Portal de Periódicos da CAPES

Identification and Characterization of Novel Lysine-independent Apolipoprotein(a)-binding Sites in Fibrin(ogen) αC-domains

2003; Elsevier BV; Volume: 278; Issue: 39 Linguagem: Inglês

10.1074/jbc.m305154200

1083-351X

Galina Tsurupa, Benoît Ho‐Tin‐Noé, Eduardo Anglés‐Cano, Leonid Medved,

Protease and Inhibitor Mechanisms

Accumulation of lipoprotein(a) (Lp(a)) in atherosclerotic plaques is mediated through interaction of fibrin-(ogen) deposits with the apolipoprotein(a) (apo(a)) moiety of Lp(a). It was suggested that because apo(a) competes with plasminogen for binding to fibrin, causing inhibition of fibrinolysis, it could also promote atherothrombosis. Because the fibrin(ogen) αC-domains bind plasminogen and tissue-type plasminogen activator with high affinity in a Lys-dependent manner, we hypothesized that they could also bind apo(a). To test this hypothesis, we studied the interaction between the recombinant apo(a) A10 isoform and the recombinant αC-fragment (Aα-(221–610)) corresponding to the αC-domain by enzyme-linked immunosorbent assay and surface plasmon resonance. Both methods revealed a high affinity interaction (Kd = 19–21 nm) between the immobilized αC-fragment and apo(a), indicating that the former contains an apo(a)-binding site. This affinity was comparable to that of apo(a) for fibrin. At the same time, no interaction was observed between soluble fibrinogen and immobilized apo(a), suggesting that, in the former, this and other apo(a)-binding sites are cryptic. Further experiments with truncated recombinant variants of the αC-fragment allowed localization of the apo(a)-binding site to the Aα-(392–610) region. The presence of ϵ-aminocaproic acid only slightly inhibited binding of apo(a) to the αC-fragment, indicating the Lys-independent nature of their interaction. In agreement, the influence of plasminogen or tissue-type plasminogen activator on binding of apo(a) to the αC-fragment was minimal. These results indicate that the αC-domains contain novel high affinity apo(a)-binding sites that may provide a Lys-independent mechanism for bringing Lp(a) to places of fibrin deposition such as injured vessels or atherosclerotic lesions. Accumulation of lipoprotein(a) (Lp(a)) in atherosclerotic plaques is mediated through interaction of fibrin-(ogen) deposits with the apolipoprotein(a) (apo(a)) moiety of Lp(a). It was suggested that because apo(a) competes with plasminogen for binding to fibrin, causing inhibition of fibrinolysis, it could also promote atherothrombosis. Because the fibrin(ogen) αC-domains bind plasminogen and tissue-type plasminogen activator with high affinity in a Lys-dependent manner, we hypothesized that they could also bind apo(a). To test this hypothesis, we studied the interaction between the recombinant apo(a) A10 isoform and the recombinant αC-fragment (Aα-(221–610)) corresponding to the αC-domain by enzyme-linked immunosorbent assay and surface plasmon resonance. Both methods revealed a high affinity interaction (Kd = 19–21 nm) between the immobilized αC-fragment and apo(a), indicating that the former contains an apo(a)-binding site. This affinity was comparable to that of apo(a) for fibrin. At the same time, no interaction was observed between soluble fibrinogen and immobilized apo(a), suggesting that, in the former, this and other apo(a)-binding sites are cryptic. Further experiments with truncated recombinant variants of the αC-fragment allowed localization of the apo(a)-binding site to the Aα-(392–610) region. The presence of ϵ-aminocaproic acid only slightly inhibited binding of apo(a) to the αC-fragment, indicating the Lys-independent nature of their interaction. In agreement, the influence of plasminogen or tissue-type plasminogen activator on binding of apo(a) to the αC-fragment was minimal. These results indicate that the αC-domains contain novel high affinity apo(a)-binding sites that may provide a Lys-independent mechanism for bringing Lp(a) to places of fibrin deposition such as injured vessels or atherosclerotic lesions. Elevated plasma levels of lipoprotein(a) (Lp(a)) 1The abbreviations used are: Lp(a), lipoprotein(a); apo(a), apolipoprotein(a); tPA, tissue-type plasminogen activator; ϵ-ACA, ϵ-aminocaproic acid; ELISA, enzyme-linked immunosorbent assay; SPR, surface plasmon resonance. and fibrinogen are independent risk factors for atherosclerotic cardiovascular diseases (1Danesh J. Collins R. Peto R. Circulation. 2000; 102: 1082-1085Crossref PubMed Scopus (821) Google Scholar, 2Danesh J. Collins R. Appleby P. Peto R. J. Am. Med. Assoc. 1998; 279: 1477-1482Crossref PubMed Scopus (1847) Google Scholar, 3Bini A. Kudryk B.J. Ann. N. Y. Acad. Sci. 1995; 748: 461-471Crossref PubMed Scopus (32) Google Scholar). Numerous in vivo experiments with transgenic animals directly proved involvement of Lp(a) and fibrinogen in the development and progression of atherosclerosis. It was demonstrated that transgenic mice expressing human apolipoprotein(a) (apo(a)), a protein component of Lp(a), are more susceptible to diet-induced atherosclerosis (4Lawn R.M. Wade D.P. Hammer R.E. Chiesa G. Verstuyft J.G. Rubin E.M. Nature. 1992; 360: 670-672Crossref PubMed Scopus (249) Google Scholar, 5Lawn R.M. Pearle A.D. Kunz L.L. Rubin E.M. Reckless J. Metcalfe J.C. Grainger D.J. J. Biol. Chem. 1996; 271: 31367-31371Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar, 6Boonmark N.W. Lou X.J. Yang Z.J. Schwartz K. Zhang J.L. Rubin E.M. Lawn R.M. J. Clin. Invest. 1997; 100: 558-564Crossref PubMed Scopus (81) Google Scholar). It was also shown that in apo(a)-transgenic rabbits, in which apo(a) is efficiently assembled into Lp(a), the latter substantially increases the development of aortic and coronary atherosclerosis and accelerates formation of advanced atherosclerotic lesions (7Fan J. Shimoyamada H. Sun H. Marcovina S. Honda K. Watanabe T. Arterioscler. Thromb. Vasc. Biol. 2001; 21: 88-94Crossref PubMed Scopus (79) Google Scholar, 8Ichikawa T. Unoki H. Sun H. Shimoyamada H. Marcovina S. Shikama H. Watanabe T. Fan J. Am. J. Pathol. 2002; 160: 227-236Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar, 9Sun H. Unoki H. Wang X. Liang J. Ichikawa T. Arai Y. Shiomi M. Marcovina S.M. Watanabe T. Fan J. J. Biol. Chem. 2002; 277: 47486-47492Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar). Finally, it was found that fibrinogen deficiency in apo(a) transgenic mice reduces accumulation of apo(a) in the vessel walls and lesion development, suggesting that fibrin(ogen) may provide one of the major sites to which apo(a) binds to the vessel wall and participates in the generation of atherosclerosis (10Lou X.J. Boonmark N.W. Horrigan F.T. Degen J.L. Lawn R.M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 12591-12595Crossref PubMed Scopus (76) Google Scholar). Although the mechanism through which Lp(a) and fibrin-(ogen) may contribute to the atherogenic processes is still not clearly understood, it seems to be connected with their structure and their ability to interact with each other. Lp(a) is a lipoprotein particle composed of a lipid core and two disulfide-linked apolipoproteins, apoB-100 and apo(a). The lipid core and apoB-100 are shared with low density lipoprotein, the major transporter of cholesterol in human plasma; at the same time, apo(a), which shows a high degree of homology to plasminogen, confers unique properties on Lp(a) (11Eaton D.L. Fless G.M. Kohr W.J. McLean J.W. Xu Q.T. Miller C.G. Lawn R.M. Scanu A.M. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3224-3228Crossref PubMed Scopus (353) Google Scholar, 12McLean J.W. Tomlinson J.E. Kuang W.J. Eaton D.L. Chen E.Y. Fless G.M. Scanu A.M. Lawn R.M. Nature. 1987; 330: 132-137Crossref PubMed Scopus (1596) Google Scholar). Because of these structural similarities, Lp(a) was implicated in the delivery of cholesterol to injured blood vessels (13Brown M.S. Goldstein J.L. Nature. 1987; 330: 113-114Crossref PubMed Scopus (179) Google Scholar) and in competition with plasminogen for binding to fibrin and cellular surfaces (14Anglés-Cano E. de la Pena D.A. Loyau S. Ann. N. Y. Acad. Sci. 2001; 936: 261-275Crossref PubMed Scopus (93) Google Scholar); interaction between Lp(a) and fibrin may play a critical role in both cases. It has been established that both plasminogen and apo(a) contain Lys-binding sites (15Wiman B. Wallen P. Thromb. Res. 1977; 10: 213-222Abstract Full Text PDF PubMed Scopus (151) Google Scholar, 16Guevara Jr., J. Jan A.Y. Knapp R. Tulinsky A. Morrisett J.D. Arterioscler. Thromb. 1993; 13: 758-770Crossref PubMed Google Scholar). It is also known that binding of plasminogen to fibrin via these sites is important for its conversion into an active enzyme, plasmin (17Wiman B. Collen D. Nature. 1978; 272: 549-550Crossref PubMed Scopus (298) Google Scholar, 18Hoylaerts M. Rijken D.C. Lijnen H.R. Collen D. J. Biol. Chem. 1982; 257: 2912-2919Abstract Full Text PDF PubMed Google Scholar, 19Verheijen J.H. Nieuwenhuizen W. Wijngaards G. Thromb. Res. 1982; 27: 377-385Abstract Full Text PDF PubMed Scopus (83) Google Scholar). Numerous studies have demonstrated that Lp(a) and apo(a) also interact with fibrin(ogen) via their Lys-binding sites and compete effectively with plasminogen for its interaction with fibrin (14Anglés-Cano E. de la Pena D.A. Loyau S. Ann. N. Y. Acad. Sci. 2001; 936: 261-275Crossref PubMed Scopus (93) Google Scholar, 20Harpel P.C. Gordon B.R. Parker T.S. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 3847-3851Crossref PubMed Scopus (262) Google Scholar, 21Loscalzo J. Weinfeld M. Fless G.M. Scanu A.M. Arteriosclerosis. 1990; 10: 240-245Crossref PubMed Google Scholar, 22Rouy D. Koschinsky M.L. Fleury V. Chapman J. Anglés-Cano E. Biochemistry. 1992; 31: 6333-6339Crossref PubMed Scopus (90) Google Scholar, 23Bas L.C. Duif P.F. Gimpel J.A. Kortlandt W. Bouma B.N. van Rijn H.J. Thromb. Haemostasis. 1992; 68: 185-188Crossref PubMed Scopus (53) Google Scholar, 24Hervio L. Durlach V. Girard-Globa A. Anglés-Cano E. Biochemistry. 1995; 34: 13353-13358Crossref PubMed Scopus (41) Google Scholar). It was suggested that such competition inhibits generation of active plasmin on fibrin associated with atherosclerotic lesions, resulting in inhibition of fibrinolysis and thereby promoting atherogenesis. Apo(a) is a one-chain multidomain glycoprotein consisting of a number of kringle domains and a serine protease domain, which are highly homologous to the corresponding domains of plasminogen (11Eaton D.L. Fless G.M. Kohr W.J. McLean J.W. Xu Q.T. Miller C.G. Lawn R.M. Scanu A.M. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3224-3228Crossref PubMed Scopus (353) Google Scholar, 12McLean J.W. Tomlinson J.E. Kuang W.J. Eaton D.L. Chen E.Y. Fless G.M. Scanu A.M. Lawn R.M. Nature. 1987; 330: 132-137Crossref PubMed Scopus (1596) Google Scholar). The COOH-terminal protease-like domain of apo(a) shares 94% sequence homology with that of plasminogen and contains the same catalytic triad; however, the Arg-Val activation cleavage site present in plasminogen is replaced with Ser-Ile in apo(a) (12McLean J.W. Tomlinson J.E. Kuang W.J. Eaton D.L. Chen E.Y. Fless G.M. Scanu A.M. Lawn R.M. Nature. 1987; 330: 132-137Crossref PubMed Scopus (1596) Google Scholar). The kringle domain of apo(a) adjacent to the protease-like domain is highly homologous to plasminogen kringle domain V, whereas the remaining multiple kringle domains share 61–75% homology with plasminogen kringle domain IV. There are 10 different types of kringle domain IV, one of which appears in a variable number generating isoforms of apo(a) of different sizes (25Lackner C. Cohen J.C. Hobbs H.H. Hum. Mol. Genet. 1993; 2: 933-940Crossref PubMed Scopus (315) Google Scholar). Kringle domain IV type 10 of apo(a) contains a Lys-binding site, which may be responsible for its interaction with fibrin(ogen) (26Klezovitch O. Scanu A.M. Arterioscler. Thromb. Vasc. Biol. 1996; 16: 392-398Crossref PubMed Scopus (20) Google Scholar). Fibrinogen is also a multidomain glycoprotein. It consists of two identical subunits, each of which is formed by three polypeptide chains, Aα, Bβ, and γ (27Henschen A. McDonagh J. Zwaal R.F.A. Hemker H.C. Blood Coagulation. Elsevier Science B.V., Amsterdam1986: 171-241Google Scholar). Both the subunits and the chains are linked together by disulfide bonds and form a number of distinct independently folded domains grouped into several structural regions, the central E-region, two identical terminal D-regions, and the αC-domains (27Henschen A. McDonagh J. Zwaal R.F.A. Hemker H.C. Blood Coagulation. Elsevier Science B.V., Amsterdam1986: 171-241Google Scholar, 28Privalov P.L. Medved L.V. J. Mol. Biol. 1982; 159: 665-683Crossref PubMed Scopus (128) Google Scholar, 29Medved L. Litvinovich S. Ugarova T. Matsuka Y. Ingham K. Biochemistry. 1997; 36: 4685-4693Crossref PubMed Scopus (48) Google Scholar, 30Weisel J.W. Medved L. Ann. N. Y. Acad. Sci. 2001; 936: 312-327Crossref PubMed Scopus (136) Google Scholar). Each D-region is formed by the COOH-terminal portions of the Bβ- and γ-chains and a middle portion of the Aα-chain, whereas each αC-domain is formed by the COOH-terminal portion of the Aα-chain (Aα-(221–610)) (27Henschen A. McDonagh J. Zwaal R.F.A. Hemker H.C. Blood Coagulation. Elsevier Science B.V., Amsterdam1986: 171-241Google Scholar, 30Weisel J.W. Medved L. Ann. N. Y. Acad. Sci. 2001; 936: 312-327Crossref PubMed Scopus (136) Google Scholar, 31Tsurupa G. Tsonev L. Medved L. Biochemistry. 2002; 41: 6449-6459Crossref PubMed Scopus (61) Google Scholar). Both the D-regions and the αC-domains contain Lys-dependent plasminogen-binding sites (32Nieuwenhuizen W. Ann. N. Y. Acad. Sci. 2001; 936: 237-246Crossref PubMed Scopus (50) Google Scholar, 33Tsurupa G. Medved L. Biochemistry. 2001; 40: 801-808Crossref PubMed Scopus (81) Google Scholar, 34Medved L. Nieuwenhuizen W. Thromb. Haemostasis. 2003; 89: 409-419Crossref PubMed Scopus (179) Google Scholar), which may potentially be involved in the interaction with apo(a) kringle domains. The existence of Lys-dependent apo(a)-binding sites in the plasminogen-binding D-regions has been demonstrated recently using the yeast two-hybrid system (35Klose R. Fresser F. Kochl S. Parson W. Kapetanopoulos A. Fruchart-Najib J. Baier G. Utermann G. J. Biol. Chem. 2000; 275: 38206-38212Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar); nothing is known about the involvement of the αC-domains in interaction with apo(a). Because the αC-domains bind plasminogen and tissue-type plasminogen activator (tPA) in a Lys-dependent manner and with high affinity, which is comparable to that for the interaction between apo(a) and fibrin (22Rouy D. Koschinsky M.L. Fleury V. Chapman J. Anglés-Cano E. Biochemistry. 1992; 31: 6333-6339Crossref PubMed Scopus (90) Google Scholar, 33Tsurupa G. Medved L. Biochemistry. 2001; 40: 801-808Crossref PubMed Scopus (81) Google Scholar), we hypothesized that they could also bind apo(a) and that the latter could compete for binding with plasminogen and possibly tPA. To test this hypothesis, we examined the interaction between the recombinant apo(a) A10 isoform and the recombinant αC-domain. We also examined the way in which that interaction is influenced by plasminogen, tPA, and ϵ-aminocaproic acid (ϵ-ACA), a synthetic analog of Lys. We found that the αC-domain bound apo(a) with high affinity; however, this binding was inhibited only slightly by excess plasminogen, tPA, or ϵ-ACA, indicating that it occurs mainly through a novel Lys-independent binding site. This site was further localized to the COOH-terminal half of the αC-domain (residues 392–610). Proteins—Plasminogen-depleted human fibrinogen was purchased from Calbiochem. It also did not contain factor XIII because its incubation with thrombin resulted in non-cross-linked fibrin as revealed by SDS-PAGE under reduced conditions. Recombinant single chain tPA was a Genentech product. Human Glu-plasminogen (form II) was prepared from citrated human plasma by affinity chromatography on Lys-Sepharose 4B (36Deutsch D.G. Mertz E.T. Science. 1970; 170: 1095-1096Crossref PubMed Scopus (1672) Google Scholar) and further purified by size exclusion chromatography on Superdex 200 (Amersham Biosciences). Bovine α-thrombin was from Sigma, and bovine serum albumin was from Calbiochem. Antibodies—The sheep anti-tPA and goat anti-plasminogen polyclonal antibodies were purchased from Chemicon International, Inc. The peroxidase-conjugated anti-sheep and anti-goat polyclonal antibodies were from Sigma. The sheep anti-apo(a) antibodies and monoclonal antibody A10.2 directed against the lysine-binding site of kringle domain IV type 10 of apo(a) were obtained as described (37Kang C. Dominguez M. Loyau S. Miyata T. Durlach V. Anglés-Cano E. Arterioscler. Thromb. Vasc. Biol. 2002; 22: 1232-1238Crossref PubMed Scopus (41) Google Scholar, 38Dominguez M. Rojas G. Loyau S. Bazurco M. Sorell L. Anglés-Cano E. Biochim. Biophys. Acta. 2001; 1548: 72-80Crossref PubMed Scopus (13) Google Scholar). Expression and Purification of Recombinant Proteins—The recombinant αC-fragments corresponding to the human fibrinogen αC-domain (Aα-(221–610)) and its NH2- and COOH-terminal halves (Aα-(221–391) and Aα-(392–610), respectively) were produced in Escherichia coli using the pET-20b expression vector as described earlier (33Tsurupa G. Medved L. Biochemistry. 2001; 40: 801-808Crossref PubMed Scopus (81) Google Scholar). They were purified and then refolded according to the procedures described (31Tsurupa G. Tsonev L. Medved L. Biochemistry. 2002; 41: 6449-6459Crossref PubMed Scopus (61) Google Scholar, 33Tsurupa G. Medved L. Biochemistry. 2001; 40: 801-808Crossref PubMed Scopus (81) Google Scholar). Recombinant apo(a) A10 was produced in adenovirus-transformed human embryonic kidney cells stably transfected with the pCMV-A10 expression plasmid and cultured in a hollow fiber bioreactor as described (39Anglés-Cano E. Loyau S. Cardoso-Saldana G. Couderc R. Gillery P. J. Lipid Res. 1999; 40: 354-359Abstract Full Text Full Text PDF PubMed Google Scholar). The culture medium was supplemented with serine protease inhibitors (20 kallikrein-inhibitory units/ml aprotinin, 1 mmol/liter aminoethylbenzenesulfonyl fluoride, 1 μm d-Phe-Pro-Arg-chloromethyl ketone, 2 mmol/liter EDTA, and 0.01% (w/v) NaN3, final concentrations) and was used for the isolation of recombinant apo(a) by affinity chromatography on Lys-Sepharose 4B equilibrated with 20 mm phosphate buffer (pH 7.4) supplemented with 0.5 m NaCl and serine protease inhibitors. After sample application and washing, the column was equilibrated with 50 mm phosphate buffer (pH 7.4) containing 80 mm NaCl and serine protease inhibitors. To isolate human recombinant apo(a) from adsorbed fetal calf plasminogen present in the culture medium, a stepwise elution procedure was used. Plasminogen was first eluted with 2mm ϵ-ACA. The column was then washed, and recombinant apo(a) was eluted with 20 mm ϵ-ACA in the same buffer. Fractions containing recombinant apo(a) were pooled, concentrated on dried polyethylene glycol 20,000 (Serva), and dialyzed against 50 mm phosphate buffer containing 80 mm NaCl and 2 mmol/liter EDTA. The recombinant apo(a) preparation was >99% pure as assessed by SDS-PAGE and NH2-terminal sequence analysis using an Applied Biosystems microsequenator equipped with a Model 610A data analysis system. Purified recombinant apo(a) A10 consisted of a serine protease-like region, kringle domain V, and nine kringle domain IV repeats representing each of the kringle domain IV types (type 1 and types 3–10) except type 2. Solid-phase Binding Assay—Solid-phase binding was performed in plastic microtiter plates using an enzyme-linked immunosorbent assay (ELISA). Microtiter plate wells (Fisher) were coated overnight with 100 μl/well of 10 μg/ml fibrinogen, αC-fragment, or its truncated variants in 0.1 m Na2CO3 (pH 9.5). To convert fibrinogen into fibrin, the wells with adsorbed fibrinogen were treated with 100 μl/well of a mixture of thrombin (1 NIH unit/ml) and aprotinin (400 units/ml) at 37 °C for 1 h as described previously (40Makogonenko E. Tsurupa G. Ingham K. Medved L. Biochemistry. 2002; 41: 7907-7913Crossref PubMed Scopus (113) Google Scholar). The wells were then blocked with 1% bovine serum albumin in phosphate-buffered saline (0.02 m sodium phosphate buffer (pH 7.4) and 0.15 m NaCl). Following washing with phosphate-buffered saline containing 0.02% Tween 20, the indicated concentrations of apo(a) in the same buffer were added to the wells and incubated for 1 h. Bound apo(a) was measured by the reaction with the sheep anti-apo(a) polyclonal antibody and the peroxidase-conjugated anti-sheep polyclonal antibody. A 3,3′,5,5′-tetramethylbenzidine microwell peroxide substrate (Kirkegaard & Perry Laboratories Inc.) was added to the wells, and the amount of bound ligand was measured spectrophotometrically at 450 nm. Data were analyzed by nonlinear regression analysis using Equation 1, A=Amax/(1+Kd/[L])(Eq. 1) where A represents absorbance of the oxidized substrate, which is assumed to be proportional to the amount of ligand bound; A max is the absorption at saturation; [L] is the molar concentration of ligand; and Kd is the dissociation constant. Biosensor Assay—The interaction of apo(a) with the αC-fragment and its variants was studied by surface plasmon resonance (SPR) using the IAsys biosensor (Fisons, Cambridge, UK), which measures association/dissociation of proteins in real time (41Johnsson B. Lofas S. Lindquist G. Anal. Biochem. 1991; 198: 268-277Crossref PubMed Scopus (1206) Google Scholar). The αC-fragment or its variants were covalently coupled to the aminosilane surface of the cuvette using the glutaraldehyde cross-linking chemistry recommended by the manufacturer. Binding experiments were performed in phosphate-buffered saline containing 0.1 mm phenylmethylsulfonyl fluoride and 0.02% Tween 20 (binding buffer). The association between the immobilized fragments and the added proteins was monitored as the change in the SPR response. The dissociation of the complex was initiated by substitution with the same buffer lacking ligand and monitored in the same manner. To regenerate the surface, complete dissociation of the complex was achieved by adding 10 mm HCl for 1 min following re-equilibration with binding buffer. The traces of the association processes were recorded, and the data were analyzed using the FASTfit™ kinetics analysis software supplied with the instrument as previously described in detail (42Gorgani N.N. Parish C.R. Altin J.G. J. Biol. Chem. 1999; 274: 29633-29640Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). Briefly, the association curves at each concentration of ligand were fitted to the pseudo first-order equation to derive the observed rate constant, k obs (termed on-rate constant in FASTfit). Then, the concentration dependence of k obs was fitted to Equation 2, kobs=kd+ka[ligand](Eq. 2) to find the association rate constant (ka) from the slope and the dissociation rate constant (kd) from the intercept. The dissociation equilibrium constant (Kd) was calculated as Kd = kd /ka. The values were examined for self-consistency of the data as described (43Schuck P. Minton A.P. Trends Biochem. Sci. 1996; 21: 458-460Abstract Full Text PDF PubMed Scopus (123) Google Scholar). Interaction of Apo(a) A10 with the αC-Fragment—To check whether the fibrinogen αC-domains are involved in binding of apo(a), we studied the interaction between the recombinant apo(a) A10 variant and the recombinant αC-fragment (Aα-(221–610)) corresponding to the αC-domain; in control experiments, we tested the interaction between apo(a) and fibrinogen. Prior to binding experiments, all three species (the αC-fragment, fibrinogen, and fibrin) were treated with carboxypeptidase B according to the protocol described (44Ellis V. Dano K. J. Biol. Chem. 1993; 268: 4806-4813Abstract Full Text PDF PubMed Google Scholar) to remove possible COOH-terminal Lys residues that could contribute to binding of apo(a). In ELISA, when increasing concentrations of apo(a) A10 were added to the immobilized αC-fragment, a dose-dependent binding was observed (Fig. 1). The binding was of high affinity with an apparent Kd of 19 nm (Table I). This Kd was very close to that obtained in the same assay for the interaction between apo(a) A10 and immobilized fibrin (Fig. 1 and Table I). This high affinity interaction was confirmed in SPR experiments, in which apo(a) A10 bound to the immobilized αC-fragment in a dose-dependent manner with a Kd of 21 nm (Fig. 2 and Table I). All these results indicate that the αC-fragment (and the αC-domain) contains a high affinity apo(a)-binding site. It should be noted that when either the immobilized αC-fragment or fibrin was cross-linked with factor XIIIa, the binding curves obtained by ELISA and the calculated Kd values were similar to those obtained with the non-cross-linked species (data not shown), suggesting that cross-linking does not impact this binding.Table IDissociation constants for the interaction of apo(a) A10 with immobilized fibrinogen, fibrin, and the recombinant αC-fragment and its subfragments determined by ELISA and SPRProteins/fragmentsKdELISASPRnmαC-fragment (Aα-(221-610))19 ± 421 ± 3Aα-(221-391) fragmentNBNBAα-(392-610) fragment22 ± 424 ± 3Fibrin13 ± 8NDFibrinogen138 ± 16ND Open table in a new tab Fig. 2Analysis of binding of apo(a) A10 to the immobilized αC-fragment by SPR. Increasing concentrations of apo(a) A10 were added to the immobilized αC-fragment, and their association was monitored in real time while registering the resonance signal (response). The concentrations of apo(a) were 6, 12, 25, 50, 100, and 200 nm. The inset shows a plot of the values of k obs determined for each association curve versus ligand concentration to derive ka and kd and thus to determine the dissociation equilibrium constants (Kd) presented in Table I. Here and in Figs. 4 and 6, a signal of 600 arc s corresponds to 1 ng of protein bound per mm2 of sensor chip surface.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Interaction of Apo(a) A10 with Fibrinogen—It is well established that the tPA- and plasminogen-binding sites of the αC-domains and the D-regions are cryptic in fibrinogen and are exposed upon its conversion to fibrin or upon its adsorption to a surface (34Medved L. Nieuwenhuizen W. Thromb. Haemostasis. 2003; 89: 409-419Crossref PubMed Scopus (179) Google Scholar, 45Medved L. Tsurupa G. Yakovlev S. Ann. N. Y. Acad. Sci. 2001; 936: 185-204Crossref PubMed Scopus (48) Google Scholar). To test whether the apo(a)-binding sites in the fibrinogen αC-domains are also cryptic, we studied the interaction of soluble fibrinogen with immobilized apo(a) A10. In ELISA, when soluble fibrinogen at a high concentration (1 μm) was added to immobilized apo(a) A10, no binding was observed (data not shown), suggesting that the apo(a)-binding sites in fibrinogen (and in its αC-domains) are cryptic. It should be mentioned that, in a reverse ELISA experiment, soluble apo(a) A10 bound to immobilized fibrinogen in a dose-dependent manner, although with much lower affinity (Kd = 138 nm) (Fig. 1, inset). However, this binding can be explained by the well established fact that, upon immobilization, fibrinogen undergoes conformational changes resulting in exposure of some fibrin-specific epitopes and binding sites, including those for plasminogen and tPA (45Medved L. Tsurupa G. Yakovlev S. Ann. N. Y. Acad. Sci. 2001; 936: 185-204Crossref PubMed Scopus (48) Google Scholar, 46Shiba E. Lindon J.N. Kushner L. Matsueda G.R. Hawiger J. Kloczewiak M. Kudryk B. Salzman E.W. Am. J. Physiol. 1991; 260: C965-C974Crossref PubMed Google Scholar, 47Moskowitz K.A. Kudryk B. Coller B.S. Thromb. Haemostasis. 1998; 79: 824-831Crossref PubMed Scopus (45) Google Scholar). The cryptic character of the apo(a)-binding sites in fibrinogen was confirmed by SPR, in which fibrinogen at the same concentration (1 μm) also exhibited no binding to immobilized apo(a) (data not shown). Thus, the above results indicate that apo(a)-binding sites are cryptic in fibrinogen and become exposed upon its immobilization or conversion into fibrin. Further Localization of the Apo(a)-binding Site to the αC-Domain—To further localize the apo(a)-binding site to the αC-domain, we tested binding of apo(a) A10 to the truncated recombinant variants of the αC-domain, the Aα-(221–391) and Aα-(392–610) fragments, corresponding to its NH2- and COOH-terminal halves, respectively. In ELISA, apo(a) A10 bound to the immobilized Aα-(392–610) fragment in a dose-dependent manner, whereas practically no interaction was observed with the immobilized Aα-(221–391) fragment (Fig. 3). This binding occurred with a Kd of 22 nm, very close to that determined for the full-length αC-fragment (Table I). This was confirmed in SPR experiments, in which apo(a) A10 bound to the immobilized COOH-terminal region of the αC-domain in a dose-dependent manner with a very similar affinity (Kd = 24 nm) (Fig. 4 and Table I). These results indicate that the apo(a)-binding site is located in the COOH-terminal half of the αC-domain (Aα-(392–610)), the same region that also binds plasminogen and tPA (33Tsurupa G. Medved L. Biochemistry. 2001; 40: 801-808Crossref PubMed Scopus (81) Google Scholar). This finding is in agreement with the above-mentioned hypothesis that apo(a) A10 could compete with plasminogen (and/or tPA) for its binding sites in the αC-domain.Fig. 4Analysis of binding of apo(a) A10 to the αC-fragment and its NH2- and COOH-terminal fragments (Aα-(221–392) and Aα-(393–610), respectively) by SPR. The association of apo(a) A10 at 200 nm with the immobilized αC-fragment (dashed curve) and its subfragments, Aα-(221–391) (dotted curve) and Aα-(392–610) (solid curve), was monitored in real time while registering the resonance signal. The inset shows dose-dependent binding of apo(a) A10 to the immobilized