Gerd J. Steffens

Thrombin inhibitory and clot-specific fibrinolytic activities of the urokinase variant, M23 (rscu-PA-40 kDa/Hir)

The recombinant bifunctional urokinase variant, M23 (rscu-PA-40 kDA/Hir), comprising the kringle and protease domain of single-chain urokinase-type plasminogen activator and a C-terminal fragment of hirudin in one single-chain molecule, was evaluated for its thrombin-inhibitory and fibrinolytic properties in vitro and in vivo. M23 inhibited thrombin-activated coagulation of human blood and thrombin-induced aggregation of human platelet rich plasma in a concentration-dependent manner. The ADP-induced aggregation of human platelet rich plasma was not influenced by M23. In contrast, recombinant single-chain urokinase-type plasminogen activator (saruplase) inhibited neither blood coagulation nor platelet rich plasma aggregation. M23 and saruplase both lysed radiolabelled human thrombi immersed in human plasma (Chandler Loop system) with equal potency. However, there was a significantly lower systemic generation of plasmin (measured as consumption of alpha 2-antiplasmin) by M23 compared to saruplase. In anaesthetized non-heparinized rabbits, experimental femoral artery thrombosis was treated with intravenous bolus injections of M23 or saruplase (6 mg/kg, each). Thrombolytic restoration of arterial blood perfusion was significantly higher in M23- than in saruplase-treated rabbits. Plasma fibrinogen concentrations were decreased markedly in saruplase-treated animals, but remained at significantly higher levels in M23-treated rabbits. In conclusion, the bifunctional molecule, M23, showed thrombin inhibitory and fibrinolytic properties in human in vitro systems and exerted superior thrombolytic effects to saruplase in rabbit femoral artery thrombosis. In vitro and in vivo data indicate that the fibrinolytic activity of M23 is highly clot-specific.

Computer Aided Protein Design: Three Dimensional Model Building of the Saruplase Structure

Modelling studies of the three-dimensional structures of the saruplase-domains are presented. The model of the N-terminal EGF-like domain highlights amino acids residues which might be involved in interactions with saruplase specific receptors. The distribution of charged residues on the surface of the kringle-model is different from other kringle- structures. The model structure of the catalytic serine-protease domain points to surface loops, which surround the active site and may participate in interactions with plasminogen. Starting from the structures of the isolated domains a model for the entire enzyme is constructed which is compatible with experimental results.

Transpeptidation in sequence analysis. Investigations concerning a native and a synthetic hexapeptide.

On sequencing a hemoglobin a peptic hexapeptide with the amino acid composition Asp2, Gly, Val, Lys2 was isolated. Cleavage of this peptide with Tos-Phe-CH2Cl-trypsin resulted in the fragments Asp-Lys, Gly-Asp, Val-Lys, Asp--Lys-Gly-Asp, Asp-Lys-Val-Lys and Val-Lys-Gly--Asp. From these data two possible but inconsistent sequences for the hexapeptide can be derived: Asp-Lys-Val-Lys-Gly-Asp and Val-Lys-Asp-Lys-Gly-Asp. Fragments obtained by other cleavage procedures, direct sequencing of the globin peptide-chain with a sequenator as well as X-ray data of this hemoglobin show, that only the sequence starting with Asp occurs in the intact protein. Therefore the peptide Asp-Lys-Gly-Asp must have been formed by transpeptidation during sequence work. In order to verify this, Asp-Lys-Val-Lys-Gly-Asp was synthesized by an unequivocal, conventional procedure. Tryptic digestion of this hexapeptide also resulted in ASP-Lys-Gly-Asp in addition to the expected fragments. Thus it has been shown for the first time, that sequencing conditions may alter the constitution of a peptide.