Lp(a) Particles Mold Fibrin-Binding Properties of Apo(a) in Size-Dependent Manner
A Study With Different-Length Recombinant Apo(a), Native Lp(a), and Monoclonal Antibody
Objective— Small-sized apolipoprotein(a) [apo(a)] isoforms with high antifibrinolytic activity are frequently found in cardiovascular diseases, suggesting a role for apo(a) size in atherothrombosis. To test this hypothesis, we sought to characterize the lysine (fibrin)-binding function of isolated apo(a) of variable sizes.
Methods and Results— Recombinant apo(a) [r-apo(a)] preparations consisting of 10 to 34 kringles and a monoclonal antibody that neutralizes the lysine-binding function were produced and used in parallel with lipoprotein(a) [Lp(a)] particles isolated from plasma in fibrin-binding studies. All r-apo(a) preparations displayed similar affinity and specificity for lysine residues on fibrin regardless of size (Kd 3.6±0.3 nmol/L) and inhibited the binding of plasminogen with a similar intensity (IC50 16.8±5.4 nmol/L). In contrast, native Lp(a) particles displayed fibrin affinities that were in inverse relationship with the apo(a) kringle number. Thus, a 15-kringle apo(a) separated from Lp(a) and a 34-kringle r-apo(a) displayed an affinity for fibrin that was higher than that in the corresponding particles (Kd 2.5 versus 10.5 nmol/L and Kd 3.8 versus 541 nmol/L, respectively). However, fibrin-binding specificity of the r-apo(a) preparations and the Lp(a) particles was efficiently neutralized (IC50 0.07 and 4 nmol/L) by a monoclonal antibody directed against the lysine-binding function of kringle IV-10.
Conclusions— Our data indicate that fibrin binding is an intrinsic property of apo(a) modulated by the composite structure of the Lp(a) particle.
High plasma levels of the Lp(a) are now recognized as a risk factor in cerebrovascular and cardiovascular diseases.1 However, the mechanism by which Lp(a) may favor the atherogenic and thrombogenic processes is not, as yet, clearly understood. Its composite structure, consisting of an inactive serine proteinase, apo(a), which is disulfide-linked to apoB-100 of an atherogenic LDL-like particle,2 may explain this phenomenon in part. Apo(a) is a highly glycosylated protein that shares a high degree of sequence identity (75% to 94%) with plasminogen, the precursor of the fibrinolytic enzyme plasmin.3,4⇓ Apo(a) and plasminogen are derived from a common ancestral gene and are constituted by disulfide-bridged structures of 80 to 90 amino acid residues called kringle (K) domains and a catalytic domain. Plasminogen consists of 5 different kringle domains, 2 of which, K1 and K4, bear a lysine-binding site (LBS) that displays affinity for lysine residues exposed on fibrin and cell membrane proteins. Binding and activation of plasminogen through these interactions leads to fibrinolysis and pericellular proteolysis. Apo(a) contains a copy of plasminogen-like K5 and multiple tandem repeats of a kringle that shares 61% to 75% sequence identity with K4, hereafter designated K-IV. The K-IV repeats of apo(a) differ by specific point mutations or deletions and have been classified as 10 different types.5 Apo(a) K-IV type 2 has no functional LBS, and its variable number gives rise to a series of apo(a) isoforms containing a single copy of the other 9 K-IV types.6 Sequence comparison and molecular modeling5,7⇓ have shown that an LBS similar to that of plasminogen K4 is present in K-IV type 10. A slightly modified LBS that is present in K-IV copies (type 5 to type 8) and appears to be masked in the intact Lp(a) particle may display lysine (fibrin) binding.8,9⇓ Kringle copies with a functional LBS may endow apo(a) with lysine-binding capabilities similar to those of plasminogen. Indeed, it has been recently shown that apo(a) accumulation in the vessel wall of fibrin(ogen)-deficient apo(a) transgenic mice is importantly reduced.10 Lp(a) particles containing distinct apo(a) isoforms may show variability in their lysine-binding function11–13⇓⇓ and display functional heterogeneity for fibrin14,15⇓ and cell16 binding. Furthermore, several studies have found a correlation between the prevalence of low-molecular-mass isoforms and coronary artery disease17,18⇓ or advanced stenotic atherosclerosis.19 In the present study, we have further explored the existence of such a structural/functional relationship in the interaction of apo(a) with fibrin by using recombinant apo(a) [r-apo(a)] isoforms containing a variable number of K-IV type 2 repeats and a monoclonal antibody (mAb) that neutralizes the LBS function.20 The results of the present study show that in contrast to native Lp(a) particles, all r-apo(a)s display similar affinity and specificity for lysine residues regardless of size, thus suggesting that the affinity of apo(a) for fibrin is conditioned by the composite structure of the Lp(a) particle.
Production of r-Apo(a)
The plasmids pCMV-A10, -A14, -A18, -A22, -A26, -A30, and -A34 were obtained as previously described21 and were stably transfected by electroporation into the adenovirus-transformed human embryonic kidney cell line 293.22 The culture medium containing the different r-apo(a)s, produced and harvested as described,23 was supplemented with proteinase inhibitors (20 kallikrein inhibitory units/mL aprotinin, 0.5 mmol/L AEBSF, 2 mmol/L EDTA, and 0.01% [wt/vol] NaN3, final concentrations) and stored at −80°C until use.
Purification and Amino-Terminal Sequence Analysis of r-Apo(a)
The r-apo(a)s were immunopurified on a Sepharose 4B–immobilized mAb directed against native apo(a) (mAb a3, kindly provided by L. Sorell, Center for Genetic Engineering, Havana, Cuba).24 The r-apo(a) was eluted with 20 mmol/L glycine-HCl, pH 2.0, containing 0.1 mol/L NaCl, and fractions were collected into a volume of 0.1 mmol/L Tris, pH 10, sufficient to neutralize the pH in each fraction. Fractions containing the r-apo(a) were pooled, concentrated on dried polyethylene glycol 20 000 (Serva), and dialyzed against buffer A (50 mmol/L sodium phosphate, pH 7.4, 80 mmol/L NaCl, 0.01% [wt/vol] Tween 20, and 0.01% [wt/vol] NaN3) containing 2 mmol/L EDTA. The concentration of the r-apo(a) was determined according to the method of Lowry et al,25 with purified human plasminogen used as a standard. Electrophoretic analysis (SDS/6% acrylamide or 3.75% polyacrylamide/0.8% agarose gels) of the r-apo(a) preparations was performed as previously described.15,23⇓ Proteins were electroblotted onto polyvinylidene difluoride sheets with a graphite electroblotter system (Millipore) according to the method of Kyhse-Andersen.26 Apo(a) protein bands were stained with Coomassie brilliant blue and submitted to microsequencing with the use of a microsequencer (model 473A, Applied Biosystem) equipped with a model 610A data analysis system.
Purification of Native Lp(a) From Plasma and Isolation of Apo(a)
Volunteer blood donors were selected among individuals attending the outpatient clinic and from the medical staff of Hôpital Robert-Debré (Université de Reims, Reims, France) for their high Lp(a) plasma concentration (≥0.5 g/L) and apo(a) phenotype. Venous blood was drawn from the forearm into sterile polypropylene tubes containing 4 mmol/L EDTA (final concentration). Plasma was separated by centrifugation at 2000g for 20 minutes at 4°C and supplemented with proteinase inhibitors and an antioxidant (butyl hydroxytoluene, 0.05 mg/mL). Lp(a) was purified by sequential ultracentrifugation and molecular sieving as previously described.16 Dissociation of apo(a) 15 kringles from the purified Lp(a) of a homozygous subject [plasma Lp(a), 1.2 g/L] was performed under mild reductive conditions essentially as described by Edelstein et al.27 After reduction, the solution was adjusted to a solvent density of 1.150 g/mL with KBr and centrifuged at 134 000g for 24 hours at 10°C in a 50-Ti rotor (Beckman). The bottom fraction (extensively dialyzed against buffer A) was shown to contain free apo(a) by SDS-PAGE and Western blotting and was conserved at 4°C until assay (<24 hours).
Purification and Radioiodination of Proteins
Plasminogen was purified from fresh-frozen human plasma as previously described,28 with modifications.29 The plasminogen preparation was considered to be >99% pure and was shown to be Lp(a) free, as assessed by SDS-PAGE and autoradiography of the 125I-labeled product and by amino terminal analysis. Plasminogen concentration was calculated by measuring the absorbance at 280 nm with a molecular weight of 93 000 and
A sheep antiserum directed against human apo(a) was prepared at the Institut National de la Recherche Agronomique (Center de Clermont-Ferrand-Theix) by immunizing the animal with 3 subcutaneous injections of purified r-apo(a) A10. The IgG fraction of this antiserum was separated by ammonium sulfate precipitation, ion-exchange chromatography on diethylaminoethyl-Tris-acryl, and affinity chromatography on protein A–Sepharose (Pharmacia). The purified IgG was further immunodepleted by using Sepharose-immobilized plasminogen, and its apparent affinity for the different r-apo(a) was determined.
The monoclonal IgG1 A10.2 recognizing K-IV type 10 of r-apo(a) was obtained, cultured, and purified on protein A–Sepharose as previously described.20 Concentration of the purified IgG1 was determined by the procedure of Lowry et al.25
Plasminogen and the purified anti-apo(a) IgG were radioiodinated with sodium 125I with the use of the Iodogen method of Fracker and Speck,31 an iodination time of 4 minutes at 4°C, and the removal of free 125I by molecular sieving on a PD-10 Sephadex column (Pharmacia). The specific radioactivity obtained was 9 to 11 nCi/ng plasminogen and ≈3 nCi/ng anti-apo(a) IgG.
Fibrin surfaces were prepared and immunocharacterized as previously described.29 Varying concentrations of purified Lp(a) or r-apo(a) preparations and isolated 15-kringle apo(a) in buffer A supplemented with 40 mg/mL BSA were incubated with fibrin for 2 hours at 22°C. Unbound proteins were removed by 3 washes with buffer A supplemented with 2 mg/mL BSA. The surfaces were then probed with a known concentration of the 125I-labeled polyclonal antibody directed against apo(a) in buffer A supplemented with 40 mg/mL BSA. After 3 washes with buffer A supplemented with 2 mg/mL BSA, bound radioactivity was counted in a γ-radiation counter and transformed into mass of antibody bound by using the specific radioactivity (dpm/mole) of the labeled IgG. Binding of ligands to fibrin in the presence of 0.2 mol/L of 6-amino-hexanoic acid was considered to be unrelated to interactions with lysine residues.
Inhibition of the Fibrin-Binding Function of r-Apo(a) and Lp(a) With a mAb
Experiments aimed at determining the fibrin-binding function of r-apo(a) and plasma Lp(a) were performed with mAb A10.2. This antibody is specifically directed against the LBS of kringle 4 in plasminogen and K-IV type 10 in apo(a)20 and does not recognize nonfunctional LBSs. Experiments were performed as indicated above except that the r-apo(a) was incubated with varying concentrations of mAb A10.2 or an unrelated mAb (7D4 directed against plasminogen activator inhibitor-1, kindly provided by P.J. Declerck, Catholic University of Leuven, Leuven, Belgium) for 30 minutes at room temperature before incubation with the fibrin surfaces.
For Lp(a) in plasma, the procedure was as described previously.32 Briefly, plasma was supplemented with protease inhibitors and incubated with varying concentrations of mAb A10.2. The samples were then incubated (50 μL per well) with fibrin for 1 hour at 37°C; the plates were washed and probed with a polyclonal antibody directed against apo(a). A peroxidase-labeled rabbit anti-sheep IgG was used as secondary antibody, and a 1 mg/mL solution of 2,2′-azinobis(3-ethylbenzthiazoline-sulfonic acid) was used as a substrate for color development. The absorbance was measured in a microplate counter at 405 nm. The amount of r-apo(a) or Lp(a) bound to fibrin was calculated by using the linear regression of the relationship between the concentration of a r-apo(a) reference and the change in absorbance at 405 nm (A405/min).
Competitive Inhibition of Plasminogen Binding to Fibrin
Solutions containing varying amounts (0 to 200 nmol/L) of r-apo(a) and a constant amount (≈4 nmol/L) of radiolabeled plasminogen were prepared in buffer A supplemented with 40 mg/mL BSA and incubated for 2 hours at 22°C with the fibrin surface (50 μL per well). In control experiments, similar amounts of LDL were added to plasminogen and incubated with fibrin. After elimination of the supernatant, the plate was washed, and the radioactivity bound to the surface was counted in a γ-radiation counter. In the absence of competitors, this amount was considered to be 100% plasminogen bound.
Analysis of Binding Data
where S · X is the equilibrium fraction of r-apo(a) bound to the surfaces, S0 is the total amount of r-apo(a) binding sites, K is the association constant of the ligand/surface interaction, and X is the total input of ligand. By relating the number of occupied sites [S · X=bound anti-apo(a) antibody] to ligand concentration [X=r-apo(a)], Equation 1 allows calculation of the maximum bound, Bmax=S0, and the dissociation constant, Kd=K−1. The amount of r-apo(a) bound to cells or fibrin was expressed by the mass of antibody bound by using an algebraic expression from Equation 1 that describes the linear relationship of antibody binding at low antibody concentrations with the amount of Lp(a) bound to fibrin34:
Production, Isolation, and Amino-Terminal Characterization of r-Apo(a) Isoforms
Stable transfected embryonic human kidney cells, line 293, were committed to express r-apo(a) isoforms differing by their variable number of K-IV type 2 repeats (from 0 to 24). All other K-IV types (K-IV type 1 and K-IV type 3 to K-IV type 10) and the serine-protease region were encoded by a common cDNA expression plasmid fragment and were present in single identical copies in all r-apo(a) isoforms (Figure 1). Conditions chosen for development of the transfected cells allowed for the harvesting of culture medium containing high concentrations of intact unmodified r-apo(a) as detected by immunoblot analysis of agarose-acrylamide gels (Figure 2A). The immunopurified r-apo(a) migrated as single protein bands in Coomassie blue–stained 6% acrylamide gels (Figure 2B). The sequence of 10 N-terminal amino acid residues of the purified r-apo(a) was similar to the deduced N-terminal sequence of human apo(a).3,4⇓
Variable Size Apo(a) Recombinants, but Not Lp(a)s, Bind to Fibrin With a Similar Affinity
The use of solid-phase fibrin as a surface acceptor for r-apo(a), native Lp(a), and plasminogen has been previously demonstrated.15,29,35⇓⇓ Binding isotherms were obtained by incubation of r-apo(a) and native apo(a) or Lp(a) with fibrin surfaces. Binding to these surfaces was quantified with a 125I-labeled polyclonal IgG raised against r-apo(a) A10. The affinity of the purified IgG for the different r-apo(a)s was in the same order of magnitude (Kd 47±15 nmol/L) as measured by using surface plasmon resonance (data not shown). Therefore, the amount of 125I-labeled IgG bound was assumed to be proportional to the amount of apo(a) bound according to Equation 2. All the isoforms tested (A10 to A34) showed dose dependence, saturation, and specificity for carboxy-terminal lysine residues of fibrin as indicated by its inhibition with 6-amino-hexanoic acid. Fitting of the data obeys the simple Langmuir equation (Equation 1) for adsorption at interfaces,33 as indicated by typical binding isotherms for isolated r-apo(a) A30 and the corresponding Lp(a) reported in Figure 3. The calculated dissociation constant (Kd) values for the different r-apo(a)s were similar (3.6±0.3 nmol/L), regardless of their size (Table). In contrast, a clear inverse relationship between apo(a) size and affinity for fibrin could be documented for Lp(a) particles, in agreement with previous published data.15,34⇓ These results suggest that the LBS function of apo(a) is modified by the composite structure of the Lp(a) particle, as previously suggested by Klezovitch et al.8 Indeed, the Kd (2.5 nmol/L) of a 15-kringle apo(a) separated from native Lp(a) was similar to the mean Kd (3.4 nmol/L) of the closest size r-apo(a) A14, thus indicating that the mild reductive conditions we used to separate apo(a) from native Lp(a) did not disturb its lysine-binding function. In contrast, the corresponding native Lp(a) particle displayed a 4-fold higher Kd (10.5 nmol/L), indicative of lower affinity. The highest differences in affinity were found between r-apo(a) A30 (Kd 3.6 nmol/L) and r-apo(a) A34 (Kd 2.1 nmol/L) and their corresponding native Lp(a) particles (Table).
All r-apo(a)s inhibited the binding of plasminogen to fibrin with a similar intensity (Figure 4). The amount of plasminogen bound to fibrin decreased as a function of the input concentration of r-apo(a) that was added, and it attained 50% (IC50) at 16.8±5.4 nmol/L These results are indicative of competitive inhibition between plasminogen and r-apo(a) for the same type of binding sites. At identical molar concentrations, the size of the different r-apo(a) preparations had no influence on the competitive binding.
Lp(a) and r-Apo(a) Fibrin-Binding Diversity Is Equally Neutralized by LBS-Blocking mAb
Binding of the different-sized r-apo(a) preparations and of native plasma Lp(a) particles was efficiently inhibited by a mAb directed against the LBS of K-IV type 10.20 Inhibition of the binding was dose dependent (Figure 5), and an IC50 of 0.074 nmol/L was calculated for r-apo(a) from data displayed in the inset to Figure 5. The IC50 values for plasma, shown in Figure 5, were 3.9 nmol/L for the inhibition of a plasma Lp(a) containing a 10-kringle apo(a) isoform and 4.2 nmol/L for an Lp(a) containing 19- and 28-kringle apo(a) isoforms. These IC50 values were higher than those obtained with the isolated r-apo(a) and most probably reflect competition/consumption by kringle 4 of plasma plasminogen, which is also recognized by mAb A10.2.
Definition of a threshold risk value for Lp(a) is complicated by variations in the distribution of Lp(a) among different populations.36 However, most studies have accepted an Lp(a) concentration >30 mg/dL as a major independent risk factor for atherosclerosis.1 The circulating concentration of Lp(a) is mainly regulated by the apo(a) gene, which varies in size as a function of the number of repetitive sequences encoding K-IV type 2. The smaller this hypervariable region, the higher the plasma concentration of Lp(a) will be. Thus, short apo(a) alleles may favor atherogenesis by increasing the concentration of Lp(a). In agreement with this known inverse relationship between the size of apo(a) isoforms and plasma Lp(a) concentration, a series of studies have shown a prevalence of small-sized apo(a) isoforms in coronary heart disease patients17,18,37-40⇓⇓⇓⇓⇓ or in those with advanced stenotic atherosclerosis.19 However, it remains to be determined whether short apo(a) alleles are, per se, an atherosclerotic risk factor.41 The possibility that apo(a) phenotypes may have different functional properties regarding fibrin and cell binding has been explored; an apo(a) size–related variability in affinity has previously been reported with the use of native Lp(a) particles.14,16,34⇓⇓ Because the main structural difference between apo(a) isoforms almost certainly reflects a variable number of K-IV type 2 repeats, the present study was aimed at determining their effect on fibrin binding. We sought to explore this phenomenon by comparing binding to fibrin of different-sized r-apo(a) and Lp(a) particles. The r-apo(a)s differed solely by the number of copies of K-IV type 2 and were similar regarding lysine-binding kringle structures. Binding of the purified r-apo(a) and Lp(a) particles to fibrin inhibited plasminogen binding and was specific for lysine residues. Of note, LDL particles were shown to be unable to modify plasminogen binding to fibrin. Furthermore, binding to fibrin of r-apo(a) and Lp(a) in plasma was neutralized with a mAb that specifically recognizes the fibrin-binding function of apo(a) K-IV type 10.20 It has been reported that regions constituting K-IV type 5 to K-IV type 88 and K-V-protease42 may also be involved in fibrin binding. These regions are present in all the r-apo(a) that we have tested, and it is possible that mAb A10.2 may also neutralize the corresponding LBS.
The dissociation constant of the different r-apo(a) for fibrin was in the same order of magnitude (3.6±0.3 nmol/L) as the value previously determined with a radiolabeled r-apo(a) containing 17 K-IV copies.35 These values are ≈150- to 250-fold lower than the Kd value that we have determined for plasminogen in previous reports (0.5 to 1 μmol/L)16,29⇓ and in the present study (0.9 μmol/L). These differences in Kd values explain the ability of r-apo(a) to compete efficiently with plasminogen for binding to fibrin. Most interesting was the finding of quantitative differences in affinity between the r-apo(a) and Lp(a) preparations tested. The functional heterogeneity of native Lp(a) particles is in contrast to the homogeneity in Kd values found for all the r-apo(a)s tested and for the isolated native 15-kringle apo(a). Thus, the purified Lp(a) particles displayed lower affinities for fibrin than did the corresponding r-apo(a)s. Of greater interest was the observation that the change in affinity for the 15-kringle apo(a) isolated from native Lp(a) was moderate (4-fold) compared with the important difference in affinity (85-fold) observed between the r-apo(a) A30 and the corresponding Lp(a) particle. The study of other apo(a) phenotypes derived from reduced Lp(a) particles is hampered by the risk of severely affecting the LBS structure and function by sulfhydryl groups.43
Our data suggest that the fibrin affinity of small-sized apo(a) isoforms is preserved in Lp(a), whereas the affinity for fibrin of high-molecular-mass isoforms (>22 kringles) is reduced in a size-dependent manner in the Lp(a) particle. We hypothesize, in agreement with Klezovitch et al,8 that interactions of apo(a) kringles with components of the Lp(a) particle may contribute to the observed changes in fibrin affinity. Indeed, Ernst et al9 have shown that binding to lysine-Sepharose of apo(a) incorporated into an LDL particle was strongly decreased. These modifications in affinity may be related to conformational changes. As a matter of fact, the extended shape of free r-apo(a)44 in solution is in contrast to the natural roughly spherical shape of Lp(a), where apo(a) is closely associated with the surface, as recently visualized by Weisel et al.45 On the other hand, homocysteine has been shown to induce a concentration-dependent increase in its affinity for fibrin46 and a concomitant release of free apo(a) from reduced Lp(a). According to our present results, the free apo(a) released by homocysteine may contribute to the observed increase in the fibrin affinity of Lp(a) through the “multiple binding with identical linkage mechanism.”34
Taken altogether, the results presented in the present study demonstrate that in contrast to native Lp(a) particles, all r-apo(a)s tested display a size-independent similar affinity and specificity for lysine residues on fibrin. Therefore, affinity for fibrin appears to be an intrinsic property of apo(a) that is not dependent on kringle number, except for apo(a) isoforms constitutive of an Lp(a) particle.
This work was supported by grant ACC-SV9 from the Ministère de la Recherche (France) and grant Adrienne et Pierre Sommer from the Fondation de France. We are grateful to Dr H.-J. Müller (Roche, Penzberg) for hosting C.K. in his laboratory during preparation of recombinant apo(a) plasmids made available to us.
↵*These authors contributed equally to the present study.
Received March 6, 2002; revision accepted April 1, 2002.
- ↵Danesh J, Collins R, Peto R. Lipoprotein(a) and coronary heart disease. Circulation. 2000; 102: 1082–1085.
- ↵Eaton DL, Fless GM, Kohr WJ, Xu QT, Miller CG, Lawn RM, Scanu AM. Partial amino acid sequence of apolipoprotein(a) shows that it homologous to plasminogen. Pro Natl Acad Sci U S A. 1987; 84: 3224–3228.
- ↵Lackner C, Cohen JC, Hobbs HH. Molecular definition of the extreme size polymorphism in apolipoprotein(a). Hum Mol Genet. 1993; 2: 933–940.
- ↵Guevara J, Jan YJ, Knapp R, Tulinsky A, Morrisett JD. Comparison of ligand-binding sites of modeled apo(a) kringle-like sequences in human apolipoprotein(a). Arterioscler Thromb. 1993; 13: 758–770.
- ↵Ernst A, Helmhold M, Brunner C, Pethö-Schramm A, Armstrong VW, Müller HJ. Identification of two functionally distinct lysine-binding sites in kringle 37 and in kringles 32–36 of human apolipoprotein(a). J Biol Chem. 1995; 270: 6227–6234.
- ↵Lou XJ, Boonmark NW, Horrigan FT, Degen JL, Lawn RM. Fibrinogen deficiency reduces vascular accumulation of apolipoprotein(a) and development of atherosclerosis in apolipoprotein(a) transgenic mice. Proc Natl Acad Sci U S A. 1998; 95: 12591–12595.
- ↵Armstrong VW, Harrach NB, Robenek NH, Helmhold NM, Walli NA, Seidel D. Heterogeneity of human lipoprotein(a) Lp(a): cytochemical and biochemical studies on the interaction of two Lp(a) species with the LDL receptor. J Lipid Res. 1990; 31: 429–441.
- ↵Fless GM, Snyder ML. Polymorphic forms of Lp(a) with different structural and functional properties: cold-induced self-association and binding to fibrin and lysine-Sepharose. Chem Phys Lipids. 1994; 67/68: 69–79.
- ↵Hervio L, Chapman J, Thillet J, Loyau S, Anglés-Cano E. Does apolipoprotein(a) heterogeneity influence lipoprotein(a) effects on fibrinolysis? Blood. 1993; 82: 392–397.
- ↵Kang C, Durlach V, Soulat T, Fournier C, Anglés-Cano E. Lipoprotein(a) isoforms display differences in affinity for plasminogen-like binding to human mononuclear cells. Arterioscler Thromb Vasc Biol. 1997; 17: 2036–2043.
- ↵Sandholzer C, Boerwinkle E, Saha N, Tong MC, Utermann G. Apolipoprotein(a) phenotypes, Lp(a) concentration and plasma lipid levels in relation to coronary heart disease in a Chinese population: evidence for the role of apo(a) gene in coronary heart disease. J Clin Invest. 1992; 89: 1040–1046.
- ↵Kronenberg F, Kronenberg MF, Kiechl S, Trenkwalder E, Santer P, Oberhollenzer F, Egger G, Utermann G, Willeit J. Role of lipoprotein(a) and apolipoprotein(a) phenotype in atherogenesis: prospective results from the Bruneck study. Circulation. 1999; 100: 1154–1160.
- ↵Brunner C, Lobentanz EM, Pethö-Schramm A, Ernst A, Kang C, Dieplinger H, Müller HJ, Utermann G. The number of identical kringle IV repeats in apolipoprotein(a) affects its processing and secretion by HepG2 cells. J Biol Chem. 1996; 271: 32403–32410.
- ↵Graham FL, Smiley J, Russell WC, Nairn R. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Virol. 1977; 36: 59–77.
- ↵Anglés-Cano E, Loyau S, Cardoso-Saldana G, Couderc R, Gillery P. A novel kringle-4 number based recombinant apo[a] standard for human apo[a] phenotyping. J Lipid Res. 1999; 40: 354–359.
- ↵Lowry OH, Rosenbrough NJ, Farr AR, Randall R. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951; 193: 265–275.
- ↵Deutsch D, Mertz ET. Plasminogen purification from human plasma by affinity chromatography. Science. 1970; 170: 1095–1096.
- ↵Soulat T, Loyau S, Baudouin V, Maisonneuve L, Hurtaud Roux MF, Schlegel N, Loirat C, Anglés-Cano E. Effect of individual plasma lipoprotein(a) variations in vivo on its competition with plasminogen for fibrin and cell binding: an in vitro study using plasma from children with idiopathic nephrotic syndrome. Arterioscler Thromb Vasc Biol. 2000; 20: 575–584.
- ↵Adamson AW. Physical Chemistry of Surfaces. 5th ed. New York, NY: Wiley-Interscience Publication; 1992.
- ↵Cobbaert C, Kesteloot H. Serum lipoprotein(a) levels in racially different populations. Am J Epidemiol. 1992; 136: 441–449.
- ↵Sandholzer C, Saha N, Kark JD, Rees A, Jaross W, Dieplinger H, Hoppichler F, Boerwinkle E, Utermann G. Apo(a) isoforms predict risk for coronary heart disease: a study in six populations. Arterioscler Thromb. 1992; 12: 1214–1226.
- ↵Wild SH, Fortmann SP, Marcovina SM. A prospective case-control study of lipoprotein(a) levels and apo(a) size and risk of coronary heart disease in Standford Five-City Project participants. Arterioscler Thromb Vasc Biol. 1997; 17: 239–245.
- ↵Kraft HG, Lingenhel A, Köchl S, Hoppichler F, Kronenberg F, Abe A, Mühlberger V, Schönitzer D, Utermann G. Apolipoprotein(a) kringle IV repeat number predicts risk for coronary heart disease. Arterioscler Thromb Vasc Biol. 1996; 16: 713–719.
- ↵Marcovina SM, Koschinsky ML. Lipoprotein(a) concentration and apolipoprotein(a) size: a synergistic role in advanced atherosclerosis? Circulation. 1999; 100: 1151–1153.
- ↵Harpel PC, Chang VT, Borth W. Homocysteine and other sulfhydryl compounds enhance the binding of lipoprotein(a) to fibrin: a potential biochemical link between thrombosis, atherosclerosis, and sulfhydryl compound metabolism. Proc Natl Acad Sci U S A. 1992; 89: 10193–10197.