Nat. Cell Biol. 2002; 4: 901-906Crossref PubMed Scopus (615) Google Scholar). Indeed we observed that activation of Rap1-induced adhesion is negatively regulated by AF6. Importantly, knockdown of AF6 in T cells results in an enhancement of Rap-mediated cell adhesion. Surprisingly, inhibition of Rap1-induced cell adhesion by AF6 correlated with an increase, rather than a decrease in RapGTP. We propose a mechanism of AF6-induced inhibition of Rap1-mediated cell adhesion that includes the sequestration of Rap1GTP in an unproductive complex with AF6 preventing the interaction of Rap1 with both Rap GAPs and effectors of Rap-induced cell adhesion. Thus, AF6 may buffer RapGTP in resting T cells and maintain them in a non-adherent state.MATERIALS AND METHODSPlasmids and Constructs—Hemaglutamin (HA)-tagged Rap1, Rap1V12 (HA-RapV12), and Rap1GAP (HA-RapGAP I) have previously been described (21Zwartkruis F.J. Wolthuis R.M. Nabben N.M. Franke B. Bos J.L. EMBO J. 1998; 17: 5905-5912Crossref PubMed Scopus (191) Google Scholar). PCR fragments containing AF6 or AF6-ΔRA (residues 347–1612) flanked by a KpnI site at the 5′ and a NotI site at the 3′ were amplified from AF6 cDNA, provided by Dr. Eli Canaani (Weizmann Institute of Science, Rehovot, Israel) and subcloned into KpnI/NotI sites of a pGEM-T vector (Promega). Subsequently, these fragments were subcloned into KpnI/NotI-digested pcDNA3-HA, and integrities of the constructs were confirmed by DNA sequencing. Myc-AF6-RA (residues 25–353) was generated by PCR amplification of a fragment flanked by a EcoRI site at the 5′ and a NotI site at the 3′ of AF6 from AF6 cDNA. This fragment was subcloned into EcoRI/NotI-digested pcDNA3-Myc vector. Full-length Myc tagged AF6 was provided by Dr. Kaibuchi Kozo (Division of Signal Transduction, Nara Institute of Science and Technology, Ikoma, Japan). The AF6L (KIAA1849) cDNA containing the coding sequence was kindly provided by the Kazusa DNA Research Institute (22Nagase T. Nakayama M. Nakajima D. Kikuno R. Ohara O. DNA Res. 2001; 8: 85-95Crossref PubMed Scopus (119) Google Scholar). Polymerase chain reaction fragments containing AF6L or AF6L-RA (residues 25–213) flanked by a SalI site at the 5′ and a NotI site at the 3′ were subcloned into the pGEM-T vector. These clones were subsequently used to generate HA-AF6L or HA-AF6L-RA by introducing the SalI/NotI fragment containing AF6L or AF6L-RA into SalI/NotI-digested pMT2-SM-HA.Rap1 Activation Assays and Immunoblotting—Rap1 activation was assayed as described previously (23de Rooij J. Bos J.L. Oncogene. 1997; 14: 623-625Crossref PubMed Scopus (420) Google Scholar). Briefly, cells were washed with cold phosphate-buffered saline and lysed with buffer containing 1% Nonidet P-40. Lysates were cleared by centrifugation, and active Rap was precipitated with glutathione-Sepharose beads precoupled to a GST fusion protein of the Ras association domain of Ral guanine nucleotide dissociation stimulator. Precipitates were washed three times with lysis buffer and solubilized in SDS sample buffer. A portion of the cell lysate was reserved for analysis of total Rap content. HA-Rap1 was detected following Western blotting with anti-HA antibodies.GST Pull-down Assays—For HA-AF6 and HA-AF6L pull-down assays, glutathione-agarose beads were loaded with GST-Rap1. HB6 cells were transfected with HA-AF6 or HA-AF6L, followed by cell lysis in lysis buffer containing 50 mm Tris/HCl, pH 7.5, 150 mm NaCl, 0.5% piridinium betain, 5 mm EDTA, 10 mm NaF, 1 μg/ml aprotinin, and 1 μg/ml leupeptin. Cell extracts were incubated with GST-immobilized proteins for 1 h at 4°C. After 4 times washing in lysis buffer, bound proteins were resuspended in Laemmli sample buffer (Bio-Rad Laboratories) and resolved by SDS-PAGE, and HA-AF6 or HA-AF6L were detected by Western blotting using anti-HA antibody.Western Blotting—Western blotting of all protein samples was carried out using polyvinylidene difluoride membranes. The antibodies used for protein detection are the monoclonal anti-HA (12CA5), monoclonal anti-Myc (9E10), monoclonal anti-AF6 (Transduction Laboratories), monoclonal anti-tubulin (Oncogene Science).Cell Culture, Cell Line, and Transfection—The Epac I monoclonal Jurkat T cell line (HB6) was generated by retroviral transduction of Jurkat cells with amphotropic virus encoding Epac-IRES-GFP. 3L. S. Price and J. L. Bos, manuscript in preparation. HB6 cells were grown at 37 °C in RPMI 1640 (Invitrogen) supplemented with 10% heat-inactivated (30 min at 56 °C) fetal bovine serum and 0.05% glutamine in the presence of penicillin and streptomycin. Cells were transiently transfected by electroporation using 35 μg of plasmid DNA in total. Cells (1.2 × 107 cells/ml in 0.4 ml of complete medium) were pulsed at 250 V and 960 μF with 5 μg of TK-luciferase plasmid DNA, construct plasmid as indicated in the figure legends, and added vector plasmid to keep DNA amounts constant. Subsequently, 24 h after transfection, cells were transferred to serum-free medium and used 42 h after transfection. For RNA interference experiments, cells were transferred to serum-free medium 48 h after transfection and used 72 h after transfection.Adhesion Assay—For adhesion assays, transiently transfected Jurkat cells serum-starved overnight were harvested, washed, and resuspended in TSM buffer (20 mm Tris/HCl, pH 8.0, 150 mm NaCl, 1 mm CaCl2, 2 mm MgCl2) at a concentration of 5×105 cells/ml. 24-Well Nunc Maxisorp plates (Corning) were coated with fibronectin (5 μg/ml) overnight at 4 °C, washed, and blocked for 1 h at 37°C with 1% bovine serum albumin, TSM. After washing, 200 μl of TSM was added per well with or without the indicated stimuli. 007 (BioLog) was used at 100 μm and Mn2+ was used at 4 mm. Subsequently, 200 μl of cell suspension was added per well. Cells were allowed to adhere for 1 h at 37°C, and nonadherent cells were removed with warmed 0.5% bovine serum albumin, TSM. Adherent cells were lysed and subjected to a luciferase assay as described previously (24Medema R.H. de Laat W.L. Martin G.A. McCormick F. Bos J.L. Mol. Cell. Biol. 1992; 12: 3425-3430Crossref PubMed Scopus (82) Google Scholar). Expression of transfected constructs was confirmed by immunoblotting of total cell lysates. Adherent cells were calculated, and the cell numbers were corrected for transfection efficiency and nonspecific effects of constructs by measuring luciferase activity of total input cells ((counts in cells bound/counts in total input cells) × 100%).Small Interfering RNA Preparation—To make pTER-AF6, gene-specific oligonucleotides were ligated into pTER vector that had been digested with BglII and HindIII according to described protocol (25van de Wetering M. Oving I. Muncan V. Pon Fong M.T. Brantjes H. van Leenen D. Holstege F.C. Brummelkamp T.R. Agami R. Clevers H. EMBO Rep. 2003; 4: 609-615Crossref PubMed Scopus (461) Google Scholar). The oligonucleotides used are, for AF6 shRNA1, 5′-gatcccgatgatcgggaaggcagatttcaagagaatctgccttcccgatcatctttttggaaa-3′ and 5′-agcttttccaaaaagatgatcgggaaggcagattctcttgaaatctgccttcccgatcatcgg-3′ and for AF6 shRNA2, 5′-gatcccggagagagctgacgggtctttcaagagaagacccgtcagctctctcctttttggaaa-3′ and 5′-agcttttccaaaaaggagagagctgacgggtcttctcttgaaagacccgtcagctctctcc-3′.GDI Measurements—Experiments were carried out as described previously (13Linnemann T. Geyer M. Jaitner B.K. Block C. Kalbitzer H.R. Wittinghofer A. Herrmann C. J. Biol. Chem. 1999; 274: 13556-13562Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar). In brief, 100 nm Rap loaded with the non-hydrolyzable, fluorescent GTP analog mGppNHp were incubated in the presence of 10 μm non-labeled nucleotide and various concentrations of the RA-domain AF6L. Fluorescence was measured with a Cary Eclipse (Varian, Australia) equipped with a temperature controlled multicuvette holder.RESULTSTo determine whether AF6 is involved in Rap1-induced cell adhesion, a Jurkat cell line stably expressing Epac (HB6 cells) was used. Cells were transfected with AF6 together with or without constitutive active Rap1V12 and analyzed for their ability to adhere to fibronectin. As shown in Fig. 1A, AF6 completely inhibited Rap1V12-induced cell adhesion. However, AF6 did not inhibit Mn2+-induced cell adhesion, where integrins are activated directly by divalent cations (26Dransfield I. Cabañas C. Craig A. Hogg N. J. Cell Biol. 1992; 116: 219-226Crossref PubMed Scopus (399) Google Scholar). Also, the stimulating effect of Rap1V12 on Mn2+-induced cell adhesion was completely inhibited by AF6. To investigate whether cell adhesion mediated by endogenous Rap1 is also inhibited by AF6, we treated HB6 cells with the Epac-specific analogue 007 to activate Epac and consequently Rap (20Enserink J.M. Christensen A.E. de Rooij J. van Triest M. Schwede F. Genieser H.G. Døskeland S.O. Blank J.L. Bos J.L. Nat. Cell Biol. 2002; 4: 901-906Crossref PubMed Scopus (615) Google Scholar). Like Rap1V12-induced adhesion, 007-induced adhesion was inhibited by the overexpression of AF6 (Fig. 1B). AF6 interacts with Rap via RA domains in its N-terminal region (11Boettner B. Govek E.E. Cross J. Van Aelst L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 9064-9069Crossref PubMed Scopus (169) Google Scholar, 13Linnemann T. Geyer M. Jaitner B.K. Block C. Kalbitzer H.R. Wittinghofer A. Herrmann C. J. Biol. Chem. 1999; 274: 13556-13562Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar). To investigate whether the interaction of AF6 with Rap is required for its ability to inhibit adhesion, an N-terminal deletion mutant lacking the RA domains was made, namely AF6-ΔRA (deletion of residues 1–346). Pull-down experiments with immobilized Rap1 demonstrated the inability of AF6-ΔRA to interact with Rap1 (Fig. 2A). In contrast to full-length AF6, AF6-ΔRA was not able to inhibit 007-induced adhesion in HB6 cells (Fig. 2B). Conversely, the isolated RA domain of AF6 (residue 1–420) was able to inhibit 007-induced adhesion (Fig. 2C). Surprisingly, full-length AF6 was much more efficient in inhibiting 007-induced cell adhesion than the isolated RA domain. From these results we concluded that in HB6 cells, expression of AF6 inhibits Rap-induced cell adhesion. Furthermore, we concluded that although the RA domains are required for this inhibition, additional domains are required for efficient inhibition.FIGURE 2Inhibition of cell adhesion by AF6 requires the RA domains of AF6. A, AF6 and Rap interaction requires the AF6 RA domains. Purified GST-Rap1 was used to precipitate HA-AF6 and HA-AF6-ΔRA from HB6 cells transiently transfected with either HA-AF6 or HA-AF6-ΔRA. The blot was probed with anti-HA antibody. Expression of each protein was determined by immunoblotting of the straight lysates. B, deletion of the RA domain abolishes AF6-induced inhibition of cell adhesion. HB6 cells were cotransfected with TK luciferase reporter plasmid and empty vector, AF6, or AF6-ΔRA. Adhesion was measured as indicated in legends Fig. 1. In the bottom panel expression levels of transfected proteins are shown. Results are representative of three independent experiments. FN, fibronectin. C, the AF6-RA domain blocks Rap-induced adhesion less efficient than full-length AF6. HB6 cells were cotransfected with TK luciferase reporter plasmid and either AF6 or indicated amounts of AF6-RA domain and adhesion was measured as indicated in legend Fig. 1. Results are representative of three independent experiments. Bottom panel, expression levels of transfected proteins. D, AF6 efficiently increases GTP level of overexpressed Rap1. HB6 cells were cotransfected with empty vector, HA-AF6, or HA-AF6-ΔRA in combination with HA-Rap1. The upper panel shows the GTP levels of the HA-Rap proteins determined using the pull-down assay (see "Materials and Methods"), the middle panel shows the level of HA-Rap1 in the total lysates, and the lower panel shows the expression of HA-AF6 and HA-AF6-ΔRA in the total lysates. The experiment was repeated three times with reproducible results. E, AF6 increases GTP level of endogenous Rap1. HB6 cells were cotransfected with empty vector or HA-AF6. The left panel shows the GTP levels of the Rap proteins determined using the pull-down assay (see "Materials and Methods"), the right panel shows the level of endogenous Rap1 in the total lysates and the overexpression of AF6 in the total lysates. The experiment was repeated three times with reproducible results. WB, Western blot.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Previously, it was shown that AF6 binds to the RapGAP SpaI and that AF6 enhanced the SpaI-induced decrease in Rap1GTP levels (19Su L. Hattori M. Moriyama M. Murata N. Harazaki M. Kaibuchi K. Minato N. J. Biol. Chem. 2003; 278: 15232-15238Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). To investigate the effect of AF6 expression on Rap1GTP levels, HA-Rap1 and HA-AF6 were cotransfected in HB6 cells, and the levels of Rap1GTP were determined by a pull-down assay. Surprisingly, AF6 strongly increased the level of HA-RapGTP (Fig. 2D). In contrast, AF6-ΔRA did not influence Rap1GTP levels. We further investigated the effect of AF6 expression on the change of endogenous Rap1GTP level in cells. Expression of AF6 resulted in an increase in endogenous Rap1GTP level (Fig. 2E). From these results we concluded that AF6 binds to Rap1 through its RA domains and stabilizes Rap1 in the GTP-bound state. This result further suggests that the observed inhibition of cell adhesion by AF6 may be because of sequestration of Rap1 in an unproductive complex.To investigate whether the negative effect of AF6 on cell adhesion is specific, we used shRNA to knock-down AF6. For that purpose HB6 cells were transiently transfected with two different shRNA constructs. Both resulted in a significant decrease of the endogenous AF6 levels (Fig. 3A). In these experiments the transfection efficiency was between 50 and 70% as measured by green fluorescent protein cotransfection, indicating that the two shRNA constructs function rather efficiently in knocking down the endogenous AF6. Importantly, both AF6 shRNAs resulted in an increased integrin-mediated adhesion (Fig. 3B). To investigate whether AF6 shRNA-induced cell adhesion is Rap-dependent, we introduced Rap1GAP to inhibit endogenous Rap (27Rubinfeld B. Munemitsu S. Clark R. Conroy L. Watt K. Crosier W.J. McCormick F. Polakis P. Cell. 1991; 65: 1033-1042Abstract Full Text PDF PubMed Scopus (203) Google Scholar). Indeed, Rap1GAP completely inhibited AF6 shRNA-induced cell adhesion (Fig. 3C). Moreover, knockdown of AF6 further enhanced 007-induced adhesion (Fig. 3D). Although overexpression of AF6 resulted in an increased level of Rap1GTP, it could well be that the increased adhesion by AF6 knockdown is caused by increased levels of Rap1GTP. We therefore tested the level of Rap1GTP in the presence of AF6 shRNA. HB6 cells were transiently transfected with AF6 shRNA and Rap1, stimulated with or without 007 and the level of RapGTP was determined by the pull-down assay. Knockdown of AF6 did not increase the level of Rap1GTP but slightly decreased particularly the basal level of Rap1GTP (Fig. 3E). From these results we concluded that endogenous AF6 is a negative regulator of Rap-induced cell adhesion.FIGURE 3AF6 depletion increases adhesion. A, AF6 shRNA decreases protein expression. HB6 cells were transfected with 30μg of indicated shRNA constructs and grown for 72 h. In control experiments using GFP we observed more than 50% transfection efficiency using this protocol. Cell lysates were analyzed by Western blotting with anti-AF6 antibody. Anti-tubulin was used to demonstrate equal loading. WB, Western blot. B, AF6 shRNA induces cell adhesion. HB6 cells were cotransfected with TK luciferase reporter plasmid and the indicated shRNA constructs. After 72 h cells were replated and seeded in triplicate wells coated with fibronectin (FN), and after 1 h the percentage of adherent cells were measured. Results are representative of three independent experiments. C, AF6 shRNA-induced cell adhesion is Rap-dependent. HB6 cells were transfected with TK luciferase reporter plasmid, together with empty vector, Rap1GAP, AF6 shRNA2, or Rap1GAP and AF6 shRNA2. Adhesion was measured as indicated above. Results are representative of three independent experiments. In the bottom panel expression of transfected Rap1GAP is shown. D, AF6 shRNA enhances 007-induced cell adhesion. HB6 cells were transfected with TK luciferase reporter plasmid, together with either empty vector or AF6 shRNA2, and adhesion was measured in the presence of 007 (concentration of 007 as indicated in the figure) as indicated above. Results are representative of three independent experiments. E, AF6 depletion slightly decreases the Rap1GTP levels. HB6 cells were transfected with HA-Rap1 and either empty vector or AF6 shRNA2. After 72 h the level of Rap1GTP was measured by the pull-down assays. The experiment was repeated three times with reproducible results.View Large Image Figure ViewerDownload Hi-res image Download (PPT)In search for effectors of the Rap family we identified a protein (KIAA1849) in the Kazusa DNA data base with similarities to AF6 (Fig. 4A). KIAA1849 is characterized by an N-terminal RA domain and a C-terminal PDZ domain. Because of its similarity to AF6 on domain structure, we called this protein AF6-like (AF6L) (Fig. 4A). We then investigated whether the AF6L can interact with Rap1. Indeed, like AF6, AF6L can be recovered from cell lysate using GST-Rap1 (Fig. 4B). This interaction was further characterized with recombinant AF6L-RA (residues 25–213). The binding of effector proteins to small G-protein often results in a reduced exchange of nucleotide bound to the G-protein (GDI-effect) (28Herrmann C. Nassar N. Prog. Biophys. Mol. Biol. 1996; 66: 1-41Crossref PubMed Scopus (47) Google Scholar). The dissociation of a fluorescently labeled nucleotide bound to Rap can be detected as a decrease of fluorescence intensity in the presence of an excess unlabeled nucleotide. Indeed, increasing concentration of AF6L-RA reduced the exchange rate of nucleotide bound to Rap1 (Fig. 4C). From this measurement an affinity for the interaction of AF6L-RA and Rap1 of 0.9 μm was determined. This is similar to the affinity of AF6-RA for Rap1 (0.25 μm) (13Linnemann T. Geyer M. Jaitner B.K. Block C. Kalbitzer H.R. Wittinghofer A. Herrmann C. J. Biol. Chem. 1999; 274: 13556-13562Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar). From these results we conclude that AF6L can interact with Rap1.FIGURE 4AF6L is not a negative regulator of Rap1 in cell adhesion. A, schematic comparison of AF6 and AF6L. AF6L consists of a RA domain in the N terminus and a PDZ domain in the C terminus. B, AF6L binds to Rap1 in a pull-down assay. HA-AF6 and HA-AF6L were expressed in HB6 cells and recovered by GST-Rap1. The blot was probed with anti-HA (12CA5) antibody. C, dissociation rate constant of mGppNHp from Rap1A in dependence of AF6L-RA concentration. The rate constant were measured at 37 °C and plotted against the concentration of the RA domain. From this data an affinity of 0.9 μm was calculated. D, AF6L does not block Rap1-induced adhesion. HB6 cells were cotransfected with TK luciferase reporter plasmid and either AF6 or AF6L, and adhesion was measured as indicated in legend Fig. 1. Results are representative of three independent experiments. Bottom panel, expression of AF6 and AF6L proteins. E, AF6L-RA domain inhibits Rap-induced adhesion. HB6 cells were cotransfected with TK luciferase reporter plasmid and either AF6L or indicated amounts of AF6L-RA domain, and adhesion was measured as indicated in legend to Fig. 1. Results are representative of three independent experiments. Bottom panel, expression levels of transfected proteins. FN, fibronectin; WB, Western blot.View Large Image Figure ViewerDownload Hi-res image Download (PPT)We next investigated whether AF6L inhibits cell adhesion. AF6L did not inhibit 007-induced adhesion to fibronectin, although it was expressed to a much higher level than AF6 (Fig. 4D
Referência(s)