Integrin ␣v3-Mediated Endocytosis of Immobilized Fibrinogen by A549 Lung Alveolar Epithelial Cells Tatjana M. Odrljin,* Constantine G. Haidaris, Norma B. Lerner, and Patricia J. Simpson-Haidaris Departments of Medicine/Vascular Medicine Unit, Microbiology and Immunology, Center for Oral Biology, Pediatrics, and Pathology and Laboratory Medicine, University of Rochester School of Medicine and Dentistry, Rochester, New York
Fibrinogen (FBG), together with its polymerized form fibrin, modulates cellular responses during wound repair and tissue remodeling. Thus, we sought to determine whether A549 lung epithelial type II–like cells would endocytose insoluble, surface-bound FBG as a potential mechanism of alveolar matrix remodeling. Surface-bound FBG was endocytosed into either lysosomes or late endosomes by A549 cells through arg-glyasp–dependent binding to ␣v3 but not ␣51 integrin receptors. Soluble FBG added to confluent monolayers of A549 cells was not endocytosed. Unlike the uptake of the extracellular matrix glycoproteins vitronectin and thrombospondin by other cell types, endocytosis of FBG by A549 cells was neither inhibited by heparin nor dependent on binding to cell-surface heparan sulfate proteoglycans. FBG did not colocalize with endocytosed transferrin, whereas dextran showed partial colocalization with FBG in endocytic vesicles, suggesting nonclathrin-mediated endocytosis. Inhibition of actin filament polymerization blocked endocytosis of both dextran and FBG but not transferrin, providing further support that FBG is endocytosed via a nonclathrin pathway. Disruption of actin polymerization inhibited integrin-mediated cell spreading, which contributed to an overall reduction in FBG clearance that was most likely due to reduced cell migration and associated pericellular proteolysis. Trasylol inhibition of extracellular plasmin activity did not inhibit endocytosis of FBG. The endocytosed FBG was degraded to trichloroacetic acid–soluble fragments that showed an electrophoretic pattern distinctly different from plasmin-degraded FBG. Together, these results suggest that endocytosis of matrix-associated FBG by alveolar epithelial cells may be involved in the processes of alveolar tissue repair and matrix remodeling.
Fibrinogen (FBG) functions in primary hemostasis in the support of platelet aggregation and in secondary hemostasis in the formation of an insoluble fibrin clot (1). FBG and fibrin are also found in tumor matrices, and in the provisional matrices of cutaneous wounds and areas of inflammation. This provisional matrix is composed of a complex of noncollagenous adhesive glycoproteins (2). These include thrombospondin, vitronectin (VN), and fibronectin (FN), which can also be crosslinked into the fibrin gel. (Received in original form October 22, 1999 and in revised form July 27, 2000) Address correspondence to: P. J. Simpson-Haidaris, Ph.D., Vascular Medicine Unit/Department of Medicine, P.O. Box 610, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Rochester, NY 14642. E-mail:
[email protected] *Current address: Learner Research Institute, Molecular Cardiology, NB 50, The Cleveland Clinic, 9500 Euclid Ave., Cleveland, OH 44127. Abbreviations: extracellular matrix, ECM; ethylenediaminetetraacetic acid, EDTA; fibrinogen, FBG; fluorescein isothiocyanate, FITC; fibronectin, FN; heparan sulfate proteoglycans, HSPG; monoclonal antibody(ies), MoAb; phosphate-buffered saline, PBS; polyclonal antibody(ies), PoAb; arg-gly-asp, RGD; sodium dodecyl sulfate polyacrylamide gel electrophoresis, SDS-PAGE; trichloroacetic acid, TCA; vitronectin, VN; 4-methylumbelliferyl-7--D-xyloside, xyloside. Am. J. Respir. Cell Mol. Biol. Vol. 24, pp. 12–21, 2001 Internet address: www.atsjournals.org
These adhesive glycoproteins are soluble in plasma and insoluble in extracellular matrix (ECM), and are released at the site of injury from platelet storage ␣-granules (1). In addition, platelet releasate is rich in growth factors, such as platelet-derived growth factor and transforming growth factor-, which bind to components of the ECM or denuded basement membrane. The provisional matrix provides a reservoir for growth factors and a structural scaffold to support cell adhesion, spreading, migration, and proliferation during wound repair (2). Models of acute lung injury focus on the role of the provisional matrix in promoting re-epithelialization of denuded basement membranes. Alveolar type II pneumocytes are the progenitor cells that restore epithelial architecture by proliferation and differentiation into type I cells during episodes of lung injury and inflammation (2). During extensive pulmonary injury, FBG and other plasma proteins flood alveoli due to increased permeability of the endothelial and epithelial barriers. Inflammatory cells recruited to the injured alveoli express procoagulant activity to initiate clot formation (3) and both FBG and fibrin support the binding of fibroblast growth factor-2 (4), which is important in wound repair and angiogenesis (2). Thus, fibrin(ogen) at the site of wound repair provides a provisional matrix to which growth factors and adhesive glycoproteins bind. The altered topology of the provisional matrix signals cells to respond to the injury in a manner to promote repopulation of a denuded basement membrane. Subsequent to tissue repair, the transient provisional matrix is replaced with established matrix and basement membrane constituents. Because fibrin(ogen) is a component of the provisional matrix, its interactions with various cell types in support of cellular processes have been extensively studied (References 1 and 5 and references therein). These cellular responses are mediated, in part, by cell-surface integrin receptors. Fibrin(ogen) binds to the 3-containing integrin receptors ␣IIb3 and ␣v3. FBG also binds to the 1 class of FN receptors, including ␣51. The ␣v3 integrin receptor participates in the adhesion of many cell types to FBG, including alveolar type II cells (6, 7). The deposition of FBG in the ECM occurs in the absence of thrombin or plasmin cleavage (8), thus exposure of pneumocytes to matrix FBG may occur without its subsequent conversion to fibrin. Whereas fibrin is known to mediate cell adhesion and spreading, migration, and new blood vessel formation (References 1 and 5 and references therein), the function of insoluble matrix FBG is not well defined. Further, although the endocytosis of surface immobilized VN involves binding to ␣v5 in addition to heparan sulfate proteoglycans (HSPG) (9, 10), turnover of FBG during remodeling of the provisional matrix is not well characterized.
Odrljin, Haidaris, Lerner, et al.: Endocytosis of Immobilized Fibrinogen by Pneumocytes
Because ␣v3 on alveolar type II cells is known to interact with FBG (6, 7), and FBG contains heparin-binding domains (11), we investigated whether integrins and HSPG promote clearance of surface immobilized FBG via endocytosis by an alveolar type II–like cell. We used a traditional in vitro cell culture model (6, 7) to evaluate the cellular response of A549 cells to a single matrix molecule immobilized on the tissue culture surface. We present evidence that: (1) lung A549 epithelial cells adhere to immobilized FBG via ␣v3 integrin receptors, (2) FBG endocytosis is mediated by arg-gly-asp (RGD)–dependent binding to ␣v3, (3) in contrast to VN and thrombospondin, endocytosis of FBG is neither inhibited by heparin nor dependent on HSPG, (4) internalization of FBG occurs via a nonclathrin pathway, and (5) subsequent endocytosis of FBG does not require plasmin proteolysis. These results suggest that the ordered process of alveolar epithelium remodeling involves integrin-dependent clearance of FBG.
Materials and Methods Cells and Culture Conditions A549 human lung epithelial carcinoma cells (CCL 185) were obtained from the American Type Culture Collection (ATCC, Manassas, VA). Cells were grown in Kaighn’s Nutrient Mixture F12 medium (Irvine Scientific, Irvine, CA) containing 10% fetal bovine serum (FBS) (Intergen, Purchase, NY), penicillin (100 U/ml), streptomycin (0.1 mg/ml), and L-glutamine (2 mM) (Life Technologies, Gaithersburg, MD). Cells were detached with 5 mM ethylenediaminetetraacetic acid (EDTA) or trypsin/EDTA (Life Technologies) for 5 min; trypsin action was stopped with addition of Kaighn’s media with 10% FBS. Experiments were performed with A549 cells at passage number 7-15 from cells obtained from ATCC.
Preparation of Surface-Immobilized FBG To remove contaminating plasminogen and FN, FBG (Calbiochem, La Jolla, CA) was purified as previously described (12). Conjugation of FBG to Oregon-Green fluorophore was performed using the Protein Labeling Kit FluoReporter Oregon Green 488 from Molecular Probes (Eugene, OR). Unlabeled FBG was used as specifically noted in the figure captions. Glass coverslips placed into wells of Corning 24-well tissue culture plates were coated with FBG diluted to 40 g/ml in phosphate-buffered saline (PBS). After overnight incubation at 4⬚C, coverslips were washed extensively with PBS to remove unbound FBG.
Immunofluorescent Staining Primary antibodies used in this study include rabbit polyclonal antibodies (PoAbs) to ␣v3 (FBG/VN-receptor) and ␣51 (FN/ FBG-receptor) (Chemicon, Temecula, CA; and Life Technologies) and antihuman FBG (Dako, Carpinteria, CA), which was further purified as described (11). Monoclonal antibodies (MoAb) used were to vinculin, CD71 (transferrin receptor) (Sigma, St. Louis, MO), and HSPG (Seikagaku, Jamesville, MD). Secondary antibodies against rabbit and mouse immunoglobulin G were conjugated to rhodamine, fluorescein isothiocyanate (FITC), or phycoerythrin (Dako or Molecular Probes). Rhodamine-phalloidin from Molecular Probes was used to visualize polymerized Factin. All antibodies were used for immunofluorescent staining at dilutions recommended by the manufacturers. Before fixation in 3.7% formaldehyde, cells were washed three times with PBS. Cells were permeabilized, when required, with 0.5% Triton X-100 for 20 min and further stained with antibodies. Immunofluorescent staining of cells was performed as previously described (13).
13
Confocal Scanning Laser Cytometry The cellular locale of actin, FBG, and integrin ␣v3 was determined using a Meridian Ultima Adherent Cell Analysis system and the Data Analysis System Master Program V3.323 (Meridian Instruments, Inc., Okemos, MI). All samples were examined and scanned using an identical set of parameters so that results from different cells could be compared (14). A 488-nm argon laser line excited samples, and a photomultiplier tube detected emitted fluorescence with an upper limit at 575 nm. Data were collected from serial, 1-m-thick optical sections along the z-axis (vertical axis) beginning below the basolateral face to the apical surface of the cell. Each optical section is labeled in micrometers, which is representative of the distance into the cell from the basolateral cell surface. From the data points collected, arbitrary-color digitized images of scanned cells were generated. A relative fluorescence intensity scale was determined for a series of vesicular structures falling within a line of query of a defined optical section.
Treatment of Cells with Inhibitors of Cellular Processes To inhibit glycosaminoglycan addition to the protein core during proteoglycan synthesis, A549 cells were treated for 24 h with 2 mM 4-methylumbelliferyl-7--D-xyloside (xyloside) before plating on FBG–Oregon Green. The cells were incubated on FBG–Oregon Green for 18 h in the continued presence of 0.5 mM xyloside. To specifically cleave HSPG, A549 cells were detached with trypsinEDTA, washed in Eagle’s minimum essential medium (MEM) ⫹ 0.1% bovine serum albumin (BSA), then treated with the same medium containing 3 U/ml heparitinase for 4 h at 37 ⬚C. To prevent expression of newly synthesized HSPG after heparitinase treatment, cells were washed three times in MEM ⫹ 0.1% BSA and then plated on FBG–Oregon Green–coated glass coverslips in complete medium containing 0.5 mM xyloside for 18 h at 37 ⬚C. Endocytosis of FBG was visualized by direct fluorescence; indirect immunofluorescent staining with anti-HSPG MoAb was used to monitor HSPG modifications by xyloside and heparitinase. Cells were metabolically labeled with 35SO4 to specifically label proteoglycans in the presence or absence of xyloside as described earlier. Newly sulfated proteoglycans were selectively precipitated with cetylpyridinium chloride to measure the amount of new proteoglycan synthesis (15). To determine whether A549 cells bound to surface immobilized FBG–Oregon Green via ␣v3, cells were stained with PoAb specific for ␣v3 and MoAb for vinculin. To block focal contact assembly on surface immobilized FBG, cells were pretreated for 15 min at 4⬚C with 1 M echistatin (Sigma) or left untreated before plating on FBG and during the subsequent 4-h incubation at 37⬚C (16). Focal adhesion plaques were monitored by immunofluorescent staining with MoAb specific for vinculin. In the experiments where kistrin (Sigma) was used to block endocytosis of FBG (17), the cells were pretreated for 15 min on ice with 1, 0.25, or 0.062 M kistrin, or left untreated. After plating on FBG–Oregon Green, cells were incubated in the continued presence of the same concentration of kistrin or left untreated for 18 h at 37 ⬚C. Alternatively, when echistatin was used (16), the cells were pretreated with 1 or 0.2 M echistatin or left untreated. After plating, cells were treated for 18 h with lower concentrations of echistatin: 0.175 M, 0.035 M, or none. To inhibit lysosomal degradation of endocytosed FBG, chloroquine (Sigma) was added to the cells at a final concentration of 50 M. A549 cells were pretreated with chloroquine for 1.5 h before plating on FBG–Oregon Green–coated coverslips, and incubated in the presence of chloroquine during the 18-h endocytosis assay. Alternatively, A549 cells were plated and incubated for 18 h on iodinated FBG prebound to the surface of six-well culture plates. The iodinated FBG starting material was greater than 95% trichloroacetic acid (TCA)–precipitable after passing over a PD10
14
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001
column before starting the experiment. After endocytosis of 125 I-FBG for 18 h, cells were detached with trypsin-EDTA and placed on ice to prevent further processing of internalized FBG. After extensive washing at 4⬚C, protein remaining on the surface of the cells was stripped in 0.2 M acetic acid, pH 2.5, containing 0.5 M NaCl at 4⬚C for 15 min. The acid-stripped cells were incubated in PBS at 37⬚C for 90 min and the 125I-FBG/FBG degradation products released into the buffer were precipitated on ice with an equal volume of 10% TCA. Total and TCA-precipitable counts were determined; the results were presented as %–TCA soluble counts on the basis of the difference in precipitable and total counts of 125I-FBG released into the buffer from the acidstripped cells. In some experiments, echistatin-treated cells were plated on 125I-FBG and treated with 50 M chloroquine during the 18-h endocytosis incubation at 37⬚C to prevent complete degradation of internalized FBG. The cells were lifted and acid-stripped as described earlier to remove cell surface–bound proteins. The cells were lysed and intracellular FBG was immunopurified and then analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and autoradiography (8). To promote actin depolymerization, cytochalasin D (10 g/ml) (Sigma) was added to the cells and incubated at 37 ⬚C for 1 h before plating of the cells on FBG–Oregon Green. Alternatively, cytochalasin D was added to the cells 2 h after plating on FBG– Oregon Green. Similar results were obtained with both treatments. To monitor endocytosis via the clathrin pathway (18), FITC-conjugated transferrin (330 g/ml) was added directly to the media at the time of cell plating. To monitor fluid phase uptake (19), dextran-70 conjugated to rhodamine was added to media at a concentration of 1 mg/ml at the time of cell plating on FBG–Oregon Green; each marker was present during the 18-h assay. To prevent plasmin-mediated proteolysis, Trasylol (Miles, Kankakee, IL) was added to A549 cells at a final concentration of 20 KIU/ml for 30 min before plating on FBG–Oregon Green, as well as during the 18-h endocytosis assay. In some experiments, A549 cells in the presence or absence of Trasylol were plated on iodinated FBG prebound to the surface of six-well tissue culture dishes (8). After 18 h of incubation at 37⬚C, the A549 cells were detached from the surface with 5 mM EDTA in PBS; the material remaining on the plates was collected in standard lysis buffer (8) and the FBG was immunopurified and then analyzed by SDS-PAGE and autoradiography. During immunopurification Trasylol was added at the same concentration to all buffers. The relative intensity of intact FBG and degradation products was measured by densitometry using the NIH Image 1.59 program (NIH, Bethesda, MD).
structures found in the cytoplasm. Additional evidence that the FBG-containing vesicles were intracellular was obtained by localization of the endocytosed FBG–Oregon Green within the same optical sections of the cell as F-actin (Figures 1b, 1d, and 1f). Note that the intensity of intracellular F-actin staining is constant along the z-axis in all of the optical sections. To determine whether A549 cells endocytosed soluble FBG, A549 cells were plated on plain glass coverslips, then soluble FBG–Oregon Green (40 g/ml) was added for 18 h. Confocal microscopy indicated that soluble FBG–Oregon Green was not endocytosed by A549 cells (not shown). Endocytosed FBG Colocalizes with Integrin ␣v3 in Lung Epithelial Cells Colocalization of FBG–Oregon Green with ␣51 or ␣v3 was assessed by indirect immunofluorescent staining and by confocal microscopy as described earlier. Results from both indirect immunofluorescence with anti-human FBG PoAb (not shown) and direct fluorescence with FBG–Oregon Green indicate that FBG was found in vesicular structures that colocalized with ␣v3 in the cytoplasmic region of the
Results A549 Cells Endocytosed Surface Immobilized FBG To determine whether A549 cells endocytosed immobilized FBG, confocal microscopy was performed. A549 cells plated on FBG–Oregon Green-coated glass coverslips were incubated at 37⬚C for 18 h. Serial optical sections were examined by confocal microscopy to localize both extracellular and intracellular FBG–Oregon Green and intracellular F-actin. Representative optical sections of the basal, middle, and apical portions of a single cell are shown in Figure 1. At the basal face of the cell, the most intense FBG staining was found extracellularly, as would be expected for uniform coating of FBG–Oregon Green on the coverslip. However, as the optical sections move along the z-axis toward the apical face of the cell, the intracellular FBG staining is retained while the extracellular FBG becomes undetectable. These results indicate that FBG (Figures 1a, 1c, and 1e) was internalized by the A549 cells into vesicular
Figure 1. Surface-bound FBG is internalized by cultured alveolar epithelial cells. Confocal scanning laser cytometry was used to analyze the intracellular versus extracellular distribution of FBG– Oregon Green (a, c, and e) compared with the intracellular marker F-actin (b, d, and f). The serial optical sections shown from left to right correspond to 1-m sections located 2, 3, and 5 m through the z-axis starting from the basolateral aspect of the cell surface and ending at the apical face. The relative fluorescence intensity is denoted by the color bar at the top, which progresses from magenta through red to orange for intense staining, to yellow, green, blue, and then black to represent low to no fluorescence. The extracellular FBG–Oregon Green immobilized on the surface of the glass coverslip fluoresces intensely outside of the cell (a). The extracellular FBG is undetectable at the apical cell surface (e, black). In contrast, the fluorescence intensity of the internalized FBG–Oregon Green, which ranges from blue-green to orange in vesicle-like structures, remains constant throughout the serial optical sections (a, c, and e, arrows). Bar: 10 m.
Odrljin, Haidaris, Lerner, et al.: Endocytosis of Immobilized Fibrinogen by Pneumocytes
cell (Figure 2); however, the FN receptor ␣51 did not colocalize with FBG in vesicular structures (not shown). Instead, the cells stained diffusely for ␣51, indicating that A549 cell adhesion and spreading on the immobilized FBG substratum was not due to engagement of ␣51 receptors. To determine the coincidence of internalized FBG with ␣v3, a relative fluorescence intensity scale was determined for a series of vesicular structures falling within a line of query in the 1-m optical section located 6 m above the basal surface of the cell (Figure 2A, panels e and f; and Figure 2B). This data indicates that peaks of FBG fluorescence correspond to abundant ␣v3 staining in the same optical sections. The data indicate further that all intracellular FBG localized to regions within the cell that stain brightly for ␣v3, whereas the converse is not true; ␣v3-positive staining did not always coincide with FBG–Oregon Green in vesicular structures (Figure 2A, panels c and d). Indirect immunofluorescence microscopy revealed that ␣v3 on the cell surface showed the expected codistribution with vinculin in focal adhesion contacts of A549 lung epithelial cells plated on immobilized FBG (not shown). Endocytosis of FBG Is RGD-Dependent and ␣v3 Integrin Receptor–Dependent To determine whether endocytosis was integrin receptor– dependent, we used the disintegrins echistatin and kistrin,
15
well-described inhibitors of RGD-dependent integrin binding to their cognate ligands, including FBG (17, 20). Both disintegrins affect ␣v3 ligand binding; echistatin also blocks ␣51 with lower affinity (21) and kistrin blocks ␣v5 but not ␣51 (17). Typical endocytosis and clearing of surfacebound FBG is shown in Figure 3A. The surface area surrounding these cells is devoid of fluorescence, indicating that all of the extracellular FBG was cleared from the pericellular matrix. This is compared with the view in Figure 3D where the FBG–Oregon Green fluorescence intensity remains high on the extracellular surfaces, indicating that kistrin effectively inhibited the clearance of the immobilized extracellular FBG. Kistrin inhibition of endocytosis was dose-dependent because endocytosis of surface-immobilized FBG was partially inhibited by kistrin (Figure 3B compared with Figure 3A) and echistatin (not shown) at low concentrations, and totally inhibited by higher concentrations of kistrin (Figures 3C and 3D) and echistatin (not shown). Further, kistrin treatment caused partial disruption of adhesion plaques as measured by the progressive loss of vinculin-containing focal adhesions (Figures 3E– 3H), indicative of the absence of integrin-mediated cell adhesion and spreading. As the concentration of kistrin was increased, cellular morphology progressed from spread (Figure 3E) to partially spread cells showing membrane ruffling, lamellipodia, or filopodia (Figures 3F–3H), sug-
Figure 2. Endocytosed FBG colocalizes with ␣v3 in A549 cells. (A) Confocal microscopy was used to analyze the intracellular distribution of endocytosed FBG–Oregon Green (a, c, e, and g) with integrin receptor ␣v3 (b, d, f, and h). The color scale was arbitrarily assigned to denote the relative intensity of staining, which ranges from low fluorescence (purple) to intense fluorescence (red). FBG– Oregon Green coating the glass coverslip appears as moderately intense blue to low purple fluorescence outside the cell (a and c, respectively). Optical sections (1 m thick) along the z-axis started at the basolateral face of the cell membrane; optical sections of 2, 4, 6, and 7 m are shown. Bar in panel c represents 10 m. Arrows indicate colocalization of FBG with ␣v3. (B) To determine the relative fluorescence intensity of both ␣v3 and endocytosed FBG, a line of query was drawn through the cytoplasmic region of the 1-mthick section located 6 m above the basal surface of the cell ( panels e and f, dotted lines) and a plot of the relative fluorescence intensity corresponding to the line distance (in micrometers) in the direction of the arrow was generated. Peak fluorescence of FBG corresponds to peaks in fluorescence of ␣v3.
16
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001
Figure 3. Endocytosis of immobilized FBG by alveolar epithelial cells requires RGDdependent binding to ␣v3. A549 cells were plated on FBG–Oregon Green–coated glass coverslips after pretreatment with no (A and E) or 0.0625 (B and F), 0.25 (C and G), or 1 M (D and H) kistrin for 15 min at 4⬚C, and then incubated an additional 18 h at 37⬚C. Endocytosed FBG–Oregon Green is visualized in A–D with a corresponding view of the same field stained with anti-vinculin followed by secondary antibody conjugated to rhodamine (E–H, respectively). Long arrow indicates vinculin containing focal adhesions (E); short arrows represent membrane ruffles (F); arrowheads denote filopodia (G). Note that in the presence of kistrin (B–D), the extracellular FBG–Oregon Green remains as a uniform coating on the surface of the coverslip. Bar in H is 25 m.
gesting that disruption of integrin-mediated signaling affects actin cytoskeleton–induced locomotion and pericellular proteolysis. To show that echistatin specifically inhibited endocytosis of FBG, A549 cells were plated on iodinated FBG for 18-h during which the cells were treated with increasing concentrations of echistatin. The intracellular FBG was purified from acid-stripped cells and analyzed by SDS-PAGE and autoradiography (Figure 4). The data indicate that the amount of iodinated FBG endocytosed by A549 cells was specifically inhibited by the disintegrin in a concentration-dependent manner. In addition, the results indicate that although less FBG was internalized, the overall pattern of intracellular FBG degradation was the same as observed in the absence of echistatin treatment. Together, these data prove that endocytosis of
Figure 4. Dose-dependent inhibition of FBG endocytosis by echistatin. A549 cells were plated on the surface, immobilized, iodinated with FBG, and treated with increasing micromolar concentrations of echistatin as indicated above each lane; all cells were treated with 50 M chloroquine to prevent complete degradation of the internalized FBG. The cells were lifted with EDTA, then acid-stripped, to remove cell-surface–bound iodinated FBG. Intracellular FBG was immunoprecipitated from cell lysates using a monospecific antihuman FBG PoAb and protein A–sepharose, then resolved by nonreducing SDS-PAGE. The gel was dried and exposed to Xomat film to obtain the resulting autoradiograph shown. Intact FBG is indicated by the arrow at the expected molecular weight (Mr) of 340 kD. The Mr markers, from top to bottom, are 440, 220, 96, 66, and 46 kD as indicated by the lines in the right margin.
immobilized FBG by A549 cells is regulated by RGD-dependent binding to ␣v3. FBG Is Processed through the Lysosomal Degradative Pathway When cells were pretreated with chloroquine, an inhibitor of lysosomal protein degradation and recycling, bright collections of fluorescently labeled ␣v3 (Figure 5A) and FBG (Figure 5B) were found colocalized in vesicular structures (Figure 5C). The characteristic dilatation of the vesicles by chloroquine treatment is indicative of lysosomes or late endosomes (22). We determined whether A549 cells recycled, i.e., exocytosed, intact FBG after uptake and processing. After endocytosis of iodinated FBG, cells were collected, acid-stripped to remove extracellular bound proteins, and then incubated in buffer to allow release of the internalized iodinated material. The 125I-FBG/FBG fragments released from the cells into the buffer were recovered and subjected to TCA precipitation. Greater than 95% of the radiolabeled material released was TCA-soluble, indicating that internalized FBG was degraded and not recycled as intact FBG. Together, these data demonstrate that surface-immobilized FBG is endocytosed into vesicular structures of lung epithelial cells via the integrin receptor ␣v3, and suggest that intracellular FBG is processed through the lysosomal degradative pathway after internalization. Endocytosis of FBG Is HSPG-Independent Previous studies have demonstrated that endocytosis of VN is dependent on both integrin ␣v5 and cell-surface HSPG (9, 10). Additional studies have shown that endocytosis of thrombospondin also depends on cell-surface HSPG (23). Because FBG contains a heparin-binding domain (11), the roles of heparin and HSPG in endocytosis of FBG–Oregon Green were examined. A549 cells were plated on FBG– Oregon Green for 18 h, then fixed and stained with antiHSPG MoAb. HSPG were found intracellularly and extracellularly both cell-surface associated and attached to the FBG-coated surface (Figure 6B). To determine whether endocytosis of FBG required an HSPG-dependent inter-
Odrljin, Haidaris, Lerner, et al.: Endocytosis of Immobilized Fibrinogen by Pneumocytes
action, the A549 cells were incubated on FBG–Oregon Green for 18 h in the presence of increasing concentrations of soluble heparin; heparin did not inhibit endocytosis of FBG (not shown). Xyloside inhibits new synthesis of proteoglycans by inhibiting addition of glycosaminoglycans to the core protein. To determine the extent to which xyloside treatment inhibited sulfation of newly synthesized proteoglycans, the A549 cells were treated with 35 SO4 and the amount of newly labeled proteoglycan was determined by cetylpyridinium precipitation. Xyloside treatment inhibited the synthesis of newly sulfated proteoglycans by greater than 75%. Although treatment of the A549 cells with xyloside reduced the amount of extracellular HSPG staining (Figure 6D, asterisk), the endocytosis of FBG–Oregon Green was not inhibited (Figure 6C). Similarly, when the cells were treated with hepariti-
Figure 5. FBG accumulates with ␣v3 in vesicular structures in chloroquine-treated lung epithelial cells. Chloroquine-treated A549 cells were plated on FBG–Oregon Green and then stained with PoAb against ␣v3. Integrin receptors ␣v3 were visualized under red fluorescence (A) and vesicles containing FBG–Oregon Green were visualized under green fluorescence (B). Using the dual band filter for both red (rhodamine) and green (Oregon Green) fluorescence, FBG colocalizing with ␣v3 was visualized as yellow fluorescence (C). Bar: 25 m. Arrowheads denote representative vesicles in which both FBG and ␣v3 were found.
17
nase (Figure 6F) to cleave HSPG on the cell surface and in the ECM, the endocytosis of FBG–Oregon Green was not inhibited (Figure 6E). To determine quantitatively whether the amount of FBG endocytosed was affected, A549 cells were plated on iodinated FBG, then treated either with soluble heparin or with heparitinase and xyloside in the presence of 50 M chloroquine to prevent complete intracellular degradation of endocytosed FBG. After 18 h of incubation, the cells were lifted and acid-stripped to remove cell surface–bound proteins. The relative amount of intracellular FBG was determined by TCA precipitation and ␥-counting. Suprisingly, the amount of endocytosed FBG was enhanced slightly, but not significantly, over control (9 to 24%) by both heparin and heparitinase treatment. These results suggest that the mechanism promoting endocytosis of surface-immobilized FBG is independent of HSPGFBG binding interactions. Further, heparitinase treatment appeared to cause A549 cell retraction (Figure 6, double arrow), and reduced overall clearance of surface-immobilized FBG was observed.
Figure 6. Endocytosis of FBG by A549 cells does not require proteoglycan-dependent binding interactions. In control cells (A and B), FBG–Oregon Green in endocytic vesicles (A) did not colocalize with cell-surface HSPG (B). Endocytosis of FBG–Oregon Green occurred in xyloside-treated cells (C) in which new synthesis of glycosaminoglycan-modified HSPG was inhibited (D). Notable clearing of the surface-bound FBG–Oregon Green occurred in xyloside-treated A549 cells (C, asterisk) compared with control cells (A, asterisk). Cells treated with heparitinase also endocytosed FBG–Oregon Green (E and F). The double arrows denote the zones of clearing of FBG–Oregon Green from the extracellular surface, indicating cell retraction or movement. Bar in F is 10 m.
18
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001
FBG Is Internalized via a Nonclathrin Pathway To determine which endocytic pathway was used for internalization of FBG, FITC-labeled transferrin was used as a marker of the clathrin pathway (18), and rhodamine-labeled dextran-70 was used as a marker of fluid phase uptake (19). Dual immunofluorescence was used to determine whether FBG–Oregon Green colocalized with dextran-70, and indirect immunofluorescent staining was used to determine colocalization of unlabeled FBG with FITC-transferrin. The results indicate that FBG–Oregon Green partially colocalized with endocytosed dextran-70 (Figures 7A and 7B, respectively); whereas there was no colocalization of FBG (not shown) with transferrin-FITC (Figure 7E) in endocytic vesicles. These findings strongly suggest that the endocytic pathway used by A549 cells for uptake of FBG was not by clathrin-coated pits. Cytochalasin D inhibits actin polymerization and specifically blocks clathrin-independent pathways of endocytosis (18). Therefore, cytochalasin D treatment was performed to determine whether inhibition of actin polymerization would also inhibit the endocytosis of FBG. The results show that cytochalasin D inhibited the endocytosis of both FBG–Oregon Green (Figure 7C) and dextran-70 (Figure 7D), but not endocytosis of transferrin (Figure 7F). These findings further support the concept that matrix-bound FBG was endocytosed via a clathrin-independent pathway. Integrin engagement activates signaling mechanisms to alter the actin cytoskeleton in support of cell adhesion and spreading. Thus, cytochalasin D disruption of
the actin cytoskeleton is also consistent with the requirement of ␣v3-dependent activation and signaling to support endocytosis of FBG by A549 cells. Endocytosis of FBG Is Independent of Plasmin-Mediated Proteolysis We used immunofluorescent staining and autoradiography to analyze the role of plasmin enzymatic degradation in the process of A549 cell endocytosis of FBG. In the absence of the plasmin inhibitor Trasylol, cells plated on FBG– Oregon Green showed the typical vesicular structures of endocytosed FBG, as well as a significant clearing of the FBG–Oregon Green surrounding the cells (Figure 8A). In contrast, cells treated with Trasylol showed little clearing of FBG–Oregon Green around the cells and no generalized clearing of surface-bound FBG–Oregon Green; however, the vesicular immunofluorescent structures were still present (Figure 8B). Trasylol inhibited degradation of extracellular immobilized 125I-FBG by 92.6% ⫾ 1.5% standard error of the mean (n ⫽ 3) into the characteristic fragments X, Y, D, and E (Figure 8C). Further, analysis of the intracellular pool of endocytosed FBG was performed by acid-stripping surface-bound proteins followed by immunopurification of the internalized iodinated FBG. The results indicate that the internalized FBG was significantly degraded; however, the electrophoretic pattern of the degraded internalized FBG was distinct from that of FBG degraded by plasmin (Figure 8C). These results suggest that
Figure 7. Endocytosis of FBG–Oregon Green requires intact actin filaments. A549 epithelial cells were left untreated (A, B, and E) or treated with cytochalasin D to inhibit actin polymerization (C, D, and F). A549 cells were plated on FBG–Oregon Green in medium containing dextran-rhodamine (A–D) or nonlabeled FBG in medium containing transferrin-FITC (E and F). FBG–Oregon Green (A, arrows) showed partial colocalization with dextran-rhodamine (B, arrows) in endocytic vesicles. TransferrinFITC in endocytic vesicles (E) did not colocalize with FBG endocytic vesicles (not shown). Endocytosis of FBG–Oregon Green (C) and dextran-rhodamine (D), but not transferrinFITC (F), was inhibited by cytochalasin D, consistent with a nonclathrin endocytic pathway. Bar in C represents 10 m.
Odrljin, Haidaris, Lerner, et al.: Endocytosis of Immobilized Fibrinogen by Pneumocytes
19
Figure 8. Endocytosis of FBG by A549 cells does not require plasmin-mediated proteolysis. In control A549 cells (A), significant clearing of surface immobilized FBG–Oregon Green and FBG-containing endocytic vesicles were observed. However, Trasylol treatment inhibited clearing of the extracellular surface-bound FBG (B) but did not inhibit endocytosis of FBG. Bar in B represents 10 m. Surface-immobilized 125I-FBG was recovered and analyzed by SDS-PAGE and autoradiography (C). In C: Lane 1, 125I-FBG starting material; lane 2,125IFBG recovered from the surface of control cells; lane 3, 125 I-FBG recovered from the surface of Trasylol-treated cells; lane 4, internalized 125I-FBG recovered from acidstripped control cells; lane 5, internalized 125I-FBG recovered from acid-stripped Trasylol-treated cells. The Mr markers are denoted in kilodaltons in the right margin; the characteristic FBG plasmin cleavage fragments X, Y, D, and E are indicated in the left margin.
whereas plasmin activity plays a role in turnover of immobilized FBG, endocytosis and intracellular degradation of FBG by A549 cells occur in the absence of plasmin cleavage.
Discussion Plasma FBG is produced by the liver; however, extrahepatic epithelial cells synthesize and secrete FBG (8, 13, 24– 27). The production of FBG by lung alveolar epithelial cells occurs primarily after induction of an inflammatory response both in vivo (27) and in vitro (25), which is indicative of a localized acute phase response to injury. It is underappreciated that approximately 25% of FBG is found in the interstitial fluid and lymph under basal conditions. However, it is well known that the levels of plasma FBG increase by 2- to 20-fold during a systemic inflammatory response (1). Moreover, the levels of extracellular FBG increase during injury (28) and both fibrin and FBG are thought to promote inflammatory responses that lead to pulmonary fibrosis (29). The conformational state of FBG is altered when it is preadsorbed to a solid surface; cryptic epitopes, including RGD domains not accessible in soluble FBG are thus exposed (30). Further, when soluble FBG is added to a confluent monolayer of fibroblasts or A549 cells, the FBG is assembled in a cell-dependent process into mature ECM fibrils on which cryptic epitopes are exposed (8). Thus, alterations in FBG conformation, such as exposure of the FBG A␣ chain RGD integrin binding domains by adsorption of FBG to a solid surface (30), will likely affect the interaction of cells with an insoluble FBG substratum. Acute lung injury leads to type I alveolar epithelial cell death, denuding of the alveolar basement membrane, and formation of an alveolar provisional matrix including FN, FBG, fibrin, and type I collagen. To restore normal lung architecture, surviving type II alveolar epithelial cells must repopulate regions of denuded alveoli (6, 7). The provisional matrix thus provides a scaffold for alveolar repair and terminal differentiation of the cells. During the resto-
ration of homeostasis after tissue injury, the provisional matrix is replaced by granulation tissue (2) that resolves into an established matrix by the proteolytic removal of the provisional matrix constituents. In addition to tissue remodeling by proteolytic degradation of provisional matrix molecules, either endocytosis coupled with intracellular degradation, or endocytosis, intracellular transport, and recycling of the extracellular molecules is known to occur. Although the clearance of fibrin by plasmin-mediated fibrinolysis is well described (1), little is known regarding the turnover of insoluble FBG in provisional matrix during wound healing. To better understand the mechanisms operative in insoluble FBG turnover, the endocytosis of surface-immobilized FBG by A549 alveolar epithelial cells was examined. In the present study, the A549 epithelial cell line, derived from a human adenocarcinoma of type II alveolar pneumocytes, adhered and spread on surface-bound FBG and engaged the integrin receptor ␣v3, but not ␣51, in focal adhesion contacts, implying RGD-mediated ligand– cell binding. To confirm that this binding event was RGDdependent, we employed the integrin inhibitory proteins, disintegrins, that contain the RGD sequence. A variety of disintegrins have been identified that have different degrees of affinity for RGD-dependent binding to integrin subclasses. Echistatin and kistrin both inhibit FBG binding to ␣v3 with high affinity, whereas echistatin causes inhibition of RGD-binding to ␣51 with lower affinity (16). In contrast, kistrin blocks the binding of RGD-ligands to ␣v5, but its affinity for ␣51 is too low to inhibit ligand binding (17). Both ␣v3 and ␣51 integrin receptors bind to FBG (Reference 1 and references therein); therefore, we used the differential binding affinities of echistatin (␣v3 and ␣51) and kistrin (␣v3 but not ␣51) to determine which integrin subclass mediated endocytosis of immobilized FBG by A549 cells. Integrin-dependent binding of A549 cells to immobilized FBG was confirmed by inhibition of RGD-dependent focal contact formation with both echistatin and kistrin. Both kistrin and echistatin were
20
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001
equally effective in inhibiting endocytosis of surface-immobilized FBG. These data, together with the immunofluorescent staining data showing colocalization of FBG with ␣v3 but not ␣51, prove that such endocytosis occurs through FBG engagement of ␣v3 and not ␣51. The endocytosis of plasma FBG, not endogenous biosynthesis, is now considered the origin of the FBG from platelets and megakaryocytes. Integrin ␣IIb3-mediated endocytosis of soluble FBG by platelets and megakaryocytes is RGD-dependent (17, 31); however, this endocytosed FBG is directed to the ␣-granule storage compartment rather than a degradative pathway. Recycling of soluble FBG endocytosed by platelets and megakaryocytes occurs; FBG stored in ␣ granules is released into the circulation as intact molecules to support ␣IIb3-mediated platelet aggregation at sites of vessel injury. In this study, soluble FBG was not endocytosed by A549 cells. Instead, surface-immobilized FBG endocytosed by A549 cells was directed toward the lysosomal degradative pathway. Further, we confirmed that internalized FBG was degraded and not recycled, as is known to occur in platelets and megakaryocytes. Together, these data indicate that the ␣v3-dependent endocytosis and subsequent degradation of surface-immobilized FBG by alveolar epithelial cells differs significantly from the ␣IIb3-mediated endocytosis and storage of soluble, intact FBG by platelets and megakaryocytes. Integrins of the 1 and 3 subclasses are important in mediating cellular responses to the adhesive glycoproteins of the provisional matrix. The 1 and 3 integrin receptors support alveolar type II cell adhesion and migration (6, 7). In the present study, ␣v3 is shown to be the predominant integrin mediating endocytosis of surface-bound FBG. Kim and colleagues have shown that primary cultures of rat alveolar type II cells adhere to surface-immobilized FN, VN, and FBG via ␣v3 (6); the engagement of ␣v3 and not ␣51 by adhesion and spreading of A549 cells on FBG as shown in this study is consistent with the results of Kim and associates (6). Further, the data in our study suggest that FBG provides an adhesive subtratum that supports cell migration and pericellular proteolysis in addition to endocytosis. In contrast, in the presence of both kistrin and echistatin the integrin-dependent cell spreading on FBG RGD sites was significantly reduced, as noted by the altered cell morphology, namely membrane ruffling and lamellipodia formation. Integrin-mediated signaling and cell spreading and migration involve changes in the actin cytoskeleton. Stable cell interactions are needed to maintain the structural integrity of tissues and active adhesion mechanisms are required to regulate the processes of cell motility and cell migration. In particular, integrin-mediated cell spreading and motility on the ECM is mediated by changes in the actin cytoskeleton (32). Indeed, treatment of the cells with cytochalasin D to inhibit actin polymerization not only reduced the endocytosis of FBG but also reduced the overall clearance of the FBG substrate. VN is cleared from the ECM, at least in part, by receptor ␣v5–mediated endocytosis followed by lysosomal degradation, suggesting that cells can regulate the levels of VN present in the matrix (9, 10). A recent study implicated the heparin-binding domain of VN in its binding to ECM and
demonstrated that its subsequent degradation by fibroblasts is dependent on HSPG (9). Our previous work has implicated the heparin-binding domain of FBG in the assembly of FBG into a detergent-insoluble fraction of the ECM (8). Thus, we wanted to determine whether the endocytosis of surface-immobilized FBG was dependent on proteoglycan binding interactions. A549 alveolar epithelial cells endocytosed FBG when treated with either xyloside, which prevents glycosaminoglycan side-chain modification of the proteoglycan core proteins, or heparitinase, which specifically digests HSPG on cell surfaces and in the ECM. In addition, soluble heparin was unable to inhibit the endocytosis of FBG by A549 cells. Together, these results indicated that endocytosis of surface-immobilized FBG by alveolar epithelium does not require proteoglycans, in particular, HSPG. Different mechanisms have been described for endocytosis of extracellular materials by mammalian cells. One of the best-characterized endocytic mechanisms occurs via clathrin-coated pits. Endocytosis of transferrin-FITC is the classical marker of the clathrin-dependent pathway, whereas dextran-70 is used as a marker of fluid phase uptake, as well as nonclathrin endocytic pathways involving the caveolar route. The nonclathrin pathways are also dependent on actin polymerization for particle uptake, or for the internalization of extracellular fluid and receptor-bound ligands (33). Cytochalasin D has been shown to block the caveolar endocytic pathway without affecting the clathrin pathway (18). In this study, we demonstrate that endocytosis of FBG by A549 alveolar epithelial cells occurred via a nonclathrin pathway as noted by the absence of FBG colocalization with transferrin and by the inhibition of FBG endocytosis by blocking actin polymerization. The clearance of fibrin involves extracellular proteolysis by fibrinolytic enzymes, primarily plasmin, a fibrinolytic proteinase generated from ubiquitously produced plasminogen by cell-derived urokinase or tissue plasminogen activators. Urokinase receptors are distributed on surfaces of many cell types, including A549 cells (34), where they focus plasmin-dependent proteolysis important in cell migration and tissue remodeling of the pericellular space. To determine whether plasmin-mediated proteolysis was required to promote endocytosis of FBG by A549 cells, Trasylol was used to inhibit pericellular proteolysis of the surface immobilized FBG. Endocytosis of FBG was not inhibited in the presence of Trasylol, although plasmin degradation of immobilized FBG was inhibited by ⬎ 92%. A limited amount of proteolysis of the FBG–Oregon Green was observed, suggesting that either non–plasmin mediated proteolysis was occurring, or that adherence of A549 cells to FBG provided a compartment beneath the cells that was inaccessible to Trasylol (35). Notably, autoradiography of internalized 125I-FBG showed that proteolysis of endocytosed FBG by A549 cells produced a unique degradation pattern distinct from the typical D and E fragments generated by plasmin cleavage of FBG. In summary, the influx of FBG and generation of fibrin in the alveolar spaces during lung injury contributes to formation of the provisional matrix that provides a structural scaffold decorated with cell adhesion domains not normally seen in the alveolar microenvironment. By provid-
Odrljin, Haidaris, Lerner, et al.: Endocytosis of Immobilized Fibrinogen by Pneumocytes
ing new adhesive sites, both FBG and fibrin, in concert with the in situ ECM and basement membrane proteins, would facilitate the repopulation of the alveolar epithelium by progenitor type II cells. The data in this report indicate that the presence of metobolically active cells, integrin receptors, and the conformational state of the molecule impart specificity in cellular processing of FBG. Thus, appropriate wound repair is directed by the spatial and temporal interplay of cells, cytokines, growth factors, and matrix constituents in the microenvironment, i.e., “wound repair in context.” Together, these observations spark new interest in understanding further the balance between deposition and turnover of matrix FBG during lung inflammation and alveolar wound repair. Acknowledgments: This work was supported by research grants HL30616, HL50615, and HL49610 from the National Institutes of Health, Bethesda, MD. The authors thank Sarah O. Lawrence for expert technical assistance.
References 1. Bini, A., P. J. Simpson-Haidaris, and B. J. Kudryk. 2000. Fibrin/fibrinogen. In Encylopedic Reference of Vascular Biology & Pathology. A. Bikfalvi, editor. Springer-Verlag, Berlin. 372. 2. Clark, R. A. F. 1996. Wound repair. In The Molecular and Cellular Biology of Wound Repair. R. A. F. Clark, editor. Plenum Press, New York. 617. 3. Idell, S., J. G. Garcia, K. Gonzalez, J. McLarty, and D. S. Fair. 1989. Fibrinopeptide A reactive peptides and procoagulant activity in bronchoalveolar lavage: relationship to rheumatoid interstitial lung disease. J. Rheumatol. 16:592–598. 4. Sahni, A., T. Odrljin, and C. W. Francis. 1998. Binding of basic fibroblast growth factor to fibrinogen and fibrin. J. Biol. Chem. 273:7554–7559. 5. Francis, C. W., L. A. Bunce, and L. A. Sporn. 1993. Endothelial cell responses to fibrin mediated by FPB cleavage and the amino terminus of the  chain. Blood Cells 19:291–307. 6. Kim, H. J., D. H. Ingbar, and C. A. Henke. 1996. Integrin mediation of type II cell adherence to provisional matrix proteins. Am. J. Physiol. 271:L277–L286. 7. Kim, H. J., C. A. Henke, S. K. Savik, and D. H. Ingbar. 1997. Integrin mediation of alveolar epithelial cell migration on fibronectin and type I collagen. Am. J. Physiol. 273:L134–L141. 8. Guadiz, G., L. A. Sporn, and P. J. Simpson-Haidaris. 1997. Thrombin cleavage-independent deposition of fibrinogen in extracellular matrices. Blood 90:2644–2653. 9. Wilkins-Port, C. E., and P. J. McKeown-Longo. 1996. Heparan sulfate proteoglycans function in the binding and degradation of vitronectin by fibroblast monolayers. Biochem. Cell Biol. 74:887–897. 10. Panetti, T. S., and P. J. McKeown-Longo. 1993. The ␣v5 integrin receptor regulates receptor-mediated endocytosis of vitronectin. J. Biol. Chem. 268: 11492–11495. 11. Odrljin, T. M., J. R. Shainoff, S. O. Lawrence, and P. J. Simpson-Haidaris. 1996. Thrombin cleavage enhances exposure of a heparin binding domain in the N-terminus of the fibrin  chain. Blood 88:2050–2061. 12. Lawrence, S. O., T. W. Wright, C. W. Francis, P. J. Fay, and P. J. Haidaris. 1993. Purification and functional characterization of homodimeric ␥B-␥B fibrinogen from rat plasma. Blood 82:2406–2413. 13. Guadiz, G., L. A. Sporn, R. A. Goss, S. O. Lawrence, V. J. Marder, and P. J. Simpson-Haidaris. 1997. Polarized secretion of fibrinogen by lung epithelial cells. Am. J. Respir. Cell Mol. Biol. 17:60–69. 14. Guadiz, G., C. G. Haidaris, G. N. Maine, and P. J. Simpson-Haidaris. 1998. The carboxyl terminus of Pneumocystis carinii glycoprotein A encodes a functional glycosylphosphatidylinositol signal sequence. J. Biol. Chem. 273: 26202–26209.
21
15. Horn, M. C., M. Breton, E. Deudon, E. Berrou, and J. Picard. 1983. The structural characterization of proteoglycans of cultured aortic smooth muscle cells and arterial wall of the pig. Biochim. Biophys. Acta 755:95–105. 16. Pfaff, M., M. A. McLane, L. Beviglia, S. Niewiarowski, and R. Timpl. 1994. Comparison of disintegrins with limited variation in the RGD loop in their binding to purified integrins ␣IIb3, ␣v3 and ␣51 and in cell adhesion inhibition. Cell Adhes. Commun. 2:491–501. 17. Handagama, P., D. F. Bainton, Y. Jacques, M. T. Conn, R. A. Lazarus, and M. A. Shuman. 1993. Kistrin, an integrin antagonist, blocks endocytosis of fibrinogen into guinea pig megakaryocyte and platelet ␣-granules. J. Clin. Invest. 91:193–200. 18. Hemmaplardh, D., S. G. Kailis, and E. H. Morgan. 1974. The effects of inhibitors of microtubule and microfilament function on transferrin and iron uptake by rabbit reticulocytes and bone marrow. Br. J. Haematol. 28:53–65. 19. von Bonsdorff, C. H., S. D. Fuller, and K. Simons. 1985. Apical and basolateral endocytosis in Madin-Darby canine kidney (MDCK) cells grown on nitrocellulose filters. EMBO J. 4:2781–2792. 20. Suehiro, K., J. W. Smith, and E. F. Plow. 1996. The ligand recognition specificity of 3 integrins. J. Biol. Chem. 271:10365–10371. 21. Pfaff, M., T. Sasaki, K. Tangemann, M. L. Chu, and R. Timpl. 1995. Integrin-binding and cell-adhesion studies of fibulins reveal a particular affinity for ␣IIb3. Exp. Cell Res. 219:87–92. 22. Tooze, J., M. Hollinshead, T. Ludwig, K. Howell, B. Hoflack, and H. Kern. 1990. In exocrine pancreas, the basolateral endocytic pathway converges with the autophagic pathway immediately after the early endosome. J. Cell Biol. 111:329–345. 23. Chen, H., J. Sottile, D. K. Strickland, and D. F. Mosher. 1996. Binding and degradation of thrombospondin-1 mediated through heparan sulphate proteoglycans and low-density-lipoprotein receptor-related protein: localization of the functional activity to the trimeric N-terminal heparin-binding region of thrombospondin-1. Biochem. J. 318:959–963. 24. Nguyen, M. D., and P. J. Simpson-Haidaris. 2000. Cell type-specific regulation of fibrinogen expression in lung epithelial cells by dexamethasone and IL-1. Am. J. Respir. Cell Mol. Biol. 22:209–217. 25. Simpson-Haidaris, P. J. 1997. Induction of fibrinogen biosynthesis and secretion from cultured pulmonary epithelial cells. Blood 89:873–882. 26. Molmenti, E. P., T. Ziambaras, and D. H. Perlmutter. 1993. Evidence for an acute phase response in human intestinal epithelial cells. J. Biol. Chem. 268:14116–14124. 27. Simpson-Haidaris, P. J., M. A. Courtney, T. W. Wright, R. Goss, A. Harmsen, and F. Gigliotti. 1998. Induction of fibrinogen expression in the lung epithelium during Pneumocystis carinii pneumonia. Infect. Immun. 66: 4431–4439. 28. Kreuzer, J., S. Denger, A. Schmidts, L. Jahn, M. Merten, and E. von Hodenberg. 1996. Fibrinogen promotes monocyte adhesion via a protein kinase C dependent mechanism. J. Mol. Med. 74:161–165. 29. Perez, R. L., J. D. Ritzenthaler, and J. Roman. 1999. Transcriptional regulation of the IL-1 promoter via fibrinogen engagement of the CD18 integrin receptor. Am. J. Respir. Cell Mol. Biol. 20:1059–1066. 30. Zamarron, C., M. H. Ginsberg, and E. F. Plow. 1990. Monoclonal antibodies specific for a conformationally altered state of fibrinogen. Thromb. Haemost. 64:41–46. 31. Handagama, P., R. M. Scarborough, M. A. Shuman, and D. F. Bainton. 1993. Endocytosis of fibrinogen into megakaryocyte and platelet ␣-granules is mediated by ␣IIb3 (glycoprotein IIb-IIIa). Blood 82:135–138. 32. Gumbiner, B. M. 1996. Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell 84:345–357. 33. Mellman, I. 1996. Endocytosis and molecular sorting. Annu. Rev. Cell Dev. Biol. 12:575–625. 34. Hasegawa, T., L. Sorensen, M. Dohi, N. V. Rao, J. R. Hoidal, and B. C. Marshall. 1997. Induction of urokinase-type plasminogen activator receptor by IL-1. Am. J. Respir. Cell Mol. Biol. 16:683–692. 35. Loike, J. D., R. Silverstein, S. D. Wright, J. I. Weitz, A. J. Huang, and S. C. Silverstein. 1992. The role of protected extracellular compartments in interactions between leukocytes, and platelets, and fibrin/fibrinogen matrices. Ann. NY Acad. Sci. 667:163–172.
