Extracellular Alix regulates integrin-mediated cell adhesions and extracellular matrix assembly
2008; Springer Nature; Volume: 27; Issue: 15 Linguagem: Inglês
10.1038/emboj.2008.134
ISSN1460-2075
AutoresShujuan Pan, Ruoning Wang, Xi Zhou, Joe Corvera, Małgorzata Kloc, Richard N. Sifers, Gary E. Gallick, Sue-Hwa Lin, Jian Kuang,
Tópico(s)Monoclonal and Polyclonal Antibodies Research
ResumoArticle17 July 2008free access Extracellular Alix regulates integrin-mediated cell adhesions and extracellular matrix assembly Shujuan Pan Shujuan Pan Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Ruoning Wang Ruoning Wang Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Xi Zhou Xi Zhou Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Joe Corvera Joe Corvera A&G Pharmaceuticals Inc., Baltimore, MD, USA Search for more papers by this author Malgorzata Kloc Malgorzata Kloc Immuno-Biology Laboratory, The Methodist Hospital Research Institute, Houston, TX, USA Search for more papers by this author Richard Sifers Richard Sifers Department of Pathology, Baylor College of Medicine, Houston, TX, USA Search for more papers by this author Gary E Gallick Gary E Gallick Department of Cancer Biology, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Sue-Hwa Lin Sue-Hwa Lin Department of Molecular Pathology, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Jian Kuang Corresponding Author Jian Kuang Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Shujuan Pan Shujuan Pan Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Ruoning Wang Ruoning Wang Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Xi Zhou Xi Zhou Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Joe Corvera Joe Corvera A&G Pharmaceuticals Inc., Baltimore, MD, USA Search for more papers by this author Malgorzata Kloc Malgorzata Kloc Immuno-Biology Laboratory, The Methodist Hospital Research Institute, Houston, TX, USA Search for more papers by this author Richard Sifers Richard Sifers Department of Pathology, Baylor College of Medicine, Houston, TX, USA Search for more papers by this author Gary E Gallick Gary E Gallick Department of Cancer Biology, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Sue-Hwa Lin Sue-Hwa Lin Department of Molecular Pathology, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Jian Kuang Corresponding Author Jian Kuang Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA Search for more papers by this author Author Information Shujuan Pan1, Ruoning Wang1, Xi Zhou1, Joe Corvera2, Malgorzata Kloc3, Richard Sifers4, Gary E Gallick5, Sue-Hwa Lin6 and Jian Kuang 1 1Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA 2A&G Pharmaceuticals Inc., Baltimore, MD, USA 3Immuno-Biology Laboratory, The Methodist Hospital Research Institute, Houston, TX, USA 4Department of Pathology, Baylor College of Medicine, Houston, TX, USA 5Department of Cancer Biology, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA 6Department of Molecular Pathology, MD Anderson Cancer Center, The University of Texas, Houston, TX, USA *Corresponding author. Department of Experimental Therapeutics, MD Anderson Cancer Center, The University of Texas, 1515 Holcombe Blvd., Box 019, Houston, TX 77030, USA. Tel.: +1 713 792 8505; Fax: +1 713 792 3754; E-mail: [email protected] The EMBO Journal (2008)27:2077-2090https://doi.org/10.1038/emboj.2008.134 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Alix (ALG-2-interacting protein X), a cytoplasmic adaptor protein involved in endosomal sorting and actin cytoskeleton assembly, is required for the maintenance of fibroblast morphology. As Alix has sequence similarity to adhesin in Entamoeba histolytica, and we observed that Alix is secreted, we determined whether extracellular Alix affects fibroblast morphology. Here, we demonstrate that secreted Alix is deposited on the substratum of non-immortalized WI38 fibroblasts. Antibody binding to extracellular Alix retards WI38 cell adhesion and spreading on fibronectin and vitronectin. Alix knockdown in WI38 cells reduces spreading and fibronectin assembly, and the effect is partially complemented by coating recombinant Alix on the cell substratum. Immortalized NIH/3T3 fibroblasts deposit less Alix on the substratum and have defects in α5β1-integrin functions. Coating recombinant Alix on the culture substratum for NIH/3T3 cells promotes α5β1-integrin-mediated cell adhesions and fibronectin assembly, and these effects require the aa 605–709 region of Alix. These findings demonstrate that a sub-population of Alix localizes extracellularly and regulates integrin-mediated cell adhesions and fibronectin matrix assembly. Introduction Mammalian ALG-2-interacting protein X (Alix), also termed as AIP1 or Hp95, is an evolutionarily conserved and ubiquitously expressed adaptor protein (Missotten et al, 1999; Vito et al, 1999; Wu et al, 2001). Studies by others have demonstrated that the yeast orthologue of Alix is a crucial component of endosomal-sorting machinery (Nikko et al, 2003; Odorizzi et al, 2003; Luhtala and Odorizzi, 2004). This function of Alix explains the involvement of Alix in viral budding (Chatellard-Causse et al, 2002; Katoh et al, 2003; Martin-Serrano et al, 2003; Strack et al, 2003; Cabezas et al, 2005; Kim et al, 2005; Lee et al, 2007), cytokinesis (Carlton and Martin-Serrano, 2007; Morita et al, 2007) and potentially its role programmed cell death (Sadoul, 2006). Our study demonstrated that Alix is an F-actin-binding protein that physically associates with multiple actin cytoskeletal structures and functionally promotes actin cytoskeleton assembly (Pan et al, 2006). This cellular function of Alix is consistent with the requirement of Alix for non-immortalized human lung WI38 fibroblasts to maintain typical fibroblast morphology. In addition to these well-defined cellular functions of Alix, Alix also directly binds lipids and regulates membrane invagination (Dikic, 2004; Matsuo et al, 2004). Alix overexpression in malignant HeLa cells restores contact inhibition and anoikis (Wu et al, 2001, 2002). Alix overexpression also reduces the strength of the static cell–matrix adhesion in HEK293 cells and inhibits endocytosis of EGF receptors in CHO cells (Schmidt et al, 2003, 2004). Mechanisms of these biological functions of Alix are yet to be understood. Several observations led us to speculate that Alix is a member of the special class of non-transmembrane proteins that have functions on both sides of the plasma membrane (Nickel, 2003). We observed that the morphological defects of Alix-knockdown WI38 cells were much more severe after subculture than before subculture (Pan et al, 2006), which could not be satisfactorily explained by the intracellular roles of Alix in actin cytoskeleton assembly. We also demonstrated that both polyclonal and monoclonal anti-Alix antibodies consistently stained the substratum of WI38 cells (Figure 1), which was not observed with antibodies against bona fide intracellular proteins such as FAK, PYK2, cortactin, α-actinin, tensin and paxillin (data not shown). Examination of the protein databases for Alix-related proteins with defined extracellular functions revealed sequence homology between Alix and adhesin in the cytolytic enteric protozoan Entamoeba histolytica, an organism that causes amoebiasis in humans (Arroyo and Orozco, 1987; Rigothier et al, 1992; Garcia-Rivera et al, 1999; Banuelos et al, 2005). The protein sequence of the 76-kDa adhesin protein aligns with the portion of Alix containing both the N-terminal Bro1 domain (aa 1–358) and the middle V domain (aa 362–702) with 42% similarity (Banuelos et al, 2005). Most importantly, adhesin localizes both on the extracellular side of the plasma membrane and cytoplasmic vacuoles and in the cytoplasm (Garcia-Rivera et al, 1999; Madriz et al, 2004). The cell surface-localized adhesin is functionally important for heterophilic cell–cell adhesions that occur during the disease process (Garcia-Rivera et al, 1999; Martinez-Lopez et al, 2004). This finding suggests the possibility that Alix may have extracellular functions in regulating cell adhesions. Figure 1.Evidence that Alix is present in the extracellular compartment of WI38 cell cultures. (A) Monolayer cultures of WI38 cells were immunogold-labelled with 3A9 antibody, and electron micrographs of a sagittal section and a cross section of embedded samples are shown. Arrowheads indicate positive staining on the cell surface. Arrows indicate positive staining on the substratum. (B) Monolayer cultures of WI38 cells fixed with the EM fixative were immunostained under identical conditions with 3A9 antibody, mouse IgG (mIgG), rabbit anti-Alix immune serum (pAb) or pre-immune serum (pre-immune). Arrows indicate positive staining on the substratum and at the cell periphery. (C) Monolayer cultures of control and Alix-knockdown WI38 cells fixed with methanol were immunostained with 3A9 antibody (green) and counterstained with PI (red). Arrows indicate positive staining on the substratum. Download figure Download PowerPoint In this study, we determined the extracellular localization of Alix in cultured mammalian fibroblasts by immunological and biochemical approaches and demonstrated the presence of Alix in the culture substratum. To investigate the biological function of extracellular Alix, we determined the effects of defined, epitope-specific anti-Alix monoclonal antibodies and recombinant Alix coated on culture substrata on adhesion, spreading and fibronectin matrix assembly of mammalian fibroblasts. Our results provide strong evidence that extracellular Alix regulates integrin-mediated cell adhesions and extracellular matrix assembly. Results Alix is localized, in part, on the substratum of WI38 fibroblasts We previously generated multiple anti-Alix monoclonal antibodies to assist in the analysis of Alix biological and biochemical functions in human non-immortalized WI38 fibroblasts of fetal lung origin. By indirect immunofluorescence, 1A12 and 3A9 antibodies stained Alix associated with the actin cytoskeleton as well as with particulate structures in the cytoplasm (Pan et al, 2006). To characterize Alix-associated structures in WI38 cells in greater detail, we performed immunogold labelling of fixed and permeabilized monolayer cultures of WI38 cells with 3A9 antibody and sectioned embedded samples for electron microscopy (EM). For the immunogold labelling, WI38 cells were fixed with 2% formalin and 3% glutaraldehyde in 0.1 M sodium cacodylate (EM fixative) as compared with 4% paraformaldehyde used in our previous study for indirect immunofluorescence staining (Pan et al, 2006). Surprisingly, 3A9 antibody poorly stained the cytoplasmic Alix under the EM fixation condition; however, it stained the substratum and scattered protein aggregates near the cell surface (Figure 1A, upper panel), some of which appeared to be in the process of secretion (Figure 1A, lower panel). We then stained monolayer cultures of WI38 cells that had been fixed with the EM fixative with both monoclonal and polyclonal anti-Alix antibodies and examined the staining by indirect immunofluorescence microscopy. Both 3A9 monoclonal antibody and the rabbit anti-Alix immune serum stained the substratum, whereas the negative control antibody mouse IgG or the pre-immune serum did not (Figure 1B). We also stained control and Alix-knockdown WI38 cells that had been fixed with methanol with 1A12 anti-Alix monoclonal antibody under identical conditions. 1A12 antibody stained both the cytoplasm and the substratum, and Alix knockdown dramatically reduced the staining in both locations (Figure 1C). These findings, in conjunction with the sequence homology between the Bro1 and V domain portion of Alix and the entire length of adhesin (Supplementary Figure S1) suggested that a sub-population of Alix is secreted from WI38 cells and deposited onto the substratum. Full-length Alix is present both in the conditioned medium and on the substratum To test the hypothesis that a sub-population of Alix is secreted from WI38 cells, we fractionated the conditioned medium collected from WI38 cell cultures and determined whether it contained Alix that could not be accounted for by cell lysis. Figure 2A shows that although Alix was undetectable in the 1000 and 10 000 g pellets, which contained dead cells and membrane debris, respectively, full-length Alix was readily and reproducibly detected in the 100 000 g pellet, presumably containing large protein complexes and small vesicles (Odorizzi et al, 2003). Low levels of cleaved Alix were sometimes detectable in the 100 000 g supernatant, and this could be due to low levels of cell lysis. Figure 2B shows that Superose 6 gel filtration of proteins extracted from the 100 000 g pellet by multiple detergent-containing RIPA buffer resulted in one peak of Alix in the void fractions (at least 5000 kDa), whereas Superose 6 gel filtration of the postnuclear lysates of WI38 cells had the majority of Alix recovered in the 158-kDa fractions and only ∼5% of Alix in the void fractions. As cell lysis is unlikely to generate a distinct peak of full-length Alix of ∼5000 kDa, the most plausible explanation for these results is that a high molecular weight complex of Alix is secreted from WI38 cells. Figure 2.Full-length Alix is present both in the conditioned medium and on the substratum of WI38 cell cultures. (A) Indicated fractions from the conditioned medium collected from WI38 cell cultures and 1/10 of cell lysates from the same cultures were immunoblotted in parallel with anti-Alix antibodies. P: pellet fraction. SN: supernatant. The asterisk indicates a cleavage product of Alix. (B) Cell lysates (CL) and protein extracts of the 100 000 g pellet fraction of the conditioned medium (CM) were fractionated by Superose 6 gel filtration, and TCA-precipitated proteins from the indicated fractions were immunoblotted with anti-Alix antibodies. (C) After live monolayer cultures of control or Alix-knockdown (Alix (−)) WI38 cells were labelled with each of the indicated antibodies, cells were fixed, permeabilized and stained with FITC-conjugated secondary antibodies (green) and TRITC-conjugated phalloidin (red). Arrows and arrowheads indicate particulate staining in the substratum and on the cell surface, respectively. (D) After live culture of WI38 cells were labelled with 1A3 antibody, fixed and permeabilized cells were labelled with anti-fibronectin (FN) antibodies. Cells were then stained with Texas-red-conjugated anti-mouse IgG for 1A3-labelled Alix (red) and FITC-conjugated anti-rabbit IgG for FN (green), and counterstained with DAPI (blue). (E) Monolayer cultures of WI38 cells were biotinylated, and derived cell lysates were immunoprecipitated with antibodies for each of the indicated proteins. Crude cell lysates and the immunoprecipitates were immunoblotted for each of the precipitated proteins (left panel) and probed with streptavidin (right panel) as indicated. Download figure Download PowerPoint To test the hypothesis that the secreted Alix is deposited onto the substratum, we labelled live monolayer cultures of WI38 cells with each of four different anti-Alix monoclonal antibodies or control antibodies at 4°C for 30 min. By immunoblotting specific GST-tagged Alix fragments (Supplementary Figure S2A), we determined that the 1A12 and 3A9 antibodies recognize the aa 605–709 region (Supplementary Figure S2B and data not shown), and the 1A3 antibody recognizes the aa 168–436 region of Alix (Supplementary Figure S2C). In contrast to these three antibodies, 2H12 antibody had been determined to recognize the three-dimensional F676 pocket in the middle V-domain, which is hidden in the cytosolic Alix (Zhou et al, 2008). The primary antibody staining was followed by labelling fixed and permeabilized cells with fluorescence-labelled secondary antibodies and phalloidin, which decorates F-actin in the cytoplasm. We observed that both 1A12 and 3A9 antibodies stained small particles that distributed across the substratum and that the particles appeared to be more concentrated in the area adjacent to the cell periphery. These immuno-positive particles were also detectable on the cell surface but at a much lower density than on the substratum. 1A3 antibody not only stained the particles on the substratum but also fibres and clumps at the cell periphery or surrounding areas. Pre-neutralization of 1A12 antibody with recombinant Alix eliminated the ability of the antibody to stain the substratum. siRNA-mediated Alix knockdown eliminated the extracellular staining by both 1A12 and 1A3 antibodies. In contrast to 1A12, 3A9 and 1A3 antibodies, antibodies against the cell surface receptor transferrin stained the cell surface but not the substratum. 2H12 antibody, mouse IgG or antibodies against the intracellular protein clathrin stained neither the cell surface area nor the substratum (Figure 2C). Taken together, these results demonstrate that Alix is present in the substratum of WI38 cells. To characterize the extracellular structures recognized by 1A3 antibody, we labelled live cultures of WI38 cells with 1A3 antibody and stained fixed cells with anti-fibronectin antibodies. The 1A3 antibody-decorated clumps and fibre overlapped with the clumps and fibres labelled by anti-fibronectin antibodies (Figure 2D), indicating that extracellular Alix is associated with assembled fibronectin. To biochemically characterize the Alix in the extracellular compartment, we biotinylated monolayer cultures of WI38 cells with a membrane non-permeable biotinylation agent and immunoprecipitated Alix in parallel with the cell surface receptor transferrin and abundant intracellular proteins p53 and GSK3β from cell lysates, followed by probing biotinylated proteins with streptavidin. As expected, streptavidin did not stain p53 or GSK3β. However, both Alix and transferrin were stained by streptavidin (Figure 2E), demonstrating that full-length Alix is present on the substratum of WI38 cells. Extracellular Alix contributes to the maintenance of fibroblast morphology of WI38 cells We previously reported that knockdown of Alix expression in WI38 fibroblasts led to a rounded cell morphology (Pan et al, 2006). To determine whether extracellular Alix contributes to the maintenance of WI38 fibroblast morphology, we utilized both loss-of-function and gain-of-function approaches. In the loss-of-function approach, we seeded WI38 cells in the presence of 1A12 or 3A9 antibody or, as a negative control, mouse IgG, and determined the effect of each of these antibodies on cell adhesion and spreading. Although 1A12 and 3A9 antibodies did not block cell attachment to the substratum, they reduced the rate of cell attachment within the first hour by ∼50 and 70%, respectively, whereas mouse IgG had no inhibitory effect (Figure 3A). Alix knockdown by transfection with Alix-specific siRNA almost eliminated the inhibitory effect of 1A12 antibody on the initial rate of cell attachment (Figure 3B), strongly suggesting that the inhibition was due to antibody binding to extracellular Alix. We also determined the effect of 1A12 antibody on the initial rate of WI38 cell attachment and spreading on culture substrata coated with fibronectin, vitronectin, collagen or the control polypeptide poly-L-lysine. After 1 h, 1A12 antibody caused 55 and 65% inhibition in cell attachment to fibronectin- and vitronectin-coated substrata, respectively, whereas this antibody had little or no effect on cell attachment to collagen- or poly-L-lysine-coated substrata (Figure 3C). When cell attachment neared completion, WI38 cells cultured in the presence of 1A12 antibody spread less than WI38 cells treated with mouse IgG on both vitronectin- or fibronectin-coated substrata (Figure 3D and E). As α5β1- and αvβ3-integrins are the major receptors for fibronectin and vitronectin and have key functions in determining mammalian fibroblast morphology (Akiyama, 1996; Giancotti and Ruoslahti, 1999), these results suggest that binding of extracellular Alix negatively impacts α5β1- and αvβ3-integrin-mediated cell adhesions. Figure 3.Anti-Alix antibodies inhibit integrin-mediated cell adhesions. (A) WI38 cells were seeded in the presence of each of the indicated antibodies, and relative cell attachments were determined at 1 h after cell seeding. Results were normalized against the value from mouse IgG (mIgG)-treated cells, and presented results are averages from three independent experiments. Error bars indicate standard errors of mean (s.e.m.). (B) Control and Alix-knockdown WI38 cells were seeded in the presence of 1A12 antibody or mIgG, and relative cell attachments were determined at 1 h after cell seeding. Presented results are averages from three independent experiments, and the error bars indicate standard errors of mean (s.e.m.). (C) WI38 cells were seeded onto the substratum that was pre-coated with FN, vitronectin (VN), collagen (CN) or poly-L-lysine (PLL) in the presence of 1A12 antibody or mIgG, and relative cell attachments on each of the coated proteins were determined at 1 h as described for (A). Results are from a representative experiment out of three, and error bars indicate standard deviations. (D, E) WI38 cells were seeded as described for (C), and cells were stained with crystal violet at 2 h after cell seeding and photographed (D). The relative spreading area per cell was determined by analysis of digitized images with Metamorph software, and the average was calculated and normalized against the value of mIgG-treated cells (E). The P-values were determined using Student's t-tests. Download figure Download PowerPoint In the gain-of-function approach, we cultured control and Alix-knockdown WI38 cells on substratum pre-coated with either GST or GST–Alix, and determined the effect of extracellular addition of recombinant Alix on cell spreading and fibronectin matrix assembly. This approach was previously used to determine the effect of fibronectin on cell morphology and proliferation (Yamada et al, 1976, 1978; Ali et al, 1977; Yamada, 1978). Although the coated Alix did not produce noticeable effects on the morphology of control WI38 cells, which were well spread, it partially rescued the spreading defect of Alix-knockdown WI38 cells (Figure 4A and B). The fact that the rescue was only partial could be explained by intracellular effects of Alix knockdown on actin cytoskeleton assembly, which is closely linked to cell morphology and fibronectin assembly (Brakebusch and Fassler, 2003). Biochemical measurement of deoxycholate (DOC)-insoluble fibronectin, which represents assembled fibronectin matrix (McKeown-Longo and Mosher, 1983), in parallel with soluble and total fibronectin showed that Alix knockdown inhibited fibronectin matrix assembly of WI38 cells without inhibiting fibronectin expression. In both control and Alix-knockdown cells, Alix coated on the substratum promoted fibronectin matrix assembly without affecting fibronectin expression (Figure 4C). These observations were further confirmed by immunofluorescence staining of fibronectin in these cells (Figure 4D). Taken together, these results demonstrate that recombinant Alix coated on the substratum promotes WI38 cell spreading and fibronectin matrix assembly. Figure 4.Recombinant Alix coated on the substratum partially rescues the defects of Alix-knockdown WI38 fibroblasts in cell spreading and fibronectin matrix assembly. (A) After control or Alix-knockdown WI38 cells were grown on GST- or GST–Alix-coated coverslips for 24 h, cells were observed under a phase-contrast microscope and images were taken at × 100 magnification. (B) After five fields of cells randomly photographed from each treatment in (A) were enlarged and printed, individual cell lengths were manually measured and the relative average cell length and s.e.m. among different fields (error bars) were calculated. Statistical analysis of the significance was performed by Student's t-tests. (C) After control or Alix-knockdown WI38 cells were grown on GST- or GST–Alix-coated coverslips for 48 h, total proteins or DOC-soluble and DOC-insoluble proteins were extracted. Whereas total proteins were immunoblotted with anti-Alix, anti-FN and anti-actin antibodies, DOC-soluble and DOC-insoluble proteins were immunoblotted with anti-fibronectin antibodies. (D) Control or Alix-knockdown WI38 cells grown on GST- or GST–Alix-coated coverslips for 48 h were immunostained with anti-FN antibodies (green) and counterstained with PI (red). Download figure Download PowerPoint Extracellular recombinant Alix promotes fibronectin matrix assembly of NIH3T3 cells To investigate the mechanism by which extracellular Alix promotes spreading, integrin-mediated cell adhesions and fibronectin matrix assembly of mammalian fibroblasts, we chose immortalized mouse NIH/3T3 fibroblasts due to their characteristic morphology, that is, less elongation, failure to align at high density and assembly of fewer fibronectin fibres at the cell–matrix interface relative to WI38 cells (Supplementary Figure S3A and B). Also, growing NIH/3T3 cells on fibronectin-coated substrata promotes both cell spreading and alignment (Supplementary Figure S3C). Although NIH/3T3 cells express similar levels of Alix as WI38 cells (Supplementary Figure S3D), three-fold less Alix was detected on the substratum of NIH/3T3 cells as compared with WI38 cells (Supplementary Figure S3E). Alix overexpression in immortalized mouse fibroblast NIH/3T3 cells promoted cell spreading and alignment (Wu et al, 2002). When NIH/3T3 cells were grown on non-coated, GST-coated or GST–Alix-coated substrata, the coated Alix had little effect on cell adhesion (Supplementary Figure S4A). However, the coated Alix promoted cell spreading and cell–cell alignment similar to coated fibronectin (Figure 5A; Supplementary Figure S4B). In parallel with these effects, cells grown on GST–Alix-coated substrata assembled more fibronectin fibres than cells grown on non-coated or GST-coated substrata, as determined by both immunofluorescence staining of fibronectin (Figure 5B) and biochemical measurement of DOC-insoluble fibronectin (Figure 5C). The promoting effect of the coated Alix on fibronectin assembly was observed at 2 h after cell seeding and was maintained for at least 48 h (Supplementary Figure S4C). The effect was produced in the absence of increases in the expression level of fibronectin or its receptor α5β1-integrin (Figure 5D). These results both support the conclusion that extracellular Alix promotes fibroblast cell spreading and fibronectin assembly and demonstrate that NIH/3T3 cells are a suitable experimental system to study the mechanism by which extracellular Alix performs these functions. Figure 5.Recombinant Alix coated on the substratum promotes NIH/3T3 cell spreading, alignment and fibronectin matrix assembly. (A) NIH/3T3 cells were seeded onto mock-, GST- or GST–Alix-coated coverslips, and cell images were then taken under a phase-contrast microscope at low and high cell densities. (B) NIH/3T3 cells were cultured on mock-, GST- or GST–Alix-coated coverslips and cultured for 48 h, and cells were immunostained with anti-fibronectin antibodies (green) and counterstained with PI (red). (C) DOC-soluble and DOC-insoluble fractions of the proteins were extracted and immunoblotted with anti-FN antibodies. (D) Total proteins extracted from NIH/3T3 cells grown on mock-, GST- or GST–Alix-coated coverslips were immunoblotted with antibodies for each of the indicated proteins. Download figure Download PowerPoint Extracellular recombinant Alix promotes α5β1-integrin-mediated cell adhesions in NIH/3T3 cells Previous studies have demonstrated that both α5β1- and αvβ3-integrins are capable of forming focal adhesions. In contrast, only α5β1-integrins that display high-affinity and translocation-competent conformations are able to translocate along with tensin from focal adhesions at cell periphery into fibrillar adhesions at cell centre. This latter conformation can be experimentally detected by conformation-sensitive antibodies such as 9EG7 and SNAKA51 monoclonal antibodies (Pankov et al, 2000; Clark et al, 2005). The translocation of α5β1-integrin ‘stretches’ the fibronectin that links to α5β1-integrin and induces fibronectin matrix assembly (Zamir et al, 1999, 2000; Pankov et al, 2000; Geiger et al, 2001; Mao and Schwarzbauer, 2005). Thus, to characterize the role of extracellular Alix in cell spreading and fibronectin matrix assembly, we determined the effect of blocking α5β1-integrin functions on adhesion, spreading and fibronectin assembly of NIH/3T3 cells on GST- or GST–Alix-coated substrata. As shown in Figure 6A and B, the α5β1-integrin functional blocking antibody did not inhibit adhesion, spreading and fibronectin assembly of NIH/3T3 cells grown on GST–coated substrata, indicating that NIH/3T3 cells have defects in α5β1-integrin-mediated cell adh
Referência(s)