Noncovalent interaction of alpha(2)-antiplasmin with fibrin(ogen): localization of alpha(2)-antiplasmin-binding sites

Abstract

Covalent incorporation (cross-linking) of plasmin inhibitor alpha(2)-antiplasmin (alpha(2)-AP) into fibrin clots increases their resistance to fibrinolysis. We hypothesized that alpha(2)-AP may also interact noncovalently with fibrin prior to its covalent cross-linking. To test this hypothesis, we studied binding of alpha(2)-AP to fibrin(ogen) and its fragments by an enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance. The experiments revealed that alpha(2)-AP binds to polymeric fibrin and surface-adsorbed fibrin(ogen), while no binding was observed with fibrinogen in solution. To localize the alpha(2)-AP-binding sites, we studied the interaction of alpha(2)-AP with the fibrin(ogen)-derived D(1), D-D, and E(3) fragments, and the recombinant alphaC region and its constituents, alphaC connector and alphaC domain and its subdomains, which together encompass practically the whole fibrin(ogen) molecule. In the ELISA, alpha(2)-AP bound to immobilized D(1), D-D, alphaC region, alphaC domain, and its C-terminal subdomain. The binding was Lys-independent and was not inhibited by plasminogen or tPA. Furthermore, the affinity of alpha(2)-AP for D-D was significantly increased in the presence of plasminogen, while that to the alphaC domain remained unaffected. Altogether, these results indicate that the fibrin(ogen) D region and the C-terminal subdomain of the alphaC domain contain high-affinity alpha(2)-AP-binding sites that are cryptic in fibrinogen and exposed in fibrin or adsorbed fibrinogen, and the presence of plasminogen facilitates interaction of alpha(2)-AP with the D regions. The discovered noncovalent interaction of alpha(2)-AP with fibrin may contribute to regulation of the initial stage of fibrinolysis and provide proper orientation of the cross-linking sites to facilitate covalent cross-linking of alpha(2)-AP to the fibrin clot.

Fig. 1

Schematic representation of fibrinogen, fibrin, and their fragments prepared for this study. Panel A: Ribbon diagram of fibrinogen based on its crystal structure (47); the individual fibrinogen chains, Aα, Bβ, and γ, are colored blue green, and red, respectively, the vertical lines denote approximate boundaries between the D and E regions. The αC regions, whose structure have not been identified, are shown schematically as two blue spheres representing αC-domains, each attached to the bulk of the molecule with the flexible αC-connector. Panel B: Schematic representation of the fibrinogen molecule and its products of plasminolysis, D₁ and E₃ fragments. Panel C: Recombinant αC region (Aα221–610 fragment), αC-connector (Aα221–391 fragment), and αC-domain (Aα392–610 fragment). Panel D: Schematic representation of fibrin and its products of fibrinolysis, the D-D:E complex, and the D-D and E₃ fragments. For the sake of simplicity, only two strands of fibrin molecules without the αC regions are shown; the molecules are linked through the non-covalent DD:E interactions and covalent γ-γ cross-linking between the D regions (shown by small horizontal bars). Small arrows in panels B and D indicate plasmin cleavage resulting in fibrin(ogen) fragments.

Fig. 2

Analysis of the interaction of α₂-antiplasmin with fibrinogen and fibrin by ELISA. Increasing concentrations of biotinylated α₂-AP were added to surface-adsorbed fibrinogen (empty circles) or fibrin (filled circles). Bound biotinylated α₂-AP was detected with avidin conjugated to alkaline phosphatase as described in Experimental Procedures. The curves represent the best fit of the data to eq. 1. All results are means ± the standard deviation of duplicate determinations.

Fig. 3

Analysis of the interaction of α₂-antiplasmin with polymeric fibrin by surface plasmon resonance. α₂-AP at increasing concentrations, 8, 16, 32, 63, 125, 250, 500, 1000 and 2000 nM, was added to immobilized fibrin polymers (see Experimental Procedures) and its association/dissociation was monitored in real time. The inset shows the results of the equilibrium analysis; the amount of bound a₂-antiplasmin at equilibrium for each concentration is presented by circles and the best fit is presented by solid curve.

Fig. 4

ELISA-detected binding of α₂-antiplasmin to immobilized fibrin(ogen) fragments. Increasing concentrations of biotinylated α₂-AP were added to the immobilized D-D (circles), Aα221–610 (diamonds), and E₃ (triangles) fragments. Bound biotinylated α₂-AP was detected with avidin conjugated to alkaline phosphatase. The curves represent the best fit of the data to eq. 1. All results are means ± the standard deviation of duplicate determinations.

Fig. 5

Analysis of the interaction between α₂-antiplasmin and the D-D:E complex by surface plasmon resonance. The D-D:E complex at increasing concentrations, 32, 63, 125, 250, 500, 1000, 1500 and 2000 nM, was added to immobilized α₂-AP and its association/dissociation was monitored in real time. The inset shows the results of the equilibrium analysis; the amount of bound D-D:E for each concentration is presented by circles and the best fit is presented by solid curve.

Fig. 6

ELISA-detected binding of α₂-antiplasmin to immobilized recombinant αC-fragments. Increasing concentrations of biotinylated α₂-AP were added to the immobilized Aα221–391 (empty circles), Aα392–610 (filled circles), Aα392–503 (empty diamonds), and Aα504–610 (filled triangles) fragments. Bound biotinylated α₂-AP was detected with avidin conjugated to alkaline phosphatase. The curves represent the best fit of the data to eq. 1. All results are means ± the standard deviation of duplicate determinations.

Fig. 7

Simultaneous binding of α₂-antiplasmin, tPA, and plasminogen to immobilized fibrin detected by ELISA. A mixture of 0.5 μM biotinylated α₂-AP, 2.5 μM tPA, and 2.5 μM plasminogen was added to surface-adsorbed fibrin in the absence (filled bars) or presence (empty bars) of 100 mM ε-ACA. Bound biotinylated α₂-AP was detected spectrophotometrically at 405 nm using avidin conjugated to alkaline phosphatase, bound tPA and plasminogen were detected spectrophotometrically at 450 nm using the anti-tPA monoclonal antibody and specific polyclonal antibodies, respectively, as described in Experimental Procedures. All results are means ± the standard deviation of two independent experiments, each performed in duplicate.

Fig. 8

The effect of tPA or plasminogen on the binding of α₂-antiplasmin to immobilized fibrin and its fragments detected by ELISA. Panel A: Increasing concentrations of biotinylated α₂-AP were added to surface-adsorbed fibrin (triangles), the D-D (circles), or Aα392–610 fragments (diamonds) in the absence (filled symbols, solid lines) or presence (empty symbols, broken lines) of 2.5 μM tPA. Panel B: Increasing concentrations of biotinylated α₂-AP were added to surface-adsorbed fibrin (triangles), the D-D (circles), or Aα392–610 fragments (diamonds) in the absence (filled symbols, broken lines) or presence (empty symbols, broken lines) of 2.5 μM plasminogen. Bound biotinylated α₂-AP was detected with avidin conjugated to alkaline phosphatase. The curves represent the best fit of the data to eq. 1. All results are means ± the standard deviation of duplicate determinations.

Fig. 9

Schematic representation of the non-covalent and covalent interactions between α₂-antiplasmin and fibrin(ogen) αC-domain. α₂-AP with flexible NH₂-termial portion containing reactive Gln14 (Q) is presented on the top. The αC region containing the flexible αC-connector with reactive Lys303 (K) and compact αC-domain consisting of two sub-domains is presented on the bottom; the structure of the αC-domain is adapted from (36). The non-covalent interaction is shown on the right, the covalent cross-linking (Q–K) between reactive Gln14 and Lys303 is shown on the left.