- Email: [email protected]
Physiological and immunological properties of small myotoxins from the venom of the midget faded rattlesnake (Crotalus viridis concolor)
Tm6cm, Vol. 26, No. 3, pp. 319-323, 1998. Printed in Great Britain
OD41-0101/88 $3.00+ .00 ® 1989 Pergamon Prea plc
SHORT COMMUNICATIONS
PHYSIOLOGICAL AND IMMUNOLOGICAL PROPERTIES OF SMALL MYOTOXINS FROM THE VENOM OF THE MIDGET FADED RATTLESNAKE (CROTALUS VIRIDIS CONCOLOR) CHARLOTTE L. OWNBY I , STEVEN D. AIRD2 and IVAN I . KAISEO* 'Department of Physiological Sciences, Oklahoma State University, Stillwater, OK 74078-0353, and 'Department of Molecular Biology, University Station, Box 3944, University of Wyoming, Laramie, WY 82071, U .S .A . (Accepted for publication 28 July 1987) C . L . OwNBY, S . D . AIRD and I. I . KAISER. Physiological and immunological properties of small myotoxins from the venom of the midget faded rattlesnake (Crotalus viridis concolor). Toxicon 26, 319 - 323, 1988 . - Myotoxins from C. v. concolor venom were isolated by gel filtration. This crude myotoxin peak was subfractionated into either two or four subfractions by cation exchange FPLC, depending upon the source of the venom . When injected at 2 pg/g, crude concolor myotoxin caused vacuolation of mouse muscle cells typical of myotoxin a from C. v. viridis and crotamine from C. d. terrkricus. All four subfractions showed qualitatively identical myotoxin activity . In double immunodiffusion studies, myotoxin a antiserum produced lines of identity when reacted with myotoxin a, crude concolor myotoxin and the four concolor subfractions. A second batch of material showed two major components when subfractionated by cation exchange FPLC . The more basic of these two components displayed approximately twice the i .v. lethality of the more acidic component . The LD for the basic component lies between 0 .625 and 0.75 ltg/g while that of the acidic component falls between 1 .00 and 1 .25 ilg/g .
GONÇALVES
(1956) reported the presence of a highly basic polypeptide from the venom of
Crotalus durissus terrificus, which he named crotamine. Since that time, crotamine-like
proteins have been reported in the venoms of C. v. viridis (CAMERON and Tu, 1977; Fox et al., 1979 ; OwNBY et al., 1976), C. v. heferi (MAEDA et al., 1978), C adamanteus (MEBS et al., 1983; MESS and KoRNALIK, 1984), C. h. horridus (MEBs et al., 1983) and C. v. concolor (BIEBER et al., 1986; ENGLE et al., 1983) . Although the exact biological mode of action of these myotoxins is not known, it is clear that they affect skeletal muscle . OWNBY et al. (1976) reported that myotoxin a from C. v. viridis venom caused necrosis of skeletal muscle cells after i.m. injection. The initial ultrastructural change was dilatation of the sarcoplasmic reticulum, seen as vacuolation at the light microscopic level. OwNBY et al. (1976) suggested that this dilatation could result from alteration of the permeability of the sarcolemma to sodium . This hypothesis has been supported by CHANG and TSENG (1978) using crotamine, and by HONG and CHANG (1985) using myotoxin a. Others, however, propose that these myotoxins act directly on the calcium channels of the sarcoplasmic reticulum Clu and MoRrrA, 1983; VOLPE et al., 1986). Although the basic mechanism by which the toxin acts is unresolved, its first microscopically observable effect on muscle *To whom correspondence should be addressed. 319
320
Short Communications
FIG .
1.
IMMUNODDTFUSION
ASSAYS OF
C. v. C0nc010r SUEFILAMONS .
CRUDE
MYOTOXIN AND MYOTOXIN
Center well: rabbit antimyotoxin a sera . Peripheral wells : (1) myotoxin a, 1 mg/ml (2) and (4) 5-200 Fraction 4, 1 mg/ml; (3), (5), (6) empty . (B) Center well : antimyotoxin a sera. Peripheral wells : (1) myotoxin a, 1 mg/ml ; (2) FPLC Fraction 1, 1 mg/ml; (3) FPLC Fraction 2, 1 mg/ml ; (4) FPLC Fraction 3, 1 mg/ml ; (5) FPLC Fraction 4, 1 mg/ml ; (6) empty. (A)
cells, i .e. vacuolation, is clear. The purpose of this work was to determine whether the minor structural differences between isotoxins isolated from one batch of venom resulted in any biological and immunological differences . Crude venom of C. v. concolor, obtained from the Miami Serpentarium (lot no. CK15SZ) and extracted from specimens collected in Rio Blanco County, Colorado was pooled and fractionated over Sephacryl S-200 equilibrated with 0.1 M sodium acetate (pH 4.0) as previously reported (AIRD and KAISER, 1985b; AIRD, 1985) . The fourth S-200 fraction (based on 280 nm absorbance), contained the myotoxins. This fraction was dialyzed against deionized water using benzoylated dialysis tubing (Sigma, 2000 MW cutoff), lyophilized, and dissolved in 0.85% NaCI to a final concentration of 1 mg/ml. This fraction was examined in mice for myotoxic activity and by immunodiffusion for cross-reactivity with rabbit polyclonal antisera raised against myotoxin a from C. v. viridis (OwNBY et al., 1979) . For biological assays, crude myotoxin (Fraction 4 from Sephacryl S-200 column) was injected in a dose of 2 Mg/g. Two female mice (CD-1, Charles River) weighing 25 g were injected with 0.1 ml i.m. into the dorsolateral aspect of the thigh; two mice were injected with 0.85% NaCI as a control . The mice were killed by cervical dislocation 24 hr after the injection, and a sample of muscle was removed from the ventromedial aspect of the injected thigh. After processing for microscopy as previously described (OWNBY et al., 1976), tissues were examined by light microscopy. Fraction 4 from 5-200 induced vacuolation in skeletal muscle cells identical to that caused by both myotoxin a (OWNBY et al., 1976) and crotamine (CAMERON and Tu, 1978). Crude myotoxin was examined by Ouchterlony double immunodiffusion (OWNBY et al., 1979). Microdiffusion plates were made with a 1 % agarose solution in 0.85% NaCl. Each well was filled twice with 15 IAl of solution (1 mg/ml), incubated at room temperature for 24 hr and then photographed . Myotoxin a and S-200 Fraction 4 show a reaction of complete identity when reacted against antiserum to myotoxin a (Fig . la). Crude myotoxin was subfractionated by cation exchange FPLC (Pharmacia) using a Mono S column. The initial buffer was 50 mM sodium acetate (pH 4.0), and the final buffer was the same plus 2 MNaCl . A small peak washed off the column in initial buffer prior to the start of the gradient (Fig. 2A). A sharp spike of non-myotoxin material eluted
Short Communications
32 1
2 . SÜBFRACTIONATION OF CRUDE C. v. concolor MYOTOXIN RECOVERED FROM GEL FILTRATION ON A 0.5 x 5 CM COLUMN OF MONO S. (A) This material was used for immunodiffusion plates and necrosis assays . Initial buffer 50 mM sodium acetate (pH 4 .0) ; final buffer, 50 mM sodium acetate (pH 4.0) plus 2 M NaCl . Flow rate was 1 .5 ml/min . (B) Mono S chromatogram of 5-200 Fraction 4 from a later batch of venom obtained from Miami Serpentarium used for LD determinations . This material was subfractionated in 20 mM acetate rather than 50 mM . Individual peaks from both batches of venom show evidence of heterogeneity .
FIG .
from the column immediately after the gradient began. Four myotoxin subfractions, which presumably differ in net charge, constituted virtually all of the A2in-absorbing material in S-200 Fraction 4 and eluted in 0.8 -1 .0 M NaCl (Fig . 2). The latter three show evidence of heterogeneity. The four subfractions were pooled separately, dialyzed exhaustively against deionized water and lyophilized. After lyophilization, stock solutions of each subfraction were prepared in physiological saline (1 .0 mg/ml) and examined by double immunodiffusion. All four subfractions reacted with complete identity to myotoxin a using rabbit antisera to myotoxin a (Fig. lb), indicating common epitopes . All four caused contracture of the hind legs in mice within five min after i.m. injection. Doses for the i.m. injections of the four subfractions were 341, 587, 723, 368 Ftg, respectively (13.8, 23 .5, 28 .9, and 14.6 ug/g). Histological assay showed that all four subfractions induced vacuolation of skeletal muscle cells identical to that caused by myotoxin a and the crude myotoxin fraction from C. v. concolor venom. Results obtained upon i.m . injections of the four subfractions suggested the possibility of lethality differences between different isomers of myotoxins. To examin e this possibility, additional material was purified as described above from crude C. v. concolor venom obtained from the Miami Serpentarium (lot no. CK17SZ). This material showed only two subfractions when the 5-200 Fraction 4 was subfractionated over Mono S in 20 mM acetate buffer (Fig. 2B). These both appeared to be heterogeneous, as had three of the four peaks shown in Fig. 2A. While approximately twice as much material was applied to the Mono S column in the second preparation (410 mg) (Fig . 2B) as in the first (Fig . 2A), a separate run of the second preparation, using approximately 3 mg (not shown),
32 2
Short Communications
failed to resolve any additional components . Thus chromatographic differences between the two preparations cannot be attributed to overloading of the Mono S Column. The two subfractions from this second batch of venom were dialyzed and lyophilized as described above. Intravenous injections were made according to the procedures outlined by AIRD and KAISER (1985a) using female mice of the same genetic strain described therein. No attempt was made to determine whether male and female mice differ in their susceptibilities to the toxin. Mice ranging from 21 to 32 g were given weight-specific doses and two to four mice were employed at each dose. The more acidic isomers (subfraction 1, Fig. 2B) proved less toxic than the more basic isomers (subfraction 2). For subfraction 1, doses x1 .25 pg/g proved fatal to all test animals . Doses < 1 .00 Mg/g were non-lethal to all test animals. By contrast, with subfraction 2, doses >0.75 pg/g were lethal to all specimens, while doses <0.625 pg/g were sublethal. This indicates that LD50S for the two subfractions differ by approximately 2-fold, and that survivorship curves for the two groups of isomers are quite steep. The LD50 was not determined more closely because the lethal and non-lethal doses reported above were so close that it would have been difficult to justify the sacrifice of additional animals to make the determination. The LD50 values reported above for subfractions 1 and 2 fall respectively within and below the ranges reported for E toxin (Crotalus h. horridus) in the presence of acetate (ALLEN et al., 1986). It may be argued that because of chromatography using acetate buffers, the basic amino acid side chains in these myotoxins have acetate bound as a counter-ion; hence their toxicity . Since the eluting cation exchange buffer contained only 20 mM acetate and 2 M NaCl, it is likely that CI - , rather than acetate, constituted the bulk of the counter-ion. Ownby (unpublished observations) has reported similar values for myotoxin a where no acetate was used in any purification step. We find the high LD50 value reported by ALLEN et al. (1986) for E toxin in the absence of exogenous acetate (6.3 pg/g) difficult to explain. It may possibly be attributed to the one-step purification procedure they employed. Trace contamination with proteases might account for the gradual loss of toxicity experienced previously with this toxin (SULLIVAN and GEREN, 1979). Several noteworthy observations were made during toxicity assays . Using a dose of 4.0 pg/g, which is close to the LD50 reported for crotamine by CHEYMOL et al . (1971), mice exhibited flaccid paralysis before the injection could be completed. Death ensued in less than 30 sec. Lethal doses closer to the LD 50 caused death within 2 min; however, ifa mouse survived as much as 5 min, recovery seemed ensured. This is in strong contrast to crotalid presynaptic neurotoxins, which continue to kill test animals as much as 18 - 24 hr postinjection, despite the fact that their toxicity exceeds that of the myotoxins by more than an order of magnitude (AiRD and KAIsER, 1985b). Our observations on mice that received a lethal dose suggest that death may result from respiratory failure, perhaps caused by paralysis of the diaphragm and skeletal musculature controlling movement of the rib cage . Such a conclusion would appear to be supported by the effects of myotoxin a on the resting membrane potential of mouse and rat diaphragms (HONG and CHANG, 1985). Myotoxins from C. v. concolor do not appear to induce death by cardiac failure. When sublethal doses of myotoxin a from C. v. viridis were injected into mice i.v., paraffin sections of cardiac muscle examined by light microscopy failed to reveal any evidence of myonecrosis (Ownby, unpublished observations). We noted that in mice. that received a lethal dose, the heart beat strongly throughout convulsions and continued for several minutes after the cessation of all attempts to breathe.
Short Communications
323
While muscle necrosis is the most significant medical problem associated with the small myotoxins, our observations on predatory behavior of C. v. concolor, C. v. viridis and C. v. helleri indicate that the biological significance of these molecules is their ability to halt the flight of a prey organism almost instantaneously . In many cases, a mouse struck by one of these snakes is unable to take a single step, and the hyperextension of the hind limbs that characterizes injections of purified myotoxins is readily observable . Acknowkdganents - This work was supported in part by the U .S . Army Medical Acquisition Activity, Contract no . DAMD 17-86-C-6061 . REFERENCES AiRD, S . D. (1985) A quantitative assessment of variation in venom constituents within and between three nominal rattlesnake subspecies . Toxkon 23, 1000. AIRD, S . D. and KAISER, I . I . (1985a) Toxicity assays . Taxicon 23, 11 . ACRD, S . D . and KAIgHR, I . I . (1985b) Comparative studies on three rattlesnake toxins. Taxleon 23, 361 . ALLEN, H . R., TucxER, R. K. and OEREN, C . R . (1986) Potentiation of the toxicity of basic peptides from rattlesnake venom& by sodium acetate . Taxkon 24, 553 . BiEwm, A . L ., McFARLAND, R . H . and BacKER, R . R . (1986) Complete sequences for myotoxins from Crotalus vWis concolor venom . Fedn Proc. Fedn Am. Socs exp. Biol. 45, 1794 . CAMERoN, D . L . and Tu, A . T . (1977) Characterization of myotoxin a from the venom of prairie rattlesnake (Crotahm vb1db viridis) . Biochemboy 16, 2546. CAmRON, D . and Tu, A . T . (1978) Chemical and functional homology of myotoxin a from prairie rattlesnake venom and crotamine from South American rattlesnake venom. Biochim. biophys. Acts Sit, 147 . CHANG, C . C . and Tsm4o, K . H . (1978) Effect of crotamine, a toxin of South American rattlesnake venom, on the sodium channel of murine skeletal muscle . Br. J. Pharmac. 63, 551 . CFmYMoL, J ., OoNruvEs, J. M ., BouRtLLer, F . and Rock-ARYEILLER, M . (1971) Action neuromusculaire eomparee de la crotamine et du venin de Crotahn derbsus terrjfkus var. crotaminkus - I . Toxkon 9, 279 . ENGLE, C . M ., BEcxER, R. R., BAILEY, T. and BmaEa, A . L. (1983) Characterization of two myotoxic proteins from venom of Crotaha vhidis concolor. J. Toxic. - Toxin Rev . 2, 267 . Fox, J . W ., ELznvoA, M . and Tu, A. T . (1979) Amino add sequence and disulfide bond assignment of myotoxin a isolated from the venom of prairie rattlesnake (Crotahis viridis virWb). Biochemistry 18, 678 . OoNÇALvEs, J . M . (1956) Purification and properties of crotamine. In: Venoms, p . 261 (BucKLEY E . E . and PoRaEs, N ., Eds) . American Association for the Advancement of Science, Washington DC . HONG, S . J. and CHANG, C . C . (1985) Electrophysiologicalstudies of myotoxin a, isolated from prairie rattlesnake (Crotahn virfdis viridis) venom, on murine skeletal muscles. Toxicon 23, 927 . MAEDA, N., TAMIYA, N ., PATrABwRAMAN, T . R . and RussELL, F . E. (1978) Some chemiéal properties of the venom of the rattlesnake, Crotales viridis viridis. Toxicon 16, 431 . MEBs, D . and KoRNAux, F. (1984) Intraspecific variation in content of a basic toxin in eastern diamondback rattlesnake (Crotaha adamanteus) venom . Toxkon 22, 831 . MEBs, D ., EHREaFELD, M . and SAmanmA, Y . (1983) Local necrotizing effect of snake venoms on skin and muscle: relationship to serum creatine kinase . Toxicon 21, 393 . OwNBY, C . L., CAMERON, D . and Tu, A . T. (1976) Isolation of myotoxic components from rattlesnake (Crotahrs viral& vhidis) venom. Electron microscopic analysis of muscle damage . Am. J. Path . 85, 149 . OwNBY, C. L ., WooDs, W . M. and ODELL, O . V . (1979) Antiserum to myotoxin from prairie rattlesnake (Crotahn viridis viridis) venom . Toxicon 17, 373 . SULuvAN, J . and OEREN, C . R . (1979) Isolation, stabilization, and characterization of a toxin from timber rattlesnake venom. Preparative Biochan. 9, 321 . Tu, A . T. and MoatTA, M . (1983) Attachment of rattlesnake venom myotoxin a to sarcoplasmic reticulum: peroxidase conjugated method . Br. J. Path. 64, 633 . VoLPE, P ., DAUUAm, E ., MAURER, A . and Tu, A. T . (1986) Interaction of myotoxin a with the Ca"-ATPase of skeletal muscle sarcoplasmic reticulum. Archs Blochem . Blophys. 246, 90.
OD41-0101/88 $3.00+ .00 ® 1989 Pergamon Prea plc
SHORT COMMUNICATIONS
PHYSIOLOGICAL AND IMMUNOLOGICAL PROPERTIES OF SMALL MYOTOXINS FROM THE VENOM OF THE MIDGET FADED RATTLESNAKE (CROTALUS VIRIDIS CONCOLOR) CHARLOTTE L. OWNBY I , STEVEN D. AIRD2 and IVAN I . KAISEO* 'Department of Physiological Sciences, Oklahoma State University, Stillwater, OK 74078-0353, and 'Department of Molecular Biology, University Station, Box 3944, University of Wyoming, Laramie, WY 82071, U .S .A . (Accepted for publication 28 July 1987) C . L . OwNBY, S . D . AIRD and I. I . KAISER. Physiological and immunological properties of small myotoxins from the venom of the midget faded rattlesnake (Crotalus viridis concolor). Toxicon 26, 319 - 323, 1988 . - Myotoxins from C. v. concolor venom were isolated by gel filtration. This crude myotoxin peak was subfractionated into either two or four subfractions by cation exchange FPLC, depending upon the source of the venom . When injected at 2 pg/g, crude concolor myotoxin caused vacuolation of mouse muscle cells typical of myotoxin a from C. v. viridis and crotamine from C. d. terrkricus. All four subfractions showed qualitatively identical myotoxin activity . In double immunodiffusion studies, myotoxin a antiserum produced lines of identity when reacted with myotoxin a, crude concolor myotoxin and the four concolor subfractions. A second batch of material showed two major components when subfractionated by cation exchange FPLC . The more basic of these two components displayed approximately twice the i .v. lethality of the more acidic component . The LD for the basic component lies between 0 .625 and 0.75 ltg/g while that of the acidic component falls between 1 .00 and 1 .25 ilg/g .
GONÇALVES
(1956) reported the presence of a highly basic polypeptide from the venom of
Crotalus durissus terrificus, which he named crotamine. Since that time, crotamine-like
proteins have been reported in the venoms of C. v. viridis (CAMERON and Tu, 1977; Fox et al., 1979 ; OwNBY et al., 1976), C. v. heferi (MAEDA et al., 1978), C adamanteus (MEBS et al., 1983; MESS and KoRNALIK, 1984), C. h. horridus (MEBs et al., 1983) and C. v. concolor (BIEBER et al., 1986; ENGLE et al., 1983) . Although the exact biological mode of action of these myotoxins is not known, it is clear that they affect skeletal muscle . OWNBY et al. (1976) reported that myotoxin a from C. v. viridis venom caused necrosis of skeletal muscle cells after i.m. injection. The initial ultrastructural change was dilatation of the sarcoplasmic reticulum, seen as vacuolation at the light microscopic level. OwNBY et al. (1976) suggested that this dilatation could result from alteration of the permeability of the sarcolemma to sodium . This hypothesis has been supported by CHANG and TSENG (1978) using crotamine, and by HONG and CHANG (1985) using myotoxin a. Others, however, propose that these myotoxins act directly on the calcium channels of the sarcoplasmic reticulum Clu and MoRrrA, 1983; VOLPE et al., 1986). Although the basic mechanism by which the toxin acts is unresolved, its first microscopically observable effect on muscle *To whom correspondence should be addressed. 319
320
Short Communications
FIG .
1.
IMMUNODDTFUSION
ASSAYS OF
C. v. C0nc010r SUEFILAMONS .
CRUDE
MYOTOXIN AND MYOTOXIN
Center well: rabbit antimyotoxin a sera . Peripheral wells : (1) myotoxin a, 1 mg/ml (2) and (4) 5-200 Fraction 4, 1 mg/ml; (3), (5), (6) empty . (B) Center well : antimyotoxin a sera. Peripheral wells : (1) myotoxin a, 1 mg/ml ; (2) FPLC Fraction 1, 1 mg/ml; (3) FPLC Fraction 2, 1 mg/ml ; (4) FPLC Fraction 3, 1 mg/ml ; (5) FPLC Fraction 4, 1 mg/ml ; (6) empty. (A)
cells, i .e. vacuolation, is clear. The purpose of this work was to determine whether the minor structural differences between isotoxins isolated from one batch of venom resulted in any biological and immunological differences . Crude venom of C. v. concolor, obtained from the Miami Serpentarium (lot no. CK15SZ) and extracted from specimens collected in Rio Blanco County, Colorado was pooled and fractionated over Sephacryl S-200 equilibrated with 0.1 M sodium acetate (pH 4.0) as previously reported (AIRD and KAISER, 1985b; AIRD, 1985) . The fourth S-200 fraction (based on 280 nm absorbance), contained the myotoxins. This fraction was dialyzed against deionized water using benzoylated dialysis tubing (Sigma, 2000 MW cutoff), lyophilized, and dissolved in 0.85% NaCI to a final concentration of 1 mg/ml. This fraction was examined in mice for myotoxic activity and by immunodiffusion for cross-reactivity with rabbit polyclonal antisera raised against myotoxin a from C. v. viridis (OwNBY et al., 1979) . For biological assays, crude myotoxin (Fraction 4 from Sephacryl S-200 column) was injected in a dose of 2 Mg/g. Two female mice (CD-1, Charles River) weighing 25 g were injected with 0.1 ml i.m. into the dorsolateral aspect of the thigh; two mice were injected with 0.85% NaCI as a control . The mice were killed by cervical dislocation 24 hr after the injection, and a sample of muscle was removed from the ventromedial aspect of the injected thigh. After processing for microscopy as previously described (OWNBY et al., 1976), tissues were examined by light microscopy. Fraction 4 from 5-200 induced vacuolation in skeletal muscle cells identical to that caused by both myotoxin a (OWNBY et al., 1976) and crotamine (CAMERON and Tu, 1978). Crude myotoxin was examined by Ouchterlony double immunodiffusion (OWNBY et al., 1979). Microdiffusion plates were made with a 1 % agarose solution in 0.85% NaCl. Each well was filled twice with 15 IAl of solution (1 mg/ml), incubated at room temperature for 24 hr and then photographed . Myotoxin a and S-200 Fraction 4 show a reaction of complete identity when reacted against antiserum to myotoxin a (Fig . la). Crude myotoxin was subfractionated by cation exchange FPLC (Pharmacia) using a Mono S column. The initial buffer was 50 mM sodium acetate (pH 4.0), and the final buffer was the same plus 2 MNaCl . A small peak washed off the column in initial buffer prior to the start of the gradient (Fig. 2A). A sharp spike of non-myotoxin material eluted
Short Communications
32 1
2 . SÜBFRACTIONATION OF CRUDE C. v. concolor MYOTOXIN RECOVERED FROM GEL FILTRATION ON A 0.5 x 5 CM COLUMN OF MONO S. (A) This material was used for immunodiffusion plates and necrosis assays . Initial buffer 50 mM sodium acetate (pH 4 .0) ; final buffer, 50 mM sodium acetate (pH 4.0) plus 2 M NaCl . Flow rate was 1 .5 ml/min . (B) Mono S chromatogram of 5-200 Fraction 4 from a later batch of venom obtained from Miami Serpentarium used for LD determinations . This material was subfractionated in 20 mM acetate rather than 50 mM . Individual peaks from both batches of venom show evidence of heterogeneity .
FIG .
from the column immediately after the gradient began. Four myotoxin subfractions, which presumably differ in net charge, constituted virtually all of the A2in-absorbing material in S-200 Fraction 4 and eluted in 0.8 -1 .0 M NaCl (Fig . 2). The latter three show evidence of heterogeneity. The four subfractions were pooled separately, dialyzed exhaustively against deionized water and lyophilized. After lyophilization, stock solutions of each subfraction were prepared in physiological saline (1 .0 mg/ml) and examined by double immunodiffusion. All four subfractions reacted with complete identity to myotoxin a using rabbit antisera to myotoxin a (Fig. lb), indicating common epitopes . All four caused contracture of the hind legs in mice within five min after i.m. injection. Doses for the i.m. injections of the four subfractions were 341, 587, 723, 368 Ftg, respectively (13.8, 23 .5, 28 .9, and 14.6 ug/g). Histological assay showed that all four subfractions induced vacuolation of skeletal muscle cells identical to that caused by myotoxin a and the crude myotoxin fraction from C. v. concolor venom. Results obtained upon i.m . injections of the four subfractions suggested the possibility of lethality differences between different isomers of myotoxins. To examin e this possibility, additional material was purified as described above from crude C. v. concolor venom obtained from the Miami Serpentarium (lot no. CK17SZ). This material showed only two subfractions when the 5-200 Fraction 4 was subfractionated over Mono S in 20 mM acetate buffer (Fig. 2B). These both appeared to be heterogeneous, as had three of the four peaks shown in Fig. 2A. While approximately twice as much material was applied to the Mono S column in the second preparation (410 mg) (Fig . 2B) as in the first (Fig . 2A), a separate run of the second preparation, using approximately 3 mg (not shown),
32 2
Short Communications
failed to resolve any additional components . Thus chromatographic differences between the two preparations cannot be attributed to overloading of the Mono S Column. The two subfractions from this second batch of venom were dialyzed and lyophilized as described above. Intravenous injections were made according to the procedures outlined by AIRD and KAISER (1985a) using female mice of the same genetic strain described therein. No attempt was made to determine whether male and female mice differ in their susceptibilities to the toxin. Mice ranging from 21 to 32 g were given weight-specific doses and two to four mice were employed at each dose. The more acidic isomers (subfraction 1, Fig. 2B) proved less toxic than the more basic isomers (subfraction 2). For subfraction 1, doses x1 .25 pg/g proved fatal to all test animals . Doses < 1 .00 Mg/g were non-lethal to all test animals. By contrast, with subfraction 2, doses >0.75 pg/g were lethal to all specimens, while doses <0.625 pg/g were sublethal. This indicates that LD50S for the two subfractions differ by approximately 2-fold, and that survivorship curves for the two groups of isomers are quite steep. The LD50 was not determined more closely because the lethal and non-lethal doses reported above were so close that it would have been difficult to justify the sacrifice of additional animals to make the determination. The LD50 values reported above for subfractions 1 and 2 fall respectively within and below the ranges reported for E toxin (Crotalus h. horridus) in the presence of acetate (ALLEN et al., 1986). It may be argued that because of chromatography using acetate buffers, the basic amino acid side chains in these myotoxins have acetate bound as a counter-ion; hence their toxicity . Since the eluting cation exchange buffer contained only 20 mM acetate and 2 M NaCl, it is likely that CI - , rather than acetate, constituted the bulk of the counter-ion. Ownby (unpublished observations) has reported similar values for myotoxin a where no acetate was used in any purification step. We find the high LD50 value reported by ALLEN et al. (1986) for E toxin in the absence of exogenous acetate (6.3 pg/g) difficult to explain. It may possibly be attributed to the one-step purification procedure they employed. Trace contamination with proteases might account for the gradual loss of toxicity experienced previously with this toxin (SULLIVAN and GEREN, 1979). Several noteworthy observations were made during toxicity assays . Using a dose of 4.0 pg/g, which is close to the LD50 reported for crotamine by CHEYMOL et al . (1971), mice exhibited flaccid paralysis before the injection could be completed. Death ensued in less than 30 sec. Lethal doses closer to the LD 50 caused death within 2 min; however, ifa mouse survived as much as 5 min, recovery seemed ensured. This is in strong contrast to crotalid presynaptic neurotoxins, which continue to kill test animals as much as 18 - 24 hr postinjection, despite the fact that their toxicity exceeds that of the myotoxins by more than an order of magnitude (AiRD and KAIsER, 1985b). Our observations on mice that received a lethal dose suggest that death may result from respiratory failure, perhaps caused by paralysis of the diaphragm and skeletal musculature controlling movement of the rib cage . Such a conclusion would appear to be supported by the effects of myotoxin a on the resting membrane potential of mouse and rat diaphragms (HONG and CHANG, 1985). Myotoxins from C. v. concolor do not appear to induce death by cardiac failure. When sublethal doses of myotoxin a from C. v. viridis were injected into mice i.v., paraffin sections of cardiac muscle examined by light microscopy failed to reveal any evidence of myonecrosis (Ownby, unpublished observations). We noted that in mice. that received a lethal dose, the heart beat strongly throughout convulsions and continued for several minutes after the cessation of all attempts to breathe.
Short Communications
323
While muscle necrosis is the most significant medical problem associated with the small myotoxins, our observations on predatory behavior of C. v. concolor, C. v. viridis and C. v. helleri indicate that the biological significance of these molecules is their ability to halt the flight of a prey organism almost instantaneously . In many cases, a mouse struck by one of these snakes is unable to take a single step, and the hyperextension of the hind limbs that characterizes injections of purified myotoxins is readily observable . Acknowkdganents - This work was supported in part by the U .S . Army Medical Acquisition Activity, Contract no . DAMD 17-86-C-6061 . REFERENCES AiRD, S . D. (1985) A quantitative assessment of variation in venom constituents within and between three nominal rattlesnake subspecies . Toxkon 23, 1000. AIRD, S . D. and KAISER, I . I . (1985a) Toxicity assays . Taxicon 23, 11 . ACRD, S . D . and KAIgHR, I . I . (1985b) Comparative studies on three rattlesnake toxins. Taxleon 23, 361 . ALLEN, H . R., TucxER, R. K. and OEREN, C . R . (1986) Potentiation of the toxicity of basic peptides from rattlesnake venom& by sodium acetate . Taxkon 24, 553 . BiEwm, A . L ., McFARLAND, R . H . and BacKER, R . R . (1986) Complete sequences for myotoxins from Crotalus vWis concolor venom . Fedn Proc. Fedn Am. Socs exp. Biol. 45, 1794 . CAMERoN, D . L . and Tu, A . T . (1977) Characterization of myotoxin a from the venom of prairie rattlesnake (Crotahm vb1db viridis) . Biochemboy 16, 2546. CAmRON, D . and Tu, A . T . (1978) Chemical and functional homology of myotoxin a from prairie rattlesnake venom and crotamine from South American rattlesnake venom. Biochim. biophys. Acts Sit, 147 . CHANG, C . C . and Tsm4o, K . H . (1978) Effect of crotamine, a toxin of South American rattlesnake venom, on the sodium channel of murine skeletal muscle . Br. J. Pharmac. 63, 551 . CFmYMoL, J ., OoNruvEs, J. M ., BouRtLLer, F . and Rock-ARYEILLER, M . (1971) Action neuromusculaire eomparee de la crotamine et du venin de Crotahn derbsus terrjfkus var. crotaminkus - I . Toxkon 9, 279 . ENGLE, C . M ., BEcxER, R. R., BAILEY, T. and BmaEa, A . L. (1983) Characterization of two myotoxic proteins from venom of Crotaha vhidis concolor. J. Toxic. - Toxin Rev . 2, 267 . Fox, J . W ., ELznvoA, M . and Tu, A. T . (1979) Amino add sequence and disulfide bond assignment of myotoxin a isolated from the venom of prairie rattlesnake (Crotahis viridis virWb). Biochemistry 18, 678 . OoNÇALvEs, J . M . (1956) Purification and properties of crotamine. In: Venoms, p . 261 (BucKLEY E . E . and PoRaEs, N ., Eds) . American Association for the Advancement of Science, Washington DC . HONG, S . J. and CHANG, C . C . (1985) Electrophysiologicalstudies of myotoxin a, isolated from prairie rattlesnake (Crotahn virfdis viridis) venom, on murine skeletal muscles. Toxicon 23, 927 . MAEDA, N., TAMIYA, N ., PATrABwRAMAN, T . R . and RussELL, F . E. (1978) Some chemiéal properties of the venom of the rattlesnake, Crotales viridis viridis. Toxicon 16, 431 . MEBs, D . and KoRNAux, F. (1984) Intraspecific variation in content of a basic toxin in eastern diamondback rattlesnake (Crotaha adamanteus) venom . Toxkon 22, 831 . MEBs, D ., EHREaFELD, M . and SAmanmA, Y . (1983) Local necrotizing effect of snake venoms on skin and muscle: relationship to serum creatine kinase . Toxicon 21, 393 . OwNBY, C . L., CAMERON, D . and Tu, A . T. (1976) Isolation of myotoxic components from rattlesnake (Crotahrs viral& vhidis) venom. Electron microscopic analysis of muscle damage . Am. J. Path . 85, 149 . OwNBY, C. L ., WooDs, W . M. and ODELL, O . V . (1979) Antiserum to myotoxin from prairie rattlesnake (Crotahn viridis viridis) venom . Toxicon 17, 373 . SULuvAN, J . and OEREN, C . R . (1979) Isolation, stabilization, and characterization of a toxin from timber rattlesnake venom. Preparative Biochan. 9, 321 . Tu, A . T. and MoatTA, M . (1983) Attachment of rattlesnake venom myotoxin a to sarcoplasmic reticulum: peroxidase conjugated method . Br. J. Path. 64, 633 . VoLPE, P ., DAUUAm, E ., MAURER, A . and Tu, A. T . (1986) Interaction of myotoxin a with the Ca"-ATPase of skeletal muscle sarcoplasmic reticulum. Archs Blochem . Blophys. 246, 90.