Construction and Functional Evaluation of a Single-Chain Antibody Fusion Protein With Fibrin Targeting and Thrombin Inhibition After Activation by Factor Xa
Background—Recombinant technology was used to produce a new anticoagulant that is preferentially localized and active at the site of the clot.
Methods and Results—The variable regions of the heavy and light chains of a fibrin-specific antibody were amplified by polymerase chain reaction (PCR) with hybridoma cDNA. To obtain a functional single-chain antibody (scFv), a linker region consisting of (Gly4Ser)3 was introduced by overlap PCR. After the scFv clones were ligated with DNA encoding the pIII protein of the M13 phage, high-affinity clones were selected by 10 rounds of panning on the Bβ15-22 peptide of fibrin (β-peptide). Hirudin was genetically fused to the C-terminus of the variable region of the light chain. To release the functionally essential N-terminus of hirudin at the site of a blood clot, a factor Xa recognition site was introduced between scFv59D8 and hirudin. The fusion protein was characterized by its size on SDS-PAGE (36 kDa), by Western blotting, by its cleavage into a 29-kDa (single chain alone) and 7-kDa (hirudin) fragment, by its binding to β-peptide, and by thrombin inhibition after Xa cleavage. Finally, the fusion protein inhibited appositional growth of whole blood clots in vitro more efficiently than native hirudin.
Conclusions—A fusion protein was constructed that binds to a fibrin-specific epitope and inhibits thrombin after its activation by factor Xa. This recombinant anticoagulant effectively inhibits appositional clot growth in vitro. Its efficient and fast production at low cost should facilitate a large-scale evaluation to determine whether an effective localized antithrombin activity can be achieved without systemic bleeding complications.
Inhibition of thrombin by either the indirect thrombin inhibitor heparin or direct thrombin inhibitors such as hirudin reduces thrombus formation after arterial injury in animal models1 and in humans with unstable coronary syndromes.2 3 Furthermore, thrombin inhibitors potentiate fibrinolysis induced by plasminogen activators.4 Several animal experiments demonstrated that hirudin is more effective than heparin in preventing platelet-dependent arterial thrombosis, rethrombosis after reperfusion, and thrombus growth.4 5 6 7 However, clinical trials with direct thrombin inhibitors have only been partially successful.8 High concentrations of hirudin were very effective in inhibiting thrombin but are associated with frequent hemorrhagic complications.4 A strategy for circumventing this problem is the targeting of hirudin to fibrin.
Fibrin targeting can be achieved with the monoclonal antibody (mAb) 59D8, which selectively binds to the amino-terminus of the fibrin β-chain that becomes exposed after cleavage of fibrinopeptide B by thrombin.9 Because exposure of this epitope is an early event in the conversion of fibrinogen to fibrin, it is likely that mAb 59D8 accumulates at sites of high thrombin activity, such as a developing arterial clot.9 Coupling of mAb 59D8 to plasminogen activators resulted in enhanced thrombolytic potency and specificity in vitro and in vivo.10 11 A chemical conjugate between hirudin and 59D8 effectively inhibited fibrin deposition on experimental clots12 and demonstrated potent antithrombotic activity in nonhuman primates.13 Nevertheless, chemical coupling of hirudin to mAbs has several limitations, the major ones being low yield and loss of hirudin activity.12 We tried to bypass these limitations by the use of recombinant technology.
Because hirudin needs a free amino- as well as a free carboxy-terminus for antithrombin activity,14 a direct fusion at the termini of hirudin was expected to result in a functional loss. Therefore, a factor Xa cleavage site was introduced between mAb 59D8 and hirudin. This cleavage site was chosen for 2 reasons. First, factor Xa cleaves at the C-terminus of its recognition sequence (Ile-Glu-Gly-Arg) and thus liberates the free amino-terminus of hirudin. Second, factor Xa is a major part of the activated coagulation system at the site of arterial clots15 and may therefore allow a preferential liberation of functional hirudin at the clot. Without an activated coagulation system, the fusion protein would be inert. However, as a clot develops, the combination of fibrin targeting and dependence on cleavage by factor Xa could result in an effective thrombin inhibition at the clot without systemic anticoagulation.
Materials and Cells
Horseradish peroxidase (HRP)–conjugated sheep anti-M13 mAb and mouse anti-E-tag mAb were obtained from Pharmacia, mouse anti-c-myc mAb from Cambridge Research Biochemicals, and goat anti-mouse HRP-conjugated polyclonal antibody (Ab) from Dianova. Bβ15-22, also termed β-peptide, with the amino acid sequence Gly-His-Arg-Pro-Leu-Asp-Lys, was purchased from MWG Biotech. Hirudin was a gift of Knoll AG (Ludwigshafen, Germany). The hybridoma secreting the fibrin-specific mAb 59D8 was generated as described previously,9 and cells were grown on DMEM, 10% fetal calf serum, 2 mmol/L l-glutamine, penicillin (10 IU/mL), and streptomycin (10 μg/mL) (all from Gibco) with 5% CO2 at 37°C.
Construction of a Functional Single-Chain Antibody
cDNA of 59D8 hybridoma cells was prepared with mRNA purification columns (oligo-dT) and M-MuLV (both from Pharmacia). Amplification of the antibody variable regions and the insertion of the linker sequence were achieved by polymerase chain reaction (PCR). Primer mixes that contained sequences from conserved regions of the variable regions of the heavy (VH) and light (VL) chains were obtained from Pharmacia. The linker sequence (Gly4Ser)3 was inserted by the addition of a linker fragment.
Clone Selection With the M13 Phage System
The PCR products encoding the functional single-chain antibody fragments (scFv) were cloned into the vector pCANTAB5E (Pharmacia). In this vector, an amber stop codon allows expression of soluble scFv in the nonsuppressor Escherichia coli strain HB2151 and display of scFv on the M13 phage surface by fusion to the pIII adsorption protein in the suppressor strain TG1. The supernatant of TG1 clones was used for the following panning procedure: A tissue culture flask with a surface area of 25 cm2 was coated with 50 μg of β-peptide at 4°C overnight, washed 5 times with PBS, and blocked for 2 hours at 37°C with 2% nonfat dry milk in PBS. The phage-containing supernatant was added and incubated for 2 hours at 37°C. Nonadhering phages were removed by washing 20 times with PBS. A TG1 culture was added to the flask for reinfection with bound phages and incubated for 1 hour at 37°C at 250 rpm. This panning procedure was repeated 9 times. Positive clones were tested for phage binding on immobilized β-peptide by use of an HRP-conjugated anti-M13 sheep mAb. The best binding clones with the expected fragment size (≈750 bp) were used to transform HB2151. Periplasmic extracts from the individual clones were analyzed for binding to immobilized β-peptide by an anti-E-tag mAb.
Cloning of scFv59D8 Into the Expression Vector pHOG21, Fusion With the Factor Xa Recognition and Hirudin Sequences, and Transfer to pOPE51
DNA of scFv clone 33 was cloned into pHOG2116 , mutated at position 6 to glutamine,17 and cloned into pOPE5118 (Figure 2⇓). DNA coding for hirudin (Biermann) was used as a template for PCR with the sense primer CAGCAAGATCTAAACTCAAGCGGC-ATCGAAGGTCGTGTTGT-TTACACCGACTGTACTG and the antisense primer AGATGATCTAGAGGATCCTTACTGCAGAT-ATTCTTCTGGG. The factor Xa recognition sequence (bold) and the restriction site BglII (underlined) are encoded by the sense primer and the restriction site XbaI (underlined) by the antisense primer. The ligation products were transformed into XL1-blue.
Preparation of scFv From Inclusion Bodies
From overnight cultures of XL1-blue, 250 μL was transferred to 5 mL of LB medium containing 100 μg/mL ampicillin and 100 mmol/L glucose and incubated at 37°C and 280 rpm until an OD600 nm of 0.8 was reached. Protein expression was induced by addition of isopropyl-β-d-thiogalactopyranoside (IPTG; 20 μmol/L) and cultured at room temperature for 4 hours. Cells were then centrifuged (6000 rpm, 15 minutes) and resuspended in 165 μL of ice-cold buffer (50 mmol/L Tris-HCl, 100 mmol/L NaCl, 1 mmol/L EDTA, pH 7.0). After freezing and thawing, the sample was centrifuged (12 000g, 4°C, 30 minutes), resuspended in 500 μL of ice-cold TE buffer (10 mmol/L Tris-HCl, 1 mmol/L EDTA, pH 7.4), and incubated for 1 hour at room temperature. Lysozyme (Boehringer Mannheim) was then added to a final concentration of 200 μg/mL, and the incubation was continued for 1 hour, followed by the addition of NaCl (0.5 mol/L) and Triton-X-100 (2.5%) and a final incubation for 1 hour. After centrifugation (12 000g, 4°C, 1 hour), the pellet was washed twice with 3 mol/L urea, and 50 mmol/L Tris-HCl (pH 7.0) and finally solubilized by rotation overnight at 4°C in 250 μL of 6 mol/L GdHCl, 100 mmol/L Tris-HCl, pH 7.0. After centrifugation (12 000g, 4°C, 1 hour), the supernatant was dialyzed against TA buffer (0.4 mol/L arginine-HCl in 100 mmol/L Tris-HCl, pH 7.0).
ELISA With Immobilized β-Peptide
Microtiter plates were coated with 1 μg of β-peptide or the control peptide GRGDSP in 100 μL of 0.05 mol/L Na2CO3 (pH 9.6) overnight at 4°C. The plate was then washed 4 times with PBS and blocked with blocking buffer (2% nonfat dry milk in PBS) for 2 hours at room temperature. Samples (100 μL) were incubated for 2 hours at room temperature. After the plate was washed 5 times with PBS, 100 μL of mAb solution (either anti-M13 mAb, anti-c-myc mAb [both 1 to 5000 diluted], or 1 μg/mL anti-E-tag mAb) in blocking buffer was added and incubated for 2 hours at room temperature. Either ABTS (Sigma) or TMB solution (Biorad) was used, and samples were measured at wavelengths of 405 or 655 nm, respectively.
Purification by Immobilized Metal Affinity Chromatography and Ionic Exchange Chromatography of scFv59D8 Expressed in the pHOG21 Vector
A 10-mL column of chelating sepharose (Pharmacia) was equilibrated with 0.1 mol/L CuSO4. The samples were loaded in 50 mmol/L Tris-HCl, 1 mol/L NaCl, pH 7.0. After the column was washed with 200 mL of 50 mmol/L Tris-HCl, 1 mol/L NaCl, 50 mmol/L imidazole, pH 7.0, the bound scFv construct was eluted with 40 mL of 50 mmol/L Tris-HCl, 1 mol/L NaCl, 250 mmol/L imidazole, pH 7.0. The eluate was dialyzed against PBS buffer and further purified by ionic exchange chromatography on a MonoS column (Pharmacia) in 50 mmol/L MES buffer (pH 6.5) with a 0 to 1 mol/L NaCl gradient.
Affinity Chromatography of scFv and Factor Xa Cleavage
Coupling of β-peptide to sepharose was performed as described previously.10 Columns containing β-peptide–conjugated sepharose were loaded and washed with TA buffer. Bound protein was eluted by 0.1 mol/L glycine, pH 2.8, and 1-mL fractions were collected and adjusted to pH 7.0 with 0.5 mol/L Tris buffer. Fractions containing significant amounts of product were pooled and dialyzed against TA buffer. For factor Xa cleavage, typically 150 μg of scFv59D8-Xa-hirudin (2 μg/μL) was cleaved by 15 μg of factor Xa (1 μg/μL, Boehringer Mannheim) for various times in 50 mmol/L Tris-HCl, 100 mmol/L NaCl, and 1 mmol/L CaCl2 (pH 8.0) at room temperature.
Measurement of Thrombin Inhibition by scFv59D8-Xa-Hirudin After Factor Xa Cleavage
Inhibition of thrombin was determined by cleavage of the chromogenic substrate S-2238 (Chromogenix). After factor Xa (0.1 μg/μL) cleavage (5 hours, room temperature), 20 μL of thrombin solution (human thrombin, 2.5 U/mL; Sigma) was added to 100 μL of sample in assay buffer (20 mmol/L sodium dihydrogen carbonate, 0.15 mol/L NaCl, and 0.1% bovine serum albumin, pH 7.4) and incubated at room temperature for 10 minutes. S-2238 (50 μL, 0.833 mg/mL) was then added, and after 10 minutes of incubation, the reaction was stopped by the addition of 50 μL of 20% acetic acid. Absorbance was measured at 405 nm.
Whole Blood Clot Assay
Except for minor modifications, clot assays were performed as described previously.12 Clots were initiated by the addition of CaCl2 (16.6 mmol/L) and 2.5 vol% of Actin7 FS-activated PTT reagent (Dade International) to anticoagulated blood (citric acid, 11 mmol/L) from healthy volunteers. Whole blood was immediately drawn into a silicone tubing (4-mm inner diameter), and clots were allowed to form at 37°C for 1 hour. Quantification of clot size was performed by labeling of blood with 125I-fibrinogen (Amersham) to a final activity of 37 500 cpm/mL. The silicon tubing was cut into 1.5-cm fragments, and the formed clots were extruded and washed 5 times in TA buffer. In each assay, the starting size of clots chosen for further experiments varied not more than ±5%. Appositional clot growth was evaluated by the incubation of clots for 10 hours at 37°C on a rotator (60 rpm). Clots were incubated in recalcified citrated whole blood (trace labeled with 125I-fibrinogen at a final activity of 112 500 cpm/clot) either with the addition of native hirudin or scFv59D8-Xa-hirudin or without addition. The clots were then washed 10 times with TA buffer, and inhibition of appositional clot growth was evaluated on a γ-counter.
To construct a single-chain antibody directed against fibrin (scFv), mRNA was prepared from 5×107 59D8 hybridoma cells and reverse transcribed with an oligo-dT primer. The variable regions of the heavy and light chains (VH and VL, respectively) were amplified by polymerase chain reaction (PCR) with primers that anneal to conserved regions at the 5′- and 3′-ends of the variable regions. PCR products of 348 and 339 bp were obtained for VH and VL, respectively (Figure 1⇓). After addition of a (Gly4Ser)3 linker by fusion PCR, the scFv product (Figure 1⇓) was cloned into pCANTAB5E (Figure 2⇓) for phage display of the scFv clones. After 10 rounds of panning on immobilized β-peptide, 144 phage clones were tested for binding to the β-peptide by phage ELISA with a horseradish peroxidase (HRP)–conjugated mAb against M13 and by DNA restriction analysis. Twenty-four clones with strong binding to β-peptide and the expected size of the insert were used for transformation of HB2151. The binding properties of soluble scFvs secreted into the supernatant were compared by ELISA with a mouse anti-E-tag mAb and a secondary HRP-conjugated goat anti-mouse antibody (Ab) (Figure 3⇓). Clone 33 demonstrated the best binding properties (Figure 3⇓) and was therefore chosen for sequencing and further characterization. Complementarity determining regions and framework regions of both variable regions and the linker region are highlighted in Figure 4⇓.
For enhanced expression and purification of soluble scFv59D8, clone 33 was transferred to the expression vector pHOG21 (Figure 2⇑). This plasmid contains a tag sequence coding for 6 histidine residues at the scFv C-terminus, thus facilitating purification by immobilized metal affinity chromatography. However, an additional purification step with ion-exchange chromatography was necessary to obtain a pure product (Figure 5A⇓). An analysis of several eluted fractions by SDS-PAGE is shown in Figure 5B⇓. The functional integrity of the highly purified scFv59D8 was tested by binding on immobilized β-peptide (Figure 5C⇓). The yield of purified scFv59D8 was 0.2 mg from 1 L of bacterial culture.
To further increase the yield, glutamic acid at position 6 of the heavy chain was mutated by PCR to glutamine, because this substitution has been shown to give increased yields of scFvs.17 Indeed, the yield of functional soluble scFv59D8 was increased ≈4 times by this single amino acid substitution.
The factor Xa recognition sequence and the hirudin sequences were fused to the scFv59D8 by PCR. However, only a low yield of soluble fusion protein was obtained with the pHOG21 expression vector. This was probably due to the high cysteine content (10%) of hirudin that might interfere with the folding process of soluble scFv. To obtain higher levels of the scFv59D8-Xa-hirudin fusion protein, we chose the expression vector pOPE51,17 which facilitates the production of large amounts of fusion proteins as inclusion bodies in the periplasmic space. When this expression system was used, up to 10 mg of highly purified scFv59D8-Xa-hirudin could be obtained from a bacterial culture of 5 L.
The fusion protein scFv59D8-Xa-hirudin was analyzed by SDS-PAGE and tested for its binding to β-peptide and its susceptibility to factor Xa cleavage. The molecular weight of the intact fusion protein scFv59D8-Xa-hirudin was 36 kDa, that of the cleavage product scFv59D8-Xa was 29 kDa, and that of hirudin was 7 kDa (Figure 6⇓).
We evaluated the functional characteristics of scFv59D8-Xa-hirudin by measuring its binding to β-peptide and by determining its antithrombin activity after binding to β-peptide. Binding to β-peptide was comparable to the binding of equimolar amounts of the Fab′ fragment of the original mAb 59D8 as measured in ELISA (Figure 7A⇓). The antithrombin activity of the scFv59D8-Xa-hirudin was determined in the presence and absence of factor Xa. ScFv59D8-Xa-hirudin was allowed to bind to β-peptide, and the nonbound fusion protein was washed away. The binding function and antithrombin activity of bound scFv59D8-Xa-hirudin could thus be evaluated simultaneously. The uncleaved scFv59D8-Xa-hirudin revealed no antithrombin activity, whereas scFv59D8-Xa-hirudin in the presence of factor Xa demonstrated marked antithrombin activity (Figure 7B⇓).
The ability of the fusion protein to inhibit clot growth was tested in a whole blood clot assay. Native hirudin and scFv59D8-Xa-hirudin were directly compared for their ability to inhibit appositional clot growth. ScFv59D8-Xa-hirudin was able to inhibit clot growth significantly better than native hirudin (Figure 8⇓).
Fibrin targeting allows for local enrichment of fibrinolytic agents at the site of the thrombus at low systemic concentrations and thus represents a strategy to increase fibrinolytic potency.10 11 Furthermore, a chemical conjugate of the fibrin-specific mAb 59D8 and the direct thrombin inhibitor hirudin inhibited fibrin deposition on experimental clots12 and demonstrated an increase in antithrombotic potency in baboons.13 To increase the yield and activity of antibody-targeted hirudin and to further improve the risk/benefit ratio of anticoagulation, we have developed a recombinant fusion molecule consisting of an antifibrin single-chain antibody and hirudin. In addition to fibrin targeting, the generation of a free N-terminus, which is essential for the antithrombin activity of hirudin, forms the basis of a unique pharmacological approach. By the addition of the factor Xa recognition sequence, the fusion protein inhibited thrombin only in the presence of factor Xa. Because this factor is part of the activated clotting system and is an important determinant of the procoagulant activity of whole blood clots and arterial thrombi,15 the designed fusion protein represents an anticoagulant that promises to be preferentially active at the site where it is needed.
The potential therapeutic use of single-chain antibody fusion proteins has several major advantages. The variable regions of antibodies comprise the smallest fragments containing a complete antibody binding site, and fusion molecules can be created without loss in binding function of the scFv. Therefore, scFvs are attractive tools for the targeting of drugs, toxins, and radionuclides. The fusion protein scFv59D8-Xa-hirudin with the small molecular size of 36 kDa is expected to be only minimally, if at all, immunogenic, and its small size may improve thrombus accessibility and penetration. It can be produced in bacteria in large amounts, in a short time, and at low cost, and it can be highly purified by affinity chromatography with β-peptide columns, thus providing an ideal situation for drug preparation on a large scale.
Fibrin is an obvious target to concentrate antithrombotic or fibrinolytic agents at the clot. Sufficient amounts of fibrin are present even in platelet-rich thrombi.13 In addition to mAb 59D8, the mAb MA-15C5, directed against the fragment d-dimer of cross-linked human fibrin, has been used successfully to target plasminogen activators to clots.19
Several reports imply that direct thrombin inhibitors may be superior to heparin.3 4 5 6 7 8 This could be explained by a number of distinct mechanisms. In contrast to heparin, which only inhibits thrombin as a soluble molecule, hirudin can also inhibit thrombin that is bound to the clot or to soluble fibrin degradation products.20 21 Heparin binds to various other partners besides thrombin and is thereby inhibited.22 In contrast to heparin, hirudin has no natural inhibitors.7 Furthermore, hirudin can displace thrombin from platelet thrombin receptors.7 In an experimental study, hirudin but not heparin was even able to dissolve preexisting mural thrombi.7 Nevertheless, the experimental advantages of hirudin compared with heparin have not been reflected by superior clinical performance. Bleeding complications with higher doses of hirudin appear to be the major limitation.4 Fusion proteins, such as the one described, provide a promising new development based on the strategy of targeting to and activation at the existing or developing thrombus. This may result in highly efficient inhibition of thrombin and at the same time in fewer bleeding complications.
In summary, a fusion protein has been developed that combines fibrin targeting and antithrombin activity after activation by factor Xa. This recombinant anticoagulant promises to be active only when and where it is needed, thus providing a pharmacological approach that may facilitate an effective anticoagulation without systemic bleeding complications.
This study was supported by the Deutsche Forschungsgemeinschaft (SFB 320) and by the Medical Faculty of the University of Heidelberg with grants to Dr Peter, Dr Little, and Dr Bode. The expert technical assistance of Simone Bauer and Armin Keller is gratefully acknowledged.
Presented in part at the annual meeting of the American College of Cardiology, New Orleans, La, March 7–10, 1999, and published in abstract form (J Am Coll Cardiol. 1999;33:2A).
- Received July 21, 1999.
- Revision received September 21, 1999.
- Accepted October 7, 1999.
- Copyright © 2000 by American Heart Association
Meyer BJ, Fernandez-Ortiz A, Mailhac A, Falk E, Badimon L, Michael AD, Chesebro JH, Fuster V, Badimon JJ. Local delivery of r-hirudin by a double-balloon perfusion catheter prevents mural thrombosis and minimizes platelet deposition after angioplasty. Circulation. 1994;90:2474–2480.
Topol EJ, Fuster V, Harrington RA, Califf RM, Kleiman NS, Kereiakes DJ, Cohen M, Chapekis A, Gold HK, Tannenbaum MA, Rao AK, Debowey D, Schwartz D, Henis M, Chesebro J. Recombinant hirudin for unstable angina pectoris: a multicenter, randomized angiographic trial. Circulation. 1994;89:1557–1566.
Cannon CP, Braunwald E. Hirudin: initial results in acute myocardial infarction, unstable angina and angioplasty. J Am Coll Cardiol. 1995;25:30S–37S.
Loscalzo J. Thrombin inhibitors in fibrinolysis: a Hobson’s choice of alternatives. Circulation. 1996;94:863–865.
Rigel DF, Olson RW, Lappe RW. Comparison of hirudin and heparin as adjuncts to streptokinase thrombolysis in a canine model of coronary thrombolysis. Circ Res. 1993;72:1091–1102.
Mruk JS, Zoldhelyi P, Webster MWI, Heras M, Grill DE, Holmes DR, Fuster V, Chesebro JH. Does antithrombotic therapy influence residual thrombus after thrombolysis of platelet-rich thrombus? Effect of recombinant hirudin, heparin, and aspirin. Circulation.. 1996;93:792–799.
Meyer BJ, Badimon JJ, Chesebro JH, Fallon JT, Fuster V, Badimon L. Dissolution of mural thrombus by specific thrombin inhibition with r-hirudin: comparison with heparin and aspirin. Circulation. 1998;97:681–685.
Chesebro JH. Direct thrombin inhibition superior to heparin during and after thrombolysis. Circulation. 1997;96:2118–2120.
Hui KY, Haber E, Matsueda GR. Monoclonal antibodies to a synthetic fibrin-like peptide bind to human fibrin but not fibrinogen. Science. 1983;222:1129–1132.
Bode C, Runge MS, Schönermark S, Eberle T, Newell BA, Kübler W, Haber E. Conjugation to antifibrin Fab′ enhances fibrinolytic potency of single-chain urokinase plasminogen activator. Circulation. 1990;81:1974–1980.
Runge MS, Quertermous T, Zavodny PJ, Love TW, Bode C, Freitag M, Shaw S, Huang PL, Chou C, Mullins E, Schnee JM, Savard C, Rothenberg ME, Newell BA, Matsueda GR, Haber E. Recombinant chimeric plasminogen activator with high affinity for fibrin has increased thrombolytic potency in vitro and in vivo. Proc Natl Acad Sci U S A. 1991;88:10337–10341.
Bode C, Hudelmayer M, Mehwald P, Bauer S, Freitag M, von Hodenberg E, Newell JB, Kübler W, Haber E, Runge MS. Fibrin-targeted recombinant hirudin inhibits fibrin deposition on experimental clots more efficiently than recombinant hirudin. Circulation.. 1994;90:1956–1963.
Bode C, Hanson SR, Schmedtje JF, Haber E, Mehwald P, Kelly AB, Harker LA, Runge MS. Antithrombotic potency of hirudin is increased in nonhuman primates by fibrin targeting. Circulation. 1997;95:800–804.
Lazar JB, Winant RC, Johnson PH. Hirudin: amino-terminal residues play a major role in the interaction with thrombin. J Biol Chem. 1991;266:685–688.
McKenzie CR, Abendschein DR, Eisenberg PR. Sustained inhibition of whole-blood clot procoagulant activity by inhibition of thrombus-associated factor Xa. Arterioscler Thromb Vasc Biol. 1996;16:1285–1291.
Kipriyanov SM, Moldenhauer G, Martin AC, Kupriyanova OA, Little M. Two amino acid mutations in an anti-human CD3 single chain Fv antibody fragment that affect the yield on bacterial secretion but not the affinity. Protein Eng. 1997;10:445–453.
Holvoet P, Laroche Y, Lijnen HR, Van Cauwenberge R, Demarsin E, Brouwers E, Matthyssens G, Collen D. Characterization of a chimeric plasminogen activator consisting of a single-chain Fv fragment derived from a fibrin fragment d-dimer-specific antibody and a truncated single-chain urokinase. J Biol Chem. 1991;266:19717–19724.
Weitz JI, Hudoba M, Massel D, Maraganore J, Hirsh J. Clot-bound thrombin is protected from inhibition by heparin-antithrombin III but is susceptible to inactivation by antithrombin III-independent inhibitors. J Clin Invest. 1990;86:385–391.
Weitz JI, Leslie B, Hudoba M. Thrombin binds to soluble fibrin degradation products where it is protected from inhibition by heparin-antithrombin but susceptible to inactivation by anti-thrombin-independent inhibitors. Circulation. 1998;97:544–552.