- Email: [email protected]
Isolation and characterization of synergistic hemorrhagins from the venom of the snake Bothrops asper
Toxicon, Vol. 31, No . 9, pp . 1137-1150, 1993 . Printed in Great Britain.
0041-0101/93 $6.00 + .00 ® 1993 Pergarnon Pren Ltd
ISOLATION AND CHARACTERIZATION OF SYNERGISTIC HEMORRHAGINS FROM THE VENOM OF THE SNAKE BOTHROPS ASPER GADI BORKOw, 1 Jost MARIA GuTd?RREzz and MICHAEL OVADIA 1
'Department of Zoology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel ; and ZInstituto Clodomiro Picado, Facultad de Microbiologia Universidad de Costa Rica, Costa Rica (Received 29 July 1992 ; accepted 25 March 1993)
G. BORKOW, J. M. GUTtlftRREZ and M. OVADIA. Isolation and characterization of synergistic hemorrhagins from the venom of the snake Bothrops asper. Toxicon 31, 1137-1150, 1993 .-Three hemorrhagic factors (BaHl, 13112 and BH3) were isolated from the venom of Bothrops asper by gel filtration on Sephacryl S-200, DEAE-Sepharose chromatography, metal chelate affinity chromatography and hydrophobic interaction chromatography . They contain 55% of the total hemorrhagic activity of the whole venom when they are mixed, but lose almost half of the activity if they are separated, indicating a synergism between the three. The main hemorrhagin is BaH l (Bothrops asper hemorrhagin 1) ; the other two are weak hemorrhagins but contribute to the synergism. They are acidic proteins with a pI of 4.5, 5.2 and 5; their mol. wt is 64,000, 26,000 and 55,000 respectively . The minimal hemorrhagic dose (MHD) of BaH1, BH2 and BH3 is 0.18, 2 and 1 .61íg, with a specific activity 55, 5 and 6.25 higher than that of the whole venom. The hemorrhagic activity of all three factors was inhibited by EDTA and ortho-phenathroline, indicating that the hemorrhagic activity is metal dependent. Phosphoramidon, soybean trypsin inhibitor, PMSF, pepstatin and aprotinin did not affect the hemorrhagic activity of the isolated factors.
THE
INTRODUCTION
MmoRITy of snakebites in Central America are caused by pit vipers . Among them locally known as `terciopelo' or `barba amarilla', is one of the most dangerous and abundant species (MINTON, 1969 ; BOLANOS, 1982). It ranges from northern Mexico to the lowlands of Colombia and Ecuador (MARTIN, 1958 ; PETERS and OREJAS-MIRANDA, 1970). The venom of this snake induces a complex series of local effects including hemorrhage, myonecrosis and edema, in addition to systemic effects such as hemorrhage, coagulation disorders, cardiovascular shock and acute renal failure (BoLANos, 1982). Among these effects hemorrhage represents an especially difficult problem mainly because it develops quickly after venom injection causing major microvasculature damage and blood loss which leads to muscle and other tissue degeneration. Similar striking consequences of envenomation, due to the occurrence of local and
Bothrops asper,
*Author to whom correspondence should be addressed. 1137
113 8
G . BORKOW et al.
peripheral hemorrhage, are caused by most crotalid and viperid snake venoms (OWNBY et al., 1978; OHsA", 1979). Forty-three hemorrhagins from 15 species were isolated and characterized (BJARNSON and Fox, 1988-89). Most of them are acidic proteins with mot. wts ranging from 20,000 to 100,000. All but two appear to be metal-dependent proteolytic enzymes. The present communication reports the isolation and characterization of three hemorrhagic factors from the venom of Bothrops asper that can act synergistically . MATERIALS AND METHODS Materials Sephacryl S-200, DEAE-Sepharose, Phenyl Sepharose CL-4B and Chelating Sepharose Fast Flow were purchased from Pharmacia Fine Chemicals (Uppsala, Sweden) ; Coomassie brilliant blue and the enzyme inhibitors were obtained from Sigma Chemical Co . (St . Louis, MO, U.S .A .) ; acrylamide and NN'-methylenebisacrylamide were obtained from BDH (Poole, U.K .); all other chemicals were analytically pure and were purchased from Merck (Darmstadt, F.R .G.). Bothrops asper venom The snakes were kept and 'milked' at Instituto Clodomiro Picado, University of Costa Rica. The aliquots of the 'milked' venom were pooled, lyophilized and stored at -20°C until used . Antisera Antiserum against Bothrops riper venom was prepared in Instituto Clodomiro Picado, Costa Rica, by injecting increasing amounts of venom at l0-day intervals s.c. into horses . The antigenic mixture contained equal parts of the venoms of Bothrops asper, Crotalus durissus durissus and Lachesis muia stenophrys (BoLANOS and CERDAs, 1980). Freund's complete adjuvant or alginate was used as adjuvant . At the end of immunization, horses were bled from the jugular vein and plasma was fractionated by ammonium sulfate precipitation (BOL .ANos and CERDAs, 1980) . Antisera against Vipera palaestinae, Walterinnesia aegyptia, Echis coloratus, Crotalus atrox and Cerastes cerastes were prepared in rabbits in Tel Aviv university as above ; the blood was drained from a vein in the ear . Antisera against Vipera persicus, Vipera lebetina and Vipera latifii were purchased from Razi Institute, Iran; antisera against Crotalus durissus terrificus from Butantan Institute, Brazil ; antisera against Bitis arietans and Bilis gabonica from Institute Pasteur, Paris; and antisera against Vipera russelli from the Thai Red Cross Society, Thailand . Determination of hemorrhagic activity The hemorrhagic activity of the whole venom and its fractions was tested by the skin injection method of KONDo et al. (1960) with some modifications of OvADiA (1978) . One-tenth of a milliliter of saline containing 0.05-20 lug of the examined material was injected into the skin at the back of white mice . Two hours later a red hemorrhagic spot was observed on the inner surface of the removed skin . The amount of venom or its fractions that produced a hemorrhagic spot of about 1 cm in diameter was defined as one minimal hemorrhagic dose (MHD) . The method is semi-quantitative . Ammonium sulfate precipitation Cold saturated ammonium sulfate was added to the whole venom or to each of the purified factors in a cold room under continuous stirring until 40% saturation was achieved . After 30 min, the suspensions were centrifuged for 20 min at 10,000 x g and the precipitate was redissolved in 0 .01 M phosphate buffer, pH 7 . More saturated ammonium sulfate was added to each of the supernatant fluids to attain 55% saturation and 30 min later the suspensions were treated as above followed by a dialysis against the same buffer. The hemorrhagic activity of the fractions was examined as described above . Determination of mol. wt The mol. wt was determined by the sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) method according to the procedure of LAEmmLi (1970) . Ten markers of known mol . wts and the hemorrhagins
Bothrops asper Hemorrhagin Isolation
113 9
were run separately in non-reducing conditions on 10% polyacrylamide gel in the presence of 0 .1% SDS dissolved in 0.05 M Tris-glycine buffer, pH 8 .3 . The gel was stained with 0 .025% Coomassie brilliant blue in 25% isopropanol and 10% acetic acid solution. Isoelectric focusing Four markers of known pI (glucose oxidase from Aspergillus niger, trypsin inhibitor from soybean, ßlactoglobulin A from bovine milk and carbonic anhydrase II from bovine erythrocytes) and the HHmorrhagic fractions were run separately in gels prepared as follows: the markers or each hemorrhagic fraction were mixed with 2 ml of 10% polyacrylamide containing 1 % carrier ampholine with a pH range of 3-10. Each mixture was then inserted to a hollow tube and left under a light bulb for 2 hr to polymerize . Two-hundred volts were applied overnight and then 300 V for 2 hr. The anode solution was 1 % phosphoric acid and the cathode 1 % NaOH . The gels were fixed with 10% TCA and the proteins were stained by 0.025% Coomassie brilliant blue as above . Immroroelectrophoresis Immunoelectrophoresis was performed on 1 % agar gels for 3 hr at 300 V in sodium-Tris barbital buffer (I = 0.04, pH 8.8) before the antiserum was placed in the central trough overnight at room temperature. After three washings with saline for 3 days, the gels were dried and the precipitin lines were stained for 10 min with Ponceau S (0.5% in TCA 5%); destaining was carried out with several changes of 5% acetic acid . Immunodiffusion Immunodiffusion was performed in 1% agar gels according to the procedure of ODCHTERLoNY (1967) . The antibodies were placed in the central or peripheral wells and the whole venom or the purified factors were poured into the alternate wells at a distance of 0 .9 cm . After overnight incubation at room temperature the gels were washed and stained with Ponceau S as above . Periodic acid Schiff (PAS) staining PAS staining of the HHmorrhagic fractions run in an SDS-PAGE was carried out according to OVADIA (1978) . After the run, overnight fixation was done with 7 .5% acetic acid, then the gels were washed with distilled water for I hr and oxidized with 0 .2% sodium metaperiodate for I hr at 4°C; the gel was then washed for 10 min with distilled water and stained with Schiff reagent for 1 hr in the refrigerator . General proteolytic activity (a) Proteolytic activity on azocoll (insoluble dye-protein complex) was examined according to the method of MooRE (1969). One hundred micrograms of the examined fraction was incubated separately with 10 nlg of azocoll suspended in 1 ml of 0 .1 M phosphate buffer, pH 7 .0, for 5 hr at 37°C in a shaking bath (120 strokes per min), and the reaction was stopped by centrifugation at 2500 x g. Proteolytic activity was estimated by measuring the absorbance of the clear supernatant at 520 rum. (b) Esterolytic activity on N-benzoyl-l.-arginine ethyl ester (BAEE) was examined according to the method of SCHWERT and TAKENAKA (1955) . Aliquots containing 5-20 jug of each fraction examined were mixed with 1 ml of 0 .05 M glycine buffer, pH 9 .5, with 0 .5 nM N-benzoyl-t.-arginine ethyl ester as a substrate . The proteolytic activity was determined by measuring the change in absorbance at 254 nm continuously while the reaction was carried out at 37°C. (c) The rate of hydrolysis of benzoyl-t.-tyrosine ethyl ester (BTEE) was determined according to Hutrtmt. (1959) with some modifications. Aliquots containing 5-20 gg of each fraction examined were mixed with 1 ml of 0.08 M Tris, 0 .1 M CaCh, pH 7.8 with 0 .5 mM N-benzoyl-l,-tyrosine ethyl ester as a substrate . The proteolytic activity was determined by measuring the change in absorbance at 256 nm continuously while the reaction was carried out at 25°C. Effect of enzyme inhibitors on the hemorrhagic activity Inhibitors of metaloproteinases, acid proteases and serine proteases were examined. Twenty MHD of whole venom, or of the purified factors, were incubated separately with 0 .5 ml aliquots of 0 .01 M phosphate buffer, pH 7, containing 0.1-0 mM of tetrasodium EDTA, 0.1-8 mM o-phenanthroline, 0.1-1 mM of phenylmethylsulfonyl fluoride (PMSF), 0.05-1 mM phosphoramidon, 0.008-0.08 mM pepstatin A, 0 .06-0 .6 mM aprotinin, or 0.2% soybean trypsin inhibitor for 2 hr at 37°C . One-tenth of a milliliter of each sample containing 5 MHD was
1140
G. BORKOW et al.
1200 1 1000
4
0
á
30
0 E
Ô
TUBE NUMBER FIG. I . GEL FILTRATION OF Bothrops asper VENOM ON SEPHACRYL 5-200 . One-hundred milligrams of venom were dissolved in 1 ml of PBS and the mixture was applied to a 2 .3 x 80 cm column of Sephacryl 5-200. The column was eluted with 0 .01 M phosphate buffer, pH 7 .2, at a flow rate of 10 mllhr. The effluent was collected in 3 ml fractions . Hemorrhagic activity of the fractions was examined by intracutaneous injections to the back of white mice . The active fractions 32-38 were pooled (SI) .
injected into the skin of the back of white mice . Whole venom, or the purified factors which were incubated with the buffer only without the inhibitors, were used as controls . The hemorrhagic activity was examined as described above . RESULTS Purification
Purification of the main hemorrhagic factors BaH 1, BH2 and BH3 involved three steps. Step 1 : Gel filtration on Sephacryl S-200 (Fig. 1). Lyophilized venom (100-300 mg) dissolved in 2 ml of 0.01 M phosphate buffer, pH 7.2, was filtered through a column of TABLE 1 . ISOLATION OF THREE HEMORRHAGIC COMFONENTs FROM THE VENOM OF
Fraction Whole venom 5-200 S1 S2 S3 DEAE Sepharose DI MCACf of D1 BHI BH2 BH3 Phenyl Sepharose of BHl BaHI Mixture of BaHl, BH2 and BH3
mg
MHD (ug)
Bothrops asper
Total MHD
SA'
Yield (%)
100
10
10,000
1
100
12 25 10
2 10 25
6000 2500 400
5
60 25 4
1 .1
0 .2
5500
50
55
0 .6 0 .1 0 .2
0 .2 2 1 .6
3000 50 125
50 5 6 .25
30 0 .5 1 .25
0 .54 0 .84
0 .18 0.18
3000 4667
55 .5 55 .5
30 46.7
`SA: Specific activity ; tMCAC : Metal chelate affinity chromatography.
Bothrops mper Hemorrhagin Isolation
300
700
soo soo
ao
200
.. .
..
0
0 le le 00 40 00w70w00100
0 0 1e 0e 0e 4e 00 00 7e 0e 90 0" Frff011011
Fraction
C
300
1:
90
100
0
so ~16
100
t00
80
~
A
180
300
100
b
ao too
400
180
0
300
114 1
300
1200
d
3000
1000
250
too
goo
too
l000
180
600
ISO
1800
100
400
!80
so
200
e
10 :e ee 40 ge "0 70 Do eet"
Fraction
3
too
so
0
i
é
2500
~
~ ~
;
i
looo
soo
0 o to ae eo 4o so eo 7o w ae tw Fri~
FIG. 2. DEAE-SEPHAROSE CHROMATOGRAPHY OF S1 . Pooled fraction S1 from the 5-200 Sephacryl chromatography was applied to a 1 x 15 cm column of DEAE-Sepharose, equilibrated with 0.01 M phosphate buffer, pH 7.2 . A: The column was eluted with a continuous gradient of salt (0 .01-0.33 M NaCl ; 100 m1-100 ml), at a flow rate of 6 ml/hr. B: 2 M NaCl was added to the column . The effluent was collected in 2 ml fractions. The HHmorrhagic, general proteolytic and esterolytic activities of the DEAF-Sepharose fractions were examined as described in the Methods section . (a) Hemorrhagic activity, (b) azocollase activity, (c) BAEEase activity, and (d) BTEE activity . The strongest enzymatic activities appeared in fractions that did not exhibit hemorrhagic activity . The hatched area was pooled (Dl) .
Sephacryl 5200 (2 .3 x 90 cm). The elution was performed with 0.01 M phosphate buffer, pH 7.2, at a flow rate of 2 ml/cmz/hr and the effluent was collected in 4 ml fractions (Fig . 1) . The hemorrhagic activity of each fraction was examined as described in Methods. Three hemorrhagic peaks were found in three different protein fractions. The strongest hemorrhagic activity was in tubes 34-37 (S1) which contained about 60% of the original venom activity with a specific activity five times higher than that of the crude venom (Table 1) . A weaker hemorrhagic activity was found in tubes 48-51 (S2) and in tubes 60-62(S3). Step 2: DEAF-Sepharose chromatography (Fig . 2) . Pooled fraction S1 of the previous step was subjected to ion exchange chromatography on 10 ml column of DEAE-Sepharose equilibrated with 0.01 M phosphate buffer, pH 7.2 . The column was washed with the equilibration buffer before the elution with 0-0.3 M NaCl gradient . The hemorrhagic activity was found in-between tubes 43-64 (Fig. 2A). The main active fractions (53-59), which contained about 55% of the original hemorrhagic activity, with a specific activity 50 times higher than that of the crude venom (Table 1), were pooled .
1142
G. BORKOW et at. 0 .15
0.10 E
ô
0.05
0.00
0
5
10
15
20
25
30
TUBE NUMBER FIG . 3 . METAL CHELATE AFFINITY CHROMATOGRAPHY OF DI .
Pooled fraction D1 from the DEAF chromatography was applied to a I x 15 cm column of Chelating Sepharose Fast Flow equilibrated with 4 mg/ml Zna 2 and 0.05 M phosphate buffer, pH 7.5 . The column was washed with 0.05 M phosphate buffer, pH 7.5, and then 10 ml of 0.5 M NaCl were added. Glycine (0 .5 M) was added after the second peak was eluted .
Step 3 . Metal chelate affinity chromatography (Fig. 3). The active fractions of step 2 (D1) were pooled and subjected on 10 ml of chelating Sepharose fast flow charged with 10 vols of 4 mg/ml ZnC1Z, washed with water and equilibrated with 0.05 M phosphate buffer, pH 7.5. The first peak (BH1), which was not absorbed by the column, contained 30% of the original activity, and had a specific activity 50 times higher than the crude
0.15
0 w w E Ô
0.10
0.05
0.00
0
5
10
15
20
25
TUSE NLI ABER
4. HYDROPHOBIC INTERACTION CHROMATOGRAPHY OF BH 1. The pooled BH1 fractions of the previous step were mixed with 4 M NaCI and applied to a I ml Phenyl Sepharose CL-4B column equilibrated with 4 M NaCl . The column was washed with 4 M NaCl . After a small peak eluted a decreasing gradient of NaCl (4-0 .01 M NaCl; 30 ml-30 ml) was applied. The column was then washed with water and a second peak eluted (BaHl) . The hemorrhagic activity was found in the second peak . FIG .
Bothrops asper Hemorrhagin Isolation
FIG. S. DETERMINATION OF THE MOLECULAR WEIGHT OF TBE ISOLATED HEMORRHAGIN3 BY SDS-ELECI1topHORESIs.
The mol. wt was determined according to the method of LAEKKU (1970). Electrophoresis was conducted under non-reducing conditions as described in the Methods, followed by staining with PAS procedure (lanes a and b) or by Coomassie brilliant blue (lanes c, d and e) . Lanes a and b: BaHl after PAS staining; lane c: Mol. wt markers; lanes d and e: BaH l stained by Coomassie brilliant blue.
FIG.
6.
IMMUNOELEc-rRopHoRh3I3 wAs PERFORMED IN 1 % AGAR GETS IN SODIUM-TRIS BARBITAL BUFFER (pH 8.8, 0.04 IONIC STRENGTH) AT 300 V, FoR 3 HR.
The purified fraction BaHI was placed in the lower well and whole venom was placed in the upper well . Antibodies against whole venom were placed in the central trough after the electrophoresis.
1143
1144
G . BORKOW et al.
43
5.4
FIG . 7 . ISOELECTROFOCUSING WAS CONDUCTED FOR 20 hr IN 2 ml OF 10% POLYACRYLAMIDE GELS CONTAINING I% CARRIER AMPHOLINES WITH A pH RANGE OF 3-10 . (a) Five micrograms of each of four markers of known pI; (b) I O ug of BaH 1 at both sides of the
markers .
FIG . 8 . IMMÜNODwFUSION WAs PERFORMED IN 1% AGAR GELS ACCORDING TO THE PROCEDURE OF OucHTERLoNY (1958) .
Antibodies against Bothrops riper were placed in the central wells . The four peripheral wells were at a distance of 0.9 cm and contained purified BaHI (a) and inactivated BaHI (b). Inactivation was carried out at 60°C for 20 min . The inactivated BaHI lost its antigenicity partially and formed a weaker precipitin line with the antibodies.
Bothrops asper Hemorrhagin Isolation
1145
TABLE 2. CHARACrFIUZATION OF THE HEMORRHAGINs BaH 1, BH2, BH3
Thermostability (up to °C) pH stability Mol. Wt Isoelectric point Ammonium sulfate precipitation (%)
BaHI
BH2
BH3
50°C 5.5-9.5 64,000 4.5 40-55
50°C 5.5-9.5 26,000 5.2 40-55
50°C 5.5-9.5 55,000 5 40-55
venom (Table 1). A second peak (BH2) was eluted from the column with 0.5 M NaCl, and a third peak (BH3) was eluted with 0.5 M glycine. BH2 and BH3 had 8-10 times weaker hemorrhagic activity than BH1 (Table 1) and a yield of 0.5% and 1 .25%, respectively. However, when various mixtures of 1/8, 1/4 or 1/2 MHD of BH1, BH2 and BH3 were mixed again, the hemorrhagic activity increased about two-fold indicating the synergism between these factors. For example, a mixture between 1/8 MHD of each hemorrhagin caused the appearance of a hemorrhagic spot of 1 MHD. Step 4: Phenyl Sepharose chromatography (Fig. 4). The main hemorrhagic factor of the previous step (BH1) was further purified by the use of hydrophobic interaction chromatography . The ionic strength of the BH1 solution was raised by mixing it with 4 M NaCl . The mixture was then applied to a 1 ml Phenyl Sepharose CL-4B column equilibrated with 4 M NaCl . A peak eluted immediately while another eluted only after the column was washed with water. The hemorrhagic activity was found in the second peak (BaHl, Bothrops asper hemorrhagin 1). The MHD of the active fraction was 0.18 ug . Homogeneity of the main hemorrhagic factor-BaHl
The main hemorrhagin BaHl appeared as one protein band when examined by the following methods: SDS-PAGE electrophoresis (Fig. 5), immunoelectrophoresis (Fig. 6), TABLE 3. CROSS-REACITVnY Of THE WHOLE VENOM AND THE ISOLATED HEMORRHAGINS WITH ANTISERA PREPARED AGAINST THE VENOMS LISTED
Antisera Bothrops asper Pseudocerastes persicus freldi Vipera labetina Vipera latifi Vipera ammodytes Vipera palaestinae Vipera russeli Bitis arietans Bilis gabonica Cerastes cerastes Crotalus atrox Crotalus durWus terricus Echis coloratus Walterirutesia aegyptia
Whole venom
BaH l
BH2
BH3
+ -
+ -
+ -
+ -
+ + + + -
+ -
+ -
+ -
+ + + + -
+ -
+ -
+ + -
1146
G . BORKOW et al . TABLE
4.
EFFE ΠOF vARIOUs ENZYME mHmrroRs HEMORRHAGIC ACrMTY oF DI FRACTION
Concentration (-M) No inhibitor Na4 EDTA Phenanthroline Phosphoramidon
Pepstatin A
Aprotinin
PMSF
Soybean trypsin inhibitor
0 .1 Z 0.5 0 .5 .1 2 0 .05 0 .1 0 .2 0 .5 1 0 .01 0 .02 0 .04 0.08 0.07 0.14 0.28 0.55 0.1 0.2 0.5 1 2 100
ON
THE
Hemorrhage + + + + + + + + + + + + + + + + + + + + + +
+ : Appearance of hemorrhage (no inhibition) . From five different DEAE Sepharose runs (second purification step, see Table 1 and Fig. 2), 3 ug of fraction D1 were incubated with the various enzyme inhibitors at 37°C for 1 hr in 0.5 ml aliquots of 0.01 M phosphate buffer, pH 7 . The hemorrhagic activity was determined by injecting 0.1 ml (3 MHD) into the back skin of white mice .
isoelectrofocusing (Fig. 7) and immunodiffusion (Fig. 8). BH2 was also homogenic but BH3 showed some minor contamination when examined by SDS-PAGE electrophoresis . Characterization Nine criteria were used to characterize the hemorrhagic activity. A. Thermostability. The stability of the hemorrhagic activity of the whole venom and of the isolated hemorrhagins was tested in different temperatures in a series of experiments. Half milliliter PBS aliquots containing 10 minimal hemorrhagic doses (MHD) of each purified factor or of the whole venom were incubated at various temperatures between 37°C and 100° C for 30 min. One hundred microliters from each sample (2 MHD) were injected into the skin of the back of mice and 2 hr later the hemorrhage was compared with that caused by the untreated samples. The hemorrhagic activity was thermolabile and was completely lost at 60° C (Table 2). B. Stability at different pH values . The stability of the hemorrhagic activity of the whole venom and of the purified hemorrhagins was tested in pH values between 1 and 12. Eight MHD of each purified factor or of the whole venom were incubated for 2 hr at 30°C in 0.3 ml buffer solutions of different pH values, followed by titration to pH 7 by 0.1 ml of
Bothrops asper Hemorrhagin Isolation
1147
0.5 M phosphate buffer. One-tenth milliliter containing 2 MHD of each purified factor or of the whole venom was injected into the skin of the back of mice. The hemorrhagic activity of all examined samples was completely lost after being incubated at pH values under 5 or above 10, but was stable at pH 5.5-9.5 (Table 2). C. Ammonium sulfate precipitation . Ammonium sulfate fractionation was carried out as described in Methods. The hemorrhagic activity appeared in the precipitates of 40-55% ammonium sulfate saturation (Table 2). D. Molecular weight. The mol. wt of the three hemorrhagins was determined by SDS-PAGE as described in Methods. The mol. wts of BaHl, BH2 and BH3 were 64,000, 26,000 and 55,000, respectively (Fig. 5 and Table 2). E. PAS staining. BaH1, BH2 and BH3 were stained by the PAS procedure as described in the Methods. They were stained by PAS, indicating that the hemorrhagic factors are glycoproteins (Fig. 5). F. Isoelectric point. The isoelectric points were determined by isoelectric focusing as described in Methods. Each of the purified hemorrhagic fractions, BaHl, BH2 and BH3, showed one acidic band. The pIs of BaHl (Fig . 7), BH2 and BH3 were 4 .5, 5.2 and 5, respectively . G. Antigenicity of the whole venom and the isolated hemorrhagins after heat inactivation. The hemorrhagic activity of the whole venom and that of the isolated factors is completely lost when heated for 15 min at 60°C (see previous section) . However, the heated fractions still form precipitin lines with the antiserum prepared against the untreated venom (Fig. 8). H. Cross-reactivity with various antibodies. Cross-reactivity of the whole venom and the isolated hemorrhagins with antisera against various venoms was examined by an immunodiffusion test and is summarized in Table 3. It could be shown that BaHI, BH2 and BH3 reacted with antisera prepared against the venoms of Vipera palaestinae and Echis colorata in addition to the reaction with the homologous antiserum of Bothrops asper. BH3 reacted also with the antisera against Cerastes cerastes. However, the crude venom of Bothrops asper also reacted with heterologous antisera. I. Effect of enzyme inhibitors on the hemorrhagic activity . The importance of bivalent cations for the hemorrhagic activity was indicated by the ability of EDTA or ophenanthroline to completely inhibit the hemorrhagic activity (Table 4). On the other hand, phosphoramidon, PMSF, pepstatin A, aprotinin and soybean trypsin inhibitor did not affect the hemorrhagic activity (the concentrations used were up to ten-fold higher than the concentrations generally used to inhibit enzymatic activities). The effect on the whole venom was similar to that on the purified hemorrhagins . DISCUSSION
Envenomation due to snakebite constitutes a relevant public health problem in most parts of the world and especially in the developing countries (W.H .O., 1981). The majority of snakebites in Central America are caused by pit vipers, and the most dangerous species among them is Bothrops asper (BOLANOS, 1982). Among the local and systemic effects caused by this venom, hemorrhage represents an especially difficult problem. It was, therefore, necessary to purify the main hemorrhagic factors from the venom of Bothrops asper in order to investigate the nature and the mechanism of action of these toxins . The hemorrhagins of Bothrops asper are similar in their characteristics to other hemorrhagins isolated from viperid or crotalid snake venoms (BJARNASON and Fox, 1988-89). They are
1148
G . BORKOW et al.
glycoproteins which are thermolabile and sensitive to extreme pH values ; the hemorrhagic activity of the isolated factors was stable only at pH 5.5-9.5 and was completely lost at 60°C . Like most hemorrhagins, the isolated toxins were acidic with similar p1s between 4.5 and 5.2, but had different mol. wts. The main hemorrhagin BaH 1 is the largest molecule (mol . wt 64,000), whereas BH2 and BH3 were 26,000 and 55,000, respectively . It was previously suggested classifying the hemorrhagins into three groups depending upon their mol. wts (BJARNASON and Fox, 1988-89) : small (20,000-30,000), medium (30,000-60,000), and large toxins (60,000-100,000) which are also the most active ones . According to this classification, the venom of Bothrops asper contains all three classes of the hemorrhagins . Moreover, the largest hemorrhagin BaHI is also the most potent . It was interesting to find that the main contribution of BH2 and BH3 was not by the direct induction of hemorrhage but rather by a synergistic activity with the main hemorrhagin BaH1 . One-quarter MHD of each hemorrhagin caused a vague hemorrhagic spot; however, a mixture between one eight MHD of each hemorrhagin caused the appearance of a hemorrhagic spot of one MHD. While BH2 and BH3 contribute only 0.5-1 .25% of the total hemorrhagic activity of the whole venom (compared to 30% of BaHl), the activity of the mixture of the three factors is almost doubled and contains 50% of the total hemorrhagic activity of the whole venom (Table 1) . This synergistic activity could explain the appearance of several hemorrhagins in many snake venoms which may act on different substrates and enhance the hemorrhage . This study also shows the existence of cross-reactivity of the isolated hemorrhagins with the antivenoms of Vipera palaestinae, Echis coloratus and Cerastes cerastes, in addition to the homologous antivenom. This is in accordance with previous reports which demonstrate the appearance of cross-reactivity between homologous and heterologous antivenoms and hemorrhagins of different snake venoms both by means of immunodiffusion (MiNToN, 1957, 1979) and by means of neutralization of the lethal, defibrinating and hemorrhagic activities (MEBS et al., 1988 ; KORNALIK and TABORSKA, 1989). The hemorrhagic activity of the whole venom and of the isolated factors is completely lost when heated for 15 min at 60°C . However, the heated fractions form precipitin lines with the antiserum prepared against the untreated venom, indicating that the antigenicity is not lost . This raises the possibility of injecting heated non-hemorrhagic venom or fractions into animals for preparing antisera which may inflict less pain and minor damage to the animals. It is generally accepted that hemorrhagins have proteolytic activity (Tu, 1983; OWNBY and GEREN, 1987 ; SANCHEz et al., 1987; BJARNAsoN and Fox, 1988-89; BARAMOVA et al., 1989). However, the question whether the proteolysis of the basement membrane is sufficient to induce the hemorrhage is still open, as OWNBY et al. (1978) have shown, i.e. endothelial cells were also lysed in addition to the disruption of the basement membrane. Ultrastructural observations made by this group also showed that the hemorrhage induced by the venom of Bothrops asper included morphological alterations of the cells together with gaps in their continuity and digestion of the basal laminae (MoREiRA et al ., 1992). Similar results were reported by other groups (BJARNASON and Fox, 1988-89) and were recently shown by RAHMY et al. (1992) with the venom of Cerastes cerastes . This may indicate activity on the cells in addition to the disruption of the extracellular matrix. However, these activities may be related, as it is well established that cell morphology is influenced by the matrix . Hemorrhagic activity was reduced by inhibitors of metalloproteinases such as EDTA and ortho-phenanthroline, but not by other protease inhibitors (Table 4) . These results are
Bothrops asper Hemorrhagin Isolation
1149
in agreement with the previous studies in which five hemorrhagins isolated from three other species of Bothrops (MANDELBAum et al., 1976, 1984) appeared to be proteolytic enzymes and all but one have been shown to be metal dependent. Further studies are, therefore, under way in this laboratory in order to elucidate the relevant activity of these isolated hemorrhagins . One study will examine the digestion of various isolated extracellular matrix components such as collagens, protoglycans, laminin and fibronectin; another will examine their activity on endothelial cells cultured in vitro, and a third investigation will examine the ultrastructural effects caused by the main hemorrhagin, BaH 1 . Acknowledgement-This study was supported by AID Grant No. DHR-5544-6-00-1064-00 .
REFERENCES BARAMOVA, E. N ., SHANNON, J. D., BJARNASON, J . B. and Fox, J . W. (1989) Degradation of extracellular matrix proteins by hemorrhagic metalloproteinases. Archs Biochem. Biophys . 275, 63-71 . BJARNASON, J . B. and Fox, J. W . (1988-89) Hemorrhagic toxins of snake venoms. J. Toxicol.-Toxin Rev . 7, 121-209. BOLANos, R . (1982) Las serpientes venenosas de Centroamerica y el problema del ofidismo . Primers parte . Aspectos zoologicos, epidemiologicos y biomedicus . Revta cost . Cienc . med. 3, 165-184 . BoLANos, R . and CERDAs, L. (1980) Produecion y control de sueros antiofidicos en Costa Rica . Boln Of. sanit. pan-am . 88, 184-196 . HuMMEL, B . C . W. (1959) Modified spectrophotometric determination of chymotrypsin, trypsin and thrombin . Can . J. Biochem . Physiol. 37, 1393 . KoNDo, H ., KoNDo, S ., IKEZAWA, H., MuRATA, R . and OHSAKA, A . (1960) Studies on the quantitative method for determination of hemorrhagic activity of habu snake venom. Jpn . J. med. Sci. Biol. 13, 43 . KORNALIK, F . and TAeoRsKA, E . (1989) Cross reactivity of mono- and polyvalent antivenoms with viperidae and crotalidae snake venoms . Toxicon 27, 1135-1142 . LAEMMLi, U . K . (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 . MANDELBAUM, F., REIcHL, A . P. and ASSAKURA, M . T . (1976) Some physical and biochemical characteristics of HF2, one of the hemorrhagic factors in the venom of Bothrops jararaca . In : Animal Plant and Microbial Toxins, Vol . 1 (OHsAKA, A ., HAYASHI, K. and SAWA], Y ., Eds) . New York : Plenum Press . MANDELBAUM, F., AssAKuRA, M . T . and REIcJn ., A . P . (1984) Characterization of two hemorrhagic factors isolated from the venom of Bothrops neuwiedi (Jararaca pintada). Toxicon 22, 193-206 . MARTIN, P. S . (1958) The biogeography of amphibians and reptiles in the Gomes Farias region, Tamaulipas, Mexico . Univ . Mich . Mise . Publ. Mus. Zool. 101, 1-102. Was, D ., POHLMANN, S. and WON TENSPOLDE, W . (1988) Snake venom hemorrhagins : Neutralization by commercial antivenoms . Toxicon 26, 453-458 . MINTON, S . A. (1957) An immunological investigation of rattlesnake venoms by the agar diffusion method . Am . J. trop. Med. Hyg. 6,1097-1107 . MINTON, S . A . (1969) In: Venomous Reptiles, p . 75. New York : Scribner's Sons . MINTON, S . A . (1979) Common antigens in snake venoms . In : Handbook ofExperimental Pharmacology, Vol. 52, Snake Venoms, pp . 847-862 (LEE, C. Y., Ed .) . Berlin: Springer. MOORE, G. L . (1969) Use of azo-dye bound collagen top measure reaction velocities of proteolytic enzymes. Analyt. Biochem . 32, 122-127. MoREiRA, L ., GunERREz, J. M ., BoRKow, G . and OvADIA, M . (1992) Ultrastructural alterations in mouse capillary blood vessels after experimental injections of venom from the snake Bothrops riper (Terciopelo). Expl . Mal . Path. 57, 124-133 . OHSAKA, A. (1979) Hemorrhagic, necrotizing and edema-forming effects of snake venoms . In: Snake Venoms, pp. 480-546 (LEE, C . Y ., Ed.) . Berlin : Springer . OucHTERLoNY, O. (1967) In : Handbook of Experimental Immunology, pp . 665-706 (WEIR, D . M ., Ed .). Philadelphia : Davis Company . OVARIA, M . (1978) Isolation and characterization of three hemorrhagic factors from the venom of Vipera palaestinae. Toxicon 16, 479-87. OwNBY, C. L . and GEREN, C . R. (1987) Pathogenesis of hemorrhage induced by hemorrhagic proteinase IV from timber rattlesnake (Crotalus horridus horridus) venom . Toxicon 25, 517-526.
1150
G. BORKOW et al.
OWNBY, C. L., BJARNASON, J. and Tu, A. T. (1978) Hemorrhagic toxins from rattlesnake (Crotales atrox) venom. Am . J. Path . 93, 201-212. PETERS, J. A. and OREIAs-MIRANDA, B. (1970) Catalogue of the Neotropical Squamata . Part 1: Snakes. Bull. U.S. Nat. Mus. 297, 1-347. RAHMY, T., Tu, A. T., EL-BANHAwEY, M. A., EL-ASMAR, M. F. and HASSAN, F. M. (1992) Cytopathologic effect of Cerastes Crastes (Egyptian sand viper) venom and isolated hemorrhagic toxin on liver and kidney: an electron microscopic study. J. Not. Toxins 1, 45-58. SANCHEZ, E. F., MAGAI HAEs , A. and DINtz, C. R. (1987) Purification of a hemorrhagc factor (LHF-1) from the venom of the bushmaster snake, Lachesis muta muta . Toxicon 25, 611-619. SCHWERT, J. A. and TAKENAKA, Y. (1955) A spectrophotomeric determination of trypsin and chymotrypsin . Biochim. Biophys. Acta 16, 570. Tu, A. T. (1983) Local tissues damaging (hemorrhage and myonecresis) toxins from rattle-snakes and other pit viper venoms . J. Toxicol.-Toxin Rev. 2, 205-210. WORLD HEALTH ORGANIZATION (1981) Progress in the characterization of venoms and standardization of antivenoms . WHO Offset Public . 51;, 1-43 .
0041-0101/93 $6.00 + .00 ® 1993 Pergarnon Pren Ltd
ISOLATION AND CHARACTERIZATION OF SYNERGISTIC HEMORRHAGINS FROM THE VENOM OF THE SNAKE BOTHROPS ASPER GADI BORKOw, 1 Jost MARIA GuTd?RREzz and MICHAEL OVADIA 1
'Department of Zoology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel ; and ZInstituto Clodomiro Picado, Facultad de Microbiologia Universidad de Costa Rica, Costa Rica (Received 29 July 1992 ; accepted 25 March 1993)
G. BORKOW, J. M. GUTtlftRREZ and M. OVADIA. Isolation and characterization of synergistic hemorrhagins from the venom of the snake Bothrops asper. Toxicon 31, 1137-1150, 1993 .-Three hemorrhagic factors (BaHl, 13112 and BH3) were isolated from the venom of Bothrops asper by gel filtration on Sephacryl S-200, DEAE-Sepharose chromatography, metal chelate affinity chromatography and hydrophobic interaction chromatography . They contain 55% of the total hemorrhagic activity of the whole venom when they are mixed, but lose almost half of the activity if they are separated, indicating a synergism between the three. The main hemorrhagin is BaH l (Bothrops asper hemorrhagin 1) ; the other two are weak hemorrhagins but contribute to the synergism. They are acidic proteins with a pI of 4.5, 5.2 and 5; their mol. wt is 64,000, 26,000 and 55,000 respectively . The minimal hemorrhagic dose (MHD) of BaH1, BH2 and BH3 is 0.18, 2 and 1 .61íg, with a specific activity 55, 5 and 6.25 higher than that of the whole venom. The hemorrhagic activity of all three factors was inhibited by EDTA and ortho-phenathroline, indicating that the hemorrhagic activity is metal dependent. Phosphoramidon, soybean trypsin inhibitor, PMSF, pepstatin and aprotinin did not affect the hemorrhagic activity of the isolated factors.
THE
INTRODUCTION
MmoRITy of snakebites in Central America are caused by pit vipers . Among them locally known as `terciopelo' or `barba amarilla', is one of the most dangerous and abundant species (MINTON, 1969 ; BOLANOS, 1982). It ranges from northern Mexico to the lowlands of Colombia and Ecuador (MARTIN, 1958 ; PETERS and OREJAS-MIRANDA, 1970). The venom of this snake induces a complex series of local effects including hemorrhage, myonecrosis and edema, in addition to systemic effects such as hemorrhage, coagulation disorders, cardiovascular shock and acute renal failure (BoLANos, 1982). Among these effects hemorrhage represents an especially difficult problem mainly because it develops quickly after venom injection causing major microvasculature damage and blood loss which leads to muscle and other tissue degeneration. Similar striking consequences of envenomation, due to the occurrence of local and
Bothrops asper,
*Author to whom correspondence should be addressed. 1137
113 8
G . BORKOW et al.
peripheral hemorrhage, are caused by most crotalid and viperid snake venoms (OWNBY et al., 1978; OHsA", 1979). Forty-three hemorrhagins from 15 species were isolated and characterized (BJARNSON and Fox, 1988-89). Most of them are acidic proteins with mot. wts ranging from 20,000 to 100,000. All but two appear to be metal-dependent proteolytic enzymes. The present communication reports the isolation and characterization of three hemorrhagic factors from the venom of Bothrops asper that can act synergistically . MATERIALS AND METHODS Materials Sephacryl S-200, DEAE-Sepharose, Phenyl Sepharose CL-4B and Chelating Sepharose Fast Flow were purchased from Pharmacia Fine Chemicals (Uppsala, Sweden) ; Coomassie brilliant blue and the enzyme inhibitors were obtained from Sigma Chemical Co . (St . Louis, MO, U.S .A .) ; acrylamide and NN'-methylenebisacrylamide were obtained from BDH (Poole, U.K .); all other chemicals were analytically pure and were purchased from Merck (Darmstadt, F.R .G.). Bothrops asper venom The snakes were kept and 'milked' at Instituto Clodomiro Picado, University of Costa Rica. The aliquots of the 'milked' venom were pooled, lyophilized and stored at -20°C until used . Antisera Antiserum against Bothrops riper venom was prepared in Instituto Clodomiro Picado, Costa Rica, by injecting increasing amounts of venom at l0-day intervals s.c. into horses . The antigenic mixture contained equal parts of the venoms of Bothrops asper, Crotalus durissus durissus and Lachesis muia stenophrys (BoLANOS and CERDAs, 1980). Freund's complete adjuvant or alginate was used as adjuvant . At the end of immunization, horses were bled from the jugular vein and plasma was fractionated by ammonium sulfate precipitation (BOL .ANos and CERDAs, 1980) . Antisera against Vipera palaestinae, Walterinnesia aegyptia, Echis coloratus, Crotalus atrox and Cerastes cerastes were prepared in rabbits in Tel Aviv university as above ; the blood was drained from a vein in the ear . Antisera against Vipera persicus, Vipera lebetina and Vipera latifii were purchased from Razi Institute, Iran; antisera against Crotalus durissus terrificus from Butantan Institute, Brazil ; antisera against Bitis arietans and Bilis gabonica from Institute Pasteur, Paris; and antisera against Vipera russelli from the Thai Red Cross Society, Thailand . Determination of hemorrhagic activity The hemorrhagic activity of the whole venom and its fractions was tested by the skin injection method of KONDo et al. (1960) with some modifications of OvADiA (1978) . One-tenth of a milliliter of saline containing 0.05-20 lug of the examined material was injected into the skin at the back of white mice . Two hours later a red hemorrhagic spot was observed on the inner surface of the removed skin . The amount of venom or its fractions that produced a hemorrhagic spot of about 1 cm in diameter was defined as one minimal hemorrhagic dose (MHD) . The method is semi-quantitative . Ammonium sulfate precipitation Cold saturated ammonium sulfate was added to the whole venom or to each of the purified factors in a cold room under continuous stirring until 40% saturation was achieved . After 30 min, the suspensions were centrifuged for 20 min at 10,000 x g and the precipitate was redissolved in 0 .01 M phosphate buffer, pH 7 . More saturated ammonium sulfate was added to each of the supernatant fluids to attain 55% saturation and 30 min later the suspensions were treated as above followed by a dialysis against the same buffer. The hemorrhagic activity of the fractions was examined as described above . Determination of mol. wt The mol. wt was determined by the sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) method according to the procedure of LAEmmLi (1970) . Ten markers of known mol . wts and the hemorrhagins
Bothrops asper Hemorrhagin Isolation
113 9
were run separately in non-reducing conditions on 10% polyacrylamide gel in the presence of 0 .1% SDS dissolved in 0.05 M Tris-glycine buffer, pH 8 .3 . The gel was stained with 0 .025% Coomassie brilliant blue in 25% isopropanol and 10% acetic acid solution. Isoelectric focusing Four markers of known pI (glucose oxidase from Aspergillus niger, trypsin inhibitor from soybean, ßlactoglobulin A from bovine milk and carbonic anhydrase II from bovine erythrocytes) and the HHmorrhagic fractions were run separately in gels prepared as follows: the markers or each hemorrhagic fraction were mixed with 2 ml of 10% polyacrylamide containing 1 % carrier ampholine with a pH range of 3-10. Each mixture was then inserted to a hollow tube and left under a light bulb for 2 hr to polymerize . Two-hundred volts were applied overnight and then 300 V for 2 hr. The anode solution was 1 % phosphoric acid and the cathode 1 % NaOH . The gels were fixed with 10% TCA and the proteins were stained by 0.025% Coomassie brilliant blue as above . Immroroelectrophoresis Immunoelectrophoresis was performed on 1 % agar gels for 3 hr at 300 V in sodium-Tris barbital buffer (I = 0.04, pH 8.8) before the antiserum was placed in the central trough overnight at room temperature. After three washings with saline for 3 days, the gels were dried and the precipitin lines were stained for 10 min with Ponceau S (0.5% in TCA 5%); destaining was carried out with several changes of 5% acetic acid . Immunodiffusion Immunodiffusion was performed in 1% agar gels according to the procedure of ODCHTERLoNY (1967) . The antibodies were placed in the central or peripheral wells and the whole venom or the purified factors were poured into the alternate wells at a distance of 0 .9 cm . After overnight incubation at room temperature the gels were washed and stained with Ponceau S as above . Periodic acid Schiff (PAS) staining PAS staining of the HHmorrhagic fractions run in an SDS-PAGE was carried out according to OVADIA (1978) . After the run, overnight fixation was done with 7 .5% acetic acid, then the gels were washed with distilled water for I hr and oxidized with 0 .2% sodium metaperiodate for I hr at 4°C; the gel was then washed for 10 min with distilled water and stained with Schiff reagent for 1 hr in the refrigerator . General proteolytic activity (a) Proteolytic activity on azocoll (insoluble dye-protein complex) was examined according to the method of MooRE (1969). One hundred micrograms of the examined fraction was incubated separately with 10 nlg of azocoll suspended in 1 ml of 0 .1 M phosphate buffer, pH 7 .0, for 5 hr at 37°C in a shaking bath (120 strokes per min), and the reaction was stopped by centrifugation at 2500 x g. Proteolytic activity was estimated by measuring the absorbance of the clear supernatant at 520 rum. (b) Esterolytic activity on N-benzoyl-l.-arginine ethyl ester (BAEE) was examined according to the method of SCHWERT and TAKENAKA (1955) . Aliquots containing 5-20 jug of each fraction examined were mixed with 1 ml of 0 .05 M glycine buffer, pH 9 .5, with 0 .5 nM N-benzoyl-t.-arginine ethyl ester as a substrate . The proteolytic activity was determined by measuring the change in absorbance at 254 nm continuously while the reaction was carried out at 37°C. (c) The rate of hydrolysis of benzoyl-t.-tyrosine ethyl ester (BTEE) was determined according to Hutrtmt. (1959) with some modifications. Aliquots containing 5-20 gg of each fraction examined were mixed with 1 ml of 0.08 M Tris, 0 .1 M CaCh, pH 7.8 with 0 .5 mM N-benzoyl-l,-tyrosine ethyl ester as a substrate . The proteolytic activity was determined by measuring the change in absorbance at 256 nm continuously while the reaction was carried out at 25°C. Effect of enzyme inhibitors on the hemorrhagic activity Inhibitors of metaloproteinases, acid proteases and serine proteases were examined. Twenty MHD of whole venom, or of the purified factors, were incubated separately with 0 .5 ml aliquots of 0 .01 M phosphate buffer, pH 7, containing 0.1-0 mM of tetrasodium EDTA, 0.1-8 mM o-phenanthroline, 0.1-1 mM of phenylmethylsulfonyl fluoride (PMSF), 0.05-1 mM phosphoramidon, 0.008-0.08 mM pepstatin A, 0 .06-0 .6 mM aprotinin, or 0.2% soybean trypsin inhibitor for 2 hr at 37°C . One-tenth of a milliliter of each sample containing 5 MHD was
1140
G. BORKOW et al.
1200 1 1000
4
0
á
30
0 E
Ô
TUBE NUMBER FIG. I . GEL FILTRATION OF Bothrops asper VENOM ON SEPHACRYL 5-200 . One-hundred milligrams of venom were dissolved in 1 ml of PBS and the mixture was applied to a 2 .3 x 80 cm column of Sephacryl 5-200. The column was eluted with 0 .01 M phosphate buffer, pH 7 .2, at a flow rate of 10 mllhr. The effluent was collected in 3 ml fractions . Hemorrhagic activity of the fractions was examined by intracutaneous injections to the back of white mice . The active fractions 32-38 were pooled (SI) .
injected into the skin of the back of white mice . Whole venom, or the purified factors which were incubated with the buffer only without the inhibitors, were used as controls . The hemorrhagic activity was examined as described above . RESULTS Purification
Purification of the main hemorrhagic factors BaH 1, BH2 and BH3 involved three steps. Step 1 : Gel filtration on Sephacryl S-200 (Fig. 1). Lyophilized venom (100-300 mg) dissolved in 2 ml of 0.01 M phosphate buffer, pH 7.2, was filtered through a column of TABLE 1 . ISOLATION OF THREE HEMORRHAGIC COMFONENTs FROM THE VENOM OF
Fraction Whole venom 5-200 S1 S2 S3 DEAE Sepharose DI MCACf of D1 BHI BH2 BH3 Phenyl Sepharose of BHl BaHI Mixture of BaHl, BH2 and BH3
mg
MHD (ug)
Bothrops asper
Total MHD
SA'
Yield (%)
100
10
10,000
1
100
12 25 10
2 10 25
6000 2500 400
5
60 25 4
1 .1
0 .2
5500
50
55
0 .6 0 .1 0 .2
0 .2 2 1 .6
3000 50 125
50 5 6 .25
30 0 .5 1 .25
0 .54 0 .84
0 .18 0.18
3000 4667
55 .5 55 .5
30 46.7
`SA: Specific activity ; tMCAC : Metal chelate affinity chromatography.
Bothrops mper Hemorrhagin Isolation
300
700
soo soo
ao
200
.. .
..
0
0 le le 00 40 00w70w00100
0 0 1e 0e 0e 4e 00 00 7e 0e 90 0" Frff011011
Fraction
C
300
1:
90
100
0
so ~16
100
t00
80
~
A
180
300
100
b
ao too
400
180
0
300
114 1
300
1200
d
3000
1000
250
too
goo
too
l000
180
600
ISO
1800
100
400
!80
so
200
e
10 :e ee 40 ge "0 70 Do eet"
Fraction
3
too
so
0
i
é
2500
~
~ ~
;
i
looo
soo
0 o to ae eo 4o so eo 7o w ae tw Fri~
FIG. 2. DEAE-SEPHAROSE CHROMATOGRAPHY OF S1 . Pooled fraction S1 from the 5-200 Sephacryl chromatography was applied to a 1 x 15 cm column of DEAE-Sepharose, equilibrated with 0.01 M phosphate buffer, pH 7.2 . A: The column was eluted with a continuous gradient of salt (0 .01-0.33 M NaCl ; 100 m1-100 ml), at a flow rate of 6 ml/hr. B: 2 M NaCl was added to the column . The effluent was collected in 2 ml fractions. The HHmorrhagic, general proteolytic and esterolytic activities of the DEAF-Sepharose fractions were examined as described in the Methods section . (a) Hemorrhagic activity, (b) azocollase activity, (c) BAEEase activity, and (d) BTEE activity . The strongest enzymatic activities appeared in fractions that did not exhibit hemorrhagic activity . The hatched area was pooled (Dl) .
Sephacryl 5200 (2 .3 x 90 cm). The elution was performed with 0.01 M phosphate buffer, pH 7.2, at a flow rate of 2 ml/cmz/hr and the effluent was collected in 4 ml fractions (Fig . 1) . The hemorrhagic activity of each fraction was examined as described in Methods. Three hemorrhagic peaks were found in three different protein fractions. The strongest hemorrhagic activity was in tubes 34-37 (S1) which contained about 60% of the original venom activity with a specific activity five times higher than that of the crude venom (Table 1) . A weaker hemorrhagic activity was found in tubes 48-51 (S2) and in tubes 60-62(S3). Step 2: DEAF-Sepharose chromatography (Fig . 2) . Pooled fraction S1 of the previous step was subjected to ion exchange chromatography on 10 ml column of DEAE-Sepharose equilibrated with 0.01 M phosphate buffer, pH 7.2 . The column was washed with the equilibration buffer before the elution with 0-0.3 M NaCl gradient . The hemorrhagic activity was found in-between tubes 43-64 (Fig. 2A). The main active fractions (53-59), which contained about 55% of the original hemorrhagic activity, with a specific activity 50 times higher than that of the crude venom (Table 1), were pooled .
1142
G. BORKOW et at. 0 .15
0.10 E
ô
0.05
0.00
0
5
10
15
20
25
30
TUBE NUMBER FIG . 3 . METAL CHELATE AFFINITY CHROMATOGRAPHY OF DI .
Pooled fraction D1 from the DEAF chromatography was applied to a I x 15 cm column of Chelating Sepharose Fast Flow equilibrated with 4 mg/ml Zna 2 and 0.05 M phosphate buffer, pH 7.5 . The column was washed with 0.05 M phosphate buffer, pH 7.5, and then 10 ml of 0.5 M NaCl were added. Glycine (0 .5 M) was added after the second peak was eluted .
Step 3 . Metal chelate affinity chromatography (Fig. 3). The active fractions of step 2 (D1) were pooled and subjected on 10 ml of chelating Sepharose fast flow charged with 10 vols of 4 mg/ml ZnC1Z, washed with water and equilibrated with 0.05 M phosphate buffer, pH 7.5. The first peak (BH1), which was not absorbed by the column, contained 30% of the original activity, and had a specific activity 50 times higher than the crude
0.15
0 w w E Ô
0.10
0.05
0.00
0
5
10
15
20
25
TUSE NLI ABER
4. HYDROPHOBIC INTERACTION CHROMATOGRAPHY OF BH 1. The pooled BH1 fractions of the previous step were mixed with 4 M NaCI and applied to a I ml Phenyl Sepharose CL-4B column equilibrated with 4 M NaCl . The column was washed with 4 M NaCl . After a small peak eluted a decreasing gradient of NaCl (4-0 .01 M NaCl; 30 ml-30 ml) was applied. The column was then washed with water and a second peak eluted (BaHl) . The hemorrhagic activity was found in the second peak . FIG .
Bothrops asper Hemorrhagin Isolation
FIG. S. DETERMINATION OF THE MOLECULAR WEIGHT OF TBE ISOLATED HEMORRHAGIN3 BY SDS-ELECI1topHORESIs.
The mol. wt was determined according to the method of LAEKKU (1970). Electrophoresis was conducted under non-reducing conditions as described in the Methods, followed by staining with PAS procedure (lanes a and b) or by Coomassie brilliant blue (lanes c, d and e) . Lanes a and b: BaHl after PAS staining; lane c: Mol. wt markers; lanes d and e: BaH l stained by Coomassie brilliant blue.
FIG.
6.
IMMUNOELEc-rRopHoRh3I3 wAs PERFORMED IN 1 % AGAR GETS IN SODIUM-TRIS BARBITAL BUFFER (pH 8.8, 0.04 IONIC STRENGTH) AT 300 V, FoR 3 HR.
The purified fraction BaHI was placed in the lower well and whole venom was placed in the upper well . Antibodies against whole venom were placed in the central trough after the electrophoresis.
1143
1144
G . BORKOW et al.
43
5.4
FIG . 7 . ISOELECTROFOCUSING WAS CONDUCTED FOR 20 hr IN 2 ml OF 10% POLYACRYLAMIDE GELS CONTAINING I% CARRIER AMPHOLINES WITH A pH RANGE OF 3-10 . (a) Five micrograms of each of four markers of known pI; (b) I O ug of BaH 1 at both sides of the
markers .
FIG . 8 . IMMÜNODwFUSION WAs PERFORMED IN 1% AGAR GELS ACCORDING TO THE PROCEDURE OF OucHTERLoNY (1958) .
Antibodies against Bothrops riper were placed in the central wells . The four peripheral wells were at a distance of 0.9 cm and contained purified BaHI (a) and inactivated BaHI (b). Inactivation was carried out at 60°C for 20 min . The inactivated BaHI lost its antigenicity partially and formed a weaker precipitin line with the antibodies.
Bothrops asper Hemorrhagin Isolation
1145
TABLE 2. CHARACrFIUZATION OF THE HEMORRHAGINs BaH 1, BH2, BH3
Thermostability (up to °C) pH stability Mol. Wt Isoelectric point Ammonium sulfate precipitation (%)
BaHI
BH2
BH3
50°C 5.5-9.5 64,000 4.5 40-55
50°C 5.5-9.5 26,000 5.2 40-55
50°C 5.5-9.5 55,000 5 40-55
venom (Table 1). A second peak (BH2) was eluted from the column with 0.5 M NaCl, and a third peak (BH3) was eluted with 0.5 M glycine. BH2 and BH3 had 8-10 times weaker hemorrhagic activity than BH1 (Table 1) and a yield of 0.5% and 1 .25%, respectively. However, when various mixtures of 1/8, 1/4 or 1/2 MHD of BH1, BH2 and BH3 were mixed again, the hemorrhagic activity increased about two-fold indicating the synergism between these factors. For example, a mixture between 1/8 MHD of each hemorrhagin caused the appearance of a hemorrhagic spot of 1 MHD. Step 4: Phenyl Sepharose chromatography (Fig. 4). The main hemorrhagic factor of the previous step (BH1) was further purified by the use of hydrophobic interaction chromatography . The ionic strength of the BH1 solution was raised by mixing it with 4 M NaCl . The mixture was then applied to a 1 ml Phenyl Sepharose CL-4B column equilibrated with 4 M NaCl . A peak eluted immediately while another eluted only after the column was washed with water. The hemorrhagic activity was found in the second peak (BaHl, Bothrops asper hemorrhagin 1). The MHD of the active fraction was 0.18 ug . Homogeneity of the main hemorrhagic factor-BaHl
The main hemorrhagin BaHl appeared as one protein band when examined by the following methods: SDS-PAGE electrophoresis (Fig. 5), immunoelectrophoresis (Fig. 6), TABLE 3. CROSS-REACITVnY Of THE WHOLE VENOM AND THE ISOLATED HEMORRHAGINS WITH ANTISERA PREPARED AGAINST THE VENOMS LISTED
Antisera Bothrops asper Pseudocerastes persicus freldi Vipera labetina Vipera latifi Vipera ammodytes Vipera palaestinae Vipera russeli Bitis arietans Bilis gabonica Cerastes cerastes Crotalus atrox Crotalus durWus terricus Echis coloratus Walterirutesia aegyptia
Whole venom
BaH l
BH2
BH3
+ -
+ -
+ -
+ -
+ + + + -
+ -
+ -
+ -
+ + + + -
+ -
+ -
+ + -
1146
G . BORKOW et al . TABLE
4.
EFFE ΠOF vARIOUs ENZYME mHmrroRs HEMORRHAGIC ACrMTY oF DI FRACTION
Concentration (-M) No inhibitor Na4 EDTA Phenanthroline Phosphoramidon
Pepstatin A
Aprotinin
PMSF
Soybean trypsin inhibitor
0 .1 Z 0.5 0 .5 .1 2 0 .05 0 .1 0 .2 0 .5 1 0 .01 0 .02 0 .04 0.08 0.07 0.14 0.28 0.55 0.1 0.2 0.5 1 2 100
ON
THE
Hemorrhage + + + + + + + + + + + + + + + + + + + + + +
+ : Appearance of hemorrhage (no inhibition) . From five different DEAE Sepharose runs (second purification step, see Table 1 and Fig. 2), 3 ug of fraction D1 were incubated with the various enzyme inhibitors at 37°C for 1 hr in 0.5 ml aliquots of 0.01 M phosphate buffer, pH 7 . The hemorrhagic activity was determined by injecting 0.1 ml (3 MHD) into the back skin of white mice .
isoelectrofocusing (Fig. 7) and immunodiffusion (Fig. 8). BH2 was also homogenic but BH3 showed some minor contamination when examined by SDS-PAGE electrophoresis . Characterization Nine criteria were used to characterize the hemorrhagic activity. A. Thermostability. The stability of the hemorrhagic activity of the whole venom and of the isolated hemorrhagins was tested in different temperatures in a series of experiments. Half milliliter PBS aliquots containing 10 minimal hemorrhagic doses (MHD) of each purified factor or of the whole venom were incubated at various temperatures between 37°C and 100° C for 30 min. One hundred microliters from each sample (2 MHD) were injected into the skin of the back of mice and 2 hr later the hemorrhage was compared with that caused by the untreated samples. The hemorrhagic activity was thermolabile and was completely lost at 60° C (Table 2). B. Stability at different pH values . The stability of the hemorrhagic activity of the whole venom and of the purified hemorrhagins was tested in pH values between 1 and 12. Eight MHD of each purified factor or of the whole venom were incubated for 2 hr at 30°C in 0.3 ml buffer solutions of different pH values, followed by titration to pH 7 by 0.1 ml of
Bothrops asper Hemorrhagin Isolation
1147
0.5 M phosphate buffer. One-tenth milliliter containing 2 MHD of each purified factor or of the whole venom was injected into the skin of the back of mice. The hemorrhagic activity of all examined samples was completely lost after being incubated at pH values under 5 or above 10, but was stable at pH 5.5-9.5 (Table 2). C. Ammonium sulfate precipitation . Ammonium sulfate fractionation was carried out as described in Methods. The hemorrhagic activity appeared in the precipitates of 40-55% ammonium sulfate saturation (Table 2). D. Molecular weight. The mol. wt of the three hemorrhagins was determined by SDS-PAGE as described in Methods. The mol. wts of BaHl, BH2 and BH3 were 64,000, 26,000 and 55,000, respectively (Fig. 5 and Table 2). E. PAS staining. BaH1, BH2 and BH3 were stained by the PAS procedure as described in the Methods. They were stained by PAS, indicating that the hemorrhagic factors are glycoproteins (Fig. 5). F. Isoelectric point. The isoelectric points were determined by isoelectric focusing as described in Methods. Each of the purified hemorrhagic fractions, BaHl, BH2 and BH3, showed one acidic band. The pIs of BaHl (Fig . 7), BH2 and BH3 were 4 .5, 5.2 and 5, respectively . G. Antigenicity of the whole venom and the isolated hemorrhagins after heat inactivation. The hemorrhagic activity of the whole venom and that of the isolated factors is completely lost when heated for 15 min at 60°C (see previous section) . However, the heated fractions still form precipitin lines with the antiserum prepared against the untreated venom (Fig. 8). H. Cross-reactivity with various antibodies. Cross-reactivity of the whole venom and the isolated hemorrhagins with antisera against various venoms was examined by an immunodiffusion test and is summarized in Table 3. It could be shown that BaHI, BH2 and BH3 reacted with antisera prepared against the venoms of Vipera palaestinae and Echis colorata in addition to the reaction with the homologous antiserum of Bothrops asper. BH3 reacted also with the antisera against Cerastes cerastes. However, the crude venom of Bothrops asper also reacted with heterologous antisera. I. Effect of enzyme inhibitors on the hemorrhagic activity . The importance of bivalent cations for the hemorrhagic activity was indicated by the ability of EDTA or ophenanthroline to completely inhibit the hemorrhagic activity (Table 4). On the other hand, phosphoramidon, PMSF, pepstatin A, aprotinin and soybean trypsin inhibitor did not affect the hemorrhagic activity (the concentrations used were up to ten-fold higher than the concentrations generally used to inhibit enzymatic activities). The effect on the whole venom was similar to that on the purified hemorrhagins . DISCUSSION
Envenomation due to snakebite constitutes a relevant public health problem in most parts of the world and especially in the developing countries (W.H .O., 1981). The majority of snakebites in Central America are caused by pit vipers, and the most dangerous species among them is Bothrops asper (BOLANOS, 1982). Among the local and systemic effects caused by this venom, hemorrhage represents an especially difficult problem. It was, therefore, necessary to purify the main hemorrhagic factors from the venom of Bothrops asper in order to investigate the nature and the mechanism of action of these toxins . The hemorrhagins of Bothrops asper are similar in their characteristics to other hemorrhagins isolated from viperid or crotalid snake venoms (BJARNASON and Fox, 1988-89). They are
1148
G . BORKOW et al.
glycoproteins which are thermolabile and sensitive to extreme pH values ; the hemorrhagic activity of the isolated factors was stable only at pH 5.5-9.5 and was completely lost at 60°C . Like most hemorrhagins, the isolated toxins were acidic with similar p1s between 4.5 and 5.2, but had different mol. wts. The main hemorrhagin BaH 1 is the largest molecule (mol . wt 64,000), whereas BH2 and BH3 were 26,000 and 55,000, respectively . It was previously suggested classifying the hemorrhagins into three groups depending upon their mol. wts (BJARNASON and Fox, 1988-89) : small (20,000-30,000), medium (30,000-60,000), and large toxins (60,000-100,000) which are also the most active ones . According to this classification, the venom of Bothrops asper contains all three classes of the hemorrhagins . Moreover, the largest hemorrhagin BaHI is also the most potent . It was interesting to find that the main contribution of BH2 and BH3 was not by the direct induction of hemorrhage but rather by a synergistic activity with the main hemorrhagin BaH1 . One-quarter MHD of each hemorrhagin caused a vague hemorrhagic spot; however, a mixture between one eight MHD of each hemorrhagin caused the appearance of a hemorrhagic spot of one MHD. While BH2 and BH3 contribute only 0.5-1 .25% of the total hemorrhagic activity of the whole venom (compared to 30% of BaHl), the activity of the mixture of the three factors is almost doubled and contains 50% of the total hemorrhagic activity of the whole venom (Table 1) . This synergistic activity could explain the appearance of several hemorrhagins in many snake venoms which may act on different substrates and enhance the hemorrhage . This study also shows the existence of cross-reactivity of the isolated hemorrhagins with the antivenoms of Vipera palaestinae, Echis coloratus and Cerastes cerastes, in addition to the homologous antivenom. This is in accordance with previous reports which demonstrate the appearance of cross-reactivity between homologous and heterologous antivenoms and hemorrhagins of different snake venoms both by means of immunodiffusion (MiNToN, 1957, 1979) and by means of neutralization of the lethal, defibrinating and hemorrhagic activities (MEBS et al., 1988 ; KORNALIK and TABORSKA, 1989). The hemorrhagic activity of the whole venom and of the isolated factors is completely lost when heated for 15 min at 60°C . However, the heated fractions form precipitin lines with the antiserum prepared against the untreated venom, indicating that the antigenicity is not lost . This raises the possibility of injecting heated non-hemorrhagic venom or fractions into animals for preparing antisera which may inflict less pain and minor damage to the animals. It is generally accepted that hemorrhagins have proteolytic activity (Tu, 1983; OWNBY and GEREN, 1987 ; SANCHEz et al., 1987; BJARNAsoN and Fox, 1988-89; BARAMOVA et al., 1989). However, the question whether the proteolysis of the basement membrane is sufficient to induce the hemorrhage is still open, as OWNBY et al. (1978) have shown, i.e. endothelial cells were also lysed in addition to the disruption of the basement membrane. Ultrastructural observations made by this group also showed that the hemorrhage induced by the venom of Bothrops asper included morphological alterations of the cells together with gaps in their continuity and digestion of the basal laminae (MoREiRA et al ., 1992). Similar results were reported by other groups (BJARNASON and Fox, 1988-89) and were recently shown by RAHMY et al. (1992) with the venom of Cerastes cerastes . This may indicate activity on the cells in addition to the disruption of the extracellular matrix. However, these activities may be related, as it is well established that cell morphology is influenced by the matrix . Hemorrhagic activity was reduced by inhibitors of metalloproteinases such as EDTA and ortho-phenanthroline, but not by other protease inhibitors (Table 4) . These results are
Bothrops asper Hemorrhagin Isolation
1149
in agreement with the previous studies in which five hemorrhagins isolated from three other species of Bothrops (MANDELBAum et al., 1976, 1984) appeared to be proteolytic enzymes and all but one have been shown to be metal dependent. Further studies are, therefore, under way in this laboratory in order to elucidate the relevant activity of these isolated hemorrhagins . One study will examine the digestion of various isolated extracellular matrix components such as collagens, protoglycans, laminin and fibronectin; another will examine their activity on endothelial cells cultured in vitro, and a third investigation will examine the ultrastructural effects caused by the main hemorrhagin, BaH 1 . Acknowledgement-This study was supported by AID Grant No. DHR-5544-6-00-1064-00 .
REFERENCES BARAMOVA, E. N ., SHANNON, J. D., BJARNASON, J . B. and Fox, J . W. (1989) Degradation of extracellular matrix proteins by hemorrhagic metalloproteinases. Archs Biochem. Biophys . 275, 63-71 . BJARNASON, J . B. and Fox, J. W . (1988-89) Hemorrhagic toxins of snake venoms. J. Toxicol.-Toxin Rev . 7, 121-209. BOLANos, R . (1982) Las serpientes venenosas de Centroamerica y el problema del ofidismo . Primers parte . Aspectos zoologicos, epidemiologicos y biomedicus . Revta cost . Cienc . med. 3, 165-184 . BoLANos, R . and CERDAs, L. (1980) Produecion y control de sueros antiofidicos en Costa Rica . Boln Of. sanit. pan-am . 88, 184-196 . HuMMEL, B . C . W. (1959) Modified spectrophotometric determination of chymotrypsin, trypsin and thrombin . Can . J. Biochem . Physiol. 37, 1393 . KoNDo, H ., KoNDo, S ., IKEZAWA, H., MuRATA, R . and OHSAKA, A . (1960) Studies on the quantitative method for determination of hemorrhagic activity of habu snake venom. Jpn . J. med. Sci. Biol. 13, 43 . KORNALIK, F . and TAeoRsKA, E . (1989) Cross reactivity of mono- and polyvalent antivenoms with viperidae and crotalidae snake venoms . Toxicon 27, 1135-1142 . LAEMMLi, U . K . (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 . MANDELBAUM, F., REIcHL, A . P. and ASSAKURA, M . T . (1976) Some physical and biochemical characteristics of HF2, one of the hemorrhagic factors in the venom of Bothrops jararaca . In : Animal Plant and Microbial Toxins, Vol . 1 (OHsAKA, A ., HAYASHI, K. and SAWA], Y ., Eds) . New York : Plenum Press . MANDELBAUM, F., AssAKuRA, M . T . and REIcJn ., A . P . (1984) Characterization of two hemorrhagic factors isolated from the venom of Bothrops neuwiedi (Jararaca pintada). Toxicon 22, 193-206 . MARTIN, P. S . (1958) The biogeography of amphibians and reptiles in the Gomes Farias region, Tamaulipas, Mexico . Univ . Mich . Mise . Publ. Mus. Zool. 101, 1-102. Was, D ., POHLMANN, S. and WON TENSPOLDE, W . (1988) Snake venom hemorrhagins : Neutralization by commercial antivenoms . Toxicon 26, 453-458 . MINTON, S . A. (1957) An immunological investigation of rattlesnake venoms by the agar diffusion method . Am . J. trop. Med. Hyg. 6,1097-1107 . MINTON, S . A . (1969) In: Venomous Reptiles, p . 75. New York : Scribner's Sons . MINTON, S . A . (1979) Common antigens in snake venoms . In : Handbook ofExperimental Pharmacology, Vol. 52, Snake Venoms, pp . 847-862 (LEE, C. Y., Ed .) . Berlin: Springer. MOORE, G. L . (1969) Use of azo-dye bound collagen top measure reaction velocities of proteolytic enzymes. Analyt. Biochem . 32, 122-127. MoREiRA, L ., GunERREz, J. M ., BoRKow, G . and OvADIA, M . (1992) Ultrastructural alterations in mouse capillary blood vessels after experimental injections of venom from the snake Bothrops riper (Terciopelo). Expl . Mal . Path. 57, 124-133 . OHSAKA, A. (1979) Hemorrhagic, necrotizing and edema-forming effects of snake venoms . In: Snake Venoms, pp. 480-546 (LEE, C . Y ., Ed.) . Berlin : Springer . OucHTERLoNY, O. (1967) In : Handbook of Experimental Immunology, pp . 665-706 (WEIR, D . M ., Ed .). Philadelphia : Davis Company . OVARIA, M . (1978) Isolation and characterization of three hemorrhagic factors from the venom of Vipera palaestinae. Toxicon 16, 479-87. OwNBY, C. L . and GEREN, C . R. (1987) Pathogenesis of hemorrhage induced by hemorrhagic proteinase IV from timber rattlesnake (Crotalus horridus horridus) venom . Toxicon 25, 517-526.
1150
G. BORKOW et al.
OWNBY, C. L., BJARNASON, J. and Tu, A. T. (1978) Hemorrhagic toxins from rattlesnake (Crotales atrox) venom. Am . J. Path . 93, 201-212. PETERS, J. A. and OREIAs-MIRANDA, B. (1970) Catalogue of the Neotropical Squamata . Part 1: Snakes. Bull. U.S. Nat. Mus. 297, 1-347. RAHMY, T., Tu, A. T., EL-BANHAwEY, M. A., EL-ASMAR, M. F. and HASSAN, F. M. (1992) Cytopathologic effect of Cerastes Crastes (Egyptian sand viper) venom and isolated hemorrhagic toxin on liver and kidney: an electron microscopic study. J. Not. Toxins 1, 45-58. SANCHEZ, E. F., MAGAI HAEs , A. and DINtz, C. R. (1987) Purification of a hemorrhagc factor (LHF-1) from the venom of the bushmaster snake, Lachesis muta muta . Toxicon 25, 611-619. SCHWERT, J. A. and TAKENAKA, Y. (1955) A spectrophotomeric determination of trypsin and chymotrypsin . Biochim. Biophys. Acta 16, 570. Tu, A. T. (1983) Local tissues damaging (hemorrhage and myonecresis) toxins from rattle-snakes and other pit viper venoms . J. Toxicol.-Toxin Rev. 2, 205-210. WORLD HEALTH ORGANIZATION (1981) Progress in the characterization of venoms and standardization of antivenoms . WHO Offset Public . 51;, 1-43 .