Lipoprotein(a), Fibrin Binding, andPlasminogen Activation

Joseph Loscalzo, Mark Weinfeld, Gunther M. Fless, and Angelo M. Scanu

Upoproteln(a) (Lp[a]> Is a complex plasma llpoproteln In which apollpoprotein(apo) B-100 Is covalerrtty linked by a dlsulfide bridge to a unique apollpoprotein,apo(a). The cDNA of apo(a) has recently been Isolated and sequenced, and aremarkable homology to human plasminogen has been noted. In this report, wedemonstrate that, like plasminogen, Lp(a) binds to fibrin. In addition, Lp(a) competeswtth plasminogen and tissue-type plasminogen activator for fibrin binding. As afunctional consequence of these binding properties, we show that Lp(a) attenuatesthe fibrin-dependent enhancement of tissue-type plasminogen activator activityagainst the native substrate, and does so as an uncompetrUve Inhibitor (K,=15 nM).Finally, we show that In a plasma milieu, Lp(a) attenuates clot lysis Induced bytissue-type plasminogen activator. None of these effects was noted with low densityllpoproteln free of apo(a). These data suggest that Lp(a) influences the flbrinolyticsystem and probably does so by virtue of the fibrin binding properties conferred bythe krtngle repeats of apo(a). (Arteriosclerosis 10:240-245, March/April 1990)

Lipoprotein(a) (Lp[a]) is a plasma lipoprotein first de-scribed by Berg.1 When present in high concentra-

tions in plasma, Lp(a) correlates strongly with an increasedrisk for coronary artery disease.2-5 Lp(a) is comprised oftow density lipoprotein in which apoKpoprotein (apo)B-100 is covalentty linked through a disutfide bridge to aunique apolipoprotefn, apo(a). While much information onthe structure of Lp(a) has been obtained6'7-8 and the cDNAsequence of apo(a) has been determined,9 the actualfunction of Lp(a) and the mechanism of its atherogenlcityremain to be defined.

The primary sequence of apo(a) has a striking similar-ity to ptasminogen9; it contains: a serine protease domainthat is 94% homologous with that of plasminogen, onecopy of the kringte-5 region, and 37 copies of thekringte-4 domain. The structural homology to the serineprotease active site of plasminogen notwithstanding, nolatent enzymatic activity can be generated because of acrucial substitution of serine for arginine at the homolo-gous activation site domain.9 In the coagulation andfibrinotytic molecules in which they were first described,the kringle domains were identified as lysine-dependentfibrin binding regions. Plasminogen contains five kringlesof which the first has the greatest binding affinity,10 the

From the Department of Medicine, Harvard Medical Schooland Bfigham and Women's Hospital, Boston, Massachusetts,and the Departments of Medicine and Biochemistry and Molec-ular Biology, the Prttzker School of Medicine, the University ofChicago, Chicago, Illinois.

This study was supported In part by NIH Grants HL4O411,HL18577, and HL43344. Joseph Loscalzo Is the recipient of anNIH Research Career Development Award HL02273.

This work was presented in part at the American Society ofHematotogy meetings In San Antonio, Texas, December, 1988.

Address for reprints: Dr. Joseph Loscalzo, Department ofMedicine, Harvard Medical School and Brigham and Women'sHospital, Boston, MA 02115.

Received July 13,1989; revision accepted October 20,1989.

second and third provide weaker affinity sites, and thefourth contains a site of intermediate affinity.11-12-13

There has been much speculation in the literatureabout the possibility that Lp(a) can Interfere with thefibrinotytic system because of the structural similaritieswith plasminogen914; however, to date little has beenpublished to support this hypothesis.16 In this report, wedemonstrate that Lp(a) binds to fibrin, competes withboth plasminogen and tissue-type plasminogen activator(t-PA) for fibrin binding sites, and attenuates the fibrin-dependent enhancement of the ptasminogen activatoractivity of t-PA in buffer and plasma.

MethodsMaterials

Human fibrinogen and S-2251 were purchased fromKabi Vrtrum, Stockholm, Sweden. Matrex Pel 102 beadswere obtained from Amicon, Danvers, MA. Plasminogen-free bovine thrombin was purchased from Miles Pharma-ceuticals, Naperville, IL Iodine monochloride was ob-tained from Kodak Chemical, Rochester, NY. t-PA wasobtained from Genentech, South San Francisco, CA.L-Glycyl-L-prolyl-L-arginyl-L-proline (GPRP), epsilon-aminocaprolc acid, aprotinin, and tranexamic acid wereobtained from Sigma Chemical, S i Louis, MO. Lysine-Sepharose and Sephadex G-25 were purchased fromPharmacia Fine Chemicals, Uppsala, Sweden, lodo-beads were obtained from Pierce Chemical, Rockford, II.Na 125I and Na 131I were obtained from Amersham, Arling-ton Heights, II. All other chemicals were reagent grade orbetter. Deionized water was used throughout

Isolation of Low Density Llpoproteln and Lp(a)Low density lipoprotein (LDL) was prepared from the

plasma of fasting normolipidemic volunteers by sequen-tial uttracentrifugation as previously described.16 Poly-

240

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from

UPOPROTEIN(a) AND PLASMINOGEN ACTIVATION Loscalzo et al. 241

acrylamide gel electrophoresis was performed to ensurepurity of the LDL preparation.

l_p(a) was prepared from blood drawn into sterilebottles that were immersed in wet ice and contained afinal concentration of 0.15% ethylenediaminetetraaceticacid (EDTA), 0.01% NaN3, and 0.4 /*M soybean trypsininhibitor. Plasma was separated immediately by lowspeed centrifugation at 4°C, and diisopropytfluorophos-phate was added to a final concentration of 1 mM tominimize proteorysis. Total lipoproteins were then pre-pared by adjusting the plasma density to 1.21 g/ml wfthsolid NaBr and centrtfuging the sample in a 60 Tl rotor at59 000 rpm for 20 hours at 15°C. Lp(a) was Isolated fromthe total lipoprotein fraction by using a combination ofrate zonal and density gradient ultracentrifugation aspreviously described.17 Lp(a) preparations were checkedfor purity by sodium dodecyl sutfate polyacrylamidegradient-gel electrophoresis. If necessary, further purifi-cation was conducted by high performance liquid chro-matography (HPLC)-ion exchange chromatography byusing a mono-Q column (Pharmacia, Uppsala,Sweden).18 The sample load varied from 1 to 10 mg.Lp(a) was eluted with a 0 to 1 M NaCI gradient in 0.01 MTris buffer (pH 7.4) at a flow rate of 1 ml/min at 8°Cperformed over 40 minutes. Lp(a) eluted at 0.41 M NaCI.The purity of isolated Lp(a) was again checked electro-phoretically as described above.

The Lp(a) preparation used in the experiments pre-sented here has two apo(a) subunits, each having amolecular weight of 280 000 daltons, of which 200 000daltons represents protein and 80 000 daltons representscarbohydrate. Since the molecular weight of apo B de-void of carbohydrate is 514 000 daltons, the Mr of thewhole protein moiety of Lp(a) is 914 000 dattons. Thesubstructure of apo(a) can be characterized from thecONA sequence; from this analysis, each apo(a) subunitof Lp(a) contains one protease domain (M,<»24 800), onekringle-5 domain (M^11 300), and 13 kringle-4 domains(M,«12 600 each).

Plasmlnogen Preparation

Glu-plasminogen was purified from freshly obtainedplasma or fresh frozen plasma thawed at 37°C using amodification of the method of Deutsch and Mertz18 wtthslight modification. Plasma was passed over a lysine-Sepharose column, and the column was washed wtth0.3 M sodium phosphate, pH 7.4, 3 mM EDTA and250 U/ml aprotJnin. Plasminogen was eluted from thecolumn with 0.2 M of epsilon-aminocaproic acid, 3 mM ofEDTA (pH 7.4), and 250 U/ml of aprotinin. The plasmin-ogen obtained from the donors was free of contaminantLp(a) and was dlalyzed before use against 10 mM sodiumphosphate, pH 7.4, 0.15 M NaCI.

Radlolodlnatlon

Glu-plasminogen, t-PA, and fibrinogen were radiolodi-nated using lodo-beads. One lodo-bead was pre-incubated with 0.5 to 1.0 mCi Na 12SI for 10 minutes at25°C, after which 1 ml of 0.1 mg/ml glu-plasminogen,fibrinogen, or t-PA was added. The incubation was al-lowed to proceed for 12 to 15 minutes with gentle

rocking, after which the solution was removed from thelodo-bead to stop the iodination reaction and was passedover a Sephadex G-25 column that had been pre-equilibrated with 1 ml of 5 mg/ml bovine serum albuminIn 10 mM Tris (pH 7.8) 0.15 M NaCI. Twelve 0.3-mlfractions were collected and assayed for total and25%-trichloroacetic acid-precipitable radioactivity. Rou-tinely, column fractions five through eight containedmaximal protein-bound counts with a specific activity ofapproximately 0.1 to 0.2 mCi/mg. These fractions wereeither used immediately or stored for up to 2 weeks at4°C without any appreciable loss of protein-boundcounts. Importantly, radioiodination of t-PA by the gentlemethod outlined here had little effect on its activity as aplasminogen activator (specific activity was 92% of native,uniodinated control); in addition, its fibrin binding prop-erties were unaffected by iodination, as indicated byequivalent competition between either uniodinatedt-PA and 126l-t-PA or 131l-t-PA and 126l-t-PA for binding tofibrin monomer on potyacrylonitrile beads (see BindingAssays below).

l_p(a) and LDL were radioiodinated by the iodinemonochloride method of McFariane20 with modifica-tions.21'22 Lp(a) and LDL-bound radioiodine was 97%precipitable with trichloroacetic acid; Lp(a) had a specificactivity of 600 to 900 cpm/ng protein, while LDL had aspecific activity of 300 to 800 cpm/ng protein.

Soluble Fibrin Monomer and MatrexBead Preparations

Soluble fibrin monomer (SFM) and fibrin monomer-immobilized Matrex beads (FM-Matrex) were prepared asdescribed previously.23

Enzymatic Activity AssaysThe t-PA activity was assayed using the native substrate

glu-plasminogen in which the plasmln-specific substrateS-2251 was used to follow the reaction. The substratehydrolysis was measured spectrophotometricaHy with aGilford Response UV/Vis spectrophotometer (Ciba-Corning, Oberiin, OH). Activity was measured at 37°C in10 mM Tris (pH 7.8) 0.15 M NaCI. The change in opticaldensity was followed for 5 minutes, and the initial reactionvelocity was determined from plots of change inabsorbance/time versus time, as described previously.24

Determination of kinetic inhibition constants was per-formed by measuring the initial reaction velocities as de-scribed above in the absence or presence of Lp(a)(53 nM, 99 nM, and 165 nM) over a range of plasminogenconcentrations (0 to 3.4 fiM). SFM (56 nM), S-2251(0.8 mM), and t-PA (48 nM) were also included in thereaction solution. Reactions were carried out in 10 mM Tris(pH 7.4), 0.15 M NaCI at STX, and the reaction wasfollowed by monitoring the change in absorbance at 405 nm.Numerical analysis was performed as described by Dixon.28

Binding Assays

The binding of Lp(a) or the competitive binding of Lp(a)with glu-plasminogen or t-PA to fibrin monomer wasmeasured using FM-Matrex prepared as describedabove. In direct binding assays, increasing concentra-

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from

242 ARTERIOSCLEROSIS VOL 10, No 2, MARCH/APRIL 1990

tions of Lp(a) (0 to 140 nM) were incubated with 2X106

beads (on the surface of which was 4x10* fibrinmonomers/bead) in a 0.3-ml total volume in 10 mMsodium phosphate (pH 7.4) 0.15 M NaCI for 30 minutes at22°C. (Prior time course experiments showed that maxi-mal binding was achieved by this incubation time.) At theend of the incubation period, 0.1 ml of the incubationsuspension was washed three times in ice-cold bufferwith centrifugation at 8700g for 30 seconds, after whichthe tip of the centrifuge tube was excised, and theradioactivity was counted to measure total binding. Non-specific binding was determined by adding a 20-foldexcess of unlabeled Lp(a) at each concentration exam-ined or 20 MM tranexamic acid. Nonspecific bindingaccounted for approximately 20% of the total bindingobserved for Lp(a). Competitive binding assays wereperformed by incubating FM-Matrex with fixed concentra-tions of radioiodinated t-PA or glu-plasminogen (at2.5xtheir estimated, apparent KoS) and increasing con-centrations of Lp(a). The amount of radioiodinated ligandbound at any concentration of Lp(a) was compared tothat bound in the absence of Lp(a) and plotted as a ratio(B/Bo) versus the total Lp(a) added.

Clot Lysis Assay

The effect of Lp(a) on lysis of preformed clots wasanalyzed by mixing known amounts of purified Lp(a) withplasma that was free of detectable endogenous Lp(a)and contained radioiodinated fibrinogen. Bovine throm-bin (0.1 U/ml) was added to this plasma to induce fibrindot formation, and the plasma was incubated at 22°C for2 hours, after which the thrombin was inhibited with1 U/ml of hirudin. The formed, radiolabeled clots werewashed twice with 10 mM of sodium phosphate (pH 7.4)0.15 M NaCI by centrifugation. The final, washed clot wasallowed to remain firmly adherent to the bottom of thecentrifuge tube, the supernatant was removed, and freshplasma was added, to which the same concentration ofpurified Lp(a) was added as in the initial plasma fromwhich the ctot was formed. Total radioactivity in theformed clot was measured, and t-PA was added to finalconcentrations of 13, 26, and 50 nM, after which theassay tube was placed on a rocker at room temperaturefor 3 hours. Aliquots of the supernatant of the suspensionwere removed at regular intervals for the determination ofextent of clot lysis.

Polyacryiamlde Gel Electrophorosls

Sodium dodecyl sutfate-potyacrylamide gel electrc-phoresis was performed as described by Weber andOsborne26 and modified by Laemmli.27 The gels werestained with Coomassie brilliant blue in 50% methanoland 5% acetic acid and were destained by diffusion. Themolecular weight standards were processed similarly,and the apparent molecular weights (Mr) were estimatedby interpolation.

Protein Determinations

Protein concentrations were determined by themethod of Lowry and colleagues.28

oB

50 100Lp(a) Bound (x 10 1 5 )

150

Figure 1. A. Binding of Lp(a) to fibrin monomer. Increasingconcentrations of purified 12*1-Lp(a) were incubated with 2x10°FM-Matrex beads, on the surface of each of which were bound4x10° molecules of fibrin monomer, for 30 minutes at 25°C in10 mM of sodium phosphate (pH 7.4) 0.15 M NaCI. At the end ofthe Incubation period, 0.1 ml of the 0.3-ml total assay volumewas washed three times with Ice-cold incubation buffer withcentrtfugatlon at 8700g, after which the tip of the centrifuge tubewas amputated and radioactivity wets measured. Nonspecificbinding (A) determined by adding either a 20-fold excess ofunlabeled Lp(a) at each concentration of Lp(a) used or 20 nMtranexamic acid accounted for 24% of the total binding (•)observed. Specific binding (O) was determined by subtractingthe nonspecific binding from the total binding at each concen-tration of Lp(a). Each point represents the results of three to fiveexperiments each performed in duplicate. B. Scatchard analysisof the data In A.

ResultsBinding of Lp(a) to Fibrin

Purified Lp(a) bound to fibrin monomer covaientlylinked to an Insoluble matrix (Matrex Pel 102), FM-Matrex(Figure 1). Binding was saturable and specific, with anestimated apparent KQ of 25 nM, and could best bedescribed as mediated by a single class of noncoopera-tive sites (Figure 1B). Binding was reversible with anexcess of unlabeled Lp(a), with tranexamic acid (20 /iM), orwith soluble fibrin (33 £tg/ml). Nonspecific binding (opentriangles) accounted for 24% of the total binding (closedcirctes) at saturation. Time course studies showed that by20 minutes into the incubation, maximal binding occurred.With approximately 4x10* fibrin monomers/bead and106 Matrex beads/assay, 96 ng of Lp(a) protein bound atsaturation representing 5.1 x1010 molecules of Lp(a), (as-suming a molecular weight of the total protein moiety of Lp(a)

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from

LIPOPROTEIN(a) AND PLASMINOGEN ACTIVATION Loscalzo et al. 243

om

10 20 30[Lp(o) or LDL] (nM)

40 50

Rgure2. Competitive binding d Lp(a) with plasminogen ex tissue-type plasminogen activator (t-PA) to fibrin monomer. Increasingconcentrations of Lp(a) were Incubated with 78 nM 125l-glu-ptasmfnogen (o) or 92 nM 1251-t-PA (A) and 2x10° FM-Matrexbeads for 30 minutes at 25°C in 10 mM of sodium phosphate(pH 7.4) 0.15 M NaCI. Assay points were processed as de-scribed in the legend to Figure 1, and the residual binding ofeither ligand Is expressed as a fraction of the total ligand boundin the absence of Lp(a) (B/Bo). Purified LDL did not displace anysignificant amounts of plasminogen (•) or t-PA (A) from FM-Matrexover the same range of molar concentrations as Lp(a).

of 914 000 daltons) per 4 x 1011 molecules of fibrin monomer.LDL did not bind to FM-Matrex when tested over a similarrange of concentrations.

Competitive Binding of Lp(a) with Plasminogenor t-PA to Fibrin

Increasing concentrations of Lp(a) were incubated with92 nM t-PA or with 78 nM glu-plasminogen and FM-Matrex(Figure 2). Lp(a) effectively competed with plasminogen(open circles), displacing more than 85% of the totalbound plasminogen at the highest concentrations exam-ined. The estimated apparent IC50 was approximately10 nM. Lp(a) was less effective in competing with t-PA(open triangles) than plasminogen. Complete displace-ment was not achieved over the range of concentrations oft-PA tested with FM-Matrex; however, approximately 40%competitive binding was noted at the highest concentra-tion tested. LDL did not displace plasminogen or t-PA fromFM-Matrex over a similar range of concentrations as Lp(a)(closed circles and closed triangles, respectively).

Effect of Lp(a) on t-PA PlasminogenActivator Activity

The effect of Lp(a) on t-PA activity against the nativesubstrate glu-plasminogen was next examined (Figure 3). Inthese experiments, 46 nM of t-PA was incubated with a rangeof concentrations of glu-plasminogen (0 to 3.4 fM) and0.8 mM S-2251 in the presence of 3.3 fig/m\ SFM in 10 mMTris, pH 7.4,0.15 M NaCI. The reactions were then monitoredover a range of concentrations (0 to 165 nM) of Lp(a). Initialrates (derived from plots of change in absorbance/t versus t)were determined in this coupled assay24 and were plottedinversely against the reciprocal of plasminogen concentra-tion (Figure 3A). These double reciprocal plots effectivelydefine a series of parallel lines as a function of increasingLp(a) concentration. A plot of 1 /K^ against the Lp(a)

165 nM Lp(o99 nM Lp(o

• 53 nM Lp(o)O WITHOUT Lp(o)

-17 -13 - 9 - 5 - 1

-20

B

60 100 140 180

[Lp(Q)] (nM)

Figure 3. The effect of Lp(a) on plasminogen activator (t-PA)activity enhanced by soluble fibrin monomer (SFM). A. In thesecoupled assays, 48 nM t-PA was incubated with 0.8 mM S-2251,56 nM SFM, and a range of concentrations of plasminogen (0 to3.4 fiM) in 10 mM Tris (pH 7.4) 0.15 M NaCI, at 37°C. Thereactions were conducted In the absence (O) or presence ofLp(a) at concentrations of 53 nM (A), 99 nM (A), or 165 nM (•)at 37°C for 5 minutes. The reactions were monitored at 405 nm, andthe initial velocities were obtained from plots of absorbance/t versus t.24 B. The x-intercepts (1 / K ^ of the double reciprocalplots of A are plotted as a function of Lp(a) concentration. The K, Isderived from the x-intercept (=-«,) .

concentration (Figure 3B) permits the estimation of K,, yield-ing a value of 15 nM.

Effect of Lp(a) on Clot Lysis In PlasmaThe fraction of clot lysed in plasma containing known

amounts of Lp(a) to which three different t-PA concentrationsare added is shown in Figure 4. Increasing concentrations ofLp(a) attenuated the fraction of clot lysed by 3 hours from65% to 53% (open circles, p<0.05) for 50 nM t-PA; from 56%to 42% (open triangles, p<0.03) for 26 nM t-PA; and from46% to 34% (open squares, p<0.03) for 13 nM t-PA In thisassay, spontaneous lysis in the absence of t-PA amounted toapproximately 13% at 3 hours and was not affected by thepresence of Lp(a) (closed symbols).

DiscussionThe kringle domains of plasminogen and t-PA are

triple-loop structures that serve as regulatory sites and

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from

244 ARTERIOSCLEROSIS VOL 10, No 2, MARCH/APRIL 1990

20 30[Lp(a)] (nM)

Rgura 4. Effect of Lp(a) on dot lysis In plasma. Radlolabeledfibrin dots formed In plasma in the presence of a fixed concen-tration Lp(a) were resuspended in plasma containing the sameconcentration of Lp(a) and incubated at 22°C with 13 nM (o),26 nM (A), and 50 nM (•) of tissue-type plasminogen activator(t-PA) for 3 hours. The fraction of clot lysed Is plotted as afunction of Lp(a) concentration. The fraction of clot lysed spon-taneously (I.e., In the absence of t-PA) was not affected by thepresence of Lp(a) (•).

are important for activation of these serine proteases.While the general structural features of these domains aresimilar, relatively minor differences in the primary se-quence impart significant differences in functional prop-erties to specific kringles. Plasminogen contains fivekringle regions, of which the first and fourth bind to lysineand fibrin(ogen),28 while the first three are important forantiplasmin binding. The second kringle of t-PA is alsoinvolved in lysine and fibrin(ogen) binding.30

Since the Lp(a) used in this study contains 26 kringleregions with significant homology to the fourth kringledomain of plasminogen, the ability of Lp(a) to competewith plasminogen for fibrin binding is not unexpected.Lp(a) also competes with t-PA for fibrin, but less well andless completely than it does for plasminogen. Less effec-tive competition with t-PA for fibrin is probably a reflectionof the additional Involvement of a nonkringle structuraldomain of t-PA in fibrin binding (the fibronectin fingerdomain).31

The 2.5-fold difference in apparent estimated K̂ forLp(a) binding to fibrin In the Matrex system and theapparent estimated ICso in the competitive binding assaywith plasminogen (25 nM versus 10 nM) is probably areflection of the steric constraints imparted by the largeLp(a) particle in the competitive binding assay. In addi-tion, the multiple kringle domains of a given Lp(a) particlemay also interact in a cooperative manner on fibrinbinding (although such cooperatrvity is not apparent fromthe binding isotherm or its derived Scatchard plot shownhere) to account for this difference between estimateddirect and competitive binding constants.

Lp(a) clearty attenuates plasminogen activation by t-PAin the presence of fibrin. No such Inhibitory effect wasnoted in the absence of fibrin, nor was any such effectnoted with apo(a)-free LDL Kinetic analysis of the datasuggests that the inhibitory mechanism is uncompetitive.This must be viewed as an operational definition sincethere is no evidence to support the binding of Lp(a)

directly to the enzyme-substrate complex. However, if weconsider that the active catalytic complex is comprised ofenzyme-substrate-activator (t-PA-plasminogen-fibrin),and that Lp(a) binds to the activator, thereby making itunavailable for binding to the catalytic complex, we areleft with the much less active enzyme-substrate complex(t-PA-plasminogen) and, in effect, the equivalent of un-competitive kinetics. The following equations can beused to define this system:

t-PA+plasminogen+fibrin?±t-PA-plasminogen-fibrin k, t-PA-plasmin-fibrin (1)

t-PA+plasminogen+fibrin+Lp(a)^±t-PA-plasminogen+fibrin-Lp7a) k2

t-PA-plasmin+fibrin-Lp(a) —> (2)

where k,^>k2.One group of investigators32 has recently demon-

strated that Lp(a) can attenuate the fibrinolytic activity ofplasma generated by addition of streptokinase, but theconcentrations of Lp(a) required were significantly higherthan noted in our experiments with t-PA (0.43 and0.86 mg/ml). From these data and the recent data ofEdelberg and colleagues33 it appears that streptokinasebinds to Lp(a) and thereby inhibits streptokinase-mediated plasminogen activation competitively as well asuncompetrtivery. In contrast to these findings in which anonphysiologic activator streptokinase was used, wewere unable to detect any inhibition of basal activity oft-PA by Lp(a). We noted only uncompetitive inhibition offibrin-dependent enhancement of basal activity of t-PA,both in a purified system and in a clot lysis assay.

The importance of these observations in regard to theatherogenictty of Lp(a) remains to be determined. Clearty,investigators have identified reduced t-PA activity34 orincreased t-PA inhibitory activity38 in the plasma of youngsurvivors of acute myocardial infarction. Inhibition by Lp(a)of t-PA or plasminogen binding to fibrin with consequentabrogation of the enhancement of fibrinolytic activity mayconfer the potential for vascular occlusion to individualswith elevated levels of this unusual lipoprotein particle.

AcknowledgmentsThe authors thank Deborah Smlck and Patricia Amarante for

technical assistance.

References1. Berg K. A new serum system In man—the Lp system. Acta

Pathol Mlcrobiol Scand 1963:59:369-3822. Berg K, Dahlen G, Frick MH. Lp(a) lipoprotein and pre-

beta,-llpoprotelns In patients with coronary heart disease.J Clln Genet 1974:6:230-235

3. Kostner GM, Avogara P, Cazzolato G, Marth E, Blttolo-Bon G, Qunlcl GB. Upoproteln(a) and the risk for myocar-dial Infarction. Atherosclerosis 1981;38:51-68

4. Koltringer P, Jurgens G. A dominant role of lipoprotein(a)In the investigation and evaluation of parameters Indicatingthe development of cervical atherosclerosis. Atherosclerosis1985:58:187-198

5. Dahlen GH, Guyton JR, Attar MA, Fanner JA, Kautz JA,Gotto AM Jr. Association of levels of lipoprotein Lp(a),plasma lipkls, and other llpoprotelns with coronary arterydisease documented by anglography. Circulation 1986;74:758-765

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from

UPOPROTEIN(a) AND PLASMINOGEN ACTIVATION Loscalzo et al. 245

6. Gaubatz JW, Hektoman C, Gotto AM Jr, Morrisatt JD,Dahlen QH. Human plasma lipoprotein(a) —structural prop-erties. J Btol Chem 1983;254:4582-4589

7. Utennann G, Weber W. Protein composition of Lp(a) lipo-protein from human plasma. FEBS Letts 1983:154:357-361

8. Fless GM, Rollh CA, Scanu AM. Heterogeneity of humanplasma lipoproteln(a). J Blot Chem 1984;259:11470-11478

9. McLean JW, Tomllnson JE, Kuang W-J, et al. cONAsequence of human apolipoproteln(a) is homologous toplasminogen. Nature 1987;330:132-137

10. Larch PG, RIcWI EE, Lergler W, Glllessen D. Localizationof individual lysine-binding regions in human plasminogenand Investigations on their complex-forming properties. EurJ Biochem 1988;107:7-13

11. Vail Z, Patthy L. Location of the Intermediate and highaffinity omega-aminocarboxylic acid-binding sites in humanplasminogen. J BicH Chem 1982:257:2104-2110

12. Plow EF, Collon D. Immunochemical characterization of alow affinity tysine binding site with plasminogen. J BlolChem 1981:256:10864-10869

13. Cole KR, Castelllno FJ. The binding of antifibrinolyticamino acids to kringle-4-containing fragments of plasmino-gen. Arch Biochem Blophys 1984;229:568-575

14. Brown MS, Goldstein JL Teaching old dogmas new tricks.Nature 1987;330:113-114

15. Eaton Dl_ Fless GM, Kohn WJ, et al. Partial amino acidsequence of apolipoproteln(a) shows that It Is homologous toplasminogen. Proc NatJ Acad Sci USA 1987;84:3224-3228

16. Havel RM, Eder HA, Bragdon JH. The distribution andchemical composition of ultracentrifugally separated lipo-proteins In human serum. J Clin Invest 1955;34:1345-1353

17. Fless GM, ZumMallen M, Scanu AM. Physlcochemicalproperties of apolipoproteln(a) and llpoprotein(a-) derivedfrom the dissociation of human plasma lipoproteln(a). J BiolChem 1986;261:8712-8718

18. Armstrong VW, Walll AK, Seldel D. Isolation, characteriza-tion, and uptake in human fibroblasts of an apo(a)-freelipoproteln obtained on reduction of lipoprotein(a). J UpidRes 1985:26:1314-1323

19. Deutsch DG, Mertz ET. Plasminogen - purification fromhuman plasma by affinity chromatography. Science 1970;170:1095-1096

20. McFarlane AS. Efficient trace-labelling of proteins withiodine. Nature 1958:182:58

21. Langer T, Strober W, Levy Rl. The metabolism of lowdensity llpoprotein in familial type II hyperilpoproteinemlaJ Clin Invest 1972:51:1528-1536

22. Shepherd J, Bradford DK, Morgan HG. RadtotodinatJon ofhuman low density lipoproteln - a comparison of four meth-ods. Clin Chlm Acta 1976;66:97-109

23. Loscalzo J, Inbal A, Handln Rl. Von Wiilebrand proteinfacilitates Incorporation of platelets in polymerizing fibrin.J Clin Invest 1986:78:1112-1119

24. Ranby M. Studies on the kinetics of plasminogen activationby tissue plasminogen activator. Biochlm Blophys Acta1982:704:461-469

25. Dbcon M. The graphical determination of Km and Ki. Bio-chem J 1972:129:172-202

26. Weber K, Osbome M. The reliability of molecular weightdeterminations by dodecyi sulfate-potyacrylamlde gel elec-trophorests. J Blol Chem 1969:244:4406-4412

27. Laemmll UK. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 1970;227:680-682

28. Lowry OH, Rosebrough NJ, Fan- AL, Randall L Proteinmeasurement with the Folln phenol reagent J Biol Chem1951:193:265-275

29. Sottrup-Jensen L, Petersen TE, Magnusson S. The pri-mary structure of human plasminogen—isolation of twolyslne-bindlng fragments and one "minl"-plasmlnogen (MW38,000) by elastase-catahyzed specific limited proteolysls.Prog Chem Flbrinolys Thromborys 1980;3:191-209

30. van Zonnenveld AJ, Veennan H, Pannekoek H. Autono-mous functions of structural domains on human tissue-typeplasminogen activator. Proc Natl Acad Scl USA 1986;83:4670-4674

31. van Zonneveld AJ, Veennan H, Pannekoek H. On theInteraction of the finger and kringle-2 domain of tissue-typeplasminogen activator with fibrin. J Blol Chem 1986;261:14214-14218

32. Karadl I, Kostner GM, Grtes A, Nlmpf J, Romlcs L, MadeE. Upoprotein(a) and plasminogen are ImmunochemteaHyrelated. Blochim Biophys Acta 1988:960:91-97

33. Edelberg JM, Gonzalez-Gronow M, Plzzo SV. Upopro-teln(a) inhibits streptokinase-mediated activation of humanplasminogen. Biochemistry 1989:28:2370-2374

34. Verheugt FWA, Wouter ten Cate J, Sturk A, et al. Tissueplasminogen activator activity and inhibition in acute myo-cardial infarction and anglographlcally normal coronary ar-teries. Am J Cardtol 1987;59:1075-1079

35. Hamsten A, Wlman B, de Fairs U, Bkomback M. Increasedplasma levels of a rapid inhibitor of tissue plasminogenactivator in young survivors of myocardial infarction. N EnglJ Med 1985;313:1557-1563

Index Terms: fibrlnogen • low density lipoproteins • kringle domains

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from

J Loscalzo, M Weinfeld, G M Fless and A M ScanuLipoprotein(a), fibrin binding, and plasminogen activation.

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 1990 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleArteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/01.ATV.10.2.2401990;10:240-245Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/10/2/240World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

document.Permissions and Rights Question and AnswerFurther information about this process is available in theis being requested is located, click Request Permissions in the middle column of the Web page under Services.Clearance Center, not the Editorial Office. Once the online version of the published article for which permission

can be obtained via RightsLink, a service of the CopyrightArteriosclerosis, Thrombosis, and Vascular Biology Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on April 5, 2017

http://atvb.ahajournals.org/D

ownloaded from