Direct thrombin inhibitors built on the azaphenylalanine scaffold provoke degranulation of mast cells.

The main structural feature of direct thrombin inhibitor LK-732 responsible for the appropriate interaction at the thrombin active site is a strong basic group. A possibility that a strong basic group of LK-732 might contribute to the mast cell degranulation effect and consequent reduction of tracheal air flow (TAF) and fall of mean arterial blood pressure (MAP) in rats was investigated in the present study. At doses up to 5 mg/kg (i.v.), LK-732 did not cause significant changes of TAF and MAP. At 7 mg/kg (i.v.), a sudden reduction of TAF and a fall of MAP was observed within 5 min after LK-732 administration (75% mortality, p = 0.007). A less basic direct thrombin inhibitor LK-658 (21 mg/kg, i.v.) did not significantly disturb TAF and MAP. A reduction of TAF and a fall of MAP caused by LK-732 (7 mg/kg, i.v.) was almost completely abolished in rats with degranulated mast cells (0% mortality, p = 0.008). LK-732 concentration-dependently degranulated rat peritoneal mast cells in vitro (pEC(50) = 1.92 +/- 0.05 muM). A structure-activity relationship (SAR) study revealed that the terminal basic groups attached to the aromatic ring are responsible for the mast cell degranulation effect. A good correlation was observed between mast cell degranulation and pK(b) of analogues of LK-732 (R(2) = 0.49), but not between mast cell degranulation and thrombin K(i) (R(2) = 0.23). LK-732-induced reduction of TAF, the fall of MAP and high mortality originate from LK-732-induced mast cell degranulation. As judged by the SAR study, this effect could be overcome by reducing the basicity of LK-732.

Antithrombotic potential of new direct thrombin inhibitors built on the azaphenylalanine scaffold in two rat venous thrombosis models.

The antithrombotic potential of new direct thrombin inhibitors built on the azaphenylalanine scaffold (LK-732, LK-639 and LK-731) and their amidoxime prodrugs (LK-658, LK-633 and LK-730) was studied in comparison to argatroban and nadroparin in two rat models of venous thrombosis, induced either by complete stasis combined with hypercoagulability (model 1) or by partial stasis combined with vessel injury (model 2). In initial experiments LK-732 was established as the most promising antithrombotic of the LK inhibitors and as such was further tested. In model 1, intravenous bolus administration of LK-732 produced a dose-dependent inhibition of thrombus formation with an ID50 value of 1.3 mg/kg. This ID50 value was approximately four times higher than the ID50 value of argatroban (0.3 mg/kg; p=0.011). However, in model 2, LK-732 and argatroban decreased thrombus weight by 50% at similar ID50 values (3.8 mg/kg vs 3.0 mg/kg, respectively; p=0.726). The ex vivo anticoagulant effect of LK-732 was substantially weaker compared to argatroban at doses that produced comparable antithrombotic effects. After subcutaneous administration, in vivo thrombus weight reduction of LK inhibitors (10 mg/kg) ranged between 22 to 48%. However, their oral antithrombotic effect at a dose of 30 mg/kg was rather low. LK amidoxime prodrugs failed to produce a substantial antithrombotic effect after subcutaneous (10 mg/kg) as well as after oral administration (30 mg/kg). In conclusion, thrombin inhibitors built on the azaphenylalanine scaffold represent a new group of intravenously effective antithrombotics. However, optimisation of the oral antithrombotic effect of amidoxime prodrug LK-658 of the lead inhibitor LK-732 is required for justifying further development of these inhibitors.