INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

The vascular response to injury requires a coordinated interaction of the hemostatic and inflammatory systems and is regulated by cytokines and growth factors that act locally to regulate cellular proliferation and tissue repair. The hemostatic response results in platelet accumulation at the site of injury, and exposure of blood to tissue factor also leads to the formation of thrombin. Thrombin then cleaves fibrinopeptides from fibrinogen converting it to fibrin, which helps prevent blood loss and also serves as a temporary matrix to support tissue healing and remodeling. The role of fibrin in the cellular response is not passive as a structural matrix only, but rather it plays an active role through specific receptor-mediated interactions with cells of the blood and vessel wall. These result in fibrin-specific responses of endothelial cells including adhesion and spreading (1), proliferation (2), protein synthesis (3) and secretion (4), and angiogenesis (5).

Cytokines and growth factors are produced in response to injury and also act locally to modulate cell responses to vascular damage. Important among these are members of the fibroblast growth factor family, which includes 13 members exerting a variety of effects on many cells and organ systems (6). In particular, bFGF¹ increases endothelial cell migration and proliferation and also stimulates angiogenesis in vitro and in vivo (6, 7). bFGF also regulates the expression of proteolytic mediators of angiogenesis including urokinase-type plasminogen activator and collagenase (8) and urokinase-type plasminogen activator receptor (9). The role of bFGF in vessel injury and repair is further supported by evidence that bFGF is released from vessel wall cells after injury (10) and that bFGF mRNA is up-regulated in atherosclerotic arteries (11) and following vessel injury (12).

The need for fibrin to support endothelial cell spreading, migration, angiogenesis, and the potent stimulation of the same responses by bFGF suggests that these processes may be interrelated. This concept is supported by evidence that fibrin clots are a good matrix to support bFGF-stimulated angiogenesis in vitro (13, 14). Little information is available, however, regarding specific interactions of bFGF with fibrin. We have, therefore, investigated the association of bFGF with fibrinogen and fibrin, and the results demonstrate high affinity specific and saturable binding.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Protein Preparation-- Plasminogen-free human fibrinogen was purchased from Calbiochem, and fibronectin in the preparation was depleted by chromatography on gelatin-Sepharose (15) (Amersham Pharmacia Biotech). Fibrinogen eluting from the gelatin-Sepharose column was further depleted of fibronectin by immunoaffinity chromatography as described elsewhere (1), and the final preparation contained less than 0.02 ng/ml of fibronectin as determined by enzyme-linked immunosorbent assay (American Diagnostica, Greenwich, CT) at a fibrinogen concentration of 1 mg/ml. Radioiodination of fibrinogen to a specific activity of 1.8 × 10⁸ cpm/mg was performed using the iodogen method (16), and unbound ¹²⁵I was removed following chromatography on Sephadex G-10 (Amersham Pharmacia Biotech). Human thrombin (3,250 NIH units/mg) was obtained from Calbiochem, and human recombinant bFGF, anti-human bFGF polyclonal antibody, and an enzyme-linked immunosorbent assay for bFGF were purchased from R & D Systems (Minneapolis, MN). ¹²⁵I-bFGF was purchased from NEN Life Science Products at a specific activity of 6 µCi/µg. Purified IgG of monoclonal antibody J88B reactive with a site within the sequence Arg⁶³-Met⁷⁸ of the human fibrinogen  chain was kindly provided by Dr. P. J. Simpson-Haidaris, Rochester, NY (17).

Binding of Fibrinogen to Immobilized bFGF-- Purified polyclonal anti-bFGF antibody (200 µg/ml) was incubated with 1 ml of Affi-gel 15 (Bio-Rad), which consists of a derivatized cross-linked agarose gel bead support with active N-hydroxy succinimide esters in 0.1 M sodium phosphate buffer, pH 7.4, containing 0.5 M NaCl and gently mixed at 25 °C for 2 h, and over 97% of antibody was bound to the beads. Residual active ester sites were then blocked by the addition of 1 M ethanolamine, pH 8.0, and the suspension was washed several times with 0.1 M sodium phosphate buffer, pH 7.4, containing 0.5 M NaCl. bFGF (50 µg/ml) was then added to this suspension and gently mixed at 25 °C for 1 h, following which the unbound bFGF was removed by washing with 0.1 M sodium phosphate buffer, pH 7.4, containing 0.25 M NaCl. The amount of bFGF immobilized on the beads was 24.3 µg/ml as determined by enzyme-linked immunosorbent assay. For binding studies, ¹²⁵I-fibrinogen at concentrations from 0.15 to 300 nM was incubated at 37 °C with a 0.02 ml suspension of immobilized bFGF in a final volume of 0.1 ml. Nonspecific binding was determined in parallel experiments using a 20-fold molar excess of unlabeled fibrinogen. Preliminary experiments demonstrated maximum specific binding after a 30 min incubation in 0.1 M sodium phosphate buffer, pH 7.4, containing 0.25 M NaCl, and these conditions were used for all subsequent experiments. Following incubation, the beads were separated by centrifugation at 3,000 × g for 10 min after which the supernatant was removed, and the beads were then washed rapidly twice with 0.1 M sodium phosphate buffer, pH 7.4, containing 0.5 M NaCl at 4 °C to minimize nonspecific association. The amount of bound fibrinogen was calculated from the radioactivity associated with the beads.

Binding of ¹²⁵I-bFGF to Fibrinogen and Fibrin Monomer-- A similar approach was used with incubation of ¹²⁵I-bFGF with fibrinogen or fibrin monomer immobilized on Sepharose beads. Affi-gel 15 beads were first incubated with purified monoclonal antibody J88B (1 mg/ml) in 0.2 M sodium bicarbonate buffer, pH 8.3, and gently mixed at 25 °C for 2 h. Residual sites were blocked by incubation in 1 M ethanolamine, pH 8.0, and the suspension was washed several times with 0.2 M sodium bicarbonate buffer, pH 8.3, containing 0.5 M NaCl. Gel containing bound antibody was incubated with fibrinogen in sodium phosphate buffer, pH 7.4, containing 0.25 M NaCl and then incubated at 25 °C with gentle mixing for 1 h. After this, the beads were washed with 0.1 M sodium phosphate buffer, pH 7.4, containing 0.25 M NaCl to remove unbound fibrinogen. This was continued until no further fibrinogen was removed as determined by monitoring the optical density at 280 nM. To convert bound fibrinogen to fibrin monomer, beads were incubated with 0.5 units/ml of thrombin at 37 °C for 90 min. Characterization of binding of ¹²⁵I-bFGF to fibrinogen and fibrin monomer was performed in the same way as ¹²⁵I-fibrinogen binding to immobilized bFGF (see above). ¹²⁵I-bFGF at concentrations from 0.05 to 100 nM was incubated with 0.02 ml suspension of beads containing 0.1 µg of fibrinogen or fibrin in a final volume of 0.1 ml. Nonspecific binding was determined in parallel experiments using a 100-fold molar excess of unlabeled bFGF. Specificity of the binding of bFGF to fibrinogen was confirmed by competition experiments in which 0.2 nM of ¹²⁵I-bFGF was incubated with 1 µg/ml immobilized fibrinogen in a final volume of 0.1 ml, and binding was competitively inhibited by unlabeled bFGF at concentrations from 0.1 to 100 nM.

Binding of bFGF to Polymerized Fibrin-- ¹²⁵I-bFGF at concentrations of 0.05 to 500 nM was added to 100 µg/ml fibrinogen in 0.1 M Tris buffer containing 0.25 M NaCl. Thrombin was then added to a final concentration of 0.5 units/ml, which resulted in clotting of the solution. Following incubation at 37 °C for 30 min, the clot and supernatant were separated by vacuum filtration using GF/C glass microfiber filters (Sigma) previously soaked overnight in a solution of 0.5% polyvinylpyrrolidone and 0.1% Tween 20 to reduce nonspecific binding. The clot on the filter was washed quickly with cold 0.1 M Tris buffer containing 0.25 M NaCl, and the associated radiolabel was measured. Nonspecific binding was determined by parallel experiments incorporating a 100-fold molar excess of unlabeled bFGF.

Data Analysis-- Unless indicated otherwise data is expressed as mean ± S.D. Scatchard analysis of the data was performed using the Ligand program (18) from Biosoft (Ferguson, MO).

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Binding of fibrinogen to immobilized bFGF was saturable and specific with nonspecific binding representing less than 20% of the total (Fig. 1A). Saturation of specific binding occurred at a fibrinogen concentration of 150 nM, and only an increase in nonspecific binding was observed at higher concentrations. In control experiments there was a maximum of 3% binding of ¹²⁵I-fibrinogen over the same range of concentrations to beads with immobilized anti-bFGF immunoglobulin only or to beads with no protein bound and active sites blocked with ethanolamine. A plot of bound versus bound/free fibrinogen (Fig. 1B) was nonlinear, suggesting the presence of more than one binding site. This was confirmed by Scatchard analysis, which indicated that binding was best described by a two-site model with apparent K_d values of 1.3 and 260 nM (Table I). B_max was 6.3 and 35 nM for the high and low affinity sites, respectively, and the maximum molar binding ratio of bFGF to fibrinogen was 4.0. 

View larger version (15K): [in this window] [in a new window]   Fig. 1.   A, binding of fibrinogen to bFGF. 125I-fibrinogen was incubated with bFGF immobilized on Sepharose beads, and the amount of bound protein was determined as radioactivity associated with the beads following centrifugation and washing. Nonspecific binding (squares) was determined in the same way in the presence of a 20-fold molar excess of unlabeled bFGF. Specific binding (triangles) was calculated by subtracting the nonspecific from the total bound (diamonds). Each point represents the mean ± S.D. of three different experiments. B, Scatchard plot. The best fit of the data was determined by analysis using the Ligand program and is consistent with presence of two binding sites of different affinities.

To further characterize the protein that bound to bFGF, ¹²⁵I-fibrinogen was passed over a column of immobilized bFGF. Following washing, the specifically bound protein was eluted with 2 mg/ml unlabeled fibrinogen (Fig. 2), and approximately 90% of bound label rapidly eluted in two fractions. SDS-polyacrylamide gel electrophoresis of the eluted protein showed bands consistent with the A, B and  chains of fibrinogen (Fig. 2, inset) establishing that the bound protein was fibrinogen and not a minor contaminant. In control experiments, less than 5% of bound radioactivity was eluted from the column with 2.0 mg/ml ovalbumin, demonstrating specificity of the elution.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Elution of bound protein from immobilized bFGF. 125I-Fibrinogen (1.0 mg/ml) was passed through a 1-ml column of Sepharose-immobilized bFGF. Following washing, the column was eluted with 2 mg/ml unlabeled fibrinogen, and fractions of 200 µl were collected. Approximately 90% of the bound radioactivity eluted in fractions 5 and 6 (1.0-1.2 ml elution volume). These fractions were pooled, and an aliquot was electrophoresed on a 7% SDS-polyacrylamide gel electrophoresis gel and used to prepare autoradiograms (inset). The polypeptide chain pattern in the eluted pool showed A, B, and  chains of fibrinogen and was similar to that in the starting material.

The association of bFGF and fibrinogen was also characterized using soluble ¹²⁵I-radiolabeled bFGF and fibrinogen immobilized on Sepharose beads (Fig. 3). With this system, saturable and specific binding was also observed, and nonspecific binding represented 20% or less of the total. Saturation of specific binding was observed at a bFGF concentration of approximately 75 nM. Scatchard analysis (Fig. 3B) indicated the presence of two binding sites of different affinities with apparent K_d values of 0.9 and 70 nM and a maximum molar binding ratio of 2.0 (Table I) as compared with 4.0 with radiolabeled fibrinogen binding to Sepharose-immobilized bFGF (Fig. 1 and Table I). Competitive inhibition of the binding was performed to further characterize the specificity and the degree of nonspecific association. ¹²⁵I-bFGF at a concentration of 0.2 nM was incubated with Sepharose-immobilized fibrinogen and then varying concentrations of unlabeled bFGF was added. The binding of ¹²⁵I-bFGF to fibrinogen was reduced in a dose-dependent manner with increasing concentrations of unlabeled bFGF (Fig. 3C), but 28% of the radiolabel remained at 100 nM bFGF, which we interpreted as nonspecific binding.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   A, binding of bFGF to fibrinogen. 125I-bFGF was incubated with fibrinogen immobilized on Sepharose beads, and the amount of bound protein was determined as radioactivity associated with the beads following centrifugation and washing. Nonspecific binding (squares) was determined in the same way in the presence of a 100-fold molar excess of unlabeled bFGF. Specific binding (triangles) was calculated by subtracting the nonspecific from the total bound (diamonds). Each point represents the mean ± S.D. of three different experiments. B, Scatchard plot. The best fit of the data was determined by analysis with the Ligand program and is most consistent with involvement of two distinct binding sites. C, competitive inhibition of binding. Increasing concentrations of unlabeled bFGF were used to competitively inhibit the binding of 125I-bFGF to fibrinogen. Each point represents mean ± S.D. of three different experiments.

Fibrinogen is converted to fibrin by thrombin, which cleaves fibrinopeptides A and B from the A and B chains respectively, forming fibrin monomer, which can then polymerize to form a branching network of fibers. To characterize the association of bFGF with fibrin in the absence of polymerization, we incubated Sepharose-immobilized fibrinogen with thrombin. The antibody-mediated immobilization of fibrinogen to the Sepharose beads prevents or limits association of the resulting fibrin, forming a surface with immobilized fibrin monomer. Binding of bFGF to fibrin (Fig. 4A) was similar to that seen for fibrinogen using the same system with specific binding approaching saturation between 75 and 100 nM bFGF (Fig. 4A). Nonspecific binding was low at bFGF concentrations below 15 nM and increased at higher concentrations. In contrast to the binding seen with fibrinogen, the bFGF binding curve at concentrations below 1 nM bFGF suggested the presence of a high affinity binding site of low capacity. This was confirmed by Scatchard analysis (Fig. 4B), indicating the presence of two binding sites with apparent K_d values of 0.13 and 83 nM (Table I).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Binding of bFGF to fibrin monomer. Fibrinogen was immobilized on Sepharose beads and then converted to fibrin monomer by incubation with thrombin. 125I-bFGF was incubated with immobilized fibrin, and bound and free ligand were then separated by centrifugation. Nonspecific binding (squares) was measured in the presence of a 100-fold molar excess of unlabeled bFGF, and specific binding (triangles) was determined by subtraction of nonspecific from total binding (diamonds). Each point represents the mean ± S.D. of three different experiments. B, Scatchard plot. The best fit of the data was determined using the Ligand program, and the presence of two distinct binding sites was indicated.

Characterization of binding to polymerized fibrin presents technical and interpretive problems because of transport of bFGF into the gel is limited, and access to potential binding sites within individual fibrin fibers may also be restricted. We chose, therefore, to add ¹²⁵I-bFGF to a solution of fibrinogen, which was then clotted by the addition of thrombin to avoid problems of transport of bFGF into the gel. Total binding was measured with this clotting system in the absence of an unlabeled competitor, whereas nonspecific binding was measured in the presence of 100-fold molar access of unlabeled bFGF (Fig. 5). Nonspecific binding represented up to 40% of the total (Fig. 5A). This was higher than that seen with binding to fibrinogen (Figs. 1 and 3) or to fibrin monomer (Fig. 4), possibly reflecting some entrapment of radiolabel within the fibrin gel. A plot of bound versus bound/free ¹²⁵I-bFGF was nonlinear (Fig. 5B), and Scatchard analysis identified two distinct binding sites with apparent K_d values of 0.8 and 261 nM, similar to those for fibrinogen (Table I). The maximum molar binding ratio of bFGF to fibrin was 2.0. 

View larger version (14K): [in this window] [in a new window]   Fig. 5.   A, binding of 125I-bFGF to polymerized fibrin. 125I-bFGF was added to a solution of 100 µg/ml fibrinogen and then clotted by the addition of 0.5 units/ml of thrombin. Bound and unbound bFGF were then separated by vacuum filtration, and nonspecific binding (squares) was determined in the presence of 100-fold molar excess of bFGF. Specific binding (triangles) was calculated by subtracting nonspecific from total binding (diamonds). Each point represents the mean ± S.D. of three different experiments. B, Scatchard analysis. The best fit of the data was determined using the Ligand program, and the presence of two distinct binding sites was indicated.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

The results presented demonstrate that bFGF binds specifically and saturably to fibrinogen and fibrin. Two different experimental systems were used to characterize the association with either bFGF or fibrinogen immobilized on Sepharose beads. The results were similar, with both systems identifying high affinity binding sites with K_d values of 1.3 and 0.9 nM and lower affinity sites with K_d values of 260 and 70 nM. The association of bFGF with fibrin was also characterized using two systems with either surface-immobilized fibrin or polymerized fibrin. The results of binding to fibrin were similar to those found using fibrinogen with two distinct binding sites. The K_d values for the high and low affinity sites were 0.13 and 0.8 nM and 83 and 261 nM for surface-immobilized and -polymerized fibrin, respectively.

The maximum molar binding ratios for bFGF to fibrinogen or fibrin were between 2.0 and 4.0 with the different systems used. Considering that fibrinogen is a dimerically symmetric molecule (19) and that two binding sites with different K_d values were identified, the ratio of 4 bFGF to 1 fibrinogen would be expected and consistent with the presence of two structurally distinct and independent sites on each half-molecule. The expected ratio of 4 was, however, only found using a system in which fibrinogen bound to immobilized bFGF, whereas the ratio was lower using the other three approaches. One potential explanation for the lower binding ratio is that the access of bFGF to potential binding sites was limited. This would be reasonable with polymerized fibrin, as sites in both the D and E domains are involved in the reciprocal binding required for polymerization (20). This could prevent concurrent binding of bFGF to any sites in close proximity or to those otherwise affected by polymerization. A similar explanation could explain reduced binding of bFGF to Sepharose-immobilized fibrinogen or fibrin if the binding site was either close to that recognized by the antibody J88B or altered by antibody binding. An alternative explanation for the lower than expected binding ratio is that one of the binding sites is present on only a minor variant of fibrinogen. There are several such fibrinogen variants including those due to heterogeneity at the carboxyl terminus of the  chain (21, 22) or to variations in serine phosphorylation. These sites are known to be important in the molecular interactions and function of fibrinogen and fibrin as the  chain site is involved both in binding to platelets (23) and in factor XIII cross-linking (24), and phosphorylation of Ser³ of the  chain affects thrombin action (25). This explanation could account for the ratio of 4 found with the system using bFGF immobilized on Sepharose as only those fibrinogen molecules with the high affinity site would bind. By contrast, the epitope recognized by J88B involves a structurally invariant site (17) so all fibrinogen molecules would be immobilized, whereas only a minority would possess the putative high affinity site. Ongoing studies designed to identify specific sites on fibrinogen and fibrin responsible for bFGF binding will be required to resolve this question.

The significance of bFGF binding to fibrinogen and fibrin must be considered in relation to both the tissue distribution of bFGF and to the availability of other sites for binding within the vasculature. bFGF has a wide tissue distribution, and it is synthesized in culture by fibroblasts, endothelial cells, glial cells, and smooth muscle cells (6). Vlodavsky et al. (26) have shown that endothelial cells synthesize bFGF, which then remains closely associated with the cells and bound to the basement membrane or cell matrix with a K_d value of 610 nM. It is inactive as a complex with heparan sulfate or heparin but is stabilized and protected against proteolytic degradation (27, 28). It can be released in active form by proteolytic degradation with plasmin (29) or heparitinase and by competition with heparin-like molecules (30). bFGF is also present normally in plasma at a concentration up to 10 pg/ml (0.6 pM), and elevated levels up to 6 pM can be found in patients after cardiopulmonary bypass (31) and chronic liver disease (32). The effects of bFGF are mediated through specific receptors, and four distinct genes encoding bFGF cell surface receptor tyrosine kinases have been identified. bFGF binds to FGFR1 and FGFR2 with similar affinities (33, 34), and bFGF binds specifically and with higher affinity to intact baby hamster kidney cells with K_d values of 20 pM and 2 nM (35).

Heparin and heparan sulfate interact with both bFGF and fibrinogen and may play a role in modulating interactions between them. Heparin and heparan sulfate represent low affinity (K_d 470 nM) receptors for bFGF present in abundance on cell surfaces and in the extracellular matrix (36). This binding increases the concentration of bFGF on the cell surface and may thereby promote the interaction of bFGF with high affinity transmembrane-signaling receptors. Heparin is not required for binding of bFGF to specific cell receptors (37), but receptors are activated by receptor dimerization, which is promoted by heparin (38). Also, the affinity of bFGF for FGFR1 is increased approximately 10-fold in the presence of heparan sulfate or heparin (39). Receptor dimerization results in activation of protein tyrosine kinase activity and autophosphorylation and further to activation of the signaling pathway for the initiation of cell proliferation (40). Heparin mediates binding of bFGF to the luminal surface of endothelial cells as indicated by its displacement by incubation of heparin in vitro (31) or by heparin infusion in vivo (31, 41). Heparin also binds to fibrinogen and fibrin, and it may, therefore, alter their interactions with bFGF. Heparin binds to a site within the fibrinogen D domain (42) and with higher affinity (K_d 0.8 µM) to the central E domain (43). Cleavage of fibrinopeptide B during conversion to fibrin exposes a new site at the amino terminus of the  chain including residues 15-42, which represent a higher affinity site (K_d 0.3 µM) that mediates fibrin-endothelial cell interactions (43).

The binding of bFGF to fibrinogen has implications regarding the distribution and actions of bFGF within the vasculature. At normal plasma concentrations of fibrinogen (7 µM) and of bFGF (up to 6 pM) nearly all bFGF should be bound to fibrinogen considering the K_d values in the nanomolar range. However, other bFGF binding proteins, ₂ macroglobulin (44) and soluble forms of FGF receptor (45), have also been identified in blood. The binding of bFGF to ₂ macroglobulin involves formation of covalent bonds and is slow, requiring up to 4 h to reach completion (44). Three soluble truncated forms of the high affinity cell receptor FGFR1 have also been identified in plasma as binding proteins for bFGF (45), but neither the plasma concentration nor binding affinities have been described. Further studies will be required to elucidate the distribution of bFGF binding to these plasma proteins and their role in influencing plasma half-life or bFGF activity.

The binding of bFGF with fibrinogen and fibrin may have effects locally at sites of vessel injury or disease. Fibrinogen is found in both normal and atherosclerotic arterial walls (46-48) and could, therefore, serve as a binding site for bFGF within the matrix. This may also occur with fibrin that is also present in atherosclerotic vessels (47) as well as at sites of injury, inflammation, or tumor growth. Binding of bFGF to fibrin could, therefore, localize both molecules to sites where they are needed to support endothelial cell migration, proliferation, and angiogenesis.

The binding of bFGF to fibrinogen and fibrin may also have effects on interactions with cell receptors and with signal transduction. Binding of endothelial cells to matrix glycoproteins through integrin receptors alters their sensitivity to growth factor-induced signaling mechanisms (49). Fibrinogen and fibrin can support endothelial cell attachment through occupancy of _v3 and the resultant formation of a focal adhesion complex. Studies in vitro demonstrate the co-localization of integrin receptors and the high affinity FGFR with multiple signaling molecules within the focal adhesion complex (50). This organization may foster signal integration between integrins and growth factor receptors. The binding of bFGF to fibrinogen and fibrin could facilitate such close association between their separate receptors. Further interactions between bFGF with fibrinogen and fibrin in vascular responses may result from the role of bFGF in modulating endothelial cell surface integrin expression (51). The binding of bFGF as described in this work may, therefore, serve to both localize its activity and coordinate with fibrinogen and fibrin in supporting angiogenesis and the vascular response to injury.