Isolation of human Val354 -plasminogen as an elastolytic fragment of human Glu1 -plasminogen

Kringle domains of human angiostatin. Characterization of the anti-proliferative activity on endothelial cells

Recently we have identified angiostatin, an endogenous angiogenesis inhibitor of 38 kDa which specifically blocks the growth of endothelial cells (O'Reilly, M. S., Holmgren, L., Shing, Y., Chen, C. , Rosenthal, R. A., Moses, M., Lane, W. S., Cao, Y., Sage, E. H., and Folkman, J. (1994) Cell 79, 315-328; Folkman, J. (1995) Nat. Med. 1, 27-31). Angiostatin was shown to represent an internal fragment of plasminogen containing the first four kringle structures. We now report on the inhibitory effects of individual or combined kringle structures of angiostatin on capillary endothelial cell proliferation. Recombinant kringle 1 and kringle 3 exhibit potent inhibitory activity with half-maximal concentrations (ED50) of 320 nM and 460 nM, respectively. Also, recombinant kringle 2 displays a significant inhibition, although decreased compared with both kringle 1 and kringle 3. In contrast, kringle 4 is an ineffective inhibitor of basic fibroblast growth factor-stimulated endothelial cell proliferation. Among the tandem kringle arrays, the recombinant kringle 2-3 fragment exerts inhibitory activity similar to kringle 2 alone. However, relative to kringle 2-3, a marked enhancement in inhibition is observed when individual kringle 2 and kringle 3 are added together to endothelial cells. This implies that it is necessary to open the cystine bridge between kringle 2 and kringle 3 to obtain the maximal inhibitory effect of kringle 2-3. An increased (<2-fold) inhibitory activity is observed for the kringle 1-3 fragment (ED50 = 70 nM) compared with kringle 1-4 (ED50 = 135 nM). These data indicate that the anti-proliferative activity of angiostatin on endothelial cells is shared by kringle 1, kringle 2, and kringle 3, but probably not by kringle 4 and that more potent inhibition results when kringle 4 is removed from angiostatin. Thus, in view of the variable lysine affinity of the homologous domains, it would appear that lysine binding capability does not correlate with the relative inhibitory effects of the kringle-containing constructs. However, as we also demonstrate, appropriate folding of kringle structures is essential for angiostatin to maintain its full anti-endothelial activity.

Inhibition by tranexamic acid of the conversion of single-chain tissue plasminogen activator to its two chain form by plasmin: the presence on tissue plasminogen activator of a site to bind with lysine binding sites of plasmin

The addition of tranexamic acid inhibited the conversion of single chain tissue plasminogen activator (sct-PA) to its two chain form (tct-PA) by plasmin. Increase in the concentrations of tranexamic acid resulted in more inhibition of the conversion, with 50% inhibition being obtained at 0.251 mM of the concentration of tranexamic acid. The addition of the fragments of plasminogen such as kringle 1 to 3 (K1-3), and K4 resulted in the facilitation of the conversion of sct-PA to tct-PA by plasmin. The addition of tranexamic acid to the mixture of sct-PA and K4 inhibited the rate of the conversion of sct-PA by plasmin. These results suggest that binding of lysine binding sites of plasmin or plasminogen with sct-PA enhanced its conversion to tct-PA by plasmin, and that there is a site on a t-PA molecule to bind to plasminogen or plasmin.

Expression of human plasminogen cDNA in a baculovirus vector-infected insect cell system

A cDNA that encodes the human plasminogen (HPg) amino acid sequence has been inserted adjacent to the polyhedrin promoter in the genome of the baculovirus, Autographa californica nuclear polyhedrosis virus, which was then used to infect cultured cells of the farm armyworm, Spodoptera frugiperda. Under the conditions of cell growth employed, recombinant (rec)-HPg was secreted into the medium after 24 h postinfection (p.i.), at which point virtually no rec-HPg antigen remained inside the cells. At 48 h p.i., a maximal level of intact rec-HPg was present in the medium, which underwent substantial proteolytic digestion after that time. The rec-HPg produced by this expression system possessed a molecular weight equivalent to that of plasma [Glu1]-plasminogen. In addition, the rec-HPg adsorbed to Sepharose-lysine, and was eluted with epsilon-aminocaproic acid (EACA). The recombinant protein also interacted with polyclonal antibodies generated to plasma HPg, as well as with a monoclonal antibody directed against a distinct region (kringle 1-3) of the plasma HPg molecule. Finally, the insect-expressed rec-HPg was activatable to plasmin (HPm) by urokinase. The results demonstrate that this expression system produces a full-length functional single-chain rec-HPg, which can be isolated intact from the culture medium, with some consideration for the temporal events that occur in secretion and longer-term degradation of the protein. The fact that this rec-HPg was converted to HPm with a plasminogen activator, and that it interacted with anti-plasma HPg polyclonal and monoclonal antibodies, as well as with the ligand, EACA, indicates that the molecule retains many of its important functional properties and is folded in an integral manner.

A combination of factor Xa and phosphatidylcholine-phosphatidylserine vesicles bypasses factor VIII in vivo

A combination of phosphatidylcholine-phosphatidylserine lipid vesicles (PCPS), as a source of coagulant active phospholipid, when infused with factor Xa bypasses factor VIII in vivo. To demonstrate this, a reproducible model of bleeding in haemophilic dogs was used. Control studies were performed in normal dogs. In initial studies, factor Xa/PCPS at a dose of 6.5 x 10(-12) and 4.0 x 10(-7) moles/kg respectively failed to correct the abnormal bleeding in the haemophilic animals and initiated a bleeding diathesis in the normal controls. Coagulation studies and immunoblotting demonstrated activation of protein C and an anticoagulant effect resulting from significant falls in the levels of factors V and VIII. Adjustment of the dose of factor Xa/PCPS to 2.6 x 10(-11) and 4.0 x 10(-8) moles/kg respectively produced an immediate haemostatic effect in both haemophilic and normal animals with bleeding stopping within 15-30 s. Despite this observation, protein C activation was again noted. It is concluded that the presence of coagulant active phospholipid and factor Xa in prothrombin complex concentrates may explain the observed factor VIII bypassing activity of these preparations and that the use of a controlled formulation of these two components may provide a more effective approach to the management of patients with factor VIII inhibitors.

Examination of the secondary structure of the kringle 4 domain of human plasminogen

The structure of a small region of human plasminogen (F4), consisting of amino acid residues Val354-Ala439 and containing its kringle 4 (K4) domain (residues Cys357-Cys434), has been predicted from Chou-Fasman calculations and hydropathy profiles, and compared to circular dichroism (CD) measurements on the isolated fragment. Calculations, by the Chou-Fasman method, of the probabilities of various types of secondary structures that exist in this region reveal that no helical structures are present. Of the total of 86 amino acid residues present in this K4-containing peptide region, 37% can adopt conformations of beta-pleated sheets, 48% of the amino acids can exist in beta-turns, and 15% of the residues can be present as coils. The structure of F4 in dilute aqueous solution has been experimentally evaluated by CD measurements. At pH = 7.4, in dilute salt solutions, a total of 64% beta-structures, 30% beta-turns, and 6% coiled structures is estimated to be present in this peptide region. Consideration of the marginal stability of many of the conformational regions of F4, as predicted by Chou-Fasman calculations, suggests that secondary structural flexibility is present in this fragment, which could result in ready adoption of new conformations. The hydropathy profile of F4 has been determined and suggests that this polypeptide is highly hydrophilic, especially in the regions of residues His387-Tyr396 and Cys406-Lys413. Thus, it appears as though a large portion of the surface of F4 can be exposed to solvent in its native conformation.

Interspecies cross-reactivity of monoclonal antibodies to various epitopes of human plasminogen

The immunological cross-reactivities of three conformationally specific monoclonal antibodies to distinct epitopes on human plasminogen toward plasminogens purified from 14 additional species have been examined. Antibody 10-F-1, which is produced against an epitope on the kringle 4 region of human plasminogen, shows a high degree (greater than 80%) of cross-reactivity against baboon, goat, monkey, ovine, and rabbit plasminogens; more limited (20-50%) cross-reactivity against bovine, equine, goose, guinea pig, mouse, rat, and porcine plasminogens; and little comparable cross-reactivity against canine and chicken plasminogens. Antibody 10-H-2, generated to an epitope of the kringles 1-3 region of human plasminogen, shows extensive cross-reactivity (72%) only toward monkey plasminogen, more limited (22-35%) cross-reactivity toward equine and rabbit plasminogens, and much less cross-reactivity toward any other of the above plasminogens. Antibody 10-V-1, also produced against an epitope on the kringle 1-3 region of human plasminogen, which is distinct from the 10-H-2 epitope, shows extensive cross-reactivity (72-100%) with baboon, monkey, and rabbit plasminogens; more limited cross-reactivity with equine (48%) and mouse (28%) plasminogens; and a low level of such reactivity with the remaining plasminogens. These studies show that the extent of interspecies cross-reactivity of various plasminogens greatly depends upon the epitope in question. The K4 region of these molecules appears more extensively conserved than the K1-3 region, at least in regard to the particular epitopes examined in this study.

The binding of antifibrinolytic amino acids to kringle-4-containing fragments of plasminogen

A monoclonal antibody to human plasminogen, 10-F-1, was found to interact with the lysine-binding site (LBS) on the kringle 4 (K 4) region of the molecule. This observation has been employed to measure the binding of various antifibrinolytic amino acid analogs of epsilon-aminocaproic acid (epsilon ACA) to its site on K 4 in appropriate elastolytic-derived fragments of human plasminogen and to other species of plasminogen to which antibody 10-F-1 cross-reacts. By analysis of the concentration dependence of epsilon ACA displacement of [125I]10-F-1 from human Glu1Pg, a KD for epsilon ACA of 7.1 +/- 1.0 mM was calculated. Similar experiments with K 4-containing fragments of Glu1Pg, viz., Lys77Pg, K 4, Lys77H and Val354Pg, yielded KD values of 6.6 +/- 1.0, 7.5 +/- 1.0, 6.6 +/- 1.0, and 12.0 +/- 2.0 mM, respectively. When baboon, goat, monkey, rabbit, and sheep plasminogens were substituted for human plasminogen, the KD values calculated ranged from 2.1 to 7.1 mM. The KD values for several analogs of epsilon ACA, i.e., 4-aminobutyric acid, 5-aminopentanoic acid, 8-aminooctanoic acid, L-lysine, and trans-aminomethyl cyclohexane-1-carboxylic acid, were measured to the K 4 region of Lys77Pg. The values obtained were 11.3 +/- 1.5, 9.0 +/- 1.0, 71.0 +/- 10, 38.0 +/- 5.0, and 1.1 +/- 0.4 mM, respectively. Additionally, the KD of trans-aminomethylcyclohexane-1-carboxylic acid towards the K 4 region of Glu1Pg, Lys77Pg, and isolated K 4 was found to be 2.4 +/- 0.5, 1.1 +/- 0.3, and 2.0 +/- 0.6 mM, respectively. These studies show directly that the LBS on the K 4 domain of plasminogen represents one of its 4-5 weak binding sites and that this site can be specifically probed with the use of monoclonal antibody 10-F-1. Furthermore, it appears as though this site is conserved in several important proteolytic fragments of plasminogen, providing additional evidence that these fragments exist as independent domains in the native molecule. Finally, this weak LBS on the K 4 domain of human plasminogen is also present in other species of plasminogen.

A probe of the human plasmin-alpha 2-macroglobulin interaction utilizing monoclonal antibodies to human plasmin

These experiments have shown that alpha 2M may have a bait region capable of accommodating proteases of a far greater molecular weight than previously recognized, since antibody-plasmin complexes apparently become entrapped. In the case of large proteases, however, it now appears that a portion of these molecules is not enclosed. Clearly, in the case of plasmin, the K1-4 region remains outside the molecular trap.

Complex formation of human Val354-plasminogen with streptokinase

A controlled digestion of native human plasminogen (Glu1-plasminogen) from either carbohydrate variant 1 or 2 with pancreatic elastase yields, among others, an activatable fragment, designated Val354-plasminogen (Pgc). This species represents the smallest human plasminogen fragment (ca., 50,000 Mw) known to bind to Sepharose-lysine. Pgc forms a tight equimolar complex with streptokinase, as determined by sucrose density ultracentrifugation, whether formed from Val354-plasminogen or Val354-plasmin. The Val354-plasmin-streptokinase complex readily activates ovine plasminogen, which is normally insensitive to streptokinase, and this complex is essentially inactive toward casein, similar to the equimolar complexes formed with Lys77-plasmin and the elastase fragment, Val442-plasmin.

The role of the lysine binding sites of human plasmin in the hydrolysis of human fibrinogen

The importance of the lysine binding sites of human plasmin for its ability to digest human fibrinogen has been assessed by analyzing the nature and rate of the products formed in the presence of epsilon-aminocaproic acid. No major differences in this regard were found when comparing Lys77-plasmin and Val442-plasmin, in the absence of epsilon-aminocaproic acid, the latter plasmin being devoid of the lysine binding sites present on residues Glu1-Val441 (K 1-4). The presence of epsilon-aminocaproic acid, at concentrations ranging from 0.5-5.0 mM, results in progressively stronger inhibition of the digestion of fibrinogen and in appearance of fibrinogen degradation products Y, D and E, for both Lys77-plasmin, and Val442-plasmin, showing the importance of lysine binding regions in this property. However, since both plasmin forms were inhibited equally well at all levels of epsilon-aminocaproic acid, these studies show that lysine binding sites other than those present on region K 1-4 of Lys77-plasmin are of primary importance to fibrinogenolysis by plasmin.

Amino acid sequence analysis of the asparagine-288 region of the carbohydrate variants of human plasminogen

The amino acid sequences of the two major carbohydrate variants in the region of Asn288 have been determined. A peptide isolated from plasminogen variant 1, which contains a complex-type Asn288-linked oligosaccharide, was found to possess the amino acid sequence-Ser280-Ala-Gln-Thr-Pro-His-Thr-His-Asn(CHO)-Arg-Thr290-Pro-Glu-, in agreement with the previously published sequence [Sottrup-Jensen, L., Claeys, H., Zajdel, M., Petersen, T. E., & Magnusson, S. (1978) in Progress in Chemical Fibrinolysis and Thrombolysis (Davidson, J. F., Rowan, R. M., Samama, M. M., & Desnoyers, P. C., Eds.) Vol. 3, pp 191-209, Raven Press, New York]. A similar peptide isolated from plasminogen variant 2 did not contain oligosaccharide but possessed an amino acid sequence identical with the corresponding variant 1 peptide. Thus, the basis for the lack of the complex-type oligosaccharide in human plasminogen variant 2 does not reside in substitution of essential amino acid residues in the region of the Asn288-linked glycosylation site.

Fibrinolysis and fibrinogenolysis by Val442-plasmin

Elastase cleavage of Lys77-plasmin results in the formation of Val442-plasmin. This result suggests that small, active plasmin fragments can be produced even under conditions of high plasminogen activator levels such as occur in vivo. We examined the effect of the generation of such fragments by studying the degradation of fibrinogen and fibrin by Val442-plasmin. Val442-plasmin lysis of fibrinogen yielded the same products as obtained with Lys77-plasmin, but at a slightly lower rate. Lysine inhibited fibrinogenolysis by both Lys77-plasmin and Val442-plasmin. The marked inhibition observed at concentrations higher than 10 mM lysine occurred to the same extent for both proteases. In addition, the products and rate of fibrinolysis were the same for both proteases. These results indicate that the lysine binding regions present in Lys77-plasmin but absent in Val442-plasmin do not determine the rate, reaction products, or lysine inhibition of fibrinolysis and fibrinogenolysis by plasmin.