JBC Papers in Press. Published on June 21, 2002 as Manuscript M203782200
Structural Elements that Govern the Substrate Specificity of the Clot-Dissolving Enzyme Plasmin*
Ryan B. Turner, Lin Liu, Irina Y. Sazonova, and Guy L. Reed‡⊥
From the Cardiovascular Biology Laboratory, Harvard School of Public Health and the ‡
Massachusetts General Hospital, Boston, Massachusetts 02114
*This work was supported in part by NIH grant HL-58496 to G.L.R.
⊥
To whom correspondence should be addressed:
Cardiovascular Biology Laboratory, HSPH II-127 677 Huntington Ave. Boston, MA, 02115 Tel: 617-432-0031 Fax: 617-432-0033 E-mail:
[email protected]
Running Title: Structural Elements of micro-Pg Substrate Specificity
Copyright 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
SUMMARY There is remarkable homology between the core structures of plasmin, a fibrin clotdegrading enzyme and factor D, a complement-activating enzyme, despite markedly different biological functions. We postulated that sequence divergence in the loop structures between these two enzymes mediated plasmin's unique substrate and inhibitor interactions. Recombinant microplasminogens chimerized with factor D sequences at loops 3, 5 and 7 were cleaved by the plasminogen activator urokinase and developed titratable active sites. Chimerization abolished functional interactions with the plasminogen activator streptokinase but did not block complex formation. The microplasmin chimeras showed enhanced resistance (ki decreased up to 2-3 times) to inactivation of microplasmin by α2-antiplasmin. Chimerization had no or minimal (~2-fold) effects on the catalytic efficiency of small substrates and there were no discernible changes in their ability to cleave fibrin substrate. However, microplasmin and the microplasmin chimeras showed enhanced abilities to degrade fibrin in plasma clots suspended in human plasma. These studies indicate that loop regions of the protease domain of plasmin are important for interactions with substrates, regulatory molecules and inhibitors. Because modification of these regions affected substrate and inhibitor interactions, loop-chimerization may hold promise for improving the clot-dissolving properties of this enzyme.
2
INTRODUCTION The serine protease plasmin cleaves fibrin to dissolve blood clots in a process known as fibrinolysis. Plasmin and its proenzyme plasminogen (Pg)1 also play important roles in cell migration, tissue remodeling and bacterial invasion (1). The function of plasmin and Pg (P(g)) are regulated through protein interactions with plasminogen activators (such as staphylokinase and streptokinase), substrates (such as fibrin), inhibitors (such as α2-antiplasmin), receptors and other molecules. Glu-Pg, the physiologic form of Pg, contains an NH2-terminal segment, 5 kringle domains and a protease domain. The kringle domains are known to play an important role in mediating protein-protein interactions, but the contribution of the protease domain to these interactions is not well understood. However, recent insights into the three-dimensional structure of the protease domain of plasmin (microplasmin) and of related serine proteases such as trypsin, chymotrypsin and factor D provide structural hypotheses about the intimate contact sites in plasmin that may interact with plasminogen activators, fibrin and α2-antiplasmin (2,3). The major physiologic substrate of plasmin is fibrin. Indeed, crystal structure analysis has suggested that the substrate cleft of plasmin is elongated to accommodate the structure of fibrin (1). The primary inhibitor of plasmin in the blood is the serpin α2-antiplasmin (4,5). Both fibrin and α2-antiplasmin interact with the kringle domains of plasmin but their contact sites with the protease domain are poorly understood. Streptokinase and staphylokinase are bacterial cofactors that complex with plasmin and convert plasmin from an enzyme that cleaves fibrin to an enzyme that cuts or ‘activates’ Pg. Among the plasminogen activators, these two proteins appear to have the most extensive contacts with the protease domains of Pg and plasmin. In the staphylokinase-microplasmin complex, microplasmin retains its susceptibility to inhibition by α2-antiplasmin (6). Conversely, in the streptokinase-microplasmin complex, microplasmin is resistant to inhibition by α2-antiplasmin (2). Taken together, these
3
experimental results and analysis of X-ray crystallographic structures of staphylokinasemicroplasmin and streptokinase-microplasmin define the surfaces of microplasmin that may interact with these cofactors and that may be necessary for interaction with α2-antiplasmin. The protease domain of plasmin (microplasmin) shares extensive sequence and structural homology with other serine protease domains such as trypsin, chymotrypsin and factor D (7). In vitro, trypsin and chymotrypsin cleave fibrin and are inhibited by α2antiplasmin, much like plasmin. Factor D, which is essential for the formation of the C3 convertase of the alternative pathway of complement activation, does not interact with fibrin or α2-antiplasmin. Still, the three-dimensional structures of these four serine proteases are highly conserved with a root mean square deviation between microplasmin and factor D structures of 1.06 angstroms squared (7). The differences among these protease structures are primarily in the loop regions of the enzymes, which suggests that divergences in these sequences are responsible for selective interactions of these proteases with substrates and inhibitors. To examine this hypothesis we chimerized specific loops of microplasmin with the corresponding structural elements of factor D and then determined whether these structures affected interactions of microplasmin with fibrin and plasminogen activators. Chimerization provides a strategy for examining structure-function relationships, with a low risk of fundamentally altering protein conformation, by swapping elements from a structurally homologous, but functionally different molecule.
4
EXPERIMENTAL PROCEDURES Cloning, Purification, & Expression of Mutant Proteins Microplasminogen was chimerized with the corresponding loop sequences of factor D by the polymerase chain reaction with overlap extension. The microplasminogen and factor D sequences are listed in Table 1 with the appropriate PCR primers. The PCR products were sequenced on both strands to confirm the target mutations and the recombinant microplasminogens were ligated into pET11d for bacterial expression (8). Recombinant microplasminogens were purified from inclusion bodies and refolded in 55 mM Tris pH 8.2, with 10.56 mM NaCl, 0.44 mM KCl, 0.055% polyethylene glycol 3350, 2.2 mM MgCl2, 2.2 mM CaCl2, 550 mM L-arginine, 1 mM reduced glutathione, 0.1 mM oxidized glutathione. Purified proteins were then dialyzed in 100 mM Tris pH 8.0 and 10 mM EDTA.
Active Site Titration The molar quantities of the active sites in the recombinant microplasmins were determined at 37oC in a Hitachi 2500 fluorescence spectrophotometer by active site titration with the fluorogenic substrate 4-methylumbelliferyl p-guanidinobenzoate (MUGB, Sigma) as described (9). Unless otherwise indicated all experiments were performed with equivalent amounts of active protein based on this titration data. Kinetic Assays of Microplasminogen Recombinants, Amidolysis The amidase kinetic parameters of the microplasmins were determined by measuring the cleavage of the paranitroanilide substrate S2251 (H-D-valyl-L-leucyl-L-lysine-p-nitroanilide dihydrochloride, Chromogenix) at 405 nm. Microplasmins (20 nM) and assay buffer (50 mM o
Tris-HCl, 100 mM NaCl, pH 7.4) were placed in a quartz cuvette at 37 C and various concentrations of S2251 (0.50-4.0 mM) were added to obtain a final volume of 300 µl. The
5
change in absorbance was measured at 405 nm for 5 min at 37oC in a thermostatted Cary 100Bio spectrophotometer. The data were plotted as V/S and analyzed by hyperbolic curve fitting using a Sigma Plot program as described (10). Plasminogen Activation Microplasminogens (50 nM) were mixed with urokinase (1U) and added to microtiter plate wells containing 0.5 mM S2251 in assay buffer. The conversion of plasminogen to plasmin was detected by monitoring the cleavage of S2251 for 20 min at 37oC. To test active complex formation by streptokinase a stoichiometric microplasminogen-streptokinase complex (50 nM) was formed in assay buffer. Active complex formation was also tested using staphylokinase in excess to create a 50 nM microplasminogen-staphylokinase complex and monitored as before. Binding Assays The binding of the microplasminogens to streptokinase was studied in microtiter plates. The microplasminogens (25 µl, 20 µg/ml) or no protein (control) were adsorbed on microtiter plates for 1 hr. The nonspecific protein binding was blocked by 1% BSA for 1 hr. After washing, an anti-plasminogen monoclonal antibody (Ab 340-11) was added for 1 hr. The 125
bound antibody was detected by I-(sheep antimouse Ab) (50,000 cpm) followed by gammacounting. The binding of microplasminogens to streptokinase was also studied in a microtiter plate. Streptokinase (10 µg/ml, 25 µl) was adsorbed to the microtiter plate for 1 hr, blocked as before with 1% BSA, and the microplasminogens (25 µl, 20 µg/ml) or no microplasminogen (control) were added to the wells for 1 hr. The bound primary antibody was detected by 125I(sheep antimouse Ab) (50,000 cpm) followed by gamma-counting as before. Inhibition of Microplasmin Mutants by α2-antiplasmin Wild-type or mutant microplasmins (20 nM) were added to cuvettes containing S2251
6
(0.5 mM) in assay buffer and the change in absorbance at 405 nm was recorded at 0.3 min. intervals prior to and after the addition of human α2-antiplasmin (120 nM, Calbiochem) as described (11). The rate constant ki of the reaction between the chimeras and α2-antiplasmin was calculated and residual enzyme concentration was plotted as described (11). Plasma Clot Lysis by Mutant Microplasmins Fresh frozen human plasma (100 µl) was mixed with trace amounts of 125I-fibrinogen and clotted with CaCl2 (20 mM) and thrombin (5 µl, 0.05 U) for 1 hr at 37oC. After washing in 1 ml of Tris-buffered saline with 10 µΜ D-phenylalanyl-L-propyl-L-arginine chloromethyl ketone (P-PACK) to inhibit residual thrombin, the clots were suspended in 50 µl plasma as supernatant. In "synchronous" experiments, the mutant microplasminogens (50 µl, 0-1 µΜ) were added to the clots followed by 5 U of urokinase. In "pre-activation" experiments, 25 U of urokinase (an appropriate catalytic molar ratio that also achieved low background fibrinolysis) o was incubated with the microplasminogens (1 µM, 250 µl) for 1 hr at 37 C. After activation
various concentrations (50 ul, 0-1 µM) of the microplasmins were added to the clots. After incubation of 5-6 hrs, the amount of the residual clot was measured by gamma-counting and the fractional fibrinolysis was determined as described (12). Fibrinogen Clot Lysis Assay Fibrinogen (5 µl, 20 mg/ml from American Diagnostica) was mixed with trace amounts of 125I-fibrinogen, and clotted by the addition of CaCl2 (2.5 µl, 0.4 M) and thrombin (2.5 µl, 1 o
unit/ml) for 1 hr at 37 C. In "synchronous" experiments, the mutant microplasminogens (200 µl) were added to the clots as supernatant followed by the addition of 5 U of urokinase to activate the microplasminogens. In "pre-activation" experiments urokinase (100 U, which was sufficient to fully activate the microplasminogens) was added to the microplasminogens (1 µM,
7
500 µl) and incubated for 1 hr at 37oC. Serial dilutions were performed and the activated microplasminogens (0-1 µM, 200 µl) were then added to the clot. The residual clot was measured at 5-6 hrs by gamma-counting and the fractional fibrinolysis was determined as above. Interaction of factor D and α2-antiplasmin The interaction of factor D with α2-antiplasmin was determined using a standard assay for factor D activity (13). Factor D (20 nM final) and 5,5'-dithiobis (2-nitrobenzoic acid) (DTNB, 0.13 mM final) were added to Tris buffer (100 mM, pH 8.0) and incubated at 37oC for 3 minutes. Then Z-Lys-SBzl (Enzyme Systems Products) was added (20 mM final) to start the reaction and the absorbance was measured at 412 nm at 0.3 min. intervals prior to and after the addition of α2-antiplasmin (100 nM) (13). The data was plotted as residual enzyme concentration versus time.
8
RESULTS Chimerization of microplasminogen In the crystal structures, streptokinase is projected to interact with microplasmin at loops 3, 5, 6 and 7, while staphylokinase interacts only with loop 7. Through these interactions streptokinase renders plasmin unable to cleave fibrin and resistant to inhibition by α2antiplasmin (2,3). This suggests that these particular loop structures may be very important for these interactions. Corresponding to the functional differences between microplasmin and factor D, there are notable differences in the sequences of these loops between these two molecules (Figure 1). For example loop 7 of factor D is longer than loop 7 of microplasmin and shows a different electrostatic potential. Negatively charged clusters are seen in factor D and positively charged clusters seen in microplasminogen. Loop 3 of factor D lacks negatively charged residues while microplasminogen has an arginine present. Loop 5 of factor D has similarities with microplasminogen with the notable difference in overall charge (positive vs. negative, respectively). Loop 6 of factor D is also similar to microplasmin though the overall charge of factor D is different (positive). To examine the contribution of these structures to mediating interactions of microplasmin, loop chimeras were created. The chimeras were expressed in bacteria, purified and refolded to obtain active protein. Interaction of chimeric microplasminogens with plasminogen activators The plasminogen activator urokinase is a serine protease that directly cleaves plasminogen to create plasmin. Each microplasminogen was treated with various amounts of urokinase for different lengths of time to determine the amount of active protein by active site titration with the fluorogenic substrate MUGB (not shown). Wild-type microplasminogen and the loop 3 and 5 chimeric microplasminogens were more readily cleaved by urokinase than was the loop 7 chimera (Fig. 2A). Following cleavage by urokinase, wild-type microplasminogen
9
and the loop 3, 5, and 7 chimeric microplasminogens developed amidolytic activity with a plasmin-selective substrate S2251 (Fig. 2B). In contrast, the loop 6 chimera was not cleaved by urokinase (Fig. 2A) even after prolonged incubation and never developed proteolytic activity (Fig. 2B). This suggested that it may not be refolded to active protein. Molecular modeling of the crystal structure predicted that loop 3, loop 5 and loop 7 of microplasmin are involved in interactions with streptokinase in the activator complex. Consistent with these observations all three loop mutants were not efficiently activated by the bacterial plasminogen activator streptokinase (Fig. 2C). Because complex formation between microplasminogen is required for activation by streptokinase, the binding of streptokinase to these domains was examined with a monoclonal antibody directed to the protease domain (34011). When compared to wells containing no microplasminogen, MAb 340-11 showed specific binding to each chimeric microplasminogen (Fig. 3A), although it bound with less avidity to the loop 3 and 7 chimeras. This permitted the use of the MAb to examine microplasminogen binding to immobilized streptokinase in a similar assay. When compared to control (no microplasminogen) all microplasminogen chimeras bound to streptokinase (Fig. 3B) in a pattern that simulated the binding of microplasminogen alone (Fig. 3A). Thus the failure of streptokinase to activate these microplasminogen chimeras was not simply due to an inability of these proteins to bind to each other. The ability of staphylokinase to activate the microplasminogen chimeras was tested to further understand the interaction of these loops with plasminogen activators. Unlike urokinase (Fig. 2B) or streptokinase (Fig. 2C), staphylokinase was unable to efficiently activate the wildtype microplasminogen (Fig. 2D). Staphylokinase was unable to activate the chimeric microplasminogens though it was able to efficiently activate Glu-plasminogen (Fig. 2D). This
10
result indicates the importance of kringle domains in the staphylokinase mechanism of activation. Effects of chimerization on amidolysis kinetics Chimerization had modest effects on the kinetic parameters of the mutant microplasmins when compared to wild-type microplasmin (Table 2). Chimerization of loop 3 did not significantly alter the kinetics of amidolysis compared with wild-type microplasmin. Chimerization of loop 7 slightly reduced catalytic efficiency by ~2-fold. The loop 5 chimera displayed an increased Km (~3-fold) and a decreased catalytic efficiency (~2-fold). Inhibition by α2-antiplasmin Under these experimental conditions, factor D was resistant to the inhibitory effects of the serine protease inhibitor α2-antiplasmin while wild-type microplasmin was rapidly inhibited (Fig. 4, ki = 19900 + 430 M-1s-1). Chimerization with factor D loops altered the resistance of -1 -1 microplasmin to α2-antiplasmin. The loop 5 (ki = 7780 ± 270 M s ) and loop 7 (ki = 6131 ±
440 M-1s-1) chimeras were 2-3 fold more resistant to inhibition than wild-type microplasminogen. The loop 3 mutant also displayed reduced kinetics of inhibition (ki = 14800 ± 900 M-1s-1) when compared to wild-type microplasminogen. Effects of chimerization on fibrin and plasma clot lysis Experiments were performed to investigate the effect of chimerization on the interactions of microplasmin with fibrin, its chief physiologic substrate. When microplasminogens were "pre-activated" to plasmin and added to fibrin clots in the absence of inhibitor, plasmin was more effective in fibrinolysis than wild-type microplasmin. However, there was no discernible difference between wild-type and chimeric microplasmins in the efficiency of fibrinolysis. When Glu-plasminogen and the microplasminogens were added to fibrin clots with urokinase,
11
Glu-plasminogen was the most potent fibrinolytic agent and the chimeras were similar to wildtype microplasminogen in their ability to cleave fibrin (Fig. 5B). Thus, introduction of the loop mutations did not alter the ability of the chimeras to cleave fibrin. Normally fibrinolysis occurs in plasma or blood which contains Glu-plasminogen, α2antiplasmin and other inhibitors of plasmin. Consequently we examined the effect of chimerization on fibrinolysis of human clots in human plasma. When Glu-plasminogen and the recombinant microplasminogens were pre-activated with urokinase and added to human plasma clots the pattern of fibrinolysis (Fig. 5C) was opposite to that seen with fibrin clots (Fig. 5A). Recombinant wild-type microplasminogen was approximately 2-fold more potent than Gluplasminogen in achieving equivalent lysis (Fig. 5C). The loop 3, loop 5 and loop 7 chimeras were up to 3-fold more potent than Glu-plasminogen and up to 1.5-fold more potent than wildtype microplasminogen. When the different plasminogens were added to plasma with urokinase a similar pattern was seen but the differences between the microplasminogens (wild-type and mutants) and Glu-plasminogen were more marked (5-6 fold, Fig. 5D). DISCUSSION The serine proteases have unique biological functions in vivo despite the extraordinary structural conservation of their protease domains. We postulated that the differences in biological activity between plasmin and factor D were due in part to sequence divergences in the flexible loops of the catalytic domains which target the function of these proteases to specific sites and substrates. The major goal of these studies was to understand the contribution of some of these structural elements to the interactions of the protease domain of plasmin with streptokinase, fibrin and α2-antiplasmin. Loops 3, 5, 6 and 7 of microplasminogen are shielded from solution when plasmin forms a complex with streptokinase while only loop 7 is shielded in the staphylokinase-microplasmin complex (2,3). The fact that streptokinase, but not
12
staphylokinase modifies the ability of plasmin to interact with fibrin and α2-antiplasmin suggested that these loops may participate in substrate and inhibitor interactions. Plasmin is inhibited by α2-antiplasmin in two steps: 1) a very fast reversible second order reaction and 2) a slower irreversible first-order reaction (14). Although the first step in this reaction involves interactions between the kringle domains of plasmin and the carboxy terminus of α2-antiplasmin, the structural interactions of the protease domain in the second step are not understood. Our studies indicated that chimerization of loops 5 and 7, and to a lesser extent loop 3, reduced the ability of α2-antiplasmin to inhibit microplasmin. This provided the first experimental evidence that structural elements in the protease domain are important for the interactions of these two molecules. We postulated that these loop residues may be integral for the proper alignment and positioning of α2-antiplasmin with respect to the catalytic triad. Chimerization may have altered intermolecular contacts between plasmin-α2-antiplasmin by deleting critical residues, altering residue charge or creating steric interference. The recent crystal structure of the α1-antitrypsin/trypsin complex provides a model for the interactions between plasmin and α2-antiplasmin. It reveals that the homologous loop 3, loop 5 and loop 7 of trypsin are in contact with α1-antitrypsin in the protease complex (15). Thus, the equivalent loops of microplasmin may be predicted to have similar interactions with its primary inhibitor α2-antiplasmin. Assuming a similar serpin-protease reaction occurs between microplasmin and α2-antiplasmin, the crystallographic data of the α1antitrypsin/trypsin complex may in part explain how the chimeras gained resistance to α2antiplasmin inhibition through disruption of residues required by microplasmin for recognition of α2-antiplasmin. The inhibition of plasmin by α2-antiplasmin also parallels the inhibition of two chain
13
tissue plasminogen activator by plasminogen activator inhibitor 1 (PAI-1) (16). Mutation of positively charged residues 296-302 made tissue plasminogen activator nearly 95% resistant to inhibition by PAI-1 and enhanced the fibrinolytic activity of this molecule in vitro (17). These residues correspond to the equivalent loop 3 structure in plasminogen. In our studies chimerization of loop 3 had only mild effects on reducing the inhibition of microplasmin by α2antiplasmin, suggesting that this is not a major site of interaction between these two molecules. These studies provide biochemical evidence for the role of loops 3, 5 and 7 in activator complex formation with streptokinase. Although streptokinase formed a complex with all three chimeras (Fig. 3), it was unable to generate an active site in plasminogen. In contrast, another plasminogen activator, urokinase, cleaved each of the loop chimeras to plasmin. The effects of the loop 3 chimera in disrupting interactions with streptokinase are particularly notable since this represented a comparatively conservative change in microplasmin with deletion of Arg 582 and substitution of homologous residues. In the loop 3 chimera of our study Thr 581 was mutated to Leu. Moreover, in the structure of the microplasmin with streptokinase, loop 3 interacts with the beta domain which, compared with the alpha and gamma domains of streptokinase has comparatively few intermolecular contacts (2). Wang and Reich found that deletion of Thr 581 in loop 3 also caused a loss of streptokinase-microplasmin complex activity (18). In the streptokinase-microplasmin structure loop 5 also contacts the beta domain and molecular modeling suggests that within this loop Arg 610 has important interactions with streptokinase. Finally, chimerization introduces a large insertion into loop 7 so that steric factors along with an increase in negatively charged residues may contribute to abolition of the ability to form a functional activator complex. The role of the loop mutations in forming an activator complex with staphylokinase could not be evaluated under the conditions of these experiments because staphylokinase,
14
though it efficiently activated Glu-Pg (the physiologic form of Pg), showed no significant ability to activate microplasminogen (even after 8 hrs of incubation, Fig. 2D). Molecular modeling would suggest, however, that loop chimeras 3 and 5 would be similar to wild-type microplasmin in their ability to form a Pg activator as these loops are not in contact with the staphylokinase molecule in the activator complex. Chimerization did not significantly affect the ability of microplasmin to cleave fibrin. Wild-type and chimeric microplasminogens had equivalent fibrinolytic effects, either when added with urokinase to fibrin clots or when pre-activated by urokinase and then added to fibrin clots. In the same experiments, greater fibrinolysis was obtained with Glu-Pg and urokinase; this probably reflects enhanced fibrin-targeting by the kringle domains (19,20). However, when fibrinolytic experiments were performed in the presence of plasma, which contains α2antiplasmin and other inhibitors, an opposite result was obtained. Wild-type and chimeric microplasminogens were significantly more effective than Glu-plasminogen as fibrinolytic agents, particularly when the plasminogens and urokinase were added to plasma together. This probably reflects the slower inhibition by α2-antiplasmin of microplasmin which lacks kringle domains. The fact that microplasmin(ogen) was more potent than Glu-plasmin(ogen) when activated in plasma (Fig. 5D), than when pre-activated and added to plasma, (Fig. 5C) suggests that microplasminogen may be a more efficient substrate for urokinase than Glu-Pg (21). The increased potency of the chimeric microplasminogens was due at least in part to the enhanced resistance of these chimeras to α2-antiplasmin. However, the superior potency of the loop 3 chimera, which displayed less resistance to α2-antiplasmin than the other chimeras, suggests that other factors such as differential rates of plasminogen activation, variable resistance to α2macroglobulin, enhanced stability in plasma, etc., may play a role in affecting fibrinolysis by these molecules.
15
In conclusion these experiments provide evidence that loop structures in the protease domain specifically modulate the functional interactions of microplasmin(ogen) with certain regulatory molecules. Further analysis is likely to reveal additional sites of interaction of plasmin with key substrates, inhibitors and regulatory molecules. Molecular manipulation of these sites may provide a rational approach to improve the therapeutic value of plasmin for treating thrombotic disorders.
16
FOOTNOTES
1
ABBREVIATIONS: The abbreviations used are: Pg, human plasminogen; Glu-Pg, Pg with
NH2-terminal glutamic acid residue; PAI-1, plasminogen activator inhibitor-1; PAGE, polyacrylamide gel electrophoresis; Km, Michaelis-Menten constant; kcat, catalytic rate constant; ki, inhibition constant; MUGB, 4-methylumbelliferyl p-guanidinobenzoate; S2251, H-D-valylL-leucyl-L-lysine-p-nitroanilide dihydrochloride; P-PACK, D-phenylalanyl-L-propyl-Larginine chloromethyl ketone; DTNB, 5,5'-dithiobis (2-nitrobenzoic acid); Ab, antibody; MAb, monoclonal antibody; BSA, bovine serum albumin.
17
ACKNOWLEDGEMENTS This study was supported by grant NIH-HL58496 (G.L.R.) and R.B.T. is a recipient of a minority graduate research assistant award (HL-58496) and an HHMI Research Training Fellowship.
18
REFERENCES 1. Parry, M. A., Zhang, X. C., and Bode, I. (2000) Trends Biochem. Sci. 25, 53-9 2. Wang, X., Lin, X., Loy, J. A., Tang, J., and Zhang, X. C. (1998) Science 281, 1662-5 3. Parry, M. A., Fernandez-Catalan, C., Bergner, A., Huber, R., Hopfner, K. P., Schlott, B., Guhrs, K. H., and Bode, W. (1998) Nat. Struct. Biol. 5, 917-23 4. Moroi, M., and Aoki, N. (1976) J. Biol. Chem. 251, 5956-65 5. Mullertz, S., and Clemmensen, I. (1976) Biochem. J. 159, 545-53 6. Collen, D. (1998) Nat. Med. 4, 279-84 7. Peisach, E., Wang, J., de los Santos, T., Reich, E., and Ringe, D. (1999) Biochemistry 38, 11180-8 8. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual., Second Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor 9. Reed, G. L., Lin, L. F., Parhami-Seren, B., and Kussie, P. (1995) Biochemistry 34, 10266-71 10. Sazonova, I. Y., Houng, A. K., Chowdhry, S. A., Robinson, B. R., Hedstrom, L., and Reed, G. L. (2001) J. Biol. Chem. 276, 12609-12613 11. Cederholm-Williams, S. A., De Cock, F., Lijnen, H. R., and Collen, D. (1979) Eur. J. Biochem. 100, 125-32
19
12. Reed, G. L., Matsueda, G. R., and Haber, E. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 1114-8 13. Green, G. D., and Shaw, E. (1979) Anal. Biochem. 93, 223-6 14. Wiman, B., Boman, L., and Collen, D. (1978) Eur. J. Biochem. 87, 143-6 15. Huntington, J. A., Read, R. J., and Carrell, R. W. (2000) Nature 407, 923-6 16. Madison, E. L., Goldsmith, E. J., Gerard, R. D., Gething, M. J., and Sambrook, J. F. (1989) Nature 339, 721-4 17. Keyt, B. A., Paoni, N. F., Refino, C. J., Berleau, L., Nguyen, H., Chow, A., Lai, J., Pena, L., Pater, C., Ogez, J., and et al. (1994) Proc. Natl. Acad. Sci. U.S.A. 91, 3670-4 18. Wang, J., and Reich, E. (1995) Protein Sci. 4, 1768-79 19. Wiman, B., and Wallen, P. (1977) Thromb. Res. 10, 213-22 20. Wiman, B., Lijnen, H. R., and Collen, D. (1979) Biochim. Biophys. Acta. 579, 142-54 21. Wohl, R. C., Summaria, L., and Robbins, K. C. (1980) J. Biol. Chem. 255, 2005-13 22. Kraulis, P. J. (1991) J. Appl. Crystallog. 24, 946-950 23. Merritt, E. A., and Murphy, M. E. P. (1994) Acta. Crystallog. sect. D 50, 869-873
20
FIGURE LEGENDS
Fig 1. Crystal structure of microplasmin. A ribbon diagram was created to illustrate those loops chosen for chimerization from PDB entry 1BML (2). The side chains of the active site residues are shown. Sequence comparisons are shown between microplasminogen and factor D loops. This figure was produced using MOLSCRIPT and Raster3D (22,23).
Fig 2. Purification, cleavage, and activation of microplasminogen chimeras. A, Purified recombinant wild-type or chimeric loop 3, 5, 6, and 7 microplasminogens (1 µM total protein, 75 µl) were cleaved with urokinase immobilized on agarose for 0 or 30 min. at 37°C. The microplasminogens were analyzed by SDS-PAGE on 10% gels under reducing conditions and stained with Coomassie blue dye. The relative migration of a 30 kDa standard is shown at left. B, The various microplasminogens (50 nM) were activated by 1 U of urokinase and added to a solution of 0.5 mM S2251. Kinetics of activation were monitored at 405 nm over 20 minutes at 37 oC. C, To test activation by streptokinase, microplasminogens (50 nM) were mixed stoichiometrically with streptokinase and S2251 was added to a final concentration of 0.5 mM S2251 to start the reaction. The 20 minute kinetics were measured at 405 nm absorbance. D, Activation by staphylokinase was tested using 50 nM microplasminogens with excess staphylokinase and S2251 (0.5 mM). The kinetics were measured for 20 minutes at 405 nm.
Fig 3. Binding of microplasminogens to streptokinase. A, binding of an anti-plasminogen monoclonal antibody to immobilized wild-type and chimeric microplasminogens. The microplasminogens (25 µl, 20 µg/ml) or no protein (control) were adsorbed on microtiter plates for 1 hr. The nonspecific protein binding was blocked by 1% BSA for 1 hr. After washing, an
21
anti-plasminogen monoclonal antibody 340-11 was added for 1 hr. The bound antibody was detected by 125I-(sheep antimouse Ab) (50,000 cpm) followed by gamma-counting. B, microplasminogen chimeras bind to immobilized streptokinase. Streptokinase (10 µg/ml, 25 µl) was adsorbed to the microtiter plate for 1 hr, blocked as before with 1% BSA, and the microplasminogens (25 µl, 20 µg/ml) or no microplasminogen (control) were added to the wells for 1 hr. The bound primary antibody was detected by 125I-(sheep antimouse Ab) (50,000 cpm) followed by gamma-counting as before.
Fig 4. Inhibition of microplasmin or factor D by α2-antiplasmin. Wild-type or chimeric microplasmins (10 nM) were added to cuvettes containing S-2251 (0.5 mM) in assay buffer and the change in absorbance at 405 nm recorded at 0.3 min. intervals prior and after the addition of human α2-antiplasmin (120 nM). The rate of loss of enzyme activity was determined as described (11) and is expressed as log residual enzyme concentration versus time; the linear fits of the data are shown. The interaction of factor D with α2-antiplasmin was determined using a standard assay for factor D activity (13). Factor D (20 nM final) and 0.13 mM DTNB final, were added to Tris buffer (100 mM, pH 8.0) and incubated at 37oC for 3 minutes. Then Z-LysSBzl substrate was added to start the reaction and the absorbance was measured at 412 nm at 0.3 min. intervals before and after the addition of human α2-antiplasmin (13). The rate of loss of the enzyme activity was determined and expressed as before.
Fig 5. Fibrinolysis by chimeric microplasminogens. A, B fibrinolysis with purified fibrin clots. Fibrinogen (5 µl, 20 mg/ml) and 3 µl 125I-fibrinogen were mixed with 37 µl TBS and CaCl2 (2.5 µl, 0.4 M) and thrombin (2.5 µl, 1 unit/ml) were mixed and added to the fibrinogen solution to form the clot for 1 hr at 37oC. After 5-6 hrs incubation, the amount of residual clot 22
was measured by gamma-counting and the fractional fibrinolysis was determined. A, In "preactivation" experiments 100 U of urokinase were added to the microplasminogens in assay buffer and incubated for 1 hr at 37oC. The resulting microplasmins (200 µl) were then added as supernatant to the fibrin clots. B, In "synchronous" experiments, the mutant microplasminogens (0-1 µΜ) were added to the clots followed by the addition of 5 U of urokinase. C, D fibrinolysis of human plasma clots in human plasma. For plasma clot lysis experiments, human plasma (100 µl) was mixed with trace amounts of 125I-fibrinogen and clotted with CaCl2 (20 mM) and thrombin (5 µl, 0.05 U) for 1 hr at 37 C. The clots were suspended in 50 µl plasma as o
supernatant. C, In "pre-activation" experiments, 25 U of urokinase was added to the microplasminogens and incubated for 1 hr at 37oC. The microplasmins (50 µl, 0-1 µΜ) were then added to the plasma supernatant. D, In "synchronous" experiments, the mutant microplasminogens (50 µl, 0-1 µΜ) were added to the clots followed by 5 U of urokinase.
23
24
Table I. Strategy of loop chimerization. ___Pg loop____ No. 3
Sequence TRFGQ
Factor D Sequence LNGA
Primers* CTGAACGGGGCACACTTCTGTGGAGGCACC TGCCCCGTTCAGTCTAAGACTGACTTGCCAGGG
5
AHCLEKSPRPSSY
AHCLEDAADGKV
CTGGAGGACGCAGCAGATGGAAAGGTAAAGGT CATCCTGGGTG TACCTTTCCATCTGCTGCGTCCTCCAGGCAGTG GGCAGCAGT
6
AHQEVNLEPHV
AHSLSQPEPSK
TCACTGTCACAGCCAGAGCCATCAAAGCAGGA AATAGAAGTGTCTAGG CTTTGATGGCTCTGGCTGTGACAGTGAGTGTGC ACCCAGGATGACC
7
EPTRKD
HPDSQPDTIDHD
CACCCAGACTCACAGCCAGATACCATCGATCAC GATATTGCCTTGCTAAAG GTGATCGATGGTATCTGGCTGTGAGTCTGGGTG CAAGAACAGCCTAGACAC
*Sense primers are listed first followed by antisense primers.
25
Table II Kinetic Parameters for Amidolysis Enzyme microplasmin type Km (µM) wild-type 2030 ± 284 loop 3 2920 ± 850 loop 5 5980 ± 852 loop 7 3310 ± 705
Amidolytic Parameters* kcat(s-1) kcat/Km (µM-1s-1) 20.5 ± 1.5 0.010 35.2 ± 5.1 0.012 27.0 ± 2.5 0.005 17.5 ± 3.7 0.005
*Experiments were carried out at 37°C in a total volume of 300 µl and kinetic parameters were determined as described in Experimental Procedures. The values represent the mean + SE.
26
Figure 1
Loop 6 Loop 3 micro-Pg TRFGQ factor D L NGA
micro-Pg AHQEVNLEPHV factor D AHSLSQPEPSK
Loop 5 micro-Pg AHCLEKSPRPSSY factor D AHCLEDAADGKV
Loop 7 micro-Pg EPTR KD factor D HPDSQPDTIDHD
30
Figure 2
A
Pg Activation (A405/min) 50 C
Wild-type Loop 3 Loop 5 Loop 6 Loop 7
MW micro-Pg micro-Pm
300
30
0
30
0
30 0
30
0
30
Cleavage by Urokinase (min.)
40 30 20 10 0
Pg Activation (A405/min) 25
B
Loop 5
Loop 7
micro-Pg
No Pg
Pg Activation (A405/min) 30
D
25
20
20
15
15
10
10
5 0
Loop 3
5 Loop 3 Loop 5 Loop 7 Loop 6 micro-Pg No Pg
0
Loop 3
Loop 5 Loop 7 micro-Pg No Pg Glu-Pg
Figure 3
cpm, micro-Pg binding 6000 A
cpm, micro-Pg binding
B 4000
4000 2000
2000
0
Wild-type
Loop 3
Loop 5
Loop 7
0
Control
32
Wild-type Loop 3
Loop 5
Loop 7
Control
Figure 4
Residual Enzyme Activity (%) 100
Factor D
loop 7 chimera loop 5 chimera loop 3 chimera
wild type micro-Pg 10 0
5
10
15
Time (min)
33
Figure 5
fibrinolysis, (%) 100 A 80
fibrinolysis, (%) 100 B
Glu-Pg wild-type micro-Pg loop 3 chimera loop 5 chimera loop 7 chimera No Pg
80
60
60
40
40
20
20
0
0 10 [plasmin(ogen)], nM
10
100
fibrinolysis, (%) 100 C
100 [plasmin(ogen)], nM
1000
fibrinolysis, (%) 100 D
80
80
60
60 2
40
40
20
20
0
0 10
100 [plasmin(ogen)], nM
1000
10
100 [plasmin(ogen)], nM
34
1000