Construction and in vitro testing of a novel fab-hirudin-based fusion protein that targets fibrin and inhibits thrombin in a factor xa-dependent manner

Engineering a "three-in-one" hirudin prodrug to reduce bleeding risk: A proof-of-concept study

An ideal anticoagulant should have at least three properties including targeted delivery to the thrombosis site, local activation or releasing to centralize the anti-thrombosis effects and thus reduce the bleeding risks, and long persistence in circulation to avoid repeated administration. In the present study, we sought to test a "three-in-one" strategy to design new protein anticoagulants. Based on these criteria, we constructed two hirudin prodrugs, R824-HV-ABD and ABD-HV-R824. The R824 peptide can bind phosphatidylserine on the surface of the procoagulant platelets and thus guide the prodrug to the thrombosis sites; albumin-binding domain (ABDs) can bind the prodrug to albumin, and thereby increase its persistence in circulation; the hirudin (HV) core in the prodrug is flanked by factor Xa recognition sites, thus factor Xa at the thrombosis site can cleave the fusion proteins and release the activated hirudin locally. Hirudin prodrugs were able to bind with procoagulant platelets and human serum albumin in vitro with high affinity, targeted concentrated and prevented the formation of occlusive thrombi in rat carotid artery injury model. Their effective time was significantly extended compared to native hirudin, and R824-HV-ABD showed a significantly improved half-life of about 24 h in rats. The bleeding time of prodrug-treated mice was much shorter than that of hirudin-treated mice. The results from the proof-of-concept studies, for the first time, demonstrate that "three-in-one" prodrug strategy may be a good solution for protein or peptide anticoagulants to reduce their bleeding risks.

Predicting the metabolic characteristics of neorudin, a novel anticoagulant fusion protein, in patients with deep vein thrombosis

INTRODUCTION

Recombinant neorudin (EPR-hirudin, EH) is an inactive prodrug that is converted to its active metabolite, hirudin variant 2-Lys47 (HV2), at the thrombus site. We aimed to investigate the mechanism underlying site-selective bioconversion of EH to HV2 at the thrombus target site and metabolic transformation of EH in patients with deep vein thrombosis (DVT).

MATERIALS AND METHODS

Metabolites in healthy volunteer plasma and urine after intravenous administration of EH were determined to elucidate how EH was metabolised after releasing HV2 at the target site in patients with DVT. After intravenous administration of EH in rats with venous thrombosis, the concentrations of EH in the blood and thrombus and the antithrombotic activity of EH were measured to predict whether EH could release HV2 at the thrombus site to exert anticoagulant effect in patients with DVT.

RESULTS

In healthy volunteers, EH and HV2 were predominantly excreted in the urine. Nine EH metabolites and ten HV2 metabolites truncated at the C-terminal were identified as N-terminal fragments, and these had the same cleavage sites. In rats with venous thrombosis, the area under the curve ratio of HV2 between the thrombus and blood was 29.5. The weight of wet thrombus was decreased with the production of HV2 by the cleavage of EH. The prothrombin time (PT) and prothrombin time (TT) changed proportionally to the concentration of EH and HV2 in the blood.

CONCLUSION

EH selectively accumulates and releases HV2 in the thrombus to exert antithrombotic effects, thus lowering the bleeding risk. Moreover, after conversion, EH may follow the same metabolic profile as that of HV2 in patients with DVT.

Evaluating prodrug characteristics of a novel anticoagulant fusion protein neorudin, a prodrug targeting release of hirudin variant 2-Lys47 at the thrombosis site, by means of in vitro pharmacokinetics

Recombinant neorudin (EPR-hirudin, EH), a low-bleeding anticoagulant fusion protein, is an inactive prodrug designed to be converted to the active metabolite, hirudin variant 2-Lys47 (HV2), locally at the thrombus site by FXa and/or FXIa, following activation of the coagulation system. Our aim was to evaluate the prodrug characteristics of EH by comparing the biotransformation of EH and HV2 in biological matrices, including rat blood, liver, and kidney homogenates, demonstrating the cleavage of EH to HV2 by FXa and FXIa, and comparing the conversion of EH to HV2 between fresh whole blood and whole-blood clot homogenate, using ultra-performance liquid chromatography-mass spectrometry (UPLC-MS/MS). Both EH and HV2 were stable in blood and unstable in the liver and kidney homogenates. Eight EH metabolites and eight HV2 metabolites identified as N-terminal fragments were found in the liver and kidney. C-terminal proteolysis is therefore the major metabolic pathway, with serine/cysteine carboxypeptidases and metallocarboxypeptidases being responsible for the degradation of EH and HV2 in the liver and kidney, respectively. EH was cleaved to release HV2 by FXIa. Higher levels of HV2 were produced from EH in the whole-blood clot homogenate, in which the coagulation system was activated compared with those in fresh whole blood. In conclusion, the metabolism of EH and HV2 shares the same cleavage pattern, and EH is transformed into HV2 when the coagulation system is activated, where FXIa is a specific enzyme. Our in vitro study revealed the anticipated prodrug characteristics of EH newly designed as an inactive prodrug of hirudin.

Delayed targeting of CD39 to activated platelet GPIIb/IIIa via a single-chain antibody: breaking the link between antithrombotic potency and bleeding?

The ecto-nucleoside triphosphate diphosphohydrolase CD39 represents a promising antithrombotic therapeutic. It degrades adenosine 5'-diphosphate (ADP), a main platelet activating/recruiting agent. We hypothesized that delayed enrichment of CD39 on developing thrombi will allow for a low and safe systemic concentration and thus avoid bleeding. We use a single-chain antibody (scFv, specific for activated GPIIb/IIIa) for targeting CD39. This should allow delayed enrichment on growing thrombi but not on the initial sealing layer of platelets, which do not yet express activated GPIIb/IIIa. CD39 was recombinantly fused to an activated GPIIb/IIIa-specific scFv (targ-CD39) and a nonfunctional scFv (non-targ-CD39). Targ-CD39 was more effective at preventing ADP-induced platelet activation than non-targ-CD39. In a mouse carotid artery thrombosis model, non-targ-CD39, although protective against vessel occlusion, was associated with significant bleeding on tail transection. In contrast, targ-CD39 concentrated at the thrombus site; hence, a dose ∼10 times less of CD39 prevented vessel occlusion to a similar extent as high-dose non-targ-CD39, without prolonged bleeding time. An equimolar dose of non-targ-CD39 at this low concentration was ineffective at preventing vessel occlusion. Thus, delayed targeting of CD39 via scFv to activated platelets provides strong antithrombotic potency and yet prevents bleeding and thereby promotes CD39 toward clinical use.

Locally activity-released bifunctional fusion protein enhances antithrombosis and alleviates bleeding risk

Despite the fact that lytic therapy of thromboembolic disorder has been achieved, reocclusion of the damaged vessels and bleeding complication frequently reduce the therapeutic effect. In order to prevent the vessel reocclusion and enhance the therapeutic effect, combining the anticoagulant with the thrombolytic was assumed. Herein, we propose that restraining but locally releasing anticoagulant activity in the vicinity of thrombus is a way to alleviate the bleeding risk. A bifunctional fusion protein, termed as SFH (Staphylokinase (SAK) linked by FXa recognition peptide at N-terminus of Hirudin (HV)), was designed. SFH retained thrombolytic activity but no anticoagulant activity in thrombus-free blood due to the extension of the N-terminus of HV. However, it could locally liberate intact HV and exhibit anticoagulant activity when FXa or fresh thrombus was present. At equimolar dose, both improved antithrombotic and thrombolytic effects of SFH were observed in kappa-carrageenin inducing mouse-tail thrombosis model and rat inferior vena cava thrombosis model, respectively. Moreover, we observed significantly lower bleeding risk in mice and rats treated with SFH than with the mixture of SAK and HV with monitoring TT (P < 0.01), aPTT (P < 0.05) and PT (P < 0.05), and bleeding time (P < 0.05). In conclusion, SFH is a promising bifunctional therapeutic candidate with lower bleeding risk.

Targeting Ligand-Induced Binding Sites on GPIIb/IIIa via Single-Chain Antibody Allows Effective Anticoagulation Without Bleeding Time Prolongation

Objective— Therapeutic anticoagulation is widely used, but limitations in efficacy and bleeding complications cause an ongoing search for new agents. However, with new agents developed it seems to be an inherent problem that increased efficiency is accompanied by an increase in bleeding complications. We investigate whether targeting of anticoagulants to activated platelets provides a means to overcome this association of potency and bleeding.

Methods and Results— Ligand-induced binding sites (LIBS) on fibrinogen/fibrin-binding GPIIb/IIIa represent an abundant clot-specific target. We cloned an anti-LIBS single-chain antibody (scFv anti-LIBS ) and genetically fused it with a potent, direct factor Xa (fXa) inhibitor, tick anticoagulant peptide (TAP). Specific antibody binding of fusion molecule scFv anti-LIBS -TAP was proven in flow cytometry; anti-fXa activity was demonstrated in chromogenic assays. In vivo anticoagulative efficiency was determined by Doppler-flow in a ferric chloride–induced carotid artery thrombosis model in mice. ScFv anti-LIBS -TAP prolonged occlusion time comparable to enoxaparine, recombinant TAP, and nontargeted mutant-scFv-TAP. ScFv anti-LIBS -TAP revealed antithrombotic effects at low doses at which the nontargeted mutant-scFv-TAP failed. In contrast to the other anticoagulants tested, bleeding times were not prolonged by scFv anti-LIBS -TAP.

Conclusions— The novel clot-targeting approach of anticoagulants via single-chain antibody directed against a LIBS-epitope on GPIIb/IIIa promises effective anticoagulation with reduced bleeding risk.