Mapping of a Minimal Apolipoprotein(a) Interaction Motif Conserved in Fibrin(ogen) β- and γ-Chains
2000; Elsevier BV; Volume: 275; Issue: 49 Linguagem: Inglês
10.1074/jbc.m003640200
ISSN1083-351X
AutoresRegina Klose, Friedrich Fresser, Silvano Köchl, Walther Parson, Andreas Kapetanopoulos, Jamila Fruchart‐Najib, Gottfried Baier, Gerd Utermann,
Tópico(s)Platelet Disorders and Treatments
ResumoLipoprotein(a) (Lp(a)) is a major independent risk factor for atherothrombotic disease in humans. The physiological function(s) of Lp(a) as well as the precise mechanism(s) by which high plasma levels of Lp(a) increase risk are unknown. Binding of apolipoprotein(a) (apo(a)) to fibrin(ogen) and other components of the blood clotting cascade has been demonstrated in vitro, but the domains in fibrin(ogen) critical for interaction are undefined. We used apo(a) kringle IV subtypes to screen a human liver cDNA library by the yeast GAL4 two-hybrid interaction trap system. Among positive clones that emerged from the screen, clones were identified as fibrinogen β- and γ-chains. Peptide-based pull-down experiments confirmed that the emerging peptide motif, conserved in the carboxyl-terminal globular domains of the fibrinogen β and γ modules specifically interacts with apo(a)/Lp(a) in human plasma as well as in cell culture supernatants of HepG2 and Chinese hamster ovary cells, ectopically expressing apo(a)/Lp(a). The influence of lysine in the fibrinogen peptides and of lysine binding sites in apo(a) for the interaction was evaluated by binding experiments with apo(a) mutants and a mutated fibrin(ogen) peptid. This confirmed the lysine binding sites in kringle IV type 10 of apo(a) as the major fibrin(ogen) binding site but also demonstrated lysine-independent interactions. Lipoprotein(a) (Lp(a)) is a major independent risk factor for atherothrombotic disease in humans. The physiological function(s) of Lp(a) as well as the precise mechanism(s) by which high plasma levels of Lp(a) increase risk are unknown. Binding of apolipoprotein(a) (apo(a)) to fibrin(ogen) and other components of the blood clotting cascade has been demonstrated in vitro, but the domains in fibrin(ogen) critical for interaction are undefined. We used apo(a) kringle IV subtypes to screen a human liver cDNA library by the yeast GAL4 two-hybrid interaction trap system. Among positive clones that emerged from the screen, clones were identified as fibrinogen β- and γ-chains. Peptide-based pull-down experiments confirmed that the emerging peptide motif, conserved in the carboxyl-terminal globular domains of the fibrinogen β and γ modules specifically interacts with apo(a)/Lp(a) in human plasma as well as in cell culture supernatants of HepG2 and Chinese hamster ovary cells, ectopically expressing apo(a)/Lp(a). The influence of lysine in the fibrinogen peptides and of lysine binding sites in apo(a) for the interaction was evaluated by binding experiments with apo(a) mutants and a mutated fibrin(ogen) peptid. This confirmed the lysine binding sites in kringle IV type 10 of apo(a) as the major fibrin(ogen) binding site but also demonstrated lysine-independent interactions. lipoprotein(a) apolipoprotein(a) lysine binding site plasminogen phenyl-methyl-sulfonyl-fluoride Chinese hamster ovary ε-aminocaproic acid HEPES-buffered saline wild type monoclonal antibody polyacrylamide gel electrophoresis Lipoprotein(a) (Lp(a))1from human plasma is composed of a low density lipoprotein core and the highly polymorphic apo(a), covalently linked to apo B-100 by a single disulfide bridge (1Brunner C. Kraft H.G. Utermann G. Muller H.J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11643-11647Crossref PubMed Scopus (148) Google Scholar, 2Gabel B. Yao Z. McLeod R.S. Young S.G. Koschinsky M.L. FEBS Lett. 1994; 350: 77-81Crossref PubMed Scopus (29) Google Scholar). apo(a) contains 10 distinct tandem repeats, named kringle IV types 1–10, closely resembling plg kringle IV followed by single plg kringle V-like and protease-like domains (3McLean 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 (1653) Google Scholar). The homology at the cDNA level between plg and apo(a) modules is 75–85% for the kringle IV domains and 94% for the protease domain (3McLean 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 (1653) Google Scholar). As a result of a size polymorphism in the apo(a) gene, more than 30 different apo(a) isoforms have been found in human plasma, differing in the number of the kringle IV type 2 repeat (4Lackner C. Boerwinkle E. Leffert C.C. Rahmig T. Hobbs H.H. J. Clin. Invest. 1991; 87: 2153-2161Crossref PubMed Scopus (367) Google Scholar, 5Koschinsky M.L. Beisiegel U. Henne-Bruns D. Eaton D.L. Lawn R.M. Biochemistry. 1990; 29: 640-644Crossref PubMed Scopus (128) Google Scholar, 6Utermann G. Menzel H.J. Kraft H.G. Duba H.C. Kemmler H.G. Seitz C. J. Clin. Invest. 1987; 80: 458-465Crossref PubMed Scopus (756) Google Scholar). Several epidemiological studies indicate that elevated Lp(a) levels are an independent risk factor for coronary heart disease (7Utermann G. Scriver C.R. Beaudet A.L. Sly W.S. Valle D. The Metabolic and Molecular Bases of Inherited Disease. McGraw-Hill, New York1995: 1887-1912Google Scholar). Lp(a) accumulates in atherosclerotic lesions of coronary bypass patients and can be cross-linked to the fibrin thrombus (8Pepin J.M. O'Neil J.A. Hoff H.F. J. Lipid Res. 1991; 32: 317-327Abstract Full Text PDF PubMed Google Scholar, 9Liu J.N. Kung W. Harpel P.C. Gurewich V. Biochemistry. 1998; 37: 3949-3954Crossref PubMed Scopus (12) Google Scholar, 10Romanic A.M. Arleth A.J. Willette R.N. Ohlstein E.H. Circ. Res. 1998; 83: 264-269Crossref PubMed Scopus (39) Google Scholar). Binding of Lp(a) to fibrin(ogen) has been hypothesized to underlie a postulated role of Lp(a) in wound healing. Bound Lp(a) may protect the thrombus from premature digestion by plasmin and serve as an important source of phospholipids and cholesterol for membrane biogenesis and cell proliferation at the site of injury (11Papagrigorakis E. Iliopoulos D. Asimacopoulos P.J. Safi H.J. Weilbaecher D.J. Ghazzaly K.G. Nava M.L. Gaubatz J.W. Morrisett J.D. Clin. Genet. 1997; 52: 262-271Crossref PubMed Scopus (36) Google Scholar, 12Yano Y. Shimokawa K. Okada Y. Noma A. J Histochem. Cytochem. 1997; 45: 559-568Crossref PubMed Scopus (46) Google Scholar, 13Nielsen L.B. Stender S. Kjeldsen K. Nordestgaard B.G. Circ. Res. 1996; 78: 615-626Crossref PubMed Scopus (49) Google Scholar). Whereas the physiological role of Lp(a) remains unknown, several hypotheses have been proposed to account for the pathogenicity of Lp(a) (14Hobbs H.H. White A.L. Curr. Opin. Lipidol. 1999; 10: 225-236Crossref PubMed Scopus (165) Google Scholar). The high degree of homology between apo(a) and plg has been suggested to form the basis for the pathogenicity of Lp(a) as a modulator of fibrinolysis (15Miles L.A. Fless G.M. Scanu A.M. Baynham P. Sebald M.T. Skocir P. Curtiss L.K. Levin E.G. Hoover-Plow J.L. Plow E.F. Thromb. Haemostasis. 1995; 73: 458-465Crossref PubMed Scopus (62) Google Scholar, 16Palabrica T.M. Liu A.C. Aronovitz M.J. Furie B. Lawn R.M. Furie B.C. Nat. Med. 1995; 1: 256-259Crossref PubMed Scopus (137) Google Scholar, 17Grainger D.J. Kemp P.R. Liu A.C. Lawn R.M. Metcalfe J.C. Nature. 1994; 370: 460-462Crossref PubMed Scopus (344) Google Scholar, 18Rouy D. Grailhe P. Nigon F. Chapman J. Angles-Cano E. Arterioscler. Thromb. 1991; 11: 629-638Crossref PubMed Google Scholar). Lp(a) effectively competes with plg for binding sites on fibrin and fibrinogen and reduces the generation of active plasmin (18Rouy D. Grailhe P. Nigon F. Chapman J. Angles-Cano E. Arterioscler. Thromb. 1991; 11: 629-638Crossref PubMed Google Scholar, 19Harpel P.C. Gordon B.R. Parker T.S. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 3847-3851Crossref PubMed Scopus (279) Google Scholar, 20Edelberg J.M. Gonzalez-Gronow M. Pizzo S.V. Thromb. Res. 1990; 57: 155-162Abstract Full Text PDF PubMed Scopus (93) Google Scholar, 21Simon D.I. Fless G.M. Scanu A.M. Loscalzo J. Biochemistry. 1991; 30: 6671-6677Crossref PubMed Scopus (76) Google Scholar). It has been demonstrated that Lp(a) increases smooth muscle cell migration and proliferation by inhibition of transforming growth factor-β activation by plasmin (16Palabrica T.M. Liu A.C. Aronovitz M.J. Furie B. Lawn R.M. Furie B.C. Nat. Med. 1995; 1: 256-259Crossref PubMed Scopus (137) Google Scholar, 17Grainger D.J. Kemp P.R. Liu A.C. Lawn R.M. Metcalfe J.C. Nature. 1994; 370: 460-462Crossref PubMed Scopus (344) Google Scholar, 22Kojima S. Harpel P.C. Rifkin D.B. J. Cell Biol. 1991; 113: 1439-1445Crossref PubMed Scopus (204) Google Scholar). In addition, Lp(a) competes with plg for binding to receptors present on endothelial cells and monocytes (23Miles L.A. Fless G.M. Levin E.G. Scanu A.M. Plow E.F. Nature. 1989; 339: 301-303Crossref PubMed Scopus (522) Google Scholar, 24Hajjar K.A. Gavish D. Breslow J.L. Nachman R.L. Nature. 1989; 339: 303-305Crossref PubMed Scopus (602) Google Scholar). The aim of the study was to identify critical motifs in fibrin(ogen) interacting with Lp(a)/apo(a) for the understanding of the functions and pathophysiological properties of Lp(a). Here we present fibrin(ogen) β- and γ-chain sequences interacting with apo(a) in the yeast two-hybrid system and the identification of a conserved 30-amino acid fibrin(ogen) minimal peptide motif that is sufficient for binding to apo(a)/Lp(a) in vitro. The genotype of theSaccharomyces cerevisiae reporter strain HF7c used for the two-hybrid screening, is MATa, ura3-52, his3-200, ade2-101, lys2-801, trp1-901, leu2-3, 112, gal4-542, gal80-538, LYS2::GAL1-HIS3, URA3:: (GAL4 17-mers)3-CYC1-lacZ. The strain CG1945, used for the β-galactosidase assays, is characterized by an additional cycloheximide resistance. The genotype of SFY526, used to test interactions between GAL4 DNA binding domain and GAL4 transcription activation domain fusions, is MATa, ura3-52, his3-200, ade2-101, lys2-801, trp1-901, leu2-3, 112, canr, gal4-542, gal80-538, URA3::GAL1-lacZ (CLONTECH Laboratories, Inc, Palo Alto, CA). Strains were grown under standard conditions in rich or synthetic medium with appropriate supplements at 30 °C. apo(a)KIV-2 and apo(a)KIV-6 were cloned as described in Ref. 25Kochl S. Fresser F. Lobentanz E. Baier G. Utermann G. Blood. 1997; 90: 1482-1489Crossref PubMed Google Scholar. apo(a)KIV-5, -7, -8, -9, and -10 were cloned likewise using polymerase chain reaction and recombinant polymerase chain reaction primers (apo(a)KIV-5: 5′-CCAAGCGAATTCGGTGGCGGTGGATCCGCACTGACTGAGGAAACCCCC-3′ and 5′-GCTTTGGTCGACTCATCATTCTTCAGAAGAAGCCTCTGTGCTTGGAT-3′; apo(a)KIV-7: 5′-CCAAGCGAATTCGGTGGCGGTGGATCCGCACCAACGGAGCAAAGCCCCA-3′ and 5′-GCTTTGGTCGACTCATCATTCTTCAGAAGGAAGCTCTGTGCTTGGAACT-3′; apo(a)KIV-8: 5′-CCAAGCGAATTCGGTGGCGGTGGATCCGCACCAACTGAAAACAGCA-3′ and 5′-GCTTTGGTCGACTCATCATTGTTCAGAAGGAGCCTCTGTGCT-3′; apo(a)KIV-9: 5′-CCAAGCGAATTCGGTGGCGGTGGATCCGCACCACCTGAGAAAAGCCCCTGT-3′ and 5′-GCTTTGGTCGACTCATCATGCTTCAGAATGAGCCTCC-3′; and apo(a)KIVtype-10: 5′-CCAAGCGAATTCGGTGGCGGTGGATCCGCACCAACTGAGCAAAC-3′ and 5′-ATTCCCGTCGACTCATCATTGTTCAGAAGGAGGCCCTAG-3′). Plasmids pCMV-A18 (A18 wt), pCMV-A18ΔVP (ΔKV-P), pCMV-A18_4174Arg (A18-Arg), and pCMV-A18Δ32–35 (ΔKIV 5–8) are described in Ref. 26Ernst A. Helmhold M. Brunner C. Petho-Schramm A. Armstrong V.W. Muller H.J. J. Biol. Chem. 1995; 270: 6227-6234Abstract Full Text Full Text PDF PubMed Scopus (116) Google Scholar. Plasmid pCMV-A18-AS (ΔKIV 8-P) is described in Ref. 27Ogorelkova M. Gruber A. Utermann G. Hum. Mol. Genet. 1999; 8: 2087-2096Crossref PubMed Scopus (62) Google Scholar. For the yeast two-hybrid screening, apo(a)KIV-6 was cotransformed with the human liver cDNA Matchmaker library in the pGAD10 vector (CLONTECH Laboratories, Inc., Palo Alto, CA) into the HF7c yeast strain, as described by the manufacturer, and the transformants were plated to synthetic dropout medium lacking leucine, histidine, and tryptophan but containing 5 mm3-amino-1,2,4-triazole. The plates were incubated at 30 °C for up to 7 days. His+colonies were assayed for β-galactosidase activity by transferring individual colonies on filters placed on selection medium. The plates were incubated for 2 days at 30 °C, and the filters lifted and immersed in liquid nitrogen for 10 s. After thawing at room temperature, the filters were placed on filter circles saturated with 0.2 mg/ml 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside in Z buffer (60 mm Na2HPO4, 40 mm NaH2PO4, 10 mm KCl, 1 mm MgSO4, 30 mmβ-mercaptoethanol) in a Petri dish (permeabilized cells up) and incubated overnight as indicated. To verify the significance of the interaction a liquid culture assay using o-nitrophenyl β-d-galactopyranoside as substrate was performed as described by the manufacturer (CLONTECH Laboratories, Inc., Palo Alto, CA). At least six individual cotransformants were assayed in the background of two different yeast strains. Only interactions of cotransformants in both yeast strains were asumed to be significant. The results are presented as the means ± S.D. Biotinylated peptides FibG207–235 and FibKA207–235 were purchased at Neosystem (Strasbourg, France). Biotinylated peptides FibG190–235 and FibKA190–235 were prepared on an automated synthesizer ABI 421A (Applied Biosystems Inc.) using standard Boc/Bzl strategy. Biotin was coupled to the peptidyl-resin viaN-hydroxy-succinimidyl-6-(biotinamido)-hexanoate in dimethyl formamide for 16 h at room temperature (28Tam J.P. Heath W.P. Merrifield R.B. J. Am. Chem. Soc. 1983; 105: 6442-6455Crossref Scopus (538) Google Scholar). The biotinylated peptidyl-resin was then washed twice with ter-amylalcohol, acetic acid, ter-amylalcohol, and diethylether. The peptides were cleaved from the vacuum-dried resins and simultaneously deprotected according to high hydrogen fluoride procedure (29Suter M. Butler J.E. Immunol. Lett. 1986; 13: 313-316Crossref PubMed Scopus (73) Google Scholar). The peptides were purified and characterized by reversed phase high pressure liquid chromatography, capillary electrophoresis, and amino acid analysis. The biotinylated control peptide derived from protein kinase Cα (RFARKGSLRQKNVY) was from Genosys (Cambridge, UK). A total of 5 ml of transformed yeast cells grown overnight in selective medium lacking tryptophan were used to inoculate 15 ml of yeast extract peptone dextrose medium. At anA600 of 0.5, the cells were pelleted, washed, resuspended at 5 × 108 cells/ml in ice-cold lysis buffer (50 mm Tris, pH 7.5, 150 mm NaCl, 1% Nonidet P-40, 0.25% sodium deoxycholate, 1 mmEDTA, 1 μg/ml aprotinin/leupeptin, 1 mm PMSF) and frozen at −20 °C. Samples were analyzed by SDS-PAGE (10%) and transferred to polyvinylidene fluoride membrane (Millipore, Vienna, Austria), and fusion proteins were detected using a GAL4 DNA binding domain specific antibody (Santa Cruz Inc, Santa Cruz, CA), followed by a rabbit antimouse IgG-peroxidase conjugate and a chemiluminescence detection kit (ECL reagent; Amersham Pharmacia Biotech). The human hepatocarcinoma cell line HepG2 (30Aden D.P. Fogel A. Plotkin S. Damjanov I. Knowles B.B. Nature. 1979; 282: 615-616Crossref PubMed Scopus (1068) Google Scholar) and CHO cells were obtained from the American Type Culture Collection (Manassas, VA) and cultured as recommended by the American Type Culture Collection. Transient transfection of cells was achieved by liposome-mediated gene transfer with the LipofectAMINE reagent (Life Technologies, Inc.) according to the manufacturer's protocol. After overnight incubation the transfection mixture was replaced by 2 ml of growth medium. After 48 h, cell culture supernatants were harvested, treated with proteinase inhibitors (1 mm PMSF, 5 μg/ml of each aprotinin and leupeptin), and centrifuged for 10 min each at 300 and 4000 × g to remove cells and cell debris, respectively. The Lp(a) content was assayed as described elsewhere (31Kraft H.G. Lingenhel A. Pang R.W. Delport R. Trommsdorff M. Vermaak H. Janus E.D. Utermann G. Eur. J. Hum. Genet. 1996; 4: 74-87Crossref PubMed Scopus (119) Google Scholar, 32Trommsdorff M. Kochl S. Lingenhel A. Kronenberg F. Delport R. Vermaak H. Lemming L. Klausen I.C. Faergeman O. Utermann G. Kraft H.G. J. Clin. Invest. 1995; 96: 150-157Crossref PubMed Scopus (144) Google Scholar) Venous blood samples from healthy donors were collected in EDTA tubes, treated with proteinase inhibitors (1 mm PMSF, 5 μg/ml of each aprotinin and leupeptin), and centrifuged for 10 min at 300 ×g to remove cells. Following enzyme-linked immunosorbent assay determination of Lp(a) plasma levels and prior to diluting for the pull-down experiments, the integrity of plasma apo(a) was analyzed by SDS-PAGE and immunoblotting. The Lp(a) isoforms used consisted of 18 apo(a) kringle units for plasma pull-down experiments and 21 apo(a) kringle units for pull-downs of purified Lp(a). Preparation of Lp(a) was performed by density centrifugation as described in Ref. 6Utermann G. Menzel H.J. Kraft H.G. Duba H.C. Kemmler H.G. Seitz C. J. Clin. Invest. 1987; 80: 458-465Crossref PubMed Scopus (756) Google Scholar. Diluted Lp(a) preparations, diluted plasma samples and cell supernatants that had been diluted with HBS (50 mm HEPES, 200 mm NaCl, pH 7.5) with or without 0.2 m EACA were precleared with110volume Pansorbin cells (Calbiochem, San Diego, CA) at 4 °C for 1 h. Precleared samples were incubated overnight at 4 °C with the indicated biotinylated peptide at 3 nm final peptide concentration followed by addition of 20 μl of Biobeads-streptavidin (Merck) for the last 30 min. Precipitates were collected on a magnetic rack and washed four times with HBS/0.05% Tween 20 buffer and protease inhibitors (1 mm PMSF, 5 μg/ml of each aprotinin and leupeptin). Analysis of experiments described was determined by densitometric quantification of nonsaturated bands of the resulting immunoblots. The linearity of density and Lp(a) concentration within the density range observed in the experiments was verified by analyzing a Western blot with serial Lp(a) dilutions (not shown). The amount of Lp(a) bound to magnetic beads alone (36% of the maximal signal obtained for FibG190–235) or the signal obtained for pCMV vector control transfections (4% of the maximal signal obtained for A18 apo(a)/Lp(a)), respectively, was subtracted. Because of some gel to gel variability, statistical analysis of data are expressed as the means ± S.E. of at least two (cell culture supernatants) and five (purified Lp(a) and human plasma) independent experiments. Relative values are expressed as percentages of the maximal binding set at 100% for calculation purposes. Cell culture supernatants adjusted to end concentration of 0.6% SDS and 1% Triton X-100 in HBS were treated with proteinase inhibitors (1 mmPMSF, 5 μg/ml of each aprotinin and leupeptin) and subsequently precleared with 100 μl of Pansorbin cells (Calbiochem) at 4 °C for 30 min. After centrifugation the supernatants were immunoprecipitated overnight with a monospecific polyclonal rabbit anti-apo(a) antibody (Behringwerke AG, Marburg, Germany) at 5 μg/ml final antibody concentration, followed by addition of 40 μl of protein A-Sepharose (Amersham Pharmacia Biotech) for the last 2 h. Immunoprecipitates were collected by centrifugation for 2 min at 10000 g at 4 °C and washed four times with 1 ml of washing buffer (0.2% SDS, 1.25% Triton X-100, 1 mm PMSF, 5 μg/ml of each aprotinin and leupeptin in HBS). The final pellet was resuspended in 15 μl of SDS-PAGE sample buffer and subjected to reducing SDS-PAGE on commercially available on 4–12% Bis-Tris- or 3–9% Tris-acetate gels (Novex, San Diego, CA). Immunoblot analysis was performed using apo(a)-specific monoclonal antibody 1A2 (33Dieplinger H. Gruber G. Krasznai K. Reschauer S. Seidel C. Burns G. Muller H.J. Csaszar A. Vogel W. Robenek H. Utermann G. J. Lipid Res. 1995; 36: 813-822Abstract Full Text PDF PubMed Google Scholar) as described (1Brunner C. Kraft H.G. Utermann G. Muller H.J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11643-11647Crossref PubMed Scopus (148) Google Scholar). In this study, we have attempted to identify novel apo(a)/Lp(a) binding proteins by the GAL4 two-hybrid interaction trap (34Chien C.T. Bartel P.L. Sternglanz R. Fields S. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 9578-9582Crossref PubMed Scopus (1237) Google Scholar) approach screening a human liver cDNA library with the unique apo(a) kringle IV type 6 as a bait. This kringle IV subtype was used for the library screening because of its low self-activation rate compared with other kringle IV subtypes (not shown). To identify cDNA clones encoding proteins that interact with apo(a) kringle IV type 6, we transformed the yeast host strain HF7c, carrying a GAL4-HIS3 selection and GAL4-β-galactosidase reporter gene with the kringle IV type 6-expression plasmid as a bait and a human liver cDNA library with the cDNA fused to the GAL4 activation domain. A total of 2.4 × 106 transformants were subjected to positive genetic growth selection on His−Leu−Trp−plates, containing 5 mm 3-amino-1,2,4-triazole. 25 colonies stained positive for β-galactosidase activity. To determine whether activation of the GAL4-dependent reporter genes reflects a specific interaction of the encoded proteins with the kringle IV type 6 bait, each cDNA clone was rescued from yeast colonies and retransformed into the same yeast strain in the absence or presence of the GAL4 kringle IV type 6 bait expression plasmid. Additionally, we transformed each of the putative ligand expression plasmids with an expression plasmid with the GAL4 DNA binding domain fused to a nonspecific protein (p53), which is not expected to interact with proteins that bind to apo(a). Four of the 25 clones showed specific interaction, activating β-galactosidase expression exclusively in the presence of the GAL4 kringle IV type 6 bait. Sequence analysis of these positive clones using primers in the pGAD10-flanking sequence revealed two cDNA sequences of so far unknown identity and two cDNA sequences showing a 100% match to fragments of the human fibrinogen β- and γ-chain, respectively. The fibrinogen β-chain clone (K6/β) extended from residues β1–310, and the fibrinogen γ-chain clone (K6/γ) extended from residues γ189–295 (Fig. 1).Figure 1Interaction of fibrinogen clones with apo(a) kringle IV type 6 bait in the two-hybrid screening. Position of the positive clones K6/γ and K6/β (arrows) obtained on screening of a human liver cDNA library are shown next to the corresponding fibrin(ogen) chain. The consensus binding domains for apo(a) kringle IV type 6 defined by the β-chain clone K6/β1 (arrow) are shaded. Also included is the amino acid sequence alignment and consensus sequence of this conserved β- and γ-chain region. Peptide sequences of FibG190–235(bold) and FibG207–235 (underlined) are marked. Lysine to alanine exchanges in the peptide sequences are indicated underneath the γ-chain sequence. Residue numbering corresponds to native human fibrinogen, in which the β-chain runs from Glnβ1 to Glnβ461 and the γ-chain runs from Tyrγ1 to Valγ411.Scissors mark the thrombin cleavage site on the fibrinogen β-chain.View Large Image Figure ViewerDownload (PPT) The matched sequences of both clones are located within the carboxyl-terminal globular domain of fibrinogen. These carboxyl-terminal sections of the fibrinogen β- and γ-chains are characterized by high sequence and structural homology (35Yee V.C. Pratt K.P. Cote H.C. Trong I.L. Chung D.W. Davie E.W. Stenkamp R.E. Teller D.C. Structure. 1997; 5: 125-138Abstract Full Text Full Text PDF PubMed Scopus (226) Google Scholar). A carboxyl-terminal fragment of K6/β containing the sequence overlapping with K6/γ was cloned in fusion with the GAL4 activation domain, and the resulting plasmid K6/β1 (Fig. 1) was assayed in yeast cells for interaction with the GAL4 DNA binding domain-apo(a) kringle IV type 6 fusion protein to localize the region of the fibrinogen β-chain clone that mediated binding to apo(a) kringle IV type 6. The positive result in the two-hybrid assay showed that the 64 carboxyl-terminal residues of fibrinogen β-chain that correspond by sequence alignment to amino acids 189–246 of fibrinogen γ-chain are sufficient for binding to apo(a) kringle IV type 6. Moreover, we performed a two-hybrid assay with the fibrinogen clones K6/β1 and K6/γ using the apo(a) kringle IV types 2, 5, 6, 7, 8, 9, and 10 as baits. Kringle subtypes 5, 7, and 10 showed a high rate of self-activation. Therefore, the interaction could not be evaluated (not shown). As expected, apo(a) kringle IV type 6 significantly interacted with the fibrinogen β- and γ-chain in both strains, despite some quantitative differences observed. However, kringle IV types 2, 8, and 9 did not interact (Fig. 2). The expression of the bait and prey proteins has been tested by analyzing the extracted yeast proteins after SDS-PAGE and Western blot by GAL4 fusion domain-specific antibody. Expressed proteins of the expected molecular weights have been detected in the corresponding yeast protein extracts (not shown). The crystal structure of the fibrinogen module encompassing residues γ144–411 (35Yee V.C. Pratt K.P. Cote H.C. Trong I.L. Chung D.W. Davie E.W. Stenkamp R.E. Teller D.C. Structure. 1997; 5: 125-138Abstract Full Text Full Text PDF PubMed Scopus (226) Google Scholar) and the crystal structure of fibrinogen fragment D (36Spraggon G. Everse S.J. Doolittle R.F. Nature. 1997; 389: 455-462Crossref PubMed Scopus (395) Google Scholar) revealed that the minimal kringle IV type 6 binding sequences of the β- and γ-chains are located in a region of similar overall structure consisting of two helices interrupted by the solvent exposed B1-loop (35Yee V.C. Pratt K.P. Cote H.C. Trong I.L. Chung D.W. Davie E.W. Stenkamp R.E. Teller D.C. Structure. 1997; 5: 125-138Abstract Full Text Full Text PDF PubMed Scopus (226) Google Scholar). We tested the biotinylated peptides comprising γ190–235 (FibG190–235) and γ207–235 (FibG207–235), respectively, as well as the corresponding lysine exchange mutants FibKA190–235 and FibKA207–235 for in vitro interaction with apo(a)/Lp(a). All four fibrinogen peptides were able to specifically pull down Lp(a) from human plasma, whereas the control peptide was not (Fig. 3). There was a reduced binding for the Lys/Ala peptide exchange mutant FibKA207–235 in relation to wt fibrinogen peptides and FibKA190–235, indicating that lysines affect binding of the shorter fibrinogen peptides (γ207–235) to Lp(a). However, and to our surprise, lysine residues appear not to be essential for this interaction. Next, employing purified Lp(a) instead of human plasma samples, again all fibrin(ogen) peptides bound Lp(a), whereas the control peptide did not (Fig. 4). Here we observed reduced binding efficiencies of the lysine exchange mutants FibKA190–235 and FibKA207–235 compared with the corresponding wt sequences FibG190–235 and FibG207–235. HepG2 cell supernatants containing distinct apo(a)/Lp(a) derivatives were obtained after transient transfection of HepG2 cells with apo(a) expression vector constructs A18 wt, ΔKV-P, and ΔKIV 8-P representing wt apo(a) with 18 kringle units and two 3′ deletions of different lenght (as outlined in Fig. 5). Equal amounts of apo(a)/Lp(a) (as measured by enzyme-linked immunosorbent assay) were used for pull-down experiments with FibG207–235 and immunoprecipitation. As a result, A18 wt Lp(a) as well as ΔKV-P Lp(a) showed a strong interaction with FibG207–235. In contrast, only marginal binding of ΔKIV 8-P apo(a) to FibG207–235was observed (Fig. 6, A andB). The three distinct forms of apo(a)/Lp(a) were expressed equally as verified by immunoprecipitation (Fig. 6 C). HepG2 cells were transfected with the apo(a) expression vector constructs A18 wt and the mutant A18-Arg (Fig. 5). The latter is A18 apo(a) with a Trp-4174 to Arg substitution in the LBS of kringle IV type 10, which renders A18-Arg Lp(a) unable to bind to lysine-Sepharose (26Ernst A. Helmhold M. Brunner C. Petho-Schramm A. Armstrong V.W. Muller H.J. J. Biol. Chem. 1995; 270: 6227-6234Abstract Full Text Full Text PDF PubMed Scopus (116) Google Scholar). Equal amounts of apo(a)/Lp(a) (as measured by enzyme-linked immunosorbent assay) from the resulting supernatants were used for pull-down experiments with FibG207–235 and the mutant peptide FibKA207–235. The influence of the LBS on the interaction with the fibrinogen peptides was further evaluated by adding the lysine analogue EACA to the binding buffer. Consistent with pull-down experiments from plasma and with purified Lp(a), pull-down of A18 wt Lp(a) from HepG2 supernatants with the mutant peptide FibKA207–235 resulted in decreased binding efficiency compared with the wt FibG207–235 peptide. Binding of FibG207–235 to A18 wt Lp(a) was significantly reduced after preventing lysine-dependent interactions by addition of EACA. Binding of both fibrinogen peptides to the A18-Arg mutant Lp(a) was greatly reduced when compared with the binding to A18 wt Lp(a). Again further reduction in binding of both peptides was observed in the presence of EACA (Fig. 7). To address the question whether the low to absent binding of FibG207–235 to ΔKIV 8-P apo(a) is due to the presence of this mutant form as free apo(a) exclusively, we transfected apoB100 negative CHO cells with the apo(a) expression vector constructs and performed pull-down experiments with equal amounts of apo(a) from the cell supernatants. We observed strong binding of A18 wt apo(a) and reduced binding of ΔKV-P apo(a) to FibG207–235 (Fig. 8,A and B). Binding to ΔKIV 8-P was below the value obtained by the control peptide. Equal expression of the three apo(a) forms was controlled by immunoprecipitation (Fig. 8 C). Moreover, we tested binding of the mutant apo(a) ΔKIV 5–8 to FibG207–235. This mutant which lacks the kringle IV types 5–8 and does not form Lp(a) particles (26Ernst A. Helmhold M. Brunner C. Petho-Schramm A. Armstrong V.W. Muller H.J. J. Biol. Chem. 1995; 270: 6227-6234Abstract Full Text Full Text P
Referência(s)