Immunogenicity of polymerizable synthetic peptides derived from avaccine candidate against schistosomiasis: the asparaginyl

endopeptidase (Sm32)

N. Chacon a, S. Losada a, H. Bermudez a, I.M. Cesari b, J. Hoebeke c, Oscar Noya a,*a Seccion de Biohelmintiasis, Instituto de Medicina Tropical, Universidad Central de Venezuela, Apartado Postal 47623, Zona Postal 1041-A, Los

Chaguaramos, Caracas, Venezuelab Centro de Microbiologıa y Biologıa Celular, Instituto Venezolano de Investigaciones Cientıficas, Altos de Pipe, Venezuela

c UPR9021C.N.R.S. Immunologie et Chimie Therapeutiques, I.B.M.C., Strasbourg, France

Received 30 March 2003; accepted 9 April 2003

Immunology Letters 88 (2003) 199�/210

www.elsevier.com/locate/

Abstract

The asparaginyl endopeptidase (Sm32) is expressed in the gastrodermal cells of the schistosome gut and in the head glands of the

cercariae. Possibly, Sm32 hydrolyzes pro-proteins involved in the degradation of host hemoglobin [Parasitol. Today 12 (1996) 125].

Preliminary evidences using an Sj32/Sm32 murine vaccine have shown a profound effect on oviposition and worm burden [Chin. J.

Schist. Control. 7 (1995) 72; Bull. Human Med. Univ. 24 (1999) 225; Vaccine 20 (2002) 439]. The importance of Sm32 as a novel

vaccine candidate is based on the possibility of preventing the maturation of other cathepsins and/or preventing schistosome skin

invasion. We studied the immunogenicity of polymerizable peptides derived from Sm32 to select potential protective epitopes. Sm32

prediction of T and B epitopes and homology studies with human legumain were performed. Among the variety of factors that

influence the antibody response, we specifically examined the effect of: (i) genetic background of mouse strain, inbred (C57BL/6)

versus outbred (Swiss) mice; and (ii) vaccination with a single peptide versus pool of peptides. Swiss mice raised antibodies to three

different regions of the Sm32, as tested by the Multiple Antigen Blot Assay (MABA): 182�/215 (peptides IMT-70 and 72), 244�/273

(IMT-64) and 336�/355 (IMT-66). None of these regions were immunogenic for C57BL/6. On the contrary, other peptides, IMT-4

(21�/40), IMT-12 (101�/120) and IMT-26 (292�/313) were highly immunogenic for this inbred strain. Only Swiss mice immunized

with a single peptide (IMT-64 and 72) or with three different pools of IMT-peptides (Pool A-II: 14, 16, 18, 70, 72, 89; pool A-III: 22,

64, 24, 26, 28 and pool A-V: 64, 66, 28, 70, 72) recognized the original protein in a crude extract of the worm antigen by Western

blot. Peptides IMT-64, 14 and 26 were responsible for this recognition. In general, the vaccination with pool of peptides was more

immunogenic for both mouse strains. Predicted B cell epitopes, with hydrophilicity scores over �/10 (IMT-12, 64, 26) were always

immunogenic after either single or combined peptide vaccination. Sm32 sequences 41�/80 (IMT-6 and 8), 141�/160 (IMT-16) and

182�/215 (IMT-70 and 72) were nearly identical to the corresponding human legumain regions and should be excluded from the

human vaccine. We can conclude that the regions of Sm32 that were recognized by antibodies of mice immunized with

polymerizable peptides depended on the mice strain and on the hydrophilicity score of the peptides.

# 2003 Published by Elsevier B.V.

Keywords: Asparaginyl endopeptidase; Vaccine; Schistosoma mansoni ; Synthetic peptides

1. Introduction

Schistosomiasis is an endemic disease that affects

around 200 million persons in the world, basically in

Africa, Eastern Mediterranean, the Caribbean and

South America [5]. Relatively safe drugs have been

developed against this parasitism, although there are

increasing experimental and field evidences indicating

the risk of resistance to these drugs [6,7]. In addition to

Abbreviations: IMT, peptides synthesized at the ‘‘Instituto de

Medicina Tropical’’ of the Venezuelan Central University; IVIC,

Instituto Venezolano de Investigaciones Cientıficas; MABA, multiple

antigen blot assay; SAWA, soluble adult worm antigen; UCV,

Universidad Central de Venezuela.

* Corresponding author. Tel.: �/58-212-605-3571; fax: �/58-212-

605-3563.

E-mail addresses: noyaoo@yahoo.com, natychacon@yahoo.com

(O. Noya).

0165-2478/03/$ - see front matter # 2003 Published by Elsevier B.V.

doi:10.1016/S0165-2478(03)00084-1

the classical control measures based on environmental

sanitation and education, alternative control measures

like the induction of protection against infection with

low-cost vaccines are a priority for development.Currently, there are five schistosome vaccine candi-

dates being evaluated [8], two of them are synthetic

antigens in the form of Multiple Antigen Peptides and

the rest are recombinant antigens. None of them reached

a protection higher than 40% when independent studies

were undertaken using both C57BL/6 and BALB/c

inbred mice. Therefore, additional molecules should be

incorporated in the anti-schistosome vaccine studies.Sm32 is the Schistosoma mansoni asparaginyl endo-

peptidase [9], a cysteine protease of the legumain family

[10]. It is released as an excretory�/secretory material

that hydrolyzes pro-proteins involved in the degradation

of hemoglobin [1], the principal source of aminoacids

for the parasite. The importance of Sm32 as a novel

vaccine candidate is based on the possibility of inducing

an antibody-mediated inhibition of its catalytic activity,therefore, preventing the processing of cathepsins like B

and L, directly involved in hemoglobin digestion [12].

Preliminary evidences using inhibitors of these activities

have shown a profound reduction on the oviposition of

the adult worms [13]. Mice DNA-vaccinated with Sm32

developed antibodies, which recognized the native

protein not only in homogenates of S. mansoni but

also in the gut of the parasites [4]. Recombinant Sj32vaccination demonstrated a protective immunity as

judged by measuring worm burden and anti-fecundity

immunity against Schistosoma japonicum [3]. Addition-

ally, the immunolocation of the Sm32 in the head glands

of the cercariae [14,15] suggests also a role of this

protease in tissue invasion.

Therefore, Sm32 could be considered as an interesting

target, because it could affect the nutrition of the adultworms, thereby reducing worm burden, viability, fe-

cundity and also, interfering with the penetration of the

cercariae. This protein has been characterized, cloned

and sequenced [9,16,17] and has shown to be highly

antigenic in infected patients [18�/20]. All these results

suggest that this molecule could be one of the antigens

responsible for the concomitant immunity characteristic

of the schistosome infection.Cattle was protected against foot and mouth disease

by the use of chemically synthesized peptides free of any

carrier, strategy first described by DiMarchi et al. [21]. A

hybrid synthetic protein polymer has also been evalu-

ated in a vaccine that protects humans against malaria

[22,23]. Similarly, we decided to use polymerizable

synthetic peptides in the present study.

Different factors affect the level of protection in anyvaccine model and may separately or in concert

stimulate antibody responses to distinct antigens. The

genetic background of vaccinated mice is one of the

factors that influence resistance to challenge [24]. The

inbred C57BL/6 mouse strain develops higher levels of

resistance to challenge following immunization with

irradiated cercariae. Therefore, it is considered as a

high responder strain [25].In this study, we evaluated the patterns of immune

recognition of Sm32 by antibodies obtained from inbred

(C57BL/6) or outbred (Swiss) mouse strains after multi-

ple vaccination with synthetic polymerizable peptides

derived from Sm32 to identify and select potential

protective epitopes.

2. Materials and methods

2.1. Parasites

The JL strain of S. mansoni has been maintained in

the laboratory using the snail host Biomphalaria glab-

rata and hamsters. Adult worm antigen (SAWA) was

prepared as described by Noya et al. [19] and protein

determination was assayed by the method of Bradford

[26].

2.2. Mice

Six-weeks-old outbred male Swiss mice weighting

between 18 and 25 g were obtained from the Instituto

de Medicina Tropical (Universidad Central de Vene-

zuela, Venezuela), and inbred C57BL/6 female mice 6�/

8-weeks-old, weighting around 18�/20 g from theInstituto Venezolano de Investigaciones Cientificas

(IVIC).

2.3. Synthetic peptides

All peptides were manually synthesized using Merri-field’s protocol [27] for the t-Boc-based solid-phase

peptide synthesis, modified by Houghten [28], for the

simultaneous multiple-peptide synthesis. Glycine and

cysteine amino acids were introduced at both carboxy

and amino termini to allow polymerization and to

increase immunogenicity. For peptide cleavage, the

protected complex peptide-resin was treated first with

low HF concentration (25%) at 0 8C in order to cleaveprotecting groups and then, with high HF concentration

(90%) to cleave the peptide from the resin [29]. After HF

evaporation under N2 stream, each peptide was washed

with cold diethyl ether and then extracted with 5% acetic

acid and lyophilized. Peptide purity was assessed by

analytical RP-HPLC and peptide quality by MALDI-

TOF mass spectrometry. This analysis was kindly

carried out at the Instituto de Inmunologıa in Bogota,Colombia. The whole sequence of amino acids of the

Sm32 described by Klinkert (1989) (GI 729709 NCBI)

was synthesized as 22 discontinuous peptides (Table 1).

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210200

2.4. Predictive studies of the Sm32

Selection of B-cell epitopes was based on predictions

made by a PC/GENE software and Hopp and Woods [30]

algorithm. All putative T-cell epitopes for three or more

mice haplotypes were selected using the software devel-oped by Guillet et al. [31].

2.5. Search for protein similarity and cross-reactive B-

cell epitopes

To assess amino acid identity, we compared themammalian legumain and the Sm32 sequences using

the ANTHEPROT V 4.3c program [32]. To estimate the

existence of potential cross-reactive B cell epitopes in the

different peptides, similarity scanning was performed

using the algorithm of Guillet et al. [31]. A six amino

acids window was used, since this is the optimal size for

a B cell epitope. Cross-reactivity was considered if the

epitopes had a similarity higher than 50% using theHopp and Woods [30] scoring system.

2.6. Immunization experiments

All mice were bled by tail before immunization. Swiss

and C57BL/6 mice were immunized with five subcuta-

neous doses of 50 mg of each individual peptide or 20 mg/peptide when pools of five peptides were used, in the

presence of Freund’s complete (first dose) and incom-

plete (successive doses) adjuvant at 15 days interval.

Immunizations were administered as shown in the

Tables 2 and 3.

2.7. Multiple antigen blotting assay (MABA)

MABA was used to evaluate the immunogenicity of

the peptides [33]. In this technique, the Sm32 peptides

were adsorbed independently and simultaneously in 24

parallel rows onto a nitrocellulose membrane and 2 mm

strips were then cut in order to expose all the 22 Sm32

peptides to immune mice sera (1:100 dilution). A 1:2000

dilution of anti-mouse IgG conjugate to horseradish-peroxidase was then added and developed with Lumi-

nol† (Amersham), a chemiluminescent substrate of the

enzyme. Luminescence was recorded with Hyperfilm†

(Amersham, UK).

2.8. SDS-PAGE and immunoblotting

The recognition of the original Sm32 was evaluated

by Western-blot. Soluble adult worm antigen (SAWA)

was separated by SDS-PAGE [34] in 10% gels under

reducing conditions, transferred onto a nitrocellulose

membrane, exposed to mice sera (1:100 dilution) and

developed with an anti-mouse IgG conjugated to

peroxidase (1:2000). Luminol† was used as previously

described by Noya et al. [19]. An inhibition assay wasdesigned for those animals immunized with cocktails of

peptides to demonstrate the peptides responsible for the

recognition of the native Sm32 molecule. In this assay,

Table 1

Synthesized peptides derived from the pro-enzyme and mature enzyme sequence of Sm32

IMT Sequences Location Hidrophilicity [30]

2 CG MMLFSLFLISILHILLVKCQ GC 1�/20 �/28.9

Mature enzyme 4 CG LDTNYEVSDETVSDNNKWAV GC 21�/40 �/6.5

Mature enzyme 6 CG LVAGSNGYPNVRHQADYCHA GC 41�/60 �/6.5

Mature enzyme 8 CG YHVLRSKGIKPEHIITMMYD GC 61�/80 �/1.5

Mature enzyme 10 CG DIAYNLMNPFLGKLFNDYNH GC 81�/100 �/9.3

Mature enzyme 12 CG KDWYEGVVIDYRGKKVNSKT GC 101�/120 �/14.3

Mature enzyme 14 CG FLKVLKGDKSAGGKVGGVLKSGK GC 121�/140 �/7.2

Mature enzyme 16 CG NDDVFIYFTDHGAPGLIAFP GC 141�/160 �/10.2

Mature enzyme 18 CG DDELYAKEFMSTLKYLHSHKRY GC 161�/182 �/7.1

Mature enzyme 70 CG YSKLVIYIEANESGSMFQQIL GC 182�/202 �/10.9

Mature enzyme 72 CG GSMFQQILPSNLSIYATTAAN GC 195�/215 �/13.9

Mature enzyme 89 CG PTECSYSTFCGDPTITTC GC 216�/223 �/9.6

Mature enzyme 22 CG LADLYSYNWIVDSQTHHLTQ GC 224�/243 �/17.4

Mature enzyme 64 CG RTLDQQYKEVKRETDLSHVQ GC 244�/273 �/11.7

Mature enzyme 24 CG VQRYGDTRMGKLYVSEFQGS GC 272�/291 �/4.9

Mature enzyme 26 CG RDKSSTENDESPMKPRHSIASR GC 292�/313 �/24.2

28 CG DIPLHTLHRQIMMTNNAEDKSF GC 314�/335 �/1.8

66 CG LMQILGLKLKRRDLIEDTMK GC 336�/355 �/9.4

30 CG LIVKVMNNEEIPNTKATIDQ GC 356�/375 �/2.3

32 CG TLDCTESVYEQFKSKCFTLQ GC 376�/395 �/0.8

34 CG QAPEVGGHFSTLYNYCADGY GC 396�/415 �/12.4

36 CG CADGYTAETINEAIIKICG GC 411�/429 �/1.6

The mature enzyme is comprised between the 9th amino acid of the IMT-4 and the 1st amino acid of IMT-26.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210 201

each serum was pre-incubated with each peptide sepa-rately for 30 min. If the serum contained antibodies

against a given peptide, they would bind to the

corresponding peptide blocking the recognition of the

Sm32 kDa band in SAWA.

3. Results

3.1. Predictive analysis of the Sm32 molecule

Table 1 shows the amino acid sequences of the

synthetic peptides of Sm32 and their respective hydro-

philicity score. Hydrophilic peptides and probably B-cell

epitopes were determined according to Hopp and

Woods [30]. IMT-12, 64 and 26 are hydrophilic peptides

and score over �/10. Other hydrophilic peptides areIMT-4, 14 and 66, but their score is lower, between �/5

and �/10. IMT-70 and 72 are hydrophobic peptides (B/

�/10) with possible n-glycosylation sites. Based on the

algorithm of Guillet et al. [31], Sm32 contains at least

seven Th cell epitopes included in the following peptides:

IMT-12, 16, 70, 72, 28, 30 and 32, which theoretically

should be recognized by three or more different mice

haplotypes (Table 4).

3.2. Protein similarity and cross-reactive B-cell epitopes

In Table 5 we show the comparison between se-

quences of three asparaginyl endopeptidases: the human

legumain [2] (GI 2842759 NCBI) and the asparaginylendopeptidases of S. mansoni sequenced by Klinkert et

al. [9] (GI 729709 NCBI) and by Caffrey et al. (GI

6851050 NCBI) [17].

The human legumain and the asparaginyl endopepti-

dases of S. mansoni proposed by Klinkert et al. [9] are

59% homologous. Also, the human legumain and

sequence proposed by Caffrey et al. [17] are 61%

homologous.The sequences 41�/60 (IMT-6), 81�/100 (IMT-10)

141�/160 (IMT-16) and 182�/215 (IMT-70 and 72),

254�/273 (IMT-64), 314�/335 (IMT-28) showed more

than 50% similarity between the proteins [2,9]. Se-

quences 61�/80 (IMT-8) and 234�/253 (IMT-22) showed

50% similarity while the other peptides showed lower

than 50% similarity.

Cross-reactions in MABA were obtained betweendifferent peptides derived from Sm32. In most cases

putative cross-reactive B cell epitopes could be demon-

strated (Table 6). An exception was the cross-reactivity

between IMT-2 and 4, which reacted in between without

presenting possible cross-reactive B cell epitopes.

3.3. Immunization with single peptides

All peptides were individually used to immunize the

high responder C57BL/6 mice. Peptides IMT-2, 4, 6, 12,

64, 26 and 30 induced a positive response (Fig. 1 and

Table 7). Antisera against IMT-2 and 4 showed a

mutual cross-reactivity while antibodies against IMT-6

Table 2

Immunization with single peptides

Mice Number of animals Groups Peptidea Concentration (mg/peptide per mouse per dose) Adjuvant

C57BL/6 3 Control �/ �/ �/ �/

3 Adjuvant �/ �/ �/ �/

C57BL/6 3/peptide II 2 4 6 8 10 12 50 �/

14 16 18 70 72 89

22 64 24 26 28 66

30 32 34 36

a Single peptide.

Table 3

Immunization with pools of peptides

Mice Number of animals Group Peptidea Concentration (mg/peptide per mouse per dose) Adjuvant

Swiss 10 Control �/ �/ �/

10 Adjuvant �/ �/ �/

10 A-I IMT-2, 4, 6, 8, 10, 12 20 �/

9 A-II IMT-4, 16, 18, 70, 72, 89 20 �/

7 A-III IMT-22, 64, 24, 26, 28 20 �/

5 A-IV IMT-66, 30, 32, 34, 36 20 �/

7 A-V IMT-64, 66, 28, 70, 72 15 �/

a Cocktail of peptides.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210202

showed a consistent cross-reactivity with peptides IMT-

10 and 30. None of the anti-peptide antisera were able to

recognize SAWA with the exception of antisera against

IMT-6 and 8. Three out of the seven peptides with a

positive hydrophilicity score (IMT-12, 26, 30) induced

antibodies while only two out of the seven peptides with

a putative Th cell epitope (IMT-12 and 30) had the same

effect.

3.4. Immunization with pools of peptides

In view of the poor immune response obtained with

single peptides in C57BL/6 mice, we assessed the

immunogenicity of different peptide pools in an outbred

strain. All the pools were able to raise antibodies,

although not to all peptides present in a given pool

(Fig. 2 and Table 8). All peptides recognized by the

C57BL/6 antisera raised against single peptides (IMT-2,

4, 6, 12, 64, 26 and 30) were also recognized by antisera

from Swiss mice immunized with the peptide pools.

Only three out of seven peptides containing a putative

Th cell epitope were immunogenic in Swiss mice (Table

4 vs. Table 8). Cross-reactions with frequencies of more

or equal to 50% were observed between (1) antisera

against pool A-I and peptide IMT-26; (2) antisera

against pool A-II and peptides IMT-2, 4 and 30; (3)

antisera against pool A-IV and peptide IMT-64; (4)

antisera against pool A-V and peptides IMT-4, 10, 30,

and 34 (Table 8).

Interestingly, all the antisera of Swiss mice immunized

with different peptide pools, were able to recognize the

antigen in a SAWA Western blot (Fig. 3, line b). To

check which of the peptides in each pool was responsible

for the induction of specific antibodies, competitive

inhibition tests were performed using single peptides as

competitors for the recognition of Sm32 molecule (Fig.

3). Only one of the peptides in each pool seemed to

induce specific antibodies against Sm32 by Western blot.

All these peptides corresponded to highly hydrophilic

regions of Sm32 (IMT-14, 64 and 26) (Table 1).

4. Discussion

The radiation-attenuated schistosome vaccine (RAV)

has been the most successful model of protection, but it

cannot be used in humans. The attenuated vaccine

induces high levels of protection in permissive hosts as

mice (80�/90%) and olive baboon (�/85%) but, moder-

ate levels of protection in chimpanzees (40%). One of the

antigens that are recognized by a moderate-responder

mouse strain (CBA/J) after RAV is Sm32 [24].

In each animal model, vaccine candidates have been

found to give different degrees of protection, serving as

a first approximation for preclinical trials.

RAV induces in non-human primates a Th2 cytokine

profile by the time of challenge and this correlated with

protection, based on worm burden [35]. Contrary to

this, in mouse models, the cell-mediated TH1 immunity

predominates over the humoral response.

The requirement of large-scale development of a

vaccine would not have been possible before the advent

of recombinant DNA techniques or synthetic peptides

that has allowed the production of a multitude of

different candidate molecules. Nowadays, more than

100 antigens are available for schistosomiasis vaccine

development, mainly from S. mansoni [36,37]. Only five

antigens are currently considered by the WHO Special

Program for Research and Training in Tropical Disease

(TDR) [38,39]. One has reached the level of clinical

trials: the S. mansoni and S. haematobium glutathione-

S-transferase (GST) recombinant antigen developed by

Capron et al. [40]. Three other antigens are at the pilot

scale-up level: recombinant paramyosin (Sm97) worked

by the Queensland Institute of Medical Research in

Australia, triose-phosphate isomerase (Multiple Anti-

genic Peptide) developed by the Harvard Medical

School and the National Institute of Health in the

United States and the recombinant Sm14 antigen

produced by FIOCRUZ (Brazil).

Asparaginyl endopeptidases are members of a novel

family of cysteine proteases, which are termed legu-

mains, since they were first characterized in seeds of

leguminous plants [41]. Based on biochemical and

Table 4

Theoric predictive T-cell epitopes [31]

Peptide, Sm32 Location Sequence Haplotypes

IMT-12 101�/120 KDWYE GVVIDYRGKKVN SKT I-Ak, I-Eb, I-Ek

IMT-16 141�/160 NDDVFIYF TDHGAPGLIAFP I-Ab, I-Ad, I-Eb, I-Ek

IMT-70 182�/202 YSKLVIY IEA NESGSMFQQIL I-Ak, I-Eb, I-Ek

IMT-72 195�/215 GSMFQQILPS NLSIYATTAA N I-Ab, I-Ad, I-Ak

IMT-28 314�/335 DIPLHTLH RQIMMTNNAEDK SF I-Ab, I-Ad, I-Ed

IMT-30 356�/375 LIVKVMN NEEIPNTKATID Q I-Ab, I-Ad, I-Ed, I-Ek

IMT-32 376�/395 TLDCTES VYEQFKSKC FTLQ I-Ad, I-Ak, I-Eb

Gray color, T-cell epitope regions of synthetic peptide.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210 203

sequence similarities, the Sm32 and Sj32 proteases were

re-named as schistosome legumains. Later on, Chen et

al. [42] characterized mammalian legumain from human

placenta and pigs.

Legumains cleave peptide bonds on the carboxyl side

of Asn residues except where the Asn occurs at the NH2-

terminus or at the second position from the NH2-

teminus of the polypeptide or where the Asn is

Table 5

Sequences comparisons between human legumain and mature Sm32

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210204

glycosylated [41]. A role in post-translational modifica-

tions of proteins has been suggested for legumains [12].

This occurs for storage proteins in legume seeds by

cleaving bonds between pro-peptides and mature pro-

teins [41].

Dalton and Brindley [1] suggested that Sm32 might

activate the schistosome cathepsin L, as an Asn residue

occurs at the point of cleavage of the pro- and mature

protease. In the same way, an Asn residue is found in the

vicinity of pro- and mature forms of other schistosome

cathepsins, like cathepsin B (Sm31), C and D [1,12]. It is

important to mention that none of the homologous

human cathepsins (L, B and D) has Asn residues in this

vicinity. The same authors hypothesized that the schis-

tosome legumains (Sm/Sj32) are involved in the activa-

tion of several pro-cathepsins directly implicated in

hemoglobin degradation. The direct effect on the

degradation of hemoglobin still remains to be estab-

lished for Sm32 and Sj32. Various cathepsins, Sm32 and

Sj32 are all expressed in the gastrodermal cells of the

schistosome gut [1]. Therefore, Sm/Sj32 is probably

involved indirectly in hemoglobin digestion, a basic

nutrient for adult schistosome. This issue was demon-

strated in mice immunized with recombinant Sj32 and

challenged with S. japonicum , showing a significant

reduction in worm burden [3]. Sj32 immunization

induced a reduction in the number of eggs in the uterus

of female worm [2] and the DNA Sm32 vaccine led to a

reduction of the number of eggs per female worm [4].

Both antigens showed a anti-fecundity effect after

challenge.

Even though Sm32 was not considered a vaccine

candidate by the TDR [8], we decided to study it.

Additionally, there are preliminary evidences that this

enzyme is present in the cephalic glands of cercariae.

Therefore, drugs or vaccines that would block the

activity of the schistosome legumain may block the

activation of other proteolytic enzymes, directly in-

volved in hemoglobin digestion and in the case of

cercariae could block the penetration of the skin. Our

goal was to determine if a protective immunity could be

induced with this antigen in terms of effect on parasite

load and fecundity. For this purpose, we chemically

synthesized the whole 50 kDa pro-protein of 429 amino

Fig. 1. A representative MABA from C57BL/6 mice immunized with single pepides IMT-2, 4, 6, 12, 26 and 30; a, pre-immune sera; b, immune sera;

C�, mouse immunized with SAWA.

Table 6

Studies of homology between cross-reactive peptides

Peptide Location Sequence Cross-reactive

B-epitope

IMT-2 1�/20 MMLFSLFLISILHILLVKCQ

IMT-4 21�/40 LDTNYEVSDETVSDNNKWAY No

IMT-72 195�/215 GSMFQQILPSNLSIYATTAAN

IMT-64 244�/263 RTLDQQYKEVKR/ETDLSHVQ Yes

IMT-30 356�/37 LIVKVMNNEE/IPNTKA/TIDQ Yes

IMT-36 416�/429 CADGYTAETINEAIIKICG

IMT-12 101�/120 K/DWYEGVVID/YRGKKVNSKT Yes

IMT-32 376�/395 TLDC/TESVYEQ/FKSKCFTLQ Yes

IMT-6 41�/60 LVAGSNGYPNVRHQADYCHA

IMT-10 81�/100 DIAYNLMNPFLG/KLFNDYNH Yes

IMT-30 356�/375 LIVKVMN/NEEIPNTKAT/IDQ Yes

Peptides were scanned for similarity using the method of Guillet et

al. [31]. A window of six amino acids was used, corresponding to a B

cell epitope and cross-reaction was suggested if the homology score

based on Hopp and Woods [30] was higher than 50%.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210 205

acid residues of Sm32 in the form of discontinuous 22-

mer peptides. The pro-enzyme appears to be processed

at both ends, amino and carboxy-termini [11]. The

mature enzyme comprises 260 amino acids beginning

at Val32 (9th amino acid of IMT-4) and ending at Arg-

292 (1st amino acid of IMT-26).

In our immunogenicity study of the Sm32 peptides,

we choose two different mice models. A highly permis-

sive host, the syngeneic (inbred) strain C57BL/6, and a

non-syngeneic (outbred) strain, also a permissive host,

as Swiss mice, in order to deal with and to compare

genetically heterogeneous populations, looking towards

human vaccination trials.

The syngeneic C57BL/6 strain, usually considered as a

high responder animal model, was a poor responder to

most of the individual peptides used as immunogens.

More importantly, none of the anti-peptide antibodies

were able to recognize the Sm32 antigen in Western

blots. This observation precludes the induction of a

protective antibody response in this strain of mice by the

peptide strategy used. It also suggests that, although

syngeneic mice are essential to unravel the immune

mechanisms by which protective immunity can be

acquired, they are probably poor experimental models

to predict the overall protective effects of the peptide

vaccine candidates used in the present study.

In contrast, the use of the outbred Swiss strain

showed quite a different immune response. The number

of peptides against which antibodies were obtained was

not only higher than that obtained in the C57BL/6

strain, but it could clearly be shown that some of the

anti-peptide antibodies from Swiss mice also reacted

with the Sm32 antigen contained in the crude adult

worm extract (SAWA) in Western blots. The main

reason of the better immunogenic potential of these

peptides in the outbred strain is probably due to the

diversity of its immune genetic background and, more

especially, the diversity of the MHC class II molecules

which can present a large panoply of peptides to the T

cell receptors of the T helper cells. It must be stressed, in

this context, that all peptide cocktails used, contained at

least one peptide in which a potential T cell epitope was

Table 7

Frequency of recognition of Sm32 peptides by C57BL/6 mice immunized with individual peptides derived from the same molecule, by MABA

Groups of mice Peptide Frequency of recognition Cross-reactivity with other peptides and SAWA Frequency of recognition of Sm32

Control �/ 0/3 0/3 0/3

Adjuvant �/ 0/3 0/3 0/3

IMT-2a 2 3/3 4 3/3 0/3

IMT-4a 4 2/3 2 2/3 0/3

IMT-6 6 2/3 SAWA 2/3 0/3

10 1/3

30 1/3

IMT-8 8 0/2 SAWA 1/2 0/2

6 1/2

IMT-10 10 0/3 �/ 0/3

IMT-12a 12 3/3 �/ 0/3

IMT-14 14 0/3 �/ 0/3

IMT-16 16 0/3 �/ 0/3

IMT-18 18 0/3 12 1/3 0/3

IMT-70 70 0/3 �/ 0/3

IMT-72 72 0/3 2 1/3 0/3

4 1/3

10 1/3

64 1/3

30 1/3

IMT-89 89 0/3 �/ 0/3

IMT-22 22 0/3 �/ 0/3

IMT-64 64 2/3 �/ 0/3

IMT-24 24 0/3 �/ 0/3

IMT-26a 26 3/3 �/ 0/3

IMT-28 28 0/3 �/ 0/3

IMT-66 66 0/3 �/ 0/3

IMT-30a 30 2/3 �/ 0/3

IMT-32 32 0/3 �/ 0/3

IMT-34 34 0/3 �/ 0/3

IMT-36 36 0/3 12 1/3 0/3

32 2/3

‘�/’ indicates no cross-reactions.a Numbers in bold indicate an intense recognition by MABA.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210206

predicted. Activation of T helper cells with these

peptides would lead to liberation of cytokines, which

are important for proliferation, and differentiation of

neighbor peptides by specific B cells. Interestingly, in the

groups of mice immunized with a peptide cocktail, only

one or two peptides seemed to elicit cross-reactive

antibodies against the Sm32 antigen. Moreover, IMT-

64 induced such a response in all animals co-immunized

with different peptide cocktails. The same peptide was

also immunogenic in the syngeneic C57BL/6 strain but

without induction of Sm32 reactive antibodies. This

peptide could thus be an interesting component in a

multi-peptide anti-Sm32 vaccine.

Another observation is the cross-reactivity of anti-

peptide antibodies with peptides not used as immuno-

gens. To increase immunogenicity, all peptides used in

this study were made polymerizable. This means that all

peptides had in common CG residues at their ends.

Probably, the common NH2- and COOH-termini that

induced the formation of disulfide bonds could also be

responsible for the observed cross-reactions, as is the

case of cross-reactive peptides IMT-2 and 4. A similar

observation has been found when in linear polypeptides

with a sequential arrangement of epitopes separated by

spacers, the junctional epitope became immunodomi-

nant [43]. One of our laboratory aims is to design

‘‘silent’’ spacers to prevent non-specific immune re-

sponse and the probable immunomodulation of the

relevant epitopes. Cross-reactivity between peptides was

observed more in the outbred than in the inbred strain.

This result reveals a strain-dependent heterogeneity in

the antibody responses to the polymerizable synthetic

peptides.

The accurate determination of the homology between

the human legumain [2] and the schistosome asparaginyl

endopeptidase is a basic aspect to be evaluated on the

pipeline towards the development of a human vaccine.

The peptides that appear 100% identical between the

human legumain and Sm32 are IMT-6, 8 (41�/80), 16

(141�/160), 70 and 72 (182�/215). IMT-6, 16 and 70

possibly form part of the catalytic sites of the aspar-

aginyl endopeptidase [41], supporting the idea that

regions of the molecule that are highly conserved

through evolution are not good candidates for vaccine

trials, because of the potential induction of an auto-

immune response. In the present study, these peptides

appeared poorly immunogenic or non-immunogenic at

all by MABA for the two mouse strains. In addition,

these regions correspond to hydrophobic areas by Hopp

and Woods [30] and may be less exposed on the

molecular surface. Consequently, peptides IMT-6, 8,

16, 70 and 72 should be excluded from a human Sm32

Fig. 2. A representative MABA from Swiss mice immunized with pools of peptides A-I, A-II, A-III and A-IV; a, pre-immune sera; b, immune sera;

C�, mouse not immunized.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210 207

Table 8

Frequency of recognition by MABA of Swiss mice immunized with pools of synthetic peptides derived from Sm32

Groups of mice Peptides Frequency of

recognition

Cross-reactivity with

other peptides

Frequency of recognition of Sm32 Peptide responsible of

inhibition of recognition of Sm32

Control �/ 0 0 0 0

Adjuvant �/ 0 0 0 0

Pool A-I 2, 4, 6, 8, 10, 12 2 4/10 0/10 0

4 2/10

6 8/10 26 5/10

8 0/10 30 3/10

10 1/10

12 10/10

Pool A-II 14, 16, 18, 70, 72, 89 14 9/9 2 3/9 6/9 2/3 (IMT-14)

14 2

16 0/9 64 3/9

18 3/9 28 4/9

70 3/9 30 3/9

72 1/9

89 0/9

Pool A-III 22, 64, 24, 26, 28 22 3/7 2 4/7 3/7 1/3 (IMT-26), 1/3 (IMT-64)

64 7/7 4 4/7

24 4/7 10 3/7

26 7/7 30 6/7

28 7/7

Pool A-IV 66, 30, 32, 34, 36 66 5/5 64 4/5 0/5 0

30 3/5

32 0/5

34 0/5

36 0/5

Pool A-V 64, 66, 28, 70, 72 64 6/7 2 2/7 7/7 3/4 (IMT-64)

66 4/7 4 4/7

28 4/7 8 3/7

70 1/7 10 4/7

72 3/7 12 3/7

14 3/7

26 3/7

30 5/7

34 4/7

Numbers in bold indicates an intense recognition by MABA, numbers in gray color a moderate recognition. Note, cross-reactions were considerated only when reactivity was seen in three or more

mice per group.

N.

Ch

aco

net

al.

/Im

mu

no

log

yL

etters8

8(

20

03

)1

99�

/21

02

08

vaccine. We can expect that the mouse legumain is also

highly conserved in these regions and that the lack of

response was probably due to auto-tolerance. Thus, in

view of the identity existing between the mammalian

and the worm sequence for peptides IMT-6, 8,16, 70 and

72, it is not surprising that the anti-peptide response was

negative, or in the case of IMT-6 and 72, did lead to an

anti-peptide response which showed no reactivity with

the original enzyme.

The peptide sequences with the lower degree of

homology between the human and the parasite legumain

were: IMT-12 (101�/120), IMT-26 (292�/313) and IMT-

30 (356�/375) and are those to be assayed as a first

choice for protection. Since IMT-26 elicits anti-peptide

antibodies able to react with the Sm32 antigen, it should

also be a good candidate for a peptide vaccine against

Sm32. However, there is a risk that antibodies raised

against regions located outside the catalytic site could

not inhibit the enzyme activity and, therefore, would not

be protective.

This work suggests that vaccine protection protocols

should be preferentially developed in outbred mouse

strains and that a final vaccine against schistosomiasis

will likely consist of a mixture of peptides, recombinant

antigens or DNA fractions to cover the genetic diversity

of immune response within populations.

Acknowledgements

We are grateful to IVIC for providing C57BL/6 used

for these studies. This work was partially supported by

FONACIT-Project S1-2000000564.

References

[1] J.P. Dalton, P.J. Brindley, Schistosome asparaginyl endopepti-

dase Sm32 in hemoglobin digestion, Parasitol. Today 12 (1996)

125.

[2] Y. Chen, X. Yi, X. Zeng, S. Wang, Schistosoma japonicum :

antifecundity immunity induced by adult worm derived 31/32 kDa

antigens in mice, Chin. J. Schist. Control. 7 (1995) 72�/74.

[3] C. Chun, X. Yi, Z. Xianfang, A. Jinchun, S. Xinhu, Schistosoma

japonicum : protective immunity induced in mice immunized with

SjGST-Sj32, Bull. Hum. Med. Univ. 24 (1999) 225�/228.

[4] K. Chlichlia, M. Bahgat, A. Ruppel, V. Schirrmacher, DNA

vaccination with asparaginyl endopeptidase (Sm32) from the

parasite Schistosoma mansoni : anti-fecundity effect induced in

mice, Vaccine 20 (2002) 439�/447.

[5] A. Bresson, I. Nutall, ‘‘Schistosomiasis’’: the disease, the current

situation, schistosomiasis control, action, important dates, con-

clusion. WHO Home Page, Division of Control of Tropical

Disease (http://www.who.org/) 1995.

[6] Y.S. Liang, G.C. Coles, M.J. Doenhoff, Detection of praziquantel

resistance in schistosomes, Trop. Med. Int. Health 5 (2000) 72.

[7] J. Kusel, P. Hagan, Praziquantel*/its use, cost and possible

development of resistance, Parasitol. Today 15 (1999) 352�/354.

[8] N.R. Bergquist, D.G. Colley, Schistosomiasis vaccine research to

development, Parasitology 14 (1998) 99�/104.

[9] M.Q. Klinkert, R. Felleisen, G. Link, A. Ruppel, E. Beck,

Primary structures of Sm31/32 diagnostic proteins of Schistosoma

mansoni and their identification as proteases, Mol. Biochem.

Parasitol. 33 (1989) 113�/122.

[10] M.Q. Klinkert, A. Ruppel, E. Beck, Cloning of diagnostic 31/32

kilodalton antigens of Schistosoma mansoni , Mol. Biochem.

Parasitol. 25 (1987) 247�/255.

[11] O. Takeda, Y. Miura, M. Mitta, H. Matsushita, I. Kato, Y. Abe,

H. Yokosawa, S. Ishii, Isolation and analysis of cDNA encoding

a presursor of Canavalia ensiformis asparaginyl endopeptidase

(legumain), J. Biochem. 16 (1994) 541�/546.

[12] J.P. Dalton, L. Hola-Jamriska, P.J. Brindley, Asparaginyl en-

dopeptidases activity in adult Schistosoma mansoni , Parasitology

111 (1995) 575�/580.

[13] M.M. Wasilewski, K.C. Lim, J. Phillips, J.M. McKerrow,

Cysteine protease inhibitors block schistosome hemoglobin de-

Fig. 3. Competitive inhibition assay by Western blot. Demonstration of the peptide responsible for the recognition of Sm32 molecule in SAWA (in

bold) for those groups of mice immunized with different pools of peptides; a, pre-immune sera; b, immune sera; Conj., conjugated control; SH,

infected S. mansoni human sera; Ir, irrelevant peptide derived from Plasmodium falciparum molecule.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210 209

gradation in vitro and decrease worm burdens and egg production

in vivo, Mol. Biochem. Parasitol. 81 (2002) 179�/189.

[14] M.A. El Meanawy, T. Aji, N.F.B. Phillips, R.E. Davis, R.A.

Salata, Y. Malhortra, D. Mc Clain, M. Aiksawa, A.H. Davis,

Definition of the complete Schistosoma mansoni hemoglobinase

mRNA sequence and gene expression in developing parasites,

Am. J. Trop. Med. Hyg. 43 (1990) 67�/78.

[15] J.P. Dalton, K.A. Clough, M.K. Jones, P.J. Brindley, The cysteine

proteinases of Schistosoma mansoni cercariae, Parasitology 114

(1997) 105�/112.

[16] A.H. Davis, J. Nanduri, D.C. Watson, Cloning and gene

expression of Schistosoma mansoni protease, J. Biol. Chem. 262

(1987) 12851�/12855.

[17] C.R. Caffrey, M.A. Mathieu, A. Gaffney, J.P. Salter, M. Sajid,

K.D. Lucas, C. Franklin, M. Bogyo, J.H. McKerrow, Identifica-

tion of a cDNA encoding an active asparaginyl endopeptidase of

Schistosoma mansoni and its expression in Pichia pastoris , FEBS

Lett. 466 (2000) 244�/248.

[18] M.Q. Klinkert, K. Bommert, D. Moser, R. Felleisen, G. Link, O.

Doumbo, E. Beck, Immunological analysis of cloned Schistosoma

mansoni antigens Sm31 and Sm32 with sera of schistosomiasis

patients, Trop. Med. Parasitol. 42 (1991) 319�/324.

[19] O. Noya, Z. Fermın, B. Alarcon de Noya, S. Losada, C.

Colmenares, T. Hermoso, Humoral immune responses of children

with chronic schistosomiasis. Isotype recognition of adult worm

antigens, Parasite Immunol. 17 (1995) 319�/328.

[20] L.H. El-Sayed, H. Ghoneim, S.R. Demian, M.H. El Sayed, N.M.

Tawfik, I. Sakr, L.M. Abou-Basha, E. Renganathan, M.Q.

Klinkert, N. Abou-Rawash, Diagnostic significance of Schisto-

soma mansoni proteins Sm31 and Sm32 in human schistosomiasis

in an endemic area in Egypt, Trop. Med. Int. Health. 3 (1998)

721�/727.

[21] R. DiMarchi, G. Brooke, C. Gale, V. Cracknell, T. Doel, N.

Mowat, Protection of cattle against foot-and mouth disease by

synthetic peptide, Science 232 (1986) 639�/641.

[22] M.E. Patarroyo, R. Amador, P. Clavijo, A. Moreno, F. Guzman,

P. Romero, R. Tascon, A. Franco, L.A. Murillo, G. Rontin, G.

Trujillo, A synthetic vaccine protects humans against challenge

with asexual blood stages of Plasmodium falciparum malaria,

Nature 332 (1988) 158�/161.

[23] M.E. Patarroyo, Study of the safety and immunogenicity of the

synthetic malaria SPf66 vaccine in children aged 1�/14 years,

Vaccine 10 (1992) 175�/178.

[24] D. Ritcher, D. Harn, Candidate vaccine antigens identified by

antibodies from mice vaccinated with 15 or 50-kilorad-irradiated

cercariae of Schistosoma mansoni , Infect. Immun. 61 (1) (1993)

146�/154.

[25] J.T. Jones, M.D. McCaffery, J.R. Kusel, The influence of the H-2

complex on responses to infection by Schistosoma mansoni in

mice, Parasitology 86 (1983) 19�/30.

[26] M.M. Bradford, A rapid and sensitive method for the quantita-

tion of microgram quantities of protein utilizing the principle of

protein-dye binding, Anal. Biochem. 72 (1976) 248�/254.

[27] R.B. Merrifield, Solid-phase peptide synthesis. The synthesis of a

tetrapeptide, J. Am. Chem. Soc. 85 (1963) 2149�/2455.

[28] R.A. Houghten, S.T. DeGraw, M.K. Bray, S.R. Hoffmann, N.D.

Frizzell, Simultaneous multiple peptide synthesis: the rapid

preparation of large number of discrete peptides for biological,

immunological and methodological studies, BioTechniques 4

(1986) 522�/528.

[29] J. Tam, Synthetic peptide vaccine design: synthesis and properties

of a high-density multiple antigenic peptide system, Proc. Natl.

Acad. Sci. USA 85 (1988) 5409�/5413.

[30] T.P. Hopp, K.R. Woods, Prediction of protein antigenic deter-

minants from amino acid sequences, Proc. Natl. Acad. Sci. USA

78 (1981) 3824�/3828.

[31] J.G. Guillet, J. Hoebecke, R. Lengagne, K. Tate, F. Borras-

Herrera, A.D. Stroeberg, F. Borras-Cuesta, Haplotype specific

homology scanning algorithm to predict T-cell epitope from

protein sequences, J. Mol. Recognition 4 (1991) 17�/25.

[32] G. Deleage. ANTHEPROT release 4. (1999) http://pbil.ibcp.fr/

ANTHEPROT. E-mail: g.deleage@ibcp.fr IBCP, 7 passage du

Vercors, 69367 Lyon cedex, France.

[33] O. Noya, B. Alarcon de Noya, The multiple antigen blot assay

(MABA): a simple immunoenzymatic technique for simultaneous

screening of multiple antigens, Immunol. Lett. 63 (1998) 53�/56.

[34] U.K. Laemmli, Cleavage of structural proteins proteins during

the assembly the head of bacteriophage T4, Nature 227 (1970)

680�/685.

[35] M. Eberl, J.A. Langermans, P.A. Frost, R.A.Q. Vervenne, G.J.

van Dam, A.M. Deelder, A.W. Thomas, P.S. Coulson, R.A.

Wilson, Cellular and humoral immune responses and protection

against schistosomes induced by a radiation-attenuated vaccine in

chimpanzees, Infect. Immun. 69 (9) (2001) 59352�/59362.

[36] W. Yang, D.C. Jackson, Q. Zeng, D.P. McManus, Multi-epitope

schistosome vaccine candidates tested for protective immunogeni-

city in mice, Vaccine 19 (2000) 103�/113.

[37] R. Bergquist, M. Al-Sherbiny, R. Barakat, R. Olds, Blueprint for

schistosomiasis vaccine development, Acta Trop. 82 (2) (2002)

183�/192.

[38] P. Hagan, M.J. Doenhoff, R.A. Wilson, M. Al Sherbiny, R.

Bergquist, Schistosomiasis vaccines: a response to a devils’

advocate’s view, Parasitol. Today 16 (8) (2000) 322�/323.

[39] A. Capron, M. Capron, D. Dombrowicz, G. Riveau, Vaccine

strategies against schistosomiasis: from concepts to clinical trials,

Int. Arch. Allergy Immunol. 124 (2001) 9�/15.

[40] A. Capron, M. Capron, G. Riveau, Vaccine develepment against

schistosomiasis from concepts to clinical trials, Br. Med. Bull. 62

(2002) 139�/148.

[41] P.J. Brindley, B.H. Kalinna, J.P. Dalton, S.R. Day, J.Y. Wong,

M.L. Smythe, D.P. McManus, Proteolytic degradation of host

hemoglobin by schistosomes, Mol. Biochem. Parasitol. 89 (1997)

1�/9.

[42] J.M. Chen, P.M. Dando, N.D. Rawlings, M.A. Brown, N.E.

Young, R.A. Stevens, E. Hewitt, C. Watts, A.J. Barrett, Cloning,

isolation and charaterization of mammalian legumain, an aspar-

aginyl endopeptidase, J. Biol. Chem. 272 (1997) 8090�/8098.

[43] B. Livingston, C. Crimi, M. Newman, Y. Higashimoto, E.

Appella, J. Sidney, A. Sette, A rational strategy to design

multiepitope immunogens based on multiple Th lymphocyte

epitopes, J. Immunol. 168 (2002) 5499�/5506.

N. Chacon et al. / Immunology Letters 88 (2003) 199�/210210