Mapping of an epitope recognized by a neutralizing monoclonal antibodyspecific to toxin Cn2 from the scorpion Centruroides noxius, usingdiscontinuous synthetic peptides

Emma S. Calderon-Aranda1,2, Barbara Selisko1, Eunice J. York3 Georgina B. Gurrola1, John M. Stewart3,*and Lourival D. Possani1,*1Department of Molecular Recognition and Structural Biology, Instituto de BiotecnologõÂa, Universidad Nacional AutoÂnoma de MeÂxico,

Cuernavaca, Mexico; 2SeccioÂn de ToxicologõÂa Ambiental, Departamento de FarmacologõÂa y ToxicologõÂa, CINVESTAV-IPN, MeÂxico;3Department of Biochemistry, University of Colorado Medical School, Denver CO, USA

The Na+-channel-affecting toxin Cn2 represents the major and one of the most toxic components of the venom of

the Mexican scorpion Centruroides noxius Hoffmann. A monoclonal antibody BCF2 raised against Cn2 has been

shown previously to be able to neutralize the toxic effect of Cn2 and of the whole venom of C. noxius. In the

present study the epitope was mapped to a surface region comprising the N- and C-terminal segments of Cn2,

using continuous and discontinuous synthetic peptides, designed on the basis of the sequence and a three-

dimensional model of Cn2. The study of peptides of varying length resulted in the identification of segments

5±14 and 56±65 containing residues essential for recognition by BCF2. The peptide (abbreviated SP7) with the

highest affinity to BCF2 (IC50 = 5.1 mm) was a synthetic heterodimer comprising the amino acid sequence from

position 3±15 (amidated) of Cn2, bridged by disulfide to peptide from position 54±66, acetylated and amidated.

Similar affinity was found with peptide SP1 [heterodimer comprising residues 1±14 (amidated) of Cn2, bridged

with synthetic peptide 52±66 (acetylated)]. SP1 and SP7 were used to induce anti-peptide antibodies in mouse

and rabbit. Both peptides were highly immunogenic. The sera obtained were able to recognize Cn2 and to

neutralize Cn2 in vitro. The most efficient protection (8.3 mg Cn2 neutralized per mL of serum) was induced by

rabbit anti-SP1 serum.

Keywords: synthetic peptide; anti-peptide sera; Centruroides noxius; immunotherapy; neutralizing epitope;

scorpion toxin.

In many parts of the world scorpion stings represent animportant public health problem. In Mexico, several species ofthe genus Centruroides are medically important [1,2]. Themolecules responsible for the toxicity of the scorpion venomare polypeptides of 4±8 kDa (reviewed in [3]). They have beenclassified according to the target they recognize. The peptidesthat bind to sodium channels are 60±70 amino acids long andstabilized by four disulfide bonds (reviewed in [4]). They are byfar the most important due to the fact that they represent themajor toxic components of the venom in terms of concen-trations and toxicity [3]. They disrupt the normal function ofexcitability causing respiratory and circulatory problems, whicheventually can lead to the death of the individual envenomed[1] (reviewed in [5±7]). In addition to the Na+-channel specifictoxins in the venom of scorpions, there are toxins specific forK+-channels [8] (reviewed in 9,10]), Ca2+-channels (reviewedin [11]) and Cl 2 channels [12].

At present the only specific treatment for envenomation withscorpion venom is the use of anti-scorpion serum. This isobtained from horses immunized with macerates of venomousglands. The horse immunoglobulins are digested with pepsin[2] and the corresponding F(ab 0)2 are purified and used as anti-venom. This product contains irrelevant F(ab 0)2, which requiresthe use of large amounts of anti-venom to neutralize the effectsof the venom. Therefore, it would be advantageous to use animproved antigen consisting mainly of a specific neutralizingepitope in order to produce a more efficient and specific anti-venom [2].

Sodium-channel-affecting toxins from scorpions of the samegenus (e.g. Centruroides) have highly homologous primarysequences [13] as well as a very well conserved three-dimensional structure [14±17]. This results in the presence ofhomologous neutralizing antigenic determinants [18±20]. Cn2is one of the most potent sodium-channel-affecting toxins(LD50 = 0.25 mg per 20 g injected intraperitoneally in mice ofthe strain CD1 [21]) in the venom of Centruroides noxius and,additionally, it represents the most abundant component [18].Cn2 was shown to act as a b-toxin, affecting the activation ofthe sodium channel [22]. Polyclonal antibodies generatedagainst Cn2 were able to recognize and protect envenomationby whole venom. One milliliter of anti-Cn2 serum neutralized57 LD50 (298 mg) of venom protein [19]. A monoclonalantibody BCF2, capable of recognizing and neutralizing thetoxic effect of purified Cn2, was obtained [18]. One mg(6.7 nmol) of BCF2 neutralizes 32 LD50 of purified Cn2, but it

Eur. J. Biochem. 264, 746±755 (1999) q FEBS 1999

Correspondence to L. D. Possani, Department of Molecular Recognition

and Structural Biology, Institute of Biotechnology/UNAM, Apartado Postal

510-3, Cuernavaca, Morelos 62210 Mexico. Fax: + 52 73 172388,

Tel.: + 52 73 171209, E-mail: possani@ibt.unam.mx

Abbreviations: Boc, tert-butyloxycarbonyl; Fmoc,

fluorenylmethyloxycarbonyl; MALDI-TOF, matrix-assisted laser

desorption ionization-time of flight.

*Note: these two laboratories contributed equally to the work reported

(Received 25 January 1999, revised 24 March 1999, accepted 8 June 1999)

q FEBS 1999 Neutralizing epitope of scorpion toxin Cn2 (Eur. J. Biochem. 264) 747

is also capable of neutralizing 28 LD50 of whole soluble venomfrom C. noxius [21].

In this publication, synthetic peptides are used to identify theneutralizing epitope on Cn2 recognized by BCF2. Previouslywe have used continuous synthetic peptides corresponding tothe sequence of Cn2 to study the immunological response ofmice to synthetic peptides [23]. One of the relevant results ofthis study suggested that the N-terminal region of Cn2 is partof a neutralizing epitope. In this paper, we use the model of thethree-dimensional structure of Cn2 to design in addition a seriesof discontinuous synthetic peptides, with emphasis on regionsnear the N-terminus. We report the results of competitionexperiments of the continuous and discontinuous peptides andnative Cn2 for binding to BCF2. Furthermore, the syntheticpeptides that were shown to be most efficient in binding wereused for immunization of mice and rabbits. The generatedantipeptide sera were tested using in vitro neutralization and invivo protection assays.

M A T E R I A L S A N D M E T H O D S

Source of toxin

The venom from scorpions of the species C. noxius Hoffmannwas extracted by electric stimulation as described [3]. The driedvenom was dissolved in double-distilled water, centrifuged, andthe soluble part was freeze-dried and kept at 220 8C untilused. Cn2 was purified by Sephadex G-50 gel filtration, andvarious cation-exchange chromatographic steps, as describedpreviously [3,18].

Preparation of synthetic peptides

Peptides synthesized are listed in Table 1 and 2. The synthesisof the continuous peptides was described previously [23].Discontinuous peptides were synthesized from the corres-ponding linear peptides either by intramolecular or inter-molecular disulfide bond formation between cysteines.Standard solid-phase methods [24] with tert-butyloxycarbonyl

(Boc)/benzyl strategy were used, except as noted below, on aBiosearch 9500 synthesizer. p-Methylbenzhydrylamine resinwas used to give the C-terminal amides. After HF cleavage anddesalting on Sephadex G-15 in 50% acetic acid, peptides werepurified both before and after disulfide bond formation byreversed phase HPLC on a Vydac C-18 column in gradients of0.1% trifluoroacetic acid in water and acetonitrile with 0.08%trifluoroacetic acid. Composition was assured by amino acidanalysis, purity was ascertained by analytical HPLC andmolecular masses were determined by matrix-assisted laserdesorption ionization-time of flight (MALDI-TOF) massspectrometry.

SP1-SP6 and SP8-SP11 were synthesized from linearpeptides having cysteine protected with the HF-cleavable4-methylbenzyl groups. Heterodimer formation (SP1±SP6,SP8 and SP9) was achieved by air oxidation of the freeSH-fragments for 24 h in 0.1 m NH4HCO3, pH 8.0, followedby acidification with acetic acid and lyophilization. Symmet-rical dimers of each linear peptide also were formed duringthe oxidation. For intramolecular disulfide formation(SP10±SP12), the peptide was dissolved in a minimumof 5% acetic acid and diluted with water to a concentra-tion of 1025 m.

The pH was adjusted to 8.2 with triethylamine. After stirringfor 2 days, the solution was acidified with acetic acid andlyophilized. SP12 was synthesized with the HF-stable acet-amidomethyl on Cys12. The side-chain amino group of Orn3was protected with fluorenylmethyloxycarbonyl (Fmoc). Aftercoupling Boc-Orn(Fmoc), the side-chain sequence was con-structed with Fmoc chemistry employing Fmoc amino acidswith the acid-stable benzyl side-chain protection. The N-terminalof the side-chain sequence was acetyled with acetic anhydride.After deprotection of the ornithine Boc, the synthesis wascompleted using Boc chemistry.

In order to improve the yield of the heterodimer formationand to eliminate formation of symmetrical dimers, SP7, SP13and SP14 were synthesized with 4-methylbenzyl protection onthe Cys12 peptide and the HF-stable 3-nitro-2-pyridinesulfenyl

Table 1. Sequences of continuous peptides tested for their ability to inhibit the formation of the complex Cn2-BCF2 in comparison with Cn2. Single-

letter codes are used for the amino acids. Cysteine residues are underlined. Toxin Cn2 (0.15 mg per well) in 20 mm sodium bicarbonate buffer, pH 9.4 was

immobilized in ELISA plates for 2 h at 378C. After blocking with BSA the wells were incubated for 15 min at 37 8C with 100 mL of a preincubated (1 h at

37 8C) mixture containing 3 mg´mL21 of monoclonal antibody BCF2 and synthetic peptide leading to a final concentration of 1025 m for the peptides and

1028 m for Cn2. BCF2 bound to Cn2 was detected by incubation (1 h at 37 8C) with 50 mL of goat IgG (H + L chains) anti-(mouse IgG). The inhibition

percentage was calculated taking as 100% the absorbance obtained where no competitor was used in solution with BCF2. The data are means of triplicates.

Peptide Sequence

Inhibition of binding

Cn2-BCF2 (%)

5 10 15 20 25 30 35 40 45 50 55 60 65

Cn2 KEGYLVDKNTGCKYECLKLGDNDYCLRECKQQYGKGAGGYCYAFACWCTHLYEQAIVWPLPNKRCS

SP1±10 KEGYLVDKNT 8

SP2±10 EGYLVDKNT 4

SP3±10 GYLVDKNT 8

SP4±10 YLVDKNT 4

SP1±14M K-GYLVDKNTGCKY 18

SP1±14 KEGYLVDKNTGCKY 20

SP4±14 YLVDKNTGCKY 6

SP1±27 KEGYLVDKNTGCKYECLKLGDNDYCLR 0

SP10±27 TGCKYECLKLGDNDYCLR 0

SP17±25 LKLGDNDYC 0

SP22±36 NDYCLRECKQQYGKG 0

SP30±40 KQQYGKGAGGY 0

SP35±48 KGAGGYCYAFACWC 0

748 E. S. Calderon-Aranda et al. (Eur. J. Biochem. 264) q FEBS 1999

group on the Cys65 peptide [25]. After HF treatment andpurification of the linear peptides, the heterodimer was formedby stirring for 5 h in 0.1 m KH2PO4.

The trimeric peptide SP15 was prepared as follows: First,peptide 5±16 was synthesized using Fmoc strategy, protectingCys12 with 4-methylbenzyl and Cys16 with triphenylmethyl(Trt). The peptide was cleaved from the resin with 95%trifluoroacetic acid, 2.5% H2O and 2.5% triisopropylsilaneleaving Cys12 protected and Cys16 as the free SH. The peptide28±41 was synthesized by Boc strategy, with Mpys for Cys41,and HF cleaved. The Cys16±Cys41 disulfide was formed bystirring in 0.1 m KH2PO4 for 3 h followed by HF treatment togive Cys12 as the free SH. This heterodimer was mixed withan equimolar amount of peptide 54±65, which had beensynthesized with Cys65 protected with 3-nitro-2-pyridine-sulfenyl, in 0.1 m KH2PO4 to form the disulfide bond betweenCys12 and Cys65.

Binding inhibition of BCF2 by synthetic peptides in ELISAassay

Polyvinyl plates (Costar, Cambridge, MA, USA) were coatedwith 100 mL of toxin Cn2 in the presence of 20 mm sodiumbicarbonate buffer, pH 9.4 for a given time and at a giventemperature (see figure legends). The wells were then blockedwith 1% BSA in 20 mm sodium phosphate buffer, pH 7.8,0.15 m sodium chloride (NaCl/Pi solution) and incubated for1 h at 37 8C. The wells were washed with NaCl/Pi containing0.1% Tween-20 (washing solution) and incubated for a giventime, at the indicated temperature (see figure legends) with100 mL of a preincubated mixture containing 3 mg´mL21 ofBCF2 plus either Cn2 or synthetic peptides at the indicatedconcentrations (1 nm to 1 mm). The preincubation wasperformed in NaCl/Pi containing 0.05% Tween-20 and 0.1%BSA for a time given in the figure legends. Afterwards, theplates were washed six times with washing solution. BCF2bound to Cn2 was detected by incubation (1 h at 37 8C) with 50

or 100 mL (see figure legends) of goat IgG (H + L chains) anti-(mouse IgG) coupled to horse radish peroxidase (dilution1 : 3000). The reaction was developed with o-phenylenedi-amine and H2O2 as substrates. The reaction was stopped by theaddition of 8 m sulfuric acid and the absorbance read at 490 nmin an ELISA reader model 1550 EIA (Bio-Rad. Lab.,Richmond, CA, USA).

Lethality tests

Lethality tests were conducted as described previously [23].Experiments were carried out in accordance with the USNational Research Council's Guide for the care and use ofLaboratory Animals.

Immunization of mice

Albino mice of the strain CD1 (4 to 8-week-old, weighing 15±18 g)were used for immunization with synthetic peptides and Cn2.Cn2 was used for immunization according to a protocolpreviously described by our group [19]. In the case of thesynthetic peptides, the first injection consisted of 50 mg ofpeptide emulsified with complete Freund's adjuvant. Threeadditional immunizations were conducted at intervals of2 weeks in the presence of incomplete Freund's adjuvant.Groups of 18 animals were used for immunization protocols.All immunized mice were bled 9 days after the thirdimmunization for serum collection. Sera were kept at 220 8Cuntil used for titration or for passive neutralization experiments.Animals that had completed the fourth immunization treatmentwere used for challenging experiments with native toxin.

Immunization of rabbits

Rabbits (female strain New Zealand, 2.0 kg weight at the startof the immunization protocol), were immunized four times atintervals of 10 days to obtain antisera against the syntheticpeptides and Cn2. The synthetic peptides were used at 25 mg

Table 2. Discontinuous peptides. Single-letter codes are used for the amino acids except for ornithine (Orn). Ac, acetylated amino group. Inhibition was

tested up to 3 � 1024 m. ±, no effect. SS, disulfide bridge.

Peptide Inhibition of binding Cn2-BCF2

SP1 [Cn2 (1±14)NH2]-C12-SS-C65-[Ac-Cn2(52±66)] IC50 70.0 mm (R = 0.987)

SP2 [Cn2(1±14)NH2]-C12-SS-C12-[Cn2(1±14)NH2] ±

SP3 [Ac-Cn2(52±66)]-C65-SS-C65-[Ac-Cn2(52±66)] ±

SP4 [Cn2(1±14)NH2]-C12-SS-C65-[Ac-Cn2(56±66)] IC50 78.0 mm (R = 0.987)

SP5 [Cn2(1±14)NH2]-C12-SS-C65-[Ac-Cn2(62±66)] ±

SP6 [Cn2(1±12)NH2]-C12-SS-C65-[Ac-Cn2(52±66)] ±

SP7 [Cn2(3±15)NH2]-C12-SS-C65-[Ac-Cn2(54±66)NH2] IC50 5.1 mm (R = 0.982)

SP8 [Cn2(1±14)NH2]-C12-SS-C65-[Ac-G59-Cn2(52±66)] ±

SP9 [Cn2(1±14)NH2]-C12-SS-C65-[Ac-A46,48-G59-Cn2(45±66)] ±

SP10 Ac-Cn2-(15±7)- - - - - - - - - - - - - - - - - (60±66) ±

|_______C12_SS_C65______|SP11 G38G39CY40GY42GW47GT49H50L51Y52Q54E53K1E2G3Y4C-NH2

|_____________________SS_______________________|_____________SS_________ ±| |

SP12 AcCY40SY42A43F44A45SW47C ±_______________|

K1E2Orn-[C12(Acm)-Cn2(4±14)NH2]

SP13 [Ac-Cn2(5±14)NH2]-C12-SS-C65-[Ac-Cn2(33,34)-GGGG-Cn2(54±65)NH2] IC50 21.0 mm (R = 0.988)

SP14 [Ac-Cn2(5±14)NH2]-C12-SS-C65-[Ac-A29-Cn2(28±34)-GGGG-Cn2(54±65)NH2] IC50 10.9 mm (R = 0.995)

SP15 [Ac-Cn2(5±16)NH2]-C16-SS-C41-[Ac-A29-Cn2(28±41)NH2]-C12-SS-C65±[Ac-Cn2(54±65)NH2] IC50 16.8mM (R = 0.988)

q FEBS 1999 Neutralizing epitope of scorpion toxin Cn2 (Eur. J. Biochem. 264) 749

per dose with Freund's adjuvant the first time, and incompleteadjuvant for subsequent applications. Cn2 was used at 7.5 mgper rabbit as described previously [23].

Titration of antisera to synthetic peptides

Sera obtained from mice and rabbits pre-immunized withsynthetic peptides or Cn2, were assayed on ELISA platescoated with antigens. Plates were coated with 0.15 mg´mL21 ofantigens in 20 mm sodium bicarbonate buffer, pH 9.4 for 2 h at37 8C, subsequently blocked with 1% BSA in NaCl/Pi during1 h at 37 8C and washed four times. After a 2-h incubation withserial dilutions of antiserum at 37 8C, the wells were washedwith washing solution. The concentration of attached antibodywas measured with a second antibody, goat IgG anti-(mouseIgG), coupled to horse radish peroxidase (dilution 1 : 3000) asdescribed above. The dilutions corresponding to 50% of themaximum absorption measured were calculated from linearregressions of the data from the linear part of the titrationcurves.

In vitro neutralization assays

The sera from immunized rabbits and mice were used forneutralization assays using groups of 8±10 naive CD1 mice.Known volumes of anti-(synthetic peptide) sera and anti-Cn2sera were preincubated for 1 h with different concentrations ofCn2 at 37 8C with agitation, and then injected intraperitoneallyor subcutaneously into mice (total volume of the mixture wasadjusted to 300 mL per animal). A control group of 10 micewas also injected with a nonimmune serum, preincubated withone LD50 of Cn2, using the same conditions. Survival rate wasobserved after 24 h of assay.

In vivo protection assays

All mice immunized with synthetic peptides or Cn2 werechallenged 9 days after the fourth immunization by intra-peritoneal injection of one LD50 of Cn2 [0.25 mg´20 g mouseweight)21 of strain CD1]. Groups of 10 nonimmunized micewere used in parallel as controls. The survival rate was recorded24 h after the challenge.

R E S U LT S A N D D I S C U S S I O N

Binding inhibition of BCF2 to Cn2 by the addition of single-chain synthetic peptides

As a first approach to identify the epitope on Cn2 that isrecognized by the neutralizing antibody BCF2, we tested aseries of continuous linear peptides. The primary structures ofthese peptides and the results of experiments testing the abilityof the peptides to inhibit the binding of BCF2 to Cn2immobilized on ELISA plates are shown in Table 1. Peptidescorresponding to the N-terminal segment of Cn2 are partiallyeffective in displacing the binding of BCF2. Thus, these couldbe part of the epitope recognized by BCF2. The best result wasachieved with SP1±14 which resulted in an inhibition of 20% ata concentration of 1025 m. This peptide might therefore containimportant `functional' residues (residues that contribute ener-getically to the affinity between the antigen and the antibody byestablishing interactions [26]). The fact that SP4±10 andSP1±10 displace the binding of BCF2 less efficiently could bedue to the need for additional residues, present in SP1±14, forthe proper conformation of the epitope. SP1±14M does not

contain residue Glu2 but gave almost the same inhibition asSP1±14, which indicates that residue 2 is probably not directlyinvolved in the binding of BCF2. The lack of displacement bySP1±27, which contains the essential sequence 1±14, mightalso be explained by a larger number of alternative structuresthat do not present the epitope in its native conformation thusdiminishing the effective concentration of the peptide.

The very large difference in affinity between the activecontinuous peptides (20% inhibition at a concentration of1025 m) and Cn2 (IC50 = 2.8 � 1029 m) might be caused fortwo reasons: (a) the epitope is continuous but the percentage ofpeptide molecules that adopt the correct conformation of theepitope is very small ± the linear peptide chain does notprovide enough stabilizing interactions to make the correctconformation energetically favorable or (b) the epitoperecognized by BCF2 is conformational, i.e. requires morethan one segment of the primary sequence, in close proximity,in the three-dimensional structure of Cn2.

Discontinuous peptides

Based on the above described results, we designed two series ofdimeric and trimeric peptides using the N-terminal segment ofCn2 as a fixed point. The structures of these peptides are givenin Table 2. As an essential tool we used an existing model ofthe three-dimensional structure of Cn2 [27] which had beengenerated using the highly homologous toxin CsEv3 fromCentruroides sculpturatus Ewing as a template. Figure 1Ashows a ribbon diagram of Cn2, which contains a three-stranded antiparallel b-sheet and a short stretch of a-helix. Thestretch of SP1±14 is highlighted in red. It comprises the firststrand of the b-sheet (residues 2±6) and a turn structure(residues 7±10).

A striking feature of the three-dimensional model is the closeproximity of the C-terminus of Cn2 to the N-terminus, heldtogether by the disulfide bond Cys12±Cys65. The first series ofdimeric peptides, SP1 to SP10, explores the importance of theresidues in these two regions. The first dimer synthesized wasSP1, which includes sequence stretches 1±14 and 52±66. In apreliminary test using the conditions given in Table 1, thispeptide inhibited binding to Cn2 by 44% (result not shown)indicating that this hetero-dimeric peptide presents higheraffinity than the continuous peptide SP1±14. During theprocess of synthesizing the peptides, it was shown that theC-terminus of native Cn2 is amidated [28]; therefore, some ofthe peptides were synthesized with an amidated C-terminus.SP8 was prepared for testing the importance of Pro59(highlighted in Fig. 1A) for the correct conformation of theC-terminal segment. In the three-dimensional model of Cn2(using as template the parameters of CsEv3) [14,16], Pro59 wasidentified as being in a cis-conformation, forming an a-acontact with residue 58. The actual structure of Cn2 obtainedby NMR (very recently published [29]) supports our initialassumption. The design of SP9 was to test for the ability of theneighboring segment 45±51 to restore a putatively stabilizingrole of Pro59. SP10 is a cyclic dipeptide containing thesequence stretch 7±15 of the N-terminus in reverse direction,that is from amino acid 15 to 7. Residue Asp7 is followeddirectly by Leu60; in the structural model they are situated8.36 AÊ (Ca±Ca) apart. The cycle is closed on the other side byCys12±Cys65. In this way it was thought to restrict theflexibility of the N- and C-terminal sequence stretchesemployed as was shown successfully for peptide mimetics [30].

The second series of peptides, SP11 to SP15 (Table 2) wasdesigned to explore regions in addition to the N-terminus and/or

750 E. S. Calderon-Aranda et al. (Eur. J. Biochem. 264) q FEBS 1999

the C-terminus that might be involved in the epitope, based onthe three-dimensional structure. SP11 and SP12 are cyclicpeptides that incorporate N-terminal residues with the sequencestretch 38±58, which comprises strands 2 and 3 of the b-sheet.Figure 1B illustrates the position of these residues on thesurface of Cn2. The region includes a hydrophobic patch

conserved in both a-toxins and b-toxins long considered to beinvolved in receptor binding [31,32], namely residues Tyr4 (nothighlighted), Tyr40, Tyr42, and Trp47. As BCF2 is aneutralizing antibody, it was presumed to recognize at leastpart of the channel-binding region of Cn2. SP11 was designedin a way that it could simulate the conformation of the residues

Fig. 1. Three-dimensional model of Cn2.

(A) Ribbon diagram illustrating secondary

structure elements: a-helix in white, b-sheet in

light brown, turns in grey. Disulfide bonds are

shown in yellow. The sequence stretch 1±14 is

shown in red. Left, `front view' of model.

Right, the same model turned 908 to the right.

(B) Design of peptides SP11 and SP12. Included

are N-terminal residues 1±4 (SP11) or 1±14

(SP12), shown in red; residues 40, 42 and 47 of

the hydrophobic patch (SP12 includes Phe44),

shown in dark grey, and the sequence stretch

49±53 (only included in SP11) in light grey.

(C) Design of peptides SP13 to SP15. The

N-terminal stretch 5±14(5±16 in SP15) shown in

red, the C-terminal 54±65 in light grey. The

additional region included comprises residues

Tyr33,Gly34 (SP13), 28±34 (SP14) or 28±41

(SP15). In SP13 and SP14 these residues are

connected to the C-terminal peptide by a bridge

of four glycine residues between Gly34 and

Gln54 (shown in dark grey). In SP15 they are

connected via the disulfide bridge C16±C41

(Cys16 in yellow). Modelling performed as

described in [27].

q FEBS 1999 Neutralizing epitope of scorpion toxin Cn2 (Eur. J. Biochem. 264) 751

in question (N-terminus 1±4, sequences stretches 38±42,Trp47, 49±53). Their order in the sequence corresponds totheir arrangement on the surface of Cn2 (Fig. 1B). SP12arranges the hydrophobic-patch residues in a cyclic peptidewhich is connected via the side chain of an ornithine residue,Orn3, to the N-terminal sequence 1±14 (Fig. 1B). The rationaleof the design of peptides SP13 to SP15 is the following: thestarting point was a sequence comparison between Cn2 andother toxins (Cn1 [28], Cn3 [18], Cn4 [33] and Clt1 [34]) thatwere not able to inhibit the binding of Cn2 to BCF2 [35]. Lys8was identified as unique in the N-terminal segment of Cn2 andit was assigned as the central residue of the putative functionalepitope. Using the graphic display program insight ii (Biosym/MSI, San Diego, CA, USA), a subset of residues was chosenthat were situated within a circle of radius 9 AÊ , taken from theCa atom of Lys8, and with a surface accessibility of above 50%as compared to the free amino acid. The subset comprisesAsn9, Ile56, Asn62, Tyr33, Gly34, Gln54, Ala55, Pro61,Arg64, Val57 and Pro59. Also included was the loop 31±35(Fig. 1C, Tyr 33 and Gly 34 are highlighted). The nearestdistance (8.4 AÊ from Ca to Ca) between the loop residueGly34 and the C-terminus Gln54 was bridged by the insertionof a flexible segment of four glycines (the distance between thenitrogen atom of the amino group of the first Gly and thecarbon atom of the carbonyl group of the last Gly was estimatedon an extended model chain to be 13.8 AÊ ). SP13 included onlythe essential residues 33 and 34 whereas SP14 included thelonger segment of residues 28±34. Finally, SP15 is a trimerderived from the actual sequence, containing the threesegments 5±16 (N-terminus), 28±41 (loop) and 54±65 (C-terminus), connected by the native disulfide bridges Cys12±Cys65 and Cys16±Cys41.

Figure 2 shows the inhibition curves for the peptides thatwere able to inhibit the binding of BCF2 to Cn2. Thecontinuous inhibition curves were generated fitting the data tologistic dose±response functions according to [36]. Thecalculated IC50 values are shown in Table 2. The controlsSP2 and SP3, homodimers of the N- and C-terminus,respectively, did not inhibit up to concentrations of3 � 1024 m, nor did peptides SP5, SP6, SP8±12 (not shown).Control peptide SP2 was expected to inhibit at concentrationshigher than that of SP1. Its inability to inhibit indicates either:(a) a larger conformational space for the dimeric peptide thatlowers the effective concentration of the correctly folded N-terminus compared to the continuous peptide 1±14; or (b) morestringent assay conditions for this second series of experiments.

The following conclusions can be drawn from the aboveresults. The epitope recognized by BCF2 includes residuessituated near the N- and the C-termini of Cn2. The peptide SP7with the highest affinity was composed of segments 3±15 and54±66. Lys1 and Glu2 as well as Tyr52 and Glu53 (compareSP1 and SP7) are not likely to be functional residues, nor areGln54 and Ala55 (compare SP1 and SP4). Glu15 might beresponsible for the higher affinity of SP7, although it carries anegative charge and the overall charge of the paratope of BCF2is negative (unpublished results). Lys13 and Tyr14 (compareSP1 and SP6) might be functional residues or contribute to thestabilization of the correct conformation of other functionalresidues. The same applies to residues 56±61 (compare SP1and SP5). The replacement of the cis-Pro59 resulted in anonfunctional peptide, showing that either Pro59 is necessary tomaintain the conformation of the C-terminal segment or itmight be a functional residue. The addition of the stretch 45±51was not able to restore the correct conformation. The cyclicpeptide SP10 did not show affinity towards BCF2; although itcontains residues shown to be important. There might be tworeasons: (a) residues 3±6 and 56±59 are involved in binding(compare negative results for SP5); or (b) the N-terminalpeptide that was synthesized in the opposite direction (residues15±7) might not be able to mimic the correct conformationalarrangement of the functional residues. This was observed forsynthetic peptides tested as antagonists of granulocyte macro-phage colony stimulating factor; only the one which keeps theN-C direction of the residues as it is in the protein resulted inactivity [30].

SP11 and SP12 did not show affinity within the testedconcentration range. SP11 contains only residues 1±4 of theN-terminus, which might be a reason for its lack of binding.SP12, with N-terminal residues 1±14, would be expected toshow some affinity, although lower than SP1. The reason for thelack of any affinity might be as discussed for SP2 (see above).The results indicate that the residues of the hydrophobic patchmay not be involved in the epitope. There are other recentindications that the hydrophobic patch might not be the regionbinding to the channel. Firstly, it was shown that the hydro-phobic patch in b-toxins is situated in a region with negativeelectrostatic surface potential [13]. Secondly, mutationalexperiments on a-toxins demonstrated that the region thatbinds to the sodium channel does not include the hydrophobicpatch [37]. The b-toxins bind to receptor site 4 whereasa-toxins bind to site 3 of the channel [38,39]. Nevertheless, thestriking similarity of the three-dimensional fold of a-toxins andb-toxins justifies the supposition that the receptor-binding sitesof both might be localized within the same surface region.

The designed peptides (SP13, SP14, SP15) that comprisethree peptide segments (the N-terminus, the loop made byTyr33-Gly34, and the C-terminus) resulted in a slight

Fig. 2. Binding inhibition of BCF2 to Cn2 by addition of synthetic

discontinuous peptides. Toxin Cn2 (0.3 mg per well in 20 mm sodium

bicarbonate buffer, pH9.4 was immobilized in ELISA plates for 24 h at

4 8C. After blocking with BSA the wells were incubated for 10 min at room

temperature with 100 mL of a preincubated (overnight at 4 8C) mixture

containing 3 mg´mL21 of monoclonal antibody BCF2 and the concentration

of synthetic peptides or Cn2 resulting in final concentrations given on the

x-axis. BCF2 bound to Cn2 was detected by incubation (1 h at 37 8C) with

100 mL of goat IgG (H + L chains) anti-(mouse IgG). The data are means

of triplicates for Cn2 and means of duplicates for the synthetic peptides.

The continuous curves were constructed according to DeLean et al. [36].

Error bars represent SD.

752 E. S. Calderon-Aranda et al. (Eur. J. Biochem. 264) q FEBS 1999

improvement of the affinity compared to the dimeric peptidesSP1 and SP4 but not to SP7. Residues Tyr33 and Gly34 (SP13)or 28±34 (SP14) might be included in the functional epitopebut the conformation they adopt in Cn2 may not be simulatedvery well by the peptides. The N-terminal residues Gly3 andTyr4 and the C-terminal residue Ser66 do not seem to befunctional residues.

In conclusion, it can be stated that segments 5±14 (possiblyto 15) and 56±65 contain important functional residues thatestablish interactions with the paratope of BCF2. Theconsiderable difference in affinity between peptides containingthese segments and Cn2 might be caused by missing functionalresidues or conformational instability, i.e. flexibility of thepeptides. The solution structure of the dimeric peptide [Cn2(1±15)NH2]-Cys12-Cys65-[Ac-Cn2(52±66)NH2] was recentlydetermined by NMR spectroscopy [40]. It was shown that theN-terminal segment Val6 to Tyr14 presented a fixed confor-mation and that the turn region between residues Asp7 andGly11 was particularly well-defined. It was concluded that thisregion, having a well-conserved conformation in toxins fromCentruroides scorpions but showing high sequence variability,might be involved in binding to the sodium channel. Thebinding to this structural element or at least to a region in itsvicinity might be the reason for the neutralizing capacity ofBCF2. The recognition of turn structures has been found invarious peptide±antibody complexes [41±43] and in structuresof antigenic peptides [44,45].

Antisera against synthetic peptides

The synthetic dimeric peptides that demonstrated the bestcapacity to bind to BCF2 (peptides SP1 and SP7) were used todetermine their ability to induce neutralizing antibodies inrabbits and mice. The sera obtained were analyzed for theirreactivity against the antigen, and in addition, their crossreactivity to Cn2. Moreover, we performed in vitro

neutralization assays using anti-SP1 and anti-SP7 sera fromimmunized rabbits to protect mice challenged with toxin Cn2,as well as in vivo protection assays of immunized mice.

Figure 3 shows the titration of sera against SP1 and SP7 fromrabbits and mice. The recognition of the homologous antigen(Fig. 3A,C) and of the toxin Cn2 (Fig. 3B,D) are shown. Thepeptides were used as test antigens in order to analyze theirimmunogenic capacity, whereas the assay of cross-reactivitywith toxin Cn2 was used to evaluate the presence of epitopes onthe surface of the toxin recognized by the anti-peptideantibodies. As a control we tested the recognition of thepeptides and of Cn2 by rabbit anti-Cn2 serum. Results of theassay of the anti-SP1 sera showed that SP1 was a very efficientimmunogen (Fig. 3A,B). It was able to induce not onlyantibodies recognizing the peptide itself (dilution of rabbitserum at 50% of maximal A492nm was 1 : 3000), but alsoantibodies showing high cross-reactivity with Cn2 (dilution ofrabbit serum at 50% of A492nm was 1 : 1300). Results fromassays using antibodies induced by peptide SP7 are shown inFig. 3C,D. Fifty percent maximal absorbance against homo-logous antigen and Cn2 were reached at dilution of rabbit seraof 1 : 800 and 1 : 500, respectively. The results show that thesynthetic peptides SP1 and SP7 constitute a structural antigenicdeterminant exposed at the surface of Cn2. The observeddifference in the immunogenicity of the peptides could be dueto various reasons: (a) the difference in the length of thepeptides; (b) the fact that individual test animals can present arather different immune response to the same antigen [46]; and(c) the efficiency of immobilization of the peptides in theELISA plates might be different.

The analysis of the sera from mice immunized with SP1 andSP7 gave results similar to the anti-peptide sera from rabbits(higher immunogenicity of SP1), however, the reactivity waslower against both homologous antigens and toxin Cn2. Thissuggests that the immunogenicity of both peptides is notvery high in CD1-strain mice. In order to improve the

Fig. 3. ELISA titration of anti-peptide sera.

(A) The ELISA plate was coated with synthetic

peptide SP1. The antisera tested were: anti-Cn2

rabbit serum (circles), anti-SP1 mouse serum

(triangles) and anti-SP1 rabbit serum (squares).

(B) The ELISA plate was coated with native

toxin Cn2. The antisera tested are the same as

shown in A. (C) the ELISA plate was coated with

synthetic peptide SP7. The antisera tested were:

anti-Cn2 rabbit serum (circles), anti-SP7 mouse

serum (triangles) and anti-SP7 rabbit serum

(squares). (D) The ELISA plate was coated with

Cn2 native toxin. The antisera tested are the same

as shown in C. Values are means of triplicates,

and standard deviation values were always below

10% of the mean value. The absorbance of

preimmune sera (controls) were subtracted.

q FEBS 1999 Neutralizing epitope of scorpion toxin Cn2 (Eur. J. Biochem. 264) 753

immunogenicity of SP1 we tested SP1 associated with apromiscuous T-cell epitope mapped to tetanus toxoid [47] onCD1-strain mice. Preliminary results suggest that it is not moreeffective (data not shown). It will be necessary to comparedifferent mouse strains in order to evaluate if the low responsein the strain used in this work is a consequence of a particularlylow recognition by the major histocompatibility complexmolecules present in this strain, and if other strains with adifferent haplotype are better responders to antigens SP1 andSP7. A similar difference between the immunogenicity of asynthetic peptide derived from the scorpion toxin AaHII inrabbits and mice was reported by Bahraoui et al. [48] andAit-Amara et al. [49].

In vitro neutralization

In order to determine if SP1 and SP7 constitute neutralizingepitopes, the protection capacities of rabbit anti-SP1 and anti-SP7 sera were evaluated. Increasing volumes of each one weremixed with different concentrations of toxin Cn2, injected intonaive mice, and the LD50 of toxin Cn2 in the presence of serawas determined. As shown in Table 3, anti-SP1 serumneutralized 8.3 mg Cn2 per mL serum (intraperitoneal route);this corresponds to about 39.5 LD50 (0.25 mg Cn2 per 20 g).The neutralizing capacity of rabbit anti-SP1 serum is similar tothat of rabbit anti-Cn2 serum (8.5 mg toxin per mL serum).However, a difference was observed with respect to theappearance and intensity of intoxication symptoms developedwhen anti-Cn2 or anti-SP1 sera were used. No symptoms wereobserved using anti-Cn2 serum whereas when anti-SP1 serumwas used, the test animals presented intoxication symptoms.

Nevertheless, the symptoms were presented relatively late withrespect to the control animals. The intensity of envenomationwas moderate and the final survival was high. The apparentcontradiction between the low affinity of SP1 to BCF2 and thehigh neutralizing capacity of anti-SP1 serum could beexplained by the generation of antibodies that recognizeepitopes which overlap with the epitope of BCF2. The assayof anti-SP7 serum showed no protection when it was tested bythe intraperitoneal route. However, when the subcutaneousroute was used, the neutralizing capacity of 3.2 mg Cn2 per mLserum was determined. The LD50 of Cn2 injected subcuta-neously was determined to be 0.6 mg per 20 g body weight ofCD1 mice. Thus, 1 mL of anti-SP7 neutralizes around 6.5 LD50

of Cn2 when injected subcutaneously. The results suggest thatthe peptide SP7 generated either fewer antibodies able torecognize Cn2 or antibodies with low affinity towards Cn2. Thereason might be its higher flexibility compared with SP1 (seeabove) that impedes the effective generation of neutralizingantibodies. Using the subcutaneous route anti-SP7 serum wasable to neutralize the effect of Cn2. This can be explained bydifferences in pharmacokinetics compared to the intraperitonealroute due to a longer resorption phase to the vascular space[50].

In vivo protection assays

In order to evaluate the capacity of the synthetic peptides SP1and SP7 to induce active protection against toxin Cn2, weconducted in vivo challenges of mice grouped as following:mice immunized with (1) SP1 (2) SP7 (3) Cn2 and (4)nonimmunized mice. The last two groups were used as positiveand negative control, respectively. The four groups describedabove were challenged using one LD50 of Cn2 (0.25 mg per20 g in CD1 mice) injected via the intraperitoneal route. Theresults are shown in Table 3. In the group immunized with Cn2the survival obtained was 100%, confirming results previouslyobtained [23]. Control (nonimmunized) mice showed a survivalof 50% as expected. The group of mice immunized with SP1and SP7 showed no protective effects (survival rate were 45%and 44%, respectively), despite the fact that the sera hadpresented cross reactivity against Cn2 (Fig. 3B,D). However,the synthetic peptides did not induce sensitization of theanimals to the toxic effect of Cn2 as was previously reported byusing sequential peptides to induce antibodies in mice [23]. Thein vivo protection results show, once more, that the neutral-ization phenomenon is not a simple process dependingexclusively on the affinity of the antibody to the toxin. Manyother factors might be important, such as the presence of anti-idiotype antibodies, liberation of endogenous cytokines,competition of multiple antibodies recognizing differentepitopes and differences in the distribution of toxin andantibodies in various body compartments [51,52]. These factorsmay explain the discrepancy between the results obtained in thein vivo protection assays versus the binding inhibition of BCF2and the in vitro assays.

In conclusion, the nonsensitization of mice immunizedwith SP1 and SP7 as well as the passive protection effect,particularly by the serum induced by SP1, encourage us tocontinue the evaluation of these discontinuous syntheticpeptides as immunogens to improve antiscorpion serapreparation.

A C K N O W L E D G E M E N T S

This work was supported partially by grants from the United-States-Mexico

Foundation for Science no. 56-CH-166 to J. S. and L. D. P., and Howard

Table 3. In vitro neutralization and in vivo protection against the toxic

effect of Cn2, by anti-SP1 and anti-SP7 sera. ND, not determined.

In vitro neutralization Neutralizing capacity

(mg Cn2 neutralized per mL of serum)

Rabbit seruma against Intraperitonealb Subcutaneousc

Toxin Cn2 8�.5 ND

SP1 8�.3 ND

SP7 0 3�.2

Control 0 0

In vivo protection In vivo protection assays

Mice immunized withd (Alive/total) (% Survival)

Toxin Cn2 10/10 100

SP1 9/20 45

SP7 8/18 44

Control 5/10 50

a Rabbits immunized with synthetic peptides SP1, SP7 or Cn2 were bled

after the fourth immunization. Serum was recovered and kept at 220 8C

until use. Neutralizing capacity determinations were performed in CD1

mice (15±18 g). The mixtures of serum and Cn2 were incubated 1 h at

room temperature under agitation before injection. Survival was recorded

24 h after challenge. Nonimmune serum was used as control. b Intra-

peritoneal route was used to challenge with Cn2 toxin mixed with anti-

peptide serum. c Subcutaneous injection was used to challenge with Cn2

toxin mixed with antiserum. d Mice immunized with synthetic peptides or

toxin Cn2 were challenged 9 days after the fourth immunization using 1

LD50 of Cn2 toxin (0.25 g per 20 g) injected intraperitoneally. Naive mice

were used as control. Survival rate was recorded 24 h after challenge.

754 E. S. Calderon-Aranda et al. (Eur. J. Biochem. 264) q FEBS 1999

Hughes Medical Institute (75197±527107) to L. D. P. A scholarship from

the Alexander-von-Humboldt Foundation, Germany was awarded to B. S.,

who also received support from Consejo Nacional Ciencia y Tecnologia,

MeÂxico (3702-N9607). We greatly appreciate the technical assistance of

Fernando Zamudio, Timoteo Olamendi and Elizabeth Mata. The English

corrections made by Shirley Ainsworth is also acknowledged.

R E F E R E N C E S

1. Dehesa-DaÂvila, M. & Possani, L.D. (1994) Scorpionism and sero-

therapy in Mexico. Toxicon 32, 1015±1018.

2. Calderon-Aranda, E.S., Dehesa-Davila, M., Chavez-Haro, A. &

Possani, L.D. (1996) Scorpion stings and their treatment in Mexico.

In Envenomings and Their Treatments (Bon, C. & Goyffon, M., eds),

pp. 311±320. Fondation Marcel Merieux, Lyon, France.

3. Possani, L.D. (1984) Structure of scorpion toxins. In Handbook of

Natural Toxins (Tu, A.T., ed.), Vol. 2, pp. 513±550. Marcel Dekker,

New York.

4. Gordon, D., Savarin, P., Gurevitz, M. & Zinn-Justin, S. (1998)

Functional anatomy of scorpion toxins affecting sodium channels.

J. Toxicol.-Toxin Rev. 17, 131±159.

5. Freire-Maia, L. & Campos, J.A. (1989) Pathophysiology and

treatment of scorpion poisoning. In Natural Toxins: Characterization,

Pharmacology and Therapeutics (Ownby, C.L., & Odell, G.V., eds),

pp. 139±159. Pergamon Press, New York.

6. Gueron, M., Margulis, G., Ilia, R. & Sofer, S. (1993) The management

of scorpion envenomation. Toxicon 31, 1071±1076.

7. Ismail, M. (1995) The scorpion envenoming syndrome. Toxicon 33,

825±858.

8. Carbone, E., Wanke, E., Prestipino, G., Possani, L.D. & Maelicke, A.

(1982) Selective blockage of voltage-dependent K+ channels by a

novel scorpion toxin. Nature 296, 90±91.

9. GarcõÂa, M.L., Hanner, M., Knaus, H.-G., Koch, R., Schmalhofer, W.,

Slaughter, R.S. & Kaczorswki, G.J. (1997) Pharmacology of

potassium channels. Adv. Pharmacol. 19, 425±471.

10. Possani, L.D., Selisko, B. & Gurrola, G. (1999) Structure and function

of scorpion toxins affecting K+-channels. In Perspectives in Drug

Discovery and Design (Darleon, H. & Sabatier, J.M., eds), pp.

15±40. Kluwer Academic Publishers, Dordrecht, the Netherlands.

11. Valdivia, H.H. & Possani, L.D. (1998) Peptide toxins as probes of

ryanodine receptor structure and function. Trends Cardiovasc. Med.

8, 111±118.

12. DeBin, J.A., Maggio, J.E. & Strichartz, G.R. (1993) Purification and

characterization of chlorotoxin, a chloride channel ligand from the

venom of the scorpion Leiurus quinquestriatus. Am. J. Physiol. 264,

C361±C369.

13. Selisko, B., Garcia, C., Becerril, B., Delepierre, M. & Possani, L.D.

(1996) An insect-specific toxin from Centruroides noxius Hoffmann

± cDNA, primary structure, three-dimensional model and electro-

static surface potentials in comparison with other toxin variants. Eur.

J. Biochem. 242, 235±242.

14. Zhao, B., Carson, M., Ealick, S.E. & Bugg, C.E. (1992) structure of

scorpion toxin variant-3 at 1.2 AÊ resolution. J. Mol. Biol. 227,

239±252.

15. Lebreton, F., Delepierre, M., Ramirez, A.N., Balderas, C. & Possani,

L.D. (1994) Primary and NMR three-dimensional structure

determination of a novel crustacean toxin from the venom of the

scorpion Centruroides limpidus limpidus Karsch. Biochemistry 33,

11135±11149.

16. Lee, W., Moore, C.H., Watt, D.D. & Rama Krishna, N. (1994a)

Solution structure of the variant-3 neurotoxin from Centruroides

sculpturatus Ewing. Eur. J. Biochem. 218, 89±95.

17. Lee, W., Jablonsky, M.J., Watt, D.D. & Rama Krishna, N. (1994b)

Proton nuclear magnetic resonance and distance geometry / simulated

annealing studies on the variant-1 neurotoxin from the new world

scorpion Centruroides sculpturatus Ewing. Biochemistry 33,

2468±2475.

18. Zamudio, F., Saavedra, R., Martin, B.M., Gurrola-Briones, G., Herion,

P. & Possani, L.D. (1992) Amino acid sequence and immunological

characterization with monoclonal antibodies of two toxins from the

venom of the scorpion Centruroides noxius Hoffmann. Eur. J.

Biochem. 204, 281±292.

19. Calderon-Aranda, E.S., Hozbor, D. & Possani, L.D. (1993) Neutraliz-

ing capacity of murine sera induced by different antigens of scorpion

venom. Toxicon 31, 327±337.

20. Bouhaouala-Zahar, B., Ducancel, F., Zenouaki, I., Ben Khalifa, R.,

Borchani, L., Pelhate, M., Boulain, J.-C., El Ayeb, M. & Karoui, H.

(1996) A. recombinant insect-specific alpha-toxin of Buthus

occitanus tunetanus scorpion confers protection against mammal

toxins. Eur. J. Biochem. 238, 653±660.

21. Licea, A.F., Becerril, B. & Possani, L.D. (1996) Fab fragments of the

monoclonal antibody BCF2 are able of neutralizing the whole soluble

venom from the scorpion Centruroides noxius Hoffmann. Toxicon 34,

843±847.

22. Dehesa-Davila, M., Ramirez, A.N., Zamudio, F.Z., Gurrola-Briones,

G., Lievano, A., Darszon, A. & Possani, L.D. (1996) Structural and

functional comparison of toxins from the venom of the scorpions

Centruroides infamatus infamatus, Centruroides limpidus limpidus

and Centruroides noxius. Comp. Biochem. Physiol. 116B, 315±322.

23. Calderon-Aranda, E.S., Olamendi-Portugal, T. & Possani, L.D. (1995)

The use of synthetic peptides can be a misleading approach to

generate vaccines against scorpion toxins. Vaccine 13, 1198±1206.

24. Stewart, J.M. & Young, J.D. (1984) Solid Phase Peptide Synthesis.

Pierce Chemical Co., Rockford, USA.

25. Bernatowicz, M.S., Matsueda, R. & Matseuda, G.R. (1986) Preparation

of Boc-[S-(3nitro-2-pyridinesulfenyl)]-cysteine and its use for

unsymmetrical disulfide bond formation. Int. J. Peptide Protein

Res. 28, 107±112.

26. Jin, L., Fendly, B.M. & Wells, J.A. (1992) High resolution

functional analysis of antibody±antigen interactions. J. Mol.

Biol. 226, 851±865.

27. Gurrola, G.B., Moreno-Hagelsieb, G., Zamudio, F.Z., GarcõÂa, M.,

Soberon, X. & Possani, L.D. (1994) The disulfide bridges of toxin 2

from the scorpion Centruroides noxius Hoffmann and its three-

dimensional structure calculated using the coordinates of variant 3

from Centruroides sculpturatus. FEBS Lett 347, 59±62.

28. Vazquez, A., Tapia, J.V., Eliason, W.K., Martin, B.M., Lebreton, F.,

Delepierre, M., Possani, L.D. & Becerril, B. (1995) Cloning and

characterization of the cDNAs encoding Na+ channel-specific toxins

1 and 2 of the scorpion Centruroides noxius Hoffmann. Toxicon 33,

1161±1170.

29. Pintar, A., Possani, L.D. & Delepierre, M. (1999) Solution structure of

toxin 2 from Centruroides noxius Hoffmann, a b scorpion neurotoxin

acting on sodium channels. J. Mol. Biol. 287, 359±367.

30. Monfardini, C., Kieber-Emmons, T., Voet, D., Godillot, A.P., Weiner,

D.B. & Williams, W.V. (1996) Rational design of granulocyte-

macrophage colony-stimulating factor antagonist peptides. J. Biol.

Chem. 271, 2966±2971.

31. Fontecilla-Camps, J.C., Almassy, R.J., Suddath, F.L. & Bugg, C.E.

(1982) The three-dimensional structure of scorpion neurotoxins.

Toxicon 20, 1±7.

32. Kharrat, R., Darbon, H., Rochat, H. & Granier, C. (1989) Structure±

activity relationships of scorpion a-toxins. Eur. J. Biochem. 181,

381±390.

33. Vazquez, A., Becerril, B., Martin, B.M., Zamudio, F., Bolivar, F. &

Possani, L.D. (1993) Primary structure determination and cloning of

the cDNA encoding toxin 4 of the scorpion Centruroides noxius

Hoffmann. FEBS Lett 320, 43±46.

34. Martin, B.M., Carbone, E., Yatani, A., Brown, M.A., Ramirez, A.N.,

Gurrola, G.B. & Possani, L.D. (1988) Amino acid sequence and

physiological characterization of toxins from the venom of the

scorpion Centruroides limpidus tecomanus Hoffmann. Toxicon 26,

785±794.

35. Zamudio-ZunÄiga, F. (1989) Preparation, characterization and use of

monoclonal antibodies for the study of the structure and function of

scorpion toxins. Master of Sciences Thesis, School of Chemistry,

National University of Mexico.

36. DeLean, A., Munson, P.J. & Rodbard, D. (1978) Simultaneous analysis

q FEBS 1999 Neutralizing epitope of scorpion toxin Cn2 (Eur. J. Biochem. 264) 755

of families of sigmoidal curves: application to bioassay, radioligand

assay, and physiological dose±response curves. Am. J. Physiol. 235,

E97±E102.

37. Zilberberg, N., Froy, O., Loret, E., Cestele, S., Arad, D., Gordon, D. &

Gurevitz, M. (1997) Identification of structural elements of a

scorpion a-neurotoxin important for receptor site recognition.

J. Biol. Chem. 272, 14810±14816.

38. Jover, E., Couraud, F. & Rochat, H. (1980) Two types of scorpion

neurotoxins characterized by their binding to two separate receptor

sites on rat brain synaptosomes. Biochem. Biophys. Res. Commun. 64,

1607±1614.

39. Wheeler, K.P., Watt, D.D. & Lazdunski, M. (1983) Classification of

Na+ channel receptors specific for various scorpion toxins. PfluÈgers

Arch. 97, 164±165.

40. Yamamoto, H., Sejbal, J., York, E., Stewart, J.M., Possani, L.D. &

Kotovych, G. (1999) An NMR. conformational analysis of a synthetic

peptide Cn2 (1±15) NH2-S-S-acetyl-Cn 2 (52±66): NH2 from the

new world Centruroides noxius 2 (Cn2) scorpion toxin: comparison

of the structure with those of the Centruroides scorpion toxins.

Biopolymers. 49, 277±286.

41. Stanfield, R.L., Fieser, T.M., Lerner, R.A. & Wilson, I.A. (1990)

Crystal structures of an antibody to a peptide and its complex with

peptide antigen at 2.8 AÊ . Science 248, 712±719.

42. Ghiara, J.B., Stura, E.A., Stanfield, R.L., Profy, A.T. & Wilson, I.A.

(1994) Crystal structure of the principal neutralization site of HIV-1.

Science 264, 82±85.

43. Tormo, J., Blaas, D., Parry, N.R., Rowlands, D., Stuart, D. & Fita, I.

(1994) Crystal structure of a human rhinovirus neutralizing antibody

complexed with a peptide derived from viral capsid protein VP2.

EMBO J. 13, 2247±2256.

44. Blumenstein, M., Matsueda, R., Timmons, S. & Hawiger, J. (1992) A.

a-turn is present in the 392±411 segment of human fibrinogen

a-chain, effect of structural changes in this segment on affinity

to antibody 4A5. Biochemistry 31, 10692±10698.

45. Campbell, A.P., Sykes, B.D., Norrby, E., Assa-Munt, N. & Dyson, H.J.

(1996) Solution conformation of an immunogenic peptide derived

from the principal neutralizing determinant of the HIV-2 envelope

glycoprotein gp125. Folding Design 1, 157±165.

46. Devaux, C., Juin, M., Mansuelle, P. & Granier, C. (1993) Fine

molecular analysis of the Androctonus australis Hector scorpion

neurotoxin II: a new antigenic epitope disclosed by the pepscan

method. Mol. Immunology 30, 1061±1068.

47. Reece, J.C., Geysen, H.M. & Rodda, S.J. (1993) Mapping the

major human T helper epitopes af tetanus toxin. J. Immunol. 151,

6175±6184.

48. Bahraoui, E., Granier, C., Van Rietschoten, J., Rochat, H. & El Ayeb,

M. (1986) Specificity and neutralizing capacity of antibodies

elicited by a synthetic peptide of scorpion toxin. J. Immunol. 136,

3371±3377.

49. Ait-Amara, D., Chavez-Olortegui, C., Romi, R., Mery, J., Brugidou, J.,

Albericio, F. & Devaux, & Granier, C. (1993) Antibodies cross-

reactive with the scorpion toxin II from Androctonus australis Hector

elicited in mice by a synthetic peptide. Nat. Toxins 1, 255±262.

50. PeÂpin-Covatta, S.t., Lutsch, Ch., Grandgeorge, M., Lang, J. &

Scherrmann, J.-M. (1996) Immunoreactivity and pharmacokinetics

of horse anti-scorpion venom F(ab 0)2-scorpion venom interactions.

Toxicol. Appl. Pharmacol. 141, 272±277.

51. CalderoÂn-Aranda, E., Riviere, G., Choumet, V., Possani, L.D. &

Bon, C. (1999) Pharmacokinetics of the toxic fraction of

Centruroides limpidus limpidus venom in experimentally enveno-

mated rabbits and effects of immunotherapy with specific F(ab 0)2.

Toxicon 37, 771±782.

52. Ada, G.L. (1989) Vaccines. In Fundamental Immunology (Paul, W.,

ed.), 2nd edn, pp. 985±1032. Raven Press, New York.