Activator of prothrombin from Echis multisqumatis venom (ecamulin) was purified by the method of Solovjev et al. as described previously [12] .

Prothrombin was isolated from human blood plasma as described in [13] .

Fibrinogen was isolated from human blood plasma as described in [14] .

The molecular weights and purity of proteins were determined using SDS-PAGE (10%) by Laemmli [15] . The molecular-weight standard proteins used were SM0671, from Fermentas (Lithuania) and the range was 7 - 170 kDa.

Meizothrombin stability was monitored by SDS-PAGE electrophoresis using 12% acrylamide gel during 30 min incubation of MT in TBS with 1 mM CaCl₂ at 37˚C.

Fibrinogen (0.4 mg/ml) clotting in the presence of FIIa, MT (0.3 nM) was detected in TBS with 1 mM CaCl₂, at 37˚C by turbidity study at 350 nm.

Fibrinogen (1 mg/ml) with trace amounts of factor XIII was incubated with thrombin in TBS with 0.015 M CaCl₂, for 10 min at 22˚C for factor XIII activation. Fibrinogen clotting was initiated by thrombin-like enzyme from Agkistrodon halys halys venom (ancistron, 0.25 NIH/ml) that clots fibrinogen but doesn’t activate factor XIII [16] ; the clot was removed and re-solved in 1.5 ml 1.5% acetic acid during 5 min. The concentration of non-cross-linked fibrin dissolved in acetic acid was detected by spectrophotometry at 280 nm. The level of cross-linked fibrin was calculated in percents.

Venous blood was collected from healthy volunteers who had not taken any medication for 7 days. As an anti- coagulant we used 38 g/liter sodium citrate (9 parts blood to 1 part sodium citrate). PRP was obtained by centri- fugation blood at 160 g for 20 min at 23˚C. Gel-sieved platelets were prepared according to the procedures of Belicer et al. [17] using size-exclusion chromatography on Sepharose 2B (Sigma-Aldrich, US) in 4 mM HEPES, 0.137 N NaCl, 2.7 mM KCl, 1 mM MgCl₂, 5.6 mM glucose, 3 mM NaH₂PO₄, 0.35 mg/ml BSA, pH 7.4 (HEPES-buffer).

Platelet aggregation was measured based on changes in the turbidity of platelet-rich human plasma [18] . We employed such experimental parameters as the lag-time between addition of collagen and the beginning of ag- gregation (lag-phase); the highest percentage of aggregation (rate) and initial speed of aggregation (speed). In typical experiment, 400 μl platelet-rich plasma was incubated with 20 μl 0.025 M CaCl₂ and 50 μl agonist (ADP 10 or 12.5 μM; collagen 2 or 3 μg/ml; adrenalin 0.8 or 0.6 μg/ml) at 37˚C. Aggregation was monitored for 10 min. ADP, collagen, adrenaline were from Tekhnologia-standard (Russia).

Activation of Gel-Sieved Platelets was detected by flow-cytometry on COULTER EPICS XL Flow-Cytometer. Gel-sieved platelets suspension (1 ml, 150 - 250 cells) was incubated with 0.5 ml of FIIa, or MT (0.28 and 0.75 nM/probe) at 37˚C during 20 min. Flow-cytometry profiles were based on cell size (forward angle light scatter) and cell granularity (side angle light scatter) during platelet activation [19] .

MT is a rather labile protein, which can be turned into thrombin by autocatalysis. All previous data of MT were obtained using recombinant MT that is resistant to autocatalytic removal of fragment 1 [6] [8] or in the presence of a specific, reversible active-site inhibitor dansylarginine N-(3-ethyl-1,5-pentanediyl)amide [7] . However, FII and FIIa are mobile molecules, so amino acid replacement can lead to structure changes and functional variation. We developed a method to get intact MT suitable for in vitro investigations.

For MT preparation, we used ecamulin (prothrombin activator from Echis multisquamatis venom) [20] [21] because it selectively cleaved FII at Arg320 with MT formation (Figure 1(a)). MT was prepared from human FII using ecamulin immobilized on BrCN-Sepharose (Amersham Biosciences, Sweden). Ecamulin was enzy- matically active after immobilization. For standardization of ecamulin-Sepharose we used chromogenic substrate S2302 (D-Pro-Phe-Arg-pNA, Chromogenix, Sweden): 1 unit of ecamulin-Sepharose cleaved S2302 with 4.5 μM/min pNA release.

To obtain MT, 1 mg of FII was incubated with 20 units of ecamulin-Sepharose in TBS with 0.2 mM CaCl₂ and 1 mM benzamidine during 15 min at 22˚C. A reversible inhibitor of thrombin, benzamidine, was used to protect MT from autocatalysis. The protein was purified from benzamidine by chromatography on PD-10 (Amer- sham biosciences, Sweden) ex temporo.

The chromogenic substrate of thrombin S2238 (H-D-Phe-Pip-Arg-pNA, Chromogenix) was used for detec- tion of the formation of active site in MT molecule (Figure 1(b)).

Obtained MT remained stable in TBS with 1 mM CaCl₂ at 37˚C during 30 minutes. Such MT was suitable for research in model systems in vitro. In further experiments we used lower concentrations of MT.

The main functions of FIIa during fibrin polymerization are the transformation of fibrinogen to fibrin and acti- vation of factor XIII, which provides formation of cross-linked fibrin clot. We measured the clotting activity of FIIa and MT by turbidity study.

MT was able to induce fibrin formation but the lag-phase of this process was distinctly longer than in the case of thrombin-induced polymerization (Figure 2(a)).

Another important process in the formation of polymeric fibrin is cross-linking by the factor XIIIa. Also, we studied the ability of MT to activate factor XIII, which so far had been neglected [8] . Fibrin polymerization was induced by thrombin-like enzyme ancistron that did not activate factor XIII by itself [22] . MT and FIIa were added to the mixture before ancistron. The amount of cross-linked fibrin was evaluated by measuring the non- cross-linked fibrin that was dissolved in 1.5% acetic acid. It was shown that MT was able to activate factor XIII but was much less effective than FIIa (Figure 2(b)).

FIIa activates platelets and triggers their aggregation, by cleaving the N-terminal region of G-protein-coupled protease-activated receptors (PARs). On the other hand, the platelet membrane glycoprotein GPIb acts as a thrombin-binding site and promotes platelet activation by decreasing thrombin concentrations locally [23] .

The role of MT in this process is still unknown [8] . We used washed human platelets and platelet rich plasma (PRP) to study MT’s influence on platelets.

Flow-cytometry showed no changes of shape and granularity of platelets in the presence of MT, though the protein didn’t activate washed human platelets (Figure 3). At the same time FIIa effectively activated washed human platelets.

The aggregation of platelets in PRP was also studied. MT induced platelet aggregation but was less effective in comparison to FIIa (Figure 4). Platelet aggregation induced by MT was reversible (Figure 4(a)). On the next stage irreversible platelet aggregation followed. Platelet aggregation in PRP was induced by MT in a concentration-dependent manner (Figure 4(b)).

To characterize the process, we also studied the aggregation of platelets in PRP induced by ADP, collagen and adrenalin. Platelet aggregation is a 2-stage process: First, cell signaling leads to GPIIbIIIa activation and grain release; then fibrin/fibrinogen binds to platelets by GPIIbIIIa, and the content of grains stimulates further platelet activation [24] . The second aggregation wave is characterized by studying of this process in the presence of light stimulants (adrenalin, serotonin, ADP in low concentration). ADP in high concentrations is a strong stimulant and can induce an one-wave irreversible aggregation similarly to FIIa. Only collagen-induced platelet aggregation has lag-phase that was shown to be limited by phospholipase C activation [25] .

We studied the aggregation of platelets in PRP induced by different stimulants in the presence of MT (Table 1). Low concentration of MT (0.1 μM) that didn’t induce platelet aggregation in PRP (Figure 4(b)) increased stimulant-induced platelet aggregation: significantly reduced the lag-phase of collagen-induced platelet aggrega- tion; increased the rate of ADP-induced platelet aggregation; raised the rate and speed of adrenalin-induced platelet aggregation. Thus MT had an enhancing effect on induced platelet aggregation.