Chemical modification of equinatoxin II, a lethal and cytolytic toxin from the sea anemone Actinia equina L

0041-0101/89 53 .00+ .00 ® 1989 Persamon Prat plc

Toxiran VoL 27, No . 3, PP . 37384. 1989. Printed in Great Britain.

CHEMICAL MODIFICATION OF EQUINATOXIN II, A LETHAL AND CYTOLYTIC TOXIN FROM THE SEA ANEMONE ACTINIA EQ UINA L. TOM TURK, 1 PETER MAL~EK~ and FRANC GUBEN~EK 2 'Department of Biology, Biotechnical Faculty and Institute of Biology, Eduard Kardelj University of Ljubljana, A~kerLeva l2, 61000 Ljubljana, Yugoslavia =Jozef Stefan Institute, Department of Biochemistry, Jamova 39, 61000 Ljubljana, Yugoslavia (Acceptedfor publication

26

October

1988)

T. TURK, P. MALEx and F. GuBEN~EK . Chemical modification of equinatoxin II, a lethal and cytolytic toxin from the sea anemone Actinia equina L. Toxicon 27, 37384, 1989 .-The role of arginine and tyrosine in cytolytic properties of equinatoxin II, isolated from the sea anemone Actinia equina L., was studied by means of chemical modifications. The toxin was modified with 2,3 butanedione and tetranitromethane, respectively . The extent of modification and physico-chemical properties of the modified proteins were checked with amino acid analysis, iscelectric focusing and circular dichroic spectra. Extensive treatment of the toxin with 2,3 butanedione modified seven arginines and also two tyrosines, with resulting loss of hemolytic activity . Modification of two out of nine arginine residues resulted in a 25% loss of hemolytic activity, whereas nitration of three out of ten tyrosines decreased hemolytic activity by 95%. The nitrated toxin had at least a 30-fold higher i .v . LDSp than the native toxin. None of the modifications significantly affected the secondary structure of the toxin as revealed by the CD spectra. It is concluded that tyrosine residues are involved in both lethal and cytolytic activity, while the role of arginine residues is not evident because of the non-specific alteration of tyrosine residues with 2,3 butanedione. INTRODUCTION

EQUINATOxIN II (i .e . equinatoxin according to FERLAN and LEBEZ, 1974), isolated from the sea anemone Actinia equina L., is a lethal protein with a mol.wt of 19,000 and a pI of 10 .5 (MALEK and LEBEZ, 1988). It is strongly hemolytic (MAL~EK and LEBEZ, 1981) and cytotoxic for different mammalian cells (GIItALDI et al., 1976; BATISTA et al., 1986, 1987). The hemolytic and lethal activities of equinatoxin II (EqT II) are inhibited by sphingomyelin and serum lipoproteins (TURK and MALEK, 1986; Strazi~ar, MaZek, Sedmak and Vozelj, unpublished observations). In some respects, EqT II is closely related to the cytolysin from the sea anemone Stoichactis (Stichodactyla) helianthus (DEVLIN, 1974 ; BERNr->~tMER and Av1GnD, 1976; BLUMENTxAL and KEM, 1983) and also to other cytolysins isolated from various sea anemone species (BERNFIi;TMER and AVIGAD, 1981 ; MnLEK et al., 1982 ; MEBS et al., 1983 ; BEAM->EIMER et al., 1984; Si-uoMi et al., 1985, 1986). The i.v . LD P Of equinatoxin II is 35 hglkg in mice . 375

376

T. TURK et al .

For data on structure-function relationships of sea anemone cytolytic toxins are available. Hemolytic activity of caritoxin II (MnLEK et al., 1982) was completely abolished by incorporation of one iodine atom per toxin molecule (SErrLI~ and MALEK, unpublished observations) . Reductive methylation of a cytolytic toxin Sh III from the sea anemone Stoichactis helitmthus (BLUMENTHAL and KE11I, 1983) did not affect binding of the toxin to erythrocytes (Kem and Doyle, unpublished observations) . The detailed mechanism of the EqT II interaction with the cell membrane is not well understood, hence the aim of the present study was to obtain insight into the role of arginyl and tyrosyl residues in hemolysis, lethality and precipitation with serum lipoproteins . MATERIALS AND METHODS Equinatoxin II was isolated in pure form according to the method of MwL`~erc and LÉSEZ (1988) . All chemicals used in experiments were of analytical grade . Modification ojEgT II with 2,3 butanedione Arginyl residues were modified as described by Rroxnwx (1973) . To 1 mg of EqT II dissolved in 1 ml of 0.05 M borate buffer, pH 8 .1, a 250-, 500 . or 1000 .fo1d molar excess of 2,3 butanedione (Fluka, Buchs, Switzerland) was added . The reaction mixtures were incubated for 60 min at 25°C. Periodically 10 pl aliquots were removed from We reaction mixtures and tested for residual hemolytic activity with a turbidimetric method (Mw~EK and LÉSEZ, 1981) . Excess of reagent was removed from the modified protein on a Sephadex G-l5 column, equilibrated and eluted with 0 .05 M borate buffer, pH 8 .1 . Toxin modified with a 1000 .fo1d molar excess of 2,3 butanedione was regenerated by applying the toxin to a Sephadex G-15 column equilibrated and eluted with 0.05 M Tris-HCl buffer, pH 8 .1 . The hemolytic activity of the regenerated toxin was assayed at different time intervals as described for modified toxins. Modification of EqT II with tetranitromethane Tyrosyl residues were modified according to the method reported by RIORDwN et al. (1967). To 1 mg of EqT II dissolved in 0 .05 M Tris-HCI buffer, pH 8 .1, 0 .28 M tetranitromethane (Sews, Heidelberg, F .R.G .) dissolved in absolute ethanol was added to achieve a 10 . or 100 .fo1d molar excess over the toxin . The reaction mixture was incubated for 20 ntin at 25°C. The modified toxin was immediately separated from the excess of reagent and side products of the reaction on a Sephadex G-25 column equilibrated and eluted with 0.05 M Tris-HCl buffer, pH 8 .1 . Residual hemolytic activity was estimated as described previously . Amino acid analysis Amino acid analyses of native, modified and regenerated toxins were performed on a l 18 CL amino acid analyser (Beckman, Palo Alto, U .S .A .) . Samples were hydrolyzed in 6 M HCI, containing I % SnCI for 24 hr at 108'C . Isoelectric focusing and sodium dodecyl sulphate-polyacrylamide electrophoresis Iscelectric focusing of native, modified and regenerated toxins was carried out on polyacrylamide gel plates with a pH gradient ranging from 3 to l0, and stabilized with Pharmalyte (Pharmacia, Uppsala, Sweden). Protein standards were purchased from Pharmacia (Uppsala, Sweden) and Serva (Heidelberg, F .R.G .) . The mol.wts of native and modified toxins were checked with SDS-PAGE electrophoresis according to the method of WESEa and Oseoxx (1969) . Circular dichroism CD spectra of native and modified toxins were obtained at 25°C with a spectropolarimeter Dichrograph Mark III (Jobin Yvon, France) . Protein concentration ranged from 0.19 to 0.82 mg/ml in a 1 cm pathlength cell and from 0.19 to 0 .41 mg/ml in a cell of 0.05 cm pathlength. Toxin modified with 2,3 butanedione was dissolved in 0.05 M borate buffer, pH 8 .1, while the tetranitromethane modified toxin was dissolved in 0.001 M acetic acid . Native toxin was dissolved in the same solvents . The CD spectra are presented as mean residue ellipticity B (degree cm'/decimole) at each wavelength using the following relationship : BMrtw -

3300 . d . s . MRW .c '

Chemical Modification of Equinatoxin II

37 7

where d=intensity of the spectrum (mm), s=sensitivity (absorb. units/mm), MRW=mean residual weight taken to be 114(8), l=pathlength (em) and c=concentration (g/liter). Interaction ojnative and mod~ed FyT II with serum lipoproteins

Interaction of native and modified toxins with serum lipoproteins was tested as described by Strati~ar, Maôek, Sedmak and Vozelj (unpublished observations). Non-immune human sera were electrophoretically separated on 1% agarose micro-slides (2 .5 x 7.5 cm) in diethylbarbital buffer, pH 8.2, for 3 hr (17 mA, 200V). Serum components were subaequcntly exposed to 80 pg of native or modified toxin dissolved in 100 ~1 of borate or Tris-HCl buffer, pH 8.1 ; respectively. After 24 hr the slides were washed with 0.9% NaCI and left overnight at room temperature and thereafter washing was repeated. The agarose slides were stained with amido black 10 B (Merck, Darmstadt, F.R .G.) or Sudan black B (Serve, Heidelberg, F.R .G.). Lethality assays

Lethality of native and modified toxins was estimated following i.v . injection in male NIH mice . One mouse ~n,° was calculated considering the weight of an individual animal and an estimated i.v . t nm of 35 l+8lkg in mice . The amount of toxin injected into experimental animals was expressed as the number of mouse t.n,° which ranged from 1 mouse t o for the native toxin to 30 and 100 r.~m for modified toxins . For each amount of native and modified toxin two experimental animals were used . RESULTS

Modification of EqT II with a 250-, 500- or 1000-fold molar excess of 2,3 butanedione for 60 min resulted in 85%, 90% and 95% loss of native hemolytic activity, respectively (Fig. lA). When a 1000-fold molar excess of the reagent was used seven out of nine arginyl residues were modified . Besides arginyl residues two out of ten tyrosyl residues were also affected, as revealed by amino acid analysis (Table 1). After the toxin modified with 2,3 butanedione was regenerated by buffer exchange, 75% of hemolytic activity was restored . Tyrosyl residues were completely regenerated while two arginyl residues remained modified (Fig. 2, Table 1).

Fyo. 1. MODIFICATION OF ~utxe~roxnv II (EqT II) waft 2,3 svrwNmtorn:wxo x~rxwxrraos~rre A. Modification of EqT II was performed in 0.05 M borate buffer, pH 8.1, with a 250 (Q), 500 (p) and 1000 (O}fold molar excess of 2,3 butanedione over the toxin. Native EqT Ii= " . B. Modification of EqT II in 0.05 M Tris-HCl buffer, pH 8.1, with a 10 (O)-fold molar excess of tetranitromethane over the toxin. Native EqT II= " .

378

T. TURK et al. TABLE I . NUMBER OF AM1N0 ACII) RESIDUES 1N NAT1VE-MODIFIED AND REGENERATED EQUINATOXIN II

ASX THR SER GLX PRO GLY ALA CYS VAL MET ILE LEU TYR PHE HIS LYS TRP ARG R.H .A .

Native EqT II

BD modified EqT II

Regenerated EqT II

TNM modified EqT II

19.9 (20) 7.6 (8) 13 .5 (14) 6 .6 (7) 4 .5 (5) 17 .6 (18) 13 .6 (14) 0 (0) 11 .3 (1 l) 2 .6 (3) 4 .l (4) I5 .8 (l6) 10 .3 (10) S .l (5) 4 .4 (4) 10 .1 (l0) n .d . 9.4 (9)

20 .4 (20) 8 .5 (8) 14 .3 (14) 6 .6 (7) 4 .5 (5) 17 .6 (18) 14 .1 (14) 0 (0) 11 .5 (I1) 2.6 (3) 4.0 (4) 16.0 (16) 8 .3 (8) 5 .4 (5) 4.2 (4) 10 .3 (10) n .d. 1 .9 (2)

20 .5 (20) 7 .6 (8) 14.0 (14) 6.5 (7) 4.5 (5) 18 .2 (18) 13 .5 (l4) 0 (0) 10.6 (11) 2 .7 (3) 3 .9 (4) 15 .5 (16) 10 .4 (!0) 5 .4 (5) 4 .3 (4) 10 .4 (10) n .d . 7 .0 (7)

20.3 (20) 7 .5 (8) 14.0 (14) 6 .5 (7) 4 .6 (5) 17 .7 (I S) 13 .6 (14) 0 (0) 11 .3 (11) 2 .6 (3) 3 .8 (4) 15 .6 (l6) 6 .9 (7) 4 .8 (5) 4.2 (4) 9.8 (10) n.d . 9.5 (9)

100%

10%

25%

5%

The number of each amino acid residue is an average number calculated from three amino acid analysis . n .d . = not determined; BD = 2,3 butanedione ; TNM = tetranitromethane ; R .H .A . = residual hemolytic activity .

3A

75

50

25

30

80

30

80

90

t(mln)

70h

FIG . 2. REGENERATION OF EQUINATOXIN II FiENMOLYTIC ACTIVITY . Regeneration of EqT II hemolytic activity modified with a 1000-fold molar excess of 2,3 butanedione was achieved following the exchange of borate with Tris-HCl buffer on a Sephadex G-15 column . (Native EqT II S, modified EqT II p, regenerated EqT II ~.) Arrows indicate the application of native and modified EqT II to the Sephadex G-IS column equilibrated with TrisHCl buffer.

Chemical Modification of Equinatoxin II

37 9

Three tyrosyl residues of ten present in the native toxin molecule were modified with a 10-fold molar excess of tetranitromethane after 20 min of incubation . The loss of hemolytic activity was 95% (Fig. 1 B, Table 1). Isoelectric focusing showed that native EqT II with a pI of 10.5 migrated to the edge of the plate because of its high positive charge. The pI of the toxin modified with 2,3 butanedione was shifted to the lower pH range. There were at least four bands covering a pH range of 7.0-8.65. The band of regenerated toxin shifted back to the higher pH range but there was still some heterogeneity characteristic for the modified toxin. Tetranitromethane modified toxin migrated as a single band to a pI value of 9.45 (Fig. 3). Only the toxin modified with a 100-fold molar excess of tetranitromethane showed the appearance of several oligomers, mostly dimers, as revealed by SDS-PAGE (data not shown) . Far ultraviolet CD spectra of modified toxins were not substantially affected, indicating that secondary structure remained practically unchanged. In the case of tyrosine modified toxin (Fig. 4A) a substantial decrease of the tyrosine band was observed. The modification of arginines also resulted in a slight decrease of the tyrosine dichroic absorption band which is in accordance with amino acid analysis where two tyrosines seem to be modified (Fig. 4B). The near u.v. CD spectra of native toxins (Fig. 4A, B) differ slightly due to the

pH g,45 8,45 7, 35

5,2

3,5

ST

1

2

3

4

5

ST

3 . L9OELECTRIC FOCUSING OF NATIVE, A~ODIFIED AND REGENERATED EQUINATbXIN II . Isoelectric focusing of native EqT II (l,2), Modified EqT II with a 1000-fold molar excess of 2,3 butanedione (3), regenerated EqT II (4) and EqT II modified with a 10 .fold molar excess of tetranitromethane (5) on polyacrylamide gel plates with a pH gradient ranging from pH 3 to pH 10 . ST represent standards . FIG.

380

T. TURK e! al.

difference in pH of the solutions. Toxins modified with 2,3 butanedione and tetranitromethane significantly differed in their ability to precipitate human serum beta-lipoproteins. Toxin modified with 2,3 butanedione as well as native toxin interacted with lipoproteins while tetranitromethane modified toxin did not cause any precipitation (Fig. 5). Modification of EqT II with both reagents affected its lethal properties . t,Ds p were increased and the survival time of experimental animals was prolonged (Table 2). In mice injected with the modified toxins general weakness, tremor and ptosis were observed . Mice that survived exhibited none of the above described symptoms . Post mortem examination of intoxicated mice revealed no major macroscopically evident changes. DISCUSSION

Equinatoxin II was subjected to chemical modification with 2,3 butanedione and tetranitromethane to elucidate the possible role of arginyl and tyrosyl residues in its hemolytic and lethal properties . Arginyl residues were chosen for modification because of their contribution to the positive charge of the protein molecule, while tyrosines have been suspected as probable candidates for interaction with membrane lipids . Furthermore, nitration of tyrosyl residues was necessary in order to discriminate the particular role of arginine and tyrosine after nonspecific alteration of tyrosine with 2,3 butanedione. The reaction of this reagent with arginyl residues of proteins is quite specific if the reaction is

HAIRW 1d~yr~~ an4 /dmol~) 250 //

x ~nn+i

zss

2~b

/ W W\ \\ // 265

27b

125

\_

/

zia

2R5

i~~

zab 2bb

-2b00

275

2liS

1

1

-b000 1 -7600

t ./

-10000

FIG. 4 . CIRCULAR DICHROIC SPECTRA OF NATIVE AND MODIFIED EQUINATOXIN II .

A. CD spectra of native (solid line) and tetranitromethane modified EqT II (dashed line) were obtained in 0.001 M acetic acid . B. CD spectra of native (solid line) and 2,3 butanedione modified EqT II (dashed line) were obtained in 0.05 M borate buffer, pH 8. l. Experimental details are described in the Materials and Methods section.

B

Chemical Modification of F.quinatoxin II

38 1

,z .?

T'

FIG . S . PRECIPITATION PROFERTIES OF NATIVE AND MODIFIED EQUINATOXIN II WITH NON-IMMUNE HUMAN SERUM BEl'A WPOPROTEIN3.

Precipitation of native F.qT II (1), 2,3 butanedione modified F.qT II (2) and EqT li modified with tetranitromethane (3) with non-immune human sera beta-lipoproteins. Sera were applied into the holes and electrophoretically separated, while toxins were applied into the slide channels. Micro slides marked with an asterisk were stained with Sudan black B, while those without asterisks were stained with amido black B.

performed in borate buffer (RIORDAN, 1973), however, we found that two tyrosyl residues in the EqT II molecule were nevertheless affected . The decreased amount of tyrosine after modification of proteins with 2,3 butanedione could be due to photooxidation as reported by Friss et al. (1979) . In our case, the resulting modification is more likely of a different character because all tyrosines were recovered when, prior to hydrolysis, the toxin had been subjected to Tris-HCl buffer. On the other hand, nitration with tetranitromethane is highly

382

T . TURK et at. TABLE 2 . LETHALITY OF NATIVE AND MODIFIED EQUINATOXIN II

Toxin

No . of mouse l .V.

Native EqT II BD modified EqT II BD modified EqT II TNM modified EqT II TNM modified EqT II 20

'One mouse i .v . g mouse.

LD s p

LD~~

1

30

100 30 100

Survival time min hr hr Survived 18 hr 1 .30 48 36

is equivalent to 0.7 hg of equinatoxin II in a

specific for tyrosine (RIORDAN and VALLEE, 1972) although under certain circumstances cysteine could also be modified (Soxor. .ovsxY et al ., 1969). In the EqT II molecule, which lacks cysteine, three tyrosyl residues could be nitrated without gross alteration of protein structure, as shown particularly with the far u.v . CD spectrum . More extensive nitration with larger excess of tetranitromethane resulted in cross-linking and insolubility of products, as has also been observed for some other extensively nitrated proteins (RIORUAN and VALLEE, 1972; MEtEIt et al ., 1979). The interpretation of the structure-function relationships needs a reliable control of protein conformation which could be affected by chemical modifications with groupspecific reagents . EqT II, treated with either reagent, showed no substantial changes in solubility, chromatographic properties or CD spectra, indicating that the secondary structure of the toxin remained essentially unchanged. Therefore, alteration of biological activity could be attributed to chemical modifications of arginyl and tyrosyl residues . Considering the accessibility for either reagent, it can be stated, that at least seven arginine and three tyrosine residues are located on the surface of the toxin molecule . Modification is also reflected in the diminution of positive charge, especially after 2,3 butanedione and in part tetranitromethane treatment of EqT II . The lower pI of the nitrated toxin is very likely due to the lower pK values of the resulting nitrophenoxide groups (RIORDAN et al., 1967a, b) . Sphingomyelin and probably some other phospholipids have been recognized as acceptor molecules for sea anemone cytolytic toxins (BExNt->En~R and AvIGAD, 1976, 1981, 1982 ; UNDER et al ., 1977 ; SFUty et al., 1979 ; BERNt-~n~t et al., 1982, 1984; MA~Ex et al., 1982 ; Ivnxov et al., 1987). Hemolytic activity of EqT II is also inhibited by Sphingomyelin (TURK and MALEK, 1986). The nature of the interaction between EqT II and Sphingomyelin was independent of ionic strength of even 2 M KCI, while the chaotropic agent KSCN was able to dissociate a Sphingomyelin-toxin complex (P . Maôek, Ph .D . Thesis, University of Edvard Kardelj, Ljubljana, 1983), thus indicating the importance of hydrophobic interactions . Therefore, the loss of activity of EqT II modified with tetranitromethane could be due to decreased hydrophobicity of nitrated tyrosyl residues . On the other hand, arginine has been found to be important for the biological activity of another class of sea anemone toxins, peptide neurotoxins, affecting sodium channel inactivation (BÀRHANIN et al., 1981 ; KOZLOVSKAYA et al., 1982) . Evidence for involvement of tyrosine in biological activity is also provided by the lethality assay of modified EqT II . It is noteworthy that lethal doses of both 2,3 butanedione and tetranitromethane-modified toxins induced in some respects different modes of pathophysiological activity than the unmodified toxin. With a lethal dose of native toxin all symptoms appeared within a short time and mice died after a few min, or they survived without any visible signs of

Chemical Modification of Equinatoxin II

38 3

intoxication when a sublethal dose was applied, as had also been observed previously (Mn~Elc and LEBEZ, 1988). Modified toxins exhibited long term actions which suggest that pathophysiological changes in the organism may be different than those induced with native toxin. Pharmacological studies in mice (LEE et al., 1988) suggested that acute death produced by EqT (i.e. EqT II) could be attributed to coronary vasospasm and hyperkalemia with or without cardiotoxicity (see also SKET et al., 1974; Ho et al., 1987). In addition to these findings, which imply primarily general lytic effects, a direct effect of EqT II on muscle and nerve ion channels was reported (~UPUT, 1986; SUPUT et al., 1988). Whatever the correlation between lethal activity and cytolytic activity, the conclusions are that tyrosine is essential for both activities of EqT II, while the loss of activity following modification of arginine residues with 2,3 butanedione is associated with nonspecific alteration of tyrosyl residues. REFERENCES Bxxxwwtrt, J ., HUGUFS, M ., $CHWII7Z, H ., Vnvc~:rrr, J . P. and Leznuxsra, M . (1981) Structure-function relationships of sea anemone toxin II from Anemorria sulcata . J. biol. Chem. 256, 764. BA77SfA, U ., Me~c, P . and Stast~tc, B . (1986) The influence of equinatoxin II on V-79-379A cell line. Period. Biol. 88, 97 . BA173IA, U ., JEZExtviK, K., McL~c, P. and Smswc, B . (1987) Morphological evidence of cytotoxic and cytolytic activity of equinatoxin II. Period. Biol. 89, 347 . Bexrn~a¢x, A . W. and Av~oen, L . S. (1976) Properties of a toxin from the sea anemone Stoichactis helianthus, including specific binding to aphingomyelin. Proc. natn . Acad. Sci. U.S.A . 73, 467 . BeRxt~x, A. W . and Avtoeu, L . S. (1981) New cytolysins in the sea anemones from the east coast of the United States. Toxicon 19, 529 . BeRivt~a~rt, A . W. and Avtoen, L . S . (1982) Toxins of the sea anemone Epiactis prolifera . Archs Biochem . Biophys . 217, 174 . BeRrtt~t~x, A . W ., AVIO~n, L. S . and Lti, C . Y. (1982) Purification and properties of a toxin from the sea anemone Condylactis gigantea. Archs Biochem . Biophys . 214, 840. BERNF~IMFR, A . W., Av~a~n, L . S ., Bw~xcrt, G., Dowot.e, E . and Lei, C . Y . (1984) Purification and properties of a toxin from the South African sea anemone, Pseudactinia varia . Toxicon 22, 183 . BLLfMENTïrAL, K . M . and Kr~, W. R . (1983) Primary structure of Stoichactis kelianthns cytolysin III . J. biol. Chem . 258, 55. DEVt .nv, J . P. (1974) Isolation and partial purification of hemolytic toxin from the sea anemone Stoichactis helianthL.s . J. pharmac. Sci . 63, 1487 . Fexux, I. and Leaez, D. (1974) Equinatoxin, a lethal protein from Actinic equina-I . Purification and characterisation . Toxicon 12, 57 . FLrss, H . and Vrsw~rr~~rru, T. (1979) 2,3 butanedione as a photosensitizing agent : application to alpha-amino acids and alpha~hymotrypsin . Can . J. Biochem . 57, 1267 . Guw.n~, T., F~eux, I . and Rouufl:o, D . (1976) Antitumour activity of equinatoxin . Chem. biol. Interact . 13, 199. Ho, C . L., Ko, J. L ., Lue, H . M ., L.ee, C. Y . and Fenurt, I. (1987) Effects of equinatoxin on the guinea-pig atrium . Toxicon 25, 659 . Iverrov, A . S., Mo~x~ue, A . A., KOZIAVSIUYA, E . P . and MoxesrvxNevw, M . M . (1987) The action of toxin from Radianthus macrodactylus on biological and model membrane permeability (in Russian) . Biol. Memb . 4, 243 . KOZLOVSKAYA, E. P ., Vozxove, H . and ELYAKOV, G . (1982) Structure~ctivity studies of neurotoxin I from the sea anemone Radiantkus macrodactylus. In: Chemistry of Peptides and Proteins I, p. 379 (VOEL'rex, M . et al., Eds). Berlin : de Gruyter . Lam, C . Y ., L~rt, W . W ., CttBV, Y . M . and Lee, S. H . (1988) On the cause of acute death produced by animal venoma and toxins . In : Progress in Venom and Toxin Research, p . 3 (Gope~ ~An~ccrnt AICOr1E, P. and Tax, C. K ., Eds). Singapore : Faculty of Medicine, National University of Singapore . Ltrmea, R., BFRNHEIMEa, A . W . and Knt, K . S. (1977) Interaction between sphingomyelin and cytolysin from the sea anemone Stoichactis helianthus . Biochim . biophys . Acta 467, 290. M~L~ex, P . and Leaez, D . (1981) Kinetics of hemolytis induced by equinatoxin, a cytolytic toxin from the sea anemone Actinic egaina . Effects of some ions and pH . Toxicon 19, 233 . MaLerc, P . and LeeFZ, D. (1988) Isolation and characterisation of three lethal and hemolytic toxins from the sea anemone Actinic equine L . Toxicon 26, 441 . M~Lerc, P., Stnvhh~, L. and LEaez, D. (1982) Isolation and partial characterisation of three lethal and hemolytic toxins from the sea anemone Actinic ceri . Toxicon 20, 181 .

38 4

T. TURK et al .

Mees, D ., LtEextctt, M ., REUL, A . and SAMEYtMA, Y . (l9ß3) Hemolysins and proteittase inhibitors from sea anemones of the Gulf of Aqaba . Toxicon 21, 257. MerEte, H ., Vmxo~, H ., HENDRIIC3, F ., SLOT'900bf, A . J . and HESS, H . (1979) Comparative studies of tyrosine modification in pancreatic phospholipases . I . Reaction of tetranitromethane with pig, horse and ox phospholipases A 2 and their zymogenes. Biochemistry 18, 3582. Rtotin~x, J . F. (1973) Functional arginyl residues in carboxypeptidase A . Modification with butanedione . Biochemistry 12, 3915 . RIORDAN, J . F . and VAL1.F~, B . L. (1972) Nitration with tetranitromethane . Meth . Enzymol. 2S, 515 . RIORDAN, J . F., SOKOLOVSRY, M . and V~t.t.eE, B . L. (1967a) Environmentally sensitive tyrosyl residues. Nitration with tetranitromethane . Biochemistry 6, 35ß . Rtotzoex, J . F ., SOKOLOVSKY, M. and VALI,eE, B . L. (1967b) The functional tyrosyl residues of carboxypeptidase A . Nitration with tetranitromethane . Biochemistry 6, 3609. Stmv, L . M ., Mtct~t,ts, D. W . and M~YEtt, M . M . (1979) Membrane damage by a toxin from the sea anemone Stoichactis helianthus. II . Effect of membrane lipid composition in a liposome system . Biochim . binphys. Acta SSS, 79 . Swot, K ., TANAKA, E ., YAMANAKA, H . and Kttcucrtt, T. (1985) Isolation and characterisation of a lethal hemolysin in the sea anemone Parasicyonis actinostoloides . Toxicon 23, 865. Swot, K ., TAKAMIYA, M ., YAMANAKA, H . and Ktxucw, T . (1986) Physiochemical properties of a lethal hemolysin isolated from the sea anemone Anthopleura japonica . Bull. Jpn . Soc. Sci . Firh. 52, 539 . Star, D., Dtt~t,~x, K . and Fateux, I . (1974) Equinatoxin, a lethal protein from Actinic equina . II . Pathophysiological action . Toxicon 12, 63 . SOKOLOVSKY, M ., Hntmt.t., D . and RIORDAN, J . F. (1969) Reaction of tetranitromethane with sulthydryl groups in proteins . Biochemistry 8, 4740 . Sueur, D. (1986) Effects of equinatoxin on the membrane of skeletal muscle fiber . Period. Biol. 88, 210 . ~urur, D ., Ruet,v, N . and MsvES, H . (1988) Effects of equinatoxins on single myelinated nerve fibres . In : Progress in Venom and Toxin Research, p.467 (GOPALAKRISHNAKONE, P . and Tax, C. K ., Eds) . Singapore : Faculty of Medicine, National University of Singapore . TURK, T. and McL`>ac, P . (1986) Effect of different membrane lipids on the hemolytic activity of equinatoxin II from Actinic equina. Period. Biol. 88, 216 . WeeFx, K . and Osnottrr, M . (1969) The reliability of molecular weight determination by dodecyl sulphate polyacrylamide gel electrophoresis . J. biol. Chem . 244, 4406.