Mice plasma fibrinogen consumption by thrombin-like enzyme present in rattlesnake venom from the North-East region of Argentina

Silvana L. Maruñak¹, Ofelia C. Acosta¹, Laura C. Leiva², Raquel M. Ruiz¹, Maria V. Aguirre², Pamela Teibler¹

¹Facultad de Ciencias Veterinarias; ²Facultad de Ciencias Exactas y Naturales y Agrimensura, Universidad Nacional del Nordeste, Corrientes

Postal address: Dra. Ofelia Acosta de Pérez. Cátedra Farmacología, Facultad de Ciencias Veterinarias - UNNE- Sgto. Cabral 2139, 3400, Corrientes, Argentina. Fax: 54-03783-425753 int. 113. E-mail: patmed@vet.unne.edu.ar

Abstract
Due to variability of venom components from the same species of snakes that inhabit different regions, particular properties of the venom of Crotalus durissus terrificus that inhabits the North-East of Argentina were studied. Gyroxin, a thrombin-like enzyme, was isolated from this venom by gel filtration and affinity chromatography, it was found to be homogeneous according to SDS-PAGE, with a molecular weight of 33 kDa. "Gyroxin syndrome" in mice was tested and it showed changes in the animal behavior, confirming that the isolated thrombin-like enzyme is gyroxin. Effects of this enzyme and the crude venom on mice plasmatic fibrinogen levels were determined. The mice plasma fibrinogen decreased rapidly until incoagulability during the first hour after thrombin-Iike enzyme injection, then reaching its normal level 10 hours after injection; whereas crude venom resulted in a 60% decrease of the mice plasma fibrinogen, reaching its normal level after the same period of time. After 1 hour of gyroxin inoculation, intravascular coagulation was observed in histological cuttings of lung, cardiac muscle and liver. The isolated enzyme showed strong hydrolyzing activity on fibrinogen and fibrin in vitro, whereas the crude venom exhibited weak hydrolyzing activity on both substrates. It is probable that this very low activity is due to the low percentage of the enzyme in the crude venom. Decreasing of plasmatic fibrinogen levels may be due to either the coagulant or hydrolyzing actions of the enzyme.

Key words: Crotalus durissus terrificus; Venom; North-East Argentina; Thrombin-like; Coagulation.

Resumen
Consumo de fibrinógeno plasmático en ratones por acción de enzima con actividad trombínica presente en el veneno de cascabel del nordeste argentino. Teniendo en cuenta la variabilidad de los componentes del veneno de serpientes de una misma especie que habitan regiones diferentes, se decidió estudiar las propiedades particulares del veneno de Crotalus durissus terrificus que habita el nordeste de Argentina, Giroxina, una enzima con actividad trombínica, fue aislada del veneno por cromatografía de filtración por gel y de afinidad; se comprobó su homogeneidad y se determinó su peso molecular, 33 kDa, por SDS-PAGE. Se ensayó el síndrome giroxina en ratones, los que mostraron cambios en el comportamiento, confirmando que la enzima tipo trombina aislada es giroxina. Se evaluó la acción de esta enzima sobre los niveles de fibrinógeno plasmático en ratones, comparándola con la del veneno crudo. Se comprobó que la enzima provoca una disminución de los niveles plasmáticos de fibrinógeno hasta la incoagulabilidad, durante la primer hora de inoculación, mientras que el veneno entero produjo una reducción del nivel plasmático en un 60%; sin embargo, en ambos casos, se evidenció una rápida reposición de fibrinógeno plasmático, alcanzando sus valores normales en un plazo de 10 horas. Se observó coagulación intravascular con la administración de giroxina una hora después de la inoculación, evidenciados en estudios histológicos de tejido pulmonar, cardíaco y hepático. En ensayos realizados in vitro, la enzima aislada mostró capacidad de degradar fibrinógeno como así también coágulos de fibrina, mientras que el veneno exhibió débil actividad hidrolítica sobre ambos sustratos. Es probable que esta baja actividad sea debida a la baja concentración de la enzima en el veneno. La disminución de los niveles de fibrinógeno plasmático observado en ratones se debería a la acción coagulante de la enzima, sin embargo no se descarta que también contribuya a este proceso una acción hidrolítica sobre fibrinógeno y fibrina por parte de la enzima.

Palabras clave: Crotalus durissus terrificus; Veneno; Nordeste de Argentina; Enzima tipo trombina; Coagulación.

     Blood coagulation by snake venom action was described at the beginning of last century¹. Thrombin-Iike enzymes from venoms of snakes belonging to the Viperidae family have been isolated and characterized^2-6.These enzymes are of great interest due to their use as anticoagulants in thrombotic diseases^7.
     Altered blood coagulation is not a major factor among the signs and symptoms of the envenoming by South American rattlesnake (Crotalus durissus terrificus). The clotting activity of this crude venom has been investigated in vitro and was shown to be thrombin-like⁸. Then, the enzyme responsible for this activity was isolated and characterized by Raw et al ⁵. This enzyme showed to be Aa fibrinogenase on human fibrinogen.
     Alexander et al.⁹ demonstrated that a thrombin-like enzyme from C. d. terrificus venom produces the gyroxin syndrome. This thrombin-like enzyme presents the barrel rotation syndrome in mice and has a variety of symptoms; some authors described it as a rolling behavior alternated with prostration periods¹⁰. Other authors mention that the animals show hypoactivity followed by loss of righting reflex, opisthotonos, spastic distortions and rotation around the long axis⁹. However, in some animals the rolling behavior was not observed but symptoms such as a brief period of hyper excitation and running, or paleness and tachypnea. Some mice show immobility, stretching of posterior limbs and grooming behavior. On the other hand, some of the animals die and others remain asymptomatic¹¹.
     Differences in toxic properties are detected in venoms of the same species that inhabit different geographic regions, different species in the same regions, and among newborn and adult specimens^{12, 13, 14}. In the North-East region of Argentina the prevalence of accidents caused by Crotalus is low, but its lethality is very high due to the action on nervous system¹⁵, for that reason it is important to know in detail the envenoming physiology caused by species of this region in order to apply an efficient treatment.
     In the coagulation, it is essential to know the periods in which this venom can affect blood clotting and the time that the liver spends to re-synthesis fibrinogen. It is also important to analyze if the low affectation of the clotting system is due to the low activity of the major responsible enzyme (thrombin-like enzyme) or to the low concentration of this enzyme in the venom.
     In this paper we describe the in vivo effect of C.d. terrificus venom from the North-East region of Argentina and the purified thrombin-like fraction on plasmatic fibrinogen levels, determining if the low injury on system is due to the low activity of the thrombin-like enzyme or to its low concentration in the venom. At the same time, the aim of this work is to distinguish if the consumption of fibrinogen is due only to the coagulant action of the enzyme or to both the coagulant and fibrinogen-fibrin hydrolyzing activities of the enzyme.

Materials and Methods

Chemicals and reagents

Crotalus durissus terrificus adult snakes proceeding from the North-East region of Argentina were kept in the Serpentarium of the local zoo, Corrientes, Argentina. Venom was collected, desiccated and immediately stored at -20 °C. Molecular weight markers (pre-stained SDS-PAGE standards, low range) were purchased from BioRad and Sigma. Standard bovine thrombin and fibrinogen were from Sigma. Sephadex G-75 and Benzamidine-Sepharose 6B, were from Sigma-Aldrich Co. The diagnostic kit for fibrinogen was from Wiener Laboratories S.A.I.C (Rosario, Argentina) and human fibrinogen type I for the specific cleavage study was from Sigma. All other reagents were laboratory grade.

Purification of thrombin-Iike enzyme

Purification of the thrombin-Iike enzyme was performed employing gel filtration and affinity chromatography⁹ with the modification introduced by Camillo et al¹⁶. In both cases the second step in the purification was performed by affinity chromatography in a Benzamidine-Sepharose 6B column, where they had used a buffer containing 50 mM Tris, 250 mM NaCI and 150 mM benzamidine at pH 9.0, to elute the thrombin-like enzyme from the column, we used a 0.1M sodium acetate buffer, pH 5.0, containing 0.15M NaCI. The elution was monitored at 280 nm and enzymatic essays were carried out as described below.

Analytical procedures

Estimation of protein

The protein determination of venom fractions was determined by measuring the absorbance at 280 nm in a UV-visible CamSpec M 330 spectrophotometer with an optical path of 1 cm, based upon the assumption that the absorbance of the 1 mg/ml of protein of venom fraction was 1.183, the same as that of crude venom.

SDS-PAGE

Purified gyroxin and crude venom samples were dissolved in phosphate buffer, pH 7.00, containing 1% (w/v) SDS, 0.05% (w/v) 2-mercaptoethanol, 5% urea and 0.005% bromophenol blue. Samples were heated for 3 min at 100°C an applied on top of 12% polyacrilamide slab gel¹⁷. Protein bands were revealed by Coomassie Brilliant Blue R-250 staining procedure. Molecular weight markers run in parallel were phosphorilase B, bovine serum albumin, ovoalbumin, carbonic anhydrase, soybean trypsin inhibitor and lyzosyme (Bio-Rad).
     Specific cleavage of fibrinogen and fibrin by crude venom and purified enzyme was performed using 12% polyacrilamide slab gel¹⁷. Samples were dissolved in 35 mM phosphate buffer, pH 7.00, containing 2% (w/v) SDS, 2% (w/v) 2-mercap-toethanol, 5% urea and 0.005% bromophenol blue and heated for 3 min at 100 °C. Molecular weight markers run in parallel were bovine serum albumin, ovoalbumin, carbonic anhydrase, trypsinogen (PMSF treated), trypsin inhibitor and a-lactalbumin (Sigma).

Enzymatic and toxic essays

Thrombin-like activity

Thrombin-like activity was assayed by recording clotting time for 0.2 ml of diluted bovine plasma (3:1 with 20 mM Tris HCI buffer, pH 8.0) with 0.1 ml of either crude venom or chroma-tographied venom fractions at 37 °C, according to the method described by Raw et al.⁵ with minimal changes in the quantities and reagents. The clotting times obtained were transformed into NIH units by means of a standard thrombin-fibrinogen curve constructed according to Lundblad¹⁸.

Gyroxin syndrome test

It was performed using swiss males and female mice with body weight 25-30 g inoculated by i.p. injection with different doses of purified protein in a rank of 0.02 to 3.2 µg/g body weight and observed during four hours.

Effects of gyroxin and C. d. terrificus venom on the mice plasma fibrinogen

The in vivo test was performed using eight groups of mice with different exposure time and one control group, each one composite by four male mice weighing 18-22 g, which were i.p injected with 0.5 µg/g body weight of purified thrombin-like enzyme. The different groups were:
     Group 1: 30 minutes; group 2: 60 minutes; group 3: 3 hours; group 4: 6 hours; group 5: 8 hours; group 6: 10 hours and group 7: 24 hours. After the time of exposure, the animals were anesthetized with chloral hydrate (300 mg/kg body weight) and two ml of blood were collected through the cava vein. The blood (1.8 ml) was anticoagulated with 0.2 ml of trisodium citrate (0.129M) in a blood collection tube. Platelet-poor plasma was obtained after centrifugation at 2500 rpm at room temperature for 15 minutes. The fibrinogen content was determined (see fibrinogen determination). Each sample was measured by triplicate and the mean value of fibrinogen concentration was calculated. The control group (group 8) was not inoculated with gyroxin; only blood samples were collected to determine standard plasma fibrinogen levels.
     The same procedure was carried out with crude venom, so the groups with different time of exposure assayed and the dose were the same. In this experience, two hours after venom inoculation, the animals were treated with anticrototalic serum to avoid mice death. This latter may happen two hours after venom inoculation.
     Plasmatic fibrinogen determination. Plasmatic fibrinogen level was determined by the diagnostic kit (Wiener, fibrinogen) based on the other method¹⁹, the procedure is a clotting rate method with the addition of a very high bovine thrombin concentration to a 10-fold diluted plasma and renders the logarithm of the clotting time directly proportional to the logarithm of fibrinogen concentration. Determinations and calibration curve were performed according to the manual of the manufacturer.

Specific hydrolysis of fibrinogen

Specific cleavage of fibrinogen by the crude venom and the isolated enzyme was shown by electrophoresis on 12% polyacrylamide gels. One hundred µl of human fibrinogen at a concentration of 5 mg/ml in 0.1 M Tris-HCl buffer (pH 8) containing 0.5 mM CaCl₂ and 0.02% NaN₃, was added to 100 µl of the isolated thrombin-like enzyme or crude venom at a concentration of 250 µg/ml in the buffer previously mentioned, and incubated at 37 °C. At various time intervals, aliquots of 30 µl were withdrawn from the digestion mixture and the reaction was stopped by addition of 30 µl of denaturing solution (35 mM phosphate buffer, pH 7.00, containing 2% SDS, 2% 2-mer-captoethanol, 5% urea and 0.005% bromophenol blue). The samples were denatured and reduced by boiling for 3 min prior the run on SDS-polyacrilamide gel²⁰.

Degradation of fibrin

Fibrinolytic activity. A modified plaque assays²¹ was used to measure fibrinolytic activity of the venom and the thrombin-like enzyme. Citrated sheep plasma was diluted 1:2 with a buffer containing 50 mM Tris, pH 7.5. Twenty-two milliliters were been transferred to plastic dishes (13 x 8 cm), 2.2 ml of 0.25 M CaCl₂ and 132 µl human thrombin (Sigma Chemical) were added to produce a clot. After 60 min of incubation at 37 °C wells of 2 mm diameter were made in the clot and 10 µl of venom solutions (ranging from 0.156-40 mg/ml) or thrombin-like enzyme (ranging from 0.125-5.0 mg/ml) were applied to each well. Incubations were carried out at 37 °C for 18 h, after which the diameter of the fibrinolytic halos was measured. The minimum fibrinolytic concentration (MFC) was defined as the concentration of venom (final, mg/ml) that induced a fibrinolytic halo of 12 mm diameter. The assay was performed three times.
     Fibrin degradation products. Fibrin hydrolysis was demonstrated by SDS-PAGE using 12% polyacrilamide gels. Two hundred and fifty microliters of thrombin 10 units/ml was added to 250 µl of a 10 mg/ml fibrinogen solution in 0.1 M Tris-HCl buffer (pH 8) containing 0.5 mM CaCl₂ and 0.02% NaN₃. The fibrin clot was allowed to form for 1 h at room temperature. After that, 500 µl of thrombin-like enzyme or crude venom (250 µg/ml) was added to the clot and incubated at 37 °C. At various time intervals, aliquots (30 µl) of incubation mixture was withdrawn and added to 30 µl of denaturing solution (35 mM phosphate buffer, pH 7.00, containing 2% SDS, 2% 2-mercap-toethanol, 5% urea and 0.005% bromophenol blue). The samples were reduced and denatured in a boiling water bath for 3 min and run on SDS-polyacrilamide gels²⁰.

Light microscopy for intravascular coagulation detection

In order to have a histological assessment of in vivo clot formation ability by the isolated thrombin-like enzyme, groups of four mice were injected i.m. in the right gastrocnemius with 0.1 ml of solution containing the enzyme isolated (9 µg). Four mice were used as control samples, receiving each of them 0.1 ml of phosphate buffered saline solution (pH 7.2). After 60 and 120 minutes of injection mice were anesthetized with cloral hydrate i.p. 300 mg/kg. Mice were sacrificed by cervical dislocation and samples of cardiac muscle, lung and liver were taken and fixed in Bouin solution for 24-48 h. Thereafter, the samples were dehydrated in a graded alcohol series and embedded in paraffin. Sections 10 µm thick were stained with hematoxylin-eosin (HE).

Results

Purification

The results from a typical purification of gyroxin are shown in Table 1. The venom was purified in a two-step procedure. The elution profile from the G-75 chromatography is shown in Fig. 1 and that of the Benzamidine-Sepharose 6B chromatography in Fig. 2. After procedure the enzyme obtained had 34-fold purification factor and a 12.1% in activity yield. Starting with 250 mg of venom, 0.9 mg of pure thrombin-like enzyme was obtained. As shown in Table 1, the Sephadex G-75 column was an efficient step in this procedure, giving 5-fold purification. The enzyme was eluted with glycine buffer, 20 mM, pH=1.9, containing 150 mM of sodium chloride. As shown in Fig. 1 most of the thrombin-like activity eluted in the higher molecular weight region. Starting with 250 mg of venom, 23 mg of this protein was obtained in this step. Those fractions containing thrombin-like activity were pooled and applied to the Benzamidine-Sepharose 6 B affinity column (step 2). There was one protein retained by the benzamidine column (Fig. 2) and eluted with sodium acetate buffer, 100 mM, pH= 5.0, containing 150 mM of sodium chloride which exhibited thrombin-like activity. This step resulted in 6-fold purification of the enzyme; besides, it was essential for obtaining a protein of higher homogeneity. SDS-PAGE attributed a value of 33 kDa for the enzyme isolated after affinity chromatography, this value is similar to those obtained by other authors (34 kDa⁹, 36 kDa¹⁶) (Fig. 3). The enzyme isolated has a thrombin-like activity as demonstrated by its ability to clot bovine plasma, exhibiting a higher activity than the one corresponding to the crude venom (Table 1)

TABLE 1.– Purification of gyroxin from Crotalus durissus terrificus venom

Fig. 1.– Elution profile of Crotalus durissus terrificus venom fractionated on Sephadex G-75. Venom (50 mg/ml, 5 ml) was applied to a column (2.0 x 75 cm) of Sephadex pre-equilibrated and eluted with glycine buffer 20 mM, 150 mM NaCl, pH 1.9. Five ml fractions were collected at a flow rate of 12 ml/h and assayed for absorbance at 280 nm ( • ) and thrombin-like activity (- - - - ).

Fig. 2.– Benzamidine-Sepharose chromatography of pooled fractions from the Sephadex G-75 gel filtration column. Elution profile of Benzamidine-Sepharose 6B (1.0 x 6.0 cm) equilibrated and run with 50 mM Tris, 400 mM NaCl, pH 9.0. The line represents absorbance at 280 nm. The pooled sample from the G-75 step (35 ml) was pH and ionic adjusted as described by Alexander et al ⁹. The sample was loaded and then the column was washed with running buffer until the absorbance at 280 nm returned to baseline. At the point indicated by the arrow the running buffer was changed to sodium acetate buffer, 100 mM, pH = 5.0, containing 150mM of sodium chloride. One protein peak eluted contained the thrombin-like activity.

Fig 3.– Electrophoretic profiles of gyroxin and Crotalus durissus terrificus venom. A: SDS-polyacrilamide 12% gel electrophoresis of: Crotalus durissus terrificus crude venom (lane 2), molecular weight markers (lane 1). B: SDS-polyacrilamide 12% gel electrophoresis of Crotalus durissus terrificus crude venom (lane 1), purified thrombin-like (gyroxin) (lane 2).

Gyroxin syndrome test

The animals injected with a low doses (0.02 to 0.25 µg/g body weight) showed in the first minutes a brief period of hyper-excitation with running and grooming behavior, after that between 10 min to 20 min was observed a sudden stretching of body and posterior limbs accompanied by opisthotonos with decrease in the time of righting reflex compared with control animals. All symptoms disappeared after half past one hour and the animal behavior returned to normal. On the other hand, mice injected with higher doses (0.34 to 0.75 µg/g body weight) developed symptoms such as immobility following by tachypnea with apparently breathing problem and stretching of posterior limbs. Not one the doses were lethal and all animals returned to normal behavior. All results confirmed that the isolated enzyme is gyroxin.

Effect of gyroxin and C. d. terrificus venom on the mice plasma fibrinogen

The effect of C. d. terrificus venom and purified thrombin-Iike enzyme were examined in vivo. The experiments showed that the mice plasma fibrinogen decreases rapidly during the first hour after the thrombin-Iike enzyme injection, until non detected values, then reaching its normal level 10 h after injection; whereas the crude venom, when injected in vivo, results in a 60% decrease of the mice plasma fibrinogen, reaching its normal level after the same period of time (Fig. 4) It is important remark again that the anticrotalic serum did not affect the evolution of plasmatic fibrinogen levels, since this latter was registered when the antivenom was not yet given to the mice.

Fig. 4.– Loss of mouse plasma fibrinogen after intravenous injection of thrombin-like enzyme isolated (- -•- -) and C. d. terrificus venom ( o ). Values are means of three determinations for each mice group (n=4).

Degradation of fibrinogen

Both crude venom and the thrombin-like enzyme exhibited fibrinogenolytic activity. SDS-PAGE analysis showed the thrombin-like enzyme isolated to be an Aa Bb fibrinogenase with the Aa chains being degraded within 2 hours followed by a slower degradation of the Bb chains (Fig. 5). The enzyme appeared to have no effect on the g chain after of 18 h of hydrolysis. Fibrinogen degradation products of different molecular weight early appeared and increased during hydrolysis, indicating extensive hydrolysis of the fibrinogen.

Fig. 5.– Fibrinogen degradation by the venom and the isolated enzyme. First lane (left) fibrinogen control incubated without enzyme for 10 min. Fibrinogen incubated for the specified times with venom (lanes A). Fibrinogen incubated for the specified times with the isolates enzyme (lanes B).

     The crude venom exhibited a weak Aa Bb fibrino-genolytic activity. Partial hydrolysis of Aa and Bb chains was only observed after 2 h, and the almost total degradation was after 18 h. At these times, fibrinogen degradation products can be observed; some of them exhibit different molecular weights from those produced by the isolated thrombin-like enzyme (Figure 5).

Degradation of fibrin

Fibrinolytic activity and cleavage specific of fibrin

The isolated enzyme showed a strong fibrinolytic activity when it was added to a fibrin clot by the plaque method. The halos were evaluated after 18 h of incubation and were proportional to the amount of enzyme; the obtained linear relationship (r = 0.997) let us determine the MFC. The concentration of purified thrombin-like enzyme that induced a fibrinolytic halo of 12 mm diameter was 2.30 mg/ml.
     The crude venom showed a weak fibrinolytic activity when it was added to a fibrin clot by the plaque method. Halos of 12 mm diameter were not observed even when high doses of 40 mg/ml were used. For that reason, the MFC for the whole venom could not be determined.
     The degradation of fibrin was examined by SDS-PAGE to determinate which subunit of this protein was affected by the crude venom and by the isolated enzyme. The hydrolysis of fibrin by the crude venom resulted in the degradation of the a-chains, while the b-chains, the g-chain and the dimmer g-g appeared unaffected throughout the incubation period examined (Fig. 6). Degradation products of high molecular weight (40 - 45 kDa), of molecular masses between 20 and 24 kDa and of molecular weight minor than 14 kDa were observed within the first two hours of hydrolysis. After 18 hours, these bands appeared more intensified.
     A similar experiment performed with the purified enzyme indicated that it was able to hydrolyze the a- and b-chains of fibrin whereas it appeared to have no effect on the g chain and dimmer g-g even after 18 h incubation (Fig. 6). Partial hydrolysis of a- and b-chains was observed after 2 h. They were completely degraded after 18 h. Degradation products of molecular weight around 33 kDa appeared after 2 h incubation. After 18 h, a new band appeared with molecular mass around 22 kDa. Therefore, fibrin degradation products obtained with the isolated gyroxin were no similar to those obtained with the crude venom.

Fig. 6.– Fibrin degradation by the venom and the isolated enzyme. Fibrinogen control incubated without enzyme for 10 min (lane A). Fibrin control incubated without enzyme for 10 min (lane B). Fibrin incubated for the specified times with venom the isolates enzyme (lanes C). Fibrin incubated for the specified times with the isolated enzyme (lanes D). Molecular weight markers (lane F).

Histological assay

After 1 hour of gyroxin inoculation, congestion of small coronary blood vessels, arterioles and capillaries among the muscular fibers were observed. After two hours, intravascular coagulation was detected.
     In cuttings of lung, after 1 hour of enzyme injection, congestion of alveolar septums and intravascular thrombosis were observed. After two hours, lesions were more intense (Fig. 7). In cuttings of liver, lesions compatible with intravascular coagulation were observed in central veins (Fig. 8).

Fig. 7.– Histological cutting of lung of mouse inoculated with 9 µg of gyroxin. Time of exposure: 2 h. Pulmonar congestion, edema and intralveolar hemorrhage (star), accompanied by thrombosis in blood vessels (arrow), were observed (40xHE).

Fig. 8.– Histological cutting of liver of mouse inoculated with 9 µg of gyroxin. Time of exposure: 2 hs. Vascular thrombosis characterized by abundant fibrin (asterisk) was observed. Platelets (star) and blood red cells (arrow) surround the clot (400xHE).

     On the whole, in every studied tissues it was verified that the process begins with the congestion of small blood vessels, arterioles and capillaries. After that, intravascular coagulation, characterized by abundant fibrin with platelets and blood red cells joined to the clot, is manifested. It demonstrates that the isolated enzyme possesses in vivo coagulant activity.

Discussion

In this paper a thrombin-like enzyme was isolated, through a modified technique as described by the literature⁹, then this enzyme was used to study its effects on haemostatic system in mice. Regarding to the methodology employed to isolate the enzyme, we observed that the markedly acidic pH in the first step, allowed to obtain the enzyme among the first fractions of eluted proteins, favoring its separation from the major proteins contained in the venom (crotoxine complex), without affecting the activity of the enzyme⁹.
     The modification introduced by Camillo et al.¹⁶ in the second step let us an easy elution of the protein without sample consumption, at the same time it allowed to elute the enzyme in its biologically active form, being the clotting activity of the isolated fraction much more intense than the one corresponding to the crude venom.
     Low recovery percentage of the isolated enzyme (0.4% of the total protein present in the original venom sample) and the reduced clotting activity of the whole venom could indicate a low concentration of thrombin-like enzyme in the venom of the rattlesnake that inhabits the northeast region of Argentina. It is important to clear up that Seki et al.²² obtained a higher output (0.8-1%) when they purified the enzyme presents in the venom extracted from C. d. terrificus specimens coming from Santiago del Estero (central region of Argentina). This different output with regard to ours, could be explained considering that venoms, obtained from specimens of different regions, have different compositions, as described Cavinato et al.¹². Also, the used methodology in that case, twenty years ago, could not have let an efficient purification of the enzyme.
     The purified thrombin-like enzyme induced changes in the animal behavior, eliciting a sintomatology that coincides with the "gyroxin syndrome", described by other authors^{9, 22}. It confirms that the isolated protein is gyroxin. Symptoms varied according to the dose inoculated presenting a more complete syndrome with the smallest dose used. It points out a non-lineal relationship between dose and effect as showed in other studies with this enzyme²².
     Gyroxin, with a higher coagulant activity than the one corresponding to the crude venom, caused plasmatic fibrinogen consumption in a way more intense than crude venom. In the first hours, the fibrinogen consumption after injection was quick, whereas with crude venom the consumption was lower. Fibrinogen levels were normal after ten hours with either thrombin-like enzyme or crude venom.
     The purified enzyme showed in vitro and in vivo plasma clotting activity. In that sense, intravascular clots observed in histological cuttings confirmed this ability. On the other hand, in vitro studies demonstrated that the isolated enzyme was capable to degrade fibrinogen and fibrin. Although Raw et al.⁵ observed that the thrombin-like isolated from Crotalus durissus terrificus only hydrolyze the Aa chains of fibrinogen, we demonstrated that our enzyme degrades the Bb chains besides the Aa ones. These results agreed with those reported by Markland Jr and Damus⁴, who described that the thrombin-like isolated from Crotalus durissus adamanteus hydrolyzes the Aa chains of fibrinogen and also slowly hydrolyzes the Bb chains. Besides, we demonstrated that the thrombin-like enzyme from Crotalus durissus terrificus exhibited hydrolytic activity on fibrin: a-b-chains were partially hydrolyzed two hours after incubation and completely hydrolyzed after 18 hours.
     These results let conclude that probably the enzyme in vivo, besides consuming fibrinogen by coagulation, directly hydrolyses fibrinogen. It is also possible the enzyme degrades fibrin clots as it was demonstrated in vitro. However, the observation of clots in histological cuttings obtained two hours after enzyme injection demonstrates an initial action of the enzyme, which then metabolizes allowing the restatement of plasma fibrinogen level, being the liver, in physiological conditions, the organ responsible of doing this latter.
     Regarding to the crude venom, it exhibited weak Aa Bb fibrinogenolytic activity and a fibrinolytic activity throughout the incubation period examined. The plaque method showed small halos, even to high doses, so the MFC for the whole venom from Crotalus durissus terrificus could not be determined. Considering that other crotaline venoms show fibrinolytic activity (MFC: 2.56 mg/ml for Crotalus durissus durissus, 1.99 mg/ml for Bothrops asper_Pacific; 3.59 mg/ml for Bothrops asper_Atlantic) ²³ as our isolated enzyme from Crotalus durissus terrificus show also (2.30 mg/ml), it is probable that this very low activity exhibited by the venom is due to the low percentage of the enzyme in the crude venom. It is also possible the presence of other proteolytic enzymes in the crude venom since fibrinogen and fibrin degradation products, different from those obtained with the enzyme isolated, were found.
     Results here obtained, show the ability of the isolated thrombin-like enzyme of consuming fibrinogen by coagulation, and probably by hydrolysis too. They also demonstrate the rapidity of fibrinogen reinstatement, even with a high amount of the enzyme. Such observations, added to the low levels of the coagulant activity detected in the venom of the rattlesnake that inhabits the northeast region of Argentina, could explain the low incidence of the clotting dysfunction present in the accidents caused by these snakes. However, it is important to amplify the knowledge about the physiopathology of the intoxication by rattlesnake in order to apply an efficient treatment, especially in patients with hepatic or clotting diseases, where the replacement of the fibrinogen could be delayed, even inhibited, leading to serious alterations in the haemostatic system that could complicate the recovery of the victim.

Acknowledgements: This work was supported by Secretaria General de Ciencia y Técnica de la Universidad Nacional del Nordeste de Argentina. The authors wish to express their gratitude to Laura Rey and her staff from the Serpentarium for giving us the venom used in this study of Corrientes, Argentina, and to Javier Lencina, Roberto Mariño and Noemi Fidanza for their collaboration in the electrophoresis analyses.

References

1. Mellamby J. The coagulation of blood. II. The action of snake venoms, peptone and leach extract. J Physiol 1909; 38: 441-503.
2. Seegers WH, Ouyang C. Snake venoms and blood coagulation. In: Lee CY, (ed). Snake venom handbook of experimental pharmacology, Saint Louis: Sigma Chemical, 1991, p. 684-700.
3. Stocker K. Defibrinogenation with thrombin-like snake venom enzymes. Handbook of Farmacology 1978; 46: 451-84.
4. Markland FS Jr, Damus PS. Purification and properties of a thrombin-tike enzyme from the venom of Crotalus adamanteus (Eastem Diamondback rattlesnake). J Biol Chem 1971; 246: 6460-73.
5. Raw I, Rocha MC, Kamiguti AS. Isolation and characte-rization of a thrombin-like erizyme from the venom of Crotalus durissus terrificus. Braz J Med Biol Res 1986,19: 333-8.
6. Smolka MB, Marangoni S, Oliveira B, Novello JC. Purification and partial characterization of a thrombin-like enzyme, balterobin, from the venom of Bothrops altematus. Toxicon 1998; 36: 1059-63.
7. Stocker K F. Snake venom proteins affecting hemostasis and fibrinolysis. ln: Stocker KF (ed). Medical Use of Snake Venom Proteins. Boston, USA: CRO Press 1990, p 97-9.
8. Nahas L, Denson KWE, Mac Farlane RG. A study of the coagulant action of eight snake venoms. Thromb Diath Haemorrh 1964; 12: 355-67.
9. Alexander G, Grothussen J, Zepeda H, Schwartzman R J. Gyroxin, a toxin from the venom of Crotalus durissus temficus, is a thrombin-like enzyme. Toxicon 1988; 26: 953-60.
10. Cohn MI, Cohn M. Barrel rotation induced by somatos-tatin in the non-lesioned rat. Brain Res 1975; 96: 139-41.
11. Barrabin H, Martiarena JI, Vidal JC, Barno A. lsola-  tion and characterizaation of gyroxin from Crotalus durissus terrificus venom. In: Rosemberg P. (Ed). Toxins: Animals, Plant and Microbial. New York: Pergamon 1978, p. 113-4.
12. Cavinato R.A, Remold H, Kipnis TL. Purification and variability in thrombin-like activity of Bothrops atrox venom from different geographic regions. Toxicon 1998; 36: 257-67.
13. Word Health Organization (WHO) 23⁰Informe Técnico de Patrones Biológicos, Publicación N⁰ 37. 1971, Ginebra.
14. Gutiérrez J M, Chaves E, Bolaños R. Comparatíve study of venoms of newborn and adult specimens of Bothrops asper. 1980. Rev Biol Trop 228: 341-51.
15. Esteso SC. Ofidismo en la República Argentina. Edit Arpon, Córdoba, 1985, p.70-71.
16. Camillo MAP, Arruda Paes PC, Troncone LRP, Rogero JR. Gyroxin fails to modify in vitro release of labelles dopamine and acetylcholine from rat and mouse striatal tissue. Toxicon 2001; 39: 843-53.
17. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680-5.
18. Lundblad RL, Kingdon HS, Mann KG. Thrombine. Methods Enzymol 1976; 45: 685-50.
19. Clauss VA. Gerinnungsphysiologishe schnellmethode zur bestimmung des fibrinogens. Acta Haematol 1957; 17: 237-46.
20. Assakura MT, Reichi A P, Mandelbaum FR. Isolation and characterization of five fibrin(ogeno)lytic enzymes from the venom of Philodryas olfersii (green snake). Toxicon 1992; 32: 819-31.
21. Rojas G, Gutiérrez JM, Gené JA, Gómez M, Cerdas L. Neutralización de las actividades tóxicas y enzimáticas de cuatro venenos de serpientes de Guatemala y Honduras por el antiveneno polivalente producido en Costa Rica. Rev Biol Trop 1987; 35: 59-67
22. Seki C, Vidal JC, Barrio A. Purification of giroxin from a south American rattlesnake (Crotalus durissus terrificus) venom. Toxicon 1980; 18: 235-47.
23. Gené JA, Roy A, Rojas G, Gutiérrez JM and Cerdas L. Comparative study on coagulant, defibrinating, fibrinolytic and fibrinogenolytic activities of Costa Rican crotaline snake venoms and their neutralization by a polyvalent antivenom. Toxicon 1989; 27: 841-8.        [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]

Received: 30-01-2004
Accepted: 10-08-2004