0161~5890/92$5.00 + 0.00

Molecular Immunology, Vol. 29, No. l/8. pp. 871-882, 1992 Printed in Great Britain.

0 1992 Pergamon Press Ltd

IMMUNOCHEMICAL ANALYSIS OF A SNAKE VENOM PHOSPHOLIPASE A, NEUROTOXIN, CROTOXIN, WITH MONOCLONAL ANTIBODIES V. CHOUMET,* G. FAURE,* A. ROBBE-VINCENT,* B. SALIOU,* J. C. MAZIB and C. BON$ *Unit& des Venins, Unit& associCe Institut Pasteur/INSERM No. 285, and tHybridolab, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris Cedex 15, France (First received 3 1 October 1991; accepted in revised form 31 December 1991)

Abstract-Crotoxin is the major neurotoxic component of the venom of the South American rattlesnake, Crotalus durissus terr$cus. The crotoxin molecule is composed of two subunits: a basic and weakly toxic phospholipase A, (PLA,) called component-B (CB), and an acidic, nonenzymatic and nontoxic subunit called component-A (CA). Crotoxin exists as a mixture of several isoforms (or variants) resulting from the association of several subunit isoforms. We prepared monoclonal antibodies (MAbs) against each isolated subunit. Six anti-CA MAbs and eight anti-CB MAbs were tested for their cross-reactivities with each subunit and with other toxic and nontoxic PLA,s. Four of the six anti-CA MAbs cross-reacted with CB, whereas only one of the eight anti-CB MAbs cross-reacted with CA. Two anti-CB MAbs were found to cross-react with agkistrodotoxin, a single chain neurotoxic PLAz purified from the venom of Agkistrodon blomhofii brevicaudus. We determined the dissociation constants of each MAb for CA and CB isoforms and their capacities to neutralize the lethality and to inhibit the catalytic activity of crotoxin. We defined three epitopic regions on CA and four on CB, and used a schematic representation of the two subunits to characterize these epitopic regions with respect to: (I) the “toxic” and the “catalytic” sites of CB, and (2) the zone of interaction between the two subunits. We propose three-dimensional structures of the crotoxin subunits in which we localize amino acid residues that might be involved in the epitopic regions described here.

et al., 1978, 1981; Renetseder et al., 1985; Brunie et al., 1985; White et al., 1990; Scott et al., 1990; Wery et al.,

INTRODUCTION Phospholipases A, (PLA,s) are widespread enzymes that catalyze the specific hydrolysis of the 2-acyl ester bond of 3-sn -phosphoglycerides. Secreted extracellular PLA, s are very abundant in mammalian pancreas as well as in snake venoms, and are also present in many other

tissues. These extracellular PLA,s have a mol. wt of approx. 14,000 and their enzymatic activity is activated by millimolar concentrations of Ca2+ (Waite, 1987). They have been grouped in three classes based on their polypeptide sequences (Davidson and Dennis, 1990; Vandermeers et al., 1991): group I contains PLA,s from mammalian pancreas and from Hydrophiidae and Elapidae venoms; group II, PLA,s from Crotalidae and Viperidue venoms and mammalian PLA,s of nonpancreatic origins; and group III, PLA,s from bee and Heloderma venoms. The three dimensional structures of secreted PLA,s from these three groups demonstrate conserved structure of their catalytic center (Dijkstra

SAuthor to whom correspondence should be addressed. Abbreviations: AGTX, agkistrodotoxin;

AMTX-A, ammodytoxin A; BSA, bovine serum albumin; CA, component-A of crotoxin; CB, component-B of crotoxin; ELISA, enzymelinked immunosorbent assay; i.p., intraperitoneal; i.v., intravenous; MAb, monoclonal antibody; PBS, phosphate buffered saline; PBS-Tween, PBS containing 0.1% Tween20; PLA,, phospholipase A*.

1991). At the same time, secreted PLA,s are different from intracellular lysosomal (Bartolf and Franson, 1990) or cytoplasmic (Kim et al., 1991) enzymes which are characterized by a much higher mol. wt (SO,OOO-100,000) and a totally different molecular structure. Certain extracellular PLA,s from snake venom, the /I-neurotoxins, evolved into potent neurotoxins which block neuromuscular transmission by acquiring a high selectivity for specific cellular targets (Hawgood and Bon, 1991). The b-neurotoxins have been classified into different subclasses according to their molecular structure. Some of them, such as ammodytoxin A (AMTX-A) from Vipera ammodytes ammodytes venom (Lee et al., 1984; Ritonja and Gubensek, 1985) or agkistrodotoxin (AGTX) from Agkistrodon blomhojii brevicaudus venom (Hsu et al., 1976; Chen et al., 1987; Kondo et al., 1989), consist of a single polypeptide chain, homologous to nontoxic extracellular PLA,s. Others, such as /3bungarotoxins from Bungarus venoms (Kondo et al., 1978a, b) are composed of two covalently-linked subunits, a PLA, subunit and a polypeptide homologous to Kiinitz-type protease inhibitors from pancreas. Finally, another group comprises the fl-neurotoxins which are characterized by the noncovalent association of several different subunits. In particular, crotoxin from Crotalus durissus terr$cus venom and crotoxin-like toxins from the venoms of Crotalus sp. (Pool and Bieber, 1981; Aird


V. CHOUMET et al


and Kaiser, 1985; Aird et al., 1990; Bieber et al., 1990) have two noncovalently coupled subunits: a basic and weakly toxic PLA,, component-B (CB), and an acidic nontoxic and nonenzymatic subunit, component-A (CA). This latter CA is made of three disulfide-linked polypeptide chains corresponding to three different regions of a PLAz precursor. Crotoxin is in fact a mixture of isotypes deriving from the various combinations of subunits isoforms (Faure and Bon, 1987, 1988). The various isoforms of CB result from the expression of different messenger RNAs (Bouchier et al., 1991). The isoforms of CA derive from a unique precursor, homologous to a PLA?, by different posttranslational events. These modifications are: (1) the proteolytic elimination of three polypeptides, at least one of them belonging to the catalytic site of PLA:, and (2) the possible cyclization of the g-NH2 of the N-terminal glutamine residues liberated by proteolytic cleavage (Bouchier et al., 1991; Faure et al., 1991). The mechanism of action of fi-neurotoxins appears to be different in each subclass. b-Bungarotoxins specifically bind to a voltage-sensitive potassium channel by the intermediate of their non-PLA? subunit (Schmidt and Betz, 1989). In the case of crotoxin, in contrast, the complex dissociates upon interaction with its target: CB binds, while CA, behaving as a chaperon to prevent nonspecific adsorption of CB, is released in solution (Jeng et al., 1978; Bon and Jeng, 1979). Immunological studies with monoclonal antibodies (MAbs) have proved very useful in the investigation of the structure and function of proteins. In this study, we describe the preparation and characterization of MAbs directed against the two crotoxin subunits. We have analyzed their specificity and their capacity to neutralize the enzymatic activity and the lethal potency of crotoxin, as well as their binding to CA and CB epitopes. We propose a map of the antigenic determinants indicating their positions in relation to the zone of interaction between CA and CB and the “toxic” and “catalytic” sites of crotoxin. MATERIALS AND METHODS Materials Crotoxin was isolated from C. durissus terrzjicus venom according to the procedure described by Faure and Bon (1987). Isoforms of each crotoxin subunit were purified as described by Faure er al. (1991). AGTX, purified from the venom of A. blomhofii brevicaudus was a gift of Y. C. Chen (Institute of Biochemistry, Shanghai, China). AMTX-A from V. ammodytes ammodytes venom was kindly provided by F. Gubensek (Josef Stefan Institute, Ljubljana, Slovenia). Taipoxin, textilotoxin and notexin were gifts from A. Coulter (Commonwealth Serum Laboratories, Australia). Bitis caudalis venom and /?-bungarotoxin were purchased from Sigma (St Louis, U.S.A.) and Crotalus atrox venom from Miami Serpentarium (Punta Gorda, FL, U.S.A.). Peroxidase-labeled rabbit anti-mouse immunoglobulins specific for IgM or IgG (IgGl, IgG2a,

IgG2b, IgG3) were obtained from Miles Laboratories (Epernon, France). Peroxidase-labeled and /?-galactosidase-labeled anti-murine immunoglobulins were from Biosys (Compiegne, France). Immunization


and preparation

of MAbs

BALB/c mice were injected S.C. with 25 pg of CA or CB (a mixture of isoforms in both cases) in 100 pl of 150 mM sodium chloride emulsified with 100 ,nl of complete Freund’s adjuvant. Two S.C. booster injections of the same materials, in incomplete Freund’s adjuvant, were given 1 and 3 weeks later. In the case of CA subunit, two additional injections were performed 5 and 7 weeks later in order to reach an appropriate antibody titer. After an interval of 1 month, mice received the last booster of CB (12 pg) or of CA (25 pg) by i.p. injection. Splenocytes were collected 3 days later and fused with the murine cell line X63-Ag8.653, according to the procedure of Kiihler and Milstein (1975). The resulting hybrid cells were selected and multiplied. The supernatant fluids were screened for antibody production by ELISA specific for CA or CB. Cells producing anti-CA or anti-CB MAbs were cloned by limited dilution and i.p. injected into mice, which had previously been primed with pristane (2,6,10,14-tetramethylpentadecane). Ascitic fluids were harvested 15 to 20 days later and centrifuged. Supernatants were collected, supplemented with 0.1% NaN, and stored frozen at -2O’C. Enzyme -linked immunosorben

t assay (ELISA


Microtiter plates (96 wells) were coated overnight at 4-C with 150 ~1 of CA (1 pg/ml) in phosphate buffered saline solution (PBS) or with 150 ~1 of CB (1 pg/ml) in phospho-borate buffer (30 mM sodium tetraboratehydrogenophosphate, pH 8.5). The plates were washed three times with PBS containing 0.1% Tween-20 (PBS-Tween) and unbound sites were saturated for 1 hr at 37’C with 3% bovine serum albumin (BSA) in PBS. The plates were washed (three times) with PBSTween and used immediately for ELISA. In order to measure titers of sera, hybridoma supernatants, ascitic fluids or purified MAbs, 100 p 1 of serial dilutions in PBS containing 3% of BSA were added to the plates and incubated for 1 hr at 37°C. The plates were washed with PBS-Tween and incubated for 1 hr at 37°C with 100 pl of peroxidaseor fi-galactosidase-labeled anti-mouse immunoglobulins, then washed again. Reaction medium (100 ~1) for p-galactosidase assay (100 mM sodium phosphate, pH 7.4, containing 1 mM Mg-Titriplex, 1 mM MnSO,, 1 mM MgSO,, 10 mM S-methylcysteine and 2.65 mM p-nitrophenyl-/I-galactoside) or for peroxidase assay (10 mM sodium phosphate pH 7.3 containing 10 mg/ml of o-phenylenediamine and 2 pi/ml of 30% H,O,) were added and the enzymatic reaction allowed to proceed for 30min at 37°C (for fi-galactosidase) or for 5 min in the dark at room temperature (for peroxidase). The absorbance was read at 405 nm (fl-galactosidase) or 490 nm (peroxidase) with a Dynatech spectrophotometer.

Immunochemical analysis of crotoxin ELISA procedures were also performed with peroxidase- and biotin-labeled MAbs. Serial dilutions of peroxidase-labeled MAbs were added to the plates and incubated for 1 hr at 37°C. After washing with PBS-Tween, substrate for peroxidase assay was added and enzymatic activity was determined. In the case of biotin-labeled MAbs, a further incubation step using streptavidin-peroxidase was performed for 1 hr at 37°C before adding the peroxidase substrate. Purljication and labeling of MAbs Anti-CA and anti-CB MAbs were purified from ascitic fluids by immunoaffinity chromatography on immunoadsorbant columns prepared by coupling CA or CB to aminohexyl Sepharose-4B according to Cuatrecasas and Anfinsen (1971). MAbs were labeled with horseradish peroxidase using the one-step glutaraldehyde method described by Avramkas (1969a). Biotinylation of purified anti-CA and anti-CB MAbs was performed according to Avramkas (19696). Measurement of dissociation constants The dissociation constants (&) of antigen-antibody interactions were determined under equilibrium conditions according to the method described by Friguet et al. (1985). Dilutions of anti-CA or anti-CB MAbs, chosen in the linear part of the ELISA titration curves (linear regression analysis was performed to assess the linearity of the curve), were incubated overnight at 4°C with various concentrations of CA or CB isoforms. The concentration of free MAbs was determined by ELISA: aliquots (100 p 1) of incubation medium were transferred into the wells of microtiter plates coated with CA or CB and allowed to react for 5 min (CA) or 10 min (CB) at 4°C. The plates were washed with PBS-Tween and the subsequent steps of ELISA were performed according to the general procedure. Dissociation constants were deduced from Scatchard plots as described by Friguet et al. (1985). Mapping of CA and CB epitopes The ability of anti-CB MAbs to interact with the same epitope on the CB molecule was determined as described by Friguet et al. (1983) by comparing the ELISA response obtained with a mixture of two MAbs with the sum of individual response of each MAb assayed alone. ELISA was carried out according to the general procedure on microtiter plates coated with CB, using the lowest concentration of each MAb that gave the maximal ELISA response. An additivity index (A) was used to estimate the degree of interaction between the two MAbs. It was defined as follows: A=

2A (I+3 A,-

1 ’

where, A, and A, are the ELISA responses obtained with MAbl and MAb2 tested separately and A,, +2j the ELISA response determined with a mixture of the two MAbs (Friguet et al., 1983). A low A value indicates that the two MAbs compete for the same or two closely MlMM 2917 8--E


associated epitopes; a high A value indicates the simultaneous binding of both MAbs and the recognition of separate and distinct epitopes. Competitive binding assays of two MAbs on CB were carried out according to the procedure of Harlow and Lane (1987) by measuring the inhibition of the ELISA response obtained with a peroxidase or biotin-labeled MAb in the presence of unlabeled MAbs. A fixed concentration of the labeled MAb was chosen in the linear part of the titration curve and incubated for 1 hr at 37°C in the presence of various concentrations of unlabeled MAb. ELISA was then carried out. Similar competitive binding assays of peroxidase or biotinlabeled and unlabeled MAbs were carried out with CA. In this case, however, sequential incubations were performed first with saturating concentrations of unlabeled MAbs then with labeled MAbs. Epitope mapping of the zone of interaction between CA and CB Microtiter plates were coated with the antigen corresponding to the specificity of the MAb to be tested (CA for anti-CA MAbs and CB for anti-CB MAbs). A fixed concentration of anti-CA MAb, or of anti-CB MAb, was incubated overnight at 4°C with a fixed concentration of CA subunit, or of CB subunit (the concentrations were chosen in order to get 90% of MAb bound to the subunit) and with various concentrations of CB (in the case of anti-CA MAbs), or of CA (in the case of anti-CB MAbs). Aliquots (100 ~1) were then transferred to a microtiter plate coated with CA, or CB, and incubated 5 min (CA) or 10 min (CB) at 4°C. Subsequent steps of ELISA were performed according to the general procedure. Inhibition of phospholipase A, actit7ity and neutralization of crotoxin toxicity by MAbs A fixed concentration of crotoxin (20 pg/ml) was preincubated for 30 min at 37°C in PBS with various concentrations of the purified MAb to be tested. The residual PLAz activity was assayed by titrimetry (Desnuelle et al., 1955) using egg lecithin solubilized by sodium cholate as substrate. The residual lethal potency was estimated by i.v. injections into mice (0.2 ml per 20 g body wt) of aliquots diluted in 150 mM sodium chloride. Carboxymethylation

of crotoxin subunits

Isolated crotoxin subunits CA and CB were reduced and carboxymethylated as described by Faure et al. (1991). However, in the case of CB, the excess reagent was removed by dialysis against 25% acetonitrile and 1% trifluoroacetic acid in water. Carboxymethylated crotoxin subunits CA and CB were then lyophilized and stored at -20°C. RESULTS General characteristics of anti-CA and anti-CB MAbs Eight cell lines secreting anti-CB MAbs were isolated from splenocytes of mice immunized with CB, while six cell lines secreting anti-CA MAbs were obtained for CA.

v. CHOUMETet al.


Table 1. Characteristics

MAbs A-54.60 A-45.6 A-77.25 A-73.13 A-56.36 A-44.13 B-142.7 B-65.5 B-62.10 B-51.20 B-71.1 B-69.9 B-103.7 B-32.13

of anti-CA and anti-C3




IgCl IgG2b IgGl IgGl IgGl IgG 1 IgGl IgG 1 IgGl IgGl IgG2b IgGl IgGl IgGl

6.3-6.7 6.9-7.3 6.7-7.0 6.8-7.1 7.2-7.6 6.8-7.2 6.669 6.1-6.3 6.6-7.0 6.8-7.2 6.1-6.5 6.9-7.2 6.2-6.6 6.0-6.2

25 50 45 300 50 125 60 50 200 30 50 375 40 120


ELISA titers (ngjml) __._. -_--_.__ CA CB

> > > > > > >

30 1.5 60 257 60 250 140 10,000 10,000 10,000 10,000 10,000 10,000 10,000


700 22 50 800 > 10,000 > 10,000 50 70 500 60 70 600 70 375

> 10,000 > 10,000 > 10,000 7400 > 10,000 > 10,000 > 10.000 > 10,000 1 10,000 > 10.000 > 10,000 35 275 > 10,000

Isotypes of MAbs were determined by ELISA using peroxidase-labeled isotype-specific antibodies, The isoelectric points (pHi) of affinity-purified MAbs were determined by isoelectric focusing using polyacrylamide gels over a pH range 3-9, using a PhastSystem (Pharmacia) electrophoresis apparatus. ELISAs were carried out using an indirect procedure on microtiter plates coated with the indicated proteins (a natural mixture of isoforms in the case of crotoxin, CA and CB). ELISA titers are the concentrations of antibodies. in ng/ml, which give half of the maximal response. Vaiues are the means of three independent dete~inations, standard errors being < 10% of the values.

The general characteristics of affinity purified MAbs are shown in Table 1. Anti-CA and anti-CB MAbs reacted strongly with crotoxin and with the corresponding isolated subunit, with ELISA titers ranging from 15 to 600 ng/ml. All anti-CA and anti-CB MAbs reacted with Mojave toxin, a crotoxin-like p-neurotoxin from Crotalus scutulatus scutulatus venom, with ELISA titers very similar to those measured with crotoxin (results not shown). Four anti-CA MAbs (A-54.60, A-45.6, A-77.25 and A-73.13) cross-reacted with the PLAz subunit CB, whilst only one anti-CB MAb (B-142.7) out of eight crossTable 2. Equilibrium MAb A-54.60 A-45.6 A-77.25 A-73.13 A-56.36 A-44.13 B-142.7 B-65.5 B-62.10 B-5 1.20 B-71.1 B-69.9 B-103.7 B-32.13







0.4 0.4 7.5 4 8 5 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000

0.12 0.15 5.4 5 5 8 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000

0.3 0.35 2 2 4 2 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000


0.35 2 2 6 2 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000 > 1000

reacted with CA (Table 1). This observation is in agreement with previous conclusions obtained with rabbit polyclonal antibodies prepared against each crotoxin subunit (Choumet et al., 1989) and is consistent with the structural similarity observed between crotoxin subunits. Table 1 further shows that two anti-CB MAbs (B-69.9 and B-103.7) reacted with AGTX, a single chain neurotoxic PLAz from A. ~10~~~~~ hrekaudus venom which possesses more than 80% sequence similarity with CB. On the other hand, none of the anti-CA or anti-C8 MAbs reacted with AMTX-A, a single chain PLAz neurotoxin from V. ammodytes ammodytes venom,

of MAbs for CA and CB isoforms and for AGTX Antigen CB,, CB,, > 1000 > 1000 > 1000 > 1000 > 1000 >lOOO 80 2 3 3 25 5 2 2

60 70 70 300 > 1000 >lOOO 20 2 5 4 30 7 3 2

CB, 230 300 600 1000 > 1000 > 1000 0.4 2 9 6 4 3 4 2

CB, > > > > > >

1000 1000 1000 1000 1000 1000 0.4 2 2 5 4 4 2 4

CB, > > > > > >

1000 1000 1000 1000 1000 1000 0.5 2 2 4 5 4 2 4

AGTX > > > > > > > > > > >

1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 0.5 5 > 1000

Dissociation constants (&) of antigen-antibody interactions were determined under equilibrium conditions, as described in Materials and Methods, using affinit~purified anti-CA and anti-CB MAbs, and purified CA and CB isoforms (or purified AGTX). Values (&), expressed in nM, are the means of three determinations, standard errors being ~30% of the values.

Immunochemical Table 3. Identification

of epitopic regions of CB by additivity indices B-I *

/ Epitopic regions



B-Ib i


analysis of crotoxin

B-Ia *
















B-142.7 B-65.5 B-62.10 B-5 1.20 B-71.1 B-69.9 B-103.7 B-32.13


0.09 0.00

0.06 0.02 0.00

0.06 0.04 0.04 0.00 -

0.32 0.36 0.32 0.30 0.00

0.69 0.52 0.50 0.50 0.55 0.00

0.74 0.67 0.50 0.50 0.60 0.50 0.00 -

0.60 0.70 0.52 0.56 0.70 0.52 0.65 0.00




for 1 hr at 37°C in microtiter plates coated with CB, at the lowest concentration which gave a maximal ELISA response, either alone or in combination with another MAb, and then ELISAs were carried out according to the general procedure. Additivity indices were calculated as described in Materials and Methods by comparing the sum of the responses obtained with MAb tested separately with the response

Each MAb was incubated

determined with a mixture of these MAbs. Values are the means of three to five independent standard errors being < 10% of the values.

although it exhibits 60% amino with CB and AGTX.

acid sequence


The affinity of anti-CA and anti-CB MAbs for various highly purified CA and CB isoforms has been determined under equilibrium conditions (Table 2). All antiCA MAbs bound all four CA isoforms with the same high affinity constants. Table 2 further shows that four anti-CA MAbs (A-54.60, A-45.6, A-77.25 and A-73.13) bound two CB isoforms (CB,, and CB,) with dissociation constants ranging from 60 nM to 1 PM, but did not interact with other CB isoforms or AGTX. This observation, which cannot be attributed to a contamination of CB,, or CB, by CA, is in good agreement with the ELISA titers determined with mixtures of CA and CB isoforms (Table 1). Additionally, it indicates that the cross-reacting anti-CA MAbs interact with epitopes present on the four CA isoforms and on isoforms CB,, and CB,, but not on the other isoforms of CB (CB,, , CB, and CB,). This is particularly interesting because CB isoforms differ from each other only by a small number of amino acid residues (Faure and Bon, 1988). The two last anti-CA MAbs (A-56.36 and A-44-13) did not recognize any CB isoform. Six out of the eight anti-CB MAbs (B-65.5, B-62.10, B-51.20, B-69.9, B-103.7 and B-32.13) bound the various CB isoforms with comparably high affinities (Table 2). MAbs B-142.7 and B-71.1 have, however, a significantly lower affinity for the two CB isoforms CB,, and CB,,, which make crotoxin complexes with the lowest lethal potency but the highest enzymatic activity (Faure and Choumet, unpublished observations). Table 2 also shows that anti-CB MAbs B-69.9 and B-103.7 bound AGTX with high affinity, whilst other MAbs did not. In the case of MAb B-69.9, the dissociation constant is even lower for AGTX than for any CB isoform (0.5 nM compared with 3-7 nM). Enhanced affinity of polyclonal and monoclonal antibodies for a heterologous antigen has been described (Makehi, 1965; Loor, 1971; Al Moudallal et al., 1982; Harper et al., 1987; Tipton et al.,


1990). These MAbs were called heteroclitic antibodies. On the other hand, we observed no cross-reaction between any anti-CB MAb and any CA isoform in solution (Table 2). However, one may note that the anti-CB MAb B-142.7, although cross-reacting with a natural mixture of CA isoforms (Table 1) or with each isolated CA isoform (results not shown) when tested by ELISA, has no affinity for any tested CA isoform in solution (Table 2). This observation might be explained, as demonstrated in the case of human growth hormone (Mazza and Retegui, 1989), by assuming that CA exposes an epitope recognized with a low affinity by MAb B-142.7 when CA is adsorbed on the solid phase of microtiter plates. None of the anti-CA or anti-CB MAbs bound with significant affinity any other tested toxic or nontoxic PLA, (i.e. AMTX-A from V. ammodytes ammodytes venom, taipoxin from Oxyuranus scutellatus scutellatus venom, textilotoxin from Pseudonaja textilis venom, /?-bungarotoxin from Bungarus multicinctus venom, notexin from Notechis scutatus scutatus venom, PLA,s from Vipera berus, Crotalus adamanteus and Naja nigricollis venoms), as measured by ELISA (results not shown). Similarly, none of the anti-CA or anti-CB MAbs interacted with reduced and carboxymethylated CA or CB (results not shown), suggesting that they recognize conformational epitopes. Identification of epitopic regions present on crotoxin subunits CA and CB In order to identify the various epitopic regions recognized by anti-CB MAbs, we examined the ability of two MAbs to bind simultaneously. As shown in Table 3, four epitopic regions can be defined on CB, one of them (B-l) being divided in two parts. The first one, B-Ia, is recognized by four anti-CB MAbs (B-142.7, B-65.5, B-62.10 and B-51.20): additivity indices, determined for each combination of MAbs, were not significantly different from zero, indicating that only one of these MAbs


V. CHOUMETet al. Table 4. Competitive B-I h

r Epitopic regions


B-142.7 B-5 1.20 B-103.7 B-32.13

B-Ia h



binding of anti-CB MAbs














3.5 4 > 500 > 500

6 1.9 > 500 > 500

8 3 > 500 > 500

4.5 1.7 > 500 > 500

17 16 > 500 > 500

> > > >

> 500 > 500 0.7 > 500

> 500 > 500 > 500 2.1

Peroxidase Peroxidase Peroxidase Biotin

A fixed concentration of labeled MAb, in CB-coated microtiter plates with out as described in Materials and expressed in ng/ml, which reduced

500 500 500 500

chosen in the linear part of the titration curve, was incubated for 1 hr at 37-C various dilutions of the indicated unlabeled MAb. ELISAs were further carried Methods. Inhibition was estimated as the concentration of unlabeled MAb, the binding of labeled MAb by 50%.

can bind at the same time to CB. The other epitopic regions (B-Ib, B-II, B-III and B-IV) were defined with a single MAb in each case (B-71.1, B-69.9, B-103.7 and B-32.13, respectively). When compared in a quantitative manner, the relative values of additivity indices indicated: (1) that regions B-II, B-III and B-IV are distinct from one another, as well as from regions B-Ia and B-Ib, as additivity indices are higher than 0.5; (2) that the two epitopic regions B-Ia and B-Ib might be partially overlapping since additivity indices varied from 0.3 to 0.4 according to the MAb pairs, indicating that when a MAb is bound to one region (B-Ia or B-Ib) binding of a second MAb to another epitope is inhibited. A low additivity index has been determined in the case of two anti-lysozyme MAbs which were shown by three-dimensional analysis of a complex Fab-lysozyme, to bind to opposite sides on the antigen (Harper et al., 1987). Thus, we used competitive binding experiments in order to examine whether this hypothesis also applies in the case of epitopic regions of CB. A labeled anti-CB MAb was incubated at a fixed and nonsaturating concentration in a microtiter plate coated with CB in the presence of various concentrations of another unlabeled anti-CB MAb. and competitive binding was determined by measuring a decrease in the ELISA response. No competition was observed between MAbs recognizing epitopic regions B-Ia, B-II, B-III and B-IV (Table 4), in agreement with the conclusions of additivity index

Table 5. Identification

measurements. This provides further evidence that these epitopic regions are distinct on CB. In the case of unlabeled MAb B-71.1, which interacts with the epitopic region B-Ib, we observed a competition with the binding of peroxidase-labeled MAbs which recognize the region B-Ia, at a MAb concentration 10 times higher than that determined with homologous MAbs (Table 4). This indicates that epitopic regions B-Ia and B-Ib are distinct but partially overlapping rather than opposite. This, however, does not exclude that MAb B-71 .l could induce a conformational change of the epitopic region B-Ia responsible for the decreased affinity of MAbs which bind to this region. A similar investigation was carried out in the case of CA with peroxidase-labeled and unlabeled anti-CA MAbs. As shown in Table 5, three independent epitopic regions, A-I, A-II and A-III, were defined. The binding of peroxidase-labeled MAb A-45.6 was efficiently prevented by the four unlabeled MAbs, A-45.6, A-54.60, A-73.13 and A-77.25, but not by the two other MAbs, A-56.36 and A-44.13, indicating that MAbs A-45.6, A-54.60, A-73.13 and A-77.25 bind to the same epitopic region, A-I, whilst the two others bind to different regions. On the other hand, the binding of peroxidase-labeled MAb A-56.36 was not inhibited by any heterologous MAb (Table 5), indicating that it recognizes an epitopic region, A-II, different from A-I and from the epitopic region A-III recognized by A-44.13.

of epitopic regions of CA

A-I Epitope MAb A-45.6-PO A-56.36-PO





A-II A-56.36

A-III A-44.13

1 97

2 98

4 96

8 99

98 2

97 99

Microtiter plates coated with CA were preincubated for 1 hr at 37’C with unlabeled MAb at a concentration which gave maximal ELISA response in control experiments. A second incubation was then performed for 1 hr at 37’C with peroxidase-labeled MAbs, at a concentration chosen in the linear part of ELISA signal. Values are expressed as percent of the ELISA response obtained in the absence of unlabeled MAb. Each determination was performed in triplicate, standard errors being ~5% of the values.


-6 CBd (Mf

CA2 (M1

Fig. 1. Inhibition of MAb binding upon the interaction of crotoxin subunits. Concentrations of anti-CA (A) or anti-CB (B) MAbs equal to their dissociation constant (Table 3) were preincubated overnight at 4°C with IO-‘M CA? isoform {A) or IO-* M CB, isoform (B) and the indicated concentrations of the other subunit CB, (A) or CA, (B). Ahquots (100 ~1) were transferred in microtiter plates coated with a mixture of CA isoforms (A) or a mixture of CB isoforms (B) and an ELISA was carried out as described in Materials and Methods. ELISA responses were determined in triplicates and expressed as percentage, using as zero the ELISA response when the assay was performed in the absence of CA, (B) or CB, (A) and as 100% the ELISA response when the assay was carried out in the absence of both subunits. (A) MAbs A-45.6, A-54.60, A-77.25 and A-73.13 (a); A-56.36 (0) and A-44.13 (u). The arrow indicates the concentration of CA2 isoform. (B) MAbs B-142.7, B-69.9 and B-32.13 (A) B-65.5 (0); B-62.10 (H); B-51.20 (a); B-71.1 (V); B-103.7 (Of. The arrow indicates the concentration of CB, isoform.


analysis of crotoxin

observed competition. This conformational change should, however, involve epitopic regions A-II and A-III without affecting region A-I. Figure l(B) shows that CA did not inhibit the binding of MAbs B-69.9 and B-32.13 which interact with epitopic regions B-II and B-IV, whilst at a concentration similar to that of CB, it prevented the binding of MAbs B- 103.7 and B-71.1, which recognized regions B-III and B-Ib, respectively. This suggests that regions B-III and B-Ib are focalized, at least in part, in the zone of interaction between the two crotoxin subunits, whereas regions B-II and B-IV are not (Fig. 2). CA prevented the binding of MAbs interacting with epitopic region B-Ia with the exception, however, of MAb B-142.7, indicating that this region might also be localized in the zone of interaction. On the other hand, as in experiments performed with anti-CA MAbs, a conformational change of CB upon binding with CA cannot be excluded. The experiments shown in Fig. 1 were carried out with isoforms CA2 and CB,. Similar results were obtained with different combinations of other isoforms of both subunits, CA, or CA, on one hand and CBa2, CB,, CB, or CB, on the other (results not shown). These results indicate that epitopic regions of CA and CB localized in the zone of interaction between the two subunits are, as expected, the same for all crotoxin isoforms. EfSect of MAbs on the lethal potency and enzymatic activity of crotoxin The three tested anti-CA MAbs (A-54.60, A-77.25 and A-73.13) which bind to the epitopic region A-I were


Identification of epitopic regions localized to the zone of interaction between crotoxin subunits The epitopic regions localized in the areas implicated in the association of the two crotoxin subunits (the zone of interaction) were identified by measuring the ability of one subunit to inhibit the binding of MAbs directed towards the other. Figure l(A) shows that CB did not compete with the binding of the four MAbs (A-45.6, A-54.60, A-77.25 and A-73.13) which interacted with region A-I of CA, indicating that region A-I is not involved in the interaction between crotoxin subunits (Fig. 2). On the other hand, stoichiometric concentrations of CB inhibited the binding of MAbs A-56.36 and A-44.13, suggesting that regions A-II and A-III, which are recognized by these two MAbs, belong, at least in part, to the zone of interaction between CA and CB (Fig. 2). A conformational change of CA induced by its binding with CB might also be responsible for the

CA removed during posttranslational

A-I A-54.60 A-73.13 A-77.23 A-45 6

Fig. 2. Schematic representation of epitopic regions on crotoxin.



Molar ratio





Molar ratio



Fig. 3. Inhibition of PLAz activity and neutralization of lethal potency of crotoxin by anti-CB MAbs. A fixed concentration of crotoxin (20 pg/ml; 20 LD,,/ml) was incubated for 1 hr at 37’C with various concentrations of anti-CB MAbs. Aliquots were tested for their residual PLA, activity using the pH-stat method and for their residual lethal potency by i.v. injection into mice (three mice tested per dose). (A) Percent inhibition of PLA, activity was plotted as a function of MAb concentration (expressed as molar ratio to crotoxin). MAbs B-71.1 (0); B-69.9 (0); B-142.7, B-65.5, B-62.10, B-51.20, B-103.7 and B-32.13 (0). (B) Residual lethal potency (expressed as percent of surviving mice) was plotted as a function of MAb concentration (expressed as molar ratio to crotoxin). MAbs B-71.1 (0); B-69.9 (0); B-103.7 (0); B-32.13 (m); B-142.7. B-65.5, B-62.10 and B-51.20 (0).

unable to decrease significantly the lethal effect of crotoxin or to reduce its PLA, activity, even in IO-fold excess (results not shown). This is in agreement with previous observations made with polyclonal antibodies directed against CA, which were unable to inhibit the enzymatic activity of crotoxin or to neutralize its toxicity (Choumet et al., 1989). It should be noted, however, that MAbs A-56.36 and A-44.13, which bind to epitopic region A-II and A-III localized in the zone of interaction between CA and CB, have not yet been tested for their ability to neutralize the lethal potency of crotoxin. On the other hand, the two anti-CB MAbs, B-71 .l and B-69.9, which recognized regions B-Ib and B-II, respectively, were found to inhibit the PLA, activity of crotoxin [Fig. 3(A)]. This suggests that interaction of MAbs with regions B-Ib and B-II interferes with the “catalytic site” of CB defined as residues involved in the catalytic mechanism and residues implicated in the binding of phospholipids (Fig. 2). Binding of B-71.1 or B-69.9 on crotoxin might also induce a conformational

et al.

change in CB leading to reduced enzymatic activity. It should be recalled that epitope region B-Ib is localized. at least in part, in the zone of interaction between the two crotoxin subunits, whereas B-II is not (Fig. 2). Three anti-CB MAbs (B-71.1, B-69.9 and B-32.13) were found to protect mice efficiently against the lethal potency of crotoxin [Fig. 3(B)]. These three neutralizing anti-CB MAbs recognize three different epitopic regions, B-Ib, B-II and B-IV. Their neutralizing effect may be due to a partial overlap of a “toxic site” on CB (Fig. 2). It is important to emphasize, however, that all the regions defined in Fig. 2, including the “toxic site” refer to phenomenological properties and not to topological structures. Two of the three epitopic regions which partially overlap the “toxic site”, B-Ih and B-II, are also localized in the vicinity of the “catalytic site” of CB, and the epitopic region B-Ib also overlaps the zone of interaction between the two crotoxin subunits (Fig. 2). With the exception of MAb B-103.7, which exerted a partial protection at a high molar ratio, all other anti-CB MAbs were unable to reduce the lethal potency of crotoxin [Fig. 3(B)]. The corresponding epitopic regions recognized by these MAbs were therefore considered as unrelated to the “toxic site” (Fig. 2). DISCUSSION

We have defined three independent epitopic regions. A-I, A-II and A-III, on the CA subunit of crotoxin. Two of these regions (A-II and A-III) are shown to be involved in the zone of interaction between the two crotoxin subunits. On the other hand, four independent epitopic regions (B-I, B-II, B-III and B-IV) were defined on the CB subunit, region B-I being subdivided into two parts (B-Ia and B-Ib) which may be partia!ly overlapping. The two regions, B-I and B-III, belong. at least in part, to the zone of interaction of CB with CA: binding of MAbs to the two regions, B-lb and B-II. interfere with the catalytic and toxic activities of crotoxin, whilst regions B-III and B-IV are involved only in the neutralization of the lethal potency of crotoxin. This is in agreement with previous observations which showed that anti-CB polyclonal antibodies are able to neutralize the toxicity of crotoxin and to inhibit its PLA, activity (Choumet et al., 1989). It is of interest to note that when an epitopic region has been defined with several MAbs, all the MAbs interacting with this region present the same properties, i.e. cross-reactivity. inhibition of enzymatic activity, neutralization of lethal potency and recognition of the zone of interaction between CA and CB. However, in contrast with the other MAbs which define the epitopic region B-Ia, MAb B-142.7 does not appear to bind within the zone of interaction between the crotoxin subunits. This may be easily rationalized by assuming that the part of region B-Ia recognized by MAb B-142.7 extends beyond the zone of interaction between the two crotoxin subunits. The existence of three and four epitopic regions on CA and CB, respectively, is in good agreement with previous studies: three and four epitopic regions have been


CA2 AMTX-A C. ad. venom



10 R-












120 ****** ooooooo QFSPENCQGESQPC







Fig. 4. Amino acid residues of CA and CB involved in different epitopes. The polypeptide sequence of CA2, the sequences of one isoform of CB, C61, deduced from tfie cDNA, AGTX, AMTX-A and of the PLA, from Crotalus udumanteus (C. ad.) venom were from Fame et al. (1991); Bouchier et al. (1991), Kondo et al. (19X9), Ritonja and Gubensek (1985) and Heinrickson et al. (1977), respectively. The numbering is taken from Renetseder ef al. (1985). 0 indicates the cyclization of N-terminal glutaminyl residues to pyrrolidone carboxylyl (pyroglutamyl) residues, {A) Residues present on CA but not on listed PLA,s are indicated by an asterisk. (B) Common residues between CB and AGTX but absent in other PLA,s are identified in boxes.


c___ C___


















c__-_ c___ C-_-





AGTX AMTX-A C. ad. venom




% !? Fi s 5'




V. CHOUMET et al.



Fig. 5. Hypothetical three-dimensional structure of CA and CB showing the amino acid residues involved in several epitopes. The three-dimensional model proposed by Dijkstra et al. (1978) for bovine pancreatic PLAz has been modified according to the structure of Crotalus atrox venom PLA, (Brunie et al., 1985) to take into account the greater similarities of CA and CB with this enzyme. The numbering of amino acid residues is that described by Renetseder et al. (1985). The positions of disulfide bridges are those indicated by Aird et al. (1985). Cystines are shown in open circles. CA: residues present on CA but not on PLA,s listed in Fig. 4 are identified in black circles. CB: common residues between CB and AGTX but absent in other PLAzs are identified in black circles.

defined, respectively, in the cases of porcine pancreatic PLA, (Meijer et al., 1978) and P-bungarotoxin (Chang and Lin, 1983). The MAbs which interact with epitopic regions A-II and A-III were found to react with CA but not with CB or with other toxic and nontoxic PLA,s. Because of the strong homology among these molecules, it is expected that the amino acid residues present in CA but absent from other PLA,s are likely to contribute to regions A-II and A-III, which are specific of CA. An examination of the amino acid sequences shows that these nonhomologous residues are spread over the entire polypeptide chain [Fig. 4(A)], suggesting that the epitopic regions are not linear. This hypothesis is confirmed by the observation that anti-CA MAbs did not react with reduced and carboxymethylated CA. On the other hand, assuming that the three-dimensional structure of CA is similar to that of PLA, from Crotafus atrox venom (Brunie et al., 1985), these residues mainly belong to two large regions localized at two extremities of the CA molecule (Fig. 5). The distance between these two areas being larger than the size of an antibody combining site (Amit et al., 1986), it is tempting to hypothesize that they correspond to the distinct epitopic regions A-II and A-III. Anti-CB MAbs which interact respectively with regions B-II and B-III, also cross-react with AGTX from A. blomhofii breuicaudus, but not with AMTX-A from V. ammodytes ammodytes or with other toxic and

nontoxic PLA,s. Thus, amino acid residues present simultaneously in CB and AGTX, but absent from AMTX-A and other PLA,s, have a higher probability to be involved in the epitopic regions B-II and B-III. As shown in Fig. 4(B), these residues are spread over the polypeptide chain. This suggests that MAbs B-69.9 and B-103.7 interact with nonlinear epitopes, as confirmed by the fact that they do not react with reduced and carboxymethylated CB. Conformational epitopes might be the general rule for PLA,s, as expected for rather globular molecules containing numerous cc-helical structures. In agreement, three MAbs interacting with the same conformational epitope have been reported in the case of /?-bungarotoxin (Strong et al., 1984). However, two MAbs raised against native AMTX-A recognize synthetic peptides corresponding to different parts of the C-terminal polypeptide sequence of the protein (Curin-Serbec et al., 1991). The residues involved in the epitopic regions B-II and B-III, which distinguish CB and AGTX from other PLA,s, have been localized on a hypothetical three-dimensional structure of CB, built by assuming similarity in the three-dimensional structures of CB and C. atrox venom PLA,, as is the case for CA. They extend over a large portion of the surface of the molecule, so that it is not easy to recognize the relative positions of the two epitopic regions, B-II and B-III (Fig. 5). Four epitopic regions of CB are involved in the neutralization of the lethal action of crotoxin. suggesting


that different mechanisms might be implicated. Binding of MAbs to two of them (B-Ib and B-II) also interferes with the enzymatic activity of the toxin. This may be the basis of their neutralizing effect. Several lines of evidence including irreversible inactivation of PLA, activity by p-bromophenacyl bromide and removal of the activating calcium ions have demonstrated the involvement of the catalytic activity of crotoxin in its neurotoxic action (Jeng et al., 1978; Marlas and Bon, 1982). It is of interest to note that a large number of amino acid residues postulated to belong to epitope region B-II (Fig. 5) are involved in the hydrophobic cleft, constituted by residues Leuz , Ile,, , Pro,, and Try,, (Radvanyi et al., 1985) and implicated in the interaction of PLA,s with their substrates (Slotboom et al., 1982; Verheij et al., 1981). Dissociation of the crotoxin complex results in the reduction of its toxicity and may be induced by MAb B-103.7, which interacts with region B-III, as this region belongs, at least in part, to the zone of interaction between the crotoxin subunits. This hypothesis agrees with the fact that the neutralization potency of MAb B-103.7 is weak, as expected if it dissociates the crotoxin complex without neutralizing the residual lethal potency of isolated CB. This mechanism, however, cannot be invoked in the case of epitopic region B-IV, suggesting that this part of the molecule might be involved in the selective recognition of the target site of CB. A similar mechanism has been proposed for MAbs and polyclonal antipeptide antibodies which neutralize the lethal potency of AMTX-A without inhibiting its enzymatic activity (Curin-Serbec et al., 1991). thank Dr Faridabano Nato (Institut Pasteur, Paris, France) for useful advice during this study and for suggestions regarding the manuscript. We are grateful to Professor Yuan-Cong Chen (Shanghai’ Institute of Biochemistry, China), Dr Franc Gubensek (Josef Stefan Institute, Ljubljana, Slovenia) and Dr Alan Coulter (Commonwealth Laboratories, Sydney, Australia) for their generous gifts of purified /I-neurotoxins. This research was supported in part by funds from the Direction des Recherches et Essais Techniques (DRET). V. Choumet was a recipient of a fellowship from DRET. Acknowledgements-We

REFERENCES Al Moudallal Z., Briand J. P. and Van Regenmortel M. H. V. (1982) Monoclonal antibodies as probes of the antigenic structure of tobacco mosaic virus. EMBO J. 1, 1005-1010. Aird S. D. and Kaiser I. I. (1985) Comparative studies on three rattlesnake neurotoxins. Toxicon 23, 361-374. Aird S. D., Kaiser I. I., Lewis R. V. and Kruggel W. G. (1985) Rattlesnake presynaptic neurotoxins: primary structure and evolutionary origin of the acidic subunit. Biochemistry 24, 7054-7058.

Aird S. D., Kruggel W. G. and Kaiser I. I. (1990) Amino acid sequence of the basic subunit of mojave toxin from the venom of the Mojave rattlesnake. Toxicon 28, 669-673. Amit A. G., Mariuzza R. A., Phillips S. E. V. and Poljak R. J. (1986) Three-dimensional structure of an antigen-antibody complex at 2.8 A resolution. Science 233, 747-753.

analysis of crotoxin


Avrameas S. (1969~) Coupling of enzymes to proteins with glutaraldehyde. Use of the conjugates for the detection of antigen and antibodies. Immunochemistry 6, 43-52. Avrameas S. (19696) Indirect immunoenzyme techniques for the intracellular detection of antigens. Immunochemistry 6, 925.

Bartolf M. and Franson R. C. (1990) Characterization and partial purification of soluble, lysosomal phosphohpase(s) A, from adrenal medulla. Biochem. biophys. Acta 1042, 247-254.

Bieber A. L., Becker R. R., McParland R., Hunt D. F., Shabanowitz J., Yates III J. R., Martin0 P. A. and Johnson G. R. (1990) The complete sequence of the acidic subunit from Mojave toxin determined by Edman degradation and mass spectrometry. Biochem. biophys. Acta 1037, 413-421.

Bon C. and Jeng T. W. (1979) Crotoxin: a possible mechanism of action. Adt. Cytopharm. 3, 231-235. Bouchier C., Boulain J. C., Bon C. and MCnez A. (1991) Analysis of cDNAs encoding the two subunits of crotoxin, a phospholipase A? neurotoxin from rattlesnake venom. Biochim. biophys. Acta 1088, 401-408.

Brunie S., Bolin J., Gewirth D. and Sigler P. B. (1985) The refined crystal structure of dimeric phospholipase A1 at 2.5 A. J. biol. Chem. 260, 9742-9749. Chang C. C. and Lin C. Y. (1983) Immunochemical characterization of b I-bungarotoxin from the venom of Bungarus 216-226.


Proc. natn. Sci. Count.

B. Rot.


Chen Y., Wu X., Zhang J., Jiang M. and Hsu K. (1987) Characterization of the structure and function of three phospholipases A? from the venom of Agkistrodon halys Pallas. Acta biochem. biophys. Sin. 3, 205. Choumet V., Jiang M. S., Radvanyi F.. Ownby C. and Bon C. (1989) Neutralization of lethal potency and inhibition of enzymatic activity of a phospholipase A2 neurotoxin, crotoxin, by non-precipitating antibodies (Fab). FEBS Lett. 244, 167-173.

Cuatrecasas P. and Anfinsen C. B. (1971) Affinity chromatography. Am. Rev. Biochem. 40, 259-278. Curin-Serbec V., Novak D., Babnick J., Turk D. and Gubensek F. (1991) Immunological studies of the toxic site in ammodytoxin A. FEBS Lett. 280, 175-178. Davidson F. F. and Dennis E. A. (1990) Evolutionary relationships and implications for the regulation of phospholipases A,: from snake venoms to human secreted forms. J. molec. Evol. 31, 228-230. Desnuelle P., Constantin J. M. and Baldy J. (1955) Technique potentiometrique pour la mesure de l’activite de la lipase pancreatique. Bull. Sot. chim. Biol. 37, 285-290. Dijkstra B. W., Drenth J., Kalk K. H. and Vandermaelen P. J. (1978) Three-dimensional structure and disulfide bond connections in bovine pancreatic phospholipase A?. J. molec. Biol. 124, 53-60.

Dijkstra B. W., Kalk K. H., HOI W. G. J. and Drenth J. (1981) Structure of bovine pancreatic phospholipase A, at 1.7 8, resolution. J. molec. Biol. 147, 97-123. Faure G. and Bon C. (1987) Several isoforms of crotoxin are present in individual venoms from the South American rattlesnake Crotalus durissus terrificus. Toxicon 25,229-234. Faure G. and Bon C. (1988) Crotoxin, a phospholipase A2 neurotoxin from the South American rattlesnake Crotalus durissus terrijicus: purification of several isoforms and comparison of their molecular structure and of their biological activities. Biochemistry 27, 730-738.


v. CH0UMET et ui

Fame G.. Guillaume J. L., Camoin L., Saliou B. and Bon C. (1991) The multiplicity of acidic subunit isoforms of crotoxin, the phospholipase A1 neurotoxin from Crotalus durissus terr$cus venom, results from post-translational modifications. Biochemistry 30, 807448083. Friguet B., Chaffotte A. F., Djavadi-Ohaniance L. and Goldberg M. E. (1985) Measurements of the true affinity constant in solution of antigen-antibody complexes by enzyme-linked immunosorbent assay. J. Immun. Meth. 77, 3055319. Friguet B., Djavadi-Ohaniance L., Pages J., Bussard A. and Goldberg M. E. (1983) A convenient enzyme-linked immunosorbent assay for testing whether monoclonal antibodies recognize the same antigenic site. Application to hybridomas specific for the 82-subunit of Escherichiu coli tryptophane synthetase. J. Immun. Meth. 60, 351-358. Harlow E. and Lane D. (1987) In Antibodies: A Laborator_r Munuui (Edited by Cold Spring Harbor Laboratory), p. 567. Cold Spring Harbor, New York. Harper M.. Lema F., Boulot G. and Poljak R. J. (1987) Antigen specificity and cross-reactivity of monoclonal antilysozyme antibodies. Molec. Immun. 24, 97-108. Hawgood B. and Bon C. (1991) Snake venom presynaptic toxins. In Handbook of‘Nuturu1 Toxins. Reptile Venoms und To.rins (Edited by Tu A. T.). Vol. 5, pp. 3-52. Marcel Dekker, New York. Heinrickson R. L., Krueger E. T. and Keim P. S. (1977) Amino acid sequence of phospholipase A2 from the venom of Crotulus udumunteus. A new classification of phospholipases A? based upon structural determinants. J. biol. Chem. 252, 4913-4921. Hsu K.. Zhang J. and Jiang M. (1976) Studies on neurotoxic components in the venom of Agkistrodon hulys PallasIsolation of neurotoxin from the venom. Arta Biophys. Sin. 8, 3577362. Jeng T. W., Hendon R. A. and Fraenkel-Conrat H. (1978) Search for relationships among the hemolytic, phospholipolytic and neurotoxic activities of snake venoms. Proc. nutn. Acad Sci. U.S.A. 75, 600.-604. Kim D. K., Suh P. G. and Ryu S. H. (1991) Purification and some properties of a phospholipase A, from bovine platelets. Biochem. biophys. Res. Commun. 174, 1899196. Kohler G. and Milstein C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495. Kondo K., Narita K. and Lee C. Y. (1978~) Amino acid sequences of the two polypeptide chains in /I I-bungarotoxin from the venom of Bungurus multicinctus. J. Biochem. 83, 101--l 15. Kondo K.. Narita K. and Lee C. Y. (1978b) Chemical properties and amino acid composition of /I I-bungarotoxin from the venom of Bungurus multicinctus (Formosan banded krait). J. Biochem. 83, 91-99. Kondo K.. Zhang J. K., Xu K. and Kagamiyama H. (1989) Amino acid sequence of a presynaptic neurotoxin. agkistrodotoxin, from the venom of Agkistrodon huj.w (Pallas). J. Biochem. 105, 196-203. Lee C. Y., Tsai M. C., Chen Y. M., Ritonja A. and Gubensek F. (1984) Mode of neuromuscular blocking action of toxic phospholipases A, from Vipera ammodytes venom. Arch. Int. Pharmucodyn. Therup. 268, 3 13-324. Loor F. (1971) On the existence of heterospecific antibodies in sera from rabbits immunized against tobacco mosaic virus determinants. Immunology 21, 557-564.

Mikela 0. (1965) Single lymph node cells producing heteroclitic bacteriophage antibody. J. Immun. 95, 378-386. Marlas G. and Bon C. (1982) Relationship between the pharmacological action of crotoxin and its phospholipase activity. Eur. J. Biochem. 125, 157-165. Mazza M. M. and Retegui L. A. (1989) The antigenic topography of human growth hormone. Molec. Immun. 26, 231-240. Meijer H., Meddens J. M., Dijkman R., Slotboom A. J. and De Haas G. H. (1978) Immunological studies on pancreatic phospholipase A:. J. biol. Chem. 23, 8564-8569. Pool W. R. and Bieber A. L. (1981) Fractionation of midget faded rattlesnake (Crotalus r:iridis concolor) venom: lethal fractions and enzymatic activities. Tovicon 19, 5 17-527. Radvanyi F., Rousselet A., Devaux P. and Bon C. (1985) Interaction of crotoxin and its isolated subunits with spinlabeled fatty acids. J. biol. Chem. 260, 876558770. Renetseder R., Brunie S., Dijkstra B. W., Drenth J. and Sigler P. B. (1985) A comparison of the structures of phospholipase A? from bovine pancreas and Crotulus atros venom. J. biol. Chem. 260, 1 I ,627-l 1,634. Ritonja A. and Gubensek F. (1985) Ammodytoxin A, a highly lethal phospholipase A> from Viperu ummodytes ummodvtes venom. Biochem. biophys. Acta 828, 306-312. Schmidt R. R. and Betz H. (1989) Cross-linking of b-bungarotoxin to chick brain membranes. Identification of subunits of a putative voltage-gated K+ channel. Biochemistry 28, 834668350. Scott D. L.. Otwinowski 2.. Gelb M. H. and Sigler P. B. (1990) Crystal structure of bee-venom phospholipase Az in a complex with a transition-state analogue. Science 250, 1563-l 566. Slotboom A. J., Verheij H. M. and De Haas G. H. (1982) On the mechanism of phospholipase A,. In Phosphohpids (Edited by Hawthorne J. N. and Ansell G. B.). Vol. 4, pp. 3599434. Elsevier Biomedical, New York. Strong P. N., Wood J. N. and Ivanyi J. (1984) Characterization of monoclonal antibodies against fi-bungarotoxin and their use as structural probes for related phospholipase A1 enzymes and presynaptic phospholipase neurotoxins. Eur. J. Biochem. 142, 145-l 51. Tipton D. A., Walker W. S. and Schonbaum G. R. (1990) Epitope mapping of horseradish peroxidase with use of monoclonal antibodies. Hvbridomu 9, 319-330. Vandermeers A.. Vandermeers-Piret M.-C., Vigneron L., Stievenart M. and Christophe J. (1991) Difference in primary structure among five phospholipases AZ from Helodermu suspecturn. Eur. J. Biochem. 196, 5377544. Verheij H. M., Slotboom A. J. and De Haas G. H. (1981) Structure and function of phospholipase A,. Rec. Physiol. Biochem. Phurmuc. 91, 91-203. Waite M. (1987) The phospholipases. In Handbook of Lipid Research. (Edited by Hanahan D. J.). Plenum Press. New York. Wery J. P., Schevitz R. W., Clawson D. K., Bobbitt J. L.. Dow E. R., Gamboa G., Goodson T., Hermann R. B., Kramer R. M., McClure D. B., Mihelich E. D., Putnam J. E., Sharp J. D., Stark D. H., Teater C., Warrick M. W. and Jones N. D. (1991) Structure of recombinant human rheumatoid arthritic synovial fluid phospholipase A> at 2.2 A resolution. Nature 352, 79-82. White S. P.. Scott D. L., Otwinowski Z., Gelb M. H. and Sigler P. B. (1990) Crystal structure of cobra-venom phospholipase A? in a complex with a transition-state analogue. Science 250, 1560-1563.

Immunochemical analysis of a snake venom phospholipase A2 neurotoxin, crotoxin, with monoclonal antibodies.

Crotoxin is the major neurotoxic component of the venom of the South American rattlesnake, Crotalus durissus terrificus. The crotoxin molecule is comp...
1MB Sizes 0 Downloads 0 Views