In vivo protection against Androctonus australis hector scorpion toxin and venom by immunization with a synthetic analog of toxin II

Elsevier PII: S026440X(97)00144-2

Vaccine, Vol. 15, No. 2, pp. 187-194, 1997 Copyright 0 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0264-410X/97 $17+0.00

ELSEVIER

In vivo protection against Androctonm australis hector scorpion toxin and venom by immunization with a synthetic analog of toxin II I. Zenouaki*, R. Kharrat*, J.-M. Sabatiert, C. Devauxt, J. Van Rietschotent, M. El Ayeb*, H. Rochatt

H. Karoui*,

A synthetic peptide mimicking the North African scorpion Androctonus australis hector toxin II was designed and produced by chemical solid-phase synthesis. It contains the entire sequence of toxin II (64 amino acid residues), with each half-cystine being replaced by the isosteric residue a-aminobutyric acid, and was thus devoid of disuede bridges. This construct was totally nontoxic in mice even if large amounts, equivalent to 1000 times the LD,, of the original toxin, were injected by the intracerebroventricular route. The synthetic peptide, either as a monomer or polymerized by means of glutaraldehyde, induced the production of antitoxin neutralizing antibodies in immunized mice and rabbits. After three injections with either the monomeric or polymerized synthetic peptide, the immunized mice were protected against several lethal doses of the corresponding native toxin or scorpion venom. Six months after immunization, the mice were completely protected against challenge with eight LD,, of the original toxin. The protection was better when the polymerized synthetic peptide was used. One month after the start of the immunization program, it showed a good correlation between antibody titer and protection. However, antibody titer decreased with time but protection remained high. This suggests that additional factors other than circulating antibodies play a role in protective activity. Copyright 0 1997 Elsevier Science Ltd. Keywords:

Scorpion

toxin: cysteine-substituted

analog;

neutralization;

Scorpion envenomation remains a serious problem in many countries. In Mexico, Dehesa-Davila and Possani’ reported 200000 cases of scorpion stings with a death rate of 310 persons per year. In Tunisia, epidemiological data collected from 1986 to 1992 suggest that 3000045000 people are stung by scorpions annually. The number of deaths varied from 35 to 105 per year, most of which were children. Scorpions of the Buthidae family are most frequently incriminated. In Tunisia, Androctonus australis hector (Aah) is regularly implicated. Its venom contains many toxins with a M,. of 6-7 kDa which are extremely noxious to mammals due to their ability to bind with high affinity to the sodium channel of nervous tissue’.

*Laboratoire des Venins et Toxines, lnstitut Pasteur de Tunis, B.P. 74, 1002 Belvedkre, Tunis, Tunisia. TLaboratoire de Biochimie, CNRS URA 1455, lngenierie des Proteines, Facultk de Mbdecine Secteur Nord, Bd P. Dramard, 13916, Marseille, Cedex 20, France. To whom correspondence should be addressed. (Received 22 December 1995; revised 17 May 1996; accepted 13 June 1996)

immuno-protection

Current therapy involves administration of specific antivenom sera’. However, the effectiveness of this treatment is strongly dependent on its rapidity because the toxic proteins diffuse throughout the body making their subsequent capture by specific antibodies difficult. The possibility of promoting a state of immunity in peopleat-risk has been explored. Possani et ~1.~and recently Chavez-Olortegui et aL5 demonstrated the feasibility of such an approach. Using animal models, they obtained in vivo protection against scorpion venoms by immunization with detoxified venoms. Bahraoui et aZ.6investigated the possibility of using protein carrier-linked synthetic peptides to generate neutralizing antibodies against scorpion toxins. Further development of this approach requires the design of a synthetic peptide including T-cell and B-cell neutralizing epitopes. In this work we assessed the’ feasibility of the design and chemical synthesis of a nontoxic peptide corresponding to the entire sequence of Aah toxin II with substitutions of the eight half-cystine residues by a-aminobutyric acid, as an anatoxin prototype to protect Swiss mice. This toxin was chosen because it is well characterized structurally7.8.9,‘0 and antigenically6.’ ‘,‘2.‘3,‘4,‘5 and is the

Vaccine

1997 Volume

15 Number

2

187

In vivo protection against scorpion venom: 1. Zenouaki et al.

most toxic [LD,, as low as 0.18 pg per 20 g of C57BL16 mouse injected by a subcutaneous (s.c.) route] and the most abundant16,‘7 of the toxins in Aah venom. Thus, its specific neutralization might confer at least partial protection against Aah venom.

MATERIALS

AND

METHODS

(Abu)S AahII synthesis and characterization Materials. Boc-His(Z)-OH and Boc-a-aminobutyric acid (Boc-Abu-OH) were obtained from Bachem and Novabiochem, respectively. All other N-a-Boc-L-amino acids and 4-methylbenzhydrylamine hydrochloride resin (MBHA) were from Applied Biosystems. Dichloromethane, from SDS, was freshly distilled over anhydrous and diisopropylethylamine potassium carbonate, (DIEA) from Merck was distilled over ninhydrin prior to use. Trifluoroacetic acid (TFA), from SDS, was refluxed overnight over chromoxide, filtered, and then distilled over free amino acid. N-methylpyrrolidone, from SDS, was distilled under reduced pressure. Other solvents and reagents were analytical grade commercial products from Sigma and Merck. Freund’s complete and incomplete adjuvant, antimouse and anti-rabbit peroxidase conjugate and HEPES choline were purchased from Sigma. Glutaraldehyde and aluminium hydroxide adjuvant were from BDH and Superfor Biosector a/s, respectively. Bovine serum albumin (BSA) and O-phenylene diamine (OPD) were from Merck. Sephadex G-50 was from Pharmacia. Na[‘251] was from Amersham. Toxins and Aah-GSO have been purified as already described by Miranda et aZ.16.AahI, AahII, and AahIII are toxins I, II, and III of Androctonus australis hector venom and Aah-G50 is the toxic fraction obtained from the water extract of Aah venom by gel filtration on Sephadex G-50. Bot I is toxin I of Buthus occitanus tunetanus. Chemical synthesis and characterization of (Abu)8 Aahll. (Abu)8 AahII (synthetic analqi of AahII) was synthesized by the solid-phase method using an automated peptide synthesizer (model 430A, Applied Biosystems). The peptide chain was assembled stepwise on 0.5 m-equiv MBHA (1% cross-linked; 0.77 m-equiv of amino group per gram) using 2 mmol of the Wbutyloxycarbonyl (Boc) amino acids. Side-chain protecting groups used for trifunctional residues were: benzyl (Bzl) for Asp, Glu, Ser, and Thr; 2-chlorocarbobenzoxy (CIZ) for Lys; 2-bromocarbobenzoxy (BrZ) for Tyr; N-im-carbobenzoxy (Z) for His; and p-toluenesulfonyl (Tos) for Arg. W-amino groups were deprotected by treatment with 33 and 50% (v/v) TFA/dichloromethane for 80 s and 18 min 30 s, respectively. After washings with dichloromethane (3 x 1 min), the free a-NH, was generated with 10% (v/v) DIEAIN-methylpyrrolidone further washings with N(2 x 1 min). After methylpyrrolidone (3 x 1 min), the Boc-amino acids were double-coupled for 48 min as their hydroxybenzotriazol (HOBt) active esters in N-methylpyrrolidone (fourfold excess). A neutralization step with 10% (v/v) DIEAIN-methylpyrrolidone was performed before each recoupling. Coupling tests were not used for monitoring the condensation step.

188

Vaccine 1997 Volume 15 Number 2

After assembly was completed and the N-terminal Boc group was cleaved, the peptidyl-resin (6.1 g) was treated for 1 h at 0°C with anhydrous HFlp-cresoll ethanedithiol (85:10:5, v/v) in a volume of 15 ml per gram of resin. The HF was then removed under reduced pressure and the crude peptide was precipitated and washed with cold diethyl ether in the presence of 1% (v/v) b-mercaptoethanol, and then extracted with 50 ml H,O prior to lyophilization. The crude product (2.4 g) was purified by preparative reversed-phase mediumpressure liquid chromatography (Labomatic, Cl 8 HDSIL 15-25 ym, 26 x 313 mm). A 90 min linear gradient of acetonitrile in 0.1% (v/v) TFA/H,O was used for elution from 0 to 60% (flow rate=10 ml min- ‘; A=206 nm). Homogeneity of the target product was assessed by analytical reversed-phase HPLC (Merck, Cl8 Lichrospher 5 pm, 4 x 250 mm) using a 40-min linear gradient of acetonitrile in 0.1% (v/v) TFA/H,O from 10 to 70% (flow rate=1 ml mm ‘; A=230 nm). The purified peptide (784 mg) was characterized by amino acid analysis after hydrolysis [6 N HCl/l% (w/v) phenol, 2048 h, 112”C]. Polymerization of (Abu)8 AahII with glutaraldehyde Aliquots of 5 ,~l (300 ,~l total volume) of 10% glutaraldehyde were added sequentially every 2 min to 10 mg of (Abu)8 AahII in 1 ml of 20 mM phosphate buffer pH 6.6. The samples were incubated for 2 h at 37°C and then the reaction was stopped by addition of 50 ~1 of 10% acetic acid. Purification and characterization of (Abu)8 AahII polymer The reaction mixture was applied to a Sephadex G-50 column (1.6 x 100 cm) equilibrated and eluted with 0.01% acetic acid at a flow rate of 10 ml h-l. The first two peaks were analyzed by SDS polyacrylamide (12%) gel electrophoresis as previously described”. Toxicity estimation Toxicity of AahII, (Abu)8 AahII, and (Abu)8 AahIIp [glutarddehyde polymerized (Abu)8 AahII] was estimated by injection into 20 g mice which were observed for 20 h16. The LD,, of AahII is 2 ng for Swiss mice and 1 ng for C57BL/6 mice when injected by the intracerebroventricular route (i.c.v.). It is 0.30 lug for Swiss mice and 0.18 pg for C57BL/6 mice injected by the S.C. route. Animals and immunization programs Animals. Swiss mice were produced in the veterinary service facilities of Institut Pasteur of Tunisia and C57BL/6 mice in the Marseille laboratory. New Zealand rabbits were from Elevage des Dombes, Romans, France. Mouse immunization programs. Six to eight week old mice were immunized, by intraperitoneal and S.C.routes, with 50 ,ug of synthetic peptides emulsified in 200 ~1 Complete Freund’s Adjuvant (CFA, protocol I) or in aluminium hydroxide gel (protocol II). Each mouse was boosted 7 and 17 days later, with 100 and 150 rug of

In vivo protection against scorpion venom: I. Zenouaki et al.

immunogen, respectively. Mice of protocol I were boosted again on days 42, 58, and 109 with 200 ,ug of immunogen in Incomplete Freund’s Adjuvant (IFA) and bled on days 30, 52, 70, 102, and 123 from the retro orbital sinus 10 days after the last boost. Immune sera were tested by ELISA for presence of antibodies against (Abu)8 AahII, (Abu)8 AahIIp, AahII, AahI, and BotI. Mice of protocol II were bled on days 30, 75, and 180 without further boosts. Immune sera were tested for binding on AaH-coated plates. Rabbit immunization program. Three rabbits were primed with 300 pg of (Abu)8 AahII emulsified in CFA by intradermal injections and boosted, five times, every 3 weeks with 200 ,ug in IFA by S.C.injection. Rabbits were bled 10 days after each boost.

ELISA assays ELISA was used both to test cross antigenicity of (Abu)S AahII and (Abu)8 AahIIp and specificity of anti-(Abu)8 AahII and anti-(Abu)8 AahIIp against AahII, AahI, AahIII, and BotI toxins. One hundred microliters of 5 pg ml- ‘, 2 ,ug ml- ‘, or 0.5 ,ug ml-’ of antigen in 0.1 M sodium bicarbonate buffer pH 9.6 were adsorbed on 96 well NUNC plates for 90 min at 37°C. After washing five times with PBS/Tween (0.05%) nonspecific sites were saturated with BSA 0.5% in PBS buffer for 1 h at 37°C and the plates washed three times with PBS/O.OS% Tween. Appropriate dilutions of sera with or without preincubation with AahII in PBSlTween were then added, and samples were incubated at 37°C for 90 min and 4°C for 15 min. After washing with PBS/O.OS% Tween, 100 ~1 of peroxidase conjugate (anti-mouse or anti-rabbit diluted lOOO-fold) was added to each well and incubated at 37°C for 90 min and 4°C for 15 min. Finally, 200 ~1 of 0.4 mg ml-’ OPD in citrate buffer pH 5.2 containing H,O, (0.03%) were added to each well and the plates incubated in the dark for 5 min at room temperature. The reaction was stopped by addition of 50 ~1 of 2 N sulphuric acid. Absorbance at 492 nm was then measured using a Titertek Multiskan photometer. The titer of immune sera corresponds to the dilution giving 50% of the maximum absorbance.

AahII iodination Five micrograms of AahII were iodinated with 1 mCi of Na[“‘I] using the lactoperoxidase procedure as previously described by Rochat et al.“.“.

RIA [“‘I]AahII was diluted to 5 x lo-” M in PBS .containing 0.1% BSA (PBSB). Various dilutions of rabbit antisera were incubated with the labeled toxin for 90 min at 37°C then overnight at 4°C. One milliliter of sheep anti-rabbit precipitating antibody (UCB Bioproducts, Belgium) was then added. After incubation for 30 min at 4°C. the immune complexes were pelleted by centrifugation at 1OOOOgand radioactivity was measured with a gamma counter (Packard Crystal II). For competitive RIA, each of a series of concentrations of unlabeled AahII or of (Abu)8 AahII was incubated with a 1:lOO

dilution of rabbit nonimmune or anti-(Abu)8 AahII for 5 h before adding the [‘*‘I]AahII. All assays were performed in duplicate. Inhibition of [‘251]AahIIbinding to rat brain synaptosomes by immune sera Svnaptosome preparations. Rat brain synaptosomes were prepared according to the method of Gra and Whittaker2’ as modified by Blaustein and Ector x and described by Jover et ai.24. Inhibition test. Two hundred microliters of rat brain synaptosomes (3 mg ml-‘) in HEPES choline buffer containing 0.1% BSA were added to lo- lo M (final concentration) of [‘251]AahII and various dilutions of mouse immune and nonimmune sera, and protein A purified immune and nonimmune rabbit IgG. Samples were incubated for 30 min at 37°C and the reaction was stopped by centrifugation at 9000g for 5 min. Pellets were washed twice with 1 ml of HEPES choline buffer containing 0.25% BSA and counted for radioactivity. Values reported are means of duplicate determinations. Neutralizing capacity of immune sera The neutralizing capacity of immune sera (from rabbits or from mice of protocol I) were tested by i.c.v. route. Each of a series of amounts of AahII equal to or higher than the value of LD,, (1 and 2 ng for C57BL/6 and Swiss mice, respectively) were preincubated for 90 min at 37°C and 2 h at 4°C with 5 ~1 of immune sera and were injected by the i.c.v. route into mice. Mortality was recorded after 20 h. Protection assay of immunized mice against challenge with toxins and venoms During the immunization program (protocol II), groups of six mice were challenged S.C. with a series of amounts of Aah venom, Aah G-50, and AahII. The amounts used were equal to or higher than those corresponding to LD,, which are 35 ,ug, 11.5 pug, and 0.3 pg for Aah venom, Aah G-50, and AahII, respectively. Mortality was recorded 20 h later.

RESULTS Characterization of (Abu)S AahII and glutaraldehyde-polymerized (Abu)8 AahII (Abu)8 AahII was obtained by the solid-phase peptide synthesis method and purified. Analytical HPLC showed a single peak eluting at 48% acetonitrile. After acid hydrolysis, the amino acid ratios were: Asx 7.8 (8); Thr 2.6 (3); Ser 1.6 (2); Glx 4.0 (4); Gly 7.5 (7); Ala 3.2 (3); a-aminobutyric acid 7.6 (8); Val 3.7 (4); Ile 0.8 (1); Leu 2.1 (2); Tyr 6.1 (7); Phe 1.0 (1); His 1.9 (2); Lys 4.9 (5); Arg .2.9 (3); and Pro 3.1 (3). The major fraction obtained from the Sephadex G-50 elution of (Abu)8 AahII polymerized with glutaraldehyde was analyzed by SDS-PAGE. Various forms of polymerized (Abu)8 AahII were observed with M, from 14 to 70 kDa (data not shown). This fraction, (Abu)8 AahIIp, was used throughout the present work without further fractionation.

Vaccine 1997 Volume 15 Number 2

189

In vivo protection against scorpion venom: 1. Zenouaki et al.

0 1

2

3

4

30

50

70

100

125

150

Time (days)

5

-log [serum dilution] Figure 1 Cross-antigenicity of (Abu)8 Aahll and (Abu)8 Aahllp with Aahll. Binding of rabbit anti-Aahll antibodies to 5 ,ug ml-’ Aahll (A), (Abu)8 Aahll (O), and (Abu)8 Aahllp (B)-coated plates was revealed using anti-rabbit peroxidase conjugate in ELISA. Values given are the mean of duplicates

C57BU6 and Swiss mice injected with 1 mg of (Abu)8 AahII (3300 LD,, of original toxin) by S.C. route and 2 pug of (Abu)8 AahII or (Abu)8 AahIIp (1000 LD,, of original toxin) by i.c.v. route did not show any symptoms of intoxication. Thus, these two preparations were considered as nontoxic to mice and appropriate for use as models of anatoxins. Cross-antigenicity of (Abu)8 AahII and (Abu)8 AahIIp with AahII was tested by ELISA using anti-AahII rabbit polyclonal antibodies (Figure 1). (Abu)8 AahII and (Abu)8 AahIIp were recognized by anti-AahII antibodies diluted 1:10000. only However, a 1:50000 dilution of anti-AahII recognized AahII.

-I 1

190

Vaccine 1997 Volume 15 Number 2

5

c

6s

-“o3l!8 s $ s

3

0.5

5:

9

8 2

8

f

0:

I

1

Characteristics of immune response against (Abu)8 AahII The sera taken at various times from mice that were immunized with (Abu)8 AahII in Freund’s adjuvant (protocol I) were tested by ELISA for their reactivity with AahII (Figure 2A). After three injections of (Abu)8 AahII (50, 100, and 150 pug) over 1 month, a titer of about 1:320 was observed. The titer increased to 1:5000 following a boost with 200 pg on day 42. The titer remained roughly constant for 4 months after the start of immunization (Figure 2A) despite further injections of 200,~g on days 58 and 109. The sera taken at 1,2.5, and 6 months from mice that were immunized with (Abu)8 AahII or (Abu)8 AahIIp in aluminium gel (protocol II) were tested by ELISA for their reactivity with AahII (Figure 2B). Antibodies from mice that had been immunized for 1 month recognized AahII-coated plates. In the same conditions, antibodies from mice that had been immunized for 2.5 and 6 months bound poorly to AahII-coated plates. Immune response to the (Abu)8 AahII was also studied in New Zealand rabbits (Figure 2C). The immune sera recognized AahII-coated plates by ELISA and [‘251]AahII in a liquid-phase radioimmunoassay. The titer deduced from ELISA were 1:5000-1:lOOOO as early as the third injection and remained constant after several boosts (Figure 2C). These results showed that a strong immune response against AahII could be obtained by immunizing mice and rabbits with (Abu)8 AahII. However, this response was not maintained without regular boosts with the immunogen. The specificity and avidity of mouse and rabbit anti-(Abu)8 AahII antibodies to AahII were tested by ELISA and. RIA, respectively. Diluted mouse anti-(Abu)8 AahII antibodies (1:9000), collected 70 days after the start of the immunization program (protocol I)

2 3 4 -log [serum dilution]

’

1

1

1

4 2 3 -log [serum dilution]

’

5

Figure 2 Reactivity of mouse anti-(Abu)8 Aahll immune sera with Aahll. Immune sera from mice immunized with (Abu)B Aahll in Freund’s Adjuvant (A, protocol I) or aluminium hydroxide (B, protocol II) were tested on 5 pg ml-’ Aahll-coated plates. The dilution of mouse immune sera, taken at various times (protocol I), and giving half-maximum binding effects (Titer) was also determined (A). Serum sample dilutions from mice immunized respectively with (Abu)8 Aahll (B, filled symbols) or (Abu)8 Aahllp (6, open symbols) in aluminium gel (protocol II) for 1 (0, O), 2.5 (A, A), and 6 months (m, 0) were tested in ELISA using 5 pg ml-’ Aahll-coated plates. Rabbit immune sera raised against (Abu)8 Aahll (serial dilutions) were tested for reactivity against 5 pg ml-’ Aahll-coated plates in ELISA (C, A) and against 11251]Aahll in RIA (C, W). Values given are the mean of duplicates. Nonspecific binding in ELISA or RIA obtained with mice or rabbit nonimmune sera was ~10%

were first incubated with free AahII at different concentrations (10~6M-10-1’ M) then reacted with AahII (0.5 ,ug ml-‘) coated plates. A binding competition curve was observed (Figure 3A). Half-maximal inhibition was obtained using lo-*M free AahII. A comparable result was obtained by RIA (competition assay) using [“‘I]AahII, AahII, (Abu)8 AahII, and rabbit anti(abu)8 AahII (Figure 3B). Binding competition curves were observed and respective concentrations of AahII and (Abu 8 AahII giving half-maximal binding effects of 1.4 x lo- 4 M and 0.9 x lop8 M were calculated. These results suggested that at least a subpopulation of anti(Abu)8 AahII antibodies recognized AahII with high avidity. Cross-antigenicity of (Abu)8 AahII with other scorpion toxins belonging to antigenic groups different from AahII was tested by ELISA (Figure 4A and B). Mouse (Figure 4A) and rabbit (Figure 4B) anti-(Abu)8 AahII antibodies mainly recognized AahII and only weakly recognized AahI, AahIII, and BotI toxins. The mouse

In vivo

0.8 1

A

E

0.6 -

8 s $

0.4 -

s B8

0.1

2

protection against scorpion venom: I. Zenouaki et al.

I

-

04 12

I

11

I

10

I

I

.

i

9

8

7

6

I 5

0

2 4 -log [serum dilution]

-log [Aahll], M

0

Figure 3 Specificity and avidity of mouse (A) and rabbit (B) anti-(Abu)8 Aahll antibodies to Aahll. Inhibition ELISA (A) were performed with Aahll-coated plates (0.5 pg ml-‘) to which Aahll (various concentrations) and a I:9000 dilution of anti-(Abu)8 Aahll were added. Anti-(Abu)8 Aahll were preincubated with free Aahll before addition to Aahll-coated plates. Half-maximal inhibition was obtained with IO-’ M free Aahll. Competitive inhibition (B) of [1251]Aahll binding to anti-(Abu)8Aahll (dilution 1:400) was tested by serial dilutions of (Abu)8 Aahll (A) or unlabeled Aahll (W). 6 and 6, are binding of [‘251]Aahll to anti-serum in presence or absence of competitor, respectively. The half-maximal inhibition effect was obtained using 0.9x10-* M and 1.4~10~’ M of (Abu)8 Aahll, and Aahll, respectively

and rabbit immune response against (Abu)S AahII seems to be restricted to the AahII antigenic group. The capacity of mouse anti-(Abu)8 AahII sera to neutralize AahII toxicity was evaluated by two approaches. First, we measured inhibition of [‘*‘I]AahII (lo- lo M) binding to rat brain synaptosomes by preincubation of the labeled toxin with anti-(Abu)8 AahII (sera from protocol I) or nonimmune sera (Figure 5). Anti-(Abu)8 AahII antibodies completely inhibited the binding of [1’51]AahII to rat brain synaptosomes and half-maximal inhibition was obtained with a dilution of 1:200. Anti-(Abu)8 AahII diluted 1:50 completely inhibited the binding of [i2’I]AahII to rat brain synaptosomes. Nonimmune mouse sera inhibited binding by only 20%. The second approach was preincubation of a given amount of AahII with anti-(Abu)8 AahII before injection into mice and evaluation of the protective effect against toxin lethality. Anti-(Abu)8 AahII sera protected mice against AahII toxicity. The minimum neutralizing titer as measured by i.c.v. injections of the toxin-antibody mixture was 800 LD,, mll’ which corresponds to 1.6 pug of toxin neutralized per ml of sera. The neutralizing capacity of rabbit anti-(Abu)8 AahII serum was similarly tested. The binding of [‘251]AahII to rat brain synaptosomes was inhibited by rabbit anti(Abu)8 AahII purified IgG with a K,,, of 10K6 M (data not shown). The protective effect of rabbit anti-(Abu)8 AahII serum was evaluated by preincubating given amounts of AahII with preimmune (control) or immune

2 4 -log [serum dilution]

6

6

Figure 4 Cross-antigenicity of (Abu)8 Aahll with Aahll, Aahl, and Botl. (A) Reactivity in ELISA of various dilutions of mouse anti(Abu)8 Aahll immune sera with 5 ,ug ml-’ Aahll (A), Aahl ( q), and Botl (0) coated plates. (B) Reactivity in ELISA of various dilutions of rabbit anti-(Abu)8 Aahll immune sera with 5 pg ml-’ Aahll (A), Aahl (0) and Aahlll (+)-coated plates. Binding was measured by reading absorbance at 492 nm

3

2

1

-log [serum dilution]

Figure 5 In vitro Aahll neutralizing capacity of anti-(Abu)8 Aahll (from protocol I). Binding of [‘251]Aahll (lo-” M) to rat brain synaptosomes is inhibited by preincubation with various dilutions of anti-(Abu)8 Aahll (W). Half-maximal inhibition was obtained with anti-(Abu)8 Aahll diluted 1:200. Mouse nonimmune sera was used as a control (A). B and B,, are binding of [‘251]Aahll to rat brain synaptosomes in the presence or absence of sera

sera followed by i.c.v. injection into C57BW6 mice, the mouse strain most sensitive to scorpion toxins (Table 1). Full protection (no toxic effect) was observed if up to 5 ng were injected per mouse (neutralization could be estimated at 1800 LD,, i.e. 1.8 pug ml-’ of serum). Immunoprotection of mice by immunization with (Abu)8 AabII and (Abu)S AahIIp

We then tested the ability of (Abu)8 AahII and (Abu)8 AahIIp anatoxins in aluminium gel to confer protection to Swiss mice against toxicity of AahII, AahG-50, and Aah crude venom. In protocol II, 100 and 150 mice were immunized with (Abu)8 AahII and (Abu)8 AahIIp, respectively, and challenged, by S.C. injections, with AahII, AahG-50, and Aah venom on days 30, 60, 75, and 180 after the beginning of the

Vaccine 1997 Volume 15 Number 2

191

In vivo protection against scorpion venom: 1. Zenouaki et al. Table 1 Neutralizing activity of rabbit anti-(Abu)8 AaH II. AaH II toxin was preincubated with antiserum before injection into C57BU6 mice Amount of AaH II injected (ng)

Nonimmune

Anti-(Abu)8

2 5 10 20

l/6= O/6 O/6 O/6

616 616 416 316

Rabbit sera

aSurvived/injected mice. The LD,, 20 g mouse by the i.c.v. route

without

antisera

AaH II

was 1 ng per

100 8 s .t; 8 S a

80 60 40 20

3

4

6

5

8 1012

3

4

8 1012

8

suggests a specific and high protection against AahII which is the most potent and the main toxin in Aah venom. The protection observed against AahG-50 and Aah venom was thus presumably due to neutralization of AaH II. The protection against AahII of mice immunized for 1 month with (Abu)8 AahII (Figure 6A) and (Abu)8 AahIIp (Figure 6B) is due to the presence of high level of circulating antibodies, respectively (Figure 2B). However, the protection against AahII of mice immunized for 2.5 and 6 months with (Abu)S AahII (Figure 6A) and (Abu)8 AahIIp (Figure 6B) could not be explained by the low-binding capacity of the corresponding anti(Abu)S AahII and anti-(Abu)8 AahIIp on AahII-coated plates (Figure 2B). The neutralizing capacity of sera from mice immunized for 1, 2.5, and 6 months was tested by preincubation with 5 ng of AahII (2.5 LD,,) before i.c.v. injection of 5 ,~l to naive Swiss mice. Sera from mice immunized with (Abu)8 AahII and (Abu)S AahIIp for 1 month protected one and four out of six mice injected, respectively. Immune sera from mice immunized for a longer period did not have any protective effect. These data suggest that components other than specific circulating antibodies may be implicated in the in viva protection after immunization with (Abu)S AahII and (Abu)8 AahIIp.

Aahll challenge dose (LD50)

DISCUSSION 8 5 .5 al 5 ri

80 60 40 20

1

3

4

5

6

5

8 1012

3

4

a 1012

8

Aahll challenge dose (LD50) Figure 6 Protection of Swiss mice immunized with (Abu)8 Aahll and (Abu)8 Aahllp (protocol II). After receiving three injections of (Abu)8 Aahll (A) or (Abu)8 Aahllp (B) emulsified in aluminim hydroxide, Swiss mice were challenged after 1, 2, 2.5, and 6 months with various doses of Aahll and the percentage of animals protected was recorded. Six mice are used for each injected Aahll challenge dose

program (see Figure 6A and B). On day 30, mice were completely protected when injected with 0.9 pugof AahII (3 LD,,). The apparent value for the LD,, measured was 1.5-1.8 pug regardless of the immunogen used. On day 60, the protection was improved, particularly with (Abu)8 AahIIp: mice were completely protected against 3.6 ,ug of AahII (12 LD,,), whereas with (Abu)S AahII, complete protection was obtained against 1.5 ,ug of AahII with an apparent LD,, of 3.6 pg. Similar results were obtained on day 75 except that the apparent LD,, for mice treated with (Abu)8 AahII was 2.4 ,ug. Finally, 6 months after the beginning of the immunization program (day 180), all mice injected were completely protected against 2.4 pug of Aah toxin II (8 LD,,). In other respects, the immunized mice were partially protected against injection of Aah venom or the toxic fraction Aah G-50 which is known to contain all the toxins directed against the sodium channel and is responsible for the lethality of the venom (Table 2). This

192

Vaccine

1997

Volume

15 Number

2

About 30000-40000 scorpion stings per year are regularly reported in Tunisia. Work involving chemical or physical detoxification of venom for use in active protection against the noxious effects of scorpion venoms and toxins has in the past obtained encouraging and promising results4.5. However, a practical problem remains in obtaining enough venom from scorpions to prepare the necessary doses. Synthetic and genetic engineered peptides may be good alternatives to the conventional approaches. Therefore, the development of immunoprotection against scorpion venoms first requires identification of the potent and antigenic toxins. A. australis hector venom contains two antigenic groups represented by toxins I and IIi2. Toxin II is the most potent toxin produced by scorpions found in Tunisia16. Synthetic peptide 50-59 and l-8 derived from AahII toxin are able to induce neutralizing antibodies in rabbits6,25 (Devaux et al., unpublished data). However, to be immunogenic, these peptides have to be coupled to a carrier protein. (Abu)8 AahII is expected to share some conformational similarity with AahII although it is devoid of residual toxicity to mice at doses as high as 2 pg by i.c.v. route. Significant antigenic cross-reactivity between AahII and (Abu)S AahII is noticed. Toxicity is more sensitive than antigenicity to the conformation of the peptide and there is no overlap between the two activities26. (Abu)8 AahII is able to elicit antibody production in mice and rabbits. AahII toxin was wellrecognized by all sera. The half-maximal inhibition assessed by ELISA using mouse anti-(Abu)8 AahII sera was obtained with lo-’ M of free AahII. The immune response against (Abu)8 AahII seems, however, to be restricted to toxins antigenically related to AahII. AahI, AahIII, and Bot I which belong to other antigenic groups’ ’ reacted poorly with rabbit and mouse anti(Abu)8 AahII antibodies. These antibodies inhibited

In vivo Table 2

Protection of Swiss mice by immunization

protection against scorpion venom: 1. Zenouaki et al.

with (Abuf8 Aahll and (Abuf8 Aahllp against challenge with Aahll, Aah G-50, and Aah venom lmmunogens (Abu)8 Aahllp

(Abu)8 Aahll Immunization period (month)

Amount of challenge dose (LD,,)

1

3

2

2.5

6 6 8 3 4 6

Toxins and venoms challenge

Aahll 616”

Aah G-50 l/6

Aahll 616

Aah G-50 216

Aah -

316 616 416 616 616

O/6 l/6 O/6

416 616 616 616 616 6/6

l/6 316 216 l/6 016

216 116 616 316 O/6

LD,, Aahll, 0.3 ,ug per 20 g mouse; Aah G-50, %urvivedlinjected

-

11.5 pg per 20 g mouse; Aah venom,

[‘251]AahII binding to its receptor sites on rat brain synaptosomal membranes and neutralized 1.6 pug or 1.8 pg (6 LD,,) of AaH II if the toxin was preincubated with 1 ml of mouse or rabbit immune sera, before injection by the i.c.v. route, respectively. We assessed the abilities of monomeric and glutaraldehyde-polymerized (Abu)8 AahII in aluminium gel to confer protection to Swiss mice against toxicity of AahII, Aah toxic fraction obtained by filtration on Sephadex G-50 (Aah G-50) and Aah crude venom. Mice immunized with 300 ,ug of glutaraldehyde-polymerized (Abu)8 AahII given in three doses were significantly protected against 2.4 pg (8 LD,,) of toxin administered 2-6 months after the beginning of the program. There was good correlation between protection and the level of circulating antibodies found just after the end of the immunization program (i.e. 1 month after the start of the immunization program). Unexpectedly, the levels of circulating antibodies were lower after 2.5 and 6 months. This was inconsistent with the good protection against AahII toxin (1.2-3.6 pug). This discrepancy between the low level of specific antibodies and high protection appeared earlier after immunization by glutaraldehyde-polymerized (Abu)S AahII than by monomeric antigen. However, 6 months after the beginning of the program, mice immunized by monomeric or glutaraldehyde-polymerized (Abu)8 AahII resisted the AahII challenge doses as high as 2.4yg (8 LD,,): in both cases the levels of anti-AahII specific antibodies were low. Possibly immunization with free or polymerized (Abu)8 AahII led to potent but low levels of neutralizing antibodies. However, this possibility is not supported by the poor neutralizing activity of immune sera collected after 6 months. These observations are consistent with the involvement of unknown resistance mechanisms in long-term immunization. Mutations of receptor-specific toxin binding sites that would be induced by free and polymerized (Abu)8 AahII in immunization did not seem to occur since partial resistance to challenge by Aah crude venom is observed: Aah venom is known to contain other toxins (AaHI and AaHIII) not antigenitally similar to AahII but which recognize the same binding sites on the sodium channel’.‘. Work is in progress to investigate whether other mechanisms such as modifications of toxicokinetic parameters, acceleration of toxin catabolism and/or specific carrier induction by immunization with synthetic peptides are involved. Significant protection against Aah venom (100%

35 ,ug per 20 g mouse.

(-)

Not determined,

protection against 105 lug at 2.5 months) is an encouraging result (Table 2). In our experience, the biggest deadly scorpions in the world would hardly give 2500 pug of venom under electrical stimulation, conditions which are known to empty the venom glands. Thus, protection against scorpion envenomation apparently is possible by immunoprotection against the most potent toxin in the venom. This result could be extended to other scorpion venoms and their most potent toxins such as toxin V from Androctonus mauretanicus mauretanicus2’ and toxin V from Leiurus quinquestriatus quinquestriatuJ6. These two scorpions are known to be very dangerous and responsible for health problems in Morocco and in the Middle East, respectively.

ACKNOWLEDGEMENTS We gratefully acknowledge Professor Koussay Dellagi Head of Institut Pasteur de Tunis for his constant encouragement and helpful advice. We would also like to thank Dr Ben Lasfar Zakaria for providing venoms and animals. This research was supported in part by funds from INSERM Grant No. 492 NS6 and the European Economic community Grant No. CIl-CT930071.

REFERENCES Dehesa-Davila, M. and Possani, L.D. Scorpionism and serotherapy in Mexico. Toxicon 1994, 32, 1015-1018 Catterall, W.A. Membrane potential-dependent binding of scorpion toxin to the action potential Na’ ionophore. J. biol. Chem. 1977, 252,8660-8668 Campos, J.A., Silva, OS., Lopez, M. and Freire-Maria, L. Signs symptoms and treatment of severe scorpion sting in children. In: Natural Toxins (Eds Eaker, D. and Wadstrom, T.). Pergamon, Oxford, 1980, pp. 61-68 Possani, L.D., Fernandez De Castro, J., Julia, J.Z. Detoxification with glutaraldehyde of purified scorpion (Centruroides noxius Hoffmann) venom. Toxicon 1981, 19,323-329 Chavez-Olortegui, C., Ait Amara, D., Rochat, H., Diniz, C. and Granier, C. In vivo protection against scorpion toxins by liposomal immunization. Vaccine 1991, 9, 907-910 Bahraoui, E., El Ayeb, M., Van Rietschoten, J., Rochat, H. and Granier, C. Immunochemistry of scorpion alpha-toxins: study with synthetic peptides of the antigenicity of four regions of toxin II of Androctonus australis hector. Molec. lmmun. 1986, 23, 357-366 Rochat, C., Rochat, H., Kopeyan, C., Miranda, F., Lssitzky, S. and Edman, P. Scorpion neurotoxins: a family of homologous proteins. FEBS Left. 1970, 10, 349-351

Vaccine

1997 Volume

15 Number

2

193

In vivo protection against scorpion venom: I. Zenouaki et al. 8

9

10

11

12

13

14

15

16

17

194

Rochat, H., Bernard, P. and Couraud, F. Scorpion toxins: chemistry and mode of action. Advances in Cytopharmacology, Vol. 3 (Eds Ceccarelli B. and Clement F.). Raven Press, New York, 1979, pp. 325-334 Delori, P., Van Rietschoten, J. and Rochat, H. Scorpion venoms and neurotoxins: an immunological study. Toxicon 1981, 19, 392407 Fontecilla-Camps, J.C., Habersetzer-Flochat, C. and Rochat, H. Toxin II from the scorpion Androcfonus australis hector orthorhombic crystals, three-dimensional structure and structure function relationships. froc. natn Acad. Sci. U.S.A. 1988, 85, 7443-7447 El Ayeb, M., Delori, P. and Rochat, H. Immunochemistry of scorpion alpha toxins, antigenic homologies checked with radio immunoassays (RIA). Toxicon 1983, 21, 709-716 El Ayeb, M., Martin, M.F., Delori, P., Bechis, G. and Rochat, H. Immunochemistry of scorpion alpha neurotoxins. Determination of the antigenic site number and isolation of highly enriched antibody specific to a single antigenic site of toxin II of Androctonus australis hector. Molec. Immun. 1983, 20, 697-708 Granier, C., Novotny, J., Fontecilla-Camps, J.C., Fourquet, P., El Ayeb, M. and Bahraoui, E. The antigenic structure of a scorpion toxin. Molec. Immun. 1989, 28, 503-513 Bahraoui, E., El Ayeb, M., Rochat, H., Van Rietschoten, J. and Granier C. Immunological studies with peptides assembled on a new polyacrylamide solid support. In: Peptides 84 (Ed. Ragnarson U.). Almquist and Wiksell Int., Stockholm, 1984, pp. 165-l 68 Fourquet, P., Bahraoui, E., Fontecilla-Camps, J.C., Van Rietschoten, I., Rochat, H. and Granier, C. Immunochemistry of scorpion toxins: synthesis and antigenic properties of a model of a loop region specific to alpha toxins. Int. J. Peptide Protein Res. 1988, 32, 81-88 Miranda, F., Kopeyan, C., Rochat, C., Rochat, H. and Lissitzky, S. Purification of animal neurotoxins. Isolation and characterization of eleven neurotoxins from the venoms of the scorpions Androctonus australis hector, Buthus occitanus tunetanus and Leiurus quinquestriatus quinquestriatus. Eur. J. Biochem. 1970, 16, 514-523 Martin, M.-F. and Rochat, H. Purification of thirteen toxins active on mice from the venom of the north african scorpion Buthus occitanus tunetanus. Toxicon 1984, 22, 279-291

Vaccine 1997 Volume 15 Number 2

18 19

20

21

22

23

24

25

26

27

Merrifield, R.B. Solid phase synthesis. Science 1986, 232, 341-347 Shagger, H. and Von Jagow, G. Tricine-sodium dodecyl sulfatepolyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Analyt. Biochem. 1987, 186,368-379 Rochat, H., Kopeyan, C., Garcia, L.G. et a/. Recent results on the structure of scorpion and snake toxins. In: Animal, P/ant and Microbial Toxins, Vol. 2 (Eds Ohsaka, A., Hayashi, K. and Sawal, Y.). Plenum Publishing Co., New York, 1976, pp. 79-87 Rochat, H., Tessier, M., Miranda, F. and Lissitzky, S. Radioiodination of scorpion and snake neurotoxins. Analyt. Biochem. 1977,85,532-538 Gray, E.G. and Whittaker, V.P. The isolation of nerve endings from brain: an electron-microscopic study of cell fragments derived by homogeneization and centrifugation. J. Anat. 1962, 96,79-88 Blaustein, M.P. and Ector, C. Carrier-mediated sodiumdependent calcium efflux from pinched-off presynaptic nerve terminals (synaptosomes) in vitro. Biochem. biophys. Acta 1976, 419,295-308 Jover, E., Couraud, F. and Rochat, H. Two types of scorpion neurotoxin characterized by their binding to two separate receptor sites on rat brain synaptosomes. Biochem. biophys. Res. Commun. 1980, 95, 1607-1614 Devaux, C., Juin, M., Mansuelle, P. and Granier, C. Fine molecular analysis of the antigenicity of the Androctonus australis hector scorpion neurotoxin II: a new antigenic epitope disclosed by the pepscan method. Molec. Immun. 1993, 30, 1061-1068 El Ayeb, M., Darbon, H., Bahraoui, E.M., Vargas, 0. and Rochat, H. Differential effects of defined chemical modification on antigenic and pharmacological activities of scorpion alpha and beta toxins. Eur. J. Efiochem. 1986, 155, 289-294 Rosso, J.P. and Rochat, H. Characterization of ten proteins from the venom of the moroccan scorpion Androctonus mauretanicus mauretanicus, six of which are toxic to the mouse. Toxicon 1985, 23, 113-125